ROBERT A. TAFT WATER RESEARCH CENTER
REPORT NO. TWRC-14
MATHEMATICAL MODEL OF
TERTIARY TREATMENT
BY LIME ADDITION
ADVANCED WASTE TREATMENT RESEARCH LABORATORY- XIV
U.S. DEPARTMENT OF THE INTERIOR
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
OHIO BASIN REGION
Cincinnati,Ohio
-------
MATHEMATICAL MODEL OF TERTIARY TREATMENT
BY
LIME ADDITION
L. Seiden and K. Patel
for
The Advanced Waste Treatment Research Laboratory
Robert A. Taft Water Research Center
This report is submitted in
fulfillment of Contract Wo.
1^-12-^16 between the Federal
Water Pollution Control Admin-
istration and General American
Research Division.
U. S. Department of the Interior
Federal Water Pollution Control Administration
Cincinnati, Ohio
September 1969
-------
FOREWORD
In its assigned function as the Nation's principal natural resource
agency, the United States Department of the Interior bears a special
obligation to ensure that our expendable resources are conserved, and that
all resources contribute their full measure to the progress, prosperity,
and security of America -- now and in the future.
This series of reports has been established to present the results of
intramural and contract research studies carried out under the guidance of
the technical staff of the FWPCA Robert A. Taft Water Research Center for
the purpose of developing new or improved wastewater treatment methods.
Included is work conducted under cooperative and contractual agreements
with Federal, state, and local agencies, research institutions, and indus-
trial organizations. The reports are published essentially as submitted
by the investigators. The ideas and conclusions presented are, therefore,
those of the investigators and not necessarily those of the FWPCA.
Reports in this series will be distributed as supplies permit. Requests
should be sent to the Office of Information, Ohio Basin Region, Federal
Water Pollution Control Administration, 46?6 Columbia Parkway, Cincinnati,
Ohio 1*5226.
-------
ACKNOWLEDGEMENT
This program was performed for FWPCA under Contract No. 1^-12-Ul6.
The Project Officer was Mr. Robert Smith, Chief, Treatment Optimization
Research Program, Advanced Waste Treatment Research Laboratory.
The authors wish to thank Messrs. Dolloff F. Bishop, Chief, FWPCA-DC
Pilot Plant, and Edward L. Berg, Municipal Treatment Research Program,
Advanced Waste Treatment Research Laboratory, FWPCA, for kindly supplying
us with information relative to their operating experience at Blue Plains,
D. C. and Lebanon, Ohio, respectively, and for their cooperation during
our visits to these installations. We also wish to express thanks to
Mr. James Zornes, Production Manager, Nevada Power Company.
In addition we wish to acknowledge the assistance of Mr. Joseph F.
Roesler, Advanced Waste Treatment Research Laboratory, FWPCA, for his
interest and helpful comments during the course of this project.
111
-------
TABLE OF CONTENTS
PAGE WO.
FOREWORD ii
ACKHOWLEDGEMEM] ill
ABSTRACT vii
INTRODUCTION 1
DESCRIPTION OF THE TERTIARY LIME TREATMENT PROCESS 2
Chemical Reactions 2
CaCO^ and Mg(OH)p Precipitation 2
Precipitation of Phosphates 2
Phosphorus Chemistry 2
Reaction with Lime 5
Physical Processes 6
LITERATURE SURVEY 9
Chemical Precipitations 9
Lime Addition 10
SURVEY OF OPERATING PLANTS AND PILOT PLANTS 11
Pomona, California 11
Blue Plains, Washington, D. C. 1^
Las Vegas, Nevada lU
Lebanon, Ohio 1^
South Tahoe, California 17
CORRELATIONS 18
Performance 18
Chemical Phosphorus Removal 18
Orthophosphate 18
iv
-------
TABLE OF CONTENTS (CONTINUED)
PAGE NO,
Total Phosphate 20
P vs. pH Relationship 20
Calcium Carbonate Supersaturation 22
Effect of Temperature 22
Coagulant Dose 2.h
Capital Cost and Sizing 2U
Operating Costs 29
METHODOLOGY USED IN COMPUTER PROGRAM 33
Input 33
Chemical Equilibria 33
Solids 36
Flow Rates 36
Costs and Sizing 38
CONCLUSIONS AND RECOMMENDATIONS 39
General Conclusions 39
Limitations of the Model 39
P vs. pH Relationship . 39
Solids Separation Ul
Sizing 4l
Coagulant ^3
Suggestions for Further Work ^3
-------
TABLE OF CONTENTS (CONTINUED)
PAGE NO.
APPENDIX A COMPUTER PROGRAM HE STING SAMPLE PRINTOUT
AND DEFINITION OF SYMBOLS A-l
APPENDIX B REFERENCES B-l
APPENDIX C BIBLIOGRAPHY C-l
VI
-------
ABSTRACT
The status of phosphorus removal from secondary effluents by lime
addition is presented. Eased on the empirical information available, a
mathematical model of the process was developed.
The factor which best correlated with phosphorus removal was the pH
of the tertiary effluent. Cost information for the model was generated
based on a solids-contact type precipitator such as the Infilco Densator.
•A computer program, in FORTRAN IV, derived from the model, was developed
for use as a subroutine in the FWPCA Executive Calling Program. Given the
input stream flow rate and analyses, and the final pH, the program computes
the necessary doses of lime and coagulant, their costs, the size and cost
of the equipment, the degree of phosphorus removal, and the output stream
analyses.
VI1
-------
INTRODUCTION
During the past decade, there has been a growing public awareness of
the necessity for more diligently conserving our water resources. The
growth of urbanization and industrialization have placed increasingly
heavier loads on what were at one time considered adequate waste-water
treatment facilities and processes. It is now quite clear that, if the
quality of our environment is to be maintained, prompt attention must be
given to the development and application of more advanced and rigorous
waste-water treatment techniques.
Excluding gross chemical pollution by specialized industries, which
may be dealt with on an individual basis, a problem common to almost all
municipal waste-treatment facilities is that of nutrient removal. It is
now well established that even moderately high levels of phosphorus and/or
nitrogen in sewage plant effluents accelerate the rate of eutrophication in
receiving waters and serve as nutrient material for undesirable algal blooms.
With phosphorus in particular, this problem has been aggravated by the
growing use of phosphorus-containing detergents.
Thus, P removal is one of several high-priority "advanced" waste-treatment
process which are being investigated by the Federal Water Pollution Control
Administration.
Reduction of phosphorus levels in sewage plant effluents may be accom-
plished by several means. The purpose of this report is to describe one
method in particular: chemical precipitation by lime addition as a tertiary
treatment process, i.e., one which follows the well-established primary and
secondary (activated sludge) treatment steps.
P removal by lime addition is accomplished via the precipitation of
calcium phosphate. It is known that a relatively high pH is necessary for
this to occur, and the reaction: CaO + HgO —^Ca2+ + 20H" accomplishes
this. The resultant high pH of the effluent also makes the process desirable
as it converts nitrogen in the form of MH^.+ ion into HHo, allowing a subse-
quent ammonia-stripping step (removal of another possible nutrient, nitrogen)
with no further pH adjustment.
The objectives of this project, sponsored by the FWPCA, were to survey
the current status of the process, to attempt a mathematical modeling of the
process, and to develop a computer program for describing the performance
and cost relationships involved. The computer program is meant to be used
as a subroutine for the Executive Calling Program being developed by the
Operations Research Unit of FWPCA.
Because of the association of General American Research Division with
Infilco/Fuller Corporation, it was requested that the modeling of the process
be in terms of solids-contact type precipitation equipment such as the Infilco
Densator.
-------
DESCRIPTION OF THE TERTIARY LIME TREATMENT PROCESS
Any attempt to describe or model the process in question must recognize
that it in reality consists of at least two groups of sub-processes. We have
first the chemical reactions which occur when lime is added to a phosphate-
bearing secondary effluent. This is followed by the physical (or physico-
chemical) processes involved in coagulation, flocculation, and sedimentation
of the solids formed by the chemical reactions. This division into discrete
"steps" is somewhat arbitrary, as there is undoubtedly much interaction of
one upon the other; it is, however, convenient for the purposes of exposition.
CHEMICAL REACTIONS
CaC03 and Mg(OH)2 Precipitation
The addition of lime (either quicklime, CaO, or slaked lime, Ca(OH)2) to
a secondary effluent will, in addition to the precipitation of phosphate,
result in the formation of a relatively large amount of solid calcium carbonate.
This is identical to the standard cold-lime softening process, in which COg
and HCOj^react, as the pH increases, to form COo , which in turn precipitates
insoluble CaCOo.
With increasing lime dose (and pH), any magnesium present in the system
will also begin to precipitate as Mg(OH)p. This usually occurs at pH's in
the vicinity of 11.0.
Precipitation of Phosphates
Phosphorus Chemistry
Before discussing the interaction of lime with phosphates, it might be
well to briefly describe and define some aspects of the chemistry of phosphates.
We may first define "ortho-phosphates" as those phosphorus-containing
compounds or ions which are derived from orthophosphoric acid, H~PO. :
H
0
HO P OH
0
This compound in solution will lose from one to all three of its protons,
yielding the orthophosphate ions HpPOi , HPO. ~, or POj/, thus:
-------
H
H+
2- _ a*. + 3-
•, H + POL
The particular ionic forms in which orthophosphats exists is determined by
the hydrogen ion activity of the solution (approximately equal to the H
concentration), according to the relations:
H PO, "1 VI o
J = K T.iOxio-3
K21 [*! -8
1 J l.. J = K?
-------
1.0
0.1
a
o
•H
-P
O
OS
0.01
0.001
8.0
9.0
10.0
11.0
12.0
PH
FIGURE 1. Fraction of Orthophosphate in Each Form vs. pH
-------
The condensed phosphate ions, In general, do not form particularly
insoluble salts with calcium, although some do combine with Ga ^ ions to
form soluble complexes.
In a well-stabilized secondary effluent, condensed phosphates normally
constitute no more than ICf? of the total phosphate present.
Reaction with Lime
The chemical interactions which occur when lime is added to a secondary
waste-water containing phosphorus compounds are quite complex and not fully
understood. This is not surprising, in that even the chemistry of a "pure"
aqueous system, containing only calcium and ortho-phosphate ions, is not
readily susceptible to simple analysis in terms of classical chemical concepts
of solubility products, etc. Indeed, the calcium-phosphate precipitate formed
may vary in composition depending on the conditions of formation, such as
temperature, pH, Ca/P ratio in the precipitating solution, and other unknown
factors. The calcium orthophosphate precipitate has the basic structure of
a hydroxyapatite, and is usually considered to be Caj_o(OH)2(POLv)6- However,
the Ca/P atomic ratios in the solid material are found to vary from 1.33 to
2.0.W
The apparent solubility product constant of this material,
Ksp - (Ca)10 (POU)6 (OH)2 (U)
as measured by several workers, shows a variation over a range of 10 . Clark^
has recently carefully measured this product to be 10~U-5. Using this criterion,
it can be shown that a moderately hard typical secondary effluent containing
approximately 50 mg/1 Ca, 10 mg/1 P01+, and at a pH of about 8, should need no
further treatment for phosphorus removal, the maximum "equilibrium" ortho-
phosphate concentration being about 10~3 mg/1 POij. As this is obviously not
the case, it appears that a solubility product-type calculation is inappropriate
for predicting P removal. The possible reasons for the deviation of this system
from "ideal" behavior are many, and will not be dealt with in detail. Among
them are: (a) sluggishness of the precipitation process, including formation
of colloidal particles which will not settle out or be caught by filters;
(b) inhibition of or interference with the reaction due to the presence of
other soluble species, particularly organics.
Going from the "pure water" case to a more realistic one, we must take
into account the presence of Mg2"1", CO^ , and HCOo" ions, solid Mg(OH)2 ani
CaCOg, and the presence of a small but significant amount of condensed phos-
phates. This increases even further the complexity of trying to deal with
phosphorus removal according to classical theoretical concepts. These species
(l) References are listed in Appendix B.
-------
will all interact with each other; the presence of solid CaCCh and Mg(OH)2
may result in more efficient precipitation of calcium phosphate by decreasing
the tendency for self-nucleation and colloid formation. The presence of
polyphosphates, on the other hand, is known to inhibit the precipitation of
calcium carbonate. Mg(OH)2 in particular is said to enhance P-removal "by
acting as a flocculating agent. Schmid(3) has demonstrated that polyphosphate
is removed from solution to a large extent "by adsorption on both Ca-PO^. and
CaCC>3 sludge particles.
PHYSICAL PROCESSES
If a solution is supersaturated with respect to a particular solid
species, this supersaturation may be relieved in one of two ways: (a) by
deposition of solid material onto the surfaces of previously formed precipi-
tate particles, or (b) by spontaneous self-nucleation, i.e., the formation of
many small crystal nuclei. This latter process is undesirable in that it
leads to the presence of a large number of very small (colloidal) particles
which do not settle, and may not be filterable. Usually this phenomenon
occurs at high degrees of supersaturation.
The chemical reactions in the preceding section have been described as
if they proceeded instantaneously. In reality, a finite time is involved in
the formation and growth of precipitate particles. In general, the rate of
a precipitation process is proportional to both the degree of supersaturation
and the surface area available for further crystallization.
The aims in the process being considered are, of course, to have the
precipitation reactions occur as rapidly as possible, yet with the least
degree of self-nucleation. The design of a solids-contact type precipitator,
such as a Densator, is meant to accomplish both of these aims. It does this
by recirculating a relatively high concentration of previously formed sludge
into the reaction zone where lime is being added to the influent stream. A
high degree of agitation is provided here, for good mixing. In this manner,
a large surface area for new precipitation is supplied, increasing the rate
of solids formation and decreasing the tendency for spontaneous nucleation.
Figure 2 shows a typical Densator configuration.
After passing through the "primary" reaction zone where lime and recycle
sludge are added, the influent stream then passes down through a "secondary"
reaction zone. Here the degree of agitation is more gentle, and coagulation
and flocculation occur, leading to particle size growth by particle-particle
collision and agglomeration. (There is provision here for so-called "split-
stream" treatment, where only a portion of the influent stream is supplied
to the primary zone, the balance entering in the secondary zone. While
desirable for softening applications, it is not clear whether there is any
advantage in using this split-stream technique in tertiary treatment.) It
is here that coagulants may be introduced, if needed.
-------
Sludge Waste
PRZ Raw Water
Lime Feed J^
Recirc.
Sludge
taw Water to SRZ
(tplit Treatment)
Recirc. Sludge
FIGURE 2. Typical Densator Configuration
7
-------
Upon exiting the secondary zone, the stream then turns upward into the
external clarification or sedimentation region. Here again, there is more
opportunity for particle-particle interaction and growth. The solids settle
to the bottom of the unit, while the clarified water is drawn off the top.
From the above description of the "physical" processes, the complexity
of the situation may be appreciated. Even assuming one knows precisely the
amounts of chemical solids formed, one must then be able to relate the degree
of clarification to the physical parameters of the system such as sludge
recycle rate, flocculation and sedimentation zone residence times, agitation
velocities, upflow rates, etc.
It is apparent how the chemical and physical processes may interact with
each other. Thus, the physical parameters such as recycle sludge concentra-
tion affect the chemistry by determining the rates of formation of precipitates
At the same time, the chemistry of the system, such as the amount of Mg(OH)2
formed (which acts as a coagulant), will affect the physical processes of
coagulation, flocculation and sedimentation.
8
-------
LITERATURE SURVEY
As a preliminary effort, a literature survey was made to determine the
state-of-the-art of phosphorus removal from secondary effluents by chemical
treatment .
There are three basic techniques which have been studied in any detail.
These include the addition of lime, alum, and iron salts. While all of these
have long been recognized for their usefulness in clarification of waste
waters, it has been only relatively recently that workers have looked at
their effectiveness for P removal. This work, the bulk of which dates from
the late 19^-0 ' s and early 1950' s, has been directed not only at P removal
in a discrete tertiary step, but also by chemical additions in earlier phases
of the conventional treatment processes. We have, however, concentrated
primarily on those articles which would be applicable in the context of a
chemical precipitation process following conventional secondary treatment.
CHEMICAL PRECIPITATIONS
The work of Sawyer^ ' and Owen^ ' established the efficacy of lime, alum,
and iron salts in removing phosphorus from secondary effluents . Sawyer
worked with all three, and his results indicated that the dose needed for a
given P removal with alum or FeClo was increased when polyphosphates were
added to the system, but that the lime requirement was not increased. Lea,
Rohlich and Katz(°) investigated aluminum sulfate, ferric and ferrous sulfate,
and copper sulfate, and concluded that of these, filter alum was the most
suitable coagulant for phosphorus removal, because of the possibility of
coagulant recovery. They also concluded that the P-removal mechanism with
these chemicals was primarily via adsorption on the hydroxide floes formed,
with possibly some precipitation of ferrous phosphate as well.
Henriksen^ ' studied the coagulation of P by iron and aluminum salts,
and found that for a given P removal, the necessary doses were dependent
on the initial P concentration. These findings were borne out for alum by
Malhotra, Lee and Rohlich(^), who also concluded that on a chemical cost
basis, lime would be the most economical chemical precipitant for the activated
sludge effluent at Madison, Wisconsin. Their prime objection to the use of
lime was the high pH of the resulting effluent. (This, of course, would not
be objectionable if a subsequent ammonia removal step were contemplated.)
^ '
has reported on the waste-water reclamation plant at South Tahoe,
California, where alum addition, followed by filtration and final polishing,
was used for P removal in a 2.5 mgd tertiary plant. Studies conducted during
construction and initial operation showed that the initially contemplated
alkaline recovery of the alum would not be feasible, and the plant has now
gone over to lime addition.
-------
LIME ADDITION
Owerr ' performed both laboratory and field studies of lime addition to
a secondary effluent. His laboratory results indicated that by dosing with
lime to a pH of 11.0, essentially complete removal of phosphorus could be
achieved. These studies also pointed out the importance of good solids
removal. Samples settled for one hour yielded P concentrations of 0.3 rag/1,
•while in those settled for 18 hours, the residual P concentration was 0.015
mg/1. His field tests were performed on the effluent from a trickling filter
secondary system; with admittedly crude mixing and dosing equipment, he
achieved soluble phosphorus residuals of 0.13 ppm with a dose of 720 ppm
Ca(OH)2 (equivalent to 520 ppm CaO).
Rudolfs^ , in studies directed primarily at clarification rather than
P removal, showed that addition of Ca alone did not result in appreciable
precipitation of phosphorus, whereas subsequently raising the pH with NaOH
decreased the P content of the solutions. More recently, MalhotraC11) ob-
tained similar results, reporting on jar tests with secondary effluents in
which P removals of 92-96$ were obtained (with an initial Ca concentration
of 63 mg/1) by adding NaOH to obtain a pH of 11.5.
Schmid^ ' performed a large number of jar tests on synthetic sewage 2
effluents in order to determine the relations and interactions between Ca ,
C0o^~, ortho-phosphates, and polyphosphates in the precipitation reactions.
The context of this study was in terms of a combined lime-biological treatment
system for raw sewage rather than tertiary treatment; therefore, many of the
experiments were run with higher orthophosphates and polyphosphate concentra-
tions, and at lower pH's, than would be typical of a tertiary process. How-
ever, a good deal of this work is of interest in elucidating the mechanism
of phosphate removal. Among his finds were: (a) at pH's higher than 9,
polyphosphates are removed from solution primarily by adsorption on calcium
carbonate and calcium orthophosphate precipitates; (b) the presence of poly-
phosphates results in marked inhibition of CaCOo precipitation (at least at
pH's up to 10.5); (c) while the presence of CaC03 precipitate is not necessary
for the precipitation of orthophosphate, it appears to be required for
efficient removal of polyphosphates (probably because it provides the necessary
adsorption surface).
In addition to the documents cited above, many more were surveyed, in-
cluding those pertaining not only to the chemistry of the process, but also
to the physical aspects of coagulation, flocculation and sedimentation. As
with the above papers, most were found to be useful in terms of general back-
ground information, but of little quantitative value in attempting to model
the actual performance of the P removal process. It was, therefore,not con-
sidered appropriate to include them in the body of this report. A reading
list covering most of these documents will be found in Appendix C, where they
are classified according to primary subject matter.
10
-------
SURVEY OF OPERATING PLANTS AND PILOT PLANTS
Weinberger ' has listed those plants or pilot plants which have
operational experience with tertiary phosphorus removal "by lime addition,
or which are doing research on this topic. These include those at South
Tahoe, California, Washington, D. G., and Las Vegas, Nevada. In addition,
pilot plant units at Lebanon, Ohio, and Pomona, California have also done
some work in this respect.
A great hindrance to the modeling process has been the scarcity of
reliable, pertinent data which is available from these sources. It is help-
ful, in a project of this sort, to have a large amount of data, taken at
many different installations and taken under many different operating con-
ditions, so as to be able to obtain correlations which would be of general
applicability in predicting performance.
Unfortunately, such has not been the case in this study. Except in
certain limited cases, it has not been possible, when studying the effects
of one variable, to ensure that other possible variables are remaining
fixed. In many cases, information concerning possible parameters of interest
has not been available. Thus, for example, effluent phosphorus analyses have
sometimes been supplied with no distinction between condensed and ortho, or
filtered and unfiltered. Little information concerning influent analyses
has been available. Those plants obtaining satisfactory effluent quality
for their particular application have done little in the way of varying their
operation parameters . In several plants, phosphorus removal has been a
secondary consideration to some other goals. Thus, the population from which
our correlations have been drawn has been an extremely limited one. As will
be explained in the following section, we have, at times, had to rely on
laboratory data which might not be applicable to full-size plant operations.
Below are listed those sources from which data has been obtained:
POMONA, CALIFORNIA
The Pomona installation is a Densator pilot plant, operated by the
Los Angeles County Sanitary District. Data was obtained during the early
part of 1968. However, its primary purpose was for conditioning water for
input to ion exchange columns, P removal being a secondary consideration.
Operating data was obtained for the period March -May 1968. These
consisted of input and output pH, alkalinity, Ca, total and ortho-P, and
lime doses (all daily averages) . The effluent values were given both un-
filtered and filtered (sand) and the pH range covered was from 9-1 to 11. k.
This information is presented in Table 1.
11
-------
TABLE 1
POMONA DATA*
PH
S
7.5
7.4
7.4
7.3
7.4
7-5
7-5
7-5
7.5
7.6
D
10.5
10.7
10.3
10.2
10.7
11.4
10.2
10.6
F
10.4
10.6
10.3
"— V • ^J
10.4
10.6
11.4
10.3
10.3
10.7 10.1
11.0
10.9
i
7-4 10.7 j 10 .6
J7.3
10.5 10.5
i ,
7-5 10.5*10.4
' 7-3;io.2;io.i
ALKALIKITY
mg/1 CaC03
S
252
248
260
264
264
280
264
280
268
272
D
192
208
212
196
216
256
208
220
220
236
260 228
252 224
269
221
I
i 264 i 228
; . (
'
7.4 10.2.10.2
7-5 10.2:10.1
252
(
232
224
i .
F
192
168
?l?
204
228
268
216
208
220
216
208
208
218
220
228
224
HARDHESS
mg/1 CaCOo
j
S
200
206
208
210
192
204
200
198
208
196
D
168
126
120
116
110
148
106
106
i4o
144
204
204
196
214
206
'198
1 i
138
138
i4l
148
F
130
132
128
132
140
176
124
_
130
130
134
132
130
136
136 | 134
150 146
CALCIUM
mg/1 CaCOg
S
-
-
-
-
D
-
-
-
-
- I *
1
io4
96
96
F
-
-
-
-
-
96
-
! i
; 1
124
98
100
165
131
|135
100
145
j ;
7.4 10.4;10.3: 260 216 2l2j J210 146
i i .
138! ;i65
! !
i .• . i ; i '
118
100
i
108 j 100
122
110
i
120 J108
102 jlOO
!
104 J102
TOTAL
PHOSPHORUS
mg/1 P
S
13.7
9-3
13.4
•*™_^ • 1
9-8
12.9
1
* _
:io.8
i
112.1
1
10.5
k
f
I
I
i
-
i -
jlO.8
12.1
'
J
104 ;100 0.4.7
102 ;100
\
9.5
110 :106 ' i -
! , i
i ' i ; ! ! 1
D
0.85
-
1.01
1.89
1.92
-
F
0.17
-
0.10
0.13
0.20
-
i
2.8
1.96
3-9
1.3
-
-
-
2.1
0.23
0.24
0.35
0.33
-
-
0.16
ORTHO-
PHOSPHORUS
j mg/1 P
0.20
0.91J0.36 !
i
- 10.15
;
-
S
11.4
-
3.8
*j • *^
8.5
9.8
-
8.5
8.8
7.8
8.1
-
-
8.5
9-5
9.8
8.2
9-5
D
0.05
-
0.07
v • \s 1
0.15
0.07
-
0.10
0.16
-
0.12
-
-
0.10
F
0.05
-
0.03
0.06
0.08
-
0.15
0.16
0.13
0.03
-
-
0.07
0.13 jo. 13
0.13
0.10
0.10
o.n;o.ii
S = Secondary Effluent
D = Densator Effluent
F = Filter Effluent
12
-------
TABLE 1
POMOHA DATA (COIWIMJED)*
PH
S
7-6
7-3
7-4
7-5
7-5
;
i 7-4
7.4
|7.5
t T H-
7 ^1-
;7.5
•7.5
; 7.4
!
!7.6
7.4
7.4
D
10.0
9.8
9.6
9-7
9-3
9-5
9.4
9-5
9-7
9-5
9-3
l
9-1
! 9-1
; 9.4
F
9-9
9-8
9-7
j
9.6
9.4
•^
9.6
9.4
i
9.4
9-6:
ALKALIMTY
mg/1 CaCOo
S
_
274
268
D
_
1
272
260
286 j 224
260
274
280
284
264
9.4; 274
9.3:
268
1
9.2 275
9-1;
272
9 -5;; 276
, 9.3 9.3 '264
,
9.2^9.2 256
244
256
240
277
240
212
280
F
_
268
244
240
236
244
248
267
244
192
272
255|202
i [
304
|288
t
264
296
288
256
,300 [288
HARDNESS
mg/1 CaC03
S
205
198
D
155
173
204 172
216
206
206
158
148
168
}
210
203
208
190
160
182
156
185
208 |l92
210
224
204 !l96
i
1212 180
204
180
1208 208
' i
1 !
1
i
F
150
i
171
•
180
156
162
CALCIUM
mg/1 CaC03
S
_
129
160
150
175
180
180
j
i
96
145
D
_
124
122
132
120
102
104
F
_
121
136
122
120
104
140
1 ;
181: l48;125 126;
168 i - i -
i t
170
212
1
124
1 *"
1
206
1
144
218
208 92
184
i
1216
120
_
135
-
124
98 104
j
-
- |
156
_
116
— i
i
152 i
.L^,
TOTAL
PHOSPHORUS
mg/1 P
S
10.2
12.4
10.0
D
-
3.1
2.8
10.0 -
10.4
13-4
10.0
2.1
4.0
-
9.112.9
10.8J3-3
13-1
21.9
10.8
10.0
13.7
6.5
3.3
8.2
5.6
12.4
12.1 6.5
- - ; 9-8
10.6
!
, ;
F
0.29
ORTHO-
PHOSPHORUS
mg/1 P
s 1
D
-t
F
8.3 JO. 20 (0.20
I
0.31 io.4
0.49
0.52
0.72
0.72
0.59
8.9
8.2
'
i 7-5
i
' 9-5
; 8.2
0.56 7.5
0.42
0.36
0.65
0.65
0.72
0.59
0.471 j 9.1 0.57
0.65 io.4
0.62
',
: 8.2
1 !
3.6 8.8
0.9
: 8.0
' 0.78 10.1
. 0.78 : 8.2
0.59 j 7-7
' i •
: ]
0.62
0.52
0.73
0.90
0.91
0.69
'o.8l
i
1
0.3l!
0.32!
i
0.39
0.32
'
1
0.52
i
0.52
0.49
0.36
0.46
0.49
0.60
0.62
0.65
0.52
0.55
* S = Secondary Effluent
D = Densator Effluent
F - Filter Effluent
13
-------
BLUE PLAINS, WASHINGTON, D. C.
The Blue Plains installation is a joint FWPCA-D.C. tertiary treatment
pilot plant. It has two Densators, operating at 5^- gpn> with provisions
for recarbonation between the two units. The bulk of the data given us (by
Mr. D. F. Bishop, Plant Chief) was for P removal around the first unit only.
Indications were that little, if any, further P removal was accomplished with
two-stage operation. The plant is described in the report "Status and Out-
look for Phosphorus Removal from Waste-Water" by Mr. Bishop(13).
Mr. Bishop supplied us with operating data and analyses for various
treatment modes, as shown in Table 2. In addition, we were given access
to the results of laboratory jar tests performed by Mr. Bishop on the
secondary effluent. These results will be presented and commented upon in
a later section of this report.
LAS VEGAS, NEVADA
The Nevada Power Company has been using lime clarification of secondary
effluent for several years to produce cooling tower makeup water. Mr. James
Zornes, Production Manager, was contacted and supplied us with typical
influent and effluent analyses and operating parameters for the Sunrise
Station. This plant utilizes an Eimco Eeactor-Clarifier (a solids-contact-
type precipitator) housed in a 60 ft. diameter basin. It is designed for
an average flow of 2.5 mgd, and utilizes an average lime dose of 1 ton/
million gallons. A flocculant aid, Nalco 603, is used at a dose of 3 mg/1.
The average effluent pH is 10.0, and a residual total P level of 1-95 mg/1
(as POii) is achieved in the clarifier. The clarifier effluent is not filtered,
but is piped to a 6 million gallon reservoir for storage and settling.
P analysis on the reservoir effluent was 1.25 mg/1 (as P01<.).
Capital and operating cost figures were given both for this plant and
another, the Clark Station. However, the capital figures did not seem
consistent with those developed by us (see p. 25) and the chemical treatment
costs were not consistent with the doses specified. Therefore, these figures
were not used in our cost correlations.
LEBANON, OHIO
The Lebanon, Ohio plant is operated by the FWPCA for study of various
advanced treatment techniques. One of the installations is a 0.1 mgd
upflow lime clarifier, the main purpose of which is to investigate removal
of suspended material rather than P removal. The effluent from the reactor
is passed through dual media filters for final clarification.
While for the most part this unit is normally run at rather low pH
levels (9-9-5)> a series of experiments has recently been performed in which
the pH was varied from 9-0 to 11.0. Mr. Edward L. Berg, the engineer in
-------
vn
TABLE 2
BLUE PLAINS PILOT PLANT
Mode
Total Phosphorus
(mg/1 as PO^)
T.O.C. (mg/1 as C)
Suspended Solids
(mg/1)
Mg++(mg/l as Mg)
Ca++ (mg/1 as Ca)
Turbidity J.T.U.
Alkalinity
(mg/1 as CaCO~)
Temperature
PH
Flow (gpm)
Lime Dose (mg/1 as CaO)
Soda Ash Dose
(mg/1 as Na2C03)
Recycle
Others
JA
Raw 1st Stage 2nd Stag
22 2.4 2.4
39.0 20.8 18.2
46 26 21
6.3 1.4 1.4
28.3 90.6 58.5
4.5 23
118 - 138
22°c
11 . 4 10 . o
34
200-300
None
None
Two Stage
2A
;e Raw Clar
20 1.5
34 15.4
58 16
6.3 1-5
36 46
2.7
108 220
24°C
11.2-11.5
54
200-300
175
None
Na C0_ to
3A 4_A 5_A
Raw Clar Raw Clar Raw Clar
21 0.04 19 1..1 21 2.2
19.4 9.6 32 15-8 36 16.0
38 2.4 76 21 42 18
8.7 o.o 8.6 o 9.0 1.6
36 84 43 38 4o 4i
2.3 - 0.75 - 8.2
132 366 120 300 138 182
24°C 28°C 27°C
11.2-11.5 11-5 11.0
54 54 54
200-300 200-300 150
175 175 175
None 10fc ICfc
NapCO., to
6A
Raw Clar
21 1.2
37 14.2
60 lU
8.1 0,0
43 45
5-7
134 296
26°C
11.5
54
200-300
175
lOfo
sec. zone
primary zone
-------
TABLE 2
BLUE PLAINS PILOT PLANT (CONTINUED)
Mode
Total Phosphorus
(mg/1 as PO^)
T.O.C. (mg/1 as C)
Suspended Solids (mg/l)
Mg++(mg/l as Mg)
Ca++(mg/l as Ga)
Turbidity (J.T.U. )
Alkalinity
(mg/1 as CaCCO
Temperature
PH
Flow (gpra)
Lime Dose (mg/1 as CaO)
Soda Ash Dose
(mg/1 as Na2CO~)
Recycle
Others
IB 2B
Raw Clar Raw Clar
18 1.5 23 1+-5
28 16.0 ho 23
1+2 1+9 60 53
8.0 0.0 9.0 1+.6
145 79 1+5 147
- 3.8 - 30
126 356 130 181;
2l+°c 2l+.5°C
11.5 11-0
5^ 5>+
200-300 150
225 225
None None
None None 5
3B
Raw Clar
2l+ 2.1+
28 15.6
50 5!+
9-5 o
1+2 56
10.7
152 17!+
2l+°C
11.0
^
150
225
None
mg/1 Fe+3
k~B 5B
Raw Clar Raw Clar
28 1+.9 26 5.5
1+3 21)- 1+1+- 20
lll+ 60 U8 32
0 2.9 17 2.1+
• 1+7 U6 7!+ 126
- 19.6 - 13-9
ll+O 238 170 36!+
23°C 22°C
11.5 11-7
5>+ 5^i
200-300 300-1+00
225 225
None None
None None
6B
Raw Clar
28 9.2
31 2h
118 97
8.8 2.5
52 16
- 25. U
130 35^
20°C
11.7
5U
300-1+00
225
None
5 mg/1 Mg
-------
charge of this unit, supplied us with phosphorus removal data (total P)
around the clarifier unit, both filtered and unfiltered. This data indicated,
as did that from Pomona, a strong dependence of P removal on pH. The specific
data points obtained will be shown in the section on correlations belcw.
SOUTH TAHOE, GAKLFORHIA
The treatment plant at South Tahoe has been amply described in the
literature'^). Briefly, it consists of a complete treatment facility including
primary, secondary, and tertiary processes. P removal is accomplished in
part by lime addition in the secondary (biological) step, and in part by
further tertiary precipitation. Steps following the P removal process are
ammonia stripping, filtration, and final polishing (activated carbon and
chlorination). The tertiary P removal is not done in a solids-contact type
unit. After lime addition, the reaction mixture is separately flocculated
and then held in a settling basin where solids separation occurs.
Mr. R. L. Gulp, General Manager, was contacted to obtain further informa-
tion with respect to P removal in the tertiary lime addition step. He reported
that optimum P removals were usually accomplished with a dose of about ^00
rag/1 lime, to a pH of 11.0. A polymer coagulant aid was also added after the
precipitation reaction but before the flocculation basin. Residual P levels
of about 0.15 mg/1 were obtained.
17
-------
CORRELATIONS
PERFORMANCE
The approach used in attempting to describe the expected performance of
a tertiary lime addition process for P removal has "been to arbitrarily
separate the chemical processes from the solids separation processes. As
explained in preceding sections of this report, this is undoubtedly a gross
oversimplification. However, in light of the limited information available,
this seems to be the most reasonable way to proceed. One should, of course,
be cognizant of the possible errors involved in the assumptions and extra-
polations which are used.
The procedure is to first determine the amount of all types formed by
the chemical reactions. No distinction is made as to particle size or
settleability. The sludge formed due to hardness, i.e., CaCOo, Mg(OH)2>
etc. is calculated on the basis of standard chemical equilibrium concepts.
As explained above, there appears to be no reasonable theoretical approach
to describing the calcium phosphate precipitation. Therefore, the amount of
calcium phosphate solids formed is derived on an empirical basis. Having
determined the total amounts of various solids in the reactor, these are all
handled equivalently in terms of physical separation, i.e., the fraction which
is settled vs. the fraction which escapes into the effluent.
Chemical Phosphorus Removal
In- examining the data concerning residual phosphorus levels, we have
concentrated primarily on filtered, rather than unfiltered, effluents, as
this gives a more realistic estimate of the conversion of P from soluble
to insoluble form, independent of the physical separation processes occurring
in the clarifier. (The distinction between solid and insoluble P, however,
is arbitrary; solid phosphates of size small enough to pass a 0.^5 micron
filter are classified as "soluble", in terms of laboratory analyses. Effluents
from plant filters will undoubtedly contain even larger particles, depending
on the filter pore size and efficiency.)
Orthophosphate
The most convenient parameter which appears to correlate with the
available data on residual soluble orthophosphate is pH. Figure 3 shows the
results of plotting filtered residual orthophosphate vs. pH, using data points
obtained from the Pomona unit, and those supplied us by Mr. D. F. Bishop for
jar tests run on the Blue Plains secondary effluent. While the Pomona points
are plant-filtered, the jar-test points are laboratory-filtered, using OA5
micron filters. The equation for a least-squares regression line for these
data is:
Log Portho (mg/1 as P) = 6.12 - (0.69*0 PH (5)
18
-------
Log P = 6.12 - 0.69^ (pH)
0.1
a o
0.01
A = Pomona
jO,* = Bishop (Blue Plains)
I" a Reported as 0.0
$ 8 J IJ
0.002
I I I
9-0
10.0
11.0
12.0
pH
FIGURE 3. Filtered Ortho-P vs. pH
-------
It is questionable, of course, whether the jar-test data points are valid in
terms of predicting plant performance, especially in view of the very fine
filters used. In other words, are the phosphate solids (> 0.^5ju ) predicted
by these data ultimately settleable even with the most efficient clarifier,
or is some fraction permanently destined to appear in the effluent stream as
colloidal material, regardless of the clarifier performance?
Total Phosphate
Residual total phosphate (ortho + condensed) values were also available
from Pomona, and Blue Plains, and, in addition, some points were supplied
by Mr. Edward Berg (Lebanon Lime Clarifier) and Mr. James Zornes of Nevada
Power Company. Again, these were plant-filtered values, except for the NPC
value, which was the effluent from a large settling basin, and for the Blue
Plains jar test values,which were 0.^5 ju-filtered. The points from Lebanon
are not raw data, but are taken from an already smoothed curve of P vs. pH.
These are plotted in Figure U. Again, a relatively linear correlation
with pH is obtained, the regression equation being:
Log Ptofcal (mg/1 as P) = 3-51 - (0.392) pH (6)
As in the previous case, we have no values at the higher pH's except
those of Bishop, which were 0.^5 jit-filtered. Again, therefore, it is not
clear how much reliance may be put on these figures, in the pH range between
11 and 12, in terms of typical plant performance. This is borne out to some
extent'by examination of the Blue Plains Plant effluent analyses (Table 2)
which show little, if any apparent pH dependence. On the other hand, it must
be pointed out that the Blue Plains Plant is somewhat atypical in that the
biological secondary process was quite unstable, leading to occasional high
levels of organic solids and polyphosphates in the effluent. In addition,
the Blue Plains water is extremely soft, which gave difficulty in maintaining
a dense sludge. Soda ash, in varying amounts, was added to this water to
increase the amount of sludge formed. For some of these cases, 5 mg/1 Mg
and/or 5 mg/1 Fe^+ were added, in an attempt to increase the efficiency of
coagulation. Mr. Bishop informed us that Fe3+ was satisfactory for a time,
but that the low density of the Fe(OH)o floe led to an increase in the
volume of the sludge zone and ultimately to high solids carryover due to "scour1
of the sludge by the water exiting the secondary reaction zone. Mr. Bishop
also told us that recycle of sludge had no appreciable effect on P removal.
As phosphate removal would appear to depend, to a large extent, on good
solids separation, it is not clear why sludge recycle should be ineffective.
P vs. pH Relationship
Examination of the relations obtained for ortho-P vs. pH, and total P
vs. pH indicates that ortho-P is removed more efficiently than condensed P.
20
-------
Pomona
Bishop (Blue Plains)
Berg (Lebanon)
Nevada Power Co.
South Tahoe
Log P = 3-51 - 0.392 (pH)
9.0
10.0
11.0
12.0
PH
FIGURE k. Filtered Total P vs. pH
21
-------
This is probably true, although, as the primary mechanism of removal of the
two species are different, i.e., precipitation and adsorption respectively,
this will depend to a large extent on the particular operating conditions.
Previous laboratory findings have indicated no significant difference between
the removal of ortho- and condensed P by lime.
Use of both relations (5) and (6) simultaneously will result in an
inconsistency; the difference between total and ortho-P (which is equal to
condensed P) rises in the pH region between 9 an<3- 10 • Because the total P
vs. pH relation is based on more actual operating data than the ortho-P
relation, and in order to be somewhat more conservative than the jar test
values in predicting ortho-P removal at the higher pH's, we have decided to
use the total P correlation only, and to assume that the ratio of ortho/
condensed P remains constant.
Calcium Carbonate Supersaturation
From the reported effluent analyses for Ca and alkalinity, it is possible
to calculate the apparent Ca2"1" and COo2" concentrations (no correction being
made for activity coefficient). Examination of aJmost all the data available
to us, both laboratory and plant, show a (Ca2+) (CO^2") concentration product
well in excess of the theoretical value, especially at high pH's.
Figure 5 is a plot of these calculated points, indicating a relatively
linear rise up to pH'-'9'75^ followed by a leveling off at a value of about
1.3 x 10~° (theoretical value at 25° is about 1 x 10~°). Such a large dis-
crepancy cannot be accounted for, even allowing for activity coefficient
corrections. This apparent supersaturation can be most easily described by
the relation:
Log K = -6.7 + (pH - 9) (1.082) (pH < 9-75) (7)
and K = 1.3 x 10~ (PH > 9-75) (8)
2+ 2-
where K is the apparent concentration product, (Ca ) (CO- ).
Effect of Temperature
The effect of temperature on chemical equilibrium constants is described
by the Van't Hoff Thermodynamic Equation (2l).
d (In K) „ AH°
dT 2
ET
where R = Gas constant, and
= Energy term.
22
-------
i r
8
10
-6
K
(Ca)(C03)
f
(moles/l)'
o o
oo A
10
-7
O = Pomona-Filtered Effluent
A = Bishop (Blue Plains)- Jar Tests
I I I I
9-0
10.0
11.0
PH
FIGURE 5. Apparent (Ca)(CO ) Solubility Product vs. pH
12.0
-23
-------
The energy term can be assumed constant in the small range of temperature,
from 5°C to 30°C. The integration of the Van't Hoff relation yields
or
log K = | + b
(22}
The values of a and b were calculated from literature datav ' and soften
ing plant data.
The relations between temperature and dissociation constants are given
in Reference (23).
Coagulant Dose
In the absence of formation of enough Mg( 011)2 to act as a coagulant and
flocculating agent, it may be necessary, depending on the water composition,
to add a coagulant. This might be one of a variety of materials including
both inorganic and organic compounds.
Not enough information is currently available to predict, on any
theoretical basis, the dose or type of coagulant which might be necessary;
this information is usually adduced from jar tests on the specific water being
treated'. However, Infilco normally recommends the addition of coagulant if
the Mg(OH)2 formed is less than 50 mg/1. Using this as a working basis, and
FeSOl| as a typical coagulant, we will set up a relation whereby the Mg(OH)o
"deficit" is made up by Fe(OH)g precipitate. On an equivalent molar basis,
then, each mg/1 difference between the Mg(OH)2 formed and 50 mg/1 will require
1.5^ mg/1 of Fe(OH)2-
CAPITAL COSTS AKD SIZIM>
The estimate of capital cost vs. treatment capacity was obtained on the
following basis:
Approximate selling prices for Densators of various sizes were obtained
from Infilco/Fuller. These prices were for the basic Densator Mechanisms,
including: mixing chamber, agitators and sludge recycle pump, but did not
include the costs of the external basins or other ancillary equipment. Also
obtained were the recommended external basin dimensions and net rise areas.
r)
Infilco recommends an overflow rate of 1.5 gal/min/ft in application
of this equipment to phosphorus removal by lime addition. Thus, one may
calculate the treatment capacity of each unit, as in Table 3-
-------
TABLE 3
DENSATOR PRICES VS. SIZE
Basin Diameter Basin Depth Wet Eise Area Approx. Selling Calculated
(ft) (ft) (ft2) Price ($) Capacity (mgd)*
20 12-1/2 250 1^,000 0.5^
UO 12-2/3 1068 25,000 2.3
60 lU 2280 U5,000 U.9
90 17-1/2 5120 70,000 11.0
120 18-1/2 9350 100,000 20.2
The approach used in estimating the costs of the construction and materials
for the external basins was as follows: For concrete basins, the necessary
wall thicknesses were calculated from the basin dimensions and water heights
by using the standard engineering formula:
where t is the minimum wall thickness in inches for a concrete tank of radius
r (inches) subject to a radial pressure p (psi) . A minimum thickness of 8"
was used for the smallest tank, and the values computed for the other tanks
were rounded to the next highest inch. The volumes of the concrete shells
were then calculated, together with a nominal 1 ft. thick tank bottom, and
multiplied by an estimated cost of $8l/cubic yard to obtain the costs for
the outer basins. Added to these were the estimated costs for wooden pouring
forms ($0.70/ft2 contact area) and excavation and backfill for setting the
basins U feet into the ground ($1.00/yd3 and $0.50/yd3, respectively).
A similar method was used for the steel basins. In this case, the
formula:
to 1/8) (10)
was used for calculating the wall thickness, with a minimum of 3/l6" for
the smaller basins.
_^_________—
* Note that, if the 1.5 gal/min/ft rise rate is not to be exceeded, these
are maximum capacities.
25
-------
The weight of the steel was then calculated (allowing 3% for overage)
and an estimated cost of $0.28/lb (erected in place) was used to compute the
total cost of the shell. Again, a 1 foot concrete floor was assumed for all
cases, and the basins were assumed to be set k feet into the ground.
Table 4 gives the results of these calculations, with the figures con-
servatively rounded:
TABLE U
ESTIMATED COSTS FOR DENSATOR INSTALLATIONS
Capacity
(mgd)
0.54
2.3
M
11
20.2
Capacity
(mgd)
0.514-
2.3
M
n
20.2
Shell
$ 1,650
U,Uoo
8,000
26, 500
33, 500
Shell
$ 1,700
3,500
7,550
21,250
ho, ooo
Bottom
$ 1,100
4,100
9,100
21, 000
37,000
Bottom
$ 950
3,800
8,500
19, 000
34, 000
CONCRETE BASIN
Excav .
Forms & Fill
$ 1,100 $ 100
2,500 300
3, 800 600
7,000 1,200
10, 000 2, 000
STEEL BASIN
Excav. & Fill
$ 100
250
550
1,100
1,900
Basin
Total
$ 4,000
11, 300
21, 500
56, ooo
102, 500
Basin
Total
$ 2,750
7,550
16, 600
^1,350
75, 900
Total w Densator
$ 18,000
36,300
66, 500
126,000
202,500
Total w Densator
$ 16,750
32,550
61, 000
ill, 350
175.900
It will be noted that the costs of the steel shells are somewhat lower
than for concrete; however, it appears that engineering practice favors
concrete over steel for basins larger than about 40 ft. in diameter (possibly
26
-------
because of problems in maintaining dimensional stability). We therefore
have used the costs for steel basins up to this size, and for concrete
basins for the larger sizes. These costs are plotted in Figure 6, together
with the second order polynomial function which was calculated for the
best least-squares fit:
Logo cost ($) = i.o/(o.099U85 - 0.005877 (Log o)) (11)
" c
where Q is the capacity of the unit in mgd.
The balance of the capital costs will also include ancillary equipment
such as lime and coagulant feeding mechanisms. Examination of previous cost
estimates for this type of system, as supplied by Infilco^1^"), indicate that
the chemical feeding and control equipment will in general constitute no
more than 3-5$ of the capital cost for the Densator installation. Therefore,
in the computer program, the capital costs derived above are multiplied by
1.05. Note that they do not include the cost of buildings, land, etc.
The capital costs are presented, both here and in the computer program,
without conversion into amortization costs; it is expected that the calcula-
tion will be performed by the FWPCA Executive Program, and appropriately
added to the total treatment cost.
It is of interest to compare the costs obtained above with those pre-
sented in previous estimates. The most recent estimate by Smith(15) appears
in a report issued in July 1968. The corresponding capital costs (in those
regions where the estimates overlap) are:
Capacity (mgd) Smith This Report
0.5 $ 50,000 $ 16,800
5 210,000 68,250
10 390,000 115,500
20 725,000 225,750
30 1,050,000 3^6,000
At first glance, there appears to be a discrepancy of about a factor
of three between the two sets of figures. However, if one examines the
references on which the first set of figures are basedC1^)^ one finds that
they are computed for average flows for the Blue Plains sewage treatment
plant, where the expected peak to average flow ratio is about 3- As stated
above, the figures derived for this report are for maximum capacity. With
this factor taken into consideration, the agreement is quite good.
27
-------
1000
100
Cost
($
10
T 1—T
Loge Cost ($) = 1.0/(0.099^85 - 0.005877(Loge q))
I I i I I I l—l
M
1.0
Capacity (mgd)
10.0
FIGUBE 6. Capital Cost vs. Capacity for Densators, not Including
Chemical Feed and Control Mechanisms.
28
-------
Examination of the cost vs. capacity curve shows that a capacity of
30 mgd is, economically,, about the maximum size for a single unit. For
capacities greater than this, it is probably more practical to have several
smaller units rather than one large unit.
OPERATING COSTS
In addition to the cost of chemicals (which -will be computed from the
chemical model), operating costs will include labor and power costs.
The cost of power has been estimated, again from figures supplied us
by Infilco, for the total connective horsepower associated with the mixing,
scraping, and recirculating mechanisms for the Densators of various sizes
as given above. Figure 7 presents these figures, together with the approxi-
mation for Power vs. Capacity derived from these points:
HP = j|j- x Q (Q, < 11 mgd) (12)
HP = li* + 0.06 (Q - 11) (Q > 11 mgd) (13)
Appropriate conversion factors may then be used to convert these into costs
per 1000 gallons treated:
FWRCO (power cost in $ per 1000 gallons) = HP x CKHH x 2h
x 0.7^6/(103 x Q) (I1*)
where CKWH is the cost of electricity in cents/kwh.
The cost of operating labor will depend in large measure on the location
of the installation with relation to other treatment processes. If it is
in an integrated treatment plant, then it is possible that operators will be
able to divide their time between Densator operation and other processes.
However, especially with a pilot-plant type installation, which may be in
a separate area, it may be necessary to keep operators on the site continually
even when their services are not wholly occupied with the Densator.
It would seem somewhat more reasonable to relate the operating labor to
the number of units to be attended, rather than to the plant capacity. (These,
of course, are related through the sizing rule given above.) For example,
it would probably take little, if any, more operating labor for a 10 mgd unit
than for a 1 mgd unit. Previous estimates, supplied by Infilco*-i^)(lb)f
indicate that at least 0.5 man per shift would be needed for one unit, while
for a multi-unit installation, the required labor appears to be about 1/10
man per shift per unit:
29
-------
16
(JO
O
Horse-
power
12 ~
10
8
T \
HP = — x Capacity
HP = lU + 0.06 x(Capacity-11
8 10
Capacity (mgd)
12 ib 16 18 20
FIGURE?. Horsepower vs. Capacity for Densators
-------
TABLE 5
OPERATORS PER SHIFT FOR MULTI -UMT INGIALLATICKG
Number of Units (ll)
1
23
32
Operators
u. ~j
2
3
This relationship may be approximated (Figure 8) by the equation:
(K)l-365
Operators per Shift = 0.5 + /A -3q ^5)
Converting this into treatment cost:
COIAB
COLKG =
[COIAB I
2600 x QINI
2500 x QIN|
where COLKG = labor cost (0/1000 gal)
QIN = plant flow (mgd).
1.3651
(16)
31
-------
10.0
5.0
Operators
Per Shift
Operators = 0.5 +
0.2
FIGURE 8.
5 10
Number of Units (w)
Operators Needed vs. Number of Units
100
-------
METHODOLOGY USED IN COMPUTER PROGRAM
On the basis of the above equations, correlations and assumptions, the
computer program (a listing for which is given in Appendix A) was derived.
A brief description of this program is presented below.
INPUT
The input stream characteristics are read in on three cards, in F10.3
format. These include QJN, SOC, SON, SOP, SFM, SBOD, VSS, TSS, DOC, DNBC,
DN, DP, DFM, ALK, DBOD, CA, MG, PH, AMO, and ORTO. (See Appendix for defini-
tions) .
It is possible that neither DP (dissolved phosphorus) nor DN (dissolved
nitrogen) are further classified as to ortho-P or ammonia -IJ . In this case,
the data fields for these (ORTO, AMO) are left blank.
The operating characteristics f.re then read in on two cards. These
include TSS1, PHF, SLDEN, COLIM, PGA, FMG, TC, CKWH, COLAB, PKTAV, and COFES.
If either one or both of the ORTO or AMO fields is empty, ORTO is set
equal to 0.9 x DP and/or AMO is set equal to '0.8 x DN. These are reasonable
assumptions, based on Neale'svlfJ survey of secondary waste effluent com-
positions .
CHEMICAL EQUILIBRIA
The ionization and solubility product constants for E^O, H^POij., H^POij. ,
HPOl|.2~, H2C03, HC03~, NHl}*, Mg(OH)2> Ca.CO$, and Fe(OH)2 are then calculated
at the operating temperature. The input concentrations of those species which
will be manipulated in the program are changed from mg/1 to moles/1.
In order to reasonably estimate the activity()f) coefficients of the various
ions, it is necessary to know the ionic strength of the solution. The activity
coefficient of an ion may be calculated from the Debye-Huckel approximation^") :
-0.508 z
1 + 0.328
where "if- is the activity coefficient of ion j,^i the ionic strength of the
solution, and r-i and z-; are, respectively, a constant associated with the ion,
and its charge. (The average experimental values of r.s are given in Ref . 2k.)
As the ionic strength of a solution is defined as/^ = i/2 £ z^2 Cj, where Cj
is the concentration of each ion j, and z^ its change, /icannot be accurately
calculated without a complete knowledge or the concentrations of all ions
present . Lacking this information, we make an initial guess of the ionic
33
-------
strength from the alkalinity (ALK) and dissolved fixed matter (DFM).
DFM, as described in Standard Methods(^9), is a measure of the soluble
residue after ignition to 600°C. We make the following assumptions:
a) During this ignition, about half the HCOo" present will be
lost as COg and H20 ^HCC^'-^HgO + COg + COg2'), yielding
30 mg fixed matter per mole "
-
b) Phosphates will be converted to pyrophosphates (P2°7 ) yielding
about 87 mg fixed matter per mmole of P.
c) Ammonium salts will decompose on ignition, leading to loss of
both MHl+ ion and the anion with which it is associated (average
d) About half the OH~ present will be lost by decomposition
(20H~— V HpO + 0 ), leading to 8 mg fixed matter per mmole
OH'.
e) The solution will contain, on the average^ ' U3 mg/1 SiO ,
which will decompose to yield 3*4- mg/1 Si02 residue. 3
f) The dissolved nitrogen which is not ammonia-nitrogen is
nitrate.
2- -
g) All orthophosphate is present either as HPCV or H^PO^ .
h) The average contribution of condensed phosphate to the alkalinity
is 20 mg/1 (as CaGO-) per mmole p(3).
i) The ionic strength due to Na , K , SOL2", and Cl" can be
approximated byjucrl.8 x 10~5 x mg/1 (see Ref. 25).
Using the above assumptions, it is possible to calculate the ionic
strength due to those ions about which no input information is given (Na ,
K+, S0l|2~, Cl~, etc.), and then to add the contributions from the known
ionic species.
Having estimated jj. , the activity coefficients for all pertinent ionic
species are then calculated. These, coupled with the ionization constants
and the input pH, allow calculation of the concentration ratios between the
various forms of orthophosphate, carbonic acid and ammonia.
The program next tests whether or not the solubility product constants
for CaCOo and Mg(OH)2 are exceeded in the input stream. This is unlikely,
but if such is the case, we assume that these precipitates are present in
colloidal, unfilterable, form. Thus, they will have contributed (erroneously)
-------
to the alkalinity, Ca, and Mg analyses on the input stream. ¥e, therefore,
solve for these quantities (COLCA and COIMG) and remove them from the "true"
ionic concentrations. A calculation is then made of the potential acidity
(POKE) of the input stream.
The final pH value (FHF) is then substituted for the initial pH, and
the amount of phosphorus remaining and precipitated is computed, using the
total-P vs. pH relationship described in a previous section. (A limit of
12.5 is put on the final pH, as this corresponds approximately to the maximum
solubility of limev^O).) The calcium associated with the phosphate precipi-
tate is "removed" from the total calcium in solution. A Ca/P ratio of 1.7,
based on the hydroxyapatite formula, is assumed for the orthophosphate
precipitate, while a Ca/P ratio of 0.5 is assumed for the condensed phosphate.
Control next reverts to a recalculation of the ionic strength, activity
coefficients, and polyprotic equilibria using the new (final pH values). The
program then bypasses the "colloid" and phosphate calculations, and proceeds
to an estimate of the amount of Mg(OH)2 which will be precipitated at the
final pH.
If the amount of Mg(OH)2 formed is less than 50 mg/1 (8.6 x 10~ moles/1),
a coagulant dose of FeSO^ is calculated which will result in the formation of
enough Fe(OH)2 to make up the difference. The potential acidity (POH) of
the final solution is then estimated, and from the values of Mg(OH)2,
Fe(OH)2 and POH, the hydroxyl demand and a corresponding test lime dose are
calculated. The Ca and Mg brought in by this dose are added to the solution,
and the amount of CaCOo precipitated is computed. An iteration process then
begins, in which the program goes back to recalculate new activity coefficients,
Mg(OH)2 and. Fe(OH)2 precipitates, final potential acidity and lime doses.
When two successive calculated lime doses are within 1/2$, of each other, the
iteration procedure is satisfied.
After leaving the iteration loop, the program calculates the final
equilibrium Ca, Mg, and carbonate concentrations. To conform to the apparent
supersaturation with respect to CaCOo observed at high pH's, some of the
CaC03 is made "colloidal" (GOLF) so as to yield a (Ca2+)(COo ) concentration
product consistent with these observations. The output CA and ALK analyses
are adjusted accordingly.
Later in the program, all chemical species are reconverted from moles/1
to mg/1 for readout.
The handling of both solid and dissolved output stream species is
relatively straightforward, with the exception of ISFM and SFM. Here, as
for the input fixed matter, correction is made for the decomposition of some
of the inorganic salts upon ignition:
Mg(OH)2 - * MgO + H20
Fe(OH) + 1/2 0
35
-------
20H
HEL + HX
etc.
SOLIDS
The total amount of solid sludge formed (SLUDG) is computed from the
sum of the CaC03, Mg(OH)2> Fe(OH)2 and calcium phosphate precipitates. To
these are added the input solids (TSS) to obtain the total solids (SOLID)
present in the reaction zone effluent.
As all solids are assumed to behave equivalently, specification of TSSl,
the suspended solids in the clarifier effluent, determines the ratio of
settleable to unsettleable solids of all types . Thus, all solid species
present are multiplied by TSS1/SOKED, which is the separation factor (SEEN)
describing the performance of the unit for solids removal. The products
obtained are put into output stream 1 (the clarified effluent).
In a similar fashion, specification of SLDEN, the solids content (mg/l)
of output stream 2, allows the calculation of SLDEN/SOLID = CONCN, which is
a concentration factor, describing the degree of sludge concentration obtained
in the unit. All solid species are multiplied by this factor to obtain their
concentrations in output stream 2 (the sludge waste).
FLOW RATES
The average flow rates in output streams 1 and 2 (Q^ and Q2) are calcu-
lated from simple mass balances, (it should be pointed out, of course, that
these average rates are not continuous. In normal operation, the clarifier
runs with no sludge waste for most of the time, the entire input flow exiting
through output stream 1, with intermittent periods of sludge waste, which
maintains the sludge inventory at a relatively constant level.)
Setting up a mass balance for solids, the mass of solids entering the
unit per day, plus the mass of sludge formed in the reaction per day, must
equal the total amount leaving per day.
i.e. (QIN)(SOLID) = (QI)(TSSI) + (Q^KSLDEN) (18)
OUT (SOLID) - Q-L (TSSl)
2 SLDEN
Making a volume balance for water, the volume of water entering must
equal the volume leaving:
OIF = Qr + f(Q2) (20)
where f = volume fraction of water in Q£.
36
-------
We have assumed here that for both QJN and Q , the solids contents are
so small that essentially all the volume is water. (We have also neglected
the water associated with the lime feed stream.) The sludge waste stream,
however, is usually quite high in solids and appropriate correction must be
made: Letting RSGCC equal the specific gravity of the solid sludge (gm/cm3),
then the volume fraction of solids in the sludge waste stream is
SLDEN x 10"
RSGCC
Therefore,the fractional volume of water in the sludge stream is
1 - SLDEN x 10~ /RSGCC, or 1 - fijiP
where RSMGL is the solid density in mg/1 (RSGCC x 10~ ). Setting the input
and output water volumes equal:
Q-L + Q£ U RSMGL'
Q = QIN . o (RSMGL - SLDEN} ( }
^1 ^ ^2 V RSMGL ' V ;
Substituting (22) in (19), and solving for Q_:
02 = QIN (SOLID - TSSI)(RSMGL)/(RSMGL)(SLDEN - TSSI)
+ (TSSI)(SLDEN) (23)
Q, may then be calculated from equation (21):
QI = QIN - Q2 (i - SLDEN/RSMGL) (2k)
The density of the solid sludge (RSGCC) is arbitrarily taken as 2.0.
This sludge will usually consist primarily of CaCOo, the density of which,
depending on its crystalline form, is quoted from 2.9 to 1.8. The 1.8 figure
is for the hexahydrate, which is probably the species which will predominate
in formation from aqueous solution. The density of solid calcium phosphate
is also not known precisely, although it is probably higher than that of
Mg(OH)p and Fe(OH)2» being gelatinous, will have rather low densities.
From SLDEN and RSGCC, the specific gravity of the sludge stream (R02) is
readily calculated:
R02 = i.o + (SLDEN)(RSGCC - I.O)/RSMGL (25)
Using the figures derived above, the program next computes the weight
and volume of sludge wasted per thousand gallons of water treated.
37
-------
COSTS AND SIZING
From the computed lime and coagulant doses and the chemical costs
(COLIM and COFE), the amounts of chemicals used per day and their cost per
thousand gallons are next calculated.
For sizing the plant, we need the ratio of peak to average flow, (it
will be remembered that the capital costs for Densators were derived on the
"basis of maximum flow.) As noted above, we have taken 30 mgd as the maximum
unit size.
To obtain the number of Densator units (NUMB), therefore, the program
multiplies the average design flow (QIN) by the expected peak to average flow
ratio (JKTAV) to obtain QMAX, and divides by 30 mgd, rounding off to the next
highest integer if there is a remainder. The capacity of each unit (QEACH)
is obtained by division of QMAX by NUMB. QEACH will of necessity fall within
the range of the capacity-cost correlation (0.5 - 30 mgd) and the formula
derived previously is then used to calculate the cost per unit (COSTE) and
the total cost (TCOST).
The electrical power is determined by calculating the power for each
unit (HPFER), and multiplying by the number of units which would be in
operation for average flow (NUMAV), to obtain total power (THP). Multipli-
cation by the cost of power, and appropriate constants, gives the power
cost per thousand gallons.
The labor cost per thousand gallons is calculated from the number of
units, the plant flow, and the operators salary, as described previously.
The program then prints the output data giving the stream parameters
for Qi and Qg, and the lime and coagulant doses, costs, sizing and sludge
waste characteristics.
38
-------
CONCLUSIONS AMD RECOMMENDATIONS
GENERAL CONCLUSIONS
The addition of lime to secondary effluents is an effective procedure
for reducing phosphorus concentrations. The residual dissolved phosphate
level declines as a function of the pH to which the waste-water is treated,
with about 0.15 mg/1 P remaining at a pH of 11 (Figure 2). It is more
correct to characterize the process by the value of the residual P level
rather than by % P removal, as the former seems to be insensitive to the
input P level.
An analysis of capital cost vs. size for a solids-contact precipitator
indicates that 30 mgd is about the maximum economical size for this type
of apparatus (Figure 6), beyond which it is desirable to use a multi-unit
installation.
The bulk of the total treatment cost will be dictated by the chemical
dose, which in turn is dependent on the chemical composition of the water
being treated. Low alkalinity waters will require less lime to reach a given
pH; however, the reduced amount of CaC03 sludge formed may result in reduced
efficiency for removal of precipitated phosphate. Waters low in Mg may also
require addition of coagulants, due to the absence of precipitated Mg(OH)2.
In view of the minimal amount of data on which the model is based, there
are several limitations which must be recognized in employing it. These are
noted below.
LIMITATIONS OF THE MODEL
P vs._ pH Relationship
The most prominent theoretical drawback to.the correlation between
residual P and pH used in this model is that it takes no account of the calcium
concentration. In theory, one could hypothesize a Ca concentration of zero,
and yet the model will still predict phosphorus removal (as calcium phosphate).
In practice, however, since we are restricting the model to a lime addition
process, this will never be the case; calcium is continuously added to the
system by the lime, so that by the time the pH is in the range where the P
removal process would be operative, there is always a supersaturation with
respect to calcium phosphate.
This can be demonstrated by worst-case calculation to find the minimum
Ca concentration which would result from lime addition. Figure 9 shows the
calculated final Ca concentrations to be expected by addition of lime to water
having an initial Ca concentration of zero, an initial pH of 7.0, and various
initial bicarbonate alkalinities (from 50-^-00 mg/1 as CaC03). The UOO mg/1
case is well in excess of any expected alkalinity.
39
-------
(moles/I)
10
11.0
PH
12.0
FIGURE 9. Minimum Ca Concentrations vs. pH.
-------
For the low alkalinity waters, there is an initial increase in Ca
content, followed by a decrease (as CaCOo is precipitated) and then followed
by a sharp increase again when the carbonate is exhausted. It can be seen
that the lowest Ca concentration is obtained with the highest alkalinity
water. Even this low calcium level, however, is sufficient for supersatura-
tion with respect to C&^OE^^O^}^, as it occurs at a Ph high enough so
that: (a) there is a high OH concentration, and (b) a larger fraction of the
orthophosphate present is in the form of (POj.) .
Figure 10 illustrates this point. Here, we have plotted the "maximum"
equilibrium values of total orthophosphate consistent with the theoretical
solubility product of Ca(OH)2 (K)i,.)g, and the Ca values shown in Figure 9
for the worst case alkalinity of ?00 mg/1.
Thus, while calcium ion concentration is not explicitly included in the
P vs. pH relationship, we may be sure that there is always enough present to
achieve supersaturation with respect to the precipitation of phosphate.
It should also be borne in mind that the P vs. pH correlation is based
primarily on points taken in the pH region from 9-5-11.0. Extrapolation
outside these limits should be made with care.
Solids Separation
Another assumption which has been made, and for which there is no con-
firmatory evidence, is that all types of solids (i.e., organics, phosphates,
CaC03j, etc.) are separated to the same degree by the clarifier. Because of
the variations in size, density, etc. of these solids, this is unlikely;
however, there is at present not enough information available to make any
other assumption for the model.
In constructing the model, we have also been forced to skirt the issue
of the physical separation process itself. Rather than predicting clarification
performance, the model takes as an input TSS1, (the effluent suspended solids),
thereby specifying performance. Again, this step has had to be taken due to
the lack of any basic knowledge of the actions of a host of physical and
chemical parameters on the processes of coagulation, flocculation and sedi-
mentation. While there are many theoretical analyses and discussions of these,
processes available in the literature, one would be hard pressed to apply
them to the case under consideration, lacking such information as the precipi-
tate particle size distributions, densities of various precipitates, relation
between "nominal" rise rates and actual liquid velocities in various parts
of the sedimentation zone, etc.
Siz:
In line with the above comments, a word should be said about the use of
the nominal rise rate of 1.5 gal/min/ft2 in sizing the apparatus. This figure
has been quoted as one which will yield efficient operation with minimal
solids carry-over. However, there has been little, if any, work done on relating
-------
10
-3
11.0
-1
Minimum Ca
(ALK =
10
12.0
PH
FIGUBE 10. Theoretical P and Ca Concentrations vs. pH.
-------
solids separation efficiency to residence times and rise velocities. If
the lime addition process is followed by a filtration step, it might well
be more economical to work at a higher throughput rate, sacrificing clari-
fication efficiency, but decreasing the necessary apparatus size and capital
cost. With no estimate of this relationship, however, it is impossible to
predict whether or not such a trade-off would be feasible.
Coagulant
As noted above, FeSOl^. was taken as a typical coagulant, and its dose
predicted on the basis of obtaining an arbitrary minimum concentration of
combined Mg(OH)2 and Fe(OH)2 to aid in flocculation. Many other materials
may also perform this function, and this report should not be interpreted
as indicating that this is the coagulant of choice. With a Kgp for Fe(OH)2
of ** 10"1 , Fe2+ ion is efficient in forming a hydroxide floe at high pH's.
However, at lower pH's (c*9-9-5) only a portion of the added Fe2+ ion will
form the hydroxide, leading to economically prohibitive calculated doses in
order to obtain the minimum specified amount of Fe(OH)2. This limitation
should be recognized, and the possibility of using other materials considered.
For example, on a cost per atom basis, ferric sulfate is much more expensive
than the ferrous salt. However, the higher charge on the ferric ion makes
it a more effective coagulant, while the much lower Kgp for the hydroxide
(~ 10-33) means that, even at low pH's, essentially all the added Fe3+ will
form Fe(OH)~ floe.
SUGGESTIONS FOR FURTHER WORK
The above assumptions and limitations of the program, together with the
reasons for them, are indicative of the areas where further work (both
theoretical and applied) should be done.
It would be of interest, for example, to examine the compositions of
both the effluent solids and the waste solids produced by an operating unit
to see whether all types of solids are settled in the same ratio, and, if not,
what the partition factor is for each type.
A study of the relationship between effluent solids and rise rate would
also be of value in obtaining a more fundamental rationale for predicting
effluent TSS.
In light of the marked variation in behavior with respect to P removal
between soft waters such as at Blue Plains, and hard waters such as at Lebanon
or Pomona, further study should be given to the effects of sludge density,
Mg(OH)2 and coagulants, on the P removal process. As Mr. Bishop at Blue
Plains has obtained excellent removals in filtered jar test effluents, the
problem appears to be associated with the solids separation process.
-------
As more such tertiary treatment units cone into use, more operational
experience will "be gained, and it is likely that more relevant experimental
work will be done in this field. The increase in the amount of information
available should then allow the development of a more generalized and complete
model of the process than has been possible based on the present limited
amount of data.
-------
APPENDIX A
COMPUTER PROGRAM LISTING
SAMPLE PRINTOUT AND DEFINITION OF SYMBOLS
-------
// FOR
*IOCS(CARD.1132PR INTER.DISK)
*BXTENDED PRECISION
*ONE WORD INTEGERS
*LIST ALL
REAL MU»MURES,MG,MGOH2
Y(R»Z»U)=10.0**<<-0.508*Z**2*U)/(1.0+0.328*R*U))
10 FORMAT(8F10.3)
20 FORMAT (1H1»37X»26HP REMOVAL BY LIME ADDITION»///•1HO.45Xt
1 9HCASE NO. »I2»/.1H0.45X,12HI!MPUT STREAM)
21 FORMATt1HO»9X»1HQ»10X»3HSOC»9X.4HSN3C»8X.3HSON»9X»3HSOP»9X.3HSFM.
1 8X»4HSBOD»/»3X»7F12.4//9X,3HVSS»9X,3HTSS»9X,3HDOC»9X.4HDMK»9X»
2 2HDN»10X»2HDP»9X*3HDFM/»3X»7F12.4//9X»3HALK»9X»4HDBOD,VX.2HCA.
3 10X»2HMG»10X»2HPH»9X»3HAMO»7X»7HORTHO-P»/«3X*7F12.4»//}
22 FORMAT (1H0.44X»23HPROCESS CHARACTERISTICSt )
23 FORMAT(1HO»7X»4HTSS1»9X»2HPH.8X.5HSLDEN.8X»5HCOLIM»8X»3HFCA.9X.
1 3HFiMG»8X»4HTEMP/3X»7F12.4//9X»4HCKWH»7X»5HCOLAB.7X»5HPKTAV»7X»
2 5HCOFES/3X.4F12.4//)
24 FORMAT (1HO»44X»15HOUTPUT STREAM 1)
25 FORMAT (iHl»44X,15HOUTPUT STREAM 2)
26 FORMAT (1HO»20X»' ORTHO-P NOT GIVEN.ASSUMED TO BE 0.9*DP.'»/)
27 FORMAT(1HO»20X»' AMMONIA NITROGEN NOT GIVEN.ASSUMED TO BE 0.8*DN'J
31 FORMAT (1HO»10X»' COSTS»SIZING*AND A'ASTE'»//»' NO. OF UNITS = '»
1 I4»//»' CAP COST PER UNIT = SF10.3,1 ^DOLLARS'•//»
2 ' TOTAL CAP COST = '.F10.3.1 ,
-------
EH3P=10.0»*< (-799.31/TO+4. 5535-10. 013486*TKJ >
EH2P=10.0**< <-1979.5/TlO+5»354l-(0.01984*TM )
EH1P=4.8E-13
EH2C=10.0**( (-3404.71/TKJ+14.8435-(0.032786*TK) )
EH1C = 10.0**( (-2902.39/TK)+6.498-(0.02379*TK) )
EMG=10.0**(1250.0/TK-15.145)
ECACO=10.0**(483.0/TK-9.68)
ENH4=10.0**{ (-2835.76/TK)+0.6322-(0.00122b*TK;j )
EFE=3.0E-14
CA=CA/40080iO
MG=MG/24310.0
ALK.=ALK;/1.0E5
ORTO=ORTO/ 30974.0
POLY=POLY/30974.0
TPHO=ORTO+POUY
AMO*/1 MO/14007.0
H=10.0**(-PH)
UVW=2.0*EH1C+H
XYZ=EH2P+H
QRS=EMH4-*-H
MURES=(1.8E-5)*DFM-1.08*AL<-ORTO*(1.57-U.54*EH2P/XYZJ-1.35*POLY
-0.616*DN/ 14007. J-0. 72 1*CA-U. 438* MG
= 2.0*ALii;*H/UVW+EH20/H*2.0*EHlC/UVW+ORTO*H/UVW*(UVW-EH2P) /XYZ
1 +AMO*H/UVW*(UVW-ENH4) /QRS-0.4*POLY/UVW*H+H
DIV=2.0*ALK/UVW*EHlC-tH20/M*EHlC/UVW+OkTO*(tHlC^H)/UVW*EH2P/XYZ
1 +POLY*(l.?*EHlC+H)/(2.0*UVW)-AMO*CHiC/UVW*EiNH4/QRS+CA+MG
TRIV=ORTO*EHlP/tEHlP*EH2P-i-EHlP*H*H**2)*EH2P
GO TO 101
100 UNIV = CAK*AL7/AL6+H+OH+AM4+OKTO*AL4/ ( AL2*AL3 )
DIV=MG+CA+CAR*AL7-i-ORTO*AL4/AL3+0.5*POLY+FEi-FES04
TRIV=AL4*ORTO
101 MU=MURES+0.5*UNIV+2.0*DIV+4.5*TRIV
U=SQRT(MU)
YH=Y(9.0»1.0»U)
YOH=Y(3.5»1.0»U)
YHC03=Y(4.25,1.0.U)
YC03=Y(4.5»2.0»U)
YCA=Y(6.0»,2.0»U)
YMG=Y(8.0.2.0»U)
YH2P=Y(4.25»1.0»U)
YP04=Yl4.0t3.0»U)
YHP=Y(4.0»2.0»U)
YNH4=Y(2.5.1.0,U)
YFE=Y(6.0.2.0»U)
AH=10.0**(-PH)
H=AH/YH
OH-EH20/(AH*YOH)
ALl=EH3P/(AH*YH2P)
AL2=EH2P*YH2P/ ( AH*YHP )
AL3=EH1P*YHP/ IAH*YP04)
AL5=EH2C/(AH*YHC03»
AL6=EH1C*YHC03/ ( YCU3*AH )
AL7 = /»L5*AL6/ { 1.0+AL&+AL5*AL6
AL8=ENH4*YNH4/AH
AL9=1.0/(l.O-i-AL81
A-2
-------
IFUNT-2) 700*100*800
700 COLMG=MG-EMG/(YMG*(YOH*OH)**2)
IF(COLMG) 710»710»720
710 COLMG=0.0
720 ALKCO=ALK-AL4*ORTO-0.5*AL4*ORTO/AL3-0.2*POLY-0.5*OH-COLMG-0.5*AM3
XYZ=1«0+1.0/(2.0*AL6)
UVW=ECACO/**2>
FES04=FEOH2+FE
143 DOSE=0.5*(POHI+2.0*MGOH2+2.0*F£OH2+0.33*SORTO-POH)
IF(DOSE) 150»150»160
150 DOSE=0400001
160 ABC=40.3*FCA+56.08*FMG
CA=BCA+DOSE*40.3*FCA/ABC
AMG=BMG+DOSE*56.08*FMG/ABC
MG=A!MG-MGOH2
CAR=BCAR
XYZ=4.0*ECACO/(AL7*YCA*YC03 )
XYZ=(CA-CAR)**2+XYZ
XYZ=SQRT(XYZ)
CAC03=(CA+CAR-XYZ)/2.0
IF(CAC03) 220*220,500
220 CAC03=0.0
GO TO 500
300 POHI=ORTO*AL4*(3.0/(AL2+AL1)+2.0/AL2+1.0)/AL3+CAR*AL7*
1 (1.0+2.0/AL5)/AL6+H-OH+AM4
IP(PHF-12.5) 310*310*320
310 IF(PHF-PH) 330*330*340
330 PHF=PH
GO TO 340
320 PHF=12.5
WRITE (3*321)
340 PH=PHF
XYZ=3.512-0.3924*PH
STPHO=TPHO-10.0**XYZ/30974.0
IF(STPHO) 430*430*440
430 STPHO=0.0
440 SORTO=STPHO*ORTO/TPHO
BCA=CA-COLCA-1.7*SORTO-0.5*(STPHO-SORTU)
CA=BCA
A-3
-------
FES04=0.0
FE=0.0
TPHO=TPHO-STPHO
ORTO=ORTO-SORTO
POLY=TPHO-ORTO
AH=10.0**(-PH>
H=AH/YH
OH=EH20/(AH*YOH)
GO TO 100
500 CONTINUE
CAR=BCAR-CAC03
CA=CA-CAC03
IF (ABS< DOSE-DOS )-0.005*DOSE) 600.600.510
51JD DOS=DOSE
GO TO 100
600 MGOH2=AMG-EMG/< (YOH*OH)**2)
IFJMGOH2) 610.610.611
610 MGOH2=0.0
611 MG=AMG-MGOH2
MGOH2=MGOH2+COLMG
C03=AL7*CAR
IF(PH-9.75) 901.902.902
901 CKSP=10.0**(-6.7-KHH-9.0)*U.813/0.75)
GO TO 903
902 CKSP=1.3E-6
903 IF(CKSP-CA*C03) 900.900.910
900 COLF*0,0
GO TO 913
910 UVW=CA+C03
XYZ=UVW**2
COLFa0.5*(-UVW-i-SQRT ( XYZ+4»0* ( CKSP-CA*C03 ) ) )
IF (COLF-CAC03-COLCA) 911»911»912
911 CAC03=CAC03+COLCA-COLK
GO TO 913
912 COLF=CAC03+COLCA
CAC03=0.0
913 CA=CA-fCOLF
AL<=COLF+C03*
-------
= SNBC*SEPiM
SON1=SON*SEPIM
SOP1= +TSS1*SLDEN
Q1=QIN-Q2*(1.0-SLDEN/RSMGL)
R02»1.0+SLDEN*(RSGCO1.0)/KSiMGL
CONCN=SLDEN/SOLID
SOC2 = SOC*COiNCN
SNBC?=SNBC*CONCN
SON2=SON*CONCN
SOP2= ( SOP+STPHO ) »CUNCN
SFM2=SFM1#(CONCN/SEPN)
SBOD2=SBOD*CONCN
VSS2=VSS*CONCN
TSS2=SLDEN
SVPKG=Q2*1000.0/QIN
WTLIM = DOSE*8.3457»QIN/2000.0
CHEMC=WTLIM*COLlM/( 10.0*QIN)
WTFES=FES04*8. 3457*0 IN/2000.0
COAGC=WTFES*COFES/< 10«0*OIN)
QMAX=QIN*PKTAV
NUMB=QMAX/30«0
NUMB=NUMB+1
OEACH=QMAX/NUMB
XYZ=ALOG(QEACH)/2.303
COSTE=EXP( l./(0. 099485-0. 005877#ALOG(QEACH) ) J/1000*
COSTE=COSTE*1.05
TCOST=COSTE*NUMB
COLKG=COLAB/(2600.0*QIN)*<1.5+(NUMB**1.351)/14.96)
NUMAV=QIN/QEACH
NUMAV=NUMAV+1
IF (QEACH-11.0) 679»679.680
679 HPPER=14.0*QEACH/11.0
GO TO 681
680 HPPER=14.0+0.06*(QEACH-11.0)
681 THP»HPPER*NUMAV
PWRCO=THP*CKWH*24.0*0»746/(1000,0*QIN)
WR'ITE(3»24)
WRITE(3»21) Q1.SOC1»SNBC1»SON1,SOP1,SFM1,SBOD1»VSS1»TSS1»DUC»
1 DNBC»DN»TPHO»DFMtALK»DBOD»CA»MG»PH»AMO»ORTO
WRITE(3»25)
WRITE(3»21) 02 »SOC2»SNBC2tSON2fSOP2,SFM2.SBOD2.VSS2,» TSS2.DOC.
i DNBC»DN»TPHO»DFMt ALK»DBOD»CA»MG»PH»AMO»ORTO
WRITE (3.31) NUMB»COSTE.TCOST.DOSE»WTLIM,CHEMC.COLKG
WRITE (3.32) FES04.WTFES.COAGC.PWRCO.R02.SWPKG.SVPKG
NCASE=NCASE-H
GO TO 998
999 CALL EXIT
END
-------
P REMOVAL BY LIME ADDITION
CASE NO. 4
INPUT STREAM
Q
250.0000
VSS
21.5260
ALK
265.4700
SOC
9*0450
TSS
24.7000
DBOD
2*9590
SNBC
4.0470
DOC
12.5820
CA
50.0000
SON
1.4940
DNBC
11.0000
MG
25.0000
SOP
0.0900
ON
21.0000
PH
7.8500
SFM
3.1740
DP
5.4170
AMO
o.uuoo
SBOD
10.0460
DFM
500.0000
ORTHO-P
0.0000
PROCESS CHARACTERISTICS
TSS1 PH SLDEN COLIM FCA FMG TEMP
25.0000 11.0000 300000.0002 20.0000 0.9800 0.0100 25.0000
CKWH
3.0000
COLAB
8000.0000
PKTAV
3*0000
COFES
25*0000
ORTHO-P NOT GIVEN.ASSUMED TO BE 0.9*DP.
AMMONIA NITROGEN NOT GIVEN.ASSUMED TO BE 0.8*DN
OUTPUT STREAM 1
Q
249,6430
SOC
0.4275
SNBC
0.1913
SON
0.0706
SOP
0.2529
SFM
23.0604
SBOD
0.4749
VSS
1.0176
TSS
25.0000
DOC
12.5820
DNBC
11.0000
DN
21.0000
DP
0.1568
DFM
46d.5456
ALK
199*6510
DBOD
2.9590
CA
60*7238
MG
0.2746
PH
11.0000
AMO
16.8000
ORTHO-P
0.1412
A-6
-------
OUTPUT STREAM 2
Q
0.4-198
VSS
12211.2379
ALK
199.6510
soc
5131.0344
TSS
300000.0002
DBOD
2.9590
SNBC
2295.7762
DOC
12.5820
CA
60.7238
SON
847.5141
DN8C
11.0000
MG
0.2746
SOP
3035.0016
DN
21.0000
PH
11.0000
SFM
276725.4190
OP
0.1568
AMO
16.8000
SBOD
5698.8802
UFM
468.5456
OKTHO-P
0.1412
COSTStSIZING.AND WASTE
NO. OF UNITS = 26
CAP COST PER UNIT = 294.091 KDOLLARS
TOTAL CAP COST = 7646.373 KDOLLARS
LIME DOSE = 266.778 PPM
LIME USED PER DAY = 273.307 TONS
LIME COST = 2.226 CENTS PER KGAL
LABOR COST = 0.085 CENTS PER KGAL
COAGULANT DOSE = 0.000 PPM FES04
FES04 USED PER DAY = 0.000 TONS
FES04 COST = 0.000 CENTS PER
-------
INPUT -OUTPUT*
Symbol
ALK
AM0
ATDS
GA
CHEMC
C0AGC
C0FES
C0LKG
C0LIM
C0LAB
C^STE
CKWH
DFM
DN
DWBC
D0C
D0SE
DP
FCA
FES01*
DEFIICTTION OF COMPUTER PROGRAM SYMBOLS
Definition
Alkalinity as CaCO,, (mg/1)
Ammonia nitrogen (mg/1 as N)
Dummy variable for incrementing TDS
Calcium (mg/1 as Ca)
Cost of lime per 1000 gallons treated
Cost of FeSO^ per 1000 gallons treated
Cost of FeSO^ ($/ton)
Cost of labor per 1000 gallons treated
Cost of lime ($/ton)
Operators salary ($/yr) (260-day year)
Capital cost of each Densator unit
Power cost (^/kwh)
Dissolved BOD (mg/1)
Dissolved fixed matter (mg/l)
Dissolved nitrogen (mg/1 N)
Dissolved nonbiodegradable carbon (mg/1 C)
Dissolved organic carbon (mg/1 C)
Lime dose (mg/1)
Dissolved phosphorus (mg/1 P)
Fraction CaO in lime
dose (mg/1
*Note; The measurement units given are those in which these quantities
appear on input and output. They may be handled in other units
internally in the program.
A-8
-------
FMG Fraction MgO in lime
MG Magnesium (mg/1 Mg)
WCASE Case number
NUMB Number of Densator units
0RT0 Ortho-phosphate (mg/1 P)
HI pH of influent
PHF Final pH
FKTAV Peak to average flow ratio
PWRC0 Power cost per 1000 gallons treated
QJN Input flow rate (mgd)
Ql Clarifier effluent flow rate (mgd)
Q2 Sludge waste flow rate (mgd)
R02 Specific gravity of sludge stream (mg/l)
SB0D Influent solid BOD (mg/1)
SB0D1 Solid BOD in effluent stream 1 (mg/l)
SB0D2 Solid BOD in effluent stream 2 (mg/l)
SFM Solid fixed matter in .influent (mg/l)
SFM1 Solid fixed matter in effluent stream 1 (mg/l)
SFM2 Solid fixed matter in effluent stream 2 (mg/l)
SNBC Solid nonbiodegradable carbon in influent
(mg/l C)
SHBC1 Solid nonbiodegradable carbon in effluent
stream 1 (mg/l C)
SKBC2 Solid nonbiodegradable carbon in effluent
stream 2 (mg/l C)
S0C Solid organic carbon in influent (mg/l C)
S0C1 Solid organic carbon in effluent stream 1
(mg/l C)
A-9
-------
S0C2 Solid organic carbon in effluent stream 2
(mg/l C)
S0N Solid organic nitrogen in influent (mg/1 N)
S0H1 Solid organic nitrogen in effluent stream 1
(mg/1 N)
S0N2 Solid organic nitrogen in effluent stream 2
(mg/1 N)
S0P Solid organic phosphorus in influent (mg/1 P)
S0P1 Solid organic phosphorus in effluent stream 1
(mg/1 P)
S0P2 Solid organic phosphorus in effluent stream 2
(mg/1 P)
SLDEN Sludge stream solids (mg/l)
SVPKG Sludge volume per 1000 gallons treated (gals)
SWPKG Wet sludge weight per 1000 gallons treated (ibs)
TC0ST Total capital cost of Densators (thousands of
dollars)
TC Operating temperature (°C)
TSS Suspended solids in influent stream (mg/l)
TSS1 Suspended solids in effluent stream 1 (mg/l)
TSS2 Suspended solids in effluent stream 2 (mg/l)
VSS Volatile suspended solids in influent (mg/l)
VSS1 Volatile suspended solids in effluent stream 1
(mg/l)
VSS2 Volatile suspended solids in effluent stream 2
(mg/l)
WIPES Weight FeSO^ used per day (tons)
WTLIM Weight lime used per day (tons)
A-10
-------
INTERNAL*
AH Hydrogen ion activity
AL1 Concentration ratio HpPO, ~/H_PO.
AL2 Concentration ratio HPO^'/H pPCv"
AL3 Concentration ratio KV /HPO. ~
AlA Concentration ratio PO^ /Total 0-phosphate
AL5 Concentration ratio HCO_~/HpCO_
P— ••
AL6 Concentration ratio CO- ~/HCO ~
o_
AL? Concentration ratio CCL "/Total carbonate
AL8 Concentration ratio NH-/NHu
AL9 Concentration ratio NHL /Total ammonia N
ALKC0 Alkalinity due to HCO ~, CO-2"
AM3 HH-, concentration (moles/l)
HH^ concentration (moles/l)
BCAR Original value of carbonate
BMG Original value of magnesium
CAC03 CaCO^ precipitate
CAR Total carbonates (moles/l)
p
CKSP Bnpirical K^-. for CaCO, (moles/l)
or o
C0LCA Influent CaCO- supersaturation (moles/l)
Note: Those symbols for which no measurement units are given are
either dimensionless, or appear in different units at various
points in the program.
A-ll
-------
Effluent CaCCX, super saturation
C0IMG Influent Mg(OH)? supersaturation (moles/l)
C0NCN Solids concentration factor
n_
C03 CO.- concentration (moles/l)
DIV Concentration of divalent ions (moles/l)
ECAC0 Kc_. for CaC00 (moles/l)2
or o
EH1C lonization constant for HCO_ (moles/l)
EH2C lonization constant for HpCO-, (moles/l)
p —
EHiP lonization constant for HPO^ " (moles/l)
EH2P lonization constant for HpPOj, (moles/l)
EH3P lonization constant for HJPCV (moles/l)
2
EH20 loniaation constant for H?0 (moles/l)
EFE Solubility product for Fe(OH)2 (moles/l)
O
EMG Solubility product for Mg(OH)p (moles/l)
ENH^- lonization constant for KHr (moles/l)
FE0H2 Fe(OH)2 precipitate
PI
FE Fe concentration (moles/l)
H H concentration (moles/l)
HPPER Horsepower per Densator unit
INT . INT = 1 for influent, INT = 2 for effluent
M30H2 Mg(OH)2 precipitate
2-
MLTRES Ionic strength .due to initial SO^ ,
Cl", NOo", Na+, K
MU Total ionic strength
NUMAV Average number of Densators in use
OH OH concentration(moles/l)
A-12
-------
P0H Effluent potential acidity {moles /I )
P0H1 Influent potential acidity (moles /I )
P0LY Poly-phosphate concentration
QMAX Maximum anticipated flow (mgd)
QEACH Capacity of each unit (mgd)
QRS Dummy variable
ESGCC Specific gravity of dry sludge solids
(gm/cm3)
RSMGL Density of dry sludge solids (mg/l)
carryover total solids
r,T^n,r « -i-j i.- -oi. f after clarifier N TSS2
SEPN Solids separation factor = (•: — .,.,,,, - r-rr- — )=^ . .•-,
initial total solids Solid
S0LID Total solids in reaction zone (mg/l)
S0RT0 Ortho-phosphate precipitate
STPH0 Total phosphate precipitate
SJjUDG Total precipitate (mg/l)
TDS Total dissolved solids
THP Total horsepower for all Densators
TPH0 Total phosphate in solution
TK Temperature (°K)
TRIV Concentration of trivalent ions (moles/l)
U Square root of ionic strength
Concentration of univalent ions (moles/l)
Dunamy variable
Dummy variable
2+
YCA Activity coefficient of Ca
2+
YFE Activity coefficient of Fe
YH Activity coefficient of H
A-13
-------
2-
YHP Activity coefficient of
YH2P Activity coefficient of
YHCJ03 Activity coefficient of HCO ~
2_
YC03 Activity coefficient of CO-
Y0H Activity coefficient of OH"
Activity coefficient of ML
2+
YM3 Activity coefficient of MG
•3_
YP01! Activity coefficient of PO..
A-lU
-------
APPENDIX B
REFERENCES
-------
REFERENCES
1. Van Wazer, J\ R., Phosphorus and Its Compounds, Volume I, Interscience
Publishers, Inc.. New York, (195b).
2. Clark, J. S., "Solubility Criteria for the Existence of Hydroxyapatite",
Can. J. Chem'. 33, 1696 (1955).
3- Schiaid, L. A., "Optimization of Phosphorus Removal with Lime Treatment",
Ph.D. Thesis, University of Kansas, (1968).
U. Sawyer, C. N., "Some New Aspects of Phosphates in Relation to Lake
Fertilization", Sewage and Ind. Wastes 2k, 768 (1952).
5. Owen, R., "Removal of Phosphorus from Sewage Plant Effluent with Lime",
Sewage and Ind. Wastes 2*4-, 768 (1953).
6. Lea, W. L., G. A. Rohlich, and ¥. S. Katz, "Removal of Phosphates from
Treated Sewage", Sewage and Ind. Wastes 26, 26l (195*0.
7. Henriksen, A., "Laboratory Studies on the Removal of Phosphates from
Sewage "by the Coagulation Process", Schweiz, Zeitschr, fur Hydrologien,
2l+, 253 (1962).
8. Malhotra, S. K., G. F. Lee, and G. A. Rohlich, "Nutrient Removal from
Secondary Effluent by Alum Flocculation and Lime Precipitation", Int.
J. Air Wat. Poll., 8, ^8? (196*0.
9. Gulp, R. L., "Wastewater Reclamation at South Tahoe Public Utilities
District", J.A.W.W.A., 6_0, Qk (1968).
10. Rudolfs, W., "Phosphates in Sewage and Sludge Treatment, II. Effect on
Coagulation, Clarification and Sludge Volume", Sewage Works Journal,
19_, 179 (19^7).
11. Malhotra, S. K., "Nutrient Removal from Secondary Effluent by Alum
Flocculation and Lime Precipitation", Ph.D. Thesis, University of
Wisconsin, (1963).
12. Weinberger, L. W., "Waste Treatment for Phosphorus Removal", Lake
Michigan Enforcement Conference (Feb. 1968).
13. Bishop, D. F., "Status and Outlook for Phosphorus Removal from Waste-
Waters", FWPCA, (Sept. 1967).
lh. Smith, R., "Capital and Operating Cost Estimates for Phosphate Removal
at the Washington, D. C. Blue Plains Sewage Treatment Plant", FWPCA,
Memo to Record (Aug. 1966).
B-l
-------
15. Smith, R., .. "Cost of Conventional and Advanced Treatment of
Waste Waters", FWPCA, (July 1968).
16. Smith, R., "Status of Cost Information of Phosphate Removal",
FWPCA, Memo to Record, (July 1967).
17. Neale, J. H., "Advanced Waste Treatment by Distillation", AWTR-7,
U. S. Public Health Service (196U).
18. Klotz, T. M., Chemical Thermodynamics, Prentice-Hall, Inc., New York,
(1950), p. 329.
19. Standard Methods, 12th Edition, American Public Health Assn., Inc.
New York (1965), p, '""*
20. Chemical Lime Facts, National Lime Association, Washington, D. C
21. Perry's Chemical Engineers' Handbook, Uth Ed., McGraw-Hill, 4-b.
22. Sillen, L. G. and A. E. Mar tell, Stability Constants of Metal-Ion
Complexes, Special Publication No. 17, London Chemical Society, (196*0
23. Robinson and Stokes, Electrolyte Solutions, London; Butterworths,
(1959), 517-
2k. Kolthoff, I. M. and P. J. Elving, Treatise on Analytical Chemistry,
Part I, Vol. I, N. Y. Interscience, (1959), 243.
25. Fair, G. M. and J. C. Geyer, Water Supply and Waste Water Disposal,
John Wiley, (1966),
B-2
-------
APPENDIX C
BIBLIOGRAPHY
-------
ADDITIONAL REFERENCE MATERIAL
GENERAL CHEMISTRY AMD P REMOVAL
AASGP Committee Report, "Determination of Orthophosphate Hydrolyzable
Phosphate, and Total Phosphate in Surface Waters", J.A.W.W.A., 50, 1563
(1958). —
"Alkalinity Equivalents", W. & S.W. Reference Number, 502 (1961).
Albertson, 0. E., and Sherwood, R. J., "Phosphate Extraction Process",
Dorr-Oliver, Inc.
Bier, M., and Cooper, F. C., Principles and Applications of Water
Chemistry, New York, John Wiley & Sons, (1967).
Earth, E. F., and Ettinger, M. B., "Mineral Controlled Phosphorus
Removal in the Activated Sludge Process", J.W.P.C.F., 39, 1362 (1967).
Butler, J. N., Solubility and pH Calculations, Addison-Wesley
Publishing Co., Inc., Reading, Mass. (1964).
Buzzell, J. C., and Sawyer, C. N., "Removal of Algal Nutrients from
Raw Wastewater with Lime", J.W.P.C.F., 39, No. 10, Part 2, R. 16 (196~7).
Clesceri, N. L., "Physical and Chemical Removal of Nutrients", Presented
at International Conference "Algae, Man and Environment" (1967)•
Cohen, J. M., "Alternative Methods of Phosphorus Removal", Talk
delivered to workshop on "Phosphorus Removal", Chicago, 111.
(May 1 and 2, 1968).
Corsaro, G., et. al., "Mechanism of Polyphosphate Threshold Action",
J.A.W.W.A., ^8, 683, (1956).
Corsaro, Lauderbach, and Schwantje, "Formation and Behavior of
Hydroxylapatite", J.A.W.W.A., 56, 3^7 (196*0.
Dean, R. B., "Forms and Measurements of Phosphorus", Notes - Chicago,
Unpublished, (May 1, 3.968).
Duff, J. H., Dvarin, R., and Salem,"Phosphate Removal by Chemical
Precipitation", Graver Water Conditioning Co., Div. Union Tank Car.
C-l
-------
Eberhardt, W. A., and Nesbitt, J. B., "Chemical Precipitation of
Phosphate Within a High Rate Bio-oxidation System", 22nd Annual
Purdue Ind. Waste Conf., Lafayette, Indiana (May, 1967).
Eliassen, and Tchobanoglous, "Chemical Processing of Wastewater for
Nutrient Removal", J. WPCF, j+0, No. 5, Part 2, R. 171 (1968).
Engelbrecht, R. S., and Morgan, J. J., "Studies on the Occurrence and
Degradation of Condensed Phosphates in Surface Waters", Sew. and Ind.
Wastes, 31, ^58 (1959)-
Faust, S. D., and Hunter, J. V., "Principles and Applications of Water
Chemistry", John Wiley & Sons, New York, (1967).
Forrest, T. H., et. al., "Method of Controlling Phosphate Concentration
in Sewage Treatment Systems", U. S. Patent 3,385,785 (May 28, 1968).
Finstein, M. S., "Nitrogen and Phosphorus Removal from Combined Sewage
Components by Microbial Activity", Appl. Microbiol., ik, 679 (1966).
Hess, John S., "Lime and Caustic Soda Softening at Fremont, Ohio",
J. AWWA, 60, 980 (1968).
Hartung, H. 0., "Calcium Carbonate stabilization of Lime-Softened
Water", J. AWWA, jj£, 1523 (1956).
Hiirwitz, E., and Beaudoin, R., and Walter, W., "Phosphates: Their
Fate in a Sewage Treatment Plant-Water-Way System", Water & Sew. Wks.,
112, 8U (1965).
Karl-Kroupa, E., Callis, C. F., and Seifter, E., "Stability of Condensed
Phosphates in Very Dilute Solution", Ind. Eng. Chem., ^9, 206l (1957).
LaMer, V. K., "The Solubility Behavior of Hydroxylapatite", J. Phy.
Chem., 66, 973 (June 1962).
Lawrence, Charles H., "Lime-Soda Sludge Recirculation Experiments at
Vandenberg Air Force Base", J. AWWA, 55, 177 (1963).
Levin, G. V., and Shapiro, J., "Metabolic Uptake of Phosphorus by
Wastewater Organisms", J. WPCF, 37, 800 (1965).
Malina, J. F., and Tiyaporn, S., "Effects of Synthetic Detergents
on Lime-Soda Ash Treatment", J. AWWA, 56, 727 (196*0.
C-2
-------
Nesbitt, J. F., "Removal of Phosphorus from Municipal Sewage Plant
Effluents", Eng. Res. Bull. B-93, Perm State University (1966).
Perloff, A., and Posner, A. S., "Preparation of Pure Hydroxyapatite
Crystals", Science, 12k, 583, (1956).
Quinby, O.T., "The Chemistry of Sodium Phosphates", Chem. Rev., U6,
Rogers, L. B., and Reynolds, C. A., "Interaction of Pyrophosphate Ion
with Certain Multivalent Cations in Aqueous Solutions", J. Amer. Chem.
So., 71,208l(19li9).
Rohlich, G. A., "Methods for the Removal of Phosphorus and Nitrogen
from Sewage Plant Effluents", Proc. 1st Int. Conf., Adv. in Wat.
Poll. Res., Vol. 2 (1962).
Rootare, H. M., Deitz, V. R., and Carpenter, F. G., "Solubility
Product Phenomena in Hydroxyapatite-Water Systems", J. Colloid
Sci., 17, 179 (1962).
Vacker, Connell, and Wells, "Phosphate Removal Through Municipal
Waste-water Treatment at San Antonio, Texas", J. WPCF, 39, 5, 750
(May, 1967).
Van Wazer, J. R., et. al., "Structure and Properties of the Condensed
Phosphates", J. ACS, 72, 639 (1950).
Black, A. P., "Split-Treatment Water Softening at Dayton", J. AWWA,
58, 97 (1966).
Gulp, G., "Chemical Treatment of Raw Sewage", Water & Waste Engineering,
k, 61 (1967).
Gould, R. F., Editor, Equilibrium Concepts in Natural Water Systems,
Advances in Chemistry Series No. 67, American Chemical Society (1967)•
Sawyer, C. N., and McCarthy, P. L., Chemistry for Sanitary Engineers,
Second Edition, New York, McGraw-Hill Book Co. (1967).
Schonfeld, E.,"Effect of Impurities on Precipitation of Calcium",
J. AWWA, 56, 767 (196*0.
Sillen, L. G., and Martell, A. E., Stability Constants of Metal-Ion
Complexes, Special Publication No. 17, London Chemical Society (196*0•
C-3
-------
Smith R., "Calculation of Ionic Equilibrium by Means of the Digital
Computer", FWPCA (Nov. 1966).
Tuepker, J. L., and Hartung, H. 0., "Effects on Accumulated Lime-
Softening Slurry on Magnesium Reduction", J. AWWA, 52, 106 (Jan. 1960).
COAGULATION AMD SEDIMENTATION
Anderson, A. A., and Sparkman, J. E., "Review - Sedimentation Theory",
Chem. Eng., 7!5/ (Nov. 1959)-
Bean, E. L., "Study of Physical Factors Affecting Flocculation",
Water Wks. Eng., 106, 33 (1953)-
Bean, E. L., "Control of Coagulation", W. & S.W. Reference Number,
R-182 (1961).
Bean, E. L., Campbell, S. J., and Anspach, F. R., "Zeta Potential
Measurements in the Control of Coagulation Chemical Doses", J. AWWA,
56, 21k (196U).
Berihnkem, Horowitz, and Katz, "Particle Growth Processes", Ind. Eng.
Chem. Fundam., 2, No. 3, 212 (Aug. 1963).
Black, A. P., "Stoechiometry of the Coagulation of Color-causing
Organic Compounds with Ferric Sulfate", J. AWWA, 5£» 13^7 (1963).
Black, A. P., "Basic Mechanisms of Coagulation", J. AWWA, 52, ^92
(I960).
Black, A. P., "Theory of Coagulation", W. & S.W. Reference Number,
R. 192 (1961).
Black, A. P., and Christman, R. F., "Electrophoretic Studies of Sludge
Particles Produced in Lime. Soda. Softening", J. AWWA, 53, 737 (1961).
Borchardt, J. A., "Coagulant Aids", W. & S.W. Reference Number, R. 173
(1961).
Camp, T. R., "Floe Volume Concentration", J. AWWA, 60, 656 (1968).
Camp, T. R., "Flocculation and Flocculation Basins", Trans. ASCE,
Paper No. 2722.
Camp, T. R., "Effects of Temperature on Rate of Floe Formation",
J. AWWA, 32, 1913 (19UO).
-------
Committee Report, "Capacity and Loadings of Suspended Solids Contact
Units", J. AWWA, ^3, 263 (1951).
Conley, W. R., and R. H. Evers, "Coagulation Control", J. AWWA, 60,
165 (1968). —
Gulp, Gordon, Hansen and Richardson, "High Rate Sedimentation in
Water Treatment Work", J. AWWA, 60, 68l (1968).
Fair, G. M., and R. S. Gemmel. "A Mathematical Model of Coagulation",
J. Colloid Sci., 19, 360 (196U).
Fitch, E. B., "The Significance of Detention in Sedimentation", Sew.
and Ind. Wastes, 29, 1123 (1957).
Fitch, E. B., "A Mechanism of Sedimentation", Ind. Eng. Chem. Fundam.,
2, No. 1, 129 (1966).
Ford, D. L., and W. W. Eckenfelder, Jr. "Effect on Process Variables
on Sludge Floe Formation and Settling Characteristics", J. WPCF, 39,
1850 (1967).
Fissora, F. V., "High Density Solids Contact in Water and Waste
Treatment", Ind. Water Eng., 26 (Nov. 1967).
Geinopolos, A., A.E. Albert, and W. J. Katz, "Considerations in
Clarifier Design", Ind. Water Eng., 3_, 19 (1966).
Gillespie, T., "The Limited Flocculation of Colloidal Systems",
J. Colloid Sci., 15, 313 (I960).
Hall, E. S., "The Zeta Potential of Aluminum Hydroxide in Relation
to Water Treatment Coagulation", J. Appl. Chem., 15, 197 (1965).
Hannah, S. A., et. al., "Control Techniques for Coagulation-Filtration",
J. AWWA, 59, 11^9 (1967).
Hansen, S. P., and G. L. Gulp, "Applying Shallow Depth Sedimentation
Theory", J. AWWA, 59, 113^ (19&7)-
Higuchi, W. I., et. al., "Kinetics of Rapid Aggregation in Suspensions",
J. Pharm. Sci., 52, ^9 (19^3).
Hornung, J., "Settling Basin Detention Time Test", W. & S.W. Reference
Wo., R. 200 (1961).
C-5
-------
Howells, D. H., and C. N. Sawyer, "Effects of Synthetic Detergents
on Chemical Coagulation of Water", ¥at. & Sew. Wks., 103, 71 (1956).
Hudson, Jr., H. E., "Flocculation and Flocculation Aids", J. AWWA,
21*2 (1957).
Hudson, Jr., H. E., "Coagulation and Flocculation of Surface Water",
J. NEWWA, 80, 232 (1966).
Hudson, Jr., H. E., "Physical Aspects of Flocculation", J. AWWA, 57,
885 (1965). ""
Hudson, Jr., H. E., and J. P. Wolfner, "Design of Mixing and Floccu-
lation Basins", J. AWWA, 59, 1257 (1967).
Katz, W. J., and J. L. Mancini, "Concepts of Sedimentation Applied to
Design", W. & S.W. Reference Number, R. 118 (1962).
Kim, W., H. F. Ludwig, andW.D. Bishop, "Cation-Exchange Capacity
and pH in the Coagulation Process", J. AWWA, 57, 327 (1965).
Kross, Fisher, and Paulson, "Solids Concentration and Performance of
Solids Contact Softeners", J. AWWA, 60, 597 (1968).
La Mer, V. K., and R. H. Smellie, Jr., "Flocculation, Subsidence, and
Filtration of Phosphate Slimes", J. Colloid Sci., 11, 70U (1956).
Langelier, W. F., H. F. Ludwig, and R. C. Ludwig, "Flocculation
Phenomena in Turbid-Water Clarification", Proc. ASCE, 78, Wo. 18, (1952)
Larson, T. E., and A. M. Buswell, ''Water Softening: Sign of Charge on
Colloidal Particles of Hydrous Alumina, Hydrous Magnesium, and Calcium
Carbonate", Ind. Eng. Chem., 32, 132 (19^0).
Matijevic, E., and K. Kerker, "The Charge of Some Heteropoly Anions
in Aqueous Solutions as Determined by Coagulation Effects", J. Phys.
Chem., 62, 1271 (1958).
Mackrle, S., "Mechanism of Coagulation in Water Treatment", J. San. Eng.
Div., Proc. ASCE, 88, SA3, 1 (19&2) .
Maier, W. J., "Model Study of Colloid Removal", J. WPCF, ^0, ^78 (1968).
Matijevic, E., and L. J. Stryker, "Coagulation and Reversal of Charge
of Lyophobic Colloids by Hydrolyzed Metal Ions", J. Colloid Interfac.
Sci., 22, 68 (1966).
C-6
-------
Maynard, S. B.. "Sedimentation", ¥. & S.¥. Reference Number, R 202 (1961).
Miller, D. G., and J. T. West, "Pi^ot Plant Studies of Floe Blanket
Clarification", J. AWWA, 60, 15^ (1968).
Morgan, J. J., and R. S. Engelbrecht, "Effects of Phosphates on Coagu-
lation and Sedimentation of Turbid Waters", J. AWWA, 32, 1303 (1960).
Ockershausen, R. W., "Coagulation Symposium", Water Wks . and Waste Eng.,
2, No. 1, 53 (1965).
Packham, R. F., "The Coagulation Process", J. Appl. Chem., 12, 56 (1962).
Pilipovich, et. al., "Electrophoretic Studies of Water Coagulation",
J. AWWA, 50, 1U6? (1958).
Priesing, C. P., "A Theory of Coagulation Useful for Design", Ind. Eng.
Chem., ^kj 38 (1962).
Rand, M. C., "General Principles of Chemical Coagulation", Sew. and Ind.
Wastes, 31, 863 (1959).
Riddick, T. M., "Zeta Potential and Its Application to Difficult Waters",
J. AWWA, 53, 100? (1961).
Rohlich, G. A., and K. L. Murphy, "Flocculation", W. & S.W. Reference
Number, R 211, (1961).
Scott, K. J., "Thickening of Calcium Carbonate Slurries", Ind. Eng. Chem.
Fund., 3, U8U (1968).
Shinnar, R., and J. M. Church, "Predicting Particle Size in Agitated
Dispersions", Ind. Eng. Chem., 52, 253 (I960).
Smith, R. , J. M. Cohen, and G. Walton, "Effects of Synthetic Detergents
on Water Coagulation", J. AWWA, U8, 55 .(1956).
Stumm, W., and J. J. Morgan, "Chemical Aspects of Coagulation", J. AWWA,
5U, 971 (1962).
Stumm, W., and C. R. O'Melia, "Stoichiometry of Coagulation", J. AWWA,
(1968).
Tolman, S. C., "Use of Models in Solving Flocculation Problems", J. AWWA,
C-7
-------
Vaughn, J. C., "Common Chemicals Used in Coagulation", W. & S.W.
Reference Number, R 207 (196l) •
Wallace, A. T., "Design and Analysis of Sedimentation Basins1', W. &
S.W. Reference Number, R 219 (196?) .'
GENERAL BACKGROUND
Betz L Handbook of Industrial Water Conditioning, Sixth Edition, Betz
Laboratories, Inc., Trevose, Pa., (1962).
Gulp, Russell L., "Wastewater Reclamation by Tertiary Treatment",
J. WPCF, 35, 799 (1963).
Gulp, Russell, L., and Roderick, R. E., "The Lake Tahoe Water Reclama-
tion Plant", J. WPCF, 38, 1^7 (1966).
Eckenf elder, Jr., W. W., Industrial Water Pollution Control, McGraw-Hill
Book Co., New York (1966).
"Equipment for a Dirty Job", C.W. Report, Chemical Week (Feb. 17, 1968).
Fair, G. M., and J. C. Geyer, Water Supply and Waste-Water Disposal,
John Wiley and Sons, New York (1966).
Gates, C. D., and McDermott, R. F., "Characterization and Conditioning
of Water Treatment Plant Sludge", J. AWWA, 60, 331 (1968).
Imhoff, K., and Fair, G. M., Sewage Treatment, Second Edition, John
Wiley and Sons, New York (1960).
James, G. V., Water Treatment, Third Edition, Technical Press Ltd,
London (1966).
Logan, J. A., Hatfield, W. D., Russell, G. S., and Lynn, W. R.,
"An Analysis of the Economics of Wastewater Treatment'1, J. WPCF,
860 (1962).
Garland, C. E., "Wastewater Renovation and Reuse", Presented at Infilco
National Seminars (1966) .
Moffett, J. W., "Comments on: Integration of the Clarification Process",
J. AWWA, 58, 91 (1966).
Montgomery, M. M., and Lynn, W. R., "Analysis of Sewage Treatment
Systems by Simulation", J. San. Eng. Div., Proc. ASCE, 90, 73 (1961*).
C-8
-------
"Phosphate Removal Processes Prove Practical". Environ. Sci. Technol.,
2, 182 (1968).
Rich, L. G.. Unit Processes of Sanitary Engineering, John Wiley and
Sons, New York (1963).
Sawyer, C. N., "Problem of Phosphorus in Water Supplies", J. AWA,
59, 1U31 (1965).
Sewage Treatment Plant Design, Washington, D. C. WPCF (1967).
Shuey, B. S., "Economics of Split Treatment Water Softening1',
J. AWWA, 58, 10? (1966).
Slechta, A. F., and Gulp, G. L., "Water Reclamation Studies at the
South Tahoe Public Utility District", J. WPCF, 39, 787 (1967).
Smith, R., "Calculation of Cost and Performance for a Water
Renovation System", FWPCA Unpublished Report (June, 1966).
Smith, R., Eilers, R. G., and'Hall, E. D., "Digital Computer Program
For Preliminary Design of Wastewater Treatment System", FWPCA
(March, 1968).
Smith, R., "Preliminary Design and Simulation of Conventional Waste
Water Renovation Systems Using the Digital Computer", WP-20-9,
Cincinnati, Ohio FWPCA (March, 1968).
Smith, R., Eilers, R. G., and Hall, E. D., "Executive Digital Computer
Program for Preliminary Design of Waste-Water Treatment Systems",
WP-20-lU, Cincinnati, Ohio FWPCA (Aug., 1968).
C-9
------- |