WATER POLLUTION CONTROL RESEARCH SERIES • 17030ESX 04/70
INVESTIGATION OF A
HIGH-PRESSURE FOAM WASTEWATER
TREATMENT PROCESS
U.S. DEPARTMENT OP THE INTERIOR • FEDERAL WATER QUALITY ADMINISTRATION
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Reports describe the results
and progress in the control and abatement of pollution of our
Nation's waters. They provide a central source of information on
the research, development, and demonstration activities of the
Federal Water Quality Administration, Department of the Interior,
through in-house research and grants and contracts with Federal,
State, and local agencies, research institutions, and industrial
organizations.
Water Pollution Control Research Reports will be distributed to
requesters as supplies permit. Requests should be sent to the
Planning and Resources Office, Office of Research and Development,
Federal Water Quality Administration, Department of the Interior,
Washington, D. C. 20242.
-------
INVESTIGATION OF A HIGH-PRESSURE FOAM
WASTEWATER TREATMENT PROCESS
by
James K. P. Miller
L. Karl Legatski
Garrett Research and Development Company, Inc.
La Verne, California 91750
for the
FEDERAL WATER QUALITY ADMINISTRATION
DEPARTMENT OF THE INTERIOR
Program #17030 ESX
Contract #14-12-176
FWQA Project Officer, Dr. C. A. Brunner
Advanced Waste Treatment Research Laboratory
Cincinnati, Ohio
April, 1970
-------
FWQA Review Notice
This report has been reviewed by the Federal Water
Quality Administration and approved for publication.
Approval does not signify that the contents neces-
sarily reflect the views and policies of the Federal
Water Quality Administration, nor does mention of
trade names or commercial products constitute
endorsement or recommendation for use.
-------
ABSTRACT
A foam fractionation-flotation process for the separ-
ation of organic materials and other contaminants from primary
and secondary sewage effluent was studied to determine its
technical and economic feasibility. Air and waste water
were mixed and held at pressures exceeding 150 psi for periods
of 6 to 25 minutes. The mixtures^were then bled to a release
vessel where pressures were reduced to atmospheric. Dis-
solved air was released as extremely fine bubbles which
formed a thick creamy froth. Soluble organic and particulate
matter collected at the bubble-water interfaces and was re-
moved with the froth at the overflow of the release vessel.
Air flow rates corresponding to between 40 and 125% of
saturation, different pressures/ and varying contact times
were studied. Several release vessels and a variety of
additives were also tested. The most promising results were
obtained with 300-400'mg/1 of either ferric chloride or alum
as coagulants, at pressures greater than 175 psi, and at
air-to-water volume ratios of 0.17 to 1. These conditions
gave chemical oxygen demand reductions of about 70%, phos-
phate reductions exceeding 90%, and suspended solids reduc-
tions of 40-80%. Estimated treatment costs, exclusive of
chemicals, would be less than $0.05/1,000 gal. for a 10 'mgd
plant*
This report was submitted in fulfillment of Program No.
17O30 ESX, Contract No. 14-12-176, between the Federal Water Quality
Administration and Garrett Research and Development, Inc.
±±
-------
TABLE OF CONTENTS
Page
INTRODUCTION 1
EXPERIMENTAL PROGRAM AND DISCUSSION
OF RESULTS 2
Experimental Equipment 2
Preliminary Experiments Without
Chemical Additives 3
Experiments Using Chemical Additives. 5
COD Reduction 6
Phosphate Reduction 7
Suspended Solids Removal.. 7
CONCLUSIONS 8
ECONOMICS 8
RECOMMENDATIONS ; 9
ill
-------
INTRODUCTION
An urgent need exists for a low cost method to reduce
contaminants in wastewater to levels that will not cause
pollution problems in receiving streams nor be objectionable
in water reuse. A combination of high pressure foam frac-
tionation and flotation offers a potential- solution to this
problem. Foam fractionation at lower pressures has been
used to remove non-biodegradable detergents; flotation is
used in many industrial processes where, particulate matter
is separated from a liquid. Preliminary laboratory experi-
mentation indicated that by increasing operating pressure
in the foam fractionation process and by.combining the pro-
cess with a flotation step large quantities of pollutants
could be removed from wastewater. When the air and water '
are contacted at high pressure, more air is dissolved than
would occur at atmospheric pressure. When the pressure is
reduced to atmospheric/ numerous very small bubbles are
released from the solution. The froth and the waste material
which it contains is then separated from the purified water
and disposed of. Such a method might be used to remove
soluble organic materialf phosphates/ and particulate matter.
The dissolving of large quantities of air during the high
pressure treatment would be beneficial in reducing Mochemical
oxygen demand ,(BCD). The process might also find industrial
application in the treatment of cannery and pulp and paper
waste or any other wastes which contain large quantities of
suspended solids.
Soluble organic and particulate matter collects at air
bubble-water interfaces and can be removed by foam fraction-
ation and flotation. Soluble phosphates can be precipitated
with alum or iron salts and removed by flotation. The pres-
sure, air-to-water ratio, and water-air contact time all
affect the physical properties of the froth and the effec-
tiveness of the process in purification of wastewater. This
investigation was intended to study each of these variables,
using equipment which might be used in a sewage treatment
facility. Additional studies were made to determine the
effect of various chemical coagulants on foam production and
removal of organic matter and soluble phosphates. All
experiments were run in a small scale pilot plant.
- 1 -
-------
EXPERIMENTAL PROGRAM AND DISCUSSION OF RESULTS
Wastewater used in this project was municipal primary
or secondary effluent obtained immediately before its use
from the Pomona, California sewage treatment plant. The
small pilot plant equipment was fabricated from commercially
available parts and is described in detail in the following
section.
Experimental Equipment
Figure 1 describes the pilot unit used in this inves-
tigation. A 200 gal., agitated feed tank was used to hold
the sewage effluent prior to introduction into the process-
ing circuit. Reagents in solution and low pressure air could
be introduced to the wastewater stream before it entered the
suction of the gear pump. The pump was driven with a variable
speed motor which allowed for.control, of both the flow rate
and pressure. The wastewater feed and the air lines had
rotameters calibrated for the operational conditions which
made it possible to control the air-to-liquid ratio. Two
dissolving tanks, fed by the pump, had a capacity of about
six gallons each. Air dissolving (retention) time was varied
by removing one of'the tanks from the circuit and piping
directly from the other to the foam separator.
A box-type foam separator (Figures 2 and 3) was origi-
nally constructed of lucite but was replaced with a stainless
.steel unit having lucite windows. The separator was a square
box having a pyramidal metal top and a volume of 22 gal.
Initial experiments were made with this separator, and a
number of problems were encountered. The froth accumulated
on the sloping sides of the top and could be seen falling
back to the bottom of the vessel. It was apparent that the
volume of the vessel was too large for the system and, as
a consequence, the foam retention time was long enough to
cause sedimentation and deposition of the froth on the walls
and top of the vessel.
A channel-type separator (Figures 4 and 5) was then de-
signed to eliminate flat surfaces which the froth could con-
tact. Difficulties in controlling the liquid level were
encountered. The foam skimmer would first remove too much
foam and then remove none at all. When replacement of the
overflow weir with three weirs and attachment of a vacuum
apparatus still proved unsatisfactory, a moving paddle was
devised to help push the foam toward two adjustable weirs
built into the channel. While this last change improved
separation to some extent, results with the channel-type
separator were not satisfactory.
- 2 -
-------
FEED
TANK
PO
REAGENT
TANK
REAGENT
PUMP
ROTAMETER
ROTAMETER °
AIR PRESSURE
REGULATOR
LOW-PRESSURE
AIR COMPRESSOR
GAUGE
DISSOLVING
TANK
DISSOLVING
TANK
HIGH-PRESSURE
VARIABLE DRIVE
GEAR PUMP
FOAM
FOAM
SEPARATOR
FOAM
RECEIVER
X
EFFLUENT
HOLDING TANK
FIGURE 1
EQUIPMENT ARRANGEMENT
FOR FOAM FRACTIONATION-
FLOTATION PROCESS
-------
18'
18"
l"LUCITE PIPE
*->'
4
T
"EFFLUENT REMOVAL
PIPE
SPARGER
FIGURE 2
BOX-TYPE FOAM SEPARATOR
-------
IP VIEW
£"s.s. i
,
f
BIPI
A *•
1
\
E
X
O
O
O
0
\,
n ~
o o o o
1
k> ^l
o o o o
13"
*-
\
o
o
o
o
^
3
I
r
EFFLUENT REMOVAL PIPE- FRONT VIEW
->,
II
FIGURE 3
INTERNALS OF
BOX-TYPE FOAM SEPARATOR
-------
52
EFFLUENT
REMOVAL
FOAM
REMOVAL
SPARGER
1
•
1
"T"
2"
t
nr" ^
_^_:.
LEGS
FIGURE 4
TROUGH-TYPE FOAM SEPARATOR
-------
SPARGER-TOP VIEW
6
FOAM REMOVAL DRAIN-SIDE VIEW
.S. TUBING
FIGURE 5
DETAILS OF TROUGH-TYPE FOAM
SEPARATOR
-------
Alternatively, a proven separator design was used to
permit adequate testing of other variables within the project
budget. A cylindrical separator with a foam skimmer is shown
in Figure 6. A similar separator used previously in experi-
mental work had performed satisfactorily. Modifications were
made to adapt this separator to the foam fractionation-
flotation process, and most of the experimental runs were
made with it.
Preliminary Experiments Without Chemical Additives
Initial experiments were made without any chemical
coagulants to determine optimum operating parameters for re-
duction in chemical oxygen demand (COD). Although chemical
additives are needed if good phosphate reductions are to be
achieved, experiments with additives were deferred until
other variables had been studied. Feed material for nearly
all of the experiments was primary sewage effluent which had
been allowed to settle for two hours. Secondary sewage
effluent was used as feed for a few experiments. Variations
in COD were noted depending on the time of day at which a
sample was obtained. Morning samples ran 100 mg/1 COD or
less, which was not considered typical of primary effluent.
Work hours were rearranged and samples were obtained in the
afternoon when the COD was usually between 150 and 210 mg/1.
In the initial experiments using secondary effluents,
an attempt was made to do material balances for the entire
system. The results, shown in Table 1, are somewhat erratic.
Because of the difficulty encountered in accurately sampling
"the high COD foam concentrates, material balances were found
to be less reproducible than desired.
Primary sewage effluent with rather low COD was used as
feed in a series of preliminary experiments to determine the
general effects of pressure, contact time, and air-to-liquid
ratios. The results are shown in Table 2. Average COD re-
ductions were 28.7% at 235 psi and 20.2% at 175 psi. Average
turbidity reductions were 29.8% and 40.2% for the same
pressures, respectively.
Two residence times in the dissolver were tested by
first operating with two pressure tanks in the circuit and
then with only one tank in the circuit. The residence times
were 13.3 minutes and 6.7 minutes. COD reductions at 13.3
minutes averaged 25.5% while at 6.7 minutes the average
reduction was 22.1%. Reductions in turbidity averaged 36.6%
and 31.8% for the two residence times, respectively.
Air flow rates corresponding to 42, 78 and 125% of
saturation at the operating pressures were tested to deter-
mine if a larger quantity of dissolved air would cause an
- 3 -
-------
VARIABLE SPEED
MOTOR
PLASTIC^
FOAM
COLLECTING
BUCKET
UNDERFLOW
DRAIN
T
FOAM
SKIMMING
PADDLES
_.» ^-
14'
SPARGER
-J
i—LJ-wJ
FIGURE 6
CYLINDRICAL
FOAM SEPARATOR
WITH FOAM SKIMMER
\\ FOAM
\\DRAIN
\r^
-------
TABLE 1
Foam Fractionation Data
Material Balance for System Using Secondary Effluent as Feed
Box-Type Foam Separator
Feed Flow Rate:
Pressure:
Time in Dissolver:
Time in Separator:
Air-to-Water Ratio:
0.9 gpm
235 psi
13.3 minutes
22.0 minutes
0.09 to 1
% of Theoretical Saturation: 42
COD, mg/1
Feed
85
65
60
60
60
Effluent
68
57
46
36
39
Foam Concentrate
896
325
134
256
157
VWJ.U111C WJ- ^Will»ClJ U4. «UG t
% of original feed
1.0
21.0
1.0
9.0
12.0
% accounted for
90
209
79
98
97
fa
CT
(->
(D
-------
TABLE 2
Foam Fractionation Data
Waste Water Treatment Without Chemical Additives
Low COD-Primary Effluent Peed
Box-Type Foam Separator
Feed Flow Rate: 0.9 gpm
Time in Separator: 22 minutes
Air Flow Rate
Pressure/
- psig
175
175
175
175
175
175
175
175
175
235
235
235
235
235
235
235
235
235
Dissolver
Time,
minutes
6.7
6.7
6.7
13.3
13.3
13.3
13.3
13.3
13.3
6.7
6.7
6.7
13.3
13.3
13.3
13.3
13.3
13.3
% of
Feed
Volume
9.5
17.0
25.5
9.5
17.0
25.5
17.0
17.0
17.0
11.5
20.0
33.0
11.5
20.0
33.0
21.0
11.5
33.0
% of
Saturation
42
78
125
42
78
125
78
78
78
42
78
125
42
78
125
78
42
125
COD
Feed/
mg/1
71
88
74
98
82
69
60
82
60
85
88
92
61
72
60
85
61
57
Reduction/
%
31
8
32
9
4
19
35
3.5
40
19
5
38
44
27
30
20
41
34
Turbidity
Feed,
mg/1
40
82
55
108
107
67
85
85
85
40
82
45
8
42
92
25
64
Reduction/
38
21
50
28
33
33
70
33
56
10
21
51
13
42
44
50
14
23
to
-------
increase in COD reduction. Average COD reductions were
28.8% at 42% of saturation/ 17.8% at 78% of saturation, and
30.6% at 125% of saturation. Turbidity reductions, however,
increased from 28.6% at 42% of saturation to 40% at both
78 and 125% of saturation.
In these runs and those that follow it was necessary
to use a number of different batches of effluent. This vari-
ability in feed composition makes interpretation of results
more difficult than when a single feed composition can be
used. Conclusions about the effect of operating variables
must be considered preliminary. An expanded experimental
program would be required to make conclusions more firm.
The above experiments suggested that only a slight gain in
COD removal would be realized by increasing the system pressure
above 175 psi or by increasing residence time in the pressure
tank to greater than seven minutes. Turbidity measurements
suggested that increased air flow, at ra'tes corresponding to
somewhat less than 78% of saturation, would provide for
better separation of particulate matter than would lower
flow rates. A limit, however, was reached at some point
below 78% of saturation where no further increase in turbi-
dity reduction was noted. COD results did not confirm this
observation. Visual observations indicated that air, in
excess of that needed to reach theoretical saturation at
the elevated pressures, caused extreme turbulence in the re-
lease vessel. The turbulence seemed to destroy much of the
froth and it was expected that COD removal would be drasti-
cally reduced. Again, the COD analyses did not confirm the
visual observations, and there was no evide'nce to indicate
that COD removal had been affected. Further experiments
were run to determine if higher COD feed material might be
used to better define the effects of operating variables.
The results of these experiments are shown in Table 3. Two
pressures/ two residence times and two air flow rates were
tried. The lower pressure of 175 psi gave .slightly better
COD removals than did 235 psi. Residence time in the pres-
sure vessel had exactly the same effect on -the high COD feed
that it had on the low COD feed. A very slight increase
(from an average of 13 to 16%) in COD removal was noted when
residence time was increased from 6.7 minutes to 13.3 minutes.
Air flow rates corresponding to 125% of theoretical saturation
gave approximately 18% greater COD reductions (average) than
did air flow rates corresponding to 42% of saturation.
Turbidity measurements on the underflow from the foam
separator varied widely when primary effluents with COD of
greater than 150 mg/1 were used as feed. An increase in
turbidity was observed for the same feed material under the
same conditions when the pump speed was increased in the
pilot plant circuit. Apparently agglomerates were broken up
at high speeds and caused the turbidity readings to increase.
Table 4 describes a series of five measurements made during
- 4 -
-------
TABLE 3
Foam Fractionation Data
Waste Water Treatment Without Chemical Additives
Primary Effluent Feed
Box-Type Foam Separator
Feed Flow Rate: 0.9 gpm
Time in Separator: 22 minutes
Air Flow Rate
Pressure,
psig
175
175
175
175
235
23!
235
235
Dissolver
Time,
minutes
6.7
6.7
13.3
13.3
6.7
6.7
13.3
13.3
% of
Feed
Volume
9.5
25.5
9.5
25.5
11.5
33
11.5
33
% Of
Saturation
42
125
42
125
42
125
42
125
COD
Feed,
mg/i
186
206
152
205
206
186
205
152
Reduction,
%
16
14
9
22
9
13
19
14
Turbidity
Feed,
mg/1 SiO.
78
78
78
108
45
78
108
45
Reduction,
%
18
31
16
0
15
4
0
-33
u>
-------
TABLE 4
Foam Fractionation Data
Effect of Pump Speed on Turbidity
.Primary Effluent Feed, (152 mg/1 COD)
Box-Type Foam Separator
Feed Flow Rate: 0.9 gpm
Pressure: 235 psi
Time in Dissolver: 13.3 minutes
Time in Separator: 22.0 minutes
Air-to-Water Ratio: 0.09 to 1
Turbidity,
Pump Speed* mg/1 SiO,,
0 76
30 87
48 91
60 98
76 102
* A variable speed motor was used on the effluent pump.
The Pump Speed figures are readings taken from the
control rheostat and are not rpm.
-------
one experiment to demonstrate the effect of pump speed on
turbidity. While pump speed was varied, all other variables
were held constant and as the table shows, there was a well
defined increase in. turbidity as pump speed was increased.
As an alternative to the turbidity analysis, a quanti-
tative suspended solids determination was attempted. This
analysis proved to be too expensive in relation to the
allowed analytical budget however, and no further consider-
ation was given it. Turbidity measurements were continued
throughout the experimental program with constant pump speed
to assure comparability of turbidity results.
Two experiments with the trough-type separator, using no
coagulants, offered no improvement over the box-type separator
used in all of the above experiments. Both runs with the
trough-type separator were made at 175 psi, with air flow
rates corresponding to 125% of saturation. The feed was
181 mg/1 COD primary effluent, at a rate of 0.9 gpm. Resi-
dence time in the pressure vessel was 13.3 minutes and
residence time in the fractionator was 10 minutes. One
experiment gave a COD reduction of 13.3% and the other, 45.6%.
The separator was difficult to control and after one experi-
ment with alum as an additive (described later) the trough-
type separator was replaced by the cylindrical separator.
The cylindrical separator was tested without chemical
additives and some promising results were obtained as shown
in Table 5. The COD in the feed for these experiments varied
from 289 to 128 mg/1. The average COD reduction for two
experiments at 235 psi was 26.5%. At 175 psi, the average
COD reduction was 34.7%; one run at 210 psi gave a reduction
of 39%. Apparently, no correlation exists between air-to-
water ratios and COD reductions; e.g., one test at 42% of
saturation gave a COD reduction of 50%, while another gave
only a 26% reduction. Another test at 109% saturation gave
a 48% reduction, while a repeat of the same test gave only
a 6% reduction.
The results of tests without chemical coagulants indi-
cated that COD reductions of as much as 50% were possible
under the proper conditions. Reductions in turbidity of up
to 71% were also observed. However, to remove phosphate and
to improve removal of organic matter, it was necessary to
study the effect of chemical coagulants (phosphate precipi-
tants) such as alum and some iron salts.
Experiments Using Chemical Additives
Before the box-type and trough-type separators were re-
placed by the cylindrical separator, a few experiments were
run using chemical coagulants. Tests with the box-type
- 5 -
-------
TABLE 5
Foam Fractionation Data
Waste Water Treatment Without Chemical Additives
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate: 0.9 gpm
Time in Separator: 2.5 minutes
Air Flow Rate
Dissolver
Pressure,
psig
210
175
175
235
235
175
Time,
minutes
13.3
13.3
13.3
13.3
13.3
13.3
% of
Feed Volume
26.4
22.4
8.7
33.8
\
11.3
22.4
COD
% of
Saturation Feed
108
109
42
125
42
109
212
128
221
187
289
220
Reduction
39
6
50
27
26
48
Turbidity
, Feed,
mg/1 Si00
163
92
242
116
102
« —
Reduction,
%
71
27
36
41
24
».*
cr
M
(D
en
-------
separator used iron salts, while those with the trough-type
used alum. Results of experiments with both separators are
shown in Table 6. Although there was enhanced froth produc-
tion over runs without additives, the resultant increase in
froth volume did not cause COD reductions to increase be-
cause of the inability of either of these separators to
separate adequately the foam from the water. The greatest
COD reduction in these six experiments was 22.1% while the
average was only 10.8%.
During experiments with the cylindrical foam separator,
using no chemical additives, the high rate of upward flow in
the 7-inch diameter vessel tended to concentrate the column
of bubbles at the center and provide a support for particles
of floe which would have otherwise fallen to the bottom of
the separator. The flow rate-in this vessel was 3.3 gal./
sq ft-min for a feed rate of 0.9 gpm as compared with a flow
rate of 0.32 gal./sq ft-min for the same feed rate using the
box-type separator. The cylindrical separator seemed to
satisfy well the residence time and foam removal requirements
for good contaminant removal.
COD Reduction. The first experiments with chemical additives
in the cylindrical separator were designed to study the
effects of low concentrations of alum and ferrous sulfate.
Results of these experiments are shown in Table 7. Because
alum was more effective, ferrous sulfate was not used in
further experiments. Three experiments with 100 mg/1 of
alum gave an average COD reduction of 27.7%. Two other alum
experiments were run with 50 and 200 mg/1 and, surprisingly,
the 50 mg/1 concentration gave the best results (30% reduc-
tion in COD), possibly because of the very high feed COD.
All of these experiments, however, gave less COD reduction
than was considered essential for the process to be very
useful.
Additional experiments were then run to determine the
relationship between increased chemical additive concen-
trations and removal of wastewater contamination. Fixed
operating conditions were used and only the type and concen-
tration of additive was varied. Feed flow rate was 0.52 gpm
with the resultant foam separator flow rate of 1.94 gal./
sq ft-min and a separator residence time of five minutes.
Operating pressure was 230 psig and an air-to-water volume
ratio of 0.173 to 1 (65% of theoretical saturation) was used
Alum and ferric chloride were first tested and the
results are shown in,Tables 8 and 9. A secondary settling
tank (with a retention time of 10 minutes) was added to the
pilot plant circuit to determine whether or not a settling
step would produce an improved effluent. As indicated by
the data in Tables 8 and 9, the secondary settler did not
generally prove useful. COD reductions, however, using
- 6 -
-------
TABLE 6
Foam Fractionation Data
Preliminary Experiments Using Chemical Additives
Primary Effluent Feed
Feed Flow Rate: 0.9 gpm
Time in Dissolver:
Time in Separator:
13.3 minutes
22.0 minutes for box-type
10.0 minutes for trough-type
Air Flow Rate
COD
Separator
Box
Box
Box
Box
Box
Trough
Additive
FeSO4
FeS04
FeS04
FeS04
FeCl3'6H20
Alum
Concentration ,
mg/1
50
50
100
200
100
100
Pressure,
psig
235
175
235
235
235
175
% of
Feed Volume
11.5
25.5
11.5
11.5
11.5
25.5
% of
Saturation
48
125
48
48
48
125
Feed,
mg/1
176
176
111
157
155
118
Reduction,
%
17.6
22.1
2.7
5.0
4.5
12.7
cr
M
n>
-------
TABLE 7
Foam Fractionation Data
Waste Water Treatment Using Low Concentrations of Chemical Additives
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate: 0.9 gpm
Time in Dissolver: 25 minutes
Time in Separator: 5 minutes
Air Flow Rate
Additive
Alum
Alum
Alum
Alum
Alum
FeS04
FeSO,
Concentration ,
mg/1
50
100
200
100
100
100
100
Pressure,
psig
175
225
225
175
175
175
175
% of
Feed Volume
22.4
22.4
11.3
22.4
22.4
33.8
22.4
% of
Saturation
109
86
44
109
109
164
109
COD
Feed
551
375
94
120
146
121
121
.Reduction,
30
21
21
44
18
7
21
cr
M
o>
-j
-------
TABLE 8
Foam Fractionation Da,ta
COD Reduction Using Alum
Primary Effluent
Cylindrical Foam
Feed Flow Rate:
Pressure:
Time in Dissolver
Time in Separator
Feed
Separator
•
Air-to-Water Ratio: 0
Theoretical Saturation :
Alum
Concentration ,
(mg/1)
157
173
222
234
250
270
300
305
340
340
350
376
400
450
450
500
550
COD
.52 gpm
230 psig
25 minutes
5 minutes
.173 to 1
65%
Foam Separator Sect
Feed Reduction , %
141
142
199
177
139
107
144
178
94
136
171
115
144
175
170
190
148
24
16
20
29
29
59
57
41
61
72
36
58
67
70
64
78
78
Feed
109
120
159
126
98
43
62
106
—
38
110
48
47
52
61
42
33
COD
Secondary Settler
Reduction, %
3.5 increase
2.0
3.0 increase
1.5
7.0
2.3
5.0
2.0
21.0
11.0 increase
10.0 increase
0.0
9.5
27.0
4.5
9.0
-------
TABLE 9
Foam Fractionation Data
COD Reduction Using Ferric Chloride
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate: .52 gpm
Pressure: 230 psig
Time in Dissolver: 25 minutes
Time in Separator:
Air-to-Water Ratio:
Theoretical Saturation:
FeCl3-6H2O
Concentration, Foam
mg/1 Feed
135 161
156 135
180 145
200 162
225 151
270 175
338 133
338 138
340 107
376 156
450 204
500 173
520 173
5 minutes
0.173 to 1
65%
COD
Separator
Reduction,%
28
29
24
24
39
44
62
65
79
47
71
80
80
COD
Secondary
Settler
Feed Reduction , %
116
96
111
124
93
98
51
49
—
83
59
34
34
0.0
11.0
11.0 increase
4.0
8.5
21.0
6.0 increase
4.0
—
5.0 increase
34.0
12.0 increase
26.0 increase
-------
300-400 mg/1 of either alum or ferric chloride were quite
encouraging. Above 375 mg/1, the rate of increase in COD
removal decreased with an increase in additive concentration.
COD reductions for the alum and ferric chloride experiments
are plotted against additive concentrations in Figures 7
and 8. The plots are almost identical for the two additives.
The optimum concentration appears to be about 350 mg/1 for
both additives.
Phosphate Reduction. Phosphate determinations were made in
some of the experiments using alum and ferric chloride to
study the effects of these additives on phosphate removal.
The data are plotted in Figures 9 and 10. Considering the
scatter in the data over much of the range of chemical doses,
there does not appear to be much difference in the removal
capability of the two additives on a weight basis. At very
high doses the removal by ferric chloride was. slightly
better. Two runs resulted in over 95% removal.
Additional experiments were done to test the effective-
ness of calcium hydroxide and zinc sulfate in phosphate
removal. Neither of these additives gave results comparable
to either ferric chloride or alum.
•Suspended Solids Removal. Turbidity measurements were made
for most of the chemical additive runs. The turbidimeter
was calibrated against known suspended solids samples to
provide data approximating actual suspended solids removal.
Results are presented in Tables 10 and 11. The results are
extremely erratic for both chemicals, but the variability is
almost certainly caused by poor analytical accuracy. If a
rapid, accurate method to quantitatively determine suspended
solids had been available, the data would have likely been
much more useful. It can be inferred from the data that
removal increased with increasing chemical concentrations.
This is more obvious with alum than with ferric chloride.
Reductions of up to 93% were obtained. It appears, however,
that somewhat lower removals are to be expected generally.
One additional bit of experimental data should be
mentioned, even though there was no follow-up work done on
it due to budget limitations. Sodium silicate was tested as
an additive at 100 mg/1 and under conditions similar to the
alum experiments described in Table 7. At 225 psig and with
air flow rates corresponding to 130, 86, and 44% of saturation,
the average COD removal was 36%. This average figure is
considerably better than the results for alum at that same
concentration and it is certainly possible that sodium silicate
might be a more effective additive than any of the others
tried.
- 7 -
-------
FIGURE 7
Foam Fractionation Data
COD Reduction Using Alum
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate:
Pressure:
Time in Dissolver:
Time in Separator:
Air-to-Water Ratio:
Theoretical Saturation:
0.52 gpm
230 psig
25 minutes
5 minutes
0.173 to 1
65%
100
80
G
0)
D
H
0)
04
c
o
O
0)
OS,
Q
O
u
60
40
20
o o
100 200 300 400 500 600
Alum, mg/1
-------
FIGURE 8
Foam Fractionation Data
COD Reduction Using Ferric Chloride
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate:
Pressure:
Time in Dissolver:
Time in Separator:
Air-to-Water Ratio:
Theoretical Saturation;
0.52 gpm
230 psig
25 minutes
5 minutes
0.173 to 1
65%
100
80
0)
O
H
0)
Oi
d
o
•H
•P
O
3
•O
0)
§
60
40
20
O O
100 200 300 400 500
'6HO, mg/1
600
-------
FIGURE 9
Foam Fractionation Data
Phosphate Removal Using Alum
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate:
Pressure:
Time in Dissolver:
Time in Separator:
Air-to-Water Ratio:
Theoretical Saturation: 65%
0.52 gpm
230 psig
25 minutes
5 minutes
0.173 to 1
100
4J
c
0)
o
V)
<0
c
o
•H
-P
O
d
•d
0)
0)
-p
id
s
to
O
x:
PI
-------
FIGURE 10
Foam Fractionation Data
Phosphate Removal Using Ferric Chloride
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate:
Pressure:
Time in Dissolver:
Time in Separator:
Air-to-Water Ratio:
Theoretical Saturation:
0.52 gpm
230 psig
25 minutes
5 minutes
0.173 to 1
65%
10Q
-P
c
0)
o
n
a)
C
O
-H
4J
O
0)
•M
CU
0)
O
x:
100
200 300 400
FeCl3-6H2O, mg/1
500
600
-------
TABLE 10
Foam Fractionation Data
Reduction of Suspended Solids Using Alum
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate: 0.52 gpm
Pressure: 230 psig
Time in Dissolver: 25 minutes
Time in Separator: 5 minutes
Air-to-Water Ratio: 0.173 to 1
Theoretical Saturation: 65%
Suspended Solids
Decrease in
Suspended Solids
Alum,
mg/1
157
173
222
234
250
270
300
305
340
340
376
400
450
450
500
550
600
in Feed,
mg/1
176
142
170
64
64
32
142
85
51
68
77
140
130
177
150
150
130
in Effluent,
%
62
20
40
41
64
41
38
33
63
72
65
43
66
75
82
75
82
-------
TABLE 11
Foam Fractionation Data
Reduction of Suspended Solids Using Ferric Chloride
Primary Effluent Feed
Cylindrical Foam Separator
Feed Flow Rate: 0.52 gpm
Pressure: 230 psig
Time in Dissolver: 25 minutes
Time in Separator: 5 minutes
Air-to-Water Ratio: 0.173 to 1
Theoretical Saturation: 65%
FeCl-,-6H0O,
«J /1
mg/1
Suspended Solids
in Feed,
mg/1
Decrease in
Suspended Solids
in Effluent,
135
180
225
270
338
338
340
376
450
500
520
128
128
175
130
128
92
X
32
114
130
205
200
16
41
63
42
70
59
41
33
48
93
56
-------
CONCLUSIONS
Experiments using the foam fractionation-flotation unit
indicate that COD reductions of up to 50% are possible with-
out the addition of chemical coagulants or precipitants.
Turbidity reductions of 70% were also observed. With modi-
fications in the foam separator, these results can almost
certainly be improved.
Results of experimentation also indicate that both alum
and ferric chloride are effective in COD reduction, phosphate
removal and removal of suspended solids from wastewater.
Concentrations of 300-400 mg/1 are required for both addi-
tives to remove 75% of the COD, 90% of the phosphates and
65% of the suspended solids. These reductions are suffi-
ciently high to consider this treatment as an alternative
to conventional methods for treatment of effluent from primary
treatment plants.
With refinement of the process equipment and under
optimum operating conditions, this process should be quite
useful in treating wastewater from any industrial process
where high concentrations of particulate matter must be
removed.
ECONOMICS
If a retention time of 10 minutes in the air-dissolving
tank were used, a vessel capacity of about 70,000 gal. would
be required for a 10 mgd plant. Assuming an operating pres-
sure of 200 psi, the air-dissolving tank would cost about
75 cents per gal. of capacity, or $52,000'. A residence time
in the foam separator of five minutes would require a capa-
city of about 35,000 gal. Since this vessel is relatively
simple, its cost is estimated at about 25 cents per gal. of
capacity, or $9,000. Pumps, piping and instrumentation are
estimated at $80,000 and installation at $150,000, giving a
total cost of approximately $290,000 for a 10 mgd plant.
The land requirement is small (approximately one-half acre),
with the cost estimated at $10,000, giving a total cost for
the plant of $300,000. The power requirement would be about
1,200 H.P. Operating costs are itemized in Table 12 and are
approximately 5
-------
TABLE 12
Foam Fractionation-Flotation Plant Operating Costs
Capacity, 10 mgd
Power, @ l«/kwh $218/day
Operating labor, 2 men @ $3.00/£hr. 48
Maintenance labor (6% of capital) 49
Operating & maintenance supplies (0.5% of capital) 4
Payroll extras (15% of labor cost) 15
Overhead, (100% of total labor cost) 97
Amortization @ 4% for 25 years (6.4% of capital) 53
Taxes & insurance (1% of capital) 8
Interest on working capital (0.72% of foregoing) 4
$496/day
or 5
-------
required would need to be added to the foaming costs, however,
to make a realistic comparison. The foaming process does
offer the advantage of requiring less land than conventional
treatment, which could be a significant factor in land-short
areas.
RECOMMENDATIONS
The amount of information developed to date, and the
success achieved, indicates that this work should be con-
tinued. The possibilities of the process expand as the work
progresses and future work might include:
1. Determination of methods (agitation, recirculation,
etc.) to get a larger percentage of the theoretically soluble
air into solution during a short residence time.
2. Investigation of combinations of additives, such as
an inorganic for phosphorus removal and a polyelectrolyte
for suspended solids removal.
3. Changing equipment and/or the flow arrangement to
improve separation of the foam from the water.
4. Investigation of the removal of nitrogen compounds
with additives. With the high percentage of dissolved or-
ganics that are removed, it is only logical that some soluble
nitrogen compounds are also being removed.
5. Batch-type studies using ten(or more) small com-
pressed gas bottles (two gallons each) which could be filled
with the same COD feed material, sealed, and pumped with an
air compressor to the desired experimental pressure. The
bottles could be agitated by shaking them or left undisturbed
for static tests. Ten or more concentrations of one addi-
tive could be tested at one time or ten different pressures
could be tried, etc. The bottles could be bled off into
identical, low cost, lucite foam separators and the analyti-
cal work could then be done. A short (two month) effort
could probably define all of the parameters necessary to
attain optimum results from the pilot scale equipment.
It is recommended that the work be continued in three
phases: a) batch-type experimentation, b) additional work
on the operational scale of the present contract, and c)
a pilot study on a fairly large scale to prove the process
on a continuous basis.
« U.S. GOVERNMENT PRINTING OFFICE : 1970 O - 405-134
------- |