WATER POLLUTION CONTROL RESEARCH SERIES • DAST-3
Foam Separation
of
Kraft Pulping Wastes
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
-------
The Water Pollution Control Research Reports describe the results and
progress in the control and abatement of pollution in our Nation's
Waters. They provide a central source of information on the research
development and demonstration activities in the Federal Water Pollution
Control Administration, in the U. S. Dept. of the Interior, both in-
house and through grants and contracts with Federal, State and local
agencies, research institutions, and industrial organizations. The
exchange of such data should contribute toward the long-range develop-
ment of economical, large-scale management of our Nation1s water
resources.
Triplicate tear-out abstract cards are placed inside the back cover to *
facilitate information retrieval. Space is provided on the card for
the user's accession number and for additional keywords. The abstracts
utilize the WRSIC system.
Water Pollution Control Research Series will be distributed to requesters
as supplies permit. Requests should be sent to the Industrial Pollution
Control Branch, Office of Research and Development, Federal Water Pollution
Control Administration, Washington, D. C. 20242.
Previously issued reports in the series pertaining to this industry are:
CRD - 1 "Joint toinicipal and Semichemical Pulping
Waste Treatment," July, 1969.
-------
FOAM SEPARATION OF KRAFT PULPING WASTES
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
DEPARTMENT OF THE INTERIOR
BY
Georgia Kraft Company
Research and Development Center
Krannert Road
Rome, Georgia 30161
Program #12040 EUG
Grant #WPRD 117-01-68
October, 1969
-------
FWPCA Review Notice
This report has been reviewed by the Federal
Water Pollution Control Administration and
approved for publication. Approval does not
signify that the contents necessarily reflect
the views and policies of the Federal Water
Pollution Control Administration.
-------
ABSTRACT
Laboratory studies of foam separation were conducted to determine the
feasibility of this process for reducing B.O.D., solids content, and
foaming tendency of clarified kraft mill effluent. Since kraft pulping
wastes have a natural tendency to foam, it was expected that the foaming
process, which has been found to be useful in treating domestic wastes,
might have applications in treatment of these effluents.
Both continuous flow and batch experiments were conducted, and liquid
and foam heights, liquid feed rates, air sparging rates, and tempera-
ture were varied over wide ranges.
The B.O.D. reduction in the treated liquid was disappointingly small,
averaging less than 5 per cent, and the B.O.D. enrichment in the foam
phase was in most cases less than 1.5 times that of the feed. Solids
removal was correspondingly low.
Foaming tendency, however, was significantly reduced by the intentional
foaming process with reductions of 40 to 60 per cent in this variable
being obtained. The reduction in foaming tendency was a strong function
of gas-to-liquid ratio with the most effective operating range being
between 1.0 and 1.5 SCFM/gallon.
The experimental results suggest that the reductions in B.O.D. and
foaming tendency were related to the separation of the tall oil com-
ponents of the waste. These components were concentrated in the foam
fraction, but they accounted for a maximum of only 10 to 12 per cent
of the B.O.D. of the raw feed. Apparently the remaining B.O.D.-
producing materials were not surface active and did not attach them-
selves to the surface-active components.
The cost of using a foam process on kraft mill wastes is estimated to
be four to five cents per 1000 gallons of feed; this cost is exclusive
of further processing of the concentrated foamate. Based on control of
foaming tendency alone, the process would be unattractive from a cost
standpoint.
This report was submitted in fulfillment of Grant No. WPRD 117-01-68
between the Federal Water Pollution Control Administration and Georgia
Kraft Company.
Key Words: Foam separation, pulping wastes, B.O.D. removal, tall oil
removal, foaming reduction, solids removal, treatment costs.
111
-------
TABLE OF CONTENTS
Page No.
Abstract iii
List of Figures vii
List of Tables ix
I. SUMMARY 1
II. CONCLUSIONS 5
III. RECOMMENDATIONS 7
IV. INTRODUCTION 9
A. Background Information 9
1. Theory and Fundamental Studies 9
2. Applications in Waste Treatment 11
3. Applications Other than Waste Treatment . . 11
4. Treatment of Kraft Mill Streams 12
B. Proposal to Investigate Foam Treatment of
Kraft Mill Wastes 12
V. EXPERIMENTAL METHODS 15
A. Equipment 15
B. Steady State Operation 18
1. Factorial Experiments 18
2. Effects of Other Design Features .... 19
C. Batch Operation 20
D. Analytical Tests and Procedures 21
1. Definition of Treatment 21
2. B.O.D. and C.O.D 24
3. Solids 24
4. Foaming Tendency 24
VI. EXPERIMENTAL RESULTS 29
A. Foaming Tendency 29
B. Foam Generation 29
-------
TABLE OF CONTENTS
Page No.
C. B.O.D., C.O.D., and Solids Removal 34
D. Batch Experiments 39
E. Chemical Analyses 41
VII. DISCUSSION OF RESULTS 43
A. Applications of Foam Separation 43
B. Relationship of Tall Oil Content to Foam Treatment . 44
C. Economics of the Process 45
D. Research for Improving Foam Treatment 46
References 49
Appendices
VI
-------
FIGURES
Figure Page No.
1. Schematic Diagram of Foam Separation Apparatus 16
2. Foam Separation Apparatus 17
3. Typical Graphical Representation of Experimental
Data for Batch Operation Using Results for Ten-
Foot Liquid Column 22
4. Graph for Evaluation of Enrichment Factor for
Ten-Foot Liquid Column 23
5. Foamability of Effluent 27
6. Reduction in Foaming Tendency 30
7. Foaming Tendency in Treated Liquid as Affected
by Reflux 31
8. Rate of Foam Production as Affected by Gas-to-
Liquid Ratio 32
9. Rate of Foam Production as Affected by Reflux
Ratio 33
10. Chemical Treatment as Function of Gas-to-Liquid
Ratio 36
11. Biological Treatment as Function of Gas-to-Liquid
Ratio 36
12. C.O.D. Reduction as Function of Temperature and
Liquid Column Height 37
13. B.O.D. Reduction as Function of Temperature and
Liquid Column Height 37
14. Reduction in B.O.D. Concentration as a Function
of Reflux 40
vii
-------
TABLES
Table Page No.
1. Summary of Results for Factorial Foam
Separation Experiments 38
2. Economic Analysis of Foam Separation 47
IX
-------
Foam Separation of Kraft Pulping Wastes
I. SUMMARY
Foam separation has been found useful to biologists, chemists, and
engineers as a means of separating and concentrating materials in
aqueous suspensions and solutions. Dissolved substances collect
at a gas-liquid interface in accordance with their surface activity,
and the more active ones can be separated from a solution by foaming
and can be concentrated by foam drainage. Less active substances
must be attached to a more active substance if they, too, are to be
removed. Separations have been achieved for a great variety of sub-
stances, including proteins, enzymes, fatty acids, detergents,
anions, cations, phenols, colloidally suspended particulates, and
biologically active components of domestic sewage. Studies of these
separations have ranqed from mere scientific inquiry to searches for
practical applications.
Since kraft pulping wastes contain natural surface-active materials,
it would appear that the dissolved and suspended components in these
wastes could likewise be removed by foaming. Previous studies have
indicated that foaming may be effective in treating kraft mill r
wastes, but there were essentially no data exploiting the potential
of the process. A proposal, therefore, was submitted to the Federal
Water Pollution Control Administration to explore the feasibility of
the method. The ultimate objective was to define an optimum process
design for removing components of high B.O.D. from kraft mill streams.
Specific objectives were to study the process as it was affected by
equipment design, gas and liquid flow rates, effluent composition,
and temperature.
The study was carried out in a cylindrical column in which liquid
descended countercurrently against rising bubbles of air which
generated the foam. The foam product was mechanically broken and
discharged from the column as a liquid. Part of the liquid was re-
turned to the column as reflux. Liquid feed entered through the
side of the column and treated liquid was removed from the bottom
of the column. The temperature of the process was adjusted by
heating the feed stream.
A series of steady state flow experiments was conducted to study
the effects of major variables. Liquid column height, liquid feed
rate, and gas sparging rate were considered the three basic operat-
ing variables and were investigated in a 33 factorial experiment.
The experimental plan was to locate the optimum combination of
these three variables and then investigate all others at this con-
dition; interactions would be clarified if necessary. The addi-
tional variables included sparger porosity, temperature, feed
location, foam height, waste stream properties, and pH.
-------
In addition to the flow experiments, batch experiments were per-
formed to ascertain the ultimate treatment attainable by foaming
and to study the effects of the individual operating variables
more precisely than was possible in the flow experiments.
The tests performed on the various streams included 6.0.0.5, C.O.D.,
total solids, total organic solids, tall oil, alkalinity, foaming
tendency, and pH. The tests were performed on the feed, treated
liquid, and foamed product streams. Experimental results showed
foaming tendency reduction to be the most significant change pro-
duced in the mill effluent by foam treatment. A reduction of at
least 40 per cent was always obtained, and under favorable condi-
tions as much as 60 per cent was found. Foaming tendency
reduction was a function primarily of the gas-to-liquid ratio,
and the most effective operating range was between 1.0 and 1.5
SCFM/gal with a reduction of 50 per cent.
The amount of foam produced by air sparging was also a signifi-
cantly varying function of the gas-to-liquid ratio and of the
reflux returned to the column. With no reflux, a certain minimum
amount of sparging air could be introduced to the foam column
without producing any foam product at all. Above this critical
air-to-liquid ratio, the foam production varied linearly with air
sparging to the maximum condition studied where 35 per cent of the
feed was converted to foam. When part of the foamed product was
returned to the column, the rate of foam production was greatly
increased. At a reflux of four (ratio of returned liquid to
product liquid) the foam production was over 90 per cent of the
feed rate, and at greater reflux values the foam production in-
creased gradually and asymptotically approached about 97 per cent
of the feed rate at infinite reflux (all foam product returned).
The reduction of B.O.D. was found to be surprisingly small in all
of these studies. Throughout the course of the experimental work,
the B.O.D. concentration in the feed stream was reduced no more
than 5 per cent, and the B.O.D. enrichment in the foam stream was
no more than 1.5 times that of the feed. No combination of physi-
cal operational variables produced any significant improvement in
treatment.
The experimental results suggested that the removal of B.O.D. is
related to the separation of tall oil by the foaming process. The
tall oil content was approximately 6 per cent of the total organic
content, and this amount was calculated to constitute approximately
12 per cent of the total B.O.D. Since only 50 per cent of the tall
oil at the most was removed by foaming in the flow experiments, the
maximum B.O.D. concentration reduction produced by the foaming pro-
cess as practiced was estimated to be around 6 per cent at best.
Apparently, the remaining B.O.D.-producing materials were not sur-
face active, did not attach themselves, to the surface of the
surface-active components, and were not removed.
-2-
-------
The C.O.D. concentration was reduced around 6 to 8 per cent and,
like B.O.D., corresponded closely to the amount of tall oil re-
moved from the effluent stream.
The experimental findings suggested that foam treatment might pos-
sibly be used for reducing foaming tendency of the mill effluent
to combat the formation of foam floes or "snow" on the aeration
ponds. The cost for such treatment was estimated to be four to
five cents per 1,000 gallons. With the present estimate, the cost
of treating effluent by foaming would be more than by using chemi-
cal defoamers, and additional costs would be incurred in the
disposal of the foamate product. The latter cost could be con-
siderable if the amount of foamate is not reduced to a minimum and
if the process is not operated as near infinite reflux as possible.
Foam separation treatment, on the other hand, would likely produce
consistent and positive control of the adverse foaming conditions
frequently occurring in the Rome mill treatment plant.
Operation of the foam separation process in a series of columns
was considered as a method for increasing the removal of B.O.D.
Calculations, however, indicated that the process was not adapt-
able to multistate operation since the number of columns required
to produce practical results would not be economical.
While foam treatment in this study was definitely limited, the pro-
cess could still have potential if chemical additives were found to
increase the surface activity of non-foamable compounds in the
liquid or to attach these compounds to others which are surface
active. If foam separation is ever to be effective on kraft mill
liquids for removing organic constituents, a program of fundamental
research should be performed to study the effects of potential
chemical additives.
-3-
-------
II. CONCLUSIONS
A. In the steady flow experiments the B.O.D. (ppm) was reduced
no more than 5 per cent in the primary clarifier effluent,
and the B.O.D. enrichment in the foam stream was no more than
1.5 times the B.O.D. in the feed. The maximum B.O.D. that
could be separated from primary clarifier effluent with ex-
tended treatment in the batch process was up to 10 to 15 per
cent. No combination of physical operational variables was
found capable of improving the B.O.D. removal.
B. The reduction of B.O.D. and C.O.D. appeared to be related to
the removal of tall oil which was the only organic constituent
specifically investigated. The tall oil amounted to about 5
to 6 per cent of the total organic content, and calculations
indicated that it could exert a maximum of 12 per cent of the
B.O.D. of the feed. Since analytical measurements indicated
that up to approximately one-half of the tall oil was removed,
the reductions in B.O.D. and C.O.D. were estimated to be
around 6 per cent, which compared favorably with the maximum
removal of 5 per cent found by foaming experiments.
C. Apparently the major B.O.D.-producing materials are not surface
active and do not attach themselves to those materials that are.
Measurements showed that the total solids and the total organic
content other than tall oil were practically unchanged by
foaming.
D. Foam separation could reduce foaming tendency up to 60 per cent,
and this reduction was probably related to about the same per-
centage of tall oil removal.
E. Foam separation could possibly be used to control foaming ten-
dency. The cost of such treatment would be four to five cents
per 1,000 gallons, exclusive of the cost for handling the
foamed product. Further development work might show a more
favorable cost estimate. The foamate disposal cost could be
minimized by operating as near infinite reflux as possible.
-5-
-------
III. RECOMMENDATIONS
A. Foam separation as it is presently understood is not recom-
mended for removing B.O.D. from kraft mill effluent.
B. Foam separation could be considered further as a potential
process for reducing foaming tendency of kraft mill effluent,
C. Future research should be conducted to determine suitable
additives for achieving the removal of non-surface-active
organic compounds in kraft mill effluent.
-7-
-------
IV. INTRODUCTION
A. Background Information
1. Theory and Fundamental Studies
Certain materials dissolved in aqueous systems have a
tendency to concentrate at a gas-liquid interface such as
found in the liquid films surrounding gas bubbles in foams.
The mechanism has been useful to biologists, chemists,
and engineers as a means of treating aqueous solutions and
suspensions for removing numerous materials, including pro-
teins, enzymes, fatty acids, detergents, anions, cations,
phenols, colloidally suspended particulates, and biologi-
cally active components of domestic sewage, to name some
of the more notable results. In special cases, surface-
inactive and weakly surface-active materials (those that
do not collect at an interface) have been removed by add-
ing foaming agents which attach to the material to be
separated. Foaming is usually produced by sparging air
through the liquid solutions.
The fundamentals of foam separation and foam fractionation
have been investigated in depth for quite a number of sys-
tems, and these processes, as currently understood, have
been recently summarized in excellent reviews by R. Lem-
lich (l & 2), R. B. Grieves, ejt. al_. (3), C. A. Brunner
and D. G. Stephan (4) and J. L. Rose and J. F. Sebald (B).
The more salient features of these reviews, together with
supporting information from other pertinent references,
served as the background for the study herein reported.
For more detailed study, the referenced articles are
recommended.
The collection of a solute at a gas-liquid interface,
under equilibrium conditions, is described by the Gibbs
equation (6)
) (1)
In this relationship, y is the surface tension of the iill
component and a\ is its activity; R is the gas constant,
T is the absolute temperature, and r^ is the "surface
excess", which is essentially the concentration of the
component i at the interface in units of mass per unit
area (e.g., grams per square centimeter). The equation
can be used to predict the concentration of a surface-
active material at an interface provided sufficient data
are available; such, however, is usually not the case for
most systems of interest because of their extreme com-
-9-
-------
plexity. Experience teaches that the variation of the
solute concentration at the surface is linearly related
to the bulk concentration in very dilute solutions and
tends to approach a constant value in more concentrated
solutions (l). Materials present in trace quantities
follow the linear portion of the curve, very dilute
metallic ions being an example, while major surfactants
in a foam usually involve the constant portion of the
surface-to-bulk liquid concentration relationship.
Although the theoretical evaluation of the relationship
might be difficult or even impossible, it can be derived
experimentally and used in theoretical analysis of foam
fractionation (1).
If at least one surface-active material is present, other
materials which may be surface inactive can, under care-
fully controlled conditions, also be removed from solu-
tion (?). In these cases the surface-inactive component
must be united to the surface-active material by some
means. Successful interactions have been achieved with
chemical processes such as chelation and hydrogen bonding
and physical actions, including ionic attraction, electro-
static attraction, and physical adsorption. Many surface-
inactive materials, including colloidal particulates (8)3
bacteria (9 & 10), metallic ions (11 & 12), proteins (13),
phenols (14), and selected organic compounds (15) have been
separated from aqueous solutions with foaming techniques.
As the surface-active properties of a dissolved substance
are responsible for its initial surface concentration
buildup, foam drainage is an equally dominant factor in
increasing the concentration further. Bubbles emerging
from a liquid entrain more liquid than can remain with
them; and, as they rise, liquid drains due to gravity and
pressure differences, leaving surface-active materials be-
hind to become more concentrated. These materials compete
in the concentration process according to their level of
surface activity and, as drained liquid flows downward
between bubbles, tend to concentrate in different parts
of the foam with the more surface active toward the top
and the less surface active toward the bottom. The result-
ing concentration and fractionation in foam drainage can
have practical significance (16).
The separation of materials from a liquid solution by in-
tentional foaming and the fractionation of these materials
have been thought to be useful in a continuous process
with properly controlled liquid and air flow rates and
product refluxing. Several fundamental analyses of the
process have therefore been made (3, 16, & 17). In general,
it is concluded that practical applications are possible
and that the operation is affected by the gas-to-liquid
-10-
-------
ratio, residence time of the foam and the liquid in the
process, the amount of reflux and its position, feed
position, bubble size, and temperature. There is also
strong evidence in one or more of the cited references
of the following:
a. Equilibrium at the gas-liquid interface in the liquid
pool is rapidly achieved and longer residence times
do not influence the separation (3).
b. The liquid behaves as a single stage, not appreciably
affected by height if the air sparger is at least one
foot submerged (13 3, & 17).
c. Foam is a good column packing and serves as a medium
for foam fractionation by liquid drainage over the
bubble surfaces, provided the liquid content is ap-
proximately 10 per cent or less (i).
d. Refluxing part of the foamed product aids fractionation
in a foam provided the liquid content of the foam does
not become too great (i).
e. A foam column can be operated as an enricher, a stripper,
or both (3).
f. Foam drainage produces enrichment and fractionation (16
& 18).
g. The overall separation in a foaming process is governed
by the separation occurring in the regions adjacent to
the foam-liquid solution interface (16).
h. Little is known about multicomponent effects (2).
2. Applications in Waste Treatment
Applied research has been conducted on the pilot plant
scale where detergents were added to foam treat domestic
wastes (43 5, & 19). In these experiments, very favorable
treatment was obtained. Removals of C.O.D. as high as 42
per cent and organic content removals as high as 65 to 75
per cent were reported. The costs were projected to be
three to five cents per 1000 gallons for most operations.
A very large plant (100 mgd) was estimated to operate for
approximately 1.5 cents per 1000 gallons.
3. Applications Other than Haste Treatment
Foaming is sometimes associated with flotation, which has
been widely employed in the mining and metallurgical in-
-11-
-------
dustries (2) and more recently in studies of water clari-
fication (20) and of domestic waste treatment (21).
Flotation is a physical process involving particles larger
than colloids that become attached to rising bubbles. While
surface-active agents may be used to enhance the attraction
of participates to bubbles, the process generally is not
one for concentrating dissolved chemicals (1). Certain
colloids, however, are an exception and can collect dis-
solved materials (8 & 11).
4. Treatment of Kraft Mill Streams
Although a considerable amount of research has been con-
ducted on foam separation processes, very little atten-
tion has been given to the foam treatment of kraft mill
liquids which foam naturally and profusely due to fatty
acid and resin acid soaps produced from the pulping of
southern soft woods (5). The National Council of the
Paper Industry for Air and Stream Improvement considered
foam separation in a study of color removal but aban-
doned it in favor of other processes (22). The use of
foam separation as a means of reducing B.O.D., however,
was recommended for further study. Harding (18) studied
foaming as it pertained to black liquor oxidation and
foam fractionation of tall oil. He found that tall oil
enrichment could be achieved. Foaming in black liquor
oxidation has been recommended as a means of increasing
tall oil yield (23). Finally, the profuse foaming in
kraft mill effluent treatment plants has caused unending
concern, and numerous additives have been developed to
combat the formation of uncontrollable amounts of foam,
but no attempt has been made to use these foams as a
method of removing the troublesome foam-making materials.
B. Proposal to Investigate Foam Treatment of Kraft Mill Wastes
If foam separation can be used to treat domestic sewage and
other aforementioned solutions of organic materials, some
with additives of a suitable surfactant, it would appear that
naturally foaming kraft mill wastes could likewise be treated.
Such a process, if effective, could be very useful in the
future when increased urbanization and industrialization place
greater demands on waste treatment facilities to preserve the
water resources of the nation. Combinations of several treat-
ment processes currently available (activated sludge, trick-
ling filters, electrodialysis, ion exchange, foaming freezing,
etc.) will undoubtedly be required to satisfy these needs of
the future. Each new treatment process which is shown to be
technically and economically feasible gives the design of
waste treatment plants added flexibility and new possibili-
-12-
-------
ties for decreased treatment costs. Foam separation is one
of the newer techniques which could offer promise for treating
industrial wastes.
Since there were essentially no data available on the treat-
ment of kraft mill effluent by foaming, a proposal was sub-
mitted to the Federal Water Pollution Control Administration
to explore the feasibility of the method. The study was
prefaced on the assumption that foaming would treat kraft
mill wastes, and the experimental program was designed to
yield data on the degree of treatment which could be expected
over a wide range of equipment design and operating conditions.
The data were to provide a basis for estimating the technical
and economic feasibility of the process and for scaling a
laboratory apparatus into a small commercial treatment facil-
ity. The overall objective was to define an optimum process
design for removing components of high B.O.D. from kraft mill
streams. The specific objectives for the study were:
1. To study the process as it was affected by
a. Foaming equipment design
b. Residence time of liquid and foamate
c. Effluent flow rate
d. Air to effluent flow ratio
e. Effluent composition
f. Temperature
2. To evaluate methods of breaking and disposing of the foam
generated.
3. To determine design data for a commercial or semi-
commercial foam separation installation.
4. To estimate the cost of treating kraft mill effluent by
foaming techniques.
As the study progressed, it was found that foaming was limited
in its potential to remove B.O.D. and that kraft mill effluent
was not treatable to a practical level in its natural state
and without supplementary chemical aids. The objectives,
therefore, were amended but were kept in the original framework
of the proposal. The amendments were to:
1. Limit the equipment design to cylindrical columns.
2. Study only fundamental problems and not attempt the devel-
opment of commercial design data.
3. Provide only very rough estimates of the cost of treating
kraft mill effluents by foaming techniques.
-13-
-------
V. EXPERIMENTAL METHODS
A. Equipment
The foam separation experiments were conducted in a cylindrical
column constructed as schematically represented in Figure 1 and
as shown pictorially in Figure 2. The design was principally
one in which liquid feed descended countercurrently against
rising bubbles of air. The column, 6.5 inches in diameter, was
made up of 24-inch flanged sections used as needed to vary the
column height from 2 to 12 feet. It contained the liquid pool
and the foam phase above it. The height of the liquid was con-
trolled by the hydrostatic leg mounted alongside the column.
Some flow was allowed to pass through the hydrostatic leg to
assure its remaining full. Liquid feed entered through a cir-
cular flow diffuser mounted through the side of one section;
changing the position of the section varied the feed position
in the column. A second diffuser was used for adding reflux
at positions other than at the feed level. Air introduced
through the sparger at the bottom of the column generated the
foam phase on rising through the liquid pool. Treated liquid
discharged from the base of the column, and foamed product rose
through the top. The air sparger was a standard fritted glass
filter available from most laboratory equipment supply houses.
Three porosities, having average pore sizes of 25-50, 70-100,
and 145-175 microns, were used. The effective sparger diameter
was 4 inches, and the assembly diameter was 5-1/4 inches. These
dimensions provided uniform coverage of sparging air across the
column and allowed a sufficient annular cross section for the
down flow of treated liquid. The upper face of the sparger was
approximately 2 inches from the column base.
The foam collector was a cubical chamber mounted above the
column. A squirrel cage air blower drew the foam from the
column and collapsed it. A perforated plate was positioned
atop the column to stabilize the foam withdrawal.
Input flow rates were measured with rotameters. The feed was
variable from 0.1 gpm to 2.0 gpm, and the sparging air rate was
variable from 0.1 cfm to 1.3 cfm. Reflux, product, and treated
liquid flows were measured by timed sample collections.
After leaving the foam breaker, the collapsed foamate was di-
vided into a reflux and a product stream. The reflux was
usually returned to the column with the feed stream, although
it could be added separately. The reflux was necessary to con-
trol the amount of product since the usually profuse foaming of
kraft mill effluent would remove more liquid as foam than could
ever be handled in a practical operation. Also, it was thought
that refluxing might induce fractionation of the surface-active
materials in the foam column.
-15-
-------
t
HF
Foam
AHL
HL
Diffuser
1Lo
Liquid
Level
Control
Foam
Breaker
Reflux
-A-
T Product
Feed
-Liquid Pool
! Rotameten
Manometer
Liquid
Feed
Heater
Feed Tank
Variable ^
Speed
Feed Pump
Air
Sparger
Air Sunply
I Treated
Liquid
FIGURE 1: SCHEMATIC DIAGRAM OF FOAM SEPARATION APPARATUS
-16-
-------
*
•„-
FIGURE 2: FOAM SEPARATION APPARATUS
-17-
-------
Nomenclature for column dimensions and operational variables
were:
#Q = Total column height, feet
#P = Foam height, feet
#Lo = Liquid height without air flow, feet
#l_ = Liquid height with air flow, feet
A#L = #L - #LQ, feet
F = Feed flow rate, GPM
G - Air flow rate, CFM
s = Bottom flow rate (bottoms meaning treated effluent), GPM
T = Tops flow rate (tops meaning total collapsed foam), GPM
P = Product flow rate, GPM
R = Reflux flow rate, GPM
Zp, ZB, Zj = Undesignated material concentration, or property
of tops, feed and bottoms, respectively, relating
to B.O.D., C.O.D., solids, etc.
B. Steady State Operation
1. Factorial Experiments
The column was designed to operate at steady flow condi-
tions, and the effects of its operating variables were to
be determined. Experiments were begun with a 33 factorial
set of experiments with the variables being liquid flow
rate, gas flow rate, and column height.
The raw material was primary clarifier effluent from the
pulping operations of Georgia Kraft Company at Rome,
Georgia; test samples from the primary clarifier were
withdrawn as needed over an extended period of time and
varied with day-to-day operation. The factorial design
was confounded to eliminate the first order effects of
the changes in feed properties (Appendix I-A). The three
levels of operation were at: (1) liquid flow rates of
0.2, 0.6, and 1.0 gal/min, (2) air flow rates of 0.13,
0.39, and 0.78 SCFM, and (3) column heights of 2, 4, and
8 feet.
-18-
-------
These conditions gave: (1) gas-to-liquid flow ratios from
0.13 to 3.9 SCF/gal, (2) gas flux rates from 0.07 to 3.9
SCFM/ft2, (3) liquid flux rates from 1.0 to 5 GPM/ft2, and
(4) liquid residence times from 3.4 to 69 min. These
levels were selected to span the ranges expected to be of
commercial importance. Work by Rose and Sebald (5) sug-
gested these levels would cover the range of practical
value. It was intended that the optimum operating con-
dition for the liquid feed flux, gas feed flux, and column
height would be derived from the factorial experiment. For
the resulting optimum, the effect of other pertinent design
variables would be investigated.
2. Effects of Other Design Features
Other features of the process which were thought to be
pertinent were investigated as follows:
a. Temperature Effects
A tube and shell, steam heated heat exchanger was pro-
vided to control the feed temperature and determine
the effects of this variable. Temperatures at three
levels, 75-850F, 110-115°F, and 135-145°F, were
investigated.
b. Feed and Reflux Location
The effects of the different positions of these streams
were investigated in experiments with a liquid column
of 4 feet and a foam column of 8 feet with:
(1) Both the feed and reflux at the liquid surface.
(2) The feed at the liquid surface and the reflux at
the top of the foam column.
(3) The feed 4 feet above the liquid surface and the
reflux at the top of the foam column.
(4) Both the feed and the reflux 4 feet above the
liquid surface.
c. Sparger Porosity
The effects of sparger porosity were studied with
spargers having average pore sizes of (1) 25-50 y
diameter, (2) 70-100 v diameter, and (3) 145-175 p
diameter.
-19-
-------
Each of the design features were separately investigated,
and due to the results as they became known, it was not
necessary to study the interactions of these variables
among themselves or with the factorial experiment
results.
C. Batch Operation
Batch operation of the foam column gave a supplementary means
of evaluating the potential of removing B.O.D. from the feed
liquid in terms of a characteristic enrichment factor and in
terms of the maximum amount of B.O.D. that could be removed
with extended foaming. The process was analyzed according to
the following development.
If B.O.D. or any similar quantity were considered as a material,
a balance was made on the process according to the relation,
(input) - (output) = accumulation. (2)
The input to the column was zero since the process was batch.
The output was proportional to the rate of foam production and
to the concentration of material in the column. Thus,
Output = kafy (3)
where M = Material in column
V = Volume of liquid in column
~ = Material concentration in liquid in column
a = Volumetric rate of foamate (liquid) production
k = Proportionality constant with the units of
Amount of B.O.D. per unit volume of colTapsed foam
Amount of B.O.D. per unit volume of column liquid
The proportionality constant, as denoted by its units, could
be shown to be the enrichment factor. The accumulation in the
column was -dw, and the overall balance was
t = -dAf (4)
which, when integrated, became
(5)
-20-
-------
V0 and M0 were the initial conditions, and by restrictions in
the integration a had to be a constant rate of removal. Since
the quantity (1 - y^) could be rearranged to
, at Vo - at _ V ,r\
i - — - — \D)
where V was the liquid left in the column at time t, the deri-
vation became
which had no restrictions on how V was removed.
The enrichment factors k were evaluated from the graphical
analysis of equation (7). The analysis was in two parts.
First, as shown by a typical graph in Figure 3, the volume of
liquid in the column and the B.O.D. in the column were plotted
as a function of air sparging time. This plot was used to
smooth the experimental data, although the actual data points
were carried in the final plot as demonstrated by Figure 4. In
this final plot and as required by equation (7), the logarithm
of the amount of B.O.D. in the column was plotted as a function
of the logarithm of the liquid volume in the column. The slope
of this line was the enrichment factor. The experimental data
and the calculated results necessary for constructing the log-
log graphs are presented in Appendix IV-A.
D. Analytical Tests and Procedures
1. Definition of Treatment
The parameters of primary importance to the study were
8.0.0.5, solids content (as total, organic, and mineral),
and foaming tendency. These tests were routinely per-
formed on the liquid feed to the column, the treated
liquid leaving the column, and the collapsed foam and
served as the basic measurements for the quantitative
evaluation of the foam treatment process. Additional
tests included determinations for C.O.D., tall oil con-
tent, alkalinity, and pH. These properties could pos-
sibly have been significant in describing the liquid
feed properties and in correlating the behavior of the
foaming process.
Treatment was measured in terms of concentration changes
and in terms of the fractional amount of B.O.D. removed
from the liquid feed. The concentration changes were ex-
pressed as enrichment, ^E., and as bottoms concentration
F
reduction, %&. The fraction of removal from the feed was
-21-
-------
QJ
(J
3
o
0)
32.0 -
24.0 -
16.0 -
8.0
0.0
0 20 40 60 80 100 120 140 160 180
Time, Minutes
FIGURE 3: TYPICAL GRAPHICAL REPRESENTATION OF EXPERIMENTAL DATA FOR
BATCH OPERATION USING RESULTS FOR TEN-FOOT LIQUID COLUMN
-22-
-------
4.6
= 4.5
o
o
en
4.4
fO
E
O)
in
E
fO
-5 4.3
o
CQ
CT>
O
4.2
4.1
1.5
• = Experimental BOD Points -
= Smoothed Data
= Initial Conditions Unknow
1.6 1.7
Log of Liquid Volume (liters) in Column
1.8
FIGURE 4: GRAPH FOR EVALUATION OF ENRICHMENT FACTOR
FOR TEN-FOOT LIQUID COLUMN
-23-
-------
p| - (|B.)-(|) where JgB was the B.O.D. in the treated
liquid and ipF was the B.O.D. in the feed. The concen-
tration ratios were indications of the separation poten-
tial of the foaming process and showed the partitioning
of surface-active constituents between the liquid layer
of the gas-liquid interface and the bulk liquid. The
fraction of removal, on the other hand, showed the actual
performance of the process in terms of how much B.O.D. was
removed from the main mill effluent liquid stream. These
treatment functions, while described specifically for B.O.D.
applied also to C.O.D., solids, and other constituents as
needed.
2. B.O.D. and C.O.D.
These determinations followed primary guidelines of national
standards. The B.O.D. procedure was that of ASTM D-2329-6ST.
The extent of dilution required was established by experi-
ment. The C.O.D. procedure was taken from ASTM D-1252-60,
except that reflux times of two hours were used. The long
reflux time seemed to produce more uniform results.
3. Solids
Solids determinations were made for the total solids, total
organic solids, and mineral content. The total solids was
the residue after evaporation of all water at 105°C; total
organic solids was taken as the material that could be
volatilized at a temperature of 600°C; and the mineral con-
tent was the final residue. The procedure was as outlined
by standard analytical procedures for waste waters (24).
The organic part of the solids included lignin, cellulose,
sugars, and fatty and resin acids, among numerous other
organic compounds in relatively smaller concentrations.
The fatty and resin acids were determined as tall oil by
the method of Saltsman and Kuiken (2S) and were reported
as tall oil. All other organic constituents were reported
collectively and were evaluated as the total organic con-
tent less the tall oil content. An attempt was made to
measure the lignin content by the absorption of ultra violet
light at 280 millimicrons following the Pearl-Benson (26)
method. The results from this test, unfortunately, were
not reliable due to color interferences from other organic
constituents in the mill effluent.
4. Foaming Tendency
Foaming tendency, or foaminess, was a property not directly
measurable and had to be expressed in terms of a related
-24-
-------
parameter. Numerous physical chemical properties have been
employed for this purpose, and some of the more frequently
considered have been foam properties--!'ncluding formation,
height, structure, drainage, and composition—and liquid
properties of the collapsed foam—including viscosity,
surface tension and optical and electrical properties.
Bikerman (27) defined foaming tendency in terms of the
average time a bubble remains entrapped in the foam layer.
His function was
l-T (8)
where I = average bubble residence time
v = volume of foam generated in time t,
and v = the volume of air introduced.
Carpenter and Gelman (28) applied this relationship to
kraft mill liquids and devised a test whereby a specified
volume of air was sparged at a very low rate into a
specified volume of liquid. The height of foam generated
was taken as a measure of foaming tendency.
The Carpenter and Gelman approach was tried, but some of
the mill effluent samples gave a foaming tendency of zero
while very profuse foaming was produced in the experimental
apparatus. The cause for these differing results was that
some liquids required a certain minimum air sparging rate
before a stable foam was generated. The air sparging rate
in the test was lower than this critical rate and the
sparging rate in the experimental foam column was larger.
The test for foaming tendency was modified in this study
so that all liquids gave a useful result. In the modified
procedure, the foaming tendency was measured in terms of
the maximum height, H*, that could be generated with a
given air rate. Equation (8) was transformed with the
substitutions
V = H* Ac,
t = Time
and V = G-AC
where H* = Maximum foam height
Ac = Column cross sectional area
G = Gas flow rate per unit time per unit cross
sectional area
-25-
-------
The transformed relationship was
y Vt_ _ H*'Ac'
*• ~ ''
and £, as a measure of foaming tendency, was the ratio of
H* to the specific gas flow rate. For a constant gas flow
rate, H* was proportionately related to J and could simi-
larly be taken as a measure of foaming tendency. The test
was evaluated experimentally and was found to reflect dif-
ferent H* values for liquids having visibly different
foaming characteristics. The test was imperfect, however,
in that H* was not truly a characterization number; it
should have been constant since the equation showed ~
G
to be constant for a selected air rate. H* was a function
of dynamics and was more nearly constant at low air rates.
In the experimental column, 6.5 inches in diameter, ^1 was
very nearly constant with less than 6 per cent variation
for gas rates up to 0.13 cubic feet of air per minute. At
double this gas rate, the ratio declined an additional 6
per cent but was beginning to fall more rapidly. Figure 5
shows these results.
H* as a measure of foaming tendency was determined in a
6. 5- inch diameter liquid column, with an air rate of 0.13
cubic feet per minute, and using a 6-inch liquid depth
above the face of the sparger. A smaller 2-inch column
was tried, but bubble coalescence caused plug flow of the
gas and gave erroneous and unreproducible results.
-26-
-------
12
10
O)
-------
VI. EXPERIMENTAL RESULTS
A. Foaming Tendency
The most significant change produced in the mill effluent by
the foam treatment process was the reduction of its tendency
to foam. As measured by the method developed for this study,
the foaming tendency of the mill effluent was reduced in most
experiments at least 40 per cent and, under favorable operat-
ing conditions, as much as 60 per cent. The extent to which
foaming tendency was reduced was controlled primarily by the
gas-to-liquid ratio as shown by the correlation in Figure 6.
The most effective operating range of the gas-to-liquid ratio
was between 1.0 and 1.5. This operating range achieved most
of the foaming tendency reduction, yet it did not require the
excessive use of air to gain a small additional increase. In
addition to being affected by the gas-to-liquid ratio, the
foaming tendency reduction was also affected significantly by
the amount of external reflux, a procedure used primarily to
control the amount of product produced. Surprisingly, how-
ever, a sizeable reduction could be obtained by foaming even
when all the product was refluxed. Figure 7 shows the foam-
ing tendency behavior of a typical treated liquid as infinite
reflux was approached. In this case the reduction at infinite
reflux was 40 per cent with a gas-to-liquid ratio of 1.5.
Other operating variables had less significant effects. In
the factorial experiments, raising the liquid pool height
from 2 to 8 feet improved the foaming tendency reduction 6
per cent. In other experiments, temperatures ranging from
83°F to 143°F gave no appreciable differences, and the air
spargers ranging in average pore size from 25-50 microns to
145-175 microns exhibited approximately the same performance.
The maximum foaming tendency reduction under the best condi-
tions was approximately 60 per cent.
B. Foam Generation
The amount of liquid that was transformed into foam depended
on the gas-to-liquid ratio and on the amount of foamed prod-
uct returned to the column as reflux. Figure 8 shows how the
volume of the collapsed foam, expressed as a fraction of the
feed, increased with increased air-to-liquid ratios when
there was no reflux. The significant feature of this result
is the intercept, which meant that a minimum amount of air
was required to produce any foaming at all. When part of the
foamed product was collapsed and returned to the column as
reflux, the rate of foamate production increased. Figure 9
shows this effect. Up to a reflux of 4, the rate of foam
generation increased 50 per cent; more reflux added only a
few percent.
-29-
-------
1.00
ITJ
a:
T3
C
O)
en
c
(13
O
a>
ai
i
o
in
e
o
o
CO
0.80 -
0.60 -
t! 0.40 -
O = 8 ft. column
A - 4 ft. column
• = 2 ft. column
1.0 2.0 3.0
G/L, Ft3 Air/Gal Feed
4.0
FIGURE 6: REDUCTION IN FOAMING TENDENCY
-30-
-------
0.6
3
CT
0.5
T3
QJ
fO
QJ
S_
-o
QJ
O)
O
*
O
*
0.4
0.3
Asymptote at _
Infinite Reflux
10 20 30
Reflux Ratio, (J - 1)
40
50
FIGURE 7: FOAMING TENDENCY OF TREATED LIQUID AS AFFECTED BY REFLUX
-31-
-------
0.30
03
Of.
-a
QJ
O)
(O
o
O)
E
0.20
0.10
0.00
e
0.2
e
Feed Foam Liquid
^Symbol gpm Ht. ft Ht.. ft
• 0.5 4 4
1.0
0.5
1.0
0.5
-------
i.oor
O)
O>
U-
I
o
o
u_
o
OJ
0.90 '
0.80 -
0.70
Asymptote at
Infinite Reflux
8
12
16
20
Reflux Ratio, (I - 1)
FIGURE 9: RATE OF FOAM PRODUCTION AS AFFECTED BY REFLUX RATIO
-33-
-------
C. B.O.D., C.O.D., and Solids Removal
The reduction of B.O.D. by foam separation was the primary
objective of this study. Unfortunately, however, the pro-
cess did not show the desired capability under the conditions
tested. Throughout the course of the experimental work, foam
separation was found to produce only limited amounts of B.O.D.
removal. In the steady flow experiments, approximately 10
per cent of the B.O.D. could be removed from the liquid
stream. The results were practically invariant with respect
to changes in liquid flow, gas flow, column height, air
sparger porosity, feed position, amount of reflux, and tempera-
ture. No combination of these physical operating variables
produced any significant improvement in treatment.
Not all of the 10 per cent B.O.D. reduction was due to con-
centration changes; some of the reduction was from the
partitioning of liquid between the treated liquid stream and
the foamed product. The total treatment was expressed by
1 . (*&)•(*)
'100
where (. • ) was the fraction of B.O.D. left in the treated
x? F liquid,
1 - (fB.) (|) was tne fraction of B.O.D. in the product
AF * stream,
^B. was treated liquid-to-feed concentration ratio,
^F
and £ was treated liquid-to-feed flow ratio.
F
The foaming column was operated with reflux so that the product
was removed to yield 5 per cent of the feed liquid as product
and 95 per cent as treated liquid or bottoms. Thus, - was 0.95.
The concentration ratio, ^B, was found also to be approxi-
*F
mately 0.95. The treatment then was
Pi - (0.95) (0.95)1 100 = 10%
The B.O.D. in the mill effluent was to some degree related to
the C.O.D. and the total solids content. Measurements of the
C.O.D. showed a reduction of approximately 15 per cent, which
was slightly greater than the B.O.D. reduction. There were no
appreciable changes produced by varying the operating variables
including gas flow, liquid flow, liquid column height, foam
-34-
-------
height, and temperature. The solids content was not measurably
reduced under any conditions.
Figures 10 and 11 show the B.O.D. and C.O.D. treatment analyzed
in terms of the gas-to-liquid ratio, and Figures 12 and 13 pre-
sent the study of temperature effects for three column heights.
The experimental points had considerable scatter (standard
error in B.O.D. ± 12 per cent); however, considering the number
of measurements taken, it is evident that a practical degree of
treatment was not obtained unless, of course, the foam separa-
tion could be applied repetitively as the stages in distilla-
tion. The data as shown included the partitioning factor, ^L,
and without it the B.O.D. and C.O.D. reductions would not F
even be as large as indicated. The unreduced experimental
results are tabulated in Appendices I-A and I-B.
The factorial experiments were designed to reveal the effects
of column height, gas rate, and liquid rate and were the
source of the data for Figures 10 and 11. When these data
were analyzed again for the first order effects of column
height, gas rate, and liquid rate, the results shown in Table 1
were obtained. Again the calculations were made with the func-
tion
1 - (£) (±R)1 , where x was the B.O.D., C.O.D., and solids
corresponding to each of the three columns of data. No statis-
tically significant variations were found. The B.O.D. reduction
(0.25) for the 8- foot column was believed to be in error. Sup-
plementary experiments confirmed that less reduction than 0.25
should have been indicated.
The effects of feed and reflux position were investigated with
a column having a liquid pool height of 4 feet and a foam
height of 8 feet. Positions were tried with the feed and re-
flux (1) both at the liquid pool surface, (2) both at the
middle of the foam column, (3) the reflux at the top of the
foam and the feed at the middle of the foam column, and (4)
the reflux at the top of the foam and the feed at the pool sur-
face. The treatment (Appendix III) was no different from that
already found in all other experiments.
Theoretical consideration indicated that refluxing of part of
the collapsed foam product could have an effect on the enrich-
ment of B.O.D. in the foam and on the stripping of B.O.D. from
the treated liquid. The reasoning was based on the fact that
some reflux would enrich the concentration of B.O.D. -contain-
ing compounds in the column and the enrichment would increase
the rate of the separation process as would be experienced in
similar operations, e.g., distillation. If the reflux were
infinite, on the other hand, there would be no reduction in
-35-
-------
a
• , — i
• ca| LL.
" ' 0.00
•r— cd Ll_
c
0 1
• r-
•(-> i—
0 ' 1
"giS 0-20
0.40
A O %
° A • o A
A »
• t? o
" 2 ^ o - 8 ft. column
^ A. A = 4 ft. column
• = 2 ft. column
A i i i
0.0
1.0 2.0
Gas-to-Liquid Ratio
3.0
4.0
FIGURE 10: CHEMICAL TREATMENT AS FUNCTION OF GAS-TO-LIQUID RATIO
o
0
CO
J_
o
+J
u
•a
n
a:
'co|ui
,
od u_
><|X
i
uTL
fC
-0.
0.
0.
0.
0.
20
00
20
40
60
^ A
A A ' O
V 9 v
• n A
A A V 0
" -A
• A ^
o = 8
0 A = 4
o • = 2
i i i
•
A
ft. column0
ft. column
ft. column
0.0
1.0 2.0 .3.0
Gas-to-Liquid Ratio
4.0
FIGURE 11: BIOLOGICAL TREATMENT AS FUNCTION OF GAS-TO-LIQUID RATIO
-36-
-------
0.00
ca|u_
o
Jm|
c x|x
c
o
3
T3
C£-
to
(O
0.20 -
0.40
-
0
2 A
is c
i i
2 4
• = 83°F
A = 113°F
0 = 143°F
1
6
9
I
8
Liquid Column Height
FIGURE 12: C.O.D. REDUCTION AS FUNCTION OF TEMPERATURE
AND LIQUID COLUMN HEIGHT
u.uu,
u.
c
colu-
o
'5 i 0-20
-o
0) in
c:
-------
TABLE 1
SUMMARY OF RESULTS FOR FACTORIAL FOAM SEPARATION EXPERIMENTS
Operating Variables
and Their
Magnitudes
Liquid Feed (6PM)
0.2
0.6
1.0
Column Height (ft.)
2
4
8
Air Rate (CFM)
0.13
0.39
0.78
Treatment in Terms of Fraction Reduced For
Foaming
Tendency
0.49
0.33
0.30
0.36
0.37
0.42
0.33
0.38
0.43
B.O.D.
0.13
0.15
0.14
0.10
0.08
0.25
0.13
0.13
0.17
C.O.D.
0.15
0.19
0.12
0.16
0.18
0.12
0.19
0.16
0.12
Solids
0.03
0.08
-0.05
0.01
0
-0.07
-0.05
0.08
0.03
-38-
-------
B.O.D. because a steady state mass balance would require all
the B.O.D. compounds to remain with the treated liquid. Con-
centration changes would exist in the column, however. Mea-
surement of the B.O.D. concentration in the refluxing liquid
did show some increase at infinite reflux, as would be ex-
pected. Somewhere between the no reflux and infinite reflux
conditions, there should have been an optimum amount of
reflux. Experimental data were collected over a wide range
of refluxing conditions. Some of the data indicated a favor-
able change, but repeated measurements failed to establish a
consistent increase in treatment at any reflux. It was con-
cluded that reflux did not have an appreciable effect. The
treatment as a function of reflux is given in Figure 14, and
the experimental data are tabulated in Appendix III.
Supplementary experiments investigated the treatment that
could be obtained with very short liquid columns. It was
found (Appendix III) that liquid columns down to one-half
foot were practically as effective as all others studied.
D. Batch Experiments
Batch experiments were designed to determine the maximum
amount of treatment that could be obtained in a single foam
separation stage and to examine more closely the B.O.D. en-
richment factor, which was the B.O.D. concentration in the
foam divided by the B.O.D. in the liquid pool. The effect
of initial liquid pool height, foam height, air sparger
porosity, air rate, temperature, and acidification (pH ad-
justment) were investigated. The treatability of several
different mill streams was also studied. The experimental
results are given in Appendix IV-A.
The enrichment factors evaluated from batch experiments are
presented in Appendix IV-B. All the factors were in the range
of 1.0 to 1.5 and indicated a very weak dependence on the
operating variables observed. Enrichment ratios in this
range correspond to the 5-10 per cent levels of treatment
found in the steady flow experiments. The exact effect of
changing the operating conditions was not discernable. For
instance, the enrichment factor as a function of temperature
was 1.22, 1.34, and 1.44 for 800, liQOf and 136QF, respec-
tively, and showed a trend toward higher enrichment with
increased temperature. This trend, however, was so slight
that a recalculation of the results on the performance of
additional experiments could have nullified or even reversed
it.
As noted in the calculated data, some experiments exhibited
a discontinuity between the feed liquor B.O.D. point and the
-39-
-------
•o
•r-
13
c
Ol
u
c
0
a
o
CO
TJ
OJ
O)
u_
c
c
o
(C
c
u
c
o
o
a
o
CO
0
0
0
0
0
.00
.10
.20
.30
.40
A
-
0
1 •
9
5
V
I • » :
_ •
s
1 1 1 1 1 1 1 1
1 2345678
Reflux Ratio (I - 1)
FIGURE 14: REDUCTION IN B.O.D. CONCENTRATION AS A FUNCTION OF REFLUX
-40-
-------
intercept of the straight line which seemed to fit the B.O.D.
measurement as a function of sparging time. In several experi-
ments when this discontinuity was present, the enrichment
factor was usually small and near 1.00 to 1.05. In others
when the discontinuity was not apparent, larger enrichment
factors were calculated tending to be closer to 1.5. It was
thought that enrichment between 1.0 and 1.5 was definitely
occurring, but differences within this range could not be
regarded as significant.
The acidification experiments were different from others in
the study in that a precipitate formed with decreasing pH.
The data in Appendix IV-A, Table 7 shows the B.O.D. to be
lowered also with decreasing pH. This B.O.D. removal was
probably due to precipitate formation which carried with it
B.O.D.-producing materials. An attempt was made in the ex-
periment to keep the precipitate in suspension by stirring
as the pH was reduced. At each pH studied the reduced B.O.D.
(at time = 0) was brought about by the air initially sweeping
out a considerable amount of the precipitate.
With regard to the maximum B.O.D. reduction, the batch experi-
ments proved conclusively that foam separation was limited. In
the total of some 20 to 30 valid batch experiments, the B.O.D.
was never reduced more than 15 per cent even after one to two
hours of sparging, and the liquid volume was reduced over 60
per cent. On the average, B.O.D. was reduced 5 to 10 per cent
E. Chemical Analyses
The basic measurements of B.O.D., C.O.D., and solids content
revealed practically no change in the treated liquid as com-
pared with the feed material and indicated there was probably
very little change in chemical composition. The fact that
foaming tendency was reduced by foam treatment would indicate
that a chemical change had taken place; however, the extent
could be small since foaming could be greatly affected by only
trace quantities of dissolved materials. It was suspected
that the fatty acid and resin acid content of southern pine
was one of the chief materials responsible for the intense
foaming in southern kraft pulp mills, and tests were devised
to measure changes in these materials (tall oil) specifically.
Analyses were made of the feed, treated liquid, and foamed
product streams in the flow experiments, of the liquid and
foam in the batch experiments and of some foams and liquids
associated with the waste treatment system at Georgia Kraft
Company's Krannert Division mill. The results are given in
Appendix V. The analyses are reported as total solids, ash
or mineral content, tall oil, and organic materials exclusive
of tall oil.
-41-
-------
The analyses for the flow experiments (Appendix V, Table 1)
confirmed that no major change in total solids or organic con-
tent took place, and they showed that the tall oil content,
amounting to about 5 to 6 per cent of the total organic
materials, was slightly, but measurably, reduced. Since the
tall oil concentration measurements were accurate to only one
significant figure, the small changes in tall oil contents
were not detected in some cases. The average of all results,
having both feed and treated liquid analysis, showed a tall oil
content reduction of about 30 per cent. In batch experiments
a definite tall oil removal was found, and the foam phase at
the beginning of the experiment was notably enriched. Further,
the liquid remaining in the column at the end of the batch
experiments was definitely reduced in tall oil by at least 50
per cent and probably much better. The batch experimental ap-
proach was believed to give more reliable data in studying the
foaming process.
It is notable that foams taken from the primary and secondary
clarifiers of Georgia Kraft Company's Krannert Division were
more dense and contained greater solids content than those
generated in the laboratory. This condition probably resulted
from very stable and well-drained foams which did not allow
the residual solids to return to the bulk liquid beneath. Foam
from the primary clarifier effluent contained 0.64 per cent
solids, of which 28.8 per cent was tall oil. The liquid from
which this foam came had 0.15 per cent solids and a tall oil
content of only 0.006 per cent. Analytical results for the
secondary clarifier foam had the same trend except the solids
concentrations were larger.
-42-
-------
VII. DISCUSSION OF RESULTS
A. Applications of Foam Separation
The ideal objective of a foam separation process for treating
kraft mill waste would be economical simultaneous reduction
of B.O.D. and foaming tendency. In such a process it would
be necessary to concentrate highly the B.O.D. and foam produc-
ing constituents in the foamate which should represent a small
fraction of the raw effluent volume; the concentrated foamate
stream should then be treated more economically than the total
volume of raw waste. If B.O.D. reductions comparable to other
treatment methods could be obtained, the added benefit of foam
prevention would make the process very attractive.
Unfortunately, the results obtained here indicate that signif-
icant B.O.D. removal does not occur upon foaming. And while
the process may find justification in some specific cases, it
is not likely that the process would generally be economical
based on foaming tendency reduction alone.
While the tendency to foam was not completely removed, the
reduction was consistent, uniform, and probably more effec-
tive than additives usually employed to control foaming. In
the operation of pulp mills, additives have been found unpre-
dictable in controlling foaming and have been exceedingly
difficult to apply in trace amounts as evidenced by the per-
sistence of foam even with liberal applications of numerous
additives, all of which presumably yield satisfactory
laboratory tests.
If it were found desirable to use the process to reduce foam-
ing tendency, the data in this report could be used for pre-
liminary design. Experiments showed that under practically
any set of conditions a reduction of at least 40 per cent
could be obtained, and as much as 60 per cent was possible
under the most favorable conditions. Since the foaming
tendency of mill effluent was found to vary from a low to a
high value by approximately this amount during normal opera-
tion, such a foam treatment system could be employed to con-
trol the foaming tendency of the mill effluent below what is
now the minimum foaming tendency. The ultimate effect would
be to aid the prevention of foam floes or "snow" which at
times are formed in the effluent treatment systems of kraft
mills.
The process could be operated with a gas-to-liquid ratio of
1.0 to 1.5 SCFM/gal and with as little as 2 feet of liquid
depth. Since the process was found effective in reducing
foaming tendency even as infinite reflux was approached, the
reflux would be adjusted to give perhaps as little a foam
-43-
-------
product rate as 0.1 per cent of the feed rate. The separated
product (collapsed liquid) would be disposed of by any of
numerous methods such as burning in the chemical recovery
boilers. For a mill with 15 million gallons of effluent per
day, the product would be approximately 10 gallons per
minute which could be easily handled.
It is conceivable that the addition of other chemicals to the
raw effluent could improve the treatment obtained by foaming.
Although study of additives was beyond the scope of this study,
one series of experiments was run in which the pH was adjusted
over a wide range. Acidification produced a precipitate which
was removed by the foam, and this resulted in a significant
B.O.D. reduction. The cost of acidification and final neutral-
ization of a large volume of kraft mill effluent would be pro-
hibitive, however. Also, it is likely that the precipitated
material could be removed more easily by filtration or
settling than by foaming. It is possible, however, that other
additives could be found which would tend to cause the dis-
solved organics to become more surface active or to attach
themselves to those surface-active materials present.
It is also possible that by using more than one foam column in
series improved treatment could be obtained. The approach,
however, does not now appear feasible for two reasons. One
is the apparent linkage between foam treatment and tall oil
content. Since the tall oil content of kraft mill effluent
is low (less than 5 per cent of organics), treatment could be
limited regardless of the process used. This is supported by
the fact that, when the liquid was given extended treatment
in the batch experiments, the B.O.D. could be reduced a maxi-
mum of only about 10 per cent. In the batch experiments, the
liquid was foamed until no more foam could be expelled from
the column and the liquid volume was decreased to about 40
per cent of its original volume.
The second reason is the treatment found experimentally is
so low that an impractical number of stages would be required
to achieve a commercially effective separation. All calcu-
lations relating to the flow experiments indicated that the
liquid treatment ranged from practically no reduction in B.O.D.
to a maximum reduction of 5 per cent with a most probable value
about the middle of the range. Calculations were made for a
number of multiple column arrangements using this degree of
treatment in each column, and the number of columns required
to obtain practical separations was always prohibitive.
B. Relationship of Tall Oil Content to Foam Treatment
Analysis for tall oil showed that this material, amounting to
approximately 5 per cent of the total organic material, was
-44-
-------
measurably reduced by foam treatment. The reduction in the
steady flow experiments was probably as much as 50 per cent,
but the accuracy in measuring such small amounts (0.004 per
cent by weight) made the result only approximate. In the
batch experiments, the tall oil content decreased from 0.004-
0.006 per cent to about 0.001 per cent after extended foaming
and the tendency to foam was depleted even with increasing
air flow. These findings indicated that foam treatment was
related at least partially to the tall oil content of the
mill effluent.
The relationship between tall oil content and the reduction
of B.O.D. and C.O.D. could be estimated from the oxidation
reactions of typical fatty acids and resin acids chemically
approximating the major tall oil constituents. With stearic
and abietic acid representing the typical components, it
could be calculated that the maximum B.O.D. attributable to
the tall oil components would be only about 10-12 per cent
of the total B.O.D. of the raw feed. Thus, a 50 per cent
removal of the tall oil would correspond to the 5-6 per cent
B.O.D. reductions which were found experimentally.
Foaming tendency was also found to be related at least
partially to the tall oil content of the mill effluent. The
batch experiments demonstrated the relationship; these studies
operated as long as foaming could be sustained even with in-
creasing air ratios and the tall oil content diminished with
time. In the studies of reflux ratios in the flow experi-
ments, where it was found that foaming tendency was reduced
even at infinite reflux, it is suspected that tall oil was
being concentrated in the foam and separating. Thus, there
is a possibility that some of this by-product might be re-
claimed in the process. However, the tall oil normally
found in the effluent amounted to only about 4 to 5 tons per
day or no more than 5 per cent of the tall oil available to
the mill, and recovery of half of this amount would not pro-
vide enough economic incentive for installing the process.
C. Economics of the Process
The removal of the B.O.D. was too small to warrant any prac-
tical considerations, The process, however, was moderately
effective in reducing foaming tendency and might conceivably
be used for this purpose.
The experimental column was approximately 0.2 square feet in
cross section and was operated successfully with liquid rates
up to 1.0 GPM. The required air rate to get optimum foaming
tendency reduction was from 1.0 to 1.5 SCFM of air per gallon
of liquid. These data reduce to 5 GPM/ft2 and 5.0 to 7.5
-45-
-------
SCFM/ft^ and may be used for preliminary sizing of a com-
mercial process for reducing foaming tendency.
A rough cost estimate was made for the Krannert Division of
the Georgia Kraft Company to relate the cost of foam treat-
ment to the cost of chemical additives to control foaming
in the effluent treatment plant. For a 15 million gallon
per day liquid rate and a 7.5 SCFM/gal air rate, the esti-
mate (Table 2) gave a unit cost of 4.97 cents/1000 gallons
exclusive of the cost for treating any foamed product that
might be removed. Chemical additives presently control
foam at a cost considerably less than this. The reduction
of foaming tendency by physical treatment, therefore, has
an unfavorable cost, although further refinements of the
process might reduce this cost to approximate that re-
quired by chemical additives.
Rose and Sebald (5) recently estimated the cost of a foam
separation process for treating domestic sewage at the rate
of 10 million gallons per day with a liquid loading of
5 GPM/ft2 and an air flow of 3 SCFM/ft2. The unit cost was
approximately two cents per 1000 gallons. Scaling this re-
sult according to their recommendations to a 15-mi11 ion-
gal lon-per-day plant with an air flow of 7.5 SCFM/ft^ gives
4.5 cents per 1000 gallons. This cost is slightly lower
than that estimated in this study for treating kraft mill
effluent. The major factor causing Rose and Sebald's
modified estimate to be lower is a much smaller power re-
quirement. Capital costs, on the other hand, were esti-
mated to be about 0.85 cents per 1000 gallons greater for
the domestic sewage system than for the kraft mill effluent
system. With all factors considered, both estimates
indicate a cost of around 5 cents per 1000 gallons, which is
too high to be of immediate interest.
D. Research for Improving Foam Treatment
Foam treatment as practiced in this study was definitely
limited. The process, nevertheless, may still have potential
if chemical additives can be found to increase the surface
activity of non-foamable compounds in the liquor or to
attach these compounds to others which are surface active.
Research in this area was beyond the scope of this project.
It is felt that a fundamental study should be undertaken
to find means for accomplishing separation; subsequent
investigations would then attempt practical applications.
-46-
-------
TABLE 2
ECONOMIC ANALYSIS OF FOAM SEPARATION*
I. Capital Costs
A. Equipment Costs Total Costs
Tanks and Footings $ 41,500
Pumps 28,000
Air Compressors 120,000
Foam Breakers 28,200
Total Equipment $217,700
B. Equipment Installation
(150% of equipment costs) 326,500
C. Total Capital 544,200
II. Annual Operating Costs
A. Capital with 20-Year Life
Depreciation - 5%
Interest - 7%
Taxes & Ins. - 6%
W $ 97,900
B. Power at $100 per HP year 145,500
C, Maintenance at 2% of Capital
Cost 10,900
D. Labor 7.000
E. Total Operating Costs $261,300
III. Unit Operating Cost; 4.97 cents/1000 gallons
*Basis: Liquid rate = 15 million gallons/day; air rate = 7.5 standard
cubic feet/minute/gallon liquid.
-47-
-------
REFERENCES
1. temlich, R., "Adsorptive Bubble Separation Methods, Foam Fractiona-
tion and Allied Techniques," Industrial and Engineering Chemistry, 60,
No. 10, pp 16-29 (1969).
2. Lemlich, R., "Questions and Answers on ... Foam Fractionation,"
Chemical Engineering, 75, No. 27, pp 96-102 (1968).
3. Grieves, R.B., and Wood, R.K., "Continuous Foam Fractionation: The
Effect of Operating Variat
No. 4, pp 456-460 (1964).
Effect of Operating Variables on Separation," AlChE Journal, 10,
(ir...
4. Brunner, C.A., and Stephen, D.G., "Foam Fractionation," Industrial
and Engineering Chemistry, 57, No. 5, pp 40-48 (1965).
5. Rose, 0. L., and Sebald, J.F., "Treatment of Waste Waters by Foam
Fractionation," TAPPI, 51, No. 7, pp 314-321 (1968).
6. Gibbs, J.W., Collected Works, Volume I, Longmans, Green, and Company,
New York (1928).
7. Sebba, F., "Organic Ion Flotation," NATURE, 188, No. 4752, pp 736-737
(1960).
8. Grieves, R.B., Bhattacharyya, D., and Crandall, C.J., "Foam Separa-
tion of Colloidal Particulates," Journal of Applied Chemistry, 17,
(June, 1967), pp 163-168.
9. Bretz, H.W., Wang, S.L., and Grieves, R.B., "Variables Affecting
the Foam Separation of Escherichia coli", Applied Microbiology, 14,
No. 5, pp 778-783 (1966).
10. Grieves, R.B., and Wang, S.L., "Foam Separation of Pseudomonas
Fluorescens, and Baccilus Subtilis Var. Niger," Applied Microbiology,
15, No. 1, pp 76-81 (1967).
11. Rubin, A.J., and Johnson, J.D., "Effect of pH on Ion and Precipitate
Flotation Systems," Analytical Chemistry, S9, No. 3, pp 298-302 (1967)
12. Grieves, R.B., and Ettelt, G.A., "Continuous, Dissolved-Air Ion
Flotation of Hexavalent Chromium," AIChE Journal, 13, No. 6, pp 1167-
1171 (1967).
13. Schnepf, R.W., and Gaden, E.L., Jr., "Foam Fractionation of Proteins:
Concentration of Aqueous Solutions of Bovine Serum Albumin," Journal
of Biochemical and Microbiological technology and Engineering, 1,
No. 1, pp 1-8 (1959).
-49-
-------
14. Grieves, R.B., and Aronica, R.C., "Foam Separation of Phenol with
a Cationic Surfactant," Air and Water Pollution International
Journal,10, pp 31-40, Pergamon Press, Great Britain (1966).
15. Karger, B.L., and Rogers, L.B., "Foam Fractionation of Organic
Compounds," Analytical Chemistry, 33, No. 9, pp 1165-1168 (1961).
16. Grieves, R.B., and Bhattacharyya, D., "The Foam Separation Process:
A Model for Waste Treatment Applications," J. Water Pollution Con-
trol Federation, 37, No. 7, pp 980-989 (1965).
17. Grieves, R.B., and Wood, R.K., "Effect of the Foam-Liquid Solution
Interface on Continuous Foam Fractionation," NATURE, 200, No. 4904,
pp 332-335 (1963).
18. Harding, C.I., Foam Fractionation in Kraft Black Liquor Oxidation,
PhD Thesis, University of Florida, Gainesville, Florida (Dec., 1963).
19. Editorial Staff, "Foam Separation, Foaming by Intention," Environ-
mental Science and Technology, 1, No. 2, pp 116-118 (1967).
20. Grieves, R.B., "Foam Separation for Water Clarification," J. of the
Sanitary Engineering Division, SA-1, Feb. 1966, pp 41-54.
21. Westinghouse Electric Corporation, commercial literature on Puripak
Compact Sewage Treatment System for Localized Disposal, Cheswick,
Pennsylvania.
22. National Council for Air and Stream Improvement, "Color Removal and
BOD Reduction in Kraft Effluents by Foam Separation," Technical
Bulletin No. 17?, September 1964.
y
23. Chemical Construction Company, Pollution Control Division, commercial
literature, New York.
24. Standard Methods for the Examination of Water and WasteiMter, 12th ed.,
American Public Health Association, Inc., New York, 1965.
25. Sa'tsman, W., and Kuiken, K.A., "Estimation of Tall Oil in Sulfate
Black Liquor," TAPPI, 42, No. 11, pp 873-874 (1959).
26. Browning, B.L., Methods of Wood Chemistry, Volume II, John Wiley &
Sons, New York (1967).
27. Bikerman, J.J., Foams, Theory, and Industrial Applications, Reinhold
Publishing Corp., New York (1953).
28. Carpenter, W.L., and Gelman, I., "Measurement, Control and Changes
in Foaming Characteristics of Pulping Wastes During Biological
Treatment," paper presented at the 20th Alkaline Pulping Conference
(TAPPI) 1966.
-50-
-------
APPENDICES
Page No.
I-A. Factorial Experiment Design 53
Table 1: Confounding of Factorial Experiments
into Three Blocks 53
I-B. Experimental Data for Factorial Experiments .... 55
Table 1: Experimental Data for Factorial Experiments. 55
Table 2: Experimental Data for Block II of
Factorial Experiments 56
Table 3: Experimental Data for Block III of
Factorial Experiments 57
Table 4: Experimental Foaming Tendency Data
for Factorial Experiments 58
Table 5: Effects of Liquid Feed Rate, Gas Rate, and
Liquid Column Height on B.O.D. Reduction. . 59
Table 6: Effects of Liquid Feed Rate, Gas Rate, and
Liquid Column Height on C.O.D. Reduction. . 60
Table 7: Effects of Liquid Rate, Gas Rate, and
Liquid Column Height on Solids Reduction. . 61
Table 8: Effects of Liquid Feed Rate, Gas Rate, and
Liquid Column Height on Foaming
Tendency Reduction 62
II. Study of Temperature Effects with Steady Flow
Operation 63
Table 1: Experimental Data for Study of Temperature
Effect with Varying Liquid Column Height. . 63
III. Study of Effect of Feed and Reflux Position and Effect
of Using Short Liquid Columns 65
Table 1: Experimental and Calculated Data for Steady
Flow Operation with Varying Feed and Reflux
Positions 65
Table 2: Experimental and Calculated Data for Steady
Flow Operation with Short Liquid Column
Heights 65
Table 3: B.O.D. Determinations for Varying Reflux
Ratios in Steady Flow Operation 66
IV-A. Experimental and Calculated Data for Batch Experiments 67
Table 1: Experimental and Calculated Data for Batch
Study of Column Height Effects 67
-51-
-------
Table 2: Experimental and Calculated Data for
Batch Study of Sparger Porosity Effects . .
Table 3: Experimental and Calculated Data for
Batch Study of Air Rate Effect
Table 4: Experimental and Calculated Data for
Batch Study of Foam Height Effects. . . .
Table 5: Experimental and Calculated Date for
Batch Study of Mill Effluent Tributary Streams
Table 6: Experimental and Calculated Data for Batch
Study of Temperature Effects
Table 7: Experimental and Calculated Data for Batch
Study of Adjusted pH Effects
IV-B. B.O.D. Treatment as Evaluated from Batch Experiments .
Table 1: Enrichment Factors and B.O.D. Discontinuity
Data for Batch Experiments
Page
Analytical Data
Table 1: Analysis of Feed, Treated Liquid, and
Product in Flow Experiments ....
Table 2: Tall Oil Analyses in Batch Experiments
Table 3: Analyses of Streams Taken from Krannert
Division Waste Treatment Plant .
69
70
71
72
74
75
77
77
79
79
80
81
-52-
-------
APPENDIX I-A
FACTORIAL EXPERIMENT DESIGN
TABLE 1 ...
CONFOUNDING OF FACTORIAL EXPERIMENTS INTO THREE BLOCKS11'
Block I Block II Block III
LQ. GO, CQ LO, GO, C2 LO, GI, CQ
LI. GI, C0 LI, GI, C2 LI, G2, C0
1-2. 62. GO Lg, 62, C2 Lg, GO, GO
LO, 62, GI LO, GI, GI LO, GO, GI
'-1, GO, Cl Ll, 62, Cl Ll, Gl, Cl
L2, Gl, Cl L2, GO, Cl L2, G2, Cl
LO, GI, Cg LO, G2, GO LO. G2, Cg
Ll, G2, C2 Ll, GO, CO Ll, GO, C2
L2, GO, C2 L2, GI , CO 1^, GI , Cg
(1) Each block run with one feed. Each experiment defined by
a liquid rate, L; a gas rate, G; and a liquid column height,
C. The values for each were:
LO • 0.2 GPM Go • 0.13 SCFM CQ » 2 ft.
LI = 0.6 GPM GI = 0.39 SCFM Cl = 4 ft.
L2 " 1.0 GPM Gg = 0.78 SCFM C2 - 8 ft.
-53-
-------
APPENDIX I-B
Experimental
Conditions
LQ, GO, CQ
I], GI, C0
LZ. GZ. Co
Lo» G2» C]
L! , GQ, C]
L£, G], C]
•
in
in
i I r r
L0, Gi, Co
EXPERIMENTAL
EXPERIMENTAL DATA
BOD
PPM
360
360
300
300
240
220
220
240
260
COD
mg/1
1005
1060
1070
895
850
1060
880
770
790
Feed
Solids
0.114
0.112
0.127
0.118
0.172
0.137
0.117
0.121
0.120
DATA FOR FACTORIAL EXPERIMENTS
FOR Bl
TABLE 1
.OCK I OF FACTOR
Collapsed Foam
Na
250
220
225
250
240
235
230
250
245
pH
8.0
8.1
8.2
7.9
8.2
8.2
7.1
8.1
8.0
BOD
PPt"
440
440
320
300
400
300
240
280
300
COD
mg/i
1310
1140
1180
1230
830
1140
1005
800
1090
Solids
0.139
0.125
0.125
0.131
0.134
0.130
0.132
0.121
0.146
IAL EXPERIMENTS
(Tops)
ppm
260
260
240
255
250
250
255
255
255
Treated
pH
8.0
8.2
8.2
8.3
8.1
8.2
7.8
8.2
8.0
BOD
380
300
200
220
340
280
180
260
240
COD
960
880
910
865
785
815
770
705
690
Liquor (Bottoms)
Solids
0.120
0.117
0.124
0.112
0.123
0.117
0.092
0.116
0.118
ppm
225
250
245
260
230
245
250
250
250
pH
7.9
8.1
8.1
8.3
7.9
8.2
8.0
8.2
8.2
-------
APPENDIX I-B
(Continued)
TABLE 2
Experimental
Conditions
i LQ, GO, C£
8? Ll, GI , C2
' L2, G2, C2
LO, Gl, C]
Ll, G2, Cj
L2, GO, Ci
LO, G2, CQ
Ll, Go, CQ
L2, GI, CQ
EXPERIMENTAL DATA
BOO
ppm
260
220
160
320
280
280
180
240
160
COD
mg/1
375
660
335
640
640
735
370
480
415
Feed
Solids
%
0.112
0.103
0.102
0.097
0.099
0.094
0.102
0.101
0.103
FOR BLOCK II
OF FACTORIAL EXPERIMENTS
Collapsed Foam (Tops)
Na
PPm
215
200
185
185
185
185
195
185
170
pH
9.1
9.2
9.2
8.2
8.8
9.0
8.2
8.1
8.1
BOD
PPm
400
220
200
340
260
500
240
320
180
COD
mg/1
1240
1002
335
850
735
1002
640
1002
480
Solids
%
0.137
0.114
0.109
0.106
0.101
0.144
0.106
0.122
0.113
Na
ppm
205
190
185
170
195
190
210
200
200
pH
8.8
9.1
8.9
8.5
8.7
8.8
8.0
8.0
8.0
Treated
BOD
PPm
240
220
180
280
240
220
160
120
120
COD
mg/1
305
530
335
610
590
590
355
465
415
Liquor (Bottoms)
Solids
%
0.107
0.102
0.109
0.099
0.114
0.097
0.104
0.105
0.100
"a
PPm
195
200
185
175
175
180
175
175
155
PH
9.1
9.1
8.8
8.7
8.6
9.0
8.1
8.1
8.0
-------
APPENDIX I-B
(Continued)
TABLE 3
en
Experimental
Conditions
LO, GI, CQ
LI, 62, CQ
1-2. GQ, C0
LO, Go,
Ll, GI,
L2» 62,
LO, 62, C2
LI, GO, C2
L2, GI, C2
EXPERIMENTAL
BOD
ppm
360
360
340
340
340
340
360
480
380
COD
mg/1
350
720
575
545
545
367
350
378
385
Feed
Solids
%
0.114
0.111
0.111
0.059
0.111
0.105
0.107
0.113
0.118
DATA FOR BLOCK III
OF FACTORIAL EXPERIMENTS
Collapsed Foam (Tops)
Na
ppm
220
230
210
205
210
210
200
200
200
pH
9.0
8.9
9.0
8.9
8.8
8.7
8.7
8.8
8.8
BOD
PPi"
360
480
380
480
400
360
380
810
420
COD
mg/1
480
800
945
990
670
400
545
610
510
Solids
%
0.116
0.116
0.131
0.153
0.120
0.115
0.114
0.169
0.128
Na
ppm
220
230
220
215
225
220
215
215
220
PH
8.9
8.8
8.8
8.8
8.7
8.6
8.6
8.6
8.7
Treated
BOD
PPti
340
220
300
300
320
320
360
360
340
COD
mg/1
320
670
495
480
447
350
335
320
350
Liquor (Bottoms)
Solids
%
0.111
0.105
0.120
0.103
0.105
0.108
0.112
0.111
0.121
Na
ppm
220
210
195
195
205
200
200
205
205
PH
8.9
8.8
8.9
8.9
8.8
8.6
8.6
8.8
8.7
-------
APPENDIX I-B - (Continued)
I
en
00
TABLE 4
EXPERIMENTAL
FOAMING TENDENCY DATA FOR FACTORIAL EXPERIMENTS
Block I
Experimental
Conditions
LO.
Ll,
L2.
LO.
Ll,
L2.
LO.
Ll,
L2.
GO,
GO,*
GI,
62.
GO,
CD
Co
CD
Cl
Cl
Cl
C2
C2
C2
H*, Inches
Feed Top
37 52
48
35 56
59
66
55
55
Bottom
26
18
16
21
25
17
30
32
20
Block II
Experimental H*,
Conditions Feed
LO.
Ll,
L2.
LO.
Ll,
L2,
LO,
Ll,
L2,
GO,
Gl,
G2.
Gl,
62,
GO,
G2,
GO,
Gl,
C2 52
C2
Ci 51
Cl
CO
CO
CO
Block III
Inches
Top
65
57
60
56
Bottom
31
38
17
44
36
25
26
Experimental
Conditions
LO.
Ll,
LO,
Ll,
L2,
LO,
Ll,
L2,
Gl,
G2.
GO.
GO,
Gl,
G2,
G2,
GO,
Gl,
Co
CD
Cl
C2
C2
C2
H*, Inches
Feed Top
37 60
58
43
64
43 47
Bottom
28
22
25
25
17
17
30
29
17
-------
APPENDIX I-B
I
en
EFFECTS OF LIQUID
Air Rate Liquid Rate
Levels Levels
(SCFM) (gpm)
0.2
0.12 0.6
1.0
0.2
0.39 0.6
1.0
0.2
0.78 0.6
1.0
(1) Percent BOD Reduction
(Continued)
TABLE 5
FEED RATE, GAS RATE, AND LIQUID COLUMN HEIGH
% BOD Reductions for Mean Value
the Column Heights for Each
2 ft 4 tt. 8 ft Air RateU)
+ 12 +21 +14
+29 - 34 +52 +13
+12 +16 +0
+15 - 21 +37
+ 5 +11 +21 +13
+ 18 +17 +10
- 7 +11 +37
- 8 +19 +42 +17
+15 +27 +16
Mean Experimental Value
for Each Column Ht.^)
+ 10 +8 +25
f— ~K
= 100 1 - j£fj |jl) . J5&J = Concentration
1 V*N \r}\ (Xc)
IT ON B.O.D. REDUCTION^1
Partial Mean Value
Liquid Rates
0.2 gpm 0.6 gpm 1
+ 16
+ 16
+ 10
+ 12
+ 14
+ 18
Total Mean Value for
Feed Rate\4)
+13 +15
Ratio. \y( = Partition
)
for
.0 gpm
+ 9
+ 15
+ 19
Each
+ 14
Ratio
(2) Equals First Order Effect of Air Rate.
(3) Equals First Order Effect of Column Height.
(4) Equals First Order Effect of Feed Rate.
-------
o
I
APPENDIX I-B - (Continued)
TABLE 6
EFFECTS
Air Rate
Level s
(SCFM)
0.13
0.39
0.78
OF LIQUID FEED
Liquid Rate
Levels
(gpm)
0.2
0.6
1.0
0.2
0.6
1.0
0.2
0.6
RATE, GAS RATE, AND LIQUID
% COD Reductions for
the Column Heights
2 ft 4 ft 8 ft
+ 17 +24 +18
+ 20 +38 +8
+22 +16 +9
+ 14 +28 +5
+ 20 +18 +21
+ 17 +9 +9
+ 5 +5 +19
+ 24 +12 +11
+ 9 +8 +9
Mean Experimental Value
for Each Column Ht.(3)
COLUMN HEIGHT
Mean Value
for Each
Air Rate(zJ
+ 19
+ 16
+ 11
ON C.O.D. REDUCTION*1'
Partial Mean Value
Liquid Rates
0.2 gpm 0.6 gpm
+ 20
+ 22
+ 16
+ 20
+ 10
+ 16
Total Mean Value for
Liquid Rate(4)
for
1 .0 gpm
+ 16
+ 12
+ 9
Each
+ 16
+18
+12
+15
+19
+12
(1) Percent COD Reduction = 100
1 -
(Mj = Concentration Ratio; |-j
(XFJ (F)
= Partition Ratio = 0.95.
(2) Equals First Order Effect of Air Rate.
(3) Equals First Order Effect of Column Height.
(4) Equals First Order Effect of Feed Rate.
-------
APPENDIX I-B
(Continued)
TABLE 7
EFFECTS OF LIQUID RATE, GAS RATE, AND LIQUID COLUMN HEIGHT ON SOLIDS REDUCTION^
Air Rate Liquid Rat*
Levels Levels
(SCFM) (gpm)
0.13 0.2
0.6
1.0
0.39 0.2
0.6
1.0
0.78 0.2
0.6
1.0
2 % Solids Reduction for Mean Value
the Column Heights for Each
2 ft
- 14
+ 1
- 15
+ 6
+ 1
- 3
+ 7
+ 10
+ 3
4 ft
+ 2
+ 32
- 75
+ 19
+ 19
+ 3
+ 2
- 9
+ 10
Mean Experimental
for Each Column Ht
Partial Mean Value for
Liquid Rates
8 ft Air RateU; 0.2 gpm 0.6 gpm 1.0 gpm
+ 7
+ 6 - 5
+ 9
+ 3
+ 6 +8
+ 25
- 2
+ 2 +3
+ 1
Value
.(3)
- 1.7
+ 13
- 27
+ 9.3
+ 8.7
+ 8.3
+ 2.3
+ 1
+ 4.7
Total Mean Value for Each
Liquid Rate(4)
8
- 5
(1) Percent Solids Reduction =100 .
(Xp) (F)
(2) Equals First Order Effect of Air Rate.
(3) Equals First Order Effect of Column Height.
(4) Equals First Order Effect of Feed Rate.
(XF)
= Concentration Ratio; \^( = Partition Ratio = 0.95.
-------
APPENDIX I-B - (Continued)
TABLE 8
EFFECTS OF LIQUID FEED RATE, GAS RATE, AND LIQUID COLUMN HEIGHT i
Air Rate
Levels
(SCFM)
0.13
0.39
0.78
Liquid Rate % Foaming Tendency Reduc- Mean Value
Levels tion for the Column Ht. for Eacfo x
(gpm)
0.2
0.6
1.0
0.2
0.6
1.0
0.2
0.6
1.0
2 ft
43
32
40
60
27
14
67
9
30
Mean
for
36
4 ft
29
29
29
51
51
15
51
-
40
Experimental
Each Column Ht
37
8 ft Air Rate**'
32 33
-
30
49
51 38
24
57
41 43
51
Value
.(3)
42
ON FOAMING TENDENCY REDUCTION*1 >
Partial Mean Value
Liquid Rates
0.2 gpm 0.6 gpm 1
35
31
53
43
58
25
Total Mean Value for
Feed Rate<4)
49 33
for
.0 gpm
33
18
40
Each
30
(1) Percent Foaming Tendency Reduction = 100
1 - ;^\;|\ ; , \ - Ratio of bottom-to-Feed Foaming Tendency
L (XF)(F)J (XF)
= Partition Ratio = 0.95; Xp = Average of all measurements in feed.
(2) Equals First Order Effect of Air Rate.
(3) Equals First Order Effect of Column Height.
(4) Equals First Order Effect of Feed Rate.
-------
APPENDIX II
STUDY OF TEMPERATURE EFFECTS WITH STEADY FLOW OPERATION
TABLE 1
EXPERIMENTAL DATA FOR STUDY OF TEMPERATURE EFFECT
UITH VARYING LIQUID COLUMN HEIGHT
Column
Height 83QF 113°F 143°F
(Ft) Sample B.O.D. C.O.D. B.O.D. C.O.D. B.O.D. C.O.D.'
2 Feed 480 880
Bottom 420 850
Top 500 975
Feed 180 915
Bottom 160 865
Top 260 1070
Feed 220 880
Bottom 180 815
Top 240 1260
4 Feed 460 945
Bottom 420 880
Top 480 1120
Feed 460 865
Bottom 420 815
TOD 480 910
Feed 440 865
Bottom 400 800
Top 440 990
8 Feed 440 815
Bottom 440 735
Top 440 975
Feed 420 815
Bottom 380 783
Top 480 910
Feed 440 865
Bottom 420 830
Top 460 970
-63-
-------
APPENDIX III
STUDY OF EFFECT OF FEED AND REFLUX POSITION
AND EFFECT OF USING SHORT LIOUID COLUMNS
TABLE 1
EXPERIMENTAL AND CALCULATED DATA FOR STEADY FLOW OPERATION
WITH VARYING FEED AND REFLUX POSITIONS'
BOD Treatment
Position as Height Experimental Concentration Ratios
Above Liquid
FeedReflux
4 4
4 8
8 0
0 0
Conditions: Foam Height = 8 feet; liquid height = 4 feet; ^ = 1.5;
B = 0.95. Data were averages of three measurements. L
******
TABLE 2
BOD Data* (ppm) for
Feed
283
315
335
300
Bottom
270
308
338
317
Product
300
325
445
383
/BOD
1BOD
0
0
1
1
of Bv
of F;
.95
.98
.01
.06
,BOD
^BOD
1
1
1
1
Of
of
.06
.03
.33
.28
p.
F;
EXPERIMENTAL AND CALCULATED DATA FOR STEADY FLOW OPERATION
Liquid Column
Height (Ft.)
0.5
2.5
4.5
6.5
WITH SHORT LIOUID COLUMN
Experimental
BOD Data* (ppm) for
Feed Bottoms Product
500 495 535
470 445 533
500 480 555
445 472 523
HEIGHTS
BOD Treatment
Concentration Ratios
/BOD of B\ (BOD of P)
BOD of F BOD of F
0.99 1.07
0.95 1.13
0.96 1.11
1.06 1.18
G R
Conditions: Foam height = 5.5 feet; £• = 1.5, ° = 0.95. Data averaged
•P v*r\m fr\i i v* ma a c 11 b*amAn + e «
from four measurements.
-65-
-------
APPENDIX III
(Continued)
TABLE 3
B.O.D. DETERMINATIONS FOR VARYING REFLUX RATIOS
IN STEADY FLOW OPERATION
B.O.D. Concentration (ppm) in
Reflux Ratio Feed Bottom Top
0 300 284 350
250 350
284 384
317 367
1.6 300 232 310
250 340
217 250
232 390
3.6 324 217 300
232 318
232 284
217 300
4.1 313 300 550
280 300
280 330
330 280
5.2 298 250 350
250 330
280 300
220 320
7.4 322 267
300
2,84
267
16.7 324 250
267
284
Infinite 270 300
250
250
-66-
-------
APPENDIX IV-A
EXPERIMENTAL AND CALCULATED DATA FOR BATCH EXPERIMENTS
TABLE 1
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF COLUMN HEIGHT EFFECTS
Experimental Data
Initial
Liquid
Column
Height
(ft)
10
8
4
Sparging
Time
(min)
0
10
30
50
70
90
110
130
143
0
5
25
55,
85
100
0
10
30
50
70
82
Liquid
Volume
in Column
(1)
65
63
59
55
51
47
43
39
38
52.1
51.1
47.1
41.1
35.1
31.0
26
24
20
16
12
10
Measured
BOD
(ppm)
400
370
380
380
380
380
380
380
385
430
390
--
380
380
390
430
390
420
400
380
420
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
_ _
380
380
380
380
380
380
380
380
•• —
385
385
385
385
385
__
400
400
400
400
400
Data Req
Evaluati
'd. by Equation (7)for
ng Enrichment Factor
Log of BOD (gms)
in Col
Measured
Points
4.416
4.368
4.352
4.320
4.288
4.252
4.214
4.172
4.160
4.352
4.300
4.194
4.126
4.083
4.049
3.972
3.925
3.807
3.682
3.622
umn
Smoothed
Points
__
4.380
4.352
4.320
4.288
4.252
4.214
4.172
4.160
__
4.294
4.270
4.200
4.132
4.078
_ _
3.983
3.904
3.807
3.682
3.603
Log of
Liquid Volume
in Column
1.814
1.800
1.772
1.741
1.708
1.673
1.644
1.592
1.580
1.718
1.710
1.674
1.614
1.546
1.492
1.416
1.381
1.302
1.205
1.080
1.000
—(Continued)—
-------
TABLE 1 - (Continued)
Experimental Data
Data Req'd. by Equation (7) for
Evaluating Enrichment Factor
en
CO
Initial
Liquid
Column
Height
(ft)
6
8
10
Sparging
Time
(min)
0
17
57
87
117
0
18
48
78
133
151
0
20
92
124
154
181
Liquid
Volume
in Column
(1)
39.1
35.8
28.0
22.1
16.9
52.1
48.3
42.5
34.4
25.5
22.6
65.1
61.3
47.1
40.6
34.5
30.4
Measured
BOD
(ppm)
567
535
500
517
517
568
435
485
517
467
467
590
568
535
500
467
467
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
534
527
522
517
497
486
477
462
460
560
516
497
477
462
Log of BOD (gms)
in Column
Measured
Points
4.347
4.283
4.146
4.059
3.942
4.472
4.314
4.315
5.251
4.077
4.024
4.586
4.542
4.402
4.308
4.208
4.153
Smoothed
Points
4.282
4.170
4.062
3.942
4.378
4.316
4.216
4.072
4.018
4.536
4.388
4.306
4.217
4J48
Log of
Liquid Volume
in Column
1.593
1.554
1.448
1.345
1.228
1.718
1.684
1.629
1.537
1.408
1.354
1.816
1.788
1.674
1.610
1.538
1.484
-------
APPENDIX IV-A
(Continued)
TABLE 2
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF SPARGER POROSITY EFFECTS
Experimental Data
Average
Sparger
Porisity
(Microns)
Sparging
Time
(min)
Liquid
Volume
in Column
(1)
Measured
BOD
(ppm)
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
25-50
IO
I
145-175
70-100
0
5
30
50
65
0
5
20
45
75
0
5
20
40
55
52.1
48.7
33.7
27.3
25.7
52.1
50.1
44.5
36.6
33.6
52.1
48.1
36.5
26.9
25.5
440
480
420
400
440
450
470
420
420
430
450
450
420
400
400
441
438
433
430
448
443
434
422
444
428
407
392
Data Req'd. by Equation (7) for
Evaluating Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.361
4.371
4.140
4.049
4.054
4.372
4.373
4.272
4.188
4.154
4.372
4.336
4.186
4.032
4.009
Smoothed
Points
4.325
4.170
4.074
4.044
4.353
4.295
4.201
4.146
4.330
4.195
4.040
4.000
Log of
Liquid Volume
in Column
1.718
1.688
1.528
1.437
1.410
1.718
1.700
1.649
1.564
1.520
1.718
1.684
1.563
1.430
1.408
-------
I
-«J
o
APPENDIX IV-A
(Continued)
TABLE 3
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF AIR RATE EFFECT
Experimental Data
Air Rate
(SCFM)
0.26
0.52
0.78
Sparging
Time
(min)
0
5
15
30
45
69
0
10
20
32
43
0
5
15
25
34
Liquid
Volume
in Column
(1)
52.1
51.1
46.1
41.6
39.3
38.6
52.1
41.3
33.5
30.6
29.9
52.1
43.1
29.8
24.0
22.7
Measured
BOD
(ppm)
310
320
330
300
280
280
300
320
280
270
270
320
280
320
270
270
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
315
310
302
294
292
292
285
276
268
310
292
275
259
Data Req'd. by Equation (7) for
Evaluating Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.210
4.206
4.182
4.093
4.052
4.047
4.195
4.122
3.973
3.918
3.908
4.223
4.082
3.980
3.812
3.804
Smoothed
Points
4.200
4.156
4.100
4.063
4.040
4.082
3.981
3.927
3.905
4.127
3.940
3.820
3.770
Log of
Liquid Volume
in Column
1.718
1.700
1.665
1.620
1.600
1.590
1.718
1.616
1.526
1.486
1.476
1.718
1.636
1.475
1.381
1.357
-------
---J
I
APPENDIX IV-A
(Continued)
TABLE 4
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF FOAM HEIGHT EFFECTS
Experimental Data
Foam
Column
Height
(ft)
8
4
2
Sparging
Time
(min)
0
5
10
15
20
23
0
5
10
15
25
0
5
10
15
26
Liquid
Volume
in Column
(1)
26
23.6
21.7
20.6
20.1
20.0
26.0
23.3
21.1
19.6
19.2
26.0
23.1
20.8
19.3
18.8
Measured
BOD
(ppm)
400
425
420
400
420
390
380
410
400
380
400
400
380
350
370
400
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
408
408
408
408
408
392
392
392
392
389
388
387
386
Data Req'd. by Equations (7) for
Evaluating Enrichment Factor
Log of BOD (gms)
in Col
Measured
Points
4.018
4.002
3.960
3.917
3.926
3.892
4.018
3.918
3.926
3.872
3.886
4.018
3.985
3.908
3.874
3.877
umn
Smoothed
Points
3.984
3.948
3.926
3.914
3.912
3.961
3.918
3.886
3.877
3.954
3.908
3.874
3.862
Log of
Liquid Volume
in Column
1.416
1.374
1.337
1.314
1.304
1.302
1.416
1.368
1.325
1.292
1.284
1.416
1.364
1.319
1.286
1.275
-------
APPENDIX IV-A
(Continued)
TABLE 5
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF MILL EFFLUENT TRIBUTARY STREAMS
Experimental Data
ro
Stream
6th Effect
Condensate
6th Effect
Condensate
Sparging
Time
(min)
0
5
10
20
30
45
0
5
20
25
40
50
60
75
Liquid
Volume
in Column
(1)
52.1
48.1
44.3
37.1
30.9
23.7
52.1
48.1
44.6
42.9
38.0
35.2
32.9
30.7
Measured
Points
(ppm)
840
880
870
820
840
840
940
870
820
800
800
820
820
845
Smoothed BOD Values
From Straight Line
Best Representing
Time Versus BOD
(ppm)
850
850
850
850
850
850
825
825
825
825
825
825
825
Data Req'd. by Equation (7 ) for
Evaluating Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.648
4.628
4.587
4.484
4.415
4.300
4.691
4.622
4.564
4.536
4.484
4.461
4.432
4.415
Smoothed
Points
4.612
4.577
4.500
4.420
4.306
4.600
4.567
4.549
4.497
4.464
4.436
4.404
Log of
Liquid Volume
in Column
1.718
1.683
1.649
1.570
1.491
1.376
1.718
1.683
1.650
1.633
1.580
1.547
1.518
1.488
—Continued—
-------
I
^1
CO
1
Experimental Data
TABLE 5
[Continued)
Smoothed BOD Values
Stream
Small
Flume
Large
Flume
Sparging
Time
(min)
0
7
20
45
60
70
85
94
0
5
15
30
45
60
90
100
129
Liquid
Volume
in Column
(1)
52.1
50.7
48.1
43.1
40.1
38.1
35.1
33.5
52.1
51.1
49.1
46.1
43.1
40.1
34.1
32.1
26.7
Measured
Points
(ppm.)
750
675
650
700
650
700
600
625
600
600
600
600
600
650
650
600
550
From Straight
Line
Best Representing
Time Versus
(ppm)
665
661
652
647
644
638
635
600
600
600
600
600
600
600
600
BOD
Data Req'd. by Equation (7) for
Evaluating Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.593
4.535
4.496
4.481
4.417
4.426
4.324
4.328
4.496
4.488
4.470
4.443
4.414
4.417
4.346
4.286
4.168
Smoothed
Points
4.529
4.504
4.450
4.415
4.390
4.351
4.329
4.488
4.470
4.443
4.414
4.382
4.312
4.286
4.206
Log of
Liquid Volume
in Column
1.718
1.706
1.684
1.635
1.604
1.582
1.546
1.526
1.718
1.710
1.692
1.665
1.636
1.604
1.534
1.507
1.427
-------
APPENDIX IV-A
(Continued)
TABLE 6
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF TEMPERATURE EFFECTS
Experimental Data
Temperature
wi
1360
110°
83°
Sparging
Time
(min)
0
5
20
50
62
0
5
25
45
68
81
0
5
25
55
70
90
107
Liquid
Volume
in Column
(1)
52.1
51.1
48.1
42.1
40.8
52.1
51.1
47.1
43.1
39.0
37.0
52.1
51.1
47.1
41.1
38.1
34.1
31.8
Measured
Points
(ppm)
480
470
450
450
415
450
450
420
380
430
400
450
435
450
430
380
400
375
Smoothed BOD Values
from Straight Line
Best Representing
Time Versus BOD
(ppm)
465
455
434
426
444
433
422
407
400
440
429
412
403
392
362
Data Req'd. by Equation (7)
for
Evaluating Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.400
4.382
4.336
4.279
4.230
4.371
4.363
4.298
4.216
4.225
4.171
4.371
4.348
4.328
4.248
4.162
4.136
4.077
Smoothed
Points
4.377
4.341
4.263
4.238
4.357
4.312
4.260
4.201
4.171
4.354
4.306
4.230
4.187
4.127
4.062
Log of
Liquid Volume
in Column
1.718
1.710
1 .684
1.626
1.606
1.718
1.710
1.674
1.636
1.592
1.569
1.718
1.710
1.674
1.615
1.582
1.534
1.503
-------
APPENDIX IV-A
TABLE 7
EXPERIMENTAL AND CALCULATED DATA FOR BATCH STUDY OF ADJUSTED pH EFFECTS
Experimental Data
PH
10.0
01
I
8.3
7.0
5.8
3.8
Sparging
Time
(min)
0
15
25
35
45
57
0
15
25
35
47
0
10
20
31
35
0
11
20
35
0
11
20
25
Liquid
Volume
in Column
(D
Measured
Points
(ppm)
52.1
40.1
32.3
26.5
22.9
19.7
52.1
40.4
33.5
29.4
27.5
52.1
44.0
37.5
34.5
34.1
52.1
45.5
40.9
38.8
52.1
51.7
51.5
51.4
600
640
600
640
585
550
530
455
450
540
475
500
480
470
450
430
450
430
420
430
370
425
410
430
Smoothed BOD Values
from Straight Line
Best Representing
Time Versus BOD
617
607
598
589
578
506
492
478
461
483
465
447
439
432
427
419
420
420
420
Data Req1
Evaluati
d. by Equation (7 ) for
ng Enrichment Factor
Log of BOD (gms)
in Column
Measured
Points
4.496
4.411
4.288
4.230
4.128
4.036
4.432
4.265
4.179
4.201
4.117
4.417
4.326
4.247
4.192
4.166
4.371
4.292
4.235
4.223
4.286
4.343
4.325
4.344
Log of
Smoothed Liquid Volume
Points
4.394
4.293
4.201
4.130
4.057
4.312
4.218
4.148
4.104
4.328
4.242
4.189
4.176
4.294
4.242
4.212
4.338
4.336
4.334
in Column
1.718
1.604
1.510
1.424
1.360
1.295
1.718
1.606
1.526
1.469
1.430
1.718
1.644
1.575
1.538
1.534
1.718
1.659
1.612
1.590
1.718
1.714
1.712
1.711
-------
APPENDIX IV-B
BOD TREATMENT AS EVALUATED FROM BATCH EXPERIMENTS
TABLE 1
ENRICHMENT
Experimental Variables
and Their Magnitudes
Column Height
4 feet
8 feet
10 feet
6 feet
8 feet
10 feet
Temperature
830F
110
136
Sparger Porosity
20-50 microns
75-100
145-175
Air Rate
0.26 SCFM
0.52
0.78
Foam Height
2 feet
8
Effect of Different
Feed Streams
Pulp Mill Effluent
Paper Mill Effluent
6th Effect Condensate
Acidification
(Effect of pH)
3.8
5.8
7.0
8.3
10.0
FACTORS AND BOD DISCONTINUITY
FOR BATCH EXPERIMENTS
Enrichment
Factor
1.01
1.00
1.02
1.04
1.11
1.32
1.22
1.34
1.44
1.13
1.10
1.04
1.36
1.12
1.15
1.00
1.00
1.12
1.01
1.00
1.00
1.11
1.22
1.19
1.06
DATA
Discontinuity
at Time Zero
8.3
5.0
9.5
-7.2
12.6
1.7
3.2
1.1
-2.1
0
0
1.8
-2.6
0
0
0
0
0
12
6
-12.5
2.7
0
0.4
5.0
-77-
-------
APPENDIX V
I
^J
i-O
ANALYTICAL DATA
TABLE 1
ANALYSIS OF FEED, TREATED LIQUID
, AND PRODUCT IN FLOW EXPERIMENTS
Composition of Each Stream as Weight, %
Experiment
1
2
3
4
5
6
7
8
9
10
11
12
13
Total
Solids
0.143
0.216
0.215
0.227
0.226
0.160
0.163
0.162
0.158
0.150
0.161
0.153
0.159
Feed
Tall
Ash Oil
0.122 0.003
0.100 0.005
0.101 0.005
0.108 0.004
0.103 0.004
0.068
0.068
0.064 0.006
0.077 0.002
0.075 0.002
0.074 0.003
0.085 0.002
0.086 0.002
Treated Liquid
All Other
Organics
0.018
0.111
0.111
0.115
0.119
0.062
0.079
0.074
0.084
0.066
0.071
Total
Solids
0.132
0.212
0.213
0.222
0.220
0.128
0.132
0.130
0.142
0.143
0.148
0.153
0.152
Ash
0.111
0.099
0.098
0.106
0.104
0.054
0.054
0.054
0.076
0.073
0.071
0.088
0.085
Tall
Oil
0.003
0.005
0.005
0.002
0.003
0.005
0.005
0.004
0.001
0.000
0.001
0.002
0.002
All Other
Organics
0.018
0.108
0.110
0.114
0.113
0.069
0.073
0.071
0.065
0.070
0.076
0.061
0.065
Total
Solids
0.133
0.231
0.220
0.246
0.232
0.134
0.141
0.134
0.155
0.162
Foamed Product
Ash
0.110
0.110
0.103
0.107
0.103
0.056
0.058
0.059
0.084
0.087
Tall
Oil
0.005
0.005
0.006
0.004
0.005
0.006
0.006
0.006
0.004
0.004
All Other
Organics
0.018
0.116
0.111
0.135
0.124
0.072
0.077
0.069
0.067
0.071
-------
APPENDIX V
TALL OIL
Sample
Feed Liquid
Foamate After Sparging
for:
5 minutes
10
22
25
30
35
50
60
64
65
80
100
102
ANALYTICAL DATA
TABLE 2
ANALYSES IN BATCH EXPERIMENTS
Weight Per Cent Tall Oil in Experiments
I II III IV V VI
0.005 0.006 0.005 0.005 0.004 0.004
0.020 0.007 0.011 0.009
0.005
0.006
0.005 0.005
0.007 0.007
0.016
0.004
0.004
0.008
0.004 0.005
0.004
0.004
0.005
Remaining Column Liquid 0.001 0.003 0.002 0.000 0.002
Conditions: Initial 8 ft. liquid column and 4 ft. foam height, 0.26 SCFM
Air Rate; room temperature.
-80-
-------
APPENDIX V
(Continued)
TABLE 3
ANALYSES OF STREAMS TAKEN FROM KRANNERT DIVISION WASTE TREATMENT PLANT
Constituents as Weight Per Cent of Sample
00
I
Sample
Primary Clari-
fier Effluent
Pulp Mill
Effluent
Paper Mill
Effluent Plus
Decker Fil-
trate
Dense Foam
From Primary
Clarifier
Dense Foam
From Secondary
Clarifier
Total Solids
Ash
Tall Oil
0.125 Avg. Value
from 78 measurements
with range 0.100 to
0.180
0.160
0.163
0.162
0.158
0.150
0.161
0.64
2.32
0.078 Avg. Value
from 9 measurements
with range 0.045 to
0.108
0.068
0.068
0.064
0.077
0.075
0.074
0.004 Avg. Value
from 9 measurements
with range 0.002 to
0.005
0.006
0.002
0.001
0.003
0.54
0.186
0.158
All Organics Other
Than Tall Oil
0.000 Avg. Value
from 0.084 measure-
ments with range
0.066 to 0.119
0.062
0.079
0.074
0.084
1.62
-------
BIBLIOGRAPHIC:
Georgia Kraft Company, Foam Separation of Kraft Pulping
Wastes, Final Report FWPCA Grant No. WPRD 117-01-68,
October, 1969
ABSTRACT
Laboratory studies of foam separation were conducted to
determine the feasibility of this process for reducing B.O.D.,
solids content, and foaming tendency of clarified kraft mill
effluent. Since kraft pulping wastes have a natural tendency
to foam, it was expected that the foaming process, which has
been found to be useful in treating domestic wastes, might
have applications in treatment of these effluents.
Both continuous flow and batch experiments were conduct-
ed, and liquid and foam heights, liquid feed rates, air sparging
rates, and temperature were varied over wide ranges.
The B.O.D. reduction in the treated liquid was disappoint-
ingly small, averaging less than 5 per cent, and the B.O.D.
enrichment in the foam phase was in most cases less than 15
times that of the feed. Solids removal was correspondingly
low.
Foaming tendency, however, was significantly reduced by
the intentional foaming process with reductions of 40 to 60
per cent in this variable being obtained. The reduction In
ACCESSION NO.
KEY WORDS:
Foam Separation
Pulping Wastes
B.O.D. Removal
Tall Oil Removal
Foaming Reduction
Solids Removal
Treatment Costs
BIBLIOGRAPHIC:
Georgia Kraft Company, Foam Separation of Kraft Pulping
Wastes. Final Report FWPCA Gram No. WPRD 117-01-68,
October, 1969
ABSTRACT
Laboratory studies of foam separation were conducted to
determine the feasibility of this process for reducing B.O.D.,
solids content, and foaming tendency of clarified kraft mill
effluent. Since kraft pulping wastes have a natural tendency
to foam. It was expected that ihe foaming process, which has
been found to be useful in treating domestic wastes, might
have applications in treatment of these effluents.
Both continuous flow and batch experiments were conduct-
ed, and liquid and foam heights, liquid feed rates, air sparging
rates, and temperature were varied over wide ranges.
The B.O.D. reduction in the treated liquid was disappoint-
ingly small, averaging less than 5 per cent, and the B.O.D.
enrichment in the foam phase was in most cases less than 1.5
times that of the feed. Solids removal was correspondingly
low.
Foaming tendency, however, was significantly reduced by
the intentional foaming process with reductions of 40 to 80
per cent In this variable being obtained. The reduction In
ACCESSION NO.
KEYWORDS:
Foam Separation
Pulping Wastes
B.O.D. Removal
Tall Oil Removal
Foaming Reduction
Solids Removal
Treatment Costs
BIBLIOGRAPHIC:
Georgia Kraft Company, Foam Separation of Kraft Pulping
Wastes, Final Report FWPCA Grant No. WPRD 117-01-68,
October, 1969.
ABSTRACT
Laboratory studies of foam separation were conducted to
determine the feasibility of this process for reducing B.O.D.,
solids content, and foaming tendency of clarified kraft mill
effluent. Since kraft pulping wastes have a natural tendency
to foam, it was expected that the foaming process, which has
been found to be useful in treating domestic wastes, might
have applications in treatment of these effluents.
Both continuous flow and batch experiments were conduct-
ed, and liquid and foam heights, Uquld feed rates, air sparging
rates, and temperature were varied over wide ranges.
The B.O.D. reduction in the treated liquid was disappoint-
ingly small, averaging less than 5 per cent, and the B.O.D.
enrichment in the foam phase was in most cases less than 1.5
times that of the feed. Solids removal was correspondingly
low.
Foaming tendency, however, was significantly reduced by
the Intentional foaming process with reductions of 40 to 60
per cent in this variable being obtained. The reduction in
ACCESSION NO.
KEY WORDS:
Foam Separation
Pulping Wastes
B.O.D. Removal
Tall Oil Removal
Foaming Reduction
Solids Removal
Treatment Costs
-------
foaming tendency was a strong function of gas-to-llquid ratio
with the most effective operating range being between 1.0
and 1.5 SCFM/gallon.
The experimental results suggest that the reductions in
B.O.D. and foaming tendency were related to the separation of
the tall oil components of the waste. These components were
concentrated in the foam fraction, but they accounted for a
maximum of only 10 to 12 per cent of the B.O.D. of the raw
feed. Apparently the remaining B.O.D.- producing materials
were not surface active and did not attach themselves to the
surface-active components.
The cost of using a foam process on kraft mill wastes is
estimated to be four to five cents per 1000 gallons of feed;
this cost is exclusive of further processing of the concentrated
foamate. Based on control of foaming tendency alone, the
process would be unattractive from a cost standpoint.
This report was submitted in fulfillment of Grant No
WPRD 117-01-68 between the Federal Waste Pollution Control
Administration and Georgia Kraft Company.
foaming tendency was a strong function of gas-to-llquid ratio
with the most effective operating range being between 1.0
and 1,5 SCFM/gallon.
The experimental results suggest that the reductions In
B.O.D. and foaming tendency were related to the separation of
the tall oil components of the waste. These components were
concentrated in the foam fraction, but they accounted for a
maximum of only 10 to 12 per cent of the B.O.D. of the raw
feed. Apparently the remaining B.O.D.- producing materials
were not surface active and did not attach themselves to the
surface-active components.
The cost of using a foam process on kraft mill wastes is
estimated to be four to five cents per 1000 gallons of feed;
this cost is exclusive of further processing of the concentrated
foamate. Based on control of foaming tendency alone, the
process would be unattractive from a cost standpoint.
This report was submitted in fulfillment of Grant No.
WPRD 117-01-68 between the Federal Waste Pollution Control
Administration and Georgia Kraft Company.
foaming tendency was a strong function of gas-to-liquld ratio
with the most effective operating range being between 1.0
and 1.5 SCFM/gallon.
The experimental results suggest that the reductions In
B.O.D. and foaming tendency were related to the separation of
the tall oil components of the waste. These components were
concentrated In the foam fraction, but they accounted for a
maximum of only 10 to 12 per cent of the B.O.D. of the raw
feed. Apparently the remaining B.O.D.- producing materials
were not surface active and did not attach themselves to the
surface-active components.
The cost of using a foam process on kraft mill wastes Is
estimated to be four to five cents per 1000 gallons of feed;
this cost is exclusive of further processing of the concentrated
foamate. Based on control of foaming tendency alone, the
process would be unattractive from a cost standpoint.
This report was submitted in fulfillment of Grant No.
WPRD 117-01-68 between the Federal Waste Pollution Control
Administration and Georgia Kraft Company.
------- |