WATER POLLUTION CONTROL RESEARCH SERIES • 12040 EEK 08/71
Treatment of Selected Internal
Kraft Mill Wastes
in a Cooling Tower
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
VATER POLLUTION CONTROL RESEARCH: SERIES
The Water Pollution Control Research. Series describe* the
results and.progress in the control and abatement of pollu-
tion of our Nation's waters. They provide a central source
of information on the research* development, and demon*
•tration activities of the Environmental Protection Agency
through inhouse research and grants and contracts with
Federal, State, and local agencies, research institutions,
and industrial organizations.
Inquiries pertaining to the Water Pollution Control Research
Reports should be directed to the Head, Publications Branch,
Research Information Division, R&M, Environmental Protection
Agency, Washington, D.C. 20460.
-------
TREATMENT OF SELECTED INTERNAL KRAFT MILL WASTES
IN A COOLING TOWER
BY
Georgia Kraft Company
Research and Development Center
Krannert Road
Rome, Georgia 30161
for the
ENVIRONMENTAL PROTECTION AGENCY
Program #12040EEK
Grant #WPRD 116-01-68
August, 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication. Approval
does not signify that the contents necessarily reflect
the views and policies of the Environmental Protection
Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendations for
use.
ii
-------
ABSTRACT
Pulp mill condensates, decker filtrate, and turpentine decanter under-
flow from an 850 ton/day kraft linerboard mill have been successfully
treated in a conventional cooling tower. These waste streams, in com-
bination with the condenser waters from a barometric type evaporator
condenser, are cooled in the tower and reused. The overall accomplish-
ments of this process are the removal of about 10,000 Ibs of BOD per
day and the reduction in overall mill water needs of about 8-10 MGD.
Theoretical, laboratory, and pilot studies investigated the BOD removal
mechanisms involved and proved that the predominant mechanism is strip-
ping of volatile components. As a part of the laboratory studies a
simple procedure called a static vapor-liquid equilibrium method was
developed for collecting and analyzing low concentration volatile com-
ponents in waste water. Mathematical relationships were developed
which allow the translation of the findings of this study to other
waste water treatment applications. The primary factors controlling
BOD removal in this system are blowdown rate, liquid-gas ratio, and
average temperature. For a blowdown rate of 15-20 per cent of the
tower influent, average treatment efficiencies for the waste streams
considered are 55-65 per cent for sixth effect condensate, 45-55 per
cent for combined condensate and turpentine decanter underflow, and
25-35 per cent for decker filtrate.
The reduction in BOD of these waste streams is believed due primarily
to the stripping of methanol. Some biological activity is evident in
the tower, however, and the addition of nutrients results in an im-
provement of 5-10 per cent in BOD removal. The system has several
advantages over the conventional surface condenser system used with
kraft mill evaporators. Both operating and capital costs compare
favorably with other waste-treatment methods.
This report was submitted in fulfillment of Project Number 12040EEK,
Grant WPRD 116-01-68, under the (partial) sponsorship of the Office of
Research and Monitoring, Environmental Protection Agency.
Key Words: Sulfate pulping, pulp mills, water pollution, waste
treatment, water reclamation, cooling towers, evapora-
tors, condensers (liquefiers), biochemical oxygen
demand, wastes, effluents, condensates, filtrates,
methanol, stripping.
-------
CONTENTS
Section Page No.
I. CONCLUSIONS 1
II. RECOMMENDATIONS 3
III. INTRODUCTION 5
Background Developments 5
Proposal to Study the Treatment of Kraft Mill
Wastes in Cooling Towers 7
Description of Proposed Cooling Tower System . . 8
IV. ANALYSIS OF WASTE WATER STREAMS 11
Standard Analysis of Waste Water Streams. ... 11
Analysis of Volatile Materials 11
V. MECHANISM OF BOD REMOVAL 19
Possible Mechanisms 19
Laboratory Sparging Studies 19
Mechanism Investigations in Laboratory Cooling
Towers 22
VI. ANALYSIS OF STRIPPING 29
Theoretical Developments 29
Experimental Developments 33
VII. FULL-SCALE MILL STUDIES 47
Design and Construction of Cooling Tower. ... 47
Collection of Experimental Data 49
Analysis of Data from Full-Scale Studies ... 53
VIII. ACKNOWLEDGEMENTS 63
IX. REFERENCES 65
X. PUBLICATIONS AND PATENTS 67
XI. GLOSSARY 69
XII. APPENDICES 73
-------
FIGURES
No. Page No.
1. Schematic Diagram of Proposed Full-Scale Cooling
Tower System 9
2. Typical Gas Chromatograph Trace for Evaporator
Condensates 16
3. Pilot Cooling Tower for Batch Stripping Experiment 24
4. Comparison of Stripping with Nitrogen and Air 25
5. Pilot Cooling Towers for Comparison of Stripping and
Biological Treatment
6. Comparison of BOD Removal by Stripping and Biological
Mechanisms in Pilot Tower
7. Schematic Diagram for Use in Mathematical Developments 32
8. Stripping Experiment: Basin Volume Versus Methanol
Concentration
9 Effect of Liquid/Air Ratio on Stripping of Methanol
in a Pilot Cooling Tower 39
10. Effect of Liquid Temperature on Stripping of Methanol
in a Pilot Cooling Tower 40
11. Stripping Experiment: Fraction of BOD Remaining
Versus Dimensionless Time, Rome 6th Effect,
July 15, 1968 Sample 43
12. Stripping Experiment: Fraction of BOD Remaining
Versus Dimensionless Time, Rome 6th Effect,
August 1, 1968 44
13. Log-log Relationship Between BOD Content in Cooling
Tower and the Dimensionless Time, *— 45
V0/Qa
14. Photographs of Complete Full-Scale Installation 50
15. Photographs of Full-Scale Installation During
Construction 51
16. Schematic of Cooling Tower-Evaporator Condenser
System 52
vi
-------
TABLES
No. Page No.
1. Chemical Composition of Streams to Be Treated
in Tower 12
2. Summary of Volatile Compounds in Kraft Mill
Evaporator Condensates 18
3. Comparison of Air Sparging and Nitrogen Sparging
to Treat Sixth Effect Condensate 21
4. Effect of Adding Nutrients to Air Sparging of Sixth
Effect Condensate in "Bio-Oxidation" Unit 22
5. Example Data from Non-Steady State Stripping
Experiments 35
6. Summary of Cooling Tower Data 54
7. Stripping Constants and Removal Efficiencies for
Full-Scale Tower 57
8. Effects of Nutrients on BOD Removal 58
9. Average Gaseous Emissions from Cooling Tower (PPM) 60
vi i
-------
SECTION I
CONCLUSIONS
Based on the theoretical and experimental results of this study, it is
concluded that:
1. A cooling tower can function very effectively in providing
reusable water for a barometric condenser used in connection
with the evaporation cycle on a kraft pulping operation.
2. Through the use of the barometric condenser-cooling tower system
a very significant reduction in water requirements for a kraft
pulp mill can be accomplished (8-10 MGD for an 850 ton unbleached
linerboard mill).
3. A cooling tower can achieve a 25-30 per cent reduction in the BOD
discharged from a kraft mill.
4. The primary mechanism of BOD removal is physical stripping of
volatile components.
5. Only minor improvement in BOD removal can be accomplished by
adding supplemental nutrients to the tower feed.
6. Sixth effect condensate, combined condensate, turpentine decanter
underflow, and decker filtrate can be treated in a cooling tower
for BOD removal. The degree of removal will be directly related
to the BOD of the volatile compounds present in the stream.
7. The cost of a cooling tower installation is essentially the same
as the cost of converting from barometric condensers to surface
condensers. However, when the reduced cost of subsequent waste
water treatment is credited to the cooling tower, it is by far
the most economical system.
8. As expected, there are emissions of reduced sulfur compounds from
the tower; however, these are in the 1 to 2 ppm range and are not
expected to cause any significant air pollution problem.
9. There have been no significant operating problems encountered in
the operation of the cooling tower-condenser system. Minor
foaming in the tower basin is easily controllable with minor
additions of defearners.
10. Preliminary studies of biological treatment of the blowdown from
the cooling tower indicated that such treatment processes are not
adversely affected by the concentrated waste waters.
-1-
-------
SECTION II
RECOMMENDATIONS
1. The cooling tower-barometric condenser is performing a successful
water pollution control demonstration. Final proof of the system's
value and its comparison with other waste water treatment tech-
niques rests in its ability to perform without significant mainte-
nance or outages over a longer time period. Therefore, the system
should be maintained in good operating condition and with normal
data collection over an extended period. This will allow continued
performance of a desired waste water treatment demonstration.
2. A more detailed study of the impact of this unit on air quality
would provide specific answers to questions raised about its impact
on odors in the mill vicinity. Additional facilities such as pre-
stripping and burning devices could provide further minimization
of atmospheric emissions from the system.
3. Future investigation of the optimization of this system with regard
to BOD removal would allow its maximum development as a waste water
treatment device.
4. Further research into relative volatilization rates of various
compounds in water solutions would allow application of this
process in other waste water treatment situations.
5. Better knowledge of the biological organisms and their life pro-
cesses in the tower would more clearly define the role that these
life forms have in the operation and maintenance function of the
tower-condenser system.
-3-
-------
SECTION III
INTRODUCTION
Background Developments
The pulp and paper industry is vitally involved with the mounting
problems in water pollution, since water performs a key role in the
pulp and papermaking processes. Current processes require an average
of approximately one-hundred tons of water to produce one ton of baft
linerboard and up to 400-500 tons of water per ton of paper is re-
quired for more refined and bleached grades. With an average of over
one-quarter ton of paper being consumed annually by each person in the
United States, the water requirement to meet this demand places the
paper industry among the foremost users of the nation's water re-
sources. The need, therefore, is urgent to seek methods for reducing
water usage and improving waste water treatment. Georgia Kraft Com-
pany recognizes this need and constantly endeavors to advance the
position of the industry, as well as its own position, with regard to
environmental problems. The use of cooling towers to conserve water
in kraft pulping came to the attention of Company personnel in previous
studies as having extensive and undeveloped potential in water conser-
vation and waste water treatment. This investigation was conceived to
pursue that potential.
Several years ago it was discovered that cooling towers potentially
might be used for reducing the water usage in pulp production. Cohn
and Tonn (l) uniquely applied a novel cooling tower system to the
multi-effect evaporators for one kraft pulp mill and found that by
cooling and reusing barometric condenser water, fresh water consumption
for the condensers could possibly be reduced as much as 95 per cent.
While not all mills could expect to attain the indicated reduction in
water usage, many mills have a serious waste treatment problem due to
the common practice of using barometric condensers in their black
liquor evaporator system. This practice leads to large volumes of
water being mixed with condensed evaporator vapors, thus giving rise
to a high-volume, low-BOD-concentration (a reference to BOD means
five-day, 20°C test) effluent. Since it is impractical to process
this large volume of dilute effluent in a waste treatment plant, many
mills which treat their effluent have replaced the barometric con-
densers with the more expensive and less efficient surface-type
condenser. While preventing contamination of the cooling water, the
surface condenser still requires large volumes of cooling water and
transmits a significant quantity of heat to the receiving stream. This
heat in some situations may be an important source of pollution.
Personnel of Georgia Kraft Company demonstrated in a pilot plant study
(2) at Macon, Georgia, that aeration in the cooling tower provides BOD
reduction as well as cooling, and the operation of the cooling tower
-5-
-------
could be interconnected with other internal mill streams to give still
greater reductions in water usage and in mill effluent BOD. The study
showed that in the Macon mill the evaporator condensate and the decker
filtrate accounted for about 40 per cent and 15 per cent of the total
BOD of the mill effluent, respectively. It was estimated that if these
streams could be used as makeup for the water evaporated in the cooling
tower, the need for using fresh cooling water from the river would be
greatly reduced. The pilot study indicated that a 45 per cent reduc-
tion of the mill's total water requirements might be expected if such a
process were installed. In addition, the cooling tower treatment
process would produce some reduction in the thermal load on the river.
Although the results of the pilot studies were very encouraging, several
important questions still needed answering before the technical and
economic feasibility of the process could be established. The more im-
portant of these were:
1. Is the BOD reduction effected by the process due to stripping of
organics by the air stream or to biological action occurring in the
tower?
2. Will organic vapors emitted from the tower contribute to an air
pollution problem?
3 How will the efficiencies of cooling and BOD reduction be affected
by the concentration of solids in the recycled stream?
4. What are the operational problems involved in a full-scale system?
5. How reliable would this process be?
6. How sensitive would the efficiency of the cooling tower be to pro-
cess upsets?
7. How do the economics of the process compare with other alternates?
Further work was clearly needed before the novel cooling tower operation
could become commercially dependable.
After it was found that cooling towers had potential for BOD treatment,
a literature review was made to determine if cooling towers had been
used for this purpose before. There was very limited published infor-
mation, but two industrial companies were achieving organic waste
treatment in cooling towers. The Sun Oil Company (3) had constructed
cooling towers to function simultaneously as water cooling and treatment
devices, and the Celanese Corporation (4) had discovered that cooling
towers were effective in removing acetone from their waste waters.
If proven feasible for use in the pulping industry, this treatment
method could be envisioned as having extensive application in indus-
-6-
-------
trial waste treatment and water conservation. With over one hundred
kraft mills in the United States using multiple-effect evaporators and
the widespread use of evaporators in the other industries, the cooling
tower treatment process could represent a major contribution to the
abatement of organic and thermal pollution of the nation's water
resources.
Proposal to Study the Treatment of Kraft Mill Wastes in Cooling Towers
The proven potential for reducing BOD and water usage and the prospec-
tive wide application of cooling towers in waste water treatment made
the continued development and demonstration of the technique highly
desirable. To carry out further work a request for a demonstration
grant from the Office of Research and Monitoring, Environmental Pro-
tection Agency was made. The proposed project was intended to evaluate
fully the technical and economic feasibility of using cooling towers to
reduce the BOD and heat content of certain selected kraft mill effluents
and to reduce the water usage of kraft mills. The specific objectives
were:
1. To determine the efficiencies of BOD reduction and cooling as
functions of cooling tower operating variables (liquid and air
flow rates, temperatures, and composition and concentration of
feed and recycled stream components).
2. To determine if organic vapors emitted from the tower would cause
air pollution problems.
3. To determine which internal process waste waters can be effectively
treated by this process.
4. To determine the mechanism of BOD reduction in the tower, i.e., how
much reduction is due to stripping of organics by the air stream
and how much is due to biological and chemical action.
5. To determine how sensitive the cooling and BOD reduction efficien-
cies will be to shock loadings and other process upsets.
6. To determine how the economics of this treatment process compare
with those of alternate treatment processes.
7. To determine what operational problems were involved in continuous
operation of a cooling tower on kraft mill wastes.
8. To collect engineering data which could be used for future design
purposes.
9. To determine how this treatment process could be integrated with
mill operations and subsequent treatment steps.
-7-
-------
Description of Proposed Cooling Tower System
The Office of Research and Monitoring, Environmental Protection Agency,
approved the project as a demonstration grant, and Georgia Kraft Com-
pany undertook an extensive study that would result in the construction
and evaluation of a full-scale cooling tower at its Macon, Georgia,
pulp mill. The proposed system, as it would be included in the mill
operation, is pictured in Figure 1. As shown by the figure, the en-
visioned installation takes barometric condenser water from the hot
wells, cools it, and recirculates the cooled water to the direct con-
tact condenser and the ejector for the non-condensable gases. Baro-
metric legs below these units produce the vacuum for the multiple
effect evaporators. Makeup water to the tower compensates for blowdown,
spray losses, and evaporation. Waste streams in the pulp mill from
which makeup water could be derived would be combined condensate from
the evaporators, decker filtrate (wash water from final washing of pulp)
and turpentine decanter underflow. Sixth effect condensate furnishes
some of the makeup volume but is not a separable stream as are the other
wastes mentioned. Since the barometric condenser consumes large volumes
of water, the use of the cooling tower would reduce greatly the intake
of fresh water from the river. Further, since several highly contami-
nated streams could be treated and used as makeup water for the cooling
tower, a very significant reduction in BOD discharged from the pulp mill
was expected. The proposed study was envisioned to evaluate which of
these available waste streams should be used, how much BOD removal could
be achieved, how much reduction in water usage would be obtained, and
how practical and economical the process would be on a comnercial scale.
In addition to the ultimate demonstration of the practicality of the
process, basic scientific data about the process would be sought.
-8-
-------
CHIPS 1
WASHERS
SCREENS
DECKER
KNOTTER
IS)
Q LJ
MULTIPLE EFFECT
EVAPORATORS
th
T
I
CONDENSER
EFFECT
rORS
I I
1 » 1
COMBINED CONDENSATE | F
1
TURPENTINE DECANTER UNDERFLOW DECKER FILTRATE
FIGURE 1: SCHEMATIC DIAGRAM OF PROPOSED FULL-SCALE COOLING TOWER SYSTEM
-------
SECTION IV
ANALYSIS OF WASTE WATER STREAMS
Standard Analysis of Waste Water Streams
To begin a fundamental study of waste water treatment in cooling
towers, a thorough chemical and biochemical analysis was required.
These data provided a basis for characterizing the waste waters, for
evaluating treatment effectiveness, and for analyzing the treatment
mechanisms involved. The characteristics usually employed for describ-
ing waste water streams include BOD, COD, solids, pH, conductivity,
alkalinity, and, in this particular type of waste, sulfide content.
Table 1 gives a typical description of the waste water streams investi-
gated in this study and the range of variation in the characterizing
measurements. The streams of interest were sixth effect condensate,
combined condensate, decker filtrate, and turpentine decanter
underflow.
The analyses given in Table 1 reveal only gross features of the nature
of the waste streams. The BOD test results reveal the existence in
the waste water of significant concentrations of biodegradable organic
compounds. As is typical of most organic bearing waste, particularly
pulp mill wastes, the COD analysis is greater than the BOD, indicating
that all the oxidizable constituents were not biodegradable. Other
features are that all of the streams are alkaline, essentially all of
the solids in the stream are dissolved, and a sizable fraction of the
dissolved solids are volatile, implying again the presence of organic
materials. It should be noted that the very volatile organic con-
stituents would escape detection in the dissolved solids test due to
the nature of the test. Sulfide, though present, constitutes a very
small fraction of the oxygen demand.
The results of Table 1 were derived from standard analytical tests.
These tests were adopted for the duration of the study and with one
exception, which is discussed in the following section, the procedures
were taken from a text on standard methods for analyzing water and
waste water streams (5). The exception was the sulfide ion content,
and this was determined by a modified TAPPI procedure (T-625ts-64)
using an Orion sulfide electrode.
Analysis of Volatile Materials
Development of Analytical Method - While standard analytical data
furnished a generalized characterization of the waste streams, a more
comprehensive examination was needed to determine the composition of
the volatile materials and to estimate their involvement in the air
stripping processes present in cooling tower operation. From previous
-11-
-------
TABLE 1
I
ro
BOD - mg/1
COD - mg/1
pH
Solids:
Total , mg/1
Dissolved, mg/1
Total Volatile, mg/1
Dissolved Volatil
Conductivity, umhos
Total Alkalinity, mg/1 as
mg/1 as
Sulfide, mg/1
CHEMICAL COMPOSITION OF STREAMS TO BE TREATED IN TOWER
6th Effect
Condensate^')
Avg.
1360
4225
9.7
2850
2820
ig/1 1133
e, mg/1 1188
1600
ig/1 as
498
alinity,
162
43
Max.
1600
5222
10.7
4036
3986
1348
1320
2600
862
378
_
Min.
810
3039
8.7
1492
1540
736
676
500
114
14
—
Combined .
Condensate^2'
Avg.
722
982
7.4
388
280
224
186
260
39
0
2
Max.
833
1192
7.4
480
384
244
196
270
52
0
_
Min.
610
771
7.3
276
176
204
176
250
26
0
*
Decker Filtrate^2)
Avg.
900
1823
9.0
1808
1632
1194
962
1150
450
54
16
Max.
1166
2068
9.9
2404
2104
1560
1268
1500
534
108
_
Min.
633
1577
8.2
1212
1160
828
656
800
366
0
_
Turpentine
Decanter
Underflow^3/
Average
8180^
7100
9.9
260
220
39
21
480
500
430
520
(1) Five sets of data from Rome plant.
(2) Two sets of data from Macon plant.
(3) One set of data from Rome plant.
(4) Seven sets of data from Macon plant.
-------
studies on the Macon pilot cooling tower, it was suspected that air-
stripping was one of the major treatment mechanisms; and it became
necessary to identify the strippable components. The qualitative
identification, quantitative estimation, and the determination of the
oxygen-demanding (BOD) characteristics of each of these volatile com-
ponents was needed to establish the basis for a fundamental evaluation
of the overall potential of the proposed processes. An analytical
method was not directly available to achieve the required measurements;
initial work, therefore, had to be devoted to the development of a
suitable method of analysis.
Studies in the air pollution control field (6, 73 8) provided some in-
sight into analytical procedures for these volatile emissions from
kraft mills. These studies were helpful in the analysis of the vola-
tile components but did not resolve the problem of determining the
concentration of such components in a waste water stream. In connec-
tion with a SEKOR (Stripping Effluents for Kraft Mill Odor Reduction)
study, Hruitfiord and McCarthy (9) identified methyl mercaptan, dimethyl
sulfide, acetone, methanol, ethanol, and methyl isobutyl ketone, as well
as the more common hydrogen sulfide and carbon dioxide gases, in
digester blow gas condensate. Bethge and Ehrenborg (10) working in
Sweden confirmed this analysis of blow gas condensate; and Ruus (21),
also a Swedish researcher, reported quantitative data on the concen-
tration of hydrogen sulfide, methyl mercaptan, dimethyl sulfide,
methanol, ethanol, and acetone found in various effluent and condensate
streams from kraft pulp mills in his country.
For the purpose of this study several methods of analysis had to be
evaluated to develop a suitable analytical procedure. Since the gas
concentrations were very low, the gas chromatograph was particularly
suited for a test method and was used in all cases. A Perkin-Elmer
Model 881 gas chromatograph, equipped with a flame ionization detector,
a recorder, and an Infotronics CRS-104 Integrator served as the analyt-
ical instruments. Separations were made on a 6 ft, 1/8 in. column of
15 per cent Carbowax 20M on 100-120 mesh Chromosorb W operated at 70°C
with a helium flow of 30 cc/minute.
The search for an analytical method began with an attempt to analyze
cooling tower off gases directly. This proved unsuccessful because
the high degree of dilution brought about by the large volumes of air
necessary for operation of the cooling tower produced concentrations
below the detectable limits of the gas chromatograph.
In the next method considered, stripping action of the cooling tower
was simulated in a bench-scale apparatus. With this unit, a gas (air
or nitrogen) was sparged slowly into a fixed volume of condensate
which was maintained at constant temperature. The off gases were con-
ducted directly to the gas sampling loop of the gas chromatograph.
This method was quite satisfactory for qualitative work; however,
concentrations of some components in the discharged gases changed
-13-
-------
very rapidly upon initial stripping, and it was impossible to relate
vapor concentrations to the concentrations in the liquid phase.
To circumvent the problem of the changing vapor concentrations, an at-
tempt was made to condense and to collect all the stripped vapors by
freezing in a dry ice-acetone mixture. It was intended that the
collected materials would be subjected to further analysis by gas
chromatography. But this procedure was also regarded as unsatisfac-
tory because the cold trap was not sufficiently chilled with dry ice
and acetone to collect all of the stripped organic vapors. Perhaps
this method could be improved by the use of liquid nitrogen in the cold
trap.
Another possible means of collecting off gases was to absorb them on
activated carbon and then extract with a solvent. The carbon readily
absorbed all of the vapors; however, the solvents generally used, e.g.,
chloroform, ethyl ether, and petroleum ether, interferred with analysis
on the gas chromatograph. The relatively large amount of solvent
present tended to hide the characteristic peaks of most of the com-
pounds of interest on the gas chromatogram. The change to a high boiI-
ing point solvent may have solved the interference problem; Jut it would
have increased the time for gas chromatographic «njlys" ^"J™?1 as
two to three times or would have been retained on the column to be
eluded slowly giving poor baseline stability. Further consideration
mght have been given to the use of carbon disulfide as an extraction
medium as this solvent reportedly does not show up on a flame loniza-
tion detector.
After these several attempts failed to provide completely satisfactory
results, a scheme—referred to as the static vapor-liquid equilibrium
method—was developed. The liquid sample was placed in a 500 ml flask,
the flask was sealed with a serum cap, and then the flask with contents
was placed in a constant temperature bath at 55°C. After sufficient
time was allowed for vapor-liquid equilibrium to be established,
usually 30 to 60 minutes being required, a vapor sample was removed by
inserting a syringe needle through the cap, and the vapor sample was
injected into the chromatograph. The method gave very reproducible
results.
A second method which was found satisfactory for quantitative analysis
of the more concentrated components involved direct injection of a
known volume of the liquid condensate sample into the gas chromatograph.
Concentrations of the gaseous components could be determined by compar-
ing the measured peak used on the chromatogram to a previously deter-
mined calibration curve.
Compound Identifications - Once a method was available which would give
reproducible results and could be relied upon to show small changes in
the composition of condensates, the problem of identifying and quantify-
-14-
-------
ing the gas chromatographic peaks required solving. Figure 2 repre-
sents a typical chromatogram.
To identify the peaks on the chromatograms of condensate samples, a
series of dilute water solutions of pure compounds were prepared.
Chromatograms of these pure compounds were made using the static
vapor-liquid equilibrium method. The retention times of the known
compounds were then compared with those of the unknown peaks of the
condensate sample. In the earlier work in which stripped vapors
were condensed in a cold trap, it was possible to obtain enough con-
centrated condensed vapors for infrared analysis, and this analysis
strongly suggested that acetone and methyl alcohol were present. Thus,
retention times were determined for these compounds along with many
others suggested by the literature and some which were used simply to
satisfy scientific curiosity. The known compounds tested, in addition
to methyl alcohol and acetone, were ethyl alcohol, methyl mercaptan,
methyl sulfide, dimethyl sulfide, formaldehyde, acetic acid, ethyl
sulfide, a-pinene, 3-pinene, isopropyl alcohol, acetaldehyde, ethyl
mercaptan, and ethyl ether. A study of the retention times of these
compounds indicated that ethyl alcohol, formaldehyde, acetic acid,
ethyl sulfide, e-pinene, isopropyl alcohol, formaldehyde, ethyl mer-
captan, and ethyl ether could be eliminated as the major components
of the evaporator condensate samples.
A further step was taken to tie the analysis down more closely. A
sample of the condensate was prepared and subjected to the static vapor-
liquid equilibrium method for determining the gas chromatogram. Then a
small amount of the suspected pure compound was added to the sample,
and after establishment of the new equilibrium, another chromatogram
was obtained. Addition of a pure compound amplified one of the peaks
without any evidence of peak splitting. This method was used to con-
firm six of the typical peaks as methyl mercaptan, methyl sulfide,
acetone, methanol, a-pinene, and methyl disulfide. The others were
eliminated as possible components when it was determined that their
peaks were not exactly coincident with those of the unknown sample.
It was recognized that the identification procedure as outlined did not
constitute complete proof of identity; it was, however, regarded as
very substantial evidence, particularly since previous investigators
had found the named compounds in their work which was similar. Further
proof of these assignments would have required the collection of suf-
ficient amounts of the separated components using a preparative gas
chromatograph followed by infrared or possibly NMR analysis.
For quantitative analyses, the static vapor-liquid equilibrium method
was used to prepare calibration curves for the six identified com-
pounds. With the pure compounds, solutions of known strength were pre-
pared and curves of concentrations versus the peak areas as determined
by the integrator were plotted. The pH of the standard solution had
to be adjusted to approximately the same value as the samples to be
-15-
-------
A.
B.
C.
D.
E.
F.
G.
H.
UNKNOWN
METHYL MERCAPTAN
METHYL SULFIDE
ACETONE
METHYL ALCOHOL
o-P'INENE
METHYL DISULFIDE
UNKNOWN
TIME
0
FIGURE 2: TYPICAL GAS CHROMATOGRAPH TRACE FOR EVAPORATOR CONDENSATES
-16-
-------
analyzed. The effect of pH was dependent upon the substance; for
example, if pH were not adjusted, an error as low as 1 per cent/pH
unit variation would be introduced for acetone and as high as 30 per
cent/pH unit variation for methyl mercaptan. The samples in this
study were constantly within 1/2 pH unit and no adjustment was made.
With the described analytical method, a number of samples of evapora-
tor condensates from each of the Georgia Kraft Company mills were
analyzed. The results are given in Table 2. Methanol appears to be
the major organic component of evaporator condensates with only minor
amounts of other compounds present.
-17-
-------
TABLE 2
SUMMARY OF VOLATILE COMPOUNDS IN KRAFT MILL EVAPORATOR COMPENSATES
Methyl Methyl
Mercaptan Methyl Sulfide Acetone Methyl Alcohol a-Pinene Disulfide
. , No- _ (JEM) __ (PPM) (ppm) __ (ppm) __ (pan) __ (ppm)
Sample - Sarngje^ Avg_^ Max._ Mln^ Ay^._ Max^ Mir^ flyg,, Max. Min. AygTMax. Min. fog. Max. Min7 Avg. Max.
6th Effect Condensate
Krannert 4 T T T .2 .2 .1 3 4 2.5 1100 1350 950 1.4 2.0 T 6.5 13.9 2.6
Combined Condensate
Krannert 10 T T T .5 .6 .1 2.0 2.2 T 400 500 310 .2 .4 T 5.2 6.5 2.5
i
oo 6th Effect Condensate
Macon 4 1-8 6.6 T T T T 1.8 3.0 1.0 950 1300 350 T T T .7 2.81.2
Combined Condensate
Macon 20 3.0 10.7 T T .1 T 1.3 4.5 .1 245 490 100 T T T .1 .81
Combined Condensate
Mahrt 2 6.211.0 T T .2 T 4.0 7.0 1.6 670 710 620 T T T 5.2 6.01.0
Turpentine Decanter
Underflow - Macon 7 1934 10000 23 224 900 3 150 210 0 4386 6500 2575 148 363 18 284 490 155
Cooling Tower Feed 10 1.0 3.6 T .1 .2 T .5 .9 T 120 190 50 T T T .8 2.0 T
Cooling Tower
B1owdown 10 -7 3.1 T .1 .3 T .3 .6 T 80 125 33 T T T .2 .4 T
T = Trace
-------
SECTION V
MECHANISM OF BOD REMOVAL
Possible Mechanisms
The identification and quantitative description of the BOD removal
mechanisms were among the primary objectives in the investigation of
waste water treatment in cooling towers, and considerable effort was
expended on this phase of the project. From the beginning it was ap-
parent that air stripping of volatile compounds was very likely a major
treatment mechanism; but it was important that all mechanisms be
thoroughly investigated. A careful evaluation of all the basic pro-
cesses operating in a cooling tower and a review of past experience
in treating waste waters suggested three possible mechanisms. These
were: air stripping of volatile compounds, biological action, and
chemical oxidation. Experimental results from the Macon pilot cooling
tower and the more recent study of volatile compounds in kraft mill
waste waters suggested that treatment was, to a great extent, due to
the stripping of volatile organic materials from waste water by atmos-
pheric air. Further, from previous experience with trickle filters in
treating kraft mill effluent CZ2J, it was known that waste water,
raining through a high void packing in a manner very similar to water
falling through a cooling tower, is reduced in BOD when bacteria are
caused to grow on the packing surfaces by the addition of suitable
nutrients. Finally, because of the intimate contact of the waste water
with atmospheric air, there was some possibility that chemical oxida-
tion could occur. A series of experiments was designed and conducted
to test for the presence and the relative effects of the three postu-
lated mechanisms.
Laboratory Sparging Studies
Comparison of Chemical Oxidation and Air Stripping - Chemical oxidation
and air stripping could be separated and their relative effectiveness
determined by sparging a small amount of liquid with nitrogen and with
air. Sparging with nitrogen would produce only stripping, while air
sparging would produce both stripping and oxidation. No biological
action would be present if the system were kept sterile and no nutri-
ents were added to support biological growth. Experiments of this type
were conducted in a sparging unit consisting of two four-liter flasks
with appropriate connections being provided for the introduction and
release of the sparging gases. The gases were released into the waste
water through a sintered glass diffuser located near the center and at
the bottom of the flask. The gases in the form of small bubbles rising
rapidly in the vicinity of the diffuser provided a continuous mixing of
the liquid contents. The flasks were immersed in a constant tempera-
ture bath to provide temperature control.
-19-
-------
So that the results of the bench experiment would be indicative of the
cooling tower processes, the gas rates and amount of liquid used were
determined from the operating parameters used in previous studies with
the pilot cooling tower. It resulted that two liters of liquid should
be sparged for 62.6 hours at a gas rate of 400 cc/min. These condi-
tions were estimated to be equivalent to a tower air flow rate of 400
ft/min, a liquid flow rate of 4 gal./min/ft^. The data from these
experiments (shown in Table 3 and Appendix A) revealed that the treat-
ment by chemical oxidation was small. The study showed that stripping
with air resulted in only slightly greater treatment than that obtained
by the stripping action of pure, inert nitrogen alone. Also, it ap-
peared from the measurements that the stripping of organic constituents
was dominantly the only significant BOD removal mechanism.
Comparison of Biological Action and Air Stripping - The sparging experi-
ments, as described for air and nitrogen, were conducted for periods up
to three days and were not suitable for investigating biological action.
Other more lengthy sparging experiments were performed for periods up to
a month in a "Bio-oxidation Unit".^^ The longer period of operation
was required to allow the biological action to establish itself and
become operative. The unit was operated as a completely mixed system
similar to an aerated lagoon. The experimental apparatus consisted
simply of a small (about 9 liters) rectangular chamber through which
waste water fortified with biological nutrients was passed. The unit
was started by seeding with biological growth from the trickling filter
in the Rome mill waste treatment plant. The unit was then fed sixth
effect condensate from the Rome mill with nutrients added. The nutrients
consisted of ammonium nitrate and phosphoric acid added in the ratio of
100:5:1 for BOD, nitrogen, and phosphorus, respectively. The liquid
feed rate was from 1 to 7 mis per minute, and the air rate was adjusted
as low as would maintain some oxygen content in the liquid. The use of
different feed rates gave different liquid retention times in the treat-
ment process. The unit was operated at room temperature.
Visual observation of the activity inside the unit revealed that the
biological organisms immediately attached themselves to all surfaces
exposed to the waste water; however, there seemed to be considerable
suspended solids also. The attaching phenomenon was not surprising
considering the source of the micro-organisms. As time passed with
continued feeding, an increase in suspended solids and a gradual de-
crease in quantity of attached organisms were noted.
(1) A standard laboratory apparatus purchased from BioDevelopment
Associates, P. 0. Box 1752, Austin, Texas 78767, for investi-
gating biological oxidation.
-20-
-------
TABLE 3
I
ro
COMPARISON OF AIR SPARGING AND NITROGEN SPARGING TO TREAT SIXTH EFFECT CONDENSATE
BOD Removal*
Sparging
Experiment Time
(hrs)
1 48.0
2 51.0
3 22.0
4 71.5
5 29.0
6 62.0
Initial
BOD.;
(mg/1)
1600
1380
1400
1366
810
843
After Sparging With
Nitrogen
(mg/1)
1460
-
1166
1033
770
850
Air
(mg/1)
1210
-
966
900
600
800
COD Removal*
Initial
COD
(mg/1)
5080
5222
4019
3704
3039
1830
After Sparging With
Nitrogen
(mg/1)
4039
4769
3541
3136
2115
1675
Air
(mg/D
3794
4731
3618
2744
2404
1618
(*) Experimental Conditions: 135°F, 400 cc/min gas flow, liquid volume 2 liters; detailed data given
in Appendix A.
-------
The bio-oxidation unit was operated for approximately one month. Table
4 summarizes the BOD reduction experienced at various liquid retention
times. These results showed that waste water would sustain and propa-
gate the growth of micro-organisms which in turn remove the dissolved
organic constituents. The high percentages of removals of 96.4 and 98.0
per cent were achieved by a combination of air stripping and biological
oxidation. With the short retention time the BOD removals were only 14.5
and 49.3 per cent. Since the liquid residence time in a cooling tower
operation is short, biological action would appear not to be a very
effective removal mechanism.
TABLE 4
EFFECT OF ADDING NUTRIENTS TO AIR SPARGING
OF SIXTH EFFECT CONDENSATE IN "BIO-OXIDATION" UNIT
Experiment
1
2
3
4
Liquid
Retention
Time
(hrs)
96.0
36.5
16.2
18.2
Feed
BOD
(ppm)
990
641
1214
1000
Effluent
BOD*
(ppm)
19.8
23.3
616
863
BOD
Removal*
(%)
98.0
96.4
49.3
14.5
*For experimental conditions, see text.
Mechanism Investigations in Laboratory Cooling Towers
Comparison of Chemical Oxidation and Air Stripping - While the basic
laboratory sparging experiments had shown that air stripping was the
major reduction mechanism and that biological action and chemical
oxidation were essentially insignificant, it was desirable that these
mechanisms be investigated further under conditions more nearly like
those found in a full-sized cooling tower. For this purpose a labora-
tory cooling tower was designed large enough to retain the principle
features of a typical industrial installation, yet small enough to
allow the processing of the air and liquid streams with conventional
laboratory devices. The tower was constructed as described in
Appendix "B" and had features similar to a conventional counterflow
cooling tower.
-22-
-------
Again, experiments were conducted to evaluate the effects of operating
with air and with nitrogen, and the experimental technique was a batch
operation very similar to the previous laboratory sparging experiments.
A known volume of waste water was added to the basin of the cooling
tower and recirculated through the tower as shown in Figure 3. The
falling of the liquid through the tower packing was a better simula-
tion of the actual tower operation than air sparging. So that a com-
parison could be made of chemical oxidation and air stripping, the
tower was operated first with compressed air and then in a subsequent
experiment with nitrogen. It was necessary to operate the tower in a
forced draft fashion rather than an induced draft because nitrogen was
furnished by commercial compressed gas cylinders and pressure feeding
was the better method of supplying the gas to the tower. Compressed
air was used for the experiment complementary to the nitrogen test so
that the results from both operations would be comparable. Two sets
of experimental data were obtained and treatment was determined both
as BOD removal and as methanol removal. These data plotted in Figure
4 provide dramatic evidence of the low contribution of chemical oxi-
dation. It was concluded that chemical oxidation for all practical
considerations was nonexistent.
Comparison of Biological Action and Air Stripping - To provide proof
that biological action also was a very minor treatment mechanism,
actual evaluations were made in the laboratory cooling tower. In
these experiments, it was necessary to employ two identical cooling
towers (Figure 5) constructed according to specifications in
Appendix "B" to separate biological action from the other mechanisms.
Operation with two identical towers provided for one tower to operate
on air stripping alone and the other, being seeded with bacteria and
fed with nutrients, to operate on air stripping and biological action.
If it could be assumed, as was done in other mechanism studies, that
no synergistic effects were produced, the treatment due to the effect
of biological action could be derived from the difference in per-
formance of the two towers.
In the tower operating with biological growth it was necessary to
operate this system continuously for about one month to prepare a well
developed biological culture within the tower. Experiments were per-
formed when the bacterial growth had covered the tower packing surface
to the extent that portions of growth would slough off and fall into
the basin.
Figure 6 presents a summary of the experimental results, showing
treatment as a function of liquid-to-gas ratio, which was the control-
ling independent variable for the experimental conditions. Slowdown
flow rate was also a significant variable and changed from a low value
of 50 mis per minute at the low liquid-to-gas ratio to a high value of
400 mis per minute at the high liquid-to-gas ratio. The direct cor-
respondence between these two variables made it possible to study the
-23-
-------
Heat
Exchanger
Pump
X
x
Packed
Section
-Air or Nitrogen
Basin
FIGURE 3: PILOT COOLING TOWER FOR BATCH STRIPPING EXPERIMENT
-24-
-------
.012
.010
.008
Air
to
Ol
•o .006
c
3
O
o.
.004
.002 -
.000
10 20
30 40
Time, Minutes
50 60
FIGURE 4: COMPARISON OF STRIPPING WITH NITROGEN AND AIR
-25-
-------
ro
CTi
HEAT EXCHANGERS
FEED
WITH
NUTRIENTS
FEED
SLOWDOWN SLOWDOWN
FIGURE 5: PILOT COOLING TOWERS FOR COMPARISON OF STRIPPING AND BIOLOGICAL TREATMENT
-------
TOO
80 -
60
o
o
CD
40
O Stripping + Biological Action
D Stripping Alone
20
1.0 1.5 2.0
Liquid-to-Gas Ratio, £
2.5
3.0
FIGURE 6: COMPARISON OF BOD REMOVAL BY STRIPPING AND
BIOLOGICAL MECHANISMS IN PILOT TOWER
-27-
-------
degree of treatment as a function of either blowdown or liquid-to-gas
ratio. The relation derived in Figure 5, therefore, distinguished
between stripping and biological action, but did not represent treat-
ment as a pure function of liquid-to-gas ratio. The experimental
points represented averages of all data for a particular liquid-to-gas
ratio. The complete experimental data are given in Appendix "C".
The basic significance of the experimental findings was that biological
action produced from 4 to 12 per cent more BOD removal than did strip-
ping alone. The effect of biological action was dependent on the
liquid-to-gas ratio and appeared to increase with decreasing liquid-to-
gas ratio, as would be theoretically expected. Similarly, with
stripping alone, a decreasing liquid-to-gas ratio increased BOD removal,
and this removal approached the maximum as the liquid-to-gas ratio de-
creased to zero. This maximum was the volatile BOD content of the waste
water system. Biological action in simultaneous operation with strip-
ping could produce greater BOD removal since the biological reactions
could attack the non-volatile portion of the BOD-producing compounds.
While this additional treatment was an advantage, the amount of increased
treatment would not appear to justify the additional operational costs
encountered to achieve it.
-28-
-------
SECTION VI
ANALYSIS OF STRIPPING
Theoretical Developments
Exploratory experimentation found extensive proof that air stripping
was the predominant treatment mechanism and the treatment process was
further elucidated by the successful identification of the volatile
materials involved and the explanation of their apparent removal from
the aqueous phase during passage through the cooling tower. It still
remained, however, to relate the stripping process quantitatively to
the operating variables of the cooling tower. Hence a mathematical
description of the process was sought.
A mathematical' framework was developed by first determining the vari-
ables that define the stripping produced in a cooling tower. The
functional dependence was expressed by the notation:
/?- f(x, B3 G, L, tla twb; (1)
where R = Fraction removal of volatile component in feed water to
the cooling tower
x = Concentration of volatile component in liquid phase
B = Slowdown rate
G = Gas rate
L = Total liquid rate to tower (feed plus recirculation)
t-, = Inlet temperature of water in tower
*wb = Wet bulb temperature of entering air
These variables were believed to be both necessary and sufficient to
describe the cooling tower system, and with them it should be possible
to determine uniquely the treatment due to stripping. The problem
then encompassed finding the specific relationship represented by the
functional notation.
As in all mass transfer processes, the composition of the transferrable
materials in the interacting phases was of prime importance. Fortu-
nately, in this analysis the description of the mass transfer process
-29-
-------
could be considerably simplified due to the low concentration of vola-
tile materials involved. Under these conditions, Henry's Law was valid
and for a gas-liquid equilibrium system the concentration of a gas
phase, volatile component (y) could be described as being directly pro-
portional to the liquid phase concentration (x) . Thus,
2/ = *h* (2)
In the cooling tower, equilibrium was not attained between the contact-
ing gas and liquid phases, but equilibrium was approached to approxi-
mately the same degree at all points along the height of the cooling
tower. With this background, it was envisioned that a small volume of
liquid progressing through the tower would not be in equilibrium with
the surrounding gas phase, but the gas phase composition of the trans-
ferring component at any height, z, in the column would be
This led to the assumption that under constant operating conditions the
amount of volatile material removed by the system (Tx-^) was a constant
fraction of that entering the tower (Lx^-,) , hence
TXj = KLxu (4)
The constant, K, was defined as the stripping constant and proved to be
a very satisfactory parameter for describing the fundamental mass trans-
fer process. By work that is described later in this report and by
other theoretical work (Appendix G), it was shown that K was independent
of the concentration level of the volatile BOD-producing materials in
the waste water being treated. The constant was dependent, however, on
variations of the chemical makeup of the volatile compounds, on the
tower packing material, and also on temperature and the gas-to-liquid
ratio.
The constant, K, as defined, represented the transfer of volatile
materials to the air from a fixed volume of liquid as it passed once
through the tower. The treatment derived from the tower then had to
be a function of the recirculation rate of waste water through the
cooling tower; and recirculation, in turn, was determined by the un-
treated water feed to the tower and blowdown from the tower. Increased
recirculation of waste water through the tower would increase the num-
ber of times a given volume of waste water would pass through the tower
and would, therefore, affect the ultimate treatment accomplished on
-30-
-------
this volume of liquid. A series of material balances on the major
streams depicted in Figure 7 showed that treatment in the tower
could be described with the relationship,
j-)
¥~ (5)
and K could be evaluated from the equation,
K = a - J-; (6)
Up to this point, the theoretical developments were based on the defi-
nition of the stripping constant, K. The heat and mass transfer pro-
cesses, however, could be analyzed by starting with the first
principles of classical theory, such as described by Treybal (IS) or
Bird et a£. (14). The classical differential equations in a packed
tower were
and ^' (y* - y)
which could be transformed with the aid of heat and material balances
over a section of the tower to
(9)
These equations had to be solved simultaneously because y* was very
temperature-dependent for volatile compounds such as methanol, which
constituted a large portion of the volatile BOD-producing materials.
-31-
-------
Exhaust Air
t
G = Lbs dry air/hr
y = Lbs Volatile mat'l./lb. dry air
T = Lbs water vapor/hour; = -\
Comp. of volatiles expressed
as Ibs volatile/lb H?0
X=
Tower Water Feed
L-|,Ibs/hr
X-i ,Comp
Feed Water
F, Ibs/hr
*p, Comp.
Recirculating Water
N, Ibs/hr
a:2 Comp.
Treated
Water
L2=Lbs/hr
XP Comp.
T2 op
+ 4
Packed
Section
•f- Height
-r 3 ft
Basin
NOTES:
1. *Cgmp denotes composition of treatable
materials or treatable BOD.
2. Average tower temperature = ]+t2
-Inlet
Air
= Lbs dry air/hr
= Lbs Volatiles/
hr = 0
d = °F
Slowdown
B = Lbs/hr
*2= Comp.
*= °F
FIGURE 7: SCHEMATIC DIAGRAM FOR USE IN MATHEMATICAL DEVELOPMENTS
-32-
-------
The analysis of a set of data with these equations began with the
determination of HTU# and HTU^ from experimental data. Then, after
HTUft and HTUft had been determined for a given tower configuration, it
was possible to use the fundamental equations to evaluate the effects
of entering liquid temperature, liquid and air flows, and wet bulb
temperature on the removal of volatile components. Due to the com-
plexity of these relationships, it was found advisable to use a
numerical procedure with a digital computer to accomplish the required
computations.
The determination of HTU[\ and HTU^ began with the assumption of a value
for HTUft. Equation (9) was then employed in a numerical procedure to
calculate the temperature profile along the column height. Upon reach-
ing the top of the tower, the calculated water temperature was compared
with the actual, experimental temperature. If there was a discrepancy,
the assumed HTU\\ was corrected and the calculation repeated until agree
ment was obtained. After the temperature profile was established, it
was possible .to construct the profile for y* versus 2. Then by using
Equation (10), HTU^ could be calculated by trial and error in a way
similar to that employed for HTU\\. The test for a current solution was
the agreement of x2/x\ and the calculated
Once the constants EW\\ and HTUM have been established, the heat and
mass transfer characteristics of the tower under study were completely
determined and a set of conditions could be examined. Thus for any
initial condition of inlet wet and dry bulb temperatures, inlet liquid
temperature, and liquid-to-gas ratio, the terminal conditions of the
air and liquid streams could be evaluated.
In performing the calculations as outlined, it was learned that
and thus K, was independent of the absolute concentration of dilute
volatile materials of interest in this study. The result was regarded
highly significant in validating the assumptions used in defining K,
and also in applying K in design calculations. The theory was applied
only to methanol rather than to all the BOD-producing compounds; but,
since methanol was the major contributor to the volatile BOD content,
it should be an excellent indicator of the overall process. The analy-
sis in Appendix "G" verifies the relationship.
The primary application of the theoretical work was to determine what
degree of treatment could be expected from conventional cooling tower
equipment when operated on waste water as was done in this study. The
theory is applied to the data described in the following section.
Experimental Developments
Experimental Technique - A very effective experimental method involving
nonequilibrium conditions was devised for laboratory studies of strip-
ping. The experimental procedure consisted of charging the cooling
tower basin with a known volume of waste water to be tested and then
-33-
-------
circulating the liquid through the tower and observing the change of
concentration with time. No feed was added to the tower after filling
the basin and no blowdown was withdrawn. Operation in this manner was
time-dependent and would not appear to embrace a simple mathematical
description. It resulted, nevertheless, that a simple balance of the
material entering and leaving the system would give an equation which
would accurately and quickly determine the stripping constant K. From
the ultimate BOD content that could be reached by prolonged treatment,
the procedure gave directly the amount of volatile BOD in the original
waste water. Analysis of the operation^) showed that the BOD in the
cooling tower as a function of time was
K-a
X2 ~
Equation (11) shows that as time progresses during the stripping opera-
tion, the concentration of methanol declines from its initial value of
x0. The quantity, a, in equation (11) was computed from the experi-
mental data by:
a = po - Vf\ tQ (12)
where V0 and Vf were the initial and final liquid volumes in the basin
(i.e., at t = o and t = t).
(2) The disappearance of BOD, x, could be expressed as a first order
process according to the relation
KQx = 0
where V was the liquid volume in the system at time, t, and Q
was the constant liquid recirculation rate. The rate of liquid
disappearance could be expressed as
dV -
where a was the constant water evaporation rate. Mathematical
manipulation of these equations gave
d* - (n v)
x ~ (a ~ K) V
which, when integrated, was Equation (11).
-34-
-------
The important utility of Equation (11) was obtaining K from the experi-
mental data. Taking the logarithms of both sides of the equation
yielded:
log x = log *0 + (X-Z-Zj log fT - $£\ (13)
a L "o J
or
log x = log (|tt) + (L^JL) log p0 - aQt\ (I3a)
It can be seen from this equation that a log-log plot of x versus
(1 - aQt/v0) resulting in a straight line would have a slope of
(K " a) from which K may be calculated.
The analysis of stripping thus far was developed for a single component.
The result, nevertheless, was equally valid for multicomponent systems.
The detailed treatment of the latter, more general case is presented in
Appendix "6".
Investigation of the Methanol Stripping Constant: (Description of
Experiments) - A comprehensive series of tests was conducted to deter-
mine the practical significance of the theoretical developments.
Initially, attention was given specifically to methanol which repre-
sented the major volatile constituents in the waste stream systems, and
experimental results were based on methanol concentrations instead of
BOD. Thirty-one experiments were performed consisting of 26 experi-
ments with Rome sixth effect condensate, two with Macon combined conden-
sate, two with Macon decker filtrate, and one with a methanol-water
mixture. The experiments were the non-steady state type usually
lasting from 40 to 60 minutes with one exception lasting 228 minutes.
The duration of the experiment was determined by the amount of initial
charge of waste water. The air flow velocities were variable in the
range of 200 to 600 feet per minute, and the liquid loading--that is
the recirculation rate—ranged between 1.00 and 4.00 gallons per minute
per square foot. A steam-heated heat exchanger maintained a liquid
temperature of 125°F at the top of the tower, and this temperature was
manually controllable within ±3°F. The top temperature was maintained
constant for all runs. The basin temperature was determined by the
process operating conditions. The temperature and humidity of the
entering air were at the ambient conditions of the laboratory. The dry
bulb temperature in all cases was close to 75°F, but the relative
humidity was more widely variant between 50 and 90 per cent. Data from
each experiment included initial charge volume, liquid flow rate, air
flow rate, final charge volume, total elapsed time of the experiment,
methanol concentration, and BOD concentration. Table 5 shows two typi-
cal sets of data from the batch stripping experiments, while complete
data are given in Appendix "D".
-35-
-------
TABLE 5
EXAMPLE DATA FROM NON-STEADY STATE STRIPPING EXPERIMENTS
Experiment 7-22
Methanol-Water
Initial Volume = 12.0 liters
Final Volume =4.0 liters
Duration of Experiment = 52 mins
Liquid-to-Gas Ratio
(L/G) = 1.35 lbs/H20
Time
(mins)
0
5
11
15
20
30
40
IDS dry air
Methanol
Concentration
590
340
170
130
75
40
35
Experiment 7-23
Rome 6th Effect Condensate
Initial Volume = 30.0 liters
Final Volume =6.24 liters
Duration of Experiment = 228 mins
Liquid-to-Gas Ratio
(L/G) =1.35
1k .
Ibs dry air
Time
(mins)
0
30
60
90
120
150
Methanol
Concentration
(mgTH
645
260
80
25
8
4
(Analysis of Data) - The experimental data were reduced to yield the
variables in Equation (13), the quantities (V0 - aQt) and x being of
primary interest. A log-log plot of these two variables resulted in
graphs as typically illustrated in Figure 8.
This plot shows several aspects of the nature of trace component re-
moval by batch stripping. The first plot shows that with increasing
time (i.e., as the system volume is reduced) the concentration of the
volatile component decreases. Also, the functional relationship
between these variables is logarithmic, demonstrated by the straight
line correlation of data on log-log coordinates. Equation (13) pre-
dicts this type of behavior if the ratio (K ~ c) is a constant.
a
All experiments were displayed on a log-log graph of the type shown
in Figure 8, and the data produced a linear graph in all cases. The
linear behavior was observed over three cycles of methanol concentra-
tion, and from this result it could be firmly concluded that K (K being
the stripping constant based on methanol concentration measurements
rather than BOD) was independent of methanol concentration. This
-36-
-------
10.OJ-
I
u>
A 6th Effect Condensate; Initial Volume = 12 liters;
pH 10.3
O 6th Effect Condensate; Initial Volume = 30 liters;
pH 10.2
D Synthetic waste composed of 10 mis methanol in
12 liters of initial water volume; pH 9.9
0.1
1.0
5.0
10.0 100.0
Methanol Concentration, mg/1
1000
FIGURE 8: STRIPPING EXPERIMENT - BASIN VOLUME VERSUS METHANOL CONCENTRATION
-------
result did not mean that K would not vary from day to day or would not
vary with different waste waters. It did mean, however, that for a
specified set of volatile components K would be constant.
(Effects of L/G) - Since concentration did not affect methanol removal
and since temperature was held constant, it was possible to concentrate
on the effects of liquid and air flow rates on treatment. These two
variables could be studied by operating the stripping experiments at
various liquid-to-gas (L/G) ratios, where L was the liquid phase flow
rate (waste water recirculation) in pounds per minute and a was the gas
phase flow rate (air) in pounds of dry air per minute.
For this analysis Equations (9) and (10) were used. As described in
the theoretical development section, the constants HTU\\ and HTUft could
be evaluated numerically with the aid of a computer. Several sets of
experimental data from the laboratory cooling tower were analyzed in
this manner. After those values were derived, they could be inserted
in Equations (9) and (10), and the performance of the laboratory cooling
tower could be evaluated for any other set of operating conditions.
Figure 9 shows the theoretical line resulting from the numerical analy-
ses of Equations (9) and (10). This theoretical evaluation was based
on an assigned inlet wet and dry bulb air temperature which was on the
average representative of the individual experiments. The individual
experiments shown on the figure were calculated from Equation (10).
Figure 9 shows that the relationship between K and L/C was a nonlinear
one. The result would be expected since the gas phase removed the
volatile constituent and the use of less air, that is higher liquid-
to-gas ratio, would produce less stripping. The scatter in the calcu-
lated data resulted from the uncontrollable variation of the wet bulb
temperature of the entering air. Although the water temperature at
the top of the tower was maintained at approximately 125°F, the bottom
(basin) temperature fluctuated with changes in the inlet wet bulb
temperature of the air. This affected the entire temperature profile
in the tower and the volatibility of methanol which directly varied K.
Despite the amount of scatter in the experimental data, it was obvious
that low L/G values (i.e., high air rates for a fixed liquid rate)
resulted in more efficient treatment.
(Effect of Temperature) - Finally, a series of experiments were con-
ducted to evaluate the effect of water temperature on the air stripping
of methanol. Figure 10 summarizes the results and clearly demonstrates
that increasing the water temperature in the tower increases treatment.
This behavior was predictable by theory as denoted by the theoretical
line, and was derived from the numerical manipulation of Equations (9)
and (10). While the results of these experiments were for methanol
only, any volatile material would be expected to behave similarly with
some variations in the slope of the curve because of difference in
vapor pressures.
-38-
-------
0.7
EXPERIMENTAL
— THEORETICAL
0.6
1 0.5
t.
o
\
u
(O
o.
a.
c
1C
0.4
0.3
0.2
\
V
0.1
0.0
0 1
2345
Liquid-Gas Ratio
FIGURE 9: EFFECT OF LIQUID/AIR RATIO ON STRIPPING
OF METHANOL IN A PILOT COOLING TOWER
-39-
-------
0.4
0.3
J=
i
s-
o
•!->
0
u_
O>
'5. 0.2
Q.
•i—
tS)
O
C
10
0.1
0.0
70
X|
X
X
X
% EXPERIMENTAL
THEORETICAL
80 90 100
Average Temperature - °F
110
FIGURE 10: EFFECT OF LIQUID TEMPERATURE ON STRIPPING
OF METHANOL IN A PILOT COOLING TOWER
-40-
-------
Investigation of Overall BOD Stripping Constant - The foregoing strip-
ping studies, involving only methanol, yielded a very significant
insight into the operation and capabilities of cooling towers as
strippers of volatile components from the water phase. Paper mill waste
water, however, consisted of many constituents, both volatile and non-
volatile. It was necessary, therefore, to investigate the effects of
the remaining constituents on treatment by stripping. An experimental
program was carried out identical to the one performed on methanol ex-
cept that BOD of the waste water was obtained in addition to the meth-
anol concentration during a batch experiment.
Due to the fact that BOD was more difficult to determine as accurately
as the concentration of methanol and, at the same time knowing that both
volatile and nonvolatile constituents were contributing to the BOD
properties of the waste waters, it was necessary to employ a slightly
different treatment of the experimental data. Equation (13) was still
the basic guideline, but being based on the material balance of volatile
components, it could not be quantitatively applied in its existing form.
It could, however, be used to suggest ways of presenting and verifying
the results of the batch experiment involving BOD measurements. Instead
of employing concentration directly, the fraction of the original BOD
in the waste, ( 1 - J?), was defined by:
where x was BOD concentration and x0 being the initial concentration
and V was the quantity of material remaining in the basin with V0 being
the initial liquid charge to the tower and basin.
This definition provided a dimensionless indication of BOD treatment.
The maximum time for an experiment (tm) was limited by the initial quan-
tity of material charged to the system (V0) , by the fraction of water
vaporized (a), and by the liquid recirculation rate (Q) and was computed
by
*•-£
(15>
This maximum time provided a basis by which experimental results involv-
ing various operating conditions could be normalized if a dimensionless
time variable were defined as
e-f-
(16)
-41-
-------
where t was the time during the experiment and t^ was the final or
maximum time of the experiment.
A total of 14 batch experiments were performed on two samples of Rome
sixth effect condensate. The results of these experiments are given
in Figures 11 and 12. Both figures show that as 0 increases (i.e.,
time increases) the fraction of the original BOD remaining decreased
rapidly at first and more slowly for longer 0. These results are
particularly significant because they show the fraction of the
original BOD that is not strippable, and this unstrippable BOD repre-
sents the nonvolatile components. The data show that 16 to 31 per cent
of the original BOD of this waste water was not volatile in one case
(Figure 11) and in the other case (Figure 12) the nonvolatile portion
was 11 to 23 per cent of the initial BOD. If the data are plotted ac-
cording to Equation (13), the resulting log-log plot is a straight line
showing that K is a straight line for the volatile BOD. Figure 13 pre-
sents such a plot for one set of data. The relationships between K and
volatile BOD are discussed in more detail in Appendix "G".
Although the experiments were performed with sixth effect condensate
exclusively, the series of experiments with the batch stripping tech-
nique proved that this selected mill waste water contained a volatile
BOD of 69 to 89 per cent of the total BOD, and this placed an upper
limit on the degree of treatment that could be expected from air strip-
ping of a waste water of this type.
Significance of Batch Stripping Experiments - The batch stripping type
of experiment had the particular advantage of providing a very conven-
ient and simple method of determining the amount of volatile BOD con-
tained in the waste water being tested. The knowledge of the volatile
BOD content indicates immediately the extent of treatment that can be
attained by stripping. From #, it is possible to predict the actual
treatment in a cooling tower. Since the thermal characteristics of
most cooling towers have been completely determined, the volatilization
rate, a, of water in the tower is known. Hence the equation
(5)
K + 2-
L
can be written in terms of relative volatility. Thus
~J
<£>
-42-
-------
00
I
c
o
o
o
o
CO
c
o
1/1
us
o
o
CD
1.0
•-r 0.8
0.6
0.4
O
o
£ 0.2
0)
0.0
0.0
Sample: 6th Effect Condensate ,
Air - 400 ft/min, Liquid - 125°* 4 gal/fr
PH
D
O
0
A
Initial Final
7.1
10.0
9.4
6.6
10.3
10.2
10.3
5.5
9.2
8.9
6.7
9.4
9.6
9.7
0.1
0.2
0.3 0.4 0.5
Dimensionless Time, t/V0/Qa)
0.6
0.7
0.8
FIGURE 11: STRIPPING EXPERIMENT - FRACTION OF BOD REMAINING VERSUS DIMENSIONLESS TIME, ROME 6TH EFFECT,
7-15-68 SAMPLE
-------
0)
+->
c
o
o
o
CO
c
o
u
to
i.
o
O)
o
o
LJ
4)
o
o
CO
Sample: 6th Effect Condensate
Temperature: 125°F
Air and Water Flows:
Air
Water
O
A
D
&
0
o
(ft/min)
200
500
400
500
200
400
500
(gal/min/ft^
8
8
8
4
2
2
2
0.1
0.2
0.5
0.6
0.7
0.8
0.3 0.4
Dimension! ess Time,
FIGURE 12: STRPPING EXPERIMENT - FRACTION OF BOD REMAINING VERSUS DIMENSIONLESS TIME, ROME 6TH EFFECT
AUGUST 1, 1968
-------
1.0
.9
.8
.7
o
0 .6
o
o
ca
c
o
u
ro
S-
.4
.3
o
o
c
0)
o
o
o
CO
Sample: 6th Effect Condensate
Temperature: 125°F
Air and Water Flows:
Water
(ft/min)
200
500
400
500
200
400
500
.03 .05 JD7 .1 .2
Dimensionless Time, t/y0/Qd)
.4
.6
.8 1.0
FIGURE 13: LOG-LOG RELATIONSHIP BETWEEN BOD CONTENT IN COOLING
TOWER AND THE DIMENSIONLESS TIME, t
-45-
-------
where (-) is the volatility of the treatable materials relative to
water. aHence for a waste water with £ known and a cooling tower for
which a is known, the amount of volatile materials that can be removed
in the cooling tower can be calculated directly with Equation (17).
The value of (&) can be determined experimentally in a simple labora-
tory experiment as outlined previously in Section VI. This experiment
is also very important because the fraction of BOD that is volatile is
evaluated and determines the limiting treatment that could be expected
in the cooling tower.
-46-
-------
SECTION VII
FULL-SCALE MILL STUDIES
Design and Construction of Cooling Tower
After the cooling tower system had been studied and proven successful
in laboratory and pilot-scale facilities, a full-scale tower was de-
signed, constructed, and operated to demonstrate its application in
reducing oxygen-demanding wastes. The design of this tower was unique
since both cooling and waste water treatment had to be considered.
Practically all past experience in the design of cooling towers had
been concerned only with cooling and the water streams involved had
been clean relative to the water streams being processed in this study.
The tower had to meet basic cooling design requirements to provide cool-
ing water for use in barometric condensers if it was to serve its major
function in the water reuse system. The factors, however, to be con-
sidered in sizing and selecting the proper tower for a given cooling
service were thoroughly documented and their effects on tower design
were well established from past experience (isf 14, 15). The criteria
necessary to establish the cooling characteristics of the tower above
included the volume of water to be processed, the temperature of the
hot water input, the temperature of the treated water, and the wet bulb
temperature of the ambient air. In addition to these variables, the
only other quantity to be fixed was the volume of air to be used or,
as otherwise stated, the liquid-to-air ratio. The air volume require-
ment, however, was more a function of BOD treatment than cooling and
was subject to other considerations.
All exploratory studies demonstrated that air stripping was the princi-
pal treatment mechanism. It was demonstrated further that stripping was
highly dependent on the liquid-to-gas ratio and the temperature in the
tower. To achieve the best possible conditions for stripping, it would
be necessary to employ the minimum liquid-to-gas ratio and the maximum
tower temperature. Physical limitations of the tower structure and the
operating expense of moving large volumes of air made it necessary to
compromise treatment with cost. Such considerations revealed that a
liquid-to-air ratio as low as 1.2 could be used. Tower temperatures
were determined by the cooled water requirement of the barometric con-
densers where the treated water was being reused. Laboratory tests,
already discussed in the experimental studies, showed that the methanol
removal corresponded to essentially the same percentage reduction in
volatile BOD. Changes in the ambient wet bulb air temperature also af-
fected the average tower temperature. Fluctuations from these effects,
however, could be compensated for by using a continuously variable speed
fan. Based on cost considerations, however, only two-speed fans were
economically feasible. This arrangement was judged capable of compen-
sating for seasonal changes in wet bulb temperatures. In experiments
-47-
-------
with the tower it was found that only one fan speed was necessary; the
maximum fan speed was used in all quantitative tests. The low fan
speed has been used effectively during low wet bulb periods.
One of the major problems anticipated with the cooling tower system
was foaming, primarily due to plans to handle decker filtrate. Every
effort was made to minimize the occurrence of foam, and a short labora-
tory study was conducted to evaluate both operational and design
methods that could be used to control foaming. From an operational
standpoint, the most critical variable was pH, and foaming tended to
be greatest when the pH was near the neutral point and was considerably
less under highly acid or basic conditions. See Appendix "E" for the
details of this study. From a design point of view, the tower basin
was arranged to keep the entire free liquid surface area under a rain
of falling water at all times, and a hanging wall was designed so that
water leaving the basin had to flow under and up into the pump sump
area. This retained the surface foam in the tower basin where the
falling waste water beat it down. Both crossflow and counter-flow in-
duced draft towers were considered for this application. Both types
of towers were found satisfactory, but two modifications were thought
to be necessary on the crossflow design. It was believed that, in the
crossflow configuration, foam generation could occur in the open pan
distribution system and in the open air plenum area. Suppliers were
asked to submit bids on a modified crossflow design which would
eliminate the exposed air-liquid interfaces at these points. The con-
ventional counterflow tower, however, was finally chosen because of
costs. As a final guard against foam, defoaming agents could be added
to the tower basin; and later in operation small quantities were
necessary.
The tower finally selected and installed was a counterflow design con-
sisting of two identical cells of 36 x 36 feet and 31 feet high. Each
cell was equipped with a 75 HP motor and a 16 foot stainless steel fan
capable of delivering 550,000 CFM in a 14 foot high fan stack. The
main structure consisted of treated redwood with corrugated fiberglass
sheathing and ten rows of plastic fill material. Because of the cor-
rosive nature of the waste water stream, stainless steel (in some cases
coated iron was substituted) was specified for all metals which would
be in contact with the liquid and air streams.
The unit was capable of providing cooled water at 90°F at the rate of
7000-8000 GPM from the tower basin to the barometric condensers on the
last effects of two sets of six-effect evaporators operating on kraft
mill black liquor. The heated water plus the condensed vapors (approxi-
mately 200 GPM) was pumped from the condenser hot wells to the cooling
tower for cooling and reuse. Evaporation losses and the continuous
blowdown of 700-1500 GPM was compensated for by adding combined conden-
sate from the first five evaporator effects (760 GPM), discharge from
the noncondensables and scrubber jets (580-640 GPM), decker filtrate
(350-700 GPM), and river water (0-600) GPM to the hot wells. The volume
-48-
-------
of water flowing to the cooling tower was expected to be in the range
of 7500-9500 GPM with a temp&rature range of 120-135°F. The tower was
designed to operate with ambient air condition having a wet bulb
temperature up to 80°F. The Fluor Products Company of Santa Rosa,
California, constructed the tower. The remainder of the tower system
was constructed by Georgia Kraft Company personnel or was sublet. Photo-
graphs of the installation are shown in Figures 14 and 15.
The cooling tower-barometric condenser system has performed extremely
well. This is an indication that the overall design of the system
has been most satisfactory. Two items are worthy of note: (1) Due to
the anticipated buildup of biological growth on the tower fill material,
it is necessary that the fill be rigid or provided with supports. In
this particular tower it has been necessary to modify the original sup-
port mechanism to assure the necessary rigidity. (2) The presence of
very small quantities of oils and fatty materials in the water causes
a reduction in cooling efficiency. This is discussed more fully in
Section VII under the heading of Cooling Characteristics.
Generally the tower produces a more uniform and controllable supply of
water for the barometric condensers than was available from the river
via the mill power house.
Collection of Experimental Data
The cooling tower evaporator system could operate in many modes with
various waste streams in or out of the system as indicated schemati-
cally in Figure 16. The tower could be operated with sixth effect
vapors with sufficient fresh water being added to balance evaporation.
Sixth-effect vapors plus combined condensate plus fresh water could be
used. Or other waste streams such as decker filtrate and turpentine
decanter underflow might be included. An extensive series of demon-
stration studies of these modes were conducted from May 1969 through
February 1970. The objectives of these studies were to determine:
1. The treatability of selected waste streams;
2. The effects of nutrients added to these streams and the ensuing
biological action generated;
3. The effects of water inlet temperature;
4. The extent of atmospheric emissions;
5. The optimum operating conditions;
6. The operating and treatment stability during process upsets;
7. Economics.
-49-
-------
FIGURE 14: PHOTOGRAPHS OF COMPLETED FULL-SCALE INSTALLATION
-50-
-------
FIGURE 15: PHOTOGRAPHS OF FULL-SCALE INSTALLATION DURING CONSTRUCTION
-51-
-------
tn
ro
i
6th EFFECT
VAPORS
OTHER WASTES
1
SEWER
COOL WATER
CONDENSER
HOT WELLS
NON CONDENSABLE
JETS
COOLING
TOWER
SLOWDOWN
FIGURE 16: SCHEMATIC OF COOLING TOWER-EVAPORATOR CONDENSER SYSTEM
-------
The data investigating the outlined items of interest were obtained
on a daily basis and summarized and reported on a period basis (a
period being approximately one month's duration). The period summaries
are presented in Appendix "F". For the purpose of data analysis and
the presentation of final results, original data have been further con-
densed in Table 6. This table contains only those data which were free
of adverse circumstances such as pulp mill startup and shutdown, pro-
duction upsets, erroneous analytical data, improper operation of the
cooling tower, and the like. While not of great overall significance,
the omission of data affected by such known adverse circumstances led
to considerable variation in the number of days of acceptable, trouble-
free operation on each separate study phase.
Analysis of Data from Full-Scale Studies
Studies of Selected Waste Streams - In analyzing the experimental re-
sults it is appropriate to begin with a review of the properties of the
streams selected for study and in particular the multi-effect evapora-
tor condensate. In the Macon mill, as in most pulp mills, the evapora-
tor systems are sextuple effect and produce three types of condensate.
One type is the condensate from the first effect which is clean stream
condensate that is returned to the process steam boiler. The other
condensates, produced in the remaining effects, are from vapors evapo-
rated from the black liquor being concentrated. These condensates are
contaminated with BOD-producing materials from condensed, volatile,
organic components and from liquid carry-over of black liquor spray
droplets. The condensate from the second through the fifth effects is
collected together and called combined condensate, and constitutes the
second type of condensate produced. The third type of condensate is
that from the sixth effect, which is considerably more contaminated
with condensed, volatile components and liquid carry-over. This con-
densate is separate from the other condensates because it goes to the
barometric condensers and mixes with the barometric condenser cooling
water. Table 1 (page 12), presented earlier, describes the typical
characteristics of the various condensates.
At the Macon mill, the differences in combined condensate and sixth
effect condensate were clearly evident. The combined condensate con-
tained up to 7,000 pounds of BOD per day, while the sixth effect, with
only one-third to one-half the volume, contained up to 12,000 pounds
of BOD per day.
The other streams studied were decker filtrate and turpentine decanter
underflow. The decker filtrate, being the wash water from the final
pulp washing operation, amounted to a relatively large volume compared
to the other wastes and contained up to 16,000 pounds of BOD per day.
Turpentine decanter underflow was a low volume flow stream containing
up to 3000 pounds of BOD per day.
-53-
-------
TABLE 6
SUMMARY OF COOLING TOWER DATA
Phase
1
2
3
4
5
6
7
8
9
10
11
12
13
Ho.
Days
5
7
7
6
25
21
13
8
14
4
17
24
15
Waste Streams to Tower
6th Effect Cond,
Vol.M)
mgd
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
0.29
BODUJ
Ibs/day
5,188
9,065
6,168
6,012
7,346
11,843
9,734
6,511
7,499
6,883
6,110
6,833
6,773
Comb. Cond.
Vol.
mgd
0.46
0
0.93
0.52
0
0
0
0.98
0
1.03
0
0
0
BOD
Ibs/day
2,743
0
6,544
3,157
0
0
0
5,847
0
5,619
0
0
0
Decker Filtrate
Vol.
mgd
0
0
1.58
1.58
1.72
1.91
1.55
1.47
1.74
1.46
1.63
1.60
1.67
BOD
Ibs/day
0
0
6,938
5,419
13,372
12,928
10,926
11,729
16,045
11.404
14,259
15,947
14,742
Turpentine Decan-
ter Underflow
Vol.
mgd
0
0
0
0
0
0
0
0
0
0
0
0
0^03
BOD
Ibs/day
0
0
0
0
0
0
0
0
0
0
0
0
2,231
Nutrients
Added*3)
0
0
0
0
0
0
Low
Low
High
High
0
0
0
Total
Flow to
Tower
mgd
9.28
10.16
11.42
11.53
11.13
11.83
10.91
11.41
10.00
12.44
11.21
11.22
11.61
Slowdown
Flow
mgd
0.42
0.33
2.39
2.03
1.56
1.79
1.49
2.37
1.63
2.27
1.65
1.60
1.67
Fresh
BOD to
Tower
Ibs/day
7,931
9,065
19,650
14,588
17.718
24,699
20,660
24,087
23,447
23.906
20,147
22.533
23.746
Avg.
BOD
Removal
Ibs/day
6,289
7,917
8,166
6,707
9,002
12,477
10.728
11,986
10,981
9,676
8,092
8,793
9,488
Tower
Temp. (4)
"P
97.2
97.8
105.9
100.9
104.5
103.6
103.6
105.5
101.2
104.2
93.7
94.0
92.9
BOD Removal
Avg.
*
79.3
87.3
41.6
46.0
50.8
50.5
51.9
49.8
46.8
40.5
40.2
39.0
40.0
Max.
*
83.7
91.9
54.3
74.1
66.5
70.5
68.2
49.6
68.8
45.4
62.2
62.3
60.5
Min.
i '
70.7
75.2
17.0
16.3
28.4
29. 5
34.2
26.6
36.2
31.7
13.2
14.6
24.8
I
en
(1) o3ume of 0.29 mgd was determined from a heat and materials balance around the 6th body on the two evaporator sets.
I?} Determined by difference from BOD balance around 6th effect condenser.
(3) Low - approximately 2.5 pounds of N and 0.5 pounds of P per 100 Ibs of BOD; high - approximately 5.0 pounds of N and 1.0 pounds of P per 100 Ibs of BOD.
(4J Average of dally temperature In tower feed and basin.
-------
In the operation of the cooling tower the waste streams were added in
varying quantities to the barometric condenser cooling water as makeup
and treated to give considerable reduction in BOD load discharged in
the total mill effluent and at the same time reduced the water demand
for the barometric condensers. It should be noted, as shown in Figure
16 (page 52), that sixth effect condensate was a part of the barometric
condenser water at all times and the waste streams which constituted
makeup were the combined condensate, decker filtrate, and turpentine
decanter underflow.
The data of Table 6 (page 54) served as the basic information for de-
termining the effect of using different waste streams, adding nutrients
and changing the cooling tower operating temperatures. Unfortunately,
however, differences in blowdown rates prevented direct comparison of
the data, and a series of theoretical assumptions and calculations had
to be made to remove this effect.
The calculations began with period No. 2 when sixth effect condensate
was the only waste stream being treated. By using Equation (5) on the
data, the stripping constant for this period could be derived. Hence
(0.873) (0.33)
0.177
This constant could then be used to evaluate the treatability of sixth
effect condensate under other blowdown rates.
The blowdown rate in period No. 1 was 0.42 mgd, and the cooling tower
recirculation rate was 9.28 mgd. Therefore, since the BOD in the sixth
effect was distributed in these streams, the percentage of treatment
was
(0.177)(1 +S4I)
R = - *•** = 0.833 or 83.3%
The BOD in the sixth effect feed to the tower was 5188 Ibs/day, and
83.3 per cent of it was removed by stripping. Therefore, the BOD
removal from the sixth effect feed stream was (5188 x 0.833) or 4322
pounds of BOD. The experimental data showed that 6289 pounds of BOD
were removed per day; hence the BOD removal from the combined conden-
sate was taken to be (6289 - 4322) or 1967 pounds of BOD per day. The
removal efficiency for the combined condensate could then be calcu-
lated as
-55-
-------
100 = 71.7%
where 2743 pounds of BOD/day was the BOD feed to the tower in the com-
bined condensate stream.
The calculation procedure used to apportion the BOD treatment between
sixth effect condensate and combined condensate in period No. 1 was
employed to apportion the treatment between sixth effect condensate and
decker filtrate in periods Nos. 5, 6, 11, and 12. A weighted average
stripping content for decker filtrate, #DF, from these four periods of
0.058 was then used as the #Dp in periods Nos. 3 and 4 to calculate a
stripping constant for combined condensate. Therefore, an average
stripping constant, K, was developed for all streams. The average con-
stant for sixth effect, #6th> was taken from Period No. 2 and assumed
to be the same for all periods at a value of 0.177. The average con-
stant for combined condensate, £QC, was obtained from a weighted average
of periods Nos. 1, 3, and 4 and is 0.119. A direct comparison between
periods Nos. 12 and 13 was used to develop the stripping constant, tfjU
for turpentine decanter underflow. This was felt to be valid because
of the close similarity of periods Nos. 12 and 13. This gives a value
of 0.115 for the stripping constant of the turpentine decanter under-
flow, £DU.
After average stripping constants had been estimated for all streams,
the percent removal for each waste stream was apportioned using the
various K's. These data are shown in Table 7 along with the actual and
calculated BOD removal accomplished in the system. As shown by Table 7,
the calculated BOD removal in pounds/day was obtained using the average
K's with each of the waste stream loadings and then adding the expected
BOD removal from each stream together. This is compared with the total
BOD removal observed. With the exception of periods Nos. 7-10, where
nutrients were added, there is good agreement between the calculated
and observed removals.
In calculating the results of Table 7, it was necessary to neglect the
effect of temperature. This, however, was not believed serious over
the temperature range of 93°F to 106°F. According to data on the labora-
tory cooling tower (Figure 10), the effect of temperature in estimating
the stripping constants for sixth effect condensate and combined conden-
sate could produce at the most a 10 to 20 per cent error.
While the data of Table 7 are estimates, it is believed that the approx-
imate range of treatabilities of the streams tested could be assigned.
It appears that with a liquid recirculation rate of about 10 mgd and a
blowdown rate of about 1.5 to 2.0 mgd, the efficiency of treatment was
between 55 and 65 per cent for sixth effect condensate, 45 and 55 per
cent for combined condensate and turpentine decanter underflow, and 25
and 35 per cent for decker filtrate.
-56-
-------
TABLE 7
en
^i
i
STRIPPING CONSTANTS AND REMOVAL
Period
1
2
3
4
5
6
7
8
9
10
11
12
13
B/C
.0452
.0325
.2090
.1760
.1401
.1513
.1365
.2080
.1630
.1824
.1470
.1420
.1440
Kfrffo
.177
.177
.177
.177
.177
.177
.212
.209
.251
.248
.177
.177
.177
Krr
.099
--
.123
.131
—
__
_ _
.406
.032
--
—
Anc
— —
~~ / 1 ^
.058
.058(1)
.056
.080
.065
.065(4)
.072
.072(5)
.052
.046
.058(1)
Pfith
83.3
87.3
55.4
59.0
63.6
62.0,0v
69.1 2
60.5 2
70.5 3
68.2V3)
62.7
63.4
63.1
EFFICIENCIES
Prr
71.7
--
44.7
50.1
—
—
_ _
79.9
__
17.9
—
—
£}£_
--
26.3
29.1
32.4
39.7
36.6
32.6
35.5
33.5
29.9
28.0
32.8
FOR FULL-SCALE TOWER
Lbs/Day
Actual
6,289~
7,917
8,166
6,707
9,002
12,477
10,728
11,986
10,981
9,676
8,092
8,793
9,488
Lbs/Day^6^
Calculated
6,398
7,917
8,108
6,146
9,138
11,467
9,943
9,265
9,430
9,880
8,451
9,610
10,242
Difference
Lbs/Day
- 109
0
+ 58
+ 561
- 136
+1010
+ 785
+2721
+1551
- 204
- 359
- 817
- 754
- 1.7
0
+ 0.1
+ 9.1
- 1.5
+ 8.8
+ 7.9
+29.4
+16.4
- 2.
- 4.
- 8.5
- 7.4
1
.2
(1) This K$f is the average of the tfpp values calculated for periods Nos. 5,
are calculated by use of Equation (5).
(2) Assumed 5 per cent increase due to nutrient addition.
(3) Assumed 10 per cent increase due to nutrient addition.
(4) Assumed same #Dp as period 7.
(5) Assumed same #pF as period 9.
(6) Calculated removal assumes £6th = 0.177, #CC = 0.119, £DF = °-058 and XDU = 0.115.
6, 11 and 12. K values
-------
Effects of Nutrients - By comparing periods Nos. 6, 7, and 9 (Table 6,
page 54), the effects of adding nutrients in the tower feed on treat-
ment of sixth effect condensate and decker filtrate can be determined.
Likewise, a comparison of periods Nos. 3, 8, and 10 should show the
effects of nutrients on a combination of three waste streams. In
periods Nos. 3 and 6, no supplemental nutrients were added, whereas
in periods Nos. 7 and 8 approximately 2.5 pounds of nitrogen and 0.5
pounds of phosphorus were added per 100 pounds of BOD, and in periods
Nos. 9 and 10 this amount was doubled. Using Equation (5) and the
average stripping constants previously discussed for sixth effect,
combined condensate, and decker filtrate, the comparison in Table 8
can be developed.
TABLE 8
EFFECTS OF NUTRIENTS ON BOD REMOVAL
Expected % BOD
Removal Without Actual %
Period Nutrient Level Nutrients* BOD Removal Difference
6th Effect Condensate plus Decker Filtrate
6 0 46.3 50.5 4.2
7 Low 48.0 51.9 3.9
9 High 40.0 46.8 6.8
6th Effect Condensate plus Combined Condensate plus Decker Filtrate
3 0 41.3 51.6 0.3
8 Low 38.5 49.8 11.3
10 High 41.3 40.5 -0.8
*Based on individual treatment efficiencies for sixth effect, combined
condensate, and decker filtrate when Xfith = 0.177, Krr = 0.119, and
* = 0.058. btn CL
The results for all combinations of sixth effect condensate, combined
condensate, and decker filtrate are rather inconsistent but do indi-
cate a general trend toward minor improvement in treatment efficiency
as found in the laboratory experiments when nutrients were added.
Bio-assays (see Appendix "H") conducted during these periods showed
that the total number of viable organisms per ml present with no nu-
trient additions was approximately 1.8 x 105. With 2.5 pounds of
-58-
-------
nitrogen and 0.5 pounds of phosphorus per 100 pounds of BOD, the or-
ganism count went up to approximately 1.6 x 10/. When this nutrient
additional level was doubled, the viable organism count exceeded 109
per ml. This increase in biological activity tends to support that
nutrient addition does improve the overall biological conditions in
the tower.
It should be noted that during the entire experimental period, con-
siderable amounts of slime were present in the tower, and even at the
zero nutrient level some biological treatment must occur.
Cooling Characteristics - In the design of the tower it was not antici-
pated that the contaminants in the waste waters circulating through the
tower would significantly affect evaporation and cooling. Based on the
standard criteria for evaluating cooling tower performance (16), however,
it has been determined that the tower is operating at about 80 per cent
of the cooling efficiency that would be expected using uncontaminated
water. It has been found that small amounts of oils and fatty materials
build up in the water, and these are believed to be responsible for the
failure of the tower to yield the expected cooling performance. Thus
it is recommended that cooling towers to be used on pulp mill wastes be
designed with an additional 20 per cent allowance for decreased evapora-
tion rates due to contaminants in the water.
Treatment of Slowdown - Because of the concentrating effect of the cool-
ing tower on the various waste streams, there was some question as to
the amenability of the tower blowdown to treatment in subsequent waste
treatment facilities. The laboratory scale bio-oxidation unit, as
described in Section V, was used to simulate treatment of the blowdown
in an aerated lagoon with five-day retention. The results of these
studies showed an average BOD removal efficiency of 86.4 per cent.
Therefore, it is evident that the character of the blowdown would not
adversely affect conventional biological waste treatment facilities.
Atmospheric Emissions - Table 1 (page 12) indicates that the waste
streams being treated in the tower contain concentrations of organic
sulfur compounds responsible for odor problems associated with the kraft
pulping process. Because of the concern for all potential environmental
pollution problems, several studies of the emissions from the tower were
made. A summary of the data on gaseous sulfur emissions is included in
Table 9. As expected, these studies show emissions of several organic
and inorganic sulfur compounds in low concentrations. No noticeable
odor problems associated with the cooling tower have been reported, and
it is believed that with the dilution and dispersion of the stack gases,
no significant odor problem would be experienced.
Table 9 shows that the noncondensable jets from the barometric conden-
sers and the turpentine decanter underflow are the most significant
contributors to the sulfur emissions. If further control of sulfur
emissions were desirable, it would be possible to reduce them to the
-59-
-------
level shown in the bottom line in Table 9. The noncondensables could
be removed and burned rather than discharged to the hot well. It
should also be possible to prestrip the turpentine underflow to remove
the highly volatile organic sulfur components prior to its introduction
into the tower.
TABLE 9
AVERAGE GASEOUS EMISSIONS FROM COOLING TOWER (PPM)
Sulfur Hydrogen Methyl Dimethyl Dimethyl
Waste Streams Treated Dioxide Sulfide Mercaptan Sulfide Disulfide
6th effect condensate,
decker filtrate, tur-
pentine decanter
underflow, noncon-
densable jet stream .047 1.4 2.3 1.5 2.1
6th effect condensate,
decker filtrate, non-
condensable jet stream .138 1.3 1.7 0.5 0.8
6th effect condensate,
decker filtrate .106 0.4 0.0 0.3 1.8
Advantages of the Cooling Tower System - The BOD removed by the cooling
tower during the experimental period amounts to about 25-30 per cent of
the entire organic pollution load discharged from the mill. Thus it is
estimated that the aeration requirements in a subsequent treatment
facility for this mill can be reduced by this amount.
After taking into account the reduction in power required for pumping
cooling water from the river to the mill, the net power required to
operate the cooling tower is about 135 HP. At the Macon mill 8-10 mgd
less river water is required, resulting in an estimated 265 HP reduction
in pumping power. The BOD reduction efficiency of the cooling tower,
then, can be estimated at about 70-75 Ibs BOD/HP/day. This compares
very favorably with the reductions in conventional aerated lagoons which
under optimum conditions can be expected to remove only about 50 Ibs
BOD/HP/day.
Another advantage which may or may not apply to other mills is that a
more constant supply of cooling water is attributable to the system,
and this has resulted in more stable evaporator operation. The evapo-
ration of waste water in the tower reduces the hydraulic load on
subsequent treatment facilities by a small amount in addition to the
more significant BOD reduction.
-60-
-------
The operation of the system has been quite reliable and relatively
trouble-free. While some foaming has occurred in the tower basin on
occasion, this has been easily controlled by addition of defearners.
The maintenance and defoamer cost of the system are probably less than
for equivalent treatment in an aerated lagoon system.
Economics - The initial cost of installing the cooling tower system was
$239,000. A breakdown of the cost by major items is as follows:
1. Cooling Tower $ 66,000
2. Basin 11,000
3. Pumps 13,000
4. Piping and Valves 34,500
5. Electrical 36,000
6. Instrumentation 15,000
7. Miscellaneous 4,000
8. Labor 36.000
Construction Total $215,500
Engineering and Supervision 23,500
Total $239,000
The estimated cost to install surface condensers for the Mead Division
mill is $229,000.
Due to the reuse of cooled water in the noncondensable jets and evapo-
ration and carry-over losses from the tower, the volume of mill
effluent is reduced by about one million gallons over what could be
achieved by surface condensers. As indicated earlier, this cooling
tower accomplishes about a 10,000 Ib/day reduction in total mill ef-
fluent BOD. The reduction in BOD is not normally achieved where sur-
face condensers are used. There would be several ways to calculate
the value of the reduction in volume and BOD, but at the Macon mill,
where treatment in a combined system is planned, the cost savings is
in the neighborhood of $250,000 in construction costs alone. The sav-
ings in annual operating cost for these volume and BOD reductions is
expected to be about $20,000. This exceeds the $13,500 per year re-
quired for the net 135 HP necessary for the cooling tower plus approxi-
mately $5,000/year defoamer cost.
In addition to the very favorable cost factors, the simpler operation
of a direct contact condenser offers operating advantages over the
surface condenser where the inherent resistance to heat transfer of
the tube walls and fouling of the tubes must be considered.
It is anticipated that this installation will continue to provide test
data and remain available as a continuing demonstration project for
all mills of the industry interested in making a similar installation.
-61-
-------
SECTION VIII
ACKNOWLEDGEMENTS
We wish to acknowledge the Research and Development Committee, its
chairman, Mr. E. V. McSwiney, and the Mead Division staff and manager,
Mr. W. E. Davey, of Georgia Kraft Company for their support and
encouragement in this project.
All of the project activities were coordinated and administered by
Dr. R. B. Estridge, Director of Research and Development, Georgia
Kraft Company.
The design, construction, and operation of the laboratory and pilot
units were performed by a team from Georgia Kraft Company's Research
and Development Center consisting of Dr. Estridge and Messrs. B. G.
Turner, James T. Van Horn, Rodney Jones, and Michael Wise.
The design, construction, and operation of the full-scale unit at
Georgia Kraft Company's Mead Division mill were performed by a team
from the Research and Development Center and Mead Division Engineering
and Technical Departments consisting of Dr. Estridge and Messrs. Turner,
Robert L. Smathers, Pat McHugh, E. L. Wilson, Ray Johnson, and Sam Rose.
The assistance of Dr. L. J. Thibodeaux of the University of Arkansas in
the pilot studies and theoretical analysis is gratefully acknowledged.
This report was prepared by Dr. J. Andrew McAlister and Mr. Turner with
assistance from Mr. Jones and Mrs. Sue Williams.
The aid provided by Mr. William J. Lacy, Mr. George R. Webster,
Mr. Ralph Scott (Project Officer), and Mr. Edmond P. Lomasney of the
Office of Research and Monitoring, Environmental Protection Agency,
was greatly appreciated.
This project was partially funded by the Environmental Protection
Agency, Office of Research and Monitoring, under a Section 6(b)
Research and Development Grant by the Industrial Pollution Control
Branch of the Division of Applied Science and Technology, Office of
Research and Monitoring, Washington, D. C.
-63-
-------
SECTION IX
REFERENCES
1. Cohn, R. 6. and Tonn, E.T., "Use of a Cooling Tower in Black
Liquor Evaporation", Tappi 47, No. 3, pp 163A-165A (1964).
2. Smathers, R. L. and Frady, J. H., "The Utilization of a Cooling
Tower for BOD Removal and Water Reclamation", paper presented at
National Paper Awards Contest (TAPPI}, February 1969.
3. Mohler, E. F., Jr., Elkin, H.F., and Kummick, L.R., "Experience
with Reuse and Bio-Oxidation of Refinery Wastewater in Cooling
Tower Systems", JWPCF S63 No. 11, pp 1380-1392 (1964).
4. Celanese Corporation, Rome, Georgia; personal visit and discussions,
5. American Public Health Association, Standard Methods for the
Examination of Water and Waste Water, 12th edition. New York:
American Public Health Association, Inc., (1965).
6. Adams, Donald F., and Koppe, Robert K., "Gas Chromatographic
Analysis of Hydrogen Sulfide, Sulfur Dioxide, Mercaptans, and
Alkyl Sulfide and Disulfide." TAPPI 42, No. 7, pp 601-605
(July 1959).
7. Cave, 6.C.B., "The Collection and Analysis of Odorous Gases from
Kraft Pulp Mills", TAPPI 46, No. 1, pp 1-20 (January 1963).
8. Blosser, Russell 0., and Cooper, Hal B.H., "Compendium of Methods
for Measuring Ambient Air Quality and Process Emissions:
Section 11A - Gaseous Emissions - Automatic Techniques - Electro-
lytic Titration." National Council of the Paper Industry for Air
and Stream Improvement, Inc., Atmospheric Pollution Technical
Bulletin No. 38 (December 1968).
9. Hruitfiord, B.F., and McCarthy, J.L., "SEKOR I: Volatile Organic
Compounds in Kraft Pulp Mill Effluents Streams", TAPPI 50, No. 2
(February 1967).
10. Bethge, Per Olaf, and Ehrenborg, Lalla, "Identification of Volatile
Compounds in Kraft Mill Emissions", Svensk Papperstidning 70, No.10,
pp 347-350 (May 31, 1967).
11. Ruus, Av. Lennart, "A Study of Waste Water from Forest Products
Industries. 7. Composition and B.O.D. of Sulfate Pulp Mill Con-
densates", Svensk Papperstidning 67S No. 19, p 751 (October 15,
1964).
-65-
-------
12. Minch, V.A., Egan, John T., and Sandlin, McDewain, "Design and
Operation of Plastic Filter Media", JWPCF 34, No. 5, pp459-469
(May 1962).
13. Treybal, R.E., Mass Transfer Operations, New York: McGraw-Hill
Book Company, Inc. (1955).
14. Bird, R. Byron, Stewart, Warren E., and Lightfoot, Edwin N.,
Transport Phenomena, New York: John Wiley & Sons, Inc. (1962).
15. McKelvey, K.K., and Brooke, M., The Industrial Cooling Tower,
New York: Elsevier Publishing Company (1959).
16. Cooling Tower Institute, "Acceptance Test Procedure for Industrial
Water-Cooling Towers", CTI Bulletin ATP-105, April 1959.
-66-
-------
SECTION X
PUBLICATIONS AND PATENTS
Three publications have been written as a result of this study.
1. Turner, B. G. and Van Horn, James T., "Identification of Volatile
Components in Kraft Mill Evaporator Condensates". Presented and
declared winner of 1969 Southeastern TAPPI and 1970 National TAPPI
Paper Awards Contests. To be submitted for publication in the
journal of TAPPI.
2. Estridge, R. B., Turner, B. G., Smathers, R. L., and Thibodeaux,
L. J., "Treatment of Selected Kraft Mill Wastes in a Cooling Tower".
TAPPI 54, No. 1 (January 1971).
3. Thibodeaux, L. J., Estridge, R. B., and Turner, B. G., "Measurement
of Relative Volatilization Rates of the Water-Miscible Fractions in
an Aqueous Effluent." Presented at National AIChE Meeting in
Cincinnati, Ohio, May 16-19, 1971.
No patents have been produced or applied for under this project.
-67-
-------
SECTION XI
GLOSSARY
Frequently Used Abbreviations:
BOD = Biological Oxygen Demand (5-day, 20°C); volatile BOD is the
total amount of BOD which can be removed by extended stripping.
cc/min = Cubic centimeters per minute.
COD = Chemical Oxygen Demand.
ft/mi n = Feet per minute.
gpm = Gallons per minute.
Ibs/day = Pounds per day.
mis = Milliliters.
mg/1 = Milligrams per liter.
mgd = million gallons per day.
ppm = Parts per million.
Symbols:
Roman
a = Evaporation rate, expressed as fraction of a unit of liquid
which is evaporated as it passes from top to bottom of tower.
B = Slowdown flow rate from tower basin.
d = Differential Operator.
G = Gas (or air) flow rate.
HTU = Height transfer unit.
7i = Enthalpy of moist air; h* indicates enthalpy of saturated air.
K = Stripping constant, expressed as fraction of volatile material
removed from a unit of liquid as it passes from top to bottom of
tower.
-69-
-------
k = Proportionality constants in Equations (2) and (3).
L = Liquid flow rate through tower.
Q = Flow rate of water recirculating to tower.
R = Removal of BOD as fraction or as percent of BOD in feed liquid.
T = Pounds of water vapor evaporated per hour from water in tower.
t = Temperature in degrees Fahrenheit; down time in minutes.
V = Volume of liquid added to cooling tower in batch experiments.
x = Liquid phase concentration.
y = Vapor phase concentration.
z = Column packing height.
Greek Symbols
0 = Dimensionless time unit definition Equation (16).
Subscripts:
CC = Combined condensate.
DF = Decker filtrate.
F = Feed liquid.
H = Heat transfer.
h = Henry's Law.
L = Liquid.
M = Methanol, also mass transfer.
m = Maximum.
Tu = Turpentine.
T = Water evaporated into exhaust air stream from tower.
wb = Wet Bulb.
-70-
-------
Z = Position (height) in tower.
o = Initial conditions in batch experiments,
1 = Condition at top of tower.
2 = Condition at bottom of tower.
6th = Sixth effect condensate.
Superscripts:
* = Equilibrium.
-71-
-------
SECTION XII
APPENDICES
Page No.
A. Laboratory Sparging Experimental Data 75
Table 1: Analysis of Sixth Effect Condensate Used in
Sparging Experiments 75
Table 2: BOD and COD Data for Sixth Effect
Condensate Sparging Experiments 76
B. Description of Laboratory Cooling Towers 77
C. Experimental Data Investigating Treatment Mechanisms in
Laboratory Cooling Towers 81
Table 1: Experimental Data for Batch Experiments
Comparing Air Stripping with Chemical Oxidation . 81
Table 2: Experimental Data for Continuous Flow Experiments
Comparing Air Stripping with Biological Treat-
ment, Full Nutrient Loading - Feed Rate, 220
cc/min; Recirculation Rate, .5 GPM 82
Table 3: Experimental Data for Continuous Flow Experiments
Comparing Air Stripping with Biological Treatment,
Full Nutrient Loading - Feed Rate, 118 cc/min;
Recirculation Rate, .25 GPM 83
Table 4: Experimental Data for Continuous Flow Experiments
Comparing Air Stripping with Biological Treatment,
Full Nutrient Loading - Feed Rate, 470 cc/min;
Recirculation Rate, 1.0 GPM 84
Table 5: Experimental Data for Continuous Flow Experiments
Comparing Air Stripping with Biological Treatment,
One-Half Nutrient Loading 85
Table 6: Experimental Data for Continuous Flow Experiments
Comparing Air Stripping with Biological Treatment,
No Nutrients 86
D. Experimental Investigations of Air Stripping 87
Table 1: Experimental Conditions for Batch Studies of
Air Stripping 87
Table 2: BOD and Methanol Determinations for Batch
Stripping Experiments 88
Table 3: Experimental Data for Continuous Flow Experiments
Investigating Effects of Water Temperature on
Air Stripping 89
E. Investigation of Foaming in Laboratory Cooling Tower ... 90
-73-
-------
Page No.
F. Experimental Data for Full-Scale Cooling Tower 93
Table 1: BOD Data for Full-Scale Cooling Tower
Operation 93
Table 2: Methanol Data for Full-Scale Cooling Tower
Operation 97
Table 3: Flow Rates for Full-Scale Cooling Tower
Operation 99
Table 4: Temperature Data for Full-Scale Cooling Tower
Operation 103
G. Measurement of the Relative Volatilization Rates of the
Water-Miscible Fractions in an Aqueous Effluent . . . . 107
H. Microbiological Report on Kraft Mill Cooling Tower Waters . 142
-74-
-------
APPENDIX A
LABORATORY SPARGING EXPERIMENTAL DATA
TABLE 1
ANALYSIS OF SIXTH EFFECT COMPENSATE
USED IN SPARGING EXPERIMENTS
Analytical Results for Experiment
Measurement No. 1 No. 2 No. 3 No. 4 No. 5 No. 6
BOD, mg/1 1600 1380 1400 1366 810 843
COD, mg/1 5080 5222 4019 3704 3039 1830
Conductivity, umhos 2300 2600 1800 1500 930 500
pH 10.7 10.2 10.1 9.6 9.0 8.7
Phenolphthalein
Alkalinity, mg/1 378 222 148
Total Alkalinity, mg/1 788 862 502 -- -- 114
Total Solids, mg/1 3500 4036 2756 2448 1492 720
Dissolved Solids, mg/1 3500 3986 2696 2394 1492 636
Suspended Solids, mg/1 0 50 60 54 0 84
Total Volatile
Solids, mg/1 1348 1254 924 990 736 368
Dissolved Volatile
Solids, mg/1 1348 1254 924 990 736 276
Suspended Volatile
Solids, mg/1 0 0 0 0 0 92
-75-
-------
APPENDIX A
(Continued)
TABLE 2
BOD AND
Experiment
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
COD DATA FOR
Elapsed
Time
(mi ns )
0
1
1.5
2
3
4
22
25.5
27
28.5
47.5
48
0
2
4
22
24.5
26
46.5
51
0
1.5
3
5
22
27
47
52
0
2
4.5
22.0
26.5
47.0
51.5
71.5
0
5.0
23.5
29
47.5
53.0
71.0
0
19
46
62.5
SIXTH EFFECT CONDENSATE
BOD
For N2
Sparging
(mg/1)
1600
--
--
--
__
—
--
--
1022
--
1460
--
1380
1330
--
—
--
--
--
--
1400
1380
1280
1185
1166
--
—
--
1366
1200
933
1133
833
--
—
1033
810
703
1066
770
940
950
920
843
810
722
850
Results
For Air
Sparging
(mg/1)
1600
1320
1200
--
—
1200
—
--
948
—
1210
--
1380
1380
1360
—
--
—
--
—
1400
1300
1250
1080
966
--
--
--
1366
1100
933
1150
1050
—
—
900
810
766
666
600
986
840
810
843
759
775
800
SPARGING
COD
For Ng
Sparging
(mg/1)
5080
4920
--
4680
4800
4720
4610
4732
__
4120
_-
4039
5222
5386
5098
4863
--
4923
4885
4769
4019
3862
3901
3861
3541
3846
3618
3732
3704
3658
3565
--
3440
3332
3185
3136
3039
2692
2500
2115
2211
2381
1905
1830
1658
1675
1675
EXPERIMENTS
Results
For Air
Sparging
(mg/1)
5080
4920
__
4680
4440
4840
4243
4406
_ _
3835
__
3794
5222
5059
5137
4902
_.
4846
4692
4731
4019
3862
3901
3861
3618
3580
3541
3503
3704
3519
3426
._
3283
3284
3038
2744
3039
3028
2308
2404
2115
2000
1809
1830
1516
1675
1618
-76-
-------
APPENDIX B
DESCRIPTION OF LABORATORY COOLING TOWERS
A. Purpose of Laboratory Cooling Towers
Basic laboratory experiments had shown that air stripping was the
major factor in the reduction of BOD in combined condensate and
decker filtrate, and that biological action and chemical oxidation
were only minor mechanisms. It was desirable that these mechanisms
be investigated in a cooling tower large enough to retain the
principal features of a typical industrial installation, yet small
enough to allow the processing of the air and liquid streams with
conventional laboratory devices. A laboratory tower design was
developed which had features similar to the Macon pilot plant and
to the proposed demonstration tower. Besides simulation, the
design provided for visual inspection of the air-liquid contacting
process, and it had the necessary physical arrangements and
auxiliary equipment to allow complete quantitative measurements of
widely varying operating conditions. So that the BOD removal
mechanisms could be separated and still be studied under identical
conditions, it became necessary to build two identical laboratory
towers.
B. Design, Construction, and Operation of Laboratory Cooling Towers
The laboratory cooling towers were counter-current flow, columnar
gas-liquid contactors of 0.25 square foot cross-sectional area
(6 inches by 6 inches) and Plexiglass construction. Overall height
was approximately 12 feet, 10 of which were packed with Poly-Grid
media stacked vertically on 1-1/8 inch centers. The sections of
packing were cut with random grid positions so that when stacked
in the tower the grid network would not form vertical channels
through the network structure. The Poly-Grid was a plastic
material of high void packing consisting of sections 1-1/4 inch
high with a grid mesh of 2 inches by 2 inches. At three equally
spaced levels along the height of the towers, liquid deflectors
were glued to the corners to force the falling liquid to distribute
itself over the tower cross-section; otherwise, much of the liquid
would run down the corners of the tower. Air for cooling was drawn
through the tower (induced draft) by two turbine type fans (Staplex
high-volume air samplers) located atop the column. A squeeze-
action, positive displacement, Randolph pump (The Randolph Company,
Houston, Texas) provided liquid flow. Auxiliary equipment included
a heat exchanger to reheat the cooled liquid plus rotameters,
thermometers, pressure devices, rheostats, and sample parts. A
schematic diagram describes the apparatus in Figure B-l.
-77-
-------
Preliminary simulation runs were performed with the flow of liquid
routed as shown in Figure B-l. Fresh waste water was fed through a
rotameter combined with recycle waste water from the lower basin,
heated and then introduced into the tower above the packing. A flow
distribution plate spread the water flow evenly over the tower
cross-section. When the liquid trickled and splashed downward as
it descended the tower, it was contacted by a stream of air moving
upward. The process of water cooling (predominant tower process)
took place as water molecules left the liquid phase and attempted
to saturate the air. This interphase transport of water molecules
and their accompanying latent heat of vaporization resulted in the
cooling of the remaining liquid water. The cooled water was col-
lected in the basin, at which point part was recycled and part was
removed as blowdown to maintain steady state operation.
Air flow through the tower was controlled by variable power trans-
formers connected in the power supply to the Staplex air samplers,
and the amount of air flow was determined from pressure drop
measurements on gas riser tubes which passed through the water flow
distribution plate. There were nine tubes, 1 inch in diameter and
6 inches long. A 2-inch inclined water manometer measured the
pressure drop across the tubes. The pressure drop was correlated
with the air flow through the tubes, and a pressure-drop/air-flow
calibration curve was prepared for routine operation. For cali-
bration purposes, the air flow through the tubes was determined
from impact pressures which, in turn, were carefully measured with
a small pi tot tube positioned at the discharge end of the gas riser
tubes. The application of conventional fluid flow laws and the
assumption of smooth tube walls afforded a calculation of gas flows
in the calibration measurements.
Cooling was not of primary concern; it was considered only as it
was involved in the BOD treatment process. Waste water treatment
parameters were monitored primarily while cooling parameters were
recorded only as support information. A list of the variables
monitored and the analyses performed while the units were in
continuous operation included:
1. Feed rate
2. Recycle rate
3. Blowdown rate
4. Air rate
5. Liquid inlet temperature
6. Basin temperature
7. Air inlet and outlet wet bulb and dry bulb temperatures
8. BOD, COD, pH, conductivity, sulfide content, and total solids
content of feed and blowdown.
-78-
-------
Feed
Rotameter
Air Press. Regulate
55 Gal. Drum
Air In
Basin Temp.
1
Inclined
Water
Manometer
Slowdown
FIGURE B-l: SCHEMATIC DIAGRAM OF LABORATORY COOLING TOWER
-79-
-------
Notes were made of the foaming conditions within the towers and
the basins, and temperatures in some experiments were recorded
at intermediate points within the tower.
-80-
-------
APPENDIX C
EXPERIMENTAL DATA INVESTIGATING TREATMENT MECHANISMS IN LABORATORY COOLING TOWERS
TABLE 1
EXPERIMENTAL DATA FOR BATCH EXPERIMENTS COMPARING AIR STRIPPING WITH CHEMICAL OXIDATION
Experiment
No. 1*
No. 2*
Elapsed
Time
(mins)
0
5
10
15
30
40
0
5
10
15
30
55
58
Stripping with Nitrogen
Inlet
Temp.
(°c)
50
50
52
45
53
52
52
45
55
54
52
--
55
Basin
Temp.
(°c)
20.0
19.5
18.0
18.0
14.5
16.0
32.0
30.0
20.0
18.0
17.0
--
17.0
Liq. Volume
in Basin
(1)
24.0
23.4
22.9
22.3
20.6
19.5
8.0
7.5
7.0
6.5
5.0
--
1.8
BOD
(ppm)
807
747
655
605
540
485
684
522
487
449
391
—
530
Methanol
(ppm)
310
290
240
225
160
no
290
174
149
94
40
—
0
- Stripping with Air
Inlet
Temp.
(°c)
40
40
57
50
55
50
52
45
53
55
52
55
™ ™
Basin
Temp.
(°c)
15.0
15.0
16.0
16.0
19.0
19.0
34.0
28.0
26.0
22.0
22.0
22.0
™ ~
Liq. Volume
in Basin
(1)
24.0
23.5
23.0
22.5
20.9
19.8
8.0
7.5
7.0
6.5
5.0
2.6
~-
BOD
(ppm)
873
710
675
677
595
570
684
490
454
416
359
325
—
Methanol
(ppm)
400
310
280
240
160
140
290
165
130
100
48
T
—
00
* Experiment conditions:
T = Trace
gas flow, 50 CFM; liquid recirculation rate, 0.5 GPM.
-------
APPENDIX C
(Continued)
TABLE 2
EXPERIMENTAL DATA FOR CWTINUOUS FLOW EXPERiyNTSCgyARINS AIR STRlPPiyMITH BIOLOGICAL TREATMENT
FULL NUTRIENT LOADING - FEED RATE. 220 CC/HIN; RECIRCULATION RATE. .5 6PH'
Exp.
1
2
3
4
5
6
1
7
8
9
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Air
ft/mi n
200
200
200
200
200
200
400
400
400
400
400
400
560
560
560
560
560
560
Temperature
Tnl.f 1"-'!-
°C
50.0
48.0
51.5
53.0
51.0
51.5
52.1
52.0
51.5
51.5
51. a
5,;, 0
50.1
51.5
51.4
52.0
51.4
52.0
*C
26.0
25.0
25.0
25.0
25.0
26.0
22.0
19.5
21.0
20.0
21.0
18.2
19.0
18.0
21.5
20.0
20.0
19.0
pH
Tn{t 1
9.25
9.25
8.95
8.95
9.45
9.45
9.34
9.34
9.3
9.3
8.5
8.5
9.4
9.4
9.3
9.3
9.3
9.3
FTnaT
8.8
fl ?
8.65
7.80
9.2
8,2
9.2
8.06
9.1
8.?
7.5
7.2
8.8
8.04
8.3
8.1
3.6
8.1
Cond.
tnU |f'-^
umhos
1200
1?00
980
980
1260
1260
930
930
960
960
740
740
770
770
820
820
760
760
umhos
1400
1260
1040
1200
1370
1650
1040
1340
1000
1320
900
1120
790
1040
860
1140
730
90C
Solids
tnft |C(«il
pp*l
672
1672
1444
1444
2016
2016
1228
1228
1208
1208
1088
1088
1076
1076
1068
1068
960
960
ppm
2180
1964
1664
1848
2420
2516
1620
1928
1532
1920
1544
1S36
1512
1404
1684
1688
1156
1188
Methanol
Content
InU |"-"1
•I/I
250
250
250
?SO
220
220
210
210
250
250
235
235
250
250
230
230
210
210
mg/1
60
60
63
60
60
63
No
Data
50
No
Data
No
Data
*
*
50
0
30
10
50
0
NH3 as N
Irfft |F*nal
•9/1
6.0
6.0
40
40
84.0
84.0
56.0
56.0
54.0
54.0
•9/1
No Dal
No FIJI
No Da
No Dal
.4
.27
3.8
14.4
No Dal
No Dat
13.8
16.3
5.4
20.2
No D
No D
17.7
7.5
Total
nnn
•9/1
a —
a
3.1
3.1
16.0
16.0
a
27.3
'7.3
16.0
16.0
ta —
17.5
17.5
P04
FTnaT
•an
9.9
6.2
16.0
22.5
36.3
41.0
20.0
27.5
25.0
22.5
STS^
down
'ate
cc/Bin
76.0
72.0
156
156
176.0
180.0
148
156
165
156
125.0
141.0
140
141
141
146
148
154
CO
nit
•9/1
2465
2465
2180
2180
2320
2320
2080
2080
1910
1910
1848
1848
2035
2035
1895
1895
1740
1740
3
'Inal
-9/1
2435
2500
880
940
2100
2245
1855
2020
1620
1800
1473
1408
1630
1665
1605
1660
1420
2280
BO
nit
ng/1
962
962
048
048
020
1020
1033
1033
1135
1135
738
738
950
950
1165
1165
1050
1050
D
Inal
•g/1
721
667
685
645
638
465
633
376
500
300
466
279
446
343
513
347
596
416
I BOO
enoval
40.5
46.2
53.5
56.2
50.0
62.5
58.2
74.5
66.6
81.0
64.2
75.8
65.0
76.9
71.6
80.0
61.8
72.3
It ReMvaTI « BOO
f Vola-**RM»val
He BOD as NeOH
65.8
75.2
76.9
80.8
81.4
101.8
83.74
107.1
95.8
116.7
83.5
98.75
84.6
100.1
93.2
104.0
80.5
94.0
20.4
20.5
19.0
19.2
16.4
16.0
16.4
—
30.2*
30.1*
22.3
25.6
17.6
18.6
16.3
19.4
Volatile
BOD
%
61.4
61.4
69.5
69.5
61.4
61.4
69.5
69.5
69.5
69.5
76.8
76.8
76.8
76.8
76.8
76.8
76.8
76.8
* " j
Removal
80.8
81.2
82.1
83.0
78.2
76.6
No
Data
83.1
Ho
Data
No
Data
98.8
98.6
87.3
100.0
91.7
97.1
84.0
100.0
00
ro
i
(*) Less than 10 mg/1.
(**)
I BOD Removal
I Volatile BOD
I Removal of Volatile BOD.
-------
APPENDIX. C
(Continued)
TABLE 3
EXPERIMENTAL DATA FOR CONTINUOUS FLOW EXPERIMENTS COMPARING AIR STRIPPING WITH BIOLOGICAL TREATMENT
FULL NUTRIENT LOADING '- FEED RATE. 118 CC/M1N; RECIRCULjiTlON RATE,' .2S~GPH
Exp.
No.
1
2
3
4
5
6
7
8
9
Tower
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
B1o
Strip
B1o
Strip
1o
trip
io
Air
Spee<
ft/ml n
200
200
200
200
200
200
400
400
400
400
400
400
560
560
560
560
560
560
Temperature
Inlet
°C
51.5
52. 0
51.4
49.0
51.6
52.0
50.9
50.0
51.2
51.0
51.3
50.0
50.2
51.5
50.5
50.0
49.5
7.0
Basin
°C
24.0
24.0
26.0
24.0
19.7
18.0
19.7
18.7
20.0
19.5
20.0
19.0
17.0
7.9
9.6
9.0
7.0
5.0
pH
Init.
9.5
9.5
9.5
9.5
9.5
9.5
9.4
9.4
9.3
9.3
9.0
9.0
9.2
9.2
9.2
9.2
8.6
8.6
Final
9.15
7.9
9.05
7.75
8.9
8.0
8.5
7.89
8.3
7.9
8.1
7.7
7.6
7.4
7.7
7.32
7.5
7.1
Cond.
Init.
umhos
660
660
650
650
563
563
810
810
780
780
870
870
910
910
800
800
850
850
Final
umhos
670
900
670
880
570
880
900
1140
840
1260
1140
1430
1060
1160
970
450
165
650
Solids
Init.
ppm
840
840
800
800
736
736
1160
1160
1204
1204
1396
1396
1200
200
144
144
996
996
Final
ppm
1072
1260
1040
1196
928
1088
1520
1588
1440
1672
1884
1968
604
736
400
784
720
068
Methanol
Content
Init.
mg/1
210
210
210
210
260
260
260
260
260
260
340
340
250
250
340
340
120
120
Final
mg/l
50
0
50
0
30
10
0
*
*
*
*
*
*
0
80
10
4
*
Total
HHi as N
Init.
mg/l
33.0
33.0
46.0
46.0
93.0
93.0
36.0
36.0
30.0
30.0
84.0
84.0
Final
mg/l
2.9
17.7
- No
— No
1.7
25.6
9.3
15.6
- No 1
- No I
5.0
2.8
5.7
2.1
No 1
No 1
7.4
3.1
Total P0«
[nit.
mg/1
13.0
13.0
Data
)ata
17.5
17.5
29.5
29.5
ata -
ata -
29.5
29.5
7.5
7.5
ta -
ta -
5.2
5.2
Final
•g/i
17.5
22.5
17.5
27.5
25.0
45.0
27.5
42.5
32.5
42.5
—
39.5
2.6
How-
down
-------
I
O3
APPENDIX C
[Continued)
TABLE 4
EXPERIMENTAL DATA FOR CONTINUOUS FLOW EXPERIMENTS COMPARING AIR STRIPPING WITH BIOLOGICAL TREATMENT
FULL NUTRIENT LOADING - FEED RATE, 470 CC/HIN; RECIROLATION RATE, 1.0 Sflt
(*)
(**)
Less tnan 10 mg/1.
I BOO Rpasval _ ,
%. Volatile BOO '
Exp.
Ho.
i
Tower
Strip
Bio
2 Strip
Slo
3
4
5
6
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
i
7 !strip
JBio
8
9
Strip
Bio
Strip
Bio
Air
Speed
ft/mi n
200
200
300
ZOO
200
200
WO
too
Temperature
Inlet'
°£
52.5
50. 3
52.8
51.0
53.6
52.5
51.1
ij.o
400 >5?.4
400 J54.0
400
400
560
560
560
560
560
560
52.4
54.0
K.1
53.5
I" '
54.5
52.0
51.0
5? 0
Basin
of.
34.0
30.0
3J.5
Z8.0
33.0
28.5
28.0
?6.0
30.0
27.9
31 0
26.0
28.0
24.0
28.0
?r).0
22.0
20.0
pH
Init.
8.94
8.94
9.9
9.9
10.35
10.35
9.0
9.0
9.0
9.0
8.78
8.78
8.6
8.6
8.6
6. 6
B.a
8.3
Final
8.8
8.1
9.7
9.3
10.09
9.99
8.8
8.7
8.7
8.4
7.95
7.9U
8.0
7.8
8.0
7.7
S.O
7.5
Cond.
Init."
umhos
1200
,1200
2660
2660
3450
3450
1120
1120
1130
mo_j
690
699
300
BOO
775
775
750
750
Final
umhos
1480
1340
2550
2460
4450
4550
1370
144Q
1320
1400
805
8-M
930
980
920
950
900
1050
Sol Ids
Init.
ppm
1728
1728
4250
4260
6084
6084
1760
1760
1772
1772
888
888
1344
1344
1052
1052
1172
1172
Final
ppm
2232
2072
4140
4C6B
7476
7040
?320
2232
2200
2252
nz4
1072
1668
1572
1520
1368
!448
T564
Methanol
Content
Init.
•9/1
320
320
320
320
360
360
780
?W1
130
130
230
230
280
280
*
*
*
*
Final
mg/1
80
100
100
100
no
no
80
100
80
SO
60
*
*
*
*
*
NHj as N
Init.
"9/1
19.0
19.0
154.0
154.0
53.0
53.0
* J25.0
* |25.0
Final
mg/1
- Mft r
- No !
- He f
-HO,
- No 1
1.4
.6
- Ho I
- to t
10.4
9.9
30.6
8.9
- No C
- No C
9.7
14.6
Total P04
tnit.
mg/1
ata -•
ata —
ata -
,t.-
3ta -•
25.3
25.3
ata -
?5.2
'5.2
21.0
21.0
ata -
23.2
23.2
Final
mg/1
----
34.8
35.2
28.7
27.5
27.7
25.0^
30.3
30.5
Blow-
down
Rate
cc/min
405
395
370
340
370
420
376
400
363
4)6
313
384
313
360
328
320
317
327
!
COO BOD
Init.
•9/1
2350
2350
5140
S140
6950
frt'tO.
2500
2500
2610
2610
1558
155S
1552
1552
1527
1527
1487
1487
Final
WJ/I
2420
2035
4360
4200
7640
74PO
2540
2610
£370
2640
1248
1119
1314
1260
1269
1180
1562
1199
InH.
»igTr
985
985
1545
1545
1440
1440_
1000
1000
1091
1091
1008
1008
673
673
792
792
697
697
Final
i»9/l
614
439
1040
950
1433
1340
592
662
635
628
567
531
390
329
452
318
441
264
I BOD
Removal
46.5
61.6
47.0
55.6
21.8
1fi.«
47.7
«3.7
55.1
«.o
62.6
57.1
61.4
62.5
60.25
72.5
57.3
73.5
I Removal!
of Vola-
tile BOD
75.75
100.4
76.6
90.7
35.5
27.6
77.6
71.2
89.7
79.8
74.6
68.0
73.9
75.2
79.6
94.3
74.6
95.6
I BOO
Removal
as MeOH
24.8
23.3
15.2
15.6
18.4
17.7
21.0
18.9
fi. 1
5.3
18.3
21.4*
39.5*
39.3*
33.5
33.5
38.1
38.1
Volatile »*rOH
son i
i IRemovai
61.4
61.4
61.4
61.4
61.4
61.4
61.4
61.4
61.4
61.4
84.0
84.0
83.2
83.2
76.8
76.8
76.8
76.8
78.5
73.7
75.4
77.4
76.0
n.-?
77.?
69.6
52.5 i
I
45.5
82.6
98.2
98.8
98.6
30.2*
31.9*
32.6*
30.4*
Removal of Volatile BOD.
-------
TABLE 5
EXPERIMEHTAL D»> FOR COHT1NUOUS ROM EXPERIMENTS •: IMPARTS A!R STRIPING WITH BIOLOGICAL TREATMENT
tj
Ho. Towe
1
2
3
4
5
6
7
8
Stri
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bio
Strip
Bin
9
Strip
Bio
ftir
Speed
TtTmi
200
200
too
400
560
560
100
200
400
400
560
560
200
200
«00
400
5«0
5(0
Feed
Rate
cc/nin
IIS
118
118
118
118
118
220
220
220
220
220
220
470
470
470
470
470
470
Reclr
Rate
~SBT
.25
.25
.25
.25
.25
.25
.SO
.SO
.50
.50
.50
.50
1.0
1.0
1.0
1.0
1.0
1.0
Tone
meT
T~
S2.0
48.0
50. S
47.0
49.4
47.0
S3.S
52. S
50. B
50.0
50.5
50.8
4.5
1.5
3.8
3.5
1.3
54.0
rature
Basil
T-
15.0
14.5
16.0
T2.0
19.0
17.0
29.0
23.5
23.0
9.0
8.5
S.O
34.0
34.0
7.5
3.0
2.0
9.0
M.
Tave
Met Bulb
— of —
55.0
55.0
54
54
59.0
59.0
60.5
60.5
57.0
57.0
lent
rature
Dry Bull
»f
6S.O
68.0
67
67
70.0
70.0
70.0
70. D
70.0
70.0
rHTT*
8.65
8.65
9.05
9.05
9.0
9.0
a. e
8.6
8.3
8.3
8.9
8.9
9.25
9.25
8.6
8.6
9.0
9.0
A
jFfnTT
8.0
7.2
8.35
8.2
9.0
8. 55
8.5
7.25
7.7
7.7
7.)
7.6
8.7
8.1
7.5
7.5
8.7
8.3
Co
TnTT
Mhos
780
780
600
600
1280
1280
880
880
520
520
520
520
690
690
520
520
895
895
nd.
TTniT
in* os
1000
1470
750
13M
1500
7000
1210
1480
625
765
675
830
730
810
650
740
070
290
Sol
ISTT
pp»
1208
1208
784
784
2268
Z2«8
1384
1384
712
712
752
752
216
216
648
648
280
280
ids
PP»
1536
2130
980
2044
2884
3572
1884
1996
900
1164
1140
1240
1300
1468
1016
932
1740
1904
Methawl
Content
'[nit fl"»l
•9/1
320
320
331
331
250
250
320
320
348
348
342
342
350
350
332
332
315
315
«9/l
95
*
61
*
28
75
95
*
69
28
58
»
25
10
56
15
#
*
NH t as If
[nit '«"->*
igTT-
19.0
19.0
24.0
24.0
18.0
18.0
9.0
9.0
26.0
26.0
16.0
6.0
4.0
4.0
9.0
9.0
6.0
6.0
•9/1
3.4
5 1
3.5
3 7
.4
.6
3.4
?,7
3.2
3.0
5.2
2.8
5.8
3,5
5.7
4.4
3.5
?.4
Total POl
In1t *4"Il
SgTT
14.7
14.7
11.5
11.5
7.0
7.0
12.5
12.5
11.5
11.5
9.7
9.7
13.2
13.2
13.0
13.0
14.2
14.2
^7T
17.2
19.7
13.7
26.7
9.8
7.0
16.7
12.2
12.5
5.0
12.7
3.1
4.2
3.4
7.1
6.7
9.2
».S
5TST-
down
cc/mn
58.0
70.0
52.2
41.0
61.3
63.4
126
120
135
lie
no
02
385
360
80
83
85
73
COD
r»9/i
2105
2105
1450
1450
—
__
2300
7VK1
--
.-
1388
1388
2238
2238
490
490
170
170
•9/1
1900
?i?n
1135
1929
—
2010
1940
—
..
997
260
885
8?9
158
153
020
W>
1 BOO ; BOD
-9/1
925
925
782
782
862
862
997
997
933
933
893
893
946
946
637
637
942
942
-ng/1
60S
364
426
468
711
673
632
353
495
493
438
419
689
580
286
???
596
507
67.6
76.6
75.9
79.2
57.0
58.0
63.7
80.6
67.44
71.65
75.47
76.24
40.33
53.03
73.8
79.4
61.7
68.8
I Remo.il 1 I BOD (Volatile: NeOH
Of VoU- l^efiova* BOD , i
89.0
100. B
90.4
94.3
92.8
94.5
81.6
103.2
74.6
79.2
83. 4
86.6
53.4
70.2
89.3
95.9
81. 2
90.6
13.7
33.0*
37.8
40.7"
26.6
26.7
25.9
30.7"
31.8
34.7
34.1
36.21,
25.4
27.3
45.6
49.3
31.9*
31.9*
— -
75
85.4
75 1 99 1
HT"
83
61
61
77
„
85
85
89
B9
69
69
83
83
75
75
91.9
99 *>
94.2
94.6
83.0
99.2
87.8
95 7
91.5
99.3
70. B
75.9 !
90.0
97.3
99.0
99.1
00
en
i
*) Lnt than 10 *g/1.
-------
TABLE 6
EXPERIMENTAL DATA FOR COMT1HUOUS FLOU EXPERIMENTS COMPARING AH STRIPPING HITH BIOLOGICAL TREATMENT
HO HUTMENTS
1 No . ' I Tower
1 Strip
Bio
2 Strip
,Bio
I 1 Strip
' JBio
1 Strip
! .Bio
S Strip
1 IBM,
i i
| 6 Strip
j Bie
7 [strip
IBIO
I
8 Strip
i I*1"
! 3 Strip
'Bio
Air
Speed
ft/mi n
200
200
400
400
560
S60
200
200
400
400
560
560
..'00
200
400
400
660
'60
Feed
Rat*
cc/«in
118
113
118
118
!I8
118
220
220
220
220
220
220
470
470
470
470
470
470
Recir.
Rate
em
.25
.25
.25
.25
.25
.25
.50
.50
.50
.50
.50
.50
1.0
1.0
1.0
!.0 j
1.0
1.0
Temperature
Inlet
oC
51.0
50.0
48.0
51.0
48.0
49.5
54.0
53.0
53.0
49.0
52.0
52.0
54.2
54.0
Basin
°C
17.0
9.0
15.0
11.0
15.0
12.0
26.0
26.0
20.0
17.0
19.0
14.0
36.0
34.0
5C.5 [27.5
50.5 '24.3
52.5 J24.0
52.0 |21.0
Ambient
Temperature
Viet Bulb
OF
52.0
52.0
55.5
55.5
59.0
59.0
57.0
57.0
59.0
59.0
59.0
59.0
55.0
55.0
62.0
62.0
53.0
58.0
Drj Sulb
°F
65.0
65.0
66.0
68.0
71.0
71.0
66.5
66.5
65.0
65.0
72.0
72.0
69.0
69.0
71.0
77.0
67.0
67.0
pH
Init.
8.6
8.6
8.8
8.8
8.9
8.9
8.5
8.5
8.33
8.38
8.4
8.4
8.9
8.9
9.0
9.0
8.8
8.8
Final
8.8
7.8
8.95
7.90
8.9
8.0
8.6
8.3
8.55
8.2
8.25
B.O
8.85
8.85
9.2
9.2
8.8
8.75
Cord.
Init.
unhos
750
750
770
770
800
BOO
795
795
810
810
620
620
980
980
1500
1500
880
B80
Final
uahos
910
1140
980
1410
1030
1600
1070
1140
1120
1240
BOO
1150
1230
1280
ZOOO
2125
1030
1170
Solids
TnU.
ppn
1168
1168
1204
1204
1228
1228
1312
1312
1128
1128
1016
1016
1504
1504
3452
2452
1268
1268
Final
ppcn
1400
1640
1564
2236
1620
2404
1676
1764
1720
1868
1272
1868
19)6
193?
3052
3316
1652
1800
Netlianol
Content
Init.
mg/1
215
216
215
215
195
195
31 5
315
285
285
285
285
291
291
235
235
245
245
Final
ing/1
15S
82
74
55
28
50
65
75
60
60
68
62
120
50
120
80
75
75
NHl as N
Init.
ng/1
5.9
5.9
7.7
7.7
7.7
7.7
8.9
B.9
9.9
9.9
7.9
7.9
1.7
1.7
10.0
10.0
9.1
9.1
Final
mg/1
1.2
2.4
1.7
2.7
0.)
1.2
0.90
1.7
.3
1.2
0.3
0.6
2.1
1.2
0.6
0.6
.3
.6
Total P0«
Init.
•9/1
0.80
0.80
*
*
1.05
1.05
*
*
*
*
*
*
0.75
0.75
0.60
0.60
0.50
0.50
Final
B9/I
1.0
.50
*
*
1.10
0.5
*
*
<*
*
*
*
0.75
0.75
1.0
1.0
0.5
0.8
Slow-
down
Rate
cc/ml n
62.0
60.0
58.6
47.0
55
55
120.0
127.5
122.8
114.2
117
no
360
375
320
312
325
305
COD
Init.
mg/1
1850
1850
1960
1960
166S
1665
1940
1940
2100
2100
1620
1620
2420
2420
3100
3100
1930
1930
Final
rng/l
1640
175O
1740
2220
1794
2415
1940
1890
2140
2200
1360
1988
2235
2285
3150
3300
1830
2060
BOD
Init.
ma/l
970
970
1200
1200
1326
1326
1112
1112
1147
1147
999
999
1H9
1149
1596
15%
1066
1066
Final
mg/1
663
526
847
862
893
945
796
780
762
797
621
738
851
862
1490
1516
755
796
i BOD
Removal
64.0
72.4
64,9
71.4
68.5
66.8
61.0
59.4
62.9
64.0
67.0
63.1
43.4
40.2
36.4
36.8
51.2
51.8
J Removal
of Vola-
tile BOD*
83. 3
94.1
84.5
92.3
89.3
87.2
73.4
71.4
75.6
,_77.0
82.0
77.2
62.5
57.9
64.8
65.5
64.8
65.6
% BOD
Rsnonal
as HeOH
13.4
17.4
14.4
15.6
13.3
12.6
24.4
23.7
21.3
21.5
24.2
24.7
16. B
21.2
WlYiTTe
BOD
75
75
76
76
76
76
77
77
70
70
78
78
58
58
9.3 50
11.1 50
17.6 i 69
17.9 : 69
leDH
Removal
62.1
60. 6
B!.<3
89. E
93.3
88.1
88.8
86.2
88.2
39.1
87.3
89.1
68.4
86.3
65.2
77.4
78. a
BO.l
I
00
Less than 0.5 mg/1
'i BOO Remowal
: Volatile BOD
•Removal of Volatile BOO.
-------
APPENDIX D
EXPERIMENTAL INVESTIGATIONS OF AIR STRIPPING
TABLE 1
Exp.
^ »
NO f
EXPERIMENTAL
Type of
Waste Water
CONDITIONS FOR BATCH STUDIES OF AIF
Volumes Inlet Water
Im't. Final Temperature
(D (D (°F)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
fc» ^
25
26
27
28
*»*^
29
30
w**
31
6th Effect
Condensate
6th Ef.
6th Ef.
6th Ef.
650 mg/1
MeOH in H20
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
6th Ef.
Macon Com-
bined Cond.
Macon Decker
Filtrate
Macon Com-
bined Cond.
Macon Decker
Filtrate
8.0
8.1
10.0
12.0
12.0
30.0
12.0
12.0
12.0
12.0
16.0
12.0
12.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
3.34
3.27
12.0
8.0
4.20
4.50
2.08
4.18
4.66
6.14
7.45
7.05
7.30
6.07
9.45
8.55
8.95
3.42
2.88
1.80
2.56
3.64
3.42
3.64
2.82
4.94
3.80
2.30
2.25
1.94
2.98
1.90
2.33
3.2
2.4
125
125
125
125
125
125
125
125
125
125
125
125
125
128
125
130
130
130
130
130
124
130
128
128
130
130
135
125
125
125
125
Air Rate
(ft3/min)
100
100
--
—
—
--
--
50
100
125
50
100
125
50
140
140
100
100
140
100
100
50
100
140
100
140
155
164
164
50
50
Recycl e
Rate
(gal/min)
1.00
1.00
__
.50
.50
.50
.50
1.00
1.00
.50
.50
.25
.25
.25
.50
.50
1.00
1.00
1.00
1.00
.25
1.00
1.00
1.00
.50
.50
1.00
1.00
1.00
.50
.50
Total
Time
(min)
38
28
48
59
52
228
30
47
23
39
90
49
35
90
60
60
36
30
30
30
90
24
30
28
52
52
23
10
10
75
60
-87-
-------
APPENDIX D
(Continued)
TABLE 2
BODs AND METHANOL DETERMINATIONS FOR BATCH STRIPPING EXPERIMENTS
Exp.
No.
1
2
3
4
5
6
7
8
Time
mlns
0
5
10
15
38
0
5
10
15
28
0
5
10
15
30
48
0
5
10
15
30
59
0
5
10
15
30
42
0
30
228
0
5
10
15
30
0
s
10
30
47
BODS
ppm
1183
..
..
482
527
1325
..
..
—
373
1124
..
..
594
..
922
1080
835
665
__
525
824
__
..
—
..
..
—
990
548
1222
1431
_.
790
--
606
1067
798
604
323
341
Methanol
ppm
580
290
160
85
--
580
285
140
57
--
600
250
100
60
35
<3S
560
410
250
__
45
<45
590
340
170
130
40
<35
645
260
< 4
690
360
200
190
40
730
560
390
50
10
:xp.
to. Time BODs Mtthanol
mlns ppm ppm
9 0 1165 470
5 529 390
10 375 150
23 247 <10
10 0 1223 780
5 870 430
10 605 180
30 433 10
39 470 0
11 0 1422 440
10 798 285
30 -- 280
45 548 220
65 - 90
90 370
12 0 1098 820
5 -- 600
10 700 440
30 -- 80
49 323 30
13 0 1098 890
5 -- 565
10 640 420
30 -- 50
35 343 30
14 0 1245
5 958
15 925
23 1030
IS 0 862 250
5 834 120
15 756 51
30 687 27
60 925 <10
16 0 933 348
5 638 128
15 437 74
30 350 <10
60 398 *10
:xp.
to. Time BOD* Methanol
mlns ppm ppm
17 0 782 331
5 520 100
15 445 <10
30 359 <10
36 386 <10
18 0 648 332
5 365 166
15 156 <10
30 244 <10
19 0 942 315
5 784 207
15 569 <10
30 529 <10
20 0 997 320
5 877 1 50
15 579 45
30 479 <10
21 0 970 215
5 745 155
15 765 110
30 588 25
45 596 35
90 637 10
22 0 1149 291
5 876 1 70
15 600 70
24 566 35
23 0 1596 235
5 1550 140
15 1274 8!,
30 1470 25
24 0 1066 ?45
5 845 70
15 670 35
28 783 <10
25 0 1147 285
5 931 130
15 670 120
30 646 65
52 686 2S !
txp.
to.
26
27
28
29
30
31
Time
mlns
0
5
15
30
52
0
5
15
23
0
5
10
0
5
10
0
5
10
15
30
75
0
5
10
15
30
60
BODs Methanni
ppm
999
916
638
627
753
1245
958
925
1030
998
365
234
612
378
405
1167
956
944
773
595
513
863
747
741
743
450
545
ppm
285
105
45
25
__
--
440
120
70
140
50
35
320
250
205
165
70
0
120
75
50
40
0
0
-88-
-------
00
10
I
APPENDIX D
(Continued)
TABLE 3
EXPERIMENTAL DATA FOR CONTINUOUS FLOW EXPERIMENTS* INVESTIGATING EFFECTS OF
WATER TEMPERATURE ON AIR STRIPPING
Exp.
No.
1
2
3
4
5
6
7
8
Water
Temperature
Inlet
OF
53
40
42
53
42
52
52
40
Basin
OF
25
26
28
31
35
40
30
30
Air
Flow
cfm
125
125
75
75
50
50
100
100
Feed
Rate
cc/min
470
470
470
470
470
470
470
470
Bl owdown
Rate
cc/min
307
400
393
327
413
369
316
377
Air Temperature
Dry Bulb
OF
67.5
67.5
67.5
67.5
79.5
80.5
84.5
85.5
Met Bulb
OF
65.5
65.5
65.5
65.5
73.0
73.0
74.0
75.0
Wastewater pH
Feed
9.9
9.9
9.0
9.8
9.3
9.3
9.4
9.4
Slowdown
8.9
8.1
8.5
8.3
9.2
9.2
9.2
9.2
BOD
Fresh
Feed
mg/1
1410
1410
1343
1305
950
1010
890
890
Bl owdown
mg/1
763
750
835
850
430
415
280
300
COD
Fresh
Feed
mg/1
2993
2993
2908
2993
2167
2276
2276
2276
Slowdown
mg/1
2400
2324
2429
2372
1701
1636
1517
1517
*Liquid Recirculation Rate = 1.0 GPM.
-------
APPENDIX E
INVESTIGATION OF FOAMING IN LABORATORY COOLING TOWER
Foaming was observed to occur at various times during the operation of
the Rome laboratory tower. This foaming occurred mainly in the packed
section of the tower and occasionally in the basin. Since a full-scale
tower would process large quantities of liquid and would use consider-
able amounts of air, an intolerable nuisance condition would likely
result. A recent report(*) concerning the use of cooling towers in
black liquor evaporation and the foaming problems encountered prompted
a study of foaming. The report stated that foaming could be con-
trolled by maintaining a continuous blowdown and thereby controlling
the solids concentration at 250 ppm. Further information contained in
the report revealed that before the tower was installed the pH of the
water was approximately 6.0 with solids concentration averaging 50 ppm;
and after installation of the tower the pH of the recycled liquid in-
creased to approximately 7.5.
Preliminary studies were undertaken to attempt to correlate solids con-
centration and pH with foaming. A quantitative measure was devised for
the foaming in the packed section. Visual observation of the action
of liquid and air in the tower made direct assessment of the foam level
possible. Foaming originated at packing support points within the
tower and could conveniently be measured by counting the number of pack-
ing grid heights (one grid height = 1.25 inches) above the support
locations upon which foaming was taking place.
An experiment with the laboratory cooling tower was performed whereby
the solids concentration was allowed to increase from 314 to 1570 ppm.
Although the solids content increased five fold, no foaming increase
was noted. Many experiments of shorter duration, where the concentra-
tion doubled, showed the same negative effect on increased foaming.
The most important variable related to foaming tendency in the tower
was pH. Data collected on 84 observations of foam height under varied
experimental conditions were fairly well correlated with liquid pH.
Figure E-l shows the foaming tendency measured as foam height versus
wastewater pH. This figure shows that the nearer the pH is to a
neutral value, the more likely a foaming condition is to occur. The
normal pH of the condensate feed was 9.5 and was typically reduced to
9.0 upon recirculation.
(*) Cohn, R.G. and Tonn, E.T., "Use of a Cooling Tower in Black
Liquor Evaporation", TAPPI 47, No. 3, pp 163A-165A (1965).
-90-
-------
" 20
15
o>
CJ
O)
e
-o
c
O)
10
£ 5
S-
O)
10
•o 0
No Data
No Data
84 Observations
1 grid height = 1.25 inches
No Data
5.0
4.0
5.0
6.0 7.0 8.0
Waste Water pH
9.0
10.0
FIGURE E-l: FOAMING TENDENCY AS A FUNCTION OF WASTE WATER pH
-------
Other experimental observations noted that increasing air flows and
decreasing temperature tended to promote foaming. Decker filtrate
produced considerably more foam than the condensate streams, although
the foaming tendency of the latter stream varied considerably among
samples.
-92-
-------
APPENDIX F
EXPERIMENTAL DATA FOR FULL-SCALE COOLING TOWER
TABLE 1
BOD DATA FOR FULL-SCALE COOLING TOWER OPERATION
BOD (mg/1)
To
Period Date Tower
1 May 14, 1969
15
16
17
18
2 May 19, 1969
20
21
22
23
24
25
3 May 26, 1969
27
28
29
30
31
June 1
4 June 2, 1969
3
4
5
6
7
5 June 9, 1969
11
12
14
15
16
17
18
19
22
(*) BOD (mg/1) for 6th
357
490
540
605
672
598
445
478
413
450
470
463
648
577
622
685
785
657
513
342
573
623
510
602
458
533
695
808
655
607
806
950
775
822
688
From
Tower
310
410
467
568
583
507
368
368
297
368
345
395
648
571
510
593
658
598
480
287
557
565
440
440
458
435
595
735
593
482
690
805
687
770
612
effect cal
Decker Combined 6th* Turpentine
Filtrate Condensate Effect Underflow
--
--
--
—
__
--
--
--
--
--
--
585
475
495
553
647
565
397
212
585
440
383
365
462
450
525
700
583
478
826
910
890
862
595
culated
665
618
445
730
853
_ _
--
--
--
_-
—
--
866
782
729
794
843
994
905
638
776
812
725
625
797
_ —
--
--
—
—
--
__
—
-_
from #BOD
970
2563
2477
1649
3066
3567
3149
4312
4474
3286
4794
2652
321
683
4387
3881
5905
1799
876
2294
594
3312
2246
6299
168
4176
5386
4353
2916
5720
4132
5574
2671
2080
3678
and flow in mgd.
-93-
-------
Period Date
5 July 4, 1969
(Cont.) 5
6
7
8
9
10
11
12
15
16
17
18
6 July 19, 1969
20
21
25
28
29
30
31
Aug. 2, 1969
3
5
6
8
11
12
13
14
15
16
17
7 Aug. 22, 1969
23
24
25
26
27
To
Tower
790
675
552
620
527
465
610
667
672
1095
1252
1103
922
962
713
777
593
745
1020
1207
1128
1017
970
712
793
1355
728
750
1042
1045
953
970
823
818
815
863
780
721
697
(
From
Tower
770
634
547
590
438
390
582
643
622
1075
1252
1067
842
947
658
680
515
735
987
1008
1042
890
873
623
702
1301
678
728
832
873
777
953
805
750
783
750
617
626
613
TABLE 1
Continuec
B(
Decker
Filtrate
807
690
572
743
418
445
625
625
685
1107
1270
1128
875
888
638
675
533
670
1012
1012
1012
1140
1018
665
1048
1283
758
738
958
910
933
1018
860
825
753
728
613
669
632
i)
)D (mq/1)
Combined 6th Turpentine
Condensate Effect Underflow
1276
1852
1061
803
4069
2961
1289
2169
2749
728
1139
2189
3914
1894
2797
4313
3629
1572
2100
9143
4688
5525
5131
3961
3935
3784
2247
1548
8257
7535
7615
1522
1999
2625
2041
4897
6622
4089
3709
-94-
-------
TABLE 1
(Continued)
BOD (mq/1)
To From Decker Combined 6th Turpentine
Period Date Tower Tower Filtrate Condensate Effect Underflow
7 Aug. 28, 1969
(Cont.) 29
30
Sept. 3
6
7
8 Sept. 9, 1969
10
11
12
13
14
15
16
9 Sept. 18, 1969
19
20
21
22
23
24
25
26
27
28
29
Oct. 1, 1969
2
10 Oct. 5, 1969
6
7
8
11 Oct. 10
11
12
15
16
998
1110
1041
682
977
985
740
820
738
932
935
795
745
755
981
802
620
823
790
981
1043
1315
1377
990
803
882
1047
1215
673
767
800
1017
1030
832
863
820
1177
865
1052
942
618
870
868
666
693
653
848
815
745
715
702
927
627
582
718
750
898
857
1260
1295
907
730
810
927
1208
627
687
765
923
982
800
797
809
1103
1010
1113
1028
745
965
945
845
887
892
1223
1067
962
797
1070
1125
727
772
837
825
1066
1155
1288
1347
972
740
908
1318
1317
797
928
1120
1392
1072
915
857
613
1082
--
--
--
--
—
626
672
672
712
806
790
732
630
__
—
—
--
--
--
--
--
--
--
--
--
--
--
650
650
620
699
— _
--
__
—
__
4761
2847
4192
2654
4401
4854
2855
4518
2663
2554
4204
1383
1443
1335
1948
5044
280
3337
1679
3094
6313
3064
3917
3440
3216
2604
1638
804
1603
2939
1133
3432
2413
1432
3134
2079
3659
-95-
-------
Period
11 Oct.
(Cont.)
Nov.
12 Nov.
Dec.
13 Feb.
Date
17, 1969
18
20
21
26
27
28
1
2
3
5
6
7
8
9
10
11
12
13
18
24
25
26
29
30
10
12
14
15
16
28
29
30
9, 1970
13
17
18
19
21
22
23
24
25
To
Tower
1230
1042
912
952
1462
1237
903
1417
1465
985
882
718
822
848
987
1043
987
1032
978
1335
920
1042
818
753
980
1137
1040
828
1147
1257
673
703
558
1155
1167
987
1072
1315
--
--
1157
1165
1150
('
TABLE 1
Continued)
BC
1
ID (mg/1)
From Decker combined 6th Turpentine
Tower Filtrate Condensate Effect Underflow
1083
1002
870
840
1395
1173
832
1405
1403
918
695
687
718
768
838
955
863
955
935
1265
917
1037
773
728
930
1127
1023
813
1125
1250
625
607
448
1093
1067
923
965
1285
—
--
1120
1108
1102
1053
935
1010
1120
1383
1195
833
1492
1463
1282
922
837
723
938
1005
992
887
1022
1108
1673
862
1035
937
772
935
1265
978
795
1078
1267
770
713
595
_—
__
--
937
1190
--
--
993
1140
1018
6293
2986
2899
4854
4168
3586
3534
1946
3295
1128
6152
524
602
2821
5657
3947
5695
4004
1877
4099
1508
1233
1651
1469
2801
755
1946
1466
2282
— _ --
2544
3728
3627
_ _ _
_ _
__
__
_- __
__ __
__
__
__
--
6467
7733
7700
9130
8650
7430
8666
-96-
-------
APPENDIX F
(Continued)
TABLE 2
METHANOL DATA FOR FULL-SCALE COOLING TOWER OPERATION
Methanol (mg/1)
To From
Decker
ppriod Date Tower Tower Filtrate
1
2
3
4
5
May 17, 1969
18
May 19, 1969
20
21
22
23
24
25
May 26, 1969
27
June 3, 1969
4
5
June 9, 1969
10
n
12
14
15
16
17
18
19
22
July 4
5
6
7
8
10
n
12
15
16
17
18
150
150
no
100
100
120
100
120
140
120
130
175
180
180
53
100
100
75
94
95
86
no
122
73
88
160
120
105
90
105
120
105
105
170
60
125
140
80
125
80
85
80
60
60
60
60
65
90
125
100
140
89
80
68
55
70
42
56
90
78
47
57
85
80
75
70
85
85
80
70
105
60
100
90
— f —
--
__
—
—
—
—
--
—
70
80
70
100
100
50
70
44
55
48
42
90
120
94
90
74
110
90
85
80
85
100
95
no
90
20
140
70
Comb. 6th Turpentine Non-
Cond. Effect Underflow Cond,
230
270
—
—
--
--
__
--
--
220
260
350
340
345
__
—
—
—
--
—
--
--
--
--
—
—
—
__
--
--
--
--
__
__
-_
—
--
-97-
-------
lo
Period Date Tower
6 July 21, 1969
25
29
30
31
Aug. 2
3
6
7
8
11
12
13
14
15
16
17
7 Aug. 22, 1969
23
29
30
Sept. 3
8 Sept. 10, 1969
11
12
13
14
15
16
9 Sept. 22, 1969
23
25
Oct. 2, 1969
13 Feb. 10, 1970
10
10
11
11
n
12
12
12
140
100
120
130
130
100
100
105
150
138
115
135
142
163
118
96
122
195
195
130
108
105
75
110
84
130
100
100
115
110
140
130
120
315
310
320
190
44
94
210
140
240
(C
From
Tower
80
85
85
no
90
85
70
67
92
98
96
97
105
115
86
88
95
129
155
82
86
62
34
84
80
84
85
39
74
110
105
__
80
190
140
200
47
40
34
195
125
225
TABLE 2
ontinue
d)
Methanol (mg/1)
Decker Comb. 6th lurpentlne Non-
Filtrate Cond. Effect Underflow Cond.
100
110
no
120
80
100
no
170
175
130
115
160
122
136
130
125
130
90
155
75
138
90
59
76
110
78
80
90
91
150
100
160
155
190
215
205
175
195
39
225
170
—
_ —
—
--
—
--
—
—
-_
--
--
--
--
--
—
—
—
—
--
—
—
--
--
140
230
220
210
220
190
210
270 2185
430 2037
--
--
275 -- 3250
335 — 6000
320 — 4250
500 — 6500
365 -- 4500
78 — 2575
415 -- 3625
--
--
180
160
240
270
215
92
155
165
-98-
-------
APPENDIX
F
(Continued)
TABLE 3
FLOW RATES FOR FULL-SCALE COOLING TOWER OPERATION
Period Date
1 May 14, 1969
15
16
17
18
2 May 19, 1969
20
21
22
23
24
25
3 May 26, 1969
27
28
29
30
31
June 1
4 June 2, 1969
3
4
5
6
7
5 June 9, 1969
10
11
12
14
15
16
17
18
19
22
To From
Tower Tower
6.84 6.37
9.36 8.65
10.05 9.29
10.51 9.72
9.68 8.89
9.64 9.37
10.33 10.01
10.34 10.02
10.40 10.09
10.29 9.99
10.33 10.05
9.76 9.49
9.99 7.96
11.80 8.87
11.66 8.69
11.81 8.87
11.78 8.54
11.61 8.73
11.39 8.54
11.79 9.31
11.00 8.48
11.58 8.96
11.85 9.23
10.47 8.62
12.48 9.83
11.12 9.09
11.18 9.03
12.24 10.19
12.50 10.43
10.63 8.79
12.10 9.85
10.67 8.63
10.83 8.77
9.94 8.18
10.44 8.40
11.21 9.02
Sewer
.44
.41
.41
.41
.44
.30
.33
.31
.36
.35
.33
.31
1.68
2.52
2.55
2.54
2.52
2.48
2.43
2.10
2.12
2.18
2.13
1.48
2.19
1.79
1.78
1.66
1.65
1.77
1.81
1.60
1.59
1.36
1.59
1.69
Flow
Decker
Filtrate
_.
--
--
__
—
__
--
—
--
--
--
—
1.02
1.62
1.68
1.68
1.68
1.68
1.68
1.68
1.68
1.68
1.68
1.11
1.68
1.68
1.68
1.68
1.68
1.68
1.96
1.75
1.77
1.47
1.76
1.89
(mgd)
Comb.
Cond.
.28
.48
.50
.50
.51
__
__
—
--
--
-_
—
.72
.99
.98
.98
.98
.95
.91
.53
.56
.56
.54
.44
.52
__
--
-_
--
--
--
__
__
__
__
--
6th Turpentine
Effect Underflow
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
-99-
-------
Period Date
5 July 3, 1969
4
5
6
7
8
9
10
11
12
15
16
17
18
6 July 19, 1969
20
21
25
28
29
30
31
Aug. 2
3
5
6
7
8
11
12
13
14
15
16
17
7 Aug. 22, 1969
23
24
25
26
27
To
Tower
9
10
10
11
11
11
11
10
10
10
12
12
11
11
11
10
10
12
14
11
11
11
12
12
12
11
11
12
11
11
11
11
12
12
12
10
9
10
10
11
11
.38
.33
.90
.19
.48
.44
.27
.00
.89
.67
.86
.26
.95
.86
.74
.67
.79
.19
.44
.23
.90
.70
.11
.55
.46
.95
.65
.87
.73
.59
.18
.59
.46
.03
.52
.58
.98
.35
.65
.30
.03
(c
From
Tower
7.
8.
8.
9.
9.
9.
9.
8.
8.
8.
10.
10.
9.
9.
9.
8.
8.
9.
9.
9.
9.
9.
9.
8.
8.
8.
9.
11.
9.
9.
9.
9.
10.
9.
9.
8.
8.
8.
8.
9.
9.
57
47
97
27
35
36
18
00
80
33
69
29
98
88
75
64
75
28
16
11
72
64
37
90
84
79
32
03
85
71
36
78
21
99
99
85
12
46
80
43
30
TABLE 3
ontinue
Sewer F
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.69
.26
.29
.29
.57
.56
.59
.54
.56
.84
.50
.33
.39
.34
.34
.42
.41
.23
.47
.76
.34
.34
.08
.97
.96
.52
.96
.47
.55
.51
.45
.45
.88
.67
.88
.44
.49
.52
.49
.51
.36
d)
Dec
ilt
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
1
1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
ker Comb.
rate Cond.
.68
.58
.64
.63
.84
.79
.80
.71
.79
.74
.87
.68
.69
.65
.70
.74
.75
.61
.00
.83
.90
.77
.50
.76
.33
.69
.66
.55
.63
.59
.53
.53
.61
.55
.96
.52
.57
.61
.57
.58
.44
6th Turpentine
Effect Underflow
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
-100-
-------
Period Date
7 Aug. 28, 1969
(Cont.) 29
30
Sept. 3
6
7
8 Sept. 9, 1969
>0
11
12
13
14
15
16
9 Sept. 18, 1969
19
20
21
22
23
24
25
26
27
28
29
Oct. 1
2
10 Oct. 5, 1969
6
7
8
11 Oct. 10, 1969
11
12
15
16
17
18
To
Tower
10.
10.
10.
12.
10.
11.
11.
10.
11.
11.
11.
10.
11.
12.
11.
8.
10.
9.
9.
10.
10.
10.
10.
10.
10.
9.
9.
10.
11.
11.
12.
12.
11.
11.
11.
9.
9.
8.
8.
23
89
81
78
87
15
61
75
17
59
09
94
58
57
35
13
45
93
74
62
91
45
41
16
14
63
01
47
94
94
95
91
72
91
82
40
26
11
51
(
TABLE 3
Continued)
Flow (mgd)
From Decker
Tower Sewer Filtrate
8.
8.
9.
10.
9.
9.
8.
8.
8.
8.
8.
8.
8.
9.
9.
6.
6.
7.
7.
8.
8.
8.
8.
8.
8.
7.
6.
8.
9.
9.
10.
10.
9.
9.
9.
7.
7.
7.
7.
36
82
06
70
14
37
86
01
41
79
37
18
86
81
47
49
58
40
47
33
49
41
38
23
09
59
49
46
22
22
23
06
81
65
85
75
89
17
49
1
1
1
1
1
1
2
2
2
2
2
2
2
2
1
1
3
2
1
1
2
1
1
1
1
1
2
1
2
2
2
2
1
1
1
1
.50
.70
.38
.71
.37
.41
.31
.38
.40
.43
.35
.39
.36
.39
.37
.22
.42
.05
.78
.79
.06
.50
.50
.54
.51
.49
.03
.50
.24
.25
.26
.32
.39
.75
.45
.21
.94
.56
.65
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
.58
.78
.46
.78
.45
.49
.46
.43
.45
.47
.41
.45
.45
.50
.59
.29
.85
.82
.74
.75
.71
.75
.75
.64
.76
.75
.23
.72
.42
.41
.40
.53
.62
.71
.68
.36
.08
.85
.82
Comb. 6th Turpentine
Cond. Effect Underflow
__
__
__
__
—
1.01
1.03
1.03
1.03
1.02
1.02
.98
.97
_ _
—
—
__
__
—
—
__
__
_-
-_
—
—
—
1.01
1.03
1.04
1.53
— fm
__
__
__
—
--
--
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
-101-
-------
TABLE 3
(Continued)
Period Date
11 Oct. 20, 1969
(Cont.) 21
26
27
28
Nov. 1
2
3
5
12 Nov. 6, 1969
7
8
9
10
11
12
13
18
24
25
26
29
30
Dec. 10
12
14
15
16
28
29
30
31
13 Feb. 9, 1970
10
13
17
18
19
21
22
23
24
25
To
Tower
9.
10.
11.
11.
11.
11.
10.
10.
11.
11.
11.
11.
11.
10.
11.
11.
10.
11.
12.
11.
10.
10.
11.
11.
10.
10.
11.
11.
11.
11.
10.
10.
10.
10.
10.
11.
11.
11.
10.
11.
11.
11.
10.
82
38
73
56
09
81
63
76
38
59
76
73
39
61
02
29
31
79
40
21
51
52
06
41
82
49
32
62
24
53
88
69
97
15
82
49
82
62
77
67
91
85
83
From
Tower
8.
8.
9.
9.
9.
9.
8.
7.
7.
8.
8.
8.
9.
8.
8.
8.
7.
10.
9.
9.
8.
9.
9.
9.
8.
8.
9.
9.
8.
9.
8.
8.
9.
9.
8.
9.
9.
9.
9.
9.
9.
9.
8.
48
71
72
51
09
44
55
58
96
29
64
62
32
59
23
56
75
04
64
51
90
10
12
53
63
33
15
42
99
35
73
14
14
05
91
48
88
65
06
99
87
89
80
Flow (mad)
Decker Comb.
Sewer Filtrate Cond.
1.'
1.
1.
1.
1.
1.
2.
2.
3.
2.
2.
1.
1.
2.
2.
2.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
2.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
96
23
54
51
54
88
58
70
85
02
52
60
55
53
34
16
05
07
56
52
60
52
43
64
63
55
57
61
93
87
86
14
55
12
65
72
68
68
67
63
65
68
66
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
.90
.84
.72
.70
.71
.80
.79
.02
.86
.92
.98
.82
.79
.73
.39
.63
.54
.11
.45
.41
.32
.13
.65
.59
.91
.87
.89
.92
.96
.89
.87
.88
__ ~_
__
__
—
__
__
__
—
__
—
—
6th Turpentine
Effect Underflow
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.03
.02
.02
.03
.03
.03
.03
.03
.03
.03
.03
-102-
-------
APPENDIX F
(Continued)
TABLE 4
TEMPERATURE DATA FOR FULL-SCALE COOLING TOWER OPERATION
Temperature (OF)
Period Date
1 May 14, 1969
15
16
17
18
2 May 19, 1969
20
21
22
23
24
25
3 May 26, 1969
27
28
29
30
31
June 1
4 June 2, 1969
3
4
5
6
7
5 June 9, 1969
10
11
12
14
15
16
17
18
19
To Tower
107.3
113.9
115.5
114.4
115.9
113.6
112.5
112.1
112.0
112.3
111.5
110.8
120.2
119.8
119.7
120.2
120.1
119.3
119.0
116.0
115.8
116.1
114.8
113.4
116.1
114.5
117.0
116.7
117.1
120.2
117.5
119.9
120.9
119.8
120.4
From Tower
78.1
79.7
82.3
82.9
82.5
82.8
83.9
83.3
83.6
84.0
83.9
82.3
85.3
86.5
86.1
87.6
87.3
87.3
88.5
86.5
82.3
84.4
86.3
84.9
88.5
87.4
88.3
89.9
90.6
90.4
90.5
88.3
90.3
87.3
88.1
Wet Bulb
61.5
62.5
65.9
67.4
66.6
69.1
69.7
69.2
69.1
67.7
69.2
67.7
68.3
66.9
65.4
67.9
68.0
69.5
70.5
67.7
58.2
61.0
66.2
67.1
70.4
71.5
70.0
71.8
72.9
74.8
73.3
71.1
74.5
72.2
72.3
-103-
-------
Period Date
5 June 22
(Cont.) July 3
4
5
6
7
8
9
10
11
12
15
16
17
18
6 July 19, 1969
20
21
25
28
29
30
31
Aug. 2
3
5
6
7
8
11
12
13
14
15
16
17
7 Aug. 22, 1969
23
24
25
TABLE 4
(Continued)
To Tower
120.2
124.5
123,0
121.0
119.5
117.0
117.5
118.5
121.5
121.0
121.0
118.5
118.0
118.5
120.0
120.0
119.0
119.0
116.5
119.2
119.4
118.9
119.6
118.2
115.4
107.2
110.6
119.1
117.2
118.2
118.6
118.9
119.3
118.4
119.5
119.2
119.4
118.5
117.8
117.9
Temperature (°F)
From Tower
90.2
89.5
90.0
90.0
91.0
90.0
90.0
90.5
89.5
91.0
90.5
91.0
89.5
89.0
90.0
90.5
88.5
89.0
89.5
89.2
88.4
88.4
89.1
89.8
88.8
86.5
87.5
90.5
91.3
90.0
89.0
88.2
89.9
90.9
91.4
92.1
87.1
84.3
85.1
86.9
Wet BulF
74.7
76.0
76.5
76.0
78.0
77.0
76.5
77.5
76.5
78.0
77.5
76.0
74.0
74.0
76.0
76.0
74.0
74.0
75.0
74.5
73.4
72.6
74.2
74.4
71.9
70.2
72.6
73.8
74.5
73.0
71.7
70.9
73.5
74.1
75.7
76.1
68.0
64.9
65.5
67.6
-104-
-------
Period Date
7 Aug. 26, 1969
(Cont.) 27
28
29
Sept. 3
6
7
8
8 Sept. 9, 1969
10
11
12
13
14
15
16
9 Sept.18, 1969
19
20
21
22
23
24
25
26
27
28
29
Oct. 1
2
10 Oct. 5, 1969
6
7
8
11 Oct. 10, 1969
11
12
13
15
TABLE 4
(Continued)
To Tower
118.7
117.1
113.1
118.3
117.5
120.7
120.7
123.0
122.2
122.3
121.7
121.7
122.0
121.9
123.1
123.1
120.6
116.9
108.8
115.3
115.9
118.1
117.5
118.3
117.6
118.7
119.2
117.4
113.5
116.8
119.5
119.0
120.5
121.2
116.5
116.3
117.3
119.5
115.0
Temperature (°F)
From Tower
88.4
88.4
83.6
86.7
91.0
90.2
90.5
90.9
88.1
86.9
86.6
86.8
86.9
86.9
91.3
93.0
91.7
83.6
83.1
84.4
84.4
87.9
87.9
86.1
85.4
86.4
87.1
82.5
85.6
86.3
87.8
88.6
92.0
91.4
88.4
87.8
88.4
89.7
83.3
Wet Bulb
70.2
72.4
66.6
68.0
72.2
73.0
73.0
73.1
64.5
61.5
62.3
60.5
62.8
62.8
71.2
73.0
69.8
67.3
63.8
64.7
64.8
69.1
68.2
65.3
64.0
66.1
67.4
58.0
70.4
66.0
62.8
63.3
69.3
68.6
66.2
66.6
66.5
68.1
62.5
-105-
-------
Period Date
11 Oct. 16
(Cont.) 17
18
20
21
26
27
28
Nov. 1
2
3
5
12 Nov. 6, 1969
7
8
11
12
13
18
24
25
26
29
30
Dec. 10
12
14
15
16
28
29
30
31
TABLE 4
(Continued)
To Tower
114.5
113.4
113.8
118.4
115.6
111.6
111.5
111.1
112.5
111.8
105.0
101.2
101.1
103.7
103.9
104.8
107.5
107.4
113.6
110.9
111.5
109.5
105.9
106.4
105.4
107.6
109.2
105.0
105.4
105.5
108.7
109.1
104.4
Temperature (°F)
From Tower
81.9
75.8
79.5
89.3
86.9
85.4
84.6
84.0
87.8
83.3
78.8
76.4
78.6
80.1
80.0
81.3
81.7
77.9
87.3
83.3
83.7
80.7
76.8
78.4
80.2
79.2
79.3
77.1
78.0
78.9
84.0
84.7
77.1
Wet Bulb
60.5
51.8
57.0
72.4
66.3
60.0
59.3
52.7
65.3
58.9
50.4
44.7
50.0
52.6
50.3
60.0
54.0
47.4
63.7
59.5
59.3
53.9
49.5
41.2
48.7
41.4
45.4
41.2
40.1
44.1
58.5
60.8
43.3
-106-
-------
SYMPOSIUM
SELECTED PAPERS, PART I
32d
MEASUREMENT OF THE RELATIVE VOLATILIZATION
RATES OF THE WATER-MISCIBLE FRACTIONS IN AN
AQUEOUS EFFLUENT
L. J. Thibodoaux, University of Arkansas
Foy«tf»vi//«, Arkansas
R. B. Estridg* and B. G. Turner, Georgia Kraft Company
Rom*, Georgia
SIXTY-NINTH NATIONAL MEETING
Cincinnati, Ohio
May 16-19, 1971
AMERICAN INSTITUTE OF CHEMICAL ENGINEERS
345 East 47 Street, New York, New York 10017
-107-
-------
Introduction
Aqueous effluents from industrial sources contain a broad spectrum of
pollutant constituents, and there are several standard measuring techniques
by which the wastewater can be analyzed and categorized. The dissolved
constituents in wastewater can be of varied nature and new measurement
techniques are continually needed to provide effective assessment of complex
treatment processes. Of the total portion of dissolved components in a
wastewater, a further sub-classification into a portion which is volatile
and a non-volatile portion is needed. The volatile portion of the dissolved
constituents in an aqueous effluent is that portion which can be transferred
to the air sphere by the mere contact of the phases, i.e., contact of the
aqueous phase and gas (air) phase. Interphase mass-transfer of dissolved
constituents will occur when the partial pressure of a constituent in the
gas phase is less than the equilibrium partial pressure of .the constituent
in the aqueous phase.
Air-water contact operations on both domestic (municipal) and industrial
aqueous effluents is common practice, as a treatment operation in itself
or as a means of obtaining molecular oxygen. Aqueous phase biochemical
oxidation of the dissolved organic constituents is a universally employed
treatment process which can be performed most economically by employing
intimate phase contact with air as the source of molecular oxygen (Eckenfelder,
1953). The auto-oxidation of hydrocarbons by molecular oxygen is a recent
innovation which employs air-water contact ( Prather, 1970 ). The use of
large volumes of air in a deliberate attempt to take advantage of interphase
mass-transfer phenomena for the removal of dissolved volatile constituents
from an aqueous effluent is currently receiving much attention. The removal
of dissolved organics from industrial wastewater (Cohn and Tonn, 1967; Mohler,
Elkin and Kummick, 1964; Smathers and Frady, 1969; Estridge, Turner, Smathers,
-108-
-------
and Thibodeaux, 1970; Burns and Eckenfelder, 1965), and the removal of ammonia
from domestic sewage (Slechta and Gulp, 1967) account for bulk of air-water
operations other than oxidation. Although these processes involving air-
water contact are in current use, there remains a need for a method of evaluating
what fraction of the dissolved constituents is amenable to interphase mass
transfer and the rate at which this transfer can be undertaken.
The ever-growing role of organics in the air sphere and their involvement
with photochemical smog formation will necessitate that all sources be pin-
pointed and examined as possible contributors. The increase in the number
of air-water contact operations which employ large volumes of air, in a once
through operation, with any of several wastewater treatment techniques, and the
current increase in wastewater treatment activities may usher in another
important source of air pollution. The gross effect is solving, or partially
solving, a water pollution problem and inadvertantly creating an air pollution
problem. There is a need for a method of assaying whether or not an aqueous
effluent is a potential air pollution source if the effluent is employed in
one of many air-water contact treatment schemes.
-109-
-------
Volatilization Measurement Devices
Interphase mass-transfer devices are employed on a large scale in the
chemical process industries. The theory and design procedures for transfer
of known components in a continuous process are well established (Treybal,
1968). These devices consist of vertical towers, either cylinderical or
rectangular in shape, packed with appropriate material that provide large
interfacial area for enhancement of interphase mass-transfer and are usually
operated in a countercurrent flow arrangement. The removal of volatile
constituents from an aqueous phase is commonly referred to as "stripping."
A device incorporating the essential features of the above mentioned full
scale devices is necessary for obtaining relative volatilization rates
of dissolved volatile components.
On the surface it seems that one could air sparge a given amount of
wastewater in a laboratory graduate and obtain the necessary information.
Although the total volatile fraction could be easily established by this
simple apparatus information concerning the volatilization rates is usually
masked by the transient behavior of this batch process with respect to the
interfacial area for mass transfer. Figure No. 1 reveals the nature of the
problem for simple batch sparging. The transient nature of this operation
is due to the simultaneous vaporization of water, the most abundant volatile
component, and hence a reduction in interfacial area due to a reduction in
total volume. Foaming frequently occurs when air is introduced into wastewater
so that it becomes the dispersed phase.
The device shown in Figure No. 2 is capable of producing experimental
data on the result of air-water contact operations involving almost any
aqueous effluent. This apparatus arrangement will allow one to evaluate the
total fraction of the dissolved constituents which are volatile, plus it will
allow him to evaluate the rate of removal of any portion of this volatile
-no-
-------
Laboratory
A1r
l}
i /, . .
o o o
0 0
Ooo
o
o
o
Q
-^ O
^B O *v
o o o
Laboratory
Ai r
Liquid Level at Start
Liquid Level at Later
Time
Zone for Interphase Mass-Transfer
is Non-Constant in a Graduated
,Cyl i nder ,
it
| 1
•
/
(
O o O
o o
o o
o o
°o °
.00°
h Oo
OOO
\
FIGURE NO. 1: SIMPLE BATCH AIR SPARGING
Air Out
Wastewater
Recirculated
Lab
Pump
>
'v
V,
lf
00
o
o
o
O 0
o
°- p.
\
?
t
•"5 '
1
* — c
p
r
^•^mi^—Bm*
/?
s.
/
^
1
•^ —
Contacting Zone Glass Column
Packed with Glass Shot
Laboratory Air
Beaker with Wastewater
FIGURE NO. 2: BATCH, RECIRCULATING AIR-WATER CONTACTOR
-111-
-------
fraction. This device also employs a fixed quantity of wastewater but is
independent of the quantity remaining since the mass-transfer operation is
performed continuously in a separate device by the use of a recirculating
pump and a small section of glass tubing packed with an appropriate material
(i.e., glass beads). This arrangement maintains a constant amount of inter-
facial area and also forces the air to become a continuous phase thus
eliminating foaming problems.
-112-
-------
Continuous Volatilization of Single Components
In order for the proposed volatilization measuring device to be of
general utility a straight forward method of analyzing the experimental
results is required. Due to the transient nature of the air-volatilization
operation and the usual complexities associated with time varying systems
a simple method of analysis seemed unlikely. However, a simplistic linearized
view of the overall process coupled with a simple mathematical analysis
yielded an equation from which the air volatilization rate could be determined
accurately and quickly.
The schematic of the air-volatilization operation shown in Figure No. 3
is helpful in visualizing the simple model used in the mathematical analysis.
This model assumes that for a constant air rate, air wet bulb temperature,
water recycle rate and water temperature that the amount of a volatile component
removed is a constant multiple of the quantity of this component entering
the tower. This constant is denoted by K. and is defined as the specific
volatilization rate for component i in the column:
Ki = (Lmixii * Lmo *io}/ Lmixii (1>
where L . and L are the volumetric flows (volume/time .area) to and from the
mi mo
column,
x.. and x. are the concentrations of component i entering and leaving the
packed column.
Definition of the remaining quantities in the schematic are:
a - specific volatilization rate for water in the column. This quantity
is constant once the aqueous phase has cooled to the air wet bulb
temperature.
M - volumetric quantity of liquid in the device. The initial quantity
charged is M .
x.x - concentration of water in the recycled liquid and the column outlet.
i ,o
t - experiment run time.
The relative volatilization rate is defined as the ratio of the specific
-113-
-------
Air Out Plus
Recycle
L x
mi1 11
i
1
Volatized Frac
H?h' 3 Lm
w/,
'Packed '
/ Column .
\
1 m
Catch'
Basin M
^ — Air In
} ' 10
FIGURE NO. 3: SCHEMATIC OF VOLATILIZATION DEVICE
-114-
-------
volatilization rate of component i to the specific volatilization rate for
water (i.e., K./a).
By assuming the catch basin is completely mixed at all times,X >»x.
H_0 i
and the temperature of the aqueous phase remains constant a series of
differential component balances plus overall balances yields upon integration
the concentration of component i as a function of run time and the other
pertinent system variables:
o • aL-t
x - xi" |i - -;r=—| " (2)
for all aL t
-------
The Concentration Independence of ^
The interphase mass-transfer phenomena occuring in the packed column is
of a transient nature but the specific voltalization rate of component i,
K , is independent of the concentration of this component in the column.
This occurence is essential for use of equation (4) to evaluate le.
The steady state flow operation of a packed column for interphase mass-
transfer of low concentration constituents being transferred to a gas phase
is given by (Treybal, 1968)
where Z is the height of a packed section,
K a is the overall mass-transfer coefficient for component i in
Y i
the gas phase,
G is the molar gas flux rate, and
m
y. is the mole fraction of component i in the gas phase.
Since most volatile components in a wastewater stream are of low con-
centration Henry's Law is sufficient for expressing the phase equilibria for
isothermal operation:
Y » H.X. . (6)
An i— component balance over any arbitrary section of the packed column
yields:
Relations (6) and (7) are sufficient to transform the integral in (5)
to one of liquid phase concentrations. After this transformation is completed
the indicated integration operation is performed and rearrangement yields:
-116-
-------
|(
(S^-2) EXP [ZT-£— (S^-Ujj- /(Sj^-1), (8)
^- ^~ HI
HiGm
where S^ =—— . Now if this expression is employed with the definition of
K. and assuming L.. = L. , a result showing that K, is independent of con-
centration is obtained, i.e.,
i (SR1~2) EXP FT y * ^Ri"11! + l (9)
L m -I
The final result showing K to be independent of concentration of i also
suggest that this equation can also be used to compute K. apriori. All
terms are easily obtained from experimental data except for K a.• This term
can be estimated by the use of published correlations (Perry, 1963), however
ideal systems consisting of binary components with no extraneous interferences
are typically employed in obtaining the correlated K a data. Even under
these conditions regression techniques are needed to correlate the resulting
data. The presence of trace components at or near the interface is known to
invalidate a literature value of K a. Due to the heterogeneous nature of
most wastewater an experimentally determined K. is most likely advantageous.
Equation (9) should be used as an estimate of K. only. K. xs also independent of
concentration for non-isothermal operations however equation (9) does not
apply since H. is strongly temperature dependent.
-117-
-------
Continuous Volatilization of Multiple Components
It is common practice to measure organic pollutants in wastewater by
gross measures such as biochemical oxygen demand (B.O.D.), chemical oxygen
demand (C.O.D.), and total organic carbon (T.O.C.), etc. Although these
gross measures indicate the concentrations of the combined volatile and non-
volatile components and not single constituents, much information concerning
the relative amounts of these two factions can be learned from these tests
combined with a volatilization experiment. The gross concentration measure
can be expressed as the sum of the concentrations of the volatile and non-
volatile fractions thus:
C° - Cj + C° (10)
where C° is a gross concentration measure (such as BOD, COD, TOC, etc.)
of the original wastewater, and
C. is the sum of the concentrations of all the volatile components in
equivalent measure as C, and
C? is the sum of the concentrations of all the non-volatile components
in equivalent measure as C.
The concentrations C° and C° and hence the total fraction consisting of
volatiles plus the relative volatilization rate (reference to water) can be
obtained from a single air-volatilization experiment.
The concentration behavior of any single volatile component during
an air-volatilization experiment is given by equation (2), where the specific
volatilization rate, K., is greater than zero. For non-volatile constituents
K. » 0, implying none of these components are removed in the packed column,
so that equation (2) becomes
(11)
-118-
-------
Just as equation (2) shows that the concentration of the volatiles decrease
with time equation (11) shows the concentration of non-volatiles to increase
with time. The quantity of wastewater in the device is related to run time
by
aL t
M - MO ( 1 - -jp- ) (12)
o
By analogy to equation (10) the gross mass quantity of constituents in the
intial wastewater aliquot is
C°M - C? M + C° M (13)
o i o j o
At any time during the experiment the quantity of constituents in the waste-
water is
CM - CjM + C.M (14)
Since the concentration of the volatile and non-volatile constituent
n m
at any time during the experiment is^Vand^ x. respectively, equation (14)
i-11 j=l 3
becomes, upon substituting (2), (11), (12) and summing over all constituents,
n m
01
where n is the number of volatile components other than water,
m is the number of non-volatile components, and
6 r all t/M is a dimensionless run time.
m o
A more useful form of the above expression is obtained by dividing equation
(15) by C°Mo:
m
F° (16)
whare P3CM/C°MO »nd is the fraction of the total amount of the gross
constituents remaining in the wastewatar/
-119-
-------
F° =x?/C° is the original fraction of volatile component i in the waste-
water, and
F° sx°/C° is the original fraction of non-volatile component j in the
wastewater.
Equation (16) shows much about the behavior of the wastewater constituents
with experiment run time. As run time increases 0 -+-1.0 the fraction of
material remaining is all non-volatile, i.e.,
m
At this point it is obvious that C° - d-F_,) C° and C? = (F ) C°.
X t^X J tr1!
Although a gross concentration measure cannot single out individual
constituents, a slight change in the form of equation (16) allows further
study of the volatile fraction. If the volatile fraction is assumed to be
made of a single pseudo volatile component equation (16) is re-interpreted as
K
-*. L1-9] * *
where F° is the total fraction of the pseudo component and K is the specific
volatilization rate of this component.
Now due to the above F + F . »1. Employing equation (17) , F = 1 - F0_1 ,
S fc^~i S vJ— X
S
and performing a logarithmic transformation results in
K r -I
log
-------
Experimental Materials and Methods
Kraft mill evaporator condensates are known to contain a broad array of
volatile organic compounds plus non-volatile organics and provide an excellent
wastewater for volatilization experiments. Kraft mill decker filtrate was
studied also. The following six volatile compounds have been identified in
the condensates: methyl mercaptan, methyl sulfide, acetone, methanol,
a-pinene and methyl disulfide. Methanol is the major volatile component.
The static vapor-liquid equilibrium method (Turner and Van Horn, 1969),
and a gas chromatograph were used for identification and analysis.
The static vapor-liquid equilibrium method is quite simple for use with
gas-chromatography. A liquid sample is placed in a 500 ml. flask, the
flask is sealed with a serum cap, and then the flask and its contents are
placed in a constant temperature bath at 55 C. After sufficient time is
allowed for vapor-liquid equilibrium to be established, usually 30 to 60
minutes, a vapor sample is removed by inserting a syringe needle through the
cap, and the sample is injected into the chromatograph. This method has
been found to give very reproducible results, and once peaks on the chromatogram
are identified, calibration curves can be prepared from pure compounds and
quantitative results obtained.
A second method which was found to be satisfactory for quantitative
analysis of the more concentrated components involved direct injection of a
known volume of liquid condensate sample into the chromatograph. Concentrations
can be determined by comparing the measured peak area of the chromatogram to
previously determined calibration curves.
The identification and analysis equipment used throughout this study
consisted of a Perkin-Elmer Model 881 Gas Chromatograph equipped with a flame
ionization detector, a recorder, and an Infotronic CRS-104 Integrator.
Separations were made on a six foot, 1/8" column of 15 percent Carbowax
-121-
-------
20M on 100-120 mesh Chromosorb W operated at 70 C with a helium flow of
30 cc/minute.
A relatively large batch, recirculating air-water contactor was employed
in this study since experiments relating to cooling were also studied,
however the simple apparatus shown in Figure No. 2 is capable of the same
general performance. This device was a countercurrent flow, columnar, gas-
liquid contactor of "j square foot cross-sectional area (6"x6") of Plexiglas
construction. Overall height was approximately twelve feet, ten of which was
packed with Poly-Grid media stacked on 14 inches centers. The Poly-Grid is a
plastic material of high void volume consisting of IS inch sections meshed
with 2"x2" square openings. Air for contacting the wastewater was drawn
through the tower (induced draft) by two turbine-type fans located atop the
column. Liquid flow is via a positive displacement, squeeze-action type
pimp. Auxiliary equipment included a heat exchanger for re-heating the cooled
recycle liquid plus rotameter, thermometers, manometers for air flow measurement,
rheostats and sample ports.
As was described briefly in a previous section, volatilization experiments
were performed by placing a known volume of wastewater into the tower catch
basin. The volume charged was normally 10 to 12 liters. One extended run
was performed with a charge of 30 liters. A total of ninteen experiments were
performed, eighteen with evaporator condensates and one run with a simulated
wastewater consisting of methanol and tap water.
After charging the basin with the test wastewater the blowers were
started and ambient air was drawn upward through the tower. Air velocity
(apparent) in the tower ranged from 200 to 600 feet per minute. The waste-
water recirculating pump was started and flow was adjusted and maintained
constant throughout the run. Liquid loading rates ranged from 2.00 to 8.00
gallons per minute per square foot of column area. Simultaneously to starting
-122-
-------
the liquid flow, steam was admitted to the heat exchanger to maintain a
liquid temperature of 125°F at the top of the column. This temperature
could be controlled (manually) to + 3 P. This top temperature was maintained
constant for all runs. No attempt was made at trying to control the basin
temperature or the temperature and humidity of the entering air. The air
was accepted as found in the laboratory. The air temperature was approximately
75°F (dry bulb) but the relative humidity varied from 50 to 90%.
Experimental data obtained from the column during the volatilization
operation was: initial charge volume, liquid flow rate, air flow rate,
final charge volume, total duration of fun, GC analysis and/or BOD concentration
with run time. All samples were withdrawn from the catch basin.
-123-
-------
Behavior o_f Selected Organic Compounds During Volatilization
By employing the above analysis techniques it was possible to study the
effects of volatilization by observing the remaining organic compounds
in the water. Each sample injected into the chromatograph provided a point
value or the "signature" of the remaining flame ionizable material. By
recording the wastewater'-s signature with time it was possible to observe
which compounds were being removed and obtain some indication of their
relative rates of removal (qualitatively) Figures No. 4 and No. 5 show the
effects of volatilization of two kraft mill wastewaters.
Evaporator condensates chromatograms are shown in Figure No. 4. This
wastewater has a large portion of volatile organics and their removal via
volatilization is shown by the disappearance of individual peaks or the reduction
in size of these peaks between the start and at some time later in the run.
Identified constituents being volatilized are methanol, acetone, methyl
mercaptan, and methyl sulfide. The compounds are known to be very volatile
and methanol is the predominant volatile constituent. The other peaks have
not been identified.
Decker filtrate chroroatograins are shown in Figure No. 5. This waste-
water has only a small portion of its dissolved organics that are readily
volatilized. These chromatograms also show the decline in concentration of
the more volatile constituents, but there remains a predominance of relatively
non-volatile organics which are not easily volatilized. a-Pinene has been
identified as one such compound. Notice the relative unchanged nature of
the prominant peaks after volatilization for 20 minutes.
-124-
-------
ui
i
B) CONDENSATE AFTER VOLATILIZATION
A. MFTHYL MERCAPTAN
B. METHYL SULFIDE
C. ACETONE
D. METHYL ALCOHOL
E. a-PINENE
F. METHYL DISULFIDE
Time
A) ORIGINAL CONDENSE
,D
Time
FIGURE NO. 4: CHROMATOGRAM OF EVAPORATOR CONDENSATE
-------
B) DECKER AFTER VOLATILIZATION
A) ORIGINAL DECKER
ro
O>
i
A.
B.
C.
D.
E.
F.
METHYL MERCAPTAN
METHYL SULFIDE
ACETONE
METHYL ALCOHOL
ct-PINENE
METHYL DISULFIDE
t
Time
Time
FIGURE NO. 5: CHROMATOGRAM OF DECKER FILTRATE
-------
Volatilization Rates of Single Components
Methanol was by far the most prominent volatile compound of concern
in the evaporator condensates. Table No. 1 shows typical average concentrations
of the identified constituents in this wastewater. Dae to its obvious
importants with respect to the volatiles in this wastewater, methanol was
used as the key component for detailed studies of the volatilization mechanism.
Methanol concentration was obtained with run time as was described above.
This experimental data was then analyzed to obtain the relative rate of
methanol volatilization to that for water (i.e., K_, ..../a) .
UH.Un
The experimental data was transformed as suggested by equation (4),
namely M -aL t , the volume of wastewater remaining vs. x_t, , the con-
o in CH ^OH
centration of methanol remaining. The former is the independent variable.
Once these two variables are obtained a log-log plot can be made. Figure
6 shows typical data for three runs.
This plot shows several aspects of the nature of trace component removal
by volatilization. First the plot shows that as run time increases (i.e.,
as system volume decreases) the concentration of methanol decreases. Also
the functional relationship between these variables is logarithmic, as predicted
by the trace component volatilization model, provided K and a are constant. This
is proved to be correct as shown by the straight line correlation of the data.
All nineteen runs were displayed on a plot of this type and all resulted in
a straight line. This straight line indicates that for approximately three
cycles of methanol concentration the slope is constant. Equation (4) reveals
that:
KCH,OH a
slope - (20)
By experiment design a is constant, therefore K_,, _u must be constant
CH_OH
also and can be obtained from equation (20).
Table No. 2 shows the values of K and a obtained on the nineteen
C.H OH
-127-
-------
10.
I I
I I I
ro
oo
3.0
o
1.0
D
£ Run 7-22
Q Run 7-23
Q Methanol Water
0.2
I i i i i i i
i i i i i I
ii iii
3.0
10.
TOO.
X, Methanol Concentration, (MG)/(Liter)
FIGURE NO. 6: METHANOL VOLATILIZATION, CONDENSATE SAMPLE 7-15-68
1000.
-------
Table No. 1
Concentration of Volatile Organics is Kraft Condensates
(Turner and Van Horn, 1969)
methanol
methyl disulfide
acetone
a-pinene
methyl sulfide
methyl mercaptan
(Big)/(I)
1100
6.5
3.
1.4
.2
Trace
-129-
-------
Table No. 2
Specific Volatilization Rates of Methanol and Water
L/G*
1.32
1.55
1.35
1.59
1.52
1.35
1.36
1.35
2.85
5.01
2.02
2.45
1.06
1.27
0.623
0.506
5.17
0.516
1.32
ave. 0.327 .0364
S 0.12
*Air wet bulb temperature also varied from run to run but was not
recorded.
K
3
.250
.268
.452
.448
.304
.324
.269
.260
.186
1232
.351
.268
.411
.176
.481
.594
.112
.484
.336
a
.0256
.0504
.0423
.0732
.0350
.0373
.0275
.0400
.0238
.0238
.0245
.0270
.0401
.0392
.0381
.0470
.0150
.0433
.0402
-130-
-------
experiments. There is a strong variation in K and a with the liquid to
air flow rates and temperature in the column. These effects are predicted
by equation (9). The average value of (K OI/a* was 9>18 *s " 2-57)
indicating that the specific rate of methanol volatilization is approximately
nine times more rapid than that for water volatilization. Table No. 3 is
helpful in interpreting the significance of the magnitude of the relative
volatilization rate.
-131-
-------
Table No. 3
Interpretation p_f the Relative Volatilization Rate
i' Character of Constituents
Specific volatilization rate is greater
>1 than water and hence i can be removed from
the water phase by air contact.
specific volatilization rate is the same
=1 as water and i cannot be removed from
the water phase by air contact.
specific volatilization rate of i is less
than that of water and air water contact
<1
will result in the concentration of
this constituent in the aqueous phase.
-132-
-------
The Volatile Fraction of Gross Pollutant Measures
It was shown above that these kraft mill wastewaters contain components
which are relatively non-volatile. The quantitative amount of the volatiles
that can be removed from a wastewater by volatilization is an important
piece of information that can be obtained from this experiment. Measuring
the concentration of the gross pollutants remaining in the basin can be
performed on the same sample employed for single component analysis. These
gross concentration measures can be: BOD_, COD, TOC, etc. The BOD of
the wastewater was employed in a series of volatilization experiments in this
study.
Equation (16) suggest that a plot of F vs. 0 may be profitable since the
fraction of the total amount of the gross constituent measure which is non-
volatile can be obtained at the limit 0 » 1. A total of fourteen volatilization
runs were analyzed for BOD on two separate samples of evaporator condensate.
The results of these runs appear in Figures No. 7 and No. 8. Both figures
show that as 0 increases (i.e., run time increases) the fraction of biodegradable
organic material remaining decreases rapidly at first and then more slowly for
larger 0 values and finally becoming invariant for 0 =0.5. The constant
F section of the curve is significant for it shows that fraction of the
original BOD that is not readily volatilized. Figure 7 shows that approximately
25% of the original BOD of this particular sample of condensate is non-
volatile and hence cannot be removed by air contact. Conversely 75% can be
removed by air contact. The sample represented by the data in Figure No. 8 is
approximately 14% non-volatile. Obviously evaporator cond^nsates contain con-
stituents that are volatile and which accounts for 75 to 86% of the oriqinil
of this wastewater.
-133-
-------
oo
-P>
i
'E 0.8
'5
C£
in
o
O
rn
r- 0.6
10
01
O
O
c 0.4
o
0
^
^"^ ^^^^^^^^^ ^^
(rp- — -A— nAn 0 o ^
^^.
^ ^ ^ A O ^
^ txj
t t i i .1 i It
0.0 0.1 0.2 O.J mnM,ncinnleee nun •Hm'i-0 0.6 0.7 0.8
FIGURE NO. 7: VOLATILE FRACTION OF EVAPORATOR CONDENSATE, SAMPLE 7-15-68
-------
oo
tn
i
o, 0.8
c
c
re
o;
in
O
§ 0.6
c
Ol
O
£ 0.4
c
o
•r-
-M
O
-------
Volatilization Rates of Combined Components
The data presented in Figures 7 and 8 can be transformed to $ vs 1-Q
by equation (19) once F . has been obtained. This has been done for the two
evaporator condensate samples and the results appear in Figure 9 and 10. A
perfunctory examination of these figures indicates that the data does not
correlate in a linear fashion and hence K is not a constant. This is not
surprising when it is realized that the wastewater has a spectrum of volatile
constituents which display a spectrum of relative volatilization rates. As
the experiment approaches complete removal of easily volatilizable components
the less volatilizables remain and therefore display lower relative volatili-
zation rates and a reduced magnitude of slope.
Extracting the slope at various points on the figures reveals some interesting
aspects about the volatile fractions of the BOD of this evaporator condensate.
The slope on Figure 9 for 0.1< $ <1.0 is 8.47 with corresponding K = 0.284
™ • S
while the slope for .01 < < 0.1 is approximately 4.53 giving K = 0.152.
"" ™ S
Note that the average value of K for methanol reported in Table No. 2
in
(0.327) is similar to that of the first volatile constituents to be removed.
The last fraction of volatiles to be removed is approximately half as easy to
volatilize as methanol. The slope in Figure 8 is computed to be 7.94 resulting
in a value of K * 0.253, again of the order of the specific volatilization
rate of methanol. It is also interesting to note the relative volatilization
rates for the most volatile fraction as computed from BOD, measurements vs.
those computed from methanol concentration measurements:
methanol analysis (several samples) 8.98
BOD5 analysis (sample 7-15-68) 8.47
BOD analysis (sample 8-1-68) 7.94
These relative rates of volatilization obtained from BOD analysis suggest
that the bulk fraction of the volatile BOD consist of methanol or constituents
of comparable volatilization nature.
-136-
-------
1.0
0.1
0.0
SYMBOL RUN
O
a
7-22
7-23
Others
_L
_L
O
Q
0.1 ' - 91.0
FIGURE NO. 9: GROSS CONSTITUENT VOLATILIZATION CHARACTERISTICS, SAMPLE 7-15-68
-137-
-------
1.0
0.1
0.01
SYMBOL
O
D
A
RUN
8-1
8-2
8-8
Others
I
Q
-I
l i i i
0.1 1-0 1.0
FIGURE NO. 10: GROSS CONSTITUENT VOLATILIZATION CHARACTERISTICS, SAMPLE 8-1-68
-138-
-------
Conclusions
A method of studying the volatile character of the dissolved constituents
in an aqueous effluent has been presented both from theoretical and
experimental points of view. The following list includes the major findings:
1. An experimental apparatus for volatilization studies was presented
which was void of the transient operation problem associated with
the volatilization of water.
2. By the use of this apparatus one can find whether or not a wastewater
contains volatile constituents.
3. The fraction of the original gross pollutant measure (i.e., BOD ,
COD, TOC, etc.) which is non-volatile can be determined experimentally.
4. The specific and relative volatilization rates of individual volatile
constituents can be determined.
5. The specific and relative volatilization rates of the combined
volatile constituents can be determined.
6. A method of interpretation is presented by which one can predict
the consequences of air-water contact operations on the volatile
constituents. This can be done by inspecting the magnitude of the
relative volatilization rate.
This project has been financed in part by the Environmental Protection Agency
pursuant to the Federal Water Pollution Control Act. Results obtained will
be confirmed in Water Quality Office Project No. WPRD 116-01-68 entitled
"Kraft Waste Treatment in Cooling Towers." The content does not necessarily
reflect the views and policy of the Environmental Protection Agency.
-139-
-------
Literature Cited
Prather, V.B., JWPCF, 42, 4, p. 596 (1970).
Cohn, R.G. and E.T. Tonn, TAPPI, 47, 3, p. 163A (1964).
Mohler, E.F., Jr., H.F. Elkin and L.R. Kummick, JWPCF, 36, 11, p. 1380 (1964)
Smathers, R.L. and J.H. Frady, paper presented at National Paper Awards
Contest (TAPPI), February 1969.
Estridge, R.B.,B.G. Turner, R.L. Smathers and L.J. Thibodeaux, paper
presented at TAPPI Air and Water Conference, June 1970.
Burns, O.B., Jr., and W.W. Eckenfelder, Jr., TAPPI, 48, 11, p. 96A (1965).
Slechta, A.F. and G.L. Gulp, JWPCF, 39, 5, p. 787 (1967).
Eckenfleder, W.W., Jr., and D.J. O'Oconnor, Biological Waste Treatment,
Pergamon, New York (1961).
Treybal, R.E., Mass Transfer Operations, McGraw-Hill, New York (1968).
-140-
-------
Nomenclature
K. - Specific volatilization rate for component i
L - volumetric or molar flux rate of aqueous phase (fc/sec.cm )
m
x. - concentration of component i in aqueous phase (mg/i)
a - specific volatilization rate for water
M - volumetric quantity of liquid in the device (i)
t - experiment run time (min)
y - concentration of component i in gas phase (mg/J.)
Z_ - height of column packed section (cm)
K a. - overall mass-transfer coefficient for component i (g-moles/sec.cm )
G - molar flux rate of gas phase in column (g-moles/sec.cm )
m
C - a gross pollutant concentration measure (rng/J.)
0 - a dimensionless run time
n - total number of volatile components in aqueous phase
m - total number of non-volatile components in aqueous phase
F - fraction of the gross pollutant constituents remaining in the
aqueous phase
4> - fraction of the gross volatile pollutant constituents remaining
in the aqueous phase
Subscripts
s -a pseudo component
i - inlet of column when double subscripts appear
o - outlet of column when double subscripts appear
i - denotes a volatile constituent
j - denotes a non-volatile constituent
Superscripts
o - indicates initial condition (i.e. t = 0)
-141-
-------
APPENDIX H
MICROBIOLOGICAL REPORT ON KRAFT MILL COOLING TOWER WATERS
Submitted September 10, 1969
Donald G. Ahearn, Ph.D.
Associate Professor Microbiology
Georgia State College
Atlanta, Georgia
Wood pulp processing waters flowing through the experimental cooling
tower at the kraft mill in Macon, Georgia were examined for their micro-
bial flora. Waters leading to and from the tower as well as those
flowing over the plastic trickling filters were examined. On July 28,
1969, the first collection day, waters were plated onto bacterial and
fungal isolation media at the site of collection. Control media were
exposed to the environment during the collection period to determine
the aerial flora. Isolation media included potato dextrose agar (PDA),
malt extract-yeast extract-peptone agar with chloramphenicol and lactic
acid (M-12), nutrient agar (NA) and blood agar (BA). Samples were col-
lected in sterile bottles and appropriate alliquots were innoculated
onto the isolation media by spread plate technique. Triplicate plates
were prepared of each sample; incubation temperatures of 24 and 43°C
were employed. Samples collected on August 7 and September 3, 1969,
were diluted in sterile deionized water prior to plating.
The sites of collection are listed below.
Code No. Site
KO Tower base (sump)
Kl Lower plastic grid, N.E. Corner
K2 Lower plastic grid, N.E. Corner
K3 Sump, adjacent to pump
K4 Top of tower, sprinkler
K5 Composite sample (to tower)
K6 Decker filtrate
K7 Tower effluent for recycling
K8 Heavy liquor effluent
K9 Total composite mill effluent
Results
Bacteria were the predominant organisms isolated from all samples. Average
bacterial numbers are given in Table I. The low populations obtained on
July 28 may be in part due to alteration of techniques. The predominant
species obtained on July 28 included representatives of the genera
-142-
-------
Bacillus, Flavobacterium. Pseudomonas. Serratia. and Aerobacter. On
August 7, the relative concentrations of the Flavobacterium-
Pseudomonas complex and Serratia groups appeared to increase. This
latter species pattern was also found on September 3, 1969. The
bacterial populations were markedly higher in samples collected on
September 3, 1969. Special attempts were made to obtain yeasts
and other fungi from the samples. No yeasts or other fungi were
isolated on July 28. Only a few filamentous fungal colonies were
obtained on August 7. The effluents collected on September 3 failed
to yield fungi; however, two separate redwood samples from the cooling
tower frame gave Trichoderma viride, a known cellulolitic fungus.
The wood samples on collection showed no visual evidence of decomposi-
tion. Tricoderma viride is common in the papermill environment and
its culturing from the wood samples may have been fortuitous. The red-
wood frame should be carefully cultured and visually examined during
the next few months.
Slime material was collected on all three collection trips from the
top of the tower (41°C) and from the base of the tower (34°C). Samples
from both sites contained about 5 nematodes per microscopic field (ca.
200 sq y) on each collection trip. Mature worms were found in slime
from the top, whereas immature specimens predominated in lower samples.
On July 27, the slime was composed of wood cells interlaced with a
large gram positive rod (Bacillus?) embedded in a matrix of gram nega-
tive rods. The nematodes, bacterial predators, were feeding on the
slime. The slime was similar on August 7, 1969, except for the presence
of motile algae or protozoan in the lower slime sample. On September 3,
1969, the slime at the base of the tower had turned reddish in hue and
contained numerous filaments of algae or fungi and profuse numbers of
rotifers. Since the filamentous slime did not yield fungi on culture,
the filaments were most probably algal. No filaments were observed in
the upper slime sample.
Comments
Determination of accurate bacterial densities in "activated sludge" or
in organically enriched waters containing particulate matter is beyond
present technology. Low bacterial densities may not reflect low activ-
ity, but rather, the presence of increasing numbers of predators. The
numbers of bacteria and bacterial predators have progressively in-
creased during the sampling period. Following enrichment, the total
biologic component of the tower showed a significant increase. There
is insufficient data to evaluate the influence of the closing of opera-
tions for the Labor Day holiday on this increase. It was noted that
the total mill sample, aside from its higher bacterial population, was
visually clearer on September 3, 1969, and was of a lower pH (pH 6
versus pH 7 on August 7).
-143-
-------
Bacterial densities reported for sites K5 to K7 should be evaluated
with consideration of the sample source. The process waters were ob-
tained from composite sample tanks. The temperature of water in
these tanks is at least 10°C below that of the cycling water and the
water has been static for varying periods prior to sampling.
The low numbers of fungi obtained may be partially due to the high
pH (pH 8-9) of most of the process waters. Further redwood samples
should be obtained to monitor the possible development of wood rot
fungi. The occurrence of Trichoderma viride is interesting, but is
of no immediate concern.
Questions
Is there a method of removing sludge from the sump at the base of the
tower?
Has addition of enrichment with a readily assimilated carbon source
been examined? This may permit co-oxidation of the more recalcitrant
sulfur derivatives.
TABLE I
Bacterial Densities in Mill Process Maters
Collection
Site July 28 August 7 September 3
KO -- (1) 1.8 x 105 1.6 x 107
(2)
Kl 3 x 105(2) 7.5 x iQb 3.5 x 106
K2
K3 " "
K4 " " 3.5 x 1Q5 2.5 x 10^
K5 5 x 105 5.3 x 10° 2.4 x 107
K6 " " 4.1 x 107 1.8 x 106
K7 " " 3.2 x 106 1.5 x 107
K8 25-50 — . — _
K9 - 1.9 x 106 3.5 x 107
(1) -- not sampled.
(2) Average numbers.
-144-
-------
APPENDIX H
(Continued)
MICROBIOLOGICAL REPORT ON KRAFT MILL COOLING TOWERS
Report No. 2 Submitted October 11
Donald G. Ahearn, Ph.D.
Associate Professor of Microbiology
Georgia State University
Atlanta, Georgia
See Report of September 10, 1969.
The fourth microbiological testing of the process waters of the kraft
mill at Macon, Georgia was performed on September 23, 1969. The col-
lection stations and the culturable bacterial counts are presented in
Table I. The bacterial numbers were markedly increased over those
reported for September 3, 1969. Fungi, almost exclusively Ceotrichum
candidum, were observed in significant numbers for the first time. A~t
the time of sampling, water from the upper lake with high microbial
populations was being passed through the tower. The presence of high
numbers of fungi in the process waters and the increased bacterial
numbers appear, at least in part, related to use of the water from
the upper lake. The bacterial densities in the decker filtrate are
probably more accurate than those of the previous report. Samples on
September 23 were taken directly from the line rather than from the
sampling tank.
Slime samples were collected from the upper and lower tower. Slime from
the open lower portion of the tower was reddish brown, upper slime
samples from the enclosed portion of the tower were dark brown. Nema-
todes were observed in both samples. On September 3, 1969, it was noted
that the lower slime sample had turned reddish and that it contained
numerous filaments of a possible alga and small motile protozoa-like
organisms. The filaments and the motile forms respectively are repre-
sentatives of the genera Beggiatoa and Chromatium. These are sulfur
bacteria which obtain their energy by oxidizing sulfides or H2S. The
genus Chromatium is composed of purple photosynthetic bacteria which
store inorganic sulfur internally. They are frequently considered
anerobic, but recent evidence indicates that they require, at least,
low levels of oxygen. The motile chromatia and large spirilla of the
Qenus Spirulina are responsible for the reddish hue of the slime. As
°ur isolation media lacks HpS, these sulfur bacteria have not been
cultured. Direct counts indicate that their numbers are in excess of
2xlQ6 cells/ml of slime. It should be noted that bacterial counts of
slime material are inaccurate. The sulfur bacteria mav be of sianifi-
cance in the reduction of odor.
-145-
-------
Four redwood samples were collected from the tower frame. All samples
yielded fungi, Fusari urn sp. Festal oti a sp. and Inch ode rma virde. Only
the latter species is established to be celluloytic. In previous wood
testings, J. virde was isolated. I did not visually observe wood
determination, however, the repeated isolation of fungi from the wood
frame suggests that the preservative has leached out from the wood
surface.
TABLE I
BACTERIAL AND FUNGAL DENSITIES IN
MILL PROCESS WATERS ON SEPTEMBER 29, 1969
KO
K4
K3
K5
K6
K6
K9
K10
Stations
Tower base (sump)
Top of tower (sprinkler
Sump (pump valve)
Composite sample (line valve
Decker filtrate (sample tank)
Decker filtrate (line valve)
Total composite mill effluent
Upper Lake
No. /ml
Bacteria
>/x!09
3.5x105
3.5x10?
5.5xl05
1.5x103
1.5xl03
2.8xl08
>lx!09
Fungi
310
60
200
0
0
0
22
367
-146-
-------
•*i-.'
GEORGIA STATE UNIVERSITY
., 33 GILMER STREET, S.E. . ATLANTA, GEORGIA 30303
January 9, 1970
Department of Biology-
Mr. Billy G. Turner
Project Manager
Research and Development Center
Georgia Kraft Company
Krannert Road
Rome, Georgia 30161
Dear Billy:
The enclosed material presents the characteristics of representative
bacteria isolated from the Macon plant. These bacteria are mainly
motile, gram negative rods of the Pseudomonas complex. These we
have attempted to fully characterize. In addition to the pseudomonads,
several other bacteria were common. The general characteristics of
these bacteria also are given.
The photomicrograph illustrates the sulfur bacteria, Chromatiuni sp. (large
cigar shaped rod with internal refractile sulfur globules) and Spirulina
sp, (corkscrew rod). See my second report submitted October 11, 1969.
The nematode species has not been identifiedj due to expense I have not
pursued this effort.
This report concludes our studies of the last sampling trip (September
23, 1969). An itemized bill for our services is attached. If I can be
of help in preparing your reports on the cooling tower please call.
Regards,
Donald G. Ahearn, Ph.D.
Associate Professor of Microbiology
DGA:am
Enclosures
-147-
-------
CHARACTERIZATION OF PSEUDOMONAD-LIKE BACTERIA FROM PROCESSING WATERS
5K-6 6K-6 7K-6 8K-6 9K-5 10K-5 H065 13K-2
Assimilation
Glucose
Xylose
Mannitol
Lactose
Sucrose
Maltose
Fermentation
Glucose
Xylose
Mannitol
Lactose
Sucrose
Maltose
Urea
NOo
Indole
M.R.
V.P.
Gelatin
Litmus Milk
10% Glucose
10% Lactose
Catalase
Oxidase
Nutrient Broth
MacConkey
SS agar
Citrate
Cetrimide
Colony Pigment
Flouracin
Pyocyanin
HIT
Growth in T6Y
5°C
25
37
42
Motility 37
25
Alk
Alk
Alk
Alk
Alk
Alk
W
Brw/Sol
A
A
A
A
A
A
A
W
A
Alk
W
W
W
A
A
A
WA
WA
W
W
W
Alk
Alk
Alk
NC
Alk
Alk
Red
A
Alk
NC
A
Alk
W
Red
Alk
Alk
Brw/Sol
Brown
Brown
Brown Yel.Brw Brown "
Green Buff Yellow
Brown Brw/Sol W/Yel
W
W
Buff
Lt.Brw.
Lt.Brw.
Brown
-148-
-------
5K-6 6K-6 7K-6 8K-6 9K-5 10K-5 H065 13K-2
<\ol 1 ov*c
TSI Slant
Butt
H2S
Gas
_________ _ _ _ _R1 II /Gv»n ------------
Alk Alk HC Alk Alk HC Alk Alk
NC NC NC NC NC NC NC NC
----- - --
W = weak; NC = No Change; AC = acid; Alk = Alkaline.
Species designations:
5-K6 Pseudomonas sp (similar to Alcaligenes faecalis)
6-K6 Pseudomonas sp. (similar to £_. diminuta)
7-K6 ? Aeromonos sp.
8-K6 Pseudomonas aeruginosa
9-K5 Pseudomonas sp.
10-K5 Pseudomonas sp. (similar to P.. stuzeri)
H-065 Pseudomonas stutzari
13-K2 Pseudomonas sp. (see 5-K6)
Non Pseudomonad bacteria:
K-8-14 Large gram positive, pink pigmented non-spore forming rod.
K-8-15 Diphtheroid-like, gram positive, non-spore forming, orange
pigmented rod.
K-8-16 Baci11 us sp. Aerobic, gram positive, spore forming rod.
Survived heat shock (800C for seven days).
-149-
-------
1
5
Accession Number
A
Subject Field &, Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Georgia Kraft Company, Rome, Georgia
Research and Development Center
Title
TREATMENT OF SELECTED INTERNAL KRAFT MILL WASTES IN A COOLING TOWER,
]Q Authorfs)
— — ' McAlister, J. A.
Turner, B. G.
Estridee. R. B.
16
21
Project Designation
Grant 116-01-68 (Project No.
12040EEK)
Note
22
Citation
23
Descriptors (Starred First)
Pulp Wastes*, Cooling Towers,* Waste Treatment, Water Reuse*, Pollution
Abatement, Pilot Plants, Analytical Techniques, Biochemical Oxygen Demand
25
Identifiers (Starred First)
Evaporators, Effluents, Condensates, Filtrates, Methanol, Stripping,
Volatilization, Liquefiers
27
Selected waste streams from a kraft linerboard mill have been successfully
treated in a conventional cooling tower. These waste streams, along with condenser
waters from a barometric type evaporator condenser, were cooled in the tower and
reused. This process accomplished BOD removals of about 10,000 Ibs per day and an
8-10 MGD reduction in overall mill water needs. Theoretical, laboratory, and pilot
investigations of BOD removal mechanisms involved proved the predominant mechanism
was stripping of volatile components. During laboratory studies a procedure called
the static vapor-liquid equilibrium method was developed for analyzing low concen-
tration volatile components in waste water. Mathematical relationships were
developed, allowing translation of study findings to other waste water treatment
applications. Primary factors controlling BOD removals were blowdown rate, liquid-
gas ratio, and average temperature. For a blowdown rate of 15-20% of tower influent,
treatment efficiencies averaged 55-65% for sixth effect condensate, 45-55% for com-
bined condensate and turpentine decanter underflow, and 25-35% for decker filtrate.
BOD reductions were believed due primarily to methanol stripping. Some
biological activity was evident, however, and nutrient additions resulted in a
5-10% improvement in BOD removal. The system has several advantages over con-
ventional surface condenser systems. Both operating and capital costs compare
favorably with other waste treatment methods. (This report contains 16 references.)
Abstractor
B. G. Turner
Institution
Georgia Kraft Company
WR: 102
WRSIC
(REV. JULY
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTFR
U.S. DEPARTMENT OF THE INTERIOR CENTER
WASHINGTON. O. C. 20240
• 6"0: 19*9-359-339
------- |