WATER POLLUTION CONTROL RESEARCH SERIES
17040 DNM 02/71
FEASIBILITY OF TREATING
WASTEWATER BY DISTILLATION
ENVIRONMENTAL PROTECTION AGENCY • RESEARCH AND MONITORING
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Water Quality
Office, Environmental Protection Agency, through inhouse
research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Room 1108,
Washington, D. C. 20242.
-------
FEASIBILITY OF TREATING WASTEWATER BY DISTILLATION
by
University of Florida
Gainesville, Florida 32601
for the
ENVIRONMENTAL PROTECTION AGENCY
Project #17040 DNM
Contract #14-12-571
February 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1
-------
EPA Review Notice
This report has been reviewed by the Water Quality
Office, EPA, 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
recommendation for use.
ii
-------
ABSTRACT
The technical feasibility of evaporation of municipal sewage treatment
plant effluent for the purpose of water reuse was investigated. The
equipment used was a long tube vertical (LTV) evaporator. The objectives
of the research were to determine the effects of feedwater quality, and
evaporation conditions on product water quality, post evaporation polish-
ing, and evaporator tube scaling.
The experimental equipment consisted of a three-effect evaporator con-
structed of 316 stainless steel. Each effect contained a single one-inch-
diameter tube with a 14-ft. effective heated length. The evaporator could
be operated as a single-effect or triple-effect unit. Possible operating
conditions varied from around 28 in. Hg vacuum to about 55 psig. All
evaporator feedwater was first acidified and degassed under vacuum to re-
move dissolved gases.
Feedwaters tested in the evaporator included extended aeration effluent,
high rate trickling filter effluent, and contact stabilization effluent.
All three units treated raw waste from a large university complex.
Results showed that an odor free product could not be produced from any of
the three feedwaters under any operating condition from 112 F to 290 F.
The odor intensity increased as the evaporator operating temperature and/or
the chemical oxygen demand of the feedwater increased. In all cases, post
treatment with activated carbon removed all odors. Aeration would not re-
move all odors. Product produced under pressure conditions using trickling
filter feed contained significantly more organic contamination than any
other products. Product contamination by ammonia could not be controlled
by adjusting the pH of the feed in the range 5.1 to 8.7. However, ammonia
in the product water was removed by ion exchange.
The scaling evaluations were carried out under pressure conditions using ex-
tended aeration and trickling filter effluent. Trickling filter feedwater
was judged unsuitable because of excessive scaling. Scaling problems with
extended aeration feedwater were minor, and with post treatment by acti-
vated carbon, a high quality product water was produced. Trickling filter
feedwater gave more severe scaling problems, mostly because of the use of
more sulfurijc^ acid in the degassing pretreatment.
Because of increased efficiencies due to higher operating temperatures and
negligible boiling point rise, wastewater evaporation should be more economi-
cal than sea water evaporation.
This report was submitted in fulfillment of Contract No. 14-12-571, Program
No. 17040 DNM, under the sponsorship of the Water Quality Office, Environ-
mental Protection Agency.
iii
-------
CONTENTS
Page
Conclusions 1
Recommendations 3
Introduction 5
Theoretical Background 9
Research Plan 21
Description of Equipment 27
Experimental Results 37
Discussion of Results 69
Appendices 75
References 81
v
-------
FIGURES
NO- PAGE
1 DISTRIBUTION OF NH3 AND NH4 AS A FUNCTION 12
OF pH AND TEMPERATURE
2 CALCIUM SULFATE SOLUBILITY 16
3 OVERALL FLOWSHEET 28
4 DEGASSING SYSTEM FLOWSHEET 29
5 EVAPORATOR FLOWSHEET 31
6 EVAPORATOR 32
7 VAPOR-LIQUID SEPARATOR 33
8 PRODUCT RECEIVER SYSTEM FLOWSHEET 34
9 CONDUCTIVITY VS. AMMONIA CONCENTRATION 52
10 PRODUCT RATES, FIRST EXTENDED AERATION 55
EFFLUENT RUN
11 OVERALL HEAT TRANSFER COEFFICIENTS 60
vi
-------
TABLES
No.
1 Dissociation Constants for Aqueous Ammonia and Water n
2 Analytical Tests on Liquid Samples 24
3 Analytical Tests on Scale 26
4 Evaporator Operating Conditions for Initial Single 39
Stage Tests
5 Initial Tests, Extended Aeration Effluent, Vacuum 40
Conditions
6 Initial Tests, Extended Aeration Effluent, Atmospheric 41
Conditions
7 Initial Tests, Extended Aeration Effluent, Pressure 42
Conditions
8 Initial Tests, Contact Stabilization Effluent, Vacuum 44
Conditions
9 Initial Tests, Contact Stabilization Effluent, Atmos- 45
pheric Conditions
10 Initial Tests, Contact Stabilization Effluent, Pressure 46
Conditions
11 Initial Tests, Trickling Filter Effluent, Vacuum Con- 47
ditions
12 Initial Tests, Trickling Filter Effluent, Atmospheric 48
Conditions
13 Initial Tests, Trickling Filter Effluent, Pressure Con- 49
ditions
14 Typical Analysis of Tap Water, Gainesville, Florida 53
15 Analytical Results - First Extended Aeration Effluent Run 57
16 Scale from First Extended Aeration Effluent Run 58
17 Analytical Results - Trickling Filter Effluent Run 59
18 Scale from Trickling Filter Effluent Run 61
vii
-------
No.
19 Activated Carbon Treatment of Products from Trickling 61
Filter Effluent Run
20 Analytical Results - Second Extended Aeration Effluent 63
Run
21 Bacteriological Tests - Second Extended Aeration Effluent 64
Run
22 Scale from Second Extended Aeration Effluent Run 65
23 Analytical Results - Three Effect Ammonia Distribution Tests 67
24 Campus Sewage Treatment Plant Test Results, High Rate 75
Trickling Filter
25 Campus Sewage Treatment Plant Test Results, Contact 76
Stabilization Plant
V11L
-------
CONCLUSIONS
Based on the results of this study the following conclusions have been
drawn:
1. Product water produced by evaporation of extended aeration, contact
stabilization, or high rate trickling filter effluent is not acceptable
for reuse without further treatment because of organic odor carry-over
from the feed.
2. Product odor tends to increase with increasing evaporator operating
temperature in the range 112-288 F.
3. The odor of products produced from secondary effluents can be ef-
fectively removed by treatment with activated carbon. However, preliminary
indications are that the odors cannot be reliably removed by aeration.
4. At high temperature (280-290 F) significantly more-organic carry-
over to the product water occurs with trickling filter effluent feed
than with extended aeration or contact stabilization effluent feed.
5. Ammonia in the evaporator feedwater can be expected to contaminate
the product water.
6. Removal of ammonia from the product water by ion exchange appears
practical.
7. Trickling filter effluent appears to be less suitable as evaporator
feedwater than extended aeration or contact stabilization effluent be-
cause of greater scaling problems.
8. At temperatures up to 280 F, scaling problems with extended aeration
effluent appear to be minor.
9. Evaporation of extended aeration effluent followed by activated
carbon treatment gives a completely demineralized odor-free product water
suitable for high quality reuse.
10. Because of increased efficiencies due to higher operating temperatures
and negligible boiling point rise, wastewater evaporation should be more
economical than sea water evaporation.
-------
RECOMMENDATIONS
In order to develop the concept of evaporation of sewage treatment plant
effluent to the demonstration plant stage, additional work is recommended
in the following areas:
1. Additional longer term pilot plant runs, lasting at least two months,
should be made to better define any scaling problems.
2. Additional pilot plant runs should be made at temperatures above
280 F to find the maximum allowable operating temperature. This maximum
may be controlled either by product quality or scaling problems.
3. Studies should be made to determine the capacity of activated carbon
when used to treat evaporator products for odor removal. This would
provide the basis for reliable economic estimates of this post treat-
ment process.
4. Corrosion tests should be made to determine the most economical
materials of construction for a full scale unit.
-------
INTRODUCTION
General Background
The need for improved water pollution control combined with an increasing
scarcity of natural, waters has led to the concept of wastewater renova-
tion for reuse. One treatment method that has potential for producing
a high quality water from wastewater is distillation or evaporation.
Extensive interest has developed in the use of distillation for producing
usable water from seawater. Development work by the U. S. Department of
Interior's Office of Saline Water and by other organizations around the
world has reduced the cost of the process to the point at which water can
be produced in some locations at a price competitive with or lower than
that for water from other sources.
While much of the technology of seawater distillation can probably be
applied to wastewater, there are differences in the two feeds which have
an important effect on operating conditions and product quality. Muni-
cipal wastewater usually has less than 1000 mg/1 dissolved solids compared
with 35,000 mg/1 for seawater. {The work here has been limited to muni-
cipal wastewater of essentially domestic origin. This does not preclude
the use of distillation on municipal wastes containing industrial wastes
or on industrial waste alone. Some of the conclusions drawn from this
study may be applicable to a number of industrial wastes.] The corrosive-
ness and scaling potential (for materials such as calcium sulfate) of
wastewater should be less than for seawater. The boiling point elevation
for wastewater is essentially zero. These characteristics of wastewater
should allow for higher distillation temperature and higher concentration
of the brine or blowdown than is practical with sea water. In contrast
to the advantages for distillation of wastewater are the disadvantages of
soluble and suspended organic materials which may foul heat transfer sur-
faces and which may contribute taste and odor to the distillate. Ammonia,
which is present to some degree in all municipal wastes, has measurable
volatility under usual distilling conditions and, therefore, also affects
product quality. Its corrosiveness toward copper presents an additional
problem. Phosphate is present in significant quantities. Its presence
could be an advantage or disadvantage depending upon its interaction with
other materials in the water.
t*
There are a number of important questions that remain to be answered be-
fore the feasibility of wastewater distillation can be realistically as-
sessed. The present work was undertaken to answer some of these questions.
Previous Work on Evaporation of Treated Sewage
Previous work on evaporation of sewage treatment plant effluent has been
limited to several desk top evaluations using presumably reasonable as-
sumptions and four studijes that involved actual experimental work.
-------
In 1960, Hickman compared distillation with other unit processes for the
reclamation of wastewaters at the request of the Advanced Waste Treatment
Research Program of the U. S. Public Health Service. Hickman concluded
that the relatively high cost of distillation (estimated at $1.00/1,000
gal.) should limit it to processing only part of the water in a total re-
cycle scheme. Other less expensive treatment procedures should be used
for the remaining parts of the system. Hickman reported the results of
one experimental run using treated sewage from a Rochester, N. Y., sewage
disposal plant. The specific evaporator used was 'not described but opera-
ting conditions were reported as 120 F (1.69 psia). In the author's opinion,
"the physical properties, flavor, and odor of the distillate were good, and
that with mild chlorination, or equivalent, the distillate would be potable
and acceptable for municipal reuse." Possible contamination of the dis-
tillate samples prevented any valid conclusions based on actual analysis.
n
In 1963 an economic analysis and a pilot plant study of wastewater dis-
tillation were made by Neale. He proposed that distillation be used
in parallel with some other form of treatment and that the fraction of
the total stream subjected to distillation be the minimum required to pre-
vent build up of inorganics in the overall system. Analysis of data from
22 cities indicated that the fraction of the total stream that would have
to be distilled ranged from 23.670 to 63.77» and averaged 42.770. A pilot
plant study was carried out in a long tube vertical (LTV) evaporator
utilizing a single tube 14 ft long. The primary objective of the tests
was to evaluate tube fouling. No provision was made to condense and col-
lect product water. Using primary effluent and no pretreatment, tube
fouling was noted after four days. Operation was at 212 F. A second
seven-day run at 212 F was made using secondary effluent with no pretreat-
ment. Again tube fouling was noted. The effervescence noted when acid
was added to this scale led Neale to conclude that it was largely calcium
carbonate. Subsequent runs were made using secondary effluent treated
with inhibited hydrochloric acid for pH adjustment. Feed pH was main-
tained at less than 5.7 most of the time. An eight day run at 212 F and
a twelve day run at 234 F were made with no scaling problems. Neale con-
cluded "that waste treatment by distillation will probably be applicable
to renovation of wastewater at temperatures equal to or exceeding those
used for sea water conversion, and that raw sewage, after removal of most
of the suspended matter, may be used for distillation plant feed."
In June, 1963, American Machine and Foundry, Inc., under contract to the
Government, undertook research on flash evaporation of treated sewage.
The equipment consisted of an eight-stage flash unit. The still body was
steel with titanium heat transfer surfaces. Due to operating problems,
very limited testing was carried out. Using secondary effluent under
relatively high temperature conditions, 250-300 F, severe fouling of the
preheater and some fouling of the feed side of the condenser tubes oc-
curred. The scale appeared to be largely organic. The product water
had a strong, disagreeable odor. However, treatment with activated car-
bon removed the odor. "
In 1965, O'Connor7 and co-workers at the Robert A. Taft Sanitary Engi-
neering Center in Cincinnati, Ohio, conducted laboratory scale tests on
-------
distillation of municipal effluents. Batch tests were made using primary
effluent, trickling filter effluent, and extended aeration effluent. Tem-
peratures used were 158, 212, and 338 °F and pH ranged from 3 to 11. At
the lower pH values it was hoped that the ammonia would be retained in the
concentrated liquid. At the higher pH values, hopefully all the ammonia
would distill over in the first small fraction of product. All feed
samples were filtered prior to evaporation to remove solid particles larger
than about 5 microns. It was found that ammonia was present in the pro-
duct water for all feeds with pH above about 3.5. With feeds containing
around 7-8 mg/1 NH3-N, ammonia in concentrations above 1.0 mg/1 NH^-N
typically was found in the first 40-50% of the distillate. Even at pH
11, ammonia in concentrations above 1.0 mg/1 NH3-N was found in at least
the first 30% of the distillate. Hence it appeared impractical to try,
on a large scale, either to eliminate ammonia from the distillate by
lowering the feed pH or to concentrate all the ammonia in the first small
fraction of the distillate by raising the feed pH.
Tests with extended aeration plant effluent having ammonia concentrations
less than 2.2 mg/1 NH3-N showed that, at feed-pH values under 5.5, all
product fractions had ammonia concentrations less than 1.0 mg/ NH3-N. No
carry-over of nitrate was observed.
Odors were present in all products and seemed to be more "prominent and
enduring" at high temperatures. Product odor seemed to be independent
of feed pH. Powdered carbon at 1 gm/1 removed all odors from products
produced under vacuum and atmospheric conditions. However, carbon treat-
ment only partially removed the odors from the products made at high tem-
perature and pressure conditions.
Economic analyses of distillation of treated wastewaters have been made
by several people. Gerster made a desk top study of multistage flash,
multiple-effect, and recompression-flash evaporation with respect to how
these processes might apply to wastewaters. He concluded that "For all
types of equipment the cost for the distillation step alone is somewhat
less than for sea water, but inclusion of costs for feed pretreatment
and ultimate disposal of blowdown, bring the cost up to about that for
seawater."
9 10
Stephan ' in 1965 predicted that wastewater evaporation should cost
about the same as seawater evaporation. Also, he concluded that a
paralle,^ renovation system utilizing evaporation on half of the recycle
stream would cost about the same as using electrodialysis on the entire
stream. **v
Objectives of this Research
Review of the previous work on evaporation of sewage treatment plant
effluent shows that additional information is needed in several areas.
First, very little work has been done on a continuous basis utilizing
equipment simulating full scale operations. Also, renovation by evapora-
tion should be considered in relationship to any additional pretreatment
-------
or post treatment processes that may be required to render the water
acceptable for reuse.
Based on these needs, the following objectives were set for this research.
1. To define the relationships between product water(quality, feedwater
quality, and evaporation conditions.
2. To define the relationships between post evaporation polishing re-
quired, feedwater quality, and evaporation conditions.
3. To define the relationships between evaporator tube scaling, feed-
water quality, and evaporation conditions.
-------
THEORETICAL BACKGROUND
Volatile Organlcs
Treated sewage contains materials that can be classified as volatile
under temperatures likely to be encountered in evaporation. These in-
clude numerous organic compounds. It would be desirable to predict how
these will affect distillate quality. If sufficient data were available,
Henry's law could be applied to make the prediction. Unfortunately, the
limited work that has been done to characterize sewage treatment plant
effluent plus the day-to-day variations in effluent quality make it vir-
tually impossible to construct a predictive model for specific compounds
in the effluent. *•>^ To complicate matters further, it appears that
some of the organics in the evaporator feed are broken down by heat in
the evaporation process to produce smaller, more volatile compounds.'
This appears to be particularly true as evaporator temperature is in-
creased. Therefore, it is impossible to predict either the amount or
the specific compounds one might expect to carry over during evaporation.
Ammonia
Nitrogen in the form of ammonia is found in significant amounts in the
effluent from almost all biological treatment processes. The exception
is treatment involving extended oxidation conditions wherein the ammonia
is oxidized to nitrate. Even in the extended aeration processes it is
not unusual to find 0.5-1.0 mg/1 NH3~N.
The 1962 Public Health Service Drinking Water Standards1^ set no limit
on ammonia concentration. Likewise the more stringent Water Quality
Goals^'^ established by the American Water Works Association make no
mention of ammonia. In a 1968 report issued by the Federal Water Pol-
lution Control Administration on raw water quality criteria for public
supplies, a permissible criteria of 0.5 mg/1 NH3-N and a desirable criteria
of less than 0.01 mg/1 NH3-N were recommended.1° xhe rationale for these
low limits was given as follows. "Ammonia is a significant pollutant in
raw water for public water supplies because its reactions with chlorine
result in compounds with markedly less disinfecting efficiency than free
chlorine. In addition, it is frequently an indicator of recent sewage
pollution."^ In order to satisfy a chlorine demand of 1 mg/1 NH3-N and
produce a free chlorine excess, i.e. "breakpoint" chlorination, about
10-mg/1 of chlorine are required.7 Hence, for economic reasons alone,
it appears undesirable to have ammonia above 1.0 mg/1 in the final product
water. The fact that no standard for ammonia in potable water exists
probably reflects that historically ammonia has not been a problem and
not that ammonia was not considered undesirable.
The volatilization of ammonia in an ammonia-water system is influenced
by pH, ammonia concentration, and solution temperature. Ammonia dissolves
in water giving the(following reaction:
-------
NH3 + H20 + NH* f OH" (1)
= ( m } I OH" ]
[ NH3 ]
If values of Kb at various temperatures are known, the relative concen
trations of NHj and NH/3 can be calculated as a function of pOH. Now,
since
KW = [ OH" ] [ H+ ] (3)
and values of K^ at various temperatures are known, the relative con-
centrations of NH^ and NHo as a function of pH can be determined.
[H+] = [NH] K,, (4)
[ NH3] K^
pH = log [ NH3J + pK,, - PKb (5)
17 18 19 20
Values of pK^ ' and pKfc, ' for ammonia are shown in Table 1. .
Using these values and equation 5, the distribution of NHo and NH, was
calculated as a function of pH at various temperatures. Figure 1 shows
that at 20°C and pH 7 more than 99% of the ammonia is present as NH4 and
hence not subject to evaporation. However, as the temperature increases,
the equilibrium shifts and there is more ammonia available for evapora-
tion. This would explain the fact that O'Connor found that he had to
lower the feed pH to under 5 in order to eliminate ammonia in the product
water .
The relative volatility of ammonia as compared to water strongly in-
fluences the relative amounts of ammonia and water in the liquid and
gas phases during evaporation. At the very low ammonia concentrations
encountered in treatment plant effluents Henry's law should apply. Un-
fortunately, no data have been published for ammonia-water systems in
the 0 to 20-mg/l NH3-N concentration range. However, considerable data
are available for ammonia-water systems in ammonia concentrations above
17o (10,000 ppm) . 1~23 These data show the relative concentration of
ammonia in the vapor phase to be much higher than in the liquid phase.
This concentration factor increases as ammonia concentration in the
liquid decreases and is around 10 at the 1% (ammonia in liquid) level.
Therefore, at the much lower ammonia concentration levels we are con-
sidering, the ammonia concentration factor from liquid to gas should
substantially exceed 10. Concentration of ammonia by evaporation and
10
-------
TABLE 1
DISSOCIATION CONSTANTS FOR AQUEOUS
AMMONIA AND WATER
Temp. pK for pK for Aqueous
w
water
.
°C w 17,18 b . 19,20
Ammonia '
0 14.9435 4.862
10 14.5346 4.804
20 14.1669 4.767
25. 13.9965 4.751
30 13.8330 4.740
40 13.5348 4.730
50 13.2617 4.723
60 13.0171
100 12.30 4.87
-11-
-------
100
.P.H
Figure 1. Distribution of NH3 and NH, as a Function
of pH and Temperature
-------
condensation is well recognized and is an accepted laboratory procedure
in the analysis of ammonia.^
The effect of temperature on ammonia volatilization frooi an ammonia-
water system is that the ammonia concentration factor decreases slightly
as temperature increases. 1 However, this effect does not appear to be
significant over the temperature range that would be considered for
evaporator operations.
From a theoretical standpoint, it appears that ammonia carryover to the
product is dependent on the pH of the feed solution and the evaporation
temperature. Since the ammonia-ammonium ion equilibrium shifts in favor
of ammonia as temperature increases, one might well expect to find signifi-
cant ammonia in the product water at pH values at least as low as 6 if
evaporation is occurring under pressures exceeding atmospheric.
Scaling and Fouling
Loss of heat transfer efficiency may result from inorganic scaling, or-
ganic fouling, or a combination of the two.
Inorganic scaling occurs when the solubility of the scaling compound
is exceeded. This may occur because of temperature increase, since
the solubilities of many scale forming salts decrease with increasing
temperature. In an evaporator, the continual evaporation of water con-
centrates the non-volatile ions in solution to the point that the entire
solution may be saturated with respect to a particular salt. However,
scaling may occur at localized points in the system even though the solu-
bility of the particular scaling salt is not exceeded in the bulk of the
water. This scaling can result from evaporation in the film of liquid
immediately adjacent to the heating surface or merely from the higher tem-
perature of the liquid in that film. The latter can occur only when
solubility decreases with increasing temperature.
In the first case, where the entire solution becomes super-saturated,
precipitation may occur throughout the bulk liquid. This precipitate
may remain suspended in the solution and be removed with the concen-
trated effluent. However, it may settle on the tube surfaces and "bake"
in place as a scale. The second case is more detrimental in that the
precipitation tends to occur directly on the heating surface and there
is little chance that the solid thus formed will be suspended in the
solution.25'^5
25
With regard to boiler scaling, the Betz Handbook makes the following
comments:
"It is well to realize that the prevention of boiler scale
cannot be predicted by any basic chemical principal. It
is the physical characteristics of the precipitate formed
in the boiler water that determines whether or not the pre-
cipitate will tend to tightly adhere to the boiler heating
13
-------
surfaces - the chemical characteristics of the pre-
cipitate are relatively unimportant. Thus, from a
chemical standpoint sodium silicate will precipitate
the calcium and magnesium salts just as well as a
phosphate or carbonate. However, it is the; physical
characteristics of the precipitate that are of im-
portance, for a precipitate of calcium silicate will
tightly adhere to the heating surfaces."
Treatment for prevention of inorganic scale may be external or internal.
External treatment involves removing the potential scale forming mate-
rials from the water prior to its introduction into the evaporator,
boiler, etc. Internal treatment involves the selective precipitation
of the potential scale forming materials in the form of non-scaling
sludges. This may include the addition of materials designed t'd* keep
the sludges fluid and in suspension so that they can be removed from
the boiler or evaporator in the blowdown or concentrated effluent. The
principal inorganic scale and/or sludge formers are calcium carbonate,
calcium sulfate, magnesium hydroxide, and silica.
Calcium Carbonate
Calcium carbonate is one of the least soluble salts that is likely to
occur in wastewater evaporation. At temperatures below boiling this
scale can be controlled by adjusting the carbonate equilibrium system
in favor of HoCOS and HCC-Q with pH control. Excellent discussions of
9R 9Q *^f) ^1
this equilibrium system have been presented by Langelier,^0'^ Dye, '
and Weber and Stumm. However, pH control of the carbonate system can-
not be accomplished under boiling conditions where the C_02 solubility
is very low. Under such conditions HCO§ converts to C0$ and ^CO^, and
CaCC>3 can precipitate according to the following reactions.
2HCO~ * H20 + C02 + CO^ (6)
-H- ' SB "*"
Ca + C03 •*• CaC03 (7)
Since CO^ solubility is very low, the continual removal of COo from
the system forces essentially complete decomposition of the HCOo.
One obvious way of controlling calcium carbonate scale is by removal of
the calcium. This could be done by hot or cold lime-soda softening or
*7 S 9*7 *^^
by ion exchange. '» However, in the case of wastewater distilla-
tion, these processes would add significant additional costs to the
overall system. In addition, ion exchange is questionable because the
effects of the organics on fouling of the exchange resin are not well
known.
Calcium carbonate scale also can be controlled by pH adjustment and de-
gassing prior to evaporation.2 This involves lowering the pH to put the
carbonate in the ^COg form and then removing it as CC>2 by degassing.
14
-------
Finally, calcium carbonate scaling may be controlled internally by con-
trolled precipitation of the calcium as a non-scaling sludge. ' '
The most common procedure is to precipitate the calcium as hydroxy-
apatite, a mixed calcium phosphate and hydroxide. Several sodium phos-
phates may be used including trisodium phosphate, disodium phosphate,
sodium metaphosphate, and monosodium phosphate. In some cases it is
necessary to add organic dispersants to prevent the sludge from growing
into aggregates. Typical types of organic materials used for this pur-
pose are tannins, lignins, glucose derivatives, and starches. ' '
Precipitation of calcium by phosphate and dispersal of the sludge by
organics has promise for use in wastewater evaporation since both phos-
phates and organics naturally occur in this water. However, individual
wastewater streams may or may not contain sufficient phosphate and the
dispersal abilities of the organics in wastewater are not known.
Calcium Sulfate
Calcium sulfate presents a more serious problem than calcium carbonate
because it forms a very hard and adherent scale.27,33 Calcium sulfate
solubility decreases with t;emperature as shown in Figure 2.34,35 it
should be noted that this is the solubility of pure calcium sulfate,
anhydrite, in pure water. In wastewater, calcium sulfate will have
a higher solubility due to the increase in ionic strength caused .by
the other ions in solution.
Control of calcium sulfate scale is accomplished in much the same way
as control of calcium carbonate. External treatment to remove or re-
duce the calcium content of the water is applicable.25,27,33 internal
treatment to remove calcium by controlled precipitation with phosphate
can likewise be used." ,27 ,33 ^e comments made previously regarding
these control measures all apply here.
Magnesium Hydroxide
Magnesium hydroxide has a solubility of about 5 ppm at 212°F decreasing
to slightly less than 1 ppm at 392°F. Control of this scale may be
external by removal of the magnesium by softening or ion exchange. ''
Internal control can be accomplished by controlled precipitation with
phosphate^ However, magnesium phosphate tends to form a sticky deposit
and requires^echanical removal. Finally, since magnesium forms a
hydroxide scale, it can be controlled by maintaining thepH and hence
hydroxide concentrations at a low enough level so as not to exceed
the solubility.
Silica
Silica tends to form complex scales which may include calcium, magnesium,
aluminum, or may be composed almost entirely of silica.25,27 These
scales are usually ver.y hard, glassy, and adherent. McCoy^^ gives a
maximum permissible concentration of silica of 250 ppm at 388°F for
15
-------
700.
850
800-
550-
500
6
£ 450-
>T
H 400-
O
CO
S
cd
o
350-
300-
250
200,
150-1
100-
50
0
• BOOTH AND BIDWELL (44)
o NEALE (4)
A LINKE (43)
SOLID PHASE - CaSO
200
I
220
I
240
I
260
280 300 320
TEMPERATURE, °F
340
360
380
400
Figure 2. Calcium Sulfate Solubility
-------
boilers. This figure drops to 175 ppm at 421 F. Presumably it increases
for temperatures below 388°F.
25
In some cases too low silica concentrations can cause problems. If
magnesium concentration is high and phosphate is present, the magnesium
will precipitate a sticky magnesium phosphate sludge. However, if suf-
ficient silica is present, magnesium silicate, which is easier to handle,
will be precipitated.
Control of silica scale may be external by hot lime-soda softening or by
coagulation with ferric sulfate. ' Ion exchange using a strongly basic
anion exchange resin regenerated with sodium hydroxide may also be used.27
Organic Fouling
Organic materials may cause heat transfer problems by any of several dif-
ferent mechanisms. Suspended organics may simply deposit on heat transfer
surfaces. Oils or greases, which have a tendency to coat metal surfaces,
may cause problems. The high temperature in the evaporator may tend to
polymerize some organics into high molecular weight insoluble compounds
that could coat tube walls. In the latter stages of evaporation, as con-
centrations of both organic and inorganic materials increase, the possi-
bility of salting out exists.
Another potential organic fouling problem exists in the early stage pre-
heaters (temperature less than 130 F). This is the problem of bacterial
growth. It is possible that this could be adequately controlled by
periodic shock chlorination.
In contrast to all the potential problems with organic fouling, there are
some possible positive benefits. Certain organics may have a dispersing
effect which would tend to keep solids suspended. Also, volatile organic
vapors carrying over from one effect to the steam chest of the next effect
may promote dropwise condensation on the evaporator tubes. If the organic
material coats the condensing surface and renders it non-wettable, con-
densation will occur dropwise and since dropwise condensation gives higher
heat transfer, the overall efficiency of the process would be improved. 36
Bacteria or Virus Contamination
«*•
Since in evaporation of sewage treatment plant effluent it is inevitable
that bacteria and virus will be present in the feedwater, a question
that must be faced is what are the possibilities of these organisms con-
taminating the product water. Three factors affect bacterial and viral
contamination of the distillate. Considered together, these factors make
contamination very unlikely. First, the organism would have to survive
the maximum feed temperature in the evaporator. The forward feed multiple-
effect evaporator design requires that jill feedwater pass from the hot end
of the system to th^ cold end. Hence, all incoming water is heated uni-
formly to the maximum operating temperature. In the case of seawater
17
-------
this temperature is 250 F. However, as pointed out previously, it is
likely that the upper temperature limit for sewage treatment plant
effluent will be higher than this. For comparison purposes, milk is
pasteurized at 161°F and microbiological laboratories sterLlize their
equipment in autoclaves at 250°F. Sterilization, tiftnes as low as 10
minutes are used for small samples. Because of the uniform heat con-
ditions obtained in an evaporator, the somewhat shorter holdup times at
the maximum temperature should be at least as efficient as 10 minutes
in an autoclave.
The second factor dictating against bacterial or viral contamination of
the product is the basic separation mechanism being used, i.e. evapora-
tion and condensation. Since bacteria and virus are nonvolatile, they
cannot contaminate the product by evaporating and condensing with the
product. To get into the product water it would be necessary for^ny
organisms to be carried over physically in the vapor stream, i.e. en-
trainment. The extent to which this would be possible would depend on
the type and design of the vapor liquid separation system in each effect.
The third factor that would be important in eliminating any danger from
bacterial or viral contamination would be chlorination. As with any
municipal potable water supply, all water from the plant should be
chlorinated.
Considering these facts, it seems reasonable to conclude that possible
danger from bacteria or virus in a forward feed multiple-effect evapor-
ator system using sewage treatment plant effluent as feed could be less
than the same danger from a conventional coagulation plant using surface
sources of 'raw water.
Ultimate Disposal
Practically all wastewater treatment processes involve separation of
the incoming liquid into a "clean" and "dirty" stream. In all cases
some final disposition must be made of the "dirty" or concentrated waste
stream. This problem is very significant in evaporation of sewage treat-
ment plant effluent. The final concentrated effluent stream from the
evaporator probably will be on the order of 3-10% by volume of the in-
coming stream and contain essentially all the inorganic S'alts and organic
material in the feed stream. Ultimate disposal of the stream will de-
pend on the individual situation; however, several relatively simple
means would be ocean outfall, underground disposal, or solar evaporation.
When these methods cannot be used, more sophisticated and expensive tech-
niques would have to be employed.
Because of the high temperature anticipated for wastewater evaporation,
the concentrated effluent should be pathogenically safe for ocean dis-
posal in a well-designed outfall. Since the effluent would have an
appreciable organic content, provision should be made in the outfall
design to release the waste far enough offshore so as to preclude the
18
-------
possibility of contamination of beach or estuary areas. This method of
disposal is becoming less popular and is almost certain to be more closely
regulated in the future.
In areas where contamination of ground water would not be a problem,
underground disposal of the concentrated effluent could be considered.
Climatic conditions and availability of land would determine whether or
not disposal by solar evaporation would be possible.
Finally, it should be noted that ultimate waste disposal problems in an
area would not be increased by evaporation of sewage treatment plant
effluent. If evaporation were not used, the more dilute waste contain-
ing the same absolute amounts of impurities would still have to be dis-
posed of. Therefore, it is likely that evaporation would reduce rather
than add to ultimate disposal problems. If methods of disposal involve
dilution, then the problem would be increased.
19
-------
RESEARCH PLAN
The objectives of this project, particularly with regard to scaling evalua-
tions, dictated that the work be done in equipment of sufficient size to
simulate full scale operation, that operation be continuous rather than
batch, and that various effluents typical of what could be expected from
municipal waste treatment plants be used as evaporator feed. Within these
constraints the following approaches were taken to various parts of the
overall research plan.
Experimental Equipment
The long tube vertical (LTV) evaporator design was selected for use in
this work. In discussing all types of evaporators, Standiford-^S points
out that more evaporation is accomplished in LTV evaporators than any
other type and makes the following comments regarding the LTV design.
The widespread use of the LTV evaporator is due partly
to the ability to build large single units, partly be-
cause of the high heat-transfer performance exhibited
under most conditions, and partly because of the sim-
plicity and hence cheapness of construction (simply
a shell and tube heat exchanger surmounted by a vapor-
liquid separator).
Because of their high capacities, and lower costs than
for any other type, rising-film LTV evaporators are
used whenever possible . . . While they cannot usually
handle crystallizing solutions, they are widely used
for viscous and mildly scaling liquors. They are well
suited to corrosive solutions because heat-transfer
coefficients are generally high requiring a minimum of
expensive heating surface, and tube replacement is
simple . . . The principal disadvantage of the rising-
film LTV evaporator is the poor heat-transfer perfor-
mance at low temperature-differences or at low tempera-
tures .
For these reaions and because the rising-film LTV design is readily
adaptable to pil^t scale size units, this type evaporator was selected
for use in this research. Consideration was given to the use of a
single-effect unit. It was concluded, however, that since any full-scale
plant would be multiple-effect, pilot plant data from a multiple-effect
unit would be far more valuable than data from a unit that would only
simulate the first stage of a multiple-effect system.
Since corrosion was not to be studied in this research, it was deemed
desirable to construct the evaporator of 316 stainless steel to minimize
the possibility of produci contamination by corrosion.
21
-------
As with the evaporator, the accessory equipment used for feed preparation
and product polishing was designed to operate continuously.
In design of the evaporator provision was made to [allow operation as a
single stage unit or to operate all three effects in series. In either
mode of operation any pressure range from about 28 in. Hg vacuum to 100
psig could be selected. Unfortunately, steam pressure available at the
site limited the upper pressure to 50-55 psig.
The evaporator was constructed by Indian River Construction Co., Jacksonville,
Florida in cooperation with Reynolds, Smith, and Hills, Architects and Engi-
neers, Jacksonville, Florida.
The experimental evaporator was set up adjacent to the 2-MGD Campus Sewage
Treatment Plant operated by the University of Florida. This plant handles
the flow from the main campus, the teaching hospital, and medical center
complex, fraternities, sororities, dormitories, an elementary school, and
several married student apartment complexes. Flow from all these sources
results in a waste reasonably typical of largely domestic municipal sewage.
The Campus Plant operates a primary treatment unit, a high rate trickling
filter, a standard rate trickling filter, and a contact stabilization
activated sludge unit. A 9,000-GPD extended aeration unit is also avail-
able for use but is not operated routinely. Effluents from any of these
units are available for experimental purposes.
(
Experimental _P Ian
The experimental work was divided into two principal phases. The first
phase was directed towards defining the relationships between feedwater
quality, product quality, and evaporation conditions. Experimental runs
were to be made operating the evaporator as a single effect unit and
using progressively poorer quality effluents as feedwater. By this pro-
cedure the possibility could be minimized of having major difficulties
with excessive organic fouling as was experienced during the work done
at American Machine and Foundry, Inc. Thus testing was to proceed
using, in order, extended aeration plant effluent, contact stabilization
plant effluent, high rate trickling filter effluent, and primary effluent
until either operational difficulties or product water quality dictated
that further testing with poorer quality effluents was not justified.
Evaporator operating conditions were to be varied over the range from
about 28 in. Hg vacuum to about 50 psig. Also, the pH of the feed was to
be varied to see the effects on ammonia carry-over of the product. In
addition to measuring the relationships between these variables and pro-
duct water quality, the first testing phase included the evaluation of
several post treatment processes. These included activated carbon treat-
ment, aeration, and ammonia removal by ion exchange.
22
-------
After completion of the first phase of the experimental work some pre-
liminary judgements were to be made regarding the practical limits for
both minimum feedwater quality and maximum evaporator operating tempera-
ture consistent with acceptable product water. Using conditions at or
near these limits, extended continuous runs utilizing all three evapora-
tor effects were to be made to evaluate scaling or fouling. It was an-
ticipated that these runs would last one to two weeks each. Scaling was
to be evaluated both directly and indirectly. Indirectly, scaling could
be detected by decrease in overall heat transfer coefficient, Uo, for
each stage. Direct measure of scaling was to be obtained by cleaning the
evaporator mechanically and weighing the amount of scale removed. The
scale was to be analyzed to determine its cause.
In addition to providing data on scaling, the extended runs would provide
considerable additional data on the quality of product that could be ex-
pected under the conditions felt to be most likely for full scale opera-
tions .
Analytical Plan and Procedures
The analytical work involved three types of waters, sewage treatment
plant effluent or evaporator feedwater, evaporator product water, and
the concentrated effluent from the evaporator. Analyses were also con-
ducted on evaporator scale.
Table 2 lists the analyses made on-the liquid samples and the procedures
used. The very large number of samples involved dictated that automated
analytical techniques be used as much as possible.
Chemical Oxygen Demand, COD, was used as an indicator of the organic
content of the samples. An attempt was made to perfect a total organic
carbon procedure that would be accurate in the 0-3 mg/1 range, the range
anticipated for product samples. Unfortunately, with the equipment avail-
able, it was not possible to develop a procedure as precise as the dilute
COD procedure and this effort was finally abandoned.
Ammonia nitrogen was one of the most important variables in this study.
Its significance has been discussed previously.
Nitrate nterogen was used to indicate the degree of nitrification in the
various sewage^treatment plant effluents. Another important use of ni-
trate results was to measure the degree of physical carry-over of material
into the evaporator product water. Since nitrate is non-volatile, a mate-
rial balance over the evaporator based on nitrate should be a very accurate
measure of physical carry-over.
The importance of pH in ammonia distribution and in scaling has been dis-
cussed previously. Measurements were made primarily on feedwater samples.
Samples were adjusted to 25°C - 5°C prior to measurement.
Odor was subjectively measured by the operating personnel at the time the
products were produced. The difficulties involved in obtaining an odor
23
-------
TABLE 2
ANALYTICAL TESTS ON LIQUID SAMPLES
Test
Procedure
COD
Ammonia Nitrogen
Nitrate Nitrogen
pH
Odor
Total Dissolved Solids
Alternate Procedure for Dilute
Samples, Standard Method^*for the 24
Examination of Water and Wastewater
Industrial Method Ind.-19-69w of
Technicon Corp. Used Technicon
AutoAnalyzer
Method of Kann and Brezenski modified
for Technicon AutoAnalyzer
Corning Model 12 pH meter
Subjective opinion of operator
Residue on Evaporation at 103 C,
Standard Methods for the Examination
O A
of Water and Wastewater
-24-
-------
free room and an experienced panel dictated against more rigorous odor
testing.
Total dissolved solids were measured on a few of the feedwaters and con-
centrated effluent samples, primarily to better characterize the eva-
porator feedwater.
Some additional test results taken from the records of the Campus Sewage
Treatment Plant were used to characterize influent and effluent materials
from the various plant processes.
The scale removed from the evaporator was first dried at 103°C and weighed
to obtain a total weight figure. The samples were then fired at 600°C
to remove the organic material. The fired residues were dissolved in
dilute hydrochloric acid and the solution of the soluble portion analyzed
as shown in Table 3.
25
-------
TABLE 3
ANALYTICAL TESTS ON SCALE
Test For Procedure Used
Ca EDTA Titrimetric Method, Standard
Methods for the Examination of Water
and Wastewater^^
Mg Atomic Absorption Method Using Beckman
DB-G Spectre-photometer
Na Same as above
Fe Phenanthroline Method, Standard Methods
for the Examination of Water and
Wastewater^
SO Turbidimetric Method, Standard Methods
for the Examination of Water and
Wastewater
4.0
P04 Method of Murphy and Riley
Note: Fired scale samples were dissolved in 0.5N HC1
26
-------
DESCRIPTION OF EQUIPMENT
A general flowsheet for the process is shown in Figure 3. Feedwater for
the system could come from either an extended aeration plant, a trickling
filter plant, or a contact stabilization plant. Before any of these ef-
fluents were fed into the evaporator, the dissolved gases and carbonates
were removed. This was accomplished by adjustment of the pH and degass-
ing under a vacuum. After evaporation, provision was made to treat the
product water either in an activated carbon column or an ammonia selective
ion exchange column or both.
A diagram of the degassing equipment is shown in Figure 4. Effluent from
the various sources was put into a 175-gal. agitated tank for mixing and
pH adjustment. The agitator was a flat blade 30, in. in diameter by 1 3/8
in. high and was driven at 100 rpm. From the mix tank, the effluent passed
through a packed column under vacuum to remove dissolved gases. The packed
column was 4-in.-O.D. glass tubing. Vapor and dissolved gases were removed
from the column through a condenser and liquid trap by a sliding vane type
vacuum pump. Degassed liquid was pumped from the bottom of the column to
one of two 55-gal. open top feed drums by a variable speed aciew type pump.
The two feed drums were fitted with polyethylene drum liners . The mixing
tank, piping, and pump in the degassing system were of steel construction.
Temperature in the column was measured by a thermometer inserted into the
column through a small port. Provision was made to introduce steam into
the bottom of the column.
Evaporator
A diagram of the evaporator is shown in Figure 5 and a picture of the
evaporator is shown in Figure 6. The evaporator was designed to operate
either as a triple-effect unit or as a single-stage unit (3rd effect
only). Possible operating conditions in the evaporator ranged from about
28 ^in. of Hg vacuum to 100 psig. Actual operating conditions under pres-
sure* conditions were limited to about 55 psig by the steam pressure avail-
able at^'the site.
Heat for the evaporator could be supplied to either the first or third
effect. The heating medium could be either steam or hot water. Steam
was taken from the campus steam line and was reduced at the inlet to the
evaporator by a steam regulator. When operating under high vacuum, it
was necessary to reduce the temperature of the heating medium to avoid
unrealistically high temperature drops. This was done by using a closed
hot water system. The water was circulated continuously through the
evaporator using 'a natural gas fired hot water heater as the heat source.
27
-------
to
00
EXTENDED
AERATION
EFFLUENT
CONTACT
STABILIZATIC
EFFLUENT
TRICKLING
FILTER
EFFLUENT
J 0
0
— ^
R
R"
•— »•
ACID
\
pH A
MB
i
DJUS
DEGA
TMENT
SSING
EVAPORATOR
(
CONCENTRATED
EFFLUENT
OR
OR
NO POST
H»
TREATMENT
ACTIVATED
CARBON
TREATMENT
•
f
AMMONIA
ION
EXCHANGE
^. PRODUCT
^ WATER
Figure 3. Overall Flowsheet
-------
TREATMENT PLANT
EFFLUENT
to
vo
SULFURIC
ACID
MIXING
TANK
THERMOMETER
GAS AND VAPOR
WATER OUT.
WATER IN
PACKED
COLUMN
DEGASSED
LIQUID
1
CONDENSER
VACUUM
PUMP
LIQUII
TRAP
55 GALLON
FEED
DRUM
STEAM
VARIABLE SPEED DRIVE
SCREW TYPE PUMP
Figure 4. Degassing System Flowsheet
-------
STEAM
PRESSURE
CONTROL
VALVE
STEM
IN
Co
o
TO VACUUM
RECEIVERS
VLS - VAPOR LIQUID SEPARATOR
LLC - LIQUID LEVEL CONTROL
HWO - HOT WATER OUT
HWI - HOT WATER.IN
CWO - COOLING WATER OUT
CWI - COOLING WATER IN
VAP.- VAPOR
LIQ.- LIQUID
CONDENSATE
RETURN
ADJUSTABLE BACK
PRESSURE VALVE
PRODUCT
WATER
TO VACCUM
RECEIVER
PRODUCT WATER
*'OR DRAIN
NOTE: NOT TO SCALE
VARIABLE SPEED PUMP
0.1 - 0.6 gpm
VARIABLE SPEED PUMP
DRAIN' (CONTROLLED BY LLC)
Figure 5. Evaporator Flowsheet
-------
-
nrm
Figure 6. Evaporator
31
-------
Feed was pumped into the evaporator by a diaphragm metering pump. A spring
loaded, back pressure valve on the downstream side of the pump prevented
liquid from being drawn through the pump when the system was operating
under vacuum conditions. Each of three effects was constructed of 1-in.
IPS Schedule 5 pipe, jacketed with 4-in. IPS Schedule 5 pipe. The effec-
tive heated length of each effect was 14 ft.
Vapor-liquid separators atop each effect separated the steam-liquid mix
into two streams. This equipment is shown in Figure 7. The demisters were
316 stainless steel mesh (Otto H. York Co., Style 326). Small vent lines
(1/4-in. O.D. tubing) with needle type control valves connected the steam
jacket of each effect with its vapor-liquid separator.
Liquid seals were maintained between stages by float controlled liquid
drainers having 7/32-in. diameter orifices. These drainers were chrome
plated cast iron with stainless steel internals. Originally*,? drainers with
1/16-in. diameter orifices were tried, but were replaced becausei^he capacity
was marginal and chances of plugging were much greater than with larger units.
To reduce heat loss to the atmosphere, the three evaporator effects and con-
necting piping were insulated with 1-in. thick asbestos type insulation. The
three vapor-liquid separators were completely enclosed with custom built foam-
glass type insulating covers with aluminum exteriors.
Removal of the concentrated liquid effluent from the last effect was accom-
plished by a diaphragm metering pump identical to the feed pump. This pump
was automatically controlled (on-off) by liquid level probes located in an
enlarged section of the liquid line from the third effect vapor-liquid separa-
tor. Cooling coils, located between the liquid level probes and the pump,
cooled the concentrated effluent to prevent flashing of the hot liquid when
released to atmospheric pressure. All concentrated effluent piping was steel.
The back pressure valve following the pump was a copper alloy.
Condensate was removed from the jackets of the three effects through controlled
disc type steam traps.
Vapor from the third effect was condensed in a water jacketed condenser. The
condenser was a 2-in. IPS pipe with an effective condensing length of 12 ft.
Product from the three stages was removed by several means, depending on the
evaporator operating conditions (See Figures 5 and 8). Under pressure con-
ditions, product no. 3 was removed from the final condenser through an ad-
justable back pressure valve. This valve was of chrome plated copper alloy
construction. Product Nos. 1 and 2 were removed through the vacuum receiver
system as shown in Figure 8 except that the system was opened at valves D
allowing the products to drain continuously to glass carboys.
Under vacuum operation all three products were removed through the vacuum
receiver system. Under normal conditions, valves B and C were closed and
valves A were open. Valves D were check valves that allowed flow from the
upper to the lower tanks only. Hence, product water would drain from the
evaporator into the upper tanks and down into the lower tanks through valves
D. Vacuum was maintained on the system by a vacuum pump. In order to remove
32
-------
PRESSURE
GAUGE
PRESSURE
RELIEF VALVE
TO TEMPERATURE RECORDER
MESH DEMISTER
4" THICK
NOTE: DEFLECTOR SUPPORTED BY
4 RODS ATTACHED TO THE''
WALL OF THE SEPARATOR
STEAM
SCALE: NONE"
EVAPORATOR TUBE
LIQUID
.FIGURE 7. Vapor-Liquid Separator
33
-------
30 GAL
SURGE
TANK
30 GAL,
SURGE
TANK
C
HgO OUT
HgO IN
PRODUCT
NO. '6
|
^H
I^H
^M
^M
HgO OUT
CONDENSER
HgO IN
PRODUCT
9
1
C
p
rv '
IMLT.g
t
a
f
^M
•^
HgO OUT
CONDENSER
HgO IN
PRODUCT
®
'
(5
3—
)
<
NO. 1
r
:
3
f
'
vixv^/vi/
(D) - G]
RE(
CONDENSER
1
'•
X(B
>--i /r7\
"©" ^
B)(C) -SOLENOID
VALVES
- CHECK VALVES
10 GALLON
RECEIVER TANKS
VACUUM PUMP
Figure 8. Product Receiver System Flowsheet
-------
the product water from any one system, solenoid valves B and C were opened
and valves A were closed. The product water either drained through valves
C by gravity or was drawn into evacuated carboys. When the lower tank was
empty, the cycle was reversed closing valves B and C and opening valves A.
Two 30-gal. tanks in the vacuum system provided surge capacity when an air
filled tank was cycled back into the system. The instrumentation system
provided for automatic cycling of the receiver system to dump product water
from each effect on a timed sequence.
All six product receiver tanks were 10-gal. capacity.
Temperature was sensed by copper-constantan thermocouples at various points
in the system and automatically recorded by a 16 point recorder. The tem-
perature range of this recorder was 0-350°F. Pressure gauges, installed in
the incoming steam line, in each vapor-liquid separator, and in the final
condenser, gave continuous visual indication of the pressure at various
points in the system.
Carbon Column
The carbon column was a 1.94-in.-I.D. pyrex tube 4 ft. long, containing
42 in. of activated carbon. The column was fed by gravity and flow rate
was controlled by a polyethylene needle valve. The activated carbon used
was a Nuchar WV-W, 12 by 40 mesh, manufactured by West Virginia Pulp and
Paper Co.
Ammonia Column
The ammonia ion exchange column was a 1.94-in. pyrex tube 13 in. long,
containing 10.25 in. of ion exchange material. The column was fed by
gravity and the flow rate was controlled by a polyethylene needle valve.
The ion exchange material used was natural clinoptilolite, 20 by 50 mesh,
from the Hector, California, area.
35
-------
EXPERIMENTAL RESULTS
Initial Testing
As stated in the research plan, the initial testing was designed primarily
to define the relationships among evaporator feedwater quality, evaporator
operating conditions, and product water quality. Also, to be included
was evaluation of several post treatment processes and determination of
how they might be included in an overall reuse cycle.
With regards to effluents to be used, it was planned to begin the testing
with the highest quality effluent to reduce the likelihood of severely
scaling or fouling the evaporator tubes. Testing would then proceed using
progressively poorer quality effluents.
The tests on each different effluent were made at three temperature and
pressure levels. These conditions are shown in Table 4. In addition to
the temperature variations, the pH values of the feed were varied at each
level to examine the effect on ammonia distribution between the concen-
trated effluent and product.
On selected evaporator products that would be unsuitable for direct reuse,
tests were made using activated carbon, ammonia selective ion exchange
material, and aeration to determine the effectiveness of these post treat-
ment processes.
Before any sewage plant effluents were used as evaporator feed, the evapora-
tor was operated using tap water and distilled water. This was done to clean
out the inside of the equipment and to develop familiarization with the opera-
ting characteristics of the equipment. Steam was used to clean the product
receiver tanks.
When the evaporator was operated as a single-effect unit (3rd stage only),
equilibrium conditions for the entire system could be reached in less than
15 minutes. The only exception occurred when the feedwater was changed
during a run. When this occurred it could take as long as one hour to
flush completely the system and obtain consistent quality product water.
In these initial tests the operating procedure was as follows: The ef-
fluent to be used was either pumped or drained by gravity from the sewage
treatment plant to the mixing tank. A sample was taken for analysis and
the pH of1 the remaining water was adjusted to the bicarbonate endpoint
(pH 4.8-5.0) with 6N sulfuric acid. The water was heated to 120-140°F and
then degassed under vacuum. The degassed water was pumped to 55-gal. drums
prior to being fed into the evaporator. If required, a pH adjustment of
the evaporator feedwater was made at this point using either sulfuric acid
or sodium hydroxide.
In the degassing operation there was a 3-5°F drop in temperature caused by
flashing as the water entered the top of the column. This corresponded to
the measured 0.3-0.570 (of incoming water) evaporation rate from the degasser.
37
-------
TABLE 4
EVAPORATOR OPERATING CONDITIONS FOR
INITIAL SINGLE STAGE TESTS
Condition
Evaporating Liquid
Temperature and Pressure
Heating M&lium
Temperature and Pressure
Vacuum
114-120 F
26.5-27 in. Hg Vac.
Hot Water
140-155°F
Atmospheric
212 °F
0 psig
Steam
->OT
228-238uF
5-9 psig
Pressure
286-288 F
39-41 psig
Steamo
300-305 F
52-58 psig
38
-------
The liquid evaporated during degassing usually had a slight odor. However,
the odor was normally no stronger than the odor of products produced under
atmospheric pressure conditions.
The average degassing rate was 2800 ml/min but ranged from 2100 to 3900
ml/min. This corresponds to an average loading rate in the degassing
column of 9.7 gpm/ft and a range of 7.2-13.4 gpm/ft .
After evaporator start-up, from-30 to 90 minutes were allowed to attain
stable operating conditions and completely flush the system. Product
sampling was then begun, usually at 60-minute intervals. Since for each
operating condition one drum of feed was required, only one feed sample
was taken. One concentrated effluent sample was taken under each set of
conditions. Total operating time, once product sampling began, was usually
2 hours, but some runs as long as 5 hours were made.
Extended Aeration_P 1ant_Effluent
Since extended aeration plant effluent was the highest quality secondary
effluent available, testing was begun using this water. The effluent
was from a 9,000-gpd-capacity plant built by Chicago Pump. This plant
was fed raw sewage that had previously been degritted and ground. Flow
to the plant was maintained at a constant rate by a constant head and weir
arrangement. Excess raw sewage from the Campus Sewage Treatment Plant was
pumped to a constant head headbox and the desired amount removed over a V-
notch weir. The excess was returned to the Campus Plant. Flow to the
plant during the period that the initial batch tests were being made was
6.2 gpm (8,920 gpd). The plant had been in operation for about, four months
prior to the initial tests and the mixed liquor suspended solids in the
aeration basin had stabilized at about 4,200 mg/1.
Thirteen runs were made using extended aeration plant effluent, five each
under atmospheric and pressure conditions and three under vacuum conditions.
The results of these runs are shown in Tables 5, 6 and 7.
All products had odors strong enough to eliminate the possibility of direct
municipal reuse. These odors ranged from very slight musty for products
produced under vacuum to moderately strong fecal for products produced under
pressure conditions.
Contact^Stabilization Plant Effluent
if-
Contact stabilization plant effluent was taken from the Campus Sewage
Treatment Plant. This unit treats about 0.7 MGD of raw sewage. Perfor-
mance data for this plant are shown in Appendix I.
Effluent from this unit was taken from the final clarifier prior to chlorina-
tion. Fourteen runs were made, five each under vacuum and pressure con-
ditions and four under atmospheric conditions. The results of these runs
are shown in Tables 8, 9, and 10.
39
-------
TABLE 5
INITIAL TESTS, EXTENDED AERATION EFFLUENT,
VACUUM CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Hot Water Temp., °F
First Stage Temp.,°F
Feed Temp., °F
Feed, Before Degassing
Alkalinity, mg/1 CaC03
pH
COD, mg/1
-------
TABLE 6
INITIAL TESTS, EXTENDED AERATION EFFLUENT,
ATMOSPHERIC CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp . , F
First Stage Temp.,°F
Feed Temp.,°F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
PH 3
COD, mg/1
NH--N, mg/1
NOg-N, mg/1
Feed, After Degassing
pH
COD, mg/1
NH3-N, mg/1
NO -N, mg/1
1
640
156
24.4
226
212
96-104
27
6.5
23
1.4
7.0
5.4
36
0.35
7.5
2
630
146
23.2
226
212
106-114
27
6.5
23
1.4
7.0
5.9
32
0.37
7.3
Run No
3
630
177
28.1
230
212
88-104
34
5.7
27
0.42
6.1
28
0.31
4
655
186
28.0
229
212
90-104
40
6.3
57
0.09
9.0
6.3
65
0.08
9.0
5
640
293
45.8
238
212
80-82
50
6.7
85
0.42
8.3
6.5
74
0.43
8.3
Product
pH
COD, mg/1
NH3-N, mg/1
NO -N, mg/1
o
Concentrated Effluent
pH
COD, mg/1
NH -^ mg/1
. NO^-N, mg/l
5.5 5.7 5.9 6.0 6.1
0.0-0.6 0.5-0.7 2.6-3.2 3.8-5.0 1.8-5.
0.05 0.10 0.16 0.03 0.36
0.00 0.00
6.
41
0.
8.
3
44
1
6.
40
0.
8.
4
47
1
6.5
37
1.8
6
87
0
13
.3
.10
6.
137
0.
14
7
38
41
-------
TABLE 7
INITIAL TESTS, EXTENDED AERATION EFFLUENT,
PRESSURE CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp., °F
First Stage Temp., F
Feed Temp., °F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
pH 3
COD, mg/1
NH -N, mg/1
N03-N, mg/1
3
Feed, After Degassing
pH
COD, mg/1
NH -N, mg/1
WO^-N, mg/1
1
630
165
26.2
299
286
96-104
6.5
52
1.0
7.7
4.8
47
1.0
7.9
Run
2
635
166
26.2
299
286
104-114
6.5
52
1.0
7.7
5.6
48
0.8
7.7
No.
3
625
146
23
300
288
100-
42
6.
97
0.
9.
5.
90
0.
8.
4
625
182
.4*' 29.1
SQSi
288
115 98-118
57
5 6.5
110
40 1.5
3 7.4
9 6.2
94
33 1.4
6 7.0
5
630
219
34.8
304
288
109-120
49
6.7
85
0.42
8.3
6.5
74
0.43
8.3
Product
pH
COD, mg/1
NH -N, mg/1
N03-N, mg/1
O
Concentrated Effluent
PH
COD, mg/1
NH -N, mg/1
NO§-N, mg/1
5.5 5.5 6.2 6.8 6.3
3.1-3.2 2.8 3.0-5.1 5.5-1&8 3.3-5.0
0.15 0.22 0.30 1.2 0.59
0.01 0.02
5.9 6.1
56 52
1.4 1.4
8.4 8.5
6.5 6.0 6.4
106 104 106
0.43 1.5 0.36
12 10 12
42
-------
TABLE 8
INITIAL TESTS, CONTACT STABILIZATION EFFLUENT,
VACUUM CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Hot Water, Temp., °F
First Stage', Temp., °F
Feed Temp., °F
Feed, Before Degassing
Alkalinity, mg/1 CaCO™
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
Feed, After Degassing
pH
COD, mg/1
NH3-N, mg/1
NOg-N, mg/1
Product
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
o
Concentrated Effluent
pH t*
COD, mg/1 ^
NH -N, mg/lv
NO^-N, mg/1
O
1
630
78
12.4
140
114
89-98
70
6.9
23
5.1
5.6
29
6.4
0.5
6.9
3.6
1.6a
7.4
29
6.4
0.6
Run
2
635
88
13.9
141
113
87-93 91
82
7.1
23
6.5
0.9
5.6
20
6.8
0.4
6.7
2.5
0.14
6.8
22
7.7
0.4
No.
3
625
90
14.4
141
114
-96
82
7.1
23
6.5
0.9
6.3
19
6.6
0.5
7.4
3.2 0
0.9
7.3
22
7.3
0.5
4
645
128
19.8
146
113
92-108
93
7.2
30
9.0
0.28
6.7
35
8.5
0.31
8.4
.9-2.9
2.9
0.00
7.4
42
7.6
0.36
5
630
79
12.5
140
114
93-106
70
6.9
23
5.1
•-
7.7
24
5.8
9.1
4.0
6.9
8.2
23
5.9
This run immediately followed Run No. 5 and the high NH,,-N result may
have been due to contamination from the previous run.
43
-------
TABLE 9
INITIAL TESTS, CONTACT STABILIZATION EFFLUENT,
ATMOSPHERIC CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp.,°F
First Stage Temp., °F
Feed Temp.,°F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
PH 3
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
Feed, After Degassing
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
Product
pH
COD, mg/1
NH»-N, mg/1
NOg-N, mg/1
Concentrated Effluent
PH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
1
635
169
26.6
230
212
95-104
6.9
24
3.3
3.1
5.1
29
3.5
3.1
5.7
4.7-5.0
0.22
6.4
30
4.0
4.1
Run
2
640
144
22.5
227
212
94-106
101
7.1
68
6.6
0.17
5.7
62
0.23
7.1
2.9-3.5
0.9
6.4
64
0.41
No.
3
630
172
27.3
230
212
97-108
6.9
24
3.3
3.1
5.9
25
4.3
3.1
6.4
2.1-3.0
0.3
6.4
27
3.7
4.3
4
*>640
163
25.4
229
212
84-108
93
7.1
27
7.0
0.49
6.5
32
9.2
0.48
6.6
1.1-2.7
4.2
0.01
7.5
32
10.0
0.64
44
-------
TABLE 10
INITIAL TESTS, CONTACT STABILIZATION EFFLUENT,
PRESSURE CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp., °F
First Stage Temp., F
Feed Temp.,°F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
O
Feed, After Degassing
pH
COD, mg/1
NH -N, mg/1
N03-N, mg/1
O
Product
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
o
Concentrated Effluent
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
1
635
152
24.0
299
286
94-103
91
7.0
23
6.7
0.1
5.2
22
7.5
0.2
6.3
1.7-2.5
2.2
6.8
24
8.5
0.3
2
630
160
25.4
299
286
86-96
69
7.1
24
4.6
5.7
23
5.7
0.1
6.0
4.2-4.
4.5
8.3
28
5.3
0.9
Run No
3
630
173
27.4
300
287
102-111
91
7.0
23
6.7
0.1
6.5
22
7.2
0.2
6.9
7 4.0-4.
3.3
7.3
25
7.8
0.4
4
615
212
34.4
301
- 284
91-104
88
7.0
35
7.5
0.38
6.6
25
8.1
0.31
6.6
9 4.8-5.2
3.5
0.02
6.8
39
8.9
0.55
5
630
170
27.0
299
286
97-108
69
• 7.1
24
4.6
,.
7.2
24
6.4
1.1
6.6
1.5-3.4
9.4
6.8
27
4.0
1.5
45
-------
All products had odors sufficiently strong to. eliminate the possibility of
direct municipal reuse. These odors ranged from musty for products produced
under vacuum conditions to a rather strong, disagreeable, fecal type odor
for products produced under pressure conditions.
Trickling Filter Plant Effluent
Trickling filter plant effluent was taken from the high rate trickling
filter operated by the Campus Sewage Treatment Plant. This unit treats
about 0.25 MGD of raw sewage. Performance data for this plant are shown
in the appendix.
Effluent from this unit was taken from the final settling basin prior to
chlorination. This effluent was noticeably different £rom the two effluents
used previously in that it had a gray color and a slight odp,r. Six runs
were made, two under each of the three temperature and pressure conditions.
The results of these runs are shown in Tables 11, 12, and 13.
All products had odors as bad or worse than odors of products produced from
extended aeration effluent or contact stabilization effluent. As in pre-
vious runs, odors tended to increase with increasing operating temperature.
The products produced under high temperature and pressure conditions had
a definite hazy appearance. With this one exception, all effluents tested
under all operating conditions produced products having a crystal clear
appearance.
Results of COD analyses made on these products indicated that organic con-
tamination was significantly worse than had been experienced during pre-
vious tests using cleaner feedwaters. Since a significant break-point seemed
to have been reached with regards to product water quality, the decision was
made, to conduct no further single stage tests with poorer quality feedwater.
Activated Carbon Tests
In order to evaluate the effectiveness of activated carbon for removal of
any odors remaining after evaporation, a bench scale granular activated
carbon column was set up. The carbon selected, Nuchar WV-W, has a large
proportion of small pores and is specifically recommended for removal of
organics causing taste and odor problems in municipal water.
Since all products produced during the initial single-stage testing had
some odor, it was necessary to test examples of products produced under
all conditions. Each test consisted of continuously feeding the product
water to the carbon column at a rate of about 1 gpm/ft until several
column volumes had been flushed through the column. After this flushing,
samples were taken of the column feed and product for analysis. The opera-
tor would also make observations to determine whether or not any odor re-
mained in the water from the column. In some cases the opinions of several
people were secured regarding the odor of the product water.
46
-------
TABLE 11
INITIAL TESTS, TRICKLING FILTER EFFLUENT,
VACUUM CONDITIONS
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Hot Water Temp., °F
First Stage Temp., °F
Feed Temp., °F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
PH
COD, mg/1
NH -N, mg/1
N06-N, mg/1
3
Feed, After Degassing
PH
COD, mg/1
NH -N, mg/1
NOC-N, mg/1
3
Product
PH
COD, mg/1
NH -N, mg/1
. NO°-N, mg/1
3
Concen-fe-ated Effluent
pH 4(N
COD, mg/1
NH -N, mg/1
NOJJ-N, mg/1
Run
1
645
100
15.5
142
113
90-95
39
6.6
116
5.4
15.2
5.3
101
6.0
13.0
5.8
2.0
0.1
0.03
6.2
6.4
13.4
No,
2
640
97
15.2
143
114
95-102
39
6.6
116
5.4
15.2
5.9
127
5.5
13.4
6.6
1.4
0.6
0.00
6.5
123
' 6.3
14.2
47
-------
TABLE 12
INITIAL TESTS, TRICKLING FILTER EFFLUENT,
ATMOSPHERIC CONDITIONS
Run No.
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp. , F
First Stage Temp.,°F
Feed Temp.,°F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
i
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
" u
Feed, After Degassing
pH
.COD, mg/1
NH. -N, mg/1
N03-N, mg/1
o
Product
pH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
Concentrated Effluent
pH
COD, mg/1
NH3-N, mg/1
NO,,-N, mg/1
645
172
26.6
228
212
90-102
50
7.0
93
5.1
11.4
5.5
93
5.9
10.0
5.8
5.8-6.2
0.9
0.06
6.5
105
7.2
12.6
650
172
26.5
228
212
101-111
50
7.0
93
5.1
11.4
6.1
93
5.7
10.8
6.5
6.2-6.6
1.6
0.02
7.0
93
6.8
13.2
-------
TABLE 13
INITIAL TESTS, TRICKLING FILTER EFFLUENT,
PRESSURE CONDITIONS
Run No.
Operating Conditions
Feed Rate, ml/min
Evaporation Rate, ml/min
% Evaporated
Steam Temp . , °F Q
First Stage Temp., F
Feed Temp . , °F
Feed, Before Degassing
Alkalinity, mg/1 CaCO
PH 3
COD, mg/1
NH -N, mg/1
N03-N, mg/1
O
Feed, After Degassing
PH
COD, mg/1
NH -N, mg/1
N06-N, mg/1
0
Product
PH
COD, mg/1
NH -N, mg/1
NO^-N, mg/1
o
Concentrated Effluent
pH
COD, mg/1
NH -N, mg/1
N03-N, mg/1
630
140
22.2
298
286
95-104
65.7
7.0
90
7.5
16
5.5
84
8.0
9.4
6.0
18.5-20.9
2.4
0.28
6.5
75
10.0
11.6
630
165
26.2
300
286
103-113
65.7
7:0
90
7.5
16
6.1
80
8.8
9.4
6.3
19.6-22.6
3.7
0.17
6.6
76
10.3
11.8
49
-------
Twenty runs were made using the activated carbon column during the initial
testing phase. Six were made using products from extended aeration effluent
ten using products from contact stabilization effluent, and four using pro-
ducts from trickling filter effluent. In all cases the odors were either
eliminated or reduced to such a level that they could be detected by only
a minority of the odor test panel. The worst sample tested was a product
produced under high temperature and pressure conditions using trickling
filter effluent. After activated carbon treatment only one out of five
members of the odor test panel detected any odor at all. COD values be-
fore and after carbon treatment for this sample were 20.7 and 3.4 mg/1.
The average COD values in and out of the carbon column for the twenty runs
were 4.4 and 1.35 mg/1. Omitting the trickling filter product noted above,
these averages drop to 3.6 and 1.25 mg/1.
Ammonia Ion Exchange Tests
A bench scale ion exchange column was used to evaluate the feasibility of
removal of ammonia from the product water. This column contained clinop-
tilolite, a natural exchange material selective for ammonium ion. In each
test, the column was operated continuously until the column had been
thoroughly flushed and then samples were taken of the column feed and pro-
duct. Surface loading rates of 1.2-1.4 gpm/ft were used. A total of six
tests was made using feedwater containing from 2.5-5.0 mg/1 NJ^-N. Ammonia
in the column effluent ranged from 0.12-0.32 mg/1 NH3-N. In all cases
ammonia removal exceeded 9070 and averaged 94.470.
Aeration -Tests
Probably the most economical means of removing odors from the evaporator
product water would be aeration. However, the effectiveness of this method
was not known. To evaluate aeration for odor removal, a 13-liter sample
of product water was aerated in a 5-gal. carboy using a diffuser stone to
disperse the air. The product water used was produced from extended aera-
tion plant effluent under atmospheric pressure and initially had only a
mild odor. The sample was aerated up to 1.5 ft-Vgal. This air volume ex-
ceeds by an order of magnitude the amount normally used in aeration for
odor removal. Little, if any change in odor could be detected in the
water.
Because of the poor results from this test, no further evaluations of
aeration for odor removal were made.
Product Water Conductivity
Conductivity measurements were made on products produced under atmospheric
and pressure conditions using extended aeration effluent and contact sta-
bilization effluent as feedwater. It was found that conductivity corre-
lated closely with the amount of ammonia in the product for products with
less than 1.5 mg/1 NE^-N. Data for 31 product samples are shown in Figure 9.
50
-------
co
"i
O
M
Q
§
O
O
M
P^
M
PL,
4 .
Product Water From Initial
Tests Using Extended Aeration
and Contact Stabilization Feed.
Measured at 4-11 C To Convert
To 20°C Multiply by 1.5
0.5
i
1.0
I
1.5
I
2.0
•NH3-N, mg/1
Figure 9. Conductivity vs. Ammonia Concentration
51
-------
A least squares regression line for these data indicates a conductivity
of 1.7 x 10~6 mho/cm at 7°C for zero ammonia. Converting this to 20°C
gives a conductivity of about 2.5 x 10~° mho/cm. By comparison, distilled
water in equilibrium with air has a conductivity of about 1 x 10"" mho/cm
and a 1.0 mg/1 solution of KC1 has a conductivity of about 2 x 10 mho/cm. "
Scaling During Check-Out and Initial Testing
Although there was no planned program of sc-ale evaluation .during the check-
out and initial testing phase, the evaporator was mechanically cleaned
several times and a few measurements made on the material removed.
During the initial check-out of the system, the evaporator was operated
as a single-effect unit (3rd stage only), intermittently, for a total of
about 30 hours using tap water from the water treatment plant, Gainesville,
Florida. Typical analysis of this water is shown in Table 14. 'The upper
end of the evaporator was inspected by removing the vapor-liquid separator
and a light tan scale was observed. The evaporator was not cleaned at this
time and was operated, again on an intermittent basis, for an additional
40 hours using tap water with the pH adjusted down to 4.1-6.4 with sulfuric
acid. Durifig this period several inspections indicated that the scale was
still present but did not seem to be getting any worse. At the end of this
period (70 hours total operating time) the evaporator tube was mechanically
cleaned with a power driven wire brush and the scale retained. A total of
16.5 grams of dry scale was removed from the evaporator. Approximate analyses
on this scale indicated that it was largely calcium carbonate.
Shortly after this cleaning, the degassing equipment was installed and all
feed was degassed prior to being put into the evaporator. -The evaporator
was cleaned a second time after about 65 hours of operation using degassed
extended aeration and contact stabilization effluents. This time only 0.92
grams of scale were removed. No analyses were made on this scale.
The evaporator was cleaned a third time after the completion of the initial
single stage tests. The scale removed was inadvertently discarded before
it could be analyzed.
Long Term Runs
Based on the results of the single stage tests, it appeared that no com-
bination of effluent feedwater and evaporator operating conditions would
give a product acceptable for municipal reuse without further treatment.
Likewise, all product water appeared amenable to treatment by activated
carbon to an aesthetically acceptable quality for reuse. Since it would be
economically desirable to operate a full scale unit at as high a tempera-
ture and pressure as possible, it was decided to carry out the long term
tests under the high temperature and pressure conditions used in the
initial tests. All three evaporator effects were used.
The major purpose of these tests was to evaluate the scaling potential of
the particular feedwater being used. Since the evaporator was operated
52
-------
TABLE 14
TYPICAL ANALYSIS OF TAP WATER,
GAINESVILLE, FLORIDA
Total Hardness . 95 mg/1 as CaC03
Calcium Hardness 48 mg/1.as CaCOg
Magnesium Hardness 47 mg/1 as CaCOg
Total Alkalinity 73 mg/1 as CaCOg
Sulfate 32 mg/1 as SO »
Chloride 20 mg/1 as Cl
Total Dissolved Solids 211 mg/1
pH 8.4
53
-------
as a three-effect unit, it was possible to evaporate a greater percentage
of the feedwater and thus observe the scaling that might occur in the latter
stages of a full scale unit. After each extended run, the scale in each
stage was mechanically removed and analyzed. In addition, the overall heat
transfer coefficients for each effect were calculated as a function of time.
Thus it was possible to get some indication of scalingi during a run without
having to shut down the equipment.
In addition to information on scaling, the long term runs gave additional
product quality data over a long period of time. Also, quality data were
obtained on products from the second and third evaporator effects. Pre-
vious single-effect tests had given information only on the first-effect
product quality.
Operating and sampling procedures during the long term runs were as follows.
Fresh feed for the evaporator was prepared every 8 hours by degassjng ap-
proximately 100 gal. of fresh effluent. Thus two 55-gal. drums of feed
were used every-8 hours. Samples of the effluent being used, before de-
gassing, were taken once every 8 hours and composited for 24 hours. Sam-
ples of the degassed feed and concentrated effluent were taken every 4 hours
and composited for 12 hours. All the product water from each stage was
collected continuously and measured and sampled each hour. The product
samples were composited for 12 hours.
First Extended Aeration Effluent Run
The first long-term run was carried out using extended aeration effluent.
It was planned to operate with 300°F steam and adjust the third stage pres-
sure to evaporate about 75% of the incoming feed. The evaporator was operated
for a total of 323 hours (13 days, 11 hours). This excluded 9 hours down-
time to make minor repairs on the equipment. Other than these two brief
shutdowns, one to tighten a leaking fitting and another to repair a liquid
seal, the run was reasonably routine. Trouble was periodically encountered
with maintaining the incoming steam to the first effect at 300°F. It was
determined that there were actually two problems. First, the first-effect
pressure was too near the supply line steam pressure, and at times there
was not enough pressure differential across the steam regulator to maintain
reliably the first-effect pressure. Second, the steam trap in the conden-
sate line from the first stage was oversized and allowed a 5-10 F temperature
drop every time it opened to discharge condensate. Consequently, on the
eighth day of the run the steam temperature to the first effect was lowered
to 292°F. This gave better control, but some temperature variation was
still caused by the oversized trap. This problem had little effect on the
validity of the scaling or product quality results, but it did make it
rather difficult to determine individual heat transfer coefficients.
During the run an attempt was made to maintain a temperature drop of 32-34°F
from incoming steam to third-stage vapor-liquid separator temperature. At
a constant feed rate of 630 ml/min this gave an average product rate of
422 ml/min or 67% of the feed rate. The product rates are shown as a function
of time in Figure 10. Problems with steam control caused some variations
in product rates. However, there were other variations that could not be
54
-------
500
400 _
300 -
E-i
O
B 200
100 _
0
TOTAL
FIRST EFFECT
I i i i i i I i r i i i i i i
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
MARCH
Figure 10. Product Rates, First Extended Aeration Effluent Run
-------
accounted for by this factor. Overall, there was a slight increase in
the total product rate over the run. This increase was due to increases
in the rates from the second and third effects. These increases apparently
were the result of decreased heat losses which resulted when the slightly
damp insulation on these effects dried. There was a cycling pattern in
the output rate that occurred within the individual effects as well as with
the total. It was noted that the product rate occasionally increased
markedly immediately following some type of temporary operational upset.
Also, there were times that the concentrated effluent contained unusually
high levels of dark-colored, suspended solids.
The analytical results for samples taken during this run are shown in
Table 15.
After the completion of the run the evaporator tubes were* mechanically
cleaned and the scale retained. Results of the analysis of tffifcis scale are
shown in Table 16.
Trickling Filter Effluent Run
Based on the good results of the extended aeration effluent run, it was
decided to make an extended run under pressure conditions using trickling
filter effluent. Prior to the start of this run the oversized steam trap
in the condensate line from the first stage was replaced with a smaller
trap. This change made it possible to control the incoming steam tempera-
ture closer to 292°F. The operational and sampling procedures were the
same as were used during the first extended aeration run.
The run lasted 285 hours (11 days, 21 hours). Total downtime was one hour.
The feed rate was held constant at 630 ml/min. For the overall run the
total product rate averaged 457 ml/min or 72.6% of the feed rate. Daily
averages for the total product rate ranged from 404- to 4-97 ml/min or 64.2
to 79.0% of the feed rate. The overall trend in total product rate was down-
ward. As in the previous test, cycling in the product output rate, and
periodic increases in suspended solids in the concentrated effluent were
observed.
Analytical results for samples taken during this run are shown in Table 17.
Close control of operating temperatures made it possible to calculate ac-
curate overall heat transfer coefficients for each effect. The 12-hour
averages for these results are plotted in Figure 11. In general, the co-
efficients decreased with time, but tended to show a cyclic pattern.
After the completion of the run the evaporator tubes were mechanically
cleaned and the scale retained. As had been experienced before, most of the
scale was removed by the power-driven wire brush in less than 5 minutes.
However, there was some harder gray scale in the upper portion of the eva-
porator tube in effects two and three. Most of this scale was removed only
after considerable additional scrubbing with the wire brush. This "hard"
scale was collected and analyzed separately from the previously removed "soft"
scale. Results of the analysis of these scales are shown in Table 18.
56
-------
TABLE 15
ANALYTICAL RESULTS - FIRST EXTENDED AERATION EFFLUENT RUN
Feed
Before
Degassing
After
Degassing
Concentrated
Effluent
Products
#1
#2
#3
rr-
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Rage
Avg
Range
COD
46.5°
35.4-63.8
39.0
28.3-49.4
105
72-211
5.9
1.2-14.7
5.0
1.8-8.6
4.3
0.8-9.5
NH3-N
0.37
0.28-0.59
0.36
0.20-0.62
1.4
0.9-2.2
0.12
0.05-0,19
0.13
0.05-0.21
0.10
0.03-0.22
NOg-N
17
12-21
18
12-24
55
42-70
0.14
0.00-0.23
0.10
0.00-0.17
0.15
0.00-0.55
Total
Dissolved
Solids a
404
398-410
383
354-428
1,084
938-1,222
t
Suspended Total ,
Solids Hardness
20
15-24
16 95.3
10-20 93.6-97.8
56 317
34-101
pH
6.6
6.3-7.0
5.8
5.5-6.2
6.2
5.8-6.6
Spot checks on less than 5 samples
As CaCO™; spot check on 8 feed samples and 1 concentrated effluent sample
CA11 concentrations in mg/1 except pH
-------
TABLE 16
SCALE FROM FIRST EXTENDED AERATION EFFLUENT RUNa
Total Weight, gms
% Benzene Soluble13
% Organic0
Analysis of Fired Residue
% Soluble in 0.5N HC1
o
% Calcium as Ca
% Magnesium aseMg
% Sodium as ga
% Iron as Fe
% Sulfate as SO e g
% Phosphate as P04
1
0.83
2.5
39.7
78.1
14.3
1.4
1.0
13.5
0.0
18.5
Stage
2
1.04
3.8
53.1
65.8
11.3
1.6
0.7
1.8
5.1
26.5
No.
3
3.65
4.3
35.2
A
91. 0Q
12.4
0.7
1.3
8.0
17.7
13.5
aSamples of 0.15 to 0.25 grams were stirred in one liter of
0.5 N HC1 for several hours
Soxhlet Extraction
cBenzene soluble + loss after firing at 600°C
More severe conditions used to dissolve this sample
eExpressed as percentage of soluble inorganic scale
58
-------
TABLE . 17
ANALYTICAL RESULTS - TRICKLING FILTER EFFLUENT RUN
Total
Feed
Before
Degassing
After
Degassing
Concentrated
Effluent
Products
#1
#2
#3
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
COD
112b
84-133
109
77-140
286
202-367
22.2
7.6-35.5
12.0
7.4-23.6
9.4
4.0-15.4
NH -N
o
12.1
2.7-17
14
7 . 6-20
40
19-66
2.9
1.1-3.8
2.4
1.0-3.1
2.7
1.2-3.9
NO -N
-------
700 -i
\2
-P
EH
PQ
CQ
EH
CO
EH
£
ffi
600
o 500 -
O
o
400
300
• FIRST STAGE
0 SECOND STAGE
A THIRD STAGE
TRICKLING FILTER RUN
EXTENDED
AERATION
RUN
200
10 11 12 13 14 15 16 17
APRIL
30 1
MAY
Figure 11. Overall Heat Transfer Coefficients
-------
TABLE 18
SCALE FROM TRICKLING FILTER EFFLUENT RUN
Total Weight
% Organic^
Analysis of Fired Residue
% Soluble in 0.5N HC1
% Calcium as Cac
% Magnesium as Mg
% Sodium as Na
% Iron as Fe
% Sulfate as 50^°
% Phosphate as P0,°
"Soft"
1
5.01
52.4
79.5
7.7
0.8
2.6
19.4
7.3
18.4
Scale,
2
3.04
63.5
68.1
15.5
1.4
3.4
3.8
20.0
17.1
Stage "Hard"
3 2
6.88
54.4
80.0
20.9
0.7
0.8
4.4
38.2
8.5
0.25
35.8
96.7
21.8
0.3
0.8
0.3
54.8
13.3
Scale, Stage
3
0.94
26.9
93.3
23.0
0.1
0.6
0.0
51.5
4.2
Samples of 0.15 to 0.25 grams were stirred in one liter of 0.5 N HC1 for
several hours.
Loss after firing at 600°C
Expressed as percentage of soluble inorganic scale
TABLE 19
ACTIVATED CARBON TREATMENT OF PRODUCTS FROM
TRICKLING FILTER EFFLUENT RUN
COD, mg/1
Before Carbon
Treatment
After Carbon
Treatment
Product from Effect #1
Product from Effect //2
Product from Effect #b '
20.1
11.6
10.6
4.9
3.3
0.1
61
-------
To confirm the initial single stage test results with regard to odor re-
moval, samples of each product were treated in the activated carbon column.
The same column set-up used previously was used in these tests. The column
loading rate for all three products was 1.0 gpm/ft. The odor was completely
removed from all three products. COD results are shown in Table 19.
Second Extended Aeration Effluent Run
A reasonable explanation for the cycling in product output and heat trans-
fer coefficients observed during the first two long term runs seemed to be
that relatively soft scale was alternately building up and then flaking off
the evaporator tube walls. The periodic increases in suspended solids in the
concentrated effluent were further evidence that this was*k)ccurring. To test
further this theory, a second run using extended aeration effluent was made.
This test was limited to 5 days. If, after 5 days, quantities of scale
were found well out of proportion of what would be expected based on the
previous 14-day test, this would further confirm that scale was alternately
building up and flaking off.
The operational and sampling procedures used in this test were identical to
those used in the first two runs. However, from the beginning of the run
there was an abnormally high product rate from the third effect. The liquid
seal between the second and third effects appeared to be leaking, allowing
a small amount of vapor from the second vapor-liquid separator to enter the
bottom of the third effect. It was felt that this would have little if any
effect on the scaling rate and the decision was made not to shut down for
repairs. However, the problem appeared to worsen with time, and on the third
day of the run the unit was shut down for 5 hours and the liquid seal repaired.
Other than the problem with this seal, the run was rather routine. The eva-
porator was operated for a total of 113 hours (4 days, 17 hours). Total
downtime was 5 hours.
Analytical results for samples taken during this run are shown in Table 20.
Heat transfer coefficients for the three effects were calculated for the
latter part of the run after the liquid seal was repaired. These results,
calculated as 12-hour averages, are shown in Figure 11.
During the latter part of the run, three complete sets of samples were
taken for bacteriological analyses. Results of these tests are shown in
Table 21. In addition to the coliform tests, a test designed to check bac-
terial s-terility was run on all samples. In this test the sample is in-
cubated at 35°C for seven days in nutrient broth. This test revealed that
all product samples from the first and second effects were sterile. All
other samples showed some biological activity.
After the run was completed the evaporator was mechanically.cleaned and
the scale retained. Results of analysis of this scale are shown in Table
22.
62
-------
TABLE 20
ANALYTICAL RESULTS - SECOND EXTENDED AERATION EFFLUENT RUN
Feed
Before
Degassing
After
Degassing
Concentrated
Effluent
Products
#1
#2
#3
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
COD
74 b
38-97
35
29-43
136
76-203
7.4
3.4-13.2
4.6
1.4-8.2
4.0
0.9-7. 5
NH -N
o
0.39
0.25-0.52
0.37
0.28-0.51
2.3
1.6-3.7
0.08
0.00-0.11
0.12
0.06-0.17
0.08
0.02-0.11
NO -N
O
14
14-15
14
14-16
79
62-98
0.09
0.05-0.13
0.13
0.04-0.26
0.00
0.00
Total
Solidsa
335.
328-341
1,643
1,265-2,075
_____
pH
6.8
6.7-7.0
5.7
5.5-6.0
5.8
5 .,5-6. 7
:;;
—
—
Spot check of 5 samples each
All concentration in mg/1 except pH
-------
TABLE 21
BACTERIOLOGICAL TESTS - SECOND EXTENDED AERATION
EFFLUENT RUN
Sample
Time
Date
Total Coliform
MPN/100 ml
Fecal Coliform
MPN/100 ml
Extended Aeration
Plant Effluent
Feed
Product No. 1
Product No. 2
Product No. 3
Concentrated Effluent
Extended Aeration
Plant Effluent
Feed
Product No. 1
Product No. 2
Product No. 3
Concentrated Effluent
Extended Aeration
Plant Effluent
Feed
Product No. 1
Product No. 2
Product No. 3
Concentrated Effluent
2400
2400
2330
2330
2330
2400
0400
0400
0330
0330
0330
0400
0800
0800
0730
0730
0730
0800
4/30/70
4/30/70
,4/30/70
4/30/70
4/30/70
4/30/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
5/1/70
1,300,000
< 2.0
< 2.0
< 2.0
< 2.0
<2.0
490,000
8.0
< 2.0
< 2.0
< 2.0
< 2.0
490,000
< 2.0
< 2.0
^ 2.0
^ 2.0
^ 2.0
790,000
< 2.0
< 2.0
< 2.0
< 2.0
< 2.0
330,000
2.0
< 2.0
< 2.0
< 2.0
< 2.0
109,000
< 2.0
< 2.0
< 2.0
< 2.0
< 2.0
64
-------
TABLE 22
SCALE FROM SECOND EXTENDED AERATION EFFLUENT RUN&
Stage No.
123
Total Weighl
% Organic
b, gins
0.92
30.7
2.54 *
36.4
1.89
28.4
Analysis of Fired Residue
% Soluble in 0.5N HC1
% Calcium as Ca °
% Magnesium as Mg
% Sodium as Na c
% Iron as Fe c c
% Sulfate as SO,
% Phosphate as PO
70.2
13.0
0.8
0.6
14.2
0.9
17.8
87 .,7
19.2
0.8
0.7
2.0
30.1
19.2
87.9
24.4
0.6
0.9
'2.0
32.4
11.9
Samples -of 0.15 to 0.25 grams were stirred in one liter of 0.5 N HC1
for several hours
Loss after firing at 600°C
Expressed as percentage of soluble inorganic scale
65
-------
Three Effec^A^gnia^Dis^tribu^ion Tests
Some indication of ammonia carry-over as a function of feed pH and eva-
porator operating conditions was gained in the single-effect tests. How-
ever, additional information was needed regarding the distribution of
ammonia between product and concentrated effluent in subsequent evaporator
effects.
To obtain this information, a series of tests was made utilizing all three
evaporator effects. Degassed contact stabilization effluent was used as
the feedwater. The pH of the feed was varied from 6.0 to 8.7 and the eva-
porator was operated at least 4 hours under each pH condition. Product
samples were taken from the last 2-hour's production and composited. Feed
and concentrated^effluent samples were taken at the end of each run. Feed
rate was held constant at 630 ml/min and the total evaporation rate varied
between 447 and 545 ml/min or 71.0-86.5% of the feed rate.
The analytical results for samples taken during these runs are shown in
Table 23.
66
-------
TABLE 23
ANALYTICAL RESULTS - THREE EFFECT AMMONIA DISTRIBUTION TESTS
Feed
After Degassing
Concentrated
Effluent
Products
#1
#2
#3
pH
NH -N, mg/1
COD, mg/1
pH
NH -N, mg/1
C0§, mg/1
NH3-N, mg/1
COD, mg/1
NH--N, mg/1
COD, mg/1
NH5-N, mg/1
COD, mg/1
1
6.0
8.3
35
5.8
35
147
2.7
16.3
1.9
3.3
2.2
4.5
2
6.7
8.0
49
18
184
5.0
4.8
2.9
8.0
2.2
4.5
Test No.
3
7.3
8.2
53
16
160
4.9
5.6
2.9
0.6
2.1
1.8
4
7.6
2.8
24
7.3
2.8
57
1.7
1.0
1.0
0.3
0.8
2.6
5
8.0
9.2
26
5.9
21
137
6.3
4.7
3.5
3.2
2.6
0.9
6
8.7
2.3
23
6.9
3.4
65
1.4
4.4
1.0
1.6
0.8
0.4
-------
DISCUSSION OF RESULTS
Product Quality
The first objective of this work was to define the relationships between
product water quality, feedwater quality, and evaporation conditions.
The second objective involved defining the relationships between the above
variables and post evaporation polishing. The three primary parameters
used to measure product quality were ammonia content, COD, and odor. Of
secondary importance in measuring product quality were conductivity and
nitrate content. Product quality with regard to each of these parameters
is discussed separately below.
Ammonia is an undesirable constituent in the product water because it con-
sumes more than six times its weight of chlorine and'produces chlpramines
which have much lower disinfecting efficiency than free chlorine. Pre-
vious laboratory investigations had indicated that only by going to extremely
acid or basic conditions could ammonia distribution between the evaporator
feedwater and the product water be controlled.' In the work reported here,
feedwater pH values between 5.1 and 8.7 were used in both single-effect and
three-effect tests. The test results show that for a given pH value, the
fraction of ammonia carrying over to the product increases as operating
temperature and pressure increase. This is consistent with the theoretical
distribution of ammonia and ammonium ion as a function of temperature.
As can be seen from the results of the single-stage tests using contact
stabilization effluent and trickling filter effluent, under vacuum conditions,
ammonia in the product was less than 1 mg/1 for feed pH values up to around
6.5. Under atmospheric conditions, feed pH had to be less than 6 to control
ammonia in the product to less than 1 mg/1. It should be noted that these
results are for the first-effect product only. In all the single-effect
tests under vacuum and atmospheric conditions using contact stabilization
or trickling filter effluent as feedwater, the pH of the concentrated ef-
fluent was significantly higher than the feedwater pH. This would mean
that under the multieffect conditions that would be used in full scale
applications, ammonia could be expected to contaminate the product water in
the later stages.
Under pressure conditions, there was more than 1 mg/1 ammonia in all products
from contact stabilization or trickling filter feedwater for pH values down
to 5.1. Tests made while operating the evaporator as a three-effect unit
showed that this contamination persisted in the second and third effects.
Raising the feedwater pH as high as 8.7 did not eliminate contamination in
the second and third effects.
There appear to be np easy ways of controlling ammonia contamination of the
product water short of removing the ammonia from the feedwater. This could
be accomplished by using a highly nitrified effluent as feedwater.
69
-------
The tests reported here and the work of other investigators ' show
that ammonia could be removed from the product water by ion exchange using
clinoptilolite. If the product water were essentially free of cations other
than ammonium, ordinary cation resins could also be used for this purpose.
Chemical Oxygen Demand
The COD test was used as an indicator of the organic content of the pro-
duct water. For the products from extended aeration effluent and contact
stabilization effluent, the GOD was generally less than 5 mg/1 for all
evaporator operating conditions. However, when trickling filter effluent
was used, the product COD increased significantly as operating temperature
and pressure increased. Thus it appears, that because of problems with
organic carry-over to the product, trickling filter effluent is a signifi-
cantly poorer feedwater than either of the other two effluents tested.
**
The COD of products from all three effluents at all operating conditions
was substantially reduced by treatment with activated carbon. However,
for product that was produced under pressure conditions from trickling
filter effluent, the final COD after activated carbon treatment was
roughly equivalent to the COD of the other products before carbon treatment.
Odor
Odor in the product water would be a particularly critical factor in any
full scale application of direct wastewater reuse because of the obvious
psychological barriers. In these tests, no combination of wastewater feed
and evaporation conditions produced completely odor free water. Product
odor tended to increase in intensity and disagreeableness as evaporator
temperature and pressure increased. Also, the more completely treated ef-
fluents tended to produce less odorous products.
Activated carbon was effective in removing the odor from all products. Aera-
tion was ineffective in removing product odors.
There was speculation regarding the effectiveness of the degassing operation
in eliminating odorous compounds. Based on the relatively mild odor of
the degasser condensate and the odor of products from the degassed feedwater ,
the degassing operation seems to have little effect on product odor.
Conductivity was measured on some of the products from the initial testing.
As shown in Figure 9, conductivity was generally related to the ammonia
content of the water. However, extrapolation to zero ammonia shows the
water was of extremely high quality with regard to inorganic salts.
Nitrate
Nitrate in the product water was used primarily as an indicator of physical
carry-over to the product, i.e. incomplete separation of vapor from liquid
70
-------
in the vapor-liquid separator. In practically all cases, nitrate test
results showed that physical carry-over was less than 0.570 of the in-
coming feedwater and in many cases much less than this.
Since it can be shown that product waters were not significantly con-
taminated by physical carry-over, the organic contamination that did occur
resulted from evaporation of volatile organics and subsequent condensation.
Bacteriological Tests
The bacteriological tests generally confirm that water safe for human con-
sumption from the standpoint of pathogen content can be produced by eva-
poration. In all cases, product water from effects one and two was sterile.
Considering this, it appears likely that the non-sterility of the product
water from effect three was due to contamination in the final condenser
and receiver system. In any case, the product water would be further
treated by chlorination before use.
Evaporator Scaling
The third objective of this work was to define the relationships between
feedwater quality, evaporator operating conditions, and scaling or fouling
of the evaporator heat transfer surfaces.
The first indication of the potential seriousness of calcium carbonate
scaling came during the initial check-out of the evaporator. Tap water
that had not been degassed proved to be ex?:remely scale forming. The
16.5 grams of scale removed from the third effect after the first 70 hours
of operation was more than the total amount of scale removed from that
effect during the entire remainder of the testing program. Thus it appears
that degassing of evaporator feedwater to remove carbonates is essential.
The-results of the three long term runs indicate that scaling from trickl-
ing filter effluent is markedly worse than scaling from extended aeration
effluent. These are discussed separately below.
>y.-Extended^ Aeration JE£ f luent
Two long t$;Fm runs, one lasting about 13 days and the other lasting about
5 days, were made using extended aeration effluent. The purpose of the
second run was to further investigate the possibility that scale was al-
ternately building up on and flaking off the evaporator heat transfer
surfaces. The results of the second run seem to confirm this. The amount
of scale in effects one and two after 5 days was more than had been removed
from these two effects at the end of the 13 day run. These results, the
cycling effects observed in product output rates and heat transfer co-
efficients, the periodic increase in suspended solids in the concentrated
effluent, and the occasional increases in product rates observed after
operational upsets, alii indicate that the scale was lightly adhering and
would periodically flake off. Thus it appears that scaling problems with
extended aeration effluent would be minor.
71
-------
Analyses of the scale showed it to be about one-third to one-half organic.
The inorganic portion contained significant quantities of calcium, iron,
sulfate, and phosphate. It is probable that the iron came from the mix-
ing tank and associated piping when the effluent pH was lowered for de-
gassing. With proper materials of construction in this part of the process,
iron should not be a problem. Calcium sulfate scale was formed only in the
second and third effects. This reflects the increase in concentration
caused by evaporation,, It should be noted that, during the long term tests,
conditions in the latter two effects were more severe than would occur in
full scale application of this process. In a full scale plant, the evapora-
tion of the same fraction of the incoming feed would occur over a larger
number of effects and hence the temperature of the more concentrated liquid
would be appreciably lower than the 260-270°F used in these tests. At the
lower temperature, calcium sulfate would be less likely to form scale.
Regarding calcium sulfate scale, consideration should be given to the fact
that the pH adjustment prior to degassing was made with sulfuric acid.
Acid additions during the two runs ranged from 11 to 35 mg/1 804. If higher
evaporation temperatures are to be attempted, this amount of extra sulfate
could become critical.
Significant amounts of phosphate were found in the scale from all three
effects. Because phosphate compounds tend to form loosely adhering sludges
rather than hard scales, a high phosphate level should be desirable.
Scalingjby Trickling. Filter Effluent
During the 12-day run using trickling filter effluent, the scale formed was
more undesirable, both with regards to quantity and quality. Significantly
more scale formed in all three effects, and the "hard" scale formed in ef-
fects two and three was very difficult to remove.
Analyses of the "soft" scale revealed that it was more than half organic,
whereas the extended aeration effluent scale was generally less than half
organic. The inorganic portion of the "soft" scale was very similar in
composition to the extended aeration effluent scale. The "hard" scale re-
moved from effects two and three contained a much greater inorganic fraction
that appeared to be largely calcium sulfate. This finding was consistent
with the physical characteristics of the material. The greater problem
with calcium sulfate experienced with this feedwater may have been caused
by the much larger amount of sulfuric acid that was required to lower the
pH to 4.9 prior to degassing than was used in the extended aeration effluent
runs. An average of 103 mg/1 804 was added to the trickling filter effluent
prior to degassing.
Heat Transfer Coefficients
The heat transfer data in Figure 11 show that during the trickling filter
effluent run, the heat transfer coefficients in effects one and three
generally declined and were higher than the rather stable coefficient in
effect two. During the extended aeration run, the coefficients decreased
72
-------
from effect one to effect three. It is likely that during the trickling
filter run the coefficients for effect three were high because .of a small
leak of steam from the second effect through the liquid seal. It was
immediately after this run that this leak really became noticeable and
was repaired. After the repair, the coefficient distribution seen in the
second extended aeration run resulted.
Theoretically, there should be a slight drop in heat transfer coefficient
from one effect to the next due to decreasing temperature. However, the
differences in coefficients seen during the extended aeration run seem
to be unrealistically high. It is probable that the coefficients cal-
culated for effects two and three are lower than the actual values. This
is because the steam side temperature in these effects was measured in the
vapor-liquid separator of the previous effect. Any temperature drop that
occurred between the vapor-liquid separator and the steam jacket of the
following effect was not accounted for in the calculations. The calcula-
tion for the coefficient in the final effect involved a correction for
heating the feed to the boiling point (see Appendix II). Although the
correction was made as accurately as possible, it may have contributed
a small error to the coefficients,, Hence, the coefficients shown in
Figure 11 are of more value as relative indicators of gain or loss of
heat transfer efficiency than as absolute measures of heat transfer co-
efficients.
73
-------
APPENDIX I.
TEST RESULTS - CAMPUS SEWAGE TREATMENT PLANT
Effluent from the High Rate Trickling Filter and the Contact
Stabilization unit of the Campus Sewage Treatment Plant was used in
this study. Shown in Tables 24 and 25 are results of tests made in
the campus plant before, during, and after the time effluent from this
plant was being used.
TABLE 24
CAMPUS SEWAGE TREATMENT PLANT TEST RESULTS,
HIGH RATE TRICKLING FILTER
Date
2-18-70'
2-19-70
2-25-70
2-26-70
3-4-70
3-5-70
3-11-70
3-12-70
4--1-70
4-2-70
4-8-70
4-9-70
4-15-70
4-22-70
4-23-70
Average ^
Alkalinity Chloride BOD
as CaC03 as Cl
In Out In Out In Out
184 47
154 82 142 140
212 38
160 84 146 144
179 36
164 76 146 126
208 51
148 74 210 192
162 37
148 88 188 168
198 36
142 72 134 144
190 40
150 84 132 140
; 152 80 157 151 191 41
i*-
Suspended
Solids
In Out
168 „ 63
166 52
181 59
136 18
182 56
125 44
177 50
__ —
162 49
All quantities expressed as mg/1
Initial tests were made during the period 2-27-70 to 3-3-70.
The long term run was made 4-6-70 to 4-17-70.
75
-------
TABLE .25
CAMPUS SEWAGE TREATMENT PLANT TEST RESULTS,
CONTACT STABILIZATION PLANT
Date
2-4-70
2-5-70 .
2-11-70
2-12-70
2-18-70
2-19-70
2-25-70
2-26-70
3-4-70
3-5-70
3-11-70
3-12-70
Average
Alkalinity Chloride BOD Suspended
as CaC03 as Cl -^ Solids
In Out In Out Ir> Out In" Out
A*
202 17.7 152 15.0
120 118 128 120
168 10.4 179 15.7
108 112 132 140
185 11.7
136 114 134 128
184 9.5 139 14.5
140 120 136 126
187 7.1 165 8.7
128 116 142 120
209 6.8 165 11.2
104 106 162 138
123 114 139 129 189 10.5 160 13.0
All quantities expressed as mg/1
The initial tests were made during the period 2-11-70 to 3-5-70.
76
-------
APPENDIX II.
CALCULATION OF HEAT TRANSFER COEFFICIENTS
In calculating overall heat transfer coefficients for the evaporator
the following procedures were used.
For the first effect, where the feed entered considerably below
its boiling temperature, two coefficients were involved. The first
coefficient was for heat transfer from condensing steam to water. The
second coefficient was for heat transfer from condensing steam to boiling
water and theoretically should be much larger than the first coefficient.
To solve for the two coefficients it was necessary to operate the
evaporator at two slightly different steam temperatures. Assuming that the
coefficients were constant over the temperature range used, the two
coefficients were calculated as follows.
Condition 1 Condition 2
Steam Temperature T T
Dl S2
Boiling Temperature T T
Bl B2
Feed Temperature I™ Tp?
Feed Rate, gpm F F
1 ^
Product Rate, gpm P, P
Heat transferred in the heating the feed to boiling was designated
2 = (F) (8.34) (60) (TR-T ) BTU/hr
h O -r
-------
If the total area for heat transfer was A and the area used for
heating the liquid to its boiling point was A^
QV, = UuA
h h h
where U = overall heat transfer coefficient ftyr heating
h
AT = log mean temperature difference
Li
-------
This was done for several sets of conditions, and it was found
that UR was consistently about three times as great as U
Using this ratio of U to U , the coefficients can be solved for
at any single condition as follows:
D =
_
h 3 (Ah) (ATj~) (15)
U =
B (A-A )(AT) (16)
h
Substituting and simplifying
(3)(Q.J(AT)(A)
A — *•!•
(QB)CATL) + (3) Qh (AT) (l?)
Substituting A^ in equation (15) or (16) gives U....
11 B
For the second and third effects where the liquid enters at very
near its boiling point, the assumption was made that only heat transfer
from condensing steam to boiling water was occurring and
Q_
(A) (AT)
(18)
To save time, all calculations were made using this procedure on
an IBM 360 computer.
79
-------
LIST OF REFERENCES
1. Hickman, K. C.' D. Jour. AWWA. J55: 1120 (1963)'.
2. Neale, J. H. U. S. Public Health Service Publication No. 999-WP-9,
AWTR-7 (1964).
3. Public Health Service Publication No. 999-WP-24, 52 (1965).
4. Water Pollution Control Research Series Publication No. WP-20-
AWTR-19, U. S. Department of the Interior, 48 (1968).
5. Brunner, C. A. Private communication, Dec. 19, 1968.
6. Stephan, D. G. Private communication, Sept. 10, 1968.
7. O'Connor, B. , et_ al. Jour. WPCF. 39: R25 (1967).
8. Gerster, J. A. U. S. Public Health Service Publication No.
999-WP-6, AWTR-6 (1963).
9. Stephan, D. G. A.I.Ch.E. - I. Chem. E. Symposium Series, No. 9,
26 (1965).
10. Stephan, D. G. Oivil Eng.. 55; 46 (1965).
11. Zuckermann, M. M. and Molof, A. H. Jour. WPCF. 42: 437, (1970).
12. Bunch, R. L., Earth, E. F. and Ettinger, M. B. Jour. WPCF, 33: 122,
(1961).
13. Public Health Service Drinking Water Standards, Revised 1962, Public
Health Service Publication No. 956, U. S. Dept. of Health,
Education, and Welfare, (1962).
14. Taylqr,, F. B., Eageti, J. H., and Maddox, F. D. Jour. AWWA.
6pj 764, (1968).
15. Jour. AWWA. 60: 1317, (1968).
16. Jour. AWWA. 61: 133, (1969).
17. Handbook of Chemistry and Physics. 54th Ed., The Chemical Rubber
Co., D-79, (1964).
18. Freiser, H. and Fernando, Q. Ionic Equilibria in Analytical
Chemistry. John Wiley, and Sons, Inc., New York, N. Y., 45, (1963)
81
-------
19. Bates, R. G. and Pinching, G. D. J. Am. Chem. Soc.. 72: 1393, (i960),
20. International Critical Tables. Vol. VII, McGraw-Hill Book Company,
New York, N. Y., 240, (1930).
21. Clifford, I. L. and Hunter, E. J. Phys. Cheml. 37: 101, (1933).
22. Sherwood, T. K. Ind. Eng. Chem.. 17: 745, (1925).
23. Perman, J. J. Chem. Soc. (London), 83: 1168, (1903).
24. Standard Methods for the Examination of Water and Wastevater. 12th
Ed., American Public Health Association, Inc., New York, N. Y.
(1965).
**.
25. Betz Handbook of Industrial Water Conditioning. Gth Ed., Bfetz
Laboratories, Inc., Philadelphia, Pa., (1962).
26. Powell, S. T. Water Conditioning for Industry. McGraw-Hill Book
Company, Inc., New York, N. Y., (1954).
27. Nordell, E. Water Treatment for Industrial and Other Uses. 2nd Ed.,
Reinhold Publishing Corp., New York, N. Y., (1961).
28. 'Langelier, W. F. Jour. AWWA, 38: 169 (1946).
29. Langelier, W. F. Jour. AWWA. 3_3: 179 (1946).
30. Dye, J. F. Jour. AWWA. 44:. 356 (1952).
31. Dye, J. F. Jour. AWWA. 50: 800 (1958).
32. Weber, W. J. and' Stumm, W. Jour. AWWA. 55: 1553 (1963).
33. McCoy, J. W. Chemical Analysis of Industrial Water. Chemical
Publishing Company, New York, N. Y. (1969).
34. Linke, W. F. Solubilities of Inorganic and Metal-Organic Compounds.
D. Van Nostrand Co., Inc., Princeton, N. J., 661 (1958).
35. Booth, H. S. and Bidwell, R. M. J. Am. Chem. Soc.. 72: 2567 (1950).
36. McCabe, W. L. and Smith, J. C. Unit Operations of Chemical
'Engineering. McGraw-Hill Book Company, Inc., New York, N. Y. ,
439 (1956).
37. Pelezar, M. J., Jr., and Reid, R. D. Microbiology. 2nd Ed.
McGraw-Hill Book Company, Inc., New York, N. Y., 300 (1965).
38. Standiford, F. C., Chem. Eng., 70: 157, Dec. 9, 1963.
39. Kahn, L. and Brezenski, F. T. Envir. Sci. Technol., 1: 492 (1957).
82
-------
40. Murphy, J. and Riley, J. P. Analyt. Chim. Acta. 27: 31 (1962)
41. Water Treatment Plant Design, American Water Works Association,
Inc., New York, N. Y. (1969).
42. Mercer, B. W. et al. Jour. WPCF. 42: R95 (1970).
43. Airles, L. L. Proc. 15th Pacific Northwest Ind. Waste Conf.,
Washington State University, 135 (1967).
83
-------
1
Access/on Number
w
2
Subject Field & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
j. I Organization
University of Florida, Gainesville, Florida
Tttlo
FEASIBILITY OF TREATING WASTEWATER BY DISTILLATION,
10lA<
ithor(a)
Sullivan, James
Singley, Edward
j & \ **
17040 DNM 02/71
2| /Vote
221citation
Final Report, 94 pages, 25 tables, 11 figures
23
Descriptors (Starred First)
*Water reuse, *Wastewater reclamation, *Long-tube vertical distillation,
Wastewater treatment, Water pollution control, Sewage treatment, Tertiary treatment
25
Identifiers (Starred First)
27
Abstract
The technical feasibility of evaporation of municipal sewage treatment plant effluent for
the purpose of water reuse was investigated. The equipment used was a three-effect long
tube vertical (LTV) evaporator. The objectives of the research were to determine the
effects of feedwater quality, and evaporation conditions on product water quality, post
evaporation polishing, and evaporator tube scaling. Feedwaters tested in the evaporator
included extended aeration effluent, high rate trickling filter effluent, and contact
stabilization effluent. All feedwater was acidified and vacuum degassed.
Results showed*'that an odor free product could not be produced from any of the three
feedwaters under any operating condition from 112°F to 290 F. The odor intensity
increased as the evaporator operating temperature and/or the chemical oxygen demand of
the feedwater increased. In all cases, post treatment with activated carbon removed all
odors. Aeration would not remove all odors. Product produced under pressure conditions
using trickling filter feed contained significantly more organic contamination than any
other products. Product contamination by ammonia could not be controlled by adjusting
the pH of the feed in the range 5.1 to 8.7. However, ammonia in the product water was
removed by ion exchange. ,
Abstractor
^ A. JBrunner
ItiNtitution
Environmental Prntp^t
Inn Agency. AWTRT.
WR:I02 (REV. JULY 1969)
WRSI C
SEND, WITH COPY OF DOCUMENT. TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
GPO: 1970 - 4O7 -691
------- |