United States Industrial Environmental Research EPA-600/ 7-79-220
Environmental Protection Laboratory September 1979
Agency Research Tnsngte Park NC 277* 1
Assessment of Three
Technologies for the
Treatment of Cooling
Tower Slowdown
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide-range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-220
September 1979
by
E. H. Houle, A. N. Rogers, M. C. Weekes,
S. C. May and V. C. Van der Mast
Bechtel National, Inc.
P. 0. Box 3965
San Francisco, CA 94119
Contract No. 68-02-2616
Task No. 8
Program Element No. INE624A
EPA Project Officer: Michael C. Osborne
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
DISCLAIMER
This report has been reviewed by the Emissions/Effluent Technology Branch,
Utilities and Industry Power Division, U.S. Environmental Protection Agency,
and approved for publication. Approval does not signify that the contents
/
necessarily reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
CONTENTS
Page
Disclaimer ħħ
Foreword vi
Acknowledgments vii
1. Introduction 1-1
2. Conclusions 2-1
3. Recommendations 3-1
4. Analysis and Comparison of Processes 4-1
4.1 Reverse osmosis 4-4
4.2 Vapor compression evaporation 4-23
4.3 Vertical tube foaming evaporation 4-35
4.4 Overall plant water balance ' 4-46
4.5 Comparison of processes 4-52
5. Test Plan for VTFE Shakedown Tests 5-1
References R-l
Appendices '
A. Basic water chemistry A-l
B. Converting units of measure B-l
iii
-------
FIGURES
Number Page
1 Reverse osmosis pretreatment process flow diagram 4-5
2 Photograph of portion of large seawater RO plant 4-8
3 Reverse osmosis process flow diagram 4-10
4 Plot plan for RO option 4-18
5 Effect of energy cost on RO treatment costs 4-20
6 Effect of pond cost on RO treatment costs 4-21
7 Photograph of a commercial VC plant 4-24
8 Vapor compression evaporation module 4-26
9 Plot plan for vapor compression evaporation option 4-31
10 Effect of energy cost on VC treatment costs 4-33
11 Effect of pond cost on VC treatment costs 4-34
12 Vertical tube foaming evaporation option downflow mode 4-38
13 Plot plan for vertical tube foaming evaporation option 4-41
14 Effect of energy cost on VTFE treatment costs 4-43
15 Effect of pond cost on VTFE treatment costs 4-44
16 Comparison of flows with and without recycle of purified
water using Water A and reverse osmosis 4-48
17 Comparison of flows with and without recycle of purified
water using Water B and reverse osmosis 4-49
18 Comparison of flows with and without recycle of purified
water using Water A and distillation 4-50
19 Photograph of VTFE pilot plant 5_2
I [
A-l Conversion from alkalinity or Ca to the respective p
values A-3
iv
-------
TABLES
Number Page
1 Comparison of Processes 2-2
2 Water A: Concentration of Water Limited by Silica Scaling 4-2
3 Water B: Concentration of Water Limited by Calcium Sulfate
Scaling 4-2
4 Water A: RO Permeate and Reject 4-14
5 Soda Ash and Sludge Handling Equipment for Waters A and B 4-15
6 Water B: RO Permeate and Reject 4-16
7 Summary of Design Criteria for RO Option 4-17
8 Effect of Recovery and Product Purity on System Design 4-53
9 Comparison of Costs and Energy Consumption 4-54
10 Overall Comparison of Processes 4-55
A-l Values of pK' and pK' at 25 C for Various Strengths and of the
Difference (pK' - pK') for various temperatures
-------
FOREWORD
The objective of this study is to examine methods for reducing the volume of
blowdown from cooling towers in order to decrease the cost of ponding or dis-
posal of the blowdown while at the same time recovering water of a quality
capable of reuse at some point within the plant.
This project was conducted under the sponsorship of the Industrial Environmental
Research Laboratory of the U.S. Environmental Protection Agency under Contract
No. 68-02-2616, entitled "Monitoring of the Vertical Tube Foam Evaporation
Demonstration (VTFE-D) and the Assessment of Various Technologies for the Treat-
ment of Cooling Tower Blowdown." The assigned work scope embraces two tasks:
1) Monitor progress, assist in formulating test plan, prepare
rough cost evaluation of VTFE-D
2) Compare economic and energy efficiency merits of VTFE, reverse
osmosis, and vapor compression evaporation.
VI
-------
ACKNOWLEDGMENTS
Several equipment manufacturers kindly contributed case histories from the com-
mercial use of their plants and equipment and operating costs.
Information on the performance and costs of commercial vertical tube evaporator
plants was supplied by Mr. R. H. Hedrick of the Goslin-Birmingham Division of
Envirotech, Mr. Don Nelson of Struthers Scientific and International Corporation,
and Mr. Malcolm Coston of the Swenson Division of the Whiting Corporation.
Technical and economic data on vapor-compression evaporators were furnished by
Mr. Wayne E. Springer of Resources Conservation Company.
Cost and performance data on reverse osmosis systems were supplied by Mr. D. C.
Brandt of DuPont's Permasep Division, Ms. Mary Jenkins of Dow Chemical Company's
Membrane Division, Ms. Amy Knapp of Polymetrics, and Mr. I. Nusbaum of UOP's
Fluid Systems.
Dr. H. H. Sephton, Principal Investigator of the University of California Sea
Water Conversion Laboratory and the inventor of the VTFE concept, cooperated in
the preparation of the test plan for the VTFE shakedown tests and provided
details and data for this report.
VII
-------
SECTION 1
INTRODUCTION
This study, performed under Contract No. 68-02-2616 of the Emissions/Effluent
Technology Branch of the U.S. Environmental Protection Agency, was directed
toward the concentration of cooling tower blowdown of power plants.
Lack of adequate supplies of cooling water and increasing concern over thermal
pollution of streams and lakes have forced the management of many power stations,
especially in western states, to consider replacing once-through cooling systems
with cooling towers. To conserve water, dry cooling towers can be used, but at
the expense of increasing turbine back pressure, which lowers power plant effi-
ciency. Consequently, either the wet or the wet-dry combination cooling tower
is preferred. For either of these choices, evaporating a fraction of the water
in the coolant cycle produces a saline blowdown stream. This stream, in the
context of a zero discharge policy being more seriously considered at the
present time, may not be discharged into surface water supplies. Ponding of
these aqueous wastes is costly, because of the necessary land area and because
of the expensive liners required to prevent seepage into the soil during pro-
longed storage.
It is apparent that the volume of material to be ponded (and therefore the cost
of ponding) can be reduced by concentrating the aqueous waste stream discharged
from cooling towers. Water can be recovered from the cooling tower blowdown
waste stream to yield a product that could be used in several locations in a
power plant. This could make the plant less dependent on fresh water supplies,
an important consideration in some water-short regions of the United States.
Possible uses are for the scrubbers of fossil fueled plants, ash sluicing in
coal burning plants, pump gland seal water, boiler makeup, and recycle to the
cooling tower loop (one of the options considered in this study).
1-1
-------
Several methods are available for treating cooling tower blowdown streams,
including:
Reverse osmosis
Vapor compression evaporation
Vertical tube foaming evaporation
Conventional softening
Multistage flash distillation
Thermal softening
Because of budgetary restrictions, it was agreed that this study would be
limited to three of the above techniques. The selection of the candidate pro-
cesses was based on the fact that the desalting-related technologies produce a
relatively pure product which could generally be used as a recyle feed at a
number of points in the plant; the softening techniques produce a product with
more restricted uses, due to the residual dissolved salts. This fact, and the
fact that multistage flash distillation technology is well known, led to an
agreement to direct the present study toward the first three of these processes.
The success of these processes or any other technique for waste water concentra-
tion depends on preventing excessive scale formation through controlling the
chemistry of the feed stream. The significance of scaling and the need to avoid
scale formation is discussed under the respective processes. For a brief discus-
sion of basic water chemistry, please refer to Appendix A.
In the following sections, each of these processes is examined from the stand-
point of technology, operating experience, and cost. The processes are then com-
pared, with a discussion of their relative merits and shortcomings. Finally,
there is a discussion on the advantages of a combined process in which the reject
stream from a reverse osmosis plant is further concentrated by means of an
evaporator-crystallizer.
1-2
-------
Bechtel cooperated in the planning of shakedown tests on the pilot plant vertical
tube foaming evaporator to be performed at the Sea Water Conversion Laboratory of
the University of California. Details of the test plan are included in this
report.
1-3
-------
SECTION 2
CONCLUSIONS
Table 1 compares the three candidate processes for the concentration of cooling
tower blowdown. It'appears at present that no process is definitely superior to
the other two in all respects. Future pilot plant work and industrial experience
may alter this conclusion. For instance, the current field testing of the VTFE
would, if successful, establish the reliability of this process, for which there
is no basis at present. Costs would also be lower if substantially higher heat
transfer rates could be demonstrated than conservatively estimated in this study.
Although not specifically studied, a combination process may be superior to any
one of the individual processes examined in this report. An interesting possi-
bility is the pre-concentration of cooling tower blowdown by a membrane process
such as RO or ED, sending the reject stream for final volume reduction to a dis-
tillation plant such as VTFE, vapor compression, or multistage flash.
2-1
-------
TABLE 1. COMPARISON OF PROCESSES
(1)
Reverse Vertical Tube
Osmosis Foaming Evaporation
Economics
Plant investment 1st quarter
'79 (million dollars)
Annual cost (million dollars
per year)(2)
Capital cost
O&M cost
Total annual cost
Cost/kgal of feed
Cost/kgal of purified water
(3)
Energy consumption
Ponding requirements
(4)
Long-term related operating
experience on cooling
tower blowdown
Modular construction of sub-
units of each train
Sensitivity to upsets
in pretreatment
Product quality
13.9
1.7
1.2
2.9
3.28
4.37
Lowest
Large
Minimal
Standard
procedure
Very
sensitive
Fair
(475 ppm)
15.9
2.0
0.6
2.6
3.41
3.48
Low (if steam is
"free")
Moderate
None
Not practical
Minor
High
(10 ppm)
Vapor
Compression
Evaporation
15.2
1.9
2.9
4.8
6.06
6.18
High
Moderate
Considerable
and successful
Not practical
Minor
High
(10 ppm)
(1) See Table 9 for details.
(2) Basis: 30-year plant life, ponds at $50,000/acre, electricity at 4c/kWh,
12.5% fixed charge rate. See body of this report for a discussion of
effects of variations in these assumptions.
(3) Energy cost is.included in O&M cost above.
(4) Ponding cost is included in the plant investment and the calculation of
annual cost above.
2-2
-------
SECTION 3
RECOMMENDATIONS
The shakedown tests of the vertical tube foam evaporator (VTFE) at the Sea Water
Conversion Laboratory (SWCL) are being initiated as this report is being com-
pleted. Consequently, no conclusion can be reached at this time on the reliabil-
ity and economy of the process under actual field conditions. It is recommended,
therefore, that both shakedown runs at SWCL and field tests at a candidate power
plant be monitored and that the resulting data be used to update the calculations
and conclusions of this report.
Simultaneously, the "first cut" analyses presented here should be refined to
include a more detailed consideration of the overall water balance of a typical
power plant. Credit should be taken for the value of product water of different
degrees of purity produced by each of the three wastewater concentration pro-
cesses. Water not required for high quality usage (such as boiler feed water)
can be returned to the cooling tower loop, thereby decreasing the size and cost
of the wastewater concentration process, as demonstrated in this report.
In future work, use of electrodialysis (ED) should be considered; under some cir-
cumstances, it is competitive with reverse osmosis (RO) with respect to both
economics and to reducing the volume of cooling tower blowdown. At the same
time, work should proceed on a study of a two-step process. Such a study should
involve a membrane plant, RO or ED, of which the reject stream is further con-
centrated by an evaporative process such as VTFE, vapor compression, or multi-
stage flash evaporation. The combined process will not only deliver a stream of
purified water for boiler makeup but will sharply reduce the volume of reject
brine requiring ponding with a potential reduction in overall cost.
3-1
-------
Another attractive alternative for future consideration is the use of a soften-
ing process such as lime or thermal softening. A portion of its product can
be further purified for boiler feed water while the balance of the product,
which is comparatively low cost, can be used for cooling tower makeup, scrubber
water, and other purposes not requiring such high purity.
3-2
-------
Section 4
ANALYSIS AND COMPARISON OF PROCESSES
In this section, three processes for concentrating cooling tower blowdown are
analyzed: reverse osmosis (RO), vapor compression (VC), and vertical tube foam-
ing evaporation (VTFE). The last two processes are operated to allow the pre-
sence of substantial amounts of suspended solids in the brine being concentrated.
Reverse osmosis, however, must be operated in a scale-free manner, and therefore,
requires pretreatment of the cooling tower blowdown (described in detail in
Subsection 4. 1.3).
It should be noted that at the time of writing of this report, no VTFE plant was
in operation, nor had any data been generated by the pilot plant. The design
parameters and economic analysis, therefore, were based on commercial experience
with vertical tube evaporators (VTE) in accordance with EPA instructions. All
references to the VTFE design in this report refer to studies on the VTE and will
require updating when the results of the VTFE pilot plant field tests become
available.
Since reverse osmosis and its costs are sensitive to the composition of the feed
stream, two water compositions were examined: one limited by silica solubility
(Water A) and one by CaSO, solubility (Water B). In each case, the ion content
of a commercial power plant cooling tower blowdown was approximated. For the
distillation processes, where costs are not sensitive to moderate changes in com-
position, only Water A was considered, and softening of the blowdown was elimi-
nated. The compositions of water A and Water B are shown in Tables 2 and 3.
The second column in each table lists the compositions after pretreatment.
The conceptual design in each case is based on a typical 700 MWe fossil fueled
power plant. The cooling tower is assumed to operate at four cycles of concentra-
tion. That is, three-fourths of the water content of the makeup stream is evap-
orated, resulting in a blowdown in which the total dissolved solids concentration
is four times that in the makeup.
-------
TABLE 2. WATER A: CONCENTRATION OF WATER LIMITED BY
SILICA SCALING
Composition Expressed as ppm of Ion
Slowdown After
Cooling Tower Softening and
Slowdown pH Adjustment
Na
Ca
Mg
Cl
SO,
1,320
400
400
208
5,000
44
150
7,522
7.3
2,018
50
236
208
5,027
43
37
7,619
6.0
Sum of ions
PH
T 80°F 80°F
TABLE 3. WATER B: CONCENTRATION OF WATER LIMITED BY
CALCIUM SULFATE SCALING
Composition Expressed as ppm of Ion
Slowdown After
Cooling Tower Softening and
Slowdown pH Adjustment
Na
Ca
Mg
Cl
SO,
3
5
710
533
193
266
,820
35
5
,562
6.7
1,265
50
174
266
3,847
43
5
5,650
6.0
Si02
Sum of ions
PH
T 80°F 80°F
4-2
-------
Each of the processes produces a reject stream requiring ponding or disposal.
In this study, it is assumed that the reject brine is discharged to plastic-
lined solar evaporation ponds. These ponds are economically attractive only in
locations where evaporation exceeds precipitation. In regions of high rainfall,
the ponds require covers, which add substantially to their cost.
This report was directed primarily toward the arid regions of the western United
States, where brine disposal is a serious problem and the conservation of cooling
tower water in a power plant is highly desirable. In these regions, net evap-
oration (evaporation minus rainfall) is high. The net evaporation from brine in
Arizona, for example, is reported to be 5.7 feet per year. ' The annual
evaporation rate, however, is not a constant at any particular location but is
influenced, among other things, by the height of the berm above the liquid level
in the pond. In this study, therefore, the conservative value of 4.0 feet of
net evaporation per year was used. The usable depth of the ponds was assumed to
be 10 feet.
For the evaporation processes, for which the reject flow was low, a pond area
was assumed adequate to hold the crystallized salts deposited from 30 years'
flow of reject brine. The surface area was found to be far in excess of that
required to evaporate all the contained water at an evaporation rate of 4.0
#
feet per year. For the RO system, however, additional pond area was required
to evaporate the large reject flow and contain the unevaporated water at the end
of 30 years.
In the economic section of each process, the cost of solar evaporation ponds was
calculated on the basis of $40,000, $50,000, and $60,000 per acre, and the pond
costs .were added to the installed costs of the respective plants.
There are several alternatives to ponding, among them deep well injection. How-
ever, injection requires extensive pretreatment of the aqueous waste stream to
ensure freedom from suspended matter and from any species that may tend to crys-
tallize or precipitate at downhole conditions. A careful analysis of all
related costs is required before deep well.injection is seriously considered.
4-3
-------
The selection of a hot desert region in the western United States as the basis
for this study led to the assumption of a comparatively high cooling tower tem-
perature and, hence, to a turbine exhaust at 125 F as the heat source to the
VTFE plant. In colder climates, the turbine exhaust temperature would be much
lower. Since the cooling tower temperature would drop by the same amount, the
thermal driving force across the VTFE would be unchanged. However, the VTFE
design and equipment costs would then reflect the increase in vapor volume and
the decrease in heat transfer coefficient resulting from evaporation at the
lower temperature. Alternatively, the VTFE could still be operated at the
125 F steamside temperature assumed in this study by supplying the process
with extraction steam from the turbine. In that case, the thermal energy sup-
plied to the VTFE could no longer be considered "free," as assumed here.
4.1 REVERSE OSMOSIS
4.1.1 Design Basis
Precipitation of scale in a reverse osmosis module may completely impede the
functioning of the membrane in a very short time. Therefore, it is essential
to treat the feed to the RO plant to reduce the concentration of scale formers
to the point where their solubilities will not be exceeded in the reject stream,
and in particular in the boundary layer between reject brine and membrane,
where the concentrations are a maximum. For example, if 75 percent of the water
is recovered in purified form from the feed stream, the reject stream then repre-
sents only one-fourth of the original volume and thus contains roughly four times
the initial concentrations of contaminants.
Consequently, the process design in Figure 1 performs the following:
Reduces the calcium concentration in the cooling tower blow-
down to a value that will avoid CaSO^ scale formation in the
reverse osmosis reject stream
Precipitates Mg as the hydroxide to carry with it sufficient
Si02 to prevent silica scaling in the reverse osmosis modules
Adjusts the pH of the softened water to ensure a pH below the
Langelier Index for CaC03 precipitation in the reverse osmosis
reject stream .
4-4
-------
M.I
LIME WEIGH FEEDER
iNK AGITATOR
T-10
BACKWASH TANK
Figure 1 . Revere osmosis pretnutmcnt proccn flow diagram.
Figure la. Reverse osmosis pretreatment process flow diagram.
-------
WATER A
WATER B
^^-^-^^STR E AM
PAR AM ETEFT- -^_
pH
TEMPERATURE °F
FLOW GPM
M3/H
LB/HR
SALINITY PPM AS ION
Ca ++
Mg "*"*"
Na +
SiO2
Cl
S04 =
HC03-
Cl ~
ss
o
7.3
80
1700
386
849,660
7522
400
400
1320
150
5000
44
208
<$>
10.3
80
48
11
23,990
32,650
5800
3975
100,000
<§>
10.3
80
1700
386
849,660
7,619
50
236
2,018
37
5,027
43
208
O
7.3
80
1700
386
849,660
5,562
533
193
710
5
3,820
35
266
<§>
10.3
80
21
5
10,496
38,517
1,515
100,000
<§>
10.3
80
1700
386
849,660
5,650
50
174
1,265
5
3,847
43
266
STREAM
<&
CHEMICAL ADDITIVE
LIME 90% CaO
SODA ASH 98% Na2CO3
CHLORINE CI2
SULFURIC ACID 66° Be
SULFURIC ACID 66° Be
WATER A
LB/HR
382
1,457
1.7
16
9
WATER B
LB/HR
1,110
1.7
17
10
Figure Ib. Material balance for RO pretreatment process flow diagram.
-------
All the pretreatment steps use equipment of proven commercial design. The reverse
osmosis unit finally recovers 75 percent of the water content of the feed, reject-
ing the remaining 25 percent as a concentrated brine to storage ponds.
4.1.2 Operating Experience
Reverse osmosis is an established process for recovering purified water from
saline feeds. During the past 10 years, it has gained wide acceptance for sup-
plying potable water from brackish feed streams. More recently, reverse osmosis
has been used to treat plating rinses, irrigation runoff, and municipal sewage.
A measure of its acceptance is the recent award of contracts by the Bureau of
Reclamation for membrane units to desalinate roughly 100 million gallons per
day of agricultural drainage water. In addition to the above plants for the
desalination of low salinity feeds (500 to 4,000 ppm of total dissolved solids),
several membrane installations now produce potable water from seawater (35,000 ppm),
Figure 2 is a photograph of a portion of a large modern RO plant which produces
potable water from seawater.
The performance of the membranes assumed in this report, both for recovery and
salt rejection, is based on actual installations having feeds of similar com-
position. The pretreatment system is standard commercial practice for soften-
ing, clarifying, and removing silica from hard waters. It contains many of the
features of the Yuma desalination plant being erected by the Bureau of Reclama-
tion for the Colorado River Salinity Control Project.
4.1,3 Process Description
In the pretreatment process flow diagram (Figure 1), 1,700 gpm of cooling tower
blowdown are delivered to the reactor-clarifier (RC) along with the solution of
sodium carbonate (soda ash) and a lime slurry. The sludge formed in the reactor-
clarif ier is discharged to a storage pond (not shown) or to landfill. Alterna-
tively, it may be thickened and filtered, and the filter cake sent to a kiln to
recover the lime for reuse. However, this alternative is not included in the
design study, since recovery of lime is not usually practical for the small
amounts produced here.
4-7
-------
f
QO
Figure 2. Photograph of portion of large seawater RO plant.
(Courtesy of Fluid Systems Division, UOP Inc.)
-------
Chlorine is injected into the overflow from the RC to prevent growth of bacteria
in the multi-media filter, the cartridge filters, and the lines downstream from
the RC.
Because of the comparatively slow kinetics of the softening process, the down-
stream filter could become clogged as a result of post-precipitation. Conse-
quently, the pH of the RC overflow is adjusted by adding acid. The acidified
stream is then passed through a multi-media filter to reduce the suspended solids
content. Periodic backwashing of the filter with part of the filtrate flushes
the entrapped solids, and the flush water is recycled to the RC.
Additional acid is metered into the filtrate to reduce its Langelier Index to
a value that will ensure freedom from calcium carbonate scale in the brine
concentrate of the reverse osmosis unit and, thus, on the membranes themselves.
(Refer to Appendix A.) The final polishing of the water is accomplished by
passing it through 5-micron or 10-micron cartridge filters.
The pretreated water is then delivered to the RO system. (See Figure 3.)
Before contact with the membranes of the RO units, the feed stream must be
dechlorinated, either by injecting sodium bisulfite or by passing it through
a bed of activated carbon. If polyamide membranes are used, residual chlorine
must be reduced to zero. Cellulose acetate membranes, on the other hand, can
tolerate up to 1.0 ppm of chlorine.
The pretreatment discussed above was aimed at maintaining a safe level of silica
and calcium hardness in the RO reject stream. To maximize the recovery of puri-
fied water and thus achieve the greatest reduction in waste volume, it is common
practice to operate the RO unit as close to the calcium sulfate scaling level as
possible. In the present study, water recovery was assumed to be 75 percent.
Under these conditions, any excursion or failure of the pretreatment system
could result in serious calcium sulfate scaling of the membranes. Consequently,
5 ppm of sodium hexametaphosphate is injected to prevent scale formation by
threshold inhibition, that is, to obstruct the growth of scale crystals.
4-9
-------
M-9
SODIUM
BISULFITE MIXER
M-10
SODIUM HtXAMETA
PHOSPHATE MIXER
G-9 A & B
SODIUM BISULFITE
METERING PUMPS
G-11 A& B
SODIUM HEXA META PHOSPHATE
METERING PUMPS
T-11
SODIUM BISULFITE
MIX TANK
T-12
SODIUM HEXA META
PHOSPHATE MIX TANK
F-2
CARTRIDGE FILTERS
PROM PRETREATMENT
SYSTEM
I
F
F
H
^
t
^
r
UR
PA
Z!A
1
PRODUCT
T-13
CKED TOWER
G 10 A & B
RO FEED PUMPS
REJECT BRINE TO POND
WATER A
WATER B
' ^__STREAM
QUANTITY -^~^_
PH
TEMPERATURE °F
FLOW GPM
M3/H
LB/HR
SALINITY PPM AS ION
Ca ++
Mg ++
Na +
SiO,
5.2
80
1275
290
637,245
475
4
17
121
8
0
306
4
15
XX
6.3
80
425
97
212,415
28,855
187
893
7708
125
0
19083
73
786
<<>
5.2
1275
290
637,245
384
0
0
126
1
0
214
7
36
NX
6.3
80
425
97
212,415
21.358
200
700
4,696
17
0
14.638
151
956
Figure 3. Reverse osmosis process flow diagram.
4-10
-------
Finally, the volume of aqueous waste is reduced by passing it through the RO
modules at 400 psig. Seventy-five percent of the feed is recovered in the form
of relatively pure water containing 6 to 7 percent of the initial salt concentra-
tion. The balance of the feed, containing all the remaining solutes, is discarded
to waste.
Details of the equipment and liquid streams are presented in Subsection A.1,4.
*
4.1.4 Plant Design
A pretreatment system is designed to supply water that meets the influent quality
requirements of the selected RO system. Pretreatment consists of softening, pH
adjustment, chlorination, filtration, sodium bisulfite addition, and sodium hexa-
metaphosphate (SHMP) addition. Each pumping system is provided with an installed
spare pump to satisfy the high reliability requirements of the treatment train.
All pumps are sized at 110 percent of required capacity.
The pretreatment system for RO is sensitive to the composition of the cooling
tower blowdown feed stream, so Water A and Water B are considered separately.
Figure 1 shows the pretreatment system. Water A (see Table 2) is softened by
the cold lime soda process. Lime and soda are mixed with cooling tower blowdown
in the reaction zone of a 60 foot diameter reactor-clarifier (RC) of steel con-
struction. An 1,870 gpm pump transfers the blowdown from the cooling tower basin
to the RC. A lime system supplies lime to the RC as a 10 percent slurry. This
lime system consists of a 140 ton capacity silo, which stores a 30 day supply of
chemical lime (90 percent CaO), a 380 pound per hour gravimetric feeder, and a
slaker with a capacity of 380 pounds per hour. The lime slaker discharges a
slurry to a 2,600 gallon capacity feed tank which provides 4 hours' retention
time. A one horsepower agitator maintains the feed tank solids in suspension,
*In the specification and purchase of water treatment equipment, English units
are commonly used for dimensions, fluid flows, and energy transfer and consump-
tion. The reader is referred to Appendix B for the conversion table to SI
units.
4-11
-------
and a 10 gpm pump supplies the lime slurry from the feed tank to the RC. A soda
ash system delivers soda ash to the RC as a 10 percent solution.
A 525 ton capacity silo stores a 30 day supply of soda ash. A gravimetric feeder
supplies soda ash at 1,450 pounds per hour from the silo to a dissolving tank with
1,270 gallons capacity and 30 minutes' retention time. A one horsepower agitator
is mounted in the dissolving tank to dissolve the soda ash in water. The soda ash
solution flows by gravity to a feed tank with 8,500 gallons' capacity and 4 hours'
retention time. A 30 gpm pump delivers the soda ash solution from the feed tank
to the RC. Instrumentation is provided for automatic adjustment of the lime and
soda ash feed rates based on the cooling tower blowdown flow rate.
As the blowdown, lime, and soda ash react in the RC, solids form. These solids
eventually settle as a sludge and flow to a sump. A 50 gpm pump discharges the
sludge from the sump to receiving facilities for final disposal at a landfill.
Clarified, softened water flows over the RC weir into a 28,800 gallon capacity
surge tank with 15 minutes' retention time. A chlorination system injects
approximately 40 pounds of chlorine per day to the surge tank influent. This
results in a chlorine concentration of approximately 2 ppm in the surge tank
contents.
An 1,870 gpm pump transfers water from the surge tank to a 19,800 gallon capacity
head box with 10 minutes' retention time. Water flows by gravity from the head
box to an upflow dual media filter. The head box outlet is situated above the
filtered water effluent to provide sufficient head for the water to flow upward
through the filter media and into the filtered water storage tank. The filter
sizing is based on a loading rate of 3 gpm per square foot. The filter is 28
feet in diameter and is divided into three compartments of equal area. Each
compartment of the filter is backwashed automatically when the pressure drop
across the compartment unit reaches a predetermined level. Under design condi-
tions, it is anticipated that each compartment will be backwashed twice a day.
The backwash cycle lasts 4 to 5 minutes and the backwash rate is 15 gpm per
square foot. A 16,000 gallon backwash holding tank is sized to accommodate the
4-12
-------
total flow from the backwash cycle of one compartment. A 10 horsepower agitator
maintains the solids in suspension in the backwash tank. A 60 gpm pump contin-
uously recycles the backwash water to the RC.
A sulfuric acid system is provided to add 66 Be sulfuric acid to the system.
A 1,700 gallon tank stores a 30 day supply of acid. A metering pump with a
range from 1 to 5 gph delivers acid at approximately 24 gallons per day (gpd) to
the surge tank effluent and 14 gpd to the gravity filter effluent. An 8 inch
diameter, 1 foot long, in-line static mixer is provided at each acid injection
point to ensure adequate mixing of acid and water.
Figure 3 shows the RO system. An 1,870 gpm pump is used to pump the gravity fil-
ter effluent through one of the two full-sized cartridge filters which are
installed in parallel. When one filter is being backwashed, the flow is
diverted to the standby filter.
A metering pump with a discharge capacity ranging from 1 to 10 gph adds sodium
bisulfite as a 10 percent solution to the cartridge filter effluent. A 1/4
horsepower mixer dissolves one day's requirement of sodium bisulfite (82 pounds
as sodium metabisulfite) in water in a 200 gallon tank. The discharge from
the metering pump is regulated by the residual chlorine concentration in the RO
feed pump discharge.
The cartridge filter effluent is pumped by an 1,870 gpm, 400 psi, 625 hp RO feed
pump through the RO system. Immediately upstream of the RO system, sodium hexa-
metaphosphate (SHMP) is added as a 10 percent solution to the RO feed pump dis-
charge by a metering pump with a 2 to 10 gph discharge range. A 1/4 horsepower
mixer dissolves one day's supply (100 pounds) of SHMP in water in a 260 gallon
tank. The SHMP feed rate is proportioned to the RO feed pump discharge rate.
The RO system consists of membranes arrayed in four banks of modules (individ-
ual vessels containing the membranes) in a three-stage configuration. The
first stage of each bank has 19 modules, the second stage has 8 modules, and
4-13
-------
the third stage has 6 modules. Each bundle of membranes is enclosed in an epoxy-
coated steel cartridge 10-3/4 inches in diameter and 48 inches long. Seventy-five
percent of the pretreated water applied to the RO system is recovered as product,
and the minimum salt rejection is 90 percent. Any hydrogen sulfide and carbon
dioxide present are removed from the permeate by forced-draft degasification in
an 8 foot diameter by 16 foot high packed tower. The packed tower is constructed
from carbon steel with coal tar epoxy coating. The permeate contains 475 ppm of
total dissolved solids and flows at 1,275 gpm. The reject stream contains 28,855
ppm of total dissolved solids and flows to solar evaporation ponds at 425 gpm.
The ponds provide an evaporative surface area of 160 acres and are sized on the
basis of 4.0 feet of assumed net evaporation per year. ' The composition
of the permeate and the reject stream appear in Table 4.
TABLE 4. WATER A: RO PERMEATE AND REJECT
Composition Expressed as ppm of Ion
Permeate Reject
Na 121 7,708
Ca 4 187
Mg 17 893
Cl 15 786
S04 306 19,083
HC03 4 73
Si02 8 125
Sum of ions 475 28,855
pH 5.2 6.3
T 80°F 80°F
A half-acre emergency pond is provided to store three days' feed to one of the
four RO banks to allow for complete shutdown and repair of one bank.
4-14
-------
The pretreatment system for Water B differs from that for Water A in that lime
is not required in the softening step, and the soda ash requirement is reduced
from 17.5 tons per day to 13.3 tons per day as 98 percent sodium carbonate.
Table 5 shows the soda ash and sludge handling equipment capacities for Waters A
and B where these values are not identical.
TABLE 5. SODA ASH AND SLUDGE HANDLING EQUIPMENT FOR WATERS A AND B
Equipment Water A Water B
Soda ash silo 17,180 ft3 14,730 ft3
Soda ash feeder 1,450 Ib/hr 11,000 Ib/hr
Soda ash feed tank 1,140 ft3 863 ft3
Soda ash feed pump 30 gpm, 1 hp 25 gpm, 0.5 hp
Sludge pump 30 gpm, 1 hp 25 gpm, 0.5 hp
The RO system required to treat Water A has 19 modules in the first stage of
each bank, while the RO system for Water B has 18 modules in the first stage
of each bank. The two RO systems are otherwise identical, with 75 percent over-
all system recovery and 90 percent minimum salt rejection in each case.
The permeate contains 384 pppi of total dissolved solids and flows at 1,275 gpm.
The reject stream contains 21,358 ppm of total dissolved solids and flows to the
solar evaporation ponds at 425 gpm. The evaporative surface area required is
the same for Water A and Water B. Table 6 shows the compositions of the permeate
and reject stream when Water B is treated by RO.
Table 7 presents the design criteria for the plant equipment. Figure 4 shows
the layout of the RO plant.
4-15
-------
TABLE 6. WATER B: RO PERMEATE AND REJECT
Composition Expressed as ppm of Ion
Na
Ca
Mg
Cl
S04
HC03
Si02
Sum of ions
pH
T
Permeate
126
0
0
36
214
7
1
384
5.2
80°F
Reject
4,696
200
700
956
14,638
151
17
21,358
6.3
80°F
4-16
-------
TABLE 7. SUMMARY OF DESIGN CRITERIA FOR RO OPTION
Equipment
Bulk storage for lime, soda ash,
and sulfuric acid
Feed tanks
Lime weigh feeder
Chemical lime slaker
Soda ash weigh feeder
Soda ash dissolving tank
Chlorine system
Reactor-clarifier
Surge tank
Head box
Dual media gravity filter
Cartridge filters
Sodium bisulfite mix tank
Sodium hexametaphosphate mix tank
RO system
Packed column
Design Criteria
30 day supply
4 hours' retention time
380 Ib/hr capacity
380 Ib/hr capacity
1,450 Ib/hr capacity
30 minutes' retention time
2 ppm chlorine dosage
2
Surface rate of 0.75 gpm/ft
with 20 percent additional area
15 minutes' retention time
10 minutes' retention time
o
Surface rate of 3 gpm/ft
2 full size systems
1 day's supply
1 day's supply
4 banks, each with 480,000 gpd
capacity
2
Loading rate of 35 gpm/ft , packed
height 8'0"
4-17
-------
i
h-t
00
y
LIME SILO
SODA
ASH
SILO
O
ACID
TANK
O
LIME
FEED
TANK
O
SODA
ASH
TANKS
SURGE
TANK
HEAD
BOX
O
ROAD
BUILDING FOR RO MODULES,
CONTROLS & SERVICES
PARKING
MEDIA
FILTER
BACKWASH TANK
ROAD
300'
o
in
Figure 4. Plot plan for RO option.
-------
4.1.5 Process Economics
The investment in the pretreatment plant is based on quoted prices for the major
equipment components. For the RO unit, quotations were solicited on the system
as a complete package, including pumps, instruments, and controls. Prices were
quoted by three vendors, differing by only a few percent. The average of these
quotations was used as the cost basis. Factors based on experience in the cost-
ing of water treatment plants were applied to cover site preparation, founda-
tions, installation costs, contractors' fees, and contingencies. Pond require-
ments were calculated on the basis of an assumed 30 year useful life. The total
installed plant cost, including ponding for the storage of blowdown, was esti-
mated at $13,900,000.
Treatment costs include operating costs, plus a capital charge based on a 30
year plant life. Membrane replacement charges assume a membrane life of three
years, in agreement with current experience in brackish water plants incorporat-
ing a thorough pretreatment of the feed. The calculated energy consumption is
5.3 million kWh per year, primarily for pumping power.
Three parameters were investigated.
The costs of lined ponds were assumed to be $40,000, $50,000,
and $60,000 per acre, respectively
Costs were developed for power at 2c, 4c, and 6c per kWh
The fixed charge rate was assumed at 12.5 percent and 16
percent, respectively
The results of the cost analysis are presented in Figures 5 and 6. Figure 5
indicates a comparatively minor dependence of RO treatment cost on the price of
energy. This is because the large ponding requirements of RO place emphasis on
plant investment and resulting capital costs. In the cases studied here, capital
costs comprise 57 to 68 percent of the total water treatment cost. Energy, on
the other hand, contributes from 4 percent to 11 percent, the actual value
depending on the fixed charge rate and kWh prices assumed.
4-19
-------
Q
LU
C3
o
o
o
4
V)
O
(J
(L
\-
FIXED
CHARGE
RATE
BASIS: POND COST $50.0007 AC RE
I
I
I
) 1 2 3 4 5 6
ENERGY COST (C/kWh)
Figure 5. Effect of energy cost on RO treatment costs.
4-20
-------
8r
7 -
6|_ BASIS: ENERGY COST
4 C / kWh
o
§
FIXED
CHARGE
RATE
16%
V)
O
u
12.5%
40
50
60
POND COST ( $1,000 / ACRE )
Figure 6. Effect of pond cost on RO treatment costs.
4-21
-------
The comparative importance of assumed pond costs is apparent in Figure 6, where
an increase from $40,000 to $60,000 per acre increases the treatment cost by
roughly 20 percent. It is possible to increase the recovery of the RO plant
from the 75 percent value assumed in this study to 85 percent or possibly higher
for the particular feed waters assumed here. As a result, the flow of brine to
the ponds by the RO plant would decrease by 36 percent. The size of the ponds
and their cost would decrease by a lesser amount, since a fraction of the pond
volume is required for the crystals precipitated from the evaporating brine.
The increased recovery of the RO plant, however, would require a more thorough
and more costly pretreatment of the cooling tower blowdown. The choice of the
exact process conditions would require a more detailed cost comparison.
For comparison of costs of the three candidate processes analyzed in this
report, please see Subsection 4.5.
4-22
-------
4.2 VAPOR COMPRESSION EVAPORATION
4.2.1 Design Basis
In contrast to reverse osmosis, the Resources Conservation Company's vapor com-
pression evaporator investigated in this study can tolerate the presence of
scale formers. Deposition on the plant components is prevented by maintaining
a comparatively high concentration of calcium sulfate crystals in the brine to
provide nuclei on which the scale will deposit in preference to the equipment
and piping. If any scale deposits on the evaporator tube surfaces, it is
scoured off by the recirculating slurry.
The seed slurry is not present, however, in the feed preheater. (See the pro-
cess description in Subsection 4.2.3.) There the concern is to avoid calcium
carbonate scale, which tends to deposit on heated surfaces because of its
inverse temperature solubility characteristics. To prevent calcium carbonate
deposition, the pH of the cooling tower blowdown is adjusted to a negative
Langelier Index by adding acid.
The feed assumed for the vapor compression plant is unsoftened cooling tower
blowdown Water A, having the composition shown in the first column of Table 2.
4.2.2 Operating Experience
Vapor compression evaporation is an old and established technique that has been
used principally for the recovery of potable water from seawater. Its use for
concentrating cooling tower blowdown was pioneered within the past decade by
Resources Conservation Company. Figure 7 is a photograph of a commercial
VC plant.
At present, ten units, ranging in feed capacity from 10 to 700 gpm, are in
operation in waste concentration service, mostly in power generating stations.
Seven more are under construction. The evaporators recover up to 98 percent
of the water in the feed in the form of a very pure product (2 to 6 micromho
conductivity, corresponding roughly to 1 to 3 ppm of total dissolved solids).
4-23
-------
JS
NJ
Figure 7. Photograph of a commercial VC plant.
(Courtesy of Resources Conservation Company)
-------
The product can be processed in a mixed-bed ion exchange polisher to yield
ultrapure water for boiler makeup.
The on-stream factor of all the plants investigated has been very high.
4.2.3 Process Description
The vapor compression evaporator (VCE) studied in this report is fundamentally
a vertical tube evaporator in which the liquid being evaporated descends as a
thin film on the inside of tubes arranged in a bundle within an evaporator
shell. Steam within the shell condenses on the exterior walls of the tubes,
thereby giving up its latent heat to evaporate a fraction of the water content
from the liquid film in the tubes. By returning the partially concentrated
fluid to the upper plenum (water box), further evaporation is achieved to pro-
duce a fairly concentrated brine.
The VCE differs from a conventional vertical tube evaporator in its lack of an
external condenser for the steam generated during the evaporation process.
Instead, the steam is compressed to raise its temperature slightly (usually
5 F to 12 F) and is then delivered to the shell of the heat exchanger section
where it condenses on the tubes as described in the preceding paragraph. That
is, the steam chest itself acts as a condenser. The source of evaporative
energy is the mechanical energy input of the compressor rather than the thermal
energy of a steam supply.
Referring to the schematic diagram in Figure 8, the cooling tower blowdown
(pH adjusted by acid addition to avoid calcium carbonate scaling of the heat
exchanger) is pumped through a plate-and-frame type heat exchanger, where it is
preheated by heat exchange with product water. Air and carbon dioxide are
stripped from the feed stream in a vacuum deaerator.
Deaerated brine is distributed over the upper tubesheet of the evaporator and
flows down as a thin film on the inside wall of each of the tubes. Hot vapor
condensing on the outer wall of the tubes contributes the heat required to
4-25
-------
*-
N3
TO MODULES
2 AND 3
I 749.700 I 1500 I
n 3407 I
FEED
FEED
FILTERS
1249.900 |500 | ,,
113.6
ACID SUPPLY
I 13 I OT026 |
1 0.006 |
VENT
nit
A
yv
209-21 2°F
^*^DQ IMA
FEEDTANK
NO. 1 RUMP | HEAT
EXCHANGER
PRODUCT
244.902 | 490 |
| 111.28 |
215-220°F
PRODUCT LEVEL
CONTROL VALVE
VENT
209-21 2°F
DEAERATOR
PRODUCT TANK
STEAM INLET
CONTROL VALVE
[2,000
STEAM BOILER
STARTUP)
I J
r<
UI-*
220-222°F
212 F COMPRESSOR
GUIDE VANE
POSITIONER
COMPRESSOR
|1 2.495.000125.000 1
1 5,678 1
4998
I 2.27 |
-*- TO POND
PRODUCT PUMP
WASTE DISCHARGE
RECIRCULATION CONTROL VALVE
PUMP
LB/HR GPM
Figure 8. Vapor compression evaporation module.
-------
evaporate a portion of the water contained in the feed. The brine that falls
to the bottom of the evaporator is pumped to the top to repeat its descent
through the tubes, thereby evaporating an additional fraction of its water con-
tent. A portion of the recirculating brine, together with any precipitates sus-
pended in it, is bled from the loop and is discharged to waste. Since it repre-
sents only a small fraction of the volume of feed to the evaporator, no attempt
is made to recover its sensible heat.
The vapor generated within the tubes is discharged through a mist eliminator and
is withdrawn by the vapor compressor and delivered to the shell side of the evap-
orator; there it condenses on the outside of the tubes as described in the pre-
ceding paragraph. The compressor serves a double purpose:
It contributes the energy required for the evaporative process
By the adiabatic compression of the vapor, it raises its tem-
perature to provide the thermal driving force across the wall
of the tubes
The small steam boiler supplied at no additional cost by the system manufacturer
(see Figure 8) is used for plant startup only.
The condensate is withdrawn from the evaporator shell and cooled by passage
through the heat exchanger, where it preheats the feed as previously described.
The purity of this condensate is monitored by a conductivity cell in the product
water line.
The vapor compression evaporator cycle is very attractive because of its high
energy effectiveness under commercially attainable operating conditions. Energy
consumption for distillation as low as 70 kWh per 1,000 gallons of cooling tower
blowdown fed to the evaporator has been achieved in operating, full-size plants.
In equivalent heat consumption this corresponds, for example, to a multistage
flash distillation plant operating at the very high economy ratio of 12 pounds
of evaporation per 1,000 Btu's supplied, compared with the more common economy
ratio of 8. The overall energy consumption, including the power to drive the
pumps, will range from 70 to 90 kWh per 1,000 gallons.
4-27
-------
There are several features inherent in a VCE that strongly affect its economics.
All other things being equal, the smaller the temperature rise and, hence, the
compression ratio, the higher the overall energy efficiency of the process. On
the other hand, a small temperature rise requires a large heat transfer area.
Consequently, the choice of operating conditions is a trade-off between cost of
energy consumed and plant investment. A second consideration concerns the
operating temperature of the VCE. The choice of a low operating temperature
will greatly decrease corrosion but will produce vapor of high specific volume,
thereby increasing the size of the compressor and vapor lines. The system
selected for analysis in this report operates at approximately 212 F.
To provide the initial charge of vapor to the compressor, a small auxiliary
boiler supplies steam to the evaporator for a short time at startup. In addi-
tion, stable operation is not achieved for the first few hours after an extended
shutdown to permit the scale crystals, suspended in the brine, to revert to the
optimum crystal form for seeding subsequent scale deposits.
4.2.A Plant Design
The raw cooling tower blowdown requires only minimal pretreatment before it
enters the vapor compression evaporator system. The pretreatment consists of
filtration and pH adjustment.
Figure 8 shows one of the three vapor compression evaporation modules. Blow-
down from the cooling system is discharged to the concentrator' at 1500 gpm.
Two 316 stainless steel 1,650 gpm pumps (one an installed spare) are provided
to transfer the blowdown from the cooling tower sump through two on-line filters
to three feed tanks. Two spare filters are provided to enable continuous opera-
tion of the evaporator during the backwash cycle. Each feed tank is constructed
from fiberglass reinforced plastic and has a capacity of 6,460 gallons to pro-
vide 10 minutes' retention time for one-third of the total cooling tower blow-
down flow. A sulfuric acid system adds about 12 gpd of 66 Be sulfuric acid to
each feed tank to lower the pH of the blowdown from 7.3 to within the range of
5.5 to 6.0. A one hp agitator provides adequate mixing of the acid and water
in each tank.
4-28
-------
The evaporator system consists of three modules, each with 500 gpm capacity.
Each feed tank supplies water to a 316 stainless steel 550 gpm feed pump which,
in turn, feeds one module. Each module consists of a heat exchanger, a deaera-
tor, an evaporator with a brine pump, a recirculating pump, a product tank, a
product pump, and a compressor. The feed pump discharge enters the heat
exchanger which brings the temperature of the feed to near its boiling point
by recovering the sensible heat contained in the hot product (condensate) stream
from the evaporator. The heat exchanger is of a plate-and-frame type with
titanium plates. The approach of the heat exchanger is 6 F.
The heated stream next enters a counterflow, packed column, atmospheric deaera-
tor measuring 18 inches in diameter and 4 feet high. The deaerator has a 316L
stainless steel shell and is packed with plastic Berl saddles. Carbon dioxide,
nitrogen, and oxygen are removed in the deaerator and vented to the atmosphere.
The feed leaves the deaerator at 209°F to 212°F and flows by gravity to the evap-
orator sump. Here, it mixes with the concentrated slurry and is continuously
recirculated by a 25,000 gpm pump. The evaporator is 12 feet in diameter and
is 80 feet high. It has a 316L stainless steel shell and a tube bundle consist-
ing of 2,600 titanium tubes, each having an outside diameter of 2 inches and a
thickness of 0.028 inch. The tubes are 50 feet long, and provide 64,000 square
feet of heat transfer surface.
The recirculated flow is distributed to the inside wall of each tube as a thin
film. Water is evaporated as the film falls down inside the tubes. The steam
formed passes through a mist eliminator and enters the suction line of the com-
pressor at approximately 209 F to 212 F. The compressor has a 3,500 hp motor and
compresses the steam by 2 psi to raise its condensation temperature to 220 F to
222 F, about 6 F above the boiling point of the recirculating brine. As the
steam condenses on the shell side of the tubes, it gives up its heat of conden-
sation and is collected in a 316L stainless steel product tank. The latent
heat of condensation provides the energy required to evaporate the brine inside
the tubes.
4-29
-------
The brine in the sump contains 301,700 ppm total dissolved solids and 82,000
ppm suspended solids. The product contains less than 10 ppm total dissolved
solids. The concentration factor of the system is 50.
The hot (220 F to 222 F) product is pumped back through the heat exchanger by a
316L stainless steel 75 psi pump. The product stream is about 98 percent of the
feed stream. A device senses the salinity of the brine in-.the sump continually,
and maintains the sump total dissolved solids concentration in the desired range
by controlling a waste flow valve. The waste flow, normally about 2 percent of
the feed stream, is discharged to a 50 acre solar evaporation pond. The pond
size is based on a net evaporation rate of 4.0 feet per year ' and is
surrounded by a dike with a 2:1 slope. The water depth in the pond is 10 feet.
At startup, each unit is seeded with calcium sulfate crystals to produce a
slurry. Twelve hours are required to heat the slurry up to the desired tempera-
ture. The slurry is then "heat soaked" for 24 hours. A boiler is provided to
supply 2,000 pounds of steam per hour at 15 psi for startup from a cold start.
The service water requirement for the evaporator system is 60 gpm. The air
requirement for the controllers is 100 cfm at 3 psi. The evaporator system
includes all controls for automatic sensing and recording of flow rates, tem-
peratures, and pressures.
A 0.85 acre emergency pond is provided to store three days' flow to one of the
three modules. The pond is lined with Hypalon and has a water depth of 10 feet.
An earthen dike with a 2:1 slope provides 2 feet freeboard. The layout of the
vapor compression evaporation plant is shown in Figure 9.
4.2.5 Process Economics
The plant investment for vapor compression evaporation is based on the turnkey
price of an analogous system recently installed for concentrating cooling
tower blowdown. Adjustments were made to include site preparation, roads,
fences, and service lines to the plant. The total installed plant cost, in-
cluding ponds for the storage of blowdown, was estimated at $15,200,000.
4-30
-------
220'
ROAD
ACID ( )
TANK
BUILDING FOR
CONTROLS, OFFICES,
& SERVICES
ROAD
FEED HEAT
TANKS EXCHANGERS
EVAPORATORS
& SUMPS
o
o
o
PRODUCT
TANKS
O
o
o
PARKING
COMPRESSORS
& MOTORS
START-UP BOILER
Figure 9. Plot plan for vapor compression evaporation option.
-------
The energy consumption was assumed to be 81 kWh per 1,000 gallons of feed,
roughly the median value for vapor compression plants treating similar feed
streams. In actual practice, the supplier guarantees the energy consumption
for each plant quoted.
Vapor compression evaporation is one of the least energy intensive of all the
distillation processes (provided, of course, that the energy for the evaporation
processes is not "free" as has been assumed in the case of the VTFE in this
report). Nevertheless, the cost of power represents a substantial fraction of
the overall cost of treating the cooling tower blowdown by this process. In
the cases analyzed here, energy consumption contributes 37 to 63 percent to the
overall treatment cost, the actual value varying with the fixed charge rate and
kWh prices assumed. The importance of energy cost is emphasized by the steep
rise in the treatment curves in Figure 10.
In contrast to the 25 percent brine rejection of RO, the VC plant rejects only
2 percent of the feed stream, recovering the remaining 98 percent as purified
water. Consequently, Figure 11 shows a very small dependence of treatment costs
on the cost of ponds.
4-32
-------
(3
o
(fi
O
O
UJ
cc
FIXED
CHARGE
RATE
16%
12.5%
BASIS: POND COST $50,000/ACRE
I
I
0123456
ENERGY COST (C/kWh)
Figure 10. Effect of energy cost on VC treatment costs,
4-33
-------
Q
ui
IU
o
_i
<
v>
O
u
u 3
BASIS: ENERGY COST
4
-------
4.3 VERTICAL TUBE FOAMING EVAPORATION
4.3.1 Design Basis
This section is devoted to discussion of vertical tube foaming evaporation (VTFE)
At the time of writing of this report, however, no VTFE units are in commercial
operation, nor have any test data been generated by the recently constructed
pilot plant. Consequently, the technical and economic analysis is based on the
vertical tube evaporator (VTE) rather than the VTFE, as agreed by the EPA.
Because of the lack of data, a very conservative approach has been adopted. If
the pilot;plant substantiates the encouraging results previously reported by the
Sea Water Conversion Laboratory of the University of California for tests on the
earlier small pilot unit, a reduction in the size and plant investment reported
in this section may be anticipated.
The vertical tube foaming evaporator (VTFE) system consists basically of a con-
ventional vertical tube, recirculating type evaporator to achieve the major
portion of cooling tower blowdown concentration. It is followed by a similar
but smaller evaporator for carrying the concentration to the crystallizing
stage. The novel feature is the addition of a small amount of surfactant to
the feed. The surfactant has been shown to provide three advantages:
The rate of deposition of scale formers on the heat transfer
surfaces and other plant components is drastically reduced
A foaming surfactant produces a comparatively stable two-phase
fluid in the evaporator tubes; this lowers hydrostatic losses
enought to permit recirculation of brine in the upflow mode
without a pump when operating at somewhat elevated tempera-
tures (above 150°F)
The overall heat transfer coefficient in the tubes is improved
by adding the surfactant.
Because this process can tolerate the presence of certain scale formers, either
of the two cooling tower blowdown compositions would be suitable as feed to the
VTFE. For this study, Water A was selected. (See Table 2 for composition).
4-35
-------
The evaporator design is based on removing approximately 90 percent of the water
contained in the feed. The crystallizer removes most of the remaining water and
delivers a slurry of crystals to the waste pond.
4.3.2 Test Experience
Vertical tube evaporators have been tested for several years in the desalination
of seawater. A 1 mgd plant was operated in the downflow mode for approximately
, r -,-,. r-i i- (Ref. 2 and 3)
two years at Freeport, Texas, using the falling film technique.
In addition, vertical tube evaporation has been widely used in crystallization
processes in the so-called Oslo type crystallizers in which, however, the flow
pattern more closely approximates full-tube flow rather than film flow.
No industrial applications of the VTFE exist at present. A 5,000 gallon per day
pilot plant evaporator-crystallizer has been subjected to extended tests at the
Sea Water Conversion Laboratory (SWCL) of the University of California at Rich-
, ,., . (Ref. 4)
mond, California.
4.3.3 Process Description
As in the case of the vapor compression evaporator, the VTFE is basically a ver-
tical tube evaporator. Two modes of operation have been tested. In one mode,
the feed (cooling tower blowdown in several tests performed at the SWCL) is pumped
to the head of the evaporator, is distributed across the tube sheet, and enters
the tubes via specially designed nozzles, permitting it to fall as a film on
the inner walls of heat exchanger tubes. In the other mode, the feed enters the
sump at the bottom of the evaporator, passes through a perforated distributor
plate, and rise a short distance up the heat exchanger tubes. In both modes,
steam condensing on the outside of the tubes causes evaporation of the liquid
within the tubes. In the second mode, referred to as upflow operation, steam
bubbles generated within the liquid in the bottom portion of each tube tend to
rise, carrying with them some liquid. This liquid is ejected from the upper ends
of the tubes, resembling somewhat the action of a coffee percolator.
The upflow mode of operation is advantageous because no recirculating pump is re-
quired, either in the single effect design under study here or for the transfer
4-36
-------
of brine to each succeeding effect in a multieffect plant. In addition to
the saving in plant investment, energy cost is also reduced. However, at low
operating temperatures such as those encountered in systems using waste heat,
the hydraulic driving force may not be adequate for upflow operation even after
surfactant is added. Therefore, at the low temperatures anticipated in the
field tests of this study, the downflow mode will be used.
For either mode of operation, the addition of a few ppm of surfactant results in
improved performance. In downflow, the additive assists in distributing the
liquid as a uniform thin film on the inner tube wall. In upflow, a suitable
surfactant stabilizes the foam produced in the tubes as a result of steam
generation, improving the operational stability and heat transfer coefficient
and decreasing hydrostatic pressure loss. A number of foaming surfactants have
been tested successfully at SWCL; an alkyl benzene sulfonate appears to be pref-
erable at present.
4.3.4 Plant Design
The flow diagram for the vertical tube foaming evaporation system is shown in
Figure 12. Six modules are provided, each consisting of an in-line static
mixer, vacuum deaerator, evaporator, recirculating pump, and condenser. Raw
cooling tower blowdown is fed to each module at 250 gpm. The pH of each stream
is adjusted from 7.3 to within the range of 5.5 to 6.0 by the addition of 66 Be
sulfuric acid. After acid injection, the feed stream flows through a 3 inch
diameter, 1 foot long, 316L stainless steel in-line static mixer. The acid addi-
tion rate is regulated by the pH reading in the static mixer effluent.
The stream next flows to a 4 foot diameter vacuum deaerator where oxygen, nitro-
gen, and carbon dioxide are removed and vented to the atmosphere. The deaerator
shell is fabricated from 316L stainless steel. After leaving the deaerator, the
stream flows into the suction line of the evaporator recirculation pump, which
recirculates brine at 6,000 gpm to the tube bundle at the top of the evaporator.
Here, a perforated plate distributes the flow across the tube bundle, which con-
sists of 2,600 titanium tubes, each of which is 26 feet long with an outside
4-37
-------
I
u>
00
J CRYSTALLI2ER . .
RECIRCULATING I HUM I 30
PUMP
IONE PROVIDED)
Figure 12. Vertical tube foaming evaporation option - downflow mode.
-------
diameter of 1.5 inches. The tube bundle is enclosed in a 106 inch diameter 316L
stainless steel shell. The brine flows down the inner walls of the tubes as a
thin film and is heated by steam, which is supplied to the tube bundle at
116,000 pounds per hour. The temperature of the steam is 125 F. A mixture of
vapor and liquid descends into a 316L stainless steel evaporator, which measures
36 feet in diameter by 12 feet. Operating conditions in the evaporator are
105 F and 55 mm absolute pressure. Noncondensibles are removed by a steam ejec-
tor supplied with 700 pounds per hour of steam at 90 psig. The liquid in the
evaporator returns to the recirculation pump and the vapor passes through a
mesh demister to a shell-and-tube condenser.
The condenser has 5,344 titanium tubes, each 18 feet long with an outside dia-
meter of 1 inch. The condenser shell measures 100 inches in diameter and is
fabricated from 316L stainless steel. Cooling water with a temperature of 80 F
flows through the condenser at 23,000 gpm. The cooling water leaves the con-
denser at 90 F, and a 250 gpm stream is <
densate leaves the deaerator at 211 gpm.
denser at 90 F, and a 250 gpm stream is diverted to feed the evaporator. Con-
As the brine is recirculated to the top of the evaporator, a 39 gpm stream is
diverted from each evaporator to make up the feed stream to the crystallizer.
The streams are combined and enter the suction line of the crystallizer recir-
culating pump, which delivers brine to the top of the crystallizer at 4,500 gpm.
Here, a perforated plate distributes the flow across the tube bundle, which
consists of 1,950 titanium tubes, each 26 feet long with a 1.5 inch outside
diameter. The tube bundle is enclosed in a 92 inch diameter 316L stainless
steel shell. As with the evaporator, 125 F steam is fed to the steam chest of
the crystallizer to heat the brine as it flows down the inner walls of the tubes,
The steam is supplied at 68,000 pounds per hour. A mixture of liquid and vapor
falls into the crystallizer body, which is 27 feet in diameter by 12 feet. The
crystallizer operates at 109 F and a pressure of 55 mm absolute. Noncondensi-
bles are removed by a steam ejector. The steam requirement for the ejector is
700 pounds per hour 90 psig steam. The liquid in the crystallizer returns to
the recirculating pump, and the vapor passes through a mesh demister to a
shell-and-tube condenser.
4-39
-------
The crystallizer-condenser has 3,150 titanium tubes, each 18 feet long with an
outside diameter of 1 inc.h. The condenser shell measures 61 inches in diameter
and is fabricated from 316L stainless steel. The cooling water requirement for
the condenser is 14,000 gpm of 80 F water. Condensate is produced at 203 gpm.
The system design was based on a heat transfer coefficient of 250 Btu/
hour-sq ft-l°F. A 30 gpm stream is diverted from the recirculating brine at
the crystallizer and discharged to 50 acre solar evaporation pond. The pond
size is based on a net evaporation rate of 4.0 feet per year. ' The water
depth in the pond is 10 feet. A half acre emergency pond is provided to store
three days' flow to one of the six evaporators.
The layout of the vertical tube foaming evaporation plant is shown in Figure 13.
4.3.5 Process Economics
No vertical tube foaming evaporation system is in use at present, nor are any
field data available to permit a realistic cost estimate of a commercial size
system at the time this report is prepared. Consequently, the economics pre-
sented here is based on a conventional vertical tube evaporator plant designed
to operate with 125 F steam, as agreed by the Environmental Protection Agency.
This low temperature imposes a cost penalty resulting from the large steam chest
and line required by the high specific volume of steam. The plant cost for the
VTE, which serves as the basis for this estimate, represents the mean of two
quotations from established suppliers.
To attain a 30 year plant life without the need for major repairs or replacement,
the quotations were based on the use of smooth titanium for the evaporator tubes
and 316L stainless steel for all portions of the equipment in contact with brine.
This is admittedly conservative, since wetted parts of carbon steel could be
used under carefully controlled operating conditions. However, numerous cases
of failure of carbon steel components in evaporators indicate that, such careful
control is seldom exercised in practice. Therefore, the conservative approach
used in this study duplicates the material selection employed in current vapor
compression plants and used as the basis for Subsection 4.2.4 of this report.
4-40
-------
250--0"
CONDENSER
D
D
BUILDING FOR
CONTROLS, OFFICES
& SERVICES
PARKING
ĞJ
Figure 13. Plot plan for vertical tube foaming evaporation option.
-------
The total installed cost of the VTFE plant, including ponds for the storage of
blowdown, was estimated at $15,900,000.
The overall heat transfer coefficients that serve as the basis for evaporators
quoted to Bechtel ranged from 225 to 250 Btu/hr-ft 1°F. A value of 250 was
assumed for this analysis. This is substantially below the value of 500 postu-
lated for commercial VTFE evaporators by the Sea Water Conversion Laboratory of
the University of California. However, their coefficient is based on pilot
plant experience at the Laboratory where the feed was reconstituted. If a coef-
ficient of 500 can be substantiated by prolonged operation under actual field
conditions without frequent tube cleaning, then the plant cost presented in
this study would decrease by approximately 8 percent and the treatment costs
by 6 percent.
When heated by turbine exhaust steam, the VTFE in effect takes the place of a
portion of the power plant condenser. In a newly designed plant, as opposed to
a retrofit, the VTFE appears to reduce the size and cost of the condenser. In
practice, however, the full condenser would be installed to handle the exhaust
steam in the event of a VTFE shutdown.
A credit has been taken indirectly for the condensing function of the VTFE: in
this study, the plant has not been charged for the energy required to deliver
the cooling tower blowdown to the evaporator or for returning the distillate to
the cooling tower. Instead, it was assumed that it would merely function as a
bypass for some of the coolant normally pumped from the cooling tower to the
power condenser.
Because of the assumption of a no-cost heat source, the energy consumption
charged to the process is quite low, consisting chiefly of pumping power require-
ments. Consequently, the dependence of treatment cost on the price of power,
as shown in Figure 14, is very small. Also, as in the case of the VC plant dis-
cussed in Subsection 4.2.5, only 2 percent of the water content of the cooling
tower blowdown is assumed to require ponding. Thus, the slope of the "treatment
cost vs. pond cost" curves in Figure 15 is very low.
4-42
-------
O 5
LU
LU
FIXED
CHARGE
RATE
16%
o
o
o
12.5%
CO
O
o
LLJ
CC
I-
BASIS: POND COST $50,000/ACRE
I
234
ENERGY COST(C/kWh)
5
Figure 14. Effect of energy cost on VTFE treatment costs,
4-43
-------
111
oc
BASIS: ENERGY COST
4 C / kWh
O
o
o
o
o
t-
z
FIXED
CHARGE
RATE
16%
12.5%
40
50
60
POND COST { $1,000 / ACRE )
Figure 15. Effect of pond cost on VTFE treatment costs.
4-44
-------
If it is subsequently decided, for the purpose of upflow operation or for other
reasons, to operate the VTFE at higher temperatures, the heat will no longer be
"free." Steam at higher temperatures can be provided by extraction from the
turbine at some point above the normal exhaust pressure. A value can be assigned
to the extraction steam by the use of curves in Reference 5, updated to reflect
current costs, and the costs in this report could be adjusted accordingly.
No charge was included to cover the cost of the surfactant which, at the low
concentrations tested to date, is comparatively minor. It has been assumed that
the surfactant accompanies the blowdown stream to the solar evaporation pond,
where its presence is not objectionable. If a particular blowdown stream con-
tains salable salts, it may be necessary to free it of surfactant. In that
case, the removal cost would have to be balanced against the value of the
recovered salts.
The VTFE costs are compared with those of the other two processes in
Subsection 4.5.
4-45
-------
4.4 OVERALL PLANT WATER BALANCE
In the concentration of cooling tower blowdown, all of the processes analyzed in
this report produce a supply of purified water suitable for many uses in the
power plant. Among the potential uses are:
Fossil-fired power plants need water for the scrubbers of the
flue gas desulfurization system. This represents a compara-
tively minor requirement, and the water used here can be high
in dissolved solids, provided that it is not excessively scal-
ing in nature.
Coal burning plants consume rather small amounts of water,
again not necessarily of high quality, for ash sluicing.
Boiler makeup and water for flushing pump glands represent,
at most, 5 percent of the power plant water demand. Since
water of very high purity is required here, two-bed ion
exchange followed by mixed bed polishing is commonly employed.
For very pure feeds, such as the distillate from the VCE or
VTFE process, only the polishing step is required.
The major water requirement is for the cooling system. In
general, most of the feed to the cooling loop is for evapora-
tive cooling. The residual water constitutes the cooling
tower blowdown.
For simplicity of this study, it has been assumed that all the water recovered
by cooling tower blowdown concentration is returned to the cooling tower. This
procedure has a threefold advantage:
The quantity of makeup water to the cooling system is reduced,
thereby decreasing the cost of raw water treatment. Although
the evaluation of raw water treatment was considered to be
outside the scope of the present study, it should be con-
sidered in a future, more detailed analysis
Since a fraction of the feed to the cooling tower loop is
pure distilled water, less dissolved salt is introduced and,
hence, the flow rate of cooling tower blowdown is reduced.
Consequently, a smaller and less costly plant is required
for the concentration of the blowdown
The rate of discharge of brine from the concentrator is
decreased. This results in a lower ponding requirement.
4-46
-------
The comparison of process conditions for operation with and without recycle of
purified water is presented in the figures for RO and distillation processes.
Figures 16 and 17 use Water A and Water B, respectively, in a reverse osmosis
system. Figure 18 shows the effect of water recycle in a distillation system
using Water A. (See Tables 2 and 3 for the composition of feed waters A and B.)
The calculations are based on the following assumptions:
The cooling tower blowdown rate in all cases represents one-
fourth of its makeup flow and, consequently, the concentra-
tion of dissolved solids ("cycles" of concentration) is four
times that in the makeup
Either distillation process, VCE or VTFE, recovers 98 percent
of the water of the cooling tower blowdown
The RO process recovers 75 percent of the water of the cool-
ing tower blowdown
The salinity of the RO product is not altered by the water
recycle scheme
For the reference power plant, the cooling tower evaporation
rate is not influenced by recycle of product water from the
concentrator ,
From the figures, it is apparent that the recycle of product water from the con-
centrator to the cooling tower when RO is used has the following advantages:
A 20 percent reduction in makeup water required for the power
plant cooling system (Column 1)
A 16 percent reduction in the size of the plant for pretreating
and concentrating the cooling tower blowdown (Columns 3 and 4)
A 16 percent reduction in the flow of reject brine to the pond
(Column 5).
For the distillation processes, the advantages are:
A 25 percent reduction in the power plant coolant makeup
(Column 1)
A 25 percent reduction in the size of the evaporation plant
. (Column 3)
4-47
-------
WITHOUT RECYCLE
WITH RECYCLE
i
.>
00
İ
COOLING TOWER SYSTEM
RO PROCESS
SiO
COOLING TOWER SYSTEM RO PROCESS
PROCESS
WITHOUT
RECYCLE
WITH
RECYCLE
"~~ -^STREAM
FLOW (gpm)
TDS (ppm)
SiO2 (ppm)
FLOW (gpm)
TDS (ppm)
SiO2 (ppm)
MAKEUP
1
8000
1880
37.5
6420
1880
37.5
EVAPORATION
2
6000
0
0
6000
0
0
SLOWDOWN
3
2000
7522
150
1700
7500
150
TREATED
SLOWDOWN 4
2000
7619
37.5
1700
7550
37.5
REJECT TO
POND 5
500
28,855
125
425
28,855
125
PRODUCT
6
1500
475
8.5
1275
475
8.5
Figure 16. Comparison of flows with and without recycle of purified
water using Water A and reverse osmosis.
-------
WITHOUT RECYCLE
'>,
2
İ
WITH RECYCLE
s~>
2
Ca
COOLING TOWER SYSTEM RO PROCESS
COOLING TOWER SYSTEM RO PROCESS
PROCESS
WITHOUT
RECYCLE
WITH
RECYCLE
-^STREAM
PARAMETER ^
FLOW(gpm)
Ca (ppm)
SO4 (ppm)
TDS (ppm)
FLOW (gpm)
Ca (ppm)
SO4 (ppm)
TDS (ppm)
MAKEUP
1
8000
133
955
1391
6410
133
955
1391
EVAPORATION
2
6000
0
0
0
6000
0
0
0
SLOWDOWN
3
2000
533
3820
5562
1700
533
3890
5730
TREATED
SLOWDOWN
4
2000
50
3847
5650
1700
50
3920
5820
REJECT TO
POND
5
500
200
14.638
21,358
425
200
15,000
22,100
PRODUCT
6
1500
0
214
384
1275
0
214
384
Figure 17. Comparison of flows with and without recycle of purified
water using Water B and reverse osmosis.
-------
WITHOUT RECYCLE
WITH RECYCLE
İ
-C-
I
COOLING TOWER
SYSTEM
DISTILLATION
COOLING TOWER
SYSTEM
DISTILLATION
PROCESS
WITHOUT
RECYCLE
WITH
RECYCLE
"--^STREAM
PARAMETeft"-^^
FLOW(gpm)
TDS (ppm)
SiO2(Ppm)
FLOW(gpm)
TDS (ppm)
SiO2 (ppm)
MAKEUP
1
8000
1880
37.5
6031
1880
37.5
EVAPORATION
2
6000
0
0
6000
0
0
SLOWDOWN
3
2000
7522
150
1500
7520
150
BRINE TO
POND 4
40
SAT
SAT
30
SAT
SAT
PRODUCT
5
1960
0
0
1470
0
0
Figure 18. Comparison of flows with and without recycle of purified
water using Water A and distillation.
-------
A 23 percent reduction in the flow of distillation plant
blowdown to the pond (Column A).
As a result of the above comparisons, the recycle of product water from the
blowdown cortcentrator back to the cooling tower has been incorporated in each
of the three process flow sheets in Subsections A.I, A.2, and A.3.
4-51
-------
4.5 COMPARISON OF PROCESSES
Before comparing processes having characteristics as diverse as the three pro-
cesses analyzed in this report, it was necessary to establish a common basis
for the analysis. The problems stem from the fact that two of the processes
use mechanical energy, and the third uses heat which, under the conditions
postulated here, is considered to have zero cost. In addition, the membrane
process produces a purified water stream substantially inferior in total dis-
solved solids content to the other two.
To compensate for the differences in product water quality, the water was
assumed to recirculate to the cooling tower in every case. Thus, the process
that delivers the purer product is automatically credited by virtue of the
resulting decrease in makeup to the cooling tower system and in the size of
plant required to treat the cooling tower blowdown. These relationships are
apparent in Figures 16, 17, and 18 and in Table 8. The ponding requirements
of the processes studied here were taken into consideration by calculating the
pond area required for a 30 year plant life and considering this cost as part
of the investment in the cooling tower blowdown treatment plant.
Table 9 summarizes the costs and energy consumption of the three processes. The
values listed there are based on an assumed pond cost of $50,000 per acre, an
energy cost of 4c per kUh, and a 12.5 percent fixed charge rate. From the
standpoint of plant investment, reverse osmosis is definitely superior to the
two distillation processes. In total operating cost, RO is only slightly
inferior to the VTFE.
If it becomes necessary to raise the operating temperature of the VTFE, the favor-
able energy consumption of RO would become even more pronounced by comparison,
and its operating cost could conceivably be lower than that of the VTFE.
These and other attributes of the three processes are listed in Table 10. From
this comparison, it appears that no process is definitely superior to the other
two in all respects. The picture promises to change with time and as a result
4-52
-------
of further work. For example, successful field testing of the VTFE will pro-
vide confidence in the process, for which there is currently no basis. It
could also result in more favorable heat transfer rates than were conservatively
assumed in this report.
It was concluded from the comparison in Table 10, that analysis should continue
in order to examine the merits of a combination process. This would involve
concentration of'cooling tower blowdown by a membrane process, which will
deliver its reject stream to a distillation plant to minimize the volume of
liquid waste requiring ponding.
TABLE 8. EFFECT OF RECOVERY AND PRODUCT
PURITY ON SYSTEM DESIGN
Reverse Osmosis Distillation
Requirement for makeup water
to power plant cooling
system (gpm) 6,420 6,031
Cooling tower blowdown
requiring treatment (gpm) : 1,700 1,500
Blowdown flow from treatment
system (gpm) 425 30
Ponding area required for
30-year plant life (acres) 160 ' 50
Basis: 700 MWe power plant.
Ponds 10 feet deep.
4-53
-------
TABLE 9. COMPARISON OF COSTS AND ENERGY CONSUMPTION
Plant investment, 1st quarter
'79 ( million dollars)
Treating plant
Ponds
Total
Annual cost (million dollars)
per year) (O
Capital cost
O&M cost
Supplies
Electrical energy
Labor
Total O&M cost
Total annual cost
(2)
Energy consumption
Electricity (million
kWh/year)
12
Steam (10 Btu/year)
Reverse
Osmosis
3.7
10.2
13.9
1.7
Vertical Tube Vapor
Foaming Evaporation Compression
0.7
0.2
0.3
1.2
2.9
5.3
0.1
0.3
0.2
12.5
3.4
15.9
2.0
0.1
2.6
0.2
11.8
3.4
15.2
1.9
0.6 2.9
2.6 4.8
6.7 (not free) 63.9
7.0 (no cost)
(1) Basis: 30-year plant life, ponds at $50,000/acre, electricity at 4p/kWh,
12.5 percent fixed charge rate. See above for a discussion of effects of
variations in these assumptions.
(2) Energy cost is included in O&M cost above.
Note: The vertical tube foaming evaporation (VTFE) costs are estimates
based on VTE experience. They may change substantially as VTFE
operating data become available.
4-54
-------
TABLE 10. OVERALL COMPARISON OF PROCESSES
Ponding requirements
(1)
Long term-related operating
experience on cooling
tower blowdown
Reverse Vertical Tube
Osmosis Foaming Evaporation
Large Moderate
Minimal None
Vapor
Compression
Evaporation
Moderate
Considerable
and successful
Modular construction of sub-
units of each train
Sensitivity to upsets
in pretreatment
Product quality
Standard
procedure
Very
sensitive
Fair
(475 ppm)
Not practical
Minor
High
(10 ppm)
Not practical
High
High
(10 ppm)
(1) Ponding cost is included in the plant investment and the calculation of
annual cost in Table 9-
4-55
-------
SECTION 5
TEST PLAN FOR VTFE SHAKEDOWN TESTS
The following test plan for the shakedown of the VTFE at the Sea Water Conversion
Laboratory (SWCL) was developed with the cooperation of Dr. Hugo H. Sephton,
principal investigator at the SWCL. See Figure 19 for a photograph of the pilot
plant. The tests are directed toward assuring the structural integrity and the
safe and smooth operation of the pilot plant equipment. Additional improvements
in the equipment and refinements in operation and controls may result from the
field test program which is to follow this shakedown phase.
Shakedown testing will be performed and improvements in process control imple-
mented for each of the facility units, i.e., boiler, VTE evaporator, VTE crys-
tallizer, and condenser. The vacuum tightness of all welds will be checked.
Process control elements will include: steam temperature control; brine distri-
bution; adequate bundle venting, couplings, valves, etc.; coolant flow and tem-
perature control for controlling the AT applied; boiler controls and steam
desuperheating; condensate and distillate removal and flow measurement; tempera-
ture measurement and recording; and startup and shutdown procedures. Calibration
will be-performed on all devices for measuring process parameters.
To protect the pilot plant, temperature sensors will be provided. The sensors
will actuate high- and low-temperature alarms and will shut the plant down in
the event of temperature excursions. Level controllers will control either the
feed or the blowdown rate of the evaporator in reponse to the level in the evap-
orator sump. The variable not regulated by the level controller will be regu-
lated by a salinity monitor. A level controller will regulate the rate of dis-
charge from the crystallizer.
5-1
-------
r
ro
Figure 19. Photograph of VTFE pilot plant.
-------
Three types of feed waters will be tested during the shakedown period:
Pure water (For convenience, tap water will be used.)
Seawater diluted with tap water to approximate a cooling tower
blowdown composition
Diluted seawater containing the preferred surfactant.
The surfactant tests have been relegated to the end of the shakedown tests,
since previous experience shows that it is very difficult to remove surfactants
from the system once they have been injected.
With each of these feeds, the feed liquor will be reconstituted by combining the
distillate and blowdown in order to minimize the cost of the shakedown tests.
It is realized, however, that reconstituting the feed does not present a realis-
tic picture of sludge and scale formation or of corrosion phenomena. Conse-
quently, it is anticipated that once field testing is initiated at a power plant
site, an actual cooling tower blowdown will be used as feed and the concentrate
will be discarded or ponded.
A matrix of tests will be instituted, incorporating the following variables:
1. Operating temperature
a. Low temperature (about 125°F), representing waste heat
from a condensing type of steam turbine
b. Medium temperature (about 150°F), representing heat from
steam discharged from a back-pressure turbine
c. High temperature (about 220°F), as would be encountered in
the use of prime steam after pressure reduction and
desuperheating.
2. Flow regime
a. Upflow, using the flashing of the brine to drive the two-
phase fluid upward and out of each tube (Note that condi-
tions la and Ib above many not provide sufficient hydraulic
driving force to permit upflow operatipn. Measurements of
pressure drops across the orifice plate and up the tubes will
define the limits of upflow operation possible without pumps.)
5-3
-------
b. Downflow, using a recirculation pump to raise the brine to
the plenum that supplies the evaporator tubes
3. Surfactant addition
a. Without surfactant, to establish a base line for comparison
b. With surfactant, using the type of surfactant and concen-
tration demonstrated as optimum in the preliminary studies
performed at the Sea Water Conversion Laboratory.
Under 2b (downflow), the effect of flow rate of the brine should also be inves-
tigated. This, in turn, controls the thickness of the fluid film on the inte-
rior surfaces of the tubes. Conversely, in upflow the liquid level in the
evaporator tubes controls the mass flow rate through the tubes and, consequently,
will be varied to the extent possible.
The measurements required, as a minimum, are:
Evaporator
Steam supply
Pressure
Temperature
Flow of condensate
Brine evaporation
Feed: Temperature
Concentration of dissolved solids
Flow rate
pH
Vapor: Temperature
Pressure
Venting rate
Concentrate: Temperature
Concentration of dissolved solids (salinity)
Concentration and composition of suspended
solids
PH
5-4
-------
Condenser
Condensate
Temperature
Flow rate
Purity (in tests with saline feeds only)*
Coolant
Temperature entering
Temperature leaving
Flow rate
Crystallizer
Steam supply
Pressure
Temperature
Flow of condensate
Brine evaporation
Vapor: Temperature
Pressure
Venting rate
Concentrate: Temperature
Concentration (salinity)
Flow rate
In addition, it is essential to determine the uniformity of brine distribution
to the evaporator tubes and the pattern of steam disengagement in the evaporator.
Quantitative data are highly desirable if attainable. At the very least, records
will be kept of periodic visual observations, preferably supported by photographs.
*If possible, it is desirable to obtain separate data on flow rate and purity of
vapors from both the evaporator and the crystallizer.
5-5
-------
Uniform brine distribution will not only improve evaporator efficiency, but will
avoid the possibility of baked deposits in the tubes. As for steam disengage-
ment, a poor velocity pattern will cause carryover of a large quantity of brine
along with the vapor.
As a check, a material balance will be prepared, comparing the feed entering the
evaporator and the concentrate leaving the evaporator. A similar material bal-
ance will be calculated around the crystallizer. Such material balances serve
as cross-checks on flow measurements and analytical results. They have the addi-
tional function of providing an early indication of buildup of deposits in the
lines and equipment.
Finally, whenever the equipment is opened, sludge and scale deposits will be
removed, examined visually, weighed, and analyzed.
The data enumerated above will:
Permit the calculation of mean effective overall heat transfer
coefficients
Permit separate calculation of economy ratio of the evaporator
and the crystallizer
Alert the operator to the scaling or fouling of the condenser
tubes
Give an indication of the malfunctioning of the plant components
Permit the calculation of the thickness of the evaporating film
of liquid on the interior tube walls
Serve as a guide in the "fine tuning" of the pilot plant
The longevity and the economics of an evaporator-crystallizer are strongly influ-
enced by corrosion. Consequently, the following tests will be performed before
startup and after completion of the shakedown tests:
Measure the wall thickness of evaporator and crystallizer tubes
from several representative locations (near the outside of the
5-6
-------
bundle, near the center, and at some intermediate location).
Such measurements will be made on the portions of the tubes
that protrude from the upper tubesheet. Measurements will
be performed before the start and at the end of the shakedown
testing and will be recorded together with visual observations
of the tube surfaces.
Install, in the vessels and lines, coupons representing the
materials of construction of the plant and also a few poten-
tially desirable alternatives. Two types of coupons are
required. One group will consist of unstressed coupons,
weighed and measured before installation and again at the
end of the shakedown tests. The second group will consist
of stressed U-bend coupons of the same materials.
Measure the thickness of the pump impeller at several speci-
fic loations. Whenever the plant is opened, inspect for rust
and pitting and take photos.
With inside calipers, check changes in the inside diameter of
piping in selected locations.
Determine corrosion/erosion of elbows by observing them at
the start and end of shakedown testing. Several nondestruc-
tive testing methods are available, among them radiography
and ultrasonic testing. Efforts will be made to locate
sources of rental equipment or testing services to determine
the most economical procedure.
The above program is directed toward the shakedown phase of the overall test
program. Many of the tests outlined in this section, however, are suitable for
continued monitoring during field testing and will be repeated at the termina-
tion of work at each power plant site.
The shakedown program was initiated during the last two weeks of activity under
EPA Contract No. 68-02-2616 with Bechtel. During the shakedown, the pilot evap-
orator was tested in the upflow mode, the crystallizer in the downflow mode.
(Please refer to Subsection A.3.) The feed consisted of fresh water. The flow
up the evaporator tubes was accomplished with the aid of a pump, in contrast to
the hydrothermal driving force planned for the field tests.
5-7
-------
In tests witnessed by Bechtel, the evaporator feed consisted of a mixture of
coolant water and distillate. The coolant water was preheated by passing
through the vent condenser of the evaporator to a temperature approaching that
of the recirculating brine, which was at 111.3 F. The preheated coolant water
and distillate were mixed with the recirculating brine and then delivered to
the evaporator at the rate of 600 gpm. No acid or surfactant was added to
the feed.
An oil fired boiler delivered steam at 25 psig. The steam pressure was reduced
by a temperature-controlled throttling valve and desuperheated by the injection
of a spray of preheated condensate. Excess moisture was stripped from the
steam in a cyclone separator designed at the SWCL. The temperature of the
desuperheated steam was 132.3 F, and the temperature of the saturated steam in
the steam chest was 131.8 F.
The vapor generated by evaporation of the feed in the evaporator (and of the
concentrate in the crystallizer when heated) was condensed in the two-pass con-
denser, where heat was rejected to cooling water flowing through the tubes at
the rate of 1,401,120 pounds per hour. The coolant temperature was maintained
at 107.5 F by varying the ratio of fresh coo!
Distillate was produced at a rate of 10 gpm.
at 107.5 F by varying the ratio of fresh coolant to water recycled from a sump.
By visual observation through sight glasses in the vapor dome, it was concluded
that the recirculating brine was distributed fairly uniformly among the evap-
orator tubes. Under the operating conditions tested, the evolved vapor did not
appear to produce excessive scatter of the droplets of liquid emerging from the
evaporator tubes.
Operation of the plant was quiet and steady. The liquid levels and salinity
were maintained reasonably constant by the control valves in the respective feed
and blowdown lines. The pH and salinity were recorded automatically on strip
charts. Plant startup was relatively simple; equilibrium conditions were attained
in approximately 90 minutes from firing of the boiler. Shutdown was uneventful.
5-8
-------
REFERENCES
1) "Concentration of Brines by Spray Evaporation," U.S. Department of the
Interior, Office of Saline Water, Research and Development Progress Report
No. 764, Contract No. 14-01-0001-2276, January 1972, p. 115.
2) Annual Report, Volume 1, Vertical Tube Evaporator Multistage Flash Test Bed
Plant, Freeport, Texas, Bechtel Corporation Report to the Office of Saline
Water, Department of the Interior, Contract No. 14-30-3060 (April 1, 1972
to March 31, 1973).
3) Houle, J. F., and W. T. Buhrig, "Performance of the Freeport, Texas VTE/MSF
Plant," Fourth International Symposium on Fresh Water from the Sea, Heidel-
berg, Germany, _1, 313-325 (September 9 - 14, 1973).
4) Sephton, H. H., "Renovation of Power Plant Cooling Tower Slowdown for
Recycle by Evaporation-Crystallization with Interface Enhancement," Report
to the Environmental Protection Agency under Project No. R-803257
(March 1977).
5) Jones, J. E., Jr., et al., "Coupling Technology for Dual-Purpose Nuclear
Desalting Plants," ORNL/TM-4471 (November 1976).
6) Goldman, E. and P. Kelleher, "Water Reuse in Fossil-Fueled Power Stations,"
Complete Water Reuse, Cecil, L. K. (ed.), New York, N.Y., American Insti-
tute of Chemical Engineers, April 1973, p. 240-249.
7) Langelier, W. F., "The Analytical Control of Anti-Corrosion Water Treatment,"
Jour. AWWA, ^8, No. 10, 1500-1521 (October 1936).
8) Langelier, W. F., "Chemical Equilibria in Water Treatment," Jour. AWWA, 38,
No. 2, 169-178 (February 1946).
9) Langelier, W. F., "Effect of temperature on the pH of Natural Waters," Jour.
AWWA, 38, No. 2, 179-185 (February.1946).
R-l
-------
Appendix A
BASIC WATER CHEMISTRY
The composition of water delivered to the cooling system of a power plant depends
on plant location and, in addition, is subject to seasonal variations. Of all
the constituents of raw water, the most important from the standpoint of both the
cooling tower and the blowdown treatment system are the scale formers. Three
types of scale merit special consideration: calcium carbonate, calcium sulfate,
and silica.
CALCIUM CARBONATE SCALING
The bicarbonate ion, which is present in many water supplies, is in equilibrium
with dissolved carbon dioxide:
2HCO~ C0~~+ H20 + C0 (1)
where C0Ğ. . represents molecular carbon dioxide dissolved in the water. The
dissolved carbon dioxide, in turn, is in equilibrium with C0_ in the atmosphere:
C°2(DIS)^ C°2(GAS) (2)
When water containing bicarbonate ion contacts air in the cooling tower, the dis
solved C02 is reduced to a value determined by the Henry's Law coefficient and
this, consequently, drives reaction (1) to the right. The resulting increase in
carbonate ion, together with the increased concentration caused by evaporation,
results in the deposition of calcium carbonate scale in the cooling tower and
associated piping:
Ca*"1" + C0~~ CaC03 (3)
After depositing the scale, the water is still saturated in CaCO~, which has a
negative temperature coefficient of solubility. Thus, when the water is heated
A-l
-------
by passage through the power plant condenser, further deposition of CaCCL scale
occurs, but now in the condenser tubes.
One solution to this problem, used in certain types of distillation processes,
is the destruction of the bicarbonate ion by acid addition. In this process,
sulfuric acid is preferred for acidification because of its low cost:
2HCO~ + H2S04 S0~"+ 2H20 + 2C0 (4)
The dissolved CO- is stripped from the acidulated water, in accordance with
reaction (2), by a countercurrent stream of air in the decarbonator .
A commonly used criterion for the stability of water against calcium carbonate
precipitation was developed
is defined by the equation:
precipitation was developed by Langelier. ' ' ' The Langelier Index
L.I. = PH - pH , where (5)
m s
L.I. = Langelier Index
pH = the measured pH of the water in question
m
pH = the "saturation" pH.
S
The "saturation" pH was shown by Langelier to be:
pH = (pKi - pK') + pCa + pAlk, where (6)
S Z S
(pK'-pK1) = a correction depending primarily on the ion
s strength of the solution
Ca = the molar concentration of Ca ions present.
Alk = the alkalinity, expressed as moles per liter of
CaC03 corresponding to the CO^", HCO^, and OH~
in the solution
p = indicates the logarithm of the reciprocal of the
quantity indicated. For example, pCa = log (I/ [Ca ]).
Both pCa and pAlk can be read directly from Figure A-l. The value of (pK'-pK1) can
z. s
be found in Table A-l, which takes the effect of temperature into consideration.
A-2
-------
4.0
3.5
o
Q
5 3.0
2.5
2.0
1.5
\
X
\
\
\
\
\
s
s
S
S
^
^
^
^
\
\
\
S
\
s
\
K
- ALKAL
-CALCIL
\\
V
INIT
JM
^
YAI
^
iCa
CO
s
\
3
\
\
S
\
s
V
2 3 4 5 6 8 10 20 30 40 50 100 200 300 500 1000
PARTS PER MILLION
Figure A-l. Conversion from alkalinity or Ca-H- to the
respective p values.
A-3
-------
Table A-l is entered at the ionic strength of the solution, defined by:
2
V = % EC v , where (7)
n n
y = ionic strength
C = the concentration of ion "n" in moles/liter
n
v = the valence of ion "n"
n
The Langelier criterion described above is useful for predicting both the scaling
and the corrosive tendency of various waters. When L.I. equals zero, the water
is stable and there is no tendency for calcium carbonate scale to form. This is
equally true for negative values which indicate, in addition, a tendency toward
corrosion of metals, particularly the ferrous metals. A positive L.I. signifies
decreased corrosiveness and a tendency to deposit calcium carbonate scale.
CALCIUM SULFATE SCALING
The solubility of calcium sulfate is quite limited. In addition, as in the case
of calcium carbonate, it has a negative temperature coefficient of solubility.
That is, the solubility of calcium sulfate decreases as the temperature of the
water rises. Thus it poses the danger of calcium sulfate scale formation in the
0
tubes of the power plant condenser.
The calcium sulfate solubility is one of the constraints on the permissible con-
centration of the cooling water in the cooling tower loop. In addition to the
sulfate ion originally in the feed water to the plant, reaction (A) contributes
additional sulfate. One alternative is to soften the feed water by addition of
slaked lime:
Ca + 2HCO+ Ca(OH)2 > 2CaC03 + 2H20 (8)
If the water contains calcium ions in excess of the HCO present, that is, if it
contains noncarbonate hardness, soda ash must be added: '
+ 2Na+ (9)
A-4
-------
TABLE A-l. VALUES OF pK' AND .pK' AT 25 C FOR VARIOUS STRENGTHS AND OF THE
DIFFERENCE (pK' - pK') FOR VARIOUS TEMPERATURES^Ref' 6)
f- S
IONIC
STRENGTH
0000
0005
.001
002
.003
.004
.005
.006
.007
.008
.009
.010
.011 '
.012
.013
.. .014
. .015
.016
.017
.018
. .019
.020
TOTAL
DI3-
BOI.VKD
8OL1D3
o
20
40
80
120-
160
200
2-40
280
320
360
400
440
480
520
560
GOO
640
680
720
760
.800
pK';
10 26
10 26
10.26
10 25
10.25
10.24
10 24
10 24
10.23
10 23
10.22
10.22
10.22
10.21
10.21
10.20
10.20
10.20
10.19
10.19
10.18
10.18
25'C.
pK'.
8 32
8 23
8 19
8 14
8.10
8 07
S 04
8 01
7.98
7 96
7.94
7.92
7.90
7.8S
7.8U
7.85
7. S3
7.81
7.80
7.78
7.77
7.76
pK t -
pK'.
1 94
2 03
2 07
2 11
2.15
2 17
2 20
2 23
2.25
2 27
2. 28
2.30
2.32
2.33
2.35
2.30
2.37
2.39
2.40
2.41
2.41
2.42
o-c.
2 20
2 29
2 33
2 37
2.41
2 43
2 46
2 49
2 51
2 53
2.54
2.56
2.58
2.59
2. 01
2.62
2.63
2.65
2.66
2.67
2.67
2. 68
io-C.
2 09
2 18
2 22
2 26
2.30
2.32
2 35
2 38
2.40
2 42
2.43
2.45
2.47
2.49
2.50
2.51
2.52
2.54
2.55
2.56
2.57
2.58
20'C.
1 99
2 08
2.12
2 16
2.20
2.22
2 25
2 28
2.30
2 32
2.33
2.35
2.37
2.39
2.40
2.41
2.42
2.44
2.45
2.46
2.47
2.48
pK', -
we.
1 73
1 82
1.86
1 90
1.94
1.96
1 99
2 03
2.05
2 07
2.08
2.10
2.12
2.13
2.15
2.16
2.17
2.19
2.20
2.21
2.21
2.22
pK'.)
60'C.
1 65
1 74
1 78
1 82
1.86
1.88
1 91
1 94
1.96
1 98
1.99
2.01
2.03
2.04
2.06
2.07
2. OS
2.10
2.11
2.12
2.12
2.13
70-C.
1 58
1 67
1 71
1 75
1.79
1.81
1 84
1 87
1.89
1.91
1.92
1.94
1.96
1.97
1 .99
2.00
2.01
2.03
2.04
2.0.3
2.05
2. 06
80'C.
1 51
1 60
1.64
1 68
1.72
1.74
1 77
1 80
1.82
1 84
1.85
1.87
1.89
1.90
1.9'J
1.93
1.94
1.96
1.97
1.98
1.9S
1.99
we.
1 44
1 53
1.57
1 61
1.65
1.67
1.70
1.73
1.75
1 77
1.78
1.80
1.82
1.83
1.85
1.86
1.87
.89
.90
.91
.91
.92
A-5
-------
Since the cooling tower water is concentrated almost to the saturation value of
calcium sulfate (unless silica is limiting see Subsection 4.1.3), a second
softening step is required before using any blowdown concentration process that
cannot tolerate scale.
SILICA SCALING
Although the solubility of silica varies somewhat with the water chemistry, many
water technologists arbitrarily set an upper limit of 150 to 200 ppm on silica
content. If not already limited by the danger of calcium sulfate scale, the
water in the cooling tower basin is generally discharged to waste when its
silica content reaches the predetermined value.
To permit evaporation in the cooling tower to a high degree of concentration
(frequently referred to as "cycles of concentration"), the feed water is sub-
jected to a silica reduction treatment consisting of coprecipitation of the
silica with magnesium hydroxide. The removal of silica by magnesium hydroxide
precipitation has been described as a chemical reaction by some investigators,
absorption by others. The magnesium can be added in the form of very fine or
"activated" magnesium oxide, or the hydroxide can be formed in situ by the pre-
cipitation of magnesium ions present in the water, by adding excess lime:
Mg"^ + Ca(OH)2 -~ Mg(OH)2 + Ca** (10)
The calcium ions introduced in this process are then removed by reactions (8)
and (9).
The softened water is finally acidified to adjust its pH to prevent post-
precipitation of calcium carbonate, as discussed in Subsection 4.1.3.
A-6
-------
Appendix B
CONVERTING UNITS OF MEASURE
Environmental Protection Agency policy is to express all measurements used in
Agency documents in metric units. In this report, however, to avoid undue costs
or lack of clarity, English units are used throughout. Conversion factors from
English to metric units are given below:
To Convert From
Btu/ft2-hr-l°F
scfm (60°F)
cfm
°F
ft
ft/hr
ft/sec
ft2
c 2/
ft /tons per day
gal/mcf
gpm
2
gpm/ft
gr/scf
in.
in. H-O
Ib
Ib-moles
Ib-moles/hr
2
Ib-moles/hr ft
Ib-moles/min
psia
To
J/m2-sec.l°K
Mm /hr (0°C)
m /hr
°C
m
m/hr
m/sec
2
m
2
m /metric tons per day
1/m3
1/min
1/min/m
gm/m
cm
mm Hg
gm
gm-moles
gm-moles/min
gm-moles /min/m
gm-moles/ sec
kilopascal
Multiply By
5.677
1.61
1.70
(°F-32)/1.8
0.305
0.305
0.305
0.0929
0.102
0.134
3.79
40.8
2.29
2.54
1.87
454
454
7.56
81.4
7.56
6.895
B-l
-------
TECHNICAL REPORT DATA
(Please read litaructiom on the reverse before completing)
1. REPORT NO. ,
EPA-600/7-79-220
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Assessment of Three Technologies for the Treatment of
Cooling Tower Slowdown
5. REPORT DATE
September 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E.H. Houle, A.N. Rogers, M.C. Weeks, S.C. May and
V.C. Van der Mast
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Bechtel National, Inc.
P.O. Box 3965
San Francisco, CA 94119
10. PROGRAM ELEMENT NO.
INE 624 A
11. CONTRACT/GRANT NO.
68-02-2616/08
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 9/78 - 7/79
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES
919/541-2898.
IERL-RTP project officer is Michael C. Osborne, Mail Drop 61,
16. ABSTRACT
The report gives results of analyses of three methods for treating cooling
tower blowdown: vapor compression evaporation (VCE), reverse osmosis (RO), and
vertical tube foaming evaporation (VTFE). The two evaporative processes produce
pure water (approximately 10 ppm dissoved solids). RO produces water of lower
purity (about 500 ppm) but adequate for many uses in the power plant or for return
to the cooling tower. VCE has been used successfully in commercial power plants;
the evaporative processes have no plants in operation on cooling tower blowdown.
Plant investment is strongly influenced by the cost of lined ponds required to
evaporate the treatment plant blowdown and store the residual salts. Consequently,
the RO plant investment is almost in the range of that of VTFE and VCE plants.
In total capital plus operating costs, VTFE has a distinct advantage over RO and
VCE becuase its source of energy is waste heat to which a zero value has been
assigned. The VTFE economics is based on vertical tube evaporator experience
(without adding surfactant). If field pilot tests substantiate previous laboratory
results, the economics of the VTFE may prove to be even more favorable.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Pollution
Industrial Processes
Analyzing
Cooling Towers
Evaporative Cooling
Evaporation
Osmosis
Foaming
Electric Power
Plants
Ponds
Pollution Control
Stationary Sources
Vapor Compression
Reverse Osmosis
13B
13H
14B
13A, 07A
07D
10B
08H
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
86
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
C-l
------- |