EPA-650/2-74-I26-Q
DECEMBER 1974
Environmental Protection Technology Series
-------
EPA-650/2-74-126-0
PROCEEDINGS:
SYMPOSIUM
ON FLUE GAS DESULFURIZATION
ATLANTA, NOVEMBER 1974
VOLUME I
ROAP NO. 21ACX-AA
Program Element No. 1AB013
Chairman: E.L. Plyler
Vice-Chairman: W.H. Ponder
Sponsored by
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, N. C. 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
December 1974
-------
EPA REVIEW NOTICE
This report has been reviewed by the National Environmental Research
Center - Research Triangle Park, Office of Research and Development,
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency , nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environ-
mental Protection Agency, have'been grouped into series. These broad
categories were established to facilitate further development and applica-
tion of environmental technology. Elimination of traditional grouping was
consciously planned to foster technology transfer and maximum interface
in related fields. These 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
9- MISCELLANEOUS
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to
develop and demonstrate instrumentation, equipment and methodology
to repair or prevent environmental degradation from point and non-
point sources of pollution. This work provides the new or improved
technology required for the control and treatment of pollution sources
to meet environmental quality standards.
This document is available to the public for sale through the National
Technical Information Service, Springfield, Virginia 22161.
Publication No. EPA-650/2-74-126-a
11
-------
PREFACE
The development of technologies to control sulfur dioxide emissions is
an issue of national importance. Indications that the use of sulfur-containing
fossil fuels to generate electricity will increase by 50 percent by 1985,
recognition that more than half of all sulfur dioxide (SC^) emissions are caused
by electric power generation, and documentation of the deleterious effects
caused by such emissions make the acceleration of development to commercial
application of SC>2 control technology one of the most important goals of the
U.S. Environmental Protection Agency (EPA). The Control Systems Laboratory
(CSL), as part of EPA's Office of Research and Development, has accelerated
the development of flue gas desulfurization (FGD) technology so that it is now
in the process of commercialization. Each year, CSL sponsors a Symposium which
brings together users and developers of the technology for open exchange and
discussion of their experience and results. The 1974 Symposium was held on
November 4-7 at the Sheraton-Biltmore Hotel in Atlanta.
Commercial application of full-scale FGD processes was the primary topic
of the Symposium which provided a means to transfer the latest information
on the status, development and assessment of current FGD processes to developers,
vendors, potential users, and those concerned with regulatory deadlines and
enforcement. The Symposium was attended by more than 600 national and inter-
national representatives of utilities, vendors, government, and universities.
One-third of the presentation was assessment and status reports from utilities
which are utilizing full-scale FGD systems. Information from small- and
large-scale pilot testing of alternate FGD systems, developments in FGD
by-product disposal and utilization, and cost studies of FGD application
completed the extensive overview of FGD technology.
In concluding remarks, it was noted by an international FGD technology
consultant that the Symposium is recognized as the world's most significant
conference on S02 pollution control by FGD technology. The growth of the
FGD Symposium to this latest, widely attended, in-depth information exchange
forum supports the conclusion that FGD is the only viable near-term alternative,
other than the burning of clean fuels, that will permit compliance with current
regulatory requirements.
The contents of these Proceedings are comprised of copies of the
participating authors' papers as received. Although the papers have been
reviewed and approved for publication by the Environmental Protection Agency,
approval does not indicate that the contents necessarily reflect the views
and/or policies of the Agency. The mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
As supplies permit, copies of the Proceedings are available free of
charge and may be obtained by contacting the Air Pollution Technical
Information Center, Environmental Protection Agency, Research Triangle
Park, North Carolina 27711.
111
-------
-------
CONTENTS
TITLE ' PAGE
VOLUME I
OPENING SESSION
Keynote Address - THE ROLE OF ENVIRONMENTAL HEALTH ASSESSMENT
IN THE CONTROL OF AIR POLLUTION
John Finklea, Environmental Protection Agency,
National Environmental Research Center,
Research Triangle Park, North Carolina ........... 1
FLUE GAS DESULFURIZATION AND OTHER ALTERNATIVES FOR PRODUCING
ELECTRICITY FROM COAL
S.J. Gage, Environmental Protection Agency,
Washington, D. C ..... . ................ 5
STATUS OF FLUE GAS DESULFURIZATION SYSTEMS IN THE
UNITED STATES
T.W. Devitt and F. Zada, PEDCo-Environmental, Inc.,
Cincinnati, Ohio ...................... } 7
STATUS OF FLUE GAS DESULFURIZATION TECHNOLOGY IN JAPAN
J. Ando, Chuo University, Tokyo, Japan ........... 125
COST COMPARISONS OF FLUE GAS DESULFURIZATION SYSTEMS
G.G. McGlamery and R.L. Torstrick,
Tennessee Valley Authority,
Muscle Shoals, Alabama ..................
NON-REGENERABLE PROCESSES SESSION
EPA/RTP PILOT STUDIES RELATED TO UNSATURATED OPERATION OF
LIME AND LIMESTONE SCRUBBERS
R.H. Borgwardt, Environmental Protection Agency,
Research Triangle Park, North Carolina ........... 225
LIMESTONE AND LIME TEST RESULTS AT THE EPA ALKALI
SCRUBBING TEST FACILITY AT THE TVA SHAWNEE POWER PLANT
M. Epstein, Bechtel Corporation,
San Francisco, California ................. 241
OPERATIONAL STATUS AND PERFORMANCE OF THE ARIZONA PUBLIC
SERVICE COMPANY FLUE GAS DESULFURIZATION SYSTEM AT THE
CHOLLA STATION
L.K. Mundth, Arizona Public Service,
Phoenix, Arizona ........ . ............. 307
WET SCRUBBER OPERATING EXPERIENCE AT LA CYGNE
STATION UNIT NO. 1
C.F. McDaniel, Kansas City Power and Light,
Kansas City, Missouri .................. . 319
-------
TITLE PAGE
DUQUESNE LIGHT COMPANY, PHILLIPS POWER STATION,
LIME SCRUBBING FACILITY
S.L. Pernick, Jr. and R.G. Knight,
Duquesne Light Company,
Pittsburgh, Pennsylvania 329
THE HORIZONTAL CROSS FLOW SCRUBBER
A. Weir, Jr., J.M. Johnson, D.G. Jones,
and S.T. Carlisle, Southern California Edison,
Rosemead, California 357
OPERATIONAL STATUS AND PERFORMANCE OF THE LOUISVILLE
FGD SYSTEM AT THE PADDY'S RUN STATION
R.P. Van Ness, Louisville Gas and Electric
Company, Louisville, Kentucky 389
DISPOSAL OF BY-PRODUCTS FROM NON-REGENERABLE FLUE
GAS DESULFURIZATION SYSTEMS
J. Rossoff, R.C. Rossi, L.J. Bornstein,
The Aerospace Corporation,
El Segundo, California
J.W. Jones, Environmental Protection Agency,
Research Triangle Park, North Carolina 399
AN OVERVIEW OF DOUBLE ALKALI PROCESSES FOR
FLUE GAS DESULFURIZATION
N. Kaplan, Environmental Protection Agency,
Research Triangle Park, North Carolina 445
INITIAL OPERATING EXPERIENCES WITH A DUAL-ALKALI
S02 REMOVAL SYSTEM
PART I - PROCESS PERFORMANCE WITH A COMMERCIAL
DUAL-ALKALI S02 REMOVAL SYSTEM
T.T. Dingo, General Motors Corporation, Parma, Ohio .... 517
PART II - EQUIPMENT PERFORMANCE WITH A COMMERCIAL
DUAL-ALKALI S02 REMOVAL SYSTEM
E.J. Piasecki, General Motors Corporation,
Parma, Ohio 539
EPA-ADL DUAL ALKALI PROGRAM—INTERIM RESULTS
C.R. LaMantia, R.R. Lunt, J.E. Oberholtzer, E.L. Field,
Arthur D. Little, Inc., Cambridge, Massachusetts
N. Kaplan, Environmental Protection Agency,
Research Triangle Park, North Carolina 549
VI
-------
TITLE PAGE
VOLUME II
REGENERABLE PROCESSES SESSION
NEW ENGLAND S0? CONTROL PROJECT FINAL RESULTS
G.R. Koehler, E.J. Dober, Chemical Construction
Corporation, New York, Nexj York 671
ASSESSMENT OF PROTOTYPE OPERATION AND FUTURE EXPANSION
STUDY - MAGNESIA SCRUBBING MYSTIC GENERATING STATION
C.P. Quigley, J.A. Burns, Boston Edison Company,
Boston, Massachusetts 709
MAG-OX SCRUBBING EXPERIENCE AT THE COAL-FIRED DICKERSON
STATION, POTOMAC ELECTRIC POWER COMPANY
D.A. Erdman, Potomac Electric Power Company,
Washington, D.C 729
POWER PLANT FLUE GAS DESULFURIZATION BY THE
WELLMAN-LORD S02 PROCESS
PART I - THE DEAN H. MITCHELL STATION
E.L. Mann, Northern Indiana Public Service Company,
Michigan City, Indiana 739
PART II - CONTINUING PROGRESS FOR THE WELLMAN-LORD
S02 PROCESS
E.E. Bailey, Davy Powergas, Inc.,
Lakeland, Florida 745
THE CAT-OX DEMONSTRATION PROGRAM
E.M. Jamgochian, The Mitre Corporation,
McLean, Virginia
W.E. Miller, Illinois Power Company,
Decatur, Illinois 761
THE SHELL FLUE GAS DESULFURIZATION PROCESS
J.B. Pohlenz, Universal Oil Products Company,
Des Plaines, Illinois 807
STATUS REPORT ON CH1YODA THOROUGHBRED 101 PROCESS
M. Noguchi, Chiyoda International Corporation,
Seattle, Washington 837
FLUE GAS DESULFURIZATION BY-PRODUCT
DISPOSAL/UTILIZATION PANEL
FLUE GAS DESULFURIZATION BY-PRODUCT DISPOSAL/UTILIZATION
- REVIEW AND STATUS
H.W. Elder, Tennessee Valley Authority,
Muscle Shoals, Alabama 85.1.
VII
-------
TITLE PAGE
LIME/LIMESTONE SLUDGE DISPOSAL - TRENDS IN THE
UTILITY INDUSTRY
C.N. Ifeadi and H.S. Rosenberg,
Battelle, Columbus, Ohio 863
ENVIRONMENTALLY ACCEPTABLE DISPOSAL OF FLUE GAS
DESULFURIZATION SLUDGES: THE EPA RESEARCH AND
DEVELOPMENT PROGRAM
J.W. Jones, Environmental Protection Agency,
Research Triangle Park, North Carolina 887
FGD SLUDGE FIXATION AND DISPOSAL
W.H. Lord, Dravo Corporation,
Pittsburgh, Pennsylvania 929
UTILIZING AND DISPOSING OF SULFUR PRODUCTS
FROM THE GAS DESULFURIZATION PROCESSES IN JAPAN
J. Ando, Chuo University,
Tokyo, Japan 955
TVA-EPA STUDY OF THE MARKETABILITY OF ABATEMENT
SULFUR PRODUCTS
J.I. Bucy and P.A. Corrigan,
Tennessee Valley Authority,
Muscle Shoals, Alabama 969
THE PRODUCTION AND MARKETING OF SULFURIC ACID FROM
THE MAGNESIUM OXIDE FLUE GAS DESULFURIZATION PROCESS
I.S. Zonis, F. Olmsted, K.A. Hoist, D.M. Cunningham,
Essex Chemical Corporation, Clifton, New Jersey 1003
SECOND GENERATION PROCESSES SESSION
SECOND GENERATION PROCESSES FOR FLUE GAS
DESULFURIZATION - INTRODUCTION AND OVERVIEW
A.V. Slack, SAS Corporation, Sheffield, Alabama 1029
PILOT PLANT TESTING OF THE CITRATE PROCESS FOR S02
EMISSION CONTROL
W.A. McKinney, W.I. Nissen, D.A. Elkins and
J.B, Rosenbaum, Bureau of Mines,
Salt Lake City, Utah 1049
TVA-EPA PILOT-PLANT STUDY OF THE AMMONIA ABSORPTION -
AMMONIUM BISULFATE REGENERATION PROCESS
C.E. Breed, Tennessee Valley Authority,
Muscle Shoals, Alabama
G.A. Hollinden, Tennessee Valley Authority,
Chattanooga, Tennessee 1069
Vlll
-------
TJILE PAGE
DESCRIPTION AND OPERATION OF THE STONE & WEBSTER/IONICS
S02 REMOVAL AND RECOVERY PILOT PLANT AT THE WISCONSIN
ELECTRIC POWER COMPANY VALLEY STATION IN MILWAUKEE
K.A. Meliere and R.J. Gartside, Stone & Webster
Engineering Corporation, Boston, Massachusetts
W.A. McRae and T.F. Seamans, Ionics, Inc.,
Waltham, Massachusetts 1109
CALSOX SYSTEM DEVELOPMENT PROGRAM PRESENTED AT
THE EPA FLUE GAS DESULFURIZATION SYMPOSIUM
R.E. Barnard and R.K. Teague, Monsanto
Enviro-Chem Systems, Inc., St. Louis, Missouri
G.C. Vansickle, Indianapolis Power & Light Company,
Indianapolis, Indiana 1127
WESTVACO ACTIVATED CARBON PROCESS FOR S0x
RECOVERY AS ELEMENTAL SULFUR
F.J. Ball, G.N. Brown, A.J. Repik, S.L. Torrence,
Westvaco Corporation, North Charleston,
South Carolina J ! 51
IX
-------
THE ROLE OF ENVIRONMENTAL HEALTH ASSESSMENT
IN THE CONTROL OF AIR POLLUTION
Dr. John Finklea
Director
National Environmental Research Center
Research Triangle Park
Environmental Protection Agency
Presented at
Symposium on Flue Gas Desulfurization
Sponsored by
Control Systems Laboratory
U.S. Environmental Protection Agency
Atlanta, Georgia
November 4-7, 1974
-------
THE ROLE OF ENVIRONMENTAL HEALTH ASSESSMENT
IN THE CONTROL OF ALR POLLUTION
ABSTRACT*
Existing legislation requires that air pollution be controlled
to protect public health. The health assessments required to evaluate
air quality standards and to support decisions on the control of mobile
and stationary source emissions present a series of complex scientific
challenges.
Assessment of the health-related air quality standards reveals
no reason for abruptly changing these air quality goals. However,
vexing uncertainties in the scientific information base remain. Thus
there are often rather large differences between upper and lower
boundary estimates of a threshold for adverse health effects. In
addition, there is a substantial body of evidence that fine particulate
aerosols like the acid-sulfate aerosols are injurious, but the broad
scientific information base necessary to establish aerosol standards
is lacking.
Assessment of the health effects of increasing sulfur oxide
emissions is likewise difficult because of the uncertainties involved
in our scientific information base. However, present rough estimates
conclude that substantial excess adverse health effects can be
expected each year if clean air requirements are not met: thousands
of premature deaths, millions of days of illness among susceptible
segments of the population, hundreds of thousands of needless acute
lower respiratory illnesses in otherwise healthy children and hundreds
of thousands of chronic respiratory disorders among adults. If the
health impact of short-term fumigations prove greater than expected or
if regional effects occur in the western power regions, the calculations
of excess adverse effects given here may prove overly conservative. It
is important to remember that the present ambient sulfur dioxide
standards would be about what is necessary to protect public health from
effects attributable to acid sulfate aerosols if dispersion conditions
were good and if acid-aerosols did not intrude from up wind sources-
One should also recall that the sulfur oxides criteria document recog-
nized the problem of acid aerosols but that the knowledge at that time
led one to believe that long range transport of aerosols was not a
major constraint for air quality.
*Full text to be published in Advances in Environmental_Science &
Technology, Volume 7, published by Wiley-Interscience, John Wiley &
Sons, Inc., New York, after July 1975.
-------
Assessment of the health impact of equipping light duty motor
vehicles with oxidation catalysts demonstrates that these emission
control devices are at best a mixed blessing. Emissions of a
number of pollutants capable of adversely effecting public health
including carbon monoxide, phenols, aldehydes and polynuclear
aromatics are dramatically reduced by these devices. Likewise,
catalytic converters will reduce other hydrocarbon emissions which
are not known to affect health adversely but which are an important
precursor of photochemical oxidants that do adversely affect health.
Unfortunately, oxidation catalysts will also alter emissions patterns
so that sulfates and sulfuric acid emissions will be greatly increased
and worsen a public health problem whose dimensions are net completely
understood. The degree of public controversy engendered by the health
assessments just described is a function of the scientific uncertain-
ties contained in each particular assessment. Another major deter-
minant of the amount of controversy involved is whether or not the
decision affects a vested industrial, environmental or governmental
interest that is politically and economically potent. Since control
costs usually rise exponentially as one approaches an emission or
ambient standard, uncertainties about exposure response relationships
can result in violent controversy and cause major economic problems.
In general, the amount of scientific information demanded seems
directly related to the degree of public controversy. Despite
attempts to augment the scientific information base for air pollution
control, it is unlikely that scientific uncertainties will be
sufficiently resolved to prevent disruptive disputes among reason-
able persons for another decade.
-------
FLUE GAS DESULFURIZATION AND OTHER ALTERNATIVES
FOR PRODUCING ELECTRICITY FROM COAL
Dr. Stephen Gage
Acting Director
Office of Energy Research
Environmental Protection Agency
Presented at
Symposium on Flue Gas Desulfurization
Sponsored by
Control Systems Laboratory
U.S. Environmental Protection Agency
Atlanta, Georgia
November 4-7, 1974
-------
Since the last symposium on flue gas cleaning, there have
been a number of developments of immense proportions in the energy
area. Some of these developments have significant implications
for this meeting and, more specifically, for the subject of my
presentation.
In June 1973, the President committed the nation to a much
expanded Federal research and development program on energy supply
and use. An interagency study chaired by the Atomic Energy Commission
recommended in December 1973 that $10 billion dollars be spent
over the next five years on energy R&D. The AEC's report, "The
Nation's Energy Future," placed greatest emphasis on coal and
nuclear power. The general thrust of the report was adopted
by the Office of Management and Budget in formulating the FY
1975 energy R&D budget which nearly doubled the FY 1974 level.
To bring improved efficiency to the Federal energy R&D
effort, the Administration proposed in July 1973 that an Energy
Research and Development Administration be formed to provide
centralized management of nuclear, fossil fuel, and other R&D.
The ERDA act xvas passed in October 1974 and soon the Atomic Energy
Commission and some programs of the Department of Interior and
National Science Foundation will be merged in the new organization.
October 1973 brought the Arab oil embargo which was followed
by the gasoline crisis, Project Independence, and the Federal
Energy Administration. The long queues of cars at gasoline stations
and the higher prices posted on the pumps brought home the realization
of how vulnerable we were becoming to international political
blackmail. Project Independence committed us on a course to
reduce that vulnerability by increasing our capability for energy
self-sufficiency. That course, as the "Project Independence
Blueprint" will tell us soon, is a complex one requiring many
different types of actions — demand reduction, increased use
of coal, offshore exploration for oil and gas, etc.
Perspective on Energy Self-Sufficiency
Research and development on new energy resources and new
energy conversion technologies will unfortunately have little
impact for a good many years on our capability for energy self-
sufficiency. The only significant advances toward this national
goal will come from application of technologies and practices
which either exist now or which have been under development for
a number of years.
-------
Near the top of any reasoned agenda for energy policy
actions is energy conservation. We are making energy conservation
work today and, of course, are being helped considerably by the
significant increases in the prices of all forms of energy during
the past year. Not only oil but coal, natural gas and electricity
have increased sharply in price. We are using energy less profligately
and also more efficiently. Smaller cars with fewer power-consuming
options, less wasteful use of energy in heating and cooling of
homes and offices, more efficient use of energy in industrial
processes — all confirm that energy conservation is an alternative
and a practical and necessary one at that.
On the supply side of the equation, we can anticipate increased
use of nuclear power, domestic oil and gas, coal, shale oil, geothermal,
and solar energy. Nuclear power expansion through the middle
1980's, however, is tied to established construction schedules;
there will be at most slight accelerations or delays from those
schedules. Rapid development of additional oil and gas supplies
is probably limited to expanded drilling near existing fields
and gathering systems and to advanced recovery techniques applied
to existing wells. There is undoubtedly much more oil on the
Alaskan North Slope and in new offshore areas but exploration
for and development of those resources will require years. Shale
oil, geothermal and solar will provide some new supplies but
within the coming decade their contribution will be marginal
at best. That leaves one major near-term alternative to imported
oil: combustion of coal in utility and industrial boilers — boilers
which are either operating or are already under construction.
The serious implications of expanded coal consumption for
the nation's air quality brings us back to the subject of this
symposium. In my opinion, there are no alternatives to the use
of both high and low sulfur coal. Both must be used to meet
the serious and continuing fuel supply crisis which faces the
nation. But both must be used in ways which do not seriously
jeopardize human health and welfare. The question then is how
we use the nation's different types of coal in environmentally
acceptable ways..
Although generally of higher heating quality, coal from
the eastern and central part of the nation typically has higher
sulfur content. Coal from the Interior Province of Western Kentucky,
Illinois and Indiana often has a sulfur content of three percent
or greater, thus requiring sulfur oxides control greater than
80 percent to meet New Source Performance Standards and many state
regulations. Appalachian coal varies considerably from metallurgical
-------
quality coal with sulfur content well below one percent to medium
quality coal with sulfur content between two and three percent.
Recent estimates of low sulfur Appalachian coal reserves have, however
indicated much larger reserves than previously thought. The
Geological Survey has estimated that'there are about 100 billion
tons of coal with sulfur content less than one percent in the
Appalachian fields, primarily in southern West Virginia, eastern
Kentucky and western Virginia. Much of this coal will meet NSPS
and if the rest is washed it too can meet the most stringent
emission standards in middle central states.
Because of the increased price of coal following the oil
embargo, I am confident that increasing amounts of this high
quality eastern coal will be produced for use in both utility
and industrial boilers. Today's differential between production
costs and selling price of coal is sufficient to provide for
full reclamation of strip mined areas and adequate health and
safety devices in underground mines. Thus much of the boiler
fuel required in the industrial heartland will be low sulfur
coal from the prime Appalachian fields with acceptable levels
of impacts to the air and land of the region.
Poorer quality Appalachian and nearly all Interior coal
will require sulfur control. This fact, of course, underlies
the raison d'etre of both the Environmental Protection Agency's
flue gas desulfurization R&D program and the Department of Interior's
(soon to be ERDA's) synthetic fuel R&D program.
Further justification for the massive Federal R&D effort
on synthetic fuels arises from the limited degree to which coal
can be substituted for other fuels. To make the point abundantly
clear, use of the vast reserves of Western coal depends to a
considerable extent on the development of technology to convert
that low sulfur Western coal to alternative fuel forms — clean
liquids and gases — and, not incidentally, on the development of
technology to remove the ash and sulfur in the process.
Western coals, typically low in both sulfur and heat content,
are not, I'm afraid, the panacea that some believe. In the next
five years, Western coals will provide important new supplies
but they are not a widely applicable solution to coal supply
and use east of the Mississippi River where most of America's
heavy industry is located. Transportation of the coal from the
Wyoming and Montana coal fields to Chicago and points east is
expensive and consumes valuable and scarce diesel fuel. Adding
$10 to $15 or more per ton to the price of coal for transportation
will in many cases tip the balance toward eastern coals with
sulfur control measures.
-------
There are other problems with using Western coals in existing
Midwestern boilers. Use of western coal in many of these boilers
would result in the derating of the boilers by 10 to 15 percent
because of limits on bulk handling.capabilities of the entire plant.
With up to 30 percent lower heat content and higher ash and moisture
content, western coals would cause a much more serious reduction
of reserve generating margin than stack gas scrubbers. There are
operational problems with western coals, such as preheater fires
due to carbon carryover and overloading of particulate collection
devices. Finally, a significant fraction of the western coals
will not meet the more stringent emission standards and in some
cases the New Source Performance Standards.
Western coals, I believe, will be important but will only
begin making significant contributions when they can be used as
feedstock for coal gasification and liquefaction process. Now
let me turn to those processes.
Alternative Technologies To Stack Gas Scrubbing
In the paper which I delivered before the last Flue Gas Desulfur-
ization Symposium (May 1973), I discussed in detail the precombustion
cleaning processes which might be considered as alternatives to
flue gas desulfurization. I also mentioned the potential of fluidized
bed combustion as an alternative but, because of the low level
of R&D activity in this area at that time, I did not expand on
fluidized bed systems.
The political and economic climate influencing the development
of alternative technologies has changed drastically in the past
year and a half. In the FY 1975 budget, $350 million were made
available to the Office of Coal Research and Bureau of Mines for
coal utilization R&D. The two technologies which appear to be the
most direct competitors of flue gas desulfurization — low-Btu
gasification and fluidized bed combustion — received the greatest
proportional increases over FY 1974. Low-Btu gasification R&D received
$52 million and fluidized bed combustion R&D received $37 million
in the FY 1975 budget. These funding levels are comparable to
the $35 million allocated to the Environmental Protection Agency
for stack gas cleaning R&D in FY 1975.
-------
Research and development, however, must proceed apace.
Adequate resources are a prerequisite for orderly progress in an
R&D program, however, progress depends on many other factors.
The advance of a project through the R&D stream cannot out-run
ideas, manpower, and construction schedules. Thus, while it
appears likely that low-Btu gasification and fluidized bed combustion
will receive adequate support, actual as opposed to paper progress
in those alternatives has been slight. Further, there have been
few technical developments in the R&D on liquefaction and high-
Btu gasification not fully anticipated a year and a half ago,
even though those two processes have also received marked funding
increases.
Before discussing low-Btu gasification and fluidized bed
combustion further, let me put liquefaction and high-Btu gasification
in perspective relative to electric power generation. There
appears to be an emerging consensus that both liquefaction and
high-Btu gasification will produce fuels that will be too expensive
for use under utility boilers. Production costs for liquid coal
products range from $1.10 to $2.50 per million Btu while costs
for high-Btu synthetic gas range from $1.20 to $1.80 per million
Btu. When these synthetic fuels become available in any quantity —•
and that will be in the 1980's — I believe that they will be
bid away from the utility market by industrial, commercial, and
residential markets where clean liquid and gaseous fuels will
seek premium uses.
From the point of view of resource conservation, it also
makes more sense to utilize the clean fuels derived from these
two processes in high efficiency end uses such as residential
heating. Coal liquefaction, it is estimated, will be 60 to 70
percent thermally efficient while high-Btu gasification will
be 55 to 65 percent efficient. It would be illogical to waste
60 percent or more of the remaining heat value in the cleaned
fuel by converting it to electricity.
Of the various clean fuel alternatives, these two are the
most advanced. The hydrogasification pilot plant being tested
by the Institute for Gas Technology in Chicago has had several
long successful runs in the past few months. The carbon dioxide
acceptor pilot plant operated by Consolidated Coal Company in
Rapid City, South Dakota has also had numerous successful runs,
exceeding the design specifications of the gasifier. The synthane
and BIGAS pilot plants are under construction in Bruceton and
Homer City, Pennsylvania, respectively. The most advanced coal
liquefaction process — Pittsburgh and Midway's Solvent Refined
Coal process — will soon be tested in a pilot plant near Fort
Lewis, Washington.
10
-------
R&D efforts for both low-Btu gasification and fluidized bed
combustion lag those for high-Btu gasification and liquefaction.
The development of a 5-year fossil fuel R&D program by the Interior
Department's Office of Research and Development during summer 1974
helped to clarify some of the goals and program thrusts for these
two technologies but many questions remain. The program plan calls
for a low-Btu gasification demonstration plant in 1978 and fluidized
bed combustion demonstration in 1979 — goals which I submit are
probably impossible to meet, given the state of the technology
and the managerial and technical expertise supporting the efforts.
One of the toughest questions to be answered in both R&D
programs is which way(s) to go. With the realization that sulfur
control measures will be required for electrical power plants and
that these may be alternatives to stack gas cleaning, there have
been a plethora of proposals on candidate processes. So much so
that in the area of low-Btu gasification the number of candidates
is overwhelming and the adequacy of the screening process — that
is, selection criteria based on technical knowledge — is frankly
underwhelming. Hopefully, these organizational problems will soon
be overcome so that both programs can move ahead. One good sign
is the development of a National Fluidized Bed Combustion R&D Plan
designed to define the program goals, translate these into research
plans, and spell out the roles of the various agencies. EPA is
participating in the development of the plan because of the agency's
important contributions to the early development of fluidized bed
boilers and because of the agency's continuing role in the environmental
assessment of these combustors.
The advantages of low-Btu gasification and fluidized bed
combustion have been widely touted in the past year. Both techniques
have the potential for improving the thermal efficiency of electrical
power production. Used in a combined cycle configuration, i.e.,
with a gas turbine topping cycle, the overall plant efficiency
may go as high as 45 percent. But such high efficiencies will
be achievable only in second or third generation plants. Early
low-Btu gasification efficiency and early fluidized bed boilers
will probably not achieve over 36 to 37 percent. And those efficiencies
assume success in the single most critical R&D element — hot gas
cleanup. Thus it is likely that early low-Btu gasification and
fluidized bed combustion plants will use coal with approximately
the same efficiency as a modern steam boiler with a low pressure
drop stack gas scrubber.
-------
Another obvious point which must be made is that since the
low-Btu gas plant, the fluidized bed boiler, and the conventional
boiler with stack gas scrubber are all located at the central electrical
generating station they are all faced with the common problem of
sulfur. Assuming that technologies will develop to the point where
the sulfur can be removed from the coal, the nagging problem of
sulfur by-product disposal remains. If dolomite, limestone, or
lime is used to remove the sulfur, then landfill techniques or
expensive and unproven regeneration techniques must be used to
handle the by-products. If, from the reducing environment of the
gasifier, the sulfur can be removed as hydrogen sulfide it can
then be converted to elemental sulfur in a Glaus process. Advanced
flue gas desulfurization processes, to which I'll return later,
also have the potential for producing elemental sulfur. The problem
of disposing of the elemental sulfur may not be trivial, however.
Predictions of low production costs for both low-Btu gasification
and fluidized bed combustion have also been touted by their proponents
as justification for rapid development and deployment. Estimates
of low-Btu power gas production costs have gone up in the past
year, reflecting more the rising estimates for high-Btu gasifiers
than detailed economic analyses of low-Btu gasifier designs. Estimates
of incremental costs for power gas production range from 50 to
80 cents per million Btu based on a 90% plant factor. Although
proponents of fluidized bed combustion claim that the process may
in fact be less expensive than conventional pulverized coal-burning
combustion, such cost savings could not be approached for the first
several generations of combustors, that is, until late in the 1980's.
One component of FBC costs which I believe has been seriously underestimated
is the waste disposal or regeneration of the reacted dolomite.
Those of you who have been working on stack-gas scrubber waste
disposal should certainly appreciate that fact.
Finally, I should make a categorical statement about the trend
of estimated costs of almost any new technology as the technology
moves from the bench scale to the pilot plant to the demonstration
plant — the estimated costs invariably go up and they go up more
rapidly as the demonstration phase is approached. There is nothing
mysterious about this — it's human nature. In the early days
of nuclear power the wiser and experienced engineers always multiplied
the cost estimates of new reactor concepts by a factor of three.
The more mathematically oriented French engineers, apochrapha has
it, always multiplied by IT.
12
-------
Certainly low-Btu gasification and fluidized bed combustion
are important and will undoubtedly provide important sources
of clean energy in the 1980's. The Environmental Protection
Agency will be encouraging substantial R&D effort on these two
technologies and will be providing back-up engineering and environmental
assessments to ERDA and private efforts. But if we're serious
about Project Independence — and we must be for the sake of our
economic security and our independence in foreign relations —
then we must turn to increased use of all of our coals as soon
as possible. And to protect public health and welfare, this means
using scrubbers on many boilers burning high sulfur coal.
ORD's Stack Gas Scrubbing R&D Program
After arriving in Atlanta last evening, I took a cab downtown
in search of a restaurant. Just a few blocks away from the hotel,
I noticed a crowd of people entering a church. The sign in front
of the church indicated that Mozart's "Requiem in D Minor" was
being performed inside. Well, after the recent on slaught of anti-
scrubber advertising by American Electric Power, there is little
that surprises me but this frankly did. I do find it hard to
believe that Donald Cook would have commissioned the performance
of Mozart's Requiem in that nearby church just to throw a pallor
over this conference. The utter ingenuity of the man!
As I'm sure you're aware, the Environmental Protection Agency
views scrubbers much differently from Donald Cook and American
Electric Power. And I think you agree or you wouldn't be here
at this meeting. Far from being the requiem mass for scrubbers,
this gathering is much more a celebration of life. I offer my
voice to the choir. Stack gas cleaning of high sulfur coals offers
the greatest single contribution, consistent with environmental
protection, to the goal of energy self-sufficiency. I feel that
if Donald Cook really believed in Project Independence and the
need to protect human health from sulfur oxides and metallic acid
sulfates he would support the use of scrubbers with high sulfur
coal in those parts of the AEP service area where air quality
is worse than ambient primary standards.
Problems remain with scrubbers. There has never been a
technological innovation which didn't require a shakedown phase.
Nuclear power —the future mainstay of the electric industry —
is still confronting and overcoming problems, 14 to 15 years after
13
-------
the first civilian demonstration power reactors were started up.
We must continue to improve reliability of the non-regenerable
(lime/limestone) scrubbers. We must continue to develop improved
sludge disposal methods and to evaluate their environmental impacts.
We must continue to reduce the capital and operational costs of
flue gas desulfurization systems. Finally we must develop regenerable
systems which are capable of producing elemental sulfur as a by-
product at a reasonable cost.
I'm sure many of you are aware that EPA's Office of Research
and Development has come under serious criticism during the past
few months. A variety of problems have been pointed out by a
National Academy of Sciences panel, the Senate Public Works Committee
and others. ORD has already begun to take measures to correct
its shortcomings and I'm sure many additional actions will be
taken once a new Assistant Administrator is named and confirmed.
The major action taken to date is the establishment of a new program
management office — reporting directly to the Assistant Administrator —
which would have responsibility for all of EPA's R&D relating
to energy. The Office of Energy Research, of which I'm now serving
as Acting Director, is faced with the challenge of meshing EPA's
research on health and ecological effects of pollutants from energy
plants with EPA's R&D on control technology for those same plants —
an imposing challenge but one which I feel the agency must confront
head-on.
One of the features of the partial ORD reorganization is
that the Office of Energy Research would concentrate on strategic
planning and management, abrogating to the laboratories increased
responsibility for detailed project management. This is directly
responsive to many of the criticisms. Such a division of responsibility,
however, can work only if both parties fully do their jobs. My
job begins with setting strategic energy-related goals for EPA's
laboratories and I want to begin that job here today with setting
goals for the agency's R&D on flue gas desulfurization.
- First, ORD will conduct an advanced lime and limestone
test program on our flexible test facility at TVA's
Shawnee plant.
This program calls for further-testing and evaluation of the recently
demonstrated reliable operating modes and of the "unsaturated
gypsum" mode recently discovered in EPA labs. The testing program
will also focus on four additional factors: (1) achieving maximum
S02 removal efficiency consistent with reasonable operating costs,
(2) maximizing alkali utilization thus minimizing sludge production,
(3) identifying mist eliminator and wash systems with more reliable
operation, and (4) evaluating sludge disposal methods, including
those offered commercially by Dravo, I.U.C.S., and Chemifix.
14
-------
Second, to provide an alternative to lime/limestone throw-
away systems, ORD will initiate a full-scale demonstration of
a Double Alkali FGD System.
This demonstration will be conducted on a 100-200 MWe coal-fired
utility boiler burning high sulfur coal. Selection of the vendor
and utility will be done on a competitive basis. ORD will provide
up to half of capital and operating costs incurred during the
test program.
Third, to provide an alternative to sulfur-producing
synthetic fuel processes, ORD will initiate a full-scale
demonstration of a regenerable FGD system which produces
by-product sulfur.
This demonstration will also be conducted on 100-200 MWe coal-
fired utility boiler burning high sulfur coal. Selection of both
the process and the vendor/utility will be done on a competitive
basis. Again EPA will cost-share up to half of both capital and
operating costs.
Those three goals define the Office of Energy Research's
agenda for stack gas scrubbing. We will be doing our best to
ensure that both financial and managerial support required to
achieve these objectives is provided. We expect ORD's Control
Systems Laboratory and other supporting laboratories to move vigorously
to meet the challenge. And we invite you to help us meet our
obligation to the American public to provide the means for using
our domestic coals in environmentally acceptable ways. Now is
the time to do it. We can't keep putting off tomorrow. All too
quickly tomorrow becomes yesterday.
15
-------
STATUS OF FLUE GAS
DESULFURIZATION SYSTEMS
IN THE UNITED STATES
Prepared by
Timothy w. Devitt and Fouad K. Zada
PEDCo-Environmental Specialists, Inc.
Suite 13, Atkinson Square
Cincinnati, Ohio 45246
Prepared for Presentation at
Flue Gas Desulfurization Symposium
Sponsored by the Control Systems Laboratory
U.S. Environmental Protection Agency
Atlanta, Georgia
November 4-7, 1974
17
-------
ABSTRACT
Data are presented on the number and total MW of flue
gas desulfurization (FGD) systems installed, under construc-
tion and planned for U.S. utility application. The reli-
ability of operating systems is briefly described. Case
histories describe FGD operations at four plants: Hawthorn
Power Station (Kansas City Power and Light), Will County
Power Station (Commonwealth Edison), Lawrence Power Station
(KCPL), and Reid Gardner Power Station (Nevada Power Company)
Description of the FGD process at each installation is
followed by an account of system performance, with emphasis
on problems encountered, solutions developed, and plans for
future operations.
CONTENTS „
Page
Status of Flue Gas Desulfurization Technology
in the United States 19
The Lawrence Power Station 30
The Will County Power Station 38
Hawthorn Power Station 48
Reid Gardner Power Station 57
Information Sources 62
Appendix A--Flue Gas Desulfurization System
Status Report, October 1974 63
18
-------
STATUS OF FLUE GAS DESULFURIZATION
TECHNOLOGY IN THE UNITED STATES
In March 1974, PEDCo-Environmental Specialists was
awarded a study by the U.S. Environmental Protection Agency
to evaluate the status of flue gas desulfurization (FGD)
technology in the United States. This study will have two
outputs. The first will be a series of reports .on indi-
vidual FGD facilities. These reports will present values
for pertinent process design and operating parameters and
describe the operability of the unit from start-up to the
time of the plant survey. To date we have visited nine
installations; we are scheduled to visit another five
facilities by February 1975.
The second study output is a series of monthly reports
over the one year study period. These reports include a
summary of the operability (reliability) for each utility-
size FGD system over the preceding month and a complete list
of all FGD units which are either under construction or
planned. A copy of the October monthly report is attached
as Appendix A.
As shown in Table 1, there are 19 operating FGD systems
on utility-size boilers, 17 units under construction and 62
units planned. The planned units include those for which
there has been a contract awarded, a letter of intent signed
or the utility is presently requesting or evaluating bids.
Of the operating systems, 13 are lime or limestone based
(including limestone injection), two are magnesium oxide
scrubbing, two are sodium carbonate, one is double alkali
and one is catalytic oxidation.
19
-------
Table 1. NUMBER AND TOTAL MW OF FGD SYSTEMS
Status
Operational
Under construction
Planned
Contract awarded
Letter of intent
Requesting/evaluating bids
Considering only FGD systems
Total
No. of
Units
19
17
12
4
9
37
98
MW
3,291
6,777
6,640
905
3,751
16,472
37,836
Figure 1 illustrates the projected increase in the
number of megawatts of power production equipped with FGD
systems. By 1975, approximately 7000 MW of capacity will be
installed of which 50 percent will be retrofit applications
and 50 percent will be on new power plants. By 1980, ap-
proximately 35,000 MW of capacity will be installed of which
30 and 70 percent will be retrofit and new installations
respectively.
Figure 2 illustrates the magawatts of capacity that
will be controlled by the FGD process. In 1974, approxi-
mately 80 percent is controlled by lime and limestone
scrubbing processes; in 1977, over 90 percent will be lime
or limestone scrubbing.
Sulfur dioxide removal efficiencies for the operating
units range from approximately 70 to 90 percent and partic-
ulate removal efficiencies generally are above 99 percent
for those units designed for particulate removal. Instal-
lation size varies from about 30 MW to over 800 MW. FGD
systems have been applied to both low (0.4 to 1.0 percent)
and high (6.0 percent) sulfur coals. The non-regenerable
systems have used a variety of methods for sludge disposal
20
-------
38_
36
34
32
LEGEND
30
28
26
24
22
20
18
16
NEwl-
RETROFIIi-
TOTAL
T
o
o
o
Q_
-------
16,000
14,000
12,000
LEGEND
(1) LIMESTONE SCRUBBING
(2) LIMF. SCRUBBING
(3) LIME/LIMESTONE SCRUBBING
(4) OTHERS (NOT SPECIFIED)
(5) WELLMAN LORD
(6) LIMESTONE INJECTION
(7) CATALYTIC OXIDATION
(8) SODIUM CARBONATE
(9) MAGNESIUM OXIDE
(10) DOUBLE ALKALI
(11) LIME INJECTION
f
10,000
8,000
6,000
4,000
2,000
1,000
/
68
70
72
76 78
YEAR
80
Figure 2. Present and projected capacities of FGD processes.
-------
including ponding, dewatering by vacuum filtration and land-
filling, and fixation followed by landfilling. The regen-
erable systems have been limited to date to recovering sul-
furic acid although a sulfur recovery unit is being installed.
The total installed capital costs of these systems have
varied from about $20/KW to over $100/KW. This cost picture
is clouded, however, because of a variety of factors which
influenced final installed costs. In some cases, the system
designer shared the installation cost and much of the system
modification cost with the utility; in other cases there was
very little cost sharing. The total installed cost for FGD
systems over about 200 MW in capacity is estimated to be in
the range of $60 to $80/KW.
Pertinent data on the operating systems are presented
in Table 2.
The reliability and availability of FGD systems has
been a major question. During our plant inspections, it was
apparent that most plants had overcome the "chemical"
problems; most operators felt they could control the process
chemistry and hence minimize or eliminate scaling and
related problems. However many units are still faced with a
number of mechanical problems (e.g., frozen venturi throat
drives, fan vibration). While most of these problems are
not major in that they occur on an infrequent basis, they
do lead to lower unit availability.
Availability data of any sort and particularly on a
common basis are scarce. Some limited data for Paddy's Run
and Will County are shown in Figure 3. The availability
factor was calculated as hours of FGD system operation
divided by hours of boiler operation. Such a definition,
however, can be misleading. For example, on a peaking unit
like Paddy's Run, the FGD system may be bypassed for short
boiler runs even though the system was available for use;
thus the availability factor would be lower than the actual
-------
Table 2. DESIGN AND OPERATING DATA FOR OPERATIONAL FGD SYSTEMS
FCO Proceaa/
Power Station
Ar 1 tor a Public Service
Choi la no. 1
City of My weat
Key "ett Po^er Plant
Connon wealth Mlaoa
will County He. 1
La Cygno *'o. 1
Line ScribMnq
Duqueine Mfht
Phillips
LouiavllU Cat t
Electric
Paddy! (tan Ko. f
Southern California
UUon
Hohave #o. 2
LiMatone Inlectioo
Sweat Cily Power t
Light
•ewUiocn Ho. 4
xeneae City Pcwer t
Light
Hawthorn No. 3
Klfiiai Pe>u*r 1 Light
Lawrence No. S
Keneas Power i tiaht
Lawrence «e. 4
Liaa In'tctlon
CttiryUfld power - Coop
AIM Station
yj*e/t,l«*tcne
Tenne**ee Valley
Authority
Shawnee Ho. 19
HoO ScrubbijMj
Voaton Cdlaon
My a ti
It
P.
R
P.
n
»
•
ii
*
*
K
It
R
>
Siie of
rco unit
KM
115
33
1(7
120
410
IS
110
100
1*0
400
" 12$
to
30
ISO
100
123
125
32
•111
Statue
(Start-
up Datel
10/73
10/72
2/72
t/73
10/73
«/«
11/73
1/72
11/72
11/73
12/tl
•/T*
4/72
4/12
»/7J
11/1J
11/11
1/14
If/Tl
Typ.
Co*l
Oil
Coal
Co*l
CM!
Co.1
Co.l
Cokl
COAl
Coal
Co.1
CMl
Co«l
CMt
Co.1
Cool
Cxi
CMl
ru«t
t Ach
&-1S
10.0
K-]»
11
14
14
14
11. S
11.9
12- JO
11-1!
10
»
KK
10
% S
0.4-
1.0
1.7S
2.0
(.0
2.1
3.7
O.J-
O.f
).TS
I.7S
>.4
1.4
1.0-
l.J
1.0
1.0
O.I
M
l.i
•rtlcul»t«
Control
Koch
Ho
Ho
T..
*>
Ho
Ho
Ho
No
Ho
T.I
No
»..
I«.
*••
V..
y*i
tit
Y.»
Mo
Y.I
rr.
Y..
Ho
Ho
Ho
NO
HO
No
Y..
HO
HO
No
Y«.
rlu* c..
VolUM
ACPH
770.000
2.100.000
Stl.OOO
20G.COO
2.100.000
44S.OOO
44S.OOO
990.000
110,000
.
10.000
211.000
471,000
471,000
X*
1.111.000
r<»p
•r
1SS
120
1(2
Me
HA
210
100
100
100
110
»0
2S»
ISO
1»
NA
110
llqul
(Vanturl)
It
11
KA
10
01
21-24
21-24
12
22
-
f 4
HA
1 10
(V«nt«ril
1
t (TOV.D
NA
M»l;e up
Ib j«ola S0j
730
ISO
NA
MA
170
170
n
•A
M
420
-
ftn
M
»0
190
KA
F.aocnt
ttan
%
99
71
HA
•A
100
HA
M
NA
"
<2
MA
(0
100
100
100
100
NA
•
rcc »>ete»
H( (icicftcy
lUrnuval
t
99
to
»S
9)*
9f.3
99
"
99
99
99
99
-
O*
39.1
99
99
»
>t
503
*
90
IS
10
10
10
10
~
70
70
75
73
70
90-95
90
10
„
IS
HA
11
Caa
(InilUectt
fittaa*
(indirect!
Stran
Cflditect
plui hot
air}
Oil
.Burnera
Caa Pired
Icnlter
Direct
Hot
Air
Stean
I Indirect!
ItflUI
{Indirect)
Not Water
Hot Hater
Hot
Applicable
etaa*
(Indirect)
(Direct
Hot
A,c
NA
M
Sludge
H.U104
OB
Fond
Per-a
FonO
Pond
VACUUM Pll-
t.r A ntul
pimd
Fond
Fond
Fond
Food
Fond
Fond
Acid Fl*nt
Acid Fltnt
Fond
oc
Dry L*k«
Fonl
Act« FlMit
rco
Coat.
14 /*»
m.l/ra
Ml/m
170-7.9/01
IS7/XV
U9.o/n>
119.0/DI
N»
NA
NA
n
iio/i.
«.
I7S/W
071/«»
-------
K)
Oi
LEGEND
(1) WILL COUNTY MODULE A
"2 WILL COUNTY MODULE B
3 PADDY'S RUN MODULE A
4 PADDY'S RUN MODULE B
a) UNIT SHUT DOWN, DEMISTER & REHEATER
PROBLEMS
b) MODIFICATIONS, & LIMESTONE BLINDING PROBLEMS
c) DEMISTER, REHEATING PROBLEMS, FREEZE UP OF
VENTURI THROAT AND LOSS OF BUILDING HEAT
d) DUE TO LOW DEMAND FOR ELECTRICITY, BOILER RUNS
WERE OF TOO SHORT DURATION TO JUSTIFY OPERATION
OF THE FGD MODULES
100
80
>H
E-»
H
^
H 6C
a
j
H
<
>
< 40
M
2
W
u
K 9ft
w zc
PU
\
\
\
S'
( 9 \
\ *_
\
\
\
\
\l
\
V|
/
a /
S~\
JAN. MAR MAY AUG. NOV.
U_ 107? J
-4
^^^
/
.X /
\ '
\ /
A I /
FEB. MAY A
-* 1 Q7
*
[
/
/
/
"-
^V / •! '
V O }
^
^^_r_,
^^^
X' (4)
/
f
.
X
x
\
\
s
^ — — — '
/
t
1
16 /
/ /
/
/
/
/
/
s
/
/~"
/^
^(D
JG. NOV. ,FEB. MAY AUG. .
3 »4* 1 Q7A M
Figure 3. FGD system availability
-------
availability of the unit. On the other hand, units with no
bypass capability would have a high availability factor
since the boiler would be down when the FGD system was in-
operative. Since complete data were not available and could
be misleading if not adequately qualified, we have presented
a qualitative description of recent performance for all the
operating systems in Table 3.
The FGD systems at the Lawrence, Will County, Hawthorn
and Reid Gardner installations are described in the fol-
lowing sections. These systems were selected since they are
not covered in other papers presented at this symposium.
26
-------
Table 3. OPERATING SUMMARY FOR FGD SYSTEMS
Utility and Unit Name
Status
Arizona Public Service
Cholla No. 1
Boston Edison
Mystic No. 6
Commonwealth Edison
Will county No. 1
Dairy Power Co-Op
Alma Station
Duquesne Light
Phillips Station
General Motors
Parraa Plant
Illinois Power
Wood River
Since start-up in October 1973, the FGD system availability has been consistently
above 90 percent. There have been a few mechanical problems, the most persistent
being vibration in the reheater section.
Scrubber availability reported for 1973 ranged from 73 percent in August to 13 per-
cent in December. The decreasing availability was due to deterioration and subsequent
overhaul of process equipment. Recent availability figures are: March 87 percent,
April 81 percent, May 57 percent and June 80 percent. This system is down with no
immediate plans for restarting since the EPA demonstration program is completed.
Boston Edison is evaluating the data collected during the demonstration period to
determine the course of future action.
Availability of Module A has increased in the past several months. It was 72-percent
in April, 93 percent in May, 54 percent in June, 95 percent in July, 91 percent in
August and 85 percent in September. Module B was removed from service until all
necessary modifications are made to Module A to permit reliable operation.
This is a demonstration project where lime mixed with water is injected into the
boiler. The longest test run was less than two days.
At present, only flue gas from Boilers 2, 3 and 4 (about 40 percent of the station
capacity) is treated because fly ash from the inefficient precipitators overloads
the clarifiers. Operating hours for Modules 1 through 4 between 3/17/74 and 6/30/74
were 1756, 762, 815, 1707 respectively. The plant tries to run Modules 1 and 4
continuously. The outage hours for these units is primarily for inspections.
System availability has been essentially 100 percent sinde 4/74. However the
availability figure has little meaning since only two of the four modules were
operating at any time because of low demand.
The unit operated only 700 hours during the past two years. Primary reason for the
downtime was the delay in converting the reheaters from natural gas to fuel oil
firing. The unit is scheduled to restart again in late November.
-------
Table 3 (continued). OPERATING SUMMARY FOR FGD SYSTEMS
Utility and Unit Name
Status
oo
Kansas City Power and Light
Hawthorn No. 3
Kansas City Power and Light
Hawthorn No. 4
Kansas City Power and Light
La Cygne
Kansas Power and Light
Lawrence No. 4
Kansas Power and Light
Lawrence No. 5
Louisville Gas and Electric
Paddy's Run
Southern California Edison
Mohave No. 2
The FGD system has undergone several major modifications. Availability has since
increased from 30 percent in 1973 to as high as 70 percent recently. Plant
operators were on strike between July and October 1, 1974, and during that time
the FGD system was down. The system was restarted on October 1st.
This unit was converted from furnace injection to tail-end scrubbing. The plant
has experienced more problems with this unit (both mechanical and chemical in
nature) than with Unit 3. The operation can be characterized as "fair."
Initial problems included fan deposits, demister and nozzle pluggage, reheater
failure and screen pluggage. Many of the original problems were attributed to
poor pH control. Recently the unit has had about 80 percent availability and
there are no outstanding problems. However it is necessary to remove each module
from service once a week to clean out accumulated solids.
The system has undergone a number of process modifications. At present the system
is capable of sustained operation although SO, removal efficiency is limited to
only 75 percent and the demisters must be washed daily (automatic washing). Manual
washing of the demisters is required every two weeks. The unit is badly corroded
and will be replaced by a new electrostatic prccipitator and FGD system in 1977.
This unit has encountered many of the same process problems as Unit 4. In addition
there is poor gas flow distribution between the eight modules and within each
module; measures are being taken to correct the distribution. The boiler burns
natural gas whenever it is available so it is difficult to assess FGD system avail-
ability.
FGD system availability is near 100 percent. However since the system is installed
on a peaking boiler there are many occasions when the boiler's run is too short to
justify start-up of the FGD system.
This is an experimental unit.
ability above 80 percent.
It has operated quite successfully with an avail-
-------
Table 3 (continued). OPERATING SUMMARY FOR FGD SYSTEMS
Utility and Unit Name
Status
Nevada Power
Reid Gardner No. 1
Nevada Power
Reid Gardner No. 2
Potomac Electric and Power
Dickerson No. 3
The FGD system started-up in March -1974 and has had an availability of over 90 per-
cent when there was sufficient trona (impure form of sodium carbonate). The lack
of trona has limited the operation of this unit. The system has operated only 900
hours since April 1974.
The FGD system started-up in March 1974 and has had an availability of over 90 per-
cent when there was sufficient trona (impure form of sodium carbonate). The lack
of trona has limited the operation of the unit. The system has operated only 900
hours since April 1974.
There is no record of FGD system's availability for the period prior to August 1974
since the unit was down frequently for process modification, equipment repairs, etc.
Availability since August is about 34 percent.
N>
vo
-------
THE LAWRENCE POWER STATION
Power Generating Facilities
The Lawrence Power Station of Kansas Power and Light Company
is located in a lightly industrialized area on the outskirts
of Lawrence, Kansas.
The plant operates two steam boilers which are equipped
to burn coal, natural gas supplemented with oil, or a combination
of these three fuels. Boiler 4 is the oldest of the two units.
It was first placed in service in 1959 and operated as a cyclic-
load boiler. The maximum electric-generating capacity of this
unit varies with the type of fuel being burned: with natural
gas the output can be as high as 143 MW; with coal plus natural
gas output decreases to 125 MW. Retrofitting of this boiler
with an FGD system in 1968 introduced additional pressure drop
in the flue gas system and further reduced the boiler capacity
to 115 MW.
The second unit at the plant is Boiler 5, with a rated
capacity of 400 MW when burning coal plus natural gas. The
unit together with the FGD system, was placed in service in
1971. Like Boiler 4, it is classified as a cyclic-load
unit.
Both boilers at the Lawrence Power Station were built by
Combustion Engineering, which also designed and installed the
FGD systems on these boilers. These FGD systems consist of
limestone furnace injection with flue gas wet scrubbing.
The present grades of coal burned at the Lawrence Power
Station has a gross heat content of 12,000 BTU/lb. Average ash
30
-------
and sulfur contents are 12 percent and 3.75 percent, respectively,
The company is in the process of switching from this high-sulfur
Kansas coal to Wyoming coal which contains 0.4 to 0.8 percent
sulfur and 10 percent ash, with a gross heating value of 10,000
BTU/lb. This change was necessitated by the curtailment of
strip mining operations at the Kansas coal supply site.
FLUE GAS DESULFURIZATION SYSTEM
FGD Process Description. The flue gas desulfurization systems
on Boilers 4 and 5 are identical in basic design and operation.
The FGD system on Boiler 4 has undergone several major modifica-
tions since its start-up in November 1968, many of which were
later incorporated into the design of the FGD system on Boiler 5.
The FGD system associated with each boiler incorporates
facilities for storing, pulverizing, and injection of finely
ground limestone rock into the boiler's furnace chamber, where
the bulk of it is calcined. The calcined limestone, along with
the fly ash, is transported by the flue gas to the tail-end
wet scrubber modules, where the S02 in the gas reacts with the
scrubbed'lime/limestone in a recirculated slurry and is substan-
tially removed, along with the fly ash, from the gas stream.
The cleaned gas is then demisted and reheated (to prevent conden-
sation in the downstream equipment) and finally discharged from
the stack by induced-draft fans.
There are two FGD modules on Boiler 4 and eight FGD modules
on Boiler 5. The 10 modules are identical in size, and each
is designed to handle roughly 150,000 scfm of flue gas. A typi-
cal module is shown in Figure 4. A single stage of 3/4-inch-
diameter glass marbles is contained in a bed about 3 to 4 inches
thick, fitted with overflow pots to collect and drain the liquor
from the top of the bed. The scrubbing liquor is sprayed through
nozzles located below the bed.
Each of the two modules on Boiler 4 is connected (through
an I.D. fan) to a separate 120-foot stack; gases from all eight
31
-------
WATER
SEAL
FROM AIR HEATER
TO STACK
ID FAN
HOT
H?0
REHEAT
DEMISTER
MARBLE B
• SOOT BLOWER AIR
WATER WASH LANCE
(2)
Wj'^^y^Vr^.i'^fy^^rA
fe^v'^^?*^':^::?^1
."•": '"'•'' ••••'•' :":v —-:v::^
OVERFLOW
RECYCLE TANK (1)
(ENLARGED)
DRAIN TANK
LAWRENCE NO. 4
CE-APCS
OCTOBER 1972
SCRUBBERS (2)
(ENLARGED DEPTH - 4')
CLARIFIED FROM POND
TO SOLID
DISPOSAL
POND
Figure 4. Sketch of a typical FGD module
at the Lawrence Power Station.
Source: Kansas Power and Light Company
32
-------
modules on Boiler 5 are discharged through a common stack, 375
feet tall.
Originally, all modules were fitted with bypass ducts
and hydraulic seal dampers. Because of extensive corrosion
and pluggage problems with the two modules of Boiler 4, the
bypass ducts on these modules were removed.
Spent liquor from the scrubber tower drains into a recycle
tank, which allows 30 to 40 minutes retention time to ensure
complete conversion of the scrubbed S07 to calcium sulfite and
calcium sulfate. Spent liquor from this tank overflows to a
drain tank, from which it is pumped to sludge disposal ponds.
There are three sludge ponds on site, covering areas of
4 acres, 16 acres, and 28 acres. The sludge first enters the
16-acre pond and overflows into the 4-and 28-acre ponds. Approx-
imately 800 gpm of sludge containing 9 percent solids, is fed
to the unlined ponds. Because the sludge contains unreacted
lime as well as fly ash (ingredients which are usually added
to stabilize limestone sludge), it sets up very hard, like
concrete, without additives.
FGD System Performance. Several major modifications completed
by Combustion Engineering and Kansas Power and Light Company
on Unit 4 appear to have significantly improved the performance
and availability of that unit. Both FGD modules on Boiler 4
have recently performed reasonably well. Availabilities close
to 100 percent were reported for July and August 1974. The
S0~ removal efficiency has been around 65 percent, which is
sufficient for the plant to comply with the applicable pollution
control regulation. S0? removal efficiencies as high as 85
percent were achieved over a short period, but only at the
expense of an accelerated rate of scale formation in the scrubbers
Because recent operation of Boiler 5 has been on natural
gas, the FGD modules have not been subjected lately to severe
operating conditions. When the boiler is coal-fired and the
33
-------
FGD system is in operation, problems experienced are similar
to those encountered with the modules on Boiler 4. Currently
the main problem with Unit 5 is improper distribution of flue
gas to the eight modules. Combustion Engineering is performing
tests on Unit 5 to alleviate this problem.
Analysis of the problems encountered during and since start-
up reveals that nearly all were due to improper control of process
chemistry. In limestone furnace injection, it is difficult
to achieve satisfactory control of the degree of limestone cal-
cination as well as the amount of lime/limestone carried in
the flue gas to the tail-end scrubbers. This situation is further
aggravated when the boiler is operating as a cyclic-load boiler
and is fired with variable combinations of coal, natural gas,
and oil.
As mentioned earlier, the FGD system on Boiler 4 underwent
several major modifications which are briefly described in this
section.
When the FGD system initially started operation in 1968,
the configuration of each of the two modules was as shown in
Figure 5. This design presented many operating problems, which
included (1) scale buildup and plugging of the hot gas inlet
duct, (2) erosion of the scrubber walls and corrosion of the
scrubber internals, (3) plugging and scaling of drain lines,
tanks, pumps, marble bed, demister, and reheater and (4) scale
build-up on I.D. fan rotors, which resulted in fan imbalance
and vibration.
In addition to the problems just mentioned, the S02 re-
moval was quite low because of everburning of limestone in the
furnace and dropout of the lime with the ash in the bottom of
the scrubbers.
After the first few months of operation, the scrubbers
were modified, as shown in Figure 6, to include (1) addition
of soot blowers in the gas inlet duct and under the reheater
-------
LAWRENCE NO. 4
CE-APCS
..XTO STACK DECEMBER 1968
^OFflN SC«<2. ^™ «
WATER. SEAL } ^ HOT
^rtr-
a.
FROM m
AIR i- -"-
HEATER
ORAI
TANK
^/VVVVWVWVVt"" WlD
*W\AA/-X/WVW- -— .. — "fc^
REHEAT
to
///////////////, DEMISTER 2
:u-T-,:.««T-^-i MARRLF RED co
rifTIT^ CLARIFIED FROM POND FROM
^/
Y
N 1 *1
: _d >ei , ^ ^
btAL «f^H
p&-
> HIK r" • '^-'
HEATER
^^/\AA/>/ViAA AA^
K'vWAAvsAA'^
Ill/lltllflllll
A 1 JL 4X1
'•••'T' •'«-"Ti>"
T T T T
\/
HOT
t— - H20
REHEAT
LAWRENCE NO. 4
CE-APCS
OCTOBER 1969
SCRUBBERS(2)
•SOOT BLOWER AIR
DEMISTER
VERHEAD'PRAY
MARBLE BED
4 oa
«=r
V 5
1 OVERFLOW . ?>
:*._
x:
ICLARIFIED FROM POND
(2) TO SOLID
-j£^\ n T c r>f*c R i
H) (2) SOLID DISPOSAL POND RECYCLE TANK DRAIN TANK PQND
(1) (1)
Figure 5.
Figure 6.
WATER,
FROM
AIR
HEATER
JO STACK
ID FAN
HOT
REHEAT -
DEMISTER
OVERHEAD SPRAY
MARBLE BED
LAWRENCE NO. 4
CE-APCS
OCTOBER 1970
SCRUBBERS(2)
•SOOT BLOWER AIR
CLARIFIED
OVERFLOW.
(2) RECYCTTTATlK
(1)
POND
DRAIN
TANK
TO
SOLID
DL
FROM"
AIR J
HEAJEK~
i
WATER
SEAL
-ir
^xTvTO S
ypffio F
'•'VWAA/VA^A^S
SAAJW*A/V\A/>—
?
^
^
' -i" f -r
y
0
M
TACK
AN
HOT
— H20
~ REHEAT
EMISTER
ARBLE BED
-OVERFLOW
LAWRENCE NO. 4
CE-APCS
OCTOBER 1972
SCRUBBERS(2)
(ENLARGED DEPTH* 4' )
•SOOT BLOWER AIR
* WATER WASH LANCE
CLARIFIED
FROM POND
i TO
(2) SOL ID
-0 nrspnsfti
(2) RECYCLE TANK (1)
(ENLARGED)
DRAIN TANK POND
Figure 7.
Figure 8.
-------
bundle to prevent plugging (2) raising of the demister to reduce
carryover of solids, (3) directing the overflow liquor from
the pots to the pond, and (4) installation of a large recycle
tank and pump to catch the highly alkaline underflow and recir-
culate it to the marble bed. Other modifications to combat
corrosion and pluggage were installation of a new type of spray
nozzles and lining of the bottom section of the scrubber tanks
with gunite.
Most of the problems were reduced but not eliminated by
these modifications. The new recirculation system did improve
the S02 removal efficiency of the scrubbers.
In an effort to further minimize corrosion, erosion, scaling,
and plugging, additional revisions were made during the summer
of 1970. The configuration of the scrubber in October 1970
is shown in Figure 7. The major revisions were the following:
1. Sandblasting and coating the interior of
the scrubbers with two coats of glass flake
lining.
2. Replacement of all internal steel pipes
with plastic and fiberglass.
3. Replacement of the stainless steel demisters
with fiberglass demisters.
4. Addition of a ladder vane under the marble
beds to improve gas distribution.
5. Modification of the pot overflow drain piping
to allow the liquor to return to the recycle
tank for a semiclosed liquor loop operation.
6. Removal of the original copper fin tubes of
the reheater coils, which because of close
fin spacing, caused the reheaters to plug
easily. Also, the fins were flattened by the
soot blower jets. The copper units were re-
placed with a carbon steel tube fin coil.
Demister pluggage continued to create serious problems.
Manual washing every other night was required to maintain the
output required of the unit.
36
-------
In the summer of 1972 the scrubbers (on both Units 4 and
5) were modified to operate using a high solid slurry crystalli-
zation process to control saturation and precipitation of scale
within the scrubber. These latest major modifications, shown
in Figure 8, included enlargement of the liquor recirculation
tank and replacement of many components, such as piping, nozzles,
pumps, and mixers. Also, the demisters were replaced with a
new two-bank fiberglass unit fitted with high-pressure wash-
water lances.
Operation of the two FGD systems since the fall of 1973
has been most successful to date. Some of the remaining problems
are isolated areas of corrosion, failure of expansion joints,
fouling of the demister, rapid wear of slurry pumps, and failure
of valves.
Future modifications of the FGD systems on each boiler
will be concerned primarily with alleviating problems inherent
in furnace injection with limestone. Therefore, Unit 5 will
be converted to a tail-end, wet limestone scrubbing process
by fall 1975.
The current schedule for revisions on Unit 4 is as follows:
a) By January 1975, construction of two 2-stage
scrubbers (venturi followed by spray) will
be started.
b) By January 1976 the two new scrubber modules
will be operational. The present scrubbers
will be kept in service while the new scrubbers
are being built.
c) By January 1977 the present scrubbers will be
razed and an ESP will be installed. It is
anticipated that the ESP/venturi/spray flue
gas cleaning system, will be operational by
January 1977, The new system will have forced
oxidation via aeration to produce sulfates.
37
-------
THE WILL COUNTY POWER STATION
Power Generating Facilities
The Will County Power Station of Commonwealth Edison Company
is located on the Chicago Sanitary and Ship Canal near the town
of Romeoville, in Will County, Illinois. Delivery of coal and
limestone to the Will County Power Station is primarily by
barges using this canal.
The station has four electric-power generating units with
a total rated capacity of 1147 MW. Only Unit 1 is retro-
fitted with a flue gas desulfurization system.
Unit 1 has a wet-bottom coal-fired boiler producing
167 MW of electricity. The boiler was manufactured by Babcock
and Wilcox and was installed in 1935.
The coal presently being burned has an average gross heating
value of 9463 BTU/lb; ash and sulfur contents are 10 and 2.1
percent, respectively.
The boiler is fitted with an electrostatic precipitator,
having a 79 percent actual particulate collection efficiency.
Normally, the precipitator is not used except when the FGD
scrubbers are out of service.
FLUE GAS DESULFURIZATION SYSTEM
FGD Process Description. The wet limestone flue gas desulfuri-
zation system was placed in service on February 23, 1972. The
system consists of limestone handling and milling facilities,
two FGD scrubber modules (identified as A and B), and a sludge
treatment and stabilization unit. Each FGD module consists
of a booster fan and a Venturi scrubber followed by a turbulent
38
-------
contact scrubber (TCA) SO- absorber tower. Operational problems,
described later in detail, have caused shutdown of Module B
and modification of Module A.
The FGD process is described in terms of the three major
operations: limestone milling, particulate and SOp scrubbing,
and sludge treatment and disposal.
The limestone milling system consists of a limestone rock
conveyor, two 260-ton-capacity limestone storage silos, two
wet ball mills, and a slurry storage tank. The total storage
capacity of the two silos is equivalent to 48 hours of the FGD
requirement of limestone at full load. The limestone is 97.5
percent calcium carbonate, 0.99 percent magnesium carbonate,
and 0.48 percent silica. It is received as coarsely ground
particles (about 1/2 inch diameter or less) and is finely ground
to 95 percent through 320 mesh. Grinding is accomplished in
two wet ball mills, each rated at 12 tons per hour. The lime-
stone is discharged from the mills in the form of a water slurry
containing 20 percent solids. The slurry is piped to a 4-hour
capacity, 62,500-gallon storage tank, which supplies the lime-
stone for the scrubber modules.
The two scrubber modules on Will County Boiler 1 are
shown in Figure 9. Each module was designed for 385,000 acfm
throughput at 355°F, each handling 50 percent of the 167-MW
boiler's flue gas load. The liquid and gas flow through a typi-
cal module is shown in Figure 10. The flue gas emerges from
the boiler, passes through an electrostatic precipitator and
is directed to the venturi scrubber. The gas is forced through
the rectangular venturi throat and comes in contact with jets
of slurry sprayed from high-pressure nozzles located on each
side of the throat. The scrubbing efficiency and consequently
the pressure drop in the venturi are regulated by actuating
a drive mechanism that varies the width of the venturi gap.
The quenched flue gas mixed with slurry droplets exits
from the venturi throat and turns through the sump, in which
39
-------
-R-
o
LIMESTONE
BUilKER
•RECYCLE AND
MAKE-UP WATER
BALL
MILLED MILL
PRODUCT
TANK
ABSORBER
RECIRCULATION
PUMPS
JO SETTLING
POND
Figure 9. General flow diagram of the FGD system at Will County Unit 1.
-------
TO SLUDGE
WASTE
POND
VENTURI
PUMPS
FLUE GAS
VENTURI
VENTURI
RECIRCULATION
TANK
CLEAN GAS
TO REHEATE.R
SUMP
ABSORBER
ABSORBER
RECIRCULATION
TANK
FROM MILL
SYSTEM
ABSORBER
PUMPS
Figure 10. Flow diagram of a typical FGD module,
-------
a great reduction in flow velocity causes the slurry droplets
to fall out of the gas stream into the venturi circulation tank.
The gas then flows upward through the S09 scrubber tower,
^>
and passes through two perforated trays, which are wetted with
a shower of limestone slurry circulated from the SO- scrubber
tank. These trays provide an extensive contact surface for
absorption of S0_. from the circulated slurry.
The scrubbed gas, now essentially free of fly ash and S0~ ,
rises up the tower and passes through a two-stage Chevron-type
demister. The fine mist droplets of liquid carried over with
the gas coalesce on the demister's vanes, and the resulting
large droplets fall through the tower by gravity. The demister
is equipped with two sets of wash water lances. It is washed
continuously from below by 120 gpm of fresh water and also
intermittently from above by 1000 gpm of pond water for 30
seconds every 2 hours. The demisted gas then enters the reheater
unit, where its temperature is raised from 128°F to about 200°F.
This temperature increase imparts buoyancy to the gas and prevents
condensation in the fans, ducts, and the brick-lined steel stack.
The bare-tube reheater is divided into three sections;
the first is made of 304 stainless steel and the other two of
corten steel. Each reheater has four soot blowers to maintain
tube cleanliness. The heat is supplied from Unit 1 boiler as
350 psig, saturated steam.
To compensate for draft loss across each scrubber module,
an induced-draft fan boosts the pressure of the heated gas and
delivers it to the suction side of the boiler induced-draft
fan.
There are two slurry tanks for each module: the S02 scrubber
circulation tanks, to which fresh limestone and make-up water
are added, and the venturi scrubber circulation tank, from which
the spent limestone slurry (or sludge) is discharged. These
two tanks are interconnected in such a way that the spent liquor
42
-------
from the SO scrubber tank overflows into the venturi circula-
tion tank. Each tank is fitted with agitators and pumps. The
slurry recirculation rate in the SO^ scrubber is about 8750
gpm. The recirculation rate through the venturi is 5800 gpm.
The sludge treatment and disposal system used at Will County
is illustrated in Figure 11. Spent slurry from all the venturi
tanks is discharged to a 65-foot-diameter clarifier tank. During
emergencies or when the clarifier or stabilizer unit is inoper-
able, the slurry can be discharged to a small clay-lined holding
pond. Overflow from the clarifier is returned to the process;
underflow is stabilized in a facility built and operated by
Chicago Flyash Company under contract with Commonwealth. About
200 pounds of lime and 400 pounds of fly ash are used per ton
of dry solids of sludge. The fixed sludge is transported by
concrete mixing trucks to a small pond on site for solidification.
The stabilized sludge solidifies in about 1 week, after which
it is hauled by Material Services Company to an offsite dump
for use as landfill. Recently, this contractor obtained State-
EPA permit to develop an offsite land for sludge disposal.
FGD System Performance. As mentioned earlier, the FGD modules
at Will County 1 were plagued with numerous problems since
February 1972. Many of these problems have since been solved,
and others have become manageable. Improved operation of the
FGD system is reflected in increasingly higher availability
figures for Module A in 1974. When Module B is modified on
the basis of experience gained with Module A, its performance
and availability should be comparable to those obtained from
Module A to date. The monthly availability factors for each
module are presented in Table 4. A. brief chronology of the
major problems encountered and their solutions is given below.
1972 - Demister plugging was a constant problem, mainly
because of heavy accumulations of limestone on the bottom of
the demister, which were partly due to low wash-water pressure
43
-------
SCRUBBER SLUDGE POND
THICKENER UNDERFLOW
HOPPER
DTD ON-SITEO
DISPOSAL BASIN
RECYCLED THICKENER OVERFLOW TO MODULES
Figure 11. Limestone sludge stabilization facilities.
-------
Table 4. FGD SYSTEM AVAILABILITY FACTORS
FOR WILL COUNTY UNIT NO. 1
Period
Month/year
March, 1972
April
May
June
July
August
September
October
November
December
Availability , %
Module A
0
33.99
69.48
8.44
78.65
0
0
0
21.77
Module B
35.4
13.68
31.82
30.91
20.60
29.48
0
0
29.69
Period
Availability, %
Month/year Module A
January, 1973
February
March
April
May
June
July
August
Septen±>er
October
November
December
0
21.
64.
6.
0
0.
51.
19.
0
32.
50.
0
94
79
17
96
36
23
00
80
Module B
0
24.26
10.69
13.08
0
0
0
0
0
0
0
0
Period
Availability, %
Month/year Module A
January ,
February
March
April
May
June
July
August
1974 0
0
20.98
72.27
93.12
54.50
95.75
91.25
Module B
0
0
0
0
0
0
0
0
-------
caused by leaks in the pond-return bypass. Because of plugging,
Modules A and B were out of service for several days each month
during March, April, June, and July 1972. The modules were
also out of service from September 26 to November 21, 1972,
because the boiler was down during that period.
As a means of solving the demister problem, the scrubbers'
slurry nozzles were lowered and the slurry circulation system
was left out of service to keep the demister clean. Because
this change gave no improvement, the demister elements were
hand-washed, a procedure that decreased demister plugging but
increased difficulties with the venturi nozzle because broken
elements from the demister entered the slurry system.
Because the reheater of Module B vibrated excessively during
high rates of gas flow, Module B was taken out of service in
April for reheater modifications. These modifications included
rebracing of the reheater tubes and installation of a baffle
plate to reduce vibrations.
Other reasons for outages in 1972 were erosion and plugging
of spray nozzles, internal and external buildup of deposits
on venturi nozzles, corrosion cracking, sulfite blinding, and
fan vibrations.
1973 - Demister plugging continued in 1973. Furthermore,
the demister on Module B broke loose from its mountings and
the resultant carryover of wash water plugged the reheater,
also leaked because of chloride pitting corrosion. Module A
was down from April 24 to May 24, 1973, and Module B has been
down from April to date. No scrubbers were in operation from
August 27 to September 26, 1973.
As a further step toward resolving the pluggage problems,
the system of continuous underspray and intermittent overspray
was installed in Module A to wash all the demister compartments.
Extra nozzles were added and a clean water supply was maintained.
The reheater bundle was also retubed.
-------
1974 - This year to date only Module A is in operation.
Although plugging of the demister and reheater has not been
completely eliminated, the improved washing system has provided
a partial solution. The other significant problems are freeze-
up of the venturi throat drive mechanism, tank screen blinding,
duct corrosion, and vibrations. Loss of building heat caused
extensive damage and kept the scrubber down the entire month
of January 1974.
Some components of Module B, such as reheater bundles have
been cannibalized for use in Module A. Modifications of Module
B are not expected to begin until performance of Module A is
satisfactory.
47
-------
HAWTHORN POWER STATION
Power Generating Facilities
The Hawthorn Power Station of Kansas City Power and Light
Company is located in a heavily industrialized area on the
north bank of the Missouri River in East Kansas City, Missouri.
The plant operates five coal-fired boilers. Boilers
1 and 2 are considered peak boilers, each rated at 100 MW.
Boilers 3 and 4 are cyclic boilers. When burning natural
gas, each is rated at 140 MW; when burning coal the capacity
of each boiler drops to 100 MW. These four boilers were built
and placed in service in early 1950. Boiler 5 was placed
in service in early 1970; it operates as a base-load boiler
at a rated capacity of 500 MW. All these boilers are dry-bottom,
pulverized-coal-fired units, designed and manufactured by Com-
bustion Engineering.
Two grades of coal are burned: the coal with higher ash
content typically contains 14 percent ash and 3 percent sulfur,
with a heat content of 11,400 BTU/lb; the lower-ash coal contains
11 percent ash and 0.6 percent sulfur, with a heat content
of 9800 BTU/lb. Natural gas is also burned occasionally.
Of the five boilers at the Hawthorn plant, only Boilers
no. 3 and 4 are fitted with flue gas desulfurization (FGD)
equipment.
FLUE GAS DESULFURIZATION SYSTEM
FGD Process Description. Initially, the FGD systems on both
boilers were designed to operate by injection of dry limestone
into the furnace, followed by tail gas scrubbing of both the
SO,, and the furnace-calcined limestone (as well as the fly ash)
48
-------
from the flue gas stream. Because of tube pluggage in Boiler
4, attributed to limestone injection, this mode of operation
was modifed. Ground limestone is no longer injected into the
furnace but instead is introduced into the flue gas near the
gas inlet to the tail scrubber tower. The FGD system on Boiler
3 continued to operate as originally designed. Both systems,
however, have undergone minor modifications to overcome such
difficulties as build-up of sediment and pluggage of various
components.
The FGD system on each boiler consists of two identical
modules, each capable of treating 250,000 acfm of flue gas at
300°F. Bypassing of the modules is possible through a system
of ducts and dampers around each module.
Details of a typical module are shown in Figures 12 and 13.
Each scrubber module is 18 feet wide, 26 feet long, and 56 feet
high. The lower 16 feet of the module is the reaction tank,
in which the materials have time to complete their chemical
reactions to remove sulfur dioxide. Two mixers are installed
in the tank to provide thorough mixing and to keep the reac-
tants and the fly ash in suspension, one on the lower back
wall and the other on the upper front wall.
A sloping, perforated plate near the top of the reaction
tanks acts as a strainer to catch large solid objects, such
as scale, mud, or marbles, and funnels them to the mouth of
a jaw crusher. Periodic operation of this equipment reduces
these pieces to fine particles, which are discharged in a slurry
to the fly ash pond.
As the flue gas enters the absorber tower above the liquid
level in the reaction tank, it comes in contact with slurry,
which is sprayed through 81 nozzles on 18 headers located under
a marble bed. The marble bed consists of a 3- to 4-inch thick-
ness of 3/4-inch-diameter glass marbles, retained on a perforated
plate. Quenched flue gas and the sprayed slurry bubble through
49
-------
STACK
LIMESTONE
HOPPER
CRUSHER
Ul
o
BYPASS
SEAL
i i
r i
i t
I i
FURNACE
AIR
DRAIN
LEGEND
AIR POLLUTION CONTROL SYSTEM
NOT APCS EQUIPMENT
Li I 111 III
DEMISTER SPRAYS
SCRUBBER CELL
OVERFLOW POTS
YYYYYyvvvYY
UNDERBED SPRAYS
j i ' i .' ' <' f i if
' ~ XT -i „ '*
, • ( I.D.U
'FANS/'
.. -'
TO DEAERATOR
FROM DEAERATOR
FROM WATER SUPPLY PUMP
QKhHtATERS
l-
DRAIN FROM
BYPASS SEAL
TO BYPASS SEAL
TO DEMISTER
SPRAY
TO GAS DEFLECTOR-*
SPRAY
TO MAKEUP
Figure 12. Process flow diagram of a typical FGD module,
similar to those used on Boilers 3 and 4,
-------
STACK
PULVERIZER
FAN
STACK GAS
SCRUBBER
RECYCLE
PUMP
TO ASH DISPOSAL
POND
Figure 13. Sketch of a dry limestone furnace injection system with tail end scrubbina.
-------
the marble bed, giving the marbles a spinning motion. The pur-
pose of the moving marbles is to provide a large amount of sur-
face area within a relatively small volume and thus to facilitate
contact of the reactants. The liquid level above the marble
bed is 10 to 13 inches high. The liquid overflow from the bed
is discharged through 48 drain pots. The scrubbed flue gases
flow upward 10 feet to a double row of Chevron separators designed
to remove water vapors carried by the flue gases. The gases
then pass around horizontal fin-tubed heaters that reheat the
gases to about 175°F before they are discharged to the stack
by the induced-draft fans. Gases from all modules on Boilers
3 and 4 are discharged to the atmosphere through a common
stack 200 feet high.
Spent liquor from the reaction tank of each module is dis-
charged to a 115-foot-diameter clarifier tank. The overflow
from this tank flows to a clear well tank, where make-up water
from the Missouri River is added. The discharge from this tank
satisfies the water requirements of the FGD system. Distribution
of the water is shown in Figure 12.
Sludge in the underflow from the clarifier tank is untreated.
The thickened sludge is pumped to a 160-acre pond, which is
also used for disposal of fly ash from the other boilers.
Inlet and outlet dampers are provided on each scrubber
module to route the gases either through or around the module.
To ensure passage of gases through the scrubber when it is in
operation, a large U-shaped bypass seal is filled with water
to positively close the bypass.
The demister wash system uses a set of eight water lances
that turn on automatically to wash mud from the demisters when-
ever the dampers close. The lances are located four above and
four below the demister vanes.
The reheater tubes are cleaned by two steam lances placed
below the finned tubes. The heating medium is hot water from
52
-------
the suction of the boiler feed pumps, delivered at 150 psig
and 325°F.
FGD System Performance. Modifications recently completed by
Combustion Engineering and Kansas City Power and Light Company
apparently have significantly improved operation of the FGD
systems. Module 3B on Boiler 3 recently operated for 550
hours with an availability factor of over 75 percent. Continu-
ous monitoring equipment showed sulfur removal efficiency in
the range of 85 to 95 percent. Particulate removal efficiency
was 98 to 99 percent.
As a result of the recent modifications, the overall avail-
ability factor for the two modules on Boiler 3 has increased
from about 30 percent to 70 percent. Availability of the FGD
system for Boiler 4 has lagged because of the many problems
attributed to dry limestone injection. Apart from those entailed
in switching of Boiler 4 from furnace injection to tail
gas injection, the major problems encountered with the two FGD
systems at the Hawthorn Power Station have been identical.
Analysis of the problems encountered during and since starting,
reveals that nearly all were due to mechanical design rather
than process chemistry.
1. Sediment^ BuiIdup: The reaction tank of each module
is equipped with two mixers to keep the 10 percent solids in
suspension. The original 15-hp mixers did not do this job
properly, since buildup of hard mud, 6 to 8 feet high, developed
in the corners of the reaction tank. Replacement of the original
mixers with 25-hp units gave no significant improvement. The
most recent modification aimed at eliminating sediment buildup
consists of rounding off the bottom corners of the tanks by
welding triangular steel plates, as shown in Figure 14. Fur-
ther, a new make-up water piping system was installed, with four
1-inch nozzles located on the walls about 6 feet from the base
of the tank and oriented to promote circulation and prevent
settling of solids.
53
-------
MAKE-UP WATER NOZZLES
Figure 14. Sketch of the reaction tank showing
the rounding off of the tank's corners and
the installation of make-up water, sediment
flushing system.
54
-------
2. Pluggage of the Marble Bed: In the initial construc-
tion, the drain pots above the marble bed drained into horizontal
headers that penetrated the module walls and emptied into the
reaction tank. These horizontal headers plugged soon after
the system was placed in operation. The headers were removed,
and sloping headers were placed inside the reaction tank. Al-
though these did not plug as fast, pluggage was still a problem.
An additional problem occurred frequently when drain pots lifted
off the marble bed allowed marbles to fall into the reaction
tank. Both of these problems have been solved by installation
of new stainless steel drain pots wich expanded metal covers;
a 3-foot rubber sock allows each drain pot to drain directly
into the reaction tank. The pots are held in place by "J" bolts
that attach to the marble bed. No pluggage of the new drain
pots has occurred.
Pluggage in the marble bed has been alleviated also by
an operational change. Increasing the liquid-to-gas ratio from
21 to 26 gallons per thousand cfm has increased the flow of
liquid sufficiently to ensure that all parts of the marble bed
are washed by at least two nozzles.
3. Excessive Wear of Spray Nozzles; Orifices of the spray
nozzles were severly worn by abrasive action of the fly ash.
The initial nozzles lasted only a few days; replacement was
required because the enlarged orifices caused buildup of mud
in the marble bed, overloading of the recycle pump, and clogging
of the demister. Although KCPL installed a series of nozzles
of different materials from different vendors, the longest life
obtained from any of those was about 3 weeks. Since these nozzles
are very expensive, KCPL designed a nozzle that could be made
in the shop from pipe parts. These nozzles also lasted only
about 3 weeks, but their cost was only 5 to 10 percent of the
cost of vendors nozzles. Recently new ceramic nozzles were
installed, and after 550 hours of operation they exhibited no
signs of wear. Although this is not an exceptional service
55
-------
time for nozzles/ it is by far the best performance to date,
since all other nozzles were completely worn in an equivalent
amount of time.
4. Demister Pluggage: During sustained operations the
scrubber modules were taken out of service every 3 days to
allow the hosing of mud from the demisters. Pluggage has been
controlled by adding retractable water lance blowers under the
demister and by moving the rotary water lance blowers from below
the demister to between the rows of demister vanes. With these
modifications the demister wash system now performs satis-
factorily.
5. Bypass Seal Pluggage: During periods when the flue
gas is bypassed around the module, the drained-out water seal
acts as a mechanical dust collector, accumulating fly ash.
When the module is returned to operation and water is turned
on, the fly ash in the water seal turns into mud, hampering
operation of the bypass seal. Pluggage in this seal was elimi-
nated by changing the flushing sequence within the seal.
56
-------
REID GARDNER POWER STATION
Power Generating Facilities
The Reid Gardner Power Station of the Nevada Power Company
is located near Moapa, Nevada, about 50 miles north of Las Vegas.
The station has two electric power generating units (Reid Gardner
1 and 2), each rated at 126 MW. The units are identical in
design, and both operate under base load conditions. The coal-
fired boilers were manufactured by Foster Wheeler. Reid Gardner
1 was placed in service in 1965, and Reid Gardner 2, in 1968.
Average gross heating value of the coal burned is 12,450 BTU/lb;
average ash and sulfur contents are 9 and 0.6 percent, respec-
tively.
Reid Gardner 3, with design capacity of 125 MW, is under
construction, with start-up scheduled for the spring of 1975.
Reid Gardner 4, also a 125-MW unit, is in the planning stage,
with operation expected in 1978.
FLUE GAS DESULFURIZATION SYSTEM
The sodium carbonate base flue gas desulfurization (FGD)
systems on Reid Gardner 1 and 2 were manufactured by Combustion
Equipment Asssociates and placed in operation during March and
April 1974. The design of these systems was based on data
gathered from operation of an 8000-acfm pilot plant, which
operated at the Reid Gardner Power Station during 1971 and 1972.
FGD Process Description. The FGD system on each boiler consists
of a single module, designed to handle 473,000 acfm of gas at
350°F. The module is made up of a twin variable-throat venturi
scrubber followed by a single-stage S02 scrubber tower. The
FGD modules on both boilers share a common trona (sodium car-
bonate ore) storage and beneficiation plant. Bypassing of each
57
-------
module is possible through a guillotine-type bypass valve. As
shown in Figure 15, the hot flue gas from the boiler passed first
through an existing mechanical collector (multicyclone), where
about 75 percent of the fly ash is removed. Pressure is then
boosted by an induced-draft fan before the stream splits as it
enters the twin variable-throat venturi scrubber. The hot flue
gas is quenched by a shower of circulated sodium-base liquor,
sprayed from tangential nozzles located on the walls of the
venturi. The gas-liquid mixture then passes through the venturi
throat and exits into the sump, where the reduction in velocity
causes the liquor droplets to fall and separate from the gas
stream. A liquid-to-gas ratio of 10 gallons/ 1000 acfm and
a pressure drop of about 15 inches of water are maintained in
the venturi for efficient removal of SO- and particulates.
The scrubbed gas then enters the SO2 scrubber tower tangentially.
The centrifugal force created causes the finer droplets to
coalesce near the tower's wall and separate from the gas stream.
The scrubbed gas, which is now essentially free of fly ash and
of most of its sulfur dioxide content, rises through the tower
and bubbles through the 3/8-inch-diameter holes of a single
sieve tray. The tray is flooded with clear water from the ash
pond at a rate equivalent to 1 gpm per 1000 acfm of gas and
is intermittently washed with fresh water from below. More
SO2 is absorbed in the clear water tray and the pH of the liquid
effluent drops to about 5. The cleaned gas continues up the
tower and passes through the radial-vane-type mist eliminator,
where the liquid droplets carried over with the gas are trapped
and removed.
The gas is reheated by direct contact with hot air to pro-
vide buoyancy and to prevent condensation of the gas on duct
walls or in the stack. Ambient air is drawn in by a separate
fan and steam-heated in a shell and tube heat exchanger before
it intermixes with the scrubbed wet gas in the outlet duct of
the module. The combined gas stream, which is now warmer, enters
the 200-foot-tall stack and discharged to the atmosphere.
58
-------
TRONA (SODIUM CARBONATE)
WATER FROM
ASH POND
\D
TRONA
SILO
BOILER
CYCLONE
/ V
\/
CLARIFIER
STACK
WATER FROM
ASH POND
TO SLUDGE POND
MODULE 1 OR 2
Figure 15. Simplified flow diagram of the FGD system on each of
Reid Gardner 1 and 2 electric power generating units.
-------
The S02 scrubbing reagent in the FGD systems is trona,
an impure form of sodium carbonate. It is mined by Morrison
and Weatherly Chemical Products Company from Owens Lake in
California and shipped to the Reid Gardner Power Station as
a powder. The trona is transferred pneumatically to a tall
silo having a 48-hour-supply storage capacity. It is then trans-
fered to the day bin from which it is conveyed by screw conveyor
to a slurry tank and mixed with fresh water to dissolve its
sodium carbonate content. The slurry, containing such impurities
as sand and sodium chloride, is pumped into a clarifier, where
the insoluble impurities settle out. The clarified sodium car-
bonate liquor is then pumped to the modules recirculation tanks.
The effluent from the FGD system is made up of three sources;
the slip stream of spent liquor discharged from the recirculation
tank, the acidic effluent from the clear water tray, and the
alkaline clarifier underflow which serves to neutralize the
pH of the combined liquor before it is discharged to the sludge
settling ponds. The spent liquor is pumped to a small 6-acre
pond, from which overflow can flow to a 45-acre pond. Because
operation of the FGD system has been limited and the evaporation
rate from the 6-acre pond is high, no liquor has yet trickled
to the larger pond. The water system operates on an open loop,
with no liquor recycled to the modules from the ponds.
FGD System Performance. Since their start-up in early 1974,
both modules at the Reid Gardner Power Station have performed
satisfactorily. Because of difficulties in obtaining sufficient
quantities of trona, however, operation of the modules has been
subjected to frequent interruptions and the flue gas has been
bypassed to conserve trona. The longest continuous operation
was on Module 2, for 18 days. The FGD system has never had
a forced shutdown due to problems originating in the equipment,
and thus can be considered as essentially 100 percent avail-
able, with operations interrupted only by the trona shortage.
Since start-up, each module has operated for 900 hours.
60
-------
Since the major problem preventing continuous operation
of the FGD modules has been and still is the inadequate supply
of trona, the Nevada Power Company has been experimenting with
the use of sodium carbonate brine. One tank truck shipment
of brine from Green River, Wyoming, has been delivered and will
soon be tested on Reid Gardner 2 unit. Larger quantities cannot
be received, since the brine freezes at ambient temperature
and the plant lacks the proper facilities for storage.
The system experienced the usual minor start-up problems,
such as rubber liners on recirculation lines becoming loose
as a result of poor fabrication, freeze-up of instruments,
plugging in bypass lines of some control valves, and leakage
in the mechanical seals of the recirculation pumps. Further
freeze-up problems were experienced with the louver type bypass
dampers, which were later replaced by a combination of louver/
guillotine-type dampers.
As expected, this soluble-base FGD system has not experi-
enced such problems as equipment scaling, demister plugging,
erosion, corrosion, which are normally associated with lime/
limestone FGD systems.
61
-------
INFORMATION SOURCES
Much of the information reported here concerning FGD systems
operation was obtained recently by visits to plant installations
and discussions with plant personnel:
Hawthorn Power Station June 6, 1974
Will County Power Station June 20, 1974
Lawrence Power Station August 13, 1974
Reid Gardner October 8, 1974
In addition, the following reports are cited for further
references :
1) Palermo, Joseph V. "Operating Experience on S02 Scrubbers,
Hawthorn Units 3 and 4" Report to Missouri Valley Electric
Association at Kansas City, Missouri on April 18, 1974.
2) Gifford, D.C. "Will County Unit 1, Limestone Wet Scrubbers,
Description and Operating Experiences: November 30, 1973.
3) Gifford, D.C. "Will County Unit 1, Limestone Wet Scrubber,
Waste Sludge Disposal" Paper presented at the Electrical
world Symposium in Chicago, Illinois on October 30-31, 1973.
4) Green, Kelly. "Operating Experience with Limestone Injection-
Wet Scrubbings Stack Gas Treatment" Paper presented at the
Frontiers of Power Technology Conference at Stillwater,
Oklahoma on October 10-11, 1973.
62
-------
APPENDIX A
FLUE GAS DESULFURIZATION SYSTEM STATUS REPORT
OCTOBER 1974
63
-------
JiGAATfiTu5J?.£P-0.l51.lo/7'+.
I .0. I\)UKHER AMU COMPANY
TABLE 1
Su,-iHA!JTUCKY b-T) ..... 6
IjCNO OHIu 1-77 b
.10-72 :i_
6-78
FA^WICJGTOW
NLXICO
FARM1UGTON NEW MEXICO
.... wHEATLAKD WYOMING
WYOMING
WYOMING
. _ CR*IG COLORADO
HAYOE.N COLOiiAUO
COLORADO
OHIO
OHIO
30.
3i
_32
"53
CO'1.V.
_0 A 1 K_r_ L A h D . . t C » t.H_t_U Of
DC.TPOIT CIUSCN
GcI-.'lAAL
37
39
"40
I'MOln^Al'OHS POWLK AND LIGHT
JS Ai\SA.S..C I I.y_['OwEIL.fi.jL,I.«i.HT __
KANSAS CITt POwlIK X LlOHf
AS CITY I:'_C.^EK a LIGHT
"
ALNA
SI.CLttIR NO
ELUAMA..
PHILLIPS
_CiiE\/r
-------
FGD STATUS RrpoRT IU/7'»
I.t'. MUMiER A«L CU.'-'iFAlJY WANE
43 KANSAS POlnCR & LIthT
44 KAf.SfiS PO.JLR Z LIGHT
45 MINFUCKY «J1 ILITIES
46 LOllif.VILLC 6AS ?, ELECThIC
47 LuUiSV ILLC l>fj i LLi.CTh.il.
•*(> LouirviLLr CAS x ELECIKIC
49 LOUISVILLE t.AS A LLELTnlC
50 LOuisviLLF GAS f. ELECTRIC
fil LOIHSVJLLE HAS X ELIOMC
b? LiH'isviLLr GAS i ELECTRIC
53~ Lt'uisv JLLL G.sS X ELLClrxIc
54 LUIilSVHLF. GAS £ tLLCTKIC
-Jfi uOUIS'J It-Lc GAb ^ El.LCTKlC
5f. LuuisviLLi t./.5. f. ELECTRIC
57 tlDlgTAIi'A f'fiwLK
to NO^TA'-IA PtiuEK
62 «'LV ;">!'> A Pf)-*tR
tJ fiv^J i-oii<
65 Nlw F. JOLA'-iU LLFC bYiiTEh
66 ?;>)IUliLKN IKOIA^-A I'llli SLMVIcE
67 NOrtTl.EPi-J JtiLUA'lA 1MB SERVICE
f.P .^uRTHKRii lijI'JA.-.A Pijii SCiJVICE
70 i.iiRT.ir.'H'j j.lAif;, f'0>Li<
71 PKVNStLVA"];; f"1!^ c-'' CO.
72 PtMiStLVA-'iiA F'OxEi* CO.
73 PLNivSYL VA'fi A 1'fV.ti? CO.
74 I'MILtilEUP-a/, ELF.ClHlC
75 ii)Tr,-ac ELF.CTKIC j. PC^.'LK
76 PuC( fC Sf'itvICtl cK jNOI,vu/\
77 1-urLlC SLHVlCt OP f. L> I^LXICO
7fc KiuLlC SCiiviCE Or r,Lx ^|LXICO
79 PUBLIC StHVICC OF i.;Ew hLXlCO
fio PUBLIC SERVICE CF NEW .^.EXICO
fll •>. CA'<(iLI,'lA F-U'i SEKV AUTHOHTY
f-2 biLT riJVCH PKGjFCI
03 S.\LT HIVtK P-'Oji'.CT
fi4 Sf\LT rtJVF.". !':"10JECT
1. OPrKATlOljAl. Ui.lTS 4.
2. U;;ITS ui^nER co^STHucrio.^i 5.
3. PLr.v-icn - coNThftcT AV.AKUEQ 6.
TABLE 1
SUM AH Y OF KGO SYSTEMS
UNIT NA^E
LAi-JKENCE NO 4
LAWhEEs.'CE I»i0 5
GKEtU KlVEi< 1283
C(. JL KUN NU 1
C»iMt HUK NO 3
CA'JE RUN NO H
CAUL KbN 'NO 5
Cji'JE. NUN NO b
f-ULL CHLEK NO l
ISiLL CKhEK l'«0 2
HILL CKLLK NO 3
rtiLU LtttH-is NO 4
PAODYS KU^ NO 6
COLSTR1P NO 1
COLa'FKlP iJO 2
CUI.ilKIH I-J0.3
COLiTNIP ;-ji),4
K^lu bAHU'-JER MO 1
ht.li) brtKrK'JLR t)0 2
KLlU ijiiKO^t-R NO 3
RElu (jA^U'^LR NO 4
UKAYTUi-J POINT Nu.3
UAILY NO. 7
BAILY fgO.8
MJTUHS.LL NO 11
SrlCH jLlni<,[. !JO 1
ShCKliOKNL i'^0 2
bnuCE -'-lAiJlif- itLU MO. 1 .
bKUCE MATgSf- IELD t;0. 2
l^rtui-E MATJSf IELD MO. 3
EijIJYSTCNL MO 1
DiCKcHSOfj NO 3
GlobOf: NO. 2
S/v 1 OUAfv NO 1
S«:i JLIAN NO 2
Sn J JOAN MO. 3
SA'I OUAN i-JO.4
GLOKGEFOwN NO 2
NAV/.JU NO 1
NAVAOC NO 2
tJ.tVAuO MO 3
PLAilivufj - LETTER OF INTENT SIGt-J
PLMNi-JECi - RilQUESTING/EVALUATING
COJfSlOERING ONLY f-G'J SYSTEhS
LOCATION
LAWRENCE KANSAS
LAURENCE KANSAS
SOUTH CARL10N KENTUCKY
LOUISVILLE KLN1UCKY
LOUISVILLE (vE^fUCKY
LOUISVILLE hEMUCKY
LOUISVILLE KENTUCKY
LOUISVILLE KENTUCKY
LOUISVILLE KENTUCKY
LouISVIuLE KENTUCKY
LOUISVILLE KLi-JlUCKY
LOUISVILLE KENTUCKY
LOUISVILLE KENTUCKY
LoblsVILLE KENTUCKY
COLSTRIP MONTANA
COLSTKii' MONTANA
COLS1RIP KOUTANA
MAOPA l-jLVAL-A
MAOPA UEVADA
SOHLRSLT MASSACHUSETTS
CMESTEKTON 1NUIANA
GAKY 1,'jOlAijA
fcECKEIi MINNESOTA
t3LLC.Ei< MINNESOTA
SMlPPI.-iGfliKT PENNSYLVANIA
SiilrHINGPoUT PENNSYLVANIA
SHIPPINOPORT Pt-HNSYLVANIA
EOOYSTONE PENNSYLVANIA
OICKCI'SOIJ f-iARYLAMU
Hhli.CETOwN INDIANA
WAIEKFLJh NEW f'.LXICO
WATLRFLOW NEW MEXICO
WATERFLOW NEW MEXICO
iCATERKLOv. NEW MEXICO
GEORGETOWN SOUTH CAROLINA
PAGE AK120NA
PAGE ARIZONA
PAGE AhliOKA
ED
BIDS
START UP DATE
12-68
11-71
4-75
fc-60
6-00
6-80
6-75
6-76
6-77
6-79
6-78
6-77
6-79
4-73
5-75
5-76
7-78
0-79
12-73
12-73
6-76
0- 0
0- 0
0- 0
6-75
b-7fa
3-77
6-75
6-75
6-78
12-71
9-73
5-75
10-77
1-77
b-76
5-80
5-77
3-76
10-76
3-77
STATUS
l
i
2
6
6
6
3
6
6
6
6
3
b
1
2
b '
6
1
1 .
2
4
6
6
6
2
2
2
2
2
6
2
1
2
3
3
6
6
5
6
6
6
-------
TABLE 1
.•yjMMARY Or I-GO SYSTEMS
I.D. NUMnER AMD COMPANY
UNIT NAME
LOCATION
START UP DATE STATUS
05
B6
67
ft*
69
91)
91
92
Ob
96
<)7
98
9CJ
SOuThL'KiM CALIKORi.U LljlSON
S')UT>if.Ff' C»Hf- 1'Ki'jI A EUISOw
bUJlHt.nr-1 CAl.lF.-'k.'jI;, LJlSON
c..)UTr-LRN CtLIKORiviA E013CN
MHITlttUM .llSblSSil'Pl PVvK COUP
SOUTh-El-U '• ISslSSlPpl PhK COOP
SJliT.,«-Jf SHK'J PliHLlC SCKVJCL
SPKriuf-'lL'Lu liTll. iTy 80AKD
TC'-ii-'CSSEr V/ALU'Y AbTnOKlTT
T •:?••,•:£::>?£ E VALLt Y AuTnOklTY
IcX/iS Ijl Jl.lTUS
TEXAS liTlLlTlc-S
TEXAS UTILITIES
Tr.XAS UTll.iTlLS
TCXAS UTILITILS
MOHftVt i»0 1
MUMAVE tiO 2
MOnAVE NO 2
MAT 1 lESIjURo NO.l
IIATflLSaURG NO. 2
HA»KII«iTON NO 1
SOUUULST fJO i
SitAwNLE WO. 10
wi'JO.
-------
C--('.'»;,Y STATUS OF Fob STSTtl'-iiJ QUkIi-,o
FtW/E/i ST/sTIO^' Cl'KkENT MOM H
I.D. fiU.'-'LLh 1 Ih^fALLMl IO,N OF A FLUL G/\S UESULFUKlZATION SYSTEM. IS faEl.NG CONSIDERED.
A!.A(:A:-A_[..l.c;c.IL<.Lk_C^_c± I(" 'V
i>!G:;i.tl :..L y HK^
£;>b -.-...- j.tlv.... ._.
CO..L 0.;,- 1.5 JTKCLf.T SULFUR
f:iUT SCLLC TFr, •/; ''__ _ _ _ ...
Li'vti>ro:a st'>ui'bi.;b
_S_T ft i< T UP 1/20 '
I.r.. :-i-••:.::> ..,.2.._._ I.'-.ornLLATICu.OF. A . Fl.OL GAS L'C.SULH'UKIZ AT ION SYSTE.^ IS 5EING COW
AL/.b-.' r, i.L:C7.-lC COOi' " Ir A SYSTEM IS I i-jS T ALLEiJ, II »ILL P.-iObAbUY ts£ ThE «LT LlMESlOi\L S
«IO- :"L .'iO. ? ...
2ti> '•'/ - :,L'.-.'
CO At. U . (• - 1.5 JJj. i< C cf J S U._P Ui<
KOT "f.t'Lf C rt'i»
L i ,"tl>Tu\L _ '<(. :-Uij!iI:.'.o_ _ .._ _ _ _ _
STA'.Tuf- ' "l/7r '" "" " " ""
j.r:.' ..!;>;,.L;- ""$"" """ " ' ..--.—. - - _
A.!i •j.o.r_-';__LLt:_cj '-ic pQft.i.ij
A^'ALHL "it?"
i:')0 •-, - :,LW _ LETTER OF_jrjTE;-.T_TO_ INSTALL AN FGD SYSTEM WAS ISSUEO TO KESEAKCH COTTRCLL.
"CC'.L '('.'{,- u.;« VfK-LL'.T SOl_FJ:\'
KCSE '••;•£>• COTT-'ELL __
LI-'C>IC-:.L st"os:;r:»j " "" "
U-^'-1' .••/?'•• .
I . u. ! L '• L. i-1- _<4 __
"AiU/JO'iA LLFCT'ilC Fui.l -. LCTTcR OF ' 1UILMT "TO If-iSTALL ArJ FGD SYSTEM WttS ISSUED TO RESEARCH COTTRELL.
APtCMt JJL *,
'{ J ^ •:*.-•' L W
COAI 0.~- , . {» Pi,-CL?.T j'^Lf-oK _ _
.^.^•.-••CM \ cvf-CLL " "" ~" "~ " '
Ll'-LiTG'.: .;C.^'.;'a::.G
STi'.r'iui' *,/!•?
— " — — ~ — ~— — ~ — —— — — • — —— — — •-«. — — •fc,———••• — •» — ^ — «- — — — — — • — »•••••• — •• — — ^••.••—"•™ — — «• — » — •*• — »*•»• — ^ — »«"«»»»^»-i — ^•.•.•••*^«,^» — »^ — ^ — ,»»^ — -.^•.^•
I.U. .!i-f-.-.J.!< 5 ~ jOlii MODULES V.LKE If; SLRVlCL CUKI.^u THL CAST 3U UAYS, EXCEPT FOR O.ME DAY
ArflJy: (i f L'-I-JC _Sr^-ulf L___ ''Lrt -VJLULE: 1U COr\, ".;: i ' ' " "" "~ ~ "i7..LL;il luk "v/F""LiAr^^.L ^L.>1 L'i In:' Trie.' IULL I' (JUCT'TO EACH HLH£ATLR~Yb OAKPEM
ll'j '• .. - : timrii XT _ fh^ vi;-KAHO.\, is CO-jSi^CrtLu A >-->«TIAL SuCCi-SS oY AP6. h GU SYSTEM AVAIL-
CO.'.L. l.-«. ;--i.;-CC-:i .Si-i r'l/rf " «oiLiTY K/.S 93 Pt.r
-------
T;-.bL£ 2
------- cc- ;,.;.rr •• •• ------------------------ ~ STATUS DFTGIT SYSTEMS uURii.5~" io"/7<+
l;C;-:i< SUTIi-i "• ...... " ...... "" ........... ..... CURhEIjT riONTH '" • '• ------- '"
— '- I.P. • r;i!Pj-rp- ' " "& --------------- ~~~~ --- CC.'jTtfACT "TO"ir.:STALir FGD SYSTEM ~OI\rCHOLLA~'NCT~3~>iAS BEEN AWARDED "TO --- ~
AKIZl.i A f"j".LTC SEiUTCE REolVrcCh COriTTnELL .
C""«'CLL«". TvC;"2 .....
2i>;; ::,-. - ' cu
"COAL. 0 .-*;. ;•;:;• CET-.-T" SI'i r\ji-. ' ........... ~"~ ...... ' " " ~~" ~ ' '
i'.L i:.r!.Cn COT 1 'JU SGML PROPOSALS HAVE BEEN
"• ....... AKI^O'iA ! v.'-LTC SERVICE ------------- KECE1VLO ANO "ARE SiCING EVALUATED. ""' ........ ----------- ' ............. ---------- ......
C:;vi!_LA .'.0 3
c?DO !•;"- ..£>. • • ' ....... - "" .......... ...........
CUAj_ 0.^« !••• i. f.1 C L ; ; 7 SL \_F(jn
;-:oT~sn riirj ;
Lir.csroi-'i. SC'-;.IB,-H';G
STAHTUF (i/V<5 ' ~ ..... . . • •-- ..... -- •- - ..... - .......... - ..... ...... - ..... - -- - •- ..... ...... - ..
•• — — — — — "•— ^— — — — — — — — — — — — — — — — — ™— » — -. — — — — —— — — ~ — — — — — — — — • — — — —— — — •. — — — — — — •• — — — — — » — — — — — — -. — •.•« — — — — — •» — — ••«•••„• — ••• — — _»••
------ l.D. ••."IT-' -.(".» " & ....... • ...... • ............... ?L«..:5 i-CK $02 COfJTKOL" SYSTEM iKE IN TilF EAHLY STACiE." flPS IS CURRENTLY
;.i "t-.'-wL ...... """ ......... " "~ ..... " ~ .......... ........ ~ "' ................... "" ..... ••-•- -— ..... •••
00 ;•;».•! :>rLi cTti. _ _
"Li-.u-STfijt. ^.o-onSiU, ..... ..... " " " ' ...... " '" ' " ..... "" ..............................
S! A'HiJr f- /l~i
I.f. :i. " . •;••' 9 LXi'jflKu SCKuntillixi ic UPGRADED TO CCISF-CY ,,-ITH APPLICABLE 502 REGULATIOI-.S. "
(-GUK CCH ...f-S ,.(, 1
175 " , - i, i-.TKOi- IT ..... ~ ....... ~ ' ••• - - - - ..... - .......... - ........ .......
C',AI. Ci . ,-i f i •'CE^,l SLI riiri
CfLM-.r- ...... .....
_ L!.vi. .,1 !-.:..'! .1.
;>T,i<>7' ' * C: / V . - - . ...... ......
^ — """"^ *.-^ — — — w — — — ^**.*_B.W«._MH— ^_~«^«^<*v^^^««w^««V^^W^H^^.»B^*l^a>^^«v^^«>*v^^«B^^^^^^«»«v^^^^^«a^^^^^a
I.D. M;:-'. f> iu ......... ....... EXISTING 'SCKUIiflErtS REMOVE AtiOUT 30 PERCENT OF THE SO;:. "'THE VEuTUKIS WILL
_ A^I^C';A , u:'LTC oE^IU. BE JHOKAbEO TO COf-.PLY w!7ri APPLICABLE S02 REGULATIONS.
F OUR rCf
STArtf'.it'
-------
TABLL 2_
'C'Cvtt'A'f:? " STATUS OF h~G if'SYSTEMS u'UKii\!fci Iu/74
PUULK siAiioi " " --- CURRENT isofcfH
i'."b."fiU!:c"i:K" " ii EXISTINGscKUObfiiks REMUVE ABOUT so PERCENT OF THE sua. THE. VENTURIS WILL
/. I? I cON i'_rllJL'j-J^_§£.'lVlt.E bC_JJ£G*ADE.y TO CJhPLY _wl T_H_ .af.PL 1_C Aj4l^_Sp_2_R£GULAJ IONS.
F>~H'I< COK;\f.Ki; .40 3
22v _.(•••» .-..KtTrvQriT. . . .._ _ ... .. ._ -
COAL r.d PE'•».
Cl^-'.-_... P..
CHtHI CO
Ll^f. SC
STARTUP 4/77
1.P. >AVOi J3 LSH'i, ArtE USED FOH PAKTICULATF. Ei-.lSSION CONTROL.' FGu SYSTEM IS STI'LL IN
AK] 7.\s.;t- r\'.f\ ,][C__
f-HjK C (;:<.,;. "f. ;:(;
600. ..''.-. - hK
C<-'AL 0..} f'r^CEMT
NOT Sr.LLCTLli ....
LIivt/i. If't.S
S.TAHTlip (>/7J
UP. •!U*'..(.;< j<* iHt uOhPANY IS EVALUATING hLl EM-1AT1VE PROCESSES FOR S02 EMISSION CONTROL.
MASH, LLt'tlrii: " '" " ' "LI.itlSTOf.L SCRUbrUUG IS THE MOST L1KCLY PROCESS TO BE SELECTED. THE FINAL
HlSSOUt'I ^asti NO 1 DEClilOfJ WILL BE ..MAUt IN A FEW MONTHS...._... ._
bbu f'w - ;;c_w
C')AL
LLfcTri.
lOr.t ^C;«vJl«i>ri..b_
ip ' ' b/T3 " " " " " "" " ""
i.l). i.i.rLtli, ~15 ""' THel COnPAlMY IS EvALUATliJG ALTERNATIVE PKOCESSES FOK SOg" EMISSION CONTROL."
tt A 5- nv_i: L LCTPTC L_I rt E S_TO I'JE_S CKU_B3lNG Is THE MO ST LIK E L Y PROCESS JT03E SELECT ED. THE FINAL
~HASi^ HO * DEC I SI ON WILL UC MA~DE IN A f-EW MONTHS".
_~ IlElr' _
"" " " "" " ' ----- - •-- - — -•-
MOT StLfLTfl^ _
"LlNtSTGNc SC'' "" """ " "~ "" "" "
STAKTUP 6/f.O
-------
TAULL 2
TTJMFrriTf SIM USTIJFT GIT S YSTePS~DUR IT-it—1077t*'
PG.JIK STnTltiVT" CUHrtENT MONTH
l.D."':U-'-;:VE.~'< 16__— THL COiM,(l;,jy i S ~E" VALUAT i.jl, ALT Ek.MAT 1VE HuOCESSLS FC,5 S02 'EMISSION CONTROL."
iV.SU: LLLCTr TC Ll^L^TOl-Jt SCKUbUlfJo IS THE KGST LIKLLY PKOCLSS TO BE SELECTED. THE FINAL
vIK.?T51jSr"TfiirrirTTO~3 LCLlSJCOK"~UTLL"riL~>!'J'f <¥-(•;_ ' ~ ------ - - - -.-- -
STrtnlUf _ r/l?_i _____ _ __
I.C. iM^t-cH 17 1-t.U SYSTEM IS uOtaig WllH MO I fit- Ll;I Al F. PLANS F-"OK KESTAHTlivG UtCAuSE i-PA
i-iOSTuf. TOiKO:.' FUi-:i)il»_ t'YSllr fvC fc COi-.t'ArjY. THL uTiLITY IS EVivLUj\T If<(, THL OftT,\ LOLLECTLD uURI.-JG i HE OclNON-
i?o i**" -'^rfft'oFiT "" ""STKAfn>iJ"i'L"riiOij" to otTtKMUvL THL "COUKSL OF PUTUKE ACTION.
UIL ^. j |'r"!_Irly __ _ _
Sff.n I'..;- 14/7,1:
"I.L. 'ul.; ;.ijv ' "ifi "~ lh^' ui-OIT' lS""f-jC)'*"u-jJjLK 'cUJgSTKUCTlON A NO IS SCHtUULtlO FOR COMPLETION IN
Ct-MKf'L i L L 1 -.. 0 IS _L i_b_H T__ C t. .MA'1 CnPftClTY IS HDQJ-1W, BUT FLuE GAS HR_OH OMLY IQQ MW
"T)'-.rC^" (.'^"("i A i <:.).! ' '""" " W'iuL LiL TKtiJT'Ei'D "lui'T I ALLY"." " " "" ' "
1UU f-',A - -,•:.,:
C"
•<]
O
STAK^UP 3/7_o
_ I.L. l-l-:.••!.(• 19 _ bluS hAVL atlLN TAKLN_ ON LIKE AMD LIMESTONE SCKUbBEKS AND AKE NOW BEING
C !.'••-T'; A L iLL'l.A. F.S "UMI'I'C" "liUhV " " LVAL'UAl CO .' "" " " "
_ .,L*TO', ;:(. . 1 _ _
t--j-i >.. - • r: v " "" ' ' " " " "
C.I/H. .?»c-j.£ r'F. i\C_Ci -1 o t i L F L' K
:•'!•! bt Li't I EC
_ __ Lii'-'E/Llf-f STl/-1 J .SC'IoiiM'-JO
STAtiTUI-' ' 5/7? "" " ~ "" ~ " " ""
• — — »— — — —-— — — — — • — — — — — — — __ — — ^ — — »*•• — • — — — — — _»« — — — — — — — — — — ^•___ — ^™__ — •__« — _«__,_««_«_««^.».«*a™— _••«. — ^_»^«»«.*,«^^www — ^^»
I.L'. '.U'-Lff ?u " ~ THL. CO,"'.PM.NY is'PKESENTL'Y ObiAiNi.Mb co^siKUcTior* 'AND OPE» Ajj Fo_u SYSTEM.
'ii/iS'T ":-L:'iTiVi." i" "
L> ij u ." • .< — "• t. 'A
COAL 0.'.'- >';.* PLKCL';T' " *"" "
f-.OT SLLFCTfu _ _
NOT ' "" "~
STAKT'.JP
-------
TABLE. 2
lMY STATUS U> I-~GD SY"Sft>|5>
powti'i* 'STATIC^ ....... ..... " ...... c'UHhLwf >idftifH ..... "" ..... " " ............ "
I.U. NUi'-I.tlK 2~1 ......... ......... " " " ....... .......... ~
C1N.CIKN/>T I __Q,\S ANf) ELECTRIC ___________________________________________
ivilAlVl ~FO~hr NO H
boo i";w_- S.EW. ______ _ ____ _ ____ _ _____
COAL 3.^"~PE'.•&
ST/iRLUP _ 1/12. ________________________________________ ________________ . ____ ...... _____ ..... ___ ___
I.Ll. UUf-'.oEK ??. ________ PLmlT HAS 'JtU-J SHUT DU*f4 SIUCE SEPT J9TH BLCAUSf. OF A BOILLR I 0 FAN NO-
ClTY OF Kt.Y -tlsT 1GK FAILURE. THtY HOPL TO i)c OPERATIONAL THE WEEK OF OCTOBER ^IST. MCCH-
K£Y Wt.ST POl-fH f'Ltv/.J _____________________ AracAL rAlLUKES IH THf. FCi^ SY-STC'1 i«tKL FIXLJ.FGD OPE'^S10:-,^ SCn.i£->i;j&...___ ___________________ ....... ___ ...... ._ _____
STAi..f.B p3 ™ AS i-;ITh h,-,yUEU 1 ^ "2 HOIH LiMl-STOiML A' Sr^nO' f-'O.) WI1n A»-I'|UM. yO HEKCt.uI EFf . UlLL hf. Kt'GUlrtED 10 MELT THE 1^70 COLORADO
H'JI. .'.-, - ii. ... __ ..... ____________________ __ i>0^ Lv'l^ilCn-l. ! CF '. I'jOPP.'i. i'.ilTH SLuOGL OlSPOSAL IN A MIME SITE.
,/iOT Sf.LLLTL'o ... . __.. ______ . _____________ ....... ........ ____________
Llt-'ilbTOIIt SCRUCtJl'IU
STAiUM.r ____ b//f! ____________ . ..... __________________ ..... ....... _________ . __ ____ _. ______ ..... _ ___ __
— - --- -— -- -....-.._.... -- _________••_..__ — __.._.-..__........__ — ____«____ __.__....__...._-.__. — —-. — — ««.« — « — _ — — -.^_»»_,. — ».^ — »_w»»_»w».» — •«
.1.1!. .,!.;, ., •-.;< £n ______________________ ..... flS viill. h/.YJE'i \ I S bUTH LlKLSTOIJL A'iU bOulUrl CAKOON.VTE SYSTEMS Arffl ijEING
f •JLO.--.,-:'.:; 'Ti. iliCTnlC LVMLJ.iTLO. USl/.-ti fl.E AiiTIt.iPAtf.L. U.45 PLKCLMT SULFuK COAL AN FbD SYSTEM
Ci-.-iK- ST;;TIO: i.-C.,, ..... ......... _____________ v.if-1 APHKOX. 70 Ht.hCt.Nl EKF..WILL bC RFUUIi( »'Cf .Hj;_^H_FuP ____ SL^'NL' J>C(• 2r'j CGLU.vAUU UTL Is PKvjCLLJIUG w!Tri L i%; G I f-; L f >. 1 lid EVALUATION OF oOTH LIMC-
CvLt;.-./ir^ jr>_rLt.CThj(.__ ___ sjo--tu. £_i;oi;_iu.s_c,v!
-------
l.AoLL i
ur F'uL"SYSTcf"iS"bUKl.Mb iU/7<4
"I.U. TjUi'ift.rt ' ?fa " CGi_OKAUU OIL IS HI-UCLl" LiliMti ,-. ITH LN& 1 i-^l T\ I ivo EVALUATION OF 1>OTH LIMC-
COLOrtAUO JTL LlCCTklC STo.Jc. & SODIUM CAKbCf«AT L SYSTEMS. UblijO THL PRESENT 0.<*b PERCENT
~H't\\ uLT: "*iC~;> SULruii CUAL AVFbu U.MT"w'lTrt" APMC*. 7"o PEnCE.i>il c.FFy~'*GULD •}£ "i\
_ 2bo__ (••-.« - RETROFIT TO -U;LT The; i9?a_coLCKAijO soa EMISSION KEGS OF isppfVi. IT is
COAL OiHb P£-RCLNT"SUi.FUR PHuHrtHLE" THAT A "LI ;S£:STOi\iE""SCRUdbEr< vJlLL'bt CHOSt'N""wn ri DISPOSAL""
~scr>UHbrMG "~ "
1.0. r;lif:t,c.K ?;7 U.o.P. IS PntSKWTLY ^ THE LNGIMELh i.'JG DLSIGh! AUD PHOCUKtHEfJT STAGE.
CJLJi'-'^LS 4" 't>i'in M fill", OHIO "CONSTRUCT IO.'M lS~ SChEDuLEO lu START :-1A«CH 1975. " "
J.yM-.'SVIl.Lt. ,,0 b
CuAL
OR PHOn
i/7'« " "~ " ' " " "~ "
"i.L. 'f.'U,--oL'n " ?i) ViOixK Oij" UuIT t^O.'b HAS bbL'N ULLAYE.C Fuk AbOUl A YEAR, CONSTRUCTION IS ;^0w
LOLUFj-l'S ;% S.luTilCRlv OHIO SC'iCUULEL. Tu STfthT JANUARY 197?. U.4IT NOW ScHtOULEO FOR COMPLETION IN
37b f.,. - NLU
COAL "" .•••-.-• - - - -•— - •— - . - -.- ^
OR P"0;\uCT_S
.1.0. fiUf.L.Lt) ?9 hOu.JLL A WAS UOWN 3 TIt-;|£S LAST hONTiU Oi^Lc TO CLEAN DEPOSITS FROM THE.
coMofjui/;'_TH"I'ui'smj" " "VEUTURI THROAT AND IUKC'DUL TO BOILER RELA^O THOUHLLS. UCMISTER conoi-
WILL. COU'.TY uO 1 Tlu.l CjETL[UOR(VTEu SLIGHTLY A^U WILL HAVE. TO BL HAND WASHED DURING SCHEO-
Ib7""f.^. - »L"51'OFl'f" UL-t_i, Ul.iL£R OUTAbt.. OP^nAliOf. MOiT Of THL ••'iC.jTH '«AS ^ITri A ,vifL,Iu|u; SULFUR
COAL 0_.fj-o.y l-'E.KLCl-1 SULFuR CO,\L liLHJ-Ju._ AVA1L<\;3_IH f Y «AS ft5 PLuCEi-T. fiUUuLL_B IS STILL UOvJN,
t'At^CuC'* (. '.- It CO
LIMLbTGf-L SCi-.'Jr-t Ii'-ly _
STARTUP 2/-12
I.D. fTuMTjLR ~3"o" " "'LXHL«If-iLNTAL WfT LI^L FUHWACE INJECTION SYblEM WITH WO TAIL ENL SCRUBBING
sAiKYI.Aj_ii. pfi',L\> _ctipjp is user,. OUTSTANDING HKOBLEHS ARC. FL-KMACE TUL»E PLUOGAGL WHICH CJECESSI-_
60 f:A - ,.Zli;OilT _ _ KErtJ'v'AL LFKiClEi'-iCY "AS FRO.-; 25 TO 4J PLRCE'^T. THE f-G J SYSTEM OID .MOT KL/N
MOiNt: TH/.N T.-.O DAYS AT A Tlr;c. AVAILABILITY SIrjCE STARTUP NLVEh EXCEEDED
_ _ 25
LIML INJLCTIO-I
STARTUP f,/7l
-------
T/.-JLE
"sf';»fus "OF" Fob
/74
POwL;' STATlfni
I.U. MUf-MER *1 " IMSTALLAT lOiJ OF 1 HE HGU SYS f El* IS LSSLN TI ALLY COMPLETE. A FAULTY iNSTRU-
()LT'!On_ L.._! MLMT I'A'-ILL .•.HIGH v^\J, 1 :--ER~ 15 TH."
;-.v., - ;-LV ?* Fuu oYjTL;-: LUuSjiTS Of 1 T/.0 STAGL i\ljD J OijL STAGE VEi'jTURi SCKU'iBEr^S. ' AT
;:.r: it S.'• • LM-:! r-hcSLM, y.'LY FLUE <3AS Fi-'O'-: tiOILERS 2.3 A-.D 4 (AoOUT HO-1* OF THt. STATION
, -ii:.i 1}-S " " ~ Cia-ACITYJ" ,\r.L I!-.' SLKVICE dLCAUSC t-LY .^h fKCM INEFFICIENT ESP • S OVERLOA'JS
slO .-', - ,.-LT'<0! i"i _ Tii;. cLfiMK ii'RS. OPLRAfl'-Ju hOL'Hii FOR MOuUL£..S 1 THROUGH H bt.T/- ?.ft ' f'i.r-tl.f-iT bUi.FUfi Af-ij 0/30/74 lr.;LKE 17bfa» 7o.', TC.L t-Gl, i,YiTt'.r': n«S i.)Lt..-J uour.1 i>I;.AS UAiSAGEO bY SOLIDS
.-icj ••••,. - ••«£. TROT IT BUiLuUp ii< THL GLAR box. KLPAIKS TO T(,t. CLARIFLER AS ^ELL AS TO OTHER
C!>iH. ir_.^ iT '<•(. f.}'l _S_^L£UI'. rtlivuii HRbliLt.i-'.S To _1_HE SYSTLvi ^ILL (jr. HAuE UUH1N6_THIS SHUTDOWN, THE FGD_
V:;CH~""" " "" SY-iTL.-:" Ii> i-.oh!-jit;G "A'T HALF CAPACITY (Si.'.Ci- i-iAKCn) DUE TO" LOlr. HE'AT OEHAl'.'DS.
— — — -'•"~'1-- — — — — — — — — — — — — — — — — — — —_— — — — «*• — « » ™ — — — — — „ — — — — — -. _ — — — -. — _.„ _ — — wv. *-MW_WWMf.«KV«H«.«M>VWWMW_«>«w_ «WWM«.*V.*.W _••« • — •. v a.v« •. •
I.U. !•)(.::•• ;..r.(< 20 " "" FIVE HROPOS^LS FOR FGO SYSli.r-,S HAVE BEElj RECEIVED AND A DECISION WILL BE.
~HU'-U.~C~I TY '. r.."$ DEbJLr'uHi^ATio.'v' IS ALSO BE~Ii';6
e>'jO_ f .- - r t ;
Cf'i'
-------
TABLE. _2.
STATUS Of Ft>U SYSTEMS DURING l'J/7t
t'OrtL'K Sl/,TJn;-i CURRENT MONTH
i'.rf.'TiuYiTLK"'~~i'f> THE" Viiu u7uT"HAS s_£!;>£ Mi I?11- !••::_- i'ETrtCf IT_ MA JUr<_KLPAlRS CONSIST OF K£.li\iF OKCLnLMT Of- THE STRUCTURAL STEEL SUPPORT
COAL 2,'i- 3.? "p'i'hCLNTS.UI.KUR TO THE! HIGH TEMPERATURE STAINLESS STEEL OUCTlgOnK AND fttPAlK OF THE ACIO
SONS Ay "1.0 ..f.'-'VllU) Ci,iLM...SY_STLf-iS COULER.. UAFFLES...... JHE _uNI_T IS SCHtpULEO . Tp KLSTART BY i\iOVEMoER_ 22...197H.
_. ...... * _ _ . «. — _»—_--- — -.—- — - — --- ^ - _._«,_«__.^»_« — -. — — — .«»«. — — — — — • — — — — — — — »- — — — • — — — — «» — — — - — — — ^- — —
i.C.i. r:ur.,(_)<•_ 37 IWjI_AiJAPOLiS POwLI< Af«D_UIGl^r HAS f,LT WITH SOME! VEIVOOKS IN AN EFFOKT TO
l^LilA-V.l'(.L!s"''CA.'F.'< A^ui LIGHT OLilAiN MORL "COMPLLTE BIDS. "THLY" AKt" f»iO^ "AWAITlfJG REPLIES.
1-0 ._3 _..._____ ...
- -iL'W
LOf'.L _3. l"-3_._fj_Pl^!_4,_ I 1GH! bLR iSJ . pUKIMG UiJS PLRIUIJ THE bUlLEtUi••, - ,-(i.T:'<0! IT _ UiMlTa l.'EKL t:OWIi DUKl:it THtl SlKIht. TH£ F&U UI-J i T KflMAKTCD OCTOBER 1ST
-o C'AL I .i,- «,'.0 fTKCi ~1 S'L-I Fo!<"~" o:\,u iiAb btLi« i(;«Oi5 i
ST."1/'••'.'.. W/7v:_... ; _
^«~ — »—" — — •••' — _ — •»•• — — — — ^^™^™_ — — ™.— • — — ,• — — — » — —^«»^_™«..— ^«_,— « — »..»™ — — — —„—.».„ — -- — ^«.^^ — «. — ^ — »-w — ^».«, — ™»* — * — ••—•» — ^** — ^ — ^^«-«^ — «-"—' — ^
I.L-. :.Uf.r,t i< ^ _ KCl'L hAU A L;\fiC/!< S1HIKL AT THE PL/iNT x'HICH LASTED Fi<0f-'. JUL t-,TH TILL OCT
* .•,'.';,,«• C 1 T Y' -JJi-f. r<' e. "LlMil "" '"" " "1ST. " bUKI.Vti THIS PLKIoi; THE UOILEKS '.Vf.KE OPEK/ifEO dY TM£ (.WAiM ' S SU-
_ n.'./rnor; > :- u _ pui E l< / J L D ;. IT H A.r.UUT BU PEKCENT " AVAlLAiilLl T Y"
f-20 v. - ' ;.* _ _ LH;i'AJOK OUTbTAKUli^ PRObLEMS. iiOWLVE.K, PHE-
CuAL b.^- r-fi-CE'iT SuLFUK "" "" " VE.-liVf. MAl^T EI-s/n'.CL. PRACTICE KLuUlKLS TM/lI ONE MODULE OE TAr^EfM COWW Oi'JCE
OAhCOCh A *IlCOX _ _A ir.HEK FOR 6 IIOUKS OF LLEArtJkG A'^U WASHING OF SOLIDi bUILiJUi>,
LlHe.STC:'it. S' "
b/73
-------
JAI.'LL 2
"CO riP AT-Y " STATl/i O'i- Kii> SYSTtrli
f-OivLli S'JaTK'l ~ CUKiHtNT MONTH
~ l.D. NUMBER """»»£TENTATIVE AGREEMENT HAS UELlM kL'ACHtO WITH CCKbUSTIO.vl ENGINEERING TO DO
KANSAS PC/.VFK ^ LIGHT PKLLjn\INjuRY_Fj7jj_j;YST^^^^
"OF FR.ivV i.u. V wouh," OU~T'HE Rib "sYsYLr-~~AKTOP o/7_o , .
I.D. n\Jt-i>r.n nZ _TE.\iTATIVt AGREE.^Eix.T HAS BELN KL,\CHEO WITH CUMhiUSTION ENGINEERING TO DO
""KA.r'.sas'PiVEf-~A LIGHT"" " " HKL-LIIII^AKY FGD svsi Lr, "OESIG^ FOR T^IS UIVIT AS PAKT OF THEIR CONTINUED
_ Ji.'Ff:LHY ;."). ? W.UK11 9'^ TML FGU SYSTEM rij_ THE LAWHf.'.'CE PLANT_.
7uo ;-.'. -Y.r.V "
"""SIAKTUP f,/yj
i.i... ,iUrr-!:H ""43 " ~ "L-OM i-'oi) rtOiie'-.ris',-ific 'i'^' OPLIUTION xiTtt "ESSENTIALLY 100 PERCENT AVAIL-'
I ,i-.:»i.;.u: '":o"~ i-.LLr.s."~ >"A.-J^<-U' ci.V/jjijp""of:"~KL~rit".AtEK "coiL'Ts'Now DUE' AFTER
Ui C'jAu o.r> r-'i_>H i.'.'rt Si/Li-Ui',
Cf'!'.i)U.ST J ».•••• E:ooIi:tC
5j.r»VvH.'P....j2/£,'.
!.h. , V.',,>.:• (.4 hCiLLK t> M/.i« btf.X uf'EKAT iftl'j Cr. C0,\!_ Sli/Ct OCFOuER l^TH. POOR GAS DISTRI-.
" K/-.f S.;'-: '••..,.:.,-. ^. LL.iiT " " ;,.l/[Iw\i ThiiOOGJi £.«(.H U.'-c. OF THE 6 .^ojULLi AS KELL AS «,''iOiviG ThE riOUJLh'S IS
L'.-.'u ..I'.i .i.i ', __ ST^Lu A Pr'Oui. £,'•;. COf;uuS f Iu,-j ENGINEER If'iG MAS COMPLETED GAS DISTRIBUTION
uOu '•'* - •:!••• IE^TS /jjo is CU!'.;*E>«TLY ^OOIKYL-JG TP..?Y.STLh'^. jJjJt.Ji.All.OJGJ'ERIENCED IN THE
'&'> :i~o lJE.'Li"v£.KY OF"sT"KULTun"AL""stEi;C.""" t'Of-JSTHUC TlON' stAHTED A06uT"SEPTEMBER' 1ST
o4 f i. - KETM>riT _. rtNy COMPLET ICfl. IS UUE lit MAHCH i97'j. _.. ..
f.OJiL i..-) t'fi'CL'-JT SlllfuK
-------
JAOLt if
S"bV"^i»"u"SYiiT
f-0«L'K Sl.tTIcii '" """ "CUKKENT MONTH --
1.0. "f:Ur iiili'uti' A 'uOr:r'LI/Ujf ~ ,'j \ ~ niv t-'uu M Si :.iv>.
110 :•:•* - KLlROfriT .._...
COM.
NO I
L"lME
STAUTUP S/JjO
MVW_«___.._V..V_..4V__«____V..V_ »••„ — »» — — — »^^«...— •—•»— — »w»»»»^» — •••••»»••••»»«»••»••••»••••— ""•••••••"••••••"-""^•"^""•^""•••"••"••"••"'""••^
j.n. ; b::; t:iv n't A tuhHLiANCt: scnLuuLt HAS BLCN subr-'jTT£;o TO THE JEFPLKSUN COUNTY AIR
Li'l'l.>V ! L.Lf.' o'.."> "i ' (.LLCTjJ'l'C" " POL-XuTloiv CONTKOu 'UlSTrcICT 'wlTH 6/6U "tISTAliLlSifLu AS THE STAkTUH 'DATc FOK'"
C^.-vL f.'i;.' •••:; ? Ai; >"i>0 SYS1CF-. __
107 ;-;•; - 'rxii'Tft'OKii ~ " "" """ ~' "" " " " ""
COAL 5.\->- ii . 0 PEKCLIvT SUL^UK
j r
>niijG
1S/-.C ~—- — --• - -- -- '-
I". D. ;'uv:-:.'. •••'«,{> A'C-Jf-iPLlANCtT'SCIiELiuLL" 'ri«S"b"-"t;W SUGMTTED ' TO THE "JEFFLRSON COUNTY AIR POLL
-: br.6 «; r.LLi.iiiic UTIO,M COI\TKOL UISTKICT »iTn &/ao tisTAiiLiSHED AS THE STAKTUP DATE FOR AN
TaTT
- HLT«0!-lT
.ri-"<»". o Pf KCCF,
•ICT J>fl.t(.Trii __ _
L1:VK Sc'-L1 u-.iViu "~" "
STf.iUL.f-1 ^-/J-^
I.D. i.uNt-r.ry u9 LOUISVILLE &AS AIJD ELLCTKIC HAS JUST RECEIVED A CERTIFICATE OF CONVEM-
LOaiS\/IlU: GflS I'fLECTUIC" Ei-JuC'FF^Oh THE PUBLIC SLRVIttl COMMISSION" TO CONSTRUCT THE tlOlLER. APPHO.
CA'a t«U.'. .-^ u VAL OF TH£. COMPLIANCE SCHEDULE FOU THE INSTALLATION OF THE FGD SYSTEM IS
1 7tf '• , - i:LTr.''i>'<'i(v- ""nff."Vil"rr.;'; "
LK-L SC»riT "FOR "CONSTRUCTION OF "THE "FGD' SYSTEM "WAS DISAPPROVED BY THE KEN-
LOUISVILLL l.-AS ^ f.LLCTKlLTUCKY PUBLIC SERVICE CUHrtlSSION. HOWEVER. THE ISSUE HAS NOT BLEN FINALLY
Gi-iT
CUAL " O. T;-w.Lib" '' " " "
r;or SELLCTCU
LIME YCHLbLUufr
STAHTUH fc-/7b
-------
TAEJLi- 2
_
~Cb"iVP/U~Y ' ' sTA"fuS"oF~ FCu SYSTEMS jUKII-.t. iu/7<
R SATION . CURRENT NONUI
51 A r-dKMlT >OR COSTRUCTION uF TuL FfcO SYSTEM WAS DlSAI-PROVEU BY TnE
Yciy..Hut-.Lic_.stLCt>T SU
"Tjof'SLLEl.TEU
Ll^f. SCr'L',(r Jr-ft .........
I.L. > U.' •, T.r< ..... 53 ..... .......... " "" A COi-iPLlANCt: SCMt-GuLt HAS lidEt-i SUbl'iiTTEiO TO THC OEFFLHSON COUNTY AIR POLL-
LUOISVlLir G/,S S. ..FLL_LT_M_IC _ UT10,-j CO^TKuL, ulbTKlCT wjjjj fe/7a__tSTABL ISHLD flS THE ..SjARTJJP DATE FOR ftN
MILL "CHELK iiO ? - - p^ gYSTtK.
33C r/. ..- ..kd'ThOJ: IT _____________________________ ....... __ ____________________ .......... __ _____ .....
COAL ;s.o- q.C fi.KCt.tjT ibLFU'<
fiOT StLLClEL . _____ ______ . ..... _ .. . ______________ _ ... •
LI H;_ SCr.,..:;;,i;-,:.
1.0. ?JU'-..tt_' ___ bt« _ _ _ _ ______ CUi-jTKACT wAS /UAKUED To AMLKICAh AlK FlLTEK. SYSTEM WILL OPERATE CLOSED
LOLlSVILLd u'fkS i' H.LC'ihlC" " " LOoi'" 01-, WAiLK AfJD SLUUbE WILL UE ST A
MILL CHtLK l-id 3
^b •••... - HdT^o'm
__ j . '->- u_. o ,r'r^C[ "T SULFUR
IC«iAi ;>Ii< ML I CM
I.U. ••jL.;-.:.-EH""5i ........... A Lr)>,f'LlAfjCE "'SCHEDULE. HAS HEEN SUbMITTED TO THL JEFFERSON" COUNTY AIR POLL
L 0 L' I !a V I L I . j. uAb J. J^LL C T K j L ___ UT10:>I CONTKOL OlSlHICT V.I_TH &/79 _AS JHE_STARTuP DATE FOK AN FGD SYSEM.
"'" ''' ' "~ " ...... " ' "
_
COAL 3.i>- t+.O PLKLcl:T SULFui<
IJC)T SfLt'CTEn
STARTUP fe/79
-------
T«BLL~ 2
STffTUS''OF~TCVO' SYSTlT-iS "SiOFiTwG"" I077V
POKE" 57. \TIoD ----------- ..... -------------- ................ "• ~ CURREwT f-,OWTli
'I. P. <'l.',"in.t( ----- So ------------------- F-'GO'SYSTL.N OPERATING SAT ISf- ACTuF-'uO"PCRc.E1vr"SlJGnDR ----- Tri£T rXrSri.VG' ROTtt^Y' VACuoil FILTERS' A'lTH~iVO fiLIfVOlrtG^CF'FiLTcIR CLOTH." '
L i«
STAKTOP 4/73
1.0. .H.'^-uP ^7 h'tu SYSTLi-1 UNDcR CONS! KUCT 10M *1TH HO MAJOR DELAYS. ANTICIPATE COMPLET-
'.xo ;T/.:;A" p;^r;r i^G' COKST RUCTION "Of; SCFIEDULCT
COLSTP.Jr' .if. 1 ____ _________ _ _____ _ ________________
ObC) vh - "f.i-.^" ..... " - - -• • - . _ _ _
('.{' f-'c1 i^CL.'JT SLLFI'K
C'0r j-iuST ! uMTT.'J f PT~7TS?OtTA~r£S
I.r. :io;,:-::rt " '!>:• ...... "~" "F-SJ" SYSTth" Ur-;CCir"COijSTrrLrc.'Efi _ _
.'*h"\ bCLtC.TFU
STAKTUt; 0/79
-------
TABLL 2
C'C^'PATY STATUS UH""Fob SYSTh>;S
"POWL'K" STATION"" ~" CURME.NT' MONTH
I.b. liUr.i-LK 61 FGu JYSfL,'! PL'KHOKKMICL IS SATISFACTORY Will) MO OUTSTANDING MECHANICAL OH
'jrfLyADjL lJl'.iJf:j< ?KUCL>S PKUoLE'VS 1U HtPutU, HOiwlVl '<, THE. PLA^l. IS. LXHEn I LUC I';'O /•,'<< If" ACE.-..
Vs/ib" <.-A';i>,t K ;;c i " UUHTL but'FLY or TKOKA tsouiuf'. CA^IJO^ATE. ot /.HICH KLSULTL.L) IN FKLUJEMT
12t> .'••*_- i-'ETKOFJ T IMt_i IN 1>L OPLKATiO.Ni Of 1 hC FG(j SYSTEM, NtVAcA PCViER IS SUIf-iG
CCAI. "'o.ij-~l .0~ Pt'.HCLivT b'Jl FUk " THt. i,lJPPLIc.i< FOK bKL'ACH OF COfgnflT ISF AC-1 OKLY l»ITH !-JU OUTS7/\NOINo KL'
'"••[ V;-,LiA t-'C-^ri." " PKCCi.-.bS H'OMLE'-.o TU K:.Pu:U. Hi;.-I-V'"L J-^'-^LY of I;;O:IA (b^jiuc- c,i!i/PP.LIE1K FOK HKr.ACji uF COMkACT AC40 ALSQ_JS KRLbfcNTLY TLsTlMGL _SO_DIUPI
coi'-'fiusTib.i~f:r,uip. rtSM-ciAitS CAKGONATE"CU'!u; ,,-.".i\ 6i
:v(, V /(f.A. (."i' ; H FG0 SYSTt^i ISUt-DLH ..CONSTRUCT IUii,
'KL u or,•<,,.;; f< .0 3
I'f.^- •>-.-< - KuTKOK 11 _ _.. . _._ .. _ _ .., __
COAL 0.;>- 1.0 PI. rtCt. UT 5ULl-Ul?
CO/-:)US.TH,;, t'^'il'.. /\S!?Oc.lA,Tf.S _ _._.. _
SOC'-IUf: C;.«! O-.-.-.Tt SLKUl uliMG
b"l .".1'TiM' r,/_y j
1.0. t;l'f|.:.>; r'« _ la.VAjA PUViEK COMPANY HAS SIOt.EO A LETTLH Of INTENT n IT H CO^HUSTION ECJU1P-
" ^ov.rx "~ nc^r ASSOCIATLS FOK Tut coKbTKucTiou ue AU FGU SYSTEM OH RLID ^AHDIMEH NO.
,«l.;jt.i! 'iO 4 >*. rnio UI/IT is SCHEDULED FOP< COHPLLTIOW. ifj iy?6.
i-.- - ,.<-,j .... . ....
C'/AI. Li-'.)- J . f. _Hr .'.•.CI.i^^S-.LaFlJn.
Cl/;iM'.CI Al L^
SCi'.UII- C,.-<..-..ATE SCKuerli;^
S I AK I IJK fe/7-
I.D. iH!:'hlH bl> UKlVLHSAL OIL HKODbCTS IS CO!-jDUCT IUG PKELIMlNnKY DESIGN wOkK Or, A QHY FGO
MLW_C.r C-LA':r- H.fC_SXS_T_Ej!. UMT_W1TH SULFUR HtCOVi-KY. M£CIEJOrj_TO__bE_ MA_DL_AROUivp_f;,0_V_ 1 WHETHER TO
iiKivY rd'i torn "?:o'. i " " " s'u.";-i AN trJoiNu'tniv^u CO.«TKACT".
f-bo -., - \LT;
-------
TAbLEl 2
C07TF7.K7 STTTTDS"•UF~GG~'SYSTL>TS~unR71J5~' i0/75
'FOV.TR STi-.Tfir-'J " ' " ~" " ' CUkRLuT fiONTil
f.C'. r;i-I"C; Uv;I7.
:;GK1,uM. I.MJii-. - l.'LrhOFjT 4^1 _
C'GM. '6"PL'KCt.iTr'SuCFTJIT" " ' "" "" " "" "'
HOT bf LLLTtt'i _ _ ___ _
U'CTF.l' " "" ' "" "~'~ "' " '"' "" ' ~ "" ~"
- U/_0
«. — — — — — — «.*. — _«.„•._ — »„«•.«»«,• „ _ — fc—, — *. — « »*.«._«.n<-~ *. _ — — — — «.»•>•> — «.«__«.«. — »^»^_ — — __^ — _^_ — — — — — — — » — — — — » — — —
1.0. (ilM',u":i 67 COl.SiOLuIiVG A LI fit. OK LlMEblONC SC.^OBljI^G UNIT.
XOPTI.ER.-: I.Mi'iAfM TuLTSfRTIce ALiU'WAITING' FOR CLRFOR.'lAXCt. OF UtLUMAf-i LORC. UNIT
tjAlLY NO.,, ONutK c^NiSTKUcT 1 OK A^T^Tj-lt-IH D«H. MITCMLLL NO.11
(^jij -^^ " " " "
C^AL ^ 'll_l
VjT"i,E'L"rC"f'
_.\'.)1 Sr.LLCt
M>.i:/.. - nCIrifjFjT
00 " CuAL ' ^.^-"i';'b'>ui--'C[TV""SULFUK " "" """ "" -••" •--•-.
° OAVY rtii-L'n-AK __
" '«!•'"""
e-ATj^
I.O. tlJ'V.U* (••? Kfcij bYsTLX UUULH COMsT»'"f»(iw"Li'.: SCivUu-jfll^S .vLRL f'L;.CHG 1,, EA^t-V 19 7j.. iXCCtWT SCi;L ELtC'iUCAL
C'^.L J.ij > ? >'Ci.:;T__Suf. FUH .. _
"CO.v."r~iii,'T I'U.TL":",", !:"!?>».• ii'sC" "" " " "" " "
Lnvf.STCft. SC^'JbbliL-
STAHTtJP _/7o " " " "
— — — ~*+ — ~~ — — f— ~ — *——. — — ~~ — ^~— ~ff.^ — ^— ~~— ^.^^^~~^~~.^^~.~ — ~~~^M.~ — — ^~~m.m.^^~m.^~~*,^.m.~~^~.^— ~ — — — ~— ~<~— — im — ~~ — ~,— —
1.0. IU"i L>? '711'"'" f-u^' JYSTLi*. uWCE:< CO.'ib TnL/CT lOfC wITn ,jO |--AuCi< DLLAYS.
.v.iiTAJ[l_SJ\V,.Lj< SCuUriiit-lS »Lt\L PL;,Ct'U Iu E^I^LY iy?i. KECLiviT SCHUuBtk bESIGiJ CHAIvGES HAVE_
'S.^i.ki-'C^i'-"''; i! ^> " ~" ~CAJs't-"d"c.LUlPi'i'Eiv.T ~L,Li.IVL::
-------
COMPANY STATUS OF"FGU""sfs'l LMS DURIiVs l~6/7<»
POv.LH ST,VT"i"f,v.i' ~~ CUKKLNT C-IOMTH
~I .[). ' ?iL'f:<:E.f> 71 FGo i.YSrtM UNDER CGf jSTrtUCT i ON , bUTENCOUNTERING SIX MONTH UELAY IN STRUC-"
P{ 'JJ'SJLV /;•<)! n PDWEK _CQ. iy*'>LJ2TLE^ A.^^^v-s%yr_'^.^iJL!=iil_'
biuJCE V.i'rsFirLp MC. i """" """ "
p.^O M,. - 'jLv. _ _ __....
COAL" 4.J PEKCENT SULFUK " """ ' "
CHCiMcO . _ . ..._
LIME scKiiHtut'G
P./.7.5 _ ___
,U. .':Uy.'EK 72 FGu_SYS_TLM uNDL'R CONSTRUCTION:, uUT _L'fJCOUNTt«IiV(i SIX MOf-jTH DLLAY IN STRUC-
t NvIblLV.'.-''I;, "i'tiWEN Cc.~ ~ "" ""VUftAL STL£L" ANiQ VLSSLL OELIVEKY.
rSFlfLO "'0. ? . ... _ -.._
iy. - ,Vc* " ~ " " "'
COAL 4 .3 F'F.iiCEINT SULFUK
ciicraco
LIML SCHOtlAfilL UEfJTUKI SLKUdbLH FOK S02 KLMOVAL) . TIE IN OF MODULES TO THC
Ef.oYsicitJt.t-ti i bOiLLK n,vs ST,-\I AND is DOL FOR COMPLKTIOW &i OLCEK^LK isr. THL' SYSTEM
320 ,v,,; - -)LTHOr;f""" ~ ~ WILL bt. Tt'STL'O »iiTH WATEi^ FOS PART 1CULA TL KLMOVAL ANO UtBUGliIiiiG CUI![]() . PQi-.^ U.A [Eh_ {• .OLLC.UE.O. B Y_ A_Sf.COUD. y EN FUK I STA.G£....FUK SQ2. SCHUBf" I.NG_
-------
TABLE 2
wLH STATION
STATUS OF FGU SYSTEMS UUHIkb
"" CUKKLNT f-.Of.iTn
UlMlf""MOr2 SHOULD "f>L "COMPLETED BY
il.^J^t-Tjr if-/_Ob1nIiMfv3 COyJ
U'UIT" 'CAPACITY is"bsVi-Vr.-".""
76
1975 . UTILITY IS EXPERIENCING
CH£MlCALS_. __
" .....
i r-iu. 2
650 |X» -
"COAL "" £.":>" 'K:!JT~
NOT_SELEC.TEO ___ _
" ~
_
'-1E.STO>;L" SCHOBfa'lNG
STARTUP 5/75
I.C. fiUr-:;,LH _J7 ____ _
"/'UcfLIC" St^Vlc'E "OF17E17" MEXICO
KGu HKOJLCT IS PKCCtLOlNG ACCORDING TO PLAN UlTH NO MAJOR PROBLEMS.
L'aun-.1c.fv,f "is'PKESLiiTLY atlkt," FrfCCUKED. "' CO'NSTKijCTIGrt TO BEGIN IN AUGUST
37- rC"
_C.O AL _ 0_. n
DAVY POXI.'U
r SULFUH
_
it/77"
7'3
CKOJLCT Ib PrtuCL'LOIKo ACCC«LINO TO
rtv. ?• LXIO
WITH NO MAJOR
rO QE_GlN I_N AUGUST
SJM JUA;.; i.n ?
340 _ ,">. - .n,_k.JKOL 1.1
COAL" ~ o.jVVn••„ -
.;T SU.LFl.jH
:-"JT iri.
SI ;»i< '«. r1
"I.C. '"'ii. >.;.''. I' ""p"o
^UHLlC ?i^
-------
__ __ __ _________ _ TAtSLE 2 _ _______ _ ___
"euf'if'VuY" ~" ' " i 7 AT US "OF Hli ' 'SYS7 fc>ii>" tiUKlNii" 10774" .....
l'0'.vt_l< SI AT ION "" ......... CUKKEivT MONTH ' " ......... __.-....
»- — ^- — ^ — » — • — -»— ^ — ^ — — •--»--• — — ••-*.•— — ^••»» — ^«-^— — -•—^^^•-• — — --•- — •»»««^.»«»»-»» — — *.»«»« — ^*,»t»»ww*»«».- — »^^»»w»»»»,»^».^»»*»«.^»fc»» — ^^^»»«»»™*»^^*
r.D. UUhiU.H "si ..... V£i,i.)UK NOT SEtCCICU, blUS ARE OUt IN OCTOBER 7<4, THE STATION CAPACITY IS
_ s._ CACCLMft POM srpv AUTHORITY _ 2uu_ .'•u. _ ALTEKI-JATJIVE oios_jiA_yE_^iEj:i\i_ RE&ULSTED FOH THE TOTAL aao m CAPAC-
GEOf<"OETP»f-rij|' 2 ITY'TNtl A i~. 9S ~SULFUh ~COAU.
__ mu .vw - t.-Eu ___ _ __ ______ _ _ ___ _ _ _ _ .... ... ___ ____
CUAU i.u pt-.hCFiJT SULFUK"
_____ NOT sn.fcrf.fi ......... ........ ________________ ____________ ...... ._ ..... .. _______ .. __ _____________ ...... ......... _________ ........ _..
L 1 ML /L I f--t -STur, E SCMibfi I i-'.G
___ STAHPJE- 5/77 _________________________
» — •„ ^^«._ — ___^ — ____ — — __ — _^_ — _^ — — fc— — ^ — __»„ — w_to — _ — _ — — ,fc___ — — — — — _^^^— ^.^ ^_ — fc — ___^^^ — — — wwww-. — ,,_ w — w^ — ^^...^^^ ^^ ^^^.^^.^^^ ^ .„„,
__ I. P. r4U/,i..i:fi_ 82 _____ AWAITING TEST KJ.SUt.TS OlSt HOKIZOwTAj. AND VLRTICAv, _SCRUBBING_SYSTE_MS_AT
SALT tuvL-T PROJECT ' "MOMAVE. coMfR'ACi~'wiLil' «E AWAKDED 6/1/75. """ ........... ~ ....... " ..... ~ "
____ CiAVAjO kO i ________________ _________________________ .. ____ ...... ____ ..... _ ..... ..... . .......... ______ ..._ ..... _ ..... ______
7-j(j
__ C.) '-.v. - ••£>•
Co C*'"''t 0..;- (. . n
^ ...,f'i>T. Sr.Li.LT;'^
l'vt STliirh
nVn Iti PKOGHESS AT IHE fiohAVt" ST AT'lQN,
6'tl; iv-^ - .nL7R.PF.lT ______ _ ______________ _______________________ ______ ....... __ ..... ._ _ __ _ __
COAL 0.5- n.fl P£.KCL>\.T Sui_FUR ~ ........... ~~ ....... "" "" .....
MOT .Sr.LEL.TFl. . ____ .. _ ________ _
"" ' " ...... """ " ""
STARTUP f./V?
-------
TAbLE 2
CCrlPV.h.'Y "" ST
1 iU. '< riuVt -tl'l< *V ..... ~ THt 'iTGi\L i>LKOHMl!':0
I.i;. .U't.fHv 07 THL TEST l'. !;::1L\- ' J. 1 / 7 i
i.e.. ' MI>!',.U? " " Vd ...... ...... THL TOTAL' CAPACITY OF buitLn is auo *<>. LXISTING EXPELMKLNTAL KGO
SOuTMLPi! CALlFOKf^lA CDlSOfi WIuL TRL'AT ONLY IbO MW. Fli\,Al- SELECTION OF THE PROCESS IS PENUlNG THE
~ — SCTJ/WE'FF"? ----- OD7clIl^L''^"ntf5>T~7liJuS'1iurnti;~ PR6&r I-SSiSSiPK-rPMrc'GTjP bUrCJi e> McuOtiNLL L , ' IS PKLStUTL Y PhtlPA^ I ^G SPECIFICATIONS FOR BIDS
<,;,L-ij
"COAL "
NOT St
STARTUP f-/79
-------
i!
t'-OC iYSl
C'JKKLwT r.Oi'.'Th
J.L). rjUH,LI> 91 '" ""PLu'MT IS UNULK CUUST i ._ _...__
l.U. flb^tt-K <;2 _ _ 1HL riOlLl-K IS UMULK COf«,STHuCT lOf-J. INSTALLATION OF FbU SYSTEM WILL SI AKT
"SPCII-.C-FULUUriLiTY "IT .HKOCoKEWtfiT SlAtiL... . .. .. ..
200 f .v - r-L'.v;
CCflL
u'p..IVH'.Si;'L'"C'TL~PlU...;UcTS
Liiv.r.S'Of%F. ^L-"ir.c;I.':y . . ._.. _. ., . .... .
b/7o - - -
l.U. f-Ut'-c't-M 93 ThXb IS Ail LXPLR1HLUTAL SY5TE.K >UNOLO BY USfclPA. THE.KL A«t THRCL TYHLS
Tr.f:'JLi^E._L_ V/.I.Lf T_/.UTjjC'f'n_V_ OF SLKbtJuLrtS f 10 j-;,. .y.V'iy.A'-j-.^l-.r A.CHJ_ "riL.1J^A_^A^.^ATLy_U.kjjLR___&...VAi(.LUi.Y .AUl:,
Ll^L/l. I vc.HTu -a Sf.niut;.4
1.1-. . Ur., t.t< V4 _ _ f.LL ,«:AJOK f->HOCLSii LCUlrr.L^I HAS BCE;;! I'UKCH;,SEU. IN-LINE
TL^r.LV^'V.f. v'i.iLi*Y" ,->"i I K'r.'li 'if'" ' AKJ JlhuCT ULAT iYbTtM IS Iu DLSIuN STftiiL.
A It .y/.i C ;• :.'L' •'• '"•?• >' _. . ._ . . -
t=hO • ... - -X T.'U')r IT
CnAL .^tj_ r_;L±'Ct.-1! 1 SULf_\.\t-.
V/il Lf Y~Au"Tt-UHlTY
SCi^'toI'iO __ . ... .
1/77
I .0. MLV | Lrt 'J-J
Tr_X/,S J.'T iwlTij.bL . .
rt,\rTPi L/KK :-j. I r\iL« bYSTt;,":, LHtlCI llvG UOILEK. CONTRACT FOK FGU SYSTLh AhARuLD.
7« . .'-,..- ;,ti% . ._. _ ... _ ......
COAL U.'* |'f
-------
2
CUHPANY STATUS OF FGti SYSTEMS DUMMi
POwtK SlAT"lC)ij~ •.-•— - cUKrtEi'JT MONTH
I.ii. liuf'i7t_k ~~"SC>
JCX/VS V]_lL n_B-___
"M'nhTIiT'L/.hL i.O.g" Ivtrf oYSTEM,"ERECT ING BOILER. CONlkACT FOK FGU SYSTEM AWARDED
7^3 !•'*••_- J;F-W
COAL U.UH H.:••/• - i\L:l-." " " """" " " ~"~' ' '
COAL n.Hu PLi^crNT SULFUR
hA'k'Tlfi T/-.r,l. "i 0.""4 (•."L'y "laYSTLTlTTTkLt'i n7G~bUrLERT""COf7iPANY~liS"OUf"Fo"R^lDS~oTi~FGTr~SYSTEM.
791 ij'i. - f.tU-
^ COAL O."iq f'L.JCrrJT oULFUK
KLS!T/;HCH COTIH^LL
L l%i'.i> lOiii, StKUhi^l'sG
_5.l'u<'_?-l...._iV»w«a««l^*B.__VBk_M«v^«M..B.W
I.C.. Mihi i « '^9 _ IIWiT«l lOfi rOK biUS ON FGO SYSTEM I'AVE b£Eiv SENT OUT rtNU SO«E
"TF./nS OtiLiTlES ' " hAVf. bLLb KLCL 1VLD." "1 MC'Cu.-iHANY "IS' PHCtJLfjTLY' STUOYlfMG AND
HOmiCHLO 1.0 3 _ _ _ THtlSE PROPOSALS. _ _ __
COAL o.,;- i.o pfHci"T suo'u.n
>T6'f"""
L) ?.
STARfuP
-------
__^
PEHFORMAfvCt" SUMMARY FOrt OPLKAt'10I\lAL FbU SYSTtffiS"10/7<+
NO. 17
UTILITY M
EQISON
UfJlT tJAMC
UMI T LOCATION
L V LK L T T h ASS AC rilli>L T TS
150 fiw
FUEL CHARACTERISTICS OIL 2.5 PERCENT SULFUR
PIIOCLSS
NEW_OR_R£rROFIT
sT A ij j_ UP . OATE
F6D STATUS
_L'LU.Clf.'vC.y...
f'AKUcULftTES"
S02
h. OXIPtL
RLTROFIT
bU PERCtl^T
90 PERCENT
WATLr! 'V«KL UP
SLUUOr. DlSPOSflL
cosr
OPEH/.TICMAL
EXPLKIEMCF
MLS/KI«H OP
•FGu SYbTLJI IS UO>N__WIT_H M0_
S FOK REbTAKT ING_13ECAUSE_EPA_
FU.\L)lI-o KAS EXPIKEL, AUU THE CALCIMNG >'LAt\iT IS .MOW btlUG USED BY ANOTHER
_CO'vif A^.L? TH.E- uLiLiiY__is_Evy\LUATi'v'G TH£. DATA^ COLLLCTEU DURING THE DLMOM-
STRATIOf-J PER'lOi; TO ULTEKMIigE THE COURSt: OF FUTURE ACTION. " "
-------
TABLE 6
PERFOH.1Ar.iCE SUMMARY f-OR OPt-KATIOfviAL FbD SYSTEMS
,'-! N'O.
22
UTILITY MAME
CITY OK KEY
UNIT
KEY WEST POWER PLANT
UNIT LOCATION
KEY WEST FLORIDA
67
FUEL CHARACTERISTICS oIL 2. H PERCENT SULFUR
FGD
iM AlH SYSTEMS
NEW CK K'i. TPOFIT
FGb STATUS
00
00
EFf-!(.IENC_Y,
" '
"UP
RATE ESTIMATED 100 GPKi "~
UNIT COST
_
1 jr<
J.'l|r_Aj_i_l_hMA iii^Ei; SHUT liOxh Sl_M(.L SEP1 ly'hi BECAUSL OF A BOlLLH I 0 FAN MO-
"" ~" r'uL'Y" 'n"o>'c"~TO P-'t- "OPLKftT'l OK/'.L" "T fit. '»"ELK"bH UC"Tl;aCR" "slVf . " 'MLc'S-
o iU Tf-E F&D SYbTLN fcLKE' HXLC.FcD OPLhAILD ABOUT A WEEK
A" HAlf UtFO«£- THL FiiESEWl BCILLK ShUlCOW^. UUHIN'G SHb J DOWN THE EX-
_P_Aijt510t\j_ jOliuTS O.M THE DUCT^.OKts JO THE SCRC'tibEK WEKE REPLACED FOR THE 2Np_
Tii-iLl" SOHE VLiJGblijG KEnOVEU AT SltbTDOUig. ..... " "" " ....... ..... ""
-------
TABLE 3
SUMMARY FOR OPERATIONAL F-'GO SYSTEMS 10/7"t
_ID E ;«f T I. F l£A_Tl Or^_NO_. _2 S_
UTILITY J;AME COMMONWEALTH EUISON
_ W_I L L _C QU WT Y_N_0_ 1
UMT LOCATIOM
ID? r--w
FULL CHA^ACTDUSTICS COAL O.b-3.0 PERCENT SULFUR
f-GCJ V'CNLiOR
PHOCLS5-.
};&• C;< f-ETT-OFIT
START UP DATE
FGD STATUS
ErFlLlf.f-.CY.
oo (Vt;n ico: ATES
VO ' ll.V.w'41 C.O
SO?
BABCOCK X ^ILCOX
Ll^'iuSTuML SCKUtiBlNG
kUKuFU
2/72
Of-ERAT10|.;AL
9fc PLHCLKT
7t/-oO PEKCt^T-AViRAGL
WATLi-,
UP
o t-AL/LD KOLL S02
SLUUGt. DISPOSAL
SLO'jGE DISPOSCC l?-i CLAY LIt-jLD
OP
lL':gJt.f* _OUCF. TO CLE/ii; ULPOSIFS FKUh THE
biULLi<~ KLL«T£L/ ThO'JbLLii.
Vf ;»tid ihnuAT A(.U tvIC
..T.iUi\ ..CETI.KIOKATLu) SLIGHTLY MU WILL HAvL TO bt. HANU CASHED DURll-,^ SCHED-
"llLLO""o"clLLK "OurAGC. "DHLKATIO^ MOST OF THE MONTH WAS WITH A MEDIUM SULFUR
..C_9.AL_.lJLEM3. ____ A.yAl_L/jBl_LIJ.L_wAS_.65 PEKCEfMr .__ MODULE B IS_STILL_ DOWN. _ _
-------
PLKKOKMAf.cL SUMMARY FOK OPLHATIGhAL Ki>U SYSTEMS 10/71*
inCM irlcATIOMi-U''Kt UP
L'Gt. i. lil'JOS/,L
UfllT COST
ORE*.'. T I()f:AL
~TxTrL ~if.i..CE
.L_hlrth '>c-! L1I!ii FuRfg?CE: _lf)Jt:c~not>l SY?Tt:!!! ^ITI) N9 TA.IL ENt? SCRUBBING
"Ti"uaLu'. uuts'l Ai<0i.'lfa~ pii UjQ U/tYS AT A TiHE. AVAlLA8iLlTY Sl^CE STARTUP NCVER_ EXCEEDEO__
?5~PLKCtf%T. "" " " """ "~"
-------
ilAnct SLMC.AHY FUK UPLKAllUUflL FJO SYSTEMS 1C/7H
IDENTIFICATION NO. 33.
UTILITY N-AHE pbQULSNE LIGHT
Ut.qj
PHlLLIPS
_ut:.I T_..kQCAJ_lQJi.
UNIT HAT TUG
SOUlh HEIGHT Pc.NI-.iSYL.VAN_lA_
FUEL CHARACTtUlSTlCS COAL 1.0- 2.6 HERCENT SULFUK
££0 J/LI>i(.Lg5 CM.-.
PKCCL.SS Ll>it:
_!!£• W-iiL RL_TJ? 9.FJJ H CJ K pf_l_T
7/.T3
(,P[.;<»«T.lpi\AU
PAKTICULATES
S02
WATLK
SLUL'GE: DISPOSAL STAblLIZEU SLUObE OISPOSCO IN UNLINED POND
'~" ''" OH " "
bYSTtr. CONSISTS OF 1 TWO STAtE. ATJU 3 ONL STAt>E VEIJfURI SCRUBbLRS. AT
E.XPLKH riCC
PfUSt.MT. ONLY f-LUc GAS KKOM bOILLRS 2ii ANU u b/.^0./7H *Er
-------
TAliLL -6
^c^TjT^VKTn^GT<~ciF'"CH^TrcrirAr~rb"j~sY"i>'Ti:H's lu/'f'T
JUSTIFICATION N-0.
UTILITY
REIV-LHAL NOTOkS
UMIT ,'Jfl,v.£
CHLVkOLET FAKtffc 1 2 3 & H
LOCATlOt-j
l-'AKhA OHIO
UNIT KATjIVG
FUEL cH/uAClTY" ('SliMCE" ;SAKCh)" DUE~7u ~LOw MEAT""
-------
_ _
FbD "SfS'TLf'iS 10/7H
NO. 3b
UTILITY fyai^E: ILLINOIS t'0*E.K
. i\io.. t .._
(if4
UfilT RflTT'v'G 110 hh
PUE:L_CHAT^AC?EKIisrics COAL 2.9-_3.^__pt;hCi:NT soL.F_yh.
FGO yt;^ijr;i* MOiMS^uTO LI\VIKO CHE.M..SYSTEMS
f'HOCLS? CA_T«XLYTIL OXl_Q;iT10M
FGQ
U)
!(.! tf-cy.
P/KT ICiJI. A
W/iTtlic f*f-rt.il up
COS!
i73/Kl'. C/'-l'l l<--L**t ,-;lLLb/Kv%H Of
i, u:.J lHA.Sit.L"i'v
TriL IJA.SI_6C DAYS TO I'CKFOKn SLR\/LRAL HLPAIRi,
itf' !-.K l(.K>b~lN 1 rit~Klitlt./>T t« bUHLjtH L)OME~HA
RE.r-.aHi> TO i>LFi< ALToH Y LliJl.u DUCTS IS Ifj PKOOKLSS. OTIitK
uF [iLt.SS bTLEL OUCT'/JOHK Ar.O KLPAIH OF THL ACID
"" iArr-'CCS"."" THL U.-jiT" IS LCHtDjLuD TO RESTART f'Y ' NuVL'Mutu 22. 19'7'»i"
-------
TABLE 3
him
MAL
_ __
T07TH
NO.
UTILITY NAME
KANSAS CITY POWER S, LIGHT
UftJlT HAKE
HAV. THORN NO 3
UNIT LOCATION
CITY MISSUORI
UMIT
mo MI»
Fut'L CHAijACTEKlSTlCS COAL 0.6- 3.0 PERCENT SULFUR
LlMLSTONE IfviJEcTiOM SWET ScRUo
WtW oi< RU
RETROFIT
START UP_ L)ATjf_
FGC STATUS
vo
99~pE7icEi\;r
70 PLRCCKT
UlSPCSAL
U,MST«ulLI2L(j SLUuGt DISPOSED IN
PONQ
EXPLHTET'Lt
.;j MLS/KWh OP~
KCPL HAk A LAbOH SFKIKL AT ThL PLAK'T WHICH LASTED FROM JULY JjTH TO OCTO-
STAFF. hOWLVLHt btcAUi£ OF THE KAr-iHOwL'< ixiOHTttGt* Hit.
!& WERE "DGW.g BURINS THE Si f-, IKE." T'dC'FGD J^lf KE5TARTEO"OCTOi>tK" 1ST
h^.S b 1-E.'-iK U/
-------
_ _ __ TAbLL _i
~FO7FTTS7TA'MrL' 5UfihAr«S_Cl 7.X ...N I.S
100 M>
FUEL CHARACTERISTICS COAL o.t- 3.0 PEhCErgr SULFUR
FCip ML'*JL>OK__ _______ cOi-ijjbSl ICU KfJGI_'JLEH.I)v_G
H; Oci; S_S ___ L l^.t'.bTQNL, lU JLCj lOi'"
_ :-it V ___0j.<... ^"TPQE.IJ ______ ^tj '^V I T .....
ST
hf-
.LF.
F
SG2 __ 70
/.".t; UP "~ ~ "
CTSHCSftL. _ ut.STABj^LI ^t. L; $LU uGt UISPQSED IW U U LJJ[4Eu__P OUg
_ ~~d:.2 MILS/KWH OP .....
Ot EPATICf at
'".0. A LABuH _STRIKE AT _T_H_L PL/>.NT WHICH LASTED FKOM JUL dTH TILL OCT
l'sf. "Ut~r<"l'i-^'"fHlS PLRlOu "THf "ijOILLKS WLKE OPEKATLD 13Y" THE COMPANY'S SU-
PLKVISOKY STAFF ._ hOkLVLKt DUE TO THL KANPO^LK_ SHOHTA6E i THE_ FGQ UNITS
•A£KE ~u6*fj juRi.jb TH£ 'STKI'KE. " "THL'FGo'uraT HAS "WOT" RES'TARTLO BECAUSE. THE"
f-OILLR IS_UCi«N TO OVERHAUL TH?; TUt
-------
PiLRFtftilAf.ct. Su;-n.AKY h'uTT~ijPLKAVTOI-mL Kbfj i>YSTilMiT 10/74
IDENTIFICATION NO. |*_0
UTILITY NiAf^E KANSAS CITY POWEK e, LIGHT
1'.N1J_JAI^E y
U.'.iIT LOG AT IOM LA _
UNIT KATiMb 620
FULL c H A aACTE: i: isTir.s COAL
FGD v'-^OOK B&UCOCK i WILCOX
.CSTOr-JE
_
73
.: r-i"H"KT"L!F'" "
SLULibL UlSPOSAL UN S TA B_1JL 12 L[.i SLUuGE L)I SHOSEp I h U f>i L_I W-U_ jPO kg
U::lT"COST i'Vi/i'-K Ci'.Hl [ At" ' " "
OPLRATIU-jAL KCPL HA» A LAPP* _STRIK_E AT THE. PL^M WHKH UASTEU_FROM JULY BI.H.TILL OCT.
EXPLrUCNCE liil- t/UMljfc This P Lhl'OU "T ML bOlLLK~WAii OHtHAt L j "nV" 1 HL "CUMI'ANV ' '5>'~"SUPEK-
ylSl.HY S1AFK. THE! FGO SYSTEM OI'LrtATLU WITH ABOUT f,0 r-tKCLliT AVAILABILITY
""LioKir.O OCIUL'CR. TMtlKL" AKE.' NtThAJOK OUfSfAKuIwu PKObLLKS. HUWEVLK, F'KE-
VL^TIVE: \-\t\\^VL\*^\tc. PRACTICE UCCOIKIIS THAT CUE :ODULL BE TAKEW OOVJN ONCE
",V "«j£tr\ "FCK o H'uUKS" OF" CLtAM'NG AfvD /.'Aihi^b OF
-------
6
___ ___ _
l.rU OKMAf-.CE SliF.VuvY FuH "uPLK AT 1 Of/AL FbU S'YSTLMii 10/Vt
ItJl'f.l IFICATIOM NO. ____ 43
in iLiTY .yAriti K AI\_SAS__PU_ u E. '<_
_UN'IT LOCATION _____________ LAUI'.ENCE .K/iNS
IJj;. It_:tflT_i£jG 12 5 l*
FUEL _.CHAi<.'*cTj:.\J LLS,__ A N £ _ IN _ 0_P E ? M T I 0>_i> .1 T H .LS S E N T I A L L Y_ 1 Q 0 £E K C E N T_ AV A I L -
" ' "
_ _
EX.JtiUi.f:CE AfulTf. J-iA.vUAi. ^ASll-UH Or Di.^:13TCnS IS PERFORMED ABOUT Oi"\)Ct. ' EVEKY TvJO
____ _____ __. __________ '*fc.!\>._. >,Af^AL CLt*,r;LP pF_KF.h-:/>TLi( COIL IS NOW. OUL AFTC_K MOKE,..THAfJ. 12
|--.CNTr:S Or " " - - ......
-------
TABLL' 3
KF OiMiAUCL SUf-'MMHY FbH OI-'LK A1 I GNAL Fbi) SYS1CMS 10/7<+
_I [: E i: jj F_I CA T IOM f>)0.
UTILITY DAME
MJAS POWt.K * LiGHT
c NO 5
LA*KLNCC KANSAS
FUEL CHARACTERISTICS CO_AL___3.5_P£.K<._«:;«T SULFUK
FGn_VK_M(jOK COMBUST ION ClMGlUtEKING
t~»S Livi:STO.'Jt I'JJLuT J.0f-i a'.JE
f.t'U C'< |« _
STAiU Lii'_y_Aj_E L1"!?..1. _
PGP STATUSOPE NATIONAL
" yj ^LKtLf.T
SO? (zcj Ht.KCE.NT
~M7.i\r~up
or ;;j-;t--jsf.u. b..sr.v,,iLi^L _
UTTiTTcsf ''
iU.iLI r<_ -_h,-,i nuC.._CHE.: "TuL H i'lUJjULLi " AS~ i^tLL 7,5'. "AI--IU'I\G' ~1 HL MuuULLS IS
STiLL >; r-iAS 0 I ST ri IbUl I ON
ILST5' ;;i:L. lb> CoKKLi.i'LY :'1UDI F Y I i JO Int. G(1S f LOJ PATTLKU THROUGH TnL
f;U !.'S OPEKATING HUUKS OK SYSTEM
'"""" " "" " " "" '
-------
_
SUMMARY FUK UPi.kAliUNAL FGD SYSTLMS 10/7H
ipcy J.
UTILITY .VAKC
LOUISVILLE. GAS a ELECTRIC
UUIT ijAMF.
UN I T_ LOG A.! L?y
UhlT K
H OCLS S
STA.K.I .u
FSfl S
EFF 1 L.I ft i_c_Y_,
PAuDYS RUN M0__b __
k°u I S V.I LL E.JK EN Tit CK Y_
3.5-ir.O PLHCEM
S CUJ E &UJ
V *?.
S02
_„. ^
SLUUC-L uispcs.a.
E LI SMC SEJ. in ONL i r-j E y _P o i-j
UuIT COST
W CA['I1«L
I.-.O.
F.'-.CL
J.HE
_ ci> ILY._TH|. P/.ST 3 ;.,Xo Or. Li'lcSTG^E (UhlT OF_F_
r,ii;j,L K.JI.S AKL i-.toe. FOK AKO f ur.(,ru I*Y cc-MiusTiofc "L.'-JGINELR-
,,y,Ml-MniuI f Y WAS t-SSL/jT lALLlr 100 PEKCtl^T wlTn t\(0 PrtOtfLtKS LXPtRI-
1IJ AioY PAKT OF ThL SYSTLn. TriL L i,v.ESTOfi£. SLUDGE IS LiLWATLHED IN
ROTAKY VACJOM TILTCHS »'1TH r;o CLIPPING OF ULIER CLOTH.
-------
I/you: 3
PLIU uh.S/HjCi- Sur.hAt'Y F-UK GPLKA1 lUNAL Fbf) SYSTfclMS 10/74
c i-jo. &i
UTILITY !>;AC:L NEVADA POwtK
LICIT rjfPt: REiU GARDfcEK NO i
UNIT RATji.'G 12
F'jEL LHAKACTEHISTICSCOAL
SOUIUM CAKbOXArt" SCKOBblf-JG
iv£ W CK RfTPOFlT Rj-lTKUFlT
FGD STATUS CPLKATICNAL
M Lf-f-IClEt-'CY,
O >-AKT:C^...Ti
S02
UP
bLUCGF.
""Ur\,iT' COST
OFri-,,T!C::,-,L _ FC;; btSTL/: F'LtiK:j:<.v;.!-;CL 13 SA I IS^ AC! OKLY WITH f;0 OuTbl ANDII-.o I'iLCHAi-JILAL OR
"TX'P L.":< IL'-Ct "P i-UL"tT^J~r'i'iOVJ'Lc. ••..•»' ~TC.'""f LXPLi
-------
Pc.r«f ui •!/•.(•• r.fc. SUrti-.AKY FU< Ol'L'AATIONAL KuU SYSrCNS 10/7U
._..L'1. _______ 62
:..\f:E. _ N-E.VAL-A
oT-JIT IJANr Ktll) oARuWtK NO
uM I T . I. ..LA f I ON ," A Or A .J.E V AU_A.__
__CS^ _ COAL_ ..0.5- 1.0 f'EKCLNT .SUL_FUk_
t-GC .vi::.:;,fK_ ........ __________ CCJ.-'i.iuSTIor.1 _CQUIP. ASSOCIATES ____
J-^ClS SOb 3 LS CAH^O.'I;. i £ SCKU
lit w L,'< '•TTJ-'OF J_'l_ _ __ KL. 1 1- yf- IJ_
STA«1 U" iltJC 12/73
!<;n iOl;AL
|-. TIC- •'-.;> L Fli SYbl Lf-'. 'LKK)«[-'.;.i\CE 1 S A 1 b FA C T OH L Y WITH NOOUTS1 AivDIgb MECANICAL
_ _ _ _
t Xf-t iflE :.C"r PKULt':J:r?;(A POWLK IS SUIriG
_ T--L. Sj.:rl_lL'< FJ.;< ortLACfi OF Cui.lxACI ANU ALSO IS PrtESLNTLY TJLSTlNo SOOJtUM
C;.K:jO. ,\rj" biMI-.c- PUKCHAiLO FI
-------
TT/U3LL 3
SunY-AK'YS!L''v' ^AS OPLrt/,lL[J Ft>0;-i OCTCl-.Lft i^D TO Tiiti itTH. >'H[1SENT
i_Y THf. SYSTTK is OOv.'fi'TO f\£f'AlH LL.AI\S LN Tr.L F1"i (;ULHt-T £LEVATOl«. &
-------
T^rrT. Suvi-.EPY ru!™;Pf.7\Tm;i,!AT~FTJiJ Sf'STtTTS
niF.t.,1 J,FIcA_TIO^_I«p. 67_
UTILITY c.'AME SUblHS.KN Cc.LI FuKNl A EDISON
_yfo i T_ri«r_c MO_HA_V L _Mo,_i
.yl*:I1_..lrP^Al I.?^ LAUt
FljE!L...CHA'.• Civ i.L"i>TiJi\t KOH i-uJ«JT Tv.'O «LLKb \\\ OCTUbLk. THE. yi'lT IS WOW bACK 01, HME,
int. Avt.iv/.CE: AVuILAlsILI IY OF Ih.lS f- v»J SYSTE-rt SIf,CL ITS IMTiAL START UP
iij JANUARY 1974 is ABOUT eo PLKCLNT.
-------
TALLL 3
S u •-: r-. i\;-. 7~~F C/H
___
7TS 10/7t
LL)Er''-T
UTILITY t-j^.
VALLLY AUTHORITY
uwIT f.'AME
NO. 10
uri f_ LOG AT ION
U ,] T K
PAUUCAH KENTUCKY
EI'ISTICS COAL
FC-D VLNL/OR
T£N;JLSSEL' VALLEY AUTHORITY
_ST_AR T_UF• _yATE
FGO STATUS
,< E T K c F1 T
H/72
OHLKATIONAL
C.FFIL1LI..CT,
""PAMICU""
S02
r, y'ftrtL UP'
SLUDOL
UilIT COST
OPfK /
Thlb IS /,N LXPLKlMLfMTflL SYSTLh FUtJUEU bY USEPA. THCKL A^tTHHE-E TYPES
" Q 0~T-ra~LTJj I V ALLMTACh ) OL I "fTG~n7"ATUATE:"irTI^UL'R A
OF OPtKATlOI-iAL H
-------
K Ai'U TOT;iL i\* Of FGU SYSTc.hii
STATUS ,'JU.UF
UNITS
I 0, i-Tix-C I lo.j le c.677
_ _ . . f .
o ""
^ COf-j5l.'J.L..t.'!.i.G._o.'''iLY...F.Ou SYSTLr.S 37
TO TAJ __^._ __2S.
-------
I'AtiLt b
•'JF FliD SYSflMS
UY COMPANY
TOT
AL GPLKr.T IONAL
co;, STRUCT i
COHlhACl LHTE.K
AUAROEU IN1ENT
UtTLlty
AL»;(.>»f-'A LlCClMlC COOP
iiRI^ljt-/-. Kl.i'.CTKlC PQI-.U;
:V
CIuCIU-i. i 1 f/iS /V'G LlFClKlC
CITY oi- KtY ,-:F.SI
tULUf-'.U.C (>IC fLt'CTKK
C0"r/ ^'.vf-l^r^u'.; °!lIU
£ A I n Y L A i , ;, t ' 0 .•.£&'. C O 0 f '
DIMKolT FT ISO,-'
Q Gti!. KAt -'OTf.ivS
ON fjl.l.L'-.'.L" i'1'i 1 if. "L'Tl'L'n i'Li.
__ iLLrr.is ce,L!i_ _
• Ii:li I Ai.Mi'TiL 1 5 t'O.'if.lv 'Vi..;i LlSHt
l\.»;;S;\i> C1JY i'f.i-'L'l' .< Llf-Hf
rtAuSfcS rOV. i K X LK-liT
LOUISVILLE. lliU.
li'l).
1UO.
obO.
"Inn!
<; 2 b 0 .
1600.
bt:0.
4472.
0
0
0
•J
c
i
0
0
0
1
0
fj
1
1
0
1
0
1
f)
3
0
1
0
0
0
0
o"
0
1
0
u
0
0
1
0
0
0
1
0
MU
0.
0.
lib.
0.
11)0.
0.
u.
0.
37.
0.
u .
loV.
60.
0.
410.
0.
llu.
'•" " o .
1060.
b2b.
0.
"6b.
0.
G!
U.
0.
0.
0.
100.
0.
0.
C.
0.
160.
0.
0.
0.
0.
NO.
U
u
u
u
u
1
u
0
0
u
u
0
0
1
1
0
o
u
o
0
0
1
2
i.
u
1
2 1
. ^ .._
1
U
1
u
0
u
1
0 ""
1
1
1
u
PIW
0
0
u
0
u
100
u
0
u
0
u
0
0
100
bio
0
0
u
0
0
0
u
0
113
ol>U
120
0
6bO
0
0
16U
0
343
You
bbO
0
r-iO.
0
0
1
0
0
0
0
0
0
0
0
. " "0
0
o
0
o
o
. "" o
0
0
0
2
0
. " " u
0
0
o
. 0
0
. u
0
2
0
u
o
0
0
0
0
. " "2 "
KM
0.
U.
" "250.
u.
0.
0.
0.
0.
o.
u.
7bJ.
U.
" 0. ""
0.
0.
u." ~
0.
" ' (J. '
0.
u.
0.
603.
0.
0.
0.
0.
0.
0.
0.
0.
71b.
Q.
C.
0.
0 .
0.
0,
0.
NU,
0
u
0
0
0
u
0
0
u
0
0
0
0
0
u
u
0
iT "
0
0
0
u
u
1
u
u
0
' 0
c
0
u
u
0
0
0
0
0
u
I
"o " •
OF KEQULSTING/
EVAL. BIDS
MW NO. MW
0.
40b.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0 •
" ' 0.
__ 0.
c.
0.
0.
0.
0.
0.
u.
0.
0.
0.
0.
0.
0.
0 •
0.
o.
0.
0.
0.
0.
0.
0
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
1
0
1 '
0
0
0
0
c
0
0
0
0
0
u
0
1
0
0
2
0
0
0
3
0.
0.
0.
o.
0.
600.
bOO.
0.
0.
0.
0,
0.
o.
0.
0.
650.
o.
515.
0.
1400.
0.
0. "
0.
o.
0.
o.
0.
0.
o.
0.
0.
0.
140.
0.
o.
446.
0.
0.
0.
2386.
CONSIDERING
Fr,0 SYSTEMS
NO. Kw""
n c.
3
0
n
0
i
0
4
n
n
n
0
n
0
0
n
0
n
n
0
fl
y
D
1
y.
n
i
n
n
n
y
n
y
0
n
0
0
"o
165C.
0.
0.
0.
bOO,
Q.
0.
0.
0.
0.
0.
o.
0.
0.
0.
0.
0.
0.
" 1899.
14UO.
0.
590.
0.
800.
0.
0.
0.
1000.
0.
22SO.
1280.
0.
0.
0.
0.
0.
TOTALS
95 3/961.
19 _ 3291.
16 6677.
3904.
530.
13 6U87. 37 16472.
-------
TAoLL b
SO'-«.V,AKY OF FGu SYbTti-lS BY VLuUOR
- STATUS
"TOTAL OPLKATIONAL c
hst.I.-'b 1 3-r3. 0. U. 1 343.
iL Jt-:JLCrioli isisLT SCHUb 4 7ob. 4 76b. 0 0.
. ;,£ . SrKu;.iP.Ii..C. _2,_13bO. 0 C. 2..1360.
TOTAL, Co-IfUiM I0i-j Ll.oJr.tLMMa & 2bi5. b 63u. 3 1703.
COi-iiL-ST KM L..-UII">. M^SOClATc'S
U.I.M. SC-..l---t.Ji.:.-. ... 2 72u,__ 0 0. 2 72C.
ScSiL-r- C-«i't-^:^.TL bCi'C'vUlisi; 3 i7b. " 2 250. 1 12b.
TOTAL, CO-ii-U'-. TlG\ tOi'lPj^ASSyCl b _10^r), 2 250, 3 645.
DAVY rciViH&r-s
.--./.r.' I.OPO 1 ii_. 0 o. i 115.
TOTAL, 0;.JY .PGv L_Rr,,%S. 1 lib... 0 0. 1 111)..
L1.VL H.^r.CTjU'; 1 oO, 1 60. 0 0.
TOTAL, FQSTLM W^ETLKK i ajij i &&_• o QJ_
KOCH
iJOUbLE ALKAl i
TOTAL, K( CM
1
1
3.J.
1
1
....32!
U
0
0.
o.
Ci.t''- SYSILf'S
CATALYTIC OXlPATIUf* 1 11U, 1 110. U 0.
TpJ_A_L., iv.Lj'.S.A.f-..Tp_E 1V..IRO _C!it-i^_SYb _ 1_ liO. 1. _110. 0 0.
-------
TAoLL i-
SUMf.n
KY Uh
FG[j oYSTfiS UY VLWOOK
TOTAL GHEKAT
M Af'il'F AC Tuf-.EK /PROCESS
NO.
riU
1,0.
--bTMUS '-
lOkAL CONSTRUCT 1C-N
hW
NO.
Ml-.
f.OT SEL.EI.TF.L)
LJNE./Lir,f.ST O.Mf SCXliL bl|.;G
TOTAL , l-if f Sf Lf CTLu
1
1
bbU.
b JU .
0
0
0.
0.
1
1
650.
650.
PLAbGCY t.MKIr.'U. iUMC;
LIKESTOI-.-E, SU'UHtilNG
T'JTALi !!LAbOnY Elf'.T.iriL EHIIvli
a
i
180.
loO.
0
0
0.
u.
1
1
180.
180.
PESF.'/ifJCH OOTTKf L I.
LliM.STC-.'.E ?.f HUMMING
TOTAL, KlSD-i'CH CCiTTKElL
i
115!
1
1
lib.
lib.
0
0
0.
0.
TILEY STC KE R/c.NVIKU.NEE«ING
LlNESTOUi; Sr.i^UboI"'^
TOTAL, R1LLY STOKf K/Ll''V IHOMEEH I
i
i
100.
100.
0
0
0.
0.
1
'• 1
100.
ICO.
M SCF/ SUii.; KiVol'K
Oo L 1 .'. E b C f\ ' Jf. r I • ' G
TOTAL' ScL/ .SltKlj «OGf k
TE'Jf-'ESSEL VALLEY /.u HiOiuT Y
LIhL/L l-.l-.S'l Ci-C SC^UhrH-ib
LIKL'STO; L Sr.K
I'iAGNESIbi-' OxirE SCiU.-r.bINO>
TOTAL, UhlTEti ENM!
LIMESTONE Sr.^UbfilliG
TOTAL, Zuf(|\ Alfi SYSTEMS
i
i
i
i
2
1
1
2
^
1
1
37
IbU .
IbO.
oO .
bDU.
boa.
•liO.
-^T^O.
3oO.
SoU.
37.
07.
101b6.
1
1
0
1
0
0
0
0
1
1
19
IbO.
30.
0.
30.
U.
0.
0.
0.
37.
37.
3291.
0
0
0
1
1
1
1
2
2
0
0
IB
0.
0.
0.
550.
b50.
120.
120.
3bO.
360.
0.
0.
fa877.
-------
TA:lLL
SUM- ARY UF f bU S
. ._,^ CAPACITY (t-.'O.'
7
YS.Tt.Mii bY P
OF' PLAI-nS)
K
GCLSi,
N(_ft Ul< Gl'CKAl IUNAL COI.bTKUCTlON CONTRACT
P'UlCKbS M. llLS !.
K
Llf't/Ll'lfSTCNE 'X CF TOTAL •*•'.: f-
F\
NO. r-iw
J 0
4 715
U (J
AJ
o Iib7
5 647
l(j . li1?^.
u ti
o D
1 32
0
~ 4
0
0
D
U
0
£ *
0.
0
J
(J
0
u
J
0
0
0
1
0
0
c
~o.
Mki
0
0
c
0
0
400.
0.
0
0
0
0
0
' 0
0
0
0
u
12b
0
a
0
530.
o . - "
76
U
HEQULSTIliu/
tVAL. tUUS
NO. Mrf
0
0
4
-•• 0
0
"12
0
0
o'
" 0"
c
0
c
0
0
1
0
0
0
0
0
13
0
0
0
1755
U
" 44B2
0
;' 6237.
0.
0
" " 0
0
0
0
0 '"
0 "
0
fabo
0
c
0
0
c
. 63B7.
0.
0
CONSJ
FGU S
NO.
1
13
3
1 "
0
2
12."
U
0
"d '
0
0
0
0
0
4
5
' 4'
' 5
lb.
21.
ICtRlNR
;YSTLM' "
MW
BOO
4333
2250
tiod"
3372
6b3i.
0
0
0
0
0
0
0
0
2500
1915
2500
6922.
"7550.
72
75
TOTAL NO,
OF PLANTS
NO.
7
21
3
26
10
34.
0
1
0
1
0
3
0
0
5
' ' I
3
1
2
50.
49.
HW
6275
14oU
13012
y/44.
0
110
0
32
0
37c
0
0
315Q
375
375
455
24V60.
13001.
85
75
-«*—
-------
1 rt'iiLL FGJ/.-iw STAUTUI' LXt-'LhlENCtl < MO. )
CATALYTIC OXIDATION
WOOD hlVLK NO <* 110 10-72 2t
110.
L'L;Uf,Lt: ALKALI
_?.«RMA l 2 3 a t 3.2 3z.?J* 7.
32., 7...
o ALMA srATjOrg ao 6-7i HO
on. HO
L1?T SCjivJi'nIN0
_HUHAV£. NC 2 _______ ifeJJ _ 11~7.3 __ 11
U'N i^o & " &s "4-73 ' ie"
PHILLIPS ___ _ HIO 7-73
SCKUbOl.'JG
SHAUNLE NO. 10 iJ 4-72 30
30. 3C
_HAwTHoKf«i _r,0_ 3 iHy 11-72 23
f r 0 "'» iUO ""S-72"~" 26
r.O H _^__ i<:'J 12-bfl 70
~"LA'A;f
-------
HP
1IVK UiMlTS />;> OF 10/7C'K U b h I"I
1
-------
TAbLfl 9
OF
"-"-'SCUuGE-- --SlbuGL'—•~ — -PDi^D--- " "-- -PONU—-"'
J Ui'MSI AblLI^klD LlfvLD
E SCHUBBING
O 2 ibo
S KUN hO (, ~ " 55 65"
N 410
TOTAL t)70. fab. 0. bib.
"TiATt"i-iof<"~iv'cr""3 mo" ~ i'»u
hA'-' 1 Hlii'^j t-jO '* 100 100
fjQ b i+ 0 0 H U 0
_ _ ______ __^ _
l £ i'b ~~ 1 1 5
__ ___ ______ ____ __ ___ _ _____
to "LA LY'jfitf i^U T • " - - _...-. ......... .- ^.^ ......... -. ...... -.-. . 620
^^IUi^—1 _________ ...... __________ ____ 167_ _ __________ ......... _____ ,lb7 ..... ...
T;iTAL 167. 1700* 167. 1737,
-------
1Ar.L£ 10
OF t-Gij inATEK LOOT' PKAC
K is. (U 1A L
• OPE:*" cLcst'j
CATALYTIC OXIL-AT10N
wuco KiVL'K ?.G 4 o a no
0 0 110
LOUfjLF «LrtAlI
PA.\-fiA 1 2 b * ^U062
LI--E: I^L-CTION
ALftA STftilOh 0
0
M
M
rtCHAVL" ivu 2 0
P/VuiJfS F-:oM f.:0 ft 0
PHILLIPS o
0
0
0
0
C
4lu
410
BC
oO
160
o5
0
225
LI^-E/L Ir LSTor-.L. SCKul-.L'I^'tJ
SKAS.MLL MO. 10 0
U
LI«ESTGf.E INJECT I0;j iwc!T StKOu
MAWTl-Ohr.: NO 3 0
HAUTHOr'l'i ijQ H 0
LAW(
-------
T_Ar:
_
OF FGO WAIE* LOCP PRACTICES
HART 1AL ______
...... ....... OPtiiJ " Cli/SLO " CLOSED
CMCLLA MO 1 IIS 0 0
KiJY UtST I'0*£K .PLAnT 37. .0 0
LA CYl-.KL .iO 1 0 £-21) 0
-.vILL C(a.'i-.1Y 1-0 1 0 167 U
907 p__
lUCKLi'SO,.; ,\:0 i CO ICO
C r.-c....b . u _u
..0 0
i-n'JlU-' CA^i'C/.'/JE i>CKUbb!.\'L-
Kt'lL' r,flKr,jth i.;o l~ ~ ..... """" ............ "~l2b "~ 'o ..... ~~" 0
Kt lu. GAi;;i<(:F>, f.O. .?„ __________________ ..... ._l*i>. ______ 0 ____________ 0..
0 _ p
-------
TAbLE 11
OPERATIONAL FGU SYSKf'iS AS OF 10/74
UT'lL I fY" CO^'7;fi~Y~ "Ktu OR S12E OF FGD PROCESS/VENDOR FULL CHARACTERISTICS
POUF.'< STATION KLTKOFJT UNIT
PUBLIC SERVICE K 115 RESEARCH COTTRLLL COAL 0.44 PERCENT SULFUR 10/73
.ChpLt.A j\0__l. _.._ Lli-H-STOWE. SCKUbBluG _ __
.150 CHtf.lCO OIL 2,5 PERCENT.SULFUR .4/72
MYSTIC NO 6 MAGNESIUM OXIDE
CITY Gf KEY fcCST N 37 ZUl(N AIK bY-SIEMS OIL 2.4 PERCENT SULFUR
KLY_'»EST p(),tR PL.A'Vl Llf.t-STOt-jE SCKU^ulMo _.
_CONrcrv»LALTvj_ EP.ISGjVi. R ..167 t-AbCOtrt N UILCOX _ COAL 0.6-3,0 PL'RCiNT SULFUR 2/72
»ILL CGu-JTY I-.Q 1 Ll.'-iESI CJ.NiL S(
LJAIKYLAI:!.' 1'CHER COOP R 60 FOf.TL'r-. wHELLER COAL 3.0- 3.5 PLKLF.NT SULFUR 6/71
ALMA STATIOM Li*>£- I^4JEc^lo^J
i)OViUtS!'t._ LItHT ii_.. "410 ChtMCO COAL ___! ..0-. .2 . 0 PERCENT SULFUR 7/73
KcxH COAL 2.b PERCENT SULFUR 3/74
OOUULE ALKALI
"" ...iLLlf^'iS, .POv.ER. R ....110. MjflSAr.TO LliVIHU ChUi'l SYSTEMS. COAL,. 2.9- 3.2 PERCENT SULFUR 10/72
u luvi.K vc it CATALYTIC uxii;;,no.v
hA|.;SA.S CITY i'0'.JFU H LT&r.l N 620 is/.tiCUCK i UILCGX CCAL 5.2 PERCENT SULFUR 6/73
.LA CYC-I'JL FJO i L^.LSIOJJE ^c
KAI;S>;S CITY PO»CH _*. L TGnT K 1.00 COf.UUST 10,'j f hGI TJLEKIMG COAL 0.6- 3.0 PERCENT SULFUR 8/72
"HAxThOK;-j >jo"~4 " "" LIML.STO.-JE i.N JL CT IG-'-J i'.vET SCKU&"
CjTY PCVir.i H LjGuT h 140 CC.vibLi 11 Jfj f'.iTGIIit-LKl.'Jb CCAL O.o- 3.0 PERCENT SULFUR 11/72
•'•-i HO 3 L i:'.LS JUI'-JL PiJLCTlOu iV.ET SCKUii
KA1.SAS >;.')>. E.<.S_L_I_l.HT (A? Pi;ufi< X LIGHT R 12b CGI'.bUSTlOU ENG I^LtKIi-iG COAL 3.b PERCET-JI SULFUR 12/68
AA.WH.E.f.CL '-0.4. .__ Ll!'LSrO!JE ifJOECTIUu iii-ET SCRUB..__.
LDJISVlLI.r. f-iS^^LLCTPJC H &b CG.v.bLST I OiJ t tjb INLEH !•'.& COAL 3.b-4.0 PERCENT SULFOR 4/73
PAOBYS «u^ r-o ^ " " " " LIM ~
NEVADA FOWLS K 12b CO:-.bUST 10.J EOUlP. ASSOCIATES COAL O.b- 1.0 PERCENT SULFUB 12/73
KEIU GAKDf.LR HO.I SOUIU.S CAKbONATE SC
-------
TM'LL 11
OPfcP/.Tlul-.f'L PTil/"SY~STC,'"S AS UF-' J.t//7H~
UTILITY" CtiKPiJ'iY
POWrK STATION
l.tV/u A
HLID &AKUM.H N'O 2
p o T o f. A c_j y. c rji i c
SOii r-,0" 3 "
bOJ I MtMf, CAI.IFONM1A
HC'H/JVE f.'J 2
"sH~fc:>('L '-o. fo
fvC* OR "SIZE OK F'tiD HHO(.i.S&/VLrvDi;K
RETROFIT UNIT
FULL CHARACTERISTICS
MO/YR
~ii-3
ICO
160
30
." A'SsC'c'iATtTS c'oA'L u.b'- i.b PC."MCENT SULFUR "i"a/7j"
i CAKbONATt SCRUbU ING
CHCMCO
i'c 'GXIDt SCHUSiHlNS
LIME SCRUBBING
/iUTHOiL _ 2.0 PEKCLNI SULFUR 9/73
TUAT. O.b- O.J~"PE"WE'ffT~~SULFUR" ff?"73~
COAL t/72
L i .'•:'£ /L i HCS'J oi\ie "s
-------
TABLE:
SYSTEMS
STATfCf.'
_ _
I r07T'flS~OF 10/71*
~K£WOK —"SIZE" OF FGD
RETROFIT UMT
' FutL CHAKACTERISTICS
MO/YH
CtHi R KLl LL 11 K> 1 b T.16I1T Cu.
'O'JCK CREEK NO.l
L'ETkC>lT
LI(,HT
KCN1UCKY UTILITIES
~RYLL T"5TOK D^TLTiVI R"CiRt
C SCHubtilMG
CUfeL 2.5-3.0 Ht.RCtNT SULFURJT7&
ItiO
PtAbOUY LN
COAL3.7 PERCE:NTSULFUR
bit)
~CR C n ICT3
LIMt. SCKUUbING
A Mi-: f< 1C AN AIR FILTLK
COAL 3.a PEKCENT SULFUR
11/74
S77T
i ATA PO»H<
iibu
360
COMHUbTIOIj ECUIP. ASSOCIAESCOAL0«0 PCKCENT SULFUR
"
115U/;V»
cvn.
COAL
5/7&
5/75
- 3-5
iiiJ-. 5-TAT( S i'Cwti-:
GOO
fcrt'J
i.c
/75-
SULFUR 6/7b
5/76
C°AL-.-. 1i° PERCENT SULFUR
3/77
! i ? TLT A •
p c v7L" K ctrr
UhUCC M;,..SFICLU '10. i:
"5775"
CKL.MC.O
COAL _ i».j PLKCENT SULFUR
UL\HHi,, rLi-'Cl K I'tT
EUorsTor,t t-o i
"G"U»sO";i r.o.'?"
"SCCTTiLTn C».i7
650
KAG'.-fc.i>iof, OXIDE
I-.OT SELLC7LD
' Lir'.L/Ll>'.LS ICN:L
"COAL 2T5~P i K cT N r~s 0 LF u R~
COAL
12/71*
TTb'iT
' 0'IL VROuuC
SCKUEBIiJG
-------
TAbLil 1.2
Ft Li .SYSTO.S UMIEK cowslTUcTToN AS OF 10/71*
UTILITY" CONPAHY
POWER STATION
b'k SIZE: OF'TGU
HLTROFIT UNIT JHW>
CHAKACTERISTICS
MO/YR
SOUTllWESTEKN PUbLlC SLuVlCL
iNETon NO i
_SPKINGFI CLP UTILITY
S"OUTnWE".sT NO 1
AUTHORITY
WIOQkS CHEEK NO
COf-bUil lOi'j
f: SCKUbBUvG
b50
U(vIVE.«SaL_OIL _
LI.v.tSTON£: SCRUBBING
CO«L 0.5 PtKCENT SULFUR
COAL
VALLEY AUTHOKITY
SCRUBBING
COAL 3.7 PERCENT SULFUR
6/76
6/76
1/77
-------
13
~PLAI."f FGD VLt^uOK/PKocEss
FUEL CHARACTERISTICS
KO/YR
CONTRACTS
LOU ISV H LE f\flS..8__LLLC_TKl_C_
CA;-JE RuY. jjjo "•»
178
. ,lR F.1LTL'H_
i>CHUbtiI!'J&
_CO.AL____5,.b-!«..g5__P_ERCENT. SULFUR ..6/75
COLUi-ipUi> a. SOUTHEMv OnIO
CCr-iLSVlLLE_N9_.5
37b UlilVLrtiiAL OIL PRODUCTS
LJME. SCRUbbING __
sou THERM &hio
7ii' 6
J7b UlvIVtriSflL OIL PROpucTS
LIME.
COAL
COAL
1/76
1/76
LOUISVlLLt GAS S LLE.CTIUC
;-'ILL CKC;:* :-.-o 3
1R FILTLK
.
CitOLLA i.O 2
2bO
COAL 3.b- ^.0 PE1KCE.NT SULFUR 6/77
_COAL 0._H^ PE_RCEhlT...SULFyR
TEXAS UTILITIES
KtbLAiTIOI(
COAL
6/76
GLNlf'.AL PUi-LlC _Ujj'.n IF S
bbO
NOT SLLECTLD
CO»L 2.8 PERCEfJT SULFUR Jt/77
-------
T/MiLL 13
PLANNLU rsii SYSKNS AS OK 10/74
POUPR STATION
NEW OR SIZE OF FGO VENDOR/PROCESS
KUEL CHARACTERISTICS
KO/YR
ARIZONA PUbLlC SERVICE
__Q HO L L A_jj 0 .3.
250
NOT SLLEC1EO
SCRUHBINr, __________
COAL o.1*'* PERCENT SULFUR
6/78
C L NT_R AL ILL J.N0IS CUULlC SCRV
ME.WTOW IvO.l
_600
lyOT_ SLLEC1EO
COAL 2.8-3,2..PLRCE:NT_.SULFUR _ 3/77 ._.
1N01ANAMOLIS fOWEH ANf) LIGHT
t'ETLRSt'UKG MO 3
515
NOT SLLLCILO
LIflL/LlMLbTOME: SCKUDUIMG
COAL 3.0-3.5 PLRCENT SULFUK
H/77
_ KANSAS POh'l.H S LIGHT f. 70g COMiUSTIUU
" ~JEf-Fr.RY 'IIP. 2" ' " "" 'LIE-.LSIONC s
.COAL
. 6/79
MISSISSIPPI PWH COOP
HATT ICSBU.RG ..MC . 2
PUB SERV/
223
NOT Sc-LL'ClLO
. SCRUBBING
COAL
6/79
N 140 f.OT SLLLCTLD . . COAL 1.0. PERCENT SULFUR ...5/77.
L I'1£/L IKEil ONE
TEX«.S UTlLlTlF S
N5
0 _ T*AS UTILIJirS
" fJ LAKE NO.'s"
793
KESLAnCH COTTKLLL
_793
RESEARCH CGHKLLL
'L SCRUublNG
COAL O.H4 PERCENT SULFUR
COAL 0.44 PERCENT SULFUR
12/79
12/78
SOUtl.r-.\ i'lSilSSlf-PJ P^H LOOP
.S UTILITIES
f;~0 ~3
Ivul i>LLCC~!LD
LlfLbTO.'ab: SCKUl'BING
COAL
600
_IJOT SELtCTLU
LiNESioNL SCRUUB'ING
COAL O.b- 1.0 PERCENT SULFUR 12/78
CPNSILLKlr.G F
. L&Uisy iLLE r.a_s z._i LI cjnu_c_
Chi-iH h'Ui, MO 1
_11C kuT StLECTLU
Lir''E SCKUbblNG
COAL 3.5- 4.0 PERCENT SULFUR 6/60
COLORADO
CCLuliAijt: l.'U. fLfCT
'Sl'/iT'jcV;' f.O.i"
fiOT SLl.L'CIEU
Llf'LSUiiL SC
;-.OT SLLLCFLfJ
COAL o.«*b PERCENT SULFUR
COAL 0.45 PERCENT SULFUR
6/78
6/78
COLORADO L'U
CRAIG STATION fM0.2
.-.TI (•AS_AU,_LLf_CTRlC_
'EAST DE-'."ij"r".V."i
SLLLC1EO
'SLwLC TLD
COAL 0.45 PERCENT SULFUR 6/78
COAL O.b- 0.8 PERCENT SULFUR 6/79
-------
13
_ ______
PLANNLU FGU bfSTLHS AS OF 11J/71*
"OR SrZE~OFTG"D VlNDCR/PKOCESSFULL' CHARACTERISTICS HO/YR
STATION RETRGf-ir UtJIT
'A LLEC1HIC COOKi5 225 MjT StlLTCTL"BC0"«L O.b-
BIGBEE NO 3 LIML'SIOUE SCRUBBING
BASIN ELECTRIC N 550 NOT SELECTED COAL
HlaSOljFl BASTN'NO 2 ' LTMES1 0,'j£ SCRUBBING"
l3ASIii ELi-CIK 1C l'i 550
MISSOURI E/vMN NO 3 Ll^'LSlONE SCRUBBING
BASIN ELECTRIC N 550 r.uT SELECTED COAL 6/79
-MISSOOK !"~Q'AiriTr~N"fr~i LTMLSI ONC
!-AS f. LLLLTRit « fZ5 TUT SLuCCTCD"
MLL Cf'EEK r-0 4 LIKE
LOUISVILLE CAS & ELECTRIC K 330 IvCT SELECTLO COAL 3.5- "*. 0 PERCENT SULFUR
fTO~l L"1HE '!>cr;OT SLLECTEC) COAL 14.3 PERCENT SULFUR 8/78
? h rrs cun-vjco " covru o".s~PF>\cci'rr SULTJR i2'/7i6"
FCUK cOK-ifl-.S i.'O ? LOT
iA PUtiLjC SERVICE t< 229 ChEF'ICO COAL 0.6 PERCENT SULFUR 3/77
CCft'hEK? NO 3 " ' "Ll'i'iL """ " ""
: .• 0H I HLi<:. ~I T-. r > iA:«t Pub SEnVICL R T51; TTOT
15AILY -JC.7 NOT SttLCTED
-------
___ _
HLArif._.U~ F"(,U
S AS 01- l~J/7t
"UTILITY COfiHA'iY"
POVE.K STATION
IV'E.* OR " "SlZL Of FGD VEfcUdh/PKUCESS
RETROFIT UNIT (M*!>
FULL CHARACTERISTICS
MO/YR
PUb
uoo NUT SLLECTLD
N-OT SL.LCCTLO
. AKIZO..:A.. PUbl_..C .St_j..V.ICL.
FGUr< COh JI.HS MO H
__ K __________ 6QO
.CHEMLG
LIFE bC
COAL 3 PEKCtNF SULFUR
..COAL _u.<. PtKCtNf SULFUR
FLECTHIC
203
r:uT SLLECTLD
COAL 0.<4b PLKCEf.T SULFUR
O/ 0
1/77
6/78
-ANTI CI.FC SY-.Jt;.L _r< (,5^ ncT SLLLCTLD COAL 0.3 PLRCE.NT SULFUR o/ o
PCI..T l.o.'i " " " MOT SLLECli-O "
jfi.i PUiUlC SCKV1CF. H 17b Li-.LMCU
FOUR CQ^'-'ES .L.f!P_! L I'"'t. bC
COAL 0.0 PEKCEWT SULFUR
.
CGLSTFlC COOP
LIOnCL ;.-.) 2
SLLECILlJ
-'.Elv'"r!E. SCKW(.;>_KiG
r SE.LLLTLD
i.-.tS iO.;C i>CKUbb'l:»G
10/76
7/7Q
CO»L 3«b- M.O PEKCENT SULFUR6/80
COAL _o,7 PERCENT SULFUR • 0/79
6/77"
SLKVKC OF ULW f-iLXlCO
SA;J JUAI. MO.J
PL'i.LIC S.Crt
SA'j jL.'\
P.F_.(vL>'_j_'LE_X_lC_0_
sou
500
I.UT vSLLLClt-U
I:OT SLLECILO
:'IOME
-------
i a
IvE,~OR -—sm-OF FGiy VO\;>G,\/P,
-------
STATUS OF FLUE GAS DESULFURIZATION TECHNOLOGY IN JAPAN
Jumpei Ando
Faculty of Science and Engineering
Chuo University
Kasuga, Bunkyo—ku, Tokyo
The regulations on S02 emission have become increasingly stringent in
Japan, forcing industry to install flue gas desulfurization (FGD) units.
Major power companies have decided to install full-scale FGD plants for
utility boilers with capacities larger than 30°MW. All of the large plants
will use wet processes by-producing gypsum. The present paper will
comprehensively describe wet lime-limestone processes and double
alkali FGD processes. In addition, three new processes, the Mitsui
Miike limestone process, the Kurabo ammonium sulfate lime process,
and the Dowa aluminum sulfate limestone process will be described in
some detail.
125
-------
STATUS OF FLUE GAS LESULFURIZATION TECHNOLOGY IN JAPAN
I Foreword
The oil crisis of late 1973 and the serious inflation resulting from it
has affected considerably Japan's energy and desulfurization policies.
Efforts more strenuous than "before are to be made for the development of
atomic energy and the import of LNG and coal. Imported crude oil which
has supplied more than 70% of the total energy needs of Japan and has
been the major source of S02 emission is expected to continue increasing
at an annual rate of 7%» as against the 20-30% in the past. On the other
hand, construction of many plants for hydrodesulfurization and gasification
desulfurization of oil to reduce sulfur contents to below 0.3% which was
planned eagerly in early 1973 wa-s postponed or given up because of the
inflation—the investment cost which was about 30 million dollars for
a plant has nearly doubled.
The regulations on S02 emission have become even more stringent and have
forced industry to install flue gas desulfurizers. Major power companies
have finally decided to install full-scale S02 removal plants for
utility boilers with capacities larger than 300MW. All of the big plants
are to use wet processes by-producing gypsum because the reliability of
the processes has been well demonstrated and the demand for gypsum is
growing. Most of the large units are to be located outside city limits,
because even 90% removal of S02 may not be satisfactory in the future
in urban areas. Tokyo Electric Power, for example, has started to burn
a considerable amount of LNG which gives off no S02 although it is
expensive. Another reason for using LNG is the low NOX emission involved.
NOX has aroused serious concern in Japan recently because of photochemical
smog which has plagued big cities for the last few years. Simultaneous
removal of S02 and NOX has been enthusiastically studied by many
organizations although there seems to be much difficulty involved in
the process.
II Introduction
At the FGD symposium of EPA last year, the author reviewed seven major
processes in Japan, namely those of Mitsubishi-JECCO, Chemico-Mitsui,
Chiyoda, Wellman-MKX, Kureha-Kawasaki, Showa Denko and NKK.1' The
present paper will describe comprehensively the technology and problems
of wet lime-limestone processes, double alkali type processes and others.
Some detailed description will be given on three new processes which
might be of interest for application in the U.S.A., namely, the Mitsui
Miike limestone process, the Kurabo ammonium sulfate lime process, and
the Dowa aluminum sulfate limestone process.
126
-------
Ill Wet lime-limestone process
S02 removal plants using wet' lime—limestone processes with a capacity
larger than 20MW equivalent are listed in Table 1. The Mitsubishi-JECCO
process has been used most widely' for oil-fired boilers, iron-ore sintering
plants, etc. while the Chemico Mitsui and Mitsui -Miike processes have
been applied to coal-fired boilers. Five other processes have also been
used mainly for flue gas from oil-fired boilers. Many of the plants use
lime to obtain a high SC>2 removal efficiency—more than 90%—which is
required in many districts in Japan. Limestone scrubbing removes
85 to 90% of the SC>2 at 0.9 to 1.2 stoichiometry by using a scrubber
nearly twice as tall as that using lime.
A flow sheet common to most processes except the Chemico-Mitsui and the
Mitsui -Miike processes is shown in Figure 1. Types of scrubbers and
examples of operation parameters are listed in Table 2.
Cooler
Scrubber
tO°C
'.Vater After burner
r -•
y_ }20°C
Waste-
water
6-
j
1
Demister
Oxidizer
k
<-• f\r*. t- Y*f
il 1 or>
A
Centri
"Air
.-'liter
Neutralizcr
V
Gyosum
Figure I A Schematic of the wet lime-limestone r-rocoar-
Cooler Flue gas from the electrostatic precipitator is first led into the
cooler or prescrubber where the gas is sprayed with water. The cooler, which
is not usually used in the United States, has two functions. (1) Cooling and
humidifying the gas in order to aid in the prevention of caking in the
scrubber. (2) Removal of dust and other impurities in the gas which were
not caught by the electrostatic precipitator. This is useful in obtaining
high purity by-products with good commercial value.
127
-------
Table 1 Sulfur dioxide scrubbing installations in Japan (lime-limestone scrubbing)
00
Process developer
Mitsubishi- JECCO
n
n
11
n
it
it
ti
n
it
it
tt
11
it
it
it
it
it
11
n
Chemico-Mitsui
Mitsui- Miike
ii
11
Bahco-Tsuki shima
Babcock-Hitachi
11
Ishikawa j ima-TCA
Sumitomo-Fuji Kasui
Chubu-MKK
"
User
Kansai Electric
Onahama Smelter
Kawasaki Steel
Kansai Electric
Tohoku Electric
Tokyo Electric
Kyushu Electric
Kawasaki Steel
"
n
Kansai Electric
Teijin
Mizushima Power
Niigata Power
Kyushu Electric
"
n
11
Chugoka Electric
"
Mitsui Aluminum
ii
"
Elec. Pow. Dev.
Yahagi Iron
Chugoku Electric
"
Chichibu Cement
Sumitomo Metal
Ishihara Sangyo
Mitsubishi Gas
Plant site
Amagasaki
Qnahama
Chiba
Hainan
Hachinoe
Yokosuka
Shinkanda
Mizushima
n
Chiba
Amagasaki
Ebime
Mizushima
Niigata
Katsura
"
Ainoura
it
Oase
"
Omuta
M
tt
Takasago
Nagoya
Mizushima
Tamashima
Kumagaya
Kokura
Yokkaichi
"
a Actual for boilers and equivalent gas flow for
Absorbent
Ca(OH)2
tt
ti
it
n
CaCOj
Ca(OH)2
11
11
"
"
"
11
CaC03
11
M
"
"
Ca(OH)2
"
"
CaCOj
11
"
Ca(OH)2
CaCOj
"
Ca(OH)2
n
CaCOx
Ca(OH)2
others (at
A Vl
MW Type of plant
30 Utility boiler
29 Copper smelter
37 Sintering plant
150 Utility boiler
125 "
130 "
175
232 Sintering plant
279 "
130 "
125 "
83 Industrial boiler
192 Utility boiler
117 "
250 "
175
250 "
250 "
375
375
119° Industrial boiler
25° "
175C
250° Utility boiler
26 Sintering plant
104 Utility boiler
500 "
62 Diesel engine
32 Heating furnace
77 Industrial boiler
22 "
1900scfm per MW).
Year of
completion
1972
n
1973
1974
"
"
"
"
1975
11
"
"
11
11
11
1976
"
"
11
it
1972
1974
1975
1974
1971
1974
1975
1972
1974
"
11
Gypsum
(tons/day)
20
450
16
22
53
20
77
130
100
30
42
93
105
47
75
50
105
105
365
365
d
22
220
220
28
28
385
24
28
55
13
b Boilers are oil-fired unless noted.
c Coal -fired.
d Waste sludge of
calcium sulfite.
-------
Table 2 Example of operation parameters of FGD plants "by-producing
gypsum and calcium sulfite
Process developer
Wet lime-limestone
Mitsubishi-JECCO
it
Chemico-Mitsui
Kitsui Miike
Babcock-Hitachi
Chubu-MKK
I shikawaj ima
Sumitomo-Fuji
Absorbent,
precipitant
(stoichiometry) lj
process
Ca(OH)2 0.9-1
CaCO^ 0.9-1
j
Ca(OH)2 1-1.05
CaCO* 0.9-1
x
CaCOj 1-1.2
CaC05
Ca(OH)2
Ca(OH)2 1-1.2
Capa-
city
Type of
Slurry or
solution
.OQOscfm absorber pH conv.%
210
230
210
44
180
150
60
37
GPa)
GP
Venturi
Venturi
ppb)
Screen
TCA
ppb)
5-6
7
6
6.2
6
6
15
12
3-5
5
7-8
10
2
10
L/G Space
gal/ velocity
L,000scf ft/sec
30-50
50-70
80-100
80-100
50
70
50
70
10
10
12
12
10
12
pressure
drop S02 Ppm
inches in
6
5
16
20
3
18
1,000
250
2,000
2,000
500
1,500
700
1,000
out
80
20
200
200
60
200
50
50
Moisture
% of
gypsum
10
8-10
10-15
8-10
10-12
10-15
10-12
Indirect lime-limestone process
Kureha-Kawasaki
Showa Denko
Nippon Kokan
Chiyoda
Kurabo
Dowa
Hitachi-Tokyo Elec.
Na/jSOzjCaCO;,
Na2S03,CaC03
(NH4)2S03,CaC03
dil.H2S04,CaC03
(NH4)2S04,Ca(OH)2
Al2(S04)3,CaC03
Carbon, CaC03
230
255
90
466
3
82
250
GP*)
Cone
Screen
Tellerette
Tellerette
Tellerette
Packed
7
6.8
6
1
3-4
3-4
20
25
30
2-4
10
10
7-15
7-15
14
200-300
40-70
20-60
10
3
6
4
1.5
8
10
2
4
19
800
1,400
700
600
1,500
600
500
20
40
30
50
80
20
80
5-6
7-8
10
6-8
8-10
10-12
10-12
a) Grid packed b) Perforated plate
Pressure drop includes that of absorber and mist eliminator
-------
Scrubber The type of scrubber is a plastic grid packed tower for
Mitsubishi-JECCO, a venturi for Chemico-Mitsui and Mitsui Miike, a TCA
with polyethylene balls for Ishikawajima, a Bahco type for Tsukishima,
a stainless steel screen type for Chubu-Mitsubishi Chubu-MKK (CM), and
perforated plates for Babcock-Hitachi and Sumitomo-Fuji Kasui processes.
The reliability of the Mitsubishi-JECCO and Chemico scrubbers has been
well demonstrated in Japan. The former has less pressure drop. The
TCA scrubber has a high S02 removal efficiency but presents the problem
of wearing of balls. The Bahco and screen type scrubbers might be more
susceptible to scaling. The perforated plate scrubber for the Sumitomo-
Fuji Kasui process (Moretana type) is designed to give extreme turbulence
producing foam layers 15-20 inches thick; a high SC>2 removal efficiency
is attained while heavy mist is formed.
For the Chemico-Mitsui and Mitsui Miike processes,which use no cooler,
the scrubber is operated with a dilute slurry at a large L/G ratio to
prevent scaling. The by-product contains a considerable amount of
impurities as will be described in the author's other paper at the
present symposium.2)
Water balance In most plants, the mist eliminator and by-product gypsum
are washed with fresh water. The wash water is then fed mainly into the
scrubber system and partly into the cooler. Some fresh water is also fed
into the cooler. The amount of total input water more or less exceeds
the output, i.e. water volatilized and taken into gypsum as water of
crystallization and moisture. Therefore, a portion of an acidic liquor
from the cooler is sent into the water—treatment system, neutralized
with lime to precipitate heavy metals and other impurities, filtered,
and then purged. The water balance is thus maintained. The use of
fresh water and the purge of some wastewater help in the prevention of
caking and corrosion.
It would be possible to apply a closed loop which would release no water
from the system and yet not cause scaling. The accumulation of chlorine
in the system, however, would cause corrosion problems. Also, the use
of highly resistant materials would add considerably to the cost. Tohoku
Electric Power Co. first planned to use a closed system for a 125MW plant
based on the Mitsubishi-JECCO process, but the investment cost increases
with the closed system; Tohoku finally changed the plan so that some
water is purged after being treated.
Oxidation The rotary atomizer invented by JECCO has been used for
the Mitsubishi-JECCO, Babcock-IIitr.chi, a;ic Chubu-MKK processes because
of its high efficiency and reliability. Different types of oxidizers
have been developed for other processes. Low pH (5«5~4«0) is desirable
for oxidation of calcium sulfite slurry. This is attained with a pH
controller by adding a small amount of sulfuric acid as is being done
at many plants, or by using flue gas and a catalyst as in the Mitsui Miike
process. The facilities to produce gypsum, pH controller, oxidizer and
centrifuge, occupy about 40}o of the total investment cost. Nevertheless,
the production of gypsum has been growing in Japan, Mitsui Aluminum Co.
130
-------
Table 3 Indirect lime-limestone processes (double alkali type)
\—I
LJ
Process developer
Nippon Kokan
Chiyoda
«•
ii
n
»
"
«
"
Kureha-Kawasaki
»
n
Showa Denko
»
"
Showa Denko-Ebara
"
n
11
n
"
11
Tsukishima
"
Kurabo Engineering
"
D
-------
which has been producing waste calcium sulfite sludge will also change
the process to by-produce gypsum as will be described later. By-
production of calcium sulfite sludge and its stabilization as it is being
done in the United States might be of much interest in the future in
Japan when there is an oversupply of gypsum.
IV Indirect lime-limestone process (double alkali type)
Many double alkali type processes that use lime or limestone as a
precipitant have been developed in Japan including those which use an
acidic solution or acid as the absorbent. The Hitachi-Tokyo Electric
process uses activated carbon as the absorbent and limestone as the
precipitant. All of those processes are classified in the category of
"indirect lime-limestone process". The installations with a capacity
larger than 20MW equivalent are listed in Table 3« The operation
parameters are shown in Table 2.
Wet process The various processes use different absorbents; a sodium
sulfite solution is used for the Showa Denko, Kureha-Kawasaki, and
Tsukishima processes, an ammonium sulfite solution for the KKK process,
an acidic ammonium sulfate solution for the Kurabo process, an aluminum
sulfate solution for the Dowa process, and a dilute sulfuric acid with
iron sulfate for the Chiyoda process. The pH of the solution is 6 to 7
ammonium and sodium sulfites, 3 "to 4 for ammonium and aluminum sulfates
and 1 for sulfuric acid. The L/G ratio is 7-15 (gal/1,OOOscf) for the
solutions with pH 6-7, 20-70 for the solutions with pH 3-4» and .200-400
for the acid at pH 1. The more acidic the solution is, the less the
S02 absorption capacity, the lesser the problem of scaling, and the easier
the reaction with limestone. Limestone can be reacted with a sodium
bisulfite solution, as in the Showa Denko and Kureha-Kawasaki processes,
but the reaction occurs slowly requiring large reaction vessels. Lime is
used for the Tsukishima, NKX, and Kurabo processes.
For the Chiyoda, Dowa, and Kurabo processes, the liquors which absorbed
S02 are contacted with air to oxidize SOj '. uto 864,"*"" Lir-esume
or lime is then added to precipitate gypsum. For other processes,
limestone or lime is added first to precipitate calcium sulfite which
is then oxidized into gypsum. By the double alkali type process,
gypsum usually grows in larger crystals than with the wet lime-limestone
process. Moisture content of the by-product gypsum after centrifugaliza-
tion ranges from 6 to 12% as compared with 8-15% f°r the wet lime-limestone
process.
The liquor from the gypsum centrifuge is returned mainly to the scrubber
system. Softening of the liquor which is usually needed to prevent
scaling, is not necessary when an acidic solution at a pH below 4 is used.
132
-------
At most plants, a small portion of the liquor is purged to maintain
the concentrations of chlorine, magnesium and other impurities under a
certain level. Calcium sulfite obtained "by the slow reaction of limestone
and sodium bisulfite grows into fairly large crystals which are not
difficult to handle,2; On the other hand, lime reacts rapidly with a
sodium bisulfite solution to give very fine crystals of calcium sulfite
like those prodiiced by the wet lime-limestone process,
Carbon absorption Another type of the indirect lime-limestone process is
dry activated carbon absorption which is used in the Hitachi-Tokyo Electric
process. The carbon which has absorbed S02 is washed with water to
give a dilute sulfuric acid of 15-20% concentration. The acid is treated
with powdered limestone to produce gypsum, which is centrifuged to 10-12^
moisture. The process requires large absorption towers because low gas
velocity (l,5 ft/sec) is used to obtain 85-90% removal of the S02«
Hitachi has recently operated a If700scfm pilot plant in which S02 is
absorbed with a carbon slurry. The resulting sulfuric acid is treated
with limestone to obtain gypsum.
V Other processor
There are about one hundred relatively small sodium scrubbing plants in
operation by-producing mainly sodium sulfite and some sodium sulfate,
Those processes were reviewed last year!) and there has not been much
progress since then. Major plants using other processes are listed
in Table 4,
Wellman-Lord process Many plants using the Wellman-Lord process have
been constructed by Mitsubishi Chemical Machinery Co, (MKK) and also by
Sumitomo Chemical Engineering Co. (SCEC), The reliability of the process
has been well demonstrated, A main problem in the process is the oxidation
of sodium sulfite into sulfate which does not absorb S02. The sulfate
has to be removed from the system, resulting in the loss of sodium and the
need for wastewater treatment. At Nishinagoya Plant of Chubu Electric
(220MW), 7 tons/day sodium hydroxide have been used for make-up giving the
equivalent amount of sodium sulfate, SCEC uses an oxidation inhibitor to
reduce the sulfate formation to below half. The inhibitor, however, might
give some trouble in wastewater treatment by disturbing the oxidation of any
reducing component in the water.
Magnesia scrubbing Two plants are in operation and one is under construc-
tion with different magnesium scrubbing processes, by Mitsui Mining,
Onahama-Tsukishima, and Chemico-Mitsui, A common feature of these
processes is the production of large MgS04'6H20 crystals (200 microns
or so) which are much easier to filter and dry than are the small
MgS04»3H20 crystals. Both the Mitsui Mining and Onahama plants have no
problem in the filtration and drying steps which were troublesome at Boston
Edison Co. A nearly completed Idemitsu Kosan plant based on the Chemico-
133
-------
Table 4 Sulfur dioxide scrubbing installations in Japan which by-produce
sulfuric acid and sulfur
Process developer Absorbent TTser
Wellman-MKK Na9SO,
^ j
n it
it ti
n it
ii ii
n 11
it n
it »
n n
Wellman-SCSC "
n n
(-•
£ n „
it ti
n n
Onahama-Tsukishima MgO
Mitsui Mining "
Chemico-Mitsui "
Sumitomo Shipbuilding Carbon
Shell CuO
Mitsubishi-IFP (NE^^SO:
TEC-IFP «
Japan Synth.
Rubber
Chubu Electric
Japan Synth.
Rubber
Toyo Rayon
Mitsubishi Chem.
Company K
Kurashiki Rayon
Company KKK
Company SKD
Toa Nenryo
Sumitomo Chiba
Chem.
Fuji Film
Sumitomo Chiba
Chem.
Sumitomo Chem.
Onahama Smelter
Mitsui Mining
Idemitsu Kosan
Kansai Electric
Showa Yokkaichi
* Maruzen Oil
Fuji Oil
Plant site
Chiba
Nishinagoya
Yokkaichi
Nagaya
Mizushima
Kawasaki
Okayama
Kashima
Yokkaichi
Kawasaki
Chiba
Fuji
Chiba
Niihama
Onahama
Hibi
Chiba
Sakai
Yokkaichi
Shimozu
Chiba
MW*
70
220
150
103
186
217
127
400
124
25
120
50
180
50
28
25
162
53
37
14
3
Year of
By-product
Type of plant completion (tons/day")
Industrial boiler
Utility boiler
Industrial boiler
"
ti
"
it
n
n
Glaus furnace
Industrial boiler
ti
n
n
Copper smelter
1*2804 plant
Glaus and boiler
Utility boiler
Industrial boiler
Glaus furnace
"
1971
1973
1974
it
it
11
it
11
it
1971
1973
1974
1975
11
1972
1971
1974
1971
1974
"
"
H2S04
11
"
n
n
11
11
it
n
S
H2S04
Liquid S02
H2S04
n
it
ii
S
H2S04
S
S
S
44
88
44
50
88
110
66
120
60
55
13
88
25
240
18
17
a Actual for boilers and equivalent gas flow for others. Boilers are oil fired.
-------
Mitsui process is to return the recovered S02 to a Glaus furnace for
sulfur production; the other plants by-produce sulfuric acid.
Dry carbon process By the Sumitomo Shipbuilding process, the S02 absorbed
on activated carbon is expelled by heating it in a reducing gas, to release
S02 gas of 10-20% concentration which is used for sulfuric acid production.
Moving beds are used for both absorption and desorption. The operation
has been carried out smoothly but there is no further plan to build
larger plants, presumably because of the requirement of large absorbers and
the considerable consumption of carbon.
Shell process In the Shell process, S02 is absorbed with copper oxide to
form copper sulfate, which is then reduced with hydrogen to expel S02«
A commercial plant started operation several months ago at Yokkaichi.
The process seems fairly costly requiring a large absorber and hydrogen
although it has the advantage of a dry process which requires no
reheating of treated gas.
Ammonia scrubbing—IFF process The Mitsubishi-IEP process uses ammonia
scrubbing with the thermal decomposition of ammonium sulfite and sulfate
to regenerate S02, which is then reacted with H2S in an IFF reactor to
produce sulfur. A plant at Shimozu started operation several months ago.
A similar plant has been recently constructed by Toyo Engineering for
Fuji Oil. The process is not simple and may need further improvement
before satisfactory operation is attained.
VI Mitsui Miike limestone gypsum process
Outline of process This process has been developed by Mitsui Miike
Machinery Co. (.1-1, 2-chome, Nihonbashi Muromachi, Chuo-ku, Tokyo).
302 is absorbed with a limestone slurry containing a catalyst which
promotes the reaction. The resulting calcium sulfite slurry is oxidized
into usable gypsum by further contact with flue gas and air. The catalyst
also promotes the oxidation.
State of development Mitsui Miike constructed the desulfurization unit at
the Omuta plant of Mitsui Aluminum Co. which has been operated successfully.
The disposal of calcium sulfite sludge, however, is a considerable problem
because of the poor nature of the sludge. In order to by-produce usable
gypsum, Miusui Miike has developed a new process based on a patent by
Dr, S* Akimoto using a catalyst that promotes both the reaction of S02 and
limestone and the oxidation of calcivm sulfite. After tests with a pilot
plant with a capacity to treat coal-fired flue gas at 2,500NmVhr
(l,500scfm), a larger test unit (75»OOONm5/hr of coal-fired flue gas)
was completed in September 1974 at Omuta. A commercial plant at the
Takasago Station of the Electric Power Development Co. with a capacity
of treating 800,OOONm3/hr flue gas from a coal-fired utility boiler
(250MW) will be completed in late 1974. Mitsui Aluminum will install a
552,OOONm3/hr unit (175MV) in 1975-
135
-------
!r_oc.ess description A flowsheet of the process is shown in Figure 2.
•'lue gas is treated in two venturi scrubbers in series with a counter-
current flow of limestone slurry containing a metallic catalyst. The
resulting calcium sulfite slurry is then fed into a pH controller where
the slurry contacts a portion of flue gas'(5-10% of the total) to lower
the pH and to complete the reaction of limestone with S02. The slurry is
then pumped into an oxidizing tower into which air is blown from the
bottom to convert the sulfite into gypsum. The gypsum slurrv is centrifugec
Most of the liquor from the centrifuge is used to prepare a limestone
slurry. A small portion of the liquor is sent to a wastewater treatment
system to prevent the accumulation of impurities. Gypsum grows into large
crystals of 50-200 microns in size and can be used for wallboard and cement.
The catalyst prevents the formation of a calcium sulfate coating on
calcium sulfite and carbonate and thus promotes both the reaction of the
carbonate with S02 and the oxidation of the sulfite into sulfate as shown
in Figure 3 and Table 5 which are based on the pilot tests. There is
virtually no loss of catalyst into gypsum because gypsum can be washed
well. In the above-mentioned wastewater treatment, the catalyst can be
precipitated in the form of hydroxide by raising the pH and returning
it to the absorbing system.
The capital required is about 50/6 higher for the lime-calcium sulfite
sludge process. The pH controller and oxidizer add about 20% to the
cost. If about 80% oxidation is satisfactory to improve the settling
property in the disposal pond, as in the partial oxidation of the sulfite
sludge, then the oxidizing system can be greatly simplified. The pond
should be sealed to prevent the leakage of water containing the catalyst.
Table 5 Example of the composition of by-product (%)
(stoichiometry 0.95)
Absorbent
QQ p
CaCO} -f Catalyst J
CaC03 only 71.0 15.3 12.3
Advantages High S02 recovery is attained with limestone slurry.
Oxidation of calcium sulfite occurs rapidly. S02 in flue gas is used
for pH adjustment instead of using sulfuric acid.
Disadvantage The facilities for oxidation add considerably to investment
cost.
136
-------
Flue ",", scrubber
After burner
1
1
Mill
iD^i U
-, '
i,v
I U UtJI
(-•
1
I
1 1
1
1
* 1
1
.-,
1
r y
V
( — •
1
1
l£
iitack
Tank
i
^
^
^
pH
controller
TOxidizer
Air
Gypsum
Figure 2 Flowsheet of Mitsui Miike limestone gypsum process
o
5 90
Q)
(M
O
0.9 1.0 1.1
StoichiometryC for inlet
1.2
Figure 3 Effect of catalyst on
removal efficiency
137
-------
4)
VII Kurabo ammonium sulfate gypsum process
Outline This process has been developed by Kurabo Industries, Ltd.
(2-41, Kitakyutaromachi , Higashi-ku, Osaka). S02 is absorbed by a
slightly acidic ammonium sulfate solution at pH 5-4. The solution is then
oxidized by air to give an acidic ammonium sulfate solution, which is then
treated with lime to precipitate gypsum and to recover aqua ammonia which
is returned to the absorbing system. Ammonium sulfate can be produced,
if desired.
State of development Kurabo has built several small ammonia scrubbing
plants in addition to many sodium scrubbing plants. Plume formation,
the greatest problem for the ammonia process, has been fairly well solved
by the use of a cold, dilute solution but this method may not be suitable
to larger plants. Kurabo has recently developed a method to use an
acidic ammonium sulfate solution as an absorbent to eliminate the plume
and has successfully operated a 2MV equivalent test unit. Two commercial
plants are presently under construction (Table 3).
Theory of absorption with ammonium sulfate solution Plume can be
eliminated by the use of an acidic absorbing liquor because the vapor
pressure of NH3 is less than Ippm with a solution at a pH lower than 4.
The acidic ammonium sulfate solution has a larger capacity of S02
absorption than plain water or a saturated calcium sulfate solution as
shown in Figure -t because of the smaller pH drop due to the following
equilibrium in the ammonium sulfate solution:
The minimum L/G ratio (liter/Nm?) required to remove 95% °f the S02 is
shown in Table 6. The relationship of L/G to S02 removal efficiency is
shown in Figure 5« Basic data for the design of S02 absorbers
obtained through the pilot tests are shown in Table 7»
Table 6 Inlet 302 concentration and minimum L/G
(gal/1, OOOscf, 60°C, 95% S02 removal)
_ Absorbing liquor _ Inlet S02 concentration, ppm
Composition Initial pH 1,000 2,000 3,000
CaS04'2H20 - H20 7.0 98 148 191
0.25 mol/1 (^4)2804 3.5 43 51 66
0.5 " 5.5 36 41 50
1.0 " 3.5 29 38 44
138
-------
VO
NH^Cmole/liter)
302 + HSO," ( mole/liter)
Figure
Relationship of 302 + HSC5
concentration to partial pressure
of S02 ( 60°C)
100
o
c
o>
•H
O
•H
90
o
E
0)
O
CO
80
30
L/G
Inlet S02
•
O
O
550ppm
BOOppm
1 ,000ppm
PH 3.7,60°C
50 70
sal./l,000scf)
Figure 5 ^Oa removal efficiency
at the test unit (2,950scfm)
Tower diameter 3.3ft
Height of packing 6.6ft
Packing Netring HA-18
Space velocity 4-5 ft/sec
-------
Table 7 Design base of S02 absorber
Pressure Space Height of
L/G drop velocity packing
Packing (gal/1,000scf) (inches Aq) (ft/sec) (ft)
Tellerette L type 56 4 6 12
Netring HA-lb 56 4 4.5 9
Process description The flowsheet of the process is shown in Figure 6.
Flue gas is first led into a KBCA scrubber and then into a packed tower
absorber. The main function of the KBCA scrubber is to cool the gas to
60°C and to concentrate the absorbing liquor (ammonium sulfate solution).
More than 90% of the S02 is removed. The gas is then reheated by the
afterburning of oil. The liquor from the packed tower absorber is sent to
the KBCA unit, concentrated, and then led into an oxidizer. The pH of the
liquor in the oxidizer is adjusted to from J to 4 by adding dilute aqua
ammonia and the sulfite in the liquor is oxidized into sulfate by small
bubbles of air formed by introducing a jet stream of circulating liquor
accompanying air into a pool of the liquor. About five times the stoichio-
metric amount of air is used. Tests have shown that Fe++ catalyst gives
an optimum oxidation rate at pH 4 and Mn at pH y (Figure 7). Actually
no catalyst is added for oxidation because it occurs fairly rapidly at a
rate of 0.8-1.Okg mole/m^/hr? presumably a small amount of vanadium and
iron derived from the fuel helps the oxidation.
Most of the liquor from the oxidizer is returned to the absorber, and
a portion is sent to a set of three reactors where the liquor is treated
with milk of lime to precipitate gypsum. In order to raise the concentration
of the slurry and to increase the retention time of gypsum for better
crystal growth, a portion of the slurry is sent to a liquid cyclone.
The gypsum is returned to the reactor and the liquor which contains some
gypsum is sent to a thickener. The overflow from the thickener is sent
to an aqua ammonia tank. The sludge from the thickener is returned to
the reactor, and the other portion of the slurry is sent to a centrifuge.
The liquor from the centrifuge and wash water are sent to an aqua ammonia
tank, and the aqua ammonia js sent to the oxidizer.
The concentration of the absorbing liquor is normally about 0.5 mole of
ammonium sulfate per liter at L/G 56. It can be raised to 2.5 moles at the
same L/G ratio without appreciable decrease of S02 removal efficiency.
The oxidation rate decreases slightly at the higher concentration. The
amount of water added to the system, namely lime dissolving water
and gypsum wash water, is usually less than that volatilized in the cooler.
Therefore, some water is introduced into the cooler to maintain the water
balance. Occasional purging may be required to prevent buildup of chloride.
The by-product gypsum contains about 9% moisture with a bulk density of
about 1.06 and pH 8. Average crystal size is 40 to 60 microns. The
gypsum can be used for both wallboard and cement additives.
140
-------
Flue gas
I
V.'ater
Cleaned gas
A
Ca(OK)2
>Vater
Air
nin
j
i f ^» /
' 1 \ I
i
.Vater
Gypsum
1_T^
8
1 K3CA scrubber
2 Cxidizer
3 Absorber
k Slaking tank
5 Seactor
6 Liquid cyclone
7 Centrifuge
3 Thickener
9 Aqua ammonia tank
Figure 6 Flowsheet of Kurabo ammonium
sulfate gypsum process
( mole/liter)
Ki-jurc 7 Oxidation rate of
sulfite ion ( W°C)
(HSO/) < 10^ mole/liter
( M++) ^xlo'*mole/liter
-------
Ammonium sulfate can be obtained easily by concentrating the solution
from the oxidizer. In this case, a nearly saturated solution (2.5-3
moles/liter) is used as the absorbent.
Cost Estimation An example of cost estimation is shown in Table 8,
Table 8 An example of cost estimation
59,000 scfm
1,590,000 gal/year
2.8%
8,500 hours/year
54.0 million
Plant:
Oil consumption:
Sulfur content:
Operation hours:
Construction cost:
Material
25% aqua ammonia
Oil for reheating
Industrial water
Power
Quick lime
Recovered gypsum
Subtotal
Depreciation
Interest
Labor
Maintenance, etc.
Total
Desulfurization cost
(l dollar =250 yen)
Unit cost.
30.3 cents/gal.
22.7 "
7.6 "
2.0 cents/kW
0.9 cents/lb.
0.4 "
Annual cost
6,000 dol.
112,400 "
1,680 "
96,000 "
63,000 "
-114,600 "
162,800 "
515,000 "
160,000 "
24,000 "
64,000 "
925,800 '•
5.9 cents/gal.oil
Advantages There is no plume and no scaling. Both oxidation and crystalli-
zation of gypsum are made at 50-60°C without external heating. The pH of
liquor is kept over 3 to prevent corrosion. All steps are carried out at
atmospheric pressure. A closed system can be achieved. Ammonium sulfate
can be produced easily, if desired. The softening step of the circulating
liquor is unneeded.
142
-------
Disadvantages The process is less simple than the wet lime process.
Limestone cannot "be used.
VTII Dowa aluminum sulfate process
Outline of the process This process has been developed by Dowa Mining Co.
(8-2, 1-chome, Marunouchi, Chiyoda-ku, Tokyo). Dowa is one of the biggest
nonferrous metal manufacturers in Japan and owns many smelters and sulfuric
acid plants. The aluminum sulfate process has been developed to
desulfurize waste gas from smelters, roasters and sulfuric acid plants.
The principle of the process is as follows:
S02 is absorbed in a solution of basic aluminum sulfate,
to form 112(304) ^12(503)3. The liquor is oxidized by air into
Al2(SO^)^, which is then treated with powdered limestone to precipitate
gypsum and to regenerate the basic aluminum sulfate solution.
Absorption: A^SO^'A^O^ + 3S02 =
Oxidation: A12(S04)5«A12(S05)5 + 3/202 =
Neutralization: A12(S04) ^'A^CSO^j + JCaCO^ + 6H20
3(CaS04«2H20)
State of development After tests with a pilot plant with a capacity of
lyOscfm, a small commercial plant (2,000scfm) was constructed at Taenaka
Mining, Mobara Works and started operation in October 1972 to treat waste
gas from a molybdenum sulfide roaster containing 7»500ppm S02 at 100 °G.
A larger pilot plant with a capacity of treating l,700scfm tail gas from
a sulfuric acid plant has also been in operation.
Two commercial units, each with a capacity of treating 140,OOONra'/hr
(82,000scfm) tail gas from sulfuric acid plants, started operation in
June 1974 at Okayama Works of Dowa. Mitsui Shipbuilding Co. has joined
Dowa for further development of this process.
Absorbing liquor The relations between the composition of basic aluminum
sulfate liquor (Table 9) and S02 absorbing capacity and the boiling point
of the liquor after the S02 absorption are shown in Figures 8 and 9»
Table 9 Composition of solutions in Figures 8 and 9
Al20j (grams/liter) in solution
No,
I
II
III
143
Free
34.4
40.0
45.5
Combined
66.1
58.2
53-2
Total
100.5
98.2
98.7
Basicity (%)
34.2
40.8
46.1
-------
120
120
IIA
113
0 2 ^ 6 8 10
S02(f/0 in waste gas
Figure 8 302 absorbing capacity
of the liquors
80 100
content in 1 liter
of liquor (gram)
Figure 9 Boiling point of
the liquor containing
Cleaned gas
A
1 Absorber
2 Tank
~5 Oxidizer
k- Reactor
5 Thickener
6 Tank
7 Centrifuge
8 Tank
Figure 10 Flowsheet of Dowa aluminum sulfate limestone
process
144
-------
An optimum concentration as well as basicity of the absorbing liquor is
selected, according to the SC>2 concentration of gas and the removal
efficiency required. Normally liquors at pH 3 to 4 which are more
dilute than those shown in Table 9 -are used.
Process description The flowsheet of the process is shown in Figure 10.
For the small commercial plant of Taenaka Mining, a TCA scrubber is used
to recover more than 90% of the 302 at L/G 35 (gal/1,OOOscf). For the
pilot tests (l,700scfm) with tail gas at 85°C from a sulfuric acid plant,
a packed tower (800mm in diameter) was used to recover more than 95% of
the S02 at L/G 20. The liquor from the absorber is led into a tank to
which make-up aluminum sulfate is added. The liquor from the tank is
sent to an oxidizer where aluminum sulfite is oxidized into sulfate by
small bubbles of air. The oxidizer, which has no moving parts, has been
developed by Dowa. Twice the stoichiometric amount of air is used.
Most of the oxidized solution is returned to the absorber, a portion is
sent to a neutralizer and treated with powdered limestone (mostly under
200 mesh in size) under a condition suitable for the crystal growth of
gypsum.
The slurry from the neutralizer passes through a thickener and is then
centrifuged. All of the liquor from the thickener and the centrifuge
as well as the wash water of gypsum are returned to the absorber. The
chemical composition of the by-product gypsum from the pilot plant
(l,700scfm) is shown in Table 10. A small amount of aluminum is contained
in the gypsum. The consumption of aluminum is about 1 Ib (as Al) for
1 ton of gypsum.
Table 10 Chemical composition of by-product gypsum (%)
No. CaO 803 Al Moisture
1 32.80 46.65 0.05 10.0
2 32.34 46.43 0.05 12.1
3 32.03 46.61 0.04 12.0
4 32.24 45.68 0.06 10.8
5 32.88 45.86 0.05 12.8
6 32.07 46.48 0.05 11.8
Operation experiences The small commercial plant of Taenaka has been in
operation since October 1973* There is no appreciable problem of scaling.
Cleaning the pipings once every several months is enough to remove soft
deposits. Continuous operation of the pilot plant (l,700scfm) for several
months also produced no appreciable scaling. The operation of the new plants
(82,000scfm) has also been satisfactory.
145
-------
Advantages (l) Limestone is used for neutralization because of the low
pH of the liquor. (2) The L/G ratio is relatively small. (3) The
oxidation occurs fairly rapidly. (4) There is essentially no scaling and
no wastewater. (5) A fully automatic operation can be achieved,
Disadvantage Aluminum sulfate is consumed although in a small amount.
IX Evaluation of the processes
Dry processes Now there are three dry processes, each being used for a
prototype plant. Those are two activated carbon processes by Hitachi-
Tokyo Electric and by Sumitomo Shipbuilding, and the Shell copper oxide
process. In addition, pilot plants using the dry sodium absorption
process followed by the decomposition of the resulting sodium sulfate
has been operated by Tsukishima Kikai Co. and also by the National Research
Institute of Pollution and Resources. There is no plan, however, to install
a commercial plant using any of those dry processes, Mitsubishi Heavy
Industries and Chubu Electric Power have recently given up the activated
manganese process because of the difficulty of achieving J0% recovery and
the economical disadvantage. The advantage of the dry process—no reheating
of gas—may not compensate for the disadvantages—the need for a large
absorber and an expensive absorbent or reducing agent.
Problems common to wet processes Numerous plants using wet processes are
in operation or under construction. Still, the wet processes have some
problems in common. One is reheating and the other is wastewater. For
reheating of gas, afterburning of oil has been used exclusively in Japan,
The oil consumption is about 5?4> of that for power generation, indicating
that when the 863 removal efficiency by scrubbing is 90%» overall efficiency
would drop to 85%. The addition of S02 and some dust to the gas is going to
present another problem as the regulations on SC>2 and dust become more
and more stringent. Studies on the use of the heat exchanger have been
started recently.
It is common in Japan to purge some water from the system in order to
maintain the concentration of impurities in the circulation liquor under
a certain level. Among the impurities, chlorine derived from fuel and
input water is most troublesome because it causes corrosion. Particularly
when the multitube heat exchanger is used for reheating, the corrosion
problem becomes significant. It may be more logical to purge some water
after treating it thoroughly than to use expensive corrosion resistant
materials. Where no water at all is allowed to be purged, some means of
removing chlorine from the circulating liquor may be desired.
Wet lime-limestone processes Wet lime-limestone processes are most
promising in Japan as in the United States. The reliability of the
146
-------
Mtsubishi-JECCO process has been well demonstrated. However, most plants
based on this process have treated gases with a low S(>2 concentration
(300-1,OOOppm) except for a plant of the Onahama Smelter which treats
waste gas from a converter containing 25,000ppm S02» Two full-scale plants
to treat flue gas with about l,600ppm S02 from oil-fired utility boilers
(375MW each) will start operation early in 1975. The operation performance
should be watched for further evaluation. The Chemico-Mitsui process has
also gained some fame from the successful operation of the Omuta plant
of Mitsui Aluminum Co. using coal-fired flue gas. There has been much
argument concerning the reason for the scale-free operation. Whether the
circulation liquor is saturated or unsaturated with gypsum is a point
which has not been made clear yet. However, there is no doubt that
careful operation control is one of the keys to success. For example,
the pH of the slurry is measured manually because an automatic pH meter
might not be quite reliable.
The Mitsui Miike process has the advantages of attaining high S02 removal
with limestone and of pH control without sulfuric acid. The Mitsui Miike
process as well as the Mitsubishi—JECCO process may be useful also in the
U.S.A. if there is any use for the by-product gypsum. The Mitsui Miike
process may be applicable also for discarding the by-product, if the
partial oxidation, which is required to improve the property of the sludge,
can be done in a simple way without much cost.
For evaluation of other wet lime—limestone processes, a longer operation
period or operation of larger plants may be needed.
Double alkali type processes Among the double aLkali type processes, the
Chiyoda process and sodium-limestone processes of Showa Deriko and
Kureha-Kawasaki have been used widely. Smooth operation and use of
limestone may be the reason for the use. The Chiyoda process is simple
and the plant operation is most easy, although it requires a big absorber
and a high L/G ratio. The operation of a pilot plant of Gulf Power Co.
in Florida which will commence shortly using the Chiyoda process should
be watched for evaluation of the applicability to flue gas from coal-
fired boilers.
The sodium-limestone processes are rather complicated including the
decomposition step of sodium sulfate. The processes would need a large
land space and relatively high investment cost. On the other hand, sodium
consumption is low—about one—fifth of that required for a plant the size
of the Wellman—Lord process plant which does not use an oxygen inhibitor.
If further crystal growth of calcium sulfite is attained, the process
may be useful in the United States to by-produce the sulfite whether it is
used for some purpose or discarded.
Both the Kurabo and Dowa processes use absorbents with an intermediate
pH (3-4) and a moderate L/G ratio (20-40). Operation of commercial plants
is about to start for both processes. Although both seem interesting, a
better evaluation can be made when the new plants have produced some
operating data.
1A7
-------
Nippon Kokan (NKK) uses an ammonium sulfite solution at pH 6.8 as the
absorbent. Therefore, NKK uses a much smaller L/G ratio than does Kurabo,
"but it is not easy to eliminate plume, which has been a problem common to
ammonia scrubbing processes. The Kurabo.process might be advantageous
where there are strict regulations on plume. Both NKK and Kurabo processes
may be useful also for the production of ammonium sulfate where there is
a demand for it.
Other processes Among other processes, the Wellman-Lord process has been
most widely used except for the sodium scrubbing processes to by-produce
sodium salts. The process is reliable and is easy to control. A main
drawback of the process is the high consumption of sodium which necessitates
the emission of a considerable amount of wastewater. Further improvement
may be desirable before application in the U.S.A. where the restrictions
on wastewater are much more severe than in Japan.
It is questionable which is better for sulfuric acid production, the
Veil man-Lord process or the magnesia scrubbing process. The latter would
give much less wastewater but needs a larger amount of energy for the
decomposition of magnesium sulfite and sulfate. Further improvements are
desired also in magnesia scrubbing for better operation and less consumption
of energy.
Sulfur by-producing processes seem costly except for refineries which
have large Glaus furnaces to which the recovered S02 can be charged.
If an oversupply of other desulfurization by-products causes problems,
the sulfur by-producing process may become more popular.
References
l) Jumpei Ando, Proceedings of Flue Gas Desulfurization Symposium (l973)»
pp 69-102 and 875-890
2) Jumpei Ando, Utilizing and Disposing of Sulfur Products from Flue Gas
Desulfurization Processes in Japan, EPA FGD Symposium (Nov. 1974)
3) C.P. Quigley, Proceedings of Flue Gas Desulfurization Symposium
(1973), PP 605-618
4) Masao Endo, Sekko to Sekkai, No. 131, 168-172 (1974)
5) Y. Yamamichi and J. Nagao, Sekko to Sekkai, No. 130, 125-129 (1974)
148
-------
COST COMPARISONS OF FLUE GAS DESULFURIZATION SYSTEMS
G. G. McGlamery and R. L. Torstrick
Tennessee Valley Authority
Muscle Shoals, Alabama
Prepared for Presentation at
Flue Gas Desulfurization Symposium
Sponsored by the Environmental Protection Agency
Atlanta, Georgia
November ^-7,
149
-------
COST COMPARISONS OF FLUE GAS DESULFURIZATION SYSTEMS1
G. G. McGlamery and R. L. Torstrick
Tennessee Valley. Authority
Muscle Shoals, Alabama
ABSTRACT
This paper describes the results of an EPA-sponsored cost
appraisal of the five most advanced flue gas desulfurization processes.
Using data available in late 1973 from demonstration-scale systems
in progress and a representative set of design and economic premises,
detailed, highly visible capital investments and operating costs were
estimated for the limestone slurry scrubbing, lime slurry scrubbing,
magnesia slurry scrubbing - regeneration to acid, sodium solution
scrubbing - regeneration to sulfur, and catalytic oxidation processes.
Itemized capital costs are presented for a base case system
(new 500-MW unit burning coal with 5.5$ sulfur, 30-year life - 127,500
hours operation, 90$ S02 removal, fly ash removal included) for each
process assuming proven technology, a 3-yeax construction schedule (1972-
1975t midpoint of activity - 197^)> no overtime, and an experienced
design and construction team. Energy costs reflecting recent worldwide
escalation are applied to the operating costs.
In addition to the detailed base cases, numerous supplementary
estimates are given to cover such variables as unit size, fuel type,
sulfur content of fuel, plant status (new vs. existing), onsite versus
offsite solids disposal and S02 removal. The five processes are ranked
according to the results; the accuracy of the estimates is also discussed.
Using results from recent in-house evaluation studies,
additional less-detailed comparisons are presented for limestone scrub-
bing with benzoic acid, two lime scrubbing options, sodium scrubbing with
production, and two double-alkali processes.
1 To be presented at the Flue Gas Desulfurization Symposium, Atlanta,
Georgia, November ^-7>
150
-------
COST COMPARISONS OF FLUE GAS DESULFURIZATION SYSTEMS
After several years of intensive process development, several
power plant stack gas S02 removal "systems are now advancing from the
pilot-plant stage to demonstration-scale applications. With the in-
stallation of these larger demonstration-size facilities on power units
around the United States, it should now "be possible to more accurately
predict their costs so that utility executives can "better choose be-
tween alternatives.
Since the Air Quality Act of 1967 and later the Clean Air Act
of 1970, the Federal Environmental Protection Agency has funded research
and development on S02 removal processes including several conceptual
design and cost studies. In these earlier efforts, many design assump-
tions were necessary and cost estimate accuracy was questionable since
technology was in a state of "flux" and available design data were
limited. Equipment costs were sketchy since most vendors had yet to
fabricate and erect the large gas scrubbing devices required for full-
scale systems. Furthermore, very little corrosion data were available
to predict materials of construction for the services involved. In
many cases, optimism of the process developers tended to maximize
process potential and to minimize problem areas such as erosion, scaling,
solids disposal, sulfite oxidation, mist elimination, gas reheat,
operational turndown, and pH control.
Finally, after many pilot-plant tests and both encouraging and
disappointing experiences, five processes have emerged from the many pro-
posed as the initial systems for installation. These are the limestone
slurry process and the lime slurry process, both of which are throwaway
systems (no salable byproduct), and the magnesia slurry - regeneration
to sulfuric acid, sodium solution scrubbing - S02 reduction to sulfur,
and catalytic oxidation - 80$ sulfuric acid processes. In cooperation
with participating utilities and process developers, EPA is currently
funding large-scale test and demonstration projects on each of these
processes. The processes and the associated projects are shown in
Table 1.
Now that many of the unknowns for these systems have surfaced,
remedies been prescribed, and large-scale projects started, more accurate
assessment of process costs should be possible. The objective of an EPA-
TVA study completed in early 19T** was to prepare a set of highly visible,
detailed capital and operating cost estimates for comparison of the five
leading processes on a common, uniform basis. Unfortunately, the rapid
construction cost escalation experienced during recent months has once
again made it difficult to accurately project total costs; however,
comparative costs should be relatively assessable.
151
-------
Table 1. EPA-SPONSORED STACK GAS DESULFURIZATION DEMONSTRATION SYSTEMS
EPA-sponsored process
(byproduct )
Limestone slurry scrubbing
(sludge)
Lime slurry scrubbing
(sludge)
Magnesia slurry scrubbing -
regeneration
(96$ sulfuric acid)
Catalytic oxidation
(80% sulfuric acid)
Sodium scrubbing -
regeneration
Cooperating
utility
TVA
T7A
Boston Edison
Illinois Power
Northern Indiana
Public Service
Process
developer
Bechtel and
others
Chemico,
Bechtel, and
others
Chemico°Ba«ic
Monsento
Davy powergas
Allied Chemical
Location
Shawnee unit 10
Paducah, Ky.
Shawnee unit 10
Paducah, Ky.
Mystic Station 6
Boston, Mass.
Wood River Station U
East Alton, 111.
D. H. Mitchell
Station 11
Unit size
and type
10 MW
coal
10 MW
coal
150 MW
oil
110 MW
coal
115 MW
coal
Expected
startup
Under way
Under way
Completed
*«*»
Late 1°T5
(sulfur)
Co.
Gary, Ind.
-------
PROCESS DEFINITION
A brief description of these five processes and the organiza-
tions supplying representative system data are given "below. The process
data represent the state of technology in late 19T3-
1. Limestone Slurry Scrubbing. Stack gas is washed with a recircu-
lating slurry (pH of 5.8-o.U) of limestone and reacted calcium
salts in water using a two-stage (venturi and mobile bed) scrubber
system for particulate and S02 removal. Limestone feed is wet
ground prior to addition to the scrubber effluent hold tank.
Calcium sulfite and sulfate salts are withdrawn to a disposal area
for discard. Reheat of stack gas to 175°F is provided. Design is
based on data taken from EPA-TVA-Bechtel Shawnee test program. A
flow diagram and layout drawings are shown in Figures 1 and 2.
2. Lime Slurry Scrubbing. Stack gas is washed with a recirculating
slurry (pH of 6.0-8.6) of calcined limestone (lime) and reacted
calcium salts in vater using two stages of venturi scrubbing.
Lime is purchased from "across the fence" calcination operation,
slaked, and added to both circulation streams. Calcium sulfite
and sulfate is withdrawn to a disposal area for discard. Reheat
of stack gas to 175°F is provided. Design is based on data pro-
vided by Chemical Construction Corporation (Chemico). The
representative flow diagram is shown in Figure 3 and layouts are
shown in Figure 4.
J. Magnesia Slurry Scrubbing - Regeneration to HPSQ4. Stack gas is
washed using two separate stages of venturi scrubbing—the first
utilizing water for removal of particulates, and the second
utilizing a recirculating slurry (pH T-5-8.5) of magnesia (MgO)
and reacted magnesium - sulfur salts in water for removal of S02.
Makeup magnesia is slaked and added to cover only handling losses
since sulfates formed are reduced during regeneration. Slurry from
the S02 scrubber is dewatered, dried, calcined, and recycled dur-
ing which concentrated S02 is evolved to a contact sulfuric acid
plant producing 98/0 acid. Design is based on data supplied by
Chemico-Basic Corporation. Shown in Figures 5 and 6 are a flow
diagram and layouts of the scrubbing area.
k. Sodium Solution Scrubbing - SOP Regeneration and Reduction to Sulfur.
Stack gas is washed with water in a venturi scrubber for removal of
particulates and then washed in a valve tray scrubber with a recircu-
lating solution of sodium salts in water for S02 removal. Makeup
sodium carbonate is added to cover losses due to handling and
oxidation of sodium sulfite to sulfate. Sodiujt sulfate crystals
are purged from the system, dried, and sold. Water is evaporated
from the scrubbing solution using a single-effect evaporator to
crystallize and thermally decompose sodium bisulfite, driving off
concentrated S02. The resulting sodium sulfite is recycled to the
scrubber and the S02 is reacted with methane for reduction to ele-
mental sulfur. The regeneration and reduction areas are designed
153
-------
srtt* n>o*
Ui
Figure 1. Limestone slurry process. Flow diagram.
-------
t'
I
• f Li I f'iA'JCir*
jjii
•1"
.*]"
y
! ", ' I
!
t
PLAN
MTIAM HI Ml Ml H
. J
nit
'—r"" p \—
! I !\L
i null' 'i \j
INI IT -
I'.fMIM
rv;!i_^.;T
i
. i--*
\H { NIMA'VJMjMI
4 rtflAl/H
i 1 _
T : •"
^•>ui i mi" i
Mli :\'
•«1 - '-•'••/ I \
:i^-,. L^r_... U j.,\
~>t-»
ELEVATION
Figure 2. Limestone slurry process. Venturi and mobile
bed scrubber system - plan and elevation - base case.
155
-------
rvtin
Figure 3. Lime slurry process. Flow diagram.
-------
ELEVATION
Figure *4. Lime slurry process. Two-stage venturi scrubber system
plan and elevation - base case.
157
-------
Ln
00
w 1 o~« •=-=
I «-"•»- I ,, ,„„ .„ ^J
— .'STT;S r"""^' I • ' n
\<* •t'—i—t j a^r^sl ^5i •—^ • "^ •* jri SH
—T-ittz- """" r^ —^^ cr3
Figure 5« Magnesia slurry - regeneration process. Flow diagram.
-------
ELEVATION
Figure 6. Magnesia slurry - regeneration process. Two-stage
venturi scrubber system - plan and elevation - base case.
159
-------
for 100/6 of power unit load. Design for the scrubbing -
evaporator - crystallizer system is provided by Davy Powergas
Inc. (Wellraan-Lord process) and data for the S02 reduction unit
are provided by Allied Chemical Corporation. Figures 7 a&d 8
describe the processing steps and scrubbing area layout.
5. Catalytic Oxidation, Stack gas is first cleaned of particulates
by a high-temperature electrostatic precipitator; then, the S02 is
catalytically converted to SOa and available excess heat is re-
covered. The SOs reacts with moisture in the stack gas to form
H2S04 mist which is scrubbed in a packed tower using a recircu-
lating acid stream to yield Qo% acid. Mist is removed by a Brink
mist eliminator and the clean 25^°F gas is exhausted to the stack.
Design is based on data supplied by Monsanto Company, developers
of Cat-Ox process. Shown in Figures 9-H are a flowsheet and the
process equipment arrangements for a new unit installation.
In the EPA-TVA report, representative flow diagrams, material
balances, control diagrams, plant layouts, and equipment arrangements are
included for the base case (new 500-MW coal-fired unit, 3-5$> S in fuel,
90% S02 removal) of each process. Together with detailed equipment
descriptions, this background defines the systems estimated.
MAJOR COST FACTORS
From previous economic studies, the following factors are con-
sidered to be the most important:
1. Project schedule and location. Project assumed to start in mid-1972
with 3-year construction period ending mid-1975- Midpoint of con-
struction costs mid-1971* ; Chemical Engineering Cost Index - 160.2.
Startup - mid-1975. A midwestern plant location is assumed.
2. Power unit size. Costs for three unit sizes—200, 500, and 1000
MW---are projected.
3. Fuel type. Systems for both coal- and oil-fired units are costed—
coal 12,000 Btu/lb, 12% ash, oil - lS,500 Btu/lb, 0.1% ash.
1*. Sulfur content of fuel. Costs for three sulfur levels are evaluated
for each fuel—2.0, 3«5, and 5-0$ for coal; 1.0, 2.5, and h.0% for
oil.
5. Plant status. Although systems for both new and existing power
units are evaluated, only a simple, moderately difficult (scrubbing
system installed on vacant space beyond the stack ) retrofit is
estimated since such systems can vary over such a wide range of
configurations and restrictions. New units are designed for a 30-
year life, 127,500 hours of operation. Costs for new and existing
systems are not directly comparable.
160
-------
.Lx—-
J^
Figure 7. Sodium solution - SOg reduction process. Flow diagram.
-------
1- $TtCf
ELEVATION
Figure 8. Sodium solution - S02 reduction process. Venturi and
valve-tray scrubber system - plan and elevation - base case.
162
-------
JW*V£>»>WJ_
U>
J:
9O>t,in fTxn*ra>
fOt UlMltUUff
f 0 rut
snwnr trr*u
f
Figure 9« Catalytic Oxidation process. Flow diagram.
-------
Figure 10. Catalytic oxidation process. S02 conversion and absorption system layout - plan - "base case.
-------
Lfl
N>UMW;
COHOCNMTt MCJTtH — ' I
CMOIumON «CIO COOLDC '
• 2S3 rr i»»»»oii
f
\^FV«
^— COOLM
\T^?TT
^
rtU« 1AMt
—
•—^
Figure 11. Catalytic oxidation process. SQ2 conversion and absorption system layout -
elevation - base case.
-------
6. S02 removal. Since all five processes are capable of 90$ S02
removal and future demands for .-mission control may exceed
present standards, 90^ removal is specified as the base value.
For those processes in which cost effective design changes could
be identified, 80^ removal is also projected.
7. Particulate removal. Costs are included for 9%.7% particulate
removal (to meet EPA standard of 0.1 Ib/million Btu heat input)
on new coal-fired systems except Cat-Ox which requires 99.9$
removal for process reasons (restricted to 0.005 gr/scf prior to
entering converter). Existing units are assumed to be already
equipped with 93-7$ electrostatic precipitators. Because of
this provision, the investment and operating cost results for
existing coal-fired systems appear lover than for new units. To
cover this disparity, a special case is examined where full
particulate removal must be added to an existing unit. Oil-
fired units do not require dust removal facilities. Existing
Cat-Ox installations require incremental additional precipitator
to reduce loading to 0.005 gr/scf.
8. Raw materials and catalysts. Assuming startup in 1975> mid-
western 1975 delivered prices are projected and sensitivity
analysis is used to evaluate variance.
9. Labor. 1975 midwestern operating labor rates are projected.
Sensitivity analysis is used to evaluate overall operating labor
cost variance. Operating labor is escalated over the life of the
project for only a few special cases to show effect.
10. Utilities. Recent energy cost escalation is recognized and 1975
values are projected. Values used in operating cost estimates for
utilities supplied by power plant cover all costs for generation
including return on investment, depreciation, and income taxes.
11. Maintenance. Various levels are analyzed by sensitivity analysis.
12. Capital charges. Regulated (profit and taxes included) economic
basis is used. 'Annual operating cost estimates utilize a base
value of 1^.9$ of fixed investment (10$ cost of money). Level
is varied by sensitivity analysis.
13. On-stream time. Annual operating costs are projected for
operating times of 7000, 5000, 3500, and 1500 hours per year.
Later, these values are used to project a lifetime cost over a
predefined 30-year declining operating schedule.
1*4. Solids disposal. Both onsite ponding and off site disposal costs
are evaluated for limestone and lime processes. Onsite ponding
includes prorated costs for calcium solids to cover pumping and
piping to and from the pond, plus a k0-foot-deep, clay-lined
166
-------
pond sized to meet requirements over the remaining life of the
power unit. Offsite disposal includes proration of a thickener,
filter, piping, pumps, cake conveyors, and loader at the power unit.
Offsite charges are levied-on a fee per ton of wet solids basis
assumed to cover all contractor expenses for hauling, treatment,
and final disposal. Sensitivity analysis is used to evaluate
variance.
15. Net sales revenue. Base values—$3/ton 100$ H£S04 as 98$ H2S04,
$6/ton 100^> H2S04 as 80$ H2S04, $25/short ton for sulfur, $20/ton
for sodium sulfate are used. Variances are covered by sensitivity
analysis.
Other important design and cost assumptions defined for con-
sistent evaluation are:
1. Fly ash disposal facilities proration (ponds, pipes, pumps) not
included. Water balance is based on closed-loop operation.
2. System design assumed not "first of kind;" no redundancy is in-
cluded; only pumps are spared; experienced design and construction
team is assumed utilized.
3. Stack gas reheated to 175 °F except Cat-Ox process.
k. Product storage - 30 days except for Na2S04 - 7 days.
5. Equipment, material, and construction labor shortages with
accompanying overtime pay incentive not considered.
PRESENTATION OF RESULTS
Capital Investment
In the detailed cost study, several methods are used to present
the results. For investment data, the following types of displays are
presented for each process:
Equipment list, cost, size-scale factor, and data source--presented
for base case (new 500-MW coal-fired unit burning coal with 3-5^ S).
See Table 2 for an example.
Summarized process area equipment and installation cost breakdown—
presented for base case and existing 500-MW limestone system. See
Table 3 an<3. k for examples.
Case variations—area investment summaries presented for each of the
16 case variations. See Tables 5-10 for examples.
Shown in Table 11 is a summary of total investment for the 16
case variations of the five leading processes. The relative ranking of
the processes for new coal-fired units (3-5$ S) in order of increasing
investment is as follows:
167
-------
Table 2. LIMESTONE SLURRY PROCESS
BASE CASE
Airj I Mjlcrul lldndling
Item No
1. Unloading 1
hopper No 1
2 Limestone 1
feeder No 1
Ivibialing)
3. Conveyor j
(bell) No 1
4 Conveyor I
(bell) No 2
5. Hopper.' 3
under pile
6. Limestone 3
feeder No. 2
(vibrating)
7. Conveyor 1
(belli No. 3
8. Tunnel 2
sump pump
9. Elevator I
No. I
10. Bin 1
11. Car shaker 1
12 Dust 1
collecting
system No. 1
13 Dust 1
collecting
system No. 2
14 Bag filter 1
system
S
Oesitirumn __
Cjp.icity 94 II '; g'-
4" side; 2 -4 "bot-
tom. 3 deep, carbon
steel
210 tonv/hr.42"
wide x 5 long pan.
ue-cosl
scale
fjclor
06K
0 58
I'ac lor Hase
source cost
diem. Enjjr 3-24-69 1.700
CulHiie
Chem Engr 3-24-69 2.691
Outline
.
Hjvccost Date
source of cost
Caulylic 1973
Inc
Rk-hjidsun Engr 1971
Services
"Projected 1974
equipment cost
each tola)
2.000 2.000
3.300 3.300
2'/i hp vibrator included.
carbon steel
210 lons/hr, 250 ft/
mm. 24" bell. 10'
long. 2'/j hp motor
included . caibon
steel
210 lons/hr. 250 ft/
mm. 24 hell. 172
long. 20 hp moioi
steel
7 '-4" top. l'-4"
botlom. 3* deep.
carbon steel
100 tons/hr. 18"
uide % 3'//long pan.
1 hp vibrator included.
caibon steel
250 fl/min. 18"belt.
1 35' lung. 3 hp moior
included , carbon
steel
5, gpm, 10' head.
V* hp motor included.
caihon steel.
neoprene lining
100 tons/In ("' 85 Ib/
ft3. 16" x 8" x 8V.
bucket, 235 ft/min,
15 hp motot included.
carbon steel
5.000 ft*. 3/8"
carbon steel plate.
plus structural sleei
Railroad trackside
vibrator
2,000 cfm inertial
separator. XO cyclone.
2 dust hoppers, fan,
and drive
6.000 cfm inertial
separator. XO cyclone.
2 dust hoppers, fan.
and drive
14.000 cfm. automatic
fabric dust collectors.
bag support, shaker «ys'
lem. isolation damper,
external shaker motor
and drive, dust hopper,
fan and motor for bag
filter system (VJ in feed
preparation area)
081
0.65
081
068
058
065
081
F:und. of Cost Engj. 1.887
1964
Oirm. Sup. 3-M-«9
Guthrie
1 und of Cost Kngr 8.948
1964
Cnetn. rngr 3-24-69
Gulhrie
Chem. Engr 3-24-69 1.300
Gulhrie
Oiem. Engr 3-24-69 1.550
Guthrie
Chem. Engr 3-24-69 11.250
Gulhrie
Fund, of Cost Engr.
1964
Depends on gpm and head 560
requirements resulting in changes
TVA work 1964
order O05C30
TVA work 1962
order D05P353
Oiem Engt - 1969
3 24-69. Guthrie
Kichardson Engr 1973
Services
Chem Engr.- 1969
3-24-69. Guthrie
Catalytic 1973
Inc
2.900 2.900
13.700 13.700
1.700 5.100
1.700 5.100
14,800 14,800
600 1.200
of motor and impeller si7.es
083
068
0.80
0.80
0.68-
Chem. Engr. 3-24*9 8.765
Gulhrie
(Them. Enp 3-24-69 1 2.650
Guthrie
4.866
Chem. Engr 3-24-69 2.392
Cuthrie
Chem. Engr 3-24-69 4.724
Guthrie
Cbem. Engr 3-24-69 7.926
Guthrie
TVA work 1964
order D05C30
TVA work 1964
order D05C30
TVA work 1965
order IXJ5C30
Richardson Engr 1973
Services
Richardson Engr. 197.3
Services
Richardson Engr 1973
Services
13.300 13.300
19.200 19.200
6.600 6.600
2.600 2.600
5.100 5.100
8.500 8.500
.Subtotal
103.400
168
-------
Table 2. LIMESTONE SLURRY PROCESS (Contd)
BASE CASE
? 1 red
1
2
3
4
5
6
7
8
9
10
11.
12.
13.
14.
Item
Mm discharge
fi-cdci
Urigh feeder
Gyralory
cnisher
(• lrva«>r
No 2
Wet hall
mill
Mills
product
lank
1 ihmg
Agll.ltor.
mills
product
tank
Pumps, mill*
product tank
Slurry feed
lank
Lining
Agilalcu.
slurry
feed
Unk
Pumps, slurry-
feed lank
Dust
collecting
system
Moist
Hag filter
system
S
No Description,
1 2'.'i Inns/Ml. I0~
2 v-idc x 2 '/i long pan.
vibrator included.
i. iilvin steel
2 C'l iims/hr. 18"
bell. 14' Innp. IV,
hp Ulolor included.
carbon sleel
- 1 2'.i|lnns/tH. 0 x l'/i
to '• . 25 hp motor
included
7 12V, lonv'hr. K5 lb/
fi3, 2.1 .s li/inm. 24'
etrs. 6" x 4" \ 4',."
binkel. 1 lip motor
uu In ded
1 3CHI luns/d.u . 8' dia
x 1 2' long, from ',."
to 200 mesh
2 450 hp molors for
hall mill
1 1.920 gal. 8'diam-
etet x 5" high.
vertical svilh open
lop. caibon sleel
Neoprene lining for
mill produc 1 tank
1 1 hp. neoprcne
coated
2 96 gpm. 58' head.
centrifug.il. wuh
variable speed drive
and 3 hp motor, carbon
1 46,080, p.il. 6.1 60 fl3.
17 -4 ' Jijmrler x 27'
high, vertical svith open
lop.<41 l'-V v.ule
baffles, carbon steel
Vt" neoprene lining
1 10 hp. neoprene
coated
2 96 gpm. 5H' head.
centrifugal, svilh
variable speed drive
and 3 hp motor, carbon
steel, neoprcnc lined
1 8.000 cfm, mental
separator, XQ
cyclone, 2 dust hoppers
fan and drive
1 5 Ion electric
1 14.000 cfm, auiomalic
fabric dus( collector .
/e-cost
si air 1 act or
aitor _ source
0.5K ("hem 1 ngr ).~>ij,~il e.ich
Richardson iKnjfr. 1973 600
Services
Catalytic I97J 8200
Inc
Denver Fquip 197? 1 1 .100
C'o
Chem 1 ngr 1969 2.400
3 24*9. Guthne
Denver Kquip 1973 108.150
Co
V.eslm(jhou« 1973 14.20(1
CATX 1971 2.500
CATX 1971 ROO
Mixing [quip 1971 1.200
C'o.. Inc.
Denver Kquip. 1973 2.200
Co
GATX 1971 19.200
GATX 1971 17.000
Mixing 1-quip 1971 5.600
Co . Inc
Denver ! quip 197.S 2.200
Co
Richardson l.ngr 1973 6.400
Services
Richardson i-lnftr. 1973 19.000
Services
Richardson Knpr 1973 8.500
Services
' 1974"
con
_ total
" T.200
16.41X1
22.600
4.KOO
216.3(XJ
2«.400
2.500
800
1.200
4.400
19.2IN1
1 7.000
5.600
4.400
6.400
19.000
8.500
hap support. sh.iKer ystcrn.
isolation d.irnpcr. c\ ernal
dusi hopper, motor nd
fan for bap Tiller sys en
area)
Subtotal
17K.700
169
-------
Table 2. LIMESTONE SLURRY PROCESS (Contd)
BASE CASE
Area 3- Paniculate Scrubbing
hem
I. Tank.
paniculate
scrubber.
effluent
hold
Lining
2. Agitator,
effluent
hold tank
3. Pumpi.
recycle
slurry
4. Venturi
scrubber
5. Venturi 4
MBA sump
6 Sool
blowers
Subtotal
No.
4
4
Sue-coil
icale Factor
Description fictor source
25,700 gal. 3.435 fU. 0.68
13 diameter x 26*
high, open top. (4)
1 wide baffles.
carbon steel
V* neopjene lining
Shp.neoprene 0.26
coated
Chem. Kngr. 3-24-69
Guthrk
r-und of Cost k'.ngi
1964
Base Bate cost
cost source
13,500 GATX
10,000 GATX
3.400 Mixing Rquip.
Co
Pi ejected 1974
Dale equipment coil
of coil each toll!
I97J 16.300 65.200
1971 12.000 48.000
1971 4.000 16.000
6 4.900 gpm. 144'
head, centrifugal,
bell drive, 300 hp
motor included, carbon
steel, neopicne lining
4 With variable throat.
36' long x 5'wide x
20'high. conveying
section & thioat
carpenter 20. remainder
W" carbon steel with
neoprene lining
4 28'long x 41* wide
x 13'high. V."
carbon steel, ncoprene
lining (Win SOj
scrubbing area)
0.50 Chem. F.ngr. 3-24-69
_, Cuthne
Depends on gpm and head
requirements resulting In
changes of motor and
impeller size
060
Universal Oil
Products
13.440 Denver Fquip.
Co
133.000 UniverulOil
Products
1973 1OOO 87.000
1971 150.000 600.000
068 Chem. Engr 3-24-69
Gulhrie
49.500
I 00 TVA
Products
3.500 Widows Creek-
TVA
1971 57.500 230.0<»
1971 4.000 80.000
1.126.200
170
-------
Table 2. LIMESTONE SLURRY PROCESS (Contd)
BASE- CASE
Area 4 SO_j Scrubbing
llem No__ Description^
K'A scrubber 4 SO, jhvoiher" mo-
bile bed. with
>i;e-cost
scale
_facloj_
OHO
driinslrr. 4 I' long x 16'
wide > 4 I1 high; V,"
cjibon stfcl wuh
nti'prenc lining.
316 S S Krids. hlj!h
density poly . spheres.
I HP spray headers.
(hi-vron v./ne mist
I acloi
source
UnivWsll Oil"'~
Products
Base It.ise cost
cost source
~22"7.<)0b" iimver'i.i OiF
Products
"Tiate"
of
COSl
1971"
1974"
lost
eaih '°'?l_
300.000 \.200.000
'Tnijeiled
equipment
2 Venlurl &
MBA sump
3 Tank. 4
absorber
effluent
hold
Lining
4 Ajilalor. SO, 4
absorber
hold lank
5 Pumr«. SO, 10
absorber
recycle
slurry
6 Pumps.
makeup
water
7 Soot
blowers
Subtotal
1
28' long x 41*
wide x I 3' high. Vt"
carbon steel, neoprene
lining (Vj in paniculate
si'rubbinp area)
240.000 pal. 32.0R8
ft 340'diameter x
26 high, open lop.
field elected.
carbon steel
Vi neoprene lining
50 hp. neoprene
coated
I l.500ppm. 105'
head. Centrifugal,
neoprene lined, belt
drrre. 500 hp motor
included
l.240gpm, 150'
head, vertical
multistage turbine.
100 hp molor
included
40
068
068
O.SO
Oieni. Ingr 3-24*9
(lulhrir
(Tiern Fnpr
C'.uthne
3-24^,S
Oiem \nfi 3-24-69
Gulhrie
Depends on ppm and head
requirements resulting in
changes of motor and
impeller si/e
Depends on ppm and head
requirements resulting in
changes of molor an-4
impeller si?e
1.0(1 TV A
49.500 Unwerul Oil
Products
.is. .127 <;ATX
32.000 C.ATX
!2.245 fatjlvt
Inc
23.000 Allen. Sherman.
Hoff
2.369 Richardson Kngi
Sen-ices
3.500 Widows Creek
TV A
1971 57.5OO 230.00O
1972 45.100 180.400
1972
1973
1971
|97?
1971
3H.600 154.400
1.1.250 53.000
27,100 271.000
2.600 5.20(1
4.00O 160.000
171
-------
Table 2. LIMESTONE SLURRY PROCESS (Contd)
BASE CASE
Aira 5 Reheat
Item
1 1,21.
re heater
2 Snol
hloueri
Suhlolal
Size-cost
«cale Factor
_No _ _ Description _ _fa£loj «oui
Tuiie type. 2.028 f(T
I3.900lb. Viof
tuhrs nude of mconel
625 and remaining Vi
made of cor-len
O.tJO
Cheni
Culhrie
Bate
cpjl _
70^000 '
Base cost
source
20
1 00 TVA
TVA
3.500 Widow; Orek
TVA
Dale Piojrclcd 1974
of cqutpmcni cost
_co«t ?'ch (ouj
1971 82.MIO ~ JJO.OOO
1971 4.000 80.000
410.000
Area 6-Ga) HandIjn^
llcm
I. Han
No,
4
Description
41"1.200rpm.
2.905 hhp. 3.250hp
molor included with
insulation (Cojl ij
difference between
4l"and 15" fan
Remainder is
allocated to boiler,)
Si/e-cost
-------
Table 2.
LIMESTONE SLURRY PROCESS (Contd)
BASE CASE
AreaJ? Solids Di^posaj
Size«M
Hem No.
1 Pond Iced 1
tank
Lining
2 Aptlaloi 1
3 Pumps, pond 4
feed tank3
4. Pump*. 1
recycle pond
water3
Subtotal
Description
63. 000 gal. field
erected, 8,42.1 ft3.
21 diameter x 26 high.
veilu'jl vvilh open top.
carbon tie el. 1 - 8"
x 2h' bafflei
V« nroprfne lining
7V) hp. nettprene
coaled
I.128f:pm (••' 75'
head, neoprene
lined with 50 hp
motor
I.OOOppm ("' ISO'
head, multistage
turbine, cast iron
bowl, stainless steel
impellers, with 75
hp motor
scale Kiclor Rate
factor source cost
0 68 Chem. Fngi .1-24-69 20.000
Gulhrie
16,120
050 Chem Engr 3-24-69 4.717
0.47 Fund of Cost Fngr 1964
Depends on gprn and head 1,850
requirements resulting in
changes of motor and
impeller lize
Depends on gpm and head 1.100
requirements resulting in
changes of motor and
impeller size
Due
Bate coil of
lourcc cost
GATX 1 97 1
GATX 1971
Mixing Fquip. 197 1
Co.. Inc
Denver Fquip. 1973
Co
Richardson Kngr 1973
Services
Projected 1974
equipment coil
each
-------
Table 2. LIMESTONE SLURRY PROCESS (Contd)
BASE CASE
Aie« 9- Scivicrt
i
2
3
4
5
Item
PiylOider
Plant
vehicles
Maml &
instrument
• hop-
equipment
Service
buildmg-
equipinent
Slorei-
cqulpment
Subtoul
No Description
1 C.a'.olinc type.
2yd-1
- t.ll.H-.mm)
- Office, ma June
tools, and
machine shop
equipment
Equipment for lab..
locker room, motor
contiol room.
reslrooms
Office equipment.
shelving, etc.
Size-cnit
scale
filCIOt
_
I'actor
source
Hue Hmr cost
_ co
-------
Table J. LIMESTONE SLURRY PROCESS - TOTAL CAPITAL INVESTMENT REQUIREMENTS
BASE CASE8 SUMMARY - PROCESS EQUIPMENT AUD INSTALLATION ANALYSIS
(Thousands
Raw
irulerub
nandlini
DvrfiCaa
Equipment
Material
Ubor
Aptnf tnd insutiDOB
Material
Uboe
Ductwork, chute*, aad supports
Material
Ubot
Concrete foundations
Material
Ubor
Excavation, ote prrp&nuon.
railroads, roads, and pond
SITU aural
Material
Ubot
Eketncal
Milenal
Ubot
Instnjmnts
Material
Ubot
Paint and mtKeOanoous
Material
Ubot
Suildmrs
MatenaJ
Ubot
Uad
Conjunction facilities
Subtotal direct investment
Indiffft Cotll
Enpneenni; deurn and supervision
Construction field expense
Contractor fees
Contingency
Subtotal faed investment
AJiovtnce for nartup and modiiVtDora
Interest dunni construction
Tout capital investment
Percent of totaj capital investment
10)
2<
i
i
15
1
14
II
II
10
)9
1)
1
4
1
5
.
«
_
419
)l
46
21
<2
566
45
45
656
26
Feed
preparation
379
57
14
IS
t
6
7
it
-
.
-
St
9J
49
23
1
6
65
67
I
-
999
II
99
45
90
1.214
97
97
1,408
56
Fartxulatc
1.126
114
102
95
7JJ
512
12
41
.
94
31
61
7)
77
31
)
15
.
2
,
).20)
298
)32
160
)20
4.)23
346
)46
S.01S
199
SO,
2JM
335
Ml
)l)
)77
411
24
II
.
95
41
101
112
162
11
)
15
-
-
2
.
4.745
427
522
237
474
6.40S
51)
SI)
7.4)1
295
of
Reheat
410
M
II
20
-
-
-
-
.
.
1
|
21
11
1
-
.
-
.
556
SO
61
29
56
751
60
60
871
)S
Dollars )
Fans
2&5
34
-
_
146
5)
4
21
.
.
Ill
155
15
7
1
.
_
-
-
954
77
94
4)
95
1.15)
92
92
1,337
5)
Soddl
dupostl
60
3
M
71
-
-
1
5
3.021
2
SO
184
II
5
2
13
-
-
395
-
3.92J
)5)
4)2
IK
)92
S.J*S
424
424
6.144
244
Uolitan
-
-
6
II
-
-
10
10
16
1
)
)
-
-
-
-
67
6
1
3
7
9!
7
7
11)5
04
Services
107
2)
.
.
-
)39
.
-
.
,
-
II)
40
14
6)1
57
70
32
64
961
69
69
m
40
Cons true DOB
(acililiei
5t Total
4.724
677
561
54«
U77
1.067
62
274
3J67
180
91
4S3
710
359
179
I)
59
I7|
107
420
765 765
765 16.069
69 1.446
84 1.768
)l *03
77 1.607
1.03) 21.693
12 1.735
12 1.7)5
1,197 1316)
41
Percent of
direct
294
42
jj
)4
79
64
04
| 7
210
I |
06
21
44
22
1 1
01
04
1 1
07
26
48
1000
90
no
50
100
:>so
101
101
1566
Percrtu of
total capita/
III
27
2J
2J
] 1
4.1
OJ
1 1
134
07
04
II
11
14
0.7
01
0.2
O.I
04
II
30
63.9
17
1.0
3.2
64
K.J
(9
69
1000
SOO-MW new coftJ-fi/eiJpowrt unit. ) i% S tn fucJ.90% SO) removal. OfHBtc MUdi dupou!
Suck fji reheat lo I 75 F bv indi/tct iteun reheat
DuptxJl pond located 1 Rule ffom po*er pUnt
MKjvnE pu/it locjiUoft (tptfWMi ptoj«i Ixipnntnj m.d-1977. endinf mid-197$ Arrrift cott bam Tor icaJtn|. mid-1974
M^n/mjm 01 peocrtt florist, only pumpi &/t tpiied
(mntmrni rrquircmenu foi otupo«Al of fit lift (tcludfd
Conitruction UbOf ihmt not conodered
-------
Table k. LIMESTONE SLURRY PROCESS - TOTAL CAPITAL INVESTMENT REQUIREMENTS
EXISTING CASE8" SUMMARY - PROCESS EQUIPMENT AND INSTALLATION ANALYSIS
(Thousands of Dollars)
Ra»
Construction
materials F««d
Equipment
Maura)
Labor
Flpint, and tnsutstioa
Material
Labor
ftictwork, chute*, and supports
Us ni«l
Labor
Concrete foundations
Material
tabor
EJtcava&on. site preparation.
rajjoada. roada, and pond
Structural
Material
Labor
DcctrlcaJ
Material
Labor
Instruments
Material
Labor
hint and nbceuaMout
Material
Labor
Buildings
Material
Labor
Land
Construction facilities
Subtotal direct investment
/idirrcf Cotn
Engineering desujn and supervision
Construction Held expense
Contractor fees
Contingency
Subtotal fixed Investment
Allowance for iiarrup and modiftcanons
Interest dunng construction
Total capital investment
Percent of total capital investment
Handling
105
17
1
I
1)
10
14
103
.
11
13
40
105
1
5
1
7
-
-
6
-
4S2
49
63
)4
S3
681
54
54
7S9
J.4
preparation
390
72
14
36
6
1
7
41
-
_
-
59
117
SO
32
1
8
66
15
1
-
1. 000
100
130
70
no
1.410
11)
113
1.636
7.1
SO,
scrubbing
2JIO
42t
353
403
363
3«7
24
103
-
M
52
102
142
165
103
3
19
-
_
2
-
3.243
324
612
367
577
7.393
592
592
J.577
371
Reheat
186
2)
9
27
-
.
_
1
.
-
-
1
II
35
16
-
_
-
.
_
-
)23
32
42
22
36
455
36
36
527
23
Fans
430
14
_
_
311
386
7
39
_
-
_
135
196
15
9
_
1
-
_
_
.
1.710
171
222
120
188
2.411
193
193
2.797
12.1
Solldi
disposal
67
4
«7
99
-
_
1
6
2.7)4
-
3
51
233
11
6
2
16
-
_
291
-
3.611
361
469
153
)97
5.091
407
407
3 90S
256
UUl/oei
31
57
12
19
-
_
1
7
4«
-
-
52
67
20
12
4
7
-
.
_
-
3)5
34
44
23
)7
473
)!
)8
549
24
facilities
Strokes 3%
101
32
_
_
-
_
-
-
4J2
.
-
_
-
.
-
-
'l!4
50
14
672
740 672
74 67
96 87
52 47
81 74
1.04) 947
8) 16
83 16
1.209 1.099
52 48
Total
3.847
735
476
385
772
791
54
301
)J02
97
68
447
m
304
183
11
38
ISO
1)5
314
672
14.116
1.412
I.85S
968
US)
19.904
IJ92
IJ92
23.088
rVrcent of
dvrct
nveitment
272
52
34
4 1
5.5
56
04
2.2
217
07
0.5
32
62
21
1 3
0 1
04
l.J
09
2 2
4 8
1000
100
13 0
1 0
II 0
141 0
11 )
11 3
16)6
rVrcentof
total capital
investment
16 7
3Jt
2.1
2J
3J
34
0.2
IJ
139
04
OJ
1 >
3(
1.3
01
01
02
06
06
1 4
29
61 1
6 1
SO
43
6 7
862
69
69
1000
alinf co»J-firsd power oroi. 3 5%^ tn fud.90% SO] rcmovkl. on-atc tobdi dupoui.
Stjck |i» rtheit to I ?5 F by direct oil-feed reheil
DupOul pond Ioc*ted I tnJe from power pUnt
Mxlwett pU/tt locibon rfprfienti project brfutnirtf m)d-l972,ntdtn{ miA-l^IS Avtrtft co«t bins for
Mini mum m proccti ttottf«.oniy pump* are fpued
Rrm.u/)in| U/e of pow« vnit. 23 jrf.
IrrvfiUTunt rtqu(/f-nTf»t» for removal tnd dupoul of Oy uh excluded.
ConitmtUon labor ihortutt with ucompanymf overttmt pty incmine rvoi conj*der«d
-------
Table 5- LIMESTONE SLURHY PROCESS
SUMMARY OF ESTIMATED FIXED INVESTMENT8
(50O-J-W new coal-fired power uiilt, }.'j% S In fuel;
9Q% SO;, removal; onoite solids disposal)
Investment, $
LJroeatotie receiving and storage (hoppers, feeders,
conveyors, elevators, and Mns) 1*19,000
Feed preparation (feeders, cruisliere, elevators, ball
mlllo, tanks, and pumps } 899,OOO
Partlculate scrubbers and Inlet ducts (t scrubbers
Including common feed plenum, effluont hold tanks,
agitators, and pimps) 3,20},OOO
Sulfur dioxide scrubbers and ducts (i scrubbere
Including mist eliminators, effluent hold tnnks,
agitators, pumps, and exhaust gas ducts to Inlet of
fan) I^T^.OOO
Stack gas reheat (U Indirect steam reheaters) 556,000
Fans (U fans including exhaust gas ducts and dampers
betveen fan and stack gas plenum) 85'i,000
Calcium solids disposal (onelte disposal facilities
Including feed tank, agitator, slurry disposal pumps,
pood, liner, and pond water return puraps) 3,923,000
Utilities (instrument air generation and supply system,
plus distribution systems for obtaining procests steam,
water, and electricity from the power plant) ('7,000
Service facilities (buildings, shops, stores, site
development, roado, railroads, and valkways) 638,000
Construction facilities 765,000
Subtotal direct Investment l6,069,OOO
Qiglneerlng design and supervision I,14|i6,000
Construction field expense 1,768,000
Contractor fees 803,000
Contingency l,fO7,000
SubtotaJ. fixed Investment 21,693,000
Allowance for startup and modi fications 1,735,000
Interest during construction (8Jk/annum rate) 1,735,000
Total capital Investment 25,163,000
Percent of subtotal
direct Inventojnt
2.6
5.6
19-9
29-5
3-5
5-3
It.O
J..8
100.0
9-0
11.0
5.0
10.0
135.0
10.0
10.8
156.6
a Basin:
Stack gas reheat to 175°F by Indirect cteaa reheat.
Disposal pond located 1 mile from power plnnt.
Midwest plant location represents project beginning mid-1972, ending mid-1975- Average
cost basis for scaling, mid-igr^.
Minimum in process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
177
-------
Table 6. LIME SLURRY PROCESS
VENTURI - VENTURE SCRUBBING SCHEME
SUMMARY OF ESTIMATED FIXED INVESTMENT"
(500-MW new coal-fired ;>ower unit, J-5# S in fuel;
90$ GO;, removal; onslte oolldn disposal)
Investment, $
Line receiving ana storage (bins, feeders, conveyors,
and elevators) 795,OOO
Feed preparation (conveyors, slakers, tanks, agitators,
and pimps) 387,000
Partlculate - sulfur dioxide scrubbers and Inlet ducts
(li r.crubbers Including conraon feed plenum and pumps) I*,017,OOO
Sulfur dioxide scrubbers and ducts (*< scrubbers
Including mist eliminators, pumps, find exhaust gas
ducts to Inlet of fans) 5,153,000
Stack gas reheat (14 indirect steam reheaters) $142,000
Fans (U fans Including exhaust gas ducts and dampers
between fan and stack gas plenum) 767,000
Calcium solids disposal (onsite disposal facilities
Including elurry disposal pumps, pond, liner, and
pond water return pimps) 3,356,000
Utilities (instrument air generation and supply
system, plus distribution systems for obtaining
process steam, vater, and electricity from the
pover plant) 67,000
Service facilities (buildings, shops, stores, site
development, roads, railroads, and walkways) 552,000
Construction facilities 682,000
Subtotal direct Investment lU,318,000
Qngineerlng design and supervision 1,289,000
Construction field expense 1,575,000
Contractor fees 7l6,OOO
Contingency l.lQg.OOO
Subtotal fixed investment 19,330,000
Allowance for startup and modifications 1,5^6,OOO
Interest during construction (8^/annum rate) 1,Sli6,000
Total capital Investment 22,^22,000
Percent of subtotal
direct Investment
2.7
28.1
22.0
5.8
23-
0.5
3.8
it.8
100.0
9.0
ia.o
5.0
10.0
135.0
156.6
a Basis:
Stack gas reheat to 175°F by Indirect steam reheat.
Disposal pond located 1 mile from power plant.
Midwest plant location represents project beginning mid-1972, ending mid-1975. Average
coat basis for scaling, mid-197tt-
Minimum In process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages vlth accompanying overtime pay Incentive not considered.
178
-------
Table 7- LIME SLURRY PROCESS
VENTURI - VENTURI SCRUBBING SCHEME
SUMMARY OF ESTIMATED FIXED INVESTMENT0
(5OO-HW existing coal-fired power unit, 3.5$ In fuel;
SOp removal; onslte solids dlspoDal)
Percent of subtotal
Investment, $ direct Investment
Lime receiving and storage (bins, feeders, conveyors,
and elevators)
Feed preparation (conveyors, sjjikers, tanks, agitators,
and pumps )
First stage sulfur dioxide scrubbers and ducte (k
scrubbers Including common feed plenum, pumpe, and
all ductwork between outlet of supplemental fans
and the scrubbers)
Second stage eulfur dioxide scrubbers and ducts (1.
scrubbers Including mist eliminators, pumpc, and all
ductwork between scrulibers and stack gas plenum)
Stack gae reheat (U direct oil-fired reheutere)
Fans C* fans including ducts and dampers between tie-
in to existing duct and Inlet to supplemental fan)
Calcium solids disposal (onslte disposal facilities
including slurry disposal pumps, pond, liner, and
pond water return pumps)
Utilities (instrument air generation and supply system,
fuel oil storage and supply system, and distribution
systems for obtaining process water and electricity
from the power plant)
Service facilities (buildings, chops, stores, site
development, roads, railroads, and walkways)
Construction facilities
Subtotal direct investment
Brglneerlng design and supervision
Construction field expense
Contractor fees
Contingency
Subtotal fixed investment
Allowance for startup and modifications
Interest during construction (8^/annum rate)
Total capital investment
676.COO
It, 565, COO
3,797,000
305,000
1,1143,000
3,01.9,000
335,000
6149,000
758,000
15,913,000
1,591,000
2,069,000
i,ui»,ooo
,1,750,000
22, i 37, 000
5-5
2.7
28.7
23-9
1.9
7-2
19-1
2.1
14.1
14.8
100.0
1,795,000
,1.795,000
26,027,000
163.6
Basle:
Stack gaa reheat to 175°F by direct oil-fired reheat.
Disposal pond located 1 mile from power plant.
Midwest plant location represents project beginning mid-1972, ending mid-1975. Average cost
basis for scaling, mid-19714.
Minimum In process storage; only pumps are spared.
Remaining life of power unit, 25 yr.
Investment requirements for removal and disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
179
-------
Table 8. MAGNESIA SLURRY - REGENERATION PROCESS
SUMMARY OF ESTIMATED FIXED INVESTMENTS
(500-MW neji coal-fired power unit, 3. 5J& E In fuel;
905Ts02 removal; 15-8 tona/hr 100% H^SOj
Investment, $
Magnesium oxide and coke receiving and storage (pneumatic
conveyor and blower, hoppers, conveyors, elevators, and
storage olios) 192,000
Psed preparation (weigh feeders, conveyora, elevators.
Blurry ing tanX, agitator, and pumps) 238,000
Particulate ocrubbero and Inlet ducta (U scrubbers
including common feed plenum, effluent bold tanks,
agitators, pumps, and fXy aoh neutralization facilities) 3,966,000
Sulfur dioxide scrubbers and ducts C» scrubbers
includJrag ndot elJmJnatorc, piznpc, and exhaust gas
ducte to Inlet of fan) 2,592,000
Stack gaa reheat (lj Indirect steam reheatere) 509,000
Fans (*4 fans Including exhaust gas dueta and dampers
between fan and stack gae plenum) 71*1,000
Slurry processing (screens, tanks, pumps, agitators and
beating colls, centrifuges, conveyors, and elevators) 711,000
Drying (fluid bed dryer, fans, combustion chamber, dust
collectors, conveyors, elevators, and MgS03 storage silo) 972,000
Calcining (fluid bed calclner, fans, weigh feeders,
conveyors, elevators, waste beat boiler, duct collectors,
and recycle /feO storage silo) 1,1O8,OGO
Sulfurlc acjd plant (complete contact unit for sulfurlc
acid production including dry gas purification systejn) J,197,000
Sulfurlc acid storage (storage and shipping facilities
for 30 dayo production of H^SO-j) 278,OOO
Utilities (instrument air generation and supply oystem,
fuel oil storage and supply system, and distribution
systems for obtaining process steam, water, and
electricity from power plant) 269,000
Service facilities (buildings, nhops, stores, site
development, roads, railroads, and walkways) 78},OOO
Construction facilities 778,000
Subtotal direct Investment 16,3311,000
Qiglneerlng design and supervision 1,797,000
Construction field expense 1,797,000
Contractor fees 817,000
Contingency 1.633.000
Subtotal fixed Investment 22,378,000
Allowance for startup and modi flcatlonc 2,238,000
Interest during construction (8^/annum rate) 1,790.000
Total capital Investment 26,'i06,OOO
Percent of subtotal
direct Invegtnent
1.2
1.5
21.. J
15.9
3-1
"•-3
5-9
6.8
19-6
1.7
1.6
U.8
100.0
n.o
ll.o
5-0
10.0
137.0
13.7
11.0
161.7
a Basis:
Stack gas reheut to 175°F by Indirect steam reheat.
Average
Mldweot plar.t location represents projects beginning mid-1972, ending mid-1975-
coot basis for scaling, mid-l97ti.
Minimum In proceoo storage; only pumps are spared.
Fly ash slurry neutralized before disposal; closed loop water utilization for first stage.
Investment requirements for disposal of fly aeh excluded.
Construction labor ohortufieo with accompanying overtime pay Incentive not conaidered.
180
-------
Table 9- SODIUM SOLUTION - S0a REDUCTION PROCESS
SUMMARY OF ESTIMATED FIXED INVESTMENT
(500-MW jneu coal-Tired power unit, 3.5^ 8 In fuel;
90?• SOj removal; It.7 tong/hr sulfur)
Soda ash and antloxidant receiving, otorage, and
preparation (pneumatic conveyor and blower,
feeders, mixing tanX, agitator, and pumps)
Particulate scrubbers and Inlet ducts (1* scrubbers
Including common feed plenum, effluent bold tanks,
agitators, pumps, and fly ash neutralization
facilities)
Sulfur dioxide scrubbers and ducts (It scrubbers
Including mist eliminators, pumps, and exhaust
gas ducts to Inlet of fan)
Stack gas reheat (It indirect steam reheatero)
FBJIB (1* fans Including exhaust gas ducts and dampers
between fane and stack gas plenum)
Purge treatment (refrigeration system, chiller-
crystalllzer, feed coolers, centrifuge, rotary dryer,
steam/air heater, fan, dust collectors, feeders,
tanks, agitators, pumps, conveyors, elevator, and
bins)
Sulfur dioxide regeneration (evaporator-crystallizers,
heaters, condensers, strippers, compressors,
desuperheater, tanks, agitators, and pumps)
Sulfur dioxide reduction unit
Sulfur storage (storage and shipping facilities for
JO days production of molten sulfur)
Utilities (instrument air generation and supply
system, and distribution systems for obtaining process
steam, water, and electricity from power plant)
Service facilities (bulldingSj shops, stores, site
development, roads, railroads, and walkways)
Construction facilities
Subtotal direct Investment
Bigineerlng design and supervision
Construction field expense
Contractor fees
Contingency
Subtotal fixed Investment
Allowance for startup and modifications
Interest during construction (8^/annum rate)
Total capital Investment
Investment. $
225,000
3,8li6,ooo
14,269,000
539,000
889,000
1,1473,000
2,717,000
2,921,000
227,000
195,000
662,000
898.000
18,861,000
2,075,000
2,075,000
9113,000
1.886,000
25,81(0,000
2,58^,000
2,067,000
30,1491,000
Percent of subtotal
direct Investment
1.2
20. U
22.6
2.9
14-7
7-8
lU.U
15-5
1.2
1.0
3-5
U.8
100.0
11.0
11.0
5.0
10.0
137.0
13-7
11.0
161.7
Basis!
Stack gas reheat to 175°F ^ Indirect steam reheat.
Midwest plant location represents project beginning mid-1972, ending mid-1975- Average
cost basis for scaling, mld-197't.
Minimum In process storage; only pumps are spared.
Fly ash slurry neutralized before disposal; closed loop water utilization for first stage.
Investment requirements for disposal of fly ash excluded.
Conetructlon labor shortages with accompanying overtime pay Incentive not considered.
181
-------
Table 10. CATALYTIC OXIDATION PROCESS
o
SUMMARY OF ESTIMATED FIXED INVESTMENT
(500-MW
new coal-fired power unit, 3-5/6 S In fuel;
t ROP removal; 15-7 tong/br 1OO% H£5O4)
Investaentf ft
Converter and abnorber startup bypass ducts and
danpers !'91,000
Electrostatic precipl tators and Inlet, ducts (U
high temperature electrostatic precipltutors
Including cotncicm feed plenum) 8,736,000
Sulfur dioxide converters and ducto C* converters
Including catalyst sifter, hopper, etoro^e bin,
conveyors, and elevatorc) 2,1^5,000
Heat recovery and ducts (^ stean/alr heaters and k
flujd/alr beaters Including ducts betveen
economizers and air heaters, and combustion air
ducts and dampers between poverhouoe and sir heaters;
Investment credit for uoe of smaller air heaters
included) 1,1<75,OOO
Fane (Ij ID fans including exhaunt gas ducto and
dampers betveen 3D fane and stacX gas plenun) 1,|JL?,000
Sulfurlc acid absorbers and coolers (2 absorbers
Including mist eliminators, coolers, tanXe,
pumps, and ducts and dampers betveen air heaters
and ID fans) 8,917,000
Sulfurlc acid storage (storage and shipping
facilities for 30 days production of H^SO.,) ^09,000
Utilities (instrument air generation and supply
system, and distribution systems for obtaining
process steam, vater, and electricity from power
plant) 57,000
Service facilities (buildings, shops, stores, site
Percent of subtotal
direct Investment
1-9
8-5
5.8
5-6
35-2
1.6
0.2
development, roads, railroads, and walkvays)
Construction facilities
Subtotal direct investment
Engineering design and supervision
Construction field expense
Contractor fees
Contingency
Subtotal fixed Investment
Allovance for startup and modifications
Interest during construction (Si^/annum rate)
Total capital Investment excluding catalyst
Catalyst
Total capital investment
518,000
1,208,000
25,368,000
2,790,000
2,Y90,OOO
1,268,000
2,517,000
3Mw,ooo
3^75,000
2,780,000
111, 008,000
1,728,000
1*2,736,000
2.0
14.8
100.0
11.0
U.n
5.0
10.0
137.0
1J-7
11.0
161.7
6.8
168.5
a Basle:
Midwest plant location represents project beginning mid-1972, ending mid-1975-
coot baols for scaling, mid-1971*.
Only punpo are spared.
Investment requlremento for disposal of fly aoh excluded.
Construction labor shortages vith accompanying overtime pay Incentive not considered.
182
-------
Table 11. SUMMARY - TOTAL CAPITAL INVESTMENT REQUIREMENTS
Case
Coal-fired pover unit
90j S02 removal; onsite solids disposal
200 MW new, 3.5$ sulfur
200 MW existing, 5-5* sulfur
500 MW existing, 3.5$ sulfur
500 MW new, 2.o£ sulfur
500 MW new, 3.5% sulfur
500 MW new, 5.0* sulfur
1000 MW existing, 3.5$ sulfur
1000 MW new, 3.5% sulfur
SOg removal; onsite solids disposal
500 MW new, 5.5^ sulfur
S0j> removaJ ; of fsite solids disposal
500 MW new, 3.5% sulfur
90% S02 removal; onsite solids disposal
(existing unit without existing
participate collection facilities)
500 MW existing, 3.5% sulfur
Oil-fired pover unit
90$ SOg reaoval; onsite solids disposal
200 MW new, 2,5£ sulfur
500 MW new, 1,0? sulfur
500 MW new, 2.5i sulfur
500 MW new, ^.0^ sulfur
500 MW existing, 2.5$ sulfur
1000 MW new. 2.5* sulfur
Years
life
30
20
25
50
30
30
25
50
50
30
Limestone process
._ .,.* ._..
13,031,000
11, 3^,000
23,083,000
??,6oo,ooo
25,163,000
27,3^3,000
55,133,000
.57,725,000
21* ,267, 000
20,532,000
s /VU
.r/_K.7
65.2
56.7
U6.2
1*5.?
50.3
^.1
35.1
37.7
k8.5
1*1.1
Lime
1
11,71*9
13,036
26,027
20,232
22,1*22
21*, 272
38,133
32,765
21,586
18,323
process
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
|/kW
58.7
65.2
52.1
1*0.^
1*1*. 8
1*8.5
33.1
3?.3
1*3.2
36.6
Magnesia process
$
l!«,139,000
I1*, 572, 000
26,026,000
22,953,000
26,1*06,000
29,355,000
38,717,000
38,865,000
25,568,000
_
J/kW
70.7
71.9
52.1
1*5.9
52.3
53.7
58.7
33.9
51.1
—
Sodium process
$
16,198,000
17,11*9,000
31,208,000
26,706,000
50,1*91,000
33,709,000
1*7,721,000
1*5,832,000
29,127,000
—
$/kW
31.0
85.7
62.1*
53.^
6l.O
67.1*
1*7.7
U5.8
58.3
^
Cat-Ox process
$
19,557,000
17,735,000
37,907,000
Ii2, 520,000
1*2,736,000
1*2,923,000
62,913,000
69,889,000
-
—
$AW
97.7
33.7
75-8
35.0
35-5
35.9
62.9
69.9
-
•>
25 29,996,000 6o.O 26,090,000 52.2 32,213,000 &*.!» 37,957,000 75-9 !*3,8l6,OOO 87.6
30
30
30
30
25
30
3,263
12,935
15,^73
17, -31
13,657
23,331*
,000
,000
,000
,000
,000
,000
M.3
25.9
50.9
35-0
37.3
23. «*
9,1*82,000
15,961,000
13,11*3,000
19,361,000
21,317,000
26,31*1,000
1*7. 1*
31.9
36.3
39-7
1*3.6
26.3
8,361,000
12,695,000
l6,oGo,000
13,765,000
20,376,000
23,656,000
U4.3
25.1»
32.2
37.5
1*0.3
23.7
10,32l*,OOO
15,196,000
18, 9^9, 000
21,893,000
2l*,'*1*5,000
28,765,000
51.6
30.fc
37.9
1*3.8
1*8.9
23.8
13,069,000
28,067,000
23,277,000
23,1*1*9,000
32, 821* ,000
1*6,356,000
65.3
56.1
56.6
56.9
65.6
»*6.l4
a Midwest plant location represents project beginning mid-1972, ending mid-1975. Average cost basis for scaling, mid-1971*. Minimum la
procesc storage; only pumps are spared. Investnent requirements for disposal of fly ash excluded. Construction labor shortages vith
acconpanying overtice pay incentive not considered.
All Cat-Ox installations require particulate removal to 0.005 gr/scf prior to entering converter. Because existing units are assumed
to already =eet EPA standard!) (0.1 lb particulote/MM Btu of heat Input). Only incremental additional precipitator is required.
-------
1. Lime slurry
2. Limestone slurry
3. Magnesia slurry - regeneration
^. Sodium solution - S02 reduction
5- Catalytic oxidation
Shewn in Figures 12 and 15 are the effects of unit size and sulfur con-
tent, of coal OP. investment for installations on new units. In comparing
these rankings, it should "be pointed out that the lime slurry process
does not provide facilities for calcining limestone, and includes a
minimum of facilities for storage of the calcined material.
An important result derived by comparing the investment pro-
jections in this study is the relatively minor investment savings
realized (3-2-^. 5 A") in designing for So^- S02 removal as compared to 90/s.
The projected investments for limestone and lime slurry pro-
cesses designed for offsite solids disposal represent savings equivalent
to approximately lS>% of the projected onsite investment. Hovever, these
projections do not include the capital for offsite waste treatment
facilities.
Accuracy. In reviewing the capital investment estimates for
these five processes, it should be understood that the base case process
definitions and estimates represent a generalized cost. As an indication
of how the project scope and corresponding investment could vary, the
effect of "add ons" to the limestone slurry process base case estimate
is shown in Table 12. Such "add ons" as delays in equipment delivery,
physical space limitations, additional redundancy requirements, taller
stacks, inclusion of a closed-loop fly ash disposal pond, and overtime to
accelerate the project completion schedule could more than double the in-
vestment for the limestone slurry process. Excluding these additional
costs, but considering only the data available for this appraisal and the
depth of investigation,the estimated investment accuracy for each of the
processes is given in Table 13.
Table 13. PROJECTED INVESTMENT ESTIMATES ACCURACY
BASED ON AVAILABLE DATA AND DEPTH OF INVESTIGATION
Process £• range
Limestone slurry +20 to -5
Liine slurry +20 to -10
Magnesia slurry - regeneration +25 to -15
Sodium solution - regeneration +25 to -10
Catalytic oxidation (Cat-Ox) +20 to -10
184
-------
100
00
Ln
•P
w
o
o
-P
c
0)
s
•P
w
+5
•H
C!
25
0
Limestone Slurry Process X
Lime Slurry Process A
Magnesia Slurry-Regeneration Process 0
Sodium Solution-S02 Reduction Process Q
Catalytic Oxidation Process n
S in coal
S02 removal
0
200
^00 600 800
Power unit size, MW
1000
Figure 12. New coal-fired, units - the effect of power unit size on
unit investment cost, dollars per kilowatt
-------
100
Limestone Slurry Process X
Lime Slurry Process A
Magnesia Slurry-Regeneration Process 0
Sodium Solution-S02 Reduction Process Q
Catalytic Oxidation Process £
90% SQS removal
00
-P
W>
O
O
-P
c
IP
a
•P
w
OJ
•H
•P
50
25 -
o
0 1 2 3 ^ 5
Sulfur in coal, %
Figure 13• 500-MW new coal-fired units - effect of sulfur content of
coal on unit investment cost
-------
Table 12. LIMESTONE SLURRY PROCESS INVESTMENT WITH MODIFIED PROJECT SCOPE
Investment,
BASE INVESTMENT - LIMESTONE SLURRY PROCESS (including Fly Ash Removal But Not Disposal)
500-MW new coal-fired unit burning coal with 3-5% S, 12% ash, 90% S02 removal, 30-year life
127,500 hours operation, onsite solids disposal, proven system, only pumps spared, DO
bypass ducts, experienced design and construction team, no overtime, 3-year program, 5% per
year escalation, mid-197U cost basis for scaling 50-30
A. Overtime to accelerate project or cover local demand requirements
(50% of construction labor requirements) 3-20
B. Research and development costs for first of a kind process technology
(as allowed by FPC accounting practice) 5.00
C. Power generation capital for lost capacity (normally covered by appropriate
operating costs for power used in process) '••JO
oo
-^ D. Reliability provisions with added redundancy of scrubbers, other equipment,
ducts and dampers, instrumentation for change over (assumes no
permission to run power plant without meeting S0$ removal emission
standards at all times) T.OO
E. Additional bypass ducts and dampers. 1.00
F. Retrofit difficulty—moderate, space available beyond stack, less than
three shutdowns required for tie-ins, field fabrication feasible 10.00
G. Fly ash pond including closed-loop provisions 5-50
H. 500-ft stack added to project cost 6.00
I. Air quality monitoring system, 2-15 mile radius, 10 stations 0-70
J. Cost escalation of 10% per year instead of 5$ b.oO
K. Possible delay of up to 2 years in equipment and material deliveries
(1977 completion instead of 1975 ) 1?/.°°
Total W'™
-------
Allfji.-1?I Operatin~ Costs
For the annual operating costs covering 7000 hours per year
operation, the following types of displays are used for each process:
Su~.7iarized effect of individual process areas on total costs--
presented for base case and existing 500-MW limestone case. See
Tables 1^ and 15.
Case variations—operating cost summaries presented for each of
the l6 case variations. See Tables 16-21.
Table 22 presents the total average annual operating costs for
the l6 case variations of all five processes excluding credit for any by-
products produced. Corresponding to new 3.5£ S coal-fired units, the
relative ranking of average annual operating costs for the processes
(base case) is as follows:
1. Limestone slurry
2. Lime slurry
3- Catalytic oxidation
k. Magnesia slurry - regeneration
5. Sodium solution - S02 reduction
Some important effects are not readily seen in comparing the
ranking of processes for only one plant size and sulfur level. As
determined in this study, the relative ranking of operating costs of the
various processes is sensitive to changes in the assumed sulfur content
of the fuel. As an illustration, a 500-MW, 1/6 S oil-fired power unit
using the catalytic oxidation process has the highest operating cost of
the five processes; but for fuel oils with sulfur contents greater than
^.O/o, this process has the lowest cost. Although the sodium process is
not the lowest cost process for any of the cases presented in this study,
it becomes more competitive for low-sulfur fuel oil-fired units. Further-
more, at the expense of small additional investment, its operating cost
could be lowered by designing the process with multiple-effect evapora-
tors thereby reducing the overall energy requirements about l8/c.
In all of the projected annual operating cost estimates,
capital charges are the most significant components followed by mainte-
nance, energy, and raw material costs in varying orders of importance.
For the magnesia and sodium processes which require thermal energy for
regeneration of absorbent, steam or fuel oil are the second most signifi-
cant components. Lime cost (which requires thermal energy in preparation)
is the second most predominant item for the lime slurry process, whereas
maintenance is second in predominance for the limestone slurry and
catalytic oxidation processes.
188
-------
Table lU. LIMESTONE SLURRY PROCESS - TOTAL AVERAGE ANNUAL OPERATING COSTS
BASE CASEa SUMMARY - AREA CONTRIBUTION ANALYSIS
CO
VO
Mwrtrai
°£z£rmMM
AattMl qwwttfy. tow
AMU! cert. S
MMi n« MMI COM
Aural IJ»— 017. «•»*»
AoDoal COM. S
UttlM
ATLi <)u.air. M ft
Aan»looM.I
rroomwmr
A«m fetal Mk4 «M*wW)
totMt of o*«ct MI im»Ml
Anulootf.1
A—lym
AaoMi omMtttty. br
Anorfoott.1
S^ionl uj«i«am n»a
»— «— '
Ann* oonl duc|> 11 14 9»
ofeiDflil m«ilm»l
OrrtxW
Ptut. 20«oreaonmmco*a
Aaiwunnt. 10% of
opontmt boat
idtoul imlnct OOM>
Tom •n.d oo««fct COM
r*oM of Ml a-ul oond^ c»
1U-
••Hrtak
Total Hndluw
16,069,000 419,000
4OO/B.
175.000
700.000
700.000
4.170
))/«00
0.01/Mpl
ooio/un
190^300
3^00
(
23JOO
\ioon*
ISM
11.200
18.200 (2JOO
7IUOO (2JOO
97.700
jjtm njoo
JJOO
3/00 I11JOO
7JI.JOO I75JOO
•1 9.37 2JI
r-t
prrpttnooa
899.000
6.470
51.700
5.150JOO
51JOO
1
71JOO
174 JOO
174 JOO
209.800
14.800
S^JOO
249.800
4 24.000
SJO
rwtk«Jib>
KTUbbmt
1.201.000
4^00
39JOO
«.440X«00
(4.400
10
320.000
7(0
9,100
4)2.700
432.700
747.400
MJOO
3.900
837.800
1J70JOO
U-.9
SO,
mubbtri
4.143000
4»0
39 JOO
174.700
14.000
!2J 30.000
225 JOO
11
322.000
1.140
13,700
814.200
8I4JOO
1.107.000
1(2^00
1,900
1.273.800
2.018.000
27.11
KtlMWI
556.000
1J50
10.000
492 JOO
345400
*
44 JOO
399 JOO
399JOO
129 JOO
79.900
1,000
210.100
1 10 JOO
792
Fui
854.000
1.337.000
IJ50
10.000
43.0W.OOO
430.MO
8
MJXM
508.800
5MJOO
199 JOO
101 .WO
1.000
302.000
810.WO
10J)
Cila»B
dupouj Ubltlte* SCTTKC*
3.911.000 61000 631.000
6.143.000 105.000 999.000
JJ40
24.700
7S.WO
(.000
700.000 2)0.000 220.000
7.000 2JOO 2JOO
(
2)5.000
)80
4,(00
279.300 2.300 2.200
279JOO 2JOO 2JOO
915.300 15/00 148.900
55.900 500 400
!,700
971.900 16.100 149.XW
1JJJJOO 11.400 151JOO
1(27 OJ4 1.97
Tout Too*
faalitia q«mt)bM dolUn i
765000
175.000
7oaooo
700X100
2*J»
2IOJM
492JOO
345^700
ISO JOO
10 JOO
71.740.000
717.400
1
UUJOO
3JOO
45/00
2/93,700
JJ93.700
171 JOO 3.749JCO
53J.700
21.000
171 JOO 4J09.000
178 JOO 7.702.700
2J2
£2Z,
BOWuf OOM
tat
tm
in
4U
tJ*
MJQ
l(*9
OJ9
3497
440*
44 U
(»9
0-27
55.94
moo
DoQjui/to* Catiymifikn OoflMVtOB
cod bunMd WiBiAWIi Bra hal Bt^at wtfur renra*«d
t )4J
"
1)* 61
«>l.90»JO,
KonMnmf ttfc of pow piiAt. 30 rr.
Col bunol. 1 J12JOOtlooVrt. 9.000 tra/in
Suet (Mrctwst 10 175*P
Pow atvl cmumu 09*. 7jOOO br/yr.
M^««f) pbuit loeaoott. 1973 op«r«an| coct*.
Silttt mx»«L 35 J80 lofu/rr nuiU dupoM. 206.000 tooVn
t Kid OOW1OBC cxM for dl^towl of fly «h
todlm omrt »T*1» -«••
-------
Table 15. LIMESTONE SLURRY PROCESS - TOTAL AVERAGE ANNUAL OPERATING COSTS
EXISTING CASEa SUMMARY - AREA CONTRIBUTION ANALYSIS
Toul
Dtrecl capital uwncmcnt. $ 14.116.000
Total capital ur-nfmeal. I 21088.000
0DTC* Corn Htm
M-TT-1 TIT Tnnmal Ulu< Cotl- * ««™li
Lunotonc 4 00/tos
Annual quantity, torn 178.900
Annual con. 1 711600
Subtotal raw material coal 711400
CuU'tiuoo com
Operating, labor and •Jper*taw>ai I OQfma»4tf
Annual quantify, mtn-tu
Annual coit, S
Fud 04 (So 6J 0.21/pl
AnnualtivaaatT.pl
Annual con. S
rroorm-aw 008/n pi
Aa/iual qua/>urr. M sal
Annual coit. S
EtectTKirr 0010/VWb
Annual quantity. a Wo
Annual con. S
Maintenance (labor tod wper*mof)l
Percent of direct urvcatrneat
Annual con. S
Analyiea UOOfttt
Aonualquaaary.hr IJ20
Annual con. S 18200
Subtotal coo»eruoa ceeta 18.200
Subtotal direct com 7] 1400
ttutrret Cao
A*craf« capital charfc* at 1 1.3%
of cipttAl t/neftzncat
Rant. 20% of cofrrtmon cocta 1400
Adnunstribve, 10% of
open tai| Ub«.OOU
6480
11.400
SJ60.000
12.600
8
so.ooo
1M.OOO
IM.OOO
UOJOO
37JOO
1.100
292.800
478.800
607
so,
tc nibbing
1.577.000
t.760
70.100
17^400
IOOO
U.010000
2JOJOO
12
«Jt 213
907
907
l.M
12.1]
OJ»
831
I4JI
0-Sa
)t 2t
«J>
44 75
7U
on
5267
100.00
DoOan/toa Crou/millkMi DoQan/UM
coa/ burned KOU/lMl
£qur*al*nt total otat oomtins coat 188 224
Brti (leal input uti/uf rtmo*«d
2411
211 17
309-I4V ei»QAf co^n/cd po»«J. 16^-40 totiVyt. »oUii dupouL IlOiOO totW)T ulcium totkU in^fudlni ontr h
lirvntPiail
-------
Table l6. LIMESTONE SLURRY PROCESS
£
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECOIOMICS
(500-MW new coal-fired power unit, 3.5£ 8 in fuel;
90?t S02 removal; onalte solid* disposal)
Direct Costa
Delivered raw material
Limestone
Subtotal raw material
Annual quantity Unit cost, $
175.0 M tons U.OO/ton
Conversion coats
Operating labor and
supervision
Utilities
Steam
Process water
Electricity
Maintenance
Labor and material, .08 x l6,o69,OOO
Analyses
Subtotal conversion costs
Subtotal direct costs
26,280 m*n-hr 8.00/man-hr
1492,800 M Ib
250,300 M gal
78,7l!0,COO kWh
0-70/M Ib
0.08/M gal
0.010/kWh
Total annual
cost, $
700,000
700,000
210,200
3^5,000
20,OOO
787,^00
1,285,500
2,693,700
3,393,700
Percsit of
total Jnnual
operatic cost
9. OS
9-09
2.73
U.1.8
0.26
10.22
16.6?
W..06
Indirect Costs
Average capital charges at I1».95t
of total capital investment
Overhead
Plant, 20^ of conversion costs
Administrative, lO'jt of operating labor
Subtotal indirect costs
Total annual operating costs
Dollars/ton
coal burned Mllls/kHl
Equivalent unit operating cost 5-87 2.20
3,7'<9,300
538,700
21,000
i», 309, ooo
7,702,700
Cents/million
i Btu heat input
21.. 145
1.8.68
6-99
100.00
Dollars/ton
sulfur removed
211.. 68
Basis:
Remaining life of power plant, 30 yr.
Coal burned, 1,302,500 tons/yr, 9,000 Btu/kWh.
Stack gas reheat to 175*F.
Power unit on-stream time, 7,OOO hr/yr.
Midwest plant location, 1975 operating coets.
Total capital investment, $25,l63,OOO; subtotal direct Investment, $16,069,000.
Working capital, $572,600.
Investment and operating cost for disposal of fly ash excluded.
191
-------
Table 17. LIME SLURRY PROCESS
VENTUKT - VENTURI SCRUBBING SCHEME
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS*1
(50O-MWnew coal-fired power unit, J.5% 3 in fuel;
removal; onslte solids rilspoial)
Total annual
_Annual quantity Unit coat, $ cost, ?
Direct Coats
Dellverci raw material
Lime
Subiotal row material
Conversim costs
Operating labor and
supervl a Ion
Utilises
Stcnn
Process water
FJ.ec',rlclty
Mslnteinnce
Labor and material, .08 x I1* ,318,000
Analyses
Subtotal conversion costs
Subtotal direct costs
81.2 H tons 22.00/ton
22,}20 man-hr 8.00/man-hr
li90,OOO M Ib
2l« 1,900 M gal
7>4,100,000 XWh
178,600
Percent of
total annual
operating cost
22.03
22.05
2.20
0.70/M Ib
0.08/H gal
0.010/kWh
3I'3,000
19,!400
71U,000
I,ll45,l400
36!500
2,1463,900
!i, 250, 300
14.23
0.2li
9.15
Il4.ll4
0.145
52. '46
Indirect Costs
Average capital charges at I1*.9^
of totfl capital Investment
Overhen«
Plant, 20^ of conversion costs
Administrative, lO^- of operating labor
Sibtotal indirect costs
T'taJ annual operating costs
3,3140,900
1492,800
17,900
3,351,600
B,101,9OO
1OO.OO
Equivalent unit operating cost
Dollars/ton Cents/million Dollars/ton
coal burned Hllla/kWh Btu heat Input sulfur removed
6.17
25.72
225.61
Basis:
Remaining life of power plant, 30 yr.
Coal burned, 1,312,500 tons/yr, 9,000 Btu/kWh.
Stack gas reheat to 175°F.
Power unit on-stream time, 7,OOO hr/yr.
Midwest plant location, 1975 operntlng costs.
Total capital Investment, $22,1422,000; subtotal direct investment, $l*i ,313,000.
Working capital, $714|4,000.
Inveotmcnt and operating cost for disposal of fly ash excluded.
192
-------
Table 18. LIME SLURRY PROCESS
VENTURI - VENTURI SCRUBBING SCHEME
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
a
(500-MW ex 1_ctlna coal-fired power unit, 3-5^ 6 In fuel;
"S0p removal; onslte solids disposal)
Direct Cocts
Delivered raw material
Lime
Subtotal rnv material
Annual quantity
83.0 M tone
Conversion costc
Operating labor and
ou}x.-rvlsion
Utilities
Fuel oil (No. 6)
Process water
Hectrlclty
Maintenance
Labor and material, . OS x 15,913,000
AnnlyneB
Subtotal conversion costo
Subtotal direct costs
22,320 mun-hr
14,236,000 gal
2'eratlng labor
Subtotal Indirect coetB
Total annual operuti nf. cor.tr-
EqulvaJent unit oj yr.
Coal burned, l,3'il,7OO tone/yr, 9,200 Htu/kWh.
f!tnck pno reheat to J75°F.
Power unit on-ntream time, 7,OOO hr/yr.
Midwest plant location, 197'; operating costs.
Total capital investment, $26,027,000; eulitotal direct Inveutment, $1';,91>,OOO.
Working capital, $863,300.
Jnventraent and operating cost for removal find dinpocal of fly aoh excluded.
193
-------
Table 19. MAGNESIA SLURRY - REGENERATION PROCESS
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS*"
(50O-MW now conl-flred power unit, 3.51k 3 in
9O% SOa removal j 110, kOO tons/yr
fuel;
Direct Costs
Delivered raw material
Lime (1st stage neutrnll zation)
Magnesium oxide (96%)
Coke
Catalyst
Subtotal raw material
Conversion coats
Operating labor and
supervision
Utilities
Fuel oil (No. 6)
Steam
Heat credit
Process water
Electricity
Maintenance
Labor and material, .0
Analyses
Subtotal conversion costs
Subtotal direct costs
Indirect Costa
Average capital charges at I
of total capital investment
Overhead
Plant, 2O?t of conversion costs
Administrative and marketing,
U.% of conversion costs
Subtotal Indirect costs
Total annual operating costs
Equivalent unit operating cost 83.
Annual quantity
;ion) 15*1 tons
1,036 tons
763 tons
1,800 liters
39,200 man-hr
5,356,000 gal
^0,000 M Ib
20,300 MM Btu
2,207,500 M gal
71,060,000 XWh
: 16,35^,000
.a
.*
ists
«>
108 tS
Dollars/ton Dollars/ton
10O% HpSO* coal burned
Unit cost, $
26.00/ton
155-00/ton
15.00/ton
1.65/liter
8.00/raan-hr
0.25/gal
0.70/M Ib
-0.60/KM Btu
0.0^ /M gal
0.010/XWh
Total annual
cost, £
3,500
168,300
n, 'too
3.000
186,200
313,600
1,231,900
300,000
(12,200)
88,500
710,600
I,lli3,l400
'102,' ooo
5,885,600
"4,071,800
5,931',500
777,100
U27,'iOO
5,159,000
9,210,8oo
Cents/ml Ulion
HlUs/kWh Btu
heat Input
Percent of
total annual
opt-mtlng cost
O.C^
1.85
0.12
0.05
2^02
3.-0
13-39
3.3^
(0.13)
0.96
7.71
12. tl
1.11
¥2.19
I.1..81
1.2.71
8.W
ij.6i
55.79
100.00
Dollars/ton
sill fur removed
7-02
2.63
29.
255. "O
8 Basis:
Remaining life of power plant, 50 yr.
Coal burned, 1,512,500 tons/yr, 9,000 Btu/kWh.
Stack gas reheat to 175*F.
Power unit on-streara time, 7,000 hr/yr.
Midwest plant location, 1975 operating costs.
Total capital Investment, $26,«06,OOO; subtotal direct Investment, $l6,331* ,OOO.
Working capital, $721,000.
Investment and operating cost for disposal of fly ash excluded.
194
-------
Table 20. SODIUM SOLUTION - S02 REDUCTION PROCESS
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS"
(50O-MW new coM-rired power unit, J.5«f 3 In fuel;
removal; >2,7OO ton«/yr sulfur)
Direct Coats
l>e!1vLTcd raw mnterlala
Lime (1st sta^e neutralization)
Soda esh
Antloxidant
Catnlyst
Subtotal re" materials
__Annun_l__g_t]«n_tltjj^_ Unit c
13^.1 tons
9,300 tons
317,100 Ib
Conversion costs
Operating labor and
supervision
Utilities
NaturaJ. gas
Steam
Heat credit
Process water
Electricity
Maintenance
Lnbor and material, .06 x l3,86l,000
Analyses
Subtotal conversion costs
SubtotaJ direct costs
1*6,5CX) ronn-hr
509,500 mcf
2,137,800 M Ib
62,900 MM Btu
9,955,''00 M gal
7^,190,000 kWh
Pel cent of
Total annual total annual
cost, $ operating coet
26.00/ton
52.00/ton
2.00/lb
8.00/mfin-hr
1.00/mcf
0.70/M Ib
-0.60/KM Btu
O.O2/M gal
0.010/kWh
3,500
** ?3 ,600
6 JJ< , 200
I-?, 000
1,133,300
372,000
509,500
1,^96,500
(37,700)
199,100
7^1,900
1,131,700
109,900
^,522,900
0.03
1|.17
5.^7
0.10
9.77
3-21
'•.59
12.90
(0.33)
1.72
6.39
9-75
0. 9^
3T.98
5,656,200
^8.75
Indirect Coats
Average capital charges at 1<< . 9^-
cf total capital investment
Overhead
Plant, 20^ of conversion costs
Admin' strative and marketing
Subtotal Indirect costs
Total nnnual operating costs
Dollars/ton
product sulfur
Equivalent unit
operating cost 35'4.79
!», 5^3,200
9o1',6oo
^97^500
5,9"5,500
11,601,500
Dollnra/i,on Cents/ciill ion
coaj. burned Mills/kWh Btu heat input
8.8U 3.31 36.83
59-16
7.3o
51.25
100.00
Dollnrs/ton
sulfur removed
323.31'
Basis:
Remaining life of power plant, 30 yr.
Coal tmrned, 1,312,500 tons/yr, 9,000 Btu/kWh.
Stack gas reheat to 175°F.
Power unit on-atream time, 7,000 hr/yr.
Midwest plant location, 1975 operating costs.
Total capital investment, $30,^91,000; subtox.nl direct investment, $18,361,000.
Worklns caoital, $1,015,500.
Investment »nd operating cost for disposal of fly ash excluded.
195
-------
Table 21. CATALYTIC OXIDATION PROCESS
TOTAL AVERAGE ACTUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
(SOO-MW new, coal-fired power unit, 3-5* 3 In fuel;
yo* S0e removal; LOy.VOO tono/yr 1(M% H;,SO« )
Toted annual
Direct, Custs
IVflivered raw ma'
Catalyst
Suttotal rav ir.«U-rial
Conversion costs
C-p<>rntin<< labor and
sup'.'.-vis ion
Utilities
_Atimjal quantity Unit cost, $ _ coatj
i,700 liters 1.65/liter 17JU3°°.
man-hr B.OO/man-hr 63,100
Percent of
total annual
operating cost
1-95
0-71
Steam 179,000 H Ib
Heat credit 967,000 MM Btu
Process water 312,000 M gal
Electricity 90, ^O, 000 KWh
Ma! nt enance
Labor <0« coal burnKii
Equivalent unit operating cc>st iJ0.7t> 6.1(<
0.70/M Ib 125,300
-0.60/MM Btu (592,200)
0.08/M gal 25,000
0.010/kWli 90l',1tOO
l.Ol't.TOO
'i^jOOO
1,5:53,300
1,761,100
6,367,700
317,700
ijT^oo
7, 11? 366
fl,fl73,900
Cents/million
Hills/kWh Btu h«.-8t lujiut
2.5,
1.1*1
(6.67)
0.28
10.19
H.^3
0. Mi
177B9
19.9k
71.76
5-59
'i.B?
8o-16
100.00
Dollars/ton
sulfur removed
*7.*
* Basis:
Remaining life of power plant, 50 yr.
Coal burned, 1,31.°,(XX) tonfl/yr, 9,000 Btu/kVfti.
:>tnck pan reheat to 17!;<>F.
l\jwer unit on-ntreir/yr.
Hliiwr:nt p] nnt location, J'TT.' npcrntlnff coats.
Totiil cnpltn] jnvefltmcnt, JJi?,73f>,OCX); subtotal direct investment,
Working capita], JV.'^,700.
Jnvcstmont and operating fdst for dlapoaa] of fly aah excluriati.
196
-------
Table 22. SUMMARY - TOTAL AVERAGE ANNUAL OPERATING COSTS
a
Excluding Credit for Byproducts
Case
Coal-fired power unit
9Ot SOa reioval; onsite solids disposal
200 HW new, 3.5J sulfur
20C KV existing, 3-5* sulfur
500 KW exls'lr^, 3.51 sulfur
500 KV aev, e.Ot sulfur
500 MV aev, 5.5it sulfur
500 KW nev, 5.04 sulfur
1000 MW existing, 3-5J sulfur
1000 KW oev, J.5J sulfur
3o% 302 reaoval; oa.°lte solids disposal
500 KW new, 3.5* sulfur
9Ot SOj ressoval; offsite solids disposal
500 MW nev, J.5J sulfur
Set 3O2 reaoval (existing unit
without existing particulste
collection facilities)
50C W existing, 3-5* sulfur
pover unit
Years
life
30
20
25
30
30
30
25
30
50
30
25
Llae stone
Total annual
operating
^oat, $
3,921,500
3,367,100
7,392,600
6,77'. ,700
7,70S,?00
3,522,200
12,752,900
11,371., 100
process
Mills/
XWh
2.30
2.76
2.26
l.QA
2.20
2. ''3
1.32
1.70
Lime process
Total annual
operating
cost, $
k, 163, 900
U, 322, 000
9.612.UOO
6,915,100
3,101,900
9,170,100
15, 301,'' oo
12,553,100
Mills/
kwh
2.97
2.75
1.93
2.31
2.62
2.19
1.79
Magnesia
Total annual
operating
cost, $
li, 776,300
5,091,200
9,607,900
7,523,'' oo
9,210,300
10,763,500
15,1.31,900
l<" ,31.7 ,000
process
Mills/
iwh
3.6k
2.75
2.15
2.63
3.03
2.21
2.05
Sodlua process
Total annual
operating
cost, $
5,971,700
7,377,700
1^,653,000
9,101,700
11,601,500
13,?35,500
25,113,500
13,391.300
Kills/
kVh
k.27
5.27
k.19
2.60
3-31
"..00
3.59
2.65
Cat-Ox process
Total annual
operating
cost. $
'",232,700
5,3ii9,koo
12,599,600
3,301,200
3,373,900
3,91.0,500
2i,'.6o,3oo
13,957,600
Mill*/
kUb
5.02
J.Jk
2.51
2.51"
2.55
3.07
1.99
7,373,000
3,576,500
9,573,1.00
2.11
2.39
7,306,900
3,#U,000
2.23
2.1.7
3,739,700 2.51 10,331.,300 3-XO
2.71. 9,723,300 2.73 U,227,300 3.21 16,389,200 !>.68 X5,593,300 3.39
5OJ SOa retsjval; anslte solids disposal
ZjQ KW new, 2.;i sulfur
5CO ;, 269,200
5,35^,700
3,305,100
10,6*0,500
10,261,600
13,636,200
5.05
1.67
2.57
5.0k
2.93
1.96
2,750,100
5,71.3,600
5,677,500
5,565,100
11,126,100
3,911,900
1.96
1.6^
1.62
1.59
3-13
1.27
* ?3v»r ur.lt on-streaa tloe, 7000 hr/yr. .Midwest plant location, 1975 operating costs. Investment and operating cost for disposal of flj ash excluded.
-------
The effect on annual operating costs of designing for 80$ S02
removal as compared to 90% is relatively small. Operating cost savings
range from 3-6 to k.2% of the projected operating costs at 90% removal
for the limestone and lime slurry processes.
The effect of excluding product credits from the annual
operating cost estimates is illustrated in Table 23 which compares the
projected annual base case operating costs for the five processes with
the revenue from sale of byproducts.
Table 23. COMPARISON OF PROJECTED ANNUAL OPERATING COSTS AND
PRODUCT CREDITS FOR BASE CASE ESTIMATES AT 7000 HOURS ANMJAL OPERATION
Process
Total Net annual
average annual Annual credit operating
operating cost, $ for byproducts, $a cost, $
Limestone
Lime
Magnesia
Sodium
Cat -Ox
a
7,702,700
8,101,900
9,210,800
11,601,500
8,873,900
1
883
,077
659
,200
,500
,^00
7
8
8
10
8
>
)
,
,
,
702
101
327
52*4
2lU
Corresponds to credit of &3/ton 100$ H2S04 as 98$ H2S04
magnesia slurry process; $25/short ton sulfur, $20/ton
Na2S04 for the sodium solution process; and $6 /ton 100$
17 O/^ . Q
-------
Table 21*. LIMESTONE SLURRY PROCESS. 500-MW NEW COAL-FIRED POWER UNIT,
3.5$ SULFUR IN FUEL, 90% SOP REMOVAL, REGULATED COMPANY ECONOMICS
FIXED INVESTHENT'
25163000
VO
vO
YEARS
AFTER
POKER
UNI t
START
1
2
3
4
1
6
7
a
9
1 fl
1 1
12
j 3
14
J C
16
17
n
1 9
?n
21
22
23
24
j e
26
27
26
29
in
TOT
ANNUAL
OPERA-
T1CN,
XW-HR/
Kk
7000
7000
7COO
7COO
JDDQ-.
7000
7COO
7000
7000
20.0.0.
5000
5COO
5COO
5COO
5nr.n
35CO
3500
3500
35CO
35J1Q
15CO
1500
1500
1500
ISQA
1500
1500
1500
1500
-15QQ-
127500
LIFETIME
PGKER UNIT POkER UNIT
HEAT fUEL
REQUIREMENT, CONSUMPTION
MILLICN uu
/YEAR
3I5CCOCO
315CCOOO
315COOCO
3150COCO
315000.00.
315COOCO
315000CO
31500000
3150COCO
^15[j^0p3
225COOO?
225CSOCC
225CGOOO
225CCOCO
22S.CC2Cfl_
1575UOCO
157500CO
1575','OGO
15750000
1525. rQCQ
6750000
67500CC
675000C
6750000
hIS.CQ.Ul-
6750000
6750000
67500CO
675COCO
6J5.GOO.O.
573750000
AVERAGE INCREASE
UOLLA8S
TONS COAL
/YEAR
131250C
1312500
1312500
1312500
1 3 L25.3Q
1312500
I31250C
1312500
1312500
1 31 2 50*3
937500
937500
937500
937500
3 J A5D*^
*56200
656200
656200
656200
65.L2 CO.
2f 1200
2312CO
261200
ie\2'lr>
2&L2.QQ.
<«I20C
2 "1210
2M200
231200
^ A i^Gfi
23505500
(DECREASE
PER TLN 3F
SULFUR
REMOVED
BY
POLLUTION
, CONTROL
PRLCESS,
TONS/YEAR
35900
35900
35900
3590C
3 5.2QQ
35900
35900
35900
35900
-15910
2560C
25600
25600
25600
256 Q.Q
17900
17900
17900
17900
1 79 10
7700
7700
7700
7700
izao
773P
77QC
770C
7700
_ 2iao_
653500
BY-PRODUCT
RA1E,
EQUIVALENT
ICNS/YEAR
WASTE
SOLIDS
206000
206COO
206CG3
206000
>0f.nnn
206000
206000
206000
206000
206QOQ
147100
147100
147100
147100
14 jir.n .,
I03COO
103000
103000
103COO
ininnn ,
44100
44100
44100
44100
,.44100. , -
44100
44100
441CO
44100
- - - dtlOJ) _
3751500
NET REVENUE
»/TON
WASTE
SOLIDS
0.0
0,0
0.0
0.0
n.n
0.0 .
0.0
0.0
0.0
Q.P
0.0
0.0
0. 0
0.0
Q-.0. _
0.0
0.0
0.0
0.0
Q.l)
0.0
0.0
0.0
0.0
o.n
0.0
o.c
0.0
0.0
. C - fi
TOTAL
OP. COST
INCLUDING
, REGULATED
ROI FOR
POkER
COMPANY ,
l/Yf AR
10320500
10146000
9971600
9797100
. . 9622f>Qn
9448200
9273700
9099200
89247CO
B 7. 50 300
7640900
7466400
7292000
7117500
6J4JQCD-
6029900
5855500
5681COO
55C6500
5332100—
4074300
3899800
3725400
355090C
33 7. 64QQ
3201900
3027500
2653000
267H50C
y Sfi4 | po
193110500
NET ANNUM
TOTAL INCREASE
NET (DECREASE!
SALES IN COST OF
REVENUE ,
t/YEAR
0
0
0
0
n
0
3
0
0
a
0
0
0
0
n
0
0
0
0
Q
0
0
c
0
n
0
0
0
0
0
POKER,
t
10320500
10146000
9971600
97971CO
}}tan
944(200
9273700
9099200
8924700
a I$Q 3CO
7640900
7466400
7292000
71 17500
fclilOQG
60299CO
5855500
56B1000
5506500
51121 O2
4074300
3899800
37254CO
35509CO
332fcfcQQ
32019CO
3027500
28530CO
26785CO
2S.Q41QQ
193,110,500
CUMULATIVE
NIT INCREASE
(DECREASE 1
IN CLST OF
POkER,
i
I032C500
20466500
30438100
40235200
&9. AS. 7 AOO
59306000
68579700
77678900
(6603600
9.5153.9.00
1029948CO
110461200
117753200
124870700
1 lldl 1700
137843600
143699100
149380100
154686600
1«,0?H jno
1642930CO
168192800
171916200
175469100
1 7 f fl4 5.5Qft
U2047400
U5074900
1*7927900
190606400
19 1110 500
1 IN UNI T 3PEIAT1NC COST
CTAL BURNEO
•MILS PER K UtkAIT-HCUli
CEM< PER MILLION
PD3CFSS COST
DULLAdS
OISUUMEO »T
PER TCN !)F
BTl HEAT INPUT
SLLFUR ,»E1'JVE
10- CX T!l IS) I!At Yf A* , D
0
"UAtS
t .08
3.03
33.66
295.50
78439900
0.0
0.0
0.0
0.0
0
8.08
3.03
33.66
295.50
78.439,000
LEVEII2EO IfiCfcfASE IDECRCASEI IH UNIT 'JPfesTIHG CCST (CUIVAICNT TO DISCCUNTEO SUCCESS CCSI OVER LIFE OF POKfR UN 11
(/CU'RI PEB '2N OF CfAL BORNE j 7.63 C.O 7.63
XULS PE« K IL^kATI-MiUH 2.86 C.O 2.86
CENTS PER "1LLICN BTL HEAT 1SPUT 31.77 0.0 31.77
COLLARS PER I UN OF SULFUR AMOVED 278.85 0.0 278.85
-------
Table 25. LIMESTONE SLURRY PROCESS, 500-MW EXISTING COAL-FIRED POWER UNIT,
3-5# SULFUR IN FUEL, 90$ S02 REMOVAL, REGULATED COMPANY ECONOMICS
YEARS ANKUAl
»f IfB OPESA-
PLUfcK TICN,
UMT Xw-HR/
START
K*
UNIT
HEAT
REQUIREMENT,
MILL I PS- UU
/YEAR
UNIT
FUEL
CtlNSUMPT H'N,
TCNS COAL
/Yf AR
SULflU
8Y
PCLLUTION
CCNTR3L
PSuCtSS,
TONS/YEAR
INVESTMENT I
PY-PRGOUCT
RATE,
EQUIVALENT
TCNS/YEAR
SOL IDS
230i>eooD
NET REVENUE,
t/ION
WASTE
SLLIDS
TOTAL
Df>. COST
INCLUDING
RECULAIEO
CO I FUR
POWER
COMPANY,
i/YEAR
TOTAL
SAL! S
Rfvfiue,
$/YEAR
NET ANNUAL
INCREASE
IOECREASEI
III COST CiF
PC»E»,
CUKULAT IVE
NET INCREASE
(DECREASE!
IN COST or
POWEI,
t
NJ
o
o
b
7
a
t.
1 'i
11
12
1 1
_ii
16
1 7
la
19
22
21
el
2 i
24
26
21
2t
2"
3D.
7CCO
icco
7COO
7COO
2CQO
5COO
50CO
5COO
5CCO
35CO
35CC
35CO
35CO
i < r Q
15GC
1SGO
15CO
1500
liCL
15CO
1*00
15CC
ISCC
1 trn
322C0003
322,C,COC1
322C3COO
322CIOCJ
32 2 GC OCa
< J3CCOCO
; joccoco
23CCCOCO
«3GCOOCG
iticcoc;
uiccorc
UKCXO
161CCOOO
1 i I ' "QfiQ
bocroc;-
69CCOCC
69CC7C7
6»c:occ
._-tSCiO-£.
bircoce
t 9 C oC C C
69( ^Ot <.
69CUOC-C
bSCCQCD
1 34(700
1 34170C
134170'J
13417'. 0
A 34 1 7/rt3
9567-51!
°56 3"H
95630P
956 JOO
67080C
67C8C.O
670»50
(-7GCOO
<• ?DP ^^'
2^7500
29 7500
2C75CO
2f ^ -3
-2i.liOJ
< • 7500
2 t 75 OC
28 750C
!t ?5^0
2 * 7 5 f T
367QO
3b73C
3673C"
367UO
3.^2 QC
262DC
2<>2 30
26230
210600
21CbOO
210600
210bOO
^1 n API
I504CO
15C400
150400
0.0
0.0
C.O
0.0
o n
0.0
0.0
0.0
2620P 150400 O.G
18300
lf3CO
18300
18300
i£\ on
7-»OC
7930
7900
7930
1D5300 0.0
1053CO
105300
105)00
insip.Q
45100
45100
45100
0.0
0.0
C.O
_CUQ
0.0
0.3
0.0
0.0
10293700
lOldfcCG
99C9500
">71 7400
t**? S40 Q
82t27CO
PC7C600
78765CO
76H6400
64658CO
6273700
608I6CO
5869600
t ^ o 7
Ll.Fu» 'fJVEO
8.67
3.32
36.13
316.95
7055GCCO
TO OISCLUNTEO PROCESS COST OVER
9.C7
3.09
33.61
295.06
C.
0.
0.
G.
Lift
C.
0.
0.
0.
0
0
0
0
0
Of
0
0
0
0
8.67
J. 32
36. 1 3
316.95
7C55C-000
POWER UNIT
8.G7
3.09
33.61
295.06
-------
Table 26. SUMMARY - CUMULATIVE DISCOUNTED PROCESS COSTS AND EQUIVALENT LEVELIZED UNIT INCREASE
(DECREASE) IN COST OF POWER OVER THE LIFE OF THE POWER
Including Credit for Byproduct!
Case
Coal-fired power unit
903» S02 resooval; onslte solids disposal
200 KW new, 5.5$ sulfur
200 MW existing, 5-5* sulfur
500 K' existing, 5-5$ sulfur
500 KW nev, 2.0* sulfur
500 MW oev, 3.5«f sulfur
500 K« new, 5.0J sulfur
1000 KW existing, 5-5$ sulfur
1000 Ktf new, 3.51 sulfur
Sot SOz reaoval; onslte solids disposal
500 MW nev, 3.5* sulfur
90$ SOa reeovttli offsite solids disposed
500 MW new, 3.5$ sulfur
90$ SOj reaoval (existing unit without
existing pertlculate collection
facilities)
500 MW existing, 3.5$ sulfur
Oil-fired power unit
Lisestone process
Lime process
Magnesia
Sodiua process
Cut-Ox process
Years
life
30
20
25
50
30
30
25
JO
30
30
Cumulative
present
worth net
Increase
(decrease)
In cost of
power, |
I'D, H2, 800
29,067,800
70,550,000
69, J^, 200
73,1-39,900
86,ii 26,900
111, 935, '•OO
120,015,500
75,259,300
90,l<26,200
Levelized
Increase
(decrease)
In unit
operating
cost,
Blll«/XVflic
3.66
^.73
J.09
2.53
2.36
3-15
2.U5
2,19
2.T1*
2.93
Cumulative
present
worth net
Increase
(decrease)
In cost of
power, b $
"•1,112,500
Jfc, 979,000
814,117,600
68,709,000
79,593,300
89,293,900
130,977,300
121,789,900
76,687,900
30,903,300
Levellzed
Increase
(decrease)
In unit
operating
cost,
oma/XHh
3-75
5.69
3-69
2.50
2.90
3.25
2.37
2.22
2.80
2.95
Cumulative
present
worth net
increase
(decrease)
In cost of
b *
power, $
M ,860,500
}6,106,200
78,292,200
71,503,600
31", 21-9, 500
95,621,900
121,1.56,200
126,808,000
3l,n9,Soo
_
I/evellzed
Incrense
(decrease)
in unit
operating
cost,
niUs/kWh0
"<.09
5.37
3."'5
2.6:
3.07
>.'*<)
2.66
2. 31
2.96
_
Cucule tlve
present
worth net
Increase
(decrease )
In cost of
b *
power, t
55,0><5,000
i'3,56B,joc
11J, 965,500
95,60^,900
1C*, 2?2,}00
121,660,300
iSS^et^OO
l6o,375,?00
96,2l>5,000
_
Levellzed
Increase
(decrease )
In unit
operating
cost,
aUls/y.Wac
5.02
7.90
5.00
3.12
3.30
k.iO
"•-«
2.92
3-59
_
Cumulative
present
worth p.et
Increase
(decrease)
In cost of
power,1" t
ill", 323, 500
"<2, "423, 000
106, 607, 900
95,730,300
S1-, 520, 900
92,305,500
IS 1,012 ,700
1^3, U7, 600
-
^
Levellzed
Increase
(decrease)
in unit
operating
cost,
ClllS /XW
">.o3
6.90
1-.67
3.1,0
3.U
• 3.53
3.77
2.70
-
.
25
57, 1^3,300 3.92 8l.,92»,200 3.72
93,875,800
130,715,900
5.73 119,12'', 300
5.22
90? SOj reaovai; onslte solids disposal
a
0
c
200 KW nev, 2.5* sulfur 30 28,231,000 £.58
500 MW new, 1.0$ sulfur 30 ue.lO^SOO 1.69
500 KV new, 2.5J sulfur 30 5l(,7'<3,900 2.00
500 KW new, U.oJ sul^a• 30 6l,3o3,l<00 2.25
500 W existing, 2-5^ sulfur 25 53,353,300 2.56
1000 KV aev, 2.5* sulfur 30 37,171,700 1.59
Basis:
Over previously defined power unit operating profile. 30 yr life;
Hldwest plant location, 1975 operating costs.
Icvestaent and operating cost for disposal of fly ash excluded.
Constant labor cost assuoed over life of project.
Discounted at 10$ to Initial year.
Equlvalri* »o discounted process cost over life of pover unit.
33,612
56,505
66,727
7<<,929
70,126
103, '•U
7000 hr -
,500
,800
,200
,500
,200
,900
10 yr,
3.06
2.06
2.1.3
2.73
3.07
1.33
5000 hr
30,089,900
11 '•,030,300
55,673,'' 00
65,572,300
61,393,300
85,962,i<00
- 5 yr, 3500 hr
2.7^
1.61
2.03
2.59
2.69
1.57
- 5 yr,
39, l1^, 900
55, 025, '"00
7i« ,20k, 600
91,337,900
32,952,700
US, 705, 700
1500 hr - 10 yr.
3.57
2.01
2.70
3.35
3-63
2.16
29,653,300
63, 57S 000
61,591,500
59,2-9,500
?6, 505, 200
96,399,300
c.
2.
2.
2.
!..
1.
"0
52
25
16
22
77
-------
Table 27. LIFETIME BYPRODUCT PRODUCTION AND CREDIT
Base Case, 500-MW New, Coal-Fired, 3.5$ S, JO Yrs Remaining Life
Net
Equivalent revenue, Cumulative revenue
lifetime production $/short Dis-
Process Product Short tons ton Actual $ counted $
Magnesia slurry -
regeneration 100% H2S04a 2,011,500 8.00 16,092,000 6,923,300
Sodium solution - Sulfur 595,000 25.00
S02 reduction Na2S04 2J7,000 20.00 19,627,500 3,^6,300
Catalytic ,
oxidation 100^ H2S04 2,002,500 6.00 12,015,000 5,163,900
£ As 9% H2S04.
As 80^ H2S04.
202
-------
o
B
(U
3
w
G
O
-P
800
600
Limestone Slurry Process X
Lime Slurry Process A
Magnesia Slurry - Regeneration Process
Sodium Solution - S02 Reduction Process
Catalytic Oxidation Process D
3«5# S in coal
90% S02 removal
127,500 hr lifetime operation
0
22.0
16.5
H
03
O
O
c
O
•P
-ea-
N;
O
U)
•P
OT
O
o
W)
G
•H
-P
0)
o
-P
•rH
G
^400
11.0
200
5.5
-p
co
O
O
bO
G
•H
P
cd
M
0)
-P
•rl
a>
a>
0)
a>
o
0
200
400 600 800
Power anit size, MW
1000
Figure I1*. New coal-fired units - the effect of power unit size on
levelized unit operating cost - regulated economics
-------
12
H
(0
o
o
c
o
-p
-p
CO
o
CJ
a
-p
a>
&
-p
•H
q
'd
a;
O)
10
8
Limestone Slurry Process X
Lime Slurry Process A
Magnesia Slurry-Regeneration Process 0
Sodium Solution - S02 Reduction Process
Catalytic Oxidation Process Q
90$ S02 removal
127,500 hr lifetime operation
3.75
•H
-P
OT
o
o
bO
c
•H
-P
cd
QJ
3.00
2.25
c!
3
o;
23 14 5
Sulfur in coal, $>
Figure 15. 500-14W new coal-fired units - effect of sulfur content
of coal on levelized unit operating cost - regulated economics
204
-------
TVA IN-HOUSE ESTIMATES
In addition to the more detailed estimates prepared for EPA, a
series of in-house studies have also been undertaken by TVA to determine
the attractiveness of several process options. These include limestone
scrubbing using benzoic acid as a slurry dissolution promoter; lime
scrubbing using a venturi - mobile-bed contactor rather than two venturi
stages; lime scrubbing including a limestone calcination unit; sodium
scrubbing with sulfuric acid production rather than sulfur; and two
double-alkali systems—one using ammonia scrubbing, the other sodium
scrubbing. Because the date available for the double-alkali process
evaluations were limited, the accuracy of these estimates is less than
the others.
Seven TVA case estimates were prepared. When evaluating lime
scrubbing with a mobile-bed contactor, both a new and a retrofit case
(500 MW, 3-5% S in coal) were estimated to determine the merit of using
such a single-stage device for S02 removal. For the other five process
options, only new systems were evaluated. Except for specific process
requirements, the same criteria used to prepare the EPA estimates were
used for these evaluations. The summarized investment and operating
cost results are given in Table 28. Additional comparative details are
given in Tables 29-^1.
CONCLUSIONS
Conclusions derived from the EPA-TVA detailed study results
are as follows;
Investment
1. For coal-fired systems, the lime scrubbing process has the lowest
investment and catalytic oxidation has the highest. Because lime
process definition specjfies venturi scrubbers and two stages are
required for S02 removal, the limestone and magnesia processes
become comparatively less expensive on oil-fired units; limestone
investment is the lowest and magnesia investment is very close.
All four wet scrubbing processes are within 21% of each other.
2. As sulfur content of fuel increases, the relative investment
ranking of the wet scrubbing processes changes; lime is still
lowest for coal-fired units but for oil-fired units, limestone
or magnesia investments are the lowest. Although these invest-
ments are very close throughout the range, magnesia investment
is slightly lower for the low-sulfur oil, and limestone investment
is slightly lower for the medium- and high-sulfur oils.
205
-------
Table 28. INVESTMENT AND OPERATING COST RESULTS FROM TVA IN-HOUSE STUDIES5
Process
Limestone-benzole acid alurry process
500 MW new 3.5$ sulfur
Lime slurry process ; veaturi-mobile bed
scrubbing scheme*5
500 MW new 3.5$ sulfur
Lime alurry process; mobile bed scrubbing
scheme01
500 MW existing, 3-5£ sulfur
Lime slurry process; venturl - venturl
scrubbing scheme with onsite calcination
500 MW new, 3.5$ sulfur
Sodium solution - HaSO^ production process
500 MWjww, 3-5£ sulfur
Ammonia-limestone/lime double-alkali
process
500 MW new 3.5$ sulfur
Sodium- limestone double -alkali process*1
500 MW new 5.556 sulfur
Years Capital investment
life . ._. $ . _' $/kW~
Average annual
operating costs
30 23,^73,000 1*6.9 T,i»39,000 2.13
30 23,363,000 1*6.7 8,11*6,000 2.33
25 20,327,000 1*0.7 8,270,600 2.36
30 27,'»52,000 5J»-9 8,101.900d 2.31d
30 30,139,000 60.3 11,05^,800 3.16
30 31,087,000 62.2 9,669,300 2.76
30 30,759,000 61.5 9,51*5,300 2.73
a Stack gas reheat to 175*F. Midwest plant location represents project beginning mid-1972,
ending mid-1975' Average cost basis for scaling, mid-1971*. Operating costs based on power
unit on-stream time of 7000 hours; costs correspond to 1975- Minimum in process storage;
only pumps are spared. Investment and operating coat for disposal of fly ash excluded.
. Construction labor shortages with accompanying overtime pay incentive not considered.
Limestone raw material cost, $l»/ton; lime raw material cost, $22.00/ton. Onsite disposal
c pond located one mile from power plant.
Assumes 93.7% effective electrostatic precipitator already installed.
Same as operating cost for lime slurry process in EPA study, since EPA study utilized esti-
mated production cost to establish purchased, cost of lime.
-------
Table 29- LIMESTONE - BENZOIC ACID SLURRY PROCESS
a
SUMMARY OF ESTIMATED FIXED INVESTMENT
(50O-MW new coal-fired power unit, 5.5$ 8 In
9O% SOP removal; onslte «olld« disposal)
8 in fuel;
Investment, $
Limestone receiving and storage (hoppers, feeders,
conveyors, elevators, and bins) It It 3,000
Feed preparation (feeders, crushers, elevators, ball
mills, benzole acid storage, conveyor, tanks, and pumps) 572,000
Partlculete scrubbers and inlet ducts (!t scrubbers In-
cluding common feed plenum, effluent hold tanks,
agitators, and pumps) 3,203,000
Sulfur dioxide scrubbers and ducts C4 scrubbers In-
cluding effluent hold tanks, agitators, pumps, exhaust
gas ducts to Inlet of fan and mist eliminators with
closed-loop wash systems end trap-out trays) 14,170,000
Stack gas reheat (^ indirect steam reheaters) 556,OOO
Fans (U fans Including exhaust gas ducts and dampers
between fan end stack gas plenum) 85^,000
Calcium solids disposal (onsite disposal facilities
including feed tanX, agitator, slurry disposal pumps,
pond, liner, and pond water return pomps) 3,772,000
Utilities (instrument air generation e.nd supply system,
plus distribution systems for obtaining process steam,
water, end electricity from the power plant) 67,000
Service facilities (buildings, shops, stores, site
development, roads, rallroada, and walkways) 638,000
Construction facilities 7l'*,OOQ
Subtotal direct investment I1*,939,OOO
Engineering design and supervision 1,31|9(000
Construction field expense 1,6^9,000
Contractor fees 71«9,000
Contingency l.^iOOO
Subtotal fixed Investment 20,235,000
Allowance for startup and roodlflcations 1,619,000
Interest during construction (8J/annun> rate) l,6l9iOOO
Total capital Investment 23,^73,000
Percent of subtotal
direct investment
2.9
3.8
21.lt
27.8
5.7
5-7
25.2
O."t
10O. 0
9.0
11.0
5.0
10.0
135.0
10.8
10.8
156.6
Basis:
Preliminary estimate from recent TVA In-house evaluation.
Stack gas reheat to 175°F by indirect steam reheat.
Msposal pond located 1 mile from power plant.
Midwest plant location represents project beginning rald-1972, ending mirt-1975- Average
cost basis for scaling, mid-1971*.
Minimum In process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
207
-------
Table 30. LIMESTONE - BENZOIC ACID SLURRY PROCESS
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
(500-MW new coal-fired power unit, 3-5$ 8 In fuel;
9O% ROP removal; onolte solids dlapunal)
Direct Coots
Delivered raw material
Llncntone
Benzole acid
Subtotal raw material
Annual quantity
l60. 1 M tons
lilO.O tone
Unit coat,
l).00/ton
600.00/ton
Total annual
Percent of
total annual
coot, t operating cost
61.0,1)00
?1»6,000
Convernlon costs
Operating labor and
supcrvioion
UtlllUen
Steam
Proceee water
Electricity
Maintenance
Labor and material, .08 x l'i,989,000
Analyses
Subtotal conversion costs
Subtotal direct conto
26,280 raan-hr
886,1.00
8.00/man-hr 230,200
2.82
1)92, 800 M Ib
260,000 N gal
70,770 M kWh
10
0.70/H Ib
0. 08/H gal
0.010/kWh
31)5,000
20,800
707,700
1,199,100
1*5,600
2,528,1)00
3,1)11), 800
0.28
9-51
16.12
0.61
")5-90
Indirect Coots
Average capitnl charges at l'i.9^
of total capital Investment
Overhead
Plant, 20$ of converolon coote
AdmJnlntratlve, 10% of operating labor
Subtotal IndJreet conto
Total annual operating costo
Equivalent unit operating cost
5,''97,500
505,700
21.000
I), 021., 200
7,1.39,000
1*7.02
100.00
Dollare/ton Cents/ml ULlon Dollars/ton
cooJ. burned Hllls/KWh Btu heat input sulfur removed
5-67
23.62
207-33
Bao1o:
Preliminary estimate from recent TVA In house evaluation.
Remaining life of power plant, JO yr.
Cool burned, 1,312,500 tono/yr, 9,000 Btu/kMh.
Stack gao reheat to 175°F.
Power unit onotream time, 7,000 hr/yr.
Hldweat plant location, 1975 operating costs.
Total capital Inveotment, $23,1)T3,000; subtotal direct Investment, $1'),909,000.
Working capital, $581,700.
Inveotment and operating cost for dlspooal of fly ash excluded.
208
-------
Table 31. LIME SLURRY PROCESS
VSNTURI - MOBILE BED SCRUBBING SCHEME
SUMMARY OF ESTIMATED FIXED INVESTMENT
(500-MW new_ coal-fired power unit, 3.5$ 3 in fuel;
905^ SOj removal; onalte solida disposal)
Percent of subtotal
Investment, $ direct Investment
Lime receiving and storage (bins, feeders, con-
veyors, and elevators)
Feed preparation (conveyors, sinkers, tnnks, agitators,
and pumps)
ParticulBte - sulfur dioxide scrubbers and Inlet ducts
(U scrubbers Including common feed plenums, effluent
hold tanks, agitators, and punipa)
Sulfur dioxide scrubbers and ducts (It scrubbers In-
cluding mist eliminators, effluent hold tanks,
agitators, pumps, and exhaust gas ducts to inlet
of fans)
Stack gas reheat (li Indirect steam reheaters)
Fans (li fans Including exhaust gas ducta and
dampers between fan and stack gas plenum)
Calcium solids disposal (onsite disposal facilities
Including slurry disposal pimps, pond, liner, and
pond vater return pumps)
Utilities (instrument air generation and supply
system, plus distribution systems for obtaining
process steam, vater, and electricity from the
power plant)
Service facilities (buildings, shops, stores, site
development, roads, railroads, and walkways)
Construction facilities
Subtotal direct Investment
Engineering design and supervision
Construction field expense '
Contractor fees
Contingency
Subtotal fixed investment
Allowance for startup and modifications
Interest during construction (8^/annum rate)
Total capital investment
795,000
387,000
3,203,000
It, 523,000
695,000
3,1.1.5,000
67,000
55?,ooo
710,000
1't, 919,000
1,3113,000
l,61il,000
7l"6,000
1.'t 92,000
20,11(1,000
i,6n,ooo
1,611,000
23,363,000
5.3
2.6
21-5
30.3
3.6
23-1
O.It
3-7
it. 8
100.0
9-0
11.0
5-0
10.0
135.0
10.8
10.8
156.6
8 Basis:
Preliminary estimate from recent TVA in-house evaluation.
Stack gas reheat to 175"F by Indirect steam reheat.
Disposal pond located 1 mile from power plant.
Midwest plant location represents project beginning mid-1972, ending raid-1975. Average
cost basis for scaling, mid-197't.
Minimum in process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overture pay Incentive not considered.
209
-------
Table 32. LIME SLURRY PROCESS
VENTURI - MOBILE BED SCRUBBING SCHEME g
TOTAL AVERAGE ANMJAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
(500-MW new -joal-flred pover unit, J.5? S in fuel;
90^ SOg reaiovnl; onolte eolido dleponal)
Direct Cooto
Delivered row material
Lime
Subtotal rav material
Annual quantity
81.2 M tone
Conversion cootB
Operating labor and
supervision
Utilities
Steam
Process water
Electricity
Maintenance
Lnbor and material, .08 x l'i,919,000
Analyses
Subtotal conversion coots
Subtotal direct coots
22,320 raan-hr
1490,000 M Ib
21(1,900 M gal
61,280,000 kWh
Unit coot,
22.00/ton
Total annual
cost, $
1.786,'i 00
I,7tt6,'4 00
8. 00/raan-hr 178, fOO
0.70/M Ib 3143,000
0.08/M gal 29,1(00
0.010/kWh 612,800
1,193,500
36,500
2,38;5,8oo
li, 170,200
Percent of
total annual
operating cost
2.19
51-19
Indirect Cootg
Average capital charges et lU
of total capital Investment
Overhead
3,l«8l.lOO
Plant, J?0?t of conversion cootfl
Administrative, 10% of operating labor
Subtotal Indirect costs
Total annual operating coete
Equivalent unit operating cost
Dollars/ton
coal burned Millo/kWb
6.21 2.33
U76,800
17,900
3, 975, «00
8,1146,000
Cents/million
Btu beat Input
25- B£>
5.85
0.22
THTBi
100.00
Dollnrs/ton
oulfur removed
227.0J
Basle:
Preliminary estimate from recent TVA in-house evaluation.
Remaining life of power plant, 30 yr.
Coal burned, 1,312,500 tons/yr, 9,000 Btu/lcWh.
Stack gas reheat to 175°F.
Power unit on-ctream time, 7,000 hr/yr.
Midwest plant location, 1975 operating coots.
Total capital Investment, J?3,j63,000; subtotal direct Investment, $114,919,000.
Working capital, $731,000.
Investment and operating cost for disposal of fly ash excluded.
210
-------
Table 33. LIME SLURRY PROCESS
MOBILE BED SCRUBBING SCHEME
SUMMARY OF ESTIMATED FIXED INVESTMENT6
(500-KW existing coal-fired power unit, 3.5% S In fuel;
90^ SOg removal; onslte solids disposal; pnrtlculates
removed by existing electrostatic preclpltntor)
Line receiving and storage (bins, feeders, conveyors,
and elevators)
Feed preparation (conveyors, slaXera, tanks,
agitators, and pumps)
Sulfur dioxide scrubbers and ducts (^ scrubbers
including common feed plenun, miot eliminators,
effluent hold tanks, ngltntors, pumps, nnd all duct-
vork between outlet of supplemental fan and stack gas
plenura)
Stack gas reheat C< direct oil-fired reheaters )
Fans (i fans including ducts and dampers between tie-
in to existing duct and inlet to supplemental fan)
Calciuji solids disposal (onslte disposal facilities
Including slurry disposal pumps, pond, liner, and
pond water return pumps)
Utilities (instrument air generation and supply
systen, plus distribution systems for obtaining
process steam, water, and electricity from the
power plant)
Service facilities (buildings, shops, stores, site
Investment, $
876,000
l<, 96^,000
305,000
1,118,000
3,152,000
335, OCX)
Percent of subtotal
direct investment
7.0
5-5
39-9
2.5
9-0
25.1.
development, roads, railroads, and walkways)
Construction facilities
Subtotal direct Investment
Engineering design and supervision
Construction field expense
Contractor fees
Contingency
Subtotal fixed Investment
Allowance for startup and modifications
Interest during construction (8^/annura rate)
Total capital Investment
6^9,000
592,000
12, I 27, 000
l,2l»3,000
1,616,000
870,000
1,367,000
17,523,000
1,1*02,000
1,1)02,000
20,327,000
5-2
C.B
10O.O
10.0
13.0
7.0
11.0
lljl.O
11-3
11. 3
163.6
8 Basis:
Preliminary estimate from recent TVA in-house evaluation.
Stack gas reheat to 175°F by direct oil-fired reheat.
Disposal plnd located 1 mile from power plant.
Kidweat plant location represents project beginning mid-1972, ending mid-1975. Average
cost basis for scaling, mid-1971).
Hinimura in process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
211
-------
Table jU. LIME SLURRY PROCESS
MOBILE BED SCRUBBING SCHEME
£
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
(500-m exiotlnK cool-fired i>crwer unit, 3.5$ 8 In fuel;
90S GOP renoval; onslte solids disposal;
partlculate removed by existing electrostatic preclpltator)
Total annual
Percent of
total annual
Annual quantity Unit cost, $ coct, I operating coat
Direct Cooto
Delivered raw material
Lime
Subtotal rav material
Conversion coots
Operating labor and
supervision
Utilities
Fuel oil (Ho. 6)
Proceoe water
Electricity
Maintenance
labor and material, .08 x 12,1*27,000
Analyceo
Subtotal conversion costs
Subtotal direct costs
83.0 H tons 22.00/ton
1.826,000
1,626,000
22,320 man-hr
8. 00/man-hr 178,600
2,763,900
,22.08
' 22.OB
2.16
1*, 236, 000 gal
21*7,300 M gal
56,056,000 kWh
,000
0. 23/gal
0.08/M gal
0.010/kWh
97l<,300
19,800
560,500
991',200
36,500
11.76
0. 2l*
6.78
12.02
55-50
Indirect Costs
Average capital charges at 15-3$
of total capital Investment
Overhead
Plant, 20$ of conversion costs
Administrative, 10$ of operating labor
Subtotal Indirect costs
Total annual operating costs
Equivalent unit operating cost
3,110,000
552,600
17.900
3,680,700
8,270,600
37.60
6.68
._ p. 22
100.00
Dollars/ton Cents/million Dollars/ton
coal burned Mlllo/XWh Dtu heat Input sulfur removed
6.16
2.36
25-69
225-
Basic:
Preliminary estimate from recent TVA In-houae evaluation.
Remaining life of pover plant, 25 yr.
Coal burned, 1,31*1,700 tons/yr, 9,200 Btu/kWh.
Stack gas reheat to 175°F.
Power unit onstream time, 7,000 hr/yr.
Hldviest plant location, 1975 operating costs.
Total capital Investment, ?20,327,000; subtotal direct Investment, $12,^27,000.
Working capital, ?8OO,OOO.
Investment and operating cost for removal and disposal of fly ash excluded.
212
-------
Table 35. LIME SLURRY PROCESS WITH ONSITE LIME PRODUCTION
VENTURI - VENTURI SCRUBBING SCHEME
SUMMARY OF ESTIMATED FIXED INVESTMENT8
(500-MW now coal-fired power unit, 3.5^ S in fuel;
90% SOy removal; onaite solids disposal)
Investment, $
Limestone receiving and storage (hoppers, feeders,
conveyors, elevatore, and bins) 751,000
Limestone calcination (rotary kiln, fan, conveyors,
elevators, dust collector, and coal storage) 3,OUl,OOO
Feed preparation (conveyors, slakera, tanks, agitators,
and pumps) 387,000
Pertlculate - sulfur dioxide scrubbers and Inlet ducts
(I* scrubbers including common feed plenum and pumps) '•,017,000
Sulfur dioxide scrubbers and ducts (!j scrubbers in-
'•luding mist eliminators, puraps, and exhaust gas
ducts to inlet of fans) 3,153,000
Stack gas reheat C< indirect steam reheaters) 5^2,000
Fans (14 fans including exhaust gas ducts and dampers
between fen and stack gas plenum) 767,000
Calcium solids disposal (onsite disposal facilities
Including slurry disposal pomps, pond, liner, and
pond water return puraps) 3,556,000
Utilities (instrument air generation and supply
system, plus distribution systems for obtaining
process steam, water, and electricity from the
power plant) 7'«,OOO
Service facilities (buildings, shops, stores, site
development, roads, railroads, and walkways) 607,000
Construction facilities 8}5,OOP
Subtotal direct Investment 17,530,000
Engineering design and supervision 1,578,000
Construction field expense 1,9?8,OOO
Contractor fees 877,000
Contingency 1.753,000
Subtotal fixed investment 23,6(>6,OOO
Allowance for startup and modifications 1,893,000
Interest during construction (Sjt/nnnum rate) 1,893|000
Total capital Investment 27,!i$2,000
Percent of subtotal
direct investment
17.3
2. Z
22.9
18.0
3-1
19-1
0.1.
3-5
"4.8
100.0
9-0
n.o
5.o
10.0
135.0
10.8
10.8
156.6
Basis:
Preliminary estimate from recent TVA in-houae evaluation.
Stack pas reheat to 175°F by indirect steam reheat.
Disposal pond located 1 mile from power plant.
Midwest plant location represents project beginning mid-1972, ending mid-1975. Average
cost basis for scaling, mid-1971'.
Minimum in process storage; only pumps are spared.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
213
-------
Table $6. SODIUM SOLUTION - H^S04 PRODUCTION PROCESS
SUMMARY OF ESTIMATED FIXED IW VESTMENT*1
(500-MW _Q£w_ coal-fired power unit, 3.5% 3 In fuel;
9($ SO-, removal; 13.3 tons/hr lOOJt
Percent of subtotal
Investment^ $ direct Investpent
Sods ash and antloxidant receiving, storage, end
preparation (pneumatic conveyor and blower,
feeders, mixing tank, agitator, and punps)
Partlculate scrubbers and Inlet ducts (I) scrubbers
Including common feed plenum, effluont hold tanks,
agitators, pumps, and fly ash neutralization
facilities)
Sulfur dioxide scrubbers and ducts (It scrubbers
Including rolst eliminators, pumps, and exhaust
gas ducts to inlet of fan)
Stack gas reheat (i Indirect stenm reheetera)
Tens (** fans including exhaust gas ducts and dampers
between fans and stock gas plenum)
Purge treatment (refrigeration system, chiller-
cry stalllier, feed coolers, centrifuge, rotary dryer,
steam/air heater, fan, dust collectors, feeders,
tanks, agltntors, pumps, conveyors, elevator, and
bins)
Sulfur dioxide regeneration (evaporator-crystnllizers,
heaters, condensers, strippers, desuperheater, tanks,
agitators, and pumps)
Sulfurlc acid plant (complete contact unit for
sulfurlc acid production)
Sulfurlc acid storage (storage and shipping
facilities for 30 days production of h^SO*)
Utilities (instrument air generation and supply
aysteo, and distribution systeus for obtaining
process steam, vater, and electricity from pover
plant)
Service facilities (bulldlngo, shops, stores, site
225, (XX)
3,81*6,000
"1,269,000
5J9,ooo
869,000
l,U73,ooo
2,622,000
2,793,000
21*2,000
195,000
1.2
20.6
22.9
2-9
7.9
1^.1
15.0
1-3
1.0
development, roods, railroads, and walkways)
Construction facilities
Subtotal direct investment
Engineering design and supervision
Construction field expense
Contractor fees
Contingency
Subtotal fixed Investment
Allowance for startup and modifications
Interest during construction (Sjt/annura rate)
Total capital investment
662,000
888,000
18,6*0,000
2,051,000
2,051,000
93?, 000
1.9&J.OOO
25, 5^1, 000
g.oio.ooo
30,133,000
3-5
l«. 8
100.0
11.0
11.0
5.0
10.0
137.0
15.7
11.0
161.7
8 Basin:
Preliminary estimate from recent TVA In-house evaluation.
Stack gas reheat to 175°F by indirect steam reheat.
Midwest plant location represents project beginning mid-1972, ending mid-1975- Average
cost basis for scaling, mid-1971*.
Minimum In process storage; only pumps are spared.
Fly ash slurry neutralized before disposal; closed loop water utilization for first stage.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay Incentive not considered.
214
-------
Table 3?. SODIUM SOLUTION - H2S04 PRODUCTION PROCESS
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS'
(500-KW new coal-fired pover unit, 3.5$ S In fuel;
90* S0? renoval; 92,890 tons/yr lOOSt H..SO.J
Annual quantity Uiilt coat, $
Total annual
coat, $
Direct Cools
Delivered rav materials
Lime (let stage neutralization) ljh.1 tons
Soda ar.h 9,300 tono
Antloxidant 317,100 Ib
Catalyst 1,800 liter
Subtotal ra» Eaterlolo
Conversion coats
Operating labor and
supervlsion
Utilities
Steam 2,137,600 M Ib
Process water 11,672,''00 M pal
Electricity 79,53'',000 kWh
Maintenance
Labor and material, .06 x !&,(>>>},OQQ
Armlynec
Subtotal conversion costs
Subtotal direct costs
26.00/ton
52.00/ton
2.00/lb
1.65/11ter
3,500
1483, foo
6.114,200
;, OOP
•45,8(0 inan-hr 8.00/inan-hr 366,900
0.70/M Ib 1,1)96,500
0.02/M gal L>33,''00
0.010/kU-b 795,300
1,118,600
108,000
i ,.U8,700
Percent of
total ftnrMJnl
op'.Tatlty, cost
0.03
li.lT
5^7
0.10
9-77
3.21
12.90
1.72
6.39
9-75
'48.75
Indirect Costs
Average capital charges at l!4.9^
of total capital Investment
Overhead
Plant, 203^ of conversion costs
Administrative and marketing
Subtotal Indirect costs
Total annual operating cost
(4,1490,600
823,700
''97.5,00
5,8.u,6oo
39-16
100. CO
Dollars/ton Dollnrn/ton Cento/mi ]J.ion Dollars/ton
-.qo,, coal burned Mllls/KVIh Btu heat Input sull'ur removed
Equivalent unit operating cost 119-01
C.1'2
5.16
35-09
306. 57
Baslo:
Preliminary eotitDate from recent TVA In-house evaluation.
Remaining life of pover plant, 30 yr.
Cool burned, 1,312,500 tono/yr, 9,000 Btu/kWh.
Stack gao reheat to 175'F.
Pover unit onstream time, 7,000 hr/yr.
Midwest plant location, 1975 operating coots.
Total capital Inveotment, $30,138,000; subtotal direct Investment, $l8,6li),000.
Working capital, $9)48,500.
Investment and operating coot for disposal of fly anh excluded.
215
-------
Table 38. AMMONIA LIMESTONE/LIME DOUBLE ALKALI PROCESS
SUMMARY OF ESTIMATED FIXED INVESTMENTa
(5OO-MVI now conl-fired power unit, 3.5?i 3 In fuel;
90>i SOS removal; onalte solids disposal)
Percent of subtotal
Investment. $ direct inveatment
Ammonia, limestone, and lime receiving and storage
(bins, feeders, conveyoro, elevators, and nmmonla
storngn sphere) 937,000 4.7
Feed prepnration (feeders, crushers, elevators, ball
mills, vaporizer, conveyors, slaker, tnnko, and pumps) 681,000 3.4
Participate scrubbers nnc! inlet ducts (^ scrubbers
Including common feed plenum, effluent hold tonka,
agltntoro, and pumps) 3,846,000 19.4
Sulfur dioxide scrubbers and ducts (U scrubbers
Including wire mesh mist eliminators, effluent
hold tanks, pumpe, and exhaust gas ducts to Inlet
of fan) 4,219,000 21-3
Stack gas reheat C< indirect steam reheaters) 539,000 2.7
Fans (b fans Including exhaust gas ducts and dampers
between fnns and stack gas plenum) 889,000 4.5
Regeneration and precipitates handling (thickeners,
filters, tanks, ogitatora, conveyors, and pumps) 3,581,000 18.0
Calcium solldo disposal (onsite disposal facilities
Including slurry disposal pumps and pond) 3,508,000 17.7
Utilities (Instrument air generation and supply
system, and distribution systems for obtaining
process steam, water, and electricity from the
power plant) 67,000 0.}
Service facilities (buildings, shops, stores, site
development, roada, railroads, and valkways) 633,000 3.2
Construction facilities 9^5,000 4.8
Subtotal direct Investment 19,§50,000 10O.O
Engineering design nnd supervision 1,787,000 9-0
Construction field expense 2,18^,000 11.0
Contractor fees 995,000 5.0
Contingency 1,935,000 10.0
Subtotal fixed investment 26,799,000 135-0
Allowance for startup and modifications 2,1^,000 10.8
Interest during construction (0%/annum rate) 2,l1tl),000 10.3
Total capital investment 31,037,000 156.6
a Bnclo:
Preliminary estimate from recent TVA In-house evaluation.
Stock gas reheat to 175"F by indirect steam reheat.
Midwest plant location represents project beginning mid-1972, ending mid-1975- Average
cost baols for scaling, mid-1974.
Minimum in process stornge; only pumps are spared.
Syntem is designed for closed loop water utilization.
Investment requirements for disposal of fly ash excluded.
Construction labor shortages with accompanying overtime pay incentive not considered.
216
-------
Table 59. AMMONIA LIMESTONE/LIKE DOUBLE ALKALI PROCESS'
TOTAL AVERAGE ANNUAL OPERATING COSTS - REGULATED UTILITY ECONOMICS
(5OO-MW new coid-fired power unit, J.5% S In fuel; 90% B0a removal)
Annual quantity Unit cost.
Direct Costs
Delivered raw materials
Limestone
Lime
Ammonia
Subtotal raw materials
1142.0 M tons
15O tons
2,110 tons
14.00/ton
22.00/ton
150.00/ton
Conversion costs
Operating labor and
supervision
Utilities
Steam
Process water
F-lectrlclty
Maintenance
Labor and material, .08 x 19,850,000
Analyses
Subtotal conversion coats
Subtotal direct costs
Indirect Costs
Averflge capital charges at lb.
of total capital Investment
Overhead
35,100 rr.nn-hr 8.00/can-hr
Total annunl
cost, ?
566,000
336,600
271;,300
1,173,"00
230,800
Percent uf
total annual
operating cost
5-87
3.1.8
2.8U
12.19
2.90
833,1400 H lb
26?, 000 M gal
63,200,000 KWh
^50,000
0.70/K lb
0.03/H gal
0.010/kWh
6l8,l400
21,000
632,000
1,583,000
51,700
3,191,900
6.140
0.22
16.142
35-01
145.20
ii,632,000
Plant, 2O% of conversion costs
Administrative, 10^ of operating labor
Subtotal Indirect costs
Total annual operating costs
Equivalent unit operating cost
633,1400
23,100
5,293,500
9,669,300
Dollars/ton Cents/million
coal burned Mills/kWh Dtu hent Input
T-37 2.76 30.70
6.60
0.29
100.00
Dollars/ton
sulfur removed
p/", Q t It O
8 Basis:
Prelloinary estimate from recent TVA in-house evaluation.
Remaining life of power plant, 30 yr.
Coal burned, 1,312,500 tons/yr, 9,000 Btu/KWh.
Stack gas reheat to 175°F.
Power unit on-strearo time, 7,000 hr/yr.
Mldweet plant location, 1975 operating costs.
Total capital investment, $31,087,OOO; subtotal direct Invest .,-
Working capital, t71'6,100.
Investment and operating coat for disposal of fly ash excluded.
i-ij, 850,000.
217
-------
Table kO. SODIUM LIMESTONE DOUBLE ALKALI PROCESS
SUMMARY OF ESTIMATED FIXED INVESTMENT8
(500-MW new coal-fired power unit, }.5$ S In fuel;
9O% SO2 removal; onaite solids disposal)
Investment. $
Limestone, soda nsh, and antioxidant receiving end
storage (bins, hoppers, feeders, conveyors,
elevators, and pumps) 319,000
Feed prepnrntion (feeders, weigh belts, agitators,
crushers, ball mills, tanks, nnd pumps) 607,000
Partlculnte scrubbers and inlet ducts (^ acrubbcjre
Including common feed plenum, effluent hold tnnks,
agitators, nnd pumps) 3,8U6,000
Sulfur dioxide scrubbers and ducts (l> scrubbers
including wire mesh mist eliminators, effluent hold
tanks, pumps, and exhaust gns ducts to inlet of fan) l»,36^,000
Stack gns reheat (l< indirect steam rehenters) 539,000
Fans (li fans Including exhnuet pas ducts and dampers
between fans nnd stack gas plenum) 1,170,000
Regeneration nnd precipitates handling (sulfnte
reoctor absorber and thickener, filter, conveyor,
Bteom coll, tnnk, agitator, nnd pumps) 3,583,000
Calclura oolidn disposal (onaite disposal facilities
including slurry disposal pumps and ponci) 3)551,000
Utilities (instrument air generation nnd supply
system, nnd distribution systems for obtaining
process otenra, water, nnd electricity from power plant) 72,000
Service facilities (buildings, shops, stores, site
development, roads, rnllroods, and walkways) 656,000
Construction facilities 9'j5,OOP
Subtotal direct investment 19»6«2,000
Engineering design and supervision 1,768,000
Construction field expense 2,l6J,000
Contractor fees 982,000
Contingency l/jfli.OOO
Subtotal fixed Investment 26,517,000
Allowance for startup and modifications 2,121,000
Interest during construction (B'J/annum rate) 2,121,OOP
Total capltnl Investment 3°»759,000
Percent of subtotal
direct Investment
1.6
3-1
19-6
22.2
2.7
6.0
16.2
18.1
O.'l
3-3
lj.8
100.0
9.0
11.0
5-0
10.0
135.0
10.8
10.8
156.6
n Boels:
Preliminary estimate from recent TVA in-house evaluation.
Stack gns reheat to 175°F by indirect steam reheat.
Midwest plant location represents project beginning mid-1972, ending mid-1975. Average
cost bnsis for scaling, mid-1971".
Minimum in process storage; only pumps are spared.
System is designed for closed loop water utilization.
Investment requirements for disposal of fly nsh excluded.
Construction Inbor nhortnp.es with accompanying overtime pny Incentive not considered.
218
-------
Table kl.
SOD1UI-; LIMESTONE DOUBLE ALKALI PROCESS
TOTAL AVERAGE ANMJAL OPERATING COSTS. - REGULATED UTILITY ECONOMICS
(500-MW_ncw coal-fired povcr unit, 3.5^ 3 In fuel; 905& S02 removal)
Total annual
Percent of
totnl annual
Annual quan11 ty Unit coot, $ coot, $ operating coot
Direct CootB
Delivered raw materials
LImestone
Soda eon
Antioxldant
Subtotal rav materials
Conversion costs
Operating labor and
supervision
Utilities
Steam
Process water
Electricity
Maintenance
Labor and material, .08 x 19,Wi?,OOO
Analyses
Subtotal conversion cofitu
Subtotal direct coots
177.5 M tons
83(4.0 tons
SliO.O M Jb
li.OO/ton
V-00. ton
710,000
'=8oiooo
8.00/nuui-hr 260,800
'jOB.'iOO M Ib
260,000 M gal
82,8110,000 kWh
',OOO
0.70/M Ib
0.08/M gal
0.010/kWb
35S900
?o,too
828.1100
1,571, ''00
lifi.YOO
3,08(1,000
11,319,'tOO
3-T3
0.22
8.68
16.146
Ji?.«
"»5.25
Indirect Costs
Average capital chargec at 1^.9^
of total capital Investment
Overhead
Plant, 20$ of conversion costs
Administrative, 10/6 of operating labor
Subtotal indirect costc
Total annual operating cost
it, 583, loo
617,?00
26,100
148.01
Equivalent unit operating cost
100.00
DolJLars/ton Conts/irJLlJon Dollars/ton
coal burned Hlllr./kWti Btu heat Input Bulfqr removed
2-73
30.30
Basle:
Preliminary estimate from recent TVA In-houre evaluation.
Remaining life of power plant, 30 yr.
Coal burned, 1,312,500 tone/yr, 9,OOO Btu/kWh.
Stack gaa reheat to 175°f.
Power unit on-etrcam time, 7,OOO hr/yr.
Mldweot plant location, 1975 operating costs.
Total capital Investment, ?30,759,OOO; subtotal direct investment, $19/Ji2,OOO.
Working capital, $739,200.
Investment and operating cost for disposal of fly aoh excluded.
219
-------
3» The Cat-Ox process has the poorest investment economy of scale
(unit size) of all five systems; however, for sulfur content of
fuel, Cat-Ox has the best scale factor. A reheat (existing)
Cat-Ox system requiring full particulate removal facilities to
.005 gr/scf was found to "be only 2.5$> higher than a new integrated
unit.
b. Plant age is an important factor only in the limestone and lime
processes where pond size depends on remaining plant life; older
units can best use limestone or lime processes.
5. Removal of only Qo% of the S02 (which would meet Federal new
source emission standards for 3«5^ S) instead of 90^> decreases
investment by only 3 to 5%.
Operating Cost
1. For both 3«5$> S coal-fired and 2.5% S oil-fired power units under
the premises used, limestone has the lowest annual operating cost
and sodium the highest. Even after credits for recovery products
are applied, the lifetime operating costs are higher for the re-
generable systems than for the throwaway systems. Cat-Ox becomes
competitive with limestone on new oil-fired units, especially
for smaller units.
2. As sulfur content of fuel increases, the relative operating cost
for all five systems changes. Cat-Ox lifetime operating costs
go from highest at low sulfur levels to third on coal-fired units
and first on oil-fired units with high-sulfur fuel levels. These
are some of the most interesting results of the study. The heat
credit for Cat-Ox becomes significant at high sulfur levels.
3. The lime and sodium processes rank high in raw material cost and
magnesia and Cat-Ox are low. Limestone has the largest range in
cost of all the raw materials.
^. Sodium scrubbing has the highest total labor cost and Cat-Ox the
lowest; however, labor is one of the smallest components (1-3/» of
total) for all five processes.
5- Energy costs are significant for all systems; applications of
sodium scrubbing on existing units require the greatest amount
of energy (35$ °f total operating cost). If a double-effect
evaporator-crystallizer is utilized in the sodium system, overall
energy costs could be reduced as much as 18$. The magnesia
process is also energy intensive; the Cat-Ox system uses the
lowest amount (5/»)«
6. Although expense for antioxidant (sodium system) is high, it is
justified to keep sulfate formation down and save Na2C03 makeup.
220
-------
7. Maintenance is quite significant ranging from 7$ of total operating
cost for a Cat-Ox case to 17$ for a limestone case.
8. Capital charges are the largest individual component of operating
cost for all five processes. For new Cat-Ox systems capital costs
are 12% of total. For the other processes, capital charges are
in the range of 39 to 50$>. A change in depreciation rate or cost
of money will obviously affect Cat-Ox the most.
9. Only about k to 5$ of total operating cost is saved when 80$ S02
removal is provided instead of 90$.
10. A Cat-Ox process on an existing plant has a ^0$ greater operating
cost than on a new system. This is caused by the high energy
required for stack gas reheat from 310 to 890°F prior to conversion
of S02 to S03.
11. As would be expected, the scrubbing steps of each process are the
highest cost operations.
12. For the energy intensive processes (magnesia, sodium), oil-fired
systems are at a disadvantage because of the high cost of fuel oil
relative to coal ($1.53/MM Btu compared to $0.5**/MM Btu).
13- Sale of byproducts at the values assumed in the study would reduce
the base case lifetime operating cost 7-2$ for the magnesia, 7-15°
for the sodium, and **.9$ for the Cat-Ox process.
l*i. Because of product revenues, the relative ranking of the magnesia
scrubbing process on oil-fired units improves under lifetime
operating costs until it is the lowest cost system above 800-MW
size.
15- Labor cost escalation (7.5$ per year) over the project life would
add about 7-10$ to total project cost.
16. Regardless of which process is utilized, the increase in the cost
of power to consumers for the base case is projected to range from
2.86 to 3.80 mills/kWh. For all case variations, projected costs
could range from 1.57 to 7-90 mills/kWh.
Although there are differences in the depth and quality of the
EPA-sponsored evaluations and the TVA in-house estimates, some additional
conclusions can be derived from comparing all the base case results as
shown in Table ^2.
1. For limestone scrubbing, process savings from the addition of
benzoic acid to reduce limestone stoichiometry (and therefore raw
material and solids disposal costs) are relatively small.
221
-------
Table 1*2. COMPARISON OF RESULTS OF EPA-SPONSORED EVALUATIONS AMD TVA IN-HOUSE ESTIMATES
NEW 500-MW, 3.5# SULFUR COAL-FIRED UNITS, JO YEARS LIFE (EXCEPTIONS NOTED)
Results
NJ
to
Process
Limestone
Limestone slurry process
Limestone-benzole acid slurry process
Lime
Lime slurry process (venturi-venturi scheme)
Lime slurry process (venturi-mobile bed scheme)
Lime slurry process (venturi-venturi scheme}*
Lime slurry process (mobile bed scheme)8
Lime slurry process with onsite lime production
(venturi-venturi scheme)
Magnesia
Magnesia slurry - regeneration process
Sodium
Sodium solution - S0£ reduction process
Sodium solution - H2S04 production process
Cat-Ox
Catalytic oxidation process
Double alkali
Ammonia-limestone/lime double-alkali process
Sodium-limestone double-alkali process
EPA-sponsored evaluations TVA In-house estimates
Capital Average annual Capital Average annual
investment^ $ operating cost, $ investment,$ operatingcost, $
25,1^3,000
26,1*06,000
1*2,756,000
7,702,700
25,1*75,000
9,210,800
50,1*91,000 11,601,500
8,875,900
50,158,000
51,087,000
50,759,000
7,!»59,000
22,U22,000
_
26,027,000*
_
-
3,101,900
-
9,6l2,l*00a
-
_
25,565,000
20,327,000*
27,1*52,000
_
3, 1U6.000
_
8,270,600*
8,101,900
11,05^,800
9,669,500
9,5'*5,300
8 Existing 500-MW, 5.5$ sulfur coal-fired unit, 25 years remaining life.
-------
2. Costs for a two-stage venturi lime system and a venturi - mobile
bed lime system on a new coal-fired unit are reasonably close;
however, for an existing coal-fired power plant or oil-fired unit
not requiring additional particulate removal facilities, the
mobile bed scrubber option requires about 22% less capital and
less operating cost.
3. Production of sulfuric acid is slightly less expensive than
production of sulfur when sodium scrubbing - regeneration is
used. Depending on prevailing market value of byproduct acid
and sulfur, the expected revenues could determine which option
is best (2.8 times as much acid as sulfur).
k . Based on the "state of the art" at this time, the costs of the
two double-alkali options are about the same, but at least 20-25/c
more costly than lime or limestone scrubbing.
223
-------
EPA/RTP PILOT STUDIES RELATED TO UNSATURATED
OPERATION OF LIME AND LIMESTONE SCRUBBERS
bv
Robert H. Borgwardt
Control Systems Laboratory
U.S. Environmental Protection Agency
National Environmental Research Center
Research Triangle Park, North Carolina
ABSTRACT
An oddity of lime and limestone $©2 scrubbers is the observation
that they can run unsaturated with respect to dissolved CaSO/-2H90
during closed loop operation. This fact offers a unique opportunity,
if the phenomenon responsible can be understood and controlled, to
effectively reduce scaling potential which has persisted as one of the
principal reliability problems of those systems. In addition to eliminating
the potential for scaling of absorber internals, unsaturated operation could
greatly increase the amount of clarified liquor available for mist eliminator
washing. Experimental studies of the process conditions necessary
to achieve unsaturated scrubbing liquor are reported, and a mechanism
is proposed that accounts for it. It is shown that the sulfate formed
by oxidation is purged, without crystallization, as a "solid solution"
within the precipitated CaSO,,. The amount of sulfate that can be thus
incorporated is related to tne sulfate activity in the scrubbing liquor
and the S02 precipitation rate in the hold tank. The effect of Cl~ on
suppressing sulfate activity and the effect of Kg"*"1" on raising sulfate
activity are evaluated and correlated with oxidation and saturation.
The observed superiority of lime vs. limestone with respect to ease of
development of unsaturated conditions is attributed to the higher
precipitation rates and low oxidation that arc character!stic of lime systems,
225
-------
EPA/RTP PILOT STUDIES RELATED TO UNSATURATED
OPERATION OF LIME AND LIMESTONE SCRUBBERS
SATURATED VS. UNSATURATED MODES
Calcium sulfate is generated in the scrubbing liquor when dissolved
S02 is oxidized either in the scrubbing tower, where the liquor is
exposed to flue gas containing about 5% oxygen, or in the ancillary
process vessels, where it is exposed to the air. Unlike calcium sulfite,
the sulfate is not precipitated from solution by the rise in pH that
occurs in the hold tank and, consequently, accumulates in the scrubbing
liquor until it is supersaturated with CaSO^-2H20, Calcium sulfate
crystallizes at a slow rate and its accumulation in the liquor of a
closed-loop scrubbing system ceases only when a level of supersaturation
is reached that yields a crystallization rate equal to the total oxi-
dation rate in all parts of the system - a necessary condition for steady
state operation. Present methods for reducing the steady state super-
saturation thus follow the crystallization approach by 1) maximizing the
rate of crystallization with high concentrations of seed crystals, and
2) providing maximum residence time for the slurry in the hold tank
(crystallizer) before returning it to the scrubber. Together, these
factors can reduce the level of supersaturation of the scrubber feed
liquor sufficiently to prevent scale growth. These principles were
successfully demonstrated by I.C.I, at 1000 ppm S02 scrubber feed
concentration, and present practice is based largely on their experience.
The possibility of achieving saturation levels close to, or even
below the equilibrium solubility of CaSO/'ZtUO in a lime scrubber liquor
was first indicated by J. Martin of Combustion Engineering, Inc. in
discussing the Louisville G&E operations at the 3rd EPA Symposium in
May 1973. Since subsaturated liquor is contradictory to the accepted
crystallization mechanism for sulfate purge, the reported unsaturated
operation was initially regarded as evidence of either open loop
conditions or total lack of oxidation. It is interesting to note that
the I.C.I, scrubber data reported in 1935, in which tests of super-
saturation were conducted on a regular basis, showed several periods
of negative supersaturation. No comment was made on the anomaly although
it sometimes persisted for several days. Inspection of the I.C.I.
reports reveals that these values were generally associated with
periods of low oxidation; 12-20%. It is a'lso significant that a
relatively short hold-tank residence time was used in that system.
In addition to the I.C.I, and L.G.5 E. scrubbers, Mitsui appears to
be another full-size unit that has achieved unsaturoted conditions. The
absence of chloride in the scrubbing liquor of each of these systems is
ail important factor they all have in common that, as will be shown, has
no doubt contributed to their success. The use of lime feed in the latter
two cases is another factor.
226
-------
The capability of operating a limestone scrubber at steady-state
unsaturated conditions, in which loop closure could be proven by
material balance, was demonstrated in the EPA/RTP model in May 1973.
These tests involved a plug flow hold tank in which no air contacted
the circulating slurry, and oxidation .values were only about 10%.
Since no chloride was present, unsaturated liquor was clearly
evidenced by calcium concentrations of only 100-200 ppm, less than
1/3 the saturated concentration. Material balances confirmed loop
closure at 62% solids and showed no unaccounted-for liquor loss.
Since that time, extensive experimentation has been carried out at
RTF with both limestone and lime scrubbers to define the process
conditions required to maintain operation in the unsaturated mode.
This report summarizes that work.
The most important potential advantage of the unsaturated mode of
Operation is that it would allow virtually unlimited washing of the mist
eliminators - the critical component in the overall reliability of most
systems at present - with liquor that would not contribute to chemical
scaling of the mist eliminator surfaces.
MECHANISM OF SULFATE PURGE IN THE UNSATURATED MODE
Proof of unsaturated operation in a closed loop scrubber requires
a mechanism of sulfate precipitation from the liquid phase that does
not involve crystallization. Given the fact that the liquor is unsat-
urated, it follows that crystallization of calcium sulfate will not
occur and cannot account for the presence of sulfate in the solid. It
is nevertheless observed that significant amounts of sulfate are present
in the solids when a scrubber is running unsaturated. Furthermore, the
incorporation of sulfate within the solid is a necessary condition for
steady-state unsaturated operation since the sulfate generated by
oxidation cannot be purged in sufficient amounts any other way. For
example, it can be shown that the maximum amount of dissolved sulfate
lost in the liquid purge from a closed loop scrubber (with filter cake
moisture) is equivalent to only about 0.2% oxidation when the chloride
level is 5000 ppm.
A mechanism whereby sulfate can be incorporated by calcium sulfite
without crystallization was proposed by I-C.I. in 1951 to account for the
unusual properties of the solid product from their Bankside scrubber.
Examination of those solids, which chemical analysis showed to contain
about 10% calcium sulfate, revealed no gypsum or anhydrite detectable
by x-ray diffraction. It was postulated from this result and from
measurements of the degree of hydration of the calcium sulfate that
it was present as a homogeneous "solid solution" within the calcium
sulfite. Laboratory experiments conducted at simulated scrubber
conditions confirmed that sulfate was brought down with the precipitating
in amounts equivalent to a maximum ratio of 0.23 mole sulfate per mole
227
o
sulfite.
-------
Tests similar to the above were carried out on solids obtained
from the EPA/RTP scrubber during unsaturated operation. These solids,
which contained 10-15 mole percent sulfate by chemical analysis, showed
no calcium sulfate when examined by x-ray diffraction in several dif-
ferent laboratories. Stirring the solids with distilled water would
not extract the calcium sulfate, confirming that it was not present as
a separate phase. Dissolution of the solid with HC1, followed by water
extraction and recrystallization, yields pure gypsum in amounts expected
from the chemical analysis of the original solid. The characteristics
of the solid conform in all respects to those reported by I.C.I, for solid
solutions and it was concluded that the formation of this product
constitutes the mechanism by which sulfate is purged from scrubbers
during unsaturated operation. The I.C.I, terminology has been adopted
here when referring to this compound although recent evidence suggests
that it is not a true solid solution in the thermodynamic sense.
OXIDATION AND SATURATION
When a scrubber is running unsaturated and the sulfate is being
purged entirely in the form of solid solution, a material balance
shows that the rate of solid solution formation must equal the overall
rate of oxidation. If this condition were not satisfied, the excess
sulfate would accumulate in the closed loop system and the liquor
would, in time, become saturated regardless of solid solution formation:
Input = Output + Accumulation (or depletion) (1)
For steady-state unsaturated operation, accumulation = 0 and since
sulfate cannot leave the system by any means other than solid solution,
equation (1) reduces to:
oxidation
1 (2)
where ratio is moles sulfate precipitated as solid solution per mole
of precipitated sulfite.
The relationship expressed by equation (2) is a necessary condition
for unsaturated operation at any level of subsaturation. If conditions
are such that equation (2) cannot be satisfied, CaSO, accumulation will
occur and the system must become saturated. Inspection of the equation
shows that this will happen if oxidation ex.ceeds the value corresponding
to the maximum limit of the value of ratio.
Figure 1 is a plot of scrubber feed saturations observed in the
EPA pilot plant over the full range of steady state oxidation values.
These data represent chloride-free operation with limestone feed,
consequently the calcium concentration is a direct index of saturation,
with Ca = 630 ppm corresponding to saturated liquor. The data show a
continuous increase in saturation from about l/6x at 10% oxidation to
Ix at about 20% oxidation which, according to the discussion above, is
equivalent to the maximum mole ratio under the test conditions. The
maximum mole ratio at 20% oxidation is thus, according to equation (2),
in agreement with the value determined by I.C.I, from lab data.
228
-------
ecc
1 -
T-
4CC-
/ •
2 / ' SCL.C SC-LTICN
- ^ •/ . • SATo,AT,ON(,.U.
3 • / ' ^^
? ,
-------
100000-
ice
30
i i i i i i i ' 1 |
i:co
TOTAL DISSOLVED WAC,
1 1 1 1 1 I .1 1
coco
' p
•ooc
Figure 2. Solubility of CaSOll-2H20 In Solutions Containing
Chloride and Magnesium. Laboratory Measurements
at 25" C.
230
-------
MAXIMUM MOLE RATIO OF SULFATE TO SULFITE IN SOLID SOLUTION
When sulfite precipitates from a liquor containing dissolved
sulfate, it is reasonable to expect that the amount of sulfate incor-
porated as solid solution is dependent upon the concentration of sulfate
in the liquor. As shown by the solubility data of Figure 2, the
equilibrium concentration of sulfate in solution can vary over two
orders of magnitude, in lime/limestone scrubbers, depending upon the
steady state concentrations of Mg"^1" and Cl~. At a given chloride level,
the magnesium concentration in a closed loop scrubber increases with
the amount of soluble magnesium fed, accumulating primarily as MgSO/°
and to a lesser extent, as CaSO,° and S0, = .
Experiments using limestone feed, to which varying amounts of Mg
were added (as MgO) to build up the concentration of total dissolved
sulfates, confirmed that solid-solution mole ratios exceeding 0.23
could be achieved by operation at high Mg"*"*" levels. Similarly, it was
shown that either MgS04 or MgO addition to a lime scrubber, which was
initially operating saturated, would force the system into the unsaturated
mode. From these results it was concluded that the mole ratio of sulfate
to sulfite in the solid is a function of the sulfate concentration in the
liquor from which it precipitates.
The presence of chloride in scrubber liquors is unavoidable in
closed-loop systems operating on coal-fired power plants that burn
coal containing chloride. This chloride enters the scrubber with the
flue gas as HC1, usually in concentrations less than 100 ppm. It
accumulates in the scrubber liquor because, unlike SC>2 and SO,, it
forms no insoluble calcium compounds and can therefore leave the system
only with the liquid purge, i.e. filter cake moisture or clarifier
underflow. Since the liquid purge from closed loop scrubbers is small,
high steady state levels of chloride must be reached in order for its
purge rate to equal the rate of chloride absorption in the tower. At
Shawnee-, Cl~ is the most abundant dissolved specie in the scrubbing
liquor.
Many runs have now been made in the unsaturated mode with both
lime and limestone scrubbers, at varying levels of chloride, magnesium
and oxidation. In Figure 3 the mole ratios' of sulfate/sulfite in the
solids are plotted against the sulfate ion activities calculated from
the average scrubber feed liquor analyses for those runs. It is apparent
from Figure 3 that the value of ratio increases in an approximately linear
manner with sulfate activity. This correlation is significant in two
respects: 1) an increase in ratio as any function of sulfate activity is
sufficient condition to qualitatively explain the formation of unsaturated
solutions (given this relationship, the existence of an unsaturated mode is
a predictable consequence); and 2) a quantitative relationship between these
231
-------
two variables will enable one to calculate the degree of subsaturation that
will result from any given levels of chloride, magnesium and oxidation.
This is accomplished through simultaneous solution of this relationship
with six other equations that define ionic strength, ion balance, sulfate
activity and saturation. Using a simple power function to represent
Figure 3, these relationships are:
Ion balance:
Ionic strength:
SO^, act. coeff.
activity:
Solid Soln.
formation:
Saturation:
Ca =
v = 0
In c
su4
In SO
Sat.
*UL96
rso
•5[24
-2
-- -t-
^- + •
,02
71
Ca
10
r
1+0.7 r
-2.227 -
64 24J
Me SO? . I
. _i_ — .lii — *— -+- -
6 16 3.
u
In
2.041
a
„= ac- ~2'3
2
a
c
fS04
L96
}
0
rCa
[40
SO^]
95+ln ratio
.906
! 1
J 3.32x10
-5
(3)
(4)
(5)
(6)
(7)
(8)
[93
For any given values of oxidation, Cl, Mg and S02 in the scrubbing
liquor, the values of the other 8 variables can be determined from
equations (2) - (9) : Ca, S04, u, C, a, S0^= activity, ratio, and
saturation.
Solution of these equations for a limestone scrubber is shown in
Figure 4, where the Mg levels required for 0.8x saturation of the
scrubber feed are plotted against oxidation at several levels of chloride.
PRECIPITATION RATE
In addition to the values of oxidation and sulfate activity
another constraint must also be satisfied for a system to operate in
the unsaturated mode. The data shown in Figures 1 and 3 were all.
obtained at residence times of about 6 minutes, inlet S02 concentrations
of 3000 ppra, and scrubber gas velocities of 9 ft/sec or more. It is
observed that, when all other variables remain constant, an increase in
hold tank residence time to 12 minutes will cause a limestone scrubber
to operate supersaturated even though the oxidation, magnesium, and
chloride levels are maintained at values that yield unsaturated liquor
at the shorter residence time. It is evident that solid solution formation
is not simply a function of the liquor composition, but kinetic factors
are also involved. Table 1 compares steady state data from duplicate
limestone/MgO scrubber runs at different hold tank residence times. The
scrubber ran supersaturated for 4 weeks at 12 rain, residence time, where-
232
-------
i.C-
UJ
< O.I
LIMESTONE •
LIVE A
.Cl
.CO'
I I I I I
.01
-cnviry , Z. ''OTVL so..;
.06
Figure 3. Molfi Ratio of Sulfate to Sulfltft In Pilot Plant
Solids, as a Function of the Sulfate Ion Activity
In the Scrubbing Liquor. SflturatIon<0.9x
233
-------
liOCC
ICOOC
900C
ecoc-
" 7COC
40CCr
200or-
iCCCr
10
3C
5. >
OXIDATlCS'
, i. <• r: f n I
Figure U . Macneslun Rnqu I ro-n^n t for.O.Kx Scrubber Feed
Saturation, as a Function of Oxidation. Values
Calculated by Equations 2-9.
234
-------
Table 1, EFFECT OF S0? PRFCI PI TAT I ON RATF ON CaSO. '2H-0 SATURATION
IN LIMF.STONE/MeO SCRUBBER
Inlet SO^ « 3000 pnn, Flue p;as 0- • U.6*;, S02 absorption • 95,
Hoi d tank
Hoi d tank
Scrubber
S02 feed
residence tine, mln.
vol. (slurry), liter
gas velocity, m./sec.
rate, Kr,./hr.
Precipitation rate, ng . S02/ml n . ( 1
Fl yash
Total sol
MgO feed,
Average F
Total oxl
Ids, ?,
r,./Kg. 1 Inestone
eed Linuor Conposl»'nn
SO^ E./ 1 1 tnr
sr..
Cn
Me
Cl
PH
da t 1 on, !j
Apr.vaj
22-26
6
1.73
2.8
2.6
i t e r ) 6U
no
7
67
19.2
3.0
0.2U
7.8
6.2
5.5
2G
Saturation(b) n.i*3x
Saturation c
0 .Ux .
Sept .
3-6
12
1090
2.1*
2.3
26
no
10
62
38.8
1.9
0.614
10.6
2.05
6.2
20
1.27x
1.3x
Auf, .
12-16
12
1090
2.1*
2.3
21*
yes
8
62
1*0. 1
6.8
0.69
12.0
3.2
6.0
28
l.lGx
1.2x
Sent.
23-27
5.3
i*8U
2.1*
2.3
61*
yes
15
73
35.2
6.0
0.27
ll.i*
2.9
5.6
12
0. U5x
0.5x
Sept.
16-20
5.3
US'*
2.U
2.3
6U
no
10
73
36.1*
7.4
0.2S
11.1*
3.5
6.0
11*
O.U3x
O.i*'x
(a) open hold tank for Apr. 22-2G run, all other runs
made with sealed tank.
(b) calculated at 50 C. with Rccht e 1 -nod i f I ed Radian
eou i 1 i b r i u-i cr,~in.iter pror-.ra-..
(c) by solid CaSn^ -2H20 solubility test at 25"c.
235
-------
as reduction of residence time to 5.3 min. resulted in saturation levels
of 0.4x. This level of subsaturation was maintained with and without
flyash addition. The choice of 5.3 rain, residence time was made to
provide the same precipitation rate that existed during the period of
testing represented by the data of Figures 1 and 3. The precipitation
rate is defined as:
n . . . Amount of SO? precipitated in hold tank
Precipitation rate = T1 , . : 7—^~ ^ j r— ; :—— r-r-
r Unit time (unit volume of slurry in hold tank)
where the amount of S02 precipitated in the hold tank is determined by
the scrubber gas throughput, S,C>2 inlet concentration, SC>2 scrubbing
efficiency and oxidation. The value of the precipitation rate for the
RTF scrubbers during unsaturated operation was 0.064 grams 50^/min.(liter)
LIME VS. LIMESTONE
The effect of precipitation rate is probably the most important
factor influencing the difference in CaSO^-Zh^O saturation levels
between lime and limestone scrubbers. The SC^ precipitation rate is
ultimately limited in the latter case by the comparatively slow
dissolution rate of limestone. Lime scrubbers designed to take
advantage of its rapid dissolution rate can operate in the unsaturated
mode with hold-tank residence times at least as short as 1 minute. The
high precipitation rates that ensue from these short residence times can
explain the following advantages of lime over limestone scrubbers that
have emerged from long-term concurrent TCA tests at the RTF facility:
1) Operating at 5000 ppm Cl~ and without MgO addition,
the limestone scrubber averaged 1.2 - 1.3x saturation
at 10 min. residence time, while the lime scrubber
averaged l.Ox saturation at 2 min. residence time.
2) At 5000 ppm Cl~, the amount of MgO addition required
to achieve a given level of subsaturation in the
lime scrubber was about 1/10 that required for the
limestone scrubber. The Mg"^" and dissolved sulfite
levels are correspondingly lower in the lime system -
an important consideration if subsaturated liquor is
to be used for washing mist eliminators where oxidation
will result in a rise in the local saturation level.
Another important advantage is the lower oxidation that is
characteristic of the lime system, which is only 4% in our TCA scrubber
at L/G 50 in the absence of Cl~. The lowest oxidation we have achieved
in the limestone scrubber is about 10%. With Cl present these minima
are about 12% in both cases. As would be expected, the chloride level
at which a lime scrubber can operate without becoming supersaturated is
much higher than that for a limestone scrubber, when no MgO is fed. In
summary, the unsaturated mode is easier to achieve and maintain in a
lime scrubber than in a limestone scrubber for any given set of operating
conditions.
236
-------
FACTORS AFFECTING OXIDATION
Low pH increases oxidation. Our experience supports this conclusion
for both lime and limestone systems. The pH is, in turn, affected by
chloride. In systems operating without chloride, where the scrubber
effluent pH's are similar (5.6-5,7), the lime scrubber yields lower
oxidation than the limestone scrubber. This follows from the higher pH
and correspondingly lower concentration of dissolved sulfite through-
out the lime system.
Addition of chloride lowered pH throughout the limestone scrubber.
In the lime scrubber, where feed pH is controlled at a set value, chloride
caused a greater drop in pH to occur across the tower resulting in a
scrubber effluent pH lower (4.8-5.0) than the limestone scrubber
(5.4-5.5). This drop in pH no doubt results from the common ion effect
of Ca++ on the solubility of CaSO-,, the dissolution of which is
responsible for the SO^ scrubbing action in lirae systems operating in
the tail-end-addition mode. As a result, lime scrubbers operating with chloride
must use higher L/G's to maintain a given scrubbing efficiency and exhibit
higher oxidation than lime scrubbers operating without chloride. For
example, a scrubber with 4% oxidation at L/G 50 without chloride showed
12% oxidation at L/G 77 with 5000 ppm Cl.
Oxidation in the hold tank has been shown to contribute significantly
to overall oxidation in the RTF scrubbers. Consequently, the hold tanks
were sealed for all experiments related to unsaturated operation. When
no MgO is fed to the scrubber, oxidation occurring in the filter/clarifier
loop is of no consequence, because the rate of SO-> circulation through
that loop in the liquid phase is insignificant compared to the total S09
absorption (and purge) rate. Thus, even if totally oxidized, the S02
circulating through the filter would not affect the overall oxidation
values. When MgO is added to either a lime or limestone system, the
equilibrium concentration of dissolved sulfite is increased (primarily
as MgSO^0) and oxidation in the filter/clarifier can no longer be
ignored. For example, operating at 10 g. Mg/liter and S'u solids, the total
dissolved S02 was 1.8 g./liter in the hold tank. After passing through the
filter loop this SOo was over 90% oxidized before reentering the hold
tank. This amounted to 11% of the total S02 absorption rate in the
scrubber and 30% of the total oxidation in the system. Raising the
percent solids proportionately lowers the rate of circulation of slurry
through the filter/clarifier loop and operation at 15% solids is
recommended to limit this source of oxidation. This can be further
reduced by filtering only settled sludge.
Oxygen in the flue gas averaged about 4.5% in this work, representing
a coal-fired power plant at normal excess air. Increasing the oxygen content
of the flue gas raised the oxidation raLe in the scrubber towor and,
although the limestone scrubber lias been operated unsaturated with up
to 6% oxygen, it appears that unacceptable levels of oxidation may be
encountered at higher levels, especially in liwestone/MgO systems.
Flyasli had no demonstrable effect on oxidation in either the lime
or limestone scrubber.
237
-------
L.G.S E. SIMULATION
In accordance with the rationale outlined above, the principal
features of the Louisville Gas £ Electric carbide-lime scrubber were
simulated at RTF with the 3-stage TCA af 11.5 ft./sec. gas velocity,
3000 ppm inlet S02 and no Cl". Using ordinary pebble lime feed, which
was slaked in filtrate liquor, the system ran unsaturated at 0.3x with
only 100-200 ppm Mg in the scrubbing liquor (no MgO was added). As
long as chloride was not present in the scrubber, this saturation level
could be maintained indefinitely. It was concluded that carbide lime
was not necessary to achieve the low oxidation characteristic of this
mode, which in our case was only 4% at L/G 50. A sealed hold tank with
2~min. residence time was used in which the pH was controlled at 7.5.
Additional tests with this tank in series with a larger, 20-min. open
hold tank showed that complete precipitation of the sulfite (to 60 ppm)
was achieved in the smaller tank; the large tank served no purpose as
far as liquor saturation was concerned.
Addition of chloride to this system sharply dropped the scrubber
effluent -pH and the S0? scrubbing efficiency could be maintained only by
raising L/G to 77. The scrubber efriuent pH was not influenced by
increasing the feed pH at this make-per-pass (560 mg. SC>2/liter). The
oxidation increased and the system became saturated. Unsaturated operation
at a 5000 ppm level of chloride was again achieved by adding
MgO to the lime feed to provide 1000-2000 ppm Mg*+ in the scrubbing liquor.
SUPPLEMENTARY EFFECTS OF MgO ADDITION
Improvement in S09 scrubbing efficiency as a result of the
accumulation of MgS040"and MgS03° in the scrubbing liquor, and/or the
dissolution of MgO in the tower, is a generally recognized fact that
is confirmed by our experience with both lime and limestone scrubbers.
For example, operating the limestone scrubber as a multigrid tower
(4 grids, 65% open area) gave 95% S02 removal with only 3.5 in. water AP
at L/G 85, and 90% removal at L/G 65. The option of eliminating
another reliability problem - TCA sphere deterioration - is thus
indicatc-d. These levels of efficiency required MgO feed rates of 1.7
lb/100 Ib limestone. TCA scrubbers operating at normal pressure drops
can achieve 98% removal under these conditions.
The high equilibrium concentration of -sulfite in liquor containing
large amounts of MgSO^0, makes possible both rapid liquid-phase oxidation
and rapid dissolution of solid CaSOo. Using a pressure oxidizer designed
by M.W. Kellogg Co., we were able to increase oxidation from 28% in the
scrubber slurry to 99% in the product sludge stream. As a result of this
oxidation, the settling rate was increased from 0.07 to 0.8 cm/rain and
the final settled volume reduced from 340 to 190 ml per liter of original slurry.
The final settled density was 54% and the volume of dry solid product
was reduced by 40%. It is our expectation that an unsaturated limestone/MgO
scrubber can be operated with such an oxidizer to achieve fully oxidized
product without affecting subsaturation in the scrubber loop.
238
-------
Magnesium must be added as MgO either as calcined dolomite or
half-calcined dolomite (to limestone scrubbers). It has been shown that
the MgCOo normally present in limestones is not soluble at the pH levels
at which these scrubbers operate. This magnesium enters and leaves the
scrubber as the mineral dolomite,'MgCOyCaCO^, without participating in
the scrubbing reactions. In our case, feeding limestone containing 5%
MgCO-} yields about 500 ppm Mg in the scrubber liquor - about 10% of the
level that would be obtained by complete dissolution of the dolomitic
component. The maximum CaCO-^ utilization obtainable from limestones of
high Mg content is necessarily reduced because an equal amount of CaCO^
is tied up in this insoluble mineral form.
CONCLUSIONS
• Uiisaturated operation is technically feasible in either lime or
limestone scrubbers. This feasibility has apparently been
demonstrated on full scale systems at Louisville G.5 E. and Mitsui.
• For any given set of operating conditions, subsaturated liquor
is easier to achieve in lime than in limestone scrubbers.
« The presence of chloride makes it more difficult to operate in
the unsaturated mode.
• Addition of magnesium makes it easier to operate unsaturated.
The magnesium concentration can be increased by adding MgO.
• High SC>2 scrubbing efficiencies can be achieved at low pressure
drop when magnesium concentrations of 10 g./liter or more are
maintained in the scrubbing liquor.
> Minimizing oxidation assists in maintaining unsaturated
conditions.
* Short hold tank residence times (1-2 min.) can be used in lime
scrubbers operating in the unsaturated mode.
ACKNOWLEDGEMENTS
The RTF pilot plant from which these data were obtained is operated
for EPA by Monsanto Research Corp. The contributions of this excellent
group, and of Mr. Robert Opferkuch in particular, are gratefully acknowl-
eged. Helpful advice and comments on this paper were received from
Dr. Michael Epstein Of Bechtel Corp. and from Prank Princiotta and
John Williams of EPA. The computer program used for liquor saturation
calculations was kindly provided by Charles Leivo of Bcchtel.
239
-------
NOMENCLATURE
L/G
x
o
ratio
CaS04°
Ca
MgS04
Mg
liquid to gas ratio in tower, gal./lOOO cfm @ 125°F
times saturation
activity coefficient of Ca and S04~ in solution
apparent ionic strength, assuming complete ionization
of dissolved CaS04 and MgS04
empirically derived factor for expressing activities
in terms of total dissolved concentrations
moles sulfate per mole sulfite in scrubber solids
oxidation total moles S02 oxidized per mole S02 absorbed
ion complex formed by Ca++ and S04 = in solution
total dissolved calcium in scrubber liquor, g./liter
ion complex formed by Mg"1"1" and S04~ in solution
sum of the concentrations of Mg
MgSOj0 excluded. g. /liter
+
and
240
-------
LIMESTONE AND LIME TEST RESULTS AT THE
EPA ALKALI SCRUBBING TEST FACILITY
AT THE TVA SHAWNEE POWER PLANT
by
M. Epstein, L. Sybert, S. C. Wang, C. C. Leivo and R. G. Rhudy
BECHTEL CORPORATION
50 Bealc Street
San Francisco, California 94119
ABSTRACT
The limestone and lime reliability testing at the Shawncc facility has shown that scrubber
internals can be kept relatively free of scale if the sulfatc (gypsum) saturation of the scrub-
ber slurry is kept below about 1 35 percent. This can be accomplished with the proper selec-
tion of percent solids recirculateci, effluent residence time and liquid to gas ratio. The only
significant reliability problem at the Shawnce facility has been associated with the scaling
and/or plugging of the mist eliminator systems. The TCA utilizes a Koch Flexitray with un-
derside steam sparging, in scries with a chevron mist eliminator which has provision for un-
derside washing. In a test run which is currently in progress, this mist elimination system has
remained essentially clean over a 790 hour period, at a superficial TCA gas velocity of 8.6
ft/sec and 14 percent solids recirculateci. The mist eliminator undcrwash utili/es diluted
clarified liquor (at a ratio of three parts raw water makeup to two parts clarified liquor) at
0.3 gpm/ft^ and the Koch tray is irrigated with the remaining clarified liquor plus mist elim-
inator wash at 0.5 gpm/ft . The spray tower utilizes a chevron mist eliminator with provi-
sion for underside and topside washing. Opcrability of this mist elimination system during
tests of up to three months has been demonstrated, at a superficial gas velocity of 6.7 ft/sec
and 8 percent solids recirculated, with intermittent or continuous bottomsidc washing. A
recent test with intermittent topside and bottomside washing has indicated that opcrability
of this mist elimination system can be extended substantially beyond three months.
Presented at the
SYMPOSIUM ON FLUE GAS DESULFU RIZATION
Atlanta, Georgia
November 4-7, 1974
241
-------
TABLE OF CONTENTS
Section Page
1 INTRODUCTION 243
2 LIMESTONE/ LIME RELIABILITY 252
TESTING CONCEPTS
2. I Scrubber Internals Scaling Potential 252
2.2 Mist Eliminator Scaling Potential 255
2. 3 Closed Liquor Loop Operation 256
2. 4 Determination of Sulfate (Gypsum) 257
Saturation
3 LIMESTONE RELIABILITY TEST RESULTS 259
3. 1 Performance Data and Test Evaluation 259
3. 2 Conclusions 278
4 LIME RELIABILITY TEST RESULTS 281
4. 1 Performance Data and Test Evaluation 281
4. 2 Conclusions 298
5 PARTICULATE REMOVAL TEST RESULTS 301
6 REFERENCES 305
242
-------
Section 1
INTRODUCTION
In June 1968, the Environmental Protection Agency (EPA), through
its Office of Research and Development (OR&D) and Control Systems
Laboratory, initiated a program to test a prototype lime and limestone
wet-scrubbing system for removing sulfur dioxide and particulates
from flue gases. The system is integrated into the flue gas ductwork
of a coal-fired boiler at the Tennessee Valley Authority (TVA) Shawnee
Power Station, Paducah, Kentucky. Bechtel Corporation of San Francisco
is the major contractor and test director, and TVA is the constructor
and facility operator.
The major goals of the test program are: (1) to characterize as com-
pletely as possible the effect of important process variables on sulfur
dioxide and particulate removal; (2) to develop mathematical models (o
allow economic scale-up of attractive operating configurations to full-
size scrubber facilities; and, (3) to perform long-term reliability testing.
The test facility consists of three parallel scrubber systems: (1) a ven-
turi followed by a spray tower; (2) a Turbulent Contact Absorber (TCA):
and, (3} a Marble-Bed Absorber. Each system is capable of treating
approximately 10 Mw equivalent (30,000 acfm @ 300 F) of flue gas con-
taining 1800-4000 ppm sulfur dioxide and 2 to 5 grains/scf of particulates.
The test facility has been described in detail in References 1, 2 and 3.
243
-------
The following sequential test blocks were defined for the program:
(1) Air/water testing
(2) Sodium carhonate testing
(3) Limestone wet-scrubbing testing
(4) Lime wet-sc rubbing testing
The limestone and lime wet-sc rubbing test blocks have been divided
into three genera] categories: (1) short-term (less than 1 day) factorial
tests, (2) longer term (over 2 weeks) reliability verification tests, and
(3) long-term (4 to 6 months) reliability tests. The object of the fac-
torial tests is to determine the effect of independent variables (e.g.,
gas rale) on SO-, removal for the scrubber systems. The primary
objective of the reliability verification tests is to define regions for
reliable (e.g., scale-free} operation of the scrubber systems. The
object of the reliability tests is to determine the long-term operating
reliability for the scrubber systems and to develop definitive process
economics data and scale-up factors. The test program has been
described in detail in References 1, 2 and 3.
This report presents the results of the limestone and lime reliability
tests at the Shawnee facility from October 1973 through October 1974.
Results from the air/water, sodium carbonate, limestone factorial and
limestone reliability verification testing have been presented in Ref-
erences 1, 2 and 3.
244
-------
The lime reliability tests have been conducted on the venturi/spray tower
system and the limestone reliability tests on the TCA system. ' Figures
1 and Z (drawn roughly to scale) show the venturi/spray tower and TCA
scrubber systems along with the mist eliminator systems selected for
de-entraining slurry droplets in the gas stream. The spray tower
utilizes a chevron mist eliminator with provision for underside and top-
side washing. The TCA employs a Koch Flexitray with underside steam
sparging, in series with a chevron mist eliminator which has provision
for underside washing. Sections of the chevron mist eliminators used
in the two scrubber systems are depicted, to scale, in Figure 3. The
system configurations used for the venturi/spray lower lime reliability
tests and the TCA limestone reliability tests are shown in Figures 4
and 5, respectively.
In June 1974, the EPA initiated a two-year Advanced Test Program at
the Shawnee facility. The advanced program will involve, primarily,
limestone testing on the TCA system and lime testing on the venturi/
spray tower system. The major goals of the advanced program are:
To continue long-term testing with emphasis on the demon-
stration of reliable mist elimination operation.
Testing on the Marble-Bed system has been discontinued since July 1973
due to problems with erosion of the bed spray nozzles and subsequent
pluggage of the marble bed.
Underside mist eliminator washing was provided in July 1974 (sor
Section 3. 1. 8).
245
-------
GAS OUT
CHEVRON MIST
ELIMINATOR
SPRAY TOWER
INLET SLURRY
THROAT
ADJUSTABLE PLUG
VENTURI SCRUBBER
WASH LIQUOR
WASH LIQUOR
INLET SLURRY
I 1
APPROX. SCALE
EFFLUENT SLURRY
Figure 1. Schematic of Venturi Scrubber and Spray Tower
246
-------
GAS OUT
MIST ELIMINATOR
WASH LIQUOR
CHEVRON MIST
ELIMINATOR
KOCH TRAY
INLET KOCH TRAY
WASH LIQUOR
EFFLUENT KOCH
TRAY WASH LIQUOR
STEAM SPARGE
RETAINING GRIDS
GAS IN
\
00 o°0
00 O
_° -Q.Q-
0 o o °o
Q_Q_o_
o ._ o°
O O Cf
00 o
O O
& o_
INLET SLURRY
MOBILE PACKING SPHERES
51
I 1
APPROX. SCALE
EFFLUENT SLURRY
Figure 2. Schematic of Three-Bed TCA Scrubber
247
-------
SPRAY TOWER
TCA
t
GAS FLOW
GAS FLOW
6 in.
Figure 3. Mist Eliminator Sections
248
-------
SEIRINC POND
O Gas Composition
® Particulale Composition & Loadinq
© Slurry or Solids Comoosidon
_ -_ Gas Stream
—— Liquor Stream
Figure 4. Typical Process Flow Diagram for Vo>nt:uri 'Spray Tower System
-------
to
l_n
O
1.0. FAN
UBBER
1 Tank
--®-fr
»>}}•>•>
w \j w
-T.-s-r'-rx.
A A A
_ r. 1
AV.W.V
i ' • ^
1 •* * ±*1u* J 1
—
])
^
1
J
X. WATtK
Bleed
O Gas Composition
S Particulate Composition & Loading
Slurry or Solids Composition
_ _ Gas Stream
_ Liquor Stream
RESIURRY
TANK
I
SIACK
SniLING POND
Figure 5. Typical Process Flow Diagram for TCA System
-------
• To investigate advanced process and equipment design varia-
tions (p.p., operation with process slurry unsaturaled with
respect to gypsum) for the improvement of system reliability
and process economics.
e To perform long-term (Z to 5 months) reliability testing on
advanced process and equipment design variations.
• To evaluate process variations for substantial increase in
limestone utilization with corresponding minimization of
sludge production.
• To determine the practical upper limits of sulfur dioxide
removal efficiency.
• To determine the effectiveness of existing technology for
producing an improved throwaway sludge product.
• To evaluate system performance and reliability without fly
ash in the flue gas.
• To evaluate the effectiveness of three commercially offered
sludge fixation processes and untreated sludge disposal.
• To develop a computer program, in conjunction with T VA, for
the design and cost comparison of full-scale limestone and
lime systems.
The Advanced Test Program is described in detail in Reference 4.
This manual, although not prepared for widespread distribution, is
available upon request from Bechtel Corporation.
251
-------
Section 2
LIMESTONE/LIME RELIABILITY TESTING CCXCEPTS
As mentioned previously, the major objective of the limestone/lime
reliability tests at the Sha\vnee facility is to determine long-term
(4 to (> months) operating reliability for the scrubber systems. Major
reliability problems for limest one/lime wet - s c rubbing systems have
been associated with scaling and plugging of scrubber internals and
mist eliminator sxirfaces (a majority of mechanical problems can be
solved by improved design).
2. 1 SCRUBBER INTERNALS SCALING POTENTIAL
The Shawnee limestone reliability verification tests have shown that
the scrubber internals can be kept relatively free of calcium sulfate
(gypsum) scale at high percent solids recirculated and/or at high
effluent residence limes (see References 1 and 2). The results have
also shown that scrubber scaling potential can be correlated with the
calculated degree of sulfate saturation of the scrubber slurry, i. e. ,
sulfate scaling is likely to occxir for a degree of sulfate saturation
In this report "scale" refers only to crystalling hard solids, and
"solids" or "soft-solids " refer to mud-like slurry solids. "Plug-
ging" refers to the accumulation of mud-like slurry solids. Also,
scale refers to sulfate base scale, unless otherwise noted.
252
-------
of the scrubber inlet liquor greater than 130 to 140 percent. ' In Fig-
ure 6, the calculated sulfate saturations of the scrubber inlet liquor
for the TCA limestone reliability verification tests are shown as a
function of percent solids recirculated and effluent residence time.
Figure 6 would indicate that, in order to remain below 130 percent
saturated in a limestone system at 1 5 weight percent solids recirculated
and at a liquid-to-gas ratio of from 64 to 80 gal/mcf, the effluent resi-
dence time would have to be equal to or greater than 10 minutes.
Results from the lime reliability tests at the Shawnee facility (to be
discussed in Section 4) have shown that the calculated sulfate saturations
of the scrubber inlet liquor can be similarly correlated with percent
solids recircxilated and effluent residence time. In addition, the lime
testing has shown tha' the sulfate saturations are strong functions of
flue gas inlet SO-, concentration (SO? absorption rate).
LJ
Results from the EPA pilot facility at Research Triangle Park
(Reference 6) have shown that it is possible to operate limestone and
lime wet-sc rubbing systems with liquors "unsaturated" with respect
to calcium sulfate, thereby completely eliminating the potential for
gypsum scaling. This is accomplished by controlling, within given limits,
the total oxidation of sulfite to sulfate and the amount of dissolved sulfate.
The Radian Corporation (Reference 5) has determined that the "critical"
sulfate saturation at 46°C is approximately 130 percent. The calculated
sulfate saturations (at 50°C) in this report were made with the use of
the Radian Equilibrium Computer Program (see Section 2.4).
TCA limestone reliability Run 526-2A (see Section 3) has been included
in Figure 6.
253
-------
210
200 -•
110 --
100 ••
90 --
80
I I I I
I T 1 I I
PERCENT SOLIDS RECIRCULATED = 8
PERCENT SOI LDS
RECIRCULATED = 15
LIQUID TO GAS RATIO = 64-80 gal/mcf
PERCENT SOLIDS DISCHARGED = 38
H 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1—I-
10 15
EFFLUENT RESIDENCE TIME, min.
510-2A
L 509-2A
502-2A
501-2A
20
25
Figure 6. Calculated Degree of Scrubber Inlet Liqxior Sulfate
Saturation from TCA Reliability Verification Tests
254
-------
Under these limits, the sulfate formed by oxidation is purged, without
crystallization, as a "solid solution" within the precipitated CaSC^.
The amount of sulfate which can-be purged in this manner is related to
the sulfate activity in the scrubbing liquor, which can be increased by
addition of magnesium ion (see Reference 6). In order to keep the
required magnesium addition as low as possible it is desirable to reduce
overall oxidation. Oxidation may be reduced by (1) sealing the effluent
hold tank (with a CC>2 or N2 insulating blanket, if necessary), (2) adding
oxidation inhibitors, and (3) reducing liquor rate. The required mag-
nesium addition for unsaturated operation appears to be higher for lime-
stone systems than for lime systems and also increases as the concen-
tration of chlorides in the process liquor increases. The high concen-
tration of chlorides in the Shawnee facility process liquors (see Tables 2
and 4 in Sections 3 and 4) are due to chlorides present in the coal which
are converted to HC1 and absorbed from the flue gas in the scrubber.
Operation in the unsaturated mode may also lead to improved process
economics, as operating conditions conservatively selected for scale
control under supersaturated liquor conditions can potentially be changed,
e.g., effluent residence time and percent solids recirculated may be
reduced. Moreover, the unsaturated process liquor may be used
effectively for the irrigation of mist eliminators and wash trays.
2.2 MIST ELIMINATOR SCALING POTENTIAL
It is theorized that mist eliminator scaling is caused, predominently,
by SO£ absorption into process liquor adhering to the mist eliminator
surfaces. Calcium sulfate scale formation is subsequently caused by
oxidation of the dissolved SO£ and the resultant super saturation of the
liquor -with respect to calcium sulfate. In addition, solids adhering to
255
-------
the mist eliminator surfaces can cause pluggage and may also act as
sites for additional scale formation. These problems may be alleviated,
therefore, by: (1) reducing the sulfate saturation of the mist eliminator
wash liquor (e. g. , utilizing raw water or "subsaturated" process liquor);
(2) washing the mist eliminator at a sufficient rate to remove all impacted
solids; (3) utilizing a wash tray to reduce the suspended solids concentra-
tion of the droplets impinging upon the mist eliminator; (4) reducing the
SC>2 concentration in the vicinity of the mist eliminator; and, (5) using
mist eliminators with improved draining characteristics (e. g. , sloped
mist eliminators). All of these concepts will be tested at the Shawnee
facility.
2. 3 CLOSED LIQUOR LOOP OPERATION
Results at the Shawnee facility have shown that scaling potential
(reliability) is significantly affected by the quantity of raw water input
to the system, i. e., the greater the raw water input, the lower the scal-
ing potential. In order to obtain significant reliability data, therefore,
the scrubber systems must be operated in a "closed liquor loop" mode.
Closed liquor loop operation is defined as operation wherein the raw
water input to the system is equal to the water normally exiting the
system in the settled sludge and in the humidified flue gas. Settled
sludge densities in lime/limestone •wet-scrubbing systems are normally
equal to or greater than 38 percent by weight of solids. Closed liquor
loop operation, of course, is also desirable from a water pollution
standpoint.
The original test facility design included slurry pumps with water seals
(Hydroseals) for bearing protection, water quench sprays for gas cooling
256
-------
and dilute limestone slurry feed (10 - 20 wt% limestone). The water input
under these conditions exceeded the makeup reqviirement for closed liquor
loop operation. The systems operated, therefore, with partially open
liquor loops during the limestone factorial tests, i. e. , sludge with less
than 40 wt% solids had to be discharged from the systems. This was
not considered to be a serious problem during factorial testing for, at
a specified scrubber inlet slurry pH, SC^ removal is not significantly
affected by liquor composition. However, little information was gained
about the effect of scaling potential on reliability during this period.
The absorbent feed systems were changed in November 1972, to provide
slurry feeds with up to 60 wt% limestone concentration. During the five
week boiler outage in February and March 1973, the Hydroseal slurry
pumps were converted to a Centriseal type (mechanical seal supplemented
with air purge) and the TCA and Marble-Bed scrubbers were provided
with process slurry gas cooling system. As a result of these modifications,
closed liquor loop operation has been maintained at the facility since the
beginning of limestone reliability verification testing in March 1973.
2.4 DETERMINATION OF SULFATE (GYPSUM) SATURATION
Three methods have been used at the Shawnee facility to determine the
degree of sulfate (gypsum) saturation of the process liquors. These
will be discussed below.
The most accurate of the three methods is a quantitative determination
of the degree of sulfate saturation by the use of the Radian Equilibrium
Computer Program (Reference 7) and the laboratory measured complete
257
-------
liquor composition of a process liquor sample. The Radian Equilibrium
Computer Program calculates the activities of calcium and sulfate ions,
based on the measured liquor composition and pH. Hence, the degree
of saturation is equal to the product of calcium and sulfate activities
divided by the solubility product of gypsum at the specified temperature.
The calculated sulfate saturations at SO°C in this report were based
upon a solubility product for CaSO.^- ZHgO of 2. 2 x 10~5 gmole^ /liter .
The second method is an approximate quantitative determination, which
is less accurate but sufficiently rapid that it may be used in the field
more easily to follow the scrubber performance. The test compares
the concentrations of dissolved calcium and sulfate in the liquor sample
with concentrations at saturation. The saturation concentrations are
determined after the liquor sample is agitated with gypsum powder for
a minimum of two hours. The degree of saturation is then calculated
as the ratio of the product of calcium and sulfate concentrations before
gypsum treatment to the product after treatment. This method assumes
that the activity coefficients of calcium and sulfate remain unchanged
after the gypsum treatment.
The third method is qualitative in nature, and does not involve any lab-
oratory analyses or calculations. Stainless steel plates or strips are
immersed in the process liquor in critical locations such as the effluent
hold tank and Koch tray effluent surge tank, etc. , to monitor the scal-
ing potential of the scrubber system. These scale probes are examined
periodically for scale formation. The scale growth rates can also be
quantitatively determined by periodically weighing the probes. This
method should be useful in a full scale facility, requiring a minimum
of cost and operator attention.
258
-------
Section 3
LIMESTONE RELIABILITY TEST RESULTS
Performance and analytical data from the TCA system limestone re-
liability testing at the Shawnee facility are presented in this section,
along with an evaluation of each reliability test and the conclusions
drawn, to date, from the testing.
3. 1 PERFORMANCE DATA AND TEST EVALUATION
A summary of the test conditions and results for each of the TCA
limestone reliability tests is presented in Table 1, along with the
run philosophies. A summary of the scrubber inlet liquor analytical
data for a majority of the tests is presented in Table 2, along with the
calculated percent sulfate saturations. Essential operating data for all
of the TCA limestone reliability tests are graphically presented in
Reference 1. The operating data for the initial 480 hours of Run 535-2A,
which is currently in progress, is graphically presented in Figure 7.
An evaluation and discussion of each test is presented below.
3.1.1 TCA Run 525-2A
On October 24, 1973, the initial limestone long-term reliability test
(Run 525-2A) was begun on the TCA system. The objective of the test
259
-------
Table 1
SUMMARY OF LIMESTONE RELIABILITY TESTS ON TCA SYSTEM
Pun No. !
SIAI-'- of- pyn D»'r
;nd-o<-Run D«.'« ;
Ort S.'Cn,
I.oopCl^.urr. «-. Solid. Di.chg.
Tot*l Di«»ol\rd Solid*, ppfri |
.\P B»nf{r. in, HjO
O*mt*<*r 4. P R»nRe. ta. H2O
>b.orbfru
StJirl-of-ftun
Method of Control
Run Phtlotophy
Sii-4A [ , f2.,-2>
10/24/7) ; ] 1/>1 /7S
U ' IS^7* I IMQ/74
M7 , Ht>o
?:.. 000 . 20. =-00
10. <- : S. t (
i;oo ' izoo '
• 0 7.1
M-lf H- U. '
10 10
i . -> - 1 . i : 1.2-1.*
71 f ^
1 !itn-40f)Q 1 !• 00-4100
rj.tm • 7^. ST
t.. i..-.. i . >. • ^- =.. "
£ . 2-1-. S ' "i. 1 .«..*•
1=.- >0 ' l^-l-
ClUrifirr : CUrtft^r
11 -42 ! i^-47
MO MO
7o no. i., no I 7«.oo.t)7Cio
^. ^-- . <• 4. S-4. ?
2. J-2. 7 110/24-1 1 M2) 1 1. 0-2. \
2.7- », 1 (llMi-H'1^1 |
' 0. IS-0. Z<-
Lim<-,t»n«- ,!llrr.rd to '.£> w-i *. L.rrtr.Ionr 3)UJ-rirrf 1« '-0 »f '.
to Fnrr. to Cf(T.
»pirKt"""a °"
TPR h»lf-«pr.rr«.» plugRed Bottom of drmiittr 1 ^r.
to be prii"» rv p robl*rn).
Hot torn of drmiater ''S*',
plURRfd.
^27-2A
1/18/74
1/24/74
IM
20. ^00
S.r,
1200
.... 7> ...
1 <>- li j
_. - .-.1° i
l.^-l.O
!.t '
2-00- JSOO |
r 0
7700-8000
4. '.-4.0
2. 1 • J- 1
0. JO- 1. 00
ro EJIV.
>*rlo
1200
7)
14-It.
:\2
1.1.1. 41
7?
'000-4000
7o. <>;>
»., 7. t,, Qt.
'=..2-^.',
10- !l>
CUrlfler
;i- ^3
no
4200-t>300
+. Q-i. J
2.0-2.2
0. It.-O. 20
to EHT.
crubbrr (oop. Dccrf*»rd
fl*l2r»
Inlrl pll* S^1-
Stoich. H*tio & 1. t.s
1o«-
-------
Table 1 (continued)
SUMMARY OF LIMESTONE RELIABILITY TESTS ON TCA SYSTEM
Run No.
Start . of- Run Rite
End-of-Run n»ce
On Sire»ni Huuri
Liquor Rite, upni
L/C. R.l/mcf
Affluent ftendence Tfmr. min.
SlQKhiomctnc lUtio. m«Ui C*
roolei SO^ »bi. /molr Ca *dd«-d
Inlet SO^ Concentr*., on. ppm
Scrubbor Inlci pH R*nge
Prrcrni Sulfur O*;d»*rd
Solid* Duposal System
LoopClo.ur... -, Sol.d, D.,chK.
Scrubber Inlrl l.tquor 6 =,0°C
Demiilrr *nd Koch Tray. In. H2C
AP R*n«c, in. H;O
Demi.t*r A P R»nKc. in. H;O
Sl«rt-of- Run
Method of Control
Results
•-•-
>
, ___ J"
20. ,00
...
u
I. U- I. U
80
""
(, ;.(, ,j
•- "..i.,.r- - ^
[H. 10
—""" 1
10-47
no
4.1.* 1
1. 9.J. 1
0. IT-0. 2X
to EHT.
*ble cUi-ified J,,»r>r l-l 4 Kp»..
N" """«"•
Overrides;
Inlrl pHsV 9-i
Stoich. Raiioi |. ,.<,
liquor-
lolidi And flight scale buildup
dcmistcr 1 9r; plun^H.
^ JO-iA
Wed'74
4/17/74
4Ti
20. '•OO
«...
i;oo
"
,4.1.-
i.1
!.:-,.•.•.
7 3
2200-4200
7S.H4
'-. l-<..4
li-il
CUr.hrr
[Crntrif'iRe only * f i r r 4'10i
10-4J
120
i lOO-t'- CO
4.9-.1
J.o-J. J
0. 17.0. 2*.
»nd »dded to EHT.
> »t»Res 14 firidil wtth «."
™n. " P
5ra>d top nf EHT.
Inlet pH s •,. 91-
Stoich. Ran,)* 1. (^
pt»«ibly c»uird by low
di-rr to Kdch tr*>". Bottom
ot demnter 441". plu([R*d.
,,,...A
«. ' 10 '74
JO. t-00
12 no
7.1
U
'•i-1i:,'.;\'l'V.v.-.;.'^"'"'
j . 4 li , ,c
L -. -- ' .' -
71.0.
V4...1 1
4.7... <
1 1.40
CUril.rr
l*.i.>
Ml
iv.OOO
t. ' -• .'
-'• ' ' '• :
*•'"•>• "
ind Jddrd 10 RHT,
run.
SrlO-S/£l 84:2".
topr<...d)c buildup on mis* rlim-
,,i.2A
ViO'74
J*H
JO. -00
*••
1 :oo
104
7.3
1.'
>..-,.<.
?:
,'fifirt. 1 JjlO
Q,..«Q
t. ' -• , 0
-.2-V-
Clarififr
12 .-11
14.*
4-. .11)0. • :). 0,irt
4. ^-4. ^
1. *'••.?. I
a. li.o. (•>
»nd *drffd t,, RUT.
• •-• - -
recycle liquor (pt\i» 1 *> (ipni
P">- •
»t S. 8;0, 2
i^,r;;r;,orPPrP
K.,ch.lrl.y'rrfVelf.|uep.
ppm Mg. Run fri-niinAtrd due
tloitom ,.f demiit.^ 1-2*1.
261
-------
Table 1 (continued)
SUMMARY OF LIMESTONE RELIABILITY TESTS ON TCA SYSTEM
Run No.
Staff- of- Run Dal*
Knd-.'f.Run t)-te
On 5. niolr* C»
Av(j ", l.iniriionp Viilt«Miun. 10o*
Prrerni SOj Hcmox*]
Scrubber Inlet ptl R*nKr
Scuibbt-r Cot»r1 pit RanRr
Prrccrtl SuUu r Cxirtirrd
Solid* Di»po*i) ri>-*trn!
Loop Cl..Mirr. "- S«Ud. 0,,chR.
Sc rubber Irtln Liquor ^ =.r>"C
Tola! f)u»olvrd Solid*, pprn
Orrni*l*r and Ktich Triy. '"- H^O
AP R»n«r, in HjO
D^iniitf r AP K^ngp. in. ItjO
s"»hb" '»""'"
Method of Control
S1J-2A
8/r./T4
VZH?4
UJ
20. '-OO
8,r.
uoo
71
14. <--!=,. S
12
t. 1- l.<.«,
Tt.
2000- J7SO
9^.90
1. 7-t.. 0
••. J-S.t-S
IS-2J
CUrifirr
2q.i«
1 i1.
^i. OOiJ.'-rt. 000
2,0<,. \. H*,
0.17-0.2S
*rtd Jddril l» MIT.
r,rw TPR -phcri.
Sv»«*^*le»hrd. (UpU^d
TPtt'».
at l.H'O. 2
CO2 bl»nhcl over EHT (.
PWHT. and mi»i elim. /KT
rrcyclr U>0p.
.^^.......y...^ s
1J4-2A
<»/ J/71
«>B/74
100
20. SOO
*.(.
1200
:i
10-12.*,
12
i. io-u to
82
2700- Jt;00
7S-00
S. t,^-^, 8^
S.2-S.4
IS- )0
CUrihrr
JO-<(J
1 )0
T'joo-S'.oa
I. 8^-2. 00
0. IS-O. JO
and artdrd lo F)1T.
run.
5y.ttmel»»r,*d.
»t '-.«l0.2
P ,
^>S-2A
9,M2/74
In Proj[r«fti
20. SOO
8-fc
1ZOO
7)
12-IS
U(«/12-q/27|. lM*rt
Stf.i«-h. HAIIO* 1. 7
^Zt..2A.
hrnlrt. \Q n\ilt i^felr »n
all Rrirta.
262
-------
Table 2
AVERAGE SCRUBBER INLET LIQUOR COMPOSITIONS AND CALCULATED
SULFATE SATURATIONS FOR TCA LIMESTONE RELIABILITY RUNS
Run No.
525-2A
S26-2A
528-2A
529-2Ala>
530-2A(b)
S32-2A
S33-2A
S34-2A
535-2A(C)
Percent
Solids
Recirculated
14-16
14-15. 5
14-16
14. 5-16
14-16
7-9
14.5-15.5
10-12. 5
11. 1-15. 5
Effluent
Residence
Time, min.
10
10
12
12
12
12
U
12
15
Percent
Solids
Discharged
31-42
35-47
25-33
30-47
30-43
32-43
29-38
30-40
35-42
Percent
Sulfur
Oxidized
15-30
15-35
10-30
15-30
12-25
7-25
15-21
15-30
15-25
Scrubber
Inlet pH
Range
5. 5-6. 1
5.65-5.9
5.7-5.95
5.2-5.5
5. 1-5.4
5.2-5.6
5.1-fc.O
5.65-5.85
5.8-6.0
Inlet Liquor Species Concentrations, mg/1
Cat+
2100
2300
1400
1900
1730
570
•us
2220
1850
MgtV
250
340
200
330
340
10,500
u.aoo
170
290
Na
80
60
30
120
50
60
GO
40
50
K*
50
140
120
520
120
no
no
180
70
so/
80
80
150
30
120
3000
2570
70
60
so4-
2000
1900
1600
2000
1900
38,700
17.200
1730
1800
(ppm)
CO/
70
210
260
ISO
320
30
200
90
80
ci-
3200
3700
1600
3300
2790
2400
1100
3540
3060
Total
7900
8700
5400
3400
7370
56,900
55.750
8040
7260
Calculated Percent
Sulfate Saturation
at 50°C«»
UO
130
110
130
120
140
135
130
120
CO
Note: Average concentrations are for steady-state operating periods.
Solids disposal sysbem: Clarifier only.
(a) Only one sample was taken during steady-state operation,
(b) Solids disposal systems: Clarifier only for first two-thirds of
lest, then centrifuge only.
(c) Test run was started on 9/12/74 with 12 minutes residence time. Residence
time wag increased to 15 minutes on 9/27. The run is still in progress as of
10/lfc. The values given in the table are for period ")fZ7 through 10/U>.
fd)
-------
§ 3
El
S§ '
s
1ION - - j REHEAIiR BURNER CltftNINC --J INSPECTION -J
VSPJCTIQN —.1
1.SOO
3000
2.000
I WO
3,000
7.VX
1.000
\ »|J I
I 9'N I 9'M I 9 M- I 9'2? ! 9'78 I 979 I 9,'X I ID'1 ]
CALEHUAR DA
Ml
4 S 5
tt * |
ijs
1 6
15
14
1 3
- X"\ -"A \ !\
• / ^ \ / \ /Xy^ J . \ /"\x /V .y -
"\ •' \.; \ ^7 \..-v,X
- ' • \ J
si I
-
I3.OOO
5 u.ooo
1
r 11.000
g 10000
•>
a 9.000
^ B.OOO
?
S 7.000
? 1.000
? 5000
S
D 4000
2 1°0°
>
a ».«»
s 1000
0
_ • TOTAL DISSOLVED SOLIDS
<^> CALCIUM IC*"J
C SUlf AT( 1503 -1
A CHLORIDE ICI "I
NOTE SPECKSWMOSE
*~ CONCENTRATIONS ARE USS
THAN SOO Wf ARC WOT
PLOTTSO
• -
-, •• •••• • ,
* * * * • 1
.
1
• A * * A - A -
************ A
"89e9D2*88ee.oege gc t Q,
"
1 U 1 — 1 1 1 1 1 l 1 1
oooo
i;.ooo
Il.OCC
10.000
•9.000
1.000
7000
(000
5000
4000
3X10
Jooo
1,000
0
TEST TIME hou.t
. | 9'1G I »MT I 9/16 ! 9/99 I 9/30 I 9>}l I 9/H I 9/2] I 9/74 1 9/75 I 917* I 9/77 I 97S I 9/29 I 9rx I 10/1
CALCNOJXROA
Gasflele- 20.500 aclm P 300 °F
Liquor Rale * 1200 gprn
L/6 • 73 gal/mcl
Gas Velocity = 8.6 li/uc
EHT (Sealed) H«il)[nM Time • 12 min(9/12-9/27).
15m.n(i)lleT9/27l
Thiee Stages. 5 in ipheres/slage
Percent Solids Redtculattd • i: 15 w %
Total Pre'^Te Drop. Excluding Qtmwtt
ano Koch Tiay = 4.M.S in H,0
Serubbei Inlet Liquor Temperature =120-l2tS°F
Liquid Conductivity 4.800-10.000 u mhoi/cm
Discharge (Clarilier) Solids
Concentration < 35 42 wt %
Figure 7. Operating Data for TCA Run 535-2A
264
-------
was to operate cont imiously for from four to six months. The operating
conditions selected for (lie run were based on the results of the relia-
bility verification tests (see Figure (>) and the tests conducted at the EPA
pilot facility at Research Triangle Park . The major test conditions
•were (see Table 1):
Gas Velocity 10. ^ ft/sec
Liquid-to-Ga s Ratio (>0 gal/mcf
Pe rcent Solids Reci rcula I ed I ^
Effluent Residence Time 10 min.
Percent SO^ Removal (controlled) 84
During this test, the Koch Floxilray was irrigated with the available
raw water makeup plxis all the clarified liquor at an approximate one
to three ratio, with the effluent irrigation liquor routed to the scrubber
effluent hold tank. The mist eliminator, at that time, had no provision
for underside wash and was washed with liquid entrainment from the
Koch tray.
For the first time, thermo-plasl ic - rubber (TPR) spheres, supplied by
UOP, were tested in 'he scrubber. They were used in the top bed while
the bottom two beds were charged with high density polyethylene (HDPE)
spheres, also supplied by UOP. The original wire mesh grids (which
eroded during the limestone reliability verifications runs) had been
replaced by sturdier bar-grids for the long-term reliability runs.
The velocities listed in this report are based upon scrubber outlet
gas flow rate at 12E>°F. Gas velocities listed in previous Shawnee
reports have been based upon scrubber inlet gas flow rate corrected
to 12S°F. The velocities al the scrubber outlet conditions are approx-
imately seven percent higher than at the scrubber inlet conditions,
due to moisture pickup.
265
-------
After 517 hours of operation the run was terminated, due to unusually
heavy solids bxiildup on the underside of the Koch Flexitray and scale
and solids buildup on the bottom vanes of the mist eliminator (about 60
percent of the free area was plugged). Numeroxis (over 200) half-spheres
of the TPR type were found in the scrubber and slurry circulating system.
TPR half-spheres were also found lodged in two of the four inlet slurry
spray nozzles. It should be noted that the scrubber stages (and bottom-
most grid) were essentially free of scale after the 517 hour operating
period, as was expected, since the calcxdated scrubber inlet liquor sulfate
saturation was 140 percent (see Table 2).
It is hypothesized that the accumulation of soft solids below the Koch
tray was due, primarily, to the partial blockage of the slurry inlet, nox-
zles, resulting in excessive entrainment of the fine slurry droplets.
This excessive entrainment and the high gas velocity (10. 5 ft/sec) most
probably contributed to the mist eliminator pluggage.
3.1.2 TCA Run 526-2A
Run 526-2A was begun on November 21, 1973. The TPR spheres in
the top bed had been replaced with FIDPE spheres , and the accumulated
scale and soft solids from Run 525-2A had been removed. The run
conditions were identical to those for Run 525-2A, except that the gas
velocity was reduced to 8. (> ft/sec. The velocity was reduced because
more detailed investigation of previous reliability verification runs
It was planned to use TPR spheres in all three stages following the
installation of strainers in the inlet slurry piping.
266
-------
(see References 1 and 2) indicated that long-term reliability for the
present Koch tray/mist eliminator configuration could not be expected
at a gas velocity of 10. 5 ft/sec.
On January 10, 1974, the run was interrupted after 1190 hours of on-
stream operation, to check the wear of the HOPE spheres in the three
beds (HDPE sphere life had been estimated to be less the 2000 hours).
Pressure drop across the chevron mist eliminator increased slightly
during the initial 800 hours of operation, and during the last 400 hours
increased more rapidly to a final level about 1. 5 times the initial value
of 0. 18 inches H2O.
The general appearance of the system was good. Scattered solids
deposits (up to 1 inch) and light scale (about 1/16 inch) was found on
the scrubber walls below the bottommost grid and on the wall areas
not in contact with the spheres. The four bar grids were covered with
10-14 mil scale on the inactive surfaces (i. e. , surfaces not in contact
with spheres).
A heavy, relatively uniform solids layer covered the underside of the
wash tray. All four inlet slurry spray nozzles were partially plugged
with debris, primarily with plastic covering from pipe insulation.
The flange of the steam sparger underneath the wash tray was found
leaking. Mist eliminator plugging was confined to the bottom two passes
only, reducing the free flow area by about 15 percent. Several small
pieces of solids fell from the outlet gas duct and rested on top of the
center section of the mist eliminator. The area restricted by these
pieces was insufficient to affect the pressure drop across the mist
eliminator.
267
-------
As with Run 525-2A, it is hypothesized that the solids accumulations
on the slurry inlet spray nozzles and header, on the adjacent scrubber
walls and on the underside of the Koch wash tray were primarily caused
by partial blockage of the TCA slurry inlet spray nozzles by debris,
which changed the spray pattern and caused excessive entrainment of
fine slurry droplets. Also, the blocked nozzles may have been a major
factor contributing to the mist eliminator pluggage observed. It should
be noted that the accumulation of sulfate scale on the bottommost bar
grid (10-14 mils) was not excessive, after the approximate 1200 hours
of operation. This was to be expected, since the scrubber inlet slurry
was only 130 percent saturated with respect to calcium sulfate (see
Table 2).
The average percent weight loss for the HDPE spheres in the bottom
two beds was approximately 32 percent, for 1707 hours (Runs 525-2A
and 526-2A) of use.
3.1.3 TCA Run 527-2A
Run 527-2A commenced on January 18, 1974, following the removal of
the soft solids from the walls (inlet slurry spray nozzles area) and from
the bottom of the Koch tray (the mist eliminator was not cleaned). It
was a scheduled short-term run to determine whether certain changes
in the scrubber internal configurations could minimize the mist elim-
inator pluggage problem and to observe the effect of the changes on
SC>2 removal and limestone utilization. The modifications included
raising the Koch tray outlet weir height from 2 to 3 inches, using two
beds (3 grids) with 71/2 inches of spheres per bed and the removal of
268
-------
the four inlet slurry spray no/.zles for minimum spray generation with
open pipe slurry distribution.
The results of the changes were inconclusive and the run was terminated
after 133 operating hours. The slxirry carryover to the Koch tray and
the mist eliminator appeared to be equal to or greater than the carry-
over observed dxiring Run 526-2A.
The mist eliminator plugging continued on the lower vanes and the free
flow area was reduced by an additional 50 percent (for a combined total
of (-5 percent for Runs 526-2A and 527-2A). It shoxild be noted that
the Koch tray, inadvertently, was not irrigated for a 2 hour period
during the run.
An additional 20 mils of scale formed on the scrubber walls below the
bottommost grid.
3.1.4 TCA Run 528-2A
Run 528-2A was started on February 6, 1974, following a complete
cleaning of the system. The run was intended to be of short duration
•and was made using raw water only for Koch tray irrigation. The
Koch tray effluent was routed to the sewer. The purpose of the run
was to observe the effect of irrigation with raw water, versus raw
water plus clarified liquor, on the mist eliminator scale and solids
deposition. Scrubber internal configurations were returned to Run
S26-2A conditions (2 inch Koch tray outlet weir height, use of inlet
slurry spray nozzles, three-bed operation). The HDPE spheres were
269
-------
replaced with ne\v, improved, TPR spheres al 5 inches of spheres in
each bed. A duel strainer (by Elliot Co. ) was installed in the scrubber
feed slurry loop to prevent the inlet spray nox,/,les from plugging by
debris.
The run was terminated on February 26 after 425 hours of operation
and the appearance of the system was generally good. The top of the
Koch Flexitray was covered with approximately 20 mils of dust type
solids. The mist eliminator was free of scale and solids with only a
slight film of dust covering the inlet vanes. The pressure drop across
the mist eliminator and Koch Flexilray had not increased during the
test. The bottom of the wash tray was relatively clean.
Evidence from this test showed that the potential for scale and solids
deposits on the mist eliminator is decreased when the degree of sulfate
satxiration of the Koch Flexilray wash liquor is reduced.
It should be noted that excessive weepage from the Koch tray irrigation
water caused the test to be made with a slightly open liquor loop (average
discharged solids of 29 percent). This explains the relatively low cal-
culated scrubber inlet liquor sulfate saturation of 110 percent (see
Table 2).
The run was interrupted for about 40 hours (Febrxiary 11-13) due to
a No. 10 boiler maintenance outage.
270
-------
3. 1. 5 TCA Run 529-2A
Run 529-2A was started on Februa'ry 26, 1974. The test conditions
were the same as for Run 528-2A, except that the Koch tray was ir-
rigated with the available raw water makeup plus all clarified liquor
at an approximately one to three ratio. The run was initially intended
to be a long-term limestone test. However, an inspection on March 7
revealed that the lower vanes of the mist eliminator were approximately
19 percent plugged with scale and soft solids. This was considered to
be significant after the relatively short period of time and the run
was terminated after only 213 operating hours.
Nearly all the TPR spheres were dimpled. Eight spheres had failed
(filled with slurry) in the bottom bed.
3.1.6 TCA Run 530-2A
Run 530- 2A was started on March 28, 1974, after a thorough cleaning.
The Koch tray was irrigated at a constant rate of 1 5 gpm, consisting
of about 8 gpm makeup water and 7 gpm clarified liquor, with the
purpose of achieving a high dilution of Koch tray irrigation liquor
and, hence, minimizing the mist eliminator and Koch tray scaling
potential (see Run 528-2A).
This partially collapsed condition is considered normal by the supplier
due to pressure and temperature cycling between operating periods
and shutdowns.
271
-------
The run was originally intended to be a long-term reliability test.
However, it was terminated after only 476 operating hours when a
rapid increase in pressure drop across the Koch tray occurred. An
inspection revealed heavy scale and solids deposits on top of the Koch
tray. A clarifier rake malfunction (which was not discovered until
April 10) caused heavy solids carryover in the clarifier overflow for
about two days. The solids carryover and the lower (15 gpni) Koch
tray flush rate probably resulted in the settling of solids on the Koch
tray. Subsequent use of the centrifuge for slurry dewatering resulted
in centrate containing about 0. 5 wt. % of solids, which may have con-
tributed to further settling of solids on the Koch tray.
The mist eliminator was heavily plugged. Inspection on April 9 (after
290 hours of operation) showed an estimated overall restriction of
8 percent which then increased rapidly in 186 hours to 44 percent at
the end of run.
Approximately 100 TPR spheres (0.4 percent of the total inventory of
the system) failed during the run.
3.1.7 TCA Run 531-ZA
Run 531-2A was begun on May 10, 1974, after a thorough cleaning.
Magnesium oxide was added to the effluent hold tank in an attempt to
operate in the sulfate unsaturated mode. Also, the TCA slurry header
was lowered four feet in an attempt to reduce the amount of slurry
droplet entrainment to the Koch tray.
272
-------
The run was terminated after 1088 operating hours due to increasing
pressure drop across the mist eliminator, Koch tray and bottommost
TCA bed. Inspection revealed considerable scale and solids deposits
on the Koch tray, mist eliminator and bottom grid. The mist eliminator
v/as 75 percent plugged at the inlet edge and within the first pass. Scale
deposits were limited to the first two passed.
Heavier than usual solids buildup was observed on the walls between
the slurry spray nozzles and the steam sparger. This could be due to
insufficient wetting of the walls by entrained slurry, resulting from the
lowering of the slurry spray header.
The run had never attained steady state (i. e. , the desired magnesium
ion concentration of 10, 000 ppm was not reached) and the final mag-
nesium concentration of 5, 000 ppm in the process slurry was insuf-
ficient to attain unsaturated operation.
A total of 638 TPR spheres (2. 4 percent of the starting inventory) had
failed during the run. The wear of the intact spheres varied from
no measurable loss in the top bed to a 6. 5 percent weight loss in the
bottom bed for over ZOOO hours of operation.
3.1.8 TCA Run 532-2A
Run 532-2A was begun on July 17, 1974, after the system was cleaned.
Again, magnesium oxide was added to the effluent hold tank in an attempt
to operate in the sulfate unsaturated mode at a 10, 000 ppm level of
magnesium ion in the process slurry. The scrubber recirculation
liquor rate was increased from 1200 to 1700 gpm.
273
-------
Also, a closed recycle irrigation liquor loop was provided for the Koch
tray/mist eliminator system which included provision for mist elim-
inator underwash (~ 8 gpm fresh water plus ~ 7 gpm recycle liquor).
The Koch tray recycle irrigation rate was set at 35 gpm. The clari-
fier liquor, which had been used for Koch tray irrigation during the
previous tests, was routed directly to the effluent hold tank.
Because of relatively high saturation (140 percent) of the inlet scrubber
slurry and the liquor in the wash tray loop, an inspection was made on
July 29, after 258 operating hours. About 200 mils thick scale covered
the walls below the bottommost grid and the top of the Koch tray was
covered with 12 mils of scale. The free flow area of the mist eliminator
was 1-2 percent plugged with solids, within the first and second passes.
About 4000 TPR spheres (15 percent of the total inventory) were found
floating in the downcomer. They had passed through the bottom grid.
In view of the above problems, the run was terminated. Note that, the
SC>2 removal efficiency for this test averaged about 98 percent, as com-
pared to 84 percent for the previous test, due to the addition of mag-
nesium (see Table 1).
3.1.9 TCA Run 533-2A
Run 533-2A was started on August 6, 1974, after a thorough cleaning
of the system. The recirculating slurry rate was reduced from 1700
to 1200 gpm, the used TPR spheres in the three beds were replaced
The 11/2 inch TPR spheres can pass through the bar-grids supplied
by UOP when they become dimpled on both sides.
The liquor rate was reduced because of increased erosion of steel
piping during the previous test.
274
-------
with new ones and the percent solids recircxilated was increased from
8 to 1 S percent. The Koch tray/mist eliminator system was the same
as in the previous test, with a closed recycle irrigation liquor loop
for the Koch tray and mist eliminator underwa sh. The scrubber slurry
magnesium ion concentration was maintained at 10,000 ppm. The
scrubber inlet liquor, however, was still supersaturated with respect
to sulfate (see Table 2).
The run was terminated after 33Z operating hours when an inspection
showed that the underside of the wash tray was covered with up to 60
mils of scale over 40 percent of the tray area. The top surface of the
tray was coated with from 50 to 100 mils of scale. The mist eliminator
was 12 percent plugged with scale/solids deposit, primarily in the
second and third passes. Also, the walls between the steam sparger
and the slurry spray nozzles had up to 2 inches of solids deposit.
3.1.10 TCA Run 534-2A
Run 534-2A was started on September 3, 1974, after the system had
been cleaned. It was decided to abandon, for the time being, attempts
to operate in the sulfate unsaturated mode with the addition of mag-
nesium oxide, until further exploratory tests were conducted at the
EPA pilot facility in Research Triangle Park. Therefore, conditions
for Run 534-2A were chosen similar to those of Run 526-2A, which had
been in operation for 1190 hours, with no apparent scaling problem on
the top of the Koch tray. The underside of the mist eliminator was
washed continuously with 15 gpm diluted clarified liquor (~ 8 gpm
makeup water plus~ 7 gpm clarified liquor) while the Koch tray was
275
-------
irrigated with the remaining ~ 8 gpm of clarified liquor plus the 15 gpm
mist eliminator wash. The main differences between Runs 534-2A
and 52(>-ZA were, therefore, the addition of direct mist eliminator
unclerwash and the four feel lower elevation of the inlet slurry spray
header. In order to reach the expected steady state chloride concen-
tration quickly, CaCl2 was added to the system before startup.
Run 534-2A was terminated after 100 on-stream hours due to scale
buildup on the Koch tray. The top of the tray, including the valve caps,
was almost completely covered with from 10 to 15 mils of scale. About
2 percent of the mist eliminator flow area was restricted by solids.
Also, about 30 percent of (lie Koch tray underside surface was covered
with solids, reaching a maximum thickness of 1 1/2 inches.
As in the previous run, the walls between the steam sparger and the
slurry spray nozzles had up to 2 inches of solids deposit.
3.1.11 TCA Run 535-2A
Run 535-2A was begxm on September 12, 1974, after the system had
been cleaned. The test conditions were nearly identical to Run 534-2A,
with the added condition that a minimum clarified liquor flow rate of
15 gpm for Koch tray irrigation and mist eliminator wash was to be
It normally takes a number of weeks for the chloride level in the process
slurry to reach a steady stale value, when beginning a rxin with raw
water. The process slurry was dumped at the beginning (if Rxm 534-2A
to dispose of the large quantity of magnesium ion in the system.
276
-------
maintained. In maintaining this rate, the fluctuating bleed flow to the
clarifier resulted in, at times, only 12-13 percent solids (as compared
to the desired 15 percent) in the reci rculating sliirry. Therefore, the
effluent residence time was increased from 12 to 15 minutes, in order
that the sulfate saturation of the scrubber liquor remain below 135
percent at all times (see Figure 6). Also, about 12 percent more
make-up water was available for mist eliminator/Koch tray irrigation,
because of the elimination of seal water from a flue gas cooling spray
pump.
After 158 hours of operation, an inspection showed about 15 mils of
scale covering about 25 percent of the top of the Koch tray and rela-
tively heavy solids accumulation (up to two inches thick in some areas)
on the underside of the tray. The mist eliminator was about 5 percent
plugged with solids (mostly fly ash) on the lower vanes. The walls
between the steam sparger and the slurry spray nozzles had up to 4
inches of solids deposit. The run was continued.
After 316 hours of operation, a second inspection revealed that the
scale had disappeared from the top of the Koch tray and that the mist
eliminator was only 1-2 percent plugged with scattered solids (mostly
fly ash) on the lower vanes. The bottom of the Koch tray, however,
still had heavy solids accumulation. At this time, it was found that
the steam pressure in the Koch tray steam sparger (see Figure 2) had
drifted to 50 psig from an original setting of 125 psig. The steam
pressure was reset and the run continued.
After 431 hours of operation, a third inspection showed that the top
of the Koch tray was still clean and that the mist eliminator was
277
-------
practically clean (approximately one percent plugged with scattered
solids on the bottom vane). The solids accumulation on the under-
side of the Koch tray had been substantially reduced, with 30 percent
of the underside clean to metal. However, solids buildup on the walls
between the steam sparger and the slurry spray nozzles had increased
to a miximum thickness of 8 inches in some areas.
After 786 hours of operation, a fourth inspection showed that the top
of the Koch tray was still clean and the mist eliminator again prac-
tically clean (approximately one percent plugged with scattered solids
on the bottom vane). The solids accumulation on the underside of the
Koch tray was slightly increased, with 20-25 percent of the under-
side clean. The solids buildup on the walls between the steam sparger
and the slurry spray nozzles had further increased. The maximum
deposit in one area had reached 15 inches in thickness. Approximately
30 mils of scale had accumulated on the grids during the 786 hours of
operation. There was no measurable increase in pressure drop across
the beds, Koch tray or mist eliminator. The run •was continued fol-
lowing the 786 hour inspection.
The average wear of the TPR spheres, tor l^uu nours ol operation,
was about 4 percent.
3. 2 CONCLUSIONS
3. 2. 1 Scrubber Internals Operability
Both the limestone reliability verification and reliability test results
have shown that scrubber internals can be kept relatively free of scale
278
-------
if the sulfate (gypsum) saturation of the scrubber liquor is kept below
about 135 percent. This can be accomplished with the proper selection
of percent solids recirculated, effluent residence time and liquid-to-gas
ratio (see Figure 6). The data indicate that, at approximately 15 percent
solids recirculated, 10 minutes effluent residence time and 73 gal/mcf
liquid-to-gas ratio, the scaling of the TCA internals will not become a
problem to maintaining long-term system operation. For example,
the four bar grids in the TCA were covered with only 10-14 mils of
scale after 1190 hours of operation in Run 526-2A.
A significant recent problem in the TCA has been associated with
solids buildup on the walls between the steam sparger and the slurry
spray nozzles (see Runs 531-2A through 535-2A), which originated
after the slurry spray nozzles had been lowered by four feet. It is
hypothesized that the solids buildup is due to insufficient wetting of the
walls by entrained slurry. The problem may be corrected, therefore,
by washing the walls between the tray and nozzles or by raising the
nozzles to their original position.
Until recently, a significant limiting factor in the long-term reliability
of the TCA scrubber has been associated with the erosion and subsequent
collapse of the UOP supplied HDPE spheres (sphere life was approxi-
mately 2000 hours). The collapsed spheres eventually filled with slurry
and settled to the bottom of the support grids. Starting with Run 528-2A,
an improved type of UOP supplied TPR sphere was used in the TCA
\
beds. After approximately 2500 hours of operation, all the-spheres
were dimpled on one side, about 2. 4 percent failed at the seam and the
weightless averaged about six percent. It is estimated that these TPR
spheres would not need replacement until after about a year of service.
The problem associated with dimpled spheres falling through the bar-
grids (see Run 532-2A) can be solved by redesigning the grids.
279
-------
There has been no evidence of significant erosion of the bar-grids in
the TCA after more than 5000 hours of operation.
3. Z. Z Koch Tray/Mist Elirninator Operability
The most significant reliability problem encountered dxiring the TCA
limestone reliability tests has been associated with scaling and/or
plugging of the Koch Flexitray and bottom vanes of the chevron mist
eliminator (see Figures Z and 3).
Run 5Z8-2A showed that the mist eliminator and top surface of the Koch
tray can be kept relatively free of scale if the irrigation liquor is low
in sulfate saturation (unsaturated).
Run 5Z6-ZA showed that the top of the Koch tray can be kept relatively
free of scale if it is irrigated with the available raw water makeup and
clarified liquor, provided the mixtxire is maintained below the critical
sulfate saturation level. The run was terminated prematurely, after
1190 hours, due to excessive wear on the HDPE spheres and pluggage
(with debris) of the scrubber slurry nozzles.
The present Run 535-ZA has been in operation for over 780 hours
without any significant scaling or plugging of the top of the Koch tray
or mist eliminator surfaces. The test conditions for this run are
similar to those for Runs 5Z6-ZA and 5Z9-ZA, with the exception of
direct mist eliminator underwash. It is likely that long-term relia-
bility of this mist elimination system -would be realized at the rxm
conditions tested.
280
-------
Section 4
LIME RELIABILITY TEST RESULTS
Performance and analytical data from the venturi/spray tower system
lime reliability testing at the Sha\vnee facility are presented in this
section, along with an evaluation of each reliability test and the con-
clusions drawn, to date, from the testing.
4. 1 PERFORMANCE DATA AND TEST EVALUATION
A summary of the test conditions and results for each of the venturi/
spray tower lime reliability tests is presented in Table 3, along with
the run philosophies. A summary of the scrubber inlet liquor analy-
tical data for a majority of the tests is presented in Table 4, along
with the calculated percent sulfate saturations. Essential operating
data for all of the venturi/spray tower lime reliability tests are graph-
ically presented in Reference 1. As an example, operating data for
Run 604-1A from June 6, 1974, to June 24, 1974, is graphically pre-
sented in Figure 8. An evaluation and discussion of each test is pre-
sented below.
The sulfate saturations and analytical data are for an average inlet
gas SO2 concentration of Z800 ppm. The effect of inlet gas SO2 con-
centration on sulfate saturation is discussed in the following sections.
281
-------
Table 3
SUMMARY OF LIME RELIABILITY TESTS
ON VENTURI/SPRAY TOWER SYSTEM
Run No.
Start-of-Run Date
End-of-Run Date
On Stream Hours
Gas Rate. acfmtp1 J30°F
Spray Tower Gas Vel. fps @ 12<>0F
Vrnturt /Spray Tower
Liquor Rates. Rpm
Spray Tower L/>tomrirjc Ratio, molts Ca
added/molf SOj absorbed
Avg ~*0 Limestone Utilization. lOOx
moles SO^ abs, /mole Ca added
(rtlet SO^ Concentration, ppm
Percent SO; Removal
Scrubber tnlel pH Range
Scrubber Outlet pH Range
Percent Sulfur Oxidized
Loop Closure. «*, Solids Oischfl,
Calculated *"- SuUale Saturation in
Dissolved Solids, ppm
Total A.P Pange, Excluding
Demister, in. HjO
Venturi A P. in. H2O
Demisier _%P. in. H2O
Absorbent
Mist Eliminator
Scrubber Internals
System Changes before Start
of Run
Method of Control
Run Philosophy
Results
',01 -IA
10/9/73
l/B/74
2m
2^.000
',.7
f.00/1200
r.O
7.T
12
Clartfier Only Clarif Jr Intermittent Clarifier «. Filter
f 10/9-1 1/7) Filter 11 1/7-1 2. Ml) (12/lS-l/P^
I. 01- 1.28 J. 02-1. 18 '- 04-J. 1*1
If. 00- 3900 1 'jQO- 4000 2 1 00-4400
«.8--9l 7S-91 7<-°*
7.4-8. 5 7. *.*.«. 7.7.R.4
4.7-S.S 4. 7-1. <• 4.8-S. *
10-30 10- 30 10-10
20-2*. 20-27 42-12
110 120 ISO
1700-7SOO 4^.00-7300 9800-12. JOO
11.0-11.1 11.0-12.0 11.7-12. 3
009
0. SI- 0. 70 0. f.0-0. 90 0. °0- I. 61
to EHT.
10/9 - 12/11; Bottom washed with available makeup water
(^-14 gpml plus available clarified liquor \~Zt> gpm). Con-
tinuous wash rate of 0.8 gpm/ft2. 12/15 - 1/8: Bottom
washed with available makeup water |~*1 gprn) plus available
Scrubber inlet pH controlled at 8. 0 * 0. 2
Initially started as lime reliability verification test. Sub-
sequcnily. due to apparent reliability of the run. decision
was made (hat test continue as long-term reliability test.
Routine inspection on 11/7/74 showed system was generally
clean after 666 hours of operation with Clarifier only for
solids disposal. Run was terminated on 1/8/74 due to ID
fan vibration and rapidly increasing pressure drop across
demister. Sulfate based scale formed on most scrubber
walls *nd in slurry piping. Top of mist eliminator SO8".
plugged with solids that fell from outlet duct-work. Mist
eliminator top vanes heavily scaled ( 300 mils avg. ).
t-02 - 1A
J/1S/74
4/1/74
m
2S.OOO
*'. 7
(. 00/1200
r.0
7. $-0.1
12
Clarifier *< Filter
1. 02-). JS
91
2100-3800
R7-97
7. fa-8. 3
4. 9-S. 4
1-28
42-48
lt-5
9100
11.0-12.0
9
0. 61-0. 73
EHT.
Bottom washed! with available
makeup 00
11.0.12.0
<)
0. It.O. 72
EHT.
Bottom wished with ivailablr
makeup wat«r 1 — 5 Rpm' plus
available clarified liquor
(~2l j;pmt. Wash rate of 1
mtn ort/l-1/2 min off.
All noceles on 4 headers
zlej/header on top J headers.
Scrubber inlet pH controlled
at 8.0 t 0.2
Intended lonfj-lerm. Recir-
culate 1 We solids in attempt
to reduce degree of sulfale
saturation. EHT s«al«d.
Degree of sulfale saturation
was reduced, but runu-as
terminated due to scale «>0
mils avg. ) and solids buildup
nn the mist eliminator lop
van«a.
282
-------
Table 3 (contirvued)
SUMMARY OF LIME RELIABILITY TESTS
ON VENTURI/SPRAY TOWER SYSTEM
Run No.
Start-of-Run Dale
End-of- Run Dale
On St ream Hour*
C*i Rate. »cfm £ 3JO°F
Spray Tower G.s Vel. fp» @ I2S°F
V«MuH/Spray Tower
Liquor Rates, upm
Spray Tower L/C. gal/mef
Percent Solids Reci rculated
Solids Disposal Sy»(cm
Stolchiomelric Ratio, motes Ca
*dded/mcrte SO2 absorbed
Avg % Limestone Utilization. 10 Ox
moles SO? abs. /mole Ca added
Inlet SO 2 Concentration, ppm
Percent SO2 Removal
Scrubber Inlet pH Range
Scrubber Outlet pM Range
Percent Sulfur Oxidised
Loop Closure. % Solids Dischg.
Calculated 7« Sulfate Saturation in
Scrubber Intel Liquor @ 50°C
Dissolved Solids, ppm
Total AP Range. Excluding
Demisjter, in. H^O
Venturi A P. in. HjO
Demtster AP. in. M2O
Absorbent
Mist Eliminator
Scrubber Internals
System Changes Before Start
of Run
Method of Control
Run Philosophy
Results
<,0-i.lA
4/2t./74
7/1S/74
1828
2S.QOO
iV 7
min. <- 1001/1200
i.O
7. S-9. 0
17
Clarifier fc Filter
l.OJ-l, JO
88
2000-3800
70-92
7. 7.8.4
4. S-5.4
8-30
SO-60
130
11.600-1 J. 700
3. 3-3.8
Plug 100*, open
0.70-1.75
Lime slurried to 20 wt ^o with
scrubber downcomer.
makeup water (^S gprrO plus
available clarified liquor
mtn on/1 I/Z min off.
All nozzles on 4 headers
zlea/header on top 3 headefd.
Mist eliminator and outlet
provided with N*2 gas purge.
EHT overAow blanked. Lime
slurry makeup added to
JC rubber downcomer.
Scrubber inlet pH controlled
at 8. 0 t O.Z
Intended 2 wks. To obierve
sulflte oxidation and Degree of
BuUate saturation with lime
add'n to downcomer, minimum
slurry rate to ventui-I. sealed
EHT purged with NZ RA«. and
8% solids rectrculated.
Degree of aulfate laturation
w*i about 130%' Solid* trorn
outlet duct fell to top of ml at
tied due to heavy ecale (500
mils avg. } and solid § buildup
on mist eliminator.
t.OS-lA
7/ JI/74
S/f/74
Ml
2^.000
-. v
mm. (-1001/1200
-.0
8. 0-Q. 3
17
Cl*ri(ier A Filter
1.10-1.17
A8
2^00-HOO
73-81
8.8-9. 2
4.9-S. 1
12-28
48-S2
IIS
b. 000-7. 400
J.2-3.9
PluR 100r» oppn
0.80-0. 8S
Lime slurried to 20 wt«» «-'ith
maVeup water only ( -S spmL
Wash rate of 0.4 (ipm/ft' on
off.
ties/header on top 3 headers.
System cleaned.
Scrubber it. let pH controlled
at 9. O i 0. 2
Intended long-term. Control
at higher pH in attempt to
reduce JM\ftte oxidation and
thereby degree of suHate
jaturation. Wash mist elim-
inator with water only.
Run was terminated due to
scale formation (vp to 1 $0
milslon cop mist eliminator
t.Ot.-lA
S/7/T4
8/ 14/74
170
2S000
".7
mm. ( -1001/1200
•-0
7. 7. 4.0
1 7
CUrifirr
1. 10-1. 1">
8<)
2400- 1200
<.7-79
7. 8-8.2
'..O-S.Z
12-22
18-2)
120
1. 000-7. 000
J. f-1.7
Plug 100^ open
0. 78-0. 80
Lime slurricd to 20 wi**e wj^
scrubber dowrxcomer.
with IS ftpm (0. 1 Kpm/ft2)
raw water only. (Rate is
water).
All noitles on 4 headers
ties/header on (op 3 headers.
Scrubber inlet pH controlled
at 8. 0 i 0.2
ntendc-d short-term. Mist
eliminator wished continuous-
ly with T*W *H*T.&Mv Ut rate
greater than available makeup
u.-ater).
Run was terminated due to
scale formation (50 mils avg.l
on top mist eliminator vanes.
t.OS-lA
ft/21/74
Q/17/74
t,tO
2S. 000
r.. 7
'•00/1200
f-0
7. 7.0. 4
12
CUrifirr \ Filter
I.OJ-1. 17
Rft
2000- J tOO
7<-. QA
7 . Q - S . 4
4.8-1. 1
1 1-24
43 - S*
MO
7. S00-9.*00
11. ^-12.0
9
0. 7<.-0. IT
Limr ?lurric
-------
Table 3 (continued)
SUMMARY OF LIME RELIABILITY TESTS
ON VENTURI/SPRAY TOWER SYSTEM
Spray ToMi-rt-Cas V.-J. fps '» 1-T^'f
UiHf Spray Tow*-r
J.iqui>r Rates. p[>m
Spray Tower L/C. j-al/
Percent Solids Reci rcul.ii
Effluent Residence Time.
Solids Disposal Sysieni
Slmchiomrtric Ratio, moles Ca
Ivdfmitle 5O2 absorbed
K T, Limestone I'lili/aMon. tOOx
rnoJrs .SC2 abs, /rnolr Ca ftddrri
Inlel SC2 Concentration, ppn
Percent SO2
Vill
Scrvibbcr Inlel pH Ranpe
Scrubbrr Outlrl pll RanR
Percent Sulfur Oxidised
Loop Closure, *"? Solirijs Dischp.
CalculiH«-(| "*„ Sulfatf Saiuraiinn
Scrubber Inlgi Liquor fi SO°C
Total A P Kange. Ex
nisler. in. HgO
VeniuriAP. in. HgQ
DemislerAP.
Mist Eliminator
5crubtx-r Internals
System Changes Before 55tar
Of Run
Method of Control
Run Philosophy
Clarifirr '- Filn-r
i.in-n- slurried to 20
rt lake up u>at«r and ad
scrubber duwnfomtr
Bottom washed with makeup at
2, 7 gpm/ft2 for ""ft min cw ry
t hrs. Simultaneous lop wa $h
^-ith remaining makeup at
lozzle covering about U ft^.
Total makeup-v^ gpni avp.
All nozzles on 4 headers
sprayed downward. 7 noz-
iles/headcr on top 1 headers.
*j nn/.xles on bottom header.
Misi eliminator and outlet duct
cleaned. A single no/,zle in-
stalled to provide top wash for
one section of mist eliminator
and several holes drilled in
[he lop vanes of a second
ection.
Scrubber inlel pH controlled
at 8.0 * 0.2
Intended 2 wks. To observe
the effect of mist eltmin.Tior
top wash (on one section) and
the effect of increased resi-
dence time on HuH.ite satur.i-
ion.
Run terminated as planned.
Sulla I e saturation reduced to
IO^,. Mis! eliminator top
an** clean M-here top washed
284
-------
Table 4
AVERAGE SCRUBBER INLET LIQUOR COMPOSITIONS AND CALCULATED SULFATE
SATURATIONS FOR VENTURI/SPRAY TOWER LIME RELIABILITY RUNS
Run No.
..ON! A1"'
,,01-1 A""
t.JJ-IA
t.04- 1A
i.O*-IA
i.OH- I A
..On-lA
Pi-rceni Effluent Percent
Solids Residence Solids
Kfcirculaled Tiou*. min. Discharced <
7--I 12 .70. 2i.
7-9 12 42-S2
11. 1-U-. 0 12 4f.-S4
7. <;-«. 0 17 M)-i.O
«.«-". 2 17 4X-S2
7. =,-<>. I) 12 -U-iH
•<-'' 24 4i. -=.2
^ercent Scrubber
Sulfur Inlet pll 1~^~| 1
Jxiili/.ed Ranpe Ca MR
10- !0 7. 4-«. i. P'OO 140
10-30 7. 1.8. 1 MOO 200
12-22 7.8-8.2 2KSO 190
S-30 7. 7-S. 4 S200 400
12-2H H. H-'i. 2 2200 so
M-2-i 7. ,t C'alcuiat..,! l',.r.,-,,i
Sulfate Satu r.t i inn
K' SC ( SC., CC; CI' Ti-ta! at 1il"<:lrl""
HO "0 2000 20 2100 t400 1=.'!
230 '.0 2iOU 2S 4SOO 10.500 ISO
i!0 i.fl |i.70 S 4770 ii'icitj | tn
27'. ^0 2000 10 =i700 11.701) | -.0
IHO M> 14^0 ^ UJ70 Vi^i) 1J\
2 in 4-". 1'ijo 10 i
K3
00
Soli'ls disposal system: Clarifier and filler. pri'duct at SO"Cl. A solubility product for ("aSC.^ .'H,»C '•'' •'••* * "" ** ^ s
used (Radian Corporation. "A Theoretical Description of the J,ime»tone-
(a) Solids disposal system: Clarifier
(bl Solids disposal system; Clarifier and filler
inlet yas SC > concent ration of 2SOO ppm.
Injection Wcl Scrubbinu Process,' N'APCrA Keport. funi' 1. |
-------
SSi
. RUN ecu IA CONTINUED
S' 90 -
VEIMTum A SCHAV TOWfll
3. WO
3.000
2.bCC
2.000
1.500
^_
r^ ^ *
3 SCO
3.000
,.*»
J.ooo
1 ft>
960 1.000 1.0*0 1.080 t.tio 1.160 ».X» 1.140 t.WO i.3» '.iW 1.4OO
risr 7iMi. hou*t
I 6,*6 I WJ I WB I 6^ i t'10 t 6,'n I fe'U 1 t'*3 I 6/M I 6lb 1 6.16 I 6 >7 t 6/118 1 &')9 I 6-'7O t &-?l I 6.7? 1 b ;j 1 &?«
CALENDAR DAY
I'll '•'
1 i 8~ ...
Ill ,,
:il I
j|l .0
il =
16000
| IS, WO
^ 14.000
13000
cc
o
3 11,000
^ 1Q.OOQ
~ s.cco
m BODO
S
£ 7.000
x 6-°°°
o 5.0»
I '•KX>
2 '.i00
S 1.000
o
«
-
-^^*^ *~^^ — -^--— *— — -^^-^^-- "^*^v-^ .___ ^..^_ __^ — -\ _ ^ -
-
>^\ ^\ r\ r\
\ x^X ^ -A -^ \l \ J ' x^-/ ' "" "*" ^"
^x/^— % / V^V" ^' i
— » \^/ [NJ pu«ct ovtit
L OVER tur J
'O CALCIUM (0 *•) A CHLORIDE ICI-) « IESS THAN WO rt™ ARE NOT PlOrTl 0
• • *
* * *
•
-
.
A
- A * * A* A* * * • A
* AA A* AAAAAA^
*
00^° ° 0^><> o 00
O O4-' D O ^ 0
_ n G 0 fj
_oaODQDulJa n n o D D g
i i i . i i i t . i i
SO 1.000 1.0*0 1.080 l,l» 1.160 1.300 l.2« 1.3» 1.JJO t.3so 1,400 1.4
TEST TIME, howt
1 8/6 1 6/1 1 6/8 1 8/9 | 0/10 1 &/1t 1 0/12 1 6/13 1 4M4 1 *"5 1 U1B 1 &/17 1 6^18 t 6/19 1 6/IO 1 6.T1 1 &/?2 1 6/73 1 &/?4 I
i J
1 ?
1 1
t,0
30
20
10
0
15.000
14.000
13.000
11.000
W.OOO
9.000
8.000
7,000
6.000
S.OOO
4.000
3.000
7.000
1.000
40
Gas Raic= 25.000 «fm@ 330 °F
Liquor Rate to Venturt B minimum (100 gpm)
Liquor Rate to Spray Tower = 1200 gpm
Venlim V/G • & gal/met
Spray Tower L/C « BOgal/mcf
Spray Tower Gai Velocity • 6.7 fi/rtc
No. ol Spray Headetj = 4
EHT (Sealed with N, Purge) Residence
Time = 17 rn'in
Percent Solidt RedrCulated = 7.&-9.0wt*i
Venturi flu,; Position 1QQ% Open
Total Pressure Drop, Excluding Demiiter = 3.3-3.8 in
SciubbeilnM Liquor Temptraiuie - ;26-l32°F
Liquid Conduciivity = 10.000-24.000 u mhoi/cm
Discharge (Clarifier and Filter) Solidj
Concentration = 50-60 wl %
Lime addition to Scrubber Downcomer
Figure 8. Operating data for Venturi/Spray Tower Run 604-1A
286
-------
4.1.1 Venturi/Spray Tower Run 601-1A
On October 9, 1973, the initial long-term reliability test (Kun bUl-I
was begun on the venturi/spray tower system. The major test condi
tions selected were (see Table 3):
Spray Tower Gas Velocity 6.7 ft/sec
Venturi Liquid-to-Gas Ratio 30 gal/mcf
Spray Tower Liquid-to-Gas Ratio 60 gal/mcf
Percent Solids Recircxilated 8
Effluent Residence Time 12 min.
Scrubber Inlet Slurry pH (controlled) 8
The pH control level for the scrubber inlet slurry was chc-;en based
on results of lime testing at the EPA pilot facility in Research Triangle
Park, which indicated reasonable lime utilization and SO2 removal at
that level.
During the entire test, the chevron mist eliminator was washed on the
underside both continuously (at a rate of 0. 8 gpm/ft ) and intermittently
(at a rate of 1 gpm/ft ) with the available raw water makeup plus
clarified liquor. Also, the mist eliminator was washed on the topside
once per week with fresh water for 5 minute durations during the last
four weeks of the test.
Throughout the initial 666 hour portion of the run the clarifier was used
as the final dewatering device and the average percent solids discharged
Twelve minutes was the minimum residence time obtainable in the
effluent hold tank for this run.
287
-------
was 23 percent, which resulted in slightly open liquor loop operation.
An inspection after 666 hours on November 7 showed that the mist
eliminator bottom vanes were clean while the top vanes were about 5
percent plugged with soft solids. The scrubbers were generally clean,
with only a thin layer of scale (20 mils) on the upper half of the spray
tower.
In order to operate under closed liquor loop conditions, the test was
continued with the clarifier plus vacuum rotary drum filter (or centrifuge)
in series. Problems with the filter cloth and mechanical difficulties
with the centrifuge, however, resulted in intermittent operation of these
pieces of equipment from November 7 to December 15. For the final
575 hours of testing (from December 15 through January 8), the filter
operated satisfactorily and the average percent solids discharged
increased from 23 to 47 percent and the total dissolved solids increased
from about 6500 to 10, 500 ppm (see Table 4).
Run 601-1A was terminated after a total of 2153 hours (3 months) of
operation, due to rapidly increasing pressure drop across the mist
eliminator. An inspection showed that the top of the mist eliminator
was covered with solids that fell from the duct-work above, plugging
about 80 percent of the surface. In addition, about 1/8 to 1/2 inch
thick scale formed in the middle and top vanes and about 5 percent of
the bottom (inlet) vanes contained heavy (1/4 to 3/4 inch) scale. It
is hypothesized that the heavy accumulation of solids in the outlet duct
occurred when slurry droplet entrainment through the chevron mist
eliminator wetted the duct walls during frequent reheater flame-outs.
The venturi/spray tower system clarifier is undersized.
288
-------
Heavy scale had accumulated on the wall of the ypray tower below the
bottom header while the middle section of the vessel was only slightly
scaled. The venturi wall above the throat was clean, but scattered
scale covered the plug. A light scale covered the venturi wall between
the throat and flooded elbow and the entire flooded elbow was covered
•with heavy scale. Also, the recirculating piping was scaled, with
heaviest scale deposits in the pump suction areas. Scale also formed
on the impeller eyes and tips of all the recirculating slurry pumps and
on the lining of the casing of some of the pumps. It should be noted that
scale formation in the scrubbers, circulating slxirry piping and pumps
did not prevent continual operation of the system or necessitate termina-
tion of the run.
The formation of most of the scale on the scrubber walls and piping
occurred between December 15 and January 8, •when the system was
operating under closed liquor loop conditions. The calculated average
sulfate saturation of the scrubber inlet liquor during this period was
180 percent. This is contrasted with a calculated sulfate saturation of
150 percent for the initial 666 hours of testing, during open liquor loop
operation (see Table 4).
An analysis of the test during the final 575 hours of closed liquor loop
operation shows that the calculated inlet liquor sulfate saturations
are strong functions of inlet gas SC>2 concentration (or SC>2 absorption
rate). For example, during test periods where the inlet gas SC>2 con-
centrations averaged 3200 and 2500 pprn, the calculated sulfate satura-
tions averaged 225 and 140 percent, respectively.
289
-------
4.1.2 Venturi/Spray Tower Rvin 602-1A
In preparation for the second lime reliability run, the system was
partially cleaned chemically (Na2 CO ^ /suga r /limestone/fly ash solution),
followed by mechanical cleaning. All the stainless steel spray nozzles
were replaced with identical, full cone, stellite-tipped Bete nozzles
(ST 48-FCN) and the original Hauck reheater, which was troublesome
throughout the previous run, was replaced with an external combustion
reheater (by Bloom Engineering Co. ).
Results from the EPA pilot facility in Research Triangle Park had
indicated that sealing the effluent hold tank could reduce the sulfite
oxidation and, thereby, the degree of sulfate saturation of the scrubbing
slurry. In an attempt to duplicate this mode of operation, the effluent
hold tank on the venturi/spray tower system was sealed prior to the
start of Run 602-1A.
Run 602-1A was started on March 15, 1974. The mist eliminator
was washed on the underside with the available raw water makeup and
clarified liquor (in an approximate one to seven ratio) at a rate of 1
gpm/fl , in an intermittent operation (4 minutes on/1 minute off cycle).
After 393 operating hours the run was terminated, when it became
apparent that sealing the effluent hold tank in this facility was not
effective in reducing the degree of sulfate saturation of the scrubber
inlet liquor. Although the sulfite oxidation had dropped somewhat (from
about 20 to 17 percent), the calculated sulfate saturation at the termina-
tion of the test was 190 percent.
290
-------
An inspection showed that a uniform (aboxit 1/8 inch) scale covered most
of the top and middle vanes of the mist eliminator. The bottom vanes
were relatively clean. The walls-of the spray tower were covered with
a thin scale. The wall of the venturi below the plxig was generally clean
and no scale was deposited in the flooded elbow.
4.1.3 Venturi/Spray Tower Run 603-1A
Run 603-1A was started on April 2, 1974, after the mist eliminator was
cleaned. The solids content of the reci rcxilaling slurry was increased
from 8 to 15 percent in order to reduce the degree of scrubber liquor
sulfate saturation. The mist eliminator was washed on the underside
with the available raw water makexip and clarified liquor (in an approx-
imate one to four.ratio) at a rate of 1 gpm/fl , in an intermittent oper-
ation (1 1/2 minutes on/1 1/2 minutes off).
After 395 hours of operation the run was terminated. dxie to increasing
pressure drop across (he mist eliminator (from 0. 55 inches 1T2O at
the start of the run to 0.70 inches IIoO). The mist eliminator had
50-70 mils of scale on the outlet edge and soft solids along the inlet
edge of the top vanes. The middle and bottom vanes had light deposits
of scale and soft solids. The venturi scrubber contained about the
same amount of scale as before the start of the run, while there ap-
peared to be slightly less scale on the spray tower walls and in the
suction piping of the slurry pumps. The calcxilated sulfale saturation
of the scrubber inlet liquor was 135 percent. This was a significant
drop from the prcvioxis run's value of 180 percent saturation at 8
percent solids recircxilated.
291
-------
4.1.4 Venturi/Spray Tower Run 604-1A
In order to decrease the quantity of soft solids buildup on the mist
eliminator which had caused termination of the previous test, it was
decided to drop back the percent solids recirculated from 1 5 to 8 per-
cent for this run. To reduce the scrubber slurry sulfate saturation
from the expected level of 180 percent (see Run 601-1A) (1) lime was
added to the scrubber downcomer (vs. the effluent hold tank), (2) the
cffruent hold tank was purged (blanketed) with N2, and (3) the venturi
scrubber was used only for gas cooling (venturi plug 100 percent open
with 100 gpm liqxior flow rate).
It was theorized that the addition of lime to the downcomer quickly
increases the pll of the scrubber outlet liquor, thereby precipitating
calcium sulfite before the sulfite can be oxidized to sulfale in the liquid
phase (the solubility of calcium sulfite is lower af higher pll). As
mentioned previously, a reduction of liquor phase sulfite oxidation
resxilts in a reduction in sulfate saturation. It was further theorized
that reducing the liquor flow to the venluri with a minimum flue gas
pressxire drop would minimize liquid atomization, oxygen absorption
and, consequently, oxidation.
Run 604-1A was begun on April 26, 1974, after the mist eliminator was
cleaned. The mist eliminator was washed on the underside with the
available raw water makeup and clarified liquor (at an approximate
one to seven ratio) at a. rate of 1 gpm/ft , in an intermittent operation
(3 1/2 minutes on/1 1/2 minutes off). The efflxient residence time was
increased to 17 minutes, since this was the minimum time obtainable
292
-------
at the reduced total liquor rate of 1300 gpm (the venturi liquor rate had
been reduced from 600 to 100 gpm).
The run was terminated after 1828 on-stream hours due to (1) increas-
ing pressure drop across the mist eliminator (from 0. 62 inches H->O
at start of run to 1. 75 inches t^O), and (2) gas flow control problems
caused by solids accumulation on the ID fan inlet dampers. The top
of the mist eliminator was 75 percent plugged with 150-200 pounds of
loose solids which had fallen from the outlet duct. There was about
1/2 inch of scale on the top vanes and 1/4 inch of scale on the bottom
vanes. Up to 2 inches of solids had accumulated in the outlet gas duct
and six of the 14 damper blades were covered with 1-2 inches of solids.
The sulfite oxidation averaged from about 14 percent with NT2 purge over
the efflxient hold tank to about 19 percent without N£ purge (from May 29
to June 5). The calculated average sulfate saturation of the scrubber
inlet liquor, however, was 130 percent. This was a significant drop
from the saturation level of 180 percent during the closed liqxior loop
portion of Run 601-1A.
As in Run 601-1A, an analysis of the data showed that the calculated
inlet liquor sulfate saturation was a strong function of the SO2 absorp-
tion rate. For example, during test periods where the inlet gas
Oxygen concentrations in the effluent hold tank in the vicinity of the gas/
liquid interface ranged from 15 to 20 percent, as compared with 21 per-
cent for air. This suggests that sealing the tank and N£ purging were
not effective in reducing oxygen absorption in the effluent hold tank.
293
-------
concentrations averaged 2700 and 2900 ppm, the calculated sulfate
saturations averaged 120 and 150 percent, respectively.
Operating data for Run 604-1A, from June 6 to June 24, 1974, has
been graphically presented in Figure 8.
4. 1. 5 Venturi/Spray Tower Run 605- 1A
Run 605-1A was started on July 31, 1974, after the mist eliminator
was cleaned. The scrubber inlet liquor pH \vas raised from 8 to 9,
in order to observe the effect of pH on sulfate saturation. The mist
eliminator was washed intermittently with raw water makeup at a rate
o
of 0.4 gpm/ft on a 1 minute on/4 mirmte off cycle. As before, lime
was added to the downcomer and the venturi liquor rate was set at 100
gpm with the venturi plug 100 percent open.
An inspection after 141 operating hours showed relatively heavy (up
to 150 mils) scale on the top mist eliminator vanes and the run was
terminated. As with the previously observed patterns, there was a
moderate buildup of scale on the middle vanes (50 mils) and only a
light dust film on the bottom vanes. No additional scale had formed
in the spray tower during the run. The calculated average percent
sulfate saturation of the scrubber inlet liquor was 115 percent. This
can be compared with a level of 130 percent for Run 604-1 A, which
had a controlled scrubber inlet liquor pll of 8. 0 (see Table 4).
4.1.6 Venturi/Spray Tower Run 606-1A
Run 606-1A was started on Axigust 7, 1974, after the mist eliminator
had been cleaned. The objective of the run, which was intended to be
294
-------
of short duration, was to investigate scale formation on the mist elim-
inator vanes with continuous washing on the mist eliminator underside
with raw water at a rate of 0. 3 gpm/ft . This raw water rate (15 gpm)
is approximately three times the quantity normally used during closed
liquor loop operation, and, therefore, the test was made under open
liquor loop conditions (about 21 percent solids discharged). The scrub-
ber inlet liquor pH was reduced from 9 to 8 for this test.
The run was terminated after 170 on-stream hours. About 95 percent
of the trailing (top) surface of the mist eliminator vanes was covered
with 30-70 mils of scale. As usual, there was light soft solids accumula-
tion on the middle vanes and the bottom vanes were relatively clean. It
should be noted that the scale growth rate on the top vanes (~50 mils/
\veek) was about the same for this test as for all of the previous tests,
in which mixture of clarified liquor plus raw water were used for under-
side washing.
4.1.7 Venturi/Spray Tower Run 608-1A
Run 608-1A was started on August 21, 1974, after the mist eliminator
had been cleaned. The main objective of the test was to investigate
scale formation on the mist eliminator vanes with intermittent under-
side washing at high pressure (nO psig), using the available raw water
makeup at a rate of 3 gpm/ft for about 9 minutes every 4 hours. Also,
the venturi scrubber was put back in service (at 600 gpm and 9 inches
H^O pressure drop), in order to investigate the effect of venturi operation
(Runs 608-1A vs. 604-1A) and lime addition in the downcomer (Runs
608-1A vs 601-1A) on scrubber liquor sulfate saturation.
295
-------
The run was terminated after 610 hours of operation due to increasing
pressure drop across the mist eliminator (from 0. 75 inches H2O at
start of run to 0. 95 inches l^O). About 80 percent of the trailing
surface of the top vanes was covered with scale, averaging 88 mils
in thickness and resulting in approximately 3 percent pluggage of the
free flow area. The middle and bottom vanes were, as usual, essen-
tially clean (3 mils of dust coverage). Note that, the average scale
growth rate on the top vanes for this test (24 mils/week) was essen-
tially half the average scale growth of the previous four tests.
The condition of the scale on the walls of the spray tower remained
essentially unchanged, but the spiral tips of most of the spray nozzles
were covered with 1/4-1/2 inch scale (scale "whiskers").
The calculated average sulfate saturation of the scrubber inlet liquor
was 130 percent (see Table 4), This is the same as the calculated
saturation for Run 604-1A and about 50 percent lower than the calculated
saturation for Run 601-1A. Therefore, it may be concluded that, for
these run conditions, venturi operation has little effect on sulfate
saturation and the addition of lime to the downcomer (vs. lime to the
effluent hold tank) significantly reduces the sulfate saturation of the
process slurry.
4.1.8 Venturi/Spray Tower Run 609- 1A
Run 609-1A was started on September 20, 1974, after the mist eliminator
had been cleaned. The primary objective of the test was to evaluate
the effect of topside mist eliminator washing on the formation of scale
296
-------
on the top mist eliminator vanes. The entire underside of the mist
9
eliminator and a 14 ft area on the topside was washed at high pres-
sure (60 psig) with makeup water at a rate of 2. 7 gpm/ft for the under-
side and 1. 0 gpm/ft^ for the topside for about 8 minutes every 4 hours.
Only a small section of the topside was washed because it was felt that
liquid carryover from top sprays could possibly cause reheater over-
loading and fan problems. Also the effluent residence time was in-
creased from 12 to 24 minutes, for this test, in order to determine
the effect of residence time on scrubber liquor sulfate saturation
(Run 609- 1A vs. 608-1A).
Run 609-1A was intended to be of short duration and was terminated
after 278 hours of operation. The lopside spray had boon successfxil
in drastically reducing the scale buildup on the lop mist eliminator
vanes. The washed area was essentially clean, with less than 1 mil
of solids accumulation, compared with an average of 40 mils scale
buildup on the rest of the topside stir faces.
The calculated average sulfate saturation of the scrubber inlet liquor
was 110 percent for the test (sec Table 4). This value is about 20 per-
cent lower than the calculated saturation for Run (>08-1A. If may be
concluded, therefore, that the effect of effluent residence lime on sulfalc
saturation is similar for lime systems as for limestone systems (see
Figure 6).
These problems could be reduced by a-sequential sectional wash of the
topside of the mist eliminator or by using a second mist eliminator
downstream of the topside sprays.
297
-------
4.2 CONCLUSIONS
4.2. 1 Scrubber Internals Operability
The lime and limestone reliability tests have shown that scrubber
internals can be kept relatively free of scale if the sulfate (gypsum)
saturation of the scrubber liquor is kept below about 135 percent.
As with limestone, this can be accomplished \vith increased percent
solids recirculated (Run 603-1A vs. 601-1A) and/or with increased
effluent residence time (Run 609-1A vs. 608-1A). Moreover, the
lime tests have shown that adding the lime to the scrubber downcomer
(which corresponds to a small residence time tank in series with the
larger effluent hold tank) can substantially reduce the sulfate saturation
(Run 608-1A vs. 601-1A), allowing for operation at reduced percent
solids and/or effluent residence time. Also, operating at higher
scrubber inlet liquor pH appears to reduce the scrubber liquor sulfate
saturation (Run 605-1A vs. 604-1A).
The lime tests have also shown that the sulfate saturation of the scrub-
ber inlet liquor is a strong function of the inlet gas SO;? concentration
(SO£ absorption rate). The data has indicated that a 100 ppm increase
in SO2 inlet concentration corresponds, roughly, to a 13 percent in-
crease in sulfate saturation, at the run conditions tested.
It should be noted that the 28 full cone stellite-tipped Bete nozzles in
the spray tower have shown no signs of measurable erosion after
approximately 4000 hours in slurry service.
298
-------
4. 2. 2 Mist Eliminator Reliability
The most significant reliability problem encountered, to date, during
the venturi/spray tower lime reliability tests has been associated with
scaling and/or plugging of the topmost vanes of the chevron mist elim-
inator (see Figures 1 and 3).
Runs 601-1A through 608-1A showed that, the scale formation rate on
the top mist eliminator vanes (25-50 mils/week) was relatively unaf-
fected by the wash cycle, wash rate or quality (sulfate saturation) of
the wash liquor. Run 608-1A showed that intermittent washing with
high pressure (60 psig) raw water at a rate of 3 gpm/ft for 9 minutes
every 4 hours, gave results that were at least as good as results with
continuous washing with relatively low pressure raw water at 0. 3 gpm/ft
(Run 606-1A). Intermittent washing can be especially important if the
availability of raw water makeup is restricted (e. g. , at high percent
solids discharged and/or high pximp seal water usage).
Run 609-1A showed that intermittent topside and bottomside washing
with high pressure raw water would most likely eliminate the scale
accumulation problem on the top vanes of the existing mist eliminator
(see Figure 3) and allow for long-term reliability of the system at the
run conditions tested (see Table 3). However, the problems caused
by entrainment of topside wash water will have to be evaluated.
The reliability of the intermittent topside and bottomside washing with
raw water has been further substantiated in a recently completed 253
hour test (Run 610-1A). The scrubber and mist elimination systems
299
-------
\vere not cleaned before the test. After a total of 531 operating hours
(Runs 609-1A and nlO-lA), the \vashed area of the top mist eliininator
vanes was still clean, compared with, an average of 34 mils scale buildup
on the rest of the topside vane surfaces.
It is planned to test a sloped (cone-shaped) closed-vane chevron mist
eliminator in the spray tower in the near future. It is expected that
the sloped mist eliminator has better draining characteristics than a
horizontal mist eliminator.
300
-------
Section S
PARTICULATE REMOVAL TEST RESULTS
The overall particxilate removal efficiencies measured during the
reliability testing on the TCA and venturi/spray tower systems are
shown in Tables 5 and 6, respectively. For these tests, a modified
EPA particulate train (manxifactxired by Aerotherm/Acurex Corporation)
was used to measure mass loading at the scrubber inlets and outlets.
As can be seen from the tables, the measured removal efficiencies
are all greater than 99 percent for the tests conducted.
In May 1973, the EPA measured overall parliculate removal efficiency
and particulate removal efficiency as a function of particle size for
the Shawnee TCA scrubber during limestone reliability verification
testing (see Section 3. 3 in Reference 2). The EPA measurements
for overall efficiencies are in good agreement with the results shown
in Table 5. The mass mean diameter of the inlet solids, as determined
by EPA, was approximately 23 microns.
The particxilate removal efficiencies shown in Tables 5 and 6 appear
to be higher than the efficiencies predicted from "impaction theory. "
These improved efficiencies could be due to condensation of water
vapor in the flue gas on the solid particles. In order to verify the
301
-------
Table 5
OVERALL PARTICULATE REMOVAL IN TCA SCRUBBER FOR
LIMESTONE RELIABILITY TESTS
Run No.
531-ZA
531-2A
532-2A
535-2A
Date
5/15
6/12
7/24
9/25
Gas
Rate,
acfm @ 300°F
20, 500
20, 500
20, 500
20. 500
Liquor
Rate,
gpm
1200
1200
1700
1200
Pressure
Drop,(a)
in. H2O
6.7
10.0
6. 7
6. 2
Grain Loading,
grains /scf
Inlet
2.82
3.47
3. 33
2.88
Outlet
0.029
0. 010
0.024
0.016(b)
Percent
Removal
99.0
99.7
99.3
99.4
(a} Including mist eliminator and Koch tray.
(b) Outlet sample taken on 9/27, two days after inlet sample.
-------
Table 6
OVERALL PARTICULATE REMOVAL IN VENTURI/SPRAY TOWER
FOR LIME RELIABILITY TESTS
Run No.
Date
Gas
Rate,
acfm @ 330°F
Liquor Rate,
gPm
Venturi | Spray Tower
Pressure Drop,
in. H2O
Venturi
604-1A 6/26 25,000 100 1200 1.9
608-1A 9/11 25,000 600 1200 9.0
610-1A 10/10 25,000 600 1200 9.0
Spray Tower
Grain Loading,
grains/ scf
Inlet
3.0 2.52
2.2 2.28
3.2 2.72
Outlet
Percent
Removal
0.024 99.1
0.023 99.0
0.021 99.2
CO
o
00
(a) Including mist eliminator.
-------
partirulate removal results obtained, to date, and identify the causes
for the observed removals, EPA has planned a new test series for the
determination of size and overall removal efficiencies in the TCA
and venturi /spray tower systems. The test series is to be conducted
by an independent party in the near future.
304
-------
Section 6
REFERENCES
1. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Phase II
and III Final Report, EPA Report, December 1974 (to be issued).
2. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Limestone
Wet Scrubbing Test Results. EPA Rcporl 650/2-74-010, January 1974.
3. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Sodium
Carbonate and Limestone Test Results, EPA Report 650/2-73-013,
August 1973.
4. Bechtel Corporation, Test Manual for Advanced Test Program,
EPA Report, September 1974.
5. P. S. Lowell, "Use of Chemical Analysis and Solution Equilibria
in Predicting Sulfate/Sulfite Scaling Potential,- presented at
Second International Lime/Limestone Wet Scrubbing Symposium,
New Orleans, Louisiana, November 8-12, 1971.
6. R. H. Borgwardt, "EPA/RTP Pilot Studies Related to Unsaturated
Operation of Lime and Limestone Scrubbers," presented at Symposium
on Flue Gas Desulfurization, Atlanta, Georgia, November 4-7, 1974.
7. Radian Corporation, A Theoretical Description of the Limestone-
Injection Wet Scrubbing Process, NAPCA (APTIC No. 22709 and
25446) Report, June 9, 1970.
305
-------
306
-------
OPERATIONAL STATUS AND PERFORMANCE
OF THE
ARIZONA PUBLIC SERVICE COMPANY
FLUE GAS DESULFURIZATION SYSTEM
AT THE CHOLLA STATION
By
Lyman K. Mundth
Arizona Public Service Company
November 4, 1974
307
-------
308
-------
OPERATIONAL STATUS AND PERFORMANCE
OF THE
ARIZONA PUBLIC SERVICE COMPANY
FLUE GAS DESULFURIZATLON SYSTEM
AT THE CHOLLA STATION
My name is Lyman Mundth and I am Vice President of Power
Production and Operations for Arizona Public Service Company. My
discussion will cover our experiences with wet scrubber systems at
our Cholla Power Plant near Joseph City, Arizona. The Cholla Plant
currently has a single coal fired unit rated at 115 MW which has been
operating since 1962. The unit burns coal from the Gallup, N.M.
area and this coal contains approximately 10 percent ash and .5 per-
cent sulfur.
We operate this unit base loaded and have historically had
trouble free operation, with a capacity factor of near 90 percent. The
original capital cost of the installation in 1962 was approximately $20
million. Mechanical collectors designed to remove 80 percent of the
fly ash provided the emissions abatement on that installation. We
recently retrofitted the unit with a Research-Cottrell sulfur dioxide/fly
ash removal process and a by-pass system at a capital cost of $6.3
million.
Construction of the Research-Cottrell system was completed in
December, 1973, one year behind the scheduled date of December 3,
1972. The system, consisting of two parallel gas cleaning trains,
began commercial operation in mid-December, 1973, and has operated
to the present with only short outages. Each gas cleaning train con-
sists of a Flooded Disc Scrubber, a wetted film absorber, mist elimi-
nators, and reheater. The limestone and fly ash slurry handling systems
are common to both trains.
For those of you who picked up a copy of my talk , you will
notice that Figure 1 illustrates a single train flow diagram. Flue gas
containing SO2 and remaining fly ash flows from the outlet of the
mechanical collectors to the inlet of the wet scrubber system. A forced
draft booster fan forces the gas into the flooded disc scrubber where
the fly ash and a portion of the SO2 are removed by a recirculating
limestone/fly ash slurry. This slurry enters into the flooded disc
scrubber tangentially above the throat and at the disc. Particulate
309
-------
CHOLLA UNIT 1 WET SCRUBBER SYSTEM
LO
t-*
O
TO
ASH DISPOSAL
POND
SECONDARY
MIST ELIMINATOR
FDS SLURRY TANK
ABSORBER TOWER
FEED TANK
-------
emissions have been tested and have been found to average 0.026
pounds per million BTU , well within state and federal emissions
regulations.
About 70 to 85 percent of the sulfur dioxide passes through
the flooded disc scrubber into a cyclonic mist eliminator where the
captured ash and entrained slurry are removed. The flue gas then
passes into a packed tower which has 2 to 5 feet of Munters packing,
a wetted film packing. A limestone slurry recirculates through the
absorber packing and absorbers SC>2. The flue gas and slurry are
contacted countercurrently. To minimize scaling, a high liquid togas
ratio and a moderate reaction products concentration are maintained.
Total SC>2 removal in the train containing packing is expected to be
around 80 percent. However, no comprehensive tests on SC>2 removal
have been conducted to verify this removal level.
At this point, I should explain that, due to the uncertainties
involved in SO2 scrubbers, only one of the trains actually contains
packing material.
A two-stage mist eliminator follows the absorber. The first
stage is an impingement vane type which is flushed intermittently with
water to avoid both supersaturation and encrustation on the surfaces.
The second stage is also an impingement vane type.
Before entering the stack a steam coil reheater constructed of
316 stainless steel heats the cleaned flue gas approximately 40° F.
The slurry handling system consists of a limestone storage silo
and absorbtion tower slurry, flooded disc and limestone mix tanks.
The system is not equipped with any special sludge and fly ash
disposal equipment. Sludge and fly ash are pumped via temporary surge
tanks to pre-existing ash disposal ponds. However, we do have plans
to provide sludge disposal capacity for this system in the new units
currently under construction at the plant.
Although APS has operated the scrubbers since completion of
construction in December of 1973, many startup problems have prevented
us from final acceptance of these units. We had scheduled acceptance
tests in February of 1974. We have not run these tests due to mod-
ifications required on the system. Some modifications have been rela-
tively minor while others have required unit outages to accomplish. We
expect acceptance tests to be completed before the end of 1974.
311
-------
Major problems encountered to date can be classified into three
categories. They are scaling and plugging, corrosion and mechanical.
We have found scaling in the scrubber vessel, in piping and in the
mist eliminator. Plugging occurs in the nozzles to the flooded disc
and the mist eliminator. Corrosion has been experienced in the re-
heater and breeching expansion joints. We have encountered mechani-
cal problems in the flooded disc control system, by-pass damper and
the reheater.
I would like to discuss some of these problems briefly. First,
let's consider our mechanical problems. On initial startup of the unit,
vibration in the reheat sections limited operation of the scrubbers to
such an extent that commercial operation had to be delayed until a
solution could be found. Research-Cottrell hired an outside consul-
tant to look at the vibration problem and to determine a solution. The
installation of baffles was recommended because Research-Cottrell's
consultant felt that improper flue gas distribution was causing the
vibration problem. This fix initially appeared to have solved the
problem. Recently, however, due to other fixes, mainly on the leak-
ing by-pass control dampers, vibration is re-occurring at full load,
but is reduced considerably in frequency. The reheater vibration is
not considered to be a major problem at this time, but we are working
toward a final solution.
We have also had problems with the by-pass damper operation.
The scrubber system is equipped with a by-pass system so that the
boiler may continue to operate should any major failures occur in the
scrubber system. Leakage due to sluggish operation and temperature
caused distortions of the by-pass dampers have reduced the effective-
ness of the scrubber system. Small amounts of untreated flue gas
pass through the closed damper and by-pass the scrubber system.
Although I have previously mentioned that the system has been tested
and found to control particulate to a level of 0.026 pounds per million
BTU , we are unable to consistently achieve this level due to improper
operation of the by-pass damper. However, we feel the scrubbers
themselves have been attaining their designed particulate removal effi-
ciency.
Since the entire installation was a retrofit, the damper system
was installed in the existing breeching well in advance of actual
scrubber operation. Fly ash build up occurred and prolonged exposure
to high temperatures caused excessive expansion and distortion. The
boiler had to be shut down to trim, clean and align damper louvers.
This did not completely solve the problem. Research-Cottrell then
reduced the pressure drop required in the scrubber system in order to
312
-------
create a recirculating effect such that a small amount of treated flue
gas would actually flow backward through the damper and thereby
eliminate leakage of untreated flue gas. Although no prolonged oper-
ation or testing has occurred, we hope this solves the problem.
A third mechanical problem involves the position control of
the flooded disc. The Research-Cottrell Flooded Disc Scrubber main-
tains a constant pressure drop at all loads by controlling the position
of the flooded disc. Some minor problems were experienced in this
area. Initially the disc kept oscillating between positions. Build-
up around the disc shaft caused binding and improper operation. There
was also a problem of control points being set too near control limits.
Fortunately, this particular problem was easily resolved by Research-
Cottrell.
Other mechanical problems which bear mentioning are leaks in
the booster fan which have been experienced due to improper welding
and erosion of pump impellers and scrubber vessels. Generally,
mechanical problems have been shakedown in nature and have not been
totally unexpected.
Next I would like to cover corrosion problems. We have had
corrosion problems in the reheater due to condensation in the duct
work leading to the reheater. Initially the duct work was uninsulated
and in cold weather acid condensation occurred. The acid run-off
caused necking of tubes at the reheater sheet and subsequently tubes
have failed. In order to solve the problem, Research-Cottrell has
insulated the breeching to minimize or eliminate the condensation. We
are not sure at this point whether this has solved the problem because
no cold weather has been experienced since the insulating job.
We have observed pitting in other tube areas, however, we do
not know whether this is due to the manufacturing process or to chem-
ical attack. Only one small leak has been observed other than those
clue to acid run-off and its cause is unknown.
Because of the condensation in the breeching, failures have
also been experienced in the expansion joints. We have just recently
replaced all of the metal expansion joints with a rubberized fabric
type. Very little operating experience is available with the new joints.
Other corrosion problems which have occurred have been due to
improper application of protective coatings and were easily repaired.
313
-------
It is our belief that plugging in parts of the scrubber system
has occurred due to excessive scaling caused either by settling or
high concentration of solids. Chemical control on the system has
been good.
Material has accumulated and partially plugged the first stage
demister, but this material has been soft and easily removed by wash-
ing with high pressure water. No adverse operating effects have been
seem due to excessive plugging, however, Research-Cottrell has re-
designed the demister wash system for better coverage and frequency
of washing. It now appears from a short operating experience that
only minimal buildup occurs.
Piping has scaled in the tangential nozzle area of the flooded
disc scrubber which in turn has caused nozzles to plug. The nozzles
are designed in such a way that they can be cleaned while the system
is in operation. The plugging is due mainly to settling out of solids
in the piping to the nozzles and the formation of a scale olug. Re-
search-Cottrell has re-designed this system with increased velocities,
however, the problem has not been totally eliminated.
Problems in most cf the areas discussed appear to be those
which will be experienced in most scrubber systems . These will un-
doubtedly require a good maintenance effort.
Scrubbers appear to require a great deal more maintenance than
other equipment in general use by utilities. We expect that we will
have additional maintenance on nozzles, scale removal, valve wear,
pump packing and impellers and process lines. Although the system
shakedown is continuing, a tremendous effort has already been put
forth by both APS and Research-Cottrell during the time that the scrub-
bers have been in operation. We hope that the same type of effort
will not be required in normal operation.
We have tried to separate the shakedown effort from the main-
tenance effort in order to project our maintenance requirements for the
scrubbers. Again, for those of you who have copies, Figure 2 illus-
trates manhour requirements for experienced scrubber maintenance. The
average daily manhour requirement through August of this year has been
about 30 hours. This constitutes 37 1/2 percent of the available
maintenance manpower at this plant. We expect that it will be neces-
sary to increase our maintenance force by approximately 50 percent to
avoid sacrificing other routine and preventive plant maintenance pro-
grams and total unit reliability.
314
-------
CHOUA UNIT 1 - SCRUBBER MAINTENANCE
Figure 2
JAN FEB MAR APR MAY JUN JUL AUG
315
-------
In addition, one auxiliary operator per shift is required for
scrubber operation. Auxiliaries for the scrubbers include 2.8 MW for
electricity and 18,000 pounds per hour of steam for reheat. Limestone
is consumed at the rate of 15 tons per- day. As indicated previously
the system currently does not include waste disposal.
Table 1 is a summary of operating costs for the scrubber at
the bus bar. Based on these costs and on our 1973 operating costs,
it is estimated that the scrubber will add approximately 15.9 percent
to the cost at the bus bar.
TABLE 1
CHOLLA SCRUBBER OPERATING COSTS
(ESTIMATED)
Mills/KWH
Auxiliaries (Steam and Electrical) 0.2
Operating Labor 0.09
Maintenance Labor 0.09
Parts and Materials 0.07
Limestone 0.15
TOTAL 0.6 mills/KWH
Total Cost 1973 FPC Form 1 3.78 millsAWH
With regard to system availability, scrubber A which contains
packing in the absorber tower has operated for a total of 4,778 hours
vs a possible 5,064 hours. Scrubber B has operated a total of 4,401
hours vs a possible 5,064 hours. These numbers do not accurately
represent the scrubber availability since a great deal of time has been
spent in resolving the problem areas in order to minimize down time.
During times when leaks have occurred in the reheater or nozzles have
plugged, the unit continued to operate so that individual items of
equipment are not necessarily represented by the above figures. No
hard availability data has been kept and it is unfair to suggest that
one time malfunctions which have been corrected by design changes
be included in any assessment of true system availability in the future.
While the scrubbing system at Cholla appears to be successful,
it does not necessarily reflect the experiences of others on large util-
ity boilers. On very large boilers, solutions which have worked at
Cholla may not work due to increased size and numbers of scrubbers.
It is our belief that improvements must be made in the future and pos-
sibly new approaches of dealing with sulfur must be developed and
316
-------
tried in order to successfully handle variations in size, fuel types
and environmental constraints.
In summary, we feel the system will ultimately work. How-
ever, new developments arc needed. The systems must be improved
both mechanically and chemically. Disposal of wastes from existing
throw-away systems is a serious problem not yot solved. Hopefully,
technology will find a way to utilize or eliminate these wastes.
Technology should be given adequate time to provide the solutions at
a realistic cost in terms of dollars and reliability of generation.
317
-------
Wet Scrubber Operating Experience
at
La Cygne Station Unit No. 1
Kansas City Power & Light Company
Kansas Gas & Electric Company
By C.F. McDaniel
Kansas City Power & Light Co.
Kansas City, Missouri
Presented at
EPA Flue Gas Desulfurization Symposium
Sheraton-Biltmore Hotel
Atlanta, Georgia
November 4-7, 1974
319
-------
320
-------
Description
La Cygne Unit No. 1 is a new generating station
located about 55 miles south of downtown Kansas
City and east of La Cygne, Kansas, figure 1. The
plant is jointly owned by Kansas City Power &
Light and Kansas Gas & Electric Company. The
plant is manned by KCP&L employees, with
output shared equally by the two companies.
Site construction was started in April, 1969, with
erection of the Air Quality Control System
initiated in mid-1971. Unit No. 1 was nearing
completion when the boiler was first fired on
December 26, 1972, and construction was
considered complete when the unit was declared
commercial on June 1. 1973.
An earth fill dam almost 7,000 feet long was
constructed to form a 2,600 acre lake for cooling.
Fly ash and spent slurry from the scrubber are
contained in a 160 acre settling pond located east
of the cooling water lake.
Coal is delivered to the mine-mouth plant with
off-the-road 120-ton trucks from a strip mine
operated by Pittsburg & Midway Coal Company.
Run of mine si/.e coal, 30 in. x 0 is conveyed from
the receiving hoppers at a rate of 2,000 tph to two
rotary breakers and reduced to 3 in. x 0 size. The
fuel is a low grade bituminous coal that has an
as-fired heating value of 9.000-10.000 Btu/lb,
20-30 percent ash, and up to 6 percent sulphur.
The coal continues on to be stacked in the
storage area with the stacker-reclaimer, or a
maximum of 1200 tph may be bypassed to the
secondary crushers.
Limestone is obtained from nearby quarries where
the rock is reduced to 3/4 in. x 0. It is delivered to
the plant in 50-ton off-the-road trucks and is
stacked and conveyed with the coal handling
equipment.
The boiler is a cyclone-fired, supercritical,
once-through, balanced draft Babcock & Wilcox
boiler. Its rating is 6,200,000 pounds of steam per
hour, 1,010° F, 3,825 psig at the superheat outlet
and 1,010° F at the reheater outlet.
The turbine-generator was supplied by
Westinghouse and is rated at 870 MW gross output
with 5 percent overpressure.
The air quality system, shown schematically in
Figure 2 was supplied by Babcock &. Wilcox, and
consists of seven, one-seventh capacity
venturi-absorber scrubber modules, designed to
handle the entire flue gas flow of 2,370,000
ACFM, (Figure 3)
Each scrubber module consists of a variable-throat
venturi followed by a two-stage countercurrent
tray absorber, with the wetted parts fabricated
from 316L stainless steel. Seven rubber lined,
7,750-GPM venturi recirculation pumps and seven
10,300-GPM absorber recirculation pumps
recirculate a flyash-limestone slurry in the system.
The original plant design called for only one of the
seven scrubber modules to be supplied with two
stages of absorber internals and absorber
recirculation pumps.
Flue gas leaving the scrubber modules is reheated
25 F with a bare-tube steam coil reheater and
discharged to a 700-foot stack by six one-sixth
capacity induced draft fans which draft both the
boiler and scrubber system.
Limestone rock. 3/4 in. x 0 pulverized to a fineness
of 95 percent minus 200 mesh in two 100 tph
Koppers wet ball mills.
The auxiliary power requirements for the AQC
system, 28 .MW, are about the same as the normal
auxiliaries for the plant. 22 MW, The total
auxiliaries result in a design plant net electrical
output of 820 MW.
Startup and Operation
The scrubber system was initially placed in service
on December 26, 1972, using two venturi scrubber
modules during initial firing of the boiler. The
turbine was rolled on January 25, 1973, and
synchronized on February 23, 1973. A total of 1(5
days of operation was obtained with the
boiler-scrubber during the first 3-month startup
period. Sustained unit load was not obtained until
the month of April, 1973. The unit was declared
commercial on June 1, 1973, only 4 weeks after
the scheduled commercial date, and a peak load of
832 MW obtained on June 2, 1973. During this
initial trial operating period, many problems were
321
-------
encountered with the plant, most of which would
not be considered abnormal when starting up an
entire new power plant facility the size of La
Cygne. These problems involved auxiliary boiler
casing leaks, boiler tube leaks, furnace slag tap
freeze-ups, generator balancing, control system
problems, coal bunker difficulties, and auxiliary
transformer problems.
Scrubber system problems during the startup
period involved mainly completing the contruction
of scrubber modules and the instrumentation and
controls.
One of the major operating problems has been
associated with the induced draft fans being
designed too close to their critical speed, which
resulted in high fan vibrations created by buildup
of wet flyash deposits on the rotors. This condition
necessitated frequent fan washing and balancing of
the rotors. The fan rotors have all been rebuilt with
increases in shaft diameter and blade thickness in
order to increase the critical speed. The problem of
wet flyash deposits on the fans is thought to be the
result of a number of conditions which exist in the
scrubber system. Some of these reasons are:
System scrubber chemistry has been very difficult
to control; that is maintaining the limestone feed
slurry fineness, concentration and feed rate within
limits that do not produce scale. The reasons for
this control problem will be explained later. When
operated outside normal chemistry limits, the
scrubbers generate large quantities of calcium
sulfate scale that cause pluggage and erosion of
venturi spray nozzles. Most of the scrubber
modules have been operated with only the venturi
recirculation circuits in service since start-up. The
nozzle pluggage condition has led to poor fly-ash
removal which has resulted in a rapid buildup of
flyash deposits on moisture separators, reheaters,
and ID fans. In addition to poor flyash removal,
these periods of nozzle pluggage also permit some
of the sulfur trioxide in the flue gas to pass
through the scrubber, resulting in a condition
where the gas entering the induced draft fans is
below the SO3 dew point, even after reheating the
flue gas 25 F. This condition results in wet, acidic
(pH 1) flyash deposits on ID fan and reheater
surfaces. This combination of flyash and sulfuric
acid has eroded fan blades and caused failure of the
304 stainless steel reheaters.
A test program is currently under way to
determine the following:
1. Degree of flue gas reheating that will result
in minimal deposits on the fans.
2. The best method of obtaining the
additional reheat.
In the meantime, 75 degrees reheat has been
temporarily provided by taking secondary air at
495° F from the air heater and blending this with
flue gas leaving the scrubbers to obtain a 200
degree flue gas entering the fans. This reheat
arrangement limits the unit load to approximately
650 MW.
"D" module is the onJy module that is complete
with both venturi and absorber recircuiation
circuits, two stages of absorber trays and a water
wash stage below the moisture separator. Testing
during the past year indicates that dust loadings
leaving this module are better by a fator of two
than dust loadings obtained from the venturi-only
modules. The average of 37 tests on "D" module
(table I) show that the dust removal performance
has been better than predicted.
Within the last few weeks, one additional absorber
stage and a water wash tray have been added to
"C" module and tests were conducted to verify its
performance. The results of 11 dust tests show
better performance than either the venturi alone or
the venturi and one absorber stage configuration.
Additional absorber and water wash trays are being
installed in the other modules as they become
available for revision.
The sulfur dioxide removal performance at La
Cygne is dependent upon many factors:
(1) module equipment (2) purity of limestone
from the quarry and (3) operation of the milling
system. This summary of SO2 performance (table
2} show from 75-83% SO2 removal has been
obtained from "D" module when the above
conditions are close to expected. Operation with
one absorber stage has resulted in 50-74% SO;
removal, and with the venturi only, we get 35-50%
percent efficiency. (Table 2)
Proper chemistry control is the most important
factor required for the successful operation of the
322
-------
La Cygne scrubber system. This control not only
includes maintaining recirculated slurry within
certain operating limits but also encompasses the
selection of raw limestone rock, reducing it to the
right fineness, and feeding the limestone slurry to
the scrubber at the right time.
The limestone and the coal at La Cygne are both
mined locally; both vary considerably in chemical
analysis; and if you don't obtain the right type,
you are going to have problems. Unfortunately,
with both the coal and limestone, you don't know
you have the wrong type until it gets into the
system. Figure 4 shows some recent variations in
limestone purity since April of this year.
During the period April-August, 1974, we went
through a period of excellent chemistry control,
seldom having to take a module out of service due
to low pH. This period the limestone purity was
generally greater than 90 percent calcium
carbonate, the magnesium carbonate concentration
was less than 1.5 percent, and the milled feed
slurry fineness in the range of 4.5-6 microns. A
number of interesting observations were also made
during this operating period.:
1. The system pH was easily maintained at pH
5.7-6.0 with limestone feed at 150-175
percent of stoichimetric requirements.
2. Little scaling was observed in the scrubber
modules.
.3. The oxidation rate of calcium sulfite to
sulfate was as low as 10-20 percent during
some periods of operation.
During this operating period large hydrocyclones
were put in service on all the module venturi
recirctilation streams. Prior to this time, it was
almost impossible to prevent venturi pump
recirculation screens from plugging with scale.
These were replaced with low-pressure-drop
hydrocyclones designed to remove particles greater
than 50 mesh in size from the entire 5,500 GPM
venturi recirculation flow. Oversize material is
discharged into a grit screen located on top of the
recirculation tanks. This system has reduced nozzle
pluggage in the Venturis and has greatly decreased
erosion of valves and spray nozzles in the scrubber
system.
With four to six ID fans operating in parallel from
the plenum common to the discharge of all seven
modules, some will assume all the load and others
unload when the negative pressure reaches 42"-45"
H2 O. Unless lowered by the Control Operator this
will lead to a boiler trip. Individual fan
characteristics are currently being studied to
correct this problem.
Present procedures call for cleaning one module
each night on a rotational schedule and attempting
to keep all seven in service during the day. Washing
requires two to three men approximately ten hours
to complete, including tagging and opening the
many sections. The predemisters and absorber
trays are normally clean unless scaled over from
erratic pH control. The venturi area usually
requires nozzle and wall wash cleaning which may
include dismantling up to 7 or 8 of the 80 nozzles.
The common sump under the venturi and absorber
usually contains from 4 to 12 inches of hard sludge
formed from flyash, carbonate, and sulphates and
often is like concrete to remove. We currently are
experimenting with this "puddle" to see if
removing the sump screens will eliminate this fluid
level.
COSTS
The operating expense for the first nine months of
1974 is as follows:
Operating labor S 187,497 0.17 mils/kW
Operating materials $ 26,353 0.12mils/kW
Maintenance labor $ 260,486 0.23 mils/kW
Maintenance material S 456,567 0.42 mils/kW
Limestone S 627,419 0.56 mils/kW
TOTAL
Capital Costs
Total Unit
Total Scrubber
$1,566,322 1.40 mils/kW
$195,000,000
$ 42,000,000
$238/kW
$ 51/kW
The problems discussed were some of the more
involved and time consuming that we confronted.
Many repairs and some design changes have been
made to the scrubber during the shutdown for
first-year turbine inspection - February 15 to April
6. These changes have given us a much more
reliable and maintenance-free operation. Startup
problems are always expected, and we believe we
now have the experience to maintain reliable
operation, while contending with any other
problems that may appear.
323
-------
Improvements being considered for the system are:
1. Addition of fan eighth module
2. Additional forced draft fan
3. Additional induced draft fan
4. Separate conveyor system for limestone
5. Different ball mill liners - manganese or rubber
instead of nihard.
6. Modernizing certain instrumentation for
reliability. Add instrumentation to the
moisture content of reheat.
7. With the remaining life of the 160 acre settling
pond estimated at 1-1/2 years, evaluation will
have to be made for an additional lake,
removing the contents to fill in a nearby quarry
or coal pits, or building the dams higher.
8. Module Improvements:
a. New improved demister installation
b. Additional absorber tray
c. New perforated predemister installation
d. Second demister tray installed in addition
to new tray
e. Re-evaluation of reheat system to include
reheat tubes and heat from an external
source such as oil burners
f. Additional wash systems to eliminate
periodic cleanings
Performance & Availability
The availability of the boiler-turbine has been
increasing steadily, from 37% in January, to 60% in
May, to 83% in September. This has consequently
put increased pressure on continuous operation of
the scrubbers. There are three demands affecting
module outages: (1) regular cleaning (2)
maintenance requiring immediate attention, and
(3) continuing modifications. Often these cannot
be done simultaneously but for the past few
months module availability has been 100%; 7
modules have been operating 20%, 6 modules 90%
and 5 modules 96% of the time the boiler is
available. In all fairness to our past efficiency tests
it makes a tremendous difference in the cleanliness
of the module, the solid control through good
instrumentation and maintaining the unit without
imbalances.
324
-------
U!
ro
Ul
c
S
La Cygne limestone wet scrubbing system
-------
Single scrubber module
326
Figure 3
-------
Particulate Performance
Module
A.B.C
C
D
No. of
tests
30
11
37
Outlet
dust
loadmp
Ib/IOOOlbgas
0.178
0-115
0.080
009!
Remarks
Ventun i 1 absorber stage
Venturi + 2 absorber swges +
wash water tray
Predicted performance equiva-
lent to EPA 0.1 Ib/MKB
Trible I
i Performance
Module
A
E
D
S02 ppm
inlet
4522
3970
4453
outlet
2710
1410
819
Removal
efficiency
%
40
64
82
Range of
SO-, removal
%
3540
50-74
7583
80
Remarks
Venturi
Ventun + 1 absorber stage
Veniori t 2 absorber stages
Predicted performance
Table
100 r-
% 90
CaC03
%
MgCO3
Average
particle
diameter
micron
©
Limestone analysis and fineness
©
©
®
MAY
©
JUNE
327
JULY
Figure 4
-------
DUQUESNE LIGHT COMPANY
PHILLIPS -POWER STATION
LIME SCRUBBING FACILITY
By
Steve L. Pernick, Or
Manager-Environmental Affairs
Duquesne Light Company
Pittsburgh, Pennsylvania
and
R. Gordon Knight
Superintendent-Technical Services
Duquesne Light Company
Pittsburgh, Pennsylvania
Prepared for presentation
at
EPA Flue Gas Desulfurization Symposium
Atlanta, Georgia
November 4-7, 1974
329
-------
DUQUESNE LIGHT COMPANY
Phillips Power Station
Lime Scrubbing Facility
CONTENTS
Page
Introduction 331
Pilot Plant Operation 332
Installation Schedule 333
Descri pti on of System 335
Initial Operation 338
Extended Shutdown for Repairs , 340
Second Start-up Operation 343
Sludge Treatment and Disposal 345
Future Operating Plans 348
Aval 1 abi 1 ity and Rel iabi 11 ty 350
Capital and Operating Expenses 350
Future Objectives 351
330
-------
DUQUESNE LIGHT COMPANY
Phillips Power Station
Lime Scrubbing Facility
INTRODUCTION
Duquesne Light Company is an investor-owned electric utility
serving approximately 1/2 million customers in southwestern Pennsyl-
vania including Pittsburgh. The Company has a net generating capa-
bility of approximately 2700 MW and presently operates three coal
burning power stations, two of which, Phillips and Elrama, are dis-
cussed in this paper.
In 1969, Gibbs & Hill, Inc. was engaged to conduct a com-
prehensive study of the most feasible means of complying with: the
emission regulations being promulgated at the time by Allegheny County,
which has control over the Phillips Station; the regulations of the
Commonwealth of Pennsylvania, which has jurisdiction over the Elrama
Station, and the most stringent regulations that could reasonably be
conceived. The study was completed in September, 1970. It repre-
sented a comprehensive evaluation of all test data concerning emis-
sions from the two power stations and an intensive investigation of
all possible means of compliance with the applicable and anticipated
regulations. The study concluded that: (1) the use of low sulfur
western coal was not a feasible means of compliance since it would
not achieve compliance with the particulate emission limitations of
the regulations with the existing dust collection equipment; (2) that
conversion from coal to low sulfur oil was not feasible due to the
uncertainties of obtaining a sufficient supply of the necessary
large quantities of such oil; and (3) that the most practical means
of sulfur dioxide emission control was desulfurization of stack
gases. The consultants then conducted a thorough study of existing
methods of sulfur dioxide removal from stack gases. At the time of
their investigation in 1969 to 1970, only two full scale stack gas
scrubbing systems were in operation in the United States. One of
these has since been abandoned and the second has had a poor record
of operating reliability. Furthermore, the injection of limestone
directly into the boiler, which was the basis of these SO? removal
processes, was not at all suitable for consideration at the Elrama
and Phillips Power Stations since it would result in additional
boiler tube erosion to which the boilers were already particularly
susceptible. Furthermore, available data indicated that S02 removal
efficiency might not be sufficiently high to comply with the regu-
lations which were among the most stringent in the country.
Because of the characteristics of our boilers and our re-
quirements for a system that would remove both particulates and sul-
fur dioxide, the report concluded that a tail-end scrubbing system
was the best available means of control and recommended that we in-
stall a dual stage venturi scrubbing system. Some of the technical
331
-------
reasons for choosing the system are as follows:
1. It was thought that a dual stage scrubbing system would enable
separating the sludges resulting .from scrubbing for fly ash
in the first stage and scrubbing for $62 in the second stage,
in the event the sludge from the SC>2 removal stage became a
problem.
2. A dual stage scrubber had the potential of being converted
to a magnesium oxide regenerative system in the event it
became desirable or practical to do so.
3. Venturi scrubbing for particulate removal was fairly well
proven in the steel industry. Furthermore, if only one
of the scrubber trains was equipped with a dual stage S02
prototype, and this was proved to be impractical or un-
workable, vie would still have a workable single stage ven-
turi scrubber system for particulate removal. We could
then investigate an alternate method of 502 removal with
minimal loss of investment.
In addition to the above reasons was the fact that preliminary
time schedules indicated that these scrubbers could be delivered with the
least delay. Duquesne accepted the recommendations of Gibbs & Hill, and
in view of the unproven nature of the wet scrubbers for sulfur dioxide
removal, decided that the most prudent approach to the installation was
to begin by equipping only a portion of the scrubber system at the Phillips
Station with sulfur dioxide removal equipment. Subsequently, Gibbs & Hill
was instructed in September, 1970, to proceed with the designing and in-
stallation of the Phillips scrubber system. The scrubber installation
at this plant was to include four scrubber trains. Three of these were
single stage scrubbers for dust removal, and one was a dual stage scrub-
ber (two single stages in series) which is the prototype for sulfur di-
oxide removal. Each scrubber was to be capable of approximately 540,000
CFM, which is the equivalent of approximately 125 megawatts. With the
station rated at 387 megawatts, this meant that the station would be
equipped with essentially a spare scrubber train. This was a criterion
specified by Duquesne Light Company to assure sufficiently high avail-
ability of the system.
It was anticipated that after a trial period, the operation of
the prototype would serve as a basis for decision as to the installation
of additional similar S02 removal equipment at this station and other
coal burning facilities to comply with the applicable regulations.
PILOT PLANT OPERATION
In February, 1971, a pilot plant size Chemico venturi type,
dual stage scrubber system with a capacity of 1500 CFM was installed
at the Phillips Power Station to study the performance and operating
characteristics of the dual stage venturi scrubber system. The pilot
332
-------
plant was connected to the duct entering the mechanical dust collector
of Mo. 5 boiler. The scrubber was operated with various types of lime-
stone and lime. In addition, other conditions such as the stoichiometry,
liquid to gas flow ratio and pressure drop were varied in order to deter-
mine the conditions which would yfeld optimum collection efficiency and
operating conditions. After determination of these parameters in approxi-
mately 3 months of operation, the pilot scrubber was dismantled on May 8,
1971.
In the meantime, in March, 1971, invitations to bid were issued
by Duquesne Light Company to Chemico and Combustion Engineering for a
scrubber system capable of removing sufficient SC^ and particulates to
comply with existing and anticipated future regulations. The bid was
subsequently awarded to the Chemico Corporation in July, 1971. The
battery limits for Chemico was limited to the scrubbers and associated
pumps and controls between the inlet hot gas duct manifold to the exit
wet gas header, including the reheater but excluding the new induced
draft fans.
In obtaining approval of our plan, compliance programs were
negotiated with Allegheny County for the Phillips Station and with the
State for the Elrama Power Station. The program included the instal-
lation of five single stage venturi scrubbing trains for particulate
removal only at the Elrama Station, to be installed concurrently with
the Phillips scrubber system as described above. The plan was to use the
prototype sulfur dioxide removal installation at Phillips as a basis for
decisions concerning additional S02 removal equipment at both stations
in order to comply with SC>2 emission limitations.
INSTALLATION SCHEDULE
Although in Duquesne's best judgment, the scrubber system could
not be made operational, even for particulate removal, prior to July 1,
1973, which was 34 months from the date of our decision to install a
scrubber system, Duquesne Light eventually acceded to pressure from the
State of Pennsylvania and established a target date of January 1, 1973.
Following startup, the schedule called for a two-month "debugging" period
followed by a ten-month test program on the prototype SO^ scrubber at
Phillips. This would determine if stack gas scrubbing with alkaline in-
jection on a plant scale basis was a feasible and reliable means of re-
moving sulfur dioxide from the gases. The plan was to study operating
conditions during this period to establish optimum S02 removal efficiency
using various types of lime, the operating mode most conducive to reliable
trouble-free operation, and other essential information. The schedule
called for a decision to be made by January 1, 1974, as to whether or not
the operation of the prototype SO? scrubber was satisfactory or if an
alternate system should be investigated. If the operation of the prototype
sulfur dioxide scrubber proved satisfactory, the schedule called for design,
procurement and installation of additional SO? removal equipment by January 1,
1975, for both stations. At that time, compliance with County and State
regulations would be achieved. (For a list of significant dates see Figure 5)
333
-------
Despite our best efforts, no part of the Phillips scrubber
installation was placed in operation until July 9, 1973, and then for
only two boilers. At that time, two of the scrubber trains were
placed in operation, and two of the six boilers were connected to the
scrubber system. This was consistent with our startup schedule which
called for connecting one boiler at a time to the scrubber system so
as not to jeopardize the availability of a significant portion of the
plant's generating capability in the event of failure of the scrubber
system. The two boilers connected have a combined steam generating
capability equivalent to approximately 80 megawatts.
The July 9, 1973 startup date represents a slippage of approxi-
mately 6 months from the original schedule. This slippage was attribut-
able to the following factors:
1. The venturi scrubber system was not a completely engineered
system at the time it was purchased, and a considerable amount
of developmental engineering was required to adapt the scrubbers
to a system that would meet our requirements. Chemico was al-
so charged to incorporate, to the limits of their responsibility the
latest available operating and maintenance experiences at the Four
Corners Plant in New Mexico, the Dave Johnston Plant in Wyoming,
Mitsui Plant in Japan, and any other Chemico installation having
experiences adaptable to our proposed installations.
2. Equipment suppliers and fabricators did not meet promised de-
livery dates due to the volume of orders, decreased produc-
tivity, and inadequate quality control. Considerable money
was spent to expedite and inspect equipment and materials for
our project. In spite of this effort and expense, many late
deliveries and mistakes in fabrication occurred. Some examples
are as follows:
a. The induced draft fans were scheduled for delivery
in Phillips by July 15, 1972. Delivery was not
completed until December 12, 1972, a delay of five
months. Not only were the fans late, but numerous
cracks and defects in the welds on the fan rotors
were found after delivery. It was necessary to
grind out all of the defective welds and make the
necessary repairs in the field.
b. An approximate five-month delay was experienced in
delivery of the ductwork. In addition, because of
the huge size and weight of the ductwork, many
structural support schemes had to be studied before
an acceptable method of reinforcing the existing
station building to accept the unusual weight load
of the ductwork could be determined. After a rein-
forcement plan 'was devised, errors were discovered
in the implementation of the plan, requiring cor-
rective design work.
334
-------
c. Delays of up to six months were experienced in the
delivery of electrical equipment such as switch-
gear and transformers. In addition to the delay
of equipment, errors were, found in the wiring dia-
grams for this equipment. Once the equipment was
received, design errors were also discovered, such
as improperly designed bus connections on the switch-
gear and improper alignment of bus bars on power
centers. In addition, failures of switchgear and
transformers were experienced during field testing.
These are only a few examples of the numerous delays and errors
by equipment suppliers which affected the final completion date
of the first phase of the Phillips project.
3. Major delays were also incurred in the field, such as the follow-
ing:
a. Because of engineering delays and late delivery of
equipment, construction schedules had to be con-
tinually revised which seriously affected pro-
ductivity.
b. Space limitations existing at the job site, more
serious than anticipated, affected productivity
because of the lack of accessability and maneuver-
ability. Special equipment and construction methods
for the installation of the equipment were required
beyond those originally anticipated, thus contribu-
ting to additional delays.
c. The availability of boiler outages for tie-in of the
new ducts, presented additional problems. Outages
had to be scheduled months in advance, and any re-
vision in schedule was dependent upon availability
of purchased power and the outage schedules of
other generating equipment, both on our system and
on interconnections.
DESCRIPTION OF SYSTEM
Design of both the Phillips and Elrama systems was a joint
effort by Gibbs & Hill and Chemico, each having certain battery limits.
Construction was the responsibility of Gibbs & Hill. Only the Phillips
system will be described in detail.
The Phillips system comprises four parallel trains, Figure 1,
and was designed to handle a total gas volume of 2,190,000 ACFM with
all trains in service. With one train out of service, the three
trains have a capacity of 1,600,000 ACFM, which is sufficient for
400 MW, assuming normal excess air.
335
-------
Three scrubber trains were provided with first stage scrub-
bing for fly ash removal, Figure 2. The fourth train, known as the
"prototype" SO- train, was fitted with a first stage particulate re-
moval scrubber and a second stage SC^ absorption scrubber as shown
on Figure 3.
The ductwork on each boiler was modified to enable operation
with the gases going either to the scrubbers or to the original gas
path after passing through the existing dust collection equipment.
A mechanical and electrical dust collector are located in series
ahead of each individual boiler stack. The existing ducts leading
to the old ID fans and stacks were blanked off with removable steel
plates to enable diversion of the flue gas to a new common duct.
The new ductwork is routed to the scrubber building where 16 foot
diameter takeoffs lead to each of the four first stage scrubbers.
The hot flue gas (about 340°F) containing fly ash and sul-
fur dioxide (about 1400 ppm $02) enters the first stage scrubber and
impinges upon the upper cone. One half of the total scrubber liquor
(8250 gpm) is introduced, Figure 4, into the vessel through the bull
nozzle where it is sprayed over the surface of the upper cone, to
initiate the scrubber action. A second stream of 8250 gpm scrubber
recycle liquor enters through the tangential nozzles at a point
above the adjustable throat damper. The flue gas and scrubbing
liquor are brought into intimate contact in the throat section of
the scrubber where the particulates and some 502 are removed- The
gas and liquor continue downward to the separator section where, after
being separated from the scrubbing liquor, the flue gas enters a de-
mister where entrained liquor is removed. The gas leaves the scrubber
and enters a new wet ID fan which is provided with water sprays to re-
move any accumulation of solids resulting from carryover from the
scrubber. Fresh water is used for spraying.
The ID fan housings are lined with 1/4 inch thick natural
rubber. Wheel material is Carpenter 20 Cb 3. The shaft is 316 L
stainless steel. Each fan is driven by a 5500 HP, 1200 RPM, 4160V
electric motor. Bearings are of the water-cooled type, served from a
closed cooling water system.
Outlet gas temperature, normally 110-120°F, from the first
stage scrubber is monitored. At 175°F a control valve is automatically
opened to admit emergency cooling water to the upper cone. Additional
temperature rise would automatically shut down the fan and close the
isolation dampers.
The gas leaving the ID fan enters one of two vessels de-
pending on the train. In the case of the three single stage scrubber
trains, the gas enters an entrainment separator in which entrained
water is separated and collected. The separators or mist eliminators
are separate vessels which will be replaced by second stage scrubber
vessels if the test program indicates this to be necessary. in the
336
-------
case of the SC^ prototype train, gas leaving the ID fan enters a
second stage scrubber vessel, which is identical in size and design
to the first stage. This second stage vessel is equipped with a
reagent (lime) injection nozzle in the bottom cone. The scrubbing
liquor is then picked up by the two recycle pumps (Carpenter 20
Cb 3). Some of the liquor is bled off to the first stage scrubber, where
some is removed from the cycle to thickeners, and the balance is
recycled.
Total recycle flow is about 16,500 gpm with a 615 gpm
bleed to the thickeners. All single stage vessels have also been
fitted with reagent nozzles and feeds to enable control of pH and
incidental S02 removal.
The exit gases leaving either the second stage scrubber
vessel orthemist eliminator vessel enter a common wet duct lined
with Flakeline 103 (a product of the Ceilcote Company consisting
of a glass flake filled modified polyester resin, subsequently
referred to as Ceilcote) which leads to a new concrete acid re-
sistant brick lined chimney. Prior to entering the 340 foot
chimney, a section of the wet gas duct, constructed of 316L SS is
equipped with a reheater, employing a direct oil-fired burner system.
The two-burner system is capable of reheating 30°F, although the in-
tent is to normally reheat 20°F.
Bleed off f-rom all scrubbers is directed to a trough feeding
two 75 foot diameter thickeners. The overflow from these thickeners
is directed to a collecting tank where it is pumped by two-2000 gpm
SS pumps to the make-up line for return to the scrubber system.
Thickener underflow, at 35 to 40% solids concentration, is
pumped by one of two 105 gpm, 15 HP pumps to one of three clay lined
sludge holding ponds. Each pond has a capacity of about 6500 cubic
yards. Just prior to discharge to the holding ponds, the sludge
enters a mixing tank where a stabilizing agent can be added at a
predetermined rate based on the density of the sludge as it leaves
the thickener. After the additive is intimately mixed with the
sludge, the mixture discharges by gravity to one of the ponds.
In the pond, both settling and curing take place. Super-
natant liquid is withdrawn and recirculated to the scrubber system
via the thickener. The sludge curing ponds provide interim storage
while stabilization is taking place. One pond received thickener
underflow, while a second is curing and the third is being excavated
for final off-site disposal.
Lime is fed from a storage silo at a controlled rate to the
lime slaker where it is slaked with fresh make-up water. The slaked
lime overflows into a slaker transfer tank, where make-up water is
added to provide a constant flow of lime slurry with a concentration
of about 15%.
337
-------
A service (river) water system independent from the power
station service water system is provided. It includes a pair of 900
gpm, 100 HP pumps taking suction from the existing condenser dis-
charge tunnel and a 10 inch distribution header for the scrubber
facility. It provides fan spray water, pump seal water, mist elim-
inator and scrubber demister spray water, instrumentation flush
water, chemical mixing tank, and emergency water for scrubbers.
INITIAL OPERATION
Although the lime addition system was not yet completed,
partial start-up of the scrubber system began on July 9, 1973. The
decision to proceed with startup, without the lime addition system,
was made in light of the fact that we thought the scrubber system
was suitable for operation at a low pH. Almost all of the major
components and piping were either Ceilcoted, rubber lined or fabri-
cated of a corrosion ^esistant material to withstand a low pH.
During the first five days of operation, a pH of approximately 1.3
to 2.0 developed in the scrubbing liquor while burning 1.8-2.0% sul-
fur coal. Because of pressure to proceed with the debugging operation,
and because of our reluctance to incur additional delays, the decision
was made to continue operation with the low pH rather than shut down
the scrubbers until the lime system became operational. Unfortunately,
as a result, the scrubber system was forced out of service due to the
corrosion failure after five days of the two thickener overflow re-
turn pumps, which were intended for operation at a pH of b to 6.. They were
promptly replaced with two 316 L SS pumps. Use of the lime system
feed to the thickeners was obtained after the first week of operation.
During the first two weeks of operation, additional problems
developed and were corrected in a two-week outage following the first two
weeks of operation:
1. A piece of rubber liner was found in the suction of one of
the recycle pumps. The individual scrubber was shut down
in an effort to locate the source; however, it could not
be determined until the corrosive nature of the scrubber
media caused the failure of a 20"/16" Tee within 10 days.
2. Leaks occurred in the lead floor lining of the stack which
were a potential threat to the stack concrete structure.
The lead sheets had been soldered rather than "burned."
The lining was replaced.
3. Leaks developed in four 405 SS expansion joints in the
gas ducts leading to the stack. The appearance of the
failed metal was variously described as that of "lace
curtain" or "swiss cheese." This resulted from partial-
load type of operation of the scrubber system requiring us
to partially open the vacuum breakers and allow ambient air
338
-------
to supplement the gas flow through the scrubbers. Conden-
sation resulted. The joints were replaced with an asbestos/
butyl material.
4. The reheater, which utilizes direct firing of No. 2 oil in
the .cold gas duct as it enters the stack, is not yet oper-
ational. We have been unable to obtain satisfactory oper-
ation of the reheater due to burner problems and oil pump
problems. We have placed no priority on solving these prob-
lems due to our reluctance to operate the reheater consid-
ering the present energy crisis and oil shortage. • Whenever
the reheater becomes operational, we plan to operate it
during periods of adverse meteorological conditions'to ob-
tain a higher effective stack height.
5. During the outage, evidence of corrosion was found on the
ID fan shaft shrouds, inlet dampers and stiffener bars - all
316 L stainless steel. Two corrective measures were taken -
redesign of the fan sprays and application of 1/4-inch
rubber coating to the affected parts. After two weeks of
operation the chloride concentration in the liquor (closed
cycle)had reached 800-900 ppm. This, no doubt, contributed
to the corrosion noted above. It was eventually to reach
3500-4500 ppm.
On August 5, 1973, the scrubber system was returned to service
with the lime feed system for pH control continuing in operation. In
addition to No. 1 and No. 2 boilers, the No. 3 boiler was also connected
to the scrubber system, which made a total of approximately 145 mega-
watts.
During the following weeks of operation, we were informed by
Chemico of a major change in the operating requirements of the scrubbers;
namely, the necessity to bleed more fluid to the thickener tanks so
that the concentration of solids in the recycling fluid could be kept
to a minimum in order to protect the recycle pumps, and to minimize the
settling of solids in the bleed lines. This caused a change in the de-
sign concept by essentially doubling the bleed rate. This meant that
the two 900 gpm thickener tank overflow return pumps, which had been
provided with the system, were inadequate since they no longer had
sufficient capacity to supply the needs of the scrubber system under
one pump operation (the second being a spare) with all boilers con-
nected to the scrubber system. Two 2000 gpm pumps were obtained for
later installation in parallel with the two 900 gpm pumps.
During the two months of operation following the August 5,
1973 start-up date, operation of the scrubbers was alternated between
trains. In other words, as problems arose on one train, it was shut
down and another train was put in service to take its place. No effort
was made to operate the dual stage prototype S02 scrubber for full S0£
removal because it was felt that debugging operations could more effect-
ively proceed with the most basic operation possible, in other words,
339
-------
with the single stage scrubbers.
EXTENDED SHUTDOWN FOR REPAIRS
During the operation of the scrubbers in August and Septem-
ber, 1973, additional problems came to light which caused an extended
shutdown of the scrubber system, and a further delay in the operating
schedule, namely:
1. A routine inspection of the new stack revealed additional
moisture leaks from condensation (pH 2,0) seeping through
the mortar joints and running down the outside of the acid
brick lining. Seepage was occurring in more than 100 areas
of the brick lining. The condensate was collecting at the
bottom of the stack in the annulus between the lining and
the shell. The acid condensate was also seeping through
construction joints and down the outside of the concrete
shell permitting it to attack the concrete, which could
possibly impair the structural stability of the stack. The
contractor advised us that a four week outage would be
necessary to permit inspection of all mortar joints and
to repoint as required. This outage for other reasons to
be described, extended for about five months.
2. Significant scaling and buildup had occurred in various
portions of one or more of the scrubbers and in the asso-
ciated piping in varying degrees as described below:
a. Buildup was found on many of the upper cone sup-
port struts. These football-size accumulations
appeared to be creating a disturbance in the water
flow pattern over the cone resulting in accumu-
lations of deposit on the cone surface.
b. Deposits were found in the throat area of two of
the scrubbers, sufficiently thick in some areas
to prevent the maintenance of the six inch de-
sign pressure drop across the throat.
c. A thin buildup of hard scale was observed on the
face of the throat dampers of one of the scrubbers.
d. Deposits of fly ash were found in and around the
operating mechanisms of the adjustable throat
dampers. These were sufficient to prevent full
adjustment of the dampers.
3. Corrosion/erosion was found in numerous areas:
a. Severe pitting and corrosion appeared on all of the 316 L
clad ID fan damper blades and frames. The cladding was
340
-------
removed and the carbon steel was Ceilooted.
b. In all scrubbers, a moist layer of fly ash was observed
on the carbon steel skirt that directs the flue gas
down the cone. Extensive rusting was found under the
fly ash and on the underside of the vessel tap as well.
The latter may have resulted from frequent outages for
inspection, during which alternate wetting and drying
occurred.
c. The upper cones of the scrubber vessels were fabricated
with a 316 L SS apex. The balance was carbon steel with
a Ceil cote coating. The uncoated apex showed erosion/
corrosion in all vessels. Similar attack was shown on
the inner surface of the 316 L bull nozzle directly
above the cones. Both affected areas were coated during
the outage.
d. Baffle plates in some tangential nozzles (12 per vessel)
were found badly corroded. The specified material of
construction was 316 L SS, but it was found that the
corroded baffles had been fabricated of 304 SS. Six-
teen baffles out of a total of 60 required replacement
with 316 L.
e. Similarly, the throat damper arms (12 per vessel) were
supposed to be 316 L SS. Thirty-two out of a total of
60 were badly corroded and were found to be 304 SS. They
were replaced with 316 L.
f. The top and bottom throat damper scraper blades were
practically destroyed. The also were found to have
been 304 SS instead of 316 L. They were replaced with
316 L.
4. An initial inspection and analysis of the ID fans by Franklin In-
stitute Research Laboratory (FIRL) indicated the existence of
chloride stress corrosion cracking of the parent metal and
structural welds, as well as general corrosion on all exposed
surfaces. At that time, they advised us that the problems re-
sulted from the hostile environment in which the fans were
operating. In addition, they made a specific recommendation
that the fans not be operated in the condition found at that
time, due to possible catastrophic failure.
As a result of their search for protective coatings, FIRL
recommended the application of a Ceilcote material, Coroline
505AB. This is a hard, impervious, acid resistant, epoxy base
material which is applied at a nominal thickness of approxi-
mately 1/4". The application of the Coroline is a tedious,
time consuming hand application by trowel and requires close
341
-------
quality control. The application of the material began in
November, 1973. Although Coroline had not been extensively
used on fan blades, or proven to any degree, we were given
reasonable assurance that it would be successful. As a
result, we began application of this material to as many
fans as manpower permitted.
On January 22, 1974, the first fan was completely coated
and was started up for balancing. After a balance move
was attempted, the fan was forced out of service due to
excessive vibration resulting from separation of a portion
of the coating from one of the blades. The Ceilcote Com-
pany informed us that the separation was localized and
probably caused by a faulty batch of material which did
not properly adhere to the metal. The Coroline was re-
applied and the fan was started again on February 5, 1974
and again was forced to shut down due to separation of the
material in a second area. At that time, the Ceilcote Com-
pany was unable to adequately explain the failure of the
coating to adhere to the fan surfaces. As a result, the
decision was made that the application of Coroline as
protection against the adverse effects of the flue qases
was not the answer to the problem. It is believed that the
flexing of the blades during operation may have contributed to
the lack of adhesion between the Ceilcote and the blades.
By this time, FIRL had been able to conduct more exhaustive
investigations and analyses on the fan blade material. As a
result, they revised their initial conclusion and reported
that the fans were suffering from a much lesser degree of
stress corrosion cracking of the weld metal and a degree of
pitting attack on all fan blades. The report also indicated
that fan failure was no longer considered imminent.
In review of the ID fan problems, it appears that the corrosion
problems which we had experienced might be partially due to the operation
of the scrubber system at a lower pH and a higher chloride concentration
(3500-4500 ppm) than had been foreseen by the scrubber designers. Several
steps were taken to try to alleviate the situation:
1. Operation of the dual stage scrubber with full lime addition,
which we felt would be adequate for maintaining optimum pH
control in that train;
2. Addition of supplemental lime feed to the other operating
scrubbers, and
3. The installation of a redesigned spray system, which had proven
effective at other installations.
342
-------
These steps in addition to frequent scheduled fan inspections
were felt to be adequate safeguards for placing the system
back in operation.
Our ability to start up the scrubber system and initiate full
S02 removal in the dual stage prototype was further improved by the
completion of the sludge stabilizer addition system. We had been re-
luctant to add large quantities of lime to the single stage scrubbers,
and equally re'iuctant to begin operation of the dual stage scrubber,
both of which result in the production of considerable amounts of
sulfate sludge, without the capability of adding a stabilizer.
SECOND START-UP OPERATION
Removal of the Coroline material from the dual stage scrubber
fan No. 1 was .completed, and startup of the scrubber system once again
began March 17, 1974. At that time, the dual stage Scrubber Train No. 1,
and Scrubber Train No. 4 were placed in operation with boilers No. 3
and 4 connected to the scrubber system. Since then there have been
these additional developments:
1. Excessive Fly Ash
The sludge handling system was designed so that after 30
days of normal operation with all boilers tied in and
with the dual stage S0"2 prototype and two of the three
single stage scrubbers, the first curing pond should be
approximately full. However, after the first two weeks,
one pond was completely filled with only two boilers con-
nected to the scrubber system during the first week and
three boilers connected to the system during the second
week. This means that, with an average of 120 megawatts
connected to the scrubber system, we were producing approxi-
mately 7,000 tons of sludge in a two week period. Or,
stated another way, if the entire plant were connected to
the scrubber system, we would fill a pond within a week.
The major factor contributing to this change in capacity
proved to be an increased fly ash loading of the scrubber
system. The sludge disposal system was based on a 90%
combined collection efficiency of the mechanical dust
collectors and electrostatic precipitators. The transition
of the new scrubber ductwork to the old ductwork is such
that the gas flow pattern of the gases through the pre-
cipitators proved to be greatly disrupted. Velocity
traverse data indicated that laminar gas flow no longer
exists. Rough estimates indicated that the amount of
particulates leaving the precipitators and entering the
scrubber system may have trippled. Additional details on
the sludge handling will be covered later.
343
-------
2. ID Fan Stress
As part of the continuing investigation to determine the
conditions of the ID fans, Structural Dynamics Research
Corporation (SDRC) was engaged to conduct a series of
strain gage tests to determine the structural stability
of the fans. Their tests indicated that the yield
strength was being exceeded in several portions of the
fan blades, and that a degree of metal deformation was
taking place. After additional strain gage testing,
SDRC recommended the installation of doubler (reinforcing)
plates on each of the blades to reduce the stresses to
acceptable levels. After actual tests with stressed welded
specimens in the fan atmosphere, it was also recommended
that the doubler plates be welded with an Inconel 112 rod,
rather than the Carpenter 20, 4 NIA or 8 N12 rods pre-
viously used. Since the installation is on a fan-by-fan
basis, so that a spare scrubber train would always be
available for emergency backup, it is expected that all
four fans will not be completed until the end of November,
1974.
3. ID Fan Attack
Frequent shut down of the ID fans for inspection revealed that
significant attack was still taking place. Despite the addi-
tion of lime for S02 removal and pH control, a low pH was being
encountered in the ID fan drains. This results from additional
SC"2 scrubbing occuring with the fan spray system. In a trial
installation, caustic was added to the fan spray water. How-
ever, an inordinate quantity is required to obtain even a pH
of 4, and the trial will soon be ended. As another attempt
to alleviate the attack, a new type of fog nozzle was installed
on one of the ID fans in an attempt to increase the effective-
ness of spraying. No evaluation is yet possible.
4. ID Fan Deposits
In addition to the cleaning being accomplished by both the
original and the new fan spray systems, it was necessary to
periodically remove the ID fans from service to remove de-
posits by manual methods. In recent weeks, this phenomenon
has not required the need to clean the blades as frequently.
5. Recycle Pumps
An unsatisfactory degree of wear (about 30% in three months)
is being experienced on the recycle pumps to the point where
pump parts are being diverted from the Elrama scrubber system
344
-------
to Phillips at such a rapid rate that the stock of installed pumps
at Elrama will soon be depleted. Experimental pump impellers and
wear rings of three different materials and design will be installed.
6. Thickener Capacity
Modifications were made to the thickener tank overflows to increase
their hydraulic capacity and thus compensate for the increased dust
loading and provide spare capacity.
7. Throat Dampers
Significant unacceptable accumulations are still being found in and
around the scrubber throat damper operating mechanisms. These have
restricted the damper operation and resulted in operation at pressure
drops of up to 10 inches instead of the normal 6 inch drop. Cleaning
requires removal of the train from service for manual removal of the
mechanism and hand cleaning. About 288 manhours per scrubber are
required.
8. Sump Pumps
The ID fan spray sump pumps have been wearing out much faster than
anticipated. This is due in part to the low pH and to the higher
solids concentration in the fan drains. Once again, we have had to
draw on the Elrama installation for replacements.
9. Closed Loop Operation
A closed-loop system may be possible at maximum load with all boilers
cut-in. However, the Phillips Power Station is not a base load station,
and as a result, the load fluctuates between 30% and full load. At the
lower loads, less water is evaporated from the system, while many of the
points of water addition (purge, seal and spray water) continue. This
results in an excess of water in the system. Temporary permission was
obtained from the Pennsylvania Department of Environmental Resources
(PDER) to discharge the excess to our existing bottom-ash settling ponds.
This permits the settling out of particulate matter and dilution of the
blowdown prior to discharge to the river,
10. S02 Guarantee Performance Tests
In July, 1974, several tests were made on the dual stage prototype scrubber
for S02 removal, with inlet S02 concentrations ranging from 1263 ppm vol.
to 1,603 ppm. at stoichiometric ratios from 1.01 to 1.15 the S02 removal
ranged from 86% to 93%. The guaranteed SOj removal is 80%.
SLUDGE TREATMENT & DISPOSAL
At the time the present period of operation started, the sludge
additive system had been completed to the point where it was
considered operable. However, it soon developed that the pumping
system for the additive was operating below design capacity. The
345
-------
necessitated pumping a slurry with higher solids content,
and the 750 ft. long transfer lines were frequently out of
service due to plugging.
This stabilization additive chosen for use is Calcilox,
a proprietary product of the Dravo Corporation. This
material was developed in the approximate period of
October through December, 1972, when we re-installed the
Chemico pilot scrubber for the sole purpose of making
sludge. Industry knowledge of sulfate-containing scrubber
sludge and means of stabilization appeared almost totally
lacking at that time. The scrubber was connected so it
could be supplied from either upstream or downstream of
the existing dust collectors of the No. 4 boiler to enable
experiments with both clean and dirty gas. The material
developed, Calcilox, is of the consistency of dry cement and,
on the basis of laboratory tests, has the property of con-
solidating the sludge so that it can be disposed of as land
fill without an objectionable leachate. The degree and time
of consolidation depends on the amount added (3-15% by dry
weight of sludge), the pH of the sludge (10.5 appears suit-
able for our curing time), and temperature (lower temperatures
slow the process). Since actual field leaching tests had not
been made, we installed two membrane-lined ponds in an inactive
ash disposal area to enable collection and retention of leachate
from Calcilox-treated sludge. Bottom drain systems were pro-
vided with valves to enable leachate sampling at intervals.
One of the ponds has 22,860 yd3 capacity, the other 5720 yd3.
These capacities may be increased by 50% if the material proves
suitable for successive layering and compacting.
During the previously noted difficulties with the additive
system, Calcilox was either added sporadically or not added
at all. As a result, the sludge after residence time in the
ponds was of a "soupy" nature, which made it difficult to
excavate with a clam shell and to transport in open trucks.
Prior to placing the material in the trucks, fly ash from the
ash silo was deposited in the rear of the truck to seal the
tailgate. Because of the difficulties in handling the sludge,
additional boilers could not be connected to the scrubber
system since it would tax our ability to remove the sludge
from the ponds as rapidly as it was being produced. Since
the sludge did not have the amount of additive which would
normally be added prior to discharging it to the ponds, no
appreciable degree of stabilization took place in the ponds.
We did not feel that the "soupy" material was representative
of that which would result from the normal sludge disposal
system, and were reluctant to use one of the lined test
ponds for this material. Therefore, we requested interim
permission from the State's Solid Waste Disposal Division
of the PDER to dispose of the material on the ash disposal
area where we normally dispose of the fly ash and bottom
346
-------
from the generating plant. The State consented to our
request, and we started disposal of the unstabilized sludge
on our ash disposal site. It was spread in thin layers,
mixed with additional fly ash and then compacted.
Handling of the sludge was extremely difficult. The trucks
could only be loaded to approximately 1/2 capacity because
of the danger of the sludge flowing over the sides if the
truck stopped or started too abruptly or made turns at
excessive speed. In addition, because it was impossible to
prevent sludge from dripping onto the sides of the truck
during the truck loading operations, it was necessary that
the trucks be washed prior to leaving the plant premises.
Even with such precautions, occasions arose when sludge was
deposited on the roadways. After several warnings from the
local police and the Pennsylvania Department of Transportation,
it was necessary to rent a street cleaner and a water sprinkler
truck to clean the streets during the hours of sludge hauling.
In approximately May, 1974, modifications were completed on
the additive system, and Calcilox was then added on a continu-
ous basis to the sludge with around-the-clcck technical super-
vision. Filling time per pond was about 10-14 days with the
increased quantity of fly ash, instead of the 28-30 days which
had been planned. Similarly, the curing time was reduced from
30-60 days to 14. However, with about ~\Q% Calcilox addition
the sludge did consolidate to the extent that drag-line ex-
cavation was possible, truck tailgates did not require sealing,
and it was possible to fully load the trucks. Observations
of the sludge during excavation showed that although the top
layer resembled hard clay in consistency, it became earth-like
and then almost fluid near the bottom. It was necessary to
mix the lower-most layer with dryer material from the upper
layers to obtain a satisfactory handling quality.
The consolidated material was trucked to and deposited in
one of the lined areas for leachate monitoring. However, the
thixotropic nature of the material was such as to prevent
either leveling or successive layering of truckloads. It
was necessary to tailgate the sludge and then let it stand
(cure?) for about six more weeks before it could be leveled,
and dozers and trucks did not get "hung-up." This, of course,
slows the disposal process considerably.
Our present procedure is to add consolidated sludge to the
lined area whenever possible. At other times, it is mixed
and compacted with dry fly ash on the normal disposal area.
We have now been monitoring the bottom drains from the lined
area at weekly intervals for six weeks. In that time, the
total dissolved solids (TDS) have varied from an original
347
-------
2203 mg/1 to the present (October 2) 3380 mg/1. A sample analyzed
by the PDER on September 22 verified the results on our own
sample of that same date. No record of total volume of leachate
has been maintained, and it is realized that this quantity can
affect the measured IDS appreciably. Similarly, it is realized
that the method of disposal by tailgating and our inability to
level the top of the sludge to a possibly impervious surface
has also had an effect.
In addition to the above tests, leachate studies were con-
ducted on a more closely controlled basis. About June 1,1974,
we used a concrete mixture to prepare four sludge mixtures
with various additives. The additives were fly ash, 2%
hydrated lime, 5% hydrated lime and 11% Calcilox. Each mix
was then added to a plastic wading pool which had previously
been fitted with a valved underdrain system for leachate
sampling. Since the tests started, we have added water to
each pool once each week so as to cover the sludge after
first sampling and draining all water from the previous week.
After fourteen weeks, the most meaningful tests are those with
the 5% lime and the Calcilox. They show that the total dissolved
solids (TDS) on the lime test started at 1450 mg/1 and are now
460 mg/1. The TDS on the Calcilox started at 3370 and are now
2000. Penetrometer readings in that time have increased to
2.7 tons/ft^ on the lime and 4.5+ on the Calcilox. In summary,
while more salts are leached from the Calcilox mix, the degree
of consolidation is higher than that of the lime.
FUTURE OPERATING PLANS
Due to improvements in the method of sludge treatment and
handling, and due to close control of the thickener tank underflow
operation, it was possible to cut-in a fourth boiler to the scrubber
system on August 18, 1974. This increased the capacity connected to the
scrubber system to approximately 180 megawatts. Our plan is to operate
the scrubber system with the four boilers for a period of time to deter-
mine the extent of any aggravation of existing problems and the occurrence
of any new problems. If the problems become intolerable, it may be neces-
sary to return the scrubber system to a three boiler operation. However,
if the problems are not severe, we anticipate the possibility of adding
a 5th boiler to the scrubber system. We are extremely cautious in our
optimism because of the number of unknowns in the sludge disposal system
such as the following:
1. Since the lime feed system is essentially operating at
maximum capacity, additional boilers will reduce the pH
in the scrubber system. We are told that a reduced pH would
adversely effect the stabilization of the sludge. However,
we do not know to what degree it will be affected. Also,
if additional boilers are connected to the scrubber system,
348
-------
the rate of sludge production may not permit optimum stabili-
zation period which now appear to be greatly in excess of 14
days.
2. With the oncoming cold winter months, it is not known to what
extent low temperatures will hinder stabilization.
3. If additional boilers are connected to the scrubber system,
and it becomes impossible to optimize stabilization of the
sludge prior to removing it from the ponds, it may be necessary
to again seal the tailgates with fly ash prior to loading the
trucks with sludge. This would then be conducive to sludge
material dropping on to the road, freezing and causing com-
plaints from the local citizens.
If we can connect five out of the six boilers to the scrubber
system by November 30, 1974, we will begin our ten-month test program
at that time to evaluate the efficiency, reliability, and practicability
of the scrubber system. A decision would then be made on September 30,
1975, concerning the installation of additional S02 removal equipment
for compliance with the S02 emission limitations. The number of
additional second stage scrubbers required for compliance with the
regulations will be determined during the ten-month test program, and
this will establish the completion date for installation of all additional
scrubbers.
In the hope of obtaining higher efficiencies of S02 removal with
the use of lime, Duquesne Light is conducting tests using Thiosorbic lime
(a Drayo development) which is a lime having a concentration of magnesium
oxide in the order of 4 to 6% Various tests on pilot scrubber systems
indicate that the higher reactivity of such a lime results in S02 single-
stage removal efficiencies in the 90's. In preparation for these tests,
one of the single stage scrubbers was isolated along with one of the
thickener tanks. This isolation was necessary to establish and main-
tain the desired chemistry, since lime slaking capacity is insufficient
to maintain the necessary conditions in all scrubbers and both thickeners.
A test run with Thiosorbic lime was made in September, 1974. However,
we encountered difficulties in achieving a sufficiently high rate of
utilization of the magnesium oxide. A few cursory tests under these
conditions indicated S0£ removal efficiencies somewhat better than those
with high calcium lime, but also significantly less than anticipated.
The tests were suspended later in September to further evaluate means
of increasing the reactivity and utilization of the Thiosorbic lime.
Resumption is planned for about October 21.
As of August, 1974, construction on the Elrama scrubber system,
which is essentially identical to the Phillips scrubber system, had pro-
gressed to a point where no additional meaningful work could be accom-
plished until the problems with the Phillips scrubbers are resolved.
With the modifications and test programs now being conducted at Phillips,
it is anticipated that solutions to the present problems may progress
to the point where startup of the Elrama scrubber system may be possible
in May, 1975.
349
-------
AVAILABILITY AND RELIABILITY
The availability of the Phillips scrubber system is meaning-
less at this point since we have essentially 100% backup. In other
words, there are two spare scrubber trains which can be placed in
service whenever a problem arises on an operating train. The service
hours for each train for the period from March 17, 1974 through
September 30, 1974 are 3430 hours for No. 1 train (dual stage),
1678 hours for No. 2 train, 1931 hours for No. 3 train, and 2803
hours for No. 4 train.
Reliability was built into the design of the system which
called for a spare scrubber train enabling us to carry almost full
station load on three scrubber trains. In addition, a degree of
reliability is built into each scrubber train through the instal-
lation of spare pumps, thickener capacity and backup electrical con-
trols.
We do not view bypassing as the solution to maintaining
acceptable reliability. However, we have incorporated a manual type
bypass into our scrubbing system as the best approach to avoid the
serious consequences of scrubber failure. The system uses blanking
plates to redirect the gas flow to the original stacks, and the
boilers have to be shut down during the plate change. Under normal
circumstances, this process would consume several days depending on
the availability of manpower. There really was no alternative in
designing a bypass system for our plants. Mechanical bypass systems
utilizing a damper arrangement require tight dampers, and sufficiently
tight dampers have yet to be developed. We also had serious space
limitations on installation of a mechanical damper system.
CAPITAL AMD OPERATING EXPENSES
As of this date, the scrubber systems at Phillips and Elrama
represent a capital investment of approximately $61 million. It is
expected that an additional $19 million will be required for equip-
ment necessary to comply with the S0£ emission limitations. This will
represent a total investment of $80 million or approximately $91 per
kilowatt, exclusive of the cost of any additional sludge disposal facili-
ties required.
A recent estimate indicates that the annual operating and
maintenance expenses, including fixed charges, at Phillips and Elrama
with full S02 scrubbing will be in excess of $30 million per year, or
5.5 mills per kilowatt hour, exclusive of additional sludge disposal
facilities which will be required for full S02 removal operation. These
expenses would represent an increase of approximately $11 per ton of
coal consumed. Expressed in different terms, these expenses, without
fixed charges, would represent approximately a 50% increase over the
actual operating and maintenance expenses for 1973.
350
-------
In an effort to estimate the total capital and operating
costs that would ultimately be incurred for the scrubber systems at
the Phillips and Elrama Stations, including provision of adequate
sludge disposal facilities for full S0£ removal, a preliminary
evaluation was made of the additional property required for develop-
ment into a disposal area and the additional expenses involved in
disposing of the sludge. This estimate indicates possible capital
expenditures of $110 million for scrubber installation and property
acquisition and development, of which $61 million has thus far been
spent. This represents an investment of approximately $124 per kw.
Present estimates of the sludge disposal costs are $7 to
$10 per wet ton, or $15 to $20 per dry ton of sludge.
Future Objectives
Although we feel we may be able to overcome the present oper-
ating and equipment problems with additional time, effort and expense,
there are five levels of performance which must be satisfactorily re-
solved if this flue gas desulfurization system is to be operationally
feasible and economically acceptable:
1. Reliability - meaning the scrubber facility should meet
the degree of availability normally expected of power
generating equipment.
2. Turndown - the capability of the flue gas desulfurization
equipment to follow the normal cycling operation of a
power generation facility without serious disruption to
both the scrubber system and the generating plant.
3. Closed Loop Operation - the ability to operate without
the discharge of objectionable liquid effluents as
dictated by the applicable water quality requirements.
4. Sludge - the technique for disposing of sludge from the
flue gas desulfurization system without any adverse
ecological effects.
5. Cost consideration - capital and operating expenses should
not be of such a magnitude as to impose unreasonable
financial burden on the operator/ owner and the consumer.
Until all these areas of concern are resolved, our system
cannot be considered a successful operation as reports by others have
indicated.
351
-------
FIGURE 1
CALCILOX CAiL/LOX
BiPG.
PUMPS ftiM£\ ft//K£\ fUM£
SILOJ \SILOJ \f,iLO/^)\
^—/ x>—/ \--S I*-BLURRY
I © I AKEA
TRAIN 4
TRAIN 3 "TKAIN 2. TRAIN 1
AIR POLLUTION CONTROL EQUIPMENT LAYOUT
PHILLIPS POWEg STATION
-------
FIGURE 2
HOT FLUE MS TO /Lr. STAGE SCKUaB£KS
SINGLE
SCRUBBER TRAIN
PHILLIPS POWER STATION
-------
FIGURE 3
PROTOTYPE DUAL STAGE SCRUBBER TRAIN-PHILLIPS POWER STATION
HOT PI. tie GAS
COLO
cv*s
OIL
AIR
A^1
•8
t
KXX*
$vv
>
tH-UWtK 1 I
T T
_[
©
VtBKArOP.
PUMP (2)
-------
*a-
LU
rs
CJ3
A PJUS TA BL E TUROA T
PAMPERS
VZNTURI SCRUBBER SECTION
355
-------
FIGURE 5
SIGNIFICANT DATES
December, 1969
September, 1970
October, 1970
February-May, 1971 -
May, 1971
July, 1971
December, 1971
July, 1973
July 9, 1973
October 8, 1973
March 17, 1974
August 26, 1974
October, 1974
Gibbs & Hill, Inc. requested to study methods
of compliance.
Gibbs & Hill study recommends scrubber and
Duquesne Light Company gave authority to
Gibbs & Hill to proceed with projects at
Elrama and Phillips Power Station.
Invitations were extended to bid on scrubber
pilot plant.
Chemico pilot scrubber tests at Phillips under-
way.
Design of full scale system started.
Contract for scrubber process including scrubber
recycle pumps, reheater and associated controls
and instrumentation awarded to Chemico.
Construction started at both Phillips and Elrama
plant sites.
Construction of Phase I at Phillips essentially
complete. Phase I consists of one dual stage
S02 scrubber train and 3 single stage particulate
scrubber train with associated lime and thickener
equipment.
Operation began at Phillips with two and eventually
three of station's six boilers.
Scrubber system shutdown for extensive repair.
Scrubber system began with three of six boilers.
Fourth boiler cut into scrubber system.
Fifth boiler cut into scrubber system.
356
-------
THE HORIZONTAL CROSS FLOW SCRUBBER
by
Alexander Weir, Jr., Principal Scientist for Air Quality
John M. Johnson, Test Modules Program Manager
Dale G. Jones, Horizontal Module Project Manager
Spencer T. Carlisle, Highgrove Horizontal Scrubber Project Manager
Southern California Edison Company
for presentation at the
United States Environmental Protection Agency's
"Flue Gas Desulfurization Symposium '
Atlanta, Georgia
November k,
357
-------
TABLE OF COKTEI'ITS
Page No.
Abstract 359
Foreword 360
Introduction 361
System Description 363
Operations: 369
Operations and Maintenance ^69
Availability 369
Turn Down Ratio 371
pH Control 372
Closed Loop Water System 372
Performance:
S02 Removal 374
Particulate Removal 377
Lime Utilization 380
Power Consumption 383
Pressure Drop 383
Acknowledgments 386
References 387
358
-------
ABSTRACT
The Horizontal Cross Flow Scrubber
The results of pilot plant testing of 9 scrubbers and the
rationale for selection of two scrubber types for construction on a
much larger scale at the Mohave Generating Station was described^ at
a previous EPA symposium. One of these referenced scrubbers, the TCA
Scrubber, was severely damaged by a fire January 2k, 197^ and is still
undergoing repair at the time of this writing (August, 197*0.
This paper presents the results obtained with the other large
scrubber, the Horizontal Cross Flow Scrubber, since the start of its
test program January 6, 1.91k. The results of operation at ^75,000
SCFM are compared with those previously presented for operation on a
much smaller scale (3000 SCFM).
Operating experience to date confirms the wisdom of the Test
Module approach as the best method for achieving reliable, continuous
compliance with the very stringent 0.15 Ib/million BTU SOg removal
regulations imposed by local and federal authorities.
359
-------
FOREWORD
The experiments reported in this paper relating to the 1 MW and.
10 MW Horizontal Scrubbers were sponsored by the Southern California
Edison Company.
The 1TO MW Horizontal Cross Flow Scrubber was designed, construc-
ted and tested as part of the NAVAJO/MOHAVE TEST MODULES PROGRAM. This
program was established as a joint venture of the Navajo/Mohave
Power Project participants to demonstrate full-scale scrubbing as a
necessary step in meeting regulatory requirements. The participants
in this joint venture are:
Salt River Project Agricultural Improvement and Power District
Arizona Public Service Company
Department of Water and Power of the City of Los Angeles
Nevada Power Company
Tucson Gas and Electric Company
Bureau of Reclamation of the United States Department of the
Interior
Southern California Edison Company
Funding for this program was provided by the participants in
accordance with their respective megawatt entitlements in the Navajo
and Mohave Power Projects.
360
-------
THE HORIZONTAL CROSS FLOW SCRUBBER
INTRODUCTION
Pilot plant testing of scrubbers was initiated at the Mohave
Generating Station in July of 1971- Tests with sodium carbonate,
ammonia, lime, limestone, and limestone with added soluble salts
were performed with various scrubbers and combinations of scrubbers
although not all reagents were tested with all scrubbers. The
scrubber combinations tested on a pilot plant scale are listed.
Scrubber
1. Conventional Spray Drier Followed by a Cyclone Separator.
2. Horizontal Lime Kiln Containing Balls or Chains.
3- Single Stage Ventura with Fixed Annual Throat Followed by
Centrifugal Separating Chamber.
U. Vertical Absorber Followed by Three-Stage Impingement Tray
Vertical Absorber.
5- Vertical Absorber Packed with Plastic Polygrid Packing.
6. Vertical Turbulent Contact Absorber Tested with 1, 2, 3 and
k Stages of Mobile Spheres.
7- Vertical Spray Tower.
8. Venturi Followed by Three Stage Turbulent Contact Absorber.
9- Horizontal Cross-Flow Scrubber.
Some of the results of this comparative testing has previously
been reported.! At that time it was indicated that two of the
scrubber types had been selected for further development on a larger
scale than heretofore had ever been built, 1450,000 SCFM each.
Formal authorization was given to contractors to design and construct
a Horizontal Cross Flow Scrubber system (called the Horizontal
Module) and a four stage Turbulent Contact Absorber system (referred
to as the Vertical Module) in December 1972 and, as previously
reported, construction started on the Horizontal Module on
February 15 > 1973. Start-up of the Horizontal Module occurred on
schedule November 1, 197** and start-up of the Vertical Module
occurred on schedule January 1, 197^. Unfortunately, a fire in the
chlorobutyl rubber lining of the TCA scrubber occurred on January 2k,
197^- The damage caused by this fire has still not been repaired
as of this writing (September 13, 197*0 , and hence no operating data
361
-------
on this scrubber in available and, therefore, the Vertical Module
will not be discussed further in this paper. Instead, this paper
will present data obtained with the full-scale Horizontal Cross
Flow scrubber compared to the pilot plant data previously reported.
The operating data obtained to date has justified the earlier
decision to use the Test Modules approach rather than proceed with
the installation of ten production modules simultaneously at the
Mohave Generating Station.
362
-------
SYSTEM DESCRIPTION
The Horizontal Module was built essentially as described, in
Figure 17 of the previous paper-*-, which is presented here as Figure 1
for convenience. A simplified flow schematic is included as Figure 2.
Photographs of the scrubber and some of the supporting systems are
shown as Figures 3 and h. Referring again to Figure 1, some idea of
the size can be obtained by comparing the 3,000 SCFM pilot plant (and
the 6 foot tall operator) shown on the left of the figure with the
Horizontal Module shown on the right. Spray headers located at the
top of each scrubbing chamber are 33 feet above grade. The scrubbing
chamber is 15 feet high and 28 feet wide. Flue gas is drawn from the
Mohave Unit 2 electrostatic precipitator outlet duct by the 1750 hp
booster fan and is pumped through the scrubbing chamber and a demister
section in that order. The scrubbed flue gas then is mixed with
heated outside air in an indirect reheater before re-entering the
precipitator outlet duct just upstream of the 500 foot stack where it
is discharged to the atmosphere.
The design of the indirect reheat section contributes to the over-
all reliability of the scrubbing system when compared to direct reheat
systems. In this system outside air provided by a 250 hp reheat fan
is passed over carbon steel steam coils and the heated air is then
mixed with the water-saturated flue gas. The steam coils are thus not
exposed to gas more corrosive than air, and reheater performance is
independent of demister performance (with direct reheater tubes, an
inefficient demister would allow entrained scrubber slurry to build up
on the reheater tubes, increasing flue gas pressure drop as well as
accelerating corrosion of the reheater tubes themselves). Downtime
for reheat cleanup and repair is thus avoided.
The remainder of the scrubbing system was also designed with
reliability in mind. Since the scrubber does not contain any in-
ternal packing or trays, buildup and plugging in the scrubber itself
with fly ash/scrubber slurry mixture does not occur, either during
operation or on shutdown. During operation the lime scrubbing
slurry is pumped in a counter-current manner from the bottom of
one stage to the top of the next stage. Three slurry pumps are used
for each stage and, as indicated in Figure 2, they are located at
the bottom of pyramidal hoppers. The use of three pumps allows the
benefits of redundancy since at nominal L/G ratios only two pumps
are required. Satisfactory operation has been obtained with only
one pump per stage, thus allowing pump maintenance to be performed
without scrubber shutdown.
While four stages of scrubbing are used, the scrubber can con-
tinue to operate with three, two or one stages o^ scrubbing,
although as may later be seen with considerable degradation of
S02 removal performance (only J0% SQp removal with two stages). This
363
-------
FIGURE 1
SIZE COMPARISON BETWEEN PILOT
AND FULL-SCALE HORIZONTAL SCRUBBERS
U)
N \_Vjt
PILOT FACILITY
FULL-SIZE HORIZONTAL TEST MODULE
-------
Figure 2
170 MW
HORIZONTAL MODULE
-------
Figure 3, 170 MW Horizontal Module with
lime silo in background
-------
Figure A, 170 MW Horizontal Module vith
thickener tank in foreground
and Mohave Unit 2 in right
background,
367
-------
redundancy allows scrubber maintenance or repair problems to be
postponed until the generating unit is shut down.
To digress for a moment, the modular approach adopted for the
Mohave Generating Station also includes the benefits of redundancy.
As previously indicated, the Mohave participants decided to test
full scale modules before proceeding with the installation of 10
or 12 scrubbers. Thus, operating problems encountered could be
solved once, rather than trying to fix the same problem simultaneously
on a number of scrubbers. While a selection of scrubber types has
not been made, present plans are either to install five vertical
modules as presently designed plus one spare per generating unit or
four Horizontal Modules as presently tested plus one spare per
generating unit. Thus an individual module availability would only
have to be 83.3% in the case of the TCA scrubber or 80.0$ in the case
of the Horizontal Scrubber to allow 100% scrubber availability to the
generating unit operating at full load. It is believed that the
"elasticity" incorporated in the Horizontal Module will allow it to
meet that goal.
368
-------
OPERATIONS
Operations and Maintenance
The test program has been scheduled to operate the scrubber
continuously for 2k hours a day, 7 days a week in order to obtain
the maximum amount of operating time and data.
All scrubber operations are performed by SCE plant equipment
operators (PEO's) and operating foremen that are also rotated to
other station activities. One Generating Station "Watch" engineer
has been assigned full time as the operating supervisor for both the
Vertical and Horizontal test programs. SCE chemical and instrument
technicians are available as required. Although many of the test
conditions that are peculiar to the test program sometimes make
operations more complex than would normally be expected, it has been
demonstrated that two PEO's and one foreman, per shift, can operate
the Horizontal Module. Maintenance support is contracted and
consists of one millwright and one pipefitter during the day shift,
5 days per week, and one laborer during the day shift, 7 days per
week.
For both test programs three SCE Research and Development
engineers have been assigned at Mohave for test planning, test
monitoring, data analysis, evaluation and reporting. Special
chemical laboratory support and particulate/gas sampling tasks have
been contracted.
It is anticipated that the operation of multiple commercial
type Horizontal Scrubbers would be a relatively simple task that can
be handled by a minimum number of PEO's who have completed the
standard apprentice training program.
Availability
The Horizontal Test Module at Mohave receives flue gas only
from Unit 2 and does not have the inlet ducting cross-connected to
Unit 1. As a result, the scrubber is only operated when Unit 2 is
in service, i.e., when the unit is down for any reason the scrubber
is shutdown. Therefore, availability has been calculated with
respect to Unit 2 operating time as opposed to calendar time or other
reference standards. Since the start of the test program on
January 16, 197^ and as of September 13, 19T^ Unit 2 has been in-
service for a total of 5,003 hours and the scrubber in operation for
a total of U,292 hours resulting in an overall availability of
l|292/5003 X 100 = 85.h%. This percentage is the accumulative total
availability and does not distinguish between scrubber outages that
were scheduled for inspections and/or test program configuration
changes, those that were necessitated because of test module problems
369
-------
or unit outages which allowed for "free" maintenance time on the
scrubber. A brief description of both categories are listed below:
Planned or Scheduled Scrubber Outages = 297 Hours
Hours
Inspect all spray nozzles 19
Install full compliment of spray nozzles for
high L/G tests 92
Remove spray nozzles to revert to design
configuration 98
Inspect demister blades 1*
Hardware configuration change and
maintenance 8^
Total 297
Test Module Problem Outages = UlU Hours
1. Clean mix tank screen - h hours.
Problem: Overflow of the h8,000 gallon mix tank.
Cause: Debris from the first stage hopper clogged
the mix tank inlet screen.
Correction: Replaced the 1/V mesh screen with an
expanded metal screen containing 3A" openings
2. Repair and clean slaker/piping - Hi hours.
Problem: Loss of lime feed slurry to the mix tank.
Cause: Lime slaker and associated piping vas plugged
with hard calcium sulfate scale. Formation
of this scale resulted from using scrubber
process water for slaking.
Correction: Converted to use of station service
water for feeding the slaker, transfer pump
seals and lime slurry feed tank level control
bubblers. Also replaced major portions of
the slaker related carbon steel piping with
rubber hose and eliminated most all of the
horizontal pipe runs and 90° elbows.
370
-------
3. Remove two hard hats from the thickener - 9^ hours.
Problem: Plugged thickener tank outlet.
Cause: Excessive winds, up to ^5 MPH, caused two operators
working at the thickener to lose their hard hats.
The hats lodged in the thickener outlet drain cone
above the discharge valve.
Correction: Two scuba divers, in wet suits, were eventually
able to retrieve both hard hats but only after the
bed and liquid levels of the thickener were lowered.
A storage box for hard hats has been placed at the
bottom of the thickener tank ladder.
k. Repair inlet isolation damper - TO hours.
Problem: Inlet isolation damper would not close.
Cause: Both push rod shafts were bent due to operation with
improperly adjusted jack screws.
Corrections: Replaced both push rods, added thrust bearings
to the jack screws and installed heavy duty limit
switches.
5. Align booster fan/motor shafts - l6T hours.
Problem: Excessive vibration in booster fan/motor shafts.
Cause: Gradual increase in vibration during the early
summer high temperatures at Mohave. Motor and
concrete foundation were placed during cooler
weather during the fall of 1973 and expanded with
the temperature increase.
Correction: Realigned the shafts and measured the alignment
numerous times to obtain and confirm the proper
settings.
6. Repair inlet duct turning vanes and scrubber hopper flange
leaks - 38 hours.
Fatigue failure which caused portions of a turning vane to fall
off and corrosion failure of two carbon steel hopper flanges
required repair work.
Turn Down Rat To
The Horizontal Module was designed for a nominal operating inlet
gas flow rate of ^50,000 SCFM. This corresponds to a superficial
linear velocity of 21.6 ft/sec in the scrubber shell, based on the
371
-------
1 MW pilot plant tests. Results of flexibility testing indicate the
scrubber can be operated satisfactorily over a wide range of inlet
flue gas flow rates. The highest gas flow obtained to date has been
^kh,000 SCFM (S02 removal of approximately 9^%} and represents the
upper limit of the booster fan capacity. The lowest gas flow rate
tested has been 130,000 SCFM (91-^% SC>2 removal) and represents what
is believed to be equivalent to one-fifth of the lowest gas flow
rate (or load) at which one Mohave boiler unit would be operated.
pH Control
The lime slurry feed rate to the reaction mix tank is auto-
matically controlled to maintain a constant slurry pH. As antici-
pated, this automatic pH control method provides a very stable and
finely tuned chemical control system for the scrubbing system.
Results from the long duration material balance tests indicate that
the control system ensures both a high S02 removal and a high utiliza-
tion of the added lime under steady state conditions. Results of
the shorter flexibility tests have demonstrated the pH control system
capable of "following" fluctuations in flue gas flow rate and inlet
S02 concentrations.
Closed Loop Water System
The front and backend systems are integrated with the scrubber
to provide a closed loop water system. This allows the makeup water
flowrate to be a function of flue gas flowrate (and S02 content) by
providing all internal water requirements with recycled water. The
scrubbing system can thus be turned down to any desired capacity
while simultaneously turning down the makeup water flowrate to avoid
operating the system out of water balance.
The makeup requirement at a flue gas flowrate of UiiO,000 SCFM
and 280°F (averaged over 1036 hours of continuous testing) is 152 gpm.
82% of this total was cooling tower blowdown water and 18$ was service
water required for slaking and dilution of lime slurry. 93% of the
makeup water goes up the outlet duct as water vapor and the remainder
is water of hydration, interstitial water and evaporation which
cannot be reclaimed from the waste sludge disposal pond. The
scrubber area is paved with concrete and provided with curbing. All
seal water drips, hose washdown water, spills and pond supernatant
water are returned to the scrubber system to maximize waste water
utilization. No water is put into the ground or the station peri-
pheral drainage system.
The most critical part of the water recycle concept required
for closed loop operation is the source of pump and instrument seal
water. The seal water requirement represents a flowrate of 50 gpm,
which is independent of flue gas flowrate or unit load.
372
-------
With high levels of dissolved solids in the makeup water, the
total dissolved solids in the scrubber slurry reach extremely high
levels - 15 to 20 weight percent. The salts represented are largely
sodium chloride (NaCl), magnesium sulfate (MgSOlj) and sodium
sulfate
Thickener overflow is filtered to provide the pump and instrument
seal water. In spite of the high dissolved solids no severe chemical
scaling or corrosion has been observed.
373
-------
PERFORMANCE
S02 Removal
The S02 removal results of the full-scale Horizontal Scrubber are
compared in Figure 5 with data from the original 1 MW pilot plant
scrubber at the Mohave Station. For both the pilot plant and the
full-scale test module, 90% SOp removal was achieved at a liquid to
gas contacting ratio (L/G) of about 17-5 gallons per 1000 SCF, using
four stages of scrubbing. The degree of repeatability between pilot
scale results and full-scale results is noteworthy. The performance
of the full-scale test module is slightly better than the pilot
plant at reduced L/G ratios.
The criterion for the Mohave Station to be in compliance with
the Clark County SC^ regulations is equivalent to about 50 ppm at full
load. The scrubber design criterion was established at an outlet S02
concentration of Uo ppm. This criterion was met or exceeded by the
full-scale Horizontal Scrubber at 200 ppm inlet SC>2 concentration with
an L/G ratio (per stage) as low as 12.5 gallons per 1000 SCF.
Percent SC>2 removal at various L/G ratios is shown plotted for
three different gas flow rates in Figure 6. These results show more
efficient S02 removal at lower gas flow rates (and increased residence
time in the scrubber) as would be expected. As mentioned earlier,
the Horizontal Module can be operated with any number of stages.
Figure 7 shows SC-2 removal efficiency as a function of the number of
stages and indicates that 90% SOp removal can be achieved with three
scrubbing stages at the specified conditions.
The sulfur content of the coal burned at Mohave has averaged
0.38% which results in an average stack gas S02 concentration of
2CO ppm. A recent coal survey of the Black Mesa mine field indicates
the highest coal sulfur content ever to be expected will be 0.&3%
which corresponds to approximately UOO ppm 862• Most of the original
pilot plant work was performed at this level and future plans are to
inject gaseous SOp to achieve this level for testing the Horizontal
and Vertical Modules. A good estimation of the performance expected
from the Horizontal Module at ^00 ppm inlet S02 (and higher) can be
obtained from the results of SCE's Highgrove Generating Station 10 MW
Horizontal Scrubber Pilot Plant Experiment ("to test the feasibility
of scrubbing high sulfur oil-fired flue gas to the equivalent of low
sulfur fuel oil"). The Highgrove 10 MW scrubber design was based on
results from the high inlet S02 experiments performed on the Mohave
1 MW pilot unit, and is essentially identical to the ^50,000 SCFM
Horizontal Test Module with respect to operating parameters.
Figure 8 presents a comparison of outlet S02 concentration vs.
L/G ratio of the 1 Ml-/ and 10 MW scrubbers at an inlet 302
374
-------
FIGURE 5
SO2 REMOVAL VS L/G RATIO
COMPARISON 1 MW PILOT&170 MW MODULE
100-
95-
90-
85-
80-
o
LLJ
60-
50-
40-
30-
20-
10-
0-
170 MW HORIZONTAL
MODULE
1MWMOHAVE PILOT
DD
RATED GAS VELOCITY
C 0.38% SULFUR COAL
10 15 20
L/G (gpm/1000 SCFM)
25
30
375
-------
100-
FIGURE 6
SO2 REMOVAL vs L/G RATIO
170 MW HORIZONTAL MODULE
LJJ
o
95-
90-
85-
80-
70-
60-
50-
40-
460,000 SCFM
30-
0.38% SULFUR COAL
20-
10-
o-
o
5
10
15
I
20
25
L/G (gpm/1000 SCFM)
30
376
-------
concentration of ^CC ppm. Again, the degree of repeatability between
the pilot and the 10 MW version is quite good. From this plot we can
reasonably expect the It50,000 SCFM Horizontal Module to meet the
90% removal criteria (at UOO ppm_ inlet 862) with an L/G ratio (per
stage) of approximately 17-5 gallons/1000 SCF.
Figure 8 also presents the curves for the pilot plant and 10 MW
scrubbers at the 1000 ppm inlet SOp level. It is obvious that the
10 MW Highgrove scrubber performs much better than expected with
respect to L/G ratio. The lower L/G ratio requirement for the High-
grove scrubber to achieve the same degree of removal as the pilot
plant was obtained by optimizing the placement and number of nozzles
in the scrubber chamber.
Experiments with the pilot plant scrubber indicated that inlet
S02 concentration was an important variable in SOg removal. Pilot
plant test results indicated a maximum £62 removal efficiency under
fixed operating conditions at UOO ppm inlet and a rapid decrease in
removal efficiency with increases in inlet SOp concentration above
kOO ppm. The effect of decreasing removal efficiency above hOO ppm
was not as pronounced with the 10 MW scrubber as shown in Figure 9-
For the Highgrove scrubber the slight reduction in S02 removal
efficiency above 1600 ppm 862 can be minimized by either a small
increase in the L/G ratio or by the addition of a fifth scrubbing
stage. Recent tests have been made at Highgrove utilizing five
steges of scrubbing at inlet concentrations as high as 3000 ppm.
Farticulate Removal
An objective of the Horizontal Scrubber Test Program equal in
importance to S02 removal has been that of particulate removal down-
stream of the 9W efficient, cold-side, electrostatic precipitators
at the Mohave Station. Since the ash content of the coal is
nominally 9%> the grain loading downstream from the precipitator is
nominally 0.07 g**/SCF, but varies with coal quality, unit load, and
operating conditions from 0.01 gr/SCF to perhaps 0.10 gr/SCF.
Previous pilot plant results with the 3000 SCFM Horizontal
Scrubber indicated about 90% particulate removal at representative
grain loading conditions. This result has been re-confirmed with
the full-scale Horizontal Module, operating above its design conditions
at a flue gas flowrate of 47*1,000 SCFM.
Inlet and outlet particle grain loadings under constant scrubber
operating conditions are presented over a wide range of inlet grain
loadings in Figure 10. As shown, the particulate removal varies
with inlet grain loading from 70$ removal at 0.01 gr/SCF to 98$
removal at 1.00 gr/SCF.
377
-------
FIGURE 7
SO2 REMOVAL vs. No. OF SPRAY STAGES
170 MW HORIZONTAL MODULE
100-
90--
80--
70--
1 60 +
OJ
o
50--
40--
30--
OPERATING CONDITIONS
460,000 SCFM
L/G = 19.6
SO2 INLET = 2IOppm
NUMBER OF SPRAY STAGES
378
-------
FIGURE 8
OUTLET SO2 vs. L/G COMPARISONS-
1 MW MOHAVE PILOT AND 10 MW
HIGHGROVE SCRUBBER
1000-
9 -
8 -
7 -
6 -
5
4 - r
\J
2-4-
E
a
^o.
CN
o
CO
I—
uu
j-j 100
ID 9 -
O 8 -
7 -
6
5
4 - -
3 --
2 -4-
10-
MW MOHAVE PILOT
vO
a
10 MW HIGHGROVE
d
o
1 MW MOHAVE PILOT
10 MW HIGHGROVE
D
O
D
ON
O
AT RATED GAS VELOCITY
O
10 15 20
L/G(gpm/1000 SCFM)
25
30
379
-------
The increased particulate removal efficiency at higher inlet grain
loadings is undoubtedly due to the fact that at higher dust loadings
the stack gas is composed of larger diameter particles due to
precipitator sections being out of service.
Particle size in the flue gas at Mohave are normally very small.
About 90 cumulative wt % of the particles are less than U microns
in diameter with 70 vt % less than 2.0 microns in diameter, Uo wt %
less than 1.0 micron in diameter and about 15 'wt % less than 0.5
microns in diameter.
Fractional size collection efficiencies previously reported in
last year's EPA symposium for the pilot plant are compared below with
full-scale test results.
3000 SCFM U7l*,000 SCFM
Horizontal Pilot Plant Horizontal Scrubber
at an inlet grain at an inlet grain
Particle Size loading of 0.02 gr/SCF loading of 0.08 gr/SCF
Greater than
1.5 microns 97$ 92%
1.0 to 1.5
microns
0.5 to 1.0
microns 87$ 85$
0.3 to 0.5
microns 75f» 16%
Lime Utilization
High calcium pebble lime delivered from Nevada indicated the
following average composition:
CaO alkalinity as CaO
(CaO, Ca(OH)2, and CaC03) 95-5$
Total alkalinity as CaO
(CaO, MgO, Ca(OH)2, CaC03
and other trace alkaline materials 97-^$
Lime utilization was calculated by three different methods over
an operating period from May 6 through May 25, 197^. Other sources
of alkalinity entering the scrubber system were also taken into
account. Chemical analysis showed that flyash and scrubber makeup
380
-------
FIGURE 9
LO
00
i
LLJ
ex.
CN
o
t/5
100-
90- -
80- -
70- -
60- -
0
EFFECT OF INLET SO2 CONCENTRATION
ON PERCENT SO2 REMOVAL
D
10 MW HIGHGROVE L/G = 20
1 MW MOHAVE PILOT L/G = 20
FROM MAY 1973 EPA PAPER
AT RATED GAS VELOCITY
200
400
600
800
1000
1200
1400
1600
1800
INLET SO2 (ppm)
-------
FIGURE 10
INLET vs. OUTLET GRAIN LOADING
170 MW HORIZONTAL MODULE
3,0-
LO
o
z
g
LU
5
—I
ID
U
1.0- -
0.3- -
0.1- -
0.03--
0.01- -
0.003- -
0.001
O
O
O
o
AVERAGE OPERATING CONDITIONS:
FLUE GAS FLOW = 474,000 SCFM inlet
584,000 SCFM outlet
L/G = 20.6
0.001 0.005 0.01 0.05
OUTLET PARTICULATE LOADING(gr/SCF)
382
-------
water contributed respectively 1.17$ and 0.85$ of the equivalent CaC
fed to the system. The three lime utilization determinations are
summarized as follows:
Method A: CaO utilization (from.S02 removed from gas) = 99-75%
Method B: CaO utilization (from unreacted CaO in sludge and material
flowrates) = 99-^0$
Method C: CaO utilization (from chemical analysis of sludge
independent of material flowrates) = 99.27$
The average utilization for the three methods is 99-^7% which is nearly
stoichiometric.
Power Consumption
The electric power consumed by the Horizontal Scrubber is metered
at the itl6o volt bus and includes transformer losses for voltage
less than itl60 volts, control room power and air conditioning,
scrubber lighting and miscellaneous welding and maintenance power
usage. Kct included are the following usages of power:
A. Reheat steam at U0,000 Ib/hr, itOO°F and 350 psig
B. Service air at ko SCFM, 90 psig
C. Electric power to operate sludge stabilization mixing
equipment (k kW)
The electric power usage was monitored during 556 hours of con-
tinuous operation. The average flue gas flowrate during this period
was it51,000 SCFM. The scrubber power consumption averaged 2.66 MW.
In general, flue gas flowrate per MW varies from station to station.
For the 790 MW Unit Two at Mchave, the flue gas flowrate at full load
is between 1.9 and 2.1 X 106 SCFM (at 60°F and 1 atm). Thus a flowrate
of it51,000 SCFM is equivalent to a range between 170 and 188 MW, and
the power consumption measured corresponds to a range between I.it2
and 1-57$ of the power generated.
Pressure Drop
One important characteristic of a scrubber which contributes a
great deal to both the capital cost and the operating cost is the
pressure drop through the scrubber. The pressure drop is, of course,
primarily a function of the design of the scrubber.
As discussed in last year's EPA symposium, the Horizontal pilot
plant was characterized by a low pressure drop compared with the other
383
-------
pilot plants. Specifically Figure Hi of Ref. 1 indicated that the
pressure drop across the 3000 SCFM Pilot Plant Horizontal Scrubber
itself was 1.0 inches of water while Figure 19 indicated that the total
system pressure drop (at nominal conditions) anticipated for a
it50,000 SCFM Module System was 6 inches of water.
In Figure 11, the total system flue gas pressure drop is plotted
versus gas flowrate at a constant L/G ratio. As may be seen, the
total system pressure drop at ^50,000 SCFM is 5-5 inches of water. This
may be compared to the 6 inches of water previously predicted. The
total system pressure drcp is the differential pressure developed by
the booster fan as required to draw flue gas from the precipitator
outlet duct and drive it through the scrubber, deraister, reheat mix
chamber and outlet ductwork into the stack breeching duct.
The test data in Figure 11 indicates that the total system pres-
sure drop under design operating conditions was approximately 6.0
inches of water. The various scrubber components contributing to this
total are roughly as follows:
Differential
Pressure Inches
Component of Water
Inlet Ductwork 1.0
Scrubbing Chamber 1.0
One Demisting Stage 0.5
Indirect Reheating Chamber 2.5
Outlet Ductwork 1.0
Total 6.0
The pressure drop through the indirect reheating chamber is somewhat
high to insure complete mixing of hot and cold gas streams.
38^
-------
FIGURE 11
SYTEM PRESSURE DROP vs.
FLUE GAS FLOWRATE
170 MW HORIZONTAL MODULE
5--
o
o
o
Cf
o
2
LU
u
LU
g
(JL
CO
O
LU
4-
O/
3--
2--
FOUR STAGES OF SCRUBBING
L/G RATIO = 22 gpm/1000 SCFM
3--
0
23456
SYSTEM PRESSURE DROP
( Inches of Wafer)
385
-------
ACKNOWLEDGMENTS
The experiments reported in this paper relating to the 1 MW and
10 MW scrubbers were sponsored by the Southern California Edison
Company. Construction and Testing of the 1?0 MW scrubber was funded
by the Mohave Project and Navajo Project participants - Arizona Public
Service Company, City of Los Angeles Department of Water and Power,
Nevada Power Company, the Salt River Project, Southern California
Edison Company, Tucson Gas and Electric Company and the United States
Department of the Interior, Bureau of Reclamation with costs
allocated in accordance with their megawatt entitlements in the Mohave
Project and in the Wavajo Project. The Salt River Project serves as
Project Manager for the Navajo Project while Southern California Edison
serves as Project Manager of the Mohave Project as well as Project
Manager for the Test Modules Program.
The design, construction and operation of the Horizontal Module
required the efforts of many individuals and organizations. The
courageous management decision tc build the Horizontal Module 150
times larger than the pilot plant was made by Mr. Tom Morong, Chief
Engineer and Assistant General Manager of the Salt River Project and
Mr. Jack B. Moore, Vice President of the Southern California Edison
Company. Mr. Richard Durning of the Salt River Project made many
contributions, particularly in the design criteria and test planning
area.
Stearns Roger, Inc. performed the engineering design according to
SCE design criteria and also procured the major equipment and provided
an engineer for test support. The Bechtel Power Corporation performed
all construction activity at the Mohave Generating Station. Routine
maintenance and design modification services have been performed
during the Test Program by Jesco, Inc. Truesdail Laboratories has
also supported the test program with field chemists, and gas
s ampling c r ews.
Many Edison personnel have contributed a great deal to the success
of this program; these include Messrs. B. C. Stowers and J. Harvey of
Construction Engineering, Mr. N. J. Dellaven of Generation Engineering,
Messrs T. L. Reed, R. L. Manley and C. Patel of Cost Schedule Engineering,
and Messrs. H. A. Kerry, E. A. Danko, S. D. Sharp, W. C. Martin and
R. B. Rolfe of Edison's R&D Organization. Mr. G. L. Fraser, Station
Superintendent and Mr. R. Young of the Mohave Generating Station Staff
have successfully supervised the operation and maintenance of the
scrubbers. The contributions of the SCE operators and foremen them-
selves cannot be overstated.
Finally, Dr. L. T. Papay, Director of Research and Development
for Southern California Edison Company has served as Chairman of
the Coordinating Committee for the Wavajo/Mohave Test Modules
Program. His encouragement and technical contributions are sincerely
appreciated.
386
-------
REFLRJM'JCES
Alexander Weir, Jr. and Lawrence T. Papay "Scrubbing Experiments
at the Mohave Generating Station" U. S. Environmental Protection
Agency's Flue Gas Desulfurization'Symposium, New Orleans, Louisiana,
May 111, 1973.
387
-------
OPERATIONAL STATUS AND PERFORMANCE OF THE LOUISVILE
FGD SYSTEM AT THE PADDY'S RUN STATION7
Mr. Robert P. Van Ness, Manager Environmental Affairs
Louisville Gas And Electric Company
November 4-7, 1974
EPA, FGD Symposium, Atlanta, Georgia
ABSTRACT
Operations at the Closed Loop Full Scale 65M'.\' SU7 System com-
menced in April 1973 and has operated in limited and extended modes
to date. Absolute reliability of the system has not been esta-
blished; however, operations to date have been extremely satis-
factory and gratifying. All data suggests high attainments of
S0_ and particulate matter reouirements without operational diffi-
culties such as scaling, plugging, and corrosion or errosion pro-
blems. High degree of operational on time at varying loads with
various levels of sulfur coal input. Extremely high removal rates
of efficiencies of SO- in 1974 with inoculations of varying amounts
of magnesium, Stoichiometric requirements of the system have been
excellent.
389
-------
OPERATIONAL STATUS AND PERFORMANCE OF THE LOUISVILLE
FGD SYSTEM AT THE PADDY'S RUN STATION
1 have been requested to present to this Symposium a status report
on the operation of the Paddy's Run Unit #6 S0? removal system to date
and the current status of the compliance program proposed to meet local
air pollution standards.
Our Ca(OH)? (carbide lime) slurry tail end wet scrubbing system
has been described in detail by both Combustion Engineering Co. and
Louisville Gas & Electric Co. in previous statements and in publica-
tions. So, in the interest of avoiding repetition, L will assume that
everyone is basically familiar with the installation. Lf not and more
information is desired, I would recommend that you obtain a copy of a
recent paper presented by Messers. Martin, Plumley and Minor at the
National Coal Association Symposium on "Coal and the Environment" in
Louisville, Ky. on October 23. The paper entitled "The C, E. Lime Wet
Scrubbing Process from Concept to Commercial Operation" covers in con-
siderable depth the data, concepts, flow conditions and charts pertaining
to the system and its operation.
As a start, flow diagrams of this system are enclosed (Exihibit A),
for your edification which basically shows the equipment and overall
design flows and enumerates the stages of control of the calcium bisul-
fite and calcium sulfite reactions using a waste product known as carbide
lime additive. Basically, the concept has been to remove the S09 from
the flue gases by the use of a liquid to gas contactor such that the S0?
is removed in a soluable form (calcium bisulfite) and Lo control the
chemistry so that the solid salts of calcium sulfite and calcium sulfato
are not formed in the scrubber. The former is a critical item in that ir.
tends to come out of solution at a pll of about 6.2. The sulfate formation
is an oxidation reaction which must be controlled by various techniques
such as pH, time, liquid and gas flows, and seed crystals. The control
concept and equipment have produced very satisfactory results.
390
-------
In our opinion the overall reliability and efficiency of this SO-
removal system to date has been good. Starting in April of last year the
unit was able to operate 70% of the time that the boiler was available for
service through 9 months of 1973 and an availability of 98% during the last
4 months of the year.
The low 70% availability factor was for the most part due to various
modifications which we chose to incorporate in the system during the early
periods of operation. These changes include:
(1) Modifications to the carbide handling system by adding
a Rietz Disintegrater which was installed to uniformly
size the carbide lime slurry before it enters the system.
(2) Modifications to the thickener circuit to improve settl-
ing characteristics of the calcium sulfite crystals by
the addition of a flocculating system. The flocculents
used in this system were Dow A-23 and Betz 1100. The
best results were obtained with the latter.
(3) Improvements to the top water washer in the demister
system which was at the start of operations found to
be inadequate.
(4) Various piping and pH control changes.
Generally, the system has operated in a satisfactory manner and very
few problems have been encountered to date. No major problems of scaling
or plugging have been encountered in this 9 month period. Once we observed
a minor build up of calcium sulfite scale that occurred when the scrubber
was allowed inadvertently, in one instance, to run at a pH above the critical
point of 6.2. This scale was easily removed by dissolving the crystals by
lowering the overall pH of the system and no further build up or scaling has
occurred. During these operations the SO removal efficiencies averaged from
80% to 90%. The best operations were accomplished on a general pH input of
approximately 8. This 8 pH area of control was selected to cover fluctuations
in SO concentration in the flue gas when the boiler operated on varying sulfur
levels in the coal. The overall range of sulfur in the coal during these test
periods were in the range of 1.5% to something greater than 4%. The particu-
late removal efficiency throughout these varying operating conditions fell
well below the standard of .05 grains/Scf. Most of the data was in the range
391
-------
of .02 and .035 grains, thus, producing a particulate efficiency well in
excess of 99%. The inlet gas to the scrubber contained a particulate load-
ing of about .2 to ,4 grains.
The stoichiometric requirements of the system appears to approach
100% based on the total SO,, removed or approximately 80 to 90 percent
depending on the inlet SO- concentrations in the flue gas to the scrubber
system.
As you are quite aware, this system has been operated in a closed
loop mode and has never been out of that mode during any of our operations
to date. This requirement of the system obviously only allows a small
period of time to wash the demisters, since excessive use of wash water
would upset the water balance and thus open the loop. Generally speaking,
this washing operation at approximately 200 GPM of river water, is utilized
8 to 12 minutes every 8 hours. In the overall water balance, the only water
that is removed from the system is that water which goes out with the re-
heated flue gas to protect the scrubber fan, which is in the circuit between
the scrubber and the stack, and the water that is removed from the system as
sludge from the filter circuit. The moisture content of the filter cake,
which for our convenience is removed from the scrubber area by trucks, rather
than some other method, approaches 55 to 65 percent water. This sludge is
conveyed to a borrow pit approximately a mile from the plant where the sludge
is mixed haphazardly with flyash to produce a rather solid landfill. In order
that no one misunderstands, this is not in our opinion the best environmentally
acceptable approach to a chemical fixation reaction. We feel that the sludge
coming off either the drum filters or the higher percentage moisture content
sludge from the thickener should be mixed with the dry flyash in the immediate
vicinty of the plant where lime or other types of additives are mixed so that
the resulting product can be pumped to the storage area. The water and the
solids will then set-up in time to hopefully produce an environmentally accept-
able solid landfill. By so doing, the water is chemically bonded in the new
resulting solids thus protecting the ground water and eliminating uhe need to
transport water back into the scrubber system. Much work remains in the deve-
392
-------
lopement of an economically feasible and environmentally acceptable method of
waste disposal.
When the unit is operating at full load the overall electrical load
demand on the system is basically equivalent to approximately 2.3% of the
output of the boiler. This is accomplished by a relatively low liquid to
gas ratio in the entire system of 30 plus (15 to 18 per bed) and a pressure
drop on the overall system of 14 to 15 inches. In future installations our
efforts will be directed toward lower capital and operating costs.
Obviously, from the above discussion we have operated the system to
date in a rather trouble free mode. No major problems have existed or
have occurred which have been identified as general problem areas such as
corrosion, errosion, scaling, plugging, and maintenance difficulties. For
the most part, the dernister banks from a visual observation appear as clean
as they were when the system first went into operation last year.
The system as you probably are aware has a bypass feature which has
worked extremely well under varying conditions. In fact, other than total
bypassing of the system, it enables a proportioning of the gases for recycle
as the boiler load varies from time to time, this provides extensive turn-
down capabilities which gives the system almost unlimited control. To be
sure the use of recycle increases power requirements on lower load condi-
tions.
After the operations were temporarily suspended in late December 1973,
a complete inspection of the system was made and very few areas of diffi-
culties were observed. During this past summer, we used the system as the
boiler requirements occurred. At no time during this period did any diffi-
culties occvr in the system. During this running period, we changed the
operating mode to a degree in that a waste product, high in MgO, was inocu-
lated into the carbide lime slurry to evaluate the effects on the system
with increased levels of MgO. This of course, was done because MgO appears
to have a greater effect on SO removal rates than do the basic calcium hy-
droxides. The main reasons for these inoculations was to ascertain if higher
393
-------
levels of SO removals could be accomplished. The results were quite interest-
ing. For the most part, the levels of SO in the flue gas to the scrubber
averaged approximately 2,000 ppm and the exit SO concentrations maintained
themselves well below 50 ppm. Generally, we had no trouble in operating in
this mode.
Overall, from this paper, it is obvious that we have established some
very encouraging results, however, it would be impossible for us to state
that absolute reliability of the system has been established.
Now to other areas: Last year as required by the Clean Air Act of
1970, we submitted a complete compliance schedule for S0_ removal systems
for units requiring control at our plants known as Cane Run and Mill Creek.
This plan basically proposed that the Primary Ambient Standard would be
attained by July 1, 1977 and the Secondary Standard by July 1, 1979 or
somewhat later. Obviously, this attempt did not satisfy the Clean Air Act
since it exceeded the guidelines in time, even though, we felt it to be a
very optimistic schedule.
After considerable time and many lengthy discussions with various
regulatory bodies, the first phase of the plan was agreed upon and a
variance was granted us by the local air pollution board since the time
of completion extended beyond the limit of local regulations. The en-
closed exhibit is a schedule of the installations included in the approved
compliance plan. With this agreement in hand the next order of business
was to request authority to expend funds by obtaining a "Certificate of
Convience and Necessity" from the Public Service Commission of Kentucky.
Initially the KPSC had granted a certificate on a new source boiler of
425 MW at Mill Creek Unit #3 which is scheduled for operations in early
1977. The next three units in the plan were units which would be retro-
fits, known as, Cane Run #4, #5 and #6. Initially, Cane Run #4 was re-
quested followed by a request for Cane Run #5. In late 1973 a formal
hearing was held on Cane Run #4 with a decision by the Commission on
March 20 of this year. This order granted the Certificate of Convience
and Necessity for Cane Run #4 but delayed any further consideration on
394
-------
other units until Cane Run #4 had been installed and had successfully
operated for one year. Thus, the dilemma of non-compliance of the plan
until some undetermined time frame beyond the sacred date of July 1, 1977.
In order to better clarify our position, we requested the KPSC for a
rehearing of their order. At this rehearing in May and June, intervenors
testified. They included the Environmental Protection Agency, the Edison
Electric Institute, The Mayor of Louisville, the County Judge of Jefferson
County, the Air Pollution Control District of Jefferson County, the Kentucky
Department of Natural Resources and Environmental Protection, and Action for
Clean Air (a local environmental group). As of this date, in late September,
when this report is being written, we have had no ruling by the KPSC. However,
in the meantime, as you can see from the approved compliance plan as noted in
exhibit B we have exceeded some of the progress dates in the actual plan. In
short, we have an enforcement order from the Air Pollution Control District of
Jefferson County to proceed, but can not proceed until Public Service Commission
approval is received.
395
-------
EXHIBIT A
Cat t* Hoc*
Cm ra
«4 Ho«
u>
JTO.OOO ei™>
Cl 3S5 f —>,
Cc»
p^
Co*nmintiror I
-O-J
J.IOO c»'
pM.4 ??OO{
BlOO Cd'n
srt.4 »ov
SC'dM'
«•*'/. 0«
Sc/^brf
drain pifre
T« Mxwm film
400 r<» oaltf (lolol)
pH*6 or ft'Cftr
5H.6J
kkh
To
{(Wltr
...
1
•«' ,M.
J
K5-J.SU.
„„„
4030$e«t/tc»gbB«r pH«93
95 40SO«p"./,c.vbb,r
yJoVr^w-NH' ?' "*" '*" 2«0 ,pm ltoro.1
* M'0«m COM') | . |
A
*
7*
pH *u:ifoc« ^ •"
o
?'*'••" i *00 «»m (IOIOII
' 4SO rip «acK
L— EMIxtnl
^— O.xllo- «
4600 lt>/>" dry Co (OH),
c^ffl H
Pump
50 1 ioi'0/o«<
C-E AIR-QUALITY CONTROL SYSTEM
•LOUISVILLE GAS & ELECTRIC COMPAKY
PADDY'S RUN NO. 6 SCHEMATIC
FLOW DIAGRAM
-------
EXHIBIT A
Overflow and effluent water
Clarifier discharge
Voouum filter Ihp
Flapper, 2 hp
Agnator, l£ hp
Connection
for air flow
Conveyor, l-jphp
To atmosphere
Overflow weir ond drnin
Underflow pump,5 hp
pH 6.5,25% slurry
80 gpm (total)
C-E AIR-QUALITY CONTROL SYSTEM
LOUISVILLE GAS & ELECTRIC COMPANY
PADDY'S RUN NO. 6 VACUUM-FILTER SCHEMATIC
-------
EXHIBIT
Cane Run #4
Cane Run #5
Cane Run #6
GJ
°°MU1 Creek #3 (Approx.)
Louisville Gas and Electric Company
Appendix VI
Sulfur Dioxide Removal System Installation
Proposed Construction Schedule*
Completion of
Order
Equipment
5-20-74
5-20-74
9- 1-74
3-27-74
Start Construction
9-20-74
12- 1-74
2- 1-75
6- 1-75
Construction
Completed
3-1-76
8-1-76
4-1-77
4-1-77
Start-up Date
4-1-76
9-1-76
5-1-77
5-1-77
Start-up Period
& Compliance By
6-1-76
11-1-76
7-1-77
7-1-77
The above schedule is predicated on manufacturers' current delivery estimates and does not consider
potential time loss due to strikes, material shortages or other unforeseen developments beyond the control of the
Company.
*Revised as of 3/21/74
-------
DISPOSAL OF BY-PRODUCTS FROM
NON-REGENERABLE FLUE GAS
DESULFURIZATION SYSTEMS
A STATUS REPORT
by
J, Rossoff
R. C. Rossi
L. J. Bornstein
Environment and Urban Division
The Aerospace Corporation
El Segundo, California
and
J. W. Jones
Control Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
Presented at the
EPA Control Systems Laboratory
Symposium on Flue Gas Desulfurization
Atlanta, Georgia
November 4-7, 1974
399
-------
This paper has been reviewed by the Environmental Protection
Agency and approved for presentation. Approval does not signify
that the contents necessarily reflect the views and policies of the
Agency, nor does mention of trade names, commercial products, or
commercial processes constitute endorsement or recommendation
for use.
CONTENTS
re
1.0 Introduction 402
2.0 Summary 403
3.0 Potential Water Pollution Problems 406
4.0 Potential Land Reclamation Problems 414
5.0 Potential Disposal Problem Solutions 422
6.0 Disposal Cost Estimates 430
7.0 Shawnee Sludge Disposal Field Demonstration 441
8. 0 References 442
400
-------
DISPOSAL OF BY-PRODUCTS FROM
NON-REGENERABLE FLUE GAS DESULFURIZATION SYSTEMS
A STATUS. REPORT
by
J. Rossoff
R. C. Rossi
L. J. Bornstein
Environment and Urban Division
The Aerospace Corporation
El Segundo, California
and
J. W. Jones
Control Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
ABSTRACT
A summary of an on-going Aerospace Corporation study for the EPA's
Control Systems Laboratory to characterize flue gas desulfurization sludges
and to assess methods for their environmentally sound disposal is given.
The sludges studied were selected from non-regenerable scrubber systems
using lime or limestone as the absorbent at several Eastern and Western
coal-burning power plants. Chemical analyses have shown that untreated
sludge liquors contained mercury, selenium, boron, chloride, sulfate, and
total dissolved solids considerably in excess of water quality criteria (most
of the constituents of concern, which without scrubbing are discharged in
the fly ash and flue gas, originate in the coal). Since soil attenuation of
selenium, boron, and chloride is known to be ineffective, and since the
remaining sludge liquor constituents prevent direct discharge to streams,
the need for environmental control of sludge disposal sites is strongly
indicated. In addition, the water-retentive properties of raw (untreated)
sludge oresent land reclamation difficulties. Currently available disposal
methods include landfilling of sludge (which has been chemically treated
for strength and the prevention of leaching of undesirable constituents to
water suoplies), and disposal of untreated sludge in oonds (lined with highly
impermeable materials). Estimates from various sources indicate disposal
costs of from $2 to $9/ton for chemical fixation and from $Z. 50 to $4. 50/ton
for ponding, excluding oossible subsequent pond reclamation costs. An
average disposal cost of $5/ton is equated to 1. 1 mills/kwhr. Other less
developed disposal techniques include drying, and conversion to gypsum.
401
-------
DISPOSAL OF BY-PRODUCTS FROM NON-REGENERABLE
FLUE GAS DESULFURIZATION SYSTEMS - A STATUS REPORT
1.0 INTRODUCTION
The Aerospace Corporation sludge disposal study discussed in this paper
is in its second year of a two-year program which began in November
1972 under EPA Contract 68-02-1010. The results of the first year's
1
work is described in detail in an initial report . This paper up-dates
that work in a summary form.
Sludge samples have been obtained from the EPA/TVA Shawnee limestone
and lime prototype scrubbers, the Southern California Edison Mohave
Station limestone pilot scale scrubber, and the General Motors Parma
facility double alkali scrubber. The major portion of the analyses of all
these samples has been completed except for the double alkali samples.
Additionally, technical and economic assessments are being made of
power plant and commercial processor operations being conducted to
develop environmentally sound sludge disposal techniques. Chemically
conditioned sludge samples from some of these sources, namely, Chemfix,
IU Conversion Systems, and Duquesne Light Company which uses a Dravo
Corporation additive have been received and are being analyzed. Reviews
are being made continually of state and federal standards which govern
water quality and the potential land disposal of sludges. Present plans
by the EPA are to expand the Aerospace study to include the sampling of
sludges from four other scrubber systems to increase the data base being
established in this study.
A sludge disposal field demonstration program of untreated and treated
sludges is being initiated by the EPA at the TVA Shawnee site in Paducah,
Kentucky. Analyses of sludges, leachate, and ground water from that
site will be conducted during the next two years in the Aerospace program
and will be integrated with the Aerospace study data and other data as
appropriate.
402
-------
2.0 SUMMARY
In this study, sludge samples have been analyzed from three scrubbers
involving lime/Eastern coal, limestone/Eastern coal and limestone/
Western coal. Additionally, two sludges from other scrubbers (limestone/
Western coal and lime/Eastern coal) were analyzed in another study for
the U.S. EPA National Research Center, Corvallis, Oregon (EPA Research
Grant R802853-01-1) and assessed as a part of the total program. Chemical
characterizations of the sludge liquors were performed to identify the con-
centrations of nine trace metals of interest and seven major soluble species,
as well as the pH and total dissolved solids (TDS) for each of the samples
collected. (Input materials such as coal, fly ash, make-up water, lime
and limestone were also characterized to identify source of constituents
found in the sludges). A preliminary assessment was made of these
samples with regard to any potential environmental problem that might
be posed if these sludge liquors entered water supplies. This assessment
was made by comparing the chemical characterizations with the criteria
established by the states, the U.S. Public Health Services Drinking Water
Standards - 1962, and the EPA Proposed Public Water Supply Intake Criteria
October 1973, The latter criteria were used in the final analysis at this
stage of the study.
In assessing the potential impact of trace metals, it was found that in each
of the sludge liquors at least one of the following trace metals exceeded
the EPA prooosed criteria: arsenic, cadmium, chromium, lead, mercury,
and selenium. Except for mercury and selenium, these trace metals
exceeded the criteria by not more than a factor of 5. Mercury and selenium
exceeded the criteria in each of the sludges analyzed by more than an
order of magnitude. Comparing the concentrations of the major soluble
soecies with the criteria, it was found that particular excesses exist
for chloride, sulfate and TDS for each of the samples analyzed. To date,
only one sample has been analyzed for boron, which was also found to be
in excess. It should be noted that most of the constituents of concern,
which without scrubbing are discharged in the fly ash and flue gas,
originate in the coal.
From the assessment just described, it appears that in all of the power
plant sludge liquors analyzed, water quality criteria are appreciably
exceeded for mercury, selenium, boron, chloride, sulfate and TDS.
Attenuation of chemical species by soil is widely accepted in many land
403
-------
disposal oractices; however, attenuation of three of the species discussed
above, namely, selenium, boron and chloride is known to be ineffective.
Therefore, the concentrations of these three species strongly indicate
that environmentally sound techniques such as lined ponds or chemical
fixation are needed for scrubber sludge disposal.
Detailed physical characterizations of various samples have also been per-
formed under the Aerospace Corporation study to determine properties
such as: ease of dewatering/settling, bulk density, viscosity, load bearing
strength and oermeability. Some significant results relating to land-use
capabilities are as follows: scrubber sludges are highly water-retentive
and have little or no bearing strength as produced; they are thixotropic;
they can be dewatered and/or compacted to produce appreciable bearing
strength and low coefficients of permeability, but after being dewatered,
they easily rewet in the presence of water.
A sulfite sludge can be dewatered in a thickener to between 35 and 45
percent solids and will not increase its solids content significantly by
settling; however, when allowed to drain freely, its solids content will
increase to about 50 percent. A filter or centrifuge will produce a solids
content of about 50-55 percent. In contrast a sulfate sludge which can be
dewatered to 35-45 percent solids in a thickener, will increase its solids
content to about 65 percent upon settling, and when freely drained or cen-
trifuged, will dewater to about 75 percent solids. It can be dewatered to
about 80 percent solids by filtration.
A better comparison of physical properties can be made when sulfite and
sulfate sludges containing equivalent moisture content are compared. For
example, at 65 percent solids content, sludges will support personnel,
and at solids content greater than 70 percent, sludges will support heavy
equipment. A settled sulfate sludge can be easily compacted such that it
will readily support personnel and equipment, but sulfite sludges by com-
parison are much more difficult to dewater to the required degree of
dryness for compaction.
From the preceding discussion, it appears that raw sludges or raw sludge
ponds which have been dewatered could be used for structural purposes.
* Sulfite and sulfate sludges described herein apply specifically to TVA
Shawnee and S.C. Edison Mohave pilot plant sludges defined in Figures
1, 2 and 3.
404
-------
However, the cost of dewatering the sludge and maintaining the site at
the necessary degree of dryness is strongly dependent on the raw sludge
properties, and may or may not compare favorably with the cost of an
alternative disposal technique such as chemical fixation. Data from
chemical fixation processors indicate that treated sludge quickly attains
a compressive strength above the limit for hard clay or compacted fine
sand as defined in the Uniform Building Code. This conditioned sludge
would have the capability to make a disposal site reclaimable for either
structural or recreational use.
Based on the sludge properties just described, the best practical methods
currently available for sludge disposal are ponding, and chemical fixation
for landfilling. Ponding which may be environmentally sound with regard
to water quality consists of lining the containment basin with an impervious
material such as clay, plastic or rubber to prevent the seepage of sludge
liquors to water supplies. There are certain problems attendant to this
approach which are still being considered; these include the inability of
the liners to last indefinitely and the difficulty of reclaiming the land after
the site is abandoned. A more technically sound approach is that offered
by industry in which the material is chemically conditioned so that leaching
of undesirable constituents is minimized and the material can be easily-
used to oroduce a structural landfill.
Preliminary cost estimates of these disposal alternatives, assuming a
sludge consisting of 50 percent solids, place the cost of ponding in the
range of $2. 50 to $4. 50 per ton. Some additional cost to provide per-
manent environmental protection or land reclamation if desired may be
incurred at the end of the service life of the pond. The cost of disposal
by fixation currently is quoted in the range of $2. 00 to $6. 00 per ton by
the fixation processors and approximately $7. 50 to $9.00 per ton by power
companies performing their own disposal task. The cost of ponding
appears to be about half that of fixation; however, ponding is not neces-
sarily a permanent disposal technique.
A disposal cost of $2. 50 per ton, in terms of power produced, is equivalent
to 0. 56 mills /kwhr; $5. 00 per ton is equivalent to 1 . 1 2 mills /kwhr. This
example is based on the use of a limestone scrubber, 85 percent
All costs quoted herein are on a wet sludge basis.
405
-------
SO removal, 1.2 CaCCX/SCX mole ratio, 50 percent solids sludge plus
ash at a station burning coal that has 3 percent sulfur and 12 percent ash,
at 0. 88 Ib/kwhr.
3.0 POTENTIAL, WATER POLLUTION PROBLEMS
3. i WATER QUALITY
The disposal of scrubber sludges presents potential water pollution and
land reclamation Problems because of soluble species and dewatering
difficulties. The measure of those problems is defined by authoritative
water quality standards and solid waste disposal regulations, or by judg-
ment of regulatory officials when standards are nonapplicable or non-
existent. The most nearly applicable water quality criteria that exist
for determining the environmental acceptability of sludge disposal are
given in Table 1. This table shows the water quality criteria of the U.S.
Public Health Service (USPHS) - 1962 Drinking Water Standards , the
1973 State of Illinois criteria (an example of one of the more compre-
hensive of all state criteria) for both public and food processing water
supply (PWS) and water effluent discharge standards - and the 1973 EPA
4
Proposed Criteria for Water Quality .
Regarding the quality of surface waters, it should be noted that all states,
in conformance with Federal EPA guidelines, do not consider dilution of
an effluent as an acceptable method of treating wastes in order to meet
the effluent standards. Regarding the quality of underground waters, in
general, states which do not have specific regulations concerning the
quality of underground water apply stream or drinking water standards
to underground water for lack of more definitive regulations. One state
which does regulate underground waters is Illinois, which requires that
underground waters that are a present or potential source of water for
public or food processing supply shall also meet the standards for those
supplies.
Standards for the regulation of solid waste disposal, as related to water
quality, are much less specific than water quality criteria. Regulations
ordinarily limit the disposal of toxic, hazardous, and harmful substances
by requiring that those materials not be disposed of on the ground, or
underground, without the approval of a designated authority such as the
State Department of Health, the State Health Commissioner, County
Health Commissioner, or the State Department of Commerce. The
406
-------
TABLE 1. WATER QUALITY CRITERIA
CONCENTRATION, ppm
ARSENIC
BORON
CADMIUM
CHROMIUM (total)
COPPER
LEAD
MERCURY
SELENIUM
ZINC
CHLORIDE
FLUORIDE
SULFATE
TDS
PH
USPHS - 1962
DRINKING
WATER
STANDARDS
0.05
--
0.01
--
1.0
0.05
--
0.01
5
250
2. 4 AT 50° F
1.4 AT 90° F
250
500
--
ILLINOIS - 1973
PUBLIC AND FOOD0
PROCESSING
WATER SUPPLY
0.01
1.0
0.01
--
0.02
0.05
0. 0005
0.01
1.0
250
1.4
250
500
6.5-9.0
EFFLUENTb
DISCHARGE
STANDARDS
0.25
--
0.15
--
1.0
0.1
0. 0005
1.0
1.0
--
2.5
3500C
5 - 10
EPA PROPOSED
PUBLIC WATER
SUPPLY INTAKE
OCT 1973
0.1
1.0
0.01
0.05
1.0
0.05
0.002
0.01
5.0
250
--
250
NO LIMITd
5.0-9.0
Point at which water is withdrawn for treatment and distribution
Based upon 24-hour composite sample. No contaminant shall at any time exceed five (5) times
the concentration limit
cNot more than 750 ppm above background concentration limits unless caused by pollution
abatement practices
No limit" indicates that insufficient data existed for prescribing limits
-------
survey undertaken to date indicates that those offices, when judging the
disposal of sludges, base their judgment on sludge quality as related to
the local or state water quality criteria'.
The limitations imposed on sludge disposal have been further intensified
by the recent oassage of legislation and the potential passage of others.
For example, the Federal Water Pollution Control Act Amendments of
197Z essentially require the revision and re-aoproval of all state water
quality standards: to change applicability from interstate waters to all
waters, to prohibit or limit direct discharges of pollutants to streams,
to limit subsurface disposal, and to include Federal E PA-prescribed
criteria for specific toxic elements. Additionally, the federal drinking
water standards are being revised, and legislation is in process to develop
new limitations for the land disposal of solid wastes.
3.2 SLUDGE SAMPLING
Sludge samples analyzed to date were taken from three scrubbers at two
different power plants representing both Eastern and Western coals and
the lime and limestone processes. These are: EPA/TVA Shawnee TCA
scrubber with limestone, EPA/TVA Shawnee Venturi scrubber with lime,
and Southern California Edison (SCE) Mohave TCA pilot scrubber with
limestone. Additionally, sludge liquors from two other plants (an Eastern
utility plant lime scrubber, and a Western utility plant limestone scrubber)
have been characterized in a separate study for the EPA NRC, Corvallis,
Oregon, and assessed along with the data determined in this program.
Samoles were taken at various points within the scrubbing system so that
the fate of trace elements and major species of interest could be "observed1'
as they passed through the system. In addition, samples of process
ingredients -- coal, process water, fly ash, lime, and limestone -- were
taken to identify the source of the elements entering the system. Samples
were taken over a short period of time in an attempt to minimize the
effect of system fluctuations.
3.3 CHEMICAL CHARACTERIZATION
Chemical analysis of the sludge samples consisted of the determination of:
408
-------
a) magnesium and trace metals by atomic absorption spectrophotometry,
b) calcium by wet chemical methods, c) sulfate by turbidimetry, and
d) chloride, fluoride, and sulfite by specific ion electrode. Results of
these analyses are presented in Tables 2 and 3 for the various power
plant sludges analyzed. In each case (except for Plant A), the analysis
is for the liquor fraction of the sludge representing the condition in which
the sludge is likely to be disposed, e.g., clarifier underflow, filter cake
or centrifuge cake (Plant A liquor is scrubber effluent and does not rep-
resent actual disposal material).
Since these analyses represent sludge liquors from a variety of scrubber
systems, certain qualifications must be made for accurate interpretations
of the data. Four principal system conditions influence the concentrations
of soluble species in system liquors. The first is the "tightness" of the
scrubber system liquor loop. The tighter the system, the less liquor is
lost with the discharge from the recirculation system to the disposal
area. Consequently, less make-up water is required and less dilution
is oossible. An example of the effects of this condition can be seen in
Table 3 where it is shown that the concentrations of chlorides and TDS
in the SCE Mohave liquor are at least an order of magnitude greater than
those in the Plant A liquor. The Mohave pilot scrubber had a tightly closed
loop system, whereas the Plant A scrubber was running open loop and
consequently experienced a high degree of liquor dilution.
If it were theoretically possible to dewater the sludge solids completely
and return all liquor to the system, no make-up water would be required
for water balance except to make-up for evaporation. Soluble species
would build up in concentration until they reach a saturation level resulting
in precipitation and subsequent removal with the solids. From the point of
view of minimizing the volume of waste material, an absolute separation of
solid and liquid might aopear desirable, but from the point of view of a
reliable system operation, an excessively tight liquor loop could present
major difficulties in the operation of the system. In practice, disposal
of liquor with solids is not only necessary but unavoidable.
The second principal system condition that can affect the concentration of
soluble species is related to the operational differences in scrubbing flue
409
-------
TABLE 2. TRACE METALS IN SLUDGE LIQUORS*
(concentration in ppm )
SCRUBBER
•SIZE, Mw
•PERCENT
SOLIDS IN
DISCHARGE
ARSENIC
BERYLLIUM
CADMIUM
CHROMIUM (total)
COPPER
LEAD
MERCURY
SELENIUM
ZINC
PH
TVA SHAWNEE
LIME
10
50-55
0,02
< 0.002
0.10
0.03
< 0.002
0.05
<0.001
<0.02
0.08
9
TVA SHAWNEE
LIMESTONE
10
35-40
0.20
0.01
0.005
0.15
0.02
0.1
0.06
0.30
0.30
8
SCE MOHAVE
LIMESTONE
<1
75-80
0.03
<0. 001
0.05
< 0.005
0.08
0.01
0.0012
0. 12
0.12
7
PLANT Ac
LIMESTONE
(western coal)
>100
15
< 0.004
0.18
0.01
0.21
0.20
0.02
0.12
2.5
0.12
4
PLANT B
LIME
{eastern coal)
>100
35-40d
0.085
0.012
0.023
0.040
0.048
0. 18
0.045
0.80
0.09
9
WATER QUALITY
CRITERIA
EPA PROPOSED
PUBLIC WATER
SUPPLY INTAKE
OCT 1973
0.1
--
0.01
0.05
1.0
0.05
0.002
0.01
5.0
5 TO 9
All sludge liquor samples, except those from SCE Mohave, contain significant amounts of fly ash
Equivalent to mg/l
cPlant A sample is scrubber effluent and does not represent actual disposal material
An undertermined amount of clarifier overflow liquor is discharged from this system,
making the effective percent solids in discharge somewhat lower than value shown
-------
TABLE 3. MAJOR SPECIES IN SLUDGE LIQUORS'
(concentration in ppmb)
SCRUBBER
•SIZE, Mw
•PERCENT
SOLIDS IN
DISCHARGE
CALCIUM
MAGNESIUM
BORON
CHLORIDE
FLUORIDE
SULFATE
SULFITE
TDSf
TVA SHAWNEE
LIME
10
50-55
2520
25
40.8
5000
3.3
800
0.9
9000
TVA SHAWNEE
LIMESTONE
10
35-40
1600
600
_e
2500
3,4
2000
110
7000
SCE MOHAVE
LIMESTONE
<1
75-80
1400
_e
_e
30, 000
3.1
2500
<0. 1
70, 000
PLANT Ac
LIMESTONE
(western coal)
>100
15
700
6
_e
1400
0.6
1150
<0. 7
3500
PLANT B
LIME
(eastern coal)
>100
35-40d
1400
410
_e
2700
2.6
2250
20
7000
WATER QUALITY
CRITERIA
EPA PROPOSED
PUBLIC WATER
SUPPLY INTAKE
OCT 1973
--
__
1.0
250 .
--
250
__
NO LIMIT9
All sludge liquor samples, except those from SCE Mohave, contain significant amounts of fly ash
K
Equivalent to mg/l
cPlant A sample is scrubber effluent and does not represent actual disposal material
An undertermined amount of clarifier overflow liquor is discharged from this system, making
the effective percent solids in discharge somewhat lower than value shown
'To be determined
f
Includes other soluble species (e.g., sodium and potassium)
a"No Limit" indicates that insufficient data existed for prescribing limits
-------
gas from Eastern and Western coals. In all cases, the solids content of
the waste bleed stream from the scrubber loop is typically 10 percent
^solids contents from 8 to 1 5 percent have been measured and appear to
depend on the presence of fly ash in the liquor and system design variables
associated with relative liquor holding capacity). However, for equiva-
lent system operation and equivalent sulfur scrubbing efficiency, the
sludge production rate (and therefore the sludge bleed rate, excluding
ash) from the scrubber recirculation loop for low sulfur Western coal is
considerably less than that for a high sulfur Eastern coal. Therefore,
less occluded liquor would be bled from the Western system than the
Eastern system. Since absorption of soluble species takes place with
each liquor pass through the scrubber, and since the scrubber liquor
"residence time" is greater in the Western coal system, the concentra-
tion of soluble species is greater. The ratio of the lowest Western coal
sulfur content to the highest Eastern coal concentrations analyzed and
reported here is 1:8 which implies that the scrubber loop liquor residence
time is about 8 times longer in this low sulfur system (see TDS for
Mohave pilot and Shawnee liquors in Table 3). This is an extreme condi-
tion which compares an Eastern high sulfur coal sludge which includes
fly ash, with a Western coal sludge that contains an insignificant amount
of fly ash. If fly ash had been collected along with SO? in the Mohave
scrubber, this effect of sludge liquor extended residence time would not
have been as pronounced.
The third principal system condition affecting the concentration of soluble
species pertains to the system design for collection of fly ash. Most of
the trace metals in the sludge originate from fly ash simultaneously
scrubbed from the flue gas; therefore, the scrubber systems that do not
collect fly ash should be expected to have significantly lower concentra-
tions of trace metals than those systems that do collect fly ash.
The fourth principal, system condition affecting the concentration of soluble
species is the pH at which the system operates. For example, the lime
scrubbing orocess operates at a higher bleed pH than the limestone process.
412
-------
As a consequence, the soluble element content in lime sludge liquors
appears to be consistently lower than the same element content in lime-
stone scrubbing systems. This trend can be most easily seen when
comparing the lime and limestone systems at Shawnee where input flue
gases are generated from the same boiler (see Table 2). In each case,
the concentration of trace elements in the lime system is less than that
in the limestone system. These data suggest that the more environmentally
acceotable sludges (from the point of view of soluble trace metals in sludge
liquors) are those disposed of at the high pH more typical of lime scrubbing
systems.
Preliminary correlation between the soluble element content of the sludges
and their concentrations in coal has been made. Arsenic, lead, and
mercury are found in higher concentrations in Eastern coals than in
Western coals, and thei-r concentration in Eastern power plant sludges
generally is also higher. Similarly, Western coals have higher concentra-
tions of copper and selenium and these elements are generally found in
greater concentrations in Western power plant sludges.
3.4 ASSESSMENT
The environmental acceptability of sulfur sludge liquors can be assessed
by the comparison of the analyses determined on actual sludges presented
in Tables 2 and 3 with the water quality criteria of various regulatory
agencies presented in Table 1. For the trace metals (see Table 2), all
elements analyzed except copper and zinc exceed water quality criteria
in at least one power plant. The remaining trace elements, except mercury
and selenium, exceed the criteria by not more than a factor of five, with
one exception (cadmium - Shawnee lime). Mercury and selenium exceed
these criteria for each of the power plants, and the excess is greater than
an order of magnitude.
Comparing Tables 1 and 3 for major soluble species, it is seen that water
quality criteria are surpassed in several cases. Particular excesses
exist for chloride, sulfate and total dissolved solids for each of the power
plants. Boron is in excess based on the one sample analyzed so far. The
413
-------
values for sulfate ions for any given plant will not vary too widely from
those given here because they are controlled by the chemistry of the
scrubbing system. On the other hand, the chloride ion and total dissolved
solids values can vary by an order of magnitude or more and are depen-
dent on factors such as "tightness" of the looo, coal composition, and
sludge bleed rate.
Summarizing the above comparisons of trace metals and soluble species
against water quality criteria, in all of sludge liquors analyzed, water
quality criteria were appreciably exceeded for mercuy, selenium, boron,
chloride, sulfate and TDS. It should be noted that most of the constituents
of concern, which without scrubbing are discharged in the fly ash and flue
gas, originate in the coal. Attenuation of chemical species by soils is
widely accepted in many disposal practices. However, soil attenuation
of selenium, boron, and chloride is known to be ineffective. Therefore,
even if excessive concentrations of other constituents were acceptable
because of soil attenuation effects, the presence of these three species
in concentrations in excess of water quality criteria strongly indicates
that environmentally sound techniques need to be employed for scrubber
sludge disposal.
4. 0 POTENTIAL LAND RECLAMATION PROBLEMS
4. 1 LAND USE CRITERIA
Criteria identifying the load suoport limits to be attained on top of a filled
sludge pond for eventual use as structural or recreational land are not
included in either the state solid waste disposal regulations or building
code requirements. However, the Uniform Building Code shows that the
range of allowable soil pressure for building construction is from Z441
kg/sq m (500 psf) to 39, 064 kg/sq m (8000 psf) depending on the type of
soil and the depth of footing used. Samples of the weakest and strongest
soils considered for structual usage are loose inorganic sand silt mixtures
and hard clay, respectively. Materials like loose organic sand and silt
mixtures and muck or bay mud are considered to have "zero" allowable
soil pressure, but it is accepted practice to reclaim municipal landfills
(e. g. , organic soils) for recreational land use.
414
-------
4.2 PHYSICAL CHARACTERISTICS
The physical characteristics of power plant sludges are primarily dependent
on the percent moisture. However, the degree to which a. sludge can be
dewatered is dependent on the relative quantities of the major ohases
present, i. e. , calcium sulfite, calcium sulfate, calcium carbonate and
fly ash. Among these, the dewatering capabilities of a sludge depend
mostly on the degree of oxidation of the sulfite phase. Primarily as a
consequence of the platelet-like nature of the sulfite crystals, water
becomes entrapped between the crystalline olates and is not easily or
readily removed by physical dewatering techniques, such as filtration
or centrifugation. In contrast, the sulfate crystals are "blocky" and do
not trap water. A sulfite sludge can be dewatered in a thickener to 35-45
oercent solids and will not increase solids content significantly by settling.
A filter or centrifuge will increase solids content to about 50-55 percent.
A sulfate sludge dewatered to 35-45 percent solids will increase its solids
content to 60-65 percent upon settling, 75 percent when freely drained or
centrifuged, and about 80 percent by filtration. The presence of signifi-
cant amounts of fly ash will slightly alter the dewatering characteristics
by increasing the solids content of sulfite sludges but will decrease slightly
the solids content of sulfate sludges. Since some degree of oxidation exists
in all power plant sludges, observed behavior tends to fall between these
values deoending on the relative quantities of the phases present. Figures
1 and 2 show the dewatering behavior of two specific sludges, Shawnee
limestone sludge and SCE Mohave limestone sludge.
Since all physical properties of a sludge are deoendent on moisture content,
it is not possible to relate directly the behavior of different sludges at
similar stages. For example, a filtered sulfate sludge will produce a cake
that is firm, easily compacted and can readily support personnel and
equipment. In contrast, a filtered sulfite sludge will produce a cake that
may have a solid-like appearance but has a consistency much like tooth-
paste. A better comparison of ohysical properties can be made when
sludges containing equivalent moisture content are compared. For example,
at 65 percent solids content, sludges are firm enough to support personnel,
415
-------
SHAWNEE
(Sludge contains about 40% sulfite,
10% sulfate, 40% fly ash, and
10% unreacted limestone)
20 40 60 80 100
SOLIDS CONTENT, weight percent
Figure 1. EPA/TVA Shawnee limestone sludge bulk dt-nsities
416
-------
1.9r-
MOHAVE (Sludge is primarily
sulfate and limestone)
80
SOLIDS CONTENT, weight percent
100
Figure
SCE Mohavc limestone sludge bulk densities
417
-------
and at solids content greater than 70 percent, equipment ca.n be supported.
The thixotrooic behavior of the sludge is a direct function of sulfite content.
Hence, while sludges with equivalent moisture content are capable of
suooorting equivalent Loads, the work applied to a sulfite sludge in walking,
or moving equipment, will tend to liquefy the sludge whereas a sulfate
sludge will tend to remain firm. The oresence of fly ash in the sulfite
sludge will tend to firm it somewhat, but has no observable effect on a
sulfate sludge. Again an example is made of the TVA Shawnee limestone
and SCE Mohave limestone sludges (Figure 3} where the load carrying
capacity of the sludges are compared at equivalent moisture content.
When a sludge is dewatered or dried to a point at -which air voids are
created, the sludge can be compacted so as to increase its bulk density.
The point at which this behavior occurs varies among sulfite and sulfate
sludges. Because of the ^article agglomeration nature of sulfite crystals,
sulfite sludges will create air voids at a solids contents as low as 60-65
oercent. Sulfate sludges develop air voids at a solids content of about
80 percent. The presence of fly ash in the sludge will affect the sulfite
sludge by increasing the solids concentration at which air void formation
is reached. As in other properties just mentioned, fly ash has little
effect on sulfate sludges. The amount of compaction is dependent upon
many factors defending on phase compositions, moisture content, and
compaction force, but dry bulk densities up to 1.5 gm/cm (93 Ib/ft )
can be obtained in many sludges. This represents a 25 percent volume
reduction on a dry solids basis for sulfite sludges but only a 7 oercent
reduction for sulfate sludges. Although the efficiency of field compaction
may not be as high as that in the laboratory, relative volume reductions
by comoaction between the two extreme types of sludges will be generally
the same. Figure 4 shows the compaction capabilities for typical sulfite
and sulfate sludges as a function of moisture content. The moisture content
for optimum compaction is about 10 percent for sulfate sludges but about
20 percent for sulfite sludges. This greater moisture requirement for
sulfites is primarily a consequence of the entrapment of moisture between
the platelet-like crystals that make up particle agglomerates.
418
-------
VO
V)
Q.
I
I-
o
l-
to
O
Z
cc:
35
30
25
20
Z 15
LU
CD
0
O
10
SHAWNEE
(sulfite sludge
with fly ash)
0
MOHAVE
(sulfate sludge
without fly ash)
30 35 40
MOISTURE CONTENT, weight percent
45
Figure 3. Sludge compaction strength
-------
2.O.—
to
o
1.8
o
1.6
LJ
Q
D
CD
1.4
1.2
SULFITE
SULFATE
1.0
0
10 15 20
MOISTURE CONTENT, percent
25
30
35
Figure 4. Preliminary compaction capabilities of sulfur sludges
-------
The density of sludge is of orimary importance because it is the major
determinant of its compaction strength, compressibility, and perme-
ability; all these factors affect the environmental acceptance of the
sludge for disposal. The density also determines the quantity of sludge
that can be stored in a given disposal area, thus affecting disposal
costs. However, cost savings gained by higher density filling can be
offset by the additional cost of compaction. On the other hand, if high
density fill would render the disposal area to a greater economic use
after reclamation, this factor must also be considered in choosing disposal
techniques.
From the dewatering data discussed earlier, it can be seen that a sulfate
sludge dewatered by filtration or centrifugation will have a moisture
content very near the requirement for optimum compaction, although
the volume gain by compaction is minimal. Sulfite sludges do not
dewater as well as sulfate sludges and will require subsequent dewatering
such as combining with dry additives, or thermal drying to attain the
moisture content for optimum compaction. It should be noted that although
a sludge can be dewatered to a degree such that it will support personnel
and equipment, it will readily rewet in the presence of water. The gain
of moisture in rewetting is limited by the volume of the voids in the sludge.
Therefore, a rewetted compacted sludge will not return to its original
moisture content or volume. By contrast, a rewetted uncompacted sludge
•will have properties nearly identical to those it had in its original wet
state. Studies are continuing to determine the degree to which different
rewetted compacted sludges approach their original state.
An additional physical parameter affecting disposal technology is the
permeability of the sludge to leachate waters. Permeability among the
sludges can vary over several orders of magnitude depending on phase
composition and degree of compaction. Sludge that is allowed to settle
and drain freely will have a permeability value equivalent to that of fly
-4
ash, about 10 cm/sec. Sulfate sludge in a settled state will drain more
readily than sulfite sludge in the same state. When these sludges are
compacted, drainability is reduced. The addition of fly ash also reduces
the drainability of sludges. The drainability of sulfite sludges is reduced
421
-------
only slightly, but the drainability of sulfate sludges is reduced by nearly
an order of magnitude. A permeability value as low as 10 cm/sec has
been obtained for a compacted sulfate sludge containing fly ash. This
appears to be the minimum permeability obtainable for untreated sludge.
4.3 ASSESSMENT
.Laboratory tests have shown that raw sludge which has been dewatered
to about 65-70 percent solids can support loads in the range of 4883 kg/
sq m (1000 psf) to 24, 41 5 kg/sq m (5000 psf). Thus, it appears that raw
sludge or raw sludge oonds which have been dewatered could be used for
structural purposes per the Uniform Building Code. However, the cost
of dewatering the sludge and maintaining the site at the necessary degree
of dryness is strongly dependent on the raw sludge properties, and may
or may not compare favorably with the cost of an alternative disposal
technique such as chemical fixation.
Data from chemical fixation processors indicate that treated sludge quickly
attains a compressive strength above the maximum limit for compacted
fine sand or hard clay; i. e. , 39, 064 kg/sq m (8000 psf). This material
has the caoability to make a disposal site reclaimable for either structural
or recreational use.
5.0 POTENTIAL DISPOSAL PROBLEM SOLUTIONS
5. 1 DISPOSAL ALTERNATIVES
The choice of disposal alternatives is governed by the criteria for an
environmentally sound disposal plan. These criteria are presently address
ing the potential for surface or ground water pollution by either runoff or
leachate percolation and the requirements for land reclamation of retired
disposal sites. Whereas these criteria are established by state regulatory
agencies and in as much as the disposal requirements are not yet fully
defined by any state, it will be oresumed that all disposal alternatives
herein discussed may be equally acceptable.
In each of the disposal plans, the basic concept of control is containment
or isolation. The extent or thoroughness of containment required will
depend on the hazard the disposal is adjudged to present to the environ-
ment. For example, a sludge from which chemical pollutants may leach
422
-------
may require containment within a disposal site with an impervious liner
made of either natural or man-made materials. The most obvious example
would be disposal in a clay or plastic lined pond. An alternative to this
approach would be to reduce leaching by one of several commercially
available chemical fixation processes. An alternative to chemical fixation
may be direct land filling of dewatered sludge by methods that restrict
rain water oercolation and leaching. The acceotability of the latter method
may deoend on the geological and hydrological conditions of the disposal
site. Variations of these methods and the advantages and disadvantages
of each are discussed in the following sections.
5. 1. 1 Chemical Fixation/Landfill
To date, three companies - Chemfix, Pittsburgh, Pennsylvania; Dravo
Corporation, Pittsburgh, Pennsylvania; and IU Conversion Systems, Inc. ,
Philadelphia, Pennsylvania - have developed a significant technology for
chemical fixation of power plant sludge materials at disposal costs that
may be considered generally acceptable. Whereas all three processes
are designed to produce a material that chemically binds soluble com-
ponents, their processes and oroducts are distinctly different. Although
not all of these processes are currently being operationally demonstrated
with oower plant sludges, each processor has sufficient experience in the
fixation of sludge materials to give credence to his performance claims.
The following discussions reflect these claims, which are now being
assessed by independent analyses in the Aerospace Corporation sludge
disposal study.
7 8
5. 1. 1. 1 Chemfix Process. Chemfix has noted ' that their company has
gained wide field exoerience in the fixation of solid, sludge, and liquid
wastes oroduced by automotive, chemical, mining, paint, petrochemical,
steel and metals industries. The Chemfix process reportedly can handle
sludge fixation over a broad range of percent solids and produces a product
with a soil-like appearance which does not prevent the oercolation of rain
water, but does have structurally stable oroperties and chemically binds
the constituents to accomplish pollution control .
The Chemfix process involves the reaction of sodium silicate and one or
a combination of setting agents--portland cement, lime, calcium sulfate,
calcium chloride--with the waste material to form chemically and mechan-
423
-------
ically stable solids. The reaction occurs at ambient temperatures and is
not affected by temperature variations. The particular chemical
choice, chemical ratios, and reagent quantities depend on the type of
waste, required speed of reaction, and the end use of the fixed material.
Either reactive or nonreactive wastes can be treated with the Chemfix
process. The reaction process involves an initial gelatinous state and
a subsequent hardening period. The gelation time, which is controlled,
is based on the distance the treated material must be pumped to the
disposal site.
The treated waste can be made either hard or soft with varying textures
depending on its ultimate disoosal requirements. The chemical system
reacts with all polyvalent metal ions oroducing stable, insoluble inorganic
compounds. The resulting material reportedly can be used directly as
landfill which, when oroperly fertilized, will readily support grasses and
olant life-
Chemfix data further indicate that the material produced by the Chemfix
process does not leach chemical constituents by rain water at concentra-
tions that exceed the natural background levels in ground waters. Its soil-
like behavior should allow it to be compacted such that land reclamation
costs can be minimized for any subsequent use of the land. Chemfix
claims chemically fixed sludges by this process should be acceptable to
regulatory agencies for disposal directly as landfill without further treat-
ment.
5. 1. 1. 2 Dravo Process . The Dravo Corporation process is now being
used by Duquesne Light Company to stabilize the sludge produced by the
lime scrubbers at Duquesne's Phillips Station near Pittsburgh, Pennsyl-
Q
vania. Dravo claims that they have gained extensive laboratory experience
on sludges from various air pollution control devices that include but are
not limited to power plant sulfur dioxide scrubbers. This process can
chemically fix sludges at various solids content and it produces a material
that is clay-like in consistency.
424
-------
The Dravo process chemically fixes sludge by the inclusion of a proprietary
admixture called Calcilox. The Calcilox is mixed with the sludge usually
just before the sludge is pumped to a disposal basin. The sludge solids
are allowed to settle and curing takes place under water over a 30-day
period. Disposal can be made directly in the final disposal site, or an
interim curing basin can be used from which the cured solids can be
removed to a oermanent disposal site.
Dravo information indicates that the product of the Dravo orocess is a
firm material which is convenient for disposal. By adding 3 percent
Calcilox (based on sludge solids) the conditioned sludge has properties
that compare with silty clay with permeability of 2 X 10 cm/sec. Clay-
like properties including high shear and compressive strength are being
obtained with Calcilox additions of 5 to 10 percent.
The low permeability, low leachability and the structural stability of
sludges conditioned by the Dravo process should allow them to be used
directly as landfill materials. The company claims that sludges chemically
fixed by this orocess have been accepted for disposal without containment
in at least one site in Pennsylvania.
5.1.1.3 IUCS Process. The IOCS process ' ' has been field tested on
sludge produced by wet scrubbing effluent gases on a lead smelter using
hydrated lime as the absorbent. This sludge has many properties similar
to power olant sludges and produces a hardened composition when chemi-
cally fixed. Laboratory tests have verified the similarity of sludge fixation
behavior between this sludge and cower olant sludges.
The IUCS process uses fly ash and a lime additive as ingredients in the
chemical fixation of sulfite/sulfate sludges. The quantity of either of
these additives depend on the moisture content of the sludge and the
reactivity of the fly ash. In some cases, dewatering of the sludge is
necessary for the desired reactions to take place at economical amounts
of the additive. Three primary reactions take place in this process:
1) the reaction of lime with soluble sulfate that originates in either of the
fly ash or sludge liquor so as to form calcium sulfate (sulfite will also
react, but more slowly), 2) the reaction of lime, sulfate and iron or
425
-------
aluminum oxides oresent in the fly ash glass forms complex sulfo-ferrites
or sulfo-aluminates (this reaction results in the formation of the crystalline
phase ettringite which contains substantial quantities of water within the
crystalline structure), 3) the reaction of lime with the glassy silica of fly
ash results in the well known pozzolanic reaction that proceeds slowly to
form the calcium silicate phase, tobermorite.
Some IUCS laboratory results of this process are shown in Table 4 for
the Shawnee limestone sludge. The data indicate that the material can
develop strengths as high as 4. 1 tons/sq ft in seven days. Falling head
permeability data indicate improvement during the first month of curing
from 10 to 10 cm/sec. The low permeability is a consequence partly
of the expansive nature of ettringite which tightly seals the set mass and
which prevents shrinkage crack formation which may otherwise allow
some leakage. Besides low leach rate, analysis of the leachate indicates
a reduction in concentration of trace elements of approximately one order
of magnitude. Thus, as a consequence of chemically combining trace
elements into new crystalline phases and the reduction in leaching rate,
the availability of soluble salts or toxic contaminants to ground waters can
be reduced many times, when comoared to unfixed sludge (see Tables 2,
3,4). An indeoendent analysis of this fixed sludge sample (as well as
others using Calcilox and the Chemfix orocess) is now being conducted
in the Aerospace Corporation laboratories, and further verifications will
be made in an EPA field disposal program.
In addition to producing material suitable for landfill, the IUCS process
may also be applied to the manufacture of synthetic aggregate having
properties suitable for road base material. This application, which
requires increased quantities of the additives and additional processing
equipment (when compared to a landfill application), is currently being
field demonstrated.
5.1.2 Dewatering/Landfill
One method of sludge disposal may incorporate dewatering, drying and
placement of the -waste solids by one of several techniques. The most
direct technique would use the method of compaction developed for sanitary
426
-------
TABLE 4. LABORATORY RESULTS OF FIXED TVA LIMESTONE SLUDGE ANALYSIS13
LEACHATE
CONCENTRATIONS,
ppma
TOTAL ALKALI
TOT. DIS. SOLIDS
so3
so4
Cl
Ca
Mg
Al
Fe
Mn
Cu
Zn
Cd
Cr+3
As
Pb
Sn
Hg
pH
PERMEABILITY,
cm/ sec
COMPRESSIVE
STRENGTH, tons/sq ft
AGE OF TEST, days
2
1068
1370
11
45.2
64
268
0.005
2.2
0.02
0.005
0.02
0.005
0.007
0.05
0.01
0.2
0.1
0.01
12.35
4
542-
730
19
41.9
12
220
0.005
3.2
0.01
0.01
0.01
0.01
0.005
0.03
0.01
0.05
0.1
0.01
12.5
6
810
1210
14
36.2
74
235
0.005
11.4
0.01
0.007
0.01
0.005
0.005
0.02
--
0.05
0.05
--
12.1
10
524
770
20
51.0
21
170
0.01
0.95
0.05
0.007
0.02
0.005
0.005
0.02
__
0.05
0.05
--
11.9
14
40
180
10
48.5
16
27.5
0.05
0.01
0.005
0.005
0.005
0.005
0.01
--
0.05
0.05
--
10.2
28
40
250
3
43.6
14
20
0. 1
< 0.01
< 0. 005
< 0.01
< 0. 005
< 0. 005
< 0.01
--
< 0.03
< 0.1
--
9.0
AGE OF TEST, days
0 7 14 21 35
1 x 10-4 6.3 x 1Q-5 " 6.2 x 10'6 5>0x 10~
4.1 4.7
Equivalent to mg/I
-------
landfill. For the oroblem of leachability, containment -n clay cells may
be necessary. In some cases, final covering with clay may be all that
would be necessary to prevent contamination of ground waters. The
soecific method of solid waste disposal will depend strongly on site
requirements.
In most cases, partial drying is required to produce a material that is
readily compacted so that the disposal site may be ready for structural
use immediately after filling. The extent of drying required would be
dependent upon the physical nature of the sludge. Several alternatives
exist and their cost effectiveness depends on sludge characteristics,
climatic conditions, and system variables. For a sulfate sludge dewatered
by filtration or centrifugation, if any additional drying is required to
reduce the water content to a. level suitable for compaction, solar drying
may be sufficient. On the other hand, for a sludge with relatively high
sulfite content, dewatering by either filtration or centrifugation is not
adequate for compaction. In this case, one alternative might include the
blending of dry collected fly ash to sludge dewatered by filtration or cen-
trifugation to reach optimum moisture content for compaction. A second
alternative might employ a thermal dryer that uses waste heat to dehydrate
the sludge. Depending on the many system variables, it is possible to
pass all the dewatered sludge solids through the dryer to reach an optimum
comoaction consistency, or to pass through the dryer, a portion which
can then be mixed with the remaining solids to reach proper consistency
for placement. To acquire the proper degree of dryness, the two alterna-
tives cited above may also be aoolied to the sulfate sludge. A third alter-
native would be to oxidize the calcium sulfite to gypsum and thereby gain
the dewatering advantages of sulfate sludges.
The environmental acceptance of direct landfilling with dewatered sludge
will be deoendent on its potential for environmental pollution by leaching.
The potential for leaching oossibly can be minimized for sulfate sludge
by cake washing and treatment of the wash liquor. However, this has not
yet been demonstrated. For most cases, the cost for containment (e.g. ,
with clay), if necessary, would equal or exceed the cost required for
direct oonding in clay or an equivalent containment material. However,
in the case of dewatered sludge disposal, the benefits of reducing the
428
-------
disoosal volume and the future land use of a compacted sludge site may
offset the dewatering costs.
5. 1.3 Environmentally Controlled Ponding
A ootential disposal alternative is the ponding of raw sludge; e.g.,
clarifier underflow or dewatered sludge cake. In this concept, the sludge
is transoorted (pumped or hauled) from the scrubbing operations to a
waste disposal pond suitably constructed to contain the raw sludge for
long oeriods of time. The pond configuration commonly used is essentially
an excavated oit or basin with the soil graded into embankments or dikes
generally 3 to 15m (1 0 to 50 ft) in height. The pond bottom is usually
flat, but may be contoured to drnin off seeoage. Impervious liner materials
are installed along the inner pond surface to inhibit any seepage of liquids
from the sludge into the ground. Both flexible (e.g., polyethylene, ooly-
vinyl chloride, butyl rubber, chlorinated polyethylene, duPont Hypalon
films) and nonflexible liners such as bituminous concrete, clay, and
pozzolan stabilized base are used.
Leakage detection concepts are usually incorporated into the pond design.
These concepts vary from a simple underdrainage basin with visual
monitoring to an electrical system designed to measure the electrical
resistivity of the ground in the pond area. In addition, many installations
have visual or remote sensing systems to monitor the nearby surface and
ground waters to determine if there is any leakage or seepage from the
pond. Overdrainage can return oond surface water to the scrubber if
appropriate to the total system design.
Although the basic design features of ponds arc generally uniform through-
out the country, there are local ordinances and environmental regulations
that govern. Even though the concept of ponding is a proven technique for
large-scale, low cost evaporation operations, the predicted useful life
of a potential sludge disposal pond liner still needs to be defined. It is
directly dependent upon the aging characteristics of the liner material and
its chemical interaction with the sludge and the supporting soil. Flexible
14
liners are generally guaranteed for 20 to 25 years , and the life expectancy
429
-------
of nonflexible liners is estimated to be somewhat longer. However, long
term service data applicable to desulfurization sludge containment do
not yet exist for either type of liner.
In addition, systems to control seepage from a sludge disposal pond after
it has completed its service life have not yet been demonstrated. Such a
system might consist of provisions for underdraiaage and overdrainage
combined with air drying to dewater the site, followed by capping with a
contoured clay cover to provide for runoff of rain water. The purpose of
such a system would be to eliminate a hydraulic head, thereby preventing
seepage from the pond.
6.0 DISPOSAL COST ESTIMATES
Disposal cost estimates were made for each of the two most prevalent
disposal methods: ponding, and landfill using chemical fixation. Infor-
mation was provided by various sources, and for each method the cost
data are widely variant. As a result, the disposal costs are given in
ranges of values consistent with appropriate qualifications based on the
many variable parameters attendant to either disposal method. Although
refinements are expected during succeeding disposal cost studies in this
orogram, the disposal costs are exnected to be always expressed as a
range of values, except when applied to a specific disposal operation.
This section summarizes this analysis to date and includes significant
disposal costs and appropriate rationale. All costs are given in units
commonly used in industry, e.g., dollars /short ton (dollar s /O. 907
metric ton) and dollars/sq yd (dollars/0. 836 sq m).
6. 1 CHEMICAL FIXATION DISPOSAL COSTS
The disposal costs by chemical fixation are dependent upon many factors
such as: the fixation process and the contractor-operating efficiency;
sludge characteristics including water, ash, and chemical content;
disposal site physical characteristics; distance from power plant to disposal
site; sludge transoort method; disposal site land costs and ownership;
residual value of disposal site; and environmental monitoring. Addi-
tionally, the costs depend upon whether the power plant uses a disposal
430
-------
contractor or does the work itself. To determine these costs, data were
obtained from personal communications, visits, published cost data, ' '
brochures, and testimony from the National Power Plant Hearings on
Scrubbers, October - November 1973. Sources of these data were: power
companies - Commonwealth Edison Company and Duquesne Light Company;
chemical fixation companies - Chemfix, Dravo Corporation, and IU
Conversion Systems, Inc. ; and the EPA Control Systems Laboratory.
Some estimated disposal costs (converted to a 50 percent solids basis)
quoted by the power companies and fixation companies are listed in Table
5.
These values, equated to sludges that are 50 percent solids, range from
$2. 00 to $9. 00/ton. All relevant factors pertaining to the values given
are not available; however, based on the data given, it appears that exceot
for extreme cases (e.g., development or demonstration), the general
range of costs will be aoproximately $2. 50 - $5. 00/ton of sludge on a 50
percent solids basis. Some additional costs may be incurred for soil
cover and/or seeding after the disposal site is filled.
Previously published fixation cost data include various limitations on the
applicability of the values, and do not include recent escalations in costs.
These data, therefore, have been augmented through personal communica-
tions with several of the companies involved for the purpose of deriving a
general cost estimate for fixation/disposal rather than to attempt a rating
of one versus another since each application would be condition and site
deoendent. Sources which provided estimates for complete disposal systems
(using quite different approaches) were Dravo Corporation which provides
a disposal service, and two power companies performing their own disposal
tasks. Appropriate Dravo data are taken from costing of a pro >osed
disposal operation for the Bruce Mansfield Station, a now 1650 Jvlw
plant, in western Pennsylvania. The costing data are based on consider-
able land acquisition and development costs and the pumping of the sludge
(32-37 percent solids) approximately Smiles to the disposal site where
it is to be chemically treated prior to disposal. In this case, although
capital costs are high, the large quantities of sludge place the cost per
431
-------
TABLE 5. SLUDGE FIXATION COSTING ESTIMATES
(all values are condition and site dependent)
SOURCE
DOLLARS/TON
(50% solids)
REMARKS
COMMONWEALTH EDISON
COMPANY, WILL
COUNTY STATION
DUQUESNE LIGHT
COMPANY,
PHILLIPS STATION
IU CONVERSION
SYSTEMS, INC.
CHEMFIX
DRAVO CORPORATION
8. 00 - 9. 00
6. 00 - 6. 50
-7.50
2.00 - 6.00
CURRENT ESTIMATE1
TARGET0
TOTAL DISPOSAL
TOTAL DISPOSAL
Company operation - on-site disposal; excludes capital costs
-------
ton as analyzed in this study at less than $5. 00. For sites requiring
less land development and shorter pumping distances, Dravo estimates
costs of less than $4.00 per ton. On the other hand, Commonwealth
Edison and Duquesne Light Company are currently conducting their own
systems on what can be termed development operations. These are, by
comparison to the Dravo example, relatively small systems (156 and
100 Mw, respectively), and they use intermediate curing ponds and
trucking of the cured sludge to disposal sites up to 1. 6 Km (1 mile)
from the fixation operation. Estimated current costs for these two are
in the range of $7. 50 to $9. 00 per ton on a 50 percent solids basis. Some
estimates of power plant operation such as the one at Commonwealth
Edison indicate that refinements in the system may bring the cost down
to approximately $6. 00 per ton on an operational basis. Additional
contacts with Chemfix and IUCS to up-date published information indicate
that for fixation disposal of 50 percent solids sludge including all capital
charges and local transport of the treated sludge, a cost of $4. 00 to $5. 00
per ton is reasonable.
6.2 PONDING ECONOMICS
The cost for constructing a sludge pond will depend not only on the technical
design details, but it will also depend upon its geographical location. The
cost for shipping the liner materials to the site frotn the factory or pro-
cessing center will vary with the distances traveled; and, the labor costs
for construction and liner installation vary with the different areas of
the country. The economic analysis undertaken in this study considered
these variations and used average values such that the costs presented
in this paper can be considered typical, but not necessarily specific.
Cost estimates are from suppliers of pond materials and builders of
1417
ponds ' . All costs are as of August 1974. The economic analyses are
based on a 1000 Mw power plant burning coal with 3 percent sulfur and
12 percent ash. The plant life is 30 years with the plant operating an
average of 4560 hours per year with an average annual sludge (50 percent
solids including ash) production of 930,000 metric tons (1,025,000 short
tons).
433
-------
The annualized costs used in this work included the costs for labor,
maintenance and capital charges at 18 percent. The capital charges
included depreciation, replacements,, insurance, cost of capital, and
taxes.
Figure 5 presents a composite cost summary for the various liner
materials considered. The values shown are typical of the range of
costs that might be expected for installed liners at the field site, but are
to be used for planning purposes only. For this study, the mid-point of
the price range was used. The widespread variations in costs resulted
because of factors such as the cost of transportation from the liner manu-
facturer's facility to the field site within a radius of 1609 Km (1000 miles),
the variation in the cost of labor throughout the country for liner installa-
tion, and price discounts for purchasing large quantities of liner materials.
It is expected that the flexible liners could probably be purchased and
installed at the lowest cost values; however, these lower limit numbers
were avoided in the cost analyses as a precautionary measure against
data which may not be reliable.
In early 1973, it was reported that the projected cost for a 46-cm (18-in)
thick clay liner installed in several large ponds being constructed in the
midwest was about $6. 00 per sq yd. This clay was transported by truck
about 48 Km (30 miles) from the clay pits directly to the pond site. If
suitable clay could have been available on the pond site, it could have been
moved into place at a cost of $1. 00 per sq yd. At today's prices, the
costs have risen to approximately $9. 00 and $1. 50 per sq yd, respectively.
The increase in installed costs during this same time period for the
flexible liners has been about 100 percent for 0. 254 mm (10 mil) thick
polyethylene liners, 50 percent for 0. 508 mm (20 mil) thick polyvinyl
chloride, and 25 percent for 0. 762 mm (30 mil) thick Hypalon. The other
flexible liners also have had appreciable rate increases. It is expected
that the prices will continue to increase in the foreseeable future, since
many of the liner materials are presently in such short supply that the
prices vary on a monthly basis. Some contractors have indicated that
the actual price will be established only at the time of delivery to the
field; they will probably also consider writing escalators or surcharges
434
-------
INSTALLED COST, $/sq yd
CM Ul
01
f-"
5=
C
U1
V.
P
d.
"i
c
cr.
LINERS J
iND THICKNESS
1 1 1 1 1
1 O 1 POLYETHYLENE - 10 mils
1 O i POLYVINYL CHLORIDE - 10 mils
LHMMM^^WV^MMB^^^^ r\f\ I V/ V/l kt\/ 1 ^* LJ 1 ^*\D 1 F^ C* OA w* • 1 «»
' ^ 1 POLYVINYL CHLORIDE - dO mils
H-O-H PETROMAT FABRIC - 125 mils
1 -Q I BUTYL RUBBER - 30
mils
CHLORINATED
I Q H POLYETHYLENE -
30 mils
ff 1 Q 1 HYPALON - 30 mils
(/>
, EPDM RUBBER
5-3 i— -0 ' 30 mils
^p | ___^_CLAY 18
Cn f~i -^ ON
3 CD CD
q {/> OT
RETE
-------
into the fixed price contracts to permit then to tag on price increases
as the materials they purchase go up in costs.
Figure 6 illustrates the 30-year averag-e disposal cost in dollars per ton
of sludge (50 percent solids) as a function of the depth of raw sludge in the
pond from 3 to 1 Z meters (1 0 to 40 ft). Two liner materials \vere con-
sidered: 0. 508 mm (20 mil) thick polyvinyl chloride (PVC) at an installed
cost of $2. 33/sq yd including a soil cover of 15. 2 cm (6 in), and 0.762 mm
(30 mil) thick duPont Hypalon at a cost of $3. 85/sq yd installed. These unit
costs are also applicable to equivalent cost liners (see Figure 5). It can
be seen that the disoosal cost is (without including the cost of land) about
$2. 50 per ton if PVC is used, and about $3.40 per ton when a higher cost
liner such as Hypalon is used.
Each pond configuration used in these analyses had a 0. 915 meters (3 ft)
freeboard. It was assumed that all of the land needed to contain the
sludge during the full 30-year life of the power plant would be purchased
initially. However, the sludge ponds would be constructed each 10 years
as the previous ones became filled. Land costs of $404 and $2020/hectare
($1000 and $5000/acre) were used. As expected, the land cost decreased
proportionately as the dike height increased because of a smaller area
needed for the ponds to contain the 30-year production of sludge. Figure
6 illustrates the effect of the cost of land on the sludge disposal costs. It
also indicates that for each pond concept there is an optimum sludge depth
at about 9. 15 meters (30 ft) which represents theoretical minimum disposal
costs.
6. 3 EFFECT OF DEWATERING
The disposal costs presented thus far were based on a desulfurization sludge
with a mixture of ash and water at a total solids content of 50 percent.
Over the 30-year operating life of the example power plant, sludge at this
solids level will occupy a volume of about 2720 hectare-meters (22, 000
acre-ft).
An investigation is underway to assess the effects (on pond size and disposal
costs) of the dewatering of the sludge prior to disposal. Figure 7 graphically
presents the results of the initial work comparing the volume occupied by:
436
-------
PVC - 20 mil
HYPALON - 30 mil
O
O
m
a
UJ
O
O
ID
TOTAL COST INCLUDING LAND
($5000/acre)
TOTAL COST
WITHOUT LAND
TOTAL COST
INCLUDING LAND
($1000 per acre)
INSTALLED LINER ONLY
I
10 20 30
SLUDGE DEPTH, feet
40
6.
o
8
O
O
a.
ili
O
O
V-TOTAL COST INCLUDING LAND
\ ($50007acre)
TOTAL COST
INCLUDING
LAND
($1000/acre)
TOTAL COST
WITHOUT LAND
INSTALLED LINER COST
10 20 30
SLUDGE DEPTH, feet
40
Dispose) costs - ponding
50 pcrct-nl solids
Yea r
-------
00
VOLUME INDEX
O — f\> OJ .&.
—
1.0
»/>
S
p
-------
only sludge at 50 percent solicit, sludge at 65 percent solids, and then
separately mixing these sludges with 80 percent solids ash. The reduc-
tions in volumes required for sludge, disposal as a result of dewatering
and/or ash removal prior to disposal are shown. For example, the
disposal of sludge plus ash (50 percent solids) requires a pond volume
3. 55 times that required for ash only; whereas, the disposal of sludge
only, at 65 percent solids, requires a volume only 1. 3 times that for
ash alone. This concept in ponding offers the potential advantage of
reducing the disposal costs by reducing the pond size and land required
for the pond.
Economic trade-off evaluations are in progress to determine: 1) the
various cost factors for conventional fly ash collection and sludge de-
watering (at numerous solids content), Z) the disposal cost savings
attributed to these factors, and 3) the net effects of 1) and Z). The de-
watering techniques currently being considered include operations such
as filtration, centrifugation, thermal drying, oxidation of the clarifier
sludge, and settling in a thickener or settling pond.
6. 4 DISPOSAL COST COMPARISONS
The disposal of scrubber sludges using a chemical fixation process is a
rather new industry; therefore, little cost data exist based on actual
practice by the power utilities. Similarly, little actual cost data exist
for scrubber sludge disposal by lined ponding. However, the art and
technology of ponding are well established, but there is little experience
with power plant sludge disposal through the use of lined ponds or of sub-
sequent pond reclamation. There are many ponding choices available that
offer a wide range of costs with apparently small differences in net results.
Therefore, at this time, comparisons are made by observing the ranges
of values available and making rational comparisons in light of the varia-
tions in both the requirements and potential disposal methods.
It should be acknowledged that these two methods, which appear to be the
best currently available or now in development, do not necessarily oifer
the same degree of environmentally sound disposal. For example, fixation,
439
-------
which may be the more expensive process of the two, appears to offer a
permanent solution and may be suitable for land reclamation, but further
study is needed to evaluate the long-term environmental acceptability of
this process. Flexible liner ponding, which may be the least expensive,
may provide safeguards at a particular site for as long as approximately
25 years, but some added costs may occur for final disposal safeguards
which may be necessary to provide permanent environmental protection
and land reclamation. Although nonflexible liner systems may protect
for periods greater than 25.years, eventually additional costs may also
be required to accomplish permanent environmental protection and land
reclamation.
Comparisons of fixation and lined ponding, therefore, cannot necessarily
be made on an equal basis since the two methods do not necessarily offer
the same results. If the two could offer the same environmental protection
indefinitely and if land reclamation were not of primary interest at a
particular disposal site, then a reasonably equal comparison could be
made. Regardless of the many variations and differences that exist between
the two processes, both are being considered throughout the power gener-
ating industry; therefore, cost comparisons are in order at this time.
Ponding costs are estimated to be in the range of $2. 50 to $4. 50 per ton
depending on factors such as the moisture content of the disposal sludge,
dike height, liner material used and the cost of land. Fixation cost estimates
vary from $2. 00 to $9. 00 per ton depending on factors such as sludge
chemistry including water content, chemical additives used, capital costs
and transport method. Neglecting development factors and unusual cir-
cumstances, it appears that average ponding costs (as qualified previously)
should be about half fixation disposal costs.
The following example considers the cost of disposal of all sludge and ash
(50 percent solids) in terms of power produced; limestone scrubber; 85 percent
SO-, removal; 1.2 GaCO,/SO? mole ratio; coal--3 percent sulfur and
u J L*
12 percent ash--burned at the rate of 0. 88 lb/kwhr. If the average disposal
cost is in the range of $2. 50 to $5/00 per ton, the cost for this example
would be 0. 56 to 1. 12 mills/kwhr respectively. If a lime scrubber is used,
440
-------
the cost of disoosal would be about 10 oercent less because the tonnage
would be reduced. Additionally, reductions in disposal costs may be
oossible as noted in Section 6- 3.
7. 0 SHAWNEE SLUDGE DISPOSAL FIELD DEMONSTRATION
The Aerospace Corporation is participating in a scrubber sludge disposal
field demonstration orogram being conducted by the EPA Control Systems
Laboratory at the TVA Shawnec Power Station in Paducah, Kentucky. This
program, which includes the monitoring of test ponds with each pond con-
taining a raw or chemically conditioned lime or limestone sludge, \vill
provide field condition data that will be integrated with the data being
developed in the Aerospace Corporation disposal study. The Aerospace
participation in the field demonstration consists of test planning, coordina-
tion, selected testing {soils, input sludges, leachate and ground \vater
well samples, and conditioned sludge cores), and reporting. A complete
description of the field demonstration is given in Reference 18.
441
-------
8.0 REFERENCES
1. Ros.soff, J. and R.C. Rossi, Aerosoace Corporation, "Disposal of
By-Products from Non -R egenerable Flue Gas Desulfurization
Systems: Initial Report, " EPA -650 11 -74 -037 -a, May 1974.
2. Public Health Service Drinking Water Standards, Publication No.
956, U. S. Department of Health, Education and Welfare, Public
Health Service, Washington 28, B.C., Revised 1962.
3. Water Pollution Regulations of Illinois, Pollution Control Board,
State of Illinois, Springfield, Illinois, July 1973.
4. Proposed Criteria for Water Quality, U.S. Environmental Protection
Agency, Washington, D. C. , 20460, October 1973.
5. "Federal Water Pollution Control Act Amendments of 1972, 92nd
Congress - Second Session, " United States Code Congressional
and Administrative News, No. 10, West Publishing Company, 1972.
6. Uniform Building Code, Table 29C, Vol. I, International Conference
of Building Officials, 1970 Edition.
7. Personal Communication, Chemfix, Division of Environmental
Sciences, Inc., Pittsburgh, Pennsylvania.
8. Conner, J.R., "Fixation/Solidification of Sludges from Lime/
Limestone SO Scrubbers"; Chemfix, Division of Environmental
j^
Sciences, Inc., November 1973.
9. Personal Communications, J.G. Sclmeczi, Dravo Corporation,
Pittsburgh, Pennsylvania, February and May 1973.
10. Personal Communications, L. J. Minnick, IU Conversion Systems,
Plymouth Meeting, Pennsylvania, March 1973.
11. Minnick, J. L. , "Fixation and Disposal of Flue Gas Waste Products:
Technical and Economic Assessment, " Paper presented at Environ-
mental Protection Agency Flue Gas Desulfurizalion Symposium,
New Orleans, Louisiana, May 14-17, 1973.
12. "Converting Stack Waste into Usable Products," POWER,
L. J. Minnick, January 1974.
13. Personal Communication, R.G. Hilton, IU Conversion Systems,
Inc., Plymouth Meeting, Pennsylvania, August 1974.
14. Personal Communication, O.H. Honsgen, Waterproofing Systems,
Inc. , Los Angeles, California, August 1974.
442
-------
15. "Truck Loads of Landfill from Waste Sludge, " Chemical Week,
January 26, 1 974.
16. Gifford, D.C., "Will County Unit 1 Limestone Wet Scrubber Waste
Disposal". Presented at the Electric World Conference, Chicago,
Illinois, October 30-31, 1973.
17. Personal Communication, W.A. Wheeler andH.C. Ellingston, Jr.,
McKittrick Mud Company, Inc., Bakersfield, California, August
1974.
18. Jones, J.W., EPA Control Systems Laboratory (CSL) "Environ-
mentally Acceptable Disposal of Flue Gas De sulfurization Sludges:
The EPA Research and Development Program", for presentation
at the EPA CSL Symposium on Elue Gas Desulfurization, Atlanta,
Georgia, November 4-7, 1974.
443
-------
AN OVERVIEW OF DOUBLE ALKALI PROCESSES
FOR FLUE GAS DESULFURIZATION
by
Norman Kaplan
Control Systems Laboratory
Office of Research and Development
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
For Presentation At:
EPA Symposium on Flue Gas Desulfurization
Atlanta, Georgia
November 4-7, 1974
445
-------
ABSTRACT
The chemistry and process design considerations applicable to
sodium/calcium double alkali systems are presented. Technical
terminology associated with these systems is defined.
The developmental efforts and full scale applications of the
technology by Envirotech Corporation, FMC Corporation, General Motors
Corporation, Zurn Industries, Arthur D. Little, Inc./Combustion
Equipment Associates, Kawasaki Heavy Industries, Ltd./Kureha Chemical
Industry Company and Showa Denko KK are discussed with reference to
appropriate flow sheets. Planned applications of the technology by
these companies are also discussed and tabulated.
446
-------
ACKNOWLEDGEMENTS
The author wishes to express appreciation to Carolyn Fowler and
Gloria Rigsbee for assistance in typing this paper, under the stress
imposed by unusual time constraints.
Appreciation is also expressed toward the personnel listed below,
without whose cooperation and informative input, a presentation of this
type would not have been possible:
Conrad Cornell , „ , _
„..,,. m . ; Envirotech Corporation
William Ellison ^
Jack Brady ,
„ n T i . ) FMC Corporation
Karl Legatski r
Thomas Dingo -, _ _ „ _
„, vr ; General Motors Corporation
Thomas Mason
John Tormev, Zurn Air Systems
Toshio lihara -, ., . . „ T ,
... ... . } Kawasaki Heavy Industries
Makoto Yanai J
447
-------
NOTES
Company Names and Products.
The mention of company names or products is not to be considered
an endorsement or recommendation for use by the U.S. Environmental
Protection Agency.
Consistency of Information.
The information presented in the section on Status of Technology
was obtained from a variety of sources (sometimes by telephone
conversation) including system vendors, users, EPA trip reports
and other technical reports. As such, consistency of information
on a particular system and consistency of information between
the several systems discussed may be lacking. The information
presented is basically that which was voluntarily submitted by
developers and users with some interpretation by the author.
The order of presentation of information or the amount of information
presented for any one system should not be construed to favor
or disfavor any particular system.
Units of Measure.
EPA policy is to express all measurements in Agency documents
in metric units. When implementing this practice will result
in undue cost or difficulty in clarity, NERC/RTP is providing
conversion factors for the particular non-metric units used in
the document. Generally, this paper uses British units of
measure.
For conversion to the Metric system, use the following equivalents
British Metric
5/9 (°F-32)
1 ft
1 grain
1 in.
1 in. 2
1 in.3
1 Ib (avoir.)
1 ton (long)
1 ton (short)
1 gal.
°C
0.3048 meter
0.0929 meters2
0.0283 meters3
0.0648 gram
2.54 centimeters
6.452 centimeters^
16.39 centimeters3
0.4536 kilogram
1.0160 metric tons
0.9072 metric tons
3.7853 liters
448
-------
TABLE OF CONTENTS
Page
Introduction and Background 451
Definition and Discussion of Terms 453
Design Considerations 460
Status of Technology 475
Envirotech Corporation 476
FMC Corporation 484
General Motors 492
Zurn Industries 498
A. D. Little/Combustion Equipment Associates 500
Kawasaki/Kureha 501
Showa Denko KK/Ebara 506
Summary and Conclusions 509
449
-------
INTRODUCTION AND BACKGROUND
Recently, there has been an increasing interest in the double
alkali processes as alternatives to wet lime/limestone slurry scrubbing
systems for utility and industrial boiler flue gas desulfurization (FGD)
applications. Presently there are no full scale utility boiler double
alkali applications in this country but there are two 150 Mw operating
systems in Japan. In the U.S. there are two operating full scale
industrial boiler applications equivalent in size to a 40 Mw and a 20 Mw
control system respectively and an industrial kiln control system
equivalent to a 10 Mw control system based on off-gas flow. The start up
of a 20 Mw prototype utility boiler system is imminent. There has been
a great deal of pilot plant testing of these systems in a variety of
applications including coal and oil fired utility and industrial boilers,
smelting plants, and in a sulfuric acid plant. There are now under
construction in the U.S. (including the 20 Mw utility prototype) at
least five full scale systems ranging in size from 3 to 150 Mw equivalent
electric generating capacity. To put application of this technology in
perspective, however, there are 11 utility boiler FGD applications now
in service using lime/limestone scrubbing totaling over 2500 Mw of controlled
capacity and at least 13 more systems under construction totaling over
7600 Mw of controlled capacity.
"Double alkali" or "dual alkali" processes as they have come to be
known, like their precursors the lime or limestone wet scrubbing processes,
are aqueous alkali scrubbing processes used for FGD. If a "black box"
view of these processes is employed, they are, with minor exception, like
the lime/limestone scrubbing systems: lime or limestone is consumed, and
a calcium sulfite/sulfate and flyash wet solid waste product is produced.
A more detailed examination inside the "black box", however, would show
that the overall process has been split into a number of intermediate steps
designed to improve upon the lime/limestone processes by increasing
reliability of operation, utilization of lime and/or limestone, and sulfur
oxide removal efficiency and, under certain circumstances, producing a
solid waste with better handling characteristics. Whereas in a lime/lime-
stone process the absorption of the SOX from the flue gas and the production
of the waste product occur to some extent simultaneously in a single reaction
system, in the double alkali processes, these two steps are separated through
the use of an intermediate soluble alkali; absorption and production of
waste product can then occur in separate system components.
Separating the absorption and waste production functions accomplishes
two very important objectives. First, it permits scrubbing the flue gas
with a soluble alkali thus limiting the SOX absorption reaction only to
gas/liquid chemical equilibrium and to the rate of transfer of SOX from the
flue gas to the scrubbing solution. In lime/limestone processes, the rate
of dissolution of lime or limestone is a third important limiting factor.
Thus SOX absorption efficiency in a double alkali system is potentially
451
-------
higher than in a lime/limestone system with the same physical
dimensions and liquid/gas flow rates. Additionally, soluble
calcium is minimized and calcium slurry is kept out of the scrubbing
apparatus thus preventing solids build-up (commonly known as scaling
and plugging) in this critical area. Another benefit of this is better
control of the lime/limestone reaction with acidic sulfur containing
compounds in separate equipment specifically designed for this reaction
thus potentially increasing lime/limestone utilization.
Although a number of other processes can technically be considered
double alkali processes, this paper is limited to consideration of the
sodium/calcium based double alkali processes. In these processes, a
soluble sodium based alkali (NaOH , Na2SCu, Na2CO^, NaHCO^) is used to
absorb SOX from the flue gas in the scrubber, and then a calcium based
alkali (Ca(OH)2, CaO, CaCO^) is reacted with the S0x~rich scrubber
effluent liquid to precipitate the insoluble CaS03«l/2 H20 and CaS04-2H2d
and regenerate sodium based soluble alkali for recycle to the scrubber
system. Double alkali systems using ammonium/calcium base have been tested
and while they might have advantages in the reaction with calcium compounds,
their main disadvantage is the potential for pollution by a visible
ammonium salt plume from the scrubbing apparatus caused by the highly
volatile ammonium compounds in the scrubbing apparatus. Another variation
that might be considered double alkali is Monsanto's "Calsox" process,
which uses an aqueous organic base as the absorbent solution in combination
with lime as the calcium supplying alkali to produce the t'nrowaway product.
This paper will discuss some of the chemistry and the status of
technology of sodium/calcium double alkali processes, concentrating for
the most part on work done and data developed by Envirotech Corporation,
FMC Corporation, Arthur D. Little, Inc., General Motors Corporation and
Kawasaki Heavy Industries, Ltd. since these companies have made significant
contributions to the development of the technology and are the front-
runners in application of the technology.
452
-------
DEFINITION AND DISCUSSION OF TERMS
As with any specialized technology, a discussion of flue gas
desulfurization in general, and double alkali technology more
specifically, involves the use of special terminology which has
evolved with the technology. While it is understandable to those
dealing with the subject on a daily basis, it can be somewhat
ambiguous to others. To clarify some of these ambiguities, and
to define terms used here and by others describing double alkali
technology development, a number of terms and concepts are defined
and discussed in a general sense.
Absorption/Regeneration Chemistry
The main chemical reactions that take place in double alkali
systems can be divided functionally into the absorption and regeneration
reactions. A number of secondary reactions which have very important
effects on the overall functioning of the system also take place.
These include oxidation, softening and sulfate removal reactions which
are discussed under the appropriate headings.
The regeneration reactions and in some cases the absorption
reactions will be dependent upon which calcium supplying regenerant
is used - lime or limestone. With lime the system can be operated
over a wider pH range than with limestone. This wider pH range allows
lime systems to operate over the complete range of active alkali
hydroxide/sulfite/bisulfite whereas limestone systems can only
operate in the sulfite/bisulfite range.
The main overall absorption reactions are described by the
following equations:
2NaOH + S02 -> Na2S03 + tUO (1)
Na SO., + S02 + HO -> 2NaHS03 (2)
The main overall regeneration reactions are described by the following
equations for lime and limestone respectively:
Lime
Ca(OH),, + 2NaHSO -> Na9SO 4- CaSO '1/2 H 04- + 3/2 HO (3)
_ _} Z j _) Z /
Ca(OH)2 + Na2SO + 1/2 H20 -> 2NaOH + CaSO^l/2 t^O-l- (4)
Ca(OH) + NaS0 + 2H0 ? CaSOO -I- + 2NaOH (5)
453
-------
Limes tone
CaCO + 2NaHS03 -f 1/2
+ CaSOyl/2 l^OI + CC>2 + ^0 (6)
Active Alkali
This term relates to concentration of NaOH, Na?CO , NaHCO ,
Na^SO- and NaHSO in the scrubbing solutions. Sodium Bisulfite is
included in this definition although it is not technically an alkali
(i.e., it cannot react with SO in these systems); however, it can be
converted to an alkali by reaction with lime or limestone. It should
also be noted that the molar capacity of each of these species for
absorption of SO,, is different, and can vary from zero to 2 moles of
SO- per mole of active alkali. This difference in molar capacity for
absorption of SO,, is illustrated by the following reaction equations:
NaHCO
+ H00 -v 2NaHSO- + CO-
j L 2 32
(Sodium carbonate molar capacity: 2 moles SO /mole)
* NaHSO- + CO-t
j z 32
(Sodium bicarbonate molar capacity: 1 mole SO /mole)
NaOH
(Sodium hydroxide molar capacity: 1 mole SO /mole)
U + SO,, + H^O -^ 2NaHSO^
(Sodium sulfite molar capacity: 1 mole S0-/mole)
- + SO- ->• No reaction
(Sodium bisulfite molar capacity: zero mole S0-/mole)
(7)
(8)
(9)
(10)
(11)
Molar capacity is simply the number of moles of SO needed to convert
1 mole of the absorbent alkali completely to sodium bisulfite. Since
there is a difference in the molar capacity of different active alkali
components to absorb SO-, active alkali is a descriptive rather than a
quantitative term. If the concentration of each of the active alkali
components (moles/liter) is known, however, the capacity of the scrubbing
liquor to absorb SO- (moles of S0-/liter of solution) can be calculated
as the sum of each of the active alkali component concentrations
multiplied by their respective molar capacities as follows:
Scrubber liquor S02 capacity (moles/liter) =
2 [Na2C03J + [NaOH] + [NaHCOj + [Na2S03]
454
-------
TOS
This is an abbreviation for "total oxidizable sulfur". It
denotes the concentration of sulfur compounds in solution in which
the sulfur is in the +4 oxidation state. Simply, this is the total
concentration of sulfite plus bisulfite.
TOS (moles/liter) = [S03=] + [HS03~]
Sulfate is not part of TOS, since the sulfur is in the +6 oxidation
state in this specie. Sulfur dioxide dissolving in scrubbing solutions
increases the TOS in solution.
Active Sodium
This is the concentration of sodium in solution which is
associated with the active alkali.
[Na+] = [NaOH] + 2 [Na CO.,] + [NaHCO ] + [NaHSO ] + 2 [Na SO ]
3CL Z J) .3 _5 £. j
If NaOH, NaHC03, Na CO , NaHSO or Na SO solids are added to double
alkali solutions, the increase in sodium ion in solution is "active
sodium". If Na SO, or NaCl, for example, is added, the increase in
sodium ion is "inactive sodium". Active sodium is not increased by
the dissolution of SO,, in scrubber solutions.
Oxidation
Oxidation in a double alkali system refers to the conversion of
TOS to sulfate by one of the following equations:
HSO ~ + 1/2 00 -> SO " + H+ (12)
J 24
SO" + 1/2 00 -> S0,= (13)
J 24
Simple oxidation of SO- to SO- in the flue gas is also considered
oxidation in the double alkali system:
S02 + 1/2 02 -> S03 (14)
Oxidation in the system has the effect of changing active sodium
to inactive sodium, or active alkali to inactive alkali.
Oxidation may occur in any part of the system: in the scrubber,
the reaction vessels, or in the solids separation equipment. Rate of
oxidation in the system is thought to be a function of rate of
dissolution of oxygen, pH of the scrubbing solution, and impurities
present in solution. Oxidation rate is thus affected by composition of
455
-------
the scrubbing liquor (scrubbing liquors containing high concentrations
of dissolved salts may absorb oxygen more slowly), oxygen content of
the flue gas, impurities in the coal and lime or limestone, and the
design of the equipment (the regeneration and solids separation
sections of the system in particular can-be designed to limit
dissolution of oxygen).
Oxidation rate is expressed as a percentage and is calculated from
an overall material balance on the system:
Oxidation rate (%) = [SO leaving the system (moles)] IQQ
[Total sulfur collected (moles)]
Sulfate leaving the system is total moles of sulfate in the solid waste
plus any sulfate in the associated liquor.
Sulfate Regeneration
This term is a misnomer. Wh^.t is really meant is sulfate removal
from the system with regeneration of active alkali from inactive sodium
sulfate. (See Sulfate Removal.)
SujLtate^ jtemovaJL
Sulfate is removed from the system with regeneration of active alkali
from inactive sodium sulfate. Examples of these sulfate removal reactions
are given below:
Na^SO, + Ca(OH)0 + 2H00 -> 2NaOH + CaSO, -2H_Ov (15)
24 2 2 42
(Gypsum)
Na.SO, + 2CaSO -1/2 H-O + H_SO. + 3H_0 ->• 2NaHSO_ + 2CaSO -21U04-(16)
24 3 2242 3 42
(Gypsum)
y Na2SO, + x NaHSO + (x+y) Ca(OH)2 + (z-x) H90 -> (17)
(x+2y) NaOH + x CaSO^y CaS04'z ^04-
(mixed crystal or solid solution)
Na0SO, + 3H00 Electrolytlc cel1 > 2NaOH + H0SO. + H0 4- 1/2 00 (18)
242 2422
Sulfate removal should be accomplished in an environmenta.ll> acceptable
manner; a simple purge of soluble Na^SO, from the system to land or water-
way disposal is not acceptable in large quantities.
456
-------
Softening
This term is used to describe various methods used to lower the
dissolved calcium ion concentration in scrubber solutions. The
purpose of softening the scrubbing liquor before recycling to the
scrubber is to assure that it is subsaturated with respect to gypsum.
This reduces the gypsum scaling potential in the scrubber. Following
are examples of softening reactions:
Ca4"1" 4- Na2C03 -> 2Na+ + CaCO.4- (19)
Ca++ + Na SO + 1/2 HO + 2Na+ + CaSO -1/2 H 04- (20)
£+ J L- -J £
Ca++ + C02 + H20 -> 2H+ + CaCO 4 (21)
In each of the above reactions, calcium ions are removed from solution
as part of an insoluble material, outside the scrubber system. Reactions
(19) and (21) are referred to as carbo.-.ate softening. Reaction (20) is
considered sulfite softening. Generally, dilute systems employ carbonate
softening while concentrated systems employ sulfite to prevent scaling
in a system.
Dilute ys_. Concentrated System
Dilute or concentrated refers to the active alkali concentration
in a particular system. This differentiation is made because, in
theory at least, based on their solubility products in water, both
CaSO, and CaSO,. should not precipitate from a solution of sulfite
and sulfate simultaneously, unless the concentrations of sulfite and
?ulfate are present in a certain ratio. This can be shown by dividing
one solubility product equation by the other:
[S03] =
from this , cancelling Ca ion concentration
= constant
[so3
The ratio of sulfate to sulfite for simultaneous precipitation of
CaSO/ and CaSO^ is shown to be a constant.
The constant in the above equation is the ratio of solubility
product constants of calcium sulfate and sulfite. Then, in theory, if
457
-------
the ratio of sulfate to sulfite is higher than this constant, only
calcium sulfate should precipitate; and if the ratio is lower than the
constant, only calcium sulfite should precipitate.
This very simplified consideration of the chemistry given above
is clouded in the "real world" by factors that contribute to non-
ideal behavior in these systems. These factors include changes in
ionic activities in solutions containing high electrolyte concentrations
and evidence of coprecipitation of calcium sulfite and sulfate in the
form of a "mixed crystal" or "solid solution" in a manner which is not
completely understood at present.^'-^
With due consideration to the non-ideal behavior of these systems,
however, under given conditions, a ratio of sulfate to sulfite in
solution can be determined at which the previously cited examples hold.
The ratio establishes a definition for "dilute" or "concentrated"
double alkali systems. When the ratio is such that either gypsum or
both gypsum and calcium sulfite will precipitate from the solution with
the addition of slaked lime, then the system is "dilute".
Lime or Limestone Stoichiometry
Lime or limestone Stoichiometry can be expressed as a percentage,
based on an overall material balance around the system:
T . c. ... moles CaO added nn
Lime Stoichiometry = —: — — 7 x 100
mole sulfur collected
„ . . . moles CaCOo added ,„_
Limestone Stoichiometry = —; ——J — x 100
mole sulfur collected
Lime or limestone Stoichiometry is an indication of the efficiency of
usage of lime or limestone in the system. Ignoring alkali components
in the flyash collected, 100% Stoichiometry is complete utilization;
stoichiometries over 100% represent less efficient utilization of
lime or limestone.
Feed Stoichiometry
This is calculated by a material balance around the scrubber.
It is usually expressed as the ratio:
, „ . , Liquor S00 Capacity (moles/liter) x Flow (liters/min)
Feed Stoxch = -3 2 ^ (mole/min)
This ratio is evaluated for streams
entering the scrubber.
458
-------
Feed stoichiometry is a measure of the ability of the incoming
liquor to react with or absorb all of the incoming SCL in the
scrubber, assuming ideal contact of gas and liquor. Feed stoichiometry
above 1.0 is required for high SO- removal. At feed stoichiometries
at or below 1.0, assuming ideal contact between the gas and liquor,
there will be significant equilibrium SO,, partial pressure above the
liquor, and thus S0? removal is theoretically limited to the value
calculated on the basis of this SO,, partial pressure in the exiting
flue gas.
459
-------
DESIGN CONSIDERATIONS
A commercial double alkali system must be designed to remove the
desired quantity of sulfur oxides from a given flue gas stream, while
operating in a reliable manner and discharging environmentally acceptable
solid waste product. In fulfilling these design objectives, cost is
also an important factor.
S02 Removal
The fact that small quantities 01 sunur aioxiae can oe removed
from large amounts of relatively inert gas by cyclic processes involving
absorption into aqueous solutions of sodium sulfite/bisulfite has been
(i
known for some time. Johnstone et al. published a paper in 1938 giving
data on the vapor pressure of S02 over solutions of sulfite/bisulfite
and methods of calculating these equilibrium values under various
conditions. The equilibrium partial pressure of S0~ above sulfite/bisulfite
solutions, the theoretical limit which a practical design can approach,
is generally a function of solution temperature, pH, concentration of
sulfite/bisulfite and total ionic strength. Since Johnstone's work a
number of organizations have pursued this technology with laboratory,
pilot plant and full scale applications for flue gas desulfurization,
and many have demonstrated the ability for high removal efficiencies.
(It should be noted that although Johnstone's work was aimed at cyclic
processes with thermal regeneration, such as the Davy Powergas system,
the vapor pressure data is also applicable to double alkali systems
which use chemical regeneration.)
Once methods have been established to determine equilibrium S09
vapor pressure over scrubbing solutions, of the various concentrations
to be encountered in an operating system, it becomes a matter of
standard chemical engineering practice to design adequate gas absorption
equipment to accomplish the desired SO,-, removal in a system. For
comparison, it should be noted that the design of lime/limestone slurry
absorption equipment is further complicated by the kinetics of dissolution
of the lime or limestone, the particle size of the suspended material,
and the crystal morphology of the lime or limestone.
Reliable Operation
System reliability can be adversely affected by two classes of
problems, chemical and mechanical.
The mechanical problems include malfunction of instrumentation
and mechanical and electrical equipment such as pumps, filters,
centrifuges, and valves. These problems in a commercial FGD system
can be minimized by carefu] selection of materials of construction and
equipment and by providing spares for certain equipment such as pumps
460
-------
and motors which are expected to be in continuous operation and are
prone to failure after a relatively short period of operation.
Another important consideration in minimizing mechanical problems
is the institution of a good preventive maintenance program.
The chemical (or physical/chemical) problems which may be
associated with a double alkali system include scaling, soluble
sulfate build-up, production of non-settling solid waste product,
water balance and build-up of non-sulfur solubles which enter the
system as impurities in the coal or lime. Each of these factors
is associated with reliable system operation, or production of
an environmentally acceptable solid waste.
a. Scaling - One of the primary reasons, and probably the most
important, for development of double alkali processes was to
circumvent the scaling problems associated with lime/limestone
wet scrubbing systems. Therefore, a double alkali system should be
designed to operate in a non-scaling manner.
Scaling is caused by precipitation of calcium compounds from
scrubbing liquors, on the surfaces of various components of the
system. When this occurs in the scrubber it is particularly
troublesome since the flue gas path through the scrubber, if
affected, could cause shutdown of the boiler-scrubber system and
lower reliability.
Since scrubbing in double alkali systems employs a clear
solution rather than a slurry, there is a tendency to ignore
potential scaling problems. Testing experience with double
alkali systems has indicated, however, that scaling can occur
and indeed the problem should be a legitimate concern in the
design of any system. Both gypsum and carbonate scale build-up
has been recognized in these systems. Gypsum scaling is caused
by the reaction of soluble calcium ion with sulfate ion formed
in the system through oxidation of the absorbed sulfur dioxide
or from absorbed sulfur trioxide according to the reaction:
Ca^ + 304= + 2H20 -> CaS04-2H20 i (22)
Gypsum scaling is controlled by softening the regenerated
liquor prior to recycling to the scrubber. Softening ensures that
the liquor rscycled to the scrubber system is unsaturated with
respect to gypsum; therefore, with proper softening even if some
sulfate is formed in the scrubber, the liquor will not be saturated
461
-------
with gypsum and cause scaling on the inside surfaces of the scrubber.
In concentrated active alkali systems, a special softening step
is not necessary since high sulfite concentration is maintained
throughout the system. This sulfite maintains a low calcium ion
concentration (sulfite softening), and thus maintains the scrubbing
solution unsaturated with respect to gypsum.
Based on experience gained in lime/limestone scrubbing testing,
a certain factor of safety in the prevention of gypsum scaling probably
exists in double alkali systems. Gypsum has been found to
supersaturate easily to about 130% saturation; thus, even if sulfate
formation is higher than expected, gypsum may not precipitate in
the scrubber until the liquor is over 130% of saturation with respect
to gypsum.
Carbonate scaling usually occurs as a result of localized high
pH scrubbing liquor in the scrubber where C09 can be absorbed from
the flue gas to produce carbonate ion. This ion subsequently
reacts with dissolved calcium to precipitate calcium carbonate
scale according to the following series of reactions:
Carbon dioxide absorption by high pH liquor:
C02 + 2 OH" -v C03= + H20 (23)
Calcium carbonate scaling:
C03= + Ca++ -* CaC03 i (24)
Based on experience with the General Motors full scale double
alkali system,5 carbonate scaling could occur with scrubber liquor
pH's above 9. At lower pH, the carbonate/bicarbonate equilibrium
system tends to limit the free carbonate ion and thus prevent
precipitation of calcium carbonate:
H+ + C03= ,: HC03~ (25)
Thus, carbonate scaling can be eliminated by control of pll in the
scrubber.
b. Production of Non-Settling Solids - Under certain conditions,
the waste product solids produced in the regeneration sections of
various double alkali systems have a tendency not to settle from the
scrubber liquors. This creates problems in the operation of settlers,
clarifiers, reactor clarifiers, filters and centrifuges. Although
-------
the phenomenon has been observed in the laboratory testing conducted
by EPA on dilute systems and in the laboratory and pilot plant work
conducted by Arthur D. Little, Inc. (ADL) on dilute and concentrated
systems, it is not completely understood.3
Some of the factors thus far identified which appear to affect
the solids settling properties are reactor configuration, concen-
tration of soluble sulfate, concentration of soluble magnesium and
iron in the liquor, concentration of suspended solids in the
reaction zones, and use of lime vs. limestone for reaction. Based
on laboratory work in dilute systems (about 0.1 M active sodium)
using limestone it appears that solids settling characteristics
degraded significantly at soluble sulfate levels above 0.5 M. Based
on laboratory work with concentrated systems (about 0.45 M TOS,
5.4 pH, 0.6 M sulfate) using limestone, marked degradation of solids
settling properties occurred at a magnesium level of 120 ppm and
virtually no settling of solids occurred at the 200 ppm magnesium
level. Equal degradation of solids settling properties also occurred
in concentrated systems when the sulfate level was raised to the
1.0 M level while maintaining low magnesium level (about 20 ppm)
and keeping other variables constant.
Envirotech advocates the recycle of precipitated solids from
the thickener underflow to the reaction zones in an effort to grow
crystals which settle faster and are more easily filtered.
ADL cites reactor configuration as being important in the
production of solids with good settling and filtration characteristics.
Their basis for this is comparative tests of a simple continuous flow
stirred tank reactor (CFSTR) with the ADL/Combustion Equipment
Associates (ADL/CEA) designed reactor system under similar conditions.
The ADL/CEA reactor system appeared to give better settling solids for
a greater variety of conditions than a simple CFSTR.
Environmentally Acceptable Solid Waste
Ideally, the solid waste product produced by double alkali FGD
systems should be environmentally acceptable. Some of the properties
which can be ascribed to such a solid waste product are listed below:
•non toxic
•low soluble solids content, non-leachable
•low moisture content
•non-thixotropic
•high corapressive or bearing strength
463
-------
In order to generate waste product solids which have properties
approaching those listed, the following design considerations are
appropriate:
•sulfate removal
•water balance/waste product (cake) wash
•gypsum vs. calcium sulfite
•sludge fixation technology
a. Sulfate Removal - In double alkali systems, some of the
sulfur removed from the flue gas takes the form of soluble sodium
sulfate due to oxidation in the system, thus changing some of the
active sodium to the inactive variety. When sodium in the system is
converted to the inactive form (Na2SO/), it is relatively difficult
to convert back to active sodium. To convert inactive sodium to
active sodium, sulfate ion must be removed from the system in some
manner, while leaving the sodium in solution. The alternative to
this is to remove the sodium sulfate from the system at the rate
it is being formed in the system. This alternative is not desirable
since it is wasteful of sodium and generally is carried out by
allowing the sodium sulfate to be purged from the system in the
liquor which is occluded in the wet solid waste product.' The solid
waste product can then potentially contribute to water pollution due
to leachability. Water run-off can lead to contamination of surface
water, while leaching and percolation of the leachate into the soil
can result in contamination of the ground water in the vicinity of
the disposal site. Failure to allow for sulfate removal from
double alkali systems will ultimately result in a) precipitation
of sodium sulfate somewhere in the system if active sodium is made
up to the system, or b) in the absence of make-up, eventual deterioration
of the S02 removal capability due to the loss of active sodium
from the system.
The equations (15), (16), (17) and (18) shown previously under the
definition of "sulfate removal" describe several sulfate removal
techniques which have been used in FGD system pilot tests.
The first equation depicts the sulfate removal technique
used in dilute active alkali systems:
Na2S04 + Ca(OH)2 + 2H20 J 2NaOH + CaS04'2H20 4- (15)
(Gypsum)
Concerning the full-scale dilute alkali system installed and
operating at the Parma, Ohio, transmission plant of General Motors,
and dilute systems in general, Phillips-" stated:
464
-------
"The presence of Na2SO^ in the scrubber effluent
is the prime factor influencing the design of the
regeneration system. Na2SO^ is not easily regenerable
into NaOH using lime. The reason being that the
product, gypsum, is relatively soluble. . . . Na2SO,
cannot be causticized in the presence of appreciable
amounts of S03= or OH~ because Ca"^ levels are held
below the CaSO/ solubility product. To provide for
sulfate causticization, the system must be operated
at dilute OH~ concentrations below 0.14 molar. At
the same time, SO/= levels must be maintained in the
system at sufficient levels to effect gypsum
precipitation. ... We selected 0.1 molar OH~ and
0.5 molar S0^= as a design criteria."
In a previous paper, Phillips^ showed a plot of equilibrium caustic
formation in Ca(OH)2~Na2S04 solutions at 120°F which is the basis
for selection of the design criteria. The essence of this discussion
is that if the active sodium concentration is sufficiently dilute,
sulfate can be removed from the system by simple precipitation as
gypsum by reaction of lime with sodium sulfate.
Since, as explained above, this reaction will not proceed to a
great extent in concentrated active alkali systems, other techniques
must be employed to effect sulfate removal in these systems.
The second equation depicts a technique which is used in the full
scale double alkali systems in Japan, and which has been pilot tested
by ADL under contract with EPA:
+ 2CaS03-l/2 H20 + H2S04 + 3H20 -> 2NaHS03 + 2CaSO^'2H20 4- (16)
(Gypsum)
This technique is used to precipitate gypsum by dissolving calcium
sulfite in acidic solution thus increasing the Ca++ in solution enough
to exceed the solubility product of gypsum. Ideally according to
equation (16) 2 moles of gypsum should be precipitated for each mole
of sulfuric acid added. In practice, however, this is not the case
since any material which functions as a base can consume sulfuric
acid and reduce the efficiency of this reaction for its intended
purpose. Unreacted lime or limestone, sulfite ion and even sulfate
ion can consume sulfuric acid thus lowering sulfate removal from the
system.
465
-------
Conceivably, this method of sulfate removal is economically
unattractive in applications with very high oxidation rates, and
where the gypsum produced must be discarded. The economic picture is
considerably changed where this system is used merely as a slip-
stream treatment to supplement other sulfate removal methods and/or
where the solid product gypsum is saleable as is the case in Japan.
The third equation describes a phenomenon which has been
referred to as mixed crystal or solid solution formation:
x NaHS03 + y Na2S04 + (x+y) Ca(OH)2 + (z-x) H20 -^ (17)
(x-2y) NaOH + x CaSO-j-y CaS04-z H20 *
(mixed crystal or solid solution)
j
This phenomenon is described by R.H. Borgwardt of EPA as it applies
to lime/limestone wet scrubbing based on pilot plant investigations.
A similar phenomenon has been observed by ADL in some of their early
pilot testing of double alkali systems in conjunction with CEA, and
later in the EPA/ADL dual alkali test program.
It appears that under certain conditons the solids precipitated
in lime/limestone and double alkali systems contain sulfate, sulfite
and calcium; however, the liquor from which these solids precipitated,
appears to be subsaturated with respect to gypsum. This is based on
the fact that pure gypsum crystal could be dissolved in the mother
liquor from which the mixed crystal/solid solution was precipitated.
In addition, the solid material was examined by X-ray diffraction
and found to contain no gypsum; infrared analysis confirmed the
presence of sulfate.
Borgwardt found that the molar ratio of sulfate to sulfite in
these solids was primarily a direct function of sulfate ion activity
in the mother liquor. In pilot test work with lime/limestone
scrubbing, with little or no chlorides present and normal magnesium
level (below 1000 ppm) in solution, the sulfate to sulfite molar
ratio in the mixed crystal solids was found to reach a maximum level
of 0.23. This is equivalent to a [SO^"]/total [SOX] ratio in the
solids of 0.19.
In pilot test work with concentrated double alkali systems, ADL
observed the simultaneous precipitation of sulfate and sulfite with
calcium in lime and limestone treatment of concentrated double alkali
scrubbing liquors. This phenomenon was surprising at first, in light
of the reasoning which led to the development of dilute double alkali
466
-------
systems; i.e., gypsum cannot be precipitated from solutions
containing high active alkali concentrations. It was a simple
technique for sulfate removal in concentrated systems. The [SO/]/
total [SOX] ratio observed in pilot double alkali work was as high
as 0.02 as a maximum.3 Coincidentlv, this is the same value
observed by Borgwardt in lime/limes tone testing. This leads
to the suspicion that the same phenomenon is occurring in both
processes. The mother liquor from which these solids were
precipitated was also found to be subsaturated in gypsum, and when
the solids were examined, pure gypsum was not found.
Based on the observed data, it appears reasonable to design
a concentrated active alkali system for a particular situation in
which the system oxidation rate is below about 20%. In this case,
sulfate can be removed at the desired rate, without the necessity
for purging Na2S04 or supplementing the system with other complex
methods of sulfate removal.
The fourth equation shows sulfate removal as sulfuric acid
in an electrolytic cell:
Electrolytic
Na2S04 + 3H20 cell ; 2NaOH + H2S04 + H2 + 1/2 02 (18)
This method is the basis for operation of the Stone and Webster/
Ionics process sulfate removal technique. In Japan, Kureha/Kawasaki
has pilot tested the Yuasa/Ionics electrolytic process for sulfate
removal in conjunction with their double alkali process. They feel
that this process will be less expensive overall than the presently
used sulfuric acid addition method. In addition, they feel that
sodium losses from the system can be cut in half through the use of
this method: from 0.018 moles Na loss/mole S02 absorbed to 0.009
moles Na loss/mole S02 absorbed.
Another approach to sulfate control is to limit oxidation. With
sufficient limitation of oxidation, by process and equipment design,
it may be possible to control sulfate by a small unavoidable purge
of Na2SO^ with the solid waste product. To design for minimum oxidation,
there should be minimum residence times in equipment where the
scrubber liquor is in contact with oxygen-containing flue gas, and
all reactors, mixers, and solids separation equipment should be
designed to minimize absorption of oxygen from air. In addition,
it has been reported^ that oxidation of scrubber liquors can be
minimized by maintaining very high ionics strength. One possible
467
-------
explanation for this is that high ionic strength liquors are poor
oxygen absorbers and that oxidation in these systems is oxygen
absorption rate limited.
b. Water Balance and Waste Product (Cake) Washing - In order
to operate a closed system to avoid potential water pollution problems,
system water balance is a primary concern. Water cannot be added to
the system at a rate greater than the normal water losses from the
system.
Generally there is a tendency to add fresh water to a scrubbing
system to serve many purposes. These include:
• saturation of flue gas
• pump seal needs
• demister washing needs
• slurry make-up needs
• waste product washing
On the other hand water should only leave the system in the
following ways:
• evaporation by the hot flue gas
• water occluded with solid waste product.
• water of crystallization in solid waste product
Careful water management, part of which is the use of recycled
rather than fresh water wherever possible, is necessary in order to
operate a closed system.
As previously indicated, disposal of wet solid waste containing
soluble salts is ecologically undesirable. In addition, allowing
active alkali or sodium salts to escape from the system is an
important operating cost factor. Sodium make-up to double alkali
systems is usually accomplished by adding soda ash (recently quoted
at $49 per ton) at some point in the system. Thus, both ecological
and economic considerations dictate that waste product washing is
desirable.
A rotary drum filter, belt filter or centrifuge is usually the
equipment in which the final solids separation is made. This
equipment can be designed to permit solids washing with fresh water.
468
-------
One concern in waste product washing is the extent to which the
cake should be washed. At present, there are no federal regulations
concerning the amount of total dissolved solids (TDS) which is
permitted in a waste product which is to be disposed of as landfill.
The future, however, may unfold stringent regulations in this area.
One obvious consideration in waste product washing is system water
balance. Unlimited waste product washing is not possible if a closed
system operation with no liquid stream discharge is a goal. Another
more subtle reason for limiting waste product washing is the potential
problem of non-sulfur/calcium solubles build-up in the system. These
non-sulfur/calcium solubles enter the system with the flyash, flue gas,
the lime and/or limestone and the make-up water. Probably the soluble
material in highest concentration would be sodium chloride which results
from the absorption of HC1 from the flue gas by the scrubber solution.
A material balance around the system at steady state necessitates
that solubles leave the system at the rate they enter. Thus, depending
upon how well the waste product is washed, a certain level of non-sulfur
solubles will be established in the system. Since the only mechanism
for these solids to leave the system is as part of the wet solid
waste, a certain purge rate is necessary. This purge also necessitates
the loss of some sodium from the system. Practical limitations in
filter design and water balance probably would limit a system to two
or three "displacement washes" of the waste product (one displacement
wash means washing with an amount of fresh water equivalent to the
amount of water contained in the final wet waste product per unit of
waste) . Depending on the characteristics of the waste product and
the design of the washing system, one displacement wash can reduce
the solubles content of the waste product by as much as 80%.
c. Gypsum vs. Calcium Sulfite - Although gypsum
is more soluble than calcium sulfite hemihydrate (CaSQ^' 1/2
gypsum may be considered a more environmentally acceptable end
product. The solubility of gypsum in water is about 0.25%; that
of calcium sulfite is on the order of 0.0025%. It is interesting to
note that while gypsum is a naturally occurring mineral, calcium
sulfite is not found in nature. In the solid x^/aste product (sludge)
from lime/limestone and double alkali FGD systems, gypsum has better
handling properties than calcium sulfite. Sludges containing a
high ratio of gypsum to calcium sulfite are less thixotropic, better
settling, more easily filtered, and can be more completely dewatered
than sludges containing a high proportion of calcium sulfite. Another
important characteristic x^hich has been attributed to high gypsum (as
opposed to calcium sulfite) sludges is their higher compressive or
bearing strength. Typically, lime/limestone systems which generate
469
-------
solids having a high proportion of calcium sulfite can be filtered
to 40-50% solids, while some double alkali systems producing high
gypsum cake can be filtered to over 65% solids.^0
Some explanation for the behavior of these sludges is given by
Selmeczi and Knight.H Although filter cakes appear dry, they still
contain a considerable amount of water and thus, upon vibration or
application of stress, they have the tendency of again becoming fluid.
This thixotropic property and high moisture content are both explained
by the morphology of calcium sulfite clusters. Due to the highly
open, porous or sponge-like nature of these clusters, a considerable
amount of water is retained in the clusters. The calcium sulfite
crystals are rather fragile and break under pressure releasing some
of the water, which results in the sludge becoming fluid.
It is possible in some double alkali systems to produce a high
gypsum product. In Japan, where by-product gypsum is a saleable
product, the calcium sulfite solids produced are oxidized completely
to gypsum in a separate oxidation process tacked on to the tail end
of the system. In applications where high excess combustion air is
present in the boiler or where low sulfur coal is burned or a
combination of these conditions is present, the oxidation rate in the
system tends to be high (possibly about 90%) and the proportion of
gypsum in the sludge tends to be high. In some dilute systems the
proportion of gypsum in the sludge can be increased by augmental
aeration of the scrubbing liquor.12 Crystal seeding techniques used
in conjunction with augmental aeration can produce relatively coarse
grained gypsum crystals with good dewatering and structural properties
in the final waste product.
d. Sludge Fixation Technology - Chemical or physical fixation of
the sludge produced in a double alkali system is another potentially
important means of producing an environmentally acceptable solid waste
product. This technology is under investigation by I.U. Conversion
Systems, Inc., Chicago Fly Ash, Dravo Corporation, and Chemfix
Corporation. Most of their efforts are concentrated on sludge produced
from the more prevalent lime/limestone systems; however, there is also
some evaluation of double alkali sludges. The objective of sludge
fixation technology is the production of a non-toxic, unleachable
solid waste product which has reasonably high load bearing strength.
If double alkali sludges are amenable to this type of treatment, the
need to reduce soluble sulfates in the solid waste product becomes
470
-------
less of a problem. Some sodium sulfate has been found to be
physically or chemically tied up in the solid calcium sulfate/
sulfite crystal lattice;-^ however, the extent of this phenomenon is
not generally considered to be adequate to remove all of the sodium
sulfate produced via oxidation. Sludge fixation technology may
be a mechanism by which additional sodium sulfate can be removed
from the system without adverse environmental effects. There is
some concern as to whether this is viable, however, since sludge
fixation chemistry involves pozzolanic reactions between calcium
compounds and flyash components in the sludge x^hich may only
involve multivalent ions rather than monovalent sodium. In other
words, monovalent ions such as sodium may either a) not take part
in the pozzolanic reactions, or b) inhibit or limit such reactions.
Further investigation is called for in this area.
Dilute vs. Ccmcentrated System
The selection of a dilute or concentrated double alkali system
is an important design consideration in any application. In general
it can be stated that the concentrated systems are more suited
to applications in which oxidation is expected to be relatively
low, and that, conversely, dilute systems are favored in applications
where oxidation rates are high. High sulfur Eastern coal applications
on utility boilers where excess air is controlled carefully and
maintained at the lowest value consistent with complete combustion,
the concentrated systems may be favored. On the other hand, in
utility or industrial boiler applications where Western low sulfur
coal is burned, and/or where control of oxidation is difficult due
to high excess air, the dilute systems may be favored.
Oxidation rate is promoted when low sulfur coal is burned, since
the ratio of oxygen to sulfur dioxide in the flue gas is higher than
in high sulfur coal applications. Since oxidation is a strong function
of the rate of absorption of oxygen, liquor which is dilute in TOS
is subject to having a greater proportion of these species oxidized
by a given amount of absorbed oxygen than one in which the TOS is
more concentrated.
Under a given set of conditons without consideration given to
waste disposal, a concentrated system can be installed at lower
capital cost than a dilute system as previously discussed; however,
471
-------
the desirability to produce a manageable solid waste (dilute systems
can be designed to produce high gypsum sludges) could, in cases,
override the economic issue.
Costs
Based on cursory design and cost analysis, with the assumption
that a single scrubber device can be used to remove particulates and
SO? to the extent required to meet new source performance standards,
the capital and operating costs of double alkali systems appear to be
significantly less than those for a lime or limestone system designed
for the same requirements.-^
Generally, the dilute active alkali systems tend to be higher
in capital costs than the concentrated systems since both equipment
size and process flows are required to be larger to accomplish the
same degree of desulfurization. Methods used for sulfate removal and
for reduction of scaling tendencies will affect both capital and
operating costs.
Since there are presently no full scale utility applications in
the U.S., a discussion of projected costs for such an application
can only be based on costs for actual industrial boiler systems,
vendor cost estimates and other engineering estimates based on
hypothetical cases. Actual capital costs for industrial boiler
applications tend to be high on a per KW basis compared to what would
be expected in a large scale utility system because a) these systems
are first-of-a-kind applications and consequently are designed for
various operating contingencies, b) they are relatively small scale
systems—economies of scale would be expected in larger scale
units, and c) there are more than the usual equipment redundancies—
e.g., four separate scrubbers to control only 32-40 Mw equivalent of
flue gas. Vendor cost estimates tend to be less than complete and
thus low, since they generally do not take into account all of the
details that need to be considered for a specific application with
its unique requirements. Engineering estimates are just that,
estimates based on certain reasonable assumptions.
With the above qualifications, some double alkali system costs
can be given.
a. Actual Industrial Boiler System^ - The General Motors
industrial boiler system was installed for about $3.5 million.
Based on flue gas rate, it is presently equivalent to a 32 Mw system;
472
-------
however, the regeneration system is designed to handle 40 Mw equivalent.
The GM system is a dilute system.
Unit Capital Cost: $88/Kw
Unusual features:
•4 separate stainless steel scrubbers
•2 separate filters
•overdesigned reactor clarifiers
•first-of-a-kind system
b. Vendor Cost Estimate - FMC Corporation1^ has estimated the
cost of a double alkali S02 removal process for a 150 Mw utility boiler
system to be about $6 million. Some of tne assumptions and components
of this estimated cost are shown in Table 1.
Unit Capital Cost Estimate: $40/Kw
c. Engineering Cost Estimate - TVA has prepared a preliminary
engineering cost estimate for a limestone double alkali process on
a 500 Mw, new, coal fired power unit burning 3.5% sulfur coal designed
for 90% S02 removal with on-site solids disposal. ->
Unit Capital Cost Estimate: $61/Kw
For comparison, on a similar basis, the TVA estimate for a
limestone slurry process is $50/Kw. Note that this comparison is
in conflict with the cursory design referred to above.
Obviously the data base for double alkali system costs is
extremely thin compared to that for lime/limestone systems. The
costs cited above are given only to establish a basis for further
consideration.
473
-------
Table 1. ESTIMATED CAPITAL COST
FMC CORPORATION S02 CONTROL PROCESS
Retrofitted Parallel Scrubbers for 1-150 Mw Utility Boiler
and
Complete Regeneration Plant
Fully Instrumented, Installed & Operable
Thousand $
Materials _ Installation Total
DIRECT COSTS
Scrubbing Section 497 430 927
Regeneration 381 706 1,087
Thickener and Offsites 465 310 775
Buildings, Foundations
and Structural 428 436 864
SUB TOTAL A 1,771 1,882 3,653
INDIRECT COSTS
Engineering
(17,000 MH @$16/hr.) 272
Field Indirects
(140% of Direct Labor) 1,330
Fee
(5% on $5,255,000) 263
SUB TOTAL B 1,865
OTHER COSTS
Proprietary Process Fee
(Based on 150 Mw) 198
Contingencies
(5% on $5,518,000) 276
SUB TOTAL C 474
TOTAL A + B + C 5,992
NOTE: Includes chemical storage, reheaters, building, structural
steel, flue gas ducting, waste solids handling and storage,
insulation, steam tracing, startup, and operator training
in addition to complete scrubbing system.
474
-------
STATUS OF TECHNOLOGY
A number of variations of double alkali FGD systems have been
tested and studied extensively during the past few years by equip-
ment vendors, potential users, and EPA. Some of the results of test-
ing and a description of some significant systems tested will be
discussed here. In addition, some of the planned full scale installa-
tions will be discussed to illustrate the commitment made by industry
to this technology.
In the United States, Arthur D. Little/Combustion Equipment
Associates (the two companies have agreed to develop and market the
technology jointly), Envirotech Corporation and FMC Corporation have
all conducted extensive pilot plant testing with the aim of marketing
sodium based double alkali systems for FGD applications on utility and
industrial boilers. General Motors Corporation has undertaken to
develop a double alkali process for use in their industrial boilers
through pilot plant testing and, later, the installation of the first
full-scale industrial boiler application of this technology in the
country. General Motors does not intend to market their system;
however, the results generated through their efforts will be in the
public domain.
In Japan, the Showa Denko KK Company and Kureha Chemical Industry
Company/Kawasaki Heavy Industries have developed sodium based double
alkali systems through pilot plant testing and have gone on to relative-
ly large full scale FGD applications which are currently operating.
Due to different economics in Japan, the processes developed and
installed by these companies produce saleable by-product gypsum rather
than a disposable solid waste product. Another interesting difference
between the development of technology in Japan and the U.S. is the use
of limestone rather than lime as the source of calcium for the Japanese
processes. In the U.S. the major emphasis in development and in
commercial application has been with lime.
This discussion on the status of technology is presented by system
developer or vendor, with emphasis on:
'System descriptions
•Test results/operating experience
•Plans for future testing or full-scale application.
475
-------
Envirotech Corporation
The Envirotech testing is being cqnducted in a 2500 ACFM pilot
plant at the Gadsby Station of Utah Power and Light Company in Salt
Lake City, Utah. The pilot plant was constructed in late 1971. Test-
ing began in January 1972. Flue gas is supplied to the pilot plant
from No. 3 boiler, a 100 Mw coal fired unit.
a. System Description - Envirotech has tested and continues to
test a number of configurations in the pilot plant. Particulate
loading in the flue gas can be varied considerably since the flue gas
take-off point from the power boiler can be upstream or downstream of
a mechanical separator and an electrostatic precipitator or between
the two. Two types of scrubber systems have been tested, a polysphere
scrubber containing two trays of balls, and a venturi/cyclone separator
scrubber. Two types of mist eliminators were tested, a spin vane in
series with a cyclone, and a Euroform chevron type (this type of curved
chevron demister allows the de-entrained liquid to flow from the system
perpendicular to the gas path). Although the majority of testing was
conducted in a dilute mode of operation which Envirotech calls the
"sulfate mode" due to employment of augmental oxidation, they have
recently also begun testing a concentrated mode of operation.
Envirotech has stressed equipment development in conjunction with
process development. Use of reactor clarifiers is stressed to main-
tain high suspended solids content in the regeneration reaction zones.
Quenching the incoming flue gas with recycle liquor (as opposed to
fresh water) to avoid water balance problems has been stressed. Liquor
introduction in a "dentist bowl" is used as a technique to avoid wet/
dry interface fouling problems, while still providing the x^ater needed
to quench the hot flue gas.
Figure 1 is a slightly modified version of the basic sulfate mode
flow scheme tested at the Gadsby station pilot plant by Envirotech. Al-
though the majority of the testing was done with a polysphere scrubber,
and the remainder with a venturi, this flow sheet incorporates both.
Augmental oxidation is accomplished in a scrubber recycle tank by sparg-
ing with atmospheric air. (In the pilot plant over 90% oxidation was
obtained with only a crudely designed sparging system; i.e., a rubber
hose blowing air below the liquid surface.) This oxidation process con-
verts most of the soluble sulfite to sulfate and thus the name, "sulfate
mode". The purpose of this is to produce a waste product consisting
mostly of gypsum and therefore having desirable settling, filtration
and dewatering characteristics. This is workable, consistent with
adequate sulfate removal, since the system operates in a dilute mode.
Sodium sulfate is essentially converted to sodium hydroxide by lime
precipitation of gypsum according to equation (15), with little inter-
ference of sulfite ion. The extent to which hydroxide can be regener-
476
-------
RECIRC. WASH WATER
FLUE GAS
IN
VENTURI
QUENCHER
1 SCRUBBER
OXIDATION jf
TANK
PUMP
FRESH WATER
_J. MAKEUP
HP
GAS
OUT
FINAL MIST
ELIMINATOR
FILTER
VACUUM
PUMP
RECEIVER
CAKE WASH
~1 RFRESH WATER)
SLAKED LIME FILTRATE
PUMP
ROTARY
VACUUM FILTER
(BELT TYPE)
SULFATE CAKE
(70-80'* SUSPEN.
SOLIDS)
WASH WATER BLEED LIQUOR
TANK TANK PUMP
REACTOR-
CLARIFIER
REACTOR
CLARIFIER -
SUSPEN.
SOLIDS
PUMP
SUMP
FIGURE 1. ENVIROTECH SULFATE - MODE DOUBLE ALKALI
-------
ated in this way is limited to the range of approximately 0.1 molar OH
concentration.
The scheme makes use of two reactor clarifiers, a thickener and
continuous rotary drum vacuum (belt type) filter for solids separation.
The reactor clarifiers are units designed for a combined job of
reaction, clarification, and solid particle growth. Crystal growth
and continued reaction are accomplished by internal circulation of
suspended solids in the central zone of the equipment. The remainder
of the equipment serves to split the feed slurry into a clarified
liquor overflow and a thickened slurry underflow product. Reactor
clarifier No. 1 serves these functions for the reaction tank effluent
slurry. Reactor clarifier No. 2 is used for the softening step. Soda
ash is added here serving the dual purposes of a) sodium make-up
reagent and b) carbonate supplying reagent for softening. The soda ash
reacts with soluble calcium, replacing some soluble calcium with soluble
sodium, and precipitating calcium carbonate according to equation (19).
This is carried out in the central reaction zone where the precipitated
solids are caused to coagulate (grow) so that they may be more easily
separated from the mother liquor. Clear liquor overflowing from reactor
clarifier No. 2 is thus softened (subsaturated with respect to gypsum)
and can be fed to the scrubber circuit as regenerated liquor with no
gypsum scaling potential. It should be noted that reactor clarifier
No. 2 was added to the pilot plant only after significant gypsum scaling
was observed in the early testing in 1972. Coagulant, an organic chemical
which may be added to the reactor clarifiers to the extent of a few parts
per million, can serve to aid solids separation.
The thickener serves to make a gross separation of product solids
from mother liquor which cuts down on the duty of the filter where the
final separation and washing of the product solids occur.
The equipment arrangement used by Envirotech for testing a concen-
trated mode of operation is somewhat simpler compared to that used in
the sulfate mode. The absorption section consists of a venturi/cyclone
separator with recirculation. The regeneration system essentially con-
sists of two series stirred tank reactors; the solids separation section
consists of a single thickener and continuous rotary filter.
b. Test Results - General conditions and performance data for the
Envirotech pilot plant operated in the sulfate mode are tabulated below:
478
-------
Flue Gas
Source: pulverized coal boiler flue gas
Fuel: coal, 0.4% Sulfur
Oxygen content: 5%
Flow Rate: 1500-3000 ACFM @ 220°F
SC>2, inlet concentration: 250-1500 ppm
Particulate, inlet concentration: 0.2-0.3 gr/SCF
Absorber System
Absorber Type: Polysphere Venturi
S02 Removal: 90% 80-85%
S02 Exit concentration: 15-40 ppm
Particulate exit
concentration, gr/SCF: 0.0.1-0.02
L/G ratio, gal./lOOO ACFtf: 20 24
Scrubber pressure drop,
in. H20: 4-10 6-8
Recycle tank/oxidizer
residence time, min.: 10-15
Scrubber liquor —
Inlet, pH: 7.5-7.7
Exit (typical), pH: 5.8-7.1 (6.5)
Oxidation rate, %: ^90
Regeneration System
Lime stoichiometry: 110-115% of S02 removed
Reactor residence time: 30 minutes, total for both reactors
Regeneration liquor composition: 6.5 pH inlet
12.2-12.5 pH exit
0.1 Molar hydroxide, exit
40,000-45,000 ppm S04=
1000 ppm Cl~
^500 ppm Ca4^" (approxi-
mately 100 ppm below
saturation of
479
-------
Solids Separation
Residence times: Reactor clarif iers : 1-2 hrs
Thickener: 3-4 hrs
Waste product composition: 30-40% water
1-2% flyash (dry basis)
2% solubles (dry basis)
10-15% unreacted CaC03 or
Ca(OH)2 (dry basis)
remainder: CaSOx solids
(mostly CaS04'2H20)
Waste product wash: 1 displacement wash,
80% reduction in solubles
Discussion - The conditions under which the pilot plant
operated, for the most part, simulated typical conditions for a utility
boiler fired with Western low sulfur coal. The actual range of S02
concentration in the boiler flue gas was 250-400 ppm since coal contain-
ing approximately 0.4% sulfur was fired in the boiler. A much larger
range of inlet S02 concentration is, however, indicated above, due to
the. fact S02 was added to the flue gas during some of the later testing
to simulate conditions typical for a utility operating with a higher
coal sulfur content. The majority of the testing with the higher S02
inlet levels was conducted with the venturi scrubber. During this test-
ing the flue gas flow rate was at the lower end of the indicated operating
range (M.500 ACFM) .
The majority of the testing with the polysphere scrubber was con-
ducted at a total pressure drop of 4 inches of water using two trays
of balls. Figure 2 shows the performance of a single polysphere tray;
S02 removal is plotted against L/G ratio in gal./lOOO ACF at three
different pressure drops (4, 7 and 10 in. H^O) at pH in the range of
5.8-7.1. Figure 3 shows a plot of S02 removal vs. scrubber effluent
pH at an L/G ratio of 19 gal./lOOO ACF at a total pressure drop of
4 in. H20 across two polysphere trays. It can be seen that, generally,
two trays operating at a pressure drop of 4 in. H20 are equivalent to
one tray operating at a pressure drop of 7 in.
S(>2 and particulate levels achieved in the treated flue gas were
well within the requirements of federal new source performance standards
based on the data reported.
Lime stoichiometry reported for most of the testing was in the range
of 110-115% based on S02 removed. Envirotech has forecasted, however,
that utilization of lime could be improved so that in a commercial
application one might expect to achieve lime stoichiometry in the range
of 100-110% based on SC>2 removed,6
480
-------
100
95-
90
LU
CC 85
r*
o
80
75 •
t
10
20
30
40
L/G — Gal/MACF
FIGURE 2. SO2 removal vs. L/G. Coal fuel—
electrostatic precipitator discharge. pH;
5.8 - 7.1, single-stage.
481
-------
100
90l
Q
111
80
LU
DC
«
o
W 70
60
o o
o
345678
SCRUBBER EFFLUENT pH
FIGURES. SO2 removal vs. pH level. Coal
fuel—electrostatic precipitator discharge.
L/G, 19 gal/1,000 cu.ft., AP = 4 in. H,O.
Two-stage operation.
482
-------
The scrubber/regeneration liquors had a chloride content in the
range of 1000 ppm at steady state. Soluble chloride is expected to
concentrate in most wet scrubbing systems due to the conversion of
coal chloride to HC1 which appears in the flue gas and is absorbed by
scrubber liquors. There were no apparent detrimental effects noted for
chloride levels in the range experienced in this double alkali appli-
cation.
The solid waste product, primarily gypsum, appeared to settle and
dewater readily. The filter cake end product contained 60-70% solids,
generally better than that produced in typical lime/limestone appli-
cations. The filter cake, however, contained about 2% solubles, some-
what higher than that expected in a typical lime/limestone system, but
probably as low a value as has been obtained in steady state closed
loop operation of a double alkali system. Experiments indicated that
a single displacement wash could achieve as much as 80% reduction in
soluble solids. The use of additional wash could conceivably lower
the cake solubles content; however, in actual practice, the design of
the filter, buildup of non-calcium/sulfur solubles in the system,
applicable regulations concerning solids disposal and economic
considerations would determine the need for additional washing.
Reliability - The pilot plant testing began in January 1972
and continued for 3 months during which period some gypsum scaling
problems were encountered. The pilot unit was shut down for a period
of about 5 months between the late spring and fall of 1972. During
the 5 month shutdown a reactor clarifier was added to the flow scheme
to accomplish the carbonate softening step which reportedly eliminated
subsequent gypsum scaling problems.
The basic sulfate mode flow scheme was then operated for almost
2 years with no shutdown-causing, operating problems between fall of
1972 and August 1974. The pilot plant was operated mostly by a single
technician per 8-hour shift for 24 hours per day, 5 days per week.
The unit was shut down on weekends for convenience; however, there was
no drainage of scrubbing or regeneration solutions or cleaning of
equipment during the weekend shutdown periods. In November/December
1973 there was a brief shutdown period in which the venturi scrubber
was tied into the system. After August 1974, testing of the sulfate
mode was discontinued so that concentrated active alkali systems which
Envirotech has termed "sulfite mode" could be tested and evaluated for
comparison with their dilute sulfate mode of operation.
483
-------
FMC Corporation
FMC began development work on a sulfur dioxide control system in
the early 1960's at the Company's Modesto, California, chemical plant.
In this plant, barium and strontium sul'fate ores are processed in high
temperature reduction kilns to produce barium and strontium sulfides
with some evolution of SO^. After some initial experimentation with
lime and limestone scrubbing processes in which severe scale formation
problems were encountered, FMC began work on the double alkali process.
They developed a concentrated active alkali scheme which was put into
service controlling S02 emissions from two reduction kilns at the
Modesto plant in December 1971. This was probably the first U.S. full-
scale application of double alkali technology.
a. System Description - A typical schematic of the "FMC/Link Belt
Alkaline Absorption Process" is shown in Figure 4.
General
Absorption - The scrubber system depicted here is what FMC
refers to as their "dual throat" venturi and cyclonic separator. Absorp-
tion of S02 occurs by conversion of sulfite to bisulfite in the venturi
(equation (2)). Due to the fact that FMC employs a very concentrated
active alkali scrubbing liquor which is highly buffered by the sulfite/
bisulfite equilibrium system, there is only a very small change in pH
between the inlet and outlet of the venturi, on the order of 0.1 pH unit
(e.g., typical operation of the FMC system might involve venturi inlet
liquor pH of 6.5 and effluent pH of 6.4). Gypsum scaling does not occur
in these systems since the calcium ion concentration is held well below
that required for gypsum precipitation by the high sulfite concentration
in solution. This can be referred to as sulfite softening (equation (20),
Carbonate scaling is not a problem in operation at these relatively low
pH's due to the carbonate/bicarbonate equilibrium system which tends to
depress the carbonate concentration and favor soluble bicarbonate.
Make-up sodium is fed as soda ash directly to the scrubber recir-
culation tank. No carbonate softening is intended in this step.
Regeneration - A relatively small bleed of scrubber liquor
is taken from the scrubber recycle loop to the regeneration system.
The regeneration system consists of a continuous flow stirred tank
reactor with a 5 minute residence time and a thickener. Regeneration
in this concentrated scheme involves only conversion of some of the
bisulfite to sulfite by reaction with lime (equation (3)). There is no
free hydroxide ion formation. Clarified liquor from the thickener rich
in sulfite is fed back to the scrubber recirculation tank to raise the
capacity of the scrubber liquor for absorption of
484
-------
120 PSIG STEAM
FLUE GAS
FAN
1
T
TO STACK
REHEATER
CONDENSATE
IN
30
SCRUBBER
RECIRCULATING
PUMP
SPARE
PUMP
SODA ASH
STORAGE
BIN
V
LIME
STORAGE
BIN
V
RECIRCULATING
TANK
30 PSIG
PLANT WATER
LIME
REACTOR
FILTRATE
RECEIVER
THICKENER
THICKENER
UNDERFLOW
PUMP
r
AND
CaSOs
EXHAUST
FILTER
VACUUM
PUMP
1
RATE
URN
MP W-
FILTRATE
RETURN
PUMP
I
SURGE TANK
FIGURE 4. FMC/LINK-BELT ALKALINE ABSORPTION PROCESS FOR SULFUR DIOXIDE CONTROL
SIMPLIFIED FLOW SHEET
-------
Solids Separation - Solids separation begins in the thick-
ener which produces a concentrated slurry underflow for feed to the
continuous rotary vacuum filter which makes the final separation of
solids. The FMC system is designed for simultaneous removal of flyash
and S02? therefore, the waste solids will contain a considerable amount
of ash if the system is employed in a coal fired boiler application in
which there is no separate provision for removal of flyash upstream of
the control system.
Modesto System Description - The double alkali system in
operation at the FMC Modesto chemical plant processes a combined
30,000 ACFM gas stream (550°F) containing 5000-8000 ppm S02 from two
barium/strontium reduction kilns. The scrubber system consists of a
tower (packed with 9 feet of Intalox saddles) and a recirculation tank.
The regeneration system consists only of a single stirred tank reactor
fed with lime and a bleedstream for the scrubber loop. Solids separa-
tion is accomplished in a rotary vacuum filter producing a solid waste
containing 40-50% solids. There are no provisions for waste product
washing. The overall system at Modesto is equivalent to 7-10 Mw based
on exit flue gas flow, and possibly up to 35 Mw based on SC^ removal
capabilities.
b. Operating Experience and Testing - FMC has operating experience
in the system at the Modesto plant since start-up in December 1971. In
addition, FMC constructed a 3500 ACFM pilot plant in a mobile 40 foot
trailer which has been tested at various locations on at least seven
different flue gas and other waste gas streams for control of S02 and
particulates. The basic flow scheme in the mobile pilot plant is that
previously described and depicted in Figure 4.
Modesto System - This system has an enviable reliability
and performance record since start-up in December 1971. As of early
September 1974, the system had been on line for 22,800 hours at 100%
availability for control of kiln off-gases. The kilns had scheduled
shutdowns approximately every 3 or 4 months for maintenance and/or
change of chemical process service; however, between December 1971
and September 1974, they were in service approximately 95% of the time.
Operation of the system is now considered routine rather than experi-
mental. With inlet S02 levels of 5000-8000 ppm, the system has achieved
as high as 99.7% S02 removal; it routinely runs with less than 100 ppm
in the exit off-gas stream.
The waste product solids from the system are discharged from a
rotary vacuum filter at about 40-50% solids but there is no waste
product washing. The discharged solids ara reslurried to about 20-25%
solids with other plant waste water streams to be sluiced out to an
on-site disposal pond. The sludge dewaters by solar evaporation and
percolation into the soil surrounding the pond. About once a year some.
486
-------
of the drier sludge is dredged from the pond for disposal elsewhere
to create more on-site "running room".
The only serious drawback to the system is the large sodium loss
due to the lack of waste product wash. The sodium loss in a concentra-
ted system without cake x^ash is probably greater than that which could
be tolerated in a large scale FGD application. The loss is both costly
and a potential water pollution problem.
Pilot Plant Testing - Data reported by FMC which were
taken in a 3 week test, 8 hours per day, in the FMC mobile pilot plant
processing a slip stream of industrial boiler flue gas are tabulated
below:
Flue Gas
Boiler type: Stoker-fired coal boiler
Fuel: Coal, 4.8% sulfur
Flue gas flow: 2,578 ACFM @ 400°F
Flue gas composition: N2 = 76.3%
02 = 7.6%
C02 - 11.4%
H20 =4.7%
100.0%
S02 inlet = 3363 ppm
S02 exit = 343 ppm
Particulate inlet = 2.4 gr/SCF
Absorption
SO2 removal: 89+%
Absorber type: Dual throat venturi
L/G ratio: 13.6 gal./lOOO ACF
Pressure drop: 10 in. H20
Liquor composition: pH 6-7
total sodium >2 Molar
active alkali >0.5 Molar
Regeneration
Lime stoichiometry: - 105% based on S02 removed
Soda ash addition: f 0.026 moles Na2CC>3/mole S02
removed
Reactor residence time: 5 minutes
Thickener rise rate: 0.3-0.4
487
-------
Solids Separation
Filter cake composition: CaSC>3-l/2 H20 47.93%
Ash 13.95%
CaC03 1.14%
Sodium as Na2SO^ 1.18%
H20 35.80%
100.00%
Solubles in waste product: 1.84% (dry basis)
Soluble sulfur loss: 0.027 moles soluble sulfur/
mole S02 removed
Filter cake wash: 2 displacement wash or less,
90% reduction in sodium
Discussion - The results reported are typical for oper-
ation of the FMC concentrated double alkali system in a high sulfur coal
application. As previously discussed, the concentrated systems are
better suited to high sulfur, low oxygen (rather than low sulfur, high
oxygen) applications due to the necessity to control sulfate formed by
oxidation in the system. Since sulfate formation rate in the system
is related to the rate of oxygen absorption, sulfate will comprise a
smaller fraction of the total soluble sulfur species in a concentrated
system than in a dilute system. Thus the soluble sulfur loss is a
small fraction of the S02 removed in these applications; however, it
can be expected to be a larger fraction in low sulfur fuel applications
if all other factors remain constant.
The FMC system is not designed for sulfate removal by any of the
chemical mechanisms described under the term "sulfate removal"; thus,
there must be a loss of soluble sulfate from the system at the rate
at which it is formed. This soluble sulfur loss occurs as solubles
contained in the waste product cake moisture. Apparently, operation
at the very high ionic strengths employed by FMC minimizes the removal
of soluble sulfate via the mixed crystal or solid solution mechanism
described by equation (17). On the other hand, the high ionic strength
liquor employed by FMC probably absorbs oxygen at a lower rate than the
less concentrated systems, and thus reduces sulfate formation by reduc-
ing oxidation rate. The soluble solids level reported in the above data
by FMC appears to be as low percentagexjise as any of the values report-
ed for other throwaway product systems in the U.S. In general, FMC has
stressed lowering sulfate formation rates rather than methods of sulfate
removal.
Another important factor which influences oxidation rate is scrubber
type. FMC has conducted tests with a packed tower, venturi, valve tray
(Koch) scrubber and a "disc and donut" scrubber, and has correlated
oxidation rate with scrubber type. As expected, the scrubbers x^ith the
larger liquor residence times exhibit the higher oxidation rates.
488
-------
Table 2 presents a summary of the FMC testing with the 3500 ACFM
mobile pilot plant. A wide range of testing experience is shown for
the mobile pilot plant. No scaling problems or other major reliability
problems have been encountered in this testing.
c. Future Applications - FMC has four double alkali systems
under construction, and a fifth in the design phase in the U.S. The
largest planned U.S. application of the FMC system is a 150 Mw system
for the control of S02 from the pulverized coal boilers at FMC's
Green River, Wyoming, plant. The system is scheduled to start up in
the summer of 1975. In addition, FMC has licensed the Ataka Construc-
tion and Engineering Company of Osaka, Japan, to install their system
in Japan. The first application of the FMC technology in Japan will
be for the control of S02 from a 45,000 Ib/hr steam boiler at a paper
manufacturing company. That installation is scheduled to start up in
the first quarter of 1975.
A summary of the FMC systems planned for near future operation is
given in Table 3.
489
-------
Table 2. FMC 3500 ACFM MOBILE PILOT PLANT TESTING
Application
1. S02 and flyash,
stoker boiler
(FMC Plant
So. Charleston,
West Virginia)
2. S02 and flyash
stoker boiler
3. S02 and flyash,
stoker boiler
(General Motors
Plant)
4. S02 and flyash
-e- on bark oil and
o coal fired boiler
5. S02 and flyash,
pulverized coal
boiler
6. S02 and flyash,
pulverized coal
electric power
boiler
7. S02 control,
sulfuric acid
plant
Pressure
Test Time S02 Removal Particulate Scrubber Drop Temp.
(Months) (%) Removal (%) Type (in.H20) (°F)
6 90+ 99+ Dual 10-15 400
Throat
Venturi
3 90+ 99+ Dual Throat 10-15 350
Venturi
4 90 99 Dual Throat 10 390
Venturi
2 90 99.5 Dual Throat 10 iOO
Venturi
1 89+ 99+ Dual Throat 10 +00
Venturi
4 90+ Dual Throat 4.5-10 150
Venturi,
Koch Tray,
Disc & Donut
1 90+ Dual Throat 4.5-10 L20
Venturi,
Disc & Donut
Flue Gas
02 Content Inlet S02
(%) (ppm)
12 400-1400
10 1800
9-10 600-1200
7 1000
7 3300
5 1000-1200
6 1200
-------
Table 3. FUTURE FMC DOUBLE ALKALI APPLICATIONS
Application
1. Stoker fired
industrial
boiler system
2. Oil and coal
fired boiler
(Firestone,
Potts town, Pa.)
3. Pulverized coal
boiler
(FMC's Green
River, Wyo.
Plant)
4. Oil fired
boilers
5. Heating plant
coal boiler
6. Paper plant,
boiler
(Japan)
Size
(Mw, Equivalent)
50
150
13
14
Scrubber
Type
Dual Throat
Venturi
Dual Throat
Venturi
Disc & Donut
Disc & Donut
Disc & Donut
S02 Inlet
(ppm)
1800
1400
1000
250
Start-Up
Date
First Quarter
1975
December 1974
Mid 1975
Construction to
start in 1975
Late 1974
First Quarter
1975
-------
General Motors
General Motors (GM) Corporation developed a dilute double alkali
system through pilot testing in a 2800 CFM cross-flow packed bed
scrubber at the GM Chevrolet, Parma, Ohio, plant. Flue gas for pilot
testing was an isokinetic sampling representing 10% of the total flue
gas flow from a boiler having a steaming capacity of 80,000 Ib/hr and
burning 2% sulfur coal with 100% excess air.
In the pilot plant, it was found that a 1 molar sodium solution
would give reasonably good S02 absorption, while also regenerating
caustic and precipitating sulfate as gypsum. It was found that a
maximum concentration of 0.1 M hydroxide could be regenerated in a
1 M sodium solution. Increasing the sodium ion concentration above
1 M did not give appreciable increase in regenerated hydroxide
concentration. Lowering the sodium ion concentration, however, gave
a decrease in regenerated hydroxide concentration.
Optimum lime utilization was found to occur with high speed mixing,
using near stoichiometric quantities of lime. Eighty percent conversion
of lime was attained in 5 minutes.
Early in the pilot plant program, GM ran into the calcium scaling
problem; but later, using soda ash softening, they were able to allevi-
ate it. Reportedly, using soda ash for sodium make-up, they were able
to reduce the calcium content in the scrubber loop from 400 to 250 ppm.
The pilot plant was operated for a period of about 4 months in
Parma. Other pilot tests were conducted concurrent with the Parma
tests at another General Motors plant in Livonia, Michigan.8
Based on the results of testing in the pilot units, General Motors
designed and constructed a full scale system for S02 and particulate
control at the Parma plant.
a. System Description - The full scale double alkali system
installed at Parma is shown schematically (only one scrubber is shown)
in Figure 5. The boiler system consists of four stoker fired boilers,
two with steaming capacities of 100,000 Ib/hr and two with 60,000 Ib/hr,
totalling 320,000 Ib/hr or approximately equivalent to 32 Mw electric
generating capacity. All of the boilers are equipped with mechanical
dust collectors.
The double alkali system was designed to handle a steaming capacity
of 400,000 Ib/hr anticipating the addition of a fifth boiler in the
future.
492
-------
SODA ASH
vo
REACTOR
CLARIFIER
=2
FIGURES. GENERAL MOTORS' DOUBLE-ALKALI SYSTEM.
-------
The absorption system consists of four Koch valve tray scrubbers
containing three trays per scrubber. The trays have a type of floating
bubble cap which rises and falls adjusting to gas flow. The upper
portion of the scrubbers contain mesh mist eliminators. The scrubbers
are constructed of 316-L stainless steel.
The regeneration system handles the bleed flow of spent liquor
from the four scrubbers. It consists of two mix tanks or reaction
tanks (CFSTR's), two s laker-feeders (one on standby) and two reactor
clarifiers. One of two continuous drum filters makes the final solids
separation (the second is on standby). The filters are equipped with
spray nozzles to allow for cake wash.
The system operates similarly to the Envirotech dilute scheme,
with the primary exception that there is no augmental oxidation;
however, due to the fact that the boilers operate with high excess
air, high oxidation rates in the system are expected, and have been
experienced.
The primary functions of some of the major equipment are given below:
Mix Tank No . 1 - Spent scrubber solution containing
and NaHS03 is mixed with the slurry underflow from reactor
clarifier No. 2 which should contain NaOH, Na2S03 and CaCC>3. This
tank was designed to increase lime utilization by making use of the
CaC03 that results from the softening reaction in reactor clarifier
No. 2 (discussed later), to neutralize some of the bisulfite in the
spent scrubber solution. The extent of CaCC^ utilization, however,
is highly dependent on the scrubber bleed stream pH. If this pH is
not low enough, very little, if any, CaC03 will be utilized as shown
in the following reaction (previously shown under limestone regeneration)
CaC03 + 2NaHSC>3 + 1/2 H20 •*
CaS03'l/2 H20 I + CC>2 * + H20
Mix Tank No. 2 - The primary reactions between lime and sul-
fite, bisulfite and sulfate begin to occur in this tank, precipitating
gypsum and calcium sulfite. Effluent from mix tank No. 1 and filtrate
from the filter are mixed with lime in this tank.
Reactor Clarifier No. 1 - Slurry feed is received from Mix Tank
No. 2. Here the reaction between lime, sulfite, bisulfite and sulfate
is continued. The overflow consists of clear regenerated liquor, and
the underflow is a thickened slurry of waste solids which is fed to the
filter for final solids separation.
494
-------
Reactor Clarifier No. 2 - The overflow from reactor clarifier
No. 1 is softened via equation (18) or (21) with soda ash or CC>2,
respectively, in this unit before it is fed back to the scrubber
system.
b. Operating Experience - General conditions of operation and
performance data for the full-scale system at Parma are tabulated
below:
Flue Gas
Source: four coal stoker fired boilers
Fuel: coal, 2.3-2.6% sulfur, 2.5% sulfur average
Boiler excess air: 50% large boilers
120% small boilers
Oxygen content: 8.5-11.5%
Flow rate: 60,000 ACFM @ 350°F - Large boilers , , . o
50,000 ACFM @ 550°F - Small boilers esign
26,000-28,000 SCFM/scrubber, typical of
recent operation
(Normally, 2 of 4 boilers and scrubbers
in operation)
S02 inlet concentration: 900-1600 ppm (1200-1300 ppm
average)
Particulate inlet concentration: 0.3 gr/SCF
Absorber System
Absorber type: Koch valve tray scrubber, 3 trays
S02 removal: 88-98%
S02 exit concentration: 20-200 ppm
Particulate exit concentration: 0.05 gr/SCF (design value)
L/G ratio: 20 gal./lOOO CF
Scrubber pressure drop: 7-1/2 in. HoO
Scrubber liquor pH: inlet ^ 9.0
outlet 5.5-6.0
Regeneration System
Lime stoichiometry: 140-165% based on S02 removal
Regeneration liquor composition:
Scrubber Recycle From Regeneration
pH_ 5.5 - 6.0 >12
S05 0.35M
OH" Trace 0.1M
HS03 0.03M Trace
Ca^ 300-400 ppm 300-400 ppm
495
-------
Solids Separation
Solid Waste composition: Ca(OH>2 10-20% dry basis
Flyash 1-2% dry basis
Solubles (Na2SO^) 4-5% dry basis
CaSOx Remainder
Moisture 50%
Solid product wash: Crude wash, 20% reduction of
solubles
Reliability - The Parma full-scale system was started up in
March 1974. There were at least two scheduled shutdowns soon after
start-up for inspection. After the second shutdown a problem developed
which kept the system off line for about 3 weeks. As of early October
1974 the system has been in operation since May 23, without a total
system shutdown. Since the total system consists of four scrubbers
and four boilers, and the steam requirements have been such that usually
only two boilers and scrubbers have been on line at any one time, it
does not necessarily follow that each boiler/scrubber system was with-
out problems for the whole period between May and October 1974. In
fact, if and when a problem developed in one boiler/scrubber system,
that system was shut down and another started up. This has been
possible since the plant steam requirements have been well below the
system capacity.
Of the problems which have been experienced since start-up, only
one seems to be recurrent and is cause for individual scrubber shut-
down: the top tray in the scrubbers has a tendency to scale with
calcium carbonate. The other problems were relatively minor, and were
possibly a result of lack of operating experience. The scaling problem
is probably due to the fact that high pH liquor (pH above 12) from the
regeneration system enters the top tray of the scrubber directly,
causing local high pH regions on the top tray. The localized high pH
regions absorb C02 from the flue gas, precipitating CaC03 scale. A
solution for this problem is planned. Regenerated liquor will be mixed
with scrubber recycle liquor (pH 5.5-6.0) so that pH on the top tray is
controlled at a level low enough to prevent precipitation of CaC03
(equations (23) and (24)). At pH's belox^ about 9, CaC03 is solubilized
due to the carbonate/bicarbonate equilibrium:
CO^ + H+ J HCO~ (25)
A more complete discussion of the GM system and the operating problems
experienced to date is given by Dingo and Piasecki.^6
496
-------
c. Discussion - As of early October 1974, the GM system at Parma
has not been in steady state operation in the strict sense of the term.
Some improvement and optimization of the system, including solution of
the carbonate scaling problem, are still intended. The sulfate level
in the system presently at 0.35M concentration is lower than the design
level of 0.5M. Fluctuations in the sulfate level are indicative of poor
control of soda ash make-up to the system. The high lime stoichiometry,
140-165%, is indicative of poor lime utilization. Closer control of the
system should bring this lime utilization more in line with the lower
lime stoichiometries in use in the other double alkali systems.
General Motors has entered into an agreement with EPA which will
allow Arthur D. Little, Inc., as a contractor for EPA's Control Systems
Laboratory, to conduct a test program at the Parma plant to more fully
characterize the double alkali system. Through careful material balances
in the system, calculated from flow rate measurement and chemical ana-
lytical data, a number of important parameters will be studied. Among
these are included:
• SC>2 removal
• Process reliability (scaling phenomena)
• Sulfate control
• Waste characteristics
• Degree of closed loop operation
• Utilization of lime and other chemicals
• Oxidation rate
497
-------
Zurn Industries
Zurn has designed, constructed,, and started up a dilute double
alkali system for control of S02 from two coal fired industrial boilers
having a total steaming capacity (180,000 Ibs/hour) equivalent to about
20 Mw electric generating capacity. The system was started up in early
October 1974; consequently, no operating data can be reported at this
time. The boilers burn coal containing about 3.0% sulfur. Generally,
the system is patterned after the General Motors system in Parma. The
scrubber system consists of two Zurn "Dustraxtor" scrubbers, containing
24 and 28 tubes, respectively. Briefly described, this is a multitube
impingement/entrainment type scrubber capable of achieving very high
internal L/G ratios. The system employs two series lime reactors and
a sodium carbonate softening step. Provisions have been made for waste
product washing on the rotary vacuum filters, to minimize sodium losses
and associated potential secondary pollution problems.
A schematic of the Zurn double alkali system is shown in Figure 6.
498
-------
FRESH WATER
t
BOILER
NO. 2
VD
VD
CT
\
CaO
n
Na2C03
DRY CHEMICAL
TRANSFER
EXISTING
COLLECTOR
EXISTING
COLLECTOR
VACUUM
FILTER
SPARE
VACUUM
RECEIVER
SPARE
'VACUUM
PUMP
SPARE
VACUUM
PUMP
MOVABLE
CONVEYOR
•H.-'
*•:••*?>.
WASTE PRODUCT
(10) PICK-UP
CONTAINERS
FIGURE 6. ZURN DOUBLE ALKALI SYSTEM
-------
A. D. Little/Combustion Equipment: Associates
Pilot testing by ADL/CEA in a 2000 ACFM pilot plant in the ADL
complex in Cambridge, Massachusetts, in 'early 1973 led to the develop-
ment of a concentrated double alkali system. Based on the results of
that testing, CEA contracted with Southern Services, Inc., a subsidiary
of The Southern Company (a Southeastern Utility Combine) to install
a 20 Mw prototype test unit at the Scholz station of Gulf Power Company
near Chatahoochee, Florida.
In May 1973, EPA contracted with ADL to conduct a laboratory and
pilot plant study of various double alkali schemes of operation. About
a year later, when construction of the prototype system was well under-
way, the EPA contract with ADL was expanded to include prototype testing
of the ADL/CEA system at Gulf Power. A good description of the ADL/CEA
pilot plant and the Gulf Power 20 Mw prototype system is given in the paper
entitled, EPA-ADL Dual Alkali Program—Interim Results. Also contained
therein are operating conditions and test results obtained under the EPA
program to date. The system developed by ADL/CEA is among the various
schemes tested under the program.
500
-------
Kawasaki /Kureha_
Kureha Chemical Industry Co., Ltd., developed a concentrated lime-
stone double alkali process in a small pilot plant. Initial testing
led to the construction of a larger pilot plant (3000 SCFM) jointly
with Kawasaki Heavy Industries, Ltd., in July 1972. Pilot testing was
completed in March 1973. Since then, these two companies have jointly
developed a commercial double alkali process which is in operation at
the Shinsendai station of Tohoku Electric Power. This unit processes
one quarter of the flue gas from a 600 Mw oil-fired boiler (150 Mw
equivalent).
a. Sy_stem Description - A schematic diagram of the system installed
at Tohoku Electric is shown in Figure 7. The system consists of an S02
absorption section, a limestone regeneration section, a sulfate removal
section and a gypsum production section.
Absorption - Flue gas from the oil fired boiler passes
through an electrostatic precipitator before entering the system. The
gas is moved by a hot fan, forced draft with respect to the scrubber
system. There is a precooler section at the entrance to the scrubber
in which the hot incoming gas is adiabatically cooled with recycle
liquor. The scrubber is a grid-packed type consisting of two separate
stages of packing; gas flows upward through the packing. Scrubber
recycle liquor and effluent from the top stage are fed to the top of the
bottom stage. The bottom stage operates at an L/G ratio of about 12 gal./
1000 CF. Regenerated liquor, only, is fed to the top stage thus maximiz-
ing the pH in this stage and achieving good countercurrent mass transfer
resulting in high S02 removal efficiency. The total S02 removal in the
system is split, 90% in the bottom stage and 10% in the top stage (this
does not imply 100% removal of the incoming S02). Flue leaving the
absorber is reheated to 285°F by direct firing of 1.2% sulfur oil.
Regeneration - The regeneration reaction consists of neutral-
ization of NaHS03 with finely ground limestone (equation (6)). The lime-
stone is ground to 10 microns in a wet vertical tower mill. Scrubber
effluent liquor bleed taken from the bottom stage recycle stream is
filtered to remove oil soot before it is pumped to the regeneration
reactors to be reacted with limestone. The reactor system consists of
five closed reactors in series which are operated at about 90°C to
promote the reaction between limestone and NaHS03 in the scrubber bleed.
The solid product produced in the reactors is separated from the mother
liquor and washed in a vacuum filter. The liquor is recycled to the
scrubber system, and the solids, mostly CaS03-l/2 t^O are fed to the
oxidizer system and to the sulfate removal section.
501
-------
O
K)
REGENERATION
REACTOR
SYSTEM
FRESH
WATER
•- I
CENTRIFUGE
GYPSUM
AIR-
FIGURE 7. KUREHA-KAWASAKI DOUBLE ALKALI PROCESS AT
TOHOKU ELECTRIC'S SHINSENDAI STATION
-------
Sulfate Removal - This system is designed to remove sulfate
formed in the scrubber as gypsum via reaction with sulfuric acid and
calcium sulfite according to reaction (16). The process is fed with a
bleed stream of scrubber effluent,, a portion of the calcium sulfite
cake (washed) from the filter and sulfuric acid. In order to maximize
sulfuric acid efficiency, a small absorber and stripper are incorporated
in this sulfate treatment loop. The incoming scrubber bleed is acidified
in the absorber with S02. SO? is supplied by the stripper which converts
NallSOj to Na2S03, stripping off S02 via thermal decomposition according
to the following reaction:
i i P^ t~
2NaHS03 -> Na2SC>3 + S02 + H20 (26)
This reaction is the thermal regeneration reaction used in the Wellman-
Lord process which is now commercialized by Davy Powergas . Feed liquor
to the stripper is the acidic mother liquor from the sulfuric acid
reactor in which the sulfate removal occurs. Stripped liquor from the
stripper is recycled to the regeneration reactor .system.
Feed to the sulfuric acid reactor is thus already at a low pH, and
therefore requires less sulfuric acid addition to achieve a given
conversion of Na2S04 to gypsum.
Effluent slurry from the sulfuric acid reaction is centrifuged to
separate the gypsum produced from acidic mother liquor. The separated
solids are transferred to the oxidizer system together with the calcium
sulfite solids produced in the regeneration reaction. The centrif ugate,
as previously stated, is recycled to the stripper to allow recovery of
acid values.
Oxidation - The oxidizer developed by Kawasaki/Kureha is
considered to be proprietary. Oxidation air is supplied at 5 psig.
Repulped calcium sulfite slurry from regeneration and gypsum from the
sulfate removal step are pumped to the oxidizer. Sulfuric acid is added
to adjust pH in the reaction to attain the desired conditions for the
oxidation reaction. Calcium sulfite is oxidized to gypsum by reaction
with atmospheric oxygen according to the overall reaction:
CaS03'l/2H20 4- 1/202 + 3/2 H20 -> CaS04«2H20 (27)
The reaction mechanism may be acidic dissolution of CaSOo'l/2H20 followed
by oxidation of the bisulfite to sulfate regenerating the acid value to
dissolve additional CaS03-l/2H20. Gypsum product is separated from
oxidizer effluent and washed in centrifuges. Spent off-gas from the
oxidizer is recycled through the scrubber to remove S02 which was
emitted to the acidic oxidation reaction.
503
-------
b. Operating Experience - Typical conditions and performance data
for the Tohoku Electric FGD system are tabulated below:
Flue Gas
Source: oil fired boiler
Fuel: 1.2-1.5% sulfur oil
Oxygen content: 3%
Flow rate: 247,000 SCFM (420,000 Nm3/hr)
SC>2 inlet concentration: 600-800 ppm
Participate inlet: 0.0009-0.0044 gr/SCF (2-10 mg/m3)
Reheat: direct fire with 1.2% sulfur oil to 285°F
(140*C)
Absorber System
Absorber type: grid packed, 2 packing stages
S02 removal: 98%
S02 exit concentration: <10 ppm
L/G ratio: 12 gal./lOOO CF (1 kg/kg), bottom stage
Scrubber pressure drop: 7.9 in. 1^0 (200 mm H20)
Scrubber liquor pH: 6.9-7.6 inlet; to top stage
6.3-6.5 bottom stage recycle
Recycle liquor composition: 5-10%
5-8%
6-9% Na2S03
Oxidation rate: 5%
Regeneration
Limestone stoichiometry: 100% of S02 inlet
Sodium make-up rate: 0.018 moles Na/mole S02 absorbed
Reactor system residence time: 3 hrs (total, 5 reactors)
Reactor % solids: 5% (essentially all CaS03'1/2H20)
Reactor filter cake solids: 40%
Regenerated liquor: 6.9-7.6 pH
20% total dissolved solids
<10 ppm Ca
Limestone particle size: lOy
SojJ.ds Production. (Gypsum)
Water content: 5%
Sodium content: <300 ppm on dry basis
Particle size: 30x200p
504
-------
Reliability - The full-scale unit started up in January/
February 1974. Based on reports as of August 1974, there have been no
significant reliability problems. The only problem noted during a visit
in Japan by a representative of EPA was the plugging of the oil soot
filter.
Discussion - The system is said to be a closed loop system in that
there is no intentional water purge; however, based on the rate of
water addition to the system, and a comparison of the rate of sodium
make-up with the sodium content of the product, it appears that there
is some sodium loss from the system other than the amount in the
product gypsum. The overall system is fairly complex when compared
to some of the systems under development in the U.S. The reasons for
the additional complexity appear to be related to a) the system pro-
duces a relatively pure saleable product, b) it is designed to achieve
extremely high S02 and particulate removal, and c) limestone rather
than lime is used as the source of calcium. Apparently, there has
been considerable effort devoted to minimization of reagent usage;
i.e., sulfuric acid, limestone and sodium.
The gypsum product has been used in cement manufacture; however,
it also meets the more stringent specs for wallboard production.
The very low oxidation (5%) in the absorption system is attributed
by Kawasaki/Kureha to proprietary know-how.
c. Plans _for Future Testing and Full-Scale Application - Kawasaki/
Kureha plans to pilot-test an electrolytic process step which was pre-
viously piloted by Yuasa/Ionics in Japan. This process step is being
evaluated to replace the sulfuric acid reactor/absorber/stripper
operation for sulfate removal. The electrolytic sulfate removal scheme
is similar to that used in the Stone & Webster/Ionics process, equation
(18). Testing was scheduled for October 1974, in a 3-5 Mw pilot plant
consisting of two 5 ft^ cells capable of decomposing 10 Ibs/hour of
Na2SO^. Use of the electrolytic step is expected to reduce sodium
make-up requirement to the system by 50%, and is anticipated to reduce
capital costs of the overall system.
A number of additional large scale utility double alkali systems
are presently being engineered and/or constructed by Kawasaki/Kureha.
These are tabulated below:
Utility Capacity (Mw) Start-Up
Sikoku Electric Power 450 August 1975
Sikoku Electric Power 450 October 1975
Tohoku Electric Power 350 October 1976
Kyushu Electric Power 250 May 1977
Kyushu Electric Power 250 May 1977
505
-------
Showa Denko KK/Ebara
Showa Denko KK and Ebara Manufacturing Co., Ltd., of Japan jointly
developed a concentrated double alkali process using limestone as the
calcium source, and producing a saleable gypsum by-product. Development
began in a 5,900 SCFM pilot plant at the Kawasaki plant of Showa Denko
in 1971. This has culminated in the construction and operation of a
150 Mw system on an oil fired boiler at Showa Denko's plant in Chiba.
a. System Description - Figure 8 is a schematic representation of
the Showa Denko/Ebara process. Basically it is similar to the Kawasaki/
Kureha system. The full scale unit (150 Mw) uses four inverted "vertical
cone" scrubbers (similar to the Zurn or Bahco type). Liquor is entrained
at the bottom of a conical draft tube (truncated apex) by the entering
flue gas and the mixture passes up the tube. At the top of the tube
(conical base) there is a disengaging section where liquor separates and
falls back to the bottom of the scrubber.
In the reactor system, limestone is reacted with spent scrubber
effluent to neutralize the NaHS03 producing Na2S03 for recycle to the
scrubber and CaS03-1/2^0 (equation (16) which is sent to an air oxidi-
zer to be converted to gypsum by air oxidation (equation (28)). A slip
stream of spent scrubber liquor is taken to a sulfate treatment section
which converts Na2S04 to gypsum and NaHS03 by reacting it with sulfuric
acid and calcium sulfite according to reaction (16). Gypsum from this
reaction containing some unconverted calcium sulfite is combined with
the calcium sulfite produced in the limestone neutralization reaction
and pumped to the air oxidizer system. Sulfuric acid is added to ad-
just the pH in the oxidizer system to achieve optimum conditions for
the calcium sulfite oxidation.
The system operates in an open loop since the amount of water
required to wash the gypsum product to below 300 ppm sodium content is
greater than the system water losses through evaporation, water of
hydration, and moisture in the final product. The sodium content
specification for wallboard manufacture is less than 300 ppm. The
corresponding specifications for gypsum used in cement manufacture is
900 ppm.
b- Operating Experience - General conditions and performance data
for the full scale system are tabulated below:
Flue Gas
Source: oil fired 150 Mw boiler
Fuel: 2.5-3.0% sulfur
Particulate inlet concentration: 0.8 gr/scf
S02 inlet concentration: 1200-1500 ppm
Reheat: direct fired with 1% sulfur oil to 285-300°F
506
-------
WASTE
LIMESTONE
FIGURE 8. SHOWA DENKO SODIUM-LIMESTONE PROCESS
-------
Absorber System
Absorber type: 4 vertical cone scrubbers
S02 exit concentration: 60-100 ppm after reheat
with 1% sulfur oil
Particulate exit concentration: 0.025-0.035 gr/SCF
L/G ratio: 7.5 gal./lOOO CF
Scrubber liquor pH: 6.3 exit
Oxidation rate: <10%
Regeneration Chemical Usage
Limestone utilization: 96%
Sulfuric acid usage: 0.2 moles H2S04/mole SO? absorbed
Sodium usage: 0.10-0.12 moles of NaOH/mole SC>2 absorbed
Oxidizer
Suspended solids content: 15%
Gypsum By-Product
Na content: <300 ppm
Water content: 7-8%
Reliability - The full scale system was started up in
June 1973. It was shut down for a period in March 1974 for inspection.
As of March 1974, the scrubbers were said to have been near 100%
reliable, and they were found to be clean in the March inspection.
The regeneration system had given some trouble but since there is
adequate solution surge capacity, regeneration difficulties have not
affected boiler reliability. The trouble has been mainly pipe plugging
which was not considered to be chemical scaling.
508
-------
SUMMARY AND CONCLUSIONS
Based on the initial performance and reliability demonstrated in
various double alkali pilot plants, and in an increasing number of full
scale applications in the U.S. and Japan, it appears that this tech-
nology will soon become a viable alternative throwaway process to the
lime/limes tone processes which are now the prevalent FGD processes used
with fossil fueled utility boilers. The total operating and planned
full-scale and prototype utility and industrial applications of double
alkali technology now number 17 with individual applications with total
gas flow control equivalent to 2375 Mw of electric generating capacity.
Summaries of significant pilot testing, and operating and planned full
scale systems are given in Tables 4 and 5, respectively.
Performance of double alkali systems with respect to S02 removal
is well established. Over 99% S02 removal has been achieved with lower
than 10 ppm S02 concentration in the treated flue gas.
Potential problems with waste disposal from these systems are
present due to the presence of a certain amount of soluble sodium salts
which could cause water pollution problems in the surface and ground
water in the vicinity of disposal sites. A number of techniques to
reduce the soluble sodium salts in the waste product have been tested;
however, it appears that there will inevitably be a higher level of
soluble salts in waste product from double alkali systems than from
lime/limestone systems. With present U.S. technology it appears that
the minimum incremental amount of solubles in a double alkali waste
product (over that typically present in lime/limestone waste product)
is about 1-2% on a dry solids basis. Development of sludge fixation
technology to change the wet solid waste product to a hard unleachable
solid could conceivably reduce this potential problem.
Actual costs for full scale utility boiler applications are not
available. Estimates for these applications based on the best available
information put double alkali system capital costs in the range of lime/
limestone system costs; however, these types of estimates are necessarily
subject to qualification.
In the U.S., development has stressed the use of lime rather than
limestone as the calcium source in all of the full scale industrial
applications and in most of the pilot plant testing in both dilute and
concentrated systems. In Japan, limestone rather than lime is used for
concentrated systems in full scale utility boiler applications apparently
for the operating cost benefit which is derived through the use of the
less expensive regenerant. Also, due to differences in the economies
of the two countries and state-of-the-art of technology, production of
by-product gypsum from the Japanese double alkali systems is prevalent;
509
-------
Table 4. SUMMARY OF SIGNIFICANT PILOT PLANT DOUBLE ALKALI TESTING
System Developer
Envirotech
FMC
General Motors
A- D. Little/
Combustion Equipment
Associates/EPA
Kawasaki/Kureha
Showa Denko KK
Pilot Plant Location
Utah Power & Light Company
electric power boiler,
coal fired
At least 7 different locations,
coal and oil fired industrial
boilers, elec. power boilers,
acid plant
GM Parma Plant
GM Livonia Plant
coal fired industrial boiler
ADL Facility
Cambridge, Massachusetts
Natural gas furnace
Oil fired boiler, Japan
Oil fired boiler, Japan
Size
(Gas rate,CFM)
2500
3500
2800
2000
3000
5900
Active Alkali
Dilute,
Cone.
Cone.
Dilute
Dilute,
Cone.
Cone.
Cone.
Calcium Source
Lime
Lime
Lime
Lime,
Limestone
Limestone
Limestone
-------
Table 5. SUMMARY OF OPERATING AND PLANNED FULL SCALE DOUBLE ALKALI SYSTEMS
System Operator
FMC
Modesto Calif. Plant
Showa Denko KK
Kawasaki Plant, Japan
Tohoku Electric
Shinsendai Sta.,
Japan
General Motors
Parma, Ohio Plant
(Unspecified)
U.S.
Gulf Power Company
Scholz Plant
(Southern Services)
Firestone
Potts town, Pa.
(Unspecified)
U.S.
(Unspecified)
U.S.
(Unspecified)
Japanese Paper Plant
System Application
Reduction kilns
Oil-fired elec.
power boiler
Oil-fired
utility boiler
4 coal fired
industrial boilers
2 coal fired
industrial
boilers
Vendor or
Developer
FMC
Showa
Denko /Ebara
Kawasaki/
Kureha
General
Motors
Zurn
Industries
(Mw,
10
150
150
40
32
20
Size
Equivalent)
(Gas Rate)
(Regen)
(Gas Rate)
Active
Alkali
Cone .
Cone .
Cone .
Dilute
Dilute
Calcium
Sources
Lime
Limestone
Limestone
Lime
Lime
Start-Up
Date3
Dec. 1971
June 1973
Jan. 1974
Mar. 1974
Oct. 1974
Coal fired
utility boiler-
prototype
Coal & oil fired
industrial boiler-
demonstration
Coal fired
industrial heating
plant
Coal fired
industrial boiler
system
Oil fired
industrial boiler
A.D.Little/ 20
Combust. Equip.
Associates
FMC 3
FMC 14
FMC 50
FMC/Ataka
Cone.
Cone.
Cone .
Cone.
Cone.
Lime
Lime
Lime
Lime
Lime
(Nov. 1974)
(Dec. 1974)
(Dec. 1974)
(early 1975)
(early 1975)
-------
Table 5 (Continued). SUMMARY OF OPERATING AND PLANNED FULL SCALE DOUBLE ALKALI SYSTEMS
System Operator
FMC
Green River, Wyo.
Plant
Sikoku Electric Power,
Japan
Tohoku Electric,
Japan
(Unspecified)
U.S.
Kyushu Electric
Power, Japan
System Application
Coal fired
electric power
boiler
2 Oil fired
utility boilers
Oil fired
utility boiler
Oil fired
industrial boiler
2 Oil fired
utility boilers
Vendor or
Developer
FMC
Kawasaki/
Kureha
Kawasaki/
Kureha
FMC
Kawasaki/
Kureha
Size Active
(Mw, Equivalent) Alkali
150 Cone.
900 (2-450 Mw) Cone.
350 Cone.
13 Cone.
500 (2-250 Mw) Cone.
Calcium
Sources
Lime
Limestone
Limestone
Lime
Limestone
Start-Up
Date3
(mid 1975)
(Aug. -Oct.
1975)
(Oct. 1976)
(Late 1976)
(May 1977)
Dates in parentheses are projected start-up dates
-------
whereas, production of a throwaway solid waste product is the general
rule with development of the technology in the U.S.
The development of double alkali technology has obviously stemmed
(and benefitted) from both lime/limestone and other soluble sodium
scrubbing technology development. Possibly as a result of this and
certain inherent advantages of soluble alkali scrubbing, it appears
that the reliability established at this point in the development of
double alkali technology is greater than that which had been estab-
lished for the lime/limestone, systems at a corresponding stage of
development.
513
-------
REFERENCES
1. Ponder, W. H., "Status of Flue Gas Desulfurization Technology For
Power Plant Pollution Control." Presented at Thermal Power
Conference, Washington State University, Pullman, Washington,
October 4, 1974.
2. Epstein, M., R. Borgwardt, et al., "Preliminary Report of Test
Results from the EPA Alkali Scrubbing Test Facilities at the
TVA Shawnee Power Plant and at Research Triangle Park."
Presented at Public Briefing, Research Triangle Park, North
Carolina, December 19, 1973.
3. La-Mantia, C., et al. , "EPA-ADL Dual Alkali Program Interim
Results." Presented at EPA Symposium on Flue Gas Desulfurization,
Atlanta, Georgia, November 4-7, 1974.
4. Johnstone, H. F., et al., "Recovery of Sulfur Dioxide from Waste
Gases." Ind. & Eng. Chem., Vol. 30, No. 1, January 1938,
pp 101-109.
5. Phillips, R. J., "Operating Experiences with a Commercial Dual-
Alkali S02 Removal System." Presented at the 67th Annual Meeting
of the Air Pollution Control Association, Denver, Colorado,
June 9-13, 1974.
6. Cornell, C. F. and D. A. Dahlstrom, "Performance Results on a
2500 ACFM Double Alkali Plant for S02 Removal." Presented at
the 66th Annual Meeting of A.I.Ch.E., Philadelphia, Pennsylvania,
November 11-15, 1973. Condensed version of the paper appeared
in December 1973 CEP.
7. Kaplan, N., "An EPA Overview of Sodium-Based Double Alkali
Processes - Part II Status of Technology and Description of
Attractive Schemes." Presented at the EPA Flue Gas Desulfurization
Symposium, New Orleans, Louisiana, May 14-17, 1973.
8. Phillips, R. J., "Sulfur Dioxide Emission Control for Industrial
Power Plants." Presented at the Second International Lime/Limestone
Wet-Scrubbing Symposium, New Orleans, Louisiana, November 8-12, 1971,
9. Brady, J. D., "Sulfur Dioxide Removal Using Soluble Sulfites."
Presented at Rocky Mountain States Section Air Pollution Control
Association, Colorado Springs, Colorado, April 30, 1974.
514
-------
10. Cornell, C. F., "Liquid-Solids Separation in Air Pollutant Removal
Systems." Presented at the ASCE Annual and National Environmental
Engineering Convention, Kansas City, Missouri, October 21-25, 1974,
11. Selmeczi, J. G. and R. G. Knight, "Properties of Power Plant Waste
Sludges." Presented at the Third International Ash Utilization
Symposium, Pittsburgh, Pennsylvania, March 13-14, 1973.
12. Ellison, W., et al., "System Reliability and Environmental Impact
of S02 Scrubbing Processes." Presented at Coal and The Environ-
ment, Technical Conference, Louisville, Kentucky, October 22-24,
1974.
13. Rochelle, G. T., "Economics of Flue Gas Desulfurization."
Presented at EPA Flue Gas Desulfurization Symposium, New Orleans,
Louisiana, May 14-17, 1973.
14. "Sulfur Dioxide and Flyash Control", FMC Corporation Technical
Bulletin. FMC Corporation, Air Pollution Control Operation,
751 Roosevelt Road, Suite 305, Glen Ellyn, Illinois 60137.
15. McGlamery, G. G. and R. L. Torstrick, "Cost Comparisons of Flue
Gas Desulfurization Systems." Presented at the EPA Symposium on
Flue Gas Desulfurization, Atlanta, Georgia, November 4-7, 1974.
16. Dingo, T. and E. Piasecki, "General Motor's Operating Experience
with a Full-Scale Double Alkali Process." Presented at the EPA
Symposium on Flue Gas Desulfurization, Atlanta, Georgia,
November 4-7, 1974.
515
-------
Initial Operating Experiences
With A Dual-Alkali SO2
Removal System
Parti
PROCESS PERFORMANCE WITH A COMMERCIAL
DUAL-ALKALI S02 REMOVAL SYSTEM
Thomas T. Dingo/General Motors Corporation
Part I I
EQUIPMENT PERFORMANCE WITH A COMMERCIAL
DUAL-ALKALI S02 REMOVAL SYSTEM
Edmund J. Piasecki/General Motors Corporation
for presentation at
ENVIRONMENTAL PROTECTION AGENCY
SYMPOSIUM ON FLUE GAS
DESULFURIZATION
ATLANTA, GEORGIA
NOVEMBER 4-7, 1974
517
-------
ABSTRACT
The first commercial demonstration of dual-
alkali wet scrubbing went on-line in March,
1974, at the General Motors Chevrolet plant
in Parma, Ohio. The S02 system was installed
at a cost of $3.5 million and was designed to
handle a steaming capacity of 400,000 pounds
of steam per hour. In this system, a dilute
sodium hydroxide solution absorbs S02/particulate
matter from the flue gas. The spent scrubbing
solution is then recausticized with lime and
reused in the scrubbers while precipitated
sulfur compounds are dewatered and sent to
landfill. The system is currently under-
going a shakedown and evaluation period.
This paper is an update of operating ex-
periences. Equipment performance, SC>2 re-
moval capabilities, process reliability and
control, sulfate formation, chemical utili-
zation, calcium ion control and waste cake
characteristics are reviewed.
518
-------
PART I
PROCESS PERFORMANCE WITH A
COMMERCIAL DUAL-ALKALI S02 REMOVAL SYSTEM
Thomas T. Dingo
General Motors Corporation
Manufacturing Development
Warren, Michigan
For Presentation at Environmental Protection
Agency Symposium on Flue Gas Desulfurization
Atlanta, Georgia
November 4-7, 1974
519
-------
PROCESS PERFORMANCE WITH A
COMMERCIAL DUAL-ALKALI SO?. REMOVAL SYSTEM
I.. INTRODUCTION
General Motors, like everyone else, has been caught in the energy
squeeze. Natural gas and fuel oil are not readily available as
in the past. Thus, because of the long term availability of coal
over oil or natural gas, the Corporation has made every effort
to use this fuel wherever economically possible within existing
pollution codes.
GM is already a large industrial user of coal in the generation of
steam for process uses. In 1973, the Corporation consumed close
to 2.0 million tons of coal. Strict air quality standards,
however, in many urban areas have made the use of coal very
difficult. In some cities, even switching to a low sulfur coal
would not result in meeting air quality requirements. These
situations have forced us to look to yet unproven sulfur dioxide
emission control technology.
In 1968, in anticipation of more stringent pollution codes coupled
with shortages and drastically high cost for low sulfur fuels, the
Corporation initiated a program to determine if control technology
could be developed for its industrial powerhouses. The criteria
in developing such technology were as follows:
1. The system must afford high process and equipment reliability.
2. It must be simple and not dictate the operation of the power-
house .
3. It must not be a chemical plant and must produce a by-product
that can be easily disposed.
4. It must be economically competitive with other alternatives.
Scrubbing with a soluble salt coupled with regeneration and reuse
of the scrubber effluent appeared to satisfy our criteria best.
Pilot work was started in 1969 and the system developed has come
to be known as the GM Double Alkali SO,, Control System. The first
commercial demonstration of this system is the $3.5 million
installation at GM's Chevrolet-Parma plant near Cleveland, Ohio,
Figure 1. The system was placed on line in March, 1974 and has
been going through shakedown and adjustment.
520
-------
It is the purpose of this paper to describe the process chemistry
and equipment and to relate the performance characteristics of
the system to date.
II. PROCESS DESCRIPTION
The General Motors dual-aklaki process involves the use of a
dilute alkaline stream of sodium hydroxide (NaOH) to remove sulfur
oxides from the boiler gases and alkaline lime (CaO) to regenerate
the spent scrubbing liquor by precipitating the absorbed sulfur
to reform an hydroxide solution. It is the purpose of this
section to describe the process chemistry through the absorbing,
regeneration, softening and neutralization sections of our system.
Figure 2 shows a process schematic.
SCRUBBER CHEMISTRY
The reactions that take place in the absorbing section of our
system are straight forward and are shown below:
2NaOH + S0 *-NaSC> + HO (1)
(2)
(3)
Sodium hydroxide reacts with sulfur dioxide to form sodium sulfite
(Reaction 1). The sodium sulfite, then can react further with
sulfur dioxide and water to form sodium bi-sulfite (Reaction 2).
In addition, some of the active sulfite is oxidized to sodium
sulfate due to the presence of excess oxygen (Reaction 3).
REGENERATION CHEMISTRY
Because of the formation of Na~SO., the regeneration section has
to be capable of regenerating ootn sulfate and sulfite. The
sulfate reaction, however, is quite difficult because of the
relative solubility of the product, calcium sulfate. In addition,
the_sodium sulfate cannpt be causticized in the presence of high
S0^~ or OH~ because Ca level are held below CaSO solubility
product. Thus, to provide effective sulfate regeneration, the
system must be operated at a dilute OH concentration (< 0.14M)
while maintaining sufficient levels of sulfate (> 0.4M) to affect
calcium sulfate precipitation. In actual practice, we try to_
maintain our feed solution at 0.1 Molar OH and 0.5 Molar SO.~.
The chemical reaction that take place in the regeneration sections
are:
521
-------
NaHS03 + Ca(OH)2 — *-NaOH + CaS03r + H20 (4)
Na S03 + Ca(OH) — *- 2NaOH + CaSC^r (5)
(6)
SOFTENING CHEMISTRY
Because of the difficulty in regenerating sodium sulfate (Reaction
6), an excess of lime must be added. This causes dissolved calcium
ion levels near SOOppm, which, if allowed to enter the scrubber,
could precipitate as calcium salts and cause subsequent scaling.
To alleviate this problem, a softening step was added to our
flowsheet. In this step, soda ash and carbon dioxide are added
to precipitate the dissolved calcium as a carbonate. The reactions
are:
Ca(OH)2 — *- 2NaOH + CaCO. (7)
C02 + Ca(OH)2— *- CaC03- + H20 (8)
The resultant calcium levels now leaving the softening clarifier
can be as low as SOppm, but are controlled to approximately 300ppm.
A double media softening is used to maintain a chemical balance
on the system. The main function of the soda ash is to replenish
the sodium losses from the vacuum filter. Because of this, the
soda ash make-up is not sufficient to lower the dissolved calcium
to the desired level. The carbon dioxide, then, supplies the
additional softening required. As shown in Reaction 8, however,
the carbon dioxide lowers the free hydroxide available for scrubbing,
From an economic and chemical utilization standpoint, the system
should be operated with the minimum softening to avoid scrubber
scaling.
NEUTRALIZATION CHEMISTRY
If the calcium carbonate formed through softening is not reused
in the system, the lime utilization suffers drastically. In a
sense, calcium would be added in one section and thrown out in the
next. The only place where the calcium carbonate could be used
back in the system is with the spent scrubbing liquor. The sodium
bi-sulfite would react with the carbonate as shown in the following
reaction.-
T +
CaC03 + 2NaHS03 — *- CaS03T + Na2S03 + H2C03 (9)
522
-------
Thus, a reaction tank to pretreat the scrubber effluent before lime
addition was added to the system. Complete use of the calcium
carbonate is dependent on the oxidation rate, low sulfate formation,
in the scrubbers.
Ill. EQUIPMENT DESCRIPTION
The Chevrolet-Parma Plant was chosen as a demonstration site
for our process because it is located near Cleveland where a
rigid S0? emission limitation exists. This code requires us
to burn an equivalent of 0.625% sulfur coal. The plant consumes
approximately 55,000 tons/year of 2.0 to 3.0% + sulfur coal. The
total steam generating capacity of the facility is 320,000 Ibs/hr.
from four spreader stoker fired boilers. In anticipation of an
additional boiler, however, the sulfur dioxide control system
was designed to handle a steam capacity of 400,000 Ibs/hr. The
equipment used in our double-alkali system is described below.
ABSORBERS
Four scrubbers are utilized in our system, one for each boiler.
They are approximately 12 ft. in dia. and 20 ft. high, Figure 3.
In these units, the boiler gases enter through a prequench-
section at the bottom of the scrubber and flow upward through
three absorption trays to exhaust. Scrubber feed and recycle
are added at the top of the scrubber, flowing counter-current to
the gases down to a recycle tank. Each of the absorption trays
contains a series of movable bubble caps that adjust to changes
in gas flows. Each unit is designed to give us a 3:1 turn down
capacity. This type of absorber gives us good S02 removal at a
relatively low recycle pH and good particulate removal. The
particulate loading from our units must be near 0.05 grain per scf.
The scrubbers are constructed of 316-L stainless steel and are
designed to operate at a total system pressure drop of 7.5 in.
H,,O. They are also rated for a maximum liquid to gas ratio of
20:1 of which about 20% of the liquid is scrubber feed.
REACTORS
The bleed from the recycle tanks, Figure 4, is pumped through
two back mix reactors in series. They are carbon steel in
construction and, are designed with a 6000 gallon capacity to
give a 5 minute retention time at full flow. In the first reactor,
a CaCO. slurry from the softener clarifier is added for neutrali-
zation. In the next reactor, lime slurry is added for recausti-
cization. Both reactors are mounted on the side of the scrubber
building as shown in Figure 5.
523
-------
CLARIFIERS
From the reactors, the regenerated solution, laden with fly ash
and calcium precipitates, flows by gravity to two 300,000 gal
reactor clarifiers in series, Figure 6. Each clarifier is 60 ft.
in dia. and has a maximum 0.46 gpm per square foot rise rate.
In the primary clarifier, the solution goes through additional
reaction, if needed, and solid separation. In the second
clarifier, the solution is softened with Na CO and CO . A
regenerated, clarified, and softened solution then overflow back
to the scrubber building.
VACUUM FILTERS
The underflow from the primary clarifier (approximately a 15%
slurry) is pumped to a sludge holding tank for batch vacuum
filtering. Two 500 square foot filters are available, each with
a maximum capacity of 50 tons of cake per day, Figure 7. The
cake from the filters is dumped into trucks for disposal as a
landfill material. The filtrate is returned to the system at
the clarifiers.
CONTROL PANEL
To assure proper operation of the system, an elaborate panel was
designed to simplify the control and monitoring functions, Figure 8.
From here, the operator has positive indication of what is happening
in the scrubbers, on the floor, and outside of the building and
can change chemical addition rates as required.
The control theory of the double alkali system is relatively
simple. First, caustic is fed to the scrubbers on the basis of
the boiler steam load and trimmed by the pH in the recycle line.
Steaming rate is a good indicator of coal usage and subsequent
pounds of sulfur dioxide being formed. Since a certain pH range
is desirable for proper SO,, removal, pH control was also combined.
Also, since the scrubber feed is proportional to the caustic
being supplied to the scrubbers, this signal was selected as a
basis for control of the chemical feeds of lime, soda ash, and
carbon dioxide. Physical titrations for caustic concentration
leaving the recausticizer and calcium concentrations across the
softening clarifier are periodically made. Chemical feed rates
are adjusted if necessary.
IV. PERFORMANCE CHARACTERISTICS
The dual alkali system at Chevrolet-Parma was officially placed
on line in early March, 1974. Since that time, it has been going
524
-------
through a checkout and shakedown period to determine the actual
performance characteristics of the system. During this period,
many interrelated and unrelated mechanical and chemical difficulties
were encountered. Some of these problems did cause complete shut-
downs, but most were overcome through minor process and operational
modifications while on stream. It is the purpose of this section
to describe the performance of various parts of the system.
jGRUBBER PERFORMANCE
The heart of the system, of course, is the absorbers for removing
the sulfur dioxide. The chemical reactions that take place in
these scrubbers, however, will greatly affect the operation and
economics of the regeneration section of our system. Thus it is
not just a matter of scrubbing out SO-, but doing it with a con-
trolled scrubber chemistry.
Sulfur dioxide tests across the scrubbers have shown removal
efficiencies from 85 to 95% over the range of scrubber outlet
pH of 5.5 to 6.5. Particulate loadings leaving the scrubbers are
still to be determined. Testing is being continued to determine
tray efficiencies for possible optimization.
As stated previously, scrubber chemistry will greatly affect the
regeneration section of our system. Specifically, if the oxidation
rate {Reaction 3) in the scrubber is too high, not enough bi-sulfite
(Reaction 2} will be formed to take care of the calcium carbonate
(Reaction 9) in our neutralizer reactor. This means that the lime
utilization of the system would be very poor. *For complete utili-
zation of the lime, the maximum oxidation rate must, be near 10%.
Actual tests, however, show that our oxidation rates fluctuate
between 30 and 80% with the recycle pH and steaming rate. The
exact effect of these variables on the oxidation rate will be
determined in hopes of controlling a lower rate.
In addition, oxidation rate will greatly affect the scaling ten-
dencies in our scrubbers. Again, a sufficient amount of bi-sulfite
must be present in the recycle liquor to partially neutralize the
incoming scrubber feed over the top tray. if the pH on the top
tray is not kept near 9.0, soluble calcium can react with the
CO- in the flue gases and form a carbonate scale, causing high
pressure drop and poor liquid-gas contact. Initially, this problem
was very acute and necessitated frequent shutdowns for manually
removing the calcium carbonate scale. Since then, however,
revisions have been made to alleviate this problem. First, a
*0xidation _ lnno/ _ Total normality of HSO0 + SO-, in recycle , n
Rate ~ UU/° Normality of OH in scrdbber reed x luu/°
525
-------
piping revision was made to assure better mixing of the acidic
recycle liquor with the incoming scrubber feed on the top tray.
Second, the recycle liquor is now controlled at a 6.0 pH instead
of 6.5 pH to assure more bi-sulfite. Finally, the soluble calcium
levels in the feed stream are maintained around 300 ppm through
softening. The combination of these changes has now allowed us
to run over a month on a scrubber without any appreciable increase
in pressure drop. It is still too early to tell whether we have
eliminated the scaling altogether or are just retarding it.
REGENERATION PERFORMANCE
The main function of the regeneration section is to produce a
0.1M OH solution. No problems have been encountered in doing
this, but, because of the inability to reuse the CaCO-, in the
neutralization reactor, the overall lime utilization nas been
very poor. The lime/sulfur stoichiometry has been in the range
of 1.4 to 1.5/1.0. Original design expectation were 1.1/1.0.
The problem in using the calcium carbonate have been both mech-
anical and chemical. As explained previously, the high oxidation
in the scrubbers rate causes a deficiency of bi-sulfite in the
scrubber bleed liquor. This deficiency then requires that only
a small volume of carbonate slurry be metered to the neutralizer
tank. This necessitates only cracking the control valve for the
low flows. The slurry then has a tendency to plug the control
valve. Thus we have been forced to over-feed the carbonate slurry,
Also working against the neutralization is the fact that we
utilize a back mix reactor. In essence, we are trying to react
an acidic stream with a basic stream in an already highly basic
media. A plug-flow reactor arrangement is being investigated.
Once these problems are overcome, our lime to sulfur stoichi-
ometry should improve to be 1.3:1.
PRIMARY CLARIFIER PERFORMANCE
From the regeneration section, the recausticized solution flows
to a primary clarifier for solid-liquid separation. If not
operated correctly, we found that this unit could cause major
problems and shutdowns.
The first problem was encountered early in the debugging period.
The total system was being operated on a Monday through Friday
basis with inspection scheduled for the weekends. At that time,
it was standard practice to shut off all pumps and completely
stop the solution flow. During one of these periods, the slurry
pumps at the bottom of the clarifier broke down and we could not
recirculate or draw off sludge. As a result of the complete
quiescence, the sludge blanket in the clarifier, approximately
526
-------
60 tons, settled rapidly to the clarifier bottom and jammed
the clarifier rake. It became necessary to completely drain the
clarifier, manually remove most of the sludge and manually free
the rake. One week was lost because of this. Now it is a
standard practice to always keep an upflow through the clarifiers
to keep the solids in suspension.
The second major problem occurred when we progressed from
scrubbing one boiler to three boilers. After operating about a
week on three boilers, solid carry over became evident, upset
the softening clarifier, and got into the scrubbers causing
scaling and plugging. At that time, it was not known whether this
carry over was due to a change in scrubber chemistry or an
inability to remove solids on a regular basis.
Originally, we thought that additional scrubbers that we placed
on line had low oxidation rates and their spent solution was
forming calcium sulfite (Reaction 5) when regenerated. The
CaSO» 1/2H-0 is much more difficult to settle and filter than
the CaS04» 2H_0. If allowed to build up then, the calcium sulfite
could form an hindered settling zone in the clarifiers. The
oxidation rates, however, in all the scrubbers were found not to
vary significantly.
Also, at the same time we placed the system completely on line,
we encountered some scheduling problems with the solid waste
hauler of our sludge. Since trucks were not available, we could
not draw off enough sludge to keep up with its formation. As
a rule of thumb, five tons of wet sludge should be removed for
every ton of lime added to the system. Consequently, we started
to build up a solid inventory in the clarifiers (400 tons), lost
the sludge blanket, forced an hindered settling zone in the
clarifiers, and pushed the solids over the top. A month was lost
while cleaning the solids from the total system.
After realizing this, several operational changes have been made
to assure that this situation does not develop again. First,
it is imperative that we be able to remove solids at a greater
rate than they are formed. Once the sludge blanket in the
clarifier is formed, it must be maintained at the desired level.
We now have the capability to remove sludge as required through
additional trucks and manpower. Second, it is important that
the sludge blanket in the clarifier be periodically checked.
The operators now regularly check the sample ports on the sides
of the clarifier to determine the solids concentration throughout
the tank. The sludge withdrawal rate is then increased or
decreased to keep the solid blanket below the clarifier cone.
Finally, to improve clarity of the clarifier overflow, a coagulant
527
-------
is being added in the contact zone. The optimum concentration
of this additive is yet to be determined. Currently, we use
about a 0.3 ppm concentration.
SOFTENER PERFORMANCE
The dissolved calcium levels leaving the primary clarifier are
around 700 ppm. We have found that the calcium ion reduction
in the softening clarifier can be controlled to any desired
level through either the addition of soda ash or carbon dioxide
(Reactions 7 & 8). Currently, we are controlling the effluent
from the softener at about 300 ppm without noticeable scaling
in the scrubbers. Material balances across the softener
indicate almost 100% chemical utilization of Na0C00 or CO~ .
Z J f.
Initially, we experienced some problems in settling the precipi-
tated carbonate. We have found it necessary to maintain a
sufficient concentration of precipitates (5% slurry) in the
reaction zone of the clarifier for effective seeding and par-
ticle growth. Also, we have started adding a coagulant to the
reactor zone to improve the solution clarity at the overflow.
Again, the optimum concentration is being determined. One
phenomena that we still must face is a build-up of carbonate
salts on the clarifier surfaces. As of yet, it is not severe
enough to force manual cleaning.
SOLIDS HANDLING PERFORMANCE
Initially, considerable difficulties were encountered in the
solids handling section of our system. Minor operational and
process changes have proven successful. First, we had to
determine the best percent solids at the discharge of the sludge
pumps for our system. This is essential for proper solids handling
because: (A) too thick a slurry can cause^plugging of the pumps
and subsequent sludge build-up in the clarifiers, and (B) too
thin a slurry can cause poor performance of the vacuum filters.
A slurry concentration of around 15% solids has been found to work
very well. Second, on the vacuum filters themselves, we have
switched from a 100% olefin to a 100% nylon cloth. The release
characteristics of the cake have improved greatly. Last, we had
to adjust the speed and vacuum on the filters to achieve a
consistent cake.
Since we have been handling sludge on a more continuous basis,
most of the vacuum filtering problems have diminished. Except
for start-up and some random occurrences, the filter can run
unattended and as designed. On an average, we have been dis-
charging a 53% solids cake with or without cake washing. The percent
528
-------
soluble in the wet cake without cake washing is about 5%. The
effect of cake washing on soluble salts is being determined
currently. Also, the optional water volume and arrangement for
the wash sprays have to be determined to measure its effects on
the system's water balance. Fortunately, contrary to original
projections, we should not have to add a repuddler to aid in
filter cake washing.
CONTROLS PERFORMANCE
In general, the control system has been able to function properly
and to provide smooth responses to changes in boiler loads.
Relatively constant scrubber bleed pH and good SO removal
efficiencies are easily maintained. Some revisions, however,
had to be made in the control theory so that a steady operation
could be assured. These changes were in the areas of lime
feeding, clarifier monitoring, and pH sensing.
Originally, lime feed was somewhat difficult to automatically
control because the magnetic flow meter used to monitor the slurry
flow is susceptible to scaling. We were never really sure what
the flow of the slurry actually was and, consequently, started
to notice wild fluctuations in the alkalinity at the causticizer
reactor. To circumvent this situation, we are now manually
controlling the rate of dry lime addition to the system and set
the slaking rate to maintain a constant level in our lime slurry
tanks. Presently, we maintain a constant volumetric flow rate
to the causticizer, but the percent lime in the slurry varies
as we change the lime addition rate. A rough addition rate is
known through experience and is adjusted after manually sampling
for alkalinity in the causticizer.
Also, at start up, we thought that the torque reading on the
clarifier rakes would be a good indication of sludge inventory
and blankets heights. Experience has shown, however, that no
direct correlation can be made. Thus we have had to implement
the manual monitoring of the sludge blanket previously described.
In the most critical area, pH sensing of the recycle stream, we
have found that the sensors have a tendency to drift and must
be recalibrated every shift. This drifting may be caused by the
way we are sampling rather than the instrument itself. Current
practice calls for comparing the solution pH at the sensor to the
pH reading of a lab instrument and adjusting accordingly. The
effects of shielding the pH probe from the abrasive action of
the recycle stream is being investigated.
529
-------
MISCELLANEOUS HARDWARE PERFORMANCE
To date, most equipment has performed as expected with no serious
problems. The fiberglass reinforced piping on the scrubber feeds
and recycles show no evidence of abrasion. Slurry pumps and feed
pumps are also holding out well. Some plugging and scaling
problems have been encountered with other pieces of equipment,
however.
The largest problem has been with the lime slakers. They have
required considerable maintenance and attention. Most of this,
we feel, is due to the fact that we are slaking with process
water. Thus we are carrying out the same reactions in our slaker
as we are in the causticizer. Gypsum scale has been evident both
in the slakers and the slurry piping. A modification has been
made so that we can slake with city water when the water balance
of the system allows. Currently, we operate the slakers half
the time on city water and half on process water. The maintenance
has decreased drastically since starting this and the practice
of running them continuously.
Also, some plugging and scaling of the gravity lines in the system
have recently become evident. The overflow line between the
primary and softening clarifiers became partially plugged and had
to be descaled with a high pressure water hose. The scaling in
this section was probably caused by the fact that our calcium
carbonate slurry is recycled through this line. A revision in
this piping arrangement is planned. In addition, the overflow
line between the two reactors has shown signs of scaling. The
exact cause is not currently known, but calcium carbonate is
again suspected. A piping revision is also planned for here.
The reactor tanks themselves also show signs of scale build-up.
A regular maintenance program may have to be set up to maintain
the gravity flow lines.
V. PROCESS ECONOMICS
The main reason for selecting the double alkali system for
Chevrolet-Parma was that it afforded an economic advantage
over conversion to lower sulfur oil or burning low sulfur coal.
After assessing our current cost, we feel that this advantage
is still real. A precise cost estimate at this time is impossible
because of fluctuating labor requirements, projected sludge
handling changes, and rising chemical cost. Currently, our best
estimate is $12/ton of coal burned. This figure represents fixed
cost, current chemical cost, utilities, labor, and maintenance
and is broken down as follows:
530
-------
Chemical Costs Cost/Ton of Coal (55,000 Tons/Yr)
$
Lime .98
Soda Ash .82
Carbon Dioxide .07
Polymer .01
Sub Total 1.88
Utilities
Water .16
Steam .37
Electricity 1.32
Sub Total 1.85
Solid Waste .81
Labor 2.04
Maintenance .67
Fixed Costs 4.55
Sub Total 8.07
Total 11.80
This cost added to the $21/ton of coal purchase price is still
some $8/ton lower than the premium price for low sulfur coal.
The cost figures for Chev-Parma have to be used with discretion
and cannot be applied across the board for every one of GM's
powerhouses. Each one is somewhat unique and thus capital costs
for installation can vary greatly just because of the physical
plant site. As an example, an estimate of installation costs for
another GM plant was made and determined to be $8 million, more
than double that of Chevrolet-Parma. This higher fixed cost
coupled with the plant's lower steam requirements would mean
operating costs double that of Chev-Parma. For this particular
site, other control alternatives may be more economical.
531
-------
VI. SUMMARY
The double alkali system is still going through intensive
evaluation to determine its complete capabilities. In general,
the trend of the system seems to be one of steady improvement.
The system's evolution has not been easy, however, and has
required a concerted effort. What must be remembered is that
we are working with a highly developmental system. Not all
the problems have been solved, and it will take time to
establish the process and equipment reliability and the exact
operating costs. Hopefully, after a complete year of operating
experience, these questions can be answered positively and we will
have provided an economic alternative to meeting the sulfur dioxide
emission requirements. General Motors will gladly keep all
interested parties informed of our progress in meeting this
challenge.
532
-------
BIBLIOGRAPHY
1. R. J. Phillips, "Operating Experiences With A Commercial
Dual-Alkali SO2 Removal System", 67th Annual Meeting
of the Air Pollution Control Association, June, 1974,
Denver Colorado.
2. R. J. Phillips, "Sulfur Dioxide Emission Control for
Industrial Power Plants", Manufacturing Development,
General Motors Corporation.
3. D. Draemel, "An EPA Overview of Sodium-Based Double-
Alkali Processes, Part I-A View of Process Chemistry
Of Identifiable And Attractive Schemes", Control Systems
Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
533
-------
Figure 1. CHEV-PARMA
SCRUBBER BUILDING
Vacuum
Filter
A
Exhaust
Neutralizer
Recausticizer
Primary
Clari fier
Soda
Ash and
?L
Scrubber
Gas
Recycle
Tank
Softener
Clarifier
Figure 2. PROCESS FLOWSHEET
534
Scrubber
Feed
-------
Figure 3, SCRUBBER
Figure 4. RECYCLE TANK
535
-------
Figure 5. REACTOR TANKS
Figure 6. CLARIFIERS
536
-------
Figure 7. VACUUM FILTER
"J""S
090090 Q
mmtmmm
OBBGBO
Figure 8. CONTROL PANEL
537
-------
PART II
EQUIPMENT PERFORMANCE WITH
A COMMERCIAL DUAL-ALKALI SO2 REMOVAL SYSTEM
Edmund J. Piasecki
General Motors Corporation
Argonaut Division
Detroit/ Michigan
For Presentation at Environmental Protection
Agency Symposium on Flue Gas Desulfurization
Atlanta, Georgia
November 4-7, 1974
539
-------
INTRODUCTION
Industrial boilers because of their size,
load fluctuation and operating character-
istics such as high excess air, relatively
high dust loads, etc., make them differ
from utility units.
When investigating possible control systems
for S02 emissions, a basic criteria was es-
tablished. This was as follows:
1. The system must provide complete equipment
reliability.
2. The system must have a minimum of com-
plexity.
3. The system must maintain control over
a wide range of steam loads.
4. The system must be economically com-
petitive with other control strategies.
Since this was the first full-scale instal-
lation of a double-alkali control, there was
no previous experience from which to draw.
However, the Argonaut Division of General
Motors Corporation has had considerable ex-
perience in the design and development of
systems for industrial waste treatment
plants and powerhouses; and it was from
this area that equipment selection and de-
sign for this plant was utilized.
540
-------
EQUIPMENT CRITERIA AND PERFORMANCE
As with any new process, one can expect any number of start-
up problems. These may be related to the process chemistry,
mechanical equipment or a combination of both. At our Chevrolet-
Parma facility we experienced some difficulties in these areas.
We will point out problems along with a presentation of basic
design information.
Booster Fans - The existing boiler I.D. fans were retained and
the increased pressure requirements for the scrubbing system
were obtained by the addition of booster fans. The fans were
located upstream of the scrubbers to allow them to run in the
dry inlet gas stream. This approach eliminates the need for
expensive corrosion resistant materials which would be required
on the outlet side of the scrubber. The design parameters for
the booster fans are:
Boilers 1 & 2 Boilers 3 & 4
Flue Gas Flow (cfm) 61,500 58,500
Static Pressure 13" 13"
Design Temperature 393°F 730°F
Gas Density 0.0474 lbs./ft.3 0.0347 lbs./ft.3
BHP 187 178
Motor HP 200 200
Fan & Motor Speed 1180 RPM 1180 RPM
Electrical Characteristics 460-3-60 460-3-60
Scrubbers - After evaluation of many scrubbers, it was felt
that the impingement type unit would be the unit most likely
to succeed in a double-alkali type process.
This type was selected because of the need for removal of both
S02 and particulate and its resistance to plugging. Packard
towers were considered but were eliminated because of a concern
for plugging which was evidenced in the pilot plant operation.
Venturi scrubbers were eliminated because there was doubt about
their ability to absorb S02 gases. Subsequently, however, a
pilot operation has demonstrated the ability of the venturi
unit to absorb S02 gases.
Basic design parameters are as follows:
541
-------
Scrubbers 1 & 2 Scrubbers 3 & 4
Flue Gas Rate Peak
Minimum
Gas Inlet Temperature
Gas Outlet Temperature
Particulate Loading
(Ibs. part./lOOO Ibs. gas)
S02
Turn Down
Scrubber material;
Non-Welded Components
Welded Components
Mist Eliminator
(Fleximesh Type)
Shell Thickness
61,500 cfm
20,000 cfm
393°F
137°F
0.6 Inlet
0.1 Outlet
1000-3500 ppm
3 to 1
58,500 cfm
19,000 cfm
550°F
152 °F
0.6 Inlet
0.1 Outlet
1000-3500 ppm
3 to 1
316 Stainless Steel
316 L Stainless Steel
316 Stainless Steel
11 Gauge
The scrubbers, to date, have been operating at an SO2 removal
efficiency above 90 percent.
We have experienced some problems with CaC03 scaling on the
top tray of the unit. Scaling is caused by the following two
conditions:
1. High pH condition on the top tray.
2. Calcium level in the scrubber feed in excess of 250 ppm Ca++.
The top tray pH is approximately 9 and is the result of the
mixing of two separate streams on this tray; scrubber recycle
pH equals 6.5 and scrubber feed of pH equals 13.
CaC03 precipitation is pH dependent; in general, the lower the
pH the more soluble CaC03 becomes. At a pH equal to 7.0, we
feel it would be possible to operate at a Ca++ hardness of
approximately 600 ppm.
We are currently preventing scrubber scaling by giving careful
attention to softening of the feed stream.
To date, we have seen no corrosion problems in the scrubber
units.
Chemical Mix Tanks - Both chemical mix tanks Nos. 1 and 2
are constructed of carbon steel. The function of these tanks
is to mix a slurry of slaked lime with the spent scrubber
liquid. Design residence time is seven minutes at peak flow,
and turbine type agitators with low RPM and high pumping rate
have been provided.
542
-------
Over a period of months, CaSO^ scale has formed in the ten
inch line between the two mix tanks. Since this is a gravity
line, we are considering installing a flume in place of the
solid pipe. This arrangement should provide easier scale re-
moval .
Reactor Clarifier No. 1 - The unit is the reactor clarifier
type which means there is a center reaction well. Mixing in
the wall is developed by a turbine-draft type pump. This
arrangement permits recycling settled sludge which acts as
seed material for further crystal growth and also allows the
addition of flocculation aid which may be necessary for good
settling.
The primary function of this unit in our process is to settle
CaS04 and CaSC>3; this is formed in the chemical mix tanks by
lime addition.
Basic design criteria is as follows:
Design Flow 1200 GPM
Diameter 60 ft.
Liquid Depth 14 ft.
Rake Drive HP 1-1/2
Surface Rise Rate 0.46 GPM/ft.2
Rake Torque Rating 19,000 ft./lb.
Mixer HP 3
Mixer Speed 4.9 RPM Minimum
14.5 RPM Maximum
Our experience has shown, that CaSO4 forms a large crystal and
settles readily. However, CaS03 precipitates as a finely
divided crystal; for this reason, settling rate is extremely
slow. In order to operate this reactor clarifier without over-
flowing solids, it is necessary to add a high molecular weight
anionic polymer to the reaction well at a concentration of
0.25 ppm.
To aid solids removal, this unit is equipped with sludge rakes
and a cone bottom sloped at 12:1. Solids removal is started
when the solids concentration in the bottom of the clarifier
is 16 percent. When solids concentration reaches 20 percent,
the sludge no longer flows freely; at 30 percent the sludge
ceases to flow. In our case, when concentration of solids
reached 30 percent, the torque rating of the unit was exceeded
and it was necessary to drain and clean out the unit.
543
-------
Reactor Clarifier No. 2 - The primary function of this unit is
to soften the regenerated caustic stream from 800 ppm Ca++ to
200 ppm Ca++. This is accomplished by adding Na2C03 and C02-
The Na2C03 is added to the reaction well as a solution; CC>2 (gas)
is added using a diffusion ring located eight feet below the surface
of the liquid in the reaction well.
Here, as in the case of the first unit, an anionic polymer is
added to the reaction well at a concentration of 0.30 ppm to
aid solids settling.
The basic design criteria for this unit is the same as for
reactor clarifier No. 1.
Vacuum Filters - These units are the rotary vacuum filter
type.
The basic filter criteria is:
Surface Area 500 ft.2
Drum Diameter 10 ft.
Drum Face Length 16 ft.
Belt Material Nylon
Drum Drive Speed 0.06 RPM Minimum
0.5 RPM Maximum
Vacuum Receiver Size 48" Diameter X 84"
Vacuum Pump Capacity 1000 cfm at 22" Hg
Filtrate Pump Capacity 300 GPM at 15 ft. Head
The system has two vacuum filters, each capable of handling
our process solids load.
Originally, the filter belt supplied was made from poly-
propylene cloth. However, after testing other materials,
better cloth release was obtained by changing to a nylon
material.
To date, it appears that the nylon material is not wearing
as well as the polypropylene. This wear seems to be caused
by fly ash particles which are not being washed off the cloth.
We are now in the process of modifying the belt washing system
in hopes of increasing belt life.
544
-------
CONTROL PHILOSOPHY
When looking at possible systems for control of our new
process, some basic criteria for evaluation had to be es-
tablished. The following items were considered basic to
satisfactory system operation:
1. The controls must be reliable.
2. Instruments must be easy to maintain and be of such
type that powerhouse maintenance personnel would feel
comfortable with them.
3. Control system must operate with a minimum of operator
manual control.
Our conclusion was that a pneumatic instrumentation system
fills these basic needs.
The control system can be broken down into three basic areas:
1. Caustic feed to the scrubber.
2. Chemical feed for caustic regeneration and softening.
3. Sludge removal and filter cake handling.
Caustic Feed - Caustic feed to the scrubber is controlled
by two inputs into the control system:
1. Steam flow from the boiler.
2. Scrubber pH.
By assuming a fixed percentage of sulfur in the coal and Btu
content per pound of coal, we can arrive at an approximate
amount of NaOH to be added. Of course this value is only
approximate because the Btu and sulfur content of the coal
can and does vary.
Therefore, these variations must be accounted for. This can
be done by measuring scrubber pH.
If the coal has more sulfur than normal for a given steam
load, then the amount of caustic being fed to the scrubber
will be insufficient to neutralize the acid generated. This
will cause the scrubber pH to be low. Sensing this, the con-
troller will increase caustic to the unit by opening the con-
trol valve a little more.
A decrease in sulfur content would similarly cause a decrease
in caustic feed to the scrubber.
We control scrubber pH at 6.0 to 6.5; this yields S02 scrubbing
efficiency in the 90 percent range.
545
-------
Chemical Feeds - Once the amount of NaOH is known, then the
total pounds of sulfur being scrubbed is also known. There-
fore, by measuring continuously total system caustic feed
rate, chemical reactants can be ratioed and controlled auto-
matically. For example, our current lime control is set at
3 pounds CaO per 100 gallons of caustic scrubber feed used.
If the gallons of caustic used goes up, the lime feed auto-
matically increases also. NaCC>3, etc., are also controlled
in this manner.
Sludge Handling - Sludge is currently removed from the bottom
of both reactor clarifiers. The bottom stream from the No. 1
clarifier contains CaS03, CaS04 and fly ash. This clarifier
is equipped with sludge rakes. Sludge blowdown is also con-
trolled by ratio to scrubber feed.
Checks on sludge density, sludge level and rake torque determine
when sludge should be taken out of the clarifiers. Sludge level
and density are determined by manual testing while rake torque
is read from a meter.
The same method is used for removing CaCC>3 sludge from the
bottom of the second reactor clarifier.
The sludge from both reactor clarifiers is then filtered to
remove water and valuable salts. The rotary vacuum filters
used are essentially operated manually.
546
-------
SUMMARY
The dual- or double-alkali process is still a highly developmental
system and will continue to be so until the questions of per-
formance reliability and cost are answered.
We have had some process difficulties which have been solved
and a few which are under intensive study but still remain a
problem.
A one-year in-depth evaluation of the total system is now in
progress.
It should be pointed out that this system was designed for
General Motors industrial type boilers.
Utility boilers are different in their firing and operating
conditions. Therefore, direct application of industrial
boiler S02 emission control technology would not necessarily
apply to utility boilers.
547
-------
REFERENCES
1. R. J. Phillips, "Sulfur Dioxide Emission Control for
Industrial Power Plants," Manufacturing Development,
General Motors Corporation.
2. J. H. Frazier, "A System for Removal of Sulfur Oxides
from Industrial Boiler Flue Gases," Industrial Coal
Conference, Purdue University, October 7, 1971, Plant
& Environmental Engineering, General Motors Corporation.
3. J. H. Frazier, "Status of Sulfur Dioxide Control in
Boiler Flue Gas at General Motors Corporation," In-
dustrial Fuel Conference, Purdue University, October 3,
1973.
4. R. J. Phillips, "Operating Experiences with a Com-
mercial Dual-Alkali S02 Removal System," 67th Annual
Meeting of the Air Pollution Control Association,
June 9-13, 1974, Plant & Environmental Engineering,
General Motors Corporation.
548
-------
EPA-ADL DUAL ALKALI PROGRAM—INTERIM RESULTS
by
C. R. LaMantia
R. R. Lunt
J. E. Oberholtzer
E. L. Field
Arthur D. Little, Inc.
Cambridge, Massachusetts
N. Kaplan
EPA Control Systems Laboratory
Research Triangle Park, North Carolina
For Presentation at:
EPA FLUE GAS DESULFURIZATION SYMPOSIUM
ATLANTA, GEORGIA
NOVEMBER 4-7, 1974
549
-------
ABSTRACT
This paper presents interim results of the Dual Alkali Program being
conducted by Arthur D. Little, Inc. (ADL), for EPA's Control Systems
Laboratory. The program consists of process research and development
at three levels of investigation:
• Laboratory reactor studies
• Pilot plant operations in a 1200 scfm
system at ADL
• Prototype test program on a 20 megawatt
Combustion Equipment Associates (CEA)/ADL
system at Plant Scholz, Gulf Power Company/
Southern Services , Inc.
Various modes of operating dual alkali systems on high and low sulfur
fuel applications are being investigated and include:
• Concentrated and dilute sodium scrubbing systems
• Lime and/or limestone regeneration
• Slipstream sulfate treatment schemes
In each mode, the objective is to characterize the dual alkali process
in terms of SC"2 removal, chemical yields, oxidation, sulfate regenera-
tion, waste solids characteristics and soluble solids losses. This
paper reports some of the laboratory and pilot plant results obtained
to date on lime and limestone regeneration of concentrated and dilute
sodium scrubbing solutions as well as work performed on the sulfuric
acid treatment scheme for soluble sulfate precipitation.
551
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge the contributions of several
individuals to the EPA/ADL Dual Alkali Program and to the
preparation of this paper.
The earlier part of the EPA Laboratory Program was conducted
under the direction of Dean Draemel, now at the University
of California. The EPA laboratory work was carried on and
completed by James MacQueen and Robert Opferkuch of Monsanto
Research Corporation under contract to the EPA.
At ADL, major contributions were made to the pilot plant and
laboratory programs by James Valentine, Bernard Jackson and
Itamar Bodek. C. Lem Kusik provided insights into the thermo-
dynamics of the high ionic strength solutions.
The cooperation of Combustion Equipment Associates in the pilot
and prototype test programs and Gulf Power Company and Southern
Services in the prototype program are also gratefully acknowledged.
Finally, Frank Princiotta made important contributions in his
continuing involvement in the review and planning of the EPA/ADL
program.
553
-------
TABLE OF CONTENTS
Page
List of Tables 556
List of Figures 557
I. INTRODUCTION 559
A. Program Objectives and Outline 559
B. Description of Chemistry and Definition of Terms 560
II. LABORATORY AND PILOT PLANT SYSTEMS 565
A. Laboratory Methods 565
B. Pilot Plant 569
III. LIME REGENERATION — CONCENTRATED MODE 575
A. Prior Work on Lime Regeneration at ADL 575
B. Laboratory Studies of Lime Regeneration of Sulfate 576
C. Pilot Plant Results 586
IV. SULFURIC ACID TREATMENT — CONCENTRATED MODE 607
A. Laboratory Results 607
B. Pilot Plant Results 614
C. Sulfuric Acid Reactor Model 624
D. Applicability of the Sulfuric Acid Treatment System 627
V. LIMESTONE REGENERATION — CONCENTRATED MODE 629
A. Initial Laboratory Studies 629
B. Initial Pilot Plant Tests and Subsequent Studies of
Factors Affecting the Physical Properties of
Product Solids 642
C. Conclusions Regarding Limestone Regeneration of
Concentrated Sodium Bisulfite/Sulfite Solutions 649
VI. LIMESTONE/LIME DILUTE MODES 650
A. EPA Laboratory Results 650
B. Dilute Mode Alternatives 652
VII. SOUTHERN SERVICES PROTOTYPE TEST PROGRAM 661
A. System Description 661
B. General Outline 662
Glossary
References
555
-------
LIST OF TABLES
Table
No. Page
II-l Scrubber System Operating Conditions 571
III-l Sulfate Precipitation in Concentrated Mode
Lime Regeneration Laboratory Continuous Reactor—
50-Minute Residence Time 578
II1-2 Effect of CSTR pH and Hold-up Time on Solids
Properties — Concentrated Lime Mode 593
III-3 Effects on Sulfate Concentration and pH on
Settling Properties of Solids Produced in a
CSTR with a 30-Minute Holdup 594
III-4 Comparative Waste Solids Properties — Concentrated
Lime Mode 596
III-5 Summary of Closed-Loop Runs 402 and 404 —
Concentrated Lime Mode 601
IV-1 Summary of Laboratory Continuous Reactor Sulfuric
Acid Treatment Experiments 612
IV-2 Summary of Sulfuric Acid Slipstream Treatment
Results 616
IV-3 Summary of Closed-Loop Runs 621
IV-4 Model Simulations of Pilot Plant Operations 625
V-l Lime Treatment of CSTR Effluent Slurry After
Regeneration with Limestone 648
VI-1 EPA 2-Liter Reactor Continuous Fredonia Limestone
Runs — Summary 651
VI-2 EPA Runs - Potential of Line Post-Treatment Step 653
556
-------
LIST OF FIGURES
Figure
No. Page
II-l Continuously Fed, Stirred Tank Reactor System 567
II-2 Schematic Flow Diagram of- Dual Alkali Pilot Plant 570
II-3 S02 Removal vs. Feed Stoichiometry (Venturi &
2 Trays) 573
III-l Sulfate Precipitation at Various Na2SOi+
Concentrations 580
III-2 Sulfate Precipitation for Various Degrees of
Regeneration Liquor 581
III-3 Effect of Active Level on Sulfate Precipitation 582
III-4 Relation of Observed Apparent Solubility Products i
Saturation Values for Gypsum 584
III-5 Relation of Observed Apparent Solubility Products t
Saturation Values for CaS03 588
III-6 (CaSOij /CaSOx) Ratio in Solids as a Function of
Sulfate Concentrations 590
III-7 (CaS04/CaSOx) Ratio in Solids as a Function of
TOS Concentration 591
III-8 Average Wt% Insolubles in Filter Cake vs.
Concentrated Lime Mode 597
III-9 Reduction of Solubles in Filter Cake vs. Wash Ratic 598
111-10 Continuous Run 402 - ADL Reactor (Na2SOit Makeup) 603
III-ll Continuous Run 404 - CSTR Reactor (Na2C03 Makeup) 604
111-12 Composite Pilot Plant Operations for 13% Overall
Oxidation in System 605
IV-1 Schematic Flow Diagram of Pilot Plant Operation
For Sulfur Acid Treatment Mode 61.5
IV-2 H2S04 Reactor Efficiency vs. Feed Stoichiometry 617
557
-------
LIST OF FIGURES (CONT.)
Figure
No.
IV-3 Extent of CaSOi, Under-Saturation in I^SOt,
Reactor Effluent — Pilot Plant and Laboratory
Data "9
IV-4 Composite Pilot Plant Operation — H2S04 Treat-
ment Mode 623
IV-5 ^SOi, Reactor Efficiency vs. Calcium Utilization
628
V-l Comparison of Limestone Reactivities in Batch
Relations 632
V-2 Effect of Limestone Feed Stoichiometry on
Regeneration Rate with Fredonia Limestone 634
V-3 Effect of Na2SO^ Concentration on Regeneration
Rate with Fredonia Limestone 636
V-4 Effect of Temperature on Regeneration Rate
with Fredonia Limestone 637
V-5 Utilization of Two Limestones in Batch and
Continuous Reactors 639
V^6 Precipitation of Sulfate by Fredonia Limestone
in 50-minute CSTR Experiments 641
V-7 Settling Curves for Solids Produced in 50-minute
CSTR Experiments Using Fredonia Limestone 644
V-8 Effect of Soluble Magnesium on Batch Regeneration
Rates with Fredonia Limestone 645
V-9 Settling Behavior Observed for Solids Produced in
a 50-minute CSTR Using Fredonia Limestone 646
VI-1 EPA Limestone Runs 21-30 Extent of CaSO^ Under-
Saturation 654
VI-2 EPA Limestone Runs 21-30 Extent of CaS03 Super-
saturation 655
VI-3 Dilute Limestone/Lime Schematic 658
VI-4 Dilute Lime/Carbonate Softening Schematic 659
558
-------
EPA-ADL DUAL ALKALI PROGRAM—INTERIM RESULTS
I. INTRODUCTION
A. PROGRAM OBJECTIVES AND OUTLINE
Dual alkali or double alkali are generic terms used to describe flue gas
desulfurization (FGD) processes involving absorption of S02 using a
soluble alkali, followed by reaction of the scrubber effluent solution
with lime and/or limestone, to produce a solid calcium sulfite/sulfate
salt and to regenerate the soluble alkali scrubber feed solution. Many
versions of dual alkali processes are under development. Most of the
advanced systems utilize sodium as the soluble alkali; but ammonia-
based systems are receiving increasing development effort. In addition
to the type of soluble alkali used, another important difference among
the dual alkali processes is the concentration of soluble alkali in the
absorption/regeneration loop. The concentration and the choice of sol-
uble alkali are importantly related to the regeneration of soluble sul-
fate produced in the system and to the control of soluble solids losses
from the system. Many of the various schemes for dual alkali processes
were described by D. Draemel and N. Kaplan at the last EPA FGD Symposium.
Work in the EPA-ADL Dual Alkali Program is limited to sodium-based dual
alkali systems, covering a wide range of active sodium concentrations
using lime and/or limestone for regeneration. Special attention is given
to regeneration of sulfate and to the control of the soluble solids con-
tent of the calcium salts which are produced for disposal.
The objectives of the EPA-ADL Dual Alkali Program are the definition of
promising sodium-based dual alkali modes of operation and characterization
of those modes with regard to S02 removal, yields on reactants, sulfite
oxidation, sulfate regeneration, waste solids characteristics, soluble
solids losses, and overall process reliability.
The program is divided into three task areas as follows:
• Task I
• ADL Laboratory Studies — Performed at ADL Laboratories,
Cambridge, Massachusetts — Regeneration of concentrated
sodium scrubbing solutions using lime regeneration, lime-
stone regeneration, sulfuric acid treatment for sulfate
removal, detailed characterization of the physical and
chemical properties of solids produced in these regener-
ation modes.
559
-------
o EPA Laboratory Studies — Performed at EPA's Control
Systems Laboratory, Research Triangle Park, North
Carolina — Regeneration of dilute sodium scrubbing
solutions using lime regeneration, limestone regener-
ation and lime/limestone regeneration.
• Task II
ADL Pilot Plant Studies — Performed at CEA/ADL Pilot
Facility, Cambridge, Massachusetts — Promising modes
of operation using dilute and concentrated sodium
scrubbing solutions with lime regeneration, limestone
regeneration, and sulfuric acid treatment for sulfate
regeneration.
• Task III
Test Program — Prototype Dual Alkali System at Plant
Scholz, Gulf Power Company/Southern Services, Inc. —
Lime regeneration of concentrated sodium scrubbing
solutions to be performed on 20 megawatt CEA/ADL system.
Consistent with the program objectives, the important process character-
istics will be determined for each of the dual alkali modes investigated
in each task.
This paper reports some of the laboratory and pilot plant results ob-
tained in the program to date under Tasks I and II.
B. DESCRIPTION OF CHEMISTRY AND DEFINITION OF TERMS
In the absorption section of sodium-based dual alkali processes, absorp-
tion of S02 in sodium sulfite solutions occurs to produce a bisulfite
scrubber effluent solution according to the overall reaction:
Na2S03 + S02 + H20 ^ 2NaHS03 (1)
Depending upon the dual alkali mode being used, the feed to the absorber
may also contain some sodium hydroxide (formed in the regeneration section
or used as sodium makeup) and/or some sodium carbonate (used as sodium
makeup to the system). Both sodium carbonate and hydroxide form sodium
sulfite on absorption of S02 :
Na2C03 + S02 -»• Na2S03 + C02
2NaOH + S02 -> Na2S03 + H20 (3)
560
-------
which is used in further absorption to produce bisulfite. The absorber
feed solution will also contain some level of sodium sulfate in solution
and may contain some sodium bisulfite if neutralization is not completed
in the regeneration section. Sodium, identified as associated with anions
involved in S02 absorption reactions, is referred to as "active" sodium
(includes sodium sulfite, bisulfite, hydroxide and carbonate/bicarbonate).
The sulfite/bisulfite content of solutions, or total oxidizable sulfur
content, is also referred to as TOS in this paper.
Some oxidation of sulfite to sulfate occurs in the absorber due to reaction
of sulfite with oxygen in the flue gas:
2Na2S03 + 02 ^ 2Na2SOu (4)
The rate of oxidation or oxygen transfer in the absorber is a function of
the absorber design, oxygen concentration in the flue gas, flue gas tem-
perature, and the nature and concentration of the species in the scrubbing
solution. For a given set of process parameters, the oxidation rate in moles of
sulfite oxidized per unit time is relatively independent of the S02 re-
moval rate. The oxidation rate can be expressed, on an equivalent basis,
as a percentage of the S02 removed. As an example, for a high sulfur
utility boiler flue gas containing about 4-5% 02, at about 2500 ppm S02
removal, approximately 10% of the S02 removed from the flue gas will show
as sulfate in the scrubbing solution and the remaining 90% will show as
sulfite/bisulfite. At steady state, this sulfate make, equivalent to 10%
of the S02 removed, must leave the system, either as calcium sulfate or
as a purge of sodium sulfate, at the rate at which it is being formed in
the system.
Under similar conditions of absorber design, oxygen concentration and
solution characteristics, but at a much lower S02 removal rate (as in a
low sulfur coal application) , a similar rate of oxygen absorption/reaction
will result in a much higher equivalent percentage of the S02 removed—perhaps
as high as 50% of the S02 removed. Also, in industrial boiler applications,
where operation at higher excess air often occurs, high equivalent oxidation
rates can be encountered even in high sulfur coal applications. As a conse-
quence, in order to avoid unacceptable sodium sulfate purge requirements
from dual alkali systems, modes of operation capable of precipitating both
sulfite and sulfate are required; the sulfate and sulfite must leave the
system as solid calcium salts with sodium sulfate and sulfite losses
minimized.
The regeneration of acid sodium sulfite/sulfate scrubber effluent solutions
can be considered as a sequential reaction first involving neutralization
of the bisulfite using either lime or limestone:
561
-------
2NaHS03 + Ca(OH)2 + Na2S03 + CaS03 + 2H20 (5)
2NaHS03 + CaC03 ^ Na2S03 -f CaS03 + C02 + H20 (6)
In theory, the lime neutralization reaction should go to completion; com-
plete neutralization of bisulfite is not possible with CaC03. Using lime,
the regeneration can be carried beyond neutralization to generate caustic:
Na2S03 + Ca(OH)2 ^ 2NaOH + CaS03 (7)
to some equilibrium hydroxide concentration. The usual form of calcium
sulfite produced is the hemihydrate, CaS03'1/2H20.
Depending upon the concentration of sulfite and sulfate and the pH of the
solution, the following reaction for sulfate removal also occurs simul-
taneously with neutralization and regeneration reactions (5)-(7) using
either lime or limestone:
+ SO i^" * CaSOit (8)
The exact form of this calcium sulfate is under investigation and will be
discussed later in this paper.
Thus, the level of sulfate precipitation in the overall scheme is given by
the ratio of calcium sulfate to total calcium/sulfur salts produced:
„ , ,- ^ „ . . . raols
Sulfate Precipitation -
moi CaSO, + CaS03
If the level of sulfate formation or sulfite oxidation given by:
c i c .. r. • mols SOq oxidized
Sulfate Formation = - : — rrr - - —
mol S02 removed
is matched by the level of sulfate precipitation, then all sulfur removed
from the flue gas can leave the system as a calcium salt and no soluble
sulfate purge is necessary to maintain a sulfate balance in the system.
In practice, even if such a balance is established, the washed calcium
sulfite/sulfate salts will contain some soluble sodium salts as well as
562
-------
soluble fly ash constituents which must be purged and some sodium makeup
to the system will therefore be required. Calcium utilization or yield
in the overall process is defined as:
n i • ,, -1- • mols CaSOq + CaSOi+ generated „
Calcium Utilization = dr—r—T~J x 100%
mol Ca fed
regardless of whether lime or limestone is used.
Dual alkali modes can be designed to accommodate varying levels of oxida-
tion, with equivalent precipitation, up to an oxidation rate of 100%. For
example, in the limit, a solution of about 0.5 molar sodium sulfate can be
reacted with lime according to the following equation:
Ca(OH)2 + 2H20 £ 2NaOH + CaSOt,'2H20 (9)
to produce gypsum and a solution containing approximately 0.1-0.15 molar
sodium hydroxide and 0.45 molar sodium sulfate. If this solution, con-
taining this relatively dilute concentration of active alkali, is re-
cycled to the absorber where 100% oxidation recurs, then a 0.5 molar sodium
sulfate solution is returned to the regeneration system and the cycle is
repeated; a dual alkali system can be designed on this basis.
As the sulfite concentration in the scrubber effluent is increased, in-
creasing the sulfite/sulfate concentration ratio in solution, the percent
sulfate precipitation decreases. However, the higher concentration of
active sodium species (sulfites, hydroxide and carbonates) reduces both
the scrubber liquid feed rate requirement and the feed rate to the re-
generation system, for a given S02 removal rate. As the concentration
of active sodium increases, the regeneration system size is thereby re-
duced as are the liquid rates throughout the system. Thus, as the sulfate
precipitation requirements decrease, the system size and recirculation
rates can be reduced by increasing active sodium concentrations.
For purposes of this program, dilute operating modes are defined as those
involving solutions containing active sodium concentrations less than or
equal to 0.15 molar active Na , where active sodium is considered to be
sodium sulfite/bisulfite, carbonate/bicarbonate or hydroxide. Concen-
trated modes are those involving solutions containing active sodium con-
centrations greater than 0.15 molar active Na"1".
As will be discussed in more detail, the sulfite and sulfate concentra-
tions, besides affecting sulfate precipitation, also have a profound
effect on the physical properties of the calcium/sulfur salts and, as
a result, on the settling and filtration characteristics of the solids
produced in dual alkali modes.
563
-------
As an alternate approach to operating at decreased active sodium concen-
trations, high levels of sulfate precipitation can be achieved using sul-
furic acid treatment for precipitation of sulfate according to the following
equation :
H2SOit + 2CaS03'l/2H20 + 3H20 £ 2NaHS03 + 2CaSO^'2H20 (10)
This reaction is carried out at a low pH (2-3) , where sulfite is converted
to bisulfite, thereby bringing calcium sulfite into solution and exceeding
the solubility product for calcium sulfate. This reaction can be applied
on a process slipstream for sulfate precipitation, as required. However,
in the overall process this scheme utilizes sulfuric acid, requiring
additional lime or limestone for ultimate neutralization of this acidity
added to the system and generating an equivalent additional amount of
waste calcium/sulfur salts. This scheme is being practiced commercially
in Japan and offers the advantages of operating a concentrated active
sodium absorption/regeneration system with the ability to sustain high
sulfate oxidation levels with equivalent precipitation of calcium sulfate.
564
-------
II. LABORATORY AND PILOT PLANT SYSTEMS
A. LABORATORY METHODS
The effects of a variety of variables on the kinetics and equilibria of
several different reactions related to dual alkali regeneration have
been studied during the course of the laboratory experimental program.
For each of the several basic reactions studied, the general approach
was first to conduct a series of batch experiments in which samples of
the reacting mixture were taken periodically and analyzed to determine
reaction rates and final equilibrium compositions. Additional studies
were then carried out in a continuously fed, stirred tank reactor (CSTR)
to investigate the reaction further under conditions more representative
of those in a continuous process situation.
Although a variety of reactions were studied, the experimental proce-
dures were generally quite similar, since in most cases a solid reagent
was reacted with a solution to produce a two-phase product slurry. The
general experimental procedures used are presented in this section. In
a few instances where a specific procedure was employed for a particular
experiment, the procedural details are discussed along with the experi-
mental findings.
1. Experimental Apparatus, Operation and Sampling Procedures
Batch experiments were carried out in a 1-liter, 3-necked, round bottom
flask, immersed in a controlled-temperature water bath. A magnetically
driven stirrer provided mixing. To prevent oxidation during the course
of the reaction, a slow constant purge of nitrogen was maintained over
the reactor headspace.
Batch experiments were initiated by weighing appropriate amounts of the
water soluble reactants (reagent grade materials were used exclusively)
into the flask, adding the required amount of water, initiating the ni-
trogen purge, and stirring to effect dissolution. When dissolution was
complete, a sample of the solution was first pipetted from the flask for
analysis to establish the solution composition accurately, and then the
solid reactant was added to the solution to initiate the reaction. Pe-
riodically, the reactor was opened momentarily to remove a sample of the
reaction slurry for analysis; procedures used for phase separation and
analysis are described below.
565
-------
The CSTR system used was a rather simple apparatus shown schematically
in Figure II-l. The reactor was a 1.5-liter glass vessel, commonly
known as a resin kettle, fitted with an overflow sidearm positioned to
yield a reactor volume of about 1 liter. Mixing was performed with an
electrically driven stirrer, operating at about 700 rpm. Four rectan-
gular pieces of Plexiglas (T.M. of Rohm & Haas Co.) about 2 cm wide and
0.3 cm thick, installed vertically around the sides of the reactor,
served as baffles to ensure good vertical mixing. The lower portion
of the reactor was immersed in a constant temperature bath for temper-
ature control.
Reactant feed lines, a thermometer, an inlet for nitrogen purge, and
the stirrer shaft all entered the reactor through air-tight, ground
glass joints in the lid; the lid and reactor bottom had mating ground
glass surfaces which could be assembled air tight. During operation,
the reactor headspace was purged with a flow of nitrogen which exited
via the overflow sidearm.
Reactants were prepared and fed from two, stirred, 14-liter glass ves-
sels with hermetically sealed stainless steel tops. The solid reagent
was slurried with water, and fed from one tank (baffled) and the sodium
sulfite/bisulfite/sulfate solution from the second. Reactants were
pumped at controlled rates by variable speed peristaltic pumps (Sigma
motor T6S and Masterflex 7545).
2. Analytical Procedures
The feed and effluent samples were collected and analyzed by procedures
which permitted mass flow rates into and out of the reactor to be de-
termined for each of the various chemical species of interest. Samples
of solutions and slurries were collected for known periods of time in
tared flasks and the total mass collected was determined gravimetrically.
The slurries were then filtered, and the solids washed thoroughly with
a saturated calcium sulfate solution. The weight of the resulting wet
cake was determined and a representative portion taken and dried at
85°C. From the weights determined above, the measured weight loss upon
drying, and the solution density, the mass flow rates of both phases of
the slurry could be calculated.
Aliquots of liquids and dried solids were characterized by a variety of
analytical techniques, most of them simple gravimetric or titrimetric
procedures. The analytical methods employed were:
Parameter Method
Sulfur (IV), TOS lodometric Titration
Total Sulfur (Solids) Oxidize, Weigh as BaS04
Total Sulfur (Solutions) Titrate with Lead Perchlorate
Acidity/Hydroxide NaOH/HCl Titration
566
-------
OK)
Solution Feed
Tank
Peristaltic Pump
C>0
Slurry Feed
Tank
Feed Sampling Valves
T—
o
Peristaltic Pump
T
Thermometer
Baffle
~\
\
Overflow
Water Bath
(stirred, electrically heated)
Figure 11-1 Continuously Fed, Stirred Tank Reactor System
-------
Parameter Method
Soluble Carbonate Titration of HC03~ with HC1
Solid Carbonate Acidify, Collect C02, Titrate
as above
Calcium EDTA Titration
Mg++", Na+, (trace Ca++) Atomic Absorption Spectrometry
The complete details of all analytical procedures employed are availa-
ble upon request from the authors.
Determination of the amount of Na2SOij which reacted with lime or lime-
stone to precipitate CaSO^ from sodium sulfite/bisulfite/sulfate solu-
tions was important in many of the laboratory experiments conducted.
This determination required that the filtered product solids be washed
thoroughly to remove entrained mother liquor without dissolving CaSO^
from the solids. It was for this reason that a wash solution saturated
with respect to CaSOif-2H20 was used for washing the solids. Prior to
using this washing procedure routinely, analyses of dissolved calcium
in the fresh and used wash solutions were performed to ascertain that
no additional calcium sulfate was being introduced into the solids by
the washing procedure.
The washed solids were subsequently analyzed for total sulfur, TOS, cal-
cium, and sodium to permit the calculation of the amount of CaSO^ pre-
sent. Even after thorough washing (1 gram of dry solids washed with
four, 25-ml portions of wash solution), a small amount of sodium, usu-
ally on the order of 3 mole percent of the total sulfur present, re-
mained in the solids. That sodium was attributed to Na2SOi4 entrapped
within the calcium sulfite crystals, and it was excluded when calculat-
ing percent calcium sulfate in the solids. The percent calcium sulfate
in the solids was calculated from the number of millimoles of CaSOx and
per gram of total solids according to the relation
Sulfate Precipitation =
~CaSOx - CaS03~|
CaSO_.
L x J
Because this calculation involves taking a relatively small difference
between two larger numbers, the result is quite sensitive to small ana-
lytical errors in the determination of CaSO and CaSOs. For example,
in one actual case where a calcium sulfate content of 9.4% was observed,
a 1% increase in the amount of CaSO accompanied by a 1% decrease in
the amount of CaSOs would have changed the calculated calcium sulfate
content to 11.2%. Analytical uncertainty of that order is not unreason-
able and its effect must be kept in mind when interpreting the experi-
mental findings relating to sulfate regeneration which are discussed
in succeeding sections.
568
-------
B. PILOT PLANT
1. Description of the Facility
The dual alkali pilot plant at the ADL complex in Cambridge, Massachusetts,
was built jointly by Combustion Equipment Associates and ADL in 1973. The
system operates on 2000 acfra (400°F) of the combustion flue gas from a
natural gas fired hot air furnace. S02 is added to the gas to provide
any desired S02 concentration up to about 4000 ppm.
The basic facility incorporates three process sections: gas scrubbing,
absorbent regeneration, and solids separation. In addition, there are
auxiliary process systems and equipment for special slipstream treatment
operations. A schematic flowsheet of the major process units is provided
in Figure II-2.
The scrubbing system includes a variable throat venturi scrubber followed
by an absorption tower fitted with a radial vane demister. The absorption
tower is equipped with removable trays and sprays so that it can be
operated alternatively as a spray tower, a tray tower (with from one to
three trays), or a deentrainment separator.
The absorbent regeneration system contains four reactor vessels ranging
in size from 17 to 700 gallons. Two lime/limestone feed systems are
available—one for metering regeneration reactant as a dry solid and one
for feeding reactant in slurry form.
The solids dewatering and separation system includes a 6 foot diameter
Dorr Oliver thickener with three overflow levels; and a Dorr Oliver rotary
drum vacuum filter with an active cloth area of 4.3 feet . The filter is
equipped with a wash water spray system which can be operated in a variety
of spray configurations.
In addition to this basic process equipment, the pilot plant has a number
of auxiliary units including two Pfaudler kettles (one glass lined and one
stainless) and a 6 inch Bird solid bowl centrifuge for operating slipstream
treatment processes.
2. Scrubber Operating Characteristics
The standard scrubber configuration for the EPA test program consisted
of the venturi scrubber followed by the absorption tower with two sieve
trays. The tray tower was operated on a once-through basis with respect
to the absorbent liquor, with the flow leaving the bottom tray being sent
directly to the venturi recycle tank. There was no recycle of any spent
absorbent solution around the tray tower. The standard operating conditions
for the scrubber circuit are provided in Table II-l.
Because the scrubber system operation does not change appreciably for
different system conditions in modes using concentrated sodium solutions,
the operational characteristics of the scrubber system are reviewed here
rather than for each particular system operating mode.
569
-------
Flue Gas
Trav Scrubber
With Demister and
2 - Trays
Ln
-J
o
Ca(OH)2 or CaCO3
Settler
Reactor System
V \ \ \
Figure 11-2 Schematic Flow Diagram of Dua! Alkali Pilot Plant
-------
Table II-l SCRUBBER SYSTEM OPERATING CONDITIONS
Flue Gas Conditions
Temperature Entering Venturi = 390 - 430°F
Inlet Flow Rate: 1100 scfm
NO Level: 800 - 1000 ppm
x
S02 Level: 500 - 2700 ppm
Oa Level: 4 - 5.5% (dry basis)
H20 Level: 18 - 23% (dry basis)
Gas Dew Point: 126 - 131°F
Venturi Scrubber Conditions
Venturi Tank Temperature: 140 - 145°F
Venturi Recycle Flow: L/G = 16 gals/Macfm sat'd
Venturi Pressure Drop: 10 - 14 in. H 0
Tray Tower Conditions (Two Sieve Trays)
Tray Tower Feed: L/G 1.5-3 gals/Macfm sat'd
Pressure Drop/Tray: 1.6 - 2.0 in. H 0
571
-------
SOg Removal The S02 removal efficiency in the pilot plant scrubber
system has been evaluated in a number of previous studies as well as during
this program. The overall S02 removal has been shown to be a function of
the system feed stoichiometry (mols active Na+ capacity/mols S02 inlet),
the scrubber configuration, and the inlet S02 level. The relationship
between S02 removal and feed stoichiometry for the venturi two tray
configuration is shown in Figure II~3. The data indicate that for a
given feed stoichiometry, lower S02 levels tend to result in slightly
lower removal efficiencies.
The plot indicates that to achieve a 90+% removal from a 2500 ppm inlet
S02 level requires a feed stoichiometry of about 1.1; while to achieve the
same 90+% removal from a 500 ppm inlet S02 level requires a feed stoichiometry
of about 1.3.
This trend would be expected since at higher S02 levels there is a greater
driving force for removal; hence, higher removal efficiencies are more
easily attained. This is somewhat different than the experience in a direct
lime/limestone scrubbing system where SC2 absorption is limited by liquid
phase and solid/liquid mass transfer. In sodium absorption above feed
stoichiometries of 1.1, gas phase mass transfer appears as a limiting
rate.
Scrubber Oxidation Rates Extensive tests in the pilot plant scrubber
system have shown that in the concentrated active sodium ranges (0.2 - 0.5M
Na+) the rate of sulfite oxidation (mols sulfite oxidized/unit time) is essentially
constant for a given scrubber configuration and flue gas condition. This
result implies that the oxidation of sulfite in the scrubber is limited by
oxygen mass transfer. Thus, as the amount of gas/liquid contacting increases
or as the level of oxygen in the flue gas increases, the rate of sulfite
oxidation will correspondingly increase—which is in fact what happens.
Oxidation in the scrubber system plays an important role in the feed stoichio-
metry. Since scrubber system oxidation is virtually constant, the sulfite
oxidation rates as a percentage of S02 absorbed (and total active sodium
throughput) increase with decreasing inlet S02 concentration. Thus, for
low S02 levels a greater percentage of the active sodium is converted to sulfate
in the scrubber.
For the pilot plant venturi/two tray scrubber system with a flue gas oxygen
concentration of 4.5 - 5.0%, the average oxidation rate is equivalent to
about 240 ppm of S02. The stability of this oxidation rate for these
conditions means that as S02 inlet levels decrease, the oxidation rate as
a percentage of the S02 inlet increases.
It should be noted that these oxidation and S02 removal data were obtained
at scrubber liquor operating temperatures which were 15 - 20°F higher than
those which would be expected in a scrubber operating on a normal utility
or industrial boiler. The higher temperatures were due to the fact that
572
-------
100
95
90
c
o
g 85
o
80
75
Scrubber Operating Conditions
Bleed Liquor Temperature = 140°F - 150°F
Active [Na+] = 0.3 - 0.5 M
Total Dissolved Solids = 5 • 15 wt%
Sulfite Oxidation = 250 ppm SC>2 Equivalent
Venturi: P = 11 - 14 in. H20
L/G =15-17 gal/Macf saturated
O
Range of SO2 Inlet Levels
O 500-700 ppm
A 800-1,000 pprn
02,250-2,650 ppm
D
-1.7
70
0.7
0.8 0.9 1.0 1.1 1.2 1.3
Scrubber Feed Stoichiometry (mols Na Capacitv/mol S02 inlet)
Figure 11-3 SO2 Removal Versus Feed-Stoichiometry (Venturi + 2 Trays)
1.4
1.5
-------
the flue gas was generated from natural gas fuel which has a much lower
carbon-to-hydrogen ratio than coal, and which when burned produces
combustion product gas with considerably higher humidity levels. These
higher humidity levels result in higher saturated gas temperatures in the
scrubber which affect both S02 removal and sulfite oxidation rates. The
equilibrium vapor pressure of S02 is on the order of 50 - 60% higher, so
it would be expected that the removal efficiencies observed in the pilot
plant would be lower than those attainable in actual boiler applications.
Similarly, the rate of sulfite oxidation is strongly dependent upon
temperature, increasing with increasing temperature. Thus, the oxidation
rates observed in the pilot plant may also be slightly high compared to
those expected in actual applications.
It should also be noted, however, that materials present in fly ash have
been cited as being catalytic to oxidation in the system, and that
the pilot unit used in these studies was free of fly ash. Conversely,
experience has been cited in which lime/limestone systems containing
fly ash have had lower oxidation rates than the same systems without fly
ash. Apparently, fly ash is capable of promoting or inhibiting oxidation
depending on its specific composition. However, the pilot plant was
operated to produce oxidation rates on the order of those which would be
expected under normal operating conditions in a full scale utility application.
574
-------
III. LIME REGENERATION—CONCENTRATED MODE
A. PRIOR WORK ON LIME REGENERATION AT ADL
Prior to initiation of the EPA/ADL Dual Alkali Program, laboratory and
pilot plant research and development were conducted at ADL on lime re-
generation of concentrated scrubbing solutions. In 1971 ADL conducted
a detailed laboratory study of the regeneration of simulated sodium
scrubbing solutions using hydrated lime. This work, sponsored by the
Illinois Institute for Environmental Quality,2 dealt exclusively with
characterizing the nature of the regeneration reaction. The objectives
of the laboratory program were the development of reaction conditions
which would produce waste calcium salts with good settling and filtra-
tion characteristics; production of scrubber feed solutions with low
concentrations of dissolved and suspended calcium salts; and good util-
ization of lime in the regeneration reaction.
The objectives of these laboratory studies were satisfied and conditions
for an effective regeneration reaction were determined. Chemical analy-
ses in the laboratory were limited to liquid phase analyses; because of
the generally high sulfate concentrations in the reactor feed, changes
in sulfate concentration x^ere usually within the experimental error of
the analyses. As a consequence, no significant sulfate precipitation
was detected in the lime regeneration of concentrated sulfite solutions.
The dual alkali process still appeared attractive because of the highly
effective S02 removal, very low calcium concentrations in the regener-
ated scrubber feed, and the effectively complete utilization of lime
in the process. Based upon the laboratory results, the dual alkali pi-
lot plant was constructed at ADL's facilities. The pilot plant was oper-
ated to generate the design data for the 20 megawatt prototype dual al-
kali system built by CEA for Gulf Power Company/Southern Services.
As part of those pilot operations, analytical techniques were developed
for the solids generated in the dual alkali system. Analyses indicated
that pilot plant solids contained 10-20% of the calcium sulfur salts
as calcium sulfate and that simultaneous calcium sulfate precipitation
was occurring, to a limited extent, with precipitation of calcium sul-
fite in concentrated scrubbing solutions. Care was exercised in analy-
sis to ensure that the sulfate found in the solids was not either oc-
cluded Na2SOit or precipitated CaS03 which had later oxidized to CaS04 .
The results of the laboratory and pilot work conducted prior to the
EPA/ADL program are summarized in a recent paper.3
575
-------
B. LABORATORY STUDIES OF LIME REGENERATION OF SULFATE
1. Introduction
The observation during ADL pilot plant tests that calcium sulfate was
present in the product solids when regenerating with lime in a concen-
trated active sodium mode aroused a great deal of interest because in
a concentrated mode the removal of sulfate produced as a result of sul-
fite oxidation was foreseen as a possibly significant problem. From a
consideration of the thermodynamic equilibria involved, one would not
expect calcium sulfate to precipitate during regeneration of concentrated
sodium sulf ite/bisulf ite/sulfate solutions.
Consider a typical solution in a concentrated mode, regenerated to a
pH in the range 9-11, containing 0.5M ^250^ and 0.25M Na2S03. Equi-
librium levels of total dissolved calcium are typically less than 100
ppm (0.0025M). For calcium sulfate to precipitate as gypsum
2H20), the apparent solubility product, "Ksp" (the product of total
dissolved calcium and total sulfate concentrations), is about 1.05 x
10"2. For the 0.5M Na2SOi4 level taken as an example, a calcium level
of 0.021M, about 10 times as much as is actually there, would be re-
quired to effect the precipitation of gypsum.
Subsequent to observing the presence of CaSO^ in the solids produced at
the ADL dual alkali pilot plant, Borgwardt*4 demonstrated quite conclu-
sively with his pilot limestone slurry scrubber that he could produce
calcium sulfite product solids containing significant amounts of CaSO^
from liquors unsaturated with respect to gypsum. He could not identify
a distinct gypsum or anhydrite (anhydrous CaSOij) crystalline phase in
the solids and proposed that the CaSOit was present as some sort of solid
solution within the CaS03'%H20 crystal. He further found that he could
correlate the ratio of CaSO^/CaSO^ in the solids quite well with the
activities of the sulfate ion in the liquor.
The laboratory program conducted to study this apparently similar phe-
nomenon in the concentrated lime dual alkali system consisted of a series
of continuously fed, stirred tank reactor (CSTR) experiments. They were
designed to first verify under "normal" operating conditions that sulfate
was in fact being precipitated and then to study the effect of Na2SOit
concentration on the amount of sulfate precipitation. Studies were also
conducted to determine the effect on sulfate precipitation of less than
complete neutralization of bisulfite, regeneration with shorter and
longer than normal reaction times, and higher and lower than normal
levels of active sodium.
Normal reactor residence time for these experiments was established at
50 minutes, a time, based on previous ADL experience, adequate for com-
pletion of reactions, even if regeneration is carried well into the hy-
droxide range. The solution to be regenerated normally contained about
576
-------
0.5M active sodium at a pH ranging from 5.4 to 5.6 (HS03~/S03~ ratio
approximately 10). As noted earlier, in the laboratory apparatus lime
was fed to the reactor as a 5 - 10% slurry which diluted the sodium
feed by 10 - 20% upon mixing in the reactor. In the subsequent dis-
cussions, when reactant concentrations are reported, they are after
dilution by the slurry water.
2. Experimental Results
The precipitation of CaSOi, observed under normal conditions at several
levels of dissolved sulfate is tabulated, along with the associated
solution compositions, in Table III-l. The first two experiments, A
and B, are essentially identical except for the change in concentration
of Na2SOi+ from about 0.2M to about 0.55M. In both cases measurable
amounts of sulfate precipitation were observed — precipitation in-
creased with increasing sulfate concentration. In experiment B, re-
generation was carried out to the point where the HS03~ fed to the re-
actor was just 100% neutralized and no free hydroxide was generated
(sulfate precipitation was 9.1%).
Since in certain cases it proved difficult to adjust reagent flows to
achieve exactly 100% neutralization of the HSOs" present, and in other
experiments, the degree of regeneration was intentionally varied, the
term "percent neutralization" has been employed to define the extent
to which regeneration was carried out. When not carried to complete
neutralization, the percent neutralization was computed from the rela-
tion
T> M i • ["[HS03-] initial - [HSO^~] final]
Percent Neutralization = |^ * [HS03-] initial |x
When regeneration was carried on into the free hydroxide region, the
neutralization was greater than 100% and was calculated from the rela-
tion
Percent Neutralization =
[HSO;T] initial + [OH~] final"
[HS03-] initial
x 100%
Experiment C was similar to B except that percent neutralization was
increased from 100% to 124%. However, the sulfate precipitation (9.0%)
was essentially identical to that observed at 100% neutralization. In
experiment D, the concentration of sulfate was further increased to
about 1.1M, and sulfate precipitation increased to 16.5%. Experiment D
was carried to only 67% neutralization; however, results which will be
presented later establish that ^280^ level, rather than incomplete
neutralization, was the more significant factor in the increased amount
of calcium sulfate precipitation observed.
577
-------
Table II1-1 SULFATE PRECIPITATION IN CONCENTRATED MODE LIME REGENERATION LABORATORY CONTINUOUS REACTOR—50-MIX RESIDENCE TIME
Expt. [HSOO.M
Ln
CO
A 0.
3 0.
C 0.
D 0.
,343
322
.322
,354
0.046
0
0
0
.041
.030
.050
[SO,. ],M
0.204
0.
0.
1.
543
531
110
£!i
5.5
5
5
5
.4
.4
.4
[HSO-<~].M
0.020
nil
nil
0.116
[Oii"],M
nil
nil
0.077
nil
[SO
0.
0.
0.
0.
1 M
•> j ,n
208
211
161
168
0.200
0.
0.
1.
,540
.518
,051
[Ca l.ppm pH Neutralized
20
60
56
84
7
8
11
6
.8
.5
.5
.8
94
100
124
67
Utilized3 in
95
103
90
97
0
0
0
0
solids
.023
.091
.090
.165
Based on (CaSO /Ca ,) x 100? in solids.
x total
-------
The sulfate precipitated in experiments A - D is shown as a function of
sulfate concentration in Figure III-l. Within experimental error, sul-
fate precipitation was directly proportional to sulfate concentration
in the reactor. (It should be remembered from the discussion of experi-
mental error in Section II-A that sulfate precipitation is subject to
some experimental error — for a true precipitation of 8%, experimental
error could cause the calculated value to be in the range from 6 - 10%.)
Included in Figure III-l are the results of additional experiments
carried out at much shorter than normal residence times (5 and 8 minutes),
but at the same active sodium level as experiments A - D and all at 100%
neutralization. With the possible exception of the experiment at 0.2M
Na2SOi+ and 5-minute residence time, calcium sulfate precipitation and
its dependence on sodium sulfate concentration was essentially the same
as that observed at the normal, 50-minute residence time. Also in-
cluded in Figure III-l is the effect of allowing portions of reactor
effluent slurry to stand with gentle stirring for a period of 3 hours
after emerging from the reactor. In every case, a decrease in the mole
ratio CaSOi^/CaSOx in the solids was observed upon standing.
The effect of percent neutralization was studied in detail with the
CSTR operating at a 5-minute residence time followed by a 3-hour hold
time to observe the effect of time on reaction products, equilibria,
and product stability. Observed sulfate precipitation is shown in
Figure III-2 as a function of percent bisulfite neutralized. In these
studies, active sodium (after dilution) and the pH of the feed were
maintained at "normal" levels; except for the several cases where spe-
cifically indicated, the concentration of Na2SOi4 was between 0.64 and
0.69M. In those cases where the Na2SOtt concentration fell outside
that range, the arrows indicate what the probable sulfate precipitation
would have been at a sodium sulfate concentration of 0.66M based on the
N32SOL, dependence shown in Figure III-l. Although there is some scat-
ter in the data, increased sulfate precipitation seemed to occur at
very high and very low percent neutralizations. Also shown in Figure
III-2 is the general decrease in CaSOi4/CaSOx in the solids when the
product slurry was allowed to react for an additional period of 3 hours.
Sulfate precipitation as a function of the concentration of TOS was
studied at an 8-minute CSTR residence time and the results observed are
shown in Figure III-3. Again, the pH of the feed solution was main-
tained constant at 5.5 in these experiments. Over the relatively small
range studied to date, no significant variation in sulfate precipita-
tion was observed.
3. Discussion
The experimental work described has been aimed primarily at developing
a practical understanding of what variables are important to, and in
what manner they influence, sulfate precipitation. Experimental re-
sults clearly establish that calcium sulfate can be precipitated when
reacting concentrated sodium sulfite/bisulfite/sulfate solutions with
lime. A limited number of samples of the solids produced under a
579
-------
00
O
o
C/3
0>
3
HI
L_
O
cc
c
o
V)
O
O
CD
O
0.20
0.16
0.12
0.08
0.04
Constant Conditions
[Na+l Active ^°-45M (diluted)
Feed pH = 5.4-5.6
% Neutralization - 100%
Temperature - 52°C
O
(70% Neutral.)
^1
O
Legend
A 5n-MinuteCSTR
A 3-Hour Reaction After 50-MinuteCSTR
O 5-Minute CSTR
• 3-Hour Reaction After 5-Minute CSTR
OS-Minute CSTR
0.2
0.4
0.6
0.8
[Na2S04l.M
1.0
1.2
1.4
1.6
Figure 1 1 1-1 Sulfate Precipitation at Various
Concentrations
-------
0.20
o 0.16
c
0)
3
LU
O
I 0.12
EC
8
<3
o 0.08
E
~^r
O
CO
<3
_w
0.04
Constant Conditions
[Na+] Aaive ==Q.45M (diluted)
Feed pH = 5.4-5.6
Temperature = 50°C
[Na2SO,j] as 0.67M (diluted), except where specified t4-.
(0.56MJ*
Legend
°5-Minute CSTR
•3-Hour Additional Reaction
_L
_L
20
40
60
140
80 100 120
Percent HSOJ Neutralized
Figure 111-2 Sulfate Precipitation For Various Degrees of Liquor Regeneration
160
180
200
-------
oo
NJ
C
3
O
G
8
cc
c
O
to
O
w
3
0.16
0.12
0.08
0.04
0.1
Constant Conditions
Feed pH = 5.4-5.6
Temperature - 52°C
[Na2SO4l = 0.6M
% Neutralization = 100%
D
0.2
0.3
I
0.4 0.5
[TOS]
Legend
° 5-Minute CSTR
D8-MinuteCSTR
0.6
0.7
0.8
Figure 111-3 Effect of Active Sodium Level on Sulfate Precipitation
-------
variety of conditions have been examined by X-ray diffraction and the
only sulfur-containing phase observed was CaSOa'^l^O. Borgwardt found
the same to be true with the CaSO^-containing solids produced by his
direct slurry scrubbing system when operated unsaturated in gypsum;
his hypothesis that CaSOi4 is in some way incorporated as a solid solu-
tion, or otherwise, in the CaSOs-^I^O crystal lattice seems to be ap-
plicable to the behavior observed in these studies as well.
While no attempt has been made to use rigorous solution thermodynamics
to explain the effect of the several independent variables studied on
observed calcium sulfate precipitation, the direct proportionality be-
tween the most significant independent variable, Na2SOi4 concentration,
and the percent CaSO^ in the product solids is not surprising since,
simplistically, increasing the S0t,=/S03= ratio should favor precipita-
tion (or inclusion) of CaSO^ in the presence of calcium sulfite. The
lower CaSOi4/CaSOx ratio observed at intermediate percent neutralizations
is also not unreasonable because S03= concentration is relatively higher
in that region than at either extreme. However, the observation that
sulfate precipitation was not changed significantly when the concentra-
tion of TOS was changed was unexpected; at constant pH, S03= concentra-
tion is directly proportional to the concentration of TOS. Additional
CSTR experiments covering a wider range of TOS concentrations are cur-
rently underway to define this effect better.
A general decrease in CaSO^/CaSOx mole ratio when the effluent slurry
from the 5-minute CSTR was allowed to continue to react obviously in-
dicates that the initial effluent was not at equilibrium. The initial
effluent liquid phase tended to be supersaturated with respect to CaS03
as evidenced by the subsequent appearance of solids (largely CaSOs-^T^O)
upon allowing samples of filtered reactor effluent to stand. However,
the change in total dissolved calcium which occurred when the intact
slurry samples were allowed to stand was, in most cases, too small to
account for the change in CaSO^/CaSOx observed if only a subsequent
precipitation of additional calcium sulfite had occurred. Thus, one
must conclude that some of the CaSO^ initially precipitated, in fact,
redissolved as the mixture was allowed to equilibrate.
In no case were the measured levels of total dissolved calcium and
sulfate high enough to even approach saturation with respect to gypsum.
Values of "K ", the "apparent solubility product", which were observed
under a number of reaction conditions are shown in Figure III-4. Also
shown is the calculated value of "K " as a function of ionic strength
for solutions saturated with respectPto gypsum. These values of "KSp"
were calculated by the method of Meissner and Kusik, which was devel-
oped to deal with solutions at high ionic strengths. The accuracy of
the predicted values of "Ksp" is confirmed by the close agreement be-
tween the calculated values as shown by the curve and experimental values
derived from EPA experiments in which dual alkali reaction effluent
slurries were saturated with gypsum (Section VI-A) during saturation
tests.
583
-------
o*
+
+
E
3
o
3
T3
O
.Q
3
O
C/5
a
a
10
-2
10
-3
10
-4
10
-5
I I I
Legend
A 50-MinuteCSTR, Vary [Na2S04)
D 8-Minute CSTR, Vary [Na+] Active
O 25% Neut., 5-Minute CSTR
• 25% Neut., 3-Hour Reaction After
5-Minute CSTR
x EPA Saturation Tests
I I I
H, Ionic Strength
Figure 111-4 Relation of Observed Apparent Solubility
Products to Saturation Values For Gypsum
584
-------
For a given experiment, one can express the degree of saturation of
the solution with respect to gypsum by expressing the observed concen-
tration of total soluble calcium as a percent of the corresponding cal-
cium level at saturation taken from the curve. The percent saturation
calculated for the three 50-minute CSTR experiments shown in Figure
III-4 ranges from 1 to 11%, increasing with increased Na2SOi4 concentra-
tion and CaSOt,/CaSOx ratio. The largest "Ksp" observed — after 25%
neutralization in a 5-minute CSTR — corresponded to a solution only
22% saturated with respect to gypsum. Experimental studies to better
understand the mechanism of CaSOij inclusion and its kinetic and/or
thermodynamic basis are continuing in the ADL laboratory program.
585
-------
C. PILOT PLANT OPERATIONS
1. Pilot Plant Test Program
The evaluation of the dual alkali process operation in the concentrated
active sodium mode using lime for absorbent regeneration was focused
primarily on the performance of the regeneration system. The operation
of the regeneration reactor ultimately determines the required calcium-
S02 stoichiometry as well as the properties of the waste solids produced,
both of which are very important factors in determining the operability
and economics of this dual alkali approach.
The pilot plant test program involved the characterization of two different
reactor systems: a continuous stirred tank reactor (CSTR) and a reactor
system previously developed by ADL (ADL reactor). Standard operating
conditions for the ADL reactor were established in pilot plant development
work prior to this EPA program; however, data on the operation of a CSTR
in this mode were not available. A preliminary study of the CSTR, there-
fore, was undertaken in the pilot plant to examine the effects of various
reactor parameters and selected process variables including sulfate and total
oxidizable sulfur (TOS) concentrations in the reactor feed liquor, on the
performance of the CSTR; and to develop a set of suitable operating criteria.
This preliminary study consists of short-term, open-loop reactor runs using
simulated scrubber bleed and dry hydrated lime. This initial phase of
testing was then followed by a number of extended, closed-loop continuous
runs to confirm the open-loop results and to evaluate the overall process
operation using both the CSTR and the ADL reactor.
The lime used in all the pilot plant runs was a 200 mesh hydrated lime
containing 95 wt% calcium (reported as Ca(OH)2) and averaging 91 wt% avail-
able calcium hydroxide. The remaining 4 wt% calcium, consisting mainly of
CaC03 and burnt CaO, was determined to be essentially unavailable for reaction
above a pH of 8.
2. Regeneration Reactor Performance
Reactor performance in both the open-loop reactor tests and the closed-loop
continuous runs was evaluated in terms of four process criteria: calcium
utilization, effluent soluble calcium concentration, reaction of sodium
sulfate to precipitate calcium sulfate, and the dewatering properties of
the solids generated. Of particular concern in the pilot plant work were the
effects of sulfate concentration in the feed liquor and the extent of
absorbent regeneration (reactor pH) on the reactor performance.
The spent scrubber solution fed to the regeneration system, whether simulated
or actual, normally contained 0.4 - 0.55M active sodium (0.35 - 0.50M TOS)
at a pH of 5.8 - 6.1. In some tests, though, active sodium concentrations
were reduced to as low as 0.25M Na+ (0.2M TOS). Sulfate concentrations in
the spent scrubber liquor were varied from 0.05M to l.OM SO 4".
586
-------
As previously indicated, all of the open-loop reactor testing as well as
much of the closed-loop continuous operation was devoted to the develop-
ment and characterization of the CSTR. However, in many respects the
performance of the ADL reactor and the CSTR were found to be equivalent.
a. Calcium Utilization Calcium utilizations in runs with both reactor
systems were high — generally greater than 95% of the available Ca(OH)2 in
the lime feed. There appeared to be no effect of reactor configuration on
the utilization of calcium; nor was there any apparent effect of sulfate
concentration on calcium utilization over a range up to l.OM SOi^. There
was, however, a slight dependence of reactor operating pH on utilization.
In the pH range 7.4 - 11.0, calcium utilizations generally approached, and
sometimes exceeded, 100% of the available Ca(OH)2 for reactor holdup times
from 25 to 60 minutes. Utilizations in excess of 100% can be attributed to
a number of factors including variations in the lime feed composition and
analytical errors; as well as the reaction of a portion of the calcium
value normally considered to be unavailable at high pH levels. The average
utilization observed in all runs below a pH of 11.5 was 99% + 4% of the
available Ca(OH)2, which is equivalent to approximately 95% of the total
calcium in the lime, as determined from the analysis of the reactor effluent
solids.
At pH levels of 11.5 - 12.5, the average utilization dropped off slightly
to 96 + 1% of the available Ca(OH)2 (or 91-92% of the total calcium value).
During continuous closed-loop operation at a pH of 11 - 11.5 and with a
reactor holdup of 25 - 30 minutes, the thickener overflow and filtrate
were observed to have a slightly higher pH level than the reactor effluent.
This increase in pH, on the order of 0.2 pH units, indicated that the reactor
effluent at the shorter residence times was not quite at equilibrium, and
that some additional reaction was probably occurring in the thickener.
b. Soluble Effluent Calcium Levels The level of soluble calcium
in the reactor effluent slurries for both reactor systems generally ranged
from 30 - 90 ppm Ca~"~. In only one instance did the calcium level rise
above 100 ppm, to 120 ppm. Although these calcium levels were quite low,
there was apparently some supersaturation with respect to CaSOs- 1/2H20
at holdup times of both 30 and 60 minutes. Figure III-5 shows the product
of the calcium and sulfite concentrations measured in the effluent solution
compared with the apparent solubility product curve predicted by the
method of Kusik and Meissner.
The general location of these experimental solubility product data with
respect to the predicted curve is consistent with data generated in both
the ADL and EPA laboratory work (see Section VI. A.). Figure III-5 suggests
supersaturation, however, this was not confirmed by the long-term closed-
loop data. In the closed-loop continuous runs the soluble calcium levels in
the thickener overflow tended to be slightly lower than or roughly equivalent
to the soluble calcium levels in the reactor effluent. Since the thickener
generally had a holdup time of about eight hours, it would be expected that
there would be a noticeable decrease in soluble calcium levels in the over-
flow if the reactor effluent slurry were supersaturated to the degree indicated
in Figure III-5.
587
-------
CO
o
GO
OS
U
ro
O
CO
to
U
o
£
>
c
0}
a
a
10
-3
10
-4
10
-5
10
-6
10
,-7
c
d>
Legend
O30 Minute Reactor Holdup
50-60 Minute Reactor Holdup
1.0
2.0
4.0
3.0
H, Ionic Strength
Figure III-5 Relation of Observed Apparent Solubility
Products to Saturation Values For CaSO,
5.0
6.0
588
-------
There are two other explanations for this apparent supersaturation. First,
the predicted curve may be slightly low. This curve was calculated based upon
literature data for low ionic strengths using the method formulated by Kusik
and Meissner for calculating activities, at high ionic strengths. The method
is only good to about 10 - 20% and any error in the data base would be
translated directly into the predicted curve.
Second, the CaS03' 1/2H20 solubility product curve may not apply, since the
calcium sulfite/sulfate solids were apparently present as some type of mixed
crystal as discussed in Section III. B. However, in no case did supersaturation
to the extent it may have been present, cause any precipitation or scaling
problems.
c. Regeneration of Active Sodium from Na?SOu A careful analysis of
the solids generated in the reactor showed that a significant amount of
calcium sulfate precipitation was occurring over a wide range of soluble
sulfate concentration. The amount of calcium sulfate precipitated was
found to be strongly related to the soluble sulfate concentration. This
sulfate dependence is shown in Figure III-5, in which observed mol ratios
of (CaSO^/CaSOx) in the reactor effluent solids, corrected for the small
amount of occluded or entrained Na2SOi4, are plotted against sulfate
concentration for TOS levels in the range 0.4 - 0.47M. The general relation-
ship indicated by the dashed line agrees quite closely with that developed
from the laboratory program for the same TOS range. The data suggests that
there is a threshold level of ~0.05M sulfate at this TOS concentration
above which precipitation of calcium sulfate occurs. This has not been
confirmed and may simply be due to experimental and analytical error.
The results shown in Figure III-6 include data from runs with both the ADL
reactor and the CSTR for reactor holdup times of 30, 45, and 60 minutes.
In general, no significant effect of either reactor configuration or holdup
time on the relative amount of sulfate precipitated could be discerned.
The results of a few runs made over a two-fold range of inlet TOS concen-
tration, though, do suggest a dependence of sulfate precipitation on TOS.
(CaSO^/CaSOx) ratios obtained in three runs at 0.74 - 0.8M S04=, with TOS
concentrations of 0.2, 0.4 and 0.52M are shown in Figure III-7. Although
there is considerable data scatter, the observed decrease in sulfate precip-
itation for the two-fold plus increase in TOS is still significant. The dashed
line shown in Figure III-7 is not meant to imply a straight line relationship
between TOS and (CaSOi,/CaSCx); rather it is merely intended to be an indication
of the trend in the data. A more definitive determination of the actual
relationship between calcium sulfate precipitation and TOS concentration
will require further study.
Laboratory studies of the stability of the CaSO^ precipitated showed that
significant redissolution of CaSO^ solids (up to 30% of the CaS04 formed)
may occur by continual intimate contacting of the solids with liquor
containing appreciable levels of sulfite. It might be expected then that
CaSOi, redissolution together with additional CaS03 precipitation will occur
in the dewatering system. No such redissolution could be substantiated in
589
-------
Ln
O
0.20
0.16
0.12
0)
CC
to
o 0.08
"o
O
to
ro
O
0.04
Conditions
[TOS1 =0.40-0.47M
Reactor pH = 7.2-11.5
D OX
n /
O
/
/
O
/D
./
/
/
.
/
/
Legend
Reactor
D CSTR
0.5 1.0
[S04=],M
Figure III-6 (CaSO^/CaSOx) Ratio in Solids as a Function of Sulfate Concentration
1.5
-------
Ln
v£>
0.20
0.16
0.12
c
o>
o
CO
!T
c
X
8
"o
Tj-
O
to
0
J. 0.04
0.08
Conditions
(S0
= 0.74-0.8M
Reactor pH = 1 1 .5-1 2.5
D
(0.8M S04"
Ratio = .213-.215[TOS]
0.1
0,2
o
(0.74M S04=)
Legend
O ADL Reactor
acsiR
0.4
(0.75MS04
0.3
[TOS], M
Figure 111-7 (CaSO4/CaSOx) Ratio in Solids as a Function of TOS Concentration
0.5
0.6
-------
the pilot plant operations based upon a comparison of solids composition
in the reactor effluent with that of the filter cake. The settler does
not provide particularly good solids/liquid mixing as compared to holding
the slurry in a well-mixed state for three hours, so the absence of redis-
solution is reasonable.
The amount of calcium sulfate precipitated in the system can be translated
directly into the tolerable levels of sulfite oxidation in the process.
In the range of sulfate concentration up to l.OM S0it~-and active sodium
levels of 0.2 - 0.4M Na , it is possible to sustain an overall system
oxidation rate equivalent to as much as 15 - 20% of the S02 removed with-
out the need for a separate sulfate purge. The data suggest that the
system could be operated at even higher sulfate concentrations with corres-
pondingly higher sulfate precipitation capability. Such an approach of
allowing sulfate levels to rise until sulfate precipitation balances sulfate
formation however, will ultimately be limited by the solubilities of the
sodium salts. Furthermore, precipitated calcium solids properties tend to
deteriorate with increasing sulfate concentrations (as will be discussed).
Therefore, the concentrated lime mode designed without special provisions for
separate sulfate removal appears to be limited to applications where
overall system oxidation is on the order of 20-25% Or less of the S02 removed.
d. Dewatering Properties of Solids Although the performance of the
ADL reactor and the CSTR were found to be essentially equivalent in terms
of both calcium utilization and the regeneration of active sodium from soluble
sulfate, there was a noticeable difference in the quality of the solids
generated. The ADL reactor is apparently capable of successfully operating
over a wider range of system conditions than the CSTR, at least with respect
to the production of waste solids with acceptable settling and filtration
properties.
CSTR Solids Properties The effects of reactor holdup and operating
pH on the dewatering properties of the solids generated in the CSTR were
initially studied in the preliminary open-loop CSTR reactor tests. Table
III-2 gives the results of two sets of CSTR tests at 0.6M sulfate. The
quality of the solids produced at a pH of 7.5 was good at both 30 minute and
60 minute reactor holdup times. As pH was increased to 12.0, the quality of
the solids deteriorated as shown by the marked decrease in the solids settling
rates and in the densities (wt% solids) of the settled solids. This pH
effect was much more pronounced for the 60 minute holdup than for the 30
minute holdup; however, in neither case was the loss of solids properties
irreversible. Reducing pH back to a low level resulted in a return to good
solids. In light of the greater sensitivity of the 60 minute reactor hold-
up, the shorter 30 minute holdup provides a more reliable design basis for
a CSTR.
The effect of sulfate concentration on solids properties over a range of
pH for the 30 minute CSTR is summarized in Table III-3. Increasing sulfate
levels results in poorer solids, an effect which is aggravated by high pH.
The observed decrease in settling rates is indicative of smaller particle
592
-------
Table 111-2 EFFECT OF CSTR pH AND KOLD-UP TIME ON SOLIDS PROPERTIES — CONCENTRATED LIME MOLE
General Reactor Conditions:
React ?r,t Fer-ei - Dry Ca(OH)2
Reactor Temperature = 105 - 120°F
Agitation Rate = 200 - 250 rpm
Liquor Feed Composition - [ICS] - 0.46M
[S04~] = 0.6H
DH = 5.8
Reactor
Hold-up(min)
60
REACTOR EFFLUENT
pH Wt% Susp'd Solids
7.2-7.4 2.5
12.0 5.0
SOLIDS SETTLING PROPERTIES
Initial Settling Wt% Insolubles in
Rate(ft/min) Settled Solids(2-3 hrs.)
0.16 - 0.23
0.05
26 - 36
9
30
7.5
11.9
2.8
4.7
0.20 - 0.25
0.15 - 0.20
24
15
-------
Table III-3 EFFECTS ON SULFATE CONCENTRATION AND pil ON SETTLING PROPERTIES
OF SOLIDS PRODUCED ON A CSTR WITH A 30-MINUTE HOLDUP
Ul
V£>
General Reactor Conditions: Reactor Holdup = 30 minutes
Reactant Feed - Dry Ca(OH)2
Reactor Temperature = 105 - 120°F
Agitation Rate = 200 - 250 rpm
Liquor Feed Composition - [TOSJ = 0.45-0.55M
pH = 5.8
REACTOR EFFLUENT
SOLIDS SETTLING PROPERTIES
Feed [SOiT] , M pH
0.05 7.5
0.5 7.5-8.5
0.6 7.5
11.9
0.75 12.0
12.5
0.8 11-11.5
0.9-1.0a 8-10
11.5-12.0
Approximate
Wt% Susp'd Solids
2.7
2.6
2.8
4.7
4.4
4.9
1.5
2.3
2.3
Initial Settling
Rate, ft/min
0.25 - 0.30
0.20 - 0.25
0.20 - 0.25
0.15 - 0.20
0.05 - 0.10
0.05 - 0.10
0.10 - 0.15
<0.05
<0.05
Approximate Wt% Insolubles in
Settled Solids (2-3hrs.)
45
25
25
15
15 - 20
10 - 15
10
5-15
5
(a) 45 minute reactor holdup
-------
sizes; and the lower densities of the partially compacted solids implies a
change in the form of the crystal agglomerates generated. Electron microscopy
and optical studies of the solids produced confirms the difference in the
precipitated crystals. The poorer settling solids are generally made up of
smaller particle sizes with more needle-like crystal structures.
These differences in the settling properties are also reflected in the
filterability of the solids. In general, low settling rates and low
densities of settled solids are consistent with low solids content in the
filter cake and poor cake wash efficiencies.
Comparative Solids Properties Table III-4 compares the settling
and filtration characteristics for solids produced in the concentrated lime
mode using both the ADL reactor and the CSTR. These results show that the
density of the compacted solids, rather than the initial settling rate, is
probably a better indication of the filterability and washability of the
solids. This is reasonable since the density of the settled solids is
simply a measure of how well particles compact under the force of gravity.
Solids which do not compact well will not dewater easily, nor will they
be easily washed.
As evidenced by the data presented in Table III-4, both reactors are capable
of producing solids with good dewatering properties over a reasonable range
of system operating conditions. In both systems, the solids properties
deteriorate with increasing sulfate concentration and pH; however, the CSTR
is more sensitive to these effects than the ADL reactor. Solids generated
in the CSTR at high pH and sulfate concentration do not settle or filter
as well as those obtained in the ADL reactor under equivalent conditions.
This difference is shown by the plot of filter cake solids content vs.
sulfate concentration in Figure III-8. The solid line represents filter data
for the system operating with the ADL reactor which were obtained in work
prior to this EPA program.
Washability of Solids The lowest levels of solubles obtained
under continuous operation were 1.5 - 3.5 wt% solubles (dry basis). This
was achieved in a run using slightly more than three displacement washes
sprayed onto the cake using two full cone nozzles arranged in series. As
indicated in Table III-4, the reduction in solubles is a function not only
of the number of displacement washes but also of the quality of the solids.
As the quality deteriorates, the moisture content of the cake increases and
the wash efficiency (and percent reduction in solubles) decreases.
The practical minimum solubles levels which can be achieved was determined
in a special set of tests conducted with the pilot plant filter. The solids
used were produced in the ADL reactor at a 0.6M sulfate concentration. The
results of these tests demonstrated that the cake can be realistically
washed to a level of about 1.5 wt% soluble solids (dry basis), if desired.
The reduction in soluble solids attained as a function of the number of
displacement washes used is shown in Figure II1-9. Here a reduction of 90%
in the soluble solids is roughly equivalent to 1.2 wt% soluble solids (dry
basis) for a cake containing 50 wt% insoluble solids. This level of sodium
595
-------
COMPARATIVE WASTE SOLIDS PROPERTIES — CONCENTRATED LLMt MOUE
VD
Reactor Systes)
Reactor Effluent
Conditions
Settling Properties
ADL --
CSTR--
50 Bin
60 din
30 oin
25 mln
45 mln
0.?
0.8
0.5
o.a
0.9-1.0
£" H.iU',
9-11
11.5-12
7.5-8.5
11-11.5
6-9
Ct/-in)
11.16
0.15
0.22
0.12
0.03
25
20
25
10
5
FtUcrabllity
Filter Cake Washability
30
wt
Flit
2 Insolublcs in
er Cak« (drv basis)
3-6
6
1.5-J.5
14
31
N'o. of Displacement
Washer," (Spray Typn)b
1.5-dFC)
1.0— (1FC)
3.1— (2FC)
2.3~(1FC)
0.4— (1FC)
% Reduction
in Solubles
55-75
60-65
65-85
40-60
5-20
Wash ,
Efficiency «)
90-120
120-130
80-100
55-85
15-6S
a So. of Displacement
Occluded Water in Cake
b Type of Spray Wash: (1FCJ-- Single, full-cone spray
(2FC)-- Two, full-cone sprays In series
c Wash Efficiency is calculated by dividing the observed reduction in solubles
by the reduction expected assuming dilution of the cake liquor, for cxaaplc,
In the first test shown, the expected reduction in solubles based upon dilution
Is 1 - (l/2.?)or 602 and the observed reduction Is 55 - 752. Therefore, the
efficiency is 90-1201.
-------
80
60
CJ
c
Legend
Correlation of Continuous Run Data
~~~~ for ADL Reactor (includes prior work)
O ADL Reactor (EPA Program). pH = 11.0-12.0
Q CSTR,pH = 7.5-11.5
o
CO
\D
D
O
40
20
0.4
0.8
1.2
1.6
2.0
[so.
, M
Figure 111-8 Average Wt % Insolubles in Filter Cake
Versus [SO4~] - Concentrated Lime Mode
-------
CO
100
90
80
70
.£ 60
o
OT 50
•g
QC
c
O>
u
V
o.
40
30
20
10
O
Test Conditions
Liquor: 0.6M Na2SO4
0.16M Na2SO3
0.07M NaOH
11% Solubles (dry basis) in Unwashed Cake
Containing 50% Moisture
_L
J_
2 34 56 78
Number of Displacement Washes {based upon water content of cake discharged)
Figure 111-9 Reduction of Solubles in Filter Cake Versus Wash Ratio
(Two Sprays in Series)
10
-------
salts (almost all of which are sodium sulfate) would leave the system at
any reasonable washing rate. This confirms similar data obtained in the
laboratory. There is apparently a low but significant quantity of soluble
salts, on the order of 1 - 1.5 wt% solubles, occluded in the cake that is not
easily leached.
The maximum cake washing capability for a medium or high sulfur coal
application would be roughly equivalent to two to three displacement washes
(0.4 gpm/Mw for a cake with 50 wt% solids). The number of displacement washes
will be limited by the amount of makeup water available for washing as
determined by the process water balance and possibly by the operation of the
filter. Assuming cake wash behavior similar to that seen in the ADL
pilot plant for two serial spray washes, 75 - 85% reduction in soluble solids
concentration in the cake liquor can be expected for this level of washing.
With a 0.6 - 0.7M sulfate concentration in the system liquor, the resultant
solubles in the cake would then be 2 - 3% (dry cake basis) for a cake
containing 50 wt% insoluble solids.
The exact amount of solubles will depend upon a large number of factors
including dissolved solids levels, the type of wash used (spray configuration),
the operating conditions for the filter, and especially the solids properties.
The degree of washing that is desirable will also depend in large part on the
presence of impurities — the rate at which they enter the system and the
tolerable steady state impurity levels within the system.
e. Reactor System Oxidation A significant amount of oxidation was
found to occur in both reactor systems operated open to the atmosphere. In
the initial open- and closed-loop continuous reactor tests, oxidation rates
ranged from 0.01 to 0.30 mols TOS/liter-min (equivalent to 10 - 280 ppm of
S02 in the pilot plant system). The rate of oxidation was found to be
essentially independent of sulfite concentration in this concentrated regime;
but it was strongly dependent upon the degree of agitation and the exposed
surface area of the vessel indicating that the oxidation is primarily a
function of the amount of gas-liquid contacting.
Subsequent batch and continuous runs showed that the oxidation in the reactor
system could be minimized by controlling agitation and modifying the
design of the vessel; and that it can be virtually eliminated by sealing
the reactor system. Based upon these results, it would normally be expected
that oxidation rates in an unsealed reactor system would be in a range
equivalent to 50 ppm S02.
3. Summary of Overall System Operation
Four closed-loop continuous runs were made in the pilot plant in the concen-
trated active sodium mode using lime for absorbent regeneration. A range of
different conditions were examined including S02 inlet concentration (2600 -
2800 ppm and 850 ppm), sulfate concentration, reactor system design (ADL
vs. CSTR) and the use of both Na^O^ and Na2C03 for sodium makeup. For
the most part, these runs were intended to confirm and extend open-loop
data and to test the system operation under a variety of conditions.
599
-------
In each run, the system was primed to predetermined concentrations and then
operated under a specified set of conditions tor at least three days (four
to eight system holdup times). Although steady state operation was not
achieved in all runs, sufficient data were collected to characterize the
operation of each section for the given set of conditions.
Under all conditions tested, the system operation was found to be quite
stable. No difficulties were encountered in attaining S02 removal
efficiencies greater than 90%. Lime utilizations exceeded 95% throughout
all runs and soluble calcium levels in the regenerated absorbent liquor
returning to the scrubber ware consistently less than 100 ppm. No scaling
occurred in the system during these runs and no increase in tray pressure
drop was observed.
The amount of oxidation experienced in the system, the properties of the
solids generated (including the cake properties) and the rate of regeneration
of active sodium from sodium sulfate differed in each run depending upon the
particular set of conditions being tested. However, two of the runs do
provide a basis for analyzing the operation of systems incorporating the
CSTR and the ADL reactor. The primary system operating conditions and key
operating results for these runs are summarized in Table III-5.
Run 404 was the last continuous run made in the concentrated lime mode. In
this run, using a new CSTR reactor, oxidation was reduced in the reactor and
maximum cake wash was used (three displacement washes). Run 402, using the
ADL system, was conducted prior to reactor redesign for oxidation control and
incorporated only minimum washing of the filter cake (one displacement wash).
As a consequence, reactor oxidation and cake solubles content were higher
in the earlier run 402 than 404.
It should be noted that not all process sections were operating under optimal
conditions in these runs, although overall they represented system operating
levels which might be achieved under steady state conditions.
Both runs 402 and 404 were made at high S0£ inlet levels and active sodium
concentrations of about 0.45. In run 402, incorporating the ADL reactor,
there was a slight decrease in active sodium throughout the run due to the
low cake wash rate and the relatively high oxidation rate in the reactor
system. Na2S(\ was also used to make up sodium value even though the
sulfate formation rate within the system was slightly higher than the
sulfate precipitation.
In run 404, the CSTR run, the system was operated at a much lower sulfate
level than run 402 to ensure the production of solids with good dewatering
properties. Although there was a lower rate of regeneration of ^£50^ to active
sodium due to the lower sulfate level, the use of a reactor vessel with a
smaller surface area than that used in run 402, which minimized oxidation in
the reactor, coupled with the slightly higher S02 inlet level, enabled the
system to essentially sustain a rate of sulfate precipitation equivalent to
the sulfate formation rate.
600
-------
Table III-5 SUMMARY OF CLOSED-LOOP RUNS 402. AND 404 — CONCENTRATED LIME MODE
Test Nc .
Operating Conditions
Inlet Gas SG'2 Level , ppm
Scrubber Bleed: pH
[TCS],M
Reactor System: Type
Sodium Makeup Form
402 404
2600 2800
5.8-5.9 5.8-6.0
0.45 0.4
0.85-0.8 0.5
ADL CSTR-30 min. holdup
11.5-12 7.5-8.5
Na2S04 Na2C03
Key Operating Results
Scrubber Feed Stoichioir.etry 1.1
SO^ Removal Efficiency (% of inlet S02) 91%
Calcium Utilization (/<,' of available'. Ca(OH)2) 96/i
Calcium Feed Ratios:
moIs Ca(OH)2mol/AS02 1.10
[Ca++] in Scrubber Feed, ppm 60
System Oxidation Rates:
Scrubber Circuit (equiv. ppn of S02) 2CC
Rftgeneration Reactor (equiv. ppm of SOz) 240
Sulfate Precipitation (equiv. ppm of S02) 330
Filter Cake Composition:
wt% insoluble solids 45wtZ
wt% soluble solids (avg.) 6wt%
No. of Displacement washes 1
1.1-1.2
93-95%
98+Z
0.92(0.96 calcium)
30-70
210
80
230
45wt%
2wt%
3
601
-------
Figures 111-10 and III-ll summarize the key operational characteristics
for runs 402 and 404 in terms of schematic block process diagrams. These
diagrams express the various system inputs and outputs in terms of the
S02 absorbed. For each of the three process sections (gas scrubbing,
regeneration reactor, and dewatering) the observed rates of sulfite oxida-
tion and/or regeneration of active sodium from sodium sulfate are included,
also expressed as a function of S02 absorption. Oxidation (and regeneration)
rates in the scrubber and reactor systems are determined from material
balances around each of these process sections individually. Oxidation
rates in the dewatering system, on the other hand, were extracted from
overall system material balances over a period of at least 24 hours,
taking into account changes in inventory within the system. The direct
determination of sulfite oxidation rates across the dewatering system were
difficult to accurately estimate since flows were not constant, and wash
water and vacuum pump water were being continually added to the system
through the filter. The best direct estimates of oxidation across the
filter are 0.02 - 0.04 gmols TOS/min; and across the entire dewatering
system is about 0.05 gmols TOS/min. This overall rate is roughly equivalent
to the oxidation of 50 ppm S02 in the scrubber and agrees closely with
the rates calculated from overall balances.
The composition of the cakes shown in Figures 111-10 and III-ll are given
in terms of wt% on a total dry basis. The value shown for CaSO^ content
in the cake, therefore, differs slightly from the rate of precipitation
shown in the regeneration system block.
As can be seen in the composite for run 402, sodium makeup is roughly
equivalent to sodium losses in the cake; however, the total sulfate
precipitated and lost in the solubles fraction of the cake amounts to only
about 20% of the S02 absorbed while the overall system oxidation plus
sulfate addition totaled 27% of the S02 absorbed. There was, therefore,
a slow decrease in active sodium in the system and a corresponding
increase in sulfate. This would have'continued until the level of sulfate rose
to a point where it would effect high enough sulfate precipitation rates
to provide a balance in the system.
It should be noted that in run 402, the total calcium feed to the system
amounted to 110% of the S02 absorbed. This feed rate was simply high,
and does not reflect a low utilization of lime. This amount of extra
lime is not needed for the reaction with the sulfur added to the system
as Na2SOi,, since this sulfur would normally leave in the cake associated
with sodium that is being discharged.
Composite Diagram of Pilot Plant Operations Figure 111-12 sum-
marizes the operational characteristics of the various pilot plant process
sections in the concentrated lime mode as determined in the various open-
and closed-loop runs. This block diagram is analogous to those for run
402 and 404, and shows the expected steady state operation of the dual
alkali system for a 2500 ppm inlet S02 loading with a total system oxida-
tion rate of 13%. The composite is based upon operation with the ADL
reactor. For a comparable operation using the CSTR, the solids content
602
-------
t 250 ppm S02
2600 ppm SO2
Scrubber System
91-92%S02 Removal
Oxidation = 8-9% of ASO2
Os
o
Na2S04
6.5% of ASO2
pH = 5.8-6.0
*a+) active = '45M
(SO4~) = 0.8M
Oewatering System
Oxidation = 2.5% of £S02
T
Regeneration System
Oxidation = 9-10% of ASO2
Sulfate Precipitation =
13-14% of ASO2
Active Line
100% of ASO2
pH = 11.5-12.0
One Displacement Wash
H2O
Cake (45% Insoluble Solids)
Average Cake Composition (wt% dry basis):
CaSOij = 13%
CaSO3 = 73%
Other Insolubles = 8%
Na2SO4= 5%
Na2S03= 1%
Figure 111-10 Continuous Run 402 — ADL Reactor (Na-j
Makeup)
-------
200 ppm SO2
2800 ppm S02
Scrubber System
90-95% S02 Removed
Oxidation = 8.0% of ASO2
ex-
es
Feed Stoich = 1.2
2.5% of A SO2
pH = 5.8-6.1
[Na+]
active
= .45M
[SO4=] =0.5M
Regeneration System
Oxidation = 3% of AS02
Sulfate Precipitation =
9% of ASO0
Active Line
96% of A SO-,
Oewatering System
Oxidation « 2% of A S02
pH = 7.5-8.5
H2O
Three Displacement Wash
Cake (45% insoluble solids!
Average Cake Composition (wt% dry basis):
CaS04 = 10%
CaS03 = 83%
Other Insolubles - 5%
Na2SO4 = 1.5%
Na2SO3 = 0.5%
Figure 111-11 Continuous Run 404-CSTR Reactor (Na2CO3 Makeup)
-------
k 200 ppm S02
2500 ppm SO,
Scrubber System
90-95% SO2 Removal
Oxidation = 8% of A SO2
Na2C03
2% of A SO-
pH = 5.9-6.1
!Na+) active = °
[S04~] = 0.6M
Regeneration System
Oxidation (unsealed) =
3% of A SO2
Sulfate Precipitation =
11.5% of AS02
Active Lime
100%ofAS00
Dewatering System
Oxidation = 2% of A S02
H2O
Three Displacement Washes
Cake (50-55% in soluble solids)
Average Cake Composition (wt % dry basis):
CaSO4 =12%
CaSOg =81%
Other Insoluble* = 5%
Na2S04 + Na2S03 = 2%
Figure 111-12 Composite Pilot Plant Operations For 13% Overall Oxidation in System
-------
of the cake would be expected to be 40 - 45% insoluble solids. These solids
contents are estimated from pilot plant data and for larger scale systems
where the dewatering equipment can be operated with better control, filter
cakes with up to 5 - 10% higher solids levels would be expected.
4. Future Pilot Plant Operations
Future pilot plant work in the concentrated lime mode will involve a number
of four to five week closed-loop continuous runs using the CSTR and/or
ADL reactor system. These runs will be aimed at demonstrating the overall
system performance and specific emphasis placed on process reliability and
stability and minimizing soluble solids losses in the cake (within practical
limitations of wash capability).
606
-------
IV. SULFURIC ACID TREATMENT — CONCENTRATED MODE
A. LABORATORY RESULTS
1. Introduction
The inherent sulfate precipitation which was observed during lime
regeneration of concentrated sodium sulf ite/bisulf ite/sulfate solutions
shows promise of being able to control sodium sulfate at manageable levels
when system oxidation rates are in the range of 10 - 20%. An alternate
method, capable of dealing with significantly higher oxidation rates,
involves the acidification of the process liquor with sulfuric acid in the
presence of solid calcium sulfite to precipitate dissolved sulfate as gypsum.
The overall reaction involved in this treatment process is,
Na2S04 + K2SQk + 2 CaS03'isH20, . 4- 3H20 -» 2 CaS04-2H20, . + 2NaHS03 (11)
\ S ) \ S )
thus, although additional sulfate is introduced as sulfuric acid, both that
sulfate and the Na2SOit should be precipitated as gypsum. On the basis of
the 1:1 relationship between sulfate (sodium) to be precipitated and sulfuric
acid required, 3 reaction efficiency (theoretically 100% based on equation 11)
can be defined by expressing the moles of Na SO removed from the process
stream as a percent of the moles of H2SOtt added:
„,,.. . ra°l NaoSOn removed -, nna/
Efficiency = - / g, — 7—;— x 100%
J mol tSOi fed
The primary acid consuming reaction is the dissolution of CaS03'!-sH20 by
converting it to the soluble calcium bisulfite
H20 -*• CaS(V2H20, , + 2HS03~ + Ca4^ (12)
\s )
The dissolved calcium ion produced in this reaction can then react with the
sodium sulfate to be precipitated to form gypsum:
Ca++ + S0k= + 2H20 -»• CaSOt+-2H20, , (13)
\s )
In addition to the primary reaction discussed above, other equilibria would
likely be involved and would result in reaction efficiencies of less than
100%. At the acidic conditions required to dissolve calcium sulfite
(pH = 2 - 3) , measureable amounts of sulfurous acid and possibly bisulf ate
607
-------
ion could exist in solution due to the acid equilibrium reactions:
HS03~ + H+ ^ H2S03 (14)
S04= + H* J HSOi^- (15)
In practice, the streams to the acid treatment could contain other sulfites,
carbonates and possibly hydroxide. In such cases, sulfuric acid consumption
would be increased and reaction efficiency reduced.
The most acidic process stream normally available from which sulfate
could be precipitated — scrubber effluent — normally contains on the order
of 0.05 M Na2S03. Even well washed filter cake could contain calcium
carbonate or unreacted lime. Depending upon washing efficiency, significant
amounts of Na2S03 and NaOH could also be present. If the calcium sulfite
were obtained as a thickened slurry from the regeneration section — attractive
because of the ease of obtaining it and transferring it to the acidification
process — significantly larger amounts of NaOH and Na2S03 would be present.
The NaOH and Na2S03 will react with ^SOij according to the reactions
2 NaOH + H2S04 -> N32S04 + 2H20 (16)
2Na2S03 + E250k -*• Na2SOt, + 2NaHS03 (17)
consuming H^Oi^ and producing additional soluble sulfate in solution. Any
lime or limestone present in the calcium sulfite feed will react with sul-
furic acid
Ca(OH)2(s) + ^SO^ -»• CaSOit • 2H20(S) (18)
CaC03(s) + H2SOi, + H20 ->- CaSOit • 2H20(S) + C02t (19)
These reactions will consume additional acid but will not add additional
sulfate to the solution because of the accompanying precipitation of gypsum.
As a point of departure for the subsequent laboratory studies, a very
approximate estimate was made of the pH to which the solution would need to
be taken for the regeneration reaction to proceed, based upon very dilute
solution solubility product and acid dissociation constants. For the
precipitation of gypsum to occur, its solubility product in water
K = [Ca++] [30^=] i. 2 x lO"1*
sp
would need to be exceeded. If [S0ij] were l.OM initially, CCa4"*"] would
need to exceed 2 x 10"1* M for precipitation to occur. For the precipitation
of gypsum to continue until CSO^] = 0.1 M, the calcium ion concentration in
solution would need to rise further to about 2 x 10~3 M.
To achieve calcium levels of this magnitude in the presence of the very
insoluble calcium sulfite, CSO^D would need to be held to a very low level
by converting it to bisulfite ion. Considering the solubility product of
608
-------
calcium sulfite,
K = CCa**: CSOID = 10-7
sp
a calcium ion concentration of about 10"3 would require that the level of
sulfite ion be held to about lO"*4 M. Now, to achieve that sulfite ion
level in a solution containing sulfite/bisulfite at about 0.5 M, the second
ionization constant of sulfurous acid
K _ [SOI] M _ in-7
K ~ [HsofT ~
would indicate that [H+U = 5 x 10~3, or a pH in the vicinity of 3, would
be required.
2. Experimental Results and Discussion
Prior to studying the acidification reaction in the laboratory in a
continuous reactor, a number of batch experiments (in which l^SOt, was
added to Na2SOi,./CaS03 slurries) were conducted in a sealed reactor to
better define the operating pH range and, more importantly, to establish
the SC>2 partial pressure levels generated within the reactor. The Na2S04/CaS03
slurries were charged into the reactor, and the headspace above the liquid
was purged briefly with SC>2 to eliminate air before sealing the reactor.
Sulfuric acid was then introduced into the sealed system from a pressure-
equalized addition funnel while monitoring the pressure within the reactor.
The batch experiments indicated that the operating pH would, in fact, be
in the range of 2.5 - 3.0. Throughout the course of sulfuric acid addition,
the pressure within the reactor indicated that the partial pressure of S02
was less than 1 atmosphere, and a sealed reactor capable of withstanding a
positive internal pressure would not be required for the reaction. The
reactor could be operated at ambient pressure vented to the atmosphere as
long as the reactor vent tube was of a sufficiently narrow bore to minimize
mixing between the internal headspace gas and the outside air.
On the basis of the background information obtained from batch experiments,
the effects of a number of experimental variables on the regeneration reaction
were studied in the laboratory using the CSTR described in Section II-A.
Since it was the only material available in quantity at that time, the calcium
sulfite used in these studies was unwashed calcium sulfite filter cake
produced in an earlier ADL pilot plant run in which concentrated sodium
bisulfite/sulfite/sulfate solutions were regenerated with lime. This material
contained relatively large amounts of calcium carbonate and calcium sulfate
along with lesser amounts of sodium salts as the following composition
indicates:
609
-------
Substance Mo I Percent
CaS03 69.2
CaC03 15.4
10.5
3.3
NaOH 1.0
Na2S03 0.5
This composition is based upon chemical analyses and material balances for
the pilot plant runs in which the material was produced. This filter cake
was slurried with a simulated acidic scrubber effluent solution containing,
in all cases, Na2S03, 0.05 M; NaHS03, 0.40 M; and Na2SOif at either 0.75 M
or 0.37 M. This slurry comprised one stream to the CSTR and either 4.5 M
or 9.0 M H2SO^ was fed as the other reactant.
Both phases of the reactor effluent slurry were analyzed in detail. In the
liquid phase, in addition to measuring total soluble calcium ion and sulfate
ion, measurements of Total Oxidizable Sulfur [TOSH and Total Acidity [H4"]]
were made by titration. The concentration of TOS is the total concentra-
tion of all sulfur (IV) species present, and the total acidity CH"1"] ,_ ,
is the sum of all acidic protons (titratable to pH = 9) in the solution.
These relationships may be expressed as follows:
[TOS] = [H2so3] + [Hso3~3 + Cso3 ]
[H+]tot = [HS03~] + 2[H2S03] + [HSO^-] + [H+]
By subtraction, the "excess acidity", ([H+3tot - [TOS]) equals
- [TOS] = [H2S03D + [HSOit'H + [H+] - [S03=]
However, under the range of conditions over which the reaction was studied
(see Table IV-1) ,_H2S03 and HS03~ are the predominant species present.
At pH 2 - 3, [S03~] is virtually zero and [H+] is less than 0.01 M.
The dilute solution value for the second ionization of sulfuric acid, pK2,
is about 2; however, dissociation of HSO^ increases as ionic strength is
increased. In the CSTR experiments performed, the ionic strength was in
two general regions — 2.25 (experiments 6-7) and 1.4 (experiments 9 - 10).
Marshall and Jones6 report values of pK2 for H2SOi+ of 0.82 and 1.1, respec-
tively, at those ionic strengths. At the conditions studied here, then,
HSOij" exists only in small concentrations — the concentration of HSO^" is
at most 2% of the H2S03 concentration in the solution. The two measurements
[H+]tot and [TOS], allow one to determine the amounts of H2S03 and HS03~
present in the reaction effluent.
610
-------
The results obtained from the CSTR studies are shown in Table IV-1. In
the three series-6 experiments, the regeneration of 0.75M NaoSO^ solution
was studied as a function of pH ranging from about 2.35 to 3.15 with re-
actor residence time and temperature held constant. In series 7, tempera-
ture and residence time were varied.' In the final three experiments, 0.37M
Na2SO|+ was studied as a function of pH.
The effect on the regeneration reaction of reducing the pH from 3.15 to
2.9 was an increased TOS concentration as additional calcium sulfite dis-
solved at the lower pH. In turn, the concentration of SO^" decreased as
the soluble calcium which was produced reacted with SO^" to precipitate
gypsum. As indicated by the lower S0l4~, the "effective Na?S04 removal"
from the simulated scrubber stream (ANa2SOil/Na2SOi+) was higher at the
lower pH (this ratio is corrected for all sulfate removed which entered
as H^OL, or in the calcium sulfite filter cake) .
As the pH was decreased to 2.9, the concentration of TOS increased by
0.1M but the total acidity increased by 0.17M as a result of the higher
concentration of H2SC>3 present at the lower pH. The ratio of H2S03/HS03~
had risen from 0.09 to 0.15 as shown in the table. The formation of
H2S03 consumed t^SO^ and should have reduced the efficiency of H2SO[t
utilization. Thus the increased effective sulfate removal at pH 2.9
was compensated for by the decrease in efficiency of H^O^ utilization
to produce an essentially constant overall "observed efficiency" of
H2SOit utilization — ANa2SOtt/total H2SOtt fed.
Since the calcium sulfite cake which was used contained a large amount
of CaC03 as well as other impurities which would consume H2SOi+, the
effects of the non-calcium sulfite species were taken into account and
a "corrected efficiency", which should have been realized if pure cal-
cium sulfite had been fed, was computed. As shown in the table, K^SO^
utilization efficiency would have been in the range of 0.6 to 0.7.
In experiment 6-c , the pH was further lowered to 2.35; but, upon analysis
of the reaction products, it was found that no solid calcium sulfite re-
mained in the effluent slurry. The results of this experiment show the
important consequence of operating "calcium sulfite limited" (less than
the stoichiometric amount of calcium sulfite) ; both the effective Na2SOi+
regeneration and H^O^ utilization dropped significantly. In experi-
ments 6-2 and 6-b , about 20% and 10%, respectively, of the calcium sul-
fite fed remained in the reaction solids.
Decreasing the reactor residence time to 15 minutes and raising the re-
action temperature from 34 to 51°C (series 7) did not produce significant
changes in effective sulfate regeneration or reaction efficiency.
When the concentration of NaaSOit was reduced to 0.37M in experiment 9,
an effective NajSO^ removal of 0.66 was observed at pH = 2.8, a value
slightly higher than in the corresponding experiment at the higher sul-
fate concentration (7-a) ; however, both the observed and corrected
611
-------
Table IV-1 SUMMARY OF LABORATORY CONTINUOUS REACTOR SULFURIC ACID TREATMENT EXPERIMENTS
Experimental Variables
Effluent Concentrations .(H)
Expt.a'b
6-a
6-b
6-c
7-a
7-b
7-c
9
10-a
10-b
[Na?
0.
0.
0.
SOJ
75
75
75
0.75
0.
0.
0.
0.
0.
75
75
37
37
37
Residence Temp
Time (min) (°C)
30
30
30
15
30
15
15
15
15
33
34
34
34
34
51
33
33
33
pH
3.1
2.9
2.35
2.8
2.85
2.8
2.8
2.7
2.2
1
1
1
1
1
1
1
1
1
1
TOS]
•395
.49
.46
.525
.595
.56
• 13s
.41s
.56
iH^SO^l
1.51
1.68
1.98
1.68
1.75
1.72
1.29
1.605
1.84
0
0
0
0
0
0
0
0
0
• 355
.29
.36
.27
.25
• 25s
.14
.073s
.055
[Ca^l
0.024
0.026
0.014
0.034
0.030
0.032
0.0415
0.069
0.107
103)
8.5
7.5
5.0
9.2
7.5
8.2
5.8
5.1
5.9
Solution to be regenerated simulated scrubber effluent:
(NaHS03] = 0.4 M
(Na2S03] - 0.05 M
[Na2SOi,] = as indicated
pH - 5.4 - 5.6
Calcium sulfite feed maintained at 110 - 120/2 of stoichiornetric except:
Expt. 6-c was deficient; no CaSOs-'-^O in effluent
Expt. 10-a, 10-b fed at 200% of stoichiometric
Mole fraction of entering Na2SOi, actually removed.
(Moles Na2SOi, in - raoles soluble SOi," out)/moles HjSOi, charged.
Q
Observed efficiency corrected for effect of alkaline impurities in CaS03 solids fed.
fHoSO?1 Effective
[HSO-*~! Na?SOu Removal0
0.090
0.15
0.55
0.11
0.11
0.11
0.16
0.16
0.22
0
0
0
0
0
0
0
0
0
.49
.57
.40
.61
.63
.62
.66
.80
.85
Efficiency
Observed
0.38
0.38
0.25
0.42
0.40
0.37
0.33
0.29
0.22
Corrected
0.70
0.62
0.37
0.70
0.63
0,57
0.47
0.55
0.35
-------
utilizations were significantly lower. (About 5% of the calcium sulfite
fed remained in the effluent solids.) An important factor in the decreased
l^SO^ utilization is the fact that as the incoming concentration of Na2SOtt
is decreased while the concentration of Na2S03 in the solution is held
constant, proportionately more of the acid is utilized for the acidification
of Na2SOo than for the regeneration of Na2SOi,. Lowering the pH to 2.2 pro-
duced an even larger effective Na2SO[t removal but sulfuric acid utilization
efficiency decreased even further.
Thus, as is not unexpected, removing a given amount of ^2804 per pass
through the reactor will result in increasingly poorer utilization of
sulfuric acid as the Na2SOi+ level in the incoming stream decreases. As
a consequence, for a fixed process oxidation rate, the Na2S04 regeneration
process will be increasingly inefficient as the required steady state ^250^
level in the process is lowered.
613
-------
B. PILOT PLANT RESULTS
The evaluation of the sulfuric acid slipstream treatment process involved
both open-loop reactor tests to confirm and extend the laboratory performance
data, and the use of the sulfuric acid treatment system in conjunction
with the complete dual alkali absorption/regeneration loop operating in
a concentrated lime mode. In both sets of the tests, the sulfuric acid
reactor system was operated in a similar manner. Filter cake slurry was
fed continuously to the reactor together with 20 - 30 wt% sulfuric acid
at a rate sufficient to maintain a prescribed reactor pH in the range of
2.3 to 3.3.
The filter cake slurry was made up as required on a batch basis. In the
integrated system operation, a portion of the filter cake generated during
the run was slurried with venturi recycle liquor. In the open-loop tests,
the slurry was prepared using filter cake saved from prior dual alkali
operations and either simulated venturi effluent liquor or simulated thickener
underflow liquor.
The sulfuric acid reactor consisted of a baffled 30-gallon Pfaudler kettle
vented to the atmosphere. The level in the reactor was adjusted to maintain
approximately a 30-minute reactor holdup time. Effluent from the reactor
was sent to a 6-inch solid bowl centrifuge for the separation of the solids.
In the full system operation, the centrate was returned to the lime regen-
eration reactor.
A schematic of the overall system operation is provided in Figure IV-1.
1. Sulfuric Acid Reactor Performance
The results of the sulfuric acid reactor pilot plant tests confirmed the
basic laboratory findings. The reaction proceeds readily to completion
with formation of a gypsum product with good settling properties. Prelim-
inary pilot tests showed that a 30-minute reactor holdup time is sufficient
for essentially complete conversion; and the optimum pH appeared to be
above 2.35.
Subsequent runs were then made over a range of pH from 2.6 to 3.3 in three
different operating regimes: ^£50^ limited, CaS03 limited, and approximately
stoichiometrics ratios of Na2SOtt to CaS03. The results of these runs are
summarized in Table IV-2 and Figure IV-2.
As would be anticipated, the runs with Na2SOit/CaS03 feed ratios nearest
to stoichiometrics (0.5) resulted in the highest reactor efficiencies.
(Here, efficiency is defined as the decrease in mols of sodium sulfate
divided by the mols of sulfuric acid used — [mols Na^O^ in-mols total
SOt^ out]/mols E2SO^ used.) The best efficiency attained in the pilot
plant operations was 0.64 at a (Na2S04/CaS03) feed ratio of 0.54. In cases
where the feed ratio was greatly different from stoichiometric, the reactor
efficiencies were found to decrease markedly. In the sulfate limited
regime, acid is wasted dissolving CaS03 which is not used; and in the
614
-------
Flue Gas
Tray Scrubber
With Demister and
2 - Trays
or Na2SO4
Ca(OH)2
f fr
3
Settler
Reactor System
f
\\\N\\\\\\\\\\\N\\\\\\^^\\\\\\k\NX\NNN•\.\\\N.\.\\\\\\\\\\N\\\X\\N\
Slurry
Makeup
H2S04 Reactor
\\\\\\\\
Centrifuge
Cake
r
Hold Tank
H2S04
Storage
N N. \ \ \ \ \ N IN NNNNNNNNNNNNNN N N \\N\\\ \\\ \\
\\|l\\\\\\\\
Figure IV - 1 Schematic Flow Diagram of Pilot Plant Operation For Sulfuric Acid Treatment Mode
-------
Table IV-2 SUMMARY OF SULFURIC ACID SLIPSTREAM TREATMENT RF.STJT.TS
Sulfuric Acid Reactor Conditions: T = 30 min (CSTR) Residence Time
T = 80 - 110CF
pH = 2.6 - 3.3
Feed Streams: Solids Source - Washed Filter Cake
Liquor - Scrubber Bleed or Thickener Underflow
Acid Strength = 2.5 - 3.0 M H2SOI|
Feed Composition
Operating
Regime
N32S04 Limited
Stoichiometric
CaS03 Limited
Run
No.
056
055
016-2
016-1
016-3
017-2
050
017-1
052
051
053
Wt %
Solids
24.
18.5
19.5
19.
19.5
18.5
13.
18.5
10.5
11.
26.5
Liquid Composition
[SOU. M
0.63
0.59
0.74
0.76
0.83
0.80
0.58
0.90
0.54
0.57
1.07
£H
11.5
6.9
6.9
7.1
6.9
7.2
6.9
7.1
11.0
7.0
11.5
Solids Composition, (wt%) mols Na2S04
CaSO-, CaSOi,
74
82
78
80
76
70
78
70
74
74
36
10
7
17
15
18
19
11
19
13
11
52
mol CaSO^
0.32
0.38
0.46
0.48
0.54
0.57
0.60
0.67
0.75
0.76
0.95
Reactor Efficiency ,
Observed
0.22
0.37
0.52
0.56
0.64
0.61
0.43
0.39
0.22
0.28
0.10
Corrected
0.33
0.45
0.55
0.59
0.68
0.70
0.52
0.44
0.27
0.32
0.13
mols NaySOu in - mols total SOt, out
Efficiency
Efficiency corrected for NaOH, CaC03 and Na2S03 alkalinity in feed slurry
-------
0.8 i
O
C/5
Jj)
O
CM
I
O
0>
QC
CSI
X
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Reactor Operating Conditions
pH = 2.6-3.3
Feed - Liquor Scrubber Bleed + Washed Cake
Legend
O-
Corrected
Observed
I
0.2 0.4 0.6 0.8
Feed Stoichiometry (mols soluble SO^/mol CaSOo) in Feed Liquor
Figure IV-2 HSO^ Reactor Efficiency versus Feed Stoichiometry
1.0
617
-------
CaS03 limited regime, acid is wasted neutralizing and adjusting the pH
of extra feed solution.
The pilot plant data plotted in Figure IV-2 cover a range of operating
conditions including sulfate and sulfite concentration, reactor pH, ionic
strength, and alkalinity content of the feed slurry. The chemistry of
the system indicates that all of these factors will influence the efficiency
of the reaction. Since all of these effects are confounded to some degree
in the pilot plant results, the correlation of reactor efficiency with
feed stoichiometry contains secondary factors other than analytical error
that contribute to the data scatter. The efficiency/feed stoichiometry
curve given in Figure IV-2 should be considered a generalized representation.
An attempt was made to extract the effects of feed alkalinity in order to
clarify the observed performance. This involved adjusting the sulfuric
acid used in each run to account for that acid used in neutralizing CaCOj,
NaOH and Na2S03 in the feed. The "corrected" reactor efficiencies based
upon these adjusted acid requirements are listed in Table IV-2 and are
superimposed in Figure IV-2. As would be expected, the magnitude of the
correction was generally greatest for the runs using simulated thickener
underflow and for runs in the CaS03 limited regime where excess solution
was fed.
The effects of sulfate concentration and reactor pH were specifically addressed
in the laboratory program. This work showed that the optimum reactor operating
pH is roughly 2.5 to 3.0, the range in which almost all pilot plant runs
were made. Furthermore, as sulfate concentrations increase in the reactor
feed slurry, reactor efficiency improves. This effect is not clearly
demonstrated in the pilot plant data due to the great range of feed stoich-
iometry, which has a more pronounced influence on reactor performance.
Finally, the ionic strength of the reactor effluent slurry has a small effect
on efficiency since it determines, along with temperature, the solubility
product of gypsum. As ionic strength increases, the solubility product
should also increase and, therefore, reactor efficiency should correspondingly
decrease. Although the effect on reactor efficiency is too small to be
observed, the increase in the gypsum solubility product with ionic strength was
observed. Figure IV-3 shows the apparent solubility product curve for
gypsum (calculated by molar concentrations) predicted by ADL along with
solubility data generated in both the pilot plant and laboratory tests.
For all but a few runs, the experimentally determined values of the apparent
solubility produce fall slightly below the predicted curve (predicted by the
method of Kusik and Meissner), indicating that the solutions' may not have
been saturated with respect to caSOit-2H20. The validity of this predicted
solubility product curve is demonstrated by the gypsum saturated data presented
in Section VI.-A.
The pilot plant solubility data shown in Figure IV-3 correspond to soluble
sulfate concentrations of 0.1M to 0.4M SO^ and calcium levels of 0.01M to
0.04M Ca4"*" (400 to 1600 ppm). The lowest effluent sulfate levels, and there-
fore highest calcium levels, were achieved in cases where the reactor effic-
iency was highest.
618
-------
- 10
O
CO
+
-f
£
D
Q.
(3
O
tt
D
O
<£
>
O
to
c
0>
Q.
Q.
ju, Ionic Strength
Figure IV-3 Extent of CaSO4 Under-Saturation in H2SO4
Reactor Effluent - Pilot Plant and Laboratory Data
619
-------
Although the highest efficiency achieved in the pilot plant tests was 0.64,
it is reasonable to assume that refinements in the operating conditions
and slight adjustments in the process configuration could raise efficiencies
to above 0.7. For example, the reactor vent gas could be recovered and
used for pre-acidifying the incoming filter cake slurry. Also, if Na2SO[t
is used for the makeup sodium value, then this could be added either as
a concentrate or solid directly to the sulfuric acid reactor feed slurry.
2. Integrated System Operation
Two closed-loop pilot plant tests were run for extended periods with the
sulfuric acid reactor system operating on a slipstream from the dual alkali
process. The overall system was operated in the concentrated active sodium
mode using hydrated lime for absorbent regeneration. The first test,
run 016, was made at an inlet S02 level of 2250 ppm. The second test, run
017, was conducted at an inlet S02 level of 650 ppm. The process conditions
and overall results are summarized in Table IV—3.
In both runs the operation of the sulfuric acid treatment system was similar
to the open-loop tests as previously described. Washed filter cake generated
during the run was used as the CaS03 source and venturi recycle liquor
provided Na2S04. The dual alkali process liquor was originally primed to a
soluble sulfate level of about 0.8M to ensure that sufficient sulfate was
present to produce high reactor efficiency. The rates of feed of cake slurry
and sulfuric acid were initially set such that the total anticipated oxidation
in the dual alkali system could be sustained by the combined sulfate precip-
itation in the absorbent regeneration reactor and sulfuric acid reactor and
by the loss of soluble sulfate in the waste cake.
The performance of the sulfuric acid system in these runs has been discussed
in connection with the sulfuric acid reactor performance in the previous
section. Overall, the results were very good. Efficiencies ranged from 0.38
to 0.64 throughout both tests, and the average efficiency over each test was
greater than 0.5. The gypsum produced had good dewatering properties.
Centrifuge cake containing up to 70 wt% insoluble solids could be obtained
by close control of the centrifuge feed rate.
The primary effect of the sulfuric acid slipstream treatment system on the
overall process operation in these two runs was in the increased lime require-
ments (for ultimate neutrali2ation of sulfuric acid added to the system).
There were no adverse effects on calcium utilization, soluble calcium levels,
or properties of solids generated in the absorbent regeneration reactor.
In all of these respects, the performance of the system was consistent with
that of a concentrated mode dual alkali process without sulfuric acid
treatment.
The consumption of lime, however, rose significantly. In run 016 where the
S02 removal equaled 2050 ppm, sulfite oxidation throughout the system amounted
to about 570 ppm, or 28% of the S02 removal. Approximately half of this
oxidation rate was sustained by sulfate precipitation in the lime reactor
620
-------
Table IV-3 SUMMARY OF CLOSED-LOOP RUNS
Sulfuric Acid Treatment Mode
Run No.
Dual Alkali Process Operation
Inlet S02 Level, ppm
S02 Removed, ppm
Scrubber Bleed: pH
[TOS], M
[so!;], M
Overall System Oxidation
Equivalent ppm S02 (avg.)
% AS02 (avg.)
Sulfate Precipitation in Lime Reactor:
Equivalent ppm S02 (avg.)
% AS02 (avg.)
Calcium Feed:
mols Ca(OH)2/mol AS02
mols Ca(OH)2/mol sulfur input
SQi4 Treatment System Operation
H^Ott Feed, (mols H2S04/mol AS02)
Reactor Efficiency
Sulfate Precipitation:
Equivalent ppm S02
% of AS02
Centrifuge Cake Composition:
Insoluble Solids (% total wt)
Soluble Solids (% dry cake)
016
2250
2050
6.1
0.35
0.7
570
28%
0.26
0.52-0.64
280
14%
60-70
4-7
017
650
530
6.1
0.5
0.8
400
"70%
250
12%
1.33
1.06
100
18%
1.83
1.01
0.63
0.38-0.61
180-300
34-57%
53
15
621
-------
and by soluble sulfate losses in the waste cakes. The remainder, about 14%
of the S02 absorbed, was removed in the sulfuric acid reactor. Since the
reactor efficiency averaged 0.57 (efficiencies ranged from 0.52 to 0.64),
the sulfuric acid requirement amounted to about 26% of the S02 absorbed. This
acid feed was directly reflected in the required increase of 26% in the lime
feed (1.33/1.06=1.26). At the lower S02 inlet level of run 017, the use of
sulfuric acid increased markedly due primarily to oxidation in the dual alkali
system being a higher percentage of the S02 absorbed. In addition, Na2SOij
was used to make up lost sodium value. The combined oxidation and sulfate
addition resulted in an increase in the lime feed of 80 - 85% over that
required for S02 removal alone; and a total sulfuric acid requirement of
63% of the S02 removed.
It is clear from the pilot plant results that the sulfuric acid treatment
scheme is a technically feasible and reliable approach to the removal of
soluble sulfate from dual alkali systems. Because the use of this sulfuric
acid treatment scheme may be costly when applied to systems with high
oxidation rates (>40% of AS02) due to the sulfuric acid and extra lime
requirements, it may be more appropriate for systems with intermediate
levels of oxidation where the rate of sulfate formation cannot be easily
handled in a simple concentrated active sodium mode.
Composite Diagram of Projected Pilot Plant Operations Figure IV-4
summarizes the operational characteristics of the various dual alkali process
sections for the concentrated lime mode operation with sulfuric acid slip-
stream treatment for a high S02 inlet condition. This composite is based
upon pilot plant operations and is analogous to the block diagrams provided
in Section III. C. for the simple concentrated lime mode. Figure IV-4
indicates the various inputs and outputs expressed in terms of S02 absorbed
for a dual alkali system with an overall oxidation rate equivalent to 30%
of the S02 absorbed. For each of the four process sections, the observed
rates of sulfite oxidation and/or active sodium regeneration are also
expressed as a function of 862 absorption.
622
-------
L200ppmSO,
2500 ppm SO2
Na2C03
5% of AS02
'
1
Scrubber System
90-95% SO2 Removal
Oxidation = 25% of AS02
J
^^
L
pH = 5.9-6.1
^active = °'5M
[S04=] 0.8M
t
\
Regneration System
Oxidation = 3% of ASO., ^ Lime
Sulfate Precipitation 1 1 5% of ASO,
= 1 5% of A S02 i
1 _L
H2S04
^-
20% of ASO2
n
i
r ~\
\
H2S04 Treatment
r
bb% ttticiency
Oxidation =» 0% of ASO,
I
\
pH = 7.5-9
r
Dewatering System H n
Oxidation
•\
~ 45%
1
Two Displacement Wash
-55%
r
Average Composition (wt.%dry basis):
CaSO4 = 85%
CaS03 = 5%
10%
Cake (45% Insoluble Solids)
Average Composition (wt% dry basis):
NaSOd
CaS03 = 77%
Other Insolubles = 5%
Na2S04 = 2%
Na2SO3 = 1%
Figure IV-4 Composite Pilot Plant Operation — h^SO. Treatment Mode
-------
C. SULFURIC ACID REACTOR MODEL
The model of the sulfuric acid reactor system was developed to simulate the
effects of various dual alkali process conditions on the sulfuric acid
system efficiency. The principal purpose of the model was to provide a
method for estimating reasonable sulfuric acid reactor operating conditions
and for determining the sulfuric acid requirements in different dual alkali
system applications. As developed, Lhe model applies to the operation of the
reactor in the CaS03 limited regime. It is based upon the reaction equations
discussed in Section IV.-A. and assumes that chemical equilibrium is achieved
in the system. The model determines the amount of sulfuric acid required
to neutralize all alkalinity values in the slurry feed and adjust the pH
into the 2.3 to 3.3 range. The levels of sulfate and calcium in solution
in the reactor effluent are determined from an estimate of the "KSp" for
CaSCH ("Ksp" = [Ca++] x [SO^] with concentrations in mols/liter). The
value of "K " used is the average of those values calculated from the
laboratory data for the range of ionic strengths expected in the pilot
plant operations. Since "KSp" is an input to the model, this can be
adjusted for any desired ionic strength.
In order to adjust for pH, it has simply been assumed that enough sulfuric
acid is added to convert about 20% of the system TOS to H2S03. This is
a rough estimate from the laboratory results. It would, of course, be
possible to try to determine the H2S03/HS03 ratio from pK for H2S03.
However, very little data exist which will allow for reasonable estimates
of the reference activity coefficients for the H2S03 and HS03~ species at
ionic strengths greater than 1.0. Furthermore, initial tests with the model
indicate that calculated reactor efficiencies were relatively insensitive
to rather large changes in this pH adjustment factor.
No consideration has been specifically given in the model to the formation
of HSO^". At the pH's involved, the amount of HSO^" should be less than a
few percent of the total S(VI) species.
Model Application The model has been used primarily to design
pilot plant experiments and to determine the sulfuric acid requirement
for various dual alkali system oxidation rates.
Table IV-4 compares the experimental reactor efficiencies from the pilot
plant operations with those predicted by the model. The model tracks the
experimental runs fairly closely and, in most cases, the predicted value
of efficiency is within about 15% of that observed. This agreement is good,
particularly in light of the sensitivity of the model to the solids level
in the feed slurry. A change of 10% in the solids level (from 20% to 22%
solids) changes the efficiency estimate by about 30% (from say 40% to 52%).
Since the solids level is an experimentally determined value that can vary
slightly during a run, this sensitivity is significant.
The effects of other operating conditions on the reactor efficiency are
smaller. Small changes in [OH~], [TOS] and CaC03 in the cake produce only
minimal changes in efficiency. The effect of changes in sulfate concentration
624
-------
Table IV-4 MODEL SIMULATIONS OF PILOT PLANT OPERATIONS
Feed Ratio
Xmols
Vmol CaS
0.54
0.57
0.60
0.67
0.75
0.76
0.95
Experimental
Reactor
Efficiency
0.64
0.61
0.43
0.39
0.22
0.28
0.10
Efficiency
Predicted
By Model
0.72
0.56
0.45
0.43
0.13
0.19
-0.12
625
-------
also tend to be small, although to some extent this is a function of the
actual level of sulfate. As sulfate concentrations in the feed decrease
toward the equilibrium effluent sulfate concentrations, changes in sulfate
concentration become more important.
Neither the pH adjustment factor nor the value of "KSp" has been found to
have a strong influence on the model performance. A 100% increase in "KSp"
from 8 x 10~^ to 16 x 10~3 generally decreases the predicted reactor efficiency
by less than 10%. Similarly a 50% increase in the pH adjustment factor, from
0.2 to 0.3, usually decreases the efficiency prediction by only about 5%.
626
-------
D. APPLICABILITY OF THE SULFURIC ACID TREATMENT SYSTEM
In using a sulfuric acid slipstream treatment system, the amount of sulfuric
acid required is important because, it directly affects the overall lime
requirement. As the sulfuric acid rate increases, the lime rate must increase
accordingly to remove the additional sulfur value added to the system. For
example, in order to precipitate sulfate at a rate sufficient to keep up with
an oxidation rate equivalent to 15% of the S02 absorbed, the lime feed
requirement will increase by 25% for a 60% reactor efficiency.
The implications of calcium utilization in the absorbent regeneration reactor
on the sulfuric acid efficiency are also important, particularly if limestone
is used for regeneration with, lower expected calcium utilization. The
influence of calcium utilization on reactor efficiency is illustrated in
Figure IV-5. In order to ensure 30% reactor efficiency under reasonable
system operating conditions, calcium utilization must exceed 75%; and in
order to achieve 50% reactor efficiency the utilization must exceed 90%.
The consequences of using the sulfuric acid slipstream treatment approach
for sulfate regeneration should, therefore, be carefully evaluated in terms
of the overall process operation. In many cases where oxidation rates are
high enough that they cannot be easily handled by a normal concentrated mode
operation, other dual alkali approaches might be more promising than a
sulfuric acid treatment scheme.
627
-------
100
80 -
Basis
Slurry Feed-20 wt % Solids
(mols S04/mols CaSOx)so|Jds = 0.1
Sol'n Cones. = IS04=) = 0.75M
[S03=] =0.10M
{HSO^l =0.30M
H2SO4 Feed Concentration - 2.56M
60
u
c
0>
'o
o
s
OC
rf
O
01
CM
I
40
H2S04 Rxtr Efficiency >30%
20
Req'd
Calcium
Util.
20
80
40 60
Calcium Utilization (%)
Figure IV-5 HoSO4 Reactor Efficiency Versus Calcium Utilization
100
628
-------
V. LIMESTONE REGENERATION — CONCENTRATED MODE
A. INITIAL LABORATORY STUDIES
Since the incremental cost of lime over the equivalent amount of lime-
stone is tied quite closely to the increasing cost of the energy re-
quired to calcine the limestone to lime, a considerable operating cost
saving would be realized if limestone could be used directly in a dual
alkali process in lieu of lime. Limestone has been used in FGD systems
which employ direct slurry scrubbing, and the chemical equilibria in-
volved have been characterized extensively. By analogy, limestone
should be applicable to dual alkali regeneration. However, because
there are significant differences between reaction conditions in direct
slurry scrubbing and dual alkali regeneration — one of the most impor-
tant is the significantly higher ionic strengths found in concentrated
dual alkali solutions — a program of laboratory experiments was under-
taken to study the reaction between limestone and concentrated sodium
sulfite/bisulfite solutions. In a parallel effort, the use of lime-
stone to regenerate dilute sodium sulfite/bisulfite solutions was stud-
ied by personnel at the Research Triangle Park laboratories of the EPA.
The results of those studies are discussed in Section VI.
Limestone is being used in two commercial dual alkali systems in Japan.
However, both systems utilize special treatment steps for sulfite pre-
cipitation; both ultimately produce gypsum and are relatively complex
processes. They are described in more detail in a paper by Kaplan.7
The ADL laboratory program, still in progress, has consisted of an ini-
tial series of batch experiments in which the effects of the type of
limestone, limestone feed stoichiometry, reaction temperature, and
sodium sulfate level on the kinetics and equilibria of the reaction
were studied. In subsequent continuous reactor experiments, limestone
utilization and the physical properties of the product solids were
studied as a function of a number of experimental variables.
As pointed out in the introduction, limestone will not react with sodium
sulfite solutions because limestone is less soluble than the calcium
sulfite which would be produced. But the bisulfite ion, HSOs", is suffi-
ciently acidic to react with limestone. This neutralization reaction,
which results in the ultimate precipitation of calcium sulfite, can be
written simplistically as follows:
CaC03, 4- 2Na2HS03 ->• CaS03, , + Na2S03 + H20 + C02 t (20)
629
-------
In fact, the bisulfite initially present is never completely neutral-
ized because at pH's where the reaction will proceed (up to about pH =
7), the species H2C03, HC03~, HS03~, and S03= all exist in significant
quantities in equilibrium with one another. Nevertheless, based on the
above equation one can express the amount of limestone fed as the per-
cent of the stoichiometric amount which would be required to neutral-
ize the HS03~ initially present.
r
,1
0.5 x mols HS03
x 100%
Thus, a feed stoichiometry of 100% is obtained when the mole ratio of
CaC03 to HS03- fed is 1:2. This definition of feed stoichiometry will
be used throughout the discussion which follows.
Regardless of the nature of the species in solution, it is the removal
of soluble sulfur-containing ions — both sulfur (IV), sulfite; and
sulfur (VI) , sulfate — from solution as insoluble calcium salts by
which the effectiveness of limestone utilization is realistically eval-
uated. In CSTR experiments, utilization was based on analyses of the
reactor product solids and calculated as follows:
„, ,-_.-,. ... f mols CaSOx in solids | 1rin°/
% Utilization = —q - X- : . , X 100%
total mols Ca in solids I
Since the initial batch experiments were comparative in nature, for the
sake of expediency the precipitation of CaSOi^ was ignored and limestone
utilization was computed on the basis of the decrease in the concentra-
tion of soluble sulfur (IV) — Total Oxidizable Sulfur [TOS] — in solu-
tion as the reaction proceeded. Thus, if limestone is fed at 100% of
stoichiometry and it is completely utilized, one would expect [TOS] to
decrease by an amount equal to one half the concentration of bisulfite
ion, [HS03~], initially present. This relationship can be expressed
as follows:
„ TT_ ., . _ . T [TOS] in - [TOS]out 1
% Utilization = -r—r .„_. , . *• , • *.—; , /^nn\ x 100%
0.5 x [HS03~]in x (% Stoich/100)
Because the formation of CaSOif is ignored in this calculation, it leads
to conservative estimates of actual utilization.
1. Batch Studies Comparing Limestone Reactivity
The initial experiments in the laboratory program compared the relative
reactivities of several types of limestone toward concentrated, acidic
sodium sulfite/bisulfite solutions. In previous studies related to
limestone slurry scrubbing, Drehmel8 found significant differences in
reactivity which were attributed to differences in both the chemical
and physical properties of the wide variety of limestones tested.
630
-------
Materials from three sources were available in quantity for testing and
subsequent use in this laboratory program. By chemical analysis all
were more than 95% CaCO-^. The materials included:
• Reagent grade CaCC>3 (precipitated chalk); Fisher Scientific Co.
• A natural calcite obtained from Pfizer, MPM Div., Clifton, New
Jersey. This material (Marblewhite 200) was an industrial grade,
lov; magnesium, limestone produced at Adams, Mass.
• Fredonia limestone ground by and obtained from the EPA/TVA
Shaxmee Test Facility.
Based on the sieve analyses shown below, the Marblewhite 200 and the
Fredonia limestones appeared to be similar in their particle size dis-
tributions while the reagent CaC03 was uniformly finer.
Mesji
- 300
-200 + 300
-100 + 200
- 60 + 100
Limestone Type
Reagent
100%
-
-
_
Marblewhite 200
87.5%
11 %
1.5%
0.2%
Fredonia
85.7%
8.9%
3.7%
0.4%
Examination under the scanning electron microscope (SEM) revealed more
subtle differences which were not evident from the sieve analysis. The
reagent grade material was composed of quite uniformly sized cubes about
8 microns (y) on a side. The Marblewhite particles also appeared crystal-
line but were very irregular and jagged in shape with a longest dimension
ranging from 2 y to greater than 50 y; an average length was estimated to
be about 20 y. The Fredonia material, which was similar to Marblewhite
by sieve analysis, appeared very different under SEM examination. It
appeared amorphous; the bulk of the particles were rough, irregular
spheres about 2 - 8 y in diameter. Particles larger than 10 y were
present but they appeared to be aggomerates of many smaller particles.
The results of batch reactions of the three limestones with similar con-
centration sodium sulfite/bisulfite sulfate solutions at 50°C are shown
in Figure V-l. The progress of the reaction as a function of time was
followed by measuring the decrease in TOS concentration in solution.
The substantially higher reactivity of Fredonia limestone (as compared
to reagent CaCOs) is immediately evident from the more rapid TOS re-
duction obtained with the Fredonia material. The solution with which
the Marblewhite was reacted had a somewhat higher initial TOS level than
the other two solutions.
631
-------
0.4
to
S3
CO
o
Conditions
Limestone Feed=100%Stoich
Initial pH = 5.5
[Na2S04J =0.75M
0.2
ATOS/At{x103)
(gm mols/litermin)
0-15 Minutes
O 6.6
A 4.0
D 4.0
Legend
QFredonia Limestone
A Marblewhite Limestone
D Reagent CaCCU
0.1
_L
_L
20
40
60
80 100
Time (minutes)
120
140
160
180
Figure V-1 Comparison of Limestone Reactivities in Batch Reactions
-------
To permit a better comparison of reactivity, initial reaction rates were
computed and the values included in Figure V-l. The overall reaction
rates shown are the change in TOS concentration per minute for the re-.
action time period 0-15 minutes. The computed overall reaction rates
for Marblewhite limestone and reagent CaC03 were essentially the same
while the rate observed for Fredonia limestone was about 50% faster than
either of the other two.
With the Fredonia limestone, TOS changed very little after the first 45
minutes of reaction. With the other two materials, a noticeable reduc-
tion in TOS, indicative of continuing slow reaction, was observed through-
out the remainder of the experiments which were terminated after three
hours.
Plotted at the extreme right-hand side in parentheses in Figure V-l are
the TOS levels which would have been obtained if the limestone which was
fed had been completely utilized to precipitate calcium sulfite. The
final TOS level observed for Fredonia limestone was much closer to the
theoretical level than it was for the other two materials. Utilizations,
based on TOS reduction, in solution, even after three hours of reaction,
were markedly poorer for Marblewhite limestone and reagent CaC03 (81 and
85%, respectively) than for the Fredonia material (96%). Utilization
of the Fredonia material after only 45 minutes of reaction was about
92%, whereas utilizations of Marblewhite and reagent CaC03 after 45
minutes were only 54 and 64%, respectively. If precipitation of CaSO^
had been taken into account, all utilizations would have been higher;
in the case of Fredonia, utilization would have been about complete.
Because of the chemical similarity of all three materials, the observed
difference in reactivity between Fredonia and the other two materials
probably resulted from the fact that the Fredonia limestone consisted
of amorphous particles with a higher accessible surface area than the
crystalline particles characteristic of the other two.
Because of its significantly higher reactivity in batch tests, the Fre-
donia limestone was used almost exclusively in the remainder of the batch
and CSTR experiments that were performed. We do realize, however, that
further consideration must be given to the effect of limestone reactivity
before a complete assessment of its potential for dual alkali regeneration
in a wide variety of situations can be made. The batch reaction method
employed seems to be a relatively simple but realistic means to study
reactivities. Additional tests of the reactivity of materials from a
variety of locations are planned as part of our continuing ADL laboratory
program.
2. Effects of Feed Stoichiometry, Sulfate Concentration, and
Temperature on Reaction Rate
The effect of changing the Fredonia limestone feed Stoichiometry on the
rate of the regeneration reaction is shown in Figure V-2. When the amount
of limestone fed was reduced from 100% to 50% of the stoichiometric amount
633
-------
0.4
Conditions
Initial pH = 5.5
[Na2S04] = 0.75M
Temperature = 50°C
0.3
CO
O
(0)
0.2
ATOS/At(x10^
(gm mols/litermin)
0-15 minutes
A 8.6
O6.6
D4.0
-A (A)
(O)
Legend
A200% Stoichiometric
Q100% Stoichiometric
D50% Stoichiometric
0.1
I
20
40
60
80 100
Time (minutes)
120
140
160
180
Figure V-2 Effect of Limestone Feed Stoichiometry on
Regeneration Rate With Fredonia Limestone
-------
required to neutralize the HSC>3 initially present, the reaction rate
during the first 15 minutes of reaction also decreased significantly,
although the percent decrease in reaction rate was not as great as the
reduction in the feed stoichiometry. The theoretical minimum TOS level
after 45 minutes of reaction was more closely approached at the lower (50%)
stoichiometry; and the resulting limestone utilization (based on TOS re-
duction only) was about 97% as compared to 92% when the feed was 100% of
stoichiometric. When the limestone feed stoichiometry was doubled from
100% to 200%, the initial reaction rate increased about 30% and the TOS
concentration after 45 minutes had fallen essentially to the minimum
theoretical level. Within experimental error, the concentration of TOS
remained constant during the remainder of the experiment. However, none
of the limestone added in excess of" 100% stoichiometric reacted — the
overall utilization was only about 50%.
The dependence of the reaction rate on limestone feed stoichiometry ob-
served in these experiments indicates that limestone dissolution is one
of the rate limiting steps in the regeneration reaction with limestone.
The effect on reaction rate of the concentration of Na2SOl+ in solution
is shown in Figure V-3. It is immediately obvious that at 1.25M Na2SOi+,
the reaction was significantly slower than at the two lower Na2SOt+ levels.
Examination of the computed reaction rates for the initial 15 minutes of
reaction reveals that initial rates were significantly different for each
of the N32S04 levels studied. Each increase in the concentration of
Na2SOit was accompanied by a corresponding decrease in initial reaction
rate. However, the reaction rate for the first 45 minutes was essentially
the same for Na2SOit levels of 0.25M and 0.75M, but the corresponding rate
at 1.25M Na2SOi+ was still significantly lower.
All of the preceding experiments were conducted at a temperature of 50°C
which is representative of the temperature at which a reactor would operate
if scrubber effluent were neither heated nor cooled prior to regeneration.
The significant decrease in reaction rate at a lower temperature, 38°C, is
shown in Figure V-4. The 12°C reduction in reactor temperature produced
essentially the same reduction in the calculated reaction rate as did in-
creasing the concentration of Na2S04 from 0.75M to 1.25M in the preceding
experiment,
3. Continuous Reactor Studies of Regeneration with Limestone
The results of the batch experiments indicated that Fredonia limestone
could be used effectively to regenerate concentrated sodium bisulfite/
sulfite solutions. The regeneration with limestone was studied further
in the laboratory CSTR (continuous , stirred tank reactor) to observe the
physical properties of the solids produced, to ascertain whether or not
sulfate would also be precipitated during the course of the reaction,
and to obtain a more realistic estimate of achievable limestone utilization.
635
-------
0.2
Conditions
Limestone Feed = 100% Stoich
Initial pH = 5.5
Temperature = 50°c
ATOS/AtlxlO-
'(gm mols/litermin)
0-15 Minutes
A 9.6
Q6.6
D4.3
Legend
A 0.25M
O 0.75M
D 1-25M
Na2S04
Na2SO4
0.1
I I
I
I
I
20
40
60
80 100
Time (minutes)
120
140
160
Figure V-3 Effect of Na^SO^ Concentration on
Regeneration Rate With Fredonia Limestone
180
-------
0.4
Conditions
Limestone Feed = 100% Stoich
Initial pH = 5.5
[Na2SO4)=0.75M
0.3
CO
O
0.2
ATOS/At(x103)
(gm mols/liter'min)
0-15 Minutes
A 4.6
O6.6
Legend
A 38° c
O 50°C
0.1
I
I
I
I
I
I
I
I
20
40
60
120
140
80 100
Time (minutes)
Figure V-4 Effect of Temperature on Regeneration Rate With Fredonia Limestone
160
180
-------
The results of the batch experiments indicated that with Fredonia lime-
stone, achieving nearly complete utilization at a temperature of 50°C and
with Na2SOtt concentrations of 0.75M or less, required about 45 minutes of
batch reaction time. In a CSTR that length of reaction would translate
to a residence time of roughly 2-2.5 hours. However, the CSTR system
available when these studies needed to be. performed had been specifically
designed to operate at shorter residence times (5-50 minutes) for study-
ing regeneration with lime, and would not operate reliably at longer
residence times. For the sake of expediency, the initial set of CSTR
studies with limestone was performed at the 50-minute maximum residence
time obtainable; a new CSTR is being constructed for additional studies
at longer residence times.
The observed utilizations of limestone (based on CaSO /total Ca in the
product solids) in the 50-minute CSTR as a function of limestone feed
stoichiometry are shown in Figure V-5. Included for reference are the
utilizations achieved after 45 minutes of batch reaction based on TOS
removal only; these are known to be somewhat low because CaSO^ formation
is not included. Even though the CSTR residence time was short, utiliza-
tions of Fredonia limestone greater than 80% were achieved with a feed
stoichiometry of about 60%. When the stoichiometric amount of limestone
fed was increased, utilization decreased significantly.
Utilization of the less reactive Marblewhite limestone was also studied
in the 50-minute CSTR at a feed stoichiometry of 100%. Utilization of
this material was only about 55%; utilization of Fredonia under similar
conditions was about 70%.
Included in Figure V-5 are the results of a number of experiments in which
the CSTR residence time was maintained at 50 minutes but a portion of the
reactor effluent slurry was collected, filtered, and the solids recycled
to the reactor. In once-through CSTR operation, the weight percent sus-
pended solids in the reactor effluent slurry was typically about 2%. In
several experiments with Fredonia limestone, the suspended solids level
in the reactor was raised to about 5% by recycling solids; and as shown
in Figure V-5, utilization of the limestone increased significantly over
that observed for once-through operation. Operation with solids recycle
also appeared to increase the utilization of Marblewhite limestone. The
two experiments with Marblewhite were conducted at different feed stoich-
iometries, but when effluent solids were recycled, Marblewhite utiliza-
tion was comparable to that obtained during once-through operation with
Fredonia; in once-through operation, Marblewhite utilization had been
significantly lower than that observed for Fredonia. Recycling solids
effectively increased the amount of time available for the limestone to
react by a factor of 2.5 - 3. A similar increase in reactor residence
time should result in a similar improvement in utilization for once-through
operation.
638
-------
100
80
01
c
o
60
Conditions
Batch Reactor-[TOS] = 0.4M
[Na2S04] =0.75M
CSTR-[TOS] =0.35M
[Na2S04] =0.6M
Common to All—Initial pH = 5.5
Temperature = 50°C
40
c
-------
0.20
o
V)
**
c
0)
3
0.16
£ 0.12
8
DC
C
"5
0.08
CJ
•§ 0.04
Conditions
Initial pH = 5.5
Initial [TOS] = 0.35M
Temperature - 50°C
Limestone Feed Stoichiometries shown (on graph)
(60%)
(100%)
o
(50%)
O
(70%)
O
I
I
0.1
0.2
0.3
0.5
0.6
0.4
[Na2SO4), M
Figure V-6 Precipitation of Sulfate by Fredonia Limestone in 50-Minute CSTR Experiments
0.7
0.8
-------
B. INITIAL PILOT PLANT TESTS AND SUBSEQUENT STUDIES OF FACTORS AFFECTING
THE PHYSICAL PROPERTIES OF PRODUCT SOLIDS
With the laboratory experimental results up to that time all indicating
that Fredonia limestone could be used to regenerate concentrated sodium
bisulfite/sulfite solutions effectively, two closed-loop tests with Fre-
donia limestone were attempted in the ADL pilot plant. In the first run,
the limestone reactor was a simple CSTR with a 90-minute holdup; in the
second run the CSTR was operated with a recycle of solids from the thicken-
er underflow. Although calcium utilizations ranged from 75% to 85%, the
dewatering properties of the solids generated in the reactors were poor.
Extremely low settling rates resulted in the carryover of a considerable
amount of solids in the thickener overflow until, after a short period
of operation, the scrubber feed liquor contained 2 - 3 wt % solids.
The filter cake reflected the poor quality of solids. The washed cake
averaged less than 40 wt % (dry cake basis).
During operating periods when the reactor effluent slurry pH fell below
6.8, bubbling began to occur in the thickener, indicating that a signi-
ficant amount of additional reaction was taking place. This contributed
to the solids carryover in the thickener overflow.
After this initial pilot plant experience, laboratory studies with the
seemingly negative goal of producing poor settling solids were under-
taken in an effort to explain the pilot plant results. Two obvious
differences between operating conditions in the laboratory and in the
pilot plant were explored initially in 50-minute CSTR experiments in
which the solutions to be regenerated were similar to those used in
earlier experiments (0.6M Na2S04 and 0.36M TOS). Limestone had been
fed as a slurry in the laboratory, but was fed dry in the pilot plant.
A CSTR experiment was conducted in the laboratory with limestone fed
dry and no deterioration of settling properties over those observed pre-
viously could be detected.
The sodium sulfite/bisultie/sulfate solutions which were regenerated in
the laboratory did not contain dissolved C02, while in the pilot plant
the scrubber effluent did contain dissolved C02 which could have possibly
impeded the reaction. However, good solids settling behavior was still
observed when the solution fed to the laboratory CSTR was saturated with
C02.
The one other significant difference between the laboratory and pilot
plant experiments was the Na2SOi+ concentration. During the pilot plant
tests, the Na2S04 level ranged from 0.9 to 0.8M. The higher Na^iSO^ level
fed to the laboratory CSTR up to that time was 0.75M which, after dilution
by the limestone slurry, resulted in a maximum Na2SOi+ concentration of
0.6M in the reactor. Subsequent laboratory CSTR experiments at higher
concentrations of Na2SOtt indicated that the higher Na2SOi+ level in the
pilot plant was probably the factor responsible for the poor settling
behavior observed.
642
-------
The dramatic change in settling behavior which occured when the concen-
tration of NagSO^ in the laboratory CSTR was changed from 0.6 to l.OM
is shown in Figure V-7. At intermediate concentrations of ^250^ (0.75 -
0.80M), the settling behavior observed for samples taken one hour after
the reactor had filled and begun to overflow was intermediate between
the two extremes in observed settling behavior. At those intermediate
Na2SOit levels, it was further observed that settling behavior tended to
improve and approach that observed for 0.6M ^250^ as the reactor was
allowed to continue to operate for an additional period of 2 -3 hours.
However, the very poor settling behavior observed at the highest ^280^
concentration did not change over a period of several hours of reactor
operation.
Clearly, on the basis of the 50-minute CSTR laboratory results to date,
the need to control sodium sulfate concentrations at a level of 0.75M
or lower appears to be an important prerequisite for successful system
performance when regenerating with limestone. Whether this restriction
can be relaxed if longer residence times are employed will be an important
element of future CSTR tests.
One additional area related to the regeneration with limestone which is
currently being studied is the effect of extraneous species on the re-
generation reaction. Limestone can contain varying amounts of magnesium;
chlorides can be absorbed from flue gases; and it is likely that smaller
amounts of other soluble species could build up to significant levels
during extended periods of closed-loop operation. To date, the effects
of magnesium have been studied in batch and CSTR experiments; the effects
of chloride and other species will be studied in future experiments.
The effect of magnesium, added as MgSO^ at a concentration of 200 ppm
(mg/liter) on the regeneration with Fredonia limestone in a batch exper-
iment is shown in Figure V-8, where TOS removal as a function of time
with and without the added magnesium is compared. The 200 ppm of Mg
reduced the reaction rate significantly. In fact, if the calculated
initial reaction rates observed with 200 ppm Mg present are compared
with those shown in Figure V-3 when the Na2SOit level was increased to
1.25M, it can be seen that the presence of the small amount of magnesium
slowed the reaction more than increasing the Na2SOi4 level from 0.75M to
1.25M
The settling behavior of solids produced during 50-minute CSTR experiments
in which about 100 and 200 ppm of Mg had been added to the solution to
be regenerated is shown in Figure V-9. Settling curves for experiments
in which no MgSOit was added are included for comparison. The 19- and 24-ppm
Mg levels shown in Figure V-9 for these experiments were the amounts of
magnesium dissolved from the limestone.
643
-------
120
100
o
o
Q.
80
60
40
20
Conditions
Limestone Feed = 100% Stoich
Initial pH = 5.5
Temperature = 50°C
[Na2SO4l = 0.98M; [TOS] in = 0.36M
after 1 hour; stable with time
[Na2S04] = 0.6M; [TOS] jp = 0.36M
after 3 hours; stable with time
10
20
30
Figure V-7
40
50
Time (minutes)
60
70
80
90
Settling Curves For Solids Produced in 50-Minute
CSTR Experiments Using Fredonia Limestone
-------
0.4
Conditions
Initial pH = 5.5
[Na2SO4l = 0.75M
Temperature = 50°C
Limestone Feed Stoich
100%
cr-
4>
Ui
to
O
0.2
0.1 •-
0
ATOS/At(x10°)
(gm mols/litermin)
0-15 Minutes
O 6.6
A 3.3
Ji.
-O
Legend
O No Mg Added
A 200 ppm Mg Added
I
I
20
40
60
80 100
Time (minutes)
120
140
160
180
Figure V-8 Effect of Soluble Magnesium on Batch Regener-
ation Reaction Rates With Fredonia Limestone
-------
120
Conditions
Initial pH = 5.5
Initial [TOS] = 0.36M
Temperature = 50°C
100 --^^.LLmestone Feed Stoich = 100%
80
ex
O
£ 60
CO
3
u
40
20
(A) 0.98M SC>4 ; 24 ppm Mg
'(B) 0.62M 504 ; 202 ppm Mg
-{C) 0.63M SO4 ; ~ 120 ppm Mg
(D) 0.43M SC> ; ~ 208 ppm Mg
\
— (E) 0.060M SO4~; 0,19 ppm Mg
_L
.L
_L
10
20
30
40 50 60
Settling Time (minutes)
70
80
90
Figure V-9 Settling Behavior Observed For Solids Produced
in a 50-Minute CSTR Using Fredonia Limestone
-------
The significant change in settling behavior when Kg"*"1" was added can be seen
by comparing the settling curves (B, C and E) for the three experiments in
which the concentration of Na2SOit was held essentially constant at about
0.6M. The effects of Na2SOi4 and Mg"1^ levels in solution appear to be simi-
lar and, after a fashion, additive/ When the concentration of Na2SOl+ was
reduced to O.A3M (curve D) , moderately good settling behavior was observed
even with 208 ppm Mg"*"*" present.
Additional CSTR studies of the regeneration reaction with limestone in
the presence of magnesium at longer reactor residence times will determine
whether or not the lower reaction rate can be accommodated given additional
reaction time. If the additional studies confirm that soluble magnesium is
really as deleterious as it appears to be at this time, and no easier means
is found to circumvent or eliminate its effect, it should be possible to
hold Mg++ at acceptable levels in the process loop by treating a portion
of the process flow with lime to remove Mg*"1" as magnesium hydroxide. The
results of initial batch experiments in which samples of the effluent
slurry from the laboratory CSTR were treated with lime to various pH's
are shown in Table V-l. At a final pH of 9.7, essentially no Mg4"1" was
removed, but when the pH was raised to 11.2 a reduction in soluble Mg*"1"
of about 98% was achieved. This approach may necessarily limit the
process to the use of low magnesium limestones.
547
-------
Table V-l
LIME TREATMENT3 OF
CSTR EFFLUENT SLURRY AFTER REGENERATION WITH. LIMESTONE
Before Treatment After Treatment
[TOS], M [Mg], ppm pH [TOS], M [Mg] . ppm pH
.262 163 6.70 .188 162 9.7
.260 161 6.70 .190 3.4 11.2
.262 165 6.75 .179 2.7 12.05
Reaction, 50-min reaction time; [Na2SOit] , 0.61 M; Temp., 46°C.
648
-------
C. CONCLUSIONS REGARDING LIMESTONE REGENERATION OF CONCENTRATED SODIUM
BISULFITE/SULFITE SOLUTIONS
The results available to date fro.m the ongoing laboratory program indi-
cate that a sodium/calcium dual alkali process using limestone may be
feasible. Using limestone in place of all or the majority of the lime
which would otherwise be required should make the economics of sodium/
calcium dual alkali FGD systems more attractive, if limestone can be
accommodated within a reasonable degree of increase in process complex-
ity or control requirements.
Because limestone is a less reactive regenerant than lime, and because
it occurs naturally in a variety of forms which differ widely in their
reactivity, a greater amount of overall attention will be required to
ensure reliable, efficient process operation with limestone than with
lime. The characteristics of regionally available limestones and the
designs of processes to be sited in those areas will probably need to
be considered together. The quality of incoming limestone will need to
be assured to prevent process upsets, and considerably more attention
may need to be given to the control of process liquor composition than
would be necessary in a lime-based process.
The investigation of the use of limestone in dual alkali processes is
continuing and will proceed to the pilot plant if a viable approach is
developed.
649
-------
VI. LIMESTONE/LIME DILUTE MODES
As explained in , Section I, above, operating with dilute sulf ite/bisulf ite
(TOS) solutions represents one possibility for increasing the proportion
of sulf ate in the cakes produced during regeneration. Against this pos-
sible benefit must be balanced the cost penalties (represented by larger
equipment and higher pumping rates) caused by operating with more dilute
TOS solutions.
A. EPA LABORATORY RESULTS
In its Research Triangle Park laboratories, EPA had already performed
some preliminary scouting runs on the regeneration of dilute TOS solu-
tions with limestone and lime. These were summarized by Draemel9, who
showed that the CaS03 exhibited a pronounced tendency to supersaturate;
i.e., the reactions
CaC03 + 2NaHS03 •> CaSO^ + Na2S03 +• C02 + H20 and (21)
Ca(OH)2 + 2NaHS03 + CaSOs* 4- Na2S03 4- 2H20 (22)
did not proceed rapidly to completion in dilute TOS solutions. His scout-
ing runs indicated that this tendency could be counteracted by recirculat-
ing the product solids through the reactor, and that maintaining a level
of, say, 5% solids in the reactor might permit reasonable utilization
(80-90% with lime, 50-70% with limestone), with 1-hour holdup times.
EPA extended this work by performing a series of continuous limestone runs
in a 2-liter reactor followed by a 5-liter clarifier, with slurry recycle
to the reactor. These were long-term runs (36-50 hours in length) at about
50° C. Satisfactory material balances were calculated for all, using an ADL
computer program adapted for the purpose; this program performs balances
around designated sections of the circuit for total weight, total volume,
and the weight of water, sulfur, sulfate, sodium, and calcium.
The pertinent results for the 10 Fredonia limestone runs are summarized
in Table VI-1. In all runs, a 17-20% CaS03 slurry was added to the TOS/SO^
feed solution, with a 1-hour residence time in the reactor and a 2-hour
residence time in the thickener. Solids recycle was maintained from the
thickener to the reactor to achieve the suspended solids concentrations
shown in the table.
650
-------
Table VI-1 EPA 2-LITER REACTOR CONTINUOUS FREDONIA LIMESTONE RUNS—SUMMARY
Feed Solution
Calcium
SOt,
Run
21
22
23
24
25
26
27
28
29
30
Composition, M
[TOS] [SO^l
0.14
0.14
0.14
0.13
0.13
0.04
0.12
0.07
0.05
0.04
0.25
0.52
0.41
0.26
0.26
0.25
0.27
0.27
0.53
0.73
Percent Solids
in Reactor
2.1
2.2
2.9
4.6
6.0
2.7
3.5
3.3
2.0
2.0
Calcium Feed
jStoichipmetry
1.13
1.10
1.19
1.02
1.20
0.88
1.25
0.78
0.72
0.84
Utilization
TOS Basis,3 %
90
64
73
116
106
133
78
110
55
52
\CaSOx/ . . ,
x x 'solids
10.1%
8.5
6.0
6.1
0.2
6.3
4.5
11.1
22.2
8.2
a. 100 x mols TOS Removed/mol Calcium Added
-------
Note that attempts to increase the sulfate content of the cake by decreas-
ing TOS and raising sulfate levels in solution met with only marginal
success. Instead, this approach resulted in poor efficiency for TOS
removal and poor Ca utilization. One interesting trend can be noted by
observing the cake sulfate content in Runs 21, 24, 25, and 27, all con-
ducted with essentially the same feeds and the same nominal stoichiometry,
but differing in percent solids level in the reactor. There appears to
be a downward trend in cake sulfate level as percent solids level (and
thus effective solids residence time) in the reactor is increased.
On the other hand, lowering the TOS and raising the sulfate levels in
solution did render a lime post-treatment more effective in precipitating
sulfate from solution. In Runs 26 through 30, samples of the thickener
overflow were agitated with excess lime for 1 hour, and the sulfate dis-
appearance from solution was monitored. Table VI-2 shows the results for
these runs, in terms of the ability of the system to reject sulfate as a
fraction of total sulfur species precipitated, both for the lime step
only and for the combination of the limestone and lime step (assuming all
limestone step overflow to go on to lime treatment). The lime treatment
solids in Runs 29 and 30 were analyzed and, as the table shows, the check
with sulfate precipitation from solution is quite good.
The calcium and sulfate levels in the limestone reactor overflow and
thickener underflow solutions of Runs 26 through 30 were multiplied for
comparison with the apparent molal solubility product for CaSO^^I^O
(predicted by the method of Kusik and Meissner5) for the ionic strength
involved. Figure VI-1 shows that in all runs (Runs 21 through 30) levels
were not high enough to precipitate CaSOi^^O'as such; however, whenever
crystals of CaSO^^t^O were shaken with samples to test for calcium sulfate
saturation, the product of the calcium and sulfate levels in the resulting
solution rose to the predicted curve almost exactly (the squares on *
Figure VI-1). Figure VI-2 shows similar calculations for the calcium
times the sulfite, compared with the analogous curve for CaS03'l/2H20.
In all runs, CaS03 supersaturation is indicated.
B. DILUTE MODE ALTERNATIVES
The results of this work have been utilized to help us project possible
flow sheets for dilute operation at 35% sulfate regeneration rates. We
have compared
• using limestone, with lime for added sulfate
regeneration, and
• using lime, with sodium carbonate for added
Ca softening.
The firs't of these has been called the "double loop-sulfite softening"
mode because the calcium level leaving the lime reactor/clarifier is
lowered by recycling the solution to the limestone reactor, in which the
precipitation of CaS03 lowers the Ca level.
652
-------
Table VI-2 EPA RUNS - POTENTIAL OF LIME POST-TREATMENT STEP
Ul
LO
LIME STEP ONLY
Run 26
27
28
29
30
LIMESTONE PLUS
Run 26
27
28
29
30
Concent
in Feed
[SO?]
0.2721
0.2845
0.2511
0.5349
0.7389
LIME OVERALL
0.2420a
0.2660
0.2626
0.5208
0.7183
rations Concentrations in ASO^ / SOi, \
Liquor, M Treated Liquor. M I 1
[TOS] [SO?] [TOS] [A(SO£ + TOS)Jliquor \CaSOx/solids
0.0184 0.2925 0.0090 (0)
0.0648 0.2886 0.0180 (0)
0.0418 0.2523 0.0214 (0)
0.0326 0.5125 0.0100 49.8% 41.9%
0.0313 0.6999 0.0106 65.3 64.5
0.03903 0.2925 0.0090 (0)
0.1183 0.2886 0.0180 (0)
0.0712 0.2523 0.0214 17.1%
0.0443 0.5125 0.0100 19.5
0.0401 0.6999 0.0106 38.4
a. These (and following) adjusted for dilution of feed with limestone slurry.
-------
A Reactor Overflow. 1 Hour
O Thickener Underflow, After
2 Hours Additional
U Samples of Both Types, Shaken
w/CaS04 • 2H20 (Done Only
on Runs 26-30)
M, Ionic Strength
Figure VI-1 EPA Limestone Runs 21-30 Extent of CaS04 Under-Saturation
654
-------
10
-3
,r 10
CO
O
C/3
ro
(J
CO
O
CO
o
D
O
<£
>-
o
CO
c
(U
a
,-4
10
-5
10
-6
10
-7
A Reactor Overflow, 1 Hour
O Thickener Underflow, After
2 Hours Additional
234
/u, Ionic Strength
Figure VI-2 EPA Limestone Runs 21-30 Extent of CaSOg Supersaturation
655
-------
Figure VI-3 shows how one would operate a dilute limestone/lime system
to precipitate sulfate to sustain a 35% oxidation level, based on the
results of EPA Run 30 (confirmed by some preliminary ADL pilot plant
runs not reported in this paper). About one-quarter volume of regener-
ated liquor would have to be sent to a lime loop (for sulfate precipita-
tion) for every volume sent to the scrubber. At these conditions, the
material returned to the scrubber would be undersaturated (with respect
to CaSOi^^I^O) by about 150 ppm Ca. The expected utilizations are only
50% for limestone and 80% for lime, based on experimental results. If
one tries to keep up with higher levels of oxidation by sending higher
fractions through the lime loop, the calcium undersaturation ("safety
factor") decreases as the material leaving the lime reactor, which is
saturated in 03504*21120, begins to overpower the sulfite softening
occurring in the limestone reactor; indeed, it does not appear possible
to keep up with more than 40% oxidation without having to accept a
regenerated scrubber liquor with a calcium level approaching within less
than 100 ppm Ca of saturation with respect to gypsum.
Figure VI-4 shows how one might operate an alternative flow sheet using
lime only, relying on sodium makeup with Na2C03 for the necessary Ca
softening. Again a 35% oxidation level has been taken as a basis. The
expected Ca(OH)2 utilization is about 80%. The amount of sodium in the
Na2C03 needed for softening to a level 100 ppm below saturation is seen
from Figure VI-4 to be less than 1% of the weight of the dry waste solids.
Since soluble Na losses in well-washed cakes have been in the range of
about 1%,the normal makeup of Na should introduce more than enough carbon-
ate for softening.
The mode of operation in Figure VI-4 has a much smaller scrubber liquor
feed; while the mode of Figure VI-3 requires 50 £/g mol S02 absorbed,
that of Figure VI-4 requires 9.1 fc/g mol S02 absorbed. The chief uncer-
tainties in the mode of Figure VI-4 are:
1) amount of solids recycle required to produce good settling
and to eliminate CaSOtt supersaturation in the lime reactor,
2) ability to produce good-settling carbonate solids in the
carbonate treatment thickener, and
3) ability to bypass some liquor around the carbonate treat-
ment step to save on the cost of the thickener (see dashed
lines on Figure VI-4).
Based upon the schematic process flow sheets developed from the laboratory
data, rough economic analyses were made comparing the two dilute systems:
limestone/lime with sulfite softening and lime with soda ash softening.
The limestone/lime system involves higher capital costs because of the
higher flow rates through the absorption/regeneration system, but it has
lower raw material costs since half of the regeneration is carried out
656
-------
FUO Evaporation HjO Maku-Up
CnCO,
•= 1.00 g mol
.730 M SO;
.040 TOS
.011 Ca
Limestone
Reactor
Thickener 1
_f\
.700^50.,
.010 TOS
.1 OH
.016 Ca
131'
631
.723.MSO;
.027 TOS
.011 Cn
Limu
Ruactor
Solids
.05 !l niol SO.,
.15 TOS
.50 CO,
1.00 Ca
Basis: ' !1 f"01 of SOp absorbed
35% Oxidation Rate (all assumed in saubbs't for c'lnvfinrncc)
Ca(OH).
- .63g mol
Solids
30 (|riH)l SO,
20 TOS
.30 OH
.63 C.i
Figure VI-3 Dilute Limestone/Lime Schematic
657
-------
HjO Evapoiation H20 Make-Up
Na:CO, .027 g mol
1.0
Ca(OH), = 1.25 g mol
.8
9.11
.700M.SO.,
.01 TOS
1 OH
.013 Ca
9.1C
.739.MSO.,
.081 TOS
.013
Solids
.355 (| mol SO.,
.645 TOS
027 CO ,
.455 OH
1 .25 Ca
Basis: ' t) mol of S02 al)sort)i!d
35";, Oxidation Rate (all assumed in scrubber for convenience!
Figure VI-4 Dilute Lime/Carbonate Softening Schematic
658
-------
using limestone, as opposed to complete regeneration using the more
expensive lime in the alternate dilute system. It was assumed that
both systems would have the same effective cake washing efficiency and
require approximately the same sodium makeup . Because of lower yields
on raw materials, the lime stone /lime system generates more waste for
disposal .
On the basis of the following analysis, it was found that the raw
material cost savings for the limestone/lime system were not sufficient
to offset the expected capital cost increases and increases in waste
disposal costs.
Two cases were examined for the economic analysis:
• a 3.5% S coal application requiring 90% 862 removal, and
• a 1.0% S coal application requiring 80% SOa removal.
In both cases a relatively large cost difference was assumed between
lime and limestone:
• 40 $/ton, CaO
• 5 $/ton, CaCC>3 (ground)
This large cost difference in materials favors the more capital-intensive
sulfite softening approach. A waste disposal cost of
• 10 $/ton waste (dry basis)
was used in all cases. A total annual fixed cost of 20% of the capital
investment (including maintenance costs) was used to calculate annual
costs.
Using this basis, we calculated the minimum capital investment difference
at which the annual capital cost savings of the simpler lime approach
equaled the annual operating cost savings of the limestone/lime approach.
For the high-sulfur coal case, a capital investment savings of about $7/kw
or more would offset operating cost savings. In the low-sulfur coal
situation, the corresponding figure dropped to only $1.8/kw. In each case,
we would expect the actual capital investments for the limestone/lime
system to easily exceed those of the lime system by more than these cal-
culated values. As a result, lime regeneration with soda ash softening
was selected as the basic approach to be investigated for dilute modes
in future pilot operations under this program.
659
-------
VII. SOUTHERN SERVICES PROTOTYPE TEST PROGRAM
A. SYSTEM DESCRIPTION
Combustion Equipment Associates, Inc. (CEA) and ADL have jointly designed
and built a dual alkali prototype system at Gulf Power Company's Scholz
Station. The prototype system is designed to handle a flue gas rate
equivalent to 20 Mw which will be drawn from one of the station's two
40 Mw coal-fired boilers.
The gas scrubbing system consists of a variable throat venturi in series
with an absorption tower that can be operated either as a spray tower, tray
tower (with up to four trays), or simply as a deentrainment separator.
Ducting allows the flue gas to be taken through the newly installed electro-
static precipitator which will be operated at varying efficiency to change
the dust loading to the scrubber system.
The absorber effluent treatment system includes a reactor system, a 40
foot thickener, and a rotary drum vacuum filter equipped with wash sprays.
This equipment has been designed to operate at a full 20-Mw load with the
boiler firing 5%S coal. The reactor system configuration also provides for
the feeding of either hydrated lime or limestone, in either slurry and
dry form.
A schematic showing the process equipment and configuration for this prototype
system is provided in Figure II-2.
660
-------
B. GENERAL OUTLINE
The test program will consist of two phases and will require approximately
one year for completion. The test program will begin early in 1975
following system checkout and startup.
Phase I will last six months and will be directed toward collecting long
term data to evaluate the overall operation of the process, including
system operability and reliability. During this test phase, the Scholz
Station boilers will be operated under "normal" plant conditions with the
electrostatic precipitator in full service. Boiler loads will generally
fluctuate from 50% to 100% of rated capacity and excess air rates will vary
from about 20% to 45%. The sulfur content of the coal fired will range from
2 to 4% S and will vary according to normal coal shipment schedules and
firing practices at the plant.
Throughout Phase I the dual alkali process will be operated in a concentrated
active sodium mode using hydrated lime for absorbent regeneration. The
process will run on a continuous around-the-clock basis with planned shut-
downs coinciding only with boiler down time.
Phase II is designed to obtain shorter term data for a number of different
system operating conditions. For planning purposes, this phase has been
broken down into four test periods, each lasting from four to six weeks.
The exact variables and system parameters to be examined in each of these
periods have not as yet been fully defined; however, it is expected that
the major variables considered in Phase II would include the use of different
coals, the effects of fly ash, and possibly the use of limestone as the
absorbent regeneration reactant. Enough flexibility will be maintained
in this second phase to allow for the modification of test schedules and
test conditions as particular needs arise. As in Phase I, the system will
be operated in the concentrated active sodium mode.
Each of the test phases will involve sufficient sampling and data logging
to support detailed material balances around the entire system and around
specific process sections as desired. The main focus of the data collection
and analysis will be the characterization of the process operation in
terms of:
scrubber efficiency
oxidation rates
sulfate regeneration
solids properties
sodium losses and
lime and other chemicals utilization.
661
-------
GLOSSARY
Active Sodium - Sodium associated with anions involved in S02 absorption
reactions and includes sulfite, bisulfite, hydroxide and carbon-
ate/bicarbonate. Total active sodium concentration is calculated
as follows :
[Na+]active = 2 x ([Na2S03] + [Na2C03]) + [NaHS03] + [NaOH] + [NaHC03]
Active Sodium Capacity - The equivalent amount of SO which can be theo-
retically absorbed by the active sodium. Active sodium capacity
is defined by:
[Na+]active capacity = [Na2S03] + 2 x [Na2C03] + [NaOH] 4- [NaHC03]
Calcium Utilization - The percentage of the calcium in the lime or limestone
which is present in the solid product as a calcium-sulfur salt.
Calcium utilization is defined as:
_ n . ,, . T . . mols (CaSO^ + CaSOu) generated , „.-.„,
Calcium Utilization = - * - 7^7 — c , *' - x 100/£
mol Ca fed
Concentrated Dual Alkali Modes - Modes of operation of the dual alkali
process in which the active sodium concentrations are greater
than 0.15M active sodium.
CSTR — Continuous Feed Stirred Tank Reactor - A well agitated, baffled
reactor vessel having a uniform concentration of species throughout.
At any time the concentrations in the effluent from a CSTR are
equivalent to those within the vessel.
Dilute Dual Alkali Modes - Modes of operation of the dual alkali process
in which the active sodium concentration is less than or equal to
0.15M active sodium.
Sulfate Formation - The oxidation of sulfite. The level of sulfate forma-
tion relative to S02 absorption is given by:
o i c -n • mols S0^= oxidized .. An<,,
Sulfate Formation = - . _, 1 - - - x 100/£
mol S02 removed
662
-------
Sulfate Precipitation - The formation of CaSOi/XH^O in soluble solids.
The level of sulfate precipitation in the overall scheme is
given by the ratio of calcium sulfate to the total calcium-sulfur
salts produced:
Sulfate Precipitation = CaSO,
mol CaSOx
TOS — Total Oxidizable Sulfur - Equivalent to the sum of all sulfite and
bisulfite species.
663
-------
REFERENCES
^Draemel, D.C. and Kaplan, N., An Overview of Sodium-Based Double Alkali
Processes, Proceedings of Flue Gas Desulfurization Symposium—1973,
EPA-650/2-73-038, (December 1973).
2LaMantia, C.R., et al, Sulfur Dioxide Control Process Study—Sodium
Scrubbing with Lime Regeneration, Report to the Illinois Institute
for Environmental Quality, (January 1972).
3LaMantia, C.R., et al, Dual Alkali Process for Sulfur Dioxide Removal,
CEP, 70., 66 (June 1974).
^Borgwardt, R.H., Preliminary Report of Test Results from EPA Alkali
Scrubbing Test Facilities at the TVA Shawnee Power Plant and at Research
Triangle Park, EPA Briefing, Research Triangle Park (December 1973).
5Meissner, H.P. and Kusik, C.L., I&EC Proc. Des. Div. . 12., 205 (1973).
6Marshall, W.L. and Jones, E.V., J. Phys. Chem.. 70(12), 4028 (1966).
7Kaplan, N., An Overview of Dual Alkali Processes for Flue Gas Desul-
furization, EPA Flue Gas Desulfurization Symposium, Atlanta, (November 1974)
8Drehmel, D.C., Wet Limestone Scrubbing of Sulfur Oxides; Limestone
Selection, Proceedings of the Sixty-Fifth Annual APCA Meeting,
June 18-22, 1972.
9Draemel, D., Regeneration Chemistry of Sodium-Based Double-Alkali
Scrubbing Processes, EPA-R2-73-186, Control Systems Division, Research
Triangle Park, N.C. (March 1973).
664
-------
APPLICABLE CONVERSION FACTORS
ENGLISH TO METRIC UNITS
VOLUME
1 gallon = 3.785 liters = 0.003785 m3
1 ft3 = 28.32 liters = 0.02832 m3
DISTANCE
1 ft = 0.3048 meters
TEMPERATURE
°C = (°F - 32) x 5/9
1F° = 5/9C°
665
-------
TECHNICAL REPORT DATA
(Please read /nuructions on the reverse before completing)
1. REPORT NO.
EPA-650/2-74-126-a
2.
3. RECIPIENT'S ACCESS!Of*NO.
4. TITLE AND SUBTITLE
Proceedings: Symposium on Flue Gas Desulfurization-
Atlanta, November 1974
5. REPORT DATE
December 1974
6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
Miscellaneous
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
1AB013: ROAP 21ACX-AA
11. CONTRACT/GRANT NO.
NA
In-House
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Proceedings; 11/4-7/74
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The proceedings document the presentations made during the symposium, which
dealt with the status of flue gas desulfurization technology, both in the U.S. and
abroad. The presentations emphasize process costs, both regenerable and non-
regenerable processes, second generation processes, and byproduct disposal/
utilization. Aim of the symposium was to provide potential users of sulfur oxide
control technology with a current review of progress made in applying processes
for the reduction of sulfur oxide emissions at the full- or semi-commercial scale.
The symposium was the sixth of such EPA-sponsored meetings, "dating back to
1966, when the topic of principal concern was the use of limestone to control sulfur
oxide emissions.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Air Pollution Byproducts
Flue Gases Disposal
Desulfurization Marketing
Sulfur Oxides Sludge
Cost Effectiveness
Regeneration (Engineering)
Air Pollution Control
Stationary Sources
13B
21B
07A, 07D, 05C
07B
14A
8. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
679
Unlimited
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
ENVIRONMENTAL PROTECTION AGENCY
Technical Publications Branch
Office of Administration
Research Triangle Park. N.C. 27711
OFFICIAL BUSINESS
AN EQUAL OPPORTUNITY EMPLOYER
POSTAGE AND PEES PAlO
ENVIRONMENTAL PROTECTION AGENCY
EPA - 335
SPECIAL FOURTH-CLASS RATE
BOOK
Return this sheet if you do NOT wish to receive this material P,
or if change of address is needed I 1. (Indicate change, including
ZIP code.)
PUBLICATION NO. EPA-650/2-74-126-Q
------- |