EPA-600/2-76-136a
May 1976                       Environmental Protection Technology Series
                                         PROCEEDINGS:
     SYMPOSIUM ON  FLUE  GAS DESULFURIZATION
                        NEW ORLEANS,  MARCH 1976
                                               Volume I
                                Industrial Environmental Research Laboratory
                                     Office of Research and Development
                                    U.S. Environmental Protection Agency
                               Research Triangle Park, North Carolina 27711

-------
               RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection  Agency, have been grouped  into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
     1.    Environmental Health Effects Research
     2.    Environmental Protection Technology
     3.    Ecological Research
     4.    Environmental Monitoring
     5.    Socioeconomic Environmental Studies

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.
                    EPA REVIEW NOTICE

This report has been reviewed by  the U. S.  Environmental
Protection Agency, and  approved for publication.  Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency,  nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                        EPA- 600/2- 76-136a

                                        May 1976
                  PROCEEDINGS:

SYMPOSIUM ON  FLUE  GAS DESULFURIZATION

          NEW ORLEANS,  MARCH  1976

                     VOLUME I
              Program Element No. EHE624
               Chairman: Richard D. Stern
  Vice-Chairmen:  Wade H. Ponder and Roger C. Christman

        Industrial Environmental Research Laboratory
         Office of Energy, Minerals, and Industry
            Research Triangle Park, NC 27711


                     Prepared for

      U.S. ENVIRONMENTAL PROTECTION AGENCY
            Office of Research and Development
                 Washington, DC 20460

-------
                              PREFACE




 More than half of all "man-made" sulfur dioxide (S02) is




emitted by electric power plants, and the use of sulfur-containing




fossil fuels to generate electricity is predicted to increase by




50 percent by 1985.  As a result, the development of S02 control




technologies is one of the most important goals of the U.S.




Environmental Protection Agency (EPA).  Flue gas desulfurization




(FGD) is the most promising technique for control of S02 that will be




available for widespread application to fossil fuel-fired electric




power plants for at least the next decade.









 The Industrial Environmental Research Laboratory - Research




Triangle Park  (IERL-RTP) of EPA's Office of Research and




Development sponsors  symposia for the transfer of information




regarding FGD  research, development and application activities




with the objective of further accelerating the development and




commercialization of  this technology.  These symposia  provide




an  opportunity for users and developers to discuss their




experiences and the  status  of development and application  of




FGD technology.








 The March  1976 symposium addressed  full-scale FGD process




applications  in the  U.S. and Japan as well as laboratory,  pilot,
                                iii

-------
and prototype research and development efforts.   The symposium




also provided an opportunity for the announcement of data and




results which were previously unreported or not  widely publicized.




The economics of FGD and the disposal, utilization, and marketing




of FGD system by-products were also discussed.  The symposium




papers were presented by a cross-section of those concerned




with FGD including users, government and private developers, and




vendors.  The electric utility industry—the principal user of FGD-




participated extensively in the symposium program.  More




than 650 people attended the symposium.









 These Proceedings are comprised of copies of the participating




authors' papers as received.  As supplies permit, copies of the




Proceedings are available free of  charge and may be obtained by




contacting lERL-RTP's Technical Information Coordinator,




Environmental Protection Agency, Research Triangle  Park, North




Carolina  27711.
                               iv

-------
                              CONTENTS

TITLE                                                             PAGE

                              VOLUME  I

OPENING SESSION  	                                 ,
Keynote Address - SULFUR OXIDE CONTROL AND ELECTRICITY
PRODUCTION
     R. E. Train, Environmental Protection Agency,
     Washington, D. C	
REMARKS
     S. J. Gage, Environmental Protection Agency,
     Washington, D. C	
STATUS OF FLUE GAS DESULFURIZATION SYSTEMS IN THE UNITED
STATES
     T. W. Devitt, G. A. Isaacs, and B. A. Laseke, PEDCo-
     Environmental Specialists, Inc., Cincinnati, Ohio 	       13

STATUS OF FLUE GAS DESULFURIZATION AND SIMULTANEOUS REMOVAL
OF SO  AND N0x IN JAPAN
     J. Ando,xChuo University, Tokyo, Japan 	       53

FLUE GAS DESULFURIZATION ECONOMICS
     G. G. McGlamery, H. L. Faucett, R. L. Torstrick, and
     L. J. Henson, Tennessee Valley Authority, Muscle
     Shoals, Alabama 	       79

STATUS OF THE EPRI FLUE GAS DESULFURIZATION DEVELOPMENT
PROGRAM
     L. W. Nannen and K. E. Yeager, Electric Power Research
     Institute, Palo Alto,  California 	      101

NON-REGENERABLE PROCESSES SESSION 	      115

IERL-RTP SCRUBBER STUDIES RELATED TO FORCED OXIDATION
     R. H. Borgwardt, Environmental Protection Agency,
     Research Triangle Park, North Carolina 	      117

RESULTS OF MIST ELIMINATION AND ALKALI UTILIZATION TESTING
AT THE EPA ALKALI SCRUBBING TEST FACILITY
     M. Epstein, H.  N.  Head, S. C.  Wang, and D.  A. Burbank,
     Bechtel Corporation, San Francisco, California 	      145
                               v

-------
TITLE                                                             PAGE

DUQUESNE LIGHT COMPANY ELRAMA AND PHILLIPS POWER STATIONS
LIME SCRUBBING FACILITIES
     R. G. Knight and S. L. Pernick, Duquesne Light
     Company, Pittsburgh, Pennsylvania 	       205

OPERATIONAL STATUS AND PERFORMANCE OF THE COMMONWEALTH EDISON
COMPANY WILL COUNTY LIMESTONE SCRUBBER
     W. G. Stober, Commonwealth Edison Company, Chicago,
     .1 llinois 	      219

MHI FLUE GAS DESULFURIZATION SYSTEMS APPLIED TO SEVERAL
EMISSION SOURCES
     M. Hirai, M. Atsukawa, A. Tatani, and K. Kondo,
     Mitsubishi Heavy Industries, Ltd., Tokyo, Japan 	      249

STATUS OF FLUE GAS DESULFURIZATION USING ALKALINE FLY ASH
FROM WESTERN COALS
     H. M. Ness, E. A. Sondreal, and P. H. Tufte, U. S.
     Energy Research and Development Administration, Grand
     Forks, North Dakota	      269

RESULTS OF THE 170 MW TEST MODULES PROGRAM, MOHAVE GENERATING
STATION,  SOUTHERN CALIFORNIA EDISON COMPANY
     A. Weir, Jr., L. T. Papay, D. G. Jones, J. M. Johnson,
     and  W. C. Martin, Southern California Edison Company,
     Rosemead, California  	      325

LA CYGNE  STATION UNIT NO. 1 WET SCRUBBER OPERATING EXPERIENCE
     C. F. McDaniel, Kansas City Power and Light Company,
     La Cygne, Kansas 	      355

RECENT SCRUBBER EXPERIENCE AT THE LAWRENCE ENERGY CENTER,
THE  KANSAS POWER AND LIGHT COMPANY
     D. M. Miller, Kansas Power and Light Company, Topeka,
     Kansas  	      373

INTRODUCTION TO DOUBLE ALKALI FLUE GAS DESULFURIZATION
TECHNOLOGY
     N. Kaplan, Environmental Protection Agency, Research
     Triangle Park, North Carolina  	      38.7

OPERATING EXPERIENCE—CEA/ADL DUAL ALKALI PROTOTYPE SYSTEM
AT GULF POWER/SOUTHERN  SERVICES,  INC.
     C. R. LaMantia and R. R. Lunt, Arthur D.  Little,  Inc.,
     Cambridge, Massachusetts; R. E. Rush, Southern Services,
     Inc., Birmingham, Alabama; T. M. Frank, Combustion
     Equipment Associates, Inc., New York, New York; and
     N. Kaplan, Environmental Protection Agency, Research
     Triangle Park, North Carolina  	      423

                                  vi

-------
TITLE                                                             PAGE

THE FMC CONCENTRATED DOUBLE-ALKALI PROCESS
     L. K.  Legatski, K. E. Johnson, and L. Y. Lee,
     FMC Corporation, Glen Ellyn, Illinois 	       471

OPERATING EXPERIENCE WITH THE ZURN DOUBLE ALKALI FLUE GAS
DESULFURIZATION PROCESS
     P. M.  Lewis,  Zurn Industries, Inc., Birmingham,
     Alabama 	       503

KUREHA FLUE GAS DESULFURIZATION "SODIUM ACETATE-GYPSUM
PROCESS"
     S. Saito,, T.  Morita, and S. Suzuki, Kureha Chemical
     Industry Company, Ltd.,  Tokyo, Japan 	       515

THE BUELL DOUBLE-ALKALI S02 CONTROL PROCESS
     H. E.  Bloss,  Buell-Envirotech, Lebanon,  Pennsylvania;
     J. Wilhelm, EIMCO-Envirotech, Salt Lake  City, Utah;
     and W. J.  Holhut, Central Illinois Public Service
     Company, Springfield, Illinois 	       545
                             VOLUME II

BY-PRODUCT DISPOSAL/UTILIZATION SESSION 	       xi

STATUS AND PLANS FOR WASTE DISPOSAL FROM UTILITY APPLICATIONS
OF FLUE GAS DESULFURIZATION SYSTEMS
     J. L. Crowe, Tennessee Valley Authority, Chattanooga,
     Tennessee;  and H.  W.  Elder, Tennessee Valley Authority,
     Muscle Shoals, Alabama 	       565

RESEARCH AND DEVELOPMENT FOR CONTROL OF WASTE AND WATER
POLLUTION FROM FLUE GAS CLEANING SYSTEMS
     J. W. Jones, Environmental Protection Agency,
     Research Triangle  Park, North Carolina 	       579

FLUE GAS CLEANING WASTE DISPOSAL - EPA SHAWNEE FIELD
EVALUATION
     J. Rossoff and R.  C.  Rossi, The Aerospace Corporation,
     El Segundo, California 	       605

CHEMICAL FIXATION OF FGD SLUDGES - PHYSICAL AND CHEMICAL
PROPERTIES
     J. L. Mahloch, U.  S.  Army Engineer Waterways Experiment
     Station, Vicksburg, Mississippi 	       627
                               VII

-------
TITLE                                                             PAGE

POTENTIAL UTILIZATION OF CONTROLLED SO  EMISSIONS FROM
POWER PLANTS IN EASTERN UNITED STATES X
     J. I. Bucy, J. L. Nevins, P. A. Corrigan, and A. G.
     Melicks, Tennessee Valley Authority, Muscle Shoals,
     Alabama	      647
REGENERABLE PROCESSES SESSION 	      701

STATUS OF DEMONSTRATION OF THE WELLMAN-LORD/ALLIED FGD SYSTEM
AT NIPSCO'S D. H. MITCHELL GENERATING STATION
   PART I:  BACKGROUND AND OVERVIEW
     E. L. Mann, Northern Indiana Public Service Company,
     Michigan City, Indiana; and R. C. Christman, TRW
     Environmental Engineering Division, Vienna, Virginia . ..      703
   PART II:  'CURRENT STATUS AND OPERATING PLAN
     S. F. Lakatos, A. W. Michener, Jr., and W. D. Hunter,
     Jr., Allied Chemical Corporation, Morristown,
     New Jersey	      709

AN UPDATE OF THE WELLMAN-LORD FLUE GAS DESULFURIZATION
PROCESS
     R. I. Pedroso, Davy Powergas, Inc., Lakeland, Florida ..      719

SUMMARY OF OPERATIONS OF THE CHEMICO-BASIC MgO FGD SYSTEM
AT THE PEPCO DICKERSON GENERATING STATION
     R. B. Taylor and P. R. Gambarani, Chemico Air Pollution
     Control Co., New York, New York; and D. Erdman, Potomac
     Electric Power Company, Washington, D. C	      735

MAGNESIUM OXIDE SCRUBBING AT PHILADELPHIA ELECTRIC'S
EDDYSTONE STATION
     J. A. Gille, Philadelphia Electric Company,
     Philadelphia, Pennsylvania 	      749

INTERIM REPORT ON CHIYODA THOROUGHBRED 101 COAL APPLICATION
PLANT AT GULF POWER'S SCHOLZ PLANT
     R. B. Dakan, Chiyoda International Corporation, Seattle,
     Washington; and R. A. Edwards and R. E. Rush, Southern
     Services, Inc., Birmingham, Alabama 	      761
ADVANCED PROCESSES 	       785
                               Vlll

-------
TITLE                                                             PAGE

STATUS AND ECONOMICS OF THE ATOMICS INTERNATIONAL AQUEOUS
CARBONATE FLUE GAS DESULFURIZATION PROCESS
     D. C.  Gehri and R. D. Oldenkamp, Atomics International,
     Canoga Park, California 	      787

ENERGY REQUIREMENTS FOR SHELL FGD PROCESS
     F. A.  Vicari and J. B. Pohlenz, UOP Inc., Des Plaines,
     Illinois 	      817

THE DOWA'S BASIC ALUMINUM SULFATE-GYPSUM FLUE GAS
DESULFURIZATION PROCESS
     Y. Yamamichi and J. Nagao, The Dowa Mining Co., Ltd.,
     Okayama, Japan 	      833

CITRATE PROCESS FOR FLUE GAS DESULFURIZATION, A STATUS REPORT
     W. I.  Nissen,  D.  A. Elkins, and W. A. McKinney, Bureau
     of Mines, Salt Lake City,  Utah 	      843

APCI/IFP REGENERATIVE FGD AMMONIA SCRUBBING PROCESS
     C. E.  Ennis, Catalytic, Inc., Philadelphia, Pennsylvania      865

BF DRY ADSORPTION SYSTEM 	      877
   PART I:   FW-BF GULF POWER DEMONSTRATION UNIT INTERIM
   RESULTS
     J. Strum and W.  F. Bischoff, Foster Wheeler Energy
     Corporation, Livingston,  New Jersey; and R. E. Rush,
     Southern Services, Inc.,  Birmingham, Alabama 	      879
   PART II:   BF-STEAG DEMONSTRATION UNIT OPERATIONAL
   EXPERIENCE AND PERFORMANCE
     K. Knoblauch,  Bergbau-Forschung GMBH, West Germany;
     and K.  Goldschmidt, STEAG  Aktengesellschaft, West
     Germany 	      899

THE CONSOL FGD PROCESS
     R. T.  Struck,  E.  Gorin, and W.  E.  Clark, Conoco Coal
     Development Company, Library, Pennsylvania 	      913
UNPRESENTED PAPERS 	       931

INFORMATION TRANSFER PROGRAM
     T.  W.  Devitt and T.  C.  Ponder,  Jr.,  PEDCo-Environmental
     Specialists, Inc.,  Cincinnati,  Ohio  	       933
                               IX

-------
TITLE
                                                                  PAGE
CATALYTIC/WESTVACO DESULFURIZATION PROCESS PROTOTYPE
DEMONSTRATION PROGRAM
     Catalytic, Inc., Philadelphia, Pennsylvania 	      945

THE SPRING-NOBEL HOECHST PROCESS FOR SULFUR DIOXIDE
RECOVERY FROM STACK GASES
     W. H. Stark, A. A. Syme, and J. C.  H. Chu,  Spring
     Chemicals, Ltd., Toronto, Canada 	      981

STARTUP OF AMERICAN AIR FILTER'S SULFUR DIOXIDE REMOVAL
SYSTEM AT THE KENTUCKY UTILITIES COMPANY'S GREEN RIVER
STATION
     A. H. Berst, American Air Filter Co., Inc., Louisville,
     Kentucky; and J. Reisinger, Kentucky Utilities Co.,
     Central City, Kentucky  	      991

TCA SPHERE DEVELOPMENT AND EVALUATION
     P. Sorenson, N. E. Takvoryan, and R. J. Jaworowski,
     UOP, Inc., Darien, Connecticut  	      999

-------
                         OPENING SESSION
Chairman:       Richard D. Stern
               Chief, Process Technology Branch
               Industrial Environmental Research Laboratory
               U.S. Environmental Protection Agency
               Research Triangle Park, North Carolina

-------
                KEYNOTE ADDRESS
SULFUR OXIDE CONTROL AND ELECTRICITY PRODUCTION
        Russell E. Train, Administrator

        Environmental Protection Agency
              401 M Street, S.W.
           Washington, D.C.   20460

-------
   I appreciate the opportunity to present some of  my thoughts  regarding
sulfur oxide control as it relates to electricity production.   As you
know, the legislative driving force for air pollution clean-up  efforts is
the Clean Air Act,  as amended in 1970.  This Act has  had,  and will continue
to have, significant impact on the burning of fossil  fuels in electric
generating facilities.

   The Act called for the establishment of national ambient air quality
standards which set limitations on the ambient concentrations of various
pollutants—SOX, NO , CO, oxidants, and particulates—to minimize adverse
effects on human health and welfare.

   In order to" achieve the ambient standards, the Act called for the
development of State Implementation Plans (SIPs) which identify abatement
measures to be taken within the several regions of  each state in order to
meet the ambient standards.  SIPs may call for, among other things,
restrictions on the emission of pollutants from various classes of facilities.
These implementation plans are tailored to the particular  air quality problems
of each state, and the emission standards that they call for can be more or
less stringent in different parts of the country.  Working together, EPA and
the states have now completed the development of these implementation plans.
The Act also set a rather stringent timetable for the achievement of ambient
standards; in most cases, such standards were to be met by mid-1975.

   Further, the Act calls for the New Source Performance Standards (NSPS) ,
which would require all new industrial sources to install the best
demonstrated pollution control technology regardless  of the source's
location.  This provision is designed to minimize the shift of  new industrial
facilities into regions of the country where emission control requirements
are  less  stringent.   It is interesting to note that,  in some states, the
emission  limits set under the SIP's are more strict than the emission limits
set by  the NSPS.

   How  does all this  affect the electrical utilities?  How do utilities
relate  to the  Clean Air Act and sulfur oxide emissions?  Using the fossil
fuels available to them — coal,  oil and natural gas  — power plants have
produced  electricity  in vast amounts.  Unfortunately, they have also produced
vast amounts of an undesirable byproduct — sulfur oxides.

   As you well know,  sulfur oxides and their derivatives present a serious
threat  to human health.   It is an established  fact that serious respiratory
disease and illness—and  even death—result from human exposure to these
compounds.  In turn,  such health  effects result in loss of work and pay by
those so  affected.  The Congress  declared that, in setting clean air standards
to protect human health—such as  the primary sulfur oxide standards—cost and
difficulty of  control were not to be issues in establishing either the standards
themselves or  the  compliance schedules with which to meet them.

   In 1971, the Environmental Protection Agency carried out the Congressional
mandate and promulgated primary  (health-related) and secondary  (welfare-related)

-------
standards for the regulated air pollutants, including sulfur oxides.   With
approximately 970 fossil-fueled power plants in this country emitting 17 million
tons or roughly 60 percent of all sulfur oxides in 1972, State Implementation
Plans focused on those plants that were in the more highly polluted regions.
The SIPs called for varying degrees of SOX emission control for each plant—
ranging from no control to control levels more stringent than the New Source
Performance Standards.

   The EPA has consistently represented that the most reasonable approach
to meeting S02 ambient standards is by means of an emission control approach
rathar than by dilution methods.  We believe this approach is particularly
appropriate since recent data indicate that the oxidation products of S02,
i.e., acidic sulfates, may cause more serious health problems than S09
itself.                                                              l

   At the present time, we do not have a detailed understanding of the
mechanism by which S02 is converted to sulfates; however, it appears to
involve the presence of other pollutants in the atmosphere and can occur
over distances greater than 100 Km.

   The problem of the regulation of S02 emissions to protect human health,
in the light of this new evidence, was recently thoroughly reviewed by the
National Academy of Sciences.  The Academy noted significant deficiences
in information available, both on the conversion of S02 to sulfates and on
the health effects associated with various sulfur components in the atmosphere.
Nevertheless, the Academy found the evidence to be persuasive enough to
justify minimizing or preventing increases in emissions of S02.

   EPA has a number of studies underway to shed more light on the
atmospheric transformation of S02 to sulfates and on the health threat
of sulfates.  The Agency is now undertaking a five-year research program
to significantly reduce uncertainty in our understanding of the sulfate
problem.

   Much has happened over the five years since the enactment of the
Clean Air amendments.  We have had an energy crisis that has caused us to
take a long, hard look at where we get our energy sources and how we utilize
that energy.  When the Clean Air Act amendments were first enacted most people
assumed that our supply of fossil fuels would remain abundant and that the
problems of high sulfur emissions could be solved by burning low sulfur fuels,
cleaning high sulfur fuels or by a combination of the two.  Before the Arab
embargo, many plants had made costly conversions to allow the burning of
cheap oil and further take advantage of available lower sulfur content.
Now all that is changed.

   The embargo is over for now and imported oil is flowing into the country
at nearly the rate it was before the embargo.  We have learned of the dangers of

-------
becoming too dependent on foreign sources of fuel and that, increasingly, we
must rely upon our own resources and our own technology.  We have also learned
that attainment of ambient air quality standards by mid-1975 was infeasible
in many instances.

   We are in the fortunate position of having within our national borders
sixty percent of the world's coal.  It seems only natural that we should seek
to take advantage of this fact.  But too many eyes are looking west to the
low sulfur variety that can be strip mined, ignoring both the environmental
consequences of such stripping and the ready availability of abundant high
sulfur coals in the East.  As you all do, EPA recognizes that under current
conditions, supplies of low sulfur fuels are insufficient to go around, that
some give and take is going to have to take place if we are going to have both
sufficient power and air that is safe to breath.  That is why new legislative
proposals were submitted by the President to the Congress during February,
1975, as limited amendments to the 1970 Clean Air Act.

   While we continue to feel that the Clean Air Act on the whole is
a reasonable and needed response to the severe problems of air pollution,
we also recognize that it perhaps lacks a certain flexibility.  In
addition, we feel that the President's proposals reflect a realistic concern
for social impacts and cost effectiveness.  Rather than weakening the air
quality standards as so many have feared, in fact, the proposed amendments
strengthen the Act by injecting a more common sense and defensible approach
toward achieving those standards.

   Let me summarize some of the most important elements of the proposed
amendments to the Clean Air Act relative to stationary sources.

   Compliance Schedule Extensions - this amendment would authorize
compliance schedule extensions for certain isolated plants until January 1,
1985, at the latest, to install and operate scrubber systems or acquire low
sulfur coal contracts.  Under the compliance schedule, these plants could
temporarily employ intermittent control systems if it could be demonstrated
that they are reliable and enforceable.  EPA estimates that there are
approximately 18 to 70 plants which could use intermittent control prior to
installation of permanent controls.

   Waiver for Technology Innovations - this amendment would authorize,
where the EPA Administrator approves in advance, a waiver of compliance
with NSPS to encourage innovative and experimental control technology,
provided primary air quality standards are met.  This amendment would have
important implications for second generation FGD and coal cleaning processes.

   Assessment of Civil Penalties - This amendment proposes providing
civil penalties of up to $25,000 per day of violation by stationary
sources.

-------
   Enforcement Order Extensions - this change would clarify the
Administrator's authority to issue enforceable compliance orders which
extend beyond the statutory air quality standards attainment date.

   Prevention of Significant Deterioration - EPA did not propose major
modifications; other Federal agencies suggested that the total
provision be deleted from the act.

   Primary Particulate Standard - this amendment would allow the EPA
Administrator to grant certain regions of the country additional time to
meet the primary air standards for particulate matter.

   New Source and Hazardous Emission Standards - this change explicitly
authorizes EPA to set design or equipment standards for certain pollutant/
source combinations if,  and for as long as, it is not feasible to set customary
emission or performance standards.

   I believe these amendments would provide the Environmental Protection
Agency with the flexibility needed to achieve compliance with air quality
standards without weakening the basic thrust of the statute.  I have been
insistent that, in developing these needed changes, we in no way undermine
the fundamental policies and objectives of the act.

   The Congress has acted deliberately in considering these proposals
and I expect that within the next three months we will see legislation
generated on this subject.  It is too early to say what the particulars of
the final legislation will be, or how closely they will reflect EPA's
recommendations.  However, based on Congressional deliberations to date,
I do not expect any significant weakening in the spirit of the Clean Air
Act.

   As you are well aware, EPA feels that the best available technology
for controlling sulfur oxides when low-sulfur fuels are unavailable are
the flue gas desulfurization systems which are the subject of this Symposium.
It would be presumptuous of me to lecture a large body of experts on flue
gas desulfurization on the viability and competitivenss of such processes.
However, I will say that the Agency is keeping abreast of alternative SO
emission control technologies such as physical and chemical coal cleaning,
fluidized bed combustion, solvent refined coal and low-BTU coal gasification.
It is clear to us that,  with the exception of physical coal cleaning which
can allow achievement of emission standards in a limited number of circumstances,
only flue gas desulfurization presents a technology which can achieve required
emission limitations consistent with the time frame mandated by the present or
proposed changes to the  Clean Air Act.  Chemical coal cleaning, fluidized bed
combustion, solvent refined coal and low-BTU gasification will be potentially
competitive with FGD no  earlier than the early-1980's.

-------
   We recognize that FGD technology is not yet a mature technology with all
problems resolved to everyone's satisfaction.  However, we believe this
technology can and is playing an important role in minimizing SO  emissions
from utility and large industrial sources.  I strongly support utility and
government efforts to utilize and improve FGD technology; I am proud of the
Office of Research and Development's Industrial Environmental Research
Laboratory's sponsorship of FGD Symposia to communicate recent developments
this fast-moving technology.  There is no question that further effort can
lower costs, and produce more desirable end products.  I trust that with
mature cooperation from the industrial and government sectors we can optimizi
and further utilize  this most important control technology.

   At this time, 45  different utilities, General Motors, and Air Force
and TVA have all committed to the use of FGD technology for one or more of
their fossil fuel plants.  Large and small utilities alike are represented.
We find their  commitment to cleaner air as well as to plentiful power
admirable, especially when we consider that the capital and operating cost
increments are not insignificant.  I feel strongly that those utilities that
lack  the  incentive to develop and install effective sulfur oxide control
equipment and  that expend great quantities of their resources to avoid such
control measures will in the end have done a grave disservice to the public
and to  themselves.

   Common sense dictates that the utilities must provide the power to
keep  America running and common sense tells us that they must use all of
the energy sources that are available to them, including high sulfur coal.
And  I can assure you that the American people will not  tolerate either
a diminution of necessary electrical power or a deterioration of their healtl
and general welfare  caused by generation of that power.  It should be clear
to all  that we cannot afford either of these alternatives.  Adequate power ii
achievable.  Clean air  is obtainable.

-------
                     REMARKS
Dr. Stephen Gage,  Deputy Assistant Administrator

    Office of Energy,  Minerals, and Industry
         Environmental Protection Agency
               401 M Street,  S.W.
            Washington, D.C.    20460

-------
     Before I hand the microphone over to Dick Stern who will make some
opening comments and introduce the first speaker, I would like to add
a few remarks of my own.  First, I should say that, in at least two
instances, the Congressional committees considering the Clean Air Act
amendments are taking a harder line than that embodied in the
Administration's proposals.  On the "significant deterioration" issue,
it appears that the Congressional position is one of requiring quite
stringent protection of those areas of the country where air quality
is currently excellent.  Also, the Congressional committees appear to
be moving toward a position of requiring Best Available Control Technolo
(BACT) for all new plants, regardless of the quality of the fuel
consumed.  However, it is too early to tell what changes will be made
in the House committee considering the bill, or what changes will be
made on the floor in both houses, and on the conference committee.

     Those of you who attended the last EPA-sponsored FGD symposium
during November, 1974 in Atlanta may remember some of my comments as
keynote speaker.  One of the points I made then was that alternative
technologies to flue gas desulfurization, particularly low-BTU gasificat
and fluidized bed combustion, cannot make commercial impact before the
1980's.  Unfortunately, nothing has happened in the last year and a half
to change my assessment.

     As Mr. Train indicated in his keynote address, flue gas desulfuriza
represents the only major alternative to the burning of scarce, low-sulfi
fuels in the next 8-10 years.  However, I would like to mention one deve
ment which I believe has the potential to make an important near-term
impact on SO  control from coal burning from plants.

     General Public Utilities Corporation (GPU) will install a physical
coal cleaning plant at their 650 megawatt plant in Homer-City, Pennsylva
This coal cleaning process utilizes a two-stage coal washing process of
coal; one stream will have low sulfur content (0.8% or less) and the seo
stream will contain about 2.2% sulfur.  Coal from both streams will be
suitable for direct firing in the Homer-City boilers (both new and exist
units) without further cleaning to meet state emission standards.  EPA
plans to participate in this important demonstration program.  Such an
approach can in certain cases achieve SIP standards or NSPS alone, or mo
generally can be used in conjunction with FGD in what may be the most
cost-effective approach for many applications.

     At the previous Symposium, I also discussed three Research, Develop
and Demonstration initiatives that EPA's ORD planned, to further advance
state-of-the-art of FGD.  First, I indicated that ORD would conduct an a
lime and limestone test program on our flexible test facility at TVA's S
plant.  I am happy to report that this advanced program was initiated an
                                  10

-------
provided many important insights into lime and limestone scrubbing.   As
you will hear in two papers this afternoon:  Operating conditions and mist
eliminator configurations capable of long term operation at high velocities
in a closed loop mode of operation have been identified.  Operating
conditions have been identified which have shown enhanced limestone
utilization (less sludge production).  Three commercial sludge fixation
processes are being evaluated at a prototype scale.

     Second, to provide an alternative to lime/limestone throwaway systems,
I indicated that OKD would initiate a full-scale demonstration of a
Double Alkali system.  Although the evaluation process has moved more
slowly than planned, I am happy to state that we have received excellent
proposals and are in the final stages of selecting utility/vendor team for final
negotiations.  It is interesting to note that the total costs for each
of the three proposals ranged from $60-75/Kw.  Such systems seem to
represent a promising alternative to lime scrubbing systems.

     Third, to provide an  alternative to sulfur-producing synthetic
fuel processes,  I indicated  that ORD would initiate a full-scale demon-
stration of a regenerable  FGD system which produces by-product sulfur.
Again, the  selection process has moved slowly, but we have received several
viable proposals and are quite  close to making our final selection.  The
selected system  will be  demonstrated at a  100 Mwe coal-fired boiler
burning high-sulfur coal.

     We hope  to  make our selections  for both demonstrations public within
a few weeks and  trust  that these  programs  will further  advance the state-
of-the-art  of FGD technology.

     Finally, I  should mention  that,  in  the  past year,  The Office of
Research and  Development has signed a  Memorandum of Understanding with
the  Electric  Power  Research Institute  in order to  facilitate sharing of
technical  information  and  cooperation  on Research, Development,  and
Demonstration projects.  We are very hopeful that  this  cooperative research
agreement  will  provide the basis for a creative approach  to solving  the
remaining  pollution control problems facing  the electric  power industry.

      It  has been my pleasure to appear before you  today to present
Mr.  Train's remarks and make some comments of my  own.   Obviously,  the
Symposium is designed  to be another success  in  the series of  successful
FGD symposia.  The  Industrial Environmental  Research  Laboratory  is to
be congratulated for sponsoring this series  and  for contributing so
much to the development of FGD technology.
                                     11

-------
                        STATUS  OF FLUE  GAS
                      DESULFURIZATION SYSTEMS
                       IN  THE UNITED STATES
             Timothy W.  Devitt,  Dr.  Gerald A.  Isaacs,
                       and Bernard A.  Laseke

               PEDCo-Environmental Specialists,  Inc.
                     Suite 13,  Atkinson Square
                     Cincinnati, Ohio   45246
ABSTRACT

     PEDCo conducts a bimonthly survey for the Industrial Environmental
Research Laboratory/RTF and the Division of Stationary Source Enforcement
of EPA on the status of flue gas desulfurization (FGD) systems.  This
paper presents an  overview of the results of this survey including
data on the number and total equivalent magawatts of FGD systems installed,
under construction,, and planned for U.S. utility application.  Information
is also presented on trends in process application and on the relia-
bility of operating systems.

     Approximately 109 FGD systems with a total equivalent rating
of 42,128 MW are either operational, under construction, or planned.
S0_ removal efficiencies are generally in the range of 80 to 90 percent.
Operability or reliability of several of the more recent systems is
approximately 90 percent although many of the older, first generation
systems continue to operate with lower values.
                              13

-------
             STATUS OF FLUE GAS DESULFURIZATION
               TECHNOLOGY IN THE UNITED STATES

     In March 1974, PEDCo-Environmental 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 has two  outputs.  The first is a
series of reports on individual FGD facilities.  These
reports present values for pertinent process designs and
operating parameters and describe the operability of the
unit from start-up to the time of the plant survey.  To date
13 installations have been visited  and another 16 will be
inspected by July 1976.
     The second study output is a series of bimonthly re-
ports listing the systems that are operational, under con-
struction or planned, and describing the operability of the
existing systems during the preceeding two month period.
     As of January 1976, the following flue gas systems were
either operational or under construction in the United
States: limestone, lime, activated carbon, double alkali,
WeiIman-Lord/Allied Chemical, Chiyoda Thorobred 101, sodium
carbonate, and magnesium oxide.  In addition,  there are
several other systems in the pilot plant stage.
     As shown in Table 1, as of January 1976 there were 21
operational FGD systems on utility-size boilers,  20 units
under construction and 67 units planned.  The planned units
include those where a contract' has been awarded or a letter
of intent has been signed and units for which the utility is
presently requesting or evaluating bids.
                              14

-------
         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
21
20

10
10
7
40
108
MW
3,796
7,026

3,761
3,911
3,837
19,797
42,128
     By late 1976, approximately 10,000 MW of FGD capacity
will be installed of which 40 percent will be retrofit
applications and 60 percent will be on new power plants.  By
1980, approximately 40,000 MW of FGD capacity will be in-
stalled of which 35 and 65 percent will be retrofit and new
installations respectively.  Figure 1 illustrates the growth
of retrofit and new applications through 1980.
     The different process applications in 1976 and 1980 are
presented in Table 2.  As is obvious from this table, the
preponderance of units, both now and anticipated through
1980, is lime or limestone based systems.  Currently, lime
based systems account for 37 percent, on an equivalent
magawatt basis, and will account for approximately 21 per-
cent in 1980.  Limestone based systems account for 44 per-
cent in 1976 and will account for 30 percent  in 1980.
Regenerable processes will remain at approximately 6 to 8
                              15

-------
                        FGD  INSTALLATIONS
3:


o
CD
CD
to
s:
(X
UJ
Q.
O
     1968 69  70  71  72  73 74  75  76  77  78 79  80 81  82

                              YEAR

       Figure 1.  Operating capacity  through 1982.
                              16

-------
percent of the total during this period although the picture
is clouded by the large number of planned systems for which
the data is not available concerning process selection.

        Table 2.  FLUE GAS DESULFURIZATION PROCESSES
Process
Lime
Limestone
Lime/ limes tone
Activated carbon
Catox
Dual alkali
Magox
Sodium carbonate
Dilute acid
Wei Iman- Lord
Not selected
Total
1976
3,442
4,077
30
20
110
20
365
375
23
115
630
9,207
1980
7,716
11,221
1,740
20
110
20
1,091
375
23
1,830
13,223
37,369
      SO9  removal  efficiencies  for the operating units range
 from approximately  40  to  90 percent and particulate removal
 efficiencies generally are above 99 percent  for those units
 designed  for particulate  removal.  SO- removal efficiencies
 of  approximately  98 percent have been attained during test
 runs at various facilities, although most systems are being
 designed  to operate in the 80  to 90 percent  efficiency
                               17

-------
range.  FGD systems have been installed on units varying in
size from about 30 MW to over 800 MW and on both low (0.4 to
1.0 percent) and high (6.0 percent) sulfur coal applica-
tions.  The non-regenerable systems have used a variety of
methods for sludge disposal including ponding, dewatering by
vacuum filtration and landfilling, and fixation followed by
landfilling.  The regenerable systems have been limited to
date to recovering sulfuric acid, although a sulfur recovery
unit is being installed.  Table 3 presents some of the
values for  key design and operating parameters for the
operational FGD systems.
     The only quantifiable process trend is that lime and
limestone systems continue to predominate for all three FGD
categories  (i.e., operational, under construction, planned).
Other apparent trends are:
      (1)  More systems are being installed to meet state
          standards that are more stringent than NSPS
          levels.
      (2)  More systems are being installed on low sulfur
          coal applications than on high sulfur coal appli-
          cations.
      (3)  There is a tendency from wet fan applications to
          dry fan applications.
      (4)  Somewhat more attention is being paid to sparing
          of critical components  (e.g., slurry pumps) al-
          though less than would be anticipated.
      (5)  There is a trend away from indirect reheat using
          tube bundles in the flue gas stream.
      (6)  Dry particulate collection systems  (e.g., a high
          efficiency mechanical collector or an electro-
          static precipitator) ahead of the FGD system are
          displacing particulate scrubber systems.
                              18

-------
Table 3.  DESIGN AND OPERATING  DATA FOR




         SELECTED FGD SYSTEMS
Item
No.
1
'2
3
4
S
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
FGD
process
Lime
scrubbing
Lime
scrubbing
Lime
scrubbing
Lime
scrubbing
Lime
scrubbing
Lime
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Limestone
scrubbing
Lime/Lime-
stone
scrubbing
Lime/Lime-
stone
scrubbing
Magnesium
oxide
scrubbing
Sodium
carbonate
scrubbing
Sodium
carbonate
scrubbing
Double
alkali
Dilute
acid
Power Station
Utility
Duquesne
Light Co.
Duguesne
Light Co.
Kentucky
Utilities
Louisville
Gas & Elec.
Montana
Power Co.
S. Calif.
Edison Co.
Arizona
Pu. Serv.
Commonwealth
Edison Co.
Kansas City
Pwr & Light
Kansas City
Pwr & Light
Kansas City
Pwr & Light
Kansas Pwr
& Light
Kansas Pwr
& Light
Key West
Utility Brd
Public Serv
of Colorado
S. Calif.
Edison Co.
Tennessee
Valley
Authority
Tennessee
Valley
Authority
Philadelphia
Elec. Co.
Nevada
Power Co.
Nevad.
Power Co.
Gulf Power
Co.
Gulf Power
Co.
Unit
Elrama
Phillips
Green River
No. 1 & 2
Paddy1 s Run
Colstrip 1
Mohave
No. 2
Cholla
No. 1
Will County
No. 1
Hawthorn
No. 3
Hawthorn
No. 4
La Cygne
No. 1
Lawrence
No. 4
Lawrence
No. 5
Stock
Island
Valmont
No. 5
Mohave
No. 1
Shawnee
No. 10A
Shawnee
No. 10B
Eddystone
No. 1
Reid Gardner
No. 1
Reid Gardner
No. 2
Scholr No. 1
Scholz
Nos. 1 I 2
Size
(MW)
510
410
64
65
360
160
115
167
140
100
820
125
400
37
50
170
10
10
120
125
125
20
23
No.
modules
5
S
2
2

1
2
2
2
2
7
2
8
2
1
1
1
1
1
1
1
1
1
New/
retrofit
Retrofit
Retrofit
Retrofit
Retrofit
New
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
New
Retrofit
New
New
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
Retrofit
Start up
date
10/75
7/73
9/75
4/73
10/75
11/73
10/73
2/72
11/72
8/72
2/73
12/68
11/71
10/72
10/74
11/74
4/72
4/72
9/75
4/74
12/73
2/75
3/75
By
pass
yes
yes

no

yes
yes
yes
yes
yes
no
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
y«.
Fuel Characteristic*
Type
coal
coal
coal
coal
coal
coal
coal
coal
coal
coal
coal
coal
coal
oil
coal/
gas
coal
coal
coal
coal
coal
coal
coal
coal
Btu/lb

11,000
11,000
11,500
8,800
11,500
12,146
9,463
9,500-
1^,500
9,800-
11,500
9,000
10,000
12,000
10,000
12,000

10,780
11,500


12,100
12,450
12,450
12,400
12,400
% S
1.0-2.8
1.0-2.8
3.8
3.5-4.0
0.8
0.4
0.44
2.1
0.6-3.0
0.6-3.0
5.2
0.6-3.75
0.6-3.75
2.4-2.75
0.72
0.5-0.8
2.9
2.9
2.3.
0.5-1.0
0.5-1.0
3.0-S.O
3.0-S.O
                    19

-------
Table 3 (continued).  DESIGN AND OPERATING DATA




           FOR SELECTED FGD SYSTEMS
Item
No.
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17


18


19


20

21

22

23

Particulate
control
device


mech, esp.
venturi
mech.

ESP



ESP

mech.

ESP,
venturi
absorber
tower
absorber
tower
venturi

absorber
tower
absorber
tower
mech.

absorber
tower
ESP

absorber
tower

venturi


mech., ESP,
ventri-rod.
scrubber
mech..
venturi
mech..
venturi
ESP

ESP,
venturi
Control System
Efficiency
Particulate


99

99.7



99

N/A

99.7

98

99

99

98

99

99

90

90

N/A







99.9


99

99

99.7

99.7

so2


60

80



40

75-98

85

82

70

70

72

75

65

85

85

N./A

75-88


80-98


90


85

85

N/A

90+

Gas reheat
Type


oil-
fired


gas-
fired


hot
air
steam

steam

steam

steam

steam/
air
steam

steam

none



steam

fuel
oil
fired
fuel
oil
fired
fuel
oil
fired
hot
air
'hot
air
fuel oil
fired
fuel oil
fired
Mode


direct



direct



indirect

indirect

indirect

indirect

indirect

indirect
/direct
indirect

indirect

none



direct

direct


direct


direct


direct

direct

direct

direct

By-product
Type
stabilized
sludge
stabilized
sludge
stabilized
sludge
unstabilized
sludge
unstabilized
sludge
stabilized
sludge
unstabilized
sludge
stabilized
sludge
unstabilized
sludge
unstabilized
sludge
unstabilized
sludge
unstabilized
sludge
unstabilized
sludge


unstabilized
sludge








magnesium
sulfite

spent
liquor
spent
liquor
filter cake

gypsum

Disposal
pond

haulaway

unlined
pond
landfill

unlined
pond
lined
pond
unlined
pond
clay lined
pond
unlined
pond
unlined
pond
unlined
pond
unlined
pond
unlined
pond
settling
pond
lined
pond
converted
FGD System
costs $/kW


84

62.5

57

50

N/A

57

103

19

19

43

N/A

N/A

21.6



N/A
to aggregate
settling
pond

settling
pond

acid plant


settling
pond
settling
pond
lined
pond
lined
pond
N/A


N/A


193


44

44

N/A

130

                        20

-------
     Several parameters have been suggested to quantify the
viability of FGD systems.  Various terms such as "availa-
bility," "reliability," and "operability" have been used
indiscriminately in reference to these parameters so that a
certain degree of confusion exists in defining these terms
when they are used.  These parameters are defined and
briefly discussed below.
     Availability  (Parameter 1):  Hours the FGD system was
available for operation  (whether operated or not) divided by
hours in period, expressed as a percentage.  This parameter
tends to overassess the viability of the FGD system because
it does not penalize for election not to operate the system
when it could have been operated.  Boiler downtime tends to
increase the magnitude of the parameter because FGD failures
are not included during  such periods.
     Operability  (Parameter 2):  Hours  the FGD  system was
operated divided by boiler operating hours in the period,
expressed as a percentage.  This parameter indicates the
degree  to which the FGD  system is actually used, relative  to
boiler  operating  time.   The principal objection is  that  the
parameter does not reflect  the extent of exertion on the FGD
system.  That  is,  the magnitude  of  the  parameter has little
or no correlation with FGD  system operating  time.  Another
objection is  that the  parameter  is  penalized when options
are  exercised  to  not use the FGD system in periods  when  the
system  is operable.
     Utilization  Factor  (Parameter  3):   Hours  that  the FGD
system  operated divided  by total hours  in  period.   This
parameter  is  a relative  stress factor for  the  FGD system.
 It  is not  a complete measure of FGD system viability because
 the  parameter can be  strongly  influenced by  conditions  that

-------
are external to the FGD system.   For example,  infrequent
boiler operation'will lower the  value of the parameter
although the FGD system may be highly dependable in its
particular application.
     Another parameter (Parameter 4) , has been defined as
the hours the FGD system was operated divided by the hours
the FGD system was called upon to operate, expressed as a
percentage.  This parameter has  been suggested in order not
to penalize the FGD system for elected outages, e.g.,
periods when the FGD system could have been run but was not
run because of chemical shortages, lack of manpower, short
duration boiler operations, and  the like.  The main problem
in using this formula is the concise determination whether
or not the system was "called upon to operate" during a
given time period.
     Thus, no single ratio yields a completely desirable
parameter to assess the degree of operability of an FGD
system.  It does seem that Parameters 2 and 3 together form
a useful index of FGD system operability.  Parameter 2 is a
measure of responsiveness of the FGD system to the boiler's
requirements, and Parameter 3 is a measure of the intensity
at which the FGD system is operated.  Perhaps "availability"
or "reliability" should be quantified by two parameters
instead of one, e.g., 78 percent operability/65 percent
utilization.  Figure 2 illustrates the difference in values
for the various parameters using data from the experimental
operation of the Mohave 170 MW horizontal unit FGD system
from January 6, 1974 to February 9, 1975.
     In most of our reports to date, we have used the
operability parameter, FGD system operating hours over
boiler operating hours, because  the data required to cal-
culate this parameter are the most readily available.  The
                             22

-------
               PARAMETER 1   AVAILABILITY
         	PARAMETER 2   OPERABILITY
         	PARAMETER 3   UTILIZATION
               PARAMETER 4.
  JAN.
 1974
FEB.
MAR.   APR.  MAY   JUN.  JUL.  AUG.  SEPT. OCT.

                        MONTHS
NOV.   DEC.  JAN.   FEB.  MAR.
          1975
Figure 2.   Comparison of FGD viability  parameters for the Mojave Station

                       of Southern California Edison.

-------
operability of units has improved over the past 2 years as
both the number of units and the operating experience has
increased.  Figures 3, 4 and 5 illustrate the operability of
some of the more successful units by month of operation.
The average operability of the units range from about 80 to
95 percent depending upon the system and the averaging
period.
     A utility industry survey was conducted recently by the
Edison Electric Intitute to determine the projected opara-
bility/reliability  and cost for FGD systems.  Any utility
known to  have an FGD system that was operational, under
construction or planned was contacted by EEI and requested
to  complete a questionnaire describing their FGD system.
Responses were received for forty-three systems and were
analyzed  by EEI and PEDCo.  For utilities with units under
construction or in  the planning stage, most estimated the
reliability of the  systems would be 90 percent or greater,
which is  essentially  that being attained by some of the more
recent  operating units, as described above.
      The  cost  side  of  the survey is also interesting.  As
anticipated,  the reported costs cover a broad range due to
both site-specific  factors and the lack of uniformity with
respect to items  included in  the cost estimates.
      The  responses  covered 30 utilities, 32 plants and  68
boilers for a total capacity  of  32,120 MW.  The  reported
costs ranged  from 33  to 197  $/kW with an average of  $94/kW
 (standard deviation = $40/kW).  Of these,  22 were  lime  or
limestone based  systems.  The cost for these  systems  ranged
from 34 to 116 $/kW with  an  average of  $78/kW  (standard
deviation =  $27/kW).
      When these  costs were  adjusted  to  reflect  a common year
dollar  value  (1975),  removal  of  costs  for  particulate emis-
                               24

-------





   60
 vS so
   «o



     J f MAMJJA SOdDJ F   AMJJASONO



                                            ,
                                           JrMTMJJASOII
                                                                              AW, CHORRA HO. 1

                                                                              KCML. LiCrSHE W. 1
                                                             MAMJJASOND
             1974
     Figure 3.  Operability
Arizona  Public Service,  and
                   1974              1975
of the FGD  systems at  the  Cholla Power  Plant,

the LaCygne Station, Kansas  City Power  and Light.

-------
N)
         100



          SO*

       I
       i
          30 ft

          •

            J  J A 5 3 N


               1975
j j A  SON'      A ft"TT~A""S"a"N"6~i"?"ftl"A""'i"j~j"A" S 0 N o j f >•

   1975      1973              1974               1975
                                                                                                   lf POWER CO.. CHirOOA lo



                                                                                                   F POWCR co.. DUAL ALKALI


                                                                                                   I E PADDY'S RUN NO. 6
                     Figure 4.   Operability  of the FGD  systems at  the Scholz  Station of

           Gulf Power Company  and the Paddy's Run Station  of Louisville Gas and  Electric,

-------



IUU
:
-..
•
o 60
0
X
-
-•
/SCRUBBER
1
-•
.
= 50
Of
LU
	 1
3
CO
1*
: 40
-
_J
CO
i
LiJ
0.
o
** 30



•
n
-
-
-
	 :


j!
—


-
-- '














'•'•'• '.•

















































^-^


















.«




"""i
















1

II
jFMANJJASONDJ

           1975
























•
'












•










'















~



-







J  F M ANJJASONOJ
          1975
                                                            NEVADA POWER,
                                                            REID GARDNER NO. 1

                                                            NEVADA POWER,
                                                            REID GARDNER NO. 2
 Figure  5.   Operability  of the FGD systems  at the

    Reid Gardner  Station  of Nevada Power Company.
                            :

-------
sion control,  addition of indirect cost items and adequate
costs for either sludge disposal or regeneration systems,
the range narrowed considerably.  The adjusted costs for all
systems with sufficient data (30 systems), ranged from 50 to
205 $/kW with an average of $91/kW (standard deviation =
$34/kW).  Both the upper end of the range and the average
costs are high because of an exceptionally high cost re-
ported by a single power company for a prototype FGD system;
the utility stated that their reported values should be
considered "upper limits."  Excluding the costs reported by
that company, the costs range from 50 to 137 $/kW with an
average value of $85/kW.  Adjusted costs for lime and lime-
stone based systems reported by nineteen utilities ranged
from 50 to $88/kW with an average of $70/kW  (standard
deviation = $10/kW).
     In summary, as operating experience increases and. as
this experience is fed back to  the design phase, systems are
coming on-line that can perform with high reliability.  A
table  from the most recent bimonthly report  is presented in
Appendix A, summarizing the status of the various FGD
systems.
                               28

-------
                         APPENDIX A

                     STATUS OF FLUE GAS

                  DESULFURIZATION SYSTEMS3
a Appendix A is Table 2 from the PEDCo November - December
  Bimonthly Report on the Status of Flue Gas Desulfuriza-
  tion Systems.
                             29

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FGD SYSTEMS DURING
12/79
                       CURRENT MONTH
l.D. NUMBER    1
ALABAMA ELECTRIC COOP
TOMBIGOEE NO. 2
  225  MM - NEW
COAL   0.6- 1.5 PERCENT SULFUR
PEABODY ENGINEERING
LIMESTONE SCRUBBING
STARTUP   3/78
    PEABOOY ENGINEERING WAS AWARDED CONTRACT TO INSTALL FGQ SYSTEM"ON
    TOM8I6BEE NO,2, SYSTEM WILL BE LIMESTONE SCRUBBING.
1.0. NUMBER    2
ALABAMA ELECTRIC COOP
TOMBIGBEE NO, 3
  225  MM • NEW
COAL   0,8- 1.5 PERCENT SULFUR
PEABOOY ENGINEERING
LIMESTONE SCRUBBING
STARTUP   3/79
    PEABOOY ENGINEERING WAS AWARDED CONTRACT TO INSTALL FGD SYSTEM ON
    TOMBIGBEE NO.3. SYSTEM WILL BE LIMESTONE SCRUBBING.
1.0, NUMBER    9
ALLEGHENY POWER SYSTEM
PLEASANTS NO. 1
  625  MM - NEW
COAL   2.0- 5.0 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   3/79
    NEW STARTUP DATE IS SET FOR MARCH 1979.  FGO SYSTEM IS IN THE PLANNING
    STAGE.  ANTICIPATE LECISION ON THE CONTRACT SOON,
I.D, NUMBER    •»
ALLEGHENY POWER SYSTEM
PLEASANT* NO, 2
  625  MU • NEW
COAL   2,0- 5.0 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   3/80
    NEW STARTUP DATE  IS SET FOR MAKCH  1980.  FGO SYSTEM  IS  IN  THE PLANNING
    STAGE.  ANTICIPATE DECISION ON  THE CONTRACT SOON.
1.0. NUMBER    5
AKIZONA ELECTRIC POWER COOP
APACHE NO 2
  200  MU - NEW
COAL  0.5- 0,8 PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP   6/79
    LETTER OF INTENT  TO  INSTALL  AN  FGO  SYSTEM  WAS  ISSUED  TO RESEARCH  COTTRELL.
    ENGINEERING DESIGN WORK  IS PROCEEDING.

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FGO SYSTEMS DURING
12/75
                       CUKKENT MONTH
1.0. NUMBER    6
AHI20NA ELECTRIC POWER COOP
APACHE NO 5
  20t>  MW - NEW
COAL   0.5- 0.8 PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP   6/79
    LETTER OF INTENT TO INSTALL AN FGD SYSTEM WAS ISSUED TO RESEARCH COTTRELL.
    ENGINEERING DESIGN WORK IS PROCEEDING,
1.0. NUMBER    7
ARIZONA PUBLIC SERVICE
CHOLLA NO 1
  115  MW - RETROFIT
COAL   O.H«,-1 PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP  10/75
    KEFEK TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
    FGU SYSTEM HAS BEEN OPERATIONAL SINCE NOVEMBER*1973. AVAILABILITY FOR
    MOUULi A HAS AVERAGED OVER 95 PERCENT FOR THE PAST 6 MONTHS. AVAILABILITY
    FOR MODULE B HAS AVERAGED OVER 6fl PERCENT FOR THE PAST 6 MONTHS. AVAIL-
    ABILITY FOR MODULE A WAS 100 PERCENT IN NOVEMBER AND DECEMBER. AVAIL-
    ABILITY FOR MODULE B WAS 80 PERCENT IN NOVEM8EK AND 100 PERCENT IN DECEM-
    BER.
1.0. NUMBER    6
ARIZONA PUBLIC SERVICE
CHOLLA NO 2
  250  MW • NEW
COAL   O.^-l PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP   6/77
    RESEARCH COTTRELL WAS AWARDED A CONTRACT TO INSTALL FGD SYSTEM ON
    CHOLLA NO. 2, AND IS PRESENTLY WORKING ON THE DESIGN PHASE OF THIS
    PROJECT. ENGINEERING IS SUBSTANTIALLY COMPLETE.  CONSTRUCTION IS
    UNDERWAY.
I.D. NUMBER    9
ARIZONA PUBLIC SERVICE
FOUK CORNERS NO. •»
  755  MW - RETROFIT
COALi 0.7 - 0.75* SULFUR 
NOT SELECTED
NOT SELECTED
STARTUP   O/ 0
     THIS 755 MW UNIT WILL BE CONTROLLED PENDING THE OUTCOME OF TESTS TO BE
     MADE ON THE PROTOTYPE 160 MW FGO MODULE NOW BEING  INSTALLED ON UNIT 5.
I.D. NUMBER   10
ARIZONA PUBLIC SERVICE
FOUR CORNERS NO. 5A
  160  MW - RETROFIT
COAL* 0,7 - 0.75, SULFUR (AVG)
SCE
LIME SCRUBBING
STARTUP  11/75
     THIS PROTOTYPE  160 MW  SYSTEM  IS  BEING MOVED  FROM  THE MOHAVE STATION OF
     SOUTHERN CALIFORNIA EDISON, WHERE THE SYSTEM HAS  BEEN TESTED PREVIOUSLY.
     IT WILL HANDLE  ABOUT 20X OF THE  EXHAUST GAS  FROM  UNIT 5 RATED AT 755MW
     (NET).  ENGINEERING IS BOX COMPLETE AND CONSTRUCTION IS 55* COMPLETE.
     PARTICULATE AND S02 EFFICIENCIES FOK THIS U IIT ARE TO BE DETERMINED, AND
     UNIT RELIABILITY WILL  BE EVALUATED.  IT HAS  NOT YET BEEN DETERMINED IF
     THIS INSTALLATION WILL BE PERMANENT.

-------
     COMPANY

     POWER STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING
12/75
                                                         CURRENT MONTH
    I.D. NUMBER   11
    ARIZONA PUBLIC SERVICE
    FOUR CORNERS NO. t>B
      595  MW • RETROFIT
    COALi 0.7 • 0.75* SULFUR 
    NOT SELECTED
    NOT SELECTED
    STARTUP   O/ 0
                                      THE BALANCE OF THIS 755* MU UNIT WILL BE CONTROLLED PENDING THE OUTCOME
                                      OF TESTS TO BE MADE ON THE PROTOTYPE 160 MW FGO MODULE NOW BEING INSTALLED
                                      NO START-UP DATE HAS BEEN SCHEOULt-0.
    I.D. NUMBER   12
    BASIN ELECTRIC POWER COOP
    MISSOURI BASIN NO 1
      550  MW - NEW
    COAL
    NOT SELECTED
    RtGCNCRABLE NOT SELECTED
    STARTUP   1/60
                                      BECAUSE OF THE STRICT WYOMING EMISSION STANDARDS OF 0.2LBS PER MILLION
                                      BTU,LOW SULFUR COAL ALONE WILL NOT MEET STANDARDS.  BASIN ELECTRIC IS NOW
                                      INVESTIGATING VARIOUS REGENERA8LE PROCESSES.
N)
I.D. NUMBER   13
BASIN ELECTRIC POWER COOP
MISSOURI BASIN NO 2
  550  MW - NEW
COAL
NOT SELECTED
RtGENERABLE NOT SELECTED
STARTUP   fc/80
                                          BECAUSE OF THE STRICT WYOMING EMISSION STANDARDS OF 0.2LBS PER MILLION
                                          BTU«LOW SULFUR COAL ALONE WILL NOT MEET STANDARDS.  BASIN ELECTRIC IS NOW
                                          INVESTIGATING VARIOUS REGENERABLC PROCESSES.
    I.D. NUMBER   14
    BASIN ELECTRIC POWER COOP
    MISSOURI BASIN NO 3
      550  MW - NEW
    COAL
    NOT SELECTED
    RtGENEHABLE NOT SELECTED
    STARTUP   t/BS
                                      BECAUSE OF THE STRICT WYOMING EMISSION STANDARDS OF 0.2LBS PER MILLION
                                      BTUiLOW SULFUR COAL ALONE WILL NOT MEET STANDARDS.  BASIN ELECTRIC IS NOW
                                      INVESTIGATING VARIOUS REGENERABLE PROCESSES.
    1.0. NUMBER   15
    CENTRAL ILLINOIS LIGHT CO.
    DUCK CRECK NO.l
      100  MW - NEW
    COAL 2.5-3.0 PERCENT SULFUR
    RILEY STOKER/ENVIKONEERING
    LIMESTONE SCRUBBING
    STARTUP   6/76
                                      THE UNIT IS NOW UNDER CONSTRUCTION AND IS SCHEDULED FOR COMPLETION IN
                                      APRIL 1976.  STATION CAPACITY IS tOOMW. BUT FLUE GAS FROM ONLY 100 MW
                                      WILL BE TREATED INITIALLY.

-------
 COMPANY
           TABLE 2
STATUS OF FCO SYSTEMS DURING  12/75
 POWER STATION
                       CURRENT MONTH
1.0, NUMBER   16
CENTRAL ILLINOIS LIGHT CO.
DUCK CREEK NO.2
  HOO  MM - NEW
COAL 2.5-3.0 PERCENT SULFUR
NOT SELECTED
LIMESTONE SCRUBBING
STARTUP   1/81
    THIS PROPOSED LIMESTONE UNIT IS DUE TO BEGIN OPERATION JANUARY 1981
I.D, NUMBER   17
CENTRAL ILLINOIS LIGHT CO,
E.O.EDWAHOS NO,3
  357  MW - RETROFIT
COAL 2.5 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   7/79
    LOW SULFUH COAL TO BE USED FROM 6/76 UNTIL STARTUP,
1.0. NUMBER   18
CENTRAL ILLINOIS PUBLIC SERV
NEWTON NO.l
  600  Mw - NEW
COAL 2.B-3.2 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP  12/77
    ALL BIDS HAVE BEEN RECEIVED AND ARE NOW BEING EVALUATED. THE PROCESSES
    BEING CONSIUEREU ARE LIME/LIMESTONE SCRUBBING AND DOUBLE ALKALI.
    A DECISION ON THE PROCESS AND VENDOR SHOULD BE MADE BY FEBRUARY 1«1976
I.D. NUMBER   19
CINCINNATI GAS * ELECTRIC CO.
EAST BEIND NO 2
  600  MW - NEW
COAL
NOT SELECTED
NOT SELECTED
STARTUP   1/80
    UTILITY WILL GO OUT FOR BIDS IN EARLY 1976. THE COMPANY IS PRESENTLY OB-
    TAINING CONSTRUCTION AND OPERATIONS PERMITS. STARTUP DATE HAS BEEN MOVED
    BACK TO JANUARY 1980. THE COAL SOURCE FOR THIS UNIT HAS NOT YET BEEN SE-
    LECTED.
 I.D. NUMBER   20
 CINCINNATI GAS & ELECTRIC co,
 MIAMI FORT NO 8
   300  MM - NEU
 COAL   1.3 PERCENT SULFUR
 NOT SELECTED
 LIME/LIMESTONE SCRUBBING
 STARTUP   l/7ft
    THE UTILITY HAS SIGNED A CONTRACT FOR THE PURCHASE OF LOW SULFUR COAL
    (APPROXIMATELY 1.3 PERCENT SULFUR). THE COAL BED FOR THE NO.e UNIT IS NOW
    IN THE PROCESS OF BEING REDESIGNED TO ACCEPT THE LOWER SULFUR COAL. CG3E
    NOW PLANS TO INSTALL AN FGD UNIT TREATING 60 PERCENT OF THE TOTAL 500 MW
    CAPACITY, NEW BIOS ARE NOW BEING REQUESTED AND ACCEPTED ON THIS SMALLER
    UNIT.

-------
 COMPANY
           TABLE 2
STATUS OF FGO SYSTEMS DURING  12/75
 POWER STATION
                                                         CURRENT  MONTH
1.0. NUMBER   21
COLUMBUS a SOUTHERN OHIO ELEC.
CONESVILLE NO 5
  HOO  MW • NEW
COAL  H.5 • i»,9 PERCENT SULFUR
UNIVERSAL OIL PRODUCTS
LIME SCRUBBING
STARTUP   6/76
    THE UTILHY SIGNED LONG-TERM CONTRACTS WITH OKAVO FUK THE PURCHASE OF
    THIOSORBIC LIME AND WITH IUCS FOR THE PURCHASE OF SLUDGE FIXATIVES. THE
    BOILER ANU FGD SYSTEM ARE STILL UNDER CONSTRUCTION. THIS MINE MOUTH PLANT
    PLANS TO BURN COAL WITH 17 PERCENT ASH CONTENT AND 
-------
    COMPANY

    POWER  STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING  12/75

                                                         CURRENT MONTH
   1.0.  NUMBER    26
   DETROIT  EDISON
   ST.CLAIR NO  6
     180 MW -  RETROFIT
   COAL   3.7 PERCENT  SULFUR
   PEABODY  ENGINEERING
   LIMESTONE SCRUBBING
   STARTUP    1/76
                                      A FOUKTH HOT FLUE GAS KUN OF 5H7 MRS, DONATION WAS TERMINATED ON DECEMBER
                                      29 DUE TO EXCESSIVE VIBRATION OF THE ID BOOSTER FAN. DUKING THE COURSE OF
                                      THE RUN MINOR OUTAGES OCCURRED PRIOR TO TERMINATION. THE OUTAGES WERE
                                      CAUSED BY BOILER SHUTDOWN DUE TO GENERATOR OVERHAULi INTERRUPTION OF THE
                                      FUEL OIL SUPPLY TO THE REHE.ATER, ANU REPAIR OF THE PACKING" OF ONE OF THE
                                      SCRUBBER PUMPS. THE SYSTEM OPERATED AT 69 PERCENT OF THE DESIGN CAPACITY.
                                      S02 REMOVAL REMAINED AT THE 90 PERCENT LEVEL FOR 1-3 PERCENT SULFUR COAL.
                                      THE SYSTEM IS SCHEDULED FOR RESTART IN MID-JANUARY.
   I.D.  NUMBER   27
   DU8UESf4E LIGHT
   ELRAMA
     510  MW - RETROFIT
   COAL   1.0 - 2.6 PERCENT SULFUR
   CHEMICO
   LIME  SCRUBBING
   STARTUP  10/75
                                      EQUIPMENT INSTALLATIONS AND MODIFICATIONS WERE COMPLETED. THE SYSTEM
                                      STARTED UP IN THE EARLY PART OF OCTOBER 1975. THE INITIAL SHAKEDOWN AND
                                      DEBUGGING PHASE PROCEEDED WITH FLUE GAS FROM ONE 100 MW BOILER BEING
                                      TREATED BY ONE SCRUBBER. THE UTILITY HAS CHECKED OUT H OF THE 5 SCRUBBER
                                      MODULES TO DATE. THE FIFTH MODULE WAS INOPERATIVE DUE TO MAlNTANENCE TO
                                      THE RUBBER-LINED RECYCLE PUMPS. THE IUCS METHOD OF SLUDGE FIXATION IS
                                      USED AT THIS STATION.
en
I.D. NUMBER   28
DUOUESNE LIGHT
PHILLIPS
  >»10  MW - RETROFIT
COAL   1.0- 2.6 PERCENT SULFUR
CHC"ICO
LIME SCRUBBING
STARTUP   7/73
                                         REFER TO BACKGROUND INFORMATION SECTION IN TABLE 9 OF THIS REPORT.  THIS
                                         FGO SYSTEM HAS BEEN OPERATIONAL SINCE JULY, 1973.
   I.D. NUMBER   29
   GENERAL MOTORS
   CHEVROLET PARMA 1 2 3 & <»
      32  MW - RETROFIT
   COAL   2.5 PERCENT SULFUR
   KOCH
   DOUBLE ALKALI
   STARTUP   3/7"»
                                      REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
                                      FGD SYSTEM HAS BEEN OPERATIONAL SINCE MARCH 1971. SYSTEM AVAILABILITY HAS
                                      AVERAGED OVER 68 PERCENT DURING THE PAST «t MONTHS, SYSTEM AVAILABILITY
                                      WAS IN EXCESS OF 56 PERCENT DURING THE REPORT PERIOD.
   1.0. NUMBER   30
   GULF POWER CO.
   SCHOLZ NO. 1A
      20  MW - RETROFIT
   COAL   3.0 PERCENT SULFUR
   AOL/COMBUSTION EQUIP ASSOCIATE
   DOUBLE ALKALI
   STARTUP   2/75
                                      REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
                                      FGD SYSTEM HAS BEEN OPERATIONAL SINCE FEBRUARY,1975. AVAILABILITY HAS
                                      AVERAGED OVER 63 PERCENT DURING THE PAST 
-------
 COMPANY
           TAULE 2
STATUS OF FGO SYSTEMS DURING  12/73
 HOWtR STATION                                           CUKHENT MONTH
I,D, NUMBER   31
GULF POWER CO.
SCHOLZ NO. -2A
   20  NW • RETROFIT
COAL   3.0 PERCENT SULFUR
FOSTER WHEELER
ACTIVATED CARBON
STARTUP   2/76.
    THE ADSORPTION AND REGENERATION SECTIONS INITIAL SHAKEDOWN Ru« ON FLUE
    GAS BEGAN ON AUGUST 11 AND WAS CONTINUED FOR A PEHIOO OF TEN DAYS.  FLOW
    OF FLUE GAS THROUGH THE ADSOKL'ER WAS CONTINUOUS DURING THAT PEHIOU. PRES-
    SUKE UHOP ACROSS THE ADSORBER WAS WLLL BELOW DESIGN LEVEL AND SU2 REMOVAL
    WAS HISH. THE Kt&LNERATION SECTION WAS OPERATIONAL ABOUT 60 PEHC£NT DUR-
    ING THE PERIOD. THE RLSOX PORTION OF THE SYSTEM RAN FOR 5 DAYS IN THE
    LATTER PA«T OF OCTOBER, THE KESOX HUN WAS SUCCESSFUL. THE UTILITY is NOW
    IN THE PROCESS OF MAKING MODIFICATIONS FOR A SCHEDULED FEBRUARY STARTUP,
1.0. NUMBER   32 '
GULF POWER CO.
SCHOLZ NOS. IB & 28
   23  MW • RETROFIT
COAL   9.0 PERCENT SULFUR
CHIYODA INTERNATIONAL
THOROUGHBRED 101
STARTUP   3/75
    REFEK TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF REPORT. THIS FGD
    SYSTEM HAS BEEN OPERATIONAL SINCE MARCH.1975. AVAILABILITY HAS AVLRAGED
    85 PERCENT FOR THE PAST <» MONTHS, AVAILABILITY WAS ALMOST 100 PERCENT IN
    NOVEMBER AND 90 PERCENT IN DECEMBER. S02 REMOVAL EFFICIENCY IS REPORTED
    TO BE IN EXCESS OF 95 PERCENT.
I.D. NUMBER   33
INDIANAPOLIS POWER * LIGHT CO,
PETERSBURG NO 3
  530  MW • NEW
COAL  3.0-3.5 PERCENT SULFUR
UNIVERSAL OIL PRODUCTS
LIMESTONE. SCRUBBING
STARTUP   H/77
    UNIVERSAL OIL PRODUCTS HAS BEEN SELECTED TO BUILD THE LIMESTONE SCRUBBING
    SYSTEM. THL ENTIRE J30MW UNIT WILL BE TREATED BY FOUR MODULES. CONSTRUC-
    TION HAS BEGUN ON ALL FOUR MODULES.  WORK ON THE FOUNDATION IS PROCEED-
    ING.
I.D. NUMBER   3H
KANSAS CITY POWER * LIGHT
HAWTHORN NO 3
  11*0  MW - RETROFIT
COAL   0.6- 3.0 PERCENT SULFUR
COMBUSTION ENGINEERING
LIMESTONE INJECTION 4WET SCRUB
STARTUP  11/72
    REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
    F6J SYSTEM HAS BEEN OPtRATIONAL SINCE NOVEMUER.1972. AVAILABILITY FOR
    BOTH MODULES WAS ZERO DURING THE PAST TWO REPORTING MONTHS. THL BOILER
    AND SCKUBBING MOUULES WERE SHUTDOWN DUKING THE MONTHS OF NOVEMBER AND DE-
    CEMBER FOK A SCHEDULED TURBINE OytKHAUL. THE SYSTEM IS SCHEDULED To GO
    BACK ON-LINE AGAIN SOMETIME IN JANUARY 1976. SYSTEM AVAILABILITY FOR
    MODULES A AND B DURING THE PERIOD OF SEPTEMBER-OCTOBER 1975 WAS 67 AND
    52 PERCENT RESPECTIVELY.
1.0. NUMBER   35
KANSAS CITY POWER ft LIGHT
HAWTHORN NO •»
  100  MW - RETROFIT
COAL   0.6- 3.0 PERCENT SULFUR
COMBUSTION ENGINEERING
LIMESTONE INJECTION «WET SCRUB
STARTUP   e/72
    REFEK TO BACKGROUND INFORMATION SECTION  IN  TABLE  3 OF THIS REPORT.  FGD
    SYSTEM HAS BEEN IN OPERATION SINCE AUGUST,1972. AVAILABILITY FUR MODULES
    A ANU B HAS AVERAGED 0 AND 58 PERCENT RESPECTIVELY FOR THE PAST TWO
    MONTHS. MODULE A WAS UNAVAILABLE  TRUUGHOUT  THE REPORT PLRIOO.  AVAIL-
    ABILITY FOR MODULE B WAS 99 PERCENT  IN NOVEMBER AND  16 PERCENT IN DECEM-
    BER. AVERAGE AVAILABILITY FOR MODULES A  ANU B DURING THE PAST  H MONTHS
    HAS BEEN 28 AND 65 PERCENT RESPECTIVELY.

-------
 COMPANY

 POWER STATION
                                                TABLE 2
                                     STATUS OF FGO SYSTEMS DURING  12/75

                                                            CURRENT MONTH
1.0. NUMBER   36
KANSAS CITIT POWER ft LIGHT
LA CYGNE NO 1
  620  MM • NEW
COAL   5,2 PERCENT SULFUR
BABCOCK £ WILCOX
LIMESTONE SCRUBBING
STARTUP   2/73
                                         REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPOKT, THIS
                                         FSO SYSTEM HAS BEEN OPERATIONAL SINCE DECEMBER.i960* AVERAGE AVAILABILITY
                                         FOK ALL 7 MODULES HAS BEEN 6<» PERCENT FOR THE PAST 8 MONTHS. AVERAGE SYS-
                                         TEM AVAILABILITY WAS 91 PERCENT IN NOVEMBER AND 66 PERCENT IN DECEMBER.
1.0. NUMBER   37
KANSAS POWER * LIGHT
JEFFERY NO. 1
  700  MW . NEW
COAL
COMBUSTION ENGINEERINS
LIMLSTONE SCRUBBING
STARTUP   6/78
                                         WORK ON THE FGD SYSTEM IS IN THE EAKLY PLANNING STAGE.  COMBUSTION
                                         ENGINEERING IS LOOKING INTO THE S02 EMISSION CONTROL NEEDS OF THIS PLANT
                                         A\U IS PREPARING A PROPOSAL TO KPL.  CONTROL OF PARTICULATES WILL BE
                                         ACCOMPLISHED IN TWO ESP UNITS TO BE INSTALLED UPSTREAM OF THE EIGHT
                                         FGD MODULES.
   I.D. NUMBER   38
   KANSAS POWER * LIGHT
0-)  JEFKERY NO. 2
     700  MW - NEW
   COAL
   COMMUST10N ENGINEERING
   LIMESTONE SCRUBBING
   STARTUP   6/79
                                      WORK ON THE FGD SYSTEM IS  IN THE EAKLY PLANNING STAGE.  COMBUSTION
                                      ENGINEERING IS LOOKING INTO THE S02 EMISSION CONTROL NEEDS OF THIS PLANT
                                      AND IS PREPARING A PROPOSAL TO KPL,  CONTROL OF PARTICULATES WILL BE
                                      ACCOMPLISHED  IN TWO ESP UNITS TO BE INSTALLED UPSTREAM OF EIGHT FliQ
                                      MODULES.
 I.D.  NUMBER    39
 KANSAS  POWER  *  LIGHT
 LAWRENCE  NO t
   125  MW • RETROFIT
 COAL    3.5 PERCENT  SULFUR
 COMBUSTION ENGINEERING
 LIMESTONE INJECTION *WET SCRUB
 STARTUP  12/68
                                         REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
                                         FGD SYSTEM HAS BEEN OPERATIONAL SINCE DECEMBER,1966*
 1.0.  NUMBER    40
 KANSAS  POWER  « LIGHT
 LAWRLNCL  NO 5
   400  MW - NEW
 COAL    0.5 PERCENT  SULFUR
 COMBUSTION ENGINEERING
 LIMESTONE INJECTION 8WET SCRUB
 STARTUP  11/71
                                         REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT. THIS
                                         FGD SYSTEM HAS BEEN OPERATIONAL SINCE NOVEMBER.1971.

-------
    COMPANY
                                             TABLE  2
                                  STATUS OF  F6D SYSTEMS  DURING   12/75
    POWER STATION
                                                         CUKKENT  MONTH
   ItO. NUMBER   m
   KENTUCKY UTILITIES
   GREEN RIVER   UNITS 1 AND
      614  MW - RETROFIT
   COAL  3.8 PERCENT SULFUR
   AMERICAN AIR FILTER
   LIME SCRUBBING
   STARTUP   9/75
                                      THE FGO SYSTEM AT  THIS FACILITY  IS  DESIGNED  TO  TREAT  THE  FLUE  faAS
                                      GENEKATEO FROM THREE.HIGH SULFUR.COAL-FIRED  BOILERS WHICH PRODUCE  A  TOTAL
                                      FLUE GAS CAPACITY  EQUIVALENT  TO  6H  MW.  THE FLUE GAS  IS  FED TO  THE  SCRUB-
                                      BINS SYSTEM VIA A  COMMON HEADER. THL SYSTEM  BECAME OPERATIONAL SEPTEMBER
                                      13.1975. SYSTEM OPERATION IS  CURRENTLY  BEING CONDUCTED  ON A HALF-LOAD
                                      BASIS BECAUSE OF A TUHBINE OVERHAUL. ADDITIONAL INFORMATION ON THIS
                                      SYSTEM IS AVAILABLE IN THE BACKGROUND INFORMATION SECTION IN TABLE 3 OF
                                      THIS KEPOKT.
   1.0. NUMBER   H2
   KEY WEST UTILITY BOARD
   STOCK  ISLAND PLANT
      37  MW - NEW
   OIL    2."* PERCENT SULFUR
   ZURN AIR SYSTEMS
   LIMESTONE SCRUBBING
   STARTUP  10/72
                                      REFEK TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT*  THIS
                                      FGO SYSTEM HAS BEEN OPERATIONAL SINCE OCTOBER,1972. FGO SYSTEM AVAILA-
                                      BILITY HAS BEEN 0 PERCENT SINCE JANUARY.1975.
CM
00
I.D. NUMBER   43
LOUISVILLE GAS ft ELECTRIC
CANE RUN NO 1
  110  MW - RETROFIT
COAL   3.5- t.O PERCENT SULFUR
NOT SELECTED
LIME SCRUBBING
STARTUP   6/80
A COMPLIANCE SCHEDULE HAS BEEN SUBMITTED TO THE JEFFERSON COUNTY AIR
POLLUTION CONTROL DISTRICT ESTABLISHING 6/60 AS THE STARTUP DATE FOR
A LIME SCRUBBING FGO SYSTEM.
    I.D. NUMBER    <»<*
    LOUISVILLE GAS ft ELECTRIC
    CANE RUN  NO  2
      107  MW -  RETROFIT
    COAL   3.5-  4.0 PERCENT SULFUR
    NOT SELECTED
    LIME SCRUBBING
    STARTUP   6/60
                                      A COMPLIANCE SCHEDULE HAS BEEN SUBMITTED TO THE JEFFERSON COUNTY AIR
                                      POLLUTION CONTROL DISTRICT WITH 6/60 ESTABLISHED AS THE STARTUP DATE FOR
                                      AN FGD SYSTEM.
    I.D. NUMBER   i*5
    LOUISVILLE GAS & ELECTRIC
    CANE RUN  NO  3
      137   MW •  RETROFIT
    COAL    3,5-  <».0 PERCENT SULFUR
    NOT SELECTED
    LIME SCRUBBING
    STARTUP  4/00
                                      A COMPLIANCE SCHEDULE HAS BEEN SUBMITTED TO THE JEFFERSON COUNTY AIR POLL-
                                      UTION CONTROL DISTRICT WITH 6/60 ESTABLISHED AS THE STARTUP DATE FOR AN
                                      FGO SYSTEM,

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FGD SYSTEMS DURING
12/75
                       CURRENT MONTH
1.0. NUMBER   H6
LOUISVILLE GAS & ELECTRIC
CANE RUN NO t
  178  MW - RETROFIT
COAL   3.5-14.05 PtRCENT SULFUR
AMERICAN AIR FILTER
LIME SCRUBBING
STARTUP   6/76
    THE UTILITY HAS SIGNED A CONSENT AGREEMENT WITH THE EPA FOR THE INSTAL-
    LATION AND OPERATION OF FGD EOUIPMENT ON THIS UNIT BY JUNE lf!976. THE
    CONTRACT HAS BEEN AWARDED TO AMERICAN AIR FILTER. THE SYSTEM WILL OPERATE
    ON A CLOSED WATER LOOP AND THE SLUDGE WILL BE STABILIZED. THIS SYSTEM IS
    PRESENTLY UNDEK CONSTRUCTION.
I.D. NUMBER   i*7
LOUISVILLE GAS & ELECTRIC
CANE KUN NO 5
  183  MW - RETROFIT
COAL   3,5-4.05 PERCENT SULFUR
COMBUSTION ENGINEERING
LIME SCRUBBING
STARTUP  12/77
    THE UTILITY HAS SIGNED A CONSENT AGREEMENT WITH THE EPA FOR THE INSTAL-
    LATION AND OPERATION OF FGD EQUIPMENT ON THIS UNIT BY DECEMBER I»l977.
    THE CONTRACT FOR THIS LIME SCRUBBING SYSTEM HAS BEEN AWARDED TO
    COMBUSTION ENGINEERING. THE UNIT IS PRESENTLY UNDER CONSTRUCTION,
I.D. NUMBER   H8
LOUISVILLE GAS X ELECTRIC
CANE RUN NO 6
  277  Mh - RETROFIT
COAL   3.5- 
-------
    COMPANY

    POWER  STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING
                          12/75
                                                         CURRENT MONTH
   1,0. NUMBER    51
   LOUISVILLE GAS * ELECTRIC
   HILL CREEK NO  3
     <*25   MM « NEW
   COAL    3,5- 
-------
   COMPANY

   POWER STATION
           TABLE 2
STATUS OF FGO SYSTEMS OUH1NG
       12/73

CURRENT MONTH
  1.0. NUMBER   56
  MONTANA POWER CO.
  COLSTRIP NO 2
    360  MW • NEW
  COAL   0.8 PERCENT SULFUR
  COMBUSTION EQUIP. ASSOCIATES
  LIME SCRUBBING
  STARTUP   7/76
    FGU SYSTEM UNDER CONSTRUCTION WITH NO MAJOR DELAYS.  ANTICIPATE COMPLET-
    ING CONSTRUCTION ON SCHEDULE. TURBINE ROLL AND SYSTEM CHECKOUT SCHEDULED
    TO COMMENCE IN MAY 1976.  NO DELAYS OF SCHEDULED STARTUP DATE FORESEEN AT
    THIS TIME.
  1.0. NUMBER   57
  MONTANA POWER CO.
  COLSTRIP NO,5
    700  MM  • NEW
  COAL   0,7  PERCENT  SULFUR
  NOT SELECTED
  NOT SELECTED
  STARTUP    7/79
    UTILITY IS WAITING FOR APPROVAL OF A CERTIFICATE TO CONSTRUCT BOIL-
    ERS NO,3 AND NO.H BEFORE PROCEEDING WITH PLANS TO  INSTALL FGD
    SYSTEMS.
   I.D,  NUMBER    58
   MONTANA  POWER  CO.
*>  COLSTRIP N0.t»
^    700 MW -  NEW
   COAL   0.7 PERCENT  SULFUR
   NOT SELECTED
   NOT SE'-ECTEO
   STARTUP    7/80
    UTILITY IS WAITING FOR APPROVAL OF A CERTIFICATE TO CONSTRUCT BOIL-
    ERS NO.3 AND NO.H BERFORE PROCEEDING WITH PLANS TO INSTAL FGO
    SYSTEMS.
   1.0,  NUMBER   59
   NEVADA POWER
   HARRY ALLEN STATION NO,  1
     500  MW - NEW
   COAL
   NOT SELECTED
   NOT SELECTED
   STARTUP   6/62
    CONSIDERING HOT  SIDE ESP  IN CONJUNCTION WITH AN FGO SYSTEM.
    TIONS HAVE NOT YET BEEN PREPARED.
                                          SPECIFICA-
   I.D. NUMBER   60
   NEVADA POWER
   HARRY ALLEN STATION NO.
     500  MW - NEW
   COAL
   NOT SELECTED
   NOT SELECTED
   STARTUP   6/63
     CONSIDERING  HOT  SIDE  ESP  IN CONJUNCTION WITH AN FGD SYSTEM.
     TIONS HAVE NOT YET  BEEN PREPARED.
                                          SPECIFICA-

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FGO SYSTEMS DURING  12/75

                       CURRENT MONTH
ItO. NUMBER   61
NEVADA POWER
HARRY ALLEN STATION NO. 3
  500  nw » NEW
COAL
NOT SELECTED
NOT SELECTED
STARTUP   6/6H
    CONSIDERING HOT SIDE ESP IN CONJUNCTION WITH AN FGD SYSTEM.
    TIONS HAVE NOT YET BEEN PREPARED.
SPECIFICA-
1.0. NUMBER   62
NEVADA POWER
HAKRY ALLEN STATION NO.
  500  MM - NEW
COAL
NOT SELECTED
NOT SELECTED
STARTUP   6/85
    CONSIDERING HOT SIDE ESP IN CONJUNCTION WITH AN F00 SYSTEM.
    TIONS HAVE NOT YET BEEN PREPARED.
SPECIFICA-
1.0. NUMBER   63
NEVADA POWER
REID GAHDNER NO 1
  12S  MW • RETROFIT
COAL   0.5- 1.0 PERCENT SULFUR
COMBUSTION EQUIP. ASSOCIATES
SODIUM CARBONATE SCRUBBING
STARTUP
    REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT.   THIS
    FGO SYSTEM HAS BEEN OPERATIONAL SINCE APRIL 1974.
1*0. NUMBER   64
NEVADA POWER
REXO GAKONER NO 2
  123  M* - RETROFIT
COAL   0.5- 1.0 PERCENT SULFUR
COMBUSTION EQUIP. ASSOCIATES
SODIUM CARBONATE SCRUBBING
STARTUP  12/75
    REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT.  THIS
    FGO SYSTEM HAS BEEN OPERATIONAL SINCE DECEMBER 1973.
1.0. NUMBER   65
NEVADA POWER
REIO GAKONER NO 3
  123  MW - NE.W
COAL   0,5- 1.0 PtRCiNT SULFUR
COMBUSTION EQUIP. ASSOCIATES
SODIUM CARBONATE SCRUBBING
STARTUP   6/76
    REFER TO MONTHLY COMMENTS IN BACKGROUND INFORMATION
    SECTION INCLUDED WITH TABLE 3 FOR THIS UNIT.

-------
 COMPANY
           TABLE 2
STATUS OF FGD SYSTEMS DURING
12/75
 POWER STATION
                       CURRENT MONTH
I.D. NUMBER   66
NEVADA POWER
REID GAHUNER NO H
  125  MW - NEW
COAL   0-5- 1.0 PERCLNT SULFUR
COMBUSTION EQUIP. ASSOCIATES
SODIUM CARBONATE SCRUBBING
STARTUP   O/ 0
    NEVADA POWEK COMPANY MAS SIGNED A LETTER OF INTENT WITH COMBUSTION EQUIP-
    MENT ASSOCIATES FOR THE CONSTRUCTION OF AN FGO SYSTEM ON REID GARDNER NO,
    >»,  HOWEVER. CONSTRUCTION OF THE BOILER HAS BEEN INDEFINITELY POSTPONEDt
I.D, NUMBER   67
NEVADA POWER
WARNER VALLEY STATION NO. X
  250  MW - NEW
COAL
NOT SELECTED
NOT SELLCTED
STARTUP   6/81
    SPECIFICATIONS ARE BEING PREPARED FOR A SCRUBBING SYSTEM,
1.0. NUMBER   60
NEVADA POWER
WARNLR VALLEY STATION NO.
  250  MW - NEW
COAL
NOT SELECTED
NOT SELECTED
STARTUP   6/B2
    SPECIFICATIONS ARE BEING PREPARED FOR A SCRUBBING SYSTEM.
I.D. NUMBER   69
NEW ENGLAND ELEC SYSTEM
BRAYTON POINT NO,3
  650  Mw - RETROFIT
COAL 3.0 PERCENT SULFUR
NOT SELECTED
RtGENERABLE NOT SELECTED
STARTUP   O/ 0
     INVESTIGATING THE POSSIBILITY OF AN ADVANCED REGENERABLE FGO SYSTEM
     HAVING ELEMENTAL SULFUR AS AN END PRODUCT WHICH HAS PROMISE OF A
     BREAKTHROUGH IN CAPITAL AND OPERATING COST.
I.D. NUMBER   70
NOKTHEHN INDIANA PUB SERVICE
BAILLY NO. 7
  190  *W - RETROFIT
COAL 3 PLRCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   O/ 0
    CONSIDERING A LIKE OR LIMESTONE SCRUBBING UNIT.  ALSO WAITING FOR PER-
    FORMANCE  OF WELLMAN LORD/ALLIED CHEMICAL UNIT UNDER CONSTRUCTION AT THEIR
    0. H. MITCHEEL NO, 11 UNIT.  LOW SULFUR COAL MAY BE BURNED TO COMPLY
    WITH S02  EMISSION REGULATIONS.

-------
 COMPANY

 POWIR STATION
           TABU 2
STATUS OF FGO SYSTEMS DURING  12/75

                       CURRENT MONTH
i.o. NUMBER   71
NORTHERN INDIANA PUB SERVICE
BAILLT NO. 6
  <*00  MW - RETROFIT
COAL 5 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   O/ 0
    CONSIDERING A LIME OR LIMESTONE SCRUBBING UNIT.  ALSO WAITING FOR PERFOR-
    MANCE OF WELLMAN LORD/ALLIED CHEMICAL UNIT UNDER CONSTRUCTION AT THEIR
    0. H, MITCHELL NO, 11 UNIT,  LOU SULFUR COAL MAY BE BURNED TO COMPLY WITH
    sos EMISSION REGULATIONS.
I.D. NUMBER   72
NORTHERN INDIANA PUB SERVICE
MITCHELL NO 11
  115  MW - RETROFIT
COAL   3.2* 3.5 PERCENT SULFuR
DAVY POWERGAS/ALLIED CHEMICAL
WELLMAN LORD/ALLIED CHEMICAL
STARTUP   3/76
    SULFUR DIOXIDE WILL BE CONVERTED TO ELEMENTAL SULFUR IN AN ADJACENT PLANT
    USING THE ALLIED CHEMICAL S02 REDUCTION PROCESS.  THIS INSTALLATION IS
    PAHTIALLY FUNDED BY AN EPA DEMONSTRATION GHANT, THIS UNIT IS STILL UNDER
    CONSTRUCTION. TH£ STARTUP DATE HAS BEEN MOVED BACK TO MARCH 1976,
I.D. NUMBER   73
NORTHERN STATES POWER CO,
SHERBURNE NO 1
  660  MM • NEW
COAL 0,« PERCENT SyLFUR
COMBUSTION ENGINEERING
LIMESTONE SCRUBBING
STARTUP   5/76
    FGO SYSTEM IS STILL UNoER CONSTRUCTION, PURCHASE ORDERS FOR SCKUBBERS
    WERE PLACED IN EARLY 1971. DESIGN MODIFICATIONS CAUSED EQUIPMENT DELIVERY
    DELAYS,HOWEVER THE OVtKALL SCHEDULE HAS NOT BEEN UPSET. PROBLEMS ENCOUN-
    TERED BY THE UTILITY THUS FAR HAyE BEEN DUE TO OBTAINING SOME ELECTRICAL
    CONTROLS, A KECCNT CONSTRUCTION STKIKE, AND A BOILEK MODIFICATION INVOL-
    VING GREATtK NIGATIVL PRESSURE TOLEKANCE, THE UTILITY IS NOW BEGINNING A
    PRELIMINARY SYSTEM CHECKOUT BY PASSING AIR AND CLEAK WATER THKOUGH THE
    FGO UNIT, A MAY STAHTUP OF THIS SYTtM IS STILL ANTICIPATED.
1.0. NUMBER   7<4
NORTHERN STATES POWER CO.
SHE.R3URNE NO 2
  600  MW - NEW
COAL 0.8 PERCENT SULFUR
COMBUSTION ENGINEERING
LIMESTONE SCRUBBINfi
STARTUP   5/77
    FGU SYSTEM UNDEK CONSTRUCTION WITH NO MAJOR DELAYS.  PURCHASE OKOtRS FOR
    SCRUBBERS WtKE PLACED  IN EARLY 1971.  RECENT SCRUBBER DtSIbN CHANGES HAVE
    CAUStO EOUIPMENT OtLIVLRY OELAYS BUT THE OVt-RALL SCHtDULE HAS NOT BEEN
    UPSET,  PROBLEMS HAVE  ARISEN IN OBTAINING SOME ELECTRICAL CONTROLS.
    CONSTRUCTION IS 8EHINO SCHLOULt AND SOME DELAY OF  STARTUP DATE MAY BE EX-
    PECTED DUE TO RECENT STRIKE. THE EXTENT OF THE DELAY CANNOT BE DETERMINED
    AT THE PRESENT TIME.
I.D. NUMBER   75
PENNSYLVANIA POWER CO.
BRUCE MANSFIELD NO. 1
  635  MW • NEW
COAL   <».3 PERCENT SULFUR
CHEMICO
LIME SCRUBBING
STARTUP   H/76
    THE UTILITY BEGAN A SHAKEDOWN AND DEBUGGING PHASE  OF  OPERATION  FOK  PART
    OF THE SYSTEM  IN DECEMBER  1975. CONSTRUCTION  IS  STILL IN  PROGRESS ON  THE
    REMAINING MOOULtS OF  THIS  SYSTEM, FULL COMMERCIAL  STARTUP IS  PRESENTLY
    PROJECTED FOK  APRIL 1976.

-------
 COMPANY
 POWER STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING
                                                                   12/75

                                                            CURRENT MONTH
I.D. NUMBER   76
PENNSYLVANIA POWER CO.
BRUCE MANSFIELD NO. 2
  635  MW - NEW
COAL   "»,3 PERCENT SULFUR
CHEMICO
LIME SCRUBBING
STARTUP   14/77
                                         FGO SYSTEM UNDER CONSTRUCTION, BUT ENCOUNTERING LONG DELAYS IN MATERIAL
                                         DELIVERIES,  STARTUP DATE HAS BEEN DELAYED 1 YEAR UNTIL f/77.
I.D. NUMBER   77
PENNSYLVANIA POWEK CO.
BRUCE (MANSFIELD NO. '3
  835  MW - NEW
COAL   H.3 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   H/79
                                         FGO UNIT IS PLANNED. PRESENTLY REQUESTING AND EVALUATING BIDS.
   I.D, NUMBER   76
   PHILADELPHIA ELECTRIC CO,
.fc,  CROMBY
^    150  MW - RETROFIT
   COAL   2-1 PERCENT SULFUR
   UNITED ENGINEERS
   MAGNESIUM OXIDE SCRUBBING
   STARTUP  10/7ti
                                      COMPANY PLAN  TO  RETROFIT  ONE  OF  THE  TWO  BOILERS  AT  CKOMBY  BUT  HAVE  NOT
                                      MADE  A FINAL  DECISION  ON  THIS POINT,   THE  FGO  PROCESS  SELECTED WILL VERY
                                      LIKELY BE AN  MGU PROCESS  OESIGNLO  JOINTLY  BY UNITED ENGINEERS  AND PHILA-
                                      DELPHIA ELECTRIC.   ENGINEERING DESIGN  WORK IS  SCHEDULED  FOR  JAN 76.
                                      STARTUP NOW SCHEDULED  FOR OCT 76.
 I.D, NUMBER    79
 PHILADELPHIA ELECTRIC CO,
 EOOYSTONE  NO lA
   120  MW  - RETROFIT
 COAL   2.5 PERCENT  SULFUR
 UNITED ENGINEERS
 MAGNESIUM  OXIDE SCRUBBING
 STARTUP    9/75
                                         THE S02 SCRUBBER HAS BEEN TEMPORARILY SHUTDOWN BECAUSE THE ACIO PLANT RE-
                                         GENERATION FACILITY AT  THE OLIN CHLMICAL SULFURIC ACID PLANT IN PAULSBORO
                                         NEW JERSEY HASPLRMflNLNI LKEASEU OPERATIONS. CONSIDERATIONS ARE NOW
                                         3EING GIVEN TO THE RELOCATION OF THt. RLGENERATION FACILITY. ONCE THIS DE-
                                         CISION is MADE A MINIMUM PERIOD OF six MONTHS WILL BE REQUIRED TO RELO-
                                         CATE. THE REGENERATION FACILITY. THE PAKTICULATE SCRUBBERS ARE CONTINUING
                                         TO OHL'RATt. THE UTILITY IS STILL EXPERIENCING PROBLEMS WITH THE FANS, RE-
                                         HEAT BURNERSi DAMPERSt  AND EXPANSION JOINTS.
 1.0.  NUMBER    80
 PHILADELPHIA ELECTRIC CO.
 tOOVSTONE  NO IB
   240   MW  - REIROFIT
 COAL  2.5 PERCENT SULFUR
 JNITEO  ENGINEERS
 MAGNESIUM  OXIDE SCRUBBING
 STARTUP 10/78
                                         THt. RETROFITTING OF SECOND STAGE SCRUBBERS ON THE BALANCE OF THE FLUE GAS
                                         FROM E.DOYSTONE: i is AWAITING PERFORMANCE OF THL EXISTING Z-STAGE UNIT
                                         WHICH IS TREATING ONE. THIRD OF THE BOILER FLUE GAS.  THIS UNIT CONSISTS
                                         OF A AND B PARTICULATE SCRUBBING TRAINS, EACH ARE 120 MW.  CHECK VALVE
                                         FAILURE ON A TRAIN HAS BEEN CORRECTED.  FAN ON B TRAIN IS STILL AT THE
                                         FACTORY FOR REPAIR.

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FSO SYSTEMS DURING  12/75

                       CURKENT MONTH
I.D, NUMBER   01
PHILADELPHIA ELECTRIC CO,
EOomoNE NO 2
  936  MW . RETROFIT
  COAL   2.H PERCENT SULFUR
UNITED ENGINEERS
MAGNESIUM OXIDE SCRUBBING
STARTUP  10/78
    COMPANY IS AWAITING PERFORMANCE OF FGO SYSTEM ON UNIT 1 BEFORt PKOCEEDING
    WITH DESIGN OF FGO SYSTEM ON THIS BOILER*  HOWEVER ENGINEERING PHASE IS
     TENTATIVELY PLANNED TO START IN JAN 76, AT WHICH TIME EVALUATION OF
    DATA FROM UNIT 1 SHOULD BE COMPLETE.
1.0. NUMBER   62
PUBLIC SERVICE CO OF NEW MEX.
SAN JUAN NO. 1
  379  MW - NEW
COAL   0,6 PERCENT SULFUR
DAVY POWERGAS/ALLIED CHEMICAL
WELLMAN LORD/ALLIED CHEMICAL
STARTUP   7/77
    FGO PROJECT IS PROCEEDING ACCORDING TO PLAN WITH NO MAJOR PROBLEMS,
    EQUIPMENT IS PRESENTLY BEING PROCURED. CONSTRUCTION TO BEGIN IN MARCH
    OR APKIL 1976.
I.D. NUMBER   63
PUBLIC SERVICE CO OF NEW MEX.
SAN JUAN NO. 2
  3<»0  MW - RETROFIT
COAL   0.6 PERCENT SULFUR
DAVY POWERGAS/ALLICD CHEMICAL
WELLMAN LORD/ALLIED CHEMICAL
STARTUP   7/77
    FGO PROJECT IS PROCEEDING ACCORDING TO PLAN WITH NO MAJOR PROBLEMS,
    EQUIPMENT IS PRESENTLY BEING PROCURED. CONSTRUCTION TO BEGIN IN MARCH
    OR APKIL 1976.
I.D. NUMBER   6<+
PUBLIC SERVICE CO OF NEW MEX.
SAN JUAN NO. 5
  500  MM . NEW
COAL 0.6 PERCENT SULFUR
DAVY POWERGAS
WELLMAN LORD
STARTUP   5/78
    A LETTER OF INTENT HAS BEEN SIGNED WITH DAVY POWERGAS FOR A WELLMAN
    LORD UNIT.
1.0. NUMBER   65
PUBLIC SERVICE CO OF NEW MEX.
SAN JUAN NO. •*
  500  MW - NEW
COAL 0.6 PERCENT SULFUR
DAVY POWERGAS
WELLMAN LORD
STARTUP   5/60
    A LETTER OF INTENT HAS BEEN SIGNED WITH DAVY POWERGAS FOR A WELLMAN
    LORD UNIT.

-------
 COMPANY
 POWER STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING
                          12/75

                   CURRENT MONTH
I.D. NUMBER   86
PUBLIC SE.KVICE CO, OF COLORADO
VALMONT NO, 5
   50  Mw - RETROFIT
COAL  0.72 PERCENT SULFUR
UOP / PU9 SERVICE OF COLORADO
LIMESTONE SCRUBBING
STARTUP  10/74
REFEN TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT* THIS
SYSTEM HAS BEEN OPERATIONAL SINCE NOVEMBER 197*.
1,0. NUMBER   87
PUBLIC SCRVICE INDIANA
GIBSON NO. 3
  650  MW - NEW
COALt 3.3S SULFUR
NOT SELECTED
NOT SELECTED
STARTUP   0/78
BID EVALUATION IS UNDERWAY.
 I.D. NUMBER    68
 PUBLIC SERVICE  INDIANA
 GIBSON NU. 4
   650  MM - NEW
 COAL
 NOT SELECTED
 NOT SELECTED
 STARTUP   0/79
BID SPECIFICATIONS ARE BEING PREPARED.
 I.D.  NUMBER    89
 RICKENBACKER  AFB
 RICKENBACKER
    20  MW - RETROFIT
 COAL    5.0 PERCENT SULFUR
 RESEARCH COTTRELL
 LIME  SCRUBBING
 STARTUP   2/76
 CONSTRUCTION  OF  THIS  SYSTEM  IS PRESENTLY PROCEEDING ON SCHEDULE. WORK ON
 THE  SLUDGE POND* FOUNDATION,  AND  TOWER  IS NEARING COMPLETION, THE
 SCHEDULED FEBRUARY  STARTUP DATE IS  STILL OPERATIVE.
 I.D. NUMBER   90
 S. CAROLINA PUB SERV AUTHORITY
 WINYAH NO.  2
   1HO  MM - NEW
 COAL  1.0 PERCENT SULFUR
 BABCOCK * WILCOX
 LIMESTONE SCRUBBING
 STARTUP   5/77
 A  LETTER  OF  INTENT  HAS  BEEN  SIGNED  WITH  BtW  FOR CONSTRUCTION OF A LIME>
 STONE  SYSTEM TREATING 1HOMW  OF  THE  TOTAL 200NW.  DIFFERENT METHODS OF
 DISPOSAL  ARE BEING  EXAMINED.

-------
     COMPANY

     POWER STATION
                                             TABLE 2
                                  STATUS OF FGO SYSTEMS DURING  12/79

                                                         CURRENT MONTH
    1.0.  NUMBER   91
    S.  MISSISSIPPI ELEC PWR ASSOC
    «.D.  MORKOW NO.l
      160  MW - NEW
    COAL  1 PtKCENT SULFUR
    RILEY STOKER/ENVIRQNEERING
    LIMESTONE SCRUBBING
    STARTUP  11/77
                                      THE CONTRACT FOR THIS LIMESTONE SYSTEM HAS BEEN AWARDED TO R1LEY
                                      STOKER/ENVIRONEERING.
    I.D, NUMBER   92
    S, MISSISSIPPI ELEC PUR ASSOC
    R.O. MORKWW NO.2
      160  MW - NEW
    COAL 1 PERCENT SULFUK
    RILEY STOKER/ENVIKONEERING
    LIMESTONE SCRUBBING
    STARTUP   6/78
                                      THE CONTRACT FOR THIS LIMESTONE SYSTEM HAS BEEN AWARDED TO RILEY
                                      STOKER/ENVIRONEERING.
oo
1.0. NUMBER   99
SOUTHERN CALIFORNIA EOISON
KAIPAROWITZ NO. 1
  750  MW - NEW
COAL 10.600.BTU. 0.5)1 S
NOT SELECTED
LIME SCRUBBING
STARTUP   0/62
                                          THIS  NEW  STATION  IS  ANTICIPATED  TO  INCORPORATE FGD,  BUT  A  SPECIFIC
                                          PROCESS HAS  NOT BEEN SELECTED. THE  STARTUP  DATE FOR  THIS UNIT  HAS BEEN
                                          DELAYED ONE  YEAR  PENDING  THE  ASSEMBLY  AND EVALUATION OF  DATA FOR ENVIRO-
                                          MENTAL CONSIDERATIONS AND LENGTHY APPROVAL  PROCESSES.
    1.0,  NUMBER    9*»
    SOUTHERN  CALIFORNIA  EDISON
    KAIPAROWITZ  NO, 2
      750  MW  -  NEW
    COAL  lOtSOOiBTUt 0.5K  S
    NOT SELECTED
    LIME  SCRUBBING
    STARTUP    0/63
                                      THIS NEW STATION IS ANTICIPATED TO INCORPORATE FGD, BUT A SPECIFIC
                                      PROCESS HAS NOT BEEN SELECTED. THE STARTUP DATE FOR THIS UNIT HAS BEEN
                                      DELAYED ONE YEAR PENDING THE ASSEMBLY AND EVALUATION OF DATA FOR ENVIRO-
                                      MENTAL CONSIDERATIONS AND LENGTHY APPROVAL PROCESSES.
    1.0.  NUMBER    95
    SOUTHERN  CALIFORNIA EDISON
    KAIPAHOWITZ NO. 3
      750   MW  - NEW
    COAL  10»600»BTUi 0.5* S
    NOT SELECTED
    LIME  SCRUBBING
    STARTUP   0/84
                                      THIS NEW STATION IS ANTICIPATED TO INCORPORATE FGD, BUT A SPECIFIC
                                      PROCESS HAS NOT BEEN SELECTED. THE STARTUP DATE FOR THIS UNIT HAS BEEN
                                      DELAYED ONE YEAR PENDING THE ASSEMBLY AND EVALUATION OF DATA FOR ENVIRO-
                                      MENTAL CONSIDERATIONS AND LENGTHY APPROVAL PROCESSES.

-------
 COMPANY

 POWER STATION
                                             TABLL 2
                                  STATUS OF FGO SYSTEMS DURING
                                                                    12/75

                                                             CURRENT  MONTH
I.D. NUMBER   96
SOUTHERN CALIFORNIA EOXSON
KAIPAROWITZ NO, *
  750  MW - NEW
COAL 10,800»BTU« 0,5X S
NOT SELECTED
LIME SCRUBBING
STARTUP   0/85
                                         THIS NEW  STATION  Is  ANTICIPATED  TO  INCORPORATE FGD,  BUT A SPECIFIC
                                         PROCESS HAS  NOT BEEN SELECTED.  THE  STARTUP DATE FOR  THIS UNIT HAS BEEN
                                         OCLAYEQ ONE  YEAR  PENDING THE  ASSEMBLY AND EVALUATION OF DATA FOR ENVIRO-
                                         MENTAL CONSIDERATIONS AND LENGTHY APPROVAL PROCESSES,
1.0. NUMBER   97
SOUTHERN CALIFORNIA EDISON
MOHAVt NO IB
  620  MW - RETROFIT
COAL   0.5- 0.8 PERCENT SUL.FUR
NOT SELECTED
LIME/LIMESTONE: SCRUBBING
STARTUP   6/77
                                          THE  TOTAL CAPACITY OF BOILER IS 790 MW. EXISTING EXPERIMENTAL FGD MODULE
                                          WILL TKEAT ONLY 170 MW,   FINAL SELECTION OF THE PROCESS HAS BEEN POST-
                                          PONEO.  SEE MOHAVE 1A.
    I.D.  NUMBER    98
    SOUTHERN CALIFORNIA EDISON
    MOHAVE NO,  2
^     790  Mta -  RETROFIT
<°   COM.    0.5  TO 0.8 % SULFUR
    NOT SELECTED
    LIME/LIMtSTONE SCRUBBING
    STARTUP   6/77
                                       A  PROTOTYPE  170  MW  LIME  SCRUBBING SYSTEM (HORIZONTAL MODULE)  ON THIS UNIT
                                       WAS  OPERATED FROM NOVEMBER  1973  UNTIL FEBRUARY 1975. AVAILABILITY FOR THE
                                       PERIOD  1/16/7H  TO 2/9/75 AVERAGED 75.558,  THIS SYSTEM IS TO BE REINSTALL-
                                       ED AT THE  FOUR  CORNERS PLANT*  OPERATED BY  ARIZONA PUBLIC SERVICE,
 I.D.  NUMBER    99
 SPRINGFIELD  CITY UTILITIES
 SOUTHWEST  NO.  1
   200  f.rf  -  NEW
 COAL  3.5 PERCENT SULFUR
 UNIVERSAL  OIL  PRODUCTS
 LIMESTONE  SCRUBBING
 STARTUP   6/76
                                          BOTH BOILER AND FGO SYSTEM ARE PRESENTLY UNDER CONSTRUCTION.  EXPERIENC-
                                          ING SOME DELAYS BUT COMPLETION Is STILL EXPECTED BY JUNE,  SCRUBBER AND
                                          CSP A.RE HOW BEING ERECTED.  10 FAN IS ALSO BEING INSTALLED
 I.D.  NUMBER  100
 TENNESSEE VALLEY AUTHORITY
 SHAWNEE NO.IDA
   10    MW - RETROFIT
 COAL  2.9 PERCENT SULFUR
 UNIVERSAL OIL PRODUCTS
 LIME/LIMESTONE SCRUBBING
 STARTUP   «*/72
                                          REFER TO BACKGROUND INFORMATION SECTION IN TABLE 3 OF THIS REPORT, THIS
                                          TURBULENT CONTACT ABSOKB£R(TCA) LIME/LIMESTONE SCRUBBING SYSTEM HAS BEEN
                                          OPERATIONAL SINCE APRIL 1972. THIS TEST PROGRAM IS FUNDED BY THE EPA,
                                          TVA IS THE CONSTRUCTOR AND FACILITY OPERATOR. THE Bt-CHTLL CORP, OF SAN
                                          FRANCISCO IS THE MAJOR CONTRACTOR, TEST DIRECTOR. AND REPORT WHITER.

-------
     COMPANY

     POWER  STATION
                                             TABLE 2
                                  STATUS OF FGD SYSTEMS DURING  12/79

                                                         CURRENT MONTH
    I.D.  NUMBER   101
    TENNESSEE  VALLEY  AUTHORITY
    SHAWNEt  NO,108
      10   MW  -  RETROFIT
    COAL  2.9 PERCENT  SULFUR
    CHCMICO
    LIME/LIMESTONE  SCRUBBING
    STARTUP    4/72
                                      REFER TO BACKGROUND INFOHMATIION SECTION IN TABLE 3 OF THIS REHOKT.  THIS
                                      VENTURI/SPRAY TOWER LIME/LIMESTONE SCRUBBING SYSTEM HAS BEEN OPERATIONAL
                                      SINCE APRIL 1972.  THIS TEST PROGRAM IS FUNDED BY THE EPA,  TVA IS THE
                                      CONSTRUCTOR AND FACILITY OPERATOR. THE BiCHTEL CORP. OF SAN FRANCISCO IS
                                      THE MAJOR CONTRACTOR*TEST DIRECTOR* AND KEPOKT WHITER.
    I.D.  NUMBER   102
    TENNESSEE  VALLEY  AUTHORITY
    WIDOWS  CREEK  NO 8
      550  MW  - RETROFIT
    COAL    3.7 PERCENT  SULFUR
    TENNESSEE  VALLEY  AUTHORITY
    LIMESTONE  SCRUBBING
    STARTUP   2/77
                                      CONTRACTS HAVE BEEN LET ON ALL MAJOR PROCESS EQUIPMENT INCLUDING THE
                                      SCRUBBER* MIST ELIMINATOR* PIPING* INSTRUMENTATION* HOT AIR INJECTION
                                      SYSTEM AND AIR HEATER FANS, THE ID FANS ARE NOW BEING INSTALLED AS is THE
                                      LIMESTONE CRUSHER. THE LAST EQUIPMENT REQUISITION WILL BE COMPLETED BY
                                      JUNE 1976. THE SCHEDULED STARTUP DATE OF FEBRUARY 1977 IS STILL OPERA-
                                      TIVE.
en
o
I.D, NUMBER  103
TEXAS UTILITIES CO,
MARTIN LAKE NO, 1
  793  MW • NEW
COAL   1.0 PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP   9/76
                                          NEW  SYSTEM*  ERECTING  BOILER.   CONTRACT FOR  FGO  SYSTEM  AWARDED.
                                          ENGINEERING  OESI8N  k-ORK  IS PROCEEDING.   FABRICATION  OF THE  TOWER HAS
                                          BEEN COMPLETED.
    1.0,  NUMBER   1Q4
    TEXAS UTILITIES CO.
    MARTIN LAKE  NO. 2
      793  MW  -  NEW
    COAL    1.0 PERCENT SULFUR
    RESEARCH COTTKELL
    LIMESTONE  SCRUBBING
    STARTUP   6/77
                                      NEW SYSTEM, ERECTING BOILER.  CONTRACT FOR FGO SYSTEM AWARDED.
                                      ENGINEERING DESIGN WORK IS PROCEEDING.  FABRICATION OF THE TOWER HAS
                                      BEEN COMPLETED.
    1.0.  NUMBER   109
    TEXAS UTILITIES CO,
    MARTIN LAKE NO. 3
      793  MW  - NEW
    COAL  1,0 PERCENT SULFUR
    RESEARCH COTTRELL
    LIMESTONE  SCRUBBING
    STARTUP 12/76
                                      UTILITY HAS SIGNED A LETTER OF INTENT WITH RESEARCH COTTRELL. THE BOILER
                                      IS CURRENTLY BEING ERECTED AT THIS NEW FACILITY.

-------
 COMPANY

 POWER STATION
           TABLE 2
STATUS OF FGO SYSTEMS DURING
       12/75

CURRENT MONTH
I.D. NUMBER  106
TEXAS UTILITIES CO,
MARTIN LAKE NO. t
  T46  MU - NEW
COAL 1.0 PERCENT SULFUR
RESEARCH COTTRELL
LIMESTONE SCRUBBING
STARTUP  12/79
    UTILITY HAS SIGNED A LETTER OF INTENT WITH RESEARCH COTTRELL, THE BOILER
    IS CURRENTLY BEIN6 ERECTED AT THIS NEW FACILITY.
I.D. NUMBER  107
UNITED POWER ASSOCIATION
COAL CREEK NO. 1
  5<*5  MM - NEW
LIGNITE • 0.63 PERCENT SULFUR
NOT SLLECTED
NOT SLLECTEO
STARTUP  11/76
    UTILITY IS INVESTIGATING BOTH REGENERABLE AND NON-REGENERABLE SYSTEMS.
1.0. NUMBER  108
UNITED POWER ASSOCIATION
COAL CREtK NO. 2
  5H5  MW - NEW
LIL.N1TE • 0.63 PERCENT SULFUR
NOT SELECTED
NOT SELECTED
STARTUP  11/79
    UTILITY IS INVESTIGATING BOTH REGENtRABLE AND NON-REGENERABLE SYSTEMS.
 1.0. NUMBER  109
 UTAH POWER ft LIGHT CO*
 HUNTINGTUN NO.l
  <*15  MW - NEW
 COAL  0.5 PERCENT SULFUR
 CHEMICO
 LIME SCRUBBING
 STARTUP   6/77
    A LETTER OF  INTENT HAS BEEN SIGNED WITH CHEMICO.  UNIT NO. 1  IS THE
    SECOND UNIT  AT HUNTINGTON.  UTAH REGULATIONS REQUIRE BOX S02  REMOVAL.

-------
    STATUS OF FLUE GAS DESULFURIZATION AND SIMULTANEOUS REMOVAL
                      OF SO- AND NO  IN JAPAN
                           L+       A.
                            Jumpei Ando

                Faculty o£ Science and Engineering
                          Chuo University
                     Kasuga, Bunkyo-ku, Tokyo
ABSTRACT

     About 100 FGD plants including several large ones with a capacity
of 250-500 MW went into operation in 1975.  Operation of most plants
has been trouble-free.  The large plants have attained more than 97|
availability.  The total FGD capacity which was about 30 million Nm /hr
(1Q,OOOMW equivalent) at the end of 1974 will exceed 70 million
Nm /hr by the end of 1976.  The rapid growth is due to the economical
advantage of FGD over the use of low-sulfur fuels and to the reliability
of plant operation.  The growth, however, may slow down after 1977
for the following reasons:

(1)  The ambient SO  concentration in big cities and industrial
     districts which was 0.05-0.09ppm in annual average several years
     ago has decreased to 0.02-0.03ppm, almost meeting the ambient
     standard.

(2)  The recent economic depression has prevented industry from
     building new plants.

(3)  A tendency of overproduction of FGD by-products has occurred
     alluring industry toward the use of  low-sulfur fuels, because
     Japan has limited landspace available for discarding useless
     by-products.

(4)  The stringent NO  regulation has started to force industry to
     carry out flue  gas denitrification.  Several processes  for simul-
     taneous removal  of NO  and SO  have been developed and  industry
     is waiting  for  the completion of new technology.

     This paper  will  describe the recent  status of FGD and simultaneous
removal of S0  and NO .
                                  53

-------
   STATUS OF FLUE GAS DESULFOBIZATION AND SIMULTANEOUS REMOVAL OF

               S02 AND HO, H JAPAN
                                              Jtnape.


        1     Major waste gas desulfurization installations


Table 1 ahovs the major wet-liae/Limestone process installations
completed or to be completed between 1974 *ad 76.  Most plants by-
produce salable gypsum except Omuta plant, Mitsui Aluminum, using the
Cheai co-Hit BUI process by-producing throw-away calcium sulfite
sludge and two plants  of Nippon Steel which use  a converter slag
as the absorbent to by-produce low-grade gypsum for discarding.

Most of the nev plants use limestone as the absorbent.  Limestone
is generally ground by wet mill to pass 325 mesh about 95$ to recover
more than yjfa of 302 .  Operation parameters of major plants are
listed in Table 2.  The total capacity of wet-lime/limeatone process
plants will reach about 36 million Nm3/br (12,OOOMV equivalent) by
the end of 1976 (Table 6).

Table 3 shows major plants with indirect lime/limestone processes
including doable alkali processes and similar processes using acidic
absorbents such as dilute sulfuric acid and aluminum sulfate.  Opera-
tion parameters of major plants are listed in Table 2.  The more
acidic the solution is, the less the S02 absorption capacity requiring'
the larger L/G ratio,  the lesser the problem of scaling, and the
easier the reaction with limestone.  The total capacity of the indi-
rect process plants will reach about 13 million Nm3/hr (4,500MW) by
the end of 1976.

Table 4 shows installations using other processes by-producing sulfuric
acid, elemental sulfur, and ammonium sulfate.  The total capacity is
about 8 million Vm3/hr.  In addition, there are nearly 200 relatively
small plants with sodium scrubbing, by-producing sodium sulfite for
paper mills and sodium sulfate for waste or industrial use.  The
total capacity of these sodium scrubbing plants will reach about 14
million Nm3/hr by the  end of 1976 (Table 6).

Table 5 shows major processes classified by developers, the number of
plants, and the total  capacity.  The combined capacity will reach 70
million Hm3/hr (24,OOOMW) by the end of 1976.  About one-half of the
gas is from utility boilers and the rest from industrial boilers and
other sources such as  sulfuric add and iron ore sintering plants.

Since most of the processes have been reported previously,1*2' the per-
formance of major new plants for utility boilers and new developments
in the steel industry will be given below.

Most plants purge wastewater as shown in Table 7.  The amount of
the water discharged from many of the plants is about the same as
that from a "closed loop" in the U.S.A. which discards a calcium
sulfite sludge containing about 5036 water (or 0.5 water ratio).
The wastewater is treated to reduce COD (chemical oxygen demand)
usually to below lOmgAiter and suspended solids to about 30mg/liter.

                                54

-------
Table I   Major wet-lime/limestone process installations (1974-1976)

Process developer
Mitsubishi (MHI)
i)
ii
ti
it
ii
ii
ii
it
ii
ti
Mi t sui- Chemi co
n
it
Babcock-Hitachi
n
n
it
Fuji Kasui-
Sumitomo
n
n
n
Chubu-MKK
Kobe Steel
n
it
Nippon Steel
Chemi co- IHI
Kawasaki H.I.
n
a
b
c

User
Kansai Electric
Tokyo Electric
Kyushu Electric
Kawasaki Steel
n
n
Kansai Electric
Mizushima Power
Kyushu Electric
n
Chugoku Electric
Mitsui Aluminum
Elec. Pow. Dev.
n
Chugoku Electric
n
Asahi Chemical
Kansai Electric
Sumitomo Metal

n
ii
n
Ishihara Sangyo
Kobe Steel
n
Nakayama Steel
Nippon Steel
Elec. Pow. Dev.
Nippon Exlan
Unitika

Plant site
Kainan
Yokosuka
Karita
Mizushima
n
Chiba
Amagasaki
Mizushima
Karat su
Ainoura
Owase
Omuta
Takasago
n
Mizushima
Tamashima
Mizushima
Osaka
Wakayama

Kashima
it
Kokura
Yokkaichi
Amagasaki
Nadahama
Osaka
Wakamatsu
Isogo
Saidaiji
Okazaki
Actual for boilers and equivalent
Boilers are oil-fired
Coal-fired.

Absorbent
CaO
CaCO*
CaO
it
n
n
n
n
CaCO,
n ?
CaO
CaCO,
.i 3
ii
ii
ii
n
n
n

n
n
n
n
CaO
it
n
if
CaCO*
n '
n
gas flow

MW*
150
130
175
232
279
150
125
192
250 & 175
250 x 2
575 x 2
175°
250°
ti
104
500 & 350
150
160
123

295
660
240
77
125
120
120
320
265°
83
83
•Lj
Type of plant
Utility boiler
ti
it
Sintering plant
n
it
Utility boiler
n
n
n
n
Industrial boiler
Utility boiler
it
n
it
Industrial boiler
Utility boiler
Sintering plant

n
n
n
Industrial boiler
Sintering plant
it
n
n
Utility boiler
Industrial boiler
ii
for others (at 3,OOONnr/hr per
Year of
completion
1974
it
n
n
1975
ii
n
n
1976
n
H
1975
n
1976
1974
1975
n
1976
1975

n
1976
n
1974
1976
n
n
n
n
1975
IT
Mtf).
Gypsum
(tons/day)
20
20
70
120
90
27
40
95
110
190
660
200
200
200
25
350 & 240
100
40
40

100
220
80
50
20
20
30
—
60



unless otherwise noted.







-------
             Table 2   Example of operation parameters of PGD plants by-producing gypsum and calcium sulfite
Absorbent ,
Process developer _ .

precipitant capacity
( stoi chiometry ) 1 . OOONmVhi
Wet lime-limestone process
MHI (Mitsubishi-JECCO)
ii
Chemi co-Mi tsui
Mitsui-Chemico
Babcock-Hi tachi
Fuji Kasui-Sumitomo
Chubu-MKK
I shikawa j ima-TCA
Kobe Steel
CaO
CaCO,
Ca(OH)2
CaCOx
CaCO*
CaCOx
CaCOz
CaO
CaOfJ
0.95-1
1
1-1.05
0.95-1
1.1 - 1.2
1.1 - 1.2


1.05
550
750
385
840 .
1,460°)
800
250
100
350
Slurry or L/G
Type of solution " ' ' '
' absorber pH conc.$

GPa> 6.4
» 6
Venturi 7
6
ppb) 6.1
ppb) 6
Screen 6
TCA
Spray 6-8

10
10
3-5
5
20
5-6
10
2
30
liters/
i Nffi3

10
10
10-15
10-15
10
5
10
7
3
Space
velocity
in/sec

3.5
3.5
3.2
4.5
4
3
3
Pressure S02 ppm
drop*
mm H?0

120
150
400
200
850
200
100

190
in

600
1,000
2,000
1,500
1,500
500
1,500
700
300
out
^•««M»

40
100
200
150
60
20
200
50
20
Moisture
'•fo of
/rypsum

8-10
CaSOj
10-15
8-10

10-12
10-15
10
Indirect lime-limestone process
vi Kureha-Kawasaki
°" Showa Denko
Nippon Kokan
Chiyoda
Kurabo
Bowa
Kureha
Na2S03,
Na2SOj,
(NE^JgSO
dil.HgSO
(NH4)2SO
A12(S04)
CELCOONa
CaCOj
CaCO*
3, CaO
4, CaCO 3
4, CaO
*, CaCO*
j Jl
f fl.f O
l,260d)
500° 1
150
1,050
100
150
5
GP*) 6.2
Cone 6.8
Screen 6
Tellerette 1
Tellerette 4
Tellerette 4
PPb) 5.5
20
25
30
2-4
10
10
20
10
2
2
55-60
6-10.
s \
7-8®
2.5

3
1
2
1.5
2-2.5
150
250
250

100
100
280
1,070
1,400
700
1,600
1,500
600
1,400
5
40
30
60
80
20
1-3
6-8
8-10
8-10
7-9
8-10
10-12
6-7
a)  Grid packed      b)  Perforated plate      c)
    For tail gas at 25°C.  L/G 6-10 for flue gas.
*  Including cooler, absorber and miet eliminator.
e)  For tail gas at 25°C.  L/G 6-10 for flue gas.      f)  In CaCl0 solution.
Four scrubbers in parallel
               2
d)  Two scrubbers in parallel
g)  Including limestone scrubbing.

-------
Table 3      Major indirect lime-limestone process installations completed in and after 1973
             (double alkali type)
Year of

Process developer
Chiyoda
n
ti

n
n
n
Kureha-Kawasaki
11
n
u
Showa Denko
u
ti
Showa Denko-Ebara
n
ti
"
Tsukishima
u
Kurabo Engineering
ii
.1
Dowa Mining
M
Absorbent ,
precipitant
dil. H2SO., CaCO,
It
It

tt
11
n
Na2 5, a ^
"
It
«
tt
M
tt
It
11
II
1!
Na2SO^, CaO
M
(NH4)9S04, CaO
it
If
A12(S04)V CaC03
"

User
Dai eel
Mitsubishi Chem.
Mitsubishi Pet.
Chem.
Hokuriku Electric
Hokuriku Electric
"
Tohoku Electric
Shikoku Electric
Shikoku Electric
Kyushu Electric
Showa Denko
Showa Pet. Chem.
Kanegafuchi Chem.
Nippon Mining
Poly Plastic
Kyowa Pet. Chem.
Toho Zinc
Kinuura Utility
Daishuwa Paper
Kurarey
Daicel
Jujo Paper
Dowa Mining
Naikai Engyo

Plant site
Aboshi
Yokkaichi
Yokkaichi

Toyama
Pukui
Toyama
Shinsendai
Anan
Sakaide
Buzen
Chiba
Kawasaki
Takasago
Saganoseki
Fuji
Yokkaichi
Annaka
Nagoya
Fuji
Tamashima
Aboshi
Ishinomaki
Okayama
Tamano


com-
Mw Type of plant pletion
31
130
230

250
350
250
150
450
450
450x2
150
62
93
37
65
46
43
63
85
31
53
65
100
30
Industrial boiler
n
"

Utility boiler
"
»
n
n
n
"
Industrial boiler
n
"
H SO . plant
Industrial boiler
"
H2 SO . plant
Industrial boiler
"
"
u
"
H2S04 plant
Industrial boiler
1973
1974
1974

1974
1975
1976
1974
1975
1975
1977
1973
1974
1974
1973
1974
1974
1974
1974
1975
1974
1975
1975
197/1
1976
Gypsum
(tons/day)
23
60
35

180
290
180
40
300
300
500
110
70
80
CaSOj
70
35
70
40
45
30
40
40
24
30
a  Actual for boilers and equivalent gas flow for other plants.  Boilers are oil-fired.

-------
                 Table  4   SO-  removal  installations which by-produce sulfuric acid and sulfur
in
oo
Year of
Process developer
Wellman-MKK
n
n
n
M
'•
»
"
Wellman-SCEC
n
"

ii
n
Onahama-Tsukishima
Mitsui Mining
Chemi co-Mi tsui
Mitsui Mining
Sumitomo H.I.
Hitachi Ltd.
"
Nippon Kokan
11
Shell
Mitsubishi-IFP
TEC-IFP
Absorbent User
Na2SO, Japan Synth. Rubber
*• ^ Chubu Electric
' Japan Synth. Rubber
1 Toyo Rayon
' Mitsubishi Chem.
1 J. N. Railways
' Kurashiki Rayon
" Shindaikyowa Oil
" Toa Nenryo
" Sumitomo Chem.
11 Fuji Film

" Sumitomo Chem.
" Sumitomo Chem.
MgO Onahama Smelter
" Mitsui Mining
" Idemitsu Kosan
ZnO Mitsui Mining
Carbon Kansai Electric
11 Tokyo Electric
11 Unitika
(NH4)2SO, Nippon Kokan
n ii
CuO Showa Yokkaichi
(NH4)2SO, Maruzen Oil
" Fuji Oil
Plant site
Chiba
Nishinagoya
Yokkaichi
Nagoya
Mizushima
Kawasaki
Okayama
Yokkaichi
Kawasaki
Chiba
Fuji

Chiba
Niihama
Onahama
Hibi
Chiba
Kamioka
Sakai
Kashima
Uji
Keihin
Fukuyama
Yokkaichi
Shimozu
Chiba
By-product
MW8* Type of plant completion (tons/day)
70
220
150
103
186
217
12?
124
23
120
50

180
50
28
25
162
16
53
150
57
50
253
40
14
3
Industrial boiler
Utility boiler
Industrial boiler
it
»
it
it
n
Claus furnace
Industrial boiler
ti

n
H
Copper smelter
H.S04 plant
Claus and boiler
HgSO. plant
Utility boiler
"
Industrial boiler
Sintering plant
n
Industrial boiler
Claus furnace
"
1971
1973
1974
it
11
1975
"
11
1971
1973
1974

1975
••
1972
1971
1974
1975
1971
1972
1975
1972
1976
1975
1974
11
HSO
II '
n
ti
n
it
it
"
S
H2S04
Liquid
so2
H2S04
it
it
"
S
H2S04
It
H2S04b
H2S04
(NH.)2SO.
n
S
S
S
44
88
44
50
88
110
66
60

55
13

88
25
240
18

12
17







   a Actual for boilers and equivalent gas flow for others.  Boilers are oil-fired.

   b Reacted with limestone to produce gypsum.

-------
Table 5   Major SO- recovery processes and plant  capacities
^vpez-asj

Process
Mitsubishi (MEL)
Chiyoda
Babcock-Hi tachi
Wellman-MKK
Fuji Kasui-Sumitomo
Oji
Kureha-Kawasaki
IHI-TCA
Mitsui-Chemico
Tsukishima-Bahco
Kurabo
Showa Denko-Ebara
Chemico-IHI
Kureha
Wellman-SCEC
Nippon Steel
Kobe Steel
Nippon Kokan
Tsukishima
MKK
IHI-TCA
Hitachi Ltd.
Kurabo
Kawasaki
Chemico-Mitsui
Dowa
Chemi co-Mi t sui
Showa Denko
Chubu-MKK
Nippon Kokan
11
Sumitomo H.I.
Shell
Onahama-Tsuki shima
Mitsui Mining
.onaj. 0.1, ena i.y(oj

Absorbent
CaO, CaCO,
H2S04-CaC02
CaCO
Na2SO,
CaO, CaCO,
NaOH
Na2SO, - CaCO,
NaOH
CaCO,
NaOH
NaOH
Na2SO -CaC05
CaCO,
NaOH
Na2SO
Slag
CaO
NH3
Na2SO,-CaO
NaOH
CaO, CaCO,
Carbon
(NH )2SO -CaO
CaO, CaCO , MgO
MgO
Al2(S04)5-CaC05
Ca(OH)2
NaOH
CaCO,
CaO
(NH4)2S05-CaO
Carbon
CuO
MgO
n


By-product
Gypsum
Gypsum
Gypsum
H2S°4
Gypsum
Na2S03
Gypsum
Na2SOj
Gypsum
Na2S05
Na2S05
Gypsum
n
Na2SO,
H2S04
Gypsum
it
(NH4)2S04
Gypsum
Na2S°4
Gypsum
H2S04,Gypsun
Gypsum
it
so2— > s
Gypsum
CaSO,
Na2SO,
Gypsum
it
ti
H2SO/
so2— * s
H2S04
n ^

Number
of plants
31
15
9
10
7
49
3
26
5
20
70
16
2
8
6
2
3
1
4
17
3
i 2
5
3
1
4
1
2
2
2
1
1
1
1
1
Total
capacity
1 .OOONm Vhr
17,908
5,488
5,341
4,120
3,954
3,839
2,940
2,787
2,739
2,655
2,580
2,458
1,800
1,431
1,298
1,170
1,070
760
714
693
650
590
588
582
500
430
385
370
311
227
150
150
120
88
80
 Total
333
70.972

-------
       Table 6   Classification of major processes and


                plant capacities (operational at end 1976)
Classification
of processes
Sodium sulfite recovery
Gypsum recovery
Wet-lime/limestone
Indirect-lime/limestone
(Double alkali type)
Sulfuric acid recovery
Other
Total
Number
of units
192

68
49
20
4
333
Capacity
(X 106Nm5/hr)*
14.5

36.4
12.6
5.9
1.8
71.0
* 10 Nnr/hr is equivalent to 330MW.
 Flue

  gas


 130° c
Water




    Cooler
         130°c
 Electrostatic

  precipitator
   Sludge
          Filter
   Waste-

    water
                     60° C
       Scrubber


           60° C


           f
                                      Water      Direct-fired

                                    (occasional) Cheater
                  CaO
 1
       CaO,CaCO,     pn.  ..     J
           ',         controller   '
           1                  Air
Neutralizer
                            Centri-

                               fuge
                                               Water(with

                                               or without)
                                                          Gypsum
Figure 1  Schematic  flowsheet of wet lime/limestone gypsum process
                                 60

-------
   Table 7   Wastewater from PGD plants

Process
Mitsubishi (MET)
it
it
Mitsui-Chemico
Babcock-Hitachi
it
Chubu-MKK
Showa Benko
Chiyoda
it
Wellman-MKK

User
Kansai Electric
Kyushu Electric
Ghubu Electric
EPDC
Chugoku Electric
it
Ishihara Chemical
Showa Denko
Hokuriku Electric
«
Ghubu Electric

Plant site
Kainan
Karita
a)
Owase
Takasago '
Mizushima
Tamashima
Yokkaichi
Ghiba
Toyama
Fukui
Nishinagoya

MW
150
188
750
250
105
500
85
150
250
350
220

Inlet
SO?
(ppm)


1
1

1
1
1

1
1
270
600
,480
,500
400
,500
,300
,400
610
,540
,800
Waste-
water
Gyp sum( t/hr)
(t/hr) (A) Solid(B)
1.5
3.7
14.0
5.0
2.5
5.0
3-5
3-5
15.0
24.0
3.0
0.
2.
29.
10.
0.
19.
2.
5.
7.
14-
-
9
2
0
0
9
5
2
2
5
0
-
Water
ratio
(A+C)
Moisture(C) (A+B+C)
0.1
0.2
2.9
1.1
0.1
1.9
0.2
0.5
0.7
1.4
—
0.64
0.64
0.37
0.38
0.74
0.26
0.62
0.45
0.73
0.64
— —
Waste-
water
(kg/M¥hr)
10
20
19
20
23
10
41
23
60
68
14
a)  Designed value; the plant is under construction.



b)  Coal-fired boiler.  All others are for oil-fired  boilers.

-------
      2    Performance of major new plants for utility boilers

2.1  Earita plant,  Kyushu Electric

         MHI  line-gypsum process (Mitsubishi-JECCO process)  '

The Karita plant with a capacity of treating 550,OOONBr/hr flue
from an oil-fired boiler  (188MW equivalent) went into operation in
November 1974.  The plant is based on the one-absorber system  (Figure
2) and uses lime.  Operation parameters are listed in Table  2.   Flue
gas is first cooled to 55-60°C  in a cooler and led into the  absorber.
The absorber inlet S02 concentration is about 500ppm and the outlet
20-30ppm.  The load fluctuates  between 550,000 and 300,OOONm3/hr
every day.  The flow rate of the slurry is kept constant while  the
amount of lime is adjusted  with the load.

The plant has been operated at  100^ availability since its start-up
except for the scheduled shutdown of the boiler from April 1 to May 14.
On February 24, 1975, a  now rate adjusting bulb was stopped up but
was repaired without interrupting the scrubber operation. In  the
inspection in April 1975 a  considerable scaling was found on the mist
eliminators.  The eliminators had been washed with circulating liquor,
but since May 1975 they have been washed alternately with fresh water
and the. liquor.  Water balance  is shown in Figure 2.

Two new units (175 and 250MV) by the MHI process using limestone at
Karatsu station, Kyushu Electric, has started operation recently.
Designed operation parameters are shown in Table 2.
Industrial
   water
(19.67mVhr)
                            Cooling

                            Pump seal
                         Mist eliminator

                          Pump seal
 Filtered
   water
        (6.71mVhr)
                          Fan cooling
* Evaporation
(21.21m*/hr)

' Wastewater
 (3.71m'/hr)

  To gypsum
                                          (0.l6mVhr)
                                                   Evaporation
                                              .30mVhr)
     Figure 2  Water balance ( Karita plant, MHI  process)
                                 62

-------
2.2  Takasago plant, Electric Power Development Co.

                                                                 1  2)
    Mitsui-Chemico limestone-gypsum process (Mitsui  Mike process)  '  '

This plant is based on the Mitsui-Chemico process developed by Mitsui
Miike Machinery Co.  It is the first Mitsui-Chemico  process unit  to
be used commercially for a coal-fired utility boiler,  and has a capa-
city of treating 840,OOONm3/hr flue gas from a 250MW boiler.  The
plant consists of two single-stage Chemico scrubbers placed in series,
a pH controller to reduce the pH of the calcium sulfite slurry by
introducing the flue gas, and two reactors for oxidation of the sul-
fite to gypsum by air.  A catalyst is used to promote  both SOg removal
and oxidation.  The flue gas passed through an electrostatic precipi-
tator containing l,500ppm S02 and 8Qmg/Nm* dust is treated to remove
90$ of the S02 and 75$ of the dust.  Limestone, 95$  under 325 mesh,
is used.

The plant went into operation in January 1975 and has  been in smooth
operation except for several short shutdowns to clean  up the mist
eliminators.  The eliminators have been washed with  the circulating
liquor and fresh water.  The availability reaches 98$  except in
periods of boiler shutdown.  The load fluctuates between 170 and  250MW
every day.  The slurry flow rate in the scrubbers is kept constant.
The daily requirements for the operation are as follows:
                                               Oil 41 liters(reheating)
                                       Industrial water 1,200 tons
     CaCOj   118 tons     Power 136,OOOkWhr
     Steam 13 tons (heating of oil)
     Catalyst $350

The coal contains about 500ppm chlorine.   In order to keep the chlo-
rine concentration of the scrubber liquor below 7»000ppm,  about  5
tons/hr of wastewater is purged.  The relationship between the amount
of purge water and the chlorine concentration of coal and  equilibrated
scrubber liquor is shown in Figure 3-
          15  -
          10
       ID

       1   5

       0>
       hO
       f-,

       CU   0
                                  3,000ppm
              0           kOO          800

                   Chlorine  in coal(ppm)
                                                   1,200
           Figure  3  Purge  water and  chlorine  concentration
                                   63

-------
Another plant with the same capacity is near completion at Takasago.
The new plant has one reactor in place of the two reactors of the
present plant.
 2.3 Tamashima plant, Chugoku Electric
                                      1  2)
               Babcock-Hitachi process  '   '

 Hitachi Ltd.  constructed a large  plant  (500MV full scale) using
 Babcock-Wilcox scrubbers and an oxidizing  system to by-produce gypsum.
 Pour scrubbers were installed of  which  three are is use and one is
 for stand-by.  The plant has been in operation  since July 1975.  Flue
 gas from an oil-fired boiler containing l,500ppm S02 is first cooled
 to 55°C in venturi scrubbers and  then led  into  absorbers with 6
 stages of perforated plates where about 95# of  S02 is  removed.  The
 pressure drop is fairly heavy~230mm H20 in the venturi and 600mm in
 the absorber.  A new type of mist eliminators designed by Hitachi
 are used.

 The plant has been used for base-load operation.  Power consumption
 reaches 3«6# of the power generated. Operation has been trouble-
 free except for minor erosion and scaling  problems.  Availability
 of the plant has reached about 97$.

 About 93 tons/hr of industrial water is used of which  13  tons are
 for washing the mist eliminator.   At the beginning of  the  operation,
 about 5 tons/hr wastewater was purged to keep  the  chlorine concentra-
 tion of the circulating liquor below l.OOOppm.  The  amount of waste-
 water has been substantially reduced recently.  Another plant  (350MW)
 at TanmoMmft has been in operation since  early 1976.
  2.4  Sakaide plant, Shikoku Electric

              Kureha-Kawasaki sodium limestone process

  The Sakaide plant with a capacity of treating l,260,OOONmV^ nue
  gas containing l,050ppm S02 from a 450MV oil-fired boiler went in
  operation in August 1975. .The process is similar to that for Shinsend
  plant, Tohoku  Electric1. 2) except that the Sakaide plant has a unit
  to remove magnesium from the circulating liquor (magnesium tends to
  delay the reaction of limestone with sodium sulfite) and that gypsum
  is washed with water  to reduce sodium, to make it available for wall-
  board production.  Operation parameters are shown in Table 2.  More
  than 9956 of S02 is removed by sodium sulfite scrubbing.  Operation
  has been smooth since start-up.  Load fluctuates between 1,260,000
  and 534,000»m3/hr every day.  The sodium sulfate concentration of
  the circulating liquor has reached 11-12$ exceeding the designed valu*
  of 8$ but a Mgh S02  recovery ratio has been attained.
                                64

-------
This plant is characterized by the absence of water purging from the
system.  At full-load operation, 65.5 tons/hr water is charged of
which 29.9 tons are used for gypsum wash.  The same amount, 63.5
tons, leaves the system, of which 62.0 tons are evaporated and the
rest is contained in the by-product gypsum.  Chlorine concentration
of the circulating liquor has reached 5,500ppm but has caused no
corrosion problem because the liquor contains little oxygen which
tends to cause stress corrosion in the presence of much chlorine.
In the Kureha-Kawasaki- process, oxidation is carried out with the
calcium sulfite separated from the mother liquor and therefore the
oxygen content of the liquor is kept low.  It seems that the chlorine
input from fuel and water and output with gypsum containing 7-9$ mois-
ture have been nearly equalized at the 3»500ppm concentration.

2.5  Pukui plant, Hokuriku Electric

         Chiyoda dilute sulfuric acid limestone process

The Fukui plant has been in smooth operation since its start-up in
summer 1975 treating l,050,OOONm3/hr flue gas from a J50MW oil-fired
boiler.  The plant has a double cylinder type absorber-oxidizer.
Operation parameters are listed in Table 2.  The requirements are
150 tons/day limestone, 3.8m3/hr oil for reheating, ll.SMW electric
power, and 2,000 tons/day industrial water.  About 580 tons/day
wastewater is purged after being treated to keep the pH at 7»9» sus-
pended solids below 5 mg/liter and COD below 10 mg/liter.  Plant
operation is easy and requires only 2 operators per shift.


          3     New developments in S02 removal processes  for

                steel industry

A major S02 source other than boilers is iron-ore  sintering plants,
which  emit 300,000-1,000,OOONm3/hr waste gas per plant containing
400-600ppm S02 and much dust.   Several plants  for  desulfurization of
the gas has been constructed by MHI  since  1972 and are in  successful
operation.  Recently three processes have  been developed mainly for
the treatment of the gas.  New  developments will be described  below.

3.1  Fuji Kasui-Sumitomo limestone-gypsum process

Fuji Kasui Kogyo has developed  a limestone-gypsum  process  jointly with
Sumitomo Metal using the Moretana  scrubber and recently completed two
                                  65

-------
plants for Sumitomo.  One is Vakayaaa plant (375,OOONmyhr) and the
other Kaahima plant (800,OOOHm3/hr).  Operation parameters are shovm
in Table 2.

The Moretana scrubber is fitted with four perforated plates made of
stainless steel.  The holes, I/O ratio, and the gas velocity are
specially designed to give extreme turbulence, producing foam layers
400 to 300mm thick which result in a high 802 recovery ratio.  An
oxidizer developed by Fuji Easui is used.  Other facilities and sys-
tems are similar to those of other Japanese processes (Figure 1).

The Vakayama plant started operation in May 1973 and has been in
smooth operation except for a defect in the plastic lining in a cool*
which was found at the beginning of the operation and was repaired.
Availability of the plant is 98$ except for the scheduled shutdown of
the sintering plant that normally occurs one day in about two months.
The mist eliminator is washed intermittently (once in JO minutes)
with the circulating liquor and fresh water alternatively.  The
pressure drop in the mist eliminator which is 30mm H20 at the beginnf
gradually increases while it is washed with the circulating liquor.
When the pressure drop reaches ^Oam, fresh water is used in place of
the liquor until the pressure drops back to 50mm.  The ratio of liqof
to fresh water is about 80 to 20.
 The KftgMi'ft plant started operation in September 1973 and has been
 operated at 100$ availability.

 3.2 Nippon Steel slag process (SSD process)

 Nippon Steel has developed a process which uses converter slag as th
 absorbent.  The  slag contains about 40$ CaO, 16$ Si02, 3$ MgO, 3$
 Al2<>5, and  33$ FeO and Fe^j and is useless.  Nippon Steel has opera
 a prototype plant with a capacity of treating 200,OOONnK/hr waste ga
 from a sintering plant since 1974 and is now constructing a commerci
 plant  (l,000,0009m3/hr) which will come on-stream in 1976.

 The process is similar to other Japanese lime/limestone-gypsum pro-
 cesses except  that it uses no oxLdizer.  The gas is cooled and led
 into two absorbers in series to remove 95$ of S02-  The slag is fed
 to the second  absorber to produce a calcium sulfite slurry which is
 led to the  first scrubber and entirely oxidized into gypsum in the
 scrubber due to  a low pH and the presence of much iron compounds
 which  act as a catalyst.  The by-product gypsum contains about 40$
                              66

-------
impurities and has been discarded.  There has been some scaling
problem to be solved to ensure a long-term continuous operation.
The process may be useful for steel producers who normally have
large amounts of useless slag.

3.3  Kobe Steel calcium chloride process (Cal process) '

Kobe Steel has developed a new process using a 30$ calcium chloride^
solution dissolving lime as the absorbent.  A pilot plant (50,OOONmyhr)
has been operated and two commercial plants (Table l) have just come
on-stream to treat waste gas from iron ore sintering plants.

Calcium chloride solution dissolves 6-7 times as much lime as does
water.  High S02 recovery is attained at a low L/G (Table 2).  The
flowsheet is similar to that in Figure 1 except that calcium sulfite
discharged from the scrubber is separated from most of the mother
liquor, reslurried with water, and sent to an oxidizer and that the
liquor from the gypsum centrifuge is sent to a cooler giving no
wastewater at all.  Since vapor pressure of the liquor is low, the
temperature of the gas after the  scrubbing reaches 70°C as compared
with the 55-60°C for the usual wet process and thus less energy for
reheating is required.

The mist eliminator is washed with the circulating liquor.  The
solubility of gypsum in the liquor is very low (nearly 1/100 of that
in water) and the evaporation of  the liquor does not cause scaling.

Continuous operation of the pilot plant for about 6 months showed
that a soft deposit formed on the wall of the absorber when the L/G
ratio was smaller than 1 but the  deposit could be washed off by
using an L/G larger than 2.  A highly corrosion-resistant material
is required for the cooler.  Operation of commercial plants will
enable further evaluation of the  process.
             4     Simultaneous removal of SOg and N0x

 4.1  Outline

 NOX removal technology has been developed in Japan remarkably since
 1973 when a stringent ambient standard for N02 (0.02ppm in daily
 average) was set forth.  Among many processes, selective catalytic
 reduction of NOX by ammonia at about 400°C has been considered most
 feasible.  Several commercial plants treating flue gas from oil-fired
                                  67

-------
industrial boilers  (200,000-450,OOONm^/W) are in operation and many
plants are under construction with the reduction process.  This dry
denitrification process, however, is not suited for use in conjunc-
tion with FGD  (for  which a wet process is economical), because it will
require a large heat  exchanger and a considerable amount of energy for
heating (No. 3, Figure  4) or an expensive hot electrostatic precipi-
tator (No. 4,  Figure  4).  Tinder this situation, many dry and wet
processes for  simultaneous removal of S02 and NOj (No. 1 and 2,
Figure 4) have been developed and many plants are in operation as
shown in Table 8.

4*2  Oxidation reduction process

One of the most promising ways for simultaneous removal is the reduc-
tion of N02  into N2 using sulfite ion formed by S02 absorption.

         4S05~"  +  2N02 = 4SO ""  +  N2
Since NO is  less reactive, it must be oxidized into NOg first.  As
the oxidizing  agent, C102 is used by Fuji Kasui and Ozone is used
by Chiyoda and HHI.  Ozone is a good oxidizing agent but is costly.
To convert NO  in flue gas from a 350MW boiler to N02, 400-600kg/hr
($500-700/hr)  of ozone  is required assuming that the gas contains
200-300ppm NO. NO  concentration in flue gas should be kept low by
combustion control.  Chlorine dioxide (ClOg) ia also very effective
and converts NO to  N0£  within a second.  Although it adds a chloride
to the by-product and complicates treatment, the cost is about one-
half that of ozone.

Fuji Kasui has constructed three plants using the Moretana scrubber
(Table 8).   Flue gas  containing 1,000-1,500ppm S02 and 200-300ppm
NOx is cooled  to 55-60sC, mixed with equimolecular 0103, and subjected
to sodium scrubbing to  remove about 90$ of NOj and 98^ of S02*  About
one-half of  the removed NOX  is converted to Ng and the rest forms
sodium nitrate and  nitrite.3)  The liquor from the scrubber is evapo-
rated to recover a  high-purity sodium sulfate for chemical use.  The
rest of the  by-product  has to be discarded.

Chiyoda has  developed a simultaneous removal process by simply adding
an ozone-oxidation  step  before the standard FGD process.  Together
with more than 90$  of S02, more than 60$ of NO* is removed, forming
N20 and some calcium  nitrate.')  Purge liquor is sent to a wastewater
treatment system.
                               68

-------
No.l
No. 2
No.3

•
400 '

AH
150


• LP
150

DBS
DDN
150
^

B
400

AH
150

EP
150 J WDS
] WON
60 ^

H
                                                         120
                                                     1
                                            160

B




400 '






AH




150
"1





El









4C
15




)0
0
V



T)D

1I[T\
•JVLJ


N

- 1 "
"


r
^00
0
->



H

i
1 >
H ! E



" 300
 No.
B
i+UU

EP
400
V
)
DDN
q-uu
•^

AH
.LOU
. . .)

WDS
                                                   60
                                                        H
                                                120
 No.5
 No. 6
B
M-VJV

DDN


AH
-LOU

EP
                                        160
                                   (With or
                                    without)
WDS
DU
r
H
                                                  120


B

400
X1


EP

400

DDS
DDN
400

AH

160
••v

        Boiler
              Air heater
                 Electrostatic
                  precipitator
    DDS
Dry
FGD
Wet
 FGD
                                             Dry denitrification
            Wet
            denitrification
                      Heater
                   Heat
                    exchanger
     Figure 4  Models of combination of FGD and denitrification

                (  Figures show gas temperature)
                               69

-------
            Table 8    Installations for simultaneous removal of S02 and NOX
                                                       Capacity
developer
Plant owner
                                          Plant site   Km
S/£r y Source of gas  Completion  By-product
Vet process
Fuji Kaaui - \
Sumitomo Metal j
/
it
„
Chiyoda
Mitsubishi H.I.
Ishikawajima H.
Kobe Steel
n
Kawasaki H.I.

Kureha Chemical

'Oxidation \
I reduction j
n
ii
"
..
I. "
»
n
/Magnesium\
* scrubbing'
Reduction
Chisso Engineering "
Mitsui S.B.
Dry process
Shell
Unitika
Sumitomo H.I.
Ebara
"

CuO, NH,
Carbon, NH,
5
Electron beam

Sumitomo Metal

To shin Steel
Sumitomo Metal
Chiyoda
Mitsubishi H.I.
Ishikawajima
Kobe Steel
n
EPDC

Kureha Chem.
Chisso P.C.
Mitsui P.C.

Showa Y.S.
Union Glass
Sumitomo Power
Ebara

AmagftBaki

Fuji
Osaka
Kawasaki
Hiroshima
Yokohama
Kakogawa
n
Takehara

Nishiki
Goi
Chiba

Yokkaichi
Hirakata
Niihama
Fujisawa

62,000

100,000
39,000
1,000
2,000
5,000
1,000
50,000
5,000

5,000
500
150

120,000
4,500
150
1,000

Boiler*)

Furnace
\
Boiler*'
n
"
"
Furnace '
n
Boiler0)

Boiler*)
Boiler*)
n

e)
Furnace '
Boiler*)
n

Dec. 73 '

Dec. 74
Dec. 74
1973
Dec. 74
Sep. 75
Dec. 73
Mar. 76
Dec. 75

Apr. 75
1974
"

1975
"
1974
tt
/ NaNO,, NaClx
1 Na9SO, J
f- *f
n
n
/ Gypsum \
^Ca(NO,)2
Gypsum, Ng
"
Gypsum, Ng
"
/ Gypsum \
^Ca(NO,)2
Gypsum, N2
(NH4)2S04
H2S04, N,

SO^ S, N2
H2S04, Ng
n
Mist, dust
a)  Oil-fired boiler         b)  Metal heating furnace
d)  Iron ore sintering furnace        e)  Glass melting furnace
                                      c)   Coal-fired boiler

-------
HHI has developed a simultaneous removal process based on the lime/
limestone gypsum process.  Ozone is used for oxidation and a catalyst
is used to convert virtually all of the removed NOX to N2.  More than
7<# of NOX is removed with more than 90^ of S02.  A portion of the
circulating liquor is treated  to remove certain compounds formed by
a side reaction.

Isblkawa j ima-Har ima Heavy Industries (iHl) has also developed a
simultaneous .removal process based on wet-lime/limestone scrubbing
using ozone ,

Kobe Steel has developed a process using a calcium chloride solution
containing Ca(OCl)2 as an oxidizing and neutralizing agent. 5)   Follow
ing tests with a pilot plant (l,OOONm3/fcr ) , a larger pilot plant
              is under construction.  The by-products are gypsum
and H2 or HNOj.

4.5 Other wet simultaneous removal processes

Kawasaki Heavy Industries has operated a pilot plant with a capacity
of treating  5,OOONm3/hr flue gas from a. coal- fired boiler using
aagnesium scrubbing and lime addition. 3 )  Magnesium sulfate and
calcium nitrate  are formed and are reacted to precipitate gypsum.
Calcium nitrate  is obtained as a by-product.

Chisso Eng.  has operated a pilot plant in which S02 and NOj are
absorbed with an ammoniacal solution containing a catalyst to reduce
the absorbed N0_ into NH} by ammonium sulfite and bisulfite to by-
produce ammonium sulfate.5'  The overall reaction may simply be ex-
pressed in  the  following way:

       2NO
 Eureha Chemical has modified the sodium acetate-limestone PGD Process
 for simultaneous removal to by-produce gypsum and N2.?'  Although the
 total system is fairly complex, an intermediate  compound— sodium
 imidodisulfonate, NH(S03Na)2~nas been found useful as a substitute
 for sodium tripolyphosphate which has been used  in large amounts as
 a builder for detergents and has caused environmental problems.  Com-
 •ercial use of the imidodisulfonate would make the Kureha process pro-
 mising.

         Mitsui  Shipbuilding has recently developed a process that
 uses a solution  containing ferrous ion and EDTA to absorb S02 and NO*.
 fhe liquor after the  absorption is sent to a reducing step to convert
 the ferric ion formed by oxygen in flue gas back to ferrous ion.
 Most of the reduced liquor is returned to the absorption step;  the
 rest is sent  to  a stripper which generates concentrated S02 and NO,
 ttoich are sent to a sulfuric acid  plant.   The NO may be reduced to N2
 by a conventional process or used  for nitric acid production.
                                   71

-------
4.4  Activated carbon process

Takeda Chemical has produced an activated carbon containing metallic
components to promote the reaction of NOX with ammonia to form N2«
Higher temperature is favorable to the reaction but decreases the
SC>2' absorbing capacity (Figure 5)»  Optimum temperature for simul-
taneous removal by this process is about 250°C.
        100
          90
      580
      t-,
      •H
      OS

      §  70
      K
         60
          100
150       200     250

   Temperature (°C)
300
350
     Figure 5   Schematic  diagram of S02  and NOX removal
                by activated carbon at  different temperatures
                and SVs.
Unitika Co. recently started operation of a pilot plant with a capacity
of treating 4,500Nm3/hr flue gas from a glass melting furnace.  About
90# of »0X and SC<2 is removed.  The carbon that has absorbed SC>2 is
heated to 350°C in a reducing hot gas to release concentrated S02 for
sulfuric acid production.  Ammonium sulfate and sulfite which tend to
fora on the carbon are decomposed to S02 and ^ in the regeneration step.
                                  72

-------
Sumitomo Heavy Industries (former Sumitomo  Shipbuilding)  has
developed a similar process using moving beds  of carbon.

Hitachi Ltd. has found that activated carbon treated with ammonium
bromide is effective even at 100°C for NOX  reduction by ammonia.4)
The low temperature activity may result in  energy saving (No. 1,
Figure 4)-

Dust, ammonium sulfate and sulfite which tend  to deposit on the
carbon, however, present a problem for carbon  processes.

4.5  Other dry simultaneous removal  processes

Copper oxide used as an absorbent of S02 in the Shell process works
as catalyst in the reaction of HOX with ammonia.  Yokkaichi plant,
SIS, treating 120,OOONm3/hr flue gas from an oil-fired boiler by the
Shell process has introduced ammonia into a reactor at 400°C since
1975.  Up to about 70$ of NOX can be removed.   Copper sulfate formed
by SOp absorption is reacted with hydrogen  to  generate concentrated
SOo, which is sent to a Glaus furnace to produce sulfur.

Ebara Manufacturing Co. has developed a unique process for simultane-
ous removal by electron beam radiation. 3)   Flue gas is introduced in
a reactor and exposed to the beam.   About  80$ of 302 and 90$ of NOX
can be removed, forming a sulfuric acid mist and a powdery product
containing S, N, 0 and H, which  are  caught  by an electrostatic pre-
cipitator.  Investment cost and  power consumption seem high.

5. Economic aspects and by-products

5.1  Flue gas desulfurization

Investment cost in battery  limits  for plants (150-300MW) using wet
and indirect lime/limestone processes by-producing gypsum was $20-30/k¥
in 1973, increased to $?0-100/kW after  the oil shock, and recently
decreased to $50-80/kW.  Plants  by-producing sulfuric acid cost 20-70$
aore.   The present desulfurization cost  by lime/limestone processes
including a depreciation for  seven years ranges from $18 to  $22/kilo-
liter  oil or  4  to 5  milAWhr  at  6»0°° hours annual operation.
The current price of a kiloliter of heavy oil is about $75 for high-
sulfur oil  (S=3$) and $100  for  low-sulfur oil (3=0.3$).  The economic
advantage of PGD over the use of low-sulfur fuels and the reliability
of PGD processes encouraged industry and power companies to  construct

-------
many plants.  But the plant building rush subsided recently with the
oversupply of by-products.


In addition to the rapid increase of by-products as shown in Figure
6, the recent economic depression in Japan which resulted in the
decrease in the demand has enlarged the oversupply (Figure 8).
Last year, the total inventory of by-product gypsum at phosphoric
acid and FGD plants exceeded 1 million tons—which is very little
compared with that in Florida, U.S.A.,but presented a problem in
Japan which has a limitation in landspace.  Efforts have been made
to develop new uses for gypsum particularly in the field of new con-
struction materials.
      10,000
 OJ
 "0
 •H
 O
 08
 P<
 
-------
   80
   60
<
o
o
•H
i-,
  20
   0
     1969         1971        1973        1975


         Figure 7  Price of by-products
QuantityC millions of tons)
o ru •£• <^
-
Demand
OU
B
C
Supply
OS
R
P

Demand
OU
B
C
>>
,-H
O<
DH
3
to
OS
R
P

Demand
OU
B
C
Supply
OS
R
P

         1970
1973
1975
 Demand:  C:Cement   B:Board   OU:Other uses

 Supply:  P:Phosphogypsum  R:Recovered  OS:Other sources


   Figure 8  Demand for and supply of gypsum in Japan
                           75

-------
The by-product gypsum from flue gas from coal-fired boilers at Omuta
plant, Mitsui Aluminum and Takasago plant,  E.P.D.C. contains less
than about 6$ impurities (fly ash,  limestone,  etc.) and has been
used mainly  for  cement and partially for wallboard.  Isogo plant,
E.P.D.C. and Wakamatsu plant, Nippon Steel, both under construction,
will produce gypsum with more than  20$ impurities to be discarded.
The first unit of Mitsui Aluminum in operation since 1972 by-producing
calcium sulfite  sludge will have to be modified to by-produce gypsum
because the  sludge pond is nearly full.  Although gypsum is in over-
supply, throw-away calcium sulfate  sludge will not increase because
the sludge needs much more land space than  does gypsum.

Elemental  sulfur is a desirable by-product  but FGD processes by-
producing sulfur seem very expensive except for oil refineries that
have  Glaus furnaces.  To by-produce elemental  sulfur, hydrodesulfuri-
 zation of heavy oil may be more economical  while it is more expensive
 than FGD by the wet lime/limestone  process  (Table 9).

 Table 9   Cost comparison at various desulfurization ratios

           (f/kl, at 6,000 hours a.*»mal  operation)

FGD (wet-lime/limestone process)
Hydrodesulfurization of oil
70*
17
17
809&
18
19
90%
19
22
97$
20
27
 Two commercial hydrodesulfurization plants recently started operation
 to reduce sulfur to below 0.3$ (over 90$ removal) using heavy oil
 with little metallic impurities  which tend to poison  the  catalyst.
 Most heavy oils are rich in impurities and difficult to desulfurize
 over 80$.  For such heavy oils, vacuum distillation followed by
 hydrodesulfurization of the distillate and decomposition of the re-
 sidue with partial gasification has been considered rational.   The
 first commercial plant with the decomposition started  operation in
 January 1976 using the Xurena process (or Eureka process).  The
 second plant is near completion using the Flexicoking  process.
                            76

-------
5.2  Simultaneous removal

Fuji Kasui claims that the plant cost for simultaneous removal ranges
from $60 to 90/kW and the operation cost is roughly f 30/kl oil in-
cluding depreciation for 7 years.  The costs for other processes are
uncertain because those are still in a test stage.

Activated carbon processes have an advantage in that they are a dry
process and can be operated at relatively low temperatures.  Carbon
for simultaneous removal is fairly expensive at present-, viz., about
$8,000/t, while carbon for PGD costs about $3,000/t.  To treat flue
gas from a 350MW boiler at a SY of 1,000, about 1,000 tons of carbon is
required.  Production of much cheaper activated carbon is desired.

The Shell process seems costly for PGD but its capability of simul-
taneous removal may compensate for the disadvantage.

Among the by-products from wet processes, sodium nitrate and nitrite
have little use.  Calcium nitrate may be used as fertilizer but its
demand is limited due to the low nitrogen content and high hygroeco-
picity.  Ammonium nitrate and nitric acid are better by-products.
 By the hydrodesulfurization of heavy oil,  30-40$ of nitrogen in the
 oil is also removed resulting in 10-20$ abatement of NOX in flue gas.
 Tests have been made to remove much more nitrogen in the oil along
 with sulfur.  To compete with possible low-sulfur low-nitrogen oil,
 the operation cost of simultaneous removal from flue gas should be
 less than about $30/kl and the process must not have difficulty in
 wastewater treatment and disposal of by-products.

 Efforts for denitrification in Japan has been exerted  mainly for flue
 gas from oil burning containing 200-300ppm NOX because oil is the
 major fuel in the country.  The need for denitrification may be larger
 for flue gas from coal burning which normally contains more than
 600ppm NO  which is not easy to reduce  to  below 400ppm by combustion
 modifications.  Denitrification will become more and more important
 in many countries as the consumption of fossil fuels increases.
                                    77

-------
                Conversion figures


               1 m^ = 35.3 cubic  feet

               1 liter =0.26  gallon

               1 kl = 6.29 barrels

               1 NnrVhr = 0.^9 scfm

             L/G   1 liter/Nm^  =7.4 gallon/1,OOOscf
                   References
l)  J. Ando, Recent Developments in Desulfurization in Japan (January 1975)
    Being published by U.S. EPA through PEDCO.

2)  G. A. Hollinden and F. T. Princiotta, Sulfur Oxide Control
    Technology, Visits in Japan (March 1974)*  Interagency Technical
    Committee.
3)  J. Ando and H. Tohata, HOx Abatement Technology in Japan for
    Stationary Sources  (March 1975).  Being published by U.S. EPA
    through PEDCO.

4)  M. Seki, Y. Sakurai and K. Yoshida, Ammonium Halide-Activated
    Carbon Catalyst to Decompose NOx in Stack Gases at Low Temperatures
    around 100°C.  American Chemical Society (August 1975. at Chicago).

 5) S. Yamada,  Y.  Watanabe and H. Uchiyama, Bench-Scale Tests on
    Simultaneous Removal of SO* and NOX by wet Lime and Gypsum Process
    Ishikawajima-Harima Engineering Review, Jan. 197&,  Vol. 16, No.i.
    (In Japanese)
                             78

-------
                FLUE GAS DESULFURIZATION ECONOMICS
 G.  G.  McGlamery,  H.  L.  Faucett, R. L. Torstrick, and L. J.  Henson

          Office of Agricultural and Chemical Development
                    Tennessee Valley Authority
                      Muscle Shoals, Alabama
ABSTRACT

     Described herein are several phases of flue gas desulfurization
(FGD) process and byproduct economic evaluations being carried out
by the TVA Office of Agricultural and Chemical Development for EPA
and the TVA Office of Power.  Included are recently updated investment
(mid-1977) and annual revenue requirements (mid-1978) for five leading
FGD processes evaluated by TVA in 1974.  Also included are results
from a cost sensitivity study of limestone scrubbing options based
on data from the EPA-Bechtel Corporation-TVA test program at Shawnee,
plus some preliminary evaluations of promising double-alkali and
regenerable processes.

     In addition to these results, several new or expanded evaluation
programs are described.  Included are:

1.   A computer program prepared by Bechtel and TVA to screen process
     options of lime-limestone scrubbing using Shawnee data.

2.   Marketing studies covering volume, pricing, and location aspects
     of various FGD byproducts.

3.   Evaluations of advanced FGD processes, N0x control processes, and
     systems for clean fuel from coal.

4.   A sludge disposal system design and cost study.

5.   Process energy optimization studies.

     The  status of these programs is also  discussed; some phases of
these programs are nearing  completion, while others  are either just
getting under way or yet to be started.
                                79

-------
                   FLUE GAS DESULFURIZATION ECONOMICS


INTRODUCTION

          Throughout the developmental stages of flue gas desulfurization
(FGD) technology, process economics has been one of the more interesting and
debatable subjects, to say the least.  From the earliest cost estimates of
the late 1960's to now, and no doubt in the future, the estimated and actual
costs of such systems have been increasing to higher and higher levels.  But,
so has the cost of most everything else in today's inflation-ridden society.
Nevertheless, not all of the increases in cost estimates for FGD systems can
be attributed to inflation, since like most new emerging technologies, as
experience and knowledge increase in the early years of application, so does
the expected real costs of such facilities.

          Some perspective can be gained from Figures 1 and 2 which show the
capital cost of the generalized limestone slurry scrubbing process as esti-
mated by TVA over the past decade.  Although the scope of these estimates has
tended to change slightly during this period, the increases are due to both
improvements in knowledge of process requirements and inflated labor and
equipment costs.

          By assuming a relatively simple example case, we have been fairly
successful in assessing the comparative economics of FGD systems.  However,
on a total magnitude basis, it has been much more difficult to predict repre-
sentative costs of specific systems.  For that matter, estimators throughout
the industry have encountered this difficulty, primarily for the following
reasons.

   1.  Variable project scope, time frame, design, and operating conditions.

   2.  Numerous process options.

   3.  Vendor optimism and user skepticism.

   4.  Confusion over regulatory requirements.

   5.  Demanding physical and chemical environment.

   6.  Variable byproducts and byproduct value/disposal cost.

   7.  Shortage of actual design and operating data.

At least for the near future-, we expect most of these reasons to remain as
impediments to total cost assessment.
                                   80

-------
              too
              80
 T
 T
                      500 MW, 3.5% S coal-fired units

                      Generalized TVA estimates
CO
           01
           (U

           c
              60
           3  40
               20
                                               Scope and technology
                1968
70
72
74
Year
76
78
                             Figure 1.    Limestone slurry scrubbing process -


                               change in investment cost during past decade.
80

-------
00
N)
            100

          0)
             80
             60
          i
          I  40
             20
                                                   3.5% S coal-fired units
                                                   Generalized TVA estimates
                                                            Mid-1977
                                                                 Mid-1972
                                     1
1
1
                         200       400      6OO      800
                                          Power unit size, MW
                     IOOO
                           Figure 2.   Limestone slurry scrubbing process -

                              range of projected investment estimates.
                    1200

-------
UPDATE OF THE PAPER "COST COMPARISONS OF FLUE GAS DESULFURIZATION SYSTEMS"

         Since 1967 the TVA  Office of Agricultural and Chemical Development
has been preparing for EPA  or its  predecessors cost estimates of various FGD
processes.  Many of these have been published or presented previously. Prob-
ably the best known set of  EPA-TVA FGD process cost estimates was  presented
ia a paper entitled "Cost Comparisons of Flue Gas Desulfurization Systems"
at the EPA symposium in Atlanta, November 1974.  Five processes were evaluated
in a single comparative study—limestone slurry and lime slurry scrubbing;
•agnesia slurry - regeneration to  98% sulfuric acid; sodium scrubbing -
regeneration to sulfur; and catalytic oxidation to 80% sulfuric acid.  The
complete data from which the  paper was prepared were later published in
January 1975 in an EPA report entitled "Detailed Cost Estimates for Advanced
Effluent Desulfurization Processes."  Because the project time frame for these
estimates was mid-1972 to mid-1975 and capital and revenue requirements were
subjected to tremendous escalation near the end of this period, TVA has up-
dated these costs to reflect  a mid-1975 to mid-1978 construction project.
Ihe revised investment costs  are shown in Table 1 and the annual revenue
requirements are shown in Table 2.  (Revenue requirements include all oper-
ating costs, such as raw materials, labor, utilities, maintenance, overheads,
capital charges, and taxes; assumes startup in mid-1978.)  The basis and data
used in updating these values are given in Table 3.

         It should be noted  that the revenue requirements shown in Table 2
do not reflect byproduct credits or sludge fixation costs.  With recent in-
creases in natural gas prices affecting the Frasch sulfur mining costs, we
have seen sulfur list prices  jump 85%  (from $35 to $65/long ton) in 1975 with
subsequent effect on sulfuric acid list prices  (from $30 to $45/ton,  100%
acid).  Although these markets are now temporarily soft and spot prices re-
flect discounting, on a  long-term basis, elemental sulfur prices can  be
expected to continue to  climb with acid prices  to follow.  Introduction of
large quantities of abatement byproducts would  certainly change the picture,
tat if smaller quantities  are introduced over a reasonable period of  time
 (at the same rate as the  total market growth), market disruption could be
held to a minimum.  Based  on recent work (paper entitled "Potential Utiliza-
 tion of Controlled SOX  Emissions from Power Plants  in Eastern United  States"
by J. I. Bucy, et al.,  presented at this meeting) TVA has completed  for EPA
on byproduct marketing,  it would not be unreasonable to expect a long-term
 net revenue of $45/long ton of sulfur or $25/ton of 100% sulfuric acid.

         Recent EPA policy statements now seem to be encouraging sludge
 fixation or similar  long-term sludge disposal  practices.  If such costs must
 be added  to those projected in Table 2 for limestone and lime scrubbing,  the
 co^>etitive position  of these processes with  respect to regenerative processes
 could change drastically.   As illustrated in  Table  4,  the effect of  byproduct
 credits  for the  regenerable processes  and sludge fixation costs  for  the non-
 regenerable systems  are significant.   Since  90% or  more of  the FGD systems
 operating or under  construction are  the nonregenerative type, the utilities
 obviously need to  take a closer look at the  regenerative systems.
                                     83

-------
                                   Table 1.   SUMMARY OF TOTAL CAPITAL INVESTMENT REQUIREMENTS

                                         FIVE LEADING FLUE GAS DESULFURIZATION PROCESSES

                                                        1975-78 COST BASIS
                                                                                                   a,b,c
00

Case
Coal-fired power unit
90% S02 removal; onsite solids disposal6
200 MW, new, 3.5% S
200 MW, existing, 3.5 %S
500 MW, existing, 3.5% S
500 MW, new, 2.0% S
500 MW, new, 3.5% S
500 MW, new, 5.0% S
1,000 MW, existing, 3.5% S
1,000 MW, new, 3.57, S
80% S02 removal; onsite solids disposal6
500 MW, new, 3.5% S
90% S02 removal; onsite solids disposal6
(existing unit without existing
particulate collection facilities)
500 MW, existing, 3.5% S
Oil-fired power unit
90% S02 removal; onsite solids disposal6
200 MW, new, 2.5% S
500 MW, new, 1.0% S
500 MW, new, 2.5% S
500 MW, new, 4.0% S
500 MW, existing, 2.5% S
1,000 MW, new, 2.5% S
Years
life


30
20
25
30
30
30
25
30

30



25


30
30
30
30
25
30
Limestone
M $


17,671
15,282
31,113
30,734
34,185
37,114
47,529
51,362

32,955



40,314


11,278
17,778
21,189
23,889
25,215
32,866
$/kW


88.4
76.4
62.2
61.5
68.4
74.2
47.5
51.4

65.9



80.6


56.4
35.6
42.4
47.8
50.4
32.9
Lime
M $


15,978
17,527
35,001
27,545
30,558
33,143
51,620
44,891

29,428



35,119


12,892
21,739
24,778
27,149
29,329
36,193
$/kW


79.9
87.6
70.0
55.1
61.1
66.3
51.6
44.9

58.9



70.2


64.5
43.5
49.6
54.3
58.7
36.2
Magnesia
M $


19,119
19,210
34,940
31,057
35,827
39,941
52,293
53,013

34,659



43,145


12,002
17,224
21,871
25,588
27,309
32,377
$/kW


95.6
96.1
69.9
62.1
71.7
79.9
5£.3
53.0

69.3



86.3


60.0
34.4
43.7
51.2
54.6
32.4
Sodium
M $


22,391
23,597
43,351
36,802
42,387
47,210
67,038
64,203

40,406



52,419


14,335
21,001
26,500
30,925
33,618
40,590
$/kW


112.0
118.0
86.7
73.6
84.8
94.4
67.0
64.2

80.8



104.8


71.7
42.0
53.0
61.9
67.2
40.6
Cat -Ox
M $


25,171
22,914
48,772
54,494
54,791
55,047
80,497
88,893

-



56,585


16,665
35,552
35,814
36,014
42,366
58,122
$/kW


125.9
114.6 .
97.5
109.0
109.6
110.1
80.5
88.9

-



113.2


83.3
71.1
71.6
72.0
84.7
58.1
       Midwest plant location represents project beginning mid-1975, ending mid-1978.  Average cost basis for scaling, mid-
       1977.  Minimum in process storage; only pumps are spared.  Investment requirements for disposal of flyash excluded.
       Construction labor shortages with accompanying overtime pay incentive not considered.
       These investment costs depend heavily on project definition.
     ^ Working capital has been included in the total capital investment.
       All Cat-Ox installations require particulate removal to 0.005 gr/acf prior to entering converter.  Because existing
       units are assumed to already meet EPA standards (0.1 Ib particulate/MM Btu of heat input), only incremental additional
       precipitator is required.
       Sludge disposal pond with clay liner.

-------
00
tn
                                                 Table  2.    SUMMARY  OF TOTAL AVERAGE ANNUAL REVENUE REQUIREMENTS


                                                        FIVE LEADING FLUE GAS DESULFURIZATION PROCESSES


                                                                         1978 COST BASIS



                                                                                 Total average annual  revenue requirements
a,b

Case
Coal-fired power unit
90% S02 removal; onsite solids disposal
200 MW, new, 3.5% S
200 MW, existing, 3.5% S
500 MW, existing, 3.5% S
500 MW, new, 2.0% S
500 MW, new, 3.5% S
500 MW, new, 5.0% S
1,000 MW, existing, 3. 5% S
1,000 MW, new, 3.5% S
80% S02 removal; onsite solids disposal
500 MW, new, 3.5% S
90% S02 removal; onsite solids disposal
(existing unit without existing
particulate collection facilities)
500 MW, existing, 3.5% S
Oil-fired power unit
90% S02 removal; onsite solids disposal
200 MW, new, 2.5% S
500 MW, new, 1.0% S
500 MW, new, 2.5% S
500 MW, new, 4.0% S
500 MW, existing, 2.5% S
1,000 MW, new, 2.5% S
Years
life


30
20
25
30
30
30
25
30

30



25


30
30
30
30
25
30
Limestone
M $


5,883
5,686
11,854
10,625
11,937
13,105
19,711
19,163

11,319



14,376


4,236
7,377
8,548
9,566
9,979
13,317
Mills/kWh


4.20
4.06
3.39
3.04
3.41
3.74
2.82
2.74

3.23



4.11


3.02
2.11
2.44
2.73
2.85
1.90
M $


6,362
7,150
14,528
11,145
12,758
14,377
23,819
20,570

12,206



14,826


5,123
8,920
10,585
11,886
12,140
17,165
Lime
Mills/kWh


4.54
5. 11
4.15
3.18
3.65
4.11
3.40
2.94

3.49



4.24


3.66
2.55
3.02
3.40
3.47
2.45
Magnesia
M $


7,036
7,257
14,052
11,572
14,082
16,448
23,169
22,789

13,406



16,639


4,615
6,848
9,019
10,967
10,683
14,734
Mllls/kWh


5.03
5.18
4.01
3.31
4.02
4.70
3.31
3.26

3.83



4.75


3.30
1.96
2.58
3.13
3.05
2.10
Sodium
M $


9,238
10,868
22,189
14,549
18,782
22,858
38,813
31, 186

17,425



24,995


6,463
8,938
1.3,075
17,042
15,322
22,223
Mills/kWh


6.60
7.76
6.34
4.16
5.37
6.53
5.54
4.46

4.98



7.14


4.62
2.55
3.74
4.87
4.38
3.17
Cat-Ox
M $


6,000
8,263
17,765
12,681
12,766
12,844
31,138
20,534

-



19,480


3,902
8,260
8,181
8,033
15,998
13,125
Mills/kWh


4.29
5.90
5.08
3.62
3.65
3.67
4.45
2.93

-



5.57


2.79
2.36
2.34
2.30
4.57
1.87
      a  Power  unit  on-stream time,  7,000  hr/yr.   Midwest  plant  location,  1978  revenue  requirements.   Investment  and  revenue  requirements  for  disposal
      ,  of  flyash excluded.
        These  revenue  requirements  reflect  capital  investments  shown  in  Table  1  (updated); byproduct  credit  and  sludge  fixation  costs  excluded.

-------
    Table  3.    DATA USED IN  REVISION OF TABLES 11  AND 22 OF THE TVA PAPER

             "COST COMPARISONS  OF  FLUE GAS  DESULFURIZATION SYSTEMS"
Data for Revision of Capital Investment

Investment basis:  500-MW, new coal-fired unit burning coal with 12% ash,
30-*yr life, 127,500 hr of operation, only pumps spared, no bypass ducts,
no overtime, experienced design and construction team, 3-yr project, 1975-78;
1977 costs for scale, includes both flyash and S02 removal, excludes flyash
disposal, reheat to 175°F.
Material index for direct costs
Labor index for direct costs
                          Mid-1974

                           153.8
                           177.9
                             Mid-1977

                              220.4
                              197.0
$/ton
$/ton
$/ton
$/ton
$/ton
$/liter
$/yr
$/lb
4.00
18.50-24.00
155.00
15.00
52.00
1.65
-
2.00
Data for Revision of Annual Revenue Requirements

  Raw materials                  Cost basis     1975
  Limestone
  Lime (varying with quantity)
  Magnesium oxide
  Coke
  Soda ash
  Catalyst (MgO, Cat-Ox)
  Catalyst (sodium)
  Antioxidant

  Conversion costs

  Labor:  Operating, supervision
          Analyses

  Utilities
    Process water (vs. quantity)
    Fuel oil, No. 6
    Fuel oil, No. 2
    Natural gas
    Heat credit, coal
    Heat credit, oil
                $/man-hr
                $/man-hr
                $/M gal
                $/gal
                $/Mcf
                $/MM Btu
                $/MM Btu
                    8.00
                   12.00
                 0.02-0.08
                    0.23
                    0.30
                    1.00
                    0.60
                    1.60
  Steam
    Coal
    Coal
    Oil
    Oil

  Electricity
    Coal
    Coal
    Oil
    Oil
1975
1978
1975
1978
1975
1978
1975
1978
$/M Ib
$/M Ib
$/M Ib
$/M Ib
$/kWh
$/kWh
$/kWh
$/kWh
0.80
1.50
1.50
2.75
0.011
0.028
0.019
0.041
0.70
1.40
1.40
2.65
0.010
0.027
0.018
0.040
                                             1978
                                              6.00
                                          33.00-42.00
                                            215.00
                                             28.00
                                             78.00
                                              2.20
                                         1.5 x 1975 cost
                                              2.75
                      10.00
                      15.00
                    0.03-0.11
                       0.35
                       0.42
                       2.00
                       1.15 ($22.80/ton coal)
                       2.40
                   Year    Cost basis    200 MW    500 MW    1,000 MW
0.60
1.30
1.30
2.55
0.009
0.026
0.017
0.039
                                       86

-------
 Table 4.   REVENUE REQUIREMENT COMPARISONS (1978) OF FGD SYSTEMS

      INCLUDING BYPRODUCT CREDITS AND SLUDGE FIXATION COSTS
Annual revenue requirements,
Process
Limestone
Lime
Magnesia
Cat-Ox
Sodium
Tons
byproduct
206,000d
175,000d
110,400e
137,400f
32,700^
As shown
in Table 2a
3.41
3.65
4.02
3.65
5.37
Includes
byproduct credit
-
-
3.23
3.04
4.99
mills /kWh
With sludge
fixation0
4.01
4.15
-
-
-
a New 500-MW, coal-fired unit, 3.5% S in coal, 7,000 hr/yr.
b 100% sulfuric acid at $25/ton, 80% acid at $16/ton, sulfur at
  $45/long ton ($40.18/ton).
c Assuming sludge fixation  service fee by contract treating sludge
  in the utility's pond at  $10/ton of 100% solids.
d 100% solids.
e 100% H2S04.
f 80% H2S04.
8 Elemental sulfur.
                               87

-------
          These updated economics unfortunately do not reflect any recent
changes in process technology which, of course, should be taken into account.
In addition to sludge fixation, numerous other developments in lime and lime-
stone technology have taken place recently including  (1) spray towers in
place of mobile-bed scrubbing devices,  (2) series hold tanks, magnesium
addition and chloride ion control, pH and stoichiometry adjustments for in-
creasing limestone utilization,  (3) mist eliminator wash cycles and hot air
injection or recycle reheat schemes to  increase operating reliability, and
(4) sludge oxidation for more effective disposal.  Furthermore, in spite of
the increased costs, some utilities have been  leaning toward scrubber re-
dundancy and equipment sparing as means for improving system reliability.

          Although the cost effects of  these changes have not been explored
in updating the 1974 EPA-TVA cost study results, some of them have been
examined as part of a separate study which will be presented as part of this
paper.

          For the  three regenerable processes, there have not been as many
noteworthy developments as with  lime-limestone; however, some new information
is available.  For instance, data from  the Boston Edison Company and Potomac
Electric Power Company  (PEPCo) magnesia scrubbing demonstrations indicate the
process chemistry  is workable  as predicted, but some  improvements are needed
in liquid-solids separation, drying, calcining and materials handling equip-
ment, and rubber-lined piping  is definitely required.   Since the piping in
the EPA-TVA estimates was rubber lined, the expected  cost effects of the above
improvements would probably be limited  to  less than 10% over the 1975-78 pro-
jected costs.

          Changes  in the Wellman-Lord - Allied process  primarily center- around
S02 regeneration where multiple-effect  evaporation can  be used to reduce
energy requirements at  the expense  of additional  capital.   Probably, a savings
of 5-6% in operating cost could  be  obtained with  2-4% increase in investment.
It is also anticipated  that  some measures  may  be  taken  to reduce or eliminate
the byproduct Na2S(>4 generated by  the process; however, such process alter-
ations are yet under study and remain  to  be firmly applied.

          Although the  Wood  River  Cat-Ox  demonstration  has  been plagued with
a variety of  problems,  many  unrelated  to  the process  itself, we are not aware
of any  improvements or  changes which would significantly alter the economics
as presented.

          Once  these and  other regeneration process  demonstrations are
carried  to  successful  completion,  there no doubt  will be greater acceptance
by  the  utilities.   The regeneration processes  badly  need a  success story,
without  any qualifications,  to improve their  image.
                                     88

-------
EXPANDED EPA-TVA PROCESS ECONOMICS PROGRAM

          In the months since the report "Detailed Cost Estimates for
Advanced Effluent Desulfurization Processes" was completed, TVA, under the
sponsorship of EPA,  has begun a larger, expanded program in evaluating FGD
process economics which we hope will produce additional valuable results.
The phases of study are as follows:

     Economics of lime-limestone scrubbing based on results of the
     Shawnee test program

     Byproduct marketing

     Advanced process evaluations

     Sludge disposal design and cost study

     Energy optimization studies


Economics of Lime-Limestone Scrubbing Based on Results of  the
Shawnee Test Program

          Joint Bechtel/TVA Computer Study.   Within  the scope of the Shawnee
lime-limestone wet scrubbing test program, EPA is  funding  a joint study  by
Bechtel Corporation and TVA to develop  a  computer  program  which will predict
investment and revenue requirements for various  designs and/or operating con-
ditions.  This project has been under way since  early 1975.  The program can
be used as a preliminary design and cost  guide in  choosing a lime-limestone
scrubbing system for a particular application  (e.g. limestone vs. lime scrub-
bing utilizing TCA scrubbers).  It can  also be used to determine the effect
of a change of independent variables on the cost of a particular system
 (e.g. 8 ft/sec vs. 12.5 ft/sec scrubber gas velocity).  The model should not
be counted on, however, to compute the  economics of a given system  to a  high
degree of accuracy.

          Bechtel's major responsibility  in the  study is to provide design
equations for relating equipment  size  or  capacity  to  the input process de-
sign variables.  TVA is responsible for (1) the  writing of the overall com-
puter program,  (2) the generation of tables and/or equations relating
equipment cost to size or capacity, and (3) the  estimation of  total invest-
ment and  revenue requirements, based upon the  equipment costs  and other
economic  inputs.

          The overall program  is  not yet  complete; however, Bechtel has
 completed the initial effort for  specifying material  flow  rates  and equipment
 sizes, and TVA has developed the  methods  for  costing  the  equipment.  The
 initial version  of the  overall program is expected to be  complete by mid-1976.
 Upon completion  of the  computer  program,  a user will  be capable of  comparing
 the effect of various design conditions on relative economics  of lime-limestone
 systems  for  coal-fired  power units ranging from 100 to 1300 MW.   As a further
 introduction to  the  program, the major input  requirements  for  the joint
 Bechtel/TVA  computer program are listed below.

                                     89

-------
 1.   Power unit size,  MW
 2.   Boiler heat rate, Btu/kWh

3a.   Coal analysis, wt % as fired

          Carbon
          Hydrogen
          Oxygen
          Nitrogen
          Sulfur
          Chloride
          Ash
          Moisture

3b.  Heating value  of coal, Btu/lb
 4.  Percent of  sulfur emitted  as S02  or  803 overhead

 5.  Percent of  ash emitted overhead

 6.  Excess air  to boiler, %
 7.  New versus  retrofit  installation
 8.  Limestone or lime chemical and physical properties

 9.  TCA scrubber operating parameters

           Number of scrubbers
           Number of beds
           Scrubber gas velocity
           Liquid to gas ratio
           S02  removal, %
           Stoichiometry

 10.  Redundancy desired
           Number of spare scrubbing trains

 11.  Mist eliminator  system options

           With or without wash tray

 12.   Effluent hold tank definition
           One, two, or three stirred tanks (series)
           Slurry residence time, min  (total)

 13.   Reheat (in-line  steam reheater)

           Outlet flue gas temperature, °F

 14.   Electrostatic precipitator or mechanical collector

           Ash removal efficiency upstream of S02 scrubbers

 15.   Sludge disposal method
           Onsite ponding:  Distance between disposal site and scrubber systa
                            Weight percent solids in settled sludge

           Fixation:   Method  or costs
                                  90

-------
  16.  Power  unit  load  factors
  17.  Remaining life of  plant
  18.  Economic factors

           Construction  cost  indices
           Raw material  and utility unit costs


          The major  outputs  of the joint Bechtel/TVA computer program are
listed below.

   1.  Conceptual  design  information
           Material balance
           Number of trains
           Equipment size,  general specifications,  and operating conditions
   2.  Capital investment
           Equipment costs
           Investment  breakdown by area (feed  preparation,  scrubbing,  disposal)
           Investment  breakdown by item (equipment,  piping,  ducts,  structure,
             electrical,  instrumentation, etc.)
   3.  Annual revenue requirement
           Direct costs:    Raw materials
                            Labor
                            Utilities
                            Maintenance
           Indirect costs:  Capital charges
                            Overheads
           Total  annual  revenue requirement
   4.  Lifetime revenue requirement
           Total
           Average  unit
           Levelized unit
          EPA-TVA Limestone  Process Economic Sensitivity Study.    In April
1975 TVA  completed a sensitivity study for  EPA showing the effect of several
selected  operating variables on the costs of limestone scrubbing.  As an
example of the  potential  use of the computer program upon completion, the
results of selected design options considered in the April 1975  sensitivity
study have been revised manually to reflect more accurately the  equipment
definition for  the computer  study.   The updated results are shown in Table 5
as ratios to  total investment and total revenue requirements of  the base case.
In the computer study, results will be calculated to give a total project
cost rather than as a ratio  to a base case.
                                  91

-------
                           Table 5.   LIMESTONE SLURRY PROCESS

                   PROJECTED INVESTMENT AND REVENUE REQUIREMENT RATIOS3
                                                               Total        Total annual
                                                             investment  revenue  requirement
Case
1
2
3
4
5a
5b
6a
6b
7
8
9
10
11
12
13a
13b
14a
Ub
15
16
Identification
Base case
12.5 ft/sec gas velocity
12.5 ft/sec gas velocity with wash trays
MgO addition
Hypalon-lined pond
Pond without clay lining
Sludge fixation service fee - $5/wet ton (50% solids)
Sludge fixation service fee - $10/wet ton (50% solids)
Slurry oxidation - low pH operation with series
hold tanks
Slurry oxidation - MgO addition
Benzole acid addition
Series hold tanks
Low pH operation
Low pH operation with series hold tanks
Hot air injection reheat - 100 F
Hot air injection reheat - 50 F
Flue gas recirculation reheat - 100 F
Flue gas recirculation reheat - 50 F
In-line reheat - 50 F
In-line reheat - 100°F - entraimnent level - 0.67%
ratio
1.000
0.915
0.931
0.911
1.125
0.981
0.839b
0.839b
1.033
1.030
0.990
0.997
0.990
1.006
1.080
1.022
1.145
1.037
0.984
1.001
ratio
1.000
0.935
0.975
0.974
1.070
0.989
1.099
1.287
1.140
1.177
1.008
0.990
0.994
1.001
1.209
0.989
1.100
0.958
0.920
1.032
3 500-MW new coal-fired power unit, 3.5% S in fuel, 7,000 hr/yr.
  85% SO  removal, mid-1977 investment, mid-1978 revenue requirements.
  Assumes service contractor provides investment for sludge transportation, fixation,  and
  disposal.  Includes only cake filtration and loading facilities - pond excluded.

-------
          Base Scrubbing System Incorporated in the Joint Bechtel/TVA
Computer Program and the Economic Sensitivity Study.   The base system re-
flected in the computer program and sensitivity study utilizes a forced-draft,
dry fan system upstream of the S(>2 scrubbers.  Costs include a low-efficiency
mechanical collector upstream of the fans to insure removal of the large ash
particles in the gas thereby preventing erosion of the fan blades.  A plenum
is used for separation of the power plant ducts from the scrubber ducts,
allowing the number of scrubber trains to be a variable, independent of the
number of power plant ducts.  Although separate fans are provided on each
side of the plenum to insure satisfactory gas distribution, only the scrubber
fan costs are included within the estimates.  The SC>2 scrubber is designed
with a presaturator compartment for partial humidification and cooling of the
gas upstream of the first bed.  The number of scrubber beds and depth of
spheres per bed are inputs.

          An in-line steam reheater is located downstream of the scrubber in
the vertical section of duct.  The reheater is designed with Inconel 625
tubes for reheating the gas to 150°F.  All reheater tubes in contact with
the gas above 150°F are Corten.  The gas is reheated to a final temperature
of 225°F.

          With the exception of the "fixed sludge" variations, each case is
designed and sized for onsite pond disposal of wet collected flyash in
addition to the SC>2 byproducts.  However, costs for transporting and dis-
posing the ash removed by the mechanical collectors are excluded.  The
distance between the scrubbers and the disposal pond for each case is 1 mi.
The concentration of solids in the settled sludge is assumed to be 40% by
weight for all cases except those designed for high oxidation of sulfites to
sulfates. The concentration of settled solids in the pond for the high oxida-
tion cases is assumed to be 60% by weight.  Disposal lines (spares included)
to the pond are rubber-lined carbon steel.  Supernate return lines are carbon
steel.  Ponds are sized for a total of 127,500 hr of operation over the life
of the plant, corresponding to an average capacity factor of 48.5% over a
30-yr plant life.
                                  93

-------
          A detailed listing of the cases and premises for these updated
costs and the initial computer program is shown below.

          Definition of Cases for Updated TVA Economic Sensitivity Analyses.

     Base Case

     Case 1:  Limestone additive
              Megawatts =  500  (4 modules)
              Sulfur in coal = 3.3 wt %; ash in coal = 12 wt %
              TCA with 3 stages  (4 grids),  5 in.  spheres/stage
              Scrubber pressure drop =4.2  in. H20  (excluding mist
                eliminator)
              Chevron mist eliminator pressure drop =0.2 in. H20
              L/G - 75 gal/Mcf
              Liquor rate  = 36 gpm/ft^
              Gas velocity = 8 ft/sec
              Limestone utilization = 65%
              S02 removal  = 85%
              Particulate  removal = 99.5%
              Solids in recirculated slurry = 15  wt %
              Scrubber inlet liquor pH =  5.8
              Effluent residence time  (single tank) = 12 min
              Pond  (clay-lined)  for sludge  disposal
              In-line  steam (500 psig pressure) reheater, 100°F reheat
              FD fan, with partial coarse ash cleanup, 33% efficient
                mechanical  collector
              Scrubber inlet particulate  loading  =  2-3 gr/scf

      Velocity Effect

      Case 2: Same  as  Case 1,  except:
              Gas velocity =  12.5  ft/sec
              Liquor  rate  = 33 gpm/ft^
              L/G = 44 gal/Mcf
              Scrubber pressure  drop  =7.2  in. 1^0  (excluding mist
                eliminator)
              Chevron mist eliminator pressure drop = 0.3 in. H20

      Case 3:  Same  as  Case 1,  except:
               Gas velocity =12.5 ft/sec
               Liquor rate - 33 gpm/ft2
               L/G = 44 gal/Mcf
              Wash tray  plus  chevron mist eliminator
               Scrubber pressure drop =  7.2  in. t^O  (excluding mist
                eliminator and wash tray)
               Chevron mist eliminator pressure drop =0.3  in.
               Wash tray pressure drop = 2.7 in.
                                    94

-------
MgO Addition Effect

Case  4:  Same as Case 1, except:
          Liquor magnesium ion concentration = 10,000 ppm
          Effluent residence time = 5 min
          L/G = 40 gal/Mcf
          Liquor rate = 19.2 gpm/ft2
          Scrubber pressure drop =3.1 in. 1^0 (excluding mist
           eliminator)

Alkali Utilization/Sludge Disposal Effects

Case 5a:  Same as Case 1, except Hypalon-lined pond

Case 5b:  Same as Case 1, except pond without clay lining

Case 6a:  Same as Case 1, except no pond; sludge fixed and disposed
           under service contract of $5/wet ton (50% solids)

Case 6b:  Same as Case 6a, except no pond; sludge fixed and disposed
           under service contract of $10/wet ton (50% solids)

Case  7:  Same as Case 1, except:
          12 in. spheres/stage
          Scrubber pressure drop = 7.2 in. H20 (excluding mist
           eliminator)
          Limestone utilization = 90%
          Scrubber inlet liquor pH = 5.2
          Three series hold tanks (4 min each)
          Complete oxidation of slurry bleed stream (pH -4.5) from the
           first hold tank, using air at 5 atm and stoichiometry of
           5 moles 02/mole S02 absorbed (similar to JECCO mode).

Case  8:  Same as Case 4, except complete oxidation of slurry bleed
           stream (from the single hold tank) using air at 3 atm and
           stoichiometry of 5 moles 02/mole S02 absorbed (similar to
           Kellogg mode).  Use 1 hr residence time in the oxidizer.

Case  9:  Same as Case 1, except:
          Organic additive
          Limestone utilization = 80%

Case 10:  Same as Case 1, except:
          Three series hold tanks (4 min each)
          Limestone utilization = 75%

Case 11:  Same as Case 1, except:
          Scrubber inlet liquor pH = 5.2
          Limestone utilization = 80%
          12 in. spheres/stage
          Scrubber pressure drop = 7.2 in. H20 (excluding mist
           eliminator)

                              95

-------
     Case  12:  Same as Case 7, except no oxidation of slurry bleed stream.

     Reheater Effect

     Case 13a:  Same as Case 1, except:
                External steam reheater with air - 100°F

     Case 13b:  Same as Case 13a, except - 50°F

     Case 14a:  Same as Case 1, except:
                External steam reheater with flue gas recycle - 100°F

     Case 14b:  Same as Case 14a, except - 50°F

     Case  15:  Same as Case 1, except - 50°F

     Case  16:  Same as Case 1, except entrainment level - 0.67% of
                 scrubber outlet flue gas rate


Byproduct Marketing

          The second most advanced project in the overall  EPA-TVA FGD  economics
program is an intensive study of marketing byproducts from FGD systems.  By-
products from both nonregenerable and regenerable processes are being evaluated
for potential market volume, location, and price.

          The initial efforts in this area were undertaken in early 1973 when
a hypothetical study was made of marketing sulfuric acid which TVA could con-
ceivably produce from its coal-fired power plants.  Using a computeriEed trans-
portation model a report entitled "Marketing I^SO,/, from S(>2 Abatement Sources—
The TVA Hypothesis" was published in December 1973 indicating that approximately
2 million tons per year of abatement acid could be produced.  Under the market
conditions prevailing at that time, a net sales revenue of $5-9/ton was indi-
cated.   With recent price increases, no doubt higher revenues could now be
expected.

          In June 1974 TVA prepared a second byproduct marketing study for EPA
entitled "Preliminary Feasibility Study of Calcium-Sulfur Sludge Utilization
in the Wallboard Industry."  Although limited in scope and very brief, this
study indicated the possible savings to be derived from marketing sludge as
gypsum.

          With these two projects as pilots, agreement was reached in 1974 to
expand the sulfuric acid analyses to cover both sulfuric acid and elemental
sulfur for all the utilities in the United States and to carry out a more com-
prehensive study of gypsum markets.  The computer program used for the TVA
hypothetical assessment was to be modified and expanded considerably for this
effort.
                                     96

-------
          During the early phases of the expanded acid study, it became
apparent that a sophisticated, useful computer program applicable to any by-
product was being developed an2 control.
Both double-alkali processes which produce sludge and regenerable processes
producing sulfur are already under study.

          In support of the EPA demonstration programs, preliminary, rough
assessments  (new 500-MW, coal-fired units, 3.5%  S in coal) were made of the
Atomics International's aqueous carbonate, UOP - Shell's copper oxide, Air
Products - Catalytic's ammonia/IFP, and Consol's potassium scrubbing pro-
cesses.  These processes all produce elemental sulfur.  Preliminary results
 (yet to be confirmed by detailed study) indicate a wide range of  investment
 ($87 to $160/kW for  1975-78 project) and  total annual revenue requirements
 (3.7 to 7.5 mills/kWh for 1978 startup).  If $45/long ton of sulfur is taken as
•byproduct credit, the net  revenue requirements range from 3.3 to  7.1 mills/
 kWh.

          In addition to the  regenerable  processes,  three double-alkali sys-
 tems— Envirotech, CEA - ADL,  and FMC—have been  studied.  The total  investment
 (including sludge pond) and revenue requirements of  the three systems  are
 relatively close, averaging about $71/kW  investment  and 3.6 mills/kWh  annual
 revenue requirement.  Additional refinement of vendor data will be  necessary
 before true  comparative costs  can be derived.


                                    97

-------
          Thus far, in all processes evaluated, capital charges (excluding
maintenance) constitutes the largest fraction of annual revenue requirements
ranging from about 40-43% for the double-alkali processes to about 45-60% for
the regenerable processes.  In the double-alkali processes, raw materials
(primarily lime) require the second greatest expenditure with energy third.
This is reversed with regenerable processes in that they require more energy
(fuels, electricity, and steam) than raw materials.  (This can be misleading
since components such as naphtha, coke, coal, and fuel oil could easily be
classified as raw materials or energy.)  Operating labor cost  for all the
processes is less than 3% of total revenue requirements.

          Plans now call for more intensive evaluation of the most promising
systems.  Detailed work is already under way on the Bureau of Mines' citrate
process for which design, investment, and revenue requirement data will be
developed along the lines of the earlier EPA-TVA studies.  After review by
the process developers for accuracy, reports will be published as with previous
evaluations.
Sludge Disposal Design and Cost Study

          Another area of work to be covered in the process economics program
is a detailed conceptual design and cost study of sludge disposal processes.
In the past the lack of detailed design and cost data for this area has re-
duced the value of our process economic assessment effort.  Included will be
cases covering several pond design and liner options, fixation processes such
as IUCS, Dravo, and Chemfix, and various sludge dewatering techniques.  In
addition, accurate design and cost data will be developed for combined and
separated flyash - scrubber sludge disposal systems.  This work is just getting
under way and the results will also be published.


Energy Optimization Studies

          The final phase of economic activity covers a careful investigation
of SC>2 removal process energy requirements and is directed toward reducing or
minimizing process energy needs.  An effort will be made to obtain data from
actual scrubber installations to compare with theoretical needs.  In addition,
key unit operations will be analyzed toward finding alternate routes to lower
energy consumption.  Work on this project is to be started soon.


ECONOMICS FOR THE TVA OFFICE OF POWER

          In addition to the work being performed for EPA, the OACD studies
staff has continued to support the TVA Office of Power with specific energy-
environmental-related process evaluations.  In the area of FGD, evaluations
are now under way on two promising Japanese double-alkali processes—
Kureha and Dowa.  The Kureha system uses a sodium acetate scrubbing solution
to react with limestone whereas the Dowa process uses a soluble aluminum
sulfate - oxide complex in water to absorb S02 for further reaction with lime-
stone.  Recent visitors to Japan have been impressed with these two processes
which produce gypsum for sale or disposal by piling.


                                   98

-------
          Results thus far indicate these two processes to be competitive
with double-alkali systems under development in the United States.  If
sludge fixation or other extra processing measures (reduction of soluble
salt losses, oxidation of sludge) are required, these processes will have
strong advantages.

          Also, during the past year, OACD has taken part in other Office
of Power energy process evaluations such as solvent refined coal, fixed-bed
coal gasification, NOX removal studies, and fluidized-bed combustion.  Some
of these results have been reported previously.  These activities have
enabled us to make comparisons of the various power plant environmental con-
trol options with perspective and consistency.

          As can be seen, TVA has been and continues to be involved in a
number of studies on FGD system economics.  In meeting these and future
responsibilities, we are striving to keep our quality high so that repre-
sentative process comparisons can continue to be made to guide the industry
in making key decisions.
                                99

-------
            STATUS OF THE EPRI FLUE GAS DESULFURIZATION
                        DEVELOPMENT PROGRAM
               Lawrence W. Nannen and Kurt E. Yeager

                 Electric Power Research Institute
                       Palo Alto, California
ABSTRACT

     This paper summarizes the present status of EPRI's program in flue
gas desulfurization for both lime/limestone and advanced regenerable
processes.  Emphasis is given to problem areas in lime-limestone scrubbing
since it offers the most widely accepted near-term solution to positive
S02 emission control.  Specific EPRI projects are discussed and general
guidelines for reliable scrubber implementation are offered.
                              101

-------
            STATUS OF THE EPRI FLUE GAS DESULFURIZATION
                        DEVELOPMENT PROGRAM
                                by
                        Lawrence W. Nannen
                                and
                          Kurt E. Yeager
                 Electric Power Research Institute
                       Palo Alto, California
The Electric Power Research Institute began operations in 1973 for the
purpose of expanding electric energy research and development under the
sponsorship of the nation's utility industry.  EPRI is presently funded
by over 500 member organizations, public, private, and cooperative.
EPRI's overall goal is to develop a broad, coordinated, advanced technology
program for improving electric power production, transmission, distribution,
and utilization in an environmentally acceptable manner.  The primary
areas of EPRI's research are fossil fuel and advanced systems; nuclear
power; power transmission and distribution; and energy systems, environment,
and conservation.

Projects of high priority at EPRI include those which develop technological
options to permit continued utilization of coal for power production in
both new and retrofit applications within environmental constraints.  A
major current issue associated with all fossil fuel combustion, particularly
coal, is the control of sulfur oxide emissions.

        Current EPRI Objectives in Flue Gas Desulfurization

The present research and development projects dealing with SO2 control
within EPRI's Environmental Control and Combustion Program are oriented
to providing electric utilities with reliable, cost-effective technology
for removing sulfur oxides from coal-fired steam boilers.  The goals of
these projects are to ultimately develop a reliable commercial design
basis for both throwaway scrubbing systems and regenerable FGD technology
so that utilities can apply such systems if they are required.
                              102

-------
                     LIME/LIMESTONE SCRUBBING

Since the technology for lime/limestone scrubbing is generally considered
to be the most advanced for present commercial applications to utility
boilers, and since the bulk of utility commitment is towards this tech-
nology (40,000 MW and $3 billion by 1980), the obvious near-term objective
of EPRI is to assure that the technical issues are resolved to allow
reliable, efficient operation of these systems.  A large amount of
operating experience now exists for lime/limestone scrubbing — some
good, some bad, and most poorly understood.  Therefore, a major task is
to collect the available data, sort out areas in which the technology
needs improvement, and develop a methodology for designing these processes
on a site-specific basis.

Table 1 lists the major problem areas continuing to face the utility
industry in integrating lime/limestone scrubbing with power production.

                              Table 1
          Major Problem Areas in Lime/Limestone Scrubbing


*    Mist Elimination

*    Reheat

*    Materials of Construction
*    Process Chemistry and System Design for Site Specific

     Applications

*    Waste Handling and Disposal

As is the case with any complex process, one must maintain the awareness
that each problem area is interrelated with another and must be addressed
with the total system in mind.

                          Mist Eliminator

The function of a mist eliminator in any wet scrubbing system is to
prevent the carryover of aqueous aerosol droplets downstream from the
absorber to prevent plugging with slurry solids and corrosion of reheaters,
scrubber I.D. fans, and flue gas duct work.  The design of the mist
eliminator system must consider the following variables:

Orientation.  The demister may be placed horizontally in a vertical gas
duct or vertically in a horizontal gas duct.  It may also be inclined  to
assist drainage during washing.  It appears that a vertical demister
configuration  (horizontal gas flow) offers advantages of superior drainage,
greater washing flexibility, and better adaptability to closed-loop
demister wash water recycle.
                                103

-------
Flow Distribution and Velocity.  Most mist eliminators used in commercial
FGD scrubbing systems are of the zig-zag "Chevron" baffle design.
Proper flow distribution and velocity of the wet gas through the demister
is essential for proper operation and efficiency.  The radial vane
collector shows promise in improving gas distribution.

Washing.  Reliable operation of mist eliminators in lime/limestone
scrubbing systems is complicated by the normally high solids content of
the scrubbing liquor.  These liquids can adhere to the demister surface,
forming a soft mud deposit or a hard chemical scale.  A large amount of
development work is presently underway to design a washing system to
minimize solids buildup on the demister.

Tie-in with the Overall System.  If the scrubber system is designed for
closed-loop operation (no water discharge), the quantity of fresh water
available to the system is limited to the water lost by saturation of
the flue gas and water remaining with the waste solids after final
disposal.  Most experience to date clearly indicates that the demister
wash water should be of the best quality, with the minimum amount of
dissolved calcium, to avoid scaling.  There is also a greater tendency
for the demister to plug when limestone is used rather than lime as the
absorbent.  This is caused by the normally poorer utilization of the
limestone which allows excess unreacted calcium to be entrained into the
demister.  The use of series hold tanks and low inlet slurry pH  (between
5.0 and 5.5} can improve limestone utilization; however, some reduction
in SO2 removal efficiency may occur at low scrubbing pH.

                             Reheater

The function of a reheater is to warm the flue gas above its wet bulb
temperature of 125°F after leaving the scrubber to a temperature normally
between 175°F and 200°F.  The warmer gas is then less likely to condense
and corrode fans, duct work, and gas dampers.  Reheat also increases
plume rise and dispersion of the effluent to minimize ground-level
concentrations of not only SO2 but also particulates and nitrogen oxides.

Degree of Reheat Needed.  There is some question as to the degree of
reheat needed for site-specific meteorological conditions.  There is no
legal requirement for reheat in present standards.

In-Line Steam Tube Reheat.  The most common design of reheat is a
series of steam tubes inserted in the flue gas stream.  Good reheater
operation then depends largely on proper mist eliminator performance.
However, corrosion is a major problem.

Indirect Reheat.  A more costly, less efficient but very reliable design
is to blend heated air with the flue gas to attain the desired temperature.
This approach results in reduced maintenance costs.
                                  104

-------
Hot Gas  Bypass.   Use of hot gas bypass is dependent on the efficiency of
the scrubber to  remove a very high percentage of the SO2 in the processed
flue gas.

Fuel Combustion.   The use of natural gas or fuel oil for direct combustion
irt the exit  gas  stream may be attractive in a few isolated cases, but is
a poor option for the future due to the risk of fuel shortages.

                     Materials of Construction

Lime/limestone process slurries contain abrasive solids (limestone, fly
ash,  and gypsum)  and corrosive dissolved species (chlorine, hydrogen,
and concentrated salts).   A major design problem is selection of reasonable
cost materials of construction to minimize corrosion and erosion of
scrubber components.

Liquor Loop.   Pumps,  valves, piping, and spray nozzles are all affected
by process slurry conditions.

Flue Gas Handling.   Fans, duct work, and the reheater must withstand the
corrosive flue gas  environment.

Absorber Internals.   Scrubber internals such as the mist eliminator,
stage partitions,  trays,  and packing must withstand both gas and liquor
contact.

                Process Chemistry and System Design

The previously discussed problem areas were primarily hardware issues.
It is important  to  recognize that the poorest understanding of lime/limestone
scrubbing is in  the process chemistry and its effect on overall system
design.  We  are  still unable to confidently predict the chemical behavior
of a scrubber for each potential utility site without a time-consuming,
expensive pilot  testing program at each site.  Table 2 lists the important
site-specific variables which affect scrubber system design.

The major scrubber  chemistry issues requiring further study are:

Degree of Sulfite to Sulfate Oxidation.  How is oxidation promoted or
retarded in  the  scrubber?  How is oxidation of the waste product achieved
to allow better  settling and dewatering?

Effect of Fly Ash Chemistry.  Fly ash provides surface area for possible
Chemical reactions.   Also,  alkaline fly ashes from Western coals can be
used beneficially in the scrubbing process.

Supersaturated vs.  unsaturated Gypsum Chemistry.  We do not fully understand
what conditions  cause operation unsaturated with respect to calcium
sulfate.
                                 105

-------
                              Table 2
               Process Chemistry and System Design
Site-Specific Variables

Degree of SO2 removal required
Inlet SO_ concentration
Excess air
Degree of oxidation
Fly ash loading
Fly ash chemistry and solubility
Degree of closed-loop operation required
Dissolved solids in  system
Tie-in with other plant waste water  streams
Constraints on sludge disposal

Chemistry Affects System  Design

Absorber type
 Reaction tank
 Liquid-to-gas ratio
 Reactant type
 Process control
 Materials selection
 Sludge composition and handling
                                106

-------
Effect, of Magnesium and Chlorine on scrubber chemistry.

Use of Lime vs. Limestone.  Limestone is the most economical, energy
efficient reagent but in many cases complicates the scrubber chemistry.
More work is needed to understand how to increase limestone utilization
while maintaining high SC>2 removals.

Use Cooling Tower Slowdown for Closed Loop Operation.  The scrubber has
the potential to act as a "garbage disposal" for the power plant.  Water
supplies can be conserved if scrubbers can feed off other plant waste
streams.

                    Waste Handling and Disposal

Handling and disposal of lime/limestone scrubber sludge is a complex
site-specific problem for which usable guidelines are not available.
Scrubber wastes consist of three general types of material:  fly ash,
calcium sulfate/sulfite salts, and scrubbing liquor associated with the
partially dewatered sludge.  Each type of material has different physical
and chemical characteristics.  For example, trace elements are found
almost exclusively in the fly ash, the calcium sulfite has very poor
physical properties resulting in inadequate dewatering and structural
stability features, and the liquor contains concentrated dissolved salts
produced from the scrubbing process.  Disposal and treatment methods
have started to evolve without a clear understanding of each sludge
component; the problems unique to one general component have been confused
to be characteristic of the total sludge.  Problems such as trace element
leachability, sludge fixation, sludge dewatering, and beneficial use of
scrubber wastes cannot properly be addressed until the characteristics
of each sludge component  are understood.

             EPRI Task Areas for Throwaway SO  Scrubbing


The  following  discussion  presents the methodology and  organization of
the  R&D tasks  in EPRI's throwaway S02 program.

1.   Evaluate  Utility Operating Experience  (1974-75)

EPRI is  funding the Battelle  Stack  Gas  Coordination  Center.   Battelle
has  completed  a state-of-the'-art  study  for EPRI  summarizing the  commercial
SO2  experience in  the U.S.  and recommending  further  areas  of research.

EPRI has  also  formed  a  utility working  group to  advise the EPA and  TVA
on the  conduct of  the  Shawnee  test  program.

 2.   Unit Analyses of  Scrubber and  Disposal  Hardware (1975-76)

 Battelle is continuing  their effort to evaluate  scrubber problem areas
with a series  of  "unit analyses"  on mist elimination,  reheat,  and cost
 variation in scrubber  bids.


                                107

-------
A new contract is under negotiation with Southern California Edison and
Steams-Roger to study improved process control capabilities for scrubbers
and the instrumentation available to measure control variables.

Proposed new projects to evaluate sludge handling techniques and to
recommend pump, fan, and valve materials and design are planned.

EPRI«is also supporting the Southern Services testing at Plant Sholz in
Florida which is operating three 20 MW pilot plants:  double alkali,
dilute acid Chiyoda, and Foster Wheeler carbon adsorption.

3.   Develop and Test Hardware and Chemistry (1976-77)

The Tennessee Valley Authority has been contracted by EPRI to study four
tasks:  a) horizontal 1 MW pilot scrubber for high sulfur coal, b) improved
methods of reheat, c) corrosion/erosion materials testing, and d) characteri-
zation of scrubber sludge as a function of scrubber operating conditions.

A contract is under negotiation'with Radian corporation to demonstrate a
novel concept in scrubber process control.

New projects include a study to characterize low sulfur/alkaline ash
scrubbing chemistry for Western coals and an investigation into the
oxidation chemistry of sulfite to sulfate.

4.   Diagnostic Capability  (1977-80)

It is planned that a large portion of EPRI's efforts in the future will
be centered around diagnostic testing of specific fuels in several
prototype scrubbers situated around the country.  This service will
benefit the power industry by providing a means to demonstrate overall
scrubber operation and behavior before committing to a specific design.
Diagnostic sites will be developed for both Eastern high sulfur coal and
Western low sulfur coal.

Figure 1 describes the current EPRI SOx Control Program and illustrates
the interrelation and timing of specific projects for both lime/limestone
scrubbing and regenerable FGD technology.
                              108

-------
OBJECTIVE
                 FIGURE 1

              SOX CONTROL PROGRAM


              PROJECT DESCRIPTION
Develop Reliable
Design Basis  for
Lime/Limestone
Scrubbing
 1.   Evaluate Utility Operating Experience

 2.   Unit Studies-Demister, Reheat, Cost Variation

 3.   1 MW Tests-Reheat, Corrosion, Horizontal  Scrubber

 4.   Evaluate FGD Process Control Capability

 5.   Design Guidelines for Lime/Limestons Scrubbing

 6.   Low Sulfur/Alkaline Ash Scrubbing  Characterization

 7.   Evaluation of Sludge Dewatering Processes

 8.   Sludge Composition & Leachability

 9.   Sludge Fixation Chemistry Guidelines

10.   Chemistry Modifications To Improve Reliability & Cost

11.   Hardware Modification To Improve Reliability & Cost

12.   Diagnostic Support - Eastern Coal

13.   Diagnostic Support - Wstern Coal
 Develop Viable
 Begenerable FGD
 Technology
14.  Comparative Evaluation of  Regenerable FGD Processes

15.  Comparative Design Of Advanced Processes

16.  Co-Sponsored Pilot  (10-40MW)  Process Construction
     & Operation

17.  20 MW Pilot Test Of Chiyoda,  Double Alkali,
     Bergbau Processes

18.  Modification To Existing Processes  To Improve
     Reliability & Cost

19 o  Test & Evaluation Of Commercial Scale (100MW+)
     Processes
                                        109

-------
   FIGURE 1  (continued)




SOX CONTROL PROGRAM




75
i A
, r,.
76

J 2 L

L_


! 14
t
1 n

700


4
n
T
1 6

1 7

77
MIL

_ /
L
A
f.
L
f^ /



_


1 8 h

h




1 15
y
— , -U
r?

L-*|

4400
r
78
ESTONES
^ - Def in
^v - Compl
\ - Updat
^y - Decis
79
it ion of K
ete State-
e State-Of
ion on Pro
10 LV — '

. , ^ _ 	 _ ^
I. 12
1 13



^. 16

18

4900

r^l
Jt ,.
II- . 1
($ x
3600
110




a_ 	 	
1000)
2500
80
ey Problem
Of-The-Art
-The-Art G
cesses To



81 | 82
Areas
Guideline
uide lines
Be Constru



i
i


2000
(5000)


2000
(5000)
1
s Document
Document
=ted
^J
	 1


2000
(5000)

-------
                  ADVANCED REGENERABLE PROCESSES

Recognizing the need to develop desulfurization processes which 1) do
not create a solid waste disposal problem, 2) are not dependent on
limestone raw materials, 3) have the possibility of being less sensitive
to site-specific design variables, and 4) can produce a useable byproduct.
EPRI is active in the evaluation of advanced regenerable processes.  The
objective of a current contract with Radian Corporation is to compare
and evaluate a large number of advanced processes on technical grounds
(material and energy balances) and to recommend further areas for process
development.

EPRI's evaluation program is considering three broad areas of regenerable
process development:

1)   Modifications to commercially available regenerable FGD processes
such as Wellman-Lord, magnesium oxide scrubbing, and CatOx to improve
performance, reliability and cost.

2)   Development of a regeneration capability for calcium sulfite/
sulfate throwaway FGD processes.  In certain circumstances, particularly
as a retrofit to lime/limestone scrubbing processes in the future, it
may be possible to convert the sludge into sulfur or acid or to convert
calcium based systems to magnesium oxide regenerable processes.

3)   Development of new regenerable processes.  A new EPRI project is
planned for 1976 which will provide a competitive site-specific design
evaluation for processes in all three of the above categories, with the
possible selection of one or two new processes  for pilot plant  (10-
40 MW) construction and demonstration with a host utility.
                              Ill

-------
                            CONCLUSIONS

For the near term, the primary utility choice for SO2 control appears to
be throwaway scrubbing processes involving the use of alkaline reagents
and the production of a calcium sulfite/sulfate waste product.  Many
utilities are reluctant to install these systems for three fundamental
reasons:

1)   Scrubbers are entirely non-profit for utilities and are expensive
to install and operate.  There is no guarantee that increased costs can
be passed on to the customer in a timely fashion.

2)   There is a questionable cost/benefit relationship and many utilities
doubt the need to control SO2 emissions to the degree many regulations
require.

3)   Early experiences with scrubbers were very poor and utilities have
been frightened by their foreign nature to conventional power plant
operation.  Only recently have utilities recognized the fact that scrubbers
are complex developmental chemical processes and operated them as such.

Considering these difficulties and the lack of positive incentive for
the utility industry to purchase and operate chemical processes which
are barely developed, scrubber technology has done well to evolve as far
as it has.

The following "challenges to industry" are given for the purpose of
maximizing the reliability of any scrubber installation in the utility
industry:

1)   Utilities must assume responsibility to make the scrubber system
work.  If the decision is made to install scrubbers, the utility must
accept the system as a developmental prototype and provide for an engineering
and maintenance intensive installation.  Provided with a proper design
basis, the utility must have its own qualified staff of chemists as well
as mechanical and chemical engineers to work with its architect/engineer
and selected vendor in building the best system to meet site-specific
conditions.  Process guarantees and fixed-cost contracts are not sufficient
to assure reliable, cost-effective scrubber operation.  Plant personnel
must give the scrubber operating and maintenance priority equal to all
other power station systems.

2)   Utilities should encourage vendors to achieve a mechanical "hybrid"
system based on the extensive body of available operational experience,
which incorporates the best components  (i.e., absorber, mist eliminator,
reheater, process equipment) of all presently available scrubber approaches.
The goal of this effort is to standardize the scrubber hardware for low
sulfur and high sulfur coals such that development priorities may be
placed on chemical design.
                               112

-------
Prom what we know today, some fundamental considerations  for maximizing
scrubber reliability include the following:

    a)   The SC>2 absorber should contain a minimum of  internals  to
    avoid plugging and scaling.  Where particulates are  to be removed
    in the scrubbing system, a two stage venturi/spray tower combination
    is recommended.

    b)   A vertical mist eliminator  (horizontal gas flow) offers advantages
    of superior drainage, greater washing flexibility, and better
    adaptability to closed-loop demister wash water recycle.  Horizontal
    mist eliminators appear to work satisfactorily for low sulfur cqal
    and for high sulfur scrubbing with lime.

    c)   The most reliable, low maintenance reheater design appears to
    be blending heated ambient air with the flue gas.  This approach is
    more capital intensive and less energy efficient,  but is very
    reliable.

    d)   Scrubber booster fans appear to operate more  reliably when
    they are dry.  Forced draft fans and induced draft fans installed
    downstream from the reheater have encountered minimum corrosion
    problems.

    e)   Spare modules and backup components improve the ability of
    scrubbers to be cleaned and serviced while on-line.

3)   Concentrate on scrubber process chemistry for site-specific design.
Capability to predict chemical behavior and performance reliability
before the scrubber is built is needed.
                            113

-------
                 NON-REGENERABLE PROCESSES SESSION
Chairman:      Michael  A.  Maxwell
              Chief, Emissions/Effluent  Technology Branch
              Industrial  Environmental Research Laboratory
              U.  S. Environmental  Protection Agency
              Research Triangle Park, North  Carolina
                               115

-------
       IERL-RTP  SCRUBBER  STUDIES RELATED TO FORCED OXIDATION
                        Robert H.  Borgwardt

           Industrial  Environmental Research Laboratory
               U.S.  Environmental  Protection Agency
          Research Triangle Park,  North Carolina   27711
ABSTRACT

     Small  scale experiments indicate that forced oxidation of FGD
scrubber sludge to gypsum is a feasible option for U.S. systems operating
with high-sulfur Eastern coals containing chloride.  Forced oxidation
was carried out as an integral part of the first stage of a two stage
system,  by  air sparging the effluent hold tank of the first stage at
pH 4.5.   Under these conditions complete limestone utilization was
achieved, together with total oxidation of the calcium sulfite at
atmospheric pressure.  The performance of the oxidizer was limited
by mass  transfer of 0_ from the air bubbles to the liquor, consequently
a tower  containing a 5.5-m depth of slurry was used to maximize oxygen
transfer efficiency.  Complete oxidation of the sludge was thus obtained
at an air stoichiometry of 2.6, without catalysts and without mechanical
atomizers.   The oxidized slurry settled at a rate ten times faster
than that of sulfite slurries at normal oxidation levels.  A settled
sludge of increased density (60% solids) also resulted, a factor
important to the environmental and economic aspects of sludge disposal.
The combined effects of high utilization and oxidation can yield a 35%
reduction in the amount of total wastes produced when the scrubber
is used  for fly ash collection.  An important objective of these tests,
to produce  a sludge that is filterable to 80% solids, was not attained.
This result, together with the observation that the settling properties
can in some cases be dominated by the fly ash, favor the dry collection
of fly ash  ahead of the scrubber as an alternative method of reducing
total waste.
                                117

-------
          IERL-RTP  SCRUBBER STUDIES RELATED TO FORCED OXIDATION

 INTRODUCTION
      A typical 600-MW power plant, burning Eastern  (3.5Z S) coal, will
 generate 15 short  tons of SO_ per hour.  When this  amount of SO- is
 reacted with limestone to form an environmentally acceptable waste
 product, it yields as much as 100 tons of wet sludge per hour, exclusive
 of fly ash.  The environmental and economic  aspects of sludge disposal
 in quantities of this magnitude were cited at the previous FGD symposium
 as the major problem in application of throwaway processes.  In view  of
 the current preference on the part of the electric utility industry for
non-regenerable FGD systems—in spite of the  sludge problem—one may assume
 that the overall economics justify the expense and effort required to  deal
 with it.  EPA's assessment of the situation  (Princiotta, 1975) pointed to
 the improvement of limestone utilization as one potential method of
 reducing the amount of sludge produced.   It is also known that the single
 factor most influencing the  amount of waste produced, per unit of S00
 scrubbed, is the dewatering  characteristics of the solid product.  This
 property has been tentatively related to  the degree of oxidation (Rossoff,
 1974), by comparison of sludges  produced  in various full scale scrubbers.
      Experiments at EPA's pilot  plant at Research Triangle Park have been
 focused on this  problem during the past year, both from the utilization
 and the oxidation standpoints.   Increasing the utilization is,  of course,
 an important objective  for its own merits due to the effect on operating
 cost and the effect on mist eliminator performance (reduced fouling) .  High
 utilization is also considered,  in our approach, to  be a necessary condition
 for efficiently carry ing out the  oxidation step; i.e.,  the  problems of
achieving maximum utilization  and  oxidation are related.   The  ultimate
objective was therefore to eliminate unreacted limestone as well as
unoxidized  calcium  sulfite from  the product solids.   The purpose of this
paper  is  to  evaluate  the  prospects for achieving this  objective  with
scrubbers operating at conditions  that correspond  to  the use of  high sulfur
coal.
                                   118

-------
BACKGROUND
     The feasibility of converting calcium sulfite scrubber slurry to
gypsum was  first indicated by I.C.I, in 1935; their original process
design included  an oxidation tower.  That design, in which the oxidation
step was carried out within the primary scrubber loop, was not efficient
with respect  to  the amount of air required.  Mitsubishi Heavy Industries
reported (Uno, 1971) several improvements in the state of the art at the
Second EPA  Limestone Scrubbing Symposium.  The MHI Lime/Gypsum Process
employs  two-stage scrubbing to adjust the pH to 4 - 4.5, followed by
pressurized (3.8 - 4,8 atm) aeration using mechanically driven atomizers
designed by the  Japan Engineering & Consulting Co. (JECCO).   The JECCO
oxidizer, operating with catalysts and low pH, has been successfully applied
in Japan to produce pure gypsum suitable for many industrial and construction
uses. The  suitability of this process to the operating conditions of U.S.
scrubbers,  which have both fly ash and chloride present in the slurry, has
not been demonstrated.  The experimental study at RTF was initiated primarily
because  of'questions regarding these factors.
     The MHI  report showed that the mass transfer efficiency of the oxidizer,
with respect  to  oxygen absorption, drops sharply at pH's greater than about
5.3 (with no  Cl~ present).  Fundamental kinetic studies conducted in the
Soviet Union  (Gladkii, 1974) showed a maximum rate for the oxidation of
calcium  sulfite  slurries at pH 4.5.  The HS03~ ion was identified as the
active species undergoing oxidation and the homogeneous reaction rate was
found to correlate directly with the concentration of that ion.  Concurrent
vork carried  out for EPA at Arthur D. Little, Inc. (Berkowitz, 1975) confirmed
the Soviet  results:  oxidation rate is proportional to the bisulfite concen-
tration  in  solution and, as a result, is an order of magnitude faster at PH 4
than at  pH  7.
pH AND UTILIZATION
     Achieving the low pH required for efficient oxidation of calcium sulfite
slurry necessarily means that the scrubber must operate at a high level of
limestone utilization.  At equilibrium the slurry will have a PH of 6.4 as
                                  119

-------
long as any unreacted limestone is present (Wen, 1975).  The fact  that  pH's
                                                                           *
this high are rarely observed in practice is a strong clue that  the system
does not operate at equilibrium, even at normal limestone feed stroichiometry.
As stoichiometry is reduced the concentration of unreacted limestone in the
slurry is lowered and kinetic factors become totally dominant (recent tests
at Shawnee with the scrubber operating at low stoichiometry showed that
90 hours of  stirring was required to raise the slurry pH from 5.7 to 6.1).
The  loss of  SC>2 removal efficiency under these conditions imposes a constraint
on both the  minimum pH and the maximum utilization that can be obtained with
scrubbers of current design.
     Techniques of optimizing the performance of  limestone  scrubbers at low
stoichiometry—a necessary condition for improving utilization—have been
examined at  RTF and are reported elsewhere (Borgwardt,  1975) in detail.  That
study  showed the kinetics of limestone dissolution/SO?  precipitation can be
accelerated  by modifying the scrubber effluent hold tank  to simulate tubular
flow.  Three or four stirred tanks,  of 2 - 4  min.  residence time each,  were
found  suitable.  The modified configuration  (Figure 1) will achieve greater
completion of the hold tank reactions in any  given  total residence time than
a single stirred tank of the same residence time.   It was shown  that a  single-
loop scrubber operating at 18 - 20 cm water pressure drop could  thus achieve
about  90% utilization at 82% S02 removal efficiency.  Ninety percent SO-
removal efficiency and 94% limestone utilization were obtained by increasing
the  tower pressure drop (holdup) to  30 cm water.  Under these conditions scrub-
ber  effluent  pH's of 4.5 and lower were achieved.  The relationship between
pH,  chloride,  and the concentration  of S02 in the scrubber liquor is shown in
Figure 2 as  observed in the RTF pilot plant.
     Attempts  to oxidize the purge slurry of a system operating  in the  scrubber
configuration  indicated by Figure  1  resulted in low conversion.   Although  the
initial pH was  less than 4.5,  it increased during aeration to 6.4.   It was
evident that the unreacted limestone in  the slurry was sufficient,  even at
utilizations of  90  -  94%,  to slowly  increase the pH so that efficient oxidation
rates (and CaSO-'^H-O dissolution  rate)  could not be maintained.
                                    120

-------
                       FLUE GAS
                     IN
                              OUT
                              A"
                            nrrmo
                            iiiiii
                                  LIMESTONE FEED
                                  SCRUBBER EFFLUENT HOLD TANK
                   SLUDGE
Figure 1.   Basic single-loop  scrubbing configuration with E.H.T.
            modified  to simulate  tubular flow.
                                121

-------
                   5.0      5.5      6.0      6.5      7.0     7.S
Figure  2.   Concentration of dissolved sulfite in RTF
            scrubbing liquor as  a  function of pH.
                          122

-------
     The rise in pH is accounted for by the reaction:
          2H+ + CaC03  -»•  Ca44" + C02 + H20                          (1)
which consumes part of the H  needed to keep the reaction cycle, that is
responsible for oxidation, going at constant pH:
          HS03~ + i$02  ->•  H+ + S04~~                                (2)
          H+ + CaS03(s)  +  Ca44" + HS03~                            (3)
          Ca44" + SO^" + 2H20  •+•  CaS04-2H20                        (4)
It was clear that additional H  must be provided during the course  of
reactions (2 - 3) if the pH was to be held constant.
STAGED SCRUBBING
     The scrubber was modified to a two-stage configuration as shown in
Figure 3 to permit a constant.pH of 4.5 to be maintained in the oxidizer.
A "staged" scrubber is defined here as one in which each gas-contacting
stage has^its own hold tank and slurry recirculation loop.  Ideally, there
would be no liquor carryover from the first to the second-stage loop by
gas entrainment, and this was the case in all experiments reported  here.
The primary loop, in which most SO^ absorption occurs, consisted of a TCA
tower operating at 18 - 20 cm water AP, and was set up with multiple hold
tanks.  The purge stream from this loop, which would normally go to the
clarifier or filter, is instead sent to the first stage loop where  the
slurry is contacted with additional S02—at maximum partial pressure—to
provide the H+ needed for reaction (1).  The low pH required for oxidation
and CaSO-(s) dissolution is thus maintained in the first stage with a
minimum sacrifice of S02 removal efficiency in the primary scrubber.  Since
little additional S02 absorption is required in the first stage it was set
up as a spray tower with low pressure drop (1 - 5 cm water) and low L/G.
In full scale application the first stage would be a venturi; several
systems, comprising a venturi in combination with a second, more efficient
S0_ scrubber, are already in operation in the U.S.
                                   123

-------

SO? 2.S3k«/lir 3830 ppm S02
HCI 73 j/hr
LIQUOR
TOTAL S AS S03 3.30 s/litet
S02 1.60
Ca 2.70
Cl 3.70
Ml 0.45
pH 4.4

SOLIDS

TOTAL SASS03 542m|/i
S02 330
CO 5
Ca 294



OXIDATION * 24 mol %
UTILIZATION » 92 mol %




SETT. RATE * 0.2 cm/min


8 liters/min
(L/6=8«al./mcf)


































M








i
502 = 2970 ppm 1

|
















»"
<»)

A


2.7
m/«~-












1
cmH20

S

1)7
26
I/I tor


pH3.8

\
-\
' \

1
£























340 mm
I

















S02 = 588 ppm
(81%)





















— •










•



.





»•
/

V
2.7
m/sec

nm
UUL.


NT
17
anH

no
1 1




in
300CD
iffli



"m
JLXJ




on
JU

pHS.O

\
\
1|21
i
S3 mers/min
(L/G = 64 gaL/mcf) ,
LIQUOR



TOTAL SASS03 1.90g/liter
S02
Ca
Cl
Mg
pH






0.52
1.40
2.30
0.33
5.2

SOLIDS 18%)
TOTAL S ASS03 519 mg/g
S02
C02
Cl

OXIDATION
343
35
3D5
= 17 mol'.
UTILIZATION = 85 mol %



LIMESTONE t FILTRATE


fate
0
\
\
n \2oiitets
^
-J




\
A 	
rieis
€
                    45min
                                      2 min
                                               2 min
                                                        2 min
Figure  3.   Staged scrubber without  forced oxidation.
                       124

-------
    When  forced  oxidation is not attempted S07 builds up in the first
stage scrubbing liquor to the concentrations shown in a typical run in
this mode  in Figure 3.  In contrast, Figure 4 shows the result of forcing
oxidation  in the  first stage by installing an air sparger in the hold
tank.  The accumulated S0_ is oxidized in the liquor, together with the
solid calcium  sulfite that is fed from the primary loop.  The latter is
redissolved, oxidized in the liquid phase, and recrystallized as gypsum.
Comparison of  the liquor compositions shown in Figures 3 and 4 indicates
that the steady-state S02 concentration in the first stage is no longer
determined solely by pH and chloride when oxidation is forced:  it is
determined instead by the difference between the rate of oxidation and
the rate of S0_ feed to the oxidizer.
    The pH was controlled at 4.5 in the oxidizer by adjusting the rate
of limestone feed to the primary scrubbing loop.  Since the utilization
was already 85% in that loop, the additional S02 absorption in the first
stage boosted  the overall limestone utilization to values approaching
100%.
    Figure 4  is  the basic scrubber configuration used at RTF for forced
oxidation.  A  major difference between this configuration and that used
by MHI is the  fact that oxidation was carried out as an integral part of
the first stage in the RTF tests.   This has the advantage of providing a
slurry of essentially pure gypsum to control scaling in the first stage,
which is most  susceptible to that problem.  We were thus able to operate
at L/G = 1.3 l./m3 (10 gal./mcf) in the first stage without scaling when
set up according  to Figure 4, but could not operate at this L/G when
oxidation was  not forced in the first stage.  In the latter case gypsum
scale formed in the upper parts of the tower where the feed slurry
"entered.  Another advantage of our approach is the ability to control pH
at the optimum value throughout the course of the oxidation reactions.
This minimizes the stripping of SO- from the liquor in the oxidizer and
increases the  oxidation efficiency to the extent that it can be conducted
at atmospheric pressure and without catalysts.

                                   125

-------
       SOZ 2.53 k|/h>
       HCI  41 }flir ~
                      2700 ppm S02
                                             SOz = 2470 ppm  (9%)
                          9 liten/min
       LiaUOB

TOTAL S AS S03 2.22 i/liter
     SO 2     0.45
     C02     0.20
     Ca      1.85
     Cl      3.3S
     M|      0.93
     pH       4.5
 SATURATION RATIO = 1.2
 TOTAL SASS03 545m|/|
       SOZ      S3
       COz    ' «
       Ci      275

  OXIDATION  -MnralX
  UTILIZATION = 99 moIX
            TO FILTER
  SETT. RATE = 1.6 em/mil.
  SETT. DENSITY-0.70|/ml
(L/G = 10 pUmcf)
                           AIR 13 A k|/hr
                                            /\
                     2.7
                    m/set
                                            I
                                          cmHzO
                     S02
                     1.1
                                           pH4.2
                                                      SOIitm/kr
                                                                                  59 liters/min
                                                         (L/G = 64 jal./mcf)
                                                               LIQUOR
    TOTAL  SASS03 1.54|/liter
                                                            S02
                                                            C02
                                                            Ca
                                                            Cl
                                                            Mi
                                                            pH
                 0.20
                 0.22
                 1.40
                 2.64
                 0.71
                 5.8
    SATURATION RATIO = 1.0

         SOLIDS (8%)

    TOTAL SASS03 S02m|/g
         S02      299
         COz      45
         Cl       297
     OXIDATION  « 26 mol X
     UTILIZATION • 85 mol X

     8.67 k|/lir
LIMESTONE + FILTRATE
                                                                              38.0% SOLIDS
                                                                   2.2 mill
                                                                                           22mm
      Figure  4.    Staged scrubber  with forced  oxidation  in  the first  stage
                      (0.91-meter slurry depth;  air  stoichiometry  =  6.3).
                                                 126

-------
OXIDIZER EFFICIENCY
     As indicated by the results shown in Figures 4 (no fly ash) and
5 (with fly ash), oxidation was successfully accomplished in the RTF
scrubber by air sparging the first stage hold tank.  The tank used for
these initial tests was 71 cm in diameter and contained a 0.91-meter
                                                     2
depth of slurry.  Air was admitted at 0.7 - 1.4 kg/cm  pressure through
a 50-cm diameter ring containing twenty-two 1.6 mm holes.  Experience
showed that the holes must point downward to prevent slurry from entering
the tube, where the solids settle and eventually fill it.  A 6.35-mm
orifice was used to measure air flow.  Water was added at 10 ml/min at
a point ahead of the orifice to inhibit scale growth in the small holes
of the sparge ring.  Satisfactory results were obtained with sparge rings
made of 316 stainless, PVC, and Teflon.  The Teflon ring appeared best
with respect to lack of corrosion and tendency to resist pluggage.  The
oxidizer tank was vented td the scrubber inlet so the air feed pressure
represents sparge ring AP almost entirely.
     Figures 4 and 5 show that the air feed rate required to complete
the oxidation of the slurry with the type of oxidizer described above
was 6-7 times the stoichiometric minimum, where air stoichiometry is
defined by:
              , .         g-atoms of oxygen fed to oxidizer
     air stozchzometry = g_moles of sOj absorbed in scrubber
                       = (kg air fed/hr) 0.21 (64.1) 2
                            29 (kg S02 fed/hr) n
where n is the combined SO™ absorption efficiency of both stages.
     Attempts to reduce the air stoichiometry below 6 resulted in a
buildup of HSO ~ in the first stage scrubbing liquor, indicating that
the rate of oxygen absorption had become the limiting factor—the rate
of oxidation in the liquid phase at pH 4.5 has been shown (Gladkii,
1974) to increase with increasing HS03~ concentration; therefore,
oxidation kinetics were not limiting.  The HS03~ concentration rises
in the first stage because, in this case, it is a dependent variable
determined by the oxidation rate—and not vice versa.  The steady-state

                                127

-------
         St>2 2.S7 ki/hr
         HCI  45 g/hr "
       LIQUOR
TOTAL S AS S03 2.30 j/liter .
                       2910ppmS02
                                 SO; = 2590 ppm
802 = 520 ppm
   (82%)
     S02
     C02
     d
     Cl
     Mg
     PH
0.87
0.03
2.52
5.44
0.77
 4.S
 SATURATION RATIO = 1.2


       SOLIDS

 TOTAL S AS S03 346 mg/g
      SO;      15
      C02       S
      Ca      179

   OXIDATION  = 95 mol %
   UTILIZATION * 97 mol %
             TO FILTER
   SETT. RATE = 2.0 cm/min
   SETT. DENSITY = 0.91 |/ml
       TOTAL SASS03 '-82g/litef
                                                                SATURATION RATIO = 1.0

                                                                     SOLIDS (15%)
                                                               TOTAL  S AS SO 3 306 mg/g
                                                                     S02      218
                                                                     C02      30
                                                                     Ca       186
                                                                             OXIDATION  =11 mol %
                                                                             UTILIZATION = 82 mol %
                                                                               11.4kg/hr
                                                                          LIMESTONE + FLY ASH
                                          1tmin
                                                                  2.2 min
                                                                              2.2 min
                                                                                          2.2 min
                Figure 5.    Forced oxidation with  fly ash  addition

                               (0.91-m  slurry depth;  air stoichiometry  =  7.0)
                                              128

-------
oxidation rate is limited, in turn, by the absorption of oxygen from
the air bubbles.  The number of bubbles per unit volume of slurry
decreases in the oxidizer with the air feed rate and thus the air/liquor
mass-transfer surface limits the amount of oxygen that can be absorbed
during the time of contact between the bubbles and slurry.
     The efficiency of oxygen mass transfer to a sulfite solution is
discussed by Urza and Jackson (1975) and their rationale was followed to
improve the oxidation efficiency of the RTF system.  The absorption of
02 from air bubbles of a given size was shown to be proportional to liquor
depth (bubble residence time) for oxidizer heights to 16 meters.  The
oxidation kinetics ir: solution influence the overall rate only via the
effect on the concentration of oxygen in the bulk liquid phase.  From
this viewpoint the high HSO,,  concentration at low pH enhances the
absorption of oxygen -by minimizing the concentration of dissolved 0- in
the liquor.
     Accordingly, the stirred-tank oxidizer was replaced with a tower
(PVC pipe) having a static slurry depth of 3.2, and later, 5.5 meters.
As shown in Figures 6, 7 and 8 these changes permitted us to reduce the
air stoichiometry ultimately to 2.6.   The air sparger used in the tower
had the same size and number of holes as that used in the stirred tank.
As a result of the increased bubble residence time afforded by greater
slurry depth in the oxidizer, oxygen transfer efficiency was improved
and more S07 was oxidized per kg of air injected.  It is reasonable to
expect that  stoichiometries lower than 2.6 can be achieved by further
increase of  the oxidizer height.   Also, as pointed out by Urza and Jackson,
one can expect a bonus in the form of a reduced energy requirement for
                                            Q
air compression,  per unit of O^ transferred.
     As indicated by Figure 6 the residence time of the slurry in the
oxidizer (with respect to the rate of recirculation to the first stage
tower) was reduced to 7  min.  without ill effect on either the oxidation
efficiency or scaling.  The CaSO^l^O supersaturation of the liquor
increased, however,  from l.lx at  16 min.  to 1.3 - 1.Ax at 7 min. residence
time.  The liquor saturation ratios shown in Figures 3-8, 11 were
determined at 25° C by the direct test method.
                                 129

-------
S02 1.93 kg/b
HCI 1009/hr
15 liter
LIQUOR
TOTAL SASS03 1.76g/liter
S02 0.16
C> 4.82
Cl 9.10
Mg 0.79
pH 4.5
SATURATION RATIO = 1.4
SOLIDS (13.8%)
TOTAL S AS S03 355 mg/g
S02 6
C02 15
Ca 1M
OXIDATION =98 mol X
UTILIZATION « 96 mol %
TO FILTER
SETT. RATE » 2.6 an/out
SETT. DENSITY = 0.89 I/ml
(50% SOLIDS)
(L/G = 20
Al
5.2k
kg/cm2
13cm r—^
"IF

2840ppmS02
s/min
g»Umcfl
g/hr
^


3.2m
102 liters
51 °C
0

ff V
S02 = 2220 ppm


FOUR
80%
OPEN
GRIDS <


ASH
1.5 kg/hi
i^^

(22%)

"A
2.0
m/s«
JP
1.5
mH20
S02
0.72
I/liter
pH3.9

1 1 II t


451





-»•
ittn/hr

1
S02

2.0
rn/stc
5.6
pH5.0

\ '


i
Sin*
e?
= 650ppm
(77%)
51 littrs/min
(L/G = 71
TOTAL S
SO
CC
Ci
Cl
FOUR Mi
65% pr!
G°R,DS SATURA
SO
TOTAL S
S<
C
C
OXIOA
UTIU2
6.27kg
LIMESTONE +
38.8HSOl.II
\
rs VI 02 liters

gal./mcf) ,
LIQUOR
AS SO 3 1.56g/liter
2 0.13
12 0.12
4.31
6.79
1 0.61
5.3
TION RATIO = 1.3
LIDS (8.2%)
AS S03 423 mg/g
J2 235
D2 119
i 302
TION = 30 mol %
ATION = 70 mol %
/hr
FILTRATE
IS

\
^204 liters
\- £
                Tana
                                     3 mill
Figure 6.  Oxidizer depth increased to 3.2 meters
           (air stoichiometry = 3.2).
                     130

-------
SOZ 3.44 kg/h
HCI 36 g/hr
25 liters
LIQUOR |
TOTAL S AS SO 3 1.49g/lit«
S02 0.07
Ca 2.08
Cl 3.70
. Mg 0.42
pH 4.3
SATURATION RATIO = 1.1
SOLIDS (15.5%)
TOTAL SASS03 306mg/8
S02 12
C02 I*
Ca 167
OXIDATION = 95 mol %
UTILIZATION =92 mol %
TO FILTER
SSTT. RATE - 3.0 cm/min
SETT. DENSITY = 0.88 n/ml
(60% SOLIDS)
, 2880 ppm SOz
min
(L/G = 20 gil./mcf)
M
k*
25cm
H20

Al
7.851
osV
cir»2
L
-


R
g/hr





5.5m
400 liters
O

FL
S02 = 2460 ppm







1 ASH
3.4 kg/hr
~T>
1 5.8 mifl


A
(15%)

3.7
m/sec

S02
0.88
g/litet
pH3.6










87 liters/hr
i

» S02
I

A
3.7
mno
IXJUU
17
jQCCXX
nmn
CJULXJQ
pH4.6

- 770 ppm
73%)
79 lilers/min
(L/G = 64 9aUmd> i
LIC
TOTAL S AS
S02
C02
Ca
Cl
M8
pH
SAT UR ATI
SOLI
TOTAL S*
SO 2
COj
Ca
OXIDA1
UTILIZ/
LIMESTONE +
\ ..


\ '
Yl 80 lite
UOR
SQ3 1.61 g/liter
0.23
D.19
1.58
1.88
0.30
5.1
)N RATIO = 1.1
OS (8.6%)
iSSOa 463mg/g
277
79
304
ION = 25 mol %
\ftOH = 76 mol %
hr
FILTRATE
37.4% SOLIDS
V --
fs \180liters
— &£*
2.2 min 2.2 min



t_w
ters
€
2.2 min
Figure 7.  Oxidizer depth increased to 5.5 meters
           (air stoichiometry = 2.9).
                     131

-------
                          SO? 3.56 kj/hr 3030j>pmS02
                          HCI lOOj/hr
S02 = 2550 ppm
SO; - 590 ppm
   (8«4)
       LIQUOR

TOTAL SASS03 2.01 g/litet
                          (L,'G = 70gal./nuf)
                              LIQUOR
                                                                                TOTAL  SASSOl 1.489/liter
 SATURATION RAT'O
                                                                                 SATURATION RATIO = 1.1
                                                                            SOLID SPHERES
                                                                            8.9 cm BED DEPTH
TOTAL SASS03 S36mj/g
     SO;       5
     CO;      10
     Ca      277
                        TOTAL SASS03 494 mg/g
                              SO?     333
                              C02      64
                              C>      302
                          OXIDATION  = 16 mol V
                          UTILIZATION = B2molS
                                                                                   !28k5/hf
                                                                              LIMESTONE* FILTRATE
  OXIDATION  = 99 mol S
  UTILIZATION . 97 mol %
 SETT. RATE « 1.1 cm/mm
 SETT DENSITY = 0.76 i/ml
           (57% SOLIDS)
                                       1S.I
                                                                      3 min
                                                                                  2 mil*
                                                                                              4 min
                 Figure 8.    No  fly  ash addition  (5.5-m  oxidizer  depth;

                                 air stoichiometry  =  2.6).
                                                132

-------
OXIDATION AND SETTLING RATE
     An important objective of forced oxidation is to improve the
settling rate of the slurry and reduce  the size of the clarifier
which, at Shawnee for example, is the largest single vessel in the
scrubber system.  An improvement in the settling rate as a result
of oxidation was first reported by TVA  (Kelso, 1971), an effect
attributed to enlargement of the crystal size.  Data accumulated
at RTF during the course of this study  (Figure 9) confirm TVA's
conclusion.  Using Shawnee (Fredonia) limestone, the fully oxidized
slurry settled 10 times faster than sulfite slurries in the normal
range of oxidation.  Our results also confirm that very high levels
of oxidation must be attained before the improvement becomes apparent.
SETTLED DENSITY AND SLUDGE PRODUCTION
     The Aerospace sludge^disposal study (Rossoff, 1974) has pointed
out the importance of sludge density to its environmental acceptability.
This property is the major determinant of compaction strength,  compres-
sibility, and permeability.  One of the reasons for the interest in
forced oxidation is the possibility of producing a sludge that  can be
dewatered by filtration to a moisture content near the optimum value
fqr compaction.   Thus,  the use of dry additives, that are required to
attain the optimum water content with sulfite sludge, might be avoided.
The load-bearing strength and drainability are both superior for a
compacted sulfate sludge, and a waste product in this form is more suit-
                                                         2
able than sulfite sludge for direct disposal as landfill.
     Figure 10 shows the density of settled sludges obtained at RTF,
which clearly increased with oxidation to a value of 0.9 g dry solids
per ml of sludge volume (equivalent to 60% solids) when fully oxidized.
In actual practice, settled densities corresponding to 35 - 45% solids
are obtained with most  sulfite sludges.    The values of Figure 10 are
somewhat lower because  the test was made under conditions that avoid
compression of the settled  solids,  and they were not mechanically
disturbed by a clarifier rake.   It  is not unlikely therefore that a

                               133

-------
     2.0
      1.5
  oc
  o
      1.0
              • WITH SHAWNEE FLYASH

              O WITHOUT FLYASH
  CO
      0.5
                 20
40        60

PERCENT OXIDATION
80
100
Figure  9.   Settling  rate of RTF  limestone scrubber  slurries

            as a function of oxidation (50° C).
                            134

-------
   1.0
                                                                            60
•a
s


1
M

i
V)

o
Ul
_l


IU
M
0.6
0.4
                                                                            GO

                                                                        50  §
                                                                            o
                                                                            CO
                                                                            LU
                                                                            O
                                                                            oc
                                                                            40
                                                                        30
    0.2
                                                       • WITHSHAWNEE FLYASH


                                                       O WITHOUT FLYASH
                  20
                           40           60
                                    OXIDATION, percent
                                                        80
100
         Figure 10.   Settled  (quiescent) density of RTF  limestone

                       scrubber  sludge,  as a function of oxidation.
                                    135

-------
large clarifier would yield a  fully oxidized  sludge of 65% solids as
expected.2  Vacuum filtration  of  the  oxidized slurry yielded a filter
cake that averaged about 67% solids in our tests.  These samples were
reslurried in acetone and dried at room temperature to ensure that no
water of hydration was  lost.   Thus, the objective of achieving 80%
solids by filtration does not  appear  attainable on the basis of RTF
                                                    2
results.  This  value is ideal  for optimum compaction  and, therefore,
is important to the  question of direct disposal as a landfill.
      It was  also apparent from our tests that settled sludges containing
 60% solids  could only be obtained with the longer oxidizer residence
 time (16  min.,  based on the slurry recirculation rate to the first stage),
 which corresponds to about  5 hours residence  time for the slurry in the
 first stage.  At shorter residence time (7 min.) we obtained only 50%
 solids.   This suggests  that the additional crystal growth that can occur
 at long residence time  is important to achieving maximum settled density.
      The combined effects of high limestone utilization and improved
 dewatering properties of the sludge can be expected to have a significant
 impact upon the rate of waste  production by a power plant.  The effects
 are shown in Table 1 for various  scrubber operating conditions, using
 the present situation defined  in  the  SOTSEP report  as a basis for
 comparison.  It is clear from  Table 1 that:   a) about 24% reduction of
 the amount of sludge can be obtained  using standard gravity settling;
 b) a greater reduction  of  sludge  is achieved  by the use of more efficient
 dewatering equipment,  such  as  vacuum  filtration, than is possible by
 forced oxidation alone—thus,  forced  oxidation is justified only if  gravity
 settling is used for dewatering.
 FLY ASH
      In general, settling rate is determined  by the size of the  smallest
 particles in suspension.  It is clear from Figure  9 that  the  settling
 rate of the mixture of fly ash and CaS04/CaS03 was dominated  by  calcium
 sulfite.  The fly ash fed to the scrubber during  these  tests,  fly  ash
 "A", had a settling rate of 5.5 cm/min at 50° C  in the  absence of
 CaSO,/CaS03.  Later tests with a different fly  ash (fly ash "B",  also

-------
                                              Table 1,   ANNUAL WASTE SLUDGE PRODUCTION  BY A 1000  MW COAL-FIRED
                                                           POWER PLANT EQUIPPED WITH LIMESTONE  FGD SCRUBBERS:
                                                                    FLY ASH COLLECTED  IN SCRUBBER

                                                                               (SHORT TONS)
Dewatering Procedure
Scrubber
Operating Conditions
Coal Ash, Dry
CaSO "'-sH.O
CaC03
Totnl Dry Solids
Solids Moisture, %
Total Wet Sludge
Reduction of Sludge
Production
Settling
Utilization - 60%
Oxidation - 10%
338,000
322,000
48,000
185,000
893,000
50
1,790,000
Base Case
Settling
Utilization - 90%
Oxidation « 10%
338,000
322,000
48,000
30, BOO
739,000
50
1,478,000
17%
Oxidation/Settling
Utilization - 100%
Oxidation = 100%
338,000
0
477,000
0
815,000
40
1,482,000
24%
Filtration
Utilization - 902
Oxidation » 10%
338,000
322,000
48,000
30,800
739,000
38
1,192,000
33%
Oxidation/Filtration
Utilization - 100Z
Oxidation - 100Z
338,000
0
477,000
0
815,000
30b
1,164,000
•35Z
0-4
~-i
          aSOTSEP  Report1,  p.  60
          b,,.
           Minimum moisture content  obtainable by  filtration in  RTP  tests

-------
from the Shawnee Test Facility) showed that the settling characteristics
of the scrubber slurry can be dominated by the fly ash instead of calcium
sulfite.  Fly ash "B" settled at 1.5 cm/min in water at 50° C and in the
scrubber yielded a slurry that settled at a rate that was independent of
oxidation.  Not only was the settling rate of this slurry determined by
the fly ash, but an adverse interaction was also apparent; e.g., a sample
of fully oxidized slurry taken from the scrubber when operating without
fly ash (and settling at a rate of 3 cm/min) settled at only 0.7 cm/min
when it was reslurried with fly ash "B" and compared at the same temper-
ature of 50° C.  The settled density was also reduced by the presence of
fly ash "B", the fully oxidized sludge containing only about 50% solids.
In view of this experience it appears that the current industry trend
toward dry collection of the fly ash prior to the scrubber is well advised.
Although the pervasiveness of "B"-type fly ashes is not known, it is clear
that minimal benefit can be-expected from forced oxidation when it is
present in the  scrubber slurry.
     Dry fly ash collection  can be considered as an alternative to forced
oxidation as a  means of reducing the waste production and, if it is mixed
with the scrubber  sludge,  as a means of  increasing the density.  In view
of the  results  at  RTF,  it  may  be the only way to achieve  the desired 80%
solids  for optimum compaction.  Table  2  summarizes the relative amounts
of waste produced  under varying conditions when the  fly ash  is  collected
ahead of  the scrubber.   Comparison of  Tables  1 and 2  shows that  the
reduction of total waste  is  greater for  dry  fly ash  collection  than  is
possible when  the  ash is  collected in the scrubber,  and  this is  true
whether or  not  forced oxidation is employed.   As  indicated by Table  2,
however,  blending the dry fly ash with the scrubber  sludge can also  yield
 the desired 80% solids if the sludge is oxidized.
 SO- ABSORPTION
      As indicated in Figure 6 the scrubber was successfully operated
. with both stages set up as spray towers at a total pressure drop of
 only 8 cm water and 77 - 80% S02 removal.  When an L/G of 2.7 l./m
                                138

-------
                                       Table 2.  ANNUAL WASTE  SLUUCE  PRODUCTION BY A 1000 MW COAL-FIRED
                                                   POWER PLANT EQUIPPED  WITH LIMESTONE FGD SCRUBBERS:
                                                                DRY FLY  ASH COLLECTION

                                                                     (SHORT TONS)
Dcwatcring Procedure
Scrubber
Onerntin;; Conditions
Coal Ash, Dry
CaSO •I:1I20
CaSO, -2H.O
!t 2
CaCO
Solids Moisture, %
Total Wet Sludjje
Total Waste
Reduction of Waste
Compared to Base Case
Settling
Utilization - 60%
Oxidation = 10%
338,000
322,000
48,000
185,000
50
1,110,000
1,448,000

19%
Settling
Utilization = 90%
Oxidation = 10% .
338,000
322,000
48,000
30,800
50
817,000
1,155,000

35%
Oxidation/Settling
Utilization » 100%
Oxidation • 100%
338,000
0
47 71, 000
0
40
795,000
1,133,000

37%
Oxidation/Filtration
Utilization - 1002
Oxidation « 100%
338,000
0
477,000
0
30
682,000
1,020,000

43%
Filtration
Utilization - 90Z
Oxidation * 10%
338,000
322,000
48,000
30,800
38
647,000
985,000

45Z
Final Sludge

Densi ly, 7, Solids
63
                                                              66
                                                                                     73
                                                                                                               80
                                                                                                                                      75
 Assuming dry fly ash  is blended with
 wet slud)./.>.

-------
 (20 gal./mcf) was maintained in  the  first  stage,  the  S02  removal  in that
 stage averaged  20Z at a gas velocity of  2.0 m/sec.  Thus, a utilization
 of only 70-80%  is required in  the  second stage  in order to control pH
 at the desired  value, and this range of  utilization is within  the
 capability of a single stirred tank.   Figure  11 summarizes a test in
 which one 9-min. hold tank was used  in the second stage.  Although a
 reduction of the overall S02 removal efficiency (to 71%)  occurred, the
 test showed that complete oxidation  could  still be obtained.
     Solid spheres were used in  most of  the tests at  3.7 m/sec gas
 velocity to permit the L/G to  be increased above  the  6.7 l./m  (50 gal./
 mcf) that is normally the upper  limit  for  hollow  (5-g) spheres.   The
 solid  (linear polyethylene) spheres were 2.5 cm diameter and formed three
 beds, each bed  7.6 cm deep.  This  tower could be operated at L/G's as
 high as 12 l./m without flooding  at 3.7 m/sec.  The  improvement  results
 from the reduction of sphere'buoyancy, which causes the light spheres
 to congregate against the upper  retaining  grid.  An alternate solution
 is to increase  the grid spacing, which in our case was 91 cm.  As shown
                                                                       3
 in Figures 7 and 8 the solid spheres gave 73% removal at L/G = 8.5 l./m ,
                                                o
 17 cm water AP  and 81% removal at  L/G = 9.4 l./m  , 25 cm AP, when
 operating at 3.7 m/sec.
 CONCLUSIONS AND RECOMMENDATIONS
     The high oxidation efficiency obtained by air sparging at atmospheric
 pressure indicates that considerable improvement in performance of forced
 oxidation systems can be realized  if the oxidation step is carried out as
 an integral part of the first  stage.  The improvement will permit the use
 of simpler, less costly oxidizer designs, will reduce the power requirements
 and will eliminate the need for  catalysts.   The air stoichiometry is
 determined primarily by the height of the oxidizer and future tests should
 be made with slurry depths to  at least 18 meters.   Stoichiometries below
 2.6 can thus be expected, and a minimum power for air compression.
     It is clear that many variations of the two-stage configuration are
workable,  including two spray  towers, multiple or single hold tanks, and
                               140

-------
SOz I.SSkjrt
HCI 36 9/hr
15 liters/
LIQUOR i
TOTAL SASS03 1.59g/liter
SOz °-34
Ca 2.69
Cl 4.97
M) 0.67
pH 4.6
SATURATION RATIO = 1.3
SOLIDS (15.1%)
TOTAL SASS03 347 mg/g
SOz 16
C02 14
Ca 184
OXIDATION =94 mol %
UTILIZATION = 94 mol %
TO FILTER
SETT. RATE = 2.0 an/mm
(L/C
fl
kg
10cm
H20
, = 20 ga
A
4.6 I
.7?V-
/cm*

If
, 2850 ppm S02
min
l./md)
R
5/hr

^


3.2m
102 liters
52°C
-o
FL\
SO






/ASH
1.63 kg/hr
"~Q

/
2 =
2350 p
(17.5%)
\
2.0
m/sec
1.5
cm H20
S02
0.83
g/liter
pH4.0

I




1






502

/^
2.0
6.6
cm H20

PH5.3
45liteis/ht
1




1
= 830 ppm
71%)
51 liters/min
(L/G = 70 jaL/mcf) j
LIQUOR
TOTAL SASS03 1.36g/litn
S02 0,19
COz 0.18
Ca 1.97
Cl 3.49
Mg 0.47
pH 5.6
SATURATION RATIO = 1.3

TOTAL SASS03 423 mg/g
S02 276
C02 118
Ca 313
OXIDATION =18 mol %
UTILIZATION = 70 mol %
t, OH Vn/hr
LIMESTONE + FILTRATE
l i
40.1% SOLIDS
460 liters
O



                   7 mm
                                            9 min
Figure 11.  Single  stirred  tank in primary scrubber loop.
                       141

-------
 gas velocities  ranging from 2.0 - 3.7 in/sec  (6.7  -  12 fps).  Our
 results favor the  following conditions:   3.7 m/sec  gas velocity,
 minimum AP in the  first stage at L/G = 1.3 - 2.7  l./m , maximum
 pressure drop (hold up) in  the  second stage, and  multiple hold tanks
 of 6 - 9 min total residence time in the  second stage.  The oxidizer
 should have a residence time of 15 min to provide a settled sludge of
 maximum density.
     The benefits  to be derived from oxidation, in  terms of improvement
 of the physical properties  of the  waste,  will depend upon the properties
 of the fly ash as well as the properties  of the gypsum produced.   In any
 case our results do not indicate  that a sludge filterable to more than
 70% solids can be expected.  Dry  fly ash  collection would, under these
 circumstances, be the preferred method of minimizing waste production
 as it would afford the only opportunity to produce a directly disposable
 landfill.
     The chemical properties of gypsum, vis-a-vis calcium sulfite as
       9
 the most desired waste product, is the principal issue upon which the
 final decision must be made as  to whether forced oxidation is to  be
undertaken on a large scale in  the U.S.
                                142

-------
REFERENCES


1.  Princiotta, F.T., "Sulfur Oxide Throwaway Sludge Evaluation Panel
    (SOTSEP):  Final Report, Volume II" EPA-650/2-?5-OlO-b (NTIS No.
    PB 2U2-619/AS), April 1975.


2.  Rossoff, J., et al., "Disposal of By-Products from Non-Regenerable
    Flue Gas Desulfurization Systems," Proceedings:  Symposium on Flue
    Gas Desulfurization-Atlanta Nov. 197k, Volume I" EPA-650/2-7h-126-a
    (NTIS No. PB 2^2-572/AS), pp. 399-kkl, December 1971;.


3.  Uno, T., et al., "The Pilot Scale R&D and Prototype Plant of MHI
    Lime-Gypsum Process," Proceedings of Second International Lime/
    Limestone Wet-Scrubbing Symposium, pp. 83U-8U9, November 1971.


lu  Gladkii, A.V., et al., "State Scientific Research Institute of
    Industrial and Sanitary Gas Cleaning (Moscow),11 Report for Protocol
    Point A-1, Development of Lime/Limestone Scrubbing for Stack Gas
    Desulfurization, US/USSR, Sulfur Oxides Technology Sub-Group, 197k.


5.  Berkpwitz, J.B., et al., "Scale Control in Limestone Wet Scrubbing
    Systems," EPA-650/2-75-031  (NTIS No. FB 2U3-309/AS), pp. 52-56,
    April 1975.


6.  Wen, C.Y., et al.,  Ibid., p. 66, April 1975.
7.  Borgwardt,  R.H.,  "Increasing Limestone Utilization in FGD Scrubbers,11
    paper presented at 68th Annual Meeting, AIChE. Los Angeles Nov. 1975.


8.  Urza, I.J., and Jackson, M.L., "Pressure Aeration in a 55-ft Bubble
    Column," Ind.  Eng. Chem., Process Des. Dev., 1_5 pp. 106-113,
    April 19751
9.  Kelso,  T.M.,  et all,  "Limestone Wet-Scrubbing Pilot Plant at Colbert
    Steam Plant," Process Engineering Branch Report,  Tennessee Valley
    Authority,  April 1971.
                                143

-------
   RESULTS OF MIST ELIMINATION AND ALKALI UTILIZATION TESTING AT
              THE EPA ALKALI SCRUBBING TEST FACILITY
       M.  Epstein,  H.  N.  Head,  S.  C.  Wang,  and D. A. Burbank

                        Bechtel Corporation
                          50 Beale Street
                 San Francisco, California   94105
ABSTRACT

     Testing with a single stage, 3-pass, open-vane, 316 stainless
steel chevron mist eliminator in both the venturi/spray tower and
TCA systems with lime and limestone has shown that the accumulation
of soft solids on the mist eliminator is a strong function of alkali
utilization.  For high alkali utilization (greater than about 85 percent),
the mist eliminator was kept free of solids deposits by use of inter-
mittent fresh water wash on both the top and bottomside.  For alkali
utilization less than about 85 percent, intermittent top and bottomside
wash with fresh water did not limit solids accumulation.  However, for
these conditions, a continuous bottom wash with diluted clarified
liquor used in combination with an intermittent topside fresh water
wash was shown to limit soft solids buildup to less than 10 percent
restriction within the mist eliminator.

     In the venturi/spray tower system with lime additive, reliable
long-term operation of the chevron mist eliminator with intermittent
top and bottomside fresh water wash was demonstrated during constant
load operation at the maximum attainable spray tower gas velocity of
9.4 ft/sec and during variable load (cycling gas rate) operation.  The
lime utilization for these tests was about 90 percent.  In the TCA
system with limestone additive, reliable long-t^rm operation of the
chevron mist eliminator with intermittent top and bottomside fresh
water wash was demonstrated at 12.5 ft/sec gas velocity at high alkali
utilization (greater than about 90 percent).  For both the lime and
limestone testing, only about half the makeup water available during
closed liquor loop operation was needed to wash the mist eliminators.

     Utilization with lime in the venturi/spray tower system was
normally greater than 85 percent for scrubber inlet liquor pH less
than 9.0.  Utilization with limestone  in both scrubber  inlet liquor
pH of 6.0 to about 95 percent at a scrubber inlet pH of 5.2.  Operation
at reduced  scrubber liquor pH, however,  inherently  causes a reduction
in SO  removal efficiency.  For the venturi/spray tower system with
a single effluent hold tank, limestone utilization  was  not affected

                               145

-------
by a change  in residence time from 20 to 12 minutes.  Limestone
utilization  declined, however, at 6 minutes residence time for scrubber
inlet liquor pH  greater than about 5.6.  For the TCA system at 12
minutes total residence time and at scrubber inlet liquor pH greater
than about 5.0,  higher limestone utilization was achieved with three
hold tanks in series than with a single hold tank.
                              146

-------
                           CONTENTS

Section                                                       Page

   1     INTRODUCTION                                      149

   2     ADVANCED TEST PROGRAM OBJECTIVES
        AND SCHEDULE                                      157

   3     LIME TESTING WITH A CHEVRON MIST
        ELIMINATOR IN THE VENTURI/SPRAY
        TOWER SYSTEM                                     161

        3. 1     Testing at Constant Gas  Velocity              161
        3.2     Variable  Load Testing                         168
        3. 3     Conclusions                                   170

   4     LIMESTONE TESTING WITH A  WASH TRAY AND
        CHEVRON MIST ELIMINATOR  IN SERIES IN
        THE TCA SYSTEM                                   172

        4. 1     Testing at 8.6 ft/sec Superficial
                Gas Velocity                                  172
        4.2     Testing at 10 and 12 ft/sec  Superficial
                Gas Velocity                                  175
        4. 3     Conclusions                                   176

   5     LIMESTONE UTILIZATION TESTING IN THE
        VENTURI/SPRAY  TOWER AND TCA SYSTEMS        177

        5. 1     Utilization  Testing  in the Venturi/Spray
                Tower System with Variable Residence Time  178
        5. 2     Utilization  Testing  in the TCA System with
                Three Hold Tanks in Series                    184
        5. 3     Conclusions                                   193

   6     EFFECT OF ALKALI UTILIZATION
        ON MIST ELIMINATOR OPERABILITY                196

        6. 1     Summary of Mist Eliminator Operability
                During Lime and Limestone Reliability
                Testing                                        196
                              147

-------
Section                                                       Page

         6. 2    Mist Eliminator Operability During
                Limestone Utilization Testing                  198
         6. 3    Conclusions                                   203

   7     REFERENCES                                        204
                                148

-------
                             Section 1


                         INTRODUCTION




In June 1968, the EPA  initiated a program to test prototype lime and


limestone wet-scrubbing systems for removing  sulfur dioxide and par-


ticulates  from coal-fired boiler flue gases.  The program -was con-


ducted in a  test facility integrated into the flue gas ductwork of boiler


No. 10 at the Tennessee Valley Authority (TVA) Shawnee Power Station,


Paducah, Kentucky.  Bechtel Corporation of San Francisco  was the


major contractor and test director, and TVA was the constructor and


facility operator.




The results of testing at the facility during the original program,


nrfaich lasted from March 1972 to  May  1974,  are presented in References


1 and 2.  The most significant reliability problem encountered during


the testing  was associated with scaling and/or plugging of mist elim-


ination surfaces.




In June 1974, the EPA, through its Office of Research and Development

                                                  **
and Industrial Environmental Research Laboratory   initiated a three-


year advanced test program at the Shawnee facility.  Bechtel Corporation


•is continuing as the major contractor and test director, and TVA  as
   The National Air Pollution Control Administration prior to 1970,

   The Control Systems Laboratory prior to 1975.
                                149

-------
the constructor and facility operator.  The major goals established

for the advanced program are: (1) to continue long-term testing with

emphasis on demonstrating reliable operation of the mist elimination

systems at increased gas velocity, (2) to investigate advanced process

and equipment design variations for improving system reliability and

process economics, and (3) to perform long-term (2 to 5 month)

reliability testing on promising process and equipment design variations.




Two parallel scrubbing systems are being operated during the advanced


program:




     •   A venturi followed by a spray tower

     •   A Turbulent Contact Absorber (TCA)




Each system has its own slurry handling facilities and is capable of

treating approximately 30,000 acfrn  of flue gas from the TVA Shawnee

coal-fired boiler No. 10.  This gas rate is equivalent to approximately

10 Mw of power plant generating capacity.  Boiler No.  10 normally

burns a high-sulfur bituminous coal which produces SO_ concentrations

of 1500 to 4500 ppm and inlet particulate loadings of 2 to 4 grains/scf


in the flue gas.




Figures 1-1  and 1-2 (drawn with major dimensions to scale) show the

two scrubber syste/ns along with the mist elimination systems currently

used for de-entraining slurry in the exit gas  streams.   The cross-

sectional area of the spray tower is 50 ft^ in both the scrubbing section

and mist elimination section.   The cross-sectional area of the TCA is
    ty                                 O               '
32 ft^ in the  scrubbing section and 49 &  in the mist elimination section.
                                150

-------
            CHEVRON MIST
             ELIMINATOR
       SPRAY TOWER
                   GAS IN
                     I
INLET SLURRY	wt^X

         THROATA
  ADJUSTABLE PLUG
 YENTURI SCRUBBER
       A>!
! PLUG/   U
                                 GAS OUT
                                       «
                               WWW
                     A   A   TV
                               A  A  A
                                        MIST ELIMINATOR
                                         WASH WATER

                                        MIST ELIMINATOR
                                         WASH LIQUOR
                               A  A  A
                     A   A   TV
r
t.
                                  INLET SLURRY
                                                       5'
                                                  APPROX.SCALE
                              EFFLUENT SLURRY
   Figure 1-1.  Schematic of Venturi Scrubber and Spray Tower
                               151

-------
MIST ELIMINATOR
  WASH WATER
        CHEVRON MIST
         ELIMINATOR
RETAINING IAR-GRIDS
           6AS IN
                           GAS OUT
T    U
                           A  7\  A
                           	o—„- -
    o
   >
 000
                            00 0°0

                               O °0
                          |2.0£L
 MIST ELIMINATOR
  WASH LIQUOR

•INLET SLURRY
                                        MOIILE PACKING SPHERES
                                              i	1
                                              APPROX. SCALE
                        EFFLUENT SLURRY
       Figure 1-2. Schematic of Three-Bed TCA  Scrubber
                               152

-------
The TCA utilizes a fluidized bed of 1  1/2  inch diameter,  light weight
(5.0-6. 5 gram) spheres which are free to move between retaining grids.

The mist elimination systems shown in Figures 1-1 and 1-2 each
consists of a 3-pass,  open-vane chevron  mist eliminator with provision
for underside and topside washing.  During the early portion of the
advanced test program, the TCA was also tested with a mist elimination
system consisting of a wash tray (Koch Flexitray) in series with a
6-pass, closed-vane chevron mist eliminator, both with underside wash.

A  typical system configuration used during venturi/spray tower lime
reliability  testing is  shown in Figure  1-3.  A typical system con-
figuration used during  TCA limestone utilization testing is  shown in
Figure 1-4.  In the TCA configuration shown, three effluent hold tanks
in series were used to increase limestone utilization.

This paper presents  the results of mist elimination and limestone
utilization  testing at the Shawnee facility  from June 1974 through
January  1976.  During this period the venturi/spray tower system was
operated with both lime and limestone and the TCA system with
 limestone.  All of the testing was performed under closed liquor loop
 conditions.  Due to the relatively high inlet gas particulate loading, the
 slurry solids contained about 40 to 50 wt % fly ash.

 For all of the tests during the reporting  period, the sulfate (gypsum)
 saturation of the scrubber inlet liquor was maintained below about
 140 percent and there was no significant accumulation of gypsum scale
 on scrubber internals.  Previous testing has shown that scrubber in-
 ternals can be kept relatively free of gypsum scale if the  sulfate
                                  153

-------
tn
THMB
1

PROCESS
WATER
HOLD
TANK
                                                                                                                              STACK
                                                                                                                         Discharge
                                                                                                                      SETTLING POND
                O  Gas Composition
                ®  Particulate Composition & Loading
                0  Slurry or Solids Composition
.. _  Gas Stream
_  Liquor Stream
                      Figure  1-3.  Typical Process Flow Diagram for Venturi/Spray Tower System

-------
en
          O  Gas Composition
          ®  Participate Composition & Loading
          ®  Slurry or Solids Composition
_ _  Gas Stream
—  Liquor Stream
                                                                                                                           SFJTLING POND
                            Figure 1-4.  Typical Process  Flow Diagram  for TCA System

-------
saturation of the  scrubber liquor is kept below about 135 percent (see
Section 7. 3 in Reference 1).  The effects of the process variables on
scrubber liquor sulfate saturation observed during the advanced test
program will be presented in a future paper.
                              156

-------
                            Section 2

   ADVANCED TEST PROGRAM OBJECTIVES AND SCHEDULE


The Shawnee Advanced Test Program is scheduled to  run from June

1974 through June 1977.  The objectives of the advanced program are:


    •   To continue long-term testing with emphasis on demonstrating
        reliable operation  of the mist elimination systems.

    •   To investigate advanced process and equipment design varia-
        tions for improving system reliability and economics.  For
        example, testing will be conducted to investigate the practical
        upper limit of the gas velocity (i. e. , minimum  scrubber
        size) at which the scrubber mist elimination  systems can be
        reliably operated.  Also, tests will be conducted to evaluate
        system performance under conditions of minimum energy
        consumption for the desired levels of SO2 and particulate removal.

    •   To evaluate process variations for substantially increasing
        limestone utilization and reducing sludge production.  Tests
        will be conducted with scrubber effluent passing through
        three  stirred tanks in series to approach a "plug-flow" con-
        dition and at reduced scrubber liquor pH to increase lime-
        stone utilization (see Reference 3).

    •   To evaluate scrubber operabiltiLy during variable load (e.g. ,
        cycling gas rate) operation.

    •   To perform long-term (2 to  5 month) reliability testing on
        advanced process and equipment design variations.
                               157

-------
To investigate methods of improving waste solids separation.
This may include testing a multiple-plate thickener, using
coagulants, attempting to relate sludge characteristics to
operating conditions,  and making operational improvements
on the centrifuge and filter.

To study schemes for oxidizing sludge in order to improve
solids settling characteristics and to reduce the chemical
oxygen demand (COD) of the sludge.   Oxidation tests will be
conducted using a two-stage scrubber system developed by
Borgwardt (Reference 4).

To evaluate the effectiveness of three commercially offered
 sludge fixation processes and of untreated sludge disposal.
 Fixed sludges (Chemfix, Dravo, and  IUCS) and untreated lime
 and limestone sludges are being continuously monitored in
 ponds at the Shawnee  site. Aerospace Corporation is the major
 contractor and test director for this effort.

 To evaluate  system performance and reliability without fly
 ash in the flue gas.  Tests will be conducted with flue gas
 taken downstream of the Shawnee boiler No.  10  electrostatic
 precipitator, i. e., with less than 0. 1 grain/scf of particulate
 in the inlet flue gas.

 To determine the practical upper limits of SO2  removal
 efficiency.  Tests will be conducted to determine the practical
 upper limit of SOo removal by increasing the scrubber slurry
 pH, increasing the slurry rate,  increasing the scrubber gas
 pressure drop,  and adding magnesium ion (MgO) to the slurry.

 To evaluate the TCA performance with lime and the venturi/
 spray tower performance with limestone.

 To characterize stack gas emissions including outlet par-
 ticulate mass loading and size distribution,  slurry entrain-
 ment,  and total sulfate  emission.

 To evaluate, under the  direction of TVA, corrosion and wear
 of alternative plant equipment components and materials.

 To develop a computer  program, in conjunction with TYA,   for
 the  design and  cost comparison of full-scale lime and limestone
 systems.

                          158

-------
The current test program  schedule,  based on the defined objectives,
is presented in Figure 2-1.  As can be seen in the figure, as of
January 1976, limestone tests were in. progress on both the venturi/
spray tower and TCA systems to: (1) demonstrate the reliability of
the mist elimination systems and (2) determine the effects of process
variables on limestone utilization.
                                 159

-------
   LIMtSTONt ADVANCED THTiNfl WTH 1CA |Y|TtM
     MflT tLM
            tOXlDI AODttlOM ritTtNQ
     ALKALI UTILIZATION UITINQ WITHOUT MfO ADDITION
     rACTofllAl TtlTINQ WITHOUT MfO ADDITION
     FACTORIAL TESTINQ WITH MfO ADDITION
     VARIABLE 10 AD Tt IT I WO
     FLV ASH FNtE TtlTINO
     Rlt(A»KITV OfMOMTHATKMf
     MINIMIZt ENtflav UTILIZATION TUT (WO
   UME ADVAftCI
     RELIABILITY TESTING
     FACTORIAL TE*T1NQ

   UME ADVANCED IWTIWO WJTH Vf NTUfll/JWUV T0W|« SVJTtM
     MIST ILIMINATOfl TESTING
     MAGNESIUM OXIDE ADDITION TESTING
     VARIABLE LOAD TESTING
     FACTORIAL TESTING'*'
     FLY ASH FRtE TESTING
     RELIABILITY Of MONSTNATION flUN
     MINIMUM ENFftOV UTILIZATION TCSTINO

   LIMESTONE
     MtlT ftlMINAT»R TISTtNC
     MADNESILtM DXIDE ADDITION TESTING
     ALKALI UTILIZATION TESTINO WITH AND WITHOUT MfO ADDITION
     FACTORIAL TfST ING WITHOUT MfO ADDITION
     FACTORIAL TESTING WITH M« ADDITION
                    ESTINO
     VARIABLE LOAD  TESTING

   tESTINQ COMMON TO BOTH TRAINS

     SLUDGE CHARACTERIZATION
     STACK OAB I MISSION CHARACTf RIZATION


6  TESTING

7  GENERAL PUBLICATION AEPOHT DRAF T SUBMITTAL DATES





          111 INCLUDES "MAXIMIZING SOj REMOVAL EFFICIENCY  TESTING! MAY INCLUDI ADDITION OF MfOI
          NOTE OASMfD LINES HEfRESI NT SEQUENCES WHICH MAT BE CONTINUED DURING THE HniOO INDICATED
                           Figure  2-1.     Current  Schedule  for  Shawnee  Advanced  Test  Program

-------
                             Section 3

      LIME TESTING WITH A CHEVRON MIST ELIMINATOR
            IN THE VENTURI/SPRAY TOWER SYSTEM
This section summarizes the results of mist elimination reliability
testing from June 1974 through mid-October 1975 with lime on the
venturi/spray tower system.   During this period, a 3-pass,  open-
vane, 316 stainless steel chevron mist eliminator was used in the
spray tower.


3. 1     TESTING AT CONSTANT GAS VELOCITY


During the period of advanced testing from June 1974 through
March 1975, several mist eliminator washing configurations  were
evaluated  under the following typical test conditions:
        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
        Weight percent solids recirculated          8
        Effluent residence time                     12-24 minutes
        Scrubber inlet slurry pH (controlled)        8
        Weight percent solids in discharge cake     43-60
        Inlet gas  SO_ concentration                 1150-4250 ppm
*
 •In this report,  all gas velocities and'liquid-to-gas ratios are at scrub-
 ber operating conditions, i. e. , saturated gas at scrubber temperature.
 With flue gas operation, the scrubber temperature is approximately
 125°F.  The gas velocities are all superficial velocities.
                                161

-------
For these test conditions, SO_ removal was 75 to 95 percent,  lime
          ;';
utilization  was approximately 90 percent,  the  scrubber outlet liquor

pH was approximately 5. 0, the sulfate (gypsum) saturation of  the

scrubber inlet liquor was 105 to 140 percent, the dissolved solids

in the  scrubber liquor was 6000 to 12000 ppm, and the total pressure

drop was about 12. 5 inches HO, including 9 inches HO across the

venturi.
                                     .JL...JU
Initially (Runs 604-1A through 608-1A") the mist eliminator was

washed from the bottomside only.  With this wash configuration,  soft

 solids accumulation did not occur on the mist eliminator, but there

was a continual problem of gypsum scale formation on the top vanes.

 The runs showed that this  scale formation rate on the top mist elim-

 inator vanes (25 to 50 mils/week) was relatively unaffected by the

 wash cycle, wash rate, or quality (sulfate saturation) of the wash

 liquor.  Run 608-1A showed that intermittent washing with high

 pressure (45 psig) raw water at a rate of 3 gpm/ft2 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/ft2 (Run 606-1A). This is significant in  that intermittent

 washing may be required in closed liquor loop  operation due to restric-

 tions in the allowable raw water makeup to the scrubber system.
  *Percent alkali utilization is defined as 100 x moles SO2 absorbed
   per mole of Ca added.   The reciprocal of the fraction alkali utiliza-
   tion is  stoichiometric ratio, defined as moles Ca added per mole
   of SO2  absorbed.


 **Detailed operating conditions for Runs 604- 1A through 624- 1A can be
   found in References 1 and 2.
                                162

-------
Starting in late September 1974, runs were made  (Runs 609-1A and
                                                               2
610-1A) in which the entire mist eliminator underside and a 14 ft
    »t-
area  on the topside were washed at high pressure (45  psig) with

makeup water at a rate of 2. 7 gpm/ft  for the underside and 1. 0

gpm/ft  for the topside in an intermittent cycle of about 8 minutes

every 4 hours.  After 530 hours of  operation, the washed area of

the top mist eliminator vanes was essentially clean (less than 1 mil

of solids accumulation) compared with an average of 70 mils scale

buildup on the rest of the topside vane surfaces.  The mist eliminator

bottom vanes were covered with smooth white scale about 10 mils

thick.
In the final run at 6. 7 ft/sec spray tower gas velocity (Run 623-1A),

begun in early March 1975, a wash system was installed in the spray

tower to provide sequential and intermittent washing of the topside

vanes of the chevron mist eliminator.   The topside wash was ac-

complished by operating  6 nozzles in sequence.  Every 80 minutes,

one nozzle was actuated for 4 minutes at a rate of 0. 5  gpm/ft  at
                                               2
13 psig.  The underside wash rate was 3 gpm/ft  at 45 psig for 3. 5

minutes every 4 hours.
*
 Only a small section of the topside was washed because of a concern
 that entrainment from the top spray might overload the reheater and
 possibly allow moisture to reach the fan.  In later tests,  the entire
 topside was  washed by resorting to  sequential washing.  Another
 possible solution would have been to use a second mist eliminator
 to intercept  entrainment from the topside sprays.
                                163

-------
After 162 hours of operation, an inspection  showed that the combina-



tion of sequential topside wash and intermittent underside wash was



successful in preventing scale and solids accumulation on the mist



eliminator.  Only scattered dust a few mils thick was  formed on some



of the vanes.







A new run (Run 624-1A) was started on March 19, 1975, without system



cleaning, to test the mist elimination system at a higher spray tower



gas velocity of 8. 0 ft/sec.  For this run, the intermittent wash cycle



for the underside of the mist eliminator was increased from 3. 5



to 4. 3 minutes every 4 hours because of the higher makeup water



rate available at the higher gas velocity.  The wash rate and cycle



for the topside were not changed from the previous run.  Run 624-1A



was terminated on April 23, 1975, after 823 hours  of operation due



to a scheduled 5.-week maintenance  outage on boiler No.  10.








A total of 4 inspections were made for this run, including one  at the



end of the  run.  The mist eliminator was found to be free of any



restriction during each of these inspections. Vane surfaces washed



earlier in the wash cycle usually held 2 mils of scattered white dust,



while recently washed vane surfaces were entirely  clean.  Operating



data for  the initial 480 hours of Run 624-1A are presented  in Figure 3-1.







Following the boiler outage, a run was begun on June 20,  1975, (Run



625-1A)  to continue testing  the mist eliminator at 8  ft/sec  spray tower



gas velocity with a reduced underside wash  rate of  1. 5 gpm/ft for 4. 3



minutes  every 4 hours.  The wash rate  and cycle for the topside were



the same as for the  previous two runs.  Only about one-half of the makeup
                                 164

-------
          ;  BEGIN RUN 62* 1;
      3.500

      3.°°°
is   ""
= "   2.000
                                                         3,500

                                                         3.000

                                                         2,500

                                                         2.000

                                                         1.500
                        I 322  I 3/23 I  3;24  I 3/25 I  3/26  I 1/27  I 3/Z8 I  3/29  I 3/30 I  3/31  I
                                                            CALENDAR DAV
jj "
: :, u
1
1 0
Hi 30
1

£ W
2 10
o
150

1 ,0.
s «
.
_ . ^ ^
"*"" /^ ^ ^^\ ^ ..*__-- S^-~--~--S * -^*-
* " ^ * ^
— f
'" ' J>v r-'"''"

* -A- *^^" / ' / *
* " ^ ^ ' '"* '/^v/N/ '
-
-
*" * -A. /*" *

'— " , _--- — -^-/
• TOTAL DISSOLVED SOiLDS NOTE SPECIES WHOSE _,

9,000
i B.OOO
1 1 '•M0
z e 6,000
3 O 5,000
§ 9
Q ^ 4.000
0 Z J.OOO
5 2,000
1.000
0 O CALCIUM (C»**l CONCENTRATIONS ARE LESS
• • U SULFATE ISO/) THAN 600 pom ARE NOT
A CHLORIDE Cl h PLOTTED jr
•**• », . •*. -
•* •«•••*

A A*
* * ** »» » A A » A * A*

- - °° ° I °'° 'y' ' ^ oo * » o o o o o o o ° > SS
_t cf C Z Dn_
i i i i i i i " i i i i
1.3
12
1.1
1.0
30


20
10
0
ISO

100
so
10.000

9,000
8,000
7.000
6.000
5,000

4.000
3.000
2,000
1.000
0 40 80 1 20 1«0 20O 240 280 320 360 400 440 «80
TEST TIMS Houn
CALENDAR DAY
                     Gas Rite; 30.000 acfm ^ 330 °F
                     Liquor Rate lo Veniuri ~- 600 gpm
                     Liquor Rate lo Spray Tower ; 1200 gpm
                     Venturi L/G = 25 gal/mcf
                     Spray Tower L/G = 50 gal/met
                     Sprav Towei Gas Vetocilv = S.OIt/iec
                     No. of Spray Headers = 4
                     EHT ftendence Time = 17 min
Peicent Soiids Recirculated = 7-10 wrt %
Venturi Pressure Drop - 9 in HjO
Total Pressure Drop, Excluding Mist Elim. - 12.5-12.6 m
ScrubbQr Inlet Liquor Temperature - 120-126 °F
Liquid Conductivity - 6,100-10,000 u, mhos/cm
Discharge (Clarifier and Filter) Solids
    Concentration = 4S-&4 wt %
Lime Addition to Scrubber Downcomer
Figure  3-1.   Operating  Data  for  Venturi/Spray  Tower  Run  624-1A
                                                              165

-------
water available during closed liquor loop operation -was needed to wash

the mist eliminator.   After 319 operating hours, the chevron mist

eliminator was essentially clean (less than 2 percent restricted with

dust).




The spray tower gas  velocity was increased to the maximum achieve-

able value of 9.4 ft/sec in the next run (Run 626-1A) which started on

July 9,  1975,  with no cleaning of the mist eliminator. The inter-

mittent wash cycle for the underside of the mist eliminator was

increased to 6 minutes every 4 hours because of the higher makeup


•water rate available at the higher gas velocity.  The topside wash rate

and cycle were unchanged. As in the previous run,  only about one-


half of the makeup water available during closed liquor loop operation

was needed to wash the mist eliminator.  An inspection after  569 hours
               «
of operation showed that the chevron mist eliminator had remained

essentially  clean during the test (less than  2 percent restricted).

Operating data for the initial 480 hours of Run 626-1A are presented


in Figure 3-2.




The  recirculation slurry solids  concentration  was increased from


8 wt % to  15 wt % in Run 627-1A which began on August 5,  1975.

The  spray tower gas  velocity was maintained at 9.4 ft/sec.  During


this  run,  the inlet gas SO_ concentration dropped to a low value
                        ^                        .o-
                                                  'i*
of about 1500 ppm from August 6 through August 8.   Asa consequence

during that  period, the outlet scrubber liquor  pH rose to about 6. 2

(the scrubber inlet liqiior pH was controlled at 8. 0).  An inspection
*
 Low sulfur Montana coal was being burned in boiler No.  10.
                               166

-------
 If*
 cii
     10.000

      9,000
 2 5   5-°°°
 82
 O -J   4.000
                                                                                          r
                                                             U CULFATE I004-|
                                                             * CHLOR1OE ICI " •
                       AHE L£S3 IM.-kN 500,1.1
                       .\HE NOT PLOTTED
                                                          .  •
                 .••    .
                                                                                                       6.000

                                                                                                       S.OOO
                                                                                                       3000

                                                                                                       7.000
                                                                              7 ?« I l 26 I ' Z6 I  1 ?7 I 7 28 I
         GaiRiti* 35.000 acfm * 330 °F
         Spray ToMtr G« Velocity " 9.4 fl/HC
         Liquor Rat» to Venturi - 600 gpm
         Liquor Rati to Spray Tower • 1,400 gpm
         V«ntutlL/G-21sd/mc(
         Sony Tomr IVG - 50 pl/mct
         No. of Spny Hndtn • 4
         EHTRMlH.net Tirni - 12 mln
Percent Solid! Recirculsterl - 8-9 wt %
Venturi Pressure Drop • 9 in. H^O
Ton) Prenure Drop. Excluding Milt Film. - 14.2-15 in.
ScrublMi Inlet Uqupr Tempmturt • 130-133 °F
Liauid Conductivity • 11,000-17,000 u mhoi/cm
Ditcharte IdirlHer ind Film) Solid!
   Concentration - 57-60 wt %
Ume Addition to Scrubbw Oowwcomw
Figure  3-2.   Operating  Data for  Venturi/Spray  Tower  Run  626-1A
                                                     167

-------
on August 8 showed that the higher level of liquor pH within the spray

tower resulted in calcium sulfite scale formation within the tower
                                                      -i-
and on the lower vanes of the chevron mist eliminator.   An inspection

at the end  of the run,  after 187 hours of operation,  showed that the

scale had diminished  somewhat during the latter half of the test,  and

that the mist eliminator was  2 to 3 percent restricted with sulfite

scale and dust. At the conclusion of Run 627-1A, the mist eliminator

had been in operation for 1075 hours without cleaning, including 319
                                i
hours at 8. 0 ft/sec and 756 hours at 9.4 ft/sec gas  velocity.



 3. 2      VARIABLE LOAD TESTING


 Run 628-1A was begun on August  16 to test the operability and con-

 trollability  of the  venturi/spray tower  system under cycling gas  load.

 The gas flow rate for this run (17, 000 to 35, 000 acfm) was varied with

 the actual Unit No. 10 boiler load (about 60 to 160 Mw),  resulting in

 spray tower gas velocities between  4. 5 and 9. 4 ft/sec.  Constant liquor

 rates were  used in the venturi and spray tower.  The pressure drop

 across the venturi was held  constant at 9 inches HO for all gas  rates
                                                 L+
 by varying the plug position.  The same mist eliminator wash scheme

 as in Runs 626-1A and 627-1A was  used during the  test.  Operating

 data for the  initial 480 hours of Run 628-1A are presented in  Figure 3-3.

 As  can be seen in the figure, the  inlet gas SO_ concentration  varied

 between  1500 and-4200 ppm during the test.
  This problem of sulfite scale formation at high outlet scrubber
  liquor pH was corrected in subsequent testing by controlling the
  outlet  scrubber liquor pH  to 5. 0_+0. 5, with the constraint that the
  scrubber inlet liquor pH remain equal to or less than 8. 0.
                                 168

-------
               : MGINRONU*!*
                                                                                 M 	-i
                                                                    V
                                    sC^
     GH Rate - 17.000-3B.OOO aclm S> 330 °F
     Spray Tower G« Velocity « 4.6-9.4 ll/stc
     Liquor Rate to Venturi - 600 gpm
     Liquor Rate to Spray Tow«r - 1.400. 1.600 [after 8/25)
     Venturi L/C - 21-««al/"
-------
An inspection after 717 hours of operation showed that the mist elim-


inator was essentially clean (only 2 percent restricted with dust).  No


problems were  experienced in controlling the system during the test.





 Rurl 628-1B was a continuation of 628-1A, except that the venturi plug


 position was fixed to give a maximum venturi pressure drop of 9


 inches H O at the maximum gas rate of 35, 000 acfm.  Operating
         Ct

 data for this test are presented in Figure 3-4.  An inspection after


 426 hours of operation (for a. total of 1143 hours for Runs 628-1A and


 628-IB) showed that the mist eliminator -was essentially unchanged


 (2 percent restricted with dust)'.





 3. 3      CONCLUSIONS





 The long-term  (6-12 month) operability of the 3-pass,  open-vane chevron


 mist  eliminator in the venturi/spray tower system with lime additive


 has been demonstrated for intermittent high pressure underside washing


 and sequential low pressure topside washing with raw water.  The


 mist  eliminator remained essentially free of solids deposit  for 756


 hours of operation at a spray tower  gas velocity of 9. 4 ft/sec and with


 8 to 15 wt % solids in the recirculating slurry.  The mist eliminator


 also remained free of solids deposit during variable load (cycling


 gas rate) tests  lasting a total of 1143 hours.  For the lime testing,


 only about one-half of the makeup water available during closed liquor


 loop operation was needed to wash the mist eliminator.   The lime


 utilization for these tests was about 90 percent.
                               170

-------
                                                                                           T~
           11}
           I'l
            i!
            si
    Gas Rate * 19,000 • 35.000 acfm @ 330 °F
    Spray Tower Gas Velocity ~ 5.1-9.4 ft/sec
    Liquor Rate to Ventun ~ 600 gpm
    liquor Rate to Spray Tower - 1.600 gpm
    Ventun L/G - 21-39 gal/md
    SpravTowet L/G - 57 105 gal/met
    No. ot Spray Headers - 4
    EHT Residence Time -12mm
Percent Solids ftecircuiated - 8.69.7 wl %
Ventun Pressure Drop  4.0-9.0 m. HjO
Total Preiiure Drop, Excluding Misi Elim. - 6.5-14.6 in.
Scrubber Inlei Liquor Temperature = 129'T31 °F
Liquid Conductivity " 8,000-11,500 u. rnhos/cm
Oisctiarge (Clanfier and Filter) Solids
    Concentration - 52-56 wt %
Lime Addvion lo Scrubber Downcomer
Figure  3-4.  Operating  Data  from  Venturi/Spray  Tower  Variable
                       Load  Run  628-1B
                                                          171

-------
                             Section 4

         LIMESTONE TESTING WITH A WASH TRAY AND
  CHEVRON MIST ELIMINATOR IN SERIES IN THE TCA SYSTEM
This section summarizes the results of mist eliminator reliability
testing from September 1974 through March 1975 with limestone in

the TCA system.  During this period the mist elimination system

consisted of a wash tray (Koch Flexitray) in series with a 6-pass,

closed-vane, 316 stainless steel chevron mist eliminator.


4. 1      TESTING AT 8. 6 FT/SEC SUPERFICIAL GAS VELOCITY


TCA limestone Run 535-2A was begun on September 12,  1974,  and

continued through December 4,  1974,  for a total of 1835 operating
      *
hours.   The TCA internals consisted of 3 beds  (4 grids), with 5

inches per bed of 1 1/2 inch diameter,  6 gram TPR (thermoplastic

rubber) spheres.  The major test conditions at the  start of the

run were:
         Gas velocity                               8. 6 ft/sec
         Liquid-to-gas ratio                         73 gal/mcf
         Percent solids recirculated                 15
         Effluent residence time                     12 min
         Percent SO- removal (controlled)           84
         Inlet gas SO? concentration                 2000-4000 ppm
         Weight percent solids in discharge sludge   35-42
 i
 TCA Run 535-2A is described in detail in Reference 2.
                                172

-------
For these test conditions, the scrubber inlet liquor pH varied between

5. 7 and 6. 0, limestone utilization was approximately 65 percent

(stoichiometric ratio of 1. 54 moles Ca added/mole SO7 absorbed),

the sulfate (gypsum) saturation of the scrubber inlet liquor averaged

110 percent,  the dissolved  solids concentration in the scrubber

liquor was 4000 to 8000 ppm, and the total pressure drop (including

the mist elimination system) was 6 to 7 inches HO.  Operating data

for the initial 480 hours of testing are presented in Figure 4-1.
The clarified liquor return flow rate was maintained at a minimum of

15 gpm for Koch tray feed and mist eliminator wash to prevent high

sulfate  supersaturation in the wash tray effluent liquor.  The under-

side of  the mist eliminator was washed continuously with 15 gpm (0. 31
      2
gpm/ft  ) diluted  clarified liquor (about 9 gpm makeup water plus about

6 gpm clarified liquor).   The tray was fed with  the 15 gpm mist elim-

inator wash plus  the remaining /^ 9 gpm (minimum) of clarified  liquor
            2
(0. 50  gpm/ft ).  The underside of the tray was  sparged with 125 psig

steam for one minute each hour.
Throughout the first half of Run 535-2A, there was a continual problem

of solids accumulation on the scrubber walls between the steam sparger

and the main slurry spray header.  This problem -was resolved by

installing four wall spray nozzles utilizing the wash tray effluent liquor.

It was  observed that the wall sprays were also effective in washing

the underside of the tray -where it was covered by the sprays.




After 1835 operating hours for  Run 535-2A,  an inspection showed that

the top surface of the wash tray was entirely clean and that the chevron
                                173

-------
 40       H       120       ISO       im      240       !BO
                                        21  I 9?? I  9 ?3 I  9>?4  ! 9 2S I  9'76 I  9'2J  I 9/28 I 9'M I  9/30  I 10'1 I
1 7
l] "
\'s '"
^ M
? | i.«

1 3
S t K
S f
H J M
Si to
g * "
13000
-
i '•

til 000

cc 10.000
; 9.000
^ 8000
£
£ 7.000
1
s '•M0
? 5.000
S
3 4000
S
2 3,000
S 2000
1.000
0

	 ^ 	 ' — ~ 	 " / \
' \ " "\ (\ /
- /' "• V /' \ T'^' w i \ /^\ /v •' •
^ \ '' ^' ^- ' \ /
* V \ / "^ "*'i«-/^
"" -• • \ .'
' ' \ / -* ~* A
A . \y

- \^ ? ' ^ \ ^.^ \ ^ * ^ ^ ^"* .j » y v \ "
x " %
,
• TOTAL DISSOLVED SOLIDS
«"• CALCIUM (C.'M
ri SULFATE IS<>3 '

A CHLORIDE (Cl I
NOTE SPECIES WHOSE
CONCENTRATIONS ARE LESS
THAN WO ppm ARf. NOT
PLOTTED
*
9

• • • 9 f



-A * * A A •, A &
-8@3*o£ fiSn^o^g* §D f  300 °F
   Liquor Rale- I200gpm
   L/G = 73 gal/mcl
   Gas Velocity = 8.6 It/sec
   EHT (Sealedl Residence Time • 12 min (9/12-9/27),
      15 mm (alter 9/271
   T^ree Stages. 5 in spheres/stage
Petcent Solids Recitculated - 12-15 wt %
Total Pressure Drop. Excluding Mist Elim.
   and Koch Tray = 4.0-4.6 in H^O
Scrubber Inlet LiQuor Temperature =120-126 °F
Liquid Conductivity • 4.800-10,000 u  rr.hos/cn
Discharge IClsrifierl Solids
   Concentration - 3642 wl %
Figure  4-1.   Operating  Data  for  TCA  Run  535-2A
                                           174

-------
mist eliminator was less than 5 percent restricted by solids (mostly
fly ash).  The underside of the wash tray had no solids accumulations
greater than 1/4 inch thick.  These deposits did not interfere with tray
operation.  Also, the scrubber walls between the tray and slurry spray
header were clean.

Run 535-2A was arbitrarily terminated on December 4, 1974, and Run
535-2B begun,  when low sulfur Montana coal was unexpectedly introduced
into Shawnee  boiler No. 10.  The wide-swings in inlet gas SO2 concentra-
tion (1200 to 4900 ppm in 8 hours) caused system upsets which resulted
in some scale formation on the wash tray and solids accumulation
in the chevron mist eliminator. During Run 535-2B, the steam  sparger
            *
was removed  and the four wall spray nozzles were replaced by a single
tray under wash nozzle, which also provided rinsing of the walls below
tiie tray.  This modification  (see Figure 3-2 in Reference 2) proved to
be effective in keeping both the underside of the wash tray and the walls
below it clean.

4.2     TESTING AT 10 AND 12 FT/SEC SUPERFICIAL GAS VELOCITY

At scrubber gas velocities of 10 ft/sec or higher, plugging of the chevron
mist eliminator by solids (mostly fly ash) became a problem.  At 10
ft/sec scrubber gas velocity (6. 5 ft/sec in the mist eliminator and wash
tray areas), the overall restriction of the mist eliminator increased
gradually to about 8 percent  during the first 500 hours of operation and
  In a full scale gas scrubbing unit, steam cleaning of the underside of
  a wash tray -would be an economic burden.
                                  175

-------
appeared to level out at 6 to 8 percent after 562 operating hours (Run



538-2A).  At 12 ft/sec scrubber gas velocity (7.8 ft/sec in the mist



eliminator and •wash tray areas), the mist eliminator was  11 percent



restricted by solids within  only 215 hours of operation (Run 539-2A).







4. 3      CONCLUSIONS







The long-term operability  of the TCA mist elimination system was



demonstrated in an 1835-hour run in limestone service at 8.6 ft/sec



superficial  gas velocity and 15 wt % solids in the recirculating slurry.



The mist elimination  system consisted of a Koch Flexitray in series



with a 6-pass, closed-vane,  chevron mist eliminator, both with under-



side wash.  Long-term operability was not demonstrated at increased



gas  velocity.   At 10 ft/sec the chevron mist eliminator was 8 percent



restricted with soft solids  after 562 operating hours,  while at 12 ft/sec



the chevron mist eliminator was 11 percent restricted in only 215 hours.
                                 176

-------
                             Section 5

             LIMESTONE UTILIZATION TESTING IN THE
            VENTURI/SPRAY TOWER AND TCA SYSTEMS
The results of limestone utilization testing from October 1975

through January 1976 on both the venturi/spray tower and the TCA

systems are presented in this section.  These tests were made as a

result of a TVA  economic  study which showed that a potential existed

for significantly improving the economics of limestone scrubbing

by improving the utilization of the  limestone feed.  Improved lime-

stone utilization not only results in a decrease in limestone feed re-

quirements but also a  corresponding decrease in waste sludge production.


Tests were conducted  primarily to determine the effect of  scrubber

inlet liquor pH,  effluent residence time, and hold tank design on

limestone utilization.   These variables were correlated with stoichio-

metric ratio (the reciprocal of fraction alkali utilization).  For the

venturi/spray tower system a single backmix effluent hold tank was

used.   For the TCA system both a single backmix hold tank and three

backmix hold tanks in  series  (to simulate a plug flow  reactor) were

used.


For each combination  of residence time and hold tank design,  tests

were conducted  to cover a  range of values of  scrubber inlet liquor

pH.  Normally,  the systems  were run for about 4 to 5 days at a
                                 177

-------
specified level of pH.  During testing, the stoichiometric ratios were
determined every 4 hours from solids analyses of the scrubber re-
circulation sVurry.

The limestone used during the advanced program was "Fredonia Valley
White", purchased from Fredonia Quarries, Fredonia, Kentucky.   The
limestone  analysis was: 96 wt % CaCOy 1 wt % MgCO ,  and 3 wt  %
inerts. The limestone size distribution was: 90 wt % less than 325
mesh, 87 wt % less than 30 microns, 82 wt % less than 22 microns,
and 52 wt % less than 6 microns.

Results of analyses presented here should be considered preliminary
as  the data reduction is still in progress.  For  example,  no attempt
has been made in this  report to adjust the pH for chloride concentration
in the slurry liquor, which ranged from 1500 to 6500 ppm during the
testing.

Both  the venturi/spray tower  system and the TCA system were operated
with a single stage, 3-pass,  open-vane,  316 stainless steel mist elim-
inator with top and bottom wash. Details of the mist eliminator systems
and the effect of alkali utilization on mist eliminator  operation are
discussed  in Section 6.
5. 1      UTILIZATION TESTING IN THE VENTURI/SPRAY TOWER
         SYSTEM WITH VARIABLE RESIDENCE TIME
In the venturi/spray tower system,  limestone tests were made with
a single backmix effluent hold tank to determine the effect of scrubber
                               178

-------
inlet liquor pH and residence time on stoichiometric ratio.  Runs

were made at 20, 12, and 6 minutes  residence time.  Major test con-

ditions maintained constant during this period were:


        Spray tower gas velocity              9.4 ft/sec
        Venturi liquid-to-gas  ratio            21 gal/mcf
        Spray tower liquid-to-gas ratio       50-57 gal/mcf
        Venturi pressure  drop                9 in. HO
        Percent solids recirculated           15


Data showing the relationship between stoichiometric ratio and scrub-

ber inlet liquor pH for the  venturi/spray tower system are plotted in

Figures 5-1, 5-2, and 5-3 for 20, 12, and 6 minutes residence time,

respectively.  As would be expected, scatter in the data was greatest

at the  shortest residence time, where pH recovery time is limited and
a small  variation in  tank level can result in a significant change  in

residence  time.


Sight average curves drawn through  the data in Figures 5-1 through

5-3 are  cross plotted in Figure 5-4, which shows the effect of effluent

hold tank residence time and scrubber inlet liquor pH on stoichio-

metric ratio.  Comparing  20 minutes residence time versus 12 in

Figure 5-4,  little effect of residence time on stoichiometric ratio

tan be seen.  However, between  12 minutes and  6  minutes, a signifi-

cant increase in stoichiometric ratio with decrease in residence time

can be seen for  scrubber inlet liquor pH's greater than or equal  to

about 5. 6.
 All but 2 tests were at 50 gal/mcf.
                                179

-------
   1.80
  1.70-
  1.60 •
V50!
8
  1.40- •
2 1-30 +
u
c
Ul
I  1.20
u
i
  1.10
  1.00
     4.80
                                                         8
VENTURI/SPRAY TOWER SYSTEM
SINGLE HOLD TANK
20 MINUTES RESIDENCE TIME
   5.00
__l	,	,	,_
 5.20       5.40       5.60        5.80
      SCRUBBER INLET LIQUOR pH
6.00
                                                                             6.20
     Figure 5-1. Stoichiometric latio versus Scrubber Inlet Liquor pH for
                  a Single Hold Tank at 20 minutes Residence Time
                                        180

-------
1.80
1.70 -
   4.80
5.00
5.20        5.40        5.60       5.80
      SCRUBBER INLET LIQUOR pH
                                                                   6.00
                                                                6.20
    Figure 5-2.  Stoichiometric Ratio versus Scrubber Inlet Liquor pH for a
                  Single Hold Tank at 12 minutes Residence Time
                                     181

-------
  1.80
  1.70- •
  1.60- •
 CM
8
  1.50- •
  1.40- •
!'•*>+
ec
u
ui
O 1.20-
O
i
  1.10 •
  1.00 - •
        VENTUR I/SPRAY TOWER SYSTEM
        SINGLE HOLD TANK
        6 MINUTES RESIDENCE TIME
                                                °00°
                                             
                         —t—
                          5.20
4.80
5.00
     5.40       5.60
SCRUBBER INLET LIQUOR pH
5.80
                                                              6.00
                                                                         ft 20
    Figure 5-3. Stoichiometric Ratio versus Scrubber Inlet Liquor pH for a
                 Single Hold Tank at 6 minutes Residence Time
                                      182

-------


1.5
IOMETRIC RATIO.
•d/mola SO 2 absorbed
-* ^
CO *
M
I | 1.2
1.1
1.0 -
i r- 	 1 	 1 	 1 i 	 1 	 1 	
VENTUR I/SPRAY TOWER SYSTEM
SINGLE HOLD TANK
^
\
o — «. 	 i
~~""*-c> 	 ~—o '

• ' ' 1 	 1 	 1 	 1 I
6 8 10 12 14 16 18 20
                  EFFLUENT HOLD TANK RESIDENCE TIME, MINUTES
Figure 5-4. The Effect of Effluent Residence Time and Scrubber Inlet
            Liquor pH on Stoichiometric Ratio
                               183

-------
Figure 5-5 shows the general relationship between SO-, removal and


stoichiometric ratio for an inlet gas SO_ concentration range between
                                      LJ

2500 and 3500 ppm.  Data points have been plotted for hold tank residence


times of 20,  12,  and 6 minutes and a  sight average line through all


the data has been drawn.  Referring back to Figure 5-4, it was shown


that, for a given stoichiometric ratio greater than about 1.25,  the


scrubber inlet liquor pH is lower at 6 minutes residence time than at


12 or 20 minutes.  Therefore,  a corresponding reduction in SO_ removal


at 6 minutes residence time was expected.  However, within the scatter


of the  data, such a decrease could not be discerned in Figure 5-5.






Figure 5-6 shows the  relationship between SO- removal and scrubber


inlet liquor pH for an  inlet gas SO- concentration range from 2500


to 3500 ppm and a  12 minute hold tank residence time.  Averages from



 Figure 5-6 and from similar plots for higher and lower inlet gas SO


 concentration ranges at 12 minutes residence time are drawn in Figure


 5-7, which shows the  effect of inlet gas SO   concentration and scrubber


 inlet liquor pH on SO   removal.  As expected,  an increase in inlet
                    C*

SO  concentration  results in a decrease in SO  removal at constant
   L*                                        £*

 inlet pH (see Equlation  14-7 in Reference 1).
 5. 2     UTILIZATION TESTING IN THE TCA SYSTEM WITH

         THREE HOLD TANKS IN SERIES
 Kinetic theory shows that for a continuous system where the reaction


 order is greater than zero,  raw materials are more completely con-


 verted in a plug flow reactor than in a backmixed reactor of the  same


 residence time.  This  concept for improving utilization was success-
                                 184

-------
  100
  95-
   90 -
   85 •-
   80 --
cc
 (SI
2
Ul
O
EC
ui
a.
   70--
   65 -
   60 -•
   55 -•
   50
                        n
                        9
                                          VENTURI/SPRAY TOWER SYSTEM


                                 INLET GAS SO2 CONCENTRATION BETWEEN 2500 & 3500 ppm
                                          SYMBOL
SINGLE HOLD TANK

 RESIDENCE TIME


    20 minutes


    12 minutes


     6 minutes
                 B
                1.00        1.20        1.40         1.60        1.80        2.00

                      STOICHIOMETRIC RATIO, moles Ca added/mole SO2 absorbed
                                                                                   2.20
    Figure  5-5. The Effect of Stoichiometric Ratio and Effluent Residence

                  Time on Percent SO, Removal with a  Single Hold Tank
                                         185

-------
 100
  95 •
  90 •
  85
  80
cc


8'5
o
oc
  70
  65
  60
  55
  SO
VENTUR I/SPRAY TOWER SYSTEM


SINGLE HOLD TANK


12 MINUTES RESIDENCE TIME


INLET GAS SO2 CONCENTRATION

BETWEEN 2500 & 3500 ppm
                                              O
                                                                   O

                                                                 o o
•+•
                           4-
•4-
4-
    4.80       5.00       5.20        5.40       5.60        5.80


                               SCRUBBER INLET LIQUOR pH
                                                          6.00
                                                                    6.20
   Figure 5-6. The Effect of Scrubber Inlet  Liquor pH on Percent SO2

                Removal in the Venturi/Spray Tower System
                                       186

-------
  100
  95--
  90--
  86--
  80--
UJ

-------
fully tested by Borgwardt (Reference 3) in a 0. 1  Mw limestone scrub-
ber with both a plug flow reactor and with 3  stirred tanks in series
to approximate plug flow.  The concept has now been tested with lime-
stone in the TCA system at the Shawnee Test Facility (see Figure 1-4)
Major conditions for these tests were:
         TCA superficial gas velocity          12. 5 ft/sec
         Liquid-to-gas ratio                   50 gal/mcf
         Number of TCA beds       ^          3
         Static sphere height per bed           5 in.
         Percent solids recirulated             15
Hold tank configurations tested were:
         Single hold tank at 12 minutes residence time.

         Three hold tanks in series at 12 minutes total residence
         time (4. 3,  2. 2,  and  5. 5 minutes).

         Three hold tanks in series at 9 minutes total residence time
         (3. 8, 1. 9,  and 3. 3 minutes).
Results of the utilization testing for these 3 configurations are plotted

as stoichiometric ratio versus scrubber inlet liquor pH in Figures

5-8, 5-9, and 5-10, respectively.  Sight drawn averages from Figures
5-8 through 5-10 are compared in Figure 5-11.  For the runs with
3 tanks in series, there was no significant difference between 9 minutes
and 12 minutes residence time.  However, there was a distinct difference
 Initially 6-gram hollow TPR (thermoplastic rubber) spheres were
 used.  These were subsequently replaced with 6. 5-gram nitrile-PVC
 solid foam spheres.
                                188

-------
  1.80
  1.70 -•
  1.60 -
 i 1.50 - •
  1.40 - -
W "*"
2
]
01.20 • •
i
i
  1.10 • •
  •1.00 • •
TCA SYSTEM
SINGLE HOLD TANK

12 MINUTES RESIDENCE TIME
                                                        o_ 8
                                                   o    p° 1  o
                                                        00  0
                                                             o
                                     o
                          -f-
                         -H
                      +
•+-
    4.80
  5.00
5.20        5.40       5.60        5.80

      SCRUBBER INLET LIQUOR pH
          6.00
6.20
    Figure 5-8.  Stoichiometric Ratio versus Scrubber Inlet Liquor pH for
                 a Single Hold Tank at 12 minutes Residence Time
                                      189

-------
 1.80
  1.70 •
  1.60
cw

2
"5
  1.50
  1-40
21.30
tc
u
cc
Ul
g 1.20
X
o
o
fc
   1.10
   1.00
      4.80
TCA SYSTEM
THREE HOLD TANKS IN SERIES
12 MINUTES RESIDENCE TIME
   —I—
   5.00
                                                           +
_|	1	.—|	
 5.20       5.40        5.60        5.80
      SCRUBBER INLET LIQUOR pH
                                                                     6.00
                                                                                6.20
       Figure 5-9.  Stoichiometric Ratio versus Scrubber Inlet Liquor PH for
                     Three Hold Tanks in Series at 12 minutes Residence Time
                                         190

-------
  1.80
 1.70-
  1.60-
N
8
  1.50-
  1.40-
  1.20-
  1.10 • •
  1.00-
TCA SYSTEM

THREE HOLD TANKS IN SERIES

9 MINUTES RESIDENCE TIME
                                                                o
    4.80
  5.00
5.20        5.40       5.60

      SCRUBBER INLET LIQUOR pH
                                                         5.80
                                                        6.00
6.20
      Figure 5-10.  Stoichiometric Ratio versus Scrubber Inlet Liquor pH for
                    Three Hold Tanks  in Series at  9 minutes Residence  Time
                                     191

-------
 1.80
  1.70 •
  1.60
1 1-50
 N

S
   1.40
<3
 8
   1.30
 
-------
between operation with a single tank and with 3 tanks in series
at pH levels greater than about 5. 0.   For example,  at a scrubber

inlet liquor pH of 5. 6, stoichiometric ratio averaged 1. 19 with a single
hold tank as compared with 1.11 with 3 tanks in series, a 6 percent

improvement in limestone utilization. At higher pH, the improvement
was greater.  For instance at 5. 8 pH the improvement in utilization

was 14 percent.


The improvement in utilization -with  3 tanks in series can also be

seen in Figure 5-12 where SO_ removal is plotted versus stoichio-
                             Cj
metric ratio for an inlet gas SO  concentration range of 2500  to 3500
                               C*
ppm.  For example,  at 85 percent SO_  removal and 12 minutes total

residence time, the stoichiometric ratio -with 3 tanks in series averaged
about  15 percent lower than with a single tank.


Data for  9 minutes residence time with 3 tanks  in series were not

included  in Figure  5-12 because the TPR spheres in the TCA  were
replaced with nitrile-PVC solid foam spheres during the 9 minute

testing.  Unfortunately, the  effect of 9 minutes  residence time on S®i
removal was confounded with the effect of the new bed of spheres on

SO,, removal.


5.3      CONCLUSIONS


Limestone utilization in the  venturi/spray tower and TCA systems
normally  varied from about  60 percent at a scrubber inlet liquor pH
of 6. 0 to about 95  percent at a  scrubber inlet pH of 5. 2.  Operation at

reduced scrubber  liquor pH,  however,  inherently causes a  reduction

in SO- removal efficiency.
                                193

-------
 100
  95 -
  90 •
  85 •
                     3 TANKS
  80
i
Ul
E
 (M

8
IU
U
c
  70
  65
   60
   55
                                                         1TANK
                                                           TCA SYSTEM


                                                   12 MINUTES RESIDENCE TIME

                                                  INLET GAS S02 CONCENTRATION

                                                     BETWEEN 2500 & 3500 ppm
                                                  SYMBOL




                                                     D
                                                                  HOLD TANKS
   50
                1.0         1.2         1.4         1.6        1-8

                     STOICHIOMETRIC RATIO, moles Ca added/mole SO2ab*orb*d
                                                                      2.0
           Figure  5-12. The Effect of Stoichiometric Ratio and Hold

                         Tank Configuration on Percent SO2 Removal
                                                                                2.2
                                         194

-------
For the venturi/spray tower system with a single effluent hold tank,
limestone utilization was not affected by a change in residence time
from 20 to 12 minutes over the measured  range of scrubber inlet liquor
pH.  Limestone utilization declined, however,  at 6 minutes  residence
time for scrubber inlet liquor  pH !s greater than about 5. 6.

For the TCA system at 12 minutes total effluent  residence time and
at scrubber inlet liquor pH's greater than about 5.0,  higher limestone
utilization is achieved with 3 hold tanks in series than with a single
hold tank.
                                 195

-------
                             Section 6
                EFFECT OF ALKALI UTILIZATION
              ON MIST ELIMINATOR OPERABILITY
Data showing the effect of alkali utilization on mist eliminator oper-
ability are presented in this section.
6. 1      SUMMARY OF MIST ELIMINATOR OPERABILITY
         DURING LIME AND LIMESTONE RELIABILITY TESTING
During lime testing on the venturi/spray tower system,  reliable
operation was achieved for the 3-pass,  open-vane, 316 stainless  steel
chevron mist eliminator  (see Section 3).  These tests were conducted
at spray tower gas velocities up to 9.4 ft/sec (maximum attainable)
and recirculated slurry solids contents  up to 15 percent.  The lime
utilization for these tests was about  90 percent.  Both the  topside
and the bottomside of the mist eliminator were washed intermittently
with fresh water.  Only about one-half of the makeup water available dur-
ing closed liquor loop operation was required for mist eliminator wash.

During limestone testing on the TCA  system with the wash tray and
6-pass, closed-vane, chevron mist eliminator in series (see Section 4),
reliable operation of the mist elimination system was achieved only
at 8. 6 ft/sec scrubber gas  velocity (5. 6 ft/sec superficial gas velocity
in the wash tray and mist eliminator  areas).  The limestone utilization
ranged from 60 to 75 percent.
                                196

-------
In view of the success of the spray tower mist eliminator operation


with lime,  a. new mist eliminator of similar design (single stage,


316 stainless steel, 3-pass,  open-vane, chevron type) was installed


in the TCA  in July 1975 for testing with limestone slurry.  However,


poor operability with this mist eliminator was experienced with  inter-


mittent topside  and bottomside wash with raw water.  For example,


at a scrubber gas velocity of 9.4 ft/sec (6. 1  ft/sec superficial gas


velocity in the mist eliminator area),  the  chevron mist eliminator


became about 50 percent restricted with soft solids after only 60 hours


of operation (Run 554-2A).  The limestone utilization was about  63


percent at the scrubber inlet liquor pH of about  6. 0.  This failure


in the TCA,  as  compared with the  successful operation in the spray


tower with lime using a similar mist eliminator design and washing


scheme, was initially attributed to the differences in the physical design


and the pattern  of mist generation within the  two scrubbers.




During  subsequent testing  of the TCA  system at scrubber inlet pH


from 5. 7 to 6. 0 (limestone utilizations from  58 to 74 percent),  stable


conditions with  less than 10 percent solids restricted were achieved


for the  mist elimination system by continuous washing of the under-


side of the chevron with diluted clarified liquor.   For these  low


utilizations, only the mist eliminator  blades  continuously contacted by


•wash water could be maintained free of solids restriction.  It was


necessary, therefore,  to  eliminate shadowing by mist eliminator


supports and to avoid maldistribution  of gas  caused by observation


ports and corrosion coupon racks.  In Run 559-2A, the chevron  mist


eliminator was  only 7 percent  restricted with soft solids after 384


operating hours using continuous underwash  with diluted clarfied liquor

             2
at 0.4 gpm/ft   and intermittent topside wash with raw water.
                                197

-------
6. 2      MIST ELIMINATOR OPERABILITY DURING LIMESTONE
         UTILIZATION TESTING
In October 1975, limestone utilization testing was  started on both the
TCA and venturi/spray tower systems (see Section 5), concurrent
with the on-going limestone mist eliminator testing in the TCA.   Tables
6-1 and 6-2 summarize the  results of mist eliminator operability during
these tests through January 1976.  These tables list the average  scrub-
ber inlet pH,  average limestone utilization and stoichiometric ratio,
mist eliminator wash scheme,  and percent of the mist eliminator
passage restricted by solids deposit at the end of each run.   All of
the available makeup water was used for tests with continuous mist
eliminator bottom wash,  while only about one-half of the available
makeup water was  used for tests with intermittent bottom wash.

During venturi/spray tower limestone Runs 701-1A and  702-1A (see
Table  6-1) with intermittent topside and bottomside raw water wash,  the
mist eliminator was heavily restricted by  soft solids within 2 to 3 days.
This was unexpected since earlier  operation wifh lime slurry under
identical operating conditions was successful.   The limestone utiliza-
tion for these two runs was only 69 percent, as compared to 90 percent
normally obtained with lime operation.  Subsequently, during Run
703-1A, the limestone utilization was increased to 93 percent by  drop-
ping the average scrubber inlet pH to 5. 2.   The mist eliminator was
found to be essentially clean  (1 percent restricted) after  319 operating
hours with intermittent underside and topside raw water wash.  The
average SO_  removal for Run 703-1A was only 58 percent,  as com-
pared with 87 percent for the previous tests.
                                 198

-------
                                                                   Table 6-1

                                 SUMMARY  OF VENTURI/SPRAY  TOWER  LIMESTONE
                                      UTILIZATION AND MIST  ELIMINATOR TESTS
                                              All rung made with a 316 stainless steel, 3-pasa, open-vane,
                                              chevron mist eliminator.  Run conditions: 1 5 wt % solids in
                                              recirculated slurry,  9.4 ft/sec spray tower superficial gas
                                              velocity, 21  gal/mcf liquid-to-gas  ratio in venturi, 50-57
                                              gal/mcf liquid-to-gas ratio in spray tower.
Run
No.
701-1A
702-1A
703-1A
704- 1A
705- 1A
706- 1A
707-1A
708-1A
709-1A
710-1A
711-1A
711-1B
712-1A
712-1B
713-1A
Single Tank
Res. Time,
min
20
20
20
20
20
12
12
12
12
12
6
6
6
6
6
A vg.
Scrubber
Inlet pH
5.9
5.8
5.2
5.8
5. 7
5. 2
5. 7
5. 6
5.9
6. 0
5. 7
5. 6
5-8
Depletion
5.2
A vg.
Stoich.
Ratio
1.45
1.45
1.07
1.45
1.25
1. 06
1. 30
1. 20
1. 35
1. 50
1. 30
1.40
1. 50
-
1. 10
Avg. Percent
Limestone
Utilization
69
69
93
69
80
94
77
83
74
67
77
71
67
-
91
Avg. Percent
SO2 Removal(g)
88
87
58
86
84
58
83
77
83
91
81
82
87
-
69
Mist Eliminator Run
Wash Scheme Hours
Top | Bottom
Intermittent Intermittent 73



\

60


'

319
66
136
180
Intermittent10' Intermittent!^' 118






1

Intermittent 138
Continuous 134




i

234
144
71
119
( 18
52
Hours Since
Cleaning Mist
Eliminator
73
60
319
385
136
180
298
436
134
368
512
583
702
720
772
Percent Mist
Eliminator
Restriction
50-60
45-50
1
45-50
17-20
1
10-15
15-20
<1
5-7
_
5-7
10-15
-
10
(a)  Intermittent, sequential top wash with makeup water at 0. 53 gpm/sqft for 4 min/8 hr/section.
(b)
(c)
    Intermittent, full face bottom wash with makeup water at 1. 5 gpm/sqft for 6 min/4 hr.
    Intermittent, sequential top wash with makeup water at 0. 53 gpm/sqft for 3 min/hr /section.
(d)  Intermittent, full face bottom wash with makeup water at 1. 5 gpm/sqft for 4 min/hr.
(e)  Continuous, full face bottom wash with diluted clarified liquor at 0.4 gprr./sqft.
(f)
    A limestone depletion run is conducted without limestone addition during SC>2 absorption.
    The scrubber inlet liquor pH is allowed to drop from about 5. 9 to 4. 8.
(g)  SC>2 removals are for 2500 to 3500 ppm inlet gas SC>2 concentration.

-------
                                                                                Table  6-2

                                              SUMMARY OF TCA LIMESTONE  UTILIZATION AND
                                                                  MIST  ELIMINATOR  TESTS
                                                  All rum rmde with a 316 itainlen steel, 3-p«n, open-vane,  chevron mitt eliminator.
                                                  Run condition!: 15 wt % solids in reclrculated ilurry,  12. 5 ft/sec TCA superficial gat
                                                  velocity, 50 gal/mcf liquld-to-gai ratio. TCA configuration:  3 beds with 5 tnche« per
                                                  bed of 6-gram TPR  spheres for Runs 562-2A through 569-2A and 6. 5-gram nitrile foam
                                                  spheres for Rani 569-2B through 582-2A.
Run Nc
No. Tan
562-2A


562-2B
563-2A
564-2A
Total
ki Res. Time,
mln
12


12
12
12
565-2A 3 12
566-2A 3 12
567-2A 3 12
568-2A 3 12
569-2A 3 9
569-2B
J 9
570-2A 3 9
571-2A 3 9
571-2B 3 9
572-2A
! 9
573-2A 3 12
573-2B 3 12
574-2A 3 9
575-2A
12
576-2A 3 12
576-2B
( 12
577-2A 3 12
578-2A 3 12
579-2A 3 9
580-2A 3 9
581-2A
582-2A
12
12
Avg. A
Scrubber St
Inlet pH R
5.9 1


5. 7 I
5.9 1
5.2 1
5.25 1
5. 9 1
6.0 1
5. 5 1
5. 5 1
5. 5 1
5. 75 1
5.8 1
Depletion'
5.25 1
5. 5 1
Depiction
Depletion
5.55 1
5. 7 1
Depletion
5.8 1
Depletion
5.25 1
Depletion
5. 5 1
Depletion
vg. Avg. Percent
olch. Limestone
atlo Utilization
.6 63


.4 71
. 7 59
.06 94
. 05 95
.25 80
.4 71
.06 94
. 08 93
.08 93
. 18 85
.25 80

.09 92
. 08 93
-
-
. 15 87
. 17 85
-
.25 80
-
.05 95
-
.15 87
-
Avg. Percent Milt Eliminator Run
SO, Removal'*' Wa.h Scheme Hours
Top ] Bottom
83 Intermittent'*1 Continuou«'b> 495


81
86
58
58
83
84
63
69
64
72
73
-
57









134
182
113
109
166
138
Intermittent ' 162






66 No wash
-
-
69
73





/. i
83 Intermittent'"
-
61
-
80
-





66
97
144
96
11
154
45
28
12
47
112
3
159
9
62
12
164
18
Houra Since
Cleaning Mist
Eliminator
155
332
495
629
811
924
109
275
413
575
641
738
882
978
989
1143
1188
1216
1228
47
159
162
159
168
230
242
406
424
Percent Mitt
Eliminator
Restriction
Z
5-7
7-9
7
7
3
0
0
2
1
<1
<1
<1
<1
-
-
<1
-
40
-
-
3
20
-
-
10
-
2
K)
O
o
           (a)    Intermittent, sequential top wain with makeup water at 0. 55-0. 83 gpm/sq ft for 3 min/hr/section.
           (b)    Continuous, full face bottom wash with duluted clarified liquor at 0.4 gpm/sq ft.
           (c)    Intermittent, full face bottom wash with makeup water at  1. 5 gpm/sq ft for 4 min/hr.
           (d)    A HmeHtone depletion run ia conducted without limestone addition during SOp absorption.  The
                 scrubber inlet liquor pH ia allowed to drop from about S. 9 to 4, 8,
           (e)    SO_ removals are for 2500 to 3500 ppm inlet gaa SO^ concentration.

-------
Subsequent venturi/spray tower Runs 704-1A through 708-1A confirmed



the observation that reliable mist elimination operation could only be



obtained at high alkali utilization (greater than about 90 percent), with



intermittent underside and topside wash using raw water.   Runs 709-1A



through 711-IB also showed that for utilization less than about  80



percent,  continuous bottomside wash with diluted clarified liquor could



reduce or limit the amount of soft  solids deposition on  the mist eliminator




vanes.







Testing with the TCA  system similarly confirmed the  strong effect



of limestone utilization on mist eliminator reliability.   Runs 562-2A,



562-2B and 563-2A  (see Table 6-2) were conducted at  average  scrubber



inlet pH's of 5. 7 to  5. 9 with limestone utilizations from 59 to 71



percent.  The mist  eliminator was washed intermittently with  fresh



water on the topside,  and continuously with diluted  clarified  liquor



on the bottomside.  The mist eliminator restriction increased  to 7-9



percent during the first 500 hours of operation and  appeared to level



out at 7 percent restriction after 811 hours at the end  of Run 563-2A.



Following these tests,  the  scrubber inlet pH was dropped to  5. 2



(Run  564-2A) and the operation continued for additional 113 hours.  At



the end of Run  564-2A the mist eliminator restriction  decreased to



 3 percent from the 7 percent at the  start of the run. The limestone



\rtilization during Run 564-2A averaged 94 percent.







As was discussed in Section  5, operation at reduced scrubber  liquor



pH and high utilization inherently  causes a  reduction in SC>2  removal



 efficiency.  This  reduction in efficiency can be compensated by:



 (1) increasing the slurry flow to the absorber,  (2) increasing the TCA
                                201

-------
packing height and gas phase pressure drop, or (3) adding MgO to the



scrubber slurry.   Another way of increasing SC>2  removal efficiency



while maintaining high limestone utilization  is to operate with three



hold tanks in series.  TVA is presently conducting an economic evalua-



tion of the proposed schemes for improving SC>2 removal efficiency



while maintaining  high limestone utilization.








 TCA Runs 565-2A through 580-2A were operated  with three hold tanks



 in series (see Table  6-2).  As with the venturi/spray tower system,



 continuous bottomside wash with diluted clarified liquor and intermittent



 topside wash with raw water limited the amount of soft solids deposition



 at low utilization  (Run 567-2A). Also, the mist eliminator was kept



 completely free of solids  with continuous bottomside wash at high



 utilization (Runs 565-2A and 566-2A).  As expected, with intermit-



 tent bottomside and topside wash at high alkali utilization (Runs  568-2A



 through 573-2A),the  chevron mist eliminator remained essentially free



 of solids.   After a total of 1188 operating hours with continuous  and



 intermittent bottomside wash and  intermittent topside  wash, the chevron



 mist eliminator in the TCA was less than one percent restricted with




 soft solids (see Run 573-2A in Table 6-2).







 A further example of a mist eliminator which has become less restricted



 during operation  at high utilization can be seen in Runs 577-2A through



 582-2A.  At the conclusion of Run 577-2A the mist eliminator was 20



 percent restricted by soft solids and after  an additional  226 hours of



 operation at 95 and 87 percent utilization (Runs 579-2A and 581-2A),



 the mist eliminator  restriction decreased to 2 percent.
                                  202

-------
6.3      CONCLUSIONS


The following general conclusions can be drawn for the performances

of the  single stage, 316 stainless steel,  3-pass,  open-vane, chevron
mist eliminator in both the TCA and venturi/spray tower systems

with lime and limestone:


    •    The reliability of the mist elimination system is a strong
        function of alkali utilization.

    •    For high alkali utilization (greater than about  85 percent),
        the mist eliminator  can be kept free of solids  deposits with
        intermittent fresh water top wash combined with either
        intermittent fresh water bottom wash or continuous bottom
        wash with diluted clarified liquor.  Intermittent top and
        bottomside fresh water  wash may be  required in closed liquor
        loop operation due to restrictions in the allowable raw water
        makeup to the scrubber system.

    •   For alkali utilization less than about  85 percent,  intermittent
        top and bottomside wash with fresh water does not limit solids
        accumulation.  However,  for these conditions, a continuous
        bottom wash with diluted clarified  liquor used in combina-
        tion with an intermittent topside  fresh water wash can limit
         soft solids buildup to less than 10 percent restriction within
        the  mist eliminator.

    •  Operation for a period  of time at high alkali utilization (greater
         than about 90  percent)  may,  in certain instances,  clean up
         an already fouled mist eliminator.


 The effect of high alkali utilization on decreasing soft  solids retention

 within a mist eliminator should have  direct application to commercial

 scrubber design and operation.   It may be possible to  change operating

 conditions in existing scrubbers to increase limestone utilization.  High

 utilization can be designed into new scrubber  systems.
                                 203

-------
                            Section 7
                          REFERENCES

1.   Bechtel Corporation, EPA Alkali Scrubbing Test Facility;  Summary
    of Testing through October 1974,  EPA Report 650/2-75-047,
    June 1975.
2.   Bechtel Corporation, EPA Alkali Scrubbing Test Facility:  Advanced
    Program-First Progress Report, EPA Report 600/2-75-050,
    September 1975.
3.   R.  H.  Borgwardt, "Increasing Limestone Utilization in FGD
    Scrubbers, " presented at the Sixty-Eighth Annual Meeting of
    the AIChE, Los Angeles, November 16-20,  1975.
4.   R.  H.  Borgwardt, "IERL-RTP Scrubber Studies Related to Forced
    Oxidation of Sludge,  " presented at the  IERL Symposium on Flue Gas
    Desulfurization,  New Orleans, March  8-11, 1976.
                               204

-------
                      DUQUESNE LIGHT COMPANY
                ELRAMA AND PHILLIPS POWER STATIONS
                     LIME SCRUBBING FACILITIES
       R.  Gordon Knight,  Superintendent,  Technical Services
       Steve L.  Pernick,  Jr., Manager,  Environmental Affairs

                      Duquesne Light Company
                     Pittsburgh, Pennsylvania
ABSTRACT

     The Elrama and Phillips Power Stations are coal burning facilities
having net generating capabilities of 494 Mw and 387 Mw, respectively.
Both plants are located in the southwest portion of Pennsylvania within
a radius of approximately 15 miles of the city of Pittsburgh.

     Venturi scrubbing systems using lime were installed at both
facilities for dust removal, with one of the scrubbers at the Phillips
facility equipped with a dual stage venturi scrubber to be used and
tested as the prototype for sulfur dioxide removal.  Startup of a
portion of the Phillips scrubber system began July, 1973.  This system
is designed to scrub the headered gases from all six boilers of the
plant, and discharge them to a new 340 ft. stack after reheating them
from 120° to approximately 140  with direct fired No. 2 oil.  The
scrubber system consists of four trains.  The first train is the prototype
for sulfur dioxide removal and consists of two identical venturi
scrubbers in series.  Each of the remaining three trains consist of
a single venturi scrubber.  A series of tests were conducted at this
facility to determine the most feasible means of complying with the
SO  emission limitations.

     Early operation of the Phillips scrubber system revealed several
problems, such as equipment erosion and corrosion, acid condensate
leakage in the new stack, and stress problems in the ID fans, which
caused outages of the scrubber system.  After an extended outage,
the scrubber system resumed partial operation in March, 1974, and has
been in continuous operation since that time with a varying number
of boilers connected to the scrubber system and with a varying number
of scrubber trains in service.  For approximately a five month period
in 1975, all six boilers were connected to the scrubber system.  During
this period, the scrubber availability was approximately 67%.

     Our operating experiences have revealed many problems, such as
scaling and deposits in the scrubber, ID fan deposits and high stresses,

                                 205

-------
a crack in an ID fan shroud, recycle pump erosion/corrosion, high water
inventory, sludge disposal problems, and lime addition problems.

     In August, 1975, the scrubber problems were adversely affecting
the plant generating capability to the point where it became necessary
to remove the largest boiler, No. 6 Boiler, from the scrubber system
and return it to its original gas path.  Operation has continued with
this arrangement until  the present time.

     Because of the  serious  nature of the problems with the Phillips
scrubber  system, startup of  the Elrama scrubber system was delayed
to permit study and  resolution of the problems at Phillips.  The Elrama
scrubber  system, consisting  of five single stage scrubber trains serving
four boilers, was  placed in  partial operation with one boiler connected
to  it  on  October 26,  1975»   A second boiler was connected on February  4,
 1976,  and operation  has continued in that manner to the present time.

      Tests conducted on the  Phillips scrubber system indicate SO^
 removal efficiencies using high calcium  lime of approximately 90-6 with
 dual  stage scrubbing and 50% with single stage scrubbing.  Tests using
 a magnesium modified lime (magnesium oxide of 8-10%) indicate removal
 efficiencies of approximately 83% with single stage scrubbing.  The
 results of the magnesium modified lime tests indicate an ability to
 comply with the S02  emission limitations with single stage scrubbing.

      More than $66 million has been spent to date on both scrubber
 systems with an additional  $20 million to be spent within the next
 several years to complete the installations for removing sufficient
 SO  to comply with the regulations.  Operating costs of the Phillips
 Station have increased approximately 35% due to the operation of these
 scrubbers.

      Although many problems  have been  and are being resolved, others
 still remain,  and  consequently,  an  assessment of the entire system
 operation, reliability, and practicability  is inconclusive at this  time.
                               206

-------
                   DUQUESNE LIGHT COMPANY
             ELRAMA AND PHILLIPS POWER  STATIONS
             	LIME SCRUBBING FACILITIES
 INTRODUCTION
          Duquesne Light Company, an  investor-owned  electric
 utility, serves about one-half million  customers  in  Southwestern
 Pennsylvania and has a net generating capability  of  approxi-
 mately 2500 Mw.  This capability is generated by  combustion tur-
 bines in simple and combined cycle modes, by nuclear plants,  and
 by coal fired power stations.  The Company is sole owner  and
 operator of three coal fired stations,  two of which  have  been
 retrofitted with wet scrubbers using  lime as a reagent.

          The history, description, and early operating experi-
 ences of the scrubbers at our Phillips  Power Station were de-
 scribed in a paper at the EPA Flue Gas  Desulfurization Symposium
 in November, 1974.  A second paper presented to APCA in June,
 1975 described the Phillips operating experiences during  the
 period March, 1974 through April, 1975.  This paper  will  update
 the Phillips experiences, giving results of tests that have been
 and are being conducted,  and it will describe the startup and
 operating experiences of  the Elrama Power Station scrubbing sys-
 tem for the period extending through January, 1976.

 DESCRIPTION OF SYSTEMS

 Phillips Power Station

          Four trains of  wet venturi-type scrubbers  have  been
 installed at the 387 Mw Phillips Power Station at a  cost  of ap-
 proximately $103 per Kw.   Gibbs & Hill, inc. was engaged  as
 architect-engineer for the entire project, with battery limits
 of the Chemico Corporation confined to the scrubbers and  asso-
 ciated pumps and controls between the inlet hot gas  duct  mani-
 fold and the exit wet gas header, including the reheater  but
 excluding the new induced draft fans.

          The four trains are located downstream of  existing  in
 series mechanical collectors and electrical precipitators in-
 stalled on each of the six pulverized coal fired boilers.  Three
 of the trains are single  stage venturi scrubbers originally in-
 tended for particulate removal.   The fourth train is dual stage
 and was the prototype for determining the feasibility of  two
 stage scrubbing for compliance with S02 emission limits.   A test
program on the prototype  and the single stage scrubbers was used
to determine what additional equipment or process will be neces-
 sary to enable us to meet regulatory requirements for S02 re-
noval,  which are essentially 83% S02 removal efficiency using 2%
sulfur coal.

-------
          Each train is equipped with a new wet-type ID fan.   A
new common duct directs emissions from all boilers to the scrub-
ber plant where they can be sent to any or all of the trains.
The scrubbed gas is exhausted through a common duct and an oil
fired reheater to a new ground supported stack which consists
of a concrete envelope, a 30 inch annulus, and an inner acid-
resistant brick stack.

          Slaked quicklime is added to the lower cone of each  of
the scrubber vessels to neutralize the recirculating liquor,
which, in single stage scrubbing with high calcium lime, reacts
with about 50% of the S02 in the flue gas.  A liquor bleed flow
of about 4% is sent to one or both of two thickeners for solids
removal.  The overflow is returned to the system, and the under-
flow is piped to the sludge treatment system.

          Startup of a portion of the Phillips scrubber system
began July, 1973.  Several problems then developed causing out-
ages of the scrubber system, and after an extended outage, the
scrubber system was returned to service in March, 1974.  The
system has been operating continuously since that time with a
varying number of boilers connected to the scrubber system, and
with a varying number of scrubber trains in service.

Elrama Power Station

          The scrubber facility at the 494 Mw Elrama Station is
identical to that at Phillips in most ways.  Gibbs & Hill is the
architect-engineer, Chemico scrubbers have been installed, and
the battery limits are the same.  Mechanical and electrical dust
removal equipment remove most particulates from the boiler emis-
sions, and the gas to and from the scrubbers is headered in the
same way.  Unlike Phillips, there are four boilers, each with
its own turbine generator.  Five single stage scrubbers were in-
stalled, and it was intended that the knowledge gained from the
test program at Phillips would be applied to Elrama to enable
compliance with the regulations.

          The first Elrama scrubber was not placed in service
until October 26, 1975.  It had been scheduled for an earlier
startup date; however, the severity and number of problems en-
countered at Phillips made it prudent to delay startup until
many of the problems at Phillips were resolved and the necessary
modifications made at both stations.  The decision to delay the
Elrama startup has proved to be a wise one.

PHILLIPS OPERATING EXPERIENCE

          A number of problem areas were described in our previ-
ous papers.  Some of these have been alleviated, and others
still persist.  The present status of a number of these  is de-
scribed below.


                            208

-------
ID Fans

          Encouraging progress has been made in this area.  The
second repair  of  the original defective welds with Inconel 112
has proved successful and resulted in very few minor or no weld
repairs  at each 1500 hour inspection.  Relocation of spray noz-
zles with the  use of a new type (Bete fog nozzle No. TF16FC) has
practically eliminated the deposit accumulations, which caused
fan imbalance  and under which pitting attack occurred.

          A problem still exists with pitting of the 316 SS
sleeves  on the fan shafts.  Recent trial use of a coating of
Neoprene-29 appears encouraging.  The rubber lining on the hous-
ings needs periodic minor repairs.

          During  a routine inspection of a fan in the last week
of January, a  crack was found in one of the shrouds.  The crack
extended about 4  inches from the outer periphery and was about
1/16 of  an inch wide at that point.  Both shrouds will be
checked  by non-destructive methods.  The defective area will be
removed  for examination.  Repairs will be made, and the fan re-
turned to service.  This fan has experienced about 11,000 serv-
ice hours.  The other fans will be inspected and checked by non-
destructive methods.

Recycle  Pumps

          The  corrosion/erosion problem with these pumps con-
tinues.   As indicated in our previous paper, we are following a
trial program  involving a number of different impeller and wear
ring materials.  Although we have still not obtained the desired
life from any  of  the materials we have tried, there are indica-
tions that a CD4MCu impeller with a 26% chrome-iron wear ring is
the best combination.  Service life with these two materials has
been up  to 4700 hours, compared to the 3000 hours with the or-
iginal Carpenter  20.  Two rubber lined pumps are on hand for
trial installation at the Elrama Station.

Bleed Valves

          Previous problems with rubber lined plug-type bleed
valves have apparently been solved by the substitution of pinch-
type valves.

Deposits

          Experience has shown that complete cleaning of a
scrubber vessel is required about every 1400 service hours.
Each cleaning with minor maintenance requires 1400-1700 man
hours.   Approximately 20-80 tons of deposits are removed during
each cleaning.  These deposits result even though  the pH of  the
recycle liquor is maintained  in the  6-7 range for about 95% of
the time.  We do not yet have redundant lime feed or automatic
pH control.  These are to be  forthcoming.
                             209

-------
           The main troublespot for deposits is in and around the
 throat  dampers.  Despite programmed, automatic exercising of the
 dampers/  deposits accumulate to the extent that the throat pres-
 sure drop increases from a normal 6 inches to 12 inches or more
 after about  1400 hours of service.  This high delta p decreases
 the amount of gas that can be scrubbed to such an extent that
 generating capability of the power plant is reduced and a clean-
 ing outage is necessary.  The scrubber manufacturer had in-
 stalled and  recommended that we keep open a 1-1/2 inch valved
 vent on each side of the damper housings in an effort to sweep
 the areas clear of deposit with outside air.  This had no ef-
 fect.   To obtain more air sweep, we next removed eleven 3/4 inch
 plugs from openings on the sides and tops of four of the twelve
 damper  housings on one of the vessels.  This, too, had no ef-
 fect.   Earlier we had introduced a continuous flow of water
 through the  same 3/4 inch openings with little or no benefit.

           Deposits accumulate in and around the twelve tangen-
 tial nozzles in each vessel.  This results in an uneven flow of
 scrubbing liquor around the upper cone.  Chemico recommended
 that alternate nozzles in one of the vessels be blanked off to
 see if  the increased velocity through the others would prevent
 the accumulations.  After 1506 service hours, this appears to
 be partially effective in the vessel where it was tried.

           Sizable deposits form in the lower cone, where the
 lime discharges into the vessel through an open end pipe.
 Chemico recommended the use of an angled nozzle at this loca-
 tion.   This  is currently being implemented.

           Appreciable deposits are also found on the lip under
 the upper cone.  Chemico is conducting scale model studies in
 an effort to determine the cause and possible corrective meas-
 ures.

           Comments on all these deposit problems should be con-
 sidered in conjunction with the results obtained in our magnes-
 ium modified lime tests.  These will be discussed later.

 PHILLIPS  SLUDGE TREATMENT AND DISPOSAL

           Previous reports have discussed our first-of-a-kind
 experience with Dravo's Calcilox as a fixating agent.   Early un-
 successful use resulted from our inability to maintain one or a
 combination of (1)  the required 10% addition (by dry weight of
 sludge),  (2)  warm ambient temperature, (3)  necessary curing
 time, and  (4) pH of 10-11.   None of these would be critical if
we could  add the fixative and then let the mixture remain in
place.   However, we have only three "curing" ponds,  each of
 about 6000 yd3 capacity, and we fill each one in 10-14 days.
Furthermore,  to be prepared for unexpected emergencies,  such as
 the draining of a scrubber or a thickener,  we must have an empty
or near empty pond at all times.   Consequently,  we must excavate
                            210

-------
 ponds within  about  4 days  of filling which precludes the neces-
 sary curing time.   The  problems involved with hauling "Lup" in
 open dump trucks are obvious.   When requirements are satisfied,
 y^r^aiV£;e 11S5* fixation with Calcilox.  The original system
 introduced the additive in slurry form.   In April, 1975, we in-
 nravi   ?h*ry add*tlve  system,  fabricated by and rented from
 wSh ™«  J   WaS lnt
-------
PHILLIPS SCRUBBER SYSTEM AVAILABILITY

          Reports by others on availability of the Phillips
scrubbers have been hopelessly confused and unduly optimistic
because of the header design of  the plant.  Until all boilers
were connected to the scrubbers,  there was at least one spare
train and as much as 100%  spare  scrubber capacity.  Therefore,
having a train out of service for maintenance did not reduce
the capability of the scrubber plant.  Until all six boilers
were connected,  meaningful availability factors were not avail-
able.  On March  17,  1975,  the last boiler was connected, and
all  four  scrubber trains were required to be in service.  Opera-
tion  in that mode continued until August 4, 1975 when it was
necessary to remove  the No. 6 Boiler  from the scrubber system
because the pH in the  system could not be maintained, and depos-
 its  were  becoming unmanageable to the point where scrubber out-
 ages were reducing  the  station generating capability.  In the
 four and  a half month period of  total scrubbing, availability
 averaged  67%.

 PHILLIPS  SCRUBBER SYSTEM TESTS

           In addition  to  our continuing performance and sludge
 tests, several other tests are noteworthy.

 Lamella Thickener

           Because of the space  limitations  at both stations  and
 because of the success of the  phosphate  industry in using the
 Lamella plate-type thickener,  we obtained and operated a 100 gpm
 model at Phillips for four weeks. A polymer was added to the
 inlet stream from the scrubber  bleed.   The  influent contained
 2.5% to 3.0% solids, and the  Lamella increased  the concentration
 to the range of 17-20%.  We require  a 30-40% concentration.   The
 overflow quality was excellent - less than  150  ppm.   The manu-
 facturer said that use of a larger bottom hopper would have
 given a greater concentration of solids in  the  underflow.   No
 good evaluation of long term deposit accumulation  and ease  of
 removal can be given.   The Lamella plates were  disassembled
 after 17 days of continuous operation,  and  soft deposits of  1/4
 to 1 inch in thickness were found near the  bottom  of  the plates. .
 These were easily removed from the plates,  after removal from
 the housing, by medium pressure jets.  Removal  of  deposit  from
 in-place plates cannot be evaluated.

 Additives

           We have been plagued with black plumes  from the  scrub-
 ber  stack at times of boiler upsets.  Although we  were  originally
 told that the scrubbers would remove unburned carbon, this has
 not  been  the case.   In an effort to clean up the  plume,  we tried
 four different wetting agents,  all from the same  supplier.   None
had  any effect.


                              212

-------
Magnesium Modified Lime

          We  ran a two and a half month test with magnesium
modified  lime in order to determine (1) if single stage scrub-
bing could remove sufficient SC>2 to enable compliance with
regulations,  (2)  the effect on deposits, and (3) the change in
sludge characteristics.  Because of our limited lime capacity,
it was necessary to reduce the quantity of emissions treated in
the scrubbers.   We did this by re-routing the emissions from a
second boiler back to the original stack - this boiler was later
taken out of  service for major maintenance - and by limiting op-
eration on a  third to supplying emergency generation require-
ments.  In order to slake the blended lime completely, a source
of hot water  was obtained so that slaking temperature was held
in the 180-190°F range.  Efforts were made to reduce the addi-
tion of fresh water in order to maintain system chemistry.  An
effective step  was the use of thickener overflow return water
to dilute the lime slurry after slaking.  A not-so-effective
step was  the  reduction of demister spray time.  Deposits re-
sulted until  normal spray time was restored.

          System requirements limited the duration of the test
so that the optimum value of some operating parameters must be
determined at a later time.  However, the test did yield some
valuable  information.

          It  was determined that the required degree of S02 re-
moval - approximately 83% - can be obtained with an MgO content
of 8-10%  in the lime with single stage scrubbing.  This will
eliminate the necessity to install any additional stages or
scrubber  vessels at both the Phillips and Elrama Stations.  This
will save approximately $10 million in capital costs.

          The two and a half month test with modified lime re-
sulted in 1612  service hours on one train and 1309 on another.
After this length of service, the throat damper pressure drop
in both trains  was still 6 inches, the same as with newly
cleaned dampers.   With the normal high calcium lime, the pres-
sure drop would have been at or approaching 12 inches, and an
outage for cleaning would have been necessary.  Inspection
showed negligible deposit accumulation in all accessible loca-
tions, and the  deposit on some previously scaled surfaces had
been removed.   For example, the maximum flow that could be ob-
tained through  the bleed line on one scrubber prior to the test
was 420 gpm.   This flow increased during the test to a value of
720 gpm at termination.  On the basis of SO2 removal efficiency
and freedom from deposits, it is unfortunate that we cannot con-
tinue the use of the modified lime.  As discussed later, this
requires  major  modifications, about 18-24 months for construc-
tion and  the  expenditure of about $20 million to convert both
the Phillips  and Elrama scrubbers.

          One of the less favorable effects of using magnesium
Modified  lime was on the sludge, which did not settle as well in

                               213

-------
 the thickeners.  The solids concentration in the underflow de-
 creased from 35-45% to 25-30%.  Although a polymer was tried,
 time did not permit a complete evaluation of its effectiveness.
 The nature of the solids is important from two considerations:
 effect on sludge treatment schemes and on the size or number of
 additional thickeners needed for the next phase of construction.
 Steps were taken to supply sludge samples to the thickener
 builders and to the three bidders for our long term waste man-
 agement contract.  The results of the thickener tests have not
 yet been submitted to us.  All three of the waste processors
 said that the sludge could be handled, with slight modification
 needed in some cases.

           Another unexpected effect of our use of magnesium
 modified lime was a reduction in the pH of the sludge (8-9 in-
 stead of 10-11}.  This evidently resulted from better lime slak-
 ing, resulting in less unreacted lime in the sludge, and from
 higher utilization of lime by better S(>2 removal.  The effect on
 sludge fixation was to leave us with soupy ponds to excavate and
 haul to the final disposal site.  Dravo recommended the use of
 15%  (instead of 10%) Calcilox addition, and they added an "alka-
 line booster" to the Calcilox.  We were not able to get the de-
 sired pH before the test ended, and, to make things worse, the
 increased SC>2 removal and consequent sludge filled the ponds
 faster and reduced the curing time.  Dravo's representatives in-
 dicated that laboratory tests showed good fixation with the
 proper pH and curing time.

 ELRAMA SCRUBBER STARTUP AND OPERATING EXPERIENCE

           As previously noted, the first boiler was connected to
 the scrubbers on October 26,  1975.   This boiler,  No. 2,  has an
 equivalent capacity of about 100 Mw, and the emissions can be
 handled by one scrubber.   However,  to insure reliability in case
 of a scrubber .malfunction, two scrubbers are operated at partial
 load to protect the boiler and its  associated turbine generator
 against a trip-off.  We have  had continuous operation of the
 boiler on the scrubber system through January,  1976 with the
 exception of two  short outages.   One was occasioned by inopera-
 tive throat dampers,  originally thought to be a result of depos-
 it buildup.   Upon inspection,  however,  the cause  was found to be
 damper scraper  blades which had been installed  with insufficient
 clearance.   The other involved failure  of the lime  feeder belt
 in conjunction  with a boiler  outage.  Four of the scrubber ves-
 sels have  been  in service  in  various combinations,  and the serv-
 ice  hours  to January  31, 1976  are:

                      No. 1 -  1169 hours
                      No. 2 -  1508 hours
                      No. 3 -   976 hours
                      No. 4 -   838 hours

No.  5 has not been  in  service  because it is  being revised for
the trial installation of  rubber lined recycle  pumps.

                              214

-------
          To this time, there have been no major problems.
There have been many startup pains, not the least of which has
been frozen pipes.  Heat tracing has been and continues to b,i
a problem at the Phillips Station, and it appears we have more
of the same at Elrama.  The most persistent problem, and one
that could take the scrubber system and generating facilities
out of service, has been experienced with the thickeners.  This
is a hardware and design problem associated with recirculation
of the sludge within the thickeners to attain the 30-40% solids
concentration.  Fortunately, station personnel have recently
found that our sludge treatment process (described below) can
handle sludge with solids concentration as low as 15%; thus.
recirculation is unnecessary.  Each of the two thickeners has
sufficient capacity for one boiler.  Therefore, we have had
spare capacity up to this point.  When another boiler is con-
nected to the scrubbers, a thickener outage will cause a boiler-
turbine generator outage.  Two boilers will be the total that
can be placed on the scrubber system until additional construc-
tion is completed.

ELRAMA LIME SYSTEM

          A report appearing in the literature in August, 1975
indicated that the Phillips scrubbers were the first in the
United States to go in service using hydrated lime.  That was
a misstatement.  However, it can now be said that the Elrama
scrubbers were placed in service and are continuing with hy-
drate.  However, we will change over to quicklime as soon as
possible since it is more economical than hydrated lime.

          The installed lime slaker, silo, and pumps had a feed
capacity of 9 Ib/minute with all scrubbers in service.  Since we
learned through experience at Phillips that this was inadequate
for proper pH control, we did not want to go through a similar
problem at Elrama, and because procurement of additional equip-
ment would have caused an impossible delay, steps were taken to
adapt the system to hydrated lime.  The slaker was converted to
a mixer,  the silo feeder changed, and new transfer pumps in-
stalled.   In addition, arrangements were made for the lime sup-
plier to leave two loaded 20 ton trailers on site at all times.
These, along with our 20 ton silo and a recently purchased 50
ton storage "blimp" and permanently installed lime blower, gave
us a total storage capacity of 110 tons and a feed rate of about
90 Ibs/minute.   The system has worked well, and a pH of 7-7.5
has been maintained with about 30 tons/day consumption for the
one boiler.   This is the equivalent of about 30 Ibs/minute of
quicklime.

ELRAMA SLUDGE TREATMENT AND DISPOSAL

          Under the original concept, it was believed that only
fly ash would be removed in the single stage scrubbers.  The un-
fixed sludge would be dewatered in three clay lined ponds, which


                             215

-------
would be excavated as required, and the sludge placed on our
normal landfill.  As we and the industry now know, there is al-
ways some SC>2 removal, and with the pH we consider necessary,
the quantity of SC>2 removed is not negligible.  And, as we have
sadly learned, unfixed sludge is not easily excavated or trans-
ported.  For this particular situation and with our recently
completed test of the IUCS prototype at Phillips, the POZ-O-TEC
system seemed to be the answer to our dilemma.  Fortunately,
IUCS had just completed their tests at the Mohave Station, and
that equipment was available.  After entering into a contract
with them,  the equipment  was moved from Nevada, and they were
ready to process sludge before we were ready to produce it.
Shortly after  startup, they were able to obtain and install a
vacuum  filter which increases the 35-40% solids in our underflow
to 50-60%.   Consequently, their process requires less dry  fly
ash to  be mixed with  the  sludge.

COSTS

           Thirty-two  million dollars has already been spent on
 the Phillips scrubber system with the expectation of an addi-
 tional  $8  million  to  be  spent for additional equipment to  come
 into compliance with  the  SO? emission limitations for the  entire
 plant.   Approximately $34 million has already been spent on the
 Elrama scrubber  system with the expectation of an additional  $12
 million to be spent  for  final compliance with the SC>2 emission
 limitations.  These  expenses equate to an  installation cost of
 $103 per Kw for  the  Phillips Power Station and $93 per Kw  for
 the Elrama Station.

           The actual  operating  costs we have been incurring  for
 the Phillips scrubber system, which is not yet complete, are  ap-
 proximately $9 million per year and are equivalent to approxi-
 mately 5 mills per Kwhr,  38* per million BTU, or  $8.50 per ton
 of coal consumed,  all including fixed charges.  Based on these
 costs,  the combined operating  costs for Elrama and Phillips when
 full SO2 compliance is achieved will be approximately  $32  mil-
 lion per year,  or  6  mills per  Kwhr, or  52$ per million BTU,  or
 $12 per ton of coal consumed,  all  including  fixed charges.

           The estimated disposal  costs  of  sludge  are  approxi-
mately $6  to $7  per wet ton,  or $12 to  $14 per dry  ton.  These
 costs  may change upon selection of  a  contractor  for  long  term
disposal and upon  determination of  the  cost  of acquiring and
developing additional disposal  areas.

           We expect that the  costs  of  operating  these scrubber
 systems,  when completed,  will  result  in an increase  in  the cus-
tomers'  bills by approximately 10%  over 1974  billing levels.

FUTURE  PLANS

           Our next phase of  construction toward  total scrubbing
for SO2  removal  at Phillips  and Elrama has been  known in our


                             216

-------
Company as Phase IA.  However, with the now recognized impor-
tance of continuous pH control, we have gone into an accelerated
Phase IA.  This is intended to give us redundant lime feed at
the Phillips Station by May, 1976, for more reliable scrubber
operation, and adequate quicklime feed at the Elrama Station for
scrubbing the gas from two boilers by June, 1976.

          Completion of the total Phase IA program will provide
the lime feed and storage facilities, thickeners, and auxiliary
equipment sufficient to enable compliance with SC>2 emission reg-
ulations.  This will include automatic pH control with the re-
dundancy required for continuous control.  This phase should be
completed by December, .1977 at both stations.  We plan to use
magnesium modified quicklime, not only to achieve compliance
without the necessity of installing additional vessels, but also
to avoid the deposits, which we experience with high calcium
lime.

          Sludge treatment and disposal is still an unsolved
problem in our opinion.  Despite glowing reports of presently
available methods, we are not convinced.  We probably have more
experience with the available methods than any other scrubber
operator, and none of them has yet been demonstrated as being
environmentally acceptable.  As mentioned previously, we are in
the process of receiving proposals for long term sludge disposal
from three different organizations.   We have specified that the
process chosen must yield a "structurally stable and environ-
mentally acceptable" product.  We want all this and a process
that we and our customers can afford.  As with all scrubber de-
velopments, this should prove interesting - and, no doubt, frus-
trating.

CONCLUSION

          In previous papers we had listed what we feel are
levels of performance that must be satisfactorily resolved if
the system is to be considered operationally feasible.  With
regard to our experience to date, we presently assess these per-
formance levels as follows:

      1.   Reliability - Although we have experienced un-
          acceptable reliability with the use of normal
          high calcium lime at our Phillips Station, our
          test program using magnesium modified lime in-
          dicates that an improved reliability can be
          achieved.   In addition, resolution of several
          of the problems at Phillips and the implemen-
          tation of those solutions  at the Elrama fa-
          cility prior to initial operation indicate
          that some improvement may be achieved in the
          reliability of the Elrama  scrubbers.   However,
          the Elrama system has not yet been run under
          full load conditions which would more accur-
          ately reflect true reliability.
                             217

-------
2.  Turndown Capability"- Over the course of opera-
    tion in the past year, it has been possible to
    follow the plant load with little problem pro-
    vided sufficient scrubber capacity is available.
    However/ we have not been able to determine if
    such cycling operation contributes to the scal-
    ing and deposit problems that have been experi-
    enced .

3.  Closed Loop Operation - Our goal is to achieve
    closed loop operation, however, this appears to
    be  several years away.  Some improvement was
    made in reducing blowdown, but it appears that
    further improvement will have to wait for the
    use of magnesium oxide lime and the addition of
    more lime feeders  and associated equipment.

4.  Sludge Disposal -  Our experiences with several
    sludge disposal techniques indicate that such
    techniques can result in acceptable stabiliza-
    tion of the sludge.  However, there is insuf-
    ficient information on long term leaching
    characteristics of this material to claim such
    techniques as being environmentally acceptable.

5.  Cost Consideration - Our installation and operat-
    ing experiences with the Phillips scrubber  system
    indicate  that upon completion of the Elrama and
    Phillips  scrubbers, the operating costs for these
    two scrubbing systems would increase the customers'
    bills by  approximately 10% over 1974.  Although
    this is not a physical performance level, the
    economic  aspect must be considered in a total
    system evaluation.
                       218

-------
      OPERATIONAL STATUS AND PERFORMANCE OF THE COMMONWEALTH
           EDISON COMPANY WILL COUNTY LIMESTONE SCRUBBER
                         Warren G. Stober

                    Commonwealth Edison Company
                         Chicago, Illinois
ABSTRACT

     The Will County Scrubber was installed in 1972 as a full scale
demonstration plant.  Over the past four years numerous problems have
been encountered, many of which have been solved.  However, reliable
operation on high sulfur coal has not been achieved.  This paper
reviews the past operating history, some of the major problems and what
has been done to solve them, scrubber availability, sludge treatment
operation, the cost to own and operate the scrubber system, and
future plans for Will County.
                                219

-------
             OPERATIONAL STATUS AND PERFORMANCE
             OF THE COMMONWEALTH EDISON COMPANY
               WILL COUNTY LIMESTONE SCRUBBER
        In the spring of 1970, Commonwealth Edison contracted
with Bechtel Corporation to investigate the sulfur removal
systems then available and to help us select a system having
a sufficient degree of development to warrant a large scale
installation on our Will County Station's Unit #1.

        After deciding upon a wet scrubber system using lime
or limestone, a specification was then prepared by Bechtel
and released for bid.  From the nine bidders that were
solicited,  seven proposals were received.  After detailed
study and bid evaluation with special consideration of the
project schedule, Babcock and Wilcox was given authorization,
in September, 1970, to begin the detailed engineering for a
limestone slurry system.  The project had a completion
deadline of December 31, 1971 which was established by the
Illinois Commerce Commission as part of a rate case.

Design

        The Babcock and Wilcox designed process was guaranteed
to remove 98% of the f" / ash and 76% of the sulfur dioxide.
These efficiencies were Msed on a dust inlet loading of
1.355 grains per standard ".ubic foot at 70°F and burning 4%
sulfur Illinois coal.

        The Will County wet  ^rubber was backfitted on a
177,000 gross kilowatt Babcock and Wilcox radiant cyclone
boiler that was put in service in 1955.  The location of the
scrubber and its associated equipment is indicated on Figure 1,
Property Plat of Will County Station.

        Construction presented a great many problems both
physically  and schedule-wise.  It was necessary to sandwich
the scrubber between the boiler house and the service building
with a substantial cantilever.  Erection of equipment began in
May, 1971 and with the use of substantial overtime, the
majority of the system was completed by February, 1972.  The
tight retrofit and quick schedule resulted in a very expensive
system:  $108 per net kilowatt in 1972 dollars including
indirect costs.
                            220

-------
       The wet  scrubber system is divided into three parts:
the limestone milling  system,  the wet scrubber and absorber,
and the sludge disposal  system.

       The milling  system as  shown on Figure 2 consists of a
limestone conveyor,  two  260-ton capacity limestone storage
silos, two full-sized  wet ball mills, each designed to
pulverize 12 tons  of limestone per hour so that 95% will pass
through a 325 mesh screen,  and a slurry storage tank.  The
required limestone is  high in  calcium carbonate, above 97%,
and low in magnesium carbonate, less than 1%.

       The wet  scrubber is made up of two identical systems,
each taking half the boiler flue gas.  Each module is designed
for 385,000 cfm  at a gas temperature of 355°F.  Each system
consists of two  recirculation  tanks, venturi and absorber
slurry recirculation pumps, a  venturi fly ash scrubber, a
sump, a two-stage  sieve  tray sulfur dioxide absorber, a two-
stage mist eliminator, flue gas reheater, and an ID booster fan.
Each system contains a by-pass damper which permits a module
to be taken out  of the gas path.   Figure 3 illustrates the
entire system showing  the relative location of each of the
above components.

       Figure 4 illustrates the flue gas path.  The flue gas
emerges from the boiler,  passes through an existing
electrostatic precipitator and on into the venturi.  The flue
gas velocity at  the  throat is  approximately 130 feet per second
with the pressure  drop being maintained at approximately nine
inches of water.   From the venturi the gas flows goes through
the sump and then  upwards into the absorber.  Here the sulfur
dioxide is removed as  the gas  is forced through two separate
sieve trays at a greatly reduced velocity, about 10 feet per
second.  From the  absorber the gas passes through a reheater
and on to the booster  fan.

       Figure 5 illustrates the slurry recirculation system.
The flow of slurry to  each venturi is 5800 gallons per minute.
and to each absorber 11,200 gallons per minute.  This gives a
liquid to gas ratio  of about 18 to 1 and 35 to 1 respectively
at full load.

       The waste  slurry is pumped to either a thickener or to
back-up settling ponds and all the clarified water runoff is
recycled to the  scrubber and milling system.
                          221

-------
        The power requirement to operate the scrubber system
is approximately 7,000 kilowatts or about 4$ of the unit
gross capacity of 177,000 kilowatts.  Over and above this
power requirement is a need for steam to reheat the gas
exiting the system.  The system was designed to reheat the
gas from 128°F up to 200°F, using about 50,000 pounds per
hour of steam at 435°F and 350 PSIG.

Performance 1972-1974

        B-module was the first module to come on line on
February 23, 1972, and wsis followed by A-module coming on line
on April 7, 1972.  Both modules were plagued by more problems
than one would expect to be normal during start-up.  Demister
fouling was very prevalent'and numerous changes were made to
the demister wash system in an attempt to solve this crippling
problem.   Other major problems that affected availability
were reheater tube vibrations, reheater fouling, scaling on
the absorber walls and grid plates, and slurry nozzle pluggage.

        Through the end of 1972, A-module operated 1444 hours,
the B-module operated 1237 hours, thus attaining availabilities
of 29.596 and 25.296 respectively.  (Availability, in this case,
is calculated by dividing the operating hours of the scrubber
by the operating hours of the boiler-turbine unit).
Simultaneous operation was 469 hours, for an availability of
only 9.6%.  The longest period of continuous operation was 21
days, this being with A-module, and the longest simultaneous
operation  of both modules was 6 days.

        By early 1973 many problems had been resolved; however,
the major  ones remained:  demister pluggage, reheater pluggage,
and scaling.  Along with the recurring major problems, new
problems were encountered, such as cracks in the booster fan
inlet cones, chloride stress corrosion cracks in the 304
stainless  steel reheater tubes and eventual failure of the Corten
reheater tubes.

        In April 1973 a decision was made to discontinue
operation  of B-module and concentrate all efforts on achieving
satisfactory operation on A-module.  If A-module could
demonstrate satisfactory operation, then B-module, which
originally had a turbulent contact absorber, would be modified
so as to be identical to A-module.
                            222

-------
        A-module, after operating for 1726 hours through
November, 1973 (22.6% availability) was taken out of  service
for extensive changes in the demister, demister wash  system,
and reheater.  The underspray water supply was changed from
pond reclaim water to house service water.  This change was
made to try to prevent the scaling that was taking place on
the demister.  The new second stage demister was designed
to knock out any wash water carryover in the gas flow from
the first stage demister top wash system.  The intent of
these changes was to eliminate demister scaling and
carryover into the reheater coils.  In an attempt to  slow
down the rate of deterioration of the reheater coils, a
steam supply line was run from the building heating system
to the coils for use during scrubber outages.  Keeping the
coils hot and dry during outages reduced the corrosion due
to chlorides deposited on the coils during operation.

        A-module returned to service in late March, 1974
following completion of the modifications and an additional
delay to repair damage caused by a freeze-up of instruments
and the pond recirculation system.

        During 1974 we changed the method of calculating
scrubber availability from the ratio of scrubber operating
hours to boiler operating hours to the ratio of scrubber
available hours to the total hours within a given time
period.  While neither method is completely satisfactory,
we believe the new approach provides a more accurate
indication of scrubber performance.   In some cases this has
increased the availability because the repairs to the
scrubber were completed before the boiler repairs.   However,
for the sake of those who still like to look at scrubber
availability based on boiler operating hours,  these numbers
have also been calculated.

        In 1974,  A-module was available for service 6025
hours for an availability of 68.894.   A-module operated 5468
hours and unit 1  was operational 7,924 hours giving a
scrubber availability based on operating hours of 69.0%.

        After achieving what seemed  like a workable system with
A-module,  we decided to proceed with the modification of
B-module,  incorporating all the changes made to A-module.   The
major changes included the  removal of the TCA ping-pong balls
and replacement with two stages of sieve trays,  the single
stage demister being replaced by two stages of a sturdier
molded design,  and the reheater bundles being replaced with
                           223

-------
5/8" OD X 0.065" 316L stainless steel tubes in the lower
three banks and similar sized carbon steel tubes in the
four upper banks.  B-module remained out of service through
the end of 1974 and well into 1975.
1975 Operation

        ¥e  entered  1975  operating  in much  the  same manner
as we  operated  in 1974.   A-module  was  operational and B-module
was  still undergoing modifications.  The first three months
of  1975 saw A-module achieve availabilities  of 99-5%, 99.4%,
and  94.0% (based on hours that the module  was  available  for
service).  Table I  summarizes 1975 availabilities computed
by both methods, one based on boiler-turbine operating hours
and the other based on the hours the scrubber  was available
 each month.

         Figures 7 through 16 are monthly charts indicating
 all the unit and scrubber outages.  While  these sheets give
 the general nature  of the outage,  they do  not  specifically
 mention or detail the work done on the scrubber during unit
 outages.  It should be noted that it was always necessary to
 do some scrubber maintenance work during those outages.

         During the first three months  of  1975  the^e  were
 eight A-module outages.  However,  only one was a forced  outage,
 the others being for no unit demand, inspections and one
 accidental trip.   The forced outage was  due to the  splitting
 of a slurry supply hose to the venturi.

         At the time the reasons for the increased availability
 during 1974 and the early months of 1975 were hard to ascertain.
 Looking back today at what took place then and what was to
 follow later in  1975, we believe that the type of coal we
 were burning at the time had a lot to do with the increased
 availability.  The addition of a second stage demister,
 automation of pH control and other minor changes had some
 influence, but were not determining factors.  As part of our
 commitment to reduce S02 emissions, only low  sulfur (approximately
 0.4?' S)  coal was being shipped to Will County Station.  As
 this type  of coal was mixed with and gradually displaced the
 high sulfur coal remaining  in the storage pile, the average
 sulfur content  of coal burned in Unit #1 dropped.  In 1974, the
 average was 1.5#; in early  1975,  the average was slightly over
 1.056.  We  believe that the  low  sulfur content  of the coal and  the
 change from sludge pond reclaim water to house service  water for
 the demister wash were the  primary  causes of  the improved
 reliability.  This conclusion was tentatively confirmed when
 we attempted operation on high  sulfur coal  later in 1975 and
 encountered severe problems.
                               224

-------
         The change from sludge pond reclaim water to house
 service water for demister wash water upset the water
 balance of the system.   The high availability of A-module during
 the  first  three months  of 1975 which resulted in part from
 this wash  water change,  caused the water level in our sludge
 pond to frequently rise  to the point where we became concerned
 about an accidental overboarding to the nearby Des Plaines
 River.   In April the scrubber was shut down until changes
 could be made that assured overboarding would not occur
 Three steps were taken to tighten the water balance and assure
 a closed loop operation:   1.  Pump gland water flows were
 reduced by 50%;   2.  The  scrubber house service water filter
 backwash was routed out  of the system;   3.  The continuous
 demister wash underspray was  changed to an intermittent spray
 of 5  minutes on  -  5 minutes off.   A-module returned to  service
 on May  5,  1975 and the noted  changes proved to be  effective  in
 maintaining  a  lower pond  level thereby assuring that overboard ne
 would not  take place.                                           &

        B-module returned  to  service  on May 20,  1975  upon
 completion of  its  modification program, and the  start up  was
 rather uneventful  with only minor problems  being encountered
 which resulted in  a  couple  of  short  outages.

        Early in the year  it was decided that  we should attempt
 to operate the scrubber on  Illinois high sulfur coal.  Because
 Will County was receiving only  low sulfur coal as part of  our
 S02 compliance program,  it was  necessary for our Fuel
 Department to arrange for special shipments of high sulfur
 coal for the test.  Difficulties in obtaining and storing
 the test coal forced us  to begin burning it before we had
 obtained sufficient experience  operating both modules
 simultaneously or with the tight water balance which had
 Oust been put into effect.  As a result, we were not totally
 prepared for the myriad  of problems which began to occur.

        During the high  sulfur coal burn program, SO? removal,
 particulate removal in the scrubber, and precipitator efficiencies
 were  measured along with extensive monitoring of cycle chemistry
 Removal efficiency tests were conducted under full and partial
 load  conditions and with the electrostatic precipitator ahead
 of the scrubber in and out of service.  S0? removal efficiencies
 averaged 78.2% on A-module and 86.8% on B-module with inlet
S02 loadings averaging 3573 ppm.  A-module removal efficiencies
were  lower  because only  one of the two absorber slurry pumps  was
 in service, resulting in  a lower liquid to gas ratio.  The other
absorber pump was out of  service for repair of the impeller and
rubber lining.
                            225

-------
        Particulate removal efficiencies were very erratic
and lower than expected.  Scrubber outlet loadings, which
were measured using the EPA Method #5 train, were high;
with a large percentage being slurry carryover rather  than
fly ash.  During the test period slurry solids in the
recirculation tanks rose to a higher level than desired.
In an attempt to reduce the solids content, additional water
was added to the cycle by changing the demister wash cycles.
Bottom wash was made continuous and the top wash frequency
was increased.  The top wash is a deluge system, utilizing
a high flow for a  short period of time.  But since this was
pond return water, the increased use of this system prompted
scale to form on the walls.  Scale and pluggage in the demister
apparently resulted in uneven gas distributions and high
velocities which permitted slurry to carryover through the
system.

        Ten days after the start of the test an abnormally
high pressure drop was noted across the A-module reheater.
An inspection showed the first stage demister starting to plug
and the reheater coils heavily fouled.  B-module was taken out
of service  13 days after the start of the test because of high
vibrations of the  booster fan.  At that time inspection of the
module  showed severe demister pluggage and the fan rotor coated
with a  dark scale.  Approximately four days were required to
clean  and balance  the fan before B-module could be returned to
service.

        By mid June, 18 days after commencing the high sulfur
coal test program,A-module was out of service due to massive
reheater leaks, reheater fouling and demister pluggage.  It
remained out of service for repairs for the remainder of 1975.

        With tight control of the water balance in order to
assure  a closed water loop, two modules in service and operation
on high sulfur coal, changes within the scrubber cycle were
occurring quicker  than we could detect and correct them.  With
the turn of events being what they were, it is difficult to
assess  the  operation or draw any firm conclusions from the
attempted high sulfur coal test other than to ss>y we were unable
to successfully operate on high sulfur coal under these conditions.

        Shortly after the test shipment of 20,000 tons of 4%
sulfur  Illinois coal was consumed,  it was decided that coal from
the storage pile would be burned in Unit #1.  Approximately
90,000  tons of coal ranging from 0.8% to k% sulfur remained
                              226

-------
segregated from the remaining 0.4% sulfur coal in storage
and was used to fuel Unit #1 when B-module was in service.
However, during the months of July, August, and early
September 1975 there were occasions when low sulfur (0.4%)
coal was used to fuel the boiler and this resulted in a large
variation of the inlet S02 concentration in the flue gas
entering the scrubber.  By mid September, B-module was taken
out of service because of massive scaling in the absorber,
in the sump area just below the absorber and in the demister.
Presumably this was caused by the pH control either not
functioning properly or the system just not being able to
handle such large variations in the inlet SO- conditions.
An outage of over one week was necessary to nand-chip the scale,
which in some places was one-half inch thick, off the various
internals.

        Another major problem that affected performance of
B-module began occurring with less than 1000~hours of operating
time on the rebuilt module.  This was the failure of the new
carbon steel reheater tubes.  We eventually experienced
failures in six of the twelve new tube bundles over a three
month period.  Peripheral cracks occurred in the tubes just
before they entered the header.  In all cases where failures
occurred, the tube supports had worked loose permitting the
tube to vibrate.  The coil manufacturer has repaired and
modified all the tube bundles by redesigning the tube supports
and their method of attachment to the frame, including an
additional support.

        Other parts of the system have also experienced
problems which are worth noting.  The horizontal plate above
the nozzles at the venturi inlet wall wash box developed numerous
holes allowing slurry to splash and drip over the scrubber
structure.  The cause was found to be slurry being deflected
onto the top plate and eventually eroding through at the point
of impingement.  Each nozzle has now been fitted with a deflector
designed by B&V to redirect the slurry flow.  The duct between
the absorber  outlet and reheater inlet has experienced corrosion
to the extent that it has eaten through the ductwork.  The
ductwork material is J" corten steel with a 60 to 80 mil
Ceilcote Flakeline 103 coating.  After the coating either fell
off the wall or was eroded away, it was not long before corrosion
took its toll.  This area was not the only area in which we
experienced problems with coatings.  In the sump, the walls between
the abosrober were lined with Flakeline and after the scale was
removed a thorough inspection of the walls showed areas where the
lining was gone and corrosion was taking place.  The other half

-------
of the sump, that area below the venturi, is lined with
1-J inches of Kaocrete refractory over the Flakeline and
appeared to be in excellent condition.  The walls below
the absorber have since been cleaned and covered with
refractory.  Severe corrosion of the reheater bundle support
frames has taken place.  In some areas there is little,
if any, of the carbon steel remaining.

        One of the absorber recirculation pumps was
damaged when a check valve in the pump discharge
piping failed and portions fell into the pump.
The pump impeller was damaged and its rubber lining
destroyed.  From the debris found in the pump, it
appeared that the rubber lining had eroded on the check,
allowing the base material to corrode and ultimately fail.
B&W reviewed the need for check valves in the pump
discharges and recommend that they be removed.  They have
since been removed and the spool pieces relined with rubber.

        Unit #1 came down for a boiler, turbine and scrubber
overhaul in mid October for an outage that lasted more than
four months.  This gave us an opportunity to conduct an
extensive inspection and maintenance program of all scrubber
components.  Approximately $200,000 was spent during this
outage on maintenance material and labor in an attempt to
have the scrubber in like-new condition upon its return to
service.

Sludge Treatment

        There has been little in the way of new development
at Will County regarding the sludge treatment and disposal.
The treatment methods remain the same as reported in previous
meetings, that is, thickener underflow or retrieved ponded
sludge is placed in concrete mixing trucks along with fly
ash and lime and then transported to a dumping site.  Until
September 1975, the treated material was dumped into a
seven acre on-site clay lined disposal basin where the
material would set up after one to two weeks depending on the
ambient air conditions.  Since September 1975 the Material
Service Company, a local quarry and concrete products firm,
has been transporting the treated sludge from our facilities
directly to a permanent disposal site which they operate.
                           228

-------
        During 1975, Illinois EPA gave a local group a landfill
permit to use the stabilized sludge already in our on-site
disposal basin as fill for a future golf course.  Over
eighty thousand tons of solidified treated sludge have been
removed from our basin and transported to the landfill site
at a cost to us of $2.35 a ton.  There is still additional
material in our basin that needs to be removed when weather
permits.

        Two years ago it was noticed that the solidified
treated sludge stockpiled in our on-site disposal basin
was crumbling at the surface.  Investigation into this
phenomenon by Dr. Berger of the Civil Engineering Department,
University of Illinois has shown that the treated material
loses its ettringite and gypsum properties after undergoing
several freeze-thaw cycles.  Analysis by x-ray diffraction
showed a greater deterioration of these properties as the
number of cycles increased.  The exact cause of this
phenomenon is not totally understood; however, Dr. Berger
feels there may be a carbonation of material during the
freeze-thaw which lowers the pH, thereby causing
deterioration of the ettringite.  Perhaps removal of a
larger portion of the water from the sludge and compaction
of the treated material will reduce the porosity enough
such that there is a reduced effect from freeze-thaw cycles.

        A belt type vacuum filter has been purchased for
¥ill County and should be in service by mid-summer of this
year.  Once this filter is in service we will have to change
our methods of handling which should reduce the cost of
treating the sludge.  We also anticipate seeing some
improvement in the characteristics of the sludge after the
removal of more water.  The one big disadvantage to this
installation will be the additional water we are going to
have in our system which will make it all the more
difficult to maintain a closed water loop.

Costs to Own and Operate

        Approximate operating costs for the Will County
scrubber are shown  in Figure 6.  These costs are based on
actual  expenses incurred for operation and maintenance
through the year 1975 and are considered approximate because
there are a few invoices still outstanding for  1975; however,
the numbers appearing in Figure 6  should be considered
indicative of the cost to own and  operate the scrubber
'system.
                            229

-------
        The fact that the scrubber did not have the same
availability as the boiler necessitated that the values
for the tons of coal consumed, BTU's and net kilowatt
hours used in the cost analysis be proportioned to the
quantity of flue gas treated by the scrubber when in
service.

        Included in the cost to own and operate are the
following items:  annual carrying charges on the invested
cost, operating labor, maintenance, limestone, auxiliary
pov/er, rehea.t steam and sludge treatment.  Included in
the auxiliary power cost is the cost for actual power
consumed and the annualized cost of replacement kilowatts
for that which is lost to the system because of the scrubber.
Reheat steam costs are based on the cost of the fuel
required to generate the steam which is removed from the
boiler cycle.

        The true meaning of maintenance costs is hard to
assess when taken over a short time interval.  One would
not expect maintenance costs to be as high the first year
as would be expected after two or three years of operation.
Therefore it is essential that a good data base be
established over a period of time to get the true picture
of what maintenance on these systems actually cost.  The
cost  for maintenance as shown in Figure 6 would have been
higher had all the 1975 charges been available for
inclusion in the dollar figure shown.  Not included in
the maintenance cost shown are the costs for the B-module
modification.  The modification costs were investment
dollars rather than expense dollars because equipment of
a  new design was installed.  We anticipate higher
maintenance costs in the years ahead when greater operating
hours are attained on the system.

        The sludge treatment cost includes labor, material
and hauling of the treated sludge to the off-site disposal
area.  Also included is the maintenance cost of the sludge
treatment facilities.  For the period January through November
27» 7^1 tons of dry solids scrubber sludge were treated
at a  cost to Edison of $546,217.  V/hen referring to dry
solids of scrubber sludge we are talking about the material
leaving the scrubber system with no additives or water
included.  This would be the calcium sulfite and sulfate,
unreacted limestone and that fly ash removed in the scrubber's
venturi section.  On this basis, it cost us $19.69 a ton  (dry
scrubber solids) to treat and dispose of the material.
                           230

-------
        To summarize, the cost to own and operate the
scrubber was about $22 a ton of coal, 1140 a million
BTU's or 13 mills per net kilowatt hour.

Future Expectations

        While the main objective at Will County has been
to demonstrate scrubber operation on a full-sized unit, we
have, out of necessity, done a considerable amount of
experimentation in our efforts to make the system
operational.   Three main reasons make it impractical to use
Will County for R&D projects that would require frequent
changes of hardware:  (1). Will County Unit #1 while not
a base load unit, is an essential part of the Edison system
and has had a capacity factpr of 505' or greater over the
past years, (2). The large size means that any major
changes would require large amounts of time and money,  and
(3). The tight physical configuration puts a constraint
on any changes that would require the addition or
re-arrangement of equipment.

        It has become painfully apparent that operation on
high sulfur coal with an assured closed water loop will
require considerably more costly experimentation.  From
past experience, just attempting to keep the system
operating will not provide solutions, if any,  to the
problems at hand.  We are in the process of developing  a
test program to attempt to control,  if this is possible,
the chemical cycle.  The information gained from this
program, in addition to possibly improving operations at
Will County,  may be of some value in designing, building
and operating the 450 MW scrubber slated for our
Powerton Station.
                           231

-------
              TABLE I



1975 OPERATING HOURS & AVAILABILITY
Generating
Unit Operating
Hours
1975
Jan.
Feb.
Mar.
April
May
June
July
August
September
October*
November *
December *
6
* 21 week
684.5
666
609
638
744
642
689
564.5
720
194.5
0
0
,151-5
Scrubber
Operating
Hours
A B
676
662
604.5
252.5
628.5
389.5
0
0
0
0
0
0
3213
0
0
0
0
276
543
547
568
451.5
194
0
0
2579-
scheduled overhaul started
Scrubber
Availability
Based On Unit
Operating Hrs.
A B
98.8%
99.4%
99.3%
39.6%
84.5%
60.7%
0
0
0
0
0
0
5 51.8%
October 11,
0
0
0
0
37.1%
84.6%
79.3%
100.0%
62.7%
99.7%
0
0
41 . 9%
1975
Scrubber
Available
Hours
A . B
740
667
699
266.5
628.5
462
0
0
0
0
0
0
3^63

0
0
0
0
276
615.5
589.5
696
451.5
240.5
0
0
2969

Scrubber
Availability
Based on The Number
Of Hours In Each
Month
A B
99 . 5%
99.4%
94.0%
37.0%
84.5%
64.2%
0
0
0
0
0
0
39.5%

0
0
0
0
37.1%
85.5%
79.2%
93.5%
62.7%
32.3
0
0
33.9%


-------
                                                          FIGURE 1
  SLUDGE
  WASTE
   POND
    Q
THICKENER
PROPERTY  PLAT
                        I38KV SWITCH YARD
                       TURBINE ROOM
                      BOILER ROOM
                     SERVICE
                    BUILDING
                                           WASTE SLURRY AND RETURN
                                           WATER PIPELINES
                                                COAL
                                              CONVEYOR
                                  COAL
                                 BREAKER
                                 BUILDING
                                           LIMESTONE
                                             MILL
                                      WET
                                   SCRUBBER
                     BUILDING AND
                      CONVEYOR
RECLAIM
HOPPER

-------
RECLAIM
HOPPER
        LIMESTONE
           SILO
                            SYSTEM
                     FEEDER
                                  _JCLASS1FIER
MILL
                                 v

                         RECYCLE
                         TANK a  ^
                          PUMPS
      r
              SLURRY
              STORAGE
               TANK
                    v
                 TO WET
                 SCRUBBER

-------
 Recycle and
make-up water
   Commonwealth Edison Company
    Will  County Station - Unit No. 1
                 APC-1

              FIGURE  3

-------
                                       FIGURE
BOILER
                 FLUE GAS  PATH
                  STACK
           EXISTING
            ID FAN
BYPASS
DAMPER

ELECTRO. /
                   r-
                    \
 NEW ID
'BOOSTER
   FAN   /
-VENTURI
             REHEATER


             ' DEMISTER
                                   SUMP
              A3SORBER

-------
                                           FIGURE 5
               SLURRY  RECiRCULATIOfel  SYSTEM
TO SLUDGE
  WASTE
  POND
 VENTUR!
  PUMPS
            VENTUR1 \
                           SUMP
  VENTURI
REC1RCULATION
    TANK
                                        ABSORBER
                                  ABSORBER
                                RECIRCULATiON
                                    TANK
                                    FROM MILL
                                     SYSTEM
                                    ABSORBER
                                     PUMPS

-------
                                                 FIGURE 6

                        'WILL COUNTY UNIT 1  WET SCRUBBER
SCRUBBER SYSTEM

Carrying Charges
on $14,900,000

Labor (Operating
and Technical)
Maintenance  (Labor
and Material)
Limestone
Auxiliary  Power
Reheat  Steam
APPROXIMATE
& ANNUAL COST
$2,280,000
.81,709
256,656
62,726
586,875
72,595
1975 COST TO
&/TON OF COAL
$ 12.54
0.45
1.41
0.54
3.23
0.40
OWN AND OPERATE
ei/MMBTU
65.3 0
2.3
7.4
1.8
16.8
2.1
MILLS/KWHR
7.47 Mills
0.27
0.84
0.21
1.92
0.24
                    $3,340,561
$ 18.37
95.7 0
                                                                10.95 Mills
 SLUDGE TREATMENT

 Carrying Charges
 on $642,000
 Sludge Treatment
98.000
546,217
0.54
3.00
2.8
15.7
0.32
1.79
                    $  644,217
$  3.54
18.5 0
2.11 Mills
 SCRUBBER & SLUDGE TREATMENT TOTAL COST

                    $3,984,778         $ 21.91
              114.2 0
           13.06 Mills
 Notes:
        1.  Scrubber system has a 14 year life.

        2.  Sludge treatment cost includes hauling to an off-site disposal area.

        3.  A-module on line 3213 hours, B-module on line 2580 hours.   181,886
            tons of coal consumed, 3-49 X 10"!2 BTU and 305.1 X 10° net kilowatt
            hours are proportioned to scrubber on line hours.

        4.  Coal average:  9604 BTU/lb, 1.596 sulfur.

        5.  Unit #1 12 month capacity factor for 1975 was 49.496.
                                                           12/30/75
                                       238

-------
           FIGURE 7
1O 11 12 13 14 15 16 17 18 19 ZO 21 22 23 24 25 26 27 28 29 3O 31
        239

-------
              n
                UNIT


o
  (0
  la









                 CYCLONE, jn
                       110

A-MODULE  SCRUBBER
                      B-MODULE SCRUBBER





                                         .


          '-fO.
      NOZZLES.
         IJN.IT.



 A.

-IOUR



                                                                                   .










                                                                                                             PI

                                                                                                           •

-------
UNIT
A-MODULE SCRUBBER
B-MODULE SCRUBBER

-------
UNIT
A-MODULE SCRUBBER
B-MODULE SCRUBBER

-------
UNIT
V-MQDULE SCRUBBER
                                                             B-MODULE SCRUBBER


















                        ICHAN
                        TO.P









     REVENT. WATER. hV.ERBORAlJlNG.
       1.11.HQIR5..

                             . TRJP-PAMPERS. .


                                     UTESi





                                             .
                             UNE)4:RSPR|AY. NOZZLES.
                                 5 HOtJRS









                                                    .
                                                           .
                                                         •


    Mm
!UNDERGO[NG
                                         iiiiPt-:
                                                              m

                                                                   REffTjRN
                                                                      '1, H
-------


VJ1
                                                                               '
                                                                             I


-------



 UNIT

!  *.
Li

  . cyciQNt;
 "I

        HOURS







>DULE  SCRUBBER
                                                            B-MODULE SCRUBBER






                          , .DEMIST]'5  ft.

                                       i-LS.

                          . REfLACEMEMl


























                                                               L.J
                                                                       o





                                                                      86 ,H

                                                                    . i

                                                                  STEW jb&



                                                                        1  .OUR




                                                                                   A
-------

                      A-MODULE SCRUBBER
B-MODULE SCRUBBER

     IT
NO. .UNIT.
CYCLONE.

-------



UNIT
A-MODULE SCRUBBER
B-MODULE SCRUBBER









                             MISTER &
                                 -R. t
                             IPLAOEMEN'T


                                              I
                                                                 SS
                                                                     .
                                                                 DEHiCSEH
                                                                      198 HOURS
                                                                BALA:
                                                                  :
                                                   , | CLjEAN

                                                                                  y3SQR_BES
                                                                         0
                                                    HQURS
                                                       T
                                                                                  'IMP
                                                               £CJRJEN



-------
UNIT
A-MODULE  SCRUBBER
B-MODULE  SCRUBBER
                           DlMIStEB.&t.
  UKUWXX-.3OIM
   R] HEATER. CDII

                            CHEDULED. JVERIJAUL
  rnr t r%*r%^*ii»t*i   n
                                            E2ULEI!
      JlAUL

-------
          MHI FLUE GAS DESULFURIZATION SYSTEMS APPLIED TO
                     SEVERAL EMISSION SOURCES
          M.  Hirai, M.  Atsukawa, A. Tatani, and K. Kondo

                 Mitsubishi Heavy Industries, Ltd.
                           Tokyo, Japan
ABSTRACT

     Since the development of its lime/limestone-gypsum recovery
process, Mitsubishi Heavy Industries, Ltd. (MHI) sold thirty-two
(32) units of MHI FGD Systems in Japan.  Twenty (20) units are operating
smoothly, eight (8) are under construction and four (4) are committed
by a contract.  The total capacity of these installations amount to
11,282,000 scfm, treating gases from boilers, sintering plants and a
copper smelter.

     This paper describes the features of MHI FGD System, operating
experience of its commercial scale units and its applicability to coal-
fired boilers.

     Further, MHI recently developed a process of removing SO  and
NO  simultaneously by lime/limestone scrubbing, and its 1,200 scfm
pilot test was completed in 1975.  Its outline is also reported
in this paper.
                                 249

-------
1.     INTRODUCTION

      The  development of MHTs lime/lime stone-gypsum  recovery
process  started 15 years ago,  and its intermediate progress  report
was made public at the Second International Lime/Limestone  Wet
Scrubbing  symposium, back in 1971.   (Reference 1)

      At that time  construction was in progress  for a large prototype
unit with the capacity to  treat  20% of the flue gas from a 156 MW
power unit.   This  prototype plant started up  in March,  1972  and  as
its operation was "very satisfactory (Reference 2), MHI has been suc-
cessful in marketing its  systems in Japan.   When limited to utilities,
MHI's share in the.market is about 56% in terms of treated  gas flow.
(Reference 3)

      MHI's system development  is shown  in  Figure 1,  and its delivery
record is listed  in Table 1.

       Among  the 20 units  in  operation,  15 units are for oil-fired
boilers,  4 units  are for  sintering plants (steel mills) and the remaining
one unit  is  for a copper  smelter.  The  capacity  of most of these  in-
stallations ranges  from 250,000 to 740,000  scfm in terms  of gas volume
treated,  and among the units  so  far  installed, the largest capacity of
one module of a scrubber is 560,000 scfm.   SOj, concentration  in the
flue  gas  ranges from 500 to 23,000 ppm.   Lime or Limestone is  used
as absobent and  high quality gypsum  is  recovered as a by-product
which is used  for  cement and wallboard production.

      The prototype plant mentioned above has so far been operated
for about 4 years, and it is now being operated  as a  156 MW full-
scale plant,  in combination with  a plant which treats the remaining
80%  of the flue gas.  Further, another  156 MW  full-scale  plant  is
under construction in the same power station.

      In evaluating a lime/limestone  scrubbing system,  the following
three items are  considered to be important.

1)    Reliability

2)    Economics

3)    Sludge disposal

      From the start point, MHI's aim were to  fulfill the above  re-
quirements  and so far have been quite successful.
                               250

-------
      This paper describes the features of the  MHI  Lime/Limestone -
Gypsum  Process,  operating  experience of its  commercial scale system
and its  applicability  to  coal-fired boilers.

      Recently,  NOX removal has become a serious problem.   For
several years,  MHI  has been  conducting  a series of R &  D programs
on both wet  and dry NOX removal processes.   One  of  the  successful
developments is a process  of  removing NOX  and SOX simultaneously
in a lime/limestone  scrubber  at high  efficiency,  by use of a water
soluble  catalyst.   As its pilot test was completed in 1975, its outline
is introduced in this paper.
                              251

-------
2.   SYSTEM DESCRIPTION

      The  MHI FGD System is  a wet non-regenerative system which
uses lime or limestone  as  absorbent and recovers high  quality gypsum
as by-product.   A  simplified flow scheme of the system is shown  in
Figure 2.

      The flue  gas  is  fed into the precooler to be cooled down,  most
of its particulates  being removed there at the  same time before  enter-
ing the SC>2  scrubber.   The absorbent  slurry  for  SC>2 absorption,  and
the gypsum "seed"  slurry  for  scale prevention, are fed to the scrub-
ber.   The cleaned flue  gas passes through the mist eliminator and is
then  reheated to restore buoyancy.

       The  scrubber slurry flows into the sulfite oxidizer where all
 the calcium  sulfite in the  slurry is oxidized by air,  yielding  gypsum
 slurry.   A portion of the  gypsum slurry is recycled to the scrubber
 and the remainder is transferred to the gypsum recovery  section  of
 the system.   Here,  the gypsum slurry is concentrated by  a thickener
 and the underflow  slurry is fed to centrifuges,  yielding by-product
 gypsum.    Most of the thickener overflow is reused for  absorbent
 preparation.   Slowdown water from  the precooler  and  part of the
 thickener  overflow are  fed to  the water treatment  section  of  the
 system.

       Each constituent of  the  MHI FGD System has the  following ob-
 jectives  and features.

 2. 1   Precooler

  Main objectives are as follows:

  l)    Removal of dust from  the incoming gas  to obtain  high quality
       gypsum.

 2)    Cooling of the flue gas  to protect plastic materials used in  the
       SC>2 scrubber.

 3)    Humidification of the flue gas  to avoid any  local evaporation
       which  may lead to deposit formation in the SC>2 scrubber.

  A spray tower is used to economically achieve the above require-
  ments.   The  spray pump capacity in  terms of L/G is normally 14
  and gas  stream pressure drop (including demisters)  is about 1  1/2
 in.
                              252

-------
2. 2   SO2 Scrubber

Important requirements of this equipment are as follows:

l)     Stable  and high SC>2  removal  efficiency even at severe load
      fluctuations.

2)     Scale-free operation.

3)     Low pressure  drop (i. e.  low power requirement).

4)     Low mist  entrainment at the  outlet.

In order  to  accomplish .the  above  requirements,  MHI has developed
a packed spray  tower employing low density "grid"  packings.
Simple structure of  the "grid" packings  enables  the  scrubbing slurry
to be  distributed uniformly  on its surface,  thus, giving  a large  ab-
sorption area.   Another feature of the "grid" packed tower is its
low mist entrainment at the outlet which contributes to the reliability
of the mist  eliminator.   The scrubbing slurry pumping rate in terms
of L/G is 50 to 140.

Scale-free operation is accomplished by  using relatively high L/G
and  adding  gypsum "seed"  slurry  to  the  scrubber.    The gas  stream
pressure drop is about 1  1/2 in.

2. 3    Mist Eliminator

The mist must  be removed from the gas at high efficiency without
scaling of the eliminator surfaces.   All  of  the  MHI  FGD Systems
that are now in operation are  using vertical chevrons,  giving very
good service.    Pressure drop is within  the  range  of 1/2 to 1 in.

2.4    Reheater

Reheating the  scrubber gas is necessary to  restore  buoyancy, to
reduce visible  steam plume,  and  to protect  downstream equipment
from  possible corrosion.

 2.5   Absorbent Preparation

 This  section of the  process is important bacause it  gives great
 influence on the following.

 1)     SC>2 removal efficiency and  absorbent utilization.


                             253

-------
2)    By-product gypsum quality.

In order  to  achieve this, the ultimate method is to make  the particle
size of the  absorbent  slurry as small as  possible.

2. 6   Sulfite Oxidizer

The oxidation reaction of calcium sulfite to calcium  sulfate depends
on dissolution and  diffusion of  oxygen into the  slurry.   MHI is using
a rotary, atomizer,  an air  atomizer of unique structure, which gene-
rates  fine air bubbles  at operating pressure  of about 50 psig.   The
oxidation rate is also improved by lowering the slurry  pH..  A small
amount of sulfuric acid is  fed  to the oxidizer for this purpose.

2. 7   Gypsum Recovery

Gypsum  crystals produced  in the oxidizer are  relatively large, blocky
 crystals  that settle easily.    Therefore,  the gypsum slurry  can be con-
 centrated in a  reasonable size thickener and dehydrated by conventional
 centrifuges. The  by-product gypsum  contains  only 5 to 10% of free
water.   Gypsum is a stable, harmless compound that can  be used  for
wallboard production and as a  retarder  for cement setting.  In the
 case  of over-supply,  the by-product gypsum  is more suitable  for  land-
fill  disposal than sulfite sludge.

2. 8   Water Treatment

Some water is disposed of  to keep the concentration of chloride and
other impurities in the system ar a safe level.

Bleed streams from the precooler and the thickener overflow  are
treated to meet the local water pollution  regulations.

2. 9    System Control

The controls for the MHI FGD System depend  upon the type of ab-
sorbent used.   In  lime scrubbing,  the slaked lime  slurry feed is
controlled by the pH of the recirculating  scrubber slurry.   In lime-
stone  scrubbing, the limestone slurry feed is  controlled by the SO2
in the flue  gas  and the pH  of the recirculating  scrubber  slurry.
                             254

-------
3.      OPERATING EXPERIENCE

       MHI has experience  in  construction and operation of the  FGD
systems  applied to  several  emission sources.   They are oil-fired
boilers,  sintering plants (steel mills) and a  copper  smelter.

       Each emission source has different flue gas  conditions,  and
consequently,  the associated FGD system runs somewhat differently
from one another.   Operating data obtained from the commercial plant
operation are described in Table 2.

       Operating experiences in the  commercial plants, problems en-
countered, and solutions to these problems are outlined below.

3.1    Precooler Operation

Problems on materials  of  construction were experienced due to  low-pH
(1  to 1.5)  and  corrosive nature  of the recirculating spray water.

1)    Pin-holes on rubber  linings due to bad -workmanship  caused corro-
     sion  on carbon steel underneath.

     This problem was overcome by setting  stringent  standards for
     lining works  and pin-hole inspection.

2)    In  an installation for a sintering plant,  the incoming gas laden
     with oily matter  caused  swelling of rubber  linings.

     This problem was solved by using oil resistant rubber linings.

3)    In  an installation which used sea water to prehumidify the  flue
     gas,  accumulated chloride ions in  the recirculating  spray  water
     caused corrosion on  stainless  steel valves.   The problem was
     solved by using rubber lined valves.

3.2  Scrubber Operation

3.2.1  SO, removal efficiency and absorbent utilization
          £*
Independent of flue  gas conditions and types  of absorbent used,  the
concentration at  scrubber outlet ranges from  10  to  60  ppm,  or  more
than 90% in terms of removal efficiency.
                               255

-------
Past operating data shows that the  absorbent utilization ranges from
90 to 99% or on the average  of 95%.  One of the  reasons why the
absorbent utilization is high even in limestone scrubbing  is that most
of the  limestone in Japan is  of high calcium in nature.

3.2.2  Scrubber  control

Using  the pH of the recirculating scrubber slurry as a control factor,
all of  the scrubbers are  running with  excellent stability.   For example,
the SO  concentration  of  flue gases  from sintering plants  changes from
800  to  1,200 ppm every 20 minutes.  Even at this kind of fluctuation,
the FGD system operated with high and stable SO2 removal.

3.2.3  Scrubber scaling

One of the important requirements  in  slurry scrubbing is scale-free
operation of scrubbers.   In the early  stage of development,  some
scaling  problems were experienced  due to  poor  operating procedures.
According to MHI's experience, the following considerations  are vital
in preventing scaling.

 1)    Addition of gypsum "seed" crystals  to  provide  seed sites  for
       calcium sulfate  crystallization.

2)    Application of relatively  high  L./G and  selection of  a scrubber
       with simple  structure,  to maintain chemical  conditions  of the
       scrubbing slurry uniform, and to  keep all of the internals wet
       and clean.

The above method  for preventing scaling was proved to be a good way,
even in units for sintering plants where the  sulfite oxidation in the
 scrubber reaches  as high as  100%,  due to high QZ concentration of
the  incoming flue  gas.

 3.3   Mist Separation

The scrubber outlet mist loading ranges from 4 to 8 gr/acf.

Using vertical chevrons,  more than 99% of the  entrained mist is
separated from the gas.   Fresh water  and a part of the  scrubbing
slurry is used for  washing the eliminator  surfaces.   Continuous
washing  and intense intermittent washing  were experienced,  both
giving  excellent service.

                              256

-------
3.4  Gas Reheating

Most of the installations reheat  the  scrubber gas directly  by burning
low-sulfur  fuel oil.  Installations for  sintering plants burn combus-
tible flue gases from steel mills.

One installation for small industrial boilers do  not reheat at all.
The reheat temperature  is  rather high in Japan,  usually in  the range
of 250  to 285 °F.

3.5  Absorbent Preparation

Both for lime and limestone,  recycle water is used for preparation of
tiie absorbent slurry.   In  lime slaking,  the grits are ground in a wet
mill and carried along  with the  slaked lime into  the scrubber.  Lime-
stone is  ground in a wet mill, but recent installations  are using fine
limestone powder.

3.6  Lime vs.  Limestone

Recent installations tend to  use  limestone rather than lime.   According
to MHI's experience,  the difference between limestone  and lime can be
outlined  as follows.

 1)   Lime is more  reactive,  and therefore lime scrubbing is much
     easier  to  control.

 2)   Sulfite  oxidation  rate is  much higher in limestone scrubbing.

 3)   Absorbent preparation is much easier with limestone powder.

     In  lime slaking,  the substitution reaction  between lime and the
     dissolved  magnesium sulfate in  recycled water may  cause calcium
     sulfate scaling.   Here again,,  the  addition  of  gypsum "seed"
     crystals in  the  slaker is very effective  for scale prevention.

 4)   Limestone is more economical,  safer and easier to handle.

 3.7 System Reliability

 All of  the  installations are  running  with high availability.
 Availability data on  some installations are shown on Table  3.
                                257

-------
4.    APPLICATION ON COAL-FIRED BOILERS

      Since most  of the utilities in Japan are using  oil in their  boilers,
all of the  MHI  FGD System installations for boilers are  treating oil-
fired gases.  If an electrostatic precipitator  is  installed before the
system,  main differences  between oil  and coal-burnt gases would be
in compositions of dust and gas.  MHI has been conducting  field pilot
tests on a coal-fired boiler in  Japan.   Test results obtained up  to
now indicates that the MHI gypsum recovery  process is  applicable
on  coal-fired boilers without problem.   On the other hand,  applica-
bility of the system to  non-recovery (throwaway) processes  has been
tested also.    If  a non-recovery process is favorable,  which is the
case  in the United States  (Reference 4), the MHI FGD System  can
meet this  requirement, and can be simplified as follows.

1)    Precooler can be eliminated.

2)    Absorbent preparation section can be simplified.

3)    Gypsum recovery section can be  eliminated.

4)    Sulfite oxidizer capacity is minimized  to  gypsum "seed" pro-
      duction.

      For  application in the United States, tests  on dolomitic lime/
limestone  scrubbing are also in progress.    Through these  tests,
influences of chloride and other impurities on the system have  been
investigated.   Test results up to now  indicates that the  system is
quite feasible.    Moreover, studies on materials of construction,
process  equipment, process economics are being continued  for  further
development of the  system.
                                258

-------
5.    MHI SIMULTANEOUS  NOV/SC>  REMOVAL PROCESS
                              •"•    X

     MHI's research for the  NOX removal process has  been aimed
at  the development of reasonable  and economical systems which can
treat flue gases ranging from clean  flue gases  of LNG-firing boilers
to  dirty flue gases of high  sulfur fuel-firing boilers,  sintering  plants
(steel mills),  etc.   MHI has  developed both dry  and  wet processes of
NOx removal  and has already obtained  orders for the NOX removal
units.   For dirty flue  gases requiring both NOX  and  SOX removal,
MHI is  considering  that the use  of the  wet process is most reasonable
for them, in view of its capability of  simultaneous NOX/SOX" removal
in  a single  scrubber and from the standpoints of economics, operation
and maintenance,  installation area,  etc.

     Recently,  MHI has successfully finished its preliminary stage
of development for  a simultaneous NOX/SO  removal process.   By
adding an 03  generator  and an  ammonia recovery section to the MHI
FGD System,  NOX and  SO  in the flue  gas are removed simultane-
ously in a single  scrubber, at high  efficiencies.   The scrubber is
identical to those used  in the MHI FGD System,  lime or limestone
can be used as absorbent,  and  by-products are gypsum and ammonia.

5.1   Process  Development

     After  a  bench  scale test (125  scfm capacity) for about a  year,
a  pilot plant test  (1,250 scfm capacity) was conducted from July, 1975.
The flue gas from an  oil-fired  boiler  contained 180  to 220 ppm NO^
800 to 1,100 ppm SOX,  and 0.04 to  0.05  gr/scf  dust.   The obtained
efficiencies were  80 to 90% NOX  removal  and over 95% SOX removal.
Since the result was satisfactory, a prototype  scale  unit is planned
in the  near future.

5.2   Process  Description

      A simplified flow scheme of the process  is shown in  Figure  3.

      The flue gas is first  cooled and dust collected in  the precooler
 and after the oxidation of  NO to  NO2 -by means of adding stoichio-
 metric  amount of O,,  NO_ and SO_  are removed simultaneously in
 the SO_/NO  scrubber.    The absorbent  slurry,  either  slaked  lime
 slurry or limestone slurry,  containing water soluble catalyst is fed
 to the scrubber.
                               259

-------
      Since the  scrubber outlet slurry  contains the N-S  compounds
(defined  as a mixture of nitrogen and  sulfur base compounds) as
dissolved components,  a portion of filtrate  from the gypsum re-
covery  section  is sent to the ammonia recovery  section where it is
decomposed to  yield gypsum and ammonia.

      Description of the remaining section of the process  is identical
to those of the MHI FGD System.

5.3   Features

      The outstanding features of the process can be summarized
as follows:

1)    Most of the main equipment  are  exactly the same as  that of the
      MHI  FGD System and their reliabilities are already known to be
      high  through actual operation.

2)    Since NO  and SO  are removed simultaneously in one  system,
      the process has  advantages in installation area, operation and
      maintenance,  economics,  etc.

3)    As the reductive  property of SO, is used in absorption, the
      process does  not require any NO  reducing agent.  (In the
      dry process,  ammonia is added  as  the reducing agent)

4)    The process is a recovery system which produces gypsum and
      ammonia as by-product.

5)    The process can  be applied to the existing MHI FGD System by
      simple reconstruction.
                              260

-------
FEASIBILITY  TESTS
FOR OIL-FIRED BOILER
3,300 scfm PILOT UNIT
                                             DATA  FEED BACK  FOR IMPROVEMENT
                                                A
                                      COMMERCIAL SCALE UNIT
                                      FOR COPPER SMELTER
                                                                                                     t\
                                        COMMERCIAL  SCALE UNITS
                                        FOR SINTERING PLANTS
                                               IT
                                                 IT
                                   HIGH S02 SCRUBBING TESTS
                                      TESTS ON THE INFLUENCE OF  IMPURITIES
APCS RESEARCH CENTER
PERMANENT  LAB. UNITS
1,200 scfm  ,125 scfm
(HIROSHIMA TECHNICAL  INSTITUTE)
SCALE-UP TESTS
FOR OIL-FIRED BOILER
62,000 scfm
PROTOTYPE UNIT
                          FEASIBILITY TESTS
                         FOR COAL-FIRED BOILER
                         1,200scfm FIELD PILOT UNIT
COMMERCIAL SCALE UNITS
FOR OIL-FIRED BOILERS
  Figure 1.  MHI  FGD  System Development
                                                   COMMERCIAL SCALE SYSTEM
                                                   DEVELOPMENT COMPLETED
                                                   FOR COAL-FIRED  BOILER
                                                   ( THROW-AWAY AND
                                                   V GYPSUM RECOVERY  SYSTEMS

-------
  FLUE GAS FROM
EMISSION SOURCE
                               FLUE GAS TO
                               CHIMNEY
                                      MIST
                                      ELIMINATOR
            S02
            SCRUBBER
PRECOOLER
REHEATER
   LIME/
  LIMESTONE
                                                      GYPSUM
                                                      RECOVERY
                           BY-PRODUC
WATER
TREATMENT
ABSORBENT
PREPARATION
OXIDIZER
       SLUDGE) (PURGE
                             GYPSUM/
     Figure 2.  MH! FGD System CMHI Lime/Limestone-Gypsum  Recovery  Process)

-------
FLUE GAS  FROM
EMISSION  SOURCE
OT
GENERATOR





c*r\ 1
PRECOOLER
S02/NO?
 SCRUBBER
                   WATER
                   TREATMENT
MIST
ELIMINATOR
              ABSORBENT
              PREPARATION
                 GYPSUM
                 RECOVERY
                                                         AMMONIA
                                                         RECOVERY
                  IF
                  D
                  D
                  D
                  D
               LIME/
              LIMESTONE
                   Figures.  MHI  Simultaneous  N0x/50x Removal  Process
             I CD CD CD
                                                             D
                                                             D
                                                             a
                                                             D
FLUE GAS TO
CHIMNEY
BY-PRODUCT
 GYPSUM
                                        f-PRODUCl
                                       .AMMONIA.
                    o

-------
                            Table  1.   DELIVERY  RECORD OF  MHI FGD SYSTEM
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Owner
Kansai Electric Power Co.
Onahama Refining Co.
Yoshino Gypsum Co.
Kawasaki Steel Corp.
Kanaai Electric Power Co.
Tokyo Electric Power Co.
Tohoku Electric Power Co.
Kyushu Electric Power Co.
Kawasaki Steel Corp.
Kansai Electric Power Co.
Teijin Limited
Kawasaki Steel Corp.
Kawasaki Steel Corp.
Kuraray Co.
Mizushima Joint Thermal
Location
Amagaaaki
Onahama
Tokyo
Chiba
Kainan
Yokosuka
Hachinohe
Kanda
Mizushima
Amagasaki
Ehime
Mizushima
Chiba
Saijo
Mizushima
Gas Source
156 MW Unit
Copper Smelter
Industrial Boiler
Sintering Plant
600 MW Unit
265 MW Unit
250 MW Unit
375 MW Unit
Sintering Plant
156 MW Unit
Industrial Boiler
Sintering Plant
Sintering Plant
Industrial Boiler
156 MW Unit
Capacity,
scfm
62,000
57,000
8,000
74,000
248,000
248,000
235,000
341,000
465,000
232,000
167, 000
558,000
260,000
131,000
379,000
/IB
Absorbent
Lime
Lime
Slaked Lime
Lime
Lime
Limestone
Lime
Lime
Lime
Lime
Lime
Lime
Lime
Lime
Lime
UJ. J «II1. , 1 /
Start-Up
1972
1972
1973
1973
1974
1974
1974
1974
1974
1975
1975
1975
1975
1975
1975
     Power Co.
16.
Toyobo Co.
Iwakuni
Industrial Boiler
124,000
Lime
1975

-------
17.
18.
19.
20.
21.
22.
23.
24.
t-o
a 25.
26.
Niigata
Power
Chubu
Chubu
Kyushu
Kyushu
Kyushu
Kyushu
Tohoku
Owner

Joint Thermal
Co.
Electric
Electric
Electric
Electric
Electric
Electric
Electric
Kawasaki Steel
Power Co.
Power Co.
Power Co.
Power Co.
Power Co.
Power Co.
Power Co.
Corp.
Kashima-minami Joint
Thermal Power
27.
28.
Kansai
Niigata
Electric
Co.
Power Co.
Joint Thermal
Location
Niigata
Owase
Owase
Karatsu
Karatsu
Ainoura
Ainoura
Niigata
Chiba
Kashima

Amagasaki
Niigata
Gas Source
350
375
375
500
375
375
500
600
MW
MW
MW
MW
MW
MW
MW
MW
Sintering
160

156
350
MW

MW
MW
Unit
Unit
Unit
Unit
Unit
Unit
Unit
Unit
Plant
Unit

Unit
Unit
Capacity,
scfrn
328,
744,
744,
353,
452,
452,
452,
260,
446,
267,

294,
328,
000
000
000
000
000
000
000
000
000
000

000
000
Km
Absorbent
Limestone
Lime
Lime
Limestone
Limestone
Limestone
Limestone
Limestone
Lime
Limestone

Lime
Limestone
o! Jan. . — WT-o
Start-TJp
1975
1976
1976
1976
1976
1976
1976
1976
1976
1976

1976
1977
      Power  Co.

29.    Sakata  Joint Thermal
      Power  Co.

30.    Sakata  Joint Thermal
      Power  Co.

31.    Kawasaki Steel Corp.

32.    Chugoku Electric Power Co.
Sakata
Sakata
350 MW  Unit
350 MW  Unit
Mizushima     Sintering  Plant

Shimonoseki    400 MW Unit
682,000      Limestone


682,000      Limestone


465,000      Lime

744,000      Limestone
1977


1977


1977

1977

-------
Table 2.    OPERATING DATA OF MHI FGD SYSTEM
ITEM """ """ — — 	 SOURCE
Inlet S02 ppm (Volume)
SO2 Removal %
Absorbent Utilization %
in Scrubber
Scrubber Control pH
Oxidation in Scrubber %
Slurry Concentration % (Weight)
Inlet Gas Composition
O2 % (Volume)
CO2 % (Volume)
H2O % (Volume)
Dust Grains /scf




Remarks




Oil-Fired Boilers
500 - 1,500
90 - 99
90 - 99

/6.5 - 7.0 (Lime)
M.O - 5.8 (Limestone)
30 - 100
10 - 14
(Pre -cooler Inlet)
3-6
9 - 12
9-12
0.0085 - 0.0175
(ESP, Outlet)
0.04 - 0.08
(Boiler A/H Outlet
One Scrubber System
, 12 Units
Two Scrubber System
1 Unit



Sintering Plants
(Steel Mills)
600 - 1,200
90 - 97
95 - 99

6.5 .- 7.5
70 - 100
10 - 14
(Pre-cooler Inlet)
12 - 17
4-8
6-12
0.0175 - 0.04
(ESP Outlet)


Inlet Cl : 20 -SOppnr
(as HC1)
Inlet Oily Matter:
-0.025 gr/scf
Dust Components:
Fe,Mn, Si, Pb,K,Na,
Ca, Mg, Al, Zn, Cu, etc.
Copper Smelter
20,000 - 29,000
90 +
- 99

,3-3.5 (T)
^9 - 10.5 ©
0-20
10 - 13
(Scrubber Inlet)
6
10
23
0.06



Two Scrubber System
(I) No. 1 Scrubber
© No. 2 Scrubber



i

-------
                                Table  3.    AVAILABILITY OF MHI FOP SYSTEM
                                                                                          As of April.  1975
Owner
(Location)
1. Kansai Electric Power
Co. (Amagasaki)
2. Onahama Refining Co.
(Onahama)
3. Yoshino Gypsum Co.
(Tokyo)
4. Kawasaki Steel Corp.
(Chiba)
5. Kansai Electric Power
Co. (Kainan)
6. Tokyo Electric Power
Co. (Yokosuka)
7. Tohoku Electric Power
Co. (Hachinohe)
8. Kyushu Electric Power
Co. (Kanda)
Start-Up
(exclusive
of test run
Mar., 1972
Nov. , 1972
Aug., 1973
Nov. , 1973
Dec., 1973
Apr., 1974
May. , 1974
Dec., 1974

Emission
Source
Operating
Mrs (A)
16,000
24, 120
12,048
11,520
8,760
8,730
4,800
2,880

FGD
System
Operating
Mrs (B)
14, 600
23,780
11,328
11,456
8,760
7,886
4,080
2,880

Availability
Q3)/tA) x 100
%
91.5
98. 6
94.0
99.4
100.0
90. 3
85.0
100.0

Remarks
(Capacity: scfm x SO£ ppm)
Boiler stop t Apr. "74-Jan. '75
(62,000 x 800)
(57,000 x 23,000)
(8,000 x 1,500)
(74,000 x 500)
Boiler stop : Dec. '74-Mar. '75
(248, 000 x 300)
Boiler stop '. May '74
(235,000 x 800)
Boiler stop : Dec. '74
(235,000 x 800)
(341,000 x 500)

I J
0-.

-------
References

1.    Uno, T.,  M.  Atsukawa, K.  Muramatsu,  834 - 849, Proceedings
      of Second International  Lime/Lime stone Wet-Scrubbing Symposium,
      New Orleans,  1971.

2.    Atsukawa,  M., K.  Matsumoto,  N. Shinoda, M.  Kirnishima,
      Y.  Ouchi,  "Removal  of SO_  from Flue Gases by Wet Processes",
      M.H.I. Technical Review, February,  1974.

3.    The data extracted from  "Heavy Industries News",  published in
      Japan,  December,  1975.

4.    Slack,  A.V.,  "Scrubber survey: a lime /lime stone trend",
      Electrical  World,  October 1, 1974.
 Figures

 1.    MHI FGD System  Development

 2.    MHI FGD System  (MHI Lime/Lime stone-Gyp sum Recovery Process)

 3.    MHI Simultaneous  NO  /SO  Removal Process
                          x   x
 Tables
 1.    DELIVERY RECORD OF MHI FGD SYSTEM

 2.    OPERATING DATA OF MHI FGD SYSTEM

 3.    AVAILABILITY OF MHI FGD SYSTEM
                             268

-------
                STATUS OF FLUE GAS DESULFURIZATION
             USING ALKALINE FLY ASH FROM WESTERN COALS
                 Harvey M. Ness, Research Chemist
             Everett A. Sondreal, Research Supervisor
                Philip H. Tufte, Chemical Engineer

        U.S. Energy Research and Development Administration
                Grand Forks Energy Research Center
                     Grand Forks, North Dakota
ABSTRACT

     It is the purpose of this paper to report on the status of flue
gas desulfurization utilizing the alkali in fly ashes derived from
Western coals.  Most of the utility scrubbers operating in the Western
United States have been installed for particulate removal.  These
particulate scrubbers, operating without an added reagent such as
lime or limestone, have been found to remove appreciable amounts of
sulfur dioxide from the flue gas and to experience sulfate scaling.
Sulfur dioxide removal occurs because of the inherent alkalinity
of the fly ashes produced from Western coals.  The deliberate use of
fly ash alkalinity for flue gas desulfurization has been investigated
in four separate pilot-scale studies; two commercial-scale installations
have been built and a third is being constructed.  The properties of
Western coals, their sulfur dioxide emissions, and the design and
performance of selected commercial and pilot scrubbers are discussed
in this paper.  An appendix lists data on operating scrubbers.
                               269

-------
                       STATUS OF FLUE GAS DESULFURIZATION
                    USING ALKALINE FLY ASH FROM WESTERN COALS
  INTRODUCTION
     The Western reserve base for measured and indicated coal in place, as
defined by the U.S. Bureau of Mines, totals 2l6 billion tons1.  Future plans for
mine expansion will increase capacity in the West above 200 million tons per
annum by 19832.  This figure does not include tentative plans for a number of
coal gasification plants.  Considering these projections, stack gas cleaning
technology for burning Western coals will assume much greater importance in the
future than at present.

     The Western coal reserves include lignite, subbituminous, and bituminous
coal, with the lower rank coals predominating.  An important property of almos^
all Western coals is that they contain far less sulfur than the 2 to 3 pet and
higher typical of Eastern and Central coals.  Unfortunately, sulfur content in
Western coals (averaging 0.7 pet) is not generally low enough to meet new source
emission standards.  Typically, an average sulfur dioxide removal of 30 to ^0
pet is required to meet the Federal standard of 1.2 lb S02/MM Btu, and higher
removals are required to meet more stringent State and local standards.  Since
sulfur oxide emission standards are based on heat released, variations in
heating value according to rank have an important effect on the coal sulfur
content that is equivalent to the emission standard, as shown in Table 1.

            TABLE 1   COAL SULFUR CONTENT EQUAL TO EMISSION STANDARDS
Coal
North Dakota
lignite
Higher
heating
value ,
Btu/lb
6,800
Coal sulfur
equal to the
Federal
standard of
1.2 lb S02/MM
Btu, pet
O.Ul
    Montana
      subbituminous                      8,600                       .52

    Arizona
      (Black Mesa)
      bituminous	11,000	.66	

     While the 0.7 pet average sulfur content in Western coals does not
satisfy the Federal standard, it does make stack gas cleaning potentially easier
to achieve and less costly than for high-sulfur Eastern coals, provided that
emission standards are not raised to cancel out the advantage.  In addition,
the alkaline nature of Western fly ash provides an opportunity for design
innovations in flue gas desulfurization.
                                   270

-------
     Ash content in Western coals can vary greatly, with the U to 20 pet shown
in Table 2,  representative of the overall range.  The ash content varies
significantly between mines, and even between locations within mines.  The
quantity of  fly ash in stack gas depends on boiler design as well as on coal ash
content.  For a pulverized coal-fired boiler, fly ash leaving the boiler represents
approximately 80 pet of the coal ash.  Resulting particulate loadings are
typically 2  to 10 gr/scfd.  For a cyclone-fired boiler, fly ash leaving the
boiler is approximately ho pet of the coal ash.  The corresponding particulate
loadings are 1 to 5 gr/scfd.

     A very  important characteristic of many Western coal ashes is their high
content of sodium oxide, magnesium oxide, and particularly calcium oxide (see
Table 2).  The alkali content tends to be highest in lignite, the lowest rank
coal, and progressively less prevalent in the subbituminous and bituminous
coals.  Variations in alkali content are also influenced by the minerology of
the overburden and the course of ground water movement.  Alkali content in
Western coal ash varies from under 10 to over 50 pet, with important variations
occurring within individual mines.

     A guideline in assessing the importance of the alkali in Western coal is
the ratio of the alkali to coal sulfur.   For a coal containing 7-5 pet ash and
20 pet alkali in the ash, the total alkali is chemically equivalent to slightly
more than 120 pet of a 0.7 pet sulfur content.   For some lignites, the alkali/sulfur
ratio can be several hundred percent.  Thus there is ample alkali to interact
importantly  with sulfur oxides in a wet  scrubber in burning Western coals.

Required Removal Efficiencies

     Required removal efficiencies are determined by coal sulfur content, sulfur
retention on ash during combustion, and  the emission limits established by law.
As stated previously, the average 0.7 pet sulfur content of Western coals does
not permit such coal to be burned without flue gas desulfurization under the
Federal emission standard of 1.2 Ib S02/MM Btu.  Retention of sulfur oxides on
ash during combustion may lower the sulfur dioxide emission by 10 to ho pet for
lignites , but this effect does not guarantee compliance with the Federal
standard.  Considering more stringent state and local standards, there are no
coals that will meet the standards of Clark County, Nevada (0.15 Ib/MM Btu), New
Mexico (0.3h Ib/MM Btu), Nevada (O.hO Ib/MM Btu), or of Colorado after 1980
(0.35 Ib/MM  Btu); it is doubtful, that any would reliably meet the Arizona
statewide standards (0.8 Ib/MM Btu).  The removal efficiencies required to meet
the more stringent standards are shown in Figure 1 for a Western subbituminous
coal.  At  the 0.7 pet coal sulfur level, the required sulfur dioxide removal is
increased from about 30 pet to meet the  Federal standard to 90 pet to meet the
Clark County, Nevada standard.

     Capital and operating costs for stack gas cleaning must be expected to rise
steeply as high percentage removals are  required for substances that are
initially present at low concentrations.  For an approximate estimate based on
principles of engineering design, it can be assumed that equipment size and
power requirements for wet scrubbing will increase in proportion to the logarithm
of one over  the required exit concentration (size and power a  log i exit), and
                                   271

-------
    100
c

o.
O
2
Ul
a:

a
UJ
oc

ID
o
UJ
ac
     80  -
     60  -
40  -
     20   -
       Subbituminous coal

         8,600 btu/lb

            500 mw
                                            .4% sulfur  in coal
            Clark County  New

             Nevada   Mexico Nevada
                 .2       .4       .6       .8      1.0      1.2      1.4


                            S02  EMISSION  STANDARD,  Ib/mm btu
                                                                       1.6
1.8
  Figure  I.  -   Removal efficiencies required  for stack gas  cleaning of Western  coai.

-------
TABLE 2   SELECTED ANALYSES OF ASH IN WESTERN COALS'
Coal 	
State 	

Sample avg. . . .
Lignite
North Dakota
212
Subb ituminous
Montana
125
Wyoming
Big Horn
12
Arizona
Black Mesa
1
New Mexico
Nava j o
2
' Bituminous
New Mexico
McKinley
1
Colorado
Hawks Nest
3
              Ash, percent of coal




6.2         9.3       1+.8       7.5




       Oxide constituents, percent of ash
20.2
8.0
SiOp 	
AloOo 	
FeoO-a 	
TiOo 	
PoOs 	
CaO 	
Ma;0 	
Na20. 	
K20 	
303 	

19.7
11 . 1
9.1
.1+
.3
2k. 6
6.9
6.5
.k
19-5

35.5
18.7
7.8
.7
.3
15.6
1+.1+
1.7
.1+
13.1+

27. k
12.7
13.9
.6
.5
16.6
5.5
2.2
.5
17.0

1+2.0
18.1
5.7
.8
.6
17.8
2.1+
1.1+
.3
8.2

55.6
26.2
6.1
.6
.5
3.9
.8
1.5
.6
3.2

5H.7
21.6
7.0
1.0
.0
6.5
1.2
1.6
.8
5.8

1+1+.8
28.3
11.5
.8
.7
5.6
1.9
.6
.5
l+.O


-------
capital cost will increase as the 0.6 pover of size.  Under these assumptions,
equipment size would triple between 50 pet and 90 pet removal, and would double
again at 99 pet removal.  Capital cost would first double and then rise by a
further 50 pet for the same increases in removal.  These figures are hypothetical
and they make no allowance for improved design or use of more effective reagents.
However, they are close enough to reality to validly demonstrate that high costs
will have to be paid to achieve improved control of stack gas emissions.

     The following sections provide up-to-date information on the capital
investment and operating experiences of the Western coal-burning utilites
experiencing sulfur dioxide removal in particulate scrubbers.  It also provides
up-dated information on utility operated scrubbers that are deliberately utilizing
fly ash alkali for sulfur dioxide removal, and also, on current research results
on fly ash alkali utilization and scaling problems.

Arizona Public Service Company, Four Corners Plant^'

     The Four Corners Plant at Farmington, NM has three pc-fired boilers (two
 175 MW, and one 225 MW) that are equipped with Chemico venturi scrubbers for
particulate removal, with two scrubber modules on each boiler.  Two additional
boilers of 755 MW each are equipped with electrostatic precipitators (ESP's).
The first scrubber began operation in December 1971.  Total cost of the scrubber
system was $30 million, or $52 per kw.

     The subbituminous coal burned at the Four Corners Plant, supplied by the
Navajo mine, has as its outstanding characteristic, an unusually high ash content
of nominally 22 pet.  This results in a high dust loading in the flue gas
leaving the boiler, 12 gr/scfd.  Initial operation of the plant using mechanical
collectors for fly ash removal resulted in dense plumes from the stacks, which
dispersed to restrict visibility over a large surrounding region.  Since
scrubbers and ESP's have been operating, the plumes are greatly improved,
although still visible.

     The Navajo coal averages 0.68 pet sulfur, which produces a sulfur dioxide
content of approximately 650 ppm.  The fly ash alkalinity is low for a Western
coal, having only 5 pet CaO in the ash.  However, the amount of calcium entering
the scrubber system is still relatively large, owing to the high percentage
of ash.  The calcium is chemically equivalent to approximately 75 pet of the
coal sulfur content; and total alkalinity, including small amounts of Na20 and
MgO, exceeds 100 pet of sulfur equivalence.

     A detailed description of the Four Corners scrubber installation is given
in the Appendix and in Figure 2.   Flue gas from the air heaters enters the
venturi and then passes consecutively through a mist eliminator, a wet ID fan,
another mist eliminator, and a steam reheater.  However, because of erosion
problems, the reheaters were removed and the units are operated as wet stacks.
There is no bypass.  Turndown is 50 pet.
                                     274

-------
Flue gas from
 air heaters
                                                          To  stack
                            Water spray
                                         Mist
                                      eliminators
                   Distribution
                      tank
                                                       Mist eliminators
                                              Lime
                                                      Make up
                                                        water
                                               Liquid transfer
                                                   tank
  figure 2.
 rubbers.
-  Simplified  flow diagram for the Four Corners fly ash
                              275

-------
     Scrubber liquor is recycled back to the venturi and mist  eliminator.   The
system vas modified so that a bleed stream is taken from the recycle pump  and
sent to the thickener.  Before the modification, blowdown was  drained by
gravity to the thickener.  However, the gravity flow line was  prone  to plugging.
Lime is now added to the overflow launders of the thickener instead  of the
centerwell, where some lime was lost by settling.  The thickener and the fly
ash transfer tank are at the same elevation and a line between them  is used for
water level control.  The thickener underflow is pumped to a fly ash transfer
tank and then pumped to a settling pond.  The sludge is reported to  have good
settling properties.  The pond is periodically dredged and disposed  of in  the
mine.

     The key operating variables are a liquid-to-gas ratio (L/G) of  18 gal/1000
scf, a pressure drop across the scrubber tower of 18-20 inches of water, and a
pressure drop across the entire scrubbing system of about 28 inches  of water.
The venturi recycle liquid has a pH of 3.2 to 3-5.  The pH of  the thickener
overflow liquid is maintained at about 7-5 by the addition of  lime.

     The operation of the venturi scrubbers has been satisfactory from the
standpoint of meeting the particulate removal goal of 99.2 pet.  Past measurements
of  sulfur dioxide removal, without supplemental lime, are reported to be in the
range of 30 to 35 pet.  The present lime addition rate is equivalent to about
7 pet of the sulfur dioxide in the inlet flue gas.  Therefore, it would be
expected to improve the sulfur dioxide removal by about 5 pet.  The  actual
sulfur dioxide removal efficiency has not yet been measured.  However, a long-
range testing program with the primary purpose of reducing sulfur dioxide
emissions, will be implemented by Arizona Public Service Company, since the
existing level of sulfur dioxide emissions will not meet new sulfur  dioxide
limits.  A secondary purpose of the test program will be to reduce scaling.

     The testing program to be implemented will modify one scrubber  system to
test several variables.  The variables to be tested are various liquid-to-gas
ratios and suspended solids.  Additional spray nozzles will be installed in
the sorubber and an additional rubber-lined recycle pump will  be added. The
sulfur dioxide removal efficiency will be investigated as a function of L/G.
The level of suspended solids will be maintained at 6 pet to provide seed
crystals for precipitation of gypsum.  Lime will be added for  pH control.

     Scrubber operating costs are not separated from the plant operating cost,
and hence, are not available.  The major operating requirements are  electrical
power equal to 3 to k pet of generating capacity, an estimated water usage
of 5-9 acre-ft/MW/yr, manpower including 8 operators plus maintenance and
supervision, and Ik tons of lime per day for control of pH.

     Availability for the scrubber system is currently estimated at  about  80 pet.  •
This is the fraction of the time that a boiler is operating or could be operating
that the scrubber modules are also operative.  Since there is  no bypass, this
level of availability involves appreciable loss in power generation.  With two
scrubbers per boiler, lost generation can involve either reduction in load when
one scrubber is inoperative or complete boiler shutdown when both scrubbers are
inoperative.

                                     276

-------
    The operating problems  have "been scale and solids buildup in the scrubber
    would break off and clog the gravity-feed blovdown line.  A severe erosion
problem in the  venturi opposite the plumb bell vas resolved by installing  silica
carbide bricks.   The bricks  are reported to be very effective in resisting
erosion.  A problem of erosion on the tangential nozzles resulted in by-passing
dirty flue gas  to the wet fan.  This, in turn, resulted in severe fan erosion.
The problem was resolved by  adding a stainless steel wear plate to  the tangential
nozzles.  Additional problems are corrosion and leakage in the recirculatory
lines and vessels and reheaters, vhere coatings on carbon steel has failed, on
the mist eliminators and on  the ducting leading into the stack.  The blowdovn
line plugging has been resolved by using a forced bleed from the recycle pump.
A nanually-operated valve has been inserted in the old existing gravity blovdown
line for emergencies.  The reheater corrosion problem was resolved  by removing
the reheaters and using a wet stack.  However, the stack liner and  mortor  were
not acid resistant and, consequently, had to be replaced with the proper liner.

    Scaling has occurred on most wetted surfaces, and it is not yet under
control.  Control measures include the use of an appreciable amount of blowdown
(open-loop operation), an increase in the percentage of fly ash solids in  the
reeirculating scrubbing liquor (from 2 to 6 pet), and addition of a new lime add
System to maintain the pH at 7-5 in the thickener.  It is not clear why this pH
aijustment insures improved  control of scaling, since in a system where the
state of oxidation is high,  with most dissolved sulfur present as sulfate,
calcium sulfate would not be expected to be precipitated in the thickener  by the
rise in pH?.

    The measures being investigated for scale control at Four Corners, if
successful,  should find wide application in scrubbing applications  involving
lew-sulfur Western coals.

 SftSD FORKS  ENERGY RESEARCH  CENTER

    The Energy Research and Development Administration's Grand Forks  Energy
 tesearch Center (GFERC) has  investigated the fly ash alkali sulfur  dioxide
 scrubbing  system since 1971.  Testing has been performed on a 130 scfm pilot
 scrubber.  The  principal objectives have been to determine sulfur dioxide
 KBOTal efficiencies and calcium sulfate scaling rates as a function of sulfur
 dioxide level,  fly ash add rate, alkali in the fly ash, supplementary  lime
 requirements, level of recirculated suspended solids, liquid to gas ratio,
 SBunt of makeup water and total dissolved solids.  Past results have  been
 l&lished  in three previous  papers"'^'10.  Research is continuing under funding
 from the Environmental Protection Agency, investigating the effect  of  high
 Jsrels of total dissolved solids on sulfur dioxide removal efficiencies and
 tflcium sulfate scaling.

    The present GFERC scrubber (Figure 3) is a 130 scfm flooded disk  venturi
       by an  absorption tower containing conical "rain and drain" trays.
 &6Bsure drop across the scrubber can be controlled by adjusting the height of
 tie flood disk.  The conical trays were installed as a modification to increase
 
-------
      S(>2  injection
        Gas furnace
NJ
^1
00
                              Water  cooled
                             heat  exchanger
            Floating  weir
                                              Rain and
                                            drain tower
                                                                                   Drip leg


: rj
To stack
,
b


                                                                                                   ID fan
                                                                                          Fly ash

^K
.>


r
o

]
o

r* —


^ L
o
*v. °
r ;
o
0 .,
                                                                    Mix tanks
               Fiaure  3. -   Pilot  olant scrubber.  Grand Forks  Enerav Research  Center.

-------
removal levels  so  that the sulfur dioxide removal observed would Toe a function
of the solution characteristics only and not of scrubber  design.   Installation
of the conical  "rain and drain" trays increased the removal efficiency by  5 pet,
from 83.3  to  88.1  pet under identical operating conditions.

     The GFERC  scrubber  system is "closed loop."  Recirculating scrubber liquor
is lost from  the system  only as mist, which is equivalent to  about 0.8 acre-
ft/MW/yr,  or  as liquor in sludge.  Efficient mist elimination is accomplished by
passing gas through both a cyclone and a stainless steel  wire mesh.  Water lost
by evaporation  is  replaced, but mist loss is not replaced during the course of a
one-week test.   The scrubber liquor is returned to a series of two fly ash mix
tanks equipped  with overflow weirs.  The overflow from the second  mix tank flows
to a settling tank where calcium sulfate precipitates and unreacted fly ash ±s
allowed to settle. A floating overflow weir in the settling tank provides  the
scrubbing  liquid to the  flooded disk.

     Early experiments at GFERC indicated that a large increase in total dissolved
solids, primarily  sodium and magnesium sulfates, during the approach to steady
state operating conditions significantly increased the scrubbing efficiency.
Since the  fly ash  derived from some Western coals are known to contain significant
amounts of soluble sodium and magnesium, it is probable that  high  concentrations
of these  species will result after long-term operation of a full-scale scrubber
employing  the fly  ash alkali scrubbing process.  Current  experiments at GFERC
are designed to investigate the properties of scrubber solutions that are  high
in sodium  (0.5  to  10 pet) and magnesium (0.5 to 10 pet).   The objectives of the
current tests are  to determine sulfur dioxide removal and scaling  rate using a
solution concentrated in sodium and magnesium and low in suspended solids  (high
levels  of  suspended solids, 6 to 12 pet, are common practice  for  scale control
in many western scrubbers).  The fly ash used in these tests  contained high
sodium  and magnesium and was produced by pc-firing of Beulah, North Dakota
lignite.   Scrubber operating conditions kept constant for all test runs were:
inlet  sulfur dioxide level of about 8UO ppm (a typical Western lignite containing
about  0.8  pet sulfur),  inlet flue  gas temperature of 350° F,  liquid temperature
of about  120° F, absorber tower pressure drop of about 13 inches  of water.

      Scaling rates reported represent the rate of weight increase in  grams per
hour  observed in a 3 ft  k inch length of 1/2 inch I.D. pipe  in the return  line
from  the  scrubber to the mix tanks.  The test position chosen was a point  of
maximum scaling, and trends in the values observed were found to be well correlated
with  operating variables.

      Tests were performed at liquid to gas ratios of 23, U5,  and 75 gal/1000
scf.   The CaO/SC>2 stoichiometric  ratio was maintained at 1.2, sodium concentration
at about  3.0 pet,  magnesium concentration at 1 to 2 pet.  The pH of the liquor
pumped to the absorber tower varied  from  5-0 to  5-5.  Previous experiments
indicated only a marginal effect when L/G was increased.  However, under the
conditions of high sodium and magnesium, removal efficiencies were affected very
significantly.  The removal efficiencies and fly ash alkali utilizations are
tabulated in Table 3.
                                        279

-------
       TABLE 3   SULFUR DIOXIDE REMOVAL EFFICIENCIES AND FLY ASH
             ALKALI UTILIZATION AS A FUNCTION OF L/G.  CaO/SOg = 1:2
L/G
23
H5
75
Removal Efficiency (pet)
81.0
88.2
98.1
Alkali Utilization (pet)
66
72
80
     The stoichiometric ratios of calcium oxide to sulfur dioxide investigated
were 0.6, 1.2 and 2.0.  These ratios correspond to particulate loadings of  2.0
gr/scf, U.O gr/scf and 6.7 gr/scf.  Operating conditions were as described
above, with an L/G of ^5.  The sulfur dioxide removal efficiency, fly ash alkali
utilization, and scaling rate are tabulated in Table U.  Removal efficiencies
are shown in Figure 6.

        TABLE k   SULFUR DIOXIDE REMOVAL EFFICIENCIES, FLY ASH
                    ALKALI UTILIZATION AND SCALING RATE AS A
                    FUNCTION OF STOICHIOMETRIC RATIO, CaO/S02
CaO/S02
0.6
1.2
2.0
Removal
Efficiency (pet)
63.3
88.2
98.0
Alkali
Utilization (pet)
100
72
U9.1
Scaling
Rate (gm/hr)
1.68
2.7
3.0
      In the above tests, some difficulty was experienced in removing the suspended
 solids to produce a  "clear" liquid, even with the addition of sodium aluminate
 to the scrubber  solution as a coagulant.  The scaling rate of 1.68  gm/hr was
 observed at a suspended solids concentration of 0.13 pet, 2.7 gm/hr at 0.18 pet,
 and 3.0 gm/hr at 0.2U pet.  Previous experience has shown that 10 gm/hr is a
 high scaling rate, and 0.2 gm/hr is a low scaling rate.

      After the foregoing tests, the absorber tower was again modified, this time
 for the purpose  of providing greater control of the pressure drop under conditions
 of severe scaling.   The change involved attaching one set of cones, those directing
 flow from the center outward, to the standpipe of the flooded disk.  Thereafter,
 movement of the  standpipe varied the spacing between the convex and concave
 conical trays as well as the spacing of the flooded disk venturi.  Thus, as
 scale buildup occurred on opposed  surfaces, all such surfaces could be moved
 further apart to maintain a constant pressure drop.  A further effect of the
 change was to distribute the pressure drop more evenly throughout the scrubber
 tower.   This last effect was believed to be responsible for a further increase
 observed in scrubber efficiency, from 88.2 to 92.9 pet, which probably occurred
 because of a more efficient use of energy in redispersing droplets  of scrubber
 liquor throughout the tower.  All  of the removal data given below are offset
 from former data due to this increased efficiency.
                                     280

-------
     Scaling rates and removal efficiencies were next investigated as a function
of sodium concentration.  The sodium levels investigated were 0.17 pet, 0.66
pet, k pet, and 9-3 pet.  The 0.17 pet represents sodium leached from the fly
ash during a 3-day test period; no make-up sodium was added.  The magnesium
concentration was kept constant at 1 to 2 pet, L/G was ^5, calcium oxide to
sulfur dioxide stoichiometric ratio was 1.2, and other operating conditions were
as described previously.  The results are tabulated in Table 5-  A typical
solution analysis at each sodium level is listed in Table 6.

        TABLE 5   SULFUB DIOXIDE REMOVAL AND SCALING RATE AS A
                        FUNCTION OF SODIUM CONCENTRATION
Sodium Concentration (pet)
0.17
0.66
h.Q
9.3
S02 Removal
Efficiency (pet)
95-2
92.9
93
96
Scaling
Rate (gm/hr)
5-8
3.51
2.7
.0
Suspended
Solids (pet)
0.066
0.07k
0.17
0.83
        TABLE 6   TYPICAL SOLUTION ANALYSIS AT SODIUM LEVELS OF
                    0.17 PCT, 0.66 POT, h.O PCT, AND 9-3 PCT
Species:
Ca (ppm)
Mg (pet)
SOl^ (pet)
Percent Sodium: 0.17
702
1.30
7-06
0.66
631
1.33
7-56
h.O
6H3
1.3
15-0
9-3
762
1.7
26.0
     The solids settling properties were observed to degrade as the ionic strength
of the scrubber solution increased. This phenomenon has been observed previously
in EPA laboratory testing on dilute double alkali systems, and by Arthur D.
Little, Inc.11 in laboratory and pilot plant work on dilute and concentrated
double alkali systems.  Factors reported to influence the solids settling properties
are reactor configuration, concentration of soluble magnesium and iron,  and the
concentration of soluble sulfate.  In this investigation, the concentration of
magnesium and iron were relatively constant.  However, the level of soluble
sulfate varied along with sodium level, due to the addition of sodium as sodium
sulfate.  The solids settling  characteristics degraded correspondingly.  A
vacuum filter is being installed on the GFERC scrubber so that the level of
suspended solids may be controlled more precisely.

     It can be seen from Table 5 that the rate of scaling decreased as the
sodium concentration increased.  The absence of  scale formations at the 9-3 pet
sodium level is thought to be  a function of sodium and not due to the higher
 (0.83 pet) level of suspended  solids, since past work at similar levels of
suspended solids (low sodium)  resulted in scaling rates of 1 to 2 gm/hr.  The
stack flue gas was also tested to determine if sulfate was being lost in the
                                       281

-------
mist.  If sulfate was lost in the mist at a rate equal to or greater than that
being absorbed into the scrubber solution (assuming constant liquid volume in
the system), scaling would not "be expected to occur.  Extensive testing indicated
this did not occur, and thus, the absence of scaling is concluded to be a result
of the high sodium concentration.

     It can also be seen from Table 5, that an increase in the level of total
dissolved solids did not have a significant effect on sulfur dioxide removal,
which contradicts previous results.  The current result showing no effect on
removal was observed after the modification of the scrubber by installation of
"rain and drain" trays, which greatly increased gas-liquid contact in the scrubber.
The  conclusion to be drawn is that a high ionic strength in terms of sodium and
magnesium sulfates increases removal for a scrubber configuration providing
minimum contact-residence time  (the flood disk venturi alone) but that it does
not  increase removal for a scrubber providing a maximum of contact-residence
time (the venturi plus trays).  The results would further indicate that the
scrubber solution having low ionic strength had a sufficient equilibrium capacity
to  absorb essentially all of the entering sulfur dioxide (at 8^0 ppm and
L/G =  1<5), but that this capacity was not fully utilized without the increased
residence-contact time.  On the other hand, the scrubber liquor having high ionic
strength was  indicated to have a greater affinity for rapidly absorbing sulfur
dioxide such that essentially all entering sulfur dioxide could be removed with
a short residence-contact time.  Thus, the final conclusion is that sulfur dioxide
removal in  ash alkali scrubbing can be materially improved by either an increase
in ionic  strength or an increase in residence-contact time, but that a substantial
increase  in either can mask the effect of the other.

      State  of oxidation, of sulfite to sulfate, was generally high for all test
conditions  (98 to 99 pet sulfate).  The percentage of sulfite was, however,
higher in the test run at 9.31* pet sodium (2 pet sulfite) than in any other
test.

      Future tests will investigate the solution effects of varying the magnesium
concentration at constant sodium levels and of scrubbing at low pH (below pH 3).
In the present investigation, the level of magnesium was maintained at a constant
value.

      The GFERC is also working under a cooperative agreement with the Square
Butte  Electric Cooperative (SBEC), Minnesota Power and Light Company (MP&L), and
Combustion  Equipment Associates (CEA), to investigate the fly ash alkali scrubbing
process using a  5*000 acfm (saturated) pilot scrubber.  The objectives of the
cooperative agreement are listed in the SBEC section.

     Work is also planned on properties of scrubber waste from ash alkali scrubbing.
The  ideal properties for a scrubber waste are:

     - absence of toxicity
     - low  soluble solids content
     - low moisture content
     - non-thixotropicity
     - high compressive or bearing strength
                                    282

-------
Solid  waste  studied will be generated from operation of the 130-sfm scrubber and
from SBEC's  5»000 acfm (saturated)  pilot plant.   Solubilities will be determined
for the major  soluble elements and  for selected trace elements under conditions
representative of disposal.
                                                                     1 *?
Minnesota Power & Light - Clay Boswell (Cohasset) and Aurora Stations

    Minnesota Power and Light operates Krebs-Elbair spray impingement scrubbers
for particulate control on two 58 MW boilers at the Aurora Station and on one 350
MW boiler at Cohasset.

    The  Krebs-Elbair scrubber consists of a stainless steel box containing
nozzles that direct a high pressure spray against baffles.  The baffles consist of
either vertical rods or a punch plate.   The liquid is atomized when the spray
impinges  upon  the baffles, and the  resulting turbulance promotes an effective
scrubbing of particulates. The Krebs-Elbair scrubber is designed to substitute
power  input  from the high pressure  spray for flue gas pressure differential, and
it operates  with only a k-incti pressure drop across the scrubber system.  This
design supposedly affects a net operational savings in power and cost when
compared  to  scrubbers employing relatively large pressure drops in the gas
stream.   An  additional important feature is a nozzle tree design which permits
sections  of  the nozzles to be removed for maintenance without scrubber shutdown.
An apparent  design limitation of MP&L's scrubbers is the erosion and plugging of
the high  pressure nozzles at other  than very low levels of suspended solids.  As
previously stated, the scrubbers are designed primarily for particulate removal,
although  sulfur dioxide is removed  because of the alkaline character of the fly
ash particulate.

    The  coal  burned at both the Aurora and Cohasset stations is a subbituminous
from the  Big Sky Mine,  located near Colstrip, Montana. The coal typically contains
1.0 pet sulfur, 10 pet ash,  and 9 to 13 pet CaO in the fly ash.  The scrubber
inlet  particulate loadings are 2 gr/scfd at the Aurora station and 3 gr/scfd at
the Cohasset station.  A typical scrubber inlet sulfur dioxide level is about 800
ppm.

    A detailed description of the  MP&L scrubbers is given in the Appendix.  A
process diagram is shown in Figure  h.  Flue gas from the air heaters passes
through three  concurrent sprays: a quench spray, high pressure spray, and the
post humidification spray.  The mist eliminator is between the high pressure
spray  and the  post humidification spray. The post humidification spray washes the
ID fan, which  discharges flue gas to a wet stack.  Scrubber liquor is pumped from
a seal tank  at the bottom of the spray chamber to two clarifiers.  Overflow from
the clarifiers is combined with makeup water and pumped back as spray.  The ID
fan is washed  only with makeup water.  Slowdown from the clarifiers is pumped to
an eighty-acre ash pond.  The scrubber operation is not closed loop.  The Aurora
station does not have clarifiers, and instead uses the clear liquor from the ash
pond.  In all  other respects, the Aurora station is similar to that at Cohasset.
                                     283

-------
          Make up water
To  stack
                                       £>X Post humidification
                                        " ^~    spray
                                               Mist eliminator
                                               Punch  plate
           Bottom  ash pond
     Figure  4.  -   Simplified  flow  diagram for the Clay  Boswell station
particulate  scrubber.
                                  284

-------
    The key operating variables at "both  stations  are  L/G of 8.3 gal/acf and a
total gas stream pressure drop of about It inches of  vater.   The high pressure
spray enters the scrubber at 200 psi. The pH of the  liquid entering the scrubber
is typically it.It and about it.O when leaving the scrubber.   Particulate removal
efficiency is about 98 pet at Aurora and  about 97  pet  at  Cohasset.   The sulfur
dioxide removal is about 20 pet at Aurora and about  15 pet at Cohasset.

    Operating costs are not available.   Operating requirements are electrical
power equal to 0.86 pet of generating capacity, water  requirements  of about  6.5
acre-ft/MW/yr at Cohasset and 30 acre-ft/MW/yr at  Aurora,  and a high labor
requirement for maintenance and operation.

    The scrubber availability was not obtainable  in terms of effective scrubber
operation during times that boilers were  operational.   However, little down-time
on boilers would be expected using this scrubber process,  since many scrubber
problems can be repaired without boiler shutdown.  Massive scaling, plugging, or
problems involving the wet ID fan would,  however,  necessitate boiler shutdown.

    The major problems that have occurred are stack gas  mist carryover, scaling
in the scrubber and liquid circuit, and heavy scale  deposits on the wet ID fan
and mist eliminators.  The scaling problem was more  severe at Cohasset, which
operates close to its rated load and is restricted on  the  amount of blowdown.
Eovever, a seasonal variance had been obtained which allows greater blowdown,
and the scaling and plugging problems have been less severe.   About U20 gpm  of
clarifier underflow is pumped to the 80-acre fly ash pond.   About 1000 gpm of
sulfate-laden overflow from the clarifier is drained to the coal pile sump where
it is diluted with drainage from selected plant drains, demineralizer waste
rater, boiler blowdown, stack drainage, and drainage from the coal  pile and  then
pumped to the bottom ash pile pond and diluted with  bottom ash slurry water.
Some sulfate removal by precipitation is  experienced.   Overflow from the ash pond
is pumped to the cooling water intake for units 1  and  2 and subsequently discharged
into the Mississipi River.

    The problem of stack gas mist carryover has been  shown to be solvable by
the use of a low velocity stack.  A low velocity stack was simulated by breaching
the brick liner of an existing high velocity stack.  The  flue gas could then
flow through the annulus of the stack in  addition  to the brick liner and thus,
in effect, simulate a low velocity stack.  Subsequent  tests on the  modified
stack indicated that mist carryover would not be a problem.   An unexpected
result of the low velocity stack tests was that the  overall sulfur  dioxide
removal was increased from about 15 pet to 37 pet.

    The increased sulfur dioxide removal in the wet stack was attributed to the
TOt ID fan and the modified low velocity  stack.  The water used to  wash the  ID
fan is shattered into droplets which, depending on their  size, would rapidly
fall out or are transported to the stack.  Depending on the transport velocity of
the droplets, they could be carried out of the stack (which they were not),  fall
to the bottom of the stack, or be absorbed onto the  walls,  with some drops
remaining in a dynamic suspension until becoming large enough to fall to the
                                       285

-------
"bottom of the stack.  A situation has  thus been created in which a water aerosal
is continuously and intimately mixed with flue gas and fly ash aerosal.  As a
consequence, an additional IT to 22  pet sulfur dioxide removal efficiency was
obtained in the wet stack and the overall removal efficiency was increased to
about 37 pet.  The liquid that was drained from the stack to the bottom ash pond
had a pH of about 3.

     Minnesota Power  and Light had planned on two TO MW retrofit scrubber
installations on units  1 and 2.  However, MP&L now tentatively plans to use
ESP's on the two TO MW  boilers and the Krebs-ELbair scrubber on the 350 MW unit
and use a  common low  velocity  stack.  With all three boilers operating, the flue
gas from the two TO MW  boilers would provide  reheat to the 350 MW scrubber exit
flue gas.

Montana Power  Company Pilot  Plant

     A pilot plant program investigating ash  alkali scrubbing was undertaken in
 19T3 for Montana Power  by  Bechtel Corporation in cooperation with Combustion
 Equipment  Associates  who designed the  scrubber.  Arthur D. Little, Inc. also
 participated in the test program, and  sulfur  dioxide and particulate testing
 services were  provided  by  York Research Corp.  Based on generally favorable
 results  in the pilot  plant program,  Montana Power is employing ash alkali scrubbing
 in a new 362 MW pc-fired generating  unit at Colstrip which started operation in
 September  19T5.  A similar unit  is scheduled  for startup in 19T6 or 19TT-

     The  information  presented on these tests was obtained primarily from Dr.
 Carlton  Grimm  of Montana Power Company, and from a report published by Combustion
 Equipment  Associates-^.  The present description of the tests as pieced together
 from the various sources is  solely the responsibility of the writers of this
 paper.

     The 3,000 acfm pilot  plant  (Figure 5) consisted of a venturi section for
particulate removal,  a  spray tower for sulfur dioxide removal, followed by mist
eliminators and a reheat section.  Provisions were made for alkali makeup as
       or  lime and the  use of  cooling  tower blowdown as makeup water.

     The principal objectives of the study were to:

     1.    Demonstrate CEA's  guarantees on sulfur dioxide and
           particulate removal.

     2.    Determine level of supplemental alkali required, if
           any,  for sulfur dioxide removal.

     3.    Optimize variables influencing system performance.

     1».    Investigate use of cooling tower blowdown as a
           demister wash spray.
                                       286

-------
                       Absorber
N)
OO
                                                               To stack
                                                                   t
                                                    Steam
                                                   reheater
                          \' V  \
                        Recycle
                          tank
                 •Q-
                                                                      Fan
Figure 5.
                                       Sludge to ash pond

                              Pilot  plant scrubber,  Montana Power Company
                                                                   Make up water
                                                                    formulation
                                                                       _L
                                                                                Alkali make up
                                                                                   tanks

-------
     The goal for particulate removal was an outlet dust loading of 0.03 gr/scf.
Results at an inlet dust loading of 2 gr/scf indicated that the 0.03 gr/scf exit
loading could be met at a venturi pressure drops of approximately 12 inches of
H20, and 0.02 gr/scf at about 17 inches of H20.  A pressure drop of 17-inch H20
was selected for the h to 6 gr/scf anticipated in the full scale units.

     The sulfur dioxide removal requirement was to comply with a standard of 1 Ib
S02/MM Btu, or a level of U25 ppm.  Compliance required removals in the range of
50 to 60 pet.  This was accomplished with fly ash alone (without supplemental
alkali) supplied at a dust loading of 2 gr/scf.  At this dust loading, however,
a pH of U  in the scrubbing liquor entering the venturi was judged lower than
desirable  for  scale control.  Using a simulated grain loading of k gr/scf, a
higher pH  of  5 to  6 was achieved and the removal criterion was also met.  An
approximate plot of sulfur dioxide removal efficiency as a function of stoichiometric
ratio alkali/S02  (inlet)  is  shown in Figure 6, for ash alone and with lime
added.

     A  suspended  solids level of 12 pet was thought to be adequate for sulfur
dioxide removal and scale control, but higher levels would furnish more residence
time for the  ash to react and provide additional nuclei for precipitation.

      Suspended solids  level  was found to have an important positive effect on
 sulfur  dioxide removal in the range of 3 to 12 pet.  Increasing L/G in the spray
 tower also increased  sulfur  dioxide removal.  L/G and pressure drop in the
 venturi had little effect on sulfur dioxide removal.

      Tests were run in which NagCC^ was added for pH control. Sulfur dioxide
 removal was improved by 60 pet or more of the stoichiometric equivalent of the
 NaoCOo  added.  There  are  no  plans to use soda ash in the full scale units at
 Colstrip.

      In other tests,  lime was added to supplement the ash alkali.  Sulfur
 dioxide removal was improved by 60 pet or more of the stoichiometric equivalent
 of lime added.  (See  Figure  6 for approximate results.)  Use of lime is planned
 in full scale units to control pH during periods when fly ash loadings are less
 than k gr/scf or  when fly ash reactivity decreases.

      Scaling in the  scrubber and liquid circuit was controlled adequately,
 mainly by the recirculation  of  ash.  Some fouling did occur in the wet-dry zone,
 and loose scale deposit accumulated in the reheater.

      Major emphasis was given to testing the chevron mist eliminators to establish
 their effectiveness  and the  wash conditions required to prevent plugging.  Three
 approaches indicated  that there was little measurable liquid entrainment through
 the dsmister.  First,  direct measurement of the degree of saturation of the  flue
 gas indicated no  carryover.   Second, outlet particulate loadings with variable
 suspended and dissolved solids  in the liquid showed no increase in outlet loading.
 Third,  outlet particulate loadings did not change upon increasing or reducing
demister wash sprays.   However, deposits did form on the reheater, indicating
that some carryover was occurring.
                                     288

-------
K)
OO
               too
                80
           o
           k.
           Q>
           Q.
                60
Ul
a:

 ftl
O
V)
40
                20
                  0.5
                                                         Montana Power (fly ash and lime)

                                                         Montana Power (fly ash only)

                                                         NSP  (fly ash and  limestone)
                                          	GFERC (fly ash only )
                                1.0
                                                   2.O
3.0
4.0
                             STOICHIOMETRIC RATIO ——	  INTO SCRUBBER
                                                        SO?
               Figure  6.  -  Sulfur dioxide  removals In  pilot plant tes'ts on alkaline fly ashes.

-------
     Simulated cooling tower blowdown was found to "be unacceptable for washing
the mist eliminators, because of scaling that occurred.   To  combat this scaling,
a Koch bubbler tray was added ahead of the mist eliminator and reheater.  Fresh
makeup water used to wash the mist eliminator dropped into this tray, and re-
entrainment of this relatively clean water from the Koch tray lowered suspended
and dissolved solids in the mist by dilution and "exchange," aiding significantly
in keeping the mist eliminator clean.  Plans call for using  this feature in the
full scale units.

     The level of dissolved solids in recycled scrub liquor  was influenced by
the amount and solubility of cations derived from the ash and by the amount and
quality of makeup water added.  Analysis at this end of a prolonged period of
closed loop operations indicated that dissolved solids reached 26,000 ppm.
Dissolved  ion concentrations were as follows:

               Magnesium                U.OOO
               Chlorine                   ^00
               Sulfate                 18,000
               Sulfur trioxide          3,000
               Calcium                    ^00

      The principal  findings of the Montana tests were that acceptable sulfur
 dioxide removal  could be achieved using fly ash alone as the source of alkali ,
 and acceptable particulate removal could be obtained.  Supplementary alkali
 could be used to control pH if fly ash quantity or quality were reduced.  The
 optimum L/G for  spray tower operation was determined to be in the range of 15 to
 20 gal/1,000 scf.

      Scale control .was achieved principally through the recirculation of ash
 solids at 12 pet and above.  Control of pH also aided in scale control.  Other
 measures found helpful were fresh water washing of the mist  eliminator and the
 use of the Koch  bubbler tray.  Cooling tower blowdown was found unacceptable  as
 a source of makeup  water for washing this mist eliminator.  Operation of the
 full scale scrubbing unit at Colstrip has been considered successful to date.

                                                     lit
 Northern States  Power Company - Black Dog Pilot Plant

      A study on  fly ash alkali scrubbing was undertaken by Northern States
 Power, at their  Black Dog Plant near Minneapolis, to establish the design of
 scrubbers for the Sherburne County Generating Plant where two 680 MW units are
 scheduled for completion in 1976 and 1977. Other parties involved in the test
 program included Combustion Engineering, the scrubber vendor, and Black and
 Veatch,  consulting  engineers.

      The test facility was a marble bed scrubber with demister and reheat
 capability,  having  a capacity of 12,000 acfm.  Attendant equipment included a
 reaction tank, thickener and ash pond, makeup and limestone  tanks.  Arrangement
 of equipment is  shown in Figure 7-
                                    290

-------
                                                 Flue gas
                                                                                     Limestone
                                                                                   slurry make up
Primary
contactor
   Air
             Figure  7.  -  Pilot  plant scrubber,   Northern States Power.

-------
    The principal objectives of the test program were to:

    1.   Demonstrate capability for an S02 removal of 50 pet
         or greater operating on alkaline ash only or on
         ash and limestone.

    2.   Demonstrate the guaranteed particulate removal capability
         to O.OH grams/scfd.

    3.   Demonstrate reliable operation with acceptable
         maintenance.

    h.   Determine optimum operating conditions, including
         blowdown  requirements, and instrumentation and
          control functions.

     5.    Demonstrate reliable operation without deposit for
         mation on the components.

     6.    Demonstrate acceptable continuous operation over
          a 30 day period.

     To obtain the guaranteed dust  removal of 0.0^ gr/scfd outlet dust loading,
or 99 pet removal,  it was necessary to modify the marble bed scrubber by  installing
a venturi rod section (commonly called the primary contactor) at the inlet.   The
venturi rod section increased the total pressure drop from 6 to over 12 inches
of water and the particulate criterion was met.  Optimum system L/G was about 30
gal/1,000 acf.  A demonstration test of the two-stage scrubber system for 50
continuous days was completed with  99 pet availability.

     A series of tests  were made wherein inlet  sulfur dioxide and supplementary
limestone were varied while the input fly ash remained relatively constant.   The
sulfur dioxide was varied from U82  ppm to 919 ppm and the limestone from  9 pet
to 100 pet of stoichiometric. The  sulfur dioxide removal efficiency varied  from
k8 to 88 pet under these conditions.  The relationship between the stoichiometric
ratio of input alkali to sulfur dioxide and the removal efficiency is shown  in
Figure 6.  The sulfur dioxide removal criterion was met with as little as 5  pet
stoichiometric limestone added.  The calcium in the fly ash played a major part
in the removal of sulfur dioxide, representing  70 to 80 pet of the alkali reagent.

     Some scaling and plugging which occurred in the pilot program was remedied
in the course of testing prior to the endurance test.  Heavy mud deposits up to
two inches thick appeared on the mist eliminator and scrubber wall surfaces,
requiring shutdown every two days  for washing.  Deposits  in both the mist
eliminator and reheater were similar  in chemical composition to the spray water
solids, consisting primarily of fly ash with CaS03, CaSO^, and CaCOg.  Deposits
on the ID fan were chemically similar but physically more amorphous.  Hard
calcium sulfate scaling occurred on overflow pots.  A redesign of the mist
eliminator washing equipment was  made to prevent plugging at that point.   In
addition, three modes of sulfate scale  control  were utilized.  Ash solids were
recirculated to provide ash and gypsum seed crystals, with a 10 pet level of ash
                                    292

-------
found to be optimum.   Supplementary limestone was reduced to a level of giving
adequate sulfur dioxide removal while minimizing scale, with a level of approximately
15 pet of stoichiometric or less found to be optimum.  A level of 30 pet
stoichiometric was used in the endurance test.  Oxidation of sulfite to sulfate
was increased by bubbling air into the reaction tank to help prevent supersaturation
with calcium sulfite in the scrubber proper.  Oxidation level was increased from
36 pet to 98 pet and above with four times stoichiometric air.  Good experience
was reported with rubber-lined process equipment and fiberglass spray headers.
slurry nozzles did not show wear or deterioration after 1700 hours of operation.

     After kO days of operation, the following concentrations of dissolved
species in the spray water were measured:

                                        PPM

          Sulfite                        0
          Sulfate                     30,000
          Chloride                       550
          Nitrate                        kOO
          Calcium                    UOO-500
          Magnesium                    T»000
          Silica                         200
          Sodium                         200

Pacific Power and Light - Dave Johnston Plant '

     The Dave Johnston Plant at Glenrock, Wyoming has one 330 MW pc-fired boiler
equipped with three parallel Chemico venturi  scrubbers.  The  initial capital
investment was $8 million, or $2U/kw.  A reported $5 million has beea spent on
improvements, bringing total cost to $39/kw.  Startup was in April 1972.

     Coal burned at the Dave Johnston Plant is Wyoming subbituminous coal from a
captive mine.  Sulfur content is 0.5 pet, resulting  in a sulfur dioxide content
of 500 ppm in the untreated flue gas.  Coal ash content is 12 pet, and the
calcium oxide content of the ash is approximately 20 pet.  The calcium is
chemically equivalent to 275 pet of the sulfur content.  Inlet dust loading is k
gr/scfd.

     A detailed description of the Dave Johnston scrubber installation is given
in the Appendix and Figure 8.  Flue gas from  the air heaters  enters the venturi
and then passes sequentially through mist eliminators, to a wet ID fan, and on
to a wet stack.  Ho reheat is used; there is  no bypass.  Turndown is to approximatel:
30 pet of scrubber design capacity.

     Scrubbing liquor is continuously recycled from  the bottom of the venturi
scrubber back to the plumb bob and to the deflector  surrounding the bob that was
installed to prevent solids buildup.  Slowdown from  this loop is pumped directly
to two fly ash settling ponds; no thickener is used.   Overflow from the settling
pond is sent to a clear pond and then pumped  back to the recycle loop.  Ash is
dredged from the settling ponds once each year and is  hauled  away for landfill.
                                     293

-------
         Flue gas from
           air heaters
Mist eliminators
                      pond ^m/
                                                        ash
  Figure  8.  -  Simplified  flow diagram  for the Dave Johnston fly ash  scrubbers.

-------
     Key  operating variables  are an L/G of 13 gal/1,000 acf and  a total pressure
drop of 15 inches of water.   The pH leaving the scrubber is 5, without supplemental
lime.

     Operating costs are not  available.  Major operating requirements are
electrical power equal to 2.3 pet of generating capacity, an estimated water
requirement of 3.6 acre-ft/Mtf/yr, lime for control of pH, and manpower.

     Particulate removal efficiency is over 99 pet, meeting the  design goal of
O.OU gr/scf.  Preliminary values for sulfur dioxide removal are  ho  pet without
lime and  somewhat higher with lime addition.

     Availability is characterized by PP&L as being less than adequate for
utility use, but no company-sanctioned percentages are available.   Availability
depends on the amount of blowdown and fresh water irrigation that are employed.

     Operation is characterized as "intermittent open loop," meaning that
operation with a minimum of blowdown is attempted as the normal  mode of operation,
with much larger amounts of blowdown and makeup used periodically to irrigate
the system.  Cooling water blowdown is mixed with service water  (North Platte
River  water) and used for washing the ID wet fan.

     Past major operational problem was scaling.  At present, it is reported
that scaling of the scrubber  is still a major problem and an occasional fresh
water  irrigation is required.  However, the previous detremental effects of wet-
dry interface scaling have been minimized by the application of  lignosulfonate,
and is now not considered a major operational problem.

     Future test plans are concerned with minimizing scale formation. The
scrubber  will be operated continuous with the recycle liquor pH  maintained at
5.5 to 6.0, and scaling rates and cleaning requirements will be  determined.  The
pH will be controlled using hydrated lime and feed rates are anticipated to be
about  1,000 Ib/hr.  If the test is successful, a major development  program will
be implemented to design and  install a permanent lime system utilizing pebble
lime to maintain the pH at a  constant level.  The hydrated lime  and future
application of pebble lime will be to the scrubber vessel.
                                                                    16
Public Service of Colorado -  Valmont, Cherokee, and Arapahoe Stations

     The  Public Service of Colorado has five turbulent contact  absorbers  (TCA)
scrubbers containing thirteen modules installed on pc-fired boilers. The  modules
were designed by Universal Oil Products primarily for particulate  removal.
However,  because of the alkali present in the fly ash, some sulfur dioxide is
also removed.  The modules consist of three stages of mobile packing,  commonly
called "ping pong balls," with the  spray directed downward through the mobile
packing,  countercurrent to the gas  flow.  A booster fan directs  the flue  gas to
the  scrubber where it is presaturated and then passes through the  mobile  "bed,^
then to  chevron mist eliminators and on to a reheater.  Reheat on most units is
accomplished by heating the flue gas with steam coils; the Cherokee No.  h unit
uses externally heated air.  All units have by-passes.  Typical turndown capability
is  from Uj to 105 pet of the rated scrubber capacity.  Detailed descriptions of
the  installations are given in the Appendix and Figure 9-
                                       295

-------
                          Flue gas to reheater
 Flue gas from
  electrostatic
  precipitator
       I
     Surge
      tank
   Clear effluent
     discharge
 *i*/*v fls.» **.•?*.+ ***'
'•»'*'*•*»«# * '•- *-• *j
•///;.'.•.:•/;;:-;•/::•
                                                          Make  up
                                                           water

                                                           Mist eliminator
                                    Sludge  pond
                                                      Ping
                                                      pong
                                                      balls
      Figure 9.   -   Simplified  particulate scrubbers at Valmont,  Cherokee
and  Arapahoe Stations, Public  Service of Colorado.
                                   296

-------
      Public Service of Colorado's installation of the TCA scrubbers represents
the last stage  of particulate removal in a system which also uses mechanical
collectors and  electrostatic precipitators.  All of the scrubber-equipped boilers
are still serviced by the previously installed mechanical collectors and ISP's
At the Valmont  Station, flue gas from the mechanical  collector on a 196 W
boiler is split into two equal streams with one stream sent to a scrubber and
the other stream sent to an ESP.  The Cherokee Station, which has four boilers
with  a gross  capacity of 115 MW, 170 Mtf, and 375 *W,  has a particulate removal
scheme in which the flue gas passes through, in series, mechanical collectors,
ESP's, and then through scrubbers.  The Arapahoe Station has a similar scheme.

      The coal burned at the Valmont and Arapahoe Stations is Wyoming subbituminous
and has a calcium content of 0.6 pet and an ash content of 5.2 pet.  The calcium
oxide content of the fly ash is about 20 pet, which is chemically equivalent to
about 99 pet of the sulfur content.  The Cherokee Station bums Colorado
subbituminous coal which has a sulfur content of 0.7  pet and an ash content of
9.U pet.  The calcium content of the fly ash is about 5 pet, which is chemically
equivalent to about 38 pet of the sulfur content.  The dust loading at all the
units range from 0.1+ to 0.8 gr/scfd.  The sulfur dioxide concentration at all
the units is nominally 500 ppm.

      Key operating variables at the Arapahoe and Cherokee Stations are a L/G
of 5^-58, a total pressure drop of 10 to 15 inches of water, an average pH of
7.0 entering the scrubbers and a pH of 2.8 to 3 leaving the scrubber.  The
scrubber liquor is recycled from the bottom of the scrubber to the spray header
above the mobile bed.   At the Arapahoe Station, slurry blowdown is pumped to an
ash settling pond after adjusting the pH of the liquor to 6.0 to 9.0 with lime.
The settled fly ash in the pond is dredged periodically and used for landfill.
Clear effluent  from the ponds is discharged under permit from the State of
Colorado.  At the Cherokee Station, slurry blowdown is neutralized and clarified
and then mixed  with the ash pond overflow for discharge.  Scrubber operation
at the Cherokee and Arapahoe Stations are considered  open loop.

      At the Valmont Station, a test program was implemented in October 1971* in
which one of the two scrubbing modules was modified so that limestone for pH
control could be added to the slurry feed.  Each module receives 50 pet of
the incoming gas,  or 25 pet of the total gas flow. Previous operation of the
scrubber system without pH control resulted in moderate to severe scaling, which
required a high degree of maintenance and thus, poor  availability.  The scaling
problem was a result,  in part, of the high alkalinity of the Wyoming subbituminous
coal  which removed U5  to 50 pet of the inlet sulfur dioxide which, in turn creates
supersaturated  conditions with respect to calcium sulfate.  Therefore, one
module was modified so that limestone could be added  to the slurry feed in order
to adjust the pH to a  range of 5.5 to 6.5.  The level of suspended solids was
desired to be about 7  pet to provide seed crystals for the precipitation of
gypsum.   Blowdown from the module is pumped to a lined settling pond.  The
settling characteristics of the sludge is reported to be good and contains about
50 pet moisture.   Overflow from the pond is pumped to a recycle tank where makeup
water is added  at  a rate of 70 gpm.  The combined water is then used to slurry
                                     297

-------
the limestone, wash demisters, and seal water for pumps.  The only water loss
in the system is due to pond evaporation, evaporation in  the module,  and the
moisture in the sludge.  The system is considered to be "closed  loop."

     Since the system has teen operated for only four months, the results are
considered to be preliminary in nature.  Additional tests will be made  in the
future.  The module operating without limestone removes about U5 to  50  pet
of the incoming sulfur dioxide and has a particulate removal efficiency of
about 96 pet.  The modified module, using limestone for pH  control,  removes
about 85 pet of the incoming sulfur dioxide and has a particulate removal
efficiency of about 96 pet, which meets particulate emission requirements of
0.02 gr/scf  (wet).  The  state of sulfur oxidation is high,  reported  to  be
nearly 100 pet.   The  degree of supersaturation is not known.  Scaling in the
modified module  is reported to be moderate, although additional  tests are
required since  some difficulty was experienced in maintaining the  suspended
solids at  7  pet.  The availability of the converted module  was about 70 pet.

      The particulate  removal efficiency of the Cherokee and Arapahoe scrubber
 systems  are  95  to 98  pet, which meets the design goal of  95 to 98  pet.   The
 sulfur dioxide  removal  efficiency at the Cherokee Station,  which burns  Colorado
 coal,  is 15  to  20 pet.   Sulfur dioxide removal at the Arapahoe Station, which
burns Wyoming coals,  is U5 to  50 pet.  The difference in  sulfur  dioxide removal
 efficiencies can be attributed, in part, to different fly ash alkalinity.

      The scrubber operating problems experienced to date  are wear  and periodic
 replacement  of the mobile bed packing, corrosive failure  of the  flue gas
 reheaters, and scaling.  At Valmont, the modified scrubber module  experienced
moderate scaling on the module walls and certain portion of the  first grid, whict
 in turn caused pluggage of pump screens.  Past experience at the other scrubber
modules using additives added  for  scale control, have not been  successful,
although the tests were limited in nature.  Scaling and plugging has occurred at
the wet/dry zone on the first  stage grids, and on the reheaters.  With the
exception of the Valmont experimental module, a higher than design blowdown is
practiced in an effort  to control  scaling and prevent pluggage.

      Direct  operating costs  for Cherokee #3 is 0.50 mill/kwh, based on 75 pet
availability.   Direct operating cost for the other stations are  not available.
Scrubber operating requirements are electrical demand equal to  k pet of the
power generated and  also, steam for reheat.  The water requirements are
approximately 3.3 to  k.l acre-ft/MW/yr.  Manpower requirements  for scrubber
operations are not available.  However, a high degree of maintenance is required.

Square Butte Electric Cooperative  Pilot Plant

      The Square Butte Electric Cooperative  (SBEC) is currently constructing a
1*50 MW cyclone-fired generating unit requiring particulate and sulfur dioxide
abatement controls.   The 1*50 MW unit is referred to as Center 2 and is being
constructed  adjacent  to the  238 MW Center 1 unit at the Milton R. Young  Station.
Particulate  control will be  provided by electrostatic precipitators  (ESP's) and
                                     298

-------
sulfur  dioxide  control  by wet  scrubbers.   SBEC and Sanderson & Porter,  Inc.,
consulting  engineers, selected Combustion Equipment Associates (CEA)  and
Arthur  D. Little,  Inc.  (ADL),  to construct a 5,000 acfm (saturated) pilot plant
and the full-scale scrubber  based partly  on previous pilot plant  experience
at Montana  Power  Company.  The full-scale scrubber will utilize fly ash alkali
vith lime supplement.   The design criteria and operating parameters of  the full-
scale scrubber  were determined in two months of testing conducted by  CEA-ADL
in cooperation  with SBEC,  GFERC, and Minnesota Power and Light, on the  5,000
acfm pilot  scrubber.  The objectives of this program were:

     1.  To demonstrate that  the stated  performance guarantee
         could be achieved  for a maximum coal sulfur content of
         1.3 pet.

     2.  To determine  design  conditions  for L/G, percent suspended
         solids  in the recycle slurry, retention tank residence
         time, and pH  of the  recycle slurry.

     3.  To determine  supplemental lime  requirments at the
         maximum sulfur content.

     ^.  To demonstrate the feasibility  of using a flue gas bypass
         for reheat.

     A long-range testing program using the 5,000 acfm (saturated) pilot  plant
will be conducted under a cooperative agreement between SBEC, Minnesota Power and
Light Company,  Combustion Equipment Associates, and the Grand Forks  Energy
Research Center.   The objectives of the cooperative program are:

     1.   To determine  whether sufficient alkali can be solubilized
          from cyclone-fired fly ash to reduce sulfur dioxide in flue
          gas below the level of State and Federal emission standards.

     2.   To determine the amount of additional alkali from lime which
         may be required to supplement fly ash alkali to meet State
          and Federal emission standards.

     3.   To determine the severity of calcium sulfite and calcium
          sulfate scale formation under normal operating conditions
          of the flue gas desulfurization pilot scrubber, and to
          investigate chemical methods of minimizing the scale formation.

     It.   To establish that the pilot scrubber can be operated on a
          closed-loop basis, and to determine the chemistry of the
          closed-loop system.

     5.   To determine what effect fly ash-derived soluble salts in
          the  scrubber solution will have on the  sulfur dioxide
          removal efficiency.
                                    299

-------
     6.    To determine and evaluate waste disposal of sulfate/sulfite
          sludge and fly ash-derived soluble salts in sludge.

     7.    To conduct corrosion tests to determine the effects of
          scrubber liquor on materials of construction to be used
          for full-scale flue gas desulfurization processes.

     8.    To determine the mass balance of all input and output
          materials, including selected trace elements and leachate
          from sludge.

     9.    To evaluate the capital and operating costs of fly ash alkali
          flue gas desulfurization for 100 MW, 500 MW, and 1000 MW
          steam generator plants based on the technical and operating
          data obtained from the pilot scrubber.

     An additional phase of testing may be concerned with dilute sulfuric  acid
scrubbing with fly ash neutralization, and sodium-magnesium and calcium  double-
alkali-type scrubbing with fly ash neutralization.

     A 5,000 acfm (saturated) pilot plant (about l.U MW equivalent) employing
spray nozzles was chosen to minimize the pressure drop (hence, energy usage)
across the absorption tower.  The pilot plant (see Figure 10) has, essentially,
two  liquid loops:  the primary sulfur dioxide scrubber loops, and the 'mist
eliminator and wash tray loops.

     The wash tray loop operates on clear liquor at approximately pH h and
consists of a wash tray above the absorber tower, a tray recycle tank, clarifier
and  clarifier overflow tanks, and a demister.  The wash tray, which is constructed
of 3l6L stainless steel, is designed to remove entrained droplets of scrubbing
liquor which could otherwise foul the demister.  Liquid from the wash tray
drains to an 8 x 8-foot flakeglass-lined recycle tank which is then pumped back
to the wash tray or to an 8 x 8-foot flakeglass-lined clarifier.  Overflow from
the  clarifier is used to wash the bottom of the wash tray.  Liquid from  the
clarifier not used for washing is drained by gravity to a 6 x 6-foot flakeglass-
lined overflow tank. Makeup water from nearby Lake Nelson is added to the  pilot
scrubber at the clarifier overflow tank at an average rate of l.U gpm (about  1.6
acre-ft/MW/yr) and the combined liquid is used to wash the polypropylene demister.
Underflow from the clarifier is pumped to a 3l6L stainless steel vacuum  filter.

     The scrubber loop operates on a slurry of alkali ash in recycled liquor  at
a characteristic pH of h to 7-  It consists of a 1*5 foot high by 3 1/2 foot
diameter flakeglass-lined absorber tower which contains six banks of 3l6L  stainless
steel nozzles spraying scrubber liquid countercurrent to the gas flow.   The
scrubber liquid drains to a 12 x 8-foot flakeglass-lined retention tank  equipped
with a 3l6L stainless steel agitator. The retention tank liquid is pumped  back to
the  spray nozzles and also to an 8 x 8-foot flakeglass-lined thickener which  is
used to control the level of suspended solids.   Reducing the liquid flow from the
retention tank to the thickener will increase the level of suspended solids;
                                    300

-------
             To stub stack
Make up water
                                                                                                       Filter cake
                                                                                                      to disposal
                 Figure 10.-  Flow diagram of 5,000 acfm saturated  pilot plant scrubber,  Square Butte
          Electric Cooperative.

-------
 increasing the flow rate will lower the level of suspended solids.  Overflow  from
 the thickener drains by gravity to a 5 x 5-foot flakeglass-lined overflow tank
 where it is mixed with liquid pimped from the vacuum filter.  Thickener underflow
 is pumped to the vacuum filter.  The vacuum filter is operated only when the
 concentration of suspended solids in the thickener underflow is above approximately
 Uo pet.   The liquid from the thickener overflow tank is pumped to a k x 5-foot
 fly ash preparation tank with the excess liquid returning to the retention tank.
 The fly ash slurry is pumped to an 8 x 8-foot feed tank for additional mixing
 time, and then to the retention tank.   Fly ash is stored in a 3 x 5-foot hopper
 and fed to the preparation tank using a screw feeder at rates up to 8 lb/min.  Lime
 from a 3 x 2-foot storage hopper and screw feeder is used for pH control and is
 fed directly into the retention tank without slaking.  The total amount of
 liquid in the entire pilot plant scrubber is about 25000 gallons.  All pumps are
 rubber lined.  Liquid flows are measured by rotometer; liquid and gas temperatures
 are measured by dial thermometers,  and pressure drops are measured by manometers.

      The pilot plant scrubber has the  necessary equipment and controls to operate
 over a wide range of variables.  The solution pH can be varied from below pH it up
 to pH 9; the retention tank residence  time may be varied from k minutes to 16
 minutes; the liquid to gas ratio may be varied from UO to 80; and a sulfur
 dioxide injection system, can adjust  the scrubber inlet sulfur dioxide to any
 desired concentration.   A duct equipped with an orifice and damper has been
 installed between the inlet and outlet  of the absorption tower and can be used to
 bypass part of the hot  inlet flue gas to mix with the cooler outlet flue gas
 leaving the absorption  tower.   The mixing of flue gases in this manner is being
 tested as a method for  reheating to  a temperature above the saturation point to
 eliminate the possibility of stack gas  rain.

      A mobile trailer,  constructed by KVB Industries  according to specifications
 supplied by the Grand Forks Energy Research Center, provided the capability of
 continuously monitoring both the inlet  and outlet flue gas  for sulfur dioxide,
 nitrogen oxides, carbon dioxide and oxygen.   In addition to the gas monitoring
 equipment, the trailer  contains a chemistry laboratory to perform most analyses
 of coal  and scrubber  liquor on site.

      The preliminary  results of the test program completed  to date  have been
 favorable.   The sulfur  dioxide  removal is required to  comply with the Federal
 emission standard of  1.2 Ib/MM Btu, which corresponds  to approximately 535  ppm
 S02  (dry).   The sulfur  dioxide removal efficiency was  investigated  as  a function
 of L/G,  suspended solids, inlet sulfur, sulfur dioxide concentration,  and  fly
 ash add  rates.   The ESP inlet fly ash particulate loading at  the  inlet to the
 ESP on Center 1  ranges  from 0.71 to 1.53 gr/scf and averages  1.13 gr/scf.   The
three ash add rates investigated were equivalent  to the  average amount collected
by the ESP on Unit 2, the combined average amount collected by the  ESP's on
Units 1  and  2,  and the maximum amount .collected on Units  1 and 2.   A typical
analysis of  the  fly ash is shown in Table 7.
                                   302

-------
          TABLE 7   A TYPICAL ANALYSIS OF THE FLY ASH PRODUCED
                       BY THE CYCLONE-FIRED CENTER UNIT NO. 1
                                                           Percent of ash,
                                                             as received
Loss on ignition at 800° C 	
Silica, Si02 	
Aluminum oxide , AlgO^ 	

Titanium oxide , TiOg 	
Phosphorous pentoxide , P20j 	

Magnesium oxide , MgO 	

Potassium oxide, KgO 	
Sulfur trioxide, S03 	
	 2.2
	 29.8
	 12.7
	 10.6
	 0.5
	 0.3
	 25.7
	 U.5
	 2.2
	 2.0
	 6.U
           TOTAL

      Figure 11 illustrates the sulfur dioxide removal efficiency at the above
fly ash add rates at L/G ratios of 60 and 80.  The dashed line corresponds to
the average fly ash production collected by both units,  hereafter referred to
as the  average ash add rate.  The solid line corresponds to the maximum fly
ash production by both units, hereafter referred to as the maximum ash add rate.
A sulfur dioxide level of about 1100 ppm (dry) would be  equivalent to about a
0.75  pet sulfur coal (HHV660J*).  A level of 1900 ppm (dry) is equivalent to
about 1.3 pet sulfur in coal.  The oulet sulfur dioxide  represents the removal
for the total scrubber system, which includes the flue gas by-passed and used
for reheat.  The total flue gas into the system was 6300 acfm, of which 1100 acfm
by-passed the absorber tower to provide reheat.  The inlet gas temperature was
about 325° 7.  The temperature of the saturated gas out  of the absorber tower
was about 135° F.  After mixing the by-pass gas, the temperature of the gas
to the  stack was about 155° F.  The flue gas reheat was  tested as an alternative
to coil reheaters.  No stack gas mist was observed to occur.

      At a L/G of 60, an inlet level of about 1100 ppm S02 (dry) using  fly  ash  at
the average add rate, the sulfur dioxide removal efficiency  for the total
scrubber system was about 6l pet (absorber tower was about  70 pet).  The total
fly ash alkali utilization, based on 25 pet CaO, k.12 pet MgO, and  1.67 pet
Na20, is about 63 pet. At the maximum ash add rate, the sulfur dioxide removal
is about 69 pet (absorber tower was about Bk pet), which represents  a  total
alkali  utilization of 51.5 pet.  Supplemental lime was not  added and the
corresponding pH of the recycle slurry was about 3-9 at the average ash  add rate
and U.9 at the maximum ash add rate.

      At a L/G of 60, inlet level of about 1850 ppm S02  (dry), using the  average
fly ash add rate with lime supplement, the total scrubber system removal efficiency
was about 70 pet (absorber tower removal of  about 8^ pet).   The  supplemental
lime  was added to maintain the pH at 6.0 to  7-0, and was chemically equivalent
from  about 90 to 100 pet of the 1850 ppm  inlet  sulfur dioxide;  the total alkali


                                     303

-------
     80O
I	1	1	1	1
              1.2 Ib SOg/mmbtu
         800    1,000   1,200   1,400   1,600   1,800   2,OOO
                           INLET S02 ,  Ppm (dry)
      800
  E
  o.
   .   600
  M
  O
  UJ
  O
      400
      200
               1.2 Ib S02 / mmbtu
                                               L/G  = 60
         800    1,000
           1,200    1,400    1,600    1,800   2,OOO

             INLET  S02 ,  ppm (dry)
    Figure II. -  Sulfur dioxide removals  in SBEC pilot plant tests
using  fly ash  alkali.  Solid  line  represents  average fly ash (1.13 gr/scf)
collected  by ESP'S and  dashed line represents maximum fly ash (1.53
gr/scf)  collected by  ESP'S.
                           304

-------
(fly ash  alkali  and lime supplement) was equivalent to about 1^0 pet of the
inlet  sulfur dioxide.   At the maximum ash add rate with lime supplement, the
sulfur dioxide removal remained at about 70 pet (absorber tower about Qh pet).
The supplemental lime  was added to maintain the pH at 6.0 to 7.0 and was chemically
equivalent  to about 68 pet of the inlet sulfur dioxide; the total alkali (fly
ash alkali  and lime supplement) was equivalent to about lh2 pet of the inlet
sulfur dioxide.   In a  separate  test using only lime chemically equivalent to
that of the total alkali, the removal efficiency increased to about 79 pet.

     At an  L/G of 80,  inlet level of about 1100 ppm SOg (dry) using fly ash at
the average add  rate,  the sulfur dioxide removal efficiency for the total
scrubber  system  increased to about 81* pet (about 86 pet in the absorber tower).
The total fly ash alkali availability was increased to about 70 pet.  The pH of
the recycle slurry was about 3.8.  No supplemental lime was used.  At the
maximum ash add  rate with'supplemental lime, the total scrubber system sulfur
dioxide removal  was about 8l pet (absorber tower was  about 97 pet).  The supplemental
lime was  added to maintain the  pH at 6.0 to 7-0, and  was equivalent to about
36 pet of the inlet sulfur dioxide; the total alkali  (fly ash alkali and
supplemental lime) was equivalent to about 162 pet of the inlet sulfur dioxide.

     At an  L/G of 80,  inlet level of about 1850 ppm S02  (dry), the removal
efficiency  for the total scrubber system was about 75 pet at the average ash add
rate  (absorber tower was about  91 pet).  Supplemental lime was added to maintain
the pH at 6.0 to 7.0,  and was chemically equivalent to about 55 pet of the  inlet
sulfur dioxide;  the total alkali (fly ash alkali and  supplemental lime) was
equivalent  to 102 pet  of the inlet sulfur dioxide. At the maximum add rate, the
removal efficiency for the total scrubber system was  about 80 pet  (absorber
tover  about 97 pet), supplemental lime was added to maintain the pH at 6.0  to
7.0,  and was chemically equivalent to about 53 pet of the  inlet sulfur dioxide;
the total alkali was equivalent to about 128 pet of the  inlet  sulfur dioxide.  A
test  using only lime chemically equivalent to about 13^  pet of the inlet  sulfur
dioxide,  the high ash add rate with  supplemental lime,  resulted in a  scrubber
system removal of about 80 pet (absorber tower about 97  pet).

      The higher  sulfur dioxide removals demonstrated in pilot  plant tests were
obtained by adding lime at rates higher than  intended for the  H50 MW  scrubber
unit,  and these high rates may not be reproduced  in practice on the  commercial
scale scrubbers.  The pilot test results do,  however, demonstrate that  the
scrubber design to be used on the full-scale  unit is capable of meeting and
exceeding removals required to comply with the  1.2 Ib/MM Btu Federal  emission
standard, and further, that required removals under normal conditions of coal
sulfur content can be achieved using fly ash  alone without lime.

      At the conclusion of the  eight-week testing  program, the scrubber system
was  inspected for scale and it was reported to  be light, with most deposits at
wet-dry interfaces.  On the basis of previous pilot plant experience at Montana
Power Company, and limited confirmatory testing in this program,  it was concluded
by CEA-ADL that a suspended solids level of 12  pet provides seed crystals
for  precipitation of gypsum, and also provides  scale control when recirculated to
                                       305

-------
the spray nozzles.  However, future reliability tests will be performed to
further investigate scaling and corrosion problems.  The state of oxidation was
high, usually greater than 97 pet.  The sludge was reported to have excellent
settling properties and cake from the vacuum filter contained ^0 to 50 pet
moisture.

     Operational problems were not numerous and were concerned with an occasional
plugging of pipes and erosion of plastic valves and spray nozzles due to the high
level of suspended solids.

     Burning coal with a sulfur content of about 0.75 pet, CEA-ADL recommends
the following full-scale scrubber operating conditions:   L/G of 80, 12 pet
suspended solids, slurry recycle pH of 5.   At the maximum fly ash add rate, no
additional alkali would be necessary.  Fifteen percent of the incoming flue
gas will be by-passed and used for reheat.

     Burning coal with the maximum sulfur  content (l.3 pet),  CEA-ADL recommends
the following full-scale operating conditions:   A L/G of 80,  12 pet suspended
solids, a pH of 6.5 to 6.8, k tons per hour of supplemental lime at the maximum
fly ash add rate.  Fifteen percent of the  incoming flue  gas will be by-passed
and used for reheat.

     Capital and operating cost estimates on the i+50 MW  full-scale  scrubber were
presented at the Mid-Continent Area Power Pool  (MAPP)  Environmental Workshop,
November 18, 19751?.  The capital  investment estimates for the  SBEC project are:

     - Material and installation cost             35.5 $/KW
     - Additional contracts and site
         preparation cost                         10.0 $/KW
     - Escalation, interest during
         construction,  consultant  fees,
         Power Company overhead                   20.0 $/KW

     The estimated annual operating costs are:

                                Unit  Cost    ^/MM Btu       $/yr

     - Fly ash                       0            00
     - Lime (CaO)                  k2  $/ton       2.13       $   555,5^0
     - Waste disposal
       transportation  (30 TPH)   0.80  $/ton      0.99         258,^60
     - Operating and main-
       tenance  staff
       (20 men)                     20,000
                                   $/man        1.23         ^00,000
    - Maintenance (k pet of
      capital  investment)                      1.71         639,000
    - Capital fixed charges                    6.k\       2,092,500

         Total Annual Cost                     12.5      $3,9UU,500
                                   306

-------
      If an equivalent lime scrubber were to be constructed,  the  following
operating costs would be affected:  an increase of $1*50,000  per  year to purchase
50  tons of lime per day, an increase of about $100,000 per year  for offsite
disposal of fly ash, a decrease of about $50,000 per year since  lower  L/G could
be  employed.  The normal operating power requirement for the 1*50,000 KW fly ash
scrubbing system is reported to be 7922 KW.  The maintenance and capital  fixed
charge would be less since it is estimated that the capital  investment for the
lime system is about 2.5 pet less than the fly ash system.  When all economic
factors are taken into consideration, the Square Butte Electric  Cooperative
expects the fly ash alkali scrubbing system to save I'M ,000 per year  in  operational
costs.

SUMMARY AND CONCLUSIONS OF WESTERN SCRUBBERS

      Most future utility scrubbers for Western coals will be designed  for the
primary purpose of sulfur dioxide removal and not for particulate removal, as
was true in the majority of past installations.  The major reason is the
necessity for having flue gas desulfurization for most low sulfur Western coals
under the developing regulatory pattern in the West.  There  is also a  trend toward
installing scrubbers in series with electrostatic precipitators  on new plants
because of the appeal of eliminating the visible fly ash plume with a  relatively
more reliable device, the ESP.  If the scrubber can then be  bypassed for  short
periods for maintenance, a high plant availability can be maintained.

      The three most significant factors to be considered in  designing  scrubbers
for Western coals are: l) the low concentration of sulfur dioxide to be removed,
2)  the alkalinity of Western coal fly ashes, and 3) the tendency to operate at a
high state of oxidation, producing sulfate and not sulfite.   It  is significant
that all Western FGD installations, existing or planned, use throwaway type
processes.  This is consistent with the remoteness of Western scrubbers from most
large markets for sulfur or gypsum, although sulfuric acid could possibly find
sizable markets in the West as an ingredient in fertilizer manufacture.   Sulfur
dioxide emissions from low-sulfur Western coals can be brought into compliance
with the Federal emission standard by treating only a portion of the flue gas
and by-passing the remainder for stack gas reheat, as planned by Square Butte
Electric Cooperative.  Cost is minimized by balancing the savings of treating
a smaller volume of gas against the cost of removing a higher percentage  of the
sulfur dioxide from the fraction treated.  Where local standards are more stringent,
the option of treating a partial flow ceases to exit.

      At low concentrations of sulfur dioxide, it can be argued that gas film
diffusion should be the rate controlling step, since the equilibrium partial
pressure of sulfur dioxide for alkaline slurry is low and the capacity of the
slurry to absorb sulfur dioxide during passage through the scrubber is not taxed
if  the amount absorbed is small.  A sufficient L/G of course plays a part in
validating this argument.  If gas diffusion does control, design should maximize
gas-liquid contact and residence time.  Long residence time  necessitates  a large
volume; and good contact requires either multiple sprays or  tower packing.  If
scaling can be resolved, packing is probably the economical  choice.  If scale
has  a tendency to form, the large empty volume with multiple sprays will  be the
better option.  The operating experience of the utilities cited  in this paper  is
that,  if blowdown is sufficiently restricted, scale will indeed  tend to form.


                                     307

-------
     The problem of scaling is inexorably tied in with  the question  of  vhat
constitutes "closed loop" operation.   The practical answer is that a system is
"closed" if no liquid blowdown is deliberately removed  and disposed  of.   Some
inadvertent loss in sludge cannot be eliminated.   Beyond this, pond  evaporation
of a saturated scrubbing liquor does remove sulfate from the system  by  precipitation,
even though liquor may be returned from the pond.  Since the sludge  and pond
losses vill vary with design and climate, every system  will be "closed" to a
different extent.

     Control of scaling must be approached differently  for low-sulfur Western
coals than for Eastern coals.  A high state of oxidation and the  slight control
that can be afforded by manipulating pH in a sulfate system offer little hope
that unsaturated operation is possible for Western coals during closed  loop
operation.  That this can be accomplished for high sulfur Eastern coals depends
on a low state of oxidation, with consequent precipitation of calcium sulfite and
coprecipitation of gypsum from a solution that is not saturated with calcium
sulfate'.  These conditions do not appear to exist for  Western coal  operations.
Control of scale formation, in the authors' opinion, will depend  most directly on
circulating a sufficiently high level of suspended solids, and operating at a
constant pH, whether high or low.  This is supported by the 6 to  12  pet level of
recirculated suspended solids, and lime or limestone for pH control,  currently
practiced by the utilities cited in this paper.

     Depending on the cost of reagent and the properties of the waste products
produced, there may be more or less motivation to improve the utilization of
alkalinity in Western fly ashes in scrubbing systems.   Laboratory tests at the
Grand Forks Energy Research Center have shown that utilization of fly ash alkali
improves significantly as the pH is dropped from 5 to 3.   Other operating variables
would have to be changed along with the pH, including the flow circuit,  gas-
liquid contact, L/G, and slurry reaction times.

     In conclusion, involvement of alkaline ash in a scrubbing system implies a
new variable which must be controlled.  Since the analyses of Western ashes
vary sufficiently, this effectively rules out one tailor-made solution  for
every scrubber application.  As fly ash varies,  the characteristics  of  sludge
will vary, and therefore extensive studies on properties of sludge,  including
leaching of major and trace elements, will be required.   Experiences  on operating
scrubbers indicate that work is needed on materials of  construction,  and on
component design to improve reliability.  The solution  to stack cleaning problems
will require continuing development effort as long as there is a  conscious
desire for improvement.
                                     308

-------
APPENDIX

Scrubber Design and Operation
Four Corners Plant
Arizona Public Service  Company


LOCATION

 1.  Farmington,  New Mexico.
 2.  Elevation is 5300  feet.
 3.  Atmospheric  pressure  is 12.1 psi.
 k.  Annual precipitation  is 8 inches.
 5.  Water supply for the  plant comes from Morgan Lake, a man-made reservoir filled
      from the San Juan River.

SCRUBBER APPLICATION

 1.  Particulate  removal,  retrofit.
 2.  Boilers equipped with scrubbers.
    - Two 175 MW Riley pc-fired boilers (Units 1 and 2).
    - One 225 MW Foster Wheeler pc-fired boiler (Unit 3).
 3.  Service date:  Units  1 and 2, December 1971; Unit 3, January 1972.
 k.  Fuel is New  Mexico subbituminous coal from the Navajo mine.
    - 8900 Btu/lb.
    - 12 pet moisture.
    - 0.68 pet sulfur.
    - 22 pet ash.
    - k pet CaO  in ash.
 5.  Flue gas entering  the scrubbers.
    - Boilers 1  and  2.
         - 8lU,000 acfm.
         - 3^0°  F.
         - 650 ppm S02-
         - 12 gr/scf particulate.
    - Boiler 3.
         - 1,030,000 acfm.
         - Conditions  are the same as on units 1 and 2.
 7.  The particulate  removal goal was set by the project at 99-2 pet.

 SCRUBBER DESCRIPTION

 1. Two venturi  scrubbers on each of boilers 1, 2, and 3.
 ,2. Vendor, the  Chemico Air Pollution Control Company.
 3.  Capital cost is  $30 million, or $52/kw.
 1*.  Operating costs  are not available.
 5. Materials of construction.
     - Scrubbers  are  carbon steel with stainless steel or plastic lining.
     - Outlet ducts were stainless steel lined, later lined with plastic over the
         stainless  steel.
     - Liquid lines and pumps are rubber lined.


                                    309

-------
     - Process vessels are plastic lined.
     - Reheaters were 3l6L stainless steel.
     - Wet fans are inconel.
 6.  No "bypass.
 T.  Turndown is to approximately 50 pet of rated scrubber capacity.
 8.  Chevron mist eliminators have 6 stages.
 9.  Wet fan.
10.  Reheaters that heated flue gas directly with steam coils  failed  because of
       corrosion.  The reheat units were removed about one year ago,  and no reheat
       has been used since.  Indirect reheat by mixing with heated air is being
       considered.

SCRUBBER OPERATING DATA

 1.  L/G is 8.5 gal/1000 acf, or 18 gal/1000 scf.
 2. AP is 20 to 22 inches H20 across the venturi, 28 inches overall.
 3.  "Open loop."  Total makeup water for the system is 1700 to 2000  gpm.
 k.  Gas residence time in the scrubber is not available.
 5.  Liquid delay time in the venturi recycle loop is about 2  minutes.
 6.  Liquid temperature leaving the scrubber is 120° F.
 T.  Solids recirculated has recently been increased from 2 pet to 6  pet.
 8.  pH of the recycle loop on the scrubber is 3.2 to 3-5.  pH at the thickener
       is U to 5 without lime added.  A level of pH 7-5 at the thickener is to
       be maintained with lime addition.
 9.  Scrubbing liquor analysis is not available.
10.  State of oxidation is not available.
11.  Degree of supersaturation is not available.

OPERATING REQUIREMENTS

 1.  Lime is added at the rate of 6 tons/day for pH control.
 2.  No dispersing agent is being used.
 3.  System makeup water requirements are about 3^00 acre ft/yr.
 h.  Power requirements.
     - Electrical requirements are 3 to k pet of generating capacity.
 5.  Manpower.
     - 8 operators.
     - Maintenance personnel not available.

OPERATING RESULTS

 1.  Particulate removal meets the goal of 99•2 pet.
 2.  S02 removal is 30 to 35 pet without lime.
     - No typical S02 removal has been determined with lime.
 3.  Availability overall is estimated at 80 pet.
 it.  Scaling has occurred on most wetted surfaces.
 5.  Methods used for scale control.
     - The level of recirculated  solids was recently increased from 2 to 6  pet.
     - A lime system has recently been installed to maintain pH at 7-5 in the
         thickener.
     - The amount of blowdown used helps to control scaling.
                                     310

-------
    Problems.
    -  The principal problem is  that  scaling  is not under control.   The  effects
        of high pH in the thickener  and  6 pet solids  in the recirculation
        scrubber liquor have not yet been assessed because of their recent
        implementation, and a long-range testing program is planned.
    -  Corrosion with resulting  leakage has occurred where coating  on mild steel
        have failed.
    -  Solids buildup has occurred  in blowdown lines.
    -  Deterioration of stack linings.
    Disposal of sludge.
    -  Sludge settles well without  a  flocculating agent.  The sludge is  concentrated
        to 30 or 35 pet solids  in  the thickener underflow and is pumped after
        some dilution with  blowdown  to decanting ponds.  Ash at present is left
        to accumulate in the ponds,  but  it may be dredged and returned  to the mine.
Scrubber Design and Operation
Aurora  Station
Minnesota Power and Light

LOCATION

 1.  Aurora, Minnesota.
 2.  Elevation 1500 feet.
 3.  Atmospheric pressure lU.l psi.
 4.  Annual rainfall approximately  25  inches.

SCRUBBER APPLICATION

 1.  Particulate removal, retrofit.
 2.  Two 58-MW pc-fired boilers.
 3.  Scrubber startup, June 1971-
 1*.  Fuel is Montana subbituminous  coal from the Big Sky Mine.
    -  8800 Btu/lb.
    -  26.7 pet moisture.
    -  1.37 pet sulfur (mine average).
    -  9 pet ash.
    -  9 to 13 pet CaO in ash.
 5.  Flue gas entering the scrubber on each boiler.
    -  291,160 acfm.
    -  3l+00 F.
    -  800 ppm S02.
    -  2.06 gr/scf.
 d.  Removal goals.
    -  Minnesota particulate standard  is  0.6 Ib/MM Btu.
    -  Scrubber guarantee is 0.03 gr/scf  or 0.078 Ib/MM Btu.

SCRUBBER DESCRIPTION

 1.  One Elbair spray-impingement scrubber for each of the two boilers.
 2.  One-stage of high pressure  spray  is  directed concurrent with  gas flow against
       vertical rods to be atomized into  fine droplets.

                                    311

-------
 3.  Vendor, Krebs Engineers.
 U.  Size of each scrubber is nominally 60 MW or 300,000 acfm.
 5.  Capital cost is not available.
 6.  Operating cost is not available.
 T.  Materials of construction.
    - Scrubber is 3l6 ELC stainless  steel.
    - Outlet  ducts are  flake-polyester coated  carbon steel.
    - Piping  is fiberglass,  or  rubber lined.
    - Pumps are rubber  lined.
 8.   Ho bypass.
 9.   Turndown is 0 to 110 pet.
10.   Demistor  consists of one bank of vertical  chevrons.
11.   No reheat, wet  fan.

SCRUBBER OPERATING DATA

 1.  L/G is 8.3 gal/1000 acf, or 13.3 gal/1000  scf.
 2  AP is 2.5 inches of H20 across scrubber, k inches  total.
 3'  Not closed loop, although scrubbing liquor is recycled from the ash pond
       back to the scrubber.  No clarifier is used.   Makeup water is approximately
       1200 gpm for each unit.
 U.  Gas residence time in the scrubber is approximately 3 seconds.
 5.  Liquid delay time in the recycle circuit is not available.
 6.  Solids circulated.
     - 0.02 pet entering scrubber.
     - 0.75 pet leaving scrubber.
 7.  pH estimated at 1*.5 in, U.1* out.
 8.  Scrubbing liquor analysis is not available.
 9   State of oxidation of dissolved sulfur is high, estimted over 90 pet sulfate.
 lo'.  Saturation of scrubbing liquor does not occur at the level of blowdown used.

 OPERATING REQUIREMENTS

 1.  No reagent is used.
 2.  Fcrubber water requirements are about 3500 acre ft/yr.
 3.  Power requirement.
      - Electrical power is about 0.5 MW per unit, or 0.0 pet.
      - No steam is used for reheat.
 U.  Manpower requirements are  not available.

 OPERATING RESULTS

 1.   Particulate removal  is  about  98 pet;  exit dust  loading is  0.0^ to  0.0^6 gr/scf.
  2.   S0? removal, occurring incidental to  particulate  removal by reaction with
        the  alkaline  fly ash and by sulfate removal  in  blowdown,  is  typically 20 pet.
  3.   Availability.                                                      .     .
      - The  Elbair  scrubber,  consisting  of a stainless  steel box containing hign
         pressure spray nozzles that are removeable section by  section  for
         maintenance, can remain on line without a bypass, even though  the spray
          system is not operating.   However, effective  operation requires a high
          level of maintenance effort.   A percentage availability in terms of
          effective operation was not obtainable.

                                         312

-------
     Scaling  and plugging.
     -  Scaling  and plugging has not  "been too  severe,  owing to the  amount of
        blowdown  used.
     Disposal of ash and spent  scrubbing liquor.
     -  Slowdown, estimated to contain 1 pet solid fly ash, is sent an  ash pond.
        Overflow  from the pond is neutralized with lime  "before  disposal.
     Problems.
     -  This unit would suffer from the same scaling and plugging problems as the
        Clay Boswell scrubber  of similar design, described  in the preceding
        section,  if the recycle loop were closed to  a similar extent.  As
        operated, it is less troublesome.
     -  The mist carryover problem is less severe than at  Clay Boswell  owing to
        operation at partial load with a resultant lower stack  velocity.
Scrubber Design and Operation
Clay Boswell Plant
Minnesota Power and Light

LOCATION

 1.  Cohasset, Minnesota.
 2.  Elevation approximately TOO feet.
 3.  Atmospheric pressure lU.3 psi.
 H.  Annual rainfall approximately 25 inches.

SCRUBBER APPLICATION

 1.  Particulate removal for a new plant.
 2.  350 MW Combustion Engineering pc-fired boiler.
 3.  Scrubber startup May 1973.
 k.  Fuel is Montana subbituminous coal from the Big Sky Mine,
     - 8800 Btu/lb.
     - 26.5 pet moisture.
     - 1.37 pet sulfur (mine average)
     - 9 pet ash.
     - 9 to 13 pet CaO in the ash.
 5.  Flue gas entering the  scrubber.
     - 1,300,000  acfm.
     - 251*0 F.
     - 800 ppm S02-
     - 3 gr/scf.
 £.  Removal  goals.
     - Minnesota  particulate standard is  0.6 Ib/MM Btu.
     - Scrubber guarantee is 0.03  gr/scf  or  0.078  Ib/MM Btu.
                                    313

-------
SCRUBBER DESCRIPTION

 1.  A single Elbair spray-impingement  scrubber.
 2.  One stage of high pressure spray is  directed concurrent  with gas flov
       against punch plate "baffles to be  atomized into  fine droplets.
 3.  Vendor, Erebs Engineers.
 U.  Size of scrubber is nominally 350  MW or  1.3  x 10°  acfm.
 5.  Capital cost is not available.
 6.  Operating cost is not available.
 7.  Materials of construction.
     - Scrubber is 3l6 LC stainless steel.
     - Outlet ducts are flake-polyester coated carbon steel.
     - Piping is fiberglass or rubber lined.
     - Pumps are rubber lined.
 8.  No bypass.
 9.  Turndown is 0 to 110 pet.
10.  Demistor consists of one bank of vertical chevrons.
11.  No reheat, wet fan.

SCRUBBER OPERATING DATA

 1.  L/G — 8.3 gal/1000 acf, or 13.3 gal/1000 scf.
 2.  AP is 2.k inches 1^0 across scrubber, k  inches  total.
 3.  Not "closed loop" although scrubbing liquor  is  recycled  from the clarifier
       back to the scrubber.  Makeup water is approximately 1^00 gpm.  About
       1*20 gpm of the makeup water compensates for clarifier  underflow to the
       fly ash pond.  The remainder is  clarifier  overflow which is pumped to
       the bottom ash pond.  The clarifier overflow  blowdown  maintains the
       sulfur level at about 3000 ppm.
 U.  Gas residence time in the scrubber is 3  seconds.
 5.  Liquid delay time in one clarifier is 2  hours,  or  U  hours if two clarifiers
       are on line.
 6.  Solids circulated.
     - 0.02 pet entering scrubber.
     - 0.75 pet leaving the scrubber.
 7.  pH 4.5 in, 4.0 out.
 8.  Scrubbing liquor analysis.
     - Ca — 600 ppm.
     - Mg — 200 ppm.
     - Na — 15 ppm.
     - SOlt — 2300 ppm.
 9.  State of oxidation of dissolved sulfur is high, estimated over 90 pet
       sulfate.
10.  Degree of supersaturation is not measured, but  it  is evident that it is
       variable depending on the amount of soluble alkali in  the ash, the amount
       of S02 absorbed, and the amount  of blowdown removed from the system.

OPERATING REQUIREMENTS

 1.  No reagent is regularly used.
 2.  Water requirements are about 2300  acre ft/yr.
                                    314

-------
 3.   Power  requirement.
     -  Electrical  power  is  about  3  MW,  or  0.86  pet  of  net  generating capacity.
     -  No steam is used  for reheat.
 k.   Manpower  requirements  are  not  available, but maintenance  is known to be
       very high because of a continuous  schedule of cleaning  on the mist
       eliminators,  which require sand  blasting to  clean.   As  of July 197U, 85
       to 100  manhours were spent each  week removing fly ash and calcium sulfate
       scale deposits.   In  the  past,  the  spray  nozzles also required continuous
       maintenance,  however,  with the increased blowdown now employed, the nozzles
       require only periodic  maintenance.

OPERATING RESULTS

 1.   Particulate removal is about 97  pet;  exit  dust loading is 0.08 pet
       gr/scf.
 2.   S02 removal,  occurring incidental  to  particulate  removal  by reaction with
       the  alkaline fly  ash and by  sulfate removal  in  blowdown, is typically
       15 to 20 pet.
 3.   Availability.
     -  The  Elbair  scrubber, consisting  of  a stainless  steel box containing
        high  pressure spray nozzles  that  are removable for maintenance section
        by section,  can remain on  line without a bypass,  even though the spray
        system is not operating.   However, effective  operation requires a very
        high  level of maintenance  effort.   A percentage availability in terms
        of effective operation was not obtainable.
 k.   Scaling and plugging.
     -  Scaling and plugging occur in  nozzles, nozzle trees, strainers, on
        punch plate baffles, in  the  wet-dry zone,  and on  the  fan and mist eliminator;
        and deposits fall  into the drains at the bottom of the scrubber.
     -  Scaling is  aggravated by any increase in CaO content in the ash of the
        coal  being burned, which causes  increased  SOg removal and an increase
        in the level of Ca++ and SO^ ions in solution.
 5.   Measures  used for scale control.
     -  Substantial amounts  of blowdown  are removed  from the system to remain
        below saturation with  CaSOlj.
     -  A very  high level of cleaning  and maintenance is carried out continuously.
        Nozzles and nozzle trees are removed and cleaned  on a rotating schedule,
        with  all  nozzles being cleaned once per week.
 6.   Disposal  of ash and spent  scrubbing  liquor.
     -  Blowdown from two clarifiers,  containing typically  5 or 6 pet fly ash,
        is sent to an 80-acre  ash  pond.   There is  no  net  evaporation, but
        rather an accumulation of  about  10 inches  of  water per annum,  z
 7.   Problems.
     -  Ho ultimate solution to  the  problem of disposing of sulfate-laden
        blowdown  water  has been  found.  The two possible  solutions would
        involve either  unrestricted  discharge  of diluted  blowdown to streams
        or closing the  loop in the scrubber circuit to eliminate blowdown.
        A  break-through in methods of  scale control would be  required to
        eliminate blowdown.
                                   315

-------
     - A stack mist problem results primarily from washing of the wet  fan to
         remove scale buildup.  Attempts to use steam soot blowers  in  place
         of washing and to apply non-stick coating have  been  unsuccessful.
         Some improvement has been achieved by reducing  the amount  of  fan
         washing.  Recent tests show favorable results by using  low-velocity
         stacks.
     Because of the difficulties experienced with flue gas mist  carryover, scale
       deposits on the mist eliminators and ID fan, disposal  of  sulfate-laden
       water and the high maintenance rate, the construction  of  similar  scrubbers
       for the two 70 MW boilers was not completed.  In  lieu  of  the particulate
       scrubbers, ESP's are presently planned.   Flue gas from all three  units
       would be discharged from one low velocity stack and the two  70  MW boilers
       should provide reheat for flue gas exiting the 350 MW  scrubber.
Scrubber Design and Operation
Dave Johnston Plant
Pacific Power and Light

LOCATION

 1.  Glenrock, Wyoming
 2.  Elevation approximately 5000 feet.
 3.  Atmospheric pressure 12.3 psi.
 k.  Annual precipitation is l*t inches.
 5.  Plant water supply is the North Platte  River.

SCRUBBER APPLICATION

 1.  Particulate removal, retrofit.
 2.  One 330 MW Combustion Engineering pc-fired boiler  (Unit no. k),
 3.  Three parallel scrubbers.
 k.  Scrubber startup was April 1972.
 5.  Fuel is Wyoming subbituminous coal  from a captive mine.  Typical analysis is:
     - 7^30 Btu/lb.
     - 26 pet moisture.
     - 0.5 pet sulfur.
     - 12 pet ash.
     - 20 pet CaO in ash.
 6.  Flue gas entering scrubber.
     - 1,500,000 acfm.
     - 270° F.
     - 500 ppm S02-
     - 12 gr/scf (design).
     - k gr/scf (actual).
 7.  Removal goal.
     - 99-7 pet removal,  or O.Oh  gr/scf  exit dust loading.
                                    316

-------
SCRUBBER DESCRIPTION

 1.   Three identical scrubbers in parallel.
 2.   Venturi  scrubbers.
 3.   Vendor,  the Chemical Construction Company (Chemico).
 U.   Initial  capital investment was $8 million,  $2^/lew.   Costs  incurred
       since  startup have increased this amount  significantly.
 5.   Operating cost  is not available.
 6.   Materials of construction.
     - Scrubbers, vessels, outlet duct, and  stack are polyester lined steel.
     - Piping and fan housing are rubber lined.
     - Fan wheels are Inconel.
 7-   No bypass.
 8.   Turndown is to  approximately 30 pet of  rated scrubber capacity.
 9.   Chevron  mist eliminator.
10.   Wet fans, no reheat,

SCRUBBER OPERATING DATA

 1.   L/G is 13.3 gal/1000 acf, or 22 gal/1000 scf.
 2.   AP is 10 inches of H20 across the venturi,  15  inches total.
 3.   Intermittently  "open loop."  Normal operation is attempted at  a  makeup rate
       of 500 gpm, which compensates for evaporation and loss in sludge.  The
       unit has been operated at times with  3000  gpm fresh water makeup to flush
       out scale.
 U.   Gas residence time in the venturi section of the scrubber  is estimated
       at about 1 second.
 5.   Liquid exit temperature is 126° F.
 6.   Liquid delay time in the venturi  recycle loop is 2  to 3 minutes.
 7.   Solids recirculated are 2 pet of scrubbing  liquor.
 8.   pH leaving the  scrubber is about  5-
       Future tests  will be conducted  maintaining the pH at 5.5 to  6.0 using
       hydrated lime.
 9.   Scrubbing liquor analysis is not  available.
10.   Degree of supersaturation is from 1.0 to 1.3.

OPERATING REQUIREMENTS

 1.   Lime is  added for pH control.
 2.   Lignosulfonate  is added to minimize scale at wet-dry zones.
 3.   Water requirements are approximately 800 acre-ft/yr in "closed loop" mode.
       Actual requirements are greater because of occasional flushing.
 k.   Power requirements.
     - Electrical power requirement is 7 to  8 MW, or 2.3 pet of
         generating  capacity.
     - No steam is used for reheat.
 5.   Manpower requirements are not available.
                                  317

-------
OPERATING RESULTS

 1.  Particulate removal meets the  outlet  grain loading  goal  of  0.1*  gr/scf.
 2.  S02 removal (preliminary values).
     - 35 to hO pet without lime.
     - Lime addition results in modest  increase in S02 removal.   Exact value
         has not teen determined.
 3.  Availability not available, but it is not considered  adequate for a utility
       power source.
 U.  Scaling and plugging.
     - Solids buildup has occurred  at the  wet-dry interface.
     - Hard gypsum scale has formed in  the scrubber vessels and  piping.
     - Solids plug bleed and recycle lines.
 5.  Methods for controlling scaling and plugging.
     - Lime for pH control has resulted in reduction but not  elimination of
         scaling.
     - Lignosulfonate addition has  resulted in a less adherent or friable wet-
         dry buildup.
     - Effects of hexamataphosphate on  scaling and wet-dry buildup were determined
         to be ineffective and its  use  discontinued.
     - Continuous fan vash has essentially eliminated buildup on fans.  The
         wash water is composed of  cooling water blowdown  and service water.
     - Fresh water washing has been required to flush ash  and scale  deposits
         from the scrubber vessel.
 6.  Additional problems.
     - Recycle pump erosion.
     - "Silting" during shutdown.
 7.  Disposal of sludge.
     - Bleed from the scrubber circuit  is  sent directly  to two ash ponds, from
         which overflow flows to a  clear pond for recycle  to  the scrubber
         circuit.  Each ash pond is dredged once per year, and the solids
         hauled out for landfill.   Excess  water resulting  from periods of flushing
         is discharged to the North Platte River under a variance from the State
         of Wyoming.
Scrubber Design and Operation
Valmont, Cherokee, and Arapahoe Stations
Public Service Company of Colorado

NOTE:  Public Service Company of Colorado has five TCA scrubbers containing
       thirteen modules installed on five boilers at three stations.  Because
       of the close similarity between these installations, they will be discussed
       collectively rather than individually.
                                   318

-------
LOCATIONS

                         	Valmopt             Cherokee             Arapahoe
                                                                Southwest Denver,
                         Boulder.  Colo.    Horth Dearer, Colo.           Colo.

Elevation, ft                 5300           Bst.  5200              Est. 5600
Atm. P,  psi                   12.1              12.1                   12.1
Annual  rainfall,  in.           Ik                lit                     -^
Plant water supply      Hillcrest Lake    South Platte River    South Platte River

SCRUBBER APPLICATION

 1.  All units are retrofits for  particulate removal; Valmont #5 has one module
        modified to study  SOg removal.
 2.  Boilers equipped with scrubbers  (all pc-fired).
 3.  Valmont #5,  196 MW,  2 scrubber modules, November 1971.
     Cherokee #1, 115 MW,  1 scrubber vessel, 2 modules, June 1973.
     Cherokee #3, 170 MW,  1 scrubber vessel, 3 modules, Bbvember 1972.
     Cherokee A, 375 MW,  1 scrubber vessel, k modules, July 197^.
     Arapahoe #U, 112 MW,  1 scrubber vessel, 2 modules, September 1973.
 It.  Arrangements of particulate  cleaning equipment.
     -  At Valmont, flue gas from  a mechanical collector is split into two parallel
          streams, with 50 pet sent to the scrubber and 50 pet to an electrostatic
          precipitator (ESP).
     -  All other units have a mechanical collector, an ESP,  and scrubber(s) in
          series, with all  flue gas entering the scrubber(s).
 5.  Coal burned at Valmont and Arapahoe is Wyoming subbituminous.
     -  8300 Btu/lb.
     -  29 pet moisture.
     -  0.6 pet sulfur.
     -  5.2 pet ash.
     -  20 pet CaO in ash.
 6.  Coal burned at Cherokee  is Colorado bituminous coal.
     -  11,000 Btu/lb.
     -  9.8 pet moisture.
     -  0.7 pet sulfur.
     -  9-^ pet ash.
     -  5 pet CaO in ash.
 7.  Flue gas entering the stack  gas cleaning train.
                    Valjmont #5  Cherokee #1  Cherokee #3  Cherokee #U  Arapahoe
#1*
Flow, acfm          H63,000      520,000      610,000     1,520,000     520,000
T» °F                  260          295         280          275         300
S02, ppm  (est.)       500          500         500          500         500
      gr/scfd         0.8          0.8         O.k          0.7         0.8
 8.  Removal goals for particulate.
     - The  applicable Colorado State Standard is 0.1 Ib/MM Btu, or about 0.05
          gr/scf .
     - The  company's desire for clean stacks requires a goal of 0.02 gr/scf.
                                         319

-------
SCRUBBER DESCRIPTION

 1.  All units are Turbulent Contact Absorbers, consisting of three stages of
       motile' packing, or "ping pong balls," with spray directed downward through
       the balls and gas passing countercurrent upward.
 2.  Vendor:  Air Correction Division, Universal Oil Products Company.
 3.  Capital cost:

     Valmont #5   Cherokee #1   Cherokee #3   Cherokee #k   Arapahoe #h

     #3,600,000   $3,810,000    $1*, 1*00,000    $12,700,000   $1*, 560,000
       $32/kw       $33/kw        $29/kw         $33/kv       $1*1/kw

 k.  Operating costs for Cherokee #3 is 0.50 mills/kwh.  Other operating costs
       are not available.
 5-  Materials of construction (typical).
     - Scrubbers are rubber-lined carbon steel with stainless steel grids.
     - Exit ducts are mild steel with stainless steel and/or fabric expansion joints.
     - Slurry piping is rubber lined.
     - Pumps are rubber lined.
 6.  Bypasses on all units.
 7.  Typical turndovn is 1*7 to 105 pet.
 8.  Mist eliminators have 2 stages, 7 passes.
 9.  Demisters for Cherokee #1 and #3 are fiberglass reinforced plastic.
     Demisters for Cherokee #1*, Arapahoe #1* and Valmont #5 are stainless steel.
10.  All units except Cherokee #1+ reheat the flue gas directly with steam coils;
     Cherokee #1* uses externally-heated air.
11.  All units have dry fans that are forced-draft with respect to the scrubber.

SCRUBBER OPERATING DATA FOR CLOSED LOOP MODULE AT VALMONT #5

 1.  L/G of about 50.
 2. AP is approximately 10 to 15 inches of J^O.
     - Three stages of mobile packing (8-12 inches per stage).
     - Stainless steel mist eliminators.
     - Direct steam coil reheater, 1 1/2 to k inches.
     - Transition ductwork, 1/2 inch.
     - Total, 16 1/2 inches maximum.
 3.  Makeup water is 70 gpm.  The water analysis is as follows:
     - SOi, — 45 ppm.
     - Ca — 35 ppm.
     - Cl — 10 ppm.
 h.  Gas residence time is not known.
 5.  pH entering is maintained at 5-5 to 6.5 by limestone addition.
 6.  State of oxidation of dissolved sulfur is nearly 100 pet sulfate.
 7.  Suspended solids maintained at 7 pet.
 8.  Liquid temperature is 110° F.
 9.  Degree of supersaturation is not known.
                                     320

-------
SCRUBBER OPERATING DATA FOR OPEN LOOP SYSTEMS

 1.  L/G is typically 5^-59-
 2.  AP is approximately 10 inches to 15 inches of ^0 depending on design
        and operating conditions.
     -  3 stages of mobile packing — 8-12 inches.
     -  Mist eliminators.
           - Fiberglass reinforced plastic, 1 1/2 to 3 inches.
           - Stainless steel — 1/2 inch.
     -  Reheat.
           - Direct steam coil reheater, 1 1/2 to k inches.
           - Hot air — negligible.
     -  Transition ductwork —       1/2 inch	
     -  Total                        16 1/2 inches maximum
 3.  "Open loop."  Amounts of makeup water from cooling tower blowdown are as
        follows:
Cherokee #1
203 gpm
Cherokee #3 Cherokee #U
380 gpm jkh gpm
Arapahoe #k
203 gpm
Valmont #5
(unmodified
module )
210 gpm
 k.   Gas residence time in the scrubber is  3.8 to 5 seconds.
 5.   Liquid temperature leaving the scrubber is 110° F.
 6.   Liquid holdup time in the recycle circuit is very short,  estimated at
        ten seconds.
 7.   pH is T to 9 entering; 2.8 to 3 leaving the scrubber.
 8.   A scrubber liquor analysis for Valmont (not known to be representative).
      - Ca — 590 ppm.
      - Mg — 350
      - Na — 1
      - SO^ — 10,000
 9.   State of oxidation of dissolved sulfur is not available.
10.   Degree of super saturation is not available.

OPERATING REQUIREMENTS

 1.   No lime or other reagent or additive is normally used.
 2.   Water requirements in acre-ft/yr (approximate).
Valmont #5
3^0
Cherokee #1
327
Cherokee 13
612
Cherokee #h
1200
Arapahoe tik
327
 3.   Power requirements.

      Electric:  Power:
      Valmont #5    Cherokee
            Cherokee #3    Cherokee #k    Arapahoe #k
        5.3 MW
        5.^ %
    MW
*. 5
6.k m
3.8 %
it.U m
3.8 %
5.2 m
U.6 %
                                       321

-------
     Steam for reheat:

     Valmont #5    Cherokee »1    Cherokee #3    Cherokee #k    Arapahoe #h

Ib/hr:  50,000        50,000         1»1,200        135,000         60,000
°F:      TOO            teO            715            ^            360
psi:     1*90            300            300           1,975           150

 U.  Manpower for scrubber operation is not identified "by the company as a
       separate category from plant oeprators.   However, it is reported that  a
       high degree of maintenance is required.

OPERATING RESULTS

 1.  Particulate removal achieves an outlet grain loading of 0.02 gr/scf.
 2.  S02 removal.
     - ^5 to 50 pet at Valmont and Arapahoe, burning Wyoming coal,
     - 15 to 20 pet at Cherokee, burning Colorado coal.
     - 85 pet at Valmont using modified module (not used continuously).
 3.  Availability.
     - Valmont #5 —  55 pet for k6 months of operation.
     - Cherokee #1 — 53 pet for 23 months of operation.
     - Cherokee #3 — 66 pet for 35 months of operation.
     - Cherokee #U — 82 pet for 10 months of operation.
     - Arapahoe #U — 8U pet for 25 months of operation.
 ^.  Problems.
     - Scaling and plugging has occurred at:
          - The wet/dry zone.
          - The first stage grid.
          - The reheater.
          - The mist  eliminators.
     - Corrosion has  caused major failures of reheaters at the Cherokee Station.
     - Wear on the balls used for packing requires replacement after 6000  hours
         or less.
     - The cost of replacing the balls are 0.06 $/ball.  The number of balls
         required is  as follows:
          - Valmont #5 —  870,000 balls.
          - Cherokee  #1 — 980,000 balls.
          - Cherokee  #3 — 1,180,000 balls.
          - Cherokee  #U — 3,070,000 balls.
          - Arapahoe  #k — 980,000 balls.
  5.   Measures for control  of  scaling.
      - Additives tried have not worked, including phosphated esters.  The  studies
         were limited in that they were of relatively  short duration.
      - Slowdown must  be maintained at an adequately high level, but otherwise
         no  specific  methods  are being used.
      - One  of two modules  at  the Valmont Station has been modified to use
         limestone  for pH  control.  The preliminary results are favorable:
         however,  additional  tests are planned for the future.
                                        322

-------
 6.  Sludge disposal.
    - At the Arapahoe Station  slurry blowdown is  pumped to an ash settling
        pond after adjusting the pH  to  6.0  - 9.0  with lime.   The settled fly
        ash is dredged periodically  and used for  landfill.   Clear effluent
        from the ponds is discharged under  permit from the State of Colorado,
        to the South Platte River  at the Cherokee and Arapahoe Stations, and
        to the cooling pond at the Valmont  Station.   At the  Cherokee Station
        only slurry blowdown is neutralized and clarified and then mixed with
        ash pond overflow for  discharge to  the river, owing  to the heavy
        loading on the ponds caused  by  discharge  from the three scrubber-equipped
        boilers at this station.


iCKNOWLEDGMENTS

    Information contained in this  paper was obtained  from published sources
and, from inquiries directed to'utilities operating wet scrubbers on
boilers burning Western coals and to  scrubber vendors.   Persons who have
contributed information include Dr. Fred Murad of  Combustion  Equipment Associates,
Mr. Ken Vig of Minnkota Power Cooperative, Mr.  Dennis  Van Tassel and
Ifr. ELdon Kilpatrick of Minnesota Power  and  Light,  Mr.  Phil Richmond of Square
Birtte Electric Cooperative, Mr. George Greene and  Mr.  Steven  Goering of Public
Service of Colorado, Mr. Thomas Ashton of Pacific  Power and Light Company,
Br. Carlton Grimm of Montana Power  Company,  Mr.  John Noer of  Northern States
Bower, Mr. Walter Ekstrom and Mr. Aubrey Parsons of Arizona Public Service.  The
Contributions of the above and others  who have freely  exchanged information in
numerous past contacts concerning flue gas desulfurization are gratefully
acknowledged.

BH'ERENCES

 1.  U.S. Bureau of Mines, Division of Fossil Fuels.   Coal—Bituminous and
      Lignite in 1973.  Mineral Industry Surveys,  January k,  1975,  p 5.

 2.  Nielsen, G.F.  Coal Mine Development  Survey.   Coal Age.  V.  80,  February
      1975, pp 130-139.

 3.  Energy Research and Development Administration.   Open file report.   Survey
      of Coal and Ash Composition and Characteristics  of Western Coals and
      Lignite.   Grand Forks, ND, 1975.

.\.  Gronhovd, G.H., et al.   Some Studies on  Stack  Emissions  from Lignite Fired
      Power Plants.  BuMines 1C 8650, 197'k,  pp  103, 133.

 5.  Facts Sheet.   Four Corners Powerplant, Farmington,  NM, April  1973.

 6.  Quig, R.H.   Chemico Experience for 50^ Emission Control on Coal-Fired
      Boilers.   Presented at the Coal and the Environment Technical  Conference,
      Louisville, KY.   October 23,  197^.

-------
 7.  Borgwardt, R.H.  EPA/RTP Pilot Studies Related to Unsaturated Operation  on
       Lime  and Limestone Scrubbers.

 8.  Tufte,  P.H.,  et  al.  Pilot Plant Scrubber Tests to Remove S02 Using
       Soluble Alkali in Western  Coal Ash.  BuMines 1C 8650, 197^, PP 103-133.

 9  Sondreal, E.A. ,  et al.  Wet  Scrubbing of S02 with Alkali in Western  Coal
       Ash.   Paper No.  7^-272,  67th Annual Meeting of the Air Pollution Control
       Association,  June  9-13,  197^,  31 pp.

10.   Sondreal, E.A.,  and  P.H.  Tufte.  Scrubber Developments in the West.   Presented
       at the Lignite Symposium,  Grand  Forks, North Dakota, May 1^-15, 1975-

11.   LaMantia, C., et al.   EPA-ADL Dual Alkali Program Interim Results.   Presented
       at EPA Symposium on Flue Gas Desulfurization, Atlanta, Georgia, November
12.  Kilpatrick, E.R., and H.E. Bacon.  Experience with a Flue Gas  Scrubber on
       Boilers Burning Subbituminous Coal.   American Society of Mechanical
       Engineers Winter Annual Meeting, New York,  NY, November 1971*.   Paper No.
       71+-WA/APC-3.

13.  LaMantia, C.R., and I. A. Raben.  Some Alternatives for S02 Control.   Presented
       at Coal and the Environment, Technical Conference sponsored  by the
       National Coal Association, October 22-21* , 197)1.

Ik.  Noer, J.A., et al.  Results of a Prototype Scrubber Program for the
       Sherburne County Generating Plant.  Presented at the IEEE-ASME Joint
       Power Generation Conference, Miami Beach, Florida, September 15-19,

15.  Ashton, T.M.  Operating Experience Report, Flue Gas Scrubbing  System,
       Dave Johnston Steam-Electric Plant Unit h, Pacific Power and Light
       Company, presented at the American Society of Mechanical Engineers
       National Symposium, Philadelphia, PA., April 1973.

16.  Green, G.P.  Operating Experience with Particulate Control Devices.
       Presented at the American Society of Mechanical Engineers National Symposium,
       Philadelphia, PA, April 1973.

17.  Murad, F.Y., et al.  Boiler Flue Gas Desulfurization by Fly Ash Alkali.
       Presented at Mid-Continent Area Power Pool  (MAPP) Environmental Workshop,
       Minneapolis, MN, November 18, 1975-
                                     324

-------
            RESULTS OF THE 170 MW TEST MODULES PROGRAM
                     MOHAVE GENERATING STATION
                SOUTHERN CALIFORNIA EDISON COMPANY
      Alexander Weir,  Jr., Lawrence T. Papay, Dale G. Jones,
              John M.  Johnson, and William C. Martin

                Southern California Edison Company
                           P. 0. Box 800
                     2244 Walnut Grove Avenue
                    Rosemead, California  91770
ABSTRACT

     This paper summarizes the performance of three different types
of scrubbers tested with lime and limestone reagents at the Mohave
Generating Station in South Point, Nevada.  Each scrubber was designed
to treat 450,000 SCFM of flue gas (170 megawatt equivalent) and was
larger than any other single scrubbing module which has been operated
in the world today.  The Horizontal Module (a horizontal cross flow
spray scrubber) was operated from January 16, 1974 to February 9, 1975.
The Vertical Module was tested in two modes; first as a Turbulent
Contact Absorber (TCA) from November 2, 1974 to April 30, 1975 when
the thermoplastic "ping pong balls" were removed and second as a Polygrid
Packed Absorber (PPA), with a plastic "eggcrate" packing with testing
continuing to July 2, 1975.

     The effects of recirculating slurry flow rate, flue gas flow rate
(turndown)  and number of scrubbing stages on S0_ and particulate removal
are presented.  Performance of two types of flue gas reheaters and two
types of mist eliminators are discussed.  Power requirements, reagent
utilization factors, scrubbing system, availability records and closed
loop water system operations are described for three types of scrubbing
systems tested at Mohave.
                                  325

-------
                           FOREWORD

     The Test Modules Program was a joint  venture  of the Navajo and
Mohave Power Project participants who are  listed below:

     Salt River Project Agriculture 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 U.S. 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.  Southern California Edison Company was  the
project manager of the Test Modules Program.

     The conclusions presented in this paper represent the personal
opinions of the authors and are not intended to represent the opinions
or position of any of the project participants.
                                  326

-------
                           INTRODUCTION

     At  the  second  EPA Flue Gas Desulfurization Symposium in New Orleans
in 1972,  Dr.  Jerry  Shapiro1 of the Bechtel Power Corporation presented
the plans of  the  Navajo and Mohave participants to test, on a pilot plant
scale, four  different  types of scrubbers with three different reagents
(lime, limestone, and  soda  ash).   The results of these tests, as well
as tests  of  four  additional scrubbers with one additional reagent (ammonia),
were presented  at the  third Flue  Gas Desulfurization Symposium in New
Orleans  in May  1973.    At that time it was indicated that contracts had
been let  in  December of 1972 to construct two 170 MW Test Modules at the
Mohave Generating Station.   Both  of these scrubbers were larger than any
other single  scrubbing module which had been previously operated in the
United States,  or in the world.  With a rated capacity of 450,000 SCFM,
that statement- is still true today.  Testing of the Horizontal Module
was initiated January  16, 1974 and the results of the first nine months
of operation  were presented at the fourth Flue Gas Desulfurization Sym-
posium in Atlanta on November 4,  1974.  Testing of this module was com-
pleted February 9,  1975 after 5927 hours of operation and it has since
been dismantled and reassembled at the Four Corners Generating Station.

     Start-up of  the Vertical Module was initiated on schedule January 1,
1974, but on  January 24, 1974 a disasterous fire burned most of the
chlorobutyl  rubber  lining and start-up was not resumed until October 1,
1974.3   Testing of  the Vertical Module was initiated November 1, 1974,
and was  completed July 1, 1975 after 3131 hours of test.  Two different
configurations  of the  Vertical Module were tested.  The first configura-
tion was  the  TCA  (Turbulent Contact Absorber) configuration in which four
Stages and later  three stages of  thermoplastic rubber balls were used.
Later, the balls  were  removed and replaced with a plastic "eggcrate"
type packing  referred  to in this  paper as the PPA (Polygrid Packed Absorber)
configuration.  The Vertical Module is presently shut down in a cold
standby  condition at the Mohave Generating Station.

     It  is the  purpose of this paper to present the results of testing
of these  three  types of scrubbers at the 170 MW scale, as well as results
of testing of two types of  reheaters, and two types of demisters.  Not
included  in  this  paper, for lack  of time, were the results of testing
various  scrubber  lining materials.  Six types of slurry pumps, as well
as rotary filters,  centrifuges and two commercially available methods
of sludge fixation  (as well as sludge ponding) were also evaluated in
this program  but  these results also will not be presented in this paper.
Since another session  of this symposium is directed to slurry and sludge
handling,  EPA has requested that  we not discuss this aspect of a total
scrubbing system  in this paper.
                                  327

-------
                        TEST MODULE DESCRIPTIONS

170 MW Vertical TCA and PPA

      A ground-level photograph of the Vertical Module is shown in
Figure 1, where the flue gas enters the scrubbing chamber at the lower
left and exits from the top of the scrubber.  Although this photograph
does not show the front end limestone preparation area,  back-end thickener
nor sludge dewatering equipment, it does indicate the height to the
top of the outlet ductwork, which is about 145 feet above grade.
The ductwork shown at the left side of Figure 1 supplies flue gas
to the.scrubber from the down stream side of the Unit 1  electrostatic
precipitator.  A 5,500 HP booster fan forces the flue gas through
the inlet ducting, scrubber packing and mist eliminator  section into
the outlet ducting where it is heated by direct-contact  steam coils
and is finally returned to the ductwork leading to the 500 foot stack.

      The majority  (2,342 hours) of the Vertical Module  test program
was conducted with  the Turbulent Contacting Absorber (TCA) configuration
shown  in Figure 2.  The scrubbing chamber dimensions are 18 feet wide
by  40  feet long and 38 feet high. The TCA configuration consisted
of  four  stages of thermoplastic rubber balls supported on stainless
steel  grids  at four foot intervals. The balls were contained in compart-
ments, with  15 compartments at each of the tour levels.  Although various
levels of ball depths were tested, the compartments were initially
filled to the  one-foot level with approximately 1,600,000 balls. The
TCA configuration was  tested with both three and four stages of balls
and at ball  depths  of  6, 10 and 12 inches in the four stage configuration.

       During that last phase of the program, the TCA configuration
was replaced with a Polygrid Packed Absorber  (PPA), as shown in Figure
3.  The  Vertical Module was operated for 789 hours in the PPA configur-
ation. The Polygrid packing consisted of plastic grids in an "eggcrate"
configuration where each grid  layer was 1 1/4"  thick with 2" square
openings.  The grid layers were stacked to a depth of approximately
17  inches  in each stage.

       The  Vertical  Module  recirculating slurry  flow rate was normally
16,000 gpm  in the TCA  configuration and 27,000  gpm in the PPA configur-
ation.  A small quantity of  the recirculating  slurry was discharged
to a  dewatering  complex to  extract solids from  the scrubber  slurry.
Water reclaimed  in  the dewatering process was  returned to the limestone
 slurry mix  tank.  The dewatering complex consisted of a thickener  tank,
with  further dewatering provided  by either a  centrifuge or vacuum  drum
 filters.

 170 Horizontal Module

       The photograph  shown in  Figure  4  shows  the size and general
 arrangement of the  basic  scrubber  shell, ductwork and major  components
 of the Horizontal  Module.  The  horizontal end-to-end length  including
 the booster fan,  demister  section and  ductwork is  150 feet  long.  The
 width varies from 28  to  32 feet and  the  top  of the scrubber  shell
 is 33 feet above grade.  The ductwork  shown on the  right  side of Figure
                                 328

-------
 ••
'• '
'
                     FIGURE1 - 170 MW VERTICAL MODULE

-------
                       Figure 2

      I70MW VERTICAL MODULE (4STAGE  TCA)
   SLURRY SPRAYS
FLUE GAS
FROM FAN
   SUMP CHAMBER
                         SCRUBBED FLUE GAS
                         TO DEMISTER

                             A
                                                 THERMO PLASTIC
                                                 RUBBER SPHERES
                                                 "QUIESCENT" SPHERE
                                                  DEPTH OF 6" PER
                                                  STAGE
                             330

-------
                        Figure 3

               I70MW VERTICAL MODULE
          ( 3 STAGE  POLYGRID PACKED ABSORBER )
                     SCRUBBED FLUE GAS
                     TO DEMISTER
 SLURRY SPRAYS
                                             17 PACKING PER
                                             STAGE
SUMP CHAMBER

-------
r.i
•<
< <
                       FIGURE 4 - 170 MW HORIZONTAL MODULE

-------
4 supplies flue gas  to  the  scrubber from the downstream side of the
Unit 2 electrostatic  precipitator.  A 1,750 HP booster fan forces the
flue gas through  the  scrubber  shell and demister section into the outlet
ducting where  it  is  heated  by  injecting hot ambient air and is finally
returned to  the precipitator outlet duct upstream from the 500 foot
stack.

    The four-stage  scrubbing  chamber of the Horizontal Module is
shown  in Figure  5 and is  48 feet long, 28 feet wide and 15 feet high.
The Horizontal Module did not  contain packing, but consisted of four
stages of  cross  flow spray. The slurry was cycled through the scrubber
in a countercurrent  manner. That is, the fresh lime slurry from the
mix tank was first sprayed across the flue gas at the fourth stage,
or discharge end  of  the scrubbing chamber. The same liquid was successively
collected  and pumped to the third, second, and first stages and success-
ively  depleted of alkalinity.  By the time the slurry reached the first
stage  collection  hopper for return to the lime mix tank, it was almost
completely depleted  of any excess alkalinity.

      The  sprays  were discharged from a row of  36 externally-mounted
nozzles  at each  stage.  The recirculating slurry  flow rate was normally
 9,000  gpm, but the slurry was mechanically pumped four  times per circuit
 for a  total  installed pump capacity  of  36,000 gpm. As in the Vertical
 Module,  a  small  quantity of recirculating slurry  was discharged to
 adewatering complex to extract solids  from the scrubbing  system.
 Water  reclaimed   in the dewatering  process was returned  to  the  scrubber
 slurry lime  mix   tank.  However, in  the  case of  the Horizontal  Module,
 the dewatering complex consisted of  a  thickener tank and a  sludge
 disposal pond.


 TEST CONDITIONS

     Both test modules were tested under  a  variety  of  configurations
 and conditions to determine the significant  parameters  affecting  SO2
 and particulate  removal, operating simplicity  and system chemistry.
 The test program for each  module  was divided  into test  blocks  and
 were designed with  specific values for  selected variables.  This  approach
 was adopted  to simplify  data  analysis and permit  the effect of
 each variable to be  determined  independently.   The  variables considered
 aost relevant are listed below:

 Type of reagent
 Demistet  wash procedure
 Number of scrubbing  stages
 Percent cooling  tower  blowdown  in makeup water
 Spray nozzle  pressure
 Flue gas  flow rate
 Circulating  liquid  flow  rate
 Inlet S02 concentration
 Inlet particulate grain  loading
 pH of the scrubber  slurry
 Method of water  balance  control
 Percent solids  in scrubber slurry
 Percent solids  in reagent  feed slurry

-------
                                  FIGURE 5
                         170 MW HORIZONTAL MODULE
                                  (4 STAGE)
  FLUE GAS
/ FROM FAN
SCRUBBER
FLUE GAS
TO DEMISTER

-------
TEST PERFORMANCE  RESULTS

S02  Removal
«l,,rrv  ^mparifon  °f S°2 removal efficiency as a function of recirculating
slurry  flow  rate  for the Horizontal and Vertical Modules is shown in
Figure  6.  Note  that  with low sulfur Western coal fired at the Mohave
Generating Station,  the S02 concentration in the stack gas and at
the  scrubber  inlet averaged only 200 ppm. At this low inlet S02 concen-
tration,  bOz  removal above 95% was achieved with all three scrubbers.
For  example,  the  recirculating liquid flow rate required for 96% SO2
removal was  9,500  gpm with the Horizontal Module 18,500 gpm with the
Vertical  Module  in the TCA mode and 27,800 gpm with the Vertical Module
in the  PPA mode.

Effect  of  Staging  on SO2 Removal

     The  percentage  S02 removal in the Horizontal Module as a function
of the  number of  scrubbing stages is shown in Figure 7. The first
stage removes 50%  of the SO2.   The second stage removes 64% of the
remaining  S02 for  a  total S02  removal with two stages of 82%. Each
successive stage  removes some  of the remaining S02 and although the
total percentage  removal increases, the total weight of S02 removed
in each additional stage decreases. The Horizontal Module design allows
a selection of  the appropriate number of scrubbing stages to obtain the
desired degree  of  S02 removal. Since the Horizontal Module can continue
to operate with  the  loss of all but one stage, it was relatively easy
to obtain  the effect of staging by selectively shutting off stages.
It was  more difficult to obtain this type of data with the Vertical
Module  since  shutdown was required to remove the packing. However,
the  Vertical  Module  was operated in the TCA mode with 3 and 4 stages
and  in  the PPA  mode  with 2 and 3 stages. The results of the effect
of staging on S02  removal were in general agreement with those obtained
with the  Horizontal  Module.

Effect  of  Turndown on S02 Removal

     All  three  scrubbers showed a slightly increased degree of S02
removal when  the gas flow rate was decreased from 450,000 SCFM and
the  circulating  liquor flow rate was allowed to remain constant. It
should  be  recognized that the  SO2 concentrations at the exit of the
scrubber  were extremely low and wet chemical analysis, rather than
instrument readings,  were used. The measured exit valves were corrected
(i.e. increased) for  the dilution effects of water vapor, and in the
case of the Horizontal Module  for the reheat air so that they were
comparable to the  inlet values.  The effect of turndown on exit S02
concentration is shown in tabular form in Figure 8.

Particulate Removal  and Turndown

     At the design operating condition of 450,000 SCFM and 0.10 gr/SCF
inlet particulate  grain loading, both the Horizontal and Vertical
Modules achieved a relatively  high degree of particulate removal.
This is shown in Figure y,  where the Vertical TCA achieved 93% removal,
the  Horizontal  Module ahieved  92.5% removal and the Vertical PPA

-------
                               FIGURE  6
               EFFECT OF CIRCULATING  LIQUOR FLOW RATE
                ON S02 REMOVAL AT CONSTANT GAS FLOW
                            G=450,000  SCFM
  100-
   99-
   98-
   97-
      TCA 4 STAGES
      LIMESTONE
UJ
cc
 CJ
UJ
UJ
Q_
UJ
   95-
   94-
   93-
HORIZONTAL
4 STAGES
LIME
                   PPA 3 STAGES
                   LIMESTONE
   92
   91-
   90-
                        10
   15
20
25
30
                     CIRCULATING LIQUOR FLOW RATE
                             (1000 GPM)           336

-------
  100
                            FIGURE 7
                      170 MW  HORIZONTAL
               S02 REMOVAL VS. NUMBER OF STAGES
   so-
il!
8
I!
   40-
   CONDITIONS
GAS FLJOW= 450,000 SCFM
L/G = 206PM/IOOOSCFM
INLET S02= 220 ppm LIME
   20
                         NUMBER OF STAGES
                              337

-------
                                FIGURE 8
                   EFFECT OF TURNDOWN ON S02 REMOVAL
00
NUMBER OF STAGES
REAGENT
CIRCULATING LIQUOR
FLOW RATE  (GPM)
INLET S02 CONCENTRATION
(ppm)

EXIT S02 CONCENTRATION
(ppm)
     at 450,000 SCFM INLET
     at 300,000 SCFM INLET
HORIZONTAL

    4

  LIME

  9,500


  220
                                  4
                                  2
                                             VERTICAL
                                               TCA
18,000
                                               220
                  II
                  9
                                                     VERTICAL
                                                       PPA
                                            LIMESTONE     LIMESTONE
27,000
              220
               8
               3

-------
  100-
                                  FIGURE 9
              EFFECT OF TURNDOWN ON ARTICULATE  REMOVAL
                  INLET GRAIN  LOADING=0.10 GRAINS/SCF
   95-
          HORIZONTAL CIRCULATING
          LIQUOR = 9000 6PM
          4 STAGES 	

-------
achieved 91.5% removal.  Note that this is a relatively  high  weight per-
centage removal for the fine particles remaining in the  flue  gas after
passing through an upstream electrostatic precipitator.

     Note that the particulate removal efficiency of the Vertical
TCA and PPA decreases as the flue gas flow rate  is reduced, while
the particulate removal efficiency of the Horizontal Module increases
as a function of turndown ratio.

     The Vertical TCA and PPA Modules achieve increased  particulate
removal at increased flue gas pressure drop levels. For  example, at
450,000 SCFM, the pressure drop across the scrubbing chamber  was 14
inches of water for the TCA and 12.2 inches for  the PPA.  At 33% of
rated capacity with the design liquid flow rate  the pressure  drop
was 3.5 inches of water in the TCA and 2.0 inches in the PPA.  Note
that at one-third of rated capacity with the design liquid flow rate,
the liquid to gas contacting ratios are three times higher than the
design value. However, the percentage particulate removal decreased
with decreasing flue gas pressure drop from 93%  to about 70%.

     In the case of the Horizontal Module, particulate removal is
primarily a function of the volume of flue gas contacted by the falling
spray droplets. As flue gas flow rate is turned  down from 450,000
SCFM to 150,000 SCFM, the pressure drop across the scrubbing  chamber
decreases from 1.0 inches of water to 0.10 inches of water. Since
the liquid flow rate is constant, a given amount of the  gas  is contacted
by three times as many droplets at one-third load as at  full  load.
The particulate removal efficiency was observed  to increase from 92.5%
at full load to 96.5% at one-third load.

Closed Loop Water System

     One of the objectives of the Test Modules Program at Mohave was
to demonstrate the operation of a closed loop water system. This means
that all the liquid water discharged from the scrubber  is recycled
for use except the water which cannot be reclaimed from  the waste
sludge.  It should be noted that two molecules of water  for each molecule
of S02 removed from the flue gas exist in the sulfate sludge  as water
of hydration, or 36 pounds of water for each 32  pounds of sulfur.
Since wet scrubbers act to cool and saturate hot inlet  flue gas, a
majority of the makeup water requirement in a closed loop scrubbing
system is to replace the water discharged as water vapor from the
stack.

     A second objective of the Test Modules Program was  to utilize
up to 75% cooling tower blowdown water in the makeup water for a
closed loop scrubbing system. The cooling tower  blowdown water contained
approximately 12,000 ppm total dissolved solids. The water balances
measured during the test period are shown in Figure 10.  The Horizontal
Module was capable of operating with 75% cooling  tower  blowdown
in the makeup and the Vertical PPA was capable of operating with  40%    g
blowdown in the makeup for scale-free operation during  extended  testing.
Prior to converting the Vertical Module to the PPA mode, considerable
scaling occurred when the Vertical Module was operated with -65%  cooling
 a.  The  two  stage PPA configuration was operated for 502 hours with  83%
     cooling  tower blowdown water as makeup.       340

-------
tower blowdown in the makeup water. This scaling may have been due-to
the presaturator configuration, later changed, rather than solely clue to
tha amount of cooling tower blowdown water used as make up. This prevented
obtaining satisfactory water balance data as is shown in Figure 10
for the other two scrubbers.

     Both modules thus demonstrated the ability to achieve over 90% of
the makeup water discharged from the stack as pure water vapor  Note
that the Horizontal Module was capable of scale-free operation at
a slurry dissolved solids level of 185,000 ppm. The Vertical PPA slurry
dissolved solids level was 70,000 ppm at the reduced consumption of
cooling tower blowdown required for scale-free operation. The. dissolved
S cn/epr!Sen^d W6re Iar9elv sodium chloride (NaCl), magnesium sulfate
(MgS04) and sodium sulfate (Na2 S04).

     Table 11-2  of the March 1975 National Academy report  indicated
that closed loop operation of the Horizontal Module was achieved at
Mohave, but indicated in a footnote "Mohave and Cholla experience
little  rainfall  and water losses due to evaporation from their sludge
ponds are significant".  It is true that Mohave experiences little
rainfall,  but it is not  believed that a 4 gpm loss by evaporation
from the sludge  pond is  significant compared to the 156 gpm makeup
water requirement.   The  presence of rainfall had no measurable influence
on the  Vertical  Module where centrifuging or filtering was used as
the dewatering technique.  The liquid water blowdown rate in the Vertical
Module  was 6.5 gpm compared to a 160 gpm makeup water rate.

     The key  to  closed loop operation is in the dewatering complex,
where water  is reclaimed from purged scrubber .solids and returned
to the  scrubbing system.  This allows the makeup water flow rate to
be a function primarily  of flue gas flow rate by providing all internal
water requirements  with  recycled water.  The scrubbing system could
thus be  turned down to any desired capacity while simultaneously turning
down the  makeup  water  flow rate to avoid operating the system out
of water  balance.   The most critical part of the water recycle concept
required  for  closed loop operation was  the source of pump and instrument
seal water.   The  seal  water requirement  for each module  represented
a  flow  rate of about 50  gpm,  which was  independent of flue gas flow
rate, S02  content  or  unit  load.  For both Horizontal and  Vertical scrubbing
systems,  thickener  overflow was filtered to provide the  pump and instrument
seal water. In spite  of  the high dissolved solids in the seal water
no severe  chemical  scaling  or  corrosion  was observed.

Operating  Conditions  at  Design  Gas Flow  Rate

     The observed operating conditions  at the design gas flow rate
of 450,000 SCFM  are  summarized  for  the  three types of scrubbers in
Figure  11.  Several  items  deserve  comment.

Electric Power Requirements

     The flue gas pressure  drop across  the  scrubbing chamber  was  an
order of magnitude  less  for  the Horizontal  Module  than for  the  Vertical
TCA  or PPA.  The electric power  requirements for  the Vertical  Module
                                  341

-------
                                FIGURE  10
                           WATER BALANCES
                    INLET GAS TEMPERATURE = 270° F.
                               450,000 SCFM
              HORIZONTAL SCRUBBER
                                       148 6PM (eq.)
                                       HgO VAPOR OUT STACK
                  PLUS THICKENER
    MAKE-UP HgO-

 39  GPM SERVICE HjO
 117  GPM COOLING
    TOWER SLOWDOWN
    12,000 PPM TDS

 156  GPM (25% FRESH H{0)
               1
  LIME SLURRY
185,000 PPM TDS
 RETURN FROM POND
«•	
               TO POND
                                       26 GPM
                                 POND
                                                         4 GPM EVAPORATION
                                                         3 GPM FREE WATER IN SLUDGE
                                                         I GPM WATER OF HYORATION
                                                         8 GPM =5.1% OF MAKE-UP
            VERTICAL PPA SCRUBBER
                 PLUS THICKENER
                                       148 GPM (eq.)
                                       H20 VAPOR OUT STACK
    MAKE-UP H20-
 96 GPM SERVICE HjO
 64 GPM COOLING
   TOWER SLOWDOWN
   12,000 PPM TDS
•••«••••••*

 160 GPM (60% FRESH HzO)
               I
LIMESTONE SLURRY
 70,000 PPM TDS
 RETURN FROM FILTER
    21.5 GPM H20
               TO FILTER
                                       33.5 GPM H20
                               342
                           6.5 GPM EVAPORATION
                           4.5 GPM FREE WATER IN FILTER CAKE
                             I GPM WATER OF HYDRATION
                                                     12 GPM =7.5% OF MAKE-UP

-------
                   FIGURE 11
OPERATING CONDITIONS AT DESIGN GAS FLOWRATE
                 450,000 SCFM
HORIZONTAL

L/6 RATIO IGPM/IOOOSCF)
SCRUBBER PRESSURE DROP (IN H20)
SYSTEM ELECTRIC POWER REQUIRED (MW)
REAGENT TYPE
SPENT SLURRY PH
REAGENT UTILIZATION
PERCENT COOLING TOWER SLOWDOWN H20
PERCENT SLURRY SOLIDS
FORMATION OF GYPSUM SCALE
(4 STAGES)
21
1.0
2.6
LIME
6.2
99%
75%
5%
NO
VERTICAL TCA
(4 STAGES)
36
14.0
3.4
LIMESTONE
5.4
75%
65%
5%
YES
VERTICAL PPA
(3 STAGES)
60
12.2
3.9
LIMESTONE
5.1
92%
41%
15%
NO
                              (ESPECIALLY ON FIRST
                                 STAGE GRIDS)

-------
were consequently 30% to 50% higher than for the Horizontal Module.
For example, at design operating Conditions, the measured electric
power consumption of the Vertical Module was 3.4 MW in the TCA mode
and 3.9 MW in the PPA mode, while the power consumption of the Horizontal
Module was 2.6 MW. This included all of the power requirement including
air conditioning of the control rooms.

Reagent Comparison

     Figure 11 indicates a lime utilization of 99% in the Horizontal
Scrubber.  The methods used to measure this were reported in a previous
paper.2 Figure 11 also indicates a limestone utilization of 65%
with 5% slurry solids in the TCA scrubber and 92% utilization with
15% slurry solids in the PPA scrubber.  The role of percent slurry
solids in influencing limestone utilization was also presented previously."
Increased slurry  solids increase limestone utilization but also increase
wear on pumps and nozzles  as well as  increase the possibility of solids
buildup in  the scrubbing system components. Both the lime and limestone
utilization data  obtained  in the 170 MW scrubbers compare well with
the data obtained on the pilot plant  scale.  Utilizing 95% pure reagents,
the data in Figure  11 can  be used to  calculate that 2.75 times as much
limestone by weight  as lime is required to remove a given amount of SO2.

     The choice  between lime and limestone is a trade off between a number
of factors  in which  the quantity of reagent, including the associated
transportation cost, is only one factor.  The operating and capital costs
of the equipment  required  to pulverize limestone to a 225 mesh size must
be balanced against  the costs of a lime slaker to slake 1/4 inch pebble
lime.  The  decreased solubility of limestone, compared to lime, results
in increased holding tank  sizes unless special attention is paid to the
chemistry of the  scrubbing solution.  Our experiments, both on the  1/2-1
MW size  and the  170  MW size indicated that a higher L/G was required with
limestone than with  lime to achieve the same degree of S02 removal, all
other  factors  being  equal.  Finally,  the unreacted limestone increases
the amount  of  sludge which must be disposed of and it's presence makes
more  difficult the  conversion of the  sludge into potentially useful
products  such  as gypsum wallboard.  All in all, the cost per ton of a
limestone versus lime  is not a  true measure of the actual costs which
will  be  incurred.

Reheaters

      It  is  well  known  that wet  scrubbers cool and saturate flue gas.
Without  flue  gas reheat, the saturated scrubber exit gas would form
a dense,  white water vapor plume.   On cold  days this plume can persist
for thousands  of feet.  Even with reheat, to 200 degrees Farenheit,
prevention  of  a  dense  water vapor  plume may be  impossible on some  winter
mornings.

      A comparison of the two methods  of flue gas  reheating tested  at
Mohave is  shown  in  Figure  12.   On  the left  is the indirect reheat  method,
where outside  air is heated to  about  375 degrees  Farenheit with steam
coils and  introduced into  the  cold flue gas at  the outlet of the  scrubber.
On the right  is  the direct reheat  method,  where  steam  coils  in direct


                                   344

-------
                                         Figure  12
                      REHEATER  COMPARISON  FOR EQUIVALENT
                             REDUCTION IN  FOG  FORMATION
              INDIRECT  REHEAT
                          DIRECT REHEAT
550.000 ACFM
0.10 lbH20/lb Eos
  (Saturated)
  120,000 ACFM 'a) 75° F
  0.01 lbH20/lb Air
710,000 ACFM Q) 170°F
 L08 lbH20/lb Gas
                             *~40,000 Ib/hr Steam
                               Carbon Steel Coils (Finned )
                              Ambient Air Fan
                                                              610,000 ACFM o&190°F
                                                               0.10 IbHjO/lb Gas
                       550,000 ACFM 0)125° F
                        0.10 lbH20/lb Gas
                          (Saturated)
                                          40,000 Ib/hr Steam

                                          E-Bright
                                          High Alloy Coils
                                          (Smooth)

-------
contact with outlet flue gas provide the desired amount of reheat.
(The so-called direct reheat method where oil or gas is burned directly
in the exit flue gas was not tested in this program.)  Both methods
of reheat require the same amount of steam for the same degree of
reduction in water vapor plume formation. Indirect reheat using relatively
dry ambient air requires an exit temperature of only 170 degrees Farenheit
to provide the same reduction in fog formation as 190  degrees Farenheit
exit temperature with direct reheat. This is because the exit gas
in the indirect reheat case is 20% diluted with relatively dry ambient
air. However, plume buoyancy or possible stack condensation problems
might dictate using a higher exit temperature gas.

      During the test program, it was found that the ambient air fan
used for indirect reheat provided a built-in means of  positive pressure
air purge when the scrubber was shut down. Guillotine  dampers at both
ends of the scrubber were under positive pressure with ambient air
due to the indirect reheat fan, which allowed safe internal access
to the ductwork without expensive pressure sealed double guillotine
dampers. The indirect reheat steam coils were not subject to the same
environment as the direct reheat coils.  These findings led to the
conclusion that indirect reheat was preferred compared with direct
reheat.

Mist Eliminators

     Two types of mist eliminators were tested at Mohave, as shown in
Figure 13.  The horizontal type (gas flow in horizontal plane)
utilized one mist eliminator section with front and back sides deluge
wasned at intermittent intervals. The normal wash cycle was about
15 minutes every eight hours. The vertical type utilized two mist
eliminator sections with no wash on the second section and continuous
downwash on the top side of the first section.  Difficulty was experienced
in keeping this mist eliminator wash pump operating continuously during
the test program. Both mist eliminators used 3-pass blades as shown,
with plastic blades in the horizontal type and stainless steel blades
in the vertical type. The plastic blades were not capable of withstanding
temperature over about 180 degrees Farenheit without warping. The
vertical type of mist eliminator suffered the disadvantage of no capability
for removing deposits from the second mist eliminator  section without
a scrubber shutdown for maintenance. Both types of mist eliminators
operated satisfactorily during the test program, but there were deposits
on the second section of the vertical-type mist eliminator blades
that continued to build up. It was concluded that a horizontal-type
mist eliminator with intermittent wash on the front and back sides
was preferred to the two section vertical-type mist eliminator with
the described wash configuration.

Unavailability - History During the Test Program

     Unavailability is defined as the total time that  the scrubber
was shut down and could not be operated due to design  deficiency or
maintenance problems. Unavailability does not depend on whether or
not the generating unit is in operation, while availability is frequently
defined as a function of generating unit operation. In an installation
which involves a number of modules connected to one generating unit,

                                    346

-------
                                               Figure 13
                                        MIST ELIMINATORS
                    HORIZONTAL MODULE TYPE
VERTICAL  MODULE  TYPE
                FRONT AND BACK SIDE
                WASHED AT 8 HR. INTERVALS
     GAS
    FROM
SCRUBBER
                 COLLECTED DROPLETS
                 FLOW BY GRAVITY TO SUMP
                                                                          CONTINUOUS DOWN WASH ON.
                                                                          ON TOP SIDE OF FIRST  SECTION
                                                                          NO WASH ON SECOND  SECTION
                                                COLLECTED
                                                 DROPLETS
                                                  FALL BY
                                             GRAVITY BACK
                                              TO SCRUBBER
                                                                          FROM  SCRUBBER
                                                           NOTE: GAS FLOW FROM SCRUBBER TENDS TO PUSH
                                                                SMALLER  DROPLETS BACK INTO  MIST
                                                                ELIMINATOR BLADES

-------
the unavailability record is the best indicator of how much of the
time a given module will not be available for service.

     The unavailability history for the test modules program is given
in Figure 14.  As can be seen, for 100% availability from a series of
modules, more modules would probably be required with the Vertical
configuration than with the Horizontal configuration.

     During the Test Program, design modifications were made to alleviate
required maintenance and decrease downtime for maintenance. In the
Horizontal Module, the mist eliminator blade wash system was improved and
a single mist eliminator section was found to be satisfactory. Successful
modifications were also made to the scrubber to alleviate inlet gas
distribution problems.  Methods were also successfully tested to repair
worn pump impellers and change worn spray nozzles without shutting
down the Horizontal Module.  Spray nozzles made with refractory materials
were found to have longer life (estimated over one year) than nozzles
manufactured with other materials.

     In the Vertical Module, many successful design modifications were
also made.  However, no methods were discovered for preventing migra-
tion of the thermoplastic rubber balls between compartments without
shutting down the scrubber and redistributing the balls.  Replacement
or cleaning of slurry distribution nozzles in' the Vertical Module
because of plugging or wear also required shutting down the scrubber.
Finally, scale formation at the bottom of the first stage grids was
aggravated by gas flow distribution problems in the Vertical Module.
The poor gas flow distribution at the first stage was never resolved,
but is  suspected to have aggravated ball migration and gas flow channeling
problems in the Vertical Module.

DISCUSSION

    There were several findings in this program which are contrary
to commonly accepted beliefs.  At the beginning of the pilot plant
program it was believed by many people that it would not be possible
to reach exit concentrations of SO2 lower than 20 ppm (90% S02 removal
from an inlet concentration of 200 ppm inlet gas) because of the re-
duced partial pressure of S02 as the concentration approaches zero.
While the pilot plant results2 refuted this, the Test Modules Program
confirmed that it is possible to achieve an exit S02 concentration
of 2 ppm (99% 302 removal) on a large scale.

     Many people believe today that chemistry plays an overwhelming
role  in scrubber technology. Vihile chemistry is important, for example
the settling rate in the thickener tank and the quantity of sludge
produced (sulfite crystals have 1/2 molecule of water associated with
them compared to 2 molecules of water with the sulfate), one of the
test program findings was that fluid mechanics also plays an important
role, particularly in low maintenance operation. The major influences
on scrubber design and operation are segregated as shown in Figure  15.

     Finally, during the Test Modules Program, twenty-four variables,
which influence visual plume opacity were identified.^  Some of these
have previously been identified (particle size, and grain loading)


                                     348

-------
                                         FIGURE 14
          UNAVAILABILITY HISTORY-MOHAVE 170 MW TEST MODULES  PROGRAM
       JANUARY 16,1974 TO FEBRUARY 9,1975
      HORIZONTAL                        HOURS
I. MODIFY AND REPAIR PLASTIC DEMISTER BLADES     503
2. CORRECT BOOSTER FAN BALANCE PROBLEMS       317
3. REPAIR CRITICAL PUMPS                     256
4. REPLACE WORN-OUT SPRAY NOZZLES            238
5. MODIFY INLET GAS FLOW DISTRIBUTION          162
6. REPAIR HOPPER LEAKS                       135
7. REMOVE HARDHAT FROM THICKENER              82
8. MODIFY SLAKING WATER TO PREVENT SCALE      45
9. CONDUCT INSPECTIONS FOR LONG RUNS          	19

                                  TOTAL    1757
     NOVEMBER 2,1974 TO JULY 2,1975
      VERTICAL                      HOURS
I . REPAIR GRIDS AND REDISTRIBUTE TCA BALLS    710
2. CLEAN SCALE FROM SCRUBBER INTERNALS      344
3. REPAIR OR REPLACE PLUGGED NOZZLES        153
4. REPAIR LEAKS IN TRAP-OUT TRAY            120
5. REPAIR /REALIGN PPA PACKING                85
6. CORRECT BOOSTER FAN  TRIP PROBLEMS        72
7. CONDUCT INSPECTIONS  FOR LONG RUNS         55
8. REMOVE  HARDHAT FROM THICKENER          46
                               TOTAL    1585
                     TOTAL CALENDAR TIME   9328
                 TOTAL CALENDAR TIME    5813
        PERCENT UNAVAILABILITY = 18.7
       PERCENT UNAVAILABILITY = 27.2

-------
                                FIGURE 15
                    MAJOR INFLUENCING TECHNOLOGY
                       IN S02 SCRUBBING SYSTEMS
CM
Ln
o
FLUID DYNAMICS
FAN
PRESATURATOR
SCRUBBER DESIGN
SCRUBBER PUMPS
SCRUBBER PACKINGS
SPRAY NOZZLES
DEMISTER
REHEATER
DUCTING
THICKENER
VACUUM FILTER
CENTRIFUGE
PIPING AND VALVES
EROSION
INSTRUMENTATION /CONTROLS
CHEMISTRY
REAGENT FEED SYSTEM
REAGENT TYPE
PH CONTROL SYSTEM
SCRUBBER LININGS
CORROSION
SCALE PREVENTION

-------
but the experimentaloiata  indicated  that  a major variable apparently
heretofore unrecognized was the  altitude  of the sun. Figure 16 indicates
the visual opacity reading which would  be obtained by a "perfect"
observer if the Mohave Generating Station were located in Florida
or in Washington.  With the absolute emissions of particulate matter
remaining constant, the predicted variation in opacity from 10% to
90% as shown in Figure 16  is  due solely to effects of the altitude
of the sun. To date, we have  not been able to convince the EPA that
the sun rises in the East  and sets in the West or that the days are
longer in the summer than  the winter. However, we have only been trying
£ot a year and hope for success  in eliminating visual opacity regulations
in the future.
                                     351

-------
Jx
H
M
H
55
w
w
100

 90

 80.

 70

 60

 50

 40

 30

 20

 10
       5:00
       AM
                         Key West,
                          lorida
                            (June  21)
       Seattle,
       Washington
           (Dec.21)
9:00
AM
                     1:00
                     PM
5:00
PM
             LOCAL STANDARD TIME
  FIGURE 16 - EFFECT OF GEOGRAPHIC  LOCATION
              AND TIME OF  DAY ON OPACITY
                       352

-------
                              BIBLIOGRAPHY
1.    Shapiro,J.L.  and  Kuo,W.L.  "The Mohave/Navajo Pilot Facility
     for  Sulfur  Dioxide  REemoval"  Second (2nd)  EPA Flue Gas
     Desulfurization Symposium,  November 8,  1971, Newe Orleans,
     Louisiana.

2.    Weir,A.,  and  Papay,L.T.  "Scrubbing Experiments at the Mohave
     Generating  Station"  3rd  EPA Flue Gas Desulfurization symposium,
     May  14,  1973  - New  Orleans, Louisiana.

3.    Weir,A.,  Johnson,J.M.,  Jones,D.G., and  Carlisle,S.T., "The
     Horizontal  Crossflow Scrubber", 4th EPA Flue Gas Desulfurization
     Symposium,  November  4,  1974,  Atlanta,  Georgia.

4.    Handler,Philip et al "Air  Quality and  Stationary Source Control-
     a  report  by the commission on Natural  Resources, National Academy
     of Sciences,  National  Academy of Engineering, and National
     Research  Council  -  Prepared for the committee on Public Works,
     United  States Senate - March, 1975  -  Serial No. 94-4 U.S.
     Government  Printing Office - Catalog Number Y4,P96/10:94-4

5.    Weir,A.,  Jones,D.G., Papay,L.T., Calvert,S. and Yung,S. "Measurement
     of Particle Size  and Other Factors influencing Plume Opacity"
     United  Nations, U.S. EPA (and other) International Conference
     on Environmental  Sensing and Assessment, September 14-19,1975,
     Las  Vegas,  Nevada.
                               353

-------
                    LA CYGNE STATION UNIT NO. 1
                 WET SCRUBBER OPERATING EXPERIENCE
     Clifford F.  McDaniel, Superintendent, Air Quality Control

                Kansas City Power and Light Company
                           P. 0. Box 211
                     La Cygne, Kansas   66040
ABSTRACT

     This paper presents a description of the Babcock and Wilcox
designed scrubbing system and reviews the status, costs, reliability
and supportive data relating to our experiences in 1975.
                              355

-------


-------
                          LA CYGNE STATION UNIT NO. 1
                      WET SCRUBBER OPERATING EXPERIENCE
DESCRIPTION
     The 820-megawatt La Cygne No. 1 unit began commercial operation on
June 1, 1973, as a joint project of Kansas Gas and Electric Company and
Kansas City Power & Light Company.  The companies share  equally in owner-
ship and output and the unit is operated by KCPL.  The 630-megawatt No. 2
unit, now about 45 per cent completed, is expected to be in service by the
spring of 1977 under an identical arrangement.
     The plant site is located about 55 miles south  of downtown Kansas City,
one-half mile west of the Missouri state line, and was selected based on
locally available coal, water and limestone.  Construction of  No. 1 unit
began in 1969 and erection of the Air  Quality Control  System was initiated
in mid-1971.
     Water for cooling purposes is furnished from a  2,600-acre reservoir
constructed adjacent to the plant site.  Fly ash  and spent slurry from the
AQC system is piped to a 160-acre settling  pond located  east of  the  reservoir.
     Coal is delivered to the plant  in off-the-road  120-ton  trucks  from
surface mines operated by The Pittsburg & Midway  Coal Mining Co.  The nearby
coal deposits are estimated to  contain 70 million tons.   The  fuel is low
grade, sub-bituminous with an as-fired heating value of  9,000  to 9,700
Btu/lb, and  an ash  content of 25  per cent  and sulfur content  of  5 per  cent.
 (Exhibit A).
     Limestone is obtained from nearby quarries  and  delivered  to the plant
 in off-the-road  50-ton trucks.
     The boiler  for No. 1 unit  is a  cyclone-fired,  supercritical,  once-
 through, balanced-draft Babcock & Wilcox  unit, with a rating of  6,200,000
 pounds  of steam  per hour, 1,010 degrees F,  3,825  psig at the superheat
 outlet  and  1,010 degrees F at  the reheater outlet.   The turbine-generator
 was supplied by  Westinghouse  and  is  rated at 874 MW gross output with  five
                                    357

-------
per cent overpressure and 3,500 psi throttle pressure.   Three auxiliary,
oil-fired boilers are used for plant start-up or for powering a 20 megawatt
house turbine-generator.  The net plant output is 820 megawatts,  adjusted
to include 24 megawatts used by the AQC system and 30 megawatts by plant
auxiliaries.

PROCESS DESCRIPTION
    The AQC system consists of seven identical two-stage Venturi-absorber
scrubber modules (Exhibit B) designed to treat the boiler flue gas flow of
2,760,000 ACFM.  (394,300 ACFM per module at 285 degrees F).   The ductwork
design does not provide for flue gas bypass of the system.  Also, the plant
does not have an alternate or secondary fuel supply. Each module can be
isolated for maintenance by individual dampers.   On site limestone grinding
and slurry storage facilities provide up to 1,000 tons  of slurry  per hour.
The unit has a balanced draft system with three 7,000 hp forced draft fans
and six 7,000 hp induced draft fans located between the AQC system and the
700 foot stack.  There is a common plenum at both the scrubber inlet and
outlet.  Spent slurry and fly ash are removed from the  module recirculation
tank through rubber lined pipes to the settling pond at the rate  of 3,500
tons of solids per day.  Clear make-up water is  pumped  from the pond and the
loop is closed by recycling ball mill and module make-up water back into the
system.
     In abbreviated terms, as the hot flue gas enters the Venturi (Exhibit C) ,
it is sprayed with slurry from 48 spray and 32 wall wash nozzles  resulting
in up to 99 per cent of the particulates agglomerated to the  sump below.
The gas continues through the sump making a 180  degree  turn up through the
absorber section.  In the reaction chamber, the SO  is  removed as the gas is
forced through a limestone slurry solution sprayed on stainless steel sieve
trays.  The chemical reaction in part combines the calcium carbonate, water
and sulfur dioxide to form two relatively insoluble calcium salts:   calcium
sulfate and calcium sulfite, which also fall to  the sump.  The cleaned gas
passes through demisters to remove moisture and  then is  reheated  to avoid
deposits on the fans and provide a plume effect  from the stack.
                                    358

-------
OPERATING EXPERIENCE
     As  a result of the continuing modification work,  the availability of
modules  has been improved substantially, and for 1975,  the total availa-
bility  (Exhibit E) averaged 84.33 per cent, including both working and
reserve  hours.
     The results of a stack sampling test conducted on May 5, 1975
(Exhibit H) with the unit continuously operating from  700 to 720 megawatts
indicated an average sulfur dioxide removal efficiency of 80.14 per. cent,
and  removal of 98.2 per cent of the particulates.
     The ambient monitoring system indicates ground level concentrations
well below the national standards for sulfur dioxide and nitrogen
compounds (Exhibit I).
     During a four hour, full load test in March (Exhibit F), the unit
maintained a minimum of 800 megawatts with all seven modules fully loaded.
The  six  induced draft fans were operating at less than maximum load and  the
three  forced draft fans had not reached maximum rating. Sulfur dioxide
removal  efficiency for the six modules from which results were obtained
was  76.2 per cent.

MAINTENANCE
      In  addition to the modifications, the improved performance of the AQC
system has depended upon a substantial maintenance effort.  Present pro-
cedures  call for cleaning one module each night on a rotational schedule
and  keeping all modules available during the daytime peak periods.  Cleaning
requires three men from 10 to 12 hours, including time taken to open  the
many sections and place holds for personal safety.  Recent modifications
have greatly improved cleaning requirements and we are looking  forward  to
modules  staying on line continuously for up to three weeks.
     Areas requiring attention are reheater pluggage;  demister pluggage;
Venturi  wall and nozzle deposits, and sump accumulation.  Hard  scale  has
in the past been an enormous problem, especially in the sieve  trays,  but by
closely  controlling pH, severe scaling is not usually one  of our  chief  troubles.

                                      359

-------
     Carry-over to the induced draft fan blades continues  to be one of our
main operating concerns.  We wash each individual fan about every four or
five days.  This requires a fan outage of from four to ten hours depending
upon the necessity of a second or even a third washing.  This is done with
very high pressure water from 4,000 to 7,000 psi.  The fan is taken off
when the vibration from inbalance exceeds 12 Mils on either bearing.
     To resist the effects of acid corrosion, various epoxy paints have
been used, the post recent a Plasite #4030 black paint with a durakane
resin base.  This very tough surface has good erosion and  corrosion
resistance but eventually cracks and should be reapplied every 8 to 10
months to maintain wheel integrity.  Currently we are considering pro-
tecting those areas most affected by acid runs by cladding with Inconel 625.

MANPOWER REQUIREMENTS
     The scrubber operating and maintenance force currently totals 51
people and the organization is separate from the rest of the plant (Exhibit
J).  This is still considered to be a research and development situation
and once routine operation is determined, the number of  scrubber personnel
will be reassessed.

COSTS
     The total cost (Exhibit G) of the AQC system to date  is $43 million,
or about 22 per cent of the $200 million total plant cost,  or about $52 a
kilowatt installed.  It is estimated that an additional  $7  million investment
will be required to reach optimum system performance.  These figures do not
include the allowance for funds used during construction.   To date, the
City of La Cygne, Kansas, has issued $30 million in tax  exempt pollution
control revenue bonds to finance the system under a lease,  sublease arrange-
ment.  The sale of an additional $39 million of the bonds  was sold
December 1, 1975, with proceeds to be applied to the remainder of the No. 1
unit system and the AQC system for No. 2 unit.
                                     360

-------
     La Cygne production costs for 1975 for energy averaged 8.82 mills
per kilowatt  hour.   The limestone and operating and maintenance expense of  the
AQC  system   totaled 1.26 mills, or about 14 per cent of the production
costs.   Labor accounted for .45 mills, limestone .55 mills, and other
costs were .27 mills.
     The total annual cost of producing power including fixed costs from
La Cygne No.  1 unit  in 1978 is estimated at 15.24 mills per kilowatt hour.
Interestingly, the  comparative cost for the No. 2 unit, which will burn
Wyoming coal  supplied by Amax Coal Company from mines in the Gillette
area, will be 21.96  mills per kilowatt hour.  As these projections indi-
cate, total cost of  producing power from No. 1 unit, even with the
complicated air quality control system, will still be 44 per cent less
expensive than for No. 2 unit.  Of the total cost, the air quality component
for La  Cygne  No. 1 unit is 3.03 mills per kilowatt hour, and for No. 2  unit
is 1.77 mills per kilowatt hour.  Unit No. 2 will be equipped with an
electrostatic precipitator.

ADDITIONAL MODIFICATIONS
     Planned  additional modifications include the installation of an
eighth  module to be  in service the summer of 1977.  This is estimated to
cost $5.2 million and will improve the cruising capability of the unit
from 700 megawatts  to 800 megawatts.  The original AQC system design
allowed space for this modification.  Changes in the induced draft fans are
also being made to  reduce or eliminate a fan paralleling problem.  Many
other experiments and minor changes are continuing to improve performance
and to  reduce operating and maintenance expenses.
                                  361

-------
                             LA  CYGNE  STATION
                         COAL  AND  ASH   ANALYSIS
                                      COAL
Proximate

Volatile
Fixed Carbon
Ash
Moisture
              100.00
 BTU/lb.
9421
 Grindability     59.59
                                                          Ultimate
Moisture
Carbon
Hydrogen
Nitrogen
Chlorine
Sulfur
Ash
Oxygen
8.60
51.93
3.43
0.94
0.027
5.39
24.36
5.33
                                                                      100.007
                                       ASH
 Analysis

 Phosphorous Pentoxide     0.15
 Silica                  46.05
 Ferric Oxide            19.23
 Alumina                14.07
 Lime                   6.86
 Magnesia                1.02
 Sulfur Trioxide           7.85
 Potassium Oxide          2.48
 Sodium Oxide            0.60
 Titania                  1.02
 Other                   0.67

                       100.00
                                              Fusion Temperature

                                              Reducing I. D.      1957
                                                Soft (H=W)      2045
                                                Soft (H=W/2)    2169
                                                Fluid           2321
                                              Oxidizing I. D.      2156
                                                Soft (H=W)      2338
                                                Soft(H=W/2)    2415
                                                Fluid           2520
                                     Exhibit A

                                           362

-------
X
zc

CO
                                   La Cygne limestone wet scrubbing system

-------
                           FIGURE  I  -  LACYGNE1  FGD  MODULE
    REHEAT
     STEAM
     550 °F


(-
1
REHEAT
1000 PPM -
-*- SO2 -j
190° F H
_C_OILS H
f n n-n o o o
x



-43"
HsO


	 J*-i ' '
   OT  Ai R
— FROM
 BOILER
INTERMITTENT
OVERSPRAY
  2150 GPM

 CONTINUOUS
 UNE5ERSPRAY
   140 GPM
         VENTURI  SPRAY
               /\  /\  /\ /\
           WALL
           WASH
           SPRAY
                                        v  V   \/  V  V
    VENTURI
    THROAT
     FLUSH
             PREDEMISTER
                                 ABSORBER
                                  SPRAY
                                             200-6OO GPM
RECIRCULATION
    CA.CO3   70 G/L
    CASO3   35 G/L
    CASO4   25 G/L
    FLYASH  4O G/L
    P H - 5 5 - 6 0
    8- 10%  SOLIDS
                                                                            T O
                                                                            FAN
  SPENT  SLURRY
   TO POND
    700 GPM

«35OO TONS/ DAY
93000 TONS/ YEAR
53 ACRE FEET/YEAR
                                            LIME
                                            STONE

                                            SLURRY
                                            FEED
                                             20%
                                           SOLIDS
               VENTURI
            RECIRC  PUMP
              5000 GPM
   TOTAL  FOR ALL MODULES
                                               EXHIBIT C
                                    ABSORBER
                                  RECIRC. PUMP
                                   90OO  GPM

                                  364

-------
               LA CYGNE SCRUBBER WATER ANALYSIS
CATIONS


CALCIUM HARDNESS SOL.  (Ca)

MAGNESIUM HARDNESS SOL.  (Mg)

SODIUM (Na)

POTASSIUM (K)
COOLING
_LAKE

 123.2

  11.0

  12.0

   4.5
SETTLING
  POND


 696.0

  48.0

  22.0

  23.0
ANIONS
BICARBONATE ALK (AS HCO )

CHLORIDE (Cl)

SULFATE (SO )

SULFITE (SO )
SILICA (Si02)
 109.8

  24.6

 247.9

 * ND

   2.4
  36.6

 177.8

1627.3

* ND

  20.6
OTHERS


pH (pH UNITS)

CONDUCTIVITY IN MICROMHOS
   7.87

 649.0
   7.0

4380.0
*ND - Not Detected
                             Exhibit D
                                                       365

-------
                 MODULE AVAILABILITY  SUMMARY

                        LA CYGNE 1975
                                                      TOTAL
        GENERATION
BOILER  CAPACITY
MONTH
January
February
March
April
May
June
July
August
September
October

Novemb er

December
A B
C
D E
F
G AVAILABILITY*
MWH HOURS FACTORS
Turbine Generator Repair
Turbine Generator Repair
82.4 96.03
Generator
94.6 85.1
87.8 85.4
78.4 89.7
74.64 88.07
78.43 83.62
66.16 77.26
Generator
92.87 90.79
Generator
90.72 87.39
89.5
Repair
94.2
83.9
89.6
87.29
84.38
46.27
Repair
80.18
Repair
80.87
76
.6 92.96
91.5
96
89.33
25 Days
89
84
83
78
84
73
.5 89.8
.9 84.1
.7 85.4
.01 92.44
.67 78.72
.62 71.91
89.3
86.1
87.4
85.00
77.71
73.07
83.4
88.6
85.2
83.06
74.24
64.69
89.4
85.8
85.6
84.07
80.25
67.57
7
244
23
332
324
297
294
239
74
,886
,873
,014
,526
,952
,870
,402
,954
,660

694

683
667
590
630
610
231

38.8

52.7
53.2
47.2
46.7
39.3
24.3
15 Days
93
.18 96.09
89.39
93.94
90.83
165
,058
346
50.6
17 Days
85
.20 86.89
88.56
83.67
86.19
278
,597
597
46.8
*Working Hours -4- Reserve
Hours In Month
                            Exhibit E

-------
                                Four Hour Load Test Results

                             (Test  for Maximum Credited Megawatts)
Date:           March  24,  1975
Time:           6:10 pm to 10:25 pm
Power:          Min. - 800 MW (e) Gross - Peak* - 830 MM (e)
Outside Temp:  26 F
                                        Data
                                                                      F
Gas Flows (cfm) 380
Throat Pos.
Reheat Temp. ( F)
Absorber Pump
Venturi Pump Flow 4
Venturi AP
Reheater AP (X2)
Abs-Dera. AP
Hot Air Damper Pos.
(% Open)
Reheat Outlet Damp.
Pos. (% Open)
Scrubber Outlet Press.
(-"H2o)
I . D . F an Amps
I.D. Fan Inlet Damper
Pos . (% Open)
F.D. Fans Amps
Lab pH**
Conductivity**
Sulfite (g/1)
Carbonate (g/1)
S02 Efficiency
Inlet (ppm)***
Outlet (ppm)
,000
22"
170
ON
,000
8.0
4.0
-
46
51
360,000
22"
180
ON
4,000
7.5
3.5
5.5
23
37
3 8. 3 (common)
540 520
100
540
5.95
2100
41.1
94.4
75.1
5700
1419
98
500
5.89
2200
36.2
84.1
79.1
5138
1075
400,000
22"
180
ON
5,000
-
3.0
-
44
98
620
98
490
5.80
2000
45.9
69.4
72.2
5516
1533
410,000+
22"
150
ON
4,000
11.0
3.5
8.5
-
84
34.0
600
100

5.68
2250
35.4
84.4
72.0
4995
1220
400,000
22"
150
ON
5,000
10.0
1.0
8.5
48
100
380
100

5.92
2000
45.1
91.9
7
5700
1
390,000
22"
170
ON
5,000
11.0
4.5
9.0
32
96
380
23

5.79
2100
29.4
109.0
82.9
5017
857
340 ,000
22"
185
ON
5,000
9.0
12.5
9.0
64
100



5.73
2200
54.7
76.6
75.7
5120
1243
                                       Exhibit F
                                                367

-------
                               COSTS
                         LA CYGNE STATION
                  Scrubber Operating Expense 1975
OPERATING LABOR
OPERATING MATERIALS
MAINTENANCE LABOR
MAINTENANCE MATERIALS
LIMESTONE

              TOTAL
    $    601,029
         195,926
         416,206
         386,397
       1,256,048

       2,855,606
0.265 Mils/KWH
 .086
0.184
0.171
0.554

1.260 Mils/KWH
TOTAL UNIT
SCRUBBER PORTION
CAPITAL COSTS

    $200,000,000
      43,000,000
$244/KW
$ 52/KW
COAL:       $7.10 TON - 37.7
-------
                                                     Module Evaluation
                                                     Test f     *
DATE:
TIME:
LOAD RANGE:
AMBIENT TEMP:
          IA CYGNE STATION
        STACK SAMPLING TEST

  5/15/75
11:30 P.M.  to 8:00 P.M.
  700 to 720  MW Continuous
  72° F
               (Data taken by independent
                      test group)
AVERAGE SO2 REMOVAL:   8^-                          ,
PARTICULATS REMOVALS:  98.2%        Inlet  9.9#  per 10  BTU
BAROMETTRIC P.:   29-2*               Outlet .18* per 10  BTU
BEL. HUM.       	  DATA LOGGED AT
gas Flow Indicated
Throat Position
Reheat Tenraeroture
Absorber Pumo
Venturi Slurry Flow
Vent or i ^ P
Reheater & p
Absorber Eerr.. A ?
Hot Air Daarper Pos.
^ Ct>en
Reheat Outlet Daaper
Pos. 4, Open
Scrubber Outlet Fress.
C-"HpO)
I.D. Fan Ar.ps (Control
RuC'IIi )
I.D. Fan Ar.cs (Breaxer?
I.D. Fan Inlet Dampers
Pos. <& Orcen
F.D. Fan Aaps (Control
Room)
F.D. Fan Arans (Breaker)
Lab T5H/C.R.
Sulfite (,-z/l)
Carbonate (p/1)
S02 Efficiency %
Inlet (DUB)
Outlet (-or*-)
A
320K
22
170
On
4,000
7
4
-
46
40
38
460

60


5.95
53,9
155.0
76
4,506
1,068
B
280 K
22
190
On
4,000
6
3
4
20
35

480

60


6.02
37.0
85.3
81
4,297
834
C
300 K
22
170
On
4.000
7
3
-
50
55

460

60


5.03
37.4
108
81
4,663
892
D
340 K
77
148
On
4,000
7
8
8
_
75
28
400

60


5.80
58.4
75.3
82
4.273
776
E
320 K
77
180
On
4,000
8
6
S
50
82

410

60


5.98
61.6
109
77
4.982
1.121
F
300 K
97
160
On
4,000
7
11
f>.
94
100

440

32


5.90
54.7
93.1
83
4 1 S6
704
G
300 K
. 22 	 ..
iqo
On
4.000
8
8.
8
100
100






5.90 i
64.8
95.9
81
L Sli
Q1 7
BOILER DATA:
  F. D. Fan Discharge ("
Air Flow %

Fuel Flow %

Feedwater Flow #/Hr.

Excess Op %

Windbox-Furn.  Diff. Press.

Furn. Pressure

Sec. Super Gas Pressure

Pend. Reheat Gas Pressure
Primary  Super  Gas Pressure
               44
                               68
                               72
                               5,200,000
                               2.5
                               31
                               -2.0
                               -3.4
                               -3.2
                               -6.0
         Horz. Reheat Gas Pressure

         Econ. Outlet Gas Pressure

         Feedwater Pressure

         Throttle Pressure

         Throttle Temperature

         Hot Reheat Temperature
         Air to Air Heater  A
                            B
         Air from Air Heater  A
                              B
         Gas to Air Heater    A
                              B
         Gas from Air Heater  A
                              B
Exhibit H         369
                                                                            -7.3
                                                                           -10.0
                                                             4,100

                                                             3,500

                                                             970

                                                             1,000°F

                                                             155.5
                                                             lib.9

                                                             561.0
                                                             598.0
                                                             655.5
                                                             668.7
                                                             308.7
                                                             30S.9

-------
                              LA CYGNE STATION
                         AMBIENT MONITORING SYSTEM
                                   STATION 1
                 STATION  2
             STATION 3
MILES FROM PLANT
PRIOR TO START UP
   CONCENTRATION S02 - ppm
                                                    10
                                 12.5
RECENT LEVELS
   CONCENTRATION S02 - ppm
                 N02 - ppm
NATIONAL STANDARDS
   CONCENTRATION SO- - ppm


                 NO- - ppm
   .009
   .019
   .030
(80mg/M3)
.008
.009

.028
.024

.030
                    .050
                  (100mg/M3)
.003
.029
.030
                            LEAD PEROXIDE CANDLES
 21 CANDLES RING THE  STATION AT DISTANCES FROM 2 to 6 MILES.  3 MORE CANDLES
 ARE LOCATED ABOUT 30 MILES NORTH OF THE PLANT, JUST SOUTH OF THE KANSAS
 CITY METROPOLITAN AREA.

 SULFATION GRADUALLY  INCREASED FROM  .029 TO  .031 mg/CM2  PRIOR TO
                                                                 2
 COMMERCIAL PLANT OPERATIONS AND HAS SINCE INCREASED TO  .082 mg/CM  .
                                   Exhibit I
                                     370

-------
  LA  CYGNE  AIR  QUALITY  CONTROL
     MANPOWER  REQUIREMENTS
          OPERATORS PER SHIFT
3 Attendants                                13
3 Clean-Up                                  14
1 Shift Foreman                              5
1 Process Attendant (Chemist)                   1
                  Exhibit J

                        371
                                          33
              MAINTENANCE

Mechanics                                   8
Apprentice Mechanics                          2
Welder                                      1
Electrician                                   1
Technician                                   1
Plant Helpers                                2
Foreman                                    1
                                          16
             ADMINISTRATIVE

 Superintendent                                1
 Engineer                                 	]_

                                            2

                  TOTAL                   51

-------
RECENT SCRUBBER EXPERIENCE AT THE LAWRENCE ENERGY CENTER
           THE KANSAS POWER AND LIGHT COMPANY
      D. M. Miller, Manager, Electricity Production

             Kansas Power and Light Company
                    818 Kansas Avenue
                 Topeka, Kansas   66612
                           373

-------
                Recent Scrubber Experience at the Lawrence Energy Center
                          The Kansas Power and Light  Company
     I would like to share with you our participation in the practical application

of the Combustion Engineering Air Pollution Control  Systems installed  on two steam

generating units located at The Kansas Power and Light Company Lawrence Energy Center.

These steam generators were both supplied by Combustion Engineering  and are capable

of firing pulverized coal, fuel oil, or natural  gas,  each or all  together.

     My Company is The Kansas Power and Light Company headquartered  in Topeka,

Kansas.  We operate wholly in the State of Kansas.   Our peak load this past summer

was just short of 1% million KW.  Our principal  electric generating  stations are

located at Tecumseh, Hutchinson and Lawrence, Kansas.   Over the past 25 years we

have primarily burned natural gas supplemented by fuel oil and/or coal when the gas

supply has been interrupted.  These interruptions have gone from  8-12% in the 50's

and 60's to over 50% in 1975 and we would expect these interruptions co exceed 65%

this year.  This shift from gas to fuel oil and  coal  is not a shock and is  an occu-

rance we had planned for - it is just here a little  earlier than  we had programmed.

     Even though in 1967, 65% of our steam generators were equipped to burn pul-

verized coal, we were still classified as a gas  fired utility.  In 1967 when the

decision was made to build the 400 MW addition at our Lawrence Energy  Center,  it

was quite clear that natural gas would not be available to fulfill its fuel require-

ment.  A 20 year contract for Kansas coal with delivery starting  in 1967  was nego-

tiated to fulfill the primary fuel requirement for this unit.   Natural gas  and  fuel

oil would supplement the coal burning the first  years of operation after  start  up

in the spring of 1971.
                                          374

-------
     Also in 1967 when  the  decision to build this coal burning addition was made,




 it was assumed that by  1971 there would be some ambient and/or emission regulations




 In effect for particulate matter  and sulfur dioxide.  Based on this assumption and




 the availability of 3-4% sulfur,  12% ash coal,  the decision was made to install as




 original equipment, facilities  to treat the stack gas to remove both fly ash and




 sulfur dioxide.  This was accomplished by the installation of a Combustion Engineering



 Limestone Injection-Wet Scrubber  System.





     Since the startup of the 400 MW unit was scheduled for 1971,  it was further




 decided to retro-fit a similar  system on  an existing 125 MW unit at Lawrence so




 that we could gain some operating experience  before  starting up the larger unit.




 This smaller unit went into service  in  the  fall of 1968 and has provided the in-




 dustry with  many  innovations which are  now  being used on newer  installations being



 planned in the  United  States.





     The  Combustion  Engineering Air Pollution Control System at Lawrence  is  designed




 to remove both  fly ash and  sulfur dioxide simultaneously from the flue gas stream.




 Both the  scrubbing of  the fly ash and the chemical reaction  to  absorb the S02 are




 accomplished  in a common vessel.





     I would like to describe for  you the system as it was initially installed on




 our 125,000 KW unit.   (Figure 1)




    The  limestone injection equipment is the existing pulverizers where the rock




 and coal  are pulverized  together and are blown into the furnace as a mixture.  The




 •et scrubber system contains  the spray nozzles,  the marble bed,  the demisters and




 reheater  coils.   The heat source for the reheater  is  extraction steam from the




turbine through the existing  feed  water heater stream.   New I.D. fans were installed




to handle the increase in head requirements.




    The actual  system has two one-half  capacity scrubbers each  with its own inlet




duct and damper, bypass duct and existing  stack.
                                          375

-------
     The bypass system was installed so that we could keep the steam generator

in service on fuels other than coal so that we could maintain reliability of KW

output while learning how to operate the Air Pollution Control System.   We have

been able to do just that as we have operated many hours with scrubbers out of

service.  This has allowed us much opportunity to maintain and modify this system

while generating KWH which still is our primary job.

     The dirty pot overflow water from both scrubbers containing about 2% solids

is pumped to the settling pond where the solids are separated.  The clarified

water is returned through an overflow structure to the clear side of the pond and

to the scrubber supply pump intake structure.  These pumps supply water to the

scrubber system and the cycle repeats.  The supply pumps and ponds will be common

to the 125 and 400 MW units.

     The scrubber on Unit 4 was started up in 1968 with the configuration shown

in Figure 2.  This was a simple system but presented many operating problems and

shortcomings:

     Problems encountered were:

     1.  Build up and plugging of the inlet duct where hot gases
         entered the scrubber.

     2.  Erosion of the scrubber walls.

     3.  Corrosion of the scrubber internals

     4.  Plugging and scaling of drain lines, tanks and pumps

     5.  Plugging and scaling of the marble bed

     6.  Plugging of spray nozzles

     7.  Plugging of demister due to  carryover from bed

     8.  Plugging of reheater due to  carryover through demister

     9.  Buildup on ID Fan  rotors due to moisture getting through
         reheater under abnormal conditions  resulting  in fan
         unbalance.
                                           376

-------
    In addition to the operating  problems  the S02 removal was quite low due to

the everburning of the limestone and  the  dropout of the lime with the ash in the

bottom of the scrubber.  Satisfactory particulate removal was achieved however.

    After the first winter's operation,  the scrubber was revised to the configu-

ration shown in Figure 3.  The  addition of  soot blowers in the inlet duct and  re-

heater helped the plugging problems there.   The demister was raised to increase

the distances between the bed and  demister.   The pot overflows were directed to

the storage pond and a large recycle  tank and pump were installed to accumulate

and pump the highly alkaline bed underflow  back into the bed.  New type spray

nozzles were installed to minimize plugging.  The bottom part of the scrubber  tanks

were lined with gunite to reduce erosion  and corrosion of the walls.  We borrowed

the coal pulverizers from an adjacent coal  fired unit and pulverized the limestone

in one of these mills and blew  the additive material into the furnace on through

the observation doors at the arch  level of  the boiler.  This did provide a reactive

•aterial for use in the scrubbers.

    Most of the problems were  reduced but  not eliminated by the first set of

revisions.  The recirculation  system did improve  the S02 removal.  Maximum S02

removal was possible under  this mode of operation but the resulting sulfur scale

formations  in  the  bed  drain lines  and pumps would soon put  the scrubber out of

business.

    Revisions were made  to the scrubbers the  following  summer in an attempt to

further minimize  the  corrosion, erosion, scaling  and plugging problems.  Figure 4

shows  the configuration of  the scrubber in October, 1970.   Major revisions made

 (ere:
     1.   Sandblasting and  coating the  interior of  the  scrubber vessels with
         a  two coat  glass  flake lining.

     2.   Replacement  of all internal  steel  pipe  systems  with plastic  and
         fiberglass.

     3.   Replacement  of stainless steel demisters with fiberglass  demisters.

                                           377

-------
     4.   Addition of  a ladder vane under  the bed to improve  gas distribution.

     5.   Modification of  the pot overflow drain piping to allow the drains  to
         return to the recycle tank allowing for a semi-closed loop operation.

     6.   The original copper fintube reheat coils were removed and replaced
         with a carbon steel fintube coil.  The copper units plugged  easily
         and the fins were flattened by the soot blower jets.

     7. .  We replaced  the  14" drain outlets from each scrubber with a  2' x 4'
         sluice way to alleviate the drain pluggage we had been experiencing.

     All the above modifications helped in reducing operating problems except  for

the demister plugging.  Scaling could be  controlled by maintaining close ph control

but only at the expense of reduced SO  removal.  Manual washing of the demisters

was necessary on an every other night basis to maintain load capability of  the

unit.

     We operated the scrubber  in this manner with a fair amount of success  through

the winter of 1970-71 and into the spring of 1971.  We were  able  to by frequently

in$pecting the system and rather continually performing small maintenance jobs,

keep the scrubber system available to meet the required coal burning  during this

period.

     We were not satisfied with the reliability of the system at  this time  but

planned to start up the 400 MW Lawrence addition under this  mode  of operation.

I will discuss that operation  later in this report.

     We continued our fight to alleviate  the nagging pains of clogging the  recycle

nozzles or the recycle piping  system and  more carryover from the  bed  than  the

demister could reasonably handle which plagued us with a  dirty  reheater which did

not lend itself to simple repeated cleaning.

     Then came the operation of  the winter of 1971-72 when we were also burning

coal on  the 400 MW Lawrence addition.

     The only thing we could consistently do was make  sulfur bearing  scale  —

on the scrubber walls — in the scrubber  marble bed  and on  the  demisters  of all

the scrubbers in both units.
                                         378

-------
     Operation of the scrubber systems remained in this mode until the late winter




of 1972.  At that time Combustion Engineering completed tests in their laboratory




scrubber at Windsor using a high solid slurry crystallization process to control




saturation and precipitation to eliminate scale within the scrubber.  The system




was quite successful in the lab so a crash program was started to modify the




Lawrence 4 scrubber for full scale testing of the high solids system.




     The temporary revamp took about 30 days to complete and the scrubber was




tested for 60 days during May-June, 1972 using the high solids slurry system to




control scaling.  The results were encouraging; so encouraging in fact that the




decision was made to spend the summer of 1972 to rebuild and modify the scrubber




systems on both Lawrence 4 and 5 to operate with the high solid-crystallization




concept.




     The modification included the installation of:




     1.  Large crystallization tank




     2.  4' enlargement of the scrubber bed on #4




     3.  New rubber lined pipe slurry system




     4.  New plastic spray nozzles




     5.  New bed drain system




     6.  New slurry pumps and strainers




     7.  New multiple mixers in tank




     8.  New two bank fiberglass demister with a power wash system




     This mode of operation is shown on Figure 5.




     The operation of this system on Unit 4 through 1973 was much improved over




anything we had been through in the past.  We still experienced some of the nagging




maintenance and cleaning problems such as:




     1.  Isolated corrosion




     2.  Unsatisfactory damper operation




     3.  Expansion joint failure






                                           379

-------
     4.  Demister fouling




     5.  Rapid slurry pump wear




     6.  Valve failures




but were able by the end of 1973 to maintain the scrubber system operable with




2-8 hour shifts of manual cleaning per scrubber per week.




     This cleaning requirement lessened a little into 1974 when for economical




reasons in the cost of coal, we completed negotiations for a supply of low sulfur




coal from SE Wyoming for these units.  We started receiving this coal: 10,000




BTU/#,  .5% sulfur and 11% ash in the fall of 1974 and had phased out all the SE




Kansas coal by late spring  1975.  The bypass duct system in the Lawrence 4 unit




was removed at this time due to complete deterioration.




     The operation of the scrubber  system on the Wyoming coal has proved to be




 somewhat better  and more economical due to  a lesser amount of sulfur removal re-




 quired.  The  scrubber system is still operating in the high solids mode as an S02




 and particulate  removal system.  Our normal manual cleaning requirements have been




 reduced to 2-4 hour  shifts  per scrubber per week.




     In 1974  this unit was  available for operation 343 days.  50% of the fuel




 consumed was  coal,  2% fuel  oil and  48% natural gas.




     In 1975  this unit was  available for operation 333 days.  64% of the fuel




 consumed was  coal,  3% fuel  oil and  33% natural gas.




     I would  estimated  this unit  to be available  for  operation 330 days in 1976




 with the following  fuel  split; 80%  coal, 8% fuel  oil  and  12% gas.




     Remember the  scrubber  system must be  in operation to burn coal.




     You may have  heard that we  are replacing this scrubber system with a rod




 section followed by a spray tower and wonder why.  You have a right  to that




 question.
                                             380

-------
    We are presently replacing this  successful  scrubber  system because since




1968 in modifying and revising the  scrubber modules  and operating many hours at




corrosion levels that were bad, we  actually consumed the  physical scrubber plant.




Now that we have achieved - at the  Lawrence Energy Center and under  the conditions




of operation there - a mode of operation that we can maintain, we have had so




much deterioration of vessel  and  equipment  that  we must build a new  system to




have one.



    We are extremely interested  in the new system since  parts of  this system




will be installed at our new  Jeffrey Energy Center which  will be discussed later.




    So much for the Lawrence #4  Unit, lets now review the operations of  the




tawrence  #5, 400,000 KW unit  - installed in 1971.




    We followed the programs as  discussed in Lawrence 4  with the scrubber  system




on this unit.   We  started with recycle to the bed - we made scale,  we modified




and installed  the high  solids mode of operation, we added 2 scrubber modules and




we operated burning  coal when it  was required.



    We experienced  the problems  of Lawrence 4 on this unit with one additional




 problem,  a more severe distribution of flue gas to and through  the marble beds




 on #5  than on #4.   This has complicated our lives on #5  unit  and has been a large




 part of the reason our operating successes on #5 have come slower and not reached




 as high a plane as #4.



     However we achieved the following performance with  Unit  #5.




     In 1974 - the unit was available  for operations 338 days - consuming 27%




  coal, 6% fuel oil and 66% gas.



     In 1975 - the unit was available for operations 352 days - consuming 42%




  ieoal, 13% fuel oil and 45% gas.



     I expect  this unit to be available for  service 330  days in 1976 - consuming




  Wl coal, 25%  fuel oil and 15% gas.
                                            381

-------
     We are studying the future of the scrubber system on Unit 5 and my best




guess today is that it will follow in the footsteps of Unit 4 and will be converted




to the rod and spray tower system.




     These experiences at our Lawrence Energy Center are the basis for our con-




clusions concerning the stack gas clean-up system we intend to install at the




Jeffrey Energy Center. (Figure 6)




     The Jeffrey Energy Center is composed of 4 - 680,000 KW coal fired units




planned to be on line 1 unit early 1978, 1 in 1980, 1 in 1982 and 1 in 1984.




All of these units will be fired by coal now under contract from the AMAX Coal




Company from the Powder River Basin in Wyoming.  The coal specifications for this




plant are designed to be 8000 BTU/#, .32% sulfur and 5.8% ash.




     We have purchased from Combustion Engineering a stack gas clean-up system




that will remove in excess of 99% of the particulate matter, and 60% of the S02




in the flue gas.  An overfire air system at the tangential fired pulverized burners




will control tk» BMI emission.




     The stack. fM&ft&Mn-** •*»«•* (Fig«r« 7) is composed of an electrostatic




praeipitator following the air heater them the induced draft fans and then through




spray towers.  Th« effluent gas fro« the spray towers will be mixed with some hot




gas from the ID fan discharge for reheating the flue gas as it flows to and out the




600 ft. chimney.




     We are of the opinion that we will be able to operate this system under the




conditions we will have at the Jeffrey Energy Center in such a manner to be good




neighbors to those who live around us in Pottawatomie County, Kansas and to satisfy




the existing requirements of The Department of Health and Environment of the State




Of Kansas and the Environmental Protection Agency.
                                            382

-------
                     ENGINEERING
   AIR POLLUTION CONTROL SYSTEM
          MNSIS ROWRUICKT COMPANY
            L«ren«  Station Unit No 5
OU*
                                                          STACK
                                                            CAS
                                                       SCRUBBER

-------
                                                •
                                                 .          ! H AIR

1
                                                                                                                                  .

1
                                                                                                                                                                 LAWRENCE  N* 4

                                                                                                                                                                  DECEMBER 1966
                                                                                                                                                                     SCRUBBER
                                                                                                                                                       SOLID DISPOSAL  '
Oi
CO
•u
                    WATER



                             ///////////////


                                             OEMISTER
                                                    SPRAY
                                                                                                                                  TO STACK
                                                                                   N* 4
                                                                                  APCS
                                                                             ')8ER  1970
                                                                           SCRUBBERS  (2)
                                                                          CLARIFIED

                                                                          POND

                                                              DRAIN TANK
                                                                                                                                          •  SOOT BLOWER AIR
                                                                                                                                            WATER WASH LANCE
                                                                                            **•
                                                                             LAWRENCE  N* 4
                                                                                    CE-APCS
                                                                                OCTOBER 1972
                                                                                 SCRUBBERS  (2)
                                                                            (ENLARGED DEPTH-4')
                                                                                                                     RECYCLE TANK (I)
                                                                                                                       C ENI. AROEO)
                                                                                                                                                  CHAIN TANK
                                                                                                                                                                       CLARIFIED
                                                                                                                                                                       FROM
                                                                                                                                                                       PONO
                                                                                                                                                                      TO
                                                                                                                                                                      SOLID
                                                                                                                                                                      DISPOSAL
                                                                                                                                                                r(J)    POND

-------
  -
  o
'    •

                                                                                                                                                                                                                                                   ---|!M!!J!|a|!!!|l" "'"'.^^'^" """T

                                                                                                                                                                                                                                                      ™  **	1  mt, MJHI - M. OB.™	1	

-------
                   INTRODUCTION TO: DOUBLE ALKALI
                FLUE GAS DESULFURIZATION TECHNOLOGY
                           Norman: Kaplan

           Industrial Environmental. Re&earch Laboratory
                Office of Research and Development
                  Environmental Protection Agency
          Research Triangle Park, North Carolina   27711
ABSTRACT

     The chemistry and significant process and design factors applicable
to sodium/calcium double alkali systems are presented.  Technical
terminology associated with these systems is defined.

     Estimates of capital and operating costs are presented in addition
to reasonable targets for reagent consumption in double alkali systems.

     The EPA development program is summarized and a status of technology
is presented.
                                387

-------
                              NOTES

1.  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.

2.  Consistency of Information.

    The  information presented 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.


3.  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,  IERL/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)                   °C
           1 ft                          0.3048 meter
           1 ft2                         0.0929 meters2
           1 ft3                         0.0283 meters^
           1 grain                       0.0648 gram
           1 in.                          2.54 centimeters
           1 in.2                        6.452 centimeters2
           1 in. ^                        16.39 centimeters^
           1 Ib (avoir.)                 0.4536 kilogram
           1 ton (long)                  1.0160 metric tons
           1 ton (short)                 0.9072 metric tons
           1 gal.                        3.7853 liters

                                   388

-------
INTRODUCTION AND BACKGROUND
     At the last EPA Symposium on Flue Gas Desulfurization (FGD),
held in Atlanta in November 1974, plans were announced for an
EPA co-funded full-scale utility boiler double alkali (D/A)
demonstration program in which the system vendor and utility
would be selected on a competitive basis.  At that time there
were two industrial boiler applications and an industrial kiln
control application of the technology in this country.  Japan
was a bit more advanced in the application of this technology
in that there'was one 150 Mw application of the technology in a
full-scale utility system.   As a result of the increased testing
of D/A systems in this country in various pilot plants, at a
20 Mw utility prototype system at Gulf Power, at various industrial
boiler systems and other industrial applications (sulfuric acid
plants, kilns,etc.) and the interest generated by EPA's request
for proposal (RFP) for a full-scale utility demonstration of D/A
technology, this technology now appears to be competitive with
lime/limestone wet scrubbing for some utility applications.

     In response to the RFP, EPA received three viable proposals
for installation of double alkali systems at full-scale coal-
fired utility systems, ranging in size from 192 to 575 Mw.

     In Japan there are now three operational full-scale utility
applications ranging in size from 150 to 450 Mw.18  Additionally,
Kureha has developed another variation of the D/A process which,
reportedly, is a significant improvement over their limestone D/A
system that they now have operating as a full-scale utility system.
This variation is called the "Sodium Acetate-Gypsum" process and
uses sodium acetate as the S02 absorbent, limestone as the regen-
erant, and produces a gypsum by-product.  This process is currently
being pilot tested at a-5000 Nm3/hr (1 Mw) pilot plant using a
perforated tray tower operating at a 280 mm  (11 inches) H20 pressure
drop, L/G of 7-8 gal./lOOO ft3 gas with an S02 inlet concentration of
1400 ppm and an exit concentration of less than 10 ppm.  This
process will be discussed in greater detail in a later paper in
this session.

     "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:
                                389

-------
lime or limestone is consumed,  and  a  calcium  sulfite/sulfate and
fly ash 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/limestone process the absorption
of the SO  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
SO  absorption efficiency  in a  double alkali  system is potentially
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 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, Na2S03, Na2C03, NaHC03)  is  used  to
 absorb SOX from  the flue  gas in  the  scrubber, and  then a  calcium based
 alkali (Ca(OH)2,  CaO,  CaO^) is reacted with the SOx-rich scrubber
 effluent liquid  to precipitate the insoluble CaS03-l/2 H20,  CaSO^-2H20
 and mixed crystal and  regenerate sodium based soluble alkali for recycle
 to the scrubber  system.   Double alkali  systems with  an ammonium/calcium
 base have been tested:  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.   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 throwaway  product.
                                   390

-------
DliFIN'TION 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  terms are  understandable  to  those
dealing with the subject on a daily  basis,  they 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 d Lscussed 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 witli  limestone.   This wider  pH range  allows
lime systems to operate over  the complete range of active alkali
hydroxide/sulfiLe/bisulfite  whereas  limestone systems  can only
operate  in the suIfite/bisulfite range.

     The main overall absorption reactions  are described by the
following oc|unl ions : .

     2NaOll + SO.,  >  Na.,S03 +  t^O

     Na,,SO  + SO   t- II 0 •> 2NaHSO^
       2.   j     t-    £•           ^
The main overall  regeneration reactions are described  by the  following

equations  for lime  and limestone respectively:

     Lime^

     Ca(0ll)2 + 2NaHS03  > Na2S03  +  CaSCyl/2 H20* + 3/2 H20           (3)

     Ca(OII)., + Na2S03 +  1/2  H^O  ->  2NaOH + CaSO^l/2 H^             (4)

     Ca(0nf + Na  SO +  2H  0 J  CaSO, -2H?0 * + 2NaOH                  (5)
     \_* rt V W 1 I / ' \  ' !»<-*.v./vy,      n   -^      ^   ^
                               391

-------
LJ mo stone

CaCO  + 2NaHS0
                 1/2
3 -I- CaScyl/2
                                                   +
                                                                     (6)
    D/A systems are frequently and erroneously referred to as
sodium ion scrubbing systems.  It should be stated that, from a
"pure" chemistry viewpoint, the reactions presented in equations
(1-11) and (15-18) do not involve the sodium ion (Na+) ; however,
the presentation is made using compounds of sodium because sodium
systems are prevalent in D/A applications and because this allows
showing the reactions using electrically neutral reactants and
products rather than charged ions.  For example, the absorption
reactions involve reaction of S02 with an aqueous base such as OH  ,
SOo, HCOo or 663 rather than with Na+ which does not take
part in the reaction, but is only present to maintain electrical
neutrality.  Thus equations  (1) and  (2) for example could have been
written as equations (la) and (2a) respectively:
2 OH" + S02  ->
      $0
                H20
                              H20

                             21IS0
                              (la)

                              (2a)
 Active  Alkali
      This  term  is  the  sum  of  the  concentrations of NaOH,  Na2C03?
 Na.,50  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
 S0? per mole of active alknli.  This  difference in molar  capacity for
 absorption of SO.,  is  illustrated by the  following reaction equations:
      Na-CO., + 2SO- + H_0 -•• 2NaHSO  + CO
        4.   -J      ~    *-          -*     *-
           (sodium carbonate molar capacity:  2 moles SO  /mole)

      NaHC0 + S0  -
           (sodium bicarbonate molar capacity:   1 mole  SO  /mole)
      NaOH  + SO   " NaHSO
           (sodium  hydroxide molar capacity:   1  mole  SO  /mole)
      Na S03 + SO  4- HO ->  2NaHSO

           (sodium sulfite  molar capacity:   1  mole  S02/mole)

      NaHSO- 4 SO  ->- No reaction
           (sodium bisulfite molar  capacity:   zero  mole S02/mole)

                                     392
                                                                 (7)


                                                                 (8)


                                                                 (9)


                                                                (10)


                                                                (ID

-------
    Molar capacity  is  simply the number of moles of S0? 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 S0? ,  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  SC>2  (moles of 80,,/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 SOo capacity (moles/liter) =
     2  [Na2C03J  + iNaOHj  + [NaHCO-j] +
 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 -t-4  oxidation state.  Simply, this is the total
 concentration of  sulfite plus bisulfite.

     TOS  (moles/liter) = [SO  =] + [HSO ""]

 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 l'ie TOS  in solution.

 Active Sodium

     This is the  concentration  of sodium in solution which is
 associated witli the active alkali.

     [Na+]    = [NaOH] +'2 [Na.,COj  + [NallCO.. J  + [NaHSO ] + 2 [Na SO ]
          act                  /   j          J          -3         ^  J
 If NaOH,  NatlCO ,  Na CO , NaHSO  or Na S03 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 uf S0? in  scrubber solutions.  Note that the term
 "active sodium" can be misleading in that the sodium ion doesn't
 participate in any of the  process reactions.
                                   393

-------
Oxidation

     Oxidation in a double alkali system refers to the conversion of
TOS to sulfate by one of the following equations:

     HSO ~ + 1/2 0  ->• S0,= + H*                                     (12)

     S03= + 1/2 02 > S04=                                           (13)

Simple oxidation of S0? to SO. in the flue gas is also considered
oxidation in the double alkali system:

     SO, + 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.  In
general, rate of oxidation in the system is thought  to be a function
of rate of dissolution of oxygen, pH of the scrubbing solution,
impurities present in solution and concentration of  reactants.  Oxida-
tion rate is thus affected by composition of 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
and number of scrubber contact stages is extremely important).

     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)]     ^QQ
                          [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.  What is really meant  is sulfate removal
from the system with regeneration of active alkali from inactive  sodium
sulfate.  (See Sulfate Removal.)
                                394

-------
Sulfate Removal

     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. 4- Ca(OH)9 + 2H 0
  /  4         i     2.
                          2NaOH + CaSO  -2H004-
                                      42
                                    (gypsum)
                                                                     (15)
Na.-SO.
  /  4
              2CaSO  -1/2' H00  + H0SO.  + 3H_0 ->• 2NaHSO, + 2CaSO, -2H-OI (16)
                    3       i      2   4      2.          3        4    /
                                                            (gypsum)
y Na?SO. + x NaHSO  +  (x+y)  Ca(OH).  + (z-x)  HO >
    £f  ^T          J                £           *~
  (x+2y) NaOH + x CaSO -y CaSO, 'z tt^Q^

                (mixed crystal or solid solution)
                                                                     (17)
Na0S0/
  24
3H00
  2
                    ElGCtr°lytic
                                > 2NaOH
                                                       1/2 0
                                                                     (18)
     Sulfate removal  should be accomplished in an environmentally acceptable
 manner; a simple  purge  of  soluble Na2SO/  from the system to land or water-
 way disposal is not acceptable in large ^quantities in most cases.
 Softening

      This term is used to describe various  methods  used  to  lower  the
 dissolved calcium ion concentration  in  regeneration 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:
      + Na2C03 -> 2Na  +

      + Na2S03 + 1/2 ti^ -»•  2Na+ +  CaSCyl/2
                                                                      (19)

                                                                      (20)

                                                                      (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  carbonate softening.  Reaction (20)  is
 considered sulfite softening.   Generally,  dilute systems employ carbonate
 softening while concentrated  systems  resist scaling due to the high
 sulfite concentrations which  prevent  high Ca++ ion in the scrubber liquor.
                                  395

-------
Dilute vs. 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 when using relatively small quantities
of lime slurry for regeneration, unless the concentrations of sulfite
and sulfate are present in a certain ratio.  This can be shown by
dividing one solubility product equation by the other:

     tea"""] [S04=] = KsPl
     [Ca  J [S03 ] = Ksp2

                        I *
from this, cancelling Ca   ion concentration,

     [so4=j
     	-—  =  constant
     iso3~]

The ratio of sulfate to sulfite for simultaneous precipitation of
CaSO. and CaSOo 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
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. >3

     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."
                             396

-------
Lime or Limestone Stoichiometry

     Lime or limestone stoicbiometry  can  be  expressed as  a percentage,
based on an overall material balance  around  the system:
     Lime Stoichiometry =
     Limestone Stoichiometry  =
moles CaO added	
mole sulfur collected

     moles CaCOn added
100
                                mole  sulfur collected
                                100
Lime or Limestone Stoichiometry  is an indication of the efficiency of
usnge of lime or limestone  in the system.   Ignoring alkali components
in the flyash collected and  the  alkalinity added with sodium make-up
to the system, 100% Stoichiometry is  complete utilization' stoichiometriee
over 100% represent less  efficient utilization of lime or limestone.
Stoichiometries under 100%  indicate alkalinity from ocher sources.
 Feed S t o i c h iome t ry

     Tins is  calculated  hy a material balance around the scrubber.
 ll  is usually  expressed  as the ratio:
     ir , , i  q,  • .1  =  .".j--i"-"r--SQo Capacity (moles/liter) x Flow (liters/min)
     r e eel  o L o i (-11             <-         /-»/^  /  i   i •  \          ——^—^——^-~—
                                       SO,, (mole/min)

                    Tliis  ratio is evaluated for streams
                    entering the scrubber.
      Feed Stoichiometry is a measure of  the  ability  of  the  incoming
 liquor to react with or absorb all of  the  incoming SO   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  S07  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.
                               397

-------
SIGNIFICANT PROCESS AND DESIGN FACTORS

     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.

SO? Removal

     The fact that small quantities of sulfur dioxide can be removed
from large amounts of relatively inert gas by cyclic processes involving
absorption into aqueous solutions of sodium sulfite/bisulfite has been
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 S02 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 SO
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 careful selection of materials of  construction and
equipment and by providing spares for certain equipment such  as  pumps
                                 398

-------
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, excessive
sulfate build-up, production of poor-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
process 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.  Bo.th 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^ + S04= + 2H20    ->    CaS04'2H20  +      (22)


     In dilute systems gypsum scaling is controlled by softening the
regenerated liquor prior to recycling to the scrubber while in con-
centrated systems gypsum is not a problem since the high sulfite concen-
tration keeps the Ca++ ion low.  Softening ensures  that the liquor
recycled  to the scrubber system is unsaturated with respect to gypsum;
therefore, with proper softening even if some sulfate  is formed  xn
the scrubber, the liquor will not be saturated with gypsum  and cause
                               399

-------
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 C0£ 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 4- 2  OH~    -    C03= +  H20             (23)

               Calcium carbonate scaling:
               C03= +  Ca++   •*   CaC03   +                     (24)


      Based  on  experience with  the General Motors  full  scale  double
 alkali  system, 5  carbonate scaling could  occur with scrubber  liquor
 pH 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=    C    HC03"                         (25)

 Thus, carbonate  scaling can  be eliminated by control of pH in  the
 scrubber.

      b.   Production  of Poor-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

                                    400

-------
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, but is thought to be a
function of reactor kinetics.^

     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  2000 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.

     Envirotech6 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
over a  greater  range  of  conditions than  a simple  CFSTR.

     c   Reliability/Availability  Data  - Three full-scale utility
and several industrial boiler double alkali  systems have started
operation  in  Japan since 1973.   Although detailed availability data
 is not  shown  here  for these  units,  few,  it any,  significant  operating
problems have been reported  for  them.   Ando  has  given more detail
on these systems  in another  paper.-1
                                 401

-------
     In the U.S. the 20 Mw utility prototype system constructed by
CEA./ADL for Southern Services at Gulf Power's Scholz station has
been in operation for almost a year.  That system started up in
February 1975 and operated intermittently through January 1976.  It
was snut down for modification and repairs from mid-July through
mid-September 1975.  Excluding that 2-month period, the system
maintained a 90% availability during the first, approximately 1 year
of operation.  Most of the problems experienced were related to
the filtration system.

     FMCrs double alkali kiln control system has been in operation
at their Modesto, California, plant for over 5 years.  That system
is equivalent to 10 Mw based on gas rate or about 30 Mw based on
the size of the regeneration system and S02 absorbed.  As of approx-
imately a year ago, that system reportedly was available for control
of SOX emissions from the barium reduction kilns 100% of the time
required.  The kilns were in service 95% of the time with scheduled
shutdowns for routine maintenance and/or change of chemical process
service every 3 or 4 months.
Environmentally Acceptable Solid Waste

     Double alkali systems should be designed to produce an environ-
mentally acceptable end product.  Some of the properties which can be
ascribed to such a solid waste product are listed below:

          -Non toxic
          •Low soluble solids content, non-leachate
          •Low moisture content
          •Non-thixotropic
          -High compressive or bearing strength
                               402

-------
     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 (Na^O^) , 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 SOo removal capability due to the loss of active sodium
from the system.

     Equations  (15), (16), (17) and  (18), shown previouslv 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  + CaS04'2H2(H     (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:
                                  403

-------
            "The presence of Na2SO^ in the scrubber effluent
          is the prime factor influencing the design of the
          regeneration system.  Na-^SO^ is not easily regenerable
          into NaOH using lime.  The reason being that the
          product, gypsum, is relatively soluble. . .  . ^280^
          cannot be causticized in the presence of appreciable
          amounts of S0-j= or OH~ because Ca"1"4" levels are held
          below the CaSO^ solubility product.  To provide for
          sulfate causticization, the system must be operated
          at dilute 01I~ concentrations below 0.14 molar.  At
          the same time, S0^= 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:
     Na2S04 + 2CaS03-l/2 H20 + H2S04 + 3H20 •> 2NaHS03 + 2CaS04'2H20-t-    (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 sulfurie acid added.  In practice, however, this is not the case
since  any material which functions as a base can consume sulfurie
acid and reduce the efficiency of this reaction for its intended
purpose.-^  Unreacted lime or limestone, sulfite ion and even sulfate
ion can consume sulfurie acid thus lowering sulfate removal from the
system.
                              404

-------
     Conceivably,  this method of sulfate removal may be 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) Na01l'+ x CaSO-j-y CaS04-z H20  4-
                           (mixed crystal or solid solution)

This phenomenon is described by R.H. Borgwardt of EPA2 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 conditions 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 [S04=]/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
                                  405

-------
systems; i.e., gypsun cannot be precipitated from solutions
containing high active alkali concentrations.  It was a simple
technique for sulfate removal in concentrated systems.  The [SO^j/
total [SO,.} ratio observed in pilot double alkali work was as
             ..
high as 0.20.-5  Coincidentally, 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,
sulfatc 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 sulfatre removal as sulfuric acid
in an electrolytic cell:

                     Electrolytic
      Na2S04 + 3H20     cell     ^   2Na01I + 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
;;scd sulfuric acid addition method.  In addition, they feel that
sodium losses from the system can be cut in half through the use of
this methodt  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 Na2S04 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 ionic  strength.   One possible


                               406

-------
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 fresh water is added to a  D/A system to serve many
purposes.  These include:

          •Saturation of flue gas
          •Pump seal needs
          'Demister washing needs
          •Slurry make-up needs
          •Waste product washing
          •Tank evaporation

     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
operating cost factor.  Sodium make-up to double alkali systems is
usually accomplished by adding soda ash  (recently quoted at $70 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.
                                 407

-------
     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 (IDS) 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.  Un-
limited 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 fly ash, flue gas,  the lime and/or
limestone and the make-up water.  Of these, probably the soluble
material  in highest  concentration would be sodium chloride which results
from the  absorption  of I1C1 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  (CaS04'2H20)
 is  more soluble than calcium sulfite hemihydrate  (CaS03'l/2 H20),
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 waste  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 which  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
                                 408

-------
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 657, solids.10

     Some explanation for the behavior of these sludges is given by
Selmeczi and Knight.11  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
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.1^  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 T.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
                                    409

-------
 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 which 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. Concentrated System

     The selection of a dilute or concentrated double alkaJi 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,
the desirability to produce a manageable  solid waste (dilute systems
can be designed to produce high gypsum sludges) could, In cases,
override the capital cost issue.
                               410

-------
Sodium Consumption

    Sodium consumption is an important performance criterion for
a double alkali system not so much from the viewpoint of economics
as from the viewpoint of producing a leachable waste product as
stressed under sulfate removal.  Sodium consumption is only a
minor factor in the operating cost of a system representing about
2% of the annual operating cost.  Thus even if soda ash make-up
to a system were to increase by 100% over the expected value,
operating cost might only be increased by a factor of 1.02.  On
the other hand the environmental consequences of higher sodium
consumption may be significant if the sodium is leachable from
the waste product to the environment.

    A logical way to measure sodium consumption is in moles of
Na consumed per mole of sulfur removed by the system.  A value of
0.05 moles of Na make-up per mole of sulfur removed (equivalent
to 0.025 moles of Na2C03 make-up per mole of sulfur removed)
appears to be a reasonable design target based on present U.S.
technology.  This target is achievable in concentrated alkali
systems burning relatively high sulfur coal (over 3% sulfur) and
having a relatively low oxygen content flue gas.  In Japan sodium
make-up is reportedly as low as 0.02 moles Na/mole sulfur removed
for some systems.

Calcium Consumption

    A logical way to specify calcium consumption is as calcium
stoichiometry moles of calcium added per mole of sulfur removed
(or collected).  A calcium consumption of 0.98 to 1.0 appears
to be a reasonable design target for concentrated double alkali
systems (values under 1.0 are possible due to the additional alkali
added with sodium make-up to the system).

Power Consumption

    Design targets for power consumption can be in  the range  of
1-2% of power plant output without reheat or under 3% even with
50°F of reheat.  These figures are based on a system which has a
scrubber pressure drop in the range of 6-8 in. H20 and a scrubber
system L/C ratio of about 10 gal./lOOO ft3.  This assumes that the
power plant is equipped with some means of efficient particulate
collection upstream o! the FGD (e.g., an electrostatic precipitator)
                               411

-------
    Low power consumption is a major selling point for double
alkali FGD systems.  Some other FGD systems are estimated to run
as high as 10% in power consumption.

Economics
    It appears  that double alkali systems are now economically
competitive with  the  "first  generation" wet alkali lime/limestone
slurry scrubbing  systems.  This  is especially true in cases where
the lime or limestone system would be required  to be equipped with
solids separation equipment  (i.e. thickener and filter).
    Some of the factors  that allow a double alkali system to be  less
expensive  than  a  lime/limestone  system both in  initial  cost and
annualized operating  cost are:

          • Lower  scrubber liquid/gas ratio (L/G)
          • Lower  scrubber nressure drop (AP)
          • Simpler scrubber  design
              -  fewer  stages
              -  no slurry in  scrubber
          • Less exotic materials of construction
          • Solid  waste with  better  handling  properties
      It is estimated that a new (or simple  retrofit)  double  alkali
  system could be installed at a 200 Mw or larger  plant  for a cost of
  $50 - 60/kw  and operated at 2.5 - 3  mills/kw-hr excluding  sludge
  disposal (which would vary with the particular application).
  These estimates are based on the following assumptions:

           200 Mw or larger system
           3-4% sulfur  coal
           80% load factor
           Capital charges @15-16% of capital investment,  annually
           Maintenance @3-4% of capital investment, annually
           Operation with two operators/shift
           Power @$0.02/kw-hr
           Soda Ash @ $65-7O/ton
           Lime @ $30-33/ton

      It appears that economics of scale for units above 200  Mw in
  size are not very significant since larger units would probably
  involve the installation of additional modules.   There would be
  some economy of scale in the savings  on design,  engineering,  site
  preparation, etc.
                                  412

-------
THE EPA DOUBLE ALKALI RESEARCH, DEVELOPMENT AND DEMONSTRATION
PROGRAM

     EPA has been Involved in the development of double alkali technology
since the Second International Lime/Limestone Wet-Scrubbing Symposium
held in New Orleans in November 1971.  Some of. the incentive to develop
this technology stemmed from a paper presented by R.J. Phillips
of General Motors (GM) concerning GM's laboratory and pilot plant
work with a dilute double alkali system which appeared very encouraging
at that time in light of the difficulties which were being experienced
with lime/limestone systems.  The development of double alkali technology
by EPA has followed an orderly progression of scale from laboratory
to pilot plant to prototype and finally to a planned full-scale utility
demonstration of the process.  Along the way, EPA also embarked on a
program to evaluate a full-scale dilute double alkali system in operation
at a GM industrial boiler system  (32 Mw equivalent).

     Some initial laboratory work on regeneration chemistry was
done in the EPA laboratories at Research Triangle Park in  addition
to an  initial feasibility study13 which indicated that double
alkali systems might be somewhat  lower in  capital and operating
costs  than  lime or limestone systems under certain  circumstances.

     After  these initial  studies, EPA contracted with Arthur D. Little
 (ADD  to  conduct a study  of  the double alkali  process.  The  scope ot
work included In the  initial laboratory and  pilot plant Program  was
subsequently  expanded  to  include  prototype testing  at  the  Scholz Plant
of  Gulf  Power Company  where  a  20  Mw FGD prototype system was  constructed
by Combustion Equipment Associates  (CEA)/ADL for  Southern  Services.   The
construction  prototype unit  was  funded by  The  Southern Company.

     The  initial goals of  the  EPA double  alkali program were to:

      •Demonstrate  reliable  system operation
      •Demonstrate  high S02  removal, 95% desirable
      •Demonstrate  environmentally acceptable sulfate removal schemes
      •Minimize  soluble material  in  disposable waste
      •Minimize  moisture in disposable waste
      •Demonstrate  closed-loop  operation
      •Minimize  costs
      •Minimize  Ca^ concentration in the scrubber

      To date a  high degree of  success with these goals has been
 achieved up to  the prototype level.
                                 413

-------
Laboratory

     Some of the areas of investigation in the laboratory program
are listed below:

     •Regeneration of simulated scrubber effluents with lime
      and limestone                                               .
     •Sulfate removal by precipitation of a mixed crystal containing
      CaSOA and  CaS03 with water of hydration
     •Sulfate removal by reaction of Na2S04, CaS03 and sulfuric acid
     •Feasible ranges of sulfate, chloride, magnesium and iron in
      solution                                    .
      •Settling characteristics of the product solids
      •Fixation of  double alkali product solids
      • Density, comparability, leachability and  permeability  of
       fixed and  unfixed  solids

 Pilot  Plant

      In the pilot  plant  both short  term and  long term  runs  (v, 5 weeks/
 run) have been conducted to examine various  modes of double alkali
 operation.  Some of these modes  of  operation are listed  below:
      •Concentrated alkali, sulfuric acid  treatment,  lime
      •Concentrated alkali, two-stage reactor system,  lime
      •Concentrated alkali, single CSTR reactor,  lime
      •Dilute alkali w/sulfite oxidation,  lime
      •Concentrated alkali, multi stage reactor,  limestone regeneration
      •Extra high concentrated alkali, two stage reactor system,  lime

      Much  of the laboratory and pilot plant work has been reported
 by LaMantia, et al.3 in a paper presented at the last EPA Symposium
 on FGD  in  Atlanta.

 Prototype

      The  results of testing to date at the 20 Mw prototype system
 at Gulf Power will be reported in a paper entitled, "Operating
 Experience — CEA/ADL Dual Alkali Prototype System at Gulf Power/
 Southern  Services, Inc."  by LaMantia, et al.17

 GM  Industrial Boiler

      The  GM test  program  at  the  32 Mw  equivalent system at the Chevrolet
 Transmission plant in Parma, Ohio  is being conducted under an agreement
 between EPA and GM.  The -test program  design and some chemical analyses
 are being done  for GM and EPA under contract by ADL.  The FGD system consists
  of  four Koch tray stainless  steel  scrubbers,  32 Mw equivalent, and a 40 Mw
  equivalent regeneration system  consisting basically of  two mix  tanks and


                                  414

-------
two reactor clarifiers.   The boiler system consists of two 60,000
Ib/hr (of  steam)  and two 100,000 Ib/hr boilers.  The GM system
is a dilute alkali system.

     The system has been operated intermittently since start-up  in
February 1974.   Availability data for the system is not easily
defined since there are four boilers and four scrubbers while
the steam demand is frequently less than the capacity of one or
two boilers.  Thus, when there is a problem with one of the
systems it can be taken out of service and replaced by one of
the stand-by units.

     The GM system cost approximately $3.5 million.

     A good account of this system was presented in the last EPA
Symposium of FGD in Atlanta by Dingo and Piasecki. 6

Full Scale Utility Demonstration

     The RFP by EPA for a full-scale utility demonstration was
issued in May 1975.  Proposals were received in response in
August 1975.  At the time of  this writing  (Feb. 1976), the proposals
are in the final stages of evaluation.

     The RFP called for a 100 Mw or larger demonstration of the
technology.  The three proposals receiving final consideration range
in size from 192 to 575 Mw.   The contract will be  let  for a four-
phase  program consisting  of:

     (I) Process design and cost estimate
     (II) Engineering design, construction and mechanical testing
     (III) Start-up and acceptance  testing
     (IV) One year  operation  and long  term testing

     A summary of  the design  criteria  requested  for  the demonstration
unit is given below:

     •Unit will meet all  applicable pollution  control regulations
     -Pulverized coal fired boiler  burning 2.5  -  4.5% sulfur  coal
     •Instrumentation must be provided to  allow accurate  material
       and  energy balances at  the demonstration plant
     •S02  control  to below  200  ppm emissions by the scrubber
     •Stack  gas  reheat  required
     •Chemical make-up  requirements requested
        -Na   0.05 moles/mole  S
        -Ca   1.2 moles/mole  S
     .Acceptable  long-range waste  disposal plan

     We anticipate conducting an extensive test program at  the  full-scale
 facility  which would  characterize  effluents, evaluate chemical  consumption,
 evaluate  sludge  disposal, optimize operating conditions and evaluate system
 reliability and  process economics.

                                415

-------
SUMMARY OF STATUS OF TECHNOLOGY

     Based on the initial performance and reliability demonstrated in
various double alkali pilot and prototype plants,  and in an increasing
number of full scale applications in the U.S.  and  Japan, it appears
that this technology has become a viable technological alternative
throwaway process to the lime/limestone processes.


      Performance of double  alkali  systems with respect  to  S02 removal
 is  well  established.   Over  99% S0£ removal has been achieved with  lower
 than 10  ppm S02 concentration in the  treated  flue gas.

      Potential  environmental  problems associated with waste disposal
 from these  systems  may occur  due to the preserce 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 the waste  product from double alkali  systems than from
 lime/limestone systems.  With present  U.S. technology  it appears  that
 the incremental amount of solubles in a double alkali waste product
 (over that  typically present  in lime/limestone waste product) is
 typically about 1-2% on a dry solid? 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 avail-
 able information put double alkali system capital costs in the range
 of lime/limestone system costs, at about  $50-60/kw capital and
 2.5-3 mills per kw-hr  operating cost.

      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;
                                    416

-------
whereas, production of a throwaway solid waste product is the
general rule with development of the technology in the U.S.

     The table presented below shows a total of approximately
3900 Mw equivalent operating and planned double alkali systems
in industrial and utility applications in the United States and
Japan representing 23 applications of this technology.  Of these,
approximately 1750 Mw equivalent representing 16 applications
are operational.  In the U.S. six operational units approximately
equivalent to 140 Mw are listed.  As a standard for comparison
the latest PEDCo Survey of FGD systems in the U.S.1^ indicates a
total of 42,160 Mw representing 109 units of operating and planned
FGD systems.  Of these 3828 Mw representing 22 units are operational.

     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 established for the lime/limestone systems at a corresponding
stage of development.
                                 417

-------
SUMMARY OF SIGNIFICANT OPERATING AND PLANNED FULL SCALE DOUBLE ALKALI  SYSTEMS

System Operator
FMC
Modesto Calif. Plant

Showa Denko KK
Chiba, Japan
Tohoku Electric
Shinsendai Sta. ,
Japan
General Motors
Parma, Ohio Plant
Caterpillar Tractor Co.
Joliet, Illinois
Showa Pet. Chetn.
Kawasaki, Japan
Kanegafuchi Chera.
Takasago, Japan
Poly Plastic
Fuji, Japan
Kyowa Pet. Chem.
Yokkaichi, Japan
Kinuura Utility
Nagoya, Japan
Daishuwa Paper
Fuj 1 , Japan

System ADp_lica t ion

Reduction kilns

Oil-fired elee.
power boiler
Oil-fired
utility boiler

4 coal-fired
InJustrial boilers '
2 coal-fired
industrial boilers
Oil-fired
industrial boiler
Oil-fired
industrial boiler
Oil-fired
Industrial boiler
Oil-fired
industrial boilar
Oil-fired
industrial boiler
Oil-fired
boiler
Vendor or
Developer

FMC

Showa
Denko
Kawasaki/
Kureha

General
Motors
Zurn
Industries
Showa
Denko
Showa
Denko
Showa
Denko/Ebara
Showa
Denko/Ebara
Tsukishii-ia

Tsukishima

Size
(Mw. Equivalent)

10 (Gas Rate-)
30 (Regen.)
150

150


40 (Regen)
32 (Gas Rate)
v 20-30

62

93

65

46

63

85

Active
Alkali

Cone .

Cone .

Cone.


Dilute

Dilute

Cone.

Cone.

Cone.

Cone.

Cone.

Cone .

Calciur. Start-Up
Sources Dite'1

Lin't Dec. 1971

Limestone June 1973

Limestone Jan. 1974


lime Mar. 1974

Lime Oct. 1974

Limestone 1974

Limestone 1974

Limestone 1974

Limestone 1974

Lime 1974

Lime 1974


-------
                             SUMMARY OF SIGNIFICANT  OPERATING AND PLANNED FULL SCALE DOUBLE ALKALI SYSTEMS  (Continued)
      System Opcra_tcrt^

Firestone
  Pottstown, Pa.
Gulf Power Company
  Scholz Plant
  (Southern Services)
Caterpillar Tractor Co.
  Mossville,  Illinois
Sikoku Electric Power
  Anan, Japan

Sikoku Electric Power
  Sakaide,  Japan

Kyushu Electric Power
  Buzen,  Japan

Caterpillar Tractor
  East Peoria,  Illinois

Central  Illinois  Publicb
  Service - Newton til
Louisville Gas  &  Electric1'
  Cane Run #6
  Louisville,  Kentucky
 Springfield Water Light*5
   and Power - Dallman  #3
   Springfield,  Illinois

 Caterpillar Tractor
   Mapleton, Illinois

System Application
Coal 6. oil-fired
industrial boiler-
demonstration
Coal-fired
utility boiler-
prototype
Coal-fired
industrial boiler
system
Oil-fired
utility boiler
Oil-fired
utility boiler
2 oil-fired
utility boilers
Coal-fired
industrial boiler
Coal-fired
utility
Coal-fired
utility
Coal-fired
utility
Vendor or
Developer
FMC


A.D. Little''
Combust . Equip .
Assoc iates
FMC


Kawasaki/
Kureha
Kawasaki/
Kureha
Kawasaki/
Kureha
FMC

Envirotech

CEA/ADL

FMC

Size
(Mw, Equivalent)
3


20


45-35


450

450

(2-450)
900
100

575

280

192

Active Calciur.i
Alkali Sources
Cone. Lime


Cone. Lime


Cor,.:. Li '.Tie


Cone . Linest oni-

Cone. Li-estoni;

Ccnc. Li.-estone

Cone. Lime

Cone. Line

Cone. Lime

CMC. Line

Start-up
Da t ea
Jar.. 19T5


Feb. 1975


•Jc t . 1 ? 7 3


1975

1975

(May 1977)

(1977)

(1977-78)

(1977-73,)

(1977-78)

Coal-fired
industrial boiler
                          FMC
                                                100
                                                                         Cone .
                                                                                       Line
                                                                                                            (1978)
  Dates in parentheses are projected start-up dates
  These units were proposed in response to the RFP for a full scale utility demonstration

-------
BIBLIOGRAPHY

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.  LaMantia, C., et al., "EPA-ADL Dual Alkali Program Interim
    Results."  Presented at EPA Symposium on Flue (las Desulfurlzation,
    Atlanta, Georgia, November 4-7, 1974.

4.  Johns tone, II. 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 S()2 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 Internation.il 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.
                                  420

-------
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.   liJ.li.fon, W. , et al., "System Reliability and Environmental Impact
     of S02 Scrubbing Processes."  Presented at Coal and The Environ-
     ineiit, Technical Conference, Louisville, Kentucky, October 22-24,
     1974.

1'3.   Rochelle, G. T., "Economics of Flue Gas Desu.l fur Iza Lion."
     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  F.llyn,  Illinois 60137.

15.   McClamery, G.  G. and  R. L. Torstrick,  "Cost  Comparisons of  Flue
     Gas Desulfurization Systems."  Presented at  the  EPA Symposium on
     Flue Gas  Desulfurination, Atlanta,  Georgia,  November 4-7, 1974.

16.   Dingo,  T.  and  E. Piasecki, "General  Motor's  Operating  Experience
     witii a  Full-Scale  Double Alkali Process."   Presented at  the EPA
     Symposium on  Flue  Gas  Desulfurizntion, Atlanta,  Georgia,
     November  4-7,  1974.

17.   LaMantia,  C.R., R.R.  Lunt, R.E. Rush,  T. Frank,  N.  Kaplan,
     "Operating Experience  — CEA/ADL  Dual Alkali Prototype
     System  at  Gulf  Power/Southern  Services,  Inc."   Presented  at
     the EPA Symposium  on  Flue Gas  Desulf urization,  New Orleans,
     Louisiana, March 8-11, 1976.

18.  Ando, J.,  "Status  of  Flue Gas  Desulfurization and Simultaneous
     Removal of  S02 and NOX in  Japan."  Presented at the EPA
     Symposium on  Flue  Gas Desulfurization, New Orleans,  Louisiana,
     March 8-11, 1976.

19.  Summary Report. Flue Gas  Desulfurization  Systems.   Nov.  -  Dec.  1975
     Prepared  by PEDCo  Environmental  Specialists for U.S.  EPA under
     Contract  No.  68-02-1321.
                                421

-------
     OPERATING  EXPERIENCE--CEA/ADL DUAL ALKALI PROTOTYPE SYSTEM
                AT GULF  POWER/SOUTHERN SERVICES,  INC.
               Charles  R.  LaMantia and Richard R.  Lunt

                       Arthur D.  Little,  Inc.
                     Cambridge,  Massachusetts
                           Randall E.  Rush

                       Southern  Services,  Inc.
                         Birmingham, Alabama
                           Thomas  M.  Frank

               Combustion  Equipment  Associates,  Inc.
                        New York,  New  York
                           Norman  Kaplan

           Industrial Environmental  Research  Laboratory
                  Environmental  Protection Agency
          Research Triangle Park,  North Carolina    27711
ABSTRACT

     As part of the Dual Alkali Program being  conducted by Arthur D.
Little, Inc. for EPA's Industrial Environmental Research Laboratory-RTP,
a one-year test is being conducted on the 20-megawatt dual alkali SO
.control process at Gulf Power Company's Scholz Steam Plant in Sneads,
Florida.  The system was developed, designed and  installed by Combustion
Equipment Associates, Inc./Arthur D. Little, Inc. for Southern Services,
Inc.  Initial startup of the system occurred in February 1975; after
the shakedown period, the EPA test program commenced in mid-May, 1975.

     This paper presents a description of the  system and its performance
during the first year of operation from the initial startup through
early January 1976, when the boiler was shut down for a scheduled overhaul,
The boiler and the control system are due to be put back in operation
by early March, 1976.

                                 423

-------
                          ACKNOWLEDGEMENTS
The authors would like to acknowledge several  individuals  for their
invaluable contributions to the test program for  the  prototype dual
alkali system.

We greatly appreciate the efforts of Indrakumar Jashnani,  Bernard
Jackson, and James Valentine of ADL who helped coordinate  and supervise
the test program at Scholz.  Steve Spellenberg and  Bruce Goodwin of
ADL provided the bulk of the on-site chemical  analytical support.

The cooperation of Combustion Equipment Associates, Gulf Power Company,
and Southern Services, Inc. in this prototype  test  program is also
gratefully acknowledged.  In particular, we  would like  to  thank
Richard White of Combustion Equipment Associates, Reed  Edwards of
Southern Services, and James Kelly of Gulf Power  for  their on-site
assistance in the operation of the system.   Their continuing concern
and attention to the various aspects of the  test  program and the
system operation have been valuable contributions to  the success of
our overall efforts.
                              424

-------
          OPERATING  EXPERIENCE--CEA/ADL DUAL ALKALI PROTOTYPE SYSTEM
                  AT  GULF POWER/SOUTHERN SERVICES, INC.


                               I.   SUMMARY

During the first  year  of operation after startup, the prototype dual
alkali system operated at an S02 concentration below the design range and
at oxygen concentrations which exceeded the design range until the last
three  to four months of operation.  Over the course of the year, the
process and equipment  performance improved significantly with excellent
system availability  during the final operating period from mid-September
1975 through early January 1976.  The process has demonstrated very good
reliability, stability of operation, and resistance to upset with no
incidents of scrubber  or mist eliminator plugging or scaling, even during
the last three months  of operation with no mist eliminator wash.

Under these conditions, the system operation relative to important
performance parameters was as follows:

     •  S02 Removal—The system generally operated at S02 removal
        efficiencies of about 95% and demonstrated the capability for
        removal of over 99%.  The amount of S02 removed is controllable
        by adjustment  of scrubber pH.

     •  Oxidation and  Sulfate Formation and Control—By co-precipitation
        of sulfate with sulfite,  this concentrated dual alkali mode can
        keep up with oxidation rates of up to 25-30% in closed-loop
        operation.

     •  Lime Utilization—Lime utilization typically ranged  from 95% to
        100%.  The overall lime  stoichiometry was  0.95-1.00  moles Ca(OH)2/
        mole of S02  removed from the flue gas.

     •  Waste Solids Properties—The system produced a  washed  filter cake
        generally containing  at  least 50% insoluble solids in  this  low to
        medium sulfur coal/high  oxidation situation.  The waste material,
        on  the average,  contained 3-5%  soluble solids  (dry cake basis,
        Period 2) with  soluble  solids reduced to  the 2-3% range when
        washing was well  controlled.  Under process conditions consistent
        with higher sulfur  coal  operation, the insoluble solids content
        of  the cake improved  markedly with improved washability and lower
        soluble solids  content.

      •   Sodium Makeup Requirements—About 0.03 moles Na2C03  are required
        per mole  of S02  removed  as makeup  for soluble  solids losses in
         the washed  cake  under these process  conditions.  About an equal
         amount of sodium makeup  was unaccounted  for, probably  due to
         leakage  from  the  system and errors  in soda ash  makeup  estimates.
         An attempt will  be  made  to reduce  leakage by maintenance performed
         during the  shutdown period in  January 1976.

  fmerally,  the system performance has  exceeded expectations  for these  flue
   as conditions.

                                     425

-------
It is planned to put the system In operation again by early March 1976
and continue operations through about June 1976.  During  this  next
operating period, high sulfur coal (3.5-4.5%) will be in  use at  Scholz,
enabling evaluation of the system performance in a high sulfur coal
application.
                                  426

-------
                          II.   INTRODUCTION
The 20-megawatt, prototype dual alkali process was designed and installed
by Combustion Equipment Associates, Inc.  (CEA) and Arthur D. Little, Inc.
(ADL) at the Scholz Steam Plant of Gulf Power Company near Sneads, Florida.
This system is one of three "second generation" 20-megawatt, prototype
flue gas desulfurization systems installed at the plant as part of a
technology evaluation program being conducted by Southern Services, Inc.
for The Southern Company (an electric utility holding company  including
Alabama Power Company, Georgia Power Company, Gulf Power Company, Mis-
sissippi Power Company, Southern Electric Generating Company and Southern
Services, Inc.).

rhe process was developed and designed jointly by CEA/ADL.  Early labo-
ratory research on the process, performed by ADL and sponsored by the
Illinois Institute for Environmental Quality, dealt exclusively with
characterizing  the nature of the regeneration reaction.  Based upon the
laboratory results, a 2,000 cfm* dual  alkali pilot  plant was conducted
at ADL's facilities by CEA/ADL and an  eight-month test  program was con-
ducted to generate the design data for  the prototype system.   The pilot
system contains the complete dual alkali  process  loop  involving:  gas
scrubbing; absorbent regeneration; and  solids  separation.   Results of
the  laboratory  program and pilot operations  for  generation  of  the proj-
totype system design have been reported previously  in  the  literature.

The  laboratory  and pilot plant investigation of  dual alkali technology
has  continued at ADL in a program  for  the U.S.  Environmental Protection
Agency's  (EPA)  Industrial Environmental Research Laboratory at Research
Triangle Park,  N. C.  The program  involves characterization of the basic
process chemistry and the various  modes of operation of sodium-based
dual alkali  processes.  The work covers a wide range of flue gas condi-
tions,  liquid reactant  concentrations,  and process  configurations
including the use of both  lime  and limestone for regeneration  of the
sodium scrubbing  liquor.   The  three  tasks of this program are  described
briefly below.   Interim results  were presented at the  1974 EPA Flue
Gas  Desulfurization  Symposium.

In Task I,  the  ADL Laboratory  Program, experiments were performed on the
regeneration of concentrated  sodium scrubbing solutions using lime re-
generation,  limestone regeneration and sulfuric acid treatment for
sulfate removal.   Work was performed at EPA's laboratories at  Research
Triangle  Park on regeneration using limestone in dilute mode operation.


 *  Despite  EPA policy,  certain non-metric units are used in the paper;
    however   a conversion table is provided at the end of the paper.
                                   427

-------
Results of this prior work have been reported.2  More recent work  still
in progress includes a characterization of the chemical, physical,  and
crystallographic properties of dual alkali solids, including work  on
the nature of calcium sulfate species co-precipitated with calcium
sulfite in dual alkali processing.  Work on physical properties includes
evaluation of the mechanical and engineering properties of untreated
dual alkali solids and solids produced from a limited number of mixtures
of dual alkali solids with fly ash, lime and other agents used in  the
treatment of solid wastes from conventional lime/limestone flue gas
desulfurization processes.  This work is scheduled to be completed  and
reported in mid-1976.

Task II, the Pilot Plant Program, was conducted at the CEA/ADL pilot
facility, Cambridge, Massachusetts.  Results of short-term pilot opera-
tions  using concentrated sodium scrubbing solutions with lime regenera-
tion,  limestone regeneration and sulfuric acid treatment for sulfate
precipitation were reported previously.2  Since then, short-term opera-
tions  have also been conducted using dilute sodium (Na+ active) scrubbing
solutions with lime regeneration; and additional pilot operations have
been conducted in an attempt to develop a successful dual alkali system
using  limestone for regeneration.

From the short-term pilot operations, three of the most promising modes
of dual alkali processes were selected for long-term (three to five
weeks  around the clock), closed-loop pilot yuns.  These long-term tests
were conducted in the following dual alkali modes:

     •  Dilute mode using lime for regeneration
     •  Concentrated mode using lime for regeneration
     •  Concentrated mode using limestone for regeneration

Work on the dilute and concentrated modes using lime has been completed.
The results indicate that these modes can be operated closed loop with
the following general performance:

     •  SQ2 removal - 90% or greater
     •  Lime utilization - 95% or greater
     •  Waste cake solids content - 45% or greater
     •  Sodium makeup requirements - less than 0.05 moles Na2C03
         per mole of SQz removed

The actual performance of the dual alkali process will vary depending
upon the SQz and oxygen concentrations in the flue gas,  the design  of
the system and the concentration of sodium solutions used in the process.
Some version of the dual alkali process can generally be designed to
exceed many or all of the above performance characteristics in most
utility applications.
                            428

-------
Aside from concentration differences, the principal difference in
operating characteristics between dilute and concentrated lime dual
alkali systems is that the dilute systems operate at or near saturation
in calcium sulfate, which requires the use of carbonate makeup to pro-
vide some softening of the regenerated solution prior to recycle back
to the scrubber.

Operations using limestone for regeneration have not been completed.
Recent results have been promising, with many of the performance charac-
teristics approaching those of lime dual alkali systems.  The problem
which remains to be solved is the production of solids with good settling
characteristics over a wide range of sulfate, magnesium and iron concen-
trations in the scrubbing liquor.

The results of these long-term runs and recent work on limestone regen-
eration will be presented in' a report scheduled for mid-1976.

Task III, the Prototype Test Program, consists of a one-year test of
the 20-megawatt CEA/ADL prototype dual alkali system at Gulf Power.
The objective of the program is to characterize the important aspects
of the prototype process performance:

         S02 removal efficiency
         Oxidation and sulfate formation and control
         Lime utilization
         Waste solids properties
         Sodium makeup requirements and degree of closed-loop operation
         Process reliability

The prototype system was completed and put in operation by CEA/ADL in
early February 1975.  In mid-May 1975, the test program was initiated
by ADL, Southern Services and CEA as part of the EPA/ADL Dual Alkali
Program.  The system was operated until early January 1976 when the
boiler was shut down for a scheduled overhaul.  The boiler and the
prototype system are due to be put back in operation by early March 1976.

This paper describes the performance of the system over this first year
of operation from startup in February 1975 through early January 1976.
The startup period, including the first three months of operation prior
to the EPA program, has already been reported in detail in the litera-
ture3; for completeness in describing the first year of operation,
information'on the first three months of operation is also included.
                                  429

-------
            III.   CEA/ADL 20-MEGAWATT DUAL ALKALI SYSTEM
A.   PROCESS CHEMISTRY

The CEA/ADL dual alkali S02 control process at Scholz Station is a
sodium solution scrubbing system in which the absorbent solution is
regenerated using lime with the provision for use of limestone.  The
use of limestone is contingent upon successful pilot plant tests of a
limestone mode.

The absorption of S02 is accomplished using a solution of sodium sulfite,
sodium hydroxide and possibly some sodium carbonate (makeup) producing
a spent sodium sulfite/bisulfite liquor:
 2NaOH + S02 -»• Na2S03 + H20

Na2C03 + S02 •*• Na2S03 + C02t
Na2S03 + S02 + H20
                     2NaHS0
                                                                     (1)

                                                                     (2)

                                                                     (3)
During absorption, and to a lesser extent through the remainder of the
system, some sulfite is oxidized to sulfate:
Na2S03
         1/2 0
                                                                    (4)
converting an "active" form of sodium to an "inactive" form.  Oxidation
in the scrubber is generally a function of the scrubber design, oxygen
content of the flue gas and the scrubber operating temperature.  At
excess air levels normally encountered in utility power plant opera-
tions burning medium or high sulfur coal, the level of oxidation is
expected to be on the order of 5-10% of the sulfur dioxide removed.

The scrubber solution is regenerated by reaction with lime (and/or
limestone) which precipitates a mixture of calcium sulfite and calcium
sulfate solids for disposal, as shown by the following overall reactions
(shown for lime):

            2NaHS03 + Ca(OH)2 -»• Na2S03 + CaS03 -1/2 H204- + 3/2 H20  (5)

             Na2S03 + Ca(OH)2 + 1/2 H20 + 2NaOH -I- CaS03-l/2 H2G+    (6)

                      Na2SOlt + Ca(OH)2 •* 2NaOH + CaSO^I             (7)
                                  430

-------
After regeneration, the solids are separated from the regenerated liquor
and washed.  The clear liquor, containing very low amounts of suspended
and dissolved calcium, is returned to the scrubbers.  Soluble calcium
levels are generally less than 100 ppm in concentrated dual alkali
processes.  Any sodium value lost with the washed waste solids is
replaced by the addition of sodium carbonate to the regenerated liquor;
sodium hydroxide can be used since carbonate softening is not required.
Since sodium sulfate is also reacted with lime in this system to regen-
erate sodium hydroxide, it should be possible to use sodium sulfate as
the sodium make-up source when low to moderate oxidation levels are
encountered.

The system is designed to operate in the concentrated active sodium
mode (active Na+ concentration greater than 0.15M).  In this mode,
sulfate removal cannot be accomplished by the precipitation of gypsum
'(CaSOit -2H20), since the high sulfite levels prevent the soluble
calcium concentration from reaching that required to exceed the gypsum
solubility product.  However, calcium sulfate  (CaSO^) is precipitated
along with calcium sulfite (CaS03-l/2 ^0) in the regeneration reactor,
resulting in a solid solution of the two salts.  The amount of sulfate
precipitated in this form is a function of the concentrations of species
in solution and the reactor pH.  Under normal operation, with sulfate
levels up to 1.5M SOi^, the system can keep up with sulfite oxidation
rates equivalent to 25-30% of the S02 absorbed without becoming satu-
rated in calcium sulfate.

Additional details of dual alkali chemistry and terminology are given
in recent publications.2 >*+

B.   SYSTEM DESCRIPTION

The prototype system at the Scholz Steam Plant is installed on Boiler
No. 1, a 40-megawatt  (nominal) Babcock and Wilcox pulverized coal-
fired boiler.  The prototype is sized to handle 50% of the flue gas
from the boiler.  The boiler has been retrofitted with a sectionalized,
high efficiency electrostatic precipitator capable of 99.5% particulate
removal.  Sections of the electrostatic precipitator can be de-energized
to study the impact of particulate input on process operation.

The design basis for  the prototype system  is given in Table 1; a
schematic flow diagram is given in Figure  1.   The system, designed as
a prototype, incorporates a high degree of flexibility aimed at
generating design and operating information for a wide variety of
applications throughout The Southern Company system.  Although the
basic mode of operation of the system is a dual alkali process with
lime regeneration, the system was designed to  accommodate limestone
regeneration and a combination of regeneration with limestone and lime.
The system was also designed to enable operation as a direct limestone
or lime scrubbing system.  As a consequence, the system contains equipment
and piping in excess  of that required to operate a dual alkali system on
a boiler already equipped with adequate particulate control.


                                431

-------
                               TABLE 1
                             DESIGN BASIS
Flue Gas Inlet

   Flow Rate (acfm)

   Temperature (°F)

   02 Concentration (% volume, dry)

   Particulate Loading (gr/scf, dry)

   S(>2 Concentration (ppm, dry)
75,000

   275

   6.5 (max.)

   0.02 (precipitator
          energized)
1800-3800
Design Performance

   S02 Removal (% reduction)

   Maximum S02 Removal Rate (Ib/hr)

   Particulate (gr/scf, dry)

   Power Consumption (% power output)

     with venturi, full spray absorber pump
        requirements

     without venturi, with tray absorber pump
        requirements
  90 (min.)

1530

  0.02 (no increase
         in loading)
  2.5-3.0
  1.0-1.5
                                 432

-------
                    Scrubbed
                      Gas
                                     Vacuum Water Recovery. (Filter
                                                                    Na2CO3
                                                                      Silo
                                                                    Solids
Figure 1,  CEA/ADL dual alkali process flow diagram - Scholz Station.

-------
The venturi scrubber  is included in the system to investigate the opera-
tion of that  type  of  scrubber in a dual alkali mode (or for direct lime
or limestone  scrubbing) when simultaneous particulate and S02 removal
is required.  The  second scrubber, an absorption tower, can be operated
as a tray scrubber, as in a dual alkali system where simultaneous
particulate control is not required; or as a spray tower, for direct
lime or limestone  scrubbing, again without simultaneous particulate
control.  In  addition to this redundancy in scrubbers, sufficient pump
capacity is provided  to operate the venturi at an L/G of 25 gallons/
Macf of saturated  gas and for an absorber L/G of 60, for operating in
a spray tower configuration.

An additional storage silo (for limestone), a mix tank and other assorted
tanks, pumps, controllers and piping were included in the system to
accommodate the high  degree of flexibility desired in the prototype
system.  To date,  except when limestone has been erroneously delivered
to the system instead of lime, the system has been operated only in
the lime dual alkali  mode as given in Figure 1.

In the lime dual alkali mode, the system design is based upon removal
of at least 90% of the S02 in the flue gas for medium to high sulfur
fuels (up to  5% sulfur) encountered in The Southern Company system.
With the precipitator energized, the system was  specified not to
increase particulate  loadings in the scrubber outlet above those in
the inlet flue gas.

The power consumed by the system (not including  reheat oil) is equiva-
lent to 2.5-3.0% of the power generated by the unit in producing the
flue gas load to the  system,  with the system operating at full fan
capacity (full gas flow at a system pressure drop of 20 inches water)
and at full venturi and absorber liquid recirculation capacity.   Cor-
recting for the excess fan and pump capacity,  the power consumed by
the equipment actually required in this application (tray tower at an
L/G of 5-10)  is roughly 1.0-1.5% of the power  generated at  the design
conditions.   In a  full scale system designed for S02 removal only,
the power consumption should be in this range.

The dual alkali system can be conveniently broken down into three
process subsystems:  gas scrubbing; absorbent  regeneration; and solids
dewatering.  The design and operation of each  of these subsystems is
discussed in  the following sections.

1.   Flue Gas Scrubbing

The gas scrubbing  system consists of  a variable  throat,  plumb-bob type
venturi scrubber followed in  series by an absorption tower.   Treated
flue gas flows through both scrubbers.   Each of  these scrubbers is
equipped with a removable liquid entrainment separator,  an  enclosed
recycle tank  to contain the scrubbing liquor,  and two recycle pumps
(one operating and one spare).   The venturi  can  be  used for absorption
                                 434

-------
and/or particulate control and can be operated on a separate liquor loop
from the absorption tower or in series with the absorber liquor loop.
The absorption tower can be operated as a tray tower  (with up to four
trays), as a. spray tower, or as a de-entrainment separator.

Gas is pulled from the exit of the electrostatic precipitator and
forced through the scrubbing circuit using the booster fan provided
with the dual alkali system.  The fan and motor have been designed
fop a total system pressure drop of 20 inches H20 at maximum flow.
Under normal conditions with the precipitator in service, the venturi
was used only for gas saturation with some S02 removal; a pressure drop
of roughly 5 to 10 inches of water was maintained across the venturi.

Gas from the fan enters the venturi scrubber, flows downward over the
wetted approach section and enters the high velocity venturi throat.
Recycled scrubbing liquor also enters the top of the scrubber through
tangential pipe inlets and through a number of vertical bull nozzles
equally spaced around the center of the venturi.

After passing through the throat, the flue gas and scrubbing liquor
continue downward through the internal downcomer; the  liquor is
collected at the bottom of the scrubber in t ie ->.nternal recycle tank.
The flue gas makes a 180 degree upward turn and contacts the chevron-
type entrainment separator  (when used).  The entrainment separator is
equipped with wash sprays above and below which can be operated con-
tinually or on a sequential timing cycle.  Either scrubbing liquor or
fresh water can be used for washing.  During dual alkali operations,
this entrainment separator system was completely removed.

Following the venturi scrubber, the saturated flue gas enters the
bottom of the absorption tower.  This tower has been  designed to
operate either as a spray tower, with one or two sets  of sprays, or
as a tray tower, with from one to four trays.  The bottom tray  is
equipped with a spray underneath to wet the bottom side of  the  tray.
Gas passes upward through the trays and then through  a final de-
entrainment separator, with a wash system similar to  the venturi
demister wash.  The wash system was used initially, but found to be
unnecessary and its use discontinued.

The clean flue gas leaving  the tower is finally reheated by the
injection of hot gas from a burner fired with No. 2 fuel oil before
being discharged through the stack on top of the absorber.  Oil-fired
reheat was specified by Southern Services for the prototype systems
in order to reserve steam for power generation.

Regenerated absorbent solution, containing  sodium hydroxide,  sodium
sulfite, sodium sulfate and some sodium carbonate,  is fed  to  the  top
 tray of the absorber.  The  solution flows countercurrent  to  the gas
 through the tray system  (two trays) and is  collected  at  the bottom
 of the absorber in the internal recycle  tank.  This  collected liquor
 supplies solution both for  spraying the bottom  tray  and  for recirculation


                                  435

-------
to the top tray, as needed for pH control in the scrubber or to maintain
liquor flow across the trays.  The pH in the combined stream of recircu-
lated liquor and freshly regenerated feed ranged from about 6 to 11.

A bleed from the collected tray tower bottoms (pH roughly 5-7) is sent
forward to the venturi recirculation loop for additional S02 removal.
In the venturi, the gases with the highest S02 concentration are con-
tacted with the most acidic liquor.  A continuous bleed stream is
drawn from the venturi recirculation loop and is sent to the absorbent
regeneration system.  The pH of the bleed stream normally ranged from
4.8 to under 6.0.

2.   Absorbent Regeneration

Spent scrubber solution from the venturi recirculation line is bled
to the regeneration reactor where it is reacted with hydrated lime.
A full-scale system would normally incorporate a lime slaker and a
slaked lime slurry tank; hydrated lime was used in this smaller
installation for simplicity.  The normal mode of operation was to feed
dry lime directly to the reactor; however, provisions have been made
to feed slurried hydrated lime as well as either slurried or dry
limestone.  The rate of the lime feed will in most cases be set accord-
ing to the boiler load and coal sulfur content and adjusted on the pH
of the reactor effluent liquor.

The lime neutralizes the bisulfite acidity in the scrubber bleed and
further reacts with sodium sulfite and sodium sulfate to produce
sodium hydroxide.  These reactions precipitate mixed calcium sulfite
and sulfate solids resulting in a slurry containing up to 5 wt %
insoluble solids.  The regeneration was normally carried out to a pH
of between 11.0 and 12.5.

The regeneration reactor system consists of a multistage design,
patented by CEA/ADL.   The system at Scholz consists of two reactor
stages in series:  a short residence time reactor followed in series
by a longer residence time reactor,  designed for reactant residence
times of 5 minutes and 35 minutes respectively at the design S02
removal rate (185 gpm flow rate through the reactor system).  This
reactor system design has been shown to generate clusters of sulfite/
sulfate crystals which are generally spherical rather than the needle-
like or platelet crystals generally associated with calcium sulfite
precipitation.   These crystal clusters have good settling,  filtration
and washing properties and can be generated in the reactor system
over a wide range of  flue gas and process conditions.
                                  436

-------
3.    Solid/Liquid Separation and Solids Dewatering

Slurry from the regeneration reactor system is fed to the center well
of  the slurry thickener.  The 40 foot diameter thickener is sized to
handle the solids produced from the treatment of flue gases from the
full 40 megawatts of Boiler No. 1.

The thickened slurry from the bottom of the settler  is  sent to a rotary
drum vacuum filter.  The solids content of the underflow was maintained
below 30% to ease underflow pumping.  The slurry is  recirculated past
the filter in a recycle loop that returns the slurry to the solids  zone
in the settler.  The feed to the filter is drawn as  the bleed from  this
recirculation loop.  The filter surface area is 75 ft2.

The solids cake is washed on the filter using two or three water sprays
arranged in series.  This wash removes a large fraction (up to 90%)  of
the occluded soluble salts from the cake and returns these salts to
the system, thereby reducing sodium losses and minimizing sodium car-
bonate makeup.

Solids from the filter  are retained in a solids storage enclosure  under
the filter from which they are loaded  into dump trucks  for transfer to
the disposal area.  The mixed  filtrate and wash liquor  from  the  filter
are returned to the thickener.

Makeup sodium carbonate may be fed  to  the  thickener  in  order  to  allow
easy removal of any CaC03 precipitated.  This carbonate is not  intended
for use as a softener,  since soluble calcium concentrations  in  the
regenerated liquor generally will run  less  that  100  ppm.  However,
 some CaCOs precipitation will  occur due  to  its very  low solubility
 limits.  The amount of  CaCOs formed will be  very  small, a  few hundred
 Bounds per day at  design conditions, which  is  less  than 0.5%  of  the
 total cake produced.

 Clear liquor overflow from  the thickener  is  collected  in the thickener
 told tank which acts as surge  capacity for  the  absorbent liquor feed
 to the scrubber system. Water can  also  be  added to  this hold tank to
 lake up  for  the difference  between  total  system water  losses (evapo-
 ration and cake moisture) and  total water  inputs  from other  sources
 {sodium  makeup solution, pump  seals,  lime  feed,  cake wash,  and
 demister wash).

 !he disposal  area  for the dual alkali  waste cake is a one-acre pit
 (450 ft  x  100  ft)  approximately 12  ft  deep.   The bottom and  sides of the
 fit are  lined  with a layer  of  clay  covered by a double thickness of poly-
 ethylene liner.   The polyethylene is  reinforced between the sheets with
 tmesh  of  nylon  fibre.   The floor of  the pit slopes to a single collec-
 tion drain constructed  of  PVC  from which leachate is discharged to the
 teh pond.   On  top  of  the  polyethylene liner is a two-inch layer of sand
 Aich  gradually  changes to  gravel near the drain.
                                    437

-------
                      IV.   OPERATING HISTORY
A.   GENERAL OPERATING CONDITIONS

The prototype  system commenced process startup on February 3, 1975 and
was shut down  on January 2, 1976 when the boiler was taken off the
line for a scheduled overhaul.  During this time the system was opera-
ted approximately 4,700 hours with shutdown periods of varying length
for system maintenance and modifications and during periods when the
boiler was'taken off the line.  Throughout the eleven months, the
system treated flue gas with lower sulfur dioxide concentrations and
higher oxygen  concentrations than the range for which the system was
designed.  This represented a difficult test for the prototype system.

In sodium-based dual alkali systems for a given scrubber design and
soluble solids level, the rate of sulfite oxidation (moles per unit
time) is a strong function of the oxygen concentration in the flue gas
and is relatively independent of the rate of sulfur dioxide removal
(moles per unit time).  As oxygen concentrations increase and SC>2 con-
centrations decrease, a higher percentage of the S02 removed from the
flue gas is converted to sodium sulfate rather than sodium sulfite/
bisulfite in the scrubber liquor.  This higher percentage oxidation
requires an increase in precipitation of sulfate relative to sulfite
and a higher calcium sulfate content in the precipitated calcium
sulfite/sulfate to enable closed-loop operation with no  intentional
purging of sodium sulfate.

Over the eleven-month period,  sulfur dioxide concentrations in the flue
gas averaged between 1,100 and 1,200 ppm,  compared  with  a minimum design
concentration  of 1,800 ppm.   The actual concentrations varied over a
range of from  600 to 1,600 ppm,  often fluctuating daily  and even hourly
over the entire range of concentrations.   Oxygen concentrations  in the
flue gas entering the scrubber ranged from 5.0% to  10.5% (by volume
equivalent to  30-100Z excess air) with oxygen  levels  increasing  with
decreasing boiler load (higher excess air).  During the  first half of
the year, air  leaks in the  boiler combustion air preheater and coal
feed tubes contributed from 0.5  to 2.0 volume  %  to  the oxygen concentra-
tions.  Oxygen levels in the flue gas were  reduced  from  the 7-10% range
down to the 5-7% range in September after  air  preheater  repairs  and when
burner box pressure was brought  under better control.

This general change in flue  gas  oxygen concentration and changes in
the system operation made at about the same time logically led to the
division of this first year  of operation into  two discrete operating
periods described in the next  section.
                                 438

-------
B.  DESCRIPTION OF OPERATING PERIODS

1.  Operating Period 1—Startup and Initial Operations

This first operating period extends from February 3, 1975 through
July 18, 1975, when the system was shut down for a two-month period
for modifications, repairs and to await receipt of replacement parts.
During this operating period, commencing on initial process startup,
the system was operated approximately 2,600 hours, 83% of the time
that the boiler was in operation.  Almost  the entire system down-time
of 518 hours relative to the boiler is accounted for by a three-week
shutdown (491 hours) in mid-April, after the start-up period, for
necessary maintenance and'equipment modifications and adjustments prior
to the start of the formal EPA test program in mid-May.  Equipment
changes are described in Section VB.

Table 2 contains a summary of the fuel and flue gas characteristics
encountered during this operating period.  The boiler was fired with
a combination of several medium and low sulfur coals ranging in sulfur
content from 0.9 to 2.2 wt %.  The weighted average sulfur content of
the coal as burned was 1.6 wt %, producing an average S02 level in the
flue gas of about 1,050 ppm.  Since it was not possible to segregate
and selectively fire the different coals,  the S02 levels in the flue
gas fluctuated daily and often hourly from a low of 600 ppm to a high
of 1,550 ppm.

The flue gas load to the prototype system  was kept within a range
equivalent to 15-22 Mw.  However, the boiler load varied from 15 mega-
watts to 40 megawatts with an attendant variation in excess air rates.

In addition to the unfavorable process  conditions resulting from low
sulfur  coal and the high oxygen concentrations,  there were frequent
process upsets.  These upsets included  occasional carry-over of fly
ash with the flue gas and the inadvertent  contamination of the lime
supply with limestone  (caused by a mix-up  in the  lime and limestone
deliveries to the prototype  systems at  the plant).  While the presence
of fly  ash in the system caused little  or  no effect on operation or
performance, the  limestone did produce  temporary  changes in the process
chemistry—particularly when a significant fraction of the calcium  feed
to the  reactor was limestone.

The limestone usually appeared mixed with  lime  at  levels of up to  50%
of the  total feed; however,  on a number of occasions  (for periods  lasting
up to one day) the feed  to  the reactor  system was  pure  limestone.   During
periods when only limestone  was fed  to  the reactor,  the pH  in  the  reactor
system  dropped, SO* removal  decreased  somewhat  and waste cake  properties
deteriorated  slightly.   Normal process  conditions  were re-established
after the limestone passed  through  the  system.
                                   439

-------
                              TABLE 2




                 SUMMARY OF SYSTEM INLET CONDITIONS




                  Operating Period 1 (2/75 - 7/75)










                                         Range         Average




Coal Fired:




     Sulfur (wt  %)                     0.9-2.2            1.6




     HHV (Btu/lb coal)               11,500-13,000        12,000



     Chloride  (wt  %)                  0.05-0.10           0.08









Inlet Gas:




     Gas Load Treated (Mw equiv)         15-22             17




     S02 Level (pptn-dry basis)          600-1550          1050




     02 Level  (Z dry vol.)              5.0-10.5           7.5
                                440

-------
There was a general rise in the flue gas oxygen concentration over the
operating period due to worsening leakage in the boiler air preheater
section.  By mid-July oxygen concentrations were in the 8-10% range,
with dilution of the already low S02 levels down to about 850-950 ppm.
At the resultant high oxidation rates  (30-50% of S02 removal), active
sodium levels were inadvertently allowed to drop below 0.15M, into the
range of dilute mode dual alkali operation.  Under these conditions,
soluble calcium levels in the regeneration system rose and the regen-
erated liquor became saturated in calcium sulfate, resulting in some
gypsum scaling of the piping at the outlet of the regeneration reactor.
At this same time, mechanical problems  in the scrubber system (failure
of two control valves and a pin hole leak in the absorber recycle tank)
required that the system be shut down  for repairs.  After shutdown it
was decided to await the repair of the  preheater leak rather than attempt
to continue operating at the low SC>2 concentrations and increasing oxygen
concentrations.

Based upon this experience, it was decided that in future operations
the active sodium concentration would be maintained well above the 0.15M
level by operating with a slightly higher total sodium concentration in
the system.  This would allow the sodium sulfate concentration to increase
in the system during periods of increasing rates of oxidation.  Sulfate
levels would continue to rise relative  to the sulfite until the rate of
sulfate precipitation as a calcium salt equaled the rate of oxidation.
This can be accomplished without reverting to dilute mode operation at
oxidation rates up to about 25-30% of  the S02 removal rate.

The system was shut down from mid-July  until mid-September.  Repairs
and revisions made during this period were of a mechanical nature rather
than involving process changes and are  discussed later.  About two
thirds of the shutdown period was to await replacement parts for the
valves that had failed.

2.   Operating Period 2—Low to Medium  Sulfur Coal Operation

The system was put back in operation on September 16, 1975.  From mid-
September to mid-October repairs were made to the air preheater during
boiler outages and adjustments were made in the boiler operation, re-
ducing flue gas oxygen levels down to  the 5-6% range.  For the remainder
of the test period through January 2,  1976, oxygen concentrations were
generally kept in the 5-6% range.  As  shown in Table 3, S02 levels were
slightly higher than those encountered  during the first operating period,
with an average level of about 1,200 ppm.  Similarly, the gas load to
the system was in the same range as that for Period 1.  Active sodium
concentrations were maintained in a range for concentrated mode opera-
tion and continuing improvements were made in the mechanical performance
of system components, particularly the  filter.  One unprogrammed period
af regeneration with limestone occurred in this interval.
                                    441

-------
                              TABLE  3
                  SUMMARY OF  SYSTEM INLET CONDITIONS
                    Operating Period  2  (9/75-12/75)
Coal Fired:
     Sulfur (wt  %)
     HHV (Btu/lb coal)
     Chloride (wt  %)
    Range

   1.5-3.1
11,900-14,100
  0.02-0.14
Average

   2.1
  13,000
   0.08
Inlet Gas:
     Gas Load Treated (Mw equiv)
     SO2 Level (ppm-dry basis)
     02 Level (% dry vol.)
16-19
800-1700
4.5-9.5
18
1220
6.0
                                442

-------
During this period, until shutdown on January  2, 1976, the prototype
system operated approximately 2,100 hours, 97% of  the time the boiler
was in operation.

C.   AVAILABILITY—FIRST YEAR PROTOTYPE  SYSTEM OPERATION

The objective of the installation of the prototype system was to test
the process chemistry and design on this relatively small scale in
order to evaluate the viability of the process technology.  While the
reliability of the process was a principal concern,  the 20-megawatt
system was not intended to be a demonstration  unit nor a test of the
ultimate availability of such systems when applied full scale.

The system design was based upon scale-up by a factor of about 40
from the CEA/ADL dual alkali pilot plant.  Although the prototype
system contains spare pumps in the scrubber  and  regeneration areas,
other key elements of the system such as the reactor system, lime feed
system and filter station were not spared in this  prototype application.
Hultiple installations and/or spare capacity for these critical items
would normally be incorporated in full-scale applications.  In spite
of these considerations and the difficult flue gas conditions encountered
during the first operating period, the availability of this prototype
unit from initial startup through the first  year of operation is impres-
sive.

Boiler and prototype system operating hours  are  graphically displayed
by month in Figure 2 and summarized by operating period  in Table 4.
The availability during the first operating  period, 83%, includes the
initial startup of the system and the three-week shutdown period for
system adjustments and modifications prior  to  the  start  of the EPA
test program.  Such adjustments would normally be  expected after process
startup.  During the second operating period,  system availability exceeded
97%.

During the two months between operating  periods, the system was down
for maintenance and awaiting receipt of  spare  parts and  repair of
air leakage at the boiler air preheater.  Including this period in  the
availability  calculation for this first  year yields an overall system
availability  of 69.3%.  Exclusion of  this interim  period from  the cal-
culation yields availability during operating  periods  of 89%.
                                    443

-------
Operating
  Hours
           800
           600
           400
           200
| Boiler Operating Hours (Available to Dual Alkali System)
Q Dual Alkali System Operating Hours
88
1^"
S «
a
fv
SS 2?
l_ *-
—

-
•
X
j
/
•



r
J
X
X
X
X




> i

8
p
/
/
/
/
{



g

en
^
^
/
/
/
/



•» 1
;
J
'
;
^
;
/











i
i
p
|
^ CMCM ""





5

g
E is




0


»
T
t
i
t
f
\
/
X
X
1





7
r
/
/
i
\
/
t





?
\
',
\
',
',
\

-

_
_

i i i i
                      2/75    4/75     6/75     8/75    10/75     12/75    2/76     4/76
                                           Month of Operation

                      Figure 2,  Availability of the CEA/ADL dual alkali system at Scholz.

-------
                                 TABLE 4
Operating Period 1
                      PROTOTYPE AVAILABILITY SUMMARY
                            Dates
2/3/75-7/18/75
  Operating Hours
Prototype   Boiler5

  2591       3109
         Availability

        Prototype Hours
         as  percent of
         Boiler  Hours

              83.3%
Interim Period

Operating Period 2

  TOTAL YEAR

  TOTAL OPERATING
    PERIODS
7/18/75-9/15/75

9/16/75-1/2/76
   0

  2153

  4744


  4744
1529

2213

6851


 5322
  0

97.2

69.2


89.1
    Hours boiler available  to  dual  alkali  system.
                                    445

-------
                            V.  SYSTEM PERFORMANCE
A.  EQUIPMENT PERFORMANCE

The equipment performance in terms of overall reliability is  reflected
in the system availability and the level of maintenance required.
Even though the  availability of the system has been  very good,  there were
numerous equipment  and instrumentation problems.  To a large  degree, these
problems were mechanical in nature and generally  reflected equipment
design or fabrication oversights commonly associated with a first-of-a-kind
prototype system.   There were only a few problems encountered that reflected
process chemistry,  and these required simple operational and/or equipment
adjustments.

The appendix contains a detailed list of mechanical  equipment and
instrumentation  problems encountered since startup.   Included in the list
are a number of  problems of the type that would normally be expected to
occur during startup or items normally requiring  maintenance.   The fact
that some of the problems shown were not immediately corrected  is  an
indication that  they were either of minor significance or did not  cause
important operational problems.  The more important  problems  are discussed
below.
                                 »

1.  Equipment Problems

The mechanical equipment problems and maintenance items of primary
concern are discussed below.

     Filter.   The  filter was the largest  source  of problems  in  the  prototype
system.  Because of anticipated corrosion  problems associated with high
chloride levels, a  large part of the filtration equipment was fabricated
out of fiberglass, which is not as sturdy  as  stainless steel  and more
prone to failures at stress points in the  construction.  Filter  problems
during Period 1 included:

     •  erosion of the fiberglass scraper  blade,  resulting in jagged
        edges which tore the cloth

     •  erosion of the bridge valve due  to solids carried through  cloth
        holes

     •  loss of vacuum due to cracks  in  the internal drum trunnion tubes

     •  cracking of the plastic caulking strips, allowing retention  ropes
        to loosen and releasing the cloth  panels

     •  failure of the fiberglass  rocker arm used to agitate the slurry
        in the filter tub
                                     446

-------
Modifications were made  to the filter during the latter part of
Period 1 and the early part of Period 2.   These modifications included
replacement of the scraper blade with one fabricated out of stainless
steel, reinforcement  of  the rocker arm with stainless steel plates,  and
the  design of a new method for retaining the filter cloth panels.  These
changes along with regular inspection and maintenance and increased
familiarity of Gulf Power personnel with the equipment have significantly
improved filter performance.

      Reactor System.   Two mechanical equipment problems were encountered
in  the reactor system operation:  plugging of the dry lime feed chute and
solids buildup in  the first reactor.  The chute plugging problem was
caused by hot vapors  drifting up the chute entrance port and wetting the
lime.  The problem was resolved by installation of a vibrator on the chute
with provision for injecting  air to prevent vapors from rising into  the
chute.

The solids buildup in the first reactor was related to the use of dry
lime feed  (as opposed to slurry lime), feed chute location, and the  poor
agitation  in  the  first reactor.  These caused a deposition of solids both
above and  below  the liquid surface, particularly in the area near the feed
chute.  A  simulation of the conditions in this reactor at the CEA/ADL pilot
plant in Cambridge has been successful in confirming the source of the
solids buildup.   A new first  reactor has been designed and fabricated, and
will be installed prior to restarting of the system.  The lime slurry feed
system is  also being activated as an alternative to the feeding of dry
lime.  Feeding  a slurry of slaked lime would be the normal practice  in
a full-scale  application.

      Scrubber.    The rubber linings in the bleed control valves on both
the absorber  and venturi failed during operations in Periods 1 and 2.
These  failures  may have been due to throttling to control flow.  These
valves were originally sized to accommodate direct slurry scrubbing at
much higher flow rates.  These rubber lifted valves are scheduled to be
replaced with stainless steel before the system is put back into operation.

 Leakage of solution  through  the venturi  and absorber pump  seals has been of
 concern primarily because it represents  a loss of sodium from the system.
 Tests were conducted by Southern Services and Gulf Power personnel to deter-
 mine the size of  the leaks under various pump packing conditions and to
 determine how best to eliminate or minimize the leaks.   A Teflon impregnated
 packing was installed just prior to shutdown and it appears  to have all but
 eliminated the leaks.   However, further testing with this  packing is required.

  A  slight  corrosion and ash buildup on the fan during extended shut-
  downs was  diagnosed to be due to flue gas leaks through the isolation
  damper.   Since  the fan is located upstream of the system it is constructed
  out of carbon  steel and significant corrosion was not expected.   The
  corrosion and  ash buildup was generally minor and the fan rotor was easily
  cleaned and  rebalanced prior to startup.  Correcting the leak through the
  isolation damper by installing a new damper or use of an air seal system
  was considered to be unwarranted.


                                      447

-------
     Thickener.    The major problems with the operation of the thickener
were plugging  of  the underflow lines (Period 1)  and leaking of liquor
through the bottom due  to lining failure (Period 2).   Plugging of the
thickener underflow lines was partly due to the  design of  the underflow
piping, and partly due  to the frequent downtime  on the filter which
allowed thickener underflow slurry concentrations to  exceed 30% solids.
Redesign of the underflow lines using flexible piping and  adjustments to
the operational procedures to maintain the underflow  slurry concentration
in the 15% to  25% range (by back-flushing with clear  hold  tank liquor to
dilute the underflow when necessary), have effectively eliminated underflow
plugging problems.

The leak that  developed in the bottom of the thickener during the
early part of  the second period of operation was small but grew worse
throughout Period 2.  After shutdown in January  all the solids were pulled
from the thickener and  the liquor drained.  Inspection of  the inside of
the thickener  showed sections of lining to be failing due  to poor curing
or poor application by  the lining supplier.  The leaks have beeti located
and patched, and  the sections of failed lining are being replaced.

 Interestingly, when the thickener was drained, pieces of rope and
a flattened paint can were found in the area of  the cone.   These items
undoubtedly contributed to difficulties with the thickener underflow
system.  A piece  of rope had previously been extracted from a seized
underflow pump during operation i'l Period 2.

2.  Instrumentation

 Instrumentation problems primarily involved the  pH units,  level
transmitters,  and soda  ash feed solution control system.   Other instru-
mentation problems  occurred, but for the most part, these  were minor.

Flow-through pH units were originally installed  throughout the system.
The piping for these units had a tendency to plug and the  electrodes coat
with a fine film,  causing a drift in pH readings.   In the  case of the pH
unit in the reactor system, the lines and probe  chamber plugged completely
with solids at low  slurry flow rates and the probes eroded away at high
flow rates.  The  flow-through unit on the reactor system was replaced with
an immersion-type unit  fitted with a sonic cleaner midway  through Period 1.
This has proved to  be much more reliable; however,  there have been problems
keeping the sonic cleaner mounted on the probe casing.   The unit now
requires routine  checking and recalibration about every week to two weeks.

The take-off lines  for  the flow-through units on the  scrubber system
have been or are  now being relocated.  Also,  somewhat higher flow rates
will be maintained  through the probes to prevent plugging  of piping.

Since close pH control  is not required, the problems  with  the pH units
have not been  critical.  Also, the scrubber system can be  operated on
either the venturi  bleed pH or outlet S02 (or for that matter, inlet S02
                                      448

-------
 and feed forward flow)  with  only  occasional  checks of the bleed pH to
 verify the SC>2  monitor.   In  fact, during  a few weeks in  the early part of
 Period 1,  when  all  pH units  and S02 monitors were out of service (due to
 delays in  obtaining replacement parts), the  system has been successfully
 operated by taking  hourly pH readings with a portable pH unit.

 The level  transmitter/controllers on a  few tanks have been unreliable.
 The particular  type of  unit  could not be  completely serviced in the field.
 Thus,  there were periods  when inaccurate  level indication and/or control
 have caused operational problems.  These  units have been replaced.

 Similarly, the  soda ash solution  feed control system has also been
 unreliable.  This,  in combination with occasional plugging of the feed
 gate on the dry soda ash  feeder, has made it difficult to close the
 overall material balance  on  sodium.  During  the last half of the second
 operating  period, the sodium makeup rate  was determined  using frequent
 checks on  the specific gravity of the soda ash solution  and recalibration
 of  the flow indicator along  with the inventory of soda ash silo and the
 quantity of soda ash delivery.  Inventory and delivery Information alone
 would  not  have  been sufficient on the short-term due to  the large storage
 capacity in the silo and  the low soda ash feed rates.  The soda ash feed
 control system  is now being  revised.
B.  PROCESS PERFORMANCE

The general process conditions for the two periods of operation are
summarized in Tables 5 and 6.  During Period 1 (February 3—January 18, 1975)
the inlet S02 concentration ranged from 600 to 1,550 ppm and averaged
about 1,050 ppm.  Flue gas oxygen levels varied from 5% to about 10% and
typically fluctuated between 6.5% and 8.5%.

The system operating philosophy during Period 1 was to maintain the
active sodium concentration in the system at or above 0.2M Na"*" and to
maintain a total sodium concentration of about 2.0M.  This resulted in
a fairly stable operation until the end of June, when the inlet S02 level
fell to 850-950 ppm and oxygen levels rose to 8.5-10.0%.  Under these
conditions, without an increase in the sodium carbonate feed rate, the
system chemistry drifted into the range of a dilute dual alkali mode with
active sodium concentrations dropping to below 0.15M Na+.  Except for the
two-week period of dilute mode operation in July, the thickener liquor
typically contained 0.20-0.25M active sodium, 0.7-0.9M sodium sulfate,
and 0.10-0.15M sodium chloride (4,000-5,500 ppm Cl~).

At the start of Period 2, it was decided to maintain the active sodium
concentration in the system above 0.3M Na+ and to allow the sodium sulfate
concentration to fluctuate to any level necessary to maintain an equili-
brium between sulfite oxidation in the system and sulfate precipitation.
This would prevent any possibility of deterioration of the system chemistry
to that of a dilute mode.
                                      449

-------
                                TABLE 5

                  SUMMARY OF SYSTEM OPERATING CONDITIONS

                                 Period 1
                             (2/3 - 7/18/75)
                                                 Range
Inlet Gas:

     Gas Load Treated  (Mw equiv.)

     S02 Level (ppm -  dry basis)

     02 Level (Z dry vol.)




Regenerated Liquor Composition:

     pH
     SO,,* (M)

     Cl~ (ppm)
       1 1
     Ca   (ppm)
   15-22

  600-1,550

  5.0-10.5


   Range3



   10-12.6

 0.10-0.35

 0.60-1.05

3,000-7,000

   20-800+b
  Average



     17

   1,050

    7.5


  Typical3



   11-12.5

  0.2-0.25

  0.7-0.9

4,000-5,500

   50-200
^or March - July 1975, not including periods when  limestone fed to reactor.

bCa++ levels above 250 ppm occurred during the  period just prior to and
 during the dilute mode operation.
                                      450

-------
                                 TABLE 6
                SUMMARY OF  SYSTEM OPERATING CONDITIONS
                                 Period 2
                            (9/18 - 12/25/75)
                                                    Range
Inlet Gas:
     Gas Load Treated  (Mw equiv.)
     S02 Level (ppm -  dry basis)
     02 Level (% dry vol.)
Regenerated Liquor Composition:
     pH
     Na+  ^ (M)
        act
     SO^" (M)
     Cl~ (ppm)
        I I
     Ca    (ppm)
   16-19
  800-1,700
  4.5-9.5

   Range3


   10-12.8
 0.25-0.6
  0.6-1.05
1,000-2,100
   30-160
Average


   18
 1,200
  6oO

Typical3


 11-12
0.3-0.35
0.8-1.0
 1,300
   70
 For October - December 1975, not  including periods when limestone fed
 to reactor.
                                  451

-------
The flue gas conditions during Period 2 improved over those experienced
in Period 1, particularly in regard to the oxygen levels.  Inlet  S02 concen-
trations ranged from 800 to 1,700 ppm and averaged about 1,200  ppm.   Flue
gas oxygen levels generally were controlled to 5-6% with only infrequent
excursions above 6.5%.

At these flue gas conditions, with the active sodium concentration
maintained above 0.3M, the sodium sulfate concentration typically ran
0.8-l.OM, and the chloride concentration fluctuated in the 0.03-0.06M
range (^1,300 ppm Cl~).  Routine analyses of coal samples performed  for
the plant from both Periods 1 and 2 show about the same chloride  content,
approximately 0.08 wt % on the average.  However, the accuracy  of these
analyses is only about ±0.05 wt %.

The various aspects of the process performance during the two operating
periods are discussed in the following sections.  The discussion  has been
broken down into six areas:

        S(>2 removal
        Lime utilization
        Oxidation and sulfate control
        Waste cake properties
        Sodium makeup, and
        Process reliability

Complete material balances could be closed only during selected intervals
in each operating period because of the frequent contamination  of the lime
supply with limestone, fluctuating flue gas and operating conditions, and
occasional instrumentation problems.  However, sufficient analytical and
operational data exist throughout both periods (even when material balances
were not complete) to allow thorough characterization of the system
operation and performance on the low and medium sulfur coal fired during
the first year.

1.  S02 Removal

The scrubber system was operated using two different configurations
for S02 removal:  the venturi and absorber together in series  (with two
trays) and the venturi alone.  In the latter configuration the  trays were
not removed from the absorber; rather, the regenerated liquor  feed to the
top tray was diverted either to the absorber recycle tank (from which it
was transferred to the venturi) or bypassed directly to the venturi
through a line Installed in May 1975.  (The bypass line was not in the
original design, since the operation on low sulfur coal was not anticipated.)
The recycle flow to the top tray and the flow to the spray on  the underside
of the bottom tray were both discontinued; however, the absorber  pumps
were maintained in operation to transfer liquor collected in the  absorber
tank to the venturi.  In order not to overly back-pressure the  absorber
pumps, a recycle was maintained through an open spray header during
intervals when there was no feed to the trays.
                                     452

-------
During Period 1, both operational configurations were  used  at  different
times.   However, during Period 2, only the combined venturi and absorber
configuration was used.

The SC>2 removal efficiency achieved with  each  of these configurations
is shown in Figure 3 for intervals when the inlet S02  ranged from 1,050 to
1,250 ppm,  the typical inlet level.  Points shown on this plot represent
data taken during the normal course of operations in both periods when
inlet and outlet 862 levels and system pH's were simultaneously available
and confirmed.  In most cases, each data  point represents at least a few
'lours of operation at the condition shown.  Where such a continuum of data
exists, outlet SC>2 levels and pH's have been averaged  and rounded to the
learest 5 ppm and 0.05 pH units, respectively.

The data in Figure 3, which reflects  the  general operating  experience
at Scholz,  confirm the high SC>2 removal capability of  sodium-based scrubber
systems operating in an equivalent concentration range of active sodium.
Achieving a given outlet S02 level (within the limit of the number of
contact stages in use) was essentially a.  matter of adjusting the operating
pH of the scrubber system by changing the feed forward rate and/or pH of
the regenerated liquor.
Using both the venturi and  absorber (with two trays) ,  outlet
levels below 50 ppm could be easily achieved  at  a venturi  liquor  pH above
5.2.  This corresponds to better  than  95% removal efficiency.   Higher
removal efficiencies were attained  with  increasingly high  pH levels;
however, there was little to be gained by operating at pH's  much  in excess
of 6.0.  Above a pH of 6.0, outlet  S02 levels dropped  to 20  ppm or less
(>98% removal), taxing the  accuracy and  operating range of the outlet  SC>2
monitor.

For the most part, when  both  the  venturi and  absorber  were operated
together, the pH of the  venturi bleed  liquor  was maintained  between 4.8
and 5.9.  Consequently,  outlet S02  levels generally ranged between 15  and
100 ppm, and typically ran  30 to  70 ppm.   In  Period 2, for example, during
December and November the outlet  S02 level ranged from 5 to  210 ppm and
averaged 45-50 ppm.  During these two  months  there were only five occasions
when the outlet S02 exceeded  100  ppm.  Three  of  these  occurred while  the
system was being operated with limestone rather  than lime  fed to  the
regeneration reactor.

Operating with the venturi  alone  (liquor bypassing absorber trays)
a pH feed of 6.4-6.5 was required to achieve  95% removal  (less than 50 ppm
outlet S02).  However, better than 90% removal could still be quite easily
achieved at pH's on the  order of  6.0.  When the  venturi alone was used in
Period 1, the bleed pH was  generally maintained  above  5.7  to keep outlet
 S02 levels at or below about  100  ppm.
                                       453

-------
           500
           400 -
           300 .
Outlet SO2
 (ppm)
                                                                       Venturi       Active
                                                                     Ap.(in. H20)   Na+,(M)
Operational Configuration
                                                                                    0.25-0.4
                                                                                    0.15-0.3
                                                                                    0.15-0.3
 Venturi + 2 Trays
 Venturi + 2 Trays
 Venturi (No Feed to Trays)
                      1050-1250 ppm
                                               5.5              6.0
                                             Scrubber Bleed Liquor pH

                               Figure 3,   SO2 Removal in the Scrubber System as a
                                         Function of pH of the Scrubber Bleed Liquor

-------
 2.   Lime Utilization

 In  general,  lime utilization was quite good,  equalling  that  predicted
 from the laboratory and pilot plant developmental work.  Under normal
 operating conditions, lime utilization ranged from  90%  to  100% of  the
 available Ca(OH)2 in the hydrated lime (the hydrated  lime  ranged froa 87%
 to  93% available Ca(OH)2).  Based upon analysis of  the  solids  produced,
 lime utilization typically ran about 95% of the available  Ca(OH)2.   These
 data are generally confirmed by overall system material balances.

 During Period 1  it is not possible to accurately calculate the overall
 Ca(OH)2/AS02  stoichiometry due to the frequent operational upsets,  including
 the occasional feeding of limestone.  The best estimate for  the first five
 months of operation (after initial startup) is 0.95 to 1.00.   This  estimate
 is  based upon the integral averages of S02 inlet and  outlet  and lime feed
 (adjusted for estimated unreacted limestone).  During November and  December
 1975,  when accurate material balances were closed,  the Ca(OH)2/AS02
 stoichiometry was 0.98.  Sodium carbonate makeup represented an additional
 alkali input  to  the system.

 3.   Oxidation and Sulfate Control

 As  would be expected, the major fraction of sulfite oxidation  in the
 system occurred  in the scrubber circuit; and  the single most important
 variable influencing the oxidation rate was the oxygen concentration in
 the flue gas.  Estimated oxidation rates in the scrubber circuit ranged
 from a low of about 150 ppm equivalent S02 (at an oxygen level  of 5%)  to
 a high of about  400 ppm (at an oxygen level of 9%).   By contrast, the
 oxidation in  the  remainder of the process (the regeneration  and dewaterlng
 systems combined)  was less than 25 ppm.  Thus, oxidation in  the scrubber
 circuit accounted  for more than 90% of the total system oxidation.

 Figure 4 shows the estimated oxidation rates  in the scrubber system
 during both periods in equivalent ppm of S02  (design gas flow basis)  as a
 function of flue  gas oxygen content.  Included in this plot are data  for both
 scrubber system operational configurations.

 It  would be expected that oxidation rates would be somewhat lower
 using  the  venturi  alone than with the combined venturi and absorber, due
 to  the decreased  gas/liquid contacting.  However, such a decrease in
 oxidation  was  not  observed.   This can be attributed to three  factors:
 first,  the carryover of entrained liquor from the venturi to  the absorber;
 second,  contacting of gas with liquor recirculated through the open spray
header;  and finally,  the flow of a small amount of regenerated liquor onto
 the  top  tray  through a leaky shutoff valve.

Considering all data,  regardless of scrubber configuration, the data
scatter  exhibited  in Figure 4 amounts to a range of 80 ppm of oxidation at
a given  flue  gas oxygen level.   This data scatter can be accounted  for by
differences in operating temperature (120-135°F), liquor flow,  and  slight
differences in pressure drop in the tray tower, as well as  sampling and
analytical errors.

                                     455

-------
*».
tn
                           500
                           400
              300
       Sulfite
      Oxidation
   (ppm of SC>2 —
design gas flow basis)
              200
                            100
 ~~                            Venturi
  Operational Configuration  AP (in.
• Venturi + 2 Trays             5-7
• Venturi + 2 Trays             8-11
a Venturi (No Feed to Trays)   11 — 13
                                                                      Operating
                                                                       Period
                                                                          2
                                                                          1
                                                                          1
                                           5678
                                               Oxygen Content of Flue Gas (% dry vol.)

                                       Figure 4,  Oxidation in the Scrubber System as a
                                                 Function of  Flue Gas Oxygen Content
                                                                                     10

-------
Assuming the median value of oxidation to represent the average
oxidation experienced, then oxidation ranged from 175 ± 40 ppm at 5% oxygen
to 380  ± 40 ppm at 9% oxygen.  For a typical S02 level of 1,200 ppm with
95% S02 removal and 90% gas load, these oxidation rates correspond to 17%
and 37% of the S0£ removed.  Adding in the oxidation through the remainder
of the  system, the total oxidation would be 20% of the S02 removed at a
5% oxygen level and 40% at 9% oxygen.  For high sulfur coals (3-4% S),
these same overall oxidation rates would correspond to roughly 8% and 16%,
respectively.

It is worth noting that these oxidation rates represent operation with
three active contact stages (apparently even when the venturi was operated
alone).  Normally in a low or medium sulfur coal application only one or
two stages should be required.  This reduction in the number of stages
should  substantially reduce oxidation rates.

At steady state the total sulfate formed must be removed from the
system by the precipitation of calcium sulfate and/or sodium sulfate losses
at the  rate it is being formed.  Calcium sulfate levels measured in the
product solids ranged from as low as a few percent of the total insoluble
calcium-sulfur salts during early startup to as high as 30% during periods
of high oxygen levels and low S02 (when the ratio of sodium sulfate to
active  sodium in the system liquor was fairly high).  The typical range of
calcium sulfate in the waste cake during Period 1 was 15% to 25% (mole basis)
of the  total calcium-sulfur salts; and in Period 2 was 12% to 20% (mole basis)
These levels of calcium sulfate precipitation indicate that the system is
capable of keeping up with oxidation rates of up to 25-30% of the S02
removal in the concentrated mode of operation.  In a situation where the
cake was thoroughly washed, the bulk of the sulfate losses would be as
calcium sulfate; sodium sulfate losses would represent less than 2-3% of
the S02 absorbed.  The system liquor composition would simply self-adjust
the ratio of sodium sulfate to active sodium to precipitate the amount of
calcium sulfate required to keep up with oxidation.

The sulfate balances established during November and December 1975
are shown in Table 7.  These months represent stable operating intervals
in which material balances were closed.  During November, 20% of the S02
absorbed was oxidized to sodium sulfate in the system.  Of this 20%, 13%
left the system as insoluble calcium sulfate.  The remaining 7% left as
sodium sulfate.  About half of the sodium sulfate was lost in the liquor
occluded with the cake (V>% sodium salts on a dry cake basis); and the
other half through uncollected pump seal leakage, the thickener leak, and
other unaccounted losses.  The total sodium sulfate loss was calculated
from the soda ash makeup during this period.

In December, 23% of the 862 absorbed was oxidized to sulfate.  About
15% left as insoluble calcium sulfate; about 3% as sodium sulfate occluded
with the cake (^3.5% sodium salts on a dry cake basis);  and the rest through
system leaks and increase in liquor sulfate concentration.
                                       457

-------
                                   TABLE 7
                      SULFATE BALANCES PORING PERIOD  2
Average Inlet S02 (ppm)

Average S02 -Removal (Z)

Average Inlet 02 (vol Z, dry)

Total Sulfate Formation (Z AS02)

Sulfate Account (Z AS02):

     In Cake - CaSOi*

             - Na2S04

     Na2SOit in System Liquor Inventory
     Other
11/3-11/23/75

    1,265

       95.5

        5.5
12/2-12/23/75^

    1,135

        96.5

         6.0
Total
13.0
4.5
-0.5
4.0
21.0*
/j
15
3
2
3
23
.0
.0
.5
.0
.54
 This NaaSOi* loss is calculated from the difference between total net
 sodium makeup and sodium losses in the cake and represents sodium sulfate
 losses in pump seal leaks, spills,  and errors in estimates of soda ash
 makeup.
                                      458

-------
Had all system leaks been stopped or returned to  the system and the
cake been washed to the same level of soluble solids,  then the sulfate to
active sodium ratio in the liquor would have increased somewhat to allow
more calcium sulfate to precipitate, thereby re-establishing  the equilibrium
between sulfate formation and losses.

4.  Waste Cake Properties

The solids content of the filter cake produced during  both periods of
operation typically ranged from 45% to 60%  solids.  In general, the  solids
content decreased with increasing calcium sulfate content of  the cake.
Process upsets (e.g., high 'levels of limestone in the  lime feed) and
mechanical problems with the filter also  tended  to depress the level of
solids in the cake to the 45% range.

During intervals of normal process conditions with the filter operating
under standard operating conditions, the  solids  content of the cake  usually
varied between 50% and 55% for calcium sulfate levels  in the  cake of
between 10% and 25% of the total calcium-sulfur  salts.  At this level of
solids, the cake generally had the appearance and consistency of a moist
powder.

During startup, when active sodium levels were fairly  high (above 0.4M)
and sulfate concentrations were below 0.6M, the  solids content of the cake
ranged from 55% to 70%, indicating the potential for producing even  drier
cake under high sulfur coal conditions.

Wash efficiency tests conducted at Scholz during Period 1 indicated
that with two banks of wash sprays in series, the soluble solids content
of the cake could be reduced to 2-3%  (dry cake basis)  under controlled
filter conditions using a wash ratio  (gals, wash  water/gal. water occluded
in cake) of about 2.5.  However, mechanical problems and process upsets
frequently prevented a reasonably continuous operation of the filter with
a thin enough cake to allow consistently  high wash ratios and, therefore,
low solubles losses.  The levels of  solubles in  the filter cake in Period 1,
therefore, ranged from 2% to 12% of  the dry cake weight with  an average of
5-8%.  The 12% solubles level occurred during periods  when the cake  was not
washed.

Prior  to the start of the second period of  operation  and during the  early
part of Period 2, a third bank of wash sprays was added and the capacity
of'the existing sprays increased.  This resulted in a  doubling of the wash
capacity from 3 to about  7 gpm, which ensured that sufficient wash water
would  be available during periods of high rates  of cake withdrawal.   In
Hovember, the level of soluble solids  in  the cake averaged 4.8%  (dry cake
basis) with a wash ratio  of 1.5-2.0.   In  December, the wash rate was
Increased to a wash ratio of 2.0-2.5 and  the average  soluble  solids  level
was decreased to  3.3%  (a  range of 1.5% to 6.0%  dry cake basis).  The
average solids content during both months was  51%.
                                     459

-------
 During Period 2, a limited amount of testing was performed with differ-
 ent types of filter  cloths.  A number of different nylon and polypropylene
 cloths In both monofilament and multifilament weaves were tried.   The original
 cloth, a polypropylene multifilament, was durable.  However, it produced
 poorer solids (45-50%) under the more adverse process conditions and tended
 to blind, requiring  cleaning of the cloth about two to three times a week.
 The purpose of the testing was to eliminate or minimize the blinding while
 maintaining the high solids content and durability.  To date, the best
 cloth has been a polypropylene monofilament with the same porosity as the
 original multifilament.  This cloth was used exclusively in December 1975.
 The durability of the cloth is still being tested.

 5.  Sodium Makeup

 The sodium makeup requirement in the operation of a dual alkali
 system is determined simply by the rate of sodium loss,  both controlled
 (solubles in the filter cake) and uncontrolled (pump seal leaks,  tank
 spills, etc.).  Under normal operating conditions the losses in the filter
 cake should be the single most important sodium loss,  and sodium  makeup
 should equal or only slightly exceed the quantity contained  in  the  cake
 (on a mole equivalent basis)  in a tight, closed-loop  operation.

 Wash efficiency tests performed at Schoiz during Period  1  indicated
 that 2-3% solubles losses in the cake cduld be achieved with a  wash ratio
 of 2.5.  This solubles loss  translates  directly into an equivalent  soda
 ash makeup requirement of 2-3%  of the S02 absorbed on a mole basis, which
 represents a reasonable lower limit  on  soda ash makeup.

 Soda ash makeup to the system has  been  consistently higher than this
 level throughout  the  first year  of operation.  During the first half  of
 the year (Period  1) mechanical problems  with the filter and Insufficient
 washing limited the level to which soluble  sodium salts could be washed
 from the cake over an extended period to a minimum of about 5-8%.   In
 addition to the cake  losses, there were  also inadvertent spills as new
 operators gained  familiarity with the system as well as the normal leaks
 from pump packing, piping, etc.  Thus, soda ash makeup requirements
 frequently exceeded 10% of the S02 absorbed on a mole basis.

 The improvements  in the filter operation and increased wash efficiency
 in  Period 2  substantially reduced sodium losses in the cake to 4.8%
 solubles  in November  and 3.3% solubles in December.   These sodium losses
 corresponded to a soda ash makeup requirement equivalent to 5.8% and  3.6%
 of  the S02 absorbed (moles Na2C03 per 100 moles of S02 removed), respec-
 tively.  The net soda ash makeup (not including that used in increasing
 the inventory of sodium In the system), though,  was 10.5% in November and
 7% in December.   The higher makeup requirement in November is partially
due to contamination of the lime with 25 tons of limestone, increasing the
amount of waste solids and associated sodium losses.
                                    460

-------
The apparent difference of 3-4% between the estimated soda ash feed
and sodium losses in the cake may be due to leakage  from the system
(pump seals and the small thickener leak) and errors in estimates of the
soda ash feed.

Entrainment losses of sodium were very small.   In particulate sampling
conducted during December, the total weight of  sodium salts in the scrubbed
gas averaged 0.002 grains/scf.  This represents a liquor loss through
entrainment of less than one gallon per hour.   The soda ash makeup required
to replace this entrainment loss is less than 0.1 mole % of the  S02 absorbed.

Efforts will continue throughout the remainder  of the testing to
minimize soda ash makeup requirements by adequate washing of the filter
cake and further control of the leakage of  liquor.   The better cake
produced with high sulfur coal should allow reduction of solubles losses
in the filter cake to below 3%.

6.  Process Reliability

Process reliability refers to  the  overall  ease  of process operation
including the resistance of the process  chemistry  to operational upsets,
the sensitivity of the process performance to  small  changes  in  the process
chemistry,  and the potential  for  scaling  in the process  equipment.   In
this regard it is  to be differentiated  from the mechanical/equipment-
related problems previously discussed.

In all  respects  the process reliability has been excellent.   The
system has  been  successfully  operated over a range of widely fluctuating
inlet S02 levels and oxygen concentrations in the flue gas,  with little
or no change  in  the S02  removal  efficiency, cake properties  or  lime
 utilization.  Throughout  Period  1 with active sodium concentrations
between 0.15  and  0.30M,  soluble  calcium levels generally fluctuated
 between 20  and 250 ppm and were  usually below 150 ppm.   In Period 2,
 when active sodium levels were maintained above 0.3M, soluble calcium
 concentrations averaged about 70  ppm and were consistently below 100 ppm.
 In this range of  calcium concentration, 20-250 ppm,  there were no problems
 with scale  formation  in the  reactor (other than the mechanical problems
 already discussed),  the dewatering, or scrubber systems.

 The low potential for scale  formation and solids deposition in the scrubber
 system was  further demonstrated in Period 2 when the wash sprays beneath
 the demister were turned off.  For the three-month  period from October
 through December the  demister was operated without  wash water.   There was
 no increase in pressure drop across the demister nor the formation of a
 scale  or deposit of any kind.

 The only period when difficulties were experienced  due to changes in  the
 process conditions was in July when the inlet  S02 levels dropped to 850-950
 5pm and the flue gas oxygen level rose to  8.5-10.0%.  The system was  allowed
 to drift into a dilute mode (active sodium levels decreased to  less than
                                        461

-------
0.15M).  The very low soluble sulfite concentrations caused  the  soluble
calcium concentrations to rise to gypsum saturation levels and above,
resulting in the precipitation of gypsum and the deposition  of gypsum
scale in the reactor vessels and reactor outlet piping.  The scale was
removed during system shutdown.  Had the active sodium been  increased by
charging additional soda ash to the system, the gypsum scale would have
slowly redisaolved, as evidenced by the softening and eventual redissolution
   gypsum scale that was not removed during the shutdown.
In addition to the fluctuating S<>2 levels and oxygen concentrations,  the
system also was operated during a number of process upsets.  These upsets
included the carryover of small amounts of fly ash (Period 1), the con-
tamination o€ the lime feed supply with limestone (Periods 1 and  2) ,  and
overfeeding of lime (Period 1).  Only the limestone had any effect on the
pksjcess performance and this effect was temporary.  During periods when
pure limestone was fed to the reactor system, the pH of the regenerated
liquor fell to below 7, causing a loss of S02 removal efficiency,  from
above 90Z down to 80-85%.  The solids content of the filter cake  also
decreased slightly and sodium losses in the cake correspondingly  increased.
Within a day or two after the limestone had passed through the system, all
of the effects were reversed and the system operation was back to  normal.
                                    462

-------
                                                    APPEMPIX
                            EQUIPMENT/INSTRUMENTATION STARTUP AND MAINTENANCE  PROBLEMS
  System
SCRUBBER
 REACTOR
                Problem
EQUIPMENT:
•  Deterioration of refractory in reheat
   burner chamber
•  Slight corrosion and ash buildup on
   fan during extended shutdown periods
•  Separation of bond on rubber linings
   in absorber and venturi bleed control
   valves
•  Hairline cracks and pinholes in
   lining on absorber recycle tank
•  Leaking of liquor through pump seals
   and piping on pump suction lines
•  Deterioration of fan thrust bearing
•  Deterioration of stack lining due to
   poor curing
INSTRUMENTATION:
•  Water  condensation in pressure taps
   on level controllers
•  Plugging of flow-through pH units on
   recycle tanks
EQUIPMENT:
•  Plugging of dry lime feed chute to
   first  reactor
•  Buildup of solids in first reactor
   due to location of dry lime feed
   chute  and poor agitation
Action
                                                        Replaced

                                                        Clean and rebalance  fan after
                                                        extended shutdowns
                                                        Replace with SS valves under
                                                        warranty

                                                        Patched

                                                        Replaced packing and flange
                                                        gaskets
                                                        Replace
                                                        Patch/replace under warranty
                                                        Relocated lines
                                                        Relocate sampling lines
                                                        Installed vibrator

                                                        Design and install new first
                                                        reactor
Status
                      Corrected  (Period 1)
                      Scheduled (1-2/76)
                      Corrected (Period 1)

                      Testing of new packing
                      (Period 2)
                      Scheduled (1-2/76)
                      Scheduled (1-2/76)
                      Corrected (Period 2)
                      Scheduled (1-2/76)
                      Corrected (Period 1)
                      Scheduled (1-2/76)

-------
                            EQUIPMENT/INSTRUMENTATION STARTUP AND MAINTENANCE PROBLEMS
 Scrubber
REACTOR
(cont.)
                Problem
FILTER/
THICKENER
•  Broken agitator blade in second
   reactor
•  Broken shaft on reactor pump due to
   piece of rubber lining from agitator
   blade caught in impeller
•  Failure of isolation valves on
   reactor pumps
INSTRUMENTATION:
•  Erosion and plugging of flow-through
   pH unit
EQUIPMENT:

•  Erosion of fiberglass scraper
•  Erosion of plastic bridge valve due
   to solids carried through cloth holes

•  Loosening of cloth retaining ropes
   out of caulking strips


•  Loss of vacuum due to cracks in
   internal fiberglass trunnion tubes
   in filter

•  Erosion/cracking of fiberglass rocker
   arm on tub agitator

•  Insufficient agitation in filter tub
   to suspend sand-like particles and
   grit

•  Plugging of thickener underflow lines


•  Deterioration of sections of lining
   in thickener and thickener hold tank
            Action
                                                                                    Status
Replaced agitator shaft and
impeller under warranty
Replace
                                                        Overhaul
                                                        Replaced with immersion type
                                                        probe with sonic cleaner
                                                        Replaced with SS under warranty
                                                        Instructed operators to shut
                                                        down filter and repair holes
                                                        immediately
                                                        Designed and installed new
                                                        temporary caulking strips and
                                                        reduced blower pressure
                                                        Patched cracks—new caulking
                                                        strip reduced stress on
                                                        internals when installing cloth
                                                        Reinforced with SS
                                                        Occasionally wash tub—(new
                                                        agitation system to be
                                                        installed)

                                                        Installed flexible lines and
                                                        new back-flushing provisions

                                                        Patch and replace lining under
                                                        warranty
Corrected (Period 2)
                                                                                          Scheduled (1-2/76)
                                  Scheduled (1-2/76)
                                  Corrected (Period 1)
                                  Corrected
                                  Corrected


                                  Corrected



                                  Corrected
          (Period 1)
          (Period 1)


          (Period 2)


          (Period 2)
                                  Corrected  (Period 1)
                                  Scheduled  (1-2/76)
                                  Corrected  (Period 1)
                                  Scheduled  (1-2/76)

-------
 Scrubber
FILTER/
THICKENER
(cont.)
SODA ASH
GENERAL
              EQUIPMENT/INSTRUMENTATION STARTUP AND MAINTENANCE PROBLEMS
                Problem                                Action
INSTRUMENTATION:
•  Poor level transmitter reliability in   Replaced
   thickener hold tank
EQUIPMENT:
•  Clogging of dry feeder gate with lumps  Replace feed control to allow
   of soda ash
•  Failure of circuitry in heat tracing
   on piping
INSTRUMENTATION:
•  Poor reliability of feed control
   system for soda ash liquor
EQUIPMENT:
•  Water freeze damage to pump seal
   water rotameters
                                                        higher gate position
                                                        Replace


                                                        Replace
                                                        Replaced and adjusted operating
                                                        procedures
        Status
Corrected  (Period 2)


Scheduled  (1-2/76)

Scheduled  (1-2/76)

Scheduled  (1-2/76)

-------
                                GLOSSARY
Active Sodium -  Sodium associated with anions  involved  in  SO-  absorption

        reactions  and includes sulfite, bisulfite, hydroxide and carbon-
        ate/bicarbonate.   Total active sodium  concentration is calculated
        as  follows :
               2  x ([Na2S03J + [Na^]) + [NaHS03] +  [NaOH]
Active Sodium Capacity - The  equivalent  amount of S02 which can be  theo-
        retically  absorbed by the  active sodium.  Active sodium capacity
        is  defined by:
             capacity - [Na2S(>3] + 2 x  [Na^] +  [NaOH] +  [NaHC03J
Calcium Utilization - The  percentage of the calcium in the lime or lime-
        stone which is present  in the solid product as a calcium-sulfur
        salt.   Calcium utilization is defined as:


        Calcium Utilization - ™ls (CaS°3 + CaSV generated
                              	mol Ca fed	X 100Z
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.

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 for-
        mation relative to SO  absorption is given by:
           Sulfate Formation » mols S°3
                                 mol S02 removed
                                       466
                                                     100%

-------
Sulfate Precipitation - The formation of CaSO,» XH 0 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 = mols CaS°4
                                          mol CaSO
                                                  X
TOS—Total Oxidizable Sulfur - Equivalent to the sum of all sulfite and
        bisulfite species.
                                    467

-------
                              REFERENCES
1.   LaMantia, C.R., Lunt, R.R., and Shah, I.S.,  "Dual Alkali Process
     for 862 Control," presented at Sixty-Sixth Annual Meeting, Amer-
     ican Institute of Chemical Engineers (November 15, 1973), Paper
     No. 25c.
2.   LaMantia, C.R., Lunt, R.R., Oberholtzer,  J.E.,  Field,  E.L., and
     Kaplan, N., "EPA-ADL Dual Alkali Program—Interim Results,"
     presented at EPA Flue Gas Desulfurization Symposium, Atlanta
     (November 4-7, 1974).
3.   Lunt, R.R., Rush, R.E., Frank, I.E.,  LaMantia,  C.R.,  "Startup and
     Operation of the CEA/ADL Dual Alkali  Process  at Gulf  Power/
     Southern Services," presented at  The  American Institute  of Chemical
     Engineers, November 16-20,  1975.
4.   Kaplan, N., "An Overview of Dual Alkali Processes for  Flue Gas
     Desulfurization," EPA Flue Gas Desulfurization  Symposium,  Atlanta
     (November 1974).
                                  468

-------
         APPLICABLE CONVERSION FACTORS
            ENGLISH TO METRIC UNITS
   British
      Metric
5/9 (°F-32)




1 ft




1 ft2  »




1 ft3




1 grain




1 in.




1 in.2




1 in.3




1 Ib  (avoir.)




1 ton  (long)




1 ton  (short)




1 gal.




1 Btu
°C




0.3048 meter




0.0929 meters2




0.0283 meters3




0.0648 gram




2.54 centimeters




6.452 centimeters2




16.39 centimeters3




0.4536 kilogram




1.0160 metric tons




0.9072 metric tons




3.7853 liters




252 calories
                         469

-------
            THE FMC CONCENTRATED DOUBLE-ALKALI PROCESS
        L. Karl Legatski, Karl E. Johnson, and Lyon Y. Lee

                          FMC Corporation
                        799 Roosevelt Road
                       Glen Ellyn, Illinois
ABSTRACT

     FMC Corporation first developed its Concentrated Double-Alkali
Process during the 1960's in response to internal needs.  The system
has since been commercialized by the Company's Environmental Equipment
Division.  The first full scale system was installed in one of the
Company's own plants in 1971.  Since then, it has been tested on a
prototype scale (- 1 Mw) in ten different applications.  A larger
(3 Mw) demonstration plant has been in continuous operation for over
one year, and the largest double-alkali system in the country was
brought on stream in October, 1975.

     FMC's development strategy and operating experience for these
installations are discussed, and the economic and technical advantages
of the system are enumerated.  The system has demonstrated outstanding
reliability and operability together with capital and operating costs
equal to or less than lime/limestone systems.
                               471

-------
INTRODUCTION

     The FMC Concentrated Double-Alkali Process was initially
developed by the Company's Industrial Chemical Division  (ICD)
in response to the need for a reliable flue gas desulfuri-
zation system for its own plants.

     As early as 1956, ICD experimented with lime and lime-
stone scrubbers and encountered many of the now well-known
problems inherent in these processes.  This early work led
to a program to develop a throwaway process that could meet
the following criteria:

          1.   High availability-

          2.   Easy to operate - within the capabilities
               of existing operating personnel; and

          3.   Capital and operating cost competitive
               with lime/limestone processes.

     The FMC Concentrated Double-Alkali Process was the
result of this program.  It was first utilized as a large-
scale prototype (about 30 Mw equivalent)  in the Company's
Modesto Chemical Plant in 1971.  Responsibility for commer-
cialization of the process for fossil fuel fired boilers
was subsequently transferred to the Environmental Equipment
Division.  Since that time, the process has been tested and
refined on a pilot plant scale and in several  different com-
mercial applications.  A 3 Mw prototype has had 94% avail-
ability for the initial year of operation, and a 50 Mw
equivalent commercial system started up in October, 1975
has also had a high availability since that time.

     In late 1975, FMC received a patent on the Concentrated
Double-Alkali Process covering what are believed to be the
most feasible operating conditions of the process.   FMC
believes that its process has been sufficiently demonstrated
in large scale installations,  and it is aggressively pursuing
large scale utility and industrial applications.   The capital
and operating costs compare very favorably to  lime and lime-
stone systems,  and the reliability and ease of operation are
demonstrated facts.
                                472

-------
PROCESS DESCRIPTION

     Figure 1 is a process schematic for the FMC Concentrated
Double-Alkali Process.  Essentially, the process entails a
sodium based scrubbing loop in which sulfur dioxide is collected
according to the reaction:

                     Na2SO3 + SO2 + H20 •> 2NaHSO3

and subsequently precipitated with calcium in a separate loop
according to the reaction:

                2NaHSO3 + Ca(OH)2 -»• CaSO3 + Na2S03 + 2H2O

     The principal advantages of the process in comparison to
other types of throwaway processes are:

          1.   No scale in the scrubber;

          2.   Ease of control of the process•

          3.   Superior chemical utilization;

          4.   Improved solid waste properties• and

          5.   Ability to control sulfur dioxide
               and flyash concurrently.

     One of the concerns in the double-alkali process
is the oxidation of the sulfite ion in the scrubbing solution
to sulfate.  In contrast to the sulfite ion, the sulfate ion
is not readily precipitated with the addition of calcium hydroxide
and the sulfate ion therefore must be purged from  the process at
a rate equivalent to  its net formation rate in  the process.  The
loss of sodium compounds from the process  is in the form of en-
trained liquor associated with the calcium sulfite and  flyash
insolubles.  This loss is a problem in that alkaline sodium
chemicals are more expensive than alkaline calcium chemicals
and the soluble salts leaving the process may pose a potential
water pollution problem.  However, in comparison to lime and lime-
stone processes, this is not a disadvantage in  that the high
magnesium limes that  have been successful  in lime  scrub-
bing systems often produce a waste  for disposal with a  solubles
content greater than  or equal to a double-alkali process.  Addi-
tionally, use of FMC's Concentrated Double-Alkali  Process  re-
sults in virtually stoichiometric utilization of chemicals
and produces a mechanically stable  filter  cake  of  low permea-
bility.  The filter cake need not be  fixed, which  results  in
total operating costs substantially less  than  those of  conven-
tional lime/limestone systems.
                                 473

-------
  FLUE
  GAS
BY-PASS
                                         TO  EXHAUST
                                         STACK
                                                                      CQ(OH)2-
              DISC
           CONTACTOR
            SCRUBBER
                                                        LIME REACTOR | C«aS03
                                                                                  SOLID TO DISPOSAL
       SCRUBBING
      SO-f NcSO
REGENERATION
                                                                          NOZS034-2HZ0
                Fiqure 1.    Schematic  of  FMC's Concentrated Double-Alkali Process.

-------
     In developing its double-alkali process, FMC was guided
by the following objectives:

          1.   Develop the scrubbing solution composition
               and scrubber design that would minimize the
               net formation rate of soluble sulfate ion;

          2.   Utilize a scrubbing solution and regeneration
               solution chemistry that would be easy to
               control within the constraints of existing
               process control technology and be responsive
               to large and rapid changes in boiler operating
               conditions ;

          3.   Develop the scrubbing chemistry that would
               minimize liquid flow rates and liquid holdup
               requirements in order to minimize the capital
               and operating costs of the facility ;  and

          4.   Develop a regeneration system that would yield
               a solid waste product physically and chemically
               acceptable for landfill disposal while minimiz-
               ing the size and  complexity of the equipment
               required for liquid-solid separation.

     To accomplish these objectives, FMC has studied the
chemical mechanisms  involved in  the process, the unit
operations, the process controls and the materials of
construction.  The rate of  soluble  sulfate  formation has
been characterized as a function of the  solution composition,
flue gas composition, and scrubber  design.   Scrubber design
has been optimized in terms of  scrubber  efficiency,  pres-
sure drop,  and mist  elimination  capabilities.  The chemistry
of the lime regeneration  step has been  studied to maximize
chemical utilization and minimize equipment  size require-
ments.  Thickener settling  rates have been  studied under a
wide variety of operating conditions and vacuum  filter per-
formance has been investigated  in terms  of  alternative
filter design features and  filter cake washing character-
istics.  The filter  cake  product has been evaluated  chemically
and physically on a  laboratory  and  prototype scale  for accepta-
bility as a landfill.  Materials of construction have been
and continue to be routinely evaluated  for  acceptability
under  the various chemical  and  physical  conditions  experi-
enced  in the process.  A  proprietary  control system for
the -process has been designed  and demonstrated by  virtually
thousands of hours of trouble  free  operation.

      The  following  sections briefly describe the state  of
FMC's  development efforts in the various areas  of  process
technology.
                               475

-------
Scrubber Performance

     The original type of scrubber utilized by FMC for sulfur
dioxide control was its proprietary Dual Throat Venturi
Scrubber.  It was felt that a scrubber capable of simul-
taneous flyash and sulfur dioxide collection in a single
contacting stage would offer significant advantages over
the scrubbers utilized with lime and limestone processes
which required flyash collection in a venturi followed by
sulfur dioxide collection in an absorber.  All of FMC's
early pilot plant experience was with a venturi type
scrubber, and the simultaneous collection capability was
adequately demonstrated.

     For stoker fired boilers, the flyash concentration could
typically be reduced to 0.1 Ibs/MM Btu simultaneously with
sulfur dioxide collection efficiencies of 90% at scrubber
pressure drops of approximately 10" w.g.  However, the ven-
turi type scrubber cannot be successfully operated at pres-
sure drops significantly less than 10" w.g. without the risk
of deterioration in collection efficiency.  Furthermore, the
tendency of utilities in recent years to utili2e high efficiency
electrostatic precipitators to comply with particulate regu-
lations has decreased the potential market for simultaneous
flyash and sulfur dioxide removal systems.

     In 1973, FMC developed and tested its proprietary Disc
Contactor Scrubber.  This scrubber, which is essentially a
baffle type scrubber, is designed to allow sulfur dioxide
collection efficiencies in excess of 90% at a relatively low
pressure drop without the use of spray nozzles.  The original
prototype was an 800 cfm unit that was tested on a boiler
operated by Colorado Public Service Company in Boulder,
Colorado.  Subsequently, a 2000 acfm prototype unit was built
and operated in conjunction with the trailer mounted pilot
plant at St. Joe Minerals Corporation.  The unit demonstrated
an average collection efficiency of 94.9% operating on a pul-
verized coal boiler burning 2% sulfur coal.

     While particulate removal was not a requirement of the
scrubber, particulate removal efficiencies were also deter-
mined.  When operating a scrubbing system with a scrubbing
solution containing substantial quantities of dissolved or
suspended solids, the possibility of the particulate emissions
being adversely affected by entrainment of the scrubbing liquor
is a potential concern.  However, in a dissolved salt system,
it is possible to utilize mesh type mist eliminators which
provide excellent entrainment removal.  This was demonstrated
by tests where the scrubber inlet grain loadings, already
below 0.01 gr/scf, were further reduced by the scrubber.
                             476

-------
     Furthermore, during a test where the precipitator was
malfunctioning and was out of compliance with the 0.02 gr/scf
guarantee, the Disc Contactor Scrubber was able to reduce the
grain loading to within the specified level.

     An additional test was conducted to determine the actual
extent of entrainment from the scrubbing solution because the
high sodium sulfate concentrations utilized in a concentrated
double-alkali process may adversely affect total sulfate emis-
sions.  The results of these tests indicated that the entrain-
ment of sodium sulfate was negligible.

     With either type of scrubber, the Dual Throat Scrubber or
the Disc Contactor Scrubber, FMC utilizes a scrubbing solutior
with a pH of approximately 6.5.  If the pH is above 7, carbon
dioxide absorption becomes significant and can lead to calcium
carbonate scaling.  A scrubbing solution pH below 6 is like-
wise avoided because the vapor pressure of sulfur dioxide
increases dramatically for concentrated systems and can lead
to equilibrium limited scrubbing conditions where outlet con-
centrations below 200 ppm are desired.  This fact is demon-
strated in Figure 2 in which the sulfur dioxide vapor pressure
is plotted as a function of pH for a solution temperature of
130°F, a typical saturation temperature for a boiler flue gas
stream.  The vapor pressure of sulfur dioxide over dilute solu-
tions is not significant, as can be seen from the line in the
figure calculated from the Henry's Constant for sulfur dioxide
and water and the first and second ionization constants for
sulfurous acid for a solution with a total oxidizable sulfur
(TOS) level of 0.03 gram moles/liter and an ionic strength of
0.3 gram moles/liter.  However, for concentrated solutions,
the sulfur dioxide vapor pressure at a pH of 6 is approximately
100 ppm and increases rapidly with decreasing pH.

     For these reasons, FMC prefers to operate the scrubbing
side of its process in the pH range of 6 to 7, and preferably
at a set point of 6.5 where the sulfite-bisulfite system is
highly buffered.  The highly buffered system is preferred be-
cause the scrubbing solution is able to adapt very well to
rapid changes in flue gas inlet conditions.  This is in con-
trast to scrubbing systems utilizing a pH of 6 or less which
can be very sensitive to rapid changes in flue gas inlet con-
centrations.

Sulfate Formation Rate

     Superficially, the concept of the double-alkali process
is quite straightforward.  Sulfur dioxide is absorbed with a
sodium sulfite solution, and the sodium sulfite solution is
regenerated by precipitating the sulfur dioxide as calcium
sulfite.  The principal difficulty with the process is that
sodium sulfite reacts very rapidly with oxygen to form sodium
                                477

-------

-------
sulfate, and the sodium sulfate cannot be readily regenerated
into an active sodium form by reaction with lime.  One of the
principal areas of diversity in double-alkali process tech-
nology is the subject of sulfate formation and sulfate removal
from the process.

     FMC has found the effects of flue gas and solution compo-
sition on the rate of sulfate formation  to be significant.  A
major part of FMC's development work has been to evaluate the
effects of these parameters on the  net formation rate of sol-
uble sulfate.  FMC first observed that,  everything else being
equal, increased ionic strength, principally in the  form of
increased sulfate concentration, reduced the net soluble sul-
fate formation rate.  The results obtained during the pilot
plant program indicated that a high ionic strength double-
alkali process can be expected to produce a net soluble sul-
fate formation rate on stoker fired boilers operating with
high excess air and flyash present  equivalent to less than
10%, but probably not less than  5%, of the sulfur dioxide
collected when operating with 2.5%  or greater sulfur coal.

     The effect of oxygen concentration  on net  soluble sulfate
formation rate was also investigated.  The results clearly
indicated that with a pulverized coal boiler operating with
4  to 7% oxygen in the flue gas stream, the equivalent of
approximately 30 to 60 ppm of sulfur dioxide will leave the
process as  sodium sulfate.  While FMC has never been able to
isolate any catalytic effect on  oxidation caused by  the presence
of flyash,  sulfate formation rates  have  generally been higher
for stoker  fired boilers with flyash present  than with pulver-
ized coal fired boilers with no  flyash present.  However, no
data are available for  low oxygen  concentrations on  stoker
 fired  boilers and very  little exists  for high oxygen concen-
trations on PC  fired  boilers.

      Inlet  sulfur dioxide concentration  has  not been found  to
 influence soluble sulfate  formation.   The sulfur dioxide  con-
 centrations encountered in  pilot operations  ranged from 400 to
 8000  ppm.   From these experiments,  no correlation between net
 Soluble sulfate formation  rate  and inlet sulfur dioxide con-
 centration  could be  found.

 time  Reactor  Performance

      For the regeneration  system,   FMC utilizes the concept of
 regenerating the sodium bisulfite  by reaction with lime according
 to the following reaction:

                Ca(OH)2 + 2NaHS03 -> Na2S03 +
                                479

-------
To accomplish this reaction, FMC utilizes  a  low  residence
time continuous stirred tank reactor.   The lime  reactor is
controlled at a pH of about 8.5, which is  effectively the
titrametric endpoint for sodium bisulfite.

     It  is possible to precipitate additional  sulfur values
at elevated pH's according to the following  reaction:
                   Ca (OH) 2 -f Na2SO3 -»• CaSO3 +  2NaOH

However,  it has been FMC's experience that it  is  difficult
to prevent excess lime addition at pH's  in excess of  about 10.
The responsiveness of the pH control system is excellent in
the presence of bisulfite ion,  but is relatively  poor at the
higher pH's existing in the lime-sulfite region.   If  a high
pH set point is used, changes in scrubber operating conditions
which lead to changes in the flow rate to regeneration can
lead to  the addition of excess lime due  to the poor respon-
siveness of the pH control system.   Excess lime leads to poor
chemical utilizations and also results in poor filter cake
quality.  Furthermore,  substantial precipitation  of sulfite
leads to increased calcium solubility in the system and poten-
tial super saturation.  The only advantage of operating the re-
generation system at a high pH is the potential for minimizing
the flow rate to regeneration by precipitation of additional
sulfur values from the sodium sulfite solution.

     FMC prefers to utilize a relatively high bisulfite con-
centration {typically 0.3 M)  in the scrubbing  solution which
provides a flow rate to regeneration equal to or  less than
that obtainable in more dilute  systems that attempt both to
regenerate the bisulfite and precipitate additional sulfite
by reaction with the sodium sulfite.   The virtue  of operating
the lime reactor at the bisulfite endpoint is  further demon-
strated  by the consistently high lime utilizations achieved
by FMC.   Independent analyses of the filter cake  conducted by
Firestone Tire and Rubber Company,  General Motors Corporation,
and St.  Joe Minerals Corporation all indicate free lime con-
centrations in the filter cake  of less than 1% on a dry basis.

     The concentrated double-alkali process has the additional
virtue of "built-in" softening  due  to the high dissolved sul-
fite concentrations always present  in the scrubbing solution.
This results in a dissolved calcium concentration an  order of
magnitude below its saturation  level.  Consequently,  calcium
sulfate  scaling in the  FMC process  is virtually impossible.

     In designing its lime reactor,  FMC  determined that for
operation in the pH range of 8  to 10,  residence times of only
a few minutes were sufficient with  reactor temperatures in
excess of 120°F.  At temperatures below  120°F  (which  may exist


                             480

-------
for applications such as acid plants) steam heat has been
found to be an effective means of elevating the temperature
to a level that will support the lime reaction.

Liquid-Solid Separation

     FMC's Environmental Equipment Division is a leading sup-
plier of thickening and clarification equipment for water
treatment applications.  The design  of  the thickening and
clarification equipment for concentrated double-alkali pro-
cesses is dependent upon many variables including  the quality
of the lime, the composition of the  scrubbing  solution, and
the reactor temperature.  FMC has experimented with rotary
vacuum filters produced by several different manufacturers
and prefers a knife discharge with an air puff to  a belt
filter because it requires less maintenance and produces a
drier cake.

     Vacuum filter performance results  obtained  from one in-
stallation recently are typical.  These show  an  average per-
cent solids in the filter cake of  61.3% for a  two  month period.
The filter cake washing efficiency was  approximately 70%.  This
was accomplished with  a wash rate of from  2 to 3 displacement
washes,  reducing the soluble content of the filter cake on a
wet basis  to  from  2 to 3%.

Landfill Studies

     As  a  regular  part of an ongoing development program, FMC
has evaluated the  filter cake  product produced from this pro-
cess for acceptability as landfill.   These tests have  been
underway in laboratory and  prototype demonstrations since  1973
 under  the direction of Dr.  Raymond Krizek,  Professor of  Soil
 Mechanics,  at Northwestern  University.   The tests  originally
 concentrated  on  the physical  properties of the material  to
 determine its handleability,  transportability, and bearing
 strength as a landfill material.   It was found in the  labor-
 atory  tests,  and later confirmed in pilot plant demonstrations,
 that the material  was  amenable to transport on conveyor belts,
 could  be loaded  into  trucks and subsequently moved, and showed
 a surprisingly good  bearing strength.

      Further  tests were conducted to determine the leachability
 of the material.  These tests showed that, while  the soluble
 salts contained in the filter cake  could be removed in shake
 tests, the material itself was highly  impermeable, and proper
 Management of a landfill could prevent displacement of the
  soluble salts by rain water.
                                481

-------
     A prototype landfill was constructed in conjunction
with the  Firestone Demonstration Plant.   The landfill was
designed  according to good engineering practice,  with appro-
priate test wells and reference wells plus provision for
periodic  core  samples.  These tests have been in progress
over one  year  and, at this time, no contribution of soluble
material  to the ground water has been observed.   Furthermore,
the physical characteristics of the landfill have been extremely
encouraging.   The solid material is distributed  in the pit area
by a front end loader which maneuvers in the area without sig-
nificant  difficulty.  Photographs of the filter  cake and the
landfill  area  are shown in Figures 3 and 4 and serve to demon-
strate the favorable overall characteristics of  the material.

     FMC  recognizes that some states may require fixation
and/or the use of pond linings for disposal of the filter
cake material, but will endeavor to prove that the material
is physically  and chemically acceptable  for disposal in a
non-chemical landfill without the use of chemical additives.
There is  no evidence to date that contradicts the above sup-
position, and  the additional tests being conducted should con-
firm or refute FMC's position.   In the worst case,  a liner
may be necessary, and initial cost estimates indicate that an
appropriate lining material could be provided for less than
$1.00/ton of disposal material.   The filter cake  produced by
FMC's process  should never have to be fixed to provide mechan-
ical stability.

Water Balance

     The  FMC process is designed to dispose of all of the
water that enters the process as either  evaporation or en-
trained moisture on the filter cake.   The degree  of difficulty
in achieving this goal depends on the sulfur dioxide concen-
tration and the temperature of the flue  gas stream.   Higher
sulfur dioxide concentrations require more lime  slaking water
and more  filter wash water.  Most of the water that leaves the
process is in  the form of evaporation, and the evaporation
loss is directly related to the inlet flue gas temperature.
Thus, a flue gas at 400°F containing 1500 ppm sulfur dioxide
does not  pose a water balance problem, while a flue gas stream
at 280°F  containing 3000 ppm sulfur dioxide will  have a very
tight water balance.   For example,  in the latter  case,  two
displacement washes on the filter cake may consume 50% of the
total makeup water requirement and lime  slaking water may con-
sume an additional 30 to 40%.   The washing and lime slaking
requirements,  therefore,  may consume virtually the entire
water makeup requirement of the  process.

     The only uses of water in  the process other  than the filter
washing and lime slaking water requirements are the pump seal
water,  soda ash solution makeup,  and mist eliminator washing
requirements.
                              482

-------
                                                V  ,«** ^«* i  •»
    Figure 3.    Landfill area  at  Firestone (note dozer tracks
                 in right foreground).
   <»".      ~fr~
         >l   >**
   <-'  ^ ,J~ v1*
m  *    m^ ^^ •
 *<•*«  -Vfc'_-4c.
                         v
Figure 4.   Climbing a recently dumped three-foot deep pile of
            waste  material demonstrates physical stability.
                            483

-------
OPERATING EXPERIENCE

     FMC's operating experience can be summarized in five
different applications or areas of experience:

          -1.   The original installation at FMC's
               Modesto Chemical Plant;

          2.   Subsequent pilot plant work-

          3.   The Firestone Demonstration Plant;

          4.   The Caterpillar/Mossville installation; and

          5.   FMC's Green River Plant installation.

The most  important facts concerning these projects are dis-
cussed below.


Modesto Chemical Plant

     In 1956, the Modesto Chemical Plant of FMC's Industrial
Chemical  Division began work on an evaluation of the various
possible  processes for sulfur dioxide removal from flue gases.
The Modesto Chemical Plant produces barium and strontium chemi-
cals, such as the oxides, chlorides, carbonates and nitrates,
by reducing barium and strontium sulfate in a reduction kiln
to produce the respective oxides, which are reacted with the
appropriate acids or soda ash in other parts of the plant.
The reduction operation produces relatively high sulfur dioxide
concentrations fluctuating in the range of 4000 to 8000 ppm.

     The  Modesto staff, with assistance from the Industrial
Chemical  Division's Princeton Research Center, initially tested
lime and  limestone scrubbers, but encountered many of the now
classical problems with these systems.  They next piloted a
sodium based scrubbing system using soda ash, since soda ash
was manufactured by FMC and already in use in the plant in
other areas.  The soda ash system proved to be very effective,
but disposal of the sodium sulfite and sulfate products posed
a problem.  FMC was ultimately able to negotiate a contract
with a nearby paper mill to sell the sodium sulfite and sulfate
for makeup to their paper process.  Unfortunately, just as
                              484

-------
detailed design of this system was completed, the paper mill
cancelled the arrangement.  Thus, the Modesto staff was faced
with finding a method of recovering the  sodium values from
the sodium based scrubbing system while  producing an accept-
able solid waste for disposal.   In response  to this necessity,
they developed the concentrated  double-alkali process.  This
process was, in fundamental areas, the same  as the process
sold commercially today by FMC.  The double-alkali process was
successfully demonstrated by  the Modesto staff on a pilot scale,
and subsequently installed and operated  on a full scale in 1971.
Figure 5 is a photograph of the  Modesto  operation.

     This unit, which processes  a flue gas  stream of 30,000 acfiu
is equivalent to approximately  30 Mw on  the  chemical regener-
ation side of the process.  While the unit  is not completely
representative of the design  FMC would utilize today, the system
has demonstrated an  extremely high degree of reliability in an
application experiencing severe  fluctuations in  the process in-
put conditions.  Since  startup,  the process  has  been available
in excess of 95% of  the  time, the majority  of the non-available
time accounted for by kiln down  time  rather than scrubber down
time.

Pilot Plants

     Shortly after the  startup  of  the Modesto  facility,  the
Air Pollution  Control Operation began an evaluation  of  the
potential for  commercialization of  sodium scrubbing  and double-
alkali processes.  The  processes were found to  offer significant
operating advantages in coal-fired  applications  in comparison
to the competing  lime and  limestone processes,  and the  capital
and operating  costs  appeared to be  very competitive.  Further,
it seemed that if  sulfur dioxide could be adequately collected
in a venturi type  of scrubber using a concentrated  sodium based
scrubbing system,  the  combining of  flyash and sulfur dioxide
collection  into one  operation would offer significant capital
cost advantages.

     In  the Spring of  1972,  a pilot test program was performed
at Modesto  using  a venturi scrubber on a slip stream from  the
kiln gas system.   This  was the  first in a long series of pilot
plant  experiments with the double-alkali process.  This pro-
gram demonstrated that a venturi type scrubber,  such as FMC's
Dual Throat Scrubber,  could  successfully collect sulfur dioxide
 in excess of 90%  efficiency  using the concentrated double-alkali
process.  Inlet sulfur dioxide  concentrations of from 4000 to
 8000 ppm were  routinely reduced to the  100-200 ppm level.

     Based  on  this encouraging  experience,  a pilot plant of the
 regeneration part of the process was built  to operate in con-
 junction with the pilot scale venturi scrubber.  This unit was
 installed on a slip stream from a stoker fired boiler at FMC's
                               485

-------

Figure 5.
Concentrated Double-Alkali System installed
at FMC's Modesto Chemical Plant.
                           .

-------
Industrial Chemical Plant in South Charleston, West Virginia
in late Spring, 1972.

     The test at South Charleston lasted for approximately
six months and consisted largely of experiments related to
scrubber performance.  The boiler was very old and seldom
fired at full load, so the oxygen concentration was very
high, typically 12 mole percent.  This fact, combined with
the relatively low inlet sulfur dioxide concentrations, which
fluctuated from 400 to 1400 ppm, led to oxidation rates that
represented a significant fraction of the sulfur dioxide col-
lected.  Thus, the majority of the effort at South Charleston
was devoted to determining the effects of solution composition
and scrubber operating conditions on the sulfur dioxide col-
lection efficiency and the rate of soluble sulfate formation.

     At the conclusion of the test program, it was decided that
the process was technically feasible for application to coal
fired boilers.  The Air Pollution Control Operation was incor-
porated into the Environmental Equipment Division, and the de-
cision was made to pursue the sulfur dioxide control market in
the industrial boiler field.  The pilot plant was installed in
a 35 ft. van to provide efficient transportation to the facili-
ties of potential customers.  In the Spring of 1973, a contract
was obtained from Caterpillar Tractor Co. for a two month
demonstration of the process on a slip stream from a stoker
fired boiler at the heating plant of Caterpillar's Mossville
Engine Plant.

     The pilot plant was operated on an 8 hour per day basis
for two months by two FMC Engineers, and the results were com-
pletely satisfactory to Caterpillar and consistent with the
prior pilot plant experience by FMC.  As a result of this suc-
cessful operation, Caterpillar Tractor Co. awarded FMC a
contract in May 1973 for turnkey installation of the sulfur
dioxide control system at the Mossville Engine Plant.

     The pilot plant program was ultimately extended to ten
different applications in which the process was applied for
periods from a few weeks to several months.  The range of
the input variables experienced by the system was substantial:

          SO2 concentrations -  400 to  8000 ppm

          0- concentrations - 5 to 12%

          Temperature - 120 to  500°F

          Particulate - Nil to  3.0 gr/sdcf

These varied  inputs allowed FMC to develop a  knowledge of the
phenomena that govern sulfite oxidation and  sodium consumption
that is unsurpassed  in the  industry.


                              487

-------
Firestone Demonstration Plant

     In January of 1975, FMC completed the installation of
a three Mw equivalent demonstration plant (Figures 6 and 7)
for The Firestone Tire & Rubber Company in Pottstown, Pennsyl-
vania.  This installation, utilizing a 10,000 acfm Dual Throat
Venturi Scrubber, is operating on a slip stream from a pulver-
ized coal boiler that is presently burning 2% sulfur oil to
meet particulate emission regulations in the State of Pennsyl-
vania.  While this unit is not of commercial scale and is not
officially a commercial operation, the experience at Firestone
demonstrates four very important points about the process.

     First, the installation is of major significance because
virtually all of the process data and relationships that were
developed in the pilot plant program have been confirmed in
this installation, which is four or five times as large as
the trailer mounted prototype unit.

     The second key feature of the project is the proven oper-
ability of the system.  Two weeks after startup, the operation
of the unit was turned over to three boiler house operators
 (one per shift) who have been responsible for the operation
since that time.  FMC provides one engineer on the day shift
to perform chemical analyses and experiments related to the
performance of the various unit operations, but he has vir-
tually no responsibility for the routine operation of the
system.  Performance has been so good and the system has re-
quired so little operator attention, that a remote alarm
system is now being installed to allow the operators to work
their normal shift at the boiler control panel and inspect
the demonstration plant area only once every two hours or
whenever a common alarm sounds at the boiler panel indicating
an upset in the process.

     Third, the landfill study program discussed in a pre-
ceding section has demonstrated the handleability and imper-
meability of the filter cake waste material.

     Finally, the availability of the operating system has
been outstanding.  For the entire first year of operation,
including most of the start-up period, the overall availability
has been 93.5%.  This is particularly impressive in view of
the lack of any spares in the system and the fact that the
operation was often unattended on the night shift.

     The various segments of the system that contributed to
the downtime are itemized in Table 1.  The single longest
outage was due to blockage of the thickener underflow pipe
by a plastic beaker that was unintentionally dropped into the
thickener.  The second longest outage was caused by improper
maintenance of the recirculation pump seal, which led to
separation of the pump lining from the pump housing.  The
                               488

-------
Figure 6.   FMC Dual-Throat Scrubber and Cyclone in
            Firestone Demonstration Plant.
                           489

-------
Figure 7.
Lime storage tank and dumpster at Firestone
Demonstration Plant.
                            490

-------
Table 1.
AVAILABILITY OF  FIRESTONE DEMONSTRATION PLANT
                                             Downtime
     Item

1.  Thickener  Pluggage

2.  Pump s

3.  Cake Conveyor

4.  Fan

5.  Lime Feeder

6.  Spray  Nozzles

7.  Instrumentation

8.  Control Valves
                 % of Total Downtime

                      20.7

                      16.1

                      16.0

                      15.4

                      13.4

                      12.4

                       4.1

                       1.9

                     100. 0
% of Operating Year

     1.35

     1.04

     1.04

     1.00

     0.87

     0.80

     0.27

     0_.13_

     6.50
     Total Annual Availability  =  93. 5i
     Note:  The  Firestone Demonstration Plant has no installed
            spares.
                              491

-------
filter  cake  conveyor problems generally related to  the
fact  that  an inappropriate conveyor was used, due to  the
unavailability of high quality equipment for the small
size  required.  The fan outage was  due to overheated
bearings,  which were in turn caused by shaft misalign-
ment.   The remaining outages were similarly due to  problems
unrelated  to the process,  and all of them could have  been
eliminated or substantially reduced by a normal commercial
sparing philosophy.

Caterpillar  Mossville Installation

      FMC's largest double-alkali installation to date is the
system for Caterpillar Tractor Co.  at its Mossville Engine
Plant outside of Peoria,  Illinois.   The unit is the largest
double-alkali system in the country and is designed to col-
lect  sulfur  dioxide and flyash from four stoker fired boilers
totaling 460,000 Ibs/hr of generating capacity.  The  operation
of this unit commenced in  October,  1975.  Figures 8 and 9 il-
lustrate portions of the system.

      During  startup and initial testing, the Caterpillar Moss-
ville unit has demonstrated a high  availability.  At  the present
time, two  of the boilers are in operation, and the  system
has not yet  been tested at full load.  Furthermore, one of
these boilers is new and has been operated only intermittently
because it is still being  debugged.   However, there has been
only  one occasion since startup in  which a boiler has been
shut  down  due to a problem with the FGD system, and the scrubbers
have  not been modified or  maintained during the boiler outages.
FMC expects  the high availability achieved during startup to
continue or  improve during sustained boiler operations.

      One of  the key features of the Mossville installation is
the simultaneous collection of  flyash and sulfur dioxide.
There are  no mechanical collectors  on these boilers,  and the
full  flyash  load is removed by  the  scrubbers.  This fact, com-
bined with the high excess air  inherent in the stoker boilers,
creates conditions that are far more technically difficult for
a  double-alkali process than a  low  excess air pulverized coal
boiler  with  a high efficiency precipitator attached.  At Moss-
ville,  most  of the problems to  date have been related to wear
with  the control valves, filter cloth, and conveyor belts.
At this time,  FMC feels that it has arrived at satisfactory
solutions  to all of these  problems.   In the long run,  the con-
centrated  double-alkali process offers the potential  for re-
liable,  simultaneous collection of sulfur dioxide and flyash
in  utility and industrial  applications.   This would result in
substantial  savings in  capital and operating costs,  and FMC
feels that this will be one  of the major  advantages of double-
alkali  systems in the future.
                              492

-------
Figure 8.   Dual-Throat  Scrubber (80,000 ACFM)  being installed
            on 150,000 lb./hr.  boiler at Caterpillar Tractor Co
                             493

-------
Figure 9.   Control panel at Caterpillar Mossville installation,
                             494

-------
     Based on FMC's performance to date on the Mossville
project, Caterpillar Tractor Co. has awarded FMC contracts
for the design of complete S02 and particulate control
systems for its East Peoria and Mapleton plants.  Each
system will have a heat input equal to approximately 100 Mw.
The engineering design for these units has been completed
and ultimate startup of the two systems is expected to be
in 1977 and 1978, respectively.

FMC Green River

     This Spring, FMC will start up its largest sodium
scrubbing system which consists of two 330,000 acfm Disc
Contactor Scrubbers on the new pulverized coal fired
boilers now being installed at FMC's Green River Soda Ash
Plant  in Green River, Wyoming.  The boilers  are two
625,000 Ibs/hr units followed by high efficiency electro-
static precipitators.  The scrubbers are designed to remove
90% of the sulfur dioxide at a pressure drop of 5.5" w.g.,
and are equipped with a bypass reheat arrangement.

     This represents the  third different type of reheat
arrangement used by FMC  in commercial installations.  The
Caterpillar Mossville facility,  due  to  spatial limitations,
uses Dual Throat Scrubbers followed  by  steam tube reheaters
operating on  an ambient  air  stream injected  into the  scrub-
ber discharge,  followed  in turn  by an  induced draft  fan.
The East Peoria and Mapleton  installations  for Caterpillar
utilize forced  draft  fans on  the scrubbers,  followed  by
provision for steam  tube air  injection  reheat. The  use  of
steam  tube air  injection reheat was  dictated by Caterpillar
to reduce its dependency on  fuels other than coal.   Due  to
the low sulfur  dioxide  collection efficiency requirement
at Green  River,  the  scrubbers are designed to utilize by-
pass  flue gas for  reheat whenever possible and oil  fired
reheat when  higher sulfur coal is burned or the boiler  is
at full load.
                              495

-------
CAPITAL AND OPERATING COSTS

     It is difficult to generalize on the capital and
operating costs  for a flue gas desulfurization system,
because of the differences in design bases and input
assumptions that can occur.  Due to significant operating
advantages of the double-alkali system, many people have
arrived at the conclusion that it is inherently more
expensive than lime/limestone.  Closer examination of both
systems indicates that this conclusion is not justified.

     In capital  cost, it is superficially obvious that
double-alkali and lime/limestone processes must be com-
parable.  In fact, the comparison in Table 2 indicates
that double-alkali may be less expensive.  Virtually
every component  in the double-alkali process has an
equal or larger  counterpart in a lime/limestone process.
The only potential disadvantage of double-alkali is that it
requires filters, but many lime/limestone systems now
use filters.

     In operating costs, the concentrated double-alkali
system is generally less expensive in all areas.  Chemical
consumption is essentially equal to lime systems; limestone
may be less depending on location.

     Power requirements are less due to lower pressure
drops and liquid rates.  Water requirements are slightly
less.  Reheat is often less because the high efficiency
of double-alkali makes bypass reheat possible.  Maintenance
and operating labor are obviously less in the highly buffered,
soluble salt, concentrated double-alkali system.  Concentrated
double-alkali systems do not scale, and scrubbing with a
solution instead of a slurry results in substantially
less wear.

     Finally, solid waste disposal is substantially less ex-
pensive.  There  is less material to begin with because of the
superior chemical utilization and the low moisture content of
the filter cake.   Chemical fixation is not required for
mechanical stability.

     A typical example of estimated operating costs for a
300 Mw Midwestern utility burning 3.13% sulfur coal is
given in Table 3.
                               496

-------
Table 2.   QUALITATIVE  CAPITAL COST COMPARISON OF

           DOUBLE-ALKALI  AND LIME/LIMESTONE SYSTEMS*
Equipment Item


1.  Scrubbers



2.  Pans


3.  Pumps


4.  Thickener

5.  Filter

6.  Tankage


7.  Materials  in




8.  Materials  out
Lime/Limestone
12 to 15" w.g.
typical

L/G = 30 to  40
typical
Less or Equal
Less  storage  for
1ime s tone,  but
substantially more
material used
FMC Concentrated
  Double-Alkali
Equal or smaller
(less mass transfer
 units)

Smaller
(6" w.g. or less)

Smaller
(L/G = 10 gal/mcf)

Equal
Less  (low residence
time)

About same as lime
                       Substantially less
     *No flyash removal included.
                              497

-------
 Table  3.    ESTIMATED OPERATING COSTS  FOR THE FMC  CONCENTRATED

            DOUBLE-ALKALI  SULFUR DIOXIDE  REMOVAL SYSTEM1
 Chemicals

 Lime  (92%  CaO)

 Soda  Ash
Consumption

0.047 Ib/KWH

0.0021 Ib/KWH
Unit Cost      C/KWH

 $33/ton       0.0776

 $64/ton       0.0067
 Utilities

 Power

 Water

 Fuel Oil2
0.0137 KWH/KWH

0.078 gal/KWH

0.0006 gal/KWH
 1.75C/KWH     0.0240

 40C/Mgal      0.0031

 35C/gal       0.0210
 Operating
 Labor  &
 Supplies
1 man/shift +
supplies
 $30,000/man   0.0067
 year +
 $50,000
 supplies
Maintenance
3% of Plant
 0.03x$21  MM   0.0240
Solid Waste
Disposal
0.204 Ib/KWH
 $4.00/ton
0.0408
TOTAL OPERATING COST  (excluding capital charges)
                                        0.2039
 300 Mw unit operating at full load with 3.13% sulfur coal;
 90% SO2 collection; system designed for 10% over heat
 input using the maximum sulfur, minimum Btu coal expected.


250°F of total reheat:  25°F with oil,  and  25°F with bypass
 flue gas.
                            498

-------
CONCLUSIONS

     In conclusion, there are substantial advantages to
the FMC process in comparison to other systems.  In com-
parison to other double-alkali systems, FMC offers the
following:

          1.   FMC has more cumulative hours of operating
               experience in pilot plant, prototype, and
               commercial systems on more different appli-
               cations than anyone else.  Due to the
               breadth of its experience, FMC is in a
               position to specify performance stan-
               dards and accept significant financial
               obligations as a consequence of potential
               nonperformance.

          2.   The sulfur dioxide collection capabilities
               of FMC's process and equipment are demon-
               strated facts.  With sulfur dioxide control
               systems there  is a potential risk that  the
               scrubbing liquor entrainment will adversely
               affect the particulate  and/or primary sul-
               fate emissions.  FMC has demonstrated that
               its Disc Contactor Scrubber does not adversely
               affect fine particulate or primary sulfate
               air emissions.

          3.   In  terms of process considerations, FMC has
               demonstrated  essentially  100% utilization of
               makeup chemicals  (soda  ash and  lime) through
               thousands of  hours of prototype and commer-
               cial operation.   Scale  prevention  is inherent
               in  the FMC process due  to  sulfite  softening
               of  the scrubbing  liquor in the  regeneration
               portion of the process.  At  no  time  in  either
               its commercial or prototype  experience  has
               FMC experienced a  scaling  problem  in its
               scrubber or  elsewhere  in its  process.   FMC
               has an excellent  knowledge of  the  causes of
               sulfate  formation and  has demonstrated  the
               capability to predict  and inhibit  the
               net sulfate  formation  rate  it will experience
               by  controlling the  process  operating conditions.

           4.   The principal objective of  all  of  FMC's develop-
               ment work  has been to  develop a process that
               takes  advantage of  the  inherent reliability of  a
               sodium scrubbing  process while striving to keep
               the process  simple enough to compete in capital
                                499

-------
              and operating costs with lime  and  limestone
              systems.  The inherent  reliability of  the
              process has been  demonstrated  in the Modesto
              Plant  of FMC and  the Firestone Demonstration
              Plant.  The overall simplicity of  the  process
              is apparent from  the process description con-
              tained in this  section.

          5.   FMC's  Double-Alkali Process is proprietary and
              is patented  (U.S.  Patent No. 3,911,084).   FMC
              believes the operating  conditions  specified in
              the patent are  optimal  for operating a concen-
              trated double-alkali process.


     In comparison to lime/limestone processes, FMC believes it
offers the following  significant advantages:

          1.   Ability  to achieve higher SO2  removal  efficiencies.
               This often allows the use of bypass reheat result-
               ing in a  substantial savings.

          2.    Sodium scrubbing  processes are much more toler-
               ant of process  upsets and variations in gas
               flow and  composition.

          3.    The reduced  liquid-to-gas ratio (usually 10 gallons
               per 1000  acfm)  results  in smaller  pumps and tanks.

          4.    The double-alkali process produces a drier and
              more mechanically stable filter cake.

          5.    The sodium  scrubbing processes produce a lower
               outlet particulate loading due to  no slurry in
               the scrubber.   This allows the use of  a mesh type
              mist eliminator without plugging.

          6.    The S02  collection efficiency  is variable  over
              wide limits  by simply adjusting the chemistry.
               This can be  significant in cases where a high
               SO- collection efficiency is not desirable or
               necessary.

          7.    The scrubbing  solution  is highly buffered  so cor-
               rosion is reduced. This also  results  in a more
               operable and less sensitive system.

          8.   Calcium  scaling does not occur. FMC has never
               shut down a  scrubber because of scaling in an
              SO2 application.
                               500

-------
          9.    One of the most significant advantages of
               sodium scrubbing is the ability to collect
               SO  and flyash concurrently.  This can re-
               sult in substantial capital cost savings.

         10.    FMC's installations are being operated by
               boiler house personnel.  All installations
               have had in excess of 90% availability, and
               the average has been about 95%.  The demon-
               strated reliability and operability of FMC's
               installations is unsurpassed in the industry,
               to our knowledge.


     FMC continues to develop and refine its basic process
through an intensive research and development program in
the following areas:

          1.    Improved scrubber design to reduce
               capital and operating costs;

          2.    Improved thickener design to reduce
               space requirements;

          3.    Improved filter cake washing to further
               reduce sodium losses; and

          4.    Continued landfill evaluation program to
               demonstrate the acceptability of disposal
               without further treatment.
                             501

-------
ACKNOWLEDGEMENTS
   FMC Corporation gratefully acknowledges the assistance
of Messrs. Roman Zaharchuk and Gary W.  Wamsley of The
Firestone Tire & Rubber Company and Mr. David C.  Dietrich
of Caterpillar Tractor Co., all of whom made material con-
tributions to the success of the projects discussed in this
paper.
                           502

-------
     OPERATING EXPERIENCE WITH THE ZURN DOUBLE ALKALI FLUE GAS
                      DESULFURIZATION PROCESS
                            P. M. Lewis

                       Zurn Industries, Inc.
                       Air Systems Division
                        Birmingham, Alabama
ABSTRACT

     The first Zurn Double Alkali Flue Gas Desulfurization went on-line
in September, 1974, at the Joliet Plant of Caterpillar Tractor Company,
Joliet, Illinois.  The system is scrubbing the flue gas from two coal
fired boilers rated at 100,000  Ib./hr, and 80,000 Ib./hr. steam.  These
boilers produce steam for building heat and are in service about seven
months a year.  The system utilizes a dilute double alkali process with
lime addition for regeneration  of the spent scrubbing solution, and
sodium carbonate addition for scrubber feed water softening.  The calcium
based sludge generated by the process is dewatered for disposal in a
remote landfill.

     This paper summarizes the  process, and discusses operating experiences
to date.  Process reliability,  equipment performance, and solutions to
problems are discussed.
                             503

-------
OPERATING EXPERIENCE WITH THE ZURN DOUBLE ALKALI FLUE GAS DESULFURIZATION
PROCESS.
                          I.   INTRODUCTION
 Due  to  the uncertain  future availability and cost of natural gas and fuel
 oil, Caterpillar  Tractor Co. decided  in 1970 to rely on coal for steam
 generation in two boilers at the Caterpillar Tractor Co. plant in Joliet,
 Illinois.

 Basically, there  are  two methods for  complying with sulfur oxide emission
 standards while burning coal.  These  are, removing the sulfur oxides from
 the  flue gas  before it  is emitted to  the atmosphere, and the use of low
 sulfur  coal.   The demand for low sulfur coal was increasing rapidly, and
 its  future price  and  availability prospects were not favorable.  The use
 of high sulfur coal and flue gas desulfurization appeared to be the most
 economical alternative  that would provide an adequate fuel supply and ensure
 compliance with current regulatory requirements.  In particular,  after
 technical and economic evaluations, they selected the Zurn Double Alkali
 Flue Gas Desulfurization System.

 It is the purpose of  this paper to describe this system and to relate the
 operating experience  at the Joliet, Illinois installation to date.
                     II.  SYSTEM DESCRIPTION
 The Zurn Double Alkali Process uses a dilute solution of sodium hydroxide
 to absorb  the sulfur oxides from flue gas.  Lime is used to regenerate the
 spent  scrubbing solution to form sodium hydroxide and insoluble sulfur
 bearing calcium compounds which can be removed by mechanical means.

 This double alkali process can be viewed as having three functionally
 different  process stages.  As shown in Figure 1, these process functions
 are gas scrubbing, chemical regeneration,  and solids removal.

 GAS SCRUBBING

 The function of the gas scrubbing section is to intimately mix the flue
 gas containing sulfur dioxide with a dilute sodium hydroxide solution.
 This reacts with the sulfur dioxide gas and converts it to water soluble,
 sulfur containing compounds (sodium sulfite,  sodium bisulfite,  and sodium
 sulfate).   The scrubber also removes a high percentage of any fly ash or
other particulate which may be contained in the flue gas.
                            504

-------
The reactions that take place  in the scrubbing section of the system
are shown below:
                  2 NaOH  +   S02 -t>- Na2S03+H20      (1)

                  Na2S03+S02-m20 -£>- 2NaHS03         (2)

                  Na2S03+l/202  -O-  Na2S04          (3)
Sodium hydroxide and  sulfur  dioxide  react  to  form  sodium sulfite (reaction
1).  The sodium sulfite  then reacts  with sulfur  dioxide and water to form
sodium bisulfite (reaction 2).   Some of  active sulfite is oxidized to
sodium sulfate by  excess oxygen in the combustion  gases (reaction 3).
After being mixed  with the flue gas,  the solution  is withdrawn from the
scrubber to an external  mix  tank for chemical regeneration of the spent
scrubbing  solution.

REGENERATION

The purpose of the chemical  regeneration section is  the conversion of the
spent scrubbing solution to  a usable form, and the separation of this
solution from the  sulfur containing solid compounds.  In the mix tank,
calcium hydroxide, from  previously slaked quick  lime, is added to the
spent scrubber solution.  The calcium hydroxide  reacts with the soluble
sodium salts  to form insoluble calcium salts  (calcium sulfite and calcium
sulfate  (or gypsum).  This reaction is effected  in two agitated process
tanks and  the resultant  mixture is then pumped to a  thickener where the
calcium salts precipitate and are removed as  a slurry.  The regenerated
sodium hydroxide  solution overflows to a clarifier.   The  reactions in
the regeneration  section are:

                   NaHS03 + Ca(OH)2 -£=-  NaOH + CaS03 y + H20     (4)

                   Na2S03 + Ca(OH)2 -t>~  2NaOH + CaS03 y           (5)
                   Na2S04 + Ca(OH)2->-  2NaOH + CaS04 Y          (6)

 Sodium carbonate is added to the clarifier to further reduce the concen-
 tration of calcium by the precipitation of calcium carbonate (limestone).
 This step is referred to as "softening".  The further precipitation of
 calcium in this step reduces the likelihood of calcium salts precipitating
 in the scrubbing section.  The softening reaction is as follows:
                   Na2C03 + Ca(OH)2 ->• 2NaOH + CaC03 1          (7)

 After the solution has passed thru the clarifier, it has been regenerated
 and is available for reuse in the scrubbing section to remove more sulfur
 dioxide   The insoluble calcium salts  have been removed in separate
 process 'equipment specifically designed for solids precipitation and
 concentrat ion .
                              505

-------
LIQUID-SOLID SEPARATION

The underflow slurry from the thickener and clarifier  is  pumped to the
solids removal section of the process for further dewatering.   The
function of the solids removal section is to recover the  usable scrubbing
solution contained in the thickened underflow slurry and  to convert the
slurry solids into a form which is easily handled and  is  acceptable for
disposal.  This is accomplished with rotary drum vacuum filters which
convert the slurry into a cake containing approximately 50 to  607o by
weight solids, with the remainder being water.   These  filters  also have
provisions for "washing" the cake to remove as much of the soluble sodium
compounds as possible before disposal.  The filter cake is composed of
calcium sulfite, calcium sulfate (gypsum),  calcium carbonate (limestone),
and the particulate removed from the flue gas by the scrubber.           *

The filter cake discharges to an automated conveyor system which alter-
nately fills waste product containers for off-site disposal as landfill.
The filtrate separated from the cake by the vacuum filters is  recycled
back to the thickener to be reused in the system.  Thus,  the process
operates in a "closed loop", which means there is  no process solution
discharged from the system other than that contained in the filter cake.
                III.  PERFORMANCE CHARACTERISTICS
DESIGN BASIS
The Zurn Industries, Inc. system installed at Caterpillar Tractor Co.'s
Joliet, Illinois plant cleans the flue gas from the  two stoker fired
boilers which produce steam for building  heat and  process needs.   These
boilers are rated at 80,000 and 100,000 Ibs. of steam  per hour and each
is equipped with a mechanical dust collector.  Total coal firing  rate to
these boilers is 9.65 tons per hour and the process  is designed to allow
the boilers to operate within the allowable emission rate of 1.8  Ibs. of
S02 per million BTUs heat input.  System  design parameters are summarized
below:

     No. of Boilers                   2
     Firing Method                    Stoker
     Total Steam Capacity             180,000 Lbs./Hr.
     Total Gas Volume                 103,500 Cu.  Ft./Min.
     Coal Sulfur Content              4%  As Received
     S02 From Boilers                 6.74 Lbs./mmBTUs
     Allowable S02 Emission           1.8 Lbs./mmBTUs

The total process design, engineering,  construction, and  start-up of this
system were accomplished by Zurn Industries, Inc.  Design,  construction,
and start-up were completed on,  or ahead  of schedule.   The chronology of
system design and installation was as  follows:
                               506

-------
     1.   Feasibility investigations                     15 Months
     2.   System design                                  12 Months
     3.   Equipment installation started February 1, 1974
     4.   Component check-out started August 12, 1974
     5.   First scrubber put in service September 25, 1974
     6.   Second scrubber put in service October 2, 1974

EQUIPMENT - PROCESS PERFORMANCE

The scrubbers, breeching, stack, and scrubber effluent piping are  con-
structed of 316 stainless steel.  The wet surfaces of the vacuum filters
are 304 stainless steel.  All other process piping and process tanks are
carbon steel.  The lime mix tanks are lined with a Urecal polyurethane
coating.

System start-up and preliminary testing was accomplished from September
25 through December 16,  1974.  The major reasons for downtime during this
period were minor mechanical equipment problems typical of system start-
ups, and the accumulation of unslaked lime and grit in the lime mix tanks.
This accumulation was due to poor slaking characteristics of the delivered
lime, high insolubles content, and some air-slaking of the lime in the
storage silo.  Accumulation rates were quite dramatic.  Approximately 10
feet of sand, gravel, and unslaked lime were deposited in the primary
lime mix tank in one ten day period.  This problem was solved by the
installation of a lime  slaker which  improved lime  slaking and had pro-
visions for separating  sand and gravel from the lime before it entered the
system.  The slaker installation was accomplished  during the scheduled
Christmas  1974 shutdown.

Design revisions were completed and  the system was put back on line January
23,  1975,  and operated  continuously  until June 1,  1975.  Process performance
and  reliability during  this period was quite good.  One of the operation
highlights observed during  the  prolonged period of operation was good system
control over a wide range  of  system  loads.  The scrubber  pressure drop and
pH were very  stable, easily controlled, and were  not adversely affected by
rapid boiler  load changes.  No  significant accumulation of solids was
observed in  the scrubbers which would  impede their efficiency or operation.
No measurable corrosion or  erosion was observed in any of the process
vessels, scrubbers, or  piping.

Various mechanical problems with  process auxiliaries were encountered during
this period  and some of these  are illustrated  below.

Chemical Feed System

The  lime and  soda ash  are  metered to the process  from  outside silos by
screw conveyors.  Two  problems  were  encountered here.   One was the  relia-
bility  of  components  in the electronic  control system  for the conveyor
drives  and the other was due  to formation  of  lumps inside the silos caused
by moisture  from  rainwater leakage and condensation on the inside tank walls,


                                   507

-------
Both of these problems cause inaccurate feed rates which in turn caused
undesirable deviations in system chemistry.  The conveyor drive control
systems were made operational by replacing faulty electronic components.
Inaccurate feed rates due to lumps of material forming inside the silos
was a continuing problem throughout this period of operation.  A dry air
purge system was installed on the lime and sodium carbonate storage silos
to prevent material degradation and lump formation due to moisture.  This
has effectively prevented lump formation in the storage silos,  and dry
material flow from the silos has been reliable.

pH Probe Reliability

Accurate pH indication is essential to maintaining good system control,
and preventing scaling and corrosion.  The pH probes as initially installed
were quite susceptible to coating and fouling which resulted in erroneous
readings.  Several methods were investigated to alleviate this problem and
reliable pH monitoring was finally achieved by the installation of parallel
probes with fresh water back flush and ultrasonic cleaning.

Sulfur Dioxide Monitoring

Stationary probes are installed in the scrubber inlet duct and in the
stack to provide continuous monitoring of flue gas for SC>2 concentration.
The  S02 analyzer was not operational during initial system check-out due
to shipping damage.  After damaged components were repaired,  the analyzer
was  calibrated and put in service.  Sample probe pluggages. which rendered
the  S(>2 analyzer inoperable, became a recurring problem.   Two system
revisions were made to improve probe reliability.  The pressure of the
blow-back air to the probes was increased to improve probe cleaning and
shields were installed around the probes.   These revisions did minimize
plugging but after a period of continuous operation the shields deteriorat-
ed.  This problem was not resolved during the first operating season before
summer shutdown.  Operation of the analyzer system has been good during the
1975-1976 heating season after revisions were made to the probe assembly
and purge cycle by the instrument supplier.

Water Balance

The process is designed for closed loop operation,  and therefore should
have zero scrubbing solution discharge during steady state operating con-
ditions.  Extended periods of zero discharge were not achieved during the
system start-up period for the following reasons:  (1)  Excessive water
was added to the system thru process pump seals and vacuum filter belt
wash sprays.  (2)  The boilers were often operated at low loads,  which
resulted in low water evaporation rates in the scrubber.

The loss of scrubbing solution was alleviated by installating rotometers
on process pump seal water supply lines and by using clarified scrubber
feed  solution for vacuum  filter belt wash.  The system requirement for
fresh water was  significantly reduced  by the  installation of  a  seal water
                                   508

-------
recirculation system on the vacuum pumps.  Closed-loop operation has
been satisfactorily demonstrated  since these modifications have been
made.

Freeze Protection

The system as originally installed, did not provide adequate protection
from freezing temperatures.  Frozen process liquor lines and control air
lines contributed to varying degrees of abnormal operation.  Installation
of additional insulation and higher quality heat tracing has provided
reliable freeze protection.

Feed Water Pump

Feed water pump impeller damage due to cavitation was experienced as a
result of insufficient suction head in the surge tank which serves as a
reservoir for the scrubber feed water pumps.  The surge tank height was
increased to provide additional system surge capacity and to provide
greater suction head to the scrubber feed water pumps.  This modification
has been very effective in preventing cavitation.

Vacuum Filter Belts

The system has two vacuum filters,, each capable of handling the total system
solids load.  The filter belt originally used was woven of multifilament
polypropylene.  Belt life was poor due to the accumulation of solids in
the interstices of the cloth.  It was thought that the belt wash sprays
were ineffective, so higher pressure water was piped to the sprays.  To
date, this modification appears to have prolonged belt life, but other
belt materials and cloth weaves are being tried in an effort to extend
belt life.

The system was put in service on  October 24, 1975, for the 1975-1976 heating
season.  After approximately two  days of operation, the sludge pump impellers
began plugging with pieces of compacted sludge that had the consistency
of stiff clay.  The entire process was idle from June until October, and
apparently the solids that could  not be removed by filtration during shut-
down had compacted on the clarifier walls and rake mechanism.  When the
process was restarted, this material fell from the metal surfaces, entered
the sludge pump, and greatly reduced pump capacity.  This situation had
never been encountered prior to the first summer shutdown, and a revised
shutdown procedure will prevent a problem of this nature in the future.

The lime slaker required considerable maintenance and attention during
the first operating season, so hydrated lime was substituted for pebble
lime to decrease the maintenance  requirement on this equipment.  Operation
with hydrated lime has been good  from a maintenance standpoint, but the
hydrated lime has different transport characteristics from the pebble lime,
and control of lime feed rate has been poor.  Modifications to the volumetric


                                   509

-------
lime feeder have greatly improved lime feed control.

Miscellaneous Hardware Performance

Most of the auxiliary equipment and controls have  given  good  service and
have performed adequately.  Operating experience and  familiarity with
the system hardware has improved component reliability through a better
understanding of equipment capability and improved preventive maintenance.

Rubber lined diaphragm valves have given excellent service in the process,
and have replaced ball valves in some areas.

The radiation type density meters intended to indicate clarifier underflow
slurry density were removed due to drift.  Slurry  density is  now monitored
by the operator.

Addition of a polyelectrolyte flocculant has yielded  the expected favorable
results.  Clarifie-r overflow clarity is excellent, slurry filterability
has improved, and preliminary results indicate that the  cake  solids content
has been increased from about 541 to around 62% by weight.

The automated sludge conveyor which alternately fills waste containers,
has performed well with little operator attention.  Periodic  cleaning  is
required to remove sludge from idlers during extremely cold weather.

The Dustraxtor  scrubbers, Figure 2, have been reliable and stable even
during  rapid boiler load fluctuations.  The entrainment  type scrubber  has
shown no tendency to scale or plug under any operating conditions.  Some
soft  solids accumulation has been found at the liquid surface in the
lower portion of the scrubber, but this does not increase pressure drop or
decrease operating efficiency.  The solids appear to dissolve away at  a
rate  which prevents any significant accumulation.   The demisting section
of  the  scrubbers is bare-metal clean with no solids accumulation whatever.

The difficulties with  the lime slaking and  lime feed control systems caused
some  scale  formation in the lime mix tanks during the first operating  season
Improved process control has eliminated this problem.

                IV.  PROCESS EFFICIENCY AMD ECONOMICS
 PROCESS EFFICIENCY

 A comprehensive system performance  test program has just been completed,
 and only preliminary results  of  these tests are available.  Preliminary
 performance tests conducted by using the installed S02 analyzer and
 independent wet chemistry  analyses  indicated that S02 removal capability
 of the process is in excess of 85%.  Particulate removal capabilities are
 yet to be determined.

                                  510

-------
PROCESS ECONOMICS

A prime reason  for selecting  the  double alkali  system was that it afforded
an economic advantage over  the use  of  low  sulfur coal or conversion to
low sulfur oil.  Precise cost determinations are difficult due to
variable costs  of raw materials and associated  services, but the best
current estimates of system operating  costs are shown below:

     Basis :     4% S coal
              75% S02 removal required                   Per Ton
                                                          of Coal

       Chemicals (lime and  sodium carbonate)              $2.56
       Water                                               0-02
       Electricity                                         0.71
       Waste                                               0.64
                                      Subtotal                        $3.93

       Labor  (operation and maintenance)                  $2.12
       Maintenance                                         0.38
       Depreciation

       Estimated Total Cost Per Ton Coal


                           V.  SUMMARY
The Zurn Double Alkali System at the Joliet plant of Caterpillar Tractor Co.
was an advance ment in the technology of flue gas desulfurization and involved
certain unknowns.  However,  the system was made operational on schedule and
has encountered few problems, none of which haven't been effectively solved.
The process is currently in  its second operating season and reliability
continues to steadily improve.  The system has completed approximately 5600
hours of operation, including 2500 hours of satisfactory continuous operation
prior to planned summer shutdown.

System development and optimization are continuing to maximize the potential
of the process.  In the near future, the remaining minor mechanical problem
will be resolved and more data will be available on system performance.
                                 511

-------
                           ZURN DOUBLE ALKALI
                        DESULFURIZATION PROCESS
                    RAW MATERIALS - UTILITIES USAGE
LIME

SOM ASH

WATER

ELECTRICITY

WASTE DISPOSAL
PER HOUR

1040    LB.

 285    LB.

  36    GAL.

   6.85 KWH

5200    LB.
PER TON COAL

  108    LB.

   29.5  LB.

    3.7  GAL.

    0.71 KWH

  539    LB.
PER 1000 LB.  STEAM

    5.78 LB.

    1.58 LB.

    0.2  GAL.

    0.038 KWH

   28.9   LB.
                    BASIS:  180,000 #/HR.  STEAM GENERATED

                               9.65 T/HR.  COAL  FIRED

                                 4% SULFUR COAL

                                75% S02 REMOVAL REQUIRED
                                   512

-------
             FRESH WATER
BOILER
 NO. 2
                                                 DRY CHEMICAL
                                                  TRANSFER
_^|VACUUM
rc°-RECEIVE
            VACUUM
             PUMP
VACUUM
FILTER
 SPARE
         EXISTING
        COLLECTOR
 VACUUM
RECEIVER
  SPARE
BOILER
 NO. 3
                  DUSTRAXTOR
                                                                                   VACUUM
                                                                                     PUMP
                                                                                    SPARE
                                              CONTROL TANK
         EXISTING
        COLLECTOR
         WASTE PRODUCT
           (10) PICK-UP
           CONTAINERS
                                FIGURE   ZURN DOUBLE ALKALI SYSTEM

-------
    I
       FIGURE 2




ZURN DUSTRAXTOR SCRUBBER
        514

-------
                  KUREHA FLUE GAS DESULFURIZATION
                  "SODIUM ACETATE-GYPSUM PROCESS"
       Shigeru Saito, Toraijiro Morita, and Shigeyuki Suzuki

                Kureha Chemical Industry Co., Ltd.
               1-8 Nihonbashi Horidome-Cho, Chuo-Ku,
                           Tokyo, Japan
ABSTRACT

     This paper deals with  a new sodium acetate-gypsum desulfurization
process.  This process  is a complete  closed  loop with a S0? removal
efficiency of more than 99% using  limestone  or lime as regenerants
and can be applied to flue  gases from oil  fired as well as coal fired
boilers.

     Furthermore, a new simultaneous  removal of S02 and N0_ is also
described.
                              515

-------
              KUREHA FLUE GAS DESULFURIZATION
             " SODIUM ACETATE-GYPSUM PROCESS "
1.  INTRODUCTION

     Several years ago, Kureha first developed a sodium
sulfite process as a single alkali process to produce sodium
sulfite for the pulp and paper industries.  Then, based on
this development, Kureha has further developed a sodium
sulfite-gypsum double alkali process in cooperation with
Kawasaki Heavy Industries, Ltd.  The first full scale plant
of this process has been in operation at Shin Sendai power
station of Tohoku Electric Power Co., Inc. since April 1974.

     Recently, it has also been proven by two full scale plants
of Shikoku Electric Power Co., Inc. having around 740,000
st.cu.ft./min. of flow rate.

     Through this experience, we have learned that this
double alkali process is superior to other existing processes
with respect to SOp removal efficiency, operability and
economy.

     However, the processes that we have developed, have been
considered mainly for flue gases from oil fired power plants
using sodium hydroxide or calcium carbonate as feed materials
to produce sodium sulfite or gypsum as by-products.

     On the other hand, there are strong demands for the
removal of S02 and particulate in the flue gas from coal fired
boilers as well as flue gas from sinter plants of steel mills.
                               516

-------
     Also, lime can, in some cases, be utilized more advan-
tageously than can lime stone.  There is a strong incentive
to develop a process to meet the above requirements.

     Key differences in the nature  of flue gases from coal
fired boilers lie in the higher oxygen content, particulates
and chlorine when compared to  flue  gases from  oil fired
boilers.

     For  the development  of  a  new  process, these factors
have been taken  into our  serious consideration and  acetic
acid has  been adopted  as  the best  absorption liquor to
satisfy the basic requirement  for  a scrubber absorbent:

1.   It should give a  weak alkaline solution with  large
     buffering capacity,  when its  aqueous  solution is
     neutralized by strong base.

 2.   It should  be a weaker acid than sulfurous acid, but
     stronger acid than carbonic acid.

 3.   It should have high solubility in water, and the vapor
     pressure should be as low as  possible.

 4.   Solubility of its calcium salt in water  should be large,

 5.    It should be  easily available in the market and the
      price should be reasonable.
                               517

-------
     Acetic acid thus selected, has been utilized success-
fully as an efficient absorbent in the process development
through stepwise tests from 60 st.cu.ft./min. bench scale
test to 5,000 st.cu.ft./min. pilot scale test for the flue
gas from oil fired power stations.

     Additionally, we have carefully investigated the
feasibility for flue gas from a coal fired boiler using
60 st.cu.ft./min. test facility.

     Through a series of tests, we could overcome problems
that had been encountered in the cleaning of flue gases  from
coal fired boilers.  We have also obtained the necessary
information for the scale-up of this process.

     We are confident that this process is superior in
operability and economy to the other known processes.  Also,
in the present situation in Japan, hydrated lime can be  used
advantageously as a regenerant for this process over lime
stone from overall process economy.

     We are now in the planning stage of several full scale
plants in Japan to demonstrate the superiority of this
sodium acetate gypsum process using either limestone or
lime for flue gases from oil fired as well as coal fired
boilers.

     We wish to report the results of our process develop-
ment of the sodium acetate-gypsum process. 1)  2)
                             518

-------
2.  PROCESS DESCRIPTION

2-1.   Lime Stone Method

     As shown in Figure I, this process is composed of three
basic sections; absorption,  oxidation and gypsum formation.
This process has been tested at 5,000 st.cu.ft./min. pilot
test facility for the flue gas from  our oil fired power
station from March 1975 through August 1975 with continuous
and successful operation.

2-1-1.   Absorption Process   As  illustrated in Figure 1,
the absorption tower consists of  two sections - S02 absorp-
tion and acetic acid elimination  section - , which are linked
in series in one absorption  tower.   The lower section is
composed of two absorption chambers  and the higher acetic
acid elimination section consists of three chambers, each
being equipped with perforated plates inside of the chambers.

     In the S02 absorption section,  S02 is removed by circulating
a sodium acetate solution through the two stages.
The following reaction takes place:
     2CH,COONa + S02 + H20 -» Na2S03 + 2CH3COOH          (l)

     A part of the acetic acid thus farmed volatilizes in
the scrubbed flue gas, and must be eliminated in the next step

     In this step, fresh lime stone slurry is added to the
top chamber, and flows down count ercurrently from chamber to
chamber to the bottom to accomplish the complete removal of
acetic acid vapor.  Also, this step serves to complete the
                              519

-------
elimination of remaining SC>2 in "the flue gas from the
absorption section, so that clean exhaust gas is released
into the atmosphere from the top of the scrubber.

     For the smooth operation of the scrubber, prevention
of scaling is essential.  Thus, the following four measures
have been taken against scaling prevention.

1.   The structure of the scrubber has been designed in
     such a way as ;there can be formed no dry parts inside
     the scrubber walls.

2.   Perforated plates have been used successfully in
     preventing scaling, because vigorous mixing of flue
     gas and absorption liquor through the plates have
     resulted in both preventing scaling to and removing
     scaling from the walls or the plates.

3.   Seed crystals have been added successfully to prevent
     scaling by the possible precipitation of calcium
     sulfate in the absorbent.  Especially, this precipita-
     tion tends to take place,when the solubility of calcium
     sulfate in the absorption liquor becomes lower at the
     outlet of the absorption section than at the inlet.

4.   When the concentration of sodium sulfite is relatively
     high in the liquor, this reacts with calcium sulfate
     in the liquor to form insoluble calcium sulfite, which
     may cause a scaling problem.   Accordingly,  it is
     necessary to keep the concentration of sodium sulfite in
     the absorption liquor lower than 2$ as can be understood
     from Figure 2.   This can be successfully accomplished in
                              520

-------
     our process by diluting the scrubber absorbent with
     a part of oxidized liquor from the oxidizer.

     By the application of these measures, we have achieved
the prevention of scaling in the scrubber and at the same
time a high efficiency of S0? removal and acetic acid
elimination for a variable loadings of gas flow rate as
summarized in Table 1.

2-1-2.  Oxidation Process   Sodium sulfite formed in the
S02 absorber is oxidized to sodium sulfate in the oxidation
tower which is provided with perforated plates to facilitate
fine dispersion of air bubbles and to promote oxidation.
Sulfite oxidation to  sulfate takes place as  follows:

        S0  + 1/2 0   -> NaS0                       (3)
     After  oxidation,  the liquor is sent to the  gypsum
formation section,  where gypsum is produced by the  addition
of lime  stone  slurry.

2-1-3.   Gypsum Production Process   In this step,  gypsum
is produced from the oxidized liquor containing  mainly
sodium sulfate and  acetic acid by the addition of lime
stone slurry.   In this gypsum formation, calcium carbonate
reacts very rapidly with sodium sulfate in the presence  of
acetic acid, whereas the reaction of calcium carbonate
with sodium sulfate is very slow in the absence  of  acetic
acid.

     In  this process, calcium carbonate is assumed  to react
at  first with acetic acid present in the liquor  to  form
                            521

-------
calcium acetate, which reacts further with sodium sulfate
to form calcium  sulfate and sodium acetate by a double
decomposition reaction.  The regenerated sodium acetate is
recirculated to  the  scrubber after separation from gypsum
by filtration.

     The reaction  can be expressed as in (4) and (5):
              2CH5COOH -> (CH3COO)2Ca + R^Q + C02    (4)
     (CH3COO)2Ca + Na2S04 -» CaS04 + 2CH3COONa      (5)
     Gypsum produced in this step has such chemical composi-
tion as represented in Table 2.

     This gypsum is high in quality and is utilized in wall
board making and in Portland cement production.

     Thus, use of acetic acid as an absorbent has made it
possible to remove S02 efficiently from flue gas, producing
high quality of gypsum as a by-product.

     Also, rapid reaction of sodium sulfate with calcium
carbonate has made this process further possible to apply
it to flue gases with high oxygen content such as flue gases
from coal fired boilers, which will be mentioned in the latter
section in detail.

2-2.   Acetate-Lime Process

     In order to  make a comparison between lime stone and
lime, tests have been conducted at 60 st.cu.ft./min.  bench
test facility to compare the date with those obtained by
lime stone process.

                             522

-------
     In principle, acetate-lime (hydrated lime) process has
the same structure with the acetate-limestone process as
given in Figure 3.  The absorbent containing sodium acetate
is in a similar way transferred from the absorption tower
through the oxidation process  to the gypsum formation step,
where lime is added instead of lime stone to produce gypsum.
This reaction takes place  more rapidly  than with lime stone
even in a very dilute lime suspension.  Also ,  the mother
liquor obtained after separation from gypsum crystals shows
a higher PH value  than  in  the  case  of lime stone process.
These two features make  it possible to  operate  the acetate-
lime process with  advantage  over lime stone  process  especially
in terms of removal efficiency of  SC>2 and  acetic acid removal
in the scrubber.

     At the same  time ,  concentration  of the  absorption  liquor
can  be reduced  considerably  to minimize the  possible loss  of
acetic acid by  the oxidation.

     Through  these experiments, we have found that the
scrubber unit  for S02 can be  simplified from two absorption
sections to  single section and as to acetic acid vapor
removal  from  three to one section, respectively.  Also,
make-up  of acetic acid has been reduced considerably.

      In the  simplified scrubber, the elimination section
 of acetic  acid by aqueous lime solution played only a
 supplementary role, because most of acetic acid vapor  in the
 flue gas had been eliminated  in the first S02  absorption
 section by the absorption liquor of high pH value (7.0 - 7.2).
 Figure 4 supports the above fact,  in which  relationships
 between pH values of the  liquor and concentrations  of  acetic
                               523

-------
acid in the gas phase are shown.  For a Vs- range of 7.0  to
7.2 with a temperature of 131°F, concentration of acetic acid
in the vapor phase can be reduced as low as 3 parts per
million.  Encouraged by the above test results, we are
planning to prove the operability and economy of this lime
process by 3,000 st.cu.ft./min. pilot plant in the near  future
 3.    DESULFURIZATION OF FLUE  GAS  FROM COAL FIRED  BOILER

      Technogical requirements for the desulfurization of
 flue gases from coal fired boilers are  increasing from the
 standpoint of environmental protection  especially in other
 countries than in Japan such  as in the  United States or in
 Germany, where they have big  coal reserves as an. energy
 source for power generation.   Also, an  increase in power
 generation from imported coals can be also anticipated in
 the future in Japan.

      In respect to these considerations,  we  have made a train
 of desulfurization tests for  flue gas from coal fired boiler
 at the 60 st.cu.ft./min. test facility  using sodium acetate-
 lime stone process in order to solve problems, which may
 encounter in the desulfurization of flue gases from coal
 fired boilers.  These tests were carried out in August of
 1975, and the results are reported in the next section.

 3-1.   Sodium Acetate-Lime Stone Process

      Following test items have been considered for the
 evaluation of the process.
                               524

-------
1.    Efficiency of desulfurization.

2.    Efficiency of particulate removal.

3.    Quality of gypsum.

4.    Construction material.

5.    Material and heat  balance.

6.    Behavior of trace  materials  and their accumulations in
     the  system.

7.   Confirmation of the system for a closed  loop without
     disposal water.

     A  flue  gas containing 600 parts per million of  S02 and
1.4xlO~5  to  2.8xlO""5 gr./st.cu.ft. of particulate was
introduced into scrubber with an inlet gas  temperature of
above  100°C.

     As to the efficiency of S02 removal, nearly complete
removal of S02 has been achieved as listed in Table  3.

      Accordingly, as far as efficiency of desulfurization
 is concerned, the experimental results are excellent.

      As to the removal  of particulate in the flue gas,  tests
 have been made for  variable contents of particulate and
 different numbers of perforated  plates in the scrubber.
 The results are shown  in Table 4.,
                                525

-------
      In these tests, flue gas containing a. high level of
 particulate (more than 0.14 gr./st.cu.ft.) has been prepared
 by adding fly ash to the flue gas to determine the  removal
 efficiency,  (s. Exp. No. 9 and 10 in Table 4.)

      As seen in Table 4, efficiency of particulate  removal
 is remarkable even to the high contents of particulate and
 a distinct relationship can be seen between particulate
 removal efficiency and the number of perforated  plates used.
 As to particulate elimination,  better results  can be
 anticipated for the scrubber with 12 perforated  plates than
 with an electrostatic precipitator.

      During the above tests,  no noticeable  troubles have
 been experienced with particulate  or chlorides even in the
 case of high particulate  loadings.

      The quality of gypsum formed  was satisfactory for
 further processing,  even  if it  had colored gray to light
 brown depending on  the  operational conditions.

     No  influence has been  observed in the crystal growth
 of gypsum,  size  of which ranged from 100 to 150 microns in
 the  long axis and 50 to 60 microns in the short axis.

     As  to materials of construction for the test,  only
 corrosion-resistant metals and plastic materials were used,
 so that no abnormality has been observed during the  tests.

     Regarding the material and utility balance of  this
process, a comparison was made between designed values and
experimentally obtained values using the 60 st.cu.ft./min.
                             526

-------
bench scale test facility.  Comparative data are listed in
Table 5.

     As can be seen from Table 5, smaller experimental
values than designed have been obtained as the material
balance for lime stone and gypsum, which can probably be
ascribed to smaller S02 content  in the flue gas than
designed at the early stage of the test.  Otherwise, an
excellent coincidence between experimental and projected
values have been attained both for material and utility
balance.

     Regarding impurities such as particulate and chlorine,
tests have been made extensively.  Acid insolubles  in the
gypsum were analyzed and found to be  composed mainly of
silica and alumina, which amount to about 4$ of the gypsum.
This suggests that the particulate in the flue gas  is
removed from the system together with gypsum.  Thus, in
our tests no build-up of particulate  has been observed
during operation.

     As to chlorine content  in  the flue gas, measurements
showed chlorine content of  0.7xlO~5  to l.0x!0"5gr./st.cu.ft.
No operational  troubles have  been  caused by  this  level of
chlorine content,  and also,  no  influence has been observed
upon the absorption of  302  and  acetic acid  in  the scrubber,
the rate of oxidation and  of gypsum  formation.   However,  if
a  closed system is adopted,  accumulation  of  chlorine will
take place in the  circulating liquor, and  must  be eliminated
from the circulating  system.   Extensive research is now
being made by us  to  solve  this  problem.
                              527

-------
     As far as this bench scale test was concerned, a
complete closed loop has been successfully demonstrated with-
out waste water disposal.  In this bench scale test, make up
of acetic acid and sodium hydroxide were not necessary.

     Besides the above items mentioned, acetic acid vapor
in the flue gas from the S02 scrubbing section has been
measured both by gas chromatography and by gas detection
tube.  The former method gave an acetic acid concentration
of 1.9 to 3-5 parts per million in the outlet gas of the
scrubber, while the concentration was Less than i part
per million for the same samples by the latter method,
showing practically complete removal of acetic acid vapor
in the flue gas.

     As to pressure drop, a total pressure drop of 11 in.
H^O has been  recorded for this process, which corresponds
to a mean drop of 2.17 in. f^O for each chamber of the
scrubber.  (The scrubber consists of 2 chambers of SOp
absorption and 3 chambers of acetic acid elimination).

     For each perforated plate, a mean pressure droi> amounts
to 0.51 in. H20.  This magnitude of pressure drop for
perforated plate can guarantee good distribution of flue gas
in the large  scale scrubber.  This was confirmed by another
separate experiment by using test facility of 18,000 st. cu.ft,/min.
gas flow rate.  Smooth flow down of the liquor and good
dispersion of flue gas into the counter-current liquor through
the perforated plates were observed in the above test.

     To determine the possible trouble in the acetic acid
eliminator by the mist of the absorbent from the absorption
                             528

-------
section, careful observation has  been made, but no
troubles have been found.

     Scaling in the scrubber was  prevented effectively by
the measures previously noted, and therefore no scaling
problem has been raised during the test.

     Through a series of tests made on the flue gas from
coal fired boiler, we have  succeeded in collecting necessary
information for the scale-up of this process, and it will be
soon demonstrated to verify the process by 3,000 st.cu.ft./min.
pilot plant, followed by demonstration of full scale plants.
4.   ECONOMY OF THE PROCESS

     On the basis of our  experience and on a series of
bench and pilot tests, process economics can be estimated
with good accuracy.

     In Table 6, a standard desulfurization cost of sodium
acetate-lime process for  the flue gas rate of 588,500
st.cu.ft./min. is listed  for 1976 costs in Japan.  This
cost should be considered as a measure of our sodium
acetate - lime process, because it is strongly dependent
on the impurity content of raw materials and on the dealing
of gypsum, whether it is  disposed or utilized.  In case
gypsum is thrown away, process can be simplified and both
the investment cost and utilities cost can be reduced,
resulting in the reduction of desulfurization cost.

     It should be further noted, that desulfurization
cost is greatly influenced by local conditions.  However,
                            529

-------
as can be deduced from Table 6, this sodium acetate-lime
process is low in desulfurization cost and excellent in
operability.
5.  SIMULTANEOUS REMOVAL OP S02 AND NOX BY SODIUM
    ACETATE-GYPSUM PROCESS.

     Taking advantage of sodium acetate-gypsum process, we
have further developed a process to eliminate S0« and NO
                                                ^       .A.
simultaneously.  NO  removal is furnished in the above
system by the addition of a catalyst, which is commonly
available in the market for a reasonable price.
     In this process, NO  is.reduced in the presence of the
catalyst to sodium imidodisulfonate by the sodium sulfite
in the absorbption liquor, and sodium imidodisulfonate
formed is further decomposed by calcium nitrite and lime in
the presence of sulfuric acid to nitrogen and gypsum.
As shown in Figure 5» a part of mother liquor containing
sodium imidodisulfonate, is drawn as a side stream from the
recirculation line to the scrubber and then, it is decomposed
by the above chemicals to nitrogen and gypsum in the reaction
vessel.
     Accordingly, this system can eliminate S0~ and NO
                                              £,       -A.
simultaneously by additional attachment of NO  removal
section to the sodium acetate-gypsum process in a complete
closed loop without no disposal water.
     This process has been proven by 3000 st.cu.ft./min.
pilot test and based on this successful results, we are
planning to build a-full scale plant in the near future.
                             530

-------
6.  CONCLUSION

     Based on long experience of research and development
in the field of flue gas desulfurization, we have finally
come to sodium acetate-lime process for the treatment
of flue gases from both coal and oil fired boilers, process
economics can be assumed to be  the most advantageous one.
Furthermore, we have recently developed a simultaneous
removal of SOp and NO  process  on the  basis of the sodium
             £—       A.
acetate-gypsum process.
Though these new systems are still on  the way of pilot test,
we are sure that these systems  will be demonstrated in a full
scale plant in the near future  as one  of the most favorable
process for various type of flue gases containing S0? and
particulate.
                              531

-------
                          References


1)  T.  Xauiinaga and K.  Noguchi
              " Kureha  Sodium Acetate-Gypsum FGD Process  and  Plant "
                Chemical  Factory  (Japan)  19, 5, 85-88  (1975)
2)  S.  Saito
              "  Kureha Sodium Acetate-Gypsum FGD j rocess  "
                Chemical  Engineering  (Japan)  20, 3, 17-19  (1975)
                                532

-------
   Table 1,   SOp REMOVAL EFFICIENCY BY SODIUM ACETATE-GYPSUM PROCESS
              AT 3000 st.cu.ft./min. PILOT  TEST FACILITY.
C/l
C/4
Gas Flow Hate
Exp.No. (st.cu.ft./min.)
1
2
3
4
2940
2350
1470
940
S02
Concentration
Inlet (parts/mill ion)
1500 -
1500 -
1500 -
1500 -
1700
1700
1700
1700
Outletl
less
less
less
less


SOp Removal
'parts/million) Efficiency
than
than
than
than
1
1
1
1
more than
more than
more than
more than
99.
99.
99.
99.
9
9
9
9

-------
Table 2,   LIME STONE AND GYPSUM QUALITY

Lime Stone

Particle Size
     200 mesh pass

Chemical Composition
     Si0
Gypsum
     CaO                         54.
     MgO (as Mg)                 0.3
     Free Water                   5.59^ (Wet base)
     Crystal Water               20.755?- (Dry base)
     CaO                         30
     SO,                         45.
     Purity                      96.88^- (as CaS04.2H?0)
     Crystal Size                50 ;i x 100 ji
                              534

-------
Table 3,   REMOVAL EFFICIENCY OF S02 BY SODIUM ACETATE-LIMESTONE  PROCESS
           IN 60 st.cu.ft./min.  BENCH SCALE TEST  FOR FLUE GAS  FROM
           COAL FIRED BOILER.
Exp. No.
1
2
3
4
5
6
Concentration of c
Gas Inlet
646
690
570
550
570
580
jOp (Darts/million)
Gas Outlet
0
1
0
1
0
2
Desulfurization
100.0
99.8
100.0
99.8
100.0
99.6

-------
Table 4,   RELATIONSHIP BETWEEN NUMBERS OF PERFORATED
           PLATES AND PARTICULATE REMOVAL EFFICIENCY FOR A
           FLUE GAS WITH A FLOW RATE OF 60 st.cu.ft./min.
Exp.No.
1
2
3
4
5
6
7
8
9
10
11
No. of
Plates
Used
3
3
3
5
5
12
12
12
12
12
12
Amount of Particulate
( gr./st.cu.ft. x 103)
Inlet
1.50
2.66
2.95
3.31
3.28
3.93
5.83
4.50
2.71
209.1
205.6
Outlet
0.90
1.02
1.16
0.88
0.85
0.48
0.51
0.23
0.28
2.04
1.47
Efficiency
of Removal
(*)
40
62
61
74
74
88
91
95
90
99
99
                               536

-------
   Table 5,    I'lATERIAL AND UTILITY BALANCE OF SODIUM ACETATE-LIMESTONE PROCESS
Gas Composition



    S0?         Inlet



    S0?         Outlet



    Acetic Acid Outlet





Haw Material
0-J
   By product



       Gypsum





   Utilities
       Process tfater



       Steam



       Power



       Air
                                        Experimental






                                   580-630 parts/million



                                   0-2       "



                                   1.6-3.5
0.55 lb/hr.










0.88 lb/hr.










0.42 gal./hr.



44.1 lb/hr.



10 kWh



141.2 st.cu.ft/hr.
                                      Designed
                                  600 parts/million



                                    1     "



                                    3
                                                                  0.59 lb/hr.










                                                                  1.01 lb/hr.









                                                                  0.42 gal./hr.



                                                                  176.3(max.)-22.0(min.)lb/hr,



                                                                  40 kWh



                                                                  141.2 st.cu.ft./hr.

-------
Table 6,   ECONOMIC EVALUATION OP SODIUM ACETATE-LIME
           PROCESS

    Flue Gas Volume:                588,500 st.cu.ft./min.
    S02 Content of Flue Gas:        1,000 parts/million
    Efficiency of Desulfurization:  more than 99^
    Operating Hours:                8,400 hrs./year.
    Investment Cost ,(BL):           15,800 ¥/KW.
    Cost of Desulfurization:        12.57 ¥/gal.
Raw Materials and Utilities
    Hydrated Lime:                  7,429.5 Ib/hr.
    NaOH :                          35-3 Ib/hr
    Acetic Acid:                    201.6 Ib/hr.
    Steam:                          8,766.2 Ib/hr.
    Process Water:                  13,210.0 gal/hr.
    Electricity:                    4,500 KW
    Gypsum:                         172,840 Ib/hr.
Man Power
    2 Men/Shift,  4 Shifts/day + 2 Men.
                            538

-------
C/n
W
1X5
             QH
          Keheater
     Acetic Acid
     Recovery
     >>ect ion
      :at i on
      '-T-t:on
   Flue Gas
                                    3_J
                                 Acetic  Acid
                                  Recovery T-mk
                         Absorbent  Tank
                Absorption
                Tower
  Oxida-
  t ion
  Tower
—Air
               Gypsum
               Recovery
               Reactor
                   Separator
                                                                                   •Lime  Stone
                                                                                   •Water
                                                                                        Slurry  Tank
                     Mother Liquor Tank
       FIGURE  1,   SCHEMATIC  FLOW  DIAGRAM  OF LIME STONE DESULFURIZATION PROCESS

-------
    FIGURE 2,    SOLUBILITY OF CALCIUM SULFATE IN THE ABSORPTION

                LIQUOR IN DEPENDENCE ON SODIUM SULFITE CONCENTRATIONS
o
3
&
•H
O

CO

C6

O




a
c:
o>
o
c
o
o
       0.5
       0.4
       0.3
       0.2
0.1
       0
                  Chemical Composition  of Liquor



                           CH5COONa


                           Na2S04


                           CHjCOOH


                           CaS04•2H?0
          0
           1
                      Concentration of
                                 540

-------
Acetic Acid
Recovery
Section
Desulfuri-
zation
Section
    Fan
                                FIGURE 3,    LIME  DESULFURIZATION PROCESS
                                 Ca(OH).
                              Oxidation
                              Tower
                                                           Liquid
                                                           Cyclone
fcQ
                                                                   Separator
                                                               A   N
—Q
                           Gypsum
                           Recovery
                           Reactor
                                                       ta.
           Absorption
           Tower
Air
Compressor
       Gypsum
               Mother
               Liquor
               Tank

-------
     FIGURE 4,   RELATIONSHIP BETWEEN P11 OP THE  LIQUOR AND
                  ACETIC ACID CONCENTRATION IN  THE VAPOR PHASE,
 o
 •H
 8

 m
 -p
 0.
 O
 O
 O
O
 C!
 O
 -H
 -P
 £0
 f-t
 -p
 A
 0>
 o
 fl
 o
 o

 Q)
 CO
 OJ
 ^3
 04

 0)
 ce
Ctl
                                           CH^CQONa SOLUTION
1   -
                                of Solution
                                      542

-------
                                       FIGURE 5,   SIMULTANEOUS REMOVAL OF S02 AND NOX

                                                  BY SODIUM ACETATE-GYPSUM PROCESS.
On
                             A
                   CaCO3
              Acetic  Acid
               Recovery
               Section
              SOX  and  NOX
              Absorption
               Section
                                       H20
                           Flue Gas
                                                                  Air
          Oxidation
            Tower
1 ------ -
                                           Ca(OH)2
   Ca(N02)2
                                           H2S04
                                               CaCO,
                         Steam
                                                    Nitrogen
                                                    Fo rmi ng
                                                                                       Gypsum Recovery
                                                                                           Reactor
                       Absorption
                             Tower

-------
            THE BUELL DOUBLE-ALKALI SO  CONTROL PROCESS
                          H. Edward Bloss

                         Buell-Envirotech
                       Lebanon, Pennsylvania
                           James Wilhelm

                         EIMCO-Envirotech
                        Salt Lake City, Utah
                         William J. Holhut

              Central Illinois Public Service Company
                       Springfield, Illinois
ABSTRACT

     This paper traces the development of the Buell Double-Alkali SO,
Control Process to the current  technology of concentrated and dilute
mode operation that are currently offered commercially.  Areas of
application are defined for the two modes,  each resulting in elimination
of chemical scaling, abrasive scrubbing  slurries and high maintenance
expenditure.  Design information is also presented for utilities
service with high chloride coals.  The Central Illinois Public Service
Company, Newton Station, Unit #1 (concentrated mode, high chloride
pulverized coal operation) currently under  design is presented and
equipment defined.  Preliminary capital  investment and operating
cost estimates are tabulated.
                                545

-------
INTRODUCTION

Scrubbing of sulfur dioxide from boiler flue gas has been
a major problem for the utilities industry due to costs,
complexities, and impact on reliability of electric gener-
ating operations.  An important solution to this problem
lies in continued advancement of process design technology
which will achieve required reduction in stack S02 emission
without reducing generating unit availability and without
creating secondary environmental pollution effects that
may jeopardize the quality of surface and ground water
bodies.

Because of variations in boiler installations and fuel
sources, a single scrubbing system design has not been
able to achieve the objectives of all applications.  For
that reason, Envirotech has invested more than four years
work in developing and evaluating the applicability of
new process design technology to control the various
operating variables encountered with operating boiler
units.
ENVIROTECH'S APPROACH

As a means of minimizing capital investment,  the utilities
industry continues to focus on flue gas desulfurization
systems of the waste "throwaway" type.   In view of their
comparatively low cost and their ability to precipitate
collected sulfur oxides as a water-insoluble  solid,
calcium-containing raw materials such as lime or limestone
are solely used to convert the collected S02  waste to the
solid form.  The basic double-alkali system shown on
Figure 1 accomplishes the above in two advantageous steps,
scrubbing first, post-precipitation later:

Equation 1 (Absorption)

  S02 + Na2SO3 + H2O                    2 NaHSO3

  Equation (la.)  SO2 + H2O             H2SO3

  Equation (Ib.)  H2S03 + Na2S03        2 NaHSO3

and/or

Equation 2 (Absorption)

  S02 + H20 + 2 NaOH                    Na2S03 + 2H20
                         546

-------
Equation 3  (Regeneration)

  2 NaHSOs + Ca(OH)2                  Na2SO3 + CaSOo.1/2 H90
                                                    +3/2 H20

and/or

Equation 4  (Regeneration of  fully oxidized collected
            sulfur oxides)

  Na2SO4 + Ca(OH)2 +  2  H2O            CaS04.2 H2O + 2 NaOH

All sodium-base double-alkali  systems perform generally
within these equations.

Beginning with  initial  work  five years  ago, Buell and Eimco
have committed  efforts  to the  advancement of the double-
alkali design approach  to SO2  control technology.  The
throwaway SO2 system  double-alkali  has  a special advantage
in avoidi-ng precipitation of the SO2  solids in the gas-
handling scrubbing system and  instead,  applies the lime
treatment externally  of a clear liquor  scrubbing circuit
that utilizes highly  water-soluble  sodium chemicals as
shown in equations above.

Envirotech's work in  double-alkali  flue gas desulfurization
has developed two similar  flow sheets that are now being
offered or  applied in commercial applications.  The first,
the dilute  mode process -  is specifically designed for
boilers operating with abnormally high  excess air or  for
low sulfur  coal applications such as in western U.S.A.
The second, the concentrated mode process - is used for
high sulfur coal  applications  in pulverized coal  fired
boilers where the excess air in the flue gas  is generally
low enough  to limit  the oxidation of the absorbed sulfur
species, as required  for its zero-effluent operation.


DILUTE MODE PROCESS

For power  plants  that have high excess  air  and/or  low
flue-gas  S02  concentrations, the oxidation  rate of the
absorbed  SO2  changes  the chemical equilibrium relation-
ships  in  the  system.   The major alkali in the process is
sodium hydroxide  not sodium sulfite and the  calcium  ion
concentration in  the regenerated liquor is  as high as
500 PPm.   in  order to control chemical scaling of the
scrubber  due  to intrusion of calcium via the regenerated
liquor return stream, a calcium softening step is utilized
as shown  in Figure 2.  Soda ash makeup required by the
system is added to this softener,  or Reactor-Clanfier,
                           547

-------
to precipitate soluble calcium and eliminate scaling
potential in the scrubber.  The liquor flow volumes in
the post-precipitation step of this system are dictated
by the concentration of alkali  (NaOH) that can be formed
in the regenerated liquor under conditions of chemical
equilibrium governed by Equation 4, above.

The normal ionic concentrations for this system are 0.1
molar NaOH and 1 molar Na.  This variation of the basic
double-alkali system provides scale control and process
reliability but has a captial cost higher than a compar-
able concentrated mode (if it were applicable), due to
the equipment for control of the soluble calcium and the
higher volume of liquor that must be handled by the
post-precipitation system.
CONCENTRATED MODE PROCESS

In the concentrated mode (concentrated active-alkali)
process, the flue gas is scrubbed with a concentrated
sodium sulfite/bisulfite solution to remove sulfur
oxides.  A bleed stream of spent scrubbing liquor at a
pH near 6.0 is reacted with lime to precipitate calcium
sulfite and the limited amount of calcium sulfate that
has been formed due to oxidation in the scrubber.  The
precipitated solids are then thickened, filtered and
water-washed (for minimization of sodium losses), to
produce a disposable waste for landfill disposal, and
the regenerated liquor is returned to the scrubber to
absorb additional SO2.  The normal ionic concentrations
for this system are approximately 0.5 molar Na2S03 and
2 molar Na.  This flow sheet is shown in Figure 3.

Due to limitations in the calcium sulfate precipitation
kinetics,  this system is used where the oxidation of
the S02 may be kept below about 25% of the total absorbed
SOX species.  The magnitude of this oxidation rate depends
upon the ratio of the S02 absorbed to the excess air or
oxygen in the flue gas.   In general, this ratio must be
at least 20 ppm of SO2 per excess air percentage in order
to adequately control the oxidation rate.  As a typical
example, a boiler operating at 30% excess air requires a
minimum of 600 ppm or about 1% sulfur in the coal to
insure that oxidation levels are sufficiently limited.  At
sulfur levels greater than the 600 ppm in gas, or 1% in
fuel, effective precipitation of the sulfate ions is assured.
                        548

-------
This double-alkali system removes at least 90 to 95%
to the SC>2 in the flue gas.  A relatively dry filter
cake waste is the sole waste product.  There are no
liquid effluents discharged from the process.  Scrubber
scaling is controlled because the regenerated liquor
contains a very low calcium ion concentration (20-80 ppm)
because of the limited aqueous solubility-product con-
stant of the calcium sulfite species.  The scrubbing
solution is a clear liquor  (containing only a small
amount of residual fly-ash), that has essentially no
abrasive characteristics and cannot produce internal
scaling.
HIGH CHLORIDE PROCESS

Absorption of hydrogen chloride from the flue  gas  can
cause special problems in all throwaway systems  including
the double-alkali  process.   As indicated above,  filter
cake washing is  employed to wash out and recycle the
soluble  sodium salts to the process.  This procedure
minimizes the amount of soda ash required for  make-up.
When chloride'is absorbed into the scrubbing liquor,
sodium chloride  concentration-buildup in the system
is magnified by  this washing and recovery of the soluble
salts from the  filter cake.  In some cases, augmental
sodium chloride  concentrations as high as 1.5  to 2.0'
molar may result,  causing sodium sulfate insolubility
problems as well as increased usage of make-up soda ash,
with higher  sodium content in the filter cake.

In order to  eliminate these problems, a modification to
the basic  flow  sheet has been developed to remove  the
chlorides  in  a  separate precooler section upstream from
the  S02  absorber.   This flow sheet is shown in Figure  4.
The  chlorides  and other secondary coal contaminants are
removed  in  a low pH scrubbing liquor.  A slipstream of
the  resulting acidic solution is neutralized with  lime
or  limestone to produce a calcium chloride solution which
 is  used  as  a final wash on a horizontal fliter of  the
 Eimco-Extractor type.  This wash is  so designed that the
 filter  cake absorbs the chlorides.

 The net  results of this chloride removal  loop when applied
 to h?gh-chloride coals  (approximately  0.1% chloride by
 weight in coal or higher),  are  to:

 1.   Reduce sodium carbonate  make-up requirements;
     thus reduce operating  costs to  normal  double-
     alkali system levels.
                           549

-------
 2.   Reduce  the  dissolved solids concentration
     in  the  SC^  scrubbing system;  thus  to  reduce
     salting out problems and achieve normal
     optimum double-alkali process operation.

 3.   Reduce  the  amount of sodium salts  in  the
     filter  cake;  thus producing more suitable
     landfill waste material.

 4.   Eliminate most of the residual fly-ash
     and other secondary  coal  contaminants from
     the SO2 removal  system.   This minimizes
     the amounts of heavy metal  ions  in the S02
     scrubbing liquor and thus reduces the
     catalytic effects of these  ions  in promoting
     oxidation of the sulfite  and  bisulfite ions.

 In  a further refinement  of this process, a fly  ash thick-
 ener is added to the precooler  loop.  Thus, the system
 can be  used as  a back-up to the electrostatic precipitators
 and still operate at a low suspended solids level with  low
 abrasion characteristics in the precooler liquor.  The
 primary function of  this additional thickener is to accommo-
 date temporary  malfunctions of the electrostatic precipitators
 such as shorts  of electrical  sets and temporary power loss.
 In  the  interim, while  the  back-up  function is actually  being
 fulfilled,  the  precooler system is not subjected to abrasion
 since solids build-up  and  related problems have been avoided:
 the system  will continue to function as designed.  This is
 the basic flow  sheet that  has been accepted and is being
 applied for use at the Newton Station of Central Illinois
 Public  Service Company,  scheduled for 1977 startup of
 operations.  The  system, currently in the engineering
 phase,  is the concentrated mode/high chloride process
 with  auxiliary fly-ash thickener loop to control the system
 during  precipitator upsets.


 PLANT DESCRIPTION & DESIGN CONDITIONS

 The double-alkali system will be installed as an adjunct
 to the Newton Station, Unit #1,  Newton, Illinois.  The
boiler is a drum-type manufactured by Combustion Engineering,
 Inc., designed for a  maximum steam generating capacity of
4,158,619 Ibs/hr. at  2,620 psig  and 1,005°F.  at the steam
outlet connection. The boiler unit and the  double-alkali
control system are scheduled for in-service  operation
December, 1977.
                         550

-------
The boiler is tangentially fired with pulverized coal.
The pulverizing mills are the bowl-type manufactured
by Combustion Engineering, Inc.  Net theoretical air
required for combustion at maximum steaming condition
is 4,192,000 Ibs/hr at 60°F. and 80% RH.  The boiler
is designed for operation at 22% excess air.  High
sulfur Illinois bituminous coals from several sources
may be burned.  In the design of the system, the
following flue gas analysis was developed from the
worst case criteria:
                          LBS/HR.

    S02                    37,800

    H2O                   330,000

    02                    556,000

    N2                  4,580,000

    C02                 1,110,000

    S03                       700

    HC1                     1,000


    TOTAL              6,615,000  d)


 The system is designed for coal properties and boiler
 system characteristics as follows:

    Sulfur Content                        4.0% in coal
    Chloride Content                      0.2% in coal
    Coal Heating Value as received        10,900 BTUS/lb.
    Boiler Heat Release                   5.5 x 109  BTUS/hr,
    95% of Sulfur in fuel Reports as S02 in flue gas
    96% of Cl in fuel Reports as HC1 in the flue gas
        represents total I.D. fan capacity including excess
  and in-bleed air leakage.
                           551

-------
PROJECT SCOPE

Buell is responsible for the design, procurement  and
delivery to the job site, of a double-alkali  flue gas
desulfurization facility (FGDU) designed for  90%
removal of S02-  The station design includes  two
induced fans which discharge into a common plenum.

Each fan is also connected to the stack.  The double-
alkali flue gas desulfurization unit, which is being
built to accommodate Unit 1, will include the following
major equipment and facilities:

A.  Four booster fans.

B.  The ductwork,  including the plenum, leading
    to the four booster fans, also including
    isolation dampers for each fan.

C.  Isolation gate dampers in the boiler ductwork
    to isolate the I.D. Fans from the stack.

D.  Four precooler towers.

E.  Ductwork from the booster fans to the four
    precoolers.

F.  Four scrubber towers.

G.  Mist eliminators between the precoolers
    and the scrubbers.

H.  Mist eliminators at the outlet of the four
    scrubber towers.

I.  Ductwork from each  of the four scrubbers con-
    necting them to two plenums,  each of which
    is common to two scrubbers.   This ductwork
    will be complete with outlet isolation gates
    for each scrubber.

J.  All ductwork to connect the scrubber system
    back to the chimney.

K.  Lime unloading and handling system including:

    1.  Equipment designed to receive pebble
        or granular"quick-lime using both
        rail and truck unloading facilities.
                           552

-------
    2. Four lime storage silos for the
       storage of pebble or granular lime.

    3. Lime feeding and ball-mill-slaking
       facilities for the addition of lime
       to the reaction tanks.

    4. A building sheltering the lime feed-
       ing and slaking facilities.

L.  Soda ash system, consisting of:

    1. Soda ash off-loading facilities de-
       signed to receive soda ash from both
       rail and truck.

    2. Two soda ash storage silos.

    3. Soda ash feeding and dissolving equip-
       ment.

M.  Two 100' diameter thickeners for the double-alkali
    post-precipitation system with concrete bottoms and
    access to bottom discharge cone using tunnel.

N.  One 50" diameter thickener for the precooler liquor
    loop, constructed of coated steel and elevated above
    grade.

O.  Three Eimco Extractor filters for dewatering and
    washing of double-alkali-system thickener underflow,
    and associated equipment including vacuum pumps,
    filtrate receivers, filtrate pumps, and moisture
    traps.  A building to house the filters and all
    related equipment is also included with the FGDU.

P.  A control and maintenance building.

Q.  Flocculant addition systems for the thickeners.

R.  Material off-loading facilities including a train
    shed  (about 250' long) which will accommodate
    four cars and material handling equipment to
    unload two cars at one time from a total of six
    hoppers.  The material off-loading facility is
    able to handle trucks which may or may not have
    their unloading auxiliaries.

The Project Scope also includes necessary accessary equip-
ment and facilities required to fulfill all performance
guarantees.
                          553

-------
GENERAL COMMENTS

Design of flange-to-flange wet-end equipment such as
reactors, thickeners, vacuum filters, is based on the
process bleed liquor flow rates in accordance with
established sizing criteria per Envirotech pilot plant
experience.  Wherever feasible, the Envirotech wet-end
system design incorporates gravity flow for operating
simplicity and to reduce oxidation, covered troughs
have been utilized in place of pipe for gravity runs
wherever possible.  Materials of construction are
generally mild steel lined with appropriate corrosion
and/or abrasion resistant materials.  Piping is sized
for flow rates of 4-7 feet per second.  Process pumps
are designed per system flow sheet with an added 20%
safety factor.  Rubber lined pumps with rubber-covered
impellers are utilized for corrosive and/or abrasive
service.  All process pumps are spared where necessary
to insure continuity of system operation.

Instrumentation is designed to permit automatic oper-
ation of the system, and includes compensation for
load changes, and- an emergency water deluge system for
high-temperature protection of scrubber internals
including non-metallic linings.  Facilities provide
for the recording of key system operating data, i.e.,
liquid flow rates, temperatures, pressures, pH, inlet
and outlet SC>2 concentrations, etc.  Manual override
is also provided for.  In the event of complete system
shutdown, provisions have been made to drain all lines
and pumps to drain sumps.  The drain sumps are then
pumped to vessel freeboard space in the post-precipita-
tion system.  Tankage is sized to allow for the minimum
liquid retention time consistent with optimum process
operation taking into account the additional volume
needed for liquid storage during system shutdown.
Materials of construction are generally non-metallic
lined, 1/4" thick, mild steel.  A vacuum filtration
system is provided and is sized based on the thickener
bottoms solids-flow rate.  Provision for a spare filter
plus variable speed drive permits extra capacity.
Should it become necessary to remove the filters from
operation for a short period of time, the thickeners
are capable of handling an amount of solids build-up.
The thickener mechanisms are designed to permit an
accumulation  (12" - 24") of settled solids in the
thickener tanks through the incorporation of automatically
operated lifting devices that raise and lower the rakes.
                           554

-------
The wet-end  (post-precipitation) system incorporates
a filter building which houses the vacuum filter system.
A separate building houses the process control room.
The latter is a dust  free, heated and air conditioned
room housing the system instrumentation and operators'
comfort facilities.   The  reaction tanks are provided
with a launder system that permits by-passing of any
one unit for maintenance  without shutting down the
process.
CAPITAL INVESTMENT AND  ANNUAL OPERATING COST  ESTIMATES

The project  is currently in the design engineering phase
and finalized costs  have yet to be developed.   It is
currently estimated  that the direct investment will be
$23 million  with  the completed cost of $40  million which
would reflect the total capital investment  estimated  at
this phase of the project.   Since the investment costs
have not been developed at this time for all  the indiv-
idual equipment  items,  an analysis of the fixed invest-
ment cannot  be made.  However, based on the estimates
and the known  feedstock/utility and manpower  requirements,
the following  annual operating cost values  have been
developed:

     ESTIMATED AVERAGE ANNUAL OPERATING COSTS

                DOUBLE-ALKALI PROCESS

 (575 MW NEW COAL FIRED BOILER UNIT, 4.0% IN FUEL; 90% REMOVAL)
 DIRECT COSTS
  ANNUAL
QUANTITIES
                                  UNIT
                                 COST $
TOTAL
ANNUAL
COST $
 Lime (93% Avail.
       Ca0)         114,975 M     35/ton  4,024,125
                    tons
3.833 M
tons
   Subtotal Raw
   Materials
                                  80/ton     306,640
                     4,330,765
PERCENT OF
TOTAL ANNUAL
OPER. COST
              30.9


               2.3



              33.2
                            555

-------
                                        TOTAL     PERCENT OF
                     ANNUAL     UNIT    ANNUAL    TOTAL ANNUAL
DIRECT COSTS       QUANTITIES  COST $   COST $    OPER. COST	


Conversion Costs

  Operating Labor
  and Supervision  42,924 Man  8.00     343,392        2.6
                   Hours       per
                               Man Hrs.


Utilities

  Steam            53,300 M    1.20/M   183,960        1.4
                   Lbs.        Lbs.

  Process Water    312,732 M   0.30/M    93,820        0.7
                   Gal.        Gal.

  Electricity      79,716,000  0.012    956,592        7.3
                   KWH         per KWH
 Maintenance

   Labor  and Material
       (.02 x  40,000,000)                 800,000        6.1

     Subtotal  Conversion
     Costs                              2,377,764       18.3

     Subtotal  Direct Costs              6,708,529       51.5
                           556

-------
                                              PERCENT OF TOTAL
INDIRECT COSTS           TOTAL ANNUAL COST   AMOUNT OPER. COSTS

Average Capital Charges
 @ 14.0% of Total
 Capital Investment          5,600,000              43.0
Overhead

 Plant
                               683,553               5.3
 Administrative, 10% of
 Operating Labor                 34,339               0.2

  Subtotal Indirect Costs     6,317,892              48.5

  Total Annual Operating
  Cost                       13,026,421             100.0
               DOLLARS/TON             CENTS/MILLION   DOLLARS/TON
               COAL BURNED  MILLS/KWH  BTU HEAT INPUT SULFUR REMOVED

Equivalent Unit
Operating Cost    8.42         3.69           38.62         233.89
ESTIMATE BASIS;

Preliminary Project Cost Estimate

Life of Boiler Plant:   30  years

Coal Burned, 1,547,064  tons/year,  10,900 BTU/hr.

Stack Gas Temperature -154°F.  (Combination  Steam Reheat
  & Bypass Mixing)

Boiler Unit on Stream Time,  6,132  hrs.  @ Full Load -
  (70% Load Factor)

Total Capital Investment,  $40,000,000  (est'd.); Subtotal
  direct investment $23,000,000  (est'd.)
                         557

-------
CONCLUSION

In order to achieve a flue gas desulfurization system
design that results in a high degree of reliability for
a specific boiler application, recognition of all key
criteria and variables is required.  The double-alkali
system operating in the concentrated mode is a system
design with no scaling potential and minimized abrasion
when applied to a modern pulverized-coal utility boiler
service.  With low-quality coals, with high chloride
content, the addition of the separate precooler absorp-
tion loop for chloride as well as fly-ash control is a
further development for that increased reliability and
reduced operating cost.

These systems currently have redundancy and/or excess
capacity in their design to accommodate operating
unknowns.  As technology develops, the redundancy can
be reduced and the complexity capacity reduced.  With
this development, investments cost and operating costs
will be reduced and reliability will continue to be
achieved.
                        558

-------
    CLEAN GAS


SO 2 LADEN
FLUE
}
GA3^
makeup wate
clear liquor
^overflow
'




S02
ABSORPTION
r
ti


,
h

REGENERATION
a
SEPARATION
1
^

	 ^^
_* regenerated sodium liquor
i
lime

1 dewatered waste solids ^
to disposal
If

REGENERATED
LIQUOR
INVENTORY
t
_ soda ash
, 	 **
fig. I - BASIC  SCHEME

-------
HOT FLUE 6AS
           PRECOOLER SCRUBSER
                                  SCRUBBED GAS
                                       lime
                                                          STACK
                                u compressed _
                                -^- 	-o	soda ash
                                            REACTORcLARIFIER
                   VACUUM! FILTER
               fig.  2 -  DILUTE   MODE

-------
Ul
  HOT FLUE GAS
i—O—
                                     SCRUBBED GAS
                  PRECOOLER SCRUBS
                        ER
                                         lime
                                                           STACK
                                      REACTION TANKS
                                             soda ash
                      fig. 3-  CONCENTRATED  MODE

-------
       HOT FLUE GAS
Ln
ON
K)
                                      SCRUBBED GAS
                                     REACTION TANKS
                                              soda ash
                                                   '*!'
                                                 THICKENER
                                 NEUTRALIZED	'v
                fig. 4-  HIGH CHLORIDE  MODE

-------
                               TECHNICAL REPORT DATA
                         (Phase read Jaitnicrions on the reverse before completing)
 REPORT NO.
EPA-600/2-76-136a
 TITLE AND SUBTITLE
 'roceedings: Symposium on Flue Gas Desulfurization-
 ew Orleans, March 1976;  Volume I
                                                      . RECIPIENT'S ACCESSION-NO.
                                 . REPORT DATE
                                 May 1976
                                 . PERFORMING ORGANIZATION CODE
 AUTHOR(S) .
        R,D. Stern, Chairman; W. H.  Ponder and
 .. C. Christman (TRW, Inc.), Vice-chairmen
                                                      8. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING OR9ANIZAT1ON NAME AND ADDRESS
Miscellaneous
                                 10. PROGRAM ELEMENT NO.
                                 EHE624
                                                      11. CONTRACT/GRANT NO.
                                                      In-house
2. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of  Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC  27711
                                 13 TYPE OF REPORT AND PERIOD COVERED
                                 Proceedings; 3/8-11/76  	
                                 14. SPONSORING AGENCY CODE

                                  EPA-ORD
 s. SUPPLEMENTARY NOTES jjgRL-RTP project officers for these proceedings are R.D.  Stern
and W.H. Ponder, Mail Drop 61,  Ext 2915.
 6. ABSTRACT The proceedingS document the presentations made during the symposium,
which dealt with the status of flue gas desulfurization technology in the United States
and abroad.  Subjects considered included: regenerable,  non-regenerable,  and
advanced processes; process costs; and by-product disposal, utilization, and
marketing The purpose of the symposium was to provide developers, vendors, users
and those concerned with regulatory guidelines with a current review o? progress
made in applying processes for the reduction of sulfur dioxide emissions at the full-
and semi-commercial scale.
                 DESCRIPTORS
                                           b.IDENTIFIERS/OPEN ENDED TERMS^
 Air Pollution
 Flue Gases
 Desulfurization
 Sulfur Dioxide
 Sulfur Oxides
 Cost Effectiveness
Byproducts
Disposal
Utilization
Marketing
Air Pollution Control
Stationary Sources
                                                                     COS AT I Field/Gioup
13B
2 IB
07A,07D
07B

14A

-------