&EPA
United States     Industrial Environmental Research EPA-600/7-79-167a
Environmental Protection Laboratory         July 1979
Agency       Research Triangle Park NC 27711
Proceedings: Symposium
on Flue Gas
Desulfurization -
Las Vegas, Nevada,
March  1979;
Volume I

Interagency
Energy/Environment
R&D  Program Report

-------
                  RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

    1. Environmental Health  Effects Research

    2. Environmental Protection Technology

    3. Ecological Research

    4. Environmental Monitoring

    5. Socioeconomic Environmental Studies

    6. Scientific and Technical Assessment Reports (STAR)

    7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

This report has been  assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND  DEVELOPMENT series.  Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare  from adverse effects of pollutants associated with energy sys-
tems. The goal of  the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data  and  control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and  ecological
effects; assessments  of, and development of,  control  technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
                        EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not signify that the contents necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products  constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                   EPA-600/7-79 167a

                                             July 1979
      Proceedings: Symposium
    on  Flue  Gas Desulfurization-
Las Vegas, Nevada,  March  1979;
                 Volume  I
                Franklin A. Ayer, Compiler

                Research Triangle Institute
                   P. 0. Box 12194
           Research Triangle Park, North Carolina 27709
                Contract No. 68-02-2612
                  Task No. 55 and 99
               Program Element No. EHE624A
            EPA Project Officer: Charles J. Chatlynne

           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

-------
                                ABSTRACT


     This publication contains the text of all papers presented at EPA's
5th PGD Symposium held in Las Vegas, Nevada on March 5-8, 1979.  Papers
cover such subjects as health effects of sulfur oxides, impact of FGD on
the economy and the energy problem, energy and economics of FGD pro-
cesses, actual operating experience, waste disposal and byproduct market-
ing, and industrial boiler applications.
                                   ii

-------
VOLUME I
                                  Table of Contents

                                                                                 Page
Session 1: Energy and the Environment	=	1
    Richard D. Stern, Chairman

Overview of Control Technology: The Bridge Between
  Energy Utilization and Environmental Goals	2
    Frank T. Princiotta and Clinton W. Hall

Remarks	14
    Leon Ring

Health Effects of SO2 and Sulfates  	21
    Bertram W. Carnow and Edward  Bouchard

Energy, Environmental, and Economic Impacts of Flue Gas Desulfurization
  Under Alternative New Source Performance Standards	48
    Andrew J. Van Horn, Richard A.  Chapman,
    Peter M. Cukor, David B. Large

Session 2: Impact of Recent Legislation  	87
    Walter C. Barber, Chairman

Session 3: Economics and Options	88
    Walter C. Barber, Chairman

Status of Development, Energy and Economic Aspects
  of Alternative Technologies	89
    P. S. Farber, C. D. Livengood,
    K. E. Wilzbach, W. L. Buck, H. Huang

Economics and Energy Requirements
  of Sulfur Oxides Control Processes  	137
    G. G. McGlamery,  T. W. Tarkington,
    S. V. Tomlinson

Combined Coal Cleaning and FGD  	215
    James D. Kilgroe

The Interagency Flue Gas Desulfurization
  Evaluation Study	258
    James C. Dickerman  and Richard  D. Stern
                                        ill

-------
 Session 4: Utility Applications	285
     Michael A. Maxwell and Julian W. Jones, Co-chairmen

 Status of Flue Gas Desulf urization in the United States	286
     Bernard A. Laseke and Timothy W. Devitt

 Recent Results from EPA's Lime/Limestone
   Scrubbing Programs	342
     H. N. Head, S. C. Wang, D. T. Rabb,
     R. H. Borgwardt, J.  E. Williams, M. A. Maxwell

 TVA Compliance Programs for SO2 Emission  	386
     G. A. Hollinden and  C. L. Massey

 S02 and NOx Removal Technology in Japan	418
     Jumpei Ando

 EPRI's FGD Program: From Problem Identification
   to Development of Solutions  	450
     G. T. Preston

 Cholla Station Unit 1  FGD System: 5 Years
   of Operating Experience 	469
     Stephen R. Travis and Frank A. Heacock, Jr.

 La Cygne Station Unit No. 1: Wet Scrubbing
   Operating Experience	486
     Terry J. Eaton

 Dry  FGD Systems for the Electric Utility Industry	508
     Stephen J. Lutz and C. J. Chatlynne

 Plan, Design and Operating Experience of FGD
  for Coal Fired Boilers Owned by EPDC  	526
     Yasuyuki Nakabayashi

 Current Alternatives for Flue Gas Desulfurization
  (FGD) Waste Disposal—An Assessment	561
     Chakra J.  Santhanam, Richard R. Lunt, Charles B. Cooper

 Marketing Alternatives for FGD Byproducts: An Update  	595
     W. E. O'Brien and W. L. Anders

Limestone FGD Operation at Martin Lake
  Steam Electric Generator	613
     Mark Richman

VOLUME II

Basin Electric's Involvement with Dry
  Flue Gas Desulfurization	629
    Kent E. Janssen and Robert L. Eriksen
                                         iv

-------
Utility Conventional Combustion Comparative Environmental
  Assessment—Coal and Oil	654
    Charles A. Leavitt, C. Shih, Rocco Orsini,
    Kenneth Arledge,  Alexandra Saur, Warren D. Peters

Operating and Status Report: Wellman-Lord
  S02 Removal/Allied  Chemical SO2
  Reduction—Flue Gas Desulfurization Systems
  at Northern Indiana Public Service Company and
  Public Service Company of New Mexico	702
    D. W. Ross, James Petrie, F. W.  Link

Citrate Process Demonstration Plant:
  Construction and Testing	761
    Richard S. Madenburg, Laird Crocker, John M. Cigan
    Laurance L. Oden, R. Dean Delleney

Design and Commercial Operation of  Lime/Limestone
  FGD Sludge Stabilization Systems	792
    Ronald J. Bacskai and Lee C. Cleveland

Power Plant Flue Gas Desulfurization  Using
  Alkaline Fly Ash from Western Coals	809
    Harry M. Ness, Philip Richmond,
    Glenn Eurick, Rick Kruger

Environmental Effects  of FGD Disposal:
  A Laboratory/Field Landfill Demonstration	835
    N. C. Mohn, A. L. Plumley,
    A. L. Tyler, R. P. Van Ness -

Physical Properties of FGC Waste Deposits at the
  Cane Run Plant of Louisville Gas and Electric Company	858
    C. R. Ullrich, D. J. Hagerty, R.  P. Van Ness

Summary of Utility Dual Alkali Systems	888
    Norman Kaplan

The FGD Reagent Dilemma:
  Lime, Limestone, or Thiosorbic Lime  	959
    Donald H. Stowe,  David S. Henzel, David C. Hoffman

Session 5: FGD Current Status and  Future
  Prospects—Vendor  Perspectives	989
    Frank T. Princiotta, Chairman

Session 6: Industrial Applications	990
    Richard D. Stern, Chairman

The Status of Industrial  Boiler FGD Applications
  in the United States	991
    John Tuttle, Avinash Patkar, R.  Michael McAdams

-------
Environmental Assessment of the Dual Alkali FGD System
  Applied to an Industrial Boiler Firing Coal and Oil  	1023
    Wm. H. Fischer, Wade A. Ponder,
    Roman Zaharchuk

Operating History and Present Status of the
  General Motors Double Alkali So2 Control System	 1067
    Thomas 0. Mason, Edward R. Bangel,
    Edmund J. Piasecki, Robert J. Phillips

R-C/Bahco for Combined SO? and Particulate Control  	1082
    Nicholas J. Stevens

Status of the Project to Develop  Standards of Performance
  for Industrial Fossil-Fuel-Fired Boilers	1115
    L. D. Broz, G. R. Offen, D. D. Anderson,
    J. D. Mobley, C.  B. Sedman

Flue Gas Desulfurization Applications
  to Industrial Boilers	1140
    James C. Dickerman

Unpresented Papers	1160

Stack Gas Reheat—Energy and Environmental Aspects	1161
    Charles A. Muela, William R. Menzies,
    Theodore G. Brna

Minimizing Operating Costs of Lime/Limestone
  FGD Systems	11 79
    Carlton Johnson

By-Product-Utilization/Ultimate-Disposal of Gas Cleaning Wastes
 from Coal-Fired Power Generation  	1187
    William Ellison and Edward Shapiro

Flue Gas Desulfurization and  Fertilizer Manufacturing:
  Pircon-Peck Process	1 204
    R. B. Boyda

Dry FGD and Particulate Control Systems	1222
    K. A. Moore, R. D. Oldenkamp,
    M. P. Schreyer, D. W. Belcher

Attendees	1 235
                                         vi

-------
         SESSION 1



ENERGY AND THE ENVIRONMENT



 RICHARD D. STERN, CHAIRMAN

-------
Overview of Control Technology - The Bridge Between Energy Utilization

                         and  Environmental  Goals
                              BY
                    Frank T.  Princiotta
             Director,  Energy Processes Division
           Office of Energy,  Minerals,  and  Industry
             U.S. Environmental Protection  Agency
                              AND
                         Clinton W.  Hall
               Director, Energy Coordination Staff
               Office of Research and Development
               U.S. Environmental Protection Agency

-------
               Energy Policy - Where Are We Going?






     The profile of U.S. energy development and use will undergo major changes




in the years ahead.  Although only slowly evolving, it appears that our




national energy policy will call for a widespread conversion of utility and




industrial power facilities from scarce oil and gas to plentiful coal,




decreased fuel consumption, particularly for the transportation sector, and,




in the longer term, the use of technologies that are only now beginning to




emerge for the production of liquid and gaseous fuels from coal and oil shale.




     Projections indicate that total U.S. coal mining activities will increase




from today's annual production of 700 million tons to nearly 1 billion tons by




1985 and will more than double by the year 2000.  In 2000, conversion of




existing utility and industrial facilities from oil and gas to coal coupled




with construction of new conventional and advanced coal utilization facilities




will consume approximately 1.4 billion tons of coal annually (Figures 1,2,




and 3).  Although conventional combustion of coal will predominate, by the




year 2000 emerging coal-based technologies are projected to consume 300




million tons of coal per year.  Furthermore, the National Highway Traffic




Safety Administration predicts that diesel powered cars, which offer 20 to




30% gains in fuel efficiency, will account for 25% of all new car sales in




1985 (Figure 4).




                         Environmental Problems




     These shifts in energy development and use pose potential significant




threats to human health in the next two decades.  Massive increases in coal




and oil shale mining, off-shore oil and gas production, and uranium extraction




are all projected by the year 2000.  Intensified mining activity will create

-------
 Figure 1
QUADRILLION BTU
                                U.S. ENERGY RESOURCE REQUIREMENTS
                                        Source: Techno ogy Assessment Modeling Project, 1978
Figure  2
                              SOURCES OF DOMESTIC ENERGY SUPPLY
                                       Source: Technology Assessment Modeling Project. 1978
                                       Mole: 1 Quadrillion IITI's = Appro*. 45 Million Ions "I Coul

-------
                   COMPONENTS OF COAL UTILIZATION
    20
     15
aa
z
o
a,
a
<
a
    10
              W>    V}
                                c   e    =   «,
                                .2   .2    °   ~
                                =   S    «   o.
              E
              o
              y
o
U
              I   -2
"55
o
U
I
U
 ..
O
     ..
     =
E    =
o    o
O   O
                                                                   0   .=
                                                                        b.
                                             o
                                             U
                               .


                               =   I
                         Mole:  1 Quadrillion B'lUs ' Approx. 45 Million Ions of tosl

               Sourer: U.S. Knvironmenlal Prolei-liim Asuni->, Tei'hnolo|;> Assessment Modelinc Projecl, 1978.
                                         Figure  3

-------
30
10
      iiiiiii
      3 383833
   80
82
84   86
 Years
88
                i
                90
   MAXIMUM PROJECTED DIESEL AUTO SALES
          Source: National Highway Traffic
           SafiMv Adminislralion
            Figure 4

-------
erosion problems and generate runoff which can contaminate surface waters.




Aquifers may also be polluted as a result of leachate or drainage from the




mines themselves, or from the improper disposal of mining wastes.  Increased




use of coal by utilities, industries, and new technologies will produce more




air pollution and solid waste residue than are currently produced (Figure 5).




The pollutants expected to increase are nitrogen and sulfur oxides, and ashes




and sludges.  Because of the way they are formed, pollutants emitted from new




technologies can be varied and complex and may prove to be even more harmful




to human health than those emitted from current technologies.  And the use




of diesel engines as an alternative to gasoline spark-ignition engines will




generate large quantities of particulate matter (Figure 6) which may be




carcinogenic to humans and which EPA research has already shown to be mutagenic.




                                   Research Needs




     A multitude of information is needed to avert massive future environmental




impacts.




     For mining activities, particularly those associated with coal, oil shale,




and uranium production, the impact of runoff on receiving streams and of mine




drainage of toxic pollutants on groundwater needs to be quantified and the




appropriate control methods developed.  Techniques to combat water and wind




erosion of reclaimed land are sorely needed.  Improved methods are also




required to mitigate radiation problems resulting from mining and milling




uranium ores.




     Expanded use of existing coal burning technologies demands that technologies




for sulfur oxide and particulate control be improved (Table 1).  Since they are




in such an early stage of development, control technologies for nitrogen oxides




must undergo vast improvement in the years ahead.  Additionally, research must

-------
250
                  GROWTH OF NET AIR EMISSION
            OVER TIME FROM STATIONARY SOURCES
                           Bas» Year (1975) Estimate < 10' Tons)
          Source: U.S. Environmental Protection Agency Technology Assessment ModelinB Project. 1978
                         Figure 5

-------


J£ 100
>«
u
0,
1
u.
O
£
z
<
CO
3
O
X
H
50










o


























„<



































*




if.
J1




%
**
jt
™ JS





.••
**T~
-J* »
A*>W~

1



«


































j
;
•
•
/
.•



v
Jfc
_w
^kT
A^£L
J» fc_ 	 _
9
^i1 --.^r^
• Ll,"«w
• , |F-' i
'te
r.'.s

- -«^~


i ;
* , ' *""

.„„,«««.


"', is*»«
4 •-







it"
W^KS"
^
<

C > J^ : '-•
-\
*&&»?.
T * '^?
V^^gp
?^ ' ' , f
^&&&f&*&f • _/

/
^j^,t*v '^
" !,8«

r'1
'













1
-^2r

' 1
f
	 J i *
««t»«jj '.||'
jgpipr

_. .



,.•*
1















































80 82 84 86 88 90
YEARS
DIESEL PARTICULATE EMISSIONS
Source: Office of Research and Development U.S. Knvironmenlul Protection Agency, 1478
Figure 6

-------
Table 1

SUMMARY OF R&D
CONTROL TECHNOLOGY
NEEDS FOR CONVENTIONAL COMBUSTION
Source:
Pnnciotla, F.T., 1977, Utility and Industri
ll Power, Energy/Environment H.
U.S. Environmental Protection Agency EPA-600/9-77-012 (Updated for Research Outlook
Description of
Pollutant

Sulfur
Dioxide
(SO )















Nitrogen
Oxides
(NOX)










Participate
Matter










Potentially
Hazardous
Materials





* Ambient Air Ouali
Standard Type of
Presently Control
Established Technology
Yes Coal Cleaning
NSPS&
AAQS*







Flue Gas
Desulf. (FGD)






Yes Combustion
NSPS & Modification
AAQS* (CM)

Flue Gas
Treatment







Yes Electrostatic
NSPS & Precipitators
AAQS*


Bag Houses

Wet Scrubbers


Novel Devices

No Undefined







y Standard (Health-Related)
Present Status Secondary
Residuals

1st Generation High S-Refuse
Demo Planned








1st Generation Sludge,
in Full Scale Purge Streams
Demo

2nd Generation
in Bench
and/or Pilot
Scale
Commercial Purge Streams
for Some New for Certain
Units Processes

Pilot Scale and
Demo in Japan
on Oil; Pilot
Scale on Coal
in U.S.




Commercial- Fly Ash

1st Generation
Demo

1st Gen. Com-
mercial
2nd Gen. Full Scale
Demo

Bench or Pilot
Scale
Undefined Undefined









, Aug. 1978)
Needed Control
Technology R&D

— Eliminate Secon-
dary Pollution
— Demonstrate
Practicability in
Conjunction with
FGD
— Develop Chemica
Processes Capable
of High Efficien-
cies
— Improve Removal
Efficiency
— Eliminate Secon-
dary Pollution
— Improve Reliabili-
ty
— Improve Energy
Efficiency
— Lower Costs
— Demonstrate Low
NO Burner
Capable of 80%
Control
— Broaden Ap-
plicability of
Combustion
Modification
Technology
— Evaluate Flue Gas
Cleaning Process
at Pilot Scale
— Improve Conven-
tional Fine Par-
ticulate Control
Technology
— Broaden Ap-
plicability
— Develop Novel
Devices with Im-
proved Capability
and Cost Effec-
tiveness

— Assess the
Magnitude of
Problems
Associated with
Unregulated
Pollutants Via
Chemical and
Biological
Characterization
   10

-------
strive to address major information gaps regarding the magnitude of the




health and environmental problems associated with trace elements, radio-




active material, and polycyclic organic emissions produced during  conven-




tional combustion of coal.




     Emerging energy technologies must also undergo careful scrutiny for




environmental impacts (Table 2).  Particularly, the level of sulfur and




nitrogen oxides, particulates, heavy metals, and toxic and carcinogenic




organic compounds, produced by the coal-based technologies of gasification,




liquefaction, and fluidized bed combustion need to be assessed.  Geothermal




energy production methods should be examined from the perspective of hydrogen




sulfide gases released, waste heat and steam plumes produced, and land-use




implications.  Solar energy systems should also be evaluated in terms of land




and water use, sludge and residual production, and the possible leakage of




toxic working fluids.




     Finally, research needs to determine the cancer-causing potential of




diesel soot and, if positive results are found, to establish the link between




ambient concentrations and the incidence of cancers in humans.  In parallel,




control technologies will be evaluated and developed which offer potential




for major reductions in particulate emissions.




                         The Bridge - Control Technology




     So we see an evolving national energy policy which through its emphasis




on increased coal combustion, conservation and emerging energy technologies




can yield massive environmental damage.  At the same time national concern.




for environmental quality remains high and stringent legislation is on the




books which calls for protection of our air, water and land resources.
                                     11

-------
                             Table 2
           SUMMARY OF EMERGING ENERGY TECHNOLOGIES
 emerging
Energy
Technologies
Status
Major
Environmental
Concerns
FOSSIL FUEL

 :oal
 Gasification
Coal
Liquefaction
Coal Fluidized
Bed
Oil Shale
 OTHF.R

 Geolhermal
Solar
Lurgi low BTU gasification is commercial in Europe for
certain non-coking coals. Various low BTU processes are
and will be demonstrated in the United States under
Department of Energy funding. Low BTU gasification
for on-site heating will be available in mid-1980's.
Methanated Lurgi and second generation high BTU pro-
cess will be available in 1990's.
Major processes near commercialization are Solvent
Refined Coal (SRC) and H-coal processes. Currently at
pilot scale; demonstration of two processes planned
by the Department of Energy.
30MW pilot FBC (atmospheric) being studied at
Rivesville, West Virginia by the Department of Energy.
Availability expected in 1990. Small pressurized EPA unit
has been operated for several years (0.63MW);
DOE plans a 14MW pilot unit in the early 1980's.
Availability expected in 1990's.
Both above ground and underground (in situ) re-
torting processes under development. Major on-going
efforts include the Navy's Program, using the Paraho
above ground process, nearing completion of its goal to
produce 100,000 barrels of shale oil and Occidental in
situ process which has produced in excess of 50,000
barrels to date.
Three principal types: Convective hydrothermal; geo-
pressurized hydrothermal and hot dry rock. Present
generating capacity of convective hydrothermal is
500MV. Hot dry rock is the largest resource but
because of the difficulty in fracturing the rock, it has
generated no commercial interest. Increased geothermal
application is expected in the 1990's.
Three major areas: Heating and cooling of buildings,
production of electricity (photovoltaic) and production of
clean fuels from biomass. Clean fuels at commercial scale
from biomass (gasohol) are currently being produced;
and space heating currently state of the art.
Sulfur and Nitrogen
 compounds
Paniculate emissions
Water contamination
Heavy metals and
 organic compounds
Acidic gases
Subsidence (in-situ
 gasification)
Aquifer disruption
 (in-situ gasification)

Sulfur and nitrogen
 compounds
Particulate emissions
Water contamination
Heavy metals and
 organic compounds
Acidic gases

Sulfur and nitrogen
 oxides and par-
 ticulates
Toxic heavy metals
 and organic com-
 pounds
Thermal pollution
Spent sorbent
 disposal problems

Sulfur and nitrogen
 compound
 emissions
Particulate emissions
Water contamination
 and availability
Overburden and
 spent shale
Toxic and com-
 bustible gases
Subsidence and
 aquifer disruption
 (in situ)

Hydrogen sulfide
1 release
Waste heat and
 stea'm plumes
Seismic effects
Subsidence
Minerals precipitation
Noise and bjowout
Land use

Toxic working  fluid
 leakage (photo-
 voltaic)
Sludge and residuals
 from silicon distilla-
 tion (photovoltaic)
                                             12

-------
In order to resolve this potential conflict, it is essential that economical



and reliable environmental control technology be developed and ultimately



applied on a widescale basis.  Control technology, then, allows the nation



to meet two of its major goals - adequacy of reliable energy supplies and



environmental protection.
                                     13

-------
                    REMARKS BY LEON RING, GENERAL MANAGER
                         TENNESSEE VALLEY AUTHORITY
                BEFORE SYMPOSIUM ON FLUE GAS DESULFURIZATION
                      LAS VEGAS, NEVADA, MARCH 5, 1979

     I appreciate this opportunity to meet with you.   I know this group
recognizes the need for technical exchange in the area of environmental
control.   And I believe that technical exchange will  occur.   On your program,
I see buyers, sellers, builders, operators, researchers, and regulators:  I
see contributors from the United States and abroad.   And I'm sure public
interest environmental organizations are well represented also.
     This type of forum is especially appropriate for indulging in frank
discussion, bringing problems to the forefront, and finding ways to resolve
them.  One area I'm sure you will discuss this week is flue gas desulfuriza-
tion system reliability.  As more and more systems are put into use, improve-
ments will be essential for the success of this technology.   Many other
things will be discussed here.  As a matter of fact,  it's entirely possible
that future regulations may be shaped from discussions held at this meeting.
It's possible, too, that company plans may be amended due to exchanges here
in Las Vegas.  Whatever area the impact is in, I'm sure it will  be construc-
tive.  Some of us who indulge at the gaming table may lose a few coins, but
I think we'll leave here winners with new information and new ideas toward
solving environmental problems.
     I guess I ought to warn you that I'm the first of five TVA speakers.
The others have told me not to cover their material.   We would hope to be
able to provide some insight into every area of TVA's FGD program.  My re-
marks will be more general and touch on TVA's policy in environmental control,
Others will tell you of our findings, our experiences, and our detailed
plans to clean up the air in the Tennessee Valley.  In fact, just before the
meeting today Jerry Hoi linden told me that he was prepared to expound on any
                                     14

-------
detail of our 50-page proposed consent decree.  So just shoot him the page,
paragraph, and line.
     I also hope that our participation at this symposium shows our support
for this type of forum and our deep interest in a subject that relates so
significantly to our nation's environmental health.   As you may know, TVA
has launched the country's largest program to clean the air and we are fully
committed to making it work.   TVA is more than a power system.   We have
broad responsibilities for the economic and social development of a vast
region.   Clean air is consistent with our purposes and an integral part of
improving the quality of life in the Valley.
     TVA was the first experiment in unified development of the total natural
resources of a river valley,  pioneering an idea that has since spread far
and wide.  TVA has built dams to regulate the Valley's floodwaters.   The
system of dams also created a waterway for barge traffic and generates
electricity.  TVA develops and demonstrates improved fertilizers which have
helped to replace erosive row crops on hillsides with pastures and cattle.
Our role in fertilizer development is a little known fact, yet half of the
fertilizer research in the world is done by TVA.   TVA also works with the
states and other organizations in developing agriculture, forestry,  recrea-
tion, and other resources.
     Protection and enhancement of environmental  quality is and has always
been an  important part of TVA's concept of integrated development of the
resources of the region.  This is no small job in the Tennessee Valley.
Protecting the environment in our 80,000-square-mile service area has become
a complex and demanding task when coupled with the expansion of power gener-
ating capacity.
     The TVA power system, right now, consists of about 1,200 MW in pumped
hydro storage, 2,500 MW in combustion turbines, 3,300 MW in hydro capacity,
3,500 MW in nuclear capacity, and 18,000 MW in coal-fired capacity.   Under
long-term contracts, we have access to an additional 1,300 MW in hydro
capacity.  With some quick addition, you'll find that's a total of close to
30,000 MW of installed capacity.   Our present construction schedule calls
for an additional 18,000 MW to be in service by 1986.  Most of that will be
nuclear—the exception is one more unit of pumped hydro.
                                     15

-------
     Forecasting growth in power loads used to be done in a simple way—
often with no more effort than drawing lines on log-log paper.  But the lead
time now required for additional capacity and the number of factors affecting
future loads have made more sophisticated forecasting mandatory.  So TVA has
gone to an extremely complex forecasting methodology.
     A number of interacting factors are avaluated for their impact on
future loads.  Among them are such factors as economic growth, population
growth, and substitution of electricity for scarcer fuels, which tend to
add to power demands.  Others such as conservation, load management, the
use of solar energy and congeneration, increased prices, and appliance effi-
ciency will make the forecasted load go down.  All these things complicate
making estimates on how much capacity we will need and by when.
     By quantifying these factors in ranges, our power supply planners come
up with a wide range of forecasts.   Load forecasts for the next 10 years
project that between 200 billion and 230 billion kilowatt-hours will be
sold in 1988.  And it seems clear that we will need additional capacity in
the 1990s.
     But TVA owes more than adequate power to the 6.7 million Tennessee
Valley residents.  We owe them a clean and safe environment, and we intend
to provide one.  In 1950, even before the first of our steam plants went
into operation, TVA began an extensive program of air quality studies and
monitoring.  That program has provided some of the best basic data available
for predicting the effects of power plant operation on air quality under
various weather conditions.  This information has been valuable to TVA and
other systems in locating, designing, and operating power plants.
     In the 1960s and early 1970s TVA went about the business of protecting
the environment by using a variety of methods.  We built tall stacks as much
as 1,000 feet high to disperse pollutants; we installed electrostatic precip-
itators to control fly ash; emissions were limited by operational  control
procedures (we called this method SDEL).   Selective use of clean fuels and
selective installation of S02 removal systems were practiced or planned.
     A master plan that would enable TVA to meet ambient sulfur dioxide and
particulate emission standards at all of its plants was proposed in 1973:
This plan included the use of tall  stacks at three plants; intermittent
                                     16

-------
controls at nine plants; and an experimental S02 scrubber at the Widows
Creek plant.  At that time, TVA felt that ambient sulfur dioxide standards
fully protected the public welfare from any known or anticipated effect of
sulfur dioxide.  And there was doubt that continuous S02 emission standards
were necessary to meet the requirements of the 1970 Clean Air Act for exist-
ing installations.  Using these methods, TVA could have met ambient standards
at about one-tenth the cost of either utilizing scrubbers or low-sulfur
coal.
     This TVA approach differed significantly from what EPA proposed.  This
difference revolved around interpretation of what was required under the
Clean Air Act and was finally settled by the Supreme Court in 1976 when they
refused to hear our case.  This decision finally settled the question and
necessitated the use of continuous emission control.  And with that decision
TVA moved into a new era of sulfur dioxide emission control.
     Since the Supreme Court ruling, TVA has faced litigation in several
different courts.  A coalition of citizens groups filed suit against TVA in
1977.   The States of Kentucky and Alabama and EPA also entered the suit to
force TVA to meet S02 emission requirements.  These suits have been consoli-
dated and are awaiting trial.  But TVA expects them to be settled with the
approval of TVA's strategy for compliance.  The TVA Board and I have already
approved this plan.  The settlement is now subject to approval by Federal
courts in Nashville and Birmingham.
     The agreement calls for the use of low- and medium-sulfur coal, coal
washing, and more scrubbers, and will bring all TVA plants into full compli-
ance with requirements of the Clean Air Act.
     Carrying out the settlement will not be without problems.  The massive
amounts of construction work that must be scheduled between now and 1982 are
staggering.   Scheduling unit outages for scrubber tie-in while maintaining
power loads will be another challenge for TVA's power system operators.   And
when we get all those FGD systems on line, operating problems will come in
multiples rather than singly as they do now from our Widows Creek scrubber.
     We realize that carrying out the settlement will also be expensive.
Our real capital investment will be in the neighborhood of a billion dollars.
(That's a ritzy neighborhood.)  Operating expenses (in 1982 dollars) are
                                     17

-------
estimated at over $400 million per year.   Yet, while there is an obvious
impact on residential customer rates, electric bills will climb gradually,
reaching a maximum increase of about 9 percent in 1983 and then will decrease
in the succeeding years.
     The benefit of the settlement is obvious--a cleaner environment for
the Tennessee Valley.  More than 970,000 tons of S02 and 85,000 tons of
particulates a year will  be removed from the skies of the Tennessee Valley
due to the pollution controls under this agreement.
     The benefits of this program will not be limited to the Tennessee
Valley though.  As we have in the past, TVA plans to share the knowledge
gained from our experience with you and the organizations you represent.  We
look for continued and increasing participation with EPA, EPRI, DOE, and EEI
in the area of technical  exchange.  I'll  mention a few examples of this type
of work but I'll look to my colleagues in TVA to give you details in their
papers later in the symposium.
     Despite our sometimes conflicting views, TVA and EPA have made some
outstanding joint contributions in FGD research.   I  can cite examples like
our Shawnee test facility, where we developed and demonstrated FGD design and
operating practices and found possible sludge oxidation techniques for
producing a better disposal product.  The list is long and the accomplish-
ments substantial.
     Our work with EPRI is no less outstanding.   The cocurrent scrubber
research that we have undertaken for many years together has provided us
with valuable data on a novel S02 scrubber device.  And we are planning
additional projects  in the next few months.   With the recent appointment of
TVA's Chairman to the EPRI Board, our longstanding cooperative relationship
should be enhanced.
     It has been more recently that TVA has stepped up its participation with
DOE.  Under a Memorandum of Understanding signed in April 1978, TVA and DOE
have begun many joint projects.  Several of these deal with the environment
and its protection.  In addition, TVA and EEI have long been involved in the
exchange of technical information.
     Let me turn now to another angle of environmental protection—not  by
cleanup approaches but by prevention.  TVA is involved in demonstrations of
                                     18

-------
those energy technologies with minimal or reduced environmental impact.  I
think you will be interested in knowing how involved we are in such areas as
fluidized bed combustion, solar energy, fuel cells, and waste heat utilization,
     Atmospheric fluidized bed combustion holds potential for dramatic
improvements in S02 emission levels over conventional coal-fired boilers.
It may allow a greater use of the predominately high-sulfur coal found in
the eastern United States.  For thise reasons, TVA has taken a lead role in
developing AFBC and is planning to build a 200-MW unit in the mid-1980s.
Preliminary design is essentially complete for this 200-MW boiler.   In
support of this program, we will build a 10-MW pilot plant.
     To address the ever-growing problem of solid waste disposal and its
environmental impact, TVA is pursuing advancements in the technologies that
recover minerals and energy from waste.  TVA and the city of Gallatin,
Tennessee, have begun a five-year development of a $7.9 million solid waste
cogeneration facility.  TVA is providing both financial and technical assist-
ance.  This type of program is also being considered for other areas of the
Tennessee Valley region.
     TVA views its solar program as a real winner in the environmental area.
A demonstration project involving the planned installation of 1,000 solar
water heaters is now being implemented in Memphis.  TVA has arranged financ-
ing for these units, providing credit for restricting the electric hookup
system to off-peak use.
     Another area where TVA has made significant progress is in our home
insulation program.  TVA, through its distributors, surveys homes,  makes
recommendations on conservation measures needed, and provides low-interest
financing for the measures.  Over 98,000 homeowners have already been
surveyed.  We are now moving to expand this type of program to commercial
and industrial customers.  We believe if we can reduce the amount of electri-
city we have to produce, we have reduced the associated pollution.
     The TVA staff has recently assessed the outlook for fuel cells placed
at dispersed sites throughout the Valley.  We are looking at a system that
would use gasified coal, distributed by pipeline from mine-mouth operations.
The outlook for this approach is promising.  Coal use efficiency should be
high and environmental effects should be minimal.
                                     19

-------
     Ways to use the low-grade heat rejected in normal power plant operation
are being explored by TVA.  TVA has operated a simulated waste heat green-
house in Muscle Shoals, and has grown catfish in waste heat water at one of
its plants.   A greenhouse at the Browns Ferry Nuclear Plant is heated with
waste heat water from that plant's condensers.   The results to date have
been positive to the point that we are planning a waste heat industrial park
at the Watts Bar Nuclear Plant.  This park would bring agricultural, aquacul-
tural, and industrial users of waste heat together—a first for the United
States.  Operation of this innovative park should begin in 1982.
     Taking this one step further, TVA is also assessing the feasibility of
building and operating a central cogenerating power plant to provide power
for the TVA grid and to supply intermediate- and high-quality process steam
to industries.  We have received inquiries from industrial customers about
this arrangement and we are encouraged by our discussions with them on
cogeneration's potential.
     We at TVA do not view environmental  protection and energy technology
development as opposing forces.  In our view they are two sides of the same
coin.  We must have energy technology development to protect our environment
adequately.   TVA will pursue a course that will enable the people we serve
to have a clean environment and energy assurance.
                                     20

-------
                       Health Effects of S02 and Sulfates
                                     by
                           Bertram W. Carnow, M.D.
                                  Professor
                   Occupational and Environmental Medicine
                           School of Public Health
                University of Illinois at the Medical Center
                   P.O. Box 6998, Chicago, Illinois 60680
                                     and
                               Edward Bouchard
                             Research Assistant
                    Environmental Health Resource Center
                           School of Public Health
                University of Illinois at the Medical Center
                   P.O. Box 6998, Chicago, Illinois 60680
                                October, 1978
A portion of this study was supported under contract with the Illinois
Institute for Environmental Quality of the State of Illinois.
                                      21

-------
                             ACKNOWLEDGMENT







     The  author acknowledge with thanks the contributions of members




of the Environmental Health Resource Center Scientific and Clerical Staff




in particular Rodney Musselman MPH  Assistant Director of the Center.




and Dr. Tsukasa Namekata who joined with me in carrying out the most





recent studies.  These were supported by EPA contract number 68-02-2492.
                                   22

-------
                                INTRODUCTION





     In preparing my discussion I had to ask myself the question, "Why was



I being asked to discuss the health effects of SO ?"  Along with particu"



lates, it was the earliest of the common pollutants studied to determine



its health impact on the community.  Our first study in the early 60's



along with a number of others used in setting the SO  standard revealed that



increased respiratory and cardiac deaths occurred at elevated levels of SO .



Other studies also revealed increased episodes of acute illness in those with



chronic disease as well as acute health effects on other segments of the



population.  In 1973 I participated in a three  (3) day conference held under



the aegis of the National Academy of Sciences.  We concluded that more studies



were needed regarding health effects, but that there was no basis for lowering



the standards for SO,, or indeed for any of the others.  A similar conclusion
                    2


was arrived at by a National Academy of Science Task force on multiple pollu-



tants. The task force on SO  which I chaired, again arrived at this general



conclusion as did a study carried out by Dr. David Rail of NIEHS for O.M.B.



     The problem,! believe, is that many studies have been carried out on



animals and humans with variable results regarding the level of effect.  Since



we will need to rely on coal as a major energy source and since SO  removal



is costly, the variability in results is confusing and thus the related question,



"Is SO  really bad for health?"  is again raised.  As you will hear from my



discussion, it is my firm conclusion based on clinical experience and epideflft-



iologie  studies carried out by our group and others that it may indeed



adversely affect health and in some marginal groups in the population be life



threatening.  A number of studies we just completed and which I will discuss



briefly, tend to confirm this conclusion.
                                     23

-------
     The confusion regarding results I believe extends from the following':



1.   Experiments have been carried out on a multitude of animal species which




have variable sensitivity to SO .



2.   The doses used and the duration of exposure have also varied greatly.




3.   Other conditions in the laboratory have also been highly  variable in




eluding variations in the age of animals, temperature, and humidity, all of




which may be important factors in assessing effect.



4.   The experiments on animals and humans generally were carried out using




SO  as the gas in the laboratory, whereas the major effect, certainly the
  ^


most severe and chronic effects of SO  are due to sulfate and sulfuric acid,
                                     ^


secondary pollutants formed from SO .
                                   £*



5.   Human experimentation in addition to being carried out under circum-



stances where no other pollutant is present, are usually carried out on



healthy, young adults.  Ethical considerations prevent us from using those



individuals at highest risk such as those with chronic heart and lung disease



or asthma.




     In view of this, my presentation should more appropriately be titled



"Health Impact of SO " because in the real world it is found along with many



other pollutants including particulates, ozone and others with which it may




act synergistically.  Further, it becomes particulate and aerosol after inter-




acting with other materials in the environment and therefore, what we may be



seeing are the effects of the off-spring and not of the parent.  Possibly most




important of all is the fact that all of these pollutants including SO  and




those with which it interacts are impacting on a heterogenous population




which includes the very young, very old with chronic diseases of the heart




and lung and other organs, millions with genetic predisposition due to




allergies, asthma, alpha 1 anti-trypsin deficiency and others.  In addition,




70 million Americans are at high risk because of cigarette smoking.  These



are all more susceptible.            ,

-------
 I-    HEALTH EFFECTS OF SO  AND SULFATES





      SO  is classified as an irritant gas,  in that it produces irritation and




 inflammation of the tissue that it contacts directly.  SO  increases flow




 resistance by constriction of the airways,  decreases the elastic recoil of




 the lung slightly, and, at higher concentrations, can decrease breathing




 frequency.  SO  may also be absorbed by the blood and has some effect through




 a central nervous system mechansim.  Additionally, chronic exposure may result




 in decreased mucociliary flow, a major defense system which I will discuss.




 Sulfuric acid is also classified as a primary irritant in that its irritant




 and corrosive action far exceeds any systemic toxic action.




II.   GENERAL PHYSIOLOGIC AND ANATOMIC CONCERN




      The lung has multiple defenses against environmental assaults.  The nose




 and upper respiratory tract warm, filter and humidify the air.  Chemoreceptors




 in the nose and airways can detect irritant gases and produce sneezing, cough-




 ing, bronchial constriction and narrowing,  and other reactions to prevent




 noxious material from reaching deep lung areas.  Lining the airways of the lung




 are cells with tiny hairs attached (cilia)  which sweep invading particles up-




 ward in waves at the rate of 1500 times per minute.  Furthermore, in response




 to irritation, the body produces mucus.  The mucus blanket lining the air




 passages trap foreign particles.  The cilia then move the mucus blanket up




 and out, in an escalator-like motion, thus removing the particles from the




 lungs.  This clearance mechanism is a primary defense of the lung.  In .a




 normal healthy individual, most of the larger particles are trapped before they




 reach the deeper, more vulnerable parts of the lung.  Particules from two to




 five microns and below in size, however, can penetrate to the alveoli,  (air




 sacsl Within the alveoli, particles can be removed by scavenger cells  (macrophages)




 which contain  digestive  enzymes such as trypsin which attempt to consume the
                                      25

-------
particle.  Trypsin is itself highly toxic however, and its release may be




related to alveolar destruction or the development of emphysema  (Bates, 1972).




People who do not have the ability to neutralize trypsin are extremely vulner-




able to all forms of air pollution (Bates, 1972).




     The defenses of the lung can fail to operate for various reasons such as




aging, illnesses, genetic effect, or simply'being overwhelmed by a toxic sub-




stance.  When it is overtaxed for any reason the respiratory system is even




more vulnerable to injury from either acute or chronic exposure to environ-




mental pollutants like SO  (Cassarett, 1975).  For example, in aging, the




capacities of the respiratory system gradually decrease (Morris,  1971).  Smoking




and/or air pollution can speed up this aging process considerably (Bates, 1972).




Thus certain older populations, such as adult males over 55 with chronic bron-




chitis , have been identified at high risk during air pollution episodes (Carnow etal 1968




 .Carnow and Feiveson, 1969).  Infants and children whose respiratory defenses




are not fully developed are similarly vulnerable (Goldstein, 1975).




     With respiratory tract impairment it takes more work to breathe.  As a




result, there is an increase in lung pressure, which places strain on the




right side of the heart.  This may enlarge the right side of the heart, a




condition known as cor pulmonale.  Additionally, aggrevation of cardiovas-




cular diseases, particulary coronary artery disease, have been associated with




high levels of SO  as a result of a reduction in available oxygen (Carnow, 1973).




III. ANIMAL EXPERIMENTAL STUDIES




     Many studies have been carried out on multiple species.  These are




extensively reviewed in the literature and summarized in  table 1   in the




text.  Many other summaries are also available.  I would only like to summarize




Dr. Amdur's conclusions after many years of studying the toxicity of the




aerosols formed by the oxidation of SO








                                     26

-------
                                         Table I      Summary of Selected Experiments with Animals Exposed to
                                                      S02 Alone or in Combination with other Pollutants
                                                      (Sulfates, acid mists and virus)
Type of Animal

Mice


Mongrel dogs




Dogs


Swine




Beagle dogs


Guinea pigs


S02
Concentration
10 ppm


7-230 ppm




22 ppiii


35 ppm




1.0 ppm


1.1 ppm
with NaCl,
and HjO mist
Expos U.re
Time
24,48 fi
72 hrs.

15-20 rain.




30-60 min.


Continuously
1-6 weeks



1 hour, twice
a day for 12
months
few hours


Type of
Test
In Vivo


In Vivo




In Vivo


In Vivo




In Vivo


In Vivo


Results

Severe injury manifested more in
the nasal cavity than in the
trachea or lungs.
Changes in flow resistance in
proportion to gas concentration.
Greater changes in pulmonary
function when S02 was administered
by tracheal cannula.
95* of SO- was absorbed in upper
airways. Portion of non-expired
gas was observed in blood stream.
Increased salivation, ocular and
nasal irritation, loss of cilia,
metaplasia, sneezing and sneezing
frequency increased with increase
in relative humidity.
Decreased removal of particles
by tracheal mucus but no significant
changes in pulmonary functions «
Pulmonary effects with increased
humidity •

Reference

Giddens and
Fair child,
1972
Frank and
Speizer, 1972



Frank et al.,
1967

Martin and
Willoughby,
1971


Hirsh, 1975


McJilton et al..
1973

to

-------
             Table I	Continued
Type of Animal
Cats
Cynomolgus
monkeys
Emphy s ematous
Syrian hamsters
Influenza infected
mice
Chickens
S02
Concentration
20 ppm
with NaCl
0.1-5.0 ppm
with H-SO , and
fly ash
650 ppm
2.9-19.3
1.0 ppm
with Newcastle
virus
Exposure
Time
few hours
few hours/day
for 78 months
few minutes
7 days
few days
Type of
Test
In Vivo
In Vivo
In Vivo
In Vivo
In Vivo
Results
Significant changes in flow
resistance.
No synergistic effects with
H2S04.
Relatively minor influence on
airway obstruction .
At low range less pneur.ionia
however at higher range more
pneumonia .
Both SO. and Newcastle virus
slowed clearance rates •
Reference
Corn, 1972
Alarie, 1975
Goldring, 1970
Fairchild, 1972,
1975, 1977
Wakabuyshi, 1977
CO

-------
     1.     Sulfuric acid has two distinct toxic actions;  a)   it promotes




            larynx spasms and bronchispasms (bronchial constriction),  and




            b)   it can also produce irreversible scarring of the bronchioles




            and alveoli.




     2.     Not all sulfates are irritants.  The irritant potency is  not




            related to the sulfate ion as such.  Of the compounds tested,




            if particle size remains the same, the order of irritant




            toxicity would be:  sulfuric acid, zinc ammonium sulfate,  zinc




            sulfate and ammonium sulfate.




     3.     The particle size of the aerosol is a critical factor in




            determining both the nature and degree of irritant response.




            For instance, when sulfuric acid mist and zinc ammonium sulfate




            were administered to guinea pigs at the same concentrations with




            nearly equivalent particle size (approximately 0.8^) sulfuric




            acid was twice as potent as zinc ammonium sulfate.  However, at




            the same concentration and when the particle size of the  zinc




            ammonium sulfate was 0.3/u  and the sulfuric acid remained at




            O.S^u  zinc ammonium sulfate was more potent by a factor of nearly




            3 to 1.




(We note that 80 to 90% of ambient sulfates have been found to be less than




2.0ju  in diameter  (USHEW, 1969a) .)




     4.     Sulfur produces a less irritant effect if it is present as




            sulfur dioxide gas than if it is present as particulate




            sulfate or sulfuric acid.  Under laboratory conditions, if SO,,




            is converted, depending on size, type of sulfur compound,  .arid




            degree of conversion, there can be up to a 20-fold increase in




            toxicity-  In ambient air, assuming only a 10% conversion rate of
                                     29

-------
            SO™ to irritant sulfate of 0.3^  particle size, Amdur predicts




            a 4-fold increase of irritant effect tAmdur et al., 1969, 1971).




     5.     Guinea Pigs appear to be more sensitive than other animal species.




IV.  HEALTH EFFECTS  EPIDEMIOLOGICAL STUDIES




     In studying the impact of air pollution on human health, we must




recognize the limitations of applying the results of laboratory studies




to the "everyday" human condition.  While toxicologic studies on animals are




valuable because these permit the careful control of the most important




variables - the use of a wide range of exposures, and the examination of




body tissue - there are species reaction differences between animals and




humans, particularly with regard to the respiratory tract.  Analogously,




experimental or laboratory exposure of human volunteers allows control of




variables, but there are obvious ethical limitations:  1)  on the whole,young




healthy  adults must be used for subjects and therefore results cannot




easily be extrapolated to a heterogeneous population;  2)  high doses cannot




be ethically used with humans, thus there are no experiments with a wide




range of exposure levels or experiments with chronic exposures.  Moreover,




additive and synergistic pollutant interactions that occur in atmospheric




chemistry are not generally duplicated in laboratory conditions, hence mis-




leading data on dose-response.




     Epidemiologic studies, however, have the advantage of being carried




out in the real environment, and much information has been learned from




them about the acute and chrohic health effects of sulfur oxides (Goldstein,




1975; Neal, 1977).  Multiple regression studies, a common epidemiologic




approach, attempt to find a correlation between certain environmental factors




and certain adverse health effects.  The health effects of air pollution, for




instance, have been observed by examining statistical relationships between
                                     30

-------
air measurements and medical data, or by comparing one community with another,
or by studying the same community at different exposures.
A.   Acute Air Pollution Episodes
     The episodes of acute air pollution which have occurred in the Meuse
Valley, Belgium, 1930; Donora, Pennsylvania, 1948; London, England, 1952;
and 1962; New York City, New York, 1953 and 1963; and in Chicago, Illinois,
1969 provide the strongest evidence of the effect of air pollution on health.
In practically all of these acute episodes there resulted significant increases
in mortality and morbidity.  In the 1952 London episode alone over 4,000
excess deaths were reported.
     Air pollution levels during the earlier acute episodes are not fully
available.  However, we do know that SO  levels reached 1.34 ppm in London
                                       £
during the 1952 episode (Logan, 1953;  (Morbidity, 1954).  In New York City
in November, 1953, from 17 to 26 excess deaths per day were reported when  a
stagnant air mass engulfed the city and SO  levels rose to an average of 0.15
                                          £•
to 0.2 ppm (Greenburg et al., 1967).  In Chicago during a thermal inversion
in November, 1969, daily city-wide levels averaged .071 ppm and the highest
city-wide hourly average was 0.295 ppm.  In the high pollution community the
hourly level reached .412 ppm.  Excess deaths from some cardiac and respiratory
diseases were attributable to pollution in this episode  (Carnow and Namekata,
1977).
     In all of these episodes, the very young and the very old experienced
more respiratory and heart dysfunctions than other age groups, and their
responses were more severe.  Those chronically ill, particularly with cardio-
vascular disease (affecting the heart and blood vessels)and respiratory disease
were the most seriously harmed. The mortality rate was much higher in these
groups than any other.  During the episodes, generally, the deaths from
cardiovascular disease occurred early and dropped off sharply, while
                                     31

-------
deaths from pulmonary disease -usually began to occur on the second or third




day and continued for a longer period of time  (Carnow et al., 1966).




     Amdur  (1969) has noted that during most of these episodes, three factors




were usually present:  1) cold temperatures, which increased the solubility




of irritant gases in liquids;  2) fog, which provided droplets allowing  for




the conversion of SO  into sulfuric acid mist; and 3) a temperature inversion,




which produced a stagnant air mass containing high concentrations of air




pollutants.  Weather variables, therefore, have a strong impact on the chemical




processes and health effects of pollutants like SO .




B.   Morbidity Studies of Lower Levels of Air Pollution Over Long Periods




     1.     The CHESS Studies




            The most extensive, relatively recent studies on the epidemiologic




association of SO , sulfates and human health were conducted by the Environ-




mental Protection Agency's Community Health and Environmental Surveillance




System  (CHESS) for 1970-71, published in 1974.




     These  studies were made in the early 1970's after pollution levels  had




already decreased from the levels of the 50's and 60's.  However, the report




shows adverse health effects from air pollution even at the lowered level.




These adverse effects were more consistently associated with exposure to sus-




pended sulfates than to sulfur dioxide or total suspended particulates.




Generally the results x^ere:




     1.     Chronic Bronchitis Rates - In four of the regions, CHES S




            reported a consistent, statistically significant pattern of




            chronic bronchitis among residents of the more polluted




            communities.  Smoking contributed more than air pollution to




            the rates of chronic respiratory disease, and there was




            considerable variation from one community to another.  The
                                     32

-------
       contribution of occupational exposures to chronic respiratory



       disease was also somewhat larger than that of air pollution,



       being one-half as large as cigarette smoking.  The effects of



       smoking, industrial exposure, and air pollution appear to be



       additive.  The report concluded that excess bronchitis rates are



       associated with SO  exposures alone, at levels of 92 to 95 /ug/m
                         3C


       (.03 ppm) S02 and 15 ug/m  suspended sulfates.



2.     Lower Respiratory Disease Rates - In the Salt Lake Basin and



       the Rocky Mountain CHESS communities, rates of lower respiratory



       disease  (LRD) were greater among children ages 0 to 12 who had



       lived in polluted communities for three or more years.  The



       report included that an "excess of respiratory disease (among



       children 0-12) may reasonably be associated with community



       exposures of approximately 95 /ug/m   (.3 ppm) SO  and 15 yug/m  SS



       (suspended sulfates)."



3.     Acute Respiratory Disease Rates - The report determined that "a



       conservative estimate" would be that exposures to 210 pg/m



       (.07 ppm) SO , with 104 /ug/m  total suspended particulates (TSP)



       and 16 ug/m  SS were associated with a 5 to 20 percent excess of



       acute respiratory disease.  In New York the study found an



       association between air pollution and susceptibility to Hong



       Kong-type influenza among otherwise healthy families.



4.     Pulmonary Function Tests - Pulmonary function studies of elementary



       school children in New York and Cincinnati demonstrated that forced



       expiratory volume (FEV  _,.) was diminished by exposure to air
                             U • / 3


       pollution.



5.     Asthma Attack Rates - Temperature changes were a stronger determinant
                                33

-------
            of asthma attack rates than any particular pollutant.  No
            relationship with SO,, and asthma attach rates was found at any
            temperature.  Suspended sulfate levels demonstrated the only
            consistent relationship with daily aggravation of asthma and
            cardiopulmonary symptoms.  The "best judgment" of the authors
            was that suspended sulfate exposure as low as 8-10 p,g/m  for
            24 hours could be a contributory factor-to "significant aggra-
            vation of pulmonary symptoms"  (Health, 1974).
      There are certain methodological weakness  in  some of the CHESS studies,  However
      the CHESS  results  are  not  to be considered invalid by  any means.

      2.     Chicago Air Pollution Studies
            For over a decade, the Chicago Air Pollution Study Group has been
examining  the  health effects of air pollution, particularly to define those
individuals in the population most sensitive to air^pollutants, and the levels
of pollutants  at which their health is adversely affected.  The results of
the earlier studies are consistent with the CHESS results, however, suspended
sulfates were  not considered.  Though the studies were independent, similar
methodologies  were employed.  Moreover, the air pollution mix in the New
York  area  is not unlike that of Chicago.
            The Chicago Chronic Bronchipulmonary Disease Registry Study began
August,1966 (Carnow et al, 1969) with a total of 571 bronchitic patients from
16 different facilities.  The patients were classified according to the
severity of their disease, and each maintained a daily record of acute chest
illness.   Their records were then correlated with data of the city of Chicago
for each sg_uare mile of the city for every 15 minutes of the day.  Patients'
exposures  were determined by the estimated level of pollution in the residence
and occupation for each 24-hour period.
                                     34

-------
Illnesses appeared to correlate with levels of sulfur dioxide, with illness



increasing at each of 7 levels of SO  pollution.  At 0.24 ppm for 24 hours,
                                    £*


there was more than twice as many acute chest illnesses as when the level was



0.04 ppm.  It appeared that when SO  was considered as a pollution index-in
                                   •Zt


males 55 and over with advanced bronchitis, there was a relationship between



levels of pollution and frequency of acute chest illnesses.



     Recently, multiple regression analysis of personal air pollution



exposure has been completed.  Of special interest to this review is the



relationship between maximum temperature, windspeed, personal SO  exposure and



acute illness for all respiratory diseases.  The dependent variable was



the percent of excess emergency room visits  (ERV) for all respiratory diseases



combined.  Using .03 ppm  (annual mean) - SO  explained 4.7% of all emergency



room visits  (ERV) for respiratory conditions   using  .14 ppm - it explains



66% of all visits.  Data from 48 days are included in the calculations.



     3.     Hamilton Ontario Air Pollution Studies



            A  Canadian retrospective hospital admissions study of Hamilton,



Ontario, a steel-producing city of about 350,000 people, found a "strong



relationship between hospital admissions for acute exacerbations among adults



with chronic respiratory illness and among children with acute respiratory



disease."  This study used an air pollution index  (API) which included SO



and particulate measurements and climatological data.



     4.     Allegheny County, Pennsylvania Air Pollution Study



            Another recent study with a very good data base and methodology



by Carpenter and associates investigated the relationship between hospital



costs and exposure to air pollution in Allegheny County, Pennsylvania



(Carpenter et al., 1977).
                                     35

-------
     After correcting for race, age, sex, smoking habits, neighborhood




income and occupation, they found that respiratory and suspect circulatory




system disease showed statistically significant increased hospitalization




rates (P < .01) and lengths of stay for those exposed to higher levels of




SO   (>99.3 ng/m  or .035 ppm) and particulates (>115 M9A1 ) compared to




those from neighborhoods meeting air quality standards.  Control diseases




were not affected by the air pollution index.  Using the area's average costs




per day for hospitalization, they estimated total increased costs of hospital-




ization for the 1.6 million persons in Allegheny County, PA, to be $9.8




million for 1972 ($9.1 million for increased hospitalization rates and $0.7




million for increased length of stay)  (Carpenter et al., 1977).



C.   Mortality Studies




     Deaths from air pollution is usally the end of a cumulative process of



stresses and insults and, as such, is a much less precise indicator of the




adverse effects of air pollution.  Two studies of mortality were carried out



by our group some years ago and a third more recently.  The first divided




square miles of Chicago into low, moderate, and high levels of SO  and
                                                                 £


compared deaths from cardiac and respiratory disease in each square mile of



the city; then grouping them into these three categories.  Excess deaths were




found in the highly polluted from both respiratory and cardiac causes.  No




effort was made to standardize in this study for socio-economic differences.




In 1969 an air pollution episode occurred in Chicago.  During this ten day




episode there were excess deaths in a number of categories, particularly




from respiratory diseases in white males and some cardiac disease categories




including rheumatic heart disease, hypertension and in older black males




ischemic heart disease.  Past mortality studies have been summarized by Gold-




stein (NAS, 1975) and Finklea  (1975).  Generally, lower  levels of air pollution
                                     36

-------
have yielded less consistent results in mortality studies than in morbidity



studies.



    Two recent studies were carried out by us examining the impact of



individual pollutants on morbidity and mortality.



     In regards to mortality - TSP - was significantly related to disease while



S02 was not when chronic exposure was considered.



     When a 25 percent reduction in TSP, which is almost equivalent to the



percentage reduction in TSP in Chicago for the period 1970-75, was applied



to the models developed, the age-adjusted death rate for all non-accidental



causes would be decreased by 5.36% (54.65 deaths per 100,000 persons) in



Chicago.  A decrease in the death rate by cause was estimated to be 8.85%



Call heart diseases), 6.42  (ischemic heart diseaseX, 16.95%  (other heart



disease), 26.16%  (emphysema) and 6.47%  (other non-accidental causes).



The implications are great and attempts are being made to extend the study.



     Models developed in daily analysis also imply that there would be possible



acute effects of daily air pollution concentrations  (both SO  and TSP, in



addition to their interaction) on daily mortality changes, controlling for



weather changes and day-of*-week effects.  Based on the model for the day of



death onset, it is estimated that a 25 percent reduction in daily levels of



each pollutant would decrease daily non-accidental deaths by 1.815%  (due to



SO ), 2.045% (due to TSP) and 0.867% (due to an interaction between SO  and
  2.                                                                   £


TSP) in the city of Chicago.



     Models for heart disease indicate that the number of daily deaths caused



by heart disease could be affected by levels of SO , TSP and their inter-



action. Based on the model for the day of death onset, it is estimated that



a 25 percent reduction in daily levels of each pollutant would decrease daily



deaths from heart disease by 1.717% (due to SO ), 2.048%  (due to TSP) and



0.940%  (due. to an interaction between SO  and TSP) in the city of Chicago.


                                     37

-------
                              THE MORBIDITY STUDY




     A morbidity study in which linear regression models have been developed




to quantitatively estimate the degree of the air pollution contribution to




emergency room visits for cardiac and respiratory diseases in two major




hospitals in the city of Chicago was also just completed.




     According to the results, sulfur dioxide based on patient exposure




levels can account for about 13% of the variation of emergency room visits




for acute bronchial and lower respiratory infections and about 22% for total




cardiac diagnoses.





     Table II summarizes other studies on the effects of SO2  on humans.
                                     38

-------
                                   Table II    Summary of Selected Humans Studies with SO. Alone or in Combination with Other
                                                Pollutants (sulfates, acid mists, and virus)
Type of Subjects

11 healthy male adults






9 healthy adults




Total Study of
25 healthy adults
as total
a. 13 adults


b. 12 adults


c. 17 adults

d. 10 adults



Compound and
Concentration
SO, 1,5,13 ppm
A





SO. 0.5, 1.0 and
5.00 ppm






SO 1.00 to 1.17 ppm


SO2 1.00 to 1.17 ppm


SO. 2.8 to 3.3 ppm

SO, 30 ppm



Exposure
Time
10-30 min.






30-60 min.




Total Study
4 year
sequential
1 hour


25 vital
capacity
breath
8 deep breaths
from a boy
10 minutes



Results

At 1 ppm slight change in pulmonary flow
resistance in one subject. At 5 ppm change in
pulmonary flow resistance in all subjects. At
13 ppm greater change in pulmonary flow rate.
No change in pulmonary compliance, tidal
volume breathing rate or pulse rate, only the
frequency increased at 13 ppm exposure.
SO, caused a decrease in maximum expiratory
flow (MEF) , and 50% decrease in vital capacity
(VC) at 1 and 5 ppm. With water aerosol a
decrease in KEF was also observed at 0.5 ppm
scy



No change in pulmonary function but signifi-
cant difference in airway resistance (SRAW)
in one subject
Difference in airway resistance


SRAW increase by a factor of 3

There was a wide range of sensitivity. SO.
by itself could be a contributing factor to
change in lung function or exacerbation of
bronchitis..
References

Frank, 1961






Snell and
Luch singer,
1969





Lawthar, 1975











VO

-------
Table  II   Continued
Type of Subjects
7 healthy adults
9 healthy non-
smoking adults
15 healthy males
5 healthy males
5 healthy males
4 healthy males
Compound and
Cone en tration
SO 16.1 ppm
(average)
SOj 5.0 ppm
SO. 1.0,5.0 and
25.0 ppm
H SO mist of
particle size
0.35u at cone.
of 5.0 mg/m
H S04 (1.5 u -
30 u particle
size) 3-
•3-38 mg/m cone.
at 91% relative
humidity
London air from
1965 to 1971
Exposure
Time
25-30 minutes for
5 exposures
20 breaths "by
mouth
6 hours in each
concentration
15 minutes
183 exposures each
of 16 min. by
mask & 31 expo-
sures each of 60
min. in chamber
2-5 km. walk to
work
Results
All SO was absorbed in nasal passages, less
coughing, less irritation to throat, fewer
and smaller increases in flow resistance than
when exposed to same cone, by mouth
No significant effect on ciliary mucus
clearance
Significant decrease in ciliary mucus flow
at 5.0 and 25.0 ppm
Respiratory rate increased. Maximum
expiratory flow rate decreased by 20%,
tidal volume decreased by 28%
Increase airway resistance 35-100% above
normal rate and much increase up to 150%
in high humidity
During first year study one subject showed
increase airway resistance with pollution/
relative humidity and lower temp. After
the 1st year no effects were found.
References
Speizer and
Frank, 1966
Wolff, 1975
Anderson,
1973
Amdur, 1952
Sim and
Pattle, 1957
Lawther,
1977

-------
Table
            Continued
Type of Subjects
a healthy mala
.students
10 human adults
Compound and
Concentration
SO. 0.37 ppm
0. 0.37 ppm
separately then
combined
S02 S ppm
Exposure
Time
2 hour
period
4 hours
Results
More adverse effect due to combination
50% decrease in nasal mucous flow rates
References
Bates, 1975
Anderson ,
1973

-------
                                    SUMMARY




     Animal and human toxicologic studies have, identified many of the




mechanisms of adverse health effects of sulfur oxides.  Pure SO  is an




irritant gas that produces irritation and inflammation of the tissue that




it directly contacts.  The principle observed effects of acute exposure are:




1) airway resistance, (2) mucociliary impairment,  (3) an acute bronchospastic




effect , and (.4) by the above, interference with breathing.  Chronic exposure




may cause chronic damage to the respiratory system.  Acid sulfates are also




classified as irritants and are more potent than SO .  The health effects of




other sulfates, sulfites and bisulfites levels  not yet been determined, but  there is  some




evidence they may be mutagenic (Hickey et al.,  1976).  The known target organs




in humans for SO  and irritant sulfates are principally the lungs and second-




arily the heart because of its high oxygen need and its interrelationship




with the respiratory system.




     The best available information on the acute and chronic adverse health




effects of sulfur oxides comes from epidemiologic studies carried out in the




real environment.  Some of these studies have shown adverse health effects at




below ambient standards.  At present epidemiologic studies have observed




associations of SO  exposure and other pollutants with:




     1. increased mortality from cardiac and respiratory disease




        (Carnow and Namekata, 1977; Goldstein,  1975; Rail, 1974).




     2. increased sputum cellularity, in healthy adults, indicating




        inflamed lung tissue (Nobutomo, 1978).




     3. increased incidence of chronic respiratory disease  (asthma, bronchitis,




        emphysema) and possibly cancer  (Goldstein, 1975).




     4. increased incidence of acute respiratory attacks among those at




        high risk because of chronic pulmonary disease  (Carnow,  1969).







                                     42

-------
     5. increased incidence of cardiac death among those at high risk



        because of cardiac disease  (Carnow, 1969).



     6. increased rate of hospitalization and increased length of hospital



        stay  (Carpenter, 1977)



     7. increased emergency room admissions on days of high pollution



        CCarnow and Namekata, 1977)



     8. increased loss of work days due to respiratory distress (CHESS, 1974).



     9. increased absence from school due to respiratory distress  (CHESS,



        1974).



     Those people at high risk include infants and children, male adults,



aged 55 years and over with severe  (advanced) chronic bronchitis, ismokers,



and "marginal" people with poor adaptive capacity, such as those suffering



from chronic  lung disease, heart disease, asthma, and certain congenital



diseases.



     Laboratory tests have generally used much higher concentrations of SO
                                                                          «^


to produce adverse response than have epidemiologic studies.  These higher



levels are required in laboratory tests because: 1) They often use SO  by



itself, whereas the ambient air SO  converts to the more toxic sulfuric acid



and often sulfur products, and in the presence of high humidity and/or other



pollutants this conversion may be relatively rapid. 2) The test use animal



species and strains of species that may be more resistant to SO  than humans.



3) When humans are used, they are usually normal young adults.



     Studies  are needed to further quantitate the impact of SO  as acid



sulfate and its contribution to morbidity and mortality as a part of a mix of



pollutants and meterologic variables.



     While the available data base is not yet sufficient for precise dose/



effect, functions, a recent Chicago air pollution study allows a projected
                                     43

-------
linear relationship between acute morbidity and personal exposure to S0_




based on actural results.  This, together with previous estimates, (Fishelson




and Graves, 1977} can assist policy makers in decision making.
                                     44

-------
                        CONCLUSIONS AND RECOMMENDATIONS



     1. There is not health justification for relaxing present SO  ambient



standards, nor is there likely to be.  Moreover, present SO  standards may



not be protecting high risk groups and hypersensitive individuals.



     2. While the data base on sulfates from epidemiologic studies is



insufficient, animal and human toxicologic studies have shown that many



sulfates are unquestionably considerably more toxic than 'sulfur dioxide.



There is also no question that there is a build-up of sulfates in the air  of  certain



regions of the U.S. notably the Northeast. It also appears that sulfur dioxide levels



are not good measure of ambient sulfates.  For those, reasons it is clear that a




separate sulfate standard is desirable to protect health.   Since adverse



health effects were observed at levels of 8 to 10 ^Ug/m  it would be prudent



to promulgate a temporary standard, possibly at half of that or of 4
annual mean, not to be exceeded more than one percent of the time during




the course of a year.




     3. Reductions of ozone and particulate which appear, to act synergesti-




cally with SO  might result in reduced negative health effects of all 3




pollutants.



     4. Additional toxicmlogic and epidemiologic research on possible mutagenic,




carcinogenic and cocarcinogenic effects of SO , sulfates, sulfites and bisulfites
                                             ^



is indicated .
                                     45

-------
                                  REFERENCES


Amdur, M.O.:  "Aerosols Formed by Oxidation of Sulfur Dioxide:  Review of
Their Toxicity".  Arch. Environ, Health.  23'459, December, 1971.

Amdur, M.O.:  Toxicologic Appraisal of Particulate Matter, Oxides of
Sulfur, and Sulfuric Acid:.  J. Air Pollution Control Assoc.  19  (9):
635, September, 1969.

Bates,D.V.:  "Air Pollutants and the Human Lung".  Amer. Rev. Resp. Pis.
105:  1-13, 1972.

Carnow, B.W. Shekelle, R.B., Lepper, M. and Stamler, J.  "The Chicago Air
Pollution Study: SO2 Levels and Acute Illnesses in Patients with Chronic
Bronchopulmonary Disease".  Arch . of Environ. Health  18, 768, 1969.

Carnow, B.W.:  "Air Pollution and Respiratory Diseases".  Scientist and
Citizens, 8:  1, May, 1966.

Carnow, B.W. and Fievenson, S.1:  Morbidity and Mortality during the
Chicago 1969 Air Pollution Episode".  Unpublished paper.

Carnow, B.W. and Meier, P.:  "Pulmonary Cancer".  Arch. Environ. Health
27: 312, September, 1973.

Carnow, B.W. and Namekata, T.:  Impact of Multiple Pollutants on
Emergency Room Admissions.  Illinois Institute for Environmental
Quality,  Document No. 77/02, February, 1976.

Carpenter, B.H., LeSourd, D.A., Chromy, J.R. and Bach, W.D.:  Health Costs
of Air Pollution Damages.  EPA-6—/5/77-006, February, 1977.

Cassarett, L.;  Chapter 9:  "Toxicology of the Respiratory System".
pp. 201-223.  In Toxicology:  The Basic Sciences of Poisons. Edited by
L. Cassarett and J. Douell.  New York:  MacMillan Co.,  1975.

Finklea, J.  "Summary of Health Effects of Increasing Sulfur Oxide
Emissions".  Publiched as appendix to Fed. Power Commission National
Power Generation Conservation, Health and Field Supply, 1975.

Fishelson, G., and Graves, P.:  "Air Pollution and Morbidity:  SO2
Damages."  An unpublished paper, prepared for the IIEQ, September, 1977.

Goldstein, B.,  Chapter  1,2,3,4:  "Introduction, General Consideration,
Basic Bioledical Effects of Sulfur Oxides, Health Effects of Sulfur
Oxides".  In Air Quality and Stationary Sources Emission Control,  A
Report by the Commission on Natural Resources, National Academy of
Sciences , National Academy of Engineering, National Research Council.
Washington, D.C.,  March, 1975:  U.S. Government Printing Office, 94-4.

Greenburg, L., Field, F., Erhardt, C.L. and Reed, J.L.:  "Air Pollution
Influenza and Mortality in New York City".  Arch, Environ. Health.  15:
430, 1967.
                                    46

-------
Health Consequences of Sulfur Oxides:  A Report from CHESS, 1970-1971,
EPA-650/174-004, Washington, D.C., U.S. Government Printing Office, May,
1974.

Hickey, R.J., clelland, R.C., Bowers, E.J. and Boyce, D.E.:  "Health
Effects of Atmospheric Sulfur Dioxides and Dietary Sulfites".  Arch.
Environ. Health, 31:  108, 1976.

Logan, W.P.D.:  "Mortality in-ftie London Fog Incident, 1952".  The
Lancet   pp. 336-338, February, 1953.

Morbidity and  Mortality During the London Fog of December, 1952.  Reports
of Public Health Medical Service No. 95.  London, Her Majesty's Stationery
Office, 1954.

Morris, J.F., Kosk, A. and Johnson, L.C.:  Spirometric Standard for
Health Nonsmoking Adults".  Amer. Rev. Resp. Pis.  103:  57-67, 1971.

Neal, A.M.D.:  "Quantitating the Effects of Environmental  Stress:
Alternation in Disease Rates".  Presented at Conference on Energy
Utilization and Environmental Health, School of Public Health, University
of Illinois, March, 1977.

Nobutomo, K.:  "Air Pollution and Cytological Changes in Sputum".  The
Eaneet.  8061:  523-526, March, 1978.

Rail, P.P.:  "Review of the Health Effects of Sulfur Oxides".  Environ.
Health Prespec.  8:97-121, 1974.
                                    47

-------
        ENERGY, ENVIRONMENTAL, AND ECONOMIC IMPACTS OF
           FLUE GAS DESULFURIZATION UNDER ALTERNATIVE
                NEW SOURCE PERFORMANCE STANDARDS

                                    by

                 Andrew J. Van Horn, Richard A. Chapman,
                    George Ferrell, Peter M. Cukor, and
                              David B. Large
                              Teknekron,Inc.
                            2118 Milvia Street
                           Berkeley, CA 94704
ABSTRACT
     The  energy, environmental,  and  economic impacts of flue  gas  desulfuri-
zation  (FGD)  under alternative  revisions  to  the  New  Source  Performance
Standards  for  coal-fired  utility boilers have been examined  using the Utility
Simulation Model (USM).  The USM simulates investment and operating decisions
related to choices of fuels and pollution control  equipment through the use of
extensive  data  bases,  and cost and  performance models for pollution control,
including control of SO-, particulates, and NO  .  For each of  the 48 contiguous
states in the U.S.,  alternative projections have been made of the impacts of
utility operations from 1976 to beyond the year 2000.

     The  SO^  control  technology and cost model  is structured to calculate
capital costs,  variable operating  costs, and capacity penalties for limestone,
lime, and magnesium oxide FGD systems installed in module sizes of between 50
and I30MW each,  except for systems of  less than 50 MW.   All systems of
IOOMW or more  include a  spare module  for  added reliability.  The model  is
applied on a unit-by-unit basis utilizing a data base with detailed information on
each existing and announced coal-fired utility boiler in the U.S.

     In  calculating FGD  system costs and penalties, the  model considers both
the gas flow  rate  and the quantity  of $62  to be removed.   This level of
sophistication makes it possible, for instance,  to compare  FGD  costs for the
same coal at various emission  limits,  or for various coals with the same sulfur
content  but different heating  values.   This model is  part of both the Coal
Assignment Model  and the USM planning module  and is used in selecting a fuel
and pollution control  strategy as well as calculating the  operating and  cost
functions for individual generating units in each  year of the simulation.

     In  this paper we summarize the recent sensitivity studies  performed for
EPA concerning the revised New Source Performance Standards to be established
for coal-fired electric utility boilers.
                                   48

-------
         ENERGY, ENVIRONMENTAL, AND ECONOMIC IMPACTS OF
           FLUE GAS DESULFURIZATION UNDER ALTERNATIVE
                NEW SOURCE PERFORMANCE STANDARDS

INTRODUCTION

     Over the past three years, under the auspices of the U.S. Environmental
Protection Agency, Teknekron has developed and applied its Utility Simulation
Model  (USM) to examine a variety of energy and environmental problems. For
the  past  18 months we have  used the  USM  to review  the economic and
environmental impacts  of various revised New Source Performance Standards
(NSPS) for SO2 and particulates from coal-fired electric utility boilers.

     This paper contains a  summary of  the recent results of  our NSPS Phase
Three analyses,    which focused on critical  uncertainties surrounding a number
of key factors that will influence future  impacts of the revised NSPS.  These
factors will affect utility costs and hence will  influence the coal choices and
pollution  control   measures  adopted  by  utilities in  response to  alternative
standards.  Teknekron has carried out city-specific analyses of utility coal and
pollution  control   choices  and their sensitivity  to  the  factors of  interest.
Complementing these sensitivity analyses  are our state, regional, and national
impact projections for alternative standards for the period from 1976  to the year
2000.

     Key elements we have varied include coal mine prices, coal transportation
rates,  coal sulfur  and Btu contents,  and  the costs and  performance of FGD
scrubbers.  In each case, the selected range of variation  reflects  the element's
degree of uncertainty and sensitivity to critical issues.

     The implications  of these  variations  for  the Environmental  Protection
Agency's (or  anyone else's) ability  to distinguish  between similar  standards are
discussed.  Also discussed are the  sensitivities  of several  cost-effectiveness
calculations (for example, cost per ton of 502 removed) which have been posited
as measures of the  worth of various standards.

-------
      The impacts of revised standards will depend not only on utility coal and
pollution  control choices  but also on such  factors  as the  future  growth  in
electricity demand, the amount of nuclear capacity, the phasing out of gas steam
plants, and the price of oil.  These factors are themselves subject to uncertainty.
In our projections for 1976 to 2000, we have used the latest assumptions made
                                                       (2)
for these parameters by the joint EPA/DOE  working group.
UTILITY SIMULATION MODEL DESCRIPTION

      The Utility Simulation Model  consists  of  a number of  interconnecting
computer modules and data bases that simulate decisions for system planning and
operation, utility finance, and the operation of individual technical  processes.
The model  is  driven by a  set of exogenous scenario elements  that include
electricity demand levels, financial market conditions, fuel  prices and availa-
bilities, advanced technology deployment, and environmental  regulations.   For
each scenario, the model calculates the following by  geographic region (county
or state) for future years up to 2010:

      •   Factor demands, including
          -      fuel use, by type and for coal by region of origin
          -      electricity generated
          —      Capital requirements, by source (e.g., debt, common
                 equity,  preferred equity)
          —      plant and equipment requirements
          -      releases of  air and water pollutants  and  generation
                 of solid wastes
      •   Financial statistics for utility firms
      •   Average electricity prices

      In order to produce these calculations at the required level of detail, the
model considers generating unit sites located in each county where electricity is
produced, fuel and water are consumed, and pollutants are  released.   Since
utilities operate as integrated systems, the  model  presently  simulates  joint
                                     50

-------
operation (i.e., dispatching) of  all generating units within a state.  Finally, the
responses of utility firms to the external environment  in  which they function
may be changed  by the model  user modifying present data bases or  specifying
alternate choices for future system planning and system operation.  For example,
the particular scenarios  evaluated in the New Source  Performance Standards
Review encompass a range of  futures for electricity demand, fuel  selection,
choices of technology,  and pollution control regulations  as specified by the U.S.
Environmental Protection Agency.

     Figure  I is  a simplified diagram of  the  Electric Utility Simulation  Model.
The model  includes the  following major components:

     •   Demand projection, including
          —     retail  and wholesale sales and purchases
          -     energy generation, i.e., average load growth
          -     peak load growth
     •   System planning, including
          —     choice of generating unit type
          -     choice of fuel  type,  quality, and for coal by region
                of origin
          -     choice of pollution control technology
          -     expansion of transmission and distribution networks
          -     siting  of generating units
     •   Dispatch, including
          -     calculation of unit capacity factors for  each typical
                day of operation, by class of unit
          —     calculation of  total  fuel, operation,  and  mainte-
                nance  expenses for electricity generation
          -     projection of fuel consumption, by type  and for coal
                by region of origin
         -     pollution  control costs and operating characteristics
                for the various types of pollution control devices
     •  Financial, including
         -     integration of  projected  production expenses with
                construction expenditures
                                     51

-------
                    Figure 1
          Electric Utility Simulation Model
                    DEMAND


PLANNING
I
.1

               EtECTRtC UTILITY
              SIMULATION MODEL
RESIDUALS

                                       1

              ENERGY, ECONOMIC
              AND ENVIRONMENTAL
                   IMPACTS
                       52

-------
          -     projection of the firm's balance sheet, income state-
                ment, sources and uses of funds, and other  financial
                statistics
          -     calculation  of  annual  revenue  requirements and
                electricity prices
          Residuals, including
          -     projection of release rates  at the generating unit
                site  for  numerous air and water pollutants and for
                solid wastes
          —     projection of  consumption  of  water  and  other
                resources
     Teknekron's coal assignment model (CAM) is used to select coals to be used
in the USM for each state and regulatory category of coal plant (e.g., SIP, NSPS,
RNSPS).  A maximum of 50 distinct coals from  12 different supply regions may
be considered for use in each of the 48 states.

     The criterion for coal selection is least levelized annual cost, where cost is
determined on an  "as-burned" basis.  Since the program  is  interactive,  the user
may exercise expertise in coal selection and/or knowledge of  historical coal
movements to force the selection of a coal  other than the  least cost coal.  To
facilitate the selection process, CAM produces a supplementary report which
includes complete cost information about the first five choices ranked on a least
cost basis.

     The determination of "as-burned" cost  considers the entire fuel-cycle. The
CAM program tracks a candidate coal along  the fuel cycle from mine production
to combustion in the utility plant boiler including particulate control and flue gas
desulfurization.  In calculating the total "as-burned" cost, the CAM considers the
following fuel cycle component costs:

     •    FOB mine price
     •    Cost of coal cleaning (where applicable)
     •    Transportation and handling charges
     •    Cost of boiler modification (where applicable)
                                     53

-------
     •   Particulate control cost
     •   Cost of flue gas desulfurization

     In addition  to  the  above costs, the user may define a "premium" to be
added to the  delivered cost of any coal.  This premium is a  surrogate for
institutional factors  or other influences upon the marketability of a coal which
are either imprecisely defined  or otherwise not amenable to economic modeling.
The total "as-burned" cost  is defined as the sum of the component costs in the
fuel cycle.
FGD COST MODEL

     Teknekron has developed FGD cost and performance models  based on
PEDCo and TVA engineering and cost estimates for  lime and limestone systems
and PEDCo cost estimates for magnesium oxide systems.  ' '    The models can
be used to predict new or retrofit FGD costs for  generating plants  of between
25 MW and 2,000 MW in size burning coal of any sulfur content and meeting any
emission limit.

     The three FGD  systems are  modular in  design, with  module sizes of
between 50 MW and 130 MW except for plants of less than 50 MW in size.  One
redundant module is included for all  systems of 100 MW or greater for a  design
reliability of 90 percent.  The design of  the three FGD systems is  based on a
three-stage turbulent contact absorber (TCA).

     The TVA and PEDCo FGD cost estimates represent a reasonable range of
costs for use in our sensitivity studies. The PEDCo costs are higher than TVA's
and are probably representative of costs that may be used by utilities without
extensive experience with FGD systems.  The TVA costs, on  the other hand, are
less conservative and represent cost estimates that may be used in the future by
utilities which have had  favorable FGD experience.  These  two cost  estimates
may also be viewed as representing two points on the FGD "learning curve" with
the lower cost estimates indicative of lower, future FGD costs.
                                    54

-------
     Tables I  and 2  present PEDCo and TVA capital  and operating cost esti-
mates for a 500 MW limestone FGD system designed for 85 percent SC^ removal
from a  medium sulfur content coal with a 2k hour averaging time.  The total
capital  investment estimated by TVA is about 60 percent of PEDCo's estimate.
Total operating and maintenance costs (less fixed costs) are estimated  at about
6.0 and  4.7 million dollars respectively by PEDCo and TVA.

     The cost of electricity and steam required to operate the FGD system is
not calculated by the FGD  cost model; instead, electricity and steam  require-
ments are used to calculate unit capacity  penalties and  are accounted for in this
manner  by the Utility Simulation Model. For the case illustrated in Tables I and
2, the TVA capacity penalty is 2.96 percent, and the PEDCo capacity penalty is
4.25 percent.  Either  of these estimates is reasonable and representitive of the
range of capacity penalties which might be expected.

     Within  the  model,  plant characteristics, coal properties,  and  emission
limits are used to determine the required rate  of sulfur dioxide removal  in
pounds  per hour and the required gas flow rate in actual  cubic feet per minute
for an FGD system having an annual average removal efficiency of 90 percent or
greater. If a given generating plant needs to remove less  than 90 percent of the
SO- produced  to meet applicable emission  limits, an  FGD system  with  an
efficiency of 90 percent will be used to scrub a portion  of the flue gas. The
remaining flue gas will be bypassed and mixed with the scrubbed gas to yield the
required SO2 emissions and to reduce or eliminate the fuel required for reheating
the flue gas.  If 90 percent or more of the SO- must be removed, an FGD system
having  the required  efficiency up  to the limits of technology will be used to
scrub the entire flue gas stream.

     The cost of  such  equipment  as  pumps,  hold  tanks,  feed preparation
equipment, and sludge ponds is based on  the sulfur dioxide removal rate, while
the cost of such  items as fans, absorbers, and soot blowers is based on  the gas
flow rate.  Likewise, operating  costs are based  on either the sulfur  dioxide
removal rate (e.g., raw material)  or the gas flow rate  (e.g., electricity, reheat
steam or oil).
                                    55

-------
    Table I.  Comparison of TVA and PEDCo Limestone FGD Capital Costs*
Capital Cost Item
                                     PEDCo**
TVA**
Direct costs
Limestone preparation
SO 2 scrubber
Sludge disposal
Sludge pond
Raw material inventory
Total direct costs
Indirect costs
Contingency and fee
Total capital investment

$ 2,423,800
21,012,600
1,201,900
5,632,800
162,600
$30,433,700
9,271,900
10,283,000
$49,988,600

$ 3,322,100
14,786,800
2,248,900
o***
0
$20,357,800
7,348,700
3,053,700
$30,760,200
*#
***
Note:
Basis:  Coal sulfur content =  2.50 Ibs S/IO  Btu
       Sulfur RSD  = 0.15, no exemptions   ,
       Design sulfur content = 3.63 Ibs S/IO  Btu
       Plant size = 500 MW
       Five scrubber modules at 125 MW each
       85 percent 24-hour average S0? removal
        1975 costs and dollars

Costs predicted by  Teknekron's SCU control  model.   Not  included are
interest during  construction,  working capital,  and taxes; these are calcu-
lated in the Utility Simulation Model's financial module.

Sludge pond capitalization included in sludge  disposal operating cost (see
Table 2).

    More recent estimates by TVA include about $7 million for the sludge
    pond and a contingency and fee of  25 percent of total direct costs.
    Total  TVA  investment  estimate  is therefore  increased  to  about
    $42 million.
                                    56

-------
   Table 2.  Comparison of TVA and PEDCo Limestone FGD Operating Costs*
Cost Item                                  PEDCo**              TVA**
Limestone
Labor
Maintenance
Water
Sludge disposal
Analysis cost
Total O&M costs
$ 804,400
406,500
3,736,600
38,000
996,100
0
$5,981,600
$ 769,900
783,400
1,816,800
21,800
1,219,700
69,400
$4,684,000
*    Basis:    Coal sulfur content = 2.50 Ibs S/IO  Btu
              Plant size = 500 MW
              85 percent 24-hour average 502 remova'
              Capacity factor =  0.65
              1975 costs and dollars

**   Costs  predicted  by  Teknekron's  SO^ control  model.  Not  included are:
     (a) steam and electricity costs, which are used in the Utility Simulation
     Model  to calculate  capacity penalties; and (b) fixed charges, which are
     calculated in the  Utility Simulation Model's financial module.

Note:     More recent  estimates by TVA include a higher cost for maintenance
          (due to higher capital cost) and sludge disposal.  Total TVA operating
          cost estimate is  now about the same as the PEDCo estimate.
                                    57

-------
     Outputs from the FGD model include:

     •    Capital cost
     •    Fixed operating cost (independent of plant capacity factor)
     •    Variable operating cost (dependent on capacity factor)
     •    Removal efficiency
     •    Scrubber size
     •    Capacity penalty (plant capacity used to operate the FGD
          system)
     •    Heat rate penalty (accounts for fuel required  to operate
          the FGD system)
     •    Water used and water cost
     •    Oil used for magnesium oxide regeneration
     •    Oil used for reheat
     •    Annual sludge generation

     SO2 emissions are calculated on the basis of the uncontrolled emission rate
and the  required removal efficiency.

     Input data required for the FGD model include:

     •    Individual generating-unit characteristics
          —     size
          -     age (new or retrofit)
          -     heat rate
     •    Coal  properties
          -     heating value
          -     composition (C, H, 0, N, S, H-O^ash)
          -     class (bituminous, subbituminous, lignite)
     •    Environmental factors
                                    58

-------
          -     emission limit (specific limits: percentage removal,
                ceiling, floor, and averaging time)
      •   Economic factors
          -     year scrubber was built (escalation, inflation)

SENSITIVITY STUDIES

      The Coal Assignment Model and the Utility Simulation Model were used to
determine the sensitivity of future  impacts of  various  revised New Source
Performance  Standards to uncertainties in  future  coal  prices, transportation
rates, coal properties, and FGD costs.
City Specific Analyses

      City specific analyses were conducted for  key locations (e.g.,  Columbus,
Ohio; Indianapolis, Indiana; Orlando, Florida; and Austin, Texas) to determine the
effect of various SO- standards on the levelized fuel-cycle cost for various coals
and  the sensitivity  of these  effects to  uncertainties  in  future  coal  prices,
transportation rates, coal properties, and FGD costs.

      The sensitivity of levelized fuel-cycle cost with respect to the annual SO^
ceiling for Powder River coal and Northern Appalachian coal used by  a utility in
Columbus, Ohio is illustrated in Figure 2. For ceilings greater  than about 0.65 Ib
SOy 10  Btu, Powder River coal is less expensive and for ceilings less than 0.65,
Northern Appalachian coal is less  costly to use.   Figures 3 and 4 show the
sensitivity of the levelized fuel-cycle cost with respect to 24-hour 50^ floor and
respectively transportation rate and FOB mine price.  As would be  expected, the
fuel-cycle cost  of Powder River coal is more sensitive to transportation rates,
and Northern Appalachian coal is more sensitive to FOB mine price.

      Another important consideration is  the sulfur content and heating  value
assumed for Powder River coal.  Figure  5  illustrates the sensitivity of levelized
fuel-cycle cost  with respect to various  24-hour $©2 floors and coal  character-
istics typical for Powder  River coals. These curves represent coals with sulfur
                                     59

-------
                                  Figure 2
          Sensitivity of Levelized Fuel-Cycle Cost with Respect to
                             Annual SO2 Ceiling
                              (Columbus, Ohio)
    400-1
?   350-
(A
o
u
IU
u
V
u
Q
U
N

IU

IU
300-
    250-
                                            —— Powder River coal
                                                  0.5% sulfur, 6.0% ash, 8,100 Btu/lb

                                            	Northern Appalachian coal
                                                  2.6% sulfur, 9.9% ash, 12,000 Btu/lb
               I         I          I         T          I
    0         0.2       0.4        0.6        0.8        1.0

                    ANNUAL SO2 CEILING (LB S02/10« BTU)

    a) Assumes no mandatory percentage removal requirement
                                                                    1.2
                                   60

-------
                                 Figure 3
        Sensitivity of Levelized Fuel-Cycle Cost with Respect to
              24-Hour SO2 Floor arid Transportation Rate
                             (Columbus,  Ohio)
    400-
o>
JL  350.

eo
&
en
O
U
LU
_l
U

U
111

"•  300-
D
HI
N
 LLJ
 >
    250-
                                      —— Powder River coal
                                            0.5% sulfur, 6.0% ash, 8,100 Btu/lb

                                      	Northern Appalachian coal
                                            2.6% sulfur, 9.9% ash, 12,000 Btu/lb
                        $ 1978
                    Rale (e/lon-mlle)
               Hall (miles)

               -250   >250
                               Water
             A 2.00
             B 2.25
             C 2.50
1.00
1.10
1.20
(all distances)

    0.4
    0.5
    0.6
         1    'Rail Only
       T-
                   i
                 0.6
                      I
                     0.8
 I
1.0
                  0.2        0.4

                    24-HOUR AVERAGE SO2 FLOOR (LB SO2/10« BTU)
1.2
a) Assumes 1.2 Ib SO2/106 BTU ceiling with 85% removal, 24-hour average with three day per
   month exemptions.
                                   61

-------
                                    Figure 4
           Sensitivity of Levelized Fuel-Cycle Cost with Respect to
                24-Hour SO2 Floor and F.O.B. Coal Mine Prices
                                (Columbus, Ohio)
       400-1
  00
  r>-
  O)
  m
  ID
  O
O
u
111
O
V
u
 I
UJ

u.
O
UJ
N

UJ
UJ
       350-
       300-
                                        ——  Powder River coal
                                              0.5% sulfur, 6.0% ash, 8,100 Biu/lb

                                        	Northern Appalachian coal
                                              2.6% sulfur, 9.9% ash, 12,000 Btu/lb
                         F.O.B. Coal Prices (S/ton)
                         S1978   -10%   -10°/o
250-
^m
^H
I
w
wm
)
PR
NA
I
0.2
6.75
23.00

7.43
25.30
I
0.4
6.08
20.70
I I
0.6 0.8

1
1.0

I
1.2
                        24-HOUR AVERAGE SO2 FLOOR (LB SO2/10« BTU)


a) Assumes 1.2 Ib S02/106 BTU ceiling with 85% removal, 24-hour average with three day per
   month exemptions.
b) Transportation Rates: Rail < 250 miles,2.25C/ton-mile; > 250 miles, 1.200/ton-mlle;
                      Water 0.5C/ton-mile
                                       62

-------
                              Figure 5
      Sensitivity of Levelized Fuel-Cycle Cost with Respect to
      24-Hour SO2 Floor and Powder River Coal Characteristics
                          (Columbus, Ohio)
   400-n
0
O
o
a
o
o

Ul
u.
O
ui
N
Ul
Ul
   350-
300-
   250-
                     "     Powder River coal (PR)
                     	Northern Appalachian coal (NA)
                          2.6% sulfur, 10% ash, 12,000 Btu/lb
                        Powder River (PR)

% sulfur
Btu/lb
Ibs S/106 Blu
a
0.4
9,000
0.44
b
0.5
9,000
0.56
c
0.5
6,500
0.59
d
0.5
8,000
0.63
e
0.5
7,500
0.67
(
0.6
7,500
o.ao
       T-
                     PR-f


                     PR-e


                     NA

                     PR-d



                     PR-c




                     PR-b




                     PR-a
                        I
                       0.4
                             r
                            0.6
 1
0.8
         0       0.2      0.4       0.6      0.8      1.0      1.2

                 24-HOUR AVERAGE SO2 FLOOR (LB SO2/10* BTU)

e)  Assumes 1.2 Ib SO2/106 BTU celling with 85% removal, 24-hour average with three day per
   month exemptions.

b)  Powder River $1978/ton = 6.75; Northern Appalachia $1978/ton = 23.00
                                63

-------
contents between 0.44 and 0.80 Ib S/IO  Btu and are representative of the coals
available in the Powder River Basin.

     The estimated cost  of FGD systems reflects perhaps the greatest  uncer-
tainty and  as illustrated in Figure 6  is of utmost importance in the selection of
the lowest levelized fuel-cycle cost strategy to meet various revised New  Source
Performance Standards. If the higher PEDCo FGD cost estimates are used, coal
selection is affected by the emission limit.  On the other hand, if the TVA costs
are used, the local coal will be selected for all emission limits.
National Utility Simulation Model Results

     The Utility  Simulation  Model  was used  to  evaluate  the economic and
environmental impacts of alternative revised New Source Performance Standards
for coal-fired  electric utility boilers.   Numerous full  and partial scrubbing
scenarios were evaluated and  compared to the current NSPS. PEDCo and TVA
FGD cost estimates were  used to  represent the likely range of uncertainty  in
total fuel-cycle costs.

     The impact  of various full and partial  scrubbing scenarios using  PEDCo
FGD costs on  national SO-, emissions through  the  year  2000 is illustrated  in
Figure 7.  National emissions  in the partial scrubbing senarios begin to  increase
again between 2000 and 2010 while the full scrubbing SO-,  emissions do not begin
to increase until sometime after 2010.  National emissions by various classes  of
coal-fired plants in 1995 are presented in  Figures 8 and 9 for high  (PEDCo) and
low (TVA) fuel-cycle costs respectively.

     The national  percent increase in total utility  cost and percent decrease in
SCU emissions  in 1995 for  alternative revised NSPS are presented in Figure  10
for the high fuel-cycle cost case.

     Utility coal production and Western coal  shipments in 1995  for the high-
fuel cycle cost  (PEDCo) and the low fuel-cycle cost (TVA) cases as  a function of
                                     64

-------
                               Figure 6
                 Sensitivity of Levelized Fuel-Cycle Cost
                       with Respect to FGD Cost
                           (Columbus, Ohio)
    400-
00
h»
o>
CD

&
V)
O
o
u
 l
Ul

u.

o
ui
N

IU

Ul
    350-
300-
    250-
                          Powder River Coal
                          Northern Appalachian Coal
                                                      FGD Cost

                                                    	PEDCo

                                                    	TVA
               FED
                  (90)

              TVA
                                                                — (77)
                  I          I         I        I
                 0.2        0.4       0.6       0.8

                      ANNUAL SO2 CEILING (LB S02/10« BTU)


        ( ) = Percentage SO2 Removal
                                                   1.0
1.2
                                  65

-------
    25-
z
o

&

-------
                             Figure 8
      National S(>2 Emissions from Coal Fired Power Plants
                         (1C6 tons) 1995

                        PedcoFGD Costs
20-i
                                                     Current NSPS

                                                     0.6 Uniform Ceiling

                                                     0.6 Floor, 1.2 Ceiling

                                                     0.2 Floor, 1.2 Ceiling
        SIP Regulated
           Plants
 Current NSPS
Regulated Plants
 Revised NSPS
Regulated Plants
                               67

-------
                                 Figure 9
        National SO2 Emissions from Coal-Fired Power Plants
                            (10* tons) 1995

                             TV A FGD Costs
V)

O
    20-i
    15-
         11.8  11.9 11 a 11.7
* *
• •
• •

• • •
• • •
• • •

                                                          Current NSPS

                                                          0.6 Uniform Ceiling,
                                                             33% Removal
                                                          0.5 Uniform Celling,
                                                             90% Removal

                                                          0.2 Floor, 1.2 Ceiling
                                                    7.1
           SIP Regulated
               Plants
 Current NSPS
Regulated Plants
 Revised NSPS
Regulated Plants
                                   68

-------
                               Figure 10

           National Percentage Increase in Total Utility Cost
     and Percentage Decrease in SO2 Emissions for Revised NSPS
                                 1995

                            PedcoFGD Costs
                                SO2 Reduction


                                Increase in Total Utility Costs
                                                      19.7
to
Q.
CO
UJ
oc
DC
3
O
O
CC
u.
HI
O

<

o
111
(3


z
UJ
u
oc
Ul
a.
          0.6 Uniform Ceiling    0.6 Floor, 1.2 Ceiling   0.2 Floor, 1.2 Ceiling
                                   69

-------
the revised New Source Performance Standard are illustrated in Figures 11-14.
These  figures  show  the  great  sensitivity  of  regional  coal  production  and
shipments to relatively small changes  in total fuel-cycle costs and  the  lower
sensitivity to the standard itself.  It is  also clear from these figures that  a full
scrubbing option  (e.g., the  proposed 1.2 ceiling, 0.2 floor, 85 percent removal,
24-hour average) will promote the greatest use of local coal and minimize the
shipment of western coal to  the East.

     Regional FGD capacity in 1995 is shown in Figure 15 for the high fuel cycle
cost case. In the low fuel  cycle cost case FGD  capacity will  be higher under the
current NSPS and  under  partial  scrubbing  options, particularly in the central
"swing" states which are more sensitive to changes in relative fuel-cycle costs.

     Estimated FGD sludge, coal ash, FGD capacity, and utility water consump-
tion in 1995 are shown in Figures 16  and 17 for the high fuel  cycle cost case.  It
is  clear from these figures  that sludge production and FGD  water consumption
are relatively insensitive  to the form of the revised NSPS.   In absolute terms
FGD  sludge  quantities will be of  the same  order of magnitude as  fly  ash
quantities and FGD water consumption is projected to be an  order of magnitude
less than consumptive cooling water  requirements.  Of course, the environmental
impacts of sludge  and water consumption will depend on  specific power  plant
locations.

     Utility fossil  fuel consumption in  1995 for the high fuel-cycle cost case is
shown  in Figure 18. Note that projected oil consumption is  independent of the
revised NSPS and that fuel  used in coal transportation is an  order of magnitude
less than boiler oil consumption.  Oil  consumption is independent  of the SO2
standard because of our plant retirement and dispatching strategy. Considerable
oil  plant retirements are projected  to  occur  in the decade between 1985  and
1995.   However,  in  our  model,  oil plants are  retired on  the  basis  of age,
announced utility plans, government coal conversion programs, and not strictly
economics, We feel that this is appropriate for  a number of reasons including:

     •  High  fuel oil  costs are  usually passed through  to the
         customer.
                                     70

-------
 522
                                               Figure 11
                                  Utility Coal Production (10s Tons)
                                                1995
                                          Pedco FGD Costs
Northern Great Plains
                   West & Gulf Coast
Current NSPS

0.6 Uniform Celling

0.6 Floor, 1.2 Celling

0.2 Floor, 1.2 Celling

-------
                                           Figure 12
                              Utility Coal Production (10s Tons)
                                            1995
                                       TVA FGD Costs
       Northern Great Plains
Current NSPS

0.6 Uniform Celling,
   33% Removal
0.5 Uniform Celling,
   90% Removal

0.2 Floor, 1.2 Celling

-------
                                                         Figure 13
                                                Western Coal Shipped East
                                                           1995
                                                      Pedco FGD Costs
u>
                                                       0.2 Lb. Floor
                                                       i     I      V
                                       Western Coal     0.6 Lb. Uniform Ceiling 136
                                     Shipped East of the
                                         Mississippi      |      \
                                           River
0.6 Lb. Floor
                                     (In Millions of Tons) r—
                                                       Current NSPS

-------
       Figure 14
Western Coal Shipped East
          1995

      TVA FGD Costs
\
Western Coal
Shipped East of the
Mississippi
River
(In Millions of Tons)
l JL^k^m. r^
0.2 Floor, 1.2 Ceiling
33
Si / W\
| \ JN^
0.5 Uniform Ceiling, 90% Removal
\ f \ '
\ \ 1 I
1 N V \
0.6 Uniform Celling, 33% Removal
_l 1 J _— r V-
1 T A^^
n riCr
Current NSPS
33
r
U^.r
66
~«~<— ' i^^
^<
72

-------
                                                          Figure  15
                                                 Regional FGD Capacity (GW)
                                                             1995
                                                      PedcoFGD Costs
Ul
                                                                         East North Central
                           Mountain & Pacific
                                                                         East South Central
West South Central
              Current NSPS

            '•I 0.6 Uniform Ceiling

              0.6 Floor, 1.2 Ceiling

              0.2 Floor. 1.2 Ceiling

-------
     100-,
(A
z
o
                                    Figure 16

                  National Sludge, Coal Ash, and FGD Capacity

                                     1995

                                PedcoFGD Costs


                  Current NSPS



                  0.6 Uniform Ceiling



                  0.6 Floor, 1.2 Ceiling


                  0.2 Floor, 1.2 Ceiling
                                     100.1


273
WPM4





r
•*
*•
i
»
»
•
&
*
r»
E
r
»
»
;
t^^^*

2£









)4








324



















-350
.300

-250
O
-200 £
5
Q.
<
O
A
-150 «

-100
-50
                 Sludge
Coal Ash
FGD Scrubber

  Capacity
                                     76

-------
                Figure 17
        Utility Water Consumption
                   1995

            PedcoFGD Costs
6.1
6-
5-
4-
UJ
UJ
I 3"
o
2-
1-
0-

^












i
1
<
(
(
i
i
5.4% Per Year
Growth Rate
(1976-1995)
(5431 TWh)
              4.9
                            5.0
                                          5.0
                                          * • • *
                                          • • • e
                                         • • • •
                                         • * • •
                                                       5.1
               Current
               NSPS
0.6 Uniform
  Celling
 0.6 Floor,
1.2 Celling
0.6 Floor.
1.2 Ceiling
                      4.3% Per Year Growth Rate (1976-1995)

                                 (4470 TWh)
                      77

-------
00
              30-i
                 Figure 18
    Utility Fossil Fuel Consumption
                   1995
             PedcoFGD Costs
                                                                               Current NSPS


                                                                               0.6 Uniform Celling


                                                                               0.6 Floor, 1.2 Celling

                                                                               0.2 Floor, 1.2 Celling
                                                                                                           -1
                                                                                                           -.75
                                                                                                           -.50
                                                                                                           -.25
                                                                                                                  fe
                         Total Fossil
                      Fuel Consumption
Coal Consumption
                                                                   Oil Consumption
   Fuel Used in
Transporting Coal

-------
     •    Oil  plants  are often located  in  urban areas where coal
          storage space is not available.
     •    It is much easier for a utility to operate an existing oil
          plant than to site, build, and operate a new coal plant.
     •    Oil  plants  are  often  located  in strategic locations in the
          distribution grid and in 1995 will be used in a cycling mode.
     •    Residual  oil  will be available as long  as petroleum is
          refined for gasoline for  use in motor vehicles, etc.
     •    Lower  reserve margins  in  1995   (about  20 percent) will
          discourage differential  retirements  of the remaining oil
          capacity  simply in response to more stringent RNSPS.

     Oil plants in the 1990's will be dispatched after coal plants because of their
high operating cost.  Since the demand  curves are assumed constant  for each
alternative NSPS, their use, and hence oil consumption, does not change with the
alternative New Source Performance Standards. If,  however,  sen !v>er reliability
is lower than assumed,  oil  plants could be utilized more -  although  it is also
possible that utilities might build more nuclear plants if this were the case.

     The costs  associated  with alternative  New  Source Performance Standards
are illustrated in Figures 19-21.   The  pollution control investment in Figures  19
and 21 are for all  pollution control investments  including those for particulate
control, water pollution control, and S02 control.  Figures 19 and 20 show that
the uncertainties in projected fuel-cycle cost  (i.e., PEDCo vs TVA FGD costs)
can lead to substantial differences in estimated  costs.  Total pollution control
investment between  1983 and 2000 is compared to total utility investment  in
Figure 21  and illustrates that pollution  control investment  will represent less
than ten percent  of the total utility investment during that period.
Cost Effectiveness of a Standard

      The cost  effectiveness of  alternative revised New Source  Performance
Standards can be measured in numerous ways, some of which are not especially
                                     79

-------
                            Figure 19
Comparison of Cumulative Pollution Control Investment 1983-2000
               Reflecting  Pedco and TVA FGD Costs
                          (Billions 1975$)
    100-
 in
 K
 o>
O
CO

O
                                          Current NSPS
0.6 Uniform Ceiling,
   33% removal
0.5 Uniform Ceiling,
   90% removal

0.2 Floor, 1.2 Ceiling
                  Pedco
  TVA
                              80

-------
                        Figure 20
National Average Residential Monthly Electric Bill in 1995
       and Percentage Increase from Current NSPS
                         (1975$)
                          Current NSPS

                          0.6 Uniform Ceiling,
                             33% removal
                          0.5 Uniform Ceiling,
                             90% removal

                          0.2 Floor, 1.2 Ceiling
  60-t
               56.21
                      57.37
57.68
(9.5%)
               Pedco
                                          TVA
                             81

-------
                                    Figure 21
            Comparison of National Pollution Control Investment
                 and Total Cumulative Investment 1983-2000
                                         197S «)
    700 -,
                                   Total Inv®8tm@irst

                                   Pollution Control
                                628.4
                612.3
    500-
    400-
o>
IL.
o
(A
;J   300-
CD
    200-
    100-
                33.9
                                 47.5
                                                 53.8
                                                                 633.7
                                                      51.8
Current NSPS   0.6 Uniform Ceiling, O.S Uniform Ceiling,
                 33% Removal       90% Removal
                                                                0.2 Floor,
                                                               1.2 Ceiling
                                    82

-------
useful  in distinguishing between alternative standards.   The different interpre-
tations of cost effectiveness measures are illustrated in Figures 22 and 23.  If
FGD costs in cents per million Btu are examined, as in Figure 22, it is clear that
it  is less expensive to remove SO^ from  low sulfur coals to achieve a  given
emission limit.  If this cost effectiveness measure is used (which relates directly
to the cost per kilowatt-hour for electricity), then a standard favoring the use of
low sulfur coals should be established. On the  other hand, if cost effectiveness is
measured in terms of dollars per ton of 502 removed, as in Figure 23, it  is clear
that  the use  of high  sulfur  coal  provides  the greatest cost  effectiveness.
Therefore, the most cost effective standard by this measure  would be one that
promotes the use of high sulfur coal.  Therein lies the dilemma presented to
decision makers who must ultimately select the revised New Source Performance
Standard.   Also therein lies the  challenge to  the  engineers and technologists
gathered at this conference who  are working to improve each  of  these cost-
effectiveness measures by reducing the future  costs of S02 removal.
                                      83

-------
                                 Figure 22
          Comparison of FGD Cost Effectiveness per Btu of Fuel Input
                Under Annual Average SO2 Control Alternatives
     160-1
eo
r-
a>
CD
CO
O
O
IU
u
IU
u.
O
IU
N
IU

a
     140-
120-
     100-
 80-
 60-
      40-
      20-
                i Bituminous Coal
                • Subbituminous Coal
                  0.2
                       0.4
0.6
0.8
 I
1.0
1.2
                    24-HOUR AVERAGE SO, FLOOR (LB SO,/10* BTU)
                                     84

-------
                                  Figure 23

                Comparison of FGD Cost Effectiveness per Ton

       of SO2 Removed under 24-Hour Average SO2 Control Alternatives

                         with a 1.2 lb/106 Btu Ceiling
5
s
o
(A
   2000-
   1800-
   1600-
   1400-
1  1200-
O


o  1000<
(9
u.

O
111
N
Si
    800-
    600-
    400-
                 Fixed Bypass

                 Variable Bypass


                 Bituminous Coal

                 Subbituminous Coal
                                                  1.33 Ib S/106 Btu
                                         *	"2.17 Ib S/106 Btu
                                                   3.87 Ib S/106 Btu
                 0.2        0.4        0.6       0.8        1.0



                       24-HOUR AVERAGE SO2 FLOOR (LB SO,/10' BTU)
                                   85

-------
                              REFERENCES
I.   Teknekron, Inc., Energy and Environmental Systems Division, "Review of
          New Source Performance  Standards for Coal-Fired Utility Boilers,"
          Phase  Three Report, "Sensitivity Studies for the  Selection of a
          Revised  Standard," R-OI3-EPA-79, Report prepared  for  the  U.S.
          Environmental Protection Agency, Office of Energy, Minerals, and
          Industry (Berkeley, California, February 1979).

2.   "Additional Information on EPA's Proposed Revision to New Source  Per-
          formance  Standard  for  Power  Plants,"   Federal   Register 43
          (8 December  1978): 57834-59.

3.   PEDCo Environmental, Inc., "Summary Report — Utility Flue Gas Desul-
          furization Systems, Oct.-Nov.  1977," Report prepared for  the  U.S.
          Environmental  Protection  Agency, Division of  Stationary  Source
          Enforcement  and Industrial  Environmental  Research Laboratory
          (Cincinnati, Ohio, 25 January 1978).

4.   PEDCo Environmental, Inc.,  "Particulate  and Sulfur  Dioxide  Emission
          Control  Costs  for Large Coal-Fired  Boilers," EPA-450/3-78-007,
          Prepared for  U.S. Environmental  Protection Agency, Office of Air
          Quality  Planning  and Standards, Research Triangle  Park,  N.C.
          (Cincinnati, Ohio, February 1978).  Includes detailed computer print-
          outs for all case studies.

5.   TVA-Bechtel Shawnee Limestone-Lime  Computer  Program:  ten printouts
          (lime 25 MW,  100 MW, 200 MW, 500 MW, 1000 MW; and  limestone
          25 MW, 100 MW, 200 MW,  500 MW, 1000 MW).  Provided by C. David
          Stephenson, National Fertilizer Development Center, Muscle Shoals,
          Alabama, December 1978.
                                   8.6

-------
             SESSION 2

  IMPACT OF RECENT LEGISLATION

   WALTER C. BARBER, CHAIRMAN
Panel:  Impact of Recent Legislation
Brief overviews of recent legislation, EPA's
approach to implementation, and potential
impacts followed by questions from the
audience.

Members:  James L. Agee
          Gary N. Dietrich

 No papers or discussions are included for
             this session.
                87

-------
        SESSION 3



  ECONOMICS AND OPTIONS



WALTER C. BARBER, CHAIRMAN
           88

-------
                                  PAPER 3A


                      STATUS OF DEVELOPMENT,  ENERGY AND

                 ECONOMIC ASPECTS OF ALTERNATIVE  TECHNOLOGIES
                        P.  S.  Farber,  C.  D.  Livengood,
                   K. E. Wilzbach,  W.  L.  Buck,  and  H. Huang
                         Argonne National Laboratory


Several energy  technologies  are under development  throughout  the world that
either totally negate  the  need for flue-gas desulfurization (FGD) or require
less than full flue-gas scrubbing.   These processes remove sulfur  either prior
to  coal  combustion  (coal  cleaning  or  conversion), during combustion (atmos-
pheric and pressurized  fluidized-bed combustion),  or "between" two combustion
stages (gasification/combined-cycle operation).

This  paper  reviews the  status  of  development  and/or  demonstration  of
these technologies with respect  to their possible  application to the genera-
tions  of  electricity.   In  addition,  the  overall  coal-to-electrical-energy
conversion efficiency and economics (capital costs  and  total annualized costs,
mills/kWh)  are  explored  and  compared  for the various  alternatives.   The
economic premises utilized conform, as much as possible, to those used by the
TVA  in  comparisons  of FGD technology.   The paper  shows,  among other things,
the importance in any  energy and economic analysis of  an energy system, be it
a postcombustion treatment  process  (FGD)  or  the total energy process  (FBC), of
taking  into  account the  cost  of  fuel  and  the  overall process  energy effi-
ciency.
                                      89

-------
                              1   INTRODUCTION

       Over  the  last  decade  there  has  been  increasing interest  in  devel-
oping  alternatives  for  the coal-to-electricity  process other  than  conven-
tional combustion.  The  impetus  for  these efforts has come from two  sources,
the OPEC oil embargo  of  the early seventies,  which led  to increased  emphasis
on  coal  use, and  environmental  regulations  such  as  the  current  New  Source
Performance Standards  for utility power plants.  As  a  consequence,  the process
development efforts  have emphasized increased overall efficiency  and burning
of  coal  in  an  environmentally  acceptable  manner.   Recently,  the  proposed
tightening of NSPS restrictions on sulfur-oxide emissions, coupled  with doubts
as  to  the  ability  of  conventional flue-gas  desulfurization to achieve  a high
degree of reliability together with an 85-90% removal efficiency,  has  brought
an  even  greater  awareness  in government  and  utility circles  of  the need to
develop  power  plant  cycles  of  greater efficiency  and  minimum environmental
impact.   To be considered  as  commercially acceptable,  however,  these power
plants must produce  electricity  at  costs  competitive  with,  or  less than,
conventional combustion with flue-gas desulfurization  (FGD).

       Three differing energy  technologies  that  will either  negate  the need
for flue-gas desulfurization or  require less than full flue-gas scrubbing are
under testing and development by industry and the government. These processes
may be broken down into three  categories:   (1) those that remove sulfur
prior  to combustion  (coal  conversion),  (2)  those  where the  sulfur  removal
takes  place  during  combustion   (atmospheric  and  pressurized fluidized-bed
combustion), and  (3) those where sulfur  removal takes  place between two
combustion stages  (gasification/combined-cycle power systems).

       As part of  a  program  sponsored  by  the Department  of  Energy,  Argonne
National  Laboratory has been evaluating the  environmental and economic aspects
of  these alternative  technologies with respect to  their  application  to elec-
tricity  generation.   This  paper  reviews  the  status  and background of these
technologies and  analyzes their relative economic attractiveness.   In  order to
allow a  direct comparison with conventional coal  to electricity systems (with
FGD),  the economic  premises used  conform,  as  much as  possible, to those used
by  the TVA.  A  comparison by  TVA  of  various  processes  used in  flue-gas
desulfurization  follows this paper in the  symposium.
                                     90

-------
                      2  TECHNOLOGY  CHARACTERISTICS
2.1  SOLVENT-REFINED COAL

       The  solvent-refined-coal  (SRC)  process has  received  considerable
attention during  the past several years,  largely  because  end  products  from
this process  can  possibly be used in  fuel-combustion sources to comply  with
environmental  standards.   At  present,  two  pilot  plants  are  in  operation:
A  6-ton/day unit  (since  1973)  by  Southern Company Services at  Wilsonville,
Alabama,  and   a  50-ton/day unit  (since  1974)  by  the  Pittsburgh and Midway
Coal Mining Company of Gulf Oil at  Fort Lewis,  Washington.   The former has
been sponsored by EPRI and DOE, with  the  primary  objective  being  collection
of  operating  information  on  the main  reactor (dissolver) and several solid-
liquid  separation devices, using  a number of different  coals.   On the other
hand,  the  operation  of  the larger  plant,  sponsored  principally  by DOE,
has  emphasized collecting technical data on a Kentucky coal  to validate
scale-up  to commercial production, and  to provide  large samples of SRC for
combustion  and market-development studies.


2.1.1  Process Description

       Basically,  two  operating  modes of  the SRC  process  have  been identi-
fied. 1   The original mode (known  as  SRC-I) is intended, with minimum hydro-
gen  consumption,  to produce  low-sulfur, low-ash solid products  that are
suitable for  use  as  boiler fuel without flue-gas-treatment  provisions.
The  slurry-recycle mode, considered as an advanced version  and known as
SRC-II,  consumes more  hydrogen  and produces a low-sulfur, ash-free fuel  oil,
along  with significant amounts of  light oil,  pipeline gas,  and  naphtha.
Although conclusive economic  figures  are not  available,  it  is  generally
believed  that  the solid SRC should be less costly (per Btu) than its liquid
counterpart.

       Figure  1 shows  a simplified  schematic  of the  SRC-I process.   The
coal is  first pulverized  and mixed with a coal-derived,  anthracene-oil-type
solvent  in  a slurry-mixing tank.  The  slurry  is  mixed  with  hydrogen  (produced
elsewhere in  the  process) and  is  then pumped  through a fired  preheater and
into  a  dissolver,  where   about  85-95% of the coal  (moisture  and ash-free)
is  dissolved.   The  process  operating  conditions vary with  the  type of  coal
that is  processed;   some  representative  values  are given  in Table 1.  Under
these  conditions,   several  reactions  — depolymerization  and  hydrode-
sulfurization  —  also  occur,  resulting  in  a  very complex  solid-liquid-gas
mixture.

       From the dissolver, the mixture flows to  a  separator, where the gases
are  separated from the  slurry of undissolved solids  and coal  solution.
The raw gas is directed to a  desulfurization unit, then to a hydrogen-recovery
unit.   The hydrogen  sulfide  is converted to  elemental sulfur, hydrocarbon
gases are given  off,  and  the recovered hydrogen is recycled.  The slurry of
undissolved solids  and  the coal solution flows  to a  solid-liquid  separator,
where the solids  are removed.   The solid  residue, composed  mainly of uncon-
verted  carbon and  ash,  is sent  along with supplemental  coal  to a  gasifier/con-
verter   to  produce make-up hydrogen.   The coal  solution flows  to a vacuum

                                      91

-------
RAW COAL
 RECYCLE
 SOLVENT
   COAL
 RECEIVING
   AND
PREPARATION
                RECYCLE HYDROGEN
UTILITIES
  AND
SERVICES
                                 PURE
                                 HYDROGEN
                                     GAS RECOVERY
                                         AND
                                     RECOMPRESSION
                                    DESULFURIZATION
                                                     HYDROCARBON GASES
                                           SULFUR
                                                     CHAR AND
  RECYCLE
  SOLVENT
                              PRODUCT
                           SOLIDIFICATION
                                                     MINERAL RESIDUE
                                                     SOLVENT REFINED
              COAL PRODUCT



              LIGHT LIQUIDS
                    Figure 1. SROI Process Schematic
                                   92

-------
                   Table 1.  SRC Operations at Ft. Lewis


                Process Parameter              Typical Value

                Coal Source:
                  State                           Kentucky
                  Mine                            Colonial
                  Seam                           #9 and #14
                  Sulfur (%)                         3.1


                Temperature (F)                      850

                Pressure (psi-g)                     1500
                Coal Conversion (% MAF Coal)          95

                SRC Output:
                  Yield (% MF Coal)                   65
                  Sulfur (%)                         0.8

                Gas Yield  (% MF Coal)                6.0

                Liquids (% MF Coal)                 13.5
                Solvent (% MF Coal)                  4.4
still for  recovery  of  light oils and recycle solvent.  The SRC is produced by
solidification.   The SRC has  a melting point  of  350 to 450  F  and  a heating
value of about 16,000 Btu/lb.


2.1.2  Product Characterization

       Large  quantities  of blended  Kentucky  coals  have  been  processed  at
Fort Lewis  in  both  the  original SRC and slurry-recycle modes.  This coal is a
blend of  #9 and  #14 seams  in Hopkins  City,  Kentucky,  as  obtained  from the
Colonial #1 mine  of P&MCM Co.   Available data regarding the physical/chemical
properties of  solid  SRC samples  as well  as the raw coal have been collected in
a  report^  prepared  for  ANL by  Air  Products and Chemicals  and  are  presented
in  Table  2. Also shown  in Table 2  for comparison are  typical  properties  of
liquid SRC oil obtained with a  similar coal  feedstock.

       It  is  obvious that  SRC  is a better  boiler  fuel than  the  raw coal in
terms of  ash,  sulfur,  and heat  content.   Also,  storage and handling tests by
Air  Products^  and  Babcock  and Wilcox^  did not  uncover  any insurmountable
problems.    Nevertheless,  it appears  that  the  use   of  solid SRC in utility
boilers without  additional  controls to meet  the proposed  NSPS standards (85%
sulfur removal and  0.03  Ib  TSP/10^ Btu)  is questionable,  though the current
standard  for  S02 can be  met.    (Utility-scale  combustion  tests are discussed
in  the following section.)   Further process improvement  or product upgrading
is  imperative  to maintain SRC  as  a  compliance boiler  fuel in  the future.
                                      93

-------
             Table 2.  Properties of Ft. Lewis SRC and Raw Coal

A.



B.






C.
D.




E.




Property
Proximate Analysis (wt % dry)
Volatile Matter
Fixed Carbon
Ash
Ultimate Analysis (wt % dry)
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen (by difference)
Heating Value (Btu/lb dry)
Sulfur Forms (wt % dry)
Pyritic
Sulfate
Organic
Total
Hardgrove Grindability Index (HGI)
Temperature (F)
10
70
150
SRC Solid

61.4-66.7
33.3-38.5
0.08-0.12

87.1-87.7
5.5-5.9
1.8-2.0
0.7-0.9
0.08-0.12
3.98-4.22
15770-15810

0.014-0.019
0.002-0.004
0.684-0.877
0.700-0.900

HGI
170-178
184-188
180-187
Coal

39.8
49.8
10.4

70.4
5.1
1.4
3.4
10.4
9.2
12,760

1.5
0.2
1.7
3.4
55-70




SRC Oil3





86.6
8.4
1.1
0.3
0.01
3.6
17,040










aAPI Gravity (60  F): 8.3, and viscosity (SSU at 140  F): 35.6

-------
These are  some  of the main reasons for the development of the  slurry-recycle
SRC process.   The liquid SRC oil is claimed  to  be  similar  to No. 4 fuel oil
in terms of  handling and burning and can be used to comply with  the proposed
NSPS, albeit at higher costs.


2.1.3  Combustion Tests

       About  3,000  tons  of  solid SRC  were burned  in a 22.5-MWe  boiler at
Georgia Power  Company's  Mitchell  Plant  in mid-19775, and about 4,500 barrels
of  liquid  SRC  oil  were consumed at  the  Consolidated  Edison's  74th-Street
station  in Manhattan in late  19786.    The  Con.  Ed. test was  reported  to be
a success,  and  all air  emissions were  below the  newly proposed  NSPS for
utility boilers  using coal-derived  fuels.  However, the  fuel characteristics
were not the same  as those  expected  for commercial  operation, having a higher
ratio  of  light  to  heavy oils.   Bench-scale  burn  tests, nevertheless,  have
demonstrated that the proposed NSPS for boilers using coal-derived fuels could
be met? with the use of  staged combustion  to suppress NOX.

       No  serious  problems  were encountered  in  the  combustion  test  of solid
SRC  at  the  Mitchell plant.5   Blowing losses experienced  in early rail-car
shipments  and  dusting during unloading  operations were successfully minimized
by simple  chemical  pretreatments.   The standard  pulverizers were modified to
the  extent  of  using' unheated  air,  reducing  ball-spring pressure,  and  in-
stalling variable-speed  feeder motors.   The  only  major  boiler modification re-
quired was the use of specially designed, water-cooled dual-register burners
to  accomodate the  low melting  point  of  solid  SRC and  to  reduce  NOX emis-
sion.

       Operation  and emissions  data  are  reported  in  Reference  5  for  solid
SRC  and raw coal.   No  CO,  C^-Cg  hydrocarbons,  or polynuclear  aromatic
compounds  were  detected  during  any  SRC  test, and  overall  boiler efficiency
at full load was essentially the same when burning either  SRC  or coal.  Of the
three  most  important air  pollutants, it  appears  that  only NOX emissions
were below both current  and  proposed  NSPS  (0.5  Ib  N02/106  Btu)  for utility
boilers  (by  about 40% and  10%,  respectively).   S02 emissions were  about 20%
below  the  current  NSPS but  barely  fall  short  of the 85% removal  mark.
The particulate  loadings  leaving the boiler were  surprisingly high due to the
high carbon  content  of the  fly ash and the unexpected higher-than-normal ash
content in SRC  (0.57  wt  percent) resulting  from  contamination by  surface dust
and other  foreign material during pit solidification and  storage.  Even with a
secondary  precipitator  of modern design,  the  plant  could not meet  the newly
proposed NSPS for particulates, probably  again due  to  the high  carbon concen-
tration of SRC  fly  ash.   Also note that  the 20% opacity  standard was not met
under full and medium loads.

       In  mid-1978,  DOE awarded  two  $6-million contracts to  consortiums
headed by  Southern  Company  Services and Gulf Oil  to  start  the preliminary
designs of two 6,000-ton-per-day plants to make solid SRC  and  liqtiid  SRC oils,
respectively.   Both design studies  are  scheduled  to  be  finished about mid-
1979.  At  that  time,  a  choice is scheduled  to  be made  as  to  which type of
plant will actually be built.
                                      95

-------
2.2  GASIFICATION/COMBINED-CYCLE POWER GENERATION

       The combination of  coal  gasification  with combined-cycle  power genera-
tion, utilizing both  a gas turbine and waste-heat  boiler,  offers  the promise
of  economically  competitive power  generation  together with  highly  effective
pollution control.   However,  these possible advantages  are  obtained  at  the
expense  of  system  simplicity,  and  many unanswered  questions  regarding  the
operability and economics of full-scale systems remain.  The results  presented
in this  study were  obtained  from  an assessment  of the technology prepared for
ANL by the United Technologies Research Center.8

2.2.1  Process Description

       The gasification/combined-cycle power plant  is an integrated  facility:
It  includes  all  the  equipment necessary  to convert  coal into electricity.
As  can  be seen in  Figure  2, the  conversion process  can be  divided  into  two
major systems, fuel processing and power production.

       The  fuel-processing system  includes  coal storage  and handling,  coal
processing, gasification,  fuel treatment,  and  by-product  streams.    Coal  is
delivered  from  the coal yard  to  a crusher, where  it is sized  to  dimensions
required by the gasifier, e.g., 1/8 x 1-1/2  in.  for  a Lurgi to pulverized coal
(70% through  200  mesh) for a Foster Wheeler gasifier.   From  the crusher,  the
coal can be dried, if necessary, and then injected into the gasifier.

       The resultant fuel  gas,  having  a heating  value from 90 to 150 Btu/scf,
is  now  sent  to the fuel-gas  treatment  portion  of the procesing plant.   Here
the gas  is cooled, and tars, if present,  are separated along with particulates
and  ammonia and  sent  to the  sulfur-removal  process,  where the  H2S,  COS,  and
other sulfur  compounds are removed.   The clean  fuel gas continues on to  the
power-production portion of the power plant  while the sulfur bearing  gases are
sent to a Glaus plant  for conversion to elemental sulfur.

       The  major  by-products   of  the  fuel-processing  system   are  elemental
sulfur,  ammonia,  slag and/or ash,  and,  for some gasifiers,  coal tars.

       The power-production  section consists  of gas  turbines,  heat-recovery
steam  generators,  steam  turbogenerators,  and heat-rejection  equipment.
The fuel gas from the fuel-processing section is burned in the gas  turbine.   A
portion of the compressor  discharge of the  gas  turbine is  sent to  the gasifi-
cation plant  via a boost  compressor to  supply  oxidant for the  process.   The
hot exhaust gases from the  gas  turbines  are  cooled  by raising steam,  which  is
sent to  a steam turbine,  expanded to  subambient conditions  producing power,
and  condensed.    The  heat  of  condensation  is   rejected  in  mechanical  draft
cooling  towers.   Steam may also be produced by  cooling  the  fuel gas  prior  to
sulfur removal.

       There are a number  of  auxiliary systems  required  in both  the fuel-pro-
cessing  and power-production  facilities.   In addition to  the normal  station-
keeping  requirements   for  heat,  light,  potable water,  compressed   air,  and
                                     96

-------
COAL
PILE
                                                      STEAM
                                                                              AMMONIA

                                                                              OIL/TAR
                                                      WASTE WATER
                                                                                                              FLUE GAS
                                                                                                              TO STACK
                    Figure  2.   Gasification/Cornbined-Cycle Power-Plar.t  Schematic

-------
instrumentation,  there are  the  requirements for boiler  feed-water  treatment,
cooling-water  treatment,  waste-water  treatment,   disposal  of  slag/ash,   and
transfer/storage  of by-products.

2.2.2  Gasifier Descriptions

       For  this  study,  four  types of  coal gasifiers were  selected.   These
include the  pressurized Lurgi dry-ash,  the IGT-Gas fluid-bed, and  the  Texaco
and two-stage Foster Wheeler entrained-flow gasifiers.   Key power-plant para-
meters are given for each  gasifier type  in Table 3.

       The  Lurgi  dry-ash  gasifier has  been  commercially  available from
Lurgi Gesellschaft  fur Warme  und  Chemie  Technik  MbH since 1936.   Presently,
there  are nearly  70  gasifiers  in commercial  operation  producing  town gas,
synthesis gas (syn gas)  or medium-Btu  fuel gas.

       In  this system,  coal is  crushed  to  between  1/8 and 1-1/2 in.   and
sent  to  one  of  two  lock hoppers.   Coal fines  are  usually  rejected, which
results in as much as a 25% loss in delivered  coal.  However-,  consideration is
being given  to briquetting fines with  recovered  tar.  Although claims are made
that all  coals, including highly  caking types,  can be  used, some  pretreatment
would probably be required for the caking  coals.   Also  the use of  a  stirrer in
the bed could be necessary.

       Coal  from the  lock hopper  is dropped into  the gasifier and distributed
evenly by  the distributor arm on a coal  bed.   Steam  and air are  introduced
into the  bottom  of the  gasifier  through a rotating grate.  The steam  and  air
pass upward  through the bed, creating  different  zones  in  the gasifier.   At  the
bottom,  carbon  is  burned, providing heat  to  the next zone,  the  gasification
zone.   The hot gases then  devolatilize the  coal  and finally provide heat  to
dry the incoming coal prior to leaving the gasifier.
                    Table 3.  Power System Characteristics
         Power System                 Lurgi    IGT       FW       Texaco
Clean Gas HHV (Btu/scf)
Gas Turbine Pwr (MW)
Steam Turbine Pwr (MW)
Net Pwr (MW)
Heat Rate (Btu/kWh)
109.4
450.9
122.1
532.8
10869
158.6
328.2
237.6
523.2
8264
177.9
320.7
206.4
488.1
8304
90.5
273. Oa
194.4
441.2
8706
        a Includes let-down turbines.
                                     98

-------
       Considerable amounts  of  tar and oils  are  produced in the upper  zones
and must be removed from  the  gas.   This  is  accomplished  in a scrubber cooler,
a waste-heat  boiler and  subsequent  coolers.   The  separated tar liquors and
oils are usually  fed to a Phenosolvan  unit  and ammonia plant to  recover  crude
phenols and anhydrous ammonia.

       The  Lurgi  gasifier  operates  at  a pressure that  is a  function of
gas-turbine pressure ratio and fuel-processing-system and fuel-control-
system  pressure  drop.   The commercially available  gasifier can process ap-
proximately 400 ton/day and has an expected turndown ratio of 4 to 1.

       The  U-Gas  process  is  being developed  by  the Institute  of  Gas  Tech-
nology, where  a  4-ft-diameter  atmospheric-pressure  unit  processing coke has
been in operation since  1974.   This  process utilizes a  fluidized-bed system
that can  produce  either  low- or medium-Btu  gas  with  either air- or oxygen-
blown operation.

       The  use  of a fluidized  bed  has many inherent advantages.  In partic-
ular,  the   bed  material  acts  as  a  catalyst for  the  gasification  reactions
and  should  permit  operation at relatively  low  temperature while  completely
gasifying the  feed.   The U-gas process is  distinguished  from other fluid-bed
processes in that  it utilizes  an  "ash  agglomeration" technique to concentrate
the ash and remove it with minimum carbon content  while operating the bed with
a relatively high carbon content.

       Raw  coal  is crushed  to 0 x  1/4  in.  size.   The  feed may contain up
to  10% <200-mesh material  as  generated in  the  crushing  step.    Noncaking,
subbituminous  coals and  lignite  can  be fed  directly  to  the  gasifier  from
the crusher.   Caking  coals (eastern bituminous,   for example) must  at present
be  pretreated  by  contact  with  air in a  fluidized  bed  operating at gasifier
pressure and  700  to 800  F.   An oxidized outer  layer  forms on  the  coal par-
ticles, preventing agglomeration and  possible blockage in the  gasifier.

       Heat  evolved during  pretreatment  is  removed by  generating  steam in
heat-transfer coils immersed  in the fluidized-bed pretreater.   Coal that has
been pretreated  is fed to  the  gasifier.   Off-gases  are  fed to  the  bottom of
the gasifier  to destroy  all  tar  and  oils that evolve during the pretreating
process.

       The  gasifier  is   a refractory-lined,  hot-metal-wall vessel.     Steam
is  generated  to provide  cooling  for  the pressure vessel while  the fluid bed
reaction takes  place  at  temperatures  as high as  2000 F.   System  pressure can
be  as  low  as  100 psia (minimum level  is  determined by economics), but in the
combined-cycle  application,  the pressure would  be  determined by gas-turbine
pressure ratio,  pressure  drop in the  fuel-processing system, and gas-turbine
fuel-control-system requirements.

       The  operating  conditions within the gasifier result  in  a product gas
free of tars and oils.   Thus, no special cleanup procedures  are  needed.
                                      99

-------
       Projected  commercial-sized U-Gas  gasifiers  would handle 3000  ton/day
of coal.   It is  estimated  that  the gasifier  would  have a  10-to-l  turndown
ratio.

       The Foster Wheeler  Energy  Corporation is  currently designing  an
air-blown version of a  two-stage, entrained-flow gasifier developed  by  Bitu-
minous  Coal  Research  (BCR).  The  oxygen-blown BCR gasifier is presently
undergoing testing in a 120-ton/day  pilot plant producing 2.5 million scf/day
of syn gas at Homer  City, Pennsylvania.

       Run-of-the-mine  coal   is  crushed,  dried  to  two-percent moisture,  and
pulverized.   Coal  is metered from  the  feed hopper  to  an injector  and  then
into  hot  transport  gas  (recycled  from  the gas-purification  section) before
being  fed  into the  upper  stage  of  the  gasifier.    In  this  stage,  the  coal
reacts with  synthesis gas from the  lower  stage and steam to  produce  methane,
carbon  monoxide,  hydrogen,  and   unreacted  char.    The  gases  leave the  upper
stage at around 1800 F.

       Entrained  residual char is removed from the gas  by cyclone separators
and  recycled via  superheated  steam to  the  lower  stage of  the gasifier.
The  char  then reacts with  steam and air at about 2800 F to  form synthesis
gas  and molten slag.   The  hot  synthesis  gas  flows to  the  upper stage  for
reaction with coal  as described  above.   Molten slag  collects and  drains  from
the  bottom of  the  lower  stage into  the  slag pot, where  it is water-quenched.

       Overall,  the upper-stage gasifier  reactions  are endothermic,  and
the  process-heat  requirement is  supplied by  combustion of  char with  air.
The  air  rate is  regulated to maintain the operating  temperature in the  upper
stage,  while the  lower-stage  temperature is   controlled by   steam  addition.
Temperature in the lower stage is fairly critical, because too high a tempera-
ture  will  damage the  refractory and too  low  a temperature will cause  the
slag  to freeze and accumulate.

       While the pilot  installation of the   FW gasifier will have only a
480  ton/day  capacity,  it would  appear  that larger  gasifiers of   a  capacity
similar  to  the  Texaco   gasifier  (1900-2000 ton/day)  would  be commercially
viable. The  operating pressure would be  dependent  upon  the gas-turbine  pres-
sure  ratio and pressure drop in the fuel-processing system  and  gas-turbine
fuel  controls.

       The Texaco gasifier has been  in commercial  use  for a  number of years,
producing synthesis gas (H2 +  CO) from a variety of liquid hydrocarbon
feeds.   Its  application  to  coal  gasification is   still in   the  development
stage,  although  as  early as the mid-19501s a 100  ton/day  pilot plant  was
operated at  300  psig on West Virginia  coal.   In  Texaco's Montibello,  Cali-
fornia,  research  facility,  a 15  ton/day pilot  plant  has provided  low-  and
medium-Btu gas  from coal to a  gas-turbine combustor.   Texaco and  Southern
California  Edison are  presently  negotiating for  a western-coal-fueled
gasification demonstration  plant  which would provide medium-Btu gas  to
a  utility  boiler.   This plant  could eventually be  converted  to  a combined-
cycle installation.
                                     100

-------
       In the Texaco  gasifier,  coal reacts with air and  steam  under  slagging
conditions  in  a refractory-lined  pressure  vessel.   The  hot gases leave  the
gasifier  and pass  through a  fire-tube-type waste-heat-recovery  steam  genera-
tor.   Slag  is  quenched at the bottom  of  the  gasifier  and removed via a  lock
hopper.

       Pulverized coal  (70%  < 200 mesh) is slurried in water and  pumped  into
the  gasifier.    The  gasifier  operates  at temperatures above  the ash fusion
point; thus, no  tars  or oils  are  produced.  Typically,  the operating  pressure
would  be  600 psig  and  above.  A  single nine-ft-diameter  gasifier  of  the  type
commercially available would gasify  approximately  1900 ton/day  of coal.

       Turndown  to  50%  of  capacity is  routinely  accomplished   in commercial
applications and the  gasifier  will operate  satisfactorily  at   15% capacity.


2.3  FLUIDIZED-BED COMBUSTION (FBC)
2.3.1  Technology Description

       The basic features of a coal-fired fluidized-bed boiler  are  depicted in
Fig. 3.9   Crushed coal (typically 1/4-in. top  size)  is  continuously fed  into
and  burned  in a bed  of crushed limestone or dolomite  (typically  1/8-in. top
size)  that  is fluidized by  a  continuous  upward flow of air through a perfo-
rated  plate at the bottom of the vessel.  Boiler tubes typically  are  submerged
in  the bed and may  also  be situated  in  the freeboard region above the  bed.
S02  released  from the  burning coal reacts  chemically with the limestone or
dolomite in  the  bed,  forming solid  CaS04  and obviating the  need for any
subsequent  flue-gas  desulfurization.   For most  effective  S02 sorption, the
bed  temperature  is maintained in  the  range  of 1500-1700°F.   This relatively
low  combustion temperature  has  the  additional  advantages of  reducing the
formation of  NO and  the volatilization of trace elements,  as well  as preclud-
ing  slag formation.  Sorbent reactivity is maintained at the required level by
continuously  feeding  fresh  stone  to the bed and withdrawing spent (sulfated)
stone  from  the  bed at equivalent rates.  A major fraction  of  the  coal ash is
entrained and carried out of the bed with the combustion gases.

       The  flow  of air  and  combustion bases upward  thru  the  bed gives  rise
to  a fairly  rapid circulation and mixing of the bed  solids.   This action, in
turn,  results  in  very effective heat  transfer  to the  submerged  boiler tubes,
nearly uniform  temperature  distribution  throughout  the bed, high volumetric
heat-release  rates,  and the  capability of burning  nearly  any type  of  solid
fuel.  The high rates of heat release and transfer also offer  the promise  that
a fluidized-bed boiler  can  be  smaller and more efficient  than a conventional
pulverized-coal boiler with the same power output.

       For  purposes  of generating electric  power,  the   energy-conversion
efficiency can, in principle,  be  still further increased by pressurizing the
f luidized-bed combustor (say to  
-------
    BASIC  FEATURES OF  A FLUIDIZED-BED  BOILER
                                  FLUE
FUEL/SORBENT
INJECTION PIPES
IN-BED
GENERATION
SUPERHEAT
OR REHEAT
SURFACE

AIR
DISTRIBUTION
GRID
 CONVECTION
(FREEBOARD)
 ECONOMIZER,
 GENERATION
 SUPERHEAT
 AND/OR REHEAT
 SURFACE
  PLENUM
                       Figure  3
                           102

-------
pressurized fluidized-bed  combustion  (PFBC),  as opposed to  atmospheric-pres-
sure fluidized-bed  combustion  (AFBC),  offers the following  additional  advan-
tages:    (1)  more compact  combustor vessels,  (2) more effective 862  sorption
(provided that dolomite or  precalcined  limestone is  employed),  (3)  inherently
lower NO emission, and (4) improved combustion efficiency.  However,  transport
of  solids  into and out of  the  pressurized  combustor and,  to an even greater
degree,  protection  of the  gas  turbine from  excessive  corrosion and erosion
when driven by  the relatively dirty  combustion gases present  formidable
technical problems.

       Figure 4  shows  simplified  schematics of an AFBC  power plant  and  three
different types of PFBC/combined-cycle  power plants.10   In this figure
"additive" means sorbent  (limestone  or  dolomite)   and  "ash"  includes   spent
sorbent.   In  the case of  the  AFBC system, note the presence of the carbon-
burnup  cell  (CBC) whose purpose  is  to recover the heating value of  unburned
coal particles and soot elutriated from the primary  combustor and collected  in
the primary cyclone.   It  is  believed  that  PFBC systems will  not  require  a CBC
to  achieve acceptable combustion efficiency.

       The water-cooled PFBC system shown in  Figure  4b is  essentially similar
to  the AFBC system  in  Figure 4a,  except that  the pressurized combustion  gases
are used to  drive a  gas  turbine  that  might  generate ^ 20% of the  electrical
output  of the plant.   Both systems would be  run  with about   15-20%  excess
air.

       The adiabatic  PFBC  shown  in Figure 4d  has  no   heat-transfer surface
in  the  combustor.   Instead,  the combustor  is  supplied with'X-  300% excess air,
which  serves  to  control the combustion temperature  and  carries off the heat
released  by  the  burning  coal.    In this case ^ 80% of  the  plant  electrical
output  would  be  generated by  the  gas turbine  and '^ 20%  by  a  steam turbine
connected to  the waste-heat boiler.   This  configuration has the disadvantage
that  the particulate-removal   equipment  between the  combustor and  the gas
turbine  would  have  to handle very large volumes of  hot, high-pressure  gases.
Also,  there  is concern that the  large amount of excess air fed to the com-
bustor  would  increase the emission of  NO and particulates in the  combustion
gases.

       The air-cooled PFBC system shown in Figure 4c  is similar  to the adiaba-
tic  system  in that  most  of the  electrical  output  is  derived  from the gas
turbine.  In  this case,  however,  the pressurized air  is split  into two
streams.   Only about 25-30% of the  air is  fed to the combustion chamber,  so
the actual amount of  excess air  and  the  quantity of gas  passing through the
particulate  removal  equipment   would  be no greater  than  in the  case of the
water-cooled PFBC system.   The  remaining  70-75%  of  the pressurized  air  passes
through  heat-transfer  tubes located in and above  the fluidized bed, and the
resulting hot,  pressurized  air is conjoined  with  the  combustion-gas  stream
between  the  particulate-removal  equipment  and  the  gas  turbine.   Thus, the
disadvantages  cited  above  for  the adiabatic  PFBC   scheme are  rather  neatly
avoided.
                                      103

-------
                    FLUIDIZED-BED  COMBUSTION POWER  GENERATION  SYSTEMS
PRIMARY
COMBUSTOR
                                  FIN.AL DUST
                                  COLLECTOR
                              AIR
                                            CONDENSER
                                                WATER
                                    BOILER
                                    FEED WATER
                                                                 COMPRESSOR
                                                                        PARTICULATE
                                                                         REMOVAL
                                                                               *
                                                                                                 STACK
                 ASH
                             DISPOSAL
                                                                            ASH DISPOSAL
     0.   ATMOSPHERIC WATER-COOLED  COMBUSTOR
                              b.   PRESSURIZED WATER-COOLED  COMBUSTOR
                STEAM TURBINE
                                  STACK
                                                                        STEAM TURBINE
             CONDENSER

                BOILER FEED WATER
          AIR
GAS TURBINE


  PARTICULATE
   REMOVAL
                                                                                           STACK
                                      CONDENSER

                                          BOILER FEED WATER
                                             WASTE HEAT'
                                               BOILER
                                                                        COMPRESSOR
                                                                        PRESSUR-
                                                                         IZED
                                                                       COMBUSTOR
                                                                                        GAS  TURBINE


                                                                                          PARTICULATE
                                                                                            REMOVAL
                                                                        ADDITIVE
                                                                        COAL
                        ASH DISPOSAL
                                                                                  ASH DISPOSAL
     C.  PRESSURIZED AIR-COOLED COMBUSTOR
                              d.  PRESSURIZED ADIABATIC COMBUSTOR
                     Figure  4
                         104

-------
2.3.2  Developmental Status and Prospects

       In Table  4  are listed  some  of  the more  important  FBC  units that are
presently in  service or  scheduled  for  early completion.   For both AFBC and
PFBC the  list includes:   (1) a  pilot  plant,  (2) a  components  test and in-
tegration unit (CTIU) or  equivalent test  facility, and  (3)  a relatively  small
but presently operational process development unit (PDU).

       The AFBC  pilot plant  at   Rivesville, West Virginia, is  located  in a
conventional, coal-fired power station of the Monongahela Power Co.  The con-
figuration consists of  three main beds,  each  10 x  12 ft in area,  plus a
slightly smaller  bed that serves as a  carbon-burnup cell.   The unit can
generate 300,000 Ib of   steam per hour at 1250  psig and 925 F.  The steam can
be  fed  to one of  the existing  turbogenerators at  the station,  producing
electricity  for  the  Allegheny Power  System's   commercial  grid.    The  first
commercial power generation with  this  unit  took place  in September  1977, and
during May 1978  the unit was  operated continuously for a period of  50 hours,
producing 800 MWh of electricity.  On  August 9, 1978,  a fire  in the air
preheater resulted  in damages to the  pilot  plant estimated at $1.5 million.
Repairs were expected to take  about six months.   Prior  to the fire,  a total of
more than 70 hours of commercial  operation had been logged.

       The PFBC pilot  plant being designed  and  built  by  Curtiss-Wright
will be  of  the  air-cooled, combined-cycle type  shown schematically  in Figure
2c  and  is expected to be  operational by  1980.   The  combustor  will  feature a
fluidized bed  12 ft  in  diameter  with  a height  of 16  ft plus  15  ft of  free-
board.  Only  1/3 of the  air from the  compressor will  be fed directly into the
combustor; the remaining  2/3  will be heated  by  passage  through tubes immersed
in  the  bed.   The  gas turbine will  generate 7  MW of  electrical  power.   The
exhaust gases will then pass through a waste-heat boiler, wherein steam will be
produced at a rate  equivalent  to  an additional  6 MW of  electrical  power.  The
gas-turbine  blades will  be   of  a special  design that  provides  a  "boundary
layer"  flow  of  cool, high-pressure air to  protect  the blades from  corrosion
and erosion.

       There  are,  of course,  certain unresolved  issues and problems in con-
nection with  FBC power generation.  Some of the more critical are  listed in
Table 5.   Nevertheless,  the   present  consensus  seems  to  be  that  FBC  power
generation — at  least  the  atmospheric-pressure  type  —   can be  brought  to
commercial status  within a reasonably short period of  time and  that it will
prove  to be  competitive  with,  or superior  to, conventional  PC/FGD power-
generation  technology with   regard  to cost,  efficiency,   and environmental
acceptability.

       Thus,   a  report recently  drafted by  a DOE task  force   concludes that
AFBC power-generation technology  is ready  for commercialization and recommends
that  a  200-MWe   commercial demonstration  plant  be  constructed and put into
operation by  1985.   Conceptual design studies  for  such a  plant have  already
been  completed,  but  detailed design  and  construction would  require  about
five years.   Assuming that the demonstration plant is  completed on schedule,
the report projects that 60-120 GW of  commercial AFBC  utility-plant generating
capacity could be on line by the  year  2000.H
                                      105

-------
                               Table 4.  Roster of Selected FBC Facilities
          Installation
  Type
Rated Power
Status (1/79)
Atmospheric-Pressure FBC Facilities
1. Pope, Evans & Robbins, Inc.
Rive svi lie, W. Va. (DOE)
2. Morgantown Energy Technology Center
Morgantown,W. Va. (DOE)
3. Babcock & Wilcox Co.
Pilot plant
CTIU
PDU
88 MW(t)
30 MW(e)
18 MW(t)
6 MW(t)
Temporarily out of service
for repairs.
Under construction;
completion in 1980-81.
Operational .
         Alliance, Ohio (EPRI)
Pressurized FBC Facilities
     4.  Curtiss-Wright Corp.
         Wood-Ridge, N.J. (DOE)
     5.  International Energy Agency
         Grimethorpe, England (IEA)
     6.  Exxon Research & Engineering Co,
         Linden, N.J. (EPA)
Pilot Plant

Flexible test
facility
PDU
("Miniplant")
38 MW(t)         Under construction
13 MW(e)         completion in 1980.
80 MW(t)         Under construction;
                 startup later in 1979.
 1.8 MW(t)       Operational.
ABBREVIATIONS:  CTIU = Components Test and Integration Unit
                PDU = Process Development Unit

-------
        Table 5.  Key Development Issues in Fluidized-Bed Combustion
           Issue
Relative Importance
  AFBCPFBC
1.  Sorbent Requirements and
    Utilization

2.  Disposal or Utilization of
    Ash and Spent Sorbent

3.  Combustion-Gas Cleanup at
    High Temperature and
    Pressure

4.  Particulate Removal from
    Stack Gas

5.  Gas-Turbine-Blade Erosion,
    Corrosion and Deposition

6.  Solids Feeding and Trans-
    port Systems

7.  Corrosion, Erosion, and
    Deposition in the
    Combustor

8.  Improved Combustion
    Efficiency

9.  Startup, Turndown, and
    Load-following Capability
                                                       ****
  ****
  N.A.
  ***
  N.A.
  ***
  **
  ***
  **
                                                                    ***
               ***
               ****
               ****
               ****
               **
               ***
NOTE:  The number of asterisks indicates relative importance in each case.
       N.A. means "not applicable."
                                      107

-------
       The TVA also has  an  active  program looking toward the design and con-
struction of AFBC power plants  at the  commercial  scale.  According to a recent
report,  their  timetable  calls  for construction of  a  200-MWe demonstration
plant by 1984 and of a full-scale commercial  plant by  1990.I2

       Although  the  prevailing view in  the U.S.  is that  PFBC/combined-
cycle  power-generation technology  is not  yet  ready  for  commercialization,
at least one U.S. utility consortium seems  to feel otherwise.  After obtaining
encouraging  results  during  a feasibility  study  conducted jointly with Stal-
Laval Turbine of Stockholm and Babcock & Wilcox, Ltd. of  Birmingham, England,
American Electric Power Co.  has  tentatively decided  to proceed with design and
construction  of  a  commercial-scale  demonstration  plant,  apparently  of  the
water-cooled PFBC/combined-cycle type.   The  coal-fired  plant,  which AEP says
might be in full operation as early as 1983,  will be built at a decommissioned
power  plant  in Brilliant, Ohio,  and will utilize a 105-MWe steam turbogener-
ator  already located at that  station.   Babcock & Wilcox will supply the
pressurized  combustor,  while Stal-Laval  will provide a  suitable  gas turbo-
generator capable  of generating  65 MWe.   Details  of  the  provisions  for
hot-gas cleanup and turbine-blade protection  are  not yet available.13


2.3.3  Environmental Implications of FBC

       Table  6   presents,  in outline  form,   a resume   of  the  prospects  for
control of stack-gas emissions  from coal-fired FBC power plants.  Of the three
pollutants covered by  the federal New Source  Performance  Standards (NSPS), it
would  seem  that  particulates  may  pose  the  most   serious  control  problems.
First  of all, particulate,  loadings in the combustion gases from  an  FBC are
inherently rather high as a  result  of the  absence of  any  slagging of the coal
ash  and  of the  unavoidable  elutriation  from the fluidized-bed of small par-
ticles  of  sorbent.    Particulate  collection  by electrostatic precipitation
(even  "hot" ESP, as used at  the Rivesville AFBC pilot  plant) is unlikely to be
very  effective,  owing  to  the unusually  high electrical resistivity  of the
particulates.  More likely,  baghouses, following conventional cyclones, will
be the preferred technology  for  AFBC  power plants.   In the case of PFBC/com-
bined-cycle plants,  the degree  of  particulate removal required to avoid
undue  erosion  in the  gas  turbine  will probably insure  that  the  stack gases
will meet particulate emission standards.   If not, then a  baghouse might still
be  required at  the  stack  for additional  collection of fine particulates.

       At  least  in  principle,  the  lower  combustion  temperatures  employed
in FBC could enhance  the  formation  and  emission  of  polycyclic organic com-
pounds,  including  various  compounds  known  to  be  mutagenic  and/or carcino-
genic.   Analyses of FBC stack gases  reported  to date have  not shown signifi-
cant  concentrations  of such  organic compounds, but further determinations of
such materials (and of respirable particulates) in the stack gases from larger
FBC units, such  as  the  Rivesville  pilot  plant, should be  given high priority.

       Tightening of  S02  emission  standards  may  pose  more  severe problems in
terms  of  solid-waste  disposal  for FBC   power plants than for  conventional
plants using flue-gas desulfurization.  It has been  estimated that for  an AFBC
plant  burning a  typical  high-sulfur coal,  attainment  of  90% sulfur retention
                                      108

-------
                  Table 6.  FBC Emissions-Control Outlook
SO? EMISSIONS:
     •  Current or proposed NSPS are attainable with either AFBC or PFBC
        for practically any coal.

     •  Up to 1 ton of limestone or dolomite sorbent per ton of coal may
        be required in some situations.

NOV EMISSIONS:

     •  Expected from AFBC:  0.4-0.6 lb/106 Btu.

     •  Expected from PFBC:  0.2-0.4 lb/10  Btu.

     •  Further reduction possible by employing staged combustion
        techniques.

TRACE-ELEMENT EMISSIONS:                  (No present or proposed NSPS)

     •  May be less than from conventional coal combustion, owing to
        lower temperature.

     •  Definitive data not yet available.

EMISSIONS OF POLYCYCLIC ORGANIC COMPOUNDS;  (No present or proposed NSPS)

     •  May be more than from conventional coal combustion,
        owing to lower temperature.

     •  Definitive data not yet available.

PARTICULATE EMISSIONS:

     •  >99.0% collection efficiency required to meet present NSPS.

     •  >99.7% collection efficiency required to meet proposed NSPS.

     •  >99.9% collection efficiency required to protect gas-turbine
        blades in PFBC/combined-cycle applications.

     •  "Best-bet" collection technology:

            Cyclones plus baghouse for AFBC

            High-efficiency cyclones plus granular-bed or porous-
            ceramic filter for PFBC/combined-cycle.
                                      109

-------
would require approximately twice  as much  limestone  (and produce twice as much
spent  sorbent)  as  attainment  of the  present SC>2 standard  of 1.2  lb/10*>
Btu.l^   The  resulting exacerbation  of  the  spent-sorbent  disposal  problem
cou^d well  delay  the commercialization  of FBC  power  generation  or  force the
premature implementation of developing technology such as sorbent pretreatment
or regeneration to reduce  the  production of solid waste.
                                     110

-------
                       3  DESIGN AND ECONOMIC  PREMISES


       In  order  that  direct  comparisons   with  flue-gas-desulfurization
processes  reported  on by  the  Tennessee Valley Authority  (TVA)  may  be made,
design  and economic  premises  compatible  with  those  used  by TVA have been
employed.15

       The base case  for  the conceptual design and detailed engineering cost
studies  is an  approximately  500 MWe  (net)  new utility  power  plant burning
Illinois No.  6  coal with  a  sulfur  content (dry) of 3.86%.   This coal has a
moisture content of  12%,  an  ash content of 8.82%,  and a higher heating value
of  12,771 Btu/lb   (dry).   A  detailed  analysis  of  this  coal is  shown in
Table 7.
                 Table 7.  Characteristics  of  Illinois No.  6 Coal
            Coal Property                                Value

            Rank                                     HVC Bituminous

            Proximate Analysis (wt% as rec'vd)
              Moisture                                   12.00
              Ash                                          8.82
              Volatile Matter                            31.41
              Fixed Carbon                               47.77

            Ultimate Analysis (wt% dry)
              Carbon                                     69.52
              Hydrogen                                     5.33
              Nitrogen                                     1.25
              Sulfur                                       3.86
              Oxygen                                     10.02
              Ash                                        10.02

            Heating Value (Btu/lb dry)
              HHV                                        12771
              LHV                                        12222

            Free Swelling Index                            4.5

            Grindability Index (Hardgrove)                  57.4

            Initial Ash Fusion, (F)                  2120-2240

            Ratio of pyritic to organic sulfur              1.14
                                      111

-------
       The projected operating  life  of the utility power  plant  is assumed to
be 30  years,  representing 127,500 hours  of generating capacity.   This  is an
average of 4,250  hours  of operation per year.   The  projected  load factor for
the first year of operation is 0.8, or 7,000 hours of operation.
3.1  EMISSION REGULATIONS

       All of  the  alternative  power plants have been  designed  to comply with
the  proposed  New Source Performance Standards  (NSPS).   For  particulate
emissions,  this  is a  0.03-lb/106  Btu input.   At  this writing,  the  proposed
EPA  standard  for  sulfur-oxide  emissions  is an  85%  removal (daily average)  of
sulfur in the  coal as  mined, with  an emission ceiling of 1.2 lb/10° Btu and a
floor  of 0.2  Ib/lO^  Btu.   The 85%  removal  requirement  has  been used  as  a
design  factor  for all 'of  the  alternative technologies.   In the  case  of the
atmospheric  fluidized-bed  combustor,  a  limestone  sorbent  was  used with  a
three-to-one  calcium-to-sulfur  ratio.     For  the  pressurized  fluidized-bed
combustor, dolomite was used as  a  sorbent  with  a two-to-one calcium-to-sulfur
ratio.   It  has been  confirmed that these mole  ratios can remove  85%  of the
sulfur in tests  performed at Argonne National  Laboratory  and  the EXXON mini-
plant.   The gasification/combined-cycle  power  systems  achieve  sulfur removal
by  using standard industrial  t^S  removal techniques,  such as the Stretford
process  with  a Glaus  tailgas-cleanup system.    Sulfur  removal  in  the  solvent-
re' Ined-coal process takes  place during the hydrogenation  of  the  coal that  is
di -solved  in  a  coal-derived  solvent.   The  sulfur;  as hydrogen  sulfide,  is
flashed  off,  separated from the recycle  hydrogen,  and  converted  to elemental
sulfur.
3.2  PLANT LOCATION

       A  southern Illinois  plant  site has  been chosen  for estimating  pur-
poses.   This would place  the  plant  near  several demand centers  and  in close
proximity  to  extensive  coal  fields,  thus  minimizing transportation  costs.


3.3  PROJECT SCHEDULE

       A  construction  start date  of 1981 and a  plant start-up  date  of  1985
have been assumed  for  all  of the  technologies.   Costs have been calculated on
the  basis  of  1980 dollars  and  have  been  scaled based  on  the  extrapolated
average annual Chemical Engineering Cost Indices  as  shown in Figure 5.


3.4  ECONOMIC ASSUMPTIONS
3.4.1  Indirect Investment Charges

       This area includes the materials  and  labor  for equipment and installa-
tion and all costs (such as architect and engineering fees, contractor expenses,
construction expenses,  and  in-house engineering) that are  necessary  for con-
struction of  a grass-roots power  plant.   The engineering  design  and contin-
gency  factors  are based  on  the  developmental  status  of the  technology  and
experience with engineering projects.
                                      112

-------
             CHEMICAL ENGINEERING  PLANT INDEX
     330

     310

     290


     270

     250


 CE  230
INDEX
     210


     190

     170

     150


     130
                           I
                 1
1
1
         1970   1972
1974  1976  1978  1980  1982  1984   1986
        YEAR

     Figure 5
                          113

-------
       The contingency for all of the alternative technologies has been  taken
as 15% of the overall project  investment, plus  an additional  10%  on  equipment
costs  for  gasification,  acid-gas  removal,   fluidized  combustion, high-temp-
erature/high-pressure  particulate  removal,  and  coal-slurry  dissolving.
Start-up and modification allowances  are  estimated  as  10% of the total  fixed
investment.    The interest during  construction  has  been  estimated  at 15% of
the  sub-total fixed  investment for each  of  the alternative processes.  This
factor  is  equivalent to  the  interest accrued  on  borrowed funds at 10% per
year  assuming a capital structure of 60/40  debt-to-equity ratio and a  four-
year project expenditure schedule as  indicated in Table  8.


3.4.2  Working Capital

       Working  capital  consists  of  the  money invested  in coal  and  other raw
materials carried in  stock,  accounts  receivable,  and  cash  kept on hand
for  payment of  operating expenses.  For these cost estimates, working capital
has  been taken  as  equivalent  to three  weeks  of raw-material  costs,   seven
weeks of direct  costs, and seven weeks of  overhead costs.
3.4.3  Indirect Costs
       Following TVA  practice,  regulated-utility-company  economics have been
used in establishing capital charges.   The breakdown  for these capital charges
is shown in Table 9.
                   Table 8.  Project Expenditure  Schedule
                                                          Year
                                                                        Total
Fraction of total expenditure
  as borrowed funds

Simple interest at 10%/yr as a
  percent of total expenditure
Year 1 debt

Year 2 debt

Year 3 debt

Year 4 debt


Accumulated interest as percent
  of total expenditure
0.10    0.20     0.20    0.10     0.60
1.0
1.0

2.0
1.0
3.0
1.0

2.0

2.0
5.0
1.0

2.0

2.0

1.0



6.0
 4.0

 6.0

 4.0

 1.0



15.0
                                     114

-------
         Jable 9.  Annual Capital Charges for Power Industry Financing


                                              Percentage of total depreciable
                                                     capital investment	
                                                    Years remaining life
                                                     30      25       20

Depreciation-straight line (based on years
  remaining life of power unit)                     3.3     4.0      5.0
Interim replacements (equipment having less
  than 30-yr life)                                  0.7     0.4
Insurance                                           0.5     0.5      0.5
Property taxes                                      1.5     1.5      1.5

     Total rate applied to original
       investment                                   6.0     6.4      7.0
                                                 Percentage of unrecovered
                                                 	capital investment3
Cost of capital (capital structure assumed
  to be 60% debt and 40% equity)
     Bonds at 10% interest
     Equity'3 at' 14% return to stockholder
Income taxes (federal and state)c

          Total rate applied to depreciation base

a.  Original investment yet to be recovered or "written off."
b.  Contains retained earnings and dividends.
c.  Since income taxes are approximately 50% of gross return, the amount of
    taxes is the same as the return on equity.
d.  Applied on an average basis, the total annual percentage of original
    fixed investment for new (30-yr) plants would be 6.0% + 1/2 (17.2%) =
    14.6%.
3.4.4  Overheads

       Plant,  administrative,  and  marketing  overheads are  costs  that  vary
from company  to company.   With consideration of the  various  methods used in
industry and  illustrated  in a variety of cost-estimating sources, the follow-
ing method for estimating overheads is used.

       Plant  overhead  is estimated  as 50%  of the subtotal  conversion costs
less utilities,  and  includes the projected  costs  for  labor,  maintenance,  and
analyses.   Administrative  overhead  is  estimated  as  10%  of  operating labor
and  supervision.   Marketing  the product  is considered in the  estimation of
overheads and is defined as 10% of sales revenue.
                                      115

-------
3.4.5  By-product Sales

       In the evaluation of the annual revenue requirements,  credit  from  sale
of by-products  (such as ammonia  or sulfur  recovered  from low-Btu  combined-
cycle gasifiers) is deducted from the annual operating  costs  to  obtain the net
annual revenue  requirements.   The selling prices for  sulfur  and  ammonia  were
taken as $60 and $100 per long ton,  respectively.


3.4.6  Raw Material Costs
       Costs for  limestone  and  dolomite (needed for sulfur-oxides removal in
fluidized-bed units)  have been  taken as  $10  per  ton.   For  the purpose of
calculating annual  revenue requirements,  a  coal price of $1/10^  Btu has
been assumed.   This  results in  a  net coal price  of  $22.50  per ton of coal.
In  addition  to  this  fixed  coal  price,  a variation  from  $0.75-2/10**  Btu was
also used,  in  order to study the  sensitivity  of busbar  power  costs  to  fuel
costs.
                                    116

-------
                  4  ALTERNATIVE TECHNOLOGIES COMPARISON
       Using  the design and  economic  premises  outlined  in the  previous
section, a series of  detailed  capital and operating-cost estimates have been
prepared for  the technologies  under  discussion  as  alternatives  to FGD.  For
fluidized-bed  combustion,  capital requirements  were  determined  from detail-
ed  cost  estimates prepared  by Westinghouse  Corporation  for the  Environmen-
tal Protection Agency.16  xhe  estimates  for the four  gasification/combined-
cycle processes  (Lurgi,  IGT,  Foster-Wheeler, and Texaco) were prepared based
on  unpublished information  given to  Argonne National Laboratory.°   Informa-
tion needed to determine capital  requirements for both  a  solvent-refined-coal
production plant  and  a power plant burning  SRC were obtained from Reference
2.
4.1  HEAT RATES

       The information noted  above  was  used to determine heat rates for each
of the alternative processes.   These rates,  which  were  used  for capital design
costing and revenue-requirement calculations,  are  shown in Table  10.

       As can  be  seen,  the heat  rate  for the pressurized  f luidized-bed com-
bustor is approximately  15%  less  than  that  of the atmospheric unit.  This is
due  to  two factors:   1) the increased  pressures  and temperatures  of the
               Table 10.  Overall Heat Rates Of Alternative  Processes
                  Process                             Heat  Rate
                                                      (Btu/kWh)
                  AFBC                                  9618a

                  PFBC                                  8688b

                  G/CC:
                    Lurgi                              10856
                    IGT                                 8258
                    Foster Wheeler                      830S
                    Texaco                              8728

                  SRC                                   9000C
                                                       13040d

               a3:l calcium/sulfur ratio-limestone  sorbent.
               b2:l calcium/sulfur ratio-dolomite sorbent.
               cBased on SRC-burning power plant.
               "Based on Btu content of coal input  to SRC production
                plant.
                                     117

-------
pressurized FBC  over  that  of the  atmospheric  FBC result in  inherently  more
efficient operation,  and 2)  the  ability of the pressurized fluidized-bed  unit
to utilize a  2-to-l  calcium-to-sulfur  absorbent  ratio as compared to  the  3-
to-1 ratio for  the  atmospheric  combustor,  results in  less  energy  being  used
in heating  and  calcining of  sorbent.    The overall heat  rates  of the  IGT,
Foster-Wheeler,   and Texaco  gasifiers  are less than  that of  the older Lurgi
gasifier.  Although  differing  by  approximately  6%,  the  heat rates of these
three gasifiers may be  considered  to be  equivalent  at  this  stage of develop-
ment.   For solvent-refined  coal,  the  heat  rate  for a power  plant utilizing
SRC  as  a fuel  is shown  as  9,000 Btu/kWh.  What  is  interesting to  note,  how-
ever, is  that when one  takes  into  account  the Btu losses incurred  during the
SRC  production process  and relates the power-plant heat  rate  back  to  the in-
let  feed coal, a new  heat rate of 13,040  Btu/kWh is found.
4.2  CAPITAL COSTS
4.2.1  Solvent-Refined Coal

       The  capital-cost  breakdown  for  an  SRC-I production  plant  is  shown
in  Table  11.   This cost  estimate  is  based on a  plant  with  a coal input  of
20,000  ton/day,  which is  the  size considered most  likely for a  commercial
facility at  this  time.   As a  basis  for  comparison,  it should be noted  that
such  a plant operating  at full  capacity would produce  enough  fuel for  a
1400-MW power plant.   The capital costs include  the equipment  necessary
for ash removal  (filtration),  sulfur removal, and  sulfur recovery  (as
elemental sulfur).
4.2.2  Gasification/Combined-Cycle Power Plants

       The  capital costs  for the  four  gasification/combined-cycle power
plants  analyzed  in this  paper  are shown  in Table  12.   Depending upon  the
gasifier chosen, total capital investment can be  seen  to  range  from slightly
over  $450 million  for  the IGT unit to well over  $600 million for  the  Lurgi-
based power  plant.   All of the plants include facilities for acid  gas (H2S)
removal and elemental sulfur recovery.  There  are also  ammonia-recovery
facilities  (except for the Texaco gasifier), because this is  a  by-product of
the gasification process.
4.2.3  Fluidized-Bed Combustors

       The  capital-cost breakdown  for  the  atmospheric  and  pressurized
fluidized-bed power plants are  shown in Tables 13 and 14,  respectively.
These  estimates  include  the  equipment necessary  for sorbent storage,  pul-
verization,  and  feed  to the  f luidized-bed  unit, as  well  as  spent-sorbent
removal.
4.3  ANNUAL REVENUE REQUIREMENTS

       The annual revenue requirements  for all of the alternative technologies
are shown in Tables 15 through  22.   In  all cases except that of the SRC plant
                                     118

-------
           Table 11.  Solvent-Refined-Coal Production Plant
                                            rj      '    '
                      Capital-Cost Breakdown
Categories                          Costs (Millions of 1980 Dollars)
Co'al Preparation                                  40.4
Coal-Slurry Dissolving, Benfield,
Cryofining, and Oil Absorption                   205.3
Filtration                                       160.4
Solvent Degassing and Recovery                    66.0
Hydrogen Plant
  Koppers-Totzek                                  96.2
  Air Separation                                  37.9
  Acid-Gas Removal                                29.8
  Shift and Purification                          25.1
Sulfur Recovery                                   18.4
Waste-Water Treatment                             17.1
Product Storage & Shipping                         9.4
Support Facilities                                29.4
       Total Plant Investment                    735.4
               A/E Home Office and Fee
                 (10% of Estimated BOP)           44.1
               Cont ingency                       14 3.4
               Interest During Construction      138.4
               Total Depreciable Investment     1061.3
               Start-up and Modifications        106.1
Land                                               1.5
Working Capital0                                  47.9
       Total Capital Investment                 1216.8
rt
 Approximately 20,000 ton/day coal input.
 15% of plant investment plus 10% additional each on Coal-Slurry Dissolving,
 Benfield, Gasifier and  Acid-Gas Removal.
Equivalent to 3 weeks of raw materials, 7 weeks of direct costs, and 7
 weeks of overhead costs.
                                  119

-------
Table 12.  Gasification/Combined-Cycle Power Plant Capital"
           Cost Breakdown
Costs (Millions of 1980 Dollars)
Categories
Coal Handling
Oxidant Feed
Gasification and Ash Handling
Gas Cooling
Acid-Gas Removal and Sulfur Recovery
Proces s~ Condensate Treatment
Steam, Condensate and BFW
Support Facilities
Combined-Cycle Components
Total
A/E Home Office & Fee @ 10% of estimated BOP
Labor, Materials & Indirects Contingency
Interest during Construction
Total Depreciable Investment
Start up and Modifications
Land
Working Capital
Total Capital Investment
Lurgi
12.77
2.47
63.01
29.97
38.73
55 98
30.32
25.50
126.91
385.66
25.27
68.02
71.84
550.79
55.0
1.5
17.60
624.89
IGT
12.60
2.58
22.50
44.05
32.29
9 78
1.35
23.45
161.29
309.39
20.28
51.89
57.23
438.79
44.0
1.5
13.79
454.08
Foster \ATheeler
18.47
2.49
27.42
25.73
29.13
7 55
1.26
22.48
152.01
286.54
18.78
48.64
53.09
407.05
41.0
1.5
12.82
462.37
Texaco
12.14
4.01
33.08
73.99
26.68

0.52
15.95
122.79
289.16
18.95
49.35
53.62
411.08
41.0
1.5
12.86
466,44

-------
Table 13.  AFBC Power Plant Capital-Cost Breakdown
Categories
Steam Generators
Turbine Generator
Process Mechanical Equipment
Electrical
Civil and Structural
Process Piping and
Ins trument at ion
Yardwork and Miscellaneous


Major
Components
47.36
27.23
12.40





86.99
Costs (Millions of
1. Direct 2. Indirect 3.
Labor Field
13.59 12.23
2.04 1.84
8.24 7.42
16.48 14.83
13.89 12.50

11.07 9.96
2.11 1.90
67.42 60.68
BOP Labor, Materials & Indirects (1 +

A/E Home
Office & Fee @ 10%
1980 Dollars)
Balance-of -Plant
Materials
5.896
0.094
27.52
11.51
12.83

9.455
1.591
68.91
2 + 3)

Contingency

Interest
during Construction

Total Depreciable Investment

Land

Working Capital
Total Capital Investment
Start-up




and modifications










Total
79.08
31.20
55.58
42.82
39.22

30.49
5.6
283.99
197.01
19.70
50.29
53.1
407.08
40.7

2.4
14.06
464.24

-------
Table 14.  PFBC Power Plant Capital-Cost  Breakdown
Categories
PFB Steam Generators
Turbine Generators
Process Mechanical Equipment
Electrical
Civil and Structural
Process Piping and
Instrumentation
Yardwork and Miscellaneous




Land
Working Capital
Total Capital Investment
Costs (Millions of 1980 Dollars)

Major 1. Direct 2. Indirect 3. Balance-of-Pl ant
Components Labor Field Materials Total
41.35 5.32 4.79 2.655
47.44 2.09 1.88 .171
17.98 7.67 6.90 22.436
11.64 10.48 9.42
12.21 10.99 9.59
16.51 14.86 17.21
1.95 1.75 1.46
106.77 57.39 51.65 62.94
BOP Labor, Materials & Indirects (1+2+3)
A/E Home Office & Fee @ 10%
Contingency
Interest during Construction
Total Depreciable Investment
Start-up and Modifications

54.11
51.58
54.99
31.54
32.79
48.58
5.16
278.75
171.98
17.20
51.45
52.11
399.51
40.0
2.4
13.12
455.03

-------
                     Table 15.   Total Average Annual Revenue Requirements - SRC Production Plant
CO

Direct Costs
Coal ($1/1B6 Btu)
Filter Aid and Other Chemicals
Raw Water ($.4/1000 gal)
Ash Disposal ($l/ton)
Electric Power (2.5£/kWh)
Maintenance and Operation
Total Direct Costs
Indirect Costs
Capital Charges
Depreciation, replacements and
insurance (6% of TDI)
Average cost of capital and
insurance (8.6% of TCI)
Overheads
Plant (50% of O&M)
Administrative (10% of O&M labor)
Total Indirect Costs
Total Costs
Hydrocarbon byproduct
credit ($2/106 Btu)
Total Annual Revenue Requirements
Unit SRC Revenue Requirements $/106 Btu>.SRC output
$/ton SRC output
$/106 Btu Coal input
$/ ton Coal input
8,760 hr/yr
(1.0 Capacity)
Millions
194.0
6.8
1.9
1.1
35.0
24.8
263.6
63.7
104.6
12.4
1.2
181.9
445.5
(36.1)
409.4
3.13
97.3
2,11
47.5
7,000 hr/yr
(0.8 Capacity)
of 1980 Dollars per year
155.2
5.4
1.5
0.9
28.0
22.3
213.3
63.7
104.6
11.2
1.2
180.7
394.0
(28.9)
365.1
3.42
108.
2.35
52.9

-------
                  Table 16.  Total Average Annual Revenue Requirements - SRC Power Plant*
Direct Costs
SRC ($3.42/106 Btu)
Maintenance and Operation

Total Direct Costs
                                                                       Annual Cost($)
                                                         7,000 hr/yr
                                                        (0.8 Capacity)
107,730,000
  4,200,000

111,930,000
                             4250 hr/yr
                           (0.49 Capacity)
65,408,000
 3,500,000

   908,000
Indirect-Costs
Capital Charges
  Depreciation, interim replacements and
  insurance at 6% of total depreciable
investment
Average cost of capital and taxes at
8.6% of total capital investment
Overheads
Plant, 50% of O&M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills/ kWh c
$/ton coal
$/10 Btu input
18,000,000
27,090,000

2,100,000
100,000
47,290,000
159,220,000
45.5
78.1
3.46
18,000,000
27,090,000

1,750,000
100,000
46,940,000
115,848,000
54.5
93.7
4.16
 500 MWe output.
 Capital charges based on $600/kW depreciable investment and $630/kW total investment.
CBased on coal feed to SRC production plant.  31% Btu loss in SRC production.

-------
                  Table 17.  Total Average Annual Revenue Requirements - Lurgi Combined-Cycle Plant'
fO
Direct Costs
Coal ($1/186 Btu)
Catalyst and Chemicals
Utilities
Ash Disposal (on-site)
Maintenance and Operation
Sulfur Credit ($60/long ton)
Ammonia Credit ($100/lbng ton)

Total Direct Costs
                                                                 7,000 hr/yr
                                                                (0.8 Capacity)
                                                                             Annual Cost ($)
40,576,430
   625,872
   732,024
   636,239
 9,314,600
(3,227,424)
(1,859,138)

46,798,603
                                                                                   4250 hr/yr
                                                                                 (0.49 Capacity)
24,853,063
   383,347
   448,365
   389,696
 6,412,380
(1,976,798)
(1,138,722)

29,371,331
           Indirect Costs
           Capital Charges
             Depreciation,  interim replacements and
             insurance at 6% of total depreciable
             investment
             Average  cost of capital and  taxes at 8.(
             of  total capital  investment
           Overheads
             Plant, 50% of  0 & M
             Administrative, 10% of operating labor
           Total Indirect Costs
           Total Annual Revenue Requirements
           Unit  Revenue Requirements mills/KWh
                                     $/ton  coal
                                     $/106 Btu input
                                                      33,047,400

                                                      53,740,540
                                                       3,744,800
                                                         182.500
                                                      90,715,240
                                                     137,513,840
                                                          36,9
                                                          76.3
                                                           3.39
                             33,047,400

                             53,740,540
                              2,293,690
                                182.500

                             89,264,130
                            118,635,460
                                 52.4
                                107.
                                  4.77
                8 MTJe niil-niil-

-------
                     Table  18.   Total Average Annual  Revenue  Requirements  -  IGT  Combined-Cycle Plant'
ON
Direct Costs
Coal ($1/106 Btu)
Catalyst and Chemicals
Utilities
Ash Disposal (on site)
Maintenance and Operation
Sulfur Credit ($60/long ton)
Ammonia Credit ($100/long ton)
Total Direct Costs
Indirect Costs
Capital Charges
Depreciation, interim replacements and
insurance at 6% of total depreciable
investment
Average cost of capital and taxes at
8.6% of total capital investment
Overheads
Plant, 50% of 0 & M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills /kWh
$/ton coal
$/106 Btu input
7,000 hr/yr
(0.8 Capacity)
30,309,250
159,843
440,000
477,404
6,933,320
(2,608,433)
(124,299)
35,587,085

26,327,400
39,050,880
3,591,660
175,000
69,144,940
104,732,025
28 ..6
77. §
3^46
Annual Cost ($)
4250 hr/yr
(0.49 Capacity)
18,564,416
97,904
269,500
292,410
4,924,784
(1,597,665)
(76,133)
22,475,216

26,327,400
39,050,880
1,587,392
175,000
67,140,672
89,615,888
40.3
109.
4.83
            523.2 MWe output.

-------
ro
Direct Costs
Coal ($1/106 Btu)
Catalyst and Chemicals
Utilities
Ash Disposal (on-site)
Maintenance and Operation
Sulfur Credit ($60/long ton)
Ammonia Credit ($100/long ton)
Total Direct Costs
Indirect Costs
Capital Charges
Depreciation, interim replacements and
insurance at 6% of total depreciable
investment
Average cost of capital and taxes at
8.6% of total capital investment
Overheads
Plant, 50% of 0 & M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills/kWh
$/ton coal
$/106 Btu input
Annual Cost ($)
7,000 hr/yr
(0.8 Capacity)
28,429,704
90,945
628,571
447,800
6,594,040
(2,377,757)
(1,315,232)
32,498,071

24,423,000
39,763,820
2,422,020
175,000
66,783,840
99,281,911
29.1
78.6
3.49
4250 hr/yr
(6.49 Capacity)
17,413,194
55,704
385,000
274,277
4,716,975
(1,456,376)
(805,580)
20,583,194

24,423,000
39,763,820
1,483,488
175,000
65,845,308
86,428,502
41.7
112.
4.96
             488.1 MWe output.

-------
                     Table  20.   Total Average Annual Revenue Requirements - Texaco Combined-Cycle Plant'
CO

Direct Costs
Coal ($1/106 Btu)
Catalyst and Chemicals
Utilities
Ash Disposal (on-site)
Maintenance and Operation
Sulfur Credit ($60/long ton)
Ammonia Credit ($100/long ton)
Total Direct Costs
Indirect Costs
Capital Charges
Depreciation, interim replacements and
insurance at 6% of total depreciable
investment
Average cost of capital and taxes at 8.6%
of total capital investment
Overheads
Plant, 50% of 0 & M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills/kWh
$/ton coal
$/L06 Btu input
Annual Cost($)
7,000 hr/yr
(0.8 Capacity)
27,015,314
51,918
847,262
425,521
7,045,160
(2,191,024)

33,194,151

24,664,800
40,113,840
4,335,080
162,500
69,276,220
102,470,371
33.2
85.3
3.79

4250 hr/yr
(0.49 Capacity)
16,546,881
31,800
518,948
260,632
4,944,848
(1,342,002)

20,961,107

24,664,800
40,113,840
3,284,924
162,500
68,226,064
89,187,171
47.6
121.
5.39
         441.2 MWe output.

-------
                      Table  21.  Total Average  Annual  Revenue  Requirements  - AFBC Power Plant*1
                                                                               Annual  Cost  ($)
                                                                     7,000 hr/yr
                                                                   (0.8  Capacity)
                                     4250 hr/yr
                                   (0.49 Capacity)
        Direct  Costs
        Coal  ($1/106 Btu)
        Sorbent ($10/ton)
        Spent-Sorbent Disposal (on-site)
        Maintenance and  Operation
        Total Direct Costs
33,673,500
 6,015,800
 2,523,920
 6,910,000
49,123,220
20,444,625
 3,652,450
 1,532,380
 5,590,000

31/219,455
VO
Indirect Costs
Capital Charges
Depreciation, interim replacements, and
insurance at 6% of total
depreciable investment
Average cost of capital and taxes at
8.6% of total capital investment
Overheads
Plant, 50% of O&M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills /kWh
$/ton coal
$/106 Btu input




24,424,800

39,924,640

3,455,000
250,000
68,054,440
117,177,660
33.5
78.3
3.48




24,424,800

39,924,640

2,795,000
250,000
67,394,440
98,613,895
46.4
109.
4.82
         500 MWe output.

-------
             Table 22.  Total Average Annual Revenue Requirements - PFBC  Power  Plant'
Direct Costs

Coal ($1/106 Btu)
Sorbent ($10/ton)
Spent-Sorbent Disposal (on-site)
Maintenance and Operation

Total Direct Costs
                                                           7,000 hr/yr
                                                         (0.8 Capacity)
                                                                       Annual Cost ($)
30,397,500
 6,136,200
 2,287,040
 7,070.000
45,890,740
                                   4250 hr/yr
                                 (0.49 Capacity)
18,455,625
 3,725,550
 1,388,560
 5,730,000
29,299,735
Indirect Costs

Capital Charges
  Depreciation, interim replacements, and
  insurance at 6% of total depreciable
investment
Average cost of capital and taxes at
8,6% of total capital investment
Overheads
Plant, 50% of O&M
Administrative, 10% of operating labor
Total Indirect Costs
Total Annual Revenue Requirements
Unit Revenue Requirements mills/ kWh
$/ton coal
$/106Btti input
23,970,600
39,132,580
3,535,000
250,000
66,888,180
112,778,920
32.2
83.5
3.71
23,970,600
39,132,580
2,865,000
250,000
66,218,180
95,517,915
45.0
116.
5.18
 500 MWe output

-------
revenue  requirements  for  both  7,000 hours  a year  (0.8  capacity) and  4,250
hours  per year  (0.49 capacity) have  been shown.   In the case of the  SRC
production plant,  the annual  revenue  requirements  necessary  for  full  capacity
(8,760 hours  per  year)  and  0.8  capacity  (7,000  hours per  year) have  been
shown.  The 7,000  hours  of operation  per  year is considered  the  norm at  which
the SRC  plant  will be operating.   Also,  unit revenue  requirements are  shown
both  in  terms  of  SRC fuel output  and  in  terms  of coal input.   It should be
noted  that  almost half  of the total annual  revenue requirement  for the  SRC
production plant  is  made  up  of  the cost of  coal  needed  to produce  the  SRC.
This large amount of coal is necessary to produce the required  amount  of  SRC-I
due in part  to the process thermal  efficiency.   In a  report  by  Air  Products
and Chemicals, Inc. and  Catalytic,  Inc.2  it was  reported  that  "A direct  ratio
of  saleable  product  to  total coal feed  gives  a plant thermal  efficiency of
78%."   This  is based on treatment of  the plant's  electrical  requirements as
an  operating  rather  than  an  energy  expense.   If  electricity  use  is  included
in  the energy balance,  the ratio  of saleable  product  to total  energy  input
gives  a plant thermal efficiency of 73%.

       Table  16  utilizes  a  fuel  cost  of  $3.42/10^ Btu for determination of
the annual  revenue requirement for a power plant  burning SRC.   It should be
noted, in this case,  that  two-thirds of  the  annual-revenue requirement  con-
sists  of  costs expended  for SRC fuel itself and that the unit  revenue  require-
ment  (mills/kWh)  for  the SRC power plant is  the highest of any  of the alter-
native technologies studied.

       The  average annual revenue  requirements  for  the  gasifier processes
covered  in  this report  can be seen  in Tables 17  through 20.   It was  found
that  the  IGT  and Foster-Wheeler gasifiers result  in nearly identical busbar-
power  costs.    These  two  alternative  technologies  also had  the  lowest  unit
revenue requirements of  any looked  at in  this study.   The Texaco gasifier  was
found  to  have  a unit revenue requirement  greater  than that  of either of  the
previous  two,  while  the  Lurgi gasifier  had the highest unit revenue  require-
ment  of  any  of the gasification/combined-cycle  power  systems.  It should be
noted  that  in all cases  (where applicable)  both  sulfur  credits  and ammonia
credits for sale  of  by-products have been factored  into  the revenue  calcula-
tions .

       Tables  21  and  22  display  the  revenue  requirements  necessary  for
operation  of   atmospheric   and  pressurized  fluid-bed   power  plants,  respec-
tively.   The analysis found that the  pressurized  fluid  bed has a smaller  unit
revenue requirement than the  atmospheric.   This is  due in part  to the higher
efficiency  of   the pressurized  fluid-bed unit,  resulting in lower capital
costs, as previously  discussed, and  lower operating  costs,  i.e., less  coal
consumed.
                                      131

-------
                               5  CONCLUSIONS


       A summary of the operating and economic characteristics of the alterna-
tive processes  is  shown in Table 23.  The  overall  heat rates of the pressur-
ized-fluid-bed,  IGT gasifier;  Foster-Wheeler  gasifier,  and  Texaco gasifier
power plants  are  approximately equivalent and  are  lower  than that of compar-
able  conventional  coal-burning power  plants  with  flue-gas  desulfurization
(about 9,900  Btu/kWh).  The  total  capital investment, in terms of dollars per
kW, would  tend to  indicate  that  the SRC power plant is the  cheapest  one to
construct.   This cost  of $630 per  kW is  less than for even a conventional
power plant,  in 1980 dollars.   It  should  be noted, however,  that  the total
capital  investment  for  the SRC power plant  does  not include that for the SRC
fuel-production  facility.    If this  cost is  included  as  part  of  the total
capital  investment,  then  greater  than $1,100 of  investment  per kW is needed.
This is  comparable  to the investment needed for a Lurgi combined-cycle plant,
which was  the most  expensive of the  units  studied.   Examination of the total
power cost,  in mills/kWh, of the alternative processes  leads to several con-
clusions,  including:  1)  at  the current  stage of development, solvent-refined
coal is not a viable option for replacement of conventional coal power genera-
tion  with  flue-gas desulfurization;  2) of the  gasification/combined-cycle
systems studied, the IGT and Foster-Wheeler processes offer the best possibil-
ities  for  replacement  of  conventional  coal-burning  power  plants  (Whether or
not the  Texaco  gasification  process  will  be able to compete successfully with
these  will depend  upon  further  cost  refinements   and  reduction  in  overall
capital  requirements.),  and  3)  pressurized  fluidized-bed  combustion  is  a
viable  option for  electric-power  generation and may become even more  so if
sorbent-regeneration processes are successfully developed.

       To  further  investigate tradeoffs  between  the technologies,  the annual
unit revenue  requirements  are  plotted as  a  function of fuel cost in Figure 6.
A  change in  the  cost  of  coal  is shown  to have little  or no  effect  on the
economic rankings,  although some of the processes show a greater dependence on
fuel cost  than do others.

       In  summary,  seven processes have been examined as possible alternatives
to  conventional  pulverized-coal  combustion with   flue-gas  desulfurization.
(Power costs  for  such a system are  estimated at 34 mills/kWh for this compar-
ison.)   SRC  is  clearly the highest-cost alternative,  at  over  45 mills/kWh,
with the Lurgi-based G/CC system  second  at  about 37 mills/kWh.  The remaining
five options  all  fall between approximately  29  and 33  mills/kWh, a variation
of  about  13%,  which  is well  within the uncertainties  inherent  in costs of
developing  technologies.   The lowest-cost options  in this  group are the G/CC
plants utilizing the  IGT  and Foster-Wheeler gasifiers.   These are followed by
the PFBC plant, which  is  also a combined-cycle  system.   This  type of cycle
helps to increase  plant  efficiency,  but at  the expense  of a  more  complex
system.

       It should be noted in closing  that all of the  processes looked at under
this study are at  best in the pilot-plant  or  demonstration stage  insofar as
electric-power  generation is concerned.   Therefore, great  caution should be
taken in using  the  costs  given in this paper as absolute rather  than relative
numbers.   Problems  with scale-up,  discoveries during development, and changes
in regulatory  constraints can deeply affect  the  results  that have been shown
here.
                                      132

-------
                  Table 23.  Alternative Processes Summary
Gasification Combined-Cycle Systems

MWe net
Overall Heat
Rate (Btu/kWh)
Total Capital
Investment ($/kW)
Capital Cost3
(mills/kWh)
Fuel costb
(mills/kWh)
Other Costs
(mills/kWh)
Total Power
Costc
(mills/kWh)
AFBC
500
9618
928
18.4
9.6
5.5
33.5
PFBC
500
8688
910
18.0
8.7
5.5
32.2
LURGI
532.8
10856
1173
23.3
10.9
2.7
36.9
IGT
523.2
8258
868
17.9
8.3
2.4
28.6
F-W
488.1
8303
947
18.8
8.3
2.0
29.1
TEXACO
441.2
8728
1057
21.0
8.7
3.5
33.2
SRC
500
9000
(13040)
630
12.9
30.8
1.8
45.5
aBased on 6% of Total Depreciable Investment and 8.6% of Total Capital Investment.
bfiased on $1.00/MBtu base coal cost.
cFirst-year costs for 7000 hours of operation.  The corresponding cost for a con-
 ventional power plant equipped with FGD is estimated to be 34 mills/kWh.
                                        133

-------
          EFFECT OF  COAL  PRICE  ON POWER  COST
         60
         55
         50
BUS  BAR
  COST,
mills/kWhr
         45
         40
         35
         30
         25
                                       SRC
              FOSTER WHEELER
1
                                             I
                0.75     1.00    1.25    1.50

                        COAL COST, $/MBtu

                        Figure 6
                    1.75    2.00
                              134

-------
 REFERENCES

 1.  B.K. Schmid and D.M. Jackson, Recycle SRC Processing for Liquid and Solid
     Fuels, presented at the Fourth Annual International Conference on Coal
     Gasification, Liquefaction and Conversion to Electricity, U. of Pittsburgh
     (Aug. 2-4, 1977).

 2.  Air Products and Chemicals, Inc., Assessment of Status of Technology for
     Solvent Refining of Coal, Argonne National Laboratory Report ANL/ECT-3,
     Appendix B (Dec. 1977).

 3.  E. N. Givens et al., Chemical Characterization, Handling and Refining of
     Solvent Refined Coal to Liquid Fuels-Final Report, FE-2003-27 (Sept. 1977).

 4.  W. L. Sage et al., Characterization of Solvent Refined Coal: Dual Register
     Burner Tests-Final Report, EPRI FP-628 (Jan. 1978).

 5.  Southern Company Services, Inc., Full-Scale Utility Boiler Test with Solvent
     Refined Coal (SRC), FE-2222-8 (April 1978).

 6.  Energy Daily (Oct. 27, 1978).

 7.  L.J. Muzio and J.K. Arand, Small Scale Evaluation of the Combustion and
     Emission Characteristics of SRC Oil, Div. of Fuel Chemistry, ACS, preprints,
     23(1): 140-150 (March 1978).

 8.  Unpublished information from United Technologies Research Center to Argonne
     National Laboratory (Jan. 1979).

 9.  Technical Notes for the Conceptual Design for an Atmospheric Fluidized-Bed
     Direct-Combustion Power Generating Plant, Vol. 1, Stone & Webster Engineer-
     ing Corp. for USDOE, Report No.  HCP/T2583-01/1 (April 1978).

10.  Problems in Applying Proposed Revised New Source Performance Standards for
     Pulverized-Coal-Fired Combustors to Fluidized-Bed Combustors, draft report
     prepared by Gilbert Associates and the MITRE Corp. for USDOE (Oct.  1977).

11.  H. Feibus et al., Commercialization Strategy Report for Advanced Electric
     Generation Technologies,  USDOE Task Force Draft Report No.  TID-28839
     (1978).

12.  H. L. Falkenberry, The Fluidized-Bed Comb,ustion Program of the Tennessee
     Valley Authority, Proc. of the Fifth International Conf. on Fluidized-Bed
     Combustion, Washington, D.C. (Dec. 1977).

13.  J. J. Markowski,  comments reported in Proc. of the Fifth International
     Conf. on Fluidized-Bed Combustion, Washington, D.C. (Dec. 1977).
                                      135

-------
14.  A. A. Jonke, Argonne National Laboratory, personal communication  (1978).

15.  R. L. Torstrick et al.,  Eoonomia Evaluation Techniques, Results, and
     Computer Modeling for Flue Gas Desulfurization, presented at USEPA
     Flue Gas Desulfurization Symposium, Hollywood, Fla. (Nov. 8-11, 1977).

16.  R. A. Newby et al., Effect of S'02 Emission Eequirements on Fluidised-Bed
     Combustion Systems: Pveliminaini Technical/Economic Assessment, EPA Report
     600/7-78-163 (Aug. 1978).
                                     136

-------
                                                                 3B
ECONOMICS AND ENERGY REQUIREMENTS OF SULFUR OXIDES CONTROL PROCESSES
                                 By

       G. G. McGlamery, T. W. Tarkington, and S. V. Tomlinson
                Emission Control Development Projects
                     Tennessee Valley Authority
                       Muscle Shoals, Alabama
                    Prepared for Presentation at
                 Flue Gas Desulfurization Symposium
          Sponsored by U.S. Environmental Protection Agency
            Industrial Environmental Research Laboratory
               Research Triangle Park, North Carolina
                               Held at
                           Caesars Palace
                          Las Vegas, Nevada
                           March 5-8, 1979
                                  137

-------
  ECONOMICS AND ENERGY REQUIREMENTS OF SULFUR OXIDES CONTROL PROCESSES
                                   By

         G. G. McGlamery, T.  W.  Tarkington, and S. V. Tomlinson
                  Emission Control Development Projects
                       Tennessee Valley Authority
                         Muscle Shoals, Alabama
                                ABSTRACT
     As part of a continuing program to evaluate the design, energy con-
sumption, and economics of sulfur oxides control processes, this paper
presents the results from three separate studies being carried out for
EPA by TVA.  Energy and preliminary economic requirements for three
physical and three chemical coal-cleaning processes are given along with
similar evaluations of a number of FGD processes.  The FGD evaluations
cover technical updates of older systems such as the lime and limestone
throwaway processes and the magnesia and sodium scrubbing to produce sul-
furic acid.  In addition, a sulfur-producing option, Wellman-Lord scrubbing
with coal reduction of S02 by Allied Chemical, is presented for the first
time.
                                    138

-------
    ECONOMICS AND ENERGY REQUIREMENTS OF SULFUR OXIDES CONTROL PROCESSES
                                 INTRODUCTION
     The EPA-TVA interagency program to evaluate the design and economics
of sulfur oxides control processes for fossil-fueled power plants continues
into its twelfth year.  Nine major reports have been distributed, three
are about to be published, and projects which will result in five more
are now underway.  In addition to conceptual design and comparative
economic evaluation of FGD processes, these projects include byproduct
marketing studies, sludge disposal economics, and an energy and economic
evaluation of physical and chemical coal-cleaning techniques.  Throughout
this period, research groups, utilities, vendors, and regulatory agencies
have utilized these studies in decision-making processes related to
sulfur oxides control.

     In this paper the preliminary energy requirements and economic
results of three EPA-sponsored studies are presented.  Data are reported
from a study of physical and chemical coal-cleaning processes.   A ground-
to-ground energy study of limestone, lime, and magnesia scrubbing FGD
economics is discussed, with emphasis on updated technology for the
magnesia process.  Also discussed is a second FGD evaluation of the
Wellman-Lord FGD process technology coupled with either sulfuric acid or
sulfur production using the Allied Chemical coal-reduction process.
These two FGD studies are evaluated on the same basis and are also
comparable to the limestone, double-alkali, and citrate process economics
presented at the November 1977 FGD Symposium by R. L. Torstrick, et al.,
Economic Evaluation Techniques, Results, and Computer Modeling for Flue
Gas Desulfurization.  (Proceedings:  Symposium on Flue Gas Desulfurization,
Volume II; EPA 600/7-78-058b, 1978)

     The premises and cost values used in these companion FGD evaluations
are listed in Appendix A.  Slightly different premises used in the coal-
cleaning study also are shown in Appendix A.

     In the future, changing conditions will require modifications to
many of the premises used in these studies.  Emission regulations,
economic conditions, and industry practices are all changing.  Premises
covering equipment redundancy; particulate, SOX, and NOX removal; waste
disposal techniques; flue gas bypassing; reduced on-stream time for
boilers; and allowance for inflation over the operating life of the
system are some which will be revised in the near future.
                                    139

-------
                      PHYSICAL AND CHEMICAL COAL CLEANING
     During recent years there have been indications that physical coal
cleaning (PCC) and chemical coal cleaning (CCC),  either separately or in
combination with FGD,; could help utilities reduce the cost of meeting
sulfur oxides emission'requirements.   At EPA's request a study was
undertaken by TVA to evaluate the economics of three PCC processes, and
three CCC processes.  Preliminary results of this study presented here
are part of a larger study to be published later  this year.   In addition
to the.results described in this paper,  the larger study will also
include a combination PCC-CCC process using two of the processes being
evaluated, a PCC process followed by FGD, and a CCC process  followed by
FGD.  In the full study coal sulfur contents of 0.7, 2.0, and 3.5% will
also be evaluated in addition to the 5.0% sulfur  coal presented here.
Premise conditions which differ from the FGD premises are (1) an update
of the time base to a 1979-1982 construction period and a 1982 startup,
(2) the use of a 2000-MW power plant for the base case, (3)  a more
detailed coal composition, and (4) a change in boiler operating time to
5500 hr/yr.  Direct comparison to other  evaluations should take these
differences in consideration.
PROCESS DESCRIPTIONS

     The three PCC processes represent widely used commercial technology
and were selected for study because they offer a relatively high level
of sulfur reduction compared to other PCC methods.  The three CCC processes
are not commercial processes, but have been developed to bench-scale or
limited pilot-plant stages.  Additional development could make signifi-
cant changes in their technical, and thus economic, potential for sulfur
reduction.  The PCC processes are somewhat limited in their desulfurization
application since they only remove pyritic sulfur.  Two of the CCC
processes remove significant quantities of organic sulfur in addition to
pyritic sulfur.

Physical Coal Cleaning

     PCC Process I.  This process uses a dense-medium vessel for the
coarse coal, a dense-medium cyclone for the intermediate-sized coal, and
froth flotation for the fine coal, as shown in Figure 1.

     The 3-inch x 0 coal is crushed and screened to three size fractions:
37% 2-inch x 3/8-inch coarse coal, 55% 3/8-inch x 28-mesh intermediate-
sized coal, and 8% 28-mesh x 0 fine coal.  The coarse coal is immersed
                                   140

-------
COAL RECEIVING
 AND STORAGE
                                                                               WATER	*
  RAW COAL
   SIZING
COARSE COAL
 CLEANING
INTERMEDIATE
   COAL
  CLEANING
  FINE COAL
  CLEANING
                                DENSE MEDIUM VESSEL
                         RINSE      WITH DRAINAGE          "7 RINSE
                        SCREEN                           |SCREEN
                        (SINK)                          JIFLOATI       ^
   REFUSE
  DISPOSAL
 CLEAN COAL
  STORAGE
                                                                              CLEAN COAL
                                                                               SHIPMENT
Figure  1 .   Flow diagram for  physical coal-cleaning  process  I.
                                            141

-------
in a 1.55 specific-gravity magnetite - water slurry in trough-type
dense-medium vessels.  The float fraction, about 85% of the feed, is
rinsed and drained, dewatered in basket centrifuges, and conveyed to the
clean coal stockpile.  The 15% sink fraction is rinsed, drained, and
discarded as refuse without centrifugal dewatering.  Integral drain
screens recover the magnetite from the clean coal and refuse.

     The intermediate-sized coal is slurried in a pulping tank and
cleaned in dense-medium cyclones at a nominal specific gravity of 1.55.
The cyclones are followed by conventional drain-and-rinse screens and by
basket centrifuge dewatering of both the clean coal float fraction,
which is sent to the clean coal stockpile, and the sink fraction, which
is discarded as refuse.

     Single-stage froth flotation is used on the entire fine coal
fraction.  For flotation feed, a pulp density of 10% solids is formed in
the flotation feed sump by diluting the coal slurry with filtrate from
the flotation process clean coal filter.  The float-fraction coal concen-
trate, at about 20% solids, is dewatered on rotary vacuum filters and
conveyed to the clean coal stockpile.  Froth flotation tailings flow to
a thickener whose underflow is dewatered on a rotary vacuum filter and
discarded as refuse.  Filtrate is returned to the thickener for further
settling to control slime.  Thickener overflow flows to a clarified
water pond from which water is returned for reuse in the plant.
         Process II.   This process, shown in Figure 2, uses dense-
medium cyclones operated at a relatively low specific gravity for the
production of a limited overflow fraction of highly cleaned coal.  The
bottoms from the low-gravity cyclones are pumped to dense-medium cyclones
operated at a high specific gravity for the production of "middling"
(medium-quality) coal and refuse.  The fine coal is recovered by froth
flotation.

     The raw coal is reduced by screening and crushing to two size
fractions.  The 3/4-inch x 28-mesh intermediate-sized coal constitutes
81% of the raw coal feed while the 28-mesh x 0 fine coal is 19% of the
raw coal feed.  The intermediate-sized coal is slurried with water and
fed to the low-gravity dense-medium cyclones operated at a specific
gravity of 1.34.  About 49% of the intermediate-sized coal fed to the
cyclones is taken off as an overflow.  It is drained on sieve bends and
vibrating screens, washed with water, again drained on the vibrating
screens, centrifuged, and conveyed to the clean coal stockpile.  Under-
flow from the cyclones is drained on sieve bends and vibrating^ screens
for further processing.

     The drained underflow from the low-gravity cyclones is slurried and
fed to high-gravity dense-medium cyclones operating at a 1.55 specific
gravity.  Overflow from the high-gravity cyclones is drained on sieve
bends and vibrating screens, washed with water on the vibrating screens,
and dewatered in basket centrifuges.  This overflow, a middling-quality
                                    142

-------
 COAL RECEIVING
  AND STORAGE
   RAW COAL
     SIZING
          2IMXO
  LOW-GRAVITY
   CLEANING
  HIGH-GRAVITY
   CLEANING
   FINE COAL
   CLEANING
   REFUSE
   DISPOSAL
                                                                           MIDDLING CO»L
                                                                            SHIPMENT
Figure  2 ,   Flow diagram  for  physical  coal-cleaning  process II.
                                         143

-------
coal product, is conveyed to a separate middling coal stockpile.   The
underflow from the high-gravity cyclones is drained, rinsed, dewatered,
and discarded as refuse.

     The fine coal fraction is pulped in the froth flotation feed sump
to a pulp density of 10% solids and pumped to the froth flotation cells.
The flotation overflow, consisting of relatively clean coal, is filtered
on rotary vacuum filters.  Since this coal stream is of lower quality
than the highly cleaned coal from the low-gravity cyclones, the flotation
product is added to the middling coal stockpile with the coal from the
high-gravity cyclones.  Flotation underflow tailings are pumped to a
thickener whose underflow is filtered on rotary vacuum filters.  To
control slimes the refuse filtrate is returned to the thickener for
additional settling.
         Process III.   This process uses dense-medium cyclones to clean
67% of the coal feed as a 1-1/2-inch x 8-mesh coarse coal fraction.   A
fine coal 8-mesh x 200-mesh fraction, amounting to 31% of the coal feed,
is cleaned on concentrating tables.   The remaining 200-mesh x 0 fine
coal fraction is thickened and filtered without cleaning and added to
the clean coal product.  The flowsheet for this process is shown in
Figure 3.

     The coarse coal fraction is pulped and cleaned in 1.55 specific-
gravity dense-medium cyclones, followed by conventional drain-and-rinse
screening and mechanical dewatering of the underflow refuse product and
the clean coal overflow, which is sent to the clean coal stockpile.  The
fine coal consists of a major stream of 8-mesh x 200-mesh size and a
minor stream of 200-mesh x 0 size.   The major stream is cleaned with
concentrating tables at a water-to-coal ratio of 3:2 in the coal feed.
Dressing water is also added along the top edge of the tables to provide
stable flow across its deck.  Including dressing water, the total water-
to-coal ratio is 2:1.  The clean coal from the tables is partially
dewatered on sieve bends followed by final dewatering in basket centrifuges.
The sieve bend filtrates are added,  along with the dilute 200-mesh x 0
slurry, to a thickener.  The thickener underflow is filtered on a
rotary vacuum filter and added to the clean coal product.

Chemical Coal Cleaning

     KVB Process.    This process is the result of several years of
research in chemical desulfurization of fuels by KVB, Incorporated,  a
Research-Cottrell Company.  According to KVB, the process removes 90 to
99% of the pyritic sulfur and up to 40% of the organic sulfur in coal.
The process was patented in September 1975 and has been demonstrated in
bench-scale equipment.  The process, shown in Figures 4 and 5, consists
of a selective oxidation of the sulfur compounds in the coal using
gaseous N02 in the presence of oxygen at a low temperature and atmos-
pheric pressure.
                                   144

-------
 R.O.M. COAL
 TREATMENT
RAW COAL
  SIZING
COARSE COAL
 CLEANING
 FINE COAL
 CLEANING
  REFUSE
 DISPOSAL
CLEAN COAL
 STORAGE
                                  COAL
                                  REFUSE
                                  DENSE MEDIUM
                             	DILUTE MEDIUM; WATER
CLEAN COAL
 SHIPMENT
   Figure  3.   Flow  diagram for physical coal-cleaning  process  III.
                                              145

-------
              «l *    '^-

HOPKRS, FEEOER5 » CONVEYOR!
                         AKER
                                      io     lol

                                     I  20%  II  10%  I
                                     DUE SLURRY  LIME SLURRY
                                       TANK  j  I MIX TANK |



                                            .  t^
                      L-tr
                            „' g
 SCRUBBER
  SLURR'
NEUTRALIZATION
   TANK
IER   \
I*    \
          THICKENER
   v     T
                                                                                       TT
                                                                                        °i
                                                                                             ?^ra
                                                                                              NEUTRALIZER
                                                                                               STAGE 2
                                       S.
                                    NEUTRALIZED
                                    I STAOE
                                      I?-2-!
                                    f", I       I
                                       NEUTRALIZER
                                       I  STAGE 4
                                       I	k.
                  EFFLUENT
                   TANK
                                                                             ^^
              UT
                                    Figure  ^ .   KVB  coal desulfurization process.

-------
                                                                                 TO POWER PLANT
Figure  5.   KVB coal desulfurization process  (continued).

-------
     The 3-inch x 0 coal from the raw coal stockpile is crushed to 1/4-
inch x 0 size and fed to fluidized-bed reactors.  Hot oxidizing gas,
containing 5% N02, 2.5% 02, N2, H20, and a trace of S02, is circulated
through the coal in the reactor and oxidizes the sulfur to sulfates and
S02 gas.  The reactions occur at 200°F and atmospheric pressure.  The
reactions are exothermic but do not provide sufficient heat to maintain
the reaction temperature.  The oxidation is carried out at a low oxygen
concentration so that the reaction effluent gas is very low in N02 and
02 and high in NO, which is reduced to N2 in the flare stack.

     The 28-mesh x 0 fine coal fraction is entrained by the oxidizing
gas stream and removed from the reactor off-gas stream by particulate
scrubbers.  The 1/4-inch x 28-mesh coarse coal fraction is removed from
the bottom of the reactor and transferred to coarse coal washing and
leaching trains.

     The S02 is removed from the oxidizing gas stream in venturi scrubbers
by scrubbing with Na2sC>3.  The NaHSC>3 solution formed in the scrubbers
is then treated with slaked lime to regenerate Na^Og and produce a
calcium sulfite sludge.  The sludge is further treated with oxygen in
the neutralizer to produce gypsum.

     The fine coal slurry from the particulate scrubbers is increased to
37% solids in a thickener and leached with 200°F water.  The water-
leached fine coal is then leached with 200°F NaOH solution and again
washed with 200°F water.  Cyclone classifiers, followed by  centrifuges,
dewater the coal to about 10% moisture.  The coarse coal from the
fluidized-bed  reactors is processed by the same method used for the
fine coal washing and leaching except that spiral classifiers are used
instead of tanks and cyclones.  The hot water wash and leaching solutions
are treated in the neutralizer with slaked lime to produce a waste
sludge of gypsum and sodium jarosite.  All of the fine coal and 15% of
the coarse coal is pelletized.  The pelletized coal, containing 5%
moisture, is combined with the unpelletized portion and stored in open-
air stockpiles.

     Potential problems with the KVB process are possible explosion
hazards involved in dry oxidation of pulverized coal and possible
nitrogen uptake in the coal.

     TRW "Gravichem" Process.   The "Gravichem" coal desulfurization
process was developed by TRW, who claim the process will remove 90 to
95% of the pyritic sulfur but none of the organic sulfur.  The process
has been demonstrated in an 8 ton/day plant at a TRW test site.

     The process, shown in Figures 6 and 7, consists of a sink-float
gravity separation, followed by selective oxidation of the sink fraction
with Fe2(SO^)3, followed by acetone leaching.  The leaching and regen-
eration reactions are both exothermic but do not supply enough heat to
maintain the reactor temperature.
                                    148

-------
                                VENT TO
                               ATMOSPHERE
Figure  6 .   TRW-"Gravichem"  coal desulfurization  process.

-------
Figure  7.  TRW-"Gravichem" coal desulfurization process (continued).

-------
     The raw coal is crushed to 14-mesh top size and the crushed coal is
slurried in recycled leach solution containing 7.5% total iron as FeSO^
and Fe^CSO^).}, plus 4% H2SO^.  The slurry is cooled to control the
specific gravity at 1.31 and pumped to cyclones for a sink-float separation.
The cyclone float fraction, which contains about 32% of the total coal and
has a low pyritic concentration, is filtered, washed, and conveyed to the
clean coal stockpile.

     The sink fraction is pumped to reactors operating at 250°F and 35
psig with a 6-hour residence time.  The oxidation reaction produces
FeSO^, I^SO^, and sulfur.  The Fe£(30^)3 solution is regenerated by
sparging with oxygen.  The reacted coal slurry is cooled, filtered, and
washed with water.  The filtered coal is then slurried with acetone,
cooled to 85°F, and filtered.  The acetone leaching removes most, but
not all, of the sulfur in the coal, which is recovered from the stripper
bottoms as a marketable byproduct.  The coal is dried and the acetone
recovered for recycling.  Approximately 80% of this coal is hot briquetted,
combined with the remainder of the dried coal and the float coal product,
and conveyed to the clean coal stockpile.

     The strong leachate bleedstream from the filter and the bottoms
from the stripper are neutralized with .slaked lime and the neutralized
slurry pumped to a settling pond.

     Potential problems with the TRW process include the presence of a
very corrosive dilute sulfuric acid and iron sulfate solution and potential
environmental problems associated with the disposal of the gypsum —
iron hydroxide sludge.

     Kennecott Process.   Kennecott Copper Corporation began development
of this process in 1970.  Development continued through May 1975, during
which time the process was demonstrated at a bench-scale level.  The
process, shown in Figure 8, consists of an oxidation system in which a
portion of the sulfur in the coal is oxidized to soluble sulfates by
sparging oxygen through pulverized coal at a high temperature and
pressure.  The reaction is exothermic and provides sufficient heat to
maintain the reaction temperature.  The soluble sulfates are removed by
washing and neutralized with slaked lime to produce a waste sludge of
Fe(OH)2 and gypsum.

     The coal is pulverized to 80% 100 mesh x 0 with crushers and wet ball
mills and the slurry is heated and pumped to agitated reactors.  The
reactors operate at 350°F and 315 psig with a 1-hour hold time.  The
reacted slurry is cooled, thickened to 35% solids, and water washed on
rotary drum filters.

     Eighty percent of the washed coal is pelletized.  The pelletized
coal is then combined with the unpelletized portion and conveyed to the
clean coal stockpile.
                                    151

-------
                                                                                 TO
                                                                                • POWER
                                                                                PLANT
Figure  8.   Kennecott coal desulfurization process.

-------
     The clear liquid  from the thickeners and filters, containing
and H2$04, is pumped to  a  neutralizer where it is treated with slaked
lime.  The neutralized slurry of gypsum and iron hydroxide is pumped to
a settling pond  from which supernate water is returned for use in the
process.

     Potential problems  with the Kennecott process include the presence
of a very corrosive dilute t^SO^-FeSO^ solution, reactor design limita-
tions due to high  operating pressures, and potential environmental
problems associated with the disposal of the gypsum - iron hydroxide
sludge.
RESULTS OF PHYSICAL AND CHEMICAL COAL-CLEANING STUDY

Cleaning Performance

     The cleaning performances of the six coal cleaning processes at the
base-case operating conditions described in the premises are shown in
Table  1.  The  PCC processes are limited to removal of pyritic sulfur and
have a considerably lower sulfur removal efficiency than the CCC processes.
In addition, there is a significant weight reduction from removal of
noncoal minerals (as well as some coal) in the PCC processes.  The CCC
processes remove most of the pyritic sulfur and the KVB and Kennecott
processes remove up to 30 to 40% of organic sulfur.
                      TABLE 1.  CLEANING PERFORMANCE.OF PHYSICAL AND CHEMICAL

                          COAL-CLEANING PROCESSES  5% SULFUR COAL

                                 (Moisture-free basis)
, Chemical coal cle.inini>
Physical coal cleaning

Total sulfur, Z
Pyritic sulfur, %
Ash, Z
Btu/lb
Btu recovery, Z
Weight recovery, %
Total sulfur, Ib/MBtu
Sulfur removed, Z Btu basis
Raw coal
5.00
3.35
16.7
12,000
-

4.17
-
PCC-I
3.67
2.02
10.1
13,000
90.7
84.2
2.84
32
PCC- II
3.51
1.86
9.3
13,100
91.4
84.0
2.68
36
PCC- [II
3.78
2.13
10.6
12,900
90;7 .
84.7 ,
2.93
30
KVB
1.30
0.09
11.4
12,600
99.4
94.7
1.00
76
TRW
Gr.ivlchem
1.86
0.09
13.6
12,300
97.7
95.3
1.50
64
Kennecott
1.80
0.40
15.6
12,000
94.3
102.9
1.50
64
     Sulfur  removal efficiencies for the physical processes range from
30 to'  36% on the  basis of raw and cleaned coal heating values.  Weight
and Btu recoveries  are about 84% and 91% respectively.  There is also an
increase in  cleaned coal heating value and a reduction in ash.  Removal
                                      153

-------
efficiencies of the chemical processes range from 65 to 79% with no
appreciable weight or Btu loss.   There is also less increase in heating
value and less reduction in ash content compared to the physical processes,

     In comparing the economics of these coal-cleaning processes the
removal efficiencies must be considered.  In contrast, FGD system
economics are based on sulfur removal to a constant level and removal
efficiencies do not enter as a cost variable.

Base-Case Economics

     Base-case capital investment and annual revenue requirement break-
downs for the six processes are shown in Appendix B.  A summary of these
data and the removal efficiencies is shown in Table 2.  Costs of a
limestone scrubbing FGD system with pond disposal of sludge are also
included in Table 2 for comparison with the coal-cleaning processes.
All of the cost data are based on processes serving a 2000-MW power
plant burning coal with 5% sulfur, as described in the premises.

     The economic results and conclusions presented here are preliminary,
based on data obtained thus far in the continuing coal-cleaning evalu-
ation.  Further data on combination processes and case variations will
better amplify and define the economics of these processes, perhaps
modifying some details of these results.
    TABLE 2.   PHYSICAL AND CHEMICAL COAL-CLEANING ECONOMIC  DATA SUMMARY

Annual


Process
PCC-I
PCC-II
PCC-III
KVB
TRW
Kennecott
FGD

% sulfur
reduction
32
36
30
79
65
65
85
Capital

$/kW
34
40
39
86
114
141
119
investment
C/lb sulfur
removed/yr
36
40
45
49
77
85
68
revenue

Mills /kWh
2.7
2.9
2.9
9.2
7.4
14.7
5.6
requirements
C/lb sulfur
removed
16
16
18
26
27
49
18

    Basis
      2,000 MW, 5.0% sulfur in coal,  5,500 hr/yr,  9,500 Btu/kWh  heat rate.
      FGD is limestone scrubbing, 25% scrubber redundancy,  with pond
      sludge disposal.  Percent sulfur reduction based on raw and cleaned
      coal heating values.
                                    154

-------
     The three physical processes have capita.1 investments of 34 to 40
$/kW and annual revenue requirements of 2.7 to 2.9 -mills/kWh.  Considering
removal efficiencies the capital investments are .36 to 45 C/lb S removed/yr
and the annual revenue requirements are 16.3 to 18.3 
-------
removal efficiencies are considered, differences between the KVB process
and the TRW process are reduced.  In all cases, however, the chemical
processes remain more expensive to operate than the physical processes.

     For comparison, a limestone scrubbing FGD unit with 85% sulfur
reduction is included in Table 2.  The capital investment for the FGD
unit is much higher than those for the PCC processes and higher than the
capital investment of the KVB process.  Annual revenue requirements for
the FGD system are higher than those of the physical processes although
lower than those of the chemical processes.

     When compared on the basis of cost in terms of sulfur removed,
however, the capital investment of the FGD system is greatly reduced
relative to the coal-cleaning processes.  On this basis the KVB process
with its high removal efficiency and relatively low capital investment
compares favorably with the FGD system in capital investment.  The PCC
processes also remain less costly in capital investment than the FGD
system in terms of cost versus sulfur removed.  The annual revenue
requirements of the FGD system are more than those of the PCC processes
in terms of cost versus sulfur removed but remains lower than the annual
revenue requirements of all the chemical processes.

Effect of Coal Sulfur Content on Economics

     Figure 9 shows the effect of coal sulfur content on the capital
costs of the six coal-cleaning processes and FGD.  Annual revenue require-
ments are shown in Figure 10.  Since the processes remove different
percentages of sulfur from the feed coal, capital and operating costs
per kilowatthour are not comparable on a direct basis.  The cost com-
parisons are shown on the basis of quantity of sulfur removed, which
incorporates removal efficiency.  The three PCC processes and the KVB
process have lower capital costs per pound of sulfur removed per year
than FGD.  The TRW and Kennecott processes have higher capital costs per
pound of sulfur removed per year than FGD.

     The three PCC processes have annual revenue requirements per pound
of sulfur removed similar to FGD except for the 0.7% coal.  All of the
CCC processes have higher annual revenue requirements per pound of
sulfur removed than FGD, with the Kennecott process being the highest.

Other Economic Benefits and Penalties of Using Cleaned Coal

     In evaluating the capital investment and annual revenue require-
ments associated with coal cleaning, it is useful to also assess the
other economic benefits and penalties for users of cleaned coal.  In
addition to the primary benefit that the cleaned coal is lower in pyritic
and, depending on the process, organic sulfur, it is generally also
lower in ash and higher in calorific value, although often higher in
surface moisture.  Combustion of coal with these characteristics has
numerous benefits as well as certain disadvantages to the user.
                                    156

-------
w
>-*

Pi
w
p-l

p
w
>


1
Pi

Pi
p
fc<
hJ
p
en

pq
w

2
H
CO
w
P-i
1000


 900



 800



 700



 600





 500





 400
 300
 200
 100


  90


  80




  70




  60




  50






  40








  30
                                                 Kennecott


                                                 TRW



                                                 FGD
                                                 KtfB
                                                 III



                                                 II
                    I
I
I
I
        012345



                          FEED COAL SULFUR  CONTENT,  %




        Figure  9-   Effect of coal sulfur content  on capital investment.
                                     157

-------
I
Pi
,-J
PQ
O
1/1
H
53
W
W
3
W
>
W
Pi
   400
    300
    200
                                       A  Kennecott
                                       •  TRW
                                       ®  KVB
                                       0  FGD
                                       X  PCC III
                                       O  PCC II
                                          PCC I
100
 90
 80
 70
 60

 50

 40


 30
      20
      10
                                I
                                        I
         01234
                           FEED COAL SULFUR CONTENT, %
        Figure 10.   Effect of coal sulfur content on annual  revenue
                    requirements.
                                     158

-------
The net effect, however, is a credit which may be of sufficient magnitude
to offset some of the increased cost of cleaned coal.  Several of the
significant economic effects of using cleaned coal are discussed below.

     Transportation Costs.   Coal beneficiation, if at the mine, decreases
the cost of coal transportation by increasing the calorific value of the
coal, consequently reducing the quantity of coal necessary to supply a
given heat requirement.

     Pension and Benefit Trust Fund.   Provisions of the 1978 UMW contract
require payment by the mine operator of $1.385 to the UMW Pension and
Benefit Trust Fund for each ton of coal shipped to a consumer.  If the
coal-cleaning plant is at the mine, the cleaned coal will be higher in
calorific value requiring a smaller tonnage to supply a required heat
requirement, thus reducing this payment.

     Pulverization Costs.   PCC, by reducing mineral matter, decreases
coal hardness and facilitates crushing.  The increased calorific value
of clean coal also reduces the quantity of coal to be crushed.  The size
of the clean coal product is considerably smaller than that of raw coal
so that significant pulverization costs, which are already covered in
the coal-cleaning costs, are saved.  Detrimentally, cleaning may con-
tribute additional surface moisture which makes pulverization more
difficult.

     Boiler Capacity.   The higher calorific value of cleaned coal
decreases the possibility that the utility boiler capacity will be
derated because of deteriorating coal quality.  Also, by reducing the
slagging tendency of the coal, coal cleaning can permit the design of
furnaces with higher heat transfer rates and correspondingly smaller
furnace volume.

     Boiler Performances.   Cleaned coal can improve boiler performance
by reducing slagging, fouling, and corrosion problems.  This can
significantly reduce the cost of boiler operation and maintenance and
increase the availability of the generating facility.

     Ash Handling.   Ash handling and disposal costs are decreased since
coal cleaning generally reduces the total amount of ash handled.  Less
sensible heat is lost in the bottom ash because of the lower ash levels.

     FGD Operation.   FGD systems generally have markedly better
operation with low-sulfur coals.  When coal cleaning is followed by FGD
scrubbing, the lower sulfur level of the cleaned coal should give less
FGD system downtime and a better overall utility availability. -

     FGD Capital and Operating Costs.   FGD systems for boilers burning
high-sulfur coal have higher capital costs because of the necessity for
a large absorbent preparation facility, scrubber system, and area for
sludge disposal.  Corresponding savings in operating costs should also
be realized with the low-sulfur cleaned coal.
                                    159

-------
     ESP  Size and Cost.  The  resistivity of  fly ash is a major  factor in
determining the collection  area of the ESP.   Resistivity is  determined
by many factors, including  ash composition and 863 level in  the gas.
With conventional ESP units,  the removal of  fly ash will generally be
more difficult with low sulfur levels of cleaned coal and ESP  costs may
be increased.  Other systems  such as hot side ESP, bag filters, or
pulsed ESP may be less expensive in certain  cases when burning  clean
coal.

     Surface Moisture.   Higher moisture levels in cleaned coal resulting
from the  smaller particle sizes increase transportation costs  and result
in a heat loss when the water is heated and  vaporized during the com-
bustion process.

     It is obvious that additional work is needed to quantify  the
economic  magnitude of these benefits and penalties to the utility that
uses cleaned coal.

Energy Requirements

     Energy usages for the  six processes are shown in Table  3.   The
comparison is made on the basis of total energy input consisting of raw
coal feed and utilities.  In  addition to the electrical, steam, diesel
fuel, or  natural gas energy consumed in the  process, there are  other
energy losses or usages that  are specific for certain systems.   Since
the product coals are generally of finer size than the raw coal feed,
they will have a higher surface moisture.  Additional energy is needed
to vaporize this extra moisture and to heat  the water vapor  to  stack
temperature.  The three PCC processes have a significant Btu loss
because part of the coal is entrapped in the refuse stream and  discarded.
                       TABLE 3.  PHYSICAL AND CHEMICAL COAL-CLEANING

                                ENERGY USAGE AND LOSSES


                                 PCC-I   PCC-II   PCC-III   KVB    TRW   Kennecott

       Total energy input,  1012 Btu/yr  115.6   115.2    115.3     110.5   114.2    125.6

       Energy Lost or Used

       Coal lost or used, % of input      9.3    8.7     9.2        0      0     1.2
       Moisture increase in product coal  0.2    0.5     0.1      0.3     1.8     0.8
       Oxygen uptake in coal                    -       -        -      -     3.9
       Utilities
         Electricity, 7. of  input        0.04   0.08    0.04      0.6     0.5     1.7
         Steam, % of input               0      0       0      4.5     6.2     9.2
         Natural gas, % of  input          000     0.02      0       0
         Diesel fuel, % of  input        0.02   0.02    0.02     	0   	0    	0

           Total                    9.6    9.3     9.4      5.4     8.5     16.8


       Basis
         2,000-MW utility power plant, 5,500 hr/yr operation, 9,500 Btu/kWh design heat rate,
         5% sulfur coal.
                                       160

-------
     The Kennecott process also has a small coal Btu usage because a
portion of the coal chemical structure is altered during the cleaning
process.  This energy aids in holding the reaction temperature at the
desired level, thereby replacing an equivalent amount of energy in the
form of steam.  In addition, the coal product from the Kennecott process
has an oxygen uptake resulting in an additional Btu loss.

     The KVB process utilizes a reaction at atmospheric pressure and at
relatively low temperatures.  As a result, the 5.4% total energy usage
for the KVB process is significantly better than for the other processes.
                                    161

-------
                          GROUND-TO-GROUND FGD STUDY
     In previous years EPA, TVA, and others have investigated the design
and economics of limestone, lime, and magnesia scrubbing processes.
Since these earlier studies, technical and operating information on
these systems has greatly increased.  Many full-scale applications of
the limestone and lime systems are now in operation and the magnesia
process has also been evaluated to a lesser extent in full-scale operation.
This study is a continuation of the earlier design and economic evaluations
but incorporates the most recently available commercial technology.

     The design of the MgO process has been updated to include recent
commercial technology.  Provision for chloride removal has been added in
the scrubbing system.  In the regeneration area process refinements
include substantial use of pneumatic conveyors instead of belts, the use
of a rotary dryer instead of a fluidized-bed dryer, redesigned slaking
and calcining systems, a simplified slurry processing system, and addi-
tional heat-recovery equipment.

     In addition, a special energy requirement assessment of the three
systems is included.  Energy costs are expected to increase more rapidly
than other costs associated with FGD processes.  Increases in energy
costs relative to other FGD costs could radically change the comparative
economics of energy-intensive processes and processes with low energy
requirements.  The energy requirement assessment is a ground-to-ground
study including all requirements for raw material preparation and waste
disposal as well as process energy usage.
PROCESS DESCRIPTIONS

     The FGD systems are assumed to be installed downstream from the
power plant particulate-control units, beginning with a single plenum
which collects all the power plant flue gas.   This plenum supplies four
parallel trains of FGD equipment  which vent  to the stack plenum.
Particulate control to meet NSPS is not included in the FGD costs.  Each
train is equipped with a forced-draft booster fan to compensate for
pressure loss in the FGD system and with indirect-steam flue gas reheating
to 175°F.  A presaturator is included upstream from the scrubber to
reduce the flue gas temperature from 300°F to 127°F and a mist eliminator
is included downstream from the scrubber to reduce the flue gas moisture
to 0.1%.  A single raw materials and feed preparation area and a single
waste disposal system serve all four FGD trains.
                                    162

-------
Limestone Process

     The widely used limestone process, Figure. 11, scrubs the flue gas
with a slurry of finely ground limestone, forming hydrated calcium-
sulfur salts which are discarded as a waste sludge.  Purchased limestone
is crushed and wet ground to form the absorbent feed slurry.  The
relatively low reactivity of limestone requires a stoichiometric ratio
of 1.3 moles of CaCO^ to 1.0 mole of sulfur removed.

     A countercurrent mobile-bed scrubber is used.  A 15% solids slurry
is circulated through the scrubber countercurrent to the flue gas and
through an external loop where makeup absorbent slurry is added and a
purge stream containing CaSO-j-l/ZI^O, CaSO^^E^O, and unreacted lime-
stone is removed.  The purge stream is pumped to an earthen-diked,
clay-lined disposal pond where it settles to a sludge of about 40%
solids.  The supernate is returned to the scrubber system.

Lime Process

     The lime process, Figure 12, is similar to the limestone process
except for details of absorbent preparation and process chemistry.
Pebble lime is slaked, forming a slurry of Ca(OH)2 which is used as
absorbent feed.  Two cases are evaluated, one in which purchased lime is
used and a second in which an onsite calciner, Figure 13, is included to
produce lime from limestone.  The higher reactivity of lime as compared
to limestone permits a stoichiometric ratio of 1.05 moles of Ca(OH)2 to
1.00 mole of SOX removed.  This results in a small reduction in slurry-
handling equipment size and pond capacity compared to the limestone
process.

     A countercurrent mobile-bed absorber is used.  The 15% solids
slurry is circulated, absorbent feed added, and a purge stream removed
in the same manner as in the limestone process.  The purge stream,
consisting of CaS03-l/2H20 and CaSO^-2H20 with a small quantity of
unreacted Ca(OH)2> is pumped to an earthen-diked, clay-lined pond where
it settles to a sludge of about 40% solids.  Supernate is returned to
the scrubber system.

Magnesia Process

     The magnesia process, Figure 14, consists of a wet-scrubbing system
using MgO as the absorbent.  The use of MgO provides a rapid absorption
reaction and a low scaling potential, allowing a wide latitude of design
and operating conditions.  The cost of MgO, however, necessitates a
regeneration process.  In this study the absorber reactants are calcined
to produce MgO and SO^ and the S02 is processed to
     A countercurrent mobile-bed scrubber is used.  The feed is an
Mg(OH)2 slurry prepared by slaking MgO in water.  The presaturator
bottoms are discarded in the ash pit to control residual fly ash and
                                    163

-------
                                                                            STEAM FROM
                                                                            STEAM PLANT
                                                                                                        1
OS
•P-
              HOPPERS, FEEDERS • COKVEVOHS
                                     Figure 11.   Limestone  slurry process.


                                                  Flow diagram.

-------
                                                              STEAM FROM
                                                              STEAM PLANT
             V7  T
C\
Ln
                     I  AIR HEATER  ~|
LIME
3TORAOE
SILO
Y
n







~~^
C)
[ J
                                   Figure  12.   Lime slurry process.

                                           Flow diagram.

-------
CRANE
             HOPPERS. FEEDERS AND CONVEYORS
             HOPPERS, FEEDERS AND CONVEYORS
                                      Figure 13.   Lime calcination process.

-------
ON
                         Figure 14.  Magnesia, slurry-regeneration process.



                                             Flow diagram.

-------
chloride buildup in  the  scrubber system.   A 15% solids purge stream, in
which MgS03'6H20 predominates,  is withdrawn from the scrubber loop for
regeneration.  The slurry  is  first centrifuged to decrease the water
content to about 15%.  The solids are dried and dehydrated in an oil-
fired rotary kiln and  decomposed in an oil-fired fluidized-bed calciner.
The MgO is collected for reuse  and the 10-15% SC>2 off-gas is converted
to r^SQ^ in a single-contact  acid plant.
RESULTS OF GROUND-TO-GROUND  FGD EVALUATION

Economics

     Capital investments  and annual revenue requirements for the processes
evaluated in this  study are  shown in Appendix B and are summarized in
Table 4.  The  ranking  of  the three processes remains the same as in
previous evaluations,  with lime scrubbing slightly lower in capital
investment than  limestone scrubbing and slightly higher in annual
revenue requirements.  The MgO  process remains higher than the lime and
limestone processes  in both  capital investment and annual revenue
requirements.
                TABLE 4.  FGD CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENTS






Annual
Capital investment0
Process
Limestone
Lime calcination
Lime
Magnesia

48
53
45
70
$
,948,000
,859,000
,319,000
,293,000
$/kW
97.90
107,72
90.64
140.59

14
15
14
18
$/yr
,375,300
,531,200
,890,500
,325,000C
Mills /kWh
4.11
4.44
4.25
5.24
revenue requirements3
$/ton
coal burned
9
10
9
12
.58
.35
,93
.22
$/MBtu
heat
0
0
0
0
input
.46
.49
.47
.58
$/ton sulfur
removed
41)
444
425
524
 a.  1980 dollars.
 b.  1979 dollars.
 c.  Includes $3,412,100. sulfuric acid sales credit.
     In  comparison to the results reported in 1976 (G. G. McGlamery,  et
al., Flue Gas  Desulfurization Economics) at the Sixth Flue Gas Desulfuri-
zation Symposium in New Orleans,  the capital investments for all three  *
processes are  greatly increased in this study, particularly for the MgO
process.  This comparison can be seen in Figure 15.  The differences  in
these results  are from premise changes, design factors, and inflation.
The MgO  process capital investment has increased more dramatically than
the lime or  limestone,  primarily because of changes in technology.
Annual revenue requirements show much smaller increases between the
1976 and the 1979 results.
                                      168

-------
Limestone
Lime
Magnesia
Limestone
Lime
Magnesia
                            50
           100
                           Capital Investment, $/kW
                       I
I
I
I
                 150
I
                      1.0      2.0      3.0      4.0      5.0




                     Annual Revenue Requirements, Mills/kWh





                      1976         II  1979
                                   6.0
       Figure  15.   Comparison of 1976 and 1979 evaluations.
                               169

-------
     One interesting result appears when comparing lime scrubbing and
limestone onsite calcination facilities with the purchased lime case.
Data available  indicate  that lime  can be purchased in most situations  at
a price less  than a utility can manufacture  its own  supply.   This, of
course, suggests that the costs of a larger  supplier are less than those
of a small onsite operation.

Energy

     Process  energy requirements are shown in Table  5.   Ground-to-ground
energy requirements for  the same processes are shown in Table 6.   Figures
16 and 17 show  the same  data graphically.  The ground-to-ground energy
results include energy required to mine or produce and transport the raw
materials required as well as the  process energy.  On the basis chosen
raw material  production  energy has little effect on  the overall energy
requirements  of the processes.  Limestone scrubbing  ground-to-ground
energy is only  slightly  higher than the process energy,  illustrating the
relatively low  energy requirements of limestone quarrying and transporta-
tion.   The use  of lime instead of  limestone  scrubbing reduces FGD process
energy slightly but increases ground-to-ground energy requirements
because of the  large heat requirements for calcination.    Interestingly,
onsite lime calcination  consumes more energy than the energy  represented
by commercial lime.  Electricity,  transportaion, and  calciner fuel
requirements  are all higher for the relatively small  onsite calciner.
                          TABLE 5. FGD PROCESS ENERGY REQUIREMENTS





Electricitv
Process
Linestone
Lime calcination
Lime
Magnesia
MBtu/hr
68.
65.
60.
67.
,9
,9
, 4
.2
7
7
6
7
kW
,655
,326
,715
,468

Steam
Reheat, Process, Oil, Coal,
MBtu/hr MBtu/hr MBtu/hr MBtu/hr
70
70.
70.
71
.0
.0 58.9
.0
.3 0.6 117.7
Total equivalent
Heat credit,
MBtu/hr

(3.6)

(11.9)
energy consumotion,
% of input energv
3
4,
3.
5,
.3
.4
.1
.6
      Based on a 500-MU boiler efficiency of 90°< for generation of steam and i gross heat rate of 9,000
      Btu/kWh for generation of electricity.
                      TABLE 6.  GROUND-TO-GROUND FGD ENERGY REQUIREMENTS


                                                                   Total equivalent
                            Steam            Natural                     energy
               Electricity   Reheat, Process,  Oil,    gas,   Coal,   Heat credit,   consumption,3 "/•
     Process	MBtu/hr kH   MBtu/hr MBtu/hr  MBtu/hr  MBtu/hr MBtu/hr   MBtu/hr
Limestone
Lime
calcination
Lime
Magnesia -
masneslte
Magnesia -
seawater
69.

66,
62.

72.

75.
. 1

.2
.8

.0

9
7,700

7,400
7,700

8,900

9,800
70.

70.
70.

71.

71.
0

0
0

3 0.6

3 0.6
8,

7.
4.

119.

121.
,9b

.3b 58.9
. 3b 50.0

,0C

.3C 1.7


(3.6)


(11.9)

(11.9)
3

4
4

5

5
.5

.6
.3

.7

.9
       Based on a 500-MW boiler efficiency of 90% for generation of steam and a gross heat rate of
       9,000 Btu/kWh for generation of electricity. Based on 1.2 Ib S02/MBtu heat input allowable emission.
       All limestone quarry and offsite lime processing plants are assumed to be 100 miles from the Chicago
       area FGD plant.
       Oil energy includes transporting MgO from Gabbs, Nevada, for nagnesite and Port St. Joe, Florida, for
       seawater to a Chicago area FGD plant.


                                        170

-------
         ENERGY USED
HEAT
CREDIT
                        LIMESTONE
                        LIME  + CALCINATION
                    iiiiiiiiiiiiii   MAGNESIA
g_| TOTAL
|/y /[ ELECTRICITY
rTTTTTI OIL
KXVt REHEAT STEAM
     PROCESS STEAM
     HEAT CREDIT
     OTHER
                1       2        3        4       5       6
            FIGURE 16.  PROCESS ENERGY REQUIREMENTS.
                        LIME  + CALCINATION
        \V\\X.\\ \VvSl
          IJIJIIIJIll]l|IJIIIIIIIlTTni   MAGNESIA - MAGNESITE
               .\\V\\]
                      IIIIMIIIHI   MAGNESIA - SEAWATER
               12        34        56
                 ENERGY REQUIREMENTS  -  % OF ENERGY INPUT
          Figure 17.   Ground-to-ground  energy requirements,
                               171

-------
     The ground-to-ground magnesia energy requirements are only slightly
higher than the process energy requirements.  The use of magnesia
produced by treatment of dolomite with seawater results in a slightly
higher energy requirement than the use of magnesia from natural magnesite
deposits.  Both types of magnesia require large amounts of energy to
produce but, because thecmagnesia is regenerated, a relatively small
amount is used and the overall energy requirements are not greatly
affected. Lime, in contrast, requires considerably less energy to produce
but the larger quantities used result in a substantially greater energy
use.
                                   172

-------
                 REEVALUATION OF THE WELLMAN-LORD FGD PROCESS
     The Wellman-Lord FGD system developed by Davy Powergas, Inc.,  has
been used in several industrial and utility applications and has under-
gone some changes since its previous economic evaluation by TVA in  1974.
It is now in operation at U.S. coal-fired utility power plants.  Also,
as a replacement for natural gas, new coal-reduction options are being
developed for processing the concentrated S02 from the Wellman-Lord
unit.  This evaluation incorporates the latest Wellman-Lord scrubbing
technology and two S02 conversion processes, the Allied Chemical coal-
reduction process which produces sulfur, and a conventional H^SO^
plant.

     Design changes in the Wellman-Lord scrubbing process represent both
continuing development and experience on U.S. coal-fired boilers.   An
improved raw material wet storage system is now used.  Chloride and 803
removal and neutralization are provided prior to S02 removal.  The
regeneration system is substantially revised to correspond to current
design practice.  Filters have been added to control ash buildup,  tank
capacity has been increased, and a high-temperature ^£$04 crystalliza-
tion system based on a revised oxidation rate is utilized.  No longer is
an antioxidant used.  Double-effect evaporators replace the previous
single-effect evaporators as a result of current energy conservation
practice.
PROCESS DESCRIPTIONS

Wellman-Lord Process

     The Wellman-Lord process, shown in Figure 18, is a regeneration
wet-scrubbing process using a solution of Na2S03 as the absorbent.
Regeneration of the absorbent produces a concentrated S02 stream which can
be processed to either sulfur or
     A spray-type presaturator using recycled water is used for chloride,
SO-j, and residual fly ash control.  Presaturator bottoms are discarded
in the power plant ash pond.  The absorber consists of a countercurrent-
flow, three-stage valve tray unit with separate recirculation in each
stage.  The scrubber effluent containing the reaction products, which
are primarily NaHSO^ and ^2804, is processed to remove Na2SO^ and
regenerate Na2SO-j and S02-  Sodium losses are made up by addition of
       to the regenerated absorbent.
                                    173

-------
AJHI
                              f     T*FWM^S
                          °UST r~~L_   E»°"-•IIT
                          .LLECTOBlJ-,
                             ^_!         STEAM PLANT
                              SULFATE
                              STORAGE
                               SILO
                       Figure  18.   Wellman-Lord  662  recovery  process,

                                            Flow diagram.

-------
     A portion of the scrubber effluent is processed to remove
by evaporation and selective crystallization in a steam-heated, forced-
circulation evaporator serving all four scrubber trains.  The clear
overflow, enriched in NaHSC^, is returned to the regeneration area.  The
bottoms, consisting of a slurry enriched in ^2864 crystals, are centrifuged
to produce a solid containing about two-thirds Na2SO^ and one-third
Na2Sp3.  The centrate is returned to the regeneration area; the solids
are dried in a steam-heated dryer and conveyed to a storage silo for
sale or discard.  There is a potential market available in the paper
industry for this material.

     The regeneration system consists of two trains of double-effect,
forced-circulation evaporators.  Scrubber effluent, combined with liquid
from the sulfate removal process, is heated and 60% is pumped to the
first-effect evaporators and 40% is pumped to the second-effect evaporators.
The first effect is steam heated; the second effect is heated by combined
first-effect vapor and sulfate crystallizer vapor.  Some ^28203 formed
in the first-effect evaporator is removed by a purge stream.  Evaporator
bottoms, consisting primarily of ^2803 are returned to the absorbent
system.  Evaporator and stripper overhead vapor, containing t^O and
802, is dried and the 802 ^s sent to a processing plant.  S02~bearing
condensate from the second-effect evaporator heater, the condensers, and
the compressor is steam stripped and returned to the absorber system.
     The 862 can ^e Processed to either sulfur or I^SO^ by several
methods.  In this study an Allied Chemical Corporation process for
producing sulfur and a single-contact, single-absorption acid plant to
produce I^SO^ are evaluated.
Allied Chemical Process

     A proprietary process, shown in Figure 19, developed by Allied
Chemical Corporation reduces the 802 to sulfur using coal.  Powdered
coal is injected into a reactor containing a bed of inert material and
the bed is fluidized with heated 802 anc* a-'-r •  ^ir ^s added because some
additional oxidation of coal is needed to maintain reaction temperature.
About 70% of the 802 ^s reduced to sulfur which is condensed from the
off-gas.  The remaining off-gas containing S02 and H2S is passed through
a Glaus-type catalytic converter to recover additional sulfur.   The
Glaus converter off-gas is oxidized and recycled to the 802 absorber.

Sulfuric Acid Plant
     For the t^SO^ alternative the 802 is converted to I^SO^ in a
single-contact, single-absorption acid plant.  A single-contact plant is
used for economy and the tail gas containing unreacted 802 is returned
to the scrubber.  A flowsheet of the acid plant is shown in Figure 20.
                                    175

-------
Figure 19.  Allied Chemical coal/S02 reduction technology.




                       Flow diagram.

-------
                                       FUEL
                                       OIL
                             -fef
                 CONDENSATE
                   FROM
                 FLASH DRUM
.EXCHANGER
SOj RICH GAS STREAM    I
     FROM     	-1
WELLMAN-LORD UNIT
                            TO
                         NO. 3 AND 4
                         EXCHANGERS
                           •—g—J
                                       NO. 4
                                     EXCHANGER
                                                    START-UP
                                                     HEATER
                                                      fe!
                   NO. I
                   EXCHANGER
                                                                          STEAM TO
                                                                        WELLMAN - LORD
                                                                           PLANT
                                                                                        NO. 2 AND 3
                                                                                        EXCHANGERS
                                                             CONDENSATE
                                                            - FROM NO. t
                                                             EXCHANGER
                                                                                                   TO NO 5 EXCHANGER"
                                                                                                                        CONDENSATE
                                                                                                                        FLASH DRUM
                                                                                                                           PRODUCT SULFURIC ACID


                                                                                                                           TO WELLMAN-LORD ABSORBER
                                                Figure  20.     Sulfuric  acid  plant.

                                                                Flow  diagram.

-------
RESULTS

      The bottom line economic results of the  Wellman-Lord process are
given in Tables 7 and 8  along with  those for  the citrate (from R.  L.
Torstrick et al.) and magnesia processes which have been repeated for
easy  comparison.   The summary tables of investment and  annual revenue
requirements for the Wellman-Lord scrubbing system with both acid and
sulfur production options  are presented in Appendix B along with those
for other processes.
                  TABLE 7.  SUMMARY OF RECOVERY FGD INVESTMENT REQUIREMENTS

                                    1979 DOLLARS

                                            Total capital investment
                  	Process  	$	$/kW
                  Magnesia
                  Wellman Lord -
                  Wellman Lord - Allied Chemical
                  Citrate
70,293,000   140.59
71,448,000   142.90
74,190,000   148.38
74,918,000   149.84
                TABLE 8.  SUMMARY OF RECOVERY FGD ANNUAL REVENUE REQUIREMENTS

                                   1980 DOLLARS


Process
Magnesia
Wellman Lord - H2S04
Wellman Lord -
Allied Chemical
Citrate
Gross average
annual revenue
requirements
21,025,100
21,752,800
23,151,900

24,820,800
Net average
annual revenue
requirements
18,325,000
19,058,800
21,478,400

23,298,000


Mills /kWh
5.24
5.44
6.14

6.58

$/ton
coal burned
12.22
12.71
14.32

15.35

$/MBtu
heat input
0.58
0.61
0.68

0.73
$/ton
sulfur
removed
524
545
614

666
                                         178

-------
     Producing sulfuric acid by the Wellman-Lord system requires about
4% less capital investment and 15% less annual revenue than the Allied
sulfur production option.  When the Wellman-Lord acid option is compared
to the magnesia process, it requires about 2% more capital investment
and 4% more annual revenue.  Comparing sulfur options, the Wellman-
Lord - Allied system requires about 1% less capital, investment and 6%
less annual revenue than the citrate process.  It should be stated,
however, that the costs derived for this paper are preliminary and the
accuracy of the estimates at this stage does not justify firm conclusions
as to which is the least expensive process.  The major point to be
derived is that the cost difference between magnesia and Wellman-Lord
scrubbing has narrowed to a point where the two processes are very
competitive.  In making a choice between the two processes, reliability,
flexibility, experience, and site-specific factors will have important
influences on the decision, as well as more than comparative economics.

     Energy requirements of the Wellman-Lord process are shown in Table
9 and Figure 21 along with the citrate and magnesia processes.  The
magnesia process has the lowest energy consumption and the citrate
process has the largest.  The energy needs of the magnesia process are
about 13% less than the Wellman-Lord system when producing sulfuric
acid.  As would be expected, the manufacture of sulfuric acid by either
magnesia scrubbing or Wellman-Lord is less energy intensive than the two
sulfur production options.
                                    179

-------
                                                              TABLE 9.  FGD PROCESS ENERGY REQUIREMENTS
CO
O
Steam
Electricity
Process
Magnesia
Welltnan Lord -
H2S04
Wellman Lord -
Allied Chemical
Citrate
MBtu/hr
67.2

64.4

62.0
33.4
kW
7,468

7,269

6,890
8,789
Reheat ,
MBtu/hr
71.3

61.2

61.2
69.9
Process, Oil, Coal, Natural gasj
MBtu/hr MBtu/hr MBtu/hr MBtu/hr
0.6 117.7

156.4 - -

164.7 12.0 10.0
76.9 - - 150.0
Heat credit,
MBtu/hr
(11.9)

(6.9)

(3.1)

Total equivalent
energy consumption,3
% of input energy
5.6

6.5

7.4
7.7
Based on a 500-MW boiler efficiency of 90% for generation of steam and a gross heat rate of 9,000 Btu/kWh for generation
of electricity.

-------
ENERGY USED
                     WELLMAN  LORD -  ALLIED CHEMICAL
                     I
         I
_L
                     345
                     % OF ENERGY  INPUT
               PROCESS ENERGY  REQUIREMENTS
     TOTAL
     ELECTRICITY

     OIL
KNNSj  REHEAT STEAM

I     |  PROCESS STEAM

       HEAT  CREDIT
            OTHER
Figure 21.  Process energy requirement  for  recovery processes
                          181

-------
                                   EPILOGUE
     In this paper, the key results from three on-going economic
evaluation studies have been given.  They are but a fraction of the
results that will be presented in the final reports which will be issued
at the end of the projects.  The complete reports will include many case
variations and sensitivities which will more fully describe the potential
of each coal-cleaning and FGD process.   Because all three projects are
still underway, some results presented  in this paper may be further
refined but no major adjustments should be necessary.
                                   182

-------
                                  APPENDIX A

                                   PREMISES



COAL-CLEANING DESIGN AND ECONOMIC PREMISES

     The design and economic premises for coal-cleaning plants follow in
most respects the assumptions and procedures developed for FGD premises
described in the following section.  There are, however, some differences
which must be recognized in making direct comparisons.  These differences
are described below.

Design Premises

     Power Plant.   The base-case conditions for coal-cleaning evaluations
are a new 2000-MW midwestern power plant with a design heat rate of 9500
Btu/kWh operating at full capacity for 5500 hr/yr.  The power plant life
is assumed to be 30 years.

     Coal Compositions.   The full study uses coals containing 0.7, 2.0,
3.5, and 5.0% sulfur representing compositions that are typical of
published information for over 350 coals with sulfur contents close to
these levels.  The 5.0% sulfur coal composition used in the evaluation
discussed here is shown below.

                  Coal composition           Wt %
               (5.0% sulfur, dry basis)   as received

                  Total sulfur               4.82
                  Pyritic sulfur             3.23
                  Sulfur as sulfate          0.06
                  Organic sulfur             1.53
                  Ash                       16.1
                  Water*                     3.5

              *Air-dried moisture.  Appropriate surface
               moistures are added depending on coal sizes.

     Coal-Cleaning Plant.   The coal-cleaning plants are assumed to be
located at the power plant and are sized to supply the power plant coal
demand.  The PCC plants are based on a 90% Btu recovery and 6000 hr/yr
of operation.  The CCC plants are based on conversion and loss data
supplied by the developers and 8000 hr/yr of operation.  The cleaning
plants have a 15-day raw coal and a 15-day clean coal storage based on
power plant usage.  The location, design, and size premises of waste
                                    183

-------
ponds where required  are  identical to the FGD pond premises.  PCC plants
will have landfill  disposal of solid wastes with mechanical compaction
and an earth  cover.   The  disposal site is located 1 mile from the coal
preparation site.

Economic Premises

     Other than  an  advanced project schedule and revised cost indexes
the economic  premises used for coal cleaning are the same as those  used
for FGD, as described in  the following section.  Costs are, of course,
based on the  coal-cleaning plant size and operating schedule.

     Project  Schedule.    The coal-cleaning projects are assumed to  begin
in mid-1979 and  end in mid-1982 with an average capital investment  cost
basis of the  end of 1980.  Annual revenue requirements are based on end
of 1982 costs.

     Direct Investment.    Chemical Engineering cost indexes through 1977
and TVA projections of these indexes through 1983 are used to determine
direct investments.   The  cost indexes and projections are shown below.

FGD Comparative  Case

     The FGD  system used  for comparison with the coal-cleaning processes
is limestone  scrubbing with 25% scrubber redundancy, 85% SOX removal, and
pond sludge disposal  at the power plant and coal conditions used in the
coal-cleaning premises.  Capital and operating costs are also based on
the coal-cleaning premises as described above.
 FLUE  GAS  DESULFURIZATION DESIGN AND ECONOMIC PREMISES

      The  premises used for the ground-to-ground economic study  and  the
 Wellman-Lord reevaluation are discussed on the following page.
Year
Plant
Material13
Laborc
1974
165.4
171.2
163.3
1975
182.4
194.7
168.6
1976
192.1
205.8
174.2
1977
204.1
220.9
178.2
1978a
221.4
240.8
194.2
1979a
240.2
262.5
209.7
1980a
259.4
286.1
226.5
1981a
278.9
309.0
244.6
1982a
299.8
333.7
264.2
1983a
322.3
360.4
285.3
       a.  TVA projections.
       b.  Same as index in Chemical Engineering for "equipment, machinery, supports."
       c.  Same as index in Chemical Engineering for "construction labor."
                                      184

-------
 Design

     Base Case^.   The base case for conceptual design and preliminary
cost estimating of FGD systems is a new 500-MW Midwestern power unit
with a heat rate of 9000 Btu/kWh.  The unit burns 3.5% sulfur coal (dry
basis) with an as-fired heating value (HHV) of 10,500 Btu/lb and an ash
content of 16%.  The as-fired coal composition and flow rate for the
base case design is shown below.

                   Coal composition         Wt %,
               (3.5% sulfur, dry basis)   as fired    Lb/hr

               Carbon                      57.56     246,800
               Hydrogen                     4.14      17,700
               Nitrogen                     1.29       5,500
               Oxygen                       7.00      30,000
               Sulfur                       3.12      13,400
               Chloride                     0.15         600
               Ash                         16,00      68,600
               Water                       ,10.74      46,000

                    Total                 100.00     428,600
     Operating Life.   The projected operating life of a new coal-fueled
power unit is assumed to be 30 years representing a total of 127,500
hours of operation during the life of the plant.  Operation during the
first year is assumed to be 7000 hours.

     Flue Gas Composition.   Flue gas composition is based on the com-
bustion of pulverized coal assuming a total air rate of the air preheater
equivalent to 133% of the stoichiometric requirement.  This includes 20%
excess air to the boiler and 13% air inleakage at the air preheater.  A
horizontal, frontal-fired, coal-burning unit is assumed.  It is assumed
that 80% of the ash present in the coal is emitted as fly ash and 95% of
the sulfur in- the coal is emitted as SOX.  One percent of the sulfur
emitted as SOX is assumed to be 863 and the remainder S02-  Flue gas
rate and composition is tabulated below.
                                    185

-------
                     Flue Gas  Composition and Properties
Component
N2
02
C02
S02
S03
NOX
HC1
H20



Vol, %
73.76
4.83
12.31
0.24
0.0024
0.06
0.01
8.79



Lb/hr
3,450,000
258,200
904,200
25,130
317
3,009
661
264,500
4,906,000
1,543,000
(approx)









Ib/hr (approx)
aft3/min at 300°F

            Fly ash loading,  gr/sft3  (60°F)  dry basis   6.65
            Fly ash loading,  gr/sft3  (60°F)  wet basis   6.06
     Degree of S02 Removal.   For the processes presented here, SC^
removal is based on meeting the current S02 emission regulation of 1.2
Ib SC>2 allowable emission/MBtu (M = one million) heat input.

     Redundancy.   No special redundancy is provided except spare pumps.
The design does not include a bypass around either the ESP or the FGD
units.

     Reheat.   Indirect steam reheat is used for all cases.  Entrainment
is estimated as 0.1% of the wet gas flow rate at the scrubber outlet for
calculating the steam required for reheat.

     Waste and Byproduct Management.   An onsite disposal pond lined
with impervious clay is used to contain the sulfite sludge from the
limestone and lime processes.  The pond is assumed to be located one
mile from the scrubbing site.  Thirty-day storage of byproduct sulfuric
acid or sulfur is provided in the other processes.

     Project Schedule.   Projects are assumed to begin in mid-1977 and
end in mid-1980, with an average capital investment cost basis of mid-
1979.  Direct investments are prepared using the average annual
Chemical Engineering cost indexes and the TVA projections shown below.
Although actual cost indexes are available for 1976-1978, TVA continues
to use its projections for these years so that consistency with past
estimates is maintained.
                                    186

-------
                         FGD Cost Indexes and Projections
Year
Plant
Material11
Labor0
1973
,144.1
141.9
157.9
1974
165.4
171.2
163.3
1975
182,4
194.7
168.6
1976a
197.9 '
210.3
183.8
1977a
214.7
227.1
200.3
1978a
232.9
245.3
218.3
1979a
251.5
264.9
237.1
1980a
271.6
286.1
259.3
1981a
293.3
309.0
282.6
        a. Projections.
        b. Same as index in Chemical Engineering for "equipment, machinery, supports."
        c. Same as index in Chemical Engineering for "construction labor."
     Mrect Investment Basis.    Direct  costs consist of materials and
labor for equipment and  installation,  services and utilities, and pond
construction.  Services,  utilities,  and miscellaneous costs are estimated
as 6% of the process  areas  subtotal.   This covers such items as mainte-
nance shops, stores,  communications,  railroad, and fire and service
water facilities.

     Indirect Inve&tment  Basis.    Indirect costs consist of in-house
engineering design and supervision,  architect and engineering contractor
expenses, contractor  fees,  and construction expenses.  Construction
facilities are considered a part of  construction expenses.  Consultant
fees are not included.   The engineering design and supervision, and the
contingency factors are  based on demonstration-level technology and
experience.  Indirect investment costs  are estimated from the number of
drawings required, man-hours of supervision and construction, and other
factors related  to the complexity of the process.

     Allowances.   Allowances are included for startup and modification,
interest during  construction,  and working capital.  Startup and modi-
fication allowances are  estimated as 10% of total fixed investment for
the recovery processes and  10% of the total fixed investment minus pond
construction cost for the processes  requiring a waste pond.  Interest
during construction is estimated as  12% of the subtotal fixed investment
for each process.  This  factor is equivalent to the simple interest
which would be accumulated  at a 10%/yr  rate assuming an incremental
capital structure of  60%  debt, 40% equity, and a 3-year project expendi-
ture schedule as indicated  below.
                          Project Expenditure Schedule	
                                                 	Year
                                                                 Total
         Fraction of  total  expenditure
          as borrowed funds                 0.15   0.30   0.15   0.60
         Simple interest  at 10%/yr
          as percent  of total expenditure
           Year 1 debt                      1.5    1.5    1.5    4.5
           Year 2 debt                       -     3.0    3.0    6.0
           Year 3 debt                      _-	   _-	   1.5    1.5

         Accumulated  interest as  percent    1.5    4.5    6.0   12.0
          of total expenditure
                                     187

-------
     Working Capital.   Working capital consists of the total amount of
money invested in raw materials and supplies carried in stock, finished
products in stock, and semifinished products in. the process of being
manufactured; accounts receivable; cash kept on hand for payment of
operating expenses such as salaries, wages, and raw material purchases;
accounts payable; and taxes payable.  For these premises, working capita]
is defined as the equivalent cost of 3 weeks of raw material costs, 7
weeks of direct costs, and 7 weeks of overhead costs.

Revenue Requirements


     Direct Costs.   Annual revenue requirements are based on 7000 hours
of operation per year.  Process operation schedules are assumed to be
the same as the power plant operating profiles.  Raw material, labor,
and utility costs are projected to 1980.  Maintenance costs are estimated
on the basis of direct investment and are varied for each process accord-
ing to the relative process complexity, and historical experience when
available.

     Indirect Costs.   Following power industry practice, regulated
company economics and the conventional method of considering the overall
life of the power plant are used to establish capital charges.  Straight-
line depreciation of 3.3% is used.

     Following Federal Energy Regulatory Commission (FERC) recommenda-
tions an interim replacements allowance factor is used in estimating
annual revenue requirements to provide for the replacement of short-
lived items.  An average allowance of about 0.35% of the total investment
is normally provided.  However, to provide for the unknown life span of
SOX control facilities, a somewhat larger allowance factor of 0.7 is
used.  An insurance allowance of 0.5% of total depreciable capital
investment is also included in the capital charges based on FERC practice.
Property taxes are estimated as 1.5% of the total depreciable capital
investment.

     Cost of capital and income tax charges of 8.6% are applied to the
unrecovered portion of capital investment, based on the debt-to-equity
ratio of 60:40, bonds at 10% interest, and a 14% return on equity.

     Overheads.   Plant, administrative, and marketing overheads are
costs which vary from company to company.  With consideration of the
various methods used in industry and illustrated in a variety of cost
estimating sources, the following method for estimating overheads is
used.

     Plant overhead is; estimated as 50% of the conversion costs excluding
utilities.  Administrative overhead is estimated as 10% of operating
labor and supervision.  Marketing byproducts is considered in the estima-
tion of overheads as 10% of sales revenue.

     Byproduct Sales.   In estimating average annual revenue require-
ments, credit from sale of byproducts is deducted from the yearly pro-
jection of operating cost to obtain the net effect of the FGD process
on the cost of power.

                                    188

-------
                    TABLE B-l.   FCC PROCESS I

                    TOTAL CAPITAL INVESTMENT

                  (Dense-medium vessel,  dense-
                 medium cyclone, froth flotation)
                    Base case - 5% S coal

                                                Investment,  $


Direct Investment

Coal receiving and storage                        8,841,000
Raw coal sizing                                   1,627,000
Coarse coal cleaning                              1,585,000
Intermediate coal cleaning                        2,249,000
Fine coal cleaning                                2,696,000
Refuse disposal as landfill                       3,058,000
Clean coal storage                                8,261,000

     Total areas                                 28,317,000

Services, utilities, and miscellaneous            1,699,000

     Total direct investment                     30,016,000


Indirect Investment

Engineering design and supervision                2,521,000
Architect and engineering contractor                600,000
Construction expense                              3,572,000
Contractor fees                                   1,009,000

     Total indirect investment                    7,702,000

Contingency                                       5,658,000

     Total fixed investment                      43,376,000


Other Capital Charges

Allowance for startup and modifications           4,337,000
Interest during construction                      6,073,000

     Total depreciable investment                53,786,000

Land                                              3,686,000
Working capital                                   9,946,000

     Total capital investment                    67,418,000


Dollars of total capital per kW of generating
 capacity                                             33.7!
Basis
  Midwest location of coal-cleaning plant with project begin-
   ning mid-1979, ending mid-1982; average basis for cost
   scaling, end-1980; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired
   power plant operating at 9,500 Btu/kWh and 5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal storage
   capacities (power plant basis).
  Working capital provides for 3 weeks.raw coal consumption,
   7 weeks  direct revenue costs (excluding Btu loss), and 7
   weeks  operating overheads.
  Landfill site for refuse disposal located 1 mile from coal
   preparation plant.
                                189

-------
                               TABLE B-2.  PCC PROCESS  I

                              ANNUAL REVENUE  REQUIREMENTS

                               (Dense-medium vessel,  dense-
                             medium cyclone,  froth flotation)
Base case -
5% S coal
Annual
quantity
Unit
cost, $
Total annual
cost, $
Direct Costs

Raw materials
  Coal loss (Btu basis)

     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Process water
    Electricity
    Diesel fuel
  Process material:  magnetite, Grade E
  Maintenance, 6% of direct investment
  Analyses

     Total conversion costs

     Total direct costs
   478,100 tons
31.58/ton
   144,000 man-hr   13.80/man-hr
    45,300 kgal
15,110,000 kWh
   145,000 gal
     2,760 tons
 0.13/kgal
 0.039/kWh
 0.70/gal
93.31/ton
     4,000 man-hr   18.70/man-hr
15,098,000

15,098,000


 1,987,000

     6,000
   589,000
   102,000
   257,000
 1,801,000
    75.000

 4,817,000

19,915,000
Indirect Costs

Capital charges
  Depreciation, interim replacements,
   and insurance at 6% of total
   depreciable investment
  Average cost of capital and taxes
   at 8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and
   supervision
  Administrative, 10% of operating labor
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
                                     3,227,000

                                     5,798,000
                                       993,000
                                       199,000
                                    10,217,000

                                    30,132,000
Byproduct Sales Revenue

None

     Total annual revenue requirements
                                                                              30,132,000
                                                       C/lb
                                       Mills/MJh  sulfur removed
Equivalent unit revenue requirements      2.74
                                                      16.3
Basis
  Midwest coal-cleaning plant location;  time basis for scaling, mid-1982; plant life,
   30 years; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired power plant operating at
   9,500 Btu/kWh and 5,500 hr/yr.
  Total-direct investment, $30,016,000;  total depreciable investment, $53,786,000; and
   total capital investment, $67,418,000.
  Raw coal (moisture-free ):  4,840,000 tons/yr, 5% sulfur, 16.7% ash, 12,000 Btu/lb
   and 4.17 Ib S/MBtu.
  Clean coal (moisture-free):  4.073,000 tons/yr, 3.67% sulfur, 10.09" ash  13 000
   and 2.84 Ib S/MBtu.                                         '           *   '    '

                                          190

-------
                   TABLE B-3.   PCC PROCESS II

                    TOTAL CAPITAL INVESTMENT

           (Low-gravity D.M. cyclone, high-gravity
                D.M. cyclone, froth flotation)
                    Base case - 5% S coal

                                                 Investment, $

Direct Investment

Coal receiving and storage                         8,841,000
Raw coal sizing                                    1,845,000
Low-gravity cleaning                               3,564,000
High-gravity cleaning                              1,782,000
Fine coal cleaning                                 4,706,000
Refuse disposal as landfill                        3,058,000
Clean coal storage                                 6,397,000
Middling coal storage                              4,632,000

     Total areas                                  34,825,000

Services, utilities, and miscellaneous             2,090,000

     Total direct investment                      36,915,000


Indirect Investment

Engineering design and supervision                 3,101,000
Architect and engineering contractor                 738,000
Construction expense                               4,393,000
Contractor fees                                    1,240,000

     Total indirect investment                     9,472,000

Contingency                                        6,958.000

     Total fixed investment                       53,345,000


Other Capital Charges

Allowance for startup and modifications            5,335,000
Interest during construction                       7,468,000

     Total depreciable investment                 66,148,000

Land                                               3,703,000
Working capital                                   10,033,000

     Total capital investment                     79,884,000

Dollars of total capital per kW of generating
 capacity                                              39.94
Basis
  Midwest location of coal-cleaning plant with project begin-
   ning mid-1979, ending mid-1982; average basis for cost
   scaling, end-1980; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired power
   plant operating at 9,500 Btu/kWh and 5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal storage
   capacities (power plant basis).
  Working capital provides for 3 weeks raw coal consumption,  7
   weeks direct revenue costs (excluding Btu loss),  and 7
   weeks operating overheads.
  Landfill site for refuse disposal located 1 mile  from coal
   preparation plant.
                              191

-------
                                   TABLE B-4.    PCC PROCESS II

                                   ANNUAL REVENUE REQUIREMENTS

                             (Low-gravity D.M.  cyclone,  high-gravity
                                   D.M.  cyclone,  froth flotation)
Base
Direct Costs
Raw materials
Coal loss (Btu basis)
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Process water
Electricity
Diesel fuel
Process material, magnetite, Grade E
Maintenance, 6% of direct investment
Analyses
Total conversion costs
Total direct costs
case - 5% S coal
Annual
quantity


458,650 tons


144,000 man-hr

85,800 kgal
27,384,000 kWh
148,000 gal
2,920 tons

4,000 man-hr


Unit
cost, $


31.53/ton


13.80/man-hr

0.13/kgal
0.039/kWh
0.70/gal
93.31/ton

18.70/man-hr


Total annual
cost, $


14,484,000
14,484,000

1,987,000

11,000
1,068,000
104,000
272,000
2,215,000
75,000
5,732,000
20,216,000
Indirect Costs

Capital charges
  Depreciation, interim replacements,
   and "insurance at 67, of total
   depreciable investment
  Average cost of capital and taxes
   at 8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and
   supervision
  Administrative, 10% of operating labor
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
                                  3,969,000

                                  6,870,000
                                    993,000
                                    199,000
                                 12,031,000

                                 32,247,000
Byproduct Sales Revenue

None

     Total annual revenue requirements




Equivalent unit revenue requirements
                 C/lb
Mills/kWh   sulfur removed
   2.93
                 16.3
                                 32,247,000
Basis
  Midwest location of coal-cleaning plant;  time basis for scaling, mid-1982; plant life,
   30 years; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired power plant operating at 9,500
   Btu/kWh and 5,500 hr/yr.
  Total direct investment,  336,915,000;  total depreciable investment, 566,148,000; total
   capital investment, $79,384,000.
  Raw coal (moisture-free):   4,820,000 ton/yr, 5% sulfur, 16.7% ash, 12,000 Btu/lb,
   and 4.17 Ib S/MBtu.
  Clean coal (moisture-free):   4,049,000 ton/yr, 3.51% sulfur, 9.25% ash, 13,100 Btu/lb,
   and 2.o8 Ib S/MBtu.
                                          192

-------
                  TABLE B-5.   FCC PROCESS III

                    TOTAL CAPITAL INVESTMENT

         (Dense-medium cyclone, concentrating table)


                    Base case - 5% S coal

                                                  Investment, $

Direct Investment
Coal receiving and storage                          8,841,000
Raw coal sizing                                     2,438,000
Coarse coal cleaning                                3,912,000
Fine coal cleaning                                  7,850,000
Refuse disposal as landfill                         2,980,000
Clean coal storage                                  8,261,000

     Total areas                                   34,282,000

Services, utilities, and miscellaneous              2,057,000

     Total direct investment                       36,339,000


Indirect Investment

Engineering design and supervision                  3,052,000
Architect and engineering contractor                  727,000
Construction expense                                4,324,000
Contractor fees                                     1,221,000

     Total indirect investment                      9,324,000

Contingency                                         6,849,000

     Total fixed investment                        52,512,000


Other Capital Charges

Allowance for startup and modifications             5,251,000
Interest during construction                        7,352,000

     Total depreciable investment                  65,115,000

Land                                                3,583,000
Working capital                                    10.007,000

     Total capital investment                      78,705,000
                   \
Dollars of total capital per kW of generating
 capacity                                               39.35
Basis
  Midwest location of coal-cleaning plant with project begin-
   ning mid-1979, ending mid-1982; average basis for cost
   scaling, end-1980; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired power
   plant operating at 9,500 Btu/kWh and 5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal storage
   capacities (power plant basis).
  Working capital provides for 3 weeks raw coal consumption,  7
   weeks direct revenue costs (excluding Btu loss),  and 7
   weeks operating overheads.
  Landfill site for refuse disposal located 1 mile from coal
   preparation plant.
                          193

-------
                            TABLE B-6.   PCC PROCESS III

                             ANNUAL REVENUE REQUIREMENTS

                     (Dense-medium cyclone, concentrating table)
                              Base case -  5%  S  coal

                                               Annual
                                          	  quantity
Direct Costs

Raw materials
  Coal loss (Btu basis)

     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Process water
    Electricity
    Diesel fuel
  Process material:  magnetite, Grade E
  Maintenance, 6% of direct investment
  Analyses

     Total conversion costs

     Total direct costs
                     Unit
                    cost, $
471,800 tons
31.58/ton
144,000 man-hr   13.80/man-hr
               Total annual
                  cost, $
14,889,000

14,889,000


 1,987,000
25,600 kgal
13,459,000 kWh
138,000 gal
1,970 tons
4,000 man-hr
0.13 /kgal
n.039/kWh
0.70/gal
93.31/ton
18. 70 /man-hr
3,000
525,000
97,000
184,000
2,180,000
75,000
5,051,000
19,940,000
Indirect Costs

Capital charges
  Depreciation, interim replacements,
   and insurance at 6% of total
   depreciable investment
  Average cost of capital and taxes
   at 8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and
   supervision
  Administrative, 10% of operating labor
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
                                  3,907,000

                                  6,769,000
                                    993,000
                                    199,000
                                 11,868,000

                                 31,808,000
Byproduct Sales Revenue

None

     Total annual revenue requirements




Equivalent unit revenue requirements
                                 31,808,000
                        C/lb
       Mills/kWh   sulfur removed
          2.89
                                                                     18.2
Basis
  Midwest location of coal-cleaning plant; time basis for scaling, mid-1982; plant
   life, 30 years; operating time, 6,000 hr/yr.
  Clean coal production capacity for 2,000-MW coal-fired power plant operating at
   9,500 Btu/kWh and 5,500 hr/yr.
  Total direct investment, $9,324,000; total depreciable investment, $65,115,000;
   total capital investment, $78,705,000.
  Raw coal (moisture-free):  4,855,000 ton/yr, 5% sulfur, 16.7% ash, 12,000 Btu/lb, and
   4.17 Ib S/MBtu.
  Clean coal (moisture-free):  4,111,000 ton/yr, 3.73% sulfur, 10.60% ash, 12,300 Btu/lb,
   and 2.94 Ib S/MBtu.
                                          194

-------
                   TABLE B-7 .   KVB PROCESS

                   TOTAL CAPITAL INVESTMENT
                      Base case -5% S coal

                                                   Investment, $

Direct Investment

Raw material handling and preparation                10,197,600
Sulfur oxidation                                      5,984,700
Reactor off-gas cleaning                             10,889,300
Fine coal leaching                                    7,426,700
Coarse coal leaching                                  6,624,800
Product agglomeration and handling                   11,328,800
Leach solution neutralization and water handling      5,913,200
Settling pond                                        16,203,000

     Subtotal                                        74,568,100

Services, utilities, and miscellaneous                4,474,100

     Total direct investment                         79,042,200


Indirect Investment

Engineering design and supervision                    6,639,900
Architect and engineering contractor                  1,586,300
Construction expense                                  9,407,800
Contractor fees                                       2.658,500

     Total indirect investment                       20,292,500

Contingency                                          19,866,900

     Total fixed investment                         119,201,600


Other Capital Charges

Allowance for startup and modifications              11,920,200
Interest during construction                         16,688,200

     Total depreciable investment                   147,810,000

Land                                                  3,611,000
Working capital                                      19,945,200

     Total capital investment                       171,366,2(30

Dollars of total capital per kW of generating
 capacity                                                  85.7
Basis
  Midwest location of coal-cleaning plant with project begin-
   ning mid-1979, ending mid-1982; average basis for cost
   scaling,  end-1980; operating time, 8,000 hr/yr.
  Clean coal production capacity for 2,000-MW  coal-fired power
   plant operating at 9,500 Btu/kWh and 5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal storage
   capacities (power plant basis).
  Working capital provides for 3 weeks raw coal consumption,
   7 weeks direct revenue costs, and 7 weeks operating
   overheads.
  Pond site  for sludge disposal located 1 mile from"coal
   preparation plant.
                               195

-------
                                TABLE B-8 .   KVB PROCESS

                              ANNUAL REVENUE REQUIREMENTS
                                Base case - 5% S coal
                                          Annual quantity
                     Unit cost, $
                                                                              Total annual
                                                                                cost, $
Direct Costs

Raw materials
  Lime
  Oxygen
  NO?
  Na5H (50%)
  Sodium lignin sulfonate
  Natural gas

     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Steam
    Process water
    Electricity
  Maintenance, 6% of direct investment
  Analysis

     Total conversion costs

     Total direct costs
    197,603 tons
    297,600 tons
        952 tons
    152,880 tons
     75,200 tons
     24,000 kft3
 43.31/ton
 21.13/ton
665.28/ton
 99.57/ton
 83.17/ton
  2.93/kft3
    152,000 man-hr   13.80/raan-hr
  5,349,838 MBtu
  2,663,074 kgal
222,739,157 kWh

     24,000 man-hr
 8,558,200
 6,288,300
   633,300
15,222,300
 6,254,400
    70,300

37,026,800
                   2,097,600
2. 54 /MBtu
0.09/kgal
0.039/kWh
18.70/man-hr

13,588,600
239,700
8,686,800
4,742,500
448,800
29,804,000
66,830,800
Indirect Costs

Capital charges
  Depreciation, interim replacements, and
   insurance at 6% of total depreciable
   investment
  Average cost of capital and taxes at
   8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and supervision
  Administrative, 10% of operating labor and supervision
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
                                       8,868,600

                                      14,73.7,500

                                       1,048,800
                                         209,800


                                      24,864,700

                                      91,695,500
Byproduct Sales Revenue

None

     Total annual revenue requirements
                                                      e/lb
                                       Mills/kWh  sulfur removed
Equivalent unit revenue requirements     8.3
             26.3
                                     91,695,500
Basis
  Midwest coal-cleaning plant location; time basis for scaling, mid-1982; plant life,
   30 years; operating time, 8,000 hr/yr.
  Clean coal production capacity for 2,000-MW  coal-fired power plant operating at
   9,500 Btu/kWh and 5,500 hr/yr.
  Total direct investment, $79,042,200; total depreciable investment, $147,810,000; and
   total capital investment, $171,366,200.
  Raw coal  (moisture-free):  4,578,400 tons/yr, 5.0% sulfur, 16.7% ash,  12,000 Btu/lb
   and 4.2 Ib S/MBtu.
  Clean coal (moisture-free):  4,336,000 tons/yr, 1.3% sulfur,  11.4% ash, 12,600 Btu/lb,
   and 1.0 Ib S/MBtu.
                                               196

-------
              TABLE  B-9\   TRW-"GRAVICHEM"  PROCESS

                    TOTAL  CAPITAL  INVESTMENT
                        Base  case    57,  S  coal

                                                Investment,  $

Direct  Investment

Raw material handling  and  preparation               7,874,600
"Gravichem" separation                             7,915,700
Float coal washing                                  7,201,100
Reactor - regenerator                              20,539,000
Acetone leaching                                   12,553,100
Acetone recovery and coal  drying                   28,202,800
Leach solution  concentration                       3,104,400
Neutralization  and pond water handling              1,792,300
Product agglomeration  and  handling                 12,188,600
Utility water handling                             1,065,200
Settling pond                                       8,219,500

     Subtotal                                    110,656,300

Services, utilities, and miscellaneous              6,639,400

     Total direct investment                    117,295,700


Indirect Investment

Engineering design and supervision                  5,738,500
Architect and engineering  contractor                1,387,900
Construction expense                              13,028,600
Contractor fees                                     3,588,600

     Total indirect investment                    23,743,600

Contingency                                       28,207,900

     Total fixed investment                      169,247,200


Other Capital Charges

Allowance for startup  and modifications           16,924,700
Interest during construction                      23,694,600

     Total depreciable investment                209,866,500

Land                                               1,988,200
Working capital                                   16,194,400

     Total capital investment                    228,049,100

Dollars of total capital per kW of  generating
 capacity                                              114.0
Basis
  Midwest location of coal-cleaning plant with project begin-
   ning mid-1979, ending mid-1982; average basis for cost
   scaling,  end-1980; operating time, 8,000 hr/yr.
  Clean coal production capacity for 2,000-MW  coal-fired
   power plant operating at 9,500 Btu/kWh and 5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal storage
   capacities (power plant basis).
  Working capital provides for 3 weeks raw coal consumption,
   7 weeks direct revenue costs, and 7 weeks operating
   overheads.
  Pond site for sludge disposal located 1 mile from coal
   preparation plant.

-------
                           TABLE B-10.   TRW-"GRAVICHEM" PROCESS

                               ANNUAL REVENUE REQUIREMENTS
Base
Direct Costs
Raw materials
Lime
Oxygen
Acetone
Copperas
Sulfuric acid
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Process water
Electricity
Maintenance, 6% of direct investment
Analysis
Total conversion costs
Total direct costs
case - 5% S coal
Annual quantity


119,200 tons
56,000 tons
2,872 tons
28,000 tons
83,200 tons


160,000 man-hr

6,728,550 MBtu
14,512,600 kgal
182,028,933 kWh

32,000 raan-hr


Unit cost, $


43.31/ton
21.13/ton
471.24/ton
72.07/ton
45.18/ton


13.80/man-hr

2.54/MBtu
0.07/kgal
0.039/kWh

18.70/man-hr


Total annual
cost, $


5,162,600
1,183,500
1,353,400
2,018,000
3,759,000
13,476,500

2,208,000

17,090,500
1,015,900
7,099,100
7,037,700
598,400
35, 049", 600
48,526,100
Indirect Costs

Capital charges
  Depreciation, interim replacement,  and
   insurance at 6% of total depreciable
   investment
  Average cost of capital and taxes at
   8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and supervision
  Administrative, 10% of operating labor and supervision
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
12,592,000

19,612,200

 1,104,000
   220,800
   123,500

33,652,500

82,178,600
Byproduct Sales Revenue

Sulfur                                    23,300 long tons    53.00/long ton

     Total annual revenue requirements

                                                       C/lb
                                       Mills/kWh  sulfur  removed
Equivalent unit revenue requirements      7.4
                                                       27.4
 (1.234.900)

 80,943,700
Basis
  Midwest coal-cleanine plant location; time basis for scaling, mid-1982; plant life, 30
   years; operating time, 8,000 hr/yr.
  Clean coal production capacity for 2,000-MW, coal-fired power plant operating at 9,550
   Btu/kWh and 5,500 hr/yr.  '
  Total direct investment, $117,295,700; total depreciable investment, $209,866,500; and
   total capital investment, $228,049,100.
  Raw coal (moisture-free):   4,578,400 tons/yr, 5.0% sulfur, 16.7% ash, 12,000 Btu/lb,
   and 4.2 Ib S/MBtu.
  Clean coal (mositure-free):  4,364,800 tons/yr, 1.86% sulfur, 13.6% ash, 12,300 Btu/lb,
   and 1.5 Ib S/MBtu,
                                               198

-------
           TABLE B-ll.  KENNECOTT PROCESS

              TOTAL CAPITAL INVESTMENT
                  Base case - 5% S coal

                                           Investment, $

Direct Investment

Raw materials handling and preparation       13,856,900
Reactor area                                 48,820,200
Coal filtration area                         24,489,000
Product agglomeration and handling           28,343,900
Neutralization and water handling             6,151,800
Settling pond                                13,961,900

     Subtotal                               135,623,700

Services, utilities, and miscellaneous        8.137,400

     Total direct investment                143,761,100


Indirect Investment

Engineering design and supervision            3,777,600
Architect and engineering contractor            877,700
Construction expense                         15,321,300
Contractor fees                               4,188,700

     Total indirect investment               24,165,300

Contingency                                  33,585,300

     Total fixed investment                 201,511,700


Other Capital Charges

Allowance for startup and modifications      20,151,200
Interest during construction                 28,211,600

     Total depreciable investment           249,874,500

Land                                          3,152,600
Working capital                              29,188,800

     Total capital investment               281,215,900

Dollars of total capital per kW of
 generating capacity                              140.6
Basis
  Midwest location of coal-cleaning plant with project
   beginning mid-1979, ending mid-1982; average basis
   for cost scaling, end-1980; operating time 8,000
   hr/yr.
  Clean coal production capacity for 2,000-MW, coal-
   fired power plant operating at 9,500 Btu/kWh and
   5,500 hr/yr.
  Fifteen-day raw coal and fifteen-day clean coal
   storage capacities (power plant basis).
  Working capital provides for 3 weeks raw coal con-
   sumption, 7 weeks direct revenue costs, and 7 weeks
   operating overheads.
  Pond site for sludge disposal located 1 mile from
   coal preparation plant.
                       199

-------
                             TABLE B-12 .   KENNECOTT PROCESS

                              ANNUAL REVENUE REQUIREMENTS
Base
Direct Costs
Raw materials
Lime
Oxygen
Sodium lignin sulfonate
Total raw materials cost
Conversion costs
Operating labor and supervision
Process Btu loss
Steam
Process water
Electricity
Maintenance, 6% of direct investment
Analysis
Total conversion costs
Total direct costs
case 5% S coal
Annual quantity
290,480 tons
1,034,400 tons
171,200 tons
168,000 man-hr
2,005,900 MBtu
12,458,440 MBtu
8,741,630 kgal
660,230,321 kWh
32,000 man-hr

Unit cost, $
43.31/ton
21.13/ton
83.17/ton
13. 80 /man-hr
1.36 /MBtu
2. 54 /MBtu
0.07/kgal
0.039/kWh
18. 70 /man-hr

Total annual
cost, $
12,580,700
21,856,900
14,238,700
48,676,300
2,318,400
2,728,000
31,644,400
611,900
25,749,000
8,625,700
598,400
72,275,800
120,952,100
Indirect Costs

Capital charges
  Depreciation, interim replacement,  and
   insurance at 6% of total depreciable
   inves tment
  Average cost of capital and taxes at
   8.6% of total capital investment
Overheads
  Plant, 50% of operating labor and supervision
  Administrative, 10% of operating labor and supervision
  Marketing, 10% of sales revenue

     Total indirect costs

     Gross annual revenue requirements
                         14,992,500

                         24,184,600

                          1,159,200
                            231,800


                         40,568,100

                        161,520,200
Byproduct Sales Revenue

None

     Total annual revenue requirements
                                       Mills/kWh
    C/lb
sulfur removed
Equivalent unit revenue requirements      14.7
    48.9
                        161,520,200
Basis
  Midwest coal-cleaning plant location; time basis for scaling, mid-1982; plant life,
   30 years; operating time, 8,000 hr/yr.
  Clean coal production capacity for 2,000-MW  coal-fired power plant operating at
   9,500 Btu/kWh and 5,500 hr/yr.
  Total direct investment, $143,761,100; total depreciable investment, $249,874,500; and
   total capital investment, $281,215,900.
  Raw coal (moisture-free):  5,249,600 tons/yr, 5.0% sulfur, 16.7% ash, 12,000 Btu/lb
   and 4.2 lb S/MBtu.                                                                 '
  Clean coal (moisture-free):  5,402,400 tons/yr, 1.8% sulfur, 13.6% ash, 12,300 Btu/lb
   and 1.5 lb S/MBtu.                                                                   '
                                              200

-------
                                TABLE B-13.  LIMESTONE SLURRY PROCESS

                               SUMMARY OF ESTIMATED CAPITAL INVESTMENT

                          (500-MW new coal-fired power unit, 3.5% S in coal;
                 1.2 Ib S02/MBtu heat input allowable emission; onsite, solids disposal)

Direct Investment
Materials handling (hoppers, feeders, conveyors, elevators, bins,
shaker and puller)
Feed preparation (feeders, crushers, ball mills, hoist,
tanks, agitators, and pumps)
Gas handling (common feed plenum and booster fans, uas ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack)
S02 absorption (four TCA scrubbers including presaturators
and entrainment separators, recirculation tanks, agitators,
and pumps)
Stack gas reheat (four indirect steam reheaters)
Solids disposal (onsite disposal facilities including feed
tank, agitator, slurry disposal pumps, and pond water return
pumps)
Subtotal
Services, utilities, and miscellaneous
Total process areas excluding pond construction
Pond construction
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction "expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment
Investment, $
1,863,000
1,651,000
4,318,000
8,974,000
1,282,000
1,680,000
19,768,000
1,186,000
20,954,000
5,145,000
26,099,000
1,218,000
270,000
3,630,000
1,145,000
6,263,000
6,473,000
38,835,000
3,369,000
4,660,000
46,864,000
1,030,000
1,054,000
48,948,000
% of
total direct
investment
7.1
6.3
16.6
34.4
4.9
6.4
75.7
4.6
80.3
19.7
100.0
4.7
1.0
13.9
4.4
24.0
24.8
148.8
12.9
17.9
179.6
3.9
4.0
187.5
Basis
  Evaluation represents  project  beginning mid-1977,  ending  mid-1980.   Average  cost  basis  for
   scaling,  mid-1979.
  Stack gas  reheat to  175°F by indirect steam reheat.
  Minimum in-process storage;  only pumps are spared.
  Disposal pond located  1 mile from power plant.
  Investment requirements for  fly ash removal and disposal  excluded;  FGD process investment  estimate
   begins with common  feed plenum downstream of the  ESP.
  Construction labor shortages with accompanying  overtime pay incentive not considered.
                                              201

-------
                                 TABLE B-14.   LIMESTONE SLURRY PROCESS

              SUMMARY OF AVERAGE ANNUAL REVENUE REQUIREMENTS   REGULATED UTILITY ECONOMICS

                           (500-MW new coal-fired power unit, 3.57, S in coal;
                1.2 Ib S02/MBtu heat input allowable emission; onsite solids disposal)

Direct Costs
Raw materials
Limestone
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Process water
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Annual
quantity


159,300 tons


25,990 man-hr

489,800 MBtu
243,400 kgal
53,588,000 kWh


3,760 man-hr


Unit
cost, $


7.00/ton


12.50/man-hr

2.00/MBtu
0.12/kgal
0.029/kWh


17. 00 /man-hr


Total
annual
cost, $


1,115,100
1,115,100

324,900

979,600
29,200
1,554,100

2,040,200
63,900
4,991,900
6,107,000
% of average
annual revenue
requirements


7.76
7.76

2.26

6.81
0.20
10.81

14.19
0.45
34.72
42.48
Indirect Costs

Capital charges
  Depreciation, interim replacements,  and
   insurance at 6.0% of total depreciable
   investment
  Average cost of capital and taxes at 8.6%
   of total capital investment
Overheads
  Plant, 50% of conversion costs less  utilities
  Administrative, 10% of operating labor

     Total indirect costs

     Total average annual revenue requirements
  2,811,800

  4,209,500

  1,214,500
     32,500

  8,268,300

 14,375,300
 19.56

 29.28

  8.45
  0.23

 57.52

100.00
                                                         $/ton coal   $/MBtu heat      $/ton
                                             Mills/kWh	burned	input	S removed
Equivalent unit revenue requirements
                                                4.11
                                                            9.58
0.46
                                                                                        411
Basis
  Midwest plant location,  1980 revenue requirements.
  Remaining life of power  plant,  30 yr.
  Power unit on-stream time,  7,000 hr/yr.
  Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed, 35,000 short tons/yr;  solids disposal 184,200 tons/yr calcium solids including only
   hydrate water.
  Investment and revenue requirement for removal  and  disposal of fly ash excluded.
  Total direct investment,  $26,099,000;  total depreciable investment,  $46,864,000;  and total capital
   investment, $48,948,000.
  All tons shown are 2,000 Ib.
                                               202

-------
                        TABLE B-15.   LIME SLURRY PROCESS WITH CALCINATION

                             SUMMARY OF ESTIMATED CAPITAL INVESTMENT

                       (500-MW new coal-fired power unit, 3.5% S in fuel;
             1.2 Ib S02/MBtu heat input allowable emission;  onsite solids disposal)




Direct Investment
Materials handling (conveyors, elevators, feeder, silo, and
bins)
Lime calcination (feeders, crusher, ball mill, fans, bins,
rotary kiln, waste heat boiler, and elevators)
Feed preparation (feeders, slakers, tanks, agitators, and
pump's)
Gas handling (common feed plenum and booster fans, gas ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack)
S02 absorption (four TCA scrubbers including presaturators and
entrainment separators, recirculation tanks, agitators, and
pumps)
Stack gas reheat (four indirect steam reheaters)
Solids disposal (onsite disposal facilities including feed
tank, agitator, slurry disposal pumps, and pond water return
pumps)
Subtotal
Services, utilities, and miscellaneous
Total process areas excluding pond construction
Pond construction
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment


Investment, $


2,570,000

3,654,000

660,000


4,318,000


8,504,000
1,232,000


1,616,000
22,604,000
1,356,000
23,906,000
4,505,000
28,465,000

1,683,000
389,000
3,944,000
1,223,000
7,239,000
7,141,000
42,845,000

3,834,000
5,142,000
51,821,000
909,000
1,129,000
53,859,000
% of
total direct
investment


9.0

12.8

2.3


15.2


29.9
4.5


5.7
79.4
4.8
84.2
15.8
100.0

5.9
1.4
13.8
4.3
25.4
25.1
150.5

13.5
18.1
182.1
3.2
4.0
189.3
Basis
  Evaluation represents project beginning mid-1977,  ending mid-1980.   Average  cost  basis  for
   scaling,  mid-1979.
  Stack gas  reheat to  175°F by indirect steam reheat.
  Minimum in-process storage;  only pumps are spared.
  Disposal pond located 1  mile from power plant.
  Investment requirements  for  fly ash removal and disposal excluded;  FGD process investment
   estimate  begins with common feed plenum downstream of the ESP.
  Construction labor shortages with accompanying  overtime pay incentive not considered.
                                           203

-------
                         TABLE B-16.   LIME SLURRY PROCESS  WITH CALCINATION

           SUMMARY OF AVERAGE ANNUAL REVENUE REQUIREMENTS    REGULATED UTILITY ECONOMICS

                        (500-MW new coal-fired  power unit,  3.5% S in coal;
              1.2 Ib S02/MBtu heat input allowable emission; onsite solids disposal)
Direct Costs
Raw materials
Limestone
Coal
Annual
quantity
129,400 tons
19,630 tons
Unit
cost, $
7.00/ton
25.00/ton
Total
annual
cost, $
905,800
490,800
7, of average
annual revenue
requirements
5.83
3.16
     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Steam
    Process water
    Electricity
    Heat credit
  Maintenance
    Labor and material
  Analyses

     Total conversion costs

     Total direct costs
37,670 man-hr    12.50/man-hr
                                                                         1,396,600
470,900
                                                                                          8.99
                                                  3.03
488,340 MBtu
235,600 kgal
51,286,000 kWh
25,100 MBtu

4,700 man-hr

2.00/MBtu
0.12/kgal
0.029/kWh
2.00/MBtu

17.00/man-hr

976,700
28,300
1,487,300
(50,200)
2,052,000
79,900
5,044,900
6,441,500
6.29
0.18
9.58
(0.32)
13.21
0.51
32.48
41.47
Indirect Costs

Capital charges
  Depreciation, interim replacements,  and
   insurance at 6.0% of total depreciable
   investment
  Average cost of capital and taxes at 8.6%
   of total capital investment
Overheads
  Plant, 50% of conversion costs less  utilities
  Administrative, 10% of operating labor

     Total indirect costs

     Total average annual revenue requirements
                                 3,109,300

                                 4,631,900

                                 1,301,400
                                    47,100

                                 9,089,700

                                15,531,200
              20.02

              29.83,

               8.38
               0.30

              58.53

             100.00
Equivalent unit revenue requirements
                                                   Mills/kWh
                       $/ton coal
                         burned
                                                     4.44
                                                                  10.35
 $/MBtu heat
    input
  $/ton
S removed
                                        0.49
                 .444
Basis
  Midwest plant location,  1980 revenue requirements.
  Remaining life of power  plant,  30 yr.
  Power unit on-stream time, 7,000 hr/yr.
  Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed, 35,000 short tons/yr;  solids disposal 153,600 tons/yr calcium solids including
   only hydrate water.
  Investment and revenue requirement for removal  and  disposal of fly ash excluded.
  Total direct investment, $28,465,000;  total depreciable investment, $51,821,000; and total
   capital investment, $53,859,000.
  All tons shown are 2,000 Ib.
                                            204

-------
                                TABLE B-17.   LIME SLURRY PROCESS

                             SUMMARY OF ESTIMATED CAPITAL INVESTMENT

                       (500-MW new coal-fired power  unit, 3.5% S  in fuel;
             1.2 Ib S02/MBtu heat input allowable emission;  onsite solids  disposal)
Direct Investment
Materials handling (conveyors, elevators, feeder, silo, and
bins)
Feed preparation (feeders, slakers, tanks, agitators, and
pumps)
Gas handling (common feed plenum and booster fans, gas ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack)
S02 absorption (four TCA scrubbers including presaturators and
entrainment separators, recirculation tanks, agitators, and
pumps)
Stack gas reheat (four indirect steam reheaters)
Solids disposal (onsite disposal facilities including feed
tank, agitator, slurry disposal pumps, and pond water return
pumps)
Subtotal
Services, utilities, and miscellaneous
Total process areas excluding pond construction
Pond construction
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment
Investment, $
1,978,000
660,000
4,318,000
8,504,000
1,282,000
1,616,000
18,358,000
1,101,000
19,459,000
4,505,000
23,964,000
1,095,000
243,000
3,391,000
1,073,000
5,802,000
5,953,000
35,719,000
3,121,000
4,286,000
43,126,000
895,000
1,298,000
45,319,000
% of
total direct
investment
8.3
2.8
18.0
35.5
5.3
6.7
76.6
4.6
81.2
18.8
100.0
4.6
1.0
14.1
4.5
24.2
24.8
149.0
13.0
17.9
179.9
3.7
5.4
189.0
Basis
  Evaluation represents project beginning mid-1977,  ending mid-1980.   Average cost  basis  for
   scaling,  mid-1979.
  Stack gas  reheat to 175°F by indirect steam reheat.
  Minimum in-process storage;  only pumps are spared.
  Disposal pond located 1 mile from power plant.
  Investment requirements for  fly ash removal and disposal excluded;  FGD process  investment
   estimate begins with common feed plenum downstream of  the ESP.
  Construction labor shortages with accompanying  overtime pay incentive not considered.
                                            205

-------
                                 TABLE B-18.   LIME SLURRY PROCESS

           SUMMARY OF AVERAGE ANNUAL REVENUE  REQUIREMENTS   REGULATED UTILITY ECONOMICS

                        (500-MW new coal-fired power  unit,  3.5%  S in coal;
              1.2 Ib S02/MBtu heat  input  allowable emission;  onsite  solids  disposal)
Annual
quantity
Direct Costs
Raw materials
Lime 68,600 tons
Total raw materials cost
Conversion costs
Operating labor and supervision 25,990 man-hr
Utilities
Steam 489,900 MBtu
Process water 232,600 kgal
Electricity 47,008,000 kWh
Maintenance
Labor and material
Analyses 3,760 man-hr
Total conversion costs
Total direct costs
Unit
cost, $


42.00/ton


12. 50 /man-hr

2.00/MBtu
0.12/kgal
0.029/kWh


17.00/man-hr


Total % o^ average
annual annual revenue
cost, $ requirements


2,881,200
2,881,200

324,900

979,600
27,900
1,363,200

1,691,900
63,900
4,451,400
7,332,600


19.35
19.35

2.18

6.58
0.19
9.15

11.36
0.42
29.89
49.24
Indirect Costs
Capital charges
Depreciation, interim replacements, and
insurance at 6.0% of total depreciable
investment
Average cost of capital and taxes at 8.6%
of total capital investment
Overheads
Plant, 50% of conversion costs less utilities
Administrative, 10% of operating labor
Total indirect costs
Total average annual revenue requirements

Mills /kWh
Equivalent unit revenue requirements 4.25



2,587,600

3,897,400

1,040,400
32,500
7,557,900
14,890,500
$/ton coal $/MBtu heat
burned input
9.93 0.47



17.38

26.17

6.99
0.22
50.76
100.00
$/ton
S removed
425

Basis
  Midwest plant location,  1980  revenue  requirements.
  Remaining life of power  plant,  30  yr.
  Power unit on-stream time,  7,000 hr/yr.
  Coal burned,  1,500,100 tons/yr,  9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed,  35,000 short  tons/yr;  solids  disposal  153,600  tons/yr  calcium solids  including
   only hydrate water.
  Investment and revenue requirement for removal  and  disposal  of fly ash excluded.
  Total direct  investment,  $23,964,000;  total depreciable  investment,  $43,126,000;  and  total
   capital investment,  $45,319,000.
  All tons shown are 2,000 Ib.
                                           206

-------
                      TABLE  B-19.  MAGNESIA SLURRY - REGENERATION PROCESS

                             SUMMARY  OF ESTIMATED  CAPITAL  INVESTMENT

                       (500-MW new coal-fired  power unit,  3.5%  S  in  coal;
              1.2 Ib S02/MBtu heat input allowable emission;  6.5  tons/hr  100%  H2SU4)



Direct Investment
Materials handling (conveyors, silos, bins, and feeders)
Feed preparation (mixer, tank, agitator, and pump)
Gas handling (common feed plenum and booster fans, ^us ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack)
S02 absorption (four spray grid scrubbers including entrainment
separators, tanks, agitators, and pumps)
Stack gas reheat (four indirect steam reheaters)
Chloride purge (four chloride scrubbers and entrainment
separators, tanks, agitators, and pumps)
Slurry processing (centrifuges, conveyor, tank, agitator, and
pumps)
Cake drying (dryer, conveyors, silos, fans, tank, and pumps)
Calcination (calciner, preheater, solids cooler, waste heat
boiler, conveyors, silos, fans, and bins)
Acid production (complete contact unit for sulfuric acid
production)
Acid storage (storage and shipping facilities for 30-day
production of sulfuric acid)
Subtotal
Services, utilities, and miscellaneous
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment


Investment, $

1,031,000
447,000


4,318,000

5,874,000
1,407,000

2,766,000

1,499,000
5,747,000

2,281,000

8,340,000

1,265,000
34,975,000
2,099,000
37,074,000

2,863,000
528,000
5,015,000
1,495,000
9,901,000
9,395,000
56,370,000

5,637,000
6,764,000
68,771,000
27,000
1,495,000
70,293,000
% c:
total direct
investment

2.8
1.2


11.6

15.8
3.8

7.5

' .0
15 5

6.:

22.5

3.4
94.3
5.7
100.0

7.7
1.4
13.5
4.1
26.7
25.3
152.0

15.2
18.3
185.5
0.1
4.0
189.6

Basis
  Evaluation represents project beginning mid-1977,  ending  mid-1980.   Average  cost  basis  for
   scaling,  mid-1979.
  Stack gas  reheat to  175°F by indirect  steam reheat.
  Minimum in-process storage;  only pumps are  spared.
  Investment requirements for  fly ash removal and  disposal  excluded;  FGD  process  investment
   estimate  begins with common feed plenum downstream  of  the  ESP.
  Construction labor shortages with accompanying overtime pay incentive not  considered.


                                           207

-------
                        TABLE B-20.  MAGNESIA SLURRY - REGENERATION PROCESS

            SUMMARY OF AVERAGE ANNUAL REVENUE REQUIREMENTS  - REGULATED UTILITY ECONOMICS

                         (500-MW new coal-fired  power unit,  3.5% S in  fuel;
             1.2 Ib S02/MBtu heat input allowable emission;  108,000 tons/yr  100%

Direct Costs
Raw materials
Magnesium oxide
Catalyst
Hydrated lime
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Fuel oil (No. 6)
Steam
Process water
Electricity
Heat credit
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Indirect Costs
Capital charges
Depreciation, interim replacements
Annual Unit
quantity cost, $


1,470 tons 300.00/ton
1,800 liters 2.50/liter
2,800 tons 54.00/ton


47,500 man-hr 12.50/man-hr

5,720,000 gal 0.40/gal
503,400 MBtu 2.00/MBtu
2,359,200 kgal 0.12/kgal
52,277,400 kWh 0.029/kWh
83,400 MBtu 2.00/MBtu


8,500 man-hr 17.00/man-hr




, and
Total % of net average
annual annual revenue
cost, $ requirements


441,000
4,500
151,200
596,700

593,800

2,288,000
1,006,800
283,100
1,516,000
(166,800)

2,595,200
144,500
8,260,600
8,857,300





2.41
0.02
0.83
3.26

3.24

12.49
5.49
1.54
8.27
(0.91)

14.16
0.79
45.07
48.33



insurance at 6.0% of total depreciable
investment
Average cost of capital and taxes
of total capital investment
Overheads
Plant, 50% of conversion cost less

at 8.6%


utilities
Administrative, 10% of operating labor
Marketing, 10% of byproduct sales
Total indirect costs
revenue

Gross average annual revenue requirements
Byproduct Sales Revenue
100% sulfuric acid

108,000 tons 25.00/ton
Net average annual revenue requirements


Equivalent unit revenue requirements
$/ton coal
Mills/kWh burned
(net) 5.24 12.22
4,126,300

6,045,200

1,666,800
59,400
270,000
12,167,700
21,025,000

(2,700,000)
18,325,000
$/MBtu heat
input
0.58
22.52

32.99

9.10
0.32
1.47
66.40
114.73

(14.73)
100.00
$/ton
S removed
524
Basis
  Midwest plant location,  1980 revenue requirements.
  Remaining life of power  plant,  30 yr.
  Power unit on-stream time,  7,000 hr/yr.
  Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed, 35,000 short tons/yr.
  Investment and revenue requirement for removal  and  disposal of  fly ash excluded.
  Total direct investment, $37,074,000;  total depreciable investment,  $68,771,000;  and total
   capital investment, $70,293,000.
  All tons shown are 2,000 Ib.
                                              208

-------
                   TABLE B-21.  WELLMAN-LQRD.SCR0BBING/SULFURIC ACID PROCESS

                             SUMMARY OF ESTIMATED CAPITAL INVESTMENT

                       (500-MW new coal-fired  power  unit,  3.5% S  in  coal;
             1.2 Ib S02/MBtu heat  input allowable emission;  14.3  tons/hr  100% H2804)


                                                                                       % of
                                                                                   total direct
                                                                   Investment,  $     investment
Direct Investment
Materials handling (conveyors, silos, feeders, bins, shaker,
tanks, and pumps)
Cas handling (common feed plenum and booster fans, gas ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack, gas ducts from chloride
scrubber to absorber)
S02 absorption (four absorbers and entrainment separators,
tanks, pumps, filters, agitators, and heat exchangers)
Stack gas reheat (four indirect steam reheaters)
Chloride purge (four chloride scrubbers and entrainment
separators, tanks, agitators, and pumps)
Sulfate crystallization (evaporator-crystallizer , heat
exchanger, pumps, agitator, tank, dryer, conveyors, centrifuge,
bin, silo, and feeder)
S02 regeneration (evaporators, heat exchangers, strippers,
tanks, agitators, pumps, blower, and condensers)
Acid production (complete contact unit for sulfuric acid
production)
Acid storage (storage and shipping facilities for 30-day
production of sulfuric acid)
Subtotal
Services, utilities, and miscellaneous
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment

1,043,000



4,699,000

8,329,000
1,179,000

2,644,000


1,907,000

7,989,000

6,820,000

1,163,000
35,773,000
2,146,000
37,919,000

2,520,000
630,000
5,110,000
1,521,000
9,781,000
9,540,000
57,240,000

5,724,000
6,869,000
69,833,000
28,000
1,587,000
71,448,000

2.8



12.4

21.8
3.1

7.0


5.0

21.1

18.0

3.1
94.3
5.7
100.0

6.6
1.7
13.5
4.0
25.8
25.2
151.0

15.1
18.1
184.2
0.1
4.1
188.4
Basis
  Evaluation represents project  beginning  mid-1977,  ending mid-1980.   Average  cost  basis  for
   scaling,  mid-1979.
  Stack gas  reheat to  175°F by indirect  steam reheat.
  Minimum in-process storage;  only pumps are  spared.
  Investment requirements for  fly ash removal and  disposal excluded;  FGD  process  investment
   estimate  begins with common feed plenum downstream  of  the  ESP.
  Construction labor shortages with accompanying overtime pay incentive not  considered.

                                            209

-------
                      TABLE B-22.   WELLMAN-LORD SCRUBBING/SULFURIC ACID PROCESS

            SUMMARY OF AVERAGE ANNUAL REVENUE REQUIREMENTS - REGULATED UTILITY ECONOMICS

                         (500-MW new coal-fired power unit, 3.5% S in coal;
             1.2 Ib S02/MBtu heat  input allowable emission; 100,300 tons/yr 100% H2S04)

Annual
quantity

Unit
cost, $
Total
annual
cost, S
% of average
annual revenue
requirements
Direct Costs

Raw materials
  Sodium carbonate
  Catalyst
  Hydrated lime

     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Steam
    Process water
    Electricity
    Heat credit
  Maintenance
    Labor and material
  Analyses

     Total conversion costs

     Total direct costs
     6,860 tons
     2,000 liters
     2,800 tons
 1,523,080 MBtu
 6,105,760 kgal
50,882,670 kWh
    48,360 MBtu
103.00/ton
  2.50/liter
 54.00/ton
    46,500 man-hr    12.50/man-hr
  2.00/MBtu
  0.12/kgal
  0.029/kWh
  2.00/MBtu
     8,500 man-hr    17.00/man-hr
  706,600
    5,000
  151.200

  862,800
  581,300

3,046,200
  732,700
1,475,600
  (96,700)

2,654,300
  144.500

8,537,900

9,400,700
 3.62
 0.03
_0.77

 4.42
 2.98

15.59
 3.75
 7.56
(0.50)

13.59
 0.74

43.71

48.13
Indirect Costs

Capital charges
  Depreciation, interim replacements, and
   insurance at 6.0% of total depreciable
   investment
  Average cost of capital and taxes at 8.6%
   of total capital investment
Overheads
  Plant, 50% of conversion cost less utilities
  Administrative, 10% of operating labor
  Marketing, 10% of byproduct sales revenue

     Total indirect costs

     Gross average annual revenue requirements
                                     4,190,000

                                     6,144,500

                                     1,690,100
                                        58,100
                                       269.400

                                    12,352,100

                                    21,752,800
                                  21.45

                                  31.46

                                   8.65
                                   0.30
                                   1.11

                                  62.97

                                 110.10
Byproduct Sales Revenue

100% sulfuric acid                     100,300 tons

Sodium sulfate                           8,110 tons

     Net average annual revenue requirements
                     25.00/ton 100% (2,507,500)      (10.27)
                           H2S04
                     23.00/ton        (186,500)       (0.83)

                                    19,058,800       100.00
                                                          $/ton coal   $/MBtu heat     $/ton
                                              Mills /kWh	burned	input	S removed
Equivalent unit revenue requirements (net)
                                                5.44
                        12.71
                                      0.61
                                                    545
Basis
  Midwest plant location, 1980 revenue requirements.
  Remaining life of power plant, 30 yr.
  Power unit on-stream time, 7,000 hr/yr.
  Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed, 35,000 short tons/yr.
  Investment and revenue requirement for removal and disposal of fly ash excluded.
  Total direct investment, $37,919,000; total depreciable investment, $69,833,000; and total
   capital investment, $71,441,000.
  All tons shown are 2,000 Ib.                210

-------
               TABLE B-23.  U'ELLMAN-LORD SCRUBBING/ALLIED CHEMICAL COAL/S02 REDUCTION PROCESS

                                  SUMMARY OF  ESTIMATED  CAPITAL  INVESTMENT

                             (500-MW new coal-fired  power  unit,  3.5%  S  in  coal;
                  1.2 Ib 502/MBtu hea; input  allowable  emission;  4.7  tons/hr  elemental  S)



Direct Investment
Materials handling (conveyors, silos, feeders, bins, shaker,
tanks , and pumps )
Gas handling (common feed plenum and booster fans, gas ducts and
dampers from plenum to absorber, exhaust gas ducts and dampers
from absorber to reheater and stack, gas ducts from chloride
scrubber to absorber)
tanks, pumps, filters, agitators, heat exchangers)
Stack gas reheat (four indirect steam reheaters)
Chloride purge (four chloride scrubbers and entrainment
separators, tank, agitators, and pumps)
Sulfate crystallization (evaporator crystallizer, heat
exchanger, pumps, agitator, tank, dryer, conveyors, centrifuge,
bin, silo, and feeder)
S02 regeneration (evaporators, heat exchangers, stripper, tanks,
agitators, pumps, blower, and condensers)
Sulfur production (complete coal reduction unit)
Sulfur storage (storage and shipping facilities for 30-day
production of sulfur)
Subtotal
Services, utilities, and miscellaneous
Total direct investment excluding pond construction
Fond construction
Total direct investment
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees
Total indirect investment
Contingency
Total fixed investment
Other Capital Charges
Allowance for startup and modifications
Interest during construction
Total depreciable investment
Land
Working capital
Total capital investment


Investment, $


1,056,000



4,699,000
8,344,000
1,179,000

2,644,000


1,919,000

8,064,000
8,400,000

593,000
36,898,000
2,214,000
39,112,000
269,000
39,381,000

2,789,000
692,000
5,286,000
1,566,000
10,333,000
9,620,000
59,334,000

5,907,000
7,120,000
72,361,000
64,000
1,765,000
74,190,000
% of
total direct
investment


2.7



11.9
21.2
3.0

6.7


4.9

20.5
21.3

1.5
93.7
5.6
99.3
0.7
100.0

7.0
1.8
13.4
4.0
26.2
24.4
150.7

15.0
18.0
183.7
0.2
4.5
188.4
Basis
  Midwest plant  location  represents  project  beginning mid-1977, ending mid-1980.  Average
   cost basis  for  scaling,  mid-1979.
  Stack gas  reheat to  175°F by  indirect  steam reheat.
  Minimum in-process storage; only pumps are spared.
  Disposal pond  located 1 mile  from  power  plant.
  Investment requirement  for  fly ash removal and disposal excluded; FGD process  investment
   estimate  begins with common  plenum downstream of the ESP.
  Construction labor shortages  with  accompanying overtime pay incentive not considered.
                                                  211

-------
          TABLE B-24.  WELLMAN-LORD SCRUBBING/ALLIED CHEMICAL COAL/S02 REDUCTION PROCESS

           SUMMARY OF AVERAGE ANNUAL REVENUE REQUIREMENTS   REGULATED UTILITY ECONOMICS

                        (500-MW new coal-fired power unit, 3.5% S in coal;
            1.2 Ib S02/MBtu heat input allowable emission; 32,690 tons/yr elemental S)
Direct Costs

Raw materials
  Sodium carbonate
  Goal
  Sand
  Catalyst
  Hydrated lime

     Total raw materials cost

Conversion costs
  Operating labor and supervision
  Utilities
    Fuel oil (No. 6)
    Steam
    Process water
    Electricity
    Heat credit
  Maintenance
    Labor and material
  Analyses

     Total conversion costs

     Total direct costs
                                          Annual
                                         quantity
                    Unit
                   cost,
               Total
               annual
               cost, $
7,440 tons
25,370 tons
180 tons

2,800 tons
103.00/ton
26.50/ton
7.50/ton

54.00/ton
766,300
672,300
1,400
3,800
151,200
46,500 man-hr
                  12.50/man-hr
                                 1,595,000
                                   581,300
             % of net average
              annual revenue
               requirements
                                                   3.57
                                                   3.13
                                                   0.01
                                                   0.02
                                                   0.70

                                                   7.43
                                                   2.71
582,580 gal
1,581,820 MBtu
5,219,260 kgal
48,230,700 kWh
21,640 MBtu
8,800 man-hr

0.40/gal
2.00/MBtu
0.12/kgal
0.029/kWh
2.00/MBtu
17. 00 /man-hr

233,000
3,163,600
626,300
1,398,700
(43,300)
2,756,700
149,600
8,865,900
10,460,900
1.08
14.72
2.92
6.51
(0.20)
12.83
0.70
41.27
48.70
Indirect Costs

Capital charges
  Depreciation, interim replacements,  and
   insurance at 6.0% of total depreciable
   investment
  Average cost of capital and taxes at 8.6%
   of total capital investment
Overheads
  Plant, 50% of conversion cost less utilities
  Administrative, 10% of operating labor
  Marketing, 10% of byproduct sales revenue

     Total indirect costs

     Gross average annual revenue requirements
                                 4,341,700

                                 6,380,000

                                 1,743,800
                                    58,100
                                   167,400

                                12,691,000

                                23,151,900
                                20.71

                                29.71

                                 8.12
                                 0.27
                                 0.78

                                59.09

                               107.79
Byproduct Sales Revenue

Sulfur
Sodium sulfate

     Net average revenue requirements
32,690 tons
 8,800 tons
45.00/ton S
23.00/ton
(1,471,100)
  (202,400)

21,478,400
 (6.85)
 (0.94)

100.00
Equivalent unit revenue requirements (net)
                                                             S/ton coal   $/MBtu heat     $/ton
                                                 Mills/kWh	burned	input	S removed
                                                    6.14
                                                                14.32
                                                                              0.68
                                                                                           614
  Midwest plant location, 1980 revenue requirements.
  Remaining life of power plant,  30 yr.
  Power unit on-stream time, 7,000 hr/yr.
  Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
  Stack gas reheat to 175°F.
  Sulfur removed, 35,000 short tons/yr.
  Investment and revenue requirement for removal and  disposal of fly ash excluded.
  Total direct investment, $39,381,000;  total depreciable investment, $72,361,000;  and total
   capital investment, $74,190,000.
  All tons shown are 2,000 Ib.
                                              212

-------
                                    TABLE B-25.  CITRATE PROCESS

                              SUMMARY OF ESTIMATED FIXED INVESTMENT3

                       (500-MW new coal-fired power unit, 3.5% S in coal;
            1.2 Ib 502/MBCu heat Input allowable emission; 4.8 tons/hr elemental S)
                                                                  Investment, $
                                                                                       % of
                                                                                   total direct
                                                                                    Investment
Direct Investment

Materials handling (conveyors and bins)
Feed preparation (conveyors, tanks, agitators, pumps,
 and feeders)
Gas handling (common feed plenum and booster fans, gas ducts
 and dampers from plenum to absorber, exhaust gas ducts and
 dampers from absorber to reheater and stack)
S02 absorption  (four packed tower absorbers including
 presaturators and mist eliminators, surge tanks, centrifugal
 pumps, compressor, and strippers)
Stack gas reheat (four indirect steam reheaters)
Chloride purge  (feeder, tank, agitator, and pump)
S02 reduction (reactor tanks, aging tanks, agitators, and
 centrifugal pumps)
Sulfur separation and removal (flotation tanks, rotary drum
 filter, pumps, slurry tank, heat exchanger, settling tank,
 heaters, and flash drum)
Sulfur storage and shipping (sulfur receiving pit, heaters,
 sulfur pump, and storage tank)
Sulfate removal (coolers, agitators, centrifuge, tanks,
 pumps, and refrigeration)
H2S generation  (battery limit plant)
H2 generation (battery limit plant)

     Subtotal

Services, utilities, and miscellaneous

     Total direct investment
   804,000

   118,000


 4,368,000
14,223,000
 1,294,000
    83,000

 1,303,000
 2,118,000

   814,000

   985,000
 5,850,000
 4.680.OOP

36,640,000

 2,198,000

38,838,000
  2.1

  0.3


 11.2
 36.6
  3.3
  0.2

  3.4
  5.5

  2.1

  2.5
 15.1
 12.0

 94.3

  5.7

100.0
Indirect Investment
Engineering design and supervision
Architect and engineering contractor
Construction expense
Contractor fees

     Total Indirect Investment

Contingency

     Total fixed investment
 3,273,000
   818,000
 5,208,000
 1.548,000

10,847,000

 9.937,000

59,622,000
  8.4
  2.1
 13.4
  4.0

 27.9

 25.6

153.5
Other Capital Charges

Allowance for startup and modifications
Interest during construction

     Total depreciable investment

Land
Working capital

     Total capital investment
 5,962,000
 7.155..00Q

72,739,000

    39,000
 2,140.000

74,918,000
 15.4
 18.4
192.9
    Basis
      Evaluation represents project beginning mid-1977, ending mid-1980:  Average
       cost basis for scaling, mid-1979.
      Stack gas reheat to 175°F by Indirect steam reheat.
      Minimum in-process storage; only pumps are spared.
      Investment requirements for fly ash removal and disposal excluded; FGD process investment
       estimate begins with common feed plenum downstream of the ESP,
      Construction labor shortages with accompanying overtime pay incentive not considered.
                                         213

-------
                                     TABLE B-26.   CITRAIE PROCESS

              TOTAL AVERAGE A.S'NUAL REVENUE REQUIREMENTS   REGULATED UTILITY ECONOMICS"

                         (500-MW new coal-cirea power unit,  3.51: S in coal
             1.2 Ib SO->/MBtu heat Input allowable cnission;  33.S40 tons/yr elemental S)
" of net average
 annual revenue
  requirement s
                                             Annual
                                            quantity
                L'nit
               cost, S
                                                                            Total
                                                                            annual
                                                                            cost, ?
Direct Coats

Raw materials
  Lime
  Soda ash
  Citric acid
  Natural gas
  Catalyst

     Total raw material cost

Conversion costs
  Operating labor and supervision
  Utilities
    Steam
    Process water
    Electricity
Maintenance
  Labor and material
Analyses

     Total conversion costs

     Total direct costs
2,870 tons
2,630 tons
230 tons
1,050,000 kftj
42.00/ton
10). DO/ton
1,340.00/ton
3.50/kctJ
120,500
270,900
308,200
3,675,000
21,000
                                                                           4,395,600
                                            67.920 aian-hr    12.50/man-hr    849,000
                                         1,027,500 MBtu
                                         2,492,500 kgal
                                        68,530,000 kWh
                                            10,600 man-hr
                  2.00/MBtu
                  0.12/kKal
                  0.029/kWh
                                                                           2,055,000
                                                                             299,100
                                                                           1,987,400
                               2,330,300
                 17.00/man-hr    180,200

                               7,701,000

                              12,096,600
                                                                                              0.52
                                                                                              1.16
                                                                                              1.32
                                                                                             15.78
                                                                                              0.09

                                                                                             18.87
      3.64

      8.82
      1.28
      8.53

     10.01
      0.77

     33.05

     51.92
Indirect Costs

Capital charges
  Depreciation, interim replacements,  and
   insurance at 6.0% of total depreciable
   investment
  Average cost of capital and taxes at 8.6%
   of total capital Investment
Overheads
  Plant, 50% of conversion costs less  utilities
  Administrative, 10% of operating labor
  Marketing, 10% of byproduct sales revenue

     Total indirect costs

     Gross average annual revenue requirements
                                                                           4,364,300

                                                                           6,442,900

                                                                           1,679,300
                                                                              84,900
                                                                             152,30U

                                                                          12,724,200

                                                                          24,820,800
                                                 13.73

                                                 27.67

                                                  7.21
                                                  0.36
                                                  0.63

                                                 54.62

                                                106.54
Byproduct Sales Revenue

Sulfur

     Net average revenue requirements
33,840 tons      45.00/tons   (1.522.800)        (6.54)

                              23,298,000        100.00

             $/ton coal   $/MBtu heat     $/ton
 Mills/kWh	burned	input	S removed
Equivalent unit revenue requirements (.net)      6,66
                15,53
                                                                           0.74
                                                                                       666
    Basis
      Midwest plant location, 1980 revenue requirements.
      Remaining life of power plant, 30 years.
      Power unit on-stream time, 7,000 hr/yr.
      Coal burned, 1,500,100 tons/yr, 9,000 Btu/kWh.
      Stack gas reheat to 175°F.
      Sulfur removed, 35,000 short tons/yr.
      Investment and revenue requirement for removal and  disposal of fly ash excluded.
      Total direct investment, $38,838,000; total depreciable investment, $72,739,000; and total capital
       investment, $74,918,000.
      All tons shown are 2,000 pounds.
                                                  214

-------
                                                                             3C
                           COMBINED COAL CLEANING
                                  AND FGD

                              James D. Kilgroe

                    U.S. Environmental Protection Agency
                Industrial  Environmental Research Laboratory
                       Research Triangle Park, N. C.
ABSTRACT

     Physical  coal cleaning (PCC) can be used to attain moderate reductions
in the ash and sulfur levels of U.S. coals.  PCC can thus be used to
generate compliance fuel for the less stringent State and Federal standards
governing fossil fuel fired steam generators.  The sulfur reduction require-
ments and emission levels which are likely to be specified in the revised
New Source Performance Standards (NSPS) for electric utility boilers will
preclude the use of coal cleaning as a sole method of complying with these
flue gas desulfurization (FGD) regulations.
     The combined use of physical coal cleaning and flue gas desulfurization
(PCC + FGD) will be the most cost effective method of complying with emission
regulations ,if the reduction in FGD and non-FGD costs which result from
using cleaned coal are greater than the costs of PCC.  Reductions in FGD costs by
PCC can result from a reduction in the volume of flue gas treated (partial scrub-
bing) or the amount of sulfur removed from the flue gas stream.  Reductions in
fuel sulfur variability by PCC can lower design safety margins needed to ensure
compliance for all fuel  sulfur values.  Non-FGD cost benefits can result from re-
duced boiler operation and maintenance costs, reduced transportation costs, re-
duced ash disposal costs,and reduced coal pulverization costs.
     Utility boilers which use high sulfur coals and which require sulfur re-
movals less than 75 percent are likely candidates for PCC + FGD.  If the revised
NSPS for utility boilers require 90 percent sulfur removal and do not specify an
emission floor, then PCC + FGD may not be competitive with FGD unless there are
substantial non-FGD cost benefits associated with cleaning.
                                     215

-------
     The range of applications for PCC + FGD in small  non-base-loaded utility
boilers and industrial  boilers may be different from those cited for base-loaded
utility boilers.   The differentials between PCC and FGD costs for these smaller
units may result in different optimal solutions for the range of alternative
control strategies.

 INTRODUCTION

      The Clean Air Act  Amendments  of 1977  will  have a  substantial  impact  on
 the  costs and technologies  used  to comply  with  State and  Federal  S02 emission
 regulations.  Regulatory  actions in  response to these  amendments  will  result
 in tightening of emission standards  in  existing boilers in order  to meet
 ambient air quality standards  in non-attainment areas.  Revised new source
 performance standards  (NSPS)  will  be set for coal  fired utility boilers and
 NSPS will be  promulgated  for  industrial boilers.
      Physical coal  cleaning (PCC)  can be used to attain moderate  reductions
 in the ash and sulfur levels of  U.S. coals.  PCC can thus be used to
 generate compliance fuel  for  the less stringent state  standards and current
 NSPS governing  fossil fuel  fired steam  generators.  The sulfur reduction  re-
 quirements and emission levels which are likely to be  specified for the
 revised NSPS  for electric utility  boilers  will  preclude the  use of coal clean-
 ing  as a sole method of complying  with  these regulations.
      Previous studies have  shown that in some  instances combinations of coal
 cleaning and  flue gas desulfurization (FGD) can be a more cost effective
 emission control  technique  than  FGD  alone.  This paper summarizes the  poten-
 tial  of PCC as a  method of  coal  desulfurization and evaluates conditions
 under which PCC  can  be  used in conjunction with FGD to reduce the cost of
 complying with S02  emission regulations.
 AIR  POLLUTION  REGULATIONS

      The  U.S.  EPA  is  developing  and  implementing  air  pollution  control  regu-
 lations in  accordance with  the provisions  of  the  Clean Air  Act  and  its
                                    216

-------
 amendments.   Stationary  source  emission  standards  are  designed  to  regulate
 the  quantities  of pollutants  emitted  from  point  sources, whereas ambient  air
 quality standards are  designed  to  regulate the concentrations of pollutants
 in the  atmosphere.

 Ambient Air  Quality Standards
      Under Section  109 of  the Clean Air  Act of 1970, the U.S. EPA  has estab-
 lished  national  primary  and  secondary ambient air  quality  standards  (NAAQS)
 to  protect human health  and  public welfare, respectively.   State Implementation
 Plans (SIPs)  which must be  approved by the  U.S. EPA are used  to  achieve permissi-
 ble  air quality levels for certain "criteria" pollutants including:  total
 particulates, sulfur dioxide  and nitrogen  oxides.
     The  1977  Clean  Air Act Amendments  require EPA  to review the NAAQS no later
 than  December 31, 1980 and at 5 year  intervals thereafter.   If warranted, re-
 visions  to the  NAAQS are to be made.   Subsequent to changes in the NAAQS, each
 state is required to modify its SIP to comply with the new air quality stand-
 ards.   Existing  coal fired boilers are regulated under SIP  regulations.
 Emission limits  for these sources may  range  from about 0.2  Ib S00/106 Btu*to
                    6
 above 6.0  Ib  S02/10 Btu depending on  the  boiler site.

 New  Source Performance Standards
      New Source Performance Standards are  issued by the U.S. EPA   in accordance
 with Section  111  of the  1970  Clean Air Act.  These standards of performance
 are  applied  to  new  and modified source categories  designated by EPA.  Many
 provisions of the Clean  Air Act Amendments of 1977 are aimed specifically
 at coal  firing  sources,  and the restrictions applied are much more rigorous
 than  in  the  past.   Where the  original  New  Source Performa-nce Standard for
 S02  emissions applying to  large coal  fired boilers permitted the emission
 of 1.2  Ib  S02/million Btu, the amended Act specified   that the revised
 NSPS  "...shall  reflect the degree of  emission limitation and the percentage re-
 duction  achievable  through application of  the best technological system of con-
 tinuous  emission  reduction...;" i.e.,  a  percentage reduction will  be required
* English to metric unit conversion factors are given at the end of this
  paper.
                                    217

-------
in addition to the maintenance of emissions below an upper limit.   Any cleaning
of the fuel which reduces the pollution characteristics of the fuel  after extrac-
tion and prior to combustion may be credited to the pollution percentage reduction
requirement.
     There is still considerable uncertainty concerning the revised NSPS for
coal fired steam electric utility boilers which are to be promulgated by EPA
(EPA, 1978a).  At least four full and seven partial S02 control  alternatives
are currently under consideration (EPA, 1978b).   The full  control  options gen-
erally specify emission limits of 0.55 to 0.80 Ib S02/10  Btu with 90 to 95
percent sulfur control    expressed on an annual  basis.  The partial  control
options generally include an emission limit, a percentage removal  requirement,
and a maximum control  or emission floor below which the full  sulfur reduction
requirement does not apply.  The partial control options typically result in
reduced emission control costs where the emission floor can be met by
burning low sulfur coals and partial scrubbing.
     EPA is also considering NSPS for industrial boilers.   Alternative standards
now being studied specify emission regulations as a function of boiler size.
Standards being studied for small and intermediate size boilers  specify emission
limits ranging from 1.2 to 2.0 Ib SO^/IO  Btu and sulfur reduction requirements
ranging from zero to 90 percent.

Prevention of Significant Deterioration
     A new Part C (Sections 160-169) was incorporated into the Clean Air Act
Amendments of 1977 for the prevention of significant deterioration (PSD) in
the present ambient air quality.  Limits on the permissible deterioration of
air  quality  are:  Class I, little or no deterioration; Class Unlimited
deterioration; and Class III, moderate deterioration.
     Any new source in an area subject to the PSD provisions of  the Act must
employ the Best Available Control Technology (BACT) for each pollutant subject
to regulation.  BACT,  which is determined on a case-by-case basis, must
consider the available technologies and the energy, environmental, and economic
impacts of each.  Thus, the BACT identified for PSD may require higher
levels of control than specified by the NSPS for that source category.
                                     218

-------
Nonattainment Areas
     Provisions were also incorporated into the Clean Air Act Amendments of
1977 to alleviate air pollution problems in areas where one or more air pol-
lutants exceed any NAAQS.  In "nonattainment areas," new sources must employ
pollution control technology which provides the "lowest achievable emission
rate" (LAER).  Before construction permits are issued, a reduction
in emissions from existing sources must be obtained to more than "offset"
the new source emissions.  Standards for new sources in nonattainment
areas are to be set by the individual states through the  SIPs on a case-
by-case basis.

^PHYSICAL DESULFURIZATION POTENTIAL OF U.S.  COALS

      The decision to use a  given set of control  technologies  as  a method
 for complying with S02 emission regulations  will  be based  on  the technical
 applicability and relative  cost of the various control  options.   The  three
 control  options  most likely to be considered for  a  wide range of regulations
 are physical coal cleaning  (PCC), flue gas  desulfurization (FGD), or  a  combina-
 tion of physical  coal  cleaning and flue gas  desulfurization  (PCC +  FGD).
 Aside from economic factors the primary constraints which  must be considered
 are the S02 emission regulation and the properties  of the  coal(s) which are
 to be used.
      FGD is a flexible technology which can  be used to comply with  a  wide
 range of S02 emission  regulations.  PCC,  while generally less costly,is
 limited in its range of application because  of the  inherent  properties  of
 coal.  An understanding of  the physical  desulfurization potential of  U.S.
 coals is essential  to  the analysis of the use of  PCC + FGD as a  S02 emission
 control  strategy.
      The sulfur  content of  U.S. coals varies considerably.   While 46  weight
 percent of the total  reserve base can be identified as low-sulfur coal
 (coal with less  than 1  percent sulfur),  21  percent  ranges  between 1  and 3
 percent sulfur an additional 21 percent contains more than 3 percent
 sulfur.   The sulfur content of 12 percent of the  coal  reserve base  is
 unknown,  largely because many coal beds have not  been adequately charac-
 terized.
                                     219

-------
     Sulfur appears in coal in two principal forms:  organic sulfur and
mineral sulfur in the form of pyrite.  Organic sulfur, which comprises from
30 to 70 percent of the total sulfur content of most U.S. coals, is an inte-
gral part of the coal matrix and can only be removed by chemical modification
of the coal structure.
     Pyritic sulfur occurs in coal as discrete particles, often of micro-
scopic size.  Pyrite is a heavy mineral which has a specific gravity of
5.0; coal has a maximum specific gravity of only 1.7.   The pyrite content
of most coals can be significantly reduced by crushing and specific gravity
separation.  However, gravimetric separation of very fine coal  and pyrite
particles  is not effective; separation techniques which depend on the sur-
face or electromagnetic properties of the particles must be used.
     Laboratory float-sink studies have been performed by the U.S. Bureau *
of Mines (USBM) on more than 455 U.S. coals to evaluate the effects of
crushing and specific gravity separation on pyrite removal (Cavallaro, 1976).
The samples tested were from mines in six major U.S. coal producing regions,
which provide more than 70 percent of the coal used in U.S. utility boilers.
In general, pyrite removal increases with decreasing coal particle size
and specific gravity of separation.
     The specific gravity desulfurization potential of U.S. coals
varies between coal regions and between coal beds within the same
region (Cavallaro, 1976).   Table 1 summarizes the average sulfur values in
coals from  six U.S. coal regions:  Northern Appalachian (NA), Southern
Appalachian (SA), Alabama (A), Eastern Midwest (EMW),  Western Midwest (WMW),
and Western (W).  Assuming that all of the pyritic sulfur could be removed by
physical cleaning, average emissions from the organic sulfur would range from
0.73 to 2.86 Ib S0«/10  Btu.  The percentage sulfur reduction (expressed in
         c        f-
Ib S02/10   Btu) achievable by removing all of the pyritic sulfur ranges from
34 to 68 percent.
     The sulfur levels which could actually be achieved by crushing these
coals to 3/8-inch top size and by gravimetrically separating them at 1.6 spe-
cific gravity are shown in Table 2.  Total sulfur emissions would range from
0.9 to 5.5  Ib S02/10  Btu.  The percentage sulfur reduction at these clean-
ing conditions ranges from about 15 to 44 percent.
                                    220

-------
N3
                               TABLE 1 .  AVERAGE SULFUR VALUES IN COALS FROM SIX U.S. COAL REGIONS1
                                                          (Ib S02/10°  Btu)
REGION
Northern Appalachian
Southern Appalachian
Alabama
Eastern Midwest
Western Midwest
Western
Total
Sulfur(St)
4.8
1.6
2.0
6.5
9.0
1.1
Standard /. v
Deviation^ '
2.7
1.0
1.5
2.1
4.5
0.6
Pyritic
Sulfur(S )
3.20
0.59
1.04
3.80
6.14
0.37
Organic
Sulfur(SQ)
1.60
1.01
0.96
2.70
2.86
0.73
yst
0.667
0.369
0.520
0.585
0.682
0.336
           (a)
               Cavallaro, 1976


           * 'Standard deviation of total sulfur values

-------
                           TABLE 2.   SUMMARY OF AVERAGE PHYSICAL DESULFURIZATION

                                            POTENTIAL OF COALS BY REGION *
                           (Cumulative analysis of float 1.60 product for 3/8-inch top size)
Coal
Region
Northern
Appalachian
Southern
Appalachian
ro
N> Alabama
Eastern
Midwest
Western
Midwest
Western
Total U. S.
No.
Samples
227
35
10
95
44
44
455
Btu
Recovery,
Percent
92.5
96.1
96.4
94.9
91.7
97.6
93.8
Ash,
Percent
8.0
5.1
5.8
7.5
8.3
6.3
7.5
Pyritic
Sulfur,
Percent
0.85
0.19
0.49
1.03
1.80
0.10
0.85
Total
Sulfur,
Percent
1.86
0.91
1.16
2.74
3.59
0.56
2.00
Emission on
Combustion,
Ib S02/1Q Btu
2.7
1.3
1.7
4.2
5.5
0.9
3.0
Calorific
Content, Btu/lb
13,766
14,197
14,264
13,138
13,209
12.779
13,530
* Callavaro,  1976.

-------
     The above cleaning conditions are representative of the physical  de-
sulfurization which can be obtained by applying technology now used  primarily
to remove mineral  matter from steam coals.   By optimization of physical  coal
cleaning processes, it is probable that from 50 to 60 percent of the total
sulfur can be removed from high sulfur coals.   Improvements in the cleaning
conditions used for low sulfur coals could  probably improve total  sulfur
removal capabilities to the range of 20 to  30 percent.
     Estimates of the potential amount of coal in the NA, EMW, and W coal
regions which could be used to comply with  various emission levels are
shown in Figures 1 through 3.  The quantities of coal from each of these
regions which can achieve an emission limit of 2.0 Ib S02/10  Btu at
varying percentages of sulfur reduction are shown in Figure 4 through
6 (Hall, 1979).
      In  evaluating  this  data  and  other  information on U.S. coals,the follow-
 ing  general  observations can  be made:
      1.    PCC  can  be  used for moderate reductions in the sulfur con-
           tents of  high  sulfur Northern Appalachian  and Midwestern coals.
           However,  few of these coals can  be  cleaned to the  1.2 Ib S02/10
          Btu  level  specified  by  the current  NSPS for coal fired steam
           generators.
      2.    Many Southern  Appalachian, Alabama,or Western coals are capable
           of meeting  the current  NSPS coal  fired  steam generators ,
           either as-mined or  after cleaning.
      3.    Emission  regulations which specify  emission limits  below about
           1.0  Ib S02/10   Btu  preclude the  use  of  physically  cleaned  high
           sulfur coal  for compliance with  these regulations.  This is a
           consequence  of the  high  organic  sulfur  contents of these coals
           and  the fine sized  pyrite which  cannot  be  removed  by PCC.
      4.    Emission  regulations which specify  sulfur  reduction requirements
           above 30  percent preclude the use of low sulfur coals from com-
           pliance with these  regulations.   The percentage of sulfur which
           can  be removed from  U.S.  coals by PCC is directly  proportional
           to the ratio of pyritic  to total  sulfur.   The fraction of pyritic
                                    223

-------
NJ
                1700
                1500
                1300
                1100


             M
             o

             3
             2  900
             M
             b
             H  700
             e
                600
                300
                100
                                              — RAW COAL

                                               • PCC, 1-1 1/2 in., 1.6 S.G.

                                               • PCC. 3/8 in.. 1.6 or 1.3 S.G.

                                               A 90% PYRITIC SULFUR REMOVED
                                                                                                       TOTAL QUADS OF RAW COAL = 1728.37
                                                                       J_
                                             1.0
            2.0                       3.0

EMISSION STANDARD lib. SOJIO6 BTU), N. APPALACHIAN
                                                                                                                          4.0
                                              FIGURE 1
                                                          ENERGY AVAILABLE IN N. APPALACHIAN RESERVE BASE AS A FUNCTION OF EMISSION
                                                               STANDARDS FOR VARIOUS PHYSICAL COAL CLEANING LEVELS

-------
             1700
             1600
             1300
— RAW COAL
 •  PCC. 1-1 1/2 in.. 1.6 S.G.
 •  PCC. 3/8 in.. 1.6 or 1.3 S.G.
 A  90% PYRITIC SULFUR REMOVAL
                                 TOTAL QUADS OF RAW COAL = 1998.69
             1100
          to
              900
Is)
10
VJ1
          fe  700
              500
              300
              100
                                                                  2.0                      3.0
                                                          EMISSION STANDARD lib. SOJ10* BTU) E. MIDWEST
                                                                                                                    4.0
                                          FIGURE 2
                                                     ENERGY AVAILABLE IN E. MIDWEST RESERVE BASE AS A FUNCTION OF EMISSION
                                                         STANDARDS FOR VARIOUS PHYSICAL COAL CLEANING LEVELS

-------
                  3600
                  3200
                  2800
                  2400
              3 2000
              g
              M
K3
N3
ON
CD

I
                  1600
                  1200
                  800
                  400
                                                                                                      — RAW COAL
                                                                                                      • PCC. 1-1 1/2 in., 1.6S.G.
                                                                                                      • PCC. 3/8 in., 1.6 or 1.3 S.G.
                                                                                                      A 90% PYRITIC SULFUR REMOVED
                                                                                                    TOTAL QUADS OF RAW COAL = 3662.29
                                                                          I
                                                                                                    I
                                               1.0
                                                           2.0                        3.0
                                                    EMISSION STANDARD (Ib. SO2/106 BTU). WESTERN
                                                                                                                             4.0
                                                  FIGURE 3
                                                             ENERGY AVAILABLE IN WESTERN RESERVE BASE AS A FUNCTION OF EMISSION
                                                                  STANDARDS FOR VARIOUS PHYSICAL COAL CLEANING LEVELS

-------
   100
    80  -
O
5
111
ff
I
Ul

111
u_
O
    60  -
S   40

E
    20 -
                                                            iiiiiiin PCC 1 - 1/2 INCH. 1.6 SG PROCESS



                                                             — —  PCC 3/8 INCH, 1.3 SG PROCESS



                                                            • •••ii 0.9 PY.S REMOVED PROCESS



                                                                   TOTAL QUADS OF RAW COAL = 1.728
                                             *v
                                        x.
                                                                              >»,.
                 10
                           20
                                                                              70
                                                                                        80
                          30         40         50         60


                               PERCENT SULFUR REMOVAL


FIGURE 4.  NORTHERN APPALACHIAN REGION; ENERGY AVAILABLE TO MEET PERCENT SULFUR REMOVAL


          STANDARDS WITH A 2.0 LB SO/106 BTU EMISSION LIMIT. (HALL, 1979)
                                                                                                  90
                                                                                                           100

-------
                 100 -I
                  80 -
                  60 -
NJ
£
i
                  20 -
                                                                        •ilium  pCC i - 1/2 INCH. 1.6 SG PROCESS

                                                                        • •• •  PCC 3/8 INCH. 1.3 SG PROCESS

                                                                        11 •• • i  0.9 PY.S REMOVED PROCESS

                                                                                TOTAL QUADS OF RAW COAL = 1.999
                                                  	
                _^______^	    *i""HMIHM........-?!?i™»

                 10         20        30        40         50        60

                                          PERCENT SULFUR REMOVAL
                              FIGURE 5  EASTERN MIDWEST REGION; ENERGY AVAILABLE TO MEET PERCENT SULFUR REMOVAL

                                        STANDARDS WITH A 2.0 LB SO/10* BTU EMISSION LIMIT. (HALL. 1979)
                                                                                                                  90
                                                                                                                           100

-------
1OO -
"

80 -


O
0
E 60 -
Z
>.
CO 5
N5 E
VO Z -
ill
u.
0
fc 40-
Ul
O
E
Ul
O.
"
20 -


-



iiiiiin PCC 1 - 1/2 INCH. 1.6 SG PROCESS
^ — — • PCC 3/8 INCH. 1.3 SG PROCESS
'\ *\ i • • 1 1 0.9 PY.S REMOVED PROCESS
**» V
\ > TOTAL QUADS OF RAW COAL = 3,662
\*
ft *
'V,
• \
\ \
**••*.*':
^ ^
\\
vs ^
*++ \'^"-\
\ \ \
X \ \

*%'C\ x*^—.-.
**« «, "**^»
**£\ X
x>x v
\ N^v x
%. V x.
lltllltl E|
10
           20         30         40        50         60

                          PERCENT SULFUR REMOVAL
                                                              70
80
                                                                                   90
                                                                                             100
FIGURE 6.  WESTERN REGION; ENERGY AVAILABILITY TO MEET PERCENT SULFUR REMOVAL
          STANDARDS WITH A 2.0 LB SO 2 /10 6 BTU EMISSION LIMIT. (HALL. 1979)

-------
           sulfur  in  low  sulfur coals  is  less  than  40  percent.   Rarely  can
           sufficient pyrite be removed from low  sulfur  coals  to achieve  a
           total sulfur reduction above 30 percent.
       5.   Emission regulations which  specify  any combination  of emission
           limit below 1.0 Ib S02/10  Btu and  sulfur reduction above  30 per-
           cent will  essentially eliminate PCC as a single control  technology
           for compliance.  For these  types of regulations, PCC  must  be used
           in conjunction with some other control technology such as  wet
           limestone  scrubbing or dry  scrubbing.
STATUS AND COSTS OF PCC

     There are more than 460 physical  coal  cleaning plants  which can process
approximately 360 million metric tons  (400  million tons)  of raw coal per year.
The principal coal cleaning processes  used  are oriented toward product stan-
dardization and ash reduction,  with increased attention to  sulfur removal as
the demand for low sulfur utility fuels grows.
      Five  general  levels of coal preparation are  used  in upgrading  coal.
 Each  level  includes one or more major  unit operations:
     Level 1   Breaking for top size control  and for the removal of
               coarse refuse.
     Level 2   Coarse beneficiation in which  the larger coal particles
               (plus 3/8-inch)  are treated.  The treated large coal
               particles are recombined with  the smaller coal  particles
               to form the final product.
     Level 3   Coarse and fine size beneficiation  in which  all  of  the
               feed is wetted.   The plus 28M is beneficiated.   The very
               fine 28M x zero material is dewatered and either shipped
               with the cleaned coal or discarded.
     Level 4   Coarse, fine and very fine size beneficiation  in which  all of
               the feed is wetted and cleaned.  The 1/4-inch  x  zero fraction
               is generally dried to limit the moisture content.
                                      230

-------
     Level 5    Full beneficiation for optimal ash and sulfur rejection.
                This may  involve crushing  the  coal to  finer sizes  and
                producing a  number of coal  products, each  with  a  different
                ash  and  sulfur  content.
     Coal cleaning costs  are sensitive to  plant capacity,  plant complexity
 (level of cleaning) and  coal replacement costs.  Coal replacement costs are
 defined as the  cost of coal energy which must be discarded with the plant
 residue.
     The percentage of coal energy  (Btu's) recovered at modern coal prep-
 aration plants  is generally greater  than 90 percent.  With current high
 coal prices,  the annualized costs of physical coal preparation are more sen-
 sitive to coal  replacement costs than to plant complexity costs (Kilgroe,
 1977).  Removal of  finely distributed pyrite from coal entails a high degree
 of complexity.  A high degree of complexity is also required to recover a
 large fraction of the fine coal which at times has been discarded by steam
 coal plants.  Thus increased plant complexity for pyrite removal  is con-
 sistent with trends to minimize annual costs by increasing the recovery of
 fine coal.
     Total coal preparation costs exclusive of coal  replacement costs may
 range from $0.04 to $0.18/106 Btu (Buder, 1977, and Holt, 1978).  For an
 11,000 Btu/lb  raw coal costing $20/ton, coal replacement costs from a  plant
 with 95 percent Btu recovery would be $0.046/10  Btu.   Typical
 costs for coal preparation are shown in Table 3 (Buder, 1977).
STATUS AND COSTS OF FGD
     FGD systems are capable of removing more than 90 percent of the S02 from
the flue gas combustion products of coals with a wide range of sulfur content.
Annualized control costs increase with the increased amounts of sulfur removed.
In a given size boiler the costs of removing a fixed percentage of sulfur from
the  flue  gas  stream are  greater for  high  sulfur  coals  than  for  low  sulfur coals,
Although total control costs increase with the volume of flue gas cleaned,
normalized control costs for small to moderate sized boilers are greater than
for large boilers.  Figure 7 illustrates the effects of boiler size and sul-
                                      231

-------
                                      Table 3.  Capital and Operating Cost Suninary For Various Coal Preparation Plants*
to
Cleaning Level
Goal
Plant Construction Cost
Pre-construction and Owners Costs
Total Depreciable Capital Cost
Land Cost
Total Capital Costs
Annual Costsa»'J
Operating and Maintenance
Capital-Related0
Total Annual Costs
Return on Investment*1
Coal Processing Costs6
§ /ton dry product
$/106 Btu dry product
Return on Investment
$/ton dry product
$/106 Btu dry product
Total Costs
$/ton dry product
$/106 Btu dry product
1 *
Montana
(Rosebud)
6,800,000
1,907,000
0,707,000
150,000
8,857,000

904,851
993,767
1,898,618
797,130
0.584
0.026

0.245
0.011

0.829
0.037
2
W. Virginia
(Cedar Grove)
12,100,000
2,651,000
14,751,000
225,000
14,976,000

1,303,067
1,674,523
2,977,596
1,327,590
0.916
0.036

0.408
0.016

1.324
0.052
3
Colorado
(ttontose City)
20,300,000
5,405,559
25,785,559
330,000
26,115,559

1,799,678
2,735,864
4,535,542
2,359,400
1.396
0.053

0.723
0.028

2.119
0.081
4
Pennsylvania
(Lower
Kittanning)
24,100,000
6,562,000
30,662,000
900,000
31,562,000

2,579,111
3,380,276
5,959,387
2,840,580
1.834
0.064

0.874
0.031

2.708
0.095
5
Pennsylvania
(Upper
nreeport)
37,800,000
10,250,500
48,050,500
600,000
48,650,500

3,896,406
5,285,128
9,181,534
4,378,545
2.825
0.097

1.347
0.046

4.172
0.143
                  *     (Buder, 1977)
                  a    Corresponds  to  3.25 million tons per year  (dry) at 250 thirteen-hour annual operating days.
                  b    Excludes coal replacement costs.
                  c    Calculated at a 7:3 debt/equity ratio repaying debt with 9% 20-year bonds.
                  d    Calculated on equity with a before tax  return of 30% with no discounting.
                  e    All costs for Case 5 reflect the combined middlings and clean coal products.  The separation of  these costs will depend
                         on market  conditions.
                  f    Excluding thermal drying option which would add $0.45 per ton dry product to the processing cost, and $6,900,000 to  the
                         capital investment.

-------
    1.00 -
   0.90 -
P   0.80 -
ea




J   0.70 -


i
0>
    0.60 -
    0.50  -
a

g   0.40 -I

a
UJ


i   0.30 -

3



<   0.20 -
    0.10 -
                                                   LEGEND



                                                   90% FGD EFFICIENCY
               1.0    ZO     3.0    4.0    5.0   6.0




                POUNDS OF SO] PER MILLION BTU REMOVED
                                                        I


                                                       7.0
       FIGURE 7   ESTIMATED COST OF LIME/LIMESTONE FGD SYSTEM

                  (MeGLAMERY, I97S AND ISAACS, 1977)
                                     233

-------
fur removal  requirements on costs for a lime/limestone  FGD system.  These  costs
were developed with the use of FGD cost data generated  from computerized cost
models (McGlamery, 1975,and Isaacs,  1977).
     Partial scrubbing can be  used to reduce control  costs in  cases where
full scrubbing is not required.   In  partial  scrubbing,  part of the flue gas
stream bypasses the scrubber.   This  reduces  the amount of flue gas to be
treated and reduces or eliminates flue gas  reheat requirements.
     Partial scrubbing can be  used if an emission limit is specified and the
fuel sulfur value is low enough to permit treatment of less than the full
flue gas stream.   Alternatively,  if  a high  level  of sulfur control is re-
quired and precombustion sulfur removal  credit is given, then  sulfur removal
by coal cleaning techniques can be used to  permit the use of partial scrubbing.
Partial scrubbing, of course,  is  only advantageous if the reduced scrubber costs
more than offset  any higher costs attributed to the use of low sulfur coals or
coal cleaning.
     Only lime/limestone FGD systems are considered in this paper.

COST TRADE-OFFS

     With combinations of PCC + FGD, a number of cost trade-offs  must
be  considered  in  assessing the cost  impacts of clean coal use.
The use of cleaned coal for power generation may impact:  the design and
operation of the  boiler, the flue gas treatment equipment,or other components
of  the overall energy generation system.  Sulfur variability,which can have a
substantial impact on FGD costs,is discussed separately.

Boiler Cost Impacts
     Some of the most important considerations in the design of a boiler are
the characteristics of a coal  and its ash.   Reliable boiler operation  de-
pends  on  the application of design techniques  utilized  to minimize  slag-
ging, fouling, and corrosion problems.  These problems in a large measure
directly affect boiler availability.  Of the reasons advanced for the use of
cleaned coal, perhaps the greatest single benefit to be obtained  (other than
                                      234

-------
the control  of emissions) is in the area  of fireside  performance.   Fireside
problems are responsible for many coal  r'ired operational  difficulties  result-
ing in both  forced and scheduled outages.   They  significantly  affect the  cost
of boiler operation and maintenance, the capacity factor, and (in
the case of  utilities) the availability of the generating facility.  By modi-
fying the coal characteristics which contribute  to  these  problems,  coal
cleaning can favorably affect the economical  use of coal.   Coal  characteris-
tics vary widely,  however, and different  methods of cleaning can have  diverse
effects on the properties of coal  ash and sulfur content.   For this reason,
the effects  of coal cleaning on boiler  operation will,  in a given circum-
stance, depend on  both the cleaning method used and the original character-
istics  of the coal.
     The net effect of coal  cleaning on the operating and maintenance  costs
of boilers  (as related to slagging, fouling, and corrosion) is difficult to
quantify because  of the many variables involved.  In a previous
study on coal  cleaning and scrubbing (Hoffman, 1976), boiler maintenance  cost
savings were postulated to be related to  the additive reduction  in  the sulfur
and ash content of the coal  (see Table  4).   These cost  savings were based
on  TVA  studies on  the  effects  of coal  quality on the operation and mainte-
nance of large central station boilers  (Holmes,  1969).

                                 TABLE 4.*
     THE EFFECT OF ASH AND SULFUR CONTENT ON BOILER MAINTENANCE  COSTS

                Additive Reduction                 Maintenance Cost
             in Ash and Sulfur, percent          Savings, $/ton Coal

                    >15                                   0.33
                     12 - 15                              0.30
                      9 - 12                              0.27
                      7-9                                0.24
                      5 - 7                                0.20
                      3-5                                0.17
                      2 - 3                                0.13

            * (Hoffman,  1976)
                                     235

-------
        Recent papers by Cole and Phillips postulate a number of cost penalties
   which may be attributed to the total  ash and sulfur content of coal  (Cole, 1978,
   and Phillips, 1979).   These penalties include increased maintenance  costs, loss
   of peaking capacity and reduction in  plant availability.  Total  cost penalties
   at 15 and 17.5 percent ash plus sulfur are given as $0.38/ton and $0.75 ton
   (see Table 5).  Above 17.5 percent ash and sulfur the postulated total  cost
   penalties rise exponentially, reaching a value of $6.41/ton at 25 percent ash and
   sulfur.

                                   TABLE 5.*
       COST PENALTIES ASSOCIATED WITH ASH AND SULFUR CONTENT OF THE COAL
Ash
Content,
percent
10.5
12.5
14.5
16.5
18.5
20.5
Sulfur
Content,
percent
2.0
2.5
3.0
3.5
4.0
4.5
Total
A&S
percent
12.5
15.0
17.5
20.0
22.5
25.0

Maintenance
Costs
0
0.38
0.75
1.13
1.50
1.88
Cost Penalty,
Peaking
Capacity
0
0
0
0.19
0.23
0.21
$/ton Coal
Rated
Capacity
0
0
0
1.08
2.08
3.00
Fired
Plant
Availability
0
0
0
0.47
0.91
1.32


Total
0
0.38
0.75
2.87
4.72
6.41
* (Phillips, 1979)

        The amount of data which can be used to correlate the effects of PCC and
   boiler operating and maintenance costs is extremely limited.   Studies are now
   underway by EPA and EPRI to develop improved correlations between coal quality
   and operating costs.  Until additional information becomes available the studies
   cited above can be used to estimate the range of boiler related cost benefits
   which may be available with improved coal quality.
                                       236

-------
FGD Cost Impacts
     If the full flue gas stream is scrubbed then it is necessary to reheat it
to avoid severe stack corrosion problems and ensure proper dispersion of the
stack plume into the atmosphere.  The use of partial scrubbing can
reduce or eliminate reheater capital and operating costs.   In
addition, after reheat requirements have been met, the use of further bypass
will lower costs by reducing the volume of flue gases scrubbed.  A study of
energy requirements of a limestone FGD system concluded that for "any fixed
set of coal and plant characteristics, FGD energy use is minimized by operating
the scrubber at high efficiency (90-93%) while bypassing as much flue gas as
permitted by the applicable emission standard'.1  An economic analysis of the
base case plant design further indicated that FGD capital costs as well  as
operating costs were reduced by partial bypass (Rubin, 1978).
     The amount of partial scrubbing which can be made available by coal
cleaning depends upon the sulfur removal efficiencies of the coal cleaning
and scrubbing processes and upon the applicable emission regulations which
must be met.  Figure 8 presents the amounts of flue gas which may be bypassed
for a number of coal cleaning sulfur reduction efficiencies and total sulfur
emission reduction requirements (a 90 percent FGD sulfur removal efficiency
is assumed).  For a 90 percent sulfur removal standard the available bypass
would probably not be sufficient to meet total reheat requirements, even at
high coal cleaning sulfur efficiencies.  If the scrubber efficiencies are
raised to 95 percent and the coal cleaning sulfur-removed efficiencies are
30 percent or larger, then the amount of  bypass available may be sufficient
for reheat requirements (see Figure 9).
     Emission regulations which contain an emission limit below which addi-
tional control is not required can use naturally occurring low sulfur coals
or cleaned coals to permit a high amount of bypass.    For example:  cleaning
a  2.0 Ib SO/2/10  Btu  coal  to remove 20 percent of the coal  sulfur would
                                      237

-------
                                   90% SCRUBBER EFFICIENCY
  100 -i
   90 -
   80 -
   70 -
LLJ
o
en
ffi

01
   50 -
   40 -
   30 -
   20 -
                                    EACH CURVE REPRESENTS PERCENT

                                    TOTAL SULFUR REDUCTION

                                    REQUIRED TO MEET REGULATION
                                       75%
   10 —
                   10
                FIGURES
     20           30           40


     SULFUR REMOVED BY PCC (PERCENT)
                                                                    50
                                                                                60
                                                                                            70
EFFECT OF PCC SULFUR REMOVAL EFFICIENCY ON ALLOWABLE
FGD BY-PASS FOR 90% SCRUBBER EFFICIENCY
                                            238

-------
   20  -
    15  -
LEGEND



EACH CURVE REPRESENTS

PERCENT FGD EFFICIENCY
p   10  H
z

3
oe
Ul
St
a>



I
a
UJ

CO
4
    5  -
                                                                      99%
                                                                      95%
                                                                      90%
                    I            i            I            I            I


                   10          20          30          40           50


                               SULFUR REMOVED BV PCC (PERCENT)



               FIGURE 9   EFFECT OF PCC AND FGD REMOVAL EFFICIENCIES ON ALLOWABLE

                         FGD BY-PASS FOR 90% SULFUR REDUCTION STANDARD
                                                                   I


                                                                   60
 I


70
                                         239

-------
allow for a 23.6 percent bypass where a 90 percent efficiency scrubber
was used to meet a 0.5 Ib S02/10  Btu emission limit (75 percent total
sulfur control).
     The potential energy cost saving available by an allowable bypass may
be estimated by the results of Rubin's studies on the energy requirements of
a limestone FGD system.   Rubin found that for the range of parameters tested,
the total FGD energy requirement was equivalent to between 2.5 and 6.1 percent of
the total power plant energy output (or input) when 100 percent .of the flue
gas was treated in the FGD system.  Sensitivity analyses for a 3.5 percent
sulfur coal used in the study showed that treating the entire flue gas
stream required 10 to 30 percent more energy to achieve the same S02 emission
standard than a system with partial by pass (Rubin, 1978).
     As a first approximation one can conclude that the energy saving avail-
able from bypass  could range from 0.25 to 1.8 percent of the coal  energy input
to the boiler.    For a coal  costing $1.00/10  Btu this would represent a cost
savings of 2.5 to 18 mills/million Btu of heat input.
Sulfur Variability Cost Impacts
     Fuel sulfur variability can have a large impact on the costs of S02 emis-
sion control.  Many current regulations specify that sulfur emissions are
"never" to exceed a stated emission limit.   Compliance with these regulations
is generally determined by one or more emission tests conducted  over a speci-
fic time period.  This time period provides an "averaging time"  over which
average emissions cannot exceed the emission limit.
     The amount of coal burned in a given size boiler with a specified aver-
aging time is the characteristic coal "lot size" for that boiler.  Previous
studies indicate that coal sulfur variability probably increases with decreas-
ing lot size (Nelson, 1977, and Versar, 1979).  Compliance with  a given regu-
lation is therefore more difficult for small boilers than for large boilers.
     The sulfur content of coal varies between coal regions, coal seams, and
locally within each coal mine.  The extreme nature of coal sulfur variability
is illustrated by Figure 10, which shows the sulfur content of Helen
Mine coal for successive mining days over a 5 year period.  As suggested by
Figure 9,sulfur variation over short and long time periods is of concern.
                                     240

-------
N>
•O
                              0.6Q
                                                                                                                                          1.600
                                                                            SUCCESSIVE MINING DAYS
                                              FIGURE
                                                         POUNDS OF SULFUR PER MILLION BTU FOR SUCCESSIVE MINING DAYS FOR HELEN MINE,
                                                         JANUARY 1970 THROUGH DECEMBER 1975. (THOMAS. 1978)

-------
In designing a sulfur emission control  system,provisions must be made for
compliance with the emission averaging  time and the maximum mean sulfur con-
tent of the coal  which is to be used.
     Data on coal sulfur variability show  distributions skewed toward the
the higher coal sulfur levels (Nelson,  1977).   This skewness can be charac-
terized by either  log-normal or inverted gamma distributions.  The skewed distri-
bution is especially apparent for  small  lot sizes:  bore hole samples are the
smallest.  As the lot sizes increase the skewness  is decreased and the distribu-
tion can be approximated by a gaussian  or normal distributions.*
                                                                   2
     Coal sulfur values exhibit different variances over hourly (0.  ),
          222
weekly (a  ), monthly (a  ), and yearly (a   )  time  periods.   The total  sulfur
variability will  be composed of the short and  long  term variance as described
by the equation:
                2     2 .    2,    2,    2,    2
               CT  - ah  + CTd  + aw  + am  + ay
     The time related variances for coal  from  a given mining reserve are also
related to the manner in which the coal  is  mined or used;  i.e., the variation
is spatially correlated.  If mining is  conducted along sulfur  isolines,
daily  and  weekly variances  will be  small.   If coal  is mined across isolines,
short  term variances may  be  large.
     Emission regulations with long averaging  times (30 days or longer)  can
 mitigate the effects of short term sulfur  variability.   Longer term variability,
which  can  be  characterized  by the  change in the monthly or yearly mean sulfur
values,  must  be  accounted for in the design of the emission control  system.
     It is convenient to characterize the variability of coal  sulfur values  and
stack emission values in terms of  a mean value (y)  and a standard deviation  (a).
The extremes in the coal sulfur values  and  stack sulfur emission values (see
Figure 10) can then be described by equations of the  form (Kilgroe,  1979):
               Coal  sulfur  extremes = y  (1 +_ a RSD  )
                                       c       c   c
               Stack sulfur emission extremes = u   (1 +_ a RSD  )
                                                 o       w   o
where RSD = a/y is the coefficient of variation or  the relative standard devia-
tion.  The product,a RSD,is defined as  the coal variability factor; a RSD  is
                                                                     C   I*
* Classical statistics do not adequately describe the variance properties
  of coal.  Geostatistics, a recent development which is capable of describing
  distributions with a spatial correlation of properties,appears to offer a
  better technique for characterizing coal sulfur variability (Thomas, 1978).

                                     242

-------
the coal variability factor and a RSDS is the scrubber variability factor.
The coefficient "a" is used as needed to express a confidence interval re-
lating to a given lot size or averaging time.
     Design constraints placed on the FGD by sulfur variability may be eval-
uated in terms of the mean sulfur reduction requirement of the emission
standard (ri).  The percentage removal requirement must hold for all cases,
the extremes as well as the mean.  The required removal conditions at the
mean are:
                         fi = 1 - y
     The removal efficiency at the extremes provides bounds for all other
cases.  The four extreme cases illustrated in Figure 11 can be represented
by the equation:
              n = 1 - v. (1 + a RSDJ = 1 - (1-n) (1 + a RSDJ
                       o    ~~  3 _ b                 ~—  J _ b
                      uc (1 ±acRSDc)             (1 ±acRSDc)

     Table 6 presents data showing the required efficiency at extreme
conditions corresponding to selected coal and sulfur variability factors.
All cases correspond to a mean FGD sulfur removal requirement of 85 percent.
Case III is not presented because the required efficiency at these extremes
is  less than those at mean conditions.
     In the worst case, Case IV, the effects of coal and scrubber variability
are compounded.  At a RSD  = 0.10 and a RSD  = 0.8 the required design removal
                     C   C             o   o
efficiency at the extreme is 97.3 percent.  For values of coal sulfur varia-
bility which may be expected from uncleaned or unmixed coals (a RSD  = 0.5)
                                                               c   c
and for the upper range of scrubber sulfur variability (a RSD  = 0.8) the
required design removal efficiency at the extreme is 98.0 percent.
     Mixing and preparation may be used to reduce variations in coal
properties.  Mixing of coal to form a more homogeneous product
will occur during mining, handling, blending, and preparation.  During mining,
coal is often removed from multiple mining faces during the same time period.
Coal from these locations is combined to form a single run-of-mine (ROM) pro-
duct.   Coal from each of the faces may have a different mean sulfur value and
                                      243

-------
  CASE I




 11-1-
                                          • *. * *,",
                                                         COAL
                                                         SCRUBBER
— — V— —
CASE II \
11
t
V

/
/
J
                                           Me-'o'e
                                           "i-S"f
                                                        COAL
                                                        SCRUBBER
CASE Ml
CASE IV
                           \
                        \

                                           Me
                                           ^ + .effc
                                           ."i
                                                        COAL
                                                        SCRUBBER
                                                        COAL
                                                        SCRUBBER
          FIGURE 11.   COAL AND FGO SULFUR VARIABILITY CASES
                        244

-------
   TABLE 6.   EFFECT OF SULFUR VARIABILITY FACTORS UPON VARIABLE EFFICIENCY
                                                                       (a)
Extreme^3'
Case I
                           Coal  Variability
                               Factor
Case II
Case IV
                             0.30
                                0.50
                             0.10
                                0.30
                             0.10
                                0.30
                                0.50
FGD Variability
Factor
(asRSDs)
0.00
0.10
0.20
0.10
0.40
0.20
0.40
0.80
0.40
0.20
0.40
0.80
0.10
0.20
0.40
0.80
0.10
0.80
Required
Efficiency at
Extreme (percent)
88.5
87.3
86.2
89.0
86.0
86.7
90.0
96.7
87.1
89.1
91.8
97.3
89.6
90.8
93.1
97.7
91.0
98.0.
(a)
(b)
For mean removal  efficiency of 85 percent
Case III not presented because required efficiency at these
extremes less than those at mean conditions.
                                     245

-------
a different degree of sulfur variability.  The mining methods and the way in
which coal is combined from the different faces will have a significant effect
on the mean sulfur values and the sulfur variability.
     The purposeful mixing of coal from two or more mines to reduce variability
or modify average coal properties can also be accomplished by "blending."
     Mixing during handling, storage,and transportation will likely result in a
more homogeneous product with reduced variability.
     Data which provide  quantitative information correlating the reduction in
sulfur variability to the individual mining, handling and blending operations
are not currently available.  However data sets from coal suppliers and util-
ity sources indicate that RSDs for large lot sizes of ROM coal  may
range from approximately 10 to 30 percent (Nelson, 1977).  Data from coal prep-
aration plants suggest that the product coal RSDs may normally be in the range
of 5 to 10 percent (Versar, 1979).
     Limited data are available on the variability of sulfur emissions from
FGD scrubbers on utility boilers (Kelly, 1978).  The RSD values for three FGD
units evaluated by OAQPS for 3 and 24 hour averaging times normally fell in
the range of 20 to 45 percent.  A fourth highly efficient FGD unit (approximately
95 percent removal efficiency) exhibited outlet RSD values from 70 to 104. These high
RSD values resulted primarily from extremely low mean emission values which
ranged from 0.233 to 0.273 Ib S02/106 Btu.  These data suggest a probable range
of FGD sulfur variability factors ranging from 40 to 100 (value of ag = 2.0 is
assumed).  Improved FGD controls should substantially reduce this variability.
     The above data suggest that FGD capital and operating costs will depend
on the coal sulfur variability, the emission regulation averaging time,and the
ability of the FGD system to control sulfur outlet concentration , whether caused
by coal sulfur variances or the basic variance of FGD process variables.  Given
the constraints of the regulation averaging time, variances in the coal sulfur
will probably cause a larger cost impact than those cost factors related to
FGD process variables (Kilgroe, 1979).  Many FGD cost components are related
to the total sulfur which must be removed from the flue gas stream (y  -y )-
                                                                     \+   j
As the peak sulfur levels increase,the FGD sulfur removal capacity must be
increased to account for the higher sulfur removal requirements.  This
increase,expressed as the ratio of costs with variability to the costs without
                                     246

-------
variability,may be approximated by the expression:

                     £= 1 (1 + acRSDc)  -(1-n) (1 - asRSDs)
                     C$  n"                  n

     Solution of this equation for a mean sulfur removal requirement of 90 per-
cent yields the fact that a 20 percent increase in the maximum peak coal sulfur
levels will result in an increase of approximately 20 percent in the FGD costs
related to amount of sulfur removed.  An 80 percent increase in peak coal  sulfur
values would raise the sulfur related FGD costs approximately 80 percent
(Kilgroe, 1979).
     The cost penalties for FGD designs which incorporate a sufficient safety
margin to account for both short and long term sulfur variabilities may be
quite high.  The use of coal  blending or coal preparation may provide cost
effective alternatives.  Additional studies are obviously needed to evaluate
these trade-offs.

ESP Cost Impacts
     The removal of fly ash from flue gas is typically done by 99.5 percent effi-
cient ESP units, usually placed downstream of the air preheaters near the
boilers.  The ash and sulfur contents of coals are major influences on ESP
capital and operating costs.   The ESP size and collection efficiency are
affected by the coal sulfur content and inlet ash loading.   Increases in
ESP size requirements which may result from coal desulfurization may be off-
set by decreased ash disposal costs or increased efficiencies resulting from
reduced fly ash loading.  The costs of controlling particulate emissions from
cleaned and uncleaned coals may not be substantially different for most cases.
Studies to identify the sensitivity of particle collection to coal cleaning
effects on a number of representative coals are needed to confirm this postu-
lation.

System Cost Impacts
     The primary system benefits which may accrue from the use of cleaned
coal include reduced transportation costs, reduced pulverizer costs, re-
duced boiler ash disposal costs,and reduced mine labor costs.

                                     247

-------
     Transportation and mine labor cost  reductions occur if the coal is
cleaned at the mine site.  Cleaning reduces the ash content of the coal and
increases its calorific value.   This reduces the weight of coal needed to
meet boiler energy requirements.  A reduction in coal shipped from the mine
site also reduces the amount paid to the UMW Pension and Benefit Trust Fund
($1.38/ton coal shipped to the consumer).
     Pulverizer operating costs will generally be reduced in proportion to
reduction in coal weight burned.  While removal of the coal  ash by clean-
ing may provide benefits beyond those attributable to the reduction coal
weight pulverized, these extra benefits cannot be generally specified.
     Reductions in transportation, ash disposal, pulverizer, and union  trust
fund costs, as estimated by Hoffman, are presented in Table 7 (Hoffman, 1976).
The total average cost savings for the 12 mine and power plant site pairs
evaluated were $0.56/ton.  Transportation and ash disposal savings were approx-
imately 4 to 10 times higher than pulverizer and union trust  fund  savings.  The
maximum total savings of $1.09/ton (approximately $0.04/10  Btu ) would provide
a substantial credit against the costs of coal cleaning.
                         TABLE 7.  SYSTEM COST BENEFITS

                                            System Cost Benefits ($/ton)
                                         Minimum      Maximum      Average
     Transportation Savings 'b'            0.15        0.52         0.29
     Ash Disposal Savings                  0.10        0.36         0.19
     Pulverizer Savings                    0.01        0.04         0.03
     UMW Trust Fund Savings ^            0.02        0.17         0.05

                   Total                   0.28        1.09         0.56

     (a) Hoffman, 1976
     * ' Average transportation distance for 12 mine power
         plant site pairs was 436 miles.
     ^ Based on costs of $0.74
                                      248

-------
COST COMPARISONS

     Economic comparisons of alternative pollution control  options are complex.
Factors unique to a given site often determine which option is the most cost
effective.
     Several comprehensive studies under the sponsorship of EPA, DOE, and EPRI
are now in  progress to evaluate the relative costs of FGD + PCC and FGD as
options for complying with S02 emission regulations.  The results of these
studies are not expected to be available for several months.   However, one may
in some part anticipate the results of the studies by using information available
in the literature to identify the instances where combinations of coal cleaning
and FGD are likely to be economically competitive.  It is emphasized that the
results are only semi-quantitative in that they indicate the relative importance
of major control system design factors which influence the costs of compliance
with a given regulation.  To assess the actual control costs and determine the
nost cost effective means of compliance require detailed studies which account
for site specific factors which influence the costs of compliance.

Analysis Method
     Cost comparisons for controlling SOp emissions were made using existing
data on the costs of FGD and coal cleaning (see Table 8).  These comparisons
were limited to the  control of S0? emissions from 500 MW boilers.   The
comparisons considered four different coal sulfur levels and two or three
emission regulations.  The coals were assumed to have compositions and proper-
ties similar to "average" coals from the Northern Appalachian, Alabama,
Eastern Midwestern and Western coal regions as specified in the U.S. Bureau of
Mines Publication on the Sulfur Reduction Potential of U.S. Coals (Cavallaro,
1976).  Sulfur emission regulations considered included those requiring
emission limits of 2.5, 2.0, 1.2, 0.8 and 0.5 Ib S02/106  Btu or (alternatively) a
90 percent  sulfur reduction .
     FGD costs were determined from Figure 7 by use of the FGD sulfur removal
requirement and the equivalent scrubber capacity in MW.  All FGD units were
assumed to  operate at 90 percent efficiency.  For the PCC + FGD cases the
equivalent  scrubbing capacity was determined by calculation of the allowable
bypass  using the clean coal, sulfur level and required sulfur emission limit.
                                     249

-------
                                                  Table  8.   PCC + FGD and PCC COST COMPARISONS
I.    A.   Coal  Region
     B.   Raw Coal  Sulfur Level,  Ib SO,
     C.   Allowable Emission Limit, Ib'
     D.   Sulfur Removal  Required,  %

II.   FGD Control  Costs
N5
Oi
O
                                                Btu
so2/io
                                                   Btu
     A.
     B.
              FGD Sulfur Removal Requirement, Ib S02/10
              Cost of 500 MW FGD at 90%  Removal
              Efficiency, $/106  Btu
                                                    Btu
III.  PCC + FGD Control  Costs
     A.
         Sulfur To Be Removed by FGD,  lb"S02/10u Btu
              Sulfur Removal by PCC^, Ib SO-/106 Btu
                                       ~*r\  1 U PA  /i r»0 r
          C.
          D.
          E.

          F.
          G.
          H.
          I.
          J.
         FGD Sulfur Removal  Requirement,  %
         Equivalent FGD Size at 90% Efficiency,  MW
         FGD Costs  at  Reduced Size  and S02 Removal
         Requirements, $/10   Btu
         Preparation Cost  Range,  $/10  Btu
         Raw Coal Costs, $/10 Btu
         Coal  Replacement  Costs,  $10  Btu
         Total  PCC  Costs,  $/10  Btu
         PCC + FGD  Control Costs,  Ib S02/10  Btu
                                                                   Emission Control  Level
                                                                 1           2           :
                                                                                                   Emission Control Level
                                                                Northern Appalachian
                                                           4.8         4.8         4.8
                                                           2.0         1.2         0.48
                                                          47.9         75.0        90.0
2.3
0.28
2.1
0.7
25.9
144
0.09
0.07-0.18
1.00
0.08
0.15-0.26
0.24-0.35
3.6
0.34
2.1
1.5
55.6
309
0.18
0.07-0.18
1.00
0.08
0.15-0.26
0.33-0.44
4.32
0.37
2.1
2.22
82.2
457
0.26
0.07-0.18
1.00
0.08
0.15-0.26
0.41-0.52
     IV.  Required Coal Cleaning Cost Benefits to Break    (0.04)-0.07(-0.01)-0.10   0.04-0.15
          Ever*15', $/106 Btu
1

2.0
1.2
40.0
0.4
0.21
0.3
0.5
29.4
163
0.10
0.07-0.18
1.00
0.04
0.11-0.22
0.21-0.32
2
Alabama
2.0
0.8
60.0
0.8
0.220.27
0.3
0.9
52.9
294
0.16
0.07-0.18
1.00
0.04
0.11-0.22
0.27-0.38
3

2.0
0.20
90.0
1.44

0.3
1.5
88.2
490
0.26
0.07-0.18
1.00
0.04
0.11-0.22
0.38-0.48
                                                                                                0-0.12   0.05-0.16   0.11-0.21
     (a)  Cleaned at 3/8 inch top size and 1.6 s.g.
     (b)  Line III-J Costs - Line II-B Costs

-------
                                                              Table 8.  (continued)
                                                                   Emission Control  Level
Ln

I. A.
B.
C.
D.
II. FGD
A.

B.

Coal Region
Raw Coal Sulfur Level, Ib S0?/10 Btu
Allowable Emission Limit, Ib S02/106 Btu
Sulfur Removal Required, %
Control Costs
FGD Sulfur Removal Requirements, Ib S0?/106
Btu
Cost of 500 MW at 90 % Removal Efficiency,
1
2
3
Eastern Midwest
6.5
2.5
38.5

4.0

0.35
6.5
1.2
81.5

5.3

0.40
6.5
0.65
90.0

5.85

0.43
              $/10   Btu
     III. PCC + FGD Control Costs
IV.   Requi
                 d Coal Cleaning Cost Benefits to Break
                         Btu
                                                     (-0.09)-0.02 (-O.l)-O.lO
                                                                                        0-0.11
     (a)  Cleaned at 3/8inch top size and 1.6 s.g.
     (b)  Line III-J Costs - Line II-B Costs
                                                                                                        Emission Control Level
                                                                                                              i         ;   2
                                                                                                                      Western
                                                                                                                  1.1          1.1
                                                                                                                  0.5          0.11
                                                                                                                 54.5        90.0
                                                                                                                  0.6

                                                                                                                  0.22
                                                                                                                        0.99

                                                                                                                        0.230
A.
B.

C.
D.
E.

F.
G.
H.
I.
J.
Sulfur Removal by PCCvaMb S02/10° Btu
Sulfur To Be Removed by FGD, IB S0?/
10 Btu
FGD Sulfur Removal Requirement, %
Equivalent FHD Size at 90% Efficiency, MW
FGD Costs at Reduced Size and S02 Removal
Requirements, $/ll Btu
Preparation Cost Range, $/10 Btu
Raw Coal Costs, $/10 Btu
Coal Replacement Costs, $/10 Btu
Total PCC Costs, $/106 Btu
PCC + FGD Control Costs, Ib SO-/106 Btu
2.3
1.7

40.5
225

0.14
0.07-0.18
1,00
0.05
0.12-0.23
0.26-0.37
2.3
3.0

71.4
397

0.27
.07-0.18
1.00
0.05
0.12-0.23
0.39-0.50
2.3
3.55

84.5
469

0.31
0.7-0.18
1.00
.05
0.12-0.23
0.43-0.54
0.2
0.4

44.4
245

0.13
.07-0.18
0.65
.02
0.09-0.20
0.22-0.33
0.2
0.79

87.8
488

0.22
.07-0.18
0.65
.02
0.09-0.20
0.31-0.42
0-0.11   0.08-0.19

-------
     A range of coal preparation costs was assumed.  The lower costs are
applicable to coals which could be easily cleaned in a relatively simple
plant configuration.  The higher costs correspond to the higher levels of
cleaning which are required for increased sulfur removal and Btu recovery.
Sulfur removal and Btu enhancement equivalent to crushing 3/8 inch top
size and separation at 1.6 specific gravity were assumed for all cases.  Coal
replacement costs were calculated for each of the coals assuming Btu recov-
eries equivalent to the average coal for each region.  Coal replacement costs
were assumed to be independent of the plant operating and maintenance costs;
i.e., they were held constant over the range of preparation costs.
     The results of the analyses were expressed in terms of the non-FGD coal
cleaning cost benefits which would be required to make PCC + FGD cost competi-
tive with FGD; i.e.,

                 ^Col? Benemsanin9   i   Cost  -Cost 
     For some conditions the use of cleaned coals reduces FGD costs  to
the point where the reduced FGD costs are equal to the coal cleaning costs.
This is the "break-even" point at which no other (non-FGD) cost benefits are
needed to make the costs of both options equal.
     FGD cost offsets  from using cleaned coal result from two factors:  a
reduction in the volume of flue gases treated and a reduction in the sulfur
which must be removed from the treated volume.  Coal cleaning can reduce the
amount of sulfur which must be removed and can at the same time provide for
allowable bypass  to reduce the volume treated.
     For high sulfur coals the break-even point is approached as the percentage
of bypass  is increased.  For low sulfur coals the potential sulfur reduction
savings may not be sufficient to offset  the cost of cleaning,  even under high
bypass  conditions.  Alternatively, regulations which require sulfur removal
requirements in excess of 90 percent may not allow for enough-bypass  (volume)
cost reductions to offset  the costs of cleaning.
     The most likely candidates for PCC + FGD are those applications which
use high sulfur coals and which do not require total sulfur removals greater
than 90 percent.  If the revised NSPS require 90 percent sulfur removal and
do not specify an emission floor, then PCC + FGD would not be competitive
                                     252

-------
with FGD unless there are substantial  non-FGD cost benefits associated with
the use of cleaned coal.
     Non-FGD cost benefits may range from $1.00/ton to $6.00/ton (approxi-
mately $0.04/106 Btu to $0.24/106 Btu).  If actua] cost benefits are at the
high end of this cost range, then PCC  + FGD may be the most cost effective
method of complying with S02 emission  regulations.  If the cost benefits are
at the lower end of this cost range, it may not be cost effective to clean
coals prior to scrubbing.

CONCLUSIONS

     Physical coal cleaning is now used in a limited number of cases to remove
sulfur for compliance-with SIP S02 emission regulations.   Tightening and strict
enforcement of the SIP regulations may increase the demand for desulfurization
of high sulfur Midwestern and Eastern coals by cleaning.
     The demand for cleaned coals for  compliance with current NSPS for utility
boilers is not expected to increase.  Boilers subject, to this regulation have
already selected a control method consisting of low sulfur coal, coal  cleaning,
or FGD.  All new utility boilers will  be subject to the revised NSPS.
     The percentage reduction specifications of the revised NSPS for utility
boilers will essentially preclude the  use of coal  cleaning as a sole method
for complying with the S02 control reouirements of these regulations.   Combinations
of coal cleaning and FGD as a compliance technique will only be used where the
combined control approach is more cost effective than FGD alone or where FGD
cannot achieve the emission requirements because of an unusually high coal sul-
fur content.  It is also possible that combinations of coal cleaning and FGD
will be required to achieve LAER requirements in non-attainment areas or BACT
requirements in clean air areas.
     It is probable that industrial boiler NSPS which are now being considered by
EPA will permit the use of cleaned coals as an S02 emission control method in
small boilers.  Large industrial boilers may find it cost effective to use com-
binations of coal  cleaning  and  FGD  for compliance with S02 emission regulations.
     The use of PCC + FGD will be the most cost effective method of complying
with emission regulations if the reduction in FGD costs and cost benefits not
related to S02 emission control are greater than the costs of cleaning.
                                      253

-------
     Reductions in FGD costs by PCC can result from a reduction in the volume
of flue gas treated or the amount of sulfur removed from the flue gas stream.
Reductions in fuel sulfur variability by PCC can lower design safety margins
needed to ensure compliance for all fuel sulfur values.
     Utility boilers which use high sulfur coals and which require sulfur
removals less than 90 percent are likely candidates for PCC + FGD.  If the
revised NSPS for utility boilers require 90 percent sulfur removal and do
not specify an emission floor, then PCC + FGD may not be competitive with
FGD unless there are substantial non-FGD cost benefits associated with
cleaning.
     The range of applications for PCC + FGD in small  non-base-loaded
utility boilers and industrial boilers may be different from those cited for
base-loaded utility boilers.  The differentials between PCC and FGD costs for
these smaller units may result in different optimal solutions for the range of
alternative permissible control strategies.
                                     254

-------
CONVERSION FACTORS

ton = 0.907 metric tons
Ib = 0.436 kg
Btu = 1055.6 Joule
Btu/lb = 2326 Joule/kg
in. = 2.54 cm
°C = 5/9 x (°F-32)
Ib./in. = 0.07 kg/cm2
Ib S02/106 Btu = 430 ng S02/Joule
                                     255

-------
                               REFERENCES


Buder, M., Environmental  Control  Implications  of Generating Electric Power from
     Coal, 1977 Technology Status  Report, Appendix  A,  Part  1,  Coal  Preparation
     and Cleaning Assessment Study:   Report  for Department  of  Energy,  Bechtel
     Corp., December 1977.

Cavallaro, J.  A., M. T.  Johnston  and  A.  W. Deurbrouck,  "Sulfur Reduction
     Potential  of U. S.  Coals:  A Revised Report of Investigations,"
     EPA-600/2-76-091  (NTIS  PB 252965) or Bureau of Mines RI 8118,
     Washington,  D.C., April  1976.

Cole, R. M., Economics of Coal  Cleaning  and  Flue Gas Desulfurization for
     Compliance with Revised NSPS  for Utility  Boilers,  Proceedings  of  EPA
     Symposium "Coal Cleaning to  Achieve  Energy and Environmental Goals,"
     Hollywood, Florida,  September 1978  (to  be  published).

U. S. Environmental  Protection Agency, 1978a,  Code  of  Federal  Regulations  40,
     Protection of Environment, September 19,  1978,  Part  V,  Electric Utility
     Steam Generating  Units, Proposed Standards of  Performance  and  Announcement
     of Public Hearing on Proposed Standards,  pp. 42154-42184.

U. S. Environmental  Protection Agency, 1978b,  Code  of  Federal  Regulations  40,
     Protection of Environment, December 8,  1978, Part  V, Standards  of
     Performance for New Stationary Steam Generating- Units,  Additional
     Information on pp.  57834-57859.

Hall, E. H., et al.,  The Use of  Coal  Cleaning  for  Compliance  with  SCL
     Emission  Regulations, Draft  Special  Report, EPA Contract  68-02-2T63,
     Battelle  Columbus Laboratories,  Columbus,  Ohio, February  1979.

Hoffman, L., S. J. Aresco, and E.  C. Holt, Jr.,  "Engineering/Economic Analyses
     of Coal Preparation  with S02  Cleanup Processes  for Keeping  High Sulfur Coals
     in the Energy Market," The Hoffman-Hunter  Corporation  for U. S. Bureau
     of Mines,  Contract  J0155171,  November 1976.

Holmes, J. G.,  Jr. , The  Effect  of  Coal Quality  on the  Operation  and  Main-
     tenance of Large  Central Station Boilers.   Paper  for presentation  at
     Annual Meeting of the American Institute  of Mining,  Metallurgical
     and Petroleum Engineers, Washington, D. C.  February 16-20,  1969.

Holt, E. C., Jr.,  An  Engineering/Economic Analysis  of  Coal  Preparation Plant
     Operation  and Cost,  EPA-600/7-78-124, (NTIS PB 285251), The Hoffman-
     Munter Corp., Silver Spring,  Maryland, July 1978.


Issacs, G. A.,  Personal  Communication, PEDCo Environmental,  July 26, 1977-


Kelly, W. E.,  et al.,   Air Pollution  Emission  Test, Volume  I:   First Interim
     Report, Continuous  Sulfur Dioxide Monitoring at Steam  Generators,  U.  S.
     Environmental  Protection Agency, Office of Air Quality Planning and
     Standards, EMB Report No.  77SPP23A,  Research triangle  Park, N.C.,  August
     1978.
                                      256

-------
Kilgroe, J. D., Coal  Cleaning for Compliance  with S02 Emission Regulations,
     Proceedings of NCA/BCR Coal  Conference  and Expo IV,  Louisville,  Ky.,
     October 1977.

Kilgroe, J. D., The Effects of Sulfur Variability on Control  Technology
     Requirements, Unpublished Technical  Notes, Industrial  Environmental
     Research Laboratory, Research Triangle  Park, N.C.  February 1979.

McGlamery, G. G., et al., "Detailed Cost  Estimates for Advanced Effluent
     Desulfurization Processes,"   TVA report  for Environmental  Protection
     Agency, EPA-600/2-75-006, (NTIS PB 242541), January 1975.

Nelson, A. C., "Preliminary Evaluation of Sulfur Variability  in Low-Sulfur
     Coals from Selected Mines,"  for U. S. Environmental  Protection Agency,
     Contract 68-02-1321, task 41, July 1977.

Phillips, P. J. and R. M. Cole, Economic  Penalties Attributable to Ash
     Content of Steam Coals, Paper prepared for Coal Utilization Symposium,
     AIME Annual Meeting, New Orleans, February 1979.

Rubin, E. S. and D. G. Nguyen, Energy Requirements of a Limestone FGD System,
     Journal of Air Pollution Control Association, Volume  28,  No. 12,
     December 1978.

Versar, Inc., Sulfur Reduction Data from  Commercial  Physical  Coal Cleaning
     Plants and Analysis of Product  Sulfur Variability, Draft  Report,
     EPA Contract 68-02-2199, Task 600, October 1978.

Versar, Inc., Effect of Physical  Coal Cleaning on Sulfur  Variability, Draft
     Report, EPA Contract 68-02-2199, Task 600, January 1979.

Thomas, R. E., Interpreting Statistical Variability, Proceedings of EPA
     Symposium on Coal  Cleaning to Achieve Energy and  Environmental  Goals,
     Hollywood, Florida, September 1978 (to be published).

-------
THE INTERAGENCY FLUE GAS DESULFURIZATION EVALUATION STUDY
                    James C. Dickerman
                    Radian Corporation
                  Durham, North Carolina
                           and
                     Richard D. Stern
           U.S. Environmental Protection Agency
       Industrial Environmental Research Laboratory
          Research Triangle Park, North Carolina
              For Presentation at the Fifth
            Flue Gas Desulfurization Symposium
                    Las Vegas, Nevada
                     March 5-8, 1979
                                258

-------
         THE INTERAGENCY FLUE GAS DESULFURIZATION EVALUATION STUDY
     President Carter's National Energy Plan of April 1977 has placed
increasing emphasis on United States industries and utilities to convert
from oil and natural gas to coal-based energy systems.  As coal utilization
increases, the potential impact of SOa emissions from coal-fired units on
air quality will become more significant.   Unless adequate steps are taken
to control these SC>2 emissions, significant degradation in air quality could
result.

     This paper presents the results of a  study that was conducted pursuant
to the President's April 1977 National Energy Plan.  The important concerns
which led to the study upon which this paper is based can be summarized as
follows.  The National Energy Policy mandates acceleration of coal usage in
the United States.  Maintenance of air quality within the context of this
increase in coal utilization will require  SOz emission controls.  FGD appears
to be the most viable method of SC>2 emission control in the near term (1985) .
President Carter, in his National Energy Plan, stipulated that the Government
would undertake a 6 month study to determine if additional Federal funding
would accelerate the commercialization and acceptance of FGD technology.
Therefore, this study was conducted to determine if additional Research,
Development and Demonstration (RD&D) efforts would accelerate the acceptance
of FGD technology.  Specific RD&D options  and funding levels were also to be
identified as an output of this study.

     This paper summarizes the procedures  used to evaluate the need for
increased Federal expenditures to accelerate the use of FGD technology and
highlights the findings of that study.  First, the project bases and organ-
ization are discussed.  Next, the approach and methodology used to achieve
the study objectives are presented.  Finally, the study conclusions,
recommendations, and current status of the evaluation are presented.
                                   259

-------
         THE INTERAGENCY FLUE GAS DESULFURIZATION EVALUATION STUDY
INTRODUCTION

     This paper is based on a draft report that was completed November 1977
pursuant to the President's National Energy Plan which mandated conversion
from an oil- and natural gas-based energy system to one based on coal.  The
President requested this study to identify if  and what government support is
necessary to accelerate the development and commercialization of flue gas
desulfurization technology in order to permit  expanded use of coal without
environmental degradation.  Although the draft report was completed over a
year ago, programs, schedules, and funding levels may be out of date;
however, its conclusions concerning flue gas desulfurization (FGD)  research
opportunities are still valid.  In fact, several research opportunities
identified in the study are currently being implemented.

     As coal usage increases across the United States, the potential impact
of S02 emissions from coal-fired units on air  quality will become more
significant.  Unless adequate steps are taken  to control these SOa  emissions,
significant degradation in air quality could result.   There are many alter-
natives for reducing S02 emissions from coal utilization.   These include
coal cleaning, coal liquefaction, coal gasification,  fluidized bed combus-
tion, and FGD.  Of these alternatives, only coal cleaning and FGD are
currently being commercially applied.  It now  appears that competing SOa
emission control technologies will not see widespread applicability before
1985.  Since coal cleaning is not attractive for all  coals, FGD must be
considered to be the most promising near-term  alternative for controlling
SOa emissions across the United States.

     This study to determine if additional Federal funding would accelerate
the commercialization and acceptance of FGD technology, utilized the follow-
ing approach.
                                     260

-------
     1)   A review of existing and proposed FGD technologies to identify
         the need for additional research, development,  and demonstration
         (RD&D)  activities.

     2)   An identification of specific RD&D options.

     3)   An estimate of the level of funding necessary to complete the
         recommended RD&D objectives.
In particular,  advanced FGD concepts were reviewed to determine if  they

offered the potential for enhancing or improving FGD technologies that  are

commercial or are nearing commercialization,  or if they offered the poten-

tial for new and improved approaches to FGD.
GROUND RULES AND ASSUMPTIONS
     The following ground rules and assumptions were used for this study:
     1)  The study would be completed in 6 months.

     2)  There would be extensive coordination and  review within the
         Federal government.

     3)  Both processes and subsystems would be evaluated.  Processes  are
         defined as complete FGD systems;  subsystems are those  parts of
         FGD processes that could be used  virtually interchangeably  in
         several FGD processes.   For example, within this definition,
         lime/limestone scrubbing is considered to  be a process.   The
         Allied Chemical S02  reduction system used  with the Wellman  Lord
         installation at NIPSCO  is a subsystem because it could be used
         with any regenerable FGD process  that produces a concentrated
         SO2 stream.
     4)  Processes and subsystems that have demonstrated and/or  have
         capability for nitrogen oxides (NO )
         would be given special consideration
capability for nitrogen oxides (NO )  or fine particle removal
                                  X
     5)   Processes and subsystems recommended for additional funding
         would have to show either economic,  environmental,  or  tech-
         nological advantages over existing or developing FGD systems.

     6)   Processes and subsystems evaluated would have to be capable  of
         being commercialized in a time frame competitive with  alternative
         technologies (by 1985).
                                    261

-------
     The 6 month time frame for the study was necessary to provide timely
inputs to EPA's FY78 and FY79 program plans.  Extensive coordination and
review within the Federal Government was desired in order to ensure maximum
objectivity and provide a broad perspective for the study.

     The decision to evaluate both processes and subsystems was made to
provide maximum flexibility.  By evaluating subsystems, it was possible to
identify potential improvements to existing and developing processes as well
as entirely new process approaches.  Processes/subsystems that had the
potential for NO  or fine particle removal were favored because of the obvious
advantage they would offer in efforts to expand coal use without environ-
mental degradation.

     The requirements that processes/subsystems must be capable of being
commercialized in a time frame competitive with alternative technologies
and must show either economic, environmental, or technological advan-
tages over existing or developing FGD systems were consistent with
accelerating the rate of application of the technology.  Any process/sub-
system development that improves the reliability, lowers the cost, or
reduces the secondary environmental impact of FGD technology could
accelerate the rate of application of the technology.  Likewise, since
FGD is the most promising means of S02 emission control until alter-
natives (such as FBC) are developed, process improvements in FGD tech-
nology are particularly important in the period between the present and
the time that alternative technologies are commercialized.  The elimina-
tion of certain processes/subsystems from consideration in this study,
however, does not imply that such processes/subsystems will jiot be
worthy of further evaluation or development at a later time.
ORGANIZATION

     This study was directed by Richard D. Stern, EPA, Industrial Environ-
mental Research Laboratory - Research Triangle Park (IERL-RTP).  To ensure
                                     262

-------
maximum objectivity and incorporate a broad perspective, the study utilized
three Interagency groups:  a Technical Working Group, an Interagency
Steering Panel, and a Liaison Group.

Technical Working Group

     This group was chaired by Stephen J. Gage, then Acting Assistant
Administrator for Research and Development, EPA.  Other members of the
group were:  Lewis Faucett, Tennessee Valley Authority (TVA); Stuart
Dalton, Electric Power Research Institute (EPRI); Myron Gottlieb, Depart-
ment of Energy (DOE); Lawrence H. Weiss, Chem Systems; and A. V. Slack
and Milton R. Beychok, independent consultants.  The functions of the
Technical Working Group were:

     1)  To assist in the development of screening criteria to be applied
         to a comprehensive list of FGD processes/subsystems.
     2)  To review the initial screening results.
     3)  To assist in the development of evaluation criteria to be applied
         to the processes/subsystems which were selected for detailed
         evaluation.
     4)  To conduct the detailed evaluation of candidate processes/
         subsystems.
     5)  To assess the schedule and costs of the process/subsystem RD&D
         options.
     6)  To make recommendations regarding RD&D opportunities.
     7)  To review the draft final report.

Interagency Steering Panel

     The Interagency Steering Panel provided overall guidance in approach
and methodology and reviewed study products with emphasis on the final
report.  Members of the Interagency Steering Panel were:  Samuel Biondo,
DOE: Bernard Chew, DOE; Howard Feibus, DOE; Stephen J. Gage, EPA; S. William
Gouse, DOE; G. R. Hall, DOE; Gerald Hollinden, TVA; David Israel, DOE;

                                     263

-------
K. H. Jones, Council of Environmental Quality (CEQ); Judy Kammins, Office
of Management and Budget (OMB); Rafael Kasper, Office of Science and
Technology (OST); Susan Hickey, Federal Energy Administration (FEA);
Richard Hertzberg, OMB; C. Morgan Kinghorn, OMB; William E. Mott, DOE;
Frank T. Princiotta, EPA; Jack Silvey, DOE; and Kenneth Woodcock, DOE.

Liaison Group

     The Liaison Group provided guidance and comment from other organizations
which were felt to have interest in the study.  Liaison with these organiza-
tions was utilized to obtain the broadest possible perspective for the study.
Included in the Group were:  R. W. Crozier, National Academy of Engineering
(NAE); Lloyd Taylor, Science Advisory Board (SAB); H. W. Elder, TVA; and
Kurt Yeager, EPRI.  Interface with the Liaison Group was handled by Richard
D. Stern of EPA's IERL-RTP-
APPROACH AND METHODOLOGY

     The approach developed for completing the program objectives is shown
schematically in Figure 1.  Shortly after initiation of the study, EPA con-
tractors, with the aid of consultants, developed an initial list of processes/
subsystems and an initial set of evaluation criteria.  A public meeting was
held in Washington to acquaint the public with the scope and objectives of
the study and to solicit public comment.  Results of this meeting added two
processes to the comprehensive list (see Table 1) for evaluation.

     A series of meetings was then held with the Technical Working Group,
contractors, Steering Panel, and the Liaison Group to review and comment
on the initial process/subsystem list shown in Table 1 and initial screening
criteria shown in Table 2.  Once these were approved, technical and economic
data were gathered on each process/subsystem and a series of Information
Survey Papers was prepared.
                                     264

-------
                                Methodology
                                Development
                                 (C)  (S) (L)
Develop Comprehensive List
 of Processes/Subsystems
 for  Initial Screening
      (C) (W) (L)
Comprehensive List
of FGD Processes/
Subsystems
i kScreenlng
   (C)  (W)
                                                  Initial
                 Selected Candidates
                 for Detailed
                 Evaluation
Ln
              Review Draft
              Report for
                                                                                Perform ' '
                                                                                                             Develop Detailed
                                                                                                             Evaluation Criteria
                                                                                                               (C) (W) (L) (S)
                                                                                Detailed
                                                                                Evaluation
                                                                                    (W)
                               Process/Subsystem
                               Summary Papers
                          Gather Technical
                         and Economical Data
                        	(C)	
                                                                                                       KEY

                                                                                                       (W)
                                                                                                       (C)
                                                                                                       (S)
                                                                                                       (L)
                                          - Working Group
                                          - Contractors
                                          - Steering Panel
                                          - Liaison Group
                               Figure  1.   Interagency  FGD Technology  Evaluation-Study Approach  and Methodology.

-------
                           TABLE  1.   COMPREHENSIVE  LIST OF POTENTIAL  CANDIDATES FOR DETAILED EVALUATION
   Throwaway Systems
 Wet Regenerable Systems
  Dry Regenerable Systems
                                                                     Subsystems
                                                                                                                                        Others
Agglomerating Cone
Alkaline Ash Scrubbing
Ammonia Scrubbing/H2SO<,
  Acidulation (Comlnco, Ammsox)
Ammonia Solution Double Alkali
Ammonia Vapor Double Alkali
  (Ugine Kuhlman, Nippon
  Kokan)
Asahi Chemical
C5,02
Calcium Chloride (Bolter, Kobe)
Calsox
Carbon Adsorption (Hitachi,
  Lurgi-Sulfacid)
Chisso Engineering
Chiyoda Thoroughbred 101
Chiyoda Thoroughbred 102
Chiyoda Thoroughbred 121
Double Alkali with Limestone
  (Showa-Denko, Kureha-
  Kawasaki)
Dowa Aluminum Sulfate
Dry Alkalis plus Fabric Filter
Ebara-Jaeri
Ishikawajiraa-Harima HI
IPRAN
Kawasaki MgO/Lime
Koyo
Krebbs-Neville
Kurabo
Kureha (S02/NOX)
Kureha Sodium Acetate
Lead-Zinc Ore
Lewis Process
Micropul
Mitsubishi Heavy Industries
Moretana Calcium
Mo retana Sodium
Red Mud
RIIC
SCRA
Sea Water
Simon Carves
Sodium Carbonate Scrubbing
Sodium Hypochlorite
Sulfurtain
Ammonia Bisulfate (ABS)
Ammonia Evaporative
  Crystallization
Ammonia/IFP (Catalytic,
  Research Cottrell)
Aquaclaus
ASARCO Dimethylanlline
Consol
Grillo
ICI Steam Stripping
Johnstone Zinc Oxide
Lurie Sodium Aluminate
Maget
McKee
Melamine
Mitsui Eng. & Shipbuilding
Molten Carbonate
Molten Potassium Carbonate/
  Thiocynate
NIIOGAZ Ammonia/Steam
  Stripping
NIIOGAZ Magnesium Oxide
NOSOX
Peabody
Potassium Bisulfite
Potassium Formate
Potassium Sulfite
Ralph Parsons
Spring-Nobel Hoechst
Stone & Webster/Ionics
Sulf-X
TSK Sulfix
UOP Sulfoxel
Alkalized Alumina
Bergbau-Forschung
Bureau of Mines Mn02
Carbon Adsorption/Inert  Gas
  Stripping (Reinluft/Chemiebau
  Sumitomo)
Copper Oxide Adsorption  (Shell,
  Esso B&W, Bureau of  Mines)
Gallery Chemical
Horrael
Houdry
In-Sltu CO Reduction (Chevron)
In-Situ H2S Reduction (Peter
  Spence, Princeton Chemical,
  Ontario Research Foundation)
Integrated Cat-Ox
Kiyoura-TIT
Mitsubishi Manganese Oxide
Purasiv S
Rohm & Haas Resin Adsorption
Shuffman
Sumitomo Heavy Industries
TOPSOE
Tyco Chamber
Unitika
Uranium Oxide
Westvaco
Allied Reduction Process
ASARCO Sulfur Plant
Calcium Sulfate Regeneration
Citrate Double Loop
,  Regeneration
Claus Process
Cocurrent Scrubbing
Dilute System Sulfate Removal
Direct Production of Sulfur
  from MgSOa
Forced Oxidation of CaSOs
IFF  Subsystems
Ionics Electrolytic
  Regeneration
Mass Transfer Additives with
  Lime/Limestone Scrubbing
Production of Reducing Gas
RESOX
Sludge Stabilization
Tampella Recovery Process
Allied Chemical Electro-
  dialysis
Allied Chemical Membrane
  Process
Barium Carbonate
Battelle Fused Salt
Boliden
Catalox
Cooper
Condensation and Reaction
  with Fly Ash
Dry Removal with Ground Lime
Dry Sorbents
Electrochemical Concen-
  tration
Esso (V203)
Goodrich
ITT
McGauley
MgO Based Double Alkali
Reduction with "Blue Gas"
S iemens
Soil Process
Sulfuric and Nitric Acid
  Recovery Method

-------
            TABLE 2.   INITIAL SCREENING  EVALUATION CRITERIA FOR
                      CANDIDATE  PROCESSES/SUBSYSTEMS
I.     Minimum Requirements

      Processes/subsystems which fail  to meet  any  of  these
      requirements will not be considered  further

      A.   A process/subsystem must be  able to  achieve compliance
          with the minimum assumed future  NSPS for SO2

      B.   A process/subsystem must be  capable  of being demon-
          strated by  1985

      C.   A process/subsystem must be  applicable to treating flue
          gases from  combustion of coal

      D.   Adequate information must be available to enable
          process/subsystem evaluation
   Go/No  Go
II.   Environmental and Energy Considerations

      A.   Compliance with S02  regulations
          (Assumes 0.2 kg S02/109  Joules;
                   0.4 Ib S02/106  Btu)
          6 - potential to comply  with  future  regulations
              with high sulfur coals (96%)a
          4 - potential to comply  with  future  regulations
              with medium sulfur coals  (94%)
          2 - potential to comply  with  future  regulations
              with low sulfur  coals  (80%)

      B.   Potential for multipollutant  removal
          3 - more than one
          1 - one
          0 - none

      C.   Performance growth potential

      D.   Secondary pollutant  problems
          3 - none
          2 - minor
          0 - major

      E.   Relative Energy Requirements
          Since material and energy  balances will not be avail-
          able for many of these processes, the following
          qualitative rating factors will  be used:
Q
 SO2 Removal
 30 points


( 6 points)
( 3 points)




( 3 points)

( 3 points)
                                    267

-------
                            TABLE 2.   (Continued)
          1)  Reheat                                              ( 3 points)
             3 - none
             2 - possibly
             1 - from saturation (52°C-79°C;  125°F - 175°F)

          2)  Reducing Gas  '                                      ( 3 points)
             3 - none
             1 - CO and H2
             0 - H2 only
          3)  L/G Ratio, Pressure Drop                            ( 3 points)
             3 - low
             1 - medium
             0 - high
          4)  Other Energy                                        ( 3 points)
             3 - low
             1 - medium
             0 - high

      F.   Raw Material Requirements                              ( 3 points)

          3 - uses less than 2 that are readily available
          2 - uses less than 2 that are not readily available
          0 - uses more than 2 that may not be readily available
III.  Dtvelopment Status                                          25 points

      A.   i tate. of Development                                   (10 points)

         10 - demonstration (>30 MW)  coal-fired
          8 - demonstration (>30 MW)  oil-fired
          7 - prototype (10-30 MW)  coal-fired
          5 - prototype (10-30 MW)  oil-fired
          4 - pilot (1-10 MW)  coal-fired
          3 - pilot (1-10 MW)  oil-fired
          2 - bench scale
          1 - conceptual

      B.   Degree of Integration                                  ( 5 points)

          5 - totally integrated
          3 - sorption and regeneration integrated
          0 - nothing integrated

      C.   Use of Currently Commercialized Technology             ( 5 points)

          5 - in all process sections
          3 - in at least two process sections
          2 - in one process section
          0 - not at all
                                    268

-------
                            TABLE 2.   (Continued)
      D.   Scale-up Problems

          5 - no apparent problems
          2 - potential problems with complexity or scale-up
              of a new or unique processing step
          0 - potential scale-up problems in several areas
( 5 points)
IV.    Economic and Technological Considerations
      (applied to major sections and subsystems)
      Capital requirements may be estimated by considering;

      A.  Relative Chemical Complexity

         15 - simple, workable
         10 - complex, workable
          5 - simple, questionable
          0 - complex, questionable

      B.  Relative Mechanical Complexity

         10 - simple
          5 - moderately complex
          0 - very complex
 25 points
(15 points)
(10 points)
V.    Applicability                                               20 points

      A.  Suitability for Utility Applications                   (10 points)

          1) Separability of Process Steps                       ( 3 points)
             3 - can be easily decoupled
             1 - difficult decoupling
             0 - decoupling not practical

          2) Load Following Capability                           ( 3 points)
             3 - follows load changes rapidly
             1 - follows load changes slowly
             0 - process sections do not readily follow load
                 changes
          3) Retrofitability                                     ( 2 points)
             2 - easy to retrofit
             1 - will require minor modifications
             0 - significant problems
          4) By-product Utilization/Marketability                ( 2 points)
             2 - readily marketable
             1 - marginally marketable
             0 - unmarketable
                                    269

-------
                     TABLE 2.  (Continued)
B.  Suitability for Industrial Application                  10 points

    1) By-product Utilization/Marketability                ( 3 points)
       3 - readily marketable
       1 - marginally marketable
       0 - unmarketable
    2) Ability to Modularize the Process                   ( 3 points)
       3 - suitable for small (>10 MW) modules)
       1 - suitable for medium (10-40 MW) modules
       0 - suitable for large (>40 MW) modules

    3) Separability of Process Steps                       ( 2 points)
       2 - can be easily decoupled
       1 - difficult decoupling
       0 - decoupling not practical
    4) Retrofitability                                     ( 2 points)
       2 - easy to retrofit
       1 - will require minor modifications
       0 - significant problems
                              270

-------
     Next, the initial screening criteria were applied to the comprehensive
process/subsystem list to select those processes that offered the most
potential for accelerating the commercialization and application of FGD
technology.  These selected, or candidate, process were then evaluated in
more detail in the next phase of the project to identify specific RD&D
opportunities.  This screening was necessary to reduce the task of detailed
evaluation to a reasonable size by eliminating those processes that appeared
to have the least potential for near term improvement of FGD technology.
The elimination of some processes from consideration in this study does not
imply that such processes will not be worthy of development or further
evaluation at a later date.  The initial screening focused on the selection
of candidate processes in two main areas:  1) those that enhance or Improve
technologies currently used in commercial and developing processes, and
2) those that, if developed, would offer new and improved approaches to FGD.
Commercial processes, as well as processes being developed under Federal
funding, formed a basis for comparison with the processes/subsystems being
screened.  For purposes of this evaluation, the following were considered to
be currently commercial processes or processes being developed under Federal
fund ing:

         Lime/Limestone Wet Scrubbing
         Magnesia Slurry Scrubbing
         Wellman-Lord
         Sodium/Lime Dual Alkali
         Citrate Buffered Absorption
         Rockwell International's Aqueous Carbonate Process

     A total of 138 processes were screened to examine their potential for
offering technological and/or economic advantages over commercialized and
developing FGD processes.  Results of the screening shown in Table 3 indi-
cated that 13 processes/subsystems should be evaluated in detail, along with
the 6 commercial and developing processes listed above, for the purpose of
identifying RD&D opportunities that would accelerate the commercialization
of. FGD.
                                     271

-------
                                  TABLE 3.  PROCESSES SELECTED FOR FURTHER EVALUATION
         Throwaway
Wet Regenerable
Dry Regenerable
Subsystems
         Calcium Chloride
         Chiyoda Thoroughbred
           121
         Dowa Aluminum Sulfate
         Kurabo
         Lime/Limestone
         Sodium/Lime Double
           Alkali
         Sodium Throwaway
           Systems
Ammonia/IFF
Atomics International
  Aqueous Carbonate
  Process (ACP)
Citrate
Magnesia Slurry
Melamine
Sorption/Steam
  Stripping
Wellman-Lord
Bergbau-Forschung
Copper Oxide
  (Shell-UOP)
Integrated Cat-Ox
Sorption/Steam Strip-
  ping (Rohm & Haas
  Resin Adsorption
  Process)
RESOX
N5
          Commercial or current federally funded processes

-------
     To assist the Technical Working Group in evaluating and comparing the
processes in a consistent impartial manner, a set of evaluation criteria
was developed for application to each of the 19 processes selected as poten-
tial candidates for RD&D funding.  In addition to providing a means of ranking
the processes, a consistent set of evaluation criteria was judged to be
necessary to evaluate RD&D options for the following reasons:
         Evaluation criteria formalize the evaluation procedure and
         allow the evaluators to compare their process evaluations on a
         consistent basis.  Any discrepancies in the evaluation can then
         be isolated and discussed.
         Specific evaluation criteria force analysis of all process
         advantages and disadvantages.
         Key process deficiencies can be identified through the use of
         evaluation criteria.
The criteria developed for the detailed process evaluations are presented
in Table 4.

     The objectives of the detailed process evaluations were to identify
specific RD&D programs and to identify specific benefits to be derived from
the successful implementation of these programs that would achieve the
overall goal of accelerating the commercialization of FGD technology.  A
list of RD&D recommendations was developed based on the detailed evaluations,
and a draft report was prepared and submitted for review.
CONCLUSIONS AND RECOMMENDATIONS

     To enhance the utilization of coal and to minimize its environmental
impact, the Technical Working Group developed the following conclusions:
                                     273

-------
      TABLE 4.  DETAILED EVALUATION FOR PROCESS/SUBSYSTEMS SELECTION


I.    Environmental and Energy Considerations                     25 points

      A.  Compliance with S02 Regulations                        ( 5 points)
          5 - ability to comply with future regulations with
              high sulfur coals (96%)
          3 - ability to comply with future regulations with
              medium sulfur coals (94%)
          1 - ability to comply with future regulations with
              low sulfur coals (80%)

      B.  Potential for Multipollutant Control                   ( 3 points)

          3 - significant removal of 1 or more
          2 - some removal of 1 or more
          1 - no removal of other pollutants

      C.  Secondary Pollution Problems - Air                     ( 2 points)

          2 - no air emission problems
          1 - potential air emission problems
          0 - major air emission problems

      D.  Secondary Pollution Problems - Liquid                  ( 2 points)
          3 - no liquid wastes
          2 - liquid wastes that can be easily treated
          1 - liquid wastes that require unusual  treatment

      E.  Secondary Pollution Problems - Solid                   ( 3 points)
          3 - no solid wastes
          2 - solid wastes that can be easily disposed of
          1 - solid wastes that require unusual treatment

      F.  Energy Intensiveness                                   (10 points)

          • Chemical and electrical energy requirements compared
            with technology currently commercialized or under
            development
            6-10 - lower energy requirements
            5    - average energy requirements
            1-4  - higher energy requirements
II.   Development Status                                          15 points

      A.  Overall Process/Subsystem Development Status           ( 5 points)

          • How long has developer studied and worked his
            process/subsystem?

          • How much of the process/subsystem is technically
            well known and well founded?

                                    274

-------
                          TABLE 4.   (Continued)
          •  At what level has  it been operated  and  for  how  long?
              Bench Scale,
              Laboratory,
              Pilot Plant,
              Small Prototype,
              Paper Study.

          •  Have all sections  been operated  as  a  chemically or
            mechanically integral system?

      B.   Process/Subsystem Operations and Use  of                (5 points)
          Commercialized Technology

          5  - operations in common practice
          3  - industrially demonstrated
          2  - some uncommon methods involved
          0  - several unproven processing  steps

      C.   Process Controllability                                ( 5 points)

          5  - simple process without many  sensitive control
              requirements
          3  - simple process with sensitive  control requirements
          2  - complex process  without many sensitive control
              requirements
          0  - complex process  with sensitive control requirements


III.   Economic and Technological Considerations                  35 points

      A.   Capital Investment Costs                               (15 points)

          15 - 40% less than median
          14 - 30% less than median
          12 - 20% less than median
          10 - 10% less than median
          8 - median
          6 - 10% more than median
          4 - 20% more than median
          2-30% more than median
          0-40% more than median

      B.   Annualized Costs                                       (15 points)

          15 - 40% less than median
          14 - 30% less than median
          12 - 20% less than median
          10 - 10% less than median
           8 - median
           6 - 10% more than median
           4 - 20% more than median
           2 - 30% more than median
           0 - 40% more than median

                                     275

-------
                           TABLE 4.  (Continued)
      C.  Economic Sensitivity to Technological Factors
           5 - little uncertainty in technological factors
               affecting cost
           3 - some uncertainty in technological factors
               affecting cost
           0 - considerable uncertainty in technological
               factors affecting cost
( 5 points)
IV.   Applicability

      A.  Suitability for Utility Application
 25 points

(10 points)
          • Retrofitability
          • By-product Utilization/Marketability
          • Separability and Process Steps
          • Load Following Capability
          • Acceptability of Processing Techniques to  the Industry

      B.  Suitability for Industrial Application                 (10 points)

          • Retrofitability
          • By-product Utilization/Marketability
          • Separation of Process Steps
          • Ability for Process Modularization
          • Acceptability of Processing Techniques to  Small
              Boiler Applications

      C.  Flexibility to Provide Alternate Products           *   ( 3 points)

          3 - can product sulfur or acid
          2 - can produce both sulfur and acid with
              significant modifications
          1 - produces either sulfur or acid
          0 - produces neither sulfur or acid

      D.  Regional/Site Considerations                           ( 2 points)

          2 - process/subsystem suitable over broad range
              of applications
          1 - process/subsystem suitable only for narrow
              range of applications
                                     276

-------
        There  are  significant  potential  benefits  to  be derived  from
        increasing the  level of  Federal  funds  for FGD  research,  develop-
        ment,  and  demonstration  (RD&D) efforts.
        The  increased Federal  funding  should be utilized  for  research
        and  development projects to  improve systems, both throwaway
        (Lime/Limestone)  and regenerable,  that have  been  or are  being
        demonstrated on large-scale  equipment.
        Resources  should be available  for  process/subsystem evaluations
        to assess  potential solutions  to common FGD  problem areas  and
        to continually  evaluate  new  or enhanced developments  in  FGD
        technology.
     It was  the consensus  of  the Technical Working  Group  that  the  highest
priority  research opportunities were in the  area of further development
to improve the economics and  applicability of systems that have been or
are being demonstrated on large-scale equipment.  Because of the emphasis
upon finding near-term solutions to existing problems,  many of the RD&D
recommendations were to further investigate  process subsystems or  process
alternatives.  These recommendations, in general, concerned RD&D cate-
gories that  have potential widespread application,  especially in wet and
throwaway FGD processes.  A secondary priority was  assigned to processes
and process  developments that provide longer-term benefits.  A discus-
sion of each recommended RD&D opportunity and its expected benefit is
presented below.

     1)   Lime/Limestone Improvements

         A)    Mass Transfer Additives - An extension of current programs
              that have focused upon evaluating magnesium,as a mass
              transfer additive for lime/limestone  processes was recom-
              mended.   Specific additives to be examined  included  both
              organic  and  inorganic mass transfer additives such as adipic
              acid,  calcium chloride, and sodium carbonate.  The use of a
              mass transfer additive has the potential  for Improving the
              S02 removal  ability of lime/limestone systems, increasing
              sorbent  utilization,  and decreasing process costs.  Thus the
                                    277

-------
     primary benefits  of  this  research could be an increased
     level of application of lime/limestone FGD technology and a
     lower process cost.   Initial efforts  should involve pilot
     scale testing followed by demonstration in a prototype unit.

B)   Forced Oxidation  - EPA has an  existing program to evaluate
     forced oxidation  at TVA's Shawnee Steam Plant.  Pilot test-
     ing of two process variations  is currently being conducted.
     A full-scale demonstration program was recommended for
     1979-1980.  A full-scale  demonstration of this concept could
     benefit lime/limestone technology in  reducing sludge disposal
     problems, thus removing a major environmental impediment to
     lime/limestone scrubbing, and  increasing its applicability.
     In addition, a potentially marketable gypsum product may be
     produced.  Since  full-scale demonstration could occur by
     modifying an existing system,  this appears to be the next
     step in commercializing this  technology.

C)   Contactors - Extension of the  current work to evaluate co-
     current flow designs was  recommended.  The use of a cocurrent
     scrubber could significantly reduce the cost and complexity
     of lime/limestone scrubbing.   Extension and expansion of
     current programs  appear necessary to  conclusively answer
     operability questions. In addition,  testing and demonstra-
     tion of the Chiyoda Thoroughbred 121  (CT-121) system were
     recommended, as this system has claimed significant cost
     reduction over conventional limestone scrubbing due to its
     simple design.

D)   Sludge Disposal - Extension of current sludge disposal activi-
     ties to evaluate large-scale disposal options (fixation, lin-
     ing, and ponding) was recommended.  Sludge disposal is a major
     environmental problem in  lime/limestone scrubbing.  Resolution
     of this problem on a large scale could enhance the applica-
     bility of lime/limestone and dual alkali systems.
                           278

-------
   E)   Hardware Improvements  -  Problems with mist  eliminators,
        reheaters,  and  stack liners  have been the cause of much of
        the  current FGD system failures.  Additional development work
        to improve  these components  was recommended as a means of
        increasing  the  reliability and operability  of wet FGD systems.
        Demonstration of improved designs for these components on
        full-scale  systems  was recommended  to gain  acceptability
        by the industry.

2)   Sulfur Production with Carbon

    A)   RESOX - EPRI is funding  a RESOX demonstration program in
        Germany to  produce  sulfur from the  Bergbau-Forschung FGD
        unit using  anthracite  coal as the reductant.  Evaluation
        of the RESOX process in  a demonstration-sized facility with
        feed from front-end systems  other than  Bergbau and evaluation
        of coals other  than anthracite for  use  as a reductant was
        recommended. The overall applicability of  the RESOX process
         suffers because it  has not been evaluated on other systems.
        Extension of the EPRI  program to test the RESOX process under
         different conditions could result in the solution of a major
        problem in  FGD:  sulfur  production  without  the use of a reduc-
         ing  gas. This  could significantly  enhance  applicability of
         FGD  due to  its  potential for reducing cost, complexity, and
         secondary pollution.

    B)    Rockwell International Regeneration - Extension of the EPRI
         program to  demonstrate and operate  the  Rockwell International
         (RI) Regeneration System in  an integrated mode to gather data
         for  full-scale  design  and construction  was  recommended..  RD&D
         efforts on  this system would demonstrate direct conversion
         of SOa to sulfur without a reducing gas plus provide addi-
         tional design data  for the Aqueous  Carbonate Process  (ACP)
         to be constructed at Niagara Mohawk under EPA funding.
                               279

-------
3)  Continuing Evaluation Effort - It was recommended that funds
    be set aside for use in technical and economic evaluation of
    specific problems or of processes that in the future may
    appear to offer advantages over current FGD technology but
    that were not considered in this study due to their lack of
    development.  This continuing evaluation effort could ensure
    that future improvements in FGD technology are given a com-
    plete and timely evaluation.  In addition, it would provide a
    source of funding for important studies on critical issues in
    FGD technology that could be of significant benefit in directing
    future RD&D activities.

4)  Magnesia Slurry Scrubbing - Demonstration of sustained sulfuric
    acid production from a full-scale system would be beneficial in
    advancing a technology that appears to have potential for very
    high S02 removal.  In addition, demonstrating the ability to
    directly produce sulfur in the calciner will significantly enhance
    process applicability.  Both concepts should be demonstrated and
    evaluated on a full-scale system.

5)  Limestone Use in Sodium Based Dual Alkali Systems - Demonstration
    of the use of limestone as the regenerant in ongoing pilot work
    and extension and modification of the EPA sponsored demonstration
    program at Louisville Gas and Electric (LG&E) to test limestone
    was recommended.  The use of limestone in dual alkali systems has
    potential for significantly reducing system operating costs
    because limestone is a less expensive raw material than lime.

6)  Sodium System Waste Handling

    A)   Disposal - Test of concepts for fixation, plus other methods
         of sludge handling and disposal from once-through sodium
         systems, were recommended.  The disposal of sodium sludge is
         a major problem in sodium throwaway processes.  A solution
                                280

-------
         to this problem would increase the applicability of  this
         process with potential reductions in both cost and environ-
         mental impacts.

    B)    Regeneration - Evaluation of methods for the regeneration  of
         sodium system by-products and eventual testing was recommended.
         This program has a similar benefit to,the option listed above.
         It could eliminate disposal problems faced by sodium systems
         and reduce raw material costs.

7)  Application of Low-Btu Gas - Evaluation of the applicability of
    coal-derived low-Btu gas to regenerate FGD systems was recommended.
    This would include feasibility studies to evaluate system require-
    ments, followed by demonstration of the applicability of  specific
    gasification systems.  The use of coal gasification could eliminate
    the dependence of some regenerable FGD processes on increasingly
    costly and scarce supplies of natural gas.

8)  Sorption/Steam Stripping - Extension of the EPRI work in  the areas
    of  laboratory studies and pilot work, to be followed by demonstra-
    tion at the 60-100 MW level, was recommended.   It was felt that
    design of a general test facility for the testing of the  large
    number of concepts currently under consideration would be beneficial.
    The Sorption/Steam Stripping concept appears to have potential  for
    more reliable, lower cost operation than some other regenerable
    systems and should be assessed.

9)  Dowa Test Facility - A pilot plant program to evaluate the feasi-
    bility of the Dowa process in coal-fired applications was recom-
    mended.  This process rated high in the evaluations and appears to
    have significant advantages over current dual alkali FGD  technology ,
    in  that it uses limestone as a regenerate and produces a  marketable
    quality gypsum by-product.  A program of this nature is currently
    being considered by EPRI and TVA.
                               281

-------
     10)  Integrated Cat-Ox - Additional support for the Cat-Ox process to
          demonstrate the Integrated Cat-Ox system at the 100 MW level was
          recommended.  The Integrated Cat-Ox process appears to have sig-
          nificant cost advantages over current FGD technology.  Successful
          demonstration of this process could have significant benefits in
          the areas of cost and process complexity.

     11)  SULF-X - An evaluation of the SULF-X process to independently
          analyze vendor claims was recommended.  The process has potential
          as a SO /NO  flue gas treatment process; however,  the claims have
                 X   X
          not been independently evaluated.  This evaluation is currently
          planned by DOE.

     The Working Group also recommended that a total level of funding of over
$100 million dollars would be necessary to complete the identified RD&D oppor-
tunities.  It was anticipated that funding of many of the recommended projects
would be on a cost-shared basis with the process vendor,  host utility or indus-
trial site, EPRI, or another government agency.   The estimated total  funding
level did not attempt to define percentage funding by participant. Figure 2
presents a prioritized listing of the recommended RD&D opportunities  along
with the then-estimated funding levels and schedules.
SUMMARY

     The approach and methodology used in this evaluation have provided an
effective means of identifying research opportunities for enhancing the near-
term commercialization of FGD technology.  Although the recommended program
has not yet been formally approved,  its impact has been felt in that several
of the recommended RD&D opportunities have already been implemented.  For
example, in the lime/limestone process area,  EPA has initiated work to eval-
uate the effects of adipic acid addition on SOz removal, and plans are being
made to demonstrate forced oxidation at TVA's Widow's Creek Plant.  EPRI has
development programs under way to evaluate both cocurrent and sparged
                                     282

-------
   RESEARCH OPPORTUNITIES
                     YEAR

  77   78   79   80   81   32    83    84   85
I     I     I     I     II     I     I     I     I
 1 LIME/LIMESTONE


        A.  ADDITIVES


        B.  FORCED OXIDATION


        C.  CONTACTORS-COCURRENT

                 CHIYODA 121

        D.  SLUDGE

        E.  HARDWARE IMPROVEMENTS

 2. SULFUR PRODUCTION  WITH CARBON

        A.  RESOX


        B.  A i


 3. CONTINUING EVALUATION EFFORT

 4. MAGNESIA SLURRY DEMONSTRATION

 5. LIMESTONE IN DOUBLE ALKAU  SYSTEM


 6. SODIUM THROWAWAY  SYSTEMS

        A.  DISPOSAL


        B.  REGENERATION

 7 GASIFICATION  APPLICATION/DEMONSTRATION


 8. SORPTION/STEAM STRIPPING
         A.   LABORATORY EVALUATION

         B.  60-100 MW DEMONSTRATION


 9.  DOWA TEST FACILITY


10.  INTEGRATED CAT-OX DEMONSTRATION


11.  SULF-X EVALUATION
                             -NUMBERS SHOWN ARE
                              ESTIMATED FUNDING
                              REQUIREMENTS FOR
                              EACH PROGRAM (t 10*)
             111111
                    25
                    25
          iiiiiiiiimiiimiiiimiiiiiiiiiiiiiiiiiiimii
                    PLANNED EPA PROGRAMS

                    PLANNED EPHI PROGRAMS
      "i"'"""" ..... iiiiiimiiiiimia PLANNED DOE PROGRAMS
                Figure  2.   Prioritized  RD&D Opportunities.
                                      283

-------
(Chiyoda 121) gas/liquid contactor designs.   In the dual alkali area, EPA
is planning to demonstrate the use of limestone as a regenerant with start-
up planned in early 1979 at the dual alkali test facility at Southern
Services Company's Scholz Plant.  In the regenerable process area, TVA and
EPA have plans to cosponsor a magnesium oxide pilot plant to gather data for
a full-scale system to be built by TVA.  EPRI is planning RESOX and sorption/
steam stripping evaluations leading to a large-scale demonstration.

     The final report documenting this project is in preparation and is
expected to be published in the near future.
                                    284

-------
           SESSION 4

      UTILITY APPLICATIONS

MICHAEL A. MAXWELL, CO-CHAIRMAN
  JULIAN W. JONES, CO-CHAIRMAN
             285

-------
                 STATUS OF FLUE GAS

                   DESULFURIZATION

                IN THE UNITED STATES
                     Prepared by

       Bernard A.  Laseke and Timothy W. Devitt

              PEDCo Environmental,  Inc.
                 11499 Chester Road
               Cincinnati,  Ohio 45246

                  For Presentation
              at the Fifth Symposium on
              Flue Gas Desulfurization
Sponsored by the U.S.  Environmental Protection Agency
  Industrial Environmental Research Laboratory-RTF
                   March 4-8,  1979

                  Las Vegas,  Nevada
                       286

-------
                            SECTION 1
                          INTRODUCTION

     Pedco Environmental, Inc., under contract to the Environ-
mental Protection Agency  (EPA), has closely monitored the growth
and use of FGD technology by utilities in the United States and
has evaluated FGD technology on both a general and a site-specific
basis.  The site-specific evaluations are based on visits to
plants with operating FGD systems, during which process design
and performance information and capital and annual cost data are
obtained.  A series of reports has been prepared and published on
the major operational installations.
     Perhaps the most significant product of this project is the
periodic summary reports that are issued.  These reports provide
updated data on the number and capacity of the systems in opera-
tion, under construction, or planned and describe the performance
of the operating systems during the reporting period.  Utility
representatives, system suppliers, system designers, regulatory
personnel, and others contribute this information voluntarily to
facilitate the timely transfer of information in this key tech-
nological area.  Information provided by utility representatives
with operating systems is reported essentially as obtained;
little attempt is made to analyze or interpret the data.  Infor-
mation provided by system suppliers and other sources is
                               287

-------
confirmed with the appropriate utility prior to publication.
     The following sections of this paper address some of the
highlights of the ongoing survey program.
                               288

-------
                            SECTION 2




                   OVERVIEW OF FGD TECHNOLOGY






     Table 1 lists the total number of FGD installations and




their equivalent electrical capacities  (in MW) as of the end of




November 1978.




    TABLE 1.  NUMBER AND CAPACITY OF U.S. UTILITY FGD SYSTEMS
Status
Operational
Under construction
Planned :
Contract awarded
Letter of intent signed
Requesting/evaluating bids
Considering FGD
Total
Number
of units
46
43
20
3
5
27
144
Capacity,
MW
16,054
17,297
10,690
1,960
3,100
13,406
62,507
     As the table shows, 144 FGD systems representing an equiva-



lent electrical capacity of 62,507 MW are in operation, under



construction, or planned.  Of these systems, 46 are operational



(16,054 MW), 43 are under construction  (17,297 MW), and 55 are



planned (29,156 MW).  There are another 55 to 60 plants that will



be using FGD systems, but information regarding these systems is



not yet ready for public release.  These systems will have an



equivalent electrical generating capacity between  36,000 and



41,000 MW.  To date, 16 systems  (1555 MW) have been shut down for



various reasons.  Three of these systems  (425 MW)  are continuing



to operate, removing primarily fly ash.  However,  the systems do
                              289

-------
remove some sulfur dioxide (35 to 50 percent) as a result of



alkaline additives put into the scrubbing solution for pH control.






GROWTH TRENDS



     Figure 1 illustrates both the number and equivalent capacity



of FGD systems as a function of year of start-up.  The number of



systems requires clarification.  A system is defined on the basis



of inlet gas ducting configuration.   A module or several modules



that are commonly ducted to one or more boilers comprise a



single system.  Thus a single FGD module that treats flue gas



from only one bo'iler is considered a system, just as multiple FGD



modules connected through a common duct to multiple boilers are



considered one system.  On the other hand, when a plant has



several boilers ducted to a number of distinct modules or groups



of modules without any common ducting between them, that plant is



considered to have separate FGD systems.



     The values in Figure 1 represent all the FGD systems installed



and operated from March 1968 to the end of November 1978, as



well as those under construction and planned for installation



from December 1978 to 1986.  Systems planned for operation beyond



1986 are excluded because they are in the preliminary planning



stage and public information is limited.



     Figure 2 shows the increase in the projected capacity of FGD



systems as a function of the year the estimate was prepared.  In



November 1974, for example, a total of 37,836 MW of capacity



could be identified as in operation, under construction, or
                                290

-------
    60




    56




    52




    48




    44




 2  40




    36
ro
 o
 s  32
    28




    24




    20




    16




    12
                       I   I  I  I   I  I  I   f  1   I  1
140




130




120




110




100




90




80




70




60




50




40




30




20




10
     196669 70 71  72 73 74 75 76 77 78 79 80 81 82 83  84 85 86

                            Year
Figure  1.   FGD operating  capacity through  1986
                          291

-------
   65


   60


   55


   50


   45


   40

 £

*£ 35
 >-

 I 30
 s

 "" 25


   20


   15



   10
                                                 T
            Total
              (27,768)
                                                  (62.507)
                                                (29.156)
                              Construction

                                 (11.8
                                        .914)
                          •(3.796)
                                      I
            1974         1976         1977
          (NOVEMBER)       (MARCH)     (NOVEMBER)
                          YEAR OF ESTIMATE
                                                   1978
                                                 (November)
Figure  2.   FGD capacity as a  function  of status
                 and  year  of estimate
                           292

-------
planned; and by the end of November 1978, 62,507 MW is accounted



for (this does not include approximately 36,000 to 41,000 MW of



planned capacity that cannot be identified at this time).



     From 1974 to late 1978 the number of operating systems



reported increased from 19 to 46, a 242 percent increase, while



the equivalent generating capacity increased from 3,291 MW to



16,054 MW, an increase of 488 percent.  The average system size



has increased from 173 MW to 434 MW in the same time period, and



the capacity associated with full-scale systems has increased



from 2,360 MW to 15,882 MW.  (Full-scale systems are defined as



those that are available for commercial operation on fossil-fuel-



fired boilers having a minimum power generating capacity of



100 MW.)



     A general uncertainty surrounds projections of the power-



generating capacity of new coal-fired boilers.  One such pro-



jection, developed by PEDCo Environmental from a number of sources



(References 1, 2, and 3), is shown in Figure 3.  Current coal-



fired capacity is indicated as approximately 265 GW, which repre-



sents 47 percent of the total power-generating capability of the



electric utility industry in the United States.  By 1986, this



figure is expected to rise to 363.2 GW and to represent 45 per-



cent of the projected total power capacity; and by 1990, it is



expected to be 440 GW and to represent 44 percent of the pro-



jected total power capacity.



     Figure 3 also shows the projected application of FGD systems



through 1986.  Two major categories of FGD systems are depicted—



committed and uncommitted.  The committed FGD systems  (listed  in





                               293

-------
450
400 -
1	1	1     I
                          COAL-FIRED
                           UTILITIES
                      UNCOMMITTED,
                       UNANNOUNCED
                       FGD SYSTEMS
                             COMMITTED,
                             FGD SYSTEMS
   75  76  77  78  79   80   81   82  83  84   85   86   87 88  89  90
    Figure 3.   Projections  of coal-fired generating capacity
      from 1975 to 1990 and FGD capacity from 1975  to 1986.
                                294

-------
Table 1) are those that are currently in service, under construc-



tion, or planned.  Virtually all of these systems are either



currently operating or are committed to become operational at



some future date.  Uncommitted FGD systems are those that cannot



be included in the committed group at this time because infor-



mation regarding their status is not ready for public release.



About 55 to 60 systems representing approximately 36,000 to



41,000 MW of generating capacity fall into this latter category.



The premature stage of their planning, developments in on-going



litigation, and the determination of applicable revised New



Source Performance Standards (NSPS) and the Clean Air Act Amend-



ments of 1977 generally preclude the inclusion of these systems.



Although the revised sulfur dioxide NSPS had not been promulgated



by the EPA at the time this paper was written, indications are



that new coal-fired units that were not under construction by



December 31, 1978 (requiring a construction permit by March 1,



1978) will be required to have some type of continuous control



device.  Impending NSPS covering sulfur dioxide are expected to



be stringent enough to require the use of FGD technology because



of its current level of commercial development.  Thus, whereas



approximately 6 percent of current total coal-fired power-generating



capacity is now controlled by FGD, it is expected that about 25



percent of the total power-generating capacity from coal-fired



facilities will be controlled by FGD by 1986.
                               295

-------
NEW VERSUS RETROFIT



     Figure 4 compares the application of new versus retrofit



systems.  Most of the initial applications were retrofitted.  In




1975, for example, 60 percent of the operational FGD systems were



retrofitted, whereas in 1980 about 70 percent will be on new



systems.  By 1985, about 75 percent of the operational FGD



systems should be new unit applications.   Any additional



retrofits will stem primarily from systems required because of



local regulatory action or small capacity demonstration projects.






PROCESS TYPES



     The three primary methods are available for categorizing



flue gas desulfurization processes:  physical mechanism, chemical



mechanism, and end-product mechanism.  Physical mechanism refers



to the phase in which sulfur dioxide removal is performed, i.e.,



wet or dry; chemical mechanism refers to the reagent used; and



end-product mechanism refers to regenerable systems (in which



sulfur dioxide is recovered in a usable, marketable form) and



nonregenerable systems (in which sulfur dioxide must be disposed



of as a nonrecoverable waste material).  Table 2 summarizes,



according to physical to mechanism, the systems that are opera-*



tional, under construction, or for which a contract has been



awarded.
                                296

-------
     70
     60
     50
     40
   o
  UJ
  Ul

  s
     30
     20
     10
          I   I   I
I   I   I
                    TOTAL
       75  76  77  78  79  80  81  82  83  84  85  86


                        Ye«r
Figure  4.   FGD operating capacity for  new and

    retrofit installations  through 1986.
                     297

-------
     TABLE 2.  COMMITTED FGD CAPACITY BY PHYSICAL MECHANISM
Physical
mechanism
Wet
Dry
Total
FGD capacity, MW
Operational
16,054
0
16,054
Under
construction
16,897
400
17,297
Contract
awarded
9,685
1,005
10,690
Total
42,636
1,405
44,041
     Table 3 summarizes the systems that are either operating or



planned, according to committed process type.



     As indicated in both Tables 2 and 3, the vast majority of



operating experience has been obtained with direct, calcium-



based, wet-phase, nonregenerable FGD systems.  Of the total



active process-committed capacity of 50,377 MW, calcium-based



systems account for 44,530 MW, or 88 percent.



     Table 3 indicates an interesting trend in the utility indus-



try's preference for limestone over lime processes.  Limestone



systems constitute approximately 59 percent of current calcium-



based operating capacity, 59 percent of the capacity under con-



struction, and 77 percent of that planned, or a total of about 65



percent.  These figures show the industry's preference for



limestone processes now and indicate an even stronger preference



in those systems committed for operation within the next 5 years.





EMISSION LIMITING STANDARDS



     Table 4 summarizes the FGD systems according to the regula-



tory standards they must meet.  Of the 144 active systems, 65



(29,713 MW) are designed to meet the existing NSPS; 61  (26,567



MW) are designed to meet state standards more stringent than the
                                298

-------
    TABLE 3.   DISTRIBUTION OF FGD SYSTEMS BY CHEMICAL PROCESS


Process
Limestone
Limeb/c
Lime/Limestone
Sodium carbonate
Magnesium oxide
Wellman Lord
Dual alkali ,
Aqueous carbonate
Citrate6
Totalf
FGD capacity, MW

Operational
8,734
6,070
20
375
120
735
0
0
0
16,054
Under
construction
8,687
6,029
330
509
0
180
1,102
400
60
17,297

Planned
10,848
3,482
330
0
1,326
940
0
100
0
17,026

Total
28,269
15,581
680
884
1,446
1,855
1,102
500
60
50,377
  Includes alkaline fly ash/limestone and limestone slurry
  process design configurations.


  Includes alkaline fly ash/lime  and lime slurry process design
  configurations.

Q
  Includes nonregenerable dry collection process design and
  nonregenerable wet scrubbing process design configurations.


  Includes nonregenerable dry collection process design and
  regenerable process design configurations.


6 This system is being installed  at St. Joseph Minerals' G.F.
  Wheaton Plant and is listed as  a utility FGD system because
  the plant is connected by a 25-MW interchange to the Duquesne
  Light Company-


  Because the processes for all planned systems are not known,
  the totals in this table are less than those in Table 1.
                               299

-------
existing NSPS* requirement; and 17  (5,427 MW) are designed to

meet regulations less stringent than the existing NSPS.  It is

interesting to notenthat half of the 46 active, operational FGD

systems  (Table 1) are now meeting standards more stringent than

the existing Federal NSPS.

       TABLE 4.  NUMBER AND CAPACITY OF ACTIVE FGD SYSTEMS
            FOR REGULATORY CLASSIFICATION CATEGORIES
Regulatory classification
Existing Federal NSPS
More stringent than existing
Federal NSPS
Less stringent than existing
Federal NSPS
Undetermined
Total
Systems
65
61
17
1
144
Capacity, MW
29,713
26,567
5,427
800
62,507
HIGH VERSUS LOW SULFUR COAL APPLICATION

     The design and operation of FGD systems for high and low

sulfur coal application represents another area of interest,

especially with regard to the viability of such systems on

boilers firing high sulfur coal.  Because of the ambiguities

inherent in the terms high and low sulfur coal, we have defined

these as follows for purposes of this paper:  low sulfur coal is

any coal whose combustion will result in emissions equal to or

less than 1.2 Ib of sulfur dioxide per 10  Btu and high sulfur

coal is any coal whose combustion will result in a higher emis-

sion value.  Using these definitions, the following observations

hold:

     0    Among the operating FGD systems, approximately 85
          percent of the MW capacity is for high sulfur coal
          application.
 The Clean Air Act NSPS value of 1.2 Ib of sulfur dioxide per
     Btu heat input to the boiler.
                               300

-------
     0    Among the systems under construction, approximately 75
          percent of the MW capacity is for high sulfur coal
          application.

     0    Among the planned systems, approximately 90 percent of
          the MW capacity is for high sulfur coal application.


SYSTEM SUPPLIERS

     Approximately 30 companies offer FGD systems for application

to utility boilers.   Table 5 lists the companies that have

supplied the systems that are in service, under construction, or

under contract and identifies the number of systems and their

capacities.  Included in these totals are systems that are no

longer used for sulfur dioxide removal.


INSTALLATION SCHEDULES

     Schedules for the installation of FGD systems can vary

greatly, primarily because of front-end activities relating to

process selection and design.  The period from startup to accep-

tance by the client can also vary depending upon system per-

formance, and contractual agreements.  The period between contract

award and initial system start-up is less variable however.  An

analysis of the schedules of 35 systems indicates a range of 18

to 60 months with a mean of 32 months.  The longer lead times are

generally associated with new systems that are being constructed

as an integral part of a new power generating system.
                              301

-------
                   TABLE  5.   MAJOR FGD SYSTEM  SUPPLIERS
System supplier
A.D. Little/Combustion
Equipment Associates
Air Correction Division, UOP
American Air Filter
Babcock l Wilcox
Hue 1 I/En v iro tech
Chemico
Chiyoda International3
Combustion Engineering
Davy Power gas
FHC
Mitsubishi International
Monsanto
Peabody
Pullman Kellogg0
Research Cottrell
Riley Stoker/Environeering
Rockwell International
United Engineers
Western/Niro
Wheelabrator-Frye/Rockwell
Current status
Operational
No.
6
5
3
2
0
7
1
9
3
0
0
0
1
0
5
2
0
1
0
0
MW
1545
1540
667
1100
0
3745
23
3073
735
0
0
0
225
0
2151
580
0
120
0
0
Under
Construction
No.
1
2
3
5
1
2
0
5
1
1
0
0
4
3
7
1
0
0
0
1
MW
277
934
985
2069
575
BOO
0
2675
180
250
0
0
1625
1525
3203
180
0
0
0
400
Contract
No.
3
1
0
4
0
1
0
3
2
0
2
0
0
1
1
0
1
2
1
0
awarded
MW
1900
720
0
1800
0
527
0
1275
940
0
980
0
0
670
1333
0
100
1510
455
0
Terminated
No. J '
1
2
0
1
0
2
1
3
0
0
0
1
1
1
0
0
0
0
0
0
MW
20
220
0
167
0
245
23
665
0
0
0
110
163
160
0
0
0
0
0
0
Total
No.
11
10
6
12
1
12
2
19
6
1
2
1
6
5
13
3
1
3
1
1
MW
3742
3414
1652
5136
575
5317
46
7703
1855
250
980
110
2013
2355
6147
760
100
1630
455
400
The Scholz  prototype was reactivated to demonstrate a  new process design configuration.  Thus,
the entries in the operational  and terminated categories refer to the same system.

The original Lawrence 4 and 5  limestone injection systems were modified/replaced with a  second-
generation  rod scrubber/spray  tower design.  Thus,  the terminated category reflects the  operational
experience  gained with these systems.

The Mohave  prototype is included  in the terminated  category although the supplier was not a
participant in the Mohave Test  Modules Program.
                                               302

-------
                            SECTION 3

            GENERAL CONSIDERATIONS IN THE APPLICATION
                AND PERFORMANCE OF FGD TECHNOLOGY
     Significant development of FGD technology in the United

States dates from about the 1950's when bench-scale and limited

pilot plant programs were initiated.  Major pilot plant investi-

gations first started in 1961, and between 1961 and 1978 more

than 60 systems representing generating capacity of approximately

75 MW were investigated at the pilot plant level in the utility

sector.  Concurrent with later pilot plant investigations, pro-

totype, demonstration, and full-scale systems were installed.

The first commercial application of an FGD system on a utility

boiler occurred in 1968.  Since then, 62 systems representing a

generating capacity of approximately 17,600 MW have been operated

at the prototype, demonstration, and full-scale levels.  As of

the end of November 1978, 46 systems representing a generating

capacity of 16,054 MW were in service, and another 98 systems

representing 46,453 MW were under construction or planned.

     Since the early investigations considerable progress has

been in the development of FGD technology, and FGD is now con-

sidered to be the most commercially developed means of continuous

control of sulfur dioxide emissions from coal-fired boilers.
                               303

-------
     The evolution of FGD technology from the limited pilot plant



level to the full-scale commercial level can be attributed to



several general process design and application considerations.



Although it is difficult to quantify the impact that many of



these factors have had on the development of the technology, an



attempt is made to identify some of the general contributing



factors in the balance of this section.  (Section 4 provides more



specific design and performance information on current technology



trends.)





PROCESS DESIGN STRATEGY



     Several general tendencies are evident in recent FGD design



strategies.  Generally, system designs incorporate an increased



degree of flexibility and reliability.  Specifically, trends are



toward the sparing of modules and ancillary components and the



designing of less interdependent systems (i.e., systems in which



major unit operations are not strongly affected by upstream com-



ponent performance).





SYSTEM APPLICATIONS



     Many recent FGD facilities are installed on large base-



loaded units designed to fire coal from a specific source.  This



generally results in a flue gas with more constant and stable



characteristics, which can improve system reliability because the



system does not have to respond to as dramatic a variation in



flue gas flow rate and composition.  In many of the original FGD



applications, the systems were required to operate on widely
                                304

-------
varying loads (cycling and peak) and coal types  (low sulfur




western, high sulfur eastern, and blends); such situations often



demanded response to conditions beyond their process control




capability.  As a result, variations in the reagent feed rate,




loss of chemical control, and the incidence of chemical and



mechanical problems caused numerous forced outages  and lower



dependabilities.






SYSTEM SUPPLIER AND ARCHITECTURAL-ENGINEERING EXPERIENCE



     Later FGD system designs have benefitted from experience



gained in the operation of first-generation systems.  Building on



this experience, system suppliers and designers are providing



better process design configurations and materials of construc-



tion.  That many suppliers now offer broader guarantees covering



sulfur dioxide removal, particulate loading, mist loading, waste




stream quality/quantity.- power consumption, water consumption,



reagent consumption, reheat energy consumption, and availability




is indicative of this trend.






UTILITY EXPERIENCE



     The utilities have also been gaining valuable operating and



design experience.  Many utilities have conducted or participated




in FGD pilot plant programs and are thus better prepared to oper-



ate demonstration and full-scale systems.  Operation of their




first demonstration or full-scale system has also led to improved



design and operation of subsequent systems.
                               305

-------
REGULATORY AGENCY ATTITUDES



     As FGD technology has evolved from a research, development,



and demonstration effort to a means of continuous compliance with



applicable regulations, local, state, and Federal regulatory



agencies have changed their attitudes toward enforcement, com-



pelling utility companies to improve the reliability of FGD



systems.






PROCESS CHEMISTRY



     Although scale and corrosion are still encountered and are



sometimes still severe, general knowledge concerning scale forma-



tion and the occurrence of corrosion has greatly improved.  As a



result, systems are being designed and operated so that problems



experienced by the earlier units will not be encountered.
                                 306

-------
                            SECTION 4




                  CURRENT TECHNOLOGICAL TRENDS






     Considerable progress has been made in the development of



conventional and emerging or advanced FGD processes.  Much of



this information has been acquired from the design and operation



of first-generation FGD systems and translated into more effective



designs and improved operation of newer systems.  This section



summarizes these emerging processes and presents a brief overview



of their current status.






EMERGING PROCESSES



     For the purposes of this discussion, processes within the



emerging or advanced category are defined as those that incorpo-



rate major design and operating changes and thereby differ



significantly from conventional direct lime/limestone processes.




Of the processes so categorized, several have been evaluated at



pilot and prototype development levels, and a few have progressed



to the installation and operation of demonstration or full-scale



units.  Table 6 provides a brief summary of the emerging pro-



cesses and highlights their current level of development and the




extent of operating experience.



     As is evident from Table 6, most of the previous operating



experience has been with wet-phase sodium- and magnesium-based
                               307

-------
             TABLE  6.   MAJOR EMERGING
..WESTIGATED IN THE UNITED STATES
o
00
• . ' -.'- .:•
•ueous
carbonate
Catalytic
oxidation
Chiyoda
Thoroughbred
101
Copper oxide
adsorption
Dual
alkali


Developer
Rockwell
International
Monsanto
Chiyoda
International
Shell/Universal
Oil Products
A.D. Little/
Combustion
Equipment
PMC
Buel I/En v iro tech
Current
level of
development
100-MW system
(planned)
100-MW system
(terminated)
20-MH system
(terminated)
Pilot plant
277-MW system
(construction)
2SO-MH system
(construction)
S7S-MH system
(construction)
Previous
operating
experience
Mohave pilot plant
test program
Wood River teat
program
Scholz prototype
test program
Big Bend test
program
Scholz prototype
test program
Industrial systems
and utility pilot
plants
Gadsby pilot plant
test program
Remarks
Full-scale application not yet
demonstrated. 100-MW demon-
stration system scheduled for
service in 1980.
No further process development.
Development of the process has
ceased in favor of a new design
concept (Thoroughbred 121 which
employs limestone reagent in a
jet bubbler reactor) .
Process available for prototype
or demonstration application.
No systems planned at present.
Full-scale application not yet
demonstrated. A 277-MW demon-
stration system scheduled for
initial operation in early 1979.
Full-scale application not yet
demonstrated. A 250-MW system
is scheduled for service in 1979.
Full-scale application not yet
demonstrated. A 575-MW system
is scheduled for service in 1979.

-------
             TABLE  6.   (Continued)
LO
O
IO
Process
Magnesiua oxide








Sodium
carbonate









W«llBan Lord





Developer
Chemico



United Engineers




A.D. Little/
Combustion
Equipment
Associates


Universal Oil
Products



Davy Powergas





Current
level of
development
150-HW system
(terminated)
95-MW system
(terminated)
600-KW system
(planned)



Three 115-MW
systems
(operational)



509-HW system
(planned)



115-MW system
(operational)
375-HW system
(operational)
340-MW system
(operational)
Previous
operating
experience
Hystic test program;
Dickerson test
program

Eddystone test
program



Reid Gardner
Station




Jim Bridger pilot
test program



Crane test program





Remarks
Process demonstrated on full-
scale oil- and coal-fired
boilers. System now offered
for commercial application.
Eddystone 120-MH prototype test
program still in progress. A
full-scale 600-HH system is now
planned for TVA's Johnsonville
station.
Three full-scale sodium carbon-
ate (trona) FGD systems have been
in service on coal-fired boilers
at the Reid Gardner Station
(Nevada Power). System perform-
ance has been good.
A full-scale sodium carbonate
(30t sodium carbonate purge
solution from soda ash plant)
FGD system is now being planned
for the Jim Bridger station.
115-MW NIPSCO/EPA test program
still in progress at Mitchell.
Two full-scale systems have
recently started operations at
Public Service of New Mexico's
San Juan Station.

-------
TABLE 6.   (Continued)
Process
Dry adsorption



Dry collection






















Developer
Foster Wheeler
Bergbau Forschung


Wheel abrator-Fr ye
Rockwell Interna-
tional

Joy/Niro






Babcock t Wilcox






Carborundum/
Delaval



Current
level of
development
20-MW system
(terminated)


410-MW system
(planned)


455-MW system
(planned)





550-MW system
(planned)





Pilot plant




Previous
operating
experience
Scholz prototype
test program


Leland Olds and
Bowen Engineering
pilot plant test
programs
Leland Olds pilot
plant test
program




Basin electric
pilot plant test
program




Leland Olds pilot
plant test
program


Remarks
No further development of system
reported. Further evaluation of
the sulfur reduction component
(RESOX) in progress in Germany.
Successful testing has resulted
in the planning of a full-scale.
spray dryer and fabric filter
FGD system at the Coyote Station.
Successful pilot plant testing
has resulted in the planning of
a full-scale system at Antelope
Valley. This system will use
lime slurry as the reagent in a
2-stage atomizer/fabric filter
design configuration.
Successful pilot plant testing
has resulted in the planning of
a full-scale system at La ramie
River. This system will use liM
slurry as the reagent in a con-
figuration which employs an ESP
as the second stage collector.
Process similar in design to the
Wheelabrator-Frye/R. I. and
Joy/Niro processes. No full-
scale applications have yet been
announced.

-------
processes that produce a recoverable, marketable byproduct or a



high-quality filter cake.  As might be expected, this substantial



experience has resulted in the commercial application of dual



alkali, Wellman-Lord, sodium carbonate, and magnesium oxide



scrubbing systems.  Three commercial dual alkali systems  (Cane



Run 6, A.B. Brown 1, and Newton 1) are now approaching startup.



One demonstration (D.H. Mitchell 11) and two commercial  (San Juan



1 and 2) Wellman-Lord systems are currently in service, and



sodium carbonate systems have been operated commercially at three



installations (Reid Gardner 1, 2, and 3).  One demonstration



magnesium oxide system (Eddystone 1) is currently in service, two



such systems (Mystic 6 and Dickerson 3) have been terminated, and



one full-scale system (Johnsonville) is planned.



     The most recent promising development in emerging processes



involves dry collection systems.  Integrated processes designed



to remove two or more pollutants simultaneously from the flue gas



of a coal-fired boiler have commanded considerable interest for



econpmic and operational reasons.  The success of fabric filters



in removing particulates from the flue gases of coal-fired boilers



prompted investigations into the feasibility of using these fil-



ters to control both particulates and sulfur dioxide.  Tests have



been conducted using the injection of dry powdered nahcolite  (a



mineral form of sodium bicarbonate), calcium oxide, and calcium



hydroxide into the flue gas stream and onto fabric filter bags



for the removal of sulfur dioxide.  Although impressive results



were obtained with nahcolite, problems in meeting government
                                311

-------
requirements regarding nahcolite excavation in the producing




areas of northern Colorado prompted research into other possible



process design configurations.  When it was found that using a



spray dryer would eliminate the exclusive need for nahcolite,



Wheelabrator-Frye and Rockwell International undertook a joint



test program at Lelands Olds involving a two-stage system that



combines a spray dryer and fabric filter.  The spray dryer, the



first stage, accomplished alkali injection and primary sulfur



dioxide removal.  The downstream fabric filter functioned as a



second-stage sulfur-dioxide absorber and collection of flue gas



particulates.  Soda ash, trona, and lime, limestone, and fly ash



slurries were tested as possible reagents in this system.  Soda



ash produced successful results.  The data indicated that sulfur



dioxide removals ranged from 48 to 98 percent a±. soda ash utiliza-



tions ranging from 96 to 65 percent for sulfur dioxide loadings



ranging from 800 to 2800 ppm.



     As a result of this successful testing, Wheelabrator-Frye



and Rockwell International were awarded a turnkey contract for a



full-scale system at Coyote 1, a 410-MW coal-fired unit.



     Additional pilot plant testing involving different dry



collection process design configurations has continued at Leland



Olds, with systems supplied by Carborundum/DeLaval and Joy/Niro



Atomizer.  Babcock & Wilcox is also involved in a similar pilot



plant evaluation of a dry-phase, two-stage collection system




which is being conducted at another station.  These programs



involve the use of less expensive reagents, such as lime slurry,
                              312

-------
 in various types of design configurations.  Based on successful

pilot plant testing, Joy/Niro has been awarded a cbntract for a

full-scale dry collection system at Antelope-Valley 1 of the

Basin Electric Company and Babcock & Wilcox has been awarded a

contract for a full-scale dry collection system at Laramie River

3 of the Basin Electric Company.


CONVENTIONAL PROCESSES

     Conventional processes include all direct lime and limestone

systems.  Because these systems are the most widely applied, they

are the ones with the most operating experience.  Furthermore,

they will have the greatest utilization in the very near future.

For these reasons, lime and limestone systems have been subjected

to extensive investigations, the results of which are summarized,

along with general conclusions concerning process design.

Process Chemistry—Scaling, Sulfur Dioxide Removal, and Reagent
Utilization
            I
     The use of reagents with low reactivity, such as limestone

or lime, for sulfur dioxide removal from gas streams that can

vary widely in flow and composition over short periods of time

has resulted in several obvious and sometimes severe chemical

limitations, including scale formation, low sulfur dioxide

removal, and poor reagent utilization.  A number of important

findings regarding process chemistry and its effects on system

performance have been made in recent years, and many of the

process design and operating features that are now being incor-
                               313

-------
porated into commercial systems are using these findings to



improve performance.   A brief summary of these features follows.



     Two basic modes of lime/limestone FGD system operation will



enable scale-free operation:  coprecipitation and desupersatura-



tion.



     Coprecipitation involves the removal of calcium sulfate from



the system as part of a calcium sulfite/sulfate solid solution.



If a system is operated so that the maximum oxidation in the



slurry circuit is about 16 percent, the scrubbing liquor remains



subsaturated with respect to calcium sulfate (gypsum),  and no



hard scale occurs.  If the degree of oxidation exceeds this



level, more calcium sulfate is formed in the slurry circuit than



can leave the system in a coprecipitated form.   This causes the



system to operate supersaturated with respect to gypsum.  If



relative saturation levels approaching the critical value of 1.4



is reached, the formation of hard scale can occur within the



system.



     Desupersaturation involves the removal of calcium sulfate



from the system through the use of calcium sulfate  (gypsum) seed



crystals, which provide nucleation sites for the precipitation of



calcium sulfate as well as through the calcium sulfite/sulfate



solid solution.  The seed crystals control sulfate  scaling in a



closed-loop system operating in a supersaturated mode.  Crystal



growth occurs on the seed crystals, the sulfate is  removed from



the system as gypsum, and relative saturation levels are kept



below the critical 1.4 value.
                                314

-------
     Another approach to improving the chemistry of lime/lime-



stone slurry systems so as to reduce scaling, increase sulfur



dioxide removal, and improve reagent utilization is the use of



additives in the slurry.



Use of Magnesium Additives—



     Magnesium additives have proved to be effective.  Increasing



the magnesium ion concentration increases the liquid-phase



alkalinity of the scrubbing slurry and increases the amount of



sulfite and sulfate the scrubbing slurry can hold without exceed-



ing solubility limits.  The overall effect is a subsaturated



operation that produces higher sulfur dioxide removal efficiencies



and higher utilization.  Experimental operating experience with



magnesium additives has been accumulated at the Shawnee TVA/EPA



Akali Scrubbing Test Facility and Paddys Run of Louisville Gas



and Electric.  Experimental and full-scale operating experience



has been obtained at Phillips and Elrama of Duquesne Light, Bruce



Mansfield of Pennsylvania Power, and Conesville of Columbus and



Southern Ohio Electric.



Use of Other Additives—



     The use of organic acids as additives is also under con-



sideration.  The use of carboxylic acids in lime/limestone



scrubbing has been tested by the Tennessee Valley Authority (TVA)



and TVA/EPA.  This research has concentrated on the use of benzoic



acid and, more recently, adipic acid at Shawnee.  These acids are



generally stronger than carbonic acid, but weaker than sulfurous



acid.  The addition of these acids has two effects:  first, it
                               315

-------
aids mass transfer by buffering the pH of the liquid film at the



gas/liquid interface; and second, because sulfurous acid is a



stronger acid, the benzoate or adipate ion acts as a base in the



sulfur dioxide absorption step.  Thus, the addition of organic



acid acids increases the total liquid phase alkalinity of the



scrubbing liquor in much the same fashion as an increase in



alkalinity because of magnesium ion.  Intensive testing with



adipic acid was recently performed at the TVA/EPA Shawnee alkali



scrubbing test facility.  In July 1978 initial test runs were



performed without adipic acid to establish base lines for both



lime and limestone scrubbing.  These runs were followed by adipic



acid testing, which continued throughout the balance of the year.



     The preliminary results of this test program indicate agree-



ment with initial expectations of higher sulfur dioxide removal



efficiencies, and higher reagent utilizations.  It has also been



shown to be effective when used in conjunction with forced oxida-



tion and when chlorides are present-conditions which adversely



affect magnesium additives.  One negative result, however, has



been the unexpectedly high deterioration or decomposition of



adipic acid that takes place in the scrubber.  Actual feed rates



of adipic acid were two to three times higher than could be



accounted for in the system discharge sludge.



Design Changes—



     A number of process design innovations also have been



developed to eliminate scaling, increase sulfur dioxide removal



efficiency, and improve reagent utilization.  Forced oxidation is
                                316

-------
one technique that has been successfully piloted and used in com-



mercial installations.  On a number of systems the use of forced



oxidation has contributed to scale-free operation, to meeting or



exceeding design sulfur dioxide removals, and to achieving high



reagent utilization levels and, most importantly, in improving



the quality of the waste sludge.  This has been especially true



for limestone systems treating flue gas with low sulfur dioxide



loadings.  For these applications, the low sulfur dioxide levels



coupled with the long liquid retention times result in high



"natural" oxidation levels.  This makes forced oxidation a



particularly attractive method to improve sludge quality and



minimize scaling, increase sulfur dioxide removal and improve



reagent utilization in these systems.



Operating changes—



     Several operating parameters have an important impact on



scaling, sulfur dioxide, and reagent utilization.



     Control of pH—Excluding all other factors, differences



in optimum operating pH can affect the performance of lime/lime-



stone slurry systems in two ways:  operation at low pH generally



promotes the formation of hard calcium sulfate scale (gypsum),



and operation at high pH generally promotes the formation of



softer calcium sulfite scale.  Operating experience indicates



that optimum pH levels are generally maintained between 8.0 to



8.5 for lime and 5.5 to 6.0 for limestone.
                              317

-------
     Solids level—If all other variables are held constant,



increases in slurry solids levels will increase the amount of



seed crystal area available for homogeneous crystallization.



This is especially true for systems that control sulfate levels



by desupersaturation.  For these systems, an optimum amount of



slurry seed crystals is maintained in the system by maintaining



an optimum level of slurry solids.  Thus, when the solids level



drops, the seed crystal level drops correspondingly and causes



the impairment or loss of homogeneous crystallization, the onset



of heterogeneous crystallization, and subsequent scale develop-



ment.  Some of these systems aid desupersaturation by forcibly



oxidizing all the sulfite to sulfate and then precipitate the



calcium sulfate with the aid of gypsum seed crystals.  Minor



episodes of sulfate scaling have also occurred in these systems,



in every case as a result of dilution of the slurry solids level



after the mist eliminator wash water rate was increased to im-



prove cleanliness.  The increased levels of makeup water in the



system decreased the slurry solids level and the seed crystal



level, which impaired desupersaturation and resulted in scale



formation.  Reestablishment of slurry solids levels prevented



further episodes of scaling.



     Liquid-to-gas ratio (L/G)—If all other variables are held



constant, increasing the L/G reduces the sulfur dioxide pickup



per volume of scrubbing liquor.  Thus, the relative saturation



in the circulating slurry can be reduced by increasing L/G,
                                318

-------
assuming that desupersaturation takes place in the hold tank.



Process Chemistry—Corrosion



     In simple terms, corrosion is the dissolving of metal



surfaces.   The incidence of corrosion in lime/limestone slurry



FGD systems and design and operating measures taken to minimize



or eliminate this problem are discussed briefly in the following



paragraphs.



     Two corrosive agents are present in the process:  (1) sul-



furous and sulfuric acids, and (2) chlorides.  These two agents



contribute to several specific types of corrosion:  general



corrosion, pitting, crevice corrosion, intergranular corrosion,



stress-corrosion cracking, and erosion-corrosion.



     A number of successful design, construction, fabrication,



and operation measures have been developed to minimize the rate



of corrosion or prevent it altogether.  These measures are sum-



marized briefly.



     A selective process design approach developed by the major



system suppliers allows highly corrosive environments to be



isolated in discrete areas of the FGD system.  This approach



involves the separation of the scrubbing loop into separate



multiple loops so that a different set of chemical conditions is



maintained for quenching or prescrubbing, sulfur dioxide absorp-



tion, and mist eliminator washing  (wash trays).  In such designs,



the quencher or prescrubber bears the first full brunt of the



incoming hot flue gas.  It encounters the total chloride content



from the fuel fired without dilution and is the area in which
                               319

-------
low-pH,  chloride-laden, return water is used freely in lime/lime-



stone slurry systems.   Isolation of such a corrosive environment



to the quencher or prescrubber is advantageous because this area



is small and discrete enough that it can be constructed of chlor-



ide-resistant materials without drastically increasing system



cost.  Alloys that have been tested and specified for full-scale



systems are listed in ascending order of molybelmum content,



pitting resistance, and cost:  317L stainless steel, Incoloy



Alloy 825, Hastelloy G, Inconel Alloy 625, and Hastelloy C-276.



     In many of the initial lime/limestone slurry systems, the



incoming hot gas contacted the reactive absorbant suspension,



which resulted in the accumulation of solids at the wet/dry



interface.  These deposits in some cases provided convenient



sites for the accumulation of chloride at concentrations



approaching 50,000 ppm.  The result was severe episodes of



pitting, stress corrosion, crevice corrosion, stress-corrosion



cracking, and erosion-corrosion.  This problem has been largely



overcome by better control of process chemistry, use of self-



cleaning devices, selective use of superior construction mate-



rials, and the use of multiple-loop designs.  Better control of



process chemistry eliminates the formation of scale in the system



and thus prevents appearance of convenient chloride ion host



sites.  A number of systems are equipped with soot blowers in the



aPProach ducts to the scrubber modules, which allows the inevit-



able buildup of solids at the wet/dry interface to be cleaned



automatically and periodically.
                                320

-------
Emission Control Strategy—

     In recent years the trend in design of particulate and

sulfur dioxide emission control systems has been toward combined

electrostatic precipitator (ESP)/FGD or fabric filter (FF)/FGD

strategies over simultaneous or two-stage wet scrubbing strat-

egies.  This preference is due to the high reliability afforded

by ESP's and FF's, which enables selective bypass of scrubber

modules without reduction of load or shutdown of the unit.  Other

benefits include the following:

     0    The potential for corrosion at wet/dry interfaces and
          erosion-corrosion in the FGD systems is minimized.

     0    Exotic construction materials can be used more selec-
          tively and in less amounts.

     0    Balanced-draft and booster fans can precede rather than
          follow the FGD system.

     0    Sludge blending and stabilization processes that use
          dry fly ash as an additive are premitted.

Equipment Design Improvements—

     Specific design and operating improvements for FGD-related

equipment are as follows:

     Balanced-draft or booster fans—In addition to placement of

these fans upstream of the FGD system, another development is the

use of variable-pitch, axial flow fans.  The main advantage of

this design is its consistently higher efficiency  (versus centri-

fugal fans)  over the entire boiler operating range, which results

in a substantial power savings.  Other advantages are superior

flow control, arrangement flexibility, easy access and main-

tenance, less severe construction requirements, and increased

design reliability.
                             321

-------
     For scrubber modules, most suppliers now prefer 316 or  316L



stainless steel because this material has demonstrated superior



resistance to corrosion, errosion, and scale development compared



with carbon steel, 304 stainless steel, and 304L stainless steel.



This preference for 316 and 316L stainless steel is based pri-



marily on the smooth mating surfaces and molybdenum content of



these steels.  The former attribute minimizes the presence of



crevices that provide convenient sites for buildup of soluble



chloride.  The molybdenum content (2.50 to 2.75 percent minimum)



of stainless steel increases corrosion resistance to localized



attack such as pitting and crevice corrosion.



     For mist eliminators, which have been very susceptible to



corrosion, most suppliers now recommend the use of fiberglass-



reinforced plastics, polypropylene,  and corrugated plastics over



stainless steels and other alloys because these materials are



relatively lightweight, inexpensive, and do not corrode.  Re-



heaters have been especially susceptible to corrosion, and the



trend is toward indirect hot-air reheat because in-line reheat



systems have been subject to corrosion and tube plugging.



Process Design



     Several advances in the process design of lime/limestone



slurry FGD systems have improved system dependability and sulfur



dioxide removal.  The following subsections describe these



advances and some methods developed to reduce problems with major



FGD equipment.
                               322

-------
     Dampers—Bypass and isolation dampers are used to regulate



the flow of flue gas into and around the FGD system.  The primary



purpose of isolation is continuance of unit operation while the



scrubber modules are under maintenance.  Efficient and reliable



dampers allow maintenance crews to service the modules in an



efficient and timely manner.  Common designs include slide-gate



(guillotine), single-blade butterfly, and multiblade parallel



(louver) dampers.  Corrosion and erosion of the various types of



dampers and damper seals have been common.  In some cases,



dampers have failed or been so inefficient that the modules could



not be maintained during bypass situations.  The current trend is



toward two-stage louver dampers having a pressurized seal-air



system that maintains a positive pressure between them.  Pres-



surized seal air increases the energy demand of the system



because of increased fan power requirements, but it contributes



significantly to successful damper operation.



     Scrubbers—Several recent design innovations have increased



dependability and removal efficiency of scrubbers.  Cooling the



gas to its adiabatic saturation temperature prior to contact with



the scrubbing slurry increases sulfur dioxide removal capability



and minimizes the potential for scaling and corrosion at the



slurry/gas interface area.  Presaturators or quenchers were not



incorporated into the design of many of the initial FGD systems.



In such systems, the incoming, hot, pollutant-laden flue gas



contacted the suspended reactive absorbent and resulted in solids



accumulation and subsequent corrosion.  For this reason, presat-
                               323

-------
urators or quenchers that use clear liquor or spent slurry for

the absorbing stage (multiple  slurry loops) are now used  (or

plans call for their use) in systems that include dry-phase

particulate precollection.

     The trend in the design of lime/limestone slurry FGD systems

is away from Venturis and packed beds to spray towers and com-

bination towers.  The venturi design was abandoned largely

because the small liquid/gas contact time results in relatively

low sulfur dioxide absorption.  Scaling, plugging, and corrosion

of internals occurred in some of the packed-bed designs (fixed

and mobile) and tray towers.  Spray towers, on the other hand,

have few internal components in the gas/liquid contact zone and

therefore offer the potential for greater dependability because

there are fewer sites for deposition of solids in the form of

scale, collected fly ash, and unused reagent.  To date, spray

tower operation has been very successful.  Although high dependa-

bility and sulfur dioxide removal have been reported for almost

all the FGD systems incorporating spray tower designs, several

limitations also have been encountered.  Mass transfer limita-

tions tend to restrict spray tower design applications so that

only low- and medium-sulfur coal can be used in conventional

lime/ limestone slurry systems.*  In addition, the greater

'tendency for slurry carryover in spray'towers requires either

increased tower height or special mist eliminator designs  (wash
*When high-sulfur coal is burned, spray towers in service and
 scheduled for operation use or plan to use special reagents
 (carbide lime, magnesium-promoted lime, and limestone) to
 compensate chemically for mass transfer limitations.

                               324

-------
trays, bulk entrainment separators), both of which increase



capital and annual cost requirements.



     These limitations have given rise to the development of com-



bination towers.  These towers combine the features of venturi,



packed, tray, and spray towers into one module.  Examples of



combination tower designs now offered by some of the major system



suppliers include spray/packed towers, venturi/spray towers, and



tray/packed towers.  These designs offer greater flexibility



because extreme operating conditions can be segregated into



discrete areas of the scrubber, allowing separate chemical and



physical conditions to be maintained.  This permits the use of



the two-loop slurry concept, in which low pH liquor contacts the



entering flue gas in an initial scrubbing loop, where some sulfur



dioxide removal takes place.   High pH liquor is contacted with



the gas in the second scrubbing loop, where the bulk of the



sulfur dioxide removal takes place.  Spent slurry from this loop



is discharged to the first loop where the unused reagent is con-



sumed.  Fresh makeup reagent is added only in the second loop.



This type of design takes advantage of the concept of contacting



the flue gas containing the highest sulfur dioxide concentration



with the lowest liquor alkalinity and the highest liquor alka-



linity with the lowest sulfur dioxide concentration.  Performance



has verified the potentially high removals and utilizations



afforded by such designs.



     Reaction tanks—Coinciding with gas-side staging is liquid-



side staging, in which hold tanks are arranged in series to simu-
                               325

-------
late plug flow reactor designs.   (A plug flow design is one  that

allows the reacting liquor to flow through the reactor without

backmixing.  A plug flow situation can be approximated by arrang-

ing agitated tanks in series.)  This design concept was orig-

inally piloted at IERL-RTP and further tested at Shawnee.  A

number of full-scale systems that incorporate liquid-side staging

have resulted; all are low-sulfur, limestone-slurry systems.

Sulfur dioxide removals and dependabilities greater than 90

percent have been reported.

     Mist elimination—Chevron and baffle-type mist eliminators

continue to be the only designs used in U.S. utility FGD systems.

Several different designs have been tested (including wire-mesh,

tube-tank, gull-wing, ESP, and radial vane), but the performance

and economics associated with these and other design alternatives

indicate that the exclusive use of chevron and baffle types will

probably prevail.  The popularity of these separators is due

primarily to design simplicity and flexibility, adequate collec-

tion efficiency for medium to large size drops, relatively low

pressure drop, open construction, easy access for maintenance,

and relatively low cost.

     Within these two preferred types of mist eliminators, a

number of specific design, construction, and operation improve-

ments have been implemented,  with the following results:

     1.    Chevron designs of  continuous-vane construction now
          predominate over noncontinuous-vane construction be-
          cause of their greater strength and lower cost.
                                326

-------
2.    Multiple-stage designs predominate over single-stage
     designs on limestone systems.  This tendency is pro-
     cess-sensitive in that limestone systems, which out-
     number lime systems in the United States, generally
     require two or three stages for effective mist entrain-
     ment separation.

3.    Single-stage designs are successful for lime systems
     because of the superior reactivity of lime and the cor-
     respondingly higher utilization.

4.    The number of passes per stage also tends to be pro-
     cess-sensitive.  Four-pass designs are generally used
     for lime and three-pass designs for limestone.  More
     passes are required for lime systems because the
     single-stage design is used, whereas fewer passes are
     required for limestone systems because the multiple-
     stage design is used.

5.    Fiberglass-reinforced plastics, polypropylene, and
     corrugated plastics are now used in almost every oper-
     ational system and specified for use in nearly every
     planned system.  These materials are preferred because
     they are relatively lightweight, inexpensive, and
     superior in resistance in corrosion.  Potential prob-
     lems associated with high temperature excursions have
     been minimized by specifying materials that can with-
     stand exposure up to 400°F.

6.    Vane spacihgs of 1.5 to 3.0 in. are generally used in
     single or first stages and 0.9 to 1.0 in. for second
     stages.  Multiple staging permits the use of finer
     spacings, which provides increased mist-separation
     capability for smaller particles.

7.    The horizontal configuration  (vertical gas flow) is
     still widely used because of its adequate performance
     (to date), its operational and design simplicity, and
     its lower capital cost.  The vertical configuration
     offers a number of advantages over the horizontal
     configuration.  For example, reentrainment due to the
     gas flow opposing the path of the drainage is elimi-
     nated, and limitations on wash water quality and quan-
     tity  (as well as wash direction) are eliminated.  Two
     systems recently started operations using vertical
     configurations.  Initial results indicate adequate
     operation performance and no major operating problems.
                         327

-------
     8.    Special features,  such as hooks and pockets on the
          vanes,  are desirable for prevention of reentrainment.

     9.    Bulk separation devices, impingement plates, single
          baffle deflectors, and gas direction changes are be-
          coming integral parts of mist eliminators because they
          increase removal efficiency and design flexibility.

    10.    Wash and knock-out trays have been incorporated into a
          number of mist eliminators to conserve freshwater and
          increase (or extend) the quantity of water available
          for washing.

    11.    Wash systems that use blended water consisting of pond
          return water or thickener overflow and freshwater are
          used over other strategies (total return or total
          makeup).  Intermittent, high-pressure, high-velocity
          wash systems are preferred to continuous wash systems
          because they have less impact on water balance and
          chemistry.

    12.    Optimum distances between stages are generally 4 to 5
          ft, and freeboard distances are 4 to 5 ft.  The former
          is the minimum distance permitting easy access for
          maintenance.  The latter is the distance at which
          carryover can be minimized without drastically increas-
          ing tower height and pressure drop.

    13.    Superior overall operation is obtained when fly ash is
          collected prior to the scrubbing system.  This is
          because scrubbing systems in which fly ash is not used
          usually have a low slurry solids content.  The lower
          the slurry solids content, the less likely the tendency
          for mist eliminator fouling.

     Reheaters—A pronounced preference for stack gas reheat

versus no reheat  (wet stack) is still evident for those systems

in service and committed for future operation.  The use of wet

stacks at a number of systems has been abandoned in favor of

reheat because problems encountered with corrosion, plume dis-

persion, and plume visibility.  Such problems are more pronounced

where high-sulfur coal is burned because sulfur dioxide loadings

are higher.  A number of developements concerning reheater design

and construction are noted:
                               328

-------
1.   Six methods are available to increase the temperature
     of the gas from a scrubber prior to discharge to the
     stack:  in-line reheat, direct combustion reheat,  in-
     direct hot air reheat, gas bypass reheat, exit gas
     recirculation reheat, and waste heat recovery reheat.

2.   Of the six methods now available only four are applied
     in commercial operations in the United States:  in-line
     reheat, direct combustion reheat, indirect hot air
     reheat, and gas bypass reheat.

3.   Among the systems that have operated or are currently
     in service, in-line reheat has proved to be the most
     popular strategy-

4.   The trend in reheat strategies, as evidenced by FGD
     systems scheduled for immediate and future operation,
     is away from in-line and direct combustion methods and
     toward indirect hot air reheat. This is largely due to
     the problems encountered with in-line reheaters and the
     need for oil or natural gas with direct combustion
     reheaters.  In-line reheat systems have been subject to
     corrosion and plugging in the tubes.  The corrosion in
     many cases has been so severe that even the heartier
     alloys have been unsatisfactory under many operating
     conditions.  Many of these problems have been attri-
     buted to upstream mist eliminator inefficiency and
     inadequate self-cleaning techniques (soot blowers).

5.   A number of the major system suppliers still recommend
     in-line reheaters, especially when minimization of the
     energy demand is desired.  It has been determined  that
     corrosion of high-alloy materials is attributed to
     stress corrosion caused by chloride, whereas carbon
     steel is more susceptible to acid corrosion caused by
     sulfur dioxide.  Therefore, if low sulfur/low chloride,
     low sulfur/high chloride, or high sulfur/low chloride
     environments can be accurately predicted, in such
     applications in-line reheaters may be used success-
     fully.

6.   Indirect hot air reheat has the undesirable effect of
     increasing the energy demand of the FGD system and
     increasing overall system cost.

7.   Bypass reheat may be used for iow-sulfur coal FGD
     applications when the required degree of reheat is not
     seriously constrained by emission standards.
                         329

-------
     8.   The use of efficient mist eliminators reduces the load
          on the reheat system by removing water droplets from
          the flue gas stream.

     9.   AT's of 14° to 28°C (25° to 50°F) adequately prevent
          downstream water condensation.

    10.   Waste heat recovery reheat (regenerative) is now being
          specified for two full-scale systems planned for future
          operation.  In this system, the sensible heat of the
          incoming flue gas is recovered in an in-line heat
          exchanger placed upstream of the air preheater.  This
          heat is then used to reheat the scrubbed gas stream.
          The in-line heat exchanger can be a direct gas-gas heat
          exchanger or a gas-liquid heat exchanger that uses a
          fluid of high heat capacity.   Experience with these
          systems has been reported for experimental, small-
          scale, pilot plant  (1 MW) tests.

     Solids separation (sludge dewatering)—The major development

in this area is the increased emphasis placed on clarification,

centrifugation, and vacuum filtration,  and the corresponding

decreased emphasis on interim ponding.   Formerly, an interim pond

was relied on to fulfill three functions:  clarification, de-

watering, and temporary or final sludge storage.  The realization

that a single pond cannot perform all three functions adequately

encouraged the development of the other techniques.  Furthermore,

the increased emphasis placed on offsite disposal for landfill

and structural fills and on attaining closed water-loop opera-

tions also stimulated the use of clarifiers, centrifuges, and

vacuum filters.  In addition to using these techniques, several

installations are using forced-oxidation strategies to enhance

solids settling and filtration properties, to improve process

chemistry, to improve waste sludge quality, and to decrease land

requirements for sludge disposal.
                                  330

-------
Process Control and Instrumentation—

     Because of the complex nature of lime/limestone scrubbing

chemistry, which has been the primary source of operating prob-

lems in full-scale systems, process control is considered a

crucial item.  The following is a brief review of some of the

essential findings and innovations in the development of process

control technology:

     1.   Virtually all of the operating full-scale systems
          regulate reagent feed rate by controlling slurry pH.
          A pH sensor provides a signal for modulating the flow
          of reagent to the FGD system in a feedback control
          mode.  The pH signal regulates the position of control
          valves for controlling the rate of reagent feed.

     2.   The major problems encountered with pH control systems
          are sensor plugging, calibration drift, breakage, false
          indication, and erosion/corrosion damage.

     3.   Sufficient operating experience has been obtained so
          that most pf the reagent feed control problems have
          been identified.  Once identified, these problems have
          been resolved for the most part through design modifi-
          cations and/or new operating and maintenance proce-
          dures.

     4.   Concerning selection of hardware for pH control, it has
          been noted that dip-type sensors are more successful
          than in-line sensors because they are easier to clean
          and calibrate.  In-line, flow-through sensors are
          generally subject to more wear and abrasion and gen-
          erally require more frequent maintenance.

     5.   Other reagent feed control systems have been or are
          being evaluated on full-scale systems.  One type
          involves feed-forward reagent control on the inlet flue
          gas flow rate and sulfur dioxide concentration, with
          trim provided by slurry pH.  Another type involves
          control of reagent feed rate by using the outlet sulfur
          dioxide as the control variable.  Limited success has
          been reported on both of these systems, primarily
          because of the difficulty in obtaining accurate and
          consistent readings from sulfur dioxide gas analyzers.
          This has been especially difficult on high-sulfur coal
          applications.
                               331

-------
Construction Materials—

     Analysis of FGD experience in the United States to date,

when examined at all levels of development, makes one point

clear:  generalization is difficult because of the many factors

and apparent contradictions in FGD operation.  This is especially

evident in the area of construction materials.  Many examples can

be cited in which seemingly inferior construction materials have

been adequate, whereas apparently adequate materials have failed.

Although construction materials have been discussed throughout

this paper with respect to process and equipment design improve-

ments, a brief review of the trend in the construction of crit-

ical elements in FGD systems is provided as follows:

     1.   The use of 316 and 316L stainless steel is generally
          preferred as the construction material in critical
          areas of the FGD system.

     2.   Some designers avoid 316 and 316L stainless steels
          whenever possible and instead use carbon steel with a
          surface lining or coating that physically shields the
          bare metal from the corrosive environment.  These
          linings are usually resins applied in liquid or semi-
          liquid form, by spray or trowel, and allowed to cure.
          The use of lined or coated carbon steel offers the
          potential benefits of being able to withstand low
          pH/high chloride environments much better than the
          316's and being considerably less expensive.  Actual
          operating experience, however, has shown that these
          materials are very susceptible to high temperature
          excursions, and often require an additional capital
          investment for an auxiliary power system.  Corrosion
          resistance is also a limiting factor for many of the
          materials used.  In addition, application and reappli-
          cation of linings have been suspect, especially reap-
          plications, where proper preparation of the metal
          surface becomes more difficult.
                              332

-------
3.    The use of chemically resistant masonry materials
     (e.g.,  silicone carbide)  and ceramic liners in the
     throat areas of venturi scrubbers (usually applied to
     316 or 316L stainless steel at the converging section
     of the venturi, where the gas velocity and erosive
     nature of the fly ash are highest),  has been quite
     successful.

4.    The use of natural rubber, neoprene, polyvinyl chlorid
     (PVC),  fiber-reinforced plastic (FRP), and flaked-glass
     polyester generally predominates in the liquid and
     thickener circuits of the FGD system (tanks, pumps,
     agitators, piping, and thickeners) where the metal
     parts are 316 stainless steel or Alloy 20.  For filtra-
     tion systems, neoprene, polypropylene, FRP, and Alloy
     20 are generally employed.

5.    Many systems are also being constructed of more re-
     sistant alloys in trouble spots (e.g., wet/dry, high-
     temperature, and high-chloride environments such as a
     prescrubber or presaturator).  Alloys such as Hastelloy
     C-276, Hastelloy G, Alloy 20, Inconel 625, Incoloy 825,
     317 low-carbon stainless steel, 904 low-carbon stain-
     less steel, Jessop JS-700, and E-Brite 26-1 are being
     selected in minimum amounts.

6.    The use of stack lining or coating materials has been
     troublesome, especially where high-sulfur eastern coal
     is burned.  Lining failures have also been reported in
     systems that included an apparently adequate degree of
     reheat.  Recent developments in this area indicate that
     some progress has been made.  A proprietary, spray-on,
     elastomer has been applied on three of the four flues
     at a station burning high-sulfur eastern coal  (without
     reheat) with apparent success.  The use of acid brick
     also appears to be successful in a similar application.
     At installations burning low-sulfur coal, adequate
     linings have not been as great a problem; many instal-
     lations can get by with unprotected carbon steel flues.
                          333

-------
                            SECTION 5



       CAPITAL AND ANNUAL COSTS OF OPERATIONAL FGD SYSTEMS






INTRODUCTION



     The cost of FGD systems is an area of intense interest and



substantial controversy.  As a result, a number of computer



models have been developed in recent years to estimate capital



and annual costs.  In an effort to provide meaningful economic



data on FGD systems, PEDCo Environmental has incorporated



reported economic data into the EPA Utility FGD Survey Report.



This information has appeared as a separate appendix to these



reports since October 1976.  Until May 1978, this cost appendix



consisted entirely of data reported by the utilities, and little



or no interpretation was provided by PEDCo Environmental.  Begin-



ning with the May 1978 report, however, the format and content of



the cost appendix were revised to include adjusted costs for the



operational FGD systems.





APPROACH



     In March 1978, a cost form was forwarded to each utility



having an operational FGD system.  The form contained all the



available cost information for the system or systems at that



particular utility.  Having been notified in advance of the



purpose of the project, each utility was requested to review the
                               334

-------
cost form thoroughly.  A followup visit by the P.EDCo Environ-

mental staff was arranged to assist in data acquisition and to

ensure completeness and reliability of information.  Results of

the cost analysis were then forwarded to each participating

utility for final review and comment.

     Analysis of the cost data centered on adjusting the esti-

mates to a common basis.  The data were analyzed solely to

determine accurate costs of FGD systems, not to critique the

design or reasonableness of the costs reported by any utility.

The primary adjustments were as follows:

     0    All capital costs were adjusted to July 1, 1977,
          dollars using the Chemical Engineering Index.  All
          capital costs, represented in $/kW, were expressed in
          terms of gross megawatts  (MW).

     0    Particulate control costs were deducted.  Since the
          purpose of the study was to estimate the incremental
          cost of sulfur dioxide control, particulate control
          costs were deducted, using either data contained in the
          cost breakdowns or a percentage of the total direct
          cost  (capital and annual).  The percentage reduction
          varied according to system design and operation.

     0    The capital costs associated with the modification or
          installation of equipment not directly involved with
          the FGD system were included  (e.g., stack lining,
          modification to existing ductwork or fans).

     0    Indirect charges were adjusted, where necessary, to
          provide adequate funds for engineering, field expenses,
          overhead, interest during construction, startup, and
          contingency.

     0    All annual costs, given in mills/kWh, were based on
          net MW.

     0    All annual costs were adjusted to a common capacity
          factor  (65 percent).
                               335

-------
     e    Replacement power costs were not included because only a
          few uitlities reported such costs and those reported were
          presented in a variety of methods.

     0    Sludge disposal costs were adjusted to reflect the
          costs of sulfur dioxide scrubber sludge disposal only
          (i.e., excluding fly ash disposal) and to provide for
          disposal over the anticipated lifetime of the FGD
          system.

     0    A 30-year life was assumed for all process and economic
          considerations for all new units.  A 20-year life was
          assumed for all process and economic considerations for
          all retrofit systems, even if the remaining boiler life
          was less than 20 years.

     Cost data were obtained for 27 of the 31 FGD systems in

operation at the time this cost analysis was conducted.   Table 7

provides a summary of the reported and adjusted capital  and

annual costs for all the operational FGD systems,  and Table 8

summarizes these results by category (process)  and application

(new/retrofit).
                               336

-------
    TABLE 7.  REPORTED AND ADJUSTED CAPITAL AND ANNUAL
            COSTS FOR OPERATIONAL FGD SYSTEMS

Choi la 1
Conesville 5
Elrama 1-4
Phillips 1-6
Petersburg 3
Hawthorn 3-4
La Cygne 1
Green River 1-3
Cane Run 4
Cane Run 5
Paddys Run 6
M.R. Young 23
Colstrip 1-2
Reid Gardner 1-2
Reid Gardner 3
D.'H. Mitchell 11
Sherburne 1-2
B. Mansfield 1-2
Eddystone 1A
Winyah 2
a
Southwest 1
Widows Creek 8
Reported
Capital,
$/kW
52.0
55.6
113.5
107.0
99.5
29.3
53.7
70.3
66.6
62.4
52.9
86.0
77.1
42.9
113.6
156.9
47.9
120.7
156.8
47.5
77.3
98.2
Annual,
mills/kWh
2.19
4.71
8.62
7.83

8.40
1.70
14.35
2.75


0.27
2.10
2.10
14.86
1.99
14.35

1.61

2.99
Adjusted
Capital,
$/kW
56.6
70.8
127.2
140.6
100.6
87.3
68.0
77.6
80.6
67.5
76.5
93.1
77.3
60.9
107.9
145.5
71.9
102.9
233.3
66.5
117.7
113.2
Annual ,
mills/kWh
2.58
7.42
7.81
8.57
6.56
4.35
3.78
5.24
5.78
5.56
6.51
5.16
4.06
3.20
4.38
12.73
2.77
8.68

2.92
6.17
5.28
Annual costs were not reported by the utility for this system
because it has not had operating status long enough to provide.

Annual costs were not reported by the utility for this system
because the peak load status of unit precludes its providing
meaningful data.

Annual cost data are being assembled by the utility.
                            337

-------
                   TABLE  8.   CATEGORICAL  RESULTS OF THE REPORTED AND ADJUSTED

                      CAPITAL AND ANNUAL  COSTS FOR OPERATIONAL FGD SYSTEMS
u>
oo

All
New
Retrofit
Noaregenerable
Regenerable
Lines tone
Line
Alkaline fly
ash/limestone
Alkaline fly
ash/lime
Sodium
carbonate
Magnesium
oxide
MelUnan-Lord
Reported
Range, SAW
29.3-156.9
47.5-120.7
29.3-156.9
29.3-120.7
156.8-156.9
47.5-99.5
29.3-120.7
47.9
77.1-86.0
42.9-113.6
156.8
156.9
Capital
average,
SAW
78.0
78.8
77.2
71.7
156.8
71.4
75.3
47.9
81.6
78.3
156.8
156.9

-------
                            SECTION 6

                             SUMMARY


     The preceding sections of this report examined the entire

gamut of FGD technology in the utility sector of the United

States.  As discussed in Sections 3 and 4, considerable progress

has been made in resolving the major problems that have plagued

the initial installations.  More work needs to be done, however,

especially in the area of evaluating trade-offs as a means of

reducing cost and parasitic energy demand (power production

penalty) without impairing dependability and efficiency.  The

report implies a number of research, development/ and demon-

stration (RD&D) needs that should be pursued to increase the

efficiency and cost-effectiveness of FGD operation.  The main

areas for RD&D work are as follows:

     0    A thorough investigation of the phenomenon of scaling;
          The mechanisms that are involved in scaling and the
          border zone between scaling and nonscaling should be
          better defined.

     0    Improvement of mist elimination.  Because this is a
          particularly troublesome area, an intensive investiga-
          tion aimed at optimizing design would be beneficial.

     0    Optimization of stack gas reheat.  A decision regarding
          whether or not reheat is required, and if so, the
          methods that provide the best results  (energy, equip-
          ment protection, plume visibility, and mechanical
          reliability) would be beneficial.
                              339

-------
Improvement of instrumentation and process control
strategies.  This would be very beneficial to cost-
effective and reliable operation.

Optimization of construction materials used in the FQD
system and related equipment.  Recent disastrous re-
sults with .alloys and linings  (especially the latter)
used on the stack accentuate the need for a major
effort in this area.

Closed-water loop operation.  This needs work as far as
ascertaining overall feasibility and determining the
best applicable methods.

Secondary environmental impacts, especially from sludge
resulting from nonregenerable systems, need further
investigation.  More work is needed to reduce land
requirements and increase utilization of the material
in land and structural fills.

Process complexity and economic and energy penalties
associated with the current generation of regenerable
systems.  It would be beneficial if these could be
minimized.

Investigation of dry collection processes.  A substan-
tial amount of investigation has already been conducted
to verify process design using lime and sodium carbonate
reagents for low- to medium-sulfur coal applications.
The feasibility of dry collection for medium- to high-
sulfur coal applications could prove beneficial and
should be investigated.

Investigation of process design configurations in which
the ESP follows rather than precedes the scrubber.
Such configuration may-offer advantages for particulate
collection.  This may prove beneficial in light of
recent concerns over scrubber slurry carryover and mist
loading in the scrubbed gas stream and compliance with
more stringent particulate emission regulations.
                      340

-------
                           REFERENCES
1.    Sixth Biennial Survey of Power Equipment Requirements of the
     U.S.  Electric Utility Industry:  1977-1986.  Sponsored by
     the Power Equipment Division,  National Electrical Manufac-
     turers Association.

2.    Temple, Barker,  and Sloan,  Inc.  Policy Testing Model for
     Electric Utilities, Exhibit II-3.

3.    Twelfth Annual Power Engineering Survey.  Power Engineering,
     April 1978.
                              341

-------
                                                                        HEAD 4B
                      RECENT RESULTS FROM ERA'S
                  LIME/LIMESTONE SCRUBBING PROGRAMS
                  ADIPIC ACID AS A SCRUBBER ADDITIVE-
                 H.N.  Head,  S.C.  Wang,  and D.T.  Rabb
              Bechtel  National,  Inc.,   50 Beale  Street
                  San  Francisco,  California 94105
                                 and
            R.H.  Borgwardt, J.E.  Williams,  and M.A.  Maxwell
             Industrial  Environmental  Research Laboratory
                U.S.  Environmental  Protection Agency
             Research Triangle Park,  North  Carolina  27711
                              ABSTRACT



Adi pic acid has been demonstrated as a powerful  scrubbing additive for
enhancing S02 removal  in lime and limestone wet  scrubbing tests both
at EPA's IERL pilot plant at Research Triangle Park,  North Carolina,
and at the EPA-sponsored Shawnee Test Facility near Paducah, Kentucky.

At adipic acid concentrations in the scrubber liquor in the range of
700 to 1500 ppm, SO? removals in excess of 90 percent have been consis-
tently obtained with limestone slurry at both facilities.  S02 removal
was effectively enhanced even when operating with dissolved chlorides
in the scrubber liquor as high as 10,000 ppm at  the Shawnee Test Facility
and 17,000 ppm at the IERL-RTP pilot plant.  S02 removal was enhanced
equally well  in systems with or without forced oxidation.  Adipic acid
was found to cause only minor differences in the dewatering and handling
properties of oxidized sludge.  No scaling problems were encountered.

Because of decomposition at scrubber operating conditions, consumption
of adipic acid has been greater than anticipated, especially in systems
with forced oxidation.  To maintain 1500 ppm of adipic acid in the slurry
liquor at Shawnee conditions, adipic acid consumption has been in the
range of 8 to 9 Ib/ton of limestone.

A preliminary economic assessment by TVA comparing a base limestone case
with adipic acid enhanced limestone for 90 percent S0£ removal or better
and using the actual adipic acid consumptions experienced at Shawnee
indicates a reduction in annual revenue requirements of about 0.3 to 0.4
mill/kWh exclusive of ponding costs for systems with 750 to 1500 ppm of
adipic acid.
                                 342

-------
                         ACKNOWLEDGEMENTS
The authors wish to express their appreciation to Dr.  Gary Rochelle,
consultant to the project,  who did the initial studies leading  to  the
adipic acid development program and to the following Bechtel  personnel
who contributed to the preparation of this paper:
           D.A.  Burbank, Jr.

           G.A.  Dallabetta

           C.L.  DaMassa

           D.G.  Derasary

           0.  Hing
D.Y. Kawahara

T.M. Martin

R.R. McKinsey

L.S. Reider

C.H. Rowland
Further acknowledgement and appreciation are extended to the  TVA staff
at the Shawnee Test Facility and to TVA's Emission Control  Development
Projects group at Muscle Shoals, Alabama, who are responsible for oper-
ation, maintenance, and engineering modification of the facility and
who prepared the preliminary economic information quoted in this report.
                               NOTE

    Although it is the policy of the EPA to use the metric system
    for quantitative descriptions, the British system is used in
    this report.   Readers who are more accustomed to metric units
    are referred  to the conversion table in the Appendix.
                            343

-------
                       RECENT RESULTS FROM ERA'S
                  LIME/LIMESTONE SCRUBBING PROGRAMS
                 -ADIPIC ACID AS A SCRUBBER ADDITIVE-
                              Section 1

                            INTRODUCTION
A primary objective of the EPA alkali  wet scrubbing test program during the
last year has been to enhance S02 removal  and improve the reliability and
economics of lime and limestone wet scrubbing systems by use of adipic acid
as a chemical additive.  This report addresses the results of testing with
adipic acid at EPA's IERL-RTP pilot plant and at the EPA Alkali Scrubbing
Test Facility, near Paducah,  Kentucky.

Adipic acid is a commercially available organic acid, [HOOC(CH2)4COOH], that
buffers the pH* in S02 absorbers when present at low concentrations in the
scrubbing liquor.  The theoretical  basis for its effect  on the performance
of limestone and lime scrubbers was developed in detail  by Rochellel):  The
buffering action limits the drop in pH that normally occurs at the gas/liquid
interface during S0£ absorption, and the higher concentration of S02  in the
surface film resulting from this buffering action accelerates the liquid-
phase mass transfer.  The capacity of the bulk liquor for reaction with S02
is also increased by the presence of calcium adipate in  solution.  S02 absorp-
tion is therefore less dependent upon the dissolution of limestone in the
absorber.  The overall result of these effects is improved S02 removal in
limestone or lime scrubbers of a given type operating at a given L/G.  In
the case of limestone, it follows that a given S02 removal  efficiency can be
achieved at a lower stoichiometric ratio.

Further analysis by Rochelle^) indicated that the use of additives would be most
attractive economically when used in scrubbers employing forced oxidation.   If
no decomposition or volatilization of the additive occurs, the makeup require-
ments for the additive would be minimized by the tightly closed loop resulting
from the better dewatering properties of oxidized sludge.  For this reason, the
scrubber tests reported here emphasized the evaluation of adipic acid in com-
bination with forced oxidation.

Adipic acid has several potential advantages over other  additives, such as MgO,
which are also known to increase SQg removal, but do so  by means of reactions
involving the sulfite/bisulfite equilibrium. Since adipic acid does not depend
on this mechanism for its buffering activity, it is not  adversely affected by
* Adipic acid has two buffering points.   In the absence of chloride these are
  at pH 5.5 and 4.5.  Chloride concentrations in the range of 5000 to 7000 ppm
  depress these buffering points to about pH 5.0 and 4.0.


                                    344

-------
oxidation of the sulfite in the scrubbing liquor.  It should therefore be es-
pecially useful in single loop scrubbers that employ forced oxidation.  Further-
more, the buffering mechanism by which adipic acid enhances SOg absorption is
not affected by the presence of chloride.  The lack of interference by chloride
means that adipic acid should be fully effective in the most tightly closed loop
systems.

Preliminary economic evaluations have shown that adipic acid can reduce the
operating cost of limestone systems while simultaneously increasing performance.
Adipic acid is a major ingredient in the manufacture of nylon and is sufficiently
available that widespread use in commercial FGD systems would have little effect
on the market.

Beginning in 1977, initial  studies conducted by EPA with the 0.1 MW pilot plant
at the Industrial Environmental Research Laboratory located at Research Triangle
Park, North Carolina (IERL-RTP) demonstrated as predicted by Rochelle that adipic
acid was indeed an attractive additive.  Results of these tests were first reported
at the EPA Industrial Briefing at Research Triangle Park in August 1978.3)   /^n
update of recent findings at the pilot plant is given in Section 2.

Based on the findings at the IERL-RTP pilot plant, a program was set up at the
EPA sponsored Test Facility located at the TVA Shawnee Steam Plant near Paducah,
Kentucky, to develop commercially usable design data for adipic acid as a chem-
ical additive.  Testing with adipic acid was initiated in July 1978 on the 10 MW
EPA prototype scrubbers and has continued since as a major part of the Shawnee
Advanced Test Program.  The results of testing with adipic acid additive at the
Shawnee Test Facility from July 1978 through January 1979 are presented in Sections
3 and 4.

Configurations successfully demonstrated during this period were:

     t  Adipic acid enhanced limestone scrubbing in a venturi/spray tower
        system with two scrubbing loops and forced oxidation in the first
        loop.

     •  Adipic acid enhanced limestone scrubbing in a Turbulent Contact Absorber
        (TCA) system with no forced oxidation.

In addition, preliminary tests have been conducted in the same configurations
using adipic acid enhanced lime slurry.


TEST FACILITY AND PROGRAM

There are two scrubber systems operating at the EPA sponsored Shawnee Test
Facility, each with its own independent slurry handling facilities.  Both systems
were tested with adipic acid additive.  The systems have the following scrubbers:

     •  A venturi followed by a spray tower (V/ST)
        (35,000 acfm capacity @ 300°F)

     •  A Turbulent Contact Absorber (TCA)
        (30,000 acfm capacity @ 300°F)


                                    345

-------
The scrubbers receive flue gas from TVA Shawnee coal-fired boiler No. 10.  The
boiler normally burns a high-to-medium sulfur bituminous coal producing S02
concentrations of 1500 to 4500 ppm.  Flue gas can be taken from either upstream
or downstream of the boiler No. 10 particulate removal equipment, allowing testing
with high fly ash loadings (3 to 6 grains/scf dry) or low loadings (0.04 to 0.2
grains/ scf dry).  All tests in the adipic acid series were made with high fly
ash loadings.

The test program was conducted with the scrubbing loop fully closed.   Chlorides
from the flue gas were concentrated in the scrubber slurry liquor over a range
of 1000 to 10,000 ppm depending on the tightness of the scrubber water balance
and the chloride concentration in the coal burned.

The Shawnee Test Facility has been operating since March 1972.   Bechtel  National,
Inc. of San Francisco is the major contractor and test director; TVA  is  the con-
structor and test facility operator.  The initial  test program  lasted through
October 19744) with the major emphasis on demonstrating reliable operation. The
forced-oxidation tests are a part of an advanced test program that is currently
scheduled to continue through December 1979.  Earlier results of the  advanced
test program and a description of the test facility are reported elsewhere.^'"'''
8,9,10)

The Advanced Test Program schedule for the period covered in this report is
shown in Figure 1.  As, can be seen, testing with adipic acid additive has  con-
stituted the major effort during this period.
                                   346

-------
CO
-o

ITEM
1. VENTURI /SPRAY TOWER SYSTEM
(TWO SCRUBBER LOOPS WITH FORCED OXIDATION)
LIME BASE CASE
LIMESTONE BASE CASE
LIME/ADIPIC ACID
LIMESTONE/ADI PIC ACID
LIMESTONE/ADI PIC ACID-VENTURI ONLY
LIMESTONE/ADI PIC ACID RELIABILITY


2. TCA SYSTEM
(SINGLE SCRUBBER LOOP WITHOUT FORCED OXIDATION)
LIMESTONE BASE CASE
LIME BASE CASE
LIME/ADIPIC ACID
LIMESTONE/ADI PIC ACID
LIMESTONE/ADI PIC ACID RELIABILITY
LIME RELIABILITY/FLUE GAS MONITORING



JULY

••
«•
mm






^™
^^
•






AUG




^^^~







•
_mmmm




19
SEPT




^^^™








mmmmm^
m



78
OCT














^mmmm




NOV















^^"



DEC















mmmmm


1979
JAN















^^^


                                  Figure  1.  SHAWNEE ADVANCED  PROGRAM  TEST  SCHEDULE  FOR
                                             PERIOD JULY  1978  THROUGH  JANUARY  1979

-------
                              Section 2

            EVALUATION OF ADIPIC ACID ENHANCED SCRUBBING
                     AT THE IERL-RTP PILOT PLANT
The initial testing of adipic acid as a scrubber additive was carried out in
the EPA pilot plant at Research Triangle Park.  A single-loop limestone scrub-
bepll) was used for this purpose, operated with forced oxidation in the scrub-
bing loop.  This configuration would be expected to provide a sensitive response
to any effect of adipic acid on oxidizer performance, because it operated at a
higher pH than the two-loop system at Shawnee.  In addition to effects on S0"2
removal and oxidation efficiencies, these tests also sought to determine whether
adipic acid caused any change in the properties of the oxidized sludge.  So that
these properties could be clearly seen, the system was operated without fly ash.
Chloride was added as HC1 and controlled at the high levels expected for tightly
closed-loop systems.

The results of the tests showed adipic acid to be very effective in improving
SOg removal efficiency, even when operating at chloride levels as high as
17,000 ppm.  A TCA scrubber, which removed 82 percent of the inlet S02 without
the additive, yielded 89 percent S02 removal with 700 ppm adipic acid, 91 per-
cent removal with 1000 ppm, and 93 percent removal with 2000 ppm adipic acid.
The limestone utilization was concurrently increased from 77 percent without
the additive, to 91 percent with 1600 ppm adipic acid.  The observed effects
thus confirmed the theoretical expectations in all respects.  In addition,
the tests showed no serious interference by adipic acid on the performance of
the oxidizer, operating at pH 6.1.

The quality of the oxidized sludge was similar to that obtained when operating
without adipic acid, although small differences were detected.  For example,
the filtered sludge averaged 80 percent solids (for 13 one-week tests) vs. 84
percent solids for 11 tests without the additive, when operating at 97 to 99
percent oxidation in both cases.  The settling rate of the slurry (fly ash free
at 50°C) averaged 2.3 cm/min during the adipic acid tests and 3.4 cm/rain without
adipic acid; bulk settled densities averaged 1.0 and 1.2 gm solids/cm^ slurry,
respectively.  It was concluded from these results that the large improvements
in sludge quality that can be achieved by forced oxidation are unaffected by the
use of adipic acid as a scrubber additive.

Tests without forced oxidation also demonstrated the efficacy of adipic acid.
Operating a TCA scrubber with 2000 ppm adipic acid and 6 inches H20 pressure
                                    348

-------
drop,  92  percent S02 removal  was obtained at a limestone utilization level  of
88 percent.   By comparison,  only 75 percent S02 removal  would be expected in
the pilot  plant at these test conditions  without the additive.  At this adipic
acid level the  unoxidized sludge filtered to 49 percent  solids;  at lower adipic
acid levels  (1500 ppm or less)  the filterability of the  slurry was the same as
that obtained without additives:  55 percent solids.

During all tests with adipic  acid, the  scrubbing liquor  had a noticeable odor
even though  the additive feed did not.   The odor has been identified as valeric
acid [CH3(CH2)3COOH], which  is  an intermediate product formed by side reactions
that degrade adipic acid at  scrubber operating conditions.   Tests conducted by
Radian Corporation12) at IERL-RTP showed  40-50 ppm valeric  acid  in the scrubbing
liquor, and  about 1 ppm in the  effluent flue gas when operating  without forced
oxidation  and a 2000 ppm adipic acid level.  Although laboratory tests show that
adipic acid  ultimately degrades to lower  molecular weight (C} to 04) paraffinic
hydrocarbons, no degradation  products other than valeric acid have been detected
(detection threshold = 10 ppb)  in the pilot plant effluents.   Efforts by Radian
to identify  the chemical  mechanism responsible for degradation are continuing.

Material  balances were carried  out with and without forced  oxidation to compare
the adipic acid makeup requirements. The results showed that the degradation
is greater when operating with  forced oxidation, in which case 64 percent of the
feed was  unaccounted for. Without forced oxidation, the estimated loss was 24
percent.   In spite of the greater rate  of degradation with  forced oxidation, the
pilot plant  material  balance  indicated  that the adipic acid makeup should still
be minimal for  this mode: the reduction in liquor loss resulting from improved
slurry dewatering properties  more than  compensates for the  additional  degradation.
                                   349

-------
                                Section 3

              ADIPIC ACID ENHANCED SCRUBBING IN THE SHAWNEE
    VENTURI/SPRAY TOWER SYSTEM WITH TWO SCRUBBER LOOPS AND FORCED OXIDATION
Since January 1977, the venturi/spray tower system has operated with two
scrubber loops in series and with forced oxidation accomplished by sparging
air into the first scrubber loop (venturi) hold tank.   Successful  demonstra-
tion of this mode of scrubbing has already been reported with three alkali
types:  limestone, lime, and limestone with added magnesium oxide.9'10)

Since July 1978, tests have been conducted with added  adipic acid  in both
lime and limestone systems.  With adipic acid,  removals as high as 98 percent
have been achieved with both alkalis while concurrently oxidizing  the product
to gypsum.  Results of these adipic acid enhanced, forced-oxidation tests  are
reported in this section.


SYSTEM DESCRIPTION

The venturi/spray tower system was modified for two-loop scrubber  operation
with forced oxidation as shown in Figure 2.  To separate the venturi and spray
tower scrubber loops, a catch funnel was installed beneath the bottom spray
header of the spray tower.   To minimize slurry  entrainment through the catch
funnel, the bottom spray header was turned upward.

The hold tank in the first  scrubber (venturi) recirculation loop was used  as
the oxidation tank.  The arrangement of this tank is shown in Figure 3.    The
tank was 8 ft in diameter and could be operated at 10, 14, or 18-ft slurry
levels.  In early tests the tank contained an air sparger ring made of
straight 3-inch 316L SS pipe pieces welded into an octagon approximately 4 ft
in diameter.  It was located 6 inches from the  bottom  of the tank.  Sparger
rings had either 130 1/8-inch diameter holes or 40 1/4-inch diameter holes
pointing downward.  The sparger ring was fed with compressed air to which
sufficient water was added  to assure humidification.  (Dry air can  evaporate
water at the sparger orifice and cause scaling).   In more recent tests the
sparger ring was replaced by a 3-inch diameter  pipe with an open elbow
discharging air downward at the center of the tank about 3 inches  from the
tank bottom.  In all tests  with adipic acid enhancement, the 3-inch pipe was
used for air discharge.

The oxidation tank had an agitator with two axial flow turbines, both pumping
downward.  Each turbine was 52 inches in diameter and  contained 4  blades.  The
bottom turbine was 10 inches above the air sparger. The agitator  rotated  at
56 rpm and was rated at 17  brake Hp.
                                    350

-------
CO
                      REHEAT
                                    FLUE GAS
                                                                              COMPRESSED
                                                                           VENT   AIR
                                                                            I
                                                                                                                 MAKEUP WATER
                                                                                     CLARIFIED LIQUOR FROM SOLIDS DEWATERING SYSTEM
                                                                                                                  BLEED TO
                                                                                                                  SOLIDS
                                                                                                                DEWATERING
                                                                                                                  SYSTEM
                              Figure 2.   FLOW DIAGRAM FOR ADIPIC ACID ENHANCED SCRUBBING
                                          IN THE  VENTURI/SPRAY  TOWER SYSTEM WITH TWO
                                          SCRUBBER  LOOPS AND  FORCED OXIDATION

-------
                                               BAFFLE
                                                 COMPRESSED AIR
            AGITATOR
                         OXIDATION TANK
                           PLAN VIEW
              COVER
             OUTLET
              18FT
              14FT   r

         ALTERNATIVE
            OUTLETS
                     L

         ~* "lOFT   r
 3 IN. DIAMETER PIPE
 WITH AIR DISCHARGE
 DOWNWARD


01  2345
  SCALE, FEET
                              L,

-------
A 10-ft diameter desupersaturation tank, operating at a 5-ft slurry level,
followed the oxidation tank to provide time for gypsum precipitation and to
provide air-free pump suction.

Provision was made to add alkali to either loop.  Adi pic acid was added as  a
dry powder to the spray tower effluent hold tank.  The dewatering system con-
sisted of a clarifier followed by a rotary drum vacuum filter.  Clarified liquor
from the dewatering system can be returned to either scrubber loop or to the
mist eliminator wash circuit.
SUMMARY OF PREVIOUSLY REPORTED TWO-LOOP FORCED-OXIDATION TEST RESULTS
Forced-oxidation test results with two scrubber loops conducted from January
1977 thrc
reported.

Key results of these earlier tests were:
1977 through June 1978 with  lime  and  limestone  slurry have  been  previously
reported.9>10)
     •  Oxidation of sulfite solids to gypsum of 90 percent or better dramati-
        cally improved the dewatering and handling characteristics of the
        waste solids.

     •  Slurry oxidation of better than 96 percent in the first of two indepen-
        dent scrubbing loops was demonstrated with simple air sparging through
        a 3-inch pipe into an agitated tank with the configuration shown in
        Figure 3.

     •  Conditions under which near complete oxidation was demonstrated were an
        oxidation tank pH range of 4.5 to 5.5, an air stoichiometric ratio of
        at least 1.5 atoms oxygen/mole of S02 absorbed, and an oxidation tank
        level of at least 14 feet.

     •  Slurries with high or low fly ash loadings oxidized equally well.

     •  A slurry solids concentration of 7 percent or higher in the spray tower
        was required to prevent calcium sulfite scaling and to maintain good
        S£>2 removal.

     •  For pH control in both scrubber loops, it was necessary to add lime to
        both loops.  With limestone, addition to the spray tower loop was
        sufficient.

     •  In a 32 day run with lime slurry, an average S02 removal of 88 percent
        (2950 ppm average inlet S02 concentration) was achieved.  Lime utilization
        was 98 percent.

     t  In a 35 day run with limestone slurry, an average S0£ removal of 86 per-
        cent (2950 ppm average inlet S02 concentration) was achieved.  Limestone
        utilization was 81 percent.

                                    353

-------
     •  In a 20 day limestone run with 5000 ppm effective Mg++ concentration*
        in the spray tower scrubber loop, an average SOg removal of 96 percent
        (2250 ppm average inlet S02 concentration) was achieved.  Limestone
        utilization was 92 percent.
TWO-LOOP FORCED-OXIDATION TEST RESULTS USING LIMESTONE SLURRY WITH ADDED
ADIPIC ACID


Beginning in July 1978, a series of 9 limestone tests with forced oxidation on
the two-loop venturi/spray tower system were conducted to demonstrate adipic
acid as an additive to enhance S02 removal.  Results of these tests are sum-
marized in Table 1.  Unless otherwise specified, controlled run conditions
common to all of the tests were:

            Fly ash loading:  High (3 to 6 grains/scf dry)
            Flue gas rate:  35,000 acfm at 300°F
            Venturi slurry rate:  600 gpm (21 gal/Mcf)
            Spray tower slurry rate:   1600 gpm (57 gal/Mcf)
            Venturi slurry solids concentration:  15 percent
            Adipic acid in spray tower liquor:  1500 ppm


After operating variables were explored in runs lasting about a week each, S02
removals above 95 percent (2500 ppm inlet S02 concentration) were achieved in
long term tests at limestone utilizations of about 90 percent and with near
complete oxidation.


Effect of Adipic Acid Concentration - The effect of adipic acid on enhancing S02
removal can be seen by comparing Run  901-1A (no adipic acid) with Run 902-1A
(nominally 1500 ppm adipic acid in the spray tower slurry).  Both runs were pur-
posely made at high limestone utilization (97 and 94 percent in Runs 901-1A and
902-1A, respectively) to demonstrate  the effect of adipic acid.  The addition
of adipic acid increased S02 removal  from 57 percent with no acid to 91 percent
at nominally 1500 ppm.  These runs were made before conditions for operating
with adipic acid were optimized.  After optimization, S02 removals in excess of
95 percent were routinely achieved at 1500 ppm adipic acid.

Actual adipic acid concentrations were 1445 ppm in the spray tower loop and 2045
ppm in the venturi loop.  Dissolved solids are concentrated in the venturi loop be-
cause water is evaporated in humidifying the hot flue gas entering the scrubber.

Results at the IERL-RTP pilot plant indicated that most of the enhancement effect
with adipic acid may be achieved at concentrations as low as 700 ppm in the slurry
liquor.  Adipic acid concentrations lower than the 1500 ppm already tested will be
explored at the Shawnee facility as time permits.
* Effective Mg   concentration is defined as the total magnesium ion concentration
  minus that magnesium ion equivalent to total chlorides.  Magnesium chloride has
  no effect on S02 removal.
                                    354

-------
                                                                                                Table  1

                                                         VENTURI/SPRAY  TOWER  TWO-LOOP  TESTS  WITH  FORCED OXIDATION
                                                                     -  ADIPIC  ACID  ENHANCED  LIMESTONE  SLURRY -
Cn
Major Test Conditions'"
Fly ash loading
Adipic acid concentration in venturi , ppm
Adipic acid concentration in ST (controlled), ppin
Gas rate, acfm 9 300°F
Venturi slurry rate, gpm
ST slurry rate, gpm
Venturi solids recirculated (controlled), wt.%
Residence times (mini/tank level (ft): Spray tower EHT
Oxidation tank
Desupersaturation tank
Venturi inlet pH (estimate, not controlled)
Venturi limestone stoichiometric ratio (controlled)
ST limestone stoichiometric ratio (controlled)
Air rate to oxidizer, scfm
Run Average Results
Start-of-Run date
Onstream hours
S02 removal, %
Inlet S00 concentration, ppm
c
Spray tower solids recirculated, wt.X
Venturi inlet pH
Spray tower inlet pH
Spray tower limestone stoichiometric ratio
Spray tower inlet liquor gypsum saturation, %
Spray tower sulfite oxidation, %
Overall Sulfite oxidation. %
Overall limestone utilization, %
Venturi inlet liquor gypsum saturation, %
Venturi inlet liquor SO.." concentration, ppm
Adipic acid concentration in venturi, ppm
Adipic acid concentration in ST, ppm
Air stoichiometry, atoms 0/mole S02 absorbed
Kilter cake solids, wti
Mist eliminator restriction, %
901-1A
High
0
0
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
(-4.5)
-
1.4
210

7/12/78
1B7
57
2800
5.9
4.50
5.45
1.36
95
28
98
97
95
80
0
0
2.30
85
0.2
902- 1A
High
(~2000)
1500
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
(-4.5)'
-
1.4
260

8/3/78
373
91
2450
4.9
4.45
5.25
1.65
110
40
98
94
100
55
2045
1445
2.05
86
0.2
902-1B
High
(~2000)
1500
35,000
600
1600<2>
15
14.7/10
11.3/18
4.7/4.8
(-4.5)
-
1.4
SCO

8/22/78
65
89
2400
4.8
4.00
5.05
1.38
115
27
99
98
90
45
2355
1615
2.15
88

903-1 A
High
3500
_
35,000
600
0(3)
15
-
11.3/18
4.7/4.8
t-5.5)
1.3

210

8/28/78
183
53
2600

4.7
-
-
-
-
98
74
100_
65
3510
-
2.65
83

904-1A
High
3500
-
18,000
600
0(3)
15
-
11.3/18
4.7/4.8
(v5.5)
1.3
-
130

9/5/78
72
67
2350
-
4.75
-
-
-
-
98
76
100
60
3690
-
2.80
80
0
905- 1A
High
(~2000)
1500<4>
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
4.5<5>
-
1.45
260

9/8/78
439
86
2200
3.2
4.55
4.75
1.53
130
63
99
91
100
60
2315
1410
2.40
85
0
906- 1A
High
(~2000)
1500
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
4.5<5>
-
1.70
260

9/27/78
153
93
2700
6.8<6>
4.5
5.15
1.49
100
28
98.5
88
110
70
2495
1485
1.80
84
0
907-1A*7'
High
t-2500)
1500
18,000-35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
4.5(5>
-
1.70
260

10/8/78
719
97.5
2350
6.l(6)
4.65
5.45
1.77
105
29
98.5
88
110
65
2360
1560
2.0-3.85
87
0
907-1B(7)
High
(-2500)
1500
20,000-35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
4.5<5>
-
1.70
260

11/13/78
1666
97
2500
5.9<6>
4.65
5.35
1.70
110
25
98
92
105
75
2180
1510
1.9-3.3
85
-
                        Notes:  (1)  All runs were made with 9 in. 1*20 venturi pressure drop except Runs 907-1A and 907-1B in which pressure drop was 9 in. HgO at 35,000 acfm.
                               (2)  Intermittent spray tower operation.  Spray tower slurry flow turned off 30 minutes every 8 hours.  S02 removal averaged •*-30 percent by
                                   venturi  only operation.
                               (3)  Venturi  alone operation.  The spray tower slurry recirculation pumps were turned off for the entire run.
                               (4)  During the initial portion of the run (unsteady state operation) adipic acid was allowed to deplete to observe S02 removal/adiplc acid
                                   relationship.
                               (5)  Venturi  inlet pH was controlled by separate limestone addition.
                               (6)  Clarified liquor in excess of mist eliminator wash was returned to spray tower EHT. except for Runs 906-1A, 907-1A, and 907-1B where a
                                   fraction of this stream was diverted to the oxidation tank to control spray tower slurry solids at 6%.
                               (7)  Long-term reliability  test.

-------
S02 Removal In the Venturl - Three runs were made to determine S02 removal in
the venturi alone.In Run 902-1B the spray tower recirculation pumps were turned
off for 30 minutes every 8 hours.  In Runs 903-1A and 904-1A, only the venturi
was operated.  Results were:
     V/ST
     Run
Flue Gas
Rate, acfm
Adi pic Acid in
Venturi, ppm
Venturi
Inlet pH
% S02 Removal
in Venturi
     902-1B       35,000

     903-1A       35,000

     904-1A       18,000
                    2355

                    3510

                    3690
                     4.0

                     4.7

                     4.8
                 30

                 53

                 67
In Run 902-1B, S02 removal was no greater than usually achieved without adipic
acid.  This is not unexpected because the operating pH was below the 4.0 to 5.0
buffering range of adipic acid.

In Run 903-1A, the pH was increased to 4.7 (limestone stoichiometric ratio of
1.35 mole Ca  /mole SO? removed) and adipic acid concentration was increased to
nominally 3500 ppm with a resultant increase in S02 removal  to 53 percent.  Finally
in Run 904-1A, the flue gas flow rate was cut in half to double the liquid-to-gas
ratio, resulting in a further increase in S02 removal to 67  percent.  These tests
showed that high removal efficiencies cannot be achieved with reasonably low
limestone stoichiometry in the venturi alone. Although not yet demonstrated, it
may be possible to increase pH and get high removal efficiency using lime with
adipic acid in the venturi alone.


System Control - In previous limestone runs without adipic acid, the limestone
stoichiometry in the spray tower loop was controlled by adding limestone to the
spray tower hold tank.  By using a spray tower limestone stoichiometric ratio
of about 1.4, there was sufficient residual limestone in the bleed from the spray
tower loop to the venturi loop to maintain the venturi inlet slurry liquor pH
above 5.0.  This was not the case with adipic acid addition.  With adipic acid,
the venturi inlet pH fluctuated between 4.0 and 4.5 which was too low to get good
removal.

Beginning with Run 905-1A, limestone was added to both the spray tower and.
venturi loops in an attempt to raise the venturi inlet pH.  The pH could not be
raised above 4.9 even with a venturi limestone stoichiometry greater than 2.0.
For reasons not yet clear, adipic acid has a depressing effect on the pH in the
venturi loop.

In subsequent runs, the spray tower limestone stoichiometric ratio was increased
to 1.7 to maintain a venturi inlet pH of 4.5.  This pH was established as the
highest pH in which an overall limestone utilization of about 90 percent could
be achieved.  Occasionally, limestone is still added to the venturi loop if the
venturi inlet pH drops much below 4.5.
                                    356

-------
 Effect of Slurry Solids Concentration - In unenhanced limestone systems, slurry
 liquor pH and, consequently, S02 removal are sensitive to slurry solids concen-
 tration  (i.e., limestone surface area available for dissolution) below 8 percent.
 Because  of increased liquor buffer capacity with adipic acid enhancement, the
 system should be less sensitive to solids concentration.  However, in early runs
 with adipic acid, the slurry solids concentration was allowed to drop to sensi-
 tive levels.  In these runs, clarified liquor from the thickener was returned
 to the spray tower loop which resulted in low solids concentrations in the
 tower slurry in the range of 3 to 5 percent.
                                                             spray
 Beginning with Run 906-1A, the spray tower solids concentration was controlled
 at nominally 6 percent by splitting the clarified liquor return between the spray
 tower and the venturi loops.  The spray tower pH and, consequently, SC>2 removal
 were increased as can be seen by the following comparison:


 V/ST      Spray Tower  Spray Tower  Spray Tower        Inlet SC>2    Percent S02
 Run       % Solids     Inlet pH     Limestone Stoich.  Cone, ppm    Removal
 905-1A

 906-1A
3.2

6.8
4.75

5.15
1.5

1.5
2200

2700
86

93
 Based on an examination of day-to-day operations, with 1500 ppm adipic acid in
 limestone slurry, it appears that S02 removal increases with spray tower slurry
 solids concentration up to about 6 weight percent.
 Forced Oxidation - Forced oxi
 equally well with or without
 liquor at a low pH favorable
 on the two-loop venturi/spray
 ratios in the range of 1.8 to
 oxidation was achieved in all
 the oxidized slurry was consi
               dation of the scrubber slurry has  been  shown to  occur
               adipic acid.   In fact, adipic acid buffers  the slurry
               for forced oxidation.   The adipic  acid  enhanced  runs
                tower system were all operated at air  stoichiometric
                2.4 atoms oxygen/mole S02 absorbed.   Near  complete
                runs and the filter cake solids concentration of
               stently high, in the range of 85 percent  or better.
 Venturi/Spray Tower Demonstration Runs 9Q7-1A and 907-1B - Run 907-1A was a month
.long adipic acid enhanced limestone run with forced oxidation designed to demon-
 strate operational reliability with respect to scaling and plugging and to demon-
 strate the removal enhancement capability of the adipic acid additive.

 This run was controlled at a nominal 1.7 limestone stoichiometric ratio (up from
 1.4 in previous runs) and 1500 ppm adipic acid in the spray tower.  Spray tower
 slurry solids concentration was controlled at 6 percent by splitting the clari-
 fied liquor return between the spray tower and venturi loops. Venturi inlet
 slurry liquor pH was nominally 4.5.  Occasionally, limestone addition to the
 venturi loop was required to maintain this pH.
                                    357

-------
Flue gas flow rate was varied from 18,000 acfm to a maximum of 35,000 acfm (spray
tower gas velocity between 4.8 and 9.4 ft/sec) to follow the daily boiler load
cycle which normally fluctuated between 100 and 150 MW.  The adjustable venturi
plug was fixed in a position such that the pressure drop across the venturi was 9
inches H20 at 35,000 acfm maximum gas rate. - Actual pressure drop ranged from
3.0 to 9.6 inches
The slurry recirculation rates to the venturi and spray tower were fixed at 600
gpm (L/G = 21 to 42 gal/Mcf) and 1600 gpm (L/G = 57 to 111 gal/Mcf), respectively.

The oxidation tank level was 18 ft and the air flow rate was held constant at
260 scfm.

The run began on October 8, 1978 and terminated November 13, 1978.  It ran for
719 on-stream hours (30 days) with no unscheduled outages.  The scrubber was
down once for a scheduled 3-hour inspection and again when the boiler came
down for 135 hours to install a new station power transformer.
Average S02 removal for the run was 97.5 percent- at 2350 ppm average inlet
concentration.  The S02 removal stayed within a narrow range of 96 to 99 percent
throughout almost the entire run.  On October 19 and on October 27, SO? removal
dropped briefly to less than 90 percent when the pump which supplies the slurry
to the top two spray headers was brought off stream for repacking and the spray
tower slurry flow rate was cut in half to 800 gpm.  .At the reduced slurry recitf-
culation rate, S02 removal was 82 to 87 percent.

Venturi and spray tower inlet pH averaged 4.65 and 5.45, respectively.  Overall
limestone utilization was 88 percent and the spray tower limestone utilization
was 56 percent, demonstrating the advantage of good limestone utilization in a
two-scrubber-loop operation.


Average adipic acid concentrations were 2360 ppm in the venturi loop and 1560 ppm
in the spray tower loop.

Sulfite oxidation in the system bleed slurry a.veraged 98.5 percent with the air
stoichiometric ratio varied between 2.0 and 3.85 atoms oxygen/mole SOo absorbed.
The filter cake solids content was 87 percent.


The mist eliminator was clean during the entire run.  Inspections after 207 hours,
603 hours, and 719 hours (at the end of the run) showed that the mist eliminator
was completely free of solids deposits.  The system was free of plugging and
scaling.  There was no increase in solids or scale deposits on the scrubber inter-
nals during Run 907-1A.


Following Run 907-1A, a second adipic acid enhanced limestone run with forced
oxidation was made during which flue gas monitoring procedures were evaluated
by EPA.  This run, 907-1B, was made under the same conditions as Run 907-1A
except that a "typical" daily boiler load cycle was established for the gas flow-
rate to follow rather than the Unit No. 10 Boiler load.  The gas rate was changed
as follows to simulate a "typical" boiler load cycle:


                                    358

-------
                Time, hours          Gas Rate, acfm @ 300°F


                  0100                       20,000

                  0500                       30,000

                  0700                       35,000

                  1100                       30,000

                  1700                       35,000

                  2300                       30,000


Run 907-1B began on November 13, 1978 and terminated January 29, 1979.  It
ran for 1666 on-stream hours (69 days) with only 27 hours of scrubber related
outages.  The scrubber was also out of service 146 hours when Unit 10 came
down for replacement of a broken turbine thrust bearing.  Scrubber related
outages were:


                  Plugged slurry line             11 hours

                  Rebuild bleed pump              10 hours

                  Miscellaneous mechanical          4 hours

                  Leak in water supply line        2 hours

                                     Total         27 hours
Excluding boiler outages and scheduled inspections, the combined Runs 907-1A
and 907-1B operated for a period of over 3 months with an on-stream factor of
98.9 percent.  No deposits whatsoever were observed in the mist eliminator for
the entire 3 month test period.  On only one occasion did solids accumulation
cause an outage.  The cross-over line carrying slurry effluent from the venturi
to the oxidation tank plugged with soft solids and had to be cleaned out.  Be-
cause of problems associated with converting the Shawnee venturi/spray tower
system to two-scrubber-loop operation, this cross-over line follows a tortuous
path (see Figure 2).  A properly designed new system would not have this problem.


Results of Run 907-1B were equally as good in every respect as those of Run
907-1A.  Average SOg removal was 97 percent at 2500 ppm average inlet SOp.
With few exceptions, S02 removal remained within a narrow band of 95 to 99
percent.  SO? removal dropped briefly (typically 30 minutes) below 90 percent
five times when one of the two spray tower recirculation pumps was taken out of
service for maintenance, effectively cutting the slurry recirculation rate in half.
                                   359

-------
Overall  limestone utilization during this run was 92 percent.  Sulfite oxidation
averaged 98 percent and the waste sludge filter cake quality was excellent,
having a solids content of 85 percent.

SC>2 emissions for Run 907-1A and 907-1B were calculated based on an assumed coal
heating value of 10,500 Btu/lb, on 100 percent sulfur overhead (none in bottom
ash), and on an assumed excess air of 30 percent. This excess air corresponds to
about 700 ppm inlet S02 per 1.0 weight percent sulfur in coal for the above con-
ditions.

Figures 4 and 5 show calculated S02 emissions for Runs 907-1A and 907-1B, re-
spectively.  Average S02 emissions for each 24-hour period (midnight-to-midnight)
are indicated by horizontal lines on the figures.  The average S02 emission for
the entire 3-month operating period was only 0.20 Ib/MM Btu.  The highest 24-hour
average S02 emission during Run 907-1A was 0.37 Ib/MM Btu (October 18) and during
Run 907-1B was 0.41 Ib/MM Btu (January 20).  These values compare with the federal
new source performance standard of 1.2 Ib/MM Btu.


Adi pic Acid Consumption - From the onset of testing, adipic acid consumption
was higher than anticipated.  A material balance calculation for the adipic
acid consumption was made for the entire Run 907-1B for a period of 76 days from
November 13, 1978 to January 29, 1979.  During this period, a breakdown of the
average adipic acid addition rate was:


                  Run 907-1B Adipic Acid Consumption

                                      Ib/hr           Ib/ton limestone feed

Discharged with filter cake            0.8                   1.8

Unaccounted                          	2._9_                  6.5

Total                                  3.7                   8.3


These losses were higher than experienced in the IERL-RTP pilot plant.  The
reasons for the differences are not yet clear.

The unaccounted adipic acid loss in this run with forced oxidation was somewhat
higher than that recorded in TCA Run 932-2A without forced oxidation (2.2 Ib/hr,
see Section 4).  Based on this comparison, it appears that forced oxidation in
the scrubber system may increase the unaccounted adipic acid losses.

As already mentioned in Section 2, Radian Corporation is investigating the mech-
anism of this unaccounted loss.  Preliminary results indicate that adipic acid
decomposes to valeric acid [CH3(CH2)3COOH] and other components.  Valeric acid
creates an odor even in the small concentrations present in the scrubber slurry.
An unpleasant odor was apparent immediately above the effluent hold tanks and
in the filter and centrifuge building where dewatering takes place.  About 600
cubic yards of the gypsum/fly ash mixture from the 3-month combined venturi/
                                    360

-------
I
                           120     IN
                                             muri—. .
                                             10/10 I 10/10 I
I 10/0 I 10/10 I 10/11 I 10/12 I 10/13 I 10/14 I 10/11 I 10/10 I 10/17 I 10/10 I 10/10 I 10/10 I 10/11 I 10/22 I 10/O I 10/24 I 10/21 I 10/M 1 10/27 I 10/20
                                 CALINDA* DAY (10701
        HUM 007 • 1« COMTIHUIO
                                                             END HUH 007- 1A|
         9(20M00000400f07207niOOMO*IOnOMO


          I 10AO I 10/30 I 10/11 I 11/1 I 11/1 I 11/3 I 11/4 I 111* I 11A I 11/7 T 11/0 I 11/0 I 11/10 I 11/11 I 11/12 I 11/13 111/14 111/11 I 11/10 I
                                           CALENOAD DAY (1070)
         Note:  Horizontal  bars  indicate 24-hour (midnight-to-midnight)
                 run  averages
                   Figure 4.    S02 EMISSIONS  DURING

                                  VENTURI/SPRAY  TOWER  RUN 907-1A
                                           361

-------
         1EGIN BUN 907 IB
                                               — G 10* PUMP MAIHTENANCE
2£
g Z    3.0 ^
                                             ,
                   I 11'" I 11/18 I 11/19 I 11/20 I 11/21 I 11/22 I 11/23 I 11/24 I 11/25 I 11/28 I 11/37 I 11/21 1 11/39 I 11/30 I 12/1 I 12/3
                                             ,
         I 12/4 | l2/« I 12/8 I 12/7 | ll/l I 12/9 I 12/10 I 12/11 I 12/12 1 12/13 I 12/14 I 12/16 I 12/10 I 12/17 [ 12/11 I 12/19 I 12720 I 12/21 I 12/23 I 12/23
                                             ,
            I 12/26 I 12/2B I 12/27 I 12/29 I 12/29 I 12/30 I 12/31 I 1/1 I 1/2 I 1/1 I 1/4  I 1/B 1 1/6 I 1/7 I  1/8 I 1/9  I 1/10 i  1/11 I 1/12
       1440    14*)    1520    1560    1600    1640    tUM    1770    1760    '*»    1840
                                        TEST TIME tow
        I 1/13 I 1/14 I I/IE i 1'16 I 1/17 I 1/19 I I'19 I 1/20 i 1/21 I 1/22  1 1/23 I 1/24 I 1/25 I 1/26 I 1/27 I 1/28 I 1/29 I 1/30 I  1/11 I 2/1
                                       CALENDAR DAV 11979)


Note:   Horizontal  bars  indicate 24-hour  (midnight-to-midnight)
          run averages
          Figure  5.    S02  EMISSIONS DURING
                          VENTURI/SPRAY TOWER  RUN  907-1B
                                   362

-------
spray tower Runs 907-1A and 907-1B were saved in a pile near the test facility.
During this winter no odor was apparent even when the pile was worked by a bull-
dozer. The pile will  continue to be monitored during the summer to determine if
an odor does exist.


TWO-LOOP KORCED-OXIDATION TEST RESULTS USING LIME SLURRY WITH ADDED ADIPIC ACID

The emphasis at Shawnee has been on adipic acid enhancement with limestone be-
cause this combination may prove to be the most economical route for achieving
high removal in throwaway systems.  However, two one^week runs made in July
1978 demonstrated that adipic acid is also effective in enhancing S02 removal
in lime systems.  Results of these lime runs are summarized in Table 2.

Effect of Adipic Acid Concentration - Two runs were conducted with lime on the
venturi/spray tower two-loop system with forced oxidation.  Run 951-1A was a
base case without adipic acid and Run 952-1A was under identical conditions
with nominally 1500 ppm adipic acid in the spray tower slurry liquor.  The
addition of adipic acid increased S02 removal  from 66 percent with no acid to
98 percent at 1500 ppm (2600 to 2750 average inlet S02 concentration).

Effect of Slurry Solids Concentration - In these lime runs, all clarified liquor
from the dewatering system was returned to the spray tower loop resulting in
relatively low spray  tower solids concentrations.  Apparently, a low solids
concentration of 4.5  percent in the adipic acid enhanced lime run was not detri-
mental because 98 percent S02 removal was achieved.  This may be attributed to
the high pH inherent  with lime systems, at which adipic acid becomes fully
effective.  However,  in the lime run without adipic acid, the low solids con-
centration was detrimental.  Run 951-1A averaged 66 percent SC£ removal at 5.7
percent spray tower slurry solids concentration.  This run can be compared with
a previous one-month  lime demonstration run under essentially the same condi-
tions which averaged  88 percent S0£ removal with 10.4 percent average spray tower
slurry solids concentration.

Additional adipic acid enhanced lime tests will  be made as time permits.
                                    363

-------
                                          Table  2

           VENTURI/SPRAY TOWER  TWO-LOOP TESTS WITH  FORCED  OXIDATION
                       - ADIPIC ACID  ENHANCED LIME  SLURRY  -
Major Test Conditions'1^
Fly ash loading
Adipic acid concentration in venturi , ppm
Adipic acid concentration in ST (controlled), ppm
Gas rate, acfm ? 300°F
Venturi slurry rate, gpm
ST slurry rate, gpn
Venturi solids recirculated (controlled), vt.%
Residence times (min)/tank level (ft): Spray tower EHT
Oxidation tank
Desupersaturation tank
Venturi inlet pH (controlled)'2'
ST inlet pH (controlled)
Air rate to oxidizer, scfm »
Run-Average Results:
Start-of-Run Date
Onstream hours
S02 removal , %
Inlet S02 concentration, ppm
Spray tower solids recirculated, wt.%
Venturi inlet pH
Spray tower inlet pH
Spray tower lime stoichiometric ratio
Spray tower inlet liquor gypsum saturation, %
Spray tower sulfite oxidation, %
Overall sulfite oxidation, %
Overall lime utilization, %
Venturi inlet liquor gypsum saturation, %
Venturi inlet liquor 503° concentration, ppm
Adipic acid concentration in venturi, ppm
Adipic acid concentration in ST, ppm
Air stoichiometry, atoms 0/mole SOg absorbed
Filter cake solids, wt.»
Mist eliminator restriction, %
951-1A
High
0
0
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
5.5
7.8
210

7/3/78
250
66
2750
5.7
5.6
7.90
1.14
90
18
95
97
95
130
0
0
2.05
86
0.5
952- 1A
High
( 2000)
1500
35,000
600
1600
15
14.7/10
11.3/18
4.7/4.8
5.5
7.8
260

7/20/78
278
98
2600
4.5
5.5
7.75
1.23
85
22
97
98
90
115
1585
1380
1.80
85
1
Notes:  (1)  All runs made with 9 in. HgO venturi pressure drop. Makeup water and clarified liquor in excess of mist
          eliminator wash were returned to spray tower EHT.
      (2)  Venturi inlet pH was controlled by separate lime addition.
                                            364

-------
                               Section 4

                 ADIPIC ACID ENHANCED SCRUBBING IN THE
              SHAWNEE TCA SYSTEM WITHOUT FORCED OXIDATION
Beginning in July and continuing to November 1978, a series of tests were
conducted.on the TCA system with adipic acid enhanced lime and limestone
slurries.  These tests were conducted without forced oxidation.  S0£ removals
ranging up to 95 percent have been achieved with adipic acid concentrations
of up to 1500 ppm in the recirculating slurry liquor.  Results of these adipic
acid enhanced tests without forced oxidation are reported in this section.


SYSTEM. DESCRIPTION

The TCA (turbulent contact absorber) was operated in a single-loop scrubbing
configuration as shown in Figure 6.  The TCA contained three beds of 1-7/8
inch diameter, 11.5 gram nitrile foam spheres retained between bar grids.
Each bed contained 5 inches static height of spheres.

The effluent hold tank was 7 feet in diameter and operated at a 17 foot slurry
level giving a 4.1 minute residence time at the slurry recirculation rate of
1200 gpm.  Alkali and adipic acid were added directly to the effluent hold  tank.

The dewatering system consisted of a clarifier followed by a solid bowl centri-
fuge.  All  clarified liquor from the dewatering system was returned to the  scrub-
ber loop either via the effluent hold tank or the mist eliminator underside sprays.
TEST RESULTS ON THE TCA WITH ADIPIC ACID ENHANCED LIMESTONE SCRUBBING
Beginning in'July 1978,  a series  of 7 limestone tests were conducted on the
TCA system to  demonstrate the effects of adipic acid on enhancing S0£  removal
in a system without forced oxidation.  Results of these tests are summarized
in Table 3.  Unless otherwise specified, controlled run conditions common to
all of the tests were:

                Fly ash  loading:   High (3 to 6 grains/scf dry)
                Flue gas rate:   30,000 acfm at 300°F
                Slurry  flow rate:   1200 gpm (50 gal/Mcf)
                Slurry  solids concentration:  15 percent
                                   365

-------
        REHEAT
FLUE GAS
ALKALI
ADIPICACID
                       1 FLUE GAS
                  /     \
                     000

                   ooooo
                    ooo
                   oo ooo
                     ooo
                   ooo oo
                              TCA
             EFFLUENT HOLD TANK
                                               MAKEUP WATER
CLARIFIED LIQUOR

FROM SOLIDS

DEWATERING SYSTEM
                                         BLEED TO SOLIDS
                                        DEWATERING SYSTEM
         Figure 6.  FLOW DIAGRAM  FOR ADIPIC ACID ENHANCED

                  SCRUBBING IN  THE TCA SYSTEM
                             366

-------
                                                                  Table  3

                                          TCA SINGLE-LOOP TESTS  WITHOUT  FORCED OXIDATION
                                              -  ADIPIC ACID  ENHANCED LIMESTONE SLURRY  -
Major Test Conditions'1'
Fly ash loading
Adipic acid concentration, ppm
Gas rate, acfm 
1
927-2A
High
300
30,000
1200
15
1.2
4.1
17
EHT

8/4/78
374
75
2300'
9.1
110
19
84
5.30
350
62
3
928-2A
High
1500
30,000
1200
15
1.2
4.1
17
EHT

8/22/78
252
93
2600
12.8
100
14
85
5.40
1600
59
0
929- 2A
High
750
30 ,000
1200
15
1.2
4.1
17
EHT

9/5/78
186
92
2300
11.2
110
10
80
5.50
885
59
0
930- 2A
High
750
30,000
1200
15
1.35
4.1
17
EHT

9/13/78
162
93
2550
12.6
80
13
75
5.60
700
58
0
931-2A
High
750
30,000
1200
15
1.05
4.1
17
EHT

9/20/78
137
77
2300
9.4
110
24
93
4.95
840
62
0
932-2A'3'
High
1500
20,000-30,000
1200
15
1.2
4.1
17
EHT

9/26/78
833
96
2450
4-18
110
21
82
5.30
1620
61
0
01
•-J
             Notes:
All runs were made with 3 beds and
EHT = in the effluent hold tank.
Long-term reliability test.
Clarifier only.
                                                   5 inches per bed of 1-7/8 inch diameter, 11.5 gram nitrile foam spheres.

-------
Effect of Adi pic Acid Concentration - TCA limestone Runs 926-2A through  929-2A
were all made under identical conditions except for adipic acid concentration.
In these runs limestone stoichiometry was controlled at 1.2 moles Ca/mole SOg
absorbed.  The effect of adipic acid on SO  removal was:

            TCA            Actual Adipic            Percent S02
            Run            Acid Cone., ppm          Removal	


           926-2A                 0                     71

           927-2A               350                     75

           929-2A               885                     92

           928-2A              1600                     93
S02 removal increased from 71 percent with no adipic acid to 92 percent with
885 ppm in the slurry liquor.  Increasing adipic acid further to 1600 ppm in-
creased S02 removal only slightly to 93 percent.  Thus, the majority of the
adipic acid enhancement was achieved in the TCA at a concentration somewhere
between 350 and 885 ppm.


Effect of Limestone Stoichiometry and pH - Limestone stoichiometry was explored
at a nominal adipic acid concentration of 750 ppm and a liquid-to-gas ratio of
50 gal/Mcf with the following results:


      TCA          Limestone Stoichiometry,       TCA          Percent S02
      Run          mole Ca/mole S02 absorbed    Inlet pH         Removal


      931-2A              1.05                     4.9             77

      929-2A              1.20                     5.5             92

      930-2A              1.35                     5.6             93


Thus, it is apparent that the system required a limestone stoichiometry of
only about 1.2 to maintain sufficiently high pH to achieve the full S02 removal
enhancement with adipic acid.  Additional limestone did not significantly
increase S02 removal.


TCA Demonstration Run 932-2A - Run 932-2A, a month long demonstration run, was
made with adipic acid enhanced limestone slurry to demonstrate both operational
reliability with respect to scaling and plugging and the removal enhancement
capability of the adipic acid additive.

The run began on September 26, 1978 and terminated on November 2,  1978 for a
total of 833 on-stream hours (35 days).  During the run, the scrubber was out

                                    368

-------
of service for 48 hours due to a boiler outage caused by a tube leak, 5 hours
for a scheduled inspection, and 8 hours for unscheduled outages to clean and
repair the scrubber I.D. fan damper.

Excluding boiler outages and scheduled inspections,  Run 932-2A operated with
an on-stream factor of 99.0 percent.

The run was controlled at a no'minal  1.2 limestone stoichiometric ratio and
1500 ppm adipic acid concentration in the slurry liquor.  Slurry solids con-
centration was controlled at 15 percent.  The flue gas flow rate was varied
between 20,000 and 30,000 acfm (8.4  to 12.5 ft/sec superficial gas velo-
city) as the boiler load fluctuated  between 100 and  150 MW.  The slurry recir-
culation rate was fixed at 1200 gpm  (L/G = 50 to 75  gal/Mcf).  As with all
runs during this test block, the effluent hold tank  residence time was only
4.1 minutes.

S02 removal during the run averaged  96 percent at an average inlet S02 concen-
tration of 2450 ppm.  Excluding the  first few days of unsteady-state operation,
S02. removal stayed within the narrow range of 94 to  98 percent as the inlet S02
concentration varied widely between  1400 and 3500 ppm.

S02 emissions were calculated for Run 932-2A on the  same basis as for the ven-
turi/spray tower Run 907-1A (see Section 3).  Figure 7 shows the calculated S02
emissions for the TCA run.  Average  S02 emissions for each 24-hour period (mid-
night-to-midnight) are indicated by  horizontal lines on the figure.

During the first seven days (September 26 through October 3), S02 emissions
were relatively high and widely varying.  The highest daily average emissions
were 1.1 Ib S02/MM Btu on September  27 and 0.9 Ib S02/MM Btu on both September
30 and October 3.  It should be noted, however, that the new source performance
standard of 1.2 Ib/MM Btu was never  exceeded on a daily average basis.

The relatively high S02 emissions resulted from frequent excursions to low pH
in the scrubber slurry liquor as the test personnel  were trying to control the
limestone stoichiometric ratio at 1.2 with widely varying inlet S02 concentra-
tions (1000 to 3500 ppm).  Beginning on October 6 after the boiler outage, a
scrubber inlet pH underride of 5.1 was implemented in addition to the limestone
stoichiometric ratio control value of 1.2.  This combined stoichiometry/pH control
produced the improved results shown  for the remainder of the run.

S02 emissions for the 27-day period  from October 6 through the end of the
run on November 2 averaged only 0.26 Ib/MM Btu.  The highest 24-hour average
    emission during this period was  only 0.44 Ib/MM  Btu.
The. mist eliminator was completely clean at the end of the run and the entire
scrubber system was free of scaling and plugging.

Limestone utilization during the run averaged 82 percent.  Discharge solids
from the centrifuge averaged about 61 percent which is typical of unoxidized
limestone sludge.

In summary,  the objectives of this run were met.  High removal was consistently
achieved at  a good limestone utilization and no fouling, scaling, or plugging
occurred.

                                    369

-------
                   40
                            80
                                   120       160      200       240      280      320       380      400      440     4M
                                                         TEST TIME, houn
               9/27 I  9/28 I 9/29 I  9/30 I 10/1 I 10/2 I  10/3 I 10/4 I  10/5 I  10/6 I  10/7 I 10/8 I 10/9 I 10/10 I 10/11 I 10/12 I 10/13 I 10/14 I  10/15 I 10/16
                                                       CALENDER DAYf1978)
<2
SS
a. a.
           RUN 932-2A CONTINUED
                    INSPECTION -
                                          -I.C. FAN DAMPER CLEANED
                                                                                      END HUN 932-2A I
                  b20
                                                                                                               2.0
                                                                                                               1.2
                                                                    760
                                                                            800       840
                                                                                                     920      960
                          560       600      640      680       720
                                                   TEST TIME, hrairi
             10/17 I lO/'S I 10/19 ! 10/20 1071 10/2? ] 10,23 ' 10/24 i 10/25 i 10/26 i  10/27 ' '.0/28 I  10/29 I  10/30 I 10/31 ' 11/1  I 11/2  I '1/3 I 11/4 I
                                                  CALENDAR DAYH978I
         Note:   Horizontal  bars  indicate  24-hour (midnight-to-midnight)
                   run  averages
                             Figure  7.    S02  EMISSIONS  DURING  TCA RUN  932-2A
                                                 370

-------
Adipic Acid Consumption - As with the forced-oxidation runs on the venturi/spray
tower system, adipic acid consumption was greater than anticipated.  An adipic
acid material balance calculation was made during Run 932-2A between October 10
and October 30, 1978, a total of 21 days.


                   Run 932-2A Adipic Acid Consumption


                                         Ib/hr      Ib/ton limestone feed

Discharged with centrifuge cake           1.9                 4,2

Unaccounted                               2.2                 5.0
Total                                     4.1                 9.2
As already discussed in Section 3, the unaccounted loss in this run without
forced oxidation of 2.2 Ib/hr compares with 2.9 Ib/hr for venturi/spray tower
Run 907-1B with forced oxidation.  Thus, it appears that air sparging for
forced oxidation may increase adipic acid losses.
TEST RESULTS ON THE TCA WITH ADIPIC ACID ENHANCED LIME SCRUBBING

Although the majority of the adipic acid enhanced tests on the TCA were with
limestone slurry, a single week-long test was conducted with lime slurry.   Table
4 summarizes the results of this test along with base case lime tests without
adipic acid.


Effect of Adipic Acid Concentration - The runs were conducted with lime on the
TCA system, two without adipic acid and one with.  Run 976-2A was a lime run
at a scrubber inlet pH of 7 with 15 percent slurry solids concentration S02
removal averaged 70 percent.  In Run 976-2B the slurry solids concentration was
dropped to 8 percent and S02 removal dropped slightly to 67 percent.  Finally,
Run 977-2A was made with 8 percent slurry solids and nominally 300 ppm (actual
average 420 ppm) adipic acid in the slurry liquor.  S0£ removal increased  to
80 percent, an enhancement of 10 to 13 percentage points over the base cases
with only a small addition of adipic acid.

It should be pointed out, however, that S02 removals in the mid 80's can be
achieved in the Shawnee TCA system with lime alone by raising the scrubber inlet
pH to 8 at 15 percent slurry solids.  Additional tests will be conducted as time
permits to work out the relationships between adipic acid concentration, scrubber
pH, and slurry solids concentration.
                                    371

-------
                                                          Table  4
                                       TCA SINGLE-LOOP TESTS WITHOUT FORCED  OXIDATION

                                            - ADIPIC  ACID ENHANCED LIME SLURRY -
ho
Major Test Conditions' '
Fly ash loading
Adipic acid concentration, ppm
Gas rate, acfm @ 300°F
Slurry rate, gpm
Solids recirculated, wt.%
TCA inlet pH controlled at
EHT residence time, min.
EHT level, ft
i o\
Lime addition pointv '
Run-Average Results
Start-of-Run Date
Onstream hours
S02 removal , %
Inlet S02 concentration, ppm
S02 make-per-pass, m-moles/ liter
TCA inlet liqudr gypsum saturation, %
Sulfite oxidation, %
Lime utilization, %
TCA inlet pH
Adipic acid concentration, ppm
Centrifuge solids, wt.%
Mist eliminator restriction, percent
976-2A
High
0
30,000
1200
15
7.0
4.1
17
DC

7/12/78
160
70
2850
10.5
90
16
91
7.0
0
63
0.2
976-2B
High
0
30,000
1200
8
7.0
4.1
17
DC

7/19/78
133
67
2950
10.7
85
12
93
7.0
0
63
1
977-2A
High
300
30,000
1200
8
7.0
4.1
17
DC

7/25/78
236
80
2700
11.6
90
10
92
6.95
420
61
1
               (1)  All runs were made with 3 beds and 5 inches per bed of 1-7/8  inch diameter, 11.5 gram nitrile foam spheres
               (2)  -----•-•
DC = in the scrubber downcomer

-------
                               Section 5

                DEWATERING CHARACTERISTICS OF ADIPIC ACID
                  ENHANCED LIMESTONE SLURRY AT SHAWNEE
Settling and dewatering characteristics of slurry solids are routinely moni-
tored in the Shawnee laboratory by cylinder settling tests and vacuum funnel
filtration tests.  A comparison of the results of these monitoring tests for
limestone slurry with and without adipic acid addition is presented in this
section and summarized in Table 5.

Cylinder settling tests are performed with slurries containing 15 percent
solids at room temperature in a 1000 ml cylinder containing a rake which
rotates at 0.16 rpm.  The initial settling rate and ultimate settled solids
concentration are recorded as indices of dewatering characteristics. The ini-
tial settling rate is only a qualitative index of the solids settling proper-
ties.  Design rates for sizing clarifiers must take into consideration the
hindered settling rate as the solids concentrate.  The ultimate settled solids
from the cylinder tests represent the highest solids concentration achievable
in a settling pond.

Funnel filtration tests are"performed in a Buchner funnel with a Whatman 2
filter paper under a vacuum of 25 in. Hg.  The funnel tests correlate well
with the Shawnee rotary drum vacuum filter when not blinded but the funnel
test cakes tend to have lower solids concentration.

Table 5 lists the effects of adipic acid on both oxidized and unoxidized lime-
stone slurries.  The data reported are for a range of adipic acid concentra-
tion of 1500 to 3000 ppm.  All samples were with high fly ash loadings in
which about 40 percent of the slurry solids was fly ash.  All tests were con-
ducted with samples containing 15 percent slurry solids.

As reported previously, settling and filtration characteristics of oxidized
slurry are much superior to the characteristics of unoxidized slurry.  The
same trend exists with the slurry samples containing adipic acid.

The average initial settling rate for oxidized limestone slurry decreased
from 1.0 cm/min with no adipic acid to 0.5 cm/min with adipic acid.  However,
this rate was still considerably higher than settling rates without forced
oxidation.  Without forced oxidation, the initial settling rate averaged 0.2
cm/min with or without adipic acid.

Adipic acid had little or no effect on ultimate settled solids or funnel test
cake solids.  The data indicated a slight decline in solids quality with adi-
pic acid but the decline was small compared with the difference between lime-
stone slurry with forced oxidation and without.

                                    373

-------
                                                       Table  5

                                     COMPARISON  OF  SHAWNEE WASTE SLURRY DEWATERING
                                 CHARACTERISTICS WITH AND WITHOUT ADIPIC ACIU ADDITION
u>
Alkali
LS
LS
LS
LS
Fly Ash
Loading
High
High
High
High
Oxidation
Yes
Yes
No
No
Adi pic
Acid
No
Yes
No
Yes
Initial Settling
Rate, cm/mi n
Avg.
1.0
0.5
0.2
0.2
Range
0.6-1.5
0.3-0.9
0.1-TJ.5
0.1-0.3
Ultimate Settling
Solids, wt %
Avg.
72
72
54
51
Range
62-86
59-83
41-67
42-69
Funnel Test Cake
Solids, wt %
Avg.
72
69
57
56
Range
65-88
59-77
48-66
49-73

-------
In addition to the funnel  filtration tests,  the filter cake solids from the
rotary drum vacuum filter  were monitored.   During forced-oxidation operation
with adipic acid enhanced  limestone, the rotary drum filter cake solids concen-
tration ranged from 80 to  87 weight percent  solids (see Table 1).  This range
is no different than that  obtained when operating with oxidized sludge in the
absence of adipic acid.

In summary, the only significant effect of adipic acid addition on dewatering
characteristics was a decline in initial settling rate with oxidized limestone
slurry.  These observations agree generally  with those at the IERL-RTP pilot
plant.
                                   375

-------
                               Section 6

                PRELIMINARY ECONOMICS OF ADIPIC ACID
                         ENHANCED SCRUBBING


At the request of the Shawnee Steering Committee, the Emission Control  Develop-
ment Projects group of TVA projected preliminary economics of adipic acid addi-
tion for forced-oxidation systems designed to achieve an average of 90  percent
SO  removal  from high sulfur flue gas.  The preliminary results indicate that
both capital and operating costs are reduced by about 5 percent for a limestone
system with 750 to 1500 ppm adipic acid compared with a limestone system with
no additive.

Conditions for these preliminary evaluations were prepared by Bechtel and are
presented in Table 6.  Results are listed in Table 7.  The evaluations  were
based on a 500 MW scrubbing facility incorporating forced oxidation and operat-
ing on flue gas from coal containing 4 wt.% sulfur.  The evaluations included
$5/dry ton for waste solids disposal but excluded land costs for a disposal
site and costs for preparing land.

The cases evaluated were:

     •  Case 1 - A limestone base case operated at relatively high liquid-to-gas
                 ratio and limestone stoichiometric ratio required to achieve
                 90 percent S02 removal.  Operation at 90 percent S02 removal
                 with limestone alone, although possible, has not been  demon-
                 strated at Shawnee.  Two effluent hold tanks were included for
                 forced oxidation in the first tank at lower pH before  limestone
                 is added in the second tank.

     •  Case 2 - A limestone case with MgO addition. Oxidation of the scrubber
                 bleed stream was chosen in this case because oxidation within
                 the scrubber loop is incompatible with magnesium enhanced
                 scrubbing.

     •  Cases 3, 4, and 5 - Limestone cases with adipic acid addition.   Adipic
                 acid concentrations in the scrubber liquor of 750, 1000, and
                 1500 ppm were evaluated.  Adipic acid consumptions of  five
                 times theoretical were used to allow for the level of  un-
                 accounted losses observed at Shawnee.  Shawnee tests made after
                 conditions were chosen for the economic evaluation indicate
                 that SO  removal for these cases should be higher than 90 per-
                 cent.  A single hold tank was chosen for these cases because
                 the low scrubber inlet pH with adipic acid is compatible with
                 forced oxidation.


                                    376

-------
                                 Table 6

               CONDITIONS  FOR PRELIMINARY ECONOMIC  ANALYSIS
                  OF ADIPIC  ACID ENHANCED LIME/LIMESTONE
                    WET  SCRUBBING WITH FORCED OXIDATION
Capacity:
Coal :
Scrubber:

Superficial Gas Velocity:
Scrubbing Mode:
Number of Scrubbing Trains:
Dewatering:

Sludge Disposal :


Onstream Factor:
Case No.
Alkali
Additive

Additive cone. , ppm
Additive rate, Ib/hr
Percent S02 removal
L/G, gal/Mcf
Alkali stoic, ratio,
mole Ca/mole S02 absorbed
Inlet pH
Filter cake solids, wt %
Sulfite oxidation, %
Air stoich., atoms O/
mole S02 absorbed
Mode of oxidation

500 MM
4 wt % sulfur
TCA with 3 beds, 4 grids, and 5 inchs
sphere height per bed
12.5 ft/sec
Single loop with forced oxidation
4
To 85 wt % solids by thickener and rotary
drum vacuum filter
Includes $5/dry ton for waste solids
disposal but excludes land costs for a
disposal site and costs for preoaring land
7000 hours of operation/year
12345
LS LS LS LS LS
MgO Adipic Adipic Adipic
Acid Acid Acid
55001 750 1000 1500
168 642 852 1282
90 90 90-95 90-95 95
58 50 50 50 50

1.55 1.20 1.20 1.20 1.20
5.8 5.4 5.4 5.4 5.4
85 85 85 85 85
99 99 99 99 99

1.7 1.7 1.7 1.7 1.7
in loop bleed in loop in loop in loop
(2 EHT) stream (1 EHT) 0 EHT) (1 EHT)













6
Lime
Adipic
Acid
1000
802
95
50

1.05
7.0
85
99

1.7
in loop
(2 EHT)
Notes:  Effective Mg++
       2Five times, theoretical
         consumption
                                   377

-------
u>
-~J
00
                                                           Table  7

                                         RESULTS  OF  PRELIMINARY  ECONOMIC ANALYSIS  OF
                           ADIPIC ACID ENHANCED  LIME/LIMESTONE  WET SCRUBBING WITH FORCED OXIDATION
Case
1
2
3
4
5
6
Alkali/Additive
Limestone
LS/MgO
LS/Adipic Acid
LS/Adipic Acid
LS/Adipic Acid
Li me/ Adi pic Acid
Additive
Cone. , ppm
_
55001
750
1000
1500
1000
Average
Percent
SO, Removal
90
90
90-95
90-95
95
95
Total Capital
Investment3
$HM(1979)
41.5
41.0
39.3
39.4
39.5
38.8
$/kW
83.1
82.1
78.5
78.7
79.0
77.6
First Year
Revenue
Requirement2
$MH(1980)
20.9
20.1
19.6
19.8
19.9
21.4
Mills/kWh
5.96
5.76
5.60
5.64
5.69
6.11
               Notes:   Effective Mg

                       2Includes 17.2% annual capital charge

                       3Does not include land costs for a disposal  site or costs for preparing land
               Raw Materials Costs:  Limestone   - $7/ton
                                    Lime       - $42/ton
                                    MgO        - $300/ton
                                    Adi pic Acid - $840/ton

-------
     •  Case 6 - A lime case with adipic acid addition.  As with the limestone
                 cases, five times theoretical  consumption of adipic acid was
                 used to allow for unaccounted  losses.  At the conditions chosen,
                 S02 removal,may be as high as  95 percent.  Two effluent hold
                 tanks were included to allow forced oxidation in the first tank
                 before lime addition in the second.

For the cases studied, as shown in Table 7, annual  revenue requirements were
lowest for the adipic acid enhanced limestone cases.  The annual revenue require-
ments included a 17.2 percent annual capital charge.

Annual revenue requirement for limestone with 1500  pprn adipic acid (Case 5) was
5.69 mills/kWh compared with 5.96 for unenhanced limestone (Case 1), a savings
of almost 5 percent.

Scrubbing economics were insensitive to adipic  acid consumption.  An increase
of adipic acid concentration from 750 ppm (Case 3)  to 1500 ppm (Case 5) increased
annual revenue requirements by only 1.5 percent from 5.60 to 5.69 mills/kWh.
These values include 5 times theoretical adipic acid consumption.

The annual revenue requirement for magnesium enhanced limestone (Case 2) was 5.76
mills/kWh, lower than unenhanced limestone.  Capital charges for the magnesium
enhanced limestone case were higher than for the adipic acid enhanced cases be-
cause of the additional equipment required for  bleed stream oxidation relative
to oxidation within the scrubber loop.

The annual revenue requirement for adipic acid  enhanced lime (Case 6) was 6.11
mills/kWh, the highest of the cases evaluated.   This high value reflects the
high cost of lime relative to limestone and the additional hold tank required
for forced oxidation.

It should be noted that the differences in annual  revenue requirements among
these cases are small.  The principal conclusion from these preliminary evalua-
tions is that adipic acid addition does not increase costs but, in fact, de-
creases them slightly.  Furthermore, costs are  insensitive to adipic acid con-
sumption over the practical range expected in FGD scrubbing.
                                   379

-------
                                Section 7

                     SUMMARY OF CHARACTERISTICS OF
                  ADIPIC ACID AS A SCRUBBER ADDITIVE


Based on testing at the IERL-RTP pilot plant and at the Shawnee Test Facility
and on the preliminary economic evaluations conducted by TVA, the characteris-
tics of adipic acid as a scrubber additive can be summarized as follows:


BENEFICIAL EFFECTS

     •  Significantly enhances S02 removal efficiency

     •  Increases alkali utilization, hence decreases waste solids disposal
        requirements

     •  When used with limestone, has projected lower capital and operating
        costs than unenhanced limestone or limestone/MgO

     •  Can be used with both lime and limestone in either conventional  or
        forced-oxidation modes for both new and existing installations

     •  Is not adversely affected by chlorides as is the limestone/MgO process

     •  Does not significantly affect solids quality (filterability/settling
        rate) as can occur with high magnesium ion concentrations

     »  Should promote use of less expensive and less energy intensive lime-
        stone rather than lime

     •  With proper pH control, steady outlet S02 concentrations can be main-
        tained even with wide fluctuations of inlet SOg concentrations


NEGATIVE ASPECTS

     •  Has unpleasant odor associated with adipic acid decomposition product

     •  Adipic acid decomposition requires adding up to 5 times that theoreti-
        cally required (However, consumption over the ranges anticipated has
        negligible economic impact)

     •  Other possible secondary environmental effects have not yet been deter-
        mined.  Separate studies are underway to determine if any such problems
        might exist.

                                    380

-------
                              Section  8

              FUTURE EPA TEST PROGRAM  WITH ADIPIC ACID
The EPA/IERL-RTP  test  program with  adipic  acid  enhanced  scrubbing  systems  is
ongoing.   Figure  8 shows  the full  program  outline.   With contingencies  it  is
anticipated that  the program will  be  completed  by the  end of  1979  or  early
1980.   The following activities  are either planned  or  proceeding:


     •  Tests  at  Shawnee  to  develop a single-loop forced oxidation system
        with adipic acid  enhancement  for both  lime  and limestone

     •  Factorial  tests at Shawnee  to develop the interrelationships  of
        operating parameters for both lime and  limestone and  both  with
        and without forced oxidation

     •  Tests  at  Shawnee  to  determine if the slurry bleed stream can  be
        oxidized  outside  the scrubber loop

     •  Full  scale demonstrations,  both  with and  without forced oxidation,
        of adipic acid enhanced  limestone  scrubbing

     •  Studies by Aerospace Corporation of the handling and  disposal
        characteristics of waste sludges produced at Shawnee  with  adipic
        acid enhanced  scrubbing

     t  Level  1 bioassay  studies by Litton Bionetics Corporation to deter-
        mine biological activity,  if  any,  from  adipic  acid addition

     •  Evaluations by TVA of the  economics of  adipic  acid addition

     •  Studies by Radian Corporation to evaluate the  unaccounted  losses
        of adipic acid and to develop better analytical  procedures

     •  A limited market  study to  determine the effect of extensive adipic
        acid consumption  in  FGD  scrubbing  on the  adipic  acid  market.
                                    381

-------

ITEM
1. SHAWNEE TEST FACILITY
V/ST LIME WITH OXIDATION

V/ST LIMESTONE BLEED OXIDATION
V/ST LIMESTONE BIOASSAY
TCA LIME
TCA LIMESTONE
TCA LIMESTONE BIOASSAY


2. FULL-SCALE DEMONSTRATION
3. SPECIAL STUDIES
UNACCOUNTED LOSSES
ANALYTICAL PROCEDURES

SCRUBBER ECONOMICS
MARKETS



JUL

• •



•
•_













AUG





•
_•!












19
SEPT






i^m












78
OCT






mm













NOV




















DEC




















JAN




•


•












FEB




•
•

•












MAR





•














APR




















MAY



















19
JUN



















79
JUL




















AUG










^m









SEPT










•m









OCT










^m









NOV










^m









DEC



















00
                                Figure  8.   SHAWNEE ADVANCED PROGRAM PROJECTED TEST SCHEDULE

-------
                                Section  9

                                REFERENCES


 1.    Rochelle,  G.T.,  "The Effect  of Additives  on  Mass  Transfer in  CaC03
      and CaO Slurry Scrubbing of  SO?  from  Waste Gases,"  Ind.  Eng.  Chem.,
      pp. 67-75, 1977.

 2.    Rochelle,  G.T.,  "Process Alternatives for Stack Gas  Desulfurization
      by Throwaway Scrubbing", Proceedings  of 2nd  Pacific  Chemical  Engineer-
      ing Congress, Vol.  I, p. 264,  August  1977.

 3.    Borgwardt, R.H.,  Significant EPA/IERL-RTP Pilot Plant  Results.  EPA
      Industry Briefing,  Research  Triangle  Park, NC, August  29,  1978.

 4.    Bechtel Corporation, EPA Alkali  Scrubbing Test Facility:   Summary of
      Testing through  October 1974,  EPA  650/2-75-047,  (NTIS  PB  244901),
      June 1975.

 5.    Bechtel Corporation, EPA Alkali  Scrubbing Test Facility:   Advanced
      Program, First ProgressTReport,  EPA-600/2-75-050.  (NTIS  PB 245279).
      September 1975.

 6.    Bechtel Corporation, EPA Alkali  Scrubbing Test Facility:   Advanced
      Program, Second  Progress Report. EPA-600/7-76-008,  (NTIS  PB 258783).
      September 1976.

 7.    Bechtel Corporation, EPA Alkali  Scrubbing Test Facility:   Advanced
      Program. Third Progress Report,  EPA-600/7-77-105,  (NTIS  PB 274544),
      September 1977.

 8.    Bechtel National, Inc., EPA  Alkali  Scrubbing Test  Facility:   Advanced
      Program, Fourth  Progress Report, to be published,  Summer  1979.

 9.    Head,  H.N. et al, Results of Lime  and Limestone Testing with  Forced
      Oxidation  at the  EPA Alkali  Scrubbing Test Facility,  in  Proceedings:
      Symposium on Flue Gas Desulfurization - Hollywood, FL, November  1977,
      EPA-600/7-78-58a, (NTIS PB 282090), March 1978  (pp.  170-204).

10.    Head,  H.N.,  Results of Lime  and  Limestone Testing  with Forced Oxidation
      at the EPA Alkali Scrubbing  Test Facility -  Second Report, EPA  Industry
      Briefing,  Research  Triangle  Park,  NC, August 29,  1978.
                                    383

-------
11.    Borgwardt,  R.H.,  Effect  of Forced Oxidation on Llmestone/SOx Scrubber
      Performance, in Proceedings:   Symposium on Flue Gas Desulfurization  -
      Hollywood,  FL, November  1977,  EPA-600/7-78-058a, (NTIS PB 282090),
      March 1978  (pp. 205-228).

12.    Meserole, F.B., Adi pic Acjd Degradation in FGD Systems,  progress  report
      for EPA Contract  68-02-2608, Task 58,  Radian Corporation, Austin, TX,
      December 1978.
                                    384

-------
                                Appendix

                       CONVERTING UNITS OF MEASURE
Environmental  Protection Agency policy is to express all measurements in
Agency documents in metric units.  In this report, however, to avoid undue
costs or lack  of clarity, English units are used throughout.  Conversion
factors from English to metric units are given below:
To Convert From
    To
                             Multiply By
scfm (60°F)
cfm
°F
ft
ft/hr
ft/sec
ft2
ft2/tons per day

gal/mcf
gpm
gpm/ft^
gr/scf
in.
in. HoO
Ib
Ib-moles
Ib-moles/hr
Ib-moles/hr ft2
Ib-moles/min
psia
m
nm/hr (0°C)
m/hr
°C
m
m/hr
m/sec
,,r
m^/metric tons
    per day
1/m3
1/min
l/min/m2
gm/m3
cm
mm Hg
gm
gm-moles
gm-moles/min
gm-moles/min/m2
gm-moles/sec
kilopascal
1.61
1.70
(°F-32)/1.8
0.305
0.305
0.305
0.0929
0.102

0.134
3.79
40.8
2.29
2.54
1.87
454
454
7.56
81.4
7.56
6.895
                                   385

-------
TVA COMPLIANCE PROGRAMS FOR S02 EMISSION
            G. A. Hollinden
            Energy Research
      Tennessee Valley Authority
        Chattanooga, Tennessee
             C. L. Massey
     Power Supply Planning Branch
      Tennessee Valley Authority
        Chattanooga, Tennessee
                386

-------
                               ABSTRACT

     On April 19, 1976, a United States Supreme Court ruling favored
mandatory constant S0_ control requirements.   Since that time, TVA has
been implementing a billion dollar program to bring its power plants into
compliance with State and Federal sulfur dioxide emission requirements.
A consent decree was approved by the TVA Board of Directors on December 14,
1978, which will become effective upon approval by the Court.  This paper
will discuss the implementation of the compliance program at the TVA plants
where flue gas desulfurization (FGD) will be used—Widows Creek, Paradise,
Johnsonville, and Cumberland Steam Plants.  Results associated with the
Widows Creek unit 8 wet limestone scrubbing system will be presented.  Coal
washing, magnesium oxide scrubbing, and other innovative FGD processes will
also be discussed.
                                387

-------
               TVA COMPLIANCE PROGRAMS FOR S02 EMISSION

     TVA has either purchased lower sulfur coal or is installing needed
control equipment to bring its coal-fired plants into compliance with
sulfur dioxide emission requirements.   The compliance plans for TVA's 12
coal-fired plants are based on the use of medium-or low-sulfur coal, the
use of conventional coal-washing, and in the case of four plants, partial
scrubbing of the plant flue gas.   Four of these plants already meet emis-
sion standards, and compliance at three other plants will be achieved in
1979.  At the five remaining plants, major installations of coal-washing
facilities, baghouse collectors,  or scrubbers are required; therefore,
compliance will not be achieved until 1981 or later.
     This paper will discuss compliance measures for those plants that
require FGD systems—Widows Creek, Paradise, Johnsonville, and Cumberland.

WIDOWS CREEK STEAM PLANT
General Plant Description
     TVA's Widows Creek Steam Plant is located on the Tennessee River in
Jackson County, Alabama.  The power plant consists of eight units (burning
pulverized bituminous coal).  These units have a combined generator capa-
city of about 1,977,985 kW.  There are six units in one powerhouse; five
of which have a capacity of 140,625 kW each, and one with a capacity of
149,850 kW.  The other two units, one with 575,010 kW and a second with
550,000 kW output, are in a second powerhouse.  The compliance schedule
requires that by September 1981 the Widows Creek Steam Plant will meet
a plantwide emission limitation of  1.2 Ibs. 803 per million Btu's heat
input on a 24-hour average basis.  To meet this emission limitation, the
six  smaller units will burn a coal that ensures compliance with an emis-
sion limit of no more than 1.6 Ibs of S02 per million Btu, and limestone
scrubbers will be used on Units 7 and 8 that will reduce S02 emissions to
no more than 0.9 Ibs SOg per million Btu's for these units.
     A full-scale wet limestone scrubbing system was installed on Unit 8.
It was completed in 1977 and is now in full operation.  The background
history and design premises were  discussed in a publication by McKinney,
Little, and Hudson.1  Operating experience was recorded in a publication
                                388

-------
by Wells,  Muirhead, and Buckner.2  The data in these publications will

not be repeated, but results previously reported will be updated with

emphasis on the operating problems associated with running this scrubbing

system.  Figure 1 is a flow diagram of the wet limestone scrubbing system.

     S(>2 removal efficiences, which result in emission levels of 1.2 Ibs

or less per million Btu's, have been achieved.  However, during the period
that the scrubber has been in operation, the unit has operated at reduced
load due to boiler problems.  Several problems have prevented sustained

operation of the scrubber system.  The major problems and solutions are
as follows:

     1.   Dampers - Each scrubber has three guillotine dampers (one each
          for the gas inlet, outlet, and bypass ducts).  Due to defective
          seals, fly ash and flue gas leakage has caused failure of elec-
          trical components, deterioration of protective boots, and jam-
          ming of the drive mechanism.  Andco dampers were originally
          installed.  Currently, a modified Andco and two other commer-
          cial dampers (Mosser and Damper Design Industries) are being
          tested.  No data are available at this time.
     2.   Expansion Joints - The 56 expansion joints in the scrubber
          system are made of a 5-ply material of asbestos, teflon, and
          stainless-steel mesh.  Solids, which have collected between
          the abrasion shield and joint, have caused the material to
          be worn through in several places with resulting gas leakage.
          Operating experience has indicated that some of the expansion
          joints can be eliminated while others will be modified by
          replacing the material with flexible stainless steel.
     3.   Rubber Linings - To prevent corrosion, the scrubbers have a
          rubber lining.  Failure of this lining, due to poor metal
          adhesion, seriously affected the operation of the scrubbers.
          The failures occurred on the sloping sections of the absorber
          and venturi hoppers.  To solve this problem, the linings in
          the sloping sections were replaced with 316-L stainless-steel
          plate welded to the carbon-steel shell.  The remainder of the
          rubber lining has operated satisfactorily.

     4.   Ball Mill - The wet limestone ball mill was designed for 50
          tons per hour, but during initial operation excessive limestone
          and ball rejection occurred at rates above 20 tons per hour.
          Inspection reve'aled that the metal helix designed to retain
          the balls and control limestone rejection was inadequate.  The
          helix was modified, and grinding rates of over 40 tons per hour
          can now be achieved with a size consisting of 90 percent minus
          200 mesh.

     5.   Absorber Grids and Nozzles - The absorber section of each train
          has five trays and grids.  Originally, the three lower grids
          were 316-L stainless steel and the  two upper ones were fiber-
          glass reinforced plastic.  Poor spray distribution from the
                                389

-------
                                 STEAM
O
                   FROM
                   TRAINS
                   B, C & D
  FROM  ESP
cr
                                          ENTRAINMENT
                                           SEPARATOR
TO ASH

* DISPOSAL POND
TO STACK PLENUM
1


- •

            AIR HEATER COILS
                                                                               AIR FROM
                                               AIR  HEATER FAN
                                                                              POWER  HOUSE
                                                            TO
                                                          TRAINS
                                                          B, C a Df
                                                                    SLURRY PUMP
                                                                    SEAL  WATER
                                                                      HEADER
                                                      RIVER
                                                                                                   O
o
  FROM B,CS D TRAIN
VENTURI CIRC TANKS

 TO  SETTLING
'    POND
                                                                                   FROM POND WATER
                     FAN
VENTURI CIRC
TANK 8 PUMPS
                                                ABSORBER CIRC
                                                TANK 6 PUMPS
                  EFFLUENT SLURRY
                 SURGE TANK  ft PUMPS
                                    TO
                                  TRAINS;
                                  B, C a

                                  LS SLURRY
                                  FEED  PUMPS
                                                                                   LIMESTONE
                                                                                    SLURRY
                                                                                  STORAGE  TANK
         Figure 1.   Flow Diagram of Widows Creek Unit 8 Scrubber System

-------
          nozzles caused erosion of the plastic grids.  The location of
          the two types of grids were reversed, and several spray nozzles
          are being tested in an effort to improve the slurry distribution.
     Although there have been several problems associated with the scrubbing
system, availability and operability have gradually increased since
startup.  The total capital cost for the Unit 8 scrubber system was
$54 million ($100/kW).
     A contract was awarded to Combustion Engineering, Inc., to supply a
limestone FGD process for the 575,010 kW Widows Creek Unit 7.  The schedule
for this unit requires completion of installation by March 1, 1981, and
shakedown and performance tests by September 1, 1981.  A bonus will be
paid if the schedule is improved and liquidated damages assessed if it is
delayed.  The cost of the system as provided by Combustion Engineering is
$23,555,200; however, TVA estimates that the total cost, including TVA's
internal expenditures, will will be $54 million or about $100/kW.
     Figure 2 is a schematic drawing of the FGD system for Unit 7.  The
system is designed to remove particulate matter and sulfur oxides.  The
flue gas, after passing through an electrostatic precipitator, is drawn
into the particulate removal section, which is a converging section con
taining staggered layers of 317-L stainless-steel rods.  The vertical
spacing between the rods Will be controlled to maintain a predetermined
pressure drop to provide maximum particulate removal.  This section, in
addition to removing a large portion of the particulate matter, is
responsible for some of the SC>2 removal.  The scrubbing slurry in the
particulate removal section is introduced by means of replaceable jar-type
ceramic spray nozzles.  A portion of the slurry is sprayed on the inlet
walls to keep them clean, and the remainder is sprayed directly on the
rods.  To prevent solid buildup at the wet-dry interface, a steam soot
blower is located in the inlet duct.  The run off slurry from the particu-
late removal section is discharged directly into the reaction tank.  From
the particulate removal section, the flue gas turns, passes through
ladder vanes to assure even distribution, and enters the spray tower.
     In the spray tower, multiple stages of sprays are used to distribute
the slurry.  The ceramic nozzles atomize the slurry into a fine spray which
provides the large surface area needed for the mass transfer of S02 into the
liquid.  Upon reaching the bottom of the absorber, the slurry is discharged
directly into the reaction tank.

                                391

-------
ro
           SCRUBBED
           FLUE GAS
        ESP
O

o
             B, C a D TRAIN
VENTURI CIRC  TANKS


 TO  SETTLING
            POND

«•*-
I '



1

*f^3—
H$—
VENTUR
TANK a
i
i
GIF
PUMF
                                                        TANK a PUMPS
                       EFFLUENT  SLURRY
                       SURGE TANK a PUMPS
   TO
 TRAINS!
B, C a

 LS  SLURRY
 FEED PUMPS
                                                                                         LIMESTONE
                                                                                           SLURRY
                                                                                        STORAGE TANK
                                                                                                          O
                                                                            SLURRY PUMP
                                                                            SEAL WATER
                                                                          -=d±EADER
                                                                                         FROM  POND WATER
                    Figure 2.  Flow Diagram of Widows Creek Unit 7 Scrubber System

-------
     The cleaned flue gas passes through the first stage separation section
called the bulk entrainment separator (BES).  The BES consists of 6-inch
fiberglass reinforced plastic vanes mounted at a 45-degree angle on 2-inch
parallel spacing.   The upper- or mist-eliminator section is constructed
of Vee-shaped fiberglass reinforced plastic vanes arranged in a series of
chevrons across the gas flow path.   Two rows of chevrons are used to mini-
mize mist carry over.  Retractable  water lances, which rotate 360 degrees
with pairs of opposed nozzles at the ends and mid points and which are
located between the BES and the lower-level chevrons, are used to clean
the mist eliminators.
     The temperature of the flue gas leaving the mist eliminators is
about 120°F, and almost all the entrained moisture has been removed.
However, the gas is saturated and contains a small amount of entrained
moisture which, if condensation occurs, would have a detrimental effect on
the induced draft fans, ductwork, and stacks.  To prevent condensation
and the resultant problems, the gas passes up through a finned-tube reheater
which increases its temperature about 50°F.  The gas travels from the
reheaters through the outlet ductwork, to the ID fans, and then to the
stack.
     The reaction tanks, located below the particulate-removal section
and the absorber, are constructed of carbon steel.  Retention of the spray
solutions in the reaction tanks allows time for the completion of chemical
reactions and the precipitation of calcium sulfite and sulfate.  The
required make up water and limestone additive are piped to the absorber
reaction tank.  The solution is recirculated from the reaction tanks to
the spray nozzles with rubber-lined spray pumps.
     To convert the calcium sulfite to calcium sulfate, forced-oxidation
equipment is being tested on one train of Unit 8.  If successful it will
be installed on both Units 7 and 8.  The calcium sulfate which is formed
when calcium sulfite is oxidized can be more easily disposed of than the
sulfite.
     A portion of the slurry is bled off from the particulate-removal
reaction tank to provide necessary solids removal from the cycle.  The
bleed rate to the disposal system is regulated to maintain the proper
slurry concentration.
                                393

-------
     Disposal of the solids produced by the FGD systems on Widows Creek
Units 7 and 8 will originally be accomplished by ponding.  The 110-acre
pond presently serving Unit 8 is expected to provide adequate storage
for disposal of the sludge from both Units 7 and 8 until late 1983.
Studies on sludge minimization, further handling, and fixation will be
conducted; and a decision about optimal long-range sludge disposal and
land reclamation will be made before commercial operation of Unit 7.
The alternatives to ponding raw sludge which are being investigated are
dewatering, mixing, layering, oxidation, and fixation.
                                394

-------
PARADISE STEAM PLANT
General Plant Description
     TVA's Paradise Steam Plant is located in Muhlenburg County, Kentucky,
on the Green River.  The plant consists of three steam-electric generating
units fueled by bituminous coal.  The total plant maximum generator name-
plate rating is 2,558,000 kW.  Units 1 and 2, which began commercial
operation in 1963, are each rated at 704,000 kW.  These two.units are
equipped with 600-foot smokestacks and electrostatic precipitators which
control particulate emissions to 0.20 Ib fly ash/106 Btu.  Unit 3, which
is rated at 1,150,000 kW, began commercial operation in 1970.  This unit
has an 800-foot smokestack and an electrostatic precipitator which is
presently limiting emissions to approximately 0.50 Ib fly ash/106 Btu.
In order to bring Unit 3 into compliance with Kentucky's particulate-
emission standard of 0.11 Ibs fly ash/106 Btu, a new precipitator is
being added in series with the existing precipitator.
     The Paradise Steam Plant is frequently called a "mine mouth" plant,
because it is located in the midst of major coal fields in western Kentucky.
The coal in this area is relatively low quality.  In 1977 the average
analysis was:  Ash 17.3 percent, Sulfur 4.2 percent, and Heating Value
10,500 Btu/lb.  The 4.2 percent sulfur is equivalent to 8.0 Ibs S02/106
Btu.  To comply with the proposed Kentucky emission standard of 3.1 Ibs/
106 Btu (3-hour average) for the plant, a coal-washing facility is being
installed that will reduce the sulfur in coal to an extent that would
limit sulfur dioxide emissions to 5.2 lbs/106 Btu.  In addition, FGD sys-
tems will be installed on Units 1 and 2 that will further reduce S02
emissions from these units to no more than 0.9 Ibs S02/106 Btu  (3-hour
average).  Thus, the weighted average for the three units will  achieve
the S02 standard for the overall plant of 3.1 Ibs S02/106 Btu (3-hour
average).  The scrubbers that will be installed on Units 1 and  2 will
also have particulate-removal sections that will reduce the particulate
emissions to the required 0.11 lbs/106 Btu.
Coal Cleaning Plant
     John T. Boyd Company was commissioned by TVA to determine  the washa-
bility of existing Paradise Coal as well as other midwestern coals most
likely to supply the plant in the future.  The Boyd Company submitted a
report of its investigation which concluded the following:
                                395

-------
     1.    The sulfur content of existing Paradise Coal could be reduced
          sufficiently by conventional washing methods to achieve a plant
          emission rate of 5.2 Ibs S02/106 Bt;u.
     2.    The burning characteristics of the washed coal would be
          compatible with the requirements of the Paradise plant.
     3.    A single coal-washing plant at the Paradise site could be
          designed with sufficient flexibility to adequately process a
          variety of coals from the major midwestern coal seams.
     Engineering design studies were made which indicate that it is
feasible to locate a coal-washing facility at the Paradise plant and
interface the plant with the coal-receiving facilities for the plant.
The type of coal-washing facility will basically consist of a mechanical-
dense medium-gravity separation process capable of removing large portions
of the ash and sulfur from the coal (see Figure 3).   The gravity separa-
tion process utilizes the fact that the mineral impurities (rock, slate,
pyrites) existing in coal typically have higher specific gravities than
the coal itself.  In fact, the mineral pyrite (FeS2), which is the source
for a major portion of the total sulfur in typical west Kentucky coal,
has a specific gravity four times that of pure coal.
     Before the coal and its impurities can be separated, it must be run
through a crusher which physically breaks the impurities away from the
coal.  This is possible because the seam joining the impurity with the
coal is usually weaker than either of the adjoining materials.  The result
is a fracture at the seam.  Once the coal is crushed and the impurities
broken away, the mixture is introduced into a dense liquid medium usually
consisting of finely ground magnetite suspended in water.  The magnetite
increases the apparent specific gravity of the water sufficiently enough
to cause the coal particles to float; yet, it allows the more dense parti-
cles of impurities to sink.  Magnetite is commonly used in the slurry
because of its magnetic properties which ease its recovery for recirculation
     After the coal and refuse have been separated, the majority of the
moisture will be drawn off by centrifuges.  The refuse from the washing
plant will be transported to a disposal area.  Thermal drying of the coal
may be required before conveying it to the steam plant for burning.  The
coal will be transported from the present plant storage and receiving
system to the washing facility and then back to the plant on belt
conveyors.
                                396

-------
vD
--J
             COAL  FEED  3"xO"
                                                           1.60 SP. G.
                                                                                   MIX TANK
               500 TPH
                     SCREEN
        U
                                  400
                                  TPH

                                   H.M.
                                VESSEL
                              3/8"xO"
                                           H.M.
                       n
 H.M.
 11.6
H.M. HEAD TANK
               TO H.M.
              MIX TANK
                 OR
              HEAD TANK
       CLARIFIER
             MAG.
               SEPARATOR
                	fc
      » RESIDUE

*—-rl    1       MAG.
         THICKENER
                L_IH20\/
I                            TO H.M.*-
                          HEAD TANK


                  RESIDUE
                  TO STORAGE
                                 CC
                            STORAGE
                 14% H90
                      C. \
                                                         3/8x28M
                                                         200 TPH
                                   SCREEN
28MxO
50 TPH
                                                  3/8 x28M.CC
           H.M.
       CYCLONE
                                              3"x3/8"CC
                                 10% H20
                                                DRYER
                                                                          H.M.
                                                                                              H20
                                                                          I
                                                               HYDROCONE\
                                              FLOATATION
                                                 -*,
                                                                                                     28MxOCC
                            Figure  3.  Heavy Media Coal Cleaning System

-------
     The washing of the coal will result in the loss of 4 to 15 percent of
the heat content of the coal as refuse.  In addition, about 0.8 percent of
the net plant generation will be required to operate the coal-washing
facilities.
     There are substantial potential benefits other than reduction of
sulfur content that may be derived from washing the coal.  Coal washing
will significantly reduce the amount of ash entering the furnace and
possibly change ash composition.  These changes have the potential of
improving reliability and availability of the plant as well as decreasing
maintenance cost attributable to the poor coal quality.  The coal-washing
costs will be offset by the value of improvements realized.  It is esti-
mated that improvements in operation of the Paradise plant by using the
clean coal will result in a savings of $16 million per year.  Capital cost
of the washing facility and related equipment is estimated to be $130
million.  Amortization of this investment for the remaining life of the
Paradise plant (32 years) will amount to an annual cost of about $17
million.  In addition, the Boyd report estimated that operation and
maintenance costs will amount to about $1.50 per ton of washed coal and
that Btu losses will average approximately 7 percent.  If the Btu losses
are evaluated based on an estimated raw coal replacement cost of $1.00/106
Btu, the annual cost of washing Paradise coal would total about $32 million.
If the savings of $16 million resulting from improvement of plant operation
is realized, the net annual cost of the washing facilities would be
reduced to $16 million.
FGD Systems for Paradise Units  1 and 2
     The flue gas desulfurization equipment  (see Figure 4) that will be
installed on each of the 704,000 kW Paradise Units  1 and 2 will consist of
venturi-absorber systems with limestone slurry as the  scrubbing medium.
Six scrubber trains will be installed  on each unit, any five of which
will accommodate all of the flue gas at full-load operation.  Thus,  1408 mW
of scrubbing capacity, plus two  spare  trains providing 20 percent  redundant
capacity, will be installed.  The spare trains will  improve the operating
reliability of the system and  reduce the frequency  of  excess emissions
resulting  from equipment malfunctions.  Induced  draft  fans  will be used
to draw the gas through the FGD system.
                                 398

-------
                            ELIMINATOR
                                 MIST  I I   ,—REHEATER
u> EFFLUENT
S     TANK
                                                             TO PLENUM
                                                                 —•   o
                   VENTURIT     ABSORBER TANK      ADDITIVE FEED
                     TANK *	1     ICOMPRESSOR        TANK
         LIMESTONE
       STORAGE SILO
          wv
                           -»ILS SURGE
                              HOPPER
                       H20-
                              BALL  MILL
                                                 LIMESTONE
                                               TRANSFER TANK
                                                                                      | THICKENER
                                                                             VACUUM FILTER
TO TRANSFER
  STATION
                            Figure 4.  Paradise FGD System

-------
     Each FGD system will consist of three major components:  limestone
preparation, venturi-absorber units, and the sludge dewatering and
disposal train.
     The limestone preparation for the two units will each consist of
three ball mills, each having a minimum capacity of 35 tph.  This will
provide an excess capacity of 50 percent.  The three wet grinding systems
will operate in a closed circuit and produce a material having a size
consisting of at least 90 percent minus 200 mesh and a slurry concentra-
tion of 40-60 percent solids.  The slurry storage tank for each FGD system
will hold enough for eight hours of full-load operation.  The limestone
slurry transfer from the storage tank to the venturi/absorber scrubber
will use a loop for recirculation with control valves for each scrubber
train.
     The venturi/absorber will be located after the existing electro-
static precipitators; the ID fans will then follow.  Plenums before and
after the venturi/absorber will be used to control the turndown in order
to follow the boiler load.  The turndown will be accomplished by removing
individual venturi/absorber trains from service.  The venturi will have a
slurry distribution system and an adjustable throat to ensure optimum
particulate removal at all boiler loads.  The high velocity areas of the
venturi will be lined with abrasion-resistant materials.
     The absorbers, in addition to spray nozzles for distribution of the
slurry, will have chevron-type demisters to remove entrained slurry and a
reheater to increase the temperature of the gas leaving the mist elimina-
tor to 50°F.  Washers to keep the mist eliminator clean and soot blowers
to remove deposits from the reheater will be provided.
     The material of all components will be selected to give long life and
the most efficient operation.  The scrubber shell, including the venturi
section, will be constructed of 317-L stainless steel to ensure resistance
to corrosion and erosion.  The mist eliminator will be chevron-type units
of two or more stages and will be constructed of either 317-L stainless
steel or fiberglass reinforced plastic.  The reheater will be bare-tube
construction and will be at least four rows of tubes deep:  the first  four
rows will be made of 317-L stainless steel and the remaining rows of carbon
steel.  Isolation dampers will be located at the inlet and  outlet of each
                                 400

-------
scrubber train to provide safe operating conditions under all operating
modes for personnel working on equipment (fans,  venturi/absorber, mist
eliminator,  reheater).
     The sludge dewatering and disposal train will be designed to produce
a filter cake >^ 80 percent solids, without the addition of lime or other
additive, for use as a  physically stable landfill.  To produce the 80-
percent-solids filter cake, forced oxidation will be used in the scrubber
system.   Forced oxidation converts the calcium products formed in the
reaction of  S02 with limestone to gypsum.  Gypsum can be dewatered to a
minimum of 80 percent solids which can be used for landfilling.  Forced
oxidation will be accomplished by injecting compressed air into the
scrubber reaction tanks through a manifold system of piping.
     The dewatering system will consist of two thickeners, one for each
unit, and three 50-percent-capacity vacuum filters for each thickener.
The scrubber slurry will have a minimum of 8 percent solids and will be
pumped to a  thickener (which is the first step in the dewatering system).
The solids settle in the thickener to a concentration of about 40 percent
before being pumped to the vacuum filters, while the clarified liquor from
the thickeners overflows into a trough and is returned to the scrubbers.
The underflow from the thickeners is further dewatered in the vacuum
filters to produce the 80-percent-solids filter cake.
     The landfilling operation will consist of:  transporting the dewatered
cake to the  disposal area; compacting, stacking, and covering the dewatered
cake with soil as the disposal area is filled.  The area will be revege-
tated with plant life compatible with the area and beneficial to wildlife.
All aspects  of the landfilling operation will be designed and operated to
be in compliance with applicable regulations required by the Resource
Conservation and Recovery Act and any other applicable regulations.
Experimental evidence indicates that the use of oxidized-dewatered sludge
will provide a method of overcoming environmental problems.  It is
estimated that approximately 200 acres of land will be needed to dispose
of dewatered sludge for the remaining life of the Paradise Plant.
     The capital cost of the FGD systems for the Paradise Units 1 and 2
is estimated to be $220 million.  The annual cost of the project, including
amortization of the capital over the remaining life of the plant and
annual operating and maintenance costs, is estimated to be $62 million.

                                401

-------
The schedule for installation of the two FGD units is as follows:

September 1, 1979

December 1, 1979
March 1, 1982


June 1, 1982



September 1, 1982
Initiate construction, Unit 1 FGD system.

Initiate construction, Unit 2 FGD system.
Complete construction and initiate
shakedown operation on Unit 1 FGD system.

Complete construction of Unit 2 FGD
system and complete shakedown and
testing of Unit 1 FGD system.

Complete shakedown and testing of Unit 2
FGD system.
                           402

-------
JOHNSONVILLE STEAM PLANT
General Plant Description
     The Johnsonville Steam Plant, located in middle Tennessee, has 10
steam generators with a total capacity of 1,450,000 kW.  Six of the gene-
rators, completed in 1953, have a capacity of 133,000 kW each and are
tangentially fired with pulverized coal.  The other four units, which
started commercial operation in 1958, have a capacity of 162,000 kW and
are rear-wall fired with pulverized coal.
     The annual capacity factor of the total plant in 1977 was about 57
percent and is expected to decrease to 46 percent in 1983 and 25 percent
in 1990, as more efficient plants come on stream.  However, these overall
annual factors do not show the peaking capability required by the
Johsonville plant in the 1980' s.
Fuel Supply
     The primary fuel is bituminous coal from western Kentucky.  The
coal analysis range is 6.1-12.6 percent moisture, 13.5-20.5 percent ash,
1.8-4.4 percent sulfur, and a high heating value of 10,100-11,400 Btu
per pound.
Compliance Requirement and Schedule
     Johnsonville 863 emissions are to be limited to 3.4 lbs/106 Btu (3-hour
average) by December 1, 1982.
     The schedule to meet this requirement is as follows:
     October 1, 1978               Award contracts for compliance coal
                                   (sulfur content equivalent to no more
                                   than 5.0 Ibs SQz per million Btu).
     March 1, 1979                 Execute contract for FGD equipment.
     September 1, 1979             Initiate onsite construction of FGD
                                   equipment.
     September 1, 1982             Complete onsite construction of FGD
                                   equipment at 600 MW of capacity and
                                   begin final shakedown operation.
     December 1, 1982              Complete shakedown operations on the
                                   600 MW of FGD equipment.  Achieve
                                   demonstrate compliance with the 3.4
                                   Ibs of SOz per million emission limit.
                                403

-------
Compliance Program and FGD Description
     It was decided that emissions at the Johnsonville plant would be
limited to 3.4 Ibs S02/106 Btu by using a combination of medium-sulfur
coal and MgO scrubbers.  The MgO scrubbers will be designed for a removal
efficiency of 90 percent.  Approximately 40 percent of the total plant
flue gas will be scrubbed.  For engineering reasons, a new 600-foot
ground-based chimney will also be constructed.
     A regenerative system was dictated because land was not readily
available for disposal of the sludge from lime/limestone systems.  Also,
a portion of the sulfuric acid produced by a regenerative system could
be used in the TVA fertilizer division.  (No decision has been made on
sulfuric acid marketing.)  An MgO scrubbing system was chosen because
the amount of commercial work with this system is greater than with any
other regenerative system.  (The plant arrangement is shown in Figure 5.)
     The flue gas from the 10 steam generators will be fed into a plenum
chamber from which 40 percent of the gas will be drawn through four
scrubber modules (one module is a spare) where 90 percent of the SQz
will be removed.  The remainder of the flue gas which is unscrubbed and
has a temperature of about 300°F will be mixed with the scrubber effluent
gas before the total gas is fed to the single 600-foot stack.  The
concentration of S02 in the stack gas will be equivalent to no more than
3.4 Ibs S02/106 Btu.
     Description of Process.   There are four scrubber modules, each of
which is sized to scrub the flue gas from 200 MW of boiler capacity; one
of the scrubber modules is a spare.  Each scrubber module consists of
two stages:  one for chloride and fly-ash removal and one for absorption
of SC>2.  (A schematic diagram of the system is shown in Figure 6.)
     The flue gas from the supply plenum is drawn into a venturi-type
particulate scrubber where most of the fly ash and all the chlorides are
removed.  The gas is cooled from 300°F to the adiabatic saturation tem-
perature (about 125°F).  A portion of the particulate scrubber recircu-
lating stream (a slurry of fly ash and mother liquor) is diverted to a
pretreatment area, neutralized, and disposed of in the power plant ash
disposal pond.
     Flue gas enters the  cocurrent spray tower after passing through a
set of mist eliminators in the particulate scrubber.  The flue gas is
                                 40A

-------
                                  L
                                    BYPASS DAMPERS
                   STACK
                   STACK
CCE
                                          10
               .SUPPLY PLENUM-
DISCHARGE PLENUM-*
          E
          E
          E
          E
          E
          E
          E
          E
          n-
                                          8
                                             JOHNSONVILLE
                                              UNITS I-IU
                                                             TAIL GAS FROM
                                                           ]r)ACID PLANT
                                                             AND DRYER
                                                         MODULE A
                                                    ^-BYPASS DAMPERS
                                 MODULE B
                                                         MODULE C
                                                         MODULE D (SPARE)
                     DRY ID FANS
      Figure 5.  Johnsonville Plant Arrangement

                                  405

-------
TO
TREATMENT
                RIVER
               WATER
                                 <•—T)FRdM ACID
                                        PLANT
                  PARTICULATE
                   SCRUBBER
                  SURGE TANK
                                       -BYPASS
                                       [DAMPER
              PARTICULATE
              SCRUBBER
                                               ENTRAINMENT
                                                  SEPARATOR
                                                                                  MOTHER LIQUOR
                                                                                FROM CENTRIFUGES
                                                                                               MOTHER
                                                                                               LIQUOR
                                                                                                TANK
                                                                            REGENERATED
                                                                                    MgO
                                                 VIRGIN
                                                  MgO
                                                                       ^COCURRENT
                                                                        ,SCRUBBER
                 SCRUBBER
                 CIRC. TAN
ENTRAINMENT SEPARATOR
CIRCULATION TANK
AND PUMP
SCRUBBER
CIRCULATION
PUMPS
                                                                                                     TO
                                                                                                  CENTRIFI
     Figure 6.  Schematic Diagram of Johnsonville MgO Scrubbers

-------
contacted with a recycling slurry of MgS03,  Mg(HS03)2,  and MgS04 for
the absorption of S02.   The slurry contains  the hydrated crystals of
MgS03 and MgS04 as well as a solution saturated with each of these
components.   A purge stream off the recycled slurry containing approxi-
mately 10 percent solids is diverted to the  centrifuges for separation
of the solids from the  mother liquor.   The centrate is  collected in the
mother liquor tank and  returned to the absorber and particulate-scrubber
recirculating loops.
     Each absorber is designed for a gas velocity of 15 feet per second,
and a liquid-to-gas ratio (L/G) of 30 gallons per 1000  ft3 of gas.
Other design features include the following:

     1.   "Wet-elbow" design for removing a  portion of  the entrained
          slurry in the gas
     2.   Six (6) grids
     3.   Mixing of unscrubbed bypass flue gas with clean gas in the
          discharge plenum for reheat
     The centrifuge cake is dried to remove  waters of hydration in an
oil-fired cocurrent rotary-kiln dryer.  The  dried (MgSOa/MgSC^) crystals
are crushed, then transferred to an inprocess storage silo and before
being fed to the fluid-bed calciner, the entrained solids in the dryer
off gas are enclosed in(a cyclone dust collector.  The  cleaned off gas,
containing some 862 from partial breakdown of the MgS03, is returned to
the scrubber area.  (The MgO regeneration system is shown in Figure 7.)
     The MgS03/MgS04 solids are weight fed to an oil-fired fluid-bed
calciner.  There they are calcined into MgO and S02.  After discharge
from the calciner, the  S02~rich off gas passes through a cyclone dust
collector where approximately 75-80 percent of the MgO dust is removed.
The gas is then cooled  from 1700°F to about 450°F by a series of heat
exchangers.   Next the cooled gas enters a baghouse, where the remaining
MgO is removed.  The S02 stream from the baghouse forms the feedstock
for a conventional contact sulfuric acid plant.  Tail gas from the acid
plant is recycled to the S02 absorbers.  All of the removed MgO is
stored before recycling to the absorption area.
     The principal objective in selecting the size of the acid plant and
the storage capacity for MgS03 is minimal risk of restricting power
plant operations.  A double- or single-acid plant module with a total
production capability of 350 tons per day is being investigated.  Facilities

                                407

-------
            FROM
o
CX5
                                                    CYCLONE
                                    FLUID BED
                                     CALCINER
                                                           S02, MgO DUST
                                                                     HEAT
                                                                 EXCHANGERS
                                                             MgO SOLIDS
                  S02 TO SCRUBBERS
      MOTHER
      LIQUOR-*-
      STORAGE
CAKE CONTAINING
MgS03
                               ROTARY
                                 DRYER
           FUEL  AND AIR
                  MgS03  SOLID
                                        ^STORAGE
                                                                                                 S02
                                                                                                ACID  PLANT
                                                                                          MgO SOLIDS
                                                                                      MgO STORAGE
TO ABSORBER
RECYCLE TANK
                                        Figure 7.  MgO Regeneration System

-------
for storing MgSOs will be provided.   Excess MgSOg storage equivalent to
several-days'  power plant operation appears quite adequate for minimizing
the risk of restricting power plant operation.   At this date,  no decision
concerning quantity has been made.
     Estimated capital cost in 1982 dollars of an MgO scrubber system is
$185,000,000.   The cost includes the following:
     1.   Sulfuric acid plant and acid handling facilities
     2.   MgS03 and MgO storage
     3.   Five (5) scrubber modules
     4.   600-foot stack
     5.   Drying and regeneration section
     6.   Ductwork tie-in for all ten units and scrubber ductwork
     7 -   Site preparation*
The total annual cost including amortization is $36,000,000.
     The annual operation and maintenance costs of the FGD system and
acid plant are estimated to be $23,000,000.  If it is assumed that $24
per ton will be received for the acid and about 256 tons per day will be
produced, the operating and maintenance costs will then be reduced
$2,240,000 to a net value of $20,760,000 (all I.a82 dollars).
     The estimated operation costs include a labor requirement of eight
men per shift for the FGD system and acid plant.  The 8-man shift does
not include the maintenance manpower required.
*When this paper was being written, TVA tentatively decided on a 4-module
 system.  Although the technical changes have been incorporated in the
 paper, revised cost estimates  are not available.
                                409

-------
CUMBERLAND STEAM PLANT
     The Cumberland Steam Plant is located approximately three miles
northwest of Cumberland City, Tennessee, on the Cumberland River.  It has
two bituminous coal-fired steam generators, each with a capacity of
1,300,000 kW.  The plant was initially placed into commercial operation
in 1973 and is TVA's newest and largest coal-fired generating facility.
The flue gas from the Cumberland plant is dispersed through two 1000-foot
concrete stacks.
     TVA is installing a heavy-media coal-cleaning system (such as the one
shown in Figure 3) which will be capable of reducing SQ% emissions from
Cumberland to about 5.0 lbs/106 Btu.  In addition, in a proposed settle-
ment of litigation involving the Environmental Protection Agency, the
States of Alabama and Kentucky, and a number of private citizen's groups,
TVA has agreed to install at Cumberland a 600-MW FGD system with 90 per-
cent removal (or the equivalent) of S02 to reduce plant emissions to 4.2
lbs/106 Btu.  The proposed settlement provides that the deadline for
installing these scrubbers may be extended by the parties if other than
conventional lime/limestone scrubbers are installed.
     To further develop advanced scrubber technologies and based upon TVA's
ongoing program for the development and demonstration of these technologies,
TVA is considering the feasibility of installing an alternative technology
for the 600 MW's at Cumberland.
     The selection of the system to install at Cumberland includes the
evaluation of data obtained by testing of advanced scrubber concepts at
three prototype levels (1 MW, 10 MW, and 20 MW) to evaluate process
operation, performance of equipment and materials, and to obtain the
optimum operating conditions for the processes.  The objective of these
tests and evaluations is to provide the most promising advanced scrubber
system suitable for full-scale demonstration on the Cumberland plant.
The following processes are being evaluated:
          Process                                 Absorbent
1.  Cocurrent Scrubber
2.  Dowa-Double Alkali
3.  Thoroughbred 121-Gypsum
4.  Absorption-Steam Stripping/Resox
5.  Aqueous Carbonate
Any
Basic Aluminum Sulfate
Sulfuric Acid-Limestone
Citrate
Sodium Carbonate
                                 410

-------
     Primarily,  these systems were chosen for evaluation because they do
not produce a disposal byproduct such as the sludge from lime or limestone
scrubbers.   It should be noted that TVA has a comprehensive study now
underway to develop the most environmentally acceptable method for disposal
of the sludges produced by the limestone scrubbers now in operation at the
Widows Creek plant and those to be installed at the Paradise plant.  However,
it is thought that the production of a useful byproduct is more desirable
than the disposal product.
     In addition to producing a useful byproduct, the five systems being
studied have the potential for achieving high S02 removal efficiencies
when compared to the limestone systems.  When more efficient scrubber sys-
tems are developed and incorporated by EPA in its New Source Performance
Standards,  the resulting reduction in emissions will allow more room for
growth of other sources.
     In summary, we believe that the following alternatives offer not only
a sound approach for reducing emissions at Cumberland but also for devel-
oping technology that will benefit the national interest as well as the TVA
system.
Cocurrent Scrubber
     In the cocurrent configuration, flue gas enters the scrubber from
the top of the tower, where the liquid absorbent is also sprayed, and both
flow cocurrently to the base of the scrubber.  A majority of the liquid
is removed from the gas by a wet-elbow scrubber bottom which forces the
gas to make a turn of 180° after impacting the entrained liquid onto the
surface of a pool of liquid maintained at the base of the scrubber.
Thereafter, the gas passes through a chevron-type mist eliminator located
in a horizontal duct with a separate water wash cycle.  The gas is then
reheated and discharged to the stack.
     The cocurrent scrubber has several advantages over the conventional
countercurrent scrubbers.  Because of the significantly higher gas veloci-
ties permitted,  the actual size of the scrubber will be smaller.  The
equipment configuration is more compatible with most power-plant duct and
fan arrangements.  The mist eliminator and reheater can be located near
ground level and thus will be more easily maintained.  Other potential
advantages  are better liquid and gas distribution and higher SC-2 removal
efficiencies.

                                411

-------
     Tests of the cocurrent scrubber have been conducted at both the 1-MW
and 10-MW levels with excellent results.   The concept was found to be very
flexible—capable of using almost any absorbent and very efficient
(> 90 percent S02 removal) over a wide range of gas velocities (18-27 feet
per second).   The cocurrent scrubber is planned for the Johnsonville MgO
system at a gas velocity of 15 ft/sec.  If chosen for Cumberland, the
cocurrent scrubber will most probably be designed for a gas velocity near
the upper limit of the range described above.
Dowa Double-Alkali
     In the Dowa process, SC>2 is absorbed in a clear solution of basic
aluminum sulfate.  The basic solution is oxidized by air in a separate
tower and then neutralized with finely ground limestone to precipitate
gypsum and to regenerate the basic aluminum sulfate solution.  The process
has been developed and tested on a 40-MW oil-fired boiler in Japan.  How-
ever, it has not been tested on a coal-fired boiler.
     Advantages of the Dowa process over the conventional lime/limestone
scrubber are higher S02 removals and reduced scaling and plugging, possible
because a solution rather than a slurry is used for scrubbing.  The Dowa
system produces a gypsum which can be stockpiled, used for landfill, or
sold for making wallboard.
     TVA plans to test the Dowa process at the 10-MW level with flue gas
from a coal-fired unit.  The prototype scrubber will be installed at the
TVA Shawnee Test Facility.  The project will be jointly funded by TVA,
EPRI, and Universal Oil Products (licensee of the Dowa technology).
Chiyoda Thoroughbred 121 Process
     In the Thoroughbred 121 process, the flue gas is quenched with water.
It is then introduced into Chiyoda's patented Jet Bubbling Reactor where
the flue gas is sparged into the absorbent through an array of vertical
spargers, generating a jet bubbling (froth) layer.  S02 is absorbed in the
jet bubbling layer and subsequently oxidized to sulfate.  The cleaned flue
gas is discharged through a mist eliminator and out the stack.
     Limestone slurry is pumped directly to the jet bubbling  reactor where
reaction takes place to produce gypsum.  The same inherent advantages
of gypsum production as discussed previously for Dowa are realized.
                                412

-------
     Tests of the Thoroughbred 121 are being conducted on a 20-MW coal-
fired unit at Gulf Power Company's Scholz Steam Plant, Sneads, Florida,
to evaluate performance, reliability, operability, and the cost and energy
effectiveness of the process.  This demonstration is a joint effort of
EPRI, Southern Company Services, and Chiyoda.  Testing began in August
1978.  No reports have been published to date, but verbal reports from
those involved indicate that excellent results are being obtained.
     TVA will continue to follow the testing on the 20-MW plant at the
Scholz steam plant.  If the results continue to be favorable, TVA may
propose to make a demonstration run under conditions more applicable to
the Cumberland Steam Plant before deciding whether or not to proceed
with a 600-MW unit at Cumberland.
Atomic International Aqueous Carbonate
     In the Atomic International Aqueous Carbonate process, SOz in the
flue gas is absorbed by sodium carbonate in a spray drier.  The dried
solids are separated from the flue gas in either an ESP or baghouse.  The
spent-absorber solids are discharged to a molten salt reactor where reduc-
tion of sulfite is accomplished.  The "green liquor" material is then
filtrated to remove insolubles and acidified in the carbonation step to
produce sodium carbonate and H2S.  The sodium carbonate is returned to the
scrubber and the I^S is sent to a Glaus plant for the production of
elemental sulfur.
     The advantages of  this  system are:  (1) relatively high S02 removal;
(2) spray drying requires no reheat; and (3) production of elemental sulfur.
     If this alternative is  selected, TVA will enter  into a contract with
Atomics International for supplying a 100-MW system at the Cumberland
plant.  Consequently, it would be necessary to use this system in combina-
tion with other FGD processes to reduce Cumberland emissions to the 4.2-
pound level.  The  100-MW size is necessary because of the advanced nature
of the process, and the lack of opportunity for testing on a small scale
makes it imprudent to commit to the full 600-MW level.
Absorption Steam Stripping/Resox Process
     This process  is a  regenerable FGD system capable of producing elemen-
tal sulfur without a reducing gas or alternate byproduct.  The process
combines aqueous absorption with steam stripping.
                                413

-------
     The flue gas first goes to a venturi for particulate removal.  From
the venturi section it enters the bottom of the absorber and flows upward
where it is contacted with a countercurrent flow of sodium citrate liquor
to remove S02.  After leaving the absorption section and mist eliminator,
the gas is reheated and discharged to the stack.
     The liquor containing the S02 is then steam stripped to remove the
S02 and to regenerate the citrate liquor for reuse in the absorber.  The
S02 is then sent to the RESOX reactor where the S02 reacts with coal to
form elemental sulfur.
     The RESOX system is being developed by EPRI on a 42-MW scale at the
Killerman Power Station of Steag A. G. in Lunen, Federal Republic of
Germany.
     The advantages of this process are:  (1) high S02 removal; (2) the
absorbent is reused; and (3) elemental sulfur is produced.
     EPRI, TVA, and Flakt, Inc., are cooperating in testing the absorp-
tion steam-stripping process at TVA's 1-MW Colbert pilot plant, which will
include short-term  (factorial) and longer term operational testing to
quantify all key process parameters.  However, testing of this system
may not be underway by the time a decision must be made for Cumberland.
Therefore, evaluation of this system must be made on data from smaller
scale studies.  This again is an advanced process with little or no data
except that obtained on the 1-MW test unit; therefore, it is thought that
it would be imprudent to commit to a 600-MW unit.
     The schedule for installation of the FGD system for Cumberland is
dependent upon the  route chosen for the type of system.  By March 1, 1979,
TVA must submit a notification of the type of FGD equipment (conventional
lime/limestone or alternative) selected.  If the conventional route is
chosen, an invitation to bid must be extended by April 1, 1979, with award
of contract by October 1, 1979.  Onsite construction will be completed by
October 1, 1982, and shakedown will be  completed by December 31,  1982.   If
the alternative route is chosen, the FGD system to be  installed will be
selected by November 1, 1979, and an appropriate construction schedule
developed.
                                 414

-------
                                 SUMMARY


     TVA has developed a program that will bring its coal-fired plants

into compliance with State and Federal sulfur dioxide emission require-

ments.   Eight of the twelve plants will meet S02 emission limits by

selection of medium- and low-sulfur coals, but four plants will require
coal washing and/or FGD systems to meet S02 standards.   The plans for

the four plants may be summarized as follows:
  Plant and Size

Widows Creek Plant
8 units, 1978 MW
Paradise
3 units, 2558 MW
Johnsonville
10 units, 1450 MW
Cumberland
2 units, 1300 MW
 Emission
   Limit

1.2 Ibs S02
per 10s Btu
(24 hour
  average)

3.1 Ibs S02
per 106 Btu
(3 hour average)

3.4 Ibs S02
per 106 Btu
(3 hour average)
4.2 Ibs S02
per 106 Btu
(24 hour
  average
     Compliance Program

a) Burn low-S coal on 6 units
   with total of 853 MW.
b) Install limestone scrubber
   on 2 units totaling 1125 MW.

a) Install coal-cleaning plant.
b) Install limestone scrubber
   on Units 1 and 2, 704 MW each.

a) Manifold all 10 units to a
   single stack.
b) Burn medium-sulfur coal.
c) Install 600 MW of MgO scrubbers
   and convert S02 to sulfuric acid.

a) Burn washed coal with sulfur
   equivalent to 5.0 Ibs S02/106 Btu.
b) Install 600 MW of alternative
   scrubbers with 90% or more S02
   removal efficiency or its
   equivalent (see text).
                                415

-------
               Start
           Construction
                                                           Initial
                                                                  2
                                                         Operation
Commercial
Operation
                  £
  Compliance Dates
 Initial   Compliance
Operation	Test
Johnsonville
MgO scrubber units 1-10
stack (600 ft) units 1-10
Widows Creek
precipitators units 1-6
limestone scrubber unit 8
limestone scrubber unit 7 (contract)
Paradise
coal washing facilities (contract)
limestone scrubber units 1-2 (contract)
precipitator unit 3 (contract)
Cumberland
precipitators units 1-2
scrubbers
9/79
9/07/788
12/10/75
2/20/73
9/01/78
6/01/77f
7/79
6/01/78n
3/79
7/01/79
12/82
NA
12/24/77
4/30/77e
2/81
12/01/80
4/82
6/80
12/81
7/01/82
3/83
2/15/78d
5/81
6/01/81
7/82
9/80
2/82
12/01/82
12/82
3/01/78
5/01/77
3/81
12/80
4/82
9/80
12/81
7/82
3/83
9/81
6/81
7/82
11/80
2/82
12/31/82
                                 (This date generally corresponds to the

                                                                     "Corap
a  Initial operation or energization of last unit in project.
   "Comp. Const" date on current TVA key date schedules.)
b  Completion of startup testing of last unit in project.  (This date generally corresponds to the
   Test" date on current TVA key date schedules.)
c  Compliance with EPA standards.
d  Completed demonstration test.
e  First gas was scrubbed 5/16/77; slurry circulation through train "A" began 4/30/77.
f  TVA site preparation work began 6/77; PA was approved 10/27/77; Contractor reported to the site 1/5/78
   and began construction 3/10/78.
g  TVA starred conduit relocation 9/7/78; TVA to start foundation preparation 3/79; Contractor to start
   work 7/79.
h  Started parking lot work 6/1/78; Contractor started work 7/10/78.
i  This schedule is for the conventional lime/limestone scrubbing option.
TABLE 1.  SCHEDULE OF COMPLIANCE METHODS

-------
                              REFERENCES
1.    Tennessee Valley Authority, Division  of Power, McKinney, B.  G.,
     Little, A. F.,  and Hudson, J. A.,  "The TVA Widows  Creek Limestone
     Scrubbing Facility,  Part  I," New Orleans:  EPA Flue  Gas Desulfuri-
     zation Symposium, May  14-17, 1973.

2.    Tennessee Valley Authority, Division  of Power, Wells, W. L., Muirhead,
     W.  B., and Buckner,  J.  H., "TVA's  Experience with  Limestone  Scrubbers
     at  the 550 MW Widows Creek Unit 8," Chicago:  American Power Conference,
     April 24-26,  1978.
                         ADDITIONAL INFORMATION
1.    Koehler,  George,  Magnesia Scrubbing Applied to  a  Coal-Fired  Power
     Plant,  EPA-600/7-77-018,  March 1977.

2.    Koehler,  George and James A.  Burns, The Magnesia  Scrubbing Process
     as Applied to an Oil-Fired Power Plant, EPA-600/2-75-057, October  1975,

3.    Sommerer, Diane K., Magnesia  FGD Process Testing  on a  Coal-Fired
     Power Plant,  EPA-600/2-77-165, August 1977.

4.    McGlamery, G. G.; Torstrick,  R.  L.; Simpson,  J. P.;  and  Phillips,  Jr.,
     J.F.:  Conceptual Design and  Cost Study; Sulfur Oxide  Removal  From
     Power Plant Stack Gas,  Magnesia Scrubbing—Regeneration: Production
     of Concentrated Sulfuric Acid, EPA-R2-73-244, May 1973.

5.    Lowell, P. S.; Meserole,  F. B.;  and Parsons,  T. B.;  Precipitation
     Chemistry of Magnesium  Sulfite Hydrates in Magnesium Oxide Scrubbing,
     EPA-600/7-77-109, September 1977.
                                417

-------
                                                    Paper No.  4D
                                                    EPA's FGD  Symposium
                                                    March 1979
                     S02 AND NOx REMOVAL TECHNOLOGY IN JAPAN
                                         Jumpei Ando
                                         Faculty of Science and Engineering
                                         Chuo University
                                         Kasuga, Bunkyo-ku, Tokyo 112
ABSTRACT
     The total operational FGD capacity in Japan has reached 33,000 MW equivalent
with about 50 plants for utility boilers and numerous smaller ones for other
flue gas sources.  Major FGD plants have been operated with over 97% operability
removing over 90% of S0?.  Through FGD, hydrodesulfurization of heavy oil, and
import of low-sulfur fuels, ambient SO  concentrations have been reduced remarkably
since 1967 — to levels low enough to meet the stringent air quality standard of
0.04 ppm in daily average or 0.02 ppm in yearly average.
     The increase in desulfurization plants has slowed down due partly to the
successful reduction of ambient SO- concentrations, and partly to oversupply of
by-products, and to the recent economic depression.  The growing usage of coal for
utility boilers, however, will necessitate more FGD plants.
    Recent efforts for air pollution control have been concentrated on NOx
abatement.  In addition to combustion modification for numerous flue gas sources,
over 60 commercial plants for selective catalytic reduction of NOx in flue gases
have been put into operation.  Selective noncatalytic reduction (thermal de-NOx)
has also been applied commercially or for large scale tests.
     The status of technologies, problems, and economics of S0« and NOx removal in
Japan will be described.

                                    418

-------
                     SO  AND NOx REMOVAL TECHNOLOGY IN JAPAN

1  FGD PROCESSES  AND PLANTS
1.1  Major Processes
     Table 1 lists major constructors of FGD plants and numbers and capacities of
plants operational at the end of 1978.  The plants totaled more than 500 and their
combined capacity reached 88,000,000 Nm /hr (equivalent to 29,000 MW).   About half
of the capacity is accounted for by utility boilers (mostly oil-fired)  and the
rest by industrial boilers, iron-ore sintering machines, nonferrous metal industry,
sulfuric acid plants, etc.
     About 50% of the plants, in terms of capacity, use the wet lime/limestone
process to by-produce gypsum, 16% the indirect lime/limestone process (double
alkali type) to by-produce gypsum, 13% the regenerable process to by-produce
sulfuric acid, ammonium sulfate and elemental sulfur,  and 24% sodium scrubbing to
                                                                               3
by-produce sodium sulfite or sulfate.  The average plant capacity is 443,000 Nm /hr
                                      3
for the wet lime/limestone, 279,000 Nm /hr for the indirect lime/limestone, 369,000
  3                                                3
Nm /hr for the regenerable processes, and 59,600 Nm /hr for the sodium scrubbing
process.  About 80% of the sodium scrubbing plants by-produce sodium sulfite for
paper mills and the rest oxidize the sulfite by air bubbling to sulfate, which is
either used in the glass industry or purged in wastewater.
     In addition to the 335 sodium scrubbing plants listed in Table 1,  there are
about 500 smaller ones operated commercially with an average capacity of about
20,000 Nm3/hr.
1.2  Status of FGD Plants by Power Companies
     Table 2 lists power companies and their capacities of steam power generation
and FGD.  The nine major power companies (Nos. 1 to 9  in the table) have produced
about 70% of total steam power using mainly oil with some LNG and a little coal.
Electric Power Development Co. (EPDC, No. 10 in the table) which was established
by the nine major companies and the Central Government has been the major consumer
of coal for power generation.  Other power suppliers have relatively small
capacities, burning mainly oil.
     Among the major power companies, Tokyo Electric,  Kansai Electric and Chubu
Electric have the largest power generation capacities  (A) and relatively small FGD
capacities (B), with a B/A ratio of only 1-8%.  Those  companies prefer clean fuel
such as LNG and low-sulfur oil to FGD, because they have power plants near large

                                       419

-------
     Table 1  NUMBERS AND CAPACITIES  (1,000 Nm  /hr)  OF FGD PLANTS  BY MAJOR  CONSTRUCTORS  (OPERATIONAL AT END 1978)
N>
O
Plant constructor
Mitsubishi Heavy Industries (MHI)
Ishikawajima H.I. (IHI)
Hitachi, Ltd.
Mitsubishi Kakoki (MKK)
Kawasaki Heavy Industries
Tsukishima Kikai (TSK)
Chiyoda Chemical Eng. & Construe.
Oji Koei
Sumitomo Metal - Fujikasui
Kurabo Engineering
Mitsui Miike-Chemico
Ebara Manufacturing
Kobe Steel
Nippon Kokan (NKK)
Kureha Chemical
Showa Denko
Gadelius
Sumitomo (SCEC)-Wellman
Nippon Steel
Mitsui Metal Engineering
Dowa Engineering
JGC
Ube Industries
Niigata Engineering
Mitsui Engineering
Total
Wet lime
limestone
34
17
13
2
4
1


7

4

6
3




2
4

1



98
(19,020)
( 4,445)
( 6,940)*
( 256)
( 756)



( 3,954)

( 2,744)

( 2,475)
( 245)




( 1,200)
( 1,006)

( 330)



(43,371)
Indirect lime 2 4*
limestone , ,

6
4
15


5

11

1






8


1

51

( 5,
(
( 4,


(

( 1,

(






(


(

(13,

450)
398)
585)


603)

914)

150)






666)


185)

951)
2
13

1



1
1


2



6

2

1
2


31
(
( 6,

(



(
(


( 1,



( 1,

(

(
(


(11,
S
SO.
4
590)
478)**

88)



18)
500)


990)



288)

130)

125)
220)


427)
Na2S°3
3
79
15
41
7
40

57
6
106

10

6
8
5
8







1
335
( 292)
( 4,351)
( 603)
( 913)
( 256)
( 4,042)

( 4,280)
( 270)
( 3,754.)

( 1,167)

( 62)
( 1,431)
( 1,372)
( 1,291)







( 160)
(19,961)
Total
37
96
30
56
17
45
15
57
13
112
5
21
6
12
8
5
8
6
2
6
8
2
2
1
1
515
(19,312)
( 8,796)
( 8,133)
( 7,643)
( 6,380)
( 4,608)
( 4,585)
( 4,280)
( 4,224)
( 4,372)
( 3,244)
( 3,081)
( 2,475)
( 2,447)
( 1,431)
( 1,372)
( 1,291)
( 1,288)
( 1,200)
( 1,136)
( 666)
( 455)
( 220)
( 185)
( 160)
(88,710)
                   * Babcock - Hitachi
** Wellman - MKK

-------
Table 2  CAPACITIES OF STEAM POWER GENERATION AND FGD OF POWER COMPANIES
                           Power generation  (MW)
FGD (MW)

No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

Power
company
Hokkaido
Tohoku
Tokyo
Chubu
Hokuriku
Kansai
Chugoku
Shikoku
Kyushu
EPDC
Niigata
Showa
Toyama
Mizushima
Sumitomo
Sakata
Fukui
Others
Total

Existing
1,270
3,925
19,167
9,933
1,412
10,672
3,777
2,687
4,500
1,430
350
550
750
462
368
0
0
5,512
66,775
Under
cons true t ion*
1,225
1,200
4,400
3,800
1,000
1,200
1,800
450
2,700
1,000
350
0
0
0
250
700
250
375
20,700

Total (A)
2,495
5,125
23,567
13,733
2,412
11,872
5,777
3,137
7,200
2,430
700
550
750
462
618
700
250
5,887
87,475

Existing
0
900
283
970
600
930
1,350
900
1,626
1,280
175
400
250
156
156
700
0
• 0
10,676
Under
construction*
525
0
0
0
500
0
700
0
0
1,000
175
0
0
0
0
0
250
0
3,150

Total (B)
525
900
0
970
1,100
930
2,050
900
1,626
2,280
350
400
250
156
156
700
250
0
13,826

B/A (%)
21.0
17.6
1.2
7.1
45.6
7.8
36.8
12.5
22.6
93.8
50.0
72.7
33.3
33.8
25.2
100.0
100.0
0.0
15.8
              * Including those decided to be constructed.

-------
cities such as Tokyo, Osaka and Nagoya and have anticipated the SOx and NOx
regulations there to become too stringent to be met by FGD and combustion
modification when high-sulfur fuels are used.  On the other hand, Hokuriku Electric,
Chugoku Electric, EPDC and some of the smaller companies, with power plants
relatively distant from large cities, have larger B/A ratios.  FGD plants of power
companies are listed in Tables 3 and 4.  All of the plants by-produce gypsum
except the three that by-produce sulfuric acid.
     The recent oversupply of gypsum and other FGD by-products and the relatively
low cost of low-sulfur fuels due to economic depression has discouraged
construction of additional FGD plants.  Four plants for coal-fired utility boilers,
175 MW (existing), 250 MW, 500 MW and 500 MW (new), will be completed between 1979
and 1981, all using the limestone-gypsum process.
2  OPERATION OF FGD PLANTS
2.1  Wet Lime/Limestone Process
     Table 5 shows operation data of major wet lime/limestone process plants.
Those plants by-produce salable gypsum except the Omuta plant, Mitsui Aluminum,
which by-produces a throw-away sludge.  For the production of gypsum, a calcium
sulfite slurry discharged from a scrubber at a pH of 6-7 is treated to lower the
pH to about 4 and then is air-oxidized.  For the pH adjustment, an additional
scrubber is used at the Owase plant, Chubu Electric, and the Takasago plant, EPDC,
while sulfuric acid is used at other plants.
     S02 removal efficiency ranges from 90 to 98%, and power required for a total
FGD system from 2.0 to 3.5% of the power generated.  The power requirement is
larger for a scrubber with a venturi because of the larger pressure drop of the
gas in the scrubber to attain high removal efficiencies for S0? and dust.
     Wastewater is purged at a rate of 3-30 tons/hr or 8-60 kg/MWhr, primarily to
prevent the accumulation of chlorine in the scrubber liquor because chlorine
increases corrosion.
     The Omuta plant, Mitsui Aluminum, has been operated with a low oxidation ratio
preventing the formation of gypsum while in other plants a considerable amount of
gypsum crystals are added as seeds to,a circulating slurry.  Scaling can be
minimized in either way.  None of the plants, except the Tamashima plant, Chugoku
Electric, has a stand-by scrubber.
     Some of the plants encountered problems at start-up but most of the problems
were solved in a few months.  All of the plants have since attained an operability
better than 97%.  Operability means an FGD plant's operating hours per cent  of  the
                                       422

-------
                     Table 3    FGD PLANTS OF POWER COMPANIES  (I) (FOR OIL-FIRED BOILERS)
Power company

  Tohoku
     II

     II
  Tokyo
     it
  Chubu
Hokuriku
Kansai
     ii
Chugoku
     ii
Boiler
Power station
Shins endai
Hachinohe
Niigata
Niigata H.
Akita
Kashima
Yokosuka
Nishinagoya
Owase
ii
Toyama
Fukui
Nanao
Sakai
Amagasaki
ii
Osaka
ii
it
Kainan
Mizushima
Tamashima
it
Shimonoseki
No.
2
4
4
1
3
3
1
1
1
2
1
1
1
8
2
1
3
2
4
4
2
3
2
2
MW
600
250
250
600
350
600
265
220
375
375
500
350
500
250
156
156
156
156
156
600
156
500
350
400
FGD
MW
150
125
125
150
350
150
133
220
375
375
250
350
500
63
i 35
156
156
156
156
150
100
500
350
400
Process developer   Absorbent, precipitant
Kureha-Kawasaki
Mitsubishi H.I.
Wellman-MKK
Mitsubishi H.I.
Kur eha-Kawasaki
Hitachi-Tokyo
Mitsubishi H.I.
Wellman-MKK
Mitsubishi H.I.
       ii
Chiyoda
   M
Not decided
Sumitomo H.I.
Mitsubishi H.I.
                                                Babcock-Hitachi
Mitsubishi H.I.
Babcock-Hitachi
                                                Mitsubishi H.I.
Na2SO~, CaC03
CaO
Na2S03
                    By-product

                      Gypsum
      , CaC03
Carbon, CaC03
Na2S03
CaO
 ii
H2S04,
Carbon
CaO
CaC03
 n
 it
CaO
CaC03

 it
 n
                                                                                              Gypsum
                                                                                                n
                      H2S04
                      Gypsum
                      H2S04
                      Gypsum
                        it
                        it
 Year of
completion

   1974
   1974
   1976
   1976
   1977
   1972
   1974
   1973
   1976
   1976
   1974
   1975
   1978
   1972
   1973
   1975
   1976
   1975
   1975
   1976
   1974
   1974
   1975
   1976
   1976

-------
                       Table 4     FGD PLANTS  OF  POWER  COMPANIES  (II)
Boiler
Power company
Shikoku
ii
Kyushu
ii
ii
n
11
it
it
EPDC
it
n
n
n
Niigata
Showa
it
Toyama
Mizushima
Sumitomo
Sakata
n
Fukui
Power station
Anan
Sakaide
Karita
Karatsu
it
Ainoura
it
Buzen
it
Takasago

Isogo

Takehara
Niigata
Ichihara
ti
Toyama
Mizushima
Niihama
Sakata
n
Fukui
No.
3
3
2
2
3
1
2
1
2
1
2
1
2
1
1
1
5
1
5
3
1
2
1
MW
450
450
375
375
500
375
500
500
500
250*
250*
265*
265*
250*
350
150
250
250
156
156
350
350
250
FGD
MW
450
450
188
188
250
250
250
250
250
250
250
265
265
250
175
150
250
250
156
156
350
350
250

Process developer
Kureha-Kawasaki
11
Mitsubishi H.I.
ii
n
"
ii
Kur eha -Kawasaki
"
Mitsui-Chemico
ii
Chemico-IHI
ii
Babcock-Hitachi
MHI
Showa Denko
Babcock-Hitachi
Chiyoda
Mitsubishi H.I.
IHI
Mitsubishi H.I.
n
Not decided

Absorbent, ^precipitant
N32S03, CaC03
"
CaO
CaC03
"
11
"
N32S03, CaC03
' ii
CaC03
it
ti
ti
n
"
Na2S03, CaC03
CaCO?
H2S04, CaC03
CaO
CaC03
CaC03
n
n

By-product
Gypsum
"
11
ii
ti
it
ii
ti
"
"
it
ii
ii
ii
it
11
n
n
11
11
ii
n
n
Year of
completion
1975
1975
1974
1976
1976
1976
1976
1977
1978
1975
1976
1976
1976
1977
1975
1973
1976
1975
1975
1975
1976
1977
1979
*  Coal-fired boilers.  Others are for oil-fired boilers.

-------
                                Table 5   OPERATION DATA OF MAJOR LIME/LIMESTONE PROCESS PLANTS
     Process
Lime scrubbing
Limestone scrubbing
->
N3
Process developer

Plant owner

Plant site
Fuel
FGD capacity (MW)
FGD capacity (1,000 Nm3/hr)
Inlet SO 2 (ppm)
Inlet dust (mg/Nm3)
Inlet gas temperature (°C)
CaO/S02 stoichiometry
Mitsubishi
H.I.
Chubu
Electric
Owase
Oil
375
1,200
1,600
15
150
1.0
Number of scrubbers in parallel 2
Prescrubber (first scrubber)
Type
L/G (liters/Nm3)
Scrubber (second scrubber)
Type
Slurry pH
Slurry concentration (%)
L/G (liters/Nm3)
Gas velocity (m/sec)
Outlet S02 (ppm)
Outlet dust (mg/Nm3)
S02 removal efficiency (%)
Mist eliminator type
Prescrubber
Pressure Scrubber
.. °^ ' Mist eliminator
2 ' Total system
Wastewater purged (t/hr)
Energy requirement (design)
Pump (kW)
Fan (kW)
Total FGD system (kW)
Per cent of power generated
Operability (%)

Spray
2

Packed
6.5-7
10
7
3.4
120
8
93
CE«0
30
150
25
375
3



7,500
2.0
98, 99
Chemico-
Mitsui
Mitsui
Aluminum
Omutaa)
Coal
156
510
2,300
630
135
1.05
2

Venturi
5-8

Venturi
7.5
5
5-8

220
40
90
Chevron
] 200
30



1,320
2,000
3,867
2.5
100
a) Throw-away process. Others by-produce gypsum.
c) Stand-by. d) Perforated
plate.
e) Chevron
Mitsubishi
H.I.
Kyushu
Electric
Karatsu
Oil
250
730
530
25
150

1

Spray
2

Packed
6.2

12
3.0
50
6
90
Chevron
60
35
65
185




5,070
2.0
99.7
Bab cock-
Hitachi
Chugoku
Electric
Tamashima
Oil
500
1,480
1,460
40
140

3+lc>

Venturi
10
-»
ppd)
6.6
12.5
10

60

96
Pwff>
225
505
25
1,080
4

4,100
11,500
17,600
3.5
97.4
b) Iron-ore sintering
and Euroform
Babcock-
Hitachi
Electric
Power D.C.
Takehara
Coal
250
850
1,550
600
150
1.05
1

Venturi
2.4
• V
ppd)
6-6.5
9
7

100
30
98
Pwff)
230
375
10
950
15

1,400
5,500
8,200
3.3
97.4
Mitsui-
Chemico
Electric
Power D.C.
Takasago
Coal
250
850
1,500
100
140
1.0
1

Venturi
6

Venturi
6.2
5-6
6

100
50
93
Chevron
150
150
25
525
10

2,800
4,200
8,000
3.2
97. 98
IHI-
Chemico
Electric
Power D.C.
I so go
Coal
265
900
450
1,500
170
1.0-1.05
1

Venturi
7

Venturi
5-6
7
7
3.0
10
50
93
Chevron
150
150
50
600
15

2,000
4,800
Sumitomo-
Fuj ikasui
Sumitomo
Metal
Kashima
Coke
(630)b>
2,000
400-600
100-200
150
1.05
2
.»
pp
7
A\
ppd)
6.7.
7-8
8
4.4
30
20
93-95
Impinger
180
130
30
530
30

4,400
8,170
7,800 13,230
2.9
100
2.1
100
plant. Others are utility boilers.
f) Pipe with fin.

-------
desired operating hours of the gas source in a year, as will be desctibed in detail
in Section 2.4.
     Figure 1 shows the relationship of the operability and inlet SO  concentration
of major wet lime/limestone process plants.  The operability is generally lower with
higher inlet SO- concentrations,  indicating a greater scaling tendency.  The
difference in operability between the plants for oil-fired and coal-fired boilers is
little and may not be significant.  Two plants for industrial boilers have 100%
operability in spite of high inlet SCK concentrations,  possibly because of easy
operation control due to the stable boiler load.
2.2  Indirect and Modified Lime/Limestone Process
     Table 6 shows the operation data of major  plants by-producing gypsum by
indirect lime/limestone processes (double alkali type)  and the Kobe Steel and
Kawasaki processes (modified lime/limestone process) which use— in addition to
lime— calcium chloride and magnesia, respectively.  The pH of the absorbing liquor
of the processes ranges from 1.0 (Chiyoda) to 6.8 (Showa Denko).  For processes
which use a liquor with a pH below 4.0 (Chiyoda, Dowa,  and Kurabo),  oxidation is
carried out with the liquor and proceeds more rapidly than with the calcium sulfite
in other processes.  The SO  removal efficiency ranges from 90 to 96%.
     Figure 2 shows the relationship of pH and plant performance.  The lower the pH,
the larger the L/C ratio and power consumption, and the higher the operability.  The
higher operability may be due to the lesser tendency of scaling.
2.3  Regenerable Processes
     Table 7 shows operation data of major regenerable process plants.  A large
ammonia scrubbing plant by the Nippon Kokan process has been operated with 100%
operability (Section 3.2).  Wellman-Lord process plants have been operated smoothly
but require extensive wastewater treatment including ozone oxidation to decompose
polythionates such as Na^S^O,-, which form mainly at the heating step of the
absorbing liquor.  Polythionates form also in other wet processes even though in
small amounts and might necessitate treatment when wastewater regulations are
tightened.
     The Chemico-Mitsui magnesium scrubbing plant of Idemitsu Kosan encountered
problems for over 1 year after its start-up in 1975.  The problems have since been
solved through improvements of the process by Idemitsu Kosan and Mitsui Miike.
Normally no wastewater is purged from the system.
     Sodium scrubbing to by-produce sodium sulfite for paper mills  (such as the
Kureha process in the Table) is most simple and requires the least energy, but
                                        426

-------
    100
        -A;QO
     99
O  98


4J
•H

•H


2   97
0)
o,
o
-     Utility

      boilers
    96
                                         Industrial

                                           boilers
         O Oil-fired  boiler

         A Coal-fired boiler

         D Iron-ore sintering machine
                                     Oil
      0
 Figure 1
                               Coal
        500     1000       1500


             Inlet S0a(ppm)
                                            2000
                                                   2500
     Relationship of inlet S02 concentration and


     operability of FGD ( lime/limestone process)
   100
     4 r
  '•> o
  6-5 3
   e
   0)

M-rl  ,  L
^  3  1 ^
flj  O*
C  
-------
        Table  6
OPERATION DATA OF INDIRECT AND MODIFIED LIME/LIMESTONE PROCESS PLANTS
ho
00
Process developer

Absorbent
Precipitant
Plant owner

Plant site
Fuel
FGD capacity (1,000 Nm3/hr)
FGD capacity (MW)
Inlet S02 (ppm)
Inlet dust (mg/Nm3)
Inlet gas temperature (°C)
Number of scrubbers in parallel
Prescrubber type
*5
L/G (liters/Nm3)
Scrubber type
Liquor pH
Concentration
L/G (liters/Nm3)
Gas velocity (m/sec.)
Outlet S02 (ppm)
Outlet dust (mg/Nm3)
S02 removal efficiency (%)
Mist eliminator type
_ Prescrubber
Pressure
drop Scrubber
, „ **.. Mist eliminator
(mmH20) Total system
Wastewater purged (t/hr)
Energy requirements (Design)
Pump (kW)
Fan (kW)
Total system (kW)
Per cent of power generated
Operability (%)
Kureha-
Kawasaki
NaOH
CaC03
Shikoku
Electric
Sakaide
Oil
1,270
450
1,270
20
135
2
None

Packed
6.2
20
1.2
2
70
10
95
Terellette

115
25
310
None



7,900
1.8
98.6
Showa
Denko
NaOH
CaC03
Showa
Denko
Ichihara
Oil
500
150
1,400
100-200
140
4
None

vcb>
6.8
25
0.5-1

50-90
Below 50
93-96


300-500
30-50
400-700
4-6



3,500
2.3
98.7
Chiyoda

H2S04
CaC03
Hokuriku
Electric
Fukui
Oil
980
350
1,800
30
140
1
Venturi

Packed
1
1-2
40-60

80

96
Euroform



680
7-24

5,300
5,500
12, 300
3.5
100
Dowa

A12(S04)3
CaC03
Naikai
Salt
Tamano
Oil
72
(25)
1,500
200
170
1
Spray
2.5
Packed
3.5

10
1.2
100
50
93
Wire mesh
10-20
70-80
20-30
170
0.3

150
300
580
2.3
99.6
Kurabo

NH3
CaO



Oil
115
(40)
1,480
150
170
1

1.4
Packed
3.8
10
8
2
130
Below 50
91
Euroform
130
120

300
None

470
190
880
2.2
99.3
Nippon
Kokan
NH3
CaO
Nippon
Kokan
Keihin
Coke*)
150
(50)
350
80
120
1
Spray
1.0
Screen
6.0
30
2
2
10-20






300
2





Kobe
Steel
CaO-CaCl2

Nakayama
Steel
Funamachi
Cokea)
375
(125)
150-250
300-400
140-155
1
Spray
3.5
Venturi
5-5.5
30+6d)
3
3
15-25
40-50
90
Euroform
15
120
35
220
None

1,200
1,000
2,600
2.1
Above 95 98.1
Kawasaki
H.I.
Mg(OH)2
CaO, CaC03
Unitika

Okazaki
Oil
200
68
1,400
200
170
1
None

MVC>
5-6

6
3
Below 140
Below 100
Above 90
Louver

115
50
230
None

430
370
1,160
1.7
99
       a) Iron-ore sintering plant   b) Vertical cone   c) Multi-venturi   d) 30% CaCl2 + 6%  CaO

-------
                              Table 7  OPERATION DATA OF REGENERABLE PROCESS  PLANTS
vO
Process developer
Absorbent
By-product
Plant owner

Plant site
Fuel
FGD capacity (1,000 Nm /hr)
FGD capacity (MW)
Inlet S02 (ppm)
Inlet dust (mg/Nm3)
Inlet gas temperature (°C)
Number of scrubbers in parallel
Prescrubber type
L/G (liters/Nm3)
Scrubber type
Liquor pH
L/G (liters/Nm3)
Gas velocity (m/sec)
Outlet S02 (ppm)
Outlet dust (mg/Nm3)
S02 removal efficiency (%)
Mist eliminator type
Pressure Scrubber
drop Mist eliminator
r
(mmH20) Total system
Was tewater purged (t/hr)
Energy requirements (Design)
Pump (kW)
Fan (kW)
Total FGD system (kW)
Per cent of power generated
Operability (%)
Nippon
Kokan
NH3
(NH. ) 0SO.
Nippon
Kokan
Ogishima
Cokea)
1,120
(380)
350
50
120
2
Spray
1.0
Screen
6.0
1.0
1.6
10-20
10
94-97
Wet EP


250
10





100
Wellman-
MKK
NaOH
S02-*H2S04
Chubu
Electric
Nishinagoya
Oil
620
220
1,600

140
1


Sieve tray

0.6
1.8
120
35
92

400
50
550
4

840
2,350

1.5*0
97.8
Wellman-
SCEC
NaOH
S02-»H2S04
Sumitomo
Chemical
Sodegaura
Oil
370
130
1,500
100
160







Below 150
Below 50
Over 90




Some




Mitsui-
Chemico
MgO
so2->s
Idemitsu
Kosan
Chiba
Oil
460b)
(160)
2,850

185
1
Venturi

Venturi



120

95
Chemico


500f)
0.1

1,960
3,400

Kureha
NaOH
Na2S03
Mitsui
Toatsu
Nagoya
Oil
190
65
1,400
200-300
170



Packed
6.5
1.2
Below 2.0
6



165
40
250
Some



560
3.4*0 0.9
Over 95
98
100
Shellc)
CuO
so2->s
Showa
Y.S.
Yokkaichi
Oil
116
38
1,250
Below 50

id)


PPe)



125
Below 50
90

200

400-500f)
Minor

140
730
870f)
2.3f»
88i)
TEP CO-
Hitachi
Carbon
HnSO.->CaSO.
24 4
Tokyo
Electric
Kashima
Oil
420
150
150

130



Packed



30

80

630

870
13

280
2,700
3,245
§) 2.2
92i)
            a)  Iron ore sintering plant   b)  From oil burner  and  Glaus  furnace   c)  Dry  process  removing up  to
            70% of  NOx simultaneously  d)  Two  reactors  are used  alternately  for S02  absorption and regeneration
            e)  Parallel passage reactor   f)  Excluding Glaus  furnace  g)  Including  energy for steam  h)  For pump
            and fan  i)  FGD  plant was  shut down for  inspection

-------
demand for the sulfite is limited.
     The Shell process plant of SYS which had a low operability  for  a  few  years
achieved ten months continuous operation recently.  Although  the process seems costly,
the capability of simultaneous removal of NOx may compensate  for the disadvantage.
     The Kashima plant, Tokyo Electric, using carbon absorption  and  water  wash, has
been operated for 6 years without appreciable problems.  Carbon  consumption was
proved low (about 2% yearly) owing to the use of a fixed bed  and to  regeneration by
water wash.
2.4  Operability
     As shown in Table 8, most FGD plants have an operability of over  97%.  Operability
means FGD operation hours per cent of the desired operation hours of FGD,  vis. the
scheduled gas source operation hours less the hours of shutdown  caused by  trouble
with the gas source.  For oil-fired boilers, when an FGD plant has to  be shut down
due to its own trouble, boiler operation is continued by switching immediately to a
low-sulfur oil.  For coal-fired boilers and sintering machines,  it can happen that
the gas source has to be shut down due to FGD trouble.  In such  a case the true
operability is less than the FGD Operation hours per cent of  the gas source operation
hours, as shown for EPDC's Takasago plant in Table 8.
     Three plants in the table have an operability of 100% but this  does not mean the
plants have no problems at all.  Minor problems can be solved without  interrupting
operation.  For example, most of the plants have a stand-by pump to  replace a pump
during FGD operation.  With corrosion, scaling and plugging under a  certain level,
the FGD plant can be operated continuously until the scheduled shutdown of the gas
source, when it can be repaired.  Normally boilers are operated  continuously
for 11 months and then undergo one month's shutdown, for annual maintenance, while
iron-ore sintering machines are operated continuously for 2 or 3 months and shut down
for several days for maintenance.
2.5  Labor Requirement
     Most larger FGD plants are operated by 2-4 persons per shift who  also carry out
minor maintenance work.  For annual maintenance to solve serious problems  of an FGD
system, a skilled maintenance staff who takes care of the whole  power  plant (or steel
plant, etc.)  looks after the FGD system.
     Typical labor requirements are shown in Table 9.  The Fukui plant, Hokuriku
Electric, by the Chiyoda process shows the least labor needs  indicating trouble-free
operation.  On the other hand, the Sakaide plant, Shikoku Electric,  shows  the largest
man-hour requirements, although the operability is relatively high,  presumably
                                        430

-------
Table  8    OPERATION HOURS AND FGD OPERABILITY IN RECENT ONE YEAR
Plant owner Plant site
Lime-Limestone process
Chubu Electric Owase (1)
" (2)
Kyushu Electric Kanda
11 Karatsu
Electric P. B.C. Takasagob)
Mitsui Aluminum Omuta
N u
Sumitomo Metal Kashima
Gas FGD capa-
source Process city (MW)

UBa>
tt
it
it
u
IBC)
It
SMd>

MHI
u
n
u
Mitsui-Chemico
Chemico-Mitsui
Mitsui-Chemico
Sumitomo-Fuj ikasui

375
375
188
240
250
156
175
(330)
Operation (hr/year)
Boiler (A) FGD (B)

7,320
7,565
7,420
7,271
8,180
8,244
8,040
8,285

7,171
7,485
7,390
7,246
8,010
8,232
8,040
8,285
Opera-
bility
(B/A)x
100

98.0
98.9
99.6
99.7
97.9^
99.9
100-0
100.0
Inlet
S02
(pptn)

1,600
1,600
800
530
1,500
2,300
2,300
500
Indirect or modified lime-limestone process
Shikoku Electric Sakaide
Hokuriku Electric Fukui
Showa Denko Ichihara
Unitika Okazaki
Naikai Salt Tamano
Nippon Kokan Keihin
Nakayama Steel Kobe
Regenerable process
Idemitsu Kosan Chiba
Chubu Electric Nishinagoya
Showa Y.S. Yokkaichi
Nippon Kokan Fukuyama
a) Utility boiler.
UB
u
IB
u
ti
SM
n

IB
UB
IB
SM

Kur eha-Kawa saki
Chiyoda
Showa Denko
Kawasaki
Dowa
Nippon Kokan
Kobe Steel

Chemico-Mitsui
Wellman-MKK
Shell
Nippon Kokan

450
350
150
67
28
(50)
(125)

170
220
40
(253)

7,441
7,044
7,885
8,232
8,001
8,202
8,419

8,016
7,247
7,656
4.2638J

7,336
7,044
7,775
8,160
7,969
8,098
8,259

7,887
7,090
6,720
4, 26 38 >

98.6
100.0
98.6
99.1
99.6
98.7
98.1

98.4
97.8 ^
87.8f)
100.0

1,270
1,800
1,400
1,400
1,500
350
200

2,850
1,500
1,250
350

Year
comp-
leted

1976
1976
1974
1976
1975
1972
1975
1976

1975
1975
1973
1975
1976
1972
1976

1975
1973
1973
1976

b) Coal-fired; other boilers are oil-fired.
c) Industrial boiler.
d) Sintering machine.


e) True operability is 97.0% because
f) FDG plant was shut down for
g) Due to production control.




boiler stopped for hours due










to FGD trouble.
inspection.









-------
             Table  9     LABOR REQUIREMENTS OF FGD PLANTS (recent one year)
        Plant owner
        (plant site)
Shikoku Electric
(Sakaide)

Mitsui Aluminum3
(Omuta)
                b)
Chugoku Electric
(Tamashima)

Chubu Electric
(Nishinagoya)

Chubu Electric
(Owase-Mita)

Hokuriku Electric
(Fukui)
      Process      FGD capaci-  Operation personnel (man-hours/year)    Operability
    (Absorbent)    ty (MW)    Skilled  Unskilled  Maintenance   Total	(%)	
Kur eha-Kawa s aki
(Na2S03-CaC03)
CaO
Chiyoda
(H2S04-CaC03)
450
Chemico-Mitsui
(Carbide lime)
Babcock-Hitachi
(CaC03)
Wellman-MKK
(Na2S03)
156
500
220
375
350
33,000    16,000     19,900     68,900     98.6


 8,040     8,040     15,360     31,440    100


17,520    17,520     not clear             97.4


      17,000         14,000     31,000     97.8


      17,000         14,000     31,000     98.5
      20,800
3,300     24,100    100
    a)  Coal-fired; others are oil-fired.
    b)  Has three scrubbers and a stand-by scrubber; others have no stand-by.

-------
because the process is not simple.
3  NEW FGD TECHNOLOGY
3.1  Chiyoda Jet-Bubbling Limestone-Gypsum Process
     Chiyoda Chemical Engineering & Construction Co. has developed a new FGD process
using a multi-function jet bubbling reactor which serves as absorber, a limestone
                                                                                    2)
reactor, an oxidizer and a gypsum crystallizer requiring no slurry circulation pump.
Following pilot plant tests in Japan,  a prototype unit with a capacity of treating
         3
85,000 nm /hr of flue gas from a coal-fired boiler (23 MW equivalent) was constructed
                                                        3)
at Gulf Power Company's Scholz Station in Florida, U. S.    The plant went into
operation in August 1978 and has been operated since at nearly 100% operability
without any scaling problem removing 90-93% of S0? with 250 mmH20 pressure drop in
the reactor and 100-101% stoichiometry of limestone.  Details of the operation will
be given by EPRI at the present symposium.
3.2  Sodium Limestone Process With Electrolytic Cell
     The Buzen plant, Kyushu Electric, with a capacity of treating half the flue gas
from a 500 MW oil-fired boiler by the Kureha-Kawasaki sodium limestone process, has
used an electrolytic cell originally developed by Ionics, U. S., to decompose
by-product Na2SO, to NaOH and N2SO,.  About 800 kg/hr of Na2SO, formed in the FGD
system is sent to the cell in a 20% solution after being treated by NaOH and Na^CO,
to remove impurities.  The by-produced NaOH is sent to the scrubber system and the
H2SO, is sent to an oxidizer of calcium sulfite to promote the oxidation.  No
wastewater is purged from the system.
     The plant has been operated since November 1977 and the cell had a corrosion
problem at the beginning. The problem has been reduced by changing the construction
materials of the cell.
3.3  Nippon Kokail Ammonia Scrubbing Process
     Nippon Kokan (NKK) recently completed two large FGD plants to treat flue gas
from iron-ore sintering machines to produce ammonium sulfate utilizing ammonia in
coke oven gas.  A flow sheet of one of them, the Ogishima plant, is shown in Figure 3.
                     3
Flue gas (1,120,000 Nm /hr) is treated with ammonium sulfite liquor to remove over
95% of S09.  The liquor discharged from the scrubbers(two in parallel)containing
ammonium bisulfite is contacted with coke oven gas to absorb ammonia.  A portion of
the resulting ammonium sulfite liquor is oxidized to produce ammonium sulfate.  The
operation parameters are shown in Table 7-
     Hydrogen sulfide in the coke oven gas is removed prior to  the ammonia absorption.

                                        433

-------
Sintering
plant

1,120,000 Nm3
^
SO- 300 ppm
Dust 500 mg/
ESP
Nms
Coke
  oven
lOO.OOONm'/hr
                           Takahax
                                                              SO, 10 ppm, NH. 2-20 ppm, Dust 1-2 mg/Nm
110°C ^~
)150°C
Dust

^LA
s~~z

i. j
90°C



FGD
50°C
x.
50°C
Wet ESP


                                                                                                  Mr
                                                                                     Cleaned coke oven gas
                                           (S, HH4HS, NH4SCN)
   Figure  3    S0_ and NH_ absorption system
                   (Ogishima plant, Nippon Kokan)
                                                                  Crystallizer
                                                                              (NH4C1)

-------
By-products from the sulfide removal unit are also oxidized and coverted to
ammonium sulfate.
     The flue gas leaving the scrubber is passed through wet electrostatic
precipitators (eight in parallel) and heated to 110°C by two Ljungstrom type heat
exchangers installed in parallel.  No plume at all is observed from the stack.
     Nippon Kokan has a similar plant at Fukuyama, with a capacity of treating
          o
760,000 Nm /hr of flue gas.  At this plant, the treated flue gas is heated by
conventional oil firing using neither wet electrostatic precipitator nor heat
exchanger.  An appreciable plume is observed at the stack.
3.4  Ammonia Scrubbing by Ube Industries Process
     Ube Industries has developed an ammonia scrubbing process and constructed two
                                                                     3
commercial units in 1977, each with a capacity of treating 110,000 Nm /hr of flue gas
from a boiler burning 2% sulfur oil.  Ammonium sulfite is by-produced which is highly
purified and used for caprolactam production.  The total investment cost was nearly 1
billion yen (13,700 yen/kW).  An appreciable plume is observed.  Tests are in
progress to reduce the plume.

3.5  Gas-Gas Heating by Ljungstrom Heat Exchanger
     In principle, a gas-gas heat exchanger as shown in Figure 3 is very useful for
FGD because it not only saves energy but also reduces the consumption of cooling
water.  This type of heat exchanger has not been used commercially for FGD because
of corrosion and solid deposition within it.
     Pilot plant tests were carried out by Gadelius Co. (Japan) jointly with EPDC
for a wet limestone FGD system for flue gas from a coal-fired boiler and also with
Tohoku Electric for a sodium limestone FGD system for flue gas from an oil-fired
boiler.  Corrosion-resistant materials and soot blow were used to solve the problems.
                                                        o
     Tests for 6,000 hours with oil-fired gas (10,000 Nm /hr) containing about
10 mg/Nm  of dust showed that a slight deposit less than 0.5 mm in thickness formed
in a zone where the soot blow was not effective.  Pressure drop did not increase
appreciably and the deposit could be removed by water wash.    Tests with flue gas
from coal containing 100-200 mg/Nm  of fly ash showed that the solid deposits were
                     I
much softer than those with oil-fired gas and could be removed by soot blowing.
Virtually no corrosion was observed in either case with the elements coated with
enamel.
     It has been decided to use Ljungstrom type heat exchangers for 3 wet limestone
process FGD plants to be completed between 1979 and 1981 for coal-fired boilers.
                                          435

-------
4  NOx ABATEMENT  )
4.1  Outline
     NOx emissions from stationary sources have been controlled by the emission
standards enforced by the Central Government and also by agreements of industry with
local governments.  The emission standards for boilers are shown in Table 10.
          Table 10  NOx EMISSION STANDARDS FOR BOILERS (ppm)

                                Capacity (1.000 Nm3/hr)
10-40
Fuel
Gas
Oil
Coal
Na
130
150
400
Eb
150
230
600
40-100
N
130
150
400
E
130
190
600
100-500
N
100
150
400
E
130
190
480
Over 500
N
60
130
400
E
130
180
480
                       a  For new boilers
                       b  For existing boilers
     Combustion modification has made much progress and has satisfied the emission
standards.  NOx concentrations in flue gases from some of the utility boilers have
been reduced to a very low level	 40 ppm for gas, 100 ppm for oil, and 200 ppm
for coal.  Stringent regulations have been applied also to mobile sources.  New
passenger cars have already met the most stringent NOx emission control, 0.25 g/km.
The ambient NO^ standard, 0.04-0.06 ppm in daily average, however, still cannot be
met in large cities and industrial regions.  The total mass regulations are to be
applied for NOx sources in polluted regions whose daily average N0? concentrations
exceed 0.06 ppm.  In addition, local governments apply very stringent control on
new large stationary sources of NOx.  Flue gas treatment for NOx removal  (flue gas
denitrification) is thus needed.  Major denitrification processes are shown in
Table 11.
                                       436

-------
       Table 11  MAJOR PROCESSES FOR FLUE GAS  DENITRIFICATION
                      r
                       Selective catalytic reduction (SCR)
           y         -I Selective noncatalytic reduction (SNR)
         processes    !
                      ^Simultaneous NOx,  SOx removal /Copper oxide process
                                                      Carbon process
                                                     lElectron beam process

         w            rOxidation absorption
         processes    I Simultaneous NOx,  SOx removal /Oxidation reduction
                                                     IReduction

4.2  Selective Catalytic Reduction (SCR)
     Among the denitrification processes,  SCR which uses  NH~ and catalyst to
reduce NOx to N2 at 200-450°C has been most popular and adopted at 60 commercial
plants because it is simple,  can give a high NOx removal  efficiency of over 90% and
does not give by-products difficult of disposal except the spent catalyst.  Major
SCR plants of power companies are listed  in Table 12.
     A small amount of 02 is  needed for the reaction of NH3 with NOx, which is
expressed in the following equation:
               4NO + 4NH_ +0  = 4N_ + 6H 0
     As the catalyst, base metals such as Fe, V,  Cr, Cu,  Co, and Mo have been used.
As the catalyst carrier (support), Al_0«  was used at the  beginning because of the
                                  6}
large surface area as shown below:
               /-A1203 > Ti02 > Zr02 > MgO > ,^-Al^ > Si02
     Since Al_0» reacts with  SOx in the gas, and this results in decreases in the
surface area and activity, most of the recent catalysts use TiO_ or its compounds.
The SOx resistance of the carrier is as shown below:
               Ti02 = Si02> 0(-Al203> ty-A!203> 
-------
                                Table 12  MAJOR NOx REMOVAL PLANTS OF POWER COMPANIES
-P-
U)
oo
Plant
constructor
Plant
Power company site
Fuel of
boiler
Selective catalytic reduction ( SCR, 80-90% removal)
Hitachi Ltd. Kansai Electric Kainan LSOa
Hitachi Ltd
Hitachi Ltd
MHI
MHI
IHI
IHI
IHI
NDG
. Chubu Electric Chita
Hokkaido Electric Tomakomai
Kyushu Electric Kokura
Chubu Electric Chita
Chugoku Electric Kudamatsu
Tohoku Electric Niigata
Company A
EPDC Takaehara
LNG
Coal
LNG
LSO
LSO
LSO
LSO
Coal
NOx removal
capacity (MW)
100
700 x 2
90
600 x 2
700
700
600
350
250
Reactor
(Catalyst)
FBb
FB
MBC
FB
HCd
HC
HC
HC
ND6
Year of
completion
1977
1977
1981
1978
1980
1979
1981
1978
1980
Selective noncatalytic reduction (SNR, 45% removal)
MHI
Combination of
Hitachi Ltd.
MHI
IHI
IHI
Chubu Electric Chita
SNR and SCR (50 - 70% removal)
Company B
Company B
Company B
Company C
LSO
LSO
LSO
LSO
LSO
375
156
156
156
350

pcf
PC
HC
HC
1977
1978
1978
1978
1978
           a  Low-sulfur oil     b


           d  Honeycomb catalyst
Fixed bed      c  Moving bed (with hot electrostatic precipitator)

  e  Not decided      f  Parallel plate catalyst

-------
   100
    90
    80
  H
    70
                I
0.7      0.8       0.9       1.0       1.1

                NH,/NOx mole ratio


   Figure 4  Typical operation data of  SCR

              ( Inlet NOx  150  - 200  ppm)
                                                  20
                                                  10
                                                    1.2
    60
2   20
                    NOx
                                NHS
                                            Inlet
                                          NOx(ppm)
                                                       80


                                                       60
                                                       a,
                                                  40  ~
                                                       20
                                                           •a
                                                            0)
      0123

                  NHa/NOx mole ratio


   Figure 5  Operation data of SNR at Chita plant,

             Chubu Electric (375  MW oil-fired boiler)
                            439

-------
     One problem with SCR as well as with SNR is the deposit of ammonium bisulfate
below about 220°C in the air preheater, which increases the pressure  drop and also
causes corrosion.
     In order to minimize the formation of the bisulfate, it is preferable to reduce
the leak NH  below 5 ppm by using less NH-, viz. 0.85-0.90 mole/mole of NOx, to
obtain 80-85% removal efficiency.
     Moving bed type and parallel flow type reactors have been developed and used
commercially in order to prevent clogging of catalyst with dust in flue gas.  The
former uses a granular catalyst which is charged from the top of the reactor and
is moved down intermittently or continuously while the gas is passed through the
catalyst layer in a cross flow.  The catalyst discharged from the bottom of the
reactor is screened to remove the dust and then returned to the reactor.  The
                                                                  3
reactor usually can treat flue gas containing up to about 0.2 g/Nm  of dust.
     On the other hand, the parallel flow type reactor uses a fixed bed of a
different type of catalyst --- honeycomb, plate, tube, or a parallel passage device.
The gas passes through a clearance between the parallel layers of the catalyst and
the reactor is expected to be able to handle even a flue gas from a coal-fired
                               3
boiler containing about 20 g/Nm  of dust although the dust may erode the catalyst
at a large gas velocity or may deposit on the catalyst at a small gas velocity.
     SCR catalysts tend to oxidize a small portion (usually below 5%) of S0? in the
gas to S0«.  Studies have been made to produce catalysts which cause no oxidation.
Catalysts reactive at 150-200°C have been developed to treat low temperature gases.
A common problem for the low- temperature catalysts is the deposit of ammonium
bisulfate on the catalyst to lower the activity.  The activity can be recovered by
occasional heating of the catalyst above 350°C to remove the bisulfate.
4.3  Selective Noncatalytic Reduction (SNR) (Thermal De-NOx)
     Ammonia rapidly reacts with NOx around 1,000°C to form N2 and HO without
catalyst.  The important keys to a high removal efficiency by SNR are a good rapid
mixing of ammonia with flue gas and a sufficient reaction time  (about 0.2 sec.) at
a suitable temperature range of 900-1050°C.  Those are attained in a laboratory but
not easily at an actual plant.  Figure 5 shows typical operation data of full  scale
tests by Chubu Electric with flue gas from a 375 MW oil-fired boiler  (0.3% sulfur
oil).    A total of 15 ammonia injecting nozzles with many holes are placed in two
locations in the boiler to cope with the gas temperature fluctuation due to the
                                       440

-------
change of the boiler load.    Use of 2 moles of NH  for each mole of NOx gives 50%
removal and also 50 ppm of leak ammonia which causes problems.  Routine operation
has been carried out using about 1.5 moles of NH- to remove about 45% of M0x giving
                         5)
leak NH- of about 30 ppm. '

     Mitsui Petrochemical has been,using, HL and NH., to remove about 40% of NOx in
flue gas from a 40 MW oil-fired boiler by reaction at 700-800°C.  A few companies
have been testing a combination of SNR and SCR 	 ammonia is injected in a boiler
at 800-1,000°C and a small amount of parallel-flow type catalyst is placed in a
duct at 350-400°C 	 to increase NOx removal to 50-70% and to reduce leak NH- to
about 10 ppm (Table 12).  With a larger amount of catalyst, higher removal efficiency
is attained but usually the pressure drop becomes too high because of the large gas
velocity in the duct.
5  SIMULTANEOUS NOx AND SOx REMOVAL
5.1  Shell Copper Oxide Process
     The Yokkaichi plant, SYS, based on the Shell copper oxide process and designed
to remove about 90% of SO- in flue gas from an oil-fired boiler (40 MW equivalent,
3% sulfur oil), has also removed up to 70% of NOx by adding ammonia to the reactor
                                                g\
utilizing the catalytic effect of CuO and CuSO,. '  Pilot plant tests (0.5 MW
equivalent) are to be made at Tampa Electric Company's Big Bend Station to remove
                                                               9)
90% of both SO- and NOx from flue gas from a coal-fired boiler. '
5.2  Activated Carbon Process
     Activated carbon adsorbs SOx and also works as an SCR catalyst, particularly
when impregnated with a small amount of metal compound.  Flue gas  injected with NH,
                                                                          -1
is passed through the carbon bed around 220°C with an SV of about  1,000 hr   for
90% removal of both SO- and NOx.  A higher temperature increases the NOx removal
but decreases the S09 removal.  SOx is adsorbed by the carbon to form H SO, and
NH.HSO,, which are removed by heating the carbon at 350°C in an inert gas produced
by incomplete combustion of fuel.  Concentrated SO,, is recovered.   Tests have been
carried out with pilot plants (0.7 and 2 MW equivalent).
5.3  Electron Beam Process
     Flue gas at about 100°C is mixed with NH- and exposed to electron beam radiation.
About 80% of both SOx and NOx are removed with 2 Mrad of the beam forming fine
crystals of ammonium nitrate sulfate double salt which are caught  by an electrostatic
precipitator for fertilizer use.  Tests are in progress at a pilot plant (1 MW
equivalent) of Nippon Steel.

                                        441

-------
5.4  Wet Simultaneous Removal Processes
     NO is not readily absorbed on absorbent liquors while NC>2 and N^O- obtained
by oxidation of NO are more readily absorbed but the resulting liquor containing
nitrate and nitrite is not easily treated.  It is possible, however, to reduce
absorbed NOx to NH  or N  by utilizing the reducing effect of SO^ present in the
flue gas .
     By the oxidation reduction process for simultaneous removal, NO is oxidized
to NO^, which is absorbed in a limestone slurry containing a catalyst, while by the
reduction process, NO is absorbed in a solution of alkaline compound containing
EDTA (ethylenediamine tetraacetic acid) and ferrous ion.  Although the reactions
are complex, forming imidodisulfonate NH(SO M)  and sulfamate NH SO M (M = Na, K,
NH , or l/2Ca) and other intermediate compounds, the overall reaction may be
expressed by the following equations:
            2N02 + 7S02 + 7CaC03 + 3H20 = 7CaS04 + 2NH3 + 7C02          19)
            2N02 + 6S02 + 6CaC03 + 02 = 6CaS04 + NZ + 6(X>2              20)
            2ND + 5S02 + 8NH3 + 8H20 = 5(NH4)2S04                       21)

     Those processes have been tested with pilot plants (0.2 - 2 MW equivalent) and
removed 80 - 90% of NOx with over 95% of SO .  There are no plans to install larger
plants, because the oxidation reduction process requires an expensive oxidizing
agent such as 0~ or C10? while the reduction process is complex.  Moreover, those
processes involve wastewater treatment problems.
5.5  Gas Composition Suitable for Processes
     The relationship of SOx and NOx concentrations in the gas to suitable processes
is  shown in Figure 6.  Although the combination of SCR followed by FGD may be used
for any composition of gas, the application is not easy for SOx-rich gas because of
the ammonium bisulfate problem.  Dry simultaneous removal processes may also suit
gases with a relatively small SOx/NOx ratio for the following reasons:  For the
copper oxide process, a high SOx concentration necessitates   frequent regeneration
which not only requires a large amount of H2 but also tends to lower NOx removal
efficiency.  For the carbon process, the carbon consumption increases with SOx
concentration.  For the electron beam process, the product quality as a fertilizer
is low when SOx concentration is high.  On the other hand, wet simultaneous removal
processes suit S02~rich gas because SO  works as the reducing agent and increases
the NOx removal efficiency.
                                       442

-------
                 §
400



300

200



100

n
—



SCR

,



SNR


SCR + FGD or
dry simultaneous


§i CuO
_2|
it
e MI
o wi
•° S!
M Oil ,'
0) r-ll "
o "i ^X
F


removal *••'
s
X
X
x^x' SCR + FGD
x'
or wet simul-
.-"' taneous removal
x
Absorption reduction


Oxidation reduction
GD
i • I
                          0          500        1,000       1,500
                                        SOx (mainly S02)  (ppm)

                     Figure  6   Gas  composition and suitable processes
2,000
6  ENERGY REQUIREMENT AND COST
6.1  Energy and heat  requirement
     Figure 7 shows various  systems  for SO, and NOx removal and the power
requirement (per cent of  power  generated by boiler)  and heat loss (per cent of heat
applied to the boiler)  involved in the operation of the systems,  with normal boiler
operation (No. 1)  taken as standard  (no requirement, no loss).  The energy required
for the production of the absorbent,  catalyst,  and chemicals such as CaCO,, NH_, H.,
°3' C1°2 is not included  in  the power requirement and heat loss.
     No. 2 illustrates  an application of FGD (wet lime/limestone process for 90%
S02 removal) requiring  2-2.5% of power with gas reheating from 55 to 75°C which
accounts for 1% heat  loss.  No. 3 shows an application of SCR to the gas after
FGD, which requires 3.0-3.5% power and 5% heat and yet is not free from deposits of
ammonium bisulfate as well as solids derived from mists in the heat exchanger and
also on the catalyst.
     No. 4 is a system  that  is  considered most practical.  The gas from a boiler
economizer is treated by  SCR and then by FGD after heat recovery and dust removal.
                                        443

-------
No.
      SOx and NOx removal system
                                         Energy
                                         require-
                                         ment(%)
Heat
loss
                                                               2.0-2.5    1.0
                                                               3.0-3.5    5.0
10   ( B
                                                      2.5-3.5  1.0
    0
Boiler
      H \ Heater
LAH )  Air preheater     (EP )   Electrostatic precipitator


(HE )  Heat exchanger
                                        ©
                                                         Cooler
     HEP  )  Hot electrostatic precipitator
                                           Wet simultaneous removal
         Figure  7   Combined  and  simultaneous SOx and NOx removal systems

                          (Figures  show temperatures, °C)
                                    444

-------
Dust removal may be carried out after FGD.   One problem with the system is the
possible deposit of ammonium bisulfate on SCR catalyst when the gas temperature
drops below about 300°C due to the lowering of the boiler load.  A temperature drop
for a few hours may not hinder the operation because the bisulfate is removed
when the gas temperature is raised above 350°C.   For temperature drops over longer
periods, a heating device such as a hot gas bypass system or an auxiliary burner may
be needed.  Another problem with the system is the accumulation of ammonia in the
wet process FGD system.   A device to remove ammonia from the system may be needed.
     No. 5 is a system using a hot electrostatic precipitator and may be suitable
for flue gas from a low-sulfur coal whose fly ash is not caught efficiently by a
cold electrostatic percipitator.  As an SCR reactor, not only the parallel flow type
but also the moving bed type can be used because the ash content in the gas may be
                          3
reduced to about 200 mg/Nm  by the precipitator.  The ammonium bisulfate problem
for the air heater may be more serious than with  system   No. 4 because the gas
contains less fly ash, which has a sweeping effect.  Other problems are common with
No. 4.
     No. 6 is a combination of SNR and FGD.  Since the NH  concentration at the
boiler outlet is high, the problems of ammonium bisulfate and ammonia accumulation
may be serious.  Consequently heat loss may be slightly higher than with Nos. 4 and
5.  It may be preferable to place a small amount of SCR catalyst in a duct to
reduce leak NH» and to increase NOx removal.
     Nos. 7-10 show dry and wet simultaneous removal processes.  The dry processes
require no gas reheating although they require hydrogen, inert gas, or electron
beam.  Further improvements are desired for commercial application.
6.2  Costs of NOx and SOx Removal
     Figure 8 shows investment costs in battery limits of NOx and SOx removal plants
                                     3
with a capacity of up to 1,000,000 Nm /hr of flue gas from a boiler, which is
equivalent to 330 MW with oil and 280 MW with coal.  Figure 9 shows annualized
operation costs for the plant at 8,000 hours'  yearly operation including 7 years'
depreciation assuming the total investment cost is 50% more than the investment cost
in battery limits.   The costs are based on investigations by the Japan Environment
Agency    and modified by Ando based on his study.  The cost per kW or kWhr for flue
gas from coal is estimated at 20% more than that for flue gas from oil because of
the larger gas volume per kW.
     The cost for SCR by the direct process, viz., treating directly the flue gas
from a boiler economizer, are 3,000-4,000 yen/kW for investment and 0.3-0.4 yen/kWhr
                                        445

-------
   10
CO


I
O
O
O
 CO
 O
 U
I
4-1
CO
0)
c
    0
  Capacity (MW, with oil)


100            200
  1             • '      i  	
                                                 300
                                                  '
   100              200

       Capacity  (MW,,with  coal)
                                                         300
                                                        FGD + SCR
                                                         FGD
                                                                      30
                                                 to
                                     SCR + FGD   o
                                                                    •5 20-
                                                                   o
                                                                   o
                                                                   o
                                                   10'
                                                       SCR
                                                      (Gas heating)



                                                  SCR(Direct,retrofit)

                                                      SCR(Direct, new)

                                                      SNR(Retrofit)
                                                      SNR(New)
                                                                           30
                                                                          - 20
C
0)
                                                            O
                                                            O
                                                            O
                                                            4J
                                                            CO
                                                            O
                                                            O

                                                            4J
                                                            0
                                                        in  w
                                                        1U  Q)
              200      400       600        800

                       Capacity  (1,000  Nm3/hr)
                                1000
      Figure 8  Investment  in battery limits of NOx and SOx removal plants
                                      446

-------
                         Capacity (MW, with oil)
  1.0"
   0.8 -
        -10
01
4J
10
0
o
ffl

c
   0.6
       . _8
   0.4-
   0.2 -
        12
           •H
           O
           c
           0)
    o
he  §
        .4
        .2
        0
                    100
        200
                            200
                             '
               100               200

                      (MW,  with coal)
 300
	L_
                                                           300
                                                    FGD
                                                                           -3
                                                                        3-
                                                                           •2
                                                    FGD + SCR  ^
                                                               r-H
                                                               ca
                                                               o
                                                               a

                                                    SCR + FGD  j-
                                                               4-1
                                                               •rl

                                                           SCR

                                                          (Gas heating)
                                                                        1 '
                                                      SCR(Direct, retrofit)

                                                        'SCR(Direct, new)

                                                        .SNR(Retrofit)

                                                        'SNR(New)
                                                                    -1
     1000
                  400       600        800


                Capacity  (1,000  Nm3/hr)


Figure 9  Annualized operation cost for  NOx and SOx removal plants
                                                                                c
                                                                                a>
                               0)
                               o
                               o
                                                                                a)
                                                                                N
                               n)
                               3
                               C
                                       447

-------
for operation with a 200 MW plant for flue gas from an oil-fired boiler.  The costs
for SCR for a cold gas about 150°C with gas heating by a heat exchanger and a
heater to 300-400°C are about double those for direct SCR.
     The costs for SNR may be between one-half and one-third those for direct SCR,
while NOx removal efficiency is about half in SCR (40-45% versus 80-90%).
     A combination of SCR followed by FGD, as shown in No. 4 of Figure 7, may cost
a little more than the sum of the costs for direct SCR and FGD because of the
requirement of ammonia removal from the FGD system, but is still considerably more
economical than a combination of FGD followed by SCR as shown in No. 3 of Figure 7.
     The costs for the dry and wet simultaneous removal processes are uncertain but
seem to be higher than those for a combination of SCR followed by FGD.
7.  CONCLUSION
     The recent remarkable progress of FGD in Japan was induced by the following
particular circumstances: (1) Severe public criticism and stringent regulations on
pollution. (2) Control by governments using telemeter systerns. (3) Government's
assistance to industry by providing low-interest funds and by allowing short-term
depreciation. (4) Most of the plants were constructed while the Japanese industry
was growing rapidly; the total investment for FGD and hydrodesulfurization of heavy
oil amounting to 1 trillion yen at the current value did not prove an excessive
burden on industry.  (5) Lime scrubbing and ammonia scrubbing which were already
applied in the 1950s provided a basis for the development. (6) Virtually all of the
by-products have been utilized.
     Under such circumstances, over 1,000 commercial FGD plants have been constructed,
operated with good performance, and contributed to the abatement of S02 concentration.
Although the desulfurization efforts have attained the goal, further studies will
be desired to improve FGD 	 particularly in simplificatidn and cost cuts.
     Concerning NOx  removal, combustion modification has made remarkable progress.
Moreover, SCR has been improved considerably and proved commercially applicable to
flue gases from gas  and low-sulfur oil burning.  Further improvements are expected
to apply SCR to more dirty gases.  SNR may be suitable for certain gas sources.
Combination of SNR and SCR may also be useful.  For simultaneous SOx and NOx removal
processes, further tests are needed to evaluate the feasibility of commercial
application.
                                    448

-------
REFERENCES
 1.   Ando,  J.,  SCL  Abatement for Stationary Sources in Japan, EPA-600/7-78-210,
     November  1978.   U.S.  EPA.
 2.   Clasen, D.  D.,  and Idemura,  H.,  Limestone/Gypsum Jet Bubbling Scrubbing System,
     EPA FGD Symposium (November 1977).
 3.   Chiyoda Chemical Engineering & Construction Co.  Ltd., The Chiyoda Thoroughbred
     121 Jet Bubbling Flue Gas  Desulfurization System	Florida Pilot Plant (July
     1978).
 4.   Sudo,  Y.,  et al., Commercialization Study of Gas Heater for Wet Sodium Gypsum
     Process FGD, Karyoku  Genshiryoku Hatsuden, 535 - 545, Vol. 28, No. 6 (June 1977)
     (in Japanese).
 5.   Ando.  J.,  and Nagata, K.  (Draft Report to be published March 1979 by EPA).
 6.   Atsukawa,  M.,  et al., Development of NOx Removal Processes with Catalyst for
     Stationary Combustion Facilities, Mitsubishi Technical Bulletin No. 124 (May 1977)
 7.   MHI Report, Noncatalytic NOx Reduction Process Applied to Large Utility Boiler
     (November 1977).
 8.   Nooy,  F.  M., and Pohlenz,  J. B., Nitrogen Oxides Reduction with the Shell Flue
     Gas Desulfurization Process, Proceedings of Second Pacific Chemical Engineering
     Congress,  Denver (August  1977).
 9.   Mobley, J.  D.,  Status of EPA's NOx Flue Gas Treatment Program, Second EPRI NOx
     Control Technology Seminar,  Denver (November 1978).
10.   Japan  Steel Federation, Status of Development of NOx Removal Technology by
     Steel  Industry (April 1978)  (in Japanese).
11.   Air Preservation Bureau, Environment Agency, Report on NOx Abatement Technology
     (April 1978) (in Japanese).
                                       449

-------
                                                                     4E
                  EPRI'S FGD PROGRAM:  FROM PROBLEM IDENTIFICATION
                            TO  DEVELOPMENT  OF  SOLUTIONS

                                   G. T. Preston
                         Electric Power Research Institute
                                Palo Alto,  CA  94303
                                      ABSTRACT

The overall objective of the Desulfurization Processes Program at Electric Power
Research Institute (EPRI) is to develop the most reliable and cost effective flue
gas desulfurization (FGD) technologies that can satisfy regulatory requirements.
Through 1978 this objective was pursued primarily through evaluation of established
techniques and development of emerging processes.  Representative projects reviewed
in this paper are lime scrubbing and sludge disposal guidelines; characterization of
full-scale operating utility FGD systems; and pilot and laboratory efforts in
absorption/steam stripping, RESOX, and an aqueous carbonate sodium regeneration
process.  In 1979, the emphasis of EPRI's program is shifting to evaluation and
demonstration of advanced technologies, ranging from prototype evaluation of the
cocurrent scrubber configuration and the Chiyoda Thoroughbred 121 and RESOX
processes, to the start of construction of one or more full-scale (100 MW) FGD
systems.
                                     450

-------
                   EPRI'S FGD PROGRAM:  FROM PROBLEM IDENTIFICATION
                             TO DEVELOPMENT OF SOLUTIONS
The purpose of this paper is to provide an overview of the flue gas desulfurization
research program at Electric Power Research Institute.  The structure and goals of
the program are described briefly and the program staff are introduced, followed by
a summary of significant results achieved in each of four subprogram areas within
the program, since our last report to this symposium.

PROGRAM STRUCTURE

Electric Power Research Institute (EPRI) is the research and development arm of the
U.S. electric utility industry.  It is a not-for-profit organization funded by
membership assessments from public and private electric utilities representing about
80% of U.S. generating capacity.  EPRI promotes the development of new and improved
technologies to help the utility industry meet present and future electric energy
needs in environmentally acceptable ways.

Responsibility for research in flue gas desulfurization by stack gas scrubbing rests
with the Desulfurization Processes Program, part of EPRI's Fossil Fuel Power Plants
Department.  The overall objective of the Program is to develop the most cost-
effective flue gas desulfurization (FGD) technologies that can satisfy regulatory
requirements.  The abatement of sulfur oxide emissions from utility boilers is
presently a technology-forced challenge; that is, clean air regulations currently
being promulgated are based on the most advanced existing technologies, even though
                                                                   ? o
some of them have been demonstrated at only relatively small scale. '    This
regulatory approach exerts pressure on EPRI to identify the FGD technologies that
are fundamentally sound and to move these toward commercial feasibility as rapidly
as possible.  Accordingly, the three-phase strategy of the Desulfurization Processes
Program is to:   evaluate on a continuing basis the established technologies and
develop design and operating guidelines; evaluate and develop emerging technologies
through pilot plant tests and operation of integrated prototype systems; and parti-
cipate in the demonstration of commercial-scale scrubbing systems.
                                      451

-------
The four subprograms of EPRI's FGD program, and the objectives of each subprogram
over the next five years, are shown in Table 1.  Activities in each subprogram
follow the three-phase strategy of evaluation, pilot and prototype testing, and
demonstration at commercial scale.

As EPRI's FGD program has grown rapidly, the breadth of identified research needs
has outstripped the funding and staff available to address those needs.  This forces
a careful evaluation of each potential research topic to insure that we make the
most effective use of our resources.  Does it address the overall objective of our
program?  Does it fit our established strategy?  And, does it address one or more of
our subprogram objectives?

There is a further requirement that we impose: that a potential research project
address one, or preferably several, of the specific technical and economic incen-
tives associated with doing research in flue gas desulfurization.  These are as
follows:

     Technical;  Compliance with SO^ removal standards

                 Maximum system reliability

                 Minimum water consumption

                 Minimum by-product volume, or maximum by-product value

                 Minimum heat rate penalty (energy consumption)

     Economic;   Minimum capital cost

                 Minimum operating and maintenance cost

AVAILABILITY OF RESULTS

One of EPRI's charges as a tax-exempt organization is to make the results of the
research available for the public benefit.  This is done in a formal way by
publishing an EPRI report at the end of each research project.  A copy of the EPRI

                                      452

-------
         Table 1.  EPRI'S DESULFURIZATION PROCESSES PROGRAM
SUBPROGRAM
OBJECTIVES  (1979-1983)
State of the Art
Maintain up-to-date design and operating
guidelines for alkali scrubbing

Characterize performance, reliability
and cost of full-scale utility FGD systems

Establish scrubber operator training
centers
Subsystem Evaluation
and Development
Evaluate and develop energy-conserving
scrubber subsystems

Compile materials of construction
experience

Develop advanced contactors and scrubber
configurations
Advanced FGD
With Recovery
Demonstrate advanced FGD to recover
sulfur using coal reductant

Demonstrate sodium regeneration subsystem
in a sulfur-recovery process
Advanced FGD
Without Recovery
Demonstrate sludge-free lime/limestone
FGD processes

Develop improved chemistry basis for
alkali scrubbing
                               453

-------
Publications List may be obtained from Research Reports Center, Post Office Box
10090, Palo Alto, CA 94303, (415) 961-9043.  Reports which have been published in
the FGD area are listed in Table 2.

Information on research in the Desulfurization Processes Program in particular is
available on an informal basis through direct contact with the program manager or
one of the project managers; they are identified in Table 3.

STATE OF THE ART

The overall goal of this subprogram is accurately and on a timely basis to assess
the status of FGD technology as it is commercially available to the utilities, so
that a utility with a need to install SO- scrubbing has up to date information on
which to base procurement decisions, and so that operators of existing FGD systems
can maximize reliability and minimize their cost.  Recent results in this area are
summarized below.

Stack Gas Emission Coordination Control Center

The objective is to maintain an up-to-date information base of scrubber operating
experience and data as an aid to EPRJ member utilities in keeping their FGD systems
operating and in planning future installations, and to identify common operating
problems as an aid to EPRI research planning.  Battelle Columbus Laboratories is the
contractor.

Since our last report to this symposium, Battelle has completed an analysis of the
causes of wide variations in the bid-price cost estimates for FGD systems.  The
principal factors are the date of the estimate and the estimating procedure. The
impact of the estimate date is due not only to inflation but also to design changes
resulting from technological advances and new regulatory requirements.

As part of this project, Battelle provides short term information search and
consulting efforts without charge to EPRI member utilities who have specific
information needs covering any aspect of.FGD technology.  Details can be obtained
from Dr. Harvey Rosenberg at Battelle, (644) 424-5010, or from Tom Morasky at EPRI.
                                      454

-------
Table 2.  PUBLISHED REPORTS - DESULFURIZATION PROCESSES PROGRAM
DATE
REPORT NO.
TITLE
Aug 1975  EPRI 209
          Part 1

Aug 1975  EPRI 209
          Part 2

Sep 1975  EPRI 202
Mar 1976  FP-272
          Vol. I,II
          Addendum

Oct 1976  FP-207
Dec 1976  FP327
Feb 1977  FP-361
Jul 1977  FP-463-SR
Dec 1977  FP-595
Dec 1977  FP-639
Mar 1978  FP-713
          Vol.  I-III

Mar 1978  FP-671
          Vol.  Ill

Mar 1978  FP-889

Oct 1978  FP-909


Jan 1979  FP-942
              Status of Stack Gas Control Technology

              Status of Stack Gas Technology for S02 Control


              Environmental Effects of Trace Elements from
              Ponded Ash and Scrubber Sludge

              Evaluation of Regenerable Flue^Gas Desulfur-
              ization Processes
              Evaluation of Dry Alkalis for Removing Sulfur
              Dioxide from Boiler Flue Gases

              Guidelines for the Design of Mist Eliminators
              for Lime/Limestone Scrubbing Systems

              Stack Gas Reheat for Wet Flue Gas
              Desulfurization Systems

              Process Synthesis and Innovation in Flue
              Gas Desulfurization

              Application of Scrubbing Systems to Low
              Sulfur Alkaline Ash Coals

              A Summary of the Effects of Important Chemical
              Variables Upon the Performance of Lime/
              Limestone Wet Scrubbing Systems

              Evaluation of Three 20 MW Prototype Flue
              Gas Desulfurization Processes

              State-of-the-Art of FGD Sludge Fixation
              EPRI/Radian Particle Balance Concept Study

              Analysis of Variations in Costs of FGD
              Systems

              Full-Scale Scrubber Sludge Characterization
              Studies
                               455

-------
Table 2.  PUBLISHED REPORTS - DESULFURIZATION PROCESSES PROGRAM

(Continued)

           REPORT NO.    TITLE
DATE

Jan 1979


Jan 1979

Jan 1979



Feb 1979


Mar 1979
           FP-671
           Vol. 1

           FP-977

           FP-941
           FP-671
           Vol. II

           FP-1030
Review and Assessment of the Existing Data
Base Regarding Flue Gas Cleaning Wastes

FGD Sludge Disposal Manual

Cocurrent Scrubber Evaluation
TVA's Colbert Lime/Limestone Wet Scrubbing
Pilot Plant

Chemical/Physical Stability of Flue Gas
Cleaning Wastes

Lime FGD Systems Data Book
                                456

-------
                    Table 3.  EPRI FGD STAFF
George T.  Preston
(415)  855-2461
Program Manager
Stuart M.  Dalton
(415)  855-2467
Project Manager
   State of the Art
   Advanced FGD with Recovery
Charles E.  Dene
(502)  443-6489
Facility Manager - Shawnee
   State of the Art
Thomas M.  Morasky
(415)  855-2468
Project Manager
   State of the- Art
   Advanced FGD without Recovery
Richard G.  Rhudy
(415)  855-2421
Project Manager
   State of the Art
   Subsystems Evaluation &
     Development
   Advanced FGD with Recovery
Dorothy A.  Stewart
(415)  855-2609
Project Manager
   Subsystems Evaluation &
     Development
   Advanced FGD without Recovery
                               457

-------
Evaluation of Improved Process Control Capability for FGD Systems

The overall objective is to evaluate the status of FGD process control practices and
instrumentation and recommend changes and/or research to develop improved methods.
The project has been completed since oxir last report to this symposium, and three
reports have been published.  The major conclusions are:

     0    By-product sludge characteristics can be improved by operating FGD systems
          so as to reduce the rate of formation of new sludge crystals.

     0    Poor operating reliability in FGD systems is often due to a spiraling
          sequence of:  scale formation, sensor malfunction, unstable control of pH
          and other chemical concentrations, and a further increased rate of scale
          formation.  Valid sampling procedures are essential in maintaining
          effective operating control.

Follow-on research is planned for late 1979 or 1980, to develop improved instrumen-
tation and sampling techniques to break the instability/scaling cycle referred to
above.

Characterization of Full-Scale Scrubbers

The objective is to characterize and publish data for four representative full-scale
utility lime and limestone wet scrubbing systems.  Field testing is being carried
out to determine removal efficiencies for regulated air and water emissions, such as
SOj and particulates, as well as currently unregulated discharges such as fine
particulate, polycyclic organics, and vapor-phase metals.  An engineering/economic
evaluation of each system is also being performed to document system operability and
costs.  The four systems currently scheduled for characterization are Pennsylvania
Power Company's Bruce Mansfield plant, Columbus & Southern Ohio's Conesville plant,
Montana Power's Colstrip station, and Northern States Power's Sherburne County
station.  These four units constitute a cross-section of lime and limestone FGD
technology on western and eastern coal.  The first phase of testing has been
completed at the Conesville plant, and testing will begin at Colstrip in April 1979.
                                      458

-------
By-Prdduct/Waste Disposal for Flue Gas Cleaning Processes

The overall objective of several research efforts under this project is to establish
a sound information base to help utilities select fly ash and scrubber sludge
disposal methods from commercially available technologies.  The project is complete;
four reports have been issued describing the following results:

     0    The existing information base on flue gas cleaning (FGC)  wastes has been
          reviewed and a laboratory program was outlined to fill in major gaps in
          the available knowledge.

     0    The laboratory program is continuing.  One useful preliminary conclusion
          is that it is possible to predict the long-term stability of fly-ash-
          stabilized sludge with reasonable confidence on the basis of its physical
          characteristics after 50 days aging.

     0    The FGD sludge fixation processes of  IUCS and Dravo are sufficiently
          developed and tested to be considered commercially available for utility
          application.  The incremental costs of sludge fixation over disposal by
          ponding or landfilling are estimated  at $6.90 and $2.50,  respectively,  per
          ton of dry sludge for a hypothetical  1000 MW generating station.  (Since
          these estimates reflect only the additional cost of fixation,  they do not
          indicate whether ponding or landfill  will be preferred at a given site.)

     0    Sludge slurry samples from six operating full-scale utility FGD systems
          were evaluated for their dewatering characteristics when  subjected to
          settling, filtration, or centrifugation.  The results confirmed that
          larger sludge particle size distributions lead to better  dewatering
          characteristics, and indicated also that the use of flocculating agents
          can usually improve dewatering performance and reduce cost.  It appears
          that quantitative understanding of the relationship between dewatering
          characteristics and scrubber operating parameters cannot  be derived from
          data obtainable in a full-scale system in normal day-to-day operation.
                                     459

-------
     0    A comprehensive guidelines manual  for disposal of  FGD  sludges  has been
          published which provides detailed  information on available  technology;
          design, specification, and procurement; and operation  of  FGD sludge
          disposal systems.  The manual incorporates the results of the  earlier EPRI
          work described above.

Lime FGD Systems Data Book

The objective is to compile lime scrubbing design and operating  guidelines for
utility FGD applications.  The project is complete, and the  final report will be
available by mid-April.  All the FGD aspects investigated in the State of the Art
and the Subsystems Evaluation and Development subprograms over the  past two years
are included in the manual as they apply to lime scrubbing systems; additional
topics included are Process Design, Equipment Design, Procurement Procedures, Lime
Handling, Slurry  Preparation  and Corrosion.  Shortly after publication of the
report EPRI will sponsor a workshop to familiarize utility staff with the contents
of the guidelines and to elaborate on how they can be used most effectively.

SUBSYSTEMS EVALUATION AND DEVELOPMENT

The overall goal of this subprogram is to identify problems and develop solutions in
areas of technology which are not directed toward a single FGD process but are
related to wet scrubbing systems in general; such areas include mist elimination,
reheat, materials of construction, and novel gas/liquid contacting  configurations.
Recent important progress in this subprogram is summarized below.

Improved Lime/Limestone Scrubbing Technology

The objectives were to evaluate two innovative gas/liquid contactor configurations,
to investigate the concept of stack gas reheat, to correlate sludge properties with
scrubber design and operating conditions, and to establish an information base on
corrosion and erosion to aid in materials of construction selection.  The work was
carried out by TVA, including 1 MW testing at the Colbert Steam Plant.  Reports on
all these subjects have been issued or are in press, and most of these efforts have
already led to larger-scope follow-on research.  The major conclusions reached are
as follows:
                                      460

-------
S02 removal efficiencies in a pilot-scale horizontal  (cross-flow) scrubber
were primarily affected by liquid rate, gas velocity, pressure drop across
the slurry nozzles, and gas/droplet contact time.  In general, the removal
efficiencies were lower than had been expected, perhaps because the
gas/droplet contact time achievable in a commercial scale module could not
be simulated in the pilot unit.  Even so, under certain test conditions,
S02 removal efficiencies of over 90% were obtained with each of the two
reagents (lime and limestone) tested.

StX, removal efficiencies in a pilot scale cocurrent flow scrubber were
sensitive to where the scrubbing liquor was introduced into the scrubber
and the presence or absence of open (grid) packing in the scrubber to
improve gas and liquid distribution and increase the gas/liquid contact
time. Liquid rate generally was a more significant variable than gas
velocity in affecting SO2 removals.  These tests provided the basis for
design of a 10 MW cocurrent scrubber at TVA's Shawnee Test Facility.

Laboratory testing of the physical and chemical properties of sludges from
operating utility FGD systems showed that the physical form in which
calcium sulfite precipitates in a scrubber system depends on whether the
source of calcium is lime or limestone.  Calcium sulfite sludge from a
limestone FGD system consists of simple, open structure, tabular crystals.
In contrast, calcium sulfite solids from lime FGD systems are complex,
interpenetrating, spheroidal aggregates.  This study showed that the
settled bulk density of FGD sludges decreases with increasing solids
surface area, meaning that calcium sulfite sludges from lime FGD systems
dewater much less readily than those from limestone systems.

Capital costs for a 1 MW flue gas recirculation reheat system were about
75% higher than for an in-line indirect steam reheater, and the estimated
operating cost was about 9% higher.  However, elimination of several
potential operating problems inherent in an in-line system, such as
plugging, corrosion, and pitting of the shell side of the reheater tubes
due to moisture carryover from the scrubber, might justify the added cost
for the flue gas recirculation reheat approach.  Careful attention to
reheater operating conditions and effective mist elimination can minimize
                           461

-------
the potential problems with in-line reheat.

In conjunction with the in-line indirect steam reheater evaluation,
several materials of construction were tested for resistance to erosion
and corrosion.  Type 316 stainless steel and Incoloy 825 showed very good
resistance.  Incoloy 800; types 304, 410, and 446 stainless; U.S. Steel
alloy 100; and 18-18-2 alloy showed fair to good resistance to erosion,
but suffered from pitting and crevice corrosion.  Cor-Ten. alloys A and B
showed high rates of erosion/corrosion.

Advanced Flue Gas Desulfurization Development and Test Facility

The overall objective is to construct and operate 10 MW prototype scrubber
facilities to evaluate advanced scrubber concepts.  This is accomplished
at TVA's Shawnee Test Facility; EPRI currently supports the operation of
one of three 10 MW prototype FGD units, a cocurrent  scrubber system which
was substantially completed in September 1978.  Testing has been completed
using sodium carbonate, magnesium oxide, lime and limestone as reagents.
Reliability testing with limestone is in progress.

In the latter half of 1979, another of the Shawnee 10 MW systems will be
used to evaluate the Dowa Mining process; this is described below under
the Advanced FGD Without Recovery subprogram.  In 1980 EPRI hopes to see
one or more of the Shawnee prototype systems in use  as a utility staff
training center and a debugging tool to solve FGD operating problems at
full-scale units.  That effective use of the Shawnee facility for such a
purpose could improve the reliability of existing systems drastically is
suggested by the strong historical correlation between high FGD reli-
ability and the presence of well-trained operating and maintenance staff.

Other Projects

Additional efforts in progress or being initiated in the Subsystems
Evaluation and Development subprogram include:
                          462

-------
     0     Construction Materials for Wet Scrubbers
     0     Scrubber Generated Particulates
     0     Cyclic Reheat
     0     Entrainment in Wet 'Stacks

ADVANCED FGD WITH RECOVERY

The overall goal of this subprogram is to encourage the commercialization of
advanced SO2 control techniques which involve recovery of the sulfur in a marketable
form such as elemental sulfur or sulfuric acid.  The two major FGD approaches EPRI
is currently supporting in this area are absorption/steam stripping/RESOX and the
Aqueous Carbonate Process.

Absorption/Steam Stripping/RESOX tm

Several projects are directed toward the objective of demonstrating an advanced FGD
process featuring sulfur recovery without the need for a gaseous reductant.   The
combination of absorption/steam stripping with RESOX has the potential to eliminate
FGD sludge disposal problems and produce elemental sulfur at a cost comparable to
conventional lime/limestone scrubbing.  Sulfur oxides are absorbed in a buffer
solution and then steam stripped from the solution in a separate vessel.  Thus the
effect of absorption/steam stripping is to concentrate SO, from several thousand ppm
to 25-95% by volume, the balance being water vapor.  In RESOX, the concentrated SO-
stream is reduced to elemental sulfur by contact with a bed of crushed coal.
Absorption/steam strip chemistries are at various stages of development by about a
half-dozen suppliers, while RESOX is a proprietary process of Foster Wheeler  Energy
Corporation.  This particular combination of the two subsystems is not unique—that
is, each of them might be used in other combinations.  However, absorption/steam
strip/RESOX appears to be among the lowest cost sulfur recovery FGD options.   Recent
results achieved in EPRI research efforts are as follows.

     0     Laboratory-scale tests were carried out to establish vapor/liquid
          equilibria for three absorption/steam stripping chemistries.  The Flakt-
          Boliden sodium citrate buffer process was selected for pilot testing.
                                      463

-------
     0    Preliminary engineering is complete and long lead time equipment items
          have been ordered for a 1 MW absorption/steam stripping pilot test program
          at TVA's Colbert Steam Plant.

     0    A 42 MW prototype RESOX'plant was constructed and started up on schedule
          in West Germany.  The concentrated SO- feed stream was converted to
          elemental sulfur of over 99% purity, using German anthracite coal as the
          reductant.  One purpose of this work is to evaluate the applicability of
          RESOX to several SO^-concentrating front-end subsystems—Bergbau-Forschung
          activated carbon, absorption/steam stripping, magnesia and Wellman Lord.

     0    Two U.S. noncaking bituminous coals were shown in lab testing to be
          suitable for use as the RESOX reductant.  Caking bituminous coals are not
          suitable unless they have been pretreated to eliminate agglomeration
          characteristics.

Aqueous Carbonate Process

Two EPRI projects address the objective of demonstrating a sodium regeneration
subsystem applicable to sulfur recovery FGD.  Both are directed toward commercial-
ization of Rockwell International's Aqueous Carbonate Process (ACP).  Sulfur oxides
are absorbed in an aqueous solution of sodium carbonate.  The contacting device is a
spray dryer.  The dry particles of spent absorbent are collected by a baghouse or an
electrostatic precipitator and charged to a molten salt reducer which uses coal to
convert sodium sulfite and sulfate to sulfide.  The reducer and the subsequent
carbonation system which regenerates the absorbent sodium carbonate solution
potentially can be applied to other sulfur-recovery FGD processes such as Wellman
Lord and magnesium oxide scrubbing.

EPRI has expressed its intention, subject to Board of Directors approval, to parti-
cipate in a 100 MW demonstration of ACP at the Huntley Station of Niagara Mohawk
Power Corporation.  Empire State Electric Energy Research Corporation is the lead
agency for this demonstration; EPA and New York State ERDA are also participants.
Preliminary engineering for the demonstration is proceeding now.
                                       464

-------
One of the EPRl projects, cofunded by Niagara Mohawk, has the objective of obtaining
test data on a 5 MW pilot scale for use in the design of the 100 MW plant.  The
other project, preparation of a test requirements document, has the objective of
assuring that the technical requirements of all the demonstration project
participants are addressed in the final design of the 100 MW plant.

ADVANCED FGD WITHOUT RECOVERY

The overall goal of this subprogram is to develop and demonstrate scrubbing
processes which offer significant improvement over conventional FGD technology in
the areas of removal efficiency, reliability, sludge disposal,  and cost.

Sludge-Free Limestone Scrubbing

The Chiyoda Thoroughbred 121 Process accomplishes limestone dissolution, SO2
absorption, sulfite oxidation to sulfate, and by product gypsum thickening,  all in
one flue gas sparged reactor vessel.  The objectives of the project are:

     0    Evaluate the performance, control characteristics, operating flexibility
          and reliability of the CT-121 process at a prototype  scale.

     •     Determine the feasibility of disposing of forced oxidation gypsum by-
          product solids by stacking.

The results achieved to date are as follows.

     0    The 23 MW CT-121 prototype facility was constructed and started up (at
          Chiyoda expense) at Gulf Power Company's Plant Scholz.

     0    Laboratory tests established the feasibility of stacking the CT-121 by-
          product gypsum as a disposal technique.

     0    A 5 month evaluation of the CT-121 process is in progress.  SO2 removals
          have usually been over 90%.
                                     465

-------
     0    The gypsum stack has been established,  and monitoring of its strength,
          permeability, and leachate composition  is in progress.   So  far,  the
          physical characteristics of the  stack confirm  the  feasibility of stacking
          the CT-121 by-product.

EPRI and TVA are planning the evaluation of another sludge-free limestone-based
advanced process, the Dowa Mining Dual Alkali Process.   SO2  is  absorbed in a basic
aluminum sulfate solution.  Limestone is used to  regenerate  the aluminum values, and
forced oxidation results again in rejection of sulfur as a gypsum  by-product.  In
addition to the potential benefits from clear solution scrubbing and production of
gypsum to decrease disposal costs, a significant  attraction  of  the Dowa process is
its potential for retrofit to existing lime or limestone scrubber  systems.  The Dowa
evaluation will be at one of the 10 MW prototype  scrubber systems  at TVA's  Shawnee
Test Facility.  EPRI anticipates that results from this project will be available in
the spring of 1980.

Chemical Basis of Alkali Scrubbing

It is obvious that the performance of FGD systems generally, and SO2 removal
efficiency and system reliability in particular,  are directly related to the
chemistry of the scrubbing process.  Shortly after our last  report to this
symposium, EPRI initiated a project whose objective was to quantify several chemical
phenomena in lime and limestone scrubbing through the development of mathematical
models from fundamental chemistry and mass transfer principles, and fitting of these
models to the available data.  Six aspects of lime and limestone FGD chemistry were
the subjects of modelling efforts:  SO- removal efficiency, oxidation,  gypsum super-
saturation, gypsum subsaturation, solids quality, and alkali utilization.  This
project is nearing completion.  A conclusion which is already apparent  is that
sufficiently reliable and internally consistent data on which to base the models are
much scarcer than we had thought originally.  The implication of this is that
although substantial effort has been directed to  data collection in prototype and
full-scale scrubber systems, those data were not  taken under conditions or in a
manner such that they are useful in arriving at a fundamental understanding of the
chemical phenomena of scrubbing.  Rather than pilot plant testing of the models,
then,  the next step in the EPRI effort will likely be bench-scale studies to fill in
the gaps that have been identified as a result of this work.
                                      466

-------
Other Projects

Other efforts in progress or being initiated in the Advanced FGD Without Recovery
subprogram include:

     0    Design of High SO, Removal FGD Processes
     0    Cost of High SO2 Removal FGD Processes
     0    Spray Drying FGD Evaluations
     0    FGD Reagent Preparation

FULL-SCALE DEMONSTRATIONS

As a result of earlier evaluation studies and some of the pilot-scale efforts
described above, four advanced FGD processes have been selected for potential major
EPRI funding and participation through the 100 MW demonstration stage.  The
processes are Chiyoda Thoroughbred 121, Dowa Mining, absorption/steam strip/RESOX,
and Aqueous Carbonate.  The planned pilot-scale and prototype evaluations will be
complete for all except absorption/steam stripping and Dowa by mid-1979.  Therefore,
EPRI is now seeking utility sites for the full-scale demonstrations.

CONCLUSION

Since our last report to this symposium, EPRI's flue gas desulfurization research
effort has evolved from subprogram to full program status; has tripled in staff,
budget, and number of active projects; and has shifted from an initial mode of
surveying the state of the art and identifying problems, to one of developing,
testing, evaluating, and demonstrating solutions to those problems.

ACKNOWLEDGMENTS

The following organizations are funding portions of the EPRI projects described
above:

Bergbau-Forschung GmbH
Chiyoda International Corp.
Deutsche Babcock AG
                                      467

-------
Foster Wheeler Energy Corp.
Niagara Mohawk Power Corp.
Steag AG
Tennessee Valley Authority
UOP, Inc.

REFERENCES

1.   Morasky, T. M., Dalton, S. M., "EPRI's Flue Gas Desulfurization Program,
     Results, and Current Work,"  EPA Symposium on Flue Gas Desulfurization,
     Hollywood, FL, 1977.

2.   Baruch, S. B., EPRI Journal, April 1978, p. 44.

3.   Preston, G. T., EPRI Journal, September 1978, p. 36.
                                    468

-------
                                                                           4F
Title              CHOLLA STATION UNIT 1  FGD SYSTEM
                    5 YEARS OF OPERATING  EXPERIENCE
Authors             Stephen R.  Travis
                    Chemical Engineer, Technical  Services
                    Electric Operations
                    Arizona Public Service Company

                    Frank.  A. Heacock, Jr.
                    Manager, Mechanical Technical Services
                    Arizona Public Service Company
                               Abstract

The Cholla Unit 1 FGD limestone throwaway system has been in commercial
operation since December 14, 1973.  The operations of this system has
been characterized by high efficiency control of flue gas S02 and
particulate matter at a high, sustained reliability factor.   Trends
in O&M costs demonstrate sound initial engineering and O&M concepts.
Five years of operation give a data base of useful information on
system component reliability and maintenance trends.  Data available
for performance of materials of construction in severe service is
presented to support the use of corrosion resistant alloy materials.

Based on the data and trends presented, extrapolations are discussed for
process capabilities and for design and selection of equipment by generic
type for reliable operation and acceptable maintenance.
                                  469

-------
                        CHQLLA STATION UNIT 1  FGD SYSTEM
                         5 YEARS OF OPERATING EXPERIENCE
INTRODUCTION

The FGD system for Unit 1  at Choi la Station of Arizona Public Service marked
five years of continuous service in December,  1978.   For the entire period,
this double-loop, limestone, throwaway FGD system has operated at a high re-
liability factor and with high efficiencies both in  required flue gas S02
removal and particulate removal.

The FGD system for this 115 MW generating unit consists of two parallel  gas
cleaning trains.  Each gas cleaning train consists of a high energy Flooded
Disc type venturi scrubber, and a  wetted-film  contact absorber.

Detailed descriptions of the FGD system components and process have been pre-
sented previously by L. K. Mundth, 1974, ^',  PEDCo  Report 1978 ^, and in
various trade journal articles.  For purposes  of this paper, only a brief
description will be given.  Reference is made  to Figure 1  for a schematic
presentation of the components and process.

The FGD system was a retrofit to the draft system of the unit and, therefore,
takes suction from the discharge of the existing ID  fans.   These fans are
preceded by multicyclone mechanical dust collectors.   The existing breeching
to the unit chimney forms a full bypass alternate for the FGD system.

As Figure 1 indicates, flue gas flows thru the FDS treatment loop for removal
of particulate materials and partial removal of S02  gases.  The flue gas then
passes vertically upward thru the combined cyclonic  entrainment separator/
wetted-film absorber tower, thru the two stage mist  eliminator, vertically
downward thru the reheater and exhausts into the chimney.
                                   470

-------
         Figure 1.   CHOLLA UNIT 1 WET SCRUBBER SYSTEM
                                         REHEATER
TO
ASH DISPOSAL-*
POND
      MAKE-UP H20
                        D
                   Q-

a
LIMESTONE FEED 	 1 t
^_— — - — -T
^j

Jl
H

                                                             SECONDARY MIST ELIMINATOR
                                                              DEMISTER WASH
                                                             PRIMARY MIST ELIM.

                                                             -ABSORBER PACKING
                                                            H-MAKE-UP H2O
                     FDSSLURRY
                       TANK
ABSORBER TOWER
   FEEDTANK

-------
Approximately 25% of the flue gas S02 is absorbed in the FDS loop in each
train along with 99+% of the participate matter.   Only train A is fitted
with the wetted-film absorber packing and, in this train, the combined
overall removal of S02 is designed for 92%.

Table 1 contains some pertinent data as reported  in Table 4 of the PEDCo
Report, 1978 ^' (by Permission).   The data  relates to hold tank design
but is at least introductory for purposes of this paper.

The flyash/sludge wastes are currently bled  from  the FDS loop and pumped for
ultimate disposal in a new evaporative disposal  area for the station.   The
FGD system operates on an open-water-loop basis with no recycle of water from
the pond.  Fresh water make-up to the FGD system  comes from the unit cooling
water lake or from deep wells.

The sustained success of this FGD application is  in great part dependent upon
the unique factors of waste disposal and open-water-loop operation and under-
writes the argument that FGD system applications  must be site specific and that
generalizations of successful experience at  one  site may not be viable for
another.
                                  472

-------
                     Table 1.
   DATA SUMMARY:   FGD SYSTEM HOLD TANKS

Total number of
tanks
Tank sizes
Retention time at
full load
Temperature
PH
Solids concentration,
Flooded disc
scrubber
hold-up tank
One
3.8 m (12.5 ft)
dia. x 4.3 m
(14 ft)
7 min
49°C (121°F)
5.2
15.5
S02 absorber
towers
hold-up tank
One (common)
8.3 m (27.3 ft)
dia. x 8.5 m
(28 ft)
5 min
49°C (121°F)
6.5
8.3
FGD system
sludge
hold-up tank
Two
5.6 m (18.5 ft)
dia. x 8.2
(27 ft)
14 hr
49°C (121°F)
5.2
25
Limestone
slurry
make-up tank
Two


32°C (90°F)

20
  percent
Specific gravity
1.102
1.049

-------
PERFORMANCE

Initial FGD system performance testing was conducted in October, 1973.
Table 2 contains the operational  data and performance results of this
testing.  This is as reported by the PEDCo report,  1978 ^' (by permission)

As shown in Table 2, the S02 removal efficiency during the initial  testing
indicated 92.4 percent in the A-side and 14.4 percent for one of two test
sets in the B-side for a combined average of 53.4 percent.  Also, the parti'
culate removal efficiency was 99.7 percent in the A-side and 99.8 percent
in the B-side.

In October 1977, further testing at the FGD outlet was conducted which  indi-
cated a combined average S0£ removal efficiency of 43.0 percent and a com-
bined particulate removal efficiency of 99.75 percent.  For these October
1977 tests, inlet loadings were estimated from coal  date.

OPERATION

Reliability

The reliability of the FGD system has been consistently high as shown in
Table 3.  During the period from January, 1974 to November 1978, the relia-
bility of the A-side averaged 93.0 percent while the B-side averaged 91.4
percent for a combined average of 92.2 percent.  In this context, relia-
bility is defined as the hours the FGD system was operated divided by the
hours the FGD system was called upon to operate.
                                  474

-------
                 Table  2.     RESULTS OF FGD SYSTEM PERFORMANCE TEST RUNS,
                                    OCTOBER 2 to 21, 1973
                                             A-side            B-side              B-side
Particulate concentration  inlet,
 g/iii3  (gr/scfd)                           4.569 (1.995)     5.810 (2.537)

Particulate concentration  outlet,        0.0190 (0.0083)   0.0231 (0.0101)     0.2631  (0.1149)
 gr/scfd

SOg concentration  outlet,  ppm                  34               357                 236

SC>2 concentration  inlet                       417               409

Configuration                                Packed            Hollow               Hollow

SO? removal, percent                         92.4              14.4                 9.2

Particulate removal  efficiency,
 percent                                      99.7              99.8

Gas inlet  to FDS.  m3/sec  (acfm)           96.9 (214,300)    96.6 (204,600)      96.4 (204,300)

Theoretical inlet  gas  to  FDS,
 nrVsec  (acfm)                            93.8 (198,000)    93.8 (198,800)      93.8 (198,800)

Apparent bypass  leakage,
 m-Vsec  (acfm)                                              7.98 (16,900)

FDS L/G ratio,
 liters/m3 (gal./lOOO  acf)                1.35(10.1)       1.42(10.6)         0.78(5.8)

Tower  L/G  ratio,
 Iiters/m3 (gal./lOOO  acf)                6.5 (48.9)

(continued)
                                             475

-------
                                        Table  2  (continued)
                                               A-side
                      B-side
                       B-side
  Gas  velocity  through  tower,
  m/sec  (ft/sec)

  Mist entrainment  from tower
  g/nr (gr/scf)

  Solids  entrainment  from  tower
  slurry g/m^  (gr/scf)

  Pressure drop FDS,  kPA (in.  H20)

  Pressure drop tower demisters,
  kPA (in H20)

  Pressure drop reheater,
  kPa (in.  H20)

  NA -Not applicable.

  Temperature tower outlet °C  (°F)

AT reheater °C  (°F)

  Mist eliminator wash  water rate,
  liters/sec  (gpm)

  Slurry  flow to  FDS,
  liters/sec  (gpm)

  Slurry  flow from  FDS,
  liters/ sec  (gpm)
 2.10 (6.9)


    0.000

0.011 (0.005)


 3.7 (14.8)


     0.0


 1.3 (5.15)
2.05 (6.6)


   0.000

    NA


3.9 (15.7)


    0.0


0.6 (2.30)
2.05 (6.6)


    NA

    NA


3.9 (15.7)
49 (121)
36 (65)
0.8 (12.5)
137 (2170)
83 (1317)
49 (121)
33 (60)
0.9 (14.0)
136 (2170)
94 (1486)
49 (121)
33 (60)
0.8 (14.0)
88 (1400)
NA
                                               476

-------
                Table 3.     YEARLY  AVERAGE  RELIABILITY
                             FACTORS  FOR  CHOLLA  FGD
          Period
     Reliability, percent

Module A       Module B
                                                            System Avg.
Jan.
Jan.
Jan.
Jan.
Jan.
74
75
76
77
78
- Dec.
- Dec.
- Dec.
- Dec.
- Nov.
74
75
76
77
78
94
91
89
93
98
88
85
89
97
98
91
88
89
95
98
                  Table 4.
 COST DATA FOR CHOLLA FGD SYSTEM
                      Period
                    (year ended
                 December 31  unless
                  otherwise noted)

                        1973
                        1974
                        1975
                        1976
                        1977
                        1978
                 (Jan.  1  to Oct.  30)
             Total  Operating
                   and
            Maintenance Costs
                $  74,600
                  627,800
                  339,000
                  363,500
                  359,000

                  313,998
NOTE:   Includes:   Operating  labor and  materials,  maintenance  labor and materials,
       limestone,  and  sludge disposal  energy.

       Excludes:   Fuel  differential  charges  and capital  investment charges
                                  477

-------
Cost
The FGD system costs shown in Table 4 include operating labor and materials,
maintenance labor and materials, limestone, and sludge disposal.   However,
these costs do not include fuel differential  charges and capital  investment
charges.

Based on the original FGD system cost of $6.5 million, the capital  invest-
ment charge amounts to about $1.5 million annually.

For a 115 MW unit this equates to an installed cost in 1973 of $56.5/KW.
Currently, unit costs are estimated to be 1.5 to 2.0 times this cost with
a resulting substantially higher annual  capital charge.  It should  also be
noted that capital charges for the new disposal facility, shared  with Units
2, 3 and 4, are not included in the $6.5 million cited.

As has been pointed out by L. K. Mundth 1974  '"', the FGD system  requires
auxiliary supply of 2.8 MW of electricity and 18,000 pounds per hour of
steam for reheat.  These requirements are operational penalties reflecting
in cost at the bus bar.  Fuel differential charges are also properly
assessed and average about $.8 million annually.

Maintenance Philosophy

During June, 1975, a preventative maintenance program was initiated on the
Choi la I FGD system.  As a part of this preventative maintenance  program,
the maintenance records of the FGD system components were analyzed  to deter-
mine the frequency and nature of system component failures.

Based on this analysis, high maintenance components were identified and
placed on a routine maintenance schedule.  By this means, the number of
emergency maintenance situations was reduced.  Consequently, this also
allowed a reduction in the overtime maintenance requirements.

This procedure has proven a  successful approach as evidenced by the sub-
sequent history of high reliability that the system  has experienced.
                                   478

-------
SYSTEM COMPONENT ANALYSIS

Initial Major Problems and On Going Solutions

Following the initial startup of the Cholla I FGD system, several  mecha-
nical and chemical  problems' were encountered.  The mechanical  problems
encountered during this shakedown period were as follows:
     1.   Vibration in the reheat sections due to improper flue gas distri-
          bution.
     2.   By-pass damper leakage due to distortions caused by  flyash build
          up and prolonged exposure to high temperatures.
     3.   Malfunction of the flooded disc position control caused  by binding
          due to build up around the disc shaft.
     4.   Booster fan leaks due to improper welding.
     5.   Erosion of the stainless steel pump impellers and liners.

In general these mechanical problems were resolved by means of equipment and
operational modifications.

The excessive reheater vibration was eliminated by the installation of
baffles in the duct work to improve the flue gas distribution.

The bypass damper leakage problem was alleviated by reducing the FGD system
pressure drop so that a small amount of treated flue gas would flow backward
through the dampers thereby reducing flyash buildup on the dampers.

However, no good solution was found for the erosion problems encountered by
the stainless steel pump impellers and liners.  These items must be replaced
every six months.

In addition to the mechanical problems specified above, the following chemi-
cal problems were encountered:
     1.   Corrosion of the reheater.
     2.   Corrosion of expansion joints.
     3.   Failure of protective coatings.
     4.   Scaling of the first stage mist eliminator.
     5.   Scaling in the tangential nozzle area of the flooded disc scrubber.
     6.   Scaling in the flooded disc differential pressure sensing lines.

                                  479

-------
The reheater corrosion problem was the result of condensation in the duct-
work leading to the reheater.  This acid run-off caused tube necking at the
reheater sheet resulting in tube failure.  This problem was corrected by
insulating the ductwork to prevent condensation.

The original metal expansion joints were replaced by a rubberized fabric
type.  However, it has been found that rubberized expansion joints wear out
as a result of continued flexing.  A complete solution to this problem has
not yet been found.

Likewise, the protective coating failure problem has not been satisfactorily
resolved.  Although it was originally believed that the initial  coating
failures that occurred were due to improper coating application, subsequent
reapplications of coating have continued to be unsuccessful on the B-side
ductwork downstream from the reheater.

Scaling of the first stage mist eliminator has been controlled by redesigning
the mist eliminator wash system for better coverage and frequency of washing.

However, solutions have not been found for scaling in the tangential nozzle
area of the flooded disc scrubber or the flooded disc scrubber differential
pressure sensing lines.  The differential pressure sensing lines have been
relocated to allow for more convenient access since these lines must be
manually unplugged about every two weeks.

It has been observed that scaling problems tend to increase if the FGD system
pH level is allowed to drop below pH 5.0.  This is believed to result from
increased oxidation of sulfites to hard-scale forming sulfates.

The pH control system has been improved by converting from the original
in-flow type slurry sampling devices to a still-well type which eliminated
sample line pluggage problems.

As the above discussions indicate, many improvements have been made on the
original Cholla I FGD system.
                                   480

-------
EXTRAPOLATIONS FOR LIMESTONE THROWAWAY SYSTEMS DESIGN AND SELECTIONS

Process Loop Separation

One of the salient features of this system is the process loop separation.
Arizona Public Service feels that the separation  is  dictated  by the require-
ment for isolation of particulate removal  from the principle  S02 removal
step.

This double-loop design affords some important advantages for its operation
as a FGD system, both from the standpoint  of process effectiveness and,
also, from the standpoint of application of materials of  construction.
As has been pointed out by Braden,  1978   ',  separation  of  the  FDS (quench)
loop from the absorber loop is important for  isolation of the higher  chlorides
to the recycling slurry of the quench loop.   This  separation affords  a more
cost-beneficial  selection of corrosion-inhibiting  materials of  construction
between the two  loops.  Recent analysis indicate that the chlorides concen-
tration of the quench loop 5 times  the concentration of  the absorber  loop.
In a later application of an FGD system at this  station, chlorides in the
quench loop are  expected to rise to 10,000 to 14,000 ppm with application of
cooling tower blowdown as make-up water and a tight station water  balance.
High molybdenum  steels are indicated for use  in  this area with  special atten-
tion paid to areas subject to abrasion.

Regarding reagent utilization, APS  feels that the  benefits  of the  double-loop
design pays dividends.  Operation of each loop at  different and discrete pH
levels provides  for open-loop operation in the absorber  tower and  maximizes
the full  utilization of the reagent in the quench  loop.
                                  HOI

-------
Open-loop operation of the absorber affords good scale control as well as
mist eliminator cleanliness conditions.  Isolated or closed-loop operation
of the quench loop affords control  of the rapid oxidation for enhancement
of disposal products characteristics and is readily accommodated in the
quench loop.  Recent pilot tests on the Cholla 1 FGD system by Research-
Cottrell gives an indication of potential for full  oxidation by air sparging.

Materials of Construction

Liberal use of corrosion resistant metal alloys as  compared to coated carbon
steel construction characterizes the Cholla Unit 1  FGD system.  The FDS
scrubber, the absorber tower, and the reheater are  316L stainless steel
construction based on results of the initial  pilot  test.   Results of sub-
sequent testing of materials in the system during operation is reported by
Brodsky and Paul, 1975 '^', and indicate that extremes of service require-
ments are present.  In the one extreme, only high-molybdenum alloys (over
6 percent) showed no local corrosion in the area of liquid-gas separation.
In the least severe, most thoroughly washed mist eliminator area, only the
sensitized 316 and 304 stainless steel  evidenced wastage  and pitting.  Good
service in the presence of acidic and chloride corrosion  attack can be real-
ized through use of nickle-based, high molybdenum alloys  or with proper atten-
tion to the molybdenum content of the stainless steels and proper fabrication
techniques.

We have had mixed success with the integrity of the remaining parts of the
system which are constructed of carbon steel  with corrosion resistant glass
flake polyester-resin based linings.  One disadvantage found for these
materials is the lower tolerance to abrasion.
                                   482

-------
The specification of alloy metal  vs rubber-lined process pumps has varied
within APS for F6D systems depending upon the specific requirement of the
application.   All of the process  pumps for the Cholla Unit 1  F6D system are
alloy metal pumps, although APS experience with pump materials at our Four
Corners Station demonstrated 500-10,000 hours of useful  life  for rubber-lined
equipment, whereas all  trial alloy metal  pumps failed in 1000-1400 hours of
service.

APS has found no panacea for selection of pump materials and  certainly least
of all first cost.  Careful selection of available designs to satisfy the
following major criteria is essential to good long-term satisfaction :  (1)
slurry abrasiveness, (2) combined corrosion/erosion mechanisms, (3) chloride/
pH factors, (4) head limitations, and (5) seal water requirements.

The state of the art design and experience with selection of  materials for
in-line reheaters in the pioneer  era of the Cholla Unit 1 FGD system was
basically experimental.  The 3161 stainless steel shell-and-tube reheaters
of this unit have been satisfactory.  Two factors have contributed to the
success that has been achieved, i.e., split coil construction and adequate
cleaning procedures.  Contrariwise, reheaters of the same materials failed
at our Four Corners Units 1, 2, & 3 application, because of difficulties
arising from the in-place cleaning of unsectionalized coil reheater con-
strucion.

In this regard, it should be noted that in the Cholla Unit No. 2 FGD system we
have taken the plunge into  Inconel 625 reheater materials for better long-term
performance.
                                  483

-------
CLOSURE

In closing, there are two compelling observations regarding FGD system appli-
cations in general  which is rooted in the Cholla experiences and which should
be emphasized.

Cost-effectiveness  for FGD systems which must be installed is heavily dependent
on experienced design with emphasis on careful  specification of process design
and of materials selection for minimizing future operational and maintenance
costs.
                                   484

-------
REFERENCES

(1)  Operational  Status and Performance of the Arizona Public Service Flue
     Gas Desulfurization System at The Choi la Station.  L.  K. Mundth
     November 4,  1974

(2)  Survey of Flue Gas Desulfurization Systems:   Cholla Steam Electric Station,
     Arizona Public Service.
     PEDCo Environmental, Inc.      March 1978
     Contract No. 68-01-4146
     Task No. 30

(3)  Double-loop Operation Offers "Best of Both Worlds" Approach To Sulfur
     Diovide Scrubbing.  Herbert H. Braden Public Utilities Fortnightly.
     August 17, 1978

(4)  Corrosive Properties of An S02~-West Limestone Scrubbing System For  A
     Coal-Fired Power Plant.
     I. S. Brodsky       G. T.  Paul     April 1975
     Presented to NACE Annual Meeting   (unpublished)
                                  485

-------
   LA CYGNE STATION UNIT NO.  1
WET SCRUBBER OPERATING EXPERIENCE
              by
          Terry J. Eaton



Supervisor of Air Quality  Control



         La Cygne Station



   Kansas City Power  &  Light  Co.
    Prepared For  Presentation
       A*
                at
          EPA CONFERENCE
         LAS VEGAS,  NEVADA




        MARCH  4  -  8,  1979
                 486

-------
                              INTRODUCTION
     This paper reviews the present operating experiences of the Babcock &
Wilcox designed scrubber system and continues to describe the trends of costs,
availabilities, modifications, manpower and other supportive data relating to
operations since June, 1973.
     Early operations proved the need for protective walls to house the modules;
heavier rotor blades and improved bearings for the induced draft fans drafting
both the boiler and scrubber; cyclone separators to prevent scale and debris
from plugging slurry nozzles; developing fast scale removal techniques and
systems to remove huge piles of debris; constant surveillance to repair internal
corrosion damage; learning the economical trade-off for exotic materials to
withstand abrasion, corrosion or violent operation; establishing the need for
minimal instrumentation for consistent operation due to terrific maintenance
requirements; and undoubtedly the most important accomplishment was to establish
an operating force that proved a scrubber burning coal with a very high sulfur
and ash could be made to work effectively.  The major problems currently
affecting the scrubber system are corrosion and lack of commercially available
instrumentation to monitor critical parameters affecting the scrubber operation.
                                 487

-------
                            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 and Light Company.  The companies share equally in owner-
ship and output and the unit is operated by KCP&L.  The 630-megawatt No. 2
Unit, in service since being declared commercial May 15, 1977, operates 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  to 6  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.  The turbine-
 generator was  supplied by Westinghouse  and is rated at 874 MW gross output  with
 five per cent  overpressure and 3,500 psi throttle pressure.   Three auxiliary,
                                     488

-------
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 eight two-stage Venturi-absorber scrubber
modules  (Exhibit B) designed to treat the boiler flue gas flow of 2,760,000
ACFM.  (345,000 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 S0? 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  buoyancy from the stack.
                                     489

-------
OPERATING EXPERIENCE

     As a result of the continuing modifications and improved operating
procedures, the module availabilities have steadily improved.  The annual
averages (Exhibit E) have been 31% for 1973; 76.3% for 1974; 84.3% for 1975;
92% for 1976; 92.5% for 1977; and 93.5% for 1978.  With the addition of the
eighth module in April 1977, continuous daytime load capability has exceeded
800 megawatts without appreciably affecting average module capability.
     The results of a full load and stack emissions test on August 26, 1977,
(Exhibit F) indicated module gas flow was still below crusing capability,
the induced and forced draft fans were loaded well below rating and most
systems were in good balance.  Sulfur dioxide removal efficiency averaged 77%
with individual modules averaging from 65 to 80%.  Although particulate emis-
sions from the plant have met EPA and Kansas State requirements, research and
development work continues in an endeavor to reduce further the particulate
emissions from Unit #1.
     The ambient monitoring system continues to indicate ground level concen-
trations within the national standards for sulfur dioxide and nitrogen com-
pounds  (Exhibit H).
     Limestone utilization has greatly improved with improved Ph control.  In
the past, it has been almost insurmountable to maintain inline glass cells
without caking the limestone during shutdown or abrading the cells during
operation wit-^ th.2 high concentration of fly ash.  By the proper maintenance
discipline of acid flush, sonic cleaning and periodic water backflush,
"straight line" Ph is resulting in approximately 30% less limestone, better
control of scaling and has eliminated one more variable which hinders analysis
in other areas.
     Demister pluggage or scaling is no longer a problem at La Cygne.  By
eliminating the intermittent wash and moving the continuous wash  (140 GPM)
from below to above the first demister with increased number of nozzles
(230 GPM), the chevrons operate "squeaking  clean".  Further experimentation
may allow a reduction in these nozzles and  perhaps sequential washing to reduce
excess water.
                                 490

-------
     Hard scale on the reheater tubes has been eliminated by the addition
of a second layer of demisters in each of the modules.   Scaling of the
reheaters continues to be a problem, however it is soft and can be removed
using fire hoses.  The previous hard scale required high pressure water to
remove the deposits.

MAINTENANCE
     Cleaning schedules continue to call for taking one module out of service
each night on a rotational schedule and keeping all modules available for
the daytime peak loads.  This allows a complete checkout of module internals
to clean steam reheater pluggage, check nozzles for debris or loose rubber
pluggage, to clean sump accumulation and to inspect for any other maintenance
that could reduce reliability during the week.  Module inspection and cleaning
is not reduced to six hours or less with reheater pluggage the greatest problem.
Water soot blowers may be the answer to cleaning on the line since steam blowers
will help scale the carryover on the steam reheat tubes.  Scaling is not one of
our chief problems and we ordinarily ignore soft scale that forms on walls, on
beams, or on the outside of nozzles.
     Carryover to the induced draft fan blades continues to require regular
washings.  Each fan now requires cleaning once every four to seven days.  Seldom
is the high pressure wash necessary any longer, a "spinning" process with low
pressure hoses has been very effective, cleaning the spare fan while out of
service.  The washings are usually done on a preventative basis, but must be
taken out of service if bearing vibrations exceed 12 mils.
     Rubber pipe linings and rubber-lined pumps have been an increasing main-
tenance problem.  After several years operation, some materials that haven't been
modified are wearing out.  Rubber linings that tear out cause damage in other
piping or pumps, plug nozzles and allow the steel pipes to wear through.  Two or
three years ago, this problem would not have been classified as serious, but
this very abrasive slurry in practically continuous operation can be detrimental
in trying to attain higher module availability.
     Corrosion of carbon steel in the ductwork, dampers, induced draft  fan
rotors and housings, breeching and stack liner is and will continue to  be our
                                 491

-------
greatest concern.  Burning extremely high sulfur coal and having the outage
problems of a large unit creates periods of enormous SO  concentrations on
these surfaces.  This "cold end corrosion" damage requires extreme surveillance
by maintenance engineers and unit outage p3.ans must consider temperatures and
time requirements for applying special coatings.

MANPOWER REQUIREMENTS
     The scrubber operating and maintenance force is being increased to 54
people by adding one electrician for a total of two and two technicians for a
total of three.  The remaining personnel will remain the same (Exhibit I).
The  continued  improvements in operating procedures and stable equipment opera-
tion should permit meaningful analysis of improved chemistry and control para-
meters .  If the  current effort to maintain Ph cells and S02 analyzers under
challenging conditions  are any indication, it will definitely require this
increased force  to make farther progress.
     Also worth  noting  are the increased  demands on present maintenance personnel
to  accumulate, record and evaluate  operating data on water saturation trends,
limestone utilization,  draft  fan wear rates, reheater bundle failures, lined
pump failures, rubber lined pipe replacements,  nozzle replacements,  spare  parts,
etc. The  operators  are also  busy updating and  extending  operating instructions,
special instructions  and  reviewing  safety and training procedures.

COSTS
     The total cost  (Exhibit  G)  of  the  AQC system to  date has  increased  to
 $46.8 million or about  22 per cent  of  the $213  million  total Unit  #1 cost, or
 about  $59  a kilowatt installed.   It is  estimated that  an additional $4 to $6
million investment will be required to reach optimum system performance.
      La Cygne Unit No.  1  production costs for  1977 for  energy  including coal
 costs  average 6.54 mills  per  K.W.H.  Production costs for the scrubber portion
 average 1.69  mills per  KW.   Discounting escalation, scrubber costs of labor and
 limestone are trending  downward but maintenance materials have increased threefold.
                                     492

-------
     Although the La Cygne Unit #2 has been commercially available since
May 15, 1977, it is still too early to make cost comparisons between a
scrubber system with local high sulphur - high ash coal and a precipitator
only system burning Wyoming coal with greater transportation costs.  Unit
#2 has had a fantastic service record with 96% availability and 76 to 84%
monthly load factors.  It begins to appear that the local coal will be
the most economical operation if probable western coal escalations are
considered and installed costs of scrubber vs precipitator are not considered.

ADDITIONAL MODIFICATIONS
     1.  An improved steam source to increase the supply for module reheater
service from 70,000 LB/HR to 120,000.  This would then permit additional steam
bundles for optimum reheat.
     2.  An additional sludge pond for deposit of scrubber spent slurry for
approximately 30 years.  A side benefit could be clear water recycled to the
scrubber for improved chemistry.
     3.  Evaluate addition of third demister.
     4.  Make study on sub micron fly ash and sulphuric acid mist passing
through scrubber without being collected.  Although most scrubbers do not have
required pressure drops for the duty, wetting agents, fogging arrangements or
other  developments could lead to a vast improvement.
     5.  Continue work to devise a better method to clean incline reheat tubes
without taking equipment out of service.
                                 493

-------
Proximate

Volatile
Fixed Carbon
Ash
Moisture
 28.63
 37.9^
 2U.36
  9.07

100.00
BTU/lb.
Grindability   59.59
 Analysis

 Phosphorous  Pentoxide
 Silica
 Ferric Oxide
 Alumina
 Lime
 Magnesia
 Sulfur ^rioxide
 Potassium Oxide
 Sodium Oxide
 Titania
 Other
         0.15
        U6.05
        19.23
        1U.07
         6.86
         1.02
         7.35
         2.U8
         0.60
         1.02
         0.67
                       100.00
                                 LA CYGNE STATION
                               COAL AND ASH ANALYSIS
                                        COAL
                                         ASH
Ultimate

Moisture
Carbon
Hydrogen
Nitrogen
Chlorine
Sulfur
Ash
Oxygen
  8.60
 51.93
  3.^3
  0.027
  5.39
 2^.36
  5.33

100.007
                                        Fusion Temperature
Reducing I.D.
   Soft  (H=W)
   Soft  (H«W/2)
   Fluid
Oxidizing  I.D.
   Soft  (H=W)
   Soft  (H=W/2)
   Fluid
  1957
  20U5
  2169
  2321
  2156
  2338
  2U15
  2520
                                       Exhibit A
                                         494

-------
     w
     X
vo
Ui
     H-
     ft

     dd
                                La Cygne limestone wet scrubbing system

-------
                     FIGURE   I  -  LACYGNE   FGD  MODULE
   REHEAT
   STEAM
    750'F
  FLUE
  GAS
                WALL
                WASH
                (32)
                                                                      CONTINUOUS
                                                                       WASH
                                                                      200 GPM
             VENTURI
              SPRAY
               (48)
              PREOEMISTER
                   f\ f\  f\ r\ f\

             4. 14. J_L UL JLL U. JJL U. J_l i.

                           JUJL
                                                               ABSORBER
                                                                SPRAY
                                                               SIEVE
                                                               TRAYS
                                                                   FEED
                                                                  SLURRY
RECIRCULATION  TANK
 CACOi
 CA SOa
 CA SO4
FLY ASH
              46G/L
              SOc/L
              I 6 G/L
              30G/L
 SPENT SLURRY
  TO  POND
  700  GPM

 3500 TONS/ DAY*
693000 TONS/ YEAR
453 ACRE  FT./YEAR
                VENTUR
              RECIRC. PUMP
               6000  GPM


  *  TOTAL  FOR ALL  MODULES
   Exhibit C

       496
                                     ABSORBER
                                 RECIRC. PUMP
                                   10000 GPM

-------
                                                             1-10-78
                      LA CYGNE SCRUBBER WATER ANALYSIS
CATIONS



CALCIUM (Ca)



MAGNESIUM (Mg)



SODIUM (Na)



POTASSIUM (K)






ANIONS



BICARBONATE ALK (AS HC0



CHLORIDE (Cl)



SULFATE (S04)



SULFITE (S03)



SILICA (Si02)






OTHERS



pH (pH UNITS)



CONDUCTIVITY IN MICROMHOS



SOLIDS, SUSPENDED



        DISSOLVED
COOLING
LAKE
126.4
16.3
31.0
5.1
03) 112.2
44.9
295.2
* ND
1.12
7.7
HOS 820.0
5.0
610.0
SETTLING
POND
808.0
106.0
52.5
41.6
79.3
314.0
1995.1
* ND
52.0
7.5
3500.0
5.0
3450.0
*ND - Not Detected
                                   Exhibit D
                                      497

-------
                                                    MODULE AVAILABILITY SUMMARY -  1973
oo
MONTH

JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
JULY
AUGUST
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
i A






20
7
79
13
28
48
42
B






21
24
64
0
41
1
20
C






40
25
65
13
34
38
5 .
D I E
1





21
41
74
13
54
4
31 .





27
27
47
13
33
63
26
F






30
25
48
0
3
59
11
G






23
31
70
0
46
49
32

AVERAGE %
AVAILABILITY*






26
26
64
7
34
37
24
31%
MWH






87,529
90,669
250,319
20,073
117,106
104,255
61,013
BOILER
HOURS






294
303
699
95
452
463
339

GENERATION
LOAD FACTOR






15.2
15.2
42.1
3.5
19.7
18.1
10.3
17.7%
      * MODULE HOURS
        HOURS IN MONTH
                                                                 Exhibit  E

-------
                                              MODULE AVAILABILITY SUMMARY - 1974
MONTH
JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
£ JULY
10
AUGUST
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER

A
49
66

67
69
92
75
90
69
71
90


•
B
32
68

70
83
84
80
90
88
61
71


C
44
59

75
78
83
80
73
73
59
60


1)
87
76

88
85
90
81
81
76
81
61


E
23
52

74
78
82
85
81
83
79
84


F
37
100

100
84
83
79
78
89
93
85


G
81
65

88
80
87
77
99
86
89
84


AVERAC7E .%
AVAILABILITY*
50
69

80
80
86
80
85
81
76
76

76.3%
MWH
35,862
85,256

83,880
157,949
185,473
110,122
231,382
209,127
230,302
130,128


BOILER
HOURS
364
364

332
500
480
313
571
606
662
386


GENERATION
LOAD FACTOR
6
16

15
27
32
19
39
36
39
23

25%
*MODULE HOURS
 BOILER HOURS
                                                     Exhibit E (Cont'd)

-------
                                               MODULE AVAILABILITY SUMMARY




                                                      LA CYGNE 1975
MONTH
JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
JULY
AUGUST
SEPTEMBER
OCTOBER

NOVEMBER

DECEMBER
A


82. A

94.6
87.8
78.4
74.64
78.43
66.16

92.87

90.72
B
Turbin
Turbin
96.03
Genera
85.1
85.4
89.7
88.07
83.62
77.26
General
90.79
General
87.39
C
a Gener
2 Gener
89.5
:or Rep
94.2
83.9
89.6
87.29
84.38
46.27
:or Rep;
80.18
or Reps
80.87
D
at or Re
ator Re
76.6
lir 25
89.5
84.9
83.7
78.01
84.67
73.62
dr 15 I
93.18
iir 17 I
85.20
E
>air
>air
92.96
)ays
89.8
84.1
85.4
92.44
78.72
71.91
)ays
96.09
ays
86.89
F


91.5

89.3
86.1
87.4
85.00
77.71
73.07

89.39

G


96

83.4
88.6
85.2
83.06
74.24
64.69

93.94

88.56||83.67
*Working Hours + Reserve
AVERAGE
AVAILABILITY*


89.33

89.4
85.8
85.6
84.07
80.25
67.57

90,83

86.19
84.3

MWH

7,886
244,873
23,014
332,526
324,952
297,870
294,402
239,954
74,660

165,058

278,597
BOILEI
HOURS


694

683
667
590
630
610
231

346

597

GENERATION
LOAD FACTOR


41.1
3.4
55.9
56.4
50.0
49.5
41.7
12.5

28.7

46.8
38.6

Hours in Month
                                                  Exhibit E (Cont'd)

-------
                                           MODULE AVAILABILITY SUMMARY




                                                  LA CYGNE 1976
MONTH
JANUARY
FEBRUARY
MARCH
APRIL

MAY

JUNE
JULY
AUGUST

SEPTEMBER
OCTOBER

NOVEMBER
DECEMBER

A
85.8
93.9
92.3
92.3

96.5

93.3
95.6
94.1


97.4

94.7
86.8
B
84.6
90.3
89.7
90.5
Schedi
92.1
Schedi
94.1
95.0
93.1
Turbir
Turbii
96.7
Turbii
93.3
88.5
C
90.7
85.8
88.4
88.7
iled Ou
93.5
iled Ou1
94.0
91.9
91.8
le Repa:
te Repa:
97.5
te Repad
93.7
81.0
D
71.8
91.2
93.0
97.1
:age 24
95.7
:age 9 :
95.0
92.9
93.4
r, Sta<
.r, Sta<
89.0
r, Stac
95.3
93.5
E
83.9
91.7
94.2
95.8
Days
89.4
Jays
92.3
93.0
91.8
:k Relir
:k Relir
96.1
k Relin
94.2
93.6
F
82.3
93.1
91.3
98.0

95.3

93.5
93.7
90.4
ing 8
ing 30
96.1
ing 18
91.3
94.7
G
84.3
94.6
91.4
94.8

96.2

90.6
94.0
87.6
Days
Days
96.1
Days
93.6
91.4

AVERAGE
AVAILABILITY*
83.3
91.5
91.5
93.9

94.1

93.3
93.7
91.7


95.6

94.0
89.9
92.0
MWH
301,641
308,361
337,468
76,810

223,048

320,701
359,028
275,014


88,925

342,236
358,338
BOILER
HOURS
620.5
594.5
643.0
143.0

436.3

656.0
688.3
521.0


255.8

626.8
706.3

GENERATION
LOAD FACTOR
50.6
55.4
56.7
13.3

37.5

55.7
60.3
46.2


14.9

59.4
60.2
46.4
*Workinfi Hours +  Reserve




    Hours in Month
Exhibit E (Cont'd)

-------
                                         MODULE AVAILABILITY SUMMARY 1977
MONTH
JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
JULY
Ln
g AUGUST
SEPTEMBER
A
94.2
93.4
94.0
96.1


95.0
88.9
93.2
OCTOBER 90.7
NOVEMBER
DECEMBER
93.1

B
90.0
93.0
92.2
93.7
C
95.0
92.6
85.9
97.0
D
95.1
93.8
94.3
94.2
E
94.5
93.3
91.4
95.2
GENERATOR REPAIR AND
STACK
92.8
55.2
93.7
95.6
96.3
TURBIN

REL1NING - 63 DAYS
94.4
93.2
89.1
89.3
93.4
E REPAi;

94.8
93.1
90.0
94.2
94.2
R Nov.

94.6
89.7
92.8
93.4
92.2
15 - De
F
91.6
93.9
94.0
96.1


94.9
92.8
95.0
93.5
92.5
c. 25
|
G
89.8
88.0
90.1
94.5


95.4
92.9
91.7
88.5
95.5


H
	 —
	

	


95.4
93.3
93.0
93.0
95.1

i
i
Availability *
92.9
92.5
91.7
95.2


94.6
87.4
92.3
92.3
94.0

92.52
i
MWH
255,822
310,748
295,420
178,226


213,334
253,605
287,701
173,979
118,439

5Ul_Le t
Hours
539
590
558
384
15

485
501
524
457
234

1
i
ocllt: i a i i <->u
Load Factor
43.0
57.8
49.6
30.9


35.8
42.6
49.9
29.2
20.6

39.9
                                                Exhibit E (Cont'd)
*Working Hours
 «Ml—^<^»^^—
 Hours in Month
Reserve Hours

-------
                                       MODULE AVAILABILITY SUMMARY 1978
                                                                                          Boiler
MONTH
JANUARY
FEBRUARY

MARCH
APRIL
MAY
JUNE
JULY

AUGUST

SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
1
A
90.2
92.4

95.3
91.4
88.9

87.9

92.1
B
94.8
93.4

95.2
92.1
91.5
0 U
97.2

92.5
1
96.1
95.9
91.7
93-9

96.0
95.5
94.9
92.9

C
94.6
95.1

90 4
92.8
91,6
T A C I
91.9

95.0

96,3
98.3
94.3
9U.O
i
D
95.1
94.3

95.4
90.8
93.1

93.9

95.7

95.8
97.0
93.3
95-0
<
E | F
93.4
90.6

94.4
90.2
91.5
6-8-78
88.4

92.7

95.9
97.0
93.6
9**-7

93.5
96.9

94.7
91.8
90.6
thru 7-
92.8

94.3

95.7
97.6
93.0
90.5

G
94.4
95.5

88.6
90.6
93.1
17-78
93.1

94.7

95.3
96.7
94.3
9k.k

H Availability* MWH
94.0
93.4

93.3
90.5
85.6

95.3

95.3

96.6
96.3
96.1
9^-7


93.8
94.0

93.4
91.3
90.7

92.6

94.0

96.0
96.8
93.9
93-3

93-5
332,033
334,897

264,961
330,571
291,651

160,847

307,378

390,826
138,126
386,402
91,7^

•^^^f ^^™- ^^^» W^^P ^^P^ ^^
Hours Load Factor
582
594

593
620
582
14
340

579

72C
255
720
239


54.2
60.5

43.2
55.7
47.6
0
26.2

50.1

65.9
22.5
65.1
13

1*2.2


















                                             Exhibit  E  (Cont'd)
* Working Hours and  Reserve Hours
Hours in Month

-------
                        LA CYGNE STATION UNIT NO. 1
                 POUR HOUR FULL LOAD £ STACK EMISSION TEST
DATE
TIME
LOAD RANGE:
AMBIENT TEMP:
August 26, 1977
11:00 A.M. - 12:00 Midnight
800 + MW
94° F
                                              NOX EMISSION:
                                              0.81 # mm BTU
                                              77*
                       PARTICULATE EMISSION:  .213 # ram BTU
                                              AVERAGE S02 REMOVAL:
MODULES
 A
                                B
                              D
                                                                  G
                             H
GAS FLOW INDICATED
THROAT POSITION
REHEAT TEMPERATURE
VENTURI A?
REHEATER AP
ADSORBER DEM. AP
REHEAT OUTLET
   DAMPER POS.
  ID FAN AMPS
ID FAN INLET
   DAMPER POS.
  FD FAN' AMPS
LAB ph
SULFITE  g/1
 CARBONATE  g/1
   INLET (PPX)
   OUTLET (PPM)
400
OPEN
170
 5
2.5
6.5
50

380
42

490
5.45
60.4
50.3
80.0
4600
920
                               350
                               OPEN
                               190
                               5.5
                               5.5
                               5.5
                               100

                               420
                               42

                               470
                               5.7
                               72.4
                               75.6
                                82.1
                               4600
                                825
                       380
                       OPEN
                       150
                        5
                       4.5
                       10
                       96

                       380
                       32
400
OPEN
190
 5
4.5
7.5
38

400
36
                       430
                       5.55   5.7
                       101.0  74.1
                       53.1   54.4
                       74.9   64.3
                       4600   4600
                       1150   2285
352
OPEN
185
 5
5
7.0
100

470
36
       5.58
       70.0
       59.4
       76.4
       4600
       1085
380
OPEN
180
 5
2.55
6.5
52

470
40
        5.77
        43.9
        83.8
        72.1
        4600
        1285
370
OPEN
160
 5
4.5
8.0
100
366
OPEN
170
 5
5.5
7.0
100
(540 MAX)
( % OPEN)

(540 MAX )
5.72    5.29
43.9    63.6
68.1    42.5
73.1    74.8
        4600
        1235
        4600
        1160
CONDENSER VAC (IN. HG)
WIMDBOX FURNACE DIFF. PRESS (IN.H20)
SCRUBBED OUTLET PRESS (IN.HjO)
FURN'ACE PRESS (IN.H20)
F.D. FAN DISCHARGE   (IN. H? Q )
PEND. REHEAT GAS PRESSURE (IN.H20)
AIR FLOW (%)
BOILER EXCESS Q?  (%)
BAROMETRIC PRESSURE  (IN.Hg)
STACK GAS TEMP  (°F)
FLUE GAS MOISTURE  (%)
STACK GAS VELOCITY   Ft /Sec
               2.5      PRIMARY SUPER GAS PRESS.  (BF.HjO) _ -R
               32       HORZ REHEAT GAS PRESS. (IN.HjO)    -9.5
               -39"     ECON OUTLET GAS PRESS. (IN.H00)   -11.5
              	                                /     ^_———
                -2      FEEDWATER PRESSURE (PSI)          4200
                41      THROTTLE PRESSURE  (PSI)          3400
                -5      THROTTLE TEMP. (°F)               1000°
                85      HOT REHEAT TEMP. (°F)             1300
                2.2    FUEL FLOW %                       68	.
               29.01   FUEL HEATING VALUE (MTB)        9800
                209    FLUE GAS VOLUME  (MCFM)            2998  _^
                13.66   STACK C02                         13.4  _
              103.15   STACK 02   %                      JLL__
              Exhibit F
               504

-------
                                   COSTS
                             LA CYGNE  STATION




               Scrubber Operating Expense June-December 1973
OPERATING LABOR




OPERATING MATERIALS




MAINTENANCE LABOR




MAINTENANCE MATERIALS




LIMESTONE




               TOTAL
$  162,934




     3,480




   189,400




   441,737




   264,514




 1,062,065
                      Scrubber Operating Expense 1974




OPERATING LABOR                          284,541




OPERATING MATERIALS                       67,032




MAINTENANCE LABOR                        401,414




MAINTENANCE MATERIALS                    335,486




LIMESTONE                                780,297




               TOTAL                   1,868,770







                       Scrubber Operating Expense  1975




OPERATING  LABOR                           6^1,029




OPERATING  MATERIALS                       195,926




 MAINTENANCE LABOR                        416,206




 MAINTENANCE MATERIALS                    386,397




 LIMESTONE                               1,256,048




                TOTAL                   2,855,606
0.223 Mils/KWH




0.005




0.259




0.604




0.362




1.453 Mils/KWH
                      0.223 Mils/KWH




                      0.053




                      0.315




                      0.263




                      0.613




                      1.467 Mils/KWH
                       0.265 Mils/KWH




                       0.086




                       0.184




                       0.171




                       0.554




                       1.260 Mils/KWH
                                  Exhibit G




                                  505

-------
                                           September 30,  1977
                         LA CYGNE STATION
MILES FROM PLANT


PRIOR TO START UP

    CONCENTRATION S02 -




RECENT LEVELS
    CONCENTRATION S02 -

       i hour high
      2k hour high
       1 hour high
      2k hour high
      Annual Average
       1 hour high
      2k hour high
      Average

    CONCENTRATION N00 - ppm
AMBIENT MONITORING
STATION
2
, - ppm .009
, - ppm 	
.098
Monthly .015
SYSTEM
1 STATION 2
10
.008
.009
Overall
.653
.135
.011
                                                            STATION 3

                                                            12.5
                                                               .003
                                                              .351
                                                              .106
                                                              .016
                               Station on Line and Wind Toward Monitors
                                 .119           .6', 3          .209
                                 .093           .135          .052
                                 .oik           .015          .016

                              Station Shut Down or Wind Away From Monitors
                                 .15^           .613          .351
                                 .098           .ot+i          .106
                                 .016           .010          .015

                                                .023
NATIONAL STANDARDS

    CONCENTRATION SO2 - ppm

     Annual average
     2k hour maximum
     1 hour maximum

    CONCENTRATION NO2 - ppm

     Annual average
                                                .030
                                                     (may exceed once/year)
                                                .500 (may exceed once/year-
                                                        secondary standard)
                                                .050
                              Exhibit H
                                506

-------
      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
                                  33
               MAINTENANCE
Mechanics                          8
Apprentice Mechanics               2
Welder                             1
Electrician                        2
Technician                         3
Plant Helpers                      2
Foreman                            1
                                  19
              ADMINISTRATIVE
Superintendent
Engineer
                 Exhibit I

                    507
               TOTAL              54

-------
             DRY FGD SYSTEMS FOR THE ELECTRIC UTILITY INDUSTRY
                           Stephen J. Lutz, P.E.
                   TRW Environmental Engineering Division
                           C.J. Chatlynne, Ph.D.
                Industrial Environmental Research Laboratory
                    U.S. Environmental Protection Agency
                           Research Triangle Park
Much research is currently being directed toward the development and utilization
of dry FGD technology because of its simplicity, lower energy requirements,
as well as its ability to produce a dry, easy to handle product.  There are
several approaches to dry FGD, each with its own particular advantages and
disadvantages. By comparing the current state of development for each of
these technologies, we will attempt to provide the reader with the technical
capability and comparative costs associated with each system, and give
suggestions for additional research.

WHY CONSIDER DRY FGD?

In today's environment, we are experiencing an increasing public awareness
of the various aspects of industrial pollution.  Utility and industrial
facilities are faced with a myriad of regulations governing atmospheric,
water borne, and solid-waste discharges.  If the U.S. is to remain com-
petitive with foreign industries while continuing to be responsive to the
environmental needs and concerns of our population, we must carefully evaluate
the long-term impact of each of the various emission control techniques.
FGD techniques vary widely in performance, reliability, and cost.  Several
of the newer approaches to flue gas S0_ control technology additionally
offer the capability of controlling several types of emissions in a unified
fashion.

Dry FGD systems may offer several economic advantages over the current
generation of wet FGD systems, but must be evaluated as part of an overall
emission control approach.  A dry FGD system providing control of both SO-
                                      508

-------
and particulate emissions can be designed and constructed for a fraction of
the cost of a comparable wet scrubber coupled to an electrostatic precipitator.
Depending on the system, it may also provide a reduction in operating costs.
A detailed comparison of these estimated costs will be provided later in
this paper.  In addition to the predicted cost savings, dry FGD systems will
provide a reduction in energy consumption due to the elimination of the need
for reheating the stack gas.  The elimination of wet sludge, an emissions
problem in its own right, will result from the utilization of dry FGD, but
may be counter-balanced by a sodium salt leaching problem from the waste
products of some of the dry processes.

WHAT IS DRY SORPTION?

Generally speaking, dry sorption refers to any process that directly produces
a dry product.  Usually one thinks of a baghouse using a dry S0_ sorbent;
however, also included in the category are processes that employ spray
dryers followed by collection equipment such as baghouses, cyclones, or
ESP's.  For completeness, direct injection of sorbents into the boiler is
included.

Baghouse FGD
Baghouse use is a simple approach in which the sorbent is either applied to
the baghouse as a precoat or injected into the flue gas downstream of the
air preheater.  The latter technique is used to increase the residence time
of the sorbent in the gas stream.  Many sorbents have been used, among which
are nahcolite and trona (naturally occuring NaHCO_ and Na^CO-, respectively),
CaCO™, Ca(OH) , as well as commercial Na2CO, and NaHCO-.  The sorbent reacts
with SO. forming sulfite salts.  The sorbent is periodically renewed so as
to always have adequate reactive species in contact with the flue gas.
                                     509

-------
The following is a brief summary of non-EPA baghouse studies (additional
details are presented later in this paper):

     The first test using a baghouse for SO  control was at Southern California
     Edison's 320 MW Alamitos Station (1.5% S residual oil) in 1965.  They
     reported successful use of dolomitic limestone for SO  removal; however,
     significant operating problems, which were not described in detail,
     were also mentioned.  Nahcolite (natural NaHCO_) was also used but not
     pursued due to lack of availability,  the station has since been converted
     to natural gas (Bechtel, 1976).

     Wheelabrator-Frye performed additional pilot-scale tests using a baghouse
     in 1967-1969 at Edwardsport Station of Public Service of Indiana.  Many
     sorbents were examined, but only Na^CO,. and NaHCO,, were found consistently
     effective.  SO  removal ranged between 13 and 72%, and utilization between
     22 and 93% (Bechtel, 1976).  Unfortunately, high SO  removal was only
     possible at unacceptably low utilizations.

     Air Preheater Company pilot tested several sorbents at Public Service
     Electric and Gas Company of New Jersey's Mercer Station in 1968-1969.
     They confirmed that lime is not an adequate dry sorbent, that nahcolite
     and commercial NaHCO  perform well, and that the operating temperature
     should be above 260°C (Bechtel, 1976).
     In 1974, Superior Oil operated a bench-scale fixed bed of nahcolite at
     Public Gas Company's Cherokee Station.  They observed 80 to 95% SO
     removal at greater than 90% sorbent utilization (Bechtel, 1976).
     Wheelabrator-Frye, Inc. tested nahcolite injection into baghouses at
     Colorado Ute Electrical Association's 11 MW-Nucla Station in 1974.
     They determined that their best removal was 69% at a 56% utilization.
     The coal was mainly 0.8% sulfur with some testing on a 1.1% coal  (Bechtel,
     1976).
                                     510

-------
     In 1976,  the Electric Power Research Corporation (EPRI) contracted with
     Bechtel Corporation to survey the use of dry alkali for removing SO
     from flue gas (Bechtel, 1976).  Bechtel found that lime was relatively
     ineffective in a baghouse, that of 12 additional reagents, only NaHCO_
     and Na2CO.appear to be sufficiently effective to warrant further consideration,
     and that  higher flue gas temperatures generally result in better S0_
     removal.   They concluded that baghouses injected with nahcolite appear
     to be the most promising method of removal in the dry state.  Their
     study did not consider the use of spray dryers.

The following is a brief summary of EPA baghouse programs:

     lERL-RTP's Particulate Technology Branch (PaTB) is co-funding a pilot
     system at Colorado Springs, Colorado, to examine the performance of
                                                    3
     nahcolite and trona on a 1000*-1500 cfm (28-42 m /min) pilot baghouse
     (0.5% S coal).  An option is available to expand this program to a full
     scale baghouse (80 MW).

     A second PaTB project is on an industrial boiler (Kerr Industries) in
     Concord,  N.C.  Testing of sorbent regeneration is planned at Concord in
     addition to the basic sorption studies.  Testing will be on two 35 acfm
         3
     (1 m /min) baghouses (0.7 %S coal).

     In 1977,  TRW began a study of dry sorbents and fabric filters (Lutz et
     al., 1979).  Their main conclusions were that dry sorbent baghouses
     exhibit economic advantages compared with current wet lime/limestone
     scrubbing processes when applied to western power plants burning low-
     sulfur coal.  Additional conclusions are noted throughout this paper.

The main advantages of dry sorbent/baghouse systems are simplicity, energy
requirement reduction, and the production of a dry, easy to handle product.
The dry, once-through approach eliminates the complication of recycle and
                                     511

-------
scaling that can occur in wet systems and, being dry, eliminates  the need  to
reheat the flue gas.  Possible disadvantages relate to the need to provide a
hotter than usual flue gas (260°C) in order to achieve S0_ removal in the
range of 90% and, in the case of sodium-based sorbents, the need  to dispose of
soluble Na?SO« in an environmentally acceptable manner.

Spray Dryers
An additional approach to increasing the contact time between the sorbent  and
the flue gas is to employ a spray dryer in which a slurry or concentrated
sorbent solution contacts the flue gas and leaves the dryer as a  dry powder.
Collection is accomplished with an ESP, cyclones, or baghouses.   Baghouses
have the advantage of allowing additional contact between the flue gas and any
unspent sorbent leaving the spray dryer.  Sorbents used include CaCO,,, Ca(QH)~,
and Na CO-.

There are currently three major commercial suppliers of spray dryer/baghouse
systems:  Rockwell International with Wheelabrator-Frye, Western  Precipitator
with NIRO (Lutz et al., 1979), and Babcock and Wilcox.  Rockwell  performed
pilot tests at Basin Electric in 1977-1978.  They (Estcourt et al., 1978)
report 92% SO  removal at a stoichiometry of 1.0 using commercial soda ash
(Na CO ) in a spray dryer followed by a baghouse.  They compare this with  74%
for a baghouse alone with dry NaHCO., injection.  Western Precipitator's spray
dryer work was in Minnesota earlier this year.  NIRO (Masters, 1978) reports
SO,., removal in excess of 90% at stoichiometric ratios as low as 1.3 to 1.5
with lime slurries in the spray dryer.

Two full scale systems have been sold in North Dakota and one in Wyoming:
Western Precipitator at Basin Electric's 455 MW Antelope No. 2 (burning 0.68%
S lignite), Rockwell at Otter Tail Power's 400 MW Coyote No. 1 (burning
0.9% S lignite), and Babcock & Wilcox at Basin Electric's 500 MW  Laramie
                                      512

-------
River Station Unit 3.  The first system is once-through soda ash while
the latter two are lime/limestone systems; Rockwell also markets a
calcium system.  Current plans for the Rockwell system are  for  the  spent
sodium salts to be disposed of by burial in the ground, but details  are
unknown; however, regeneration a1 la the Aqueous Carbonate  Process  is a
future option.  The first installations are scheduled to begin  operation
in mid-to-late 1981.

In addition to Rockwell International/Wheelabrator-Frye, Western Preci-
pitator (Joy)/NIRO, and B&W, American Air Filter and Koch are also
involved with spray dryers for FGD.

lERL-RTP's Process Technology Branch (PrTB) has initiated a full-scale
demonstration of Rockwell International's Aqueous Carbonate Process  at
Niagara Mohawk's 100 MW Huntley Station in Tonawanda, New York, co-
funded by the Empire State Electric Energy Research Corporation.  This
process utilizes sodium carbonate in a spray dryer to absorb SCL and has
the added feature of regenerating spent Na_CO« and producing elemental
sulfur using coal directly as the reductant.  A 2-year test program  is
expected to start in early 1982.  Perhaps the main advantage of a spray
dryer over a baghouse alone is the increase in sorbent residence time.
When compared with lime/limestone scrubbing, any possibility of scaling
is eliminated.  It appears, also, that the use of a spray dryer is  the
only method of utilizing CaCO_/Ca(OH)  in a dry process since these
salts are not sufficiently reactive to be used in a baghouse alone.
When utilizing a spray dryer for FGD, sufficient S0_ removal is exper-
ienced upstream of the collection device to allow the use of collection
devices other than fabric filters.

Direct Injection
The approach here is to inject the sorbent, limestone, directly into the
boiler along with the coal.  EPA investigated direct injection  in the
early 70's; however, boiler fouling caused the program to be discon-
tinued.  There do not appear to be any non-EPA programs; however, EPA is
back in business, this time on a staged combustion, low NO  burner.
                                                          x
Apparently, the mechanics of combustion in the low NO  burner eliminates
                                                     X
the boiler fouling that had previously taken place.
                                   513

-------
The main advantage of direct injection is to minimize capital costs since a
separate scrubber will not be required.  One disadvantage is that high SCL
removal efficiency has not been demonstrated; however, not being a capital-
intensive process, it can be combined with other processes such as coal
cleaning to achieve the required SO  removal.

OPERATING EXPERIENCE

Several projects have been undertaken, some of which are still in operation,
and additional installations are planned for the near future.  The greatest
amount of experience has been obtained in baghouse-related testing, although
the advancement of this technology is not progressing rapidly due to the
current inability to obtain a sufficient supply of sorbent for full-scale
demonstrations.  Spray dry technology is currently the most advanced of the
various dry FGD approaches, with several commercial units under development.
Little operating experience exists with combustion-zone injection other than
several EPA test burners.  This technology is currently the furthest from
commercial acceptability.

Combustion Zone Injection
No commercial applications of combustion-zone injection for dry FGD are
planned.  Table I lists the relevant operating experience with this technology.
Of particular note are the positive results demonstrated recently using dry
limestone as a sorbent material.  These tests, conducted by the lERL-RTP's
Combustion Research Branch, involve the mixing of ground limestone with the
coal under combustion conditions designed to minimize NO  formation by
                                                        X
controlling the temperature/stoichiometry history of the reactants.  This
combustion condition provides prolonged reactant residence times under fuel-
rich conditions and lower peak flame temperatures.  Currently available
preliminary data from this program are encouraging but not conclusive and
will be subject to verification by additional testing.
                                      514

-------
                                                       TABLE I

                                              COMBUSTION ZONE INJECTION
                                                OPERATING EXPERIENCE
       FACILITY
     DATE
SORBENTS TESTED
          COMMENTS
   EPA Test Burner
Early 1970's
Dry limestone
The test program was not successful due to
recurring problems with injection system.
   Superior Oil Test
   at Cherokee Station
   at Public Service
   of Colorado
    1974
Nahcolite, commercial
sodium bicarbonate,
predecomposed sodium
bicarbonate, soda ash
Ln
h-'
Ul
Test was run on a pilot boiler burning No. 2
fuel oil.  Nahcolite was the most efficient
for S02 removal with commercial sodium bicar-
bonate, predecomposed sodium bicarbonate, and
soda ash being less efficient.  A major
finding of these tests was that nahcolite and
commercial sodium bicarbonate particles ex-
ploded (thermal comminution) because the CO
and water formed by decomposition could not
be liberated fast enough.
   EPA Low NO
             X
   Test Burners
  Current
Dry limestone
LERL-RTP's Combustion Research Branch is
currently running tests on three small coal-
fired test burners (30 kW, 30 MW, 300 MW heat
input).   Although the primary purpose of these
tests is to evaluate a distributed mixing
burner (DMB) for minimizing NO  formation,
limestone addition to the coal was also eval-
uated for S0? control.  The limestone was found
to have an apparent effectiveness, providing
greater than 50% reduction in SO™ with a Ca/S
mole ratio of 1.  No clogging or slagging was
experienced.
   EPA Pilot Plant
   (Fluidized Bed
   Combustion)
   Current
Limestone
lERL-RTP's Advanced Process Branch is currently
evaluating limestone injection for SO  control
in FBC technology.  The pilot plant is a
38 cm x 38 cm bed with a gas flow of 600 scfm
(17 m3/min).   Test results are not yet available.

-------
Current testing is limited to several EPA test programs.  Three small coal-
fired experimental burners (30 kW, 30 MW, 300 MW heat input) are available.
Additional testing is being performed in conjunction with the fluidized bed
combustion (FBC) program.  Early operating experience with this dry FGD
process resulted in numerous operating problems, including materials handling
difficulties, boiler slagging, and generally poor performance. The current
EPA testing program represents an advance in the design of the sorbent
addition system and, based on the results from the initial phase of testing,
appears to have overcome these operating difficulties. It must be understood,
however, that these tests are pilot plant scale studies and may not be
representative of operating characteristics found in full size utility
boilers.

Baghouse FGD
The development of dry FGD baghouses has fostered research in two distinct
areas:  the evaluation of possible sorbent materials, and The- development of
techniques to utilize dry sorbents in practical operating systems.  The
earliest testing, conducted by Southern California Edison at the 320 MW
Alamitos Station (1965), Wheelabrator-Frye at Edwardsport (1967-1969) (pilot),
and Air Preheater at Mercer Station (1968-1969) (pilot), was primarily
concerned with the evaluation of sorbents.  The development of dry FGD
baghouse technology has proceeded through several test programs:  Wheelabrator-
Frye at Nucla Station (1974) and Basin Electric (1977). and by KVB (1978);
and through several major engineering studies (Bechtel, 1976; Lutz et al.,
1979).  Table II reviews the relevant operating experience with this technology.

Operating experience with these test systems has been satisfactory but,
because they have all been designed for discrete testing, no long-term
continuous reliability data have been available.  Baghouses have, however,
been used in the electric utility industry for control of particle emissions
over a considerable time period and have been accepted as reliable.  The
addition of a dry sorbent material to the collected ash is not expected to
significantly decrease the reliability of these devices.
                                     516

-------
      TABLE II

     BAGHOUSE FGD
OPERATING EXPERIENCE
AGILITY
Southern Cal.
Edison
Wheelabrator-
Frye at
Edwardsport
Air Preheater at
Mercer Station
Wheelabrator-
Frye at Nucla
Wheelabrator-
Frye at Basin
Electric
KVB Bench Test
DATE
1975
1967-
1969
1968-
1969
1976-
1977
1976-
1978
SORBENTS TESTED
Dolomitic limestone
nahcolite
Sodium bicarbonate,
soda ash, potassium
permanganate, calcium
hydroxide, and 12
others
Commercial sodium bi-
carbonate, nahcolite,
and hydrated limes
Nahcolite
Nahcolite
Commercial sodium bi-
carbonate, trona,
nahcolite
COMMENTS
Significant operating problems were experienced at this
facility. Testing was discontinued due to these problems
and the inability to obtain an adequate supply of nahcolite.
This was a side stream pilot scale test which ran for 2 years.
SO™ removals ranged to 72% with utilizations of 22-93%. Due
to the small scale of this pilot program, the operating ex-
perience is not considered applicable to full-scale installations.
Tests were run on an existing baghouse which had been used
for particle removal testing. Baghouse flow varied from
7,500 to 15,000 cfm (200-400 m3/min) . Considerable operating
experience was obtained using a variety of injection techniques.
16 independent 90 minute tests were run on a small (11 MW)
coal fired unit. Flow rate was 65,000 scfm (1800 m^/min) .
Various system configurations were tested over a 4-month
period.
Bench scale test demonstrating the Buell-Horbid
baghouse design.

-------
EPA's current and planned testing of the dry FGD baghouses is limited to two
test programs.

lERL-RTP's Particulate Technology Branch has initiated a baghouse program at
an 80 MW generating system owned by the City of Colorado Springs, Colorado.
The fuel used is 0.5% sulfur western coal.  The chief program objectives of
baghouse evaluation and dry sorbent injection will be accomplished with a
                      o
1000-1500 cfm (28-42 m /min) pilot baghouse, fully instrumented, with a
                                      23          2
typical air/cloth ratio of 2/1 (cfm/ft  or 0.3 m /min per m ) of filter.
The current testing program is planned for 15 months.  The contract contains
an option for continued testing of the pilot baghouse, and/or the construction
of a full scale baghouse (80 MW) which would run an additional 15 months.
The utility itself has recently decided to install a pilot-scale spray
dryer.

The second test program currently underway is the expansion of an IERL-RTP
industrial baghouse project to evaluate S0« sorbents.  The location is at
Kerr Industries in Concord, N.C.  The installation includes a 35,000 acfm
      3
(100 m /min) baghouse on each of two 60 MW boilers, using a 0.7 to 0.8%
                                     3          2
sulfur coal.  The air/cloth ratios (m /min per m ) of filter for the baghouses
are capable of ranging from 1/1 to 3/1; 2/1 is a typical operating ratio.
The dry sorbent injection studies are scheduled to begin by spring 1979,
using a number of sorbents and a range of operating conditions.  Further
work is being contemplated that would include a regenerable sorbent process.
The industrial boiler project may be the most likely candidate for regenerable
sorbent investigations, since higher sulfur fuel usage will generate more
solid waste when once-through sorbent methods are used.

No commercial applications of dry FGD baghouses are contemplated in the near
future.  Nahcolite, the most reactive of the dry sorbent materials, is not
currently available in quantitites necessary for a full-scale baghouse
installation, and the owners of nahcolite reserves are reluctant to open a
commercial-size mine (at least 500,000 tons per year).
                                      518

-------
Spray Dryers
Spray dryers, a combination of a spray contactor and a particle collector,
rely heavily on the operating experience obtained from traditional approaches
to particle collection (baghouses, cyclones and electrostatic precipitators),
and from some of the newer FGD research programs, including the dry sorbent
baghouse pilot demonstrations and the spray contactor wet scrubbing systems.
The combination of these individually proven technologies has been demonstrated
in a relatively few pilot plant test programs, identified in Table III,
which have yielded valuable results. Basing their designs on these test
results, Rockwell International, Western Precipitator, and B & W have each
sold one full scale installation.  Testing on these installations should
provide adequate performance, reliability, and operating data from which
other utilities can evaluate their options for flue gas desulfurization.  In
addition, IERL-RTP and the Empire State Electric Energy Research Corporation
are co-funding the demonstration of Rockwell International's Aqueous Carbonate
Process—a regenerable Na»CO_-spray dryer process that produces sulfur.

DRY SCRUBBING AND PROPOSED NEW SOURCE PERFORMANCE STANDARDS

The NSPS revisions proposed in the September 19, 1978 Federal Register for
steam-electric generating facilities include a requirement of 85% S0_ removal
(24 hour average) with maximum emissions of 520 ng/J (1.2 Ib/million Btu).
In addition, the percentage removal will not apply if S0_ emissions are
reduced to 86 ng/J (0.2 Ib/million Btu).

Recently, because of what appears to be some distinct advantages in meeting
the revised SO  standards, a great deal of interest has developed in dry
scrubbing processes.  It appears that these processes may offer improved
reliability and reduced capital and operating costs for selected applications.
Among these applications are sources burning fuels with sulfur contents
which would allow meeting the 0.2 Ib/million Btu limit with less than 85%
removal treating all or a portion of the flue gas.  Dry sorption has the
potential for "full" scrubbing where less than 100% of the flue gas is
                                     519

-------
                                                     TABLE III

                                                  SPRAY DRYER FGD
                                                OPERATING EXPERIENCE
   FACILITY
 DATE
 SORBENTS TESTED
           COMMENTS
   Rockwell International
   and Wheelabrator-Frye
   at Leland Olds Station
1977-
1978
Soda ash, lime,  flyash,
flyash/lime mixtures
Pilot plant testing for 2 months to obtain
scale-up data for future installations.
NJ
o
   Western Precipitator
   at Hoot Lake Station
1978
Soda ash, trona,  pot-ash,
limestone, lime,  flyash
Pilot plant testing on 20,000 acfm
(570 m /min) at 150°C to obtain design
data for later applications.
   Babcock & Wilcox
Conti-
nuing
Lime, trona
Pilot test program utilizing Hitachi Ltd.
and B&W_spray dryers.  A new, 120,000 acfm
(3400 m /min) prototype is under construction.

-------
scrubbed.  Additionally, a dry sorbent fabric filter system also removes
partlculate matter, and some dry sorption approaches conserve energy because
the flue gas temperature is not appreciably lowered, nor is the gas saturated
as in wet scrubbing.  These advantages result in potentially simpler, less
energy consumptive (due to minimization or elimination or reheat) and less
costly systems for compliance with NSPS.
COSTS
Despite all analyses of the technical and environmental advantages of one
system over another, the choice of FGD systems by an electric utility usually
boils down to answers to just two questions:  (1) Which system will satisfy
the EPA? and (2) What does it cost?  The capital and annualized operating
costs for each of the dry FGD systems is compared with a similarly designed
wet scrubber (limestone) in Table IV.

Combustion Zone Injection
Of the various dry FGD techniques discussed in this paper, combustion zone
injection offers the greatest potential for cost savings.  Current EPA
research is progressing into a design which will control NO , S0?, and
                                                           X    £.
particulate matter with a single, low capital investment design.  Although
much research is needed before a definitive cost analysis can be performed,
the primary components of this system appear to be a system for mixing the
dry sorbent material with the coal upstream of the pulverizers, a specially
designed burner, and a particle collection device.  Capital costs are expected
to compare very  favorably with baghouse FGD or spray dryer systems.

Operating costs should reflect the use of inexpensive sorbents, such as
limestone, and the only utilities required will be for sorbent transport and
particle collection.  Operating labor should be minimal since there will be
                                     521

-------
                                                      TABLE IV


                                                COMPARATIVE COSTS3 FOR

                                                SELECTED FGD SYSTEMS

                                                    500 MW BOILER

Capital Costs: $10
$/kW
Annualized $10 /yr
Operating Costs:
mills /kWh
COMBUSTION ZONE
INJECTION
Unknown

Unknown

BAGHOUSE
FGD
23
46
9.1
2.6
SPRAY
DRYER
43
86
10.9
3.1
WET
LIMESTONE
59
118
14.3
4.1
Ui
ro
K5
              In 1977 dollars.

-------
no new process systems such as injectors or contactors to require maintenance
and, because of the totally dry nature of this system, a maximum amount of
flue gas sensible heat can be recovered by air preheaters.

Baghouse FGD
Although no full-scale systems have been sold at this time, sufficient data
have been obtained from the extensive dry sorbent testing programs to establish
accurate cost parameters for baghouse FGD systems.  TRW's Environmental
Engineering Division has developed a FGD baghouse design for a 500 MW pulverized
coal-fired utility boiler (Lutz et al., 1979) based on typical utility
requirements.

Spray Dryers
Three full-scale spray dryer FGD systems have been sold to date.  A comparison
of costs for the spray dryer versus a wet scrubber has been made for the
Laramie River Station Unit No. 3 and will be presented in a later paper at
this symposium (Janssen, 1979).

Wet Scrubbers
Comparison costs were established for wet scrubbing by the TVA, utilizing a
computer cost analysis program modeled on their Shawnee wet limestone scrubber
(Torstrick et al., 1977).

FUTURE RESEARCH NEEDS

Further development of the various dry FGD techniques is anticipated and can
be expected to increase performance and reliability and decrease costs of
future designs.  Some of this research will be funded by the EPA, primarily
in the areas of developing technologies.  Competition among the various
suppliers of commercial dry FGD systems is expected to reduce costs through
design improvements as more of these systems are sold.
                                     523

-------
Combustion zone injection FGD technology is still in its infancy.  The EPA
is providing some basic research in this field and will continue to develop
it in future programs.  One interesting possibility is the application of
combustion zone injection FGD to the fluidized bed combustion concept now
being developed.

Dry sorbent baghouse FGD technology has developed to the point where the
basic FGD process is well understood.  Commercialization has not been forthcoming
due to the unobtainability of nahcolite in sufficient quantity.  Research
needs to be established into possible ways to reduce or eliminate the large
consumption of this material by increasing utilization.  Sorbent regeneration
appears to be the most likely candidate but must be proven in pilot-scale
testing before it can be applied to any commercial installations.  Other
research needs associated with the dry sorbent baghouse FGD system include
further development of methods of insoluabilization of the waste sorbent
so that it can be disposed of in an environmentally acceptable manner.

Spray dryer FGD technology has advanced at a very rapid rate and is now
available commercially.  Improvements in reducing the costs for these systems
can be expected but will require extensive development. Regeneration of
sorbents offers great promise in this area, but will require significant
research and development before it is practical on a commercial scale.
                                     524

-------
REFERENCES

Bechtel Corporation, 1976.  "Evaluation of Dry Alkalis for Removing
     Sulfur Dioxide from Boiler Flue Gases."  EPRI FP-207.

Estcourt, V. F., R. 0. M. Grutle, D. C. Gehri, and H. J. Peters, 1978.
     "Tests of a Two-Stage Combined Dry Scrubber/SO,, Absorber Using
     Sodium or Calcium."  Presented at the 40th Annual Meeting, American
     Power Conference, Chicago, April 26.

Janssen, K., 1979.  "Report on Dry Flue Gas Desulfurization by Kent
     Janssen on Behalf of Basin Electric Power Cooperative to the
     Environmental Industry Council."  Presented at the Symposium on
     Flue Gas Desulfurization, Las Vegas, NV.  March 6, 1979.

Lutz, S. J., R. C. Christman, B. C. McCoy, S. W. Mulligan, and K. M.
     Slimak, 1979.  "Evaluation of Dry Sorbents and Fabric Filtration
     for FGD."  EPA-600/7-79-005 (NTIS PB 289921) January 1979.

Masters, K., 1978.  "The Niro Atomizer Spray Dryer for SO  Absorption
     from Flue Gas."  Presented at the Western Precipitation Seminar,
     Durago, Colorado, May 21.

Torstrick, R. L., L. J. Henson, and S. V. Tomlinson, 1977.  "Economic
     Evaluation Techniques, Results, and Computer Modeling for Flue Gas
     Desulfurization," in "Proceedings:  Symposium on Flue Gas Desulfuri-
     zation, Hollywood, FL,  November 1977 (Volume I)  EPA-600/7-78-058a
     (NTIS PB 282090), March 1978.
                                     525

-------
                                          Paper No. 41
                                          EPA's FGD Symposium
                                          March 1979
      PLAN, DESIGN AND OPERATING EXPERIENCE OF FGD

                           FOR

            COAL FIRED BOILERS OWNED BY EPDC



                           by

                  Yasuyuki Nakabayashi

                Assistant General  Manager

                Thermal  Power Department

          Electric Power Development Co., Ltd.

                      Tokyo,  Japan
                        ABSTRACT
EPDC has built Wet Limestone-Gypsum FGD systems for coal
fired boilers which are comprised of five 250MW class
capacity, totaling 1,280MW.

Each of the FGD systems has  been in service for three to four
years with the same operating reliability as the boiler
availability.  EPDC is planning and constructing two
500MW and one 700MW coal  fired plants where EPDC is plan-
ning to use imported coal  and install FGD Systems
accordingly.

EPDC is planning to start  project of another 1,OOOMW coal
fired plant in the near future.

This paper is presented to describe basic philosophy for
design, specific features  on design, improvement of
design resulted from operating experience, operating
reliability, major problems  and treatment of by-product.
                         526

-------
              PLAN, DESIGN AND OPERATING EXPERIENCE OF FGD

                                  FOR

                    COAL FIRED BOILERS OWNED BY EPDC
1.   BASIC PHILOSOPHY OF FGD SYSTEM

     It is indispensable to evaluate in advance which type of process is
     the most feasible for application of FGD.

     The following factors  are to be evaluated in  advance

     o Technical reliability

     o High performance capability

     o Good economy

     o No secondary pollution

     o Superior operating characteristics

     o Security of stable supply of absorbent and  disposal  of by-product

     The optimum FGD process shall be selected on  basis  of the total  evalua-
     tion on above six factors.

     As EPDC judged Limestone-Gypsum process as the best one, EPDC  adopted
     the process.


2.   EPDC'S PRE-EVALUATION

     The following is an outline of EPDC's  philosophy as to how the Limestone-
     Gypsum process has been evaluated.

(1)   Selection of  Wet Process or Dry Process

     At the time when EPDC  was planning to  adopt FGD system (19&7 - 1968),
     dry processes were developed in Japan  with Government Subsidy.

     EPDC concluded that wet processes were superior to  dry processes on  the
     basis of pre-evaluation of the six factors previously mentioned after  a
     complete study of the  development status of dry processes and  the details
     of new technical development on wet processes.

     Table 1  show  its outline.
                                   527

-------
             Table 1.  Comparison of Wet and Dry FGD Processes
Item
Desulfurization
efficiency
Temperature of
treated exhaust
gas

Material of
equipment
Scaling up
Application to
coal-fired
boiler
Reaction velocity
Investment cost
Equipment
Pressure loss
Consumption of
utilities
Water
Absorbent
Electricity
Fuel
Steam
Wet FGD Process
90 percent or more
(Influence of variation
of flue gas volume on
efficiency is very
small .)
50° to 60°C
(Reheat is required
for prevention of
white plume and better
diffusion to the
atmosphere)
Mainly plastic lined
material
(Anti-corrosive measures
are needed)
Easy
Suitable
Fast
Small
Small (smaller space is
required)
Small

Much
Cheap (limestone)
Much
Much
Much
Dry FGD Process
Around 80 percent
(Efficiency varies accord-
ing to volume of flue gas
or operating hours)
Reheat is not necessary
as boiler flue gas
temperature is suffi-
ciently high.

Mainly metal material
(Heat-proof and anti-
corrosive measures are
needed)
Difficult
Not suitable because of
high dust concentration
Slow
Large
Large (larger space is
required
Large

Little
Expensive (activated
carbon)
Little
Not required
Not required
(2)   Selection  of  the  Optimum Process  out of Wet  Processes

     The  wet  process can  be  sub-classified  to several ones  depending  on
     the  kind of absorbent and  the  treatment method of  by-product.

     Each process  possessed  advantages and  the disadvantages, respect!vely,
     however, the  Limestone-Gypsum  process  has been evaluated  as  the  best
     one.   Table 2 summarizes  the pre-evaluation  results.
                                   528

-------
                  Table  2.   Pre-evaluation of Wet FGD Process
~~^~^^ Evaluation Item
Name of Process ^^~\^^
Throw-away
Process
Gyps urn
recovery
Process
Sulfuric acid
recovery
Process
(1)
Calcium base
(2)
Sodium base
(3)
Lime-Limestone
(4)
Double Alkali
(5)
Sulfuric acid
Dilution
(6)
Ammonia-Calcium
(7)
Sodium base
(8)
Magnesium base
Process
Desulfuri-
zation
Efficiency
0
©
o
©
o
©
© '
©
Simplicity
©
©
o
A
O
A
X
X
Secondary
Pollution
A
A
O
O
O
A
O
©
.Facility
of
Operation
©
©
• O
A
O
A
X
X
Actual
use
O
o
©
o
o
A
O
o
Cost
Construct
-ion
©
©
O
A
A
O
X
X
Operation
O
A
O
O
A
A
O
O
           Remarks ;
©  Better
O  Good
A  Bad
X  Worse
3.    OUTLINE OF EPDC'S FGD SYSTEM

     Since EPDC started operation on February 2nd, 1975, of EPDC's first
     FGD system at Takasago Power Station No.l unit, EPDC has adopted FGD
     Systems to all  of coal  fired power plants owned by EPDC.   All EPDC's
     FGD Systems -can treat the full  capacity of boiler flue gas and the total
     capacity  is equivalent  to 1,280MW generating capacity in terms of facil-
     ities.  The FGD process  is  that which uses limestone as absorbent and
     produce gypsum  as by-product.

     The boiler fuel is mainly the coal which  is produced in Japan.

     The FGD systems are the extension of facilities to those of  existing
     power stations, Table 3 shows the outline of power stations  and  those
     FGD systems.
                                   529

-------
Table 3.  Outline of Design  Specifications
^ 	 	 ____I'ower Station
Item ^^^— -^^^Jnit
Location
Generation Equipment
Flue Gas Desulfurization
Equipment
Approved Output (MW)
Turbine
Boiler
I— 1
s
fu
Type
Output (MW)
Manufacturer
Type
Max. Evaporation (t/h)
Manufacturer
Fuel
S Content in Fuel
Fuel Consumption (t/h)
Operation Started on
Type
Volume (Nm3/h)
Volume (MW Equivalent)
Absorbent
Manufacturer
Waste
Water
Trpat-
u Type
c
e Volume (t/h)
Operation Started on
Isogo Thermal P/S
No.l Unit
No. 2 Unit
Yokohama City, Kanagawa Pref.
265
265
Cross compound 4-Turbine
Chamber, 4-Shunt, Suction,
Reheating System.
265
265
Tokyo Shibaura Electric
IHI-FW Single Drum Radiation
Reheating Water Pipe System.
840
840
Ishikawajima-Harima Heavy
Industries
Coal
0.6
100
May 25, 1967
Takasago Thermal P/S
No.l Unit
No. 2 Unit
Takasago City, Hyogo Pref.
250
250
Reheating, Regenerating,
Circulation System.
250
250
Mitsubishi Heavy Industries
Forced Circulation Reheating
Readiation Water Pipe System.
825
825
Mitsubishi Heavy Industries
Coal Coal
0.6
100
Sep. 23, 1969
Wet Limestone-Gypsum Process
900,000
265 (Whole
Energy)
900,000
265 (Whole
Energy)
Calcium Carbonate
Ishikawajima-Harima Heavy
Industries
Coagulating Sedimentation
plus adsorption
15
Mar. 3, 1976
15
May 21, 1976
1.8
96
Jul. 1, 1968
Coal
1.8
96
Jan. 18, 1969 .
Wet Limestone-Gypsum Process
842,000
250 (Whole
Energy)
842,000
250 (Whole
Energy)
Calcium Carbonate
Mitsui Miike Works
Coagulating Sedimentation
plus adsorption
15
Feb. 5, 1975
15
Mar. 24, 1976
Takehara Thermal P/S
No.l Unit
Takehara City, Hiroshima Pref.
250
Tandem, 3-Turbine Chamber,
4-Shunt, Suction, Reheating
system.
250
Hitachi
Hitachi B & W Natural Circula-
tion Single Drum Radiation
Reheating System.
810
Babcock-Hitachi Hitachi
Coal
2.0
100
Jul. 25, 1967
Wet Limestone-Gypsum Process
852,000
250 (Whole Energy)
Calcium Carbonate
Hitachi
Coagulating Sedimentation plus
adsorption
15
Feb. 1977

-------
 Photo 1  Isogo's FGD
        epl
 Photo 2 Takasago's  FGD
           ^w >-.$*

       ESfJn.
Photo 3  Takehara's  FGD
      531

-------
A.   DESCRIPTION ON FGD PROCESS

(1)  Specific Features of Process at Each Plant

     Specific features of process at each location are described as
     fo11ows:

     At Takehara Power Station, two scrubber towers were adopted in parallel
     and cooling towers were used for removal  of dust.  At Takasago Power
     Station,  a PH adjustment tower is installed for adjustment of PH of
     slurry from first scrubber by introducing part of flue gas into the PH
     adjustment tower so that high desulphurization efficiency may be main-
     tained and unreacted limestone may be converted to gypsum without
     applying any sulphuric acid.

     At Isogo  Power Station, the receiving facilities of limestone and gypsum
     are small because of low SOx concentration in flue gas.   The FGD system
     is designed and guaranteed for higher dust removal efficiency because
     emmission control of dust is very sever at Isogo Power Station.

     Table k shows the comparison of FGD Processes.
          Table k.  Comparison of FGD Processes  in Actual  Operation

Scrubber
Cool ing tower
PH Adjusting tower
Absorbent excess
ratio
H2SOj,
SOx in flue gas
El iminator washing
water
After burner
(gas temp.)
1 sogo
2
ser ies
None
None
1 -v 1.1
little
300 ppm
Raw water
used
(85 ^ 95°C)
Takasago
2
series
None
1
1 '\< 1.1
None
1500 ^
Raw water £
mother water
used
(80 -v- 85°C)
Takehara
2
paral lei
2
None
1 ^ 1.1
1 ittle
1500 %
Raw water
used-
(120°C)
(2)   Description of the Process at Each Plant

     (a)   Process description of IHI's FGD system at Isogo P/S

          Flue gas boosted up by fan is sent to the first scrubber, then,  the
          gas goes through the second scrubber.  SOx contained in boiler flue
                                   532

-------
    gas  is  removed in the first and second scrubbers and the dust in it
    is also removed in the same scrubbers.

    The wet gas flowing through the second scrubber is mixed in the
    mixing  chamber with the hot air warmed in the reheater, and is
    discharged from the stack after heated up by the after burner to a
    temperature of 85°C ^ 95°C.  The absorbent is 98% pure-J imestone
    with a  particle size of minus  325 mesh.

    The absorbent is carried by container trucks f«r exclusive use at
    the power station and is stored  in a silo.
    After weighing, it is kept ir a tank.  After measuring the concen-
    tration, the volume of absorbent slurry is automatically determined
    to match with the saif- ^olume of flow from the scruSber.

    The amount of feed is about 1.0 -\> 1.1 times of fie '- «t SC>2 by
    molecular weight.

    The circulation slurry from second scrubber is supj.  ad to the
    fir-^t scrubber.

    .ne bleed slurry from the first scrubber is fed to the oxidation
    tower and then the slurry is oxidized with normal  pressure air
    for production of gypsum.

    The 5%  gypsur slurry after oxidation  is fed to a gypsum separation
    apparatus.

    After concentrating the slurry by a thickener, the slurry is
    processed by centrifuges for production of 10% moisture cake.

    The product is stored in a gypsum warehouse and a part of the thickner
    overflow water is discharged out of the FGD system in order to
    prevent build up of such impurities as Cl, F etc.  The mist
    eliminator is washed  intermittently with industrial raw water.
    Figure 1 shows the flow chart of  Isogo's FGD.


(b)  Process description of Mitsui Mi ike's FGD System at Takasago P/S

    Flue gas boosted up by about ^50 mmAq by use of fan is split into
    three divisions for introduction  into 1st and 2nd oxidation towers
    (No.2 Unit has only one oxidation tower), pH adjustment tank and
    1st scrubber.   Then, the gas  goes through 2nd scrubber for
    completion of S02 removal,  dust removal  and pH adjustment.  The wet
    gas  coming through the 2nd  scrubber is mixed in the mixing chamber
    with the hot air generated  in  the reheater, and is discharged from
    the  stack after heated up to  a temperature of 80°C -\> 85°C.

    The absorbent of limestone of purity over 98% and particle size of
    minus 325 mesh is received from a ship for exclusive use and
                              533

-------
      Purified gas
 From suction    Booster fan
 fan at the boiler
      Industrial
       water
  Concentrated
  sulfuric acid
  tank
             Circulation
              pump
                                 Mixing chamber
                                       Ok.
To NO.2 scrubber
     X
       Oxidizing tower
                   Circulatiorig-
                                   Pump  No.2 scrubber
Calcium
carbonate
hopper

  I	1
                                                          - Absorbent
                                                          "Slurry tank"
                Oxidizing
                blower
                                       To waste water treatment
         Figure 1    The Flow  Chart  of Isogo's  FGD
some container 'trucks, and stored in silo.  After  weighing,  it  is
kept in  the  tank as about  15%  slurry.  After measurement of  density,
volume of  the absorbent slurry is automatically  determined to meet
the volume of flue gas treated,  inlet concentration of S02 and  pH
of oxidation tower.  Then,  it  is  fed into 2nd  scrubber.  The  amount of
the feed is  about 1.0 ^ 1.05 times of inlet S02  by molecular weight.

The 2nd  scrubber is being operated at pH 6.0 ^ 6.3 and liquid/gas
ratio 6.0  ^  7.0 5,/m'.  The circulation slurry  from the 2nd scrubber
is supplied  to the 1st scrubber.   The 1st absorption tower is being
operated at  pH 5-6 ^1.1 and liquid/gas ratio  5-5  ^ 6.0 &/m3.

The pH value of the bleed  slurry  from the 1st  scrubber is adjusted
to pH 5-^  ^  5-8 in the pH  adjusting tower.   In the oxidation tower,
the air  is blown into the  bleed  slurry and  the slurry  is oxidized
under normal  pressure for  gypsum  production.

A very small  quantity of catalyst solution  is  supplied to the 2nd
scrubber to  improve the S02  removal, to increase the oxidation
speed and  to obtain quality  gypsum.
                           534

-------
    The 5 '^ 1%  gypsum  slurry  after  the  oxidation is sent to gypsum
    separating  apparatus.   After  concentration to about 2Q% slurry by
    the thickner,  this slurry is  processed  by the 7 centrifugal  separa-
    tors to produce  cakes  of  5  ^  8% moisture.  The product is stored in
    the gypsum  warehouse.

    The thickner overflow  containing 500  ppm or less suspended solids
    and the filtrate of centrifugal  separators are circulated for
    adjustment  of-absorbent density, adjustment of liquid  level  of the
    scrubber and for washing  of mist eliminator.   A portion of the
    liquid is discharged out  of the  system  for the purpose of preventing
    accumulation of  Cl~.

    The mist eliminator of a  A  stage chevron type is being washed
    intermittently by  use  of  raw  water  and  circulating mother liquor.

    Figure 2 shows the flow chart of Takasago's FGD.
 Industrial
  water
Absorbent
silo
                                          To waste water
                                            treatment
              Figure 2    The Flow Chart of Takasago's FGD
(c)   Process  description  of Hitachi's  FGD.system at  Takehara P/S

     Flue  gas boosted  up  by use  of  fan is  split into two divisions,
     A cooling  tower and  B  cooling  tower.

     In  the case  of Takehara P/S,  the FGD  is composed by 2 trains.
                              535

-------
           Flue gas temperature  is  reduced at the cooling  tower from 150°C to
           50°C ^ 60°C and the flue gas  is sent to scrubber.   Dust is removed
           at  the cool ing tower  and SOx  is removed at  the  scrubber.

           Limestone as the absorbent  is  the same specification of the others,
           and unloading system  of  limestone and Gypsum  Production System  is
           the same as Isogo's process.

           Figure 3 shows the flow  chart  of Takehara's FGD.
              Cooling tower   £,bw°rrption
              circulation tank  cjrcuiation
                           tank
                                                       Limestone

                                                         	  Limestone freight
                                                         """"A r"—i
                                                  Limestone slurry tank

                                              -Industrial water
                                    To waste water treatment
                                                              /vGypsum
                 Figure 3-   The Flow Chart of  Takehara's FGD
5.   SPECIFIC  FEATURES  OF DESIGN AND OPERATION  OF FGD SYSTEM

     There are a  number of factors to be considered in the design  and
     operation of FGD  system in order to keep a high desulphurization
     efficiency and  maintain the equipment  in good conditions.


(1)   Specific  Features  of Design

     The following specific features were considered in the design of
     EPDC's FGD systems.
                                     536

-------
(a)   Specific features of process

     •  The  1fmestone  absorbent  Is  purchased  in  the  form  of  powder with
       a  size of  minus 325  mesh  directly  from the  limestone production
       mine.   There are no  crushing  facilities  of wet  mill  for  FGD
       process at the power plant  site  for.

       The  reasons following:

       o  Problem  exists in  the  reliability of wet  mill crushing facilities

       o  FGD systems  must be installed  within  the  limited area  of  power
         stations as  the FGD system is  the extension of  existing facili-
         ties

       o  There are crushing facilities  already  at  limestone mine so it
         will be  double investment

       o  Satisfactory countermeasures are needed for prevention of  noise
         generated from the crushing facilities

       o  Quality  control (purity and particle  size) was  quite possible
         by purchasing in a form of powder and  there is  no influence to
         the desulphurization efficiency

     •  Production facilities of gypsum as by product are installed  in
       power stations.

       Following  is  the reason:

       o  Gypsum is a  marketable item and sales  is  possible

       o  Gypsum i.s a  stabilized  chemical  material.
     •  The scrubber is  of the Venturi  type which possesses  high  dust
       removal  capabi1i ty

       Following is the reason:

       o Limitation is  found in the capability of electrostatic
         preci pi tators

       o Severer regulations apply to dust emission


(b)  Consideration for  corrosion

     •  Epoxy lining or  rubber lining is applied in the scrubber and its
       connecting pipes and pumps
                               537

-------
       Following  is  the  reason:

       o  PH  in  the scrubber  line  is  less  than  7-
         Corrosive slurry  moves  in  the  line so direct  contact  to
         metallic surface  must  be avoided.

       o  Epoxy  material  or Titanium metal  is used  for  the parts  where
         lining is not possible.
(c)   Coordination  with  the  plant

     *  A flue  gas  by-pass duct was  installed  for

       the  following  reasons:

       o No influence to  the  boiler  is  caused  by  a  B.U.F.  trip

       o No necessity is  found for  flue gas  to flow through  the  FGD  system
         at the time  of plant start  up.


(d)   Waste  water treatment

     •  Waste water discharged from  FGD  system is  treated.

       Following is the reason:

       o FGD waste water  must be  treated as  PH,  SS  and  COD etc.  exceeds
         the limit of Japanese waste water standards


     •  Specific COD treatment system is provided  for  FGD system

       Following is the reason:

       o With  a conventional  COD  system, the COD  materials discharged
         from  FGD system cannot  be  treated satisfactorily and therefore
         the specific COD treatment  system was developed for FGD system.


     •  Calcium substances are used  as a coagulate sedimentation  agent
       for  treatment  of fluorine

       Following is the reason:

       o Treatment of heavy metal  in FGD waste water is made by  such
         coagulate sedimentation  agents as Na-substances and Ca-substances,
         Fluorine cannot  be treated  by  Na-substances.
                              538

-------
(2)   Consideration  on  Operation

     Following  consideration  shall  be  paid  for  operation of  FGD  system


     (a)   Continuous  blowing  water  out of  the  system

          Following is the reason:

          o Coal  contains  chroline  and some tens  ppm chroline exists  in  flue
            gas.   Chroline is feared  to concentrate in  the scrubber which
            causes  corrosion  of  the equipment,  therefore  the constant chroline
            concentration  must be kept with the continuous blowing water.
     (b)   Good  control  shall  be  achieved  on  the  concentration  of  absorbent
          siurry

          o Desulphurization  efficiency  is  influenced  by  particle size of
            absorbent  and  PH  of  absorbent slurry.

            The particle size is almost  constant as  the absorbent is  purchased
            in  the form of powder.
            The PH control of absorbent  slurry is  achieved  satisfactorily  by
            a good control of slurry concentration as  it  changes  considerably
            depending  on the  input  amount of absorbent.


          o The desulphurization efficiency  goes up  for the increase  of  input
            amount of  absorbent  and it causes the  scale build  up, therefore
            the optimum input amount exists.


     (c)   The amount of absorbent is determined  according to plant load

          o Unless the absorbent is supplied  following the  load change,  the
            material balance  of  absorbent which  contributes to reaction  with
            SOx will be broken then the  desulphurization  efficiency will be
            affected.
     (d)   Sulphur content  in  the  coal  shall  be  kept  constant

          Following  is  the reason:

          o If the sulphur content  in  the coal  changes  considerably,  the pH
            control  according to  the change  of  sulphur  content shall  become
            impossible  as  the mass  in  FGD system is  too large even though the
            matching amount of absorbent  is  fed accordingly.
                                   539

-------
            Therefore the variation of sulphur content shall  be controlled to
            the minimum and a certain range of sulphur content coal  must be
            used when different coals are blended.
     (e)   Raw water shall  be applied for washing eliminator

          Following is  the reason:

          o In case of  desulphurization of high sulphur content flue gas,  it
            is often feared that scale build up takes place at the Venturi  and
            mist eliminator of the  scrubber and further it will-cause plugging.

            If the process water from dehydrator is  utilized for the washing
            water of eliminator, the scale build up  occurs but it was prevented
            by changing the process water to raw water.
6.   OPERATING RELIABILITY

     The operating  reliability  of  the  FGD  system  can  be  judged  by  the  following
     factors.

     o Availability of FGD system  has  equalled  the  availability of plant

     o Maintained high performance

     o Stable operation

     Assurance has  been obtained on the statisfactory reliability  of FGD
     systems which  EPDC is operating,  based  on  the  past  operation  experience
     of FGD systems.

     Operating experience is described as  follows.


(1)   Operating Hours

     Table 5 shows  the operating hours of  each  unit at each plant  since the
     start of operation through fiscal  year  end of  1977.

     Following facts have been  proven.

     o Total operating time of  all FGD systems  came to approximately 80,000
       hours which  are equivalent  to 99-3% of availability of EPDC's plants.

     o The longest  operating time  per  unit is approximately 25,000 hours  and
       its availability is 98.7%.   (Takasago No.l unit)

     o 100% is the  availability of FGD systems  at I sogo No.l  and No..2  and
       Takehara No.l  for the fiscal year of 1977-
                                    540

-------
    The abpve data were  taken  from the figures through March 31,  1978 and FGD
    systems have  been  in service at present without any problem.

    The estimated operating  time will be as follows by March 31,  1979-

    o Total availability time:  Approx. 120,000 hours

    o The  longest availability time per unit:  Approx. 30,000 hours
             Table  5.  Actual Hours of Plant and  FGD Operations
                                                                [Unit:  hour]
Plant
Isogo
Takasago
Takehara
Total
Unit
1
2
Total
1
2
Total
1

1974
A
"^
^
^
975
^
975
^
975
B
\
\
\
957
\
957
K
957
1975
A
676
^
676
8,206
^
8,206'
^
8,924
B
676
\
676
8,053
\
8,053
\
8,729
1976
A
7,699
7,031
14,730
7,515
8,143
15,523
1,056
31,741
B
7,699
7,031
14,730
7,438
8,008
15,446
1,056
31,252
1977
A
8,115
7,641
15,756
8,381
7,703
16,084
7,895
39,735
B
8,115
7,641
15,756
8,328
7,595
15,923
7,895
39,574
Total
A
16,490
14,672
31,162
25,077
15,846
40,923
8,951
81,036
B
16,490
14,672
31,162
24,776
15,603
40,399
8,951
80,512
Avail-
ability
100 %
100 %
100 %
98.7%
98.4%
98.6%
100 %
99.3%
Remarks
o 1974 means
period of
April 1, 1974
to March 31,
1975.
o A means hours
of Plant
operation.
o B means hours
of FGD
operation.
(2)  Average Desulphurization Efficiency

    Table 6 indicates average desulphurization efficiency by the fiscal year
    end of 1977 according to plant and unit.

    Following facts have been  learned.

    o The average desulphurization efficiency  is about 91% through approximately
      80,000 hours of FGD operation.
                                   541

-------
    o The average desulphurization efficiency at  Isogo Power Station  is about
      36% through approx. 30,000 hours of FGD operation.

    o The average desulphurization efficiency at Takasago Power Station is
      about 88% through approx. 40,000 hours of FGD operation.

    o The average desulphurization efficiency at Takehara Power Station is
      about 9U through approx. 9,000 hours operation.
            Table 6.  Actual Average Desulphurization Efficiency
                                                                    [Unit:
Plant
1 sogo

Takasago

Takehara
Average
Unit
1
2
Total
1
2
Total
1
-
1974

90

90

90
1975
94
94
87

87

88
1976
96
96
96
88
88
88
91
91
1977
96
96
96
90
90
90
91
92
Average
96
96
96
88
88
88
91
91
Remarks
Year means
fiscal
year


(3)   The Load  Following  Capability of  FGD System to  the Load Change of  Power
     Plant

     According to  a  load pattern, EPDC's coal  fired  power  plants are  in
     service at full  load  during day and at about  1/2  load during  night, but
     at  the minimum  load in  the week end due  to  reduced demand.

     This load pattern for a week  is repeated.  The  following  capability of
     FGD system to load  change  is dependent upon how to pontrol  the pH  value
     in  the scrubber  at  constant  level.

     To  maintain a constant  desulphurization  efficiency during  load swing,
     the feed  rate of absorbent into the scrubber  is controlled  in proportion
     to  the produced  gypsum  quantity thus the  PH value is  kept  constant.

     As  a result of  investigation of desulphurization efficiency for  random
     consecutive 60 days in  1978, according to load  change of  the  plants at
                                   542

-------
three locations, the variation of desulphurization efficiency is summa-
rized in the following Table 7-
       Table 7.  The Variation of Desulphurization Efficiency
                                                                [Unit: %]
Plant
Isogo
Takasago
Takehara
Load
Variation
265MW 'V 135MW
250MW ^ 125MW
250MW * 100MW
Max.
DeSOx
Efficiency
Variat ion
1.2
(95.8 * 97.0)
3.9
(90.8 * 3k. 7)
B.k
(88.0 * 96. k)
Min.
DeSOx
Efficiency
Variation
0
(37. k * 97. k)
1.7
(93-3 *> 95.0)
0.8
(93. k ^ 3k. 2)
Average
DeSOx
Efficiency
Variat ion
0.1
(96.8 -v 96.7)
2.7
(92.2 * 3k:3)
2.9
(93.1 * 96.0)
The following facts have been learned.

o Difference exists in the variation according to FGD system at each
  plant.

o As the sulphur concentration in flue gas changes considerably and
  desulphurization efficiency is affected accordingly, the efficiency is
  not always assured to be more than 90%.

o The average desulphurization efficiency value has increased compared to
  those of past years.  The increase has bean realized due to the efforts
  exerted in developing EPDC's know-how of the absorbent control line.
The variation of Desulphurization Efficiency at Load Change

Figure k shows the variation of desulphurization efficiency at the time
the plant load decreases from 250MW,  rated value,  to 125MW at Takasago
Power Station.

The, load changes by 3 MW/min. so  it takes about k2 minutes to reduce
250MW down to 125MW,

Figure 5 shows the change of SOx  and  Q£ en Chart Recorder, 02%  increase
along with load down from 250MW to  125MW so SOx concent rat i'on at  the
                               543

-------
                           250 MW load
                                        24
                           Time
 Figure k.
The Variation of Desulphurization
Efficiency at Load Change
inlet of FGD system
decrease.

SOx concentration  at
the outlet of FGD
system decreases
accordingly, the
desulphurization
efficiency changes
in an almost constant
range without showing
unstable characteris-
tics.

The stable desulphuri-
zation efficiency  is
maintained after load
change has been
completed.
As for SOx concentration, no calibration  has  been  made based on
concentration.
                      o CHART No.MR-861(IU)
                                         SOz Outlet (0~500ppm)
                     DeSOx Outlet SOx(0 SOOppm)
      Figure 5.  The Change  of  SOx and 02 on Chart Recorder
                               544

-------
7-    CHANGE OF SULPHUR CONTENT DUE TO BLENDING OF DIFFERENT COALS

(1)   Method of Blending Coals

     Three to four kinds of coals  are blended for boiler fuel  at  each power
     station,  and the variation range of sulphur content is intended to be
     as small as possible.

     As for a blending method, the coals are stockpiled  in  the predetermined
     area of power station, according to the kind of coal  then the certain
     coals are scrapped by  bulldozers and thrown  into a  underground  hopper
     where several kinds are blended.

     The variation range of sulphur content  becomes  as follows.
Local ion
1 sogo
Takasago
Takehara
Actual data
0.5 T. 0.6%
1.6 *• 1.85?
1.5 ^ 1.9*
Blend ratio
A:B:C 8:1:1
D:B:A - 'l.25:'i.25: 1 .5
D:B:A = 3:6:1
Sulphur content
A 0.3*
B 2.5 2.7*
C 0.6 1.0*
D 1.0 1 M,
B 2.5 2.7%
A 0.6 1 .0%
D 0.9 1 .5*
B 2.1 2.7*
A 0.3 0.5
8.   MAJOR IMPROVEMENT ON DESIGN

     A problem arose in the basic design  of FGD  system at  Takasago  Power
     Station  during its operation as  scale  was built  up in Venturi  and eliminator
     of the scrubber.

     The scale build up was monitored by  an increase  of pressure  drop, and was
     manually removed,  stopping  the  FGD system  if  necessary.

     This has been  the major trouble  of FGD system and there  have no other
     troubles in  the operating experience.

     Following is the description on  the  solution  of  scale build  up at Takasago
     Power Station.

     Overflow water (mother liquid)  from  a  thickener was utilized for the wash-
     ing water of eliminator at  the  time  the plant operation  started.  Since
     CaSO/j is saturated in the mother liquid,  it tends to  be  crystal ized  as
     scale onto the eliminator  in the scrubber.

     Following improvement was performed  in June 1978.

     o Raw water  was supplied in addition to the mother liquid  to the washing
       water  line of eliminators of  the first  scrubber and the  second
       scrubber.   (Takasago P/S)

     o It is  desirable to change the  mother liquid to raw  water also  for  the
       washing water of eliminators,  but  it was  found to cause  further burden
                                    545

-------
  to the existing waste water treatment system because  the  volume  of
  blowing water out of the system must be  increased due  to  the  balanced
  volume of total water.
  With the modification, the scale build up on the eliminators was
  considerably prevented.

  Figure 6 shows the  improvement of eliminator washing water  line.
                                    New line
                                                     i j Raw water
      PH adjusting Tower    reactor
        * Gas line is omitted.
    Figure 6.  The  Improvement of Eliminator Washing Water  Line
As photo k and 5 shows, there is no sign of scale build up after approxi-
mately 4,500 hours since the start of supplying paw water to the washing
water.
                               546

-------
        Photo k   Before  improvement
Photo 5   After improvement
9.    MAJOR  PROBLEMS  OF FGD  SYSTEM AT  EACH  PLANT

     One  problem of  FGD system in EPDC's experience which  has  resulted  in
     shutdowns  of FGD  system is  caused  by  the  scale build  up  in  the  scrubber
     when the  S02 concentration  in flue gas  exceeds  1,000  ppm.

     No trouble has  been encountered  in the  FGD  system at  ISOGO  Power  Station
     where  S02  concentration is  very  low.

     Nevertheless,  troubles related  to  scale build  up  have already been
     solved as  previously mentioned.

     The  troubles encountered till the  end of  fiscal  year  of  1977  is summa-
     rized  according to the number of troubles to have shut down FGD system
     and  the description and the plant.

     The  shutdown time of FGD system  means only  the time when  FGD  system is
     out  of service, meanwhile the plant is  in service with  low  sulphur
     fuel,  but  does  not include  the case when  a  part  of FGD system or  one
     process line is out of service.

     Table  8 shows  major troubles of  FGD system  at  each plant.
                                      547

-------
              Table  8.   Major Problems of  FGD  System  at  Each Plant
Plant


ISOGO
(Stop HourO)

TAKASAGO
(Stop Hour
544)


TAKEHARA

(Stop HourO)
Absorption Process
Description
Failure of circulation
Pump in Absorbing Tower
(Poor Manufacture)
Scaling Build up in (
Absorbing Tower, PH 1
Adjusting Tower 1
(Stop Hour 498) '-
BUF Trouble
(Stop Hour 44)
Trouble of Spray Nozzle
in Cooling Tower
/Half process \
\Line stopped/
No.

1

#1
12
#2
5
#1
2

2


Gypsum Product
Process
Description

None

Trouble on
Gypsum
Conveyor
(Stop Hour 5)



None


No.

—


1




	


Absorbent Receiv-
ing Process
Description

None


None




None


No.

—


—




—


10.    FGD WASTE WATER TREATMENT SYSTEM

      Japan has very severe waste water standards, therefore any waste water
      to be discharged from power stations must be treated to the extent it
      meets the standards.

      The following  items need to be controlled in the FGD waste water.

      o PH

      o SS

      o COD

      o Heavy  Metal

      o F
(I)    Description  Water  Treatment
                                  0
      In  a  wet  FGD  process,  corrosion  can occur  because of  the  absorbed  and
      concentrated  chlorine  ion  (Cl~)  in the  fuel.  Accordingly,  in order  to
      maintain  the  chlorine  ion  (Cl~)  concentration below a certain  level,
      blow  down water  in  the process  is  required.  Of  primary  importance in
      the blow  down operation is  the  treatment of  SS,  fluorine  ion  (F~), COD
      and some  other heavy metal  ions  and the neutralization of discharge.
                                   548

-------
     The SS, and some other heavy metals, are easily treated by using
     coagulative precipitation-filtering system; however, the required COD
     level  in the FGD waste water (10 or 20 mg/£) can not be achieved even by
     using activated charcoal absorption or oxidation with chlorine.  It has
     been found, however, that the FGD waste water can be absorbed and removed
     by using a certain plastic absorbent.  This process  is actually in
     practical use at this time.

     A water treatment system for FGD process is illustrated below:
                                                   10
                                                                 13
      1  Alkaline coagulant

      2  Coagulant aids

      3  Original water

      4  PH adjustment tank

      5  Caogulative  tank

      6  To dehydration equipment
      7  Precipitation tank
 8  Filter

 9  Acid

10  PH adjustment  tank

11  COD absorption bed
12  Alkali

13  Neutralization tank
14  Discharge
11.    UTILIZATION OF BY-PRODUCT

      Gypsum as a by-product  is mainly taken by cement producing companies.
      Utilization of gypsum  is indicated as follows and the market of gypsum
      is changeable  in Japan.

      o Tempering agent  in cement
      o Plaster
      o Construction material as gypsum board

      Table 9 shows annual product of EPDC's FGD system.

      Certain type of catalyst is used to produce good gypsum  in coal fired
      FGD process at Isogo and Takasago power stations so that size of gypsum
      crystal is large and the higher purity of gypsum is obtained compared
      to natural gypsum.

      Table 10 shows the quality of gypsum to be unloaded from the power
      stations.
                                   549

-------
      Table 9.   Annual  Gypsum Production  of  EPDC's  FGD System
                                                         (Un i t:  metric ton)
Plant
Isogo
Takasago
Takehara
Total
1971*
-
5,519
-
5,519
1975
-
30,207
-
30,207
1976
30,755
107,087
7,815
1^5,657
1977
36,608
121,880
66,890
225,378
Total
67,363
26A.693
7^,705
^06,761
                  Table 10.  The Quality of Gypsum from FGD
                       Compos it ion
   Wt
                   CaSO/t-2H20

                   CaS03-l/2H20

                   CaC03

                   R203

                   Residue insoluble
                   by acid
96% or over

 0.5% or less

 0;5 ^ U

 \% or less


 ]% or less
12.   PROBLEM IN  DISPOSAL  OF  GYPSUM

     Table 11  shows  FGD  systems  owned  by  power  utility  companies out of  those
     in service  in Japan  at  present.

     The processes of  FGD systems are  almost of limestone-gypsum and the
     production  capacity  of  gypsum  from these FGD  system  goes  up considerably.
     The market  of the gypsum  produced from  FGD systems changes heavily
     according to the  increase and  decrease of  supply and demand balance in
     the cement  industry  and the construction material  industry.

     EPDC asked  in the past  a  cement company to take  FGD  gypsum with freight
     paid by EPDC when the demand of gypsum was quite low.

     Problem will arise  in future when new FGD  systems  are built because
     sales of gypsum itself  may  become impossible  and even the disposition of
     gypsum shall be questionable.
                                   550

-------
                        Table 11.  Desulfurization  Installations  in the Electric Utilities of Japan
Power Company



E P D C



Hokkaido



Tohoku


Tokyo


Chubu






Kansal











Kyushu













Desulfurization Process



rfet, Limestone-Gypsum





rf t I' t G
, imes one ypsum
Jet , Double Alkali-Gypsum
Met, Sodium- Sulfuric Acid
Wet, Double Alkali-Gypsum
Dry, Active Carbon
Jet, Limestone-Gypsum
Wet, Sodium-Sulfuric Acid
Wet Lime G sum
yp
Jet, Sulfuric Acid
Dilution- Gyp sum

Met, Lime-Gypsum



Jet, Lime-Gypsum
Jet, Limestone-Gypsum








Jet, Limestone-Gypsum



' yP
Jet, Lime-Gypsum







Wet, Suifuric Acid Dilu-
tion Gypsum 	
Maker
Mitsui Miike Machinery
Mitsui Miike Machinery
I H I
I H I
Babcock-Hitachi K.K.
I H I
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Kawasaki Heavy Inds .
Mitsubishi-Kakoki K.
Kawasaki Heavy Inds .
Babcock-Hitachi K.K.
Mitsubishi Heavy Inds.
Mitsubishi-Kakoki K.
Mitsubishi Heavy Inds .
Mitsubishi Heavy Inds .
Chiyoda Chemical Eng.
& Construction Co., Ltd
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Mitsubishi Heavy Inds.
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Babcock-Hitachi K.K.
Mitsubishi Heavy Inds.
Kawasaki Heavy Inds .
Kawasaki Heavy Inds .
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Kawasaki Heavy Inds .

Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
I H I
Mitsubishi Heavy Inds.
Mitsubishi Heavy Inds.
Chiyoda Chemical E & C
Power Plant
Takasago
Takasago
Isogo
Isogo
Takehara
Matsushima
Matsushima
Date
Tomakomai-ShigasH
Hachinohe
Higashi-Niigata
Shin-Sendai
Niigata
Akita
Kashima
Yokosuka
Nish-Nagoya
Owase-Mita
Owase-Mita
Toyama-Shinko
Fukui
Amagasaki-Higashi
Kainan
Amagasaki-Higashi
Osaka
Osaka
Amagasaki-Higashi
Osaka
Mizushima
Tamashima
Tamashima
Shimonoseki
Anan
Sakaide
Karita
Karatsu
Ainoura
Ainoura
Karatus
Buzen

Mizushima-Kyodo
Niigata-Kvodo
Niigata-Kvodn
Sakata-Kyodo
Sakata-Kyodo
Sumitomo-Kyodo
Fukui-Kvodo
Kashima-Minami-
Kyodo
Toyama-Kyodo
Unit
1
2
1
2
1
1
2
1
1
4
1
2
4

3
1
1
1
2
1
1
2
4
2
3
2
1
4
2
3
2
2
3
3
2
2
1
2
3
1

5
1
2
1
2
3
1
2
1
Out Put
(MW)
250
250
265
265
250
500
500
350
350
250
600
600
600

600
265
220
375
375
500
350
156
600
156
156
156
156
156
156
500
350
400
450
450
375
375
375
500
500
500

156
350
350
350
350
156
250
-
250
Fuel



Coal



H&C oil
Coal

Heavy
Oil


Heavy
Oil
Heavy
Oil

Heavy &
C. Oil


Heavy
Oil



H. Oil
HSC
Oil
H. Oil
Heavy
Oil


Heavy
Oil




1976

Heavy
Oil




Start - up
1975 - 2
1976 - 3
1976 - 3
1976 - 6
1977 - 2
1980 - 1
1980 - 7
1978 - 12
1980 - 8
1974 - 2
1976 - 6
1974 - 3
1977 - 3
1977 - 9
1972 - 7
1974 - 1
1973 - 5
1976 - 3
1976 - 5
1974 - 10
1975 - 6
1972 - 3
1973 - 12
1975 - 1
1975 - 3
1975 - 12
1976 - 10
1976 - 10
1974 - 4
1975 - 7
1976 - 3
1977 - 4
1975 - 8
1975 - 10
1974 - 6
1976 - 3
1976 - 4
1976 - 5
1976 - 6
1977 - 12

1976 - 1
1976 - 1
1977 - 3
1977 - 10
1978 - 10
1975 - 12
1978 - 8
1976 - 9
1975 - 9
Gas Volume
(Nm3/H)
840,000
840,000
900,000
900,000
852,000
1,300,000
1,300,000
260,000
610,000
380,000
420,000
420,000
760,000
1,050,000
420,000
400,000
620,000
1,200,000
1,200,000
750,000
1,050,000
100,000
400,000
375,000
500,000
500,000
475,000
500,000
310,000
1.460.000
1,000,000
1,200,000
1,260,000
1,260,000
550,000
570,000
730,000
730.000
730,000
736.300

611,000
530.000
530.000
1,100,000
1,100,000
450,000
750,000
431,000
750,030
Capacity
(%)
100
100
100
100
100
75
75
25
50
50
25
25
50

25
50
100
100
100
50
100
25
25
75
100
100
100
100
66
100
100
100
100
100
50
50 ,
75
50
50
50

100
50
50
100
100
100
100
—
100
Efficiency
(%)
93.3
93.3
90
90
94.2
95
95
90
90
90
90
96
90

80
90
90
90
90
90
96
90
90
90
90
90
90
90
80
96
96
90
97
97
90
90
90
90
90
90

90
90
90
90
90
90
95
90
92.5
Remarks


Ltd.
IHI : Ishikawa J ima-Harima
Heavy Industries
Co., Ltd.



h













Additional I 11 '








tie up with Kureha
Chemical Industries





tie up with Kureha
Chemical Industries









Ul

-------
     Following  studies  are  under  way  as  a  solution  of  this  problem.

     o  Utilization  of gypsum  for  reclamation  material

     o  Change to the more marketable  material,  i.e.  elementary sulphur,  as
       one of FGD by-products.


(1)   Utilization of Gypsum  for  Reclamation Material

     Tests were performed  in  1976 by  EPDC  to  determine if  the gypsum can be
     used  as  reclamation material.

     The tests  have proved  as follows.

     o  No  dissolution of heavy  metal  was observed from the  gypsum, which was
       solidified by coagulator,  to exceed the  range of the waste water
       standards.

     o  Gypsum itself cannot help  being dissolved.

     The research in this  regard  is now  discontinued at present because  any
     effective  coagulator has been found to prevent  dissolution of gypsum
     itself.
13.   CONSTRUCTION PLAN OF NEW FGD SYSTEMS

     EPDC is now constructing two 500MW coal  fired  power  plants,  and  the
     plants are equipped with FGD systems  which  treat  approx.  3A flue gas.
     Meanwhile EPDC is also planning one 700MW coal  fired  power  plant with
     which a full capacity FGD system is scheduled  to  be  equipped.   Outline
     of new FGD systems is shown  below.

(1)   Matsushima Thermal Power Station

     Matsushima Power Station is  the one of coal  fired plants  to use  imported
     coal.  Generating capacity goes up 1,OOOMW  consisting of  two 500MW units,

     No.l plant will  be operated  commercially from  January 1981  and No.2
     plant from July  1981.

     At the Power Station, EPDC will install  wet limestone-gypsum FGD systems
     which have been  proven at the other existing Power Stations of EPDC.

     The capacity of  the FGD systems is approx.   3/^t of the generating
     capacity of plants.
                                   552

-------
(a)   Special  features of Matsushima's  FGD System

     The special  features of the FGD system for Matsushima Power Station
     are described  below.

     o In order to  burn  the various  kinds of imported  coal,  the pre-
       scrubber is  installed to remove dust and impurities in  flue  gas
       before S0£ removal.

     o The prescrubber and absorber  are spray tower type which have
       very low pressure drop and high liquid-to-gas ratios.

     o In order to  reduce consumption  of energy and plant water,  the
       reheating system  is changed from after burning  system to gas-gas
       heater of "Ljungstorm" type.

     o The process  water treatment equipment is installed to reduce
       consumption  of plant water, and to prevent  scaling troubles.


(b)   Design basis

     The FGD system is designed based  on our operating experiences  in
     existing plants, and the results  of research  and  development
     achieved by pilot plant tests.

     The pilot plants for this system were installed at Isogo  P/S and
     Takehara P/S.   The  tests were conducted for about one year.

     The design basis of the FGD system for each 500MW power generating
     coal-fired boiler at Matsushima P/S is shown  in Table 12.
(c)   Process description

     o Prescrubber

       The flue gas is introduced from the two gas-gas heaters to a
       prescrubber through the two booster-fans, and is quenched to
       saturated temperature by contacting with cooling water.

       At the same time, dust and impurities in flue gas such as HC1 ,
       HF, etc. removed in the prescrubber.

       Removal of these impurities is very important in FGD process,
       so as to be described below.

       o to maintain high desulphurization efficiencies for the various
         gas conditions.

       o to improve the quality of gypsum.
                              553

-------
     o to reduce limestone stoichiometry.

     o to prevent scaling problems,  and corrosion of equipments,


         Table 12.   Outline of FGD System  at Matsushima P/S
           Item
           Specifications
Gas flow to prescrubber


Flue gas temperature


Inlet gas S02 concentration

Inlet gas particulate

Performance

   S02 Concentration

   Particulate loading


Absorbent


By-product


Make-up water
1,300,000 Nm3/Hr (Wet)
(Approx. 970,000 ACFM)

Booster-fan outlet: 135°C
Prescrubber inlet:  Max. 110°C, Nor 85°C

1,000 ppm Dry

300 mg/Nm3 Dry (0.13 grain/SCFD)
50 ppm Dry (Absorber outlet)

30 mg/Nm3 Dry (0.0125 grain/SCFD)
(Absorber outlet)

38% CaCO?, 5-5 Ton/Hr (Approx.)
Size: 325 Mesh pass over 95%

Gypsum: 10.5 Ton/Hr (Approx.)
Purity: Over 95% CaSOlj 2H20

Raw water: ^9 Ton/Hr
       The quenched gas is sent to the absorber through the vertical mist
       eliminator which is located in outlet duct of the prescrubber
       in order to prevent entrainment of mist accompanied in quenched
       gas.

     • Absorber

       The outlet gas from the prescrubber is led to a absorber, and the
       S0£ in the flue gas is absorbed in the absorber by droplets of
       the limestone slurry which is sprayed from absorption spray
       nozzles.

       At the top of absorber, a horizontal  type mist eliminator is
    *   equipped.
                              554

-------
  The cleaned  gas  is  fed  to gas-gas  heaters through this mist
  eliminator,  where the entrained slurry droplets are separated.

• Reheating system

  The desulfurized gas should be reheated for the purpose of
  protection of duct  and  stack corrosion, improvement of the
  atmospheric  diffusion effect of flue gas from stack, and preven-
  tion of the  white plume generation.

  A "Ljungstom" type  gas-gas heater  is designed based on the results
  of research  and  development achieved by pilot scale tests at
  Takasago P/S.

• Gypsum production process

  The slurry bled  from the absorber  includes calcium sulfites,
  calcium sulfates and limestone unreacted in the absorber.

  The Calcium  sulfites are oxidized  to gypsum by air in the oxida-
  tion tower.

  The limestone unreacted is converted to gypsum by addition of
  sulfuric acid.

  The gypsum is concentrated in the  thickner, and then is dehy-
  drated by the centrifuges.

• Process water treating system

  A part of process water is bled from the thickner to keep water
  balance of process, because the fresh water is supplied to the
  mist-eliminator  in  the absorber.

  The process  water treating equipment is equipped for reusing.

  The SS and Ca+ in the bleed water  are treated by this equipment
  and then is  supplied to the prescrubber.

          Water quality (outlet)      100 ppm (as Ca+)

          Water quality               15 m /Hr

• Waste water  treating system

  A part of the prescrubber recycle  liquor is bled from the pre-
  scrubber to  keep the dissolved chlorine concentration in recycle
  liquor at 5,000  ppm or less, and to discharge the dust removed,
  which  is  preferable for the materials of construction.

  The waste water  is  treated in this system and then is discharge
  to the sea.
                         555

-------
To Sea
     Boiler     Hot E.  P.  Air Heater  I. D. Fan
                                                                                                                     Limestone
                                                                                                                     Slurry Pit
                                                                  Gypsum
                                        Figure  7.   The  Flow  Chart  of Matsushima's FGD

-------
(2)   Takehara  Thermal  Power  Station

     EPDC  is  planning  to extend  No.3  plant  at  Takehara  Power  Station.
     Table 13  is  the outline of  extension  plant  but  the specification  has
     not  been  finalized.
                Table 13.   Takehara  No.3  Plant  Extension  Plan
              Generating  capacity

              Fuel

              FGD capacity

              Process

              Commercial  operation
700 MW

Coal

2,700,000 Nm3/H (Full  capacity)

Limestone-gypsum

July, 1982
     Specific features of the FGD system design  for  No.3  plant  are  summarized
     as Table 14.
                Table ]k.   Specific Features  of FGD  System for
                           Takehara No.3 Plant
           o Treatment  of Nitrogen Removal  from FGD  Waste  Water

           o New Position of DeSOx Fan
     (a)   Treatment of nitrogen removal  from FGD waste water

          As DeNOx system is installed for No. 3 plant, ammonia  leaked  from
          DeNOx system goes into FGD system where it is dissolved  in the  process
          water and discharged  in the FGD Waste Water after concentrated.

          Biological  treatment  is introduced to remove nitrogen ion such  as
                contained in the FGD waste water.
          The success of developing biological  treatment system is due  to
          the effort of some Japanese manufacturers which have been studying
          jointly with EPDC the removal  of nitrogen from the FGD waste  water.


     (b)   New position of DeSOx fan

          Gas-gas heater (called as GGH) will  be applied like the FGD system
                                   557

-------
     of Matsushima Power Station, however, the severer SOx regulation
     will  be imposed to No.3 plant of Takehara Power Station by a  local
     autonomous body and any boost up fan cannot be installed  in series
     of ID Fan like that of Matsushima as shown in Figure 8, because the
     leakage of flue gas in GGH will  cause a drop of total desulphuriza-
     tion  efficiency.

     With  this reason, if the position of the fan is moved to  the one
     between GGH and FGD system (called as DeSOx Fan)  the leakage through
     GGH will occur from the treated  gas side to untreated gas side, then
     the desulphurization efficiency  of the FGD system becomes the same
     value of total  desulphurization  efficiency of the plant.
          Matsushima P/S

         IDF  B.U.F.
          Stack
Takehara No.3

IDF         DeSOx Fan
-^
u
n

I
/
iPi
s d uct
G
G
H






FGD






                 leak gas

                Figure 8.  Location of DeSOx Fan
                                                New Position   leak gas
STUDY OF NEW FGD PROCESS

It is indispensable to take full countermeasures for flue gas treatment
of coal  fired power stations in Japan along with enforcement of severer
popullation regulations.

EPDC has achieved a good prospect of commercialization of NOx Removal
System through the past researches and developments.

EPDC's papers were presented in this regard at DeNOx Seminar in Denver
Colorado in October 1978 sponsored by EPRI.

Problems are expected to arise on the FGD  Systems  installed at coal
fired power stations because the waste water treatment system will become
complicated with an introduction of ammonia treatment system in conjunc-
tion with NOx removal system.

The other problems may arise in future on  the disposal of FGD gypsum
which will  be produced excessively in the  market.

EPDC is challenging further studies on new FGD systems bearing  the back-
grounds in mind.
                              558

-------
(1)   Studies of New FGD System

     New Dry FGD has recently been  developped.   The process applies activated
     carbon with ammonia  injection  which evaluated as superior as to wet
     process in terms of  performance  capability and operation economy.

     The dry process will prove  better  evaluation for EPDC with a combination
     with NOx Removal System.

     Table 15 shows the advantage of  the new dry process.
                  Table 15.  Advantage of The New Dry Process
          o Simpler Flue Gas Treatment System at a coal fired power  plant

          o Elemental Sulphur  as  a  by-product

          o Simpler Waste Water Treatment System
(2)  New FGD System

     Figure 9 shows the comparison of flow diagrams for New FGD system  and  the
     FGD system and the FGD  system which is planned for Takehara No.3 Plant.
             Wet Type
                        NHi
(p
\s


vv

[i







-o-
IDF

T
-o-
DeSOx
Fan

Wet
J|
FGD
              Boiler
                     H.ESP   SCR   A/H
                                       GGH
              New Dry Type
                                                        Stack
                                             Waste Water Treatment
                                             (with De-N treatment)
                                                 V
                   BUF
             Boiler   H.ESP
                             SCR
A/H
                                           Dry FGD   BH
                                                          Stack
  H.ESP: High Temperature  Electrostatic Pricipitator   A/H:  Air Preheater
  SCR: Selective Catalytic Reduction System  GGH: Gas/Gas  Exchanger
  IDF: Induced Draft  Fan   BUF:  Boost-up Fan   BH: Bas  House

           Figure 9.    New FGD  System and FGD System for Takehara No.3
                                        559

-------
EPOC is conducting test  of 10,000 Nm3/H at a EPDC's power station
starting the test from November 1978.

EPDC will  present details after observing tests status and others.


SUMMARY

(1)  EPDC  has judged Wet Limestone-gypsum process as the best process.

(2)  Basically no problems have been encountered on Wet Limestone-gypsum
     process.  The FDG systems of its process have been in service
     satisfactorily and its availability is the same as the boiler
     availabi1ity.

(3)  Desulphurization efficiency changes slightly according to boiler
     load  variation but at least more than 85% of the efficiency is
     maintained  in average.

(4)  GGH and Denitrogen Treatment  is applied, as new peripheral
     technology, to FGD systems at newly built power stations.

(5)  New FGD is  under study in connection with the introduction  of NOx
     Removal Systems and its researches and developments are underway.
ACKNOWLEDGEMENT

The fact that FGD systems have been in service with high  reliability and
good performance is mainly dependent upon special  attention and
maintenance of the engineers and the operators working at EPDC's Power
Stations a's well as the efforts exerted by the manufacturers which
have been making joint development with EPDC.
REFERENCE:

1)  T.  Hayase,  K.  Mouri  ;
"State of Development and use of the
 Desulphurization in Japan"

 Symposium of state of development
 and proving in operational tests
 of desulphurization of Waste Gases
 of Power Stations by Czechoslovac
 Scientific and Technical Society.

 March 1978
                              560

-------
CURRENT  ALTERNATIVES FOR FLUE  GAS DESULFURIZATION  (FGD)

              WASTE DISPOSAL - AN ASSESSMENT



                                By

    Chakra J. Santhanam, Richard R.  Lunt and  Charles  B.  Cooper

                      Arthur D.  Little,  Inc.
                     Cambridge,  Mass.    02140



                             ABSTRACT

       With increasing coal utilization  in industrial and
       utility boilers, generation of coal ash  (fly ash  and
       bottom ash), and flue gas desulfurization  (FGD) wastes,
       which together comprise flue  gas  cleaning  (FGC)
       wastes, is expected to increase dramatically over
       the next twenty (20) years.   Since  most  of  the FGC
       wastes generated will be  disposed of,  rather than
       utilized, these wastes represent  significant poten-
       tial sources of environmental pollution  unless proper
       disposal technology is employed.  Continuing research
       and development efforts sponsored by EPA, EPRI, and
       others in recent years have provided substantial  infor-
       mation on environmentally sound disposal techniques.

       This paper discusses the  current  state-of-the-art of
       flue gas desulfurization  (FGD)  waste disposal  with
       focus on wastes from nonrecovery  systems.   The paper
       includes a review of the  following  areas:

          •  Production and categorization of FGC  wastes;

          •  Disposal options presently  in use  and/or with
             future potential; and

          •  Environmental issues  associated  with  various
             disposal options.
                               561

-------
1.0  INTRODUCTION

     As coal utilization in utilities and large industrial boilers
increases, the quantity of flue gas cleaning (FGC) wastes, particularly
those associated with flue gas desulfurization, will increase dramatically.
The preponderant part of these FGC wastes will be discharged disposal.
Over the long term, utilization is expected to grow but at a slower rate
than that of FGC waste generation.  Table 1 shows projections of coal ash
and flue gas desulfurization (FGD) wastes through 2000.

     In the past, utilities operating FGC systems have typically disposed
of wastes by storage in ponds, often without provision for control of over-
flows or seepage into groundwater.  However, several factors will dra-
matically influence disposal options in the coming years.

     a.  An increase in coal-fired capacity in the United States.
         In 1976 the total U.S. coal-fired electric utility gen-
         erating capacity was estimated at over 191,000 MW in 399
         plants  (3).  The estimated capacity is expected to increase
         by 1986 to over 326,000 MW (4).  Use of coal in large industrial
         boilers (+25 MW equivalent or larger)  is likely to further
         increase the total coal-fired capacity.(1)

     b.  A major increase in the application of scrubber technology
         by utilities and a consequent increase in FGD waste genera-
         tion.  At present over 16,000 MW of generating capacity at
         some thirty plants utilize FGD systems.  As of September
         1978, over 59,000 MW of capacity have been committed (5).
         Future  increases are likely to be even more dramatic.

     c.  Advances in stabilization technology for FGD wastes which
         permit  landfill disposal of partially dewatered solids
         instead of ponding of difficult to handle sludges.  In the
         future, disposal of wastes in managed  fills is likely to
         be encouraged.  In many cases this will require stabiliza-
         tion prior to disposal.

     d.  Regulatory developments including the Clean Air Act of
         1977 and the Resource Conservation & Recovery Act of 1976
         (RCRA).  New Source Performance Standards (NSPS)  for cri-
         teria pollutants are now under review by the EPA and may be
         significantly tightened.  Similarly, the recent issuance of
         proposed guidelines provides impetus to environmentally
         sound disposal of FGC wastes.

     R&D efforts and regulatory development have focused on characteri-
zation and categorization of FGC wastes, determination of the mechanisms
by which pollutants may enter the environment from a disposal site,
assessment of the environmental impacts and development of disposal
criteria and guidelines.
                                   562

-------
UJ
                                                         Table 1




                                     Projected Generation of Coal Ash and FGD Wastes
                                                                                PROJECTED
Coal Ash
Industrial
Utility
Total
FGD Wastes
Industrial
Utility
Total
- 1975 1985
10 Metric Tons % of Total 10 Metric Tons
8,590
64,440
52,060 - 73,030
1,090
21,050
6,200 - 22,140
% of Total
12
88
100
5
95
100
2000
10 Metric Tons
19,950
84,800
104,750
5,260
29,860
35,120
% of Total
19
81
100
15
85
100
        Source:  (1,2)

-------
2.0  GENERATION OF FGD WASTES

2.1  Overview on FGC Technology

     Ash Collection Technology

     Coal-fired utility and industrial boilers generate two types of coal
ash—fly ash and bottom ash.  (Economizer ash and mill rejects are lumped
into the two major categories here.)  Both constitute the non-combustible
(mineral) fraction of the coal and the unburned residuals.  Fly ash, which
accounts for the majority of the ash generated, is the fine ash fraction
carried out of the boiler in the flue gas.  Bottom ash is that material
which drops to the bottom of the boiler and is collected either as boiler
slag or dry bottom ash, depending upon the type of boiler.

     The total amount of coal ash produced is directly a function of the
ash content of the coal fired.  Thus, the total quantity of ash produced
can range from a few percent of the weight of the coal fired to as much as
35%.  The partitioning of ash between fly ash and bottom ash ususally de-
pends upon the type of boiler.  Standard pulverized coal-fired boilers
typically produce 80-90% of the ash as fly ash.  In cyclone-fired boilers,
which are frequently used to burn lignite, the fly ash fraction is usually
somewhat less; in some cases bottom ash constitutes the majority of the
total ash created.

     Collection of bottom ash (or boiler slag) does not involve systems out-
side the boiler itself.  The key technology issue is the handling of
bottom ash.  Fly ash, however, is a major source of particulate emissions
and with regulatory requirements has required major collection systems.
Control of particulate emissions from pulverized-coal-fired steam genera-
tors is rapidly becoming a significant factor in the siting and public
acceptability of coal-burning power plants.  The particulate emissions
limit under current NSPS set by the EPA for large, new coal-fired boilers
is 0.043 grams/106 joules (0.1 lb/106 Btu).  Some states have requirements
more restrictive than this.  Furthermore, the NSPS are now under review
and are expected to be tightened significantly.

     Fly ash carried in the flue gas stream can be collected in a number
of ways to meet the current particulate emission control limitations as
noted above.   Typical methods historically employed include mechanical
collection, electrostatic precipitation, fabric filtration and wet scrub-
bing.  However, the tightening regulatory requirements support two
criteria for future fly ash collection systems:

     •  The collector must be efficient in removing sub-micron
        particulate matter.  This criterion eliminates all
        mechanical collectors and many wet scrubber systems from
        consideration as the only systems.  Mechanical collectors
        may,  however, function as a first unit followed by a more
        efficient collector.
                                   564

-------
     •  The collector must be available commercially and be proven
        in a utility boiler application.  This constraint eliminates,
        for the immediate future, many hybrid wet scrubber systems
        and novel collectors that are now under development.  In the
        long run, however, it is conceivable that such advanced sys-
        tems may be used at least in some instances.

     It appears that electrostatic precipitators and fabric filters will
be the only systems capable of meeting the requirements in the foreseeable
future.  In addition, developments in effluent guideline standards strongly
point to dry fly ash handling systems in new plants (at present, the major
portion of ash is handled by wet sluicing where dry collectors are employed).

     FGD Technology

     The implementation of  flue  gas desulfurization (FGD) technology for
the control of SC>2 emissions from the combustion of fossil  fuels in
industrial and utility boilers is rapidly growing in the United States.
At present, FGD systems are in operation on over 16,000 megawatts of
utility generating capacity at some 30 different plants throughout the
country, and more than 40 industrial steam plants are equipped with FGD
systems.  By the end of 1979, the total capacity of FGD systems in opera-
tion on utility and industrial boilers is expected to exceed 25,000
megawatts (equivalent).  The degree of SC>2 control ranges from less than
50% SC>2 removal efficiency  to over 90%, depending upon the  type of FGD
systems, the sulfur content of the fuel, and the applicable S02 emission
regulations.

     The growth in FGD systems on fossil-fuel-fired boilers in the United
States over the next 20 years will be principally dependent upon the growth
in utility and industrial boiler capacity, current and future S02 emission
regulations, and the impact of alternative desulfurization  approaches to
current and developing FGD  technology.  An important factor may be the use
of existing and enhanced coal-cleaning techniques.

     A wide variety of FGD processes have been developed for application
on utility and industrial boilers.  In general, the technology can be
grouped into two categories:  nonrecovery, or throwaway systems, which
produce a waste material for disposal; and recovery systems, which produce
a saleable byproduct (either sulfur or sulfuric acid) from  the recovered
S02.  Nonrecovery processes make up the overwhelming majority of the
technology.  Nine different processes and process variations can be con-
sidered to be commercially available, seven of which are nonrecovery systems,
These seven processes constitute more than 95% of the capacity currently in
operation on utility and industrial boilers, a trend which  is expected to
continue for the foreseeable future.  Table 2 summarizes the applications
of FGD process technologies for  systems expected to be in operation by the
end of 1979.
                                   5:65

-------
                                                Table 2
                          Summary of FGD Systems Expected To Be In Commercial
                          Operation on Utility and Industrial Boilers in 1979
Utility
                                                                                  Industrial
Nonrecovery
  Direct Limestone-Conventional
                   Forced Oxidation
  Direct Lime
  Alkaline Ash
  Dual Alkali
  Once-Through Sodium
  Ammonia
    Total
Recovery
  Wellman-Lord
  Citrate
        Q
  Mag-Ox
    Total
No. of Plants
19
13
2
3
1
0
38
2
0
0
2
Capacity (MW)
11,780
7,305
1,170
1,105
510
—
21,870
735
—
—
735
No. of Plants
1
1
0
8
26
_7
43

0
1
_0
1

3
Capacity (10 scfm)
50
85
—
1,082
4,954
552
6,722
(^3000 MW-eq)
	
104
—
104
(^50 MW-eq)
 Two systems have been commercially operated on utility coal-fired
 boilers but these are not currently in operation.
Source:  Arthur D. Little, Inc.

-------
     Nonrecovery Systems;  Nonrecovery processes in general can be sub-
divided into two groups, wet processes and dry processes.  Wet processes
involve contacting the flue gas with aqueous slurries or solutions of
absorbents and produce wastes in the form of solutions or slurries for
direct discharge or further processing prior to disposal.  In some cases,
waste slurries are partially dewatered and further processed to produce a
soil-like material for landfill.  Dry processes, on the other hand,-pro-
duce essentially moisture-free solids through dry injection of absorbents
into the flue gas or the use of spray dryers.  All nonrecovery processes
now in operation as well as those due to come on line in 1979 involve wet
scrubbing.  However, a number of contracts have been signed for the appli-
cation of dry systems to utility boilers which will start up in the early
1980's.

     Of the seven different types of nonrecovery processes now in com-
mercial operation on industrial and utility boilers, five involve conver-
sion of the SC>2 to some form of solid waste  (sludge) for disposal in
either wet ponds or landfills:

     •  Conventional direct lime scrubbing,
     •  Conventional direct limestone scrubbing,
     •  Limestone scrubbing with forced oxidation,
     •  Alkaline fly ash scrubbing, and

     •  Dual alkali.

     Two systems produce a soluble waste which is discharged as an aqueous
liquor to holding ponds or wastewater treatment systems:

     •  Once-through sodium scrubbing, and
     *  Ammonia water scrubbing.

     As shown in Table 2, essentially all utility applications of non-
recovery technology involve solid waste-producing systems.  In contrast,
a large majority of industrial boiler applications of FGD involve the
production of liquid wastes.

     Both  types of wet  scrubbing nonrecovery systems  can usually withstand
relatively high levels  of  particulate,  and  many  in  the  past have been
designed  for  simultaneous  S02  and  particulate  removal.   Approximately 40%
of  FGD systems currently in  operation on  utility boilers and  about  80% of
those  in operation on industrial boilers  serve as combined particulate and
S02  control systems.  However,  most  systems  being installed today on
utility-scale boilers follow high  efficiency electrostatic precipitators
in  order  to ensure reliable  service  of  the  FGD system.

     Dry nonrecovery processes  have  not yet been commercially demonstrated
in  the United States, although  at  least three  systems  for full-scale
utility applications are in  the early stages of  planning or design.
                                   567

-------
     Three different approaches to dry scrubbing for producing solid
wastes have been actively pursued (6):
     •  Injection of solid sorbents into the flue gas stream with
        collection of sorbents downstream in a particulate control
        device;
     •  Injection of solid sorbents into the boiler combustion
        zone; and
     •  Contacting of flue gas with alkali sorbent slurries in
        a spray dryer.

     All of these approaches involve simultaneous particulate and S0~
control, and all offer the advantage of not requiring flue gas reheat,
which wet processes generally do require.

     Recovery Processes;  As in the case of nonrecovery processes, re-
covery processes can also be categorized into wet and dry according to
the mode of SO- removal.  They can be further classified according to the
type of byproduct produced:  concentrated SCL for conversion to sulfur or
sulfuric acid; sulfur only; or acid only.

     At present, only two process technologies have been commercially
demonstrated on large industrial - or utility-scale boilers—the Wellman-
Lord process and magnesium oxide scrubbing.  Another, the citrate scrub-
bing process, is currently being commercially tested on a large industrial
boiler.  All three of these are wet scrubbing processes.

     The total capacity attributable to these three technologies
(including magnesium oxide system not now in operation) is less than 5%
of the total FGD operating capacity in 1979; and market share is expected
to remain below 5% of the total installed FGD capacity on boilers in the
United States through the mid-1980's.  Since recovery systems represent
such a small fraction of the market, and produce only a small amount of
wastes in comparison to nonrecovery systems, discussions throughout the
remainder of this paper will focus on wastes from nonrecovery systems.

2.2  General Composition and Categorization of Nonrecovery FGD Wastes

     Major Components

     The quantity and characteristics of FGD wastes produced from a com-
bustion system depend on a variety of factors including:

     •  Composition of the coal (ash and sulfur content);

     •  Type of combustion (boiler) system and its operating
        conditions;

     •  Type of particulate collection system and its operating
        conditions;

     •  Type of FGD system and its operating conditions; and

     •  Degree of SO™ control required.
                                   568

-------
     The principal substances making up the solid phase of FGD wastes are
calcium-sulfur salts (calcium sulfite and/or calcium sulfate) along with
varying amounts of calcium carbonate, unreacted lime, inerts, and fly ash.
The ratio of calcium sulfite to calcium sulfate is a key parameter (the
latter, usually present as CaS04 • 1/2 H20 or as gypsum, CaSC>4 ' 2H20) and
will depend principally upon the extent to which oxidation occurs within
the system.  Oxidation (and thereby sulfate content) is generally highest
in systems installed on boilers burning low sulfur coal or in systems
where oxidation is intentionally promoted.  When the sulfate content of
the waste solids is low, calcium sulfate can exist with calcium sulfite
as a solid solution of hemihydrate crystals (CaSOx • 1/2 H20).  Data from
pilot plant, prototype, and full-scale FGC system operations indicate
that up to about 25% of the total calcium-sulfur salts can be present as
CaSO^ • 1/2 H20 in solid solution with CaS03 • 1/2 H20.  At higher calcium
sulfate levels, gypsum (CaSO^ • 2H20) becomes the predominant form of
calcium sulfate.  It is expected that at very high levels of oxidation
(greater than 90% oxidation of the S02 removed) calcium sulfite can also
form a solid solution with gypsum (CaSO   • 2H20) analogous to the solid
solution of hemihydrate salts formed at low sulfate levels.

     Because the  differences  in  the  crystalline morphology  of hemihydrate
and dihydrate  solids not  only reflect  the chemical  composition but also
to a large  extent dictate the physical  and engineering properties of FGC
wastes,  it  is  convenient  to  classify FGC  wastes on  the basis of  the cal-
cium sulfate content.   Three  such categories have been selected, as follows:

           Category                       Predominant  Crystalline Form
Sulfate-rich (CaS04/CaSOx <  0.90)           Dihydrate
Mixed  (0.25 >  CaS04/CaSOx ^  0.90)           Dihydrate  and hemihydrate

Sulfite-rich (CaS04/CaSOx <  0.25)           Hemihydrate

where  CaSOx is  the  total  calcium-sulfur salt content.  This categorization
will be  employed  in the ensuing  discussions throughout this paper.

     Factors which  tend to influence the  amount of  sulfite  in FGC wastes
(i.e., the extent  of oxidation)  include:   boiler excess air, type of
scrubber, use  of  forced oxidation, presence of oxidation inhibitors or
catalysts in fly  ash,  reagents  (or water  makeup), type of reagent, pH in
the scrubber loop, sulfur content of the  coal and the  degree of S02 removal.

     In  general,  it is  possible  to relate the three  general categories
of wastes indicated above and their  associated crystalline morphologies
with various types of  FGC process technologies and  their applications
according to the  coal  sulfur  content.   Such a matrix  relationship is
shown  in Table  3.  As  indicated, dual alkali and conventional direct
lime scrubbing  systems  using  either  carbide or Thiosorbic lime almost
exclusively produce sulfite-rich wastes.   Such systems are  generally
applied  to medium and  high sulfur coal-fired boilers,  and attempts^are
made to minimize  oxidation.   On  the  other hand, alkaline ash and limestone
                                   569

-------
                                                  Table 3
No.  Waste Type

1.   Sulfite-Rich  (CaSO  .1/2
                       A
     Mixed Sulfite/Sulfate
      (CaSOx-l/2 H20
                                 Matrix  of  Untreated FGD Wastes Generation-
                                 Nonrecovery  Solid Waste Producing Systems
Chystalline
Morphology

Needles

Platelets

Agglomerates


Needles or Platelets
Low/Med. Sulfur
DLd/AAG  DLSf
                                                                              LSFO8  DAh
                                                                                       DL
Sulfur Coal
DLS   LSFO  DA
Sulfate-Rich (CaSO,.-1/2
 CaS0..2H00)
     4   2
                                        Platy

                                  +/or  Needles
  Sulfite-Rich = CaSO./CaSCL. x < .25
                     4     x   —
  Sulfate-Rich = CaSO,/CaSOx x ^ .9

 bLow/Med.  Sulfur Coal £ 2% S

 'lligh Sulfur Coal >2% S

  Conventional Direct Lime Process

 £
  Alkaline  Ash Scrubbing
                                           or
                                        Platy
                                                          Conventional Direct Limestone Process
                                                         o
                                                          Limestone with Forced Oxidation

                                                          Dual Alkali Process

                                                          Resembling rhombohedral cleavage fragments

                                                         Notes:

                                                           J Refers to the particular waste type as the
                                                             common waste product from the type of coal
                                                             and process.

                                                           1 Some question on this.

-------
forced oxidation systems produce  sulfate-rich wastes  almost  exclusively.
And conventional direct lime  (using  commercial  lime)  and limestone  systems
can produce either sulfite-rich,  sulfate-rich,  or mixed wastes  depending
upon the sulfur content of  the coal  and the manner in which  the scrubber
systems are operated.

     Fly ash will be a principal  constituent of FGD wastes only if  the
scrubber serves as a particulate  control device in addition  to  S09  re-
moval, or if fly ash separately collected is admixed  with the wastes from
SC>2 scrubbing.  The amount  of fly ash, therefore, can range  from nil to
as much as 80% of the total dry weight of the wastes  produced.   More than
85% of the total weight of  fly ash is made up of silica, alumina, and iron,
calcium and magnesium oxides.  These will appear in the wastes  to the ex-
tent that fly ash is present.

     Minor and Trace Components

     FGD waste solids from wet scrubbing processes carry with them  oc-
cluded liquor which contains dissolved solids.  The amount of liquor is
a function of the degree of dewatering prior to discharge.   The  major
soluble ions usually present include calcium, chloride, magnesium, potassium,
sodium, sulfite and sulfate.  Together, these can amount to  as much as
10-15% of the two solids (dry basis).

     A variety of trace elements  are also present in  FGD wastes  and derive
from a number of sources:   coal,  where they are present either  in mineral
impurities or as organometallic compounds; makeup chemicals  for  the FGD
system; and FGD process makeup water.  The principal  source  of  trace ele-
ments, though, is from the  coal;  and the levels of trace elements depend
primarily on their level in the coal, the amount, if  any, of ash that is
collected or admixed with the wastes, and the efficiency of  the  scrubber
system in capturing trace metal vapors and fine particulate.  Since most
of the elements in coal are not highly volatile and will be  retained in
the ash matrix (either as fly ash or bottom ash), the presence and  concen-
trations of most trace elements will depend upon whether fly ash is simul-
taneously removed with SC^ or admixed x^ith the waste  calcium-sulfur salts.
The concentrations in the wastes  of those elements that are most highly
volatile (notably arsenic, mercury, selenium, beryllium, chloride,  and
fluoride) will be a function of the extent to which they are present and
released from the coal and, more  importantly, the efficiency with which
they are captured in the scrubber.

2.3  Dewatering of FGC Wastes

     Most unthickened slurry wastes produced by FGC systems  contain on the
order of 5-15 wt% suspended solids.  In order to avoid the unnecessary
discharge of large amounts of process liquor, these wastes are  frequently
mechanically dewatered prior to being discharged from the process.  Primary
dewatering is usually accomplished using thickener/clarifiers or settling
ponds.   Primary dewatering is virtually universally practiced in order to
reduce sludge volume and conserve water.   Secondary methods  of dewatering are
also sometimes employed.   These include vacuum filtration and centrifugation.
Secondary dewatering is only employed as  a precursor  to dry  impoundment

                                   571

-------
in order to improve the handling properties of the wastes prior to truck
transport or stabilization.  In general the dewaterability of FGD wastes
varies with the sulfite/sulfate content of the waste and the amount of
fly ash present.  Sulfite-rich wastes can typically be dewatered to 40-65
wt% solids, while sulfate-rich wastes can usually be filtered to 65-85%
solids.  The presence of fly ash and unreacted limestone can often improve
the dewaterability, but significant improvements usually only occur for
wastes with poor dewatering characteristics (within each general category
of waste type).

     Table 4 summarizes dewatering practices for full-scale FGC systems in
operation on utility boilers as of November 1978 and shows some interesting
trends in dewatering practices in the utility industry:

     •  No simultaneous S02 and fly ash control systems or wet par-
        ticulate scrubbing systems employ secondary methods of
        dewatering (i.e., filtration or centrifugation) for FGC waste
        dewatering, although a number of the plants do dispose of
        wastes via dry impoundment of wastes reclaimed from secondary
        settling ponds.
     •  The overwhelming majority of the FGD capacity  for S02 removal
        only  involves  thickening and filtration or centrifugation  for
        dry impoundment of the wastes.  This trend is  expected to
        continue for the foreseeable future.  About 6,700 megawatts
        of new, nonrecovery FGC capacity producing solid wastes are
        expected to be on-line in 1979, all of which will be devoted to
        SC>2 control only.  Of this total, approximately 85% will utilize
        some  form  of dry impoundment for waste disposal, and more  than
        two-thirds of  these will employ either filtration or centri-
        fugation for waste dewatering.

2.4  Stabilization Processes

     There are now more than two dozen processes for solidification/
stabilization of many types of sludges and difficult to handle wastes.
The state of  development of these processes ranges from laboratory-
scale on a number  of different types of wastes.

     There are basically three methods by which stabilization processes
can improve the disposability of wastes.

     •  First, through improvements in the physical characteristics
        of the wastes to the extent that they are more easily
        handled.   This frequently leads to better control/management
        of the disposal area, resulting in reduced impacts relating
        to physical stability and contamination of ground and surface
        waters.

     •  Second, through decrease in the exposure of the wastes by
        reducing surface area and/or permeability or by encapsu-
        lating the wastes, thus limiting the contact of groundwater
        (or infiltration water) with the waste.
                                   572

-------
Ui
                                                         Table 4
                        Summary of FGC Waste Dewatering Practices  for Operating  Utility  Scrubbers'
                                                               Dewatering  Practices Employed
       Scrubber System Mode
S CL Removal
- Low Sulfur Coal
- High Sulfur CoalC
S02 + Ash Removal
- Low Sulfur Coal
- High Sulfur CoalC
            Wet Particulate Removal
            - Low Sulfur Coal
            Total
Pond Settling

3/1570
1/550
2/1085
2/885
7/1220
15/5310
Thi ckening
(No. of
2/365
3/185
3/2185
1/1650
3/865
12/6250
Thickening/
Pond Settling
Thickening/
Filtration
Thickening/
Centrifugation
Plants /Total Capacity, MW)

—
—
—
2/1175
2/1175
3/1045
6/2515e
—
—
—
9/3560
1/1585
—
—
—
—
1/1585
        Basis:   November 1978
       3Generally _< 1.5% sulfur
       "Generally > 1.5% sulfur
        In addition to dewatering,  settling pond acts  as  final  disposal site in 10/3330 of those indicated.
       ^Includes two plants (totaling 920 MW)  whose  scrubber  systems remove ash but have ESP's for primary ash removal.

-------
     •  Finally, by chemical reaction with the waste, limiting the solu-
        bility of chemical constituents that would otherwise be
        readily accessible either through flushing of interstitial
        liquor or solubilization.

     Different techniques usually emphasize one or two of these factors.
The applicability and "success" of a particular process, therefore, will
depend importantly upon the chemical and physical properties of the waste,
the disposal site characteristics, and the waste-handling constraints.

     At the risk of oversimplification, most all stabilization processes
generally can be categorized into one of about six groups, according to
the manner in which the wastes are treated.  Table 5 lists the principal
processes of each type, indicating the vendors and status of the process.

     Stabilization of FGD Wastes

     A number of the processes listed in Table 5 have been tested on FGD
wastes, mostly in the bench scale.  Justification of the use of additives
to improve the physical characteristics of FGD wastes has been based on
improvement in strength, reduction in compressibility, and reduction in
permeability caused by an increase in solids content or the formation of
permanent bonds between particles.  The additives most advantageous then
would be those available at low cost in large quantities (e.g., fly ash)
and those effective as cementing agents (e.g., Portland cement).   Combi-
nations of additives may produce both types of improvement (e.g., fly
ash plus lime).  A limited amount of study has been devoted to the evalu-
ation of simple additives such as fly ash, lime, and Portland cement.
These studies are discussed later.

     At present, there are two approaches which have achieved commercial
applicability for calcium-based FGD wastes:  addition of lime and fly ash
for dry impoundment systems (currently marketed by IU Conversion Systems,
Inc. , and others) and the proprietary technology developed by the Dravo
Corporation involving the use of processed blast furnace slag as the
additive for stabilization in wet ponds.  Other additives and stabiliza-
tion approaches for calcium-based wastes have been laboratory- and field-
tested but are not being actively marketed at present.

     The economic evaluation of the use of additives for waste stabiliza-
tion is site specific, at best, and must take into account not only the
applicable disposal regulations, but also the type of waste and the dis-
posal area hydrogeology.  In some cases, for example(dry impoundments)
it may be possible to stabilize materials to form containment dikes and
basal layers into which unstabilized materials could be placed.  This
would, of course, depend upon the handling properties of the untreated
wastes.
                                   574

-------
                                                                                  Table 5

                                                                        Waste Treatment Processes
             Process Type
                                                                                                  Degree of
Ln
^j
Ui
Cement (Lime) -Based I-UCS , Inc.
Dravo Lime Co.
TJK, Inc.
Chem-Nuclear System, Inc.
Commonwealth Edison/American
Admixtures, Inc.
Aerojet Liquid Rocket
Sludge Fixation Technology, Inc.
Research Cottrell
Envirotech (Chemfix)
fiuu-L La. ve&
Ash & Lime
Slag & Lime
7
Cement

Lime & Fly Ash
Cement
?
Lime & Fly Ash
Lime & Fly Ash
liUUUUel UJ-<1XJ.£(1LXOI1
Commercial/U.S.
Commercial/U.S.
Commercial/ Japan
Comnercial/U. S .

Commercial/U.S.
Laboratory (FGD)
Laboratory (FGD)
Field (FGD)
7
       Self-Cementing
        (plaster of paris)

       Silicate-Based
Marston Associates

Environmental Technology Corp.
Envirotech (Chemfix)
Ontario Liquid Was~te Disposal, Ltd.
United Nuclear Industries
Stablex Corporation
                                       None
Silicate + Cement

Silicate + Cement
Silicates
Pilot (FGD)

Commercial/U.S.
Commercial/U.S.
Commercial/Canada

Commercial/U. K.
                                                                                                                   Commercial Waste Application

                                                                                                                   FGD, Fly Ash
                                                                                                                   FGD, Fly Ash,  Mine Tailings
                                                                                                                   Industrial Inorg., Dredge Spoils
                                                                                                                   Utility Radwastes

                                                                                                                   FGD

                                                                                                                   Industrial Heavy Metal Sludges
                                                                                                                   None
                                                                                                                   None
Metal Hydroxide
Org. & Inorg. Industrial, Sewage
Industrial
(Radwastes)
Industrial
Level of Testing/Operation
     with FGD Wastes	

     Full-Scale
     Full-Scale
  Full-Scale (Japan)
     Unknown

     Full-Scale
     None
     None Reported
     Field
     Lab?

     Field

     None Reported
     Field
     Field
     None
     None Reported
       The rmoplas t i c



       Organic Polymer




       Inorganic Precipitation

       Unknown
Werner & Pfleiderer Corp.
Southwest Research Institute
Chem-Nuclear System, Inc.
AMEFCO Co.
TRW Systems
Protective Packaging (Teledyne)

Industrial Resources, Inc.

Wehran Engineering Corp.
Asphalt
Epoxy Composites
Sulfur

UF
UF
Polybutadiene
Waste Acid + Iron
                                                                                 Radwastes
                                                            Laboratory
                                                            Laboratory
Laboratory
7

Field?

Laboratory?
                                          Nuclear Wastes?
                                        None Reported
                                        None
                                        None

                                        None
                                        None
                                        None
                                        Unknown

                                        Lab

                                        None
       Notes:  1.  This is a generic listing for all wastes
               2.  The list is a partial listing
       Source:  (7,8)

-------
3.0  DISPOSAL OF NONRECOVERY FGD WASTES

3.1  Disposal Options

     A number of methods are potentially available for the disposal of
FGD wastes either on land or in the ocean.  Applicability of disposal
options for FGD wastes can be broadly categorized on the basis of the
nature of the wastes and the type of disposal.

     Table 6 lists potential disposal options for the various types of
wastes.  In this table sulfur is included as a potential waste product;
however, it is more likely that sulfur as a final product from recovery
FGD systems will be produced for utilization.  More importantly, recovery
FGD processes are likely to require prescrubber systems to remove particu-
lates, chlorides, and other flue gas constituents which might contaminate
absorbent liquors.  Prescrubber blowdown from these systems will result
in wastes analogous to the wastes from nonrecovery FGD systems (although
in smaller quantities).  Hence, in the future if recovery processes are
used, it will thus reduce, not eliminate, FGD wastes.

     At present, all FGD wastes generated are disposed of on land.  To
provide a perspective on the current state of FGD waste disposal, Table 7
presents the breakdown on current disposal practices for operational FGD
systems on utility boilers as of November 1978.

     In addition to the above commercially operating units, a number of
FGD systems and associated disposal systems are in operation for testing,
development and/or data gathering purposes.  A list of such current field
testing programs on FGD wastes and associated data on the systems involved
is presented in Table 8.

3.2  Regulatory Considerations

     The disposal of FGD wastes is subject to regulations at both Federal
and state levels.  State regulations governing waste disposal on land can
be more stringent than corresponding Federal regulations.   At present,
FGD wastes are disposed of exclusively on land.   Ocean disposal may be a
technically feasible alternative.   In the future, ocean disposal may be
carried out to a limited extent in regions where there are no mines
available and disposal sites for land impoundments are scarce.

     Disposal on Land

     There are four major impact issues concerning land disposal:

     c  Waste stability/consolidation;

     9  Groundwater contamination;

     •  Surface water contamination; and

     •  Fugitive emissions.
                                   576

-------
                                                                             Table  6

                                                         FGC Waste Types Versus  Potential  Disposal Options
                                   Basis:   All potential methods for disposal of
                                           wastes from particulate control and flue
                                           gas desulfurization methods  are listed.
                                                 LAND DISPOSAL
                                                                                                               OCEAN DISPOSAL
NO.
Ponding
With
Water
Impound-
ment W/0
Water
Cover
Surface Mine Disnosal
Land
Fill
Spoil
Pit Bank
Under-
ground
Mine

Conven-
tional
Shallow
Concen-
trated

Dispersed

Conven-
tional
Deep
Concen-
trated

Dispersed
       1.  UNSTABILIZED
             Sulfite Rich and
             Mixed Sulfite/Sulfate

               By Itself
               With Ash
               With Soil
Ui             With Tailings

             Sulfate Rich

               By Itself
               With Ash
               With Soil
               With Tailings

               Sulfur

       2.  STABILIZED
             Sulfite-Rich and
             Mixed Sulfite/Sulfate
               Soil Like
               Concrete Like

             Sulfate Rich & Gypsum

               Soil Like
               Concrete Like
J
J
NA
NA
7
./
NA
NA
./
•1
J
NA
J
J
NA
7
/
/
NA
7
/
J
NA
7
j
j
NA
NA
•1
7
NA
NA
7
/
NA
NA
J
J
NA
NA
•1
7
NA
NA
7
7
7
NA
7
7
NA
7
7
7
NA
J
1
7
NA
7
7
7
NA
NA
7
7
NA
NA
7
7
NA
NA
                                NA
                                NA
                               7
                               NA
                               •J
                                        NA
J
J
•J
7
                   7
                   NA
7
NA
                    7
                    NA
J
NA
                     NA
                     7
NA
7
                                                                                         NA
7
7
                                                                                                   NA
                               7
                               7
7
7
                                                                                                             NA
                                                                                                                           NA
                                                                                                                           NA
                                                                                    7
                                                                                    7
                                                                                    NA
                                                                                    NA

                                                                                    NA
                                            /
                                            7
                                                                                                 NA
                                                                                                 NA
                                                                                        7
                                                                                        7
                                                                                        NA
                                                                                        NA

                                                                                        NA
                                                /
                                                7
                                                                                                    7
                                                                                                    7
                                                                                                    NA
                                                                                                    NA
                                                                                          7
                                                                                          NA
                                                                                          NA

                                                                                          NA
7
/
        Notes:     / =  Applicable
                  NA =  Note  Applicable
         Source:   Arthur D.  Little,  Inc.

-------
                                                                      Table 7





                          Summary  of  Disposal  Practices  for  Operational  FGC Systems on Utility Boilers as of November  1978




                                                                                 Number of Plant/Plant Capacity
Waste Form System Type
FGD Waste Only Lime-Based
Limestone-Based
Total
^j Codisposal Lime-Based
00 Limestone-Based
Wet Particulate Scrubbing
Total
Stabilized FGD Waste Lime-Based
Limestone-Based
Wet Particulate Scrubbing
Total
TOTALS
	 __ — _ 	
Wet Pond
Dry Fill
—
0/0
2/865
1/1585
5/1685
8/4135
0/0
8/4135
Lined
—
0/0
2/2040
1/50
3/2090
0/0
3/2090
Unlined
1/200
1/200
2/960
4/2460
4/1195
10/4615
0/0
11/4815
Total
0/0
1/200
1/200
4/1825
7/6085
10/2930
21/10840
0/0
0/0
0/0
0/0
22/11040
Dry Fill
o7o~
1/65
1/180
2/245
3/1720
2/730
1/165
6/2615
8/2860
•p . .^
Wet Pond
Lined
1/225
1/225
1/165
1/165
oTo"
2/390
Unlined
1/140
1/140
3/850
2/950
5/1800
1/1650
1/1650
7/3590
Total
0/0
2/365
2/365
4/915
3/1130
1/165
4/3370
2/730
1/165
7/4265
17/6840
TOTALS
0/0.
3/565
3/565
8/2740
10/7215
11/3095
29/13050
4/3370
2/730
1/165
7/4265
39/17880
Source:  Arthur D. Little, Inc.

-------
Ln
SUMMARY OF CURRENT FIELD TESTING PROGRAMS FOR FGC WASTE DISPOSAL
Basis: Status as of November 1978

Location Utility (Plant)* Sponsor(s)a
LAND DISPOSAL:
Mlnnfcota Power (M.R. Young) EPA
Gulf Power (Scholz) EPRI
Gulf Power (Scholz) EPA/EPRI
Colunbus & S. Ohio EPRI
(Conesville)
Louisville Gas & Electric EPA
(Paddy's Run)
Louisville Gas & Electric EPA
(Cane Run)
TVA (Shawnee) EPA/TVA


OCEAN DISPOSAL:

Duquesne (Elrama/ DOE/EPA/EPRI/
Phillips) NYSERDA/PASNY
	 EPA

*ADL - Arthur D. Little
CE - Combustion Engineering
CEA - Combustion Equipment Associates
CIC - Chiyoda International
DOE - Department of Energy
EPA - U.S. Environmental Protection Agency
EPRI - Electric Power Research Institute
IUCS - IU Conversion

Contractor^;?

UND/ADL
CIC/Radlan
CEA/ADL
MBA/Battelle
CE/LGE/UL
Bechtel
TVA/Bechtel




SUNY/ IUCS

NEA/ADL

MBA
NEA
NYSERDA -
PASNY
SUNY
TVA
UL
USD
Scrubber System
Mode Type

S02 Only Alkaline Ash
S02 Only Limestone
S02 Only Dual Alkali
S02 Only Lime (Thiosorbic)
S02 Only Lime (Carbide)
SO2 Only Dual Alkali
SO, & Lime & Limestone
S02 + Ash
S02 Only Limestone (Forced
Oxidation)


S02 + Ash Lime (Thiosorbic)

S02 S Many
S02 + Ash
Michael Baker Associates
New England Aquarium
New York State Energy Research
Power Authority of the State of
State University of New York
Tennesse Valley Authority
Univeristy of Louisville
University of North Dakota

Type

Sulfate-Rich
Gypsum
Sulfite-Rich
Sulfite-Rich
Sulfite-Rich
Sulfite-Rich
Sulfite-Rich
Gypsum



Sulfite-Rich

Many



& Development
New York




Waste Characteristics
Form

Filter Cake (Unstabilized)
Thickened Slurry (Unstabilized)
Filter Cake (Stabilized & Unstabilized)
Filter Cake (Stabilized)
Filter Cake (Stabilized & Unstabilized)
Filter Cake (Stabilized)
(Filter Cake )
{Centrifuge Cake [.(Stabilized f. Unstabilized)
(Thickened Slurry)
Filter Cake



Filter Cake (Stabilized)

Many



Authority






Disposal Mode teat Area
Surface Mine Section of Mine
Stacking -v-l-Acre Area
Dry Impoundment 1-Acre Pit
Dry Impoundment 50-Acre Site
Dry Impoundment Small Pits/Ponds
Dry Impoundment 7
Wet & Dry Impoundments 6 Pits (<.l Acre)
Dry Impoundment 4 Pits/Area



Reef Construction 10.2 Acre

Concentrated Dump 1/2-Acre Pond











Program
Status
Underway
Underway
Planning
Planning
Underway
Planning
Underway
Underway



Underway

Underway











-------
     These are essentially regulated under the federal legislative frame-
work listed in Table 9.  All these legislative acts impose an element of
constraint on FGD waste disposal.  However, the Resource Conservation and
Recovery Act is the major federal environmental legislation regulating
disposal in mines, landfills and impoundments.  According to proposed
regulations under RCRA, Section 3001 defines criteria for determining
whether wastes are hazardous or not.  The most pertinent of these for
FGD wastes are the toxicity-related tests.  If a waste fails these tests
the disposal of that waste would be regulated under Section 3004 of RCRA.'
If an FGD waste fails these tests, its disposal would be regulated as a
special case under Section 3004.  If a waste passes these tests, the waste
would be considered nonhazardous.  Proposed regulations include guidelines
under Section 4004 for nonhazardous waste disposal.  Further definition
of regulations including design standards and criteria are expected.

     Disposal in the Ocean

     Regulation of dispersed ocean dumping of stabilized and unstabilized
FGD waste falls under the Marine Protection Research and Sanctuaries Act
and is administered by the Environmental Protection Agency.  If stabilized,
brick-like FGD waste is used to create artificial fishing reefs with EPA
concurrence, the activities would not be subject to ocean disposal criteria.


3.3  Land and Ocean Disposal Methods

     Land Disposal

     The principal methods of land disposal are:

     •  Wet ponding;

     •  Dry impoundment; and
     •  Mine disposal.

     Wet Ponding:  This method is at present more widely used than any
other.  Ponding can be employed for a wide variety of FGD wastes including
unstabilized materials; however, ponding has been employed with the Dravo
stabilization process.  Ponds can be designed based,on diking or incision
and can even be engineered on slopes.  But the construction of dikes or
other means of containment for ponds is usually expensive.  In the future,
particularly if stabilization of FGD wastes is widely practiced, ponding
will probably be limited to those sites that can be converted to a pond
with minimal construction of dams or dikes.  A special case of wet ponding
is gypsum stacking now under evaluation.  In this case, if the operation
were analogous to that for  phos-gypsum, gypsum slurry (typically from forced
oxidation systems) would be piped to a pond and allowed to settle and the
supernate recycled.  Periodically the gypsum would be dredged and stacked
around the embankments, thus building up the entrainment.
                                   580

-------
                                 Table 9

            Major Regulatory Framework For FGD Waste Disposal
Impact Issue

Groundwater
  Contamination
Surface Water
  Contamination
Physical
  Stability
Fugitive Air
  Emissions
         Legislation

•  Resource Conservation
     and Recovery Act of
     1976

•  Safe Drinking Water
     Act of 1974
•  Federal Water Pollution
     Control Act Amendments
     of 1972

•  Marine Protection Research
     and Sanctuaries Act
•  Surface Mining Control
     and Reclamation Act of
     1977
                   •  Dam Safety Act of 1972
                      Federal Coal Mine Health
                        and Safety Act of 1969
                      Occupational Safety and
                        Health Act of 1970
•  Clean Air Act of 1970
     and its Amendments of
     1977

•  Federal Coal Mine Health
     and Safety Act of 1969
                   •  Occupational Safety and
                        Health Act of 1970
         Administrator
•  Environmental Protection
     Agency, Office of
     Solid Waste

•  Environmental Protection
     Agency, Office of
     Water Supply

a  Environmental Protection
     Agency, Office of
     Water Programs
0  Environmental Protection
     Agency, Office of
     Marine Protection

•  Office of Surface
     Mining Reclamation
     and Enforcement,
     Department of Interior

o  Army Corps of Engineers,
     Department of Defense

•  Mining Enforcement
     Safety Administration,
     Bureau of Mines,
     Department of Interior

o  Occupational Safety
     Health Administration,
     Department of Labor

•  Environmental Protection
     Agency, Office of Air
     Programs

o  Mining Enforcement
     Safety Administration,
     Department of Labor

e  Occupational Safety and
     Health Administration,
     Department of Labor
                                   581

-------
     Leaching from wet ponds is likely to be an important environmental
issue that must be addressed in pond design and operation.  Recent R&D
efforts on wet ponding have centered on:

     •  Most effective means of containing pollutants within the
        disposal area; i.e., study of potential liner material.

     •  Better definition of leaching mechanism from lined and
        unlined ponds.

     Among the more important studies on liner materials relating to the
disposal of FGD wastes are:

     a) The U.S. Army Corps of Engineers Waterways Experiment Station
        (WES) is conducting a program to:  (1) determine the compati-
        bility of 18 liner materials with flue gas cleaning (FGC)
        wastes and associated liquors and leachates; (2) estimate the
        length of life for the liners; and (3) assess the economics
        involved with purchase and placement (including disposal area
        construction) of various liner materials.   The liners that
        WES is testing include:

          i)  Admixture types (cement, lime, fly ash)

         ii)  Prefabricated liner membranes (polymer, neoprene coated, etc.)

        iii)  Spray-on types (polyvinyl acetate, latex, asphalt, cement, etc.)

        Results of this investigation are expected to be available in 1979.

     b) In 1978 EPRI initiated a program to evaluate leachate control
        and monitoring systems for solid waste disposal facilities.
        The objective is to evaluate liner materials for utility solid
        wastes.  This 36-month program which will  be underway in 1979
        is expected to yield substantial technical data on a number of
        liner materials.

     Proposed RCRA Guidelines (9) provide some indication of potentially
appropriate impoundment design.  The extent of leaching of pollutants from
disposal ponds is dependent on several factors:  the hydrostatic head in
the pond, which forces percolation through the pond bottom; the nature of
the waste—primarily its permeability and the solubility of contaminants
it contains; and the characteristics of the soil around the pond.  At
present, monitoring wells exist in some of the FGD waste disposal ponds,
particularly the larger, more recent ones like the disposal pond for the
Bruce Mansfield plant.  However, insufficient time has passed for adequate
data to be available.  Efforts are continuing in this field and addi-
tional insights on field-scale leaching is likely to be available in the
future.

     Dry Impoundment Methods;  These may include any of the following
variations:
                                   582

-------
     •  Interim ponding followed by dewatering and sometimes
        excavation and landfilling;

     •  Mechanical dewatering and landfilling of FGD wastes;

     •  Blending with fly ash and landfilling of FGD wastes;  and

     •  Stabilization through the use of additives (non-proprietary
        or otherwise).

     Typically, for dry impoundment type of disposal, the wastes are
thickened and dewatered to a high solids content level and  blended with
fly ash and lime, thus forming a material with cementitious properties.
This material is transported to the disposal site where it  is spread  on
the ground in  1 to 3 foot lifts and compacted by wide track  dozers,
heavy rollers or other equipment.  Layering proceeds in 1 to  3 foot
lifts in segments of the site.  The ultimate height of a disposal fill
is site-specific but may be 30 feet to as high as 80 feet or  more.  A
properly designed and operated dry impoundment system can potentially
enhance the