&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 value of the disposal site after termination or at least
permit post operational use.
Important engineering considerations in dry disposal of FGD wastes
in landfills include:
a) Physical instability is a potential problem for all FGD wastes,
including stabilized materials. Geometric factors such as
height and slope angle in a landfill are interrelated;
stability depends on the combination of fill height, slope
angle, wastes density, degree of saturation, effective co-
hesion, effective angle of shearing resistance, and behavior
during shearing (dilatant vs. densifying). For a given
material, safe fill height decreases with increasing slope
angle. For proper design, data is required in which maximum
safe fill height is related to slope angle and soil shearing
behavior. Such relatiors are not fully developed yet for
FGD wastes because adequate data from proper tests (triaxial
compression tests on consolidated samples with measurement of
porewater pressures) have not been available.
Instability problems may be ameliorated by compaction since
compaction may produce several changes: voids may be
eliminated (between chunks, not between individual particles);
waste density may be increased; particles moved closer together
may be bonded more effectively in stabilization reactions; and
effective stress and residual total stress levels may be increased.
If the wastes resemble sandy soils, an increase in density during
compaction may create an important increase in angle of shearing
resistance.
b) Climate could affect the stability of a wastes deposit in
several ways. The total amount of rainfall and the intensity
of rainfall both affect surface erosion, but these factors
583
-------
influence the rate of erosion; i.e., erodible materials will
erode if wind and rain act on them but increase in amount/duration
and intensity of wind and rain will increase the rate of surface
erosion. Of course, as mentioned above, infiltration of water
into wastes deposits can lead to saturation, decrease in apparent
strength, and, possibly, to failure. This would be most likely
to occur with wastes of high permeability in an unsaturated
condition, but it could also occur in any FGD waste deposit.
Freeze-thaw cycles may also impact waste strength and stability.
Stabilization could be retarded, disrupted or destroyed by an
episode of freezing soon after mixing and placement. Freezing
could produce cracks in the near-surface layers of a wastes
deposit (frost polygon behavior). Cracks in the wastes deposit
could yield greater mass permeability and infiltration rate
even though waste blocks between cracks had been compressed
and even dewatered.
c) Post Closure Land Use: Often it is difficult to specify the
ultimate use of the land on which FGC waste disposal is planned
or practiced. However, it is important to recognize engineering
constraints on post-closure land use. Post-closure land use
would tend to be limited by nature of the loads created or by
the sensitive nature of the structures or facilities built on
the wastes fill. For example, placement of some sort of fill
in a uniform layer over the entire waste deposit should be
feasible, but imposition of concentrated loads (e.g., footings
in a building) may not be feasible.
d) Run-Off: To limit the impact of water run-off from waste
deposits, grading and drainage are required as in any other
type of construction activity. Consideration should also
be given to run-off control following completion of the
fill, including proper overall site grading and soil cover.
Surface run-off caught in temporary retention basins could be
recycled or discharged after appropriate treatment.
Mine Disposal; A disposal method that is receiving increased
attention is mine disposal. It appears that surface coal mines and
underground room and pillar mines for coal, limestone or lead/zinc ores
offer particular potential (10). Of the four categories of mines noted
above, coal mines, and in particular surface area coal mines, are the
most likely candidates for waste disposal. Coal mines offer the greatest
capacity for disposal, and they are frequently tied directly to power
plants. In fact, many new coal-fired power plants are "mine-mouth"
(located adjacent to the mine within a few miles) and the mine provides
a dedicated coal supply. Since the quantity (volume) of FGC wastes pro-
duced is considerably less than the amount of coal burned, such mines
usually would have the capacity for disposal throughout the life of the
power plant.
5S4
-------
In general, inactive surface mines are considerably less promising
than active mines for FGD waste disposal. Unreclaimed surface mines can
be used for disposal of wastes between remaining spoil banks, and these
may offer suitable sites for disposal. However, because of recent surface
mine reclamation legislation, the number of sites and total capacity for
wastes available in the future will be limited.
In active surface mines, there are basically three options for the
placement of FGD wastes:
• In the working pit, following coal extraction and prior to
return of overburden;
• In the spoil banks, often return of overburden but prior
to reclamation; and
• Mixed with or sandwiched between layers of overburden.
In any disposal operation in an active mine, though, the general
overriding consideration is that disposal should cause minimal disruption
of mining or reclamation activities. This provides a number of con-
straints on the disposal system.
a) First, the amount (volume) of sludge disposed of in any surface
mine should not greatly exceed the amount of coal removed. The
objectives of strip mine reclamation include returning the
mined terrain to topographic configurations similar to original
terrains. Also, returning significantly more, waste to a mine
than the coal extracted could slow down the mining and reclama-
tion activities. In most cases, this does not really represent
a constraint, since the wastes returned to the mine will usually
be only those resulting from the coal removed. The amount of
FGD waste generated from the combustion of coal will be consider-
ably less than the quantity (weight or volume) of coal removed.
Depending upon the type of coal, FGC system, and emission stan-
dards to be met, the volume of total waste (ash plus calcium-
sulfur solids) will range from less than 10% of the coal burned
to slightly over 50%. With Western coals, which are relatively
low in sulfur, less than 20% is the rule. With higher sulfur
Eastern and Interior coals, the total amount of wastes will
typically run 30% or more.
b) Second, the physical condition (consistency) of the wastes must
be amenable to ease of handling, transport, and placement using
earth-moving equipment with minimal potential impact on the
mining operations. For pit-bottom disposal this means that the
wastes at the time of placement or immediately thereafter should
have as a minimum the consistency of a soil-like material with
little or no liquefaction potential. A slurry-like material or
a waste with a tendency to flow either when placed or when over-
burden is dumped on top would present significant operational
problems or unacceptably high costs for containment measures. A
585
-------
little more leeway exists for disposal in V-notches (between
spoil banks); however, here again, soil-like materials or
relatively cohesive materials that are relatively easily handled
and transported will result in the least cost and minimal dis-
ruption of reclamation activities. At the least, the wastes
must be well filtered (55% solids or higher), and may need to
be admixed with dry fly ash, or stabilized.
c) Finally, minimal use should be made of existing mine equipment
for transport and placement of the wastes at the mine. Dedicated
equipment is preferred and, in most cases, mandatory. In almost
all scenarios, waste is most easily placed by truck dumping. The
use of coal trucks for this purpose could lead to unacceptable
delays in coal mining operations due to the additional time for
waste loading and discharging (and possible cleaning operations)
Furthermore, most large mines use large bottom-dump trucks for
coal haulage. These are designed to carry as much as 150 tons
of coal and are usually constructed of aluminum. Not only can
the waste corrode the aluminum, but the bottom dumping of wastes
would be Impractical. These trucks are not designed for ease of
maneuvering, and operating them (or any other equipment) on a
freshly dumped layer of waste would be difficult at best. The
type of truck used for transport and placement will greatly affect
the quantity of waste that can easily be disposed of.
At present there are only two commercial operations involving mine
disposal of FGD wastes in surface coal mines—one at Texas Utilities'
Martin Lake Station and the other at Square Butte's Milton R. Young
Station (North Dakota). Both stations fire lignite and involve returning
combined fly ash and calcium-sulfur solids from SC>2 removal to the re-
spective mines. The operation at the Baukol-Noonan mine which supplies
coal to the Milton R. Young Power Station is currently an EPA mine dis-
posal demonstration project. At this mine both pit-bottom and spoil bank
disposal are being employed. Mine disposal of FGD wastes can potentially
be employed for subsidence control. Acid mine drainage neutralization,
reclamation of mine areas or as soil amendments for tailings disposal
from mining operation. Thus, there could be subsidiary benefits from
this type of disposal.
Ocean Disposal
Ocean disposal of FGD wastes is not practiced today. However, if
it could be practiced under environmentally acceptable conditions, it
could represent an important option, particularly in Federal Regions 1
and 2 (the Northeast) where land for disposal is limited. For this and
other reasons, EPA has been studying the disposal of FGD wastes in the
ocean (10). From a technical viewpoint, ocean disposal may be considered
in the shallow ocean (i.e. , on the continental shelf) or deep ocean (off
shelf). Each of these has a different ecosystem with a different set of
potential impacts. At present a number of viable techniques exist for
transporting FGC wastes to offshore disposal sites. These include:
586
-------
• Self-propelled hopper ship with throttled discharged
disposal or with a sudden dump capability;
• Tow^barge transportation and controlled dispersal over
a great expanse of water or a sudden total bottom dump; and
• Submarine pipeline transportation and dispersal at a
preselected offshore disposal site.
All of these approaches are technically feasible systems which have
been utilized on a full scale for disposal of other types of wastes.
Selection among them depends upon both the characteristics of the FGC
wastes to be dumped and any environmental conditions and/or constraints
that might exist.
At present, 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.
The dumping would be required to be limited to a'n EPA-prescribed dumpsite
under the following conditions:
• Trace contaminant (e.g., Hg, Cd) content of the dumped
materials would be no higher than 50% above that of
background sediments at the dumpsite;
• Concentrations of the dumped material in the water column
four hours after release would not exceed 1% of the 96-hour
LC5Q of the material to local sensitive species; and
• No feasible alternatives to ocean disposal are available.
Stabilized, brick-like FGD waste may potentially be used to create
artificial fishing reefs with EPA concurrence. Artificial fishing reefs
are not subject to ocean disposal criteria but FGD waste disposal may be
a special case. While ocean disposal of FGD sludges is an option that
is perhaps available to throwaway system users with economic access to the
ocean, new ocean disposal initiatives are now discouraged by the regulatory
agencies.
At present, the major studies under EPA sponsorship or participation
relating to the ocean disposal of FGD wastes are:
a) The Arthur D. Little study (10) for EPA on the technical, economic
and environmental feasibility of the ocean disposal of stabilized
and unstabilized FGD wastes. This study includes both laboratory
and small-scale field testing relating to impact issues.
b) The State University of New York (11) with funding provided from
Power Authority of the State of New York, New York ERDA, DOE,
EPRI, and EPA, is studying the use of stabilized brick-like
FGD wastes (using IU Conversion Systems process) to create
artificial reefs for marine habitats. This study is expected
to continue for 2 to 5 years.
587
-------
3.4 Disposal Options Versus Potential Impact Issues
On balance, technology exists for environmentally sound disposal
of FGD wastes. The environmental impact issues requiring consideration
in handling and disposal of FGD wastes are:
• Air-related. These include fugitive particulate emissions,
emissions of S0~ and RyS and emissions of trace metal compounds;
• Water-related. These include groundwater contamination, surface
water point source discharges and run-off;
• Land-related. These include physical stability (subsidence,
liquefaction or other structural failure, erosion, etc.) and
land use considerations; and
• Biological impacts both in the site and adjacent areas and
consequential effects.
Potential impact issues are highly site- and system-specific. With
that understanding, Table 10 illustrates the major types of impact issues
associated with various disposal options. As the matrix illustrates, the
range of waste types and possible disposal conditions is sufficiently
broad to eliminate the potential for "generally significant" issues to be
associated with any of the disposal options. Further, site-specific
application of appropriate control technology can be employed to mitigate
adverse impacts. In other words, issues of potential significance in
FGD waste disposal can best be defined in terms of specific waste types,
disposal practices, and disposal environments. The significance of many
potential impact issues may be better quantified by additional field-scale
operating experience (and environmental monitoring) with FGD waste disposal.
As indicated in the matrix, this is particularly desirable for defining
potential issues in the categories of water quality and biological impacts.
3.5 Assessment of Present Control Technology
It is expected that much of the difference between potential and
actual impacts for the FGD waste disposal options discussed above will be
determined by the degree to which presently available control technology
becomes incorporated as "good design" and "good practice" in typical dis-
posal operations. Good design and practice could also minimize the poten-
tial for adverse impact from abnormal events. Important considerations
in the application of present control technology are briefly discussed
below.
a) Site Selection: Site selection may or may not be considered
control technology. However, there is no question that proper
site selection could by itself ameliorate or eliminate most
of the potential disposal impacts discussed above. Specifically,
the following mitigative combinations of site characteristics
and impact issue categories are considered applicable:
588
-------
Table 10
Disposal Options Vs. Potential Environmental
Impact Issues for FGC Wastes
Potential Environmental Impact Issues
Ln
00
VO
Disposal Options
Wet Ponding
Dry Disposal
Surface Mine Disposal
Underground Mine Disposal
Shallow Ocean Dumping
Deep Ocean Dumping
Land Use
2
2
2
3
3
3
Surface Water
Quality
2*
2*
2*
3
2*
2*
Groundwater
Quality
2*
2*
2*
2*
3
3
Air
Quality
3
2*
2*
3
3
3
Biological
Impact
2*
2*
2*
2*
2*
2*
Significance highly uncertain due to data gaps.
Key: 1 = Issue of potential general significance for all FGC wastes at all disposal sites.
2 = Issue of potential significance for specific types of FGC wastes and/or specific
disposal sites.
3 = Issue of minor or no potential significance.
-------
Potential Impact Issue Mitigative Site Characteristics
Land Use Proper topography, geology and
hydrology; absence of nearby
conflicting land uses.
Water Quality As above for land use, plus
absence of nearby sensitive
receiving waters (surface or
aquifers). For example, a small
stream or very pure aquifer may
impose greater constraints than
a relatively large stream or
impure aquifer.
Air Quality Absence of "non-attainment area
and Class I Prevention of Signif-
icant Deterioration designations
for total suspended particulates.
Usually this is even more important
for the Power Plant Siting.
Biological Effects Absence of sensitive biological
resources.
b) Dewatering; As discussed earlier, dewatering of FGD waste prior
to processing or land disposal can result in major improvements
in physical stability and. reduce water quality impacts regardless
of which disposal approach is employed, including those discussed
below.
c) Stabilization: Stabilization appears to be highly relevent to
the mitigation of land use issues, including the potential for
abnormal events (i.e., disposal area liquefaction or other cata-
strophic failure modes), and the suitability of disposal sites
for a broader range of post closure uses requiring increased
bearing strength. Stabilization techniques resulting in decreased
waste permeability can be considered mitigative of potential water
quality impacts due to leachate migration. This factor should be
considered in balance with the requirements for disposal area run-
off control on a site-specific basis.
Stabilization reduces permeability and hence reduces rate of
contaminant transfer from a disposal site. However, long-term
cumulative contaminant migration could be important. In particu-
lar, it is not clear that reductions in long-term trace contamin-
ant availability would take place when fly ash is used as a
stabilization additive to a waste initially containing no ash.
However, migration of contaminants to the environment at a
slower rate is more desirable.
590
-------
Cement!tious stabilization processes, because of increased
particle size, may also be considered mitigative of the
potential for post-disposal fugitive particulate emissions
from dry FGC waste disposal operations, and may minimize or
prevent gaseous emissions by reducing exposure of waste to
water and biological organisms.
In ocean disposal, cementitious stabilization may remove
liabilities of FGC wastes as benthic substrates and as sources
of sulfite-related depletion of dissolved oxygen. However,
questions of sulfite and trace contaminant availability, among
others, preclude definitive judgment on this issue at this time.
d) Forced Oxidation; The intentional production of sulfate-rich,
rather than sulfite-rich FGC wastes, is presently a subject of
considerable interest. In ocean disposal, the sulfate-rich
products of forced oxidation would have the obvious advantage
of mitigating the potential for sulfite-related depletion of
dissolved oxygen. This advantage would be shared in land dis-
posal operations (especially wet impoundments), but its relative
importance is less clear. A dominant question concerning the
mitigative potential of forced oxidation for land disposal is
whether or not the process results in increased or decreased
physical stability. Based on experience with soils, gypsum
FGD wastes comprised of relatively uniform, sand-sized particles
may exhibit considerable failure potential in the absence of:
1) effective compaction and dewatering, and/or 2) co-disposal
with materials of varying particle size (i.e., fly ash). How-
ever, if FGD gypsum is analogous to phos-gypsum, recrystalliza-
tion mechanisms occurring in the disposal pile may improve
stability.
e) Co-disposal of Wastes and Creation of Waste/Soil Mixtures:
Although the term co-disposal is often used in reference to the
creation of disposal mixtures of two waste streams (e.g., FGD
wastes and coal ash), it is used here to imply a broader range of
potential opportunities. Specifically, for land disposal of FGD
wastes, "co-disposal" might also include the application of tech-
nologies for the creation of soil/waste mixtures. If soils with
the proper characteristics are available, the creation of soil/
waste mixtures may be an alternative to the addition of fly ash
where only limited increases in physical stability are desired
in a disposal operation, or where trace contaminant availability
needs to be reduced to facilitate revegetation or decrease water
quality impacts. Traditional co-disposal involving fly ash plus
FGD waste appears to have substantial advantages over independent
disposal in terms of improved physical stability and (potentially)
decreased permeability. This might be especially relevant to
sulfate-rich FGC wastes of uniform particle size. However, in
some situations the extent to which the ash serves as a reservoir
of certain trace contaminants could prove a liability from the
standpoint of potential water quality degradation.
591
-------
3.6 Future Research Needs
A number of programs have been undertaken (and are in progress) by
the Environmental Protection Agency (EPA), the Department of Energy (DOE),
the Electric Power Research Institute (EPRI), and others. These efforts
have provided much of the baseline information for environmental assess-
ment. Provided these programs continue, additional data and insight
permitting better environmental assessment will be possible.
Research needs pertinent to environmental assessment of FGC disposal
are:
a) Acquisition of field data on the actual impacts of full-scale
disposal operations under varying environmental conditions.
Field-scale monitoring of large disposal operations: over a
period of several years is warranted. EPRl's proposed program
at Conesville Plant of Columbus and Southern Ohio Power is one
such example. EPA is also planning an extensive 2-year study
on characterization and environmental monitoring of full-scale
utility disposal sites.
b) A corrollary of (a) above would be the development of correla-
tions and tools of extrapolation to relate existing lab/pilot
scale data on physical stability and water quality impacts to
full-scale field data.
c) Integrated study and evaluation of the environmental trade-offs
in co-disposal of various FGD wastes and various coal combus-
tion ashes. (It appears that this type of initiative could
emphasize laboratory work with limited pilot and full-scale
field verification.)
d) Development of basic data (laboratory and field-scale) on the
biological impact potential of principal land-based FGC waste
disposal options, especially data relating to water-related
impacts of major soluble species and trace contaminants.
Typical questions are:
• What are the biological and health effects of mixtures
of trace metals (in the form found in liquors), such as
zinc, copper, lead, mercury, cadmium or nickel in com-
bination with selenium in particular, but also in other
combinations?
• What is the uptake of potentially toxic materials by
vegetation and wildlife associated with disposal areas?
• What are the levels of ambient concentration of waste-related
potentially toxic materials in vegetation and surface water
that may produce chronic health problems for wildlife?
EPA is presently supporting biological testing work on FGC
wastes at Oak Ridge National Laboratory.
592
-------
e) Development of basic (laboratory and field) data on the
potential for fugitive particulate emissions from areas
previously used for the dry disposal of FGC wastes.
f) Socio-economic impacts of FGC waste disposal on land need
to be better defined.
In the future, FGD waste generation will not be limited to those by
utility systems. Coal utilization in industrial boilers (25 MW or larger)
is also likely to grow substantially. FGD wastes from such industrial
boilers (while analogous in composition to those from utility boilers)
present additional waste management issues due to differences in dis-
tribution of generation facilities, in quantity of FGD wastes generated
at each facility and other factors. These issues also require further
evaluations and study.
593
-------
REFERENCES
1. Annual Environmental Analysis Report, prepared by Mitre Corporation,
Consad Research, Control Data Corporation, and International Research
and Technology; Report to ERDA, under Contract EE-01-77-0135,
September 1977-
2. "Steam Electric Plant Air and Water Quality Control Data - For the
Year Ended December 31, 1975 - Summary Report," Federal Energy
Regulatory Commission, Washington, D.C., January 1979.
3. National Electric Reliability Council 7th Annual Review of Overall
Reliability and Adequacy of the North American Bulk Power Systems,
NECR, Research Park, Terhune Road, Princeton, New Jersey, 1977.
4. 1977-1978 Electrical World Directory of Electrical Utilities,
McGraw-Hill Book Company, New York, New York, 1978.
5. "EPA Utility FGD Survey - April-May 1978," by B. Laseke, M. Melia,
M. Smith and W. Fisher, PEDCo Environmental Inc., EPA 600/7-78-D57c,
September 1978.
6. "Evaluation of Opportunities to Accelerate the Commercialization of
Dry Sorbent Technology for S02 Control," Lutz, S.J., Cotton, J.E.,
Houser, C.N., TRW under EPA Contract 68-02-2613, Draft Report,
November 1978.
7. "State of the Art of FGD Sludge Fixation," Duvel, W.A., Gallagher,
W.R., Knight, R.G., Kolarz, C.R., and McLaren, R,J., Michael Baker,
Inc., EPRI RP-671, Electric Power Research Institute, Palo Alto,
California, January 1978.
8. "Survey of Solidification/Stabilization Technology for Hazardous
Industrial Wastes," Waterways Experiment Station under Interagency
Agreement, EPA-IAG-D4-0569, Draft Report to EPA-MERL, Cincinnati,
Ohio, February 1979.
9. Hazardous Wastes - Proposed Guidelines & Regulations and Proposal on
Identification & Listing, Federal Register, Monday, December 18, 1978,
Part IV, pages 58946-59028.
10. Lunt, R.R., et al, "An Evaluation of the Disposal of Flue Gas
Desulfurization Wastes in Mines and the Ocean: Initial Assessment,"
Arthur D. Little, Inc., EPA Report EPA-600/7-77-051, May 1977.
11. Duedell, I.L., et al, "Environmental Assessment of Coal Waste Disposal
in the Ocean Waters," Marine Sciences Research Center, State University
of New York and iu Conversion Systems, submitted to the EPA, July 1978.
594
-------
MARKETING ALTERNATIVES FOR FGD BYPRODUCTS
AN UPDATE
By
W. E. O'Brien and W. L. Anders
Emission Control Development Projects
Tennessee Valley Authority
Muscle Shoals, Alabama
ABSTRACT
The latest FGD byproduct marketing programs conducted by TVA for EPA
are "First Annual Update - Potential Abatement Production and Marketing
of Byproduct Sulfur and Sulfuric Acid in the U.S." and "Flue Gas
Desulfurization Byproduct Production and Marketing System Users Manual."
The first is an annual update to 1983 of an extensive computer system
to project FGD byproduct sulfur and sulfuric acid market potential.
Results of this update are presented. Boilers projected to come on-line
between 1978 and the end of 1983 are important candidates for marketing
FGD byproduct sulfuric acid. A small reduction in total FGD byproduct
sulfur costs could result in emergence of some competitive FGD sulfur
production.
The users manual describes programing, procedures, and techniques
required to utilize the computer production and marketing system. The
users manual makes the byproduct production and marketing computer
system available on a commercial basis through TVA under a technology
transfer agreement with EPA. Procedures for access are described.
595
-------
MARKETING ALTERNATIVES FOR FGD BYPRODUCTS
AN UPDATE
INTRODUCTION
For several years the Tennessee Valley Authority (TVA), in conjunc-
tion with the U.S. Environmental Protection Agency (EPA) and others, has
conducted economic studies related to flue gas desulfurization (FGD)
processes. As a part of these studies a computer-based byproduct
marketing system was developed to evaluate the potential of marketing
sulfur byproducts from FGD processes. It compares the economics of
using clean fuel with the economics of several FGD processes, including
limestone scrubbing with waste sludge production, magnesia scrubbing
with sulfuric acid production, and the Wellman-Lord - Allied Chemical
scrubbing process which produces sulfur. The system contains data on
over 900 U.S. power plants and the locally applicable regulatory informa-
tion required to calculate scrubbing costs for each power plant in the
system. In the case of sulfuric acid production, the marketability of
the acid is determined using data on U.S. sulfur and sulfuric acid trans-
portation and the U.S. sulfuric acid manufacturing industry. Provisions
for sulfur marketing are being incorporated in the computer system.
General or specific cost determinations can be made for different alterna-
tive clean fuel levels (ACFL) and different scrubbing methods, including
the possibilities and effects of byproduct marketing.
A scrubbing system that produces a potentially marketable byproduct
such as sulfuric acid is not practical for all power plants because the
incremental cost of recovery can exceed the market value. 'A scrubbing
system that produces waste byproducts is not satisfactory in every case
because of disposal problems, delivered raw material costs, and plant
operating characteristics. Neither will the exclusive use of a clean
fuel solve the problem for all plants. Consideration must, therefore,
be given to engineering, economics, transportation, and marketing factors
which allow in-depth analyses in selecting a suitable strategy for
particular power plants. Since several strategies may be selected that
result in compliance with clean air requirements, information must be
available to make decisions that result in the least cost of compliance
in the long run.
A definitive TVA report on a marketing analysis of byproduct
sulfuric acid economics projected to 1978 was issued by EPA in 1978
(Bucy et al. 1978). Power plant abatement compliance decisions must,
596
-------
however, be made to conform with anticipated conditions several years
in advance. The need for regulatory agency:approval and longer construc-
tion periods combine to extend the effects of decisions into a more
distant, and thus less certain, future. It is essential, therefore, that
a project of this nature use the latest available data projected as far
in the future as can be reasonably substantiated. For this reason an
update of the 1978 projection for byproduct sulfuric acid market potential
was recently projected through 1983 using data available up to September
1978 (O'Brien and Anders, in press). This update,is a continuing refine-
ment of the computer programs of the original byproduct marketing model.
In addition, a manual analysis of byproduct sulfur marketing as an alter-
nate to byproduct sulfuric acid is included.
A users manual (Anders, in press) was also prepared to make the
byproduct marketing computer system available to the power industry on
a commercial basis. This manual provides the information and procedures
necessary to use the system. ;It is primarily intended for the.analyst-
programer but it should also be useful to those requiring detailed informa-
tion about the system who do,not have extensive systems or programing
experience.
The byproduct marketing system is available through TVA under a
technology transfer agreement with EPA. Procedures for releasing the
system are initiated upon receipt of a written request. At the present
time, under the same technology transfer agreement, selected runs of the
system based on user-supplied data can be made by: TVA. All inquiries
concerning the byproduct marketing system and this manual should be
directed to Emission Control Development Projects, Tennessee Valley
Authority, Muscle Shoals, Alabama 35660, telephone No. (205) 383-4631,
extension 2516.
SYSTEM DESCRIPTION
The users manual provides the only description of the complete
byproduct marketing system. The system description presented herein is
a condensed version of the comprehensive description presented in the
users manual. The byproduct marketing system consists of a number of
integrated computer programs, models, and data bases which can be used
to make cost comparisons of FGD strategies designed to meet clean air
regulations. For strategies which produce a salable byproduct, the
marketability of the byproduct is determined and its effect on FGD costs
is included in the cost comparisons. The system can use this data or
user-supplied data to develop situations for comparison of alternate FGD
strategies. For comparisons based on the use of clean fuel without FGD,
an ACFL is used to reflect the cost differential between a complying fuel
and a noncomplying fuel. In the cases of magnesia scrubbing and the
Wellman-Lord - Allied Chemical process, the system determines the incre-
mental cost of sulfuric acid or sulfur production. Incremental production
cost is defined as the production cost per ton of sulfuric acid or sulfur
597
-------
above the cost of either limestone scrubbing or the ACFL value, whichever
is lower. For sulfuric acid the system determines the marketability in
terms of net back from sales to specific locations at which the delivery
and incremental production costs are offset. Provisions for distribution
of sulfur are being incorporated into the system.
The entire system can be used to compare the magnesia-scrubbing
strategy with either the limestone-scrubbing or clean fuel strategies
for any combination of geographic and power plant situations. The
numerous subsystems, programs, and data bases can be used separately or
in various combinations to provide information on a wide range of related
FGD processes and on sulfuric acid manufacture, transportation, and
marketing.
The byproduct marketing system is shown diagramatically in Figure 1.
It can be divided into four subsystems. The supply subsystem consists of
data bases and programs which provide data on power plants, emission
control regulations, raw material costs, and FGD design and cost data.
These can be used to determine scrubbing costs on a boiler-by-boiler
basis for each power plant in the data base. The demand subsystem consists
of programs and data bases on sulfur transportation costs and acid plant
operating costs which are used to determine acid plant avoidable production
costs. Avoidable production cost is the expenditure that could be avoided
by shutting a sulfur-burning acid plant down. This savings is the break-
even price that can be paid for FGD byproduct acid. The transportation
subsystem consists of data bases and programs to provide rail mileages,
tariffs, and rate-basing information for power plants and acid plants from
the other subsystems. It is used to calculate acid transportation costs.
The fourth subsystem consists of a linear program model generator, a
linear program model solution generator, and various optional report
generators. It uses the results of the other three subsystems to select
the least-cost option for each power plant considered for marketing.
The data used in the system have been compiled from a wide range of
U.S. Government, TVA, and published sources. The data include over 3500
boilers representing over 900 power plants and all acid plants and smelters
in the United States. The automated model is limited to the eastern 37
states for which a railroad rate-basing system, necessary to the transpor-
tation subsystem, exists. Excess acid supply from the 11 western states
and Canada is included in the linear programing model as a manually
calculated factor. The calculation of scrubbing costs is based on design
and economic premises developed by TVA and EPA to compare the economics
of scrubbing systems (McGlamery et al. 1975).
The byproduct marketing system is designed on the basis of several
assumptions important to a conceptualization of the evaluation process.
They are listed as follows:
598
-------
POWER PLANT
DATA
I
SCRUBBING COST
GENERATOR
RATE
DATA
I
TRANSPORTATION
COST
GENERATOR
ACID PLANT
DATA
ACID PRODUCTION
COST
GENERATOR
MARKET SIMULATION
LINEAR PROGRAMING
MODEL
EQUILIBRIUM SOLUTION
RESULTS
BYPRODUCT MARKETING SYSTEM
Figure 1. Byproduct marketing system.
599
-------
1. The cost of sulfuric acid produced in the magnesia-scrubbing
system is based on the incremental cost difference between the
magnesia system and the limestone or clean fuel strategy used
for comparison, plus the cost of transportation to the consumer.
2. All byproduct acid is assumed to be sold to sulfur-burning acid
plants which can reduce their costs by buying acid, at a price
determined as described above, instead of manufacturing it
themselves.
3. It is assumed that total acid consumption is unaffected by
byproduct acid production. All byproduct acid is assumed to
replace acid manufactured from sulfur. As a corollary, smelters
are assumed to be producers of necessity. All smelters are
assigned an acid production based on their compliance with
applicable emission regulations and this is included in the
total supply.
4. The marketing model does not consider elements of profit or
maximization of benefits for a particular industry. Each model
solution is an optimum situation in which all acid producers,
transportation networks, and acid plant consumers are integrated
into a system which provides for the greatest byproduct acid
revenue within the restrictions imposed by the system design.
5. The additional cost assigned to ACFL as compared to noncomplying
fuel is also used as a screening technique in the model construc-
tion process. By assigning particular clean fuel costs, the
structure of the model, in terms of power plants included and
comparisons made, can be varied to compare the effects of
different costs of compliance.
FGD PROCESS DESCRIPTIONS
All of the FGD systems used in this evaluation are scrubbing
processes in which the flue gas is contacted with a suspension or solu-
tion of absorbent in water. The SOX in the flue gas reacts with the
absorbent to form sulfur salts. A purge stream is removed and fresh
absorbent added to maintain equilibrium concentrations in the scrubber
system. All of the processes are in use or have been tested in full-
scale operation (Herlihy 1977). The FGD systems are assumed to be
installed downstream from existing air heaters and particulate matter
removal equipment. All FGD equipment is provided, including raw material
handling systems, auxiliary processing equipment to produce and store
byproducts, and waste disposal facilities.
The limestone FGD system consists of a scrubbing system in which
a suspension of finely ground limestone is contacted with the flue gas
to form calcium sulfite and calcium sulfate. The purge stream is
600
-------
pumped to a disposal pond without further treatment. The system is
characterized by large raw material and land requirements and relatively
low energy, capital, and operating costs.
The magnesia FGD system is a similar scrubbing process using
magnesia as the absorbent. The magnesium sulfite formed is removed
from the purge stream, dried, and calcined to regenerate magnesia and
sulfur dioxide. The magnesia is returned to the scrubber system and
the sulfur dioxide is processed to sulfuric acid in an onsite acid
plant. The system is characterized by lower raw material and land
requirements and higher energy, capital, and operating costs, relative
to the limestone system.
The Wellman-Lord - Allied Chemical process uses a sodium sulfite-
bisulfite solution as the absorbent. Sodium sulfate is removed from
the purge stream by selective crystallization. The bisulfite-rich
purge is then thermally treated to regenerate sodium sulfite and sulfur
dioxide. The sulfur dioxide is converted to sulfur by an onsite Allied
Chemical proprietary reduction process. The characteristics of the
system are similar to the magnesia system but raw material, energy,
capital, and operating costs are higher for most situations.
FUNCTION AND USE OF THE USERS MANUAL
As the EPA-TVA byproduct marketing studies progressed, the automated
system o£ programs and data bases that had been developed were expanded
and refined in order to meet increasingly detailed requirements. It
became apparent that, in addition to the direct benefits to the actual
studies, the expanded system presented opportunities for others to
utilize the resources that were being developed. An additional objective
was defined in conjunction with the continuing studies to prepare a
manual to describe how the system is used, and to make the complete
system available for access on a commercial, nationwide, interactive
time-sharing/remote batch data processing network when justified by
potential usage.
The manual is designed for the needs of a programer-analyst; it is
a how-to guidebook providing information and procedures necessary to
use the system. It should also be useful to those requiring information
about the system who do not have extensive systems or programing experi-
ence. It is primarily a users manual, however, and does not provide
all of the concepts and background information necessary for the use
of the system.
MODEL UPDATE
Following is a brief description of the major changes from the
1978 projection which were incorporated in the 1983 projection.
601
-------
Scrubber Cost Generator
Power Plant and Boiler Screens. Past observations have shown that
older, smaller power plants with small boilers are very poor candidates
for FGD. In order to eliminate them from the model a screening method
was adopted. It provided that individual boilers of less than 25-MW
capacity or over 20 years old would automatically be eliminated from
consideration. It also eliminated power plants of less than 100-MW
total capacity. In this manner the model is' run on a more cost-efficient
basis and emphasis can be placed on the more logical candidates.
New Power Plants and Boilers through 1983. Projected boilers
scheduled for completion between 1978 and the end of 1983 were added to
the data base. There were 147 of these boilers identified and 26 were
directly included in the 1983 model. The 121 projected boilers that
were not considered are summarized as follows:
Number of boilers
not considered
13
25
Reason
Total
23
10
49
1^
121
Location could not be identified.
These boilers were located in the 11 western states
where transportation rates were impractical to
automate. They are included in summary form by
manual analysis.
An alternate scrubbing strategy was already
selected.
Fuel data were incomplete.
A clean fuel strategy was already selected.
The total plant size was less than 100 MW.
Updated Regulations and Compliance Plans. The 1978 projection was
based on regulation data available from EPA through June 30, 1976. The
1983 projection used equivalent data through July 15, 1977, the latest
available from EPA. It does not provide for regulation changes which may
result from the revisions proposed in 1978 by EPA. The compliance status
of power plants was also updated. A special report from EPA based on
regulations assumed to be in effect December 31, 1978, was used as a
basis.
602
-------
Cost Escalation. Capital costs were escalated to 1983 using the
Chemical Engineering cost indexes. The capital costs used in the 1978
projection were based on the same indexes. Operating costs were esca-
lated to 1983 based on TV A projections.
The escalated cost data increased magnesia-scrubbing costs to a
greater degree than limestone-scrubbing costs. The result was a
relatively higher cost for magnesia scrubbing compared to limestone
scrubbing in the 1983 projection.
Transportation Cost Generator
Rail Rate Increase. Rail rates were projected to 1983 based on the
1973-1977 5-year historical pattern and the anticipated effects of the
"Railroad Revitalization and Regulatory Reform Act of 19.76fi (see
Figure 2).
Barge Rate Increase. Barge rates were projected to 1983 based on
the 1973-1977 5-year historical pattern, and the estimated effects of
the passage of HR-8309. HR-8309 was a bill in the House of Representa-
tives which had already passed the Senate at the time of the update
report. This bill was signed by the President October 24, 1978, in
essentially the same form used in this report. This legislation assesses
a users tax on fuel used for commercial traffic on the inland waterways
system as shown in Figure 3.
Transportation Cost Inflation. Transportation costs were projected
to increase at an average annual rate of 13.4% from 1978 through 1983.
This is an overall transportation cost increase of approximately 88% over
the 1978 projection. It is significantly above other cost escalation
rates and is expected to make transportation an increasingly important
element of FGD byproduct marketing.
Acid Production Cost Generator
Sulfuric Acid Plant Demand Base. New sulfuric acid plant capacity
between 1978 and 1983 was added to the demand data base. Plants projected
to be shut down by 1983 were eliminated. Demand was equated to 75% of
sulfur-burning sulfuric acid plant capacity. This update reflects actual
fiscal year 1977-1978 production levels projected to 1983.
Sulfuric Acid Avoidable Production Cost. This cost, the expenditures
in $/ton of sulfuric acid that could be avoided by shutting down the sulfur-
burning acid plant, was modified substantially. The major change was an
update of the byproduct steam credit to a sulfur-burning sulfuric acid
plant. In addition, the price of sulfur at Port Sulphur was increased from
$60/long ton to $70/long ton.
Market Simulation Linear Programing Model. The 1983 model was
basically unchanged from the 1978 model with two exceptions: (1) The
$0.35/MBtu (M = million) ACFL model was eliminated because of escalated
603
-------
o
o
CO
M
in
<
m
in
w
H
:Actual 1973-1978
X :Projected 1978-1983
350
300
250
200
150
100
I I I I I
I I
73 74 75 76 77 78 79 80 81 82 83
YEAR
Figure 2. Rail rate increases, actual and projected, 1973-1983.
604
-------
-o :Actual 1973-1978
X-- :Projected 1978-1983
350 _
300
CTi
250
o
o
co
M
CO
3
co
w
w
200
150
100
73 74 75 76 77 78 79 80 81 82 83
YEAR
Figure 3. Barge rate increases, actual and projected, 1973-1983.
605
-------
costs. (2) In the 1978 projection models a plant was considered in
the equilibrium solution only if limestone scrubbing was less than the
ACFL. In the 1983 models a plant was considered under an additional
condition: when limestone scrubbing cost was greater than the ACFL
a direct comparison was made between the ACFL and magnesia scrubbing.
If the resulting incremental unit cost of acid did not exceed $30/ton
the plant was considered for marketing rather than a clean fuel strategy
being automatically selected.
RESULTS
FGD Byproduct Sulfuric Acid
Selection of Power Plants for Marketing Consideration. For 1983,
94 power plants with 165 boilers were calculated to be out of compliance
and were considered in all model runs. In the 1983 $0.50 ACFL solution
five power plants were considered for marketing. In the $0.70 ACFL
solution 26 out of a total of 94 power plants were considered for
marketing. Because the number of plants at the $0.50 ACFL was limited
and these plants were also included in the $0.70 ACFL solution, a
detailed analysis was limited to the $0.70 solution.
The $0.70 ACFL model generation prescreen that resulted in 26
plants being considered for marketing is shown below:
Number of plants^
All clean fuel 43
All limestone scrubbing 24
Mixed clean fuel and
limestone scrubbing 1
Potential magnesia scrubbing 26
Total 94
The $0.70 ACFL solution potential power pla*nt supply from these 26 power
plants was 4,600,000 tons as shown in Table 1.
1983 $0.70 ACFL Solution. The solution indicated a market potential
for 7 out of the 26 plants considered. At two of the seven plants market
potential was limited to only a part of the projected production and a
mixed magnesia-scrubbing and clean fuel strategy was selected. A summary
of the strategy selection process for plants in the equilibrium solution
is shown below:
606
-------
Table 1. TWENTY-SIX POWER PLANTS INCLUDED IN 1983
$0.70 ACFL SOLUTION FOR ANALYSIS OF MARKET POTENTIAL
Plant No.
0720000900
0785000500
1040000200
1095000200
1095000600
1115001300
1145001200
1145001900
1400000600
1415000150
1655000300
2185000900
2225000800
2260000100
2260000500
2455000250
2755000600
2755000650
3080000400
3795000350
3840000500
4045000250
4480000075
4530000850
4740000300
4820001800
Location
NC
IL
OH
OH
OH
IL
MI
MI
PA
KY
FL
TX
IL
IN
IN
KY
KY
KY
MS
PA
PA
IN
SC
TX
FL
MI
MW
considered
1440
550
680
787
413
1271
1283
1185
525
800
1280
2820
2342
438
1239
2011
682
495
877
650
615
2587
595
951
1136
3247
Incremental cost ,
$/ton acid
0.00
22.61
37.76
28.34
27.70
26.31
22.59
13.09
51.66
26.27
12.15
0.00
20.24
46.35
25.96
6.43
40.22
24.37
37.84
24.65
37.44
13.39
29.51
27.77
27.83
13.26
Tons of acid
considered
67,592
74,394
137,133
363,075
67,243
321,051
197,390
117,946
69,209
132,361
196,573
156,384
402,939
73,221
156,607
338,976
161,584
99,482
67,862
82,161
68,664
334,927
100,205
159,425
226,708
449,826
4,622,938
607
-------
Number of plants
Magnesia scrubbing 5
Mixed magnesia scrubbing
and clean fuel 2
All clean fuel 1
All limestone scrubbing 18
Total 26
The seven power plants marketing sulfuric acid in the 1983 $0.70
ACFL solution had total sales of 1,243,000 tons as shown in Table 2.
The ranked order shown is the result of application of four criteria:
(1) the balance between the production capacity of the power plant and
its potential market, (2) the balance between the acid capacity of each
consumer and the byproduct acid supply from both power plants and
smelters, (3) the potential sales margin per ton of acid from each power
plant, and (4) the sales indicated as also occurring in the $0.50 solution.
Projected new boilers were a factor at six of the seven plants
selected for marketing. They contributed almost 60% of the sales even
though they made up only 15% of the total number of boilers considered
for marketing.
The eastern smelter and Canadian supply was absorbed by the market
in the 1983 equilibrium solution. No sales potential for the additional
western supply through transshipment terminals was indicated because of
higher transportation costs.
The solution would be fairly stable in the event of a sulfur price
reduction. Less than 200,000 tons out of a total of 1,200,000 tons
would be affected by a $20/long ton sulfur price reduction.
FGD Byproduct Sulfur
There is a considerable interest in elemental sulfur as a potential
power plant FGD byproduct. Marketing advantages over sulfuric acid
include over three times the equivalent sulfur concentration; nonhazardous,
noncorrosive properties; easy stockpiling characteristics; less market
competition in many locations; and recoverable energy, in the production
of sulfuric acid, of approximately 8 MBtu/short ton of sulfur.
The market for FGD byproduct sulfur is wider and potentially greater
than the market for FGD byproduct sulfuric acid. Sulfur can be marketed
as a raw material to all acid plants regardless of their size and produc-
tion costs. FGD byproduct sulfuric acid marketing is typically limited
to relatively small high-production-cost acid plants.
608
-------
Table 2. SEVEN POWER PLANTS MARKETING ACID IN THE 1983 $0.70 ACFL SOLUTION
Ranked Tons
order Plant No. Location MW of acid
1 0720000900 NC 1440 67,000
2 2455000250 KY 2011 339,000
3 2185000900 TX 2820 156,000
4 1145001900 MI 800 75,000
5 4820001800 MI 2431 335,000
6 1655000300 FL 1280 197,000
7 0785000500 IL 550 74,000
1,243,000
Consumer
Royster
Swift
Weaver
Allied
American Cyanamid
Army Ammun. Plant
Marion
Olin
American Cyanamid
American Cyanamid
Swift
Dupont
Dupont
Allied
Kerr-McGee
Royster
US I Chemical
Location
Norfolk, VA
Norfolk, VA
Norfolk, VA
Nitro, WV
Hamilton, OH
Radford, VA
Indianapolis, IN
Pasadena, TX
Joliet, IL
Ka lama zoo, MI
Calumet City, IL
North Bend, OH
Cleveland, OH
Cleveland, OH
Cottondale, FL
Mulberry, FL
Tuscola, IL
1,
Tons
of acid
15,000
26,000
26,000
101,000
71,000
125,000
42,000
156,000
38,000
7,000
30,000
131,000
150,000
54,000
11,000
186,000
74,000
243,000
-------
Although no byproduct sulfur sales were shown in either the 1983
or the 1978 model solutions a sensitivity analysis showed that it could
become competitive in certain locations with a relatively small (10%)
reduction in total FGD byproduct sulfur costs. At one power plant, for
example, a reduction of 8.5% ($1.18 M/yr) of the total production cost
($13.91 M/yr) of FGD sulfur would result in a reduction in incremental
cost sufficient to enter the market in competition with sulfur delivered
from Port Sulphur.
Of the 16 power plants that have relatively low estimated sulfur
production costs compared to limestone scrubbing, the first plant becomes
competitive with a reduction in total FGD sulfur production costs of
only 3.1%. The other 15 plants become competitive with reductions
ranging from 5.4 to 23.7%. The combined production of these plants
would be 265,723 short tons/year, equivalent to over 800,000 tons of
sulfuric acid. This represents slightly over 3.0% of the projected
1983 sulfuric acid demand from sulfur-burning plants.
Marketing sensitivity to FGD byproduct sulfur is based on sulfur
delivered from Port Sulphur at $70/long ton plus delivery and handling
expenses escalated to 1983. At the level of projected 1983 Port Sulphur
delivered costs, Canadian sulfur recovered from sour gas may be more
competitive in the Midwest and Great Plains where 6 of the 16 plants
analyzed are located.
CONCLUSIONS
The need of an extended lead time was apparent. The increasing
period between the decision to build and plant completion requires a
corresponding increase in projection lead time. It is critical, therefore,
that projections be extended as far into the future as is practical.
This may require more recent data than have been available through
conventional channels.
The marketing potential of boilers projected for completion between
1978 and 1983 dominated the solution. Projections must emphasize future
boilers since they are expected to be the most likely candidates for FGD
byproduct marketing.
Although boilers reported to be in compliance are excluded from the
model on the basis that strategy selection is final, this may not be
necessarily true for future boilers if the strategy selected is clean
fuel. Proposed regulations, higher clean fuel premiums, and scrubbing
technology improvements could alter the choice of clean fuel as a total
compliance strategy for future boilers. Model solutions should include
future boilers if compliance is projected as a result of clean fuel
usage.
Application of more stringent environmental controls must be
analyzed for effects on model solutions.
610
-------
ri.s FGD, transportation, and sulfuric acid avoidable production
costs increase, correspondingly higher ACFL values will be required
to fully analyze potential market interactions for extended projections.
Greater emphasis should be given to potential transportation
advantages of water traffic. Backhaul potential should also be considered
where applicable.
FGD byproduct sulfur is not yet competitive with FGD byproduct
sulfuric acid. A relatively small reduction (10% or less at nine power
plants) in total FGD byproduct sulfur costs, however, could result in
emergence of competitive FGD sulfur production.
Projections of avoidable construction costs for future sulfuric
acid plants should be developed. The current model only considers the
costs which can be avoided by shutting down an existing or future sulfuric
acid plant operation. For future considerations it also includes all
costs which can be avoided by not building new plants.
611
-------
REFERENCES
1. Anders, W. L. Flue Gas Desulfurization Byproduct Production and
Marketing System Users Manual. Tennessee Valley Authority, Office
of Agricultural and Chemical Development, Muscle Shoals, Alabama.
Prepared for the U.S. Environmental Protection Agency (in press).
2. Bucy, J. I., R. L. Torstrick, W. L. Anders, J. L. Nevins, and
P. A. Corrigan. Potential Abatement Production and Marketing of
Byproduct Sulfuric Acid in the U.S. Bulletin Y-122, Tennessee
Valley Authority, Office of Agricultural and Chemical Development,
Muscle Shoals, Alabama. EPA-600/7-78-070, U.S. Environmental
Protection Agency, Office of Research and Development, Washington,
D.C., 1978. (NTIS PB 284-200)
3. Herlihy, J. Flue Gas Desulfurization in Power Plants. Status
Report, U.S. Environmental Protection Agency, Department of
Stationary Source Enforcement, Washington, D.C., 1977.
4. McGlamery, G. G., R. L. Torstrick, W. J. Broadfoot, J. P- Simpson,
L. J. Henson, S. V. Tomlinson, and J. F. Young. Detailed Cost
Estimates for Advanced Effluent Desulfurization Processes.
Bulletin Y-90, Tennessee Valley Authority, Office of Agricultural
and Chemical Development, Muscle Shoals, Alabama. EPA-600/2-75-006,
U.S. Environmental Protection Agency, Office of Research and
Development, Washington, D.C., 1975. (NTIS PB 242-541/1WP)
5. O'Brien, W. E., and W. L. Anders. First Annual Update - Potential
Abatement Production and Marketing of Byproduct Sulfur and
Sulfuric Acid in the U.S. Tennessee Valley Authority, Office of
Agricultural and Chemical Development, Muscle Shoals, Alabama.
Prepared for the U.S. Environmental Protection Agency (in press).
612
-------
LIMESTONE FGD OPERATION
AT
MARTIN LAKE STEAM ELECTRIC STATION
BY
MARK RICHMAN
MANAGER, PRODUCT IMPROVEMENT
RESEARCH-COTTRELL
613
-------
ABSTRACT
The startup of the Flue Gas Desulfurization (FGD) System at Texas
Utilities' Martin Lake Steam Electric Station, Unit #1, in April
of 1977 marked the first use of FGD technology on a full-scale
lignite installation. The station also features fully closed
loop FGD operation with no liquid waste streams generated, disposal
of the truckable FGD sludge-fly ash solid waste product at the*
mine site, and the use of bypass gases for all scrubbed gas re-
heating.
Since the startup, the Double-Loop, , FGD Systems for Unit #1
and Unit #2 (June of 1978) have operated at well in excess of 95%
absorber efficiencies and 90% reagent utilization while consuming
less than 1.3% of the station's electrical power production. In
addition the FGD Systems have been extremely reliable, causing
only 5 days of unscheduled boiler downtime combined as of
December 1, 1978.
This paper, prepared by Research-Cottrell - the supplier of the
FGD and Solids Handling Systems, presents a discussion of the
Martin Lake S.E.S. FGD Systems design, startups, operating exper-
ience to December 1, 1978, performance and operating requirements.
614
-------
DESIGN CRITERIA
Martin Lake Steam Electric Station consists of four 750,000
kilowatt lignite-fired units. The station, located in Rusk County,
Texas, is surrounded on three sides by Martin Lake, which serves
as a source of makeup and cooling water. The station is also less
than ten miles from the source of its lignite fuel.
The Texas lignite fuel for the Martin Lake S.E.S. has the
following characteristics:
Heating valve, BTU/LB
Ash Content, %
Sulfur Content, %
Moisture content, %
MINIMUM
AVERAGE
MAXIMUM
6,972
5.6
0.5
29.0
7,380
8.0
0.9
33.0
7,894
13.2
1.5
37.9
To produce 750,000 kilowatts of power with this fuel, each
unit requires approximately 1,000,000 Ibs/hr of lignite fired.
To meet this demand, the utility operates a drag line with a 94
cubic yard bucket at the mine source.
The particulate emissions code for Martin Lake is 0.1 Ibs/MMBTU.
A 99.4% efficiency Research-Cottrell Double-Deck Electrostatic
Precipitator is provided to achieve this requirement. The SO
emissions code for Martin Lake is 1.2 Ibs/MMBTU. This computes
to about 71% SO removal for the worst case (i.e. maximum sulfur)
fuel. Other requirements for the FGD System included: totally
closed loop operation; a minimum of 25 F reheat at. the stack entry;
and the production of a truckable, dumpable disposal product that
can be transported back to the mine site.
615
-------
FGD SYSTEM DESIGN
The process provided by Research-Cottrell for SO2 removal at
Martin Lake is the Double-Loop. Limestone System. A schematic
of the basic process design is shown in Figure #1. The Double-Loop
process differs from simple single-loop processes in that two sep-
arate sets of chemical conditions are maintained. In a single loop
process, shown below, the limestone slurry contacting the flue gas
is kept at a fixed set of chemical conditions (i.e. solids composi-
tion and pH) for the required level of SO removal. Since high levels
SINGLE LOOP
OUTLET
FLUE GAS
T
ABSORBER
t
INLET
FLUE
GAS
ALKALI
LIMESTONE
1
ABSORBER
TANK
1
I
WASTE EFFLUENT
FLY ASH PRECIPITATION -
SO2 ABSORPTION SYSTEM
Reagent
Storage/Feed
FIGURE #1
Absorber
Absorber FeedPumps
Feed
Tank
oo"—'oo
616
-------
of SO removal efficiency can only be obtained with sufficiently high
inventories of limestone reagent present, and since the solids blow-
down to the solids handling/dewatering system is at the same chem-
ical conditions as the scrubbing slurry, it is difficult to achieve
both high SO removal efficiencies and high reagent utilization with
this type of system. As a result, typical first generation systems
of this type achieve reagent utilization levels no better than 70 to 80%
In a two loop process, the first slurry loop (i.e. the Absorber
Loop) operates identically as the single loop process operates to
insure high SO removal efficiencies. The second slurry loop
(i.e. the Quencher Loop) receives the slurry discharge from the
first loop and re-uses it to obtain reagent utilization levels
exceeding 90%, even with absorber tower SO removal efficiencies
exceeding 95%.
TWO LOOP
LIMESTONE FEED
1
WASTE EFFLUENT
Using the Double-Loop,_,... approach, six absorber towers each
were provided for Martin Lake #1 and for Martin Lake #2. A
schematic of the arrangement of the absorber area is shown in
Figure #2. Each absorber tower is designed to treat 12.5% of
the maximum boiler.flue gas output. Therefore the entire FGD
System can treat up to 75% of the maximum boiler output with all
six towers in operation. In order to achieve the 71% overall SO
removal requirement for the worst fuel case, this means that each
tower must be capable of achieving 95% SO removal efficiency.
The Martin Lake absorber tower design is presented in Figure #3.
Each absorber tower, 28 feet in diameter by 100 feet tall, features
three stages of SO removal units followed by a two stage mist
617
-------
FIGURE #2
MARTIN LAKE #1
FGD SYSTEM
ABSORBER AREA -\
ARRANGEMENT
elimination system. Boiler flue gas at 335 F enters the tower
tangentially into the cyclonic quencher, a co-current spray cham-
ber feeding quencher (i.e. second slurry loop) slurry. Here the
gas is quenched to saturation, most of the fly ash not removed in
the ESP is removed, and 25-30% of the SO is removed. After
passing through a liquid-gas separator which recycles absorber
slurry back to the Absorber Feed Tank, the gas passes through three
levels of absorber sprays and two feet of the fixed bed wetted film
contactor. These two SO removal sections remove virtually all of
<£•
the remaining SO by contacting the flue gas with absorber (i.e.
first slurry loop) slurry. The cleaned gas then passes through
two chevron mist eliminator stages and exits the absorber tower.
Once the cleaned gases exit the absorber towers, they are
recombined with the portion of the boiler flue gases that was not
scrubbed to reheat the treated gases. Since no more than 75% of
the gases is treated at any one time, a minimum of 48°F of reheat
(well above the required 25 F) is obtained. Note that at lower
sulfur levels, where less than 71% overall SO removal is required,
less gas is treated so that more gas is bypassed giving even greater
levels of reheat.
618
-------
FIGURE #3
SO: Stripped
Flue Gas
Upper
Demister
Lower
Demister
Wetted Film
Contactor
Spray
Tower
Liquid Gas
Bowl-Separator
Slurry Return
to Tank
Cyclonic
Quencher
1 Quencher
Sump
DOUBLE-LOOP (TM)
QUENCHER-ABSORBER TOWER
619
-------
SOLIDS HANDLING SYSTEM DESIGN
Waste slurries from each absorber tower at 15% solids combine
into a single stream to feed a 140 ft. diameter by 12 ft. sidewall
gravity thickener. The thickener underflow at about 35% solids
is fed to one of three centrifuges for additional dewatering.
Cake from the centrifuges at 68-70% solids then discharges into
a Muller-type blender. Here the cake combines with fly ash
collected in the ESP to to form a truckable, dumpable blend. Rail-
cars receive the blend for ultimate disposal.
Thickener overflow and centrate streams are collected in wet
wells for recycle back to the FGD System to insure fully closed
loop operation. First generation FGD systems experienced process
SINGLE LOOP CLOSED LOC
Oulle
1 Flue Gas
1 l~
1 ^.__ ,
1 Absorber
1 *
i i7n,rr ~~ ~~i
1 Flue Gas I
1
1
Absorber
Tank
, ,
Dewatering
System
)P SYSTEM
i
1 Alkali Limestone
| Fresh Makeup
1
1
1
Return I
Water
!____! _,
y Dewatered
Sludge
Absorber Loop = Closed Loop
Total Process = Closed Loop
DOUBLE-LOOP CLOSED LOOP SYSTEM
1 Outlet
Flue Gas
_^
1 Absorber
1 *
---t
1
Quencher
) ^
1 t ,
I Inlet ~
Flue Gas |
1
L _
Absorber Loop = Open Loop
Quencher Loop = Closed Loop
Total Process = Closed Loop
Absorber
Tank
-1-
Quencher
Tank
\
Dewatering
System
| Alkali Limestone
| Fresh Makeup
1 Water
1
| Fresh Makeup
Water
i
i
Return |
Water
1 Dewatered
Sludge
620
-------
problems due to closed loop operation. As shown below, these
single loop systems to not isolate the return water to any part
of the process, but instead expose the entire process to disposal
return water.
The Double-Loop. closed loop system works differently in
that disposal return streams are concentrated only in the Quencher
Loop. The Absorber Loop, with the wetted film contactor and demist-
ers, runs as an open loop, even though the total process is closed
loop. As a result, the Absorber Loop can operate unsaturated in
calcium sulfate, minimizing the potential for harmful scale
formation.
621
-------
STARTUP AND OPERATING HISTORY
Research-Cottrell provided the equipment, design engineering
and advisory services during construction, checkout and startup
for the FGD System. H. B. Zachry Co. constructed the FGD System.
TUGCo was responsible for checkout of equipment and instrumentation,
startup, operation and maintenance of the FGD System.
A chronological account of the first nineteen months of FGD
System operating at Martin Lake is presented below. Following
this we will discuss mechanical and instrument related operating
problems and solutions.
April 21, 1977
April 26, 1977
April 29, 1977
June 5, 1977
June 10, 1977
June 24, 1977
Flue gas admitted to tower 1C-T200.
Flue gas admitted to tower 1C-T300.
Flue gas admitted to tower 1B-T200 and 1B-T300;
stack SO monitor indicates the unit is in SO
£ £
compliance.
Boiler shutdown for tube leak permits completion
of construction on A module towers. Inspection
of towers reveals wetted film contactors and
demisters with little or no solids accumulation.
Boiler back on line at 800-810 MW with all six
towers now in operation.
Preliminary acceptance tests at 750 MW indicate
system can achieve compliance with four towers
up to 900ppm SO and with five towers up to
l,200ppm SO,
at 98-99%.
Tower SO removal efficiencies
August 7, 1977
Boiler operation resumed after month-long shut-
down resulting from furnace damage.
August 16, 1977 - Slurry introduced to the thickener; thickener
overflow returned to FGD System.
622
-------
August 25, 1977 - Sludge introduced to centrifuges; blending oper-
ation begins.
September, 1977 - Environmental Protection Agency certifies
Martin Lake #1 as being in full compliance
with all air quality regulations.
November 3, 1977 - Due to low absorber tower gas flows and an in-
ability to fully isolate the towers, the
boiler was shut down. Inspection of the scrub-
bers revealed wetted film contactor (WFC)
support beams in tower 1C-T200 had failed with
WFC falling into the liquid-gas separator and
tower sump. The inspection also showed tower
inlet wet-dry line buildup sufficient to have
contributed to the low gas flows.
November 8, 1977 - Boiler and all six towers restarted with all
WFC removed to prevent additional support beam
failures.
NOTE THAT THIS FIVE DAY PERIOD WAS THE ONLY
BOILER SHUTDOWN PERIOD DURING THE FIRST NINETEEN
MONTHS OF BOILER OPERATION CAUSED BY FGD SYSTEM
MALFUNCTION.
(Note also that two other shutdowns (March 15, 1977
for three days and April 8, 1977 for two days),
were related to bypass damper failure.)
January 28, 1978 - Tower 1A-T200 was the first to replace FRP support
beams with alloy WFC support beams; other towers
subsequently received alloy support beams.
Inspection of this tower after 2^ months of
operation with no WFC revealed significant demister
buildup not present when the tower was operated
with the contactor. This buildup necessitated
the replacement of much of the demister packing
623
-------
April 2, 1978
May 1, 1978
June 22, 1978
June 23, 1978
August, 1978
and some demister beams. Note that an inspec-
tion of a tower with two feet of WFC amde in
the Spring, 1978 after seven weeks of operation
revealed the WFC and demisters to have only
insignificant buildup.
- Unit #1 taken off line for annual two week outage.
- Unit |1 taken off line due to generator problems;
boiler operation resumed on June 21, 1978.
- Unit #2 startup begins with 2 towers (B module)
placed into service.
- Unit #2 C module placed into service.
- Unit #1 FGD System acceptance tests conducted by
Texas Utilities; results show the Absorber Towers
to be removing 98-99% of the incoming SO .
October 11, 1978 - Tower 1B-T200 taken out of service for 2 weeks
to repair lining; Unit #2 remained in compliance
during this period with the other 5 towers in
operation.
October 19, 1978 - Unit #1 FGD System accepted by Texas Utilities
November 23, 1978- Unit #1 taken off line for month long scheduled
outage.
624
-------
FGD OPERATING PROBLEM.S AND CORRECTIVE ACTION
The following discussion presents the major mechanical, struc-
tural and instrumentation related problems encountered during the
first nineteen months of operation,of the Martin Lake #1 FGD System
and the first five months of operation of the Martin Lake #2 FGD
System. Devices and pieces of equipment not discussed can be
assumed to have performed satisfactorily, but should not be assumed
to have been totally trouble-free.
Dampers - Dampers provided for the Martin Lake FGD System include
a single-louver bypass damper, two consecutive louver dampers at
each tower inlet, and single link-louver dampers at each tower
out-let. Even with the addition of seal air flowers between the
two tower inlet dampers, isolation of individual towers for per-
sonnel access and maintenance has been extremely difficult, but
is now possible. Areas of malfunction have included bearings,
seal strips, and linkages.
Expansion Joints - All gas-side expansion joints originally pro-
vided have been replaced due to the original joints' deterioration
and inability to withstand exposure to liquids. The replacement
joints are performing adequately to date.
Liquid Flow Switches - Sonic flow switches, originally provided
to monitor flow deviations in key process loops, were found to be
inadequate for electric generating station environments. They were
eliminated, where possible, or replaced with magnetic flow meters.
pH Meters - Submersion devices in the quencher instrument wells
experienced calibration drifts and electrical problems. These
were replaced, following an extensive test program, by flow-through
devices with ultrasonic cleaners. These have performed extremely
well for 6 months.
Liquid Level Devices - Sonic level devices in the quencher instru-
ment wells experienced problems related to electronics, poor cal-
ibration, slurry foaming, fouling of transducers and excessive
liquid level variations. These problems were addressed and the
devices have functioned adequately since October, 1977.
625
-------
Gas Flow Meters - Poor gas distribution at the absorber tower inlets
led to the relocation of these devices to the tower outlets. High
instrument air moisture content and probe pluggage also contributed
to poor operating performance.
On-Off Valve Operators - Electric operators on reagent feed and
density control valves experienced numerous failures in early oper^
ation. Some of these were replaced with pneumatic valve operators
and others were modified to improve performance.
WFC Support Beams - The original FRP support beams were replaced
with 316SS beams in January and February, 1978. The FRP beams had
failed completely in one tower and were on the verge of failure
in several other towers when the WFC was removed from all towers
on November 7 and 8, 1977.
Failure of these beams was due to inadequate structural design
but was accelerated due to solids accumulations in the wetted film
contactor. These solids accumulations were caused by operating
the FGD System out of chemical balance due to problems with pH
devices, on-off valve operators and the difficulty in manual
control that these malfunctions created. Poor slurry distribution
to the contactor may also have accelerated this buildup. As stated
earlier, recent operation with two feet of WFC and with the new
valve operators has indicated the WFC and demister packings to have
had only insignificant buildup.
Wet-Dry Line - Buildup at the cyclonic inlets of the absorber towers
was discovered in November, 1977. Model testing determined modifi-
cations to the inlets to allow operation of the towers for a full
year without cleaning. These modifications have been successful
and are being added to all absorber towers.
626
-------
FGD SYSTEM PERFORMANCE
Several test programs were conducted by Research-Cottrell for
TUGCo to evaluate the Martin Lake FGD System and to optimize
operating setpoints. The following summarizes key results of
the testing:
S0? Removal - Towers with four feet of wetted film contactor achieved
greater than 99% SO removal. Towers with no wetted film contactor
achieved 80-85% removal at peak tower gas velocities. Based on
this testing, two feet of WFC was installed in each tower and the
maximum tower gas throughput was increased by 10%. Unit #1 acceptance
tests in August of 1978 revealed absorber tower SO removal to be
greater than 98% with two feet of WFC at design velocities.
Reagent Utilization - Typically greater than 90%; average of 90-92%
utilization. Tests on tower 1C-T200 on in situ forced oxidation
showed that the tower can be operated with a reagent utilization
in excess of 98% on a continuous basis.
Scale Control in Closed Loop Operation - Testing of the saturation
levels of sulfate in each loop indicated an average Absorber Loop
sulfate saturation level '^ 0.90 and an average Quencher Loop sulfate
saturation level ^ 1.20. This indicates that, even during closed
loop operation, the Absorber Loop operated unsaturated in SO., pre-
venting sulfate scale formation in the absorber section of the
towers. The Quencher Loop sulfate level shows, as expected, the
sulfate level to be super-saturated but below the 1.3 times satura-
tion critical level.
Disposal % Solids - Each system operated within 1 - 2% of design
% solids with the thickener underflow ^ 35% solids, the centrifuge
cake ^ 68% solids and the fly ash sludge blend at ^ 82% solids.
Tower Pressure Drop - The absorber tower pressure drop at the
design gas flow rate was within 0.1 IWC of the predicted value at
4.5 IWC. This does not include the two tower inlet louver dampers.
Degree of Oxidation - During normal operation, the quencher discharge
has a sulfite:sulfate ratio of about 50:50 (or 1:1). A test pro-
gram, using in situ forced oxidation on tower 1C-T200, produced
100 tons of solids with a sulfite:sulfate ratio of 1:99.
627
-------
FGD SYSTEM OPERATING REQUIREMENTS
The following are operating and maintenance requirements for
Martin Lake #1 FGD and Solids Handling Systems based on the first
nineteen months of operation. Consumption rates are based on
750MW, 0.9% sulfur and 80% boiler availability per year. (Note
that the unit normally operates at 800 - 820MW.)
Reagent Consumption - 240 tons/day or 70,000 tons/year of 95%
CaCO limestone.
Makeup Water Consumption - 550 gpm or 700 acre-ft/year.
Sludge Generation - 500 tons/day or 150,000 tons/year of 68% solids
centrifuge cake.
Blended Waste Product - 13 railcar loads/day or 3,800 railcar loads/
year.
Power Consumption - 10MW or 1.3% of station generated power -
This includes ID Fan horsepower to account for FGD System AP,
quencher pumps, spray tower pumps, WFC pumps, slurry agitators,
reagent preparation system, solids handling system, scrubber control
room operation, HVAC and lighting and all auxiliary equipment and
motor control centers related to the FGD System. The Electrostatic
Precipitator is not included in this total.
Operating Personnel - 2 Scrubber Area Operators/shift; 1 Reagent
Area Operator/day (1 Shift); 2 Solids Handling System Operators for
two units/shift.
Therefore, for Unit #1 FGD and Sludge Systems, a total of 3 1/3
Operators per shift are required.
Maintenance Personnel - Actual maintenance manpower has averaged
1,700 manhours/month. Actual chemical technician manpower has
averaged 200 manhours/month. This computes to about 12 manmonths/
month for maintenance and chemical technicians.
628
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing/
1. REPORT NO.
EPA-600/7-79-167a
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE Proceedings: Symposium on Flue Gas
Desulfurization--Las Vegas, Nevada, March 1979;
Volume I
5. REPORT DATE
July 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Franklin A. Ayer, Compiler
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMEN'
E HE 62 4 A
NO.
11. CONTRACT/GRANT NO.
68-02-2612, Tasks 55 and 99
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD I
Proceedings; 3/5-8/79
COVERED
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES jERL-RTP project officer is Charles J. Chatlynne, Mail Drop 61,
919/541-2915.
16. ABSTRACT The publication, in two volumes, contains the text of all papers presented
at EPA's fifth flue gas desulfurization (FGD) symposium, March 5-8, 1979, at Las
Vegas, Nevada. 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 processes, actual operating experience, waste disposal and byproduct market-
ing , and industrial boiler applications.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
Pollution Energy
Flue Gases Waste Disposal
Desulfurization Byproducts
Sulfur Oxides Marketing
Environmental Biology
Economics Boilers
Pollution Control
Stationary Sources
Health Effects
Industrial Boilers
13 B 14B
2 IB
07A,07D
07B
06F
05C 13A
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
21. NO. OF PAGES
635
22. PRICE
EPA Form 2220-1 (9-73)
628a
* U.S. GOVERNMENT PRINTING OFFICE--1979/64O-013/3937
------- |