United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S9-85/033 Mar. 1986
Project Summary
Proceedings: Ninth
Symposium on Flue Gas
Desulfurization, Cincinnati,
Ohio, June 1985
Franklin A. Ayer and Anna W. Wallace
The proceedings of the Ninth Sympo-
sium on Flue Gas Desulfurization (FGD)
consists of two volumes: Volume 1,
Opening Session, Commercial Status of
FGD, Limestone FGD/Organic Acid En-
hancement, FGD Reliability Improve-
ment, Chemistry/Reagent Preparation,
and Materials of Construction; and Vol-
ume 2, Panel Discussion on Retrofitting
FGD Systems, Dual Alkali, Emerging
Technologies, Spray Dryer FGD, FGD
By-product Disposal/Utilization, Unpre-
sented Papers (8), and Attendees List-
ing. EPA and EPRI cosponsored this
symposium which was held in Cincin-
nati, OH, June 4-7,1985.
The meeting was a forum for the ex-
change of technical and regulatory in-
formation and developments regarding
systems and processes applicable to
utility and industrial boilers. Addresses
at the opening session: (1) compared
regional declines of forests in Europe
with those of North America and the
possible role of airborne chemicals in
these declines, (2) examined the cur-
rent legislative/regulatory situation,
and (3) gave results of a study of the
economics of FGD systems in a variety
of SO2 reduction strategies for coal-
fired power plants. Subsequent techni-
cal sessions dealt with new and exist-
ing FGD technologies that enabled
utility and industrial users of the tech-
nology, consultants, engineering firms,
equipment manufacturers, process
suppliers, government agencies.
academia, and research and develop-
ment firms to share their experiences.
This Project Summary was devel-
oped by EPA's Air and Energy Engineer-
ing Research Laboratory, Research Tri-
angle Park, NC, to announce key
findings of research projects that are
more fully documented in two separate
reports of the same title (see Project
Report ordering information at back).
Introduction
This Summary consists primarily of
abstracts of presentations made and pa-
pers presented at the Ninth Symposium
on Flue Gas Desulfurization. Titles and
authors of several papers prepared for
the meeting, but not formally pre-
sented, are also included.
Abstracts
Volume 1
Session 1: Opening Session
Comparison of Regional De-
clines of Forests in Europe and
North America: The Possible
Role of Airborne Chemicals
Ellis B. Cowling, North Carolina
State University, Raleigh, NC
Human activities of many sorts have
been changing the chemical climate of
Europe and North America especially
since the Industrial Revolution. Re-
gional declines of forests also have
-------
been occurring in Europe for the past
200 years and in North America for at
least the past century. A total of 18 re-
gional declines of forests have been re-
ported on the two continents. These de-
clines are generally believed to have
been caused by a combination of com-
petitional, physical, biological, and
chemical stress factors acting simulta-
neously or sequentially. Airborne chem-
icals have been implicated as an impor-
tant causal factor in only 4 of the
18 cases. A consensus regarding the
possible role of selected airborne chem-
icals is described. This judgement sug-
gests that ozone and biologically avail-
able nitrogen compounds are among
the chemical stress factors of greatest
probable importance to forests. Other
chemical stress factors that may also be
involved include toxic gases other than
ozone, toxic metals, acids, acidifying
substances, and growth-altering or-
ganic chemicals. Implications of this
concensus are discussed briefly with re-
gard to management of air quality in
Europe and North America.
Address on Current Legislative/
Regulatory Situation
Joseph A. Cannon, Pillsbury,
Madison, and Sutro, Washing-
ton, DC
No written material furnished.
The Role of FGD in SO2 Reduc-
tion Strategies for Coal-Fired
Power Plants
E. S. Rubin, M. Cushey, R. J.
Marnicio, C. N. Bloyd, and J. F.
Skea, Carnegie-Mellon Univer-
sity, Pittsburgh, PA
A newly developed computer model
for detailed analysis of S02 control
strategies and costs at the unit, state,
and regional levels is described. Illustra-
tive results are presented in which a
broad range of parameters affecting
FGD retrofits at existing coal-fired
power plants are examined for an eight-
state region of the midwest, using site-
specific plant and fuel characteristics.
The future cost and capacity of FGD
retrofit systems are seen to depend
strongly on S02 reduction require-
ments. For the highest levels of control,
wet limestone scrubbing in conjunction
with physical coal cleaning often pro-
vided the most economical means of
compliance, particularly if only "local"
coals were used. The future availability
of dry S02 removal processes (lime
spray dryer and LIMB) is seen to have a
potentially significant influence on the
choice of S02 control technology, if
emission reductions of 50% or more be-
come required in the next decade. The
potential effects of plant life extension
programs on the use of conventional
and advanced FGD systems also are sig-
nificant. The way in which overall S02
emission reductions are allocated to in-
dividual sources, and the degree to
which utility coal choices are con-
strained, also have a pronounced effect
on control technology selection and
cost.
Session 2: Commercial Status
of FGD
The Present and Future Status
of FGD in the United States
Robert W. Mcllvaine, The
Mcllvaine Company, North-
brook, IL
More than 50,000 MW of FGD sys-
tems are in operation in the U.S. More
than 100,000 MW will be in operation by
1991. An upsurge in new boiler orders
beginning in the next few years will
raise the FGD total to close to 175,000
MW in the U.S. by the late 1990's. Acid
rain control could be important in deter-
mining this total, but variations in peak
load demand and boiler replacement
rates will be more significant in the long
term. The commercial situation is con-
stantly changing. Architect/engineers,
boiler companies, scrubber system de-
signers, and component suppliers are
attempting to take more responsibility.
This is spurring competition between
these groups and between individual
suppliers. Integration, new technology,
and internationalization are also shap-
ing the competitive environment.
Survey Report and Assessment
of Utility Flue Gas Desutturiza-
tion System Installations in
Europe
William Ellison, Ellison Consul-
tants, Monrovia, MD
This paper gives details of recent sur-
veys of major programs for FGD and
NOX removal installation in West Ger-
many. The paper: provides an under-
standing of the principal types of con-
trol system designs that are being
applied, outlines technological ad-
vancements that are being achieved,
and describes operating experience
gained to date in expanded use of FGD
and selective catalytic reduction (SCR)
in Europe in the 1980s. Significant dif-
ferences between FGD and NOX re-
moval in the U.S. and Japan are re-
viewed, and information that may
improve the operation and reliability of
new and retrofit installations in U.S. is
offered.
Recent Developments in SO2
and NOX Abatement Technol-
ogy in Japan
Jumpei Ando, Chuo University,
Tokyo, Japan
More than 1400 FGD plants have been
built in Japan with a total capacity of
treating 130 million Nm3/h of flue gas
(43,000 MW equivalent). Thirty fijGD
plants have been operated treating
about 90% of total flue gas from coal-
fired utility boilers. All of the FGD plants
for coal by-produce salable gypsum. On
the other hand, the total capacity of se-
lective catalytic reduction (SCR) of NOX
exceeded 90 million Nm3/h (30,000
MW). Twenty coal-fired boilers use SCR
in addition to FGD. All of the FGD and
SCR plants have been operated with
over 99% reliability. The investment
cost for limestone/gypsum process FGD
plants for coal-fired boilers has been
lowered from about 30,000 yen/kW* in
1979-1980 to about 20,000 yen/kW in
1983-1984. The power consumption of
the FGD processes was also reduced
from 2.1-2.6% to 1.7-2.1%. The annual-
ized cost of the wet limestone/gypsum
process is currently about 1.4 yen/kWh
including 7 years depreciation and 10%
interest. The investment cost of SCR for
coal-fired utility boilers is about 6,000
yen/kW, while the annualized SCR cost
is 0.5-0.6 yen/kWh including 7 years de-
preciation and 10% interest. For the re-
moval of acid (H+), NOX abatement by
combustion modification is the most
economical, while SCR is more costly.
There have been no significant develop-
ments in simultaneous SO2 and NOX re-
moval technology except that two test
plants have been operated with acti-
vated carbon processes. Several indus-
*250yen~$1 (U.S.)
-------
trial coal-fired boilers using fluidized-
bed combustion have been put into
operation, reducing both S02 and NOX.
EPA's Stationary Source Con-
trol Technology Research
Program
Carl R. Gerber, U.S. Environ-
mental Protection Agency,
Washington, DC
The reduction or control of S02 emis-
sions from coal combustion—in a cost-
effective manner—is a topic of wide-
spread interest today, just as it was at
the first symposium on FGD held over
12 years ago. More recently S02 as well
as NOX emissions have been tied to the
problems of acidic deposition and visi-
bility degradation. Research to reduce
NOX emissions has focused on combus-
tion modification. EPA's major thrust
has been to develop technology for new
Iow-N0x burners that may be retrofitted
to existing boilers or incorporated into
new designs. EPA's research on particu-
late technologies is directed toward re-
ducing costs by as much as 50% and
minimizing operational uncertainties.
Two approaches under development
are the use of a multistage electrostatic
precipitator (ESP) and electrostatic en-
hancement of fabric filtration.
EPA's engineering program has been
involved in many of the technological
advances in FGD. EPA continues to con-
duct research to develop higher reliabil-
ity and lower cost S02 emission reduc-
tion methods. An EPA in-house pilot
plant program is focusing on spray dry-
ing FGD—particularly on reducing
costs, such as the use of cheaper or
more reactive sorbents. EPA is also de-
veloping several innovative concepts in
SO2 control which will offer low-cost
options for acidic deposition reduction
if proposed legislation calling for this
reduction becomes law. EPA's largest
effort among these combined emission
reduction technologies is demonstra-
tion of the technology known as LIMB
(limestone injection combined with
low-NOx multistaged burners). Another
approach to lower cost technology
combines S02 control with particle
emission control, a concept that EPA
has developed, involving spray drying
in a modified ESP; this concept has
been dubbed "E-SOX."
In addition to its hardware-related
control technology research, EPA has a
continuing effort to assess the perform-
ance and engineering costs of both
commercial and emerging technolo-
gies.
Finally, a major effort in EPA's engi-
neering program has been what many
refer to as "technology transfer." This is
EPA's way of disseminating data and in-
formation from technological develop-
ments by EPA and others via published
papers, technical reports, industry brief-
ings, and symposiums. Over 3 years
ago, EPA joined forces with EPRI in con-
tinuing the FGD symposium series. This
joint sponsorship has been highly suc-
cessful in making these meetings more
responsive to the needs and interests of
those who need this data and informa-
tion in their work, particularly the users
of FGD technology. EPA also benefits
from the other work discussed at these
symposiums.
Session 3: Limestone FGD/Or-
ganic Acid Enhancement
Results of Using Organic Acid
in San Miguel Electric's Flue
Gas Desulfurization System
Jack M. Burke, Radian Corpora-
tion, Austin, TX; Robert Cmiel,
San Miguel Electric Coopera-
tive, Jourdanton, TX; and
J. David Mobley, U.S. EPA, Air
and Energy Engineering Re-
search Laboratory, Research
Triangle Park, NC
This paper summarizes San Miguel
Electric Cooperative's first full year of
operation after conversion of the Unit 1
FGD system to an organic acid en-
hanced limestone system. Plant operat-
ing data from 1983 (without organic
acid addition) and 1984 (with organic
acid addition) were reviewed. Based on
that review, changes which occurred
after organic acids were added to the
system were identified and cost savings
resulting from those changes esti-
mated. The cost estimates showed that
savings directly attributable to organic
acid use exceeded the cost for organic
acid by $109,000 in 1984. Further, addi-
tional cost benefits resulting from or-
ganic acid use may have been in excess
of $3 million. Because of this positive
experience, San Miguel continues to
use organic acid in its FGD system.
State-of-the-Art Design Appli-
cations on a Closed-Loop FGD
System
Dennis Laslo and Even Bakke,
Peabody Process Systems, Inc.,
Norwalk, CT
Seminole Electric Cooperative's
Units 1 and 2 in Palatka, FL, are capable
of 1240 MW of generating power. The
limestone-based FGD systems for both
boilers incorporate an organic acid ad-
dition system for operation at high dis-
solved salt concentrations, and, on Unit
1, a unique double draw-off crystallizer
unit operation for improved cake de-
watering, and a two-tank forced oxida-
tion system for production of gypsum.
Discussions include full-scale closed-
loop water management and operating
experience, results from a commercial
gypsum dewatering pilot plant, results
of double draw-off product removal on
both calcium sulfite and calcium sulfate
crystals, and operating experience and
economics of organic acid addition.
Also, results of a successful test of
sodium thiosulfate addition as a gyp-
sum scale inhibitor are presented.
Dibasic Acid Test and Chemical
Process Evaluation at Peters-
burg Unit 3 FGD System
D. Guetig, Indianapolis Power
& Light Company, Petersburg,
IN; and S. Ou and C.P. Wedig,
Stone & Webster Engineering
Corporation, Boston, MA
The use of dibasic acid (DBA) in a FGD
system buffers the scrubber chemistry,
resulting in increased scrubber S02 re-
moval efficiency. The use of DBA may
also result in reduced scrubber electric-
ity consumption. It is expected that Indi-
anapolis Power & Light Company's (IPL)
Petersburg Unit 3 can burn higher sulfur
coal when using DBA in the scrubber,
while maintaining its current 30-day av-
erage S02 control. However, long-term
effects of factors that may limit scrubber
performance (e.g., demister pluggage
or dewatering equipment performance)
have yet to be determined. Other effects
of DBA addition (e.g., increased corro-
sion rates) also must be considered in
long-term use of DBA. The use of DBA
at Petersburg Unit 3 did not reduce
limestone consumption. The use of
DBA on an annual basis or on an "as-
needed basis" may result in increased
scrubber reliability in that scrubber S02
-------
emission excursions may be minimized.
In addition, operating the scrubber with
DBA, lower booster fan inlet vane set-
tings, and continued use of presently
available coal should increase scrubber
reliability by minimizing wet gas recir-
culation to the booster fans.
Operating Results of Toyama
Kyodo Electric Power's Chiy-
oda Thoroughbred 121 Flue
Gas DesuKurization System
K. Wataya and A. Hori, Toyama
Kyodo Electric Power Com-
pany, Yokohoma, Japan;
N. Hashimoto and H. Koshi-
zuka, Chiyoda Chemical Engi-
neering and Construction Com-
pany, Ltd., Yokohama, Japan;
and D. D. Clasen, Chiyoda In-
ternational Corporation, Seat-
tle, WA
This paper reviews the design and ini-
tial operating experience of two Chiy-
oda Thoroughbred 121 FGD plants at
Toyama Kyodo Electric Power Com-
pany's Toyama Shinko Kyodo power
station to treat flue gas from two
200 MW coal-fired boilers. This is the
first application of the second genera-
tion CT-121 process to large electric-
utility coal-fired boilers. Each CT-121
unit consists of a single scrubber, and
limestone slurry preparation and gyp-
sum dewatering equipment. Each unit
also has a Ljungstrom type gas-gas heat
exchanger for reheating scrubbed flue
gas. Plant operation, since startup in
July 1984 (Unit 1) and August 1984
(Unit 2), has been smooth and trouble
free. The plants are operated at a SO2
removal efficiency of 85%. Limestone
utilization is greater than 99%, and the
dry gypsum by-product is sold to ce-
ment and wallboard manufacturers.
Plant reliability, including shakedown
operations, has been superb: 99.9% for
Unit 1 and 100% for Unit 2. Inspection of
the system following shakedown re-
vealed no 'scaling, plugging, or corro-
sion. Maintenance has been limited to
routine and general servicing of
equipment.
Session 4, Part I: FGD Reliabil-
ity Improvement
The Sulfur Dioxide Removal
System Improvement Program
at Seven Units of Louisville
Gas and Electric Company
M. L Mclnnis, Louisville Gas &
Electric Company, Louisville,
KY; D. A. Froelich, S. A.
Bjorklun, and G. M. Graves,
Burns & McDonnell Engineer-
ing Company, Kansas City, MO
During early 1984, Louisville Gas &
Electric Company (LG&E) implemented
a program to upgrade its mid-1970
vintage SO2 removal systems (SDRS) to
current design standards. The goal of
the project was to improve SDRS reli-
ability to a minimum levet of 90% while
reducing excessive operating and
maintenance expenses. This program is
at its midpoint, with design and con-
struction proceeding at the Mill Creek
Station. Problem identification and en-
gineering recommendations are in the
evaluation stage for two units at Cane
Run. Design is proceeding for a third
Cane Run unit. This paper summarizes
the project and highlights the unique
problems encountered with the SDRS
improvement program.
Improving the Reliability of In-
dianapolis Power & Light Com-
pany's Petersburg 3 Limestone
FGD System
J. David Colley, Radian Corpo-
ration, Austin, TX; Steve Wol-
siffer, Indianapolis Power &
Light Company, Petersburg, IN;
Dorothy Stewart and Richard
Rhudy, Electric Power Research
Institute, Palo Alto, CA; and
David Balfour, Radian Corpora-
tion, Austin, TX
Preliminary results of a comprehen-
sive program to improve the reliability
of Indianapolis Power and Light Com-
pany's Petersburg 3 FGD system are
presented. The objective was to identify
the causes of the process problems and
recommended solutions to the system's
reliability problems. FGD operating
guidelines developed from the informa-
tion gathered during on-site testing are
given. Severe mist eliminator scaling
was investigated: recommended proce-
dures were to select better control in-
strumentation for the scrubbers, to de-
termine proper operating setpoints for
pH and density, and to modify the mist
eliminator wash system. Other testing
and modifications to the system are to:
(1) install and document operation of
three different mechanical seals for re-
cycle pumps, (2) modify the limestone
grinding circuit and to install a particle-
size monitor to determine if the monitor
can indicate circuit maintenance fre-
quency, (3) optimize adipic acid use as
other process changes are made, and
(4) study the problem of stack opacity as
process modifications are imple-
mented.
Results of an FGD Process
Troubleshooting Program at
Texas Utilities, Martin Lake
Steam Electric Station
David Colley, Radian Corpora-
tion, Austin, TX; Don Mzyk and
Mike Wadlington, Texas Utili-
ties Generating Co., Tatum, TX;
and Dorothy Stewart, Electric
Power Research Institute, Palo
Alto, CA
The objectives of the troubleshooting
program at the Martin Lake Station
were to reduce the operating and
maintenance costs and to improve reli-
ability of the limestone FGD system.
The problem areas addressed include
chemical scaling in the absorber pack-
ing and mist eliminators, process con-
trol and instrumentation, system water
balance, and limestone utilization. The
troubleshooting methodology used to
solve the problems is outlined. A 3-
month demonstration test was under-
taken to verify the success of recom-
mended process changes to reduce
gypsum scale formation rates and to
improve process stability. The results of
the demonstration testing showed that
scale formation in the absorber packing
and mist eliminator can be virtually
eliminated. Recommendations to con-
vert the FGD system from a water pro-
ducer to a water consumer are listed.
Also, recommendations to increase
limestone utilization are presented.
Availability Analyses for Flue
Gas Desulfurization Improve-
ment Decisions
Albert R. Cunningham,
Philadelphia Electric Company,
Philadelphia, PA; and Morton
J. Smith, Pickard, Lowe, and
Garrick, Inc., Newport Beach,
CA
A recent study was sponsored by
EPRI to determine the usefulness of
availability analysis techniques for
-------
guiding decisions involving improve-
ments on FGD systems. The study con-
cluded that availability analyses can
provide significant insight into the con-
sequences of alternative decisions
faced by responsible FGD plant person-
nel. These decisions can relate to mat-
ters of FGD design, operation, and/or
maintenance, and can affect various
levels of the system hierarchy from the
overall FGD system to individual com-
ponents. In addition, the results indi-
cated that these analyses can provide
plant personnel with additional value at
the process variable level if the
availability analysis can be combined
with FGD performance models.
Session 4, Part II: Chemistry/
Reagent Preparation
A Study of Sulfur-Nitrogen
Compounds in Wet Lime/Lime-
stone FGD Systems
James B. Jarvis and Peter A.
Nassos, Radian Corporation,
Austin, TX; and Dorothy A.
Stewart, Electric Power Re-
search Institute, Palo Alto, CA
This paper summarizes the results of
an EPRI-sponsored program to study
the relationship between sulfur-
nitrogen (S-N) compounds and the wet
lime/limestone FGD process. S-N com-
pounds have been identified in the Ara-
pahoe pilot-scale FGD system and are
formed through a liquid-phase reaction
between S02 and NOX absorbed from
boiler flue gas. The program objectives
were to identify the factors which pro-
mote S-N compound formation in wet
FGD systems and determine if these
compounds change system perfor-
mance. A three-phase program was
conducted which included analytical
methods development, bench-scale
process simulation, and pilot testing at
EPRI's Arapahoe test facility. The test
results include the identification of vari-
ous S-N species in solution and a de-
scription of how FGD process condi-
tions affect NOX removal, S-N
formation, and the steady-state concen-
trations of S-N related species in FGD
systems. The test results indicate that
the accumulation of S-N compounds
can decrease both SOj removal effi-
ciency and operating pH. In most sys-
tems, however, the concentrations of
S-N compounds are expected to be low
enough so that the impact on SO2 re-
moval efficiency is not significant.
Thiosulfate as an Oxidation In-
hibitor in Flue Gas Desulfuriza-
tion Processes: A Review of
R&D Results
Gary T. Rochelle and David R.
Owens, University of Texas at
Austin, Austin, TX; John C. S.
Chang, Acurex Corporation, Re-
search Triangle Park, NC; and
Theodore G. Brna, U.S. EPA,
Air and Energy Engineering Re-
search Laboratory, Research
Triangle Park, NC
Sodium thiosulfate (Na2S203) has
been tested in a pilot plant as an oxida-
tion inhibitor in FGD by lime and lime-
stone slurry scrubbing with and without
MgO and adipic acid additives. The ef-
fectiveness of thiosulfate is propor-
tional to the inhibitor product, defined
as the product of thiosulfate concentra-
tion (M), calcium concentration (M), and
the moles of S02 absorbed per hour per
liter of hold tank volume. Gypsum satu-
ration was less than 100%, and scaling
was eliminated when the inhibitor
product exceeded 0.3 x 10~6 M3/hr.
Thiosulfate was relatively more effec-
tive in systems with chlorides and less
effective in systems promoted by MgO.
An inhibitor product greater than 10~6
M3/hr significantly enhanced dewater-
ing of solids from limestone scrubbing.
S02 removal and/or limestone utiliza-
tion were increased in systems that
started with less than 10 mM dissolved
calcium.
Lime Slaking Using Stirred
Mills
N. N. Dharmarajan and R. D.
Forbus, Central and South
West Services, Inc., Dallas, TX
The large tonnages of reagent con-
sumed in lime-based FGD processes re-
quire that the lime reagent be slaked at
the point of use for maximizing econ-
omy of operation. Analysis of commer-
cially available slaking equipment indi-
cated design, operating, and
maintenance limitations for our spray
dryer FGD project. This necessitated our
investigation of new methods for slak-
ing lime. The vertical stirred mill ap-
peared to offer several advantages that
are desirable. Some of the perceived ad-
vantages were simplicity of design,
ease of construction, reduced installed
cost, ease of operation and mainte-
nance, and reduced energy costs. This
paper presents the equipment features
and performance results of a pilot eval-
uation of a vertical stirred mill in lime
slaking service. The many desirable fea-
tures of the vertical stirred mill are high-
lighted to support its selection for Co-
leto Creek Unit 2 dry FGD lime-based
process. This is the first known use in
the U.S. of lime slaking using a stirred
mill for a commercial FGD system.
Limestone Selection and
Preparation for FGD
Dorothy Stewart, Electric Power
Research Institute, Palo Alto,
CA; and J. David Colley, 0. W.
(Buddy) Hargrove, and A. J.
Jones, Radian Corporation,
Austin, TX
Limestone properties and preparation
for conventional wet scrubbing FGD
systems have been investigated in the
laboratory and pilot plant with some
full-scale tests for comparison. Major
factors affecting limestone performance
are the pH of the scrubber liquor and the
particle size distribution of the lime-
stone. Utilization of the reagent is above
90% when the stone is ground to >90%
passing through 325 mesh. To accom-
plish this grinding, the wet ball mill and
tower mill have the greatest efficiency
and lowest system cost. Because of the
decreased requirement for -325 mesh
limestone, a system to grind to 90%
-325 mesh is less costly then a system
producing only 80% -200 mesh. Lime-
stone tests at the Duck Creek station of
Central Illinois Light Co. show that two
of the stones tested are less expensive
to use than current supply and can
maintain desired S02 removal. Labora-
tory evaluation of these limestones did
not completely define their behavior at
full scale.
Session 5: Materials of Con-
struction
Failure Analysis of FGD System
Components
Gary D. Jones, Radian Corpora-
tion, Research Triangle Park,
NC; Peter F. Ellis II and Dennis
M. Anliker, Radian Corporation,
Austin, TX; and Dorothy A.
Stewart, Electric Power Re-
search Institute, Palo Alto, CA
As part of a program sponsored by
EPRI, Radian has analyzed equipment
-------
failures at several FGD system installa-
tions. Both metallic and non-metallic
material failures were encountered.
Failures of non-metallic components in-
clude a variety of organic coatings ap-
plied to vessel walls and rubber linings
applied to vessel walls and pump im-
pellers. The metallic failures include air
sparger anchors, nozzles, a limestone
storage tank, and an agitator shaft. Al-
though some failures were associated
with installation, most were attributed
to inappropriate materials or improper
process operation. For each failure, the
most probable cause was determined
and recommendations were made to al-
leviate future occurrences. Several of
the analyses and solutions are applica-
ble to other systems where similar fail-
ures have occurred. The results illus-
trate how failure analysis of chronic
problems can reduce maintenance re-
quirements of the FGD system.
Performance of Dust and Stack
Materials in Wet FGD Systems
H. S. Rosenberg, G. H. Koch,
C. W. Kistler, Jr., and J. A.
Beavers, Battelle Columbus
Laboratories, Columbus, OH;
M. L. Meadows, Black & Veatch
Engineers-Architects, Kansas
City, KS; and D. A. Stewart,
Electric Power Research Insti-
tute, Palo Alto, CA
Previous field surveys of construction
materials for wet FGD systems on utility
boilers found that the major problem
areas are outlet ducts and stacks. These
components are critical in that a failure
may result in complete loss of generat-
ing capacity for lengthy periods. As part
of an EPRI-sponsored study, Battelle an-
alyzed the field performance of two out-
let duct linings and three stack materi-
als. The duct linings included a
glass-flake-reinforced polyester at San
Juan and an hydraulically bonded con-
crete at Gibson. The stack materials in-
cluded a mica-flake-reinforced
polyester lining at Duck Creek, an Alloy
625 flue at Bruce Mansfield, and an acid-
resistant brick and mortar flue at
Phillips. The last two materials did not
represent actual failures. Field visits
were made to each site by specialists to:
(1) obtain background information,
(2) inspect the duct or stack, and
(3) collect samples of the materials for
detailed physical and chemical analy-
ses. Results of the analyses were used
to evaluate the performance of the ma-
terials. Performance was found to de-
pend on the environmental conditions
within the duct or stack. The most
severe condition seems to occur when
wet scrubbed gas is mixed with hot by-
passed gas. Characterization of the en-
vironmental conditions is important in
all cases in order to identify potential
materials problems and solve them.
Leaning Crick Chimney Liners
at Coal Fired Plants
E. R. Dille and D. L. Krueger,
Burns & McDonnell Engineer-
ing Company, Kansas City,
MO; and R. G. Rhudy, Electric
Power Research Institute, Palo
Alto, CA
Leaning brick liners in stacks down-
stream of wet FGD systems are a seri-
ous problem that has recently surfaced
in the utility industry. While available
information suggests that so far the
problem affects only a few stacks, the
modifications necessary to repair or re-
place an existing stack liner represent a
major investment. To date, little infor-
mation has been available to determine
the nature and extent of the problem.
Accordingly, EPRI has initiated a project
to evaluate the severity of the problem,
identify the cause or causes, and de-
velop preventive measures. This paper
documents currently known informa-
tion on leaning stacks, including what is
known about the magnitude and cause
of the problem.
The Economics of Clad Metallic
Fabrication Versus Solid Alloy
Metal Fabrication
E. H. Cloth and R. Kreider,
Stone & Webster Engineering
Corporation, Cherry Hill, NJ
The purpose of this analysis of FGD
materials is to provide economic guid-
ance for selecting a clad metallic fabri-
cation system versus a solid metal fabri-
cation. A simplified FGD tower shell was
designed to provide a common basis for
the costs. The material and fabrication
costs of solid alloys were compared
with clad metal fabrications (i.e., roll-
bonded, alloy lining) to determine the
potential economical savings. Solid
alloy materials costs spanned the $2-1M
Ib range, which included 316L, an
austenitic stainless steel, to C-276, a
nickel base alloy. The costs included
plate or sheet material, weld filler mate-
rial, and welding man-hour costs. The
investigation indicates possible savings
for clad alloy fabrications over solid
alloy fabrications when the solid alloy
cost is in the $2-3/lb range with savings
approaching the potential of 30-60% in
the higher-priced alloys such as 625 or
C-276. The welding of thin alloy sheets,
1/16- and 1/8-in. thick, on a carbon steel
shell can also be commercially advanta-
geous in the higher-priced alloys.
Laboratory Evaluation as a
Technique for Predicting the
Behavior of Organic Coatings
in FGD Systems
Malcolm L. White and Henry
Leidheiser, Jr., Lehigh Univer-
sity, Bethlehem, PA; and G. H.
Koch, Battelle-Columbus Labo-
ratories, Columbus, OH
A laboratory evaluation of several
neat resin systems commonly used as
the basis for formulated organic coat-
ings in FGD systems was carried out by
exposure of coated carbon steel sub-
strates to dilute sulfuric acid at 60°C
(140°F) for as long as 3000 hours. Mea-
surements of corrosion potential, ac
conductance, tensile adhesion, and
weight gain revealed distinct differ-
ences in the behavior of the resin types
in protecting the steel against corro-
sion. These results are compared with
those obtained in an 18-month field
study of formulated coatings (using the
same resins as a base material) exposed
to the extremely aggressive environ-
ment of the outlet mixing zone of the
R.D. Morrow, Sr., FGD system. The cor-
relations of resin/coating performance
in the actual and simulated FGD envi-
ronments were investigated with re-
spect to both corrosion protection of the
underlying steel and in the modes of
failure for those materials that showed
poor performance.
Volume 2
Session 6: Panel Discussion on
Retrofitting FGD Systems
TVA's SO2 Control Program
Gerald A. Hollinden, Tennessee
Valley Authority, Chattanooga,
TN
In the past few years, TVA has imple-
mented various S02 emission control
procedures on its coal-fired plants; e.g.,
-------
coal cleaning, limestone-scrubbing
FGD, and switching to low-sulfur coals.
These emission reduction strategies, in
combination with conservation and re-
duced power demand, improved heat
rates, and additional nuclear genera-
tion, have greatly reduced the quantity
of S02 emitted by TVA: the 1983 S02
emissions are less than half the 1977
S02 emissions. Although great prog-
ress has been made, potential changes
in S02 emission regulations to control
acid rain could have profound effects on
the use of FGD by TVA and other east-
ern utilities; e.g., increased use of FGD,
upgrading of existing FGD systems, and
the accelerated adoption of new tech-
nology. With a total FGD capacity of
2500 MW already installed and test facil-
ities for advanced FGD and combustion
technology (AFBC) planned, TVA is con-
tinuing to assess controlling S02 emis-
sions in the future. Even though lime-
stone scrubbing is expected to be the
primary method, at least through the
year 1999, innovations, such as the use
of additives (adipic and dibasic acid),
and new processes, such as limestone
double alkali and dry FGD (spray dryer),
should increase in importance and
move into commercial use in the near
future, particularly if the need for wider
use of FGD occurs. AFBC, in which SO2
removal is an integral function, is reach-
ing an advanced stage of development
and could reduce the use of FGD at new
facilities by the latter half of the 1990's.
Randall E. Rush, Southern
Company Services, Birming-
ham, AL
No abstract furnished.
A/E Perspective of Retrofitting
FGD Systems
Paul A. Ireland, Stearns Cata-
lytic Corporation, Denver, CO
From our perspective as engineers,
we feel it is important to recognize the
true cost of retrofitting an FGD system
to an existing plant. We also feel that
many acid rain related studies may be
underestimating these costs. We also
share utilities' concerns of minimizing
outage time and maintaining existing
plant services while the retrofit con-
struction is underway. Often retrofit
items can be overlooked; e.g., having to
take special precautions when dealing
with existing asbestos insulation or
having to hand-dig foundations be-
cause of underground obstructions. Of-
ten, FGD process selection or design
will be affected by the space available.
Finally, if an FGD system retrofit is re-
quired for acid rain compliance, an op-
portunity may exist to minimize the
overall cost by burning a cheaper higher
sulfur coal.
The Changing FGD Market-
place
James R. Martin, Combustion
Engineering Inc., Birmingham,
AL
The demand in the U.S. for FGD sys-
tems at the beginning of this decade
was concentrated primarily in the utility
sector and consisted primarily of new
systems in conjunction with utility ca-
pacity additions. During the last 5 years,
we have seen that demand shrink at an
average rate of 50% per year to a very
low level. During the next 5 years, how-
ever, we can expect an increase in de-
mand for FGD systems. In part, this new
demand will be caused by utility capac-
ity additions; but an increasing demand
for emission control systems for refuse
recovery, cogeneration, and hazardous
waste incineration will have a major im-
pact, and FGD system rehabilitation and
modifications will also be important.
This presentation outlines the forces
that are reshaping the demand for
gaseous emission control equipment
and how we can expect the technology
to adapt to that changing demand.
Retrofitting FGD Systems
Howard Feibus, U.S. Depart-
ment of Energy, Washington,
DC
If legislation is enacted requiring con-
trol of SO2 emissions from pre-NSPS
coal-fired power plants in order to re-
duce acid rain precursor emissions,
commercially available technology is
essentially limited to coal cleaning, wet
scrubbers, and spray dryers. DOE analy-
ses indicate that commercial coal clean-
ing is of importance but limited value
and that the last two would be pro-
hibitively expensive. Developmental
technologies that could be available in
the late 1980s to mid-1990s offer the po-
tential of improved performance in
terms of cost per unit weight of S02.
These approaches include advanced
coal cleaning processes, furnace sor-
bent injection, fluidized-bed combus-
tion, and duct injection FGD.
Jack F. Stewart, Babcock &
Wilcox Co., Barberton, OH
No abstract furnished.
Retrofit Technology Options
for Reducing Sulfur Emissions
Stephen M. Katzberger, Sar-
gent & Lundy Engineers,
Chicago, IL
A wide variety of technologies are
available for reducing sulfur emissions
from existing industrial and utility boil-
ers. These technologies may reduce or
capture sulfur: (1) prior to combustion
by fuel switching, fuel blending, coal
gasification, mechanical coal washing,
or chemical coal cleaning; (2) during
combustion by furnace sorbent injec-
tion or fluidized-bed boilers; or (3) after
combustion by downstream sorbent in-
jection, wet FGD, or spray drying. Each
of these technologies has site-specific
advantages and disadvantages in terms
of level of development, cost, retrofit-
ability and sulfur reduction potential.
Even fuel switching will require equip-
ment modifications. Switching from
high sulfur Illinois coal to low sulfur
eastern Kentucky coal, for example,
may require only minor modifications
to ensure safe operation. However,
switching to a low sulfur Powder River
Basin coal can seriously impact coal
handling, boiler performance, gas vol-
ume which affects both the fans and
ESP, ash resistivity, and ash handling.
Unit outage time should be considered
in evaluating retrofit options. Since a
wet FGD system can be erected with the
unit on line, only a 2-3 month outage
would be needed to tie into existing
ducts. A fluidized-bed boiler retrofit
could require a unit outage of 11/2 to
2 years to modify most of the boiler.
The sensitivity of the total evaluated
cost of each alternative to the remaining
life of the unit may show that high capi-
tal cost technologies, such as wet FGD,
are not economically viable for older
units without unit life extension. Low
sulfur coal conversion, a lower capital
cost/higher operating cost option may
be selected for units with limited re-
maining lives or low capacity factors.
Furnace sorbent injection might be se-
lected for certain boiler designs and
lower sulfur reduction requirements.
Similarly, fluidized-bed retrofits may be
-------
favored for units under 200 MW. In
other cases, unit retirement may be the
best choice.
Issues in Utility SO2 Control
Retrofit
Stuart M. Dalton, Electric
Power Research Institute, Palo
Alto, CA
Cost, space, risk, energy, and waste
disposal are the main issues facing util-
ities that are considering a retrofit FGD.
Most debates have centered on FGD
cost, and have normalized the cost in
terms of dollars per ton of S02 control.
Early estimates used in many political
debates may have understated costs.
EPRI estimates range from $560 to
$1120/ton S02 removed for conven-
tional wet FGD and $425 to $935 for ad-
vanced concepts such as furnace sor-
bent injection and Chiyoda 121
scrubbing (levelized over 30 years).
Several factors cause costs to vary
widely. Space constraints can cause
costs to increase by over 60% and will
preclude FGD on a few installations.
Utilities often are unwilling to accept
high risk technologies since they cannot
shut down if the process doesn't work.
Electric energy used for FGD (1-3% of
the plant output) is unavailable to meet
the peak loads and must be replaced by
utilities at premium replacement rates.
Waste disposal quantities for midwest-
ern plants can double or even triple in
volume compared to untreated plants.
These issues make it prudent for utilities
to study the alternatives available to
them very carefully before committing
to a single strategy for their systems.
Session 7: Dual Alkali
Development and Demonstra-
tion of a Limestone Dual Alkali
FGD Process at Central Illinois
Public Service Company's
Newton Power Station
Jerry L. Simpson, Central Illi-
nois Public Service Company,
Springfield, IL; and James H.
Wilhelm, Codan Associates,
Sandy, UT
The paper describes the develop-
ment, implementation, and demonstra-
tion of the CIPS/CODAN limestone dual-
alkali process at the 595 MW Newton
Power Station of CIPS. The paper de-
scribes the process, the changes re-
quired to convert the original lime dual-
alkali process to the use of limestone,
operating cost savings, performance
goals, and achievements.
Current Operating Experience
and Operating Costs of Double
Alkali FGD at NIPSCO's R.M.
Schahfer Generating Station
Unit 17
R. J. Biolchini, W. L. Boward,
and B. S. Camponeschi, FMC
Corporation, Schaumburg, IL
This paper presents operating experi-
ence and 1984 operating costs of FMC's
double-alkali FGD system, for the 393
MW, high-sulfur coal-fired boiler at
NIPSCO's R.M. Schahfer Generating
Station. Test results have shown that
the FGD system has met the EPA's 1979^
Revised New Source Performance
Standards on high sulfur coal and has
demonstrated collection efficiencies
well above 90%. Average availability
during the first 2 years of operation has
been 99.97%. FGD system operating
costs for 1984 were on the order of
4 mills/kWh.
Current Progress at the FMC
Limestone Double Alkali
Demonstration Plant
Willard L. Boward, Jr., and
Robert J. Biolchini, FMC Corpo-
ration, Schaumburg, IL; and
Beth A. Wrobel, Northern Indi-
ana Public Service Company,
Hammond, IN
The Limestone Double Alkali (LSDA)
FGD process combines the reliability
and efficiency of sodium scrubbing with
the economical use of limestone as the
main reagent. Studies sponsored by
EPA and EPRI have shown LSDA to be
economically desirable for full scale
plants when evaluated against other
types of wet scrubbers. An LSDA
demonstration plant has been in opera-
tion since September 1984, at Northern
Indiana Public Service Company's R.M.
Schahfer Station. The demonstration
plant treats a flue gas slipstream from
Schahfer Unit 17 which burns high sul-
fur coal. Results of the first test period is
discussed. Period 1 was used to charac-
terize the operation of the system and
determine the preferred process set-
tings for the next test. Period 2 will be a
long term test of the dependability of
the system while maintaining 90% SO2
removal and following changes in
boiler operation. Plans for Period 3,
which will determine how well the sys-
tem responds to stresses, are also dis-
cussed. Results in the areas of SO2 col-
lection, limestone utilization, soda ash
consumption, and filter cake percent
solids have been encouraging.
Effects of Limestone Type and
Grind on Dual Alkali System
Performance
John C. S. Chang, Acurex Cor-
poration, Research Triangle
Park, NC; and Theodore G.
Brna and Norman Kaplan, U.S.
EPA, Air and Energy Engineer-
ing Research Laboratory, Re-
search Triangle Park, NC
Pilot plant tests were conducted to
evaluate the dual-alkali system per-
formance with respect to five types and
grinds of limestones. The data show
that high purity, high calcium, fine grind
limestone should be used for the regen-
eration of dual-alkali scrubbing solu-
tion. Limestone containing more than
0.5% magnesium is not suitable for the
dual-alkali process. The accumulation
of dissolved magnesium caused per-
formance deterioration and system
upset. Experimental techniques which
can be used to characterize the lime-
stone reactivities were also evaluated.
The reactivities measured by sodium
bisulfite titration agreed with pilot plant
data.
Session 8: Emerging Technolo-
gies
Pilot Evaluation of Combined
Particulate and SO2 Removal
Using a Fabric Filter System
Richard G. Hooper, Electric
Power Research Institute, Palo
Alto, CA; Verle Bland, KVB,
Denver, CO; Franz G. Pohl,
SoRI, Birmingham, AL; and
Michael McElroy and Richard
Rhudy, Electric Power Research
Institute, Palo Alto, CA
The process of combining S02 and
paniculate control using dry reagents is
gaining increased utility interest. This
a
-------
paper reviews and updates events rele-
vant to the injection of both sodium and
calcium reagents into a flue-gas stream
ahead of a fabric filter. Results are re-
ported from tests conducted at EPRI's
Fabric Filter Test Facility at the Arapa-
hoe station of the Public Service Co. of
Colorado in Denver, CO, and at EPRI's
High-Sulfur Coal Pilot Plant at the
Scholz station of the Gulf Power Com-
pany near Sneads, FL.
With dry-sodium injection, sodium
reagent is fed into the flue-gas stream
ahead of a baghouse and downstream
of the air heater. In the ductwork,
sodium carbonate partially reacts with
S02 in the flue gas to form a dry powder
that collects along with the fly ash on
the bags as part of the dust cake where
it continues to remove S02. This paper
gives results for three sodium re-
agents—sodium bicarbonate, sodium
sesquicarbonate, and sodium carbon-
ate. For each reagent, SO2 removal is
determined as a function of reagent par-
ticle size and normalized stoichiometric
ratio (NSR), and sodium utilizations are
compared.
A dry calcium reagent for SO2 re-
moval is injected in the same manner as
sodium, except that water or steam is
also injected to condition the flue gas to
a desired temperature and humidity. In
the duct-work, a portion of the lime re-
acts with S02 to form calcium sulfite
and calcium sulfate. This paper:
(1) reports on preliminary tests with
pressure-hydrated dolomitic lime, and
(2) compares calcium utilization and
S02 removal for five calcium reagents—
quicklime, pressure-hydrated high-
calcium lime, conventionally hydrated
lime, pressure-hydrated dolomitic lime,
and pre-calcined pressure-hydrated
dolomitic lime.
Investigation of Combined Par-
ticulate and SO2 Using E-SOX
Leslie E. Sparks, Norman Plaks,
Geddes H. Ramsey, and
Richard E. Valentine, U.S. EPA,
Air and Energy Engineering Re-
search Laboratory, Research
Triangle Park, NC
Research aimed at developing a low
cost retrofit system (called E-SOX, for
combined paniculate and S02 removal)
is described. The E-SOX concept cen-
ters on recent advances in ESP technol-
ogy that make it possible to reduce the
size of ESPs required for paniculate
control. Results of mathematical model-
ing and limited pilot scale experiments
on the feasibility of collecting S02 in the
freed space are discussed. The results
show that the concept is technically fea-
sible. Modest (40-65%) levels of S02 re-
moval are possible with lime as the
reagent. High (over 90%) levels of S02
removal are possible with sodium car-
bonate as the reagent. Order of magni-
tude economic analysis shows that the
process is economically feasible. Plans
for additional experimental and theoret-
ical work are presented.
Mitsui-BF Simultaneous SOX
and NOX Removal System
Yoshiro Ito, T. Fujimoto, and
0. Nagaoka, Mitsui Mining
Company, Ltd., Tokyo, Japan
Mitsui Mining Company, Ltd. (MMC),
has been developing the Mitsui-BF sys-
tem jointly with Mitsui Miike Engineer-
ing Corporation (MMEC) by using acti-
vated coke to improve a process
originated by Bergbau Forschung
GmbH (BF) of West Germany. A pilot
plant with a capacity of 1000 Nm3/h was
operated from February 1981 to Sep-
tember 1983 at Tochigi Works of MMEC.
Successful long-term continuous opera-
tion of the pilot plant was achieved at an
efficiency of more than 99% SOX re-
moval and more than 80% NOX re-
moval. Based on the results of the pilot
plant test performance, MMC gained
confidence to apply the new technology
to a commercial plant. MMC con-
structed a plant of 30,000 Nm3/h capac-
ity at the MMC Coal-fired Power Station
in Omuta, Japan. Its commercial opera-
tion began on October 1, 1984 and the
plant has been operated satisfactorily
since then. MMC has been operating an
activated coke (AC) production pilot
plant with a capacity of 0.8 ton AC/day
in an attempt to produce less expensive
AC with better performance. This AC
was used in the Mitsui-BF plant at
Omuta and has shown very satisfactory
results. After performance test opera-
tion based on a design condition of 99%
DeSOx and 50% DeNOx was achieved,
higher NOX removal efficiency opera-
tion (70-80%) began in February 1985.
Operating Experiences and De-
velopments with Ljungstroem
Gas-gas Heaters in West Ger-
man FGD Plants
David E. Clay, Kraftanlagen
Aktiengesellschaft Heidelberg,
Heidelberg, Federal Republic of
Germany
Regenerative gas-gas heaters have
been adopted by the German power in-
dustry as the state-of-the-art for reheat-
ing scrubbed flue gases. Anticipated
fouling and corrosion have proved con-
trollable, and the first large-scale gas-
gas heater entered operation in 1982 at
NWK Wilhelmshaven. Since then more
than 11,000 operating hours have been
achieved. Two novel features are incor-
porated in the design—a cold-side plas-
tic heating element and an on-line high-
pressure washing system. Anticipated
future S02 emission limits of 200 mg/
Mm3 have led to a low leakage design
(0.3-0.5%). The operating experiences
and developments with the regenera-
tive gas-gas heater are discussed.
SOXAL" Process
K. N. Man! and F. P. Chlanda,
Allied Corporation, Mt. Bethel,
NJ
The SOXAL process is a regenerative
wet scrubbing system that produces a
concentrated S02 stream and is applica-
ble for low to high sulfur coals. The S02
can be liquefied and converted to sulfu-
ric acid or elemental sulfur. The process
is based on the use of a high pH sodium
solution as a scrubbing medium to re-
move the SOX from the flue gas. The
bisulfite salt in the product solution
from the absorber is converted to the
original sulfite and hydroxide (that are
recycled) and an aqueous solution of
S02 (that is readily stripped to recover
concentrated SO2). The heart of the
process is the "Membrane Water Split-
ter" (AQUATECH cell) that electrodialyt-
ically converts salt solutions to their
constituent acids and bases at low en-
ergy consumption. Key to the operation
of the AQUATECH cell is the "bipolar
membrane" developed by Allied Corpo-
ration. The SOXAL process and the as-
sociated membrane technology have
been under development at Allied Cor-
poration for over 12 years and are cur-
rently being commercialized by its
AQUATECH Systems business unit.
-------
Session 9: Spray Dryer FGD
Spray Dryer FGD Experience:
Joy-Niro Installations
J. R. Donnelly, A/S Niro Atom-
izer, Soeborg, Denmark;
S. Wilson, Joy Manufacturing
Company, Los Angeles, CA;
L. P. Matis, Northern States
Power Company, Minneapolis,
MN; R. Eriksen, Basin Electric
Power Cooperative, Bismarck,
ND; R. D. Emerson, Sunflower
Electric Cooperative, Hays, KS;
and J. C. Fooks, Platte River
Power Authority, Ft. Collins, CO
Joy-Niro lime-based Spray Dryer Ab-
sorption (SDA) FGD systems are in com-
mercial operation on the following coal-
fired utility boilers: (1) Riverside
Generating Station—Units 6 and 7;
(2) Antelope Valley Station—Unit 1;
(3) Holcomb Generating Station—
Unit 1; and (4) Rawhide Generating Sta-
tion—Unit 1. More than 9 years of oper-
ating experience have been accu-
mulated by these stations. These
systems employ spray dryer absorbers,
equipped with single rotary atomizers,
and reverse air bagfilters for SO2 and
paniculate removal. The systems range
in size from 0.6 to 2.1 million acfm (110
to 450 MW) and treat flue gases from
the firing of North Dakota lignite, Pow-
der River Basin subbituminous coal,
and a blend of subbituminous coal and
petroleum coke. Each system incorpo-
rates unique site specific design fea-
tures for the flue gas cleaning systems
and for waste conditioning/disposal.
This paper compares the plant designs
and performances. Compliance/per-
formance test results are compared to
design values. Start-up problems are re-
viewed, and current plant reliability is
discussed. Waste conditioning and dis-
posal practices at each site are de-
scribed, and results of physical/chemi-
cal tests of the end product are
presented.
Problem Solving and Results of
Performance Test at United
Power Association Stanton
Station Dry FGD System
Ghassem B. Manavizadeh and
Stefan Negrea, Research-
Cottrell, Inc., Somerville, NJ;
Gordon Westerlind, United
Power Association, Elk River,
MN; and Alan W. Ferguson,
Black & Veatch Engineers-
Architects, Kansas City, MO
Use of data acquisition and manage-
ment techniques of solving problems
associated with spray driers and fabric
filters is discussed. Methods of cooper-
ation between the owner, A/E, and the
system supplier are highlighted. Results
of the first FGD performance test, a
60 day reliability test, its inconclusive
results and the improvements gener-
ated by its data, as well as the prelimi-
nary final performance test (1 year after
the initial test) are presented. The sys-
tem design, field modifications, and its
operation are described. Using meth-
ods specified by EPA for compliance
testing, guaranteed SO2 removal effi-
ciency and paniculate emission were
achieved at the UPA Stanton dry FGD
unit. Preliminary data are also pre-
sented on use of sonic horns for reduc-
ing pressure drop in fabric filters and
their effect on S02 removal and particu-
late emission.
High-Sulfur Spray Dryer
Evaluations
Russell F. Robards and Robert
W. Aldred, Tennessee Valley
Authority, Chattanooga, TN;
Thomas A. Burnett and Lynn R.
Humphries, Tennessee Valley
Authority, Muscle Shoals, AL;
and Michael J. Widico,
Research-Cottrell, Inc.,
Somerville, NJ
The TVA initiated a program in 1983
to evaluate a pilot lime spray dryer/bag-
house FGD system treating flue gas
from the combustion of high-sulfur coal
(3.2% S as-fired) at the Shawnee Test
Facility. Testing was performed to de-
termine the effects of the major spray
dryer operating variables on the spray
dryer, the baghouse, and the overall
S02 removal efficiencies of the system.
Variables which were evaluated include
lime stoichiometry, approach-to-
saturation temperature, recycled solids,
flue gas residence time in the spray
dryer, atomizer speed, and inlet flue gas
temperature. As expected, S02 removal
efficiency is a strong function of lime
stoichiometry and approach-to-
saturation temperature. Recycled solids
and the inlet flue gas temperature are
also important, but atomizer speed and
residence time seem to have little, if
any, impact over the ranges tested. The
most important conclusion is that a
spray dryer/baghouse FGD system can
attain a sufficiently High S02 removal
efficiency to meet a 1.2 Ib S02/106 Btu
emission regulation (currently imposed
at TVA's Shawnee Steam Plant) when
treating flue gas from the combustion of
high-sulfur coal. However, careful con-
sideration of the major independent
variables is necessary to attain this goal.
Recent Results from the EPRI
2-1/2 MW Spray Dryer Pilot
Plant
Richard G. Rhudy, Electric
Power Research Institute, Palo
Alto, CA; and Gary M. Blythe,
Radian Corporation, Austin, TX
Since January 1984, testing on the
EPRI 2-1/2 MW spray dryer/fabric filter
FGD pilot plant has centered on varia-
tions of the basic FGD system. Results
presented in this paper demonstrate the
effects of the use of several cooling
tower blowdown waters for lime slaking
and/or process dilution, the effects of
calcium chloride as a system perform-
ance additive, and the impacts on SO2
removal of the use of warm (spray dryer
inlet) gas bypass around the spray dryer
to reheat the fabric filter inlet gas by
20-40°F. Additionally, results are pre-
sented which relate process variations
to fabric filter performance; e.g., corro-
sion tendencies and tube-sheet pres-
sure drop. Data are presented which
compare the tube-sheet pressure drop'
history for bags in the fabric filter com-
partment downstream of the spray
dryer to that for bags in a control com-
partment that is filtering hot, untreated
flue gas. Finally, data are presented
which relate these process variations to
the disposal properties of solids col-
lected in the fabric filter.
JO
-------
Performance of a Spray Dryer/
ESP Flue Gas Cleanup System
During Testing at the Pitts-
burgh Energy Technology
Center
James T. Yeh, Charles J. Drum-
mond, and James I. Jouberg,
U.S. Department of Energy,
Pittsburgh, PA; and Dennis M.
Tulenko, Robert R. Crynack,
and Martin L. Hochhauser,
Wheelabrator Air Pollution
Control, Pittsburgh, PA
Wheelabrator Air Pollution Control
and the U.S. DOE jointly sponsored a
project to characterize the performance
of a spray dryer FGD system using flue
gas from the combustion of different
types of coal, and to study the effective-
ness of an electrostatic precipitator
(ESP) in controlling participate emis-'
sions from this system. Three-different
coals (low-, medium-, and high-sulfur,
coal) were burned in a 500 Ib/hr
pulverized-coal combustion test facility
at the Pittsburgh Energy Technology
Center. S02 was removed from the flue
gas by injection of a lime slurry in a
spray dryer. Paniculate emissions were
controlled using a mobile ESP. For each
fuel, a three-part test program was con-
ducted: (1) base-line ESP performance
tests without spray dryer operation to
determine requirements for fly ash col-
lection; (2) parametric tests defining
SO2 removal efficiency and ESP collec-
tion efficiency at various spray dryer op-
erating conditions; and (3) sorbent recy-
cle tests in which the particulate
collected by the ESP was recycled by
injection into the spray dryer with the
fresh sorbent to improve reagent utiliza-
tion. Test results show that S02 removal
efficiencies of 90% can be achieved for
each of the coals tested using the com--
bined spray dryer and ESP system for
emissions control. Performance criteria
for the ESP were met during all tests.
Application of Electrostatic En-
hancement of Fabric Filtration
to Spray Dryer Byproduct
L S. Hovis and Bobby E.
Daniel, U.S. EPA, Air and En-
ergy Engineering Research Lab-
oratory, Research Triangle
Park, NC; Yang-Jen Chen, Joy
Industrial Equipment Co., Los
Angeles, CA; and R. P. Dono-
van, Research Triangle Insti-
tute, Research Triangle Park,
NC
Experiments conducted at small pilot
scale show that the pressure drop in-
crease during the fabric filtration of re-
dispersed spray dryer by-product
(chiefly calcium salts and fly ash) is sig-
nificantly reduced through electrostatic
enhancement of the filtration. The pres-
sure drop rise for a typical electrostatic
augmented fabric filtration (ESFF) is
only 25% or less of that of the rise for a
conventional filtration cycle. The ESFF
takes advantage of the relatively lower
electrical resistivity of the spray dryer
by-product attributable to the high
moisture content and to the low temper-
ature at which the filtration of spray
dryer by-product is carried out. A low
resistivity allows application of greater
electrode potentials than can be used at
normal fly ash filtration conditions.
Comparison of Pilot Plant Data
with Predictions from the U.S.
EPA SPRAYMOD Computer
Program
Wayne T. Davis, Gregory D.
Reed, and George P. Partridge,
The University of Tennessee,
Knoxville, TN; and Theodore G.
Brna, U.S. EPA, Air and Energy
Engineering Research Labora-
tory, Research Triangle Park,
NC
The use of computer models allows
the user to predict conditions which
may not have been tested in spray
dryer/fabric filter applications. This
paper presents a preliminary compari-
son of the U.S. EPA-developed SPRAY-
MOD computer program with five dif-
ferent data bases collected at the
University of Tennessee on a 1000 cfm
spray dryer utilizing a Ca(OH)2 slurry.
Results are presented from a variety of
operating conditions for the spray dryer
without recycle, including:
S02 concentration: 1000-3100 ppm
Inlet temperature: 138-177°C (280-
350°F)
Approach to satu- 9-20°C (16-35°F)
ration (AT):
Conclusions regarding the ability of
the program to predict the actual behav-
ior of the pilot plant are: (1) with the
proper choice of a reaction rate coeffi-
cient, the plug flow model reliably pre-
dicted the performance for low S02 con-
centrations (1000-1100 ppm) and
approaches to saturation in the range of
16-35°F, and (2) the model tended to
overpredict performance for higher S02
concentrations (2000-3000 ppm).
Session 10: FGD By-Product
Disposal/Utilization
Gypsum By-Product FGD
System
W. A. Liegois and D. A. Wicks,
Stanley Consultants, Inc., Mus-
catine, IA
This paper summarizes the start-up of
the FGD system on Unit 9 of Muscatine
Power and Water. This FGD system has
a guaranteed removal efficiency of 96%
of the S02 from the flue gas and was the
first unit in the U.S. designed to produce
a gypsum by-product. Operating prob-
lems associated with the system are dis-
cussed, along with system availability
and by-product quality.
By-Product Gypsum Production
at a 2300 MW Power Plant
Don Mzyk, Texas Utilities Gen-
erating Co., Tatum, TX; and
Jan Zmuda, Research-Cottrell,
Inc., Somerville, NJ
The Martin Lake Station of Texas Util-
ities consists of three 770 MW boilers
firing lignite coal. Each boiler is
equipped with a Research-Cottrell lime-
stone FGD system. The FGD systems
were originally designed to produce a
sulfite sludge to be blended with fly ash
for disposal. In early 1983 each FGD sys-
tem was retrofitted with equipment for
in-situ oxidation to produce a calcium
sulfate (gypsum) by-product. This paper
discusses start-up and testing of the
forced oxidation system. Utility operat-
'ing experience, which led to the signing
of a contract for the sale of commercial
grade gypsum for wallboard manufac-
ture, is also discussed.
-------
Evaluation of Engineering
Properties and Wet Stacking
Disposal of Widows Creek FGD
Gypsum-Fly Ash Waste
John E. Garlanger, Ardaman &
Associates, Inc., Orlando, FL;
Sal H. Magliente, Tennessee
Valley Authority, Chattanooga,
TN; Thomas S. Ingra, Ardaman
& Associates, Inc., Orlando, FL;
and James L. Crowe, Tennes-
see Valley Authority, Chat-
tanooga, TN
Wet stacking of by-product gypsum
has been practiced by the phosphate
fertilizer industry for more than
25 years. The use of wet stacking for
disposal of FGD gypsum was first
demonstrated during an EPRI-
sponsored project on Chiyoda Thor-
oughbred 121 FGD gypsum produced at
the Scholz Electric Generating Station
of Gulf Power Company in Sneads, FL
Wet stacking of FGD gypsum containing
fly ash, however, has not been previ-
ously demonstrated. Accordingly, as
part of an overall project investigating
various FGD waste disposal alterna-
tives, the TVA constructed a pilot-scale
wet-stacking disposal facility to evalu-
ate the feasibility of wet stacking FGD
gypsum-fly ash waste produced at the
Widows Creek Steam Plant in Steven-
son, AL. Operational experience and re-
sults from geotechnical laboratory test-
ing performed on the waste are
presented. The results indicate that, al-
though the Widows Creek FGD gypsum-
fly ash had settling, dewatering, and
structural characteristics not as favor-
able for stacking as phosphogypsum or
CT 121 FGD gypsum, they were ade-
quate for wet stacking. Therefore, the
project findings should extend the abil-
ity of the utility industry to employ wet
stacking disposal to facilities that also
use FGD/forced oxidation systems as
the primary particulate removal
process.
Solids Handling and Dewater-
ing Optimization at Colorado
Ute's Craig Station
O. W. (Buddy) Hargrove and
David R. Owens, Radian Corpo-
ration, Austin, TX; Dorothy A.
Stewart, Electric Power Re-
search Institute, Palo Alto, CA;
and Jim Renner, Colorado Ute,
Montrose, CO
Units 1 and 2 at Colorado Ute's Craig
Station include wet limestone scrub-
bers for S02 emissions control. Because
the SO2 concentration in the gas is very
low, the solids formed are completely
oxidized. Normally, gypsum produced
in a limestone FGD system dewaters
and handles relatively well. However,
the centrifuge product at Craig fluidizes
when shear forces are applied and, as a
result, cannot be transported in an open
truck. EPRI, Radian, and Colorado Ute
have been investigating the poor han-
dling properties of these solids over the
past 3 years. The fluid properties of the
solids are a result of the fine crystals in
the product cake which may be caused
by a number of factors. This paper dis-
cusses possible causes of the cake prop-
erties and methods to optimize the per-
formance of the dewatering equipment.
A comparison of full-scale centrifuge
and pilot-scale vacuum filter results is
included.
EPRI Research on Management
of Wastes from SO2 Control
Processes
Dean M. Golden, Electric Power
Research Institute, Palo Alto,
CA; and Russell H. Boyd, Envi-
rosphere Company, Norcross,
GA
EPRI is keenly aware of the many is-
sues facing the electric utility industry in
the area of controlling flue gas emis-
sions from coal-fired power plants, par-
ticularly SOX. Virtually all facets of the
question of how to control SO2 emis-
sions have been or are under consider-
ation by EPRI in one or more research
programs. In closing the loop on the
question of S02 control technologies,
however, EPRI recognized that the man-
agement of solid wastes from the new
control technologies had not been ex-
amined in depth. This paper describes
the research EPRI is conducting with re-
gard to management of these wastes.
Preliminary results of the research proj-
ect are discussed. The research project
focuses on spray drying, atmospheric
fluidized-bed combustion, limestone
furnace addition, dry sodium addition,
and advanced physical coal cleaning.
Within the project are activities dealing
with information surveys and assess-
ments, waste characterization, leachate
testing, conceptual designs, case stud-
ies, and utilization options. The project
is comprehensive in scope and will be
providing topical reports on various ac-
tivities during the next 3 years.
Thermal Oxidation of Spray
Dryer FGD Waste Product
S. Bengtsson and S. Ahman,
Flakt Industri AB, Zaxjo, Swe-
den; T. Lillestolen, Flakt, Inc.,
Knoxville, TN; and G. Koudijs,
Dorr-Oliver B.V., Stamford, CT
Spray dryer FGD for coal-fired boilers
using low- and medium-sulfur coals has
become an important technology. Typi-
cally, the FGD end product has been dis-
posed of, untreated, in landfills. In some
countries in Europe, disposal by these
means is not permitted; i.e., the FGD
end product must be reclaimed for alter-
nate use. To meet this requirement,
Flakt entered into a cooperative agree-
ment with Dorr-Oliver to develop a
thermal oxidation system based on the
Dorr-Oliver FluoSolids Technology.
This system provides for the calcination
of the dry FGD end product to produce
a technical grade, anhydrous calcium
sulfate, which can be used as a setting
retarderfor Portland cement. This paper
describes the process and presents op-
erating experience and test results from
both a pilot plant program and a
demonstration plant (2.5 metric ton/
hour) test program. It also describes
how this process has been integrated
into a commercial spray dryer FGD sys-
tem recently sold by Flakt in the Federal
Republic of Germany.
Unpresented Papers
Reliability Problems and Solu-
tions of FGD Systems
D. C. Agarwal, Cabot Corpora-
tion, Kokomo, IN
Four Corners 4 and 5 Waste
Processing System: An Inte-
grated and Unique Retrofit
Alexander P. Simko, Arizona
Public Service Company,
Phoenix, AZ; Richard R. Lunt,
United Engineers and Construc-
12
-------
tors. Inc., Philadelphia, PA; and
Charles S. White, Conversion
Systems, Inc., Horsham, PA
Corrosion Resistance Compari-
son of Duplex Stainless Steel
with Type 317LM in Simulated
Flue Gas Scrubber Environ-
ments
Ronald E. Van Hoose and John
B. Guernsey, Eastern Stainless
Steel Company, Baltimore, MD
Corrosion Resistance of a Cr-
Ni-Mo-Cu-Fe Alloy in Scrubber
Environments and Other
Chemicals
Terry A. DeBold, Douglas G.
Frick, and Gerald A. Bauer, Car-
penter Technology Corporation,
Reading, PA
The Chemistry of Sodium Dry
Sorbent Injection
A. W. Mueller and A. E. Win-
ston, Church & Dwight Co.,
Inc., Piscataway, NJ
Techno-Economic Investigation
into the Hybrid Technology of
FGD and PCC Applied to Large
Boilers of Firing Canadian
Coals
R. L Wang, Consultant, St.
Catharines, Ontario, Canada
F. AyerandA. Wallace are with the Research Triangle Institute, Research Triangle
Park, NC 27709.
Julian W. Jones is the EPA Project Officer (see below).
The complete report consists of two volumes, entitled "Ninth Symposium on Flue
Gas Desulfurization:"
"Volume 1. Sessions 1 through 5," (Order No. PB 86-138 641/AS; Cost:
$34.95
"Volume 2. Sessions 6 through 10 & Appendix," (Order No. PB 86-138
658'/AS; Cost: $46.95)
The above reports will be available only from: (cost subject to change)
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC27711
Size Measurements in Aerosols
Produced in Sulphur Dioxide
Removal by Electron Beams
M. D. Carabine, P. G. Clay, and
G. Sisniega, Imperial College,
London, England
The NOXSO Process Develop-
ment: An Update
J. L Haslbeck and L G. Neal,
NOXSO Corporation, Library,
PA
13
U. S. GOVERNMENT PRINTING OFFICE 1986/646-116/20781
-------
c,
•<
OIVJ30C O
aeutm o
o
O
O
n
5'
I.
O
I
o>
oo
61 ZZ M
OO < O
mvm-4
» 70
OW90
303OO
o
w as
O 35
*• O
------- |