United States
Environmental Protection
Agency
Industrial Environmental Researc
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81-097 Aug. 1981
Project Summary
Third Survey of Dry SO2
Control Systems
Mary E. Kelly and S. A. Shareef
This report is the third in a series of
reports on the status of dry flue gas
desulfurization (FGD) processes in the
United States. This updated assess-
ment of dry FGD systems is based on a
review of current and recently com-
pleted research, development, and
commercial activities. Dry FGD sys-
tems covered include: (1) spray dryers
with a fabric filter or an electrostatic
precipitator (ESP), (2) dry injection of
alkaline material into flue gas combined
with particulate collection in an ESP
or a fabric filter, and (3) combustion of
coal/alkali fuel mixtures.
Spray drying continues to be the only
commercially applied FGD process.
Since the last survey was completed in
the Fall of 1980, two new utility and
two new industrial spray drying sys-
tems have been sold. All of these new
systems use a lime sorbent and include
a fabric filter for particulate collection.
No new commercial systems have
come on line since the last survey
report, but the first utility system is
scheduled to start up in the Spring of
1981.
A number of pilot-scale demonstra-
tion programs funded by vendors,
utilities, and/or government agencies
have been completed in the last few
months and several similar programs
are continuing currently. The Environ-
mental Protection Agency is currently
funding a spray drying demonstration
program and a program for the devel-
opment of a process for combustion of
coal/limestone fuel pellets. In a pro-
gram jointly funded by the EPA and
the Department of Energy, the com-
bustion of a pulverized coal/alkali fuel
mixture in a low-NO, burner is being
investigated. The Department of
Energy and a few vendors, as well as
EPA, are continuing to investigate dry
injection through pilot-scale demon-
stration programs. In addition, a sub-
stantial amount of work has begun in
the area of dry FGD waste disposal.
Waste disposal projects are being
funded by the DOE, the Electric Power
Research Institute (EPRI), and several
vendors.
This Project Summary was devel-
oped by EPA's Industrial Environmen-
tal Research Laboratory. Research
Triangle Park, NC. to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
This report is the third in a series of
reports on the status of dry flue gas
desulfurization (FGD) processes in the
United States for both utility and indus-
trial applications. Throughout this re-
port, dry FGD is defined as any process
which involves contacting an SOa-con-
taining flue gas with an alkaline material
and which results in a dry waste product
for disposal. This definition includes (1)
systems which use spray dryers for a
contactor with subsequent baghouse or
electrostatic precipitator (ESP) collec-
tion of waste products; (2) systems
which .involve dry injection of alkaline
material into the flue gas with subse-
quent baghouse or ESP collection; and
-------
(3) other varied dry systems which are
concepts that primarily involve addition
of alkaline material to a fuel prior to
combustion. This definition of dry sys-
tems excludes several dry adsorption or
"acceptance" processes, such as the
Shell/UOP copper oxide process and
the Bergbau-Forschung adsorptive char
process. The status of these processes
has been documented in previous EPA
reports. Fluidized-bed combustion has
also been excluded.
Also excluded was the regenerable
Rockwell Aqueous Carbonate Process
(ACP) which, although based on SO2
removal with a spray dryer, does not fit
the limitation of this study as being a
"throwaway" system. However, the
open loop, spray dryer contactor portion
of the Rockwell process has been adapted
for a "throwaway" system and, as such,
has been included here.
The report isdivided into four sections.
The first section presents an introduc-
tion to the report and a general process
description of each of the three types of
dry scrubbing technology (spray drying,
dry injection, and combustion of a coal/
alkali fuel mixture). The second section
contains an overview of the dry FGD
systems and includes: (1) a discussion
of the commercial and developmental
activities for each type of process and (2)
the highlights of the results of dry FGD
test work completed recently.
Details of commercial, current, and
recently completed research work, and
demonstration programs are provided in
the third section. The activities of each
organization or vendor involved with dry
FGD processes are discussed with
respect to current and future research
and development programs and com-
mercial system sales.
The final section presents some of the
findings of the research work in the area
of dry FGD waste characterization and
disposal. In addition, the scope of on-
going studies in this area and the
current and planned disposal methods
for commercial-sized dry FGD systems
sold to date are presented.
Recent Developments
Spray drying continues to be the only
commercially applied dry FGD technol-
ogy. Since Fall 1980, two new industrial
and two new utility systems have been
sold, bringing the total number of
commercial spray drying systems to 17:
11 utility and 6 industrial. (See Table 1
on pages 4, 5, 6, and 7.) At this time,
only two commercial systems are oper-
ating and they are both applied to indus-
trial boilers: at Celanese Fibers Co., in
Cumberland, MD, and at Strathmore
Paper, Inc., in Woronco, MA. No utility
system is in commercial operation, al-
though initial start-up procedures have
begun at Otter Tail Power Co.'s Coyote
Station in Beulah, ND.
Several spray drying demonstration
programs have been completed or are
nearing completion. It should be noted
that none of these programs, which
ranged in size from 1000 acfm to
120,000 acfm, were conducted on high
sulfur coals. Two new demonstration
programs are scheduled to begin soon.
Northern States Power's Riverside
Station is in start-up and the EPRI-
sponsored spray drying demonstration
program at Public Service of Colorado's
Arapahoe Station is in the design and
construction stage.
Dry injection development programs
are continuing under DOE and EPRI
funding, and Buell recently completed
EPA-f unded tests at the City of Colorado
Springs' Martin Drake Station. Recently
reported results have shown that a
substantial degree of SOa removal (up to
90 percent) is achievable with nahcolite
at stoichiometric ratios (moles Na2 O/
mole inlet S02) of less than 2. But even
with this improved performance, rela-
tive to earlier tests, the commercial
application of dry injection has not
occurred. Primary restrictions on the
commercial development of this tech-
nology are sorbent cost and availability,
relatively high stoichiometric require-
ments, and waste disposal concerns
related to the undesirable solubility and
leachability properties of the sodium-
based waste solids.
Development of a process to reduce
SO2 and nitrogen oxides(NOx) emissions
through combustion of pulverized coal/
alkali mixtures in a Iow-N0x burner is
continuing under a 5-year program
jointly funded by the DOE and EPA. The
primary emphasis of this program will
be to assess the retrofit potential of the
technology for existing boilers.
The EPA is also continuing to fund
development of the coal/limestone
pellets for control of SO2 emissions from
industrial stoker-fired boilers. However,
a continuous 14-day test of the pellet,
scheduled for November 1980, was
cancelled due to inadequacies in the
pellet production process.
Recent pilot-scale test work with both
of the combustion technologies has
shown that they are capable of achiev-
ing at least 50 percent SO2 removal.
Combustion of coal/alkali fuel mixtures
offers a low cost alternative to scrubbing
for industrial boilers, and interest in
these technologies remains strong in
the light of potential new source regula-
tions for industrial boilers and the
concern over acid rain. Much work
remains, however, to fully characterize
the impact of these technologies on
boiler and particulate control system
design and operation.
Current Status of Dry FGD
Processes
Spray Drying
Eleven utility systems (totalling over
3200 MWe) and six industrial systems
had been sold as of January 1981. Only
two systems were fully operational a
this writing: the Rockwell/Wheelabrator-
Frye system at Celanese Fibers Com-
pany's Amcelle plant in Cumberland,
MD, and the Mikropul system at
Strathmore Paper, Inc. in Woronco, MA.
The first utility system to become opera-
tional will likely be another Rockwell/
Wheelabrator-Frye system, this one al
the Coyote Station of Otter Tail Power
Company. Initial start-up procedures al
Coyote began in January 1981. Con-
struction on several other utility sys-
tems, including Antelope Valley Unit 1
(Basin Electric Power Cooperative),
Laramie River (also Basin Electric), and
Colorado Ute Power Company's Craig
Station, is reportedly on schedule.
The two new utility spray drying
systems will both be applied to low-
sulfur coal-fired boilers. Basin Electric
has awarded Joy/Niro a contract for the
utility's 430 MWe Antelope Valley Unit
2. The North Dakota lignite-fired boiler
and lime-based spray dryer/fabric filter
system will essentially be a duplicate ol
the Antelope Valley Unit 1 system. The
second new utility system has been
awarded to the Buell/Anhydro join!
venture. The Marquette (Ml) Board o1
Light and Power has purchased the
system to treat flue gas from a 44 MWe
coal-fired boiler located at Marquette's
Shiras Municipal Power Plant. This
system will also use a lime sorbent anc
include a fabric filter for collection of fl\
ash and waste solids. Both new utility
system designs will have provisions foi
recycle of the waste solids.
A new industrial system was recently
awarded to Joy/Niro by the Department
of Energy's Argonne National Labora-
tories in Argonne, IL. Argonne is con-
verting an existing boiler to fire a
nominal 3.5 percent sulfur Illinois
-------
bituminous coal. This system will be the
first commercial application of spray
drying to a boiler firing coal with a sulfur
content of greater than 3 percent. (How-
ever, the Mikropul system at Strathmore
Paper, Inc. is scheduled to undergo
compliance/performance testing in
February 1981 while the boiler is firing
a 3 percent sulfur coal.) Sulfur dioxide
emissions from the Argonne unit are to
be reduced by about 80 percent to meet
an SOz emission limit of 1.2 Ib/million
Btu thermal input. Ecolaire. Systems,
Inc. has sold a lime-based spray dryer/
fabric filter system to Container Corpo-
ration. The system will treat flue gas
from a 170,000 Ib steam/hr coal-fired
boiler at Container Corporation's facility
in Philadelphia, PA.
Contracts for at least two new indus-
trial systems (one for a high sulfur coal-
fired boiler) and one utility system are
expected to be awarded soon. Also, at
least five uti lilies have specified only dry
FGD or are considering both dry and wet
FGD for new units that will go on-line
before the end of the decade.
Part of the reason for the increasing
commercial application of dry FGD is the
potential cost savings the technology
' offers when compared with conventional
wet lime/limestone scrubbing. Results
of a recent EPA-funded TVA conceptual
cost study support this premise. The
study compared the capital and annual
costs of a lime spray dryer/fabric filter
system with that of a wet limestone
scrubbing system for a new 500 MWe
power plant. Costs were developed for
three fuels: low sulfur western coal, low
sulfur eastern coal, and high sulfur
eastern coal. The spray dryer designs
were based on vendor information,
while the limestone scrubbing design
(which includes an ESP upstream of the
scrubber) was .based on information
from the EPA Shawnee test facility and
general industry information. Table 2
shows the results of the cost compari-
son. Lime spray drying is reported to be
less expensive on a capital and annual
cost basis for all three coals. The higher
absorbent cost associated with the
spray dryer process, particularly for the
high sulfur coal case, is offset by the
lower spray dryer equipment costs. Note
that the spray drying operating costs are
based on limited data and may change
as more experience is gained with
commercial utility systems.
Another EPA-funded conceptual eco-
nomic study was performed to estimate
the cost of retrofitting a spray drying
Table 2. Comparison of Capital Costs and Annual Revenue Requirements for Wet
and Dry FGD Systems for a New 500 MW Power Plant"
Capital Cost Level/zed Annual Revenue
($/kW) Requirements (mills/kWhr)
Low Sulfur Western Coal
Lime spray drying1'
Limestone wet scrubbing?
Low Sulfur Eastern Coal
Lime spray drying*
Limestone wet scrubbing*
High Sulfur Eastern Coal
Lime spray drying/1
Limestone wet scrubbing*
"Source: Burnett, T.A., et al.
Assessment. (Presented at E 3A
31.) Tennessee Valley Authority.
^Includes fabric filter.
"Includes ESP upstream of scrubber.
144 to 152
168 to 176
144 to 152
180 to 188
180 to 188
236 to 234
8.7 to 9.1
10.5 to 10.9
8.4 to 8.7
11.3 to 11.6
14.5 to 14.9
16.4 to 16.7
Spray Dryer FGD: Technical Review and Economic
's Sixth FGD Symposium. Houston, TX. October 28-
Muscle Shoals, AL.
system to a northeastern utility power
station. The costs were assumed to be
associated with conversion to low sulfur
coal-firing from oil-firing. The host
estimates were based on vendor Bud-
getary quotes. The total installed costs
for the spray dryer/fabric filter sviitem
ranged from $89/kW to $118/kW
Fora
generalized northeastern location.
Developments in spray dryer sv:;tem
design include increased use of off-
product recycle and closer approaches
to the adiabatic saturation temperature
at the dryer outlet. Reslurryinjj and
recycling of dry fly ash/product solids
mixture has been shown to significantly
increase SO2 removal for a giver stoi-
chiometric ratio of fresh sorbent to nlet
S02 and to reduce fresh sorbent con-
sumption for a given SO2 removal
efficiency. However, recent tests have
shown conflicting results with respect
to the role of fly ash alkalinity in the
improved system performance observed
with recycle.
Recent tests have also shown that a
20°F approach to saturation at the c ryer
outlet appears to be the optimum control
point in spray dryer system design. This
approach has been shown to sijiifi-
cantly increase SOz removal relative to
a 30 or 40°F approach, while still
allowing for an adequate margin of
safety to protect the downstream coijitrol
device from moisture condensatioi and
also maintaining dry free-flowing
product solids. l
These and other.
more subtle, refine-
ments of spray drying technology have
occurred partly as
a result of several
recent demonstration programs. A
number of the programs are now com-
plete, and it appears that future testing
may be somewhat limited to high sulfur
coal applications. However, at least five
demonstration programs, of varying size
and focus, will be conducted during
1981. Buell/Anhydro is continuing
tests at the City of Colorado Springs'
Martin Drake Station, and Joy/Niro has
begun a 2-to 3-year large-scale demon-
stration program at Northern States
Power Company's Riverside Station.
Also, Combustion Engineering, Inc.
recently started tests on a 100,000 acf m
unit at the Gadsden Station of Alabama
Power Company. DOE's Pittsburgh
Energy Technology Center (PETC) will
soon begin pilot-scale tests to evaluate
the performance of a spray dryer/fabric
filter system treating flue gas from a
high-sulfur coal-fired test furnace.
Finally, the Electric Power Research
Institute (EPRI) has contracted with
Stearns-Roger for demonstration tests
at Public Service of Colorado's Arapahoe
Station.
The demonstration programs also
involve evaluation of sorbents other
than lime. In an effort to evaluate the
viability of less expensive limestone for
the spray drying process, Buell/Anhydro
has run tests using a pulverized lime-
stone slurry with adipic acid addition.
Adipic acid appears to benefit the reac-
tion between CaC03 and S02, but the
maximum S02 removal achieved on a
straight-through basis was less than 40
percent at an inlet S02 concentration of
1000 ppm. Buell/Anhydro has also run
tests with trona, which exhibited better
S02 removal and sorbent utilization
-------
Table 1. Key Features of Commercial Spray Drying Systems Sold to Date
System
Purchaser/Vendor
Utility
Otter Tail Power Co./
Rockwell/Wheelabrator-Frye
Basin Electric Power Coop/
Joy-Niro
Basin Electric Power Coop/
Babcock & Wilcox
Basin Electric Power Coop/
Joy-Niro
Location/Size
Coyote Station, Beulah. NO/
Unit 1, 41 OMW 11, 890.000
acfmj
Antelope Valley, Beulah, NO/
Unit 1, 430 MW
12,200.000 acfm)
Laramie River, Wheat/and.
WY/Unit 3. 500 MW
(2.810.000 acfm)
Ante/ope Valley. Beulah,
ND/Unit 2. 430 MW
(2.200.000 afcml
System Description
Four parallel spray dryers,
with 3 centrifugal atomi-
zers each, followed by
fabric filter with Dacron
bags. Will initially use
commercial soda ash. Sorbent
utilization guarantee of 80%.
Five parallel spray dryers
(one spare), single rotary
atomizer per dryer, followed
by fabric filter with Teflon-
coated fiberglass bags. Lime
sorbent with partial recycle
of solids. Ball mill slaker.
Four parallel reactors (one
spare) with 12
fluid nozzles each. Each
reactor followed by an ESP.
Lime sorbent. no solids
recycle.
Design identical to Antelope
Valley Unit 1 (see above).
Coal
North Dakota lignite;
0. 78% S average; 70BO
Btu/lb; 7% ash.
North Dakota lignite;
0.68% S average; 1.22%
S maximum.
Wyoming subbituminous.
0.54% S average;
0.81% S maximum;
8 140 Btu/lb; 8% ash.
North Dakota lignite;
0.68% S average;
1.22% S maximum.
SO* Removal
Guarantee
70% lor all fuels.
62% for average
S coal; 78% for
maximum S coal.
82% for average
S coal; 90% for
maximum S coal.
62% for average
S coal; 78% for
maximum S coal.
Mar queue Board of Light
and Power/Buell-Anhydro
Tucson Electric/
Joy-Niro (2)
United Power Association/
Research-Cottrell
Plane River Power Authority/
Joy-Niro
Colorado-Ute Association/
Babcock & Wilcox
Sunflower Electric Coop/
Joy-Niro
Industrial
Celanese Fibers Co. /
Rockwell/Wheelabrator-Frye
Strathmore Paper Co./
Mikropul. Inc.
University of Minnesota/
Kennecott Development Co.
(Environmental Products
Division)
Shiras Municipal Power
Plant. Marquette. Ml/
44 MW (226.000 acfm)
Springerville Station/
Units 1 and 2; 350 MW
each
Stanton Station. Stanton,
ND/65 MW
Rawhide Station/Unit 1
250 MW
Craig Station/Unit 3
450 MW
Ho/combe Station/Unit 1
310 MW
Amcelle plant, Cumberland.
MO/65,000 acfm
(110,000 Ib steam/hr)
Woronco. MA/
40.000 acfm (85.000
Ib steam/hr)
Univ of Minnesota/2 units
at 120,000 acfm each
Single spray dryer with
rotary atomizer. Reverse-air
fabric fitter.
Lime sorbent. Solids recycle.
Spray dryer/fabric filter
design. Lime sorbent.
Rotary atomization.
Spray dryer/fabric filter
rotary atomizers, possibly
multiple atomizers per
dryer. Lime sorbent.
Spray dryer/fabric filter
design. Rotary atomizers.
Lime Sorbent.
Horizontal spray dryers
with nozzle atomizers,
followed by fabric filter.
Solids recycle. Ball mill
slaker for lime sorbent.
Spray dryer/fabric filter.
Rotary atomization. Lime
sorbent.
Spray dryer with single
rotary atomizer followed
by fabric filter with felt/
fiberglass bags. Paste
slaker for lime sorbent.
No solids recycle.
Spray dryer with four
two-fluid nozzles, followed
by fabric filter with specially
finished acrylic bags.
Two spray dryers-one with
single, other with multiple
rotary atomizers-followed by
fabric filter with fiberglass
bags. Lime sorbent.
1.5%S western
subbituminous 15% ash.
7700 Btu/lb.
New Mexico coal;
0.69% S.
Low and intermediate
sulfur subbituminous
Montana coal.
Western subbituminous
coal; 1.3% S.
0.70% S. 8950 Btu/lb,
14% ash design coal;
0.40% S. 10250 Btu/lb.
Western subbituminous
coal.
1.5% Sand2 to2.5%
S eastern coals.
2.3 to 3% S
eastern coal.
Subbituminous coal;
0.6 to 0.7% S.
80% design efficiency.
61%.
Not available.
80%.
87% for design
coal.
80%.
70% for 1.5% S coal.
87% for 2.0% S coal.
75%.
70%.
-------
Reported Capital
Cost
Reported Operating
Cost
Status
$32.000,000
$49,665.100
($113/kWf.
$49.807,000
(83/dWf.
$54.000.000°.
($126/kW>
Not available.
Not available.
$5.000.000
t$77/kW).
Not available.
($100/kW)
Not available.
$1.250.000*.
S1.400.000'1''.
$3,300,000fa.
$6.580,000 ($2.5
mills/kWh/'. Does
not include waste
disposal.
$2.270.834/yr ($0.8
mills/kWhf. Lime
cost-$1.102.500
($6O/ton). Does not
include waste disposal.
$2,571,000/yr ($0.7
mills/kWhf. Lime
cost - $1.396,570
($60/tonl. Does not
include waste disposal
costs.
Same as Antelope Valley
Unit 1 (see above).
Not available.
Not available.
Not available.
Not available.
Not available.
Not available.
Not available.
$600/day.
Not available.
Start-up scheduled for
mid-1981.
Stan-up scheduled for
April 1982.
Start-up scheduled for
Spring 1982.
Start-up scheduled for
1985.
Start-up scheduled for
Fall 1982.
Unit 1 scheduled to start
up in late 1984; Unit 2 in
1986.
Start-up scheduled for 1981.
Start-up scheduled for 1983.
Initial operation in November
1982. Commercial operation
in April 1983.
Start-up scheduled for 1983.
Operational. Passed
Maryland state compliance
tests in February 1980.
Has achieved guaranteed
removal.
Operational. Now achieving
removal guarantee.
Commercial operation in
Fall 1981.
than lime. However, the soluble nature
of the waste products and the cost
and/or availability of trona may restrict
the use of this sorbent in commercial
systems.
The current focus of on-going and
planned demonstration programs ap-
pears to be toward higher inlet $62
concentrations and obtaining a better
definition of spray dryer performance
limits. And, of course, many of the
programs will evaluate more subtle
refinements to the technology such as
lime quality, slaking technique, and
recycle methods. Another area of
emphasis will likely be development of
more sophisticated process control
techniques to improve system reliability
and reduce sorbent-related operating
costs.
Table 3 shows the status of eight
major spray drying demonstration pro-
grams. Results from three of these
programs, Buell/Anhydro at Martin
Drake, Research-Cottrell at Comanche,
and Babcock & Wilcox at Jim Bridger,
were presented at .the recent EPA-
sponsored FGD Symposium. Also, Flakt,
Inc. has recently published some results
of its Jim Bridger tests. Many of these
results are presented in detail in the full
report.
Dry Injection
Presently, there are no plans for the
construction of any commercial dry
injection systems. However, several
demonstration programs are being
conducted. Demonstration-scale dry
injection systems have been or are
being operated through funding by the
Department of Energy (DOE), the Envi-
ronmental Protection Agency (EPA), and
the Electrical Power Research Institute
(EPRI). The current status of several
demonstration programs is presented in
Table 4.
Investigations with a number of sor-
bents have shown that only sodium-
based sorbents, such as sodium bicar-
bonate, nahcolite, and trona ores, provide
satisfactory SO2 removal. (However,
nahcolite appears to be the most reactive
sorbent. Removal efficiencies of up to
90 percent have been reported with a
nahcolite sorbent. Many important
variables influence S02 removal during
dry injection. These include: stoichio-
metric ratio, injection temperature,
sorbent pretreatment, sorbent particle
size, and the mode of injection.
Buell, a division of Envirotech Corpo-
-------
Table 1. Key Features of Commercial Spray Drying Systems Sold to Date
System
Purchaser/Vendor
Utility Location/Size System Description Coal
Calgon/Niro-Joy 57,000 acfm
Argonne National Labs/ Argonne, IU
Niro-Joy 1 70.000 Ib steam/ hr
Spray dryer/ fabric filter.
flotary atomizer. Soda ash
sorbent. Removing SOx and
HCIfrom 1 700° F gases.
Solids recycle.
Single spray dryer with
rotary atomizer followed by
1 to 2% S coal,
6000-8000 ppm SOi,
8000 ppm ha/ides.
3.5% S Illinois
bituminous coal.
SO] Removal
Guarantee
76% SOa
90%HCI.
78.8% SOi removal
(1.2lb/10'Btu).
Container Corporation/
Ecolaire Systems
Philadelphia, PA/
170,000 Ib steam/hr
pulse-jet fabric filter.
Lime sorbent, solids recycle.
Spray dryer with single
rotary atomizer followed
by pulse-jet fabric filter.
Lime sorbent, so/ids recycle.
1% S coal.
Not available.
^Capital cost for turnkey installation from airpreheater outlet to stock connection, excludingI.D. fans (1977$). Source: Johnson, O.B. etal. "Coyote Station,
First Commercial Dry FGD System." (Presented at 41st Annual American Power Conference, Chicago, IL. April 23-25, 1979.)
''Based on 35-year plant life. 75% capacity factor 11981$). Source: Janssen.K.E. andR.L Eriksen. "BasinEtectric's Involvement withDryFlueGasDesulfurization.
" In Proceedings; Symposium on Flue GasDesuHurization - Las Vegas. NV, March 1979; Volume II. EPA - 600/7-79-167b fNTIS PB 80-133176). July 1979.
pp 629-653.
"(1980$)
"Stern, J.L. "Dry Scrubbing for Industrial Flue Gas Desulfurization: State-of-the Art. 1980." (Presented at the 89th National Meeting ofAIChE. Port/and, OR.
August 17-20. 1980.)
"From "ground-up." 11979$)
'Kelly, M.E. and S.A. Shareef. Meeting notes at Babcock & Wilcox. Barberton. OH. June 1980.
'"Straight-through system." 11980$)
Table 3. Major Spray Drying Demonstration Activities'*
Vendor/Agency Location
Size
Comments
Babcock & Wilcox0
Buell/Anhydroc
Flakt. Inc.
Pacific Power & Light's Jim Bridger 120,000 acfm
Station
Colorado Springs-Martin Drake 20,000 afcm
Station
Pacific Power & Light's Jim Bridger 15,000 acfm
Station
Combustion Engineering Alabama Power's Gadsden Station 100,000 acfm
Ecolaire Systems, Inc.
Research-Cottrellc
(Cottrell Environmental
Sciences)
Electric Power Research
Industry (EPRI)
(Steams-Roger will
conduct the tests)
Joy-Niro
Nebraska Power's Gerald Gentleman 10,000 acfm
Station mobile pilot plant
Public Service of Colorado's
Comanche Station
Public Service of Colorado's
Arapahoe Station
Northern States Power's
Riverside Station
10,000 acfm
Testing completed.
EPA-funded testing still in
progress. Bulk of program has
been completed.
Testing completed.
Testing has begun and is expected
to run for remainder of 1981.
Testing near completion.
Testing completed in January 1981.
2.5 MWe equivalent System in design and construction
of flue gas
680,000 acfm
phase. (Spray dryer and
associated equipment will be
supplied by Stork-Bowen.)
In start-up.
"More information on each of these programs can be found in Section 3 of full report.
''Several other smaller demonstration tests are also being conducted by private firms, and the Department of Energy's Pittsburgh
Energy Technology Center will begin tests soon on a 2500 acfm unit.
cResults presented in FGD Symposium - related papers.
6
-------
Reported Capital
Cost
Reported Operating
Cost
Status
$1.600.000°'''.
Not available.
Not available.
Not available.
Not available.
Not available.
Under construction.
Start-up scheduled for
Winter of 1981-82.
Construction scheduled for
April 1981. start-up in
October 1981.
ration, recently completed EPA-funded
dry injection testing at the City of
Colorado Springs' Martin Drake Station.
The tests were begun in late October
1979 and completed in May 1980.
Experiments with three sorbents,
nahcolite, raw trona, and refined trona,
were conducted on the 4500 acfm dry
injection baghouse system. The sorbents
were ground to less than 74 /urn in
diameter before being injected into the
duct leading to the fabric filter. These
tests were performed in an attempt to
1 characterize the effects of sorbent type,
stoichiometric ratio, temperature, and
air-to-cloth ratio on SOz removal.
The results of these parametric tests
showed that nahcolite provided the best
SOz removal, followed by refined trona.
Raw trona exhibited the weakest S02
removal capability. It was also found
that an increase in the stoichiometric
ratio caused SOz removal to increase.
However, removal began to level off at
stoichiometric ratios between 1.5 and
1.75 for these sodium sorbents. The
effects of changes in temperature were
also characterized. It was found that
SOz removal decreased with increasing
temperature for both nahcolite and
refined trona. However, removal with
raw trona increased with increasing
temperature. Results of tests character-
izing the effect of air-to-cloth ratio have
shown that variation of the air-to-cloth
ratio from 1.5 to 3.0 ft/min had no
noticeable effect on SOz removal.
The DOE has conducted research on
dry injection systems at both the Grand
Forks Energy Technology Center (GFETC)
and Pittsburgh Energy Technology
Center (PETC). These tests were per-
formed to characterize several process
parameters, such as sorbent type, inlet
SOz concentration, inlet gas tempera-
ture, bag materials, and air-to-cloth
ratios.
Work conducted at GFETC with a 200-
scfm dry injection system has shown
that up to 90 percent utilization (based
on NO, and SOz) has been obtained with
trona and nahcolite sorbents. Low
sulfur western coals were used in these
tests, which resulted in inlet SO2 con-
centrations of 650 to 1100 ppm. The
results reported here are preliminary; a
final report on these tests has not been
published.
GFETC has just completed the instal-
lation of a new 130-scfm pulse jet
Table 4. Current Dry Injection Programs
Vendor Location
Size
baghouse. This baghouse will be used
exclusively for dry injection studies (the
present baghouse was also used for
particulate characterization studies).
Parametric tests with both nahcolite
and trona will be conducted.
Dry injection work at the Pittsburgh
Energy Technology Center (PETC) was
completed in the Fall of 1980. A final
report on the dry injection studies,
which evaluated the performance of
hahcolite, trona, and commercial sodium
bicarbonate, is in preparation. The
average baghouse temperature for the
dry injection tests was 400°F. The fabric
filter was equipped with Nomex bags
and was operated at an air-to-cloth ratio
of 4 ft/min. In general, the tests indi-
cated that nahcolite showed the greatest
SO2 removal capability of the three
sorbents evaluated.
In tests conducted with 1.1,1.6, and
3.5 percent sulfur coals, dry injection of
nahcolite resulted in SOz removals of up
to 95 percent with a stoichiometric ratio
of 1.5 moles Na20 per mole of inlet
sulfur. The tests also indicated that SOz
removal decreased as inlet SOz concen-
tration increased. However, 90 percent
S02 removal was reportedly achieved
with a stoichiometric ratio of 1.5, even
when 3.5 percent sulfur coal was
burned.
EPRI's Air Quality Control Program
involves research on the technical
aspects of a dry injection/baghouse
system. Detailed laboratory scale te*sts
have been completed. The results of
these tests were summarized in the
Second Survey of Dry SOz Control
Systems. EPRI has been conducting
large-scale dry injection tests at the
Public Service of Colorado's Cameo
Station since the Fall of 1980. These
tests are being performed by KVB, Inc.
on the 22 MWe Unit 1 boiler at Cameo
using the existing fabric filter. Although
no results are currently available for
Comments
EPA /Buell-Envirotech
DOE/'Grand Forks Energy
Technology Center
DOE/Pittsburgh Energy
Technology Center
EPRI/KVB
Colorado Springs -
Martin Drake Station
GFETC Labs
PETC Labs
Public Service
Company of Colorado
Cameo Station
4500 acfm
200 acfm
500 Ib coal/hr furnace
20 MWe
Testing completed in May
1980. EPA funded.
Testing completed. Report is
being prepared.
Testing completed. Report
is being prepared.
Testing in progress.
Report on initial
testing is expected
in June 1981.
-------
publication, the first phase testing with
nahcolite has been completed and a
final report is scheduled for June.
Further testing will continue through
the end of 1981,
Combustion of Coal/Alkali
Fuel Mixtures
Currently, two processes are being
developed, based on the combustion of
coal/alkali fuel mixtures to control SO?
emissions: combustion of coal/lime-
stone pellets in an industrial stoker-
fired boiler, and combustion of a pul-
verized coal/alkali fuel mixture in a low-
NOx burner.
Recent large-scale tests with a 3.5:1
calcium-to-sulfur molar ratio coal/lime-
stone pellet demonstrated that 50 per-
cent SOz removal was achievable with
this technology. Although earlier labo-
ratory tests had indicated that SO2
removals of up to 75 percent were
possible, the higher, less controllable
bed temperatures in the full-size indus-
trial test unit (60,000 Ib steam/hr)
resulted in lower removal efficiencies.
Inadequacies in the pellet production
process have hampered development
and progress in recent months, but EPA
hopes to resolve these problems in the
near future. Meanwhile, Battelle-Co-
lumbus Laboratories is continuing to
test the properties and S02 removal
capabilities of the pellets on a smaller
scale (25,000 Ib steam/hr boiler).
The DOE and EPA have proposed a
jointly funded 5-year development pro-
gram to further investigate the concept
of firing a coal/alkali fuel mixture in a
low-NOx burner. SC>2 removals of 50
percent and greater have been demon-
strated in small-scale tests.
Reasons for the accelerated develop-
ment of both these technologies include
the potential cost savings offered by
reduced equipment requirements rela-
tive to conventional wet FGD and the
retrofit potential of the technology for
existing boilers.
Dry FGD Waste Disposal
The collective experience of U.S.
utilities and industries in operating wet
scrubbers and disposing of scrubber
waste (sludge, fly ash) has been studied
and documented fairly extensively.
Studies have been aimed at (1) develop-
ing a data base on sludge and ash
handling procedures, (2) providing
independent evaluations of the sludge
fixation processes, (3) quantifying vari-
ables affecting the solubility of trace
elements and potentially toxic species
from solid waste by-products, and (4)
establishing guidelines for the construc-
tion of sludge disposal facilities.
Similar efforts directed at character-
ization, disposal, and utilization of
waste from dry S02 control systems are
only now getting underway. As more
commercial dry FGD systems begin to
operate, these efforts can be expected to
gain momentum, and a broader data
base should begin to emerge. The
results of some of the studies are
summarized below:
(1) Chemico, a division of Envirotech
Corporation, has conducted studies to
determine various physical and chemi-
cal properties of the dry waste solids
mixed with varying amounts of water. A
mixture of 80 percent solids and 20
percent water displayed the most desir-
able properties in terms of disposal
requirements. The unconfined compres-
sive strength of the mixture was found
to be about 12,000 Ib/ft2, and the
permeability was less than 10~5 cni/sec.
These values are similar to those for
conventional fly-ash-stabilized FGD
sludges. The leachability of heavy me-
tallic compounds, as determined in
laboratory tests, was found to be well
within the limits set by EPA as guide-
lines for definition of a hazardous
waste.
(2) Battelle-Columbus Laboratories,
under a contract to Buell's Emission
Control Division, has conducted a tech-
nical and economic feasibility study of
the Sinterna* process for disposal of dry
FGD wastes. Laboratory-scale studies
were conducted on the powdered solid
waste that was generated during nah-
colite dry injection tests conducted by
Buell at Colorado Springs' Martin Drake
Station.
Sodium-based wastes present a dis-
posal problem because of their high
leaching potential. Untreated sodium-
based dry FGD wastes are not considered
suitable for disposal by conventional
landfill methods. The Sinterna process
produces stabilized pellets from the
untreated wastes. These pellets are
considered suitable for landfill disposal
because of reduced leaching potential.
Laboratory tests showed that the
sulfate leaching could be reduced from
the 60 percent typical of a dried unsin-
tered pellet to about 20 percent after
sintering. The amount of leaching is
measured by placing the pellets in a
•The Patent for the Sinterna process is held by
Industrial Resources, Inc.
continuously stirred beaker of water and
sampling the water at predetermined
intervals, with the analysis at 100 hours
used as the standard for comparing
sintered and unsintered pellets. Using
this method, it was observed that pellets
(dried at 105°C) showed 60 percent
sulfate leachability after 100 hours,
while the comparable figure for pellets
sintered at 1000°C was about 20 per-
cent, and about 30 percent of the sulfur
was converted to S02. Reducing the
sintering temperature to 925°C reduced
the conversion of sulfur to S02to about
10 percent, while 65 percent of the
sulfate was found to have leached out of
the pellet (at 100 hours). A sintering
temperature of 1000°C was found to
provide the best balance between re-
duced sulfate leaching and conversion
of sulfur to S02.
Although the Sinterna process appears
to be technically feasible, and the
sintered pellets have properties more
suitable for landfill than untreated
waste or. dried pellets, it does not appear
to be economically feasible. The esti-
mated annualized cost (including capital
charges for pelletizing, drying, and
sintering equipment) is $100/tonof dry
waste. The capital cost of the process for
a conceptual 500 MWe plant (producing
about 20 tons/hour of nahcolite-based
dry. injection wastes) was estimated to
be about $20 million.
(3) Research-Cottrell has evaluated
the characteristics of the spray dryer
waste generated at Public Service
Company of Colorado's Comanche Sta-
tion. The Comanche fly ash is highly
cementitious and has a high reactive
CaO content. Results of the preliminary
tests indicate that the dry wastes (from
tests with lime) are similar to wet FGD
sludge/fly ash mixtures. Initial perme-
abilities of the laboratory samples were
in the 10~5 cm/sec range, but after 28
days of curing at ambient conditions,
permeability dropped to the 10"7cm/sec
range.
(4) Niro Atomizer, Inc. has also con-
ducted research on disposal of wastes
from dry FGD systems. The properties of
dry wastes have been investigated and
alternatives for disposal, depending on
the properties of the specific waste
material, have been developed. The
composition of the wastes was found to
vary with coal composition and process
conditions in the scrubber. For low
sulfur western coals, fly ash dominates
the waste material and its characteristics
are, therefore, very important.
8
-------
Fly ash from western coals has pozzo-
lanic properties. (A pozzolan is a material
containing silicon or aluminum and
silicon compounds. Alone it has no
hardening qualities, but exposed to
water at normal temperatures it reacts
with calcium hydroxide to form a cement-
like material.)
The critical parameters for a landfill
material are density, compressive
strength, permeability, and the compo-
sition of the leachate and run-off gener-
ated. Niro claims that 80 to 90 percent of
the laboratory-measured dry density
and compressive strength can be ob-
tained in a landfill. High density is
desirable in that it permits more material
to be disposed of in a given amount of
land. Compressive strength is important
in that the landfill must support its own
weight, sometimes in a thickness of 50
to 100 ft, and should allow trucks and
earth-moving equipment to move on its
surface without getting stuck. Low
permeability reduces the risk that rain
water will leach through the landfill and
contaminate the groundwater under-
neath with soluble salts.
Niro has found that the addition of
sufficient water to lubricate the grains
, of dry material will allow a greater final
density for the same amount of com-
pactive effort. Once the material is
wetted and compacted into the landfill,
it undergoes several chemical reactions
that bind the individual particles together
and fill the remaining void space with
impermeable reaction products. The
first reactions to take place are hydra-
tions. The fly ash contains considerable
calcium oxide which combines rapidly
(in a few minutes) with water to form
calcium hydroxide. Calcium sulfate
hemihydrate combines with water
somewhat more slowly to form the
dihydrate. Much slower reactions form
calcium-silicon-aluminatesandsulfo-
aluminates over a period of days and
months. Much of the ultimate strength
of the cured product comes from these
compounds.
The compressive strength of the
landfilled material, as reported by Niro,
is normally 100 to 150 psi, permeability
is 10"* cm/sec or less, and density is
100 Ib/ft3. (A permeability of 10~6
cm/sec equals 1 ft/year and a density
of 100 Ib/ft3 is a normal value for
compacted soil.)
Niro concludes that the outcome of
landfill disposal is a block of solidified
waste 50 to 100 ft deep. It is essentially
inert and impermeable, may be covered
with soil and revegetated, is strong
enough to support light construction,
and may be excavated or drilled like soft
rock or hard soil.
An alternative method of dry FGD
waste disposal, that may have less
impact on the environment and could
result in lower disposal costs, is also
being developed by Niro, Inc. If the
powder is exposed to a higher degree of
compression at lower water content, it
is possible to form pellets of high
strength suitable for commercial uses.
The "synthetic gravel" is produced by
adding 10 to 20 percent water to dry
FGD waste and compressing it to be-
tween 2.5 and 3.0 times its loose
density. This is reported to result in a
cured density of about 120 Ib/ft3 and a
compressive strength of over 10,000
psi.
The gravel reaches 50 percent of its
strength in 2 days and 90 percent in 10
days. The pellets are reportedly not
affected by water, ordinary heat, freeze,
thaw, or mechanical handling. Cold
weather as well as excessive dryness
does slow the curing process. However,
the return of heat or moisture is reported
to permit curing to continue. Therefore,
Niro claims that the gravel may be silo
cured a few days before use or may be
stock-piled outdoors for years.
Niro has estimated the capital cost of
a dry FGD waste disposal system to be
about $9/kW and the operating cost to
be about 0.8 mills/kWh. These estimates
compare favorably with TVA's estimate
for wet FGD (lime/limestone) waste
disposal costs of about $17/kW for
capital cost and 1.08 mills/kWh for
operating cost. According to Niro, the
difference in the costs is primarily due to
the elimination of the thickening and
filtration modules for a dry system.
Scope of On-going Waste
Disposal Studies
In addition to the EPA-sponsored
study being conducted by Cottrell Envi-
ronmental Services, Inc., two other
major studies aimed at characterizing
dry FGD wastes are currently underway.
The scope of these studies is briefly
described below.
(1) EPRI has recently begun a study to
characterize the physical and chemical
properties of solid waste from spray
drying and dry injection system. A total
of 30 solid samples will be used to
identify those parameters which impact
the handling and disposal of dry FGD
wastes. The parameters will also provide
comparative data concerning the effects
of system design and coal type on the
physical and chemical make-up of the
solids.
(2) DOE's Grand Forks Energy Tech-
nology Center (GFETC) is characterizing
the chemical and physical properties of
waste material produced by the com-
bustion of low-rank coals. This will
include waste from dry FGD systems as
well as wet systems and coal fly ash.
The waste material will be characterized
chemically and physically through the
following analyses:
• Collection of the waste materials.
• Fixation of FGD sludges.
• Chemical and physical analysis of
all the waste materials, including
FGD sludges before and after fix-
ation.
• Extraction and column leaching of
the various waste materials.
• Evaluation of current disposal
techniques and future disposal
requirements.
• Assessment of the GFETC column
leaching program results.
Current and Planned Disposal
Methods for Commercial Units
The studies performed by most dry
FGD system vendors to date have not
shown any special treatment require-
ments for disposal of dry FGD wastes.
Consequently, the current and planned
disposal methods for dry FGD commer-
cial systems are not much different from
the established disposal methods for
wet FGD systems. At present there are
only two commercial dry FGD systems
in operation. Wastes from both the
Strathmore Paper Company system at
Woronco, MA, and the Celanese Fibers
Company System at Cumberland, MD
are being trucked to landfills.
Tentative waste disposal plans by
utilities with contracts for commercial
systems range from dry landfill to
ponding of wetted solids. Eight of the 10
utilities report the following disposal
plans:
Natural clay-lined landfill (1).
PVC-linedlandfilH1),
Wetted transport to landfill (2).
Landfill with fixation (1).
Clay-lined ponding of wetted solids
(1).
Unspecified landfills (2).
-------
Mary E. Kelly and S. A. Shareef are with Radian Corporation, 3024 Pickett Road,
Durham, NC 27705.
Theodore G. Brna is the EPA Project Officer (see below).
The complete report, entitled "Third Survey of Dry SOz Control Systems," f Order
No. PB 81-218 976; Cost: $11.00. subject to change) will be available only
from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
•(l U S. GOVERNMENT PRINTING OFFICE, 1981 — 757-012/7310
10
-------
q =
Is
D E
3 Q
B> =
=•- c
O1
to
o>
oo
t-s
r? ^
5 C
& Z,
C;
r«3
^
o
•D
8
Q)
03
O
O
JT
m > TJ m •
CO O O O -
C0< S. 3 c
wi o 3 2
3 0) I
3
0)
Q
0)
$
39
0)
CD
------- |