United States
Environmental Protection
Agency
Industrial Environmental
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-84-045 May 1984
Project Summary
Pilot-Scale Parametric Testing of
Spray Dryer S02 Scrubber for
Low-to-Moderate Sulfur Coal
Utility Applications
N.J. Stevens, G.B. Manavizadeh, G.W. Taylor,
and M.J. Widico
A comprehensive dry SO, scrubbing
test program was conducted which in-
volved an in-depth field pilot study at the
Comanche Station of Public Service
Company of Colorado. The program in-
vestigated the effects of a number of
process variables on SO, removal. The
ranges of process variables tested dur-
ing the program were: inlet flue gas SO,
concentration, 185-2150 ppm; stoichio-
metric ratio, 0.5-6.4 moles lime/mole
SO,; recycle ratio, 0-4.3 Ib* recycle
solids/lb fresh lime; inlet flue gas
temperature, 226-340°F; spray dryer flue
gas outlet temperature, 128-210°F; and
fabric filter temperature, 117-200°F.
In the spray dryer, stoichiometric
ratio, flue gas temperature approach to
adiabatic saturation, and temperature
drop across the spray dryer significantly
influenced SO, removal. In the fabric
filter, stoichiometric ratio and tempera-
ture approach to adiabatic saturation
controlled SO, removal. Recycling of
flue gas desulfurization (FGD)/flyash
product solids enhanced SO, removal
over that of lime-only once-through
operation. In the absence of fresh lime,
recycle solids, and flyash solids in
separate tests each produced about 20
percent SO, removal at a stoichiometry
of 1 mole alkali/mole SO,. Over the
range investigated (30 to over 90 per-
cent), SO, removal correlated well with
the key spray dryer and fabric filter pro-
cess parameters.
1 EPA policy is to express all measurements in Agency
documents in metric units. This project summary uses
English units to improve clarity of presentation. Con-
version factors are provided at the end of this summary.
The final phase of the field test pro-
gram consisted of a continuous demon-
stration of dry SO, scrubbing tech-
nology. The 5-day continuous run
demonstrated that the spray dryer/
fabric filter system can achieve the 70
percent SO, removal level required to
meet the New Source Performance
Standards for low sulfur coal.
Pilot test results related to stoichio-
metry, recycle ratio, and unit operating
temperatures provided the basis for a
technoeconomic evaluation that
showed that a spray dryer SO, removal
system is less costly than limestone wet
FGD/particulate control systems (fabric
filter/limestone scrubber, ESP/lime-
stone scrubber) for coal sulfur levels up
to about 1.5-1.8 wt percent.
Dry SO, scrubbing solid waste char-
acteristics also were evaluated. Spray
dryer product solids are coarser than
fabric filter solids (30-35 \an versus about
10 urn mean diameter). Product solids
chemical compositions from the two
sources are similar, but spray dryer
solids contain higher concentrations of
unused reagent and fabric filter solids
contain increased percentages of FGD
products. Curing for 1 month at 72°F and
about 100 percent relative humidity in-
creased the cohesive strength of the
waste solids by a factor of 3-6. Product
solids leachate heavy metal contents are
significantly less than hazardous waste
maximum allowable levels.
This Project Summary was developed
by EPA's Industrial Environmental Re-
search 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
Dry S02 scrubbing has emerged as a new
technology and appears to be a cost-
effective desulfurization process for low-
sulfur western coals. Contracts for at least
10 first generation dry scrubbing systems
already have been awarded, and many addi-
tional projects are being evaluated. The at-
tractiveness of the dry S02 scrubbing system
compared to wet FGD lies in its physical
simplicity, moderate pressure drop, very low
water use, reduced reheat requirements, and
the dry condition of the reduced volume of
waste solids produced.
The first example in the U.S. of dry S02
scrubber testing using a spray dryer for utility
coal-fired boiler application occurred in the
early 1970s. A spray dryer was used to
remove S02 with an aqueous sodium car-
bonate reagent. However, major spray dryer
S02 scrubbing test activity began only as
recently as 1977. At this time, several com-
panies participated in spray drying and
baghouse pilot testing at the Leland Olds
Station of Basic Electric Power Cooperative.
Sodium reagents were tested initially, but
potential disposal problems and high costs
turned the investigations toward other
alkaline compounds. Additional testing in-
dicated that slaked lime held the most pro-
mise as an economical reagent for dry SO2
scrubbing.
Pilot tests at the Hoot Lake Station of
Otter Tail Power Company demonstrated
greater than 90 percent overall S02 removal
at a stoichiometric ratio of: 2.0-3.0 moles
lime/mole SO2, using lime-only operation;
and 1.0-1.5 moles lime/mole S02, with re-
cycle. The study also identified the impor-
tance of operating the spray dryer near the
adiabatic saturation temperature to enhance
S02 removal. The Hoot Lake results also
showed that an alkaline flyash, especially
under recycle operation, contributes
significantly to S02 removal.
Field and laboratory pilot tests utilizing a
horizontal flow reactor and dual fluid
atomization identified several variables that
affect S02 removal. They include stoichio-
metry, approach to adiabatic saturation
temperature, inlet gas temperature to the
system, and flyash alkalinity. Lime stoichio-
metry was the primary correlating variable
with S02 removal, but flyash alkalinity also
played a strong role. S02 removal using
flyash alone was 15-65 percent with highly
alkaline Laramie River flyash.
In 1981-82, parametric pilot tests at the
Martin Drake Station of the City of Colorado
Springs showed the importance of operating
near the adiabatic saturation temperature on
S02 removal, but found that in|et gas tem-
perature to the spray dryer had a negligible
effect. Using lime-only operation, S02
removal was considerably reduced at 2000
and 2500 ppm inlet S02 concentrations from
that at 1000 and 1500 ppm S02. Fabric filter
S02 removal in the Martin Drake tests was
strongly affected by the approach to
adiabatic saturation temperature and was
also a function of stoichiometric ratio.
The present program examines the sys-
tematic application of dry SO2 scrubbing
technology to treat utility flue gas using a
spray dryer and fabric filter. A pilot test pro-
gram at the Comanche Station of Public Ser-
vice Company of Colorado investigates the
effect of key process variables and recycled
solids on S02 removal. Study of the fabric
filter variables on S02 removal is confined
to the effects that occur simultaneously with
the testing of the spray dryer. Also, the pre-
sent investigation does not address the effi-
ciency of particulate removal in the dry
scrubbing system.
The final phase of the field test program
is a continuous demonstration of spray
dryer/fabric filter technology. The
demonstration run is to verify that S02
removal can be achieved on a sustained basis
to meet the New Source Performance Stan-
dards (NSPS) for utility boilers operating on
flue gas generated from low sulfur fuels.
Field test results are utilized in a techno-
economic evaluation to establish the areas
where dry FGD technology may be applied
economically.
The characteristics of the solid wastes pro-
duced from the dry S02 scrubbing system
Bypass
are also evaluated. Materials generated at the
pilot test site are examined at the CES
laboratory facilities. This limited scope pro-
gram of testing and evaluation is to deter-
mine 1} the basic composition of the waste
products, and 2) the soil-mechanical and
leaching properties of the FGD solids/flyash
mixtures, to determine their suitability for
landfill disposal or reuse. Solid samples are
also analyzed for trace heavy metals to pro-
vide hazardous waste information.
Pilot System Description
The pilot test system, Figure 1, is designed
to treat up to 10,000 acfm (nominal) of flue
gas. The system consists of a spray dryer to
remove S02 followed by a fabric filter to col-
lect dry FGD solids and flyash. An induced
draft fan moves the flue gas through the
system and a second "reverse air" fan is
used to clean the fabric filter. Feed tanks and
metering pumps supply reagent to the
system. An S02 tank and delivery system
provide additional SO2 to the flue gas for
tests at higher inlet flue gas S02 concen-
trations.
Spray Dryer
The pilot spray dryer is 8 ft in diameter,
35 ft high, and equipped with a variable-
speed rotary-disc atomizer. The atomizer
used in this program has a titanium body disc
with silicon carbide ports around the
periphery. The bottom plate of the disc is
coated with aluminum oxide for protection
against slurry abrasion.
Dirty flue gas containng S02 and flyash
enters the top of the spray dryer where it is
intimately contacted with finely atomized
lime slurry. The intimate contact and large
Bypass
Fabric Filter
Compartments
I
SOi
Injection
Reagent
Preparation
Figure 1. Dry SOt scrubbing pilot flow schematic.
\ / \
ml H
J. "^ A JL
| | | | j | I.D. Fan Stack
—I > U To Ash
Disposal
-------
interfacial area of the spray dryer result in
very rapid S02 absorption by the lime slurry.
Most of the S02 removal in the overall
system occurs in the spray dryer. Before
leaving the dryer, the solids approach com-
plete dryness. In the pilot unit, coarse solids
settle in the conical bottom of the dryer and
are discharged through a rotary valve to
receiving drums. The scrubbed flue gas con-
taining finer particles leaves the dryer
through a side port and flows to the fabric
filter.
Fabric Filter
The fabric filter is 10 x 15 x 55 ft and is
operated with two compartments each
designed to process about 5,000 cf m of flue
gas. Commercial Teflon-coated fiberglass
bags were used in the fabric filter. Each
fabric filter bag is 12 in. in diameter, 30 ft
high, and contains about 94 ft2 of bag sur-
face. Twenty to 24 bags (10 to 12 per com-
partment) were used in various tests during
the pilot program.
In the fabric filter, flyash and FGD solid
paniculate are removed from the flue gas
and additional S02 is removed. From the
fabric filter, the flue gas flows to the induced
draft fan and then to the stack. Flue gas from
the spray dryer continuously enters the bot-
tom of the fabric filter unit and leaves from
the top.
To limit pressure drop across the filter
bags from the accumulation of collected
solids, a flow of air periodically is passed
through the bags in the reverse direction for
a short period of time (1 to 2 min/hr). The
"reverse air" flow dislodges most of the
deposited solids from the bag surface; they
drop into the collection hoppers, from where
they are discharged through rotary valves.
During the brief bag cleaning period, flue gas
is bypassed around the fabric filter. Valves
are operated from the control room either
manually or automatically.
Pilot Test Program
The pilot test program consists of process
variable parametric studies and two con-
tinuous process demonstration runs (one
short- and the other long-term).
Process Variable Tests
The process variable tests were empirical
investigations designed primarily to identify
the dominant variables that affect SO2
removal and to establish their relative impor-
tance, rather than to determine why they are
important or the underlying mechanisms in-
volved. The parametric studies were carried
out using two operating modes: (1) process
variable tests using lime reagent in once-
through operation with no product solids
recycle; and (2) process variable tests incor-
porating recycled solids.
The process variables considered relevant
to S02 removal were studied over the ranges
of conditions shown in Table 1.
Short-Term Process
Demonstration
A continuous process demonstration run
was conducted to verify that S02 removal
levels can be achieved on a sustained basis
to meet New Source Performance Stan-
dards. Specific operating conditions for the
demonstration were selected, based on
results of the process variable tests. Dry
scrubbing system operability and control as
well as S02 performance were observed over
the course of the run.
Conclusions of Major Process
Variable Studies
Spray Dryer S02 Removal
1. Parameters that dominate spray dryer
S02 removal performance are: stoichio-
metric ratio, SR; approach to adiabatic
saturation temperature, ATAS/SD:
temperature drop across the spray
dryer, ATSO; and recycle ratio, RR.
2. Spray dryer S02 removal increases
directly with fresh lime stoichiometry.
At lower stoichiometries, SO2 removal
in the spray dryer is quite sensitive to
stoichiometric ratio. At higher stoichio-
metries, S02 removal levels off and
stoichiometric ratio has much less ef-
fect on spray dryer S02 removal.
3. S02 removal efficiency in the spray
dryer increases as the flue gas temper-
ature approaches the adiabatic satura-
tion temperature.
4. S02 removal in the spray dryer in-
creases as the temperature drop across
the spray dryer is increased. Spray dryer
temperature drop directly reflects the
liquid-to-gas ratio and the quantity of
water fed to the dryer.
5. Spray dryer solids moisture content in-
creases as the flue gas temperature ap-
Table 1. Range of Pilot Test Variables
Variable
proaches the adiabatic saturation
temperature.
6. The optimum spray dryer operating
temperature strikes a balance between
high SO2 removal and smooth trouble-
free discharge and handling of
moisture-laden product solids.
7. Spray dryer S02 removal is enhanced
as the amount of recycle material (re-
cycle ratio) increases. S02 removal in-
creases as recycle ratio increases up to
a value of about 2.5 Ib recycle solids/Ib
fresh lime, where it levels off, indicating
no further benefit to recycling additional
solids.
8. Spray dryer flue gas residence time of
8-16 sec and spray dryer inlet S02 con-
centration of 185-2150 ppm have very
little effect on S02 removal.
Fabric Filter S02 Removal
I.The process parameters that
significantly affect fabric filter S02
removal are: stoichiometric ratio, ap-
proach to adiabatic saturation temper-
ature, and recycle ratio.
2. S02 removal in the fabric filter increases
proportionately with increasing fresh
lime stoichiometry.
3. S02 removal in the fabric filter increases
as the flue gas temperature approaches
the adiabatic saturation temperature.
4. Product solids recycle enhances S02
removal across the fabric filter.
5. Air-to-cloth ratio and fabric filter inlet
S02 concentration have negligible ef-
fects on SO2 removal in the fabric filter.
Overall S02 Removal
1. The dry scrubbing system can attain the
70-90 percent (see main report) S02
removal required by the New Source
Performance Standards (NSPS). Lime
stoichiometry is minimized by operating
with recycle at high flue gas inlet
temperature to the spray dryer, low
spray dryer outlet temperature, and low
fabric filter inlet temperature.
Minimum
Maximum
Stoichiometric Ratio, moles CalOHh/mole SO2 in
SO Met Flue Gas Temperature, °F
SO Outlet Flue Gas Temperature, °F
Fabric Filter Gas Temperature, °F
Inlet SO2 Concentration, ppmv
Atomizer Disc Diameter, in.
Atomizer Disc Speed, rpm
Inlet Flue Gas Rate, acfm
Fabric Filter Air-to-C/oth Ratio, ft/min.
Lime Slurry Feed Concentration, wt %
Recycle Slurry Solids Concentration, wt %
Recycle Ratio, Ib recycle solids/lb fresh lime
0.5
226
128
117
185
7%
10,600
3,000
1.3
3
10
0
6.4
340
210
200
2, ISO
8%
13,920
7,500
2.8
25
53
4.3
-------
2. The spray dryer is the primary S02 con-
trol unit in the system. The fabric filter
is the primary participate control unit.
In the present study, more than 75 per-
cent of the total S02 was removed in
the spray dryer, and considerably more
than half of the total solids were col-
lected in the fabric filter.
Long-Term Process
Demonstration
A 5-day, 120-hour, continuous process
demonstration run was successfully con-
ducted without interruption and with negligi-
ble operating problems.
The continuous run demonstrated that the
spray dryer/fabric filter system can achieve
the 70 percent S02 removal level required to
meet the New Source Performance Stan-
dards for low sulfur coal (Figure 2).
Waste Characterization Studies
As part of the dry S02 scrubbing test pro-
gram, a brief study was undertaken to char-
acterize the solids wastes produced by the
spray dryer/fabric filter dry scrubbing pro-
cess. Only a modest amount of information
has been published at the present time on
the nature of the waste solids or their
disposal characteristics.
The FGD solids/flyash samples were col-
lected at the Comanche test site during the
pilot runs and sent to CES laboratory
facilities for the characterization. To generate
the samples, the pilot unit was operated
steady-state at carefully chosen levels of
stoichiometry and atomization until about
100 Ib of representative waste was collected.
A wide spectrum of operating conditions
is represented in the runs sampled. The
operating conditions for each run in which
samples were taken are summarized in Table
2. In most cases, a blend of 70 percent
baghouse solids and 30 percent spray dryer
solids (by weight) was tested. The 70/30 split
is normal for typical operating conditions,
such as those given in Table 3. In three
cases, only fabric filter solids were tested;
in one other run, only spray dryer solids were
evaluated.
Conclusions of Waste
Characterization
1. Optimum compacted densities of 78-90
Ib/ft3 are obtained at dry waste solids
moisture contents of 25-35 wt percent
H20. Dry scrubbing waste product
compacted densities are lower, and cor-
responding moisture contents are
higher than values reported for
sludge/flyash blends from wet FGD
systems (115-120 Ib/ft3 and 15-20 wt
percent H20).
2. The cohesive strength and angle of in-
ternal friction of waste products from
the lime-only runs are 40-50 percent
greater than for the waste products
from recycle runs. Cohesive strength in-
creases 3- to 10-fold and angle of in-
ternal friction increases 30-50 percent
upon curing of waste solids for approx-
imately 1 month at 72 °F and about 100
percent relative humidity (Table 3).
3. Permeability values for uncured dry
scrubbing product solids are of the
order of 10"5 cm/sec. For samples cured
approximately 1 month at 72°F and 100
percent relative humidity, permeability
is 10-MO-7 cm/sec (Table 3).
4. The total dissolved solids content of the
leachate from the dry scrubbing waste
products generally exceeds 1200 ppm;
most of it is attributable to CaS04.
Technoeconomic Study
A technoeconomic study was performed
to compare a dry S02 scrubbing system us-
ing lime reagent with wet FGD systems us-
ing limestone. The S02/particulate pollution
control systems evaluated in this study in-
12/12/80
12/13/80
12/14/80
12/15/80
12/16/80
ffU
7n
eo
S
"5
O
1 40
Overall SOi 1
8
20
1O
> 1 1
f • Target Value
* 1 * 0lr •
• » , •
q • Overall % SOz Removal
-
• • •
1 1 1 1 1 1 1
l
•
Test Conditions
SOz IN = 480-520 ppm
AT — OC OJ7OC
**' AS/SD *• -tO-oO r
ArAS/FF=45-55°/:
Overall Stoichiometric Ratic
\ i 1
.
"*"
-
-
f. -
1
to
20
30
40
50
60 70
Elapsed Time, hr
80
90
100
110
120
130
Figure 2. Process demonstration test results—overall SO2 removal and overall Stoichiometric ratio.
4
O
to
•8
I
I
Q
5
.e>
I
.
I
5
to
-------
eluded: a spray dryer followed by a fabric
filter {SD/FF), a fabric filter followed by a
single-loop wet scrubber (FF/LS), and an
electrostatic precipitatof followed by a single-
loop wet scrubber (ESP/LS).
Capital investment and annual operating
costs were estimated and compared for the
three pollution control systems studied. In
addition, sensitivity analysis clarified the ef-
fect of system size (MW) and percent S in
the coal on unit capital and unit operating
costs. The effects of percent S in the coal
and reagent price on lime reagent costs were
also determined.
Study Design Basis
For the technoeconomic studies, a power
plant in Colorado was used as the basis since
the pilot test work was performed at the
Comanche Station of Public Service of Col-
orado in Pueblo. The Powder River Basin
coal used in this study had a heating value
of 8230 Btu/lb and contained 0.6 percent
sulfur, 5.77 percent ash, and 30.5 percent
water. The S02 and particulate removal ef-
ficiencies were based on New Source Per-
formance Standards of 0.6 Ib SOz/106 Btu
and 0.03 Ib particulate/108 Btu. The system
design basis specified for these conditions
is presented in Table 4.
Economic Basis
The technoeconomic comparisons were
based on estimates of capital and operating
costs for the three S02/particulate control
systems. Capital costs include direct, in-
direct, and contingency costs, but not such
charges as allowance for system start-up and
modification, interest for system construc-
tion, land, working capital, and royalties. The
capital cost estimates for the spray dryer,
fabric filter, and electrostatic precipitator are
based on Research-Cottrell cost estimate in-
formation. Capital cost estimates for the
limestone scrubber are based on TVA values.
Operating costs, including direct and indirect
costs, are presented as first-year annual
revenue requirements. Capital and operating
cost estimates are calculated based on first-
quarter 1981 dollars.
Equipment included in each pollution con-
trol system was divided into functional areas
for cost estimating purposes. The direct cost
of each area was estimated independently.
The total direct investment for each system
is the sum of the six process area direct
costs. Indirect investment was assumed to
be 30 percent of the total direct system in-
vestment. Contingency costs were of the
sum of direct and indirect costs.
Capital Cost Estimates
Comparison of the total capital investment
costs for the three S02/particulate pollution
control systems shows that the dry scrub-
bing system is less costly than the two wet
FGD systems. Table 5 shows that the total
capital investment for a 500 NW spray
dryer/fabric filter system treating flue gas
generated from 0.6 percent S coal is
Table 2. Operating Conditions for Pilot Plant Runs Chosen for Waste Characterization Sampling
Run
No.
220
314
317
405R2
506
624
923
923
923
Sample
Source
Blend
Blend
Blend
Fabric
Filter
Fabric
Filter
Blend
Spray
Dryer
Fabric
Filter
Blend
Inlet
Concen-
tration
ppmv
1350
1900
2150
760
850
780
1380
1380
1380
Over-
all
*?o
OV2
Re-
moval
%
39.4
37.1
43.1
59.3
50.9
39.9
72.2
72.2
72.2
Stoichio-
Ratio
moles lime/
moles SOt in
3.01
2.16
1.05
1.07
2.01
1.04
1.14
1.14
1.14
Recycle
Ratio
Ib recycle
solids/lb
makeup lime
0
0
0
2.27
0
1.06
2.62
2.62
2.62
Spray
Dryer
Tempera-
ture
In-
let
°F
268
264
345
340
258
295
255
255
255
Out-
let
°F
188
183
185
180
178
177
147
147
147
hTAS/SD
°F
71
68
70
50
48
48
22
22
22
Spray
Dryer
Resi-
dence
Time
sec
10.0
10.2
10.4
10.2
9.5
9.9
9.4
9.4
9.4
Table 3. Product So/ids Triaxial Compression and Permeability Test Results
Uncured
Cured3
Permeability
Run
No.
220
314
317
405-R2
BOB
624
923
923
923
Operation
Mode
Lime-Only
Lime-Only
Lime-Only
Recycle
Lime-Only
Recycle
Recycle
Recycle
Recycle
Sample
Source
Blend*
Blend
Blend
Fabric
Filter
Fabric
Filter
Blend
Spray
Dryer
Fabric
Filter
Blend
Cohesion
psi
16.3
18.2
15.8
6.0
14.0
13.0
12.0
12.1
J0.2
Angle of
Internal
Friction
degrees
36
39
37
33
34
32.5
24
21
22
Cohesion
psi
110
_
63
63
34
—
_
-
Angle of
Internal
Friction
degrees
51
_
51
48.5
42
_
_
-
Uncured
cm/sec
0.3 x 10-*
0.4 x 10-*
1.2 x 10-*
0.2 x 1fr*
3.3 x 10-*
8.9 x 10-*
7. 1 x 10-*
9.3 x 70-6
5.3 x 10-*
Cured
cm/sec
3.5 x 70-«c
_
3.5 x 10+d
0.7 x 10**
0.9 x 10**
—
-
8 Cured for 28 days at 72°F and about 100 percent relative humidity.
b Blend refers to a physical mixture of 70 wt percent fabric filter solids and 30 wt percent spray dryer solids.
c Cured for 35 days at 72°F and about 100 percent relative humidity.
d Cured for 38 days at 72°F and about 100 percent relative humdity.
-------
estimated at $42,300,000. The fabric filter/
limestone scrubber system cost estimate is
$59,046,000. The ESP/limestone scrubber
system cost estimate is $63,699,000.
The conclusions on capital investment
reached during the study are similar to those
presented by a TVA study. In each study,
the total fixed capital investment for the dry
scrubbing system is about 30 percent less
than for an ESP/limestone scrubber system.
However, the fixed capital investment figures
developed in the present study are 15-20 per-
cent less than the corresponding TVA
estimates. A significant part of the difference
arises because the capital investment esti-
mates of the present study are based on ac-
tual first-quarter 1981 dollars while the TVA
study uses projected mid-1982 dollars. In
view of the ±30 percent error range
associated with study or preliminary esti-
mates, the capital investment costs
generated for these and the TVA evaluations
agree fairly well.
These cost estimates show that dry
system costs are 30-35 percent lower than
wet FGD/particulate systems costs for low
sulfur coal applications. Unit costs for the
three systems at a 500 MW size are
$84.6/kW for the spray dryer/fabric filter,
$118.1/kW for the fabric filter/limestone
scrubber, and $127.4/kW for the
ESP/limestone scrubber case.
Operating Cost Estimates
Annual operating costs (first-year annual
revenues) were estimated for the three pollu-
tion control cases for a 500 MW power plant
burning 0.6 wt percent S coal and operating
7000 hr/yr. Cost comparisons in Table 6
show that dry scrubbing system operating
costs are about 25-30 percent lower than wet
FGD/particulate control systems costs under
the conditions investigated. All operating
cost items except raw materials, or reagent,
costs are less for the dry scrubber system.
Reagent costs are much higher for dry scrub-
bing than for wet scrubbing primarily
because of the large cost differential be-
tween lime and limestone, $70 vs. $8/ton.
Essentially all of the operating cost difference
between dry and wet scrubbing systems
determined in the present study is at-
tributable to capital investment-related items,
maintenance, and capital charges.
The results of the present study and those
of TVA indicate that dry scrubbing annual
operating costs are 25-30 percent less than
ESP/limestone scrubber costs. Annual
operating cost estimates of this study are
only 70-75 percent of those developed by
TVA. Much of the annual operating cost dif-
ference is attributable to lower capital
charges used in the present study compared
to the TVA work. Lower operating labor
Table 4. Study Design Basis
Item
Design Value
Flue Gas Rate, acfm
SO2 Met Concentration, ppm
Flue Gas Inlet Temperature, °F
Atmospheric Pressure, psi
SO2 Removal Efficiency, %
Paniculate Inlet Content, gr/acf
Particulate Removal Efficiency, %
2,040,000
600
275
12.3
70
2.3
99.7
Table S. Comparison of System Total Capita/ Investments
(500 MW, 0.6% S Coal, 70% SOt Removal)
Total Cost, 1000$ (1981$)
Investment Area
Material Hand/ing
Feed Preparation
Gas Hand/ing
SO2 Absorption
Particulate Removal
Waste Disposal
Total Direct
Investment
Indirect Investment
Contingency
Total Fixed Investment
Unit Cost ($/kW)
Lime Spray Dryer/
Fabric Filter
3,391
850
5,975
7,145
9,230
524
27,115
8,135
7,050
42,300
84.6
Fabric Filter/
Limestone Scrubber
800
1,100
8,554
11,802
10,990
4,604
37,850
11,355
9,841
59,046
118.1
Electrostatic
Precipitator/
Limestone Scrubber
800
1,100
8,554
11,802
13,973
4,604
40,833
12,250
10,616
63,699
127.4
Table 6. Comparison of System Annual Operating Costs
(500 MW, 0.6% S Coal, 70% S02 Removal, 7000hr/yr Operation)
Total Cost, 1000$ 11981$)
Item
Raw Materials
Electricity
Water
Maintenance
Operating Labor
Overhead
Administration
Capital Charges
Total Cost, 1000$
Unit Cost, mil/s/kWh
Lime Spray Dryer
Fabric Filter
1,026
1,820
40
1.654
649
832
65
6,345
12,431
3.6
Fabric Filter/
Limestone Scrubber
225
2,282
73
3,217
774
1,269
77
8,857
16,774
4.8
Electrostatic
Precipitator/
Limestone Scrubber
225
2,487
73
3,062
749
1,274
75
9,555
17,500
5.0
hourly rates and costs of this study also
account for some of the operating cost
difference.
Lime Reagent Cost
For the present study, lime reagent costs
are somewhat more sensitive to increases in'
the coal sulfur content than to raw material
price. Lime costs increase 5-7 times to meet
the reagent requirements as coal sulfur in-
creases from 0.6 to 1.5 wt percent. In this
coal sulfur range, lime reagent costs (at a
price of $70/ton) increase from less than 10
percent of the total operating costs at 0.6
percent to about 30 percent at 1.5 percent
S. Since lime reagent cost is very sensitive
to sulfur level, dry scrubbing enjoys its
greatest economic advantage over wet FGD
systems for low sulfur coal applications. As
coal sulfur content increases, the cost dif-
ference between wet and dry systems de-
creases. The operating cost crossover point
is estimated at somewhat greater than 1.5
percent S coal for the bases used in the pre-
sent study (Figure 3). Note that the spray
dryer/fabric filter operating cost increases
more rapidly as the coal sulfur content in-
creases. This is due to the greater lime re-
quirement (higher stoichiometric ratios) as
the sulfur content increases.
-------
I
5
«
I
Fabric Filter/Wet Scrubber
(Limestone)
ESP/Wet Scrubber
(LSi
Spray Dryer/Fabric Filter
(Lime)
500 MW System
1981 $
0.6
1.0
1.5
1.9
Percent S in Coal
Figure 3. Unit operating cost sensitivity.
Conclusions of Technoeconomic
Study
1. The capital investment required for a
dry S02 scrubbing system (spray dryer/
fabric filter) is 30-35 percent less than
for either of the two wet FGD systems
(fabric filter/limestone scrubber, ESP/
limestone scrubber) based on estimates
for a 500 MW unit burning Powder
River Basin coal with 0.6 percent sulfur
and complying with the NSPS.
2. Annual operating costs (first year an-
nual revenues) for a dry S02 scrubbing
system (spray dryer/fabric filter) are
25-30 percent lower than for either wet
scrubbing system (fabric filter/lime-
stone scrubber, ESP/limestone
scrubber).
3. Reagent costs for a dry scrubbing
system using lime are much higher than
for wet scrubbing systems using lime-
stone. Lime costs 4-5 times as much as
limestone for S02 control of flue gas
generated from 0.6 wt percent sulfur
coal and is a major operating expense
for dry scrubbing systems. Lime
reagent cost is affected by coal sulfur
content, stoichiometric ratio, and raw
materials price.
4. Unit capital costs are only moderately
sensitive to system size (MW) for all
three systems and increase as coal
sulfur content increases (see main
report). Dry scrubbing system costs are
more sensitive to coal sulfur content
than are wet scrubbing system costs.
5. For each of the three desulfurization
processes, unit operating costs de-
crease as system size (MW) increases
but are not very sensitive to size. Unit
operating costs increase as coal sulfur
content increases because of increased
reagent use. Dry scrubbing system
operating costs are more sensitive to
coal sulfur content than are wet scrub-
bing system operating costs because
dry systems use more-expensive lime as
the reagent.
6. As coal sulfur content increases, the
cost advantage of a dry system over a
wet system decreases. The crossover
range is about 1.5-1.8 percent sulfur for
the conditions used in this study.
Conversion Factors
To Convert From
English
cfm
ft
gr/scf
in.
in. H2O
Ib
gal.
gal./1000 ft3
Btu
short ton
°F
To
SI
nWhr
m
kg/m3
m
Pa
kg
m3
liters/m3
joule
tonne
°C
Multiply
By
1.70
0.305
0.00229
0.0254
249
0.454
3.79
0.13
0.252
0.91
5/9(°F-32)
-------
N. J. Stevens, G. B. Manavizadeh, G. W. Taylor, and M. J. Widico are with
Cottrell Environmental Sciences, Inc., Sommerville, NJ 08876.
Theodore G. Brna is the EPA Project Officer (see below).
The complete report, entitled "Pilot-Scale Parametric Testing of Spray Dryer S02
Scrubber for Low-to-Moderate Sulfur Coal Utility Applications," (Order No. PB
84-175 959; Cost: $22.00. subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use S30O
U.S. GOVERNMENT PRINTING OFFICE: 1984-759-102/0943
------- |