-------
clarifed liquor from the dewatering system was returned to the venturi loop or the
spray tower loop. Beginning with Run 863-1A, the returning clarified liquor was
split between the two scrubber loops to control the spray tower slurry solids con-
centration at about 10 percent. This solids level was a compromise between low
solids (below about 7 percent) where scaling and a drop in S02 removal is experi-
enced and high- solids (above about 15 percent) where it becomes more difficult to
keep the mist eliminator clean.
F1 ue Gas Rate - Earlier runs were made at a reduced flue gas flow rate of 25,00.0
acfm (at 300°F) because it was assumed that forced oxidation would reduce .S0£ removal
efficiency in the venturi loop at lower pH to the extent that an overall S0~ re-
moval efficiency of at least 80 percent could not be achieved. This proved not to
be the case. Beginning with Run 862-1A, the flue gas flow rate was increased to
maximum achievable of 35,000 acfm (at 300°F) which corresponds to a spray tower
superficial velocity of 9.4 ft/sec. In this run and subsequent runs at 35,000 acfm,
S02 removal averaged about 85 percent at 2000 to 3000 ppm inlet SO^ concentration.
Spray tower slurry liquor pH - In earlier runs the spray tower inlet slurry liquor
pH was controlled at 8.0. In some runs, especially those at low slurry solids con-
centration as mentioned previously, this level of inlet pH resulted in an outlet
pH approaching 6, causing sulfite scaling.
While the pH drop across the spray tower depends on the SO? removal and the inlet
S0£ concentration, it has been generally observed that sulfite scaling does not
occur if the spray tower outlet pH stays below about 5.5 Therefore, during
Run 863-1A, the inlet pH was adjusted downward slightly to 7.R. In this run and
in subsequent runs, small patches of scale were observed to appear and disappear
in a cyclic manner. This cyclic appearance of scale did not interfere with
scrubber operation. No effect on S02 removal was discernable as a result of the
slight adjustment in inlet pH.
This slight adjustment in pH is significant in that it demonstrates the need for
good pH control in commercial installations and it demonstrates one of several
operating adjustments that can be made to eliminate a scaling problem.
Slurry Level in Oxidation Tank - In Runs 864-1A through 867-1A, the effect of oxi-
dation tank slurry level (and consequently air and slurry residence times) was
explored. In these tests, air was discharged into the bottom of the oxidation tank
through an open 3-inch pipe as previously described (see Figure 3). Major test
22
-------
conditions are listed in Table 1. All runs were nade at an oxidation pH of 5.5.
The effect of the tank level is summarized below:
Oxidation Air Stoichiometry, Percent
Run No. Tank Level, ft atoms 0/mole 50~ absorbed Sulfite Oxidation
864-1A 18 1.8 98
867-1A 14 1.8 98
865-1A 10 2.1 89
866-1A 10 3.8/2.7 98/R1
Oxidation efficiency was hiqh at 18 and 14-ft tank levels but dropped ofV at a
10-ft level. In Run 866-1A, hiqh oxidation efficiency was achieved at a 10-ft tank
level by increasing air Stoichiometry. Part of Run R6R-1A was made at lower air
Stoichiometry with subsequent loss in oxidation efficiency.
These runs demonstrated that 98 percent sulfite oxidation can he achieved at 14
to 18-ft tank levels at an air Stoichiometry of 1.8 atoms 0/mole S02 absorbed.
At a 10-ft tank level an air Stoichiometry approaching 3.8 is required.
It must be pointed out that the oxidation at a 10-ft tank level is not directly
comparable with those at 14 and 18 feet because the top turbine of the agitator
is located at the 11-ft level. In 10-ft slurry level tests, the top turbine is
not in contact with the slurry and a different aqitation pattern results.
Filter cake solids concentration during these tests was about 85 percent when
the oxidation efficiency was 98 percent. In test periods when oxidation efficiency
dropped below 90 percent, the filter cake solids concentration tended toward a lower
range of 80 percent.
Lime Reliability Run - From mid-December 1977 through mid-January 1978, Run R63-1A,
a one-month lime-slurry reliability run, was made with the venturi/spray tower
system in a two-scrubber-loop configuration with forced oxidation in the venturi
scrubber loop. Onstream operation for this run totaled 779 hours (32 days). The
run was designed to demonstrate operating reliability of the scrubber system with
respect to scaling and plugging and to determine if the F.PA New Source Performance
Standards for SOp and particulate emissions could he met.
To simulate variable boiler load, the flue gas flow rate was varied between
18,000 and 35,000 acfm (4.8 and 9.4 ft/sec spray tower superficial gas velocity)
as the boiler load varied between 100 and 150 MW. Flue gas with high fly ash
loading was used. The venturi plug was fixed at a position to give 9 inches H20-
23
-------
pressure drop across the venturi at full 35,000 acfm flue qas flow rate. The
actual venturi pressure drop ranged from 2 to 9 inches HpO. The slurry recircu-
lation rates to the venturi and spray tower were held constant at 600 and 1600
gpm, respectively. The venturi inlet pH was controlled at 5.5. The oxidation
tank level was 18 ft and the oxidation air flow rate was 210 scftn discharqed
through a 3-inch pipe. As previously discussed, the spray tower slurry inlet pH
was adjusted downward from 8.0 to 7.8 to eliminate an observed sulfite scale buildup.
During the run, the scrubber was shut down a total of 57 hours; 46 hours were due to
boiler outages, 7-1/2 hours were for scheduled scrubber inspections, and 3-1/? hours
were unscheduled downtime. This resulted in a scrubber availability of 99.6 percent,
excluding the interruptions due to boiler outage and the scheduled inspections. The*
unscheduled downtime included 2 hours for mist eliminator cleaning and l-l/? hours
for air compressor repair.
Average SOp removal for the entire run was 88 percent at 2950 ppm average inlet SOo
concentration. This corresponds to an average emission of 0.9 Ib S02/MM Rtu, well
within the EPA standard of 1.2 Ib S02/MM Btu. However, due to unusually wide fluc-
tuations in inlet S02 concentration and slow system response time, the S02 emissions
at times exceeded the EPA standard for periods greater than the three hours allowed
by EPA regulations.
The fluctuations in inlet SOo concentration, ranging up to 4700 ppm, resulted from
the wide variety of coals being burned during the 1977-78 coal strike. Normally,
inlet SOo concentration ranges between 2000 and 3000 ppm. These high S02 concentra-
tions were beyond the capacity of the venturi/spray tower system to remove with its
limited slurry recirculation rates (liquid-to-gas ratios of 57 and 21 gal/Macf at
35,000 acfm full gas flow rate in the venturi and spray tower, respectively).
Average particulate loading was 0.046 grain/dry scf corresponding to an average
emission of 0.09 Ib particulate/MM Btu (assuming 30 percent boiler excess air).
Although the EPA standard of 0.10 Ib particulate/MM Btu was not exceeded on the
average, a few measurements exceeded this value.
Sulfite oxidation averaged 97 percent during the run with the air stoichiometric
ratio varying between 1.4 and 2.8 atoms 0/mole S02 absorbed. The filter cake was
excellent throughout the run with solids concentration averaging 85 percent. Lime
utilization was 90 percent in the spray tower and 98 percent overall, reflecting
the high utilization to be expected in a two-scrubber-loop system.
At the first scheduled inspection after 160 operating hours, the mist eliminator
was found to be 15 percent restricted by solids. After a review of the history
of the mist eliminator exposure, the restriction was attributed to excess calcium
carbonate from the previous limestone run (limestone stoichmetric ratio of 1.65
24
-------
in the spray tower) and a failure to activate the intermittent underwash for the
first eiqht hours of the reliability run. At the beginning of the reliability run,
the mist eliminator underwash had been changed from continuous with diluted clari-
fied liquor (needed for the limestone run conditions) to intermittent with makeup
water (1.5 gpm/ft? for 6 minutes every 4 hours - satisfactory for lime runs). The
mist eliminator was cleaned and the run was continued. This mishap was frustrating
in that it broke a record of 41R3 hours of operation under widely varyinq conditions
without cleaning the mist eliminator.
At subsequent inspections at 399 operating hours and at the end of the run, the
mist eliminator was entirely clean.
In summary, the operatinq reliability of the venturi/snray tower system in a
two-scrubber-loop configuration with forced oxidation in lime slurry service
has been demonstrated with a system availability of 99.6 percent. However, under
the conditions selected, the system was unable to continually meet F.PA New Source
Performance Standards for S02 and particulate emissions even thouqh the average
emissions for the run met the standards.
UPDATE ON TWO-SCRUBRER-LOOP TEST RESULTS WITH LIMESTONE SLURRY
Since the last report, 6 runs have been made with limestone slurry in a two-scrubber-
loop configuration with forced oxidation. Results of the tests are summarized in
Table 2. In these tests, several operatinq problems were solved and operational
reliability was established. Filter cake solids concentration stayed consistently
above 85 percent throughout the tests. The following discussion highlights the
new information developed from the recent runs.
Flue Gas Rate - As with lime testing, earlier limestone runs were made at a reduced
flue gas flow rate under the assumption that, with forced oxidation, high S0£
removal (in the range of 85 percent) could not be achieved at full gas rate. Be-
ginning with Run 815-1A, the flue gas flow rate was increased from the reduced rate
of 25,000 acfm (at 300°F) to the maximum rate of 35,000 acfm. S02 removal for this
run was 86 percent under the test conditions listed in Table ?., which is about 5
percentage points below S02 removal achieved on an identical run at the lower flue
gas flow rate. All subsequent runs were made at the higher flue gas flow rate.
System Control - In the limestone tests with two scrubber loops, the control
philosophy was to hold the venturi inlet pH (oxidation tank pH) at 5.5 by adjust-
ing the limestone slurry feed rate to the spray tower effluent hold tank. Control
in this manner proved to be difficult and wide fluctuations were experienced in both
pH and limestone st.oichiometry. For example, in Run 815-1A, the venturi inlet pH
varied between 4.9 and 6.3 with corresponding fluctuations in the limestone stoichio-
metric ratios of 1.1 to 1.9 in the venturi loop and 1.2 to 2.1 in the spray tower
loop.
25
-------
Table 2
RESULTS OF FORCED OXIDATION TESTS WITH TWO SCRUBBER LOOPS
ON THE VENTURI/SPRAY TOWER SYSTEM USING LIMESTONE SLURRY
Major Test Conditions
Fly ash loading
Flue gas rate, acftn 9 300°F
Slurry rate to venturl, gpra
Slurry rate to spray tower, gpm
Venturi percent solids redrculated (controlled)
Residence times, m1n: Oxidation tank
Desupersaturation tank
Spray tower EHT
Venturl Inlet (oxidation tank) pH (controlled)
Spray tower limestone stolchiometrlc ratio
Venturl pressure drop, In. H20
Oxidation tank level, ft
Air rate to oxidation tank, scfm
Clarified liquor returned to '3'
Selected Results
Onstream hours
Percent SO, removal
Inlet SO, concentration, ppm
Spray tower percent sol Ids redrculated
Spray tower Inlet pH
Spray tower limestone stolchiometrlc ratio
Spray tower inlet liquor gypsum saturation, X
Spray tower sulflte oxidation, X
Overall sulflte oxidation, t
Overall limestone utilization, X
Venturl Inlet liquor gypsum saturation, X
Venturl Inlet liquor sulflte concentration, ppre
A1r stolchionetry, atoms 0/roole S02 absorbed
Filter cake solids, wtl'4'
Mist eliminator restriction, t'5'
815-1A
Low
35 ,000
600
1400
15
8.8
4.7
13.4
5.5
-
9
14
210(1)
S.T
306
86
2500
8.4
5.85
1.98
105
26
96
67
105
40
1.70
87
3
816-1A
Low
35,000
600
1400
15
11.3
4.7
13.4
5.5
-
9
18
210(D
S.T.
142
86
2350
7.7
5.75
1.68
105
27
98
83
100
25
1.80
86
3
817-1A
High
35,000
600
1400
15
11.3
4.7
16.8
5.5
-
9
18
210<2>
Vent
188
83
2500
8.9
S.9
1.60
100
21
97
82
105
25
1.75
86
1
818-1A
High
35 ,000
600
1600
15
11.3
4.7
14.7
5.5
-
7.5-9
18
210*2^
Vent
141
86
2550
9.6
5.9
1.64
100
19
98
81
105
25
1.70
86
2
819-1A
High
Variable
600
1600
15
11.3
4.7
14.7
5.5
-
*9
18
210^2^
Vent
840
86
2950
10.0
5.85
1.65
100
21
98
81
100
25
1.45-2.80
87
3
819-18
High
Variable
600
1600
15
11.3
4.7
14.7
-
1.6
tg
18
210<2>
Vent
126
85
3000
9V6
5.9
1.65
110
19
98
83
105
25
1.45-2.80
86
3
NJ
dotes:
1 A1r discharged downward through sparger ring with 40-% Inch diameter holes about 3 Inches from tank bottom.
Air discharged downward through 3-Inch diameter pipe with an open elbow at center of oxidation tank about 3 Inches from tank bottom.
Spray tower loop (effluent hold tank) or venturl 3oop (oxidation tank).
, Ctarlfler and filter 1n series used for solids dewaterlng 1n all runs.
5) Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gpm/ft2. Sequential top wash with makeup water with one of
6 nozzles on at 0.53 gpm/ft2 for 4 minutes every 80 Minutes.
-------
In Run 815-1A, the oxidation tank level was 14 feet, which was satisfactory for
forced oxidation (96 percent oxidation at an air stoichiometric ratio of 1.7
atoms 0/mole S0£ absorbed). In Run 816-1A, the fluctuation in venturi inlet pH
was reduced to a range of 5.2 to 5.8 with corresponding reduction in fluctuation
in limestone stoichiometry by increasing the oxidation tank level to the maximum
of 18 feet and-thus increasing hold tank residence time.
In Run 819-1B, the control philosophy was changed. In this run, the limestone
stoichiometry in the spray tower was controlled at 1.6 moles calcium per mole SO?
absorbed and the venturi inlet pH was allowed to vary. With direct control on the
spray tower stoichiometry, the fluctuation in venturi inlet pH was 5.2 to 5.8, no
greater than in the previous runs with venturi inlet pH control.
Based on these runs, control of limestone stoichiometry in the primary scrubbinq
loop (spray tower) is recommended over control of pH in the oxidation loop
(venturi). v
Mist Eliminator - In previous runs with limestone slurry and high fly ash loadings
(Runs 805-1A through 808-1A), problems with mist eliminator plugging occured. In
these runs, the spray tower solids concentration was maintained at 15 percent, which
required that the clarified liquor from the solids dewatering system be returned to
the venturi loop and which allowed only enough makeup water in the spray tower system
for an intermittent mist eliminator underside wash. Such a wash was inadequate at
the limestone utilizations experienced in the spray tower (60 to 70 percent) and
the mist eliminator plugged within a matter of days.
Beginning with Run 817-1A, the mist eliminator was washed continuously at 0.4 qpm/
ft* with clarified water diluted with available makeup water. Excess clarified
water was returned to the venturi loop. This wash scheme (coupled with a sequen-
tial top wash - see Table 2) proved adequate and the mist eliminator no longer
plugged.
The continuous wash diluted the spray tower solids concentration to about 9 percent.
At 9 percent solids concentration, SOg removal dropped a few percentage points to
83 percent at 2500 ppm inlet S02 concentration.
In Run 818-1A, the slurry recirculation rate in the spray tower loop was increased
from 1400 gpm to the maximum controlled rate of 1600 gpm. With this modification
S02 removal was increased to 86 percent at 2550 ppm inlet concentration.
Limestone Reliability Run - During November 1977, Run 819-1A, a one-month limestone
slurry reliability run, was made with a two-scrubber-loop configuration on the
venturi/spray tower system and with forced oxidation in the venturi loop. This
run operated for a total of 840 hours (35 days). As with the lime reliabilty
run, the run was designed to demonstrate operating reliability of the scrubber
system and to determine if the EPA New Source Performance Standards for SOp and
particulate emission could be met.
27
-------
Flue gas and slurry flow rates were the same as with the lime reliability run.
Flue gas with high fly ash loading was varied in rate between 18,000 and
35,000 acfm (at SOOOf) to follow the boiler load. The venturi plug was fixed to
give 9 inches HgO pressure drop at 35,000 acfm flue gas rate. Slurry recircula-
tion rates were held constant at 600 gpm and 1600 gpm in the venturi and spray
tower loops respectively. The venturi inlet pH was controlled at 5.5 by control-
ling the limestone feed rate to the spray tower hold tank.* The oxidation tank
was maintained with an 18-ft slurry level and an air flow rate of 210 scfm dis-
charged through a 3-inch pipe.
During the run, the scrubber was shut down for 18 hours due to a boiler outage,
5 hours total for weekly inspections, plus 3-1/2 hours of unscheduled downtime'for
a total of 26-1/2 hours. Based on unscheduled downtime, the scrubber system had
an availability of 99.6 percent. The unscheduled downtime included 3 hours to
clean a partially plugged slurry return pipe from the venturi to the oxidation
tank and 1/2 hour to clean a plugged mist eliminator nozzle.
The plugged mist eliminator nozzle was discovered after 391 hours of operation.
The mist eliminator in the vicinity of the slurry nozzle was severely restricted
by slurry solids (7 percent overall mist eliminator restriction). The nozzle
was cleaned but the mist eliminator was not disturbed. By the end of the run
(840 hours), the mist eliminator restriction had dropped to 3 percent, demon-
strating that a restricted area can be self cleaning.
For the entire run, the SO? removal averaged 86 percent at 2950 ppm average
inlet SO? concentration. This removal efficiency corresponds to an average emis-
sion of 1.0 Ib S02/MM Btu which meets the EPA New Source Performance Standard of
1.2 Ib S02/MM Btu. However, as with the lime reliability run, fluctuations to
unusually high inlet SOg concentrations were experienced and the standard was
at times exceeded for periods greater than the three hours allowed by EPA regu-
lations.
The outlet particulate loading ranged from 0.021 to 0.063 grain/dry scf with an
average of 0.042 grain/dry scf. Assuming 30 percent excess air to the boiler
the average outlet particulate loading corresponds to 0.08 Ib/MM Btu which meets
the EPA New Source Performance Standard of 0.1 Ib/MM Btu. However, a few of the
outlet particulate loading measurements exceeded the standard.
Sulfite oxidation averaged 98 percent during the run with the air stoichiometric
ratio varying between 1.4 and 2.8 atoms 0/mole SO? absorbed. The filter cake
solids concentration averaged 87 percent. Overall limestone utilization was 81
percent while the spray tower limestone utilization was 61 percent, again demon-
strating the advantage of a two-scrubber-loop system in achieving high alkali
utilization.
* As previously discussed, this mode of control was later changed to Sto1ch1ometr1e
ratio control in the spray tower (Run 819-1B).
28
-------
To summarize, the operating reliability of the venturi/spray tower system in a
two-scrubber-loop configuration with forced oxidation in limestone slurry service
has been demonstrated with a system availability of 99.6 percent. However,
under the conditions selected, the system was unable to continually meet EPA New
Source Performance Standards for SO? and particulate emissions even though the
average emissions for the run met the standards.
TWO-SCRUBBER-LOOP TEST RESULTS USING LIMESTONE SLURRY WITH ADDED MAGNESIUM OXIDE
Beginning in March 1978, a series of six runs were made in which magnesium oxide
was added to the spray tower hold tank along with the limestone slurry. The pri-
mary purpose of the magnesium oxide addition was to enhance S0£ removal efficiency
in the spray lower loop by increasing the sulfite ion concentration in the liquor
for S02 scrubbing. In a two-scrubber-loop configuration as shown in Figure 2,
the magnesium ion concentration in the venturi loop is higher than that in the
spray tower loop because of the water loss in humidifying the flue gas in the
venturi loop. But because the sulfite ion is converted into nonscrubbing sulfate
ion by forced oxidation, the higher magnesium ion concentration in the venturi
loop does not enhance $62 removal in the venturi loop. The secondary purpose of
the magnesium oxide addition was to determine whether the presence of magnesium
ion had an effect on oxidation efficiency.
Typical operating conditions and results of these tests are summarized in Table 3.
The expected enhancement of S02 removal was achieved and oxidation efficiency was,
if anything, improved. Thus, magnesium oxide addition is compatible with a two-
scrubber-loop forced oxidation system.
S02 Removal - Run 820-1A was made under identical conditions to Run 818-1A
(Table Z) except for the addition of magnesium oxide. Effective magnesium
ion concentration* averaged 5150 ppm in the spray tower. The anticipated removal
enhancement was achieved as the average S02 removal was 96 percent at 2250 ppm
average inlet S02 concentration compared with 86 percent removal at ?550 inlet
ppm for Run 818-1A.
The spray tower inlet slurry liquor was 100 percent saturated in gypsum and no
scale was observed. This condition was typical of all the tests in the limestone/
magnesium oxide, forced-oxidation test block.
* Effective magnesium ion concentration is defined as the total magnesium ion
minus that magnesium ion concentration equivalent to total chlorides. Magnesium
chloride has no effect on S02 removal.
29
-------
OJ
o
Table 3
RESULTS OF FORCED OXIDATION TESTS WITH TWO SCRUBBER LOOPS
ON THE VENTURI/SPRAY TOWER SYSTEM USING LIMESTONE SLURRY WITH ADDED MAGNESIUM OXIDE
Ma.lor Test Conditions
Fly ash loading
Flue gas rate, acfm 9 300°F
Slurry rate to venturl , gpm
Slurry rate to spray tower, gpm
Venturl percent solids redrculated (controlled)
Residence times, m1n: Oxidation tank
Desupersaturatlon tank
Spray tower EHT
Venturl Inlet (oxidation tank) pH (controlled)
Spray tower limestone stolchiometrlc ratio (based on solids)
Effective Mg** concentration (S.T. loop), ppm
Venturl pressure drop, 1n. HjO
Oxidation tank level, ft
Air rate to oxidation tank, scfm' '
Clarified liquor returned to "'
Selected Results
Onstream hours
Percent SO, removal
Inlet S02 concentration, ppm
Spray tower percent sol Ids redrculated
Spray tower Inlet pH
Spray tower limestone stoichiometrlc ratio (based on total slurry)
Spray tower Inlet liquor gypsun saturation, X
Spray tower sulflte oxidation, X
Effective Hg concentration (S.T. loop), ppm
Overall sulf1te oxidation, X
Overall limestone utilization, X
Venturl Inlet liquor gypsum saturation, t
Venturl Inlet liquor sulflte concentration, ppm
A1r stolchlometry, atom 0/mole SO., absorbed
Filter cake solids, wtX (3)
M1st eliminator restriction, x'4'
820- 1A
High
35,000
600
1600
15
11.3
4.7
14.7
5.5
-
5000
9
18
210
Vent
462
96
2250
6.0
6.05
1.16
100
30
5150
98
92
130
50
1.70
85
-
820- IB
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6<6)
5000
9
18
150
Vent
137
94
2500
8.3
5.9
1.28
105
17
4985
92
90
130
950
1.10
82
0
820- 1C
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
0
Vent
134
91
2750
10.5
5.9
1.52
90
20
4700
36
82
145
5585
0
63
0
821-1A
High
35,000
600
0
15
11.3
4.7
-
5.5
-
5000
9
18
210
Vent
(7)
0
822- 1A
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
210
Vent
232
91
2750
8.0
5.75
1.55
100
21
4985
97
79
125
735
1.45
85
-
822- IB
High
35,000
600
1600<5>
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
210
Vent
85
90
2400
5.6
5.55
1.21
110
23
4845
98
93
130
410
1.70
B5
0
Notes:
Air discharged downward through 3-Inch diameter pipe with an open elbow at center of oxidation tank about 3 Inches from tank bottom.
Venturl loop (oxidation tank).
Clarlfler and filter used for solids dewaterlng In all runs.
Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gp»/ft2. Sequential top wash with makeup water with one of
6 nozzles on at 0.53 gpn/ft2 for 4 minutes every 80 minutes.
Spray tower turned off for 30 minutes every 8 hours to obtain SO; removal with venturl alone. Venturl SO? removal averaged 29X.
In runs with control by spray tower stolchjnetrlc ratio, the venturl Inlet pH mriged 5.0.
Run failed due to low NoO dissolution rate In spray tower effluent hold tank.
-------
In Run 821-1A, an attempt was made to determine the S02 removal in the venturi by
turning off the slurry recirculation to the spray tower. Unfortunately, the
magnesium oxide, added to the spray tower hold tank, would not dissolve without
recirculation and the run was aborted.
A second effort was more successful. Run 822-1R was an extension of 822-1A in
which the spray tower slurry recirculation was turned off once a shift for only
30 minutes. This short time period did not upset the system balance. S02 removal
in the venturi loop was found to be 29 percent which is typical of removal effi-
ciency with limestone slurry in the absence of magnesium ion. Thus, it has been
demonstrated that magnesium ion does not enhance S02 removal in a scrubber loop
with forced oxidation.
Run 822-1A was made in an effort to improve removal efficiency by minor
changes in piping configuration to locate makeup and bleed streams at their
optimum locations i-n the venturi slurry recirculation loop. Referring to
Figure 2, the bleed from the spray tower loop was sent to the desupersaturation
tank instead of the oxidation tank as shown. Also, the bleed to the solids
dewatering system was taken from the oxidation tank instead of the desupersatura-
tion tank as shown. Improvement in SOp removal efficiency, if any, was too small
to observe.
Oxidation Efficiency - In this whole test block, the 3-inch pipe was used for dls-
charging air into the oxidation tank and an oxidation tank level of 18 feet was
maintained. Runs 820-1A, B, and C were a series to explore the air stoichiometry
required to achieve near complete oxidation. Results were as follows:
Air Stoichiometric Ratio, Percent
atoms 0/mole S02 absorbed Sulfite Oxidation
820-1A 1.7 98
820-1B 1.1 q?
820-1C 0 36
The oxidation efficiency was marginally acceptable at an air Stoichiometric ratio
of 1.1. Although the oxidation efficiency averaged 92 percent, it fluctuated
widely, indicating that barely enough air was available. During the last 40 hours
of Run 820-1B, air Stoichiometric ratio increased to 1.3 and the oxidation efficienc
was steady at 98 percent. Thus, sulfite oxidation efficiency appears to be unaf-
fected, if not improved, by the addition of magnesium oxide.
31
-------
Filter cake solids concentration at 98 percent oxidation averaged 85 percent,
demonstrating that magnesium oxide addition does .not adversely affect dewatering
characteristics of oxidized sludge. This series of runs also demonstrated the
effect of forced oxidation on solids dewatering characteristics. Filter cake
solids concentration decreased from 85 percent to 63 percent as the oxidation
of sulfite decreased from 98 percent to 36 percent.
An additional observation in this series of runs was that overall limestone
utilization decreased from 92 to 82 percent as the air rate to the oxidation
tank was reduced from 210 to 0 scfm. Presumably, the higher air rate gave better
agitation of the slurry and promoted the limestone dissolution.
32
-------
Section 3
FORCED OXIDATION WITH ONE SCRUBBER LOOP ON THE TCA SYSTEM
Forced oxidation with good S0£ removal in a sinqle scrubber loop has been
demonstrated in the TCA system using limestone slurry. In this arrangement,
sulfite oxidation is achieved by contacting the slurry with air in the scrubber
hold tank. A compromise must be made in the scrubber liquor pH between a higher
pH desired for good SOg removal and a lower pH desired for good oxidation.
Although the optimum oxidation rate occurs at about 4.5 pH, it has been found
that the oxidation rate is adequately fast up to a pH of about 6. Thus, the
oxidation pH range is compatible with the limestone scrubbing pH range of 5
to 6.
Forced oxidation in a single scrubber loop is detrimental to lime slurry scrubbing
because sulfite ion, a major scrubbing species in a lime based scrubbing system,
is removed in the oxidation process. Thus, forced oxidation substantially re-
duces SOg removal efficiency in a single loop lime system.
The single loop configuration is of prime interest commercially because the
majority of commercial installations, both operating and planned, are of this
type. Modification of these installations for forced oxidation would require as a
minimum a compressor (or blower) plus an air sparger in the scrubber hold tank.
Two devices for air/slurry contact have been tested on the TCA system. From
late June through early October 1977 an air eductor was tested. Experience with
the air eductor has been previously reported. 3) Because of erosion problems and
high energy consumption, the eductor has been replaced with an air sparger similar
to the one used in the venturi oxidation tank. Tests with the air sparger were
conducted from early December 1977 through late January 1978. All tests were
conducted with flue gas containing high fly ash loadings.
SYSTEM DESCRIPTION
Two operating configurations were used in the single loop tests. With one hold
tank as shown in Figure 4, effluent slurry from the scrubber is discharged to
the oxidation tank where limestone is added and the slurry is recycled back to
33
-------
MAKE UP WATER
FLU EG AS
LIMESTONE
CLARIFIED LIQUOR ^
\
BLEED TO SOLIDS
_ DEWATERING SYSTEM
REHEAT
r-LUb laAJS
/ \
* * *
Siii.!
>>M55!>
LLTJ
1 N
f
1
000 TCA
O OO O O
oo o
o o o o o
o O O
o o o o_c
D
<=3C
cdc
'
3
WATER
3 I COMPRESSED
^ f AIR
^^ OXIDATION TANK
FIGURE4. FLOW DIAGRAM FOR SINGLE LOOP
FORCED OXIDATION IN THE TCA SYSTEM WITH ONE TANK
34
-------
the scrubber. With two tanks in series as shown in Fiqure 5, effluent slurry
is discharged to the oxidation tank and the slurry then passes to a second tank
where limestone is added. Slurry is recycled from the second tank back to the
scrubber. Although the one-tank configuration is simpler, the two-tank configura-
tion allows the oxidation to take place at the lower pH of the scrubber effluent
before limestone is added. The two-tank configuration also provides longer
residence time for better limestone utilization.
The oxidation tank arrangement is shown in Figure 6. The tank is 7 ft in diameter
and was operated at a 17 to 18-ft level. All tests were conducted with an air
sparger ring made of straight 3-inch 316L SS pipe pieces welded into an octagon
of approximately 4-ft diameter. It was located 8 inches from the bottom of the
tank and had 40 1/4-inch diameter holes pointed downward. The sparger ring was
fed with compressed air to which sufficient water was added to assure humidifi-
cation.
A major shortcoming of this oxidation system was the agitator which was rated at
only 3 Hp and rotated at 37 rpm (compared with 17 brake Hp and 56 rpm for the ven-
turi oxidation tank). This agitator was similar in configuration to the agitator
in the venturi oxidation tank with two axial flow turbines (49 inches in diameter)
pumping downward. Because of the weaker agitation, runs with similar oxidation
tank environment (pH, air stoichiometry, tank level, percent slurry solids, and
limestone utilization) had lower oxidation efficiency in the TCA oxidation tank
than in the venturi oxidation tank.
A 20 Hp variable speed agitator is on order and will be used to develop the rela-
tionship between oxidation tank agitation and air requirements.
A second shortcoming was the existing Shawnee air compressor which did not have
sufficient capacity to serve the venturi and the TCA oxidation tanks simultaneously
at full flue gas load. To circumvent this problem, several of the TCA runs were
made at reduced flue gas flow rates. An additional air compressor has been ordered
to correct this limitation.
A clarifier was used for dewatering in all runs except Run 821-2A where a clarifier
followed by a centrifuge was used.
SUMMARY OF PREVIOUSLY REPORTED TEST RESULTS WITH AIR EDUCTOR
Forced oxidation test results with one scrubber loop conducted from late June
through early October. 1977 with limestone slurry using the air eductor have been
previously reported.-3) These tests were conducted at 30,000 acfm (300 F) flue
gas rate which corresponds to a superficial gas velocity in the TCA of 12.5 ft/sec.
The slurry recirculation rate was 1200 gpm. Each run averaged about 5 to 6 days.
All runs were made with flue gas having high fly ash loadings.
35
-------
REHEAT
MAKEUP WATER
FLUE GAS
LIMESTONE
CLARIFIED LIQUOR
BLEED TO SOLIDS
DEWATERING SYSTEM
FLUE GAS
A
1
popoo
OOOOO
OOJDOOO
TCA
COMPRESSED
AIR
A ?Vft
EFFLUENT HOLD
TANK
OXIDATION
TANK
FIGURE 5. FLOW DIAGRAM FOR SINGLE LOOP
FORCED OXIDATION IN THE TCA SYSTEM WITH TWO TANKS
36
-------
FIGURE 6
ARRANGEMENT OF THE TCA
OXIDATION TANK WITH AIR SPARGER
AGITATOR
BAFFLE
SPARGER
COMPRESSED AIR
OXIDATION TANK
PLAN VIEW
1
J
OUTLET *— '
BAFFLE v.
^s.
SPARGER WITH
401/4-inch HOLES
(DOWNWARD DISCHARGE) v
01 2345
^
\
1 i
c.
F™5'
1 t
V
»
• — •• i ii _
!3-
j
y
[ 1
7
/
"
1
•
AGITATOR
^^(37 rpm, 3 Hp)
— - OYinATIOM TAIJIT
x COMPRESSED AIR
_ INLET
r
ELEVATION VIEW
37
-------
Key results from these earlier tests were as follows:
t The dewatering and handling characteristics of slurry solids oxidized
to 90 percent or better in a single loop system were as good as those
in a double loop system.
• Sulfite oxidation to 98 percent with qood S0£ removal was demonstrated
in a single scrubber loop with two hold tanks using an air eductor for
air/slurry contact.
• Conditions under which near complete oxidation was demonstrated were
slurry feed to the eductor from a small downcomer hold tank at 5.15 pH,
eductor discharge to the oxidation tank held at 8-ft slurry level and
5.5 pH, and an air stoichiometric ratio of about 2.5 atoms 0/mole SOg
absorbed;
• S02 removal was enhanced slightly by single loop forced oxidation with
limestone scrubbing.
• The rubber lined eductor diffuser eroded severely in less than 1500
hours of operation.
ONE-SCRUBBER-LOOP TEST RESULTS WITH AIR SPARGER
Eight forced oxidation runs with limestone slurry were made on the TCA system in
a one-scrubber-loop configuration with an air sparger. Results of these tests
are reported in Table 4. Despite agitator and air compressor limitations, forced
oxidation with an air sparger in a single scrubber loop was demonstrated.
Air Stoichiometrv - Runs 815-2A through 818-2B were made with two hold tanks in
series as shown in Figure 5. The primary effort during these runs was to identify
the air stoichiometric ratio required for near complete oxidation. In the first "
two tests, run at the maximum achievable flue gas flow rate of 30,000 acfm, it
was found that the air compressor did not have a high enough capacity to supply
both the venturi/spray tower system and the TCA system. With an air rate of 210
scfm to the venturi oxidation tank only 180 scfm was available for the TCA system.
Further tests were conducted at reduced flue gas flow rates (20,000 to 25,000 acfiri\
to allow higher air stoichiometry at the available air rate. Results of these tests
conducted over an oxidation tank pH range of 5.4 to 5.7 were as follows:
38
-------
Table 4
RESULTS OF FORCED OXIDATION TESTS
WITH ONE SCRUBBER LOOP ON THE TCA SYSTEM USING LIMESTONE SLURRY
Major Test Conditions
Fly ash loading
Flue gas rate, acfm e 300°F
Slurry flow rate to TCA, gpm
Percent solids recirculated
Residence times, min: Oxidation tank
EHT
Oxidation tank level, ft
Airflow rate to sparger, scftr1'
Limestone stoichicmetric ratio (controlled)
TCA inlet pH (controlled)
Effective Mg*+ concentration, ppm
Limestone addition point
Total static height of spheres, inches
Selected Results
Onstream hours
Percent S02 removal
Inlet S02 concentration, ppm
Percent sulfite oxidation
Air stoichiometry, atoms 0/mole SO^ absorbed
TCA Inlet pH
Oxidation tank pH
Limestone utilization, X
Gypsum saturation in TCA inlet liquor, %
Mist eliminator restriction, X(
815-2A
High
30,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
20
75
89
3000
40
1.0
6.25
-
80
110
0.5
816-2A
High
30,000
1000
15
5.2
14.4
18
180
1.3
-
-
EHT
22.5
48
91
2850
54
1.40
6.25
5.7
76
100
-
817-2A
High
20,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
22.5
161
79
3000
94
1.70
5.8
5.45
81
100
-
818-2A
High
25,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
22.5
131
85
3000
67
1.25
6.2
5.65
77
95
-
818-2B
High
25,000
1000
15
5.2
14.4
18
0
1.3
-
-
EHT
22.5
140
82
3300
24
0
6.2
-
81
no
0
819-2A
High
20 ,-000
1000
15
4.9
-
17
130
1.3
-
-
Oxid. Tk
22.5
164
75
2800
94
1.90
5.55
5.55
77
110
-
820- 2A
High
20,000
1000
15
4.9
-
17
130
-
5.9
-
Oxid. Tk
22.5
259
79
2500
92
2.0
5.65
5.65
62
115
0
821-2A
High
30,000
1200
15
4.1
-
17
170
1.2
-
5000
Oxid. Tk
15
182
84
2500
95
1.65
5.35
5.35
79
110
1.5
Notes:
1) Air discharged downward through sparger ring with 40->s inch diameter holes about 8 inches from tank bottom.
2) Clarifier used for solids dewatering except for Run 821-2A where clarifier and centrifuge was used.
3) Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gprn/ft? for Runs 815-2A & 816-2A and at 0.3gpm/ft< for Runs 817-2A
through 820-2A. Intermittent bottom wash with makeup water for Run 821-2A at 1.5 gprn/ft^ for 4 minutes each hour. Sequential top wash for all
runs using makeup water with one of 6 nozzles on at 0.55 gpm/ft' for 3 minutes every 10 minutes.
-------
Air Stoichiometric Ratio Percent
Run atoms 0/mole S02 absorbed Sulfite Oxidation
817-2A 1.7 94
816-2A 1.4 54
818-2A Io25 67
815-2A 1.0 40
818-2B 0 24
Thus, with two hold tanks in series, an air Stoichiometric ratio of about 1.7 was
required to achieve greater than 90 percent oxidation,. Under similar conditions
in the venturi oxidation tank, higher oxidation efficiency was achieved. This
better performance in the venturi oxidation tank was attributed to the superior
agitation in the venturi tank.
Runs 819-2A and 821-2A were made with the oxidation tank as the only hold tank as
shown in Figure 4. In these runs, the pH in the oxidation tank was higher because
of the limestone addition. Because of the higher pH, a higher air.stoichiometr.y
was required. This effect can be seen by comparing Runs B17-2A (2 hold tanks)
and 819-2A (1 hold tank) made at essentially the same operating conditions. Ninety-
four (94) percent sulfite oxidation was achieved in both runs. An air Stoichiometric
ratio of 1.7 atoms 0/mole S02 absorbed was used in the run with two hold tanks (5.4
oxidation tank pH) while an air Stoichiometric. of 1.9 was required in the run
with one hold tank (5.65 oxidation tank pH).
S02 Removal Efficiency - S02 removal efficiency in these runs appeared to be in-
dependent OT oxidation efficiency. S02 removal efficiency was primarily a func-
tion of flue gas flow rate and inlet S02 concentration, closely following pre-
viously developed correlations for the TCA system in limestone service without
forced oxidation. At 3000 ppm inlet S0£ concentration and a limestone Stoichio-
metric ratio controlled at 1.3 moles Ca/moles SO? absorbed, S0£ removal efficiency
ranged from about 90 percent at 30,000 scfm to about fin percent at 20,000 scfm.
Limestone Utilization - Limestone utilization was higher in the runs using two
tanks in series tnan in the single tank runs. Again comparing runs 817-2A and
819-2A, limestone utilization with one tank was 77 percent while with two tanks
it was 81 percent, SOg removal efficiency was also improved from 75 percent with
one tank to 79 percent with two tanks. The improvement can be attributed to higher
40
-------
residence time (19.6 minutes with two tanks versus 4.9 minutes with one tank) and
the approach to plug flow reaction inherent with tanks in series.
Because of the relatively poor S02 removal efficiency in Run 819-2A, the next run
(820-2A) was made at a slightly higher pH. The oxidation tank pH was increased
from 5.55 to 5.65. S02 removal increased only slightly from 75 percent at 2800
inlet ppm to 79 percent at 2500 inlet ppm. However, the limestone utilization
decreased from 77 percent to 62 percent.
Magnesium Oxide Addition - The addition of magnesium oxide should not enhance
S02 removal in a scrubber loop with forced oxidation. This was demonstrated in
Run 821-2A. Magnesium ion in the scrubber liquor improves $02 removal by in-
creasing the sulfite ion, an effective SOg scrubbing component. Rut forced oxi-
dation converts the sulfite to sulfate which is non-reactive.
In Run 821-2A, with 5000 ppm effective magnesium ion concentration and with forced
oxidation, the S02 removal efficiency averaged 84 percent, no higher than expected
without magnesium oxide-addition. In a previous run with magnesium oxide addition
equivalent to Run 821-2A but without forced oxidation, S02 removal averaged 92
percent. Thus, the enhancement on SO? removal with magnesium oxide addition is
not achieved in a scrubber loop with forced oxidation.
41
-------
Section 4-
FORCED OXIDATION OF THE VENTURI/SPRAY TOWER'BLEED STREAM
Forced oxidation within the scrubber loop requires a compromise between the
conditions needed for good oxidation and those heeded for qood SOn removal.
Such would not be the case if it were possible to oxidize the slurry bleed
stream by simple air/slurry contact. Unfortunately, tests at the IERL-RTP
pilot plant^) and at the Shawnee Test Facility-*) have shown that air sparging
of the bleed stream increases the rate of dissolution of the residual alkali
and causes the pH to rise, slowinq down the oxidation rate to an impractica-1
level. Furthermore, tests conducted with sulfuric acid addition to control the
bleed stream pH have produced oxidized sludqe with inferior dewaterinq and
handling characteristics.3)
Despite the generally unfavorable results, batch oxidation tests at the Shawnee
Laboratory indicated that near complete sulfite oxidation could be achieved by
simple air sparging of lime or limestone slurry when magnesium ion was present
in concentrations of 1,000 ppm or higher. Magnesium ion apparently has two
effects: it tends to buffer the pH rise from dissolving residual alkali in the
waste slurry solids; and it tends to promote dissolved sulfite availability,
allowing oxidation to take place at a higher pH.
Starting in mid-May 1978, bleed stream forced oxidation with limestone slurry
and added magnesium oxide was successfully demonstrated in a month Ipnq series
of tests on the venturi/spray tower system. Oxidized slurry from-these tests
had good dewatering properties with filter cake solids concentration averaging
about 85 percent. Thus, it is commercially feasible to improve the quality and
reduce the volume of waste solids in installations incorporating maqnesium ion
in the slurry liquor by simple air/slurry contact of the bleed stream.
SYSTEM DESCRIPTION
The venturi/spray tower system was arranged as shown in Figure 7 for the bleed
stream oxidation tests. Both the venturi and the spray tower slurries discharged
into a single hold tank to which limestone and magnesium oxide were fed. A bleed
stream was taken from the spray tower downcomer to take advantage of the low pH
42
-------
FLUE GAS
IV\gO
LIMESTONE
REHEAT
VENTURI
FLUE GAS
A
**»*»»
\L- xU ^l/
MAKE-U P WATER
SPRAY TOWER
D
OVERFLOW
EFFLUENT
HOLD TANK
COMPRESSED
VENT AIR
t
CLARIFIED LIQUOR
OXIDATION
TANK
BLEED TO
SOLIDS
DEWATERING
SYSTEM
FIGURE 7. FLOW DIAGRAM FOR BLEED STREAM OXIDATION IN THE VENTURI/SPRAY TOWER SYSTEM
-------
at that point. The bleed stream was discharged to the oxidation tank which was
arranged as shown in Figure 3 and is described in Section 2. During these tests,
the 3-inch pipe was used to discharge air into the oxidation tank. All tests were
conducted at an 18-ft oxidation tank level. Bleed from the oxidation tank was
dewatered by a clarifier and a filter in series.
BLEED STREAM OXIDATION TEST RESULTS
Four bleed stream, oxidation runs were made on the venturi/spray tower system
using limestone with added magnesium oxide. All tests were conducted with approx-
imately 5000 ppm effective magnesium ion concentration in the slurry liquor.
Test results are reported in Table 5. Percent S0£ removal was high as expected
in runs with magnesium oxide enhancement. Oxidized slurry solids in all runs had
good dewatering properties, averaging about 85 percent filter cake solids con-
centration.
In Runs 823-1A and 824-1A, conducted at 18,000 acfm and 35,000 acfm, respectively,
97 to 98 percent sulfite oxidation was achieved at an air stoichiometry of about
1.6 atoms 0/mole S0£ abosorbed. Oxidation was consistently high even though the
oxidation tank pH averaged 6.3 in Run 823-1A and at times rose as high as 6.7.
In these runs, 30 gpm of oxidized slurry was recycled from the oxidation tank back
to the scrubber hold tank. The purpose of this recycle was to reduce the pH
difference between the oxidation tank and the scrubber hold tank. However, the
opposite occurred. The hold tank pH was depressed, requiring excess limestone
feed to maintain a pH of 5.3. The net result was a limestone utilization of less
than 40 percent for these runs.
Runs 825-1A and 826-1A (at 18,000 acfm and 26,500 acfm, respectively) were con-
ducted without this recycle. In both of these runs the pH difference between
the scrubber hold tank and the oxidation tank was only about 0.1 and 0.2 with the
oxidation tank pH averaging 5.65 or less. Near completL oxidation (high 90's)
was easily achieved in both runs with air stoichiometric ratios of 1.6 and 2.0,
respectively. Time was not available in the test block to determine minimum
air stoichiometry. Control of limestone feed was poor in these runs resulting in
relatively low limestone utilization (64 and 61 percent, respectively).
This short series of runs has shown that in systems containing magnesium ion,
the slurry bleed stream can be readily oxidized. Furthermore, oxidation of the
bleed stream does not interfere with enhancement of S0~ removal by the magnesium
ion as was experienced when oxidation was accompli sheer within the scrubber loop.
44
-------
Table 5
RESULTS OF FORCED OXIDATION TESTS ON THE VENTURI/SPRAY TOWER BLEED STREAM
USING LIMESTONE SLURRY WITH ADDED MAGNESIUM OXIDE
Major Test Conditions
Fly ash loading
Flue gas rate, acfm ? 300°F
SI urry rate to venturi , gpm
Slurry rate to spray tower, gpm
Percent solids recirculated (controlled)
EHT residence time, min.
Spray tower inlet pH (controlled)
Scrubber limestone stoichiometric ratio (control led) (based on solids)
Effective Mg+ concentration, ppm
Venturi pressure drop, in. H-0
Oxidation tank level, ft
Air rate to oxidation tank, scfnr '
Recycle flow from oxidation tank to EHT, gpm
Selected Results
Onstream hours
Percent SO, removal
Inlet S02 concentration, ppm
Scrubber percent solids recirci/lated
Scrubber inlet liquor pH
Oxidation tank pH
Limestone utilization, % (based on total slurry)
Sulfite oxidation in oxidation tank, %
Sulfite oxidation in scrubber inlet slurry, %
Gypsum saturation in scrubber inlet liquor, %
Gypsum saturation in oxidation tank, %
Effective Mg++ concentration in scrubber inlet liquor, ppm
Oxidation tank liquor sulfite concentration, ppm
Air stoichiometry, atoms 0/mole S02 absorbed
Filter cake solids, wt* '2^
Mist eliminator restriction, % '3)
823-1A
High
18,000
600
1600
15
11.2
5.3
-
5000
9
18
110
30
205
94
2600
13.3
5.25
6.30
36
98
86
120
115
4990
65
1.55
83
0
824-1A
High
35,000
600
1600
15
11.2
-
1.9
5000
9
18
210
30
159
88
2600
14.1
5.25
5.90
38
97
49
85
90
5215
105
1.60
85
0
825- 1A
High
18,000
600
1600
15
11.2
-
1.4
5000
9
18
110
0
229
95
2500
14.7
5.45
5.65
64
97
39
105
115
5380
220
1.60
85
0.5
826-1A
High
26,500(4)
600
1600
15
11.2
-
1.4
5000
9
18
210
0
246
89
2750
15.2
5.35
5.45
61
96
29
105
115
4970
230
2.00
84
0.1
Notes:
1) Air discharged downward through 3-inch diameter pipe with an open elbow at center of oxidation tank about 3 inches from tank bottom,
2) Clarifier and filter in series used for solids dewatering in all runs, ,, 2
3) Continuous mist eliminator bottom wash with diluted clarified liquor at 0,4 gpm/ft (0.3 gpm/ft for Run 823-1A), Sequential top wash
with makeup water with one of 6 nozzles on at 0.53 gpm/ft^ for 4 minutes every 80 minutes.
4) Desired flow rate was 35.000 acfm but problems with the venturi lifting mechanism limited the rate to 26,500 acfm.
-------
Additional testing will be conducted to fully characterize the limestone/MqO
bleed stream oxidation system. Better operational control is required to improve
the limestone utilization. Higher limestone utilization, however, is not expected
to have an adverse effect on the oxidation efficiency because of the reduced amount
of residual alkali and the correspondingly less possibility of pH rise in the
oxidation tank.
The possibility of bleed stream oxidation on a lime/MqO system must also be investi
gated.
46
-------
Section 5
DEWATERING CHARACTERISTICS OF THE OXIDIZED SLURRY SOLIDS
The settling and dewatering characteristics of slurry solids are routinely
monitored in the Shawnee laboratory by cylinder settling tests and vacuum
funnel filtration tests. Results of these monitoring tests are presented in
this section. The test results are summarized in Table 6.
Results of sludge disposal studies at the Shawnee Test Facility are presented
on a separate paper by the Aerospace Corporation.
Cylinder settling tests are performed in 1000 ml cylinder containing a rake which
rotates at 0.16 rpm. The initial settling rate and ultimate settled solids con-
centration are recorded as indices of dewatering characteristics. The initial
settling rate is a qualitative index of the solids settling properties only. De-
sign rates for sizing clarifiers must take into consideration the hindered settling
rate as the solids concentrate. The ultimate settled solids from the cylinder
tests represent the highest achievable solids concentration in a settling pond.
Funnel filter tests are performed in a Buchner funnel with a Whatman ?. filter paper
under a vacuum of 25 in. Hg. The funnel tests correlate well with the Shawnee
rotary drum vacuum filter when not blinded but the funnel test cakes tend to have
lower solids concentrations.
As can be seen in Table 6 the benefits of forced oxidation are clear, the dewatering
characteristics of oxidized sludge are markedly better than those of unoxidized
sludge. The initial settling rate is higher by a factor of 4, and both the settled
and filtered solids concentrations are higher by a factor of 1.4.
Without forced oxidation, the average initial settling rate was about 0.2 cm/min.
With forced oxidation the average initial settling rate was higher, ranging from
0.42 cm/min. to 1.20 cm/min.
The presence of magnesium ion tended to decrease the initial settling rate - slightl
with oxidized slurry and more with unoxidized slurry. For oxidized slurrv in the
2-loop mode of oxidation, the average initial settling rate was 1.0 cm/minute
47
-------
Table 6
SUMMARY OF THE DEWATERING CHARACTERISTICS
OF THE SHAWNEE WASTE SLURRY
OO
Oxidation
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
Ho
No
No
Fly Ash
Loading
High
High
High
High
High
Low
Low
High
High
High
High
High
Low
Low
Alkali
LS
LS
LS
LS
L
LS
L
LS
LS
LS
L
L
LS
L
Onldatlon Hade
1-loop
2-loop
Bleed Stream
2-loop
2-loop
2-loop
2-loop
-
-
-
-
-
-
-
Initial Settling Rate. ca0i1n|uUlMte Settling Solids, wtt
Avg. | Range | Hvg. | Range
1.06 0.63-1.27 74 67-84
0.70(1) 0.55-0.8711'
1.20(Z1 0.96-1.41(2> 72 62-86
1.12(3) 0.81-).47(3)
0.42 0.29-0.61 71 61-84
0.75 0.23-1.19 66 46-73
0.98 0.77-1.23 73 61-85
0.88 0.61-1.15 74 61-87
1.20 0.45-2.44 70 60-81
D.20 0.07-0.54 54 41-67
0.20 0.07-0.37 45 30-60
0.05 0.01-0.11 41 32-46
0.20 0.19-0.49 50 48-66
0.79 0.22-1.15 42 31-52
0.17 0.05-0.47 43 33-54
0.35 0.09-0.87 40 30-55
Funnel Test Cake Solids. wtS Slui
Avg. | Range Sol
76 73-80 16
72 65-88 15
73 71-76 15
70 46-76 15
71 64-78 15
73 64-82 15
76 64-83 15
57 48-66 15
57 45-64 15
55 47-69 15
53 51-55 15
52 43-63 8
50 41-59 15
45 40-50 8
'ry Effective Hg
ds Concentration, ppn
0
0
5000
8000 l
0
0
0
0
5000
9000
0
2000
0
0
Note: Values for forced oxlditlon runs are only from data where solids oxidation
was greater than or equal to 90 percent,
(1) Ox1d1«r pH
(Z)
4.5
OxIdUer pH « 5.0
(3) Oxldlzer pH - 5.5
-------
without magnesium and 0.75 cm/min. with 8000 ppm effective magnesium ion concentra-
tion. For unoxidized limestone slurry with high fly ash loading, the average initial
settling rate was reduced from 0.20 cm/min. without magnesium to 0.05 cm/min. with
9000 magnesium ion, a decrease by a factor of 4. This magnesium effect is probably
the result of an increase in liquor viscosity and density due to the increased
amount of total dissolved solids.
Figure 8 is a plot of percent oxidation versus settling rate for a lime system with
low fly ash loading. The presence of fly ash in oxidized slurry appears to decrease
the settling rate slightly. For example, oxidized lime slurry with low fly ash
loading has an average initial settling rate of 1.20 cm/min., whereas oxidized lime
slurry with high fly ash loading has an average initial settling rate of 0.98 cm/min.
Similar results were obtained with the ultimate settled solids and the funnel test
cake solids. Without forced oxidation, the ultimate settled solids were generally
in the range of 40 to 50 weight percent solids; with forced oxidation, the range was
65 to 80 weight percent. Funnel test results indicated 45 to 60 weiqht percent solids
without forced oxidation and 65 to 85 weight percent with forced oxidation. On the
rotary drum vacuum filter used in the scrubber dewatering system, cake solids con-
centration was always above 80 percent with oxidized slurry while averaging 50 to 60
percent with unoxidized slurry.
The unoxidized solids tended to be thioxotropic, like quicksand, while the oxidized
solids were more like moist soil.
The presence of magnesium ion did not affect the ultimate settled solids or the
funnel test cake solids results. This result was also seen in the test facility's
operating data.
The product solids at Shawnee are a mixture of unoxidized calcium sulfite hemihydrate
(CaS03'l/2H20), the oxidation product calcium sulfate dihydrate (CaSO^Zt^O), fly
ash, unreacted alkali, and other inert materials. In slurry that is 10 percent oxi-
dized, the main component is calcium sulfite in platelets and rosettes of only a few
microns in diameter. The solids in slurry 95 percent oxidized are mainly calcium
sulfate crystals having a bulky rectangular shape and ranging in size from 20 to
100 microns.
It is currently thought that in the case of unoxidized slurry, the calcium sulfite
fines are the limiting factor in the inital settling rate. Rut in the case of oxi-
dized slurry, the fly ash may be the limiting factor. Fly ash is extremely fine
when compared with the calcium sulfate particles and hence settles at a slower rate.
At the Shawnee Test Facility, the limit for oxidized slurry with high fly ash
loading appeared to be 1.5 cm/min. For oxidized slurry with low fly ash loading,
the limit was 2.4 cm/min. Since, at Shawnee, solids from flue gas with low fly ash
loading contain up to 1 weight percent fly ash, the limits of the calcium sulfate
settling rate may be even higher.
The residence time in the oxidation tank may also affect the size of the gypsum
crystal; the longer the residence time the larger the crystal. This effect has yet
to be thoroughly explored at Shawnee.
49
-------
FIGURE 8
EFFECT OF OXIDATION ON INITIAL SETTLING RATE
2.5
2 -
e
E
1.5 -
Z
-I
UJ
CO
1 -
.5 -
10 20 30 40 50 60 70
%SULFITE OXIDIZED
80
90 100
50
-------
Section 6
FUTURE TESTING
Testing with forced oxidation will be continued with emphasis on:
• More fully developing forced oxidation within a single scrubber loop
• Exploring the conditions under which bleed stream oxidation is applicable
t Determining compatibility of forced oxidation with chemical additives
such as adipic acid
Based on the encouraging results at the IERL-RTP pilot plant, an extensive program
to develop adipic acid as an additive for enhancing S02 removal efficiency has re-
cently been initiated. Adipic acid acts as a buffer to limit the drop in pH,
thereby improving the liguid-phase mass transfer. The advantages of adipic acid
are listed below:
• Lower cost compared to MgO based on the quantity needed. For example, for
a similar degree of Sr>2 removal enhancement in limestone scrubbing:
With MgO: 6,000 ppm Mg++ at $0.17/lb MgO requires
$14/1,000 gal of discharged liquor
With adipic acid: 1,000 ppm adipic acid at $Q.42/lb
acid requires only $3.50/1,000 gal of
discharged liquor
• Optimum adipic acid concentration for effective improvement in S02 removal
is only 5-10 m-moles/liter (700 - 1,500 ppm)
• Adipic acid improves SO? removal and may also improve 1imestone utilization,
whereas MgO may reduce the limestone dissolution rate
t In the limestone scrubbing system, S02 removal efficiency is no longer
limited by the limestone dissolution rate when adipic acid is present in
sufficient quantity
51
-------
• Forced oxidation does not affect the effectiveness of adipic acid.
Forced oxidation reduces the effectiveness of the maqnesium ion by
converting the scrubbing S03= into non-scr.ubbinq S04= species
• Unlike MgO addition, where two chloride ions tie up a maqnesium ion
to form neutral MgC^, adipic acid is not affected by chloride
• Adipic acid is nontoxic (used as a food additive)
• Both IERL-RTP pilot plant and preliminary Shawnee results show that
the solids quality (filterability, settling rate) is not affected by
adipic acid
Tests with adipic acid as an additive are scheduled with and without forced
oxidation and with both lime and limestone slurries.
Tests are also planned to investigate the effect of limestone type and grind on
SO? removal and limestone utilization. Initial screening tests have already been
scheduled for the IERL-RTP pilot plant.
A 3-month test block is planned for the Shawnee spray tower with various internal
configurations (different number of headers, number of nozzles, type of nozzles
nozzle pressure drop, etc.). The primary objective of the testing will be to
provide a better basis for designing full-scale spray towers.
Long-term (over one month) lime and limestone tests are also planned, which will
combine the most promising operating conditions, including forced oxidation and
organic acid addition, to demonstrate system reliability and conformance to the
existing Federal emission standard.
Other concurrent future activities include:
• Transfer of Shawnee-developed technology to full-scale plants, including
if necessary, simulation of commercial plant operation at Shawnee '
• Continued development and updating of the Economic Study Computer Program
in conjunction with TVA
• A study of the overall power plant water management as it relates to FRO
plant operation
52
-------
Section 7
REFERENCES
1. Barrier, J.W. et al, Comparative Economics of FGD Sludge Disposal.
Presented at the 71st Annual Meeting of the Air Pollution Control
Association, Houston, Texas, June 25-30, 1978.
2. Borgwardt, R.H., Sludge Oxidation in Limestone FGD Scrubbers
EPA-600/7-77-OG1, June 1977. ~~~
3. Head, H.N. et al, Results of Lime and Limestone Testing with Forced
Oxidation at the EPA Alkali Scrubbing Test Facility. Proceedings:
Symposium on Flue Gas Desulfurization - Hollywood, Fl., November 1977
EPA-600/7-78-58a, March 1978. '
4. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Summary of
Testing through October 1974. EPA 650/2-75-047, June 1975.
5. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Advanced
Program. First Progress Report. EPA-6no/2-75-050, September 1975.
6. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Advanced
Program. Second Progress Report. EPA-600/7-76-008, September 1976.
7. Bechtel Corporation, EPA Alkali Scrubbing Test Facility: Advanced
Program. Third Progress Report. EPA-600/7-77-105, September 1977.
53
-------
SELECTED TOPICS FROM
SHAWNEE TEST FACILITY OPERATION
Presented bv
David T. Rabb
EPA Program Site Manager
Bechtel National, Inc.
50 Beale Street
San Francisco, California 94119
at the
EPA Industry Briefing
Research Triangle Park, North Carolina
August 29, 1978
EPA Contract 68-02-1814
John E. Williams
Project Officer
Industrial Environmental Research Laboratory
Office of Research and Development
Research Triangle Park, North Carolina 27711
54
-------
.LNTROBUCTI-ON
The Shawnee Test Facility is an EPA funded wet lime/limestone scrubbing test
facility that has operated since 1972. The facility currently consists of two
10 MW equivalent scrubbing systems each treating approximately 7 percent of the
flue gas produced by a medium-to-high sulfur coal-fired 150 MW boiler. One
system is a venturi followed by a spray tower (V/ST) the other is a Turbulent
Contact Absorber (TCA).
The information presented here reflects selected topics from the operating
experience obtained between June 1977 and May 1978 at this facility. These
include:
a) Scrubber operation and maintenance
b) Oewaterinq systems
c) Forced oxidation systems
d) Automatic control of limestone addition
e) Operational development
SCRUBBER OPERATION -AND MAINTENANCE
SCRUBBER OPERATION
Over the past year the V/ST and the TCA have maintained a high operational
availability that is summarized in Table 1. The V/ST system operated 7040
hours during the year or 80 percent of the time. The TCA system operated 7272
hours or 83 percent of the time. The systems downtimes were attributed to
four categories:
a) Boiler outages
b) Weather affecting scrubber operations
c) Scheduled scrubber system inspections and modifications
d) Scrubber equipment failures
The unit 10 boiler operated 91 percent of the time and consequently caused 9
percent downtime for each of the two systems for the year. Of the over ROO hours
of boiler downtime approximately 350 hours were the result of a scheduled major
overhaul. The hours of downtime differ for the two systems because the V/ST and TCA
had to start-up separately due to operating personnel limitations.
55
-------
TABLE 1
SCRUBBER OPERATION
OPERATING TIME. Hours
DOWN TIME, Hours
BOILER & DOE
WEATHER
INSPECTION & MODIFICATION
EQUIPMENT PROBLEMS
TOTAL PERIOD*. Hours
VENTURI/SPRAY TOWER
7040 (80%)
TCA
7272 (83%)
832(9%)
220 (3%)
421 (5%)
247 (3%) ^
810 (9%)
171 (2%)
) 1720 (20%)
323 (4%)
184(2%)^
) 1488(17%)
8760 (100%)
8760 (100%)
*JUNE 1977 THROUGH MAY 1978
-------
The second major cause of downtime was attributed to routine scrubber inspec-
tions and system modifications inherently associated with the goals of this test
facility. Typically a test would last 5 to 6 days with occasional runs lasting
as long as 9 days to insure sufficient steady state data. The time spent between
test runs varied from a few hours for a scrubber inspection to 3 or 4 days for a
system modification. The factors caused downtimes for the year of 5 percent for
the V/ST and 4 percent for the TCA.
The remaining two causes of outages were scrubber equipment failures and
weather related problems. Equipment failures and subsequent maintenance
will be discussed in depth in the next section. The weather related problems
were the result of the abnormally severe winter experienced in Kentucky last
January. Alkali addition streams were the most troublesome in that the one-
inch utility hoses used as alkali addition lines froze repeatedly. Maintenance
was seriously hampered by the cold. The downtime for the year caused by the weather
was 3 percent for the V/ST and 2 percent for the TCA.
Curtailed power generation because of scrubber inavailability is of serious concern
to utility companies. For this paper, scrubber availability is introduced and is
defined as the percent of time that a commercial scrubber operates while the
accompanying boiler is operating. Scrubber equipment failures and weather created
problems were used in calculating the availability term at Shawnee. Modifications
and inspections due to test requirements were not included because they were in-
consistent with the concept of a commercial installation.
For the V/ST and TCA, the average availability for the last year was 94 and 95
percent, respectively. On a monthly basis the availability ranged from R2 to
100 percent for the V/ST and 81 to 100 percent for the TCA.
MAINTENANCE
At the Shawnee Facility, the high availability has been strongly influenced by
an effective maintenance program and an adequate spare parts inventory. These
factors are particularly important when considering pump maintenance; the
facility does not have stand-by spare pump capacity as is common practice in
industrial systems. The total hours and frequency associated with equipment
maintenance that resulted in system shut-downs are outlined in Tables 2 and 3
for the V/ST and TCA, respectively.
In the V/ST system the primary cause of downtime was scrubber related problems
that consisted, in part, of a corroding air sparge pipe, pipe plugging and
leaks, nozzle plugging, and solids build-up in the outlet ductwork! The
corroding sparge pipe occurred only once and was caused by mistakenly using
304 stainless steel as part of the material of construction.
57
-------
TABLE 2
V/ST EQUIPMENT MAINTENANCE CAUSING DOWN TIME
JUNE 1977 TO MAY 1978
Ul
00
ITEM
SCRUBBER INTERNALS AND PIPING
I. D. FAN
INSTRUMENTATION
PUMPS
ALKALI FEED
OTHER
TOTAL
FREQUENCY OF EVENT
TOTAL HOURS
12
4
1
2
1
1
114
78
39
12
3
1
21
247
-------
TABLE 3
TCA EQUIPMENT MAINTENANCE CAUSING DOWN TIME
JUNE 1977 TO MAY 1978
Ul
VO
ITEM
SCRUBBER INTERNALS AND PIPING
AGITATOR
PUMPS
ALKALI FEED
INSTRUMENTATION
OTHER
FREQUENCY OF EVENT TOTAL HOURS
6
108
39
4
2
1
2
17
8
7
5
TOTAL
16
184
-------
The longest downtime requirement for the TCA was the installation of the
Penberthy eductor. The agitator shaft replacement in the main hold tank was
second, followed by routine maintenance of the main slurry pump which consisted
of repacking Allen-Sherman-Hoff Centri-Seals. Other operating problems included
plugging of the alkali feed lines, bearing failures of the I.n. fan and general
solids build-up in the outlet duct.
The personnel available for maintenance requirements and system modifications
are listed in Table 4. To avoid possible conflicts of priorities between the
powerhouse and scrubbing facility and to allow the crafts people to develop
expertise with specific equipment, the maintenance personnel are assigned exclu-
sively to the test facility.
DEWATERING SYSTEMS
The primary dewatering of the purged slurry in both scrubbers is achieved bv
clarifiers. further dewatering of the clarifier sludges is accomplished in the
V/ST system by a filter and in the TCA system by a centrifuge. The information
discussed below represents actual data that might be used as a guide in understand
ing the operating problems and costs associated a with centrifuge or filter.
CENTRIFUGE
A continuous centrifuge is one process used to dewater scrubber waste sludge
and to recover the dissolved scrubbing additives. The normal operating condi-
tions usually consist of a feed stream flow of 15 gpm at 30 to 40 wt. percent
solids, a centrate of 0.1 to 3.0 wt. percent solids, and a cake of 55 to 65 wt,
percent solids, for unoxidized slurry. Approximately 30 percent of the total
solids is fly ash; the remaining solids are predominantly calcium sulfate
and sulfite.
The machine is a Bird 18" x 28" solid bowl continuous centrifuge which operates
at 2050 rpm. The material of construction is 316L stainless steel with stellite
hardfacing on the feed ports, conveyor tips and solids discharge ports. The howl
head plows and case plows are replaceable. The pool depth is set at 1-1/2 inches
No cake washing is performed in this machine.
The centrifuge was inspected in June 1978, after 6460 hours of operation since
the previous factory servicing. The inspection was prompted by the gradual and
continued increase of centrate suspended solids to a level of approximately 3
wt. percent. The machine was judged to be in generally fair condition but
certain components were in need of factory repair. Serious wear was observed
at the conveyor tips on the discharge end and at the junction of the cylinder
and the 10* section of conveyor. Wear was also present at the casing head plows
and solids discharge head near the discharge ports. The bowl and effluent head
were in good condition.
60
-------
TABLE 4
SHAWNEE FACILITY
MAINTENANCE PERSONNEL
I MAINTENANCE FOREMAN 1
BOILERMAKER-WELDER 2
CARPENTER 1
ELECTRICIAN 1
PIPEFITTER-WELDER 2
TEAMSTER 1
LABORERS 2
HEAVY EQUIPMENT OPERATOR 1
PAINTER 1
MACHINIST 1
INSULATOR 1
II INSTRUMENT FOREMAN 1
SENIOR INSTRUMENT MECHANIC 2
JOURNEYMAN INSTRUMENT MECHANIC 2
III TOTAL 19
61
-------
TABLE 5
CENTRIFUGE MAINTENANCE & POWER REQUIREMENT
JUNE 1977 TO MAY 1978
MAINTENANCE
ESTIMATED ESTIMATED
TOTAL ONSITE TOTAL MATERIAL
FREQUENCY OF LABOR COST
EVENT OCCURRENCE (MAN-HOURS) ($)
FEED PIPE REPAIR 1 16 20
II POWER REQUIREMENT - 30 HORSEPOWER
-------
The current plans call for the centrifuge to be shipped to the factory for an
overhaul. The following items will he accomplished:
a) Inspect and service the gear and bearing unit
b) Rebuild all worn conveyor surfaces and add hardfacing on the tips
c) Rebuild and add hardfacing to the discharge ports
d) Replace all seals and bushings in the effluent and discharge head
e) Replace case plows and discharge plows as necessary
In an attempt to improve performance and machine life, tungsten carbide
hardfacing wi.ll be applied to the conveyor tips instead of the previously
used stellite. The estimated cost for the complete factory service, including
the hardfacing, is $17,000.
The machine has been a minimum maintenance item; as shown in Table 5 the only
maintenance of the past year has been the replacement of the feed pipe in the
centrifuge. The power requirement for the machine is 30 horsepower.
FILTER
An Ametek 31 x 6' vacuum drum filter without cake wash is operated at the facility
for waste sludge dewatering and dissolved scrubbing additive recovery. The feed to
the filter is usually 15 gpm of 30 to 40 wt. percent solids.
The filtrate generally contains less than 0.02 wt. percent solids. The filter
cake varies from 55 to 85 wt. percent solids depending mainly on whether the
sludge is unoxidized or oxidized.
The filter, with the exception of the filter cloth, has been a moderate
maintenance item. Table 6 is a breakdown of maintenance categories, frequency,
approximate total manhours required, and approximate replacement material
costs. Also, included in Table 6 is the power requirement for the filter and
vacuum pump.
Contrary to experiences at other scrubbing facilities, filter cloth replacement
as noted in Table 7 has been a serious problem at Shawnee. The causes of cloth
blinding and fraying are not satisfactorily understood as yet. However, in the
last few months operating experience indicates a relationship between cloth
63
-------
TABLE 6
FILTER MAINTENANCE & POWER REQUIREMENT
JUNE 1977 TO MAY 1978
I MAINTENANCE
EVENT
SPEED CONTROL REPAIR
CAKE DISCHARGE AIR REPAIR
CLOTH REPLACEMENT
FREQUENCY OF
OCCURRENCE
15
ESTIMATED
TOTAL ONSITE
LABOR (MAN-HOURS)
ESTIMATED
TOTAL MATERIAL
COST ($)
2
2
16
16
1000
0
90
1500
II POWER REQUIREMENT - 20 HORSEPOWER
-------
CLOTH TYPE*
TABLE 7
FILTER CLOTH SERVICE
DATE INSTALLED
HOURS IN SERVICE
COMMENT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
TFI
LAMPORTS
LAMPORTS
LAMPORTS
LAMPORTS
LAMPORTS
TFI
TFI
LAMPORTS
AMETEK
AMETEK
LAMPORTS
LAMPORTS
AMETEK
AMETEK
6-6-77 292
7-7
7-13
7-19
8-3
8-15
9-19
9-23
10- 18
127
142
302
249
540
187
501
187
10-26* 2096
2 - 6 78 188
2-14
2 22
3-21
4-19
190
290
535
•
1197
BLINDED
HOLE IN CLOTH
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
HOLE IN CLOTH
BLINDED
HOLE IN CLOTH
BLINDED
BLINDED
HOLE IN CLOTH
*AMETEK - AMETEK OLEFIN (STE - F9D8 - HJO)
LAMPORTS- LAMPORTS POLYPROPYLENE (7512 - SHS)
TFI -TECHNICAL FABRICATORS INCORPORATE, POLYPROPYLENE (9162)
65
-------
life and the technique by which the cloth is fitted to the drum. A carefully
controlled amount of looseness in the fit between the dividers appears to be
desireable for cake discharge and non-blinding. The looseness evidently allows
the cloth to "snap" the cake off when the air puff of the cake discharge cycle
is applied to the given filter cloth section. Two additional observations have
been that oxidized sludge exhibits less tendency towards cloth blinding and
Ametek olefin appears to provide the most satisfactory service of these cloths
tested. The reason for the better service life of Ametek is suspected to be
attributable to the looseness of weave that the Ametek has in comparison with
Lamports and TFI.
As noted, some limited progress has been achieved towards understanding and re-
ducing cloth blinding. Future operations at Shawnee will include continued efforts
to improve filter cake performance.
OXIDATION -SYSTEM-DISCUSSION
The desirability of an oxidized calcium base sludge has for a number of years
been recognized when considering ease of sludge filtration, handling and disposal
Forced oxidation has been determined to be one method in achieving those ends. i
forced oxidation, air is introduced and dispersed into the slurry as required to
oxidize the calcium sulfite to calcium sulfate.
At Shawnee, forced oxidation has been investigated in a series of tests using
a Penberthy eductor and an air sparger. The intended goal of these tests was
to investigate the operating parameters necessary to achieve "near complete"
oxidation, i.e., greater than 90 percent total sulfur as sulfate.
One important operating experience resulting from these tests was a tight water
balance with no observable deterioration of mechanical components or scruhbinq
chemistry.
PENBERTHY EDUCTOR
The Penberthy eductor is a device similar in concept to a laboratory aspirator.
A high velocity slurry passes through a constricted nozzle, into an eductor
chamber and then through a moderately restricted jet throat. In the chamber
the slurry induces a vacuum that draws ambient air into the chamber via an entry
pipe located at right angles to the slurry flow path. The air is then entrapped
and mixed in the slurry as the fluid leaves the eductor chamber and passes throuah
the jet throat. *n
At Shawnee, a Penberthy Model ELL-10 eductor was tested. The materials of con-
truction were stellite for the nozzle and neoprene-lined carbon steel for the
eductor chamber and exit jet throat. The system was operated at 1600 gpm.
66
-------
Test results indicated that near complete oxidation could be achieved, hut at
the same time serious erosion problems developed. The neoprene lininq in the
jet throat was observed to be "chipped off" after only 620 hours of operation.
After approximately 1800 hours of operation, bare carbon steel was exposed, and
after an epoxy patch failed after 2055 hours of operation, the tests were
terminated.
Currently, no plans exist to resume testing of eductors primarily because no
advantages in oxidation capabilities were seen in comparison with sparqe air
systems. Secondary reasons included the materials erosion problem and the unfa-
vorable operating and capital costs in comparison to sparge air systems.
AIR SPARGER
The sparge atr system used in sludge oxidation is a simple concept that involves
bubbling air into the bottom of a slurry tank in conjunction with simultaneous
slurry agitation. At Shawnee the oxidation tanks are purposely tall and thin
(7 to 8 feet diameter x 20 feet tall). This shape enables a long air/slurry
contact time.
In the TCA an octagonal sparge ring with 40 holes is currently installed at the
bottom of the effluent hold tank. The holes are 1/4 inch diameter and located
on the underneath side of the ring; no sparge ring plugging had occurred to
date. The ring is constructed of 316L stainless steel'and is materially in
good condition except for minor erosion at the 1/4 inch air holes. Currently
a conventional tank agitator (37 rpm, 3 hp) is in service. Future tests will
use a variable, high speed agitator. Near complete oxidation has been achieved
with the current configuration.
Testing was done on the V/ST system with an octagonal sparge ring similar in
design to that of the TCA. After 2400 hours of operation the ring was removed
from service and replaced with a 3-inch diameter open-ended sparge pipe. The air
and slurry are mixed with an axial flow agitator (56 rpm, 20 hp). Near complete
oxidation has been obtainable with both the sparge ring and sparge pipe in con-
junction with the agitator.
Both V/ST and TCA sparge systems operate from the same Worthington oil-free air
compressor under the following conditions: 50 psig, 270 F, and flow rates normally
at 210 scfm per scrubber. The compressor loading/unloading cycle is normally
4 seconds/10 seconds for 210 scfm. The oil-free compressor was chosen to minimize
the possibility of slurry contamination by oil which contains oxidation inhibitors.
Maintenance on the sparge systems has been mainly confined to the compressor. The
suction valve has been serviced and the compressor cooling water jacket has been
flushed periodically.
Further testing of forced oxidation with the sparqe ring and sparge pipe configura-
tions is planned.
67
-------
AUTOMATIC LIMESTONE ADDITION CONTROL
An automatic limestone feed control system was installed and initially tested on
the TCA in April-June 1977. The control loqic was based on a material balance
concept of maintaining a desired stoichiometric limestone feed in relation to the
amount of S02 absorbed in the scrubber. The results of the tests indicated that
limestone addition was satisfactorily controlled durinq normal fluctuations of
S02 inlet mass flow rates. As will be discussed, one observed weakness of the
control system was that the stoichiometric ratio was independent of the SO? inlet
mass flow rates and therefore the control system could not effectively compensate
for unusually large fluctuations in S02.
The basic control scheme is represented by the following equation and by Figure 1
L=Gx(S-K)xR
where: L = limestone slurry addition rate, gpm
G = flue gas flow rate, acfm
S = inlet S02 concentration, ppm
K = a manually adjustable constant related to destred outlet
S02 concentration, ppm
R = a manually adjustable constant proportional to the desired
stoichiometri c rati o
= (unit conversion factor) x (stoichiometric ratio)
Q
= 2.34 x 10 x (stoichiometric ratio)
The factor G(S-K) represents the amount of S02 absorbed per unit time. Thus
at a set value of R which is proportional to the desired stoichiometric ratio,
the limestone addition rate is automatically adjusted tc maintain the desired*
stoichiometry.
In practice, the inlet S02 concentration, S, and the flue qas flow rate, G
may vary within a wide range depending on the sulfur content in the coal and
the boiler load. Therefore, the control scheme also includes overrides which
are activated when the following situations arise:
a) If the measured outlet S02 concentration exceeds a set maximum, the
limestone addition rate will be stepped up to a preset maximum!
b) If the measured outlet S02 concentration drops below a set minimum
the limestone addition rate will be maintained at a preset minimum!
68
-------
FIGURE 1
ALKALI FEED CONTROL SCHEME
SCRUBBER
INLET GAS
SCRUBBER
OUTLET GAS
INLET
GAS FLOW
FE
DESIRED
OUTLET
S02
HIC
LIMESTONE
TO
SCRUBBER
1
PUMP
/OUTLET\
(SO2 LIMIT?
-------
The former provision insures compliance with the SOg emission standard,
while the latter reduces the posibility of pipe plugging.
Operating under typical conditions of 30,000 acfm, 1200 gpm liquor rate, and 3
beds each with 5 inches static height nitrile foam spheres, the desired'limestone
stoichiometric eatio for the TCA was set at 1.35 moles Ca added/mole S02 absorbed
(R = 2.34 x 10"° x 1.35 = 3.16 x 10"8). Actual stoichiometric ratio varied
between 1.18 and 1.55, with an average of 1.37. Actual limestone addition rate
was generally within 10 percent of the rate calculated from the control equation
presented above. A desired outlet S02 concentration, K, was set at 430 ppm.
Initially, overrides were set at 500 ppm and 200 ppm outlet S02 concentration.
Because of wide variation in the inlet S02 concentration (2100 to 3400 ppm),
the outlet S02 concentration frequently exceeded the 500 ppm upper override*!imit
This resulted in the actuation of the override control limit and interferred with'
the testing of the major control logic. Subsequently, the upper override limit
was raised to 800 ppm and the proportional control functioned more smoothly.
In theory, the higher the inlet S02 concentration, the higher the required
percent S02 removal in order to meet the S02 emission standard. The higher
required percent S02 removal would, in turn, call for higher limestone stoich-
iometric ratio. In the present setup, the constant R can only be reset manually
Therefore, as a second generation control logic, R as a function of the inlet
S02 concentration would be desirable.
OPERATIONAL -DEVELOPMENT
In the 6 years of operation the intent of the test facility has been to accelerate
the development and application of lime/limestone scrubbing technology. In addi-
tion to better understanding the chemistry of scrubbing, improving the sludge
disposal properties and enhancing S02 removal, countless developments and improve-
ments have been made in the operation of the facility. A few are listed here:
a) Mist eliminator plugging at Shawnee is extremely infrequent. Wash patterns
and wash sequences have been refined to such a degree that an excess of 7700
hours of operation was obtained recently between cleanings on the TCA mist
eliminator. The reason for the cleaning was that 12% plugging was present
and a new series of tests using adipic acid were to begin. The V/ST mist
eliminator was cleaned only once during the last 8000 hours of operation due
to problems with the wash system that caused 15% plugging. Considering the
wide range of operating conditions for the systems, these are outstanding
records.
b) The variable speed Allen-Sherman-Hoff rubber lined pumps have demonstrated
their effectiveness as slurry pumps under a wide range of operating condi-
tions.
70
-------
c) The process pH is measured using Uniloc Model 321 submersible electrode
assemblies. Originally, Uniloc Model 320 flow-through meters were installed.
But because of line plugging problems and frequent sensing electrode breakage
this type of sensor was abandoned. Current service requirements for the sub-
mersible assemblies consist of periodic cleaning and buffering of the electrodes,
generally every 2 or 3 days to insure accuracy. Also to minimize the service
requirement, the instrument electrodes are placed in water when the scrubbers
are not operating. Years of operation have shown that to insure the accuracy
of the process pH meters, a laboratory measured pH should be taken once every
four hours for comparison purposes. This procedure enables a normal operation
to within _+ 0.2 pH units of any desired set point.
d) A Dupont Model 400 UV split-beam photometer is used to measure S02 concentrations
In the last few years the instrument has been accurate and reasonably trouble-
free. Maintenance requirements are limited to cleaning the sample cell and
sample lines approximately once every 1 or 2 months and cleaning the particulate
filter usually once every 3 to 4 weeks. Ultraviolet lamp failure has been
the only component problem and has been caused by uncontrollable and
momentary power fluctuations due to the switching of station power. The
effective particulate filter for the instrument at Shawnee is a cylindrical
chamber constructed of a fine mesh screen. The screen cylinder is surrounded
by a solid protective cylinder. The gas sample lines have operated leak free;
the lines are 316L stainless steel tubing with heat tracing.
e) Oxygen in the inlet flue gas is measured with a Teledyne Model 9500 which uses
a micro-fuel cell. Operating performance has not been acceptable. A fre-
quent problem has been the rapid deactivation of the special micro-fuel cells.
Service life has varied from one day to approximately 1-1/2 months. In some
cases, the cells have arrived from the factory in a deactivated condition. The
causes of the cell deactivation might be due to exposure to a COg-free environ-
ment or factory defective cells, the problem with cell life continues and is
being studied.
f) The Foxboro 2800 series and 1800 series magnetic flow meters have shown no
serious problems. Periodic scale cleaning is required to improve accuracy
and sensitivity but the meters are considered reliable, acceptably accurate
and easily serviced.
g) Both Dynatrol Model CL-10HY U-tube density meters and Ohmart radioactive density
meters are used at Shawnee. Both meters provide acceptably accurate and de-
pendable service. From an operations point of view, the U-tube meters did have
some initial problems with line plugging. The cause was attributed to operator
error in setting too low a flow through in instrument. The problem has since
been resolved.
In the future, operational areas of interest will include but not be limited to
the continued effort to understand and control scale growth, improve dewatering
equipment operation, reduce the frequency of routine maintenance, and optimize
pump seal water requirements.
71
-------
STATUS REPORT OF
SHAWNEE COCURRENT AND DOWA SCRUBBER PROJECTS
AND
WIDOWS CREEK FORCED OXIDATION
J. L. Crowe
G. A. Hollinden
Energy Research
Tennessee Valley Authority
Chattanooga, Tennessee
and
Thomas Morasky
Desulfurization Processes Program
Fossil Fuel Power Plants Department
Electric Power Research Institute
Palo Alto, California
Prepared for Presentation at
Industry Briefing Conference
Results of EPA Lime/Limestone Wet Scrubbing Test Programs
Sponsored by the U.S. Environmental Protection Agency
Royal Villa Motel in Raleigh, North Carolina
August 29, 1978
72
-------
Introduction
In 1970, the Tennessee Valley Authority and the Environmental Protec-
tion Agency began a joint program at TVA's Shawnee Steam Plant to evaluate
processes which would remove sulfur dioxide and particulates from the
gaseous emissions of a coal-fired power plant. The major areas of concern
investigated are: the cost of removal, the reliability of the process,
the availability of materials needed and waste disposal of the byproducts
from the removal processes.
Three 10-MW scrubbers were constructed for testing by TVA. Each has
the capability of pulling flue gas from the No. 10 boiler either before
or after the electrostatic precipitator. Simultaneous testing has con-
tinued on various lime and limestone removal processes in an attempt to
lower capital and operating costs to make each process more reliable, to
optimize the processes, and to stabilize or utilize the waste products
from the processes.
Two advanced systems are presently being prepared for testing at a
10-MW size. The first is a recurrent scrubber which has the potential
advantage of a smaller scrubber vessel, thus lower capital cost over a
conventional scrubber. Testing of the cocurrent scrubber has been com-
pleted on an 1-MW pilot plant at Colbert Steam Plant and the 10-MW
prototype is in initial testing phase. The other process is the DOWA
process. This Japanese process is being marketed in the U.S. by Universal
Oil Products (UOP). This process can produce a stable, storable product
(gypsum) and, where demanded, a sellable product.
Forced Oxidation has been demonstrated at Shawnee Steam Plant on a
10-MW facility. TVA plans to demonstrate forced oxidation on a full-
scale facility at Widows Creek Steam Plant. Testing will include technical
73
-------
feasibility and environmental acceptability of forced oxidation as a
method of sludge disposal.
Cocurrent 1-MW Pilot Plant Results
EPRI funded TVA to evaluate the cocurrent scrubber concept at the
Colbert 1-MW pilot plant and to provide design data for a 10-MW prototype
to be constructed and operated at the Shawnee Scrubber Test Facility.
The emphasis of the evaluation was to study (1) the gas-liquid distribu-
tion characteristics of the absorber, (2) S(>2 and particulate removal
efficiencies as a function of gas velocity and liquid rates, and (3) the
effect of spray nozzle type and location and scrubber internals, such as
grids and packing, on S02 removal.
The gas-liquid distribution study consisted of operating the absorber
with (1) air only to determine the gas distribution and (2) air and water
to determine the liquid distribution. With the scrubber containing no
internals and operating at gas velocities of 12.6 and 19 ft/sec, the air-
flow was unsymmetrical at the scrubber inlet but became symmetrical at
the lower portions of the tower. Five superimposed sections of bar grids
(straightening vanes) were installed at the scrubber inlet, which improved
the flow profile down the absorber. Waterflow traverses during the air-
water tests revealed a poor liquid distribution at all levels. The total
flow rate and nozzle pressure had little effect on liquid distribution.
The data indicates that a large portion of the waterflow was on the walls
at the upper levels but disengaged from the wall as it proceeded down the
tower.
Gas velocity profiles were not as good and the liquid distriuution was
still poor, after the addition of six inches of Poly Grid packing to each
of the five bar grids. However, the packing did tend to cause less flow
on the scrubber walls.
74
-------
Tests were performed using sodium carbonate as the absorbent to
determine the maximum efficiency of the scrubber. With no internals
but the straightening vanes in the absorber the SO. removal ranged from
90 to 97 percent with an average of 94 percent which indicated good mass
transfer. The liquid rate had the greatest effect on S0_ removal; increasing
the liquid rate increased S02 removal. The spray nozzle type (pressure
drop) and its location had a lesser effect; an increase in nozzle pressure
drop increased S02 removal. A higher SO- removal was observed when the
total liquid flow was routed through the uppermost nozzles at the scrubber
inlet than when the liquid was distributed at various levels down the
absorber.
With lime as the absorbent, five 6-inch sections of the Poly Grid
packing were added to the bar grids to increase SO. removal from the
62 percent obtained with no internals to a prerequisite 85 percent at
established conditions (gas velocity of 19 ft/sec, liquid rate of 212 gpm,
and liquid to gas ratio of 56). During the test program, the SO- removal
ranged from 79 to 95 percent and averaged 88 percent. The effect of
the liquid rate was more dominant than that of the gas velocity. Increasing
the liquid rate was more dominant than that of the gas velocity. Increasing
the liquid rate increased SCL removal, while decrease in SO. removal
occurred when the gas velocity was increased. Higher SO- removals occurred
when the slurry was routed through the top than when it was distributed
throughout the tower. The nozzle pressure drop had no significant effect
on SO^ removal. Additional lime tests were made to determine the effect
of the number of scrubbing stages (gas-liquid contact time) on SO. removal.
With the same internals, the scrubbing slurry was successively introduced
above each grid down the absorber. Increasing the gas-liquid contact time
was found to improve SO. removal.
75
-------
Comparing the liquid distribution data with the SO. removal effi-
ciencies for these tests indicates that good SCL removals can be achieved
with a less than ideal gas-liquid distribution in the absorber. Other
scrubber types may also exhibit a similarly poor gas-liquid distribution
but there is little published data for comparison.
When limestone was used as the absorbent, an additional 3 inches of
Poly Grid packing (five 9-inch sections) were added to obtain the pre-
ferred S0_ removal efficiency of 85 percent at set conditions. During the
test program, the S02 removal ranged from 77 to 92 percent with an average
of 84 percent. The gas velocity had the greatest effect on SO- removal.
Increasing the gas velocity from base conditions lowered SO, removal
efficiencies. An increase in liquid rate increased SO. removal. The
nozzle pressure drop had no significant effect on S02 removal.
The particulate removal averaged 99.4 percent for both the lime and
limestone tests.
TVA completed the lime/limestone-scrubbing tests with the 1-MW cocurrent
pilot scrubber in mid-1977. These tests successfully demonstrated that the
cocurrent scrubber is an effective S02 scrubber. The results of these
pilot-plant tests were used to guide the design of a 10-MW improved proto-
type scrubber which was installed in the idle Hydro-Filter scrubber train
at the Shawnee facility. Figure 1 is the process flow diagram for the
new 10-MW prototype scrubber in the cocurrent mode. The possible advan-
tages of a cocurrent scrubber system over a conventional countercurrent
system are:
o The equipment configuration is more compatible with most power
plant duct and fan arrangements. The gas would enter the scrubber
at a higher elevation and exit near ground level. The entrainment
76
-------
FLUE GAS
INLET
RIVER
WATER
LIMESTONE SLURRY
FEED PUMP
ENTRAINMENT
SEPARATOR
CIRCULATION
PUMP
RIVER WATER -»
ENTRAINMENT
SEPARATOR
CIRCULATION
TANK
s
T
o
O
:RUBBER ,
r
-£
C
TANK
TO POND DISPOSAL
SCRUBBER
CIRCULATION
PUMPS
THICKENER
UNDERFLOW
PUMP
DISPOSAL
PUMP
FILTER CAKE
RESLURRY
TANK
VACUUM BELT FILTER
FILTRATE
PUMP
RECYCLE LIQUOR
SURGE TANK
RECYCLE LIQUOR
RETURN PUMP
FIGURE I
Process Flow Diagram - Improved Prototype Scrubber (CocurrenE Mode)
-------
separator and reheat systems (likely to require the most attention) would
be located near ground level. Likewise, the induced-draft (ID) fans would
be on the ground and the connecting ductwork to the stack would be shorter
and probably less complex.
o The physical arrangement of the proposed cocurrent system causes
the gas to change direction in the base of the unit before it enters the
entrainment separator. The change in direction, together with the vertical
position of the entrainment separator, promotes good liquid separation
and drainage. Also, a separate entrainment wash loop can be used if
needed.
o Scrubbing liquid would tend to coalesce into larger droplets before
it disengaged from the gas stream near the base of the scrubber. This
would further facilitate efficient operation of the mist eliminator.
o Flooding of the unit and associated high-pressure loss and
excessive entrainment of scrubbing slurry, even if grids are added to
improve gas-liquid contact, is less likely. Also, during normal cocurrent
operation, the gas-side pressure loss would be lower since some liquid-
side energy would be recovered.
o Higher gas velocites (smaller scrubbers) are expected because of
the reduced tendency to flood and because more efficient mist elimination
is likely. (The Colbert pilot tests were successfully performed at
approximately 30 ft/sec superficial gas velocity.) Therefore, smaller
or fewer scrubber modules would be required in a full-scale system.
To evaluate further the merits of the cocurrent scrubbing concept and
to obtain additional data for scale-up to a full-scale facility, EPRI
contracted with TVA to design, procure, erect and test the improved
prototype wet-scrubbing system (gas flow equivalent to 10-MW of generating
capacity) at the TVA Shawnee Steam Plant, Paducah, Kentucky.
78
-------
Test Objectives
The objectives of the scrubber test program are as follows:
1. To perform an operating variable study that will identify the
optimum operating conditions of the improved prototype scrubber
with respect to SO- and particulate removal efficiency.
2. To develop the scale-up similarities and differences between the
Colbert Test Program (1-MW) and the Shawnee Test Program (10-MW).
3. To determine design parameters for scale-up to a full-scale FGD
system.
4. To perform a long-term reliability demonstration of a cocurrent
prototype limestone scrubber system at optimum operating conditions.
5. To evaluate the collection efficiency, reliability, and materials
of construction of the entrainment separator.
6. To evaluate the performance of an inline-indirect steam reheater
with respect to heat transfer characteristics, materials of
construction, and reliability.
7. To determine the physical properties of the waste solids which
are produced during the reliability demonstration.
8. To evaluate the performance of other mechanical components within
the scrubber system, such as pumps, piping materials, valves,
and instrumentation.
General Test Program Outline and Schedule
The type of tests and order of testing will be generally as outlined
below:
1. Preoperational Testing (Final Equipment Checkout).
2. Sodium carbonate-scrubbing factorial tests to establish the
maximum removal efficiency of the scrubber in various operating
modes.
79
-------
3. Limestone-scrubbing operating variable study with a cocurrent
scrubber.
a. Factorial testing.
b. Short-term reliability tests.
4. Lime-scrubbing operating variable study with a cocurrent scrubber.
a. Factorial testing.
b. Short-term reliability tests.
5. Reliability demonstration at optimum operating conditions—
cocurrent limestone scrubbing.
A proposed schedule for the test program is shown in Figure 2.
The schedule is included to indicate the order of testing and relative
time to be assigned to each test block. It will be subject to change as
the test program proceeds. The length of individual tests will vary from
an 8-hr shift (factorial tests) to a month (reliability demonstration).
However, most tests will probably be 1-2 weeks in length to allow the
composition of the process liquor and solids to approach equilibrium.
TVA technical personnel will determine when the test objective has been
achieved and when the tests are to be terminated.
DOWA Process
In this process, SO- is absorbed in a clear solution of basic aluminum
sulfate. The spent absorbent is oxidized with air, then neutralized with
limestone to remove the sulfur in the form of gypsum. The regenerated basic
aluminum sulfate solution is recycled to the absorber.
PROCESS DESCRIPTION
The flue gas from the boiler flows directly into the absorber
(Figure 3). SO is absorbed by a counterflowing stream of basic aluminum
sulfate (Al (S0,)3 • A12 03). The absorption mechanisms are given in
80
-------
Proposed Test Program Schedule
Improved Prototype Scrubber
TA** Dl/i*«k
1 CSI BIOCK
Preoperationat Tests
Sodium Carbonate Factorial Tests
•
Cocurrent Mode Tests
Limestone - Factorial
Short-Term Reliability Tests
Lime- Factorial
Short -Term Reliability Tests
Long -Term Reliability Demonstration
Months
1
••
-
2
^••i
3
4
5
MM
6
••
7
8
9
10
II
12
13
14
Topical Reports
I-Report on sodium carbonate tests
2-Report on limestone tests
3-Report on lime tests
Figure 2
81
-------
Limestone Silo
Reheater
Desulfurized
Flue Gas
Mist /\
Eliminator
Gas from
Boiler—
OO
ro
Absorber
1
*—Water
AI2(S04)3
Aluminum Sulfate
Make-up Tank
•Air
Absorber Liquor
Mold Tank
Feeder
Limestone
Slurry Storage
Neutralizing
Tanks Thickene7
Vacuum
Vacuum Receiver
Reclaiming
Absorbent
Tank
To Pond
Waste Slurry
Tank
Figure 3. Dowo Typical Process Flow Diagram
-------
Reaction 1 (below). The cleaned gas passes through a mist eliminator
before entering the stack.
The spent absorbent is delivered to an oxidizer into which fine
air bubbles are injected. The sulfites in the solution are oxidized
to sulfates by Reaction 2. The bulk of the resulting solution is
recycled to the absorber.
The reminader of the oxidizer effluent is channeled to neutralizing
tanks. Limestone (CaCO,) is added to recover the basic aluminum sulfate
and precipitate the sulfur in the form of gypsum (Reaction 3). The
liquor from the neutralizing tank overflows into a thickener, where the
absorbent liquor is separated from the gypsum. Further separation of
the gypsum slurry from the thickener takes place in a filter. The
resulting gypsum is sent to disposal. The liquor from the filter is
mixed with that from the thickener and recycled to the absorber.
REACTIONS
Basic aluminum sulfate is used in the absorber to remove SO-. The
oxidizer is used to convert the resulting sulfites to sulfates. The
products are then regenerated in the neutralizer.
Absorber:
A12(S04)3 • A1203 + 3SO > Al (S04)g • A12(S03)3 (1)
Oxidizer:
2A12(S04)« A12(SO«)- + 30 £ > 4A1-(SO,)^ (2)
Neutralizer:
3CaC03 + 6H20 > (3)
• A120- + 3CaS04 • 2H20 + 3CO
83
-------
Advantages
(1) The process is not very complex.
(2) Since limestone is used as a neutralizing agent, operating costs
are low.
(3) Due to the low liquid-to-gas ratio, equipment size requirements
are low.
(4) Recovered gypsum my be of high quality and, thus, sellable,
wherever a demand for it exists. In this project, however,
the gypsum will be a disposal byproduct.
DISADVANTAGES/PROBLEMS
(1) Gypsum is not a very desirable commodity for recovery, since
other sources can easily supply the existing market.-
STATUS
TVA has negotiated with EPA for the use of the TCA for testing of
this process.
Initial engineering design and procurement are underway. TVA is
negotiating contracts with both UOP and EPRI for the project. An optimis-
tic schedule (Figure 4) calls for modification to the TCA to begin
during October with operation of the scrubber to begin in early January.
Test Objectives
The major objectives of this test and demonstration program are
as follows:
1. To demonstrate that the DOWA process can effectively treat
flue gas from a boiler which is fired with high-sulfur coal
and meet current emission standards.
84
-------
oo
Months from
date of UOP-
TVA contract
Phase I
Engineering and
Procurement
Phase II
Engineering and
Procurement
Phase III 8
Phase IV
Startup and
Operation
8
10
II
12
13
14
15
Figure 4. Dowa Project Schedule
-------
2. To evaluate the physical properties of the gypsum byproduct
and its suitability as a landfill material.
3. To determine the S02 removal efficiency and particulate removal
efficiency of the DOWA process over a broad range of operating
conditions.
4. To determine or confirm parameters for scale-up to a full-scale
FGD system.
General Test Program Description and Schedule
The types of test blocks and operating variables which will be con-
sidered for investigation during the test program are outlined below:
1. Equipment shakedown with air and water.
2. System shakedown and process demonstration at operating condi-
tions which are based on previous commercial experience in
oil-fired boiler applications.
3. Factorial tests - S02 and particulate removal efficiencies;
determine the effect of the following variables upon SO. and
particulate removal efficiencies:
a. Al concentration
b. Basicity
c. TCA pressure drop
d. TCA liquid recirculation rate
e. TCA superficial gas velocity
f. Absorber hold tank retention time
4. Factorial tests - Oxidation; determine the effect of the following
variables upon the oxidation efficiency:
a. Oxygen stoichiometry in the absorber loop
86
-------
c. Al concentration
d. Basicity
e. Absorber hold tank retention time
5. Factorial tests - Neutralization and byproduct production;
determine the effect of the following variables upon neutraliza-
tion reaction rate, settling rates of precipitates, filtera-
bility and final settled bulk density of byproduct gypsum, and
aluminum losses.
6. Short-term reliability tests.
Following the system shakedown and initial process demonstration,
process variable studies will begin with factorial tests which are designed
to screen the effect of each of the above listed variables. If technically
feasible and to conserve time, several of the factorial tests may be
performed simultaneously. Following the factorial tests, a series of
short-term reliability tests will be performed. The selection of operating
conditions for these tests will be based upon the results of the factorial
tests. The purpose of the short-term reliability tests is to obtain more
definitive information on the effect of operating conditions upon the
reliability of system components and to obtain operating data which will
be the basis for selection of operating conditions for a long-term
reliability demonstration. The conditions of the scrubber equipment will
be evaluated at the end of each test. Although the operating conditions
for all tests will be specified before each test block begins, the test
conditions will be subject to change as the results of the initial tests
are evaluated.
If results from this test program are favorable, additional funding
will be needed for more extensive variable studies and a long-term relia-
bility demonstration of the DOWA FGD system. As indicated above, the
87
-------
selection of the operating conditions for the long-term reliability test
should be based on the short-term reliability test results. System relia-
bility and economy of operation will be the major criteria which should be
used in the selection of operating conditions. Also, only operating
conditions which have met the current emission standards should be selected
During the reliability test, all the scrubber operating conditions should
be held constant except for the flue gas rate which will be varied in
proportion to the boiler load. During this test the physical and chemical
properties of the gypsum byproduct should be routinely determined, including
settling rate, final settled bulk density, compressive strength, filtera-
bility, particle size distribution, chemical analysis, and general crystal
form. To the extent possible, pertinent physical properties should be
correlated with the operating conditions of the process unit.
The preoperational tests (equipment shakedown) and the test program
will be conducted during a 4-month period. A test program schedule which
indicates the order of tests and the relative amount of time assigned to
each test will be proposed to EPRI and UOP one month prior to startup. The
test program schedule will be subject to revision as the test results are
evaluated. All schedule changes will be approved by EPRI and UOP.
88
-------
DEMONSTRATION OF FORCED OXIDATION AT
WIDOWS CREEK UNIT 8
The purpose of oxidizing lime/limestone FGD product sludges is to
convert calcium sulfite (CaSO_), the normal product of these scrubber
processes to calcium sulfate (CaSO,). Calcium sulfate, which is commonly
known as gypsum, is a more desirable waste product because it improves the
settling and dewatering properties of the sludge. This in turn reduces
the volume of material for disposal and makes a material which may be
suitable for landfill without need of additives or the use of mixing or
blending equipment.
This project is designed to develop, demonstrate, and evaluate, at the
full-scale level, the technical feasibility and environmental acceptability
of utilizing oxidation as a method of sludge disposal. This will be
accomplished by testing a forced oxidation system on the "D" train
(140-MW equivalent) of the Widows Creek unit 8 wet limestone scrubber.
A slipstream of 25 percent of the full load flow of the effluent slurry
will be treated in a thickener and vacuum filter.
Tests will include two-stage forced oxidation, single-stage forced
oxidation, and oxidation in both stages (venturi and absorber). The effect
of such variables as air stoichiometry, pH, and limestone stoichiometry
will also be evaluated.
Combustion Engineering, Inc., under a contract with the Tennessee
Valley Authority will be responsible for the design, procurement, erection,
testing, and reporting of this oxidation demonstration program.
A. Widows Creek Simulation at Shawnee
Essentially complete oxidation of calcium sulfites from the scrubbers
operated at the Shawnee Test Facility has been routinely achieved during
89
-------
forced oxidation testing on both the TCA and the venturi/spray tower systems.
However, in pilot-plant scale tests conducted at the TVA Colbert plant,
the sludge could not be oxidized in any practical manner. These opposed
results indicate some undetermined site-specific conditions that either
promotes oxidation at Shawnee or conversely deters oxidation at Colbert.
Since the natural oxidation at Widows Creek and Colbert is essentially
the same and is significantly lower than that obtained at Shawnee, con-
siderable concern has arisen over the Colbert-Shawnee correlation since
extrapolation may be possible to the forced oxidation mode. Should this
be the case, little to no oxidation could be achieved on the Widows Creek
system if a forced oxidation should be desired for use as a method of
treating the sludge. In order to predict achievable oxidation at Widows
Creek, a special coal burn at Shawnee using Widows Creek coal, scrubbing
with Widows Creek limestone, and using the venturi-spray tower oxidation
mode as operated in the EPA test program is planned. The main problem
would be the difference in boilers for the two systems and the possible
effect this could have on oxidation; Shawnee—B&W, Widows Creek unit 8—
CE. These tests would simulate the Widows Creek scrubber design as much
as possible so that problem areas could be identified.
B. Widows Creek Forced Oxidation
TVA will also evaluate oxidation on one of the scrubber trains at the
Widows Creek unit 8 facility such that data can be collected on the
feasability of oxidizing the sludge under Widows Creek operating condi-
tion. Figure 5 is a flow diagram of the Widows Creek Forced Oxidation
System. Should oxidation prove to be technically feasible an economic
evaluation will be made for comparison to other methods of disposal,
i.e., sludge-fly ash blending. This project is designed to develop,
90
-------
Reheat
Absorber Circ. Tank
Compressor
Entrapment
Separator
t
7
I
I
1
>« >* Tr
/
> Stack
From Powerhouse
—Venturi Circ. Tank
Bleed Stream
Thickener
Thickener Overflow
*• To Effluent
Slurry Surge
Tank
Vacuum
Filter
To Truck
Figure 5. Flow Diagram - Widows Creek Forced Oxidation
-------
demonstrate, and evaluate the technical feasibility and environmental
acceptability of oxidation as a method for disposal of sludge from the
No. 7 and No. 8 wet/limestone scrubber units at Widows Creek. The primary
purpose for oxidizing the scrubber solids is to improve waste solids
dewatering and landfill disposal characteristics. This objective will be
accomplished through specific tests which will be made to meet the following
criteria.
1. The project should result in an evaluation that will provide a
reference point for making confident decisions on the feasibility
of converting the scrubbers at the Widows Creek station to the
forced oxidation mode.
2. Provide the results of this evaluation by March 1979 so that a
sludge disposal system can be installed and be ready for use
before the Widows Creek pond is filled.
Figure 6 is a schedule for the Widows Creek Forced Oxidation System.
This study will also include a comparative economic study of forced
oxidation vs. other sludge disposal processes as they relate to the Widows
Creek system.
Other major factors that will be included for evaluation are:
1. Transportability of the oxidized solids after dewatering.
2. Primary application for the dewatered solids.
3. Material handling equipment required.
4. Methods of disposal or reuse of waste water resulting from
dewatering.
5. Effect of dewatering for landfill treatment.
92
-------
Engineering and Procurement
Construction
Testing
AUG
1978
SEPT
1978
OCT
1978
NOV
1978
DEC
1978
JAN
1979
FEB
1979
MAR
1979
Figure 6. Widows Creek Forced Oxidation Schedule
-------
CURRENT STATUS OF DEVELOPMENT OF THE SHAWNEE LIME-LIMESTONE
COMPUTER PROGRAM
C. D. Stephenson and R. L. Torstrick
Emission Control Development Projects
Office of Agricultural and Chemical Development
Tennessee Valley Authority
Muscle Shoals, Alabama
Prepared for Presentation at
Industry Briefing Conference
Results of EPA Lime/Limestone Wet Scrubbing Test Programs
Sponsored by the U.S. Environmental Protection Agency
Royal Villa Motel in Raleigh, North Carolina
August 29, 1978
94
-------
CURRENT STATUS OF DEVELOPMENT OF THE SHAWNEE LIME-LIMESTONE
COMPUTER PROGRAM
GENERAL SCOPE AND PURPOSE
In conjunction with the U.S. Environmental Protection Agency (EPA)-
sponsored Shawnee test program, Bechtel National, Inc., and the Tennessee
Valley Authority (TVA) have jointly developed a computer program capable
of projecting comparative investment and revenue requirements for lime
and limestone scrubbing systems. The computer program has been developed
to permit the estimation of relative economics of these systems for varia-
tions in process design alternatives (i.e., limestone vs lime scrubbing,
alternative scrubber types, or alternative sludge disposal methods) or
variations in the values of independent design parameters (i.e., scrubber
gas velocity and L/G ratio, alkali stoichiometry, slurry residence time,
reheat temperature, and specific sludge disposal design). Although the
program is not intended to compute the economics of an individual system
to a high degree of accuracy, it is based on sufficient detail to allow
the quick projection of preliminary conceptual design and costs for
various lime-limestone case variations on a common design and cost basis.
PROGRAM DEVELOPMENT
The responsibility in the development of the computer program was shared
by Bechtel and TVA. Bechtel*s major responsibility was to analyze the
results of the Shawnee scrubbing tests and develop models for calculating
the overall material balance flow rates and "tream compositions. Bechtel
provided TVA with a complete computer program for specifying this informa-
tion. TVA was responsible for determining fhe size limitations of the
required equipment for establishing the minimum number of parallel equip-
ment trains, accumulating cost data for the major equipment items, and
developing models for projecting equipment and field material costs as a
function of equipment capacity. Utilizing these relationships TVA
developed models to project the overall investment cost breakdown and a
procedure for using the output of the material balance and investment
models as input to a previously developed TVA program for projecting
annual and lifetime revenue requirements.
95
-------
PROGRAM DESCRIPTION AND DOCUMENTATION
TVA has presented two papers which reported the status of the program
development and displayed its capabilities. The first, titled "Shawnee
Limestone-Lime Scrubbing Process Computerized Design-Cost Estimate
Program: Summary Description Report," was given at the Industry Briefing
Conference sponsored by EPA at Raleigh, North Carolina, October 14, 1976.
The second, titled "Economic Evaluation Techniques, Results, and Computer
Modeling for Flue Gas Desulfurization," was presented at the EPA Flue
Gas Desulfurization Symposium in Hollywood, Florida, November 8-11, 1977.
A significant number of revisions have been incorporated into the program
since these earlier reports. The present paper describes the current
capabilities of the computer program. Since the design basis for the
lime and limestone scrubbing systems has not changed appreciably from
the earlier publications, it will not be included in this paper. TVA
is currently in the process of preparing a users manual for the overall
program which will include all the information required for running the
Turbulent Contact Absorber (TCA) program option. It is anticipated that
the users manual will be available within the next few months.
CURRENT PROGRAM SCOPE
Uses and Limitations
The present computer program has the capability of projecting a complete
conceptual design package for lime or limestone scrubbing utilizing a
TCA with any one of four sludge disposal options (discussed later). The
program is designed to consider new coal-fired power units ranging in
size from 100-1300 MW. Equipment size and layout configurations are
modeled based on coals ranging in sulfur contents from 2-5%. To
limit the extremely wide variations in equipment sizes and layout con-
figurations which can result with changes in other key independent
variables, the following range of values for these variables was
established.
Scrubber gas velocity 8-12.5 ft/sec
Liquor recirculation rate 25-75 gal/kft9
Slurry residence time in hold tank 2-25 min
However, operating parameters and plant sizes outside of these ranges
will not necessarily be invalid.
It is expected that the results may within limits also be valid for
extrapolation of sulfur content of coal beyond the range actually tested
at Shawnee. The Shawnee models are based on scrubbing results over an
S02 concentration range of approximately 1500 to 4000 ppm.
The effect of variations in any of the inputs, such as scrubber gas
velocity, degree of S02 removal, reheat temperature, alkali stoichi-
ometry, L/G ratio, etc. , on process design and economics may be determined.
96
-------
For both lime and limestone scrubbing, SQ2 removal, stoichiometry (pH for
the lime option), and L/G ratio may all be specified and results projected.
Alternatively, S02 removal and stoichiometry (or pH) may be specified and
L/G calculated, or S02 removal and L/G ratio may be specified and stoichi-
ometry (or pH) calculated. An additional option is being incorporated
into the program allowing for the calculation of S02 removal based on
input values of L/G and stoichiometry (or pH).
The outputs of the overall computer program include (1) a detailed
material balance including properties of the major streams, (2) a detailed
water balance Itemizing water availability and water required, (3) speci-
fications of the scrubber system design, (4) a revised method for dis-
playing overall pond design and costs, (5) specifications and costs of the
process equipment by major processing area, (6) a detailed breakdown of
the projected capital investment requirements, (7) an itemized breakdown
of the projected revenue requirements by component for the first year of
operation of the system, and (8) a lifetime revenue requirement analyses
showing projected costs for each year of operation of the plant, as well
as lifetime cumulative and discounted costs and equivalent unit revenue
requirements.
New Program Options and Modifications Since the November 1977 FGD Symposium
To illustrate the current program inputs and outputs, an example run of the
updated computer program is shown in the appendix for limestone scrubbing
with an onsite pond disposal option. Discussions of the modifications incor-
porated into the program since the November 1977 FGD symposium are given below,
Particulate Removal—
The quantity of ash in the coal which is emitted overhead as fly ash is an
input. Additional inputs are required to specify ash removal upstream of
the scrubber and within the scrubber. Th^se are input either as a removal
efficiency in percent or as an equivalent 'iitlet emission in pounds per
million Btu. If removal is not input, a 33% efficient mechanical collector
is provided for protecting the fan from abrasion by large fly ash particles.
A cost model is available for optionally including the costs for the mechani-
cal collector. As discussed later, cost models for a high-efficiency
electrostatic precipitator (ESP) or baghouse are presently being incorpo-
rated into the program, but are not yet available for the current version.
An example output of the fly ash removal option is shown on pages 10 and 13 of
the appendix under the heading "Fly Ash Removal."
SO2 Removal—
The percentage of the sulfur in coal which is emitted overhead as S02 is an
input to the program. The degree of S02 removal may be input by specifying
either
1. % S02 removed,
2. Ibs S02 emitted per million Btu of heat input to the boiler or
3. ppm S02 in outlet flue gas.
97
-------
For each computer run, equivalent S02 emissions are displayed on all
three basis, regardless of the method for inputting the degree of S02
removal. The alternative methods for specifying S02 removal are
illustrated on page 14 under the heading "Flue Gas to Stack."
Redundancy—
In addition to designing the flue gas desulfurization (FGD) system with
spare operating pumps and optionally with spare scrubbing trains as
described in the earlier publications, spare feed preparation units may
now be specified. This is applicable for both limestone scrubbing in
which spare ball mill trains can be provided, and lime scrubbing in
which spare slakers can be specified. The number of redundant alkali
preparation units and redundant scrubber trains are specified under
"Raw Material Handling Area" (pp.n and 20 ) and "Scrubber System Variables"
(pp. 11, 16, and 22) respectively.
Water Balance—
The program has previously assumed no net accumulation or loss of water
due to rainfall, evaporation, and seepage. The current version of the
program allows for specific rainfall, evaporation, and seepage rates to
be input for accurately projecting makeup water requirements. The water
balance inputs and outputs for an example run are shown on page 15.
Waste Disposal Options—
The program allows for four alternate waste disposal alternatives to be
assessed including:
1. Onsite ponding
a. Unllned pond
b. Clay-lined pond (cost of clay and depth of lining input)
c. Synthetic-lined pond (cost of liner input)
2. Thickener - ponding
3. Thickener - fixation fee
4. Thickener - filter - fixation fee
The onsite ponding options may also be run with fixation fees applied to
them. For alternatives 3 and 4, the fixation fee must include costs for
transportation and disposal of the fixed sludge offsite. For alternatives
1 and 2, however, only the costs for fixation need be provided since the
fixed sludge can be disposed of at the existing pond site.
For the ponding alternatives the program allows for the onsite pond to
be sized larger or smaller than the normal projected lifetime capacity.
-------
This option, hag been incorporated (1) to account for variations in the
sulfur content of fuel, (2) to evaluate design philosophy in construc-
tion ponds for less than the total amount of sludge to be disposed of
(requires assessment of additional cost for expanding pond later), or
(3) to allow the feed preparation and scrubbing areas to be sized based
on maximum sulfur contents expected, while sizing the pond based on
average sulfur contents. An example output of the revised onsite ponding
mpdel is shown on page 19.
Equipment Cost Breakdown—
The program has been modified to provide a breakdown showing projected
equipment specifications and cost for each equipment item. Both material
and labor components of equipment costs are displayed for each of the
three major areas. The equipment list and costs for limestone scrubbing
and onsite ponding are illustrated on pages 20-23. Although a complete
printout for the lime and alternate sludge disposal options is not
included in this paper, an equipment list for those options is illus-
trated on pages 27-30.
Operating Profile—
The current version of the program allows for the specification of three
alternative operating profiles as indicated below for projecting lifetime
revenue requirements:
1. Profile similar to that utilized in the report Detailed Cost
Estimates for Advanced Effluent Desulfurization Processes"
by G. G. McGlamery, et al. (EPA-600/2-75-006, January 1975)
2. Historical power plant operating profile based on FPC Form 67 data
3. Variable profile with annual load factors as input
See pages 31-32 for illustration.
FUTURE PROGRAM DEVELOPMENT
Further additions to the program are expected to be made as additional
test data from Shawnee become available. Bechtel and TVA are currently
incorporating the results of the venturi-spray tower tests at Shawnee
into a design and cost model for that option. In addition, cost models
for upstream fly ash removal by hot or cold side ESP, baghouse collectors
and venturi scrubbers a,re being incorporated into the program. Other
options which are being considered for incorporation as sufficient data
become available include (1) series scrubbers/high alkali utilization
systems and (2) forced oxidation systems.
-------
PROJECTED PROGRAM USE
Upon completion of the overall effort, the program will be useful for
projecting a complete conceptual design package for lime-limestone scrub-
bing including material balance, capital investment estimate, and
projected revenue requirements. It is expected that the program will
be used by utility companies and architectural and engineering con-
tractors involved in the selection and design of S02 removal facilities
for specific applications. It is not intended to be used for projecting
a final design of a given system, but to assist in the evaluation of
system alternatives prior to development of a detailed design. Also,
the program will be useful for evaluating the potential impact of various
process variables on economics as a guide for planning research and
development activities.
Although the program was not meant to be used for comparing projected
lime-limestone economics with economics for alternate processes, these
comparisons should be valid as long as the basis for the alternate process
economics are comparable to those included in the computer program for
lime-limestone systems.
Method for Attaining Results
TVA is in the process of loading the current version of the program on
a Control Data Corporation (CDC) timesharing computer system and publish-
ing a users manual for utilizing the program. After this effort is
complete, outside users will be allowed to access the computer program
for making computer runs. Until these activities are completed, TVA,
under a Technology Transfer Contract with EPA, can upon request make
computer runs for interested users, or can release copies of the program
to interested users along with available documentation for running it.
Current Users and Program Applications
A significant number of responses have already been handled under the
above arrangement.
Given below is a list of people who have been provided tape copies of
the Shawnee lime-limestone computer program.
Robert H. Boeckmann, Gibbs and Hill, Inc.
Paul S. Farber, Argonne National Laboratory
D. J. Hagerty, University of Louisville
J. Scott Hartman, PEDCo
R. G. Knight, Michael Baker, Jr., Inc.
M. Lieberstein, The City of New York,
Environmental Protection Administration
F. Y. Murad, Combustion Equipment Associates, Inc.
Edward S. Rubin, Carnegie-Mellon University
J. G. Stevens, Exxon
John Valente, Air Correction Division, UOP
John Wysocki, Burns and Roe, Inc.
100
-------
A list of people who have been provided results of specific computer
runs is shown below:
Randy M. Cole, Tennessee Valley Authority
Wilson Cramer, U.S. Steel Corporation
Richard Furman, Florida Power and Light
Robert Gosik, Environmental Protection Agency (Denver)
Robert Lane, Illinois Commerce Commission
S. J. Lutz, TRW, Inc.
M. F. Patterson, Linde Division, Union Carbide Corporation
A. V. Slack, SAS Corporation
John Wile, National Economic Research Associates, Inc.
To date several uses of the program other than those for which it was
intended have been tried. The program has been run simulating both
industrial and utility boilers, smelter situations, partial scrubbing
situations, plant optimization studies, and for comparisons of front
end coal cleaning economics with total scrubbing. Probably the most
important use to date has been support work for the National Economic
Research Association's assessment of the impact of possible NSPS
revisions on the electric utilities industry on a national scale.
[EPA issued Federal Standards of Performance for New Stationary Sources
(often called "new source performance standards" or NSPS).] Examples
of some of the sensitivities which might be assessed are shown on
pages 33-46.
101
-------
APPENDIX
INPUT DATA FOR EXAMPLE
PROGRAM RUN
11111
111111111011000
1 1 1
INDUSTRY BRIEFING-500 MW
0 500 9000 10500 33 300 2 175 470 751
57.56 4.14 7.00 1.29 3.12 0.15 16.00 10.74 95 80 1 97 5 an
50 20 12.5 25 2 1.2 12 1 0.0 1 0 0 2.85 500
15 40 .2 40 30 60 1.2 7.0 1 100 0
2 3 4 5 51 .000001 42 10 1.35 142.1
1 12 9999 3500 25 25 5280 1 10 2.5
9 16 5 10 8 12 11.6 8 3 50 10 17.2 1.17
8 50 12.5 2.0 0.12 0.029 17 220.9 178.2 1977 1978
END
102
-------
OUTPUT OF EXAMPLE RUN
*** INPUTS ***
BOILER CHARACTERISTICS
MEGAWATTS = 500.
BOILER HEAT RATE = 9000, BTU/KUH
EXCESS AIR = 33. PERCENT, INCLUDING LEAKAGE
HOT GAS TEMPERATURE - 300. DEC F
COAL ANALYSIS, UT 7. AS FIRED :
C H 0 N S CL ASH H20
57.56 4.14 7.00 1.29 3.12 0.15 16.00 10.74
SULFUR OVERHEAD = 95.0 PERCENT
ASH OVERHEAD = 80.0 PERCENT
HEATING VALUE OF COAL = 10500. BTtl/LB
EFFICIENCY, EMISSION,
FLYASH REMOVAL % |_BS/M RTU
UPSTREAM OF SCRUBBER 97.5 0.30
UITHJN SCRUBBER 80.0 0.06
COST OF UPSTREAM FLYASH REMOVAL EXCLUDED
ALKALI
LIMESTONE ?
CAC03 = 97.15 I.JT 7. DRY BASIS
SOLUBLE MGO = 0.0
INERTS = 2,85
MOISTURE CONTENT = 5.00 LB H20/100 LBS DRY LIMESTONE
LIMESTONE HARDNESS WORK INDEX FACTOR = 10 00
LIMESTONFT DEGREE OF GRIND FACTOR = 1,35
FLY ASH :
SOLUBLE CAO = 0.0 U] %
SOLUBLE MGO •- 0.0
INERTS = 100.00
103
-------
RAW MATERIAL HANDLING AREA
NUMBER OF REDUNDANT ALKALI PREPARATION UNITS =
SCRUBBER SYSTEM VARIABLES
NUMBER OF OPERATING SCRUBBING TRAINS = 4
NUMBER OF REDUNDANT SCRUBBING TRAINS = 2
NUMBER OF BEDS =3
NUMBER OF GRIDS = 4
HEIGHT OF SPHERES PER BED = 5,0 INCHES
LIQUID-TO-GAS RATIO = 55. GAL/1000 ACF
SCRUBBER GAS VELOCITY = 12.5 FT/SEC
S02 EMISSION LIMIT = 1.20 LB S02/M BTU
STOICHIOMETRY RATIO TO BE CALCULATED
ENTRAINMENT LEVEL = 0.10 WT X
EHT RESIDENCE TIME - 12.0 MIN
S02 OXIDIZED IN SYSTEM = 30.0 PERCENT
SOLIDS IN RECIRCULATED SLURRY = 15.0 WT %
SOLIDS DISPOSAL SYSTEM
COST OF LAND = 3500.00 DOLLARS/ACRE
SOLIDS IN SYSTEM SLUDGE DISCHARGE = 40,0 WT %
MAXIMUM POND AREA = 9999. ACRES
MAXIMUM EXCAVATION = 25.00 FT
DISTANCE TO PONP = 5280, FT
POND LINED WITH 10.0 INCHES CLAY
104
-------
STEAM REHEATER (IN-LINE)
SATURATED STEAM TEMPERATURE = 470. KEG F
HEAT OF VAPORIZATION OF STEAM = 751. BTU/LB
OUTLET FLUE GAS TEMPERATURE = 175. HEG F
SUPERFICIAL GAS VELOCITY (FACE VELOCITY) = 25.0 FT/SEC
IT SR SROLD
1 1,24 1.50
2 1.24 1.24
3 1.24 1.24
PAPER PRINT 01 r POND FEE
105
-------
*#X< OUTPUTS *.**
HOT GAS TO SCRUBBER
C02
HCt
S02
02
N2
H20
S02
MOLE PERCENT LB-HOLE/HR
12 . 3 15 0.2054E+05
0*011 0.1813E+02
0.238 0.3962l£i03
4,827 0.8050E+04
73.867 0.1232E+06
8.743 0.1458E+05
CONCENTRATION IN SCRUBBER INLET
FLYASH EMISSION = 0.30 LBS/MILLION
= 0.152 GRAINS/SCF
SOLUBLE CAO IN FLY ASH =
SOLUBLE MGQ IN FLY ASH =
LB/HR
0.9040E+06
0.6612E+03
0.2538E+05
0.2576E+06
0.3452E+07
0.2627E+06
GAS = 2376. PPM
BTU
(WET) OR 1371. LB/HR
0. LB/HR
0.
HOT GAS FLOW RATE = .1054E+07 SCFM (60 DEC FT 1 ATM)
= ,1540E+07 ACFM (300. DEG Fr 1 ATM)
CORRESPONDING COAL FIRING RATE = .4286E+06 LB/HR
HOT PAS HUMIDITY = 0.057 LB H20/LB DRY GAS
WET BULB TEMPERATURE = 127. BEG F
UET GAS FROM SCRUBBER
MOLE PERCENT LB-MOLE/HR
C02
S02
02
N2
H20
11.683
0.047
4.480
68.969
14.820
S02 CONCENTRATION
FLYASH
EMISSION =
«
0.2087E+05
0.8430E+02
0.8003E+04
0.1232E+06
0.2647E+05
IN SCRUBBER OUTLET
0.06 LBS/MILLION
0.02R GRAINS/SCF
LB/HR
0.9185E+06
0.5400E+04
0.2561E+06
0.3452E+07
0.4769E+06
GAS = 472. PPM
BTU
(WET) OR 274. LB/HR
TOTAL WATER PICKUP = 439. GPM
INCLUDING 10.2 GPM ENTRAINMENT
WET GAS FLOW RATE = .1128E-1-07 SCFM (60 DEG Fj 1 ATM)
- .1274E+07 ACFM (127. DEG F» 1 ATM)
WET GAS SATURATION HUMIDITY = 0.103 LB H20/LB DRY GAS
106
-------
FLUE GAS TO STACK
MOLE PERCENT
C02 11,665
S02 0.047
02 4.473
N2 68,860
H20 14,955
LB-MOLE/HR
0,20S7E-f-05
0.8430E402
0.8003E+04
0.1232E+06
0.2676E+05
LB/HR
0.9185E+06
0.5400E+04
0.2561E-f06
0.3452E+07
0.4820E+06
CALCULATED S02 REMOVAL EFFICIENCY = 78.8 %
SPECIFIED S02 EMISSION = 1.20 POUNDS PER MILLION BTU
CALCULATED R02 CONCENTRATION IN STACK GAS = 471,
FLYASH EMISSION = 0.06 LBS/MILLION BTU
- O.O28 GRAINS/SCF
-------
ATER BALANCE INPUTS
RAINFALL(IN/YEAR)
POND SEEPAGE(CM/SEC)*10**3
POND EVAPORATION(IN/YEAR)
51.
100.
42,
IATER BALANCE OUTPUTS
JATER AVAILABLE
RAINFALL
ALKAL.I
TOTAL
658, GPM
4. GPM
662, GPM
328778. LB/HR
1995. LB/HR
330773. LB/HR
•JATER REQUIRED
HUMIDIFICATION
ENTRAINMENT
DISPOSAL WATER
HYDRATION UATER
CLARIFIER EVAPORATION
POND EVAPORATION
SEEPAGE
TOTAL WATER REQUIRED
429, GPM
10, 0PM
157, GPM
11. GPM
0, GPM
575, GPM
160. GPM
1342, GPM
214217.
5103.
78373.
5307.
0.
287557.
80040.
LB/HR
LB/HR
LB/HR
LB/HR
LB/HR
LB/HR
LB/HR
670596. LB/HR
!ET UATER REQUIRED
680, GPM
339823. LB/HR
108
-------
SCRUBBER SYSTEM
TOTAL NUMBER OF SCRUBBING TRAINS (OPERATING+REDUNDANT) = 6
S02 REMOVAL = 78.7 PERCENT
PARTICULATE REMOVAL IN SCRUBBER SYSTEM = 80.0 PERCENT
TCA PRESSURE DROP ACROSS 3 BEDS = 8.6 IN. H20
TOTAL SYSTEM PRESSURE DROP = 14.8 IN..H20
SPECIFIED LIQUID-TO-GAS-RATIO = 55. GAL/1000 ACF
LIMESTONE ADDITION = 0.3990E+05 LB/HR DRY LIMESTONE
CALCULATED LIMESTONE STOICHIOMETRY = 1.24 MOLE CAC03 ADDED AS LIMESTONE
PER MOLE S02 ABSORBED
SOLUBLE CAO FROM FLY ASH = 0.0 MOLE PER MOLE S02 ABSORBED
TOTAL SOLUBLE MGO = 0.0 MOLE PER MOLE S02 ABSORBED
TOTAL STOICHIOMETRY = 1.24 MOLE SOLUBLE (CA+MG)
PER MOLE S02 ABSORBED
SCRUBBER INLET LIQUOR PH = 5.34
MAKE UP WATER = 680. GPM
109
-------
SYSTEM SLUDGE DISCHARGE
SPECIES
H20
CAS04 .2H20
CAC03
INSOLUBLES
H20
CA++
MG-H-
S03—
S04—
CL-
LB-MOLE/HR LB/HR
0.2182E+03
0.9272E+02
0.6629E+02
0.4350E+04
0.9937E+01
0.0
0.1254E4-00
0.8489E+00
O.1791E-H02
0.2817E+05
0. 1596E4-05
0.6635E+04
0.2234E+04
0.7837E+05
0.3983E+03
0.0
0.1004E4-02
0.8154E4-02
0.6351E+03
SOLID
COMF»
UT X
53.16
30.11
12,52
4.22
LIQUID
COUP,
PPM
5010.
0.
126.
1026.
7988.
TOTAL DISCHARGE FLOW RATE = 0.1325E+06 LB/HR
200. GPM
TOTAL DISSOLVED SOLIDS IN DISCHARGE LIQUID = 14150. PPM
DISCHARGE LIQUID PH = 7.24
SCRUBBER SLURRY BLEED
TOTAL FLOW RATE = 0.3533E+06 LB/HR
642. GPM
TOTAL SUPERNATE RETURN
TOTAL FLOW RATE = 0.1820E+Q6 LB/HR
364. GPM
110
-------
SUPERNATE TO WET BALL MILL
TOTAL FLOW RATE = 0.2461E+05 LB/HR
49. GPM
LIMESTONE SLURRY FEED
TOTAL FLOW RATE = 0.6A51E+05 LB/HR
=84. GPM
SUPERNATE RETURN TO SCRUBBER OR EHT
TOTAL FLOW RATE = 0.1574E+OA LB/HR
= 315. GPM
RECYCLE SLURRY TO SCRUBBER
TOTAL FLOW RATE = 0.3859E+08 LB/HR
= 70087. GPM
FLUE GAS COOLING SLURRY
TOTAL FLOW RATE = 0.2807E+07 LB/HR
5097. GPM
111
-------
POMD DFSIGM
OPTIMIZED TO MINIMIZE TOTAL COST PLUS OVERHEAD
POND DIMENSIONS
DEPTH. OF POND
DEPTH OF EXCAVATION
LENGTH OF PERIMETER
LENGTH OF DIVIDER
AREA OF BOTTOM
AREA OF INSIDE WALLS
AREA OF OUTSIDE WALLS
AREA OF POND
AREA OF POND SITE
AREA OF POND SITE
VOLUME OF EXCAVATION
VOLUME OF SLUDGE TO BE
DISPOSED OVER LIFE OF PLANT
20.fc2 FT
3.24 FT
13193. FT
2403. FT
1074. THOUSAND YD2
129. THOUSAND YD2
9
-------
RAW MATERIAL. HANDLING AND PREPARATION
INCLUDING 2 OPERATING AND 1 SPAKE PREPARATION UNITS
ITEM
DESCRIPTION
NO. MATERIAL
LABOR
CAR SHAKER AND HOIST
CAR PULLER
UNLOADING HOPPER
UNLOADING VIBRATING FEEDER
UNLOADING BELT CONVEYOR
UNLOADING INCLINE PEl.T
CONVEYOR
UNLOADING PIT MIST COLLECTOR
UNLOADING HOPPER
UNLOADING VIBRATING FEEDER
UNLOADING KELT CONVEYOR
UNLOADING INCLINE BELT
CONVEYOR
UNLOADING PIT MIST COLLECTOR
UNLOADING PIT SUMP PUMP
STORAGE" BELT CONVEYOR
STORAGE CONVEYOR TRIPPER
MOBILE EQUIPMENT
RECLAIh HOPPER
RECLAIM VIBRATING FEEDER
RECLAIM BELT CONVEYOR
RECLAIM INCLINE BELT CONVFYOR
RECLAIM PIT DUST COLLECTOR
RECLAIM PIT SUMP PUMP
RECLAIM BUCKET ELEVATOR
FEED BELT CONVEYOR
FEE!" CONVEYOR TRIPPER
20HP SHAKER 7.5HP HOIST
25HP PULLER , 5HP RETURN
14FT DIA, 10FT STRAIGHT
SIDE HT, CS
3.5HP
20FT HORIZONTAL r f-HP
310FT, 50HP
POLYPROPYLENE BAGTYPE,
2200 CFM,7.5HP
J6FT niA, 10FT STRAIGHT
SIDE HT, CS
3.5HP
20FT HORIZONTAL* 5HP
310FTr 50HP
POLYPROPYLENE. BAGTYPE,
2200 CFM»7.5HP
60GPM, 70FT HEAD, 5HP
200FT, 5HP
30FPM, 1HP
SCRAPPER TRACTOR
1
1
1
1
1
1
1
1
1
1
1
28582.
49345.
4160.
12134.
1 7527 .
AO&70.
5258.
4180.
12134.
17527.
60670,
1866.
1866.
7711,
1866.
0,
24875,
12438.
7711.
1866.
0.
24875.
7FT UIDE» 4.25FT HT, :>FT 2
WIDE BOTTOMt CS
3.5HP
200FT. 5HP
193FTt 40HP
POLYPROPYLENE BAG TYPE
60bPM» 70FT HEAD, 5HP
90FT HIGH, 75HP
60.FT HORIZONTAL 7.5HP' 1
30 FPM, IMP j
113
5250.
3371.
57974,
13482.
136171,
1079.
24268.
40447,
37750.
5258.
3371,
80894.
20223.
13482.
12438.
746.-
16167.
2488.
0.
1741.
3731.
8706.
13930.
12438.
746.
1617.
1368.
2488.
-------
FEED BIN
BIN WEIGH .FEEDER
GYRATORY CRUSHERS
BALL MILL DUST COLLECTORS'
BALL MILL
MILLS PRODUCT TANK
MILLS PRODUCT TANK SlURRY
PUMP
SLURRY FEED TANK
SLURRY FEED TANK AGITATOR
SLURRY FEED TANK PUMPS
TOTAL EQUIPMENT COST
13FT DIAr 2JFT STRAIGHT
SIDE HT» COVERED. CS
14FT PULLEY CENTERS, 2HP
75HP
POLYPROPYLENE BAG TYPE
2200 CFKr 7.5HP
16179.
29851 .
135.HP
1O.OTPH.
5500 GAL 10FT PIAf 10FT
HT. FLAKEGLASS LINED CS
MILLS PRODUCT TANK AGITATOR . 10HP
42.GPM. 60FT HEAD*
2.HP» 2 OPERATING
AND 1 SPARES
44334.GAL» 19.6FT DIAi 1
19.AFT HTt FLAKEGLASS-
LINED CS
41.HP 1
21.GPM, 6,0 FT HEAD» 9
l.HP. 4 OPERATING AND
5 SPARE
3
3
3
3
3V
3
3
54603.
1597(45.
15774.
442088.
14561.
24673.
7719.
3731.
15672*
37313.
40784.
22761.
1119.
1493.
10621.
29197.
22342.
22762.
2155.
4478.
1412984. 298907.
114
-------
SCRUBBING
INCLINING A OPERATING AND 2 SPARE SCRUBBING TRAINS
ITEM
DESCRIPTION
NO. MATERIAL LABOR
F.D, FANS
SHELL
RUBBER LINING
MIST ELIMINATOR
SLURRY HEADER AND NOZZLES
GRIDS
SPHERES
TOTAL SCRUBBER COSTS
REHEATERS
SOOTBLOUERS
EFFLUENT HOLD TANK
EFFLUENT HOLD TANK AGITATOR
COOLING SPRAY PUMPS
ABSORBER RECYCLE PUMPS
MAKEUP MATER PUMPS
TOTAL EQUIPMENT COST
14.SIN H20i WITH 1195.
HP MOTOR AND DRIVE
6 2135508. 123075.
231288.GAL» 34.OFT DIAt
34.OFT HTt FLAKEGLASS-
LINED CS
63. HP
1274.GPM 100FT HEADf
59.HPi 4 OPERATING
AND 6 SPARE
8741.GPMr JOOFT HEAD?
406.HP* 8 OPERATING
AND 10 SPARE
6
6
72
6
6
12
974744.
1439962.
442464.
376418.
566157.
210822.
4010566.
1256471.
485362.
109948.
415024.
141024.
334259.
51940.
358206.
206827.
183777.
20822.
18
2549.GPM» 200.FT HEAD. 2
715.HP. 1 OPERATING
AND 1 SPARE
790391.
19790.
62O76.
1826.
9364080. 1343607.
115
-------
WASTE DISPOSAL
ITEM
DESCRIPTION
NO. MATERIAL LABOR
ABSORBER BLEED RECEIVING
TANK
ABSORBER BLEED TANK AGITATOR
POND FEED SLURRY POMPS
POND SUPERNATE PUMPS
TOTAL EQUIPMENT COST
57761.GAL. 17.OFT DIA»
34.OFT HTt FLAKGLASS-
LINED CS
36. HP
642.GPMf 130.FT HEAD
39. HPi. 1 OPERATING
AND 1 SPARE
1
2
364.GPM» 192.FT HEAD, 2
29.HPf 1 OPERATING
AND 1 SPARE
14579.
20467.
14366.
7144.
31229.
1511.
2718.
659,
56557. 3&116.
116
-------
LIMESTONE SLURRY PROCESS — BASIS: 500
PROJECTED CAPITAL INVESTMENT fcE
UNIT. 1978 STARTUP
- INDUSTRY MHIEF1NG'
INVESTMENT, THOUSANDS OF 1977 DOLLARS
DISTRIBUTION
EOUTPMFNT
MATERIAL
LABOR
PIPING
MATERIAL
LAROR
DUCTWORK
MATERIAL
LA^OC)
FOUNDATIONS
MATERIAL
LAPOR
POND CONSTRUCTION
STRUCTURAL
M4TE4MI
LABOR
ELECTRICAL
MATERIAL
LAPOrt
INSTRUMENTATION
MATERIAL
L«POR
BUILDINGS
MATERIAL
LAPOR
SERVICES AND MISCELLANEOUS
SUBTOTAL DIRECT INVESTMENT
ENGINEERING DESIGN AN-:) SUPERVISION
CONSTRUCTION EXPENSES
CONTRACTOR FEE^
CONTINGENCY
SUBTOTAL FIXFO INVESTMENT
ALLOWANCE FOR STARTUP AND MODIFICATIONS
INTEREST DURING CONSTRUCTION
SUBTOTAL CAPITAL INVESTMENT
LAND
WORK-ING CAPITAL
TOTAL CAPITAL INVESTMENT
PAW MATERIAL
HANOI Ir»G AND
Mi-! PAR AT ION
1197.
255.
197.
79.
0.
0.
10*.
442.
0.
??7.
HS.
141.
261.
9*.
21.
28.
44.
103.
3300.
297.
*>?«.
Ib1^.
330.
4620.
370.
554.
5544.
7.
12H.
5680.
! SCRUBBING
7800.
1116.
2431.
70?.
V
19*9.
1347.
94.
281.
0.
1 95.
436.
457.
701.
743.
124.
0.
0.
595.
1H991.
1709.
3039.
950.
1%99.
26588.
?1?7.
31*1.
31906.
3.
738.
32647.
WASTE
DISPOSAL
46.
29.
729.
270.
0.
0.
10.
30.
398*.
1.
4.
83.
191 .
7.
2.
0.
C.
174.
5561.
500.
B90.
278.
556.
7785.
623.
934.
9342.
979.
216.
10538.
TOTAL
9042.
1400.
3357.
1052.
1969.
13*7.
208.
754.
3984.
423.
525.
681 .
1173.
844.
147.
28.
44.
872.
27852.
?507.
4<*5r>.
1 3-»3.
2765.
38993.
3119.
4679.
*6792.
989.
1083.
48864.
PERCENT
OF OHECT
INVESTMENT
32.5
5.0
12.1
3.8
7.1
4.6
0.7
2.7
14.3
1.5
1.9
2.4
4.2
3.0
0.5
C.I
0.2
3.1
100.0
•5.0
16.0
5.0
10.0
140.0
11.2
16.3
168.0
3.<,
3.9
175.*
-------
LIMESTONE SLURRY PROCESS — BASIS-- -*oo -• UNIT. i<»7H
PROJECTED REVENUE REQUIREMENTS - INOUSTrtY KR1EF 1*6-500
DISPLAY SiFET FOR YEAH
ANNUAL OPERATION K«—
i
7000
27.44 TONS PtP
TOTAL FIXEO INVESTMENT
DRY
D1BECI.CDSIS
l*fS.3 K TONS
0.0 *> TONS
B.OO/TON
50.00/TON
SUBTOTAL RAW MATERIAL
AND
UTILITIES
25990.0 HAN-HR
6*1550.0 K L«
21*81*0.0 K GAL
12.50/HAN-HP
2.00/K LB
o.l2/< GAL
3760.0 rl»
17.00/MP
PROCESS hATEP
ELECTRICITY
MAINTENANCE
LA40H ANO MATERIAL
ANALYSES
SUBTOTAL CONVERSION COSTS
SUBTOTAL DIRECT COSTS
1MQ15ECJ-CDSIS
DEPRECIATION
COST OF CAPITAL AND TAXES* 17.20* OF UNDEPRECIATED INVESTMENT
INSURANCE *, INTERIM REPLACEMENTS. 1.17* OF TOTAL CAPITAL INVESTMENT
OVE3HEAO
PLANT, 50.o* OF CONVERSION COSTS LESS UTILITIES
ADMINISTRATIVE. RESEARCH, AND SERVICE.
10.0* OF OPERATING LA^OR AND SUPERVISION
SUBTOTAL INDIRECT COSTS
SUBTOTAL ANNUAL REVENUE REOUIHEMtiMT
SLUDGE FIXATION COSTS 192100.0 TONS
TOTAL ANNUAL REVENUE REUUIftEMENT
EQUIVALENT UNIT REVENUE PE
-------
VO
L1MF5TONE SLURRY P^OCFSb — ?USIS: 500 Mi UNIT. 1<»7* STARTUP
PROJECTED LIFETIME REVENUE RtOUlREMtNTS - INOUSTKY HwlEFlNf»-SOO "id
TOTAL CAPITAL INVESTMENT: *
488o4000
SULFUR
YEAPS ANNUAL POtaFR UNIT POWfP UNIT F)Y
AFTER OPERA- HEAT FUtL POLLUTION
POWER TION, REQUIREMENT, CONSUMPTION* CONTROL
UNIT KW-HR MILLION flTu TONS COAL PROCESS,
STAPT^ /KK /YEAR /Y£A*> TONS/YKA^
1 7000
2 70PO
3 7000
4 7000
— 5.11_jn2fi
6 7000
7 7000
8 7000
9 7000
1.2 7&U2
11 5000
12 5000
13 5000
14 5000
.ll.ll.5Q2a;
14 3500
'.7 3500
1« 3500
19 3500.
?] 1500
23 1500
23 15CO
2* . 1500
?6 1500
27 1500
28 1500
29 1500
ltt-ll_15J2fl".
OT 1?7500
LIFETIME
EVNUE «EOL>
LEVEL IZEO
31500000 1^00000
31500000 1500000
3 1 o o 0 r. i o ' Iriooooo
31500000 1500000
315DQ0.2P. 150010.0.
31500000 1500000
31500000 1500000
31510000 1500000
31500000 1500000
3.l2°°222 1§2°220
22500000 1071400
22500000 1071400
22500000 1071400
2P500000 10714CO
15750000 75001)0
15750000 75fiOO)
15750QOO 7">0000
1^750r,QO 750000
6750000 321400
•S7DOfiOO 3314QO
A750000 J?1'00
*>7SOOOO 3?1400
6/60000 S/Jl^OO
67-.0000 3?MOO
«-.7bflOOO 321400
6750000 321400
£ 7 s o o o o 32i'*l!^
573750r.0fl 27321000
AVERAGE I'.CREiSE IN UNIT *FVENJF
DOLLARS PER TON OF COAL
MILLS PFR KILOWATT-HOUR
CFNTS PFP MILLION ^Tll MF
DOLLARS PER TON OF bULFn
I&rMfNT DJSCOUNTtO IT 11. ft* TO
I'. CREASE IN UMIT REVENCF Rtt-^l»
HOLLARS PER TON OF COtL i
MILLS PC& K ILO*-TT-«OU-<
3r>000
3SOOO
35000
3SOOO
35QOO
35000
35UOO
35000
35000
350.00
2SOOO
25000
?5000
I7»no
17500
17500
17^00
75UO
7500
7300
7600
7500
7500
7500
7500
BYPROOUCT
ADJUSTED GROSS
ANNUAL REVENUE
•*AT^ • SLUUGE REOUIWEMENT TOTAL NET ANNUAL CUMULATIVE
EQUIVALENT FIXATION FEE EXCLUDING ANNUAL INCREASE NET INC«ElSf
TONS/YEAR WTON SLUDRE SLUDGE IN TOTAL IN TOT*1
FIXATION FIXATION REVENUE REVENUE
DRY Dhy COST, COST, REQUIREMENT, REQUIREMENT,
SLUIXit SLUDGE S/YE4R $/YEA« $ §
iv!i!S
1^2100
103 100
1^2100
192100
192100
I1' ''I 00
137200
1.17200
137200
13720ft
96000
96000
96000
9^22
41200
4 1 ? 1 0
41200
41*00
41200
41200
41200
12.00
12.00,
12.00"'
12.00
'~12.00
12.00
12.00
12.00
15 QQ
12.00
12.00
12.00
12.00
12.00
12.00
12.00
12.00
~12*00
12.00
12.00
12. UU
~12*00
12.00
12.00
12.00
18078900
17bl(>600
175^2400
17274100
"~16737500~
1646V300
16201000
15932700
i "jhf4^(30
"136^2300"
13354000
13085flOO
12017500
10b77300
1060SOOO
10340700
10U72500
~~747l700~
7203500
6935200
b t* (5^>^0 0
h J *^^ *S 00
6130400~
5862100
5593SOO
5325600
2305200
2305200
2305200
2305200
2305200
2305200
23C5200
2305200
2305200
Ib4b400
1646400
1646400
1 6** 'i^t 0 0
J 646400
1152000
115200C
11520CO
1152000
494400
494400
494400
494400
494400'
494400
494400
494400
20384100
201 15SCO
19847600
19042700
18774500
18506200
18237900
1526H/00
15000400
14732200
K463900
12029300
1 17olOOO
11492700
11224500
79661i>0
7697900
742960U
7161300
6624dOO
6356500
6046200
5820000
20384100
60347500
' ' -\ "- -»
^ 22 jZ^QQ.
1 1 H dh 050 0
137055000
ISSSfrlc-OO
173799100
"20/037500~
222037900
23ft770! 00
2512J4COO
2774bd900
2"92 ' 9930
3007! ?f> j.l
311937100
33G039-.30
3365S7300
^'•!*:'"::.'
36666^000
373 0?2500
j7y 1 1 0700
lifiQ £1222 12*0.0 T-StSJjflO. 49.J4JJP. 55512(12 __3.iU4d2*0.2_
637^)00
t>tOuI«FM£
dUWNtO
IT INPUT
R wtMilVErt
34V9000
NT
INITIAL YEA«, HOLLARS
t^tNT ECUI
rJU^'lti>
v«Lf.NT TO OISCOUNTEO
REQUIREMENT
348494400
12.76
5.47
bn.74
546. TC
125705100
OVf< LIFE
11.80
5.06
CENTS PER MILLION HIU HFAT INPUT 56. 21
DOLLARS PFR TON OF iULFi"
< MfMOVFD
505.86
419A8000
1.53
0.66
7.32
65.86
16366000
OF PO»ER
1.54
0.66
7.32
65. 8*
390482400
14.29
6.13
68 • Of)
V fc» . V -f
612.52
142071100
UNIT
13.34
5.72
63.53
571.71
-------
INCLUDING 1 OPERATING AND 1 SPARE PREPARATION UNITS
ITEM
DESCRIPTION
NO. MATERIAL LABOR
CONVEYOR FROM CALCINATION
PLANT
1500FT HORIZONTAL* 30HP
153293.
46144,
CD ~O
NJ
C
H- STORAGE SILO ELEVATOR
CONCRETE STORAGE SILO
STORAGE SILO HOPPER BOTTOM
RECLAIM VIBRATING FEEDER
RECLAIM BELT CONVEYOR
FEED BIN ELEVATOR
FEED BIN
29,FT HIGH» 50 HP
659.FT3r 8.2FT DIA »
12.4FT STRAIGHT SIDE
STORAGE HT
60 DEGREEr CS
3.5HP .
83.FT HORIZONTAL, 5HP
SOFT HIGH» 50HP
10FT DIA» 15FT STRAIGHT
SIDE HT» COVERED, CS
BIN.VIBRATING FEEDER
BIN WEIGH FEEDER
SLAKER
SLAKER PRODUCT TANK
SLAKER PRODUCT TANK AGITATOR 10HP
3, 5HP
12FTr 12IN SCREWr IMP
O.TPHt O.HP
1
1
1
1
1
1
2
2
2
2
2
2
32047.
3161.
401,
12134.
15232.
52311.
5393.
9168,
11864.
24960.
12134.
14291 .
650,
7624.
1194.
1866.
3953,
995.
9950,
3234.
1244,
2303.
18905.
746.
i > i **->
5 3
r- o
CO
— 1 3>
yo
-------
LIME SYSTEM DUST COLLECTORS
SLAKER PRODUCT TANK SLURRY
PUMPS
SLURRY FEED TANK
SLURRY FEED TANK AGITATOR
SLURRY FFiED TANK PUMPS
~ TOTAL EQUIPMENT COST
POLYPROPYLENE BAG TYPE
2200 CFhi7.5HP
l.GPMi 60FT HEAD»
O.HPi 1 OPERATING
AND 1 SPARES
680,GALF 4.9FT DIA»
4.9FT HTf FLAKEBLASS-
LI NED CS
1 .HP
O.GPMt 60 FT HEADi
0,HP» 4 OPERATING AND
5 SPARE
1
9
21032,
4448,
656,
2867.
19444.
SCRUBBING
INCLUDING 4 OPERATING AND 2 SPARE SCRUBBING TRAINS
49751,
995,
1406.
212,
4476,
394838. 155649,
CO
o o
o :z
z rn
c: a>
ITEM
DESCRIPTION
NO, MATERIAL LABOR
-------
WASTE DISPOSAL
ITEM
DESCRIPTION
NO, MATERIAL- LABOR
to
to
ABSORBER BLEED RECEIVING
TANK
ABSORBER BLEED TANK AGITATOR
THICKENER FEED PUMP
THICKENER
THICKENER OVERFLOW PUMPS
THICKENER OVERFLOW -TANK
SLUDGE FIXATION FEED PUMP
52510. GAL f 16.5FT DIAi
32.9FT HTi FLAK GLASS-
LINED CS
34, HP
665.GPMr 60FT HEAD.
lO.HPf 1 OPERATING
AND 1 SPARE
18248.SQ.FT.rl52.FT DIAi 1
B.6FT HT
438.GPMF 75, OFT HEAD* 2
14. HP* 1 OPERATING
AND 1 SPARE
13682. 29306,
1
2
1
18967.
13078.
536285 .
1400.
2752.
366827,
m
•— •
3
O
-a
—i
o
CO
CO
rn
o
CO
7224, GAL i
B.6FT HT
12. OFT DJAr 1
207.GPMr 5 OFT HEADr
6, HP, 1 OPERATING
AND 1 SPARE
5913.
1637,
10575.
3V27,
1816,
TOTAL EQUIPMENT COST
600137. 406574.
m
CO
CO
-------
WASTE. DISPOSAL
ITEM
DESCRIPTION
NO* MATERIAL LABOR
to
ABSORBER BLEED RECEIVING
TANK
ABSORBER BLEED TANK AGITATOR
THICKENER FEEH PUMP
THICKENER
THICKENER OVERFLOW PUMPS
TH 1CKENER 0VERFl 0U TANK
FILTER FEED SLURRY PUMP
FILTER
FILTRATE PUMP (PER FILTER)
FILTRATE SURGE TANK
FILTRATE- SURGE TANK PUMP
52510.GALr 16.5FT DIAr 1
32.9FT HT, FLAKGLASS-
LINEP CS
34.HP 1
652.GPMF 60FT HEAD, 2
IB.HP, 1 OPERATING
AND i SPARE:
17913.SQ.FT.,151.FT MIA, 1
8.5FT HT
430.6PMr 75.OFT HEADr 2
14.HP, 1 OPERATING
AND J SPARE
7092.GAL, 11.9FT D1A, 1
8.5FT HT
l()2.GPMr SOFT HEADi 3
3.HP, 2 OPERATING
AND 1 SPARE
269.SQ FT FILTRATION 2.
AREA
45.GPM, 20.OFT HEAD, 4
O.HP» 2 OPERATING
AND 2 SPARE
1482. GAL,
6.3FT HI
6.3FT DIA, 1
13682.
18967.
13053.
90.GPM, 85.OFT HEAP)' 2
3.HP,' 1 OPERATING
AND 1 SPARE
5877.
1618.
10750.
180058.
6211.
594.
3874,
29306.
1400,
2734.
532294. 362329.
542.
3880 .
2114,
16611 .
573.
1424.
357,
CO
(ZD
^^ C^ C3
~o -o •—•
13 —I CO
m *—i ~o
s: CD o
—I 2: co
r^ -t i—
CO .T>
-------
FIVE PROCESS STUDY OPERATING PROFILE
60
20
0
' '
0
l
10
T
20 30 40
BOILER ME - YEARS
50
1 ' I
60
I I
70
-------
FPC OPERATING PROFILE
ALL BOILERS AVERAGE CAPACITY FACTOR vs. BOILER AGE-
BASED ON 1969-1973 FPC DATA
BOILER AGE-YEARS
125
-------
NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY STUDY
NJ
160
120
100
GO
80
60
300-MW UNIT
LIMESTONE SYSTEM
I
COAL SULFUR
CONTENT, LB
S02/MBTU
,8 1,2 1,6
SULFUR REGULATION, LB S02/MBTU
2,0
-------
NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY STUDY
ro
110
1
300-MW UNIT
LIME SYSTEM
cr>
120
LU
100
80
60
COAL SULFUR
CONTENT, LB
S02/MBTU
,8 1,2 1,6
SULFUR REGULATION, LB S02/MBTU
2,0
-------
oo
NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY STUDY
1
)-W
600-MW UNIT
LIMESTONE SYSTEM
130
CD
110
90
C/5
70
50
COAL SULFUR
CONTENT, LB
S02/MBTU
,8 1,2 1,6
SULFUR REGULATION, LB S02/MBTU
2,0
-------
NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY STUDY
130
1
600-MW UNIT
LIME SYSTEM
110
01
90
CO
70
50
COAL SULFUR CONTENT
LB S02/MBTU
8
,8 1,2 1.6 2,0
SULFUR REGULATION, LB S02/MBTU
-------
COSTS OF REDUNDANCY
100
90
70
UJOOfW
8 OPERATING TRAINS
SYSTEM
SULFUR
60
234
REDUflATfT TRAINS
-------
o
00
T
I-]
M
S
w
M
o*
w
z
[I]
ONSITE POND OPTION
3.5% S COAL
1.2 LB S09/MBTU
INCLUDING^FLYASH
DISPOSAL
OPERATING PROFILE
H 4
1
1
I
200
400 600 800
POWER UNIT SIZE, MW
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND OPERATING PROFILE ON UNIT REVENUE REQUIREMENT
-------
T
T
CO
N5
O
CO
^
2
H
Z
w
w
z
w
ONSITE POND OPTION
3.57. S COAL
1.2 LB S02/MBTU
LIMESTONE
I
I
200
400 600 800
POWER UNIT SIZE, MW
1,000
EFFECT OF POWER UNIT SIZE AND PROCESS
ON UNIT REVENUE REQUIREMENT
-------
CO
u>
o
oo
ON
•-* 7
». j *
us
CO
I 6
H
z
w
3
2
H
M
2
ONSITE POND OPTION
3.5% S COAL
1.2 LB S02/MBTU
INCLUDING FLYASH
DISPOSAL
EXCLUDING FLYASH
DISPOSAL
200
400 600 800
POWER UNIT SIZE, MW
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND FLYASH ON UNIT REVENUE REQUIREMENT
-------
CO
120
z
1100
en
>
M
H
M
Z
80
200
400
600
T
T
ONSITE POND OPTION
3.57. S COAL
INCLUDING FLYASH
DISPOSAL
800
POWER UNIT SIZE, MW
VAL
I
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND % SO- REMOVAL ON UNIT INVESTMENT REQUIREMENT
-------
O
oo
r i r
ONSITE POND OPTION
3.5% S COAL
INCLUDING FLYASH DISPOSAL
CO
Ln
7. S02 REMOVAL
I
I
I
200
400 600 800
POWER UNIT SIZE, MW
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND 7. S02 REMOVAL ON UNIT REVENUE REQUIREMENT
-------
140
•co-
H
Z
120
100
80
H
M
2
I
I
200
T
ONSITE POND OPTION
1.2 LB S02/MBTU
INCLUDING FLYASH
DISPOSAL
I
I
400 600 800
POWER UNIT SIZE, MW
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND S IN COAL ON UNIT INVESTMENT REQUIREMENT
-------
o
oo
Cfl
h-J
1-1
W
w
G
2
W
H
M
3
T
T
ONSITE POND OPTION
1.2 LB S07/MBTU
INCLUDING FLYASH
DISPOSAL
2.0__
I
1
I
I
200
400 600 800
POWER UNIT SIZE, MW
1,000
LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
AND S IN COAL ON UNIT REVENUE REQUIREMENT
-------
COAL CLEANING VERSUS 100% FGD
CASE 1 ROM COAL •*• 2000 MW •*• FGD
CASE 2 ROM COAL •» PREP PLT -> 2000 MW + FGD
CASE 3 ROM COAL •* CHEM COMM & PREP PLT + 2000 MW •> FGD
138
-------
CASE 1 CASE 2 CASE 3
CAPACITY FACTOR 45 56 56
GENERATION (KWH/YR) 7.88 x 109 9.81 x 109 9.81 x 109
PRODUCTION COSTS
(MILLS/KWH) 15.2 12.7 12.9
FGD REVENUE REQUIREMENT
(MILLS/KWH) 7.1 6.1 5.8
TOTAL GENERATION COSTS
(MILLS/KWH) 22.3 18.8 18.7
-------
LANDFILL AND PONDING CONCEPTS
FOR FGD SLUDGE DISPOSAL
by
Jerome Rossoff, Paul P. Leo, and Richard B. Fling
The Aerospace Corporation
El Segundo, California
Presented at the
U. S. Environmental Protection Agency
Industrial Environmental Research Laboratory
Industry Briefing Conference on
Technology for Lime/Limestone Wet Scrubbing
Research Triangle Park, North Carolina
August 29, 1978
140
-------
ABSTRACT
This paper is concerned with the environmentally
sound disposal of flue gas desulfurization (FGD)
sludges. The environmental considerations and
the technology and costs associated with the dis-
posal of FGD sludges by landfilling and ponding
are summarized. Concepts discussed are lined
ponds, unlined ponds equipped with underdrainage,
chemical treatment and landfilling, and conversion
to gypsum. The need for environmental control is
reviewed. The capabilities of each concept to pre-
vent water pollution and the environmental consid-
erations that require site maintenance are discussed.
Bearing strengths associated with landfill concepts
are included, and the status of developments of non-
operational concepts, i. e. , ponding with under-
drainage and the disposal of FGD gypsum, are
discussed, Additionally, disposal site volume
requirements and estimated disposal costs are
given.
141
-------
Introduction
This paper summarizes current concepts of landfilling and ponding for the
environmentally sound disposal of flue gas desulfurization (FGD) sludges. The
techniques discussed herein represent the results of studies and assessments
performed by The Aerospace Corporation under contract to the Industrial
Environmental Research Laboratory of the U.S. Environmental Protection
Agency (EPA), Research Triangle Park, North Carolina. These techniques
do not constitute endorsement or approval by the EPA, but are presented as
the authors' assessments of the best available methods for the disposal of
FGD sludges by landfilling or ponding.
With the passage of the Resource Conservation and Recovery Act (RCRA),
public law 94-580, October 1976, guidelines and criteria are forthcoming for
application to FGD sludges. Determinations will be made by the EPA as to
whether these sludges are to be considered hazardous, and, depending on
those determinations, criteria will be developed for FGD sludge disposal.
Without federal criteria applicable specifically to FGD sludges, almost all
studies and developments up to this time have used drinking water criteria
as the basis for establishing requirements for disposal. Because the trace
element and salt content of most samples analyzed exceeded the drinking
water criteria at least for some of the constituents, the general approach
taken has been to dispose of FGD sludges such that no direct discharge to any
water supply would be permitted, that any seepage would be minimized or
perhaps totally eliminated, and that runoff would be controlled. Additionally,
a strong effort has been made by industrial and government agency develop-
ment and evaluation programs to determine disposal techniques that would
not only be environmentally sound from the standpoint of water quality control
and, when practical, would also result in reclamation of the land area selected
for the disposal site. As a result, all disposal techniques that have been de-
veloped for FGD sludge are intended for the control of water quality, but not
all of them produce reclaimable disposal sites. The techniques discussed
herein consist of the following: (a) ponding of untreated sludges, (b) disposal
of untreated sludges in ponds equipped with underdrainage systems, (c) chem-
ical treatment and landfilling, and (d) conversion to gypsum and subsequent
disposal.
The basic characteristics of each of these approaches are discussed as to
protection of water supplies, land reclamation, and disposal costs.
Water Quality Criteria
A comparison of chemical constituents from a large number of analyses of
sludge liquors in a discharge stream with the National Interim Primary Drink-
ing Water Regulations (40 CFR 141) is given in Table I as a ratio of constituent
concentration to water criteria. These ratios are given for the range of con-
stituents from a composite of data for ten eastern and western sludges, with
and without fly ash, and for the ten independent samples. It should be noted
that the values used in this comparative analysis represent the initial concen-
trations that would seep from the base of an untreated sludge pile.
142
-------
Table I. Comparison of Sludge Liquors with Water Criteria
NIPDWR
Drinking
Criteria,
mg/t
As 0. 05
Cd 0.01
Cr 0.05
Pb 0.05
Hg 0.002
Se 0. 01
F 2
TDS 500
pH (actual
values)*3
Concentration -f- Criteria (Nondimensional)
All
Samples
< 0. 8 - 2. 8
0.4 - 11
0.2Z - 5
0.2 - 6.6
0.03 - 2.5
0.28 - 20
< 0. 5 - 5
6.6 -48.5
6. 7 - 12.2
Sample
A
0.6
5.0
5.0
0.8
2.5
10.0
_ _
36
6.7
B
0.4
1.2
0.8
3.0
_ _
3. 3
0. 5
6.6
6.8
C
2.0
0.4
1.8
4.6
10.0
3. 3
30.0
8.0
D
0.04
_ _
<0. 2
<0. 1
4. 2
- -
13.4
12.2
E
0.4
11
0.6
6.6
< 0. 5
< 2
1.7
18.8
8.7
F
1. 2
1. 3
0. 2
0. 2
< 0.001
7. 8
1
20. 5
8.0
G
2.8
--
--
< 0.2
< 0. 1
20
--
28
7.8
H
0. 1
--
--
< 0. 2
< 0. 1
14
--
18.4
7. 3
I
0.8
5
--
0.8
0. 1
2.8
5
8.4
10.7
J
0.2
2. 5
1. 1
< 0. 1
0.03
0. 3
< 0. 5
48. 5
8.9
*-
OJ
Sample data are as follows:
Sample Station
A
B
C
D
E
F
G
H
I
J
Mohave
Cholla
Shawnee
Shawnee
Shawnee
Shawnee
Shawnee
Shawnee
Duquesne Phillips
LG&E Paddy's Run
EPA-proposed secondary regulation
GForced-oxidized to gypsum.
Absorbent
Limestone
Limestone
Limestone
Limestone
Lime
Lime
Lime c
Limestone
Lime
Carbide lime
is 6. 5 to 8. 5.
Ash
Sampling Date
3
59
40
6
40
6
6
6
60
12
Mar
Nov
Jun
Jan
Jun
Sep
Oct
Aug
Jun
Jul
1973
1974
1974
1977
1974
1976
1976
1977
1974
1976
-------
In Table I, the ratios for the range of constituent concentrations of the com-
posite data show that all elements analyzed, as well as the total dissolved
solids (TDS) and pH, exceed drinking water criteria. However, in observing
the ratios for the ten independent samples shown in the table, it can be seen
that, except for selenium in two samples and cadmium in one, no trace ele-
ment exceeds the criteria by a factor greater than 10. (Water criteria for
barium, nitrate, and silver are 1, 10, and 0. 05 mg/.e, respectively. Limited
field evaluation leachate data show maximum concentrations for these elements
to be about 5, 1, and 0.5 times the criteria, respectively.) The TDS are high
for most of the samples, and the pH is excessive for two of the samples.
Although trace elements are not eliminated as a matter of concern for some
sites by these data, there are indications that in many cases the concentra-
tions are quite low and that, generally, the concern may be for the concen-
tration of dissolved solids and, in some cases, pH. Chemical oxygen demand
(COD) was considered somewhat differently. Values of COD in fresh sludge
ranged between 40 and 140 mg/£, but, because of the rapid oxidation charac-
teristic of sulfite sludge, the COD after one pore volume displacement by
leaching was 10 mg/f or less and rapidly decreasing. Therefore, because
COD is significant only for fresh sludge and because the sludge is not dis-
charged directly to streams, it was concluded that COD is not a critical
parameter.
Because of depletion of the material with leaching time, cation exchange and
adsorption in the soil, and dilution between the disposal site and the consumer
tap, it is difficult at this time to specifically quantify the degree of pollution
potential at a given site. Therefore, because of the comparatively large con-
centration of dissolved solids and the identification of random values of high
concentrations of trace elements, methods for disposal of these materials to
prevent their access to public water supplies were assessed.
Results
Operational Modes
By the end of 1977, SO2 scrubbers were operating at 22 power stations having
a scrubbing capacity of approximately 10,375 MWe; 20 have nonregenerable
scrubbers, and 2 have regenerable scrubbers. In tne nonregenerable category
7 stations (scrubbing 4460 MWe) use chemical treatment disposal processes,
and 13 stations (scrubbing 5680 MWe) dispose of the sludge untreated. The two
regenerable systems have a total capacity of 235 MWe. A breakdown of the
disposal modes is as follows:
Treated, Untreated, Gypsum, Stabilized,21 Untreated,
Treated, Lined Unlined Lined Unlined Solar
Unlined Pond Pond Pond Pond Evaporation
No. of Plants 61714 i
Total MWe 4293 167 3250 1420 635 375
Stabilized, e.g., dewatered with fly ash addition; not necessarily the final
disposal mode.
144
-------
Disposal Alternatives
The general categories of disposal and the considerations required for environ-
mental control are shown in Table II. In each case, seepage of rainwater
through the sludge and eventual contamination of groundwater pose an environ-
mental concern for all disposal methods. Runoff is a potential source of en-
vironmental pollution for landfill sites because these sites are open and do not
necessarily return water to the scrubber. Only in the case of ponding is it
clear that the disposal site is not directly amenable to land reclamation efforts,
although even in some of these cases it may be possible upon retirement to air-
dry» cap, and vegetate the site. Consideration of each of these effects are
n in the following discussions.
Ponding. In general, the simplest and the least cost (though not neces-
sarily the most environmentally sound) approach to FGD sludge disposal is
ponding. This method requires that, if the pond does not contain a base mate-
rial considered to be impermeable, a liner must be added to prevent seepage.
Operationally, sludge ponds exist today which contain either naturally imper-
meable soils or clay liners transported and placed in the base and on the slopes
of the pond. Because of the highly thixotropic nature of these sludges, ponds
Table II. Environmental Effects of Disposal Alternatives
Type of
Disposal
Pond
Basin
Landfill
Condition
of Waste
Untreated3-
or
chemically
treatedb
Untreated3-
or
conditioned0
Conditioned0
or
chemically
treated0
Primary
Drainage
Supernate
Supernate
Underdrainage
Runoff
Environmental Effect
Seepage
Yes
Yes
Yes
Yes
Runoff
No
No
No
Yes
Land Reuse
No
Yes
Yes
Yes
aUntreated waste refers to FGD sludges as emitted from primary or secondary
dewatering equipment.
Chemically treated sludges refer to the waste treated by one of several com-
mercial processes that make these wastes suitable for landfill disposal.
cConditioned waste refers to sludge treated by techniques other than chemical
treatment and includes oxidation to gypsum and dewatering by mixing with dry
fly ash or other agents that allow the material to be handled in a manner
similar to that for soils.
145
-------
are nonstructural sites and generally are not considered amenable to recla-
mation, except possibly in areas of low rainfall and high evaporation. Also,
if ponded sludges are not dewatered, larger land areas are needed to contain
the m ate rial.
Ponding with Underdrainage. This approach to ponding is still under
evaluation and is not being used operationally at this time. Underdraining and
collection of all seepage for return to the scrubber system maintains control
of leachate at all times and has been shown at small scale evaluation sites to
produce a material structurally capable of supporting personnel and construc-
tion equipment. Present evaluations1 are being made to determine the feasi-
bility of such an alternative regarding (a) site reclamation and (b) relaxation
of requirements on the degree of water-tightness of the base material, inas-
much as no appreciable hydraulic head exists. A site of this type collects
rainfall via the seepage system and returns it to the scrubber. As a result it
is necessary to limit the size of each disposal basin to maintain an acceptable
•water balance in the scrubber loop. This would be accomplished by dividing
the site into sections of approximately 35 to 50 acres at a depth of about
30 feet. Figure 1 shows an underdrained, untreated pond supporting a general
purpose farm tractor within one day after a 3-inch rainfall.
Chemical Treatment. The stabilization of FGD sludges by chemical
treatment offers the most positive solution to the disposal problem. It con-
verts the sludge to a structural material; decreases its coefficient of perme-
ability to a range of approximately 10-5 to 10~7 cm/sec, which as a minimum
is one order of magnitude better than untreated sludges; reduces the concen-
tration of salt constituents in the leachate by approximately 50%; is amenable
to subgrade or above-grade landfilling; and allows the disposal site to be
reclaimable. Chemically treated sludges have not been shown to appreciably
reduce concentration of trace elements in leachate, and, even though the con-
centration of major species is reduced, leaching of chemically treated sites
should be avoided unless it can be assured that the leachate can be diluted by
local groundwater and streams. A general procedure for managing rainfall
runoff from a chemically treated site is to collect the runoff in a peripheral
ditch which directs the water to a settling pond. Depending on the quality of
the water in this pond, it can be decanted to a stream or returned to the
scrubber system.
Chemically treated sludges have solids content of approximately 45 to 65 wt%
(or possibly higher depending on the dewatering potential and the treatment
process used) and attain load bearing strengths in the range of 75 to 300 psi
(5.4 to 21.6 tons /ft2\
Test ponds containing chemically treated sludges are pictured in Figures 2
and 3.
Gypsum. The forced oxidation of sulfite sludges to gypsum or the pro-
duction of high sulfate sludge from the use of western coal results in a waste
material which is readily dewatered by vacuum filtration or by centrifuging
to a solids content in the approximate range of 75 to 85 wt%. Leachate from
gypsum is similar to that of sulfite sludges and therefore should be prevented
from entering water supplies. Because gypsum tends to form a protective
146
-------
Figure 1. Untreated, unstabilized
sludge ponded with underdrainage,
one day after 3-inch rainfall.
Figure 2. Chemically treated sludge
(IU Conversion Systems process).
Figure 3. Chemically treated sludge
(Dravo process).
147
-------
surface scale capable of shedding rainwater, tests are currently being con-
ducted to determine the applicability of the disposal of gypsum on the ground
without the added benefit of liners or impoundment dikes. Limited results
have shown that gypsum sludges crack badly under freeze-thaw conditions,
thereby allowing rainwater to enter into the material. Additionally, gypsum
sludge slumps in its freshly deposited condition when exposed to rainfall
and produces a runoff containing potentially high concentrations of dissolved
solids from the sludge, as well as a condition which requires machinery to
replace the material on the disposal site. These preliminary results indicate
that considerable site maintenance may be required on an operational scale
to reconfigure the disposal pile after weathering (freeze-thaw and erosion)
and to control the runoff. Tests are continuing for the determination of what
control (if any) should be exercised at the site during and after disposal, A
gypsum test pile before and after weathering is pictured in Figures 4 and 5.
Sludge Volume Prediction
Landfill volume requirements are strongly affected by the solids content of
sludges. A comparative analysis of sludge production in acre feet annually is
shown for a 500-MW plant in Figure 6. (This figure neglects the approximate
25% increase in acreage requirements to account for berm slopes and access
roads.) If it is assumed that an untreated sludge settles to approximately
50% solids, the acre feet produced in one year for this case would be 250. The
advantage for gypsum in this regard (neglecting other environmental factors)
would be that approximately 155 acre ft would be produced, providing that the
sludge is dewatered to a solids content of 80%. In the case of chemical treat-
ment, if it is assumed that the material is disposed of at a solids content in
the range of 60 to 70%, the volume to be disposed of would be in the range of
165 to 190 acre ft.
Disposal Cost Estimates
Cost estimates for ponding and chemical treatment for landfilling have been
made and reported by The Aerospace Corporation on several occasions. Dur-
ing studies associated with the EPA Shawnee field disposal evaluation project,
Aerospace cost estimates were made of chemical treatment disposal and were
reported in the initial report on that study. ^ The Aerospace estimates for
lined-pond costs were presented in the initial and second progress reports on
sludge disposal^'^ and at EPA flue gas desulfurization symposiums. 5, 6, 7 ^11
estimates have been updated in a report to EPA** on new source performance
standards on a July 1977 basis; these cost estimates are summarized in
Table III.
Conclusions
Constituent concentrations of FGD sludges require disposal controls to prevent
direct discharge, seepage, or runoff to water supplies. The methods used
operationally today are (a) disposal of untreated sludges in ponds with highly
impermeable liners or (b) chemical treatment prior to sub-grade or above-
grade landfilling. Other methods being evaluated are (a) disposal of untreated
sludges in ponds equipped with underdrainage and (b) conversion to gypsum
for disposal.
148
-------
Figure 4. Gypsum filter cake immediately
after placement, September 1977,
Paducah, Kentucky.
Figure 5. Gypsum filter cake after first
winter season, March 1978,
Paducah, Kentucky.
149
-------
300 r
280
260
£ 240
a
O
cc
* 220
3
^ 200
180-
160-
140
90%S02 REMOVAL
L!ME UTILIZATION 90%
LIMESTONE UTILIZATION 80%
GYPSUM (45% ash)
40 50 60 70 80
SOLIDS CONTENT, wt %
90
100
Figure 6. Sludge produced annually
(500-MW plant, 3.5% sulfur coal,
12,000 Btu/lb, 14% ash).
Untreated sludge ponds have the disadvantage of not being reclaimable. Those
equipped with underdrainage maybe reclaimable, depending on evaluations
now in progress.
Chemical treatment improves the impermeability of sludges by one order of
magnitude or more, reduces the dissolved solids concentration by about 50%,
and attains a bearing strength greater than 5 tons/ft2. Chemically treated
sites must be maintained to control seepage or runoff, depending on the process
used. Above-grade sites generally require maintenance for runoff control only.
Gypsum sludges dewater readily to 75 to 85 wt% solids. These materials when
stacked have exhibited severe surface cracks after freeze-thaw cycling. There-
fore, piling or stacking gypsum without considerable site maintenance may not
be a feasible disposal method, on the basis of preliminary field tests. Further
testing is under way.
Volume production for a 500-MW eastern plant, on the average, is approxi-
mately 250, 175, and 155 acre ft annually for untreated, chemically treated,
and gypsum sludges, respectively. For a 1000-MW plant, these values would
be increased by about 93%. Landfill requirements for these volumes are in-
creased by approximately 25% to account for berm slopes and access roads.
Disposal cost estimates in mills per kilowatt hour (July 1977 dollars) for pond-
ing on indigenous clay, ponding with liner added, and chemical treatment are
0.55, 0.80, and 1.05, respectively, fora 1000-MW plant burning typical
eastern coal.
150
-------
Table III. Disposal Cost Comparison'
Cost Basis ,
Mid- 1977 $
Mills/kWh
$ /ton of
sludge (dry)
S / ton of
coal
Ponding
Indigenous
Clay
0. 55
4. 90
1. 50
Liner
Added
0. 80
7.25
2. 20
Landfill,
Chemical
Treatment
1.05
9.70
2.95
Gypsum
1. 10
10. 30b
3. 10
Notes:
Dollar base:
Plant characteristics;
Coal burned:
Annual average
operating hours:
Plant and disposal site
lifetime:
removal, with
limestone absorbent:
Limestone utilization:
Sludge generated:
Average annual capital
charges, 30-yr average:
Cost of land used for
disposal:
Land depreciation:
Disposal site:
July 1977
1000 MW, 8700 Btu/kWh
(0. 73-lb coal/kWh)
3.5% sulfur, 12,000 Btu/lb, 14% ash
4380 hr/yr (30-yr average)
30 yr
90%
80% of all cases except for gypsum,
which is 100%
4. 8 X 10 short tons/yr untreated
waste (dry) including ash
18% of total capital investment
$5000/acre; all land assumed pur-
chased initially; sludge depth, 30 ft
Total depreciation in 30-yr; straight-
line basis
Within one mile of the plant
Cost of forced oxidation and disposal of gysum sludge converted to
cost/ton of equivalent quantity of nonoxidized sludge. Divided by
1.08 to convert to gypsum cost. Includes fly ash; disposal is in an
indigenous clay pond.
151
-------
REFERENCES
1. R. B. Fling etal.. Disposal of Flue Gas Cleaning Wastes: EPA
Shawnee Field Evaluation: Second Annual Report7 EPA-600/7-78/024,
U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, Feb 1978.
2. R. B. Fling _et_al_. , Disposal of Flue Gas Cleaning Wastes; EPA
Shawnee Field Evaluation: Initial Report, EPA-600 /2-76-070, U.S.
Environmental Protection Agency, Research Triangle Park, North
Carolina, March 1976.
3. J. Rossoff and R. C. Rossi, Disposal of By-Products from Non-
regenerable Flue Gas Desulfurization Systems; Initial Report,
EPA-650/2-74-037a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, May 1974.
4. J. Rossoff et al. , Disposal of By-Products from Nonregenerable Flue
Gas DesulfurTzation Systems: Second Progress Report, EPA-6QO-7-
77-052, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, May 1977.
5. J. Rossoff and R. C. Rossi, "Flue Gas Cleaning Waste Disposal, EPA
Shawnee Field Evaluation, "presented at the EPA Flue Gas Desulfuriza-
tion Symposium, New Orleans, Louisiana, March 1976.
6. J. Rossoff etal. , "Disposal of By-Products from Non-Regenerable
Flue Gas Desulfurization Systems: A Status Report, " presented at
the EPA Flue Gas Desulfurization Symposium, Atlanta, Georgia,
Nov 4-7, 1974.
7. P. P. Leo, R. B. Fling, and J. Rossoff, "Flue Gas Desulfurization
Waste Disposal Field Study at the Shawnee Power Station, " presented
at the EPA Symposium on Flue Gas Desulfurization, Hollywood, Florida,
Nov 8-11, 1977.
8, P. P. Leo and J. Rossoff, Controlling SO? Emissions from Coal-Fired
Stream Electric Generators; Solid Waste Impact, EPA-600/7-78-044b,
Vol. II, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, March 1978.
152
-------
COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
J. W. Barrier
Emission Control Development Projects
Office of Agricultural and Chemical Development
Tennessee Valley Authority
Muscle Shoals, Alabama
Prepared for Presentation at
Industry Briefing Conference
Results of EPA Lime/Limestone Wet Scrubbing Test Programs
Sponsored by the U.S. Environmental Protection Agency
Royal Villa Motel in Raleigh, North Carolina
August 29, 1978
153
-------
COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
J. W. Barrier
Emission Control Development Projects
Office of Agricultural and Chemical Development
Tennessee Valley Authority
Muscle Shoals, Alabama
ABSTRACT
Several series of studies to evaluate the economics of various systems
associated with the control of fly ash and sulfur dioxide emissions from
power plant flue gases are being conducted by the Tennessee Valley
Authority (TVA) for the U.S. Environmental Protection Agency (EPA). One
group of studies involves the preparation of economics for the comparison
of flue gas desulfurization (FGD) sludge disposal alternatives. Two
studies are complete—one report is published and one report is being
reviewed by EPA before publication—and a third study is underway. The
results of the two completed studies are described in this report.
Six disposal alternatives have been evaluated to date. A base case for
each process was established and complete conceptual designs of the
systems were prepared for use as a cost estimating basis. Cost estimates
and conceptual designs are based on common premises used for all TVA-EPA
studies.
The six alternatives evaluated are (1) untreated ponding, (2) Dravo
Corporation's process, (3) Chemfix process, (4) IU Conversion Systems'
process, (5) untreated sludge - fly ash blending, and (6) gypsum landfill.
For each alternative total capital investments and annual revenue require-
ments were estimated for the base case and major case variations.
154
-------
COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is sponsoring an exten-
sive research and development program to evaluate, develop, and demonstrate
sludge disposal alternatives that are environmentally and economically
acceptable to the utility industry for flue gas desulfurization (FGD)
sludge (1). A major program area that involves the field testing of
potential processes for commercial-scale use is The Aerospace Corporation's
study being conducted at the Shawnee power plant of the Tennessee Valley
Authority (TVA). All of the alternatives evaluated at Shawnee are also
considered in the TVA economic studies for sludge disposal options.
Two general categories of FGD processes are available for use by the
utility industry: nonregenerable or throwaway processes which produce a
waste material for disposal and regenerable or recovery processes that
produce a saleable byproduct. Many processes are available in both cate-
gories; however, most utilities are selecting the lime or limestone
process which produces a throwaway sludge (2). Two categories of waste
disposal processes are being used: wet and dry. Wet processes normally
involve pond disposal and dry processes usually involve landfill of
sludge (3,4). The alternatives evaluated by TVA are representative of a
range of disposal options and include both wet and dry disposal processes.
The six alternatives evaluated are (1) untreated ponding, (2) Dravo
Corporation's process, (3) Chemfix process, (4) IU Conversion Systems, Inc.,
(IUCS) process, (5) untreated sludge - fly ash blending, and (6) gypsum
disposal. For each process considered, a base case was established and
definitive estimates of total capital investments and total annual revenue
requirements were calculated. The estimates were all made using a set of
carefully defined common premises and are directly comparable. Cost esti-
mates are based on process background information, flowsheets, material
and energy balances, equipment and system requirements, and raw material,
labor, and utility costs. All estimates of capital investment are
projected to mid-1979 and revenue requirements to mid-1980.
155
-------
BACKGROUND AND DESCRIPTION OF ALTERNATIVES
Many power plants with nonregenerable FGD systems that are now in opera-
tion in the United States use a sludge disposal method involving some
form of onsite ponding or impoundment of untreated material. This method
of disposal, although popular, will not necessarily be the best option
for future installations. Drawbacks, such as potential disposal regula-
tions and limited land availability, have made necessary the development
of other disposal options for FGD wastes (1). Several other treatment
options involving chemical and physical stabilization are available to
the utility industry.
The technology associated with the six disposal alternatives evaluated in
the TVA-EPA work and discussed in this paper is assumed to be proven, but
in many cases is in the development stage and is not actually proven in
full-scale application (i.e., forced oxidation, simultaneous sulfur
dioxide (S02) and fly ash removal, etc.). The primary emphasis of the
work was to evaluate the economics of the disposal alternatives rather
than the process technology (5).
Untreated Ponding
As stated earlier, the untreated ponding option is the alternative
selected most often by the utility industry. Effluent from the scrubber
system is pumped directly to a pond and allowed to settle. Excess water
is recycled to the scrubber system. Very few items of equipment are
required if this option is used, but the capital investment for the dis-
posal pond is very high (7).
Dravo Process
Dravo offers two basic processes for FGD sludge disposal (pond or land-
fill). Although the pond or impoundment aJternative was the base case
for TVA studies, the more recently promoted landfill process may be more
economically attractive. Effluent from the scrubber system is partially
dewatered using a thickener before mixing with Dravo"s fixation additives
(Thiosorbic lime and Calcilox). The treated material is then pumped to
an impoundment area where the material settled and is eventually stabi-
lized. Excess water is recycled to the scrubber system. Dravo's fixation
agents and their entire fixation process are patented (Synearth process)
(6).
IUCS Process
The IUCS system is called the Poz-0-Tec process and involves chemical
stabilization of calcium-based waste materials by mixing with lime and
fly ash. Scrubber system effluent is dewatered using a thickener and
rotary drum filter. The dewatered material (containing about 60£ solids)
is then mixed with fixation additives (lime and fly ash) and trucked to a
disposal site for landfill disposal. Fly ash is a necessary ingredient
for stabilization and can be blended with the sludge cake at the additive
156
-------
mixing stage of processing or included with the sludge following removal
simultaneously with the S02 in the preceding and scrubbing stages of the
FGD process. IUCS reports that the stabilized material is claylike and
can be easily handled, transported by truck, placed, and compacted, and
that the landfill is structurally suitable for future reclamation (7).
Chemfix Process
Chemfix has been applying their technology for sludge stabilization to
wastes generated by metal finishing, automotive assembly, and electronics
operations for several years (8). Chemfix offers a process that yields
a treated stabilized sludge that is reported to be suitable for landfill
disposal. Effluent from the scrubber system is dewatered using a thickener
and rotary drum filter. The filter cake is mixed with two chemical addi-
tives (Portland cement and sodium silicate), during which stabilization of
the sludge is achieved.
Untreated Sludge - Fly Ash Blending
Many power plants are meeting particulate emission requirements for fly
ash by installing equipment for dry fly ash collection. Dry fly ash can
be used in many cases as an additive for blending with dewatered scrubber
sludge to yield a physically stable material. This process of sludge
treatment would allow the utility to dispose of both fly ash and scrubber
wastes in one operation and also to produce a waste product that is
suitable for landfill disposal.
Effluent from the scrubber system is dewatered using a thickener and
rotary drum filters. The filter cake is mixed with dry fly ash which is
pneumatically conveyed from the fly ash collection system to the sludge
disposal facility. The blended material Ls transported by truck to a
landfill disposal site. TVA studies inui ite that this procedure can be
used to produce a product suitable for I;1 Mill disposal and that handling
with trucks and earthmoving equipment is : .?ible (9).
The lime and limestone FCiJ processes can be modi I Led m include a
processing step to force i he oxidation of calcium sulfite sludge (the
normal product ol these processes) to gypsum. Gypsum is a more desirable
waste product because of improved settling properties (settling rate is
about 10 times greater than CaS03) and therefore a reduced volume of
material can be attained through dewatering. The landfill disposal of
gypsum can be accomplished without the use of blending or mixing equip-
ment and fixation additives. Underflow from the scrubber system is
dewatered, using a thickener and rotary drum filter, before it is hauled
by truck to a landfill disposal site. Tests conducted at Shawnee power
plant indicate Lhat SO^ .:iul fly ash can be removed simultaneously in the
scrubber system and therefore the equipment for dry fly ash collection
is not needed (10,11).
157
-------
EVALUATION OF ALTERNATIVES
A comparative economic evaluation of several processes requires that the
basis for the capital investment and revenue requirement estimates be the
same. All TVA studies are made using a predetermined set of design and
economic premises for the power plant, fuel, FGD system, and estimate
calculation procedures. These premises that allow the comparison of
estimates are summarized in the following paragraphs.
Base Case Design Premises
Power Plant—
1. The plant is newly constructed and has a 30-year life.
2. The single coal-fired unit has an output of 500 MW.
3. The total operating life is 127,500 hours with an average
annual capacity of 4,250 hours.
4. The power unit heat input requirement is 9,000 Btu/kWh.
5. The coal heating value is 10,500 Btu/lb.
6. The coal contains 3.5% (by wt) sulfur (dry) and 16% (by wt) ash.
FGD System—
1. A limestone scrubbing process is used for S02 removal.
2. S02 and fly ash are removed to meet NSPS. [EPA issued Federal
Standards of Performance for New Stationary Sources (often called
"new source performance standards" or NSPS).] The allowable S02
emission is 1.2 Ib/MBtu heat input and the particulate emission,
0.1 Ib/MBtu heat input.
3. Eighty-five percent of the ash present in the coal is emitted as
fly ash.
4. Ninety-five percent of the sulfur in the coal is emitted as S02.
5. Effluent from the scrubber system contains 15% solids.
6. All storage facilities have a 30-day capacity and feed bins,
intermediate storage tanks, etc., have an 8-hour capacity.
Untreated Ponding—
1. Effluent (15% solids) from the scrubber system is pumped to a
clay-lined disposal pond.
158
-------
2. The pond is located 1 mile from the scrubber facilities.
3. The sludge settles to 50% solids in the pond and excess water
is recycled to the scrubber system.
4. The FGD process stoichiometry is 1.5 raols calcium oxide per mol
S02 removed.
5. Fly ash and S02 are removed simultaneously in the scrubber system;
therefore the sludge contains both fly ash and calcium wastes.
6. Fifteen percent of the S02 removed is converted to gypsum and the
remaining 85% calcium sulfite.
Dravo Process—
1. Thickened sludge (35% solids) is treated with Dravo additives:
Calcilox (7% of dry solids) and Thiosorbic lime (1% of dry solids).
2. Treated sludge is pumped 1 mile to a clay-lined pond for disposal.
3. Stabilization as a soillike material occurs over a 2- to 4-week
period. Fixed sludge is 50% solids and excess water is recycled
to the scrubber system.
4. The FGD process stoichiometry is 1.5 mols calcium oxide per mol
S02 removed.
5. Fly ash and S02 are removed simultaneously in the scrubber system;
therefore the sludge contains both fly ash and calcium wastes.
6. Fifteen percent of the S02 removed is converted to gypsum and the
remaining 85% calcium sulfite.
IUCS Process—
1. Dewatered sludge (60% solids) is treated with lime (4% of dry solids).
2. Trucks are used to transport the treated material to a landfill
disposal site located 1 mile from the scrubber facilities.
3. Treated sludge is assumed to have claylike properties and can be
placed and compacted in a landfill with typical earthmoving equip-
ment.
4. The FGD process stoichiometry is 1.5 mols calcium oxide per mol S02
removed.
5. Fly ash and S02 are removed simultaneously in the scrubber system;
therefore the sludge contains both flyash and calcium wastes.
6. Fifteen percent of S02 removed is converted to gypsum and the
remaining 85% calcium sulfite.
159
-------
Chemfix Process—
1. Thickened sludge (35% solids) is transported by pipeline (1 mile)
to the disposal site where additional dewatering, mixing with
fixation additives, and landfill placement occurs.
2. Dewatered sludge (60% solids) is stabilized by mixing with two
Chemfix additives: Portland cement (7% of dry solids) and sodium
silicate (2% of dry solids).
3. Treated material is placed and compacted as landfill using
typical earthmoving equipment.
4. The FGD process stoichiometry is 1.5 mols calcium oxide per mol
S02 removed.
5. Fly ash and S02 are removed simultaneously in the scrubber system;
therefore the sludge contains both fly ash and calcium wastes.
6. Fifteen percent of the S02 removed is converted to gypsum and the
remaining 85% calcium sulfite.
Untreated Sludge - Fly Ash Blending—
1. Dewatered sludge (60% solids) is blended with dry fly ash to
yield a physically stable material.
2. Fly ash is removed from the flue gas to meet NSPS using an
electrostatic precipitator (ESP) and pneumatically conveyed to
the sludge treatment area for blending with sludge.
3. The blended material (about 75% solids) is transported to a
landfill disposal site by truck (1 in Lie).
4. Typical carthmovinR equipment is used for placement and compac-
tion in a landfill.
5. The FGD process stoichiometry is 1.5 mols calcium oxide per mol
S02 removed.
6. Fifteen percent of the S02 removed is converted to gypsum and
the remaining 85% calcium sulfite.
Gypsum—
1. The limestone FGD process is modified to provide forced oxidation
of calcium sulfite sludge to gypsum. The FGD process stoichi-
ometry is 1.1 mols calcium oxide per mol S02 removed.
2. Ninety-five percent of the S02 removed is converted to gypsum
and the remaining 5% calcium sulfite.
160
-------
3. Fly ash and S02 are removed simultaneously in the scrubber loop
to meet NSPS; therefore, sludge contains both gypsum and fly ash.
4. Dewatered gypsum (about 80% solids) is transported by truck
(1 mile) to the landfill disposal site.
5. Typical earthmoving equipment is used for placement and compaction
of the gypsum in the landfill.
Economic Premises
A midwestern plant location was selected because of coal availability
for the large number of coal-fired plants in this region. Other economic
assumptions are summarized as follows:
1. All capital cost estimates are based on Chemical Engineering cost
indices (labor index - 237.9, material index - 264.9). Capital
costs are project-i to mid-1979 using these indices. Construction
on the project is assumed to have started in mid-1977 and to be
completed in mid-1980.
2. Direct capital costs cover process equipment, piping and insula-
tion, transport lines, foundations and structural, excavation and
site preparation, roads and railroads, electrical instrumentation,
buildings, and trucks and earthmoving equipment. Material and
labor (fabrication and installation) costs for each of these items
were estimated. These estimates are based on costs obtained from
vendors and on related literature information.
3. Indirect capital costs include engineering design and supervision,
architect and engineering contractor expenses, construction
expenses, contractor fees, contingency, allowance for startup and
modifications, and interest during construction. Two other capital
costs not included as indirect costs, but in the total capital
investment, are working capital and land. These estimates are
based on current industry practice and authoritative literature
sources.
4. Direct costs for revenue requirements include raw materials, labor,
electricity, equipment fuel and maintenance, and analyses. These
costs are projected to mid-1980.
5. Indirect costs for revenue requirements are capital charges and
overheads.
6. Capital charges are based on regulated utility economics.
7. Revenue requirements are projected for an annual 7000 hr/yr (first
year) operation. Other estimates are made for lifetime revenue
requirements that are based on the declining operating profile
of the plant.
161
-------
Case Variations in Design Premises
The base case design premises were altered for selected variables in
order to evaluate the effects of changes in operating conditions and
site-specific design factors. Several of the variations which were
considered are as follows:
1. Plant size: 200 and 1500 MW (the 1500-MW plant is assumed to
be three 500-MW units).
2. Coal composition: Sulfur content, 2.0% and 5.0%; ash content,
12% and 20%.
3. Remaining life of an existing plant: 25, 20, and 15 years.
4. Distance to disposal: 5 and 10 miles.
5. Availability of land for disposal site. Construction: 50% and
75% of optimum.
RESULTS
Two TVA-EPA studies to evaluate the economics of six FGD sludge disposal
alternatives are complete. The capital investments and revenue require-
ments of the base cases and major case variations for the six options
are discussed in this paper. Additional details concerning the cost
estimates can be obtained by reviewing the two TVA-EPA reports (1)
(one of the two reports is not yet published, but details are available
from the author of this paper).
Total System Costs
The total cost of S02 and particulate emission control can be obtained
by combining the cost estimates of the FGD system with waste disposal
system costs. Estimates of FGD system costs (total capital investment
and annual revenue requirements) are available from other TVA-EPA
studies (5,13) and are suitable for combining with the waste disposal
system costs discussed in this report. These costs are summarized in
Tables 1 and 2. The FGD costs presented in this paper apply only to the
base case conditions and therefore cannot be used with waste disposal
systems other than the base cases.
Unit Revenue Requirements
Unit revenue requirements for the base case system and several major
case variations are shown in Table 3.
162
-------
TABLE 1. SUMMARY OF CAPITAL INVESTMENTS FOR
COMBINED FGD AND SLUDGE DISPOSAL SYSTEMS
Disposal process'
Total capital investments
Variation
Site
description
FGD
system
Disposal
system
k$
$/kW
k$
$/kW
Combined
system
k$
u>
Untreated Ponding
Dravo Ponding
Dravo Landfill
Chemfix Landfill
IUCS Landfill
Gypsum Landfill
Untreated sludge - Landfill
fly ash blending
$/kW
36,368C 72.8 17,211 34.4 53,579 107.2
36,368C 72.8 24,114 48.2 60,482 121.0
36,368C 72.8 12,670 25.3 49,038 98.1
36,368C 72.8 13,531 27.2 49,899 99.8
36,368C 72.8 10,717 21.4 47,085 94.2
38,671c'd 77.3 5,411 10.7 44,082 88.2
45,982e 92.0 8,605 17.2 ^4,587 109.2
a. Dewatering equipment for all cases included in the disposal system.
b. The amounts shown are for the base case (mid-1979 costs).
c. Costs are for an FGD system which removes both S02 and fly ash in the scrubber
loop.
d. Cost includes additional equipment required for forced oxidation ($2,300,000).
e. An electrostatic precipitator (ESP) is used to remove fly ash and its installed
cost is included ($9,614,000).
-------
TABLE 2. SUMMARY OF REVENUE REQUIREMENTS FOR
COMBINED FGD AND SLUDGE DISPOSAL SYSTEMS
Revenue requirements3
Disposal process
Variation
Untreated
Dravo
Dravo
Chemf ix
IUCS
Gypsum
Untreated
fly ash
sludge -
blending
Site
description
Ponding
Ponding
Landfill
Landfill
Landfill
Landfill
Landfill
FGD system
Total
annual $
ll,841,500b
ll,841,500b
ll,841,500b
ll,841,500b
ll,841,500b
12,846,800b
13,816,500d
Mills/
kWh
3.38
3.38
3.38
3.38
3.38
>c 3.67
3.94
Total
annual
3,280,
6,701,
6,620,
6,988,
5,291,
3,117,
3,735,
Disposal system
$
000
000
000
000
000
500
000
Mills/
kWh
0.94
1.91
1.89
2.00
1.51
0.89
1.07
$/ton
dry solids
8.08
15.32
15.16
16.51
12.55
7.86
9.20
Combined systems
Total
annual
15,121,
18,542,
18,461,
18,829,
17,132,
15,964,
17,551,
$
500
500
500
500
500
300
500
Mills/
kWh
4.32
5.30
5.27
5.38
4.90
4.56
5.01
a. The amounts shown are for the
b. Costs
c. Cost
are for
includes
d. An ESP is used
base case (mid-1980 costs).
an FGD system which removes
that associated
to remove fly
with forced
both S02 and fly ash in
oxidation
ash and its associated
equipment
($1
the scrubber loop.
,005,300).
operating costs are included ($1,975
,000).
-------
TABLE 3. UNIT REVENUE REQUIREMENTS - ALL PROCESSES
Disposal process
Untreated
Base case*3
Variation from base case
200 MW
1500 MW
Existing, 25-year life
Existing, 20-year life
Existing, 15-year life
12% ash in coal
20% ash in coal
2% sulfur in coal
5% sulfur in coal
5 miles to disposal
10 miles to disposal
Constrained acreage
(50% of optimum)
Constrained acreage
(75% of optimum)
Mills/
kWh
0.94
1.44
0.64
0.55
0.45
0.38
0.83
1.03
0.75
1.10
1.58
2.14
1.18
0.96
$/dry
ton
8.08
12.12
5.55
4.69
3.80
3.19
8.68
7.44
9.37
7.35
13.61
18.48
10.15
8.29
Dravo
Mills/
kWh
1.91
2.60
1.36
1.32
1.21
1.16
1.69
2.12
1.52
2.29
2.32
2.67
2.60
2.25
$/dry
ton
15.32
20.41
10.87
10.30
9.50
9.04
16.43
15.58
17.45
14.08
18.57
21.39
20.82
18.06
IUCS
Mills/
kWh
1.51
2.55
0.99
1.01
1.02
1.04
1.30
1.71
1.33
1.77
1.85
2.14
S/dry
ton
12.55
20.68
8.23
8.24
8.26
8.43
13.05
11.84
15.57
11.29
15.40
17.73
Chemf ix
Mills/
kWh
2.00
3.24
1.37
1.40
1.41
1.43
1.78
2.17
1.70
2.36
2.48
2.86
S/dry
ton
16.51
26.14
11.31
11.36
11.39
11.59
17.86
15.00
20.19
15.01
20.49
23.63
Untreated sludge -
fly ash blending
Mills/
kWh
1.07
1.96
0.65
1.07
1.06
1.06
1.02
1.11
0.91
1.20
1.25
1.39
S/dry
ton
9.20
16.51
5.64
9.01
8.97
8.94
10.77
8.03
11.26
7.88
10.81
11.96
Gypsum
Mills/
kWh
0.89
1.79
0.47
0.88
0.88
0.88
0.86
0.92
0.77
0.93
1.06
1.22
S/dry
ton
7.86
15.42
4.17
7.63
7.62
7.61
9.23
6.75
9.74
6.45
9.37
10.80
a. Basis
Midwest plant location, mid-1980 costs; 7,000 hr/yr plant on-stream time; S02 and fly ash removed to meet NSPS.
b. Base case
New 500-MW plant with 30-year life.
-------
Sludge Disposal System Costs
Major case variations and their effects on costs are discussed in the
following section of the paper. The costs shown in these tables represent
only the costs associated with the sludge disposal area.
Power Plant Size—
The power plant size has an almost direct effect on sludge disposal
costs. A slight economy of scale is seen for the plant sizes evaluated.
Table 4 is a summary of sludge disposal process costs for the alternatives
evaluated.
Coal Composition—
The sulfur and ash contents of the coal also have a direct effect on
the quantity of sludge for disposal. Cost estimates were made for
variable sulfur and ash percentages of the coal. These estimates are
summarized in Table 5.
Remaining Plant Life—
In many cases existing power plants (5-15 years old) are required to
install FGD systems to meet emission regulations. Several cost estimates
were made to evaluate the sludge disposal costs for plants with remaining
operating times of less than 30 years (15, 20, and 25 years). Capital
investments for these cases were considerably less if the disposal
alternative involved ponding. Unit revenue requirements were increased
because the depreciation of capital was taken over a shorter period of
time. Table 6 summarizes the remaining life case variation estimates.
Distance to Disposal Site—
Case variations were considered to determine the effect of the distance
to the waste disposal site on capital investment and revenue require-
ments. These results are summarized in Table 7. The capital investment
and revenue requirements increase rapidly with the increasing distance
to disposal for alternatives using pipelines for slurry transport. Costs
increase for alternatives for using trucks for transport, but not as much
as the pipeline transport alternatives.
Availability of Land—
The quantity of land available for construction of a disposal pond for
untreated sludge can be a significant factor in selecting a disposal
alternative. Several cost estimates were made to evaluate the effect of
land availability on costs. Estimates are normally made in TVA studies
by determining the minimum total pond cost by optimizing between land
cost and construction costs. The quantity of land is therefore the
amount that should be used to obtain the lowest overall pond cost.
166
-------
TABLE 4. TOTAL CAPITAL INVESTMENTS AND ANNUAL REVENUE REQUIREMENTS
FOR PLANT SIZE CASE VARIATIONS
Total capital investment, k$a Annual revenue requirements, k$
Disposal process
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
Power
200b
9,800
13,942
7,193
9,259
6,126
3,988
plant size, MW
500C
17,211
24,114
10,717
13,531
8,605
5,411
1500b
36,455
48,235
20,105
24,104
18,282
9,826
Power
200b
2,014
3,643
3,567
4,529
2,742
2,502
pj.ant sizet MW
500°
3,280
6,701
5,291
6,988
3,735
3,118
1500D
6,746
14,264
10,411
14,362
6,867
4,961
c.
New plant with 30-year life; Midwest plant location; mid-1979 capital costs;
mid-1980 revenue requirements; 7,000 hr/yr on-stream time; coal analyses
(by wt): 3.5% sulfur (dry basis), 16% ash; fly ash and S02 removed to meet
NSPS; 1 mile to disposal site.
Base case premises except plant size.
Base case.
-------
TABLE 5. TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE
REQUIREMENTS FOR COAL COMPOSITION CASE VARIATIONS
oo
Annual revenue requirement. k$
Sulfur in coal, %
Disposal process
Untreated
Dravo
IUCS
Chemf ix
Untreated sludge -
fly ash blending
Gypsum
2b
13,390
19,251
9,345
11,879
7,356
4,782
5C
20,655
28,523
11,957
14,192
9,534
5,884
Ash in
12^
15,031
21,466
9,025
11,123
7,917
5,042
coal, %
20e
19,055
26,028
12,283
14,854
9,309
5,707
Sulfur in coal, %
2b
2,639
5,314
4,654
5,935
3,186
2,707
5C
3,869
8,007
6,118
8,263
4,199
3,252
Ash in
12d
2,902
5,924
4,533
6,229
3,581
3,018
coal, %
20e
3,609
7,406
5,971
7,600
3,896
3,206
a. New plant with 30-year life; Midwest plant location; mid-1980 operating costs; mid-1979
capital costs; 7,000 hr/yr on-stream time; fly ash and S0a removed to meet NSPS; 1 mile
to disposal site.
b. Base case premises except percent sulfur in coal and coal heating value (10,700 Btu/lb).
c. Base case premises except percent sulfur in coal and coal heating value (10,400 Btu/lb),
d. Base case premises except percent ash in coal and coal heating value (11,100 Btu/lb).
e. Base case premises except percent ash in coal and coal heating value (9,900 Btu/lb).
-------
TABLE 6. TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENT
FOR REMAINING POWER PLANT LIFE CASE VARIATIONS
Total capital investment,
, k$a
Remaining power plant life, jrear
Disposal process
Untreated
Dravo
IUCS
Chemf ix
Untreated sludge -
fly ash blending
Gypsum
17
24
10
13
8
5
3Qb
,211
,114
,717
,531
,605
,411
14
21
10
13
8
5
25c
,578
,416
,591
,400
,528
,174
11
18
10
13
8
5
20C
,399
,281
,402
,204
,381
,115
15C
&, 822
15,553
10,269
13,077
8,276
5,076
Annual
revenue
Remaining power
30b
3,280
6,701
5,291
6,988
3,735
3,118
25C
2,906
6,377
5,402
7,152
3,739
3,097
requirements , k$a
plant life, year
20C
2,135
5,941
5,430
7,191
3,724
3,091
15C
2,130
5,728
5,559
7,359
3,712
3,087
a. Midwest plant location; mid-1979 capital costs; mid-1980 revenue requirements;
7,000 hr/yr on-stream time; fly ash and S02 removed to meet NSPS; 1 mile to dis-
posal site.
b. Base case.
c. Same as base case except remaining plant life and boiler heat rate (9,200 Btu/kWh).
-------
•-J
o
TABLE 7. TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENTS
FOR DISTANCE TO DISPOSAL SITE CASE VARIATIONS
Q
Total capital investment, k$
Distance to disposal site, mile
Disposal process
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
lb
17,211
24,114
10,717
13,531
8,605
5,411
5C
26,836
30,994
11,377
18,313
8,969
5,750
10C
37,420
37,765
11,891
20,227
9,334
6,514
rt
Annual revenue requirement, k$
Distance to disposal site, mile
lb
3,280
6,701
5,291
6,988
3,735
3,118
5C
5,527
8,124
6,490
8,675
4,389
3,719
10C
7,504
9,360
7,475
10,003
4,855
4,286
a. New plant with 30-year life; Midwest plant location; mid-1979 capital costs; mid-
1980 revenue requirements; 7,000 hr/yr on-stream time; fly ash and S02 removed to
meet NSPS.
b. Base case.
c. Same as base case except distance to disposal site.
-------
Estimates shown in Table 8 are for systems with disposal ponds constructed
on a less than optimum acreage. Although total land costs are less for
these cases, pond construction costs are much higher than for the optimum
ponds.
Other Variations—
Several other variations from the base case design and economic premises
were considered in TVA sludge studies. Since these case variations had
a lesser effect on the costs than the variations discussed, the results
are not included in this paper.
Lifetime Revenue Requirements
Estimates of the total revenue requirements of waste disposal processes
over the 30-year system life were estimated. These costs, as shown in
Table 9, are cumulative over the 30-year plant life.
CONCLUSIONS
Several conclusions can be derived from the results generated by the
TVA-EPA sludge disposal economic studies.
1. The base case sludge disposal system requiring the lowest capital
investment and annual revenue requirement was gypsum disposal.
This alternative requires a much smaller investment for equipment
than any other alternative except untreated ponding which requires
a very expensive disposal pond. The selection of this alternative
would require that a typical limestone FGD system be modified to
include the forced oxidation of sulfite (S03) compounds to gypsum.
This requires an additional capital investment of $2,300,000.
2. In all case variations, the gypsum process had the lowest total
capital investment and annual revenue requirements.
3. The alternatives involving pond disposal (untreated and Dravo)
required the highest capital investments. All other processes
were for landfill disposal.
4. The three processes involving chemical treatment (Dravo, IUCS,
and Chemfix) all had higher annual revenue requirements than the
three processes involving no chemical treatment.
5. Both unit capital investment and unit revenue requirements were
slightly lower for large plant size.
6. Capital requirements and revenue requirements vary almost directly
in proportion to the quantity of sludge for disposal. A slight
economy of scale is seen. Cases involving coal, ash, and sulfur
content variations are examples of this effect.
171
-------
TABLE 8. TOTAL CAPITAL INVESTMENTS AND
ANNUAL REVENUE REQUIREMENTS FOR LAND AVAILABILITY
CASE VARIATIONS FOR UNTREATED PONDING DISPOSAL
Case
variation
Optimum landc
75% optimum land"
50% optimum landd
Land
requirement,
acre
407
305
204
Total capital
investments, k$
17,211
17,985
22,676
Annual revenue
requirements , k$
3,280
3,365
4,119
b
a. New 500-MW plant with 30-year life; Midwest plant location; mid-
1979 costs; fly ash and S02 removed to meet NSPS; 1 mile to
disposal site.
b. Same as footnote "a" except costs are mid-1980.
c. Base case for untreated disposal option.
d. Same as base case except acreage and cost of disposal pond.
172
-------
TABLE 9. SUMMARY OF LIFETIME REVENUE REQUIREMENTS FOR ALL PROCESSES*
10
Disposal process
Actual cumulative
lifetime
revenue requirements, $
Lifetime average
unit revenue
requirements,
mills/kWh
Discounted
cumulative
lifetime revenue
Levelized unit
revenue
requirements,
mills/kWhc
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
97,757,800
175,764,900
131,224,200
167,942,300
96,526,800
78,072,400
1.53
2.76
2.06
2.63
1.51
1.22
33,612,100
62,052,600
45,381,700
59,099,300
32,801,900
216,513,400
1.35
2.50
1.83
2.38
1.32
1.07
Basis
New plant with 30-year life; Midwest plant location; mid-1980 costs; fly ash and S02
removed to meet NSPS; operating profile: 7,000 hr/yr for 10 years, 5,000 hr/yr for 5 years,
3,500 hr/yr for 5 years, 1,500 hr/yr for 10 years; coal analysis (wt %) - 3.5% sulfur (dry),
16% ash.
Discounted to initial year at 10%.
Equivalent to discounted process cost over life of power plant.
-------
7. The remaining life of a power plant has a significant effect on
the relative ranking of capital investments for the two alterna-
tives involving pond disposal. As the plant life is reduced,
these alternatives become more favorable.
8. The distance to the disposal site greatly increases the capital
investments for the untreated, Dravo, and Chemfix alternatives.
These increases are primarily due to the additional costs for
pumps and pipelines (other alternatives involve truck trans-
portation to disposal site).
9. Case variations for disposal of untreated sludge in ponds con-
structed on less than the optimum acreage have higher total
capital investments than the base (optimum acreage) case. These
variations illustrate the potential problems for plants with a
limited quantity of land available for pond construction.
10. Alternatives involving truck transport and landfill disposal
generally had higher revenue requirements, but lower capital
investments than the alternatives involving pipeline transport
and pond disposal.
The results presented in this paper do not take into account site-specific
waste disposal conditions that a utility may encounter when selecting a
system for installation. Results are based only on predetermined design
and economic premises and should not be interpreted to represent a sLte-
specific disposal situation.
174
-------
REFERENCES
1. Jones, J. W. Research and Development for Control of Waste and
Water Pollution from Flue Gas Cleaning Systems. In: Proceedings
of Symposium on Flue Gas Desulfurization, Vol. II, New Orleans,
Louisiana, March 8-11, 1976. EPA-600/2-76-136b (NTIS PB 262 722),
May 1976. pp. 579-604.
2. Crowe, J. L., and H. W. Elder. Status and Plans for Waste Disposal
from Utility Applications of Flue Gas Desulfurization Systems. In:
Proceedings of Symposium on Flue Gas Desulfurization, Vol. II,
New Orleans, Louisiana, March 8-11, 1976. EPA-600/2-76-136b (NTIS
PB 262 722), May 1976. pp. 565-577.
3. Fling, R. B., W. M. Graven, F. D. Hess, P. P. Leo, R. C. Rossi,
and J. Rossoff. Disposal of Flue Gas Cleaning Wastes: EPA Shawnee
Field Evaluation - Initial Report. EPA-600/2-76-070 (NTIS PB
251 876), March 1976. 221 pp.
4. Leo, P. P., and J. Rossoff. Control of Waste and Water Pollution
from Power Plant Flue Gas Cleaning Systems: First Annual R and D
Report. EPA-600/7-76-018 (NTIS PB 259 211), October 1976.
5. Barrier, J. W., H. L. Faucett, and L. J. Henson. Economics of
Disposal of Lime-Limestone Scrubbing Wastes: Untreated and
Chemically Treated Wastes. TVA Bull. Y-123, EPA-600/7-78-023a,
February 1978. 452 pp.
6. Selmeczi, J. G. Flue Gas Desulfurization and Stabilization.
Dravo Lime Company, Pittsburgh, Pennsylvania, May 1975.
7. Poz-0-Tec Process for Economical and Environmentally Acceptable
Stabilization of Scrubber Sludge and Ash. IU Conversion Systems,
Inc., Philadelphia, Pennsylvania.
8. Conner, J. R. Ultimate Disposal of Liquid Wastes by Chemical
Fixation. In: Proceeding of 29th Annual Purdue Industrial Waste
Conference, Purdue University, West Lafayette, Indiana, May 7-19,
1974. pp. 906-922.
9. Kelso, T. M. Monthly progress report. Tennessee Valley Authority,
Plant Operations Section, Emission Control Development Projects,
Muscle Shoals, Alabama, November-December 1976 and January-
February 1977.
10. Bechtel Corporation. Progress report for work conducted at EPA
Alkali Scrubbing Test Facility at TVA Shawnee Steam Plant,
Paducah, Kentucky, March 1977 to May 2, 1977.
11. Borgwardt, R. H. Sludge Oxidation in Limestone FGD Scrubbers.
EPA-600/7-77-061, June 1977.
175
-------
12. McGlamery, G. G., R. L. Torstrick, W. J. Broadfoot, J. P. Simpson,
L. J. Henson, S. V. Tomlinson, and J. F. Young. Detailed Cost
Estimates for Advanced Effluent Desulfurization Processes. TVA
Bull. Y-90, EPA-600/2-75-006 (NTIS PB 242 541), January 1975.
418 pp.
13. Torstrick, R. L., L. J. Henson, and S. V. Tomlinson. Economic
Evaluation Techniques, Results, and Computer Modeling for Flue Gas
Desulfurization. In: Proceedings of Symposium on Flue Gas
Desulfurization, Vol. I, Hollywood, Florida, November 8-11, 1977.
EPA-600/7-78-058a, March 1978. pp. 118-168.
176
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-092
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Proceedings: Industry Briefing on EPA Lime/Lime-
stone Wet Scrubbing Test Programs (August 1978)
5. REPORT DATE
March 1979
6, PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John E. Williams, Conference Chairman
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
INE624A
See Block 12
11. CONTRACT/GRANT NO.
N.A. (Inhouse)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Proceedings: 8/29/78
14. SPONSORING AGENCY CODE
EPA/600/13
15.SUPPLEMENTARY NOTES BERL-RTP project officer is John E. Williams, MD-61, 919/541-
2483.
16. ABSTRACT
The proceedings document presentations made during the August 29, 1978 industry
briefing conference which dealt with the status of EPA/IERL-RTP's flue gas desul-
furization (FGD) research, development, and application programs. Subjects con-
sidered included: lime/limestone scrubbing test results, forced oxidation, process
cost and energy requirements, by-product disposal options, and future test plans.
The conference provided developers, vendors, users, and those concerned with
regulatory guidelines with a current review of progress made in lERL-RTP's FGD
technology development program.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Flue Gases
Sulfur Oxides
Desulfurization
Scrubbers
Calcium Oxides
Calcium Carbo-
nates
Oxidation
Waste Disposal
Power
Operating Costs
Pollution Control
Stationary Sources
Forced Oxidation
Energy Requirements
13B
21B
07B 07C
07A,07D
131 14G
14A,05A
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport}
Unclassified
21. NO. OF PAGES
180
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------