EPA-650/2-74-010
January 1974
Environmental Protection Technology Series
:;:t-:: JJiS&WftWftWS1' •:•:•:•:•:•:•
"••"•"-
;-lil§
-------
EPA-650/2-74-010
EPA ALKALI
SCRUBBING TEST FACILITY:
LIMESTONE WET SCRUBBING
TEST RESULTS
by
Dr. Michael Epstein, Louis Sybert,
Dr. Shih-Chung Wang, and Charles C. Leivo
Bechtel Corporation
50 Beale Street
San Francisco, California 94119
Contract No. PH 22-68-67
ROAP No. 21ACY-32
Program Element No. 1AB013
EPA Project Officer: Frank T. Princiotta
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
January 1974
-------
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
ABSTRACT
The report describes test results from a prototype lime/limestone
scrubbing test fa.cility for removing SC^ and particulates from flue
gases. The facility consists of three parallel scrubbers — a venturi/
spray tower, a Turbulent Contact Absorber (TCA), and a Marble-Bed
Absorber--each able to treat a 10-Mw equivalent (30, 000 acfm) of
flue gas from a coal-fired boiler at TVA's Shawnee Station. The
short-term (less than 1 day) limestone factorial tests, completed
in February 1973, were conducted at high (6. 0-6. 2) scrubber inlet
liquor pH. Longer term (about 500 hours) limestone reliability ver-
ification tests, completed in September 1973, were conducted at
reduced (5. 6-5. 8) scrubber inlet liquor pH, to increase system
reliability and limestone utilization. A s of early January 1974, more
than 1000 hours of a long-term limestone run on the TCA and more
than 2000 hours of a long-term lime run on the venturi/spray tower
system were completed. The objective of testing since February
1973 has been to identify the most economically attractive lime/
limestone system operating conditions, consistent with reasonable
performance.
This report presents the results, through early January 1974, of
limestone and lime reliability verification and long-term reliability
testing at the Shawnee Prototype Facility.
111
-------
ACKNOWLEDGEMENT
The authors -wish to acknowledge the various personnel from Bechtel,
he Environmental Protection Agency and the Tennessee Valley Authority
yho contributed to the preparation of this report. Specifically,
I. H. Borgwardt of EPA was the major contributor for Section 6,
r. T. Princiotta of EPA was the major contributor for Sections 7 and
i, and R. M. Statnick and D. C. Drehmel of EPA -were the major
:ontributors for Section 3.3. A. H. Abdul-Sattar of Bechtel contributed
o Sections 3 and 5 and J. E. Williams of EPA contributed to Section 4.
Appendix D was written by G. L. Crow and H. R. Horsman of TVA.
The authors also wish to acknowledge the contributions of the Bechtel
.nd TVA on-site personnel at the Paducah Test Facility.
IV
-------
CONTENTS
Section Page
1 SUMMARY 1-1
1.1 EPA Shawnee Test Facility 1-1
1.2 EPA Pilot Facility at Research
Triangle Park 1-10
2 INTRODUCTION 2-1
3 LIMESTONE TEST RESULTS AT THE
SHAWNEE FACILITY 3-1
3. 1 Reliability Verification Limestone
Testing 3-1
3.2 Long-Term Reliability Limestone
Testing 3-25
3. 3 Particulate Removal Efficiencies 3-31
4 OPERATING EXPERIENCE AT THE
SHAWNEE FACILITY DURING LIMESTONE
TESTING 4-1
4. 1 Closed Liquor Loop Operation 4-1
4.2 Demisters 4-2
4. 3 Hot Gas/Liquid Interface 4-8
4.4 Reheaters 4-9
4.5 Fans 4-11
4.6 Pumps 4-13
4.7 Waste Solids Handling 4-14
-------
Section Page
4.8 Scrubber Internals 4-16
4.9 Linings 4-19
4. 10 Instrument Operating Experience 4-21
4. 11 Materials Evaluation 4-26
5 LIME TEST RESULTS AT THE SHAWNEE
FACILITY 5-1
6 TEST RESULTS FROM THE EPA PILOT
FACILITY AT RESEARCH TRIANGLE PARK 6-1
6.1 Description of Equipment and Operation 6-1
6. 2 Material Balances as a Basis for
Evaluating Performance 6-2
6. 3 Improvement of Limestone Utilization 6-6
6.4 Fate of MgCOj in Limestone Feed 6-9
6. 5 Control of Sulfite Scaling 6-9
6.6 Control of Sulfate Scaling 6-12
6. 7 Utilization in Lime Scrubbers 6-16
6. 8 Conclusions 6-20
7 FINDINGS TO DATE 7-1
8 FUTURE TEST PLANS AT THE SHAWNEE
TEST FACILITY 8-1
9 REFERENCES 9-1
Appendices
A Converting Units of Measure A-l
B Graphical Operating Data from Limestone
Reliability Verification Tests B-l
C Test Run Inspection Summary Tables C-l
D TVA Interim Report of Corrosion Studies:
EPA Alkali Scrubbing Test Facility D-l
VI
-------
ILLUSTRATIONS
Figure Page
2-1 Schematic of Venturi Scrubber and Spray Tower 2-3
2-2 Schematic of Three-Bed TCA 2-4
2-3 Schematic of Marble-Bed Absorber 2-5
2-4 Shawnee Test Schedule 2-7
3-1 TCA Inspection 3-29
3-2 Particle Size Distributions at TCA Inlet and
Outlet 3-39
3-3 TCA Particulate Removal Efficiency as a
Function of Particle Size 3-40
4-1 Test Facility Demister Configurations 4-4
5-1 Operating Data for Venturi Run 601-1A 5-3
5-2 Venturi Inspection 5-8
5-3 Spray Tower Inspection 5-9
6-1 Operating Data for Limestone Feed 6-3
6-2 Rate of Sulfite Precipitation in Backmixed
Effluent Hold Tank as a Function of SO2
Concentration in the Liquid Phase 6-7
6-3 Rate of Sulfite Precipitation in Backmixed
Effluent Hold Tank as a Function of
CaCO3 Slurry Density 6-8
6-4 Calcium Sulfate Saturation as a Function of
Sulfite Oxidation in Scrubbers Operating with
Limestone: Chloride Concentration =0 6-15
6-5 Operating Data for Lime Feed 6-17
8-1 Preliminary Schedule for Advanced Shawnee
Program 8-2
Vll
-------
TABLES
Table Page
2-1 Topical and Final Report Description 2-9
3-1 Summary of Limestone Reliability Verification
Tests: Venturi System 3-4
3-2 Summary of Limestone Reliability Verification
Tests: TCA System 3-6
3-3 Summary of Limestone Reliability Verification
Tests: Marble-Bed System 3-8
3-4 Limestone Reliability Verification Test Run
Evaluations: Venturi and Spray Tower 3-10
3-5 Limestone Reliability Verification Test Run
Evaluations: TCA 3-12
3-6 Limestone Reliability Verification Test Run
Evaluations: Marble-Bed Absorber 3-14
3-7 Average Scrubber Inlet Liquor Compositions
for Limestone Reliability Verification Runs 3-19
3-8 Summary of Material Balances for Sulfur and
Calcium from Limestone Reliability Verifica-
tion Tests 3-23
3-9 Overall Farticulate Removal in Venturi and
Spray Tower Scrubber During Factorial Tests 3-33
3-10 Overall Particvdate Removal in TCA Scrubber
with Five Grids and No Spheres During Factorial
Tests 3-34
3-11 Overall Particulate Removal in Marble-Bed
Scrubber During Factorial Tests 3-35
3-12 Overall Particulate Removal in TCA Scrubber
During Limestone Reliability Verification Tests 3-37
IX
-------
Table Page
4-1 Test Facility Demister Specifications 4-3
4-Z Corrosion Test Results 4-28
6-1 Typical Performance Characteristics of TCA
and Multigrid Scrubbers as Determined by
Material Balance 6-4
6-2 Comparison of Backmixed and Plug-Flow
Effluent Hold Tank Designs 6-10
6-3 Effect of Feed pH on Lime Scrubber Operation 6-18
-------
Section 1
SUMMARY
1.1 EPA SHAWNEE TEST FACILITY
The EPA Shawnee test facility consists of three parallel scrubber
systems: (1) a venturi followed by a spray tower; (2) a Turbulent
Contact Absorber (TCA); and, (3) a Marble-Bed Absorber. Each
system is capable of treating approximately 10 Mw equivalent (30,000
acfm* @ 300°F) of flue gas containing 1800-4000 ppm sulfur dioxide
and 2 to 5 grains/scf of particulates.
The object of the limestone short-term (less than 1 day) factorial
tests is to determine the effect of independent variables (e. g. , gas
rate) on SO? removal for the scrubber systems. These tests were
conducted at a scrubber inlet liquor pH range of from 6. 0 to 6. 3.
The results of the limestone factorial tests have been presented in
References 1 and 2.
Although it is the policy of the EPA to use the Metric System for
quantitative descriptions, the British System is used in this
report. Readers who are more accustomed to metric units are
refferred to the conversion table in Appendix A.
1-1
-------
The major objective of the limestone and lime longer-term (greater
than 2 weeks) reliability verification tests is to define regions for
reliable (e.g., freedom from scaling, plugging, erosion, corrosion,
etc. ) operation of the scrubber systems. The limestone reliability
verification tests -were conducted at reduced scrubber inlet liquor pH
(5. 6-5. 9), in order to reduce sulfite scaling potential and increase
limestone utilization (lOOx moles SO_ absorbed/mole CaCO added).
u J
For the limestone verification runs on the venturi/spray tower system,
SO removals from 67-82 percent have been obtained, with an average
Ct
limestone utilization of 68 percent, for pressure drops from 10 to
14.5 inches HO (9 inches HO across venturi). The venturi/spray
£ C*
tower runs showed that washing the demister from the underside
(upstream) can be effective in reducing soft solids accumulations on
the demister blades. However, some accumulation of solids has
occured even with improved -washing procedures. Spiral tip stain-
less steel nozzles have operated in the spray tower for over 2100
on-stream hours of limestone scrubbing without significant erosion.
However, erosion was significant after an additional 2150 on-scream
hours of lime scrubbing operation; of a total of 28 nozzles, 9 were
severely eroded and 15 were considerably worn.
For the limestone verification runs on the three-bed TCA system,
SO_ removals from 77-88 percent have been obtained, with an average
L*
limestone utilization of 77 percent, for pressure drops from 6. 5 to
10. 5 inches ^O (includes about 2 inches ^O across Koch Flexitray).
The TCA runs showed that, the Koch Flexitray was effective in mini-
mizing soft solids accumulations on the chevron demister blades at
1-2
-------
or below a superficial gas velocity of 8 ft/sec. Deposits of scale and
soft solids continued to occur, however, and could have resulted in
plugging during prolonged operation. Significant problems with the
TCA scrubber system during the limestone reliability verification
tests have been:
(1) Erosion and subsequent failure of the plastic spheres.
Present sphere life is under 2000 hours.
(2) Erosion of wire mesh support grids. Sturdier support
grids (parallel 3/8 inch rods) have been installed dur-
ing subsequent long-term reliability testing.
(3) Plugging of the inlet gas duct at the hot gas/liquid
interface. The hot flue gas is cooled with process
slurry before entering the neoprene rubber lined
TCA scrubber. The problem of soft-solids buildup
in the cooling zone has been solved by careful selec-
tion of the proper size, orientation, location and
number of cooling spray nozzles, and by careful
selection of the sootblowing schedule.
(4) Plugging of the mist eliminator. Progress has been
made in reducing the magnitude of this problem but
it has not been completely resolved at this time.
For the limestone verification runs on the single-stage Marble-Bed
system, SO removals from 65-77 percent have been obtained, •with
L4
an average limestone utilization of 67 percent, for pressure drops
from 7.5 to 11 inches H_O. A severe operating problem has been
L*
associated with the erosion of the bed spray nozzles and subsequent
pluggage of the marble bed and demister. This system has not been
operated long enough to have solved these problems.
1-3
-------
The slurries contain substantial quantities of fly ash (approximately
40 weight percent of solids is fly ash). There is evidence based on
preliminary information generated at the EPA-RTP pilot facility that
the presence of fly ash accelerates erosion.
The slurry analytical data for the reliability verification tests has
indicated that the process liquors are supersaturated with respect to
sulfate. The sulfate supersaturation increases with decreasing efflu-
ent residence time and/or with decreasing percent solids. The dis-
solved solids concentrations have ranged from about 7000 ppm (clari-
fier only) to 18, 000 ppm (clarifier with centrifuge or filter). The
major component in the slurry liquor is chloride. The chlorides
present in the coal are absorbed from the flue gas in the scrubber.
The objective of the long-term test is to operate continuously for four
to six months. On October 24, 1973, a limestone long-term reliability
test was begun on the TCA system. Based on the results of the relia-
bility verification tests and the tests conducted at the EPA pilot facility
in Research Triangle Park, the following conditions were chosen for
the long-term test:
Gas Rate, acfm @ 300°F 25,000
Gas Velocity, ft/sec 9.8
Liquor Rate, gpm 1200
L/G, gal/mcf 64
Percent Solids Recirculated 15
Effluent Residence Time, min. 10
Total Pressure Drop, in. f^O 8. 5
Percent SO2 Removal (controlled) 80-85
1-4
-------
Three beds of spheres were used, with five inches of spheres/bed.
The top bed used Universal Oil Products (UOP) supplied thermo-
plastic-rubber (TPR) spheres and the bottom two beds UOP supplied
high density polyethylene (HDPE) spheres. A summary of the operat-
ing data for the test is as follows:
Percent Limestone Utilization 71
Inlet SO2 Concentration, ppm 1600-4000
Scrubber Inlet pH Range 5. 7-6. 0
Scrubber Outlet pH Range 5.3-5. 6
Percent Solids Discharged 42
Dissolved Solids, ppm 8000
After approximately 500 hours of operation the run was terminated,
due to unusually heavy solids build-up on the underside of the Koch
Flexitray, on the scrubber walls between the top bed and the Koch
tray and on the demister blades. Also, numerous (over 200) half-
spheres of the TPR spheres were found in the scrubber and slurry
circulating system. Half-spheres of TPR were also found lodged in
the TCA inlet slurry spray nozzles. It should be noted that the scrub-
ber beds (and bottommost .grid) were free of scale after the 500 hour
operating period, as was expected. The HDPE spheres had lost from.
8-14 percent of their original weight and the TPR spheres about 2. 6
percent.
It is hypothesized that the soft solids accumulations below the Koch
tray were due to the partial blockage of the TCA slurry inlet nozzles
by the TPR half-spheres, which produced high pressure drops across
the nozzles, resulting in excessive entrainment of fine slurry droplets.
The partially blocked nozzles may have also contributed to a large
degree to the mist eliminator pluggage, but excessive gas velocity is
a contributing factor.
1-5
-------
On November 22, 1973, a new TCA limestone long-term reliability
test was begun. The TPR spheres in the top stage had been replaced
with HDPE spheres. The run conditions were identical to those for
the initial test, excepting that the gas velocity has been dropped to
8 ft/sec (20,500 acfm). The velocity was reduced, because more de-
tailed investigation of previous reliability verification runs indicated
that long-term reliability for the present Koch tray-demister configu-
ration should not be expected at a gas velocity of 9. 8 ft/sec. The
operating data for this test -was essentially the same as for the initial
test.
After 1190 hours of operation the run was interrupted, in order to
check the wear rate of the HDPE spheres in the three beds. Pressure
drop across the chevron demister increased slightly during the initial
800 hours of operation, and during the last 400 hours increased more
rapidly to a final level about 1. 5 times the initial, value of 0. 18 inches
HO. An inspection of the system showed that the general appearance
of the scrubber was good, with only very slight scale on the scrubber
walls and bar-grids. Heavy solids deposits, however, covered the
inlet slurry spray nozzles and header and adjacent walls. Also a
heavy, relatively uniform (about 1 inch thick) solids layer covered the
underside of the Koch tray; approximately 5 percent of the valves were
completely plugged. All four inlet slurry spray nozzles were partially
plugged with debris, primarily with plastic covering from pipe insula-
tion. Demister plugging was confined to the bottom two passes only,
reducing the free flow area by about 15 percent.
As with the previous TCA run, it is hypothesized that the solids ac-
cumulations on the inlet slurry spray nozzles and header and on the
1-6
-------
underside of the Koch tray were primarily caused by partial blockage
of the TCA slurry inlet nozzles by debris, which produced high pres-
sure drops across the nozzles, resulting in excessive entrainment of
fine slurry droplets. Also, the blocked nozzles may have been a ma-
jor factor contributing to the demister pluggage observed. The next
long-term limestone reliability test on the TCA will be conducted with
a strainer in the recirculating slurry line to catch debris and with
TPR spheres in all three beds.
An initial lime reliability verification test on the venturi/spray tower
system at the Shawnee facility was begun on October 9, 1973. The
test was conducted at a scrubber inlet liquor pH of 8.0, a total (ven-
turi and spray tower) liquid-to-gas ratio of 96 gal/mcf, an effluent
residence time of 12 minutes and a percent total solids recirculated
of 8 percent. The lime utilization was approximately 90 percent and
the SO_ removal approximately 85 percent, with a pressure drop of
£i
12. 5 inches H~O (9 inches H_O in venturi included). The scrubber
outlet pH was between 4. 6 and 5. 4.
After 2153 hours (3 months) of on-stream operation the lime test was
terminated due to a rapid increase in demister pressure drop through-
out the final month of operation. This period of operation represents
the longest continous run for prototype or full-scale installations of lime/
limestone scrubbing systems in the United States, to date. Inspection
after shutdown showed that scale formation in the scrubber vessel
was not sufficiently heavy to interfere with the gas flow. The mist
eliminator was substantially blocked, partly by solids that had fallen
down from the outlet duct work and partly by scale formation, mainly
1-7
-------
on the bottom (inlet) vanes. The rubber-lined shell of the spray tower
was covered with from 1/8 to 3/8 inch thick sulfate scale. Also, about
3/16 inch thick sulfate scale was found in the rubber lined variable speed
pumps. It should be noted that scale formation in the spray tower,
circulating slurry piping and pumps did not prevent continual operation
of the system or necessitate termination of the 3 month long lime re-
liability test. However, scale particles in nozzles and strainers re-
quired periodic maintenance.
It is hypothesized that the sulfate scale formation occurred, primarily,
during the latter month of testing, when the clarifier and filter were
used for solids dewatering (~-47 percent solids discharged) and the cal-
culated liquor sulfate saturation was about 190 percent (90 percent
supersaturated). An earlier inspection of the system, after 666 hours
of operation, revealed a relatively clean scrubber. During this early
portion of the test the clarifier was used for solids dewatering and the
percent solids discharged (~23 percent) was lower than desired; the
calculated liquor sulfate saturation was 150 percent. The next long-
term lime reliability test will be made under conditions intended to re-
duce recirculating liquor sulfate saturation and the accumulation of
solids in the outlet ductwork.
During the factorial limestone testing, overall particulate removal ef-
>'"
ficiencies of 99. 4 to 99. 8 percentr were obtained for the Chemico ven-
turi at a gas flow rate of 30, 000 acfm (330°F) and liquid-to-gas ratios
from 13 to 27 gal/mcf, with venturi plug pressure drops from 6 to 12
For an average scrubber inlet grain loading of 3. 5 grains/scf, a par-
ticulate removal of 99. 0 percent would correspond to 0. 07 Ibs particu-
late discharged per 106 BTU.
1-8
-------
inches HO. For the spray tower, the removal efficiency was about
98. 5 percent at a gas velocity of 4 ft/sec and a liquid-to-gas ratio of
40 gal/mcf. For the TCA scrubber with 5 grids and no spheres, the
overall removal efficiencies were 98. 6 to 99. 8 percent at a gas velocity
of 7. 5 ft/sec and a liquid-to-gas ratio of 50 gal/mcf, with total pres-
sure drops (includes Koch tray, demister and inlet duct) from 4 to
7 inches H_O. The Marble-Bed scrubber gave an overall particulate
L*
removal efficiency range of 98. 8 to 99. 6 percent during the factorial
limestone tests, at a gas velocity of 5 ft/sec and a liquid-to-gas ratio
of 54 gal/mcf, with 12 inches HO total pressure drop.
A limited series of particulate removal tests on the TCA system was
performed by EPA during the reliability verification limestone testing.
Overall removal efficiencies of 98. 7 to 99. 9 percent were achieved at
gas velocities from 8 to 10 ft/sec, liquid-to-gas ratios from 40 to
80 gal/mcf, and total pressure drops from 5. 5 to 10 inches HO. The
L*
higher pressure drops generally gave higher overall removal efficien-
cies. For the submicron particles (0. 11 to 0.99 microns), the effici-
encies were 95 to 99 percent (increasing with particle size) at 9. 8 inches
H2O total pressure drop, 93 to 95 percent at 7.6 inches H^O, and 71
to 90 percent at 5. 6 inches H_O. Because of the limited number of
tests, conclusions regarding submicron collection efficiency should
be reserved until additional testing can be carried out. The inlet mass
loadings during these tests were 2 to 3 grains/scf with a mass mean
diameter of about 23 microns. The mass mean diameters for the out-
let particles were about 0. 5 to 0. 75 microns, depending upon the pres-
sure drops.
1-9
-------
The pumps used for slurry service at the test facility are rubbed lined,
variable speed, centrifugal pumps with Hydroseals or Centriseals. In
general, the pumps have performed satisfactorily and, in general, the
rubber linings have been found to be in excellent condition. Operation
of the Moyno pumps (for makeup limestone and lime slurry service)
has also been satisfactory.
The major problem with the induced draft fans at the facility has been
high fan vibration. This problem has been controlled by (1) addition
of shims to the bearings or replacement of bearings, (Z) insulation of
the fan housings, (3) welding balance weights onto the fan shrouds, and
(4) adding additional bracing to the outboard pedestals.
The reheaters originally employed at the facility were fuel oil fired
units with separate combustion air supply and with combustion
occurring in the flue gas stream. Quenching of the flame by the
cold (1Z8 F) flue gas and subsequent formation of soot in the exiting
flue gas were effectively reduced by providing an isolated combus-
tion zone (stainless steel sleeves) in each reheater and by replacing
the turbulent mixing type fuel oil nozzles with mechanical atomizing
nozzles. These modifications performed at the test facility are
acceptable as an expedience for operation. However, operation is
still not consistent with the requirements for long-term sustained
reliability.
The rubber linings (in the spray tower, TCA, Marble-Bed, process
water hold tanks, pumps, circulating slurry piping and agitator
blades) have been found, generally, to be in excellent condition.
Essentially no erosion or deterioration has been noted, except
1-10
-------
slight wear on some of the rubber-coated agitator blades. Hairline
cracks have been noted in the Flakeline glass lining on the effluent
hold tanks and clarifiers. However, the cracks did not appear to
penetrate the entire thickness of the lining.
The concentration of solids in the underflow of the larger (30 foot
diameter) TCA clarifier approaches the expected final settled density
of the sludge (about 40 percent by weight). However, periodic high
solids carryover in the overflow continues to occur in the two smaller
(20 foot diameter) venturi and Marble-Bed clarifiers. Also, the
highest solids concentration in the underflow streams from these
smaller units averages about 25 percent by weight.
Operation of the centrifuge had been satisfactory. The cake dis-
charged contained 56-62 weight percent solids. However, severe
erosion on the bowl and casing necessitated repair of the unit after
about 1400 on-stream hours of operation.
Tests have indicated that the dewatering capability of the filter
corresponds to 50-55 and 45-50 weight percent solids in the cakes
from limestone and lime slurries, respectively. Cake discharge
from the nylon filter cloth has been satisfactory without the use of
mechanical equipment (scraper or wire). However, the useful
life of the nylon and polypropylene filter cloths tested to-date is
unsatisfactory.
Operating experience has been gained with two types of pH meters:
(1) a Uniloc Model 320 in-line flow-through type and (2) a Uniloc
1-11
-------
Model 321 submersible type. The performance of the in-line flow-
through type meter has been unsatisfactory due to erosion of the
glass cells by the slurry, their high rate of failure and the frequent
plugging of the sample lines. For the submersible type pH meters,
cell erosion, cell breakage and sample line pluggage have not been
experienced during the approximately 1300 hours of operation.
Experience with the Ohmart radiation-type density meter indicates
a loss of calibration in the range of about 1 to Z percent per week.
Also, the meter accuracy is affected by the accumulation of scale
in the sample line which can be removed only during the scrubber
shutdowns. The sample lines and the probes of the bubble-type
(differential pressure) density meters plug frequently and require
significant maintenance. However, these meters are accurate when
clean and can be used to check the calibration of the radiation-type
density meters. Dynatrol density cells (using the vibration principle
of the U-tube for continuous response to density changes) have been
installed to measure the densities of the lime slurry feed to the
venturi/spray tower system and the circulating limestone slurry
to the TCA system. The performance of the type of density meter
has thus far been encouraging.
Operating experience with control valves in slurry service has
generally been unsatisfactory. Severe erosion in a short time has
been caused by the increasing velocity during throttling operation.
This deterioration has been observed in the stainless steel plug
valves, globe valves and rubber pinch valves. Satisfactory and
trouble free flow control has been experienced only with variable
speed pumps.
1-12
-------
1.2 EPA PILOT FACILITY AT RESEARCH TRIANGLE PARK
Two pilot-scale TCA scrubber systems (400 acfm @ 300°F each)
have been installed at the EPA facility in Research Triangle Park,
N. C. , in support of the Shawnee testing activities. The following
conclusions have been drawn from the limestone and lime results
of the TCA pilot-plant testing with inlet SC^ feed concentrations
of 3000 ppm:
• For limestone scrubbing, control of the scrubber effluent
pH below 6. 2 will prevent calcium sulfite scaling. However,
this conclusion is based on tests run without chloride in the
scrubbing liquor.
• Acceptable scrubber inlet pH (reasonable SC>2 removal and
reasonable lime utilization) for lime scrubbing is in the
7-8 range.
• Limestone dissolution kinetics are improved by plug flow
reaction; higher utilizations can be achieved by effluent
hold tank designs such as U-tubes or a series of stirred
tanks that approximate plug flow.
• For limestone scrubbing, dissolution of CaCOg is the
rate controlling step for SC>2 absorption. For high-
calcium lime scrubbing, dissolution of CaSO, is the
rate controlling step.
• Hold tank residence time must exceed 5 minutes in a
limestone system. 10 minutes appears to be a good
choice. For a lime system, 5 minutes appears adequate.
• The scrubber effluent will always, be supersaturated with
CaSO^/2H2O when a scrubber is operated with saturated
feed.
• Supersaturation is maximum at the first bed TCA support
grid and rapid scaling of the grid by calcium sulfate
occurs at liquid-to-gas ratios less than 65 gal/mcf with
no fly ash in the system. The liquid-to-gas ratio must
exceed 65 for reliable operation with saturated calcium
sulfate feed.
1-13
-------
The presence of fly ash reduces the rate of scaling by
calcium sulfate.
Sulfate scaling can be eliminated by operating a scrubber
in the unsaturated mode. Closed liquor loop unsaturated
operation can be achieved in a chloride-free scrubber by
reducing oxidation below 19 percent. Under these condi-
tions the sulfate generated by oxidation is purged entirely
as solid solution (i.e., calcium sulfate in calcium sulfite
crystal lattice). Additional work is needed for a system
with chlorides present in the process liquor, since all
commercial systems will have caloride.
For limestone scrubbing, oxidation is a controllable
variable within the limits required for unsaturated operation.
It can be reduced by eliminating air contact in the effluent
hold tank (e.g. , using sealed stirred tank or plug flow
tanV^. Further reduction can be attained by circulating high
percent solids (\vhich affects the amount of liquor circulating
through the clarifier or filter).
Sulfate bound as solid solution is less soluble than the pure
salt and, therefore, the potential for water pollution is
reduced.
The dolomitic component of limestone feedi.; is essentially
inert and leaves the scrubber in the same form within the
sludge.
1-14
-------
Section 2
INTRODUCTION
In June 1968, the Environmental Protection Agency (EPA), through
its Office of Research and Development (OR&D) and Control Systems
Laboratory, initiated a program to test a prototype lime and lime-
stone wet-scrubbing system for removing sulfur dioxide and particulates
from flue gases. The system is integrated into the flue gas ductwork
of a coal-fired boiler at the Tennessee Valley Authority (TVA)Shawnee
Power Station, Paducah, Kentucky.
Bechtel Corporation of San Francisco is the major contractor and
test director, and TVA is the constructor and facility operator.
Three major goals of the test program are: (1) to characterize as
completely as possible the effect of important process variables on
sulfur dioxide and particulate removal; (2) to develop mathematical
models to allow economic scale-up of attractive operating configura-
tions to full-size scrubber facilities; and, (3) to perform long-term
reliability testing.
The test facility consists of three parallel scrubber systems: (1) a
venturi followed by a spray tower; (Z) a Turbulent Contact Absorber
(TCA); and, (3) a Marble-Bed Absorber. Each system is capable of
2-1
-------
treating approximately 10 M-w equivalent (30,000 acfm @ 300°F) of
flue gas containing 1800-4000 ppm sulfur dioxide and 2 to 5 grains/
scf of particulates. Each system can be operated with any combin-
ation of clarifier/filter/pond or clarifier/centrifuge/pond for solids
disposal. The test facility has been described in detail in Refer-
ences 1, 2 and 3.
The venturi scrubber (manufactured by Chemical Construction Co. )
contains an adjustable throat that permits control of pressure drop
under a wide range of flow conditions. Although a venturi is ordin-
arily an effective particulate removal device, gas absorption is
limited (in limestone wet-scrubbing systems) by low slurry residence
time. For this reason the after-absorber (spray tower) was included
for additional absorption capability. The TCA scrubber (manufac-
tured by Universal Oil Products) utilizes a fluidized bed of low den-
sity plastic spheres which are free to move between retaining grids.
The Marble-Bed scrubber (supplied by Combustion Engineering Co. )
utilizes a packing of 3/4-inch glass spheres (marbles). A "turbulent
layer" of liquid and gas above the glass spheres enhances mass
transfer and particulate removal. Figures 2-1, 2-2 and 2-3 (drawn
roughly to scale) show the three scrubber systems along with the
demisters selected for de-entraining slurry droplets in the gas stream.
The following sequential test blocks were defined for the program:
(1) Air/water testing
(2) Sodium carbonate testing
2-2
-------
GAS OUT
CHEVRON DEMISTER
AFTER-SCRUBBER
INLET SLURRY
THROAT
ADJUSTABLE PLUG
VENTURI SCRUBBER
DEMISTER WASH
DEMISTER WASH
INLET SLURRY
EFFLUENT SLURRY
5'
APPROX.SCALE
EFFLUENT SLURRY
Figure 2-1. Schematic of Venturi Scrubber
and Spray Tower
2-3
-------
GAS OUT
CHEVRON DEMISTER
INLET KOCH TRAY
WASH LIQUOR
KOCH TRAY
* EFFLUENT KOCH
TRAY WASH LIQUOR
STEAM SPARGE
RETAINING GRIDS
GAS IN
3
3 O _
00 0°0
00 O
oo
0 O °0
Q?5_<£j
\ I /
INLET SLURRY
-MOBILE PACKING SPHERES
t-
APPROX.SCALE
EFFLUENT SLURRY
Figure 2-2. Schematic of Three-Bed TCA
2-4
-------
GAS OUT
DEMISTER WASH
INLET SLURRY
INLET SLURRY
GAS IN
DEMISTER WASH
CHEVRON DEMISTERS
TURBULENT LAYER
GLASS SPHERES
EFFLUENT SLURRY
EFFLUENT SLURRY
5'
I 1
APPROX.SCALE
Figure 2-3. Schematic of Marble-Bed Absorber
2-5
-------
(3) Limestone wet-scrubbing testing
(4) Lime -wet-scrubbing testing
The limestone and lime wet-scrubbing test blocks have been divided
into three general catagories: (1) short-term (less than 1 day)
factorial tests, (2) longer term (over 2 weeks) reliability verifica-
tion tests, and (3) long-term (4 to 6 months) reliability tests. The
1 object of the factorial tests is to determine the effect of independent
variables (e.g. gas rate) on SC>2 removal for the scrubber systems.
The primary objective of the reliability verification tests is to define
regions for reliable (e. g. scale free) operation of the scrubber
systems. The object of the reliability tests is to determine the long-
term operating reliability for the scrubber systems and to develop
more definitive process economics data and scale-up factors.
The test program schedule is presented in Figure 2-4. As can be
seen in the figure, the air/water, sodium carbonate, limestone
factorial and limestone reliability verification tests have been com-
pleted. As of early January 1974, a three month lime reliability
verification test has been completed on the venturi/spray tower
system, as -well as seven weeks of a limestone reliability test on
the TCA system.
Two smaller scrubbing systems (400 acfrn @ 300°F each), which
are capable of operating over a wide range of operating conditions,
have been installed at the EPA facility in Research Triangle Park,
North Carolina, in support of the Shawnee prototype testing activities.
2-6
-------
TEST PROGRAM FUNCTIONS
SYSTEM CHECK-OUT
AIR-WATER & SODIUM CARBONATE TESTING
LIMESTONE WET-SCRUBBING TESTING:
Short -Term Factorial Tests
Reliability Verification Tests
Reliability Tests
LIME WET-SCRUBBING TESTING:
Reliability Verification Tests
Short-Term Factorial Tests
Reliability Tests
ENGINEERING & COST ESTIMATE STUDIES
1972
MAMJJASOND
123456789 10
^•M
—
1973
JFMAMJJASOND
11 12 13 14 15 16 17 18 19 20 21 22
— 1 1
1 1
' A "
i f i
_T
BOILER OUTAGE & '
SYSTEM MODIFICATIONS
«^^—
1974
J. F M A M J
232425262728
•
-
Figure 2-4. Shawnee Test Schedule
-------
The small pilot scale scrubber systems are capable of simulating
the TCA scrubber system and have generated large quantities of
closed liquor loop data on certain TCA configurations.
This report presents the. results, through early January 1974, of
(1) limestone and lime reliability verification and long-term relia-
bility testing at the Shawnee Prototype Facility, and (2) limestone
and lime testing at the EPA Pilot Facility at Research Triangle
Park. In Table 2-1, a description of the reports which are presently
scheduled for general distribution is presented. Results from the
air/water, sodium carbonate and limestone short-term factorial
testing at the Shawnee facility have been presented in the August
1973 Topical Report (Reference 1), listed in Table 2-1.
2-8
-------
Table 2-1
TOPICAL AND FINAL REPORT DESCRIPTION
Report Title
Information to be Included
Estimated
General
Publication
Date
vO
EPA Alkali Scrubbing Test
Facility: Sodium Carbonate
and Limestone Test Results
EPA Alkali Scrubbing Test
Facility: Limestone Wet-
Scrubbing Test Results
EPA Alkali Scrubbing Test
Facility: Lime Wet-
Scrubbing Test Results
EPA Alkali Scrubbing Test
Facility: Final Report
Summary of operational problems and resolutions,
planned and actual test designs, results of air/
water and Na.,CO, testing, utilization of data for
model development, results of factorial limestone
testing with interpretation of data.
Summary of operating problems and resolutions
associated with reliability verification testing,
planned and actual test design, interpretation of
data, status of process model development and
selection of parameters for limestone long-term
reliability testing.
Summary of operational problems and resolutions
associated with lime reliability verification test-
ing, planned and actual test designs, results of
factorial lime testing, status of process model
development, interpretation of data and status of
limestone reliability testing.
Summary of total test program with particular
emphasis on lime and limestone reliability test
results, mathematical models, scale-up design
and economic studies.
August, 1973
(actual date)
January. 1974
(actual date)
April, 1974
July, 1974
-------
Section 3
LIMESTONE TEST RESULTS AT THE SHAWNEE FACILITY
Performance data from limestone reliability verification and long-
term reliability testing at the Shawnee facility are presented in this
section, along with an evaluation of each reliability verification test.
Operating experience associated with specific system components
(e.g., demisters, reheaters, SO? gas analyzers) will be discussed
in detail in Section 4. Results from the limestone factorial tests
have been presented in References 1 and 2.
3. 1 RELIABILITY VERIFICATION LIMESTONE TESTING
The major objective of the closed liquor loop limestone and lime re-
liability verification tests is to identify areas or regions for reliable
long term (4-6 month) operation consistent with reasonable SO?
removal.
A majority of the reliability verification tests were on-stream for
approximately 500 hours (3 weeks). It should be noted that, it is dif-
ficult to assess long-term reliability from runs lasting 500 hours,
especially when small quantities of scale* or soft solids are present
oil system components.
*
In this report "scale" refers only to crystalline hard solids, and
"solids" or "soft-solids" refer to mud-like slurry solids. "Plug-
ging" refers to the accumulation of mud-like slurry solids.
3-1
-------
Summaries of the limestone reliability verification test results for
the venturi/spray tower, TCA and Marble-Bed systems are presented
in Tables 3-1, 3-2, and 3-3, respectively. Operating data for Runs
506-1A (venturi/spray tower), 510-2A (TCA) and 506-3B (Marble-
Bed) are graphically presented in Appendix B. Graphical presenta-
tion of the operating data for all of the limestone reliability verifica-
tion runs will be included in the Final Report (see Table 2-1).
The significant data from the detailed inspection reports prepared at
the test facility during the reliability verification tests are presented
in Appendix C. The reliability data summarized in Tables 3-1, 3-2,
and 3-3 have been obtained from the data in Appendix C.
3. 1. 1 System Reliability
Reliability verification tests presented in Tables 3-1, 3-2, and 3-3
has been evaluated in Tables 3-4, 3-5, and 3-6.
The major system problems addressed at the test facility are asso-
ciated with scaling and plugging of the equipment. A majority of the
component problems can be solved by improved design. The major
variables affecting scaling and plugging tendencies are: (1) effluent
residence time, (2) percent solids recirculated, (3) percent oxidation
of sulfite to sulfate, (4) gas velocity, (5) liquid-to-gas ratio, and
(6) scrubber pH.
The limestone reliability verification tests have been run at reduced
scrubber inlet liquor pH (5. 6-5.9) to decrease the potential for scaling
3-2
-------
and to increase limestone utilization. A modest reduction in SC>2 re-
moval, from high-pH'1' performance, is the price of the increased
system reliability and limestone utilization.
The test conditions for the initial reliability verification runs were
selected in order to give maximum probability for reliable operation,
consistent with reasonable SC>2 removal. Subsequent tests were made
to observe the effects of reduced effluent residence time, reduced
•
percent solids recirculated, increased gas velocity (decreased liquid-
to-gas ratio) and increased scrubber inlet liquor pH on system reli-
& *•'•*
ability."' In addition, the effects of demister type (polypropylene vs.
stainless) and demister wash location (upstream vs. upstream and
downstream washing) were investigated in the venturi/spray tower
system.
Venturi System. The initial venturi system test run 501-1A (see
Tables 3-1 and 3-4) was conducted, as mentioned previously, under
conditions selected to maximize the probability for reliable operation,
i.e. , an effluent residence time of 20 minutes, a percent solids recir-
culated of 15-16 percent' ''"^ and a spray tower gas velocity of 5 ft/sec.
As expected, the relative condition of the system at the end of the test
was good (see Table 3-4). Subsequent venturi system runs were made at
more economically attractive operating conditions, as mentioned pre-
viously. Run 507-1A (see Tables 3-1 and 3-4) indicated relatively
*
The limestone short-term factorial tests were run at a scrubber
inlet liquor pH range from 6. 0 to 6. 3.
**
Minimizing effluent residence time and percent solids recirculated,
and maximizing gas velocity is, of course, economically attractive.
#**
Approximately 40 percent of solids is fly ash.
3-3
-------
Table 3-1
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: VENTURJ SYSTEM
Run No.
Test Objectives
Start-of-Run Date
End -of -Run Date
On Stream Hours'*'
Gas Rate, acfm @ 330°F
Spray Tower Gas Vel., fpa @125°F
Venturi/Spray Tower
Liquor Rates, gpm
Spray Tower L/C, gal/mcf
Percent Solids Recirculated
Effluent Residence Time, min.
Stoichiometric Ratio,
moles Ca/ moles SO^ absorbed
Average % Limestone Utilization,
lOOx moles SO2 absorbed/mole Ca
Inlet SO2 Concentration, ppm
Percent SO2 Removal
Scrubber Inlet pH Range
Scrubber Outlet pH Range
Percent Sulfur Oxidized
Solids Disposal System
Loop Closure, % Solids Diech.
Clear Liquor to Demister, gpm
Make-Up Water to Demister, gpm
EMsBolved Solids, ppm
Total AP Range, in. H2O*b'
Demister AP Range, in. H2O
Demister Condition
at End of Run (c) («)
Venturi and Spray Tower
Condition* at End of Run
501-1A
Reliability verifi-
pH with Chevron
316 S. S. demister.
4/9/73
5/9/73
645
20,000
5
600/600
•40
15-16
20
1.4-1,6(4/9-4/27)
1.9-2, l(4/Z7-5/9)(dl
67 (4/9-4/27)
50(4/27-5/9) (d)
2,400-3, 300
70-75
5. 8-6.0
5.4-5. 7
5-25
Cla rifle r
20-28
14-20
12-14
5,000-8.000
10-11
0.40-0. 65
Scattered 1/8"
scale on top and
20 mil scale on
bottom.
Thin scattered
pcale and eroded
guide van t bolts in
venturi section.
Scattered 15-25
mil scale on after-
scrubber walls
above trapout tray.
9 of 28 slurry
ST48FCN spray
nozzles in after-
scrubber were
plugged.
50Z-1A
Same as SOl-lA
plaatlc demister.
6/13/73
6/26/73
278
20, COO
5
600/600
40
14-lb
20
1. 5-1.9
59
2.200-1.000
67-73
5.7-5.9
5.3-5.5
15-35
Clarilier
21-39
14-20
12-14
-
10.0-10. 5
0. 40-0.55
1/16" scale on top.
Light scale and
eolldB between
vanes. Approxi-
mately 1/3 ft3
BOlldB buildup at
4 locations (junc-
tion of support
bars).
5 mil • cale on
walls below
plug and about
1/6 belts heads on
guidevanea in
venturi section
eroded. 10 mil
scale on after-
scrubber walla
below trapout tray.
S03-1A
Same as 502-1A
rate and lower
percent aolids
recirculated.
6/29/73
7/11/73
256
20.000
5
600/1200
•BO
8-9 .
20
1.4-1.6
67
2,200-3,000
74-82
5.6-5.8
5.1-5.3
20-40
Clarifier
27-41
27-33
10-12
7,300
10.4-11.6
0.45-0. 55
No scale or solids
on top. 1/2 to 1"
non-uniform scat-
tered Bolidfi de-
posit on top of
bottom vane and
on 2nd pass of
vane. Center
portions clean.
20 mil scale on
walla below plug
and noticeable
erosion of guide-
vane croaa braces
in venturi section.
1 5 mil ecale on
afte r- ac rubber
walls below trap-
out tray. 3 of 28
alurry spray noz-
zles plugged.
506-1A
Same aa 503-1A
rate and lower ef-
fluent residence
time. Top and
bottom demiater
«..h.(«l
7/25/73
8/13/73
417
30,000
7.5
600/1200
53
8-10.5
12
1.4-1.6 (1.6-1.8
from 8/3-6)<8>
67 (59 from
8/3 to 8/6](f|l8)
2.500-3, 300 (2,000-
2,200 on 7/26 Ci 28)
68-76
5.35-5.65
4.85-5.1
20-35
Clarifier and
centrifuge
55-65
40-55
8-11
14,900 1"
12. 5-14. 5
1.0-1.25 (hl
Some plastic vanea
damaged by chunks
of solids broken off
from outlet duct.
5 mil solids buildup
on westside vanea,
Moderate solids
buildup on dead
• pota on aupport I
beams k adjacent
vanes.
1 5 mil scale on walls
of flooded elbow It
below plug. Cont'd
erosion of guldevane
croia braces. 20
mil light scattered
scale on after-
scrubber walls b
spray headers. No
slgnlf. sollda accum.
above k below trap-
out tray. Severe
reheater outlet duct.
(a) Includes line-out.
(b) Spray tower and venturi, excluding demister.
(c| Chevron plastic four-pass open demister in all runs
except 501-LA -where Chevron 316S.S. three-pass
open demister wa§ used.
(d) Reactivity of limestone decreased after 4/27 due to
(e) Demister wan washed from bottom only for all runt)
except Run 50o-lA where it was washed from both
top and bottom.
(f) The stoichiometry range was higher at 1. 6-1. 8
(59% average limestone utilization) from 8/3 to 8/6
with correspondingly higher scrubber inlet pH range
of 5.5 to 5.65.
(g) As of 7/27, a new limestone containing approx.
1.25 mole % MgCO3 was uied. Prior to this time
a limestone having approx. 5 mole % MgCOj had
been used.
(h) Range given is for period 7/25-8/8. Increased
steadily from 1. 25 to 1. 40 in. H2O ftitar 8/8.
(i) Increasing steadily during run from 8, 000 to 14, 900
ppm.
3-4
-------
Table 3-1 (Continued)
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: VENTURI SYSTEM
Run No.
Test Objectivea
Start-of-Run Date
End-of-Run Date
On Stream Houra(a)
Gas Raie, acfm @ 330° F
Spray Tower Gas VeL.fps @125°F
Venturi/Spray Tower
Liquor Rates, gpm
Spray Tower L/C, gal/mcf
Percent Solids Recirculated
Effluent Residence Time, min.
moles Ca/moles SO£ absorbed
Average % Limeslonc Utilization,
lOOx moles SO2 absorbed/mole Ca
Inlet SO, Concentration, ppm
Percent SO2 Removal
Scrubber Inlet pH Range
Scrubber Outlet pH Range
Percent Sulfur Oxidized
Solids Disposal System
Loop Closure, % Solids Disch.
Clear Liquor to Demiater, gpm
Make-Up Water to Demister, gpm
Dissolved Solids, ppm
Total AP Range, in. H2O(b)
Demister AP Range, in. r^O
Demioter Condition
at End of Run
77 (63 from
8/31 to 9/7)
Clarifier and
centrifuge(e)
53-68
-------
Table 3-2
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: TCA SYSTEM
Run No.
Test Objectives
Start-of-Run Date
End-of-Run Date
On Stream Hours *a>
Gas Rate, acfm @ 300°F
Gas Velocity, fps @ 125°F
Liquor Rate, gpm
L/C, gal/mcf
Percent Solida Recirculated
Effluent Residence Time, min.
Stoichiometric Ratio,
moles Ca/molea SO? absorbed
Average % Limestone Utilization,
lOOx moles SO^ absorbed/mole Ca
Inlet SO2 Concentration, ppm
Percent SO- Removal
Scrubber Inlet pH Range
Scrubber Outlet pH Range
Percent Sulfur Oxidized
Solida Disposal System
Loop Closure, % Solids Disch.
Clear Liquor to Koch Tray, gpm
Make-Up Water to Koch Tray, gpm
Dissolved Solids, ppm
Total AP Range, in. H2O
Demister and Koch Tray
AP Range, in. H2O
Demister Condition
at End of Run
Bed Condition
at End of Run
Inlet Duct Condition
at End of Run
Other Problems
or Comments
501- 2A
Reliability verifi-
cation test @ low
PH.
3/22/73
4/23/73
580
20, 000
7.6
1EOO
80
14-16
20
1. 1-1.4
80
2,200-3. 300
80-86
5. &-&. 0
5. 1-5.7
ZO-4O
Clarifier
25-48
20-30
7-12
4, 000-10,000
4. 5-5. I
1. 6-2. 0
1 /16" scale on
jottom vanes
only.
Spheres in middle
>ed fell down to
2 holes in bottom
grid. Replaced 3
dozen collapsed
spheres.
Slight solids
>uilt-up up-
Btream and heavy
olids buildup
[ownstream of
cooling nozzles.
1 intermediate
shutdowns due to
cooling nozzle
and Ventri-Rod
>luggage. Ven-
ri-Rod pluggage
ranged from 12-
70%. Replaced
Ventri-Rod with
4 ST24FCN SS
iete nozzles on
4/16/73.
502-2A
Same ai 501-2A
with high ftoichi-
ornetry and pH.
4/27/73
5/21/73
557
20,000
7.8
1200
SO
14-16
20
1. 5-1.9 (2. 1-2. 7
from 5/5-12l(cl
59 (42 from
5/5-12)
2,300-3, 300
90-97
5. 9-6. 1
5.4-5.7
17-35
Clarifier
31-3<)
5-20
8-12
4,300-8.800
5. 5-6. 5<">
1.9-2.2
1/16" scale on
vanes. About S07o
of west quadrant
plugged.
No grid failure
but significant
grid wire con-
tinued.
About 60% of duct
a rea plugged.
Solids buildup at
and upstream of
nozzles.
4 intermediate
shutdowns due to
cooling nozzles
pluggage. Used
Spraco 7LB 316
5/5/73.
508-2A
Replicate of
501 -2 A.
5/25/73
5/29/73
98
20,000
7.8
1200
80
15-16
20
1. 2-1.4
77
2. 300- J, 000
82-87
5. 6
5.0-5. 2
20-35
Clarifier
31-43
5-18
7-12
-
5.0-6.0
1. 9-2. 3
Clean.
Replaced about
5% of collapsed
placed 3 damaged
grid sections.
Solids buildup up-
stream of cooling
nozzles.
509-2A
Same as 501 -2A
with lower per-
cent Bolids recirc.
6/5/73
6/25/73
465
20,000
7.8
1200
80
8-9
20
1. 2-1.45
75
2,000-3, 100
80-88
5.5-5. 8
5.0-5.4
32-50
Clarifier
26-45
20-45
7-11
8, 300-11,800'
5.0-5.8
1. 9-2.0
1/16" scale on
west quadrant.
Replaced the da-
maged SW section
placed about 20%
of collapsed
spheres.
Clean.
Modified soot-
blowe r head on
5/29/73 to have
2 jets blowing air
forwa rd only.
Capped bottom
cooling spray
nozzle.
(a) Includes line-out.
(b) Total, excluding demieter and Koch Tray.
(c) Reactivity of limestone decreased during the period
of higher stoichiometry range and lower average lime-
stone utilization, due to larger average limestone
particle aiz«.
(d) Range given is for period before 5/15. Increased
gradually from 6 to 9 in. H2O from 5/J5 to 5/21.
3-6
-------
Table 3-2 (Continued)
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: TCA SYSTEM
Run No.
Te»t Objective*
Start-of-Run Date
End-of-Run Date
On Stream Hours**'
Gaa Rate, acfm @ 300°F
Gas Velocity, fpa @ 125°F
Liquor Rate, gpm
L/C, gal/mcf
Percent Solids Recirculated
Effluent Residence Time, mln.
Stolchiometric Ratio,
molea Ca/moles SO2 absorbed
Average % Limestone Utilization,
lOOx moles SO, absorbed/mole Ca
Inlet SO2 Concentration, ppm
Percent SO2 Removal
Scrubber Inlet pH Range
Scrubber Outlet pH Range
Percent Sulfur Oxidized
Solids Disposal Syotem
LoopCloaure, % Solidi Diach.
Clear Liquor to Koch Tray, gpm
Mike -Up Water to Koch Tray, gpm
Diasolved Solids, ppm
Total AP Range, in. H2O(b)
Demister and Koch Tray
AP Range, in. H2O
Demister Condition
at End of Run
Bed Condition
at End of Run
Inlet Duct Condition
at End of Run
Other Problems
or Comments
510-2A
Same as 509-2A
with higher gae
6/27/7J
7/10/73
297
25.000
9. 8
1200
64
7. 5-9. 5
20
1.2-1.5
74
2,000-2.700
78-89
5.4-5.5
4.85-5.2
35-55
Clarifier
26-44
25-45
9-13
8.800-11,300
6.6-7.6
2.2-2.5
60 mil scale on
moat of demister.
partially plugged
with 1/2" solid.
Between bottom
vanea.
Several loose grid
wire.. No grid
replacement
needed.
Clean.
5I4-2A
Same as 510-2A
with lower effluent
d hi h
eolids recirculated.
7/22/73
8/13/73
493
25,000
9.8
1200
64
13.5-16
4. 4
1.25-1. 55(7/22-8/5
1. 15-1.3I8/6-13)1''
71 (7/22-8/5)
82 (8/6-l3)(»'
2, 100-3. 100(1,800-
2. 100. 7/24. 264 28
77-89
5.2-5.55
4. 7-5.0
20-50
Clarifier
27-52
5-26
9-13
9,800-11,400
7.0-8. 5
2.2-2.5(2.5-2.8
from 8/ 10 to 8/13)
20% area plugged
with .olid.. NW
ged) with hard
cryst. solids.
Solid* buildup
mainly on bottom
vanea. Pitting
continued.
14% damaged
sphere o In lower 2
beda. Light scale
on 75% of bottom
grid.
7 ft3 a olid a build-
up upstream of
nozzlea.
AB of 7/27 a new
limestone having —
1.25 mole %
MgCO3 was used.
Prior to this time,
limestone contain-
ing — 5 mole %
MgCO3 was used.
515-2A
Same as 51 4-2A
with lowe r
8/16/73
9/10/73
571
25.000
9.8
1200
64
7. 0-8. 5
4.4
1.2-1.4(1.4-1.55
from 8/31 to 9/3llcl
77 (68 from
8/31 to 9/3)(cl
2.400-3,300(1,450-
1,700. 8/31-9/3)
80-88
5.2-5.4
4.9-5. 1
25-50""
Clarifier"1
34-44<"
36-50
10-12
11,000-13, 300'fl
7. 0-8. 5
2.2-2.5(2.5-2.9
from 9/6 to 9/10]
407oarea plugged.
Solids buildup
yo .„
solids buildup on
underaide of Koch
tray. 6 tray valves
partially plugged.
Broken arid
wire.. In) 2-3" «ta-
l&ctite on btm. grid.
1/32-3/16" scale on
wall..
Clean.
Underside of Koch
tray not properly
washed due to loose
steam sparger
jranch headerfl.
(a) Includes line-out.
(b) Total, excluding demister and Koch tray.
(c) High stoichiometry range of 1.4-1. 55 (68% average
limestone utilization) caused by excess limestone in
scrubber ilurry during the p«riod of low inlet SO^
concentration (av. 1600 ppm) from 8/31 through 9/3.
(d) 48-80% oxidation from 8/31 through 9/3 when average
inlet SO2 concentration dropped to 1600 ppm.
(e) With Clarifier and centrifuge before 1235
hours on 8/20. Percent aolids discharged
from centrifuge was 58-59%.
(f) No analytical data before 8/20, when Clarifier
and centrifuge were used.
(g) Indicates effect of utilization of new limestone
(see Other Problems or Comments).
(h) About 6 in2 of support grid for top bed broken.
Some of top bed spheres dropped into middle bed.
3-7
-------
Table 3-3
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: MARBLE-BED SYSTEM
Run No.
Telt Objectives
Start-of-Run Date
End-of-Run Date
On Stream Mourn1'1
Gat Rate, acfm @ 330°F
Ga. Velocity, fps @ 125°F
Liquor Rates to Top/Bottom
Sprays, gpm
L/C. gal/mcf
Percent Solids Reclrculated
Effluent Residence Time, min.
Stoichiometric Ratio,
molea Ca/molea SO^ absorbed
Average % Limestone Utilization,
lOOx moles SO2 absorbed/mole Ca
Inlet SO2 Concentration, ppm
Percent SO2 Removal
Scrubber Inlet pH Range
Scrubber Outlet pH Range
(Weir/Downcomer)
Percent Sulfur Oxidized
Solid* Disposal System
Loop Closure, % Solids Diech.
Clear Liquor to Demister, gpm
Make-Up Water to Demister, gpm
Dissolved Solida, ppm
Total AP Range, in. HjO""
Demister AF Range, in. HO
Demister Condition
at End of Run
Bed Condition
at End of Run
Inlet Duct Condition
at End of Run
Other Problems
or Comments
50I-3A S, 3B
Reliability verifi-
cation test @ Low
pH.
3/14/73
4/33/73
771
20. 000
5. 1
200/600
53
10-12
30
1.15-1.45(3/14-4/7
1.3-1.6 (4/8-Z3)(8)
77 (3/14-4/7)
69 (4/8-23)osit bottom aide.
30% of bed either
>lugged or ma r-
blea in stratified
>attern.
2 ft3 of solids de-
tosit between
spray header and
scrubber.
Both soot blower
airjets projecting
orward. Several
shutdowns due to
>lugged cooling and
>ottom apray noz-
zles. Swirl vanea
n 13 of 16 CE bot-
om bed spray
nozzles eroded
away.
502- 3A
Same as 501-3A I
3B with high stoi-
chiometry &i pH.
4/25/73
5/7/73
285
20.000
5. 1
200/600
53
11-14
30
1.5-2.1 (4/25-291
1. 9-2.7(4/2?-5/7><
2,600-3,200
67-77
5.8-6. 1
S.4-5.7/
10-30
Cla rifie r
19-25
13-23
7-20
3,700-8,000
7.5-11.0
0. 16-0. 35
1/4" slurry scale
and some scale on
mister.
607i of marbles
stratified. 1 ft2
area was plugged
with solida.
6 ft3 of solids de-
posit blocking 60-
70*^ of duct between
leader and
scrubber.
4 ST20FCN cooling
placed by ST24FCN
nozzles at start of
run. Swirl vanes
in all 16 CE bottom
led nozzles dis-
appeared. Swirl
vanes in all 6 CE
top bed nozzles
lightly eroded.
50J-3A 81 3B
Same as 50I-3A ii
3B with lower per-
cent solids recirc.
5/11/73
5/22/73
267
20.000
5. 1
200/600
53
8-11
30
1. 8-2.4
48
2. 500-3, 500
67-71
5.7-5.9
S.4-5.7/
5. 1-5. 3
15-30
^larlf. &t centrifuge
57-60 (used venturi
clarifier 5/16-17)
43-53
3-5
11.000(cl
8. 0-9. 3
0.15-0.22
1/8" sclide de-
>oait on top side.
side of demister.
25% of bed waa
>lugged with solids
and had stratified
rows of marbles.
-1/2 ft3 of solids
[eposit between
leader and scrub-
>er.
2 cooling spray
plugged. All CE
>ottom bed nozzles
operated without
awirl vanea.
Swirl vanes in CE
op bed nozzles
still intact.
504-3A fc 505-3A
Same as 501 -3A &
3B with Spraco bed
spray nozzles.
5/25/73
6/4/73
233
20. 000
5. 1
200/600
55
10-12
30
1.5-2.0
51
2, 300-J. 100
67-72
5.6-5.8
5. 3/
5.2-5.4
10-JO
Centrifuge''1
60-65 (23-24
from 5/30-31|("'
20-30
3-4
!0.500
-------
Table 3-3 (Continued)
SUMMARY OF LIMESTONE RELIABILITY
VERIFICATION TESTS: MARBLE-BED SYSTEM
Teat Objectives
Start-of-Run Date
End-of-Run Date
On Stream Hours
Gas Rate, acfm @ 330°F
Cas Velocity, fps @ 125°F
Liquor Rates to Top/Botton
Sprays, gpm
L/C, gal/mcf
Percent Solida Recirculated
Effluent Residence Time, min.
Stolchiometric Ratio,
mole
Average % Limestone Utilization,
Inlet SO- Concentration, ppn
Scrubber Inlet pH Range
Scrubber Outlet pH Range
(Weir/Downcomer)
Percent Sulfur Oxidized
Solids Disposal System
Loop Closure, % Solids Disch.
Clear Liquor to Demiater, gpm
Make-Up Water to Demiater, gpn
Dissolved Solids, ppm
Total AP Range, in.
Demleter AP Range, in.
Demister Condition
at End of Run
Bed Condition
at End of Run
Inlet Duct Condition
at End of Run
Other Problems
or Comments
Same as 505-3A
with bed under spray
higher.
6/15/73
20,000
1.6-2. 1
2,100-3,200
5. 5-5. 6/
5.0-5.2
20-40
Centrifuge
60-66
15.500-20.000
7.8-9.0
0.17-0.20(c
1/3 of the demister
(particularly SE
corner) completely
plugged.
10% of bed plugged.
15% of marbles
stratified.
Extremely light
solids accumulation
at duct walls down-
stream of nozzles.
One shutdown due to
reheater pilot mal-
function. Ono shut-
down due to Instru-
ment air compres-
sor valve leakage.
Extended cooling
apray nozzle* to
reach 2" from
scrubber Inlet.
(a) Includes line-out.
(b) Total, excluding demister.
(c) Range given IB for period before 6/23. Increased
steadily from 0. 20 to 0. 80 in. H2O from 6/23 to 7/2.
3-9
-------
Table 3-4
LIMESTONE RELIABILITY VERIFICATION TEST
RUN EVALUATIONS: VENTURI AND SPRAY TOWER
PARAMETER
Demister Scaling or
Plugging
Venturi and Spray Tower
Mechanical Condition at
End of Run
Venturi Scaling or
Plugging
Spray Tower Scaling or
Plugging
TEST RUN
501-1A (645 Operating Hours)
Comments
Light scale on bottom
vanes. Scattered scale on
top vanes.
Negligible solids deposits.
bolts and surrounding area
in venturi.
Scattered light scale on
walls belo-w plufi. Moder-
ate scale on walla of flood-
ed elbow.
Negligible solids deposits.
Scattered light scale on
walls.
Scattered moderate solids
er, on bottom demister
wash header, and on bottorr
of trapout tray.
Relative
Condition
at
End-of-Run
Fair
Good
Good
50Z-1A (276 Operating Hours)
Comments
Moderate scale on top
vanes.
Scattered light solids de-
posits on bottom vanes at
support bar junctions.
Moderate erosion of guide
vanes, bolts and cross
braces in venturi.
below plug. Moderate
scale on walls of flooried
elbow.
Negligible solids deposits.
Light scale on walls below
trapout tray.
Scattered light solids de-
posits on top slurry header,
on bottom demister wash
header, and on bottom of
trapout tray.
Relative
Condition
at
EnJ-of-Run
Fair
Fair
jood
Good
503-1A (256 Operating Hours)
Comments
No scale.
Non -uniform, light, scat-
tered solids deposits on
Slight e"rosion of guide vane
cross braces in venturi.
Light scale on walls below
plug. Moderate scale on
walla of flooded elbow.
Negligible solids deposits.
Light scale on walls below
trapout tray.
Light solids deposits on
bottom of trapout tray.
Relative
Condition
at
End-of-Run
Good
Fair
Good
Good
506-1A (417 Operating Hours)
Comments
No scale. Moderate solids
deposits around support I
beams. About 1 0% of plas-
tic top vanes damaged by
solids dislodged from stack
(Solids due to high entrain-
ment of demister top
flush).
Slight erosion in guide vant
area of venturi.
Slight erosion of top splash
seal flangp in venturi.
Light scale on walls below
plug and on walls of
flooded elbow.
Negligible solids deposits.
Light scale on all four
spray headers and on about
5O% of wall area above
trapout tray.
Light solids on bottom of
trapout tray and adjacent
walls.
Relative
Condition
at
End-of-Run
Poor
Fair
Good
Fair
I
H-«
O
-------
Table 3-4 (Continued)
LIMESTONE RELIABILITY VERIFICATION TEST
RUN EVALUATIONS: VENTURI AND SPRAY TOWER
Demister Scaling or
Plugging
Venturi and Spray Tower
Mechanical Condition at
End of Run
Venturi Scaling or
Plugging
Spray. Tower Scaling or
Plugging
TEST RUN
507-1A (434 Operating Hours)
Comments
No scale.
Negligible solids deposits
on vanes. Scattered par-
tial plugging of about B% of
top vane flow area.
Slight erosion of guide
vane assembly.
No scale.
Moderate solids buildup at
wet-dry interface due to
discontinued soot blowing.
Negligible scale and solids
deposits.
Relative
Condition
at
End-of-Run
Good
Fair
Good
Good
508-1A (Z8 Operating Hour si
Comments
No scale.
Negligible solids deposits.
High entraihment of
demister top flush.
—
Relative
Condition
L at
End-of-Run
Poor
—
LO
I
-------
Table 3-5
LIMESTONE RELIABILITY VERIFICATION
TEST RUN EVALUATIONS: TCA
PARAMETER
Demieter and Koch Tray
Scaling or Plugging
Scrubber Mechanical ^
Condition at End of Run
Scrubber Scaling or
Plugging
Inlet Duct Plugging
| TEST RUN
501 -2A (580 Operating Hours)
Comments
Negligible scale and solids
deposits.
Spheres from middle bed
dropped to bottom bed
through eroded grid wires.
Negligible acale.
Scattered solids deposits
on walls below Koch tray.
Slight solids buildup up-
stream and heavy solids
buildup downstream of
cooling sprays.
Relative
Condition
at
End -of -Run
Good
Bad
Good
Bad
502-2A (553 Operating Hours)
Comments
Light scale on vanea.
About 1 5% of bottom vane
flow area plugged by solids.
Significant erosion of grid
•wires.
Negligible scale.
Some Bolida deposits on
slurry nozzles only.
About 60% of duct area
plugged immediately up-
stream of cooling sprays.
Relative
Condition
at
End-of-Run
Poor
Bad
Good
Bad
508-2A (98 Operating Hours)
Comments
Negligible scale and solids
deposits.
Loose, bent and eroded
wires in two grids.
Negligible scale.
Scattered moderate solida
deposits on walls immedi-
ately below Koch tray.
Moderate solids buildup
upstream of cooling
sprays.
Relative
Condition
at
End-of-Run
Good
Bad
Fair
Poor
509-2A (465 Operating Hours)
Comments
Moderate scale on bottom
two rows of vanes.
Negligible solids deposits.
Broken and eroded wires
in several grids.
Moderate scale on walls
below bottom bed, light
scale on walls of bottom
two beds.
Negligible solids deposits.
Clean
Relative
Condition
at
End-of-Run
Good
Bad
Fair
Excellent
w
I
No attempt was made to modify run conditions in order to solve the continuing problem of
support grid erosion. The wire meeh grids (0. 148 inch diameter wires) were replaced with
sturdier 3/8 inch diameter rods (1 -1/4 inch on center) prior to the long-term reliability test.
-------
Table 3-5 (Continued)
LIMESTONE RELIABILITY VERIFICATION
TEST RUN EVALUATIONS: TCA
PARAMETER
Demister and Koch Tray
Scaling or Plugging
Scrubber Mechanical *
Condition at End of Run
Scrubber Scaling or
Plugging
Inlet Duct Plugging
TEST RUN
510-2A
-------
Table 3-6
LIMESTONE RELIABILITY VERIFICATION TEST
RUN EVALUATIONS: MARBLE-BED ABSORBER
PARAMETER
Dem later Scaling or
Plugging
Scrubber Mechanical
Condition at End of Run
Scrubber Scaling or
Plugging
Inlet Duct Plugging
TEST RUN
501 -3A & 3B (771 Operating Hours)
Comment
Light scale on all vanee.
Intermittent, moderate
solids deposits on bottom
vanes.
Swirl vanes in 80% of bot-
tom slurry nozzles com-
pletely eroded. Nozzle
plugging by marbles drop-
ped through loose grid.
Light scale on -walla below
bed.
Intermittent heavy solids
deposits on bottom spray
beaders fc bottom of bed.
About 30% of bed plugged
or in stratified pattern.
About Z5% of duct plugged
downstream of cooling
sprays.
Relative
Condition
at
End-of-Run
Fair
Bad
Bad
Bad
502-3A (285 Operating Hours)
Comment
Moderate scale and solids
deposits on top vanes.
Light scale on bottom
vane a.
All bottom slurry nozzles
without swirl vanes.
Light scale on v^alls above
and below bed and on all
spray headers.
About 2% of bed plugged,
60% stratified.
About 70% of duct plugged
immediately downstream
of cooling sprays.
Relative
Condition
at
End -of -Run
Fair
Bad
Poor
Bad
503-3A &' 3B (267 Operating Hours)
Comment
Mo scale.
Light solids deposits on
top vanes.
About 20% of bottom
slurry nozzles plugged
with debris.
Light, intermittent scale -
solids deposits on headers
and walls above bed. Mod-
erate scale-solids deposits
on bottom headers.
About 25% of bed plugged.
About 20% of duct plugged
sprays.
Relative
Condition
at
End-of-Run
Good
Bad
Bad
Poor
504-3A I S05-3A (233 Operating Hours)
Comment
Moderate scale on top
vanes.
Light solids deposits on
bottom vanes .
About 35% of bottom
Spraco slurry nozzles
partially plugged with
marbles.
Moderate scale on walls
above demister and bed.
Scattered, light scale-
solids deposits on headers
below bed.
About 1 Z% of bed plugged,
60% stratified.
About 30% of duct plugged
downstream of cooling
sprays.
Relative
Condition
at
End-of-Run
Fai r
Poor
Bad
Rad
CO
I
-------
Table 3-6 (Continued)
LIMESTONE RELIABILITY VERIFICATION TEST
RUN EVALUATIONS: MARBLE-BED ABSORBER
LO
t—•
(Jl
PARAMETER
Demiuter Scaling or
Plugging
Scrubbing Mechanical
Condition at End of Run
Scrubber Scaling or
Plugging
Inlet Duct Plugging
TEST RUN
506-3B (380 Operating Hours)
Comments
Light scale on bottom
vanes. Moderate scale on
top vanes.
About 30% of middle vanes
completely plugged, 30%
partially plugged.
About 80% of bottom Spraco
slurry nozzles severely
eroded
Moderate scale on walls
throughout scrubber. Inter
mittent, heavy scale on all
headers and walls below
bottom headers.
About 10% of bed plugged,
15% stratified.
About 45% of duct plugged
downstream of cooling
sprays. No solids down-
stream of open nozzle.
Relative
Condition
at
End-of-Run
Bad
-------
good system condition at the end of 434 operating hours at an effluent
residence time of 12 minutes, a percent solids recirculated of approx-
imately 9 percent, a spray tower gas velocity of 7. 5 ft/sec and a
liquid-to-gas ratio of 53 gal/mcf. Effluent residence times below 12
minutes could not be obtained during the testing due to system
constraints.
The venturi/spray tower runs also showed that washing the demister
from the underside (upstream) can be effective in reducing the rate
of soft solids accumulations on the demister blades, at demister
superficial velocities at or below 7. 5 ft/sec (see Runs 503-1A and
507-1A in Tables 3-1 and 3-4). In addition, the runs showed that
washing the demisters from the topside (downstream) can cause
droplet entrainment -within the exiting high velocity flue gas and re-
sultant accumulation of solids on the walls of the outlet duct (see Runs
506-1A, 507-1A and 508-1A). Demister operability will be discussed
in greater detail in Section 4. 2.
The spray tower has used spiral tip Bete No. ST-48 FCN stainless
steel full cone nozzles for the duration of the reliability verification
tests. No significant erosion of these nozzles had been observed for
over 2000 hours of operation.
TCA System. As expected, the relative system condition at the end
of the initial TCA reliability verification test 501-2A was good, as far
as scrubber or demister scaling and plugging was concerned (see
Tables 3-2 and 3-5). There -was, however, difficulty experienced due
to mechanical failure of a TCA support grid and pluggage of the inlet
3-16
-------
duct at the hot gas-liquid interface. During subsequent tests (see Runs
509-2A through 515-2A), the problem of inlet duct plugging was appar-
ently solved, although there was still some accumulation of solids in
the inlet duct during Run 514-2A (the resolution of the inlet duct plugg-
ing problem will be discussed in Section 4. 3). The continual problem
of erosion of TCA support grid wires was not solved during the remainder
of the reliability verification tests. Sturdier support grids (parallel 3/8-
inch rods) have been installed in the TCA scrubber during subsequent
long-term testing.
The results of the subsequent testing, under more economically attrac-
tive operating conditions, showed relatively good system condition
after 465 on-stream hours with an effluent residence time of 20 min-
utes, a percent solids recirculated of approximately 9 percent, a gas
velocity of 7. 8 ft/sec and a liquid-to-gas ratio of 80 gal/mcf (see Run
509-2A). A subsequent test at a gas velocity of 9. 8 ft/sec and a liquid-
to-gas ratio of 64 gal/mcf (Run 510-2A) gave some indication of scale
buildup within the scrubber and partial pluggage of the demister. Two
final test runs at an effluent residence time of 4. 4 minutes (Runs
514-2A and 515-2A) gave indications of severe scale buildup within the
-i,
scrubber^ and demister, and severe solids buildup in the demister and
on the underside of the Koch tray. Effluent residence times between
4. 4 and 20 minutes could not be obtained during these tests due to sys-
tem constraints (the larger effluent hold tank was used for the 20 min-
ute tests and a smaller recirculation tank for the 4.4 minute tests).
The maximum sulfate supersaturation within the scrubber system
occurs at the scrubber effluent. Hence, scale formation is heaviest
on the bottommost grid of the TCA (see Section 6).
3-17
-------
The acceptable effluent residence time, therefore, is between 4.4 and
20 minutes, at a percent solids recirculation of approximately 9 per-
cent (with 40 percent of solids fly ash), a gas rate of approximately
8 ft/sec and a liquid-to-gas ratio of approximately 80 gal/mcf. Tests
at the EPA TCA pilot facility at Research Triangle Park (which will be
discussed in Section 6) have indicated that an effluent residence time of
10 minutes is satisfactory, for a liquid-to-gas ratio equal to or greater
than 65 gal/mcf and a percent solids recirculated of 10 percent, pro-
vided that an appreciable amount of fly ash is present (40 percent of
the solids are fly ash).
Marble-Bed System. The condition of the Marble-Bed scrubber at the
termination of the reliability verification tests was poor (see Tables
3-3 and 3-6), because of problems associated with erosion of the CE
bed nozzles, and the resultant insufficient wetting of the underneath of
the marble-bed and subsequent pluggage of the bed. The use of hollow-
cone Spraco nozzles also caused insufficient wetting of the bed and sub-
sequent pluggage (see Runs 504-3A, 505-3A, and 506-3B). Future lime
reliability verification tests will be conducted with the Marble-Bed
scrubber with new, improved CE full-cone nozzles.
3. 1. 2 Analytical Data
A complete summary of scrubber inlet liquor analytical data for the
limestone reliability verification tests is presented in Table 3-7. Ex-
cept where noted, most of the dissolved species appear to have
approached steady-state concentrations for these runs. The liquid ana-
lytical data are tested by inputting the measured compositions and pH's
into a modified Radian Equilibrium Computer Program (Reference 4),
3-18
-------
Table 3-7
AVERAGE SCRUBBER INLET LIQUOR COMPOSITIONS
FOR LIMESTONE RELIABILITY VERIFICATION RUNS
Run No.
Venturi
501-1 A
506-1A
507-lA(b)
TCA
501 -ZA
502-2A
509-2A
510-2A
514-ZA
515-ZA
Marble-Bed
Percent
Solids
Recirculated
16
8
8
16
16
8
8
16
. 8
S01-3A8.3B 11
50Z-3A
505-3A(c)
506-3B
11
11
11
Effluent
Residence
Time, min.
20
12
1Z
20
20
ZO
ZO
4.4
4.4
30
30
30
30
Percent
Solids
Discharged
20-Z8
55-65
53-68
ZS-48
31-39
Z6-45
26-44
27-52
34-44
20-29
19-25
60-65
60-66
Percent
Sulfur
Oxidized
5-25
20-35
25-45
20-40
17-35
32-50
35-55
20-50
25-50
15-35
10-30
10-30
20-40
Sc rubbe r
Inlet pH
Range
5.8-6.0
5.4-5.7
5.3-5.7
5.7-6.0
5.8-6.1
5.5-5.8
5.4r5.5
5.2-5.6
5.2-5.4
5.8-6.0
5.8-6.1
5.6-5.8
5. 5-5.8
Inlet
Ca++ | Mg"
2000 220
4200(a) 480(a)
5100 510
1800 300
1600 200
2700 350
3000 400
3000 350
3300 310
2200 200
1500 120
2800 300
5000 450
Liquor Species Concentrations, mg/1
Na+
50
140
140
50
100
100
100
100
120
50
50
150
150
K+ | «;
200
85(a> 200
180 120
100
40 150
70 220
70 180
80 190
120 210
250
200
50 200
120 200
so4=|co3=
1300 200
2100 40
2500 25
1600 100
1250 Z50
1900 250 .
2000 80
2000 70
2500 20
1500 100
1000 200
1400 250
1BOO 250
(ppm)
Cl" 1 Total
3000 7000
7800 15,000
10,200 18,800
2600 6500
2200 5800
4800 10,400
5000 10,800
5000 10,800
5800 12,400
3500 7800
2200 5300
5500 10,600
9400 17,400
Calculated Degree of Saturation. %(d)
CaSO3-l/2H2O|e) CaSO4-2H2O(f)
83 96
68 167
43 204
35 109
61 87
73 138
47 149
46 152
46 193
105 113
80 74
74 106
88 150
(a) Concentration at end of run. Concentration increasing throughout run.
(b) Concentrations for first half of run are listed. Percent solids discharged
was 20-30% during second half (using clarifier only) and total dissolved
solids decreased gradually to 11,300 ppm.
(c) Only one liquor analysis taken.
(activity Ca ) x (activity anion) / (solubility product at 50 °C)
(e) Based on a solubility product for CaSO3' 1 /Z H2O of 4. 5 x 10~? which
was fit to previous pilot plant data (C. Y. Wen. private communication.
January 1973).
(f)
Based on a solubility product for CaSO4*2H2O of 2. 2 x 10~5(Radian
Corporation, "A Theoretical Description of the Limestone-Injection
Wet Scrubbing ProceflB." NAPCA Report. June 9, 1970).
-------
which then calculates the ionic imbalance. For the data shown in
Table 3-7, the calculated ionic imbalances were all less than 13
percent.
The large concentrations of chloride ions are attributable to chlorides
present in the coal which were converted to HC1 and absorbed from
the flue gas in the scrubber. A. Saleem (Reference 5) of Ontario
Hydro has reported similar chloride concentrations during limestone
wet-scrubbing tests with flue gas from a coal-fired boiler.
Venturi Runs 506-1A and 507-1A and Marble-Bed Runs 503-3A
through 506-3B demonstrate the higher degree of closed-loop opera-
tion that is achieved by use of the clarifier and centrifuge, rather
than the clarifier alone, to separate solids (53-68 percent solids dis-
charged vs. 19-29 percent). Total concentration of dissolved species
was more than doubled for these runs.
The calculated values for the degree of liquor saturation with
CaSO3' 1/2 H2O and CaSO4« 2H2O presented in Table 3-7 were made
with the use of the modified Radian program. The calculated degrees
of sulfate saturations for the scrubber effluent, liquors (which have
not been presented in this report) are, of course, greater than the
predicted saturations for the scrubber inlet liquors (see Section 6).
The calculated degrees of sulfite saturation are subject to large error,
due to the uncertainty in the value of the solubility product and the
large experimental error associated with measuring the sulfite con-
centrations. It is likely that, at steady state, with solid CaSC>3' 1/2
H2O present, the liquid phase would be saturated with respect to
sulfite.
3-20
-------
The TCA data from Table 3-7 indicates an increase in the degree of
sulfate super saturation of the scrubber inlet liquor when effluent resi-
dence time is decreased from 20 minutes to 4.4 minutes (Run 515-2A
vs. 509-2A and 510-2A) and when the percent solids recirculated is
decreased from 16 to 8 percent at 4.4 minutes residence time (Run
515-2A vs. 514-2A). This result is consistent with the TCA reliabil-
ity verification test results discussed in Section 3. 1. 1. Sulfate scal-
ing of the bottommost TCA grid occurred when the effluent residence
time was decreased from 20 minutes to 4.4 minutes, and increased
in severity when the percent solids was decreased from 16 to 8 per-
cent at 4.4 minutes (see Table 3-2). From the data in Tables 3-2
and 3-7, it would appear that sulfate scaling is likely to occur in the
TCA scrubber for a degree of sulfate saturation of the scrubber inlet
liquor greater than approximately 150 percent, at a liquid-to-gas ratio
of 64 gal/mcf and a percent solids recirculated between 8 and 16 per-
cent (40 percent of solids is fly ash).
A small amount of analytical data has been analyzed for the wash liquor
to the spray tower and Marble-Bed demisters and the TCA Koch tray.
The results have shown that the inlet wash liquors, which are composed
of mixtures of clarified process liquor and available raw water makeup,
have approximately 60 percent of the degree of sulfate saturation of the
scrubber inlet liquor, at typical conditions. Even though the wash
liquor may be less than saturated when introduced, SC>2 absorption and
oxidation on the mist eliminator surfaces can cause supersaturation
and scaling.
3-21
-------
3. 1. 3 Material Balances
The results of material balances for calcium and sulfur for many of
the limestone reliability verification runs, during continuous (uninter-
rupted) on-stream operating periods, are given in Table 3-8, The
SC>2 absorbed was computed from the measured inlet gas rate, the in-
let and outlet gas SC>2 concentrations and the estimated gas outlet rate.
The calcium added was computed from the measured volumetric rate
of limestone slurry additive and the solids concentration in the slurry.
The sulfur and calcium discharged were computed from the measured
rate of slurry discharged from the system and the concentrations of
sulfur and calcium in the discharge.
The computed inlet and outlet rates for calcium and sulfur are in good
agreement. The average stoichiometric ratios, based on solids analy-
ses, are probably more accurate than the values based on limestone
addition rate and SO£ absorption, due to uncertainties in the measure-
ment of limestone slurry feed rate. The ionic imbalances for the bleed
stream solids analyses, from which the calcium and sulfur discharge
rates were calculated, averaged less than +5 percent (more cations
than anions).
The continuous operating periods were broken up into "computational
periods" of from 8 to 48 hours, and material balances made for each
computational period and the results summed. The computed inlet and
outlet rates for calcium and sulfur did not necessarily balance during
each computational peripd, due to the unsteady conditions which prevail
at any point in time (e.g., changing percent solids) and the resultant
3-22
-------
Table 3-8
SUMMARY OF MATERIAL BALANCES FOR SULFUR AND
CALCIUM FROM LIMESTONE RELIABILITY VERIFICATION TESTS
Run No.
Ventu-i
501-1A
502- 1A
TCA
501-2A
50Z-2A
509-2A
M-B
501-3A
501-3B(a|
506-3B
Material
Balance
Period,
hours
605
210
150
170
465
150
140
360
Sulfur Balance
S02
Absorbed,
Ib-moles/hr
4. 3
3.9
4.7
5.6
4.7
4.2
4. 3
3.8
SOX in Solids
Discharged,
Ib-moles/hr
4. 5
4.0
4.3
6.0
4. 1
4. 1
4.4
4.2
Percent
Error
- 5
- 4
•t 8
- 7
+ 14
+ 3
- 2
-10
Calcium Balance
Ca in L-S
Feed,
Ib-moles/hr
7. 1
6.2
4.5
11.6
4.9
4. 5 '
6.2
6.8
Ca in Solids
Discharged,
Ib-moles/hr
7.8
7.4
5.0
10.6
5. 7
5.2
6.2
7.4
Percent
Error
- 9
-13
-11
+ 9
-15
-13
0
- 8
Average Stoichiometric Ratio,
Moles Ca Added/Mole SOj Absorbed
Based on Lime-
stone Added
and SO2 Absorbed
1.67
1.59
0.95
2.07
1.03
1.06
1.45
1.78
Baaed on
Solids
Analysis
1.73
1.82
1.15
1.77
1.38
1.26
1.42
1. 77
I
to
UJ
(a) Because of turbid clarifier overflow, some of the solids in the clarifier feed ia returned to the scrubber. The values for SOx and Ca
discharged have been corrected for solids returned and are net discharge from the system.
(b)
The values for SOX and Ca discharged have been corrected for solids in the centrate returned to the scrubber.
-------
accumulation (or depletion) of the species within the system. How-
ever, over a longer period of time (>150 hours) the accumulation term
becomes negligible as compared to the total input or output of species.
3.1.4 SO2 Removal and Limestone Utilization
The results of the limestone short-term factorial tests (see References "
1 and 2), showed that SOo removal is a strong function of liquor rate,
inlet liquor pH and, of course, scrubber geometry (e.g. , number of
stages in the TCA). For the venturi, TCA and Marble-Bed scrubbers,
SO2 removal was not significantly affected by gas rate (gas velocity),
while for the spray tower, SO2 removal was slightly affected by gas
rate (increasing SO^ removal for increasing gas rate at constant liquor
rate). SC>2 removal is also a weak function of SC>2 inlet concentration
(higher removal for lower concentration) and scrubber temperature
(higher removal for lower temperature).v
The results from the EPA TCA pilot facility at Research Triangle Park
have shown that limestone utilization (100 x moles SC>2 absorbed/moles
CaCO3 added) is a strong function of limestone "reactivity" (i. e. , aver-
age particle size) and scrubber inlet liquor pH (see Section 6).
For the limestone reliability verification tests (see Tables 3-1, 3-2,
and 3-3) SO2 removals from 67 to 82 percent, 77 to 88 percent and 65
to 77 percent were obtained with the venturi/spray tower, TCA and
Marble-Bed scrubbers, respectively. Corresponding average lime-
stone utilizations of approximately 68 percent, 77 percent and 67
A 10 percent change in inlet SCU concentration or a 10°F change in
liquor temperature would correspond to about a one percent change
in SO2 removal.
3-24
-------
percent •were obtained for the three scrubber systems. Not included
in the above averages are runs in which there was an apparent decrease
in limestone reactivity (due to larger limestone average particle size)
and in which the effect of "high-pH" was being investigated (Runs 502-2A
and 502-3A).
An example of a decrease in limestone utilization and increase in SO
L*
removal due to an increase in inlet liquor pH can be seen by comparing
TCA Runs 501-2A and 502-2A (see Table 3-2). An increase in average
*
scrubber inlet pH from 5. 8 to 6. 0 resulted in a decrease of utilization
from approximately 80 to 60 percent and an increase in SO removal
from approximately 83 to 93 percent.
An example of changes in limetsone reactivity (and, hence, in limestone
utilization), due to changes in the average size of the limestone particles
(limestone "grindability"), can be seen in Runs 501-1A (Table 3-1),
502-2A (Table 3-2), and 502-3A (Table 3-3). In Section 6, the effect of
limestone particle size on limestone reactivity is discussed.
3.2 LONG-TERM RELIABILITY LIMESTONE TESTING
3.2.1 TCA Run 525-2A
The objective of the long-term test is to operate continuously for four
to six months. On October 24, 1973, a limestone long-term reliability
test (Runs 525-2A) was begun on the TCA system. Based on the re-
sults of the reliability verification tests and the tests conducted at the
EPA pilot facility in Research Triangle Park, the following conditions
The pH data from the Shawnee facility were not considered reliable
during the initial reliability verification runs.
3-25
-------
•were chosen for the long-term test:
Gas Rate, acfm @ 300°F 25, 000
Gas Velocity, ft/sec 9.8
Liquor Rate, gpm 1200
L/G, gal/mcf 64
Percent Solids Recirculated 15
Effluent Residence Time, min. 10
Total Pressure Drop, in. H^O 8. 5
Percent SO2 Removal (controlled) 80-85
Solids Disposal System Clarifier
Three beds of spheres were used, with five inches of spheres/bed.
The top bed used UOP supplied thermo-plastic-rubber (TPR) spheres
and the bottom two beds UOP supplied high density polyethylene
(HDPE) spheres. Also, the wire support grids in the scrubber
were replaced by sturdier bar-grids for the long-term run. A
summary of the operating data for Run 525-2A is as follows:
Average Percent Limestone Utilization 71
Inlet SO2 Concentration, ppm 1600-4000
Scrubber Inlet pH Range 5. 7-5. 6
Scrubber Outlet pH Range 5.3-5.6
Percent Solids Discharged 42
Dissolved Solids, ppm 8000
Predicted Percent Sulfate Saturation 120
After approximately 500 hours of operation the ,run was terminated,
due to (1) unusually heavy solids buildup on the underside of the
Koch Flexitray and on the scrubber walls between the top stage and
the Koch tray, and (2) scale and solids buildup on the bottom vanes
of the demister. Also, numerous (over 200) half-spheres of the TPR
spheres were found in the scrubber and slurry circulating system.
Half-spheres of TPR were also found lodged in the TCA inlet slurry
3-26
-------
spray nozzles. It should be noted that the scrubber stages (and
bottommost grid) were free of scale after the 500 hour operating
period, as was expected. The HDPE spheres had lost from 8- 14
percent of their original weight and the TPR spheres about 2. 6 percent.
It is hypothesized that the soft solids accumulated below the Koch tray
were due to the partial blockage of the TCA slurry inlet nozzles by the
TPR half-spheres, which produced high pressure drops across the
nozzles, resulting in excessive entrainment of fine slurry droplets.
The partially blocked nozzles may have also contributed to a large
degree to the mist eliminator pluggage, but excessive gas velocity
is a contributing factor.
3. 2. 2 TCA Run 526-2A
On November 21, 1973, a new TCA limestone, long-term reliability
test (Run 526-2A) was begun. The TPR spheres in the top bed had
*
been replaced with HDPE spheres , and the accumulated scale and
soft solids from Run 525-2A had been removed. The run conditions
were identical to those for Run 525-2A, excepting that the gas velocity
has been dropped to 8 ft/sec (20, 500 acfm). The velocity was reduced
because more detailed investigation of previous reliability verification
runs indicated that long-term reliability for the present Koch tray/
demister configuration should not be expected at a gas velocity of
9.8 ft/sec (compare Runs 509-2A and 510-2A in Tables 3-2 and 3-2).
A summary of the operating data for Run 526-2A is as follows:
\f
r
Once strainers are installed in the TCA system lines, it is planned
to replace all three stages of spheres with TPR spheres.
3-27
-------
Stoichiometric Ratio, moles Ca/mole 1.3-1.6
SO_ absorbed
Average Percent Liimestone Utilization 70
Inlet SO Concentration, ppm 1800-3400
Scrubber Inlet pH Range 5.6-5.9
Scrubber Outlet pH Range 5.2-5.5
Percent Solids Discharged 43
Dissolved Solids, ppm 8000
Predicted Percent Sulfate Saturation 120
On January 9, 1974, Run 526-2A was interrupted after 1190 hours
of on-stream operation, in order to check the wear rate of the HDPE
spheres in the three beds (HDPE sphere life had been estimated to be
less the 2000 hours). Pressure drop across the chevron mist eliminator
increased slightly during the initial 800 hours of operation, and during
the last 400 hours increased more rapidly to a final level about 1. 5
times the initial value of 0. 18 inches HO.
L*
A inspection was conducted on January 9, 1974, and the accumulation
of sulfate scale and solids within the scrubber system is shown in
Figure 3-1.
The general appearance of the scrubber was good. Scattered solids
deposits (up to 1 inch) covered the walls of the scrubber below the
bottommost bar-grid. Abour 1/16 inch thick scale was found on the
walls below the bottommost grid and on the wall areas not in contact
with the spheres. The 4 bar grids were covered with 10-14 mil scale.
Heavy solids deposit covered the inlet slurry spray nozzles and header
and adjacent walls between the elevation of the nozzle tips and Koch
•wash tray. A heavy relatively uniform (about 1 inch thick) solids
3-28
-------
DATE OF INSPEQION:
INSPEaOR /g g. TVt.tS
J>EAf/fT£g Sen. /OS
T»
, '/4'Tk. TOTAL-
SECTION 'C1
I.D.FAN DAMPERS
SPRAY
HEADER
A7~ SECTION 'B1
I. D. FAN
DAMPER •.
COOLING
SPRAY
NOZZLES
REHEATER
C
veer.
£#£# fe>**/f'
£/S T0 /"Tit
SEaiON 'A' -47-
TCA SCRUBBER
August 1973
Figure 3-1. TCA Inspection
-------
layer covered the underside of the -wash tray. Approximately 5 per-
cent of the tray valves were covered with solids. All four inlet slurry
spray nozzles were partially plugged with debris, primarily with
plastic covering from pipe insulation. Demister plugging was confined
to the bottom two passes only (1/4 inches thick solids), reducing the
free flow area by about 15 percent.
The flue gas outlet duct was covered with scattered (up to 1 inch thick)
solids, with approximately 20 percent of the surface clean to metal.
About 1/4 inch thick solids covered the underside (upstream) surfaces
of the ID fan damper louvres. The downstream sides of the ID fan
blades were covered with up to 30 mils of dry solids.
The average weight loss of the HDPE spheres from each bed was
as follows:
>!<
Hours in Use Percent Weight Loss
Top Bed 1190 28
Middle Bed 1707 40
Bottommost Bed 1707 23
As with Run 525-2A, it is hypothesized that the solids accumulations
on the inlet slurry spray nozzles and header and on the underside of
the Koch tray were primarily caused by partial blockage of the TCA
*
Some of the discrepancy in the weight loss results between the
middle and bottom beds could be attributed to the difference in
sphere material quality in the beds. The HDPE spheres have
been supplied by two different manufacturers.
3-30
-------
slurry inlet nozzles by debris, which produced high pressure drops
across the nozzles, resulting in excessive entrainment of fine slurry
droplets. Also, the blocked nozzles may have been a major factor
contributing to the mist eliminator pluggage observed, It should be
noted that the accumulation of sulfate scale on the bottommost bar
grid (10-14 mils) is not excessive, after the approximate 1700 hours
of operation without cleaning. This is not surprising, since the
scrubber inlet liquor was only 20 percent supersaturated with respect
to sulfate.
3. 3 PARTICULATE REMOVAL EFFICIENCIES
3. 3. 1 Equipment
For the overall particulate removal in the three scrubber systems
during the limestone factorial testing, a modified EPA particulate
train (manufactured by Aerotherm/Acurex Corporation) was used
to measure mass loading at the scrubber inlets and outlets.
During the limestone reliability verification testing, a special series
of tests using a Brink impactor were conducted by EPA to measure
the TCA inlet and outlet aerodynamic size distributions. In order to
utilize the Brink impactor at scrubber inlet mass loading conditions,
a modified EPA particulate mass sampling train was used. The train
is of 316 stainless steel construction and consists mainly of a heated
sample probe (6 feet x 1/2 inch outside diameter), a cyclone, and the
Brink impactor with a 144 mm glass fiber filter. The impactor draws
a sample from the gas stream exiting the cyclone. Previous work
3-31
-------
had established that the particulates collected in the cyclone had a
mass mean diameter of approximately 5 microns at a flow rate of one
cubic foot per minute. At the scrubber outlet, the Brink impactor
was used directly in the flue gas duct (without sample probe and
cyclone).
3. 3. 2 Overall Removal Efficiencies
The overall particulate removal efficiencies for the three scrubbers
obtained during the limestone short-term factorial testing (see Figure
2-4), are presented in Tables 3-9, 3-10, and 3-11. Only those data
which were taken at close-to-isokinetic sampling conditions have been
included in the tables. All of the outlet particulate data have been
-i*
fi*
corrected for s oot-contamination from the flue gas reheaters. The
soot amounted to less than 30 percent of the total mass of the outlet
particulates.
From Table 3-9, it is seen that overall particulate removal efficiencies
i'c;';
of 99. 4 to 99.8 percent were obtained for the Chemico venturi at
a gas flow rate of 30, 000 acfm (330 F) and liquid-to-gas ratios from
13 to 27 gal/mcf (300-600 gpm), with venturi plug pressure drops
from 6 to 12 inches HO. For the spray tower, the removal efficiency
£1
was about 98. 5 percent at a gas velocity of 4 ft/sec and a liquid-to-gas
ratio of 40 gal/mcf (15, 000 acfm and 450 gpm).
*
This problem of soot contamination from the oil-fired reheaters
has been temporarily solved (see Section 4.4).
For an average scrubber inlet grain loading of 3. 5 grains/scf, a
particulate removal of 99. 0 percent would correspond to 0. 07 Ibs
particulate discharged per 10° Btu.
3-32
-------
Table 3-9
OVERALL PARTICULATE REMOVAL IN VENTURI AND SPRAY TOWER SCRUBBER
DURING FACTORIAL TESTS
Run No.
415-1A
414-1D
414-1D
414-1C
417-1A
414-1E
418-1C
453-1B
454-1B
456-1A
Date
11-09-72
11-12-72
11-14-72
11-15-72
12-22-72
12-25-72
12-27-72
12-31-72
1-04-73
1-05-73
Gas
Rate,
acfm @ 330°F
30,000
30,000
29,900
29,900
30,000
30,000
14,900
14,900
14,900
14,900
Liquor Rate,
gpm
Venturi
305
305
305
305
605
300
600
12
12
12
Spray Tower
0
0
0
0
0
0
0
460
450
450
Pressure Drop,
in. H20
Venturi
9.0
9.0
9.0
6.4
9.5
12. 0
12. 5
2. 5
0. 75
0. 70
Spray Tower
2.0
1.9
1.9
1.9
!-9
1.9
0.4
0.45
0.45
0.45
Grain Loading,
grains/scf
Inlet
4. 38
2. 1
3. 32
3. 40
3. 38
4. 17
6.39
2.6
4. 62
3. 38
Outlet
0. 012
0. 010
0. 013
0. 02
0. 012
0. 009
0. 114
0. 004
0.07
0.056
Percent
Removal
99. 7
99:5
99.6
99.4
99. 6
99. 8
98. 2
99. 8
98. 5
98. 3
OJ
I
OJ
Including demister.
-------
Table 3-10
OVERALL PARTICULATE REMOVAL IN TCA SCRUBBER WITH FIVE GRIDS
AND NO SPHERES DURING FACTORIAL TESTS
CO
1
co
Run No.
Date
Gas
Rate,
acfm @ 330°F
Liquor
Rate,
gpm
Total
Pressure
Drop, in. H2O
Grain Loading,
grains /scf
Inlet
WC-5 12-21-72 19,200 730 3.8 1.70
WC-5A 1-06-73 19,300 730 4.7* 4.16
WC-5A 1-09-73 19,300 730 5.5* 1.32
WC-11 1-12-73 19,400 745 7.0* 3.29
WC-12 1-14-73 19,300 375 7.1* 3.65
Outlet
Percent
Removal
0.004 99.8
0.029 99.3
0.019 98.6
0.017 99.5
0.022 99.4
"High total pressure drop (including Koch tray, demister, and inlet duct) due to the pluggage
of inlet gas duct by solids deposit.
-------
Table 3-11
OVERALL PARTICULATE REMOVAL IN MARBLE-BED SCRUBBER
DURING FACTORIAL TESTS
Run No.
427-3A
4Z7-3A
426-3B
4Z7-3C
427-3B
428-3A
428-3A
428-3A
438-3A
440-3A
440-3A
Date
11-13-72
11-16-72
11-28-72
12-02-72
12-24-72
12-28-72
12-29-72
12-30-72
1-07-73
1-11-73
1-13-73
Gas Liqi
Rate, Rat
acfm @ 330°F gpj
20,000 81
20,000 81
20, 000 81
no r Total
e, Pressure
m Drop, in f^O
0 12.2
0 12.2
0 10.2
20,000 800 12.7
20,000 805 11.2
20,000 81
20,000 81
20, 000 81
0 11.7
0 11.7
0 11.7
19,900 400 7.2
12,500 600 6.9
12,500 600 6.9
Grain Loading,
grains /scf
Inlet
2.6
3.32
4.43
4.24
2.19
3.78
4.12
3.63
4.20
3.82
3.59
Outlet
0.030
0.035
0.032
0.033
0.027
0.025
0.016
0.035
0.020
0.042
0.066
Percent
Removal
98.8
98.9
99.3
99.2
98.8
99.3
99.6
99. 0
99.5
98.9
98.2
U)
Ul
-------
Table 3-10 shows that, for the TCA scrubber with 5 grids and no
spheres, the overall removal efficiencies were 98.6 to 99.8 percent
at a gas velocity of 7. 5 ft/sec and a liquid-to-gas ratio of 50 gal/mcf
(19, 300 acfm and 730 gpm), with total pressure drops (includes Koch
tray, demister and inlet duct) of 4 to 7 inches H?O.
The Marble-Bed scrubber (see Table 3-11) gave an overall particulate
removal efficiency range of 98. 8 to 99. 6 percent, at a gas velocity of
5 ft/sec and a liquid-to-gas ratio of 54 gal/mcf (20, 000 acfm and 810
gpm), with 12 inches f^O total pressure drop.
During the limestone reliability verification testing, a series of par-
ticulate removal tests with the TCA scrubber (3 stages, 5 inches of
spheres/stage) were conducted by EPA. Results from these tests
are presented in Table 3-12. The overall removal efficiencies of
98. 7 to 99. 9 percent were achieved at gas velocities from 8 to 10
ft/sec (20,000-25,000 acfm), liquid-to-gas ratios from 40 to 80 gal/
mcf (600-1200 gpm), and total pressure drops from 5. 5 to 10 inches
H^O. The higher pressure drops generally gave higher overall re-
moval efficiencies.
The overall particulate removal efficiencies show.n in Tables 3-9
through 3-12 appear to be higher than the efficiencies predicted from
the "impaction theory. " These improved efficiencies could be due to
(1) the condensation of water vapor in the flue gas on the solid particles'
and (2) solids accumulations upon the demisters or underneath the TCA
Koch tray during the duration of particulate testing. More testing is
planned to provide a better definition of the mechanisms.
The condensation of water vapor per unit mass of inlet solids has been
estimated to be from 1-5 grains water/grain inlet particulates within
the scrubber.
3-36
-------
Table 3-12
OVERALL PARTICULATE REMOVAL IN TCA SCRUBBER
DURING LIMESTONE RELIABILITY VERIFICATION TESTS
Gas
Run No. Date Rate,
acfm @ 300°!
503-2A 5/22-23 25,000
506-2A 5/24 20,000
505-2A 5/23 20,000
Liquor Pressure Grain Loading,
Rate, Drop, grains /scf
gpm in. H20 Inlet Qutlet
1200 9.8 3.16 0.00852
3.00 0.00375
1200 7.5 2.89 0.0143
2.13 0.0152
600 5.6 2.34 0.031
2.61 0.020
2.28 0.010
Percent
Removal
99.7
99.9
99.5
99.3
98.7
99.2
99.6
-------
3. 3. 3 Particulate Size Distribution in the TCA
For the runs listed in Table 3-12, the particle size distributions of the
particulates at the TCA inlet and outlet were also determined. The re-
sults are shown in Figure 3-2.
As shown in Figure 3-2, the mass mean diameter of the inlet solids is
approximately 23 microns, which is slightly greater than the "normal"
range of 10 to 20 microns. The data for the outlet size distribution
shows some scatter. The mass mean diameter ranges from about 0. 5
to 0. 75 micron, for a total pressure drop range of 5. 5-10. 0 inches
t^O. Generally, the higher pressure drops give smaller outlet mass
mean diameters.
3. 3. 4 Particulate Removal Efficiency in the TCA as a Function of
Particle Size
The particulate removal efficiency as a function of particle size was
determined by EPA for the TCA runs shown in Table 3-12. In Figure
3-3 the percent penetration (100-percent removal) is plotted vs. parti-
cle diameter in microns, for different ranges of total pressure drop.
From Figure 3- 3, it is seen that for the submicron particles (0. 11 to
0. 99 micron), the removal efficiency drops rapidly with decreasing
particle size, especially at low total pressure drop. The efficiencies
were 95 to 99 percent at 9. 8 inches E^O total pressure drop, 93 to
95 percent at 7.6 inches t^O, and 71 to 90 percent at 5.6 inches t^O.
Because of the limited number of tests, conclusions regarding sub-
micron collection efficiency should be reserved until additional testing
can be carried out.
3-38
-------
PARTICLE SIZE, microns
0.6 0.8 1
I 1 1
TOTAL PRESSURE DROP = 5.5-10.0 in. Hj
GAS VELOCITY = 7.8-9.8 ft/sec
LIQUID-TO-GAS RATIO = 40-80 9ol/mef
H 1 1
8 10 20
PARTICLE SIZE, microns
•+•
-f-
60 80 100
Figure 3-2. Particle Size Distributions
at TCA Inlet and Outlet
3-39
-------
i
z
UJ
O
LU
O.
8
ii
z
o
60
40
20 4-
10
8
6
4 4-
2 -•
5 i
£ 0.8
3 0.6
0.4 -•
0.2 -•
0.1
—D
5.5-5.7 in.
9.7-9.9 in. H2O
TOTAL PRESSURE DROP:
O
A
9.7-9.9 in. H2O
7.4-7.7 in. H20
D 5.5-5.7 in. H2O
GAS VELOCITY = 7.8-9.8 ft/sec
LIQUID-TO-GAS RATIO = 40-80 gal/mcf
0.04 0.06 0.1 0.2 0.4 0.6 1
PARTICLE DIAMETER, microns
Figure 3-3. TCA Particulate Removal Efficiency
as a Function of Particle Size
3-40
-------
Section 4
OPERATING EXPERIENCE AT THE SHAWNEE FACILITY
DURING LIMESTONE TESTING
In this section, the operating experience at the test facility during
the open liquor loop short-term factorial testing and the closed
liquor loop limestone reliability verification testing are summarized.
The results of a material evaluation program are also summarized.
Scaling and plugging tendencies on the scrubber internals have been
discussed, primarily, in Section 3.
4. 1 CLOSED LIQUOR LOOP OPERATION
For closed liquor loop operation, the raw water input to the system
is equal to the water normally exiting the system in the humidified
flue gas and the waste product. The original test facility design
included slurry pumps with water seals (Hydroseals) for bearing
protection, water quench sprays for gas cooling, -water sprays for
mist eliminator washing, a water wash for the Koch tray in the
TCA scrubber, and dilute limestone slurry feed (10-20 wt % lime-
stone). The water input under these conditions exceeded the makeup
requirement for closed liquor loop operation. The systems operated,
therefore, with partially open liquor loops during the limestone
short-term factorial tests, i.e. process liquor had to be discharged
4-1
-------
from, the systems, in order to maintain the overall water balances.
This was not considered to be a serious problem during factorial
testing for, at a specified scrubber inlet liquor pH, SC>2 removal is
not significantly affected by liquor composition. However, little
information was gained about the effect of scaling potential on re-
liability during this period.
The absorbent feed systems were changed in November 1972, to
provide slurry feeds with up to 60 wt % limestone concentration.
During the 5 week boiler outage in February and March 1973, the
Hydroseal slurry pumps were converted to a Centriseal type (mech-
anical seal supplemented with air purge); quench spray systems
using circulating slurry were provided for the TCA and Marble-
Bed scrubbers; and, the Koch tray wash system on the TCA
scrubber and the mist eliminator wash systems on the spray tower
and the Marble-Bed scrubber were converted to use clarified
liquor plus raw water makeup. Required revisions to bleed control,
flow measurements and control instrumentation were also made
during this period.
As a result of the modifications, closed liquor loop operation (based
on discharge of thickener underflow) has been maintained at the facility
since the beginning of limestone reliability verification testing in
March 1973.
4. 2 DEMISTERS
The specifications for the demisters tested on the three scrubber
systems are given in Table 4-1. The demisters are depicted, to
4-2
-------
Table 4-1
TEST FACILITY DEMISTER SPECIFICATIONS
Material of Construction
a
Design
Number of Vanes (Passes)
Total Depth of Demister
Center-to-Center Distance
Between Vanes
Included Vane Angle
Spray Tower
Stainless Steel
Chevron, open
3
7-11/16 in.
3-9/16 in.
100°
Polypropylene
Chevron, open
4
10 in.
3 in.
110°
TCA
Stainless Steel
Chevron, closed
6
14 in.
1-1/8 in.
120°
Marble-Bed
Stainless Steel
Chevron, closed
3
7-1/8 in.
3 in.
80°
Open-vanes not joined, closed-vanes joined.
-------
SPRAY TOWER
STAINLESS STEEL
SPRAY TOWER
POLYPROPYLENE
TCA
STAINLESS STEEL
MARBLE-BED ABSORBER
STAINLESS STEEL
GAS FLOW
GAS FLOW
GAS FLOW
GAS FLOW
6 in.
Figure 4-1. Test Facility Demister Configurations
-------
scale, in Figure 4-1. The spray tower polypropylene chevron demister
was used only during reliability verification Runs 502-1A, 503-1A
and 506-1A (see Table 3-1).
In order to remedy demister solids accumulation problems encoun-
tered during the early stages of factorial testing, the following
modifications were made to the systems:
(1) In November 1972, a Koch Flexitray wash tray was in-
stalled in the TCA scrubber between the inlet liquor spray
header and the chevron demister, and a steam sparger
was provided for cleaning the underside of the wash tray.
At first, irrigation was obtained with raw water. A sub-
sequent modification in February 1973 allowed for irri-
gation with process liquor, diluted with the available raw
water makeup.
(2) During the boiler outage in early 1974 the spray tower
and Marble-Bed demister systems were modified to allow
for washing from both the upstream (underside) and down-
stream directions with process liquor, diluted with the
available raw water makeup.
Wash-Tray Operation. The Koch Flexitray wash tray has been
successful, to date, in significantly reducing the solids accumulation
on the TCA demister blades at or below a scrubber superficial gas
velocity of 8 ft/sec. However, heavy solids buildup did occur on
the underside and below the Koch tray with intermittent steam
sparging for one minute per 8 hour shift. Subsequent to the 5 week
boiler outage, the steam sparging was increased to one minute per
hour, which substantially reduced the solids accumulation below the
tray. There was also an accumulation of solids underneath the Koch
tray during a recent long-term reliability test (see Section 3.2 for a
description of the problem).
4-5
-------
Wash Liquor. During the limestone reliability verification tests on
the three scrubber systems, the liquor -wash to the demisters (and
to the Koch tray) has varied from a ratio of about one part fresh
j^
water and six parts clarified liquor to half and half mixtures.
Occasionally, the undersides of the Marble-Bed and spray tower
demisters have been washed intermittently, on a cycle that has
averaged about 3 minutes "on" and 2 minutes "off", at an average
rate of about 0. 5-1 gpm/ft during the spray cycle.
Polypropylene vs. Stainless Demister. The advantages of a poly-
propylene demister over a stainless steel demister are: (1)
reportedly greater resistance to corrosion and erosion, (Z) lighter
weight, and (3) easier cleaning characteristics. These advantages
are, however, largely offset by the plastic demister1 s poor impact
resistance, which makes it more vulnerable to breakage during
installation, operation, removal and cleaning. All of these charac-
teristics were observed during venturi system Runs 502-1A through
506-1A (see Table 3-1).
Chevron vs. Centrifugal Demister. As part of the equipment
evaluation program, both chevron and centrifugal (or whirl vane)
demisters were tested in the venturi system after-scrubber. The
pressure drop across the centrifugal demister was found to be
prohibitive: about 5. 0 and 7. 0 inches H2O at superficial gas veloc-
ities of 5. 0 and 6. 3 ft/sec, respectively. Apparently, the centrifugal
unit supplied with the scrubber was not properly designed.
The mixture ratio is dependent upon the percent solids discharged,
the percent solids recirculated and the gas flow rate.
4-6
-------
Top (Downstream) Demister Wash. The results of venturi system
Runs 506-1A, 507-1A and 508-1A showed that the use of top (down-
stream) demister wash resulted in a considerably reduced rate of
solids buildup on the demister vanes. However, the carryover
of slurry solids, which ultimately deposited on the reheater
sleeve, the exhaust duct wall and the ID fan dampers, was substan-
tial. It is believed that the abnormally high slurry droplet entrain-
ment was primarily due to the short distance (14 inches) between the
top wash nozzles and the top tangent of the spray tower and the
resultant rapid acceleration of the gas as it approaches the converg-
ing outlet (96 to 40 inches diameter). The extent of solids deposition
in the duct and on the fan dampers downstream of the reheater
during runs with bottom wash only had been limited to light solids
coatings.
Minimizing Demister Pluggage Problems. Based on the results to
date at Shawnee, it appears that the following design provisions should
be effective in minimizing plugging problems:
• Minimize scrubber superficial velocity, consistent
with cost, turndown, space and other factors.
• Utilize a wash tray between the uppermost scrubber
stage and demister.
• Wash demister from underside (downstream) with a
' mixture of clarified liquor and all available makeup
water, and assure complete surface irrigation.
4-7
-------
4. 3 HOT GAS/LIQUID INTERFACE
The hot (=» 320°F) flue gas feed was humidified with raw water
•f*
during the open-loop'1" factorial testing before entering the neoprene
rubber lined TCA and Marble Bed scrubbers, to reduce its temper-
ature below 190°F, the maximum, permissible for liner protection.
Cooling of the feed gas is not required in the venturi system, since
the venturi scrubber itself is a very efficient humidifying device.
sjojc
In spite of soot blowing with air ' for 90 seconds (45 seconds in each
direction of travel) at four hour intervals, there was a continual prob-
lem of soft solids buildup at the hot gas/liquid interface sections of
the TCA and Marble-Bed scrubbers. There has been little evidence
of any solids buildup within the venturi scrubber when the sootblowers
have operated properly.
To facilitate closed-loop operation by minimizing raw water addition,
the TCA and Marble-Bed scrubbers were provided with process slurry
cooling systems during the "boiler outage in February-March 1973.
The TCA system was equipped with a Ventri-Rod presaturator in the
horizontal gas duct, two feet upstream of the scrubber entrance. The
performance of the Ventri-Rod was not satisfactory in the horizontal
*See Section 4. 1
One blower per scrubber with two 3/4-inch diameter venturi nozzles
leading and trailing at 15° from the vertical at a rated air consump-
tion of 1920 scfm and a blowing pressure of 160 psig.
4-8
-------
flow configuration (rapid buildup of soft solids occurred both on and
downstream of the Ventri-Rod) and it was replaced, during reliability
verification Run 501-2A, with a humidification section consisting of
four full cone Bete nozzles (ST-24 FCN).
By subsequent careful selection of the proper size, orientation,
location, and number of the spray nozzles, modification of the soot
blower head (both nozzles leading at 45°), air blowing during for-
ward travel only, and installation of a Y strainer in the process
slurry line to the cooling spray nozzles, the buildup of solids in
the inlet duct was eliminated? These improvements resulted in
the wet-dry interface being moved to within 12 inches of the scrubber
entrance, from where accumulated solids are easily blown into the
scrubber and discharged through the 36 inch downcomer for re-
slur rying in the effluent hold tank.
On the Marble-Bed scrubber, the modifications of the cooling slurry
system resulted in moderate but encouraging results. The effec-
tiveness of the latest modifications will be verified following the
resumption of testing on the Marble-Bed system.
4. 4 REHEATERS
Flue gas is reheated after evolving from the scrubber to prevent
condensation and corrosion in the exhaust system, to facilitate
isokinetic and analytical sampling, to protect the induced draft
fans from solid deposits and droplet erosion, and to increase
plume buoyancy. The reheaters originally employed were fuel oil
*
Excepting for Run 514-2A (see Section 3.
4-9
-------
fired units with separate combustion air supply and with combustion
occurring in the flue gas stream. The reheaters had been difficult
to start and operate during the short-term factorial testing and
combustion had been incomplete, which led to a visible plume con-
taining significant quantities of soot. This made it difficult to
interpret outlet particulate data and affected gas sampling by the
duPont SO photometric analyzers. Moreover, significant quantities of
oil-soaked soot accumulated in the duct work and, on two occasions, re-
sulted in fires. The difficulty appeared to result from quenching of the
flame by the cold (128°F) flue gas before complete combustion could
occur, and from operating with the same fuel nozzles over a wide range
of flow rates.
The reheater systems were modified during the scheduled boiler
outage in early 1973. Internal stainless steel sleeves (10 gage,
304 SS, 40 inches in diameter by 4 feet high) were installed to pro-
vide approximately 50 cubic feet of isolated combustion zone for
each reheater. Also, the turbulent mixing type nozzles supplied
originally were replaced with mechanical atomizing nozzles. These
new nozzles are designed for a relatively narrow range of oil flow
rate and have to be changed when the reheat requirements change
significantly. Nozzle replacement, however, is a simple job.
The above modifications appear to have been effective. Little or
no soot is visible in the stack gas, and the outlet particulate samples
have shown no appreciable quantities of carbon from the reheaters.
Therefore, plans for installation of an external combustion system
on one of the reheaters have been deferred.
To assess their durability, the internal stainless steel sleeves were
inspected periodically. On all three reheaters, sleeve failure
4-10
-------
in the form of warping, girth •weld failure and excessive oxidation
of the metal in the proximity of the burners occurred after 2000 to
3000 hours of operation. The warpage of the sleeves is attributed
to unequal thermal stresses while operating with only two of the
three burners and the excessive metal oxidation is caused by local-
ized hot spots in the areas of flame inpingement. All three sleeves
were replaced during July (Marble-Bed) and September 1973 (spray
tower and TCA) with 1/4 inch thick 310 stainless steel ones of the
original dimensions.
In September 1973, both the venturi and TCA reheaters were re-
lined in preparation for the long-term test runs.
The modifications performed at the test facility are acceptable as an
expedience for operation. However, operation is still not consistent
with the requirements for long-term sustained reliability. Therefore,
full scale application of direct-fired reheat, in applications where
the flame is subject to quenching, is to be avoided.
4. 5 FANS
Initially, considerable difficulty was experienced with the induced
draft fans. Some of the problems included high fan vibration, fan
motor failure, fan damper control failure and fan blade deformation.
All of the problems, except for blade deformation, necessitated
repeated shutdowns of the affected scrubber systems.
The unacceptable high vibration problem of all three fans was
greatly reduced in June 1972, by insulating the fan housing, adding
additional bracing to the outboard pedestals, and welding balance
weights on the fan shrouds. However, occasional high fan vibra-
tion continued to hinder scrubber operation, particularly on the
4-11
-------
venturi system, and required either addition of shims to the bearings
or replacement of the bearings (the high venturi fan vibration was
due to excessive clearance between the outer bearing race and the
bearing housing on the inboard bearing).
The motors of the venturi and TCA fans had to be returned once to
the supplier for repair and correction of serious manufacturing
problems.
Stable flue gas flow control -was achieved by increasing the "fully
open" to "full closed" fan damper response time from 1 0 to 100
seconds with new actuators. Three scrubber system shutdowns
were caused by inoperable linkage and a broken shear pin.
Distortions of several blades (arc shapes as contrasted to the
original straight line configuration) of the Marble-Bed and TCA fans
were observed in March 1973. The maximum deformation was 0. 55
inch on a blade of the Marble-Bed fan. The manufacturer indicated
that the deformation was probably caused by stress relieving during
fan operation and that the warping of the blades did not interfere
with efficient, safe operation. No significant continuing deformation
of the blades has 'been observed to date.'
Erosion, corrosion, pitting, scaling, etc. to date have been negligible
on all three fans. However, past operation has been only with 125°F
flue gas reheat to give a fan outlet temperature of 250°F. Future
plans include operation with only 50°F reheat.
4-12
-------
4. 6 PUMPS
Most of the pumps used at the Shawnee test facility in alkali slurry
service are rubber lined variable speed centrifugal pumps. Most
of the original Hydroseal type pumps were converted to either
Centriseal type or were modified to include a mechanical seal during
the boiler outage during February 1973.
Both the Hydroseal and Centriseal pumps have performed satisfac-
torily. It was necessary, however, to install pump discharge-to-
suction recirculation lines to eliminate the vapor-lock problem of
Centriseal pumps at low flow rates. More frequent replacement
of packing material and shaft sleeves has also been noted in the
Centriseal pumps.
In general, the rubber linings have been found to be in excellent
condition. Part of one slurry spray nozzle was found in the effluent
hold tank recirculation. pump on the Marble Bed Scrubber, and had
damaged the rubber lining. Wear rates of the rubber linings will be
determined during the long-term reliability runs.
One stainless steel variable speed centrifugal pump was originally
used in the limestone addition system. Severe erosion of the housing
and impeller were noted (with a corresponding loss of 20% pumping
efficiency) in only twenty days of actual operation. Subsequently,
this pump was replaced by individual Moyno pumps to each scrubber
system. Operation of the Moyno pumps (stainless steel rotor and
4-13
-------
butyl rubber stator) has been satisfactory with normal maintenance
required. Wear rates will be determined during the test program.
4. 7 WASTE SOLIDS HANDLING
The test facility is equipped to study alternate methods of waste
solids dewatering and disposal. Separate clarifiers are provided
for each scrubber system, and a rotary drum vacuum filter, a
horizontal solid bowl centrifuge and a slurry settling pond are com-
mon to the three systems. Solids separation for any scrubber system
can be achieved with any combination of clarifier/filter/pond or clar-
ifier/centrifuge/pond.
4. 7. 1 Clarifiers
The clarifiers are conventional solids contact units with a heavy
duty rake and scraper mechanism supported from a bridge. The
vessels are flake-glass lined with a stainless steel rotating mech-
anism. The venturi and Marble-Bed systems have 20-foot diameter
units while the TCA clarifier is 30 feet in diameter.
The performance of the clarifiers during the short-term factorial
test period was unsatisfactory. Solids carryover in the overflow
of the two smaller units -was a problem and the solids concentration
in the underflow streams of all three units could not be controlled.
Following extensive system modifications during the February 1973
power outage, so that a. purge stream could be routed from each
scrubber slurry recirculation loop directly to the clarifiers, the
4-14
-------
performance of the clarifier s improved significantly. The concen-
tration of solids in the underflow of the larger TCA unit approaches
the expected final settled density of the sludge (approximately 40
percent by weight). However, the poor settling characteristics of
certain solids components, particularly calcium sulfite and fine fly
ash, and the high solids loading in the bleed continued to result in
periodic high solids carryover in the overflow of the 20 foot diameter
venturi and Marble-Bed units (about 0. 2 to 5.0 percent by -weight at
14.5 to 25.0 tons per day solids loading, respectively). Also, the
highest solids concentration in the underflow streams from these
smaller units averaged about 25 percent by weight.
Preliminary test data with lime slurries indicate that, with the
reduced solids loading to the venturi clarifier (due to better lime
utilization), the solids concentration in the overflow averaged less
than 0. 1 percent by weight. The solids level in the underflow stream
remained unchanged at about 25 percent by weight.
4.7.2 Centrifuge
Following some exploratory tests in April 1973, the centrifuge was
used for solids dewatering in the venturi and Marble-Bed systems,
directly or in series with a clarifier. The average feed rates were
25 to 35 gpm with solids concentration varying between 8 to 20 per-
cent by weight. The operation of the centrifuge was quite satisfac-
tory at 1 - 1/4 inch pool depth and 2000 RPM speed. The cake solids
were in the 56 to 62 weight percent range and the centrate solids
averaged 0. 5 to 1.0 weight percent.
4-15
-------
The centrifuge failed after about 1400 hours of operation. Detailed
inspection revealed "above average" -wear of the bowl and conveyor,
requiring their return to the factory for repair. The eroded casing
was repaired on site.
4.7.3 Filter
Following modifications to the system to facilitate trouble free
discharge of the cake-wash-water mixture, the rotary vacuum filter
was operated continuously for about 200 hours. The cake discharge
from the nylon cloth was satisfactory without the use of mechanical
equipment (scraper or wire). Tests have indicated that the dewater-
ing capability of the filter corresponds to 50-55 and 45-50 weight
percent solids in the cake from limestone and lime slurries, re-
spectively.
Filter operation has been significantly affected by the life of the
filter cloth. The useful life of the polypropylene and nylon filter
cloths tested to-date is unsatisfactory.
4. 8 SCRUBBER INTERNALS
4.8.1 TCA Wire Grids
The original wire mesh grids (0. 148 inch diameter stainless steel
wires) deteriorated considerably during the approximately 3000
hours of operation in slurry service. Vibration caused by plastic
sphere activity resulted in the rubbing together of the grid wires at
4-16
-------
their perpendicular junctions and the grids failed at several locations
due to subsequent erosion of the wires in slurry service.
The wire mesh grids were replaced with sturdier "rod grids" (3/8
inch in diameter, SS rods,l -1/4 inch on centers) prior to the long-
term limestone reliability ru*n.
4.8.2 TCA Spheres
A significant limiting factor in the long-term reliability of the TCA
scrubber has been associated with the erosion and subsequent collapse
of the UOP supplied 1-1/2 inch polypropylene and polyethylene
plastic spheres used as packing. The collapsed spheres eventually
fill with slurry and settle to the bottom of the support grid. Random
samples of collapsed spheres, subsequent to the termination of the
limestone factorial runs, showed about 60 percent weight loss. Other
data has indicated a weight loss of about 27 percent for the plastic
spheres, after approximately 1000 hours of operation.
L/
During the initial phase of the long-term reliability limestone run
on the TCA (see Section 3.2), the top-most bed utilized UOP supplied
thermo-plastic-rubber (TPR) spheres, and the bottom two beds
UOP supplied high density polyethylene (HDPE) spheres. After 500
hours of operation, the TPR spheres had lost approximately 2.6
percent of their original weight and the HDPE spheres from 8-14
percent. Also, about one percent of the TPR spheres in the top
stage had come apart at the seams. Some of these half-spheres
eventually lodged in the slurry recirculating nozzles, causing pre-
mature termination of the run.
4-17
-------
4. 8. 3 Nozzles
Nozzle reliability at the test facility has been greatly reduced by the
frequent plugging of spray nozzles with foreign material (TCA spheres,
marbles, debris, etc. ), and the erosion of some spray nozzles by the
abrasive solids in the circulating slurries. It has become apparent
that nozzle plugging could be reduced substantially by placing screens
over open vessels in the scrubber systems and/or within the circu-
lating slurry lines.
Spray Tower. Limestone factorial testing in the spray tower started
with the use of spiral tip, 316 SS, full cone, Bete No. ST-24 FCN
nozzles (capacity: 12 gpm @ 12 psig) manufactured by Bete Fog
Nozzles, Inc. Because of frequent plugging with slurry and/or
debris, these nozzles were replaced in September 1972, with Bete
No. ST-32 FCN nozzles (capacity: 21 gpm @ 10 psig). Plugging of
the larger Bete nozzles became less frequent. Neither type of nozzle
showed any significant sign of erosion.
To allow for increased liquor flow to the four-header spray tower,
Bete No. ST-48 FCN stainless steel nozzles (capacity: 47 gpm @
10 psig) were installed during the February 1973 shutdown. During
the first limestone reliability verification test (Run 501-1 A), five of
the 28 nozzles became totally plugged with debris and four nozzles
became partially plugged. Although erosion of these stainless steel
nozzles has not been observed to date, after approximately 3500
hours of slurry service, they will be replaced with identical stellite-
tipped ST-48 FCN nozzles in the near future.
4-18
-------
TCA. The large Spraco 1969, full cone, 316 SS, open-type slurry
feed nozzles have performed satisfactorily and without significant
erosion since the original startup of the unit. Occasional partial
pluggage by large debris did not necessitate premature termination
of any test run (excepting pluggage by TPR spheres, see Section 3.2).
Marble-Bed. The 22 original slurry feed spray nozzles lined with
Solathane 291 and equipped with internal Adiprine LD 315 swirl vanes
failed in various ways during short-term factorial testing. The
swirl vanes in all 22 nozzles eroded, the liners of four bottom
nozzles collapsed, and two bottom nozzles disintegrated. The nozzles
frequently became plugged with slurry and debris.
The original slurry feed nozzles were replaced (during the February
1973 shutdown) with improved nozzles supplied by Combustion
Engineering (stronger, Adiprine LD 3056 lining with improved bonding
using Thixon 1244 between the liner and the body of the nozzle and a
locking groove to hold the vanes in place). The diffusion vanes of 13
(of the 16) bottom spray nozzles failed during the initial limestone
reliability verification test run (Run 501-3A), after 764 hours of
operation.
4. 9 LININGS
Two types of lining material were used throughout the scrubber
systems. The Marble-Bed and TCA scrubbers, the venturi scrubber
downstream of the plug, the venturi after-scrubber, the process
water tanks, the pumps, the circulating slurry piping and the tank
4-19
-------
agitator blades are rubber lined. The effluent hold tanks and clarifiers
are lined with Flakeline glass.
The rubber linings have been found, generally, to be in excellent
condition. Essentially no erosion or deterioration has been noted.
However, slight wear has been noted on some of the rubber-coated
agitator blades. This type of wear is believed to be caused primarily
by foreign objects striking the agitator blades (rubber lined pumps
are discussed in Section 4. 6 ).
Hairline cracks have been noted in the Flakeline lining of the effluent
hold tanks and clarifiers. The cracks did not appear to penetrate the
entire thickness of the lining. The cracks are more prevalent at the
junction between the baffles and the tank walls. Isolated areas on
the bottom of the TCA effluent hold tank also show wear by erosion.
These areas are also near the wall baffles where eddy currents are
formed.
Flakeline patching material is available for lining repair but has not
been used to date. The eroded areas on the bottom of the TCA effluent
hold tank were painted with epoxy.
Prior to starting the long term reliability run with limestone on the
TCA, the agitator in the effluent hold tank was lowered four feet. As
a precautionary measure, a steel \vear plate was also installed on
the bottom of the effluent hold tank covering the area under the
agitator.
4-20
-------
4. 10 INSTRUMENT OPERATING EXPERIENCE
4.10.1 Sulfur Dioxide Analyzers
Essentially trouble-free operation was experienced with the duPont
Model 400 UV sulfur dioxide analyzers following the modification of
the sampling system and the replacement of interference filters in
November 1972. Initially, the sampling system was particularly
vulnerable to condensation, solid particulates, oil, soot, corrosion,
or the combinations of these factors which led to leakage or plugging
of the sampling lines, plugging of the filters, or coating of the
optical lens. All of these effects caused erroneous sulfur dioxide
analyzer readings.
To eliminate the problem areas, the sampling handling system was
modified as follows:
All heat sinks and sharp bends in the sample lines were
eliminated. A new 3/8 inch diameter Dekeron sample
line was installed to replace the original 1/4 inch stainless
steel line. Heat tracing was installed along the full length
of the sample line.
Stainless steel shields furnished by duPont were installed
around the probe filters. The original ceramic probe
filters were replaced by probe filters made from 316
stainless steel and recently developed by duPont.
An automatic zero and air blow-back system was installed
on the SC>2 analyzers in the inlet gas ducts, similar to
those provided originally in the scrubbed gas ducts.
Stainless steel lines and fittings were replaced with Dekeron
or Teflon plastic wherever possible.
4-21
-------
Calibration methods were changed to use a stainless steel
•wire mesh reference filter rather than bottled standard
reference gas.
One additional problem associated with all six analyzers was the
deterioration of the interference filter in the optic section. All of
these filters, •which filter out all except the desired light wave
lengths, were replaced. The failure and subsequent deterioration of
the filter was attributed by duPont to the exposure of the analyzers
to freezing conditions prior to their installation. It was theorized by
duPont that the freezing caused minute cracks which then deteriora-
ted with time.
4.10.2 pH Meters
Operating experience has been gained with two types of pH meters.
These are Uniloc Model 320 in-line flow-through type and Uniloc
Model 321 submersible type pH meters.
The performance of the in-line flow-through type pH meters has been
unsatisfactory due to the erosion of the glass cells by the slurry,
their high rate of failure, and the frequent plugging of the sample
lines. All the flow-through type meters at the scrubber inlet and
outlet have subsequently been replaced by the submersible type
meters.
For the submersible type pH meters, cell erosion, cell breakage,
and sample line pluggage have not been experienced during the
approximately 1300 on-stream hours of operation. Routine cleaning
4-22
-------
and calibration of the cells measuring the slurry pH at the scrubber
inlet are made twice a week to maintain the desired meter accuracy
with 1 0. 1 pH unit. However, routine calibration checks during the
scrubber operation have not been possible for the submersible cells
located at the scrubber outlets (inside the downcomers). Studies are
being made to effect the routine calibrations of the scrubber outlet
pH meters during the operation of the scrubber systems.
4.10.3 Density Meters
Operating experience has been obtained with the Ohmart radiation-
type, the bubble-type (differential pressure), and the Dynatrol Model
CL-10HY U-tube type density meters.
Experience with the radiation-type density meter indicates a loss
of calibration in the range of about 1 to 2 percent per week. Also,
the meter accuracy is affected by the accumulation of scale in the
sample line which can be removed only during the scrubber shut-
downs.
The sample line and the probes of the bubble-type density meter
plug frequently and require significant maintenance. However, the
meter is accurate when clean and can be used to check the calibra-
tion of the Ohmart radiation-type density meters.
The Dynatrol density cells (using the vibration principle of the U-tube
for continuous response to density changes) have been installed in
September 1973, to measure the densities of the lime slurry feed to
4-23
-------
the venturi/spray tower system and of the circulating limestone
slurry to the TCA system. The performance of this type of density
meters has thus far been encouraging.
All three types of meters require further study and modification to
achieve adequate reliability in their respective control service.
4.10.4 Flowmeters
Foxboro magnetic type and differential pressure type (both orifice
and annubar) flowmeters are used at the test facility.
Operating experience with the magnetic flowmeters has generally
been good. The main problem has been in obtaining accurate
flow measurements at very low flow rates with meters designed to
measure flow over a wide range. To assure accuracy, Foxboro
recommended a minimum linear velocity of 3 ft/sec through the flow
element.
The magnetic flowmeters smaller than 4 inches have a tendency to
drift in calibration and frequent flow checks are required to verify
the accuracy. Meters larger than 4 inches are more reliable; how-
ever, scale accumulations on the electrodes over an extended period
influence the accuracy and necessitate periodic cleaning.
4-24
-------
The Scothane liners in the magnetic flowmeters are vulnerable .to
failures attributable to leaving the power on for long periods during
shutdowns when the meters are drained of liquid.
The orifice type flowmeters appear to function reasonably well in
slurry service (6 to 1 5 percent solids by weight), provided that the
problem of the plugging of pressure taps is resolved by using dia-
phragms as close as possible to the slurry piping. Preliminary
inspection results indicate relatively little erosion of the 316 stainless
steel orifice plates after about 2500 hours of operation.
Experience at the test facility indicates that annubar meters should
be used only in non-scaling, clear liquid service (containing a maxi-
mem of 0. 5 percent by weight of suspended solids) to prevent frequent
plugging and associated maintenance work.
4.10.5 Control Valves
Operating experience with control valves in slurry service has generally
been unsatisfactory. Severe erosion in a short time has been caused
by the increasing velocity during throttling operation. This deteriora-
tion has been observed in the stainless steel plug valves, globe valves
and rubber pinch valves. Satisfactory and. trouble free flow control
has been experienced only with variable speed pumps.
4-25
-------
4.11 MATERIALS EVALUATION
TVA has conducted a study for the evaluation of corrosion and wear
of plant equipment and test specimens (coupons) at the Shawnee
facility. A detailed interim report of the results of this study, by
G. L. Crow and H. R. Horsman, is presented in Appendix D. The
following is a summary of some of the results of that study. Linings
have been previously discussed in Section 4. 9.
4.11.1 System Components
A thorough inspection of all system components was conducted during
the extended February and March 1973, boiler outage. Each of the
three scrubber systems had been operated for about 1800 hours during
the factorial limestone scrubbing tests.
Localized deposits of loose fly ash accumulated in the mild steel
gas ducts between the boiler and scrubber structure. The surfaces
were coated with a thin iron oxide scale and moderate pitting had
occurred at the uninsulated connections. The flanges and access
doors have been insulated.
The most severe corrosion was found on Type 316 stainless steel
surfaces, particularly on the mist eliminator blades in the TCA
system. In general, the corrosion was in the form of pitting with
some pits as large as 1/16-inch diameter and 30 to 35 mils deep.
4-Z6
-------
Significant erosion was noted on the pump sleeves, at the intersections
of the wire of support grids in the TCA scrubber, and on the impeller
and casing of the 316 SS Gould limestone slurry pump.
4.11.2 Test Coupons
Test coupons of several different materials of construction, together
•with stressed and welded specimens, were exposed for periods of
1700 hours or longer to various slurry and gas environments. The
corrosion rates observed are presented in Table 4-2.
Corrosion of Hastelloy C-276 was from negligible to 5 mils per year.
This alloy showed no evidence of localized attack in any test location.
Next in resistance to corrosion were Inconel 625, Incoloy 825, Car-
penter 20 Cb-3, and Type 316L stainless steel alloys. The corrosion
rates for each material ranged from negligible to 5, 7, 14 and 15 mils
per year, respectively. These alloys had few minute corrosion pits
and/or crevice corrosion. Type 316L, the fifth alloy in corrosion
resistance, is the least expensive of this group of materials.
Three nonferrous alloys, Cupro-Nickel 70-30, Monel 400, and
Hastelloy B, each had minimum corrosion rates of less than 1 mil.
Maximum corrosion rates were 49, 57 and 100 mils per year, re-
spectively. Only one or two specimens pitted. In three tests of
Monel and in one test of Cupro-Nickel 70-30, the welds were in-
ferior to the parent metal.
4-27
-------
CORROSION TEST RESULTS
Metals (a)
1. Hastelloy C-276
2. Iconel 625
3. Incoloy 825
4. Carpenter 20Cb-3
5. Type 316L SS
6. Cupro-Nickel 70-30
7. Monel 400
8. Hastelloy B
9. Type 446 SS
10. E-Brite 26-1
11. Incoloy 800
12. USS 18-18-2
13. Type 304L SS
14. Type 410 SS
15. Aluminum 3003
16. Mild Steel A-283
17. Cor-Ten B
Corrosion,
mils/yr
Neg. to 5
Neg. to 5
Neg. to 7
Neg. to 14
Neg. to 15
=>1 to 49
=>1 to 57
=»1 to 100
Neg. to <:140
Neg. to <190
Neg. to < 190
Neg. to 1 to<:250
=»1 to<500
>1 to <1400
>1 to -C1400
Number of
Pitted
Samples"5)
_
1
-
2
3
1
1
2
9
10
6
11
14
15
9
2
5
Pitted Depth, (c)
mils
Min.
_
-
-
-
-
-
-
-
Minute
Minute
Minute
Minute
Minute
Minute
2
-
Minute
Max.
_
Minute
-
Minute
Minute
18
2
Minute
19
18
19
16
25
16
70
Minute
5
Number of
Samples With Nu
Crevice Attack S
_
-
1
-
Z
-
1
-
11
2
3
11
11
16
5
Z
4
Other Types of Attack
mber of Area of
amples(b> Attack
_
-
-
-
1,1 CIS," Weld
1 Weld
3 Weld
-
-
-
1 (e)
-
1 (e)
-
-
-
1 Weld
I
IN)
00
Non-Metals'f)
Plastics Bondstrand 4000
Flakeline 200
Qua-Corr
Rubbers Butyl 1375
Natural 9150
Neoprene 26,666
Ceramic Transite
Evaluation, Number of Samples
Good
12
Z
5
6
6
6
14
Fair Poor
9
14 5
1
2 5
Note: Test samples of each material were tested for 1680 hours, or more.
(a) Metals are listed in approximate order of decreasing corrosion
resistance.
(b) Samples of each metal were tested in 21 locations.
(c) Depth of penetration in mils during total exposure period.
(d) Groove in parent metal is 18 mils deep.
(e) Severe localized attack of parent metal.
(f) Samples of Bondstrand, Flakeline and Transite were tested in
21 locations. Samples of Qua-Corr and rubbers were tested in
6 locations.
-------
The corrosion rates of Type 446 stainless steel, E-Brite 26-1,
Incoloy 800, USS 18-18-2, and Type 304L stainless steel, ranged
from negligible to values which indicated that the alloy specimen
•was completely destroyed at one or more test locations. The values
for the specimen failures ranged from greater than 140 mils per
year for Type 446 to greater than 200 mils for both USS 18-18-2
and Type 304L stainless steels. These five alloys were highly
susceptible to localized corrosion.
Another group of alloys, Type 410 stainless steel, 3003 aluminum,
A-283 mild steel, and Cor-Ten B, had minimum corrosion rates of
less than 1 mil per year and maximum corrosion rates of greater
than 250 mils for Type 410 to greater than 1400 mils for A-283 and
Cor-Ten B. Pitting and crevice corrosion occurred on the four
alloys.
In general, the stressed specimens (five alloys only) were not corroded
at rates higher than their counterpart disk-type specimens.
Specimens of Bondstrand 4000, Flakeline 200, and Transite materials
were tested at 21 locations. Bondstrand 4000 showed good corrosion
resistance in 12 tests and poor resistance in nine tests. Only six
specimens of each of the following materials were tested: Qua-Corr
plastic, butyl natural rubber and neoprene rubber. The results
were: five good specimens and one poor specimen for Qua-Corr
plastic, and six good specimens for each type rubber.
4-29
-------
With few exceptions, mainly in the TCA system, the greatest loss
of -weight from metal specimens occurred in areas where the velocity
of the unscrubbed, partially humidified flue gas was comparatively
high. Impingement on the specimens of the slurry caused erosion
and corrosion. Pitting and crevice corrosion were not important
factors where erosion and corrosion kept the specimens clean. In
other areas of the three scrubber systems where solids accumulated,
the frequency of localized corrosion was high. However, each of the
17 alloys tested showed good corrosion resistance at one or more
test locations in each scrubber system.
4-30
-------
Section 5
LIME TEST RESULTS
AT THE SHAWNEE FACILITY
On October 9, 1973, a reliability verification test run (Run 601-1A)
was begun on the venturi/spray tower system. The primary variable
selected for control was the pH of the recirculating (scrubber inlet)
slurry. This value was automatically controlled at 8. 0 t 0. 3 by coup-
ling the pH meter to the lime addition system. This pH control level
was chosen based upon the results of lime testing at the EPA pilot
facility in Research Triangle Park (see Section 6), which indicated
reasonable lime utilization (lOOx moles 803 absorbed/mole Ca(OH)2
added) and SO£ removal at that level. Other operating conditions were
selected based on results from limestone reliability verification testing.
The test conditions for Run 601 - 1A were:
Gas Rate, acfm @ 330°F 25, 000
Spray Tower Gas Velocity, ft/sec 6. 3
Liquor Rate to Venturi, gpm 600
Liquor Rate to Spray Tower, gpm IZOO
Venturi Liquid-to-Gas Ratio, gal/mcf 32
Spray Tower Liquid-to-Gas Ratio, gal/mcf 64
Percent Solids Recirculated 8
Effluent Residence Time, min. 12
Number of Spray Headers 4
Venturi Pressure Drop, in. ^O 9
Spray Tower Pressure Drop, in. H£O 2,5
Scrubber Inlet pH (Controlled) 8.0
5-1
-------
On October 28 the system was shut down for 24 hours due to a problem
with the lime slaker and on November 7 there was a 15 hour scheduled
shutdown for an inspection (after 666 on-stream hours). The November
7 inspection revealed a very clean scrubber, except for minor scale
and solid deposits in parts of the flooded elbow (between the venturi and
spray tower and the upper half of the scrubber). However, occasional
cleaning of accumulated scale from the venturi tangential nozzles was
required. All the bottom demister vanes were clean while the top
vanes were about 5 percent plugged with soft mud-like solids. This
minor pluggage was not considered to be serious. Downstream of the
demister some scattered solids-deposits (1/16 inch deep) were ob-
served on the duct between the demister and the reheater and on the
fan inlet dampers. Aside from these minor deposits, both the fan and
outlet duct were clean.
Throughout this initial portion of the run (till the November 7 inspec-
tion), the clarifier was used as the final dewatering device. This re-
sulted in an average percent solids discharged of 23 percent, which is
somewhat lower than desired. The desirable solids concentration to
be discharged is 40 percent or greater.
Graphical representation of the operating data for the first 30 days of
Run 601-1A is given in Figure 5-1. As can be seen, the control of
scrubber inlet pH and stoichiometry was not as good during the first
20 days of operation as during the latter 10 days. A summary of the
operating data for lime Run 601-1A, as of the November 7 inspection,
is as follows:
5-2
-------
i IECINHJN W1-IA
es'^^^V.^^
s" \ ''!
— iJ ) CCO
is.'
o o°
00°
VI
/.
°o
'.TO
».*0
1.00
7.UJ
10/11 I 10/1?
Till 1IME. feur.
10/1] I TO/14 I IO/T3
CALENDAI DAY
10/14 I 10/17 I 10/11
On Dm • 15.000 icfm » 330 °F
Llquw B«t Id Viitturl • i!00 ipm
Liquor Ritt n Sprty Towr . 1200 ipm
Vinturl 170 • 5J (il/mcf
8pnyTmnrUa-e4gil/incl
Spray Tonr On Viiocrty • U.J fi/M
Vinturl Pnoin Crop - 9 In. HjO
B.H.T. Rtjdtnci Tlfiil -1J nta
8cnbbvlnMLI<|«iiTiinp.-Jn-IMl>F
LlquM ContfuctMty • 3,700-S.MO u mhoi/cm
Dhdargi{QirKM ttolMl Cone.• 2I-78 wl K
tO/TO I 10/11
TUT TIME, houn
10/T] I 10/14 I 10/13 I tO/ld
CALENDAR DAY
10/17 I 10/18
• TOTAL DISSOLVED SOLIDS * MAGNESIUM [Mg ** )
O CAlCrjM [Co " ) £ SODIUM (No * )
Q SULFAlt < M> ih^n.
DA
A
s
B7
JL
TEST TIME, heun
I 10/14 I
CALENDAI DAY
to/15 I 10/16 I 10/17
Figure 5-1
OPERATING DATA FOR VENTURI
RUN601-1A
5-3
-------
UN Ml U CONTIMJED
Gn Ritt • 26,000 icf m 0 330 °F
Liquor Rita to Vtnturf • 600 gpm
Liquor Ritt to Spny Tontr • 1200 gpm
Vinturi UG • 33 gtf/mcf
SpnyTowUG-Moil/mcf
Sony Towtr Gn Vilodry - 6.3 HJmc
Vtnturl Prwtwn Drop - 9 in. H20
E.H.T. RnUvna Tim - 12 mln
No.ofSpnyHndn-4
Scrubbir Inltt Liquor Ttmp. • 123-126 °F
Liquid Conduct(vtry - 5,200-8,600 u tnhu/cm
Obchargi (Cltrfflfrt SolMi Cone. - 22-28 wt %
14,300
12,000
10,000
1
=^ 8,900
*
S 6,000
|
| .,000
1 2.000
z o
* TOTAL DISSOLVED SOLIDS * WAGNEIIUM (Mg « )
O CALCIUM (C. **) i JODlUMtNa*)
D SULFAIE BO( * ) 7 POTAJJIUM (K ' )
A CHIO*IDE{C|- ) • SULFlT(nOj')
0 CAHONATt (C03' )
•
•
• *
•
.
'
0 9 =$ I
A •
M,COO
12,000
10.000
1.000
4,000
4.000
1,000
0
10/21 I 10/71 I 10i/73 I
TIST TIMH, houn
10/74 I 10/21
CALCNOAI DAY
10/18 I 10/W
Figure 5-1 (Continued)
OPERATING DATA FOR VENTURI
RUN601-1A
5-4
-------
Gn Hit. • 25,000 Kim « MO °F
Liquor Hill to Vtrrnirl • 600 Jpm
Liquor Rltl to Sprey Towfr • 1200 gpm
VmturlL/0-ngal/mtl
Spray TowtrL/0-64 gil/mcf
Spny Towv On Vfloclty • 6.3 fl/ne
Vinturi Prraura Drop • 8 In. H jO
EJ4.T.Rttldtnc
-------
On-Stream Hours as of November 7 666
Average Stoichiometric Ratio, moles 1.10
Ca/:rnoles SO2 absorbed
Average Percent Lime Utilization 91
Average Percent SC>2 Removal 80
Inlet SO2 Concentration, ppm 1600-4000
Scrubber Outlet pH Range 4. 5-5.5
Solids Dewatering Clarifier
Average Dissolved Solids, ppm 7000
Predicted Sulfate Saturation 150
It is of interest to note, from the data in Figure 5-1, the effect of SOo
inlet concentration on SO2 removal; a drop of SO2 inlet concentration
is invariably accompanied by an increase in SO2 removal, and vice
versa.
Subsequent to the November 7 inspection, system control remained ex-
cellent, despite the varying SO2 inlet load; stoichiometric ratio was
controlled between 1. 02 and 1.16 and scrubber inlet pH between 7. 8 and
8.4.
In order to evaluate the effect of higher percent solids discharged
(higher dissolved solids concentration) upon system reliability, Run
601 -1A was restarted on November 7 with the vacuum rotary drum
filter in series with the clarifier. However, problems with the filter
(filter cloth) resulted in intermittent operation of thid device and on
November 23 the centrifuge was put in service. The centrifuge, how-
ever, also had mechanical difficulties (high vibration) and was taken
out of service after a short while. On December 15, the filter was put
back in service and remained in operation, with periodic replacement
of the filter cloth, until the termination of the test. A summary of the
operating data for Run 601-1A, with both the clarifier and filter in ser-
vice for dewatering, is as follows:
5-6
-------
On-Stream Hours with Clarifier 1487
and Filter
Average Stoichiometric Ratio, 1.11
moles Ca/moles SO2 absorbed
Average Percent Lime Utilization 90
Average SO2 removal 85
Inlet SO 2 Concentration, ppm 1700-4000
Scrubber Outlet pH Range 4. 6-5.4
Solids Dewatering Clarifier/Filter
Average Dissolved Solids, ppm 11,000
Predicted Sulfate Saturation 190
Throughout Run 601-1A, the pressure drop across the chevron demis-
ter had been somewhat variable (0. 63 1" 0. 1 inches f^O) and, after
about December 1, 1973, there was a continual increase. On January
8, 1974, Run 601-1A was terminated after 2153 hours (3 months) of
on-stream operation due to high I. D. fan vibration. The pressure
drop across the demister had increased from about 0.6 to about 1.65
inches H~O. An inspection was conducted on January 9 and the accum-
ulation of scale and solids within the scrubber system is shown in
Figures 5-2 and 5-3.
The mist eliminator was substantially blocked, partly by solids that
had fallen down from the outlet duct work and partly by scale formation,
mainly on the bottom (inlet) vanes. The bottom demister vanes were
covered with 30 to 40 mil scale. The outlet ducting between the re-
heater and I. D. fan damper was covered with from 1 to 3 inches of solids.
The shell of the spray tower was covered with a 3/8 inch thick sulfate
scale below the bottom spray header and a 1/8 inch thick sulfate scale
between the top and bottom spray headers. The rubber lining was vis-
ible in an approximate 2-1/2 ft wide band immediately below the
demister. Noticeable scale and solids buildup occurred on the tips of
5-7
-------
ui
i
00
LEGEND:
/ffff/ff SCALE
SOLIDS
CLEAfii
E -•
DATE OF INSPECTION: ^A*/-
INSPEaOR: -^ ^
SEaiON'A1 SHOWING t>£D /Ooo
SLIDING GUIDE BOLTS /W ytf//> -
w/M 3/6
DETAIL OF SPLASH SEAL
VENTURI
August 1973
Figure 5-2. Venturi Inspection
-------
01
i
vO
m SOLIDS
X PLUGGED NOZZLE
/ PARTIALLY PLUGGED NOZZLE
W-*
ff^ff se*t.f £-
- M
SECTION 'A'
DATE OF INSPECTION:
INSPEQOR: *• c-
ttF 10 MO*^LCS
VAftti ±-±
SECTION 'B'
HEADERS 1 (BOTTOM) 43
SEaiON
HEADERS 2 &4(TOP)
FUNNEL SAMPLER FfLL
IMTO H'/OI TAfifK.
SECTION 'C1
I.D. FAN DAMPERS
FAN
DAMPER
> d~~^ C**-'0S
' ^/'~2mf~^ #ee£.
} \ 3 " TK. *f£#«
TO
r"i
^c
r -.
REHEATER
VENTURI AFTER SCRUBBER
August 1973
Figure 5-3. Spray Tower Inspection
-------
several of the spray nozzles, especially on the bottom header. Some
of the Bete stainless steel ST-48 FCN spiral tip nozzles were found to
be significantly eroded after approximately 4300 hours of slurry ser-
vice; of a total of 28 nozzles, 9 were severely eroded and 15 were con-
siderably worn.
Light 1/16 inch thick scale covered the venturi plug and a 1/8 to 1/4
inch thick scale covered the venturi walls below the plug. A 1/8 to
3/8 inch thick scale covered the walls of the flooded elbow.
About 3/16 inch thick sulfate scale was found in the rubber lined circu-
lating slurry piping. Scale buildup, varying from 1/64 to 1/8 inch in
thickness, was also noted in the rubber lined variable speed pumps.
It should be noted that scale formation in the venturi, spray tower, cir-
culating slurry piping and pumps did not prevent continual operation of
the system or necessitate termination of the 3-month long lime relia-
bility test. However, scale particles in nozzles and strainers required
periodic maintenance.
It seems likely that the scale formation occurred, primarily, during
the latter portion of the test, when the clarifier and filter were used
for solids dewatering and the calculated liquor sulfate saturation was
about 190 percent. As mentioned previously, an earlier inspection of
the system on November 7, 1973, revealed a relatively scale-free
scrubber system. Prior to this inspection the calculated liquor sulfate
saturation was 150 percent.
5-10
-------
The next long-term lime reliability test will be made under conditions
intended to reduce recirculating liquor sulfate super saturation and the
accumulation of solids in the outlet duct-work. Sulfate supersatura-
tion can be reduced (1) by decreasing oxidation of sulfite to sulfate,
(2) by increasing percent solids recirculated, and (3) by increasing
effluent residence time. Oxidation can be reduced by increasing the
scrubber inlet liquor pH and/or by sealing the effluent hold tank. Also,
the eroded stainless Bete spiral tip spray nozzles in the spray tower
will be replaced by Stellite-tipped Bete spray nozzles for the next test.
5-11
-------
Section 6
TEST RESULTS FROM THE EPA PILOT FACILITY
AT RESEARCH TRIANGLE PARK
It is recognized that operating reliability is crucial to the successful
application of limestone and lime scrubbers to utility boilers. This
fact was the central consideration in planning an experimental program
for the EPA pilot facility at Research Triangle Park. The approach to
the reliability question assumes that scaling by calcium sulfate and
sulfite is one of the most important problems. Therefore, the primary
objective is to identify the operating conditions that eliminate scale
formation. The secondary objective is to maximize the limestone and
lime utilization, which will strongly influence operating cost when
reliability is established. This section summarizes progress toward
the improvement of process performance in these two areas.
6.1 DESCRIPTION OF EQUIPMENT AND OPERATION
Two 300-cfm scrubbers are operated concurrently with flue gas ob-
tained from a natural gas/oil-fired boiler. Sulfur dioxide is fed from
cylinders to provide a constant inlet concentration of 3000 ppm. One
TCA scrubber is operated with lime, while the other with limestone ob-
tained from the Shawnee test facility. Limestone tests are conducted
Ln both TCA and multigrid scrubber configurations. Except for selected
tests, fly ash is not present in the slurry and the liquor is free of chlor-
ide. Both scrubbers operate with closed liquor loop using a rotary drum
filter for solids disposal (about 63 percent solids in sludge discharge).
6-1
-------
6. 2 MATERIAL BALANCES AS A BASIS FOR EVALUATING
PERFORMANCE
It is essential that the primary reactions that occur in the two process
components, the scrubber and effluent hold tank, be determined from
among the various possibilities that have been proposed. For this
purpose, extensive material balances were made on limestone
scrubbers over a range of operating conditions. Typical analytical
data acquired for these material balances are indicated in Figure 6-1.
In order to set up the simultaneous equations that will define all
reactions in both parts of the system, it is necessary to measure the
rate of CO£ evolution from the effluent hold tank, This was accom-
plished by sealing the tank and purging with air at a constant measured
rate. Orsat analyses of the purged gases provide direct data on CO2
evolution and O2 absorption rates at the liquid surface in the tank.
Table 6-1 illustrates the type of information obtained from such
material balances. Among other things, the material balances show
that: (1) under normal operating conditions (stoichiometry of 1.25
based on SO2 feed) about 70 percent of the total limestone dissolution
occurs in the scrubber; and, (2) most of the oxidation occurs in the
hold tank, about 60 percent of the total with a TCA, and 80 percent of
the total with a multigrid scrubber.
The reaction scheme consistent with the data from these material
balances shows two principal reactions occurring in the scrubber:
SO2 + H2O *- H+ + HS03" (6-1)
H+ + CaCO3 ^ Ca++ + HCO3" (6-2)
6-2
-------
GAS OUT
SULFITE OXIDATION =21%
LIMESTONE UTILIZATION = 70%
STOICH. RATIO BASED ON SO2 ABSORBED = 1.45
STOICH. RATIO BASED ON SO2 FEED = 1.13
-*— h
300 cfm 1^
@125°F
-U
78%
SO2 REMOVAL
S02
7.25 Ib/hr
I
GAS IN 1
— '"**¥
^
A
11.6
ft/sac
3200 ppm 1
280° F I j
19.5gpm(L/G = 65)
°2
CO2
LIQUID PHASE SOLID PHASE
SO3 2820 mg/l 94 mg/g
SO2 774 281
CO2 247 124
Ca 600 314
Mg 770
= 12.4%
= 15.8%
13.5 Ib/hr LIMESTONE
CaCO3 =94.9%
MgCO3 = 4.9%
INERTS=1.7%
f
I AIR
\ pH = 5.4 / 0-56 cfm
LIQUID PHASE
SO3 3090 mg/l
SO2 1370
CO2 151
Ca 774
E. H. T. /
5 min. /
4% SOLIDS
C^/ pH = 5.8
^sO " ""*•
Figure 6-1. Operating Data for Limestone Feed
6-3
-------
Table 6-1
TYPICAL PERFORMANCE CHARACTERISTICS
OF TCA AND MULTIGRID SCRUBBERS AS DETERMINED
BY MATERIAL BALANCE
Absorbent:
Inlet SO2 Concentration:
Effluent Residence Time:
Stoichiometric Ratio, moles CaCO3/
mole SO2 feed
E. H. T. O2 Absorption Rate,
acfm @ 93°F
E. H. T. CO2 Evolution Rate,
acfm @ 93°F
SO2 Make Per Pass, mg/1
CaCO3 Dissolution in E. H. T. , mg/1
CaCO3 Dissolution in Sc rubber, mg/1
Dissolution in Scrubber, % of total
Total Oxidation, mole %
SO2 Oxidized in Scrubber, mg/1
Oxidation in E. H. T. , % of total
SO2 Ppted in Scrubber as CaSO^, mg/1
803 Ppted in Scrubber as CaSO4, mg/1
L/G, gal/mcf
Scrubber Effluent pH
Scrubber Feed pH
Limestone
2800 ppm
mn.
TCA
4% Solids
1. 13
0. 035
0. 084
Multigrid
16% Solids
1.25
0. 035
0. 062
2.5
0. 014
0. 046
630
357
626
64
21
57
57
-23
380
65
5.4
5.8
774
316
814
72
27
44
78
218
184
48
5. 8
6.3
863
259
1090
81
23
139
30
454
294
48
6.2
6.4
6-4
-------
SC>2 absorption by Equation 6-1 is enhanced as a result of the dissolution
of limestone by Equation 6-2. Dissolution of CaSO3 does not occur as
long as sufficient limestone is present to maintain the scrubber effluent
pH above 5.4 (see Table 6-1). Dissolution of limestone by Equation 6-2
is completed in the scrubber effluent hold tank where it raises the pH.
As a consequence of the increased pH the bisulfite equilibrium is
shifted and sulfite is precipitated. In addition to Equation 6-2, the
primary reactions in the effluent hold tank are:
HS03- ^=r H+ + S03= (6-3)
S03= + Ca++ - — CaS03 (6-4)
The net unreacted species from the above sequence of reactions
(Equations 6-1 through 6-4) are H"1" and HCO3" which build up in the
liquor returned to the scrubber at high pH (~6. 3). The third major
reaction in the scrubber thus occurs when the pH is again dropped by
Equation 6-1; the lower pH shifts the bicarbonate equilibrium irrever
sibly to CC>2f completing the reaction cycle:
H*- + HC03- - — C02 H- H20 (6-5)
The material balances show that about 90 percent of the CC>2 is
released in the scrubber by Equation 6-5. The remainder is evolved
from the effluent hold tank.
6-5
-------
6.3 IMPROVEMENT OF LIMESTONE UTILIZATION
The slowest reaction in the reaction cycle (Equations 6-1 through 6-5)
is the dissolution of limestone (Equation 6-Z). Two important con-
sequences result: (1) the rate of limestone dissolution controls the
reactions that follow it in sequence; and, (2) the extent to which the
dissolution and precipitation reactions go to completion in the effluent
hold tank is determined by the kinetics of limestone dissolution in the
effluent hold tank.
The rate of CaSO3 precipitation is controlled by and directly related
bo the CaCO3 dissolution rate. The kinetics of the overall rate-
Limiting reaction (Equation 6-2) can thus be examined by measurements
:>f the rate of disappearance of SO2 in the effluent hold tank. This was
ione in the pilot plant by operating at various effluent hold tank residence
times, percent solids and limestone stoichiometries. Figures 6-2 and
6-3 summarize these data, showing CaSOg precipitation rate, r(mg
5O2/liter/min. ), as a function of SO2 concentration in the liquid phase,
CSQ? (mS SO2/liter), and limestone density in suspension within the
Dackmixed tank, [CaCOj] in g CaCOj/liter. The analysis indicates
tiiat the overall reaction can be approximated by a rate expression
ihat is second order in sulfite concentration and first order in CaCO3
suspension density.
The high sensitivity of the dissolution/precipitation reaction to SO2
concentration suggests that greater total conversion would be expected
in a reactor without backmixing. This can be shown mathematically
yy integration of the standard reactor design equation with the indicated
6-6
-------
30
10 --
fc
•
o
8 J-
LIMESTONE PARTICLE SIZE:
O 7% GREATER THAN 325 MESH
A 16% GREATER THAN 325 MESH
O
O
1.0 --
0.3
O
O
H 1 I I II
50
100
1,000
3,000
•SO
Figure 6-2. Rate of Sulfite Precipitation in Backmixed
Effluent Hold Tank as a Function of SO2
Concentration in the Liquid Phase
6-7
-------
4x10
-3
l
-------
second order rate expression for backmixed and plug-flow modes of
operation. An experimental comparison of these two reactor types
was carried out using vertical pipes arranged in a series of U-tubes
as a plug-flow hold tank. As shown in Table 6-2 the normal 65-75
percent limestone utilization characteristic of backmixed effluent hold
tanks was increased to 85-90 percent by plug -flow. Scrubber feed pH
was likewise consistently higher in plug-flow tests.
6.4 FATE OF MgCO3 IN LIMESTONE FEED
The limestone used in the pilot scrubber is the same as that used at
the Shawnee test facility, and contains about 5 percent MgCOg. Lime-
stone utilizations are calculated on the basis of the total CaCO3
content of the limestone. Since the MgCO3 component is present as
dolomite (CaCO3« MgCOg), a like number of moles of CaCO3 is bound
in this mineral form. Operating experience, based on Mg*+ levels
(ca. 700 ppm) observed in the closed-loop scrubbing liquor and petro-
graphic analysis of the unreacted spent solid, indicates that a maximum
of about 10 percent of the dolomite component of the limestone feed
dissolves. The remaining solid dolomite passes through the scrubber
without participating in any scrubber reactions. Thus, the utilization
achievable with a dolomitic stone is less than that which can be expected
with a pure limestone. Likewise, the spent sludge from the scrubber
contains no leachable
6. 5 CONTROL OF SULFITE SCALING
Insofar as limestone dissolution kinetics controls the steady state
liquor composition within the scrubber system, the rate of sulfite
6-9
-------
Table 6-2
COMPARISON OF BACKMIXED AND
PLUG-FLOW EFFLUENT HOLD TANK DESIGNS
Scrubber Type: Multigrid
Limestone: Shawnee (Fredonia), 16% f325 mesh
Backmixed Plug- Flow
SO2 Feed Rate, Ib/hr 4.84 4.93
Limestone Feed Rate, Ib/hr (40% Slurry) 17.4 17.6
E. H. T. Residence Time, min. 6 6
Slurry Circulation Rate, gpm 12.7 12.7
Slurry Solids, % 77
Scrubber Feed pH 5.9 6.3
Scrubber Effluent pH 5.6 6.0
Scrubber Effluent Temperature, °F 127 134
Scrubber Feed Temperature, °F 126 124
SO2 Feed Concentration, ppm 2930 3300
Stoichiometry (based on SO2 feed) 0. 87 0. 87
Stoichiometry (based on SO2 absorbed) 1. 33 1.14
Percent SO2 Removal 65.5 76.1
Limestone Utilization, % 75 88
Sulfite Oxidation, mole % 17 10
Ca++ in Scrubber Feed, mg/1 .440 100
6-10
-------
precipitation (Equation 6-4) exceeds that of limestone dissolution
(Equation 6-2) and super saturation by sulfite cannot occur. As indi-
cated by the kinetic analysis, the limestone dissolution rate is
accelerated by increasing the SO2 concentration or suspended limestone
density, or by reducing the limestone particle size. Should the lime-
stone dissolution rate locally exceed the precipitation rate of CaSC>3,
super saturation and scaling of CaSO3 can occur. Tests with the pilot
scrubber at high limestone feed stoichiometries (which increase
suspended limestone density) at L/G of 65 and 3000 ppm SO, in the
inlet flue gas (which determine the effective SC>2 concentration in the
scrubber liquor) showed excessive scaling of the scrubber walls by
CaSOj when the scrubber effluent pH rose above 6.2. This condition
corresponded to a stoichiometric ratio of 2. 5 moles CaCO3/mole SO2
feed, with 93 weight percent of the limestone particles less than 325
mesh. As shown by Table 6-1, the proportion of CaCC>3 dissolution
occurring in the scrubber is increased under these conditions. In
some cases material balances showed as much as 90 percent of the
total limestone dissolution occurring in the scrubber at high feed
stoichiometry. In such situations nearly all of the reaction sequence
is completed within the scrubber rather than delaying precipitation
until the slurry reaches the effluent hold tank as is normally the case.
Reduction of the effluent hold tank residence time will, according to
the reaction scheme discussed, buildup the steady-state sulfite
concentration in the liquor, accelerating the rate of dissolution to the
point where it may no longer control the sulfite precipitation reaction.
The scrubber feed can then become supersaturated. Experience in-
dicates that this condition is not reached until the SC>2 concentration
6-11
-------
Ln the effluent hold tank liquor exceeds 800 mg/1 which occurs only
it very short residence times (less than 5 minutes). On this basis,
residence times shorter than 7 minutes are not recommended with
Limestone scrubbing, particularly when oxidation is inhibited.
D.6 CONTROL OF SULFATE SCALING
A.S shown by the material balances, the dissolution of limestone in the
scrubber increases the concentration of Ca by about 150 ppm at
L/G of 65. When the scrubber feed is saturated with CaSO4* 2H2O
ihis represents a 25 percent increase in the calcium concentration
it the bottom of the scrubber. Thus, even if no oxidation occurs, the
scrubber effluent will be supersaturated with calcium sulfate. In a
TCA scrubber, maximum super saturation occurs at the first stage
support grid and rapid scaling at this point has been experienced in
;he scrubbers with either limestone or lime feeds. In the absence
Df fly ash, the grid openings are plugged within 50 hours at L/G of
52. The presence of fly ash significantly reduces the rate of scaling,
is does increasing L/G. Other factors, such as humidification of
Flue gas (128°F quench) and CaSO4 seeding, had no appreciable effect
when scrubbing flue gases containing 3000 ppm SO?. It was concluded
;hat L/G's must exceed 65 for TCA operation with saturated feed, and
;he presence of fly ash in the slurry is a positive benefit from the
scale-control point of view.
A. limestone or lime scrubber is customarily visualized as operating
with two coexisting, but independent, liquid/solid systems involving
precipitated CaSO3' 1 /2 H2O, crystallized CaSO4* 2H2O, and saturated
6-12
-------
mother liquor. Any sulfate generated by oxidation in the scrubber
must, according to this view, build up in the recirculating solution
until it is saturated with calcium sulfate. Crystallization of gypsum
thereupon removes the sulfate from the system.
It was shown by Imperial Chemical Industries (I.C.I. ) in 1951 that
significant amounts of calcium sulfate are incorporated into the calcium
sulfite crystal lattice when it precipitates from scrubber liquors. The
amount of sulfate that could be so incorporated was experimentally
established at 0. 225 mole SOj/mole SO? for precipitation in a limestone
slurry. The resulting compound was referred to as a "solid solution. "
Although unsaturated operation of the I. C. I. Bankside scrubber was
never reported, the formation of the solid solution clearly affords an
alternate mechanism by which CaSO^ can be purged from the system
without crystallization. Thus, if the rate of generation of sulfate by
oxidation is less than the rate at which sulfate is incorporated by the
precipitating CaSO,, the recirculating liquor is no longer constrained
to saturation by calcium sulfate.
That a closed-loop scrubber can operate unsaturated by the above
mechanism was first noted during the plug-flow hold tank experiments
at the EPA pilot facility. Unsaturation was evidenced by steady state
Ca"1"1" concentrations of only 100 ppm in the scrubbing liquor (about
1/6 the level of saturation with CaSO^.* 2H2O). Material balances
confirmed closed-loop operation and direct dissolution of solid
CaSO4'2H7O in the scrubber liquor proved unsaturation. Tests of the
solid solution verified the characteristics reported by I. C.I. , i. e. , no
detectable gypsum by x-ray diffraction, and no gypsum extractable by
water. Therefore, the sulfate shown to be present by chemical analysis
6-13
-------
(9-17 percent of total sulfur) does not exist as a separate phase.
Confirmation of the presence of sulfate by chemical analysis was made
by dissolution of the solid solution in HC1, extraction and recrystalli-
zation of pure gypsum from -water in the amounts indicated by analysis.
Intense x-ray peaks at d-values of 2.67 and 5, 34 have also been noted,
which are similar to those of calcium thiosulfate, Since thiosulfate
•was shown to be absent by chemical analysis, these peaks are probably
characteristic of the solid solution.
Tests with the backmixed effluent hold tank showed that it could also
be operated in the unsaturated mode by sealing the top of the tank to
prevent air contact with the slurry. As indicated by the material
balances, most oxidation occurs in the effluent hold tank. Sealing the
tank eliminates this source and the total oxidation is reduced to about
10-15 percent (CC^ evolution from the tank provides a self-generating
blanket). Unsaturation of the liquor occurs when the oxidation falls
below the level corresponding to the maximum 803/SO2 ratio in the
solid solution.
The oxidation level at which unsaturation occurs is shown in Figure 6-4
where calcium concentrations of the scrubber feed liquor are plotted
over the full range of oxidation in which the pilot plant has been oper-
ated. It is clear that calcium concentrations correspond to saturated
CaSO^' 2H2O above 19 percent oxidation and drop progressively as
oxidation is reduced below that level. The relationship between oxida-
tion and the SO3/SC>2 mole ratio in the solid is given by:
Sulfite Oxidation = SO2 Oxidized/(SO2 Oxidized + SO2 Precipitated)
= Ratio/(Ratio + 1)
where (6~6)
Ratio = SO3/SO2 mole ratio in solid
6-14
-------
800
700--
600--
Z
UJ v
U £*
z o
O ^
u O
a I
5 £
—• *
< U
U vi
400--
300--
200 ••
100 ••
O
O
.0.
I.C.I. SOLID SOLUTION
SATURATION
SATURATED LIQUOR
0_Q_
-8-
O TCA SCRUBBER - STIRRED TANK
A MULTIGRID SCRUBBER - STIRRED TANK
D MULTIGRID SCRUBBER - PLUG FLOW
30
40
50
60
70
80
MOLE PERCENT SULFITE OXIDATION
Figure 6-4.
Calcium Sulfate Saturation as a Function of Sulfite Oxidation in Scrubbers
Operating -with Limestone: Chloride Concentration = 0
-------
Therefore, 19 percent oxidation corresponds to the maximum SO3/SC>2
mole ratio of 0. 23, which is in good agreement with the value obtained
by I. C. I. from laboratory measurements.
Elimination of sulfate super saturation of the scrubber effluent liquor
by operation in an unsaturated mode has been demonstrated in the
TCA pilot scrubber with both limestone and lime feeds. In either
case, the limitation of oxidation below 19 percent is the critical con-
dition for desaturation. Other factors that may affect solid solution
formation and the ability to operate unsaturated, particularly in the
presence of high concentrations of chloride, are less well defined.
The current work at the EPA pilot facility is focused in this area.
6. 7 UTILIZATION IN LIME SCRUBBERS
When CaO is the scrubber feed rather than CaCOo» dissolution kinetics
is no longer a constraint and higher utilizations can be achieved in a
stirred tank of given residence time. Comparison of Figure 6-1
(limestone) and Figure 6-5 (lime) illustrates the difference at scrubber
feed pH of about 6 for 5 minutes residence time and 4 percent solids.
For lime scrubbing, raising the scrubber feed pH will increase the
amount of CO2 absorbed from the flue ga,s and utilization drops as
Ca(OH)2 is recarbonated to CaCO^. This side reaction must be limited
if the advantage of CaO over CaCO3 is to be maintained. Table 6-3
summarizes a series of tests made in the pilot scrubber to determine
the effect of feed pH on lime utilization. The values of utilization
shown are based on the solids analysis and do not reflect losses in
the slaking step, -which average about 4 percent of the total CaO content
6-16
-------
GAS OUT
SULFITE OXIDATION = 26%
LIME UTILIZATION =98%
STOICH.RATIO BASED ON SO2 ABSORBED = 1.0
300 cfm
@125° F
GAS IN
u-
R1%
>-
S02 REMOVAL
SO2
6.10lb/hr
I
I
T
V
L
'
I
*
I
t
11.5
ft/sec
L
19.5gpm(L/G°65)
LIQUID PHASE SOLID PHASE
SO3 2980 mg/l 1 25 tng/g
S02 216 378
C02 29 7
Ca 585 305
Mg 546
5.01 Ib/hr LIME
CaO = 96%
MgO = 04%
/
LIQUID PHASE
S03
S02
CO?
Ca
3300 mg/l
1110
42
boo
E. H. T. /
mm. /
4% SOLIDS
/
/
*-v pH=6-1 .
^ki *
Figure 6-5. Operating Data for Lime Feed
6-17
-------
Table 6-3
EFFECT OF FEED pH ON LIME SCRUBBER OPERATION
Scrubber Type: TCA
Pressure Drop: 6. 5 in. H^O
Effluent Residence Time: 5 min.
Percent Solids: 4%
Scrubber Feed pH
i I 1 12.
CO2 in Scrubber Effluent, mg/1 42 43 47 68
CO2 in Scrubber Feed, mg/I 29 26 25 8
CO2 Precipitated in E. H. T. , 13 17 22 60
mg/1
CO2 in Solid, mg/g 7 14 " 33 37
Lime Utilization, % 98 97 89 81
Ca++ in Scrubber Feed, mg/1 585 570 845 896
SO2 Make Per Pass, mg/1 506 569 500 570
ASO2 Across Scrubber, mg/1 732 866 674 600
ASO2 - Make Per Pass, mg/1 226 297 174 30
ACa Across Scrubber, mg/1 298 255 240 136
SO2 in Scrubber Feed, mg/1 216 188 64 33
Percent SO2 Removal 81* 78 80 84
Sulfite Oxidation, mole % 26 20 19 13
Mg++, mg/1 546 1000 40 5
Scrubber Effluent pH 4.9 5.1 5.1 5.0
* AP = 8. 5 in.
6-18
-------
of pebble lime. It is clear from Table 6-3 that recarbonation is
significant at inlet pH of 9 and above. Also, magnesium is precipi-
tated from the liquor at pH of 9. These facts indicate that the optimum
scrubber inlet pH1 s are in the range of 7-8.
The data in Table 6-3 confirm the observation reported by others
that the primary scrubbing reaction in the lime system is the dis-
solution of CaSC>3:
CaS03 + H+ — Ca++ + HS03~ (6-7)
This is evidenced by the fact that the change in SC>2 concentration in
liquid ( ASC>2) across the scrubber exceeds the SC>2 make per pass in
all cases together with the increase in Ca concentration across the
scrubber.
It is notable that the reaction shown by Equation 6-5, which contributes
to pH buffering in limestone scrubbers, does not occur in.a lime system.
Consequently the pH drops to a greater degree when SC>2 is absorbed.
As indicated in Table 6-3, change in liquor pH across the scrubber
(1-5 units) far exceeds the 0. 5 unit typical of limestone scrubbers.
This lack of buffering is presumed to be responsible for the generally
lower SC>2 scrubbing efficiency that has been experienced with lime
scrubbing at the EPA pilot facility, compared to limestone at a given
pressure drop.
At SC>2 feed concentrations in excess of 2500 ppm, the increase in
calcium concentration in the scrubber effluent resulting from CaSO,
dissolution (Equation 6-7) is responsible for CaSC>4 super saturation
6-19
-------
and TCA grid scaling in lime system, in the same manner as
dissolution (Equation 6-2) in limestone systems. The most effective
technique for dealing with this problem appears to be the unsaturated
mode of operation discussed in the previous section. Control of pH,
oxidation and Mg++ concentration to yield an unsaturated scrubber feed
and marginally saturated (to unsaturated) scrubber effluent is an
achievable goal for either limestone or lime scrubber systems.
6.8 CONCLUSIONS
The following conclusions have been drawn from the limestone and
lime results of the TCA pilot-plant testing at Research Triangle
Park, -with inlet SO2 feed concentrations of 3000 ppm:
• For limestone scrubbing, control of the scrubber effluent
pH below 6. 2 will prevent calcium sulfite scaling.
• Optimum scrubber inlet pH (reasonable SO^ removal and
reasonable lime utilization) for lime scrubbing is in the
7-8 range.
• Limestone dissolution kinetics are improved by plug flow
reaction; higher utilizations can be achieved by effluent
hold tank designs such as U-tubes or a series of stirred
tanks that approximate plug flow.
• For limestone scrubbing, dissolution of CaCO3 is the rate
controlling step for SO^ absorption. For high-calcium lime
scrubbing, dissolution of CaSOj is the rate controlling step.
• Hold tank residence time must exceed 5 minutes in a lime-
stone system. 10 minutes appears to be a good choice.
For a lime system, 5 minutes appears adequate.
• The scrubber effluent will always be supersaturated with
CaSO4* ZHoO when a scrubber is operated -with saturated
feed.
6-20
-------
• Super saturation is maximum at the first stage TCA support
grid and rapid scaling of the grid by calcium sulfate occurs
at liquid-to-gas ratios less than 65 gal/mcf with no fly ash
in the system. The liquid-to-gas ratio must exceed 65 for
reliable operation with saturated calcium sulfate feed.
• The presence of fly ash reduces the rate of scaling by
calcium sulfate.
• Sulfate scaling can be eliminated by operating a scrubber
in the unsaturated mode. Closed liquor loop unsaturated
operation can be achieved in a chloride-free scrubber by
reducing oxidation below 19 percent. Under these conditions
the sulfate generated by oxidation is purged entirely as solid
solution (i. e. calcium sulfate in calcium sulfite crystal
lattice).
• For limestone scrubbing, oxidation is a controllable variable
within the limits required for unsaturated operation. It can
be reduced by eliminating air contact in the effluent hold
tank (e.g. , using sealed stirred tank or plug flow tank).
Further reduction can be attained by circulating high percent
solids (which affects the amount of liquor circulating through
the clarifier or filter).
• Sulfate bound as solid solution is not extractable from the
spent sludge by water leaching. Therefore, the potential
for water pollution is reduced.
• The dolomitic component of limestone feeds is essentially
inert and leaves the scrubber in the same form within the
sludge.
6-21
-------
Section 7
FINDINGS TO DATE
Based on Shawnee and EPA pilot facility testing to date, the following
items are preliminary findings regarding (1) scrubber design and oper-
ating parameters, (2) process chemistry/advanced concepts, and (3)
equipment/mate rials/instrumentation.
Scrubber Design and Operating Parameters
The TCA and the venturi/spray tower have potential for reliable op-
eration for both limestone and lime scrubbing. Due primarily to noz-
zle wear problems, insufficient data has been generated on the Marble-
Bed scrubber to assess its reliability potential under Shawnee conditions.
Conditions for best overall potential for reliable limestone scrubbing
operation with configurations comparable to the Shawnee scrubber sys-
tems are: an effluent residence time equal to or greater than 10 min-
utes and a percent solids recirculated equal to or greater than 10 per-
cent, for a liquid-to-gas ratio equal to or greater than 60 gal/mcf.
Variables which appear particularly important are liquid-to-gas ratio
and scrubber superficial gas velocity. Shawnee liquid-to-gas ratios
should be considered minimum values, since there is evidence that
values higher than attainable at Shawnee would further minimize the
potential for scrubber scaling and increase SC>2 removal and limestone
7-1
-------
utilization. Superficial velocity is considered important since there
appears to be a certain value above which excessive demister entrain-
ment occurs. For the Shawnee demister/Koch tray configuration in
the TCA, this maximum velocity appears to be about 8 ft/sec. The
spray tower demister/wash system limitation appears to be about 6
ft/sec.
For lime scrubbing, the venturi/spray tower conditions which have
given good results in the operation to date are: (1) a scrubber inlet pH
of 8, (2) an effluent residence time of 12 minutes, (3) a percent solids
recirculated of 8 percent (40 percent of solids is fly ash), and (4) a
spray tower liquid-to-gas ratio of 64 gal/mcf. In addition to the im-
portance of liquid-to-gas ratio and superficial gas velocity discussed
for limestone systems, inlet scrubber pH is considered important in
lime systems; the value of 8. 0 has resulted in good performance.
The following factors should be effective in minimizing pluggage prob-
lems associated with chevron-type demisters: (1) minimize scrubber
superficial velocity consistent with cost, turndown, space and other
factors, (2) utilize a wash tray between the uppermost scrubber stage
and demister, and (3) wash demister from upstream (bottomside) di-
rection with a mixture of clarified liquor and all available makeup water
and assure complete surface irrigation.
All three scrubbers have been effective dust collectors, since they have
reduced inlet particulate loadings from about 2-5 grains/scf to 0. 01-
0. 04 grains /scf.
7-2
-------
Limited cascade impactor distribution data for the TCA indicates
effective particulate removal in the submicron range. For example,
at the highest pressure drop tested (9. 7-9. 9 inches f^O), TCA re-
moval efficiencies were about 95 percent in the 0. 1 to 0. 2 micron di-
ameter size range; decreased pressure drops resulted in significantly
reduced removal efficiencies. Further testing, however, is neces-
sary to validate this apparent high performance in the submicron range.
It should be noted that combined scrubber collection of SC>2 and fly ash
(no mechanical or ESP collection upstream of the scrubbers) seems to
have a major disadvantage. Namely, that roughly 30-50 percent of
slurry solids is fly ash, -which appears to be the most abrasive solid
component. Rubber lined components have shown little evidence of
wear. The following components, however, appear particularly sensi-
tive to erosion: TCA spheres, TCA support grids, slurry spray noz-
zles, centrifuge and venturi throat guide-vane assembly. Work with
vendors is in progress to improve these components.
Although it is too early for a definitive finding concerning the relative
advantages of lime versus limestone, the following are preliminary
observations: (1) lime utilizations are substantially higher than lime-
stone utilizations; in the order of 90-93 percent for lime vs. 65-75 per-
cent for limestone, (2) pH control appears easier with lime since pH
is more sensitive to alkali addition than in the limestone system, where
the required excess is higher, and (3) preliminary at least, lime opera-
tion seems less prone to plugging and scaling problems, although rea-
sons for this apparent difference are not well understood. As mentioned
previously, the moisture content of the discharge slurry is higher than
desired for the lime testing.
7-3
-------
Process Chemistry/Advanced Concepts
Generally operation at Shawnee for both limestone and lime has yielded
scrubbing liquors that were either saturated or supersaturated with re-
spect to CaSO4- 2H2O and CaSO3' 1/2H2O. The following combination
of process variables appear effective in minimizing scaling potential:
(1) high liquid-to-gas ratios, (2) greater than 1 percent CaSO^/2H2O
seed crystals in slurry, and (3) effluent residence times greater than
or equal to 10 minutes. To minimize demister scaling, dilution of de-
mister or wash tray liquor with fresh water appears to be effective.
Research on the small pilot scrubber at EPA-RTP has indicated that
it may be possible to operate in an unsaturated sulfate mode with re-
spect to CaSOj' 2H2O, with potential freedom from gypsum scaling.
Low oxidation is essential to operation in this mode.
Plug flow effluent hold tank designs appear substantially more effec-
tive than conventional stirred tanks in enhancing limestone utilization.
They also help achieve unsaturated operation by reducing hold tank oxi-
dation. Further work is necessary.
It has been determined that the dolomite fraction of dolomitic limestone
is very slowly soluble and does not supply alkalinity under normal scrub-
ber operating conditions.
With properly calcined dolomite, magnesium does go into solution and,
when using dolomitic limes, SC>2 removal efficiencies are substantially
enhanced; this appears to be due to the increased amounts of solution
alkalinity in the form of magnesium sulfite (relative to sulfurous acid).
7-4
-------
In limestone systems it is important to avoid high stoichiometries with
corresponding high pH's, because CaSC>3* l/ZJ^O formed in the scrubber
precipitates and can lead to scaling. Lower limestone stoichiometries
are also effective in enhancing limestone utilizations, since limestone
dissolution is much more rapid at the lower pH's.
It appears practicable to substantially increase oxidation by air sparg-
ing in the stirred hold tank; this could lead to improved sludge charac-
teristics. Further evaluation is necessary.
The scrubbers are effective in absorbing HC1 from the flue gas (coal
has from 0. 1 to 0. 3 percent Cl) leading to relatively high liquor chlor-
ide levels during closed loop operation. Although the effects of chloride
are not well understood, they appear, for example, to significantly de-
crease scrubber pH with subsequent loss in absorption performance.
Equipment/Materials/Instrumentation
Rubber lined, variable speed, centrifugal pumps with Hydroseals
or Centriseals are reliable for slurry service.
For a fuel oil fired reheater, an isolated or external combustion
chamber should be used to avoid quenching of the reheater flame
by cold flue gas.
The rubber linings of the scrubbers, pumps, pipes, etc. , give
satisfactory erosion and corrosion resistances for slurry and flue
gas (quenched) services. Flakeline linings on the effluent hold tanks
and clarifiers are also satisfactory.
7-5
-------
Type 316L stainless steel gives irmch better corrosion resistance
than type 304L in slurries containing chlorides.
A centrifuge gives satisfactory de-watering capability. However,
erosion on the metal surfaces is a major problem.
The dewatering and cake discharge capabilities of the filter, using
a nylon filter cloth, are satisfactory. However, the useful life of
the nylon filter cloth tested to-date is unsatisfactory.
Uniloc Model 321 submersible type pH meters gives better perfor-
mance than Model 320 in-line flow-through types.
The performance of Ohmart radiation-type and bubble-type density
meters is unsatisfactory. The performance of Dynatrol density
meters has thus far been encouraging.
Control valves in slurry service have generally been unsatisfactory.
Variable speed pumps should be used for slurry flow control.
7-6
-------
Section 8
. FUTURE TEST PLANS AT THE SHAWNEE FACILITY
Planni ig has been performed to formulate a follow-up test program
to the present Shawnee activities. It should be noted that these plans
are quite preliminary in nature; they require additional funding which
has not officially been authorized and hopefully will be modified based
on utility and vendor inputs. Figure 8-1 presents the overall test
schedule for advanced limestone and lime testing which will extend
the present program through June 1976. The folio-wing are the ob-
jectives of the advanced program:
Evaluate the effectiveness of automatic control systems
for scrubbers experiencing widely varying flue gas flow
rates and inlet SO? concentrations.
Investigate advanced process variations which offer
promise of enhanced reliability by completely eliminating
potential for gypsum scaling.
For limestone systems, evaluate process variations which
offer potential for substantial increases in limestone utili-
zation, with subsequent decrease in sludge production.
For limestone and lime systems determine the upper limit
of SC>2 removal efficiencies within the constraints of facility
scrubber configuration, pressure drop and liquid flow rate
limitations. This may be important if future air pollution
regulations require more stringent SO2 control for power
plants.
8-1
-------
00
I
tv
TEST PROGRAM FUNCTIONS
PRESENT TEST PROGRAM
LIMESTONE LONG TERM TESTING (TWO 2000 Hi. TESTS)
LIME LONG TERM TESTING (TWO 2000 Mr. TESTS)
LIME FACTORIAL TESTING
LIMESTONE ADVANCED TESTING
FACILITY CAPABILITY SCREENING TESTS (CONTROL. OXIDATION)
FACILITY MODIFICATIONS (OXIDATION, EHT. CONTROL)
VARIABLE LOAD TESTING (pH CONTROL. 2500 Hn.l
LIMESTONE SLUDGE FIXATION
SEALED HOLD TANK (TWO 500 Hi. TESTS!
PLUG FLOW HOLD TANK TESTS (THREE 500 Hr. TESTS)
ADDITIVE TESTS (THREE 500 Mr. TESTS)
MAXIMIZE OXIDATION TESTS (THREE 500 Hi. TESTS)
MAXIMIZE UTILIZATION (THREE 250 Hr. TESTS)
MAXIMIZE SOj REMOVAL EFFICIENCY (THREE 500 Hi. TESTS)
ADV. CONCEPT RELIAB. TEST (1500 Hr CONST.. 1500 Hrv VARIABLE!
LIME/DOLOMITIC LIME ADVANCED TESTING
FACILITY CAPABILITY SCREENING TESTS (CONTROL. OXIDATION)
FACILITY MODIFICATIONS
VARIABLE LOAD TESTING (LIME)
LIME SLUDGE FIXATION
LIME TESTING
Plug Flow HOW T«nk (Two 5OO Hr. Ton)
Seated HM Tank (Two 500 Hi. Tan)
OOLOMITIC LIME TESTING
Salad Hold Tmk (One 500 Hi. Tall
Open Hold Tank (Two 500 Hi. Tntll
Pluj Flow Hokl T.nk (One 500 Ht. Tail
MAXIMIZE SO2 REMOVAL EFFICIENCY (THREE 500 Hi. TESTS)
MAXIMIZE OXIDATION TESTS (THREE 500 Hi. TESTS)
ADVANCE CONCEPT RELIABILITY TESTING (SAME AS LIMESTONE)
GENERAL PUBLICATION REPORTS
1973
N| 0
1974
J F M | A JMJ J J A S\O\ N\D
,,^.«.*, . f.«. ..
4
"_
4
1975
J|F|M|A|M|J J | A | s O|N|D
1976
J|F|M|A|M|J J)A s | o N]O
LEGEND
• •
^^m
•M^M
4 4
• TESTING OR MODIFICATION EFFORTS
• ENVIRONMENTAL EFFECTS EVALUATION
IM
•
*
Figure 8-1. Preliminary Schedule for Advanced Shawnee Program
-------
Investigate process variations and/or de-watering equipment
•which are capable of producing a more acceptable sludge
product.
Perform long term reliability testing on advanced limestone
and lime process variations which offer substantial improve-
ment over present process variations in one or more of the
following areas: potential reliability, limestone utilization,
SO£ removal efficiency and improved sludge product charac-
teristics.
In order to attempt to achieve these objectives, the following rep-
resents our preliminary thinking regarding the elements of the
advanced program:
Limestone Testing
For limestone, based on the results of the present test program, a
single scrubber train will be selected for advanced testing; in all
probability this will be the Turbulent Contact Absorber. Design of
system modifications will be initiated during early 1974 with the aim
of supplying a plug flow hold tank and other modifications necessary
for future testing. Facility capability screening tests will be the
first tests performed following the present test program. These
will attempt to evaluate the facility's present capability (without
modifications) in terms of (1) supplying the necessary oxidation
capability (via air sparging in the present effluent hold tank) to
produce an improved gypsum-rich sludge product and (2) ability of
the present pH control-system to effectively adjust limestone feed
rates to correspond to wide variations in flue gas and inlet SO?
concentration variations. Approximately two months are planned
for the necessary system modifications.
8-3
-------
Subsequent to this, variable load testing for a period of approximately
2500 hours will be performed on the process variation subjected to
earlier reliability testing. The pH control system selected will
attempt to keep the chemistry of the system in balance despite the
wide variations in SC>2 inlet concentrations associated with normal
Shawnee coal supplies, and a pre-programmed flue gas flow rate
daily history (via duct damper settings) which will simulate a widely
varying boiler output.
At the conclusion of this test, the effluent hold tank will be sealed
and process parameters selected to attempt to operate with the
lowest practicable oxidation; this will hopefully enable operation
in the "unsaturated-gypsum" mode which has potential for elimina-
ting gypsum scaling as a potential operating problem. After comple-
tion of this testing, the plug flow hold tank -will be tested as an
alternative to the present stirred tank; this configuration offers the
following potential advantages: (1) elimination of hold tank oxidation
allowing greater potential for unsaturated operation, (2) improved
limestone utilization and/or SO^ removal efficiencies, and (3)
potentially improved pH control, since the more sensitive scrubber
outlet pH can be used as the limestone feed rate control variable.
In another attempt to increase limestone utilizations and/or SC>2
removal efficiencies, additives will be tested for their ability to
increase limestone dissolution rates with corresponding improve-
ments in performance.
8-4
-------
Subsequent to this, process variations will be selected which offer
the potential for producing sludge with more desirable properties.
For example, oxidation of scrubber slurry will be enhanced (via
air sparging in a hold tank or by a separate oxidizer utilizing
Japanese technology) to produce a predominately gypsum/fly ash
product which should have vastly improved settling characteristics
over calcium sulfite-rich sludges. Advanced de-watering equipment
and liquor bleed stream treatment equipment might also be tested
with the aim of producing a more acceptable landfill material with-
out the need for a fixation treatment process.
Based on the results obtained prior to this point, two separate test
series will be performed; they will attempt to find conditions
associated with achieving maximum limestone utilization (minimum
sludge production) and maximum SO2 removal efficiencies, respec-
tively, consistent with constraints imposed by scrubber type, pressure
drop limitations and reasonable process economics.
The last scheduled limestone run would involve long term reliability
testing on an advanced process variation selected based on results
of prior evaluation and optimization testing. The process variation
will be selected with the aim of maximizing reliability, limestone
utilization and SC>2 removal efficiency, and improving sludge charac-
teristics consistent with reasonable process economics.
Lime Testing
A single scrubber type will be selected for lime testing; based on
8-5
-------
results to date this will probably be the venturi/spray absorber. As
discussed under limestone testing, facility capability screening
tests and system modifications will be performed during mid-1974
•with subsequent performance of variable load testing using hydrated
lime [ Ca(OH)2 ] as the alkali.
Subsequent testing will be divided into hydrated lime and slightly
dolomitic hydrated lime [ Ca(OH)2« — 5% Mg(OH)2 ] test blocks.
This is planned based on the research data generated at the RTP
scrubber facility and communications with commercial vendors
•which indicate that dolomitic lime offers the following potential
advantages over low-magnesium lime: (1) higher SO., removal
efficiency potential due to higher concentration of the sulfite anion
in solution, and (2) higher potential for operation in the unsaturated-
gypsum mode, since higher concentrations of dissolved sulfates
have been shown to favor formation of CaSO^/CaSO^ solid solutions.
Two series of lime tests are planned which-will evaluate sealed
stirred tank and plug flow hold tank variations on unsaturated oper-
ation and,in the case of the plug flow hold tank,on SO? removal
efficiencies and lime utilization.
A similar test series will be performed using dolomitic lime as the
alkali; however, three effluent hold tank variations will be evaluated:
the sealed stirred tank, the plug flow hold tank, and the conven-
tional stirred tank.
8-6
-------
Subsequent to this testing, two test series will be performed with
the objective of maximizing SC>2 removal efficiencies, and improving
sludge characteristics, respectively* These will be similar in scope
to the limestone tests described earlier.
Finally, the last lime test would involve long term reliability testing
on an advanced lime (or dolomitic lime) process variation, selected
as the most promising based on a review of prior testing.
8-7
-------
Section 9
REFERENCES
1. Bechtel Corporation, EPA Alkali Scrubbing Test Facility:
Sodium Carbonate and Limestone Test Results, EPA Report
650/2-73-013, August 1973.
2. M. Epstein, et al. , "Limestone Test Results at the EPA
Alkali Wet-Scrubbing Test Facility at the TVA Shawnee Power
Plant, " presented at the Sixty-Sixth Annual Meeting of the
A.I.Ch.E., Philadelphia, Pa., November 11-1 5, 1973.
3. M. Epstein, et al. , "Test Results from the EPA Lime/Limestone
Scrubbing Test Facility, " presented at the Flue Gas Desulfuri-
zation Symposium, New Orleans, Louisiana, May 14-17, 1973.
4. Radian Corporation, A Theoretical Description of the Limestone-
Injection Wet Scrubbing Process, NAPCA (APTIC No. 22709 and
25446) Report, June 9, 1970.
5. A. Saleem, J. Air Pollution Control Assoc. , Vol. 22, No. 3,
March 1972.
9-1
-------
Appendix A
CONVERTING UNITS OF MEASURE
Environmental Protection Agency policy is to express all measurements
in metric units. When implementing this practice will result in undue
costs or lack of clarity, conversion factors are provided for the non-
metric units. Generally, this report uses British units of measure.
For conversion to the metric system, use the following conversions:
To Convert From
acfm
OF
ft
ft/sec
gal/mcf
gpm/ft
gr / scf
n
in. H2O
Ib-moles
Ib-moles/hr
Ib-moles/min
To
nm^/hr
°C
m
m/sec
min/m^
o
gm/m
cm
mm Hg
gm-moles
gm-moles /min
gm-moles /sec
Multiply By
1.70
subtract 32 then
'* 1.8
0.305 •
0. 305
0. 134
3.79
40.8
2.29
2.54
1.87
454
7. 56
7.56
A-l
-------
Appendix B
GRAPHICAL OPERATING DATA FROM LIMESTONE
RELIABILITY VERIFICATION TESTS
B-l
-------
BOILtR WMNTENANCl I
VENIURI & SFTtAV tOWER INLET (LAB)
• VENTUII & SKAY TOWEI OUTLET (LAB)
Gil Rut. • 30.000 icfrn 8 330 °F
Liquor RUB to Vanturi = 600 gpm
Liquor Rltt to Spny Tower = 1.200 gpm
Vinturi L/G = 27 pl/mcf
Sony Tower L/G= 53 jal/mcf
Sony Tower GUI Velocity • 7.5 ft/at
Vemuri Prasure Drop = 9 in r^O
E.H.T. Rnidenci Time - 12 min
No. of Spray Haderi = 4
Gn Inlet SQ2 Cone. • 2,500-3.200 ppm
12,000-2.200 ppm during 7/26 end 7/28)
Scrubber Inlet Liquor Temp. = 126-131 °F
Liquid Conductivity = 9.500 21.500 ji mhoi/en
Discharge ICIerHier& Centrifuge) Solids
Cone. * 57-65 wt %
eppnuimetely 1-1/4 moleXMg C03naibeen
mad it the tea facility. Prior to ttiil tinw a limntone
havingepproximately 5 mole % Mg C03 habainusad.
7/28 1 7/79 I
TESTTIMf, heun
VJO I
CALENMI OAV
£ a O
IS'.
11 =
, IN SOLUBLES (ASH)
/27 I 7/2B I 7/39 I '/30 I 7/3) I 8/1
14,000
12,000
10.000
_ 8,000
5 6,000
? -,000
2 2,000
• TOTAL DISSOLVED SOLIDS • rAAGNiSIUM (Mtj ** 1
^> CALCIUM [Co ** ) A SODIUM (f^i * 1
O 5UIFA1E (SO^" ) "7 POTASSIUM
-------
Gil Ran - 30.000 artm » 330 °F
Liquor Ran to Vinturi • 600 gpm
Liquor Rail to Spray Towir • 1,200 gpm
Vmlitri L/C - 27 nil/mri
SPTBY Tower L/G » 53 gal/md
Spray Tower Gai Vilociry - 7.5 rt/ac
Vinluri Prinuro Drop • 9 in r^O
E.H.T. HeudenceTime-Umin
No. ol Spriy Hearlan - 4
Gil Inlet S02 Cone. - 2.600-3.300 ppm
Soubbir Inlet Liquor Temp, • 128-132 °F
Llnuid ConductivilY- 22,000-27,000 u mhoi/cm
Dbclwrta ICIiritinl Cinlrifugi) Solidi
Cone.-55-65 wl%
CALENDAI DAY
SI:
. INSOIUUES (ASH)
f*'-
ls\ '
S ? 8
2= "
Si'
B|' "
8§I ..
IEII riMi. h«,
I l/v I
CALENDAI DAY
I a/ii l
14,000
M,000
11.000
10,000
B.OOO
6,000
7,000
0
.
; o
0
A
*
a
V
•
: * °
A
Q 0
TOTAL DISSOLVED SOllDS
CALCIUM (C. ** )
SULFATE (SO4 ' )
CHLORIDE (Cl ' )
MACNCSUM [Mg **)
SODIUM [Mo * )
POTASSIUM (K '1
SUlFIll (SO. " )
CARIONATKCO, '»
-
-
'
t*,ooo
11,000
10,000
1,000
6,000
1,000
0
i .
IESI TIME, teun
Ifl I 1/10
CALENDAX DAY
Figure B-] (Continued)
OPERATING DATA FOR VENTURI
RUN 506-1A
B-3
-------
KG IN «UtNl.Q-ZA
TEST TIME .ton
7/1 I f/J I
CAUNOAI DAY
14,000
11,000
10,000
3
_ 1,00X3
l" ..«>
li
E «.ooo
i
§ 2,000
C o
O CAlCIUM [Co ** > ^ MAGNESIUM (Mfl " ) '
D SULFAIE (SO/ ) ^ SODIUM (No M •
^ PorASstuM IK *)
• • SU1FITE BOj' ) • «
O CAtlONAIE ICOj' )
• -
"
-
A * A A
o o o o -
a • D n • •
14,000
12,000
10,000
i.ooa
4,000
4,000
2,000
0
Gn Riu • 25.000 icfm « JOO °F
Liquor Rne-UOOgpm
L/G-Mpl/mcf
Cm Velocity 9.8 WBC
E.H.T. Roidsna Tims = 20 min
Three Stagti, S in tfttmt/tUf
Go Inln S02 Cone. = 2.000-2.700 ppm
Scrubber Into liquor T«mp. • 121-1!7°F
Liquid Condunnrity - 11.000-20,000 Ji mhoi/cm
Dinhirjt [Chrifiirt Solid. Cone. = 26-M M %
•a
S7
&
_i Q_
6/1? I */»
CALENDAR DAY
7/7
Figure B-2
OPERATING DATA FOR TCA
RUN 510-2A
B-4
-------
ENOIUN JIO-2A '
2 a
P ,
w
o~Q f u
BO
75
10
•
SS3- •
E s ^
7
...
Sii»
5 s
1.0
m"
5i°"''
Hi-
7
B * 40
III -
3 is".
Z So
s 5 r »
C S 5
85 »
14
s - l!
Si*. '0
lli' •
S «s *
3 2 ~"
2
S~l x ,..
Is*
1 | 5 '•«
SSS o
a i *
30
2
a 11
11' ,
ill •
o ? 10
£
1
14,000
17,000
10,000
J
** 8,000
\
O 4.000
- 4,000
z
B 1.000
1
K o
2
5 600
C. 500
S
6 .a,
5
300
TOO
m
0
^^^v/V/
^^y \ /
, TOTAL, EXCLUDING OEMISTEB & KOCH HLAY
/_ '
s^
, OEMIiTEl 1 KOCH TMY
/ ;
t INLET (IN-LINE M£TE»)
/_^ — " — '
nT^Q-^^^O O O OL
^S* INLET (LAI)
LOO<>O<><>CL J
^^^_^-*> ^ OUTLET (LAIt
10 160 280 300 170 340 1«0 380 400 470 *40 440 4fl
TEST TIME, houn
C* LINDA* DAY
_
"^ ~~*
, TOTAL
r-^ L-— ^"
. (NiOtUILtl (AlHl
/
L— — _/—-
^— ]
^— j
j CALCIUM (CoO)
/ / TOTAL SULFUB (SO.)
J /
c^t^ ^A^A.^ /_/\
"^e^^^^e^^d
/ -9JLFITE (SOj)
^ O^-,_, _t-O
i i i i i i i i i 1 i
10 760 780 300 370 MO 360 380 400 4 20 440 440 44
TEST TIME, houn
CALtNDAl DAY
•
.
•
.
.
-
•
•
•
_
NO ANAIYTICAL WMnES TAKEN
•
•
•
, , 1 I I I I . . I
91
«0
81
BO
7S
10
B
4
*
1
..0
»
3.0
1.3
1.3
1.1
0 '
3D
40
10
10
14
1}
10
a
4
4
1
1.0
o
10
11
»
13
10
j
10
14,000
11,000
10,000
8,000
6,000
4,000
1,000
0
600
500
400
300
no
100
0
Gat Rail • 25,000 Kf rn 0 300 °f
Liquor Rili - 1,200 gpn
L/G - 64 gil/mrf
Gn Velocity • 8.8 II/K
E.H.T. RnidenoTim>- 20 min
Thrm Stagei, 5 in ipheres/itage
Gu Inlet SO2 Cone." 2.200-2.600 ppm
Scrubber Inlet Liquor Temp." 123-127 °F
Liquid Conductivity " 18,000 - 22,400 XL mhot/cm
e (CUriled Solidi Cone. - 36-38 wt %
Figure B-2 (Continued)
OPERATING DATA FOR TCA
RUN 510-2A
?40 160 280 300 310 340 160 330
TEST TIME, hwn
I 7/8 1 7/V I 7/'0 I 7/11 I 7/11 I 7/13
CAUNDAJI DAY
7/1S I 7/16
B-5
-------
°"gf
1EGIN (UN 506-31
//
70,000
11.000
16,000
- 14,000
J
^ 11.000
. 10,000
O
E 6,000
^ 4,000
s
s 2'tTO
« 0
• TOTAL CmitXVtD 1OLIDS A CHLOKIDt (Cl ' ) ^7 POTASSIUM (KM • SULFITE (SO - }
- O CA1.CKJM (Co** ) 4 MAGNESIUM {Mg ** ) Q CAR»ONAn (CO ' )
O SULfATE (504 ' ) i SODIUM (No * ) •
' *
.
•
A
A A
o
o - o
! 8 8 8 "
70.030
18.000
16.000
14,000
11,000
10.000
1.000
e.oop
4,000
0
Gil Rltl • 20.000 icf m » 330° F
Liquor Rm • 800 ipm Itotil)
U0-63|ll/mcf
Gn Velocity • 5.1 ft/ac
E.H.T. Undines Time-3) min
Mtrt.ll Bid Height - 3.5 in
Ga Inlat S02 Cone. • 2,300-3,100 ppm
Scrubber Inlet Liquor Tlmp. - 122-129 °F
Liquid Conductivity • 23.000,32.000 M. mhot/cm
Dietergi (Cimrrfugi) Solidi Cone. • 60-66 wt %
TEST TIME, houn
t/17 I 4/18
CALENDAI DAY
M Figure B-3
110 OPERATING DATA FOR MARBLE-BED
! RUN 506-3A & 3B
B-6
-------
BED & TOP SPHAY HEADERS (LJPPtR TAP FOR POI 30CW
RAISED If In. ON JUNE ?B, 1*73)
Gu Rate = 20,000 icfm 9 330°F
Liquor Rate* BOOgpm
L/C-S3 B«l/md
GBI Velocity- 5.t ft/we
E.H.T. Reiidenci Tims • 30 min
Marble Bad Height - 3.5 In
Gai Inlet S02 Cone,» 2,100-3,200 ppm
Scrubber Inlet Liquor Temp. - 122-130 °F
Liquid Condudiviiy « 25,000-30,600 u, mhot/cm
Otscnarge (Cintrif ugi) Solidl Cone. • 60-66 wt %
nn TIME, h
4/23 I 4/26 I 4/77 I 6/M
CAIENDAI DAY
4/?9 I 4/» I 7/1 I
10.000
18,000
14,000
-£ 14,000
r- 17,000
f
. 10,000
2 8,000
2 6.000
z
; «,ooo
1 J-wo
3
• IOTA! DISSOLVED SOUK A CHLOHM (Cl ' ) O CARIONATE (COj ' )
. O CAlCIUM(d** 1 * MAGNESIUM (Mo *+) a
D suLFAn no, • ) fi SODIUM (N,, * j •
^ fOIAJJtUM(K + J
• • SULFITE (50 - )
3 •
.
. A
A
*
o o ^
a a o a
70,000
18,000
14,000
U.OOO
17,000
10,000
8,000
6,000
4,000
7,000
0
TEST TIME, taun
4/7? I 4/1
CALENDAR DAY
Figure B-3 (Continued)
™ OPERATING DATA FOR MARBLE-BED
? RUN 506-3A & 3B
B-7
-------
Appendix C
TEST RUN INSPECTION SUMMARY TABLES
C-l
-------
Table C-l
Hun No. 501-1A (Depletion Stage) on Strata 16 Hours
Operating Date April 9 thru April 10, 1973
Conponent
Gas Inlet Duct
Vanturl
Scrubber
Afterscrubber
HoMlAB.
TO _spr<..
uemister
Kluah (2)
Chevron Stain-
less Stml
Demlster
Rsheater
IB Fan
Miscellaneous
(1) Bete Fog
(2) Top - 8p:
Bottom -
Scale
*
10 mil scale on walls
and in flooded elbow.
5 mil Intermittent
acale in afterscrubber
section, on and below
trapout tray, on and
below demlster scrubber
at second from bottom
Blurry header.
5 all scale on fourth
(Tap) and second slurry
header systems and
bottoa demiater spray
headers. (SO,-50.6J vt.
C02-2.0( Yt.fso2-36.St
«t., Ash-10.« vt. ).
5 mil scale precipitate
on bottom vanes.
*
*
*
T? us res
tying Systems Co. No. 1
Spraying Systems Co. Ho.
Solids Deposits
*
Negligible
Negligible
Negligible
Top completely free of solids.
*
»
»
I 7
3/1* H 6 w
Deterioration During Test Run Or At Time
First noticed
«
Negligible
Negligible
Negligible
Negligible
»
»
»
•Denotes not applicable or not inspected.
C-3
-------
Table C-2
Run Ko. 501-1A
On Stream 629 Hours
Operating Date April 10 thru Kay 9, 1973
Component
Gas Inlet Duct
Vraturi
Scrubber
Aftericrubber
Nozzle I. \
C 'Spn..
Demister
Plush w
Chevron Stain-
less Steel
Denister
Reheater
Component
ID Far
Miscellaneous
(1) Bete For, TF
(2) Top - Sprayl
Bottom - Spi
Scale
Negligible
Scattered thin scale
precipitate on valla;
flooded elbo« had 35
mil scale. (30,-53.9*i*.
CaO-lU.TK rt,, SO2-29.
6f, wt., Ash-1.7* wt.)
Intermittent scale in
afterscrubber section;
25 nil on and below
trapout tray;?? nil at
second from bottom
header spray area. 15
mil on and below denJate:
(S03-57.8j6 wt., SOj-
15.3* wt., CaO-21.lt*
vt., Ash-4.9< wt.)
2^ mil scale on second
header nozzles; 15 mil
soale on top header
nozzleo.
15 mil scale on bottom
vanes; scattered scale
deposit on top of vanes
(S03-66.6£ vt. , COp-SDjt
wt., SOo-Trsce, Ash-
25.7* wt. )
«
Scale
*
*
18 FC1!
ig Sys ems Co. Mo. 1 H 7
lying Systems Co. Ho. 3/
Solids Deposits
3 ft. 3 flyash accumulation In
horizontal section of ductwork
upstream of saturation sprays.
(Result of 2lt?5 operating hours)
Negligible
12 ft.3 solids deposit on bottom
of trapout tray. Scattered 3 inel
deep deposits on top slurry and
demlster bottom vash headers.
_totally
^iFive slurry nozzles' plugged/ foul
slurry nozzles partially plugged
in afterscrubber section.
Negligible
1/8 inch dry solids in duct abort
reheater; 3 burners cleaned.
Rollds Deposits
Thin dry solids film on fan
Blades
*
H 6 w
Deterioration During Test Pun Or At Time
First noticed
negligible
Guide vane bolts and surrounding area continue to
erode; two of eight annealed 316 stainless steel
bolts warrant removal.
Negligible
Slurry afterscrubber nozzles vere worn but In good
coalition .
The 316 stainless steel demlster was lightly pitted
and corroded with sons bent vanss from bMflUng but
demlster was in good condition. Stainless steel
demister was removed and polypropylene demister In-
stalled.
Stainless steel sleeve was severely deformed on the
north side In teardropahape, several SMil cracks
£ l^rw^r&t^^fW itfaffir-01" OTar
Deterioration During Test Run Or At Tim
First Noticed
Negligible
*
^Denotes not applicable or not inspected.
C-4
-------
Table C-3
Run Ho. 502-1A
276 Hours
Operating Data June 13 thru June 26, 1973
Conporwnt
Qaa Inlet Duct
Vcnturl
Scrubber
Aftersc rubber
Nor.z3.eB /^
Spray
Dealster
Flush (2)
Chevron Poly-
propylene
Demi ate r
BeheaUr
ID Pan
Conpcnent
MleoeLlAnaoufl
'!) Beto ?oC TT
(2) Top - Spra;-'l
Bottom - ^-p*
Scale
*
5 mil scale on walls;
30 nil scale on walls
of flooded elbow.
(S03-l*3.6£ wt.,
002-7.9* «t..
S02-19.3]t wt., Ash-
6.5* wt.)
10 oil acale on v&lls
bene&th trapout tray
Negliclble
1/16 inch scale on top
vanes.(S03-l(7.2t vt.,
SOg-H.5* wt., OOj-
2.BJ vt., C»O-1.6| «t.,
A«h-33.9* wt.)
«
*
Scale
*
8 FCT
ig Systems Co. Ko. U^7
ylng Syatcma Co. I.o. 3/1
Solids Depoaiti
«
Negligible
U ft. 3 Bollds depoolt on bottom
of trmpout ti«y. Scattered ^2
Inch solids deposlta on top
slurry header and bottom demlBtei
wash header.
One plugged Blurry nozzle
Scattered Infrequent solids de-
poBltB on bottom vanes; about
1/3 ft.2 flow area IB blocked at
four corners at demlstor aupport
bar Junctions.
1/16 Inch dry Bolide In duct
above reheater
Thin dust coating on fan blades]
fan danpers had no significant
solids deposition.
Sollda Deposits
*
H6w
Deterioration During Test Run Or Jit Tlffl»
First Rotlcei
*
north and eaet guide vane cross braces were eroded
In arc shapes of 5 Inch length t!3/l6 inch maxima
depth and 5i Inch length, 1 Inch mxlnnm depth
respectively. Th« tonth and vest cross braces vere
eroded to arc ahtps but las* severely. The guide
vanes, guide vane bolts, sod splash seal nuts and
oolti vere alao significantly eroded. Most of the
erosion took place In 923 hours slnoe the February "Q
*.gli#,l. OUtaee'
AfterBcrubber slurry nozclsa vere eroded but still
In good condition.
negligible
StainleBs steal sleeve continues to deteriorate but
at slover rate than in put. Sleeve 1* severely
bulged and varped on the north side vlth tvo cracks,
eachtafi Inches long. Refractory has deteriorated
slightly, IB severely cracked, but it 111 Intact.
Burners are oil coated but not daoaeed or deformed.
Negligible
Deterioration During teat Run Or At Tine
First noticed
*
•Denotes not applicable or not Inspected
C-5
-------
Table C-4
Run Bo. 503-1*
On Streaa
Operating Date June 29 thru July 11, 1973
Component
Gas Joint Duct
Venturl
Scrubber
Afterscrubber
Nozzles . .
^epray
Deniistei
Flush!2)
Chevron Poly-
propylene
Demiater
Reheater
ID Fan
Miscellaneous
(1) Bete Fog
(2) Top - Spl
Bottom -
Scale
*
SO mil scale on walla;
35 mil scale on walla to
flooded elbow.
(S0,-79.1< vt., 000-831
wt.? 80,-0.9* wt., A«h-
15. C* wt., CaO-2.6* wt.
15 mil scale on walls
beneath trapout tray.
Much of scale on wall
below demlster had dis-
solved and disappeared,
during test ^03-lA.
Hone
None
*
*
*
TT W FCN
ayiug Systems Co. Mo. 1
Spraying Systems Co. No.
Solids Deposlta
*
negligible
£~1 Inch solids deposit on bottos
of trapout tray.
Three slurry spray noieles were
plugged by debris.
Scattered, noounlform £-1 inch
solids deposits on top of bottom
vane and on tecond vans ; four
co rner sections were heaviest
covered as about 1/3 of their
flow area Has blocked rt'top of
bottom vane.
*
*
*
17
3/U H6W
Deterioration During Test Run Or At Tloe
First Noticed
*
Erosion on guide vane crosi braces continues.
negligible
negligible
Negligible
«
*
*
•Denotes not applicable or not inspected.
Table C-5
and
On Stream
23 Hours
Operating Data July U thru July 12, 1973
Component
Oas Inlat Duet
Venturl
Scrubber
Afterscrubber
BozEles
(l)Bpn
Demistei
Fi.ush (2)
Chevron Poty-
propylene
Demlster
Reheater
ID Fan
Miscellaneous
(1) Bete Fog
(2) Top - Spr
. Bottom -
Scale
.
3; oil scale on walls;
30 mil scale on walls ol
flooded elbow;
(S0,-(te.li5[ wt., C02-1.7ll
wt., S0,-2.lllt wt., Aoh-
13.3* wt.)
30 mil scale on walls
beneath trapout tray.
Some of scale on walls
from previous runs had
dissolved and diaapperei
Negligible
Negligible
«
*
*
tT 148 FCN
ylng Systems Co. No. 1 I
praying Systems Co. No.
Solids Deposits
*
negligible
0-3/8 Inch soft solids on bottom
of trapout tray.
One plugged slurry noizle in
afterscrubber.
£-1 inch solids present on 7/11
are slightly leas prevalent on
7/12/73.
1/8 inch dry solids deposit In
gas duct above reheater.
1*2 inch non-uniform solids ds-
poslt on fan Inlet dampers.
•
7
3/U H 6 W
Deterioration During Test Run Or At Time
First Noticed
*
Erosion of guide vans cross braces has contlned; the
*n*^H"ffln depth of the arc shape pattern on the north
and east aides hale increased to 3/16 and 1/8 Inch
respectively in 278 operating hours. The south and
east guide vaae cross braces had not significantly
eroded during this tijne period.
Sams of 316 stainless steel piping supports were
algnlgleantly pitted but still In good condition.
Negligible
Some broken and warped plastic denlster vanes, but
demlster Is still In good condition. It has been In
place since May 10th Inspection; 556 operating hours.
Three large cracks In stainless steel sleeve (1-2 ft.
long, 2-1 ft. lonf) at section Joints. Steel reheater
shell showed no discoloration since 6/26/73 (278 nrs.)
negligible
*
•Denotes not applicable or not Inspected.
C-6
-------
Huo Ro.
506-1A
Table C-6
On Stream Ul7 hour« Operating Date 7/29/73 - 6/13/73
Co-ponent
Gas Inlet Duct
Venturl
Scrubber
After Scrubber
Nozclea
Spray U)
De misters
Chevron
Polypropylene
Demlster (a)
Reheater
ID Fan
Miscellaneous
W Used both t
Scale
*
15 oil scale on walla;
15 mil scale on flooded
elbow walla.
20 mil acale on all 't
alurry spray headers ani
on bottom half of wall
expanse between trapout
tray and demlster.
20 mil scale on after
scrubber slurry nozzles.
5 mil deposit on
scrubber walls above
demister.
SO,-V».25(HT Aah-27.2#fl
SO|-27.9*HT CCj- .7*W
*
*
bp and bottom demister fl
Sollda Deposits
*
Small amount of aollda under
splaah seal flange.
Three k ft* of solids on bottom
of trapout tray and on walls
adjacent to trapout tray.
3 plugged alurry nozzlea in
after scrubber.
Large aollda fell on top from
gaa duct above.
Several deposits of ntflCOin^
at reheater outlet. &: 2 inch
aollda on gaa duct walls above
reheater.
5 inch solids deposits on bottom
of inlet dampers.
*
ush * Denotes not applicable)
Deterioration During Test Run Or At Tin*
First noticed
•
Erosion continues on north and east guide vans
crosa braces, guide vanes , and guide vane nut-bolt
assemblies. Erosion continuing on top splash seal
flange. Hirlmim depth of arc shapes on north and
east guide vane cross braces bad lncreajeO/32
and 3/1^ inches respectively during U17 hours.
negligible
negligible
Large solids deposits (one of 3/U ft5) fell on
plastic demlater from gaa duct above and did
significant damage. 10$ of top vanes wen shatter**
by these sollda. Plastic demlater was removed
and SS demister was installed.
Refractory was severely cracked but still intact.
Sleeve was distorted badly at north side,* cracks
between section Joints.
Experienced considerable difficulty during test run
in controlling gaa flow by gaa damper positioning
and had several shear pin failures on damper linkage
between dampers and damper control drive.
Venturl SO? outlet probe (AF1020) was partially
covered with solids; too extensive a covering on
or not inspected.
CT) Bete Fog IF W FCH
(2) Top - Spraying Systems Co. No. 1 H 7 ,
Bottom - Spraying Systems Co. No. 3A H 6 W
Table C-7
Run No.
507-1A
On Stream
Operating Date
6/22/73 -
Conponent
Oas Inlet Duct
Venturl
Scrubber
After Scrubber
Nozzles
Spray U)
Demlster
Flush (2)
Chevron SS
Demister
Reheater
ID Fan
Miscellaneous
* Denotes not
(1) Bete Fog
(2) Top - Sp:
Bottom -
Scale
*
Negligible
Negligible
Negligible
20-UO mil loose
crystalline scale on
walla above demiater.
*
*
»
ppllcable or not inspect
IF Ii8 FCN
aylng Systems Co. No. 1
Spraying Systems Co. No.
Solids Deposits
*
^ 5 mil solids on walls and
flooded elbow.
Negligible
One alurry nozzle was plugged.
Light dust coating on top and
bottom vanes was* 5 mil thick.
Several 2 Inch thick small
aollda deposits fell from gas
duct and lay on top.
Carbon deposit in No. 3 burners.
Solids in duct above reheater.
Negligible
*
ed.
1 7
3/"t H 6 H
Deterioration During Teat Run Or At Tins
First noticed
»
No additional eroalon on guide vane cross braces
during this 163 hour period. Some erosion of
corrosion specimens and aplash seal flange* Haynea
6B and 316 SS test wear bars were Installed on top
of north and east cross bracea.
Negligible
Negligible
Negligible
Refractory was severely cracked»a 1/2 ft hole
been oxidlied in the S3 sleeve opposite the No
burner; another 1/2 ft2 hole next to It due to
sleeve joint separation.
has
. 2
Negligible
*
C-7
-------
Table C-8
507-U
On Stnu 271 Sour*
Operating Data 8/89/73 - 9/9/73
Component
Oea Inlat Duct
Venturl
Scrubber
After scrubber
•OUlM
Bprajr I1)
"• lat»5»
Flush (2)
Chevron 8S
Dosdater
Reheater
ID Tan
Miscellaneous
•Denotes not a;
(1) Bate yog
(2) torp - S]
Bottom •
Seal*
*
Bon*
Ion*
Segllgtble
*
•
»
?llcable or not inspecte
IT 48 KB
raying System Co. Bo. 1
Spraying Systems Co. He
Solid! Deposits
*
3-lt Inch lollla on Venturl kails
around bull noult. 8«aU
asBunt (
-------
Table C-10
Run No.
509-1*. 5H-1*
520-1A, S32-U
Oa 8tr»a»
80 BOOM
Operating Date 9/U/73 - 9/16/73
Component
Oai Inlet
Duct (a)
Venturl
Scrubber (b)
After
Scrubber
Roat lea
Spray (1)
Deals ter
flush (2)
Chevron 88
Demi.ter
Reheater
ID Fan
Miscellaneous
fa) saturatioi
(b) Ho slurry
' Denotes no
Scale
Negligible
Hone
10 mil acale on walla
beneath trapout tray.
Hone
•one
*•
•
eprays (raw water) were
low to Venturl scrubber
applicable or not InapJ
Solids Deposits
1 inch thick, 6 inch wide band
at saturation spray area.
Some of solids around boll
nozzle of 9/10/73 inspection
were gone.
Hone
3 plugged slurry nozilea in
after scrubber section.
Heavy aollda on top and bottom,
Top - 3/k inch aollda in HI
corner
. i/li inch in 81
. 3/8 Inch in 8V
- 1/2 inch In RV quadrant.
Bottom had 1/2-1 inch wedge
deposit blocking 50j of flow
area on east aide, l/k inch
aolida on weat aide. *
1/16 inch aollda in duct above
reheat er. Carbon deposit in
burner Ho. 2.
1-1/Z inch (avg) 3 inch (mx)
solids on bottom of damper
blades.
•
uaed aa there waa no flow to the >
cted
Deterioration During Teat Run Or At TIM
First noticed
Hegliglble
Continued erosion on apper flange of aplaah aeal
and ita accompanying nut-bolt aaseabllesi. 1/3 of
nuta on Inalde position of north and ea*t guide TSUM
cross bra/sea eroded away. 316 88 wear bar waa more
severely eroded than Raynea 6B but eroaion waa more
aevere on eaat side than on north aide.
negligible
Hegliglble
negligible
83 aleeve has 2-1/2 ft* hol», one waa Bade by
oxidation, by Bo. 2 burner. Refractory eracka
up to 2inchea wide but not aa deep aa to reheater
shell. Hew aleere and refractory Is being Install*
Hegliglble
•
enturl scrubber.
«te Pog TF 1)8 rai
:op - Spraying Systeoa Co. Ho. 1 H 1, Bottom - Spraying System Co. Ho. 3/k H 6 V
C-9
-------
Table C-ll
Hun go. 501-2A (Depletion Phase)
On Streaa Hour*
22 Hour*
Operating Bates 3/22/73 - 3/83/73
Component
Oas Inlet
Duet
TO Scrubber
Boules
Spray (1)
Cooling (g)
Chevron SB
Deadster
Beheater
n> Fan
Miscellaneous
(1) Spraco
(2) Ceraml
* Denotes not
Scale
negligible
5 mil scale precipitate
on mill beneath bottom
bed. (803-19.7* Wt,
BOz - 32.9* Wt, Aab -
39.1* Wt).
Hone
1-2 nil scale on
bottoB veaes.
»
*
*
full cone, free flow noi
nozale (5/8 Inch openlu
applicable or not Inapet
Solids Deposits
1/5 of ventri-rod gas flow
airtta was plugged by 3 ft3
of solids.
•one
Bone
Bone
Hone
Light limestone dust on
blade*.
*
let, Bo. 1969.
).
ted.
Deterioration During Kit Rim Or it Tim
nret noticed
Begllglbls
Begllglble
Bone
Begllglble
Befractoiy and 88 ileeve were in T»ry good condition.
Begllglble
*
Table C-12
Run Bo.
On Strean Hour! 253 Hbure ________ Operating Da.te» 3/23/73 - V6/73
(xy» emce i»et cleaning ventri-rod) ~~^^^~~^""~~^^^^~1
Component
Oaa Inlet
Duct
TCA Scrubber
Boulee
Spray u)
Cooling (2)
Chevron SS
Demlater
Beheater
ID ran
Mlecellajieoua
Denotes no*
(1) Spraco :
(2) Ceraaic
Open pi
Scale
*
5 ail iron oxide scale
on valli inmllately
beneath Koch tr&y.
Hone
l/l£ Inch scale
precipitate on bottoa
vanea.
Begligible
Bone
*
applicable or not Inipc
•all cone, free flow noci
noule (5/8 i°ch opening
je nipple (1.0 inch openl
Bolide Depoaita
70* of ventrl-rod aisenbly
blocked by 5 ft3 of solid!.
Bone
Bone
•egllglbl.
negligible
Light dust coating.
*
cted.
ea, lo. 19&9.
3/83/73 - 3/2T/73
tig)3/27/73 - W73
Deterioration During Test Bun Or At Tins
First Hoticed
Begllglble
Bone
negligible
Beoliglble
88 sleeve hid not warped or deformed. Beheater refractory
above the burners had cracked and exposed expanded astal
retaining grid; rerraetory *as repaired.
MiTlnrnm fan blade defonatlon from straight line patten
mm .167 inch.
*•
C-10
-------
Table C-13
•o. 501-a*
On Stream Hours
92 Bouri
Opening D.t.._VVra-Vn/73_
Component
Oas inlet
Duet
TCA Scrubber
Hollies . .
spray t1/
Cooling '2'
Chevron 88
Demliter
Reheater
10 Fan
Mlscellaneoui
(I) Spraco
(2) Ceramic
Denote! c
Scale
negligible
negligible
*
*
*
*
»
Full cone, free flow noil
nosile (5/8 Inch openlni
t applicable or not Ins]
Solid. Depoeltl
6of of ventrl-rod flow
area was blocked by about
U-l/2 ft3 solids deposit.
legliclble
*
»
*
*
•
,e«, Bo. 1969
ected.
Deterioration During Teet HOD Or At Tiae
ririt Noticed
Ie«ll«lble
negligible
«
«
«
*
*
Run HO._501-2A
Table C-14
On Stream Hours 86 Hours
Operating Date. V^/73 - V1&/73
Component
das inlet
Duct
TCA Scrubber
Hossles
Spray (O
Cooling (2
Chevron 88
Dealster
Reheater
10 ?an
Mlscellaneoui
(1} Spraco
^2) Ceramic
Denotes nc
Scale
•egllglble
negligible
negligible
negligible
*
»
•
'ull cone, free flow nosi
noule (5/8 Inch opening
applicable or not insp<
Solids Deposits
5 ft3 of solids had
blocked 500 of ventri-rod
flow area (solid vaa
deposited In 86 hours).
k - 5 Inch solids deposit
on west wall beneath Koch
tray.
Ventrl-rod was renored
and replaced with the
U original Bete Tog nommle
(8T 2U KS\ . Riree of four
slurry nossles were
partially plugged.
negligible
*
*
*
es, Ho. 1969.
cted.
Deterioration During Test Run Or At Tlsu
first noticed
negligible
The flange of the a«ln steam sparger line Lasida the
scrubber bad partially separated. Tare* of fear nut-bolt
assemblies required replacement.
Ceramic cooling spray noule for Twatrl-rod was found
broken.
negligible
•
*
*
C-ll
-------
Table C-15
Run HQ. 501-2*
On Stream Bouri
127 hoars
Operating Date. V"/73 - V23/73
Component
Oea Inlet
Duct
TCA Scrubber
Nozzles
8p«y I l>
Cooling *z
Chevron 83
Demister
Reheater
ID Faa
Miscellaneous
(1) Spraco t
(2) Four, Be
Denotes no
Scale
Negligible
negligible
Negligible
Negligible
«
*
*
all cone, free flow nozz!
be Tog 8t-2U FCN nozzles,
b applicable or not Inapt
Solids Depoiltl
1» ft3 solidi deposit
upstream of cooling spray
nozzles.
2-3 inch ilurry solids
on the eatt and vest walla
beneath Koch tray.
The top cooling spray
nozzle, a Bete Fog
ST 32 FCN nozsle vss
plugged. Replaced Bete
Foe nozzle »lth Spraco TIB
nozzle.
negligible
1/3 gallon of carbon vas
deposited at Ho. 2 burner.
1/16 inch solids covered
duct above reheater.
Light dust coating on
blades.
*
e>, Ho. 1969.
cted.
Deterioration During Test Run Or At Tias
Flrit Sotlced
H«gll4iM«
Entire 5 Inch saddle bed had fallen into the bottom bed
due to t holei In support grid, roar of six (rid sections
at this elevation were replaced.
negligible
negligible
S3 ileeve vas itlll of circular shape and vai not deformed.
Decently repaired refractory (April 6th) vas in excellent
condition.
Negligible
About one dozen colLapead spheres in each bed were replaced
Table C-16
502-2A
On Stream Hours 889 Hours
Operating Petes VCT/7? - 5/10/73
Component
3as Inlet
Duct
FCA Scrubber
Nozzles
Spray (1)
Cooling (2
atevroo SS
Demist er
Reheater
CP Fan
4iscellaneovu
fl) Spraco
(2) Four, S
* Denotes nc
Scale
Negligible
5 mil scale on walla
beneath bottom grid.
*
Negligible
*
*
*
ull cone, free flow nozc
>raco 7 LB 316 SS nozzles
applicable or not insp<
Solids Deposits
A «edge ahapea, It- 1/2 ft3
solids deposit had accumu-
lated on the bottom of the
gafl duct upstream of cool-
Ing gpray.
Scattered 0-3 inch solids
accumulation on bottom of
Koch tray.
North cooling spray nozzle
vat plugged.
Negligible
*
»
»
es, No. 1969.
cted.
Deterioration During Test Run Or At Tine
First Noticed
Negligible
negligible
negligible
Negligible
*
*
0-201 puop shaft seal vas repacked.
C-1Z
-------
Table C-17
Bun go. 502-2A
On Stream Hour« 868 Hour«
Operating Dates
5/10/73 - 5/21/73
Component
0»f Inlet
Duct
TCA Scrubber
HosBlea *.*
Spray 1J>
Coollng(2)
Chevron 88
Denloter
Reheater
ID FAN
Hlacellaneoui
* Denotea not
(I) Spraco :
(2) rour, Bl
Scale
negligible
Insignificant localise
icallng.
Negligible
Negligible Additional
Scale.
»
*
*
applicable or not Inape
ull cone, free flow noiE
te Fog ST-32 FCN no»le>
Sollda Deposltl
60-70t of gas duct flow
area immediately upstream
of cooling spray noiiles
blocked by 7-1/2 ft3 of
solids.
One Inch solids deposit on
slurry noislee; no other
additional solids.
3 of U cooling spray
nosslea plugged.
2 south slurry Inlet spray
noBclea partially plugged.
Bottom vane flov area of
west quadrant 50$
blocked by solids.
»
*
*
ted.
es, No. 1969.
Deterioration During Test Run Or At Tine
First Noticed
Negligible
Grid wires In several areas of the top and bottoa bed grids
were noticeably eroded during test run) however, no (rldl
were removed.
Negligible
Negligible
•
*
*
Table C-18
, Bo. 503-2A - 506-2A
On Stream Hours l63 Hour«
Operating Dates 5/&/T3 - 5/29/73
Component
Oaa Inlet
Duct
TCA Scrubber
Ho"lM m
Spray I1'
Cooling (2
Chevron 38
Demlater
Reheater
ID Fan
Miscellaneouj
(l) Spraco
(2) Four, Bi
* Denotes i
Scale
Negligible
Negligible
Negligible
Negligible
Negligible
*
»
ull cone, free flow nois
te ?og ST-32 rCN notsles
ot applicable or not ins
Solids Deposits
$-1/2 ft3 wedge shape
deposit on bottom of gas
duct Immediately upstream
of cooling sprays.
NOTE: To prevent solids
buildup In the gas duct,
the sootblover noule
and blowing cycle was
altered as well as the
bottom cooling spray
noule was capped.
2-3 Inch solids on west
wall, and 1 Inch on east
wall, Immediately below
Koch tray.
South cooling spray
nofile and aasoc lated
header were plugged.
Southeast slurry nossle
was partially plugged.
Negligible
Negligible
1 inch solids accumulation
on bottoB of Inlet dampers
*
es, No. 1969.
wcted.
Deterioration During Test Run Or At Time
First Noticed
Negligible
Loosening, bending and erosion of grid wires continues.
Two grid sections of the bottom k»d were replaced; in
one, k lineal inches of wire were missing.
Negligible
Negligible
Reheater refractory above burner* cracked and closed
expanded metal to flame. Reheater sleeve is still
circular end in good condition. Several small section
cracks at welds.
Negligible
*Z> 1% of plastic spheres were damaged. Berth steam
sparge header under the loch tray had vibrated loose
the main header.
from
C-13
-------
Table C-19
Run Ho.
509-2A
On Stream Hour!
1465 Hours
Operating Bates 6/5/73 - 6/25/73
Component
Gas Inlet
Duct
TCA Scrubbei
Nozzles . .
Spray
Cooling''
Chevron SS
De mister
Reheater
ID Pan
Mlscellaneou
* Denotea n
(1) Spraco
(2) Three,
Scale
Negligible
60 nil scale on valla
beneath bottom bed.
(S03-71.6t Wt. S02 -
2.2% Wt, CaO-5.8t Wt
Ash-20.ltl( Wt) 35 mil
acale on walls be-
neath top and bottom
beds.
Negligible
1/16 Inch acale on
bottom and second fro
bottom vanes (SO?-
67.9* Wt, CaO-OJt Wt,
302 -2.7* Wt, ABh-
28.6* Wt).
»
Negligible
3t applicable or not ins]
rull cone, free flov nozi
ete Fog ST-32 FCN no»l<
Solids Deposits
Negligible
1/2 ft2 of scale-solid!
stalactites on HE corner
of bottom grid (S03-l|8.8*
Wt, S02-9.3*Wt, Ash-lH. 7
Wt).
North cooling spray
notile yes plugged; all
1* slurry spray nocsles
were partially plugged.
Negligible
1/16 Inch dry flaky solid
above reheater.
Light dust coating.
»
ected.
lea, No. 1969.
i.
Deterioration During Teit
First noticed
Hun or At Tla»
Negligible
Grid vires continue to deteriorate
had broken vires; U grid sections were
paired In place by tackveldlng.
Pin top grid Motions
replaced,* one was re-
Negligible
Negligible
Tvo small cracks (p:3 and 6 Inches long) at section joints
of S3 aleeve. Refractory in excellent condition since its
repair during 5/30-6/5 outage. Burner shrouds are in good
condition, not oxldlced.
Negligible
K, 20% spheres
otherwise) damaged.
collapsed or vitro,
Run Ro.
510-2A
Table C-20
On Stream Hours a97 Hours
Operating Eats, 6/S7/T3 - 7/10/73
Component
Sal Inlet
Duct
ICA Scrubber
Hollies
Bpray (1)
Cooling (2)
Chevron SS
Demiater
Reheater
ID Fan
Miscellaneous
\
* Denotea no'
(1) Spraco
(2) Three,
Scale
Negligible
30 oil scale on walls
beneath bottom grid (SO
7ll.5* Wt, S02-1.5»Wt,
Ash-22.7% Vt} 20 mil
scale on walla above
bottom bed.
65 mil acale on slurry
noaslea (S0,-6l.9jt Wt,
CaO-0* Wt, 802-10.14% Wt
Ash-25.1% Wt).
60 mil scale covered
bottom vanes (so,-75.2-jt
Wt, CaO-0% Wt, SOo-
2.8t Wt, ABh-21.i(J Wt)
*
1/3 of the bottom grid
waa covered with 1/U
inch stalactite acale.
(303-91. 8)1 Wt, Ash-9.2(
Wt, SOa-1.8* Wt)
applicable or not inape
ull cone, free flow noil
lete Fog ST-32 FCH nozllc
Solids Deposits
Negligible
Negligible
Top and south cooling spray
noailea were partially
plugged. All It slurry
noaalea were partially
plugged.
1/2 inch Bolida between
bottom vanes In SW corner.
Other sections partially
plugged.
1/8 inch dry solids in gas
duct above reheater.
Light dust coating on
bladea. Nonunlform 1-2
Inch sollda accumulation on
inlet daapers.
•
ted.
les, No. 1969.
I.
Deterioration During Test Run Or At Ties
first Noticed
Negligible
Loose and bent grid vires were discovered on seven 1 grids,
but grlda did not require replacement.
All U Blurry notile throats were eroded and contained 1/16
inch grooves; noulea are still In good condition, however.
Negligible
S3 sleeve had 5 cracks, only 2 are significant ( - It and
8 laches loag))001y one email crack above Ho. 2 burner in
refractory.
Negligible
*
C-14
-------
Table C-21
Run No.
511-aA
On Stream Hourfl
15 Hours
Operating Pates 7/10/73 - 7/11/73
Component
Oas Inlet
Duct
TCA Scrubber
Hollies
Spray (1)
Cooling (2)
Chevron S3
De mister
Reheater
ID Fan
Miscellaneous
* Denotes no
(1) Spraco :
(2) Three, 1
Scale
Negligible
5-10 mil scale on botto
of bottom grld,Jn 15 he
period 50% of grit
Previous italactlte
ecale had Increaaed In
length .OSlnch (SOj-771
Wt, S0,-2.17* Wt, 00,-
3.2< Wt, Ash-17.7* Wt
present composition)
0.3 Inch acattered
acale vaa more denaely
populated on eaat and
north valla than on
7/10/73.
Negligible
•
»
*
*
applicable or not Inapt
ull cone, free flov noxi
ate Fog flT-32 FCH noule
Solids Depoalta
Negligible
a *
ir
UX14 SCoIrd .
Negligible
*
*
*
*
cted.
.as. No. 1969.
Deterioration During Teat Run Or At Tina
First Hotlced
Negligible
Negligible
Negligible
*
*
*
*
Table C-22
Run No. 5H-2A. 512-2A. 513-2A On Str
Operating Date July 11 thru July 11, 1973
Component
Gaa Inlet Duet
TCA Scrubber
ironies
(1) Spray
(2) Cooling
Chevron Stain-
leaa Steel
Demlater
Reheater
ID Fan
Miscellaneous
(1) Spraco fu
(2) Three, Be
Scale
Negligible
15 mil acale on vails
beneath bottom bed. Hie
5-10 mil scale on 50J
of the bottom crld on
July 11 had disappeared,
•
4
*
W
»
1 cone, free flov noiili
e Fog BT-32 KB noillea,
Solids Deposits
Negligible
3/l6 Inch soft solids on valve
rims on bottom of Koch tray. ^
Inch solids on vest vail beneath
Koch tray.
I'orth cooling epray nocele part-
ially plugged.
Additional £ inch solids on
bottom vanes In southvest corner
remainder had 1/8 inch additions
aollds buildup.
*
*
Top of Koch tray had ^5 oil
aolids deposit even vitb no Irrl.
gatlon vater during 6£ hours of
testa 512-2A and 513-2A.
a, Ho. 1969.
Deterioration During Test Run Or At Tioa
First noticed
Negligible
Five grid sections vere replaced due to loose and/or
bent vires. One lineal inch of vire vas missing from
the vest central section of the bottom grid.
Cooling spray nettles (ST-32 FCN) have eroded signifi-
cantly after 1340 hours operation, but are still quite
operative .
Negligible
4
*
•Denote! not applicable or not Inspected.
C-15
-------
Table C-23
Run Ho.
51U-2A
On stream Hours
1*93 HOUTI
Operating Dates 7/22/73 - 8/13/73
Component
Oes Inlet
Duct
TCA Scrubber
Howies
Spray (1)
Cooling (2)
Chevron SS
Demlater
R Chester
ID ran
Mlacellaneou.
* Denotes m
(1) Spraco :
(2) Three ,B<
Scale
Negligible
3 mil scale below
bottom bed. 30 mil ac
between second and
third beds. 60 mil
scale between top bed
and Koch tray.
«O> 20% or flow area
waa plugged by 1/6
inch (avgj scale on
bottom two vanes.
(SO.-50.9* Wt, 80,-
O.lif Wt, Ash-US. T% Wt;
*
*
75% of bottom grid was
covered with varying
scale-solids deposits;
some areas were covers
with as much aa 3/*+ an
1 inch stalactites.
(S0,-n.7* Wt, SOa-
3. U", Wt, CaO-2U.3% Wt,
Aah-Ojt Wt).
t applicable or not ins[
ull cone, free flow nozi
e Pog 3T-32 ror aoules
Solids Deposits
7 Pt^ deposit at elbow on
bottom or duct.
1 inch solids on west wall
le beneath Koch tray. Botto
or Koch tray was clean ere
for 2 ft* in SW corner.
Top cooling spray nosile
waa plugged.
Negligible
Carbon deposit In No. 3
burner. Light solids in
duct sbove reheater.
negligible
*
cted.
es, Bo. 1969
Deterioration During Test Run Or At Time
rirst noticed
negligible
Two grid aectlons were replaced because or broken wires.
i Several other sections eontsined loose and bent wires.
Pt
Cooling spray nosilea (8T-32 rCN) are eroded but still in
good operating condition.
Continued corrosion of vanes (particularly topatoat vanes).
93 sleeve is defomlng on northeast side. Eight cracks at
section joints of aleeve -onlf 3 over 6 Inchesiseveral
I/It inch wide vertical cracks In refractory.
negligible
114 of spheres of bottoa two beds were punctured or dimpled.
All of high density polyethylene spheres In top bed (milky
white) were in good condition after 1*93 hours. O-203
yaag shaft ileeve was grooved and required replacement.
C-16
-------
Table C-24
Run No.
515-2A
On Stream 571 Hours
Operating Dote August 16 thru September 10, 1973
Component
Gia Inlet Duct
TCA Scrubber
Scattered
Nozzles
U) Spray
(2) Cooling
Chevron St&ln-
less Steel
Demi ate r
(1) Spraco fu
(2) Three, Be
Component
Reheater
ID Pan
Miscellaneous
Seal*
negligible
New .168" scale unlf on-
ly covered the walls
beneath bottom bed. Ken
scale between bottom be
and Koch tray avg. .062
Inch. 2-3 Inch stalac-
tite acale on bottom
grid. Scattered non-unl
form scale-solids of 1
Inch max. depth. (F»r--
tlculsrly OB west side]
Negligible
filial of gas flow area
at bottom two vmnes wac
blocked by scale and
solids.
feall scale
strip about three IndK
wide above demlster.
(S03-80.3* wt., 80,-
1.91 wt., Ash-17.8f wt)
1 cone, free flow nozili
e rog 8T-32 PCS nomzloe
Scale
*
*
•
Solids Deposits
Solids on TE-2007; remainder of
solids In -gas duct Inlet vflC ne-
gligible.
1
Top and south cooling apray noz-
zles were plugged. TSiree slurry
sprey nozzles were partially
plugged.
s, Ho. 1969.
Solids Deposits
l/li and 1/8 inch solids In gas
duct entering tad leaving re-
heater, respectively.
Ron-uniform 3 Inch deep (max. )
solids on bottom of dampers. 5
mil dry dust on back of blades.
*
Deterioration During Test Run Or At Time
First noticed
negligible
1-6 In.' and 2-U In.2 holes In wire grid sections of
third grid. Spheres from top bed had fallen Into
middle bed. Second grid had two lineal Inches of wi«e
missing. Four of five steam sparge trench headers
and main head flange wen Loose.
Erosive grooves In alurry spray nozzle throats were
«1 Inch deep (max.); nozzle e were otherwise In good
condition.
Negligible
Deterioration Daring Teat Run Or At Tine
First noticed
Stainless steel sleen was distorted; new eleeve
will be installed. 1/U Inch vertical oracki in
refractory.
negligible
G-201 pump Impeller and inner oaflifig WCTI eroded
and pitted to various depths of .200, .279, .300
Inches at Impeller rim, hub aid at suction inlet
respectively. G-206 pump inner casing wai loose.
0-203 aid 0-20$ pinnj sleeves grooved.
•Denotes not applicable or not Inspected.
C-17
-------
Table C-25
Bun go. 501-3* (Depletion Hia««)
stream Hour*
3"t Eovuri
Operating Datea 3/W73 - 3/15/73
Component
lea Inlet
)uct
eurole Bad
Icrubberl
rom.lBS f .
Slurry (1A
Cooling -;
Sievron 88
Knitter
teOMter
:D ran
Uacellaneoua
> Denotes no
1) Bottom nee
Top Headez
?) Bete Foe £
Seal*
Hegliglble
5 •!! scale oo Blurry
piping and valla.
(7
Cooling1 z>
Chevron 83
De mister
Reheater
CD ran
Uacellaneoiu
^ Denotes no
1) Bottom Hes
Top Headei
2) Bete Fog E
Scale
»
Bottom slurry headers a
Inch acale - solids (A
so2 - 11.1* wt).
Bottom slurry spray no
acale - solids (Asn - (
S02 - 11.1* Wt).
Hegllglble
Negligible
*
*
: applicable or not Inspi
ler - CE new Ijnproved no
- CE new improved nozzl
T-20FCH nozzles (Four).
Sollda Deposit!
1 ft3 of lolida on north
aide.
id noziles covered vlth 3/8
1-60.1* Wt, 803 - 22.7* Wt.
Eles cowred «ltb 3/8 Incb
).l* Wt, 803 - 22.it Wt,
1/8 Inch (lurry deposit on
bottom vanet. LKht dust
covering on top vuaa.
O - 1/U Inch dry lolida in
outlet duct above rehsater
Light solids coating.
*
eted.
zles.
s.
Deterioration During Test Bun Or At Tlaa
Wrst noticed
legliglble
BegllglbLe
The cooling apny header and three of four spray oosilas
were obstructed by iaaill plscea of «ood and scale. Tiro
of the bottosi slurry noules w«ra plugged.
Hegllglble
negligible
Tvo of eight fan bladea vere aignlflcantly warped.
blade had a sudaum dsronmtion of O.Jo Inches from
•tralght Una pattern.
On*
a
*
C-18
-------
Table C-27
Run Ho.
501-3B
On Stream Hours
gJHour. Operating Date. 3/30/73 - 3/31/73
Component
0«s Inlet
Duct
(tortile Bed
Scrubber
Monies , .
Blurry^'
Cooling* z>
Chevron 88
Demlater
Reheater
10 ran
Mlacellaneoui
* Denotea no
(1) Bottom He«
Top lieader
'Si Bete Foe E
Scale
*
10 nil acale on valla ,
Blurry headera, and
slurry notilea (SO, -
60. OJ Wt, 80, - 2"tT3<
Wt, Aah - ?.» Wt).
10 mil acale on the to;
and bottom outer aurfa<
of tbe slurry headera.
1/10 Inch solids-scale
(aee scale analysis ab(
solids on top vanes.
*
*
*
applicable or not Lnapi
ler - CE new Improved no:
- CE new Improved nozzl<
T-20FC1! nozzles (Four).
Bollda Deposits
Small solids deposits on
the north. ajid louth aides
«l/2 ft3).
Bagllglble
*
ee
accumulation on bottom vuea
ve). 1/16 Inch aoft flaky
«
Light Dust Coating.
*
cted.
zles.
s.
Deterioration During Test Boa Or At Tlsja
rirat Hotlc«d
•egllglble
negligible
•
negligible
as sleeve hxo defomed and distorted slightly but othenrlsie
In good condition.
Ho further deformation observed or neaiured.
•
Table C-28
Run Ho. 501-3B
, Stream Houri B* H°"' Operating Pate. V2/73 ' U/6/73
Cosponent
Oaa Inlet
Duct
Marble Bed
Scrubber
Rozilea , . v
Slurry (2)
Coollngu;
Chevron S3
De mister
Reh eater
ID Fan
Mlacellaneout
* Denotes ni
(1) Bottom Hei
l^p Heodei
(2) Bete FOR £
Scale
*
Negligible
•
1/16 Inch scale on top
vanes. (Ash - 5O.2* W
SO, - "A.!* Wt, 30, -
5.#Wt). Z
*
*
«
applicable or not Insp
er - Cf. new Improved no
- CE new Improved nozzl
-20FCK nozzles (Four).
Solids Deposits
1 ft3 (total) solids
deposit on north and south
side.
Negligible
*
negligible
Negligible
Thin duat coating.
*
cted.
zles.
8.
Deterioration During Test Run Or At TlSH
Tlrst noticed
negligible
negligible
*
negligible
S3 sleeve has varped and deformed to an elongated oval
shape. Refractory deterioration rate Increasing.
Presently has significant number and slie of cracks.
Negligible
*
C-19
-------
Table C-29
Run Ho. 501-3B
On Str
i Houri
390 Hours
Operating Dates V6/73 - Vag/73
Component
Oas Inlet
Duct
(tortile Bed
Scrubber
Houl.es / . x
Blurry ^L
Cooling1'1''
Chevron 88
Demleter
Reheater
ID Tan
Klacellaneous
* Denotes no
'!) Bottom Hea
Ttop Headei
<2) Bete Pog S
Scale
negligible
10 nil scale sfter 26
noun! 5 nil addltlooial
•eale after U73 noure.
Up to 1 Inch of acale-i
nozzles. (Aeh-l>l».U* Mt
Wt, CaO - 12.6)1 Wt).
Top bad 1/16 Inch nonui
oulation.
*
»
*
applicable or not InajM
er - CE new improved no:
- CE nev improved nozzl<
-2QFCN nozzles (Pour ).
Solids Deposits
2 tt* at Blurry-ash soft
solids on north and south
Bides.
PZ, 12 ft3 of scale-sollda
removed frost slurry piping?
nozzles, bottom of
bed and walla.
ilids on slurry piping tod
, S02 - U.7* Wt, 803 - 27.8)1
.form scale and solids accu-
Bottom covered vlth 1/8 ln<
slurry solids.
1/16 Inch of dry soot and
solids in duct above
reheater. Burners were
cleaned.
Thin soot and solids coatii
ted.
zles.
Deterioration During Test ROD Or At Tls»
First Botic«d
Negligible
29* of Ifcrble bet was plugged or In stratified patten
(initial plugging stage).
Ths swirl vanes in 13 of the 16 bottom slurry noBiles
have completely disappeared, toe regaining 3 vere only
remants of their original site Jl shmpe after 70t hours
operation. The 6. top nossles vere only lightly t.-Med.
negligible
Refractory continues to crack.
deform on north side,* currently
g. Jo further blade deformation.
88 sleere continues to
haa "teardrop" shape.
rroa random sables In the glass sphere bed, the average
sphere velght loss during this test was about 6%.
Table C-30
Run Bo. 502-3*
do Stream Hours
28? hours
Operating Dates U/83/73 - 5/7/73
Component
Oas Inlet
Duct
Marble Bed
Scrubber
•°"lel !1\
Blurry* 1'
Coollng(-)
Owrron 88
Dealster
Baheater
ID ran
Micellae eoui
* Oanot
Scale
Negligible
5 nil scale beneath bet
(not on walls) 10 mil
scale above bed on
piping walls, nozzles.
(80«-66.5f wt, C05-7.0
wt, 80 -2U.I1S wt, Aah-
6.6f wt).
5 •!! scale on nozzles
below bed, 10 mil seal
on nozzles above bed.
Bottom venes covered
with 10 mil scale.
Top vanes covered with
1/b solids and brovn
scale. (S0,-55.3t wt,
COe-lt.oi ^, 80-6.9*
wt, Ash-33.7t
-------
Table C-31
Run No.
503-3A
On Stream Hour*
267 hour*
Operating Date. 5/11/73-5/82/73
Component
Qas Inlet
Duct
Marble Bed
Scrubber
Nozzles / ; i
Slurry f>
Cooling1 3>
Chevron BS
De milter
Rebeeter
ID Fan
Miscellaneous
* Denotes no
Seal*
Negligible
Bottom Blurry headers a
under the bed were cove
aollds deposits. (SO?-:
15.9* wt, Ash-20.3* Wt
Slurry nozzles covered
nith 1/2 - 1 Inch aluri
scale deposits.
Hone
Rone
»
*
applicable or not losp<
Solids Deposits
1-1/2 ft3 of solids on
north side and south aide.
id nozzles and west wall
>ed with 1/2 - 1 Inch scale
1.2% Wt, CaO-30* Wt, 802-
North and south most coolir
1 spray noilles were partial!
Plugged. 3 of 16 bottom
slurry nozzles were plugget
C-E noztles removed, Spraci
8H nozzles Installed.
Bottom covered with light
dost coating. Top of top
vanea covered with 1/8
Inch soft solids deposit.
1/8 Inch solids deposit
la duct above raheater
(Ash-21.3* wt, H-0-11.7* w
HC-67* wt).
1-2 Inch BolldB deposit
on bottom of damper blade*
Thin solids-moisture coatli
on fan blades.
Marble Bed waa plugged in
stratified row pattern
over 25% of Its area.
Stratified areas varied In
severity of plugging.
ted.
Deterioration During Test Run Or At 7iJSa>
71rat Noticed
Negligible
Bed retaining grid was further secured la U area*).
I Cooling apray nozzles In good condition. Svirl vanes) In
r top slurry spray nozzles (Combustion Engineering) still
Intact.
Negligible
88 sleeve continues to wa-fb and refractory to crack but
at slower rate than la past.
Do further fan blade deformation.
The outer casing rubber liner of O-30k pump waa seventy
eroded and pitted around the suction inlet (up ta 1/U
inch deep) due to a piece of slurry nozzle being lodged la
the suction line next to the Impeller. Two thrust besrlngs)
and one inboard had normally wore out and were replaced.
1) Bottom Header - CE new improved nozzles.'
Top Header - CE new Improved nozzles.
?) Bete Fog ST-20FCM nozzle« (four).
Table C-32
Run Ho.
50U-3A and 505-3A
On Stream Hours
23mn*MI.
6 ft^ (12*) of marble bed
was plugged. Another 60*
of bed is In stratified ro<
Deterioration During Teat Run Or At TlaM
Flrat Noticed
' 15 mil Iron oxide scale buildup during 3752 operating hour*.
negligible
Hone of the Spraco 8 M nozzles had eroded or deteriorated
during this period. Tour, 16 inch long, 1/2 inch oianeter
nlpplea were added to ST-21* FCH nozzles in an attempt to
keep inlet gaa duct solids deposit* at a -'•*'•"-
negligible
Reheater refractory Is badly cracked but (till intact. 80
sleeve has warped but ha* no crack*. Ho. 2 burner abroad
has oxidized significantly.
The small hole In the 8V corner of the expansion joint abovt
the ID fan that vaa repaired at the end of teat JO2-3A, 5OO
operating hour* earlier, needs repairing again.
(1) Bottom Header -* Spraco 8M nozzles.
Top Header - Spraco &{ nozzles.
pattern. (2) j^te Fog ST-207CH nozzles (F-ur).
C-21
-------
Table C-33
Run Ho. 506-3*
On Stream Hour*
operating DatM 6/8/73-6/9/73
Coxponent
}aa Inlet
Duet
Itorble Bed
Scrubber
ioiilei
Blurry^.
Cooling*2'
Chevron 83
Deniater
Jeheater
[0 Fan
CLacellaneoui
^ Oenotei nc
) Bottom Hea
Top Header
) Four Bete
Scale
«
Negligible
*
negligible
#
*
*
applicable or not Inap
er - Spraco GM nozzles.
- Spraco GM nozzles.
og ST-2QFCW nozzles.
8olld> Depoiiti
Bone
Negligible
3 coollog ipray noislei van
completely plugged; reaaln-
Ing one «ai partially
plugged.
Negligible
*
*
50% of Mu-ble Bed «a» in
Btratlfled row pattern.
(incipient plugging atage).
cted.
Deterioration During Teit Run Or At Tim
ririt noticed
negligible
negligible
The ipray of 8 of the 16 bottom • lurry nojalei OB west and
eut (Idea of icrubber vaa defleetal by onrflon w«lr
domcoBer piping. Tbeae noulei »ere extended 10-1/2 Ijicnei
tovard the bed. Cooling apray ayitea >ta •odlfled >lth
3-7 Inch long, 1 Inch dlaaatar nlpplei and 81-32 KS
noiilei.
Negligible
»
»
»
Table C-34
On Stream Hour.
Hour" Operating Date. 6/15/73 - 6/19/73
Component
KB Inlet
Duct
arble Bed
Scrubber
oCElei ^
CoolingC2)
hevron S&
Denis ter
eheater
D Fan
[Iscellaneoui
Denotes n<
) Bottom Hea
Top Header
) Three Bete
Scale
Negligible
1/8 inch ecale- slurry i
bed. Walls below and i
3-10 mil scale depoalti
*
*
*
.*
*
t applicable or not inap
er - Spraco flM nozzles.
- Spraco StM nozzles.
Fog ST-32PCH nozzles.
Sollda Deposit*
1 ft3 of Blurry solids
downstream of the cooling
spray B.
olidB on slurry piping below
bove bed bad 10 mil fcnd
, rcflpectiv«ly.
Berth and middle cooling
spray nociles were plugged.
South noEEle was open.
*
*
*
1 ft2 of marble bed was
plugged. 12-11*4 was in
stratified row pattern.
cted.
Deterioration During Test Run Or At Tis»
First Noticed
negligible
*
*
*
Fart of propane gas supply piping was cracked and propane
fuel adjusting screw was inoperative; could not relight
reheater.
«•
*
C-22
-------
Table C-35
506-JB
On Stream Hour*
130 Hour!
Operating Dates_
6/26/73 - 7/8/73
Component
Oas Inlet
Duct
Marble Bed
Scrubber
Houle* fl)
Blurr/ Ly
Cooling l '
Chevron 88
Demliter
Reheater
ID fan
Miscellaneous
* DenoteB n
Sole
Hegllglble
3/8 Inch average acale-
plplng beneath bed (80,
2.6H wt, aah-22.7* vt)?
valli above and belov b
vt, SOj-1.3* «t, Ce0-0i
3/8 Inch acale-alurry
deposit on bottom
•lurry spray nettles.
2O nil scale OD bottom
vanea; 90 Dill scale
on top vanea of
composition (80, -81. 54
vt, CaO-OJt wt, gOo-1.1'
vt, aah-16.8*; vt).
*
»
2O-U5 mil scale In
•action llnei of O-30J
and B pumpi. (00,-lOf
vt, 80,- 1* vt, s6,-72.!
vt, aah-5.6*; vt, CaO-
1.5* vt).
t applicable or not Ins]
Solid! Deposit!
It ft3 solids depoalt on
north and south ildei dovn-
'etream of the cooling iprayi
ollds deposit on slurry
67.2* vt, S02-5.5* vt, CaO-
50 and 60 til icale on
id, respectively. (80,-77.5i
vt, a«h-21.3* vt). 3
Horth and south cooling
•pray noiileia van plugged.
At middle vane, 1/3 of
demliter vai completely
plugged vhlle 1/3 vaa
partially plugged.
1/8 Inch solids depoalt In
duct above reheater.
2-3 inch depoalt on bottom
of Inlet damper. Light .
solids coating on bladea.
About 9% of bed vaa plugged;
114 In stratified pattern.
,
cted.
Deterioration luring Teat Run Or At Tine
mat noticed
Hegllglble
Hegllglble
The throat and vblrl chambers of 13 of 16 bottom Bpraco 8M
•lurry «pray noiiles van severely eroded after l»93 operat-
ing bouri. (The ramalnlng 3 bottom noiilej vere plugged).
The top 6 nosilsi were in good condition. 8 of 16 bottom
•lurry nosslei vera • replaced vlth nev Spraco 8M noiBlei.
negligible
Reheater refractory crackj continued to enlarge oa
north aide. 63 aleevv va» not cracked but vaa deformed.
Sleeve vaj replaced.
Sipanalon Joint above ID fan (till leaking. (See run 5O5-3A
report of June l>th). Dlicbarge !• of lov pH (about i.k).
Tlake lining in tank 0-301 (craped (0.105 inch groove) by
the 8 Inch overflow weir dovncomer. The domcoaur vu
installed during the Hireb 31, 1973 outage.
(1) Bottom Header - Spraco tn noczles.
Top Header - Spraco BMnozzles.
(2) Three Beta Fog, ST 32iW noziiea.
Table C-36
Run Bo. 507-3A and 5O8-3A
On Stream Hours
37 Hours
Operating Date» 7/10/73 - 7/11/73
Component
O«B Inlet
Duct
Marble Bed
Scrubber
Houles
Slurry t1'
Cooling (2)
Chevron 38
Demliter
Reheater
n> ran
Klacellaneou
* Danotea n
(1) Bottom H«
Top Heade
(2) Tvo, Bet<
Scale
Negligible
2O and 10 mil acale
deposited on valla and
alurry piping beneath
and above marble bed
during the above perlo<
•
15 mil scale on
extreme north and
south aectiona of top
vanes.
*
*
*
t applicable or not imp
ider - Spraco 8M noczles
- SpracoflfcL noxElea..
fog BT-32TCB BOMBSr.
Solids Deposits
2 n2 of 1 Inch deep solids
3/U - 1 Inch solids on vest
tvo bottom slurry pipes.
*
1/8 and 1/U Inch solids on
north and south quadrants
of bottom vanea. Remainder
of bottom vanes covered
only vlth dust coating.
•
•
30£ of the (fertile bed vaa
plugged. A p across bed eC
7.1» in HgO.
cted.
Deterioration During Test Run Or At Time)
First Noticed
Hegllglble
•
5 of 16 bottom elurry nomiles vere plugged. lorth cooling
•pray nestle vaa partially plugged.
»
•
»
*
C-23
-------
Appendix D
TVA INTERIM REPORT OF CORROSION STUDIES:
EPA ALKALI SCRUBBING TEST FACILITY
by
G. L. Crow
H. R. Horsman
October 1, 1973
D-l
-------
EPA ALKALI-SCRUBBING TEST FACILITY— SHAWHEE POWER PLANT
Interim Report of Corrosion Studies
Identification and solution of corrosion and erosion problems
associated with construction materials are important goals in a program
for the design and evaluation of limestone - vet-process systems for
removing sulfur dioxide from stack gas at coal-fired power plants. The
program at the Shawnee Power Plant is a cooperative effort among the
Environmental Protection Agency (EPA), Bechtel Corporation, and TVA.
Earlier corrosion tests made in pilot-plant studies "by the
Process Engineering Branch of limestone - wet-scrubbing systems at
Colbert Power Plant showed that some materials of construction were
durable while others were severely attacked under plant operating
conditions (Process Engineering Branch reports—Sept. 1971, Dec. 1971,
July 1972, and Aug. 1972).
At the request of the EPA in 1972, the Process Engineering
Branch of TVA started corrosion tests of 17 alloys and 7 nonmetals at
21 strategic locations in three parallel scrubber systems at the Shawnee
Power Plant. The systems were the venturi, the Turbulent Contact
Absorber (TCA), and the Hydro-Filter; each of these had the capacity
to handle JO,000 acfm of gas.
After the systems had been operated from 1700 to 220O hours,
the results of corrosion tests and of plant inspections showed that
greatest corrosion had occurred in areas such as inlet ducts and venturi
where wetted gas or gas and slurry flowed at high velocity. Typically,
the stack gas contained O.J$ S02 and 3 to 5 grains of fly ash per cubic
foot. Deposits of solids, such as limestone and fly ash, prevented
erosion but caused corrosion of the concentration cell type in some
areas. The most resistant alloys tested were Hastelloy C-276, Inconel 625,
Incoloy 825, Carpenter 20 Cb-3, and Type 3l6L stainless steel. Hastelloy
C-276 was the most durable and also the most expensive; Type 316L stain-
less steel ranks fifth in durability but eleventh in cost of the alloys
tested. Rubbers, such as butyl, natural, and neoprene, showed good
resistance. With some exceptions, units of plant equipment made of
Type 316L stainless steel or lined with neoprene or with polyester-
inert flake material were durable. Testing work is being continued.
*Process Engineering Branch of the Tennessee Valley Authority.
D-3
-------
ogram and Flans
Program: The limestone - vet-scrubbing program for sulfur
oxide removal at Shavnee is funded and directed "by EPA. The Bechtel
rporation designed the plant facility and TVA built it. TVA is operating
e plant under a test program developed and directed by Bechtel. Evalua-
on of construction materials by exposure of test specimens at strategic
cations and by inspection of the plant equipment is an important goal in
e program.
Plans: Responsibility for conducting the evaluation program
Shavnee vas assigned TVA in March 1972. /Report to Air Pollution
ntrol Office, EPA (Contract No. PH 22-68-6?, June 28, 1968), by Bechtel
rporation (March 2, 1972 )J The Process Engineering Branch of the
vision of Chemical Development was given this task. /Informal memo-
ndums—H. W. Elder to R. D. Young (April 5, 1972) and Ronald D. Young
H. W. Elder (April 1, 1972)_./ This vork includes procuring test
terials, making test specimens, fabricating suspension equipment for
ools and racks of specimens in the plants, and reporting test results.
so included are periodic inspection and evaluations of plant equipment
r corrosion and wear.
Bechtel Corporation specified 20 materials of construction that
nsisted of 17 alloys and 3 nonmetals to be tested at 2k- designated loca-
ons in the three plants. ^aterial List of Corrosion Coupon Test Rack
/L5/72) and Drawings SK-M-102 through 109 and SK-M-111 (Job 6955),
chtel Corporation^
Plant Facility; Figure 1 is a view of the plant shoving the
ree parallel scrubbing systems—the venturi, the TCA, and the Hydro-
Iter.
Power plant stack gas at an average temperature of J20°F (300°-
0°F) flows through a lj-0-inch duct to a system vhere it is sprayed for
midification and for cooling. It then passes through limestone slurry
a particular type of test scrubber for sulfur dioxide removal. After-
rd, it is freed of mist in a separator, reheated to between 235° and
5°F to vaporize mist and eliminate a plume, and discharged through a fan
d duct to the atmosphere. Scrubber effluent is clarified to remove
lids which are discarded and the liquor is then recirculated.
Some features common to all the systems are described below. A
-inch duct is used to carry the stack gas at 320°F from the No. 10 boiler
the power plant to a test system; each duct is made of 10-gage carbon
eel, ASTM A-283, and is insulated except at flanged joints. The 40-inch
ct connects to another gas duct made of Type 316L stainless steel. This
ct is equipped with two sets of spray nozzles (the first for humidifying
d the second for cooling the gas) and an air-operated soot blower.
D-4
-------
Downstream from each sulfur dioxide absorber and mist eliminator unit
there are a stainless steel duct, a refractory-lined reheater fired with
fuel oil, an induced-draft fan of stainless steel, and a stack of stain-
less steel. For liquor handling there are a slurry recirculation tank,
a scrubber effluent tank, and a liquor clarification system. The effluent
hold tank and a clarifier tank are made of carbon steel A-283 and coated
inside with Flakeline 103 which is a Bisphenol polyester resin-fiber glass
coating manufactured by the Ceilcote Company. The recirculation tank,
clarified water storage tank, and reslurry tank are made of carbon steel
and lined with neoprene.
Distinguishing features of the systems are as follows. In the
venturi scrubber system shown in Figure 2, the gas is scrubbed in a venturi
unit made of Type 516 stainless steel and then passed through a neoprene-
lined spray tower (afterscrubber) with a chevron-type separator in the top
for mist recovery. In the TCA system, shown in Figure 3> Sae is scrubbed
in a mobile bed of wetted balls, and the mist is removed in a Koch
FlexiTray and chevron-type separator in a tower lined with neoprene. In
the Hydro-Filter system shown in Figure k, gas is scrubbed in a flooded
bed of marbles, and the mist is removed in a chevron-type mist separator
in a neoprene-lined scrubber tower.
Preparations for Corrosion Tests
With Bechtel's approval, several improvements were made in plans
for the design, preparation, and installation of test specimens. Non-
metallic materials added to the test materials list included Qua-Corr, a
fiber glass-reinforced furan resin, and the rubbers—butyl, natural, and
neoprene. Stressed specimens of five alloys were also added to detect
stress-corrosion cracking under plant operating conditions.
Disks: Disk-type specimens, 2 inches in diameter, were prepared
from the 17 metals. A weld was made (according to manufacturer's recom-
mendations) across the diameter, and after being welded, the metal was
cooled slowly in still air to simulate conditions of constructing or
repairing large equipment. Whenever it was available, metal stock of
1/8-inch minimum thickness was used, and the surfaces were machined
smooth after the welding. Some alloys available only in thinner gages
could not be machined, so the weld beads were smoothed by grinding. A
hole, 2J/6k inch in diameter, was drilled in the center of each disk for
mounting.
D-5
-------
Three nonmetallic materials, Bonds-brand 4000, Flakeline 200,
and Transite, vere also prepared as 2-inch disks and mounted on spools
along with the metal disks. Flakeline 200, a coating material, was
applied on mild steel disks "by the manufacturer. Bondstrand UOOO and
Transite are self-supporting materials and are obtained in sheet form
for disk preparation.
Stressed: A strip approximately 1/8 by 1 by 5-1/2 inches was
welded at midlength, machined to smooth all the surfaces, and formed into
a U shape. One-half-inch holes were drilled in each end of the strip to
accommodate a bolt (Type yi6 stainless steel, 1/4 inch) fitted with
Teflon insulators for applying static stress in the specimen.
Coated; Because of their large sizes of approximately 4-1/2 by
4-1/2 inches, the plate specimens of Qua-Corr plastic and the butyl,
natural, and neoprene rubbers were mounted on a separate rack. Qua-Corr
is a self-supporting material; the rubbers were applied on mild steel
specimens by the manufacturer. The durometer "A" hardness values of the
rubber-coated specimens as received were as follows: natural, 3^-37;
butyl, 54-56; and neoprene, 64-65-
Mounts and Suspensions: Spools and racks for mounting the test
specimens and also the suspension equipment for installing them in the
plants were constructed mainly of Type 316 stainless steel. Bolts and
nuts were annealed to remove stresses caused.by cold-working in threading
operations. To prevent loss of fasteners through vibration of equipment,
two nuts were locked by forcing them together.
At some test locations inside plant equipment, brackets were
attached as permanent fixtures by welding, and then the spools of specimens
were bolted to them. In other locations, spools were fastened to existing
pipes by the use of band-type clamps. In a tank, spools were suspended by
means of a 1/8-inch strip or a ~5-i.uch pipe that was bolted to the top.
Sleeves (3/8-inch wall by 6 inches long) of soft butyl rubber were placed
around the J-±nch specimen support pipe as cushions to prevent abrasion
damage to the Flakeline coating or neoprene lining on a tank wall. No
specimens were installed inside pipelines or fittings.
Figure 5 shows the three types of assemblies used for mounting
the corrosion test specimens. These were:
(A) Stressed—with 5 U bends
(B) Rack—with three rubber-coated plates and one plastic plate
(C) Spool—with 20 disks consisting of 17 alloys and J nonmetals
D-6
-------
A Teflon sleeve was used to insulate the specimens from the supporting
stainless steel bolt, and Teflon spacers or washers were used to prevent
contact of the dissimilar materials.
Figure 6 shows prepared specimens and support equipment before
shipment from the Office of Agricultural and Chemical Development (OACD)
to Shawnee in August 1972. A few racks of specimens are shown attached
to 3-inch pipe and to 1/8-inch-thick strap.
Test Exposures, Conditions, and Procedures
Test specimens of materials listed in Tables I, II, and III were
installed in the three plant systems in August 1972. Table IV gives the
analysis of each of the 17 metals tested. Specimens were exposed at test
locations identified by series 1000, 2000, and 3000 as shown in Figures 2,
3, and k; however, specimens (100U-6) were omitted in the venturi after-
scrubber tower that was to be modified. All specimens remained in the
scrubber systems from August 12, 1972, to February 3> 1973^ except those
in the TCA system which were temporarily removed for preliminary
evaluation in November 1972.
Plant Operation: Usually, one system was operated at a time,
although all three could be operated simultaneously. Operating hours
in the exposure period are shown below.
Hours
System _ Idle Operated
Venturi 23*«D 1840
Turbulent Contact Absorber 2535 1667
Hydro-Filter81 1950 2203
a
Also called marble bed.
Plant Process Materials and Deposits: Typical compositions
of inlet and outlet gas at the scrubber systems are tabulated below.
Stack Scrubbed
Component gas gas
so2, % 0.3 0.05-0.2
C02, % 13 12
N2, £ 74 69
02, # ^-5 b
H^O, % 8 15
Fly ash, gr/std ft3 3-5 0.02
D-7
-------
Temperature of the inlet stack gas from unit 10 boiler averaged 320 °F
(300°-350°F) and that of the exhaust gas after being reheated was 235°
to 265°P.
Ranges in properties of liquor in the different tanks of the
three scrubber systems are summarized below.
Liquor in tanks
Recycle Effluent Clarifier
Temperature, °F 70-125 75-130 70-100
Solids, % by weight 4-15 ^-10 0-30
pH 5.3-6A 5.6-6.8 6.0-7-6
Composition, $ by weight
CaS04-2H20 1.0-3.0 1.0-2.0 0-6
CaS03-l/2H2p 1.5-5.0 1.5-3.5 0-10
Unreacted limestone plus
fly ash 1.5-7.0 1-5-4.5 0-15
Water 85-96 90-96 70-100
Table V shows analyses of deposits from the three systems in
the plant. These scale and solid deposits from tanks and scrubber equip-
ment exposed to the limestone scrubbing liquor in the three systems were
composed mainly of calcium, suLfite, sulfate, and carbonate in the ranges
of percentages shown below.
Percent
Component by weight
CaO 26-41
S02 9-18
S03 24-47
C02 0-5-6
Soot that deposited in stacks above the gas reheater contained the
following on a dry basis: 27 to 64$ ash and 36 to 73$ hydrocarbon.
Moist soot contained 2 to 13$
Exposed Specimens: Pictures were made of specimens when removed
from the plant as shown in Figures 7-12. Then the specimens were cleaned
and their corrosion rates and physical condition were determined as shown
in Tables I through III along with properties of gas and liquor at various
test points .
D-8
-------
Inspections of Plant; Equipment in the plant systems was
inspected for corrosion and erosion damage daring the first week of
February 1973-
Durometer "A" hardness values of rubber lining on equipment
and on test specimens were measured with a Shore instrument—Type "A-2,"
ASTM 22*K). Unfortunately, hardness of most lined plant equipment was not
determined before plant operation; so data from the rubber vendors were
ordinarily used as reference values. Temperature of the atmosphere varied
from 35° to 60°F as did the temperature of equipment during the plant
inspection. A decrease in temperature would be expected to increase
rubber hardness. Values for neoprene linings are ^imniarized in Table VI
for the plant equipment and in Table VII for test specimens after exposure
to plant operating conditions.
Results of Plant Inspections
and Corrosion Tests
In this section, plant inspections are described first, and
then the results of corrosion tests under different exposures in equip-
ment are given. Some of the observations on plant shipment were made
by or in collaboration with R. E. Wagner and R, C. Tuxis, engineers with
TVA at Shawnee Power Plant.
Carbon Steel Ducts for Inlet Stack Gas—Pla^t Equipment: A
product of general corrosion thinly coated the inside walls of these
ducts where they had been insulated. A thicker corrosion product covered
inner walls of uninsulated duct sections (at flanges) because heat loss
through bare metal to air cooled the stack gas and condensed corrosive'
liquid containing carbon dioxide, oxygen, and sulfur oxides. Such
localized corrosion was pronounced in the Hydro-Filter duct; and subse-
quently in February 1973> the plant personnel fully insulated this duct
as well as those to the other systems.
Small quantities of fly ash had deposited in ductwork areas
where the gas flow changed directions, but this caused no apparent
problem.
Stainless Steel Ducts for Inlet Stack Gas—Plant Equipment:
In each duct between the carbon steel section and the scrubber unit
there are: three nozzles of Type 316 stainless steel for spraying
liquid to humidify gas, and one nozzle of Type 309 stainless steel
for blowing air to dislodge soot. At the TCA and the Hydro-Filter
(but not the venturi) scrubbers, there are four nozzles of Carpenter
20 alloy for spraying recycle slurry to cool the inlet gas.
D-9
-------
The ducts, in general, vere not appreciably corroded. Slight
abrasion occurred in areas vhich vere not coated "by solids, but corrosion
Df the concentration cell type was present under the accumulations of
solids.
Spray nozzles for gas homidification in these ducts were operated
for the number of hours shown below.
Percent of
Duct to Spray hoars operating hours
Venturi 56? 31
TCA 0 0
Hydro-Filter 2203 100
rhe conditions of the soot "blower nozzles were as follows: at TCA—good,
md at Hydro-Filter—severely corroded. (The nozzle in the venturi system
fas not inspected.) Nozzle corrosion at the Hydro-Filter was attributed
to the use of the water sprays for gas humidification upstream which would
/ield hot corrosive mist containing carbon dioxide, oxygen, and sulfur
Dxides.
Two of the four nozzles of Carpenter 20 stainless steel used
For cooling gas to the TCA scrubber were plugged and two were severely
eroded internally. Erosion was caused by high-velocity flow of cooling
Blurry consisting of water, limestone, and fly ash.
Stainless Steel Ducts for Inlet Stack Gas—Corrosion of Test
Specimens: Specimens located in ducts below gas humidifier sprays
:orroded as follows in mils per year: 1 to more than 330 in venturi
system, 1 to 17 in TCA, and 1 to more than JOO in Hydro-Filter. (See
points 1002, 2002, and 3002 on Figures 2-k.) The high rates in ducts to
bhe venturi and Hydro-Filter systems are attributed to previously men-
tioned hot corrosive spray from humidifier spray operation. (Compare
L002, 2002, and 3002 on Figures 1, 9, and 11, respectively. ) Type 3l6L
showed good resistance in the venturi and TCA ducts but had localized
ittack (l8-mil groove and minute pits) in the Hydro-Filter duct. Other,
nore expensive alloys, such as Hastelloy C-2?6 and Inconel 625, showed
rood resistance in the ducts to the three systems as well as in other
)arts of the systems as described later.
In the duct to the TCA system where no humidification was used,
;he temperature of the gas was 260° to 330 °F, and the conditions of the
lonmetal specimens were: Transite—good, Flakeline 200—fair, and
Bonstrand—poor. All three of the materials were in poor condition in
lumidified gas to the venturi and Hydro-Filter units.
D-10
-------
Venturi Scrubber—Plant Equipment: Bolts and nuts of Type 304
stainless steel used to assemble internal parts of the venturi scrubber
had failed twice in plant operation before they were replaced with ones
of fully annealed Type Jl6 stainless steel.
The neoprene-lined duct between the venturi unit and the after-
scrubber tower was in good condition. Durometer A hardness of the lining
was 67.
Venturi Scrubber—Corrosion of Test Specimens: The specimens
were installed directly below the vertically mounted venturi as shown at
point 1C11 of Figure 2. Gas.and slurry (laden with compounds of sulfur
oxides) at a high velocity caused more severe corrosion and erosion
damage to specimens in this location than in any other in the three
systems. Specimens of nine alloys and three nonraetals failed. Figure 7
shows that spool 1011 was clean and only 8 of the 20 test specimens
remained at the end of the 1840-hour test period. The five alloys that
showed the lowest corrosion in mils per year were: Hastelloy C-276--5 mils,
Inconel 625—5 mils, Incoloy 825—7 mils, Carpenter 20 Cb-3—14 mils, and
Type 316L stainless steel—15 mils.
The other three remaining alloys and their corrosion rates
were: Cupro-nickel 70-30—49 mils, Monel kOO—57 mils, and Hastelloy B—
100 mils.
The three rubbers (butyl, natural, and neoprene) were in good
condition, but the plastic Qua-Corr failed as shown fourth from the left'
on 1011 in Figure 8. Both Figures 7 and 8 show severe erosion damage to
the chemically resistant Teflon spacers on the spools at location 1011.
Towers in the Venturi, TCA, and Hydro-Filter Systems—Plant
Equipment: In general, the neoprene lining on the wall of each tower
was in good condition (Table VI). Hardness values and comments are
listed below.
D-ll
-------
Tower
Venturi
(afterscrubber)
Durometer hardness
Original8- Measured*3
60-65
TCA
Hydro-Filter
60-65
60-65
52-60
53-63
65-72
Comment
Wear was apparent in a small area
near a nozzle. The highest hard-
ness was near the top and the
lowest was near bottom of the
tower.
Highest hardness was in the mid-
section and the lovest was near
bottom and top of the tower.
Slight impact damage was probably
caused by foreign sharp objects.
Highest hardness was in the mid-
section and the lowest was near
the top of the tower.
From vendor's data--hardness was not measured in the plant before tower
operation.
Measurements, in plant were not made at same temperature because of
weather changes.
Solids deposition in the towers varied as described below:
Venturi
(afterscrubber)
TCA
Hydro-Filter
A heavy deposit was present as follows: on the walls
below trapout tray in bottom; on a 30-inch-wide band
of the wall below the mist eliminator (chevron); and
on bottom (l/2 the area) of the mist eliminator near
top of tower.
Multilayered deposits of solids covered the walls of
the tower. These decreased in thickness from 1 inch
at bottom to 1/16 inch at top. The mist eliminator
(chevron) was partly clogged. No loose solids were
present because the unit was cleaned 2 weeks before
the inspection.
Scale, 1/16 inch thick, was on walls, piping, and
spray nozzles. Slurry deposit was 1/4 to 1/2 inch
thick on a narrow band of wall below and adjacent
to the mist eliminator and on the bottom of the mist
eliminator.
D-12
-------
Corrosion of Type 31.6 stainless steel components of the towers
ranged from mild to severe as described belov.
Venturi . Surfaces under solid deposits had generally developed
(afterscrubber) small pits. The wall of the outlet duct below the
gas reheater, although clean, was pitted.
TCA Grids that supported packing showed negligible corrosion,
but their top surface showed some abrasion from the
moving bed of wetted balls. The Koch FlexiTray was
clean and showed no apparent corrosion after 985 hours'
service. However, the top side of the mist eliminator
(chevron) had undergone severe general corrosion and
pitting after 1667 hours1 service.
Hydro-Filter Corrosion of the mist eliminator (chevron) and other
components of Type 316L stainless steel in this tower
was not detected.
Spray nozzles of Type 316 stainless steel were generally in good condition
after handling slurry in the venturi afterscrubber and the TCA towers, but
nonmetal nozzles in the Hydro-Filter tower were damaged. Four of 16 nozzle
locations had been previously blanked when nozzles had failed and no spares
were available. Some of the plastic nozzles beneath the glass sphere bed
in the Hydro-Filter were damaged and had been replaced with nozzles of .
improved design. The remaining original nozzles were badly worn, and two
of four improved design nbzzles had failed. Of six soft rubber nozzles
at a higher level in the Hydro-Filter, four were badly eroded; in one,
the lining of the whirl chamber had torn so as to plug the outlet, and the
casing was cracked.
Towers in the Venturi, TCA, and Hydro-Filter Systems—Corrosion
of Test Specimens: In the afterscrubber of the venturi system, specimens
were not installed because of plans to alter the arrangement of sprays.
In the TCA scrubber tower, test specimens were mounted at three
elevations (see Figure 3 an(i Table II). Those above the third grid for
holding mobile packing hollow plastic spheres at location 2006 were
exposed to gas and liquor. Those below the FlexiTray at 2005 were
exposed to gas and droplets, and those below the chevron mist eliminator
at 200k were exposed to gas and mist. Figure 9 shows that spools of
specimens which had been exposed at 2006 and 2005 were partly covered by
solids, but those at 2004 were clean. In the period August 12 through
November J, 1972, movement of the mobile bed had caused some erosion at
2006. The eroded specimens were replaced with new ones that were placed
within a wire mesh container to protect them from erosion by the balls.
At all three locations, the corrosion rates were 1 mil per year or less
for Carpenter 20 Cb-5, Hastelloy C-276, Incoloy 825, Inconel 625, and
D-13
-------
lype 31^'ti stainless steel. G-as and mir.t at 200^-1 below the mist eliminator
caused crevice corrosion on Incoloy 825 and minute pitting on [type J16L
stainless steel. It also caused the greatest corrosion of mild steel
(250 mils) and Cor- Ten B (268 mils per year). Pitting and/or crevice
corrosion occurred on most of the other alloys exposed in this tower.
The corrosion of stressed specimens 200*4- shown in Figure 10 vas about
equal to that of the counterpart disk specimens in Figure 9-
The condition of the nonmetallic materials tested in the TCA
tower ranged from poor to good. The three specimens of rubber and two
Df the three specimens of Bondstrand were in good condition.
In the tower of the Hydro-Filter system, tests of corrosion
specimens were conducted at two locations (see Figure k- and Table III).
Dne was in the liquor and inlet gas at 3006 below the marble support grid,
and the other was in the gas and liquor at 3005 above the marble bed (see
Figures 11 and 12) . All of the test specimens were coated with scale and
deposit. The following alloys were corroded at rates of 1 mil per year
3r less: Carpenter 20 Cb-3, cupro-nickel 70-30, Eastelloy C-276, Incoloy
325, Inconel 625, and- Type 3l6L stainless steel. Monel 400 and Hastelloy B
tiad rates of less than 1 to k mils per year in the two locations. Mild
steel and Cor- Ten B had the greatest rates — 37 and kO mils per year,
respectively, above the marble bed; and 14 and 13, respectively, below
the bed. Pitting and/or crevice corrosion occurred on the other alloys.
En the liquor and inlet gas at J006, Bondstrand was good, and the Flakeline
and Transite were fair. In the gas and liquor at 3005^ Bondstrand Qua-Corr,
and the three rubbers were good, and the Flakeline was fair.
Exhaust Gas Systems — Plant Equipment; Each exhaust gas reheater
for heating the scrubbed gas to between 235° and 265°F is identical in the
three systems (Figures 2, 3> an(3- ^) • The refractory lining, 3 inches
thick, in all the reheaters had cracked, mainly near the burner ports .
Ehe lining of the venturi reheater had the largest cracks and was coated
fuel oil.
In the venturi stack, soot saturated with oil had deposited, and
Dn two occasions such a deposit had ignited and burned. In the TCA exhaust
stack, soot saturated with oil was also found, but no fires had occurred.
En the Hydro-Filter system, the soot deposit in the exhaust duct was
thinner, indicating that combustion of fuel oil in the heater had been
nore efficient than in the other two systems .
Downstream from the reheater in each system, there was no
apparent corrosion of the stack made of Type 316 stainless steel.
D-14
-------
At the induced-draft (l. D.) fern of each system, soot and fly
ash accumulated on the fan blades and housing to depths of 1/16 to lA inch.
In general, the thickest deposits were on stationary parts, and the trailing
faces of the "blades accumulated a thicker deposit than other areas of moving
parts. Deposits were smallest in the fan for the Hydro-Filter where no oil
was detected. Measurements of the "blades and shrouds of Type 316 stainless
steel showed only slight variation in thickness from the original values
determined before the plants were operated. Slight bends on the periphery
of two blades on the fan for the Hydro-Filter and one blade on the fan for
the TCA system probably occurred because of stress relieving, but these
bends caused no apparent problems.
A stainless steel sleeve (*tO-inch diameter by k feet high) has
subsequently been installed in each reheater. Also, burner nozzles of
different design and having much better atomizing characteristics were
installed. The sleeve and nozzles should promote essentially complete
combustion of oil before hot combustion gases combine with the scrubber
exhaust gas and thus should minimize problems with oil and soot deposits
in the stack and fan.
Exhaust Gas Systems—Corrosion of Test Specimens: Corrosion
test specimens were mounted in the exhaust stacks in each system 8 to 10
feet downstream from the reheater as shown at points 1007, 2007, and 3007
(Figures 2, 3> 9-nd. h). Temperature of heated exhaust gas in contact with
the specimens was usually between 235° to 265°F. Tables I, II, and III
give corrosion data. Figures 7 through 12 show the soot- and ash-covered
specimens after exposure.
In the stack of the venturi system, the corrosion of the test
specimens was slightly more severe than in other systems. Oil-saturated
soot had caught fire and destroyed the Teflon insulators and spacers.
(See item 1007 on Figures 7 and 8.) Five of the highly alloyed materials
and Type 3l6L stainless steel were durable, however, corroding at rates
of less than 1 mil per year. Five other alloys had corrosion rates of
1 to 5 mils per year, and the rates for mild steel and Cor-Ten B were 16
and 18 mils. Aluminum 3003 was pitted to a depth of 70 mils during the
exposure period. Transite was in good condition after the test, but
Bondstrand and Flakeline failed apparently because of overheating.
In the TCA exhaust stack, the corrosion rate was either
negligible or less than 1 mil per year for eight alloys including
Type 316L. (See item 2007 on Figures 9 and 10.) Cor-Ten B and mild
steel had rates of 2 and 3 mils per year with minute pits. Pitting of
other alloys ranged from minute to depths of 12 mils. Flakeline 200
and Transite were in good condition but Bondstrand failed.
D-15
-------
In the stack of the Hydro-Filter system, corrosion of specimens
as slightly greater than in the TCA system "but less than in the venturi
ystem. (See item 3007 on Figures 11 and 12. ) Five alloys, including
ype 316L stainless steel which had minute pits, were corroded at rates
ess than 1 mil per year. Several alloys were pitted, and the deepest
it was 18 mils in E-Brite 26-1. Attack of mild steel, Cor-Ten B, and
luminum 3003 was 4 to 5 niils per year with crevice corrosion under the
eflon insulator. Flakeline and Transite were in good condition, "but
ondstrand failed.
Corrosive attack of the stressed specimens by reheated stack
as was about equal to that of the counterpart disks in each test.
Effluent Hold Tanks—Plant Equipment: An effluent hold tank
0 feet in diameter and 21 feet tall is located directly under each
crubbing tower: IX-101 for the venturi, D-201 for the TCA, and D-301
or the Hydro-Filter systems. The shells are made of A-283 carbon steel
oated inside (80 mils minimum thickness) with Flakeline 103 manufactured
y the Ceilcote Company. This coating is a Bisphenol-A type of polyester
esin filled with flake glass (25-35$).
Each tank was in good condition except for minute cracks at the
unction of some baffles with the tank walls. Stains of iron rust indi-
ated that the cracks penetrated the Flakeline coating. All cracks were
ithin 8 feet of the bottom of a tank. The neoprene-lined agitators were
n good condition, and only slight wear was noted'on the leading edge of
he blades. The hardness of the neoprene had changed little if any (see
able VI). The Bondstrand 5000 and the Type 3l6L stainless steel down-
omers showed no evidence of attack in either tank. In tanks D-101 and
-301 there were slightly worn areas where the butyl rubber insulator on
he specimen suspension pipe (15 feet long) had rubbed the vail.
Effluent Hold Tanks—Corrosion of Test Specimens: Corrosion
est specimens were mounted in the effluent hold tanks 15 feet below the
op. Figures 2 through 4 identify the locations and Figures 7 through 12
how pictures of test specimens by numbers—1008 for the venturi, 2008 for
he TCA, and 3008 for the Hydro-Filter systems. Tables I through III show
hat corrosion was less than 1 mil per year for several alloys in the
hree tanks.
In the venturi system tank, nine alloys, including Type 316L
tainless steel, had corrosion rates of 1 mil per year or less without
ocalized attack. The eight other alloys were attacked locally and had
orrosion rates of 1 to 18 mils per year. Four specimens were pitted
p to 2U mils deep. Crevice corrosion occurred on eight specimens. The
ate for mild steel was 18 mils and that for Cor-Ten B was 1^ mils per
ear, both with crevice corrosion.
D-16
-------
The TCA effluent hold tank was in use only daring the last
^5 hours of the 1667-hour operating period, BO corrosion rates are less
representative than those at the two other tanks which were used con-
tinually during operating periods. Corrosion was less than 1 mil per
year for nine alloys. Appreciable corrosion in mils per year occurred
to several metals as follows: aluminum 3003, 20; Cor-Ten B, 170; and
mild steel, 210. Apparently significant general corrosion of these
alloys occurred during the extended period that the tank was idle,
but there was only minute pitting and no crevice corrosion.
In the Hydro-Filter tank, corrosion was less than 1 mil per
year for 10 alloys, including Type ~5\.6L stainless steel, without
localized corrosion. Aluminum 3003, mild steel, and Cor-Ten B were
corroded at the greatest rates—2 to 5 mils per year. Pitting occurred
on three alloys, and the deepest pat was 5 mils on Type 304L stainless
steel. Crevice corrosion occurred on 7 alloys.
In all three of the effluent hold tanks, the following shoved
good resistance: the butyl, neoprene, and natural rubbers; the Bondstrand
and Qua-Corr plastics; and the Transite. Because of abrasion on one face,
the specimens of Flakeline 200 were in only fair condition. Flakeline 200
is similar to Flakeline 103 except that it is formulated for application
by "brush or spray" instead of by "trowel or spray." Apparently, the
application of Flakeline 103 coating inside the effluent tanks was
superior to that of Flakeline 200 on the test specimens.
Recirculation Tanks—Plant Equipment; Each of the scrubbing
systems has a recirculation tank 5 feet in diameter by 21 feet tall as
follows: D-10^ for the venturi, D-204 for the TCA, and D-30^ for the
Hydro-Filter. These tanks were lined with neopreiie sheet 1/4 inch thick,
and the blades and shaft of their agitators were also lined with neoprene.
The linings on all of the tank walls and the agitators were in
good condition. A thin scale had deposited that would protect the surface.
Durometer A hardness values for the neoprene linings, however, were not
consistent (see Table VI). The hardness values were higher than the
original for the lining in Hydro-Filter tank D-304, and they Were lower
for the agitator blades in TCA tank D-204.
Recirculation Tanks--Corrosion of Test Specimens; Corrosion
test specimens were suspended 8 feet below the top in recirculation tanks
D-10U (venturi) and D-304 (Hydro-Filter); they were 15 feet below the top
in D-20^ (TCA). See Figures 2, 3, and k. Corrosion was negligible or
less than 1 mil per year for several alloys in the three tanks (Tables I
through III). The greatest attack occurred on Cor-Ten B and mild steel.
Items 1012, 2012, and 3012 on Figures 1, 9, and 11, respectively, show
the spools of specimens after exposure.
D-17
-------
In venturi tank D-104, the rate for Cor-Ten B was 12 and that
>r mild steel was 19 mils per year. The rate of attack on the other
.loys was less than 1 mil per year. Pitting and crevice corrosion
:curred only on Type klO stainless steel.
In TCA tank 204, the corrosion rate was 5 mils per year for
>r-Ten B, 5 mils for mild steel, and 2 mils for aluminum. Pitting
icurred on seven alloys -with the maximum depth of 10 mils on Type
;ainless steel. Crevice corrosion occurred on four alloys.
In Hydro-Filter tank D-304, the corrosion rate was 10 mils for
>r-Ten B and 11 mils per year for mild steel. Type 410 had pits 7 mils
:ep and three alloys underwent crevice corrosion. The other alloys were
;tacked less than 1 mil per year.
In general, localized attack was less prevalent in the recircula-
.on tanks where agitation was more vigorous than in the effluent hold tanks
:xcept in D-201 that was used only ^5 hours).
Bondstrand and Transite were in good condition after the test in
Lch tank, "but Flakeline 200 was only fair "because of abrasion on one face
' each specimen.
Clarifier Tanks--Plant Equipment; Clarifier tanks D-102 for the
>nturi and D-302 for the Hydro-Filter are 20 feet in diameter and 15 feet
ill; and tank D-202 for the TCA is 30 feet in diameter "by 15 feet tall.
ich tank has a coned bottom that is positioned J to 5 feet above the
>undation elevation (Bechtel drawings M-8 and M-9). The tanks are of
• 283 carbon steel coated inside with Flakeline 103. Mechanical equip-
:nt inside the clarifiers is made of Type J>16L stainless steel. Tank
•302 was not inspected because it was in use for testing filter
luipment.
The Flakeline 103 coating in tanks D-102 and D-202 was in good
>ndition except for cracks at the junction of the overflow weir and the
ill. Iron rust had "bled through the cracks. The stainless, steel equip-
jnt had not been attacked. However, four carbon steel "bolts used to
ichor the underflow cone at the bottom of the two tanks had rusted.
Clarifier Tank—Corrosion of Test Specimens: A spool of corrosion
;st specimens was suspended in the fluid 5 feet belov the weir in clarifier
inks D-102, D-202, and D-302. These tanks are not shown in Figures 2
irough k (see Bechtel drawings M-8 and M-9). Items 1013, 2013, and 3013
i Figures 7, 9, and 11 are pictures of specimens after exposure. Tables I
irough III show corrosion data.
D-18
-------
In the venturi. tank D-102, seven alloys including Types
and 316L stainless steel showed negligible corrosion, and two other alloys
had rates of 1 mil per year or less without localized attack. Cor-Ten B
and mild steel had rates of 5 mils per year. Pitting occurred on four
alloys, and the greatest depth was 12 mils on aluminum JOOJ. Five alloys
had undergone crevice corrosion.
In the TCA tank D-2O2, six alloys including Type JO^L and
Type 3l6L stainless steel showed negligible attack and seven other alloys
had.rates of 1 mil per year or less without localized attack. Cor-Ten B
and mild steel were corroded at rates of 6 and 8 mils per year. Two
alloys were pitted; the deepest pit was h mils on Type kW stainless
steel. Localized corrosion occurred on Cor-Ten B and Type 410 stainless
steel.
In the Hydro-Filter tank D-302, a total of nine alloys including
Type JO^L and Type 316L stainless steel corroded at less than 1 mil per
year without localized attack. The rates were 7 and 9 mils for Cor-Ten B
and mild steel. Four alloys were pitted; Type UlO stainless steel had the
deepest pit, 16 mils, and four alloys had undergone crevice corrosion.
Transite was in good condition in the three clarifier tanks;
Bondstrand was good in the venturi and the TCA tanks but poor in the
Hydro-Filter tank because of spalling; Flakeline 200 was fair in the
three tanks.
Clarified Process Water Storage Tanks—Plant Equipment; Clarified
water storage tank D-103 for the venturi and D-303 for the Hydro-Filter
systems are 10 feet in diameter and 9 feet tall. Tank D-203 for the TCA
system is 13 feet in diameter and 9 feet tall. Each tank has four baffles
and a shell of carbon steel lined with 1/k inch of neoprene of durometer A
hardness of 55-60. Each tank has a three-blade agitator with diameter as
follows: Ik inches in D-103 and D-303 and about k2 inches in D-203. The
agitators and shafts are lined with neoprene.
Conditions of linings in the clarified water tanks are described
below.
D-19
-------
Tank Tank No. Condition of lining on
renturi D-10J Tank; Excellent
Agitator: Noticeable wear
'CA D-203 Tank: A lap joint, 8 inches long, vas
loose where "bottom liner extends upvard
1-1/2 inches to make overlap on wall
liner near a "baffle on west side.
Agitator: Slight wear
ydro- Filter D-303 Tank; Good
Agitator: Noticeable wear. Cuts at
several places were possibly made by sharp
foreign obj ects.a
Two pieces of thin gage metal were on floor of tank.
It appears that the durometer A hardness of the neoprene might
ave increased slightly up to 11 units above 60 as shown in Table VI.
owever, the temperature (60DF) of the linings during the inspection was
ower than the standard (73°F) specified in the ASTM designation, 022^0-68.
Re slurry Tank—Plant Equipment: Tank D-401 is used for reslurrying
aste solids removed in the clarifier. It is identical in size and in con-
traction to storage tank D-10J already described. All the neoprene linings
ere in good condition and hardness tests were not made.
Neoprene-Lined Centrifugal Pumps--Plant Equipment: During the
orrosion tests in the scrubbing systems, Hydroseal pumps were in service;
mpeller diameters were 12, IT, or 20 inches. These centrifugal pumps were
anufactured by the Allen-Sherman-Hoff Company. All wetted parts were
ined with neoprene of a durometer A hardness specified to be ^h to 56 •
When the pumps were dismantled, inspection showed that the linings
ere not damaged severely in any except pumps discussed later. However,
ear of varying degrees was found. The grooving of neoprene linings on
mpellers and casings was least in the TCA system and greatest in the Hydro-
ilter system. General wear of the linings was slight, but a little more
oticeable in the Hydro-Filter system. A durometer A hardness of the
iners ranged up to 16 above the specified maximum; this was fairly general
or the three systems. (The temperature of the linings when tested was
elow the standard of 73°F; this would cause higher values.)
D-20
-------
Packing glands caused severe wear on the stainless steel
components of pumps G-102 and G-202; these are the thickener underflow
pumps for slurry containing about 30$ solids.
The output of slurry feed pumps, G-108 and G-208, had decreased
over a period of weeks. These are rotary screw- type pumps (Moyno) used
for pumping limestone slurry containing about 60$ solids. Inspection
revealed that increased clearance between the stator and the rotor,
because of wear of the rubber lining of the stator, allowed excess
leakage. This was corrected by replacement of worn parts.
Pump G-toL, reslurry tank pump, was dismantled for modification .
The rubber-lined impeller and casing were neither grooved nor worn
appreciably. Hardness values of the linings were not determined.
Some of .the Hydroseal pumps have been replaced since February
with Centriseal pumps produced by the same manufacturer. Sealwater
required at the Hydroseal pumps had added more water to the system than
could be tolerated for closed- loop operation.
Valves — Plant Equipment : The stainless steel check valves at
the discharge of several pumps for each scrubber system were inspected.
These valves are ASTM A-351, Grade CF-8M body, Type 316 plate, and
neoprene seal. Generally, these valves had worn slightly and their
surfaces were smooth and polished.
The U-inch neoprene pinch valve upstream (of FE 106l) from the
bottom slurry header in the afterscrubber tower of the venturi system
showed no signs of chemical attack and only slight evidence of wear.
Piping — Plant Equipment: Neoprene-lined piping was inspected
at the inlet to all pumps that were dismantled for inspection or seal
modification in the three scrubber systems . Elbows, tees, and open ends
of the piping showed no evidence of wear or deterioration. The neoprene
lining 3/16 inch thick with a specified durometer A hardness of 50 plus
or minus 5 was applied by the Rubber Applicators, Houston, Texas.
Hardness values were not determined during the inspection.
Discussion
Process Materials: In the S02 removal plant, the inlet stack
gas, the limestone absorbent, and their reaction products are corrosive
or abrasive. Components of stack gas, such as C02, QS, and SOa^ dissolve
sparingly to make condensate or water corrosive. Fly ash in stack gas
D-21
-------
and the limestone in absorbent slurry are abrasive, especially in high-
velocity streams. Slurry containing limestone, sulfite, sulfate, and fly
ash forms deposits on metal to cause localized corrosion, (in future
tests, chloride in gas or in makeup water should be considered along with
compounds of sulfur as a likely corrosive in areas where it might be
concentrated in a residue.)
Materials of Construction: Materials in the plant consist mainly
of: carbon steel in the inlet duct for stack gas from the power plant;
stainless steel, Type 516L in the scrubbing system ducts, the venturi
scrubber, removable internal parts of scrubber towers, the outlet gas
duct, the fans, and stack; neoprene-lined carbon steel in the venturi
afterspray, TCA, and Hydro-Falter towers; neoprene-lined carbon steel in
the recirculation, clarified process water, and reslurry tanks; Bondstrand
and Type 316L stainless steel downcomers to the effluent hold tank;
Flakellne 103-lined carbon steel in the effluent hold and clarifier tanks;
refractory-lined gas reheater; and neoprene-lined pumps and piping.
Corrosion—Plant Equipment: In general, materials used in
construction of the three scrubbing systems showed good resistance to
attack. Carbon steel ducts vere slightly attacked by inlet stack gas
when at temperature below the dew point.
Inlet stack gas, after being humidified by spray water, attacked
stainless steel ducts and nozzles as follows: slight erosion of bare duct
surfaces; concentration cell-type corrosion (pitting and crevice) of
surfaces underlying deposits; and severe corrosion' and erosion of surfaces
(nozzles or projections) subjected to impingement.
In the venturi scrubber, the limestone slurry and gas discharging
at high velocity corroded and eroded stainless steel parts,.but apparently
iid not damage neoprene lining in the duct.
In the towers of the three systems, slurry and gas flowing at
Low velocity caused only slight corrosion and erosion of bare removable
parts of stainless steel, such as packing supports and FlexiTrays.
Movement of the mobile packing (hollow plastic spheres similar to Ping-
Pong balls) caused some erosion of grid wire in the TCA absorber.
Cause for severe corrosion on the top surface of a chevron-type
Tiist eliminator in the TCA tower is not known. However, it is likely
that some mist, passing through a Koch FlexiTray located below would collect
:>n the chevron mist eliminator and evaporate to form a residue high in
compounds of chlorine and sulfur which would te corrosive. Pits observed
in the outlet duct from the venturi afterscrubber might also have been
caused by such a residue of chlorine and sulfur compounds. Periodic
washing to remove residue might decrease the corrosion.
D-22
-------
Nozzles of stainless steel were more durable than those of
rubber or plastic for spraying limestone slurry in the tovers.
Rubber lining on the tower shells, though coated usually with
slurry solids, was generally in good condition.
Exhaust gas stacks of type Jl6L stainless steel were apparently
in good condition after exposure to gas reheated to between 235° and 265°F.
Sleeves and improved burner nozzles installed at the gas reheaters should
improve fuel oil combustion and thereby minimize troublesome soot deposi-
tion and a potential fire hazard in exhaust gas stacks.
Flakeline 103 linings in the effluent hold tanks and clarifier
tanks were generally in good condition except for cracks near attachments,
such as baffles and weirs, to the walls. Bondstrand downcomers were in
good condition to effluent hold tanks.
Neoprene linings were in good condition in the recirculation,
clarified water, and reslurry tanks. Slight to noticeable wear was
apparent on neoprene-lined agitators in these tanks. Neoprene-lined
piping was inspected near pumps and it appeared to be in good condition.
The neoprene linings of casings and impellers in centrifugal pumps were
not severely damaged, and the least wear was on those in the TCA system
and the greatest on those in the Hydro-Filter system. Decreased output
of limestone slurry (60$ solids) by two rotary screw-type pumps (Moyno)
required that neoprene-lined stators be replaced. The rotors are of
stainless steel.
Corrosion—Test Specimens: In general, the least attack occurred
to test specimens in the TCA system and the greatest to those in the venturi
system. With few exceptions, the greatest loss of weight from metal speci-
mens occurred in inlet duct areas exposed to wetted flue gas and in venturi
outlet area exposed to slurry and gas at comparatively high velocities.
Damage to some nonmetallic materials occurred in these areas also. Slurry
impingement on the specimens caused erosion and corrosion. Pitting and
crevice-type corrosion were minor where erosion and general corrosion
kept the specimens clean.. Generally, in most areas where solids accumulated
in the three scrubber systems, the surface of the underlying specimens
showed localized corrosion. However, each of the 17 alloys tested showed
good resistance at one or more test locations in each scrubber system.
The three rubbers were tested at only six locations; they showed good
resistance in all tests. The maximum service temperature was exceeded
for some nonmetallic materials in the inlet gas duct and exhaust gas duct.
D-23
-------
Table VIII is a summary of data from all the corrosion tests
conducted in the three different scrubber systems. It shows the com-
parative resistance of the materials tested without identifying the test
conditions. Hie metals are grouped into four categories with respect to
decreasing corrosion resistance. The evaluation is based on corrosion
rates determined "by weight loss and/or resistance to pitting, crevice
corrosion, and other types of localized attack.
Corrosion of Hastelloy C-276 was negligible to 5 mils per year;
this alloy showed no evidence of localized attack in any test location.
Next in resistance were the alloys Inconel 625, Incoloy 825., Carpenter
20 Cb-3 and Type 316L stainless steel with corrosion rates ranging from
negligible to 5, 1, ~Lk, and 15 mils per year, respectively. These alloys
had very few pits and/or corrosion crevices. One specimen of Type Jl6
stainless steel was grooved and the weld of another was attacked.
Three nonferrous alloys, cupro-nickel 70-30, Monel 400, and
Hastelloy B, had minimum rates of less than 1 mil and maximum rates of
U9, 57, and 100 mils per year, respectively, with one or two specimens
pitted. In three tests of Monel and in one test of cupro-nickel 70-30,
the welds were inferior to the parent metal.
A group of five alloys that included Type k-k6 stainless steel,
E-Brite 26-1, Incoloy 800, USS 18-18-2, and Type 30U stainless steel, had
rates that ranged from negligible to a "greater than" value which indicates
that the specimen was completely destroyed at one or more test locations.
The values for failures ranged from greater than 1^*0 mils per year for
Type 41*6 to greater than 200 for both USS 18-18-2 and Type 30l*L stainless
steels. These five alloys were highly susceptible to localized corrosion.
Another group of alloys which consisted of Type klO stainless
steel, aluminum 3003} mild steel A^283, and Cor-Ten B had minimum rates
of less than 1 mil per year and maximum rates of greater than 250 for
Type 4lO to greater than 1^00 for mild steel and Cor-Ten B. Pitting
and crevice corrosion occurred on these four alloys.
In general, the stressed specimens (5 alloys only) were not
corroded at rates higher than their counterpart disk-type specimens,
and no cracks were observed.
Of all the alloys tested, Hastelloy C-276 was the most durable
and also the most expensive. Type 3l6L ranked fifth in durability and
about eleventh in cost. Tne values for cost comparison are based on
costs of tubing and sheet with Type 30^ stainless steel as unity (l.OO).
(See Table VIII.)
D-24
-------
Specimens of Bondstrand kQOO, Flakeline 200, and Transite were
tested at 21 locations. Bondstrand showed good resistance in 12 tests
and poor in 9 tests. The evaluations for Flakeline were: 2 good, 14
fair, and 5 poor; and those for Transite were: 14 good, 2 fair, and
5 poor. Only six specimens of each of'the plastic Qua-Corr and of the
rubbers, butyl, natural, and neoprene, were tested. The results were
five good and one poor for Qua-Corr and six good for each of the rubbers.
Summary
Test specimens and equipment exposed for about 6 months in three
test SOs removal systems at Shawnee Power Plant were evaluated for corrosion
and wear.
The most severe damage occurred in plant areas exposed to
humidified stack gas containing fly ash, CO2) ®s.> an
-------
TABLE I
CorruBton Teate Conducted in Che Vanturl Syotem of the Llncatoite - Wet-Scrubbtog
ProeoHu for Sulfur Dioxide Removal
(Teet period --Aug. 12, 1972, to Feb. 2
day a; end Idle ILme-
Corroslon specimens
frun Stack Oau at Shawrioc Povor Plant
, 1975; operating time — l8UO houra or 76.7
-25liO hours or 97.5 daya)
Exhaust
Exposed In ; Inlet gas
Locations (See Fig. 2), Reference No. 1002
Oae
Velocity, ft/sue 2O-6O
Flow rate, 1000 'a of actual ft3/nlo 10-50
Conpoa 1 1 1 on , % by voluine
SOa 0.5
COe 15
0» -l, veld E-Brl te 26-1
E-Drltc, 26-1, weld E-Brtte 26-1, stressed
llantelloy 11, weld lU'.aU-Hoy B
Hastelloy C-276, weld tliintelloy C-S76
Incoloy 000, weld Inconel 82
Ineoloy 825 , we Id In^oloy 65
Inconel 625, weld Im:on.-l625
Mild otci-1 A-285, weld Ef.012
Moncl UOO, weld Munel UOO
Type )Ol>L, w«U Type JOflL
Typ«: 5O4L, wolu Type 5O6l, stix'.ised
Type J16L, weld Type )idL
Typo HO, we Id Typ.- J09
Type W>, weld Type 50V
USS l6-lft-2, weld Inronol 82
Evnluatlon ( Po Iv li'oi rrin> )
Nnojirirni; VIVO (Cliloivprene polymer)
Ce rujtilc
Tr*mii I te (Portland ceun.Tit aad uubetlLou)
.
-
_
.
_
.
> 160
< 1
.
> 290
13
> Ito
_
28
< 1
W, -o
< 1
-
< 1
> 550
10
65, -e
-
1
-
> 1)0
> 105
> Ho
Poor
Poor
-
-
-
-
Poor
Oas and
a pray j
1011
60-170
15-1.5
8-25
0.2
12
69
it
15
0.02
_
.
.
.
.
.
> 550
lit
_
> Hoc
1.9
> 190
.
100
5
> 190
7
-
5
> 1MX)
57f
> 200
-
15, -f
-
> 250
> li*O
> 200
Poor
Poor
Pool-
Good
Oood
Good
Poor
gas Effluent
heated) liquor
1007 1008
255-265
SO -f>0
1O-50
0.2
12
69
15
0.02
90-150
_
.
.
.
_
.
P7o
< i
<£ i
16, -*
pie, -J
Pm, -4
<: i
-^
< 1
£. 1
1
< 1
16, -<=
"<
1
2
< 1
£. \
5, -4
i
2
Poor
poor
-
-
-
-
Good
H-10
5.6-6.8
1.0-2.0
1.5-5.5
90-96
P2^, -C
Neg.
Neg.
I1!, -C
1
< 1
Neg.
2, Pm
Neg.
< 1
< 1
Neg.
< 1
18, -c
1
< 1. -c
Neg.
Heg.
P12, -c
P5, -c
Good
Fair*1
Oood
Good
Oood
Oood
Good
Liquor
Recycle In
liquor clsrlfler
L012 1013
70-125 70-1OO
b-10
5.6-6.5
1.0-2.0
1.5-5.5
90-96
< 1
Heg.
-
12
< 1
Meg.
-
< 1
< 1
< 1
< 1
-
Neg.
19
< 1
Neg.
-
< 1
-
Pi, -e
Neg.
Heg.
Good
Fair*
-
-
-
-
Oood
0-20
6.0-7.0
0-k
0-7
O-10
80-100
P12, -c
Neg.
-
5, -°
1
Meg.
.
< 1, PB
Heg.
Neg.
< 1
-
Neg.
5
1, P2
Neg.
-
Neg.
-
< 1, -C
P10, -c
Good
Fair1"
•
-
-
-
Oood
11 No uprny w>»ii.T wmi uiiud to humidify the «"" when the temperature was 275'-55O*F; temperature waa 125°F for
•j&'t htAirn when .'Iir'ny water wan uneJ.
b The 'Vi-enier limn" (>) lUrfu I" uued wiiun u upeclmun wae completely deatroyed. "P" preceding a number Indl-
i:m.i!n plitliirf durliiK tliu uxpoiiurc period to the depth In mils shown by the numbsr, und "Pa" Indicates minute
pita. "NeK.," ne^Udll'le, no weight loua or localized attacK.
c Crevlue corroiilun "t Tuflun Inuulntor.
d llu-ru wou oorao. welijnt looo ol1 auacloeu due to wear at ceatar hola aftor Tefloa Inaulatore failed vhea over-
heated.
e Bevciv loi:»Hzf!d Bli.iu:k of parent metal .
f AttMCk of weld.
8 Evaluation: Oo.)d, little or no i:han«je In condition of specimen) fair, definite change, probably could be
used; poor, I't.lled or ueveruly domiiged.
Evidence of ubrimlon on edge of opeclraen.
D-27
-------
Con-onion Taata Conducted In the TCA System of the Umeetone - Wet-Scrubbing
Pi-ocees for Sulfur Dioxide Removal from Stack Oaa at Shawnei; Power Plant
(Teat period—Aug. 12, 1972, to Feb. J, 1973; operating Llme--l667
houru or 69.5 days; and idle time—2555 houra or 1O5-6 days)
>slon ape.'lneqs
joeed In
:atlons (See Fig. 3), Reference No.
Liquor
Exhaust In
Inlet Gao and Gas KOI) Oaa and gas Effluent Recycle clarl-
gae liquor dropleta mist (heated) liquor 1Iquor fler
2O02a 2006fe 2005 "2005 2007 2008 c 2012 201}
iperature, °F
.oclty, ft/sec
jw rate, lOOO'n of actual ft3/mln
ipOBltlnn, % by volume
iOa
Ply ash, gr/standard ft3
iperature, *F
Llde , (by weight
iposltlon, % by weight
:a303 'l/SUgO
/ater
>aion rate of metals , mlla/yr
imlnum 3O03, veld ER11OO
-penter 20Cb-5, weld Carpenter 2CCb-3
rpenter 20Cb-J, weld Carpenter 20Cb-}, streased
•-Ten B, weld E8018-C3
>ro-aickel 70-30, weld B259 RCuMi
Srlte 26-1, weld E-Brlte 26-1
i ielloy B, veld Hastelloy B
i telloy C-276, weld Raatelloy C-276
:oloy 80O, weld Inconel 82
Ld ateel A -283, weld E6012
vel Uoo, veld Konel UOO
>e 3QltL, weld Type 308L
>e y>^L, weld Type 30flL, atreaaed
>e J16L, weld Type JlCL
>e 316L, weld Type 316L, stressed
>e 'ilO, veld Type 509
>o '<'i6, ueld Type 309
! 18-18-2, weld Inconel 82
Ltlon of nonmctalllc materials J
tatlca
ionds t, rand '•000 (Fiber glass-reinforced epoxy).
Tlakellne 200 (Inert flakes and polyester resin).
iua-Corr (Fiber glans— reinforced furan restn)..
ibers
Jutyl 36,666 (Copolymer of loobutylene-lsoprane ) .
latural 1}75 (Polylsoprene)
leoprene 9150 (Chloroprene polymer)
runic
Pranaitc (Portland cement and asbestos)
260-310
25-60
15-30
0.3
13
U.5
a
3-5
-
2
< 1
15, Po
2
< 1
< 1
< 1
< 1
< 1
-
< 1
17
1
< l, Pm
.
< i
_
1, Pa
c 1
< 1
Poor
Fair
-
_
-
.
Oood
70-125
6-12
11-22
O.O5
12
69
15
O.O2
-
"•, -'
Neg.
^
< 1 -^
P2, -•
6
< 1, -*
-
Neg.
26
1
P6, -•
Keg.
Pm, -•
P5, -•
pit, -e
Poor
Poor
-
_
-
-
Poor
70-120
5-10
11-22
0.05
69
it
15
0.02
-
I.
< 1
21, Pi
e, -»
PS
.
6
Heg.
-
Heg.
23
1
P13, -e
-
< 1
-
P6, -c
P19, -"
P8, -e
Good
Fair
-
-
-
-
Fair
70-120
it -8
11-28
0.05
12
69
it
15
0.02
-
26, P20
< 1
268
17
1, Pa
Pa
13
Neg.
P19, -e
1, -
Keg.
£50
15, -*
**, -e
Pm
Pl6. -
P16, -e
Good
Fair
Good
Good
Good
Good
Poor
235-265
25-60
15-30
0.05
69
It
15
0.02
-
1, P2
< 1
< 1
2, Fa
< 1
P2
Pm
< 1
Heg.
Pm
Neg.
< 1
< 1
3, Pfc
< 1
P16
< 1
< 1
< 1
P12, -
P6
P5
Poor
Oood
-
-
-
-
Good
110-125
6-10
5.9-6.3
1.0-2.0
2.0-3.5
3.0-U.5
90-9'"
20
•eg.
2
170, Pm
3
Beg.
< 1
2
Neg.
Neg.
Neg.
< 1
Neg.
210, Pa
2
Pa
Neg.
< 1
< 1
Pa
< 1
- 1
Qood
Fair
Oood
Oood
Good
Oood
Oood
85-125
7-15
5. 6-6. It 6
1.5-3.0
2.5-5.0
3.0-7.0
85-93
2, P2
< 1
5, Pm
< l
Neg.
-
< 1
< 1
P2
Neg.
-
Neg.
J
< 1
P10, -
-
< 1
fte, .«
P7, -°
Pm, -e
Good
Fair
-
-
-
-
Oood
85-100
0-30
i. 3-7.6
0-6
0-10
0-15
70-100
1
Neg.
6, >
< l
Neg.
-
< 1
Heg.
Pm
Beg.
-
< 1
8
< 1
Neg.
-
Heg.
-
P'*, -e
< 1
< 1
Good
Fair
-
-
-
-
Good
spray water wan used at teat location ^OO2 during the corroalon t«at period to humidify the gas.
:aufie teat specimens were worn liy movement of plastic balls during the period 8/12 to l]/}/7£, new epeclmens were
italled 11/17/72. The ilata given were determined from the last 995 hours of operation.
: specimens were Iranersed In the Blurry only during the laat kU. 5 hours of operating time, and the corrosion rate was
:ermlned on this basis. However, the high rates for aluminum, Oor-Ten B, and mild steel Indicate that these alloys
re corroded durlr.p, idl« time also.
t "greater i.han" (-•) sign la used whon a specimen was completely destroyed "P" preceding a number indicates pitting
ring exposure period to depth In mils shown by number, and "Pn," minute pita. "Neg.," negligible, no weight losa or
:allzed attack.
2vicc corrosion ar. contact with TeHf lir.,At—aiTcctnil zr,ny of weld.
xr on u'i^i- oi' :;;t<.'(jl[fi'.-n iiue to movement of plastic balls.
/e of specimen iltunagi-'l liy impact tit sharp object.
Lack or wslrt.
iluat.i..u: 'i'jo.l. lliMc or im L-htjinr- in condition ot' apeclmeiv, fair, del'ir.lte change, probably .:oiold be us^.-d;
ir. failt:'i oi- .(t.>/ei-eJy -luiLa^c..!.
D-28
-------
TAM£ III
Cui-ixiiiliin Tnnl.ii CunduirU'd In tin; Hydro-niter flyalon i.f the Llmonluie - Wet-Scrubbing
I'roi-.Miii I'or Bull'ur ntoxlilt; Hmncjvii.l fna ntunk GBH a<, llhaunun Puwr,r Plant
(IVnt period--Aug. IS, I'/f?, to Fob. 1, 197>; opnmtlrg Llrn.:--S!?0'i
hiiurM or 91.0 dayai and Idle (.loo-.1950 houro or 8l.J dnytt)
uyrcimuno
Kxpoiiod in Ii
LorntloiiR (See Fig. 'i), Reference No ~
0«a
Velocity, tl/Hvf ...................
Flow i-nte, 1000 "n of ai-1.ua 1 fl.3/nln
Composition, % by volume
COB
Ne
H«0
fly ash, gr/ntandard ft'
120-150
20-1*5
10-S5
0.)
1}
6
3-5
0.1
12
69
U
15
0.02
Liquor Exhaust
and Inlet Gas and gss Effluent Recycle
(hasted)
3007
95-135 95-130 2)5-265
}-8 )-8 25-60
1O-8J 1O-2J 12.5-50
0.1
12
69
15
O.O2
0.1
12
69
k
i5
0.02
Temperature, *F
Bollrtn, * by weight
pH
Ccrapo.Mtlon, % by weigh',
CoSO, -rilaO
CwSOn' 1 /2Hj>0
Uni-ca.-i.frt l.lm>.>nt.uiu> plus fly nnh
Wnti:r
75-125 75-125 75-100
14-10 fc-10 0-20
5.6-6.U 5.J-6.U 6.1-7.<»
CorroBlon rate nf \
'^ mtls//r
Aluminum 3001, veld ERUOO
Cni-pci>''.r 20Cl>-:>, weld Curpenter 20Cb-3
Cm-pouter 20Cb->, vc)<1 Cnrpenter 2OCb-3, stressed
Cot--Tr:n B, weld EflOlO-C3
Cupro-nlckel YO-.W, wold BJ59 RCuNl
E-BrHe Mi-l, wrld B-Pi-1 tc 26-1
E-Brllc 26-1, weld E-Drlte 26-1, stressed
Kac.t.rlloy B, weld Hnnt.el.loy B
Hustclloy C-276, weld Hastelloy C-276
Incoloy 800, weld Inconel 82
Innoloy R25, wnld IncoAoy 65
Inc-oloy f)25, weld Incoloy 65, stressed
Inconel 625, weld Inconel 625
Mild steel A-?8j, weld E6012
Monel ''OO, weld Honel <(OO
Type SO'iL, weld Type VD6L
Type 30!iL, weld Type 30flL, a tressed
Type 316L, weld Type }16L
Type 316L, weld Type J16L stressed
Type 1*10, weld Type 309
Type l<>46, weld Type 509
USS lfl-18-2, weld Inconel 82
Evaluation of nonmetalllc materials6
Plastics
Bondstrand 1)000 (Fiber glass reinforced epoxy).
Flakellne 200 (Inert flakes and polyester resin).
Qus-Corr (Fiber glsse reinforced ftiran resin)..
Rubbers
Butyl 26,666 topolyner of Isobutylene-lsopreue).
Natural 1575 (Polylsoprene)
Ncoprone 9150 (Chlgroprene polymer)
Cornjnlc
Trannlte (Portland cement and asbestos)
1.0-2.0 1.0-2.O
- 1.5-3-5 1.5-3.5
90-96
> 160
U, Pa
.
> 300
35d
> 130
71
< 1
93
2
.
< 1
> 300
30, -d
> Ik)
.
018, Fro
_
> 100
> 90
> 120
P9
Neg.
_
UO
1
< 1, PI
.
2
Heg.
< 1, P19
Neg.
-
Heg.
37
< 1
P12, -°
.
Neg.
-
P8, -c
P9, -'
*, -c
pie
Neg.
< 1
13
1
Neg.
< 1 -c
U
Neg.
< 1
< 1
< 1
< 1
it
2
P10
P18, -c
Neg.
< 1, -c
P3, -c
PiO, -c
P9, -c
5,-c
PD
< 1
14 -C
3
P7
<1, P18
2
< 1
1
< 1
1
< 1. Pm
u, -c
3
1, Po
< 1, Pn
< 1
< 1, Fa
3, -c
1, Pm
2
Poor
Poor
Poor
Good
Fair
Fair
Oood
Fair
Ooodr
Oood
Good1
Ooodr
Oood
Poor
Good
Good
2, -
Neg.
< 1
5, -c
< 1
Neg;
< 1
< 1
< 1
< 1
< 1
< 1
Neg,
< 1
< 1 -C
PS! -c
< i
< i
Pm, -c
Pn,' -c
Good
Fair
Good
Good
Oood
Good
Oood
0-U
0-7
o-:o
90-96 60-100
< 1
Neg.
10
< 1
< 1
< 1
Neg.
Neg.
Neg.
Ne8.
11
< 1
Neg.
P7, -<=
_c
< 1 -c
PS
Heg.
7
1
< 1
< 1
< 1
Neg.
< 1
Neg.
9
< 1
Neg.
Pl6, -c
P9 >c
K, -c
Good P1J, Poor
Fair Fair
Good
Good
a Rpray water waa used at all tlmee at test point J002 to humidify the gas.
*> The Vrcater than" (>) elgn Is ueed when a specimen was completely destroyed. "P preceding a number Indicates pitting
during' the exposure period to the depth In mils shown by the number, and "Pn" Indicates minute pits. Ntg- , negligible,
no weight loss or localized attack. "0," groove of depth In mils shown.
c Crevice corrosion at Teflon Insulator.
d Attack at weld. . ... , . .
« Evaluation: Oood, little nr no change In condition of specimen-, fair, definite change, probably could be used;
poor, failed or severely damaged.
T Bpecimen was damaged by Impact of shorf! Instrument during the exposure period.
D-29
-------
TABLE IV
Compositions of Allovs Tested in the Limestone - Wet-Scrubbing Systems for Sulfur
1.
2.
5-
4.
5.
6.
7.
D
I* 8-
o
9.
10.
11.
12.
15-
14.
15-
16.
17.
Allovs
A 1 liml mm 3003
Carpenter 20Cb-3
Cor -Ten Bb
Cupro- nickel 70-30
E-Srite 26 -lb
Hastelloy 3b
Hastelloy C-276b
Incoloy 800b
Incoloy 825b
Inconel 625
Mild Steel A-283b
.Monel 400b
Type 504L
Type 5l6L
Type 4lOb
Type 446b
USS lB-l3-2b
Dioxide Removal from Stack Gas at Shavnee Pover Plant
viicniicfil sins lv si s t ^
c
_
0.07a
0.066
-
< 0.001
< 0.01
0.002
0.04
0.04
o.ia
0.17
0.09
0.050a
0.030*
0.062
0.10
0.065
Cr
.
19-21
0.52
-
26.17
0.19
15-87
21.11
22.28
20-23
-
-
18-20
16-18
12.7
24.6
18.2
31
_
52-53
0.018
31.00
0.08
Bal.
Bal.
51.52
42.22
Bal.
-
64.66
8-12
10-14
0.16
0.50
18.0
Fe
0.7tt
Bal.
Bal.
0.55
Bal.
5.75
5-96
45.01
28.30
5.00a
Bal.
1.00
Bal.
Bal.
Bal.
Bal.
Bal.
Cu Mo
0.2a
3-4 2-3
0.31 o.oio
67-79
0.01 i.oo
26.20
16.52
0.40 _
2.12
8-10
0.057
33.06
-
- 2.0-3.0
0.05 0.054
0.045 0.10
0.05 0.018
"j-.
1.0-1.5
2.00a
1.20
0.52
0.01
0.53
0.49
0.84
0.56
0.5"
0.48
1.03
2.00a
2.00a
0.45
0.71
1.50
a
0
1
0
0
0
;
.6*
.00°
.29
-
.19
.01
< 0.01
0.31
0.34
0
0
.5a
.070
P
_
0.055*
0.012
0.003
0.010
0.005
0.012
0.015a
0.015
0.08
1
1
0
0
1
.00*
.00a
.40
.37
.94
0.04 5a
0.04 5a
0.014
0.018
O.OC7
S Al Ti Ctr.-rs
a
Bal. _ Zn 0.1 , Total C.lj
0.035* - - Cb + TaSxC
0.031 0.056 - v 0.05
0.005 - - Zn 0.054, Pb C.002
0.012 - N 0.010
0.006 - CO 0.35, V 0.26
0.010 - Co 1.34, W 3.51, V C.25
0.007 0.48 0.46
0.007 0.06 0.66
0.015* 0.4a 0.4a Co l.oa, Cb + Ta 3.1= - t. 15
0.024 0.005
0.008 o.oo4
0.050a
0.050*
0.018 0.069 - s 0.034, v < 0.03
0.010 0.008 < 0.02 K C.13, V < 0.03
0.009 0.001 - N 0.04
, Maximum,
Analysis was supplied with the material received for use in corrosion tests.
-------
Table V
Analyses of T-eposits in Llaestone - Vet -Scrubbing Systecs for Sulfur Dioxide Repoval from Stack Gas at Shavnec Power Plant
Identification of sample
tion, '•-. by veight
Bat* JJueber
Venturl Systen
2/1,73 VD 2173
2/23/73 VD 2 2373
TCA System
2/3/73 TCA D 12373
2/3/73 TCA D 22373
2/5A5 TCA D 1257J
2/5/73 TCA D 22573
2/22/73 TCA D 122273
2/12/73 TCA D 121273
2/12/73 TCA D 221273
Hydro -Filter, Systen
2/2/73 HTD-2273
2/3/73 KFD-2373
2/23/73 KFD-22373
Location
Scale from recirculation tank D-10U.
Soot from gas duct about 25 feet above reheater.
Scale froa recirculation tank D-20« (Test 2012).
Scale froa spool of corrosion specimens (Test 2006)
Scale froa scrubber wall below and near Koch tray.
Scale from grid-vail junction, elevation 396 feet
1-1/2 inches.
Rust-colored scale froo corroded deraister.
Tar -like material from duct 25 feet downstream
from reheater.
Tar -like material downstream fron and near reheater.
Scale from corrosion spool above marble bed.
Deposit fron botton slurry nozzle.
Soot from gas duct about 25 feet above reheater.
-C3i» or Siua^e a»pos:t Sect aa^-erialO
CaO SC, SC^ CCfe M*0 <^r.ers Asb IbC hydrocarbon
25. U 9.0 37.3 1.2 - 25-5
- - - - - - 56.0 12.8 31-2
(6U.2) - (35.3)
31.3 10.1* 23.7 6.3 - 23-l> -
23. 5 17.6 27.1* 0.5 - 25-9
39.7 16.8 25-9 2.0 0.22 15. li - -
1»1.2 15-8 21*. * 6.2 0.21 12-5 -
3.6 l» .2 83.2 o.o 9.0
25-2 7.8 67.0
(27.3) - (72.7)
39.3 1-7 59-0
(UO.O) - (60.0)
29.5 17.6 27.1* 0.5 - 2k. 9 - -
23.8 9.6 Ui.7 2.2 - 17.7 -
- 36.1* 11. T 51-9
(M.2) - (53.8)
General Operation Equipment
3/1/73 5173
Scale from reslurry pun? C-^01.
39.3 0.5t 12.3 20.1 - 26.7 -
Information taker- fron reports dated .'/arch and May 197? by R. E. Wagner of inspections made February 1, 2, and 3, 1973 of the Hydro-Filter, the venturl,
and the TCA scrubber systems.
Values In parentheses are on a dry basis.
-------
VI
Hnrdnei.o of Neoriviio Linings of Equipment In the Three Limestone - Wet-Scrubbing
for Sulfur Dioxide Removal frcm Stack Quo nt Shnwnog Power Plant
(Exposure period! Aug. 1?, 19Y2, to Feb. }, 1975)
Teet
Locntton of hordnena tent
temp . , DuroBotar "A" hardness
• p 8 Original Final"
Venturl System (iflliO Operating llouro)
Aflorecruhl'er Tower:
Eight, Inches below Type JlCL stainless steel at venturi section ... .............. . 60 to 65 67
Above tropout. l.niy (npproxlmntc- elevation 588 foot) ............................. . 60 to 65 53 to 60
Threp tntthCR beJow mint ellminat,or .............................................. . 60 to 65 5U to 60
Throe fret nbove mint ollmlnntor ................. . .............................. . 60 to 65 53 to 55
Four Inches below Type 5J6L utalnleos eteel duct to reheater ..................... . 60 to 65 52 to 5<4
Clarified Process Water Storragc Tank, D-105:
Above liquid level [[[ 60 55 to 60 6U to 65
Below liquid level [[[ 60 55 to 60 6l to 6j
Reclrculntlon TnrtX:
Five feet above bottom . [[[ . 55 to 60 65 to 69
Elides of ftgl In tor [[[ . . .. 60 to 70 6} to 67
Bladoo of AgUnlor In:
Effluent hold t.unk D-101 [[[ 60 to 70 65 to 66
Reclrculot Ion tnr* D-10'i . . . . [[[ . 60 to 70 6 J to 67
Pumps 0-1O1, 0-10?, G-10J, nnd 0-101!:
Impollcrn find cnnliigg [[[ . 511 to 56 60 to 70
New ImppllerB nn<\ fnalngn (nparPB ) .............................................. . - 61* to 72
TCA Syatom (1^fi7 Operating Hours)
Sci-ubber Towur-
HliVllu of w,il I nnil 6 Inchoo bi;low nuinwn/ ........................................ - 60 to 65 55
Six In'-M'T, nbnvu bottom | lnr;hcu below reducer ....................................... }6 60 to 65 61* to 67
Ifcrun Inchon bulow Type )l6L atalnleee steel stack ..................... , ........ ' 56 60 to 65 65 to 70
Clarified Prcii.'uHn Wntor Storrnjc Tunk, D-)05:
Above 11 qu Id level [[[ 60 55 to 60 60 to 61
B'llow liquid levul [[[ 60 55 to 60 58 to 60
n^clrculRtloij T'ink, D-iO1':
Five fuut nooxo bottom [[[ - 55 to 60 75 to DO
Bladuu of Bgltntor [[[ 60 to 70 67 to 70
Blnilcu of Atflt.ntor In:
-------
TABLE VII
Hardness8" of Rubber Lining*1 Specimens Tested in the Limestone - Wet—Scrubbing
Systems for Sulfur Dioxide Removal from Stack Gas at Shawnee Power Plant
(Exposure period—^Aug. 12, 1972, to Feb. 3, 1973)
Location of
specimens0
As received
Butyl
(26.666)
(55)
Durometer "A" hardness
Natural
(1375)
3**-37 (35)
Neoprene
(9150)
Venturi System (181+0 Operating Hours)
1008
1011
56-58 (57)
5^-58 (56)
Ui-43
60-63 (61)
62-63 (62)
TCA System (1667 Operating Hours)
56-59 (57)
5^58 (56)
2008
38-42 (Uo)
36-39 (38)
59-61 (60)
59-6U (62)
Hydro-Filter System (2203 Operating Hours)
3005
3008
55-58 (57)
(56)
35-39 (38)
37-^0 (38)
58-62 (60)
61-65 (63)
a Four tests were made of each specimen in the laboratory at 78°F with a
durometer type "A2, " ASTM D2240, manufactured by The Shore Instrument and
Manufacturing Company. Values in parentheses are averages.
*> Specimens of butyl, natural, and neoprene liners were applied on mild steel
coupons by the Gates Rubber Company.
c See reference numbers on Figures 2-4.
D-33
-------
TABLE VIII
Compilation of Corrosion Data of Materials Tested in the Three Limestone - Wtt-gcrubblng Systems
for Sulfur Dioxide Removal from Stack Gas at Shavnee Power Plant
Metals'*
stelloy C-276
:onel to'j
:oloy (<25
•pentcr 20Cb-5
, sr,
Cost comparlBonb
A B
9.29
6.59*
li.'ifif
U.21
>ro-nickel 70-30 1.80*1
2.9V
icl liOO
Jlelloy H
>e li'ifi RS
irlte 26-1
:oloy BOO
! lfl-18-2
>e JOUL GS
9.'"7
2.5^
1.11
>e 'ilO SS 1.92
uninum "5OOJ
.d Steel A-28J O.Jlif OflOf
•-Ten B
6.05
3.73
3.73
1.61
3.61
1.85
2.70
1.11
0.93
Corrosion*
On basis of
weight loss,
mlls/yr, range
Neg. to 5
Neg. to 5
Ne«. to 7
Neg. to I1!
Neg. to 15
< 1 to ^9
< 1 to 57
< 1 to 100
Neg. to liO
Neg. to 190
Neg. to 190
Neg. to 200
Neg. to 200
< 1 to > 250
< 1 to > 550
< 1 to > ll*00
< 1 to > lliOO
Specimens pitted
No.
_
1
_
2
3
1
1
2
9
10
6
11
Ik
15
9
2
5
Depth,
Minimum
.
_
.
-
-
-
Minute
Minute
Minute
Minute
Ml nute
Minute
Z
.
Minute
mils"
Maximum
Minute
_
Minute
Minute
18
2
Minute
19
18
19
16
23
16
70
Minute
3
Specimens
with crevice
attack. No.
.
1
_
2
.
1
-
11
2
3
11
11
16
5
2
li
Specimens vlth other
types
No.
.
.
.
I, 1
1
J
-
.
.
1
.
1
.
-
.
1
of attack
Area
-
_
.
olft8, weld
Weld
Weld
-
—
.
_h
_
_h
_
-
_
Weld
Condition
.lie Materials Good Fair Poor
idatrand <4OOO
.kellne ZOO
i-Corr
12
2
5
lit
9
5
1
,urfll 1575
iprene 9150
6
6
fi
nalte
1U
ompllatlon la baaed on 21 tests of each material except for Qua-Corr and the three rubbers (butyl, natural, and
ene) vlth which only six teats were conducted.
comparison values are baaed on Type 30li stainless ateel having a value of 1.00. "A" la based on commercial quality
f»--10,000 feet of 3/li-lnch outside diameter by O.C65-lnch average wall which la cut to 20-foot 0-lnch lengths.
irmatlon from Carpenter Technology Corporation (August 7, 1973K7 "?" i« baaed on 1/8-lnch sheet In 20,000-pound
_/Informatlon from J. M. Tull, Atlanta, Georgia, by telephone (July 2, 1975)j7
ctual depth of penetration In mils during expoiure periods Is indicated in Tables I, II, and III.
s are Hated In their approximate order of decreasing corrosion resistance.
(1.
.ess.
iove In parent metal was IB nils deep.
•e localized attack of parent metal.
D-34
-------
HYDRO-FILTER
(FLOOOED-BED
OF MARBLES)
TCA
(MOBILE BED OF
PING-PONG BALLS)
VENTURI
(FOLLOWED BY
AFTER-SCRUBBER)
FIGURE I
THREE PARALLEL SYSTEMS OF LIMESTONE-WET-SCRUBBING
TEST FACILITY AT SHAWNEE POWER PLANT
D-35
-------
TOP OF STACK
DUCT-40 01 A.
(TVPE3I6LSS)
LOCATION OF TEST
— * SPECIMENS
a (RACK.SPOOL OR STRESSED)
SPECIMEN OMITTED IN CURRENT RUNS
(AUG 8, 1972--- FEB 2. 1973)
CARBON STEEL ASTM A- 283
e TEST 1013 WAS CONDUCTED IN
CLARIFIER TANK O-IO2 NOT SHOWN
PRESSURE
SAFETY
VALVE (PSV.
24" BUTTERFLY)
CATWALK
(TO POWER
BUILDING)
GAS INLET OUCT(4O"DIA .IOGA
•>
CARBON STEEL
TYPE 3I6L S.S.
( A TO B )
LOWER
ACCESS OOOR
SPOOL
VENTURI SECTION
(TYPE3I6L SS)
NEOPRENE LINED
(CARBON STEEL..B-C-O
REClRCULATION -»
TANK (D-104.
NEOPRENE LINED
CARBON STEEL)
GROUND LEVEL
EL. J4J'-0"
FIGURE 2
VENTURI SCRUBBER SYSTEM. (C-IOI)
D-37
-------
,TOP OP STACK
LEGEND:
C h X LOCATION OF TEST
^—-. SPECIMENS.
0 (RACK. SPOOL OR STRESSED)
0 CARBON STEEL ASTM A-283
b TEST ZOIS WAS CONDUCTED IN
CLARIFIER TANK D-SO2 NOT SHOWN
REHEATER (P-20I.REFRACTOFW
LINED CARBON STEEL
SHELL. INSULATED)
73Ve"00 ,67 -
FLEXITRAY - ^
OAS INLET DUCT (40" .^
DIA.. 10 CA. CARBON ~\
°
EL. 397'-IO"
TYPE 316 L S. S - : -
I A TO B)
ACCESS DOOR
SPOOL
•' • 1
D. FAN
YPE 3I6LSS)
UCT-40"DIA
YPE 3I6L SS)
OOL
loon
.TRESSED"
YPE. 3I6L SS)
3(
1
?" "'-
L
82
)'-6'
b"
6'-H"SO INSIDE
-RUBBER LI NINO
-------
/TOP OF STACK
LEGEND:
C5ZX LOCATION OF TEST
^~T"" SPECIMENS
D( RACK, SPOOL OR STRESSED)
0 CARBON STEEL ASTM A-2B3
b TEST 3013 WAS CONDUCTED IN
CLARIFIER TANK D 302 NOT SHOWN
i
STRESSED -
REHEATERIF- 301. REFRACTORY
LINED CARBON STEEL
SHELL IMSUl ATFO) ....
73V 0.0. 67'/z"' D
OAS INLET OUCT(40"
DIA..IOGA. CARBON — \
STEEL") ._\
'-0' ^S
TYPE 316 L S3- »
(® TO ®)
(355ZS..,
ACCESS DOOR— --^U
SPOOL - — "
6'-ll'/j"SO. INSIDE ^
^>s._.
1
GRID
(MARBLE SUPPORT) —
T~~"
GSDC?
SPOOL —
RECIRCULATION
TANK (0-304,
NEOPRENE-LINEO *
CARBON STEEL)
®
E
a"
Tfi
V
i
r
i
^-
U
i
p
-------
*>.
UJ
FIGURE 5
TYPICAL ASSEMBLIES OF CORROSION TEST SPECIMENS
-------
0
I
**
Ul
h
FIGURE 6
CORROSION TEST ASSEMBLIES AND SUPPORTS READY FOR INSTALLATION IN PLANTS
-------
1007
d
i
DISK SPECIMENS AFTER EXPOSURE
FIGURE 7
IN VENTURI SYSTEM (AUG. 12, I972--FEB. 2, 1973)
-------
1007
d
i
FIGURE 8
STRESSED AND COATED SPECIMENS AFTER EXPOSURE IN VENTURI SYSTEM
AUG. 12, I972-- FEB. 2, 1973
-------
FIGURE 9
DISK SPECIMENS AFTER EXPOSURE IN TCA SYSTEM (AUG. 12, I972--FEB. 3, 1973)
-------
c
I
U1
FIGURE 10
STRESSED AND COATED SPECIMENS AFTER EXPOSURE IN TCA SYSTEM
AUG 12, I972--FEB. 3, 1973
-------
d
I
un
Ul
FIGURE II
DISK SPECIMENS AFTER EXPOSURE IN HYDRO-FILTER SYSTEM (AUG. 12, I972--FEB. I, 1973)
-------
u
Ul
FIGURE 12
STRESSED AND COATED SPECIMENS AFTER EXPOSURE IN HYDRO-FILTER SYSTEM
AUG. 12, I972--FEB. I, 1973
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
i. REPORT NO.
EPA-650/2-74-010
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
EPA Alkali Scrubbing Test Facility: Limestone
Wet Scrubbing Test Results
5. REPORT DATE
January 1974
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Dr. Michael Epstein, Louis Sybert,
Dr. Shih-Chung Wang, and Charles C. Leivo
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Bechtel Corporation
50 Beale Street
San Francisco, California 94119
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ACY-32
11. CONTRACT/GRANT NO.
PH 22-68-67
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; Through January 1974
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT!^ report describes test results through early 1/74 from a prototype lime/
limestone scrubbing test facility for removing SO2 and particulates from flue gases.
The facility consists of three parallel scrubbers—a venturi/spray tower, a Turbulent
Contact Absorber (TCA), and a marble-bed scrubber—each able to treat a 10-Mw
equivalent (30,000 acfm) of flue gas from a coal-fired boiler at TVA's Shawnee Stat-
ion. The short-term (less than 1 day) limestone factorial tests , completed in 2/73,
were conducted at high (6.0-6. 2) scrubber inlet liquor pH. Longer term (about.500
hours) limestone reliability verification tests, completed in 9/73, were conducted ?.t
reduced (5. 6-5. 8) scrubber inlet liquor pH, to increase system reliability and lime-
stone utilization. As of early 1/74, more than 1000 hours of a long-term
limestone run on the TCA and more than 2000 hours of a long-term lime run
on the venturi/spray tower system were completed. The objective of testing since
2/73 has been to identify the most economically attractive lime/limestone system
operating conditions, consistent with reasonable performance.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution
Coal
Calcium Oxides Boilers
Limestone
Washing
Sulfur Dioxide
Flue Gases
Spray Tanks
Test Facilities
Prototypes
Scrubbers
Absorbers
Air Pollution Control
Stationary Sources
Particulates
Venturi/Spray Tower
Turbulent Contact
Absorber
Marble -Bed Scrubber
13B
7A
14D
8. DISTRIBUTION STATEMENT \
19. SECURITY CLASS (ThisReport)
21. NO. OF PAGES
Unlimited
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (9-73)
------- |