Tennessee
Valley
Authority
United States
Environmental Protection
Agency
Office of Power
Power Research Staff
Chattanooga, Tennessee 37401
Office of Research and Development
Office of Energy, Minerals, and Industry
IERL, Research Triangle Park, NC 27711
PRS-19
EPA-600/7-77-019
March 1977
TVA'S 1-MW PILOT PLANT:
FINAL REPORT ON
HIGH VELOCITY SCRUBBING AND
VERTICAL DUCT MIST ELIMINATION
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
A. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 - Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations Include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
PRS-19
EPA-600/7-77-019
March 1977
TVA'S I-MW PILOT PLANT:
FINAL REPORT ON
HIGH VELOCITY SCRUBBING AND
VERTICAL DUCT MIST ELIMINATION
by
G. A. Holliden (Project Director)
R. F. Robards (Project Director)
N. D. Moore
T. M. Kelso
R. M. Cole
Tennessee Valley Authority
Power Research Staff
Chattanooga, Tennessee 37401
and
Office of Agricultural and Chemical Development
Muscle Shoals, Alabama 35660
Interagency Agreement No. EPA-IAG-D5-072I
Program Element No. EHB528
EPA Project Officer: John E. Williams
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
This study was conducted
as part of the Federal
Interagency Energy/Environment
Research and Development Program
Prepared for
OFFICE OF ENERGY, MINERALS, AND INDUSTRY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
DISCLAIMER
This report has been prepared by the Tennessee Valley Authority and
reviewed by the U.S. Environmental Protection Agency and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the Tennessee Valley Authority or
the U.S. Environmental Protection Agency, nor does mention of trade
names or commercial products constitute endorsement or recommendation
for use.
ii
-------
ABSTRACT
TVA has recently demonstrated washing techniques that maintain
continuous mist eliminator performance for lime/limestone closed-loop
scrubbing systems at its 1-MW pilot plant at the Colbert Power Plant.
The systematic test program which developed these washing techniques
is reviewed for both the limestone and lime systems. High velocity
scrubbing tests were also performed in conjunction with the mist
eliminator tests. Continuous operation of the Chevron-type mist
eliminator, positioned horizontally in a vertical duct, in the lime-
stone system was maintained (after extensive testing at 12.6 ft/sec)
by washing the bottom of the mist eliminator intermittently with all
the available clarified liquor immediately followed by an allocated
amount of makeup water. The top of the mist eliminator was washed
intermittently with the remaining allocation of allowable makeup water.
At a gas velocity of 16 ft/sec, the scrubber operated more efficiently
and mist eliminator performance was improved. Continuous mist elimi-
nator performance in the lime system was maintained at 12.6 and 16 ft/
sec by washing the bottom of the mist eliminator intermittently with
an allocated amount of allowable makeup water. The remainder of the
allocated makeup water was used to intermittently wash the top of the
mist eliminator.
1X1
-------
-------
CONTENTS
Abstract ill
Figures vi
Tables ...... vii
1. Introduction 1
2, Summary and Conclusions 3
3. Test Results 6
k. Discussion of Results 25
5. Expenditures 39
References U-0
Conversion Factors Ul
Appendix ^-3
-------
FIGUBES
Figure Page
1 TCA limestone scrubbing flow diagram 6
2 Top and bottom views of mist eliminator after 202
hours of operation (test ME-10) 8
3 Top and bottom views of mist eliminator after 120
hours of operation (test ME-11) 9
h Side and bottom views of mist eliminator after 96
hours of operation (test ME-12) 10
5 Side and bottom views of mist eliminator after Bit-
hours of operation (test ME-13) 11
6 Top and bottom views of mist eliminator after 213
hours of operation (test ME-lU) 12
7 Top and bottom views of mist eliminator after 500 <
hours of operation (test ME-15) 13
8 Side, top, and bottom views of mist eliminator
after 100 hours of operation (test ME-15) 1^
9 Side view of mist eliminator after 13^ hours of
operation (test ME-17) 15
10 Side, top, and bottom views of mist eliminator
after 560 hours of operation (test ME-18) 17
11 Modified TCA scrubber configuration for the high
velocity tests 18
12 Side, top, and bottom views of mist eliminator
after 500 hours of operation (test ME-19) 20
13 Top and bottom views of mist eliminator after 300
hours of operation (test ME-20) 21
lU Side, top, and bottom views of mist eliminator
after 162 hours of operation (test ME-21) 23
15 Perspective view of Colbert pilot plant 26
16 Chevron mist eliminator 26
17 Mist eliminator test module 27
18 Comparison of planned and actual expenditures 39
VI
-------
TABLES
Performance of the Three Pass, 90 Bend Chevron Mist
Eliminator During Air/Water Tests 29
2 Summary of Mist Eliminator Operations 31
3 Comparison of TVA Lime Scrubbing Data 33
U Summary of SOp Removals and Entrainment Measurements
for Short Term High Velocity Tests 35
5 Comparison of TVA Limestone Scrubbing Data 37
vii
-------
1. INTRODUCTION
TVA has operated the 1-MW wet lime/limestone pilot plant at the Colbert
Steam Plant since February 1971. Until recently, the pilot plant was
used primarily for testing limestone scrubbing with major attention
directed toward the full-scale 550-MW Widows Creek limestone scrubbing
system.
Results from this testing suggested utilizing a vertical mist elimi-
nator in a horizontal duct. This type of mist elimination design
permits the use of a wash system for the mist eliminator separately
from that of the scrubber. More recently, the pilot plant tests were
designed to evaluate similar methods for washing this type of mist
eliminator without exceeding the water balance of the closed-loop
limestone wet scrubbing process. The quantity of fresh water required
for effective washing of the mist eliminator on a once-through basis
was four to five times the amount required for makeup to the closed-
loop slurry system. One such method of maintaining continuous opera-
tion of the mist eliminator in the horizontal duct while operating in
a closed-loop slurry system is the use of sodium carbonate in the
recyclable mist eliminator wash system. This additive increases the
solubility of sulfates thus reducing the tendency for scale formation.
The additive method may not be applicable for washing the mist elimi-
nator in the vertical duct because of the difficulty in separating the
wash liquor from the scrubber liquor. Such separation is necessary to
avoid loss of sodium carbonate which would be prohibitively expensive.
The current project was primarily aimed at finding a washing technique
for maintaining continuous reliable mist eliminator performance in a
closed loop lime/limestone scrubbing system where the mist eliminator
was positioned vertically in a horizontal duct. At a gas velocity of 9
-------
ft/sec, the problem of plugging has generally been resolved. At gas
velocities greater than 9 ft/sec, mist eliminator plugging was still
a problem. Scrubbing techniques had been established for a gas
velocity of 12.6 ft/sec, and this velocity was the preferred operating
velocity. At 16 ft/sec, testing of an advanced high velocity scrubber
was done in conjunction with the mist eliminator study.
-------
2. SUMMARY AND CONCLUSIONS
Pilot-plant tests have demonstrated washing methods that maintain
continuous mist eliminator performance. Continuous performance of a
horizontally mounted mist eliminator operating in the limestone mode
is difficult "because the wash water required for proper washing cannot
be separated from the slurry system. Closed-loop operation precludes
using copious amounts of fresh water to maintain reliable mist elimi-
nator performance since the allowable makeup water rate is approxi-
P
mately 0.7 gpm or 0.2 gpm/ft of duct area. The vendors recommended a
P
wash rate of 5 gpm/ft . An intermittent wash using additional sources
of wash liquor had to be used.
Washing the mist eliminator intermittently in the limestone mode with
fresh makeup water was not successful. This may partly be attributed
to operating the limestone mode at a stoichiometry of 1.5. Operation
of this stoichiometry increases the plugging potential with soft mud-
like solids due to excess limestone in the scrubbing slurry being
entrained into the mist eliminator. The accumulation was at such a
rate that another liquid source was needed to wash the mist eliminator.
The clarified liquor to the scrubber system was accumulated and used
to wash the mist eliminator along with the allowable makeup water.
Intermittent washing with clarified liquor immediately followed by the
makeup water allocation was successful in removing the soft mud-like
solids and the calcium-sulfur salts from the bottom of the mist elimi-
nator—thus preventing the formation of scale. The top was washed
intermittently with the remaining allocation of allowable makeup water.
Continuous mist eliminator performance was maintained with this method.
Washing the mist eliminator intermittently in the lime mode with fresh
makeup water was successful. The lime system operated at a stoichio-
-------
metry of 1.0 which alleviated having excess alkali material present in
the scrubbing slurry being entrained into the mist eliminator. Contin-
uous mist eliminator performance in the lime mode was maintained by
washing the bottom of the mist eliminator intermittently with an allo-
cated amount of allowable makeup water. The remainder of the allocated
makeup water was used to wash the top of the mist eliminator
intermittently.
The pilot plant was modified for the advanced high velocity scrubbing
tests by inserting additional slurry inlets within the beds of spheres.
S02 removals in the lime mode were higher at the high gas velocity
(16 ft/sec) than those observed in the previous tests at 12.6 ft/sec.
This high removal is believed to result from improved mixing and a
higher dissolution of the lime. The mist eliminator was washed with
fresh water only- using the washing technique of the successful lime
run. This washing technique kept the mist eliminator clean for the
500-hour long-term run at both 12.6 ft/sec and 16 ft/sec. '
In the limestone mode, S0p removal was also higher at the higher gas
velocity. The effect of the sprays in the beds was shown to improve
SO removal. However, the mist eliminator, washed intermittently with
only fresh water, showed signs of solids buildup. The need of more
wash water is evident.
The conclusions drawn from this study are as follows:
• Proper application of the available wash water and the wash
frequency is the key contributor in maintaining continuous
mist eliminator performance.
• Limestone scrubbing requires additional washing (clarified
liquor) than the fresh makeup water available.
• A higher limestone stoichiometry (1.5 vs. 1.25) requires
additional mist eliminator attention.
• Lime scrubbing requires only fresh makeup water for mist
eliminator washing.
-------
• The results do not say that all mist eliminator problems are
solved, but the progress in that direction is certainly
encouraging.
• High velocity scrubbing appears to improve scrubber operation.
Additional testing is needed to investigate velocities higher
than those attainable at Colbert.
-------
3. TEST RESULTS
A flow diagram of the TCA limestone scrubbing system for the mist
eliminator tests is shown in Figure 1. The pilot plant operated on
a closed loop basis such that the only liquor purged from the system
was that contained in the discarded spent solids (filter cake). This
quantity of liquor amounted to about 1^0 pounds per hour when the
filter cake contained 60 to 65 percent solids. Fredonia limestone
(75$ - 200 mesh) was fed to the system at a rate sufficient to main-
tain a Ca:S02 mole ratio of 1.5 based on the concentration of S0p in
the inlet flue gas. Scrubbing slurry containing an average of 15
percent suspended solids and 1.25 percent dissolved solids was recir-
culated to the venturi and absorber at a liquid-to-gas rat;Lo (L/G) of
10 and 50 gallons per 1000 cubic feet, respectively. The scrubber
contained two stages (each 12 in deep) of 10-gram thermoplastic rubber
spheres manufactured by Moldcraft. The absorber operated at a
REHEAT SYSTEM
MAKE-UP VOTER
Z-STAGE TCA ABSORBER
BLOWER
• MIST ELIMINATOR
- MAKE-UP WATER
/'
DISTRIBUTOR
fl FLUE GAS i
(FROM UPSTREAM OF UMT 4ESP)
LIMESTONE LMESTONE
SLURRY SLURRY
PREPARATION FEEDTANK
TANK p. 2
F-l
~£3!S? '
rs*
TANK
F-4
SETTLING P-8
TANK
Figure I. TCA limestone scrubbing flow diagram.
6
-------
superficial gas velocity of 12.6 to 13.8 ft/sec. The concentration of
SOp in the inlet gas varied from 1600 to 2700. The SOp removal for the
limestone tests averaged 72 percent. This low S0? removal is attributed
to the absorber inlet spray header being lowered (approximately 15 ft)
to just beneath the third grid so that the mist eliminator could be
installed between the third and fifth grids. The particulate loading
in the inlet and outlet averaged lj-,5 and 0.02 grains per standard (60 F)
cubic foot, respectively. The pressure drop across the venturi and TCA-
type absorber (containing 3 grids and 2 stages of the TPR spheres)
averaged 9 a-0-^ 7 inches of water, respectively.
The initial run was designed to observe drainage and mist entrainment
from the mist eliminator washes at superficial gas velocities of 5>
7.5> 10, 12.6, and 16 ft/sec using only air and water. Proper drainage
of the mist eliminator occurred at 12.6 ft/sec—the planned operating
velocity. The approximate loadings of entrained water entering and
leaving the mist eliminator were determined by exposure of a specially
treated filter paper. The filter paper, attached to a holder, was
moved across the duct. To prepare the test paper, Whatman No. 1 filter
paper was saturated with a 1 percent solution of potassium ferricyanide,
thoroughly dried and dusted with finely divided ferrous ammonium sulfate.
The dry, treated paper was pale yellow. A sharply defined, dark blue
stain appeared when the paper was wetted by a water droplet. This gave
an indication of water entrainment quantity. The approximate loading
was measured by weighing the paper before and after exposure. The
approximate loading of entrained water entering and leaving the mist
eliminator with the wash sprays off was 5.3 and 0.32 grains per standard
cubic foot, respectively.
Run ME-10 began on July 29, 1975, using flue gas and limestone slurry
with an initial pressure drop across the mist eliminator of 0.2 inch
ILO. The mist eliminator was washed intermittently with only fresh
makeup water—0.7 gal/min or 336 gal/shift. Two-thirds of this water
was used to wash the bottom section with only fresh makeup water.
The bottom was washed with 11.7 gpm at a pressure of 20 psi for 2.^
-------
Figure 2. Top and bottom views of mist eliminator
after 202 hours of operation (test ME-10).
minutes every hour. The top was washed with 8 gpm at a pressure of
10 psi for 3.5 minutes every two hours. The pressure drop gradually
increased and leveled off several times before the run was terminated
at a pressure drop of 1.5 inches HpO. The run lasted approximately
200 hours. During this run, the top two passes remained clean indi-
cating a sufficient, if not excessive, top wash. The bottom pass was
plugged with soft, mud-like deposits mainly on the lower lip of the
mist eliminator. Figure 2 shows the accumulation of this material on
the mist eliminator. It was decided that a pressure drop of 1.5
inches H?0 was excessive as a terminating point—no stopping point had
been determined up to this time. A more realistic termination point
of 0.5-inch HpO was selected for the next run.
Run ME-11 began on August 6, 1975, with an initial pressure drop of
0.1-inch HpO. Since the top passes remained clean during the previous
run, half of the makeup water used to wash these passes was added to
the bottom wash whose frequency was also increased. The bottom wash
became 11.7 gpm at a pressure of 20 psi for 1.5 minutes every half hour,
The top wash became 8.2 gpm at a pressure of 10 psi for 1.6 minutes
every two hours. The pressure drop rose slowly until reaching the
8
-------
Figure 3. Top and bottom views of mist eliminator
after 120 hours of operation (test ME-11).
termination point of 0.5 inch HpO after about 120 hours. Figure 3
shows the appearance of the mist eliminator at the end of this run.
Its appearance was similar to that in the previous run. The lower
lip was partially plugged with soft solids, but there were indications
of scale deposits in this mud. The scale formation was probably due
to the remaining SOp in the flue gas reacting with soft solids already
attached to the mist eliminator thus forming the hard calcium sulfate
salts. From experience at Colbert and Shawnee and from calculations
on pressure drop versus percent pluggage of the mist eliminator, a
pressure drop increase of 0.1-inch HpO when starting at 0.1-inch HpO
gives a 30 percent pluggage. Therefore, a new termination point of
0.2-inch HpO was chosen for all subsequent runs.
Run ME-12 began on August 10, 1975» with an initial pressure drop of
0.1-inch HpO. Since the top two passes remained clean during the
previous run and reducing more water from this wash to add to the
bottom wash would be an almost insignificant addition, some of the
available clarified liquor was used to supplement the bottom wash.
Previous mist eliminator tests, prior to this project, had indicated
-------
Figure 4. Side and bottom views of mist eliminator
after 96 hours of operation (test ME-12).
that washing with a mixture of clarified liquor—saturated with CaSCV-
and fresh water would result in scale formation on the blades. With
this experience, it was decided to use clarified liquor as the "bottom
wash and to immediately follow this wash with the fresh makeup water.
Such operation reduces the possibility of any clarified liquor
remaining on the blades for any length of time and becoming super-
saturated with CaSO) and cause scaling to occur. The bottom wash
•"T
was 11.7 gpm. of clarified liquor at a pressure of 20 psi for 1.8
minutes every half hour followed immediately with 11.7 gpm of fresh
water at a pressure of 20 psi for 1.53 minutes. The pressure drop
during this run remained fairly constant for approximately 100 hours
when the center of the lower lip began plugging thus causing termina-
tion of the run. Figure k shows the plugged area that caused the
increase in pressure drop. The small area that had accumulated the
soft deposits was believed to be caused by an uneven distribution of
the wash and additional nozzles were recommended for subsequent runs.
Run ME-13 began on August 19, 1975, with an initial pressure drop of
0.1-inch HLO. Two nozzles replaced the single bottom wash. The
10
-------
i
Figure 5. Side and bottom views of mist eliminator
after 84 hours of operation (test ME-13).
bottom wash became 11.h gpm of clarified liquor at a pressure of 20
psi for 1.8 minutes every half hour followed immediately with 11.k gpm
of fresh water at a pressure of 20 psi for 1.53 minutes. The top wash
remained the same as the previous run. The pressure drop remained
constant for approximately 100 hours when it suddenly rose to 0. It-inch
HUO. The reason for this increase is not certain. Figure 5 shows the
appearance of the mist eliminator at the end of this run.
Run ME-1^ began on August 23, 1975 > with an initial pressure drop of
0.1-inch HpO. It was decided to increase the clarified liquor wash
to the maximum available—approximately 100 gal/shift or three times
that used previously. The bottom wash thus became ll.lt- gpm of clari-
fied liquor at a pressure of 20 psi for 5.U minutes every half hour
with the fresh water wash of 11.k gpm for 1.55 minutes at a pressure of
20 psi immediately following. There was some concern about main-
taining adequate solids concentration with the addition of this
clarified liquor. However, the concentrations were held fairly close
to design conditions. After approximately 200 hours of operation,
the run was terminated after reaching a pressure drop of 0.2-inch HO.
11
-------
m
Figure 6. Top and bottom views of mist eliminator
after 213 hours of operation (test ME-14).
This termination may have teen premature because of the variability of
the pressure drop readings-. Figure 6 shows the accumulation of mud on
the mist eliminator at the end of this run.
Run ME-15 began on September 1, 1975, with an initial pressure drop of
0.1-inch HgO. The only difference between this run and the previous
run is the frequency of the bottom wash was changed from every 30
minutes to every 15 minutes. The washing duration was therefore
adjusted to maintain the same total wash on the bottom. The pressure
drop of 0.1-inch HpO was maintained for 1000 hours. The mist elimi-
nator was momentarily removed after 500 hours for photographing.
Figure 7 shows the appearance of the mist eliminator at that time.
Figure 8 shows the appearance of the mist eliminator after completion
of the long-term run.
Run ME-16 began on October I1*, 1975, to test the effect of washing
the mist eliminator with absorber slurry and fresh makeup water. An
initial pressure drop across the mist eliminator was 0.1-inch HpO.
The bottom of the mist eliminator was washed at 15-minute intervals
12
-------
Figure 7. Top and bottom views of mist eliminator
after 500 hours of operation (test ME-15).
with absorber slurry at a rate of 6 gpm for 2.7 minutes at a pressure
of 15 psi with a fresh water wash of 11.k gpm for 0.78 minutes at a
pressure of 20 psi immediately following. The top wash remained
unchanged. A buildup of soft mud-like solids occurred in the second
pass of the mist eliminator. More than likely, this buildup was
caused by an accumulation of slurry solids that were carried into the
mist eliminator during the slurry wash sequence. The same header was
used for the slurry wash as the makeup water wash. High-pressure air
was used frequently to dislodge slurry solids from the two spray
nozzles. This resulted in a portion of the solids which originally
accumulated in the second pass to be blown into the third pass where
they were removed by the top wash. The net effect of using high-
pressure air to aid in cleaning the mist eliminator is uncertain.
After 166 hours of operation, the pressure drop reached 0.2-inch ELO
and the run was terminated.
Run ME-17 began on October 21, 1975, using the same washing techniques
as ME-16 with the exception that a separate spray header with larger
nozzles was used for the slurry wash. These nozzles plugged with
13
-------
Figure 8. Side, top, and bottom views of mist eliminator
after 1000 hours of operation (test ME-15).
1U
-------
Figure 9. Side view of mist eliminator
after 134 hours of operation (test ME-17).
slurry solids and high-pressure air was unsuccessful to unplug these
nozzles because a sufficient back pressure could not be maintained.
The solids accumulated in the second pass of the mist eliminator at a
faster rate than the previous run. This may be due to the elimination
of high-pressure air hitting the mist eliminator. The run was termi-
nated after 13^ hours of operation when the pressure drop reached 0.2-
inch HpO. Figure 9 shows the spray headers and the mud deposits in
the second pass at the end of the run.
The absorbent in the scrubber was changed from limestone to lime. The
flow diagram for the lime tests is essentially the same as the lime-
stone tests (Figure 1). The system operated on a closed-loop basis
such that the only liquor purged from the system was that contained in
the discarded filter cake. This quantity of liquor amounted to about
180 pounds per hour when the filter cake contained 55 to 60 percent
solids. Calcium hydroxide, manufactured by the Longview Lime Company,
was used as the calcium source. Fresh calcium hydroxide slurry (30
percent suspended solids) was fed to the absorber circulation mix tank
to maintain the pH of the slurry being fed to the absorber between 8.0
15
-------
and 8.5. The amount of fresh calcium hydroxide required was equivalent
to a Ca:S02 mole ratio of about 1.0. The scrubbing slurry containing
10 percent suspended solids and 1 percent dissolved solids was recir-
culated to the venturi and absorber at an L/G of 10 and 60 gallons per
1000 cubic feet respectively. The scrubber contained two stages (each
12 in deep) of mobile packing spheres. The Moldcraft 10-gram TPR
spheres were replaced by 6-gram nitrile foam spheres manufactured by
Universal Oil Products. A small mix tank was used to bring the
scrubber effluent slurry into contact with the fresh calcium hydroxide
slurry before it overflowed into the absorber circulation tank. The
retention time of the slurry in the mix tank was about 10 seconds
based on the scrubber effluent slurry rate of iMj- gallons per minute.
The use of the mix tank in the configuration was effective in obtaining
93 percent utilization of the calcium hydroxide. The absorber operated
at a superficial gas velocity of 12.6 to 15 ft/sec. The concentration
of SO^ in the inlet flue gas varied from about 2*4-00 to 2600 gpm. The
*- i
SOp removal averaged 85 percent. This low removal is attributed to
the height of the tower as discussed previously. The inlet and outlet
dust loadings remained constant as in the limestone tests.
The chemistry of the absorption of SOp is different in the lime mode
than in the limestone mode. This difference 'results in higher utili-
zation of the absorbent and, therefore, a reduction in the amount of
water removed with the sludge. The makeup water requirement for the
lime mode is approximately 0.6 gal/min/MW as compared to 0.7 gal/min/MW
for limestoneo This reduces the quantity of water available to wash
the mist eliminator.
Run ME-18 began on November 29? 1975. As a base case, the mist elimi-
nator was washed intermittently with fresh makeup water only. The top
was washed every 2 hours for 1.6 minutes at a rate of 11.h gpm and a
pressure of 10 psi. The bottom was washed every 15 minutes for 0.65
minutes at a rate of 11.k gpm and a pressure of 20 psi. The pressure
drop across the mist eliminator remained 0.1 in E^O through 560 hours
of operation. At that time, the pilot plant was shut down for the
16
-------
Figure 10. Side, top, and bottom views of mist eliminator
after 560 hours of operation (test ME-18).
17
-------
Reheat System
Soot Blower
.(From Up""""11 °f Unit4 ESP>
Mist Eliminator
Wash System
(Make-up Water)
-""••7 jiui«j =~~ Absorber —— ——
Preparation Feed Tank f~* Retention Venturl M P-6
Tank F-2 7a* Retenlioi
F-l F"3 Tank _
F-4 Settling Tank JT,
Vacuum Filter D^?.'?'d
Cake
Figure 11. Modified TCA scrubber configuration for the high velocity tests.
Christmas holidays. Figure 10 shows the appearance of the mist elimi-
nator at the end of the lime run. This run completed the mist eliminator
tests in the vertical duct. Trends in operating data for all runs are
graphically displayed in the Appendix.
To conduct the high velocity scrubber and mist eliminator portion of
the project, the scrubber was modified to improve the gas/liquid
mixing characteristics. This was done by installing a slurry inlet
10 inches below the stages that sprayed upward into each stage. The
existing slurry inlet and the two additional inlets were each fitted
with a low pressure nozzle (13 psi at 6k gpm). The slurry would be
equally divided among the nozzles for a total L/G of 60 (L/G of 20/
nozzle or 6k gpm/nozzle). The absorber contained two stages (each 9
inches deep) of 6-gram nitrile foam spheres manufactured by Universal
Oil Products.
A series of air/water tests were performed to observe the drainage of
the mist eliminator and the action of the TCA spheres. At 16 ft/sec,
the planned operating velocity, the spheres remained fluidized, and
18
-------
the drainage of water within the mist eliminator was satisfactory.
Run ME-19 "began on June 16, 1976, using lime slurry and flue gas at
a velocity of 16 ft/sec. The closed loop flow diagram is shown in
Figure 11. The only liquor purged from the system was that contained
in the filter cake. This quantity of liquor amounted to about 330
pounds per hour when the filter cake contained 55 percent solids.
Fresh calcium hydroxide slurry (30 percent suspended solids) was fed
to the absorber circulation mix tank to maintain the pH of the slurry
being fed to the absorber between 8.0 and 8.5. The amount of fresh
calcium hydroxide required was equivalent to a Ca:SO? mole ratio of
about 1.0. The scrubbing slurry containing 10 percent suspended solids
and 2 percent dissolved solids was recirculated to the venturi and
absorber at a L/G of 10 and 60 respectively. The concentration of SOp
in the inlet flue gas varied from about 1900 to 2700 ppm. The SOp
removal averaged 91 percent and particulate removal 99«^ percent.
Calcium hydroxide utilization was 90 percent. The pressure drop across
the venturi and absorber was 6 and 5 in HpO respectively. The mist
eliminator was washed intermittently, as in ME-18, with the extra water
(available from the increase in gas throughput) added to the absorber
circulation tank to control the solids concentration. This washing
technique permitted continuous reliable mist eliminator performance
for the 500-hour long-term test. The pressure drop across the mist
eliminator varied from 0.1 to 0.2 in HJ3 during this test. Figure 12
shows the appearance of the mist eliminator at the end of this run.
Run ME-20 began on July 9, 1976, with limestone as the alkali source
and a gas velocity of 16 ft/sec. The flow diagram for this limestone
test is similar to that for the proceeding test (Figure 11). The only
liquor purged from this closed-loop system was that contained in the
filter cake (365 Ibs HpO/hr @ 5^ percent solids). Fredonia limestone
(75$ - 200 mesh) was fed to the system to maintain a Ca:SOp mole ratio
of 1.3. The scrubbing slurry contained 10 percent suspended solids
and l.U percent dissolved solids. The internals and slurry distribu-
tion within the absorber remained the same as in the proceeding test.
19
-------
Figure 12. Side, top, and bottom views of mist eliminator
after 500 hours of operation (test ME-19).
-------
Figure 13. Top and bottom views of mist eliminator
after 300 hours of operation (test ME-20).
The concentration of SOp in the inlet flue gas varied from 2000 to 2700
ppm. The S0? removal averaged 8l percent and the particulate removal
averaged 99-3 percent. The mist eliminator was washed intermittently as
in ME-18 and ME-19 with the extra water (available from the higher gas
throughput) being used to control the solids concentration in the absorber
circulation tank. During the 300 operating hours in this test, the
pressure drop across the mist eliminator varied between 0.1 and 0.2 in R~0,
At the end of this test, the first pass of the mist eliminator had a light
accumulation of soft mud-like solids. Figure 13 shows the appearance of
the mist eliminator at this time.
On August 27, 1976, run ME-21 began. In this test, the effect of all of
the slurry being introduced from the top spray (fitted with two nozzles)
was studied. A stoichiometry of 1.3 was maintained with 12 percent
suspended solids and 0.9 percent dissolved solids in the scrubbing system.
The concentration of SOp in the inlet flue gas varied between 2^00 and
2800 ppm. The SOp removal averaged 76 percent and particulate removal
averaged 99.6 percent. The mist eliminator was washed intermittently as
runs ME-19 and ME-20. The pressure drop across the mist eliminator
21
-------
averaged 0.2 in H?0 for 162 hours of operation. At that time the first
pass of the mist eliminator had a light accumulation of soft mud-like
deposits. Side, top, and bottom views of the mist eliminator at the end
of the run are shown in Figure Ik.
22
-------
f'ff'fiiiiiiJiiiiilllllllllUl
w
Figure 14. Side, top, and bottom views of mist eliminator
after 162 hours of operation (test ME-21).
23
-------
-------
DISCUSSION OF RESULTS
Previous limestone tests at the Colbert pilot plant have indicated
that the mist eliminator requires continuous washing with fresh water
n
at a rate of 1 gpm/ft face area as compared to the vendor recommended
p
wash rate of 5 gpm/ft . This amounts to about 3.^ gpm of fresh water.
To operate the scrubbing system closed loop, the allowable makeup water
amounts to 0.7 gpm, less than one-fifth the amount specified by vendors
and prior tests. In a previous TVA study^ ' on entrainment separators,
the mist eliminator was placed vertically in a horizontal duct and the
wash water was recycled to maintain closed-loop operation. This method
resulted in hard gypsum scale formation on the mist eliminator blades.
The addition of sodium carbonate to this wash system allowed the
sulfate formed to remain in solution thus preventing scale formation
on the blades. The separation of this wash water from the scrubbing
liquor is easy in the configuration used. A mist eliminator placed
horizontally in a vertical duct has been the standard practice, however,
a lot of problems have plagued its reliability. Therefore, TVA with
funding by EPA, undertook the task of finding a washing technique to
maintain continuous reliable mist eliminator performance at the 1-MW
level.
The mist eliminator in the pilot plant was repositioned between the
third and fifth grids of the TCA absorber (highest point possible
before leaving the tower). Observation windows were also installed
to visually detect the effectiveness of the washing patterns. A
perspective view of the Colbert pilot plant showing the major pieces
of equipment is shown in Figure 15.
The mist eliminator tested during the project was the three-pass, 90-
degree bend, Chevron-type with 1^-inch spacings as shown in Figure 16.
25
-------
Figure 15. Perspective view of Colbert pilot plant
Figure 16. Chevron mist eliminator.
26
-------
Figure 17. Mist eliminator test module.
Other mist eliminator types—Chevron with larger spacings, open vane,
and Koch tray—were planned to be tested if the Chevron type could not
be kept unplugged. The Chevron-type mist eliminator was placed in the
horizontal position in the vertical absorber tower. Figure 17 shows a
photograph of the test module and its dimensions for viewing the
operation of the mist eliminator.
Air water tests were performed to observe the drainage of the 90 bend,
3-pass, Chevron-type mist eliminator when washing from the top and/or
bottom. Within the passes of the mist eliminator, the entrained water,
bottom wash water, and water from the top wash spray drained satisfac-
torily at gas velocities up to 16 ft/sec (larger spacings of Chevron
type were not as effective in mist removal).
27
-------
During the top wash sequence, water entrainment from the mist elimi-
nator amounted to over 25 percent by weight of the gas entering the
reheater. This water content is unacceptable for reheat requirements
if continuous washing from the top is used. However, intermittent
washing was planned, and the average water entrainment, 0.2 to 0.3
percent by weight, is acceptable for efficient reheat performance.
The water grain loading leaving the Chevron mist eliminator is shown
in Table 1.
Previous testing of limestone scrubbing at Colbert has shown that
continuous washing with fresh makeup water only was insufficient to
keep the mist eliminator clean. Blending of the fresh makeup water
and clarified liquor was found to be disasterous because additional
S02 was absorbed by this liquor and hard gypsum scale formed on the
mist eliminator blades. Tests at TVA's 10-MW Shawnee test facility
did not observe scale formation or excessive solids deposition when
using a clarified liquor/makeup water wash, ' however, tnis TCA
scrubber had a gas velocity of 8.6 ft/sec with a much lower gas velo-
city through the mist eliminator whereas the Colbert pilot plant had
a gas velocity through the mist eliminator in excess of 12.6 ft/sec.
Testing by organizations outside of TVA found an improvement to clean-
ing the mist eliminator with high pressure sprays and copious amounts
of water. Based on the above information, a test program was set up
to wash the mist eliminator intermittently from the top and bottom
with fresh water—as a base case. Over $0 percent of this water was
used for the hourly bottom wash and the remainder for the top wash
every two hours. The pressure drop across the spray nozzles was 20
and 10 psi respectively. As expected, the bottom pass of the mist
eliminator plugged with soft mud-like solids rapidly while the top
remained clean. Cutting the top wash in half and adding this water to
the bottom and increasing the washing frequency did not help. Since
more water was needed for the bottom wash, a portion of clarified
liquor was used. The clarified liquor wash was followed immediately by
the fresh water wash. In this manner, the fresh water washed the clari-
fied liquor off the blades before scale formation could occur. Increasing
28
-------
TABLE 1. PERFORMANCE OP THE THREE PASS, 90 BEND CHEVRON
MIST ELIMINATOR DURING AIR/WATER TESTS
Superficial
Velocity
ft/sec
5
7.5
10
12.6
16
5
7.5
10
12.6
16
5
7.5
10
12.6
16
5
7.5
10
12.6
16
5
7.5
10
12.6
16
Washing Rates
gal/min
Bottom Top
a
a
a
a
a
3.3C
3.3
3.3
3.3
3.3
8d
8
8
8
8
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
3.3C
3.3
3.3
3.3
3.3
Exit Gas Loading
Grains Water Per
Cubic Foot
.007
.008
.008
.001
.001
.007
.008
.008
.003
.02
.007
.008
.008
.006
.Ok
b
b
b
b
b
b
b
b
b
b
a Mist eliminator not washed.
b Entrainment too heavy to measure.
c Bete TF11* FC nozzles with pressure drop of 8 and 10 psi at 3.3 and
k gal/min respectively.
d Bete TF20 FC nozzle with pressure drop of 10 psi at 8 gal/min.
Note: A constant L/G of 50 was maintained in the scrubber for each
gas velocity.
-------
the clarified liquor wash, to the maximum available and increasing the
frequency to every fifteen minutes for the bottom wash, keeping the top
wash the same, was found to be very effective. A 1000 hour long-term
run was accomplished while maintaining a constant pressure drop across
the mist eliminator element. At this time, the mist eliminator had
only minor deposits of mud-like deposits on areas missed by the wash
sprays. A summary of the test conditions and other information leading
to this test is shown in Table 2. More detail of these tests is
provided in the previous section.
Tests were also conducted to observe the effect of washing the bottom
of the mist eliminator with limestone slurry followed by fresh water
wash. Soft mud-like solids accumulated in the second pass of the mist
eliminator because the pressure of the sprays forced the limestone
slurry into this pass and the fresh water wash, from a separate header,
could not effectively reach this area. During these tests,rhigh
pressure air was used to clean the slurry wash spray nozzles. This air
apparently dislodged part of the solids accumulated in the second pass
and these could therefore be washed away by the top wash. The use of
high pressure air may be an aid in cleaning the mist eliminator, but its
use needs to be explored in more detail than the p-reXitninary observations
made here.
When the alkali source was changed from limestone to lime, the washing
technique of intermittent fresh makeup water was again used as a base
case. This washing technique was sufficient to maintain a constant
pressure drop across the mist eliminator element for a 560-hour long-
term run. Visual inspection of the mist eliminator during this run
revealed no accumulation of solids occurring. The lower wash water
requirement is believed to result from the lime mode operating at unit
stoichiometry rather than a 1.5 stoichiometry in the limestone mode.
This lower stoichiometry apparently reduces the amount of unreacted
alkali being entrained into the mist eliminator where it can accumulate
and/or react with residual S02 to form hard scale.
30
-------
TABLE 2. SWWARK OF MIST EUMIHATOH OPERATIONS
Top Wash
Bottom Wash
U)
H
Duration A? Pressure Rate Duration
Run of Teat (in. H,0) (psi) (gem) (mln)
ME-9 12 b
MB-10 202 h 0.2—1.5 10 8 3.5
ME-11 120 h 0.1-»0.5 10 8.2 1.6
MB-12 96 h 0.1—0.2 10 8.2 1.6
ME-13 8U h 0.1-»0.l» 10 8.2 1.6
ME-lU 213 h 0.1—0.2 10 8.2 1.6
ME-15 1000 h 0.1— »0.1 10 8.2 1.6
ME-16 166 h 0.1—0.2 10 8.2 1.6
ME-17 13U h 0.1—0.2 10 8.2 1.6
Interval Type Pressure
(h) Waah (pai)
2 FW 20
2 FW 20
2 FW 20
20
2 FW 20
20
2 FW 20
20
2 FW 20
20
2 FH 15
20
2 FW 15
20
Rate
(BE») .
11.7
11.7
11.7
11.7
ll.U
ll.lt
ll.lt
ll.lt
ll.lt
n.U
6.0
ii.it
6.0
U.It
Duration
(min)
2.U
1.5
1.8
1.53
1.8
1.53
5.1*
1.55
2.7
0.78
2.7
0.78
2.7
0.78
Interval
(h)
1
0.5
-i
0.5
0.5
0.5
0.5
0.5
0.5
0.25
0.25
0.25
0.25
0.25
0.25
Type
Wash
FW
FW
CL*
FW
CL*
FW
CL»
FW
CL«
FW
SL*
FW
SL*
FW
Total ME Wash
(gal/shin)
337
333
337
339
328
332
985
335
985
338
518
338
518
338
Cements
Air/water teat.
Top two paseea clean.
Lower lip plugged with mud-like deposits
Top two pases clean. Lover lip
plugged with mud-like deposits
with indications of scale in mud.
Top two paaaes clean. Bottom center
plugged probably due to spray pattern.
Used 2 nonles at 5.7 gpm. Suddenly
started plugging. Upset unknown.
Similar to ME-13. May begin having
problems maintaining solids
concentration. May have been
stopped premature — one small area
causing pluggage— rest clean with
little mud on bottom edges
(always there).
Nozzles were raised to cover a missed
area 9-9-75. ME momentarily removed
after 500 hours for photographing.
Units 3,>t, and 5 off-line— pilot
plant shut down 10-6 to 10-8. ME
remained clean through 1,000 hours.
Second pass plugging. Nozzles plugging
and cleaned by blast of air (30 psi) —
may nave helped clean HE.
Nozzles eroded.
Separate and rotating slurry header.
Plugging in second pass.
ME-18 560 h 0.1 — 0.1 10 8.2 1.6
ME-19 500 h 0.2 -»0.2 10 8.2 1.6
ME-20 300 h 0.2 -.0.2 10 8.2 1.6
ME-21 162 h 0.2 -. 0.2 10 8.2 1.6
FW: Fresh makeup water 0.7 gpm: 336 gal/shift
CL: Clarified liquor
SL: Slurry
•Followed Immediately with fresh water wash
FW
FW
20
20
ll.lt
ll.lt
0.65
0.65
0.25
0.25
20 ll.lt 0.65 0.25
Hf 20 n.lt 0.65 0.25
FW 289 Lime mode: 0.6 gpm makeup—288 gal/shift
ME remained clean while using only
fresh makeup water
FW 289 Lime mode at higher gas velocity—16 ft/sec.
Extra fresh water available going to
absorber circulation tank. ME remained
clean.
FW 289 Limestone mode at 16 ft/sec. Light
accumulation of soft solids on first
pass.
IV 289 Limestone mode at 16 ft/sec. Light
accumulation of soft solids on first
pass.
-------
The above mist eliminator tests have indicated that proper implementa-
tion of the available wash water and wash frequency is the key contri-
butor in maintaining continuous mist eliminator performance. These
results do not say that all mist eliminator problems are solved, but
the progress in that direction is certainly encouraging.
The last phase of the project was designed to test the mist eliminator
in an advanced high velocity scrubber. The TCA scrubber was modified
with the addition of a slurry spray below each of the two beds of TCA
spheres spraying cocurrent to the gas flow. The scrubber liquor was
equally distributed between these two inlets and the previous slurry
inlet above the top bed (also installed with a low pressure nozzle).
The same mist eliminator and positioning were used.
Air/water tests were done to observe the drainage of water from the
mist eliminator and action of the TCA spheres. At 16 ft/sec, the mist
eliminator drained properly. The TCA spheres remained mobile even at
this high gas velocity.
A comparison of the TCA lime scrubbing data at 12.6 and 16 ft/sec is
shown in Table 3» The SOp removal at the higher gas velocity was
better—around 91 percent versus about 85 percent at the lower gas
velocity. This higher S0p removal is believed to be attributed to
improved mixing characteristics resulting from the higher gas veloc-
ity and the effect of the modification of the slurry sprays. Previous
work at the Shawnee test facility also indicates that increasing the
gas velocity improves the mixing characteristics of the gas and slurry
thus improving S0p removal. The better mixing apparently allows more
dissolution of the lime and, with the basic material being replenished
more rapidly, improvement of SOg removal. The same effect is believed
to have occurred here at Colbert. Modifications made to the scrubber
are also believed to improve the S0? removal efficiency. The new inlets
spray the slurry cocurrently with the gas into the beds of spheres where
the retention time of the slurry is apparently increased by the force of
the gas and sprays. In addition, the inlets were fitted with lower
pressure spray nozzles which produce smaller droplets than the previous
32
-------
TABLE 3. COMPARISON OF TCA LIME SCRUBBING DATA
GAS VELOCITY
(ft/sec) 12.6 16
L/G ,,
(gal/1000 ftj)
Venturi 10 ICL
Absorber 60a 60
STOICHIOMETRY
(Ca:S02 mole ratio) 1.0 1.0
SOLIDS
($by wt.)
Suspended 10 10
Dissolved 1 2
PH 8.5 8.5
PRESSURE DROP
(in
Venturi 9 6
Absorber 7 5
SLURRY NOZZLE PRESSURE
(psi) 5 13
PARTICULATE REMOVAL
99.5 99*
S02 REMOVAL
a all at top
b L/G = 20 below first bed
L/G = 20 below second bed
L/G = 20 top
c open pipe distributor
80 91
33
-------
open pipe distributor. The smaller drops provide more surface area
available for contacting the SOp in the gas.
A series of short tests were conducted to determine the S0p removal
efficiencies of the sprays in the modified configuration. The results
of these tests are shown in Table k. The venturi removes 2k percent
of the incoming 20^0 ppm SOpj hence, approximately 1550 ppm SOp enters
the main absorber vessel. With only the bottom absorber spray operat-
ing, 38 percent of the 1550 ppm S02 was removed leaving 959 ppm SOp.
The second spray removed 1*9 percent of the remaining S0p to leave 490
ppm in the gas. With the third spray in service, 5^ percent of the
^90 ppm was absorbed by this spray. The sprays obtain the liquor from
the same source so the dissolved basic material is virtually the same
in all sprays. Hence, as a portion of the S0? is removed in each stage,
the ratio of basic material in the next stage to the SOp remaining in
the gas increases. Therefore, the effective stoichiometry in the liquid
phase is higher. Another less dominant but important consideration that
may influence the increase"in S02 removal per stage is a higher residence
time of the slurry which is believed to result from one or more of the
following: (1) the bottom two nozzles spray the slurry the same direc-
tion as the gas flow thus the absorber acts like a cocurrent scrubber
for about two-thirds of the scrubbing section, (2) the slurry from the
second spray has an .additional four feet to react with the SOp than the
slurry from the bottom spray, and (3) interaction of the sprays occurs,
e.g., bottom spray keeps the liquor from the second spray in the bed for
an additional length of time. The top spray proved to be beneficial not
only from an S0p removal viewpoint but also as a mechanism that assisted
mist pliTrrina.-t-.or performance. Apparently, the downward spray promoted
the coalescing of the slurry droplets thus reducing the entrainment. In
addition, the force of the top spray helped prevent the TCA spheres from
becoming impinged upon the top retaining grid.
The mist eliminator was kept clean for a 500-hour long-term run with
lime by the same washing technique as the previous successful lime run.
Since the scrubber operated at a higher gas throughput, there would be
-------
Ul
TABLE k. SUMMARY OF S02 REMOVALS AND ENTRAINMENT MEASUREMENTS FOR THE HIGH-VEIiOCITY ABSORBER TESTS
Entrainment
Test
No.
1
2
3
k
5
SOp removal,
Description %
Top, middle, and bottom 89
sprays in service
(each at 6k gal/min)
Middle and "bottom sprays 76
in service (each at
6k gal/min)
Bottom spray in service 53
(at 6k gal/min)
Top, middle, and bottom 2k
sprays out of service;
venturi operating at
32 gal/min
Middle and bottom sprays 86
in service (each at
8U gal/min, max. flow)
Mist eliminator,
In
k.k7
23.77
22.38
^.39
15.17
gr/stdft^
Out
0.053
0.2kO
0.050
Q.tikQ
0.155
NOTE: The absorber was operated at 3200 acfm at 120°F with 2 stages (each 9 in deep) of
spheres.
-------
more makeup water available for washing the mist eliminator. This extra
water was not required for the mist eliminator wash and was used to help
control the solids content of the clarified liquor. The higher gas
throughput increased the liquor flow through the settling tanks thus
decreasing the residence time of the slurry in these tanks—designed
for a 1-MW scrubber. Prior to the start of these tests, it was
believed a problem of solids content control of the clarified liquor
would arise. However, with proper implementation of water, the solids
content was controlled satisfactorily. For larger scale scrubbers of
this type, the size of the thickener or settling areas need to be in
proportion to the operating conditions expected.
A comparison of the limestone scrubbing data at 12.6 and 16 ft/sec is
shown in Table 5- The S02 removal efficiency was higher at 16 ft/sec
than that observed in the previous limestone runs at lower gas
velocities—8l percent versus 73 percent. This higher S0p removal is
attributed to the improvement of the gas/liquid mixing characteristics
as discussed previously. In addition, the L/G ratio for the high
o
velocity tests was increased from 50 to 60 gal/1000 ft gas. A higher
L/G generally increases SOp removal. When all of the scrubbing liquor
was sprayed on top of the bed, the S0? removal efficiency decreased to
76 percent. An explanation of this phenomenon'may be the nozzles
spraying the slurry onto the bottom of the beds provide gas/liquid
mixing characteristics unobtainable from the top slurry spray alone.
During the high velocity limestone tests, the stoichiometry was
reduced from 1.5 to 1.3 to increase limestone utilization and reduce
the amount of unreacted material being carried into the mist eliminator
thereby reducing plugging problems. This reduction in stoichiometry
increased utilization from about k$ to 62 percent. Along with the
higher utilization was an increase in SOp removal efficiency. The
reasons for the increase in SOp removal was discussed previously but
it is interesting that this occurred with a lower amount of limestone
fed to the system. Improvement of the mixing characteristics and the
dissolution of limestone presumably more than compensated for the
reduction in stoichiometry.
36
-------
TABLE 5. COMPARISON OF TCA LIMESTONE SCRUBBING DATA
GAS VELOCITY
(ft/sec) 12.6 16 16
L/G .
(gal/1000 ft0)
Venturi 10 10 1CL
Absorber 50 60a 60°
STOICmOMETRY
(Ca:S02 mole ratio) 1.5 1.25 1.3
SOLIDS
(% by wt.)
Suspended 15 12 10
Dissolved 1.3 0.9 l.U
PRESSURE DROP
(in
Venturi 9 77
Absorber 7 67
SLURRY NOZZLE PRESSURE
(psi) 5 15 13
PARTICULATE REMOVAL
99.^ 99.6 99.3
REMOVAL
73 76 81
a all at top
b L/G = 20 below first bed
L/G = 20 below second bed
L/G = 20 top
c open pipe distributor
37
-------
The lower stoichiometry apparently helped the performance of the mist
eliminator which was washed intermittently with fresh makeup water
only. After 300 hours of operation, the first pass of the mist elimi-
nator had a light accumulation of soft mud-like solids. This buildup
was much lower than that experienced before when only fresh makeup
water was used for washing. Therefore, a reduction in the amount of
available, unreacted material being blown into the mist eliminator
should reduce the probability of pluggage—as indicated by the results
(3)
of this test. This result is similar to tests conducted at Shawnee.x '
The particulate removal for the high velocity tests averaged 99-3
percent as compared to 99 • 5 percent removal for the lower velocity tests.
38
-------
5. EXPENDITURES
A comparison of planned and actual expenditures is graphically dis-
played in Figure 18. The total monies allotted to this project was
$600,000. This amount was necessary to complete the project.
700
60O
500
1 400
~ 300
ZOO
100
PLANNED COST •-
ACTUAL COST • •• -•> •
April May June July Aug Sept Oct Nov Dec Jan Feb
1975 |
April May Jane July Aug Sept Oct Nov Dec Jan Feb Mai
I9T6 | 1977
Figure 18. Comparison of planned and actual expenditures.
39
-------
REFERENCES
1. Schultz, J. J., Kelso, T. M., Glasgow, S. L., Cole, R. M.
Performance of Entrainment Separators in Slurry Scrubbing Processes.
TVA Bulletin Y-93, Tennessee Valley Authority, Muscle Shoals,
Alabama, June 1975, 39 pp.
2. Epstein, M. EPA Alkali Scrubbing Test Facility: Advanced Program--
First Progress Report, EPA-600/2-75-050, U.S. Environmental Protec-
tion Agency, Washington, DC, September 1975, 172 pp.
3. Head, H. N. EPA Alkali Scrubbing Test Facility: Advanced Program-
Second Progress Report, EPA-600/7-76-008, U.S. Environmental Protec-
tion Agency, Washington, DC, September 1976, 377 pp.
-------
CONVERSION FACTORS
The Environmental Protection Agency policy is to express all measure-
ments in metric units. Implementing this practice will result in
undue lack of clarity. The following conversion factors are provided
to convert the nonmetric units to the International System of Units
(SI).
To Convert From
Inches HpO
o
Pound/inch (psi)
Pound (Ib)
Gallon/minute (gpm)
Gallon (gal)
Foot/second (ft/sec)
Foot2 (ft2)
Foot3 (ft3)
Degree Fahrenheit ( F)
To
Millimeter Mercury (mm Hg)
Atmosphere (atm)
Kilogram (kg)
Liter/minute (1/min)
Liter (1)
Meter/second (m/sec)
Meter2 (m2)
Meter3 (m3)
Degree Centigrade (°C)
Multiply By
1.868 x 10°
6.805 x 10"2
U.536 x 10"1
3.785 x 10°
3.785 x 10°
3.0U8 x 10"1
9.290 x 10"2
2.832 x 10"2
t. = (t_ - 32)/1.8
-------
-------
APPENDIX
TRENDS IN OPERATING DATA
-------
-------
Sj o«r
^blk
! •«
IS »
is 70
*§ «»
50
S
1
120
no
•0
•0
r CLEM UOUOR RfTUBN
USORWR SU/BKY
_
1 1 1 1 , . 1 1 . 1 1 . 1 . L 1 1 1 1 1 . • 1 1 1 .
» r s 9 n ii 5 a S is ie rr S S 55
28 29 SO
ME- 15
-------
I 0«|
is03
iz-02
i* 0.1
06
s *•'
_i 1 1 i
60
90
120
110
CO
CLEAR LIQUOR RETURN
ABSORBER SLURRY
i I I 1 L
.a
20
I i I 1 L
-I I I'll
_! 1 i I I
I I I 1 1 L.
I I I I I
J 1 1 I I 1 L
I I I I 1 1 L
1 I I 1
1J
2.0
1.3
£ 1:0
'e '•»,
I I I I I U
20[
T
101—
i I I i I L
I I I I
_l I 1 1 1 1
i I i i i
I I 1 1 1 1
-UNIT 3 OUT OF SERVICE
10,
sol
I I I I I 1
I I I I—I 1
Z 22 23 24 ZS 26
i 2 3
, IOCTII
96 7 8 9 O II 12 13
14 IS IS t7 18 19
ME-16 -
-------
•L
K>l—
IX •
ao •
no -
no -
9O -
80 -
70 •
60-
CLEiR UOUORRETUDN
ABSORBER FED
«-
0 P 4
I ME-I7-*
-------
10r
u-
SMUT
DOM
lonmiR
B-7
-i—i
-1 1 L.
i—i—i—i i
o« -
i 03 .
02 -
I t i
: to
TO
•0
120
110
IOO
90
/kBSOfOER 3UURRY
O_E»R LJOUOR RETUIN
5 ao
§i§.
IP- -
*Ste 9
I I ' I 1 1 U L
J—I—I—I—I—I—I
-I—I—I—I— II
v
_J I 1
I I I 1 1 1 1 1 1 1 1 1
T a t(OIII2l3«l»F«irWI»ZO
(JULY I)
-ME-f»-
25 14
I JULY T)
-------
z
§ o.i
o
7
s
K <
-
g '00
5 90
1?"
8- 70
I .0
« 50
40
30
a .200
.175
ABSORBER SLURRY
CLEAR LIQUOR RETURN
i*.
.150
o .125
Sgo 0.5
0
Z
z
<
I 2 3 4 5 6 7 6 9 10 II 12 13 14 IS l«
(JULY 9) (JULY 21)
7
i i i i i i i
i i I I i i I I I i
i i i i i
i i i i i i i i i i
12345678910
(AUG 27) (SEPT I)
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-800/7-77-019
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
TVA's 1-MW Pilot Plant: Final Report on High-Velocity
Scrubbing and Vertical Duct Mist Elimination
5. REPORT DATE
March 1977
6. PERFORMING ORGANIZATION CODE
7.AUTHOR(S) G.Hollinden, R.Robards, N.Moore (TVA/
Chatt.); and T.Kelso and R.Cole (TVA/M. Shoals)
8. PERFORMING ORGANIZATION REPORT NO.
PRS-19
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TVA, Power Research Staff, Chattanooga, TN 37401
and TVA, Office of Agricultural and Chemical Devel-
opment, Muscle Shoals, AL 35660
10. PROGRAM ELEMENT NO.
EHB528
11. CONTRACT/GRANT NO.
EPA-TAG-D5-0721
TVA-TV-41967A
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 1-9/76
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES T£RL_RTp project officer for this report is J.E. Williams, Mail
Drop 61, 919/549-8411 Ext 2915.
16. ABSTRACT
The report describes the systematic test program that led to the development
of washing techniques that maintain continuous mist eliminator performance for lime/
limestone closed-loop scrubbing systems. TVA recently demonstrated the techniques
at its 1-MW pilot plant at the Colbert Power Plant. The report also describes high-
velocity scrubbing tests performed in conjunction with the mist eliminator tests. Con-
tinuous operation of the chevron mist eliminator, positioned horizontally in a vertical
duct, in the limestone system was maintained (after extensive testing at 12. 6 ft/sec)
by washing the bottom of the mist eliminator intermittently with all the available clar-
ified liquor, immediately followed by an allocated amount o.f makeup water. The top
of the mist eliminator was washed intermittently with the remaining allocation of
allowable makeup water. At a gas velocity of 16 ft/sec, the scrubber operated more
efficiently and mist eliminator performance was improved. Continuous mist elimin-
ator performance in the lime system was maintained at 12. 5 and 16 ft/sec by washing
the bottom of the mist eliminator intermittently with an allocated amount of allowable
makeup water. The remainder of the allocated makeup water was used to intermitt-
ently wash the top of the mist eliminator.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Washing
Scrubbers
Calcium Oxides
Limestone
Tests
Air Pollution Control
Stationary Sources
Mist Eliminators
Vertical Dacts
High-velocity Scrubbing
13B
13H, 07A
07B
14B
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
58
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
50
------- |