EPA-450/2-73-007
OPERATION AND PERFORMANCE
OF THE LIME SCRUBBING
SYSTEM AT MITSUI
ALUMINUM COMPANY, LTD.
.S. ENVIRONMENTAL PROTECTION AGENCY
-------
EPA-450/2-73-007
OPERATION AND PERFORMANCE
OF THE LIME SCRUBBING
SYSTEM AT MITSUI
ALUMINUM COMPANY, LTD.
by
James A. McCarthy and James F. Durham
Emission Standards and Engineering Division
December 1973
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations •* as supplies permit - from the
Air Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina'27711, or from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22151.
Publication No. EPA-450/Z-73-007
11
-------
PREFACE
From July 4 through July 7, 1973, an inspection team from the U. S.
Environmental Protection Agency (EPA) visited the Mitsui Aluminum Company,
Ltd., in Omuta, Japan, for the purpose of obtaining first-hand information
on the performance of the Mitsui Aluminum system for controlling sulfur
dioxide emissions. The visit was timed to coincide with the annual 24-hour
shutdown of the power plant boiler for maintenance so that the control
equipment could be inspected during shutdown as well as during operation.
The EPA inspection was conducted by Mr. James A. McCarthy and Mr. James F.
Durham. This report presents the documentation of the observations and
data collected by the EPA team on the operation and perfoijmance of the
Mitsui sulfur dioxide control system.
111
-------
CONTENTS
Page
LIST OF TABLES v
JLIST OF FIGURES vi
SUMMARY AND CONCLUSIONS 1
INTRODUCTION 3
GENERAL PROCESS DESCRIPTION 3
OPERATION OF THE MITSUI POWER PLANT SCRUBBER SYSTEM 7
OPERATING HISTORY 7
ON-SITE INSPECTION OF SYSTEM 13
Inspection During Operation .,...' -.* 13
Inspection During Shutdown 22
OPERATING PARAMETERS AND PERFORMANCE DATA 23
APPENDIX A. DATA TABLES. ''....' A-l
APPENDIX B. DETAILED REPORT ON TROUBLE IN ASH POND WATER PUMPS ... B-l
APPENDIX C. SUMMARY OF SIGNIFICANT LOAD VARIATIONS C-l
APPENDIX D. RELATIONSHIP BETWEEN ASH POND WATER AND LEAKAGE WATER.
D-l
APPENDIX E. CLIMATIC DATA FOR OMUTA AREA E-l
IV
-------
LIST OF TABLES
Table Page
1. Operating History of Mitsui Aluminum Scrubber Module. ...... 8
2. Scrubbing System Electric Power Consumption 12
3. Water Analyses . 15
4. SO- Concentrations in Scrubber and Stack Gas Streams 16
5. Scrubber Pressure Drop 19
6. Measurement of Liquid Stream Parameters 20
7. On-Site Inspection of Scrubber Internals During Shutdown. ... 24
8. Characteristics of Power and S02 Control Plants at Mitsui
Aluminum . . * 25
9. Operating Data 27
10. Analysis of Coal Burned at Mitsui Aluminum Company 28
4
A-l. Schedule of Scrubber Installation and Operation at Mitsui
Aluminum Company, Oimata, Japan. A-l
A-2. Summary of Operating Data for 13 Months at SO- Control System
for Mitsui Aluminum A-2
A-3. Summary of Operating Data for Two-Scrubber S02 Control System
at Mitsui Aluminum A-3
A-4. Mitsui Performance Test Data A-4
E-l. Monthly Climatic Data for the City of Omuta E-l
E-2. Rainfall for June 18 through July 7, 1973, in Omuta E-2
-------
LIST OF FIGURES
Figure Page
1. Schematic of Two-Stage Venturi Scrubber (Chemico Design)
at the Mitsui Aluminum Company 5
2. Process Flow Sheet 6
3. Water Balance for Scrubber (Estimated for 75% Gas Treatment) . 12
4. Location of Mitsui Aluminum Company Slurry Ponds as of
July 5, 1973 14
5. Boiler Gas Flow Schematic 17
6. Schematic of Liquid Streams on Mitsui SO- Scrubbing System
as of July 4, 1973 18
VI
-------
OPERATION AND PERFORMANCE
OF THE LIME SCRUBBING
SYSTEM AT MITSUI
ALUMINUM COMPANY, LTD.
SUMMARY AND CONCLUSIONS
Specific conclusions reached by the inspection team after on-site
inspection of the system'and an examination of the operating and performance
data are:
1. Since initial startup in March 1972 until July 1973:
a. S02 removal efficiencies have been maintained at 86 to
93 percent.
b. No boiler outages or shutdowns have been caused by scrubber
malfunction or by any cause attributable to the scrubber
installation.
c. No scrubber malfunction has occurred, so that it has not
been necessary to bypass the scrubbing system.
2. From initial startup until June 12, 1973, only one of the two
scrubbing modules was in operation at a time and it was used to
treat 60 to 75 percent of the total boiler flue-gas volume. Since
June 12, 1973, both scrubbers have been operated and 92 percent of
the total boiler flue-gas volume has been treated. There has been
no need to switch from one scrubber to the other at any time because
of malfunction.
3. Available data indicate that the scrubbing system has operated
under closed-loop conditions during all but a few days of the year,
-------
at which time heavy rainfalls created overflow conditions and open-
loop operation was used.
4. Control of pH adequate to prevent scale formation and to achieve
efficient S02 removal has been maintained during significant load
changes by manually adjusting the rate of addition of fresh carbide
sludge to the scrubber system.
5. Further inspection on shutdown revealed that there were no significant
deposits or deterioration in the system.
6. The system's performance has met all of Mitsui Aluminum's require-
ments and they are planning .installation of another, similar system
that will use limestone.
7. Gypsum production is planned for the future.
8. Installation of the lime scrubbing system has ensured that Mitsui
Aluminum will achieve compliance with air pollution control regula-
tions but will not jeopardize the long-term stable supply of cheap
fuel. This has upgfaded the economy of the coal mining area by
permitting the use of a locally available low-grade, relatively
high-sulfur and high-ash coal. This "slime" coal is a byproduct
from the coal preparation plant that processes the coal from the
Mitsui Mining Company's Omuta mine.
-------
INTRODUCTION
The Mitsui Aluminum Company, Ltd., owns and operates a primary aluminum
smelter in the City of Omuta on the Island of Kyushu in Japan. The electrical
energy required for operation of the aluminum smelter is supplied by a 156-MW
coal-fired boiler, which is also owned and operated by Mitsui Aluminum. This
is one of the largest private power stations in Japan. One of the primary
purposes for locating the aluminum plant at Omuta was to make use of the low-
grade slime coal that is being produced locally by the Mitsui Mining Company,
Ltd. Slime coal is a low-heat-content coal from the coal cleaning plant that
has little or no marke't value. Utilization of this slime coal by Mitsui Aluminum
will ensure that the company has a stable supply of relatively cheap fuel and,
in addition, will upgrade the economy of the coal mining area.
In 1968, when the boiler was under construction, Japan had no stringent SO^''
regulations; however, in 1969, in anticipation of future regulations, Mitsui
Aluminum initiated an S02 control program for their coal-fired boiler in Omuta
in order to be able to continue to use the low-grade slime coal. This program
included a literature search, fundamental studies, and visits to Europe and the
U.S.A. On June 9, 1971, an order was placed with the Mitsui Miike Machinery
Company, Ltd., to install an S02 control system that was based on technology
developed by the Chemical Construction Company (Chemico), an American chemical
engineering firm. In July 1971, pilot-scale testing with a 2500 NnrVhr test unit
was initiated. Construction of the foundation for the full-scale S02 control
system began on August 10, 19.71; the absorption towers were erected on October
10, 1971; and operation began on March 29, 1972. A log of important dates in
the installation and operation of the Mitsui system is shown in Table A-l (in
the Appendix).
GENERAL PROCESS DESCRIPTION
Flue gas from the 156-MW coal-fired boiler passes through an electrostatic
precipitator and then enters the scrubbing system or bypasses it and goes directly
to the stack. Carbide sludge, a byproduct of acetylene manufacturing that is
similar in chemical composition to commercial lime, is used to react with the S02
in the stack gas. Specific details describing the basic design, cost, and opera-
ting parameters of the Mitsui Aluminum Company's lime scrubbing system have been
documented in a recent paper authored by the Mitsui Aluminum power station manager;
-------
his information on basic design, along with information collected by the in-
spection team, is included in this report.* The SO- control process incor-
porates many features that are designed to achieve both reliable operation
and efficient SC^ removal. These features are discussed below.
The scrubbing system is designed with 50 percent spare gas-handling
capacity. Two identical and independent fixed-throat venturi scrubber modules
(A and B) are installed in parallel; each is designed to handle 75 percent of
the total boiler flue gases at full load, The installation of spare gas-handling
capacity and parallel modules enables repairs to be made without completely
shutting down or bypassing the scrubbing system, or shutting down the boiler.
A schematic of the scrubber and a flow diagram of the scrubbing system are shown
in. Figures 1 and 2.
Multistage gas-slurry contacting has been incorporated into-the scrubber
design for high-efficiency SC^ removal. Each-scrubbing module contains two
stages of gas and slurry contacting. The sides of the scrubber throats are
kept wet to improve gas-slurry contacting and to reduce scale buildup. This
is done by introducing the slurry through bull nozzles and tangential inlets.
9
Several features have been incorporated into the process design to reduce
scale formation and buildup of solids on scrubber internals, including opera-
tion at high liquid throughputs, installation of a delay tank, and the use of
unsaturated makeup water to wash the demisters.
The simple straight-through flow design of the venturi scrubber minimizes
the surface area upon which scale can form. The flue gases passing through
the first and second stage throat change direction before passing through the
mist eliminator (demister). "The centrifugal action that results from reversing
the gas flow reduces jthe quantity of solid and liquid that must be removed by
the mist eliminator and therefore reduces the possibility of plugging the
demister. In addition, the top and bottom of the demisters of both stages are
washed intermittently with either fresh water or makeup water mixed with pond
water to further minimize plugging of the demister.
1Sakanishi, Jun and R.H. Quig. "One Year's Performance and Operability of the
Chemico/Mitsui Carbide Sludge (Lime).Additive S02 Scrubbing System." Presented
at the EPA Flue Gas Desulfurization Symposium, New Orleans, Louisiana, May 15-17,
1973.
-------
TANGENTIAL
SLURRY '
DEMISTER SPRAY
WATER
COUNTERCURRENT
SLURRY SPRAY
TANGENTIAL
SLURRY
DEMISTER
SPRAY
WATER
COUNTERCURRENT
SLURRY SPRAY
BULL
NOZZLE-
SLURRY
BOILER
FLUE GASES
BULL
NOZZLE
SLURRY
FIRST STAGE
SLURRY DRAIN-
SECOND STAGE
SLURRY DRAIN
Figure 1. Schematic of two-stage venturi scrubber (Chemico design) at the
Mitsui Aluminum Company.
-------
ELECTROSTATIC
PRECIPITATOR
SECOND STAGE
RECYCLE
COAL
WET CARBIDE PIT
• k "PUMP
• I
DELAY
TANK
(EFFLUENT
HOLD TANK)
FIRST STAGE
RECYCLE
PUMP
-
MAKEUP SLURRY
FEED PUMP
BLEED SLURRY TRANSFERTUMP
__
ASH POND LIQUOR
RETURN PUMP
DRY CARBIDE PIT
RECYCLE SLURRY
MAKEUP SLURRY
BLEED SLURRY
RETURN LIQUOR
WASTE DISPOSAL POND
Figure 2. Process flow sheet.
-------
The delay tank, a 1800-jn^ agitated vessel, increases reliable scrubber
operation, for it reduces the degree of scrubbing liquor supersaturation.
Approximately 30 minutes' design residence time is provided so that the super-
saturated salts contained in the liquor leaving the scrubber can precipitate
from solution in the delay tank before the liquor is recycled to the scrubber.
Residence time is presently somewhat less than 30 minutes because flow
rates are greater than originally planned.
The scrubbing liquor is pumped through the scrubber at a high rate in
order to prevent the droplets from becoming highly supersaturated. Erosion
problems that can result from the presence of solids in the scrubbing solution
have been minimized through maintenance of adequate flow rate. The solids content
is held at a low level (5.5 percent),, which provides adequate "seed crystals"
upon which supersaturated salts can precipitate, thus further reducing the
potential for scale buildup.
Additional precautions have been taken to prevent or at least minimize
downtime that could occur as the result of erosion, corrosion, and pump failure.
To prevent corrosion, the scrubber is constructed of carbon steel, the
interior is lined with a glass-flake reinforced polyester material, and erosion-
prone areas, such as the recycle liquor line piping and pumps, are rubber-lined.
Where the reinforced polyester materials are subject to abrasion, 316-stainless-
steel liners are used.
Finally, spare pumping capacity has been installed in critical areas such
as the first- and second-stage recycle pumps.
OPERATION OF THE MITSUI POWER PLANT SCRUBBING SYSTEM
Operating History
The lime scrubbing system was placed in commercial operation on March 29,
1972, and only one scrubbing module was in service at a time from initial startup
until June 12, 1973. On June 12, both modules were placed in service. The
service period for each scrubbing module is summarized in Table 1. From startup
in March 1972 until June 12, 1973, 60 to 75 percent of the total boiler flue
gas was being treated by the scrubbing system, and 92 percent of the gas was
treated from June 12 to July 5, 1973. The objective has been to minimize
operating costs by treating the minimum quantity of gas required for compliance
with existing regulatory requirements. Since June 12, 92 percent has been
treated in order to optimize operation under conditions necessary to meet future
S02 control regulations. From initial startup until July 5, 1973, no interruptions
in boiler operation have occurred as a result of problems with the system.
-------
TABLE 1. OPERATING HISTORY OF MITSUI ALUMINUM SCRUBBER MODULES
Service Scrubbing module Stack gas
dates in service being treated, %
March 29, 1972 to B module 75
April 10, 1972
April 10, 1972 to A module 60 to 75
October 17, 1972a
November 1972 to B module 75
March-, 1973b
March 3, 1973 to A module 75
June 12, 1973
June 12, 1973 to A § B modules 92
July 6, 1973
July 5, 1973 Scrubber shut down because of scheduled
maintenance in boiler
Q
Scrubber out due to boiler outage.
Boiler was shut down because of boiler tube failure in March 1973, but
scrubber internals were not cleaned.
-------
Similarly, it has not been necessary to bypass or shut down the scrubbing
system in order to perform maintenance on the system. The scrubber internals on both
modules were inspected and cleaned for the first time in October 1972, 6
months after startup, when the scrubbing system was shut down for annual
boiler inspection and maintenance. This inspection revealed that a minor
modification in the mist eliminator spray pattern was- required to. reduce
solids buildup in specific sections of the demister.. No significant erosion
or corrosion, was found in the piping system. This- includes the scrubbing
liquor recycle piping to and from the scrubber and the piping to and from
the sludge pond.
Since startup, with one exception, there has been no significant increase
in pressure drop across either Module A or B because of solids, buildup in the
scrubber internals. The one exception occurred in June 1972 when wooden chips
plugged, the suction side of the sludge pond water return, pump. The substituted
pump did not have sufficient pressure to effectively wash the demister, and
a pressure drop increase was observed across the first-stage mist eliminator
until the sludge pond water'return pump was restored to service. Details
on this pressure-drop problem are given in Appendix B.
Removal of S02 is achieved by circulating the scrubbing liquor through
both the first and second stages of each module at a constant, rate. Approxi-
mately 5200 m /hr (23,000 gal/min) of scrubbing liquor is circulated through
the scrubbing system with one module on line treating 75 percent of the flue
gas. The lime concentration in the scrubbing liquor is also maintained at a
constant level through tight pH control. Control of pH, which is critical to
both SO- removal and scale prevention, is accomplished manually. The pH of
the first- and second-stage recycle liquor is measured with a lab pH meter
every hour, and the quantity of makeup slurry is then manually adjusted as
necessary to maintain the specified pH level. This technique enables control
to be maintained within close tolerance, generally within ±0.2 pH unit, which
is more than, adequate for SC>2 removal as well, as scale prevention. Mitsui
Aluminum considers their specific operating. pH value to be confidential, but
they have operated with a. variety of levels of pH between 6 and 9. The control
of pH to within ±0.2 is merely for optimum operation.
-------
Maintenance of the liquid flow rate to the scrubber and the lime concentration
in the scrubbing solution determines the quantity of lime that is available to
react with the S0_ at any given time. Thus, the actual SO- removal that will be
achieved for any given scrubbing device depends on the quantity of flue gas
treated and the S02 concentration present in the flue gas. For example, decreasing
the quantity of flue gas treated will increase the S0_ removal efficiency. The
S02 removal efficiency depends on the pH values of the scrubbing liquor and,
consequently, on the stoichiometric lime feed rate. At low S02 concentrations,
with the same lime feed rate, the ratio of lime to SO- (stoichiometry) is greater
and, therefore, removal efficiency is greater. The Mitsui management thinks that re-
moval,' efficiency would be equal or slightly greater for higher S02 concentrations if
stoichiometry were constant. At Mitsui Aluminum lime feed is varied
to maintain a constant stoichiometry and a desired pH. Under this mode of
operation, S0_ removal is kept relatively constant with varying load and S02
concentrations. Until June 12, 1973, the gas volumes treated in each module
have ranged from 60 to 75 percent of the total boiler flue gas flow (193,000
scfm to 241,000 scfm). The scrubber inlet S02 concentrations have ranged from
1800 ppm to 2300 ppm, and the corresponding scrubber outlet SC^ concentrations
from 115 ppm to about 300 ppm. The S02 removal efficiencies have varied from
about 86 percent to 93 percent. The low S02 concentration occurred during the
first 3 or 4 months of operation when lower-sulfur coal was the primary fuel.
However, the percentage of the higher-sulfur slime coal used as boiler fuel
has increased from 30 percent to^over 80 percent by weight (dry basis) and has
caused a corresponding increase in the inlet S02 concentration. An inlet S02
concentration of 2000 to 2100 ppm is more representative of recent operations.
Since startup, the boiler has operated at or close to its design output
of 156 MW and does not fluctuate significantly from the design output during
normal operation. Therefore, the gas volumes and the quantity of S02 being
fed to the scrubber have remained fairly constant. Several instances have
occurred, however, in which the S0_ load to the scrubber has changed significantly.
These load changes have resulted when the boiler has been shut down or started up,
or when special situations have required that the load be increased or decreased.
During major load changes, it has been necessary to gradually increase or decrease
the volume of gas to the scrubber and at the same time manually adjust the lime
makeup to the scrubbing system to keep the pH within safe operating limits.
10
-------
Because the risk of possible scale formation in the scrubbing system was recognized,
a several-hour period was initially allowed for startup and shutdown. However,
as confidence developed and experience was obtained, the time required to start
up, shut down, or decrease the load to the scrubbing system decreased
significantly. During the shutdown on July 5, 1973, with both modules on line
and 92 percent of the stack gases being treated, the scrubbing system was shut
down in less than 90 minutes. During this period, manual reduction of the carbide
sludge makeup enabled the pH to be maintained to within ±0.2 pH unit from its
control value. Specific details describing the load changes and a 13-month summary
of operating data are contained in Appendix C and Appendix A, respectively.
Two full-time operators are required to operate the scrubbing system; one to
operate the carbide sludge makeup system and the other to man the scrubbing unit coritn
room.
The electrical power required to operate the scrubbing system with one module
on line treating 75 percent of the gas flow is about 2450 kl\. This is equivalent
to about 1.6 percent of the power plant capacity of 156 MW. Almost 55 percent of
the power requirement is fot operation of the booster fans that force the flue
gas through the scrubbing system. A more detailed breakdown of the power require-
ments is presented in Table 2.
Water loss resulting from evaporation and entrainment is inherent in wet
scrubbing processes. Mitsui Aluminum loses approximately 14 m^/hour of water
out of the stack, which must be replaced in order to maintain the required
scrubbing liquor volume. The majority of this makeup water is mixed with the
slurry pond return water, part of which is used to wash the demister. The
remaining makeup comes from'water contained in the carbide sludge and from
industrial quality water. An estimated water balance is presented in Figure 3.
When two modules are running, the water makeup will increase by about 20
percent because of greater evaporative loss. Thus, with both modules on, more
industrial water is used for makeup watei1 in the form of mist eliminator washing.
In order to remove spent reaction products, fly ash, and other solids from
the system, approximately 95 m /hr (419 gal/min) of recirculating scrubbing liquor
is pumped to the sludge disposal pond, where the solids settle out. Super-
natant liquid is then returned from the pond to the scrubbing system at a rate
of 100 m3/hr (440 gal/min).
11
-------
Table 2. SCRUBBING SYSTEM ELECTRIC POWER CONSUMPTION
Actual power
Total motor
Equipment
Pieces consumption, kW capacity, kW
Booster fans 2 1350 2000
First-stage recycle pumps 2 442 480
Second-stage recycle pumps 2 380 400
Pond return water pump 1 38 45
Other 236 378
Total
INDUSTRIAL
WATER MAKEUP
7.5 m3/hr
POND WATER
INCLUDING
WATER MA
100 m3/hr
2446 3303
BOTTOM SEAL ON
STANDBY SCRUBBER
ENTRAINMENT EVAPORATION
CARBIDE SLUDGE 1 f f
WATER CONTENT 1 1.5m3/hr| 1 12.5 m3/hr
l.SnrVhr f ..
*" OPERATING .
SCRUBBER !
•*
RETURN , , ,.
MINE MINE WATER. ^EVAPORATION 9laftA"
KEUP 20m3/hr • tVAKUKA I ION
J. 1 , , ,. •• * ffl*
— Y -T 1 I 95 m3/hr
1
RECYCLE
i/hr
SETTLED SOLIDS
8 m3/hr
Figure 3. Water balance for scrubber (estimated for 75% gas treatment).
12
-------
A diagram of the current pond design is presented in Figure 4. At the
present time, pond B is filled with sludge and the scrubber wastes are
disposed of in pond C. The saturated supernatant from pond C percolates
through the fly ash dike and into pond A, from which it is returned to the
scrubbing system.
In addition to spent scrubbing liquor, waste mine water from the nearby
coal mines is also discharged into pond A. This mine water thus provides part
of the makeup water that is required for the scrubbing system. The sludge
disposal pond provides a convenient place for disposing of this mine drainage
waste stream and has the additional advantage of providing makeup water in place of
industrial water. The mine water has a high sulfate concentration. See Table 3 for ai,
analysis.
As part of their expansion plans, Mitsui Aluminum is going to build a
175 MW boiler at their Omuta plant that will also be equipped with an SO-
scrubbing system, but, since there is insufficient carbide sludge to use at
both the existing unit and the new unit, limestone will be used as an absorbent
along with a proprietary catalyst developed by Mitsui Miike to improve limestone
utilization. Because of a shortage of storage space and available markets, a
gypsum plant will be constructed that will use as a feedstock the byproducts
from the planned S02 control systems as well as possibly the byproduct from
the existing scrubber. Another option being considered for disposing of the
existing sludge is to use it as a sealant in the mined-out coal mines. No decision
between these options has been made.
ON-SITE INSPECTION OF SYSTEM
Inspection During Operation
The inspection team had complete freedom to look at all portions of the
system, to take samples and photographs, and to ask any questions. Because of
this cooperation, it was possible during our 4-day visit to obtain what we
believe is a full understanding of the system.
Another investigator^ performed some stack sampling tests on July 4, 1973,
and made his data available to the EPA inspection team. Tests for S02 were made
2
Dr. Howard E. Hesketh, Associate Professor of Engineering, Air Pollution Control,
Southern Illinois University, was visiting the plant concurrently but independently.
This information was presented in testimony before the Pollution Control Board of
the State of Illinois on July 23, 1973.
* Mention of commercial products or company names does not constitute endorsement
by the U. S. Environmental Protection Agency.
13
-------
WOOD
STORAGE
YARD
(S'ALT WATER)
UNDERGROUND MINE
WATER DRAIN
CLEAR WATER RETURN
LINE TO PLANT
X^ SPARE LINE
LINE
FROM PLANT
WOOD
STORAGE
YARD
150ft
MIIKE GOBAN CO.
VENEER FACTORY
ARIAKE
INLAND1
BAY
NOTE: NOT TO SCALE
Figuce4. Location of Mitsui Aluminum Company slurry ponds as of July 5,.1973.
14
-------
Table 3. WATER ANALYSES
Analysis
pH
Conductivity, yV/cm
Ions, ppm
Ca++
Mg+*
Mn**
so4"
Cl"
N03-
Si03~~
Sludge disposal pond,
April 1972 to June 1973
7.2 to 8..6
2600 to 5690
276 to 797
48 to 105
Trace
959 to 2070
405 to 1192
Trace to 4.58
-
Mine
water
7.6
3540
238
99.6
-
808
662.5
-
-
Industrial
water
7.1
173
14.2
3.56
-
23.8
6.06
3.9
18.5
Suspended solids, ppm 3.0 to 11.6
COD, ppm 3.2 to 20.8
15
-------
using an Unico #400 precision gas detector* and Kitagawa colorimetric detector
tubes #103bf* (expiration date May 25, 1974) for S02 in flue gas (0.02 to 0.30 percent)
Samples in lines under positive pressure were taken at the points indicated
on the boiler gas flow schematic in Figure 5. .
Sulfur dioxide is monitored continuously at the three positions marked
on Figure 5. The scrubber inlet SC>2 concentration before the induced-draft
fan is measured by a conductivity meter and the scrubber exit and stack S02
concentrations are measured by an infrared meter. One meter is used for inlet
and outlet readings by switching the meter every 2 hours from one sample to
the other. It is assumed that the gas concentration is essentially uniform
in duplicate ducts. The plant meters are calibrated daily using calibration
gas mixtures.
The gas tube readings are listed in Table 4 with the plant meter readings for
comparable sampling times on July 4, 1973.
Table 4. SO CONCENTRATIONS IN SCRUBBER AND STACK GAS STREAMS
(ppm)
Measurement method
Time
3:40 pm
4:15 pm
3:50 pm
3:50 pm
Sample
. 1
3
2
Gas stream
Scrubber inlet
Scrubber outlet
Stack inlet
Stack
Gas tube Meter
-v!600 2160
%200 290
^400
a
^520 510
a This value is estimated using our scrubber inlet and stack inlet values,
knowing that 90 percent of the gases are being scrubbed:
(0.90)(400) + (0.10)(1600) = 520.
These data indicate that the plant's analyses of S02 gas concentrations are
correct. Note that the stack gas contains higher S02 concentrations than
the exit gas from the scrubber because of added S02 from the oil-fired reheater
and the mixing with unscrubbed flue gases that occurs.
The gas flow rate to each scrubber system is 45 percent of 319,000 scfm, or
144,000 scfm.
Dr. Hesketh, the other investigator, personally sampled all available liquid
streams for pH and for anything of visual significance. The sampling points he
used are marked on Figure 6.
16:
-------
BOILER
CONDUCTIVITY
METER
SAMPLE
ELECTROSTATIC PRECIPITATOR
IXING
CHAMBER
AND
REHEAT
GAS TUBE
SAMPLE NO. 2
GAS TUBE
SAMPLE NO. 1
SAMPLING
POINTS
INFRARED
METER
SAMPLE
Figures. Boiler gas flow schematic.
17
-------
CARBIDE
SLUDGE
NO. 4
MAKE UP
SLURRY
TANK
9 NO. 5
' ll %Dl
Flip" Tl
MINE SHAFT
NO. 6
INDUSTRIAL WATER
TO DEMISTERS
TO SCRUBBER B
DEMISTERS
•TO BAY
LIQUID SAMPLE POINTS
•NO. 1
DELAY
TANK
(EFFLUENT
HOLD TANK)
Figure 6. Schematic of liquid streams on Mitsui S02 scrubbing system as of July 4, 1973.
18
-------
Table 5. SCRUBBER PRESSURE DROP*
Pressure drop, mm H^
Location
First-stage scrubber throat
First -stage mist eliminator
Second-stage throat
Second-stage mist eliminator
Total scrubber
A Module
85
9.0
63
4
145
B Module
70
7.0
75
2.5
130
O '
These readings were taken by the EPA inspection team from the control room at
10:30 a.m. on July 5, 1973. Both modules A and B were on line treating 92
percent of the total boiler flue gas. For comparison purposes, the design
pressure drop under similar operating conditions is approximately 150 mm
19
-------
The pH measurements were made immediately with pHydrion paper and after
several minutes with the plant electronic pH meter. Results of the pH tests
made July 4, 1973, by Dr. Hesketh are as follows:
Table 6. MEASUREMENT OF LIQUID STREAM PARAMETERS
11:05
Location or
Liquid sample
Time
10:35
10:45
Sample
1
2
substance sampled
a
Delay tank
Bottom second-
pH
d
6.0-9.0
A
Temperature, "C
36
Solids, %
5.5
Liquid color
^Colorless
stage A 5 B
Makeup slurry
6.0-9.0
.47
Colorless
11:30 4a
11:00 4b
11:15 5
11:55 6
12:30 7
12:45
-
(15% solids)
Wet carbide sludge
Dry carbide sludge
Pond recycle water
. to slurry makeup
and scrubber A
demisters
Mine water discharge
Pond A water pumps
Industrial water
Ocean water
12.25 30
. . .
-
8.0b 30
7.55 31
8.1b 31
7.05 27
8.0C
15
High
100
0
Low
0
0
-
-
-
(Grey powder
^Colorless
Brownish
Colorless
Colorless
-
This is slurry to pond C plus recycle to scrubbers A and B and contains 5.5 percent solids.
The slurry comes from the bottom of the first stage and from the second stage bleed from
both A'and B scrubbers.
The high ion concentration in this solid-free water caused the paper to indicate pH = 7
for both of these.
Ocean water data from Mr. Sakanishi's reports.
Exact pH was recorded but is confidential.
20
-------
Considerable time was spent studying the slurry ponds. The slurry ponds
are located about 2 miles south of the power station. Three pipes, approximately
6 inches in diameter, connect the ponds with the plant. One pipe returns the
clear water to the plant, the second pipe transports the 5.5 percent solids
slurry to the ponds, and the third pipe is a spare.
The three slurry ponds are now arranged as shown in Figure 4 (page 14).
Pond A contains the clear water that is returned to the system for slurry makeup
and demister spraying. In July, pond C was being filled and the slurry appeared
to be solidifying in one-fourth of the pond nearest the fill pipe. Pond B was
full of set-up slurry and had not been used since last February.
The wall between ponds A and C had only recently been completed. It is
made of fly ash and permits water from the slurry added to C to percolate
through to pond A. The south side of pond A appeared slightly colored as if
some slurry had penetrated the fly ash wall; however, it was pointed out that
pond C was being filled before the wall was completed and that the coloration
was probably a result of diffusion in the pond before the wall was present.
The northeast end of p'ond A was slightly brownish on July 4, 1973, indicating
that mine water was flowing into the pond because there had not been enough rain
to raise the level of pond A to the overflow level at the pumping station. Note
in Figure 4 that excess mine water discharges into a wood storage pond via an
underground drain. During the evening of July 4, 1973, a substantial rain fell
(0.67 inch) so that on July 5, 1973, no mine water was observed entering pond A
but rather a small amount (compared with that overflowing into the pumping
station for return to the scrubber) of water was seen to be leaving the pond
through the mine water entrance. At Omuta, the rainfall, as shown in the climatic
data given in Appendix E, is not evenly spread over the entire year; instead,
over half comes during a 3-month rainy season. Because of its design, the
drainage ditch acts as a level-control device for the pond. Rainfall at the
Miike power station is significant because during the dry season mine waters
flow into the clear slurry pond A; but after periods of heavy rain the flow
reverses and excess pond water leaves through the mine water entrance. Thus,
during rainy periods, the rainfall is larger than the losses resulting from
humidification, entrainment, and evaporation so that a small design over-
flow condition is created. Mitsui[Aluminum personnal indicated that this
overflow occurs only a few times during the rainy season and then in only
a small amount for a short duratios. During the first day of inspection (July
4, 1973), mine water was flowing i^to the pond. During the night of July 4-5, there
21
-------
was a heavy rainfall (0.67 inch). The next day a small flow (relative to
the flow returning to the scrubber) was noticed going from pond A out the mine
drainage ditch into the wood pond.
The west ends of ponds A and C and the south side of C are lined with
polyethylene, although the lining was not visible in July. In addition to
this liner, there is about 100 feet of dirt between the west ends of ponds A
and C and the wood storage pond. There is about 300 feet of dirt between the
bay to the south of pond C. The wood storage pond is open to the bay water and
is lower than pond A, depending on the tide. Pond A is the lowest; C is higher;
and B is highest.
Several series of water samples were taken on the shore along the south
side of the veneer factory at low tide by Mr. Sakanishi of the Mitsui Aluminum
Company. Results are given in Appendix D.
Pond B, which has not been used since February, appeared to be completely
solid; however, it was actually thixotropic (became liquid if agitated).
The pond was solid enough for Mr. Kodama of the Tokyo Chemico office to stand
•
on its surface before and after a heavy rain. Some Mitsui Miike Mining Company
employees were digging at the north end of pond B to gather slurry samples to
test for use as fill in abandoned mine shafts. The hole they dug showed what
appeared to be a solid clay-like material. When struck with a shovel, the slurry
was hard like clay, but when it was worked in the hands it was found to be
thixotropic, turning into a pasty fluid.
Inspection During Shutdown
The inspection team was fortunate in being able to observe the scrubber
while it was shut down for boiler maintenance as well as while it was operating.
The boiler began shutting down at 10:00 p.m. on July 5, 1973, and the booster
fans were shut down at 11:37 p.m. Demister water washing was stopped when the
booster fans were shut off and the venturi slurry pumps.were operated until the
scrubbers were opened for inspection on the morning of July 7, 1973. Prior to
the inspection, B module had been in service about 5 months and A module for 4
months since the last cleaning in November 1972. Small quantities of solids
were observed in the first four demister layers in the first-stage mist eliminator
of module B. This buildup, however, did not cause any increase in pressure drop
readings as shown by the data in Table 5 taken on July 5, 1973, the morning before
the shutdown. A more detailed description of the July 7 inspection is summarized
in Table A-3 (in Appendix A).
22
-------
The only buildup of scale in scrubber A was observed on the throat
braces and consisted of a piece of scale about 1-1/2 by 3 inches. The scale
on one of the braces of scrubber B was slightly larger (approximately 4 by 4
inches). Very little scale existed even at the slurry inlet of the first
stage of scrubber A.
The first-stage polypropylene mist eliminators of scrubber B were almost
completely clean. A light hairy type of crystalline growth was observed on
the first-stage cone above the mist eliminators. Areas on the cone near a mist
eliminator spray nozzle were completely clean. The mist eliminators and cone
were washed in our presence with a mild stream of water from a hose and the
material present washed away as if it were ice. Chunks of scale on the throat
braces, however, did not easily wash away. It should be noted that this scale
apparently had no effect on operation.
Inspection of the second-stage mist eliminators and cone showed they were
almost perfectly clean. Close inspection of the first-stage mist eliminators
in scrubber A showed that very slight buildups were present in areas of the
demister not located direct'ly below a spray nozzle. At the point of maximum
scaling, scale was removed from as many layers as possible. Of the six layers
comprising the demister, scale samples were obtained from the top layer, second
layer, and fourth layer. The scale was measured as 0.5 to 2 mm thick on the
second layer and 1 to 6 mm thick on the fourth layer. The scale was about 2
by 5 inches in size.
The only significant scaling found in the scrubbers was found in scrubber
A, at about the center of the unit in which holes drilled in an inner manhole
allowed slurry to pass from the inner to outer sections and allowed it to
form a piece of scale about 1 inch thick by 8 inches wide by 28 inches long.
This scale buildup did not and would not result in any operational problem
and is only noted for interest.
Pumps and lines were checked by inspecting one slurry pump section. The
slurry pump and lines are original to the system and are located at the north-
west end of the scrubber area. The stainless steel pump impeller and lines appeared
to be in perfect condition. Inspection of the rubber-lined pump volute showed
that some wear had occurred above the discharging opening. A summary of information
obtained from out on-site inspection of the scrubbers during shutdown is given in
Table 7.
23
-------
Table 7. ON-SITE INSPECTION OF SCRUBBER INTERNALS DURING SHUTDOWN0
Scrubbing module
Inspection point
Observations
Module A
First-stage scrubber No significant deposits
throat .. or scale was observed
Top of first-stage
mist eliminator
Top of mist eliminator
was essentially clean.
Scale deposits were found
in the lower layers of the
demister.
Module B
First-stage scrubber
throat
Top of first-stage
mist eliminator
Top of second-stage
mist eliminator between
first and second stage
throat
No significant scale deposits
were observed. Slightly more
scale was observed on a sup-
porting cross member.
Wo significant deposits
could be observed either
on the mist eliminator or
the throat body.
Cleaner than first stage
mist eliminator. No deposits
were observed.
Inspection conducted by the Power Station Manager, Mr. Jun Sakanishi;
Professor Howard Hesketh, Southern Illinois University, Carbondale,
Illinois; Masao Kodama, Chemico, Tokyo, Japan; and James A. McCarthy
and James F. Durham of the U. S. Environmental Protection Agency,
Durham, North Carolina.
24
-------
Table 8. CHARACTERISTICS OF POWER AND S02 CONTROL
PLANTS AT MITSUI ALUMINUM
Plant and equipment
Characteristics
Electric power plant
Generator
Turbine
Boiler
Combustion system
Coal treatment
Electrostatic precipitator
Stack
SO- control plant
Process
Capacity
Chemico guarantee
Equipment details
Disposal system
Solids in recycle slurry
Delay tank residence time
174,000 kVA, 15,000 V, 60 Hz.
156,250 kW, 3,600 rpm.
490 tons/hr steam production,
manufactured by F.W. - 141 - I.H.I.,
single-drum natural-circulation type.
Pulverized-coal firing, front-wall
firing.
Two slime coal dryers, having a
capacity of 24 tons/hr and 45 tons/hr
each.
Dust removal efficiency: 98.7%.
Outlet dust loading: 0.25 gr/scfd.
425 ft. high, concrete outer shell.
Chemico/Mitsui Miike Machinery process,
wet type and calcium base.
2 units each handling 241,000 scfm of flue
gas. Total boiler flue gas is 319,000
scfm.
Dust removal efficiency: 90% or more
from gas cleaned by electrostatic
precipitator.
S02 removal efficiency: 90% or more.
Stoichiometric requirement of absorbent:
120% for 90% S02 removal.
Booster fans: 1,000 kW each; 2 units
Recycle pumps: approximately 200 kW each;
6 units.
Reheat furnace: 1 unit including fan, etc.
Throwaway system^ waste slurry is trans-
ferred to ash pond and supernatant is
returned from ash pond to S02 control
plant.
7.5 % (first stage)
5 % (second stage)
30 minutes
25
-------
Table 8 (continued). CHARACTERISTICS OF POWER AND S02
CONTROL PLANTS AT MITSUI ALUMINUM
Scrubbing 92% of gas Scrubbing 75% of gas
Scrubber A
First-stage slurry
Bull nozzle
Tangential
Countercurrent spray
Second-stage slurry
Bull nozzle
Tangential
Countercurrent spray
Inlet gas pressure (gauge)
First-stage throat, pressure drop
First-stage mist eliminator,
pressure drop
Second-stage throat, pressure drop
Second-stage mist eliminator,
pressure drop
Outlet pressure
Temperature
Scrubber inlet
Scrubber outlet
Mixer-reheater discharge
Stack inlet
Makeup slurry density, g/cc
Makeup slurry flow
Flow to first-stage recycle, m3/hr
Flow to second-stage recycle, m3/hr
Pond H20 to mist eliminators, m-Vhr
Industrial H20 to mist eliminators,
m3/hr
Bleed slurry to pond, m3/hr
Pond water return rate, m3/hr
First-stage L/G, gal/1000 scf
Second-stage L/G, gal/1000 scf
520
830
800
400
800
770
280
85
9
65
4
135
159°C
55
76
76
(A+B) 1.
(A+B)
(A+B)
88
(A+B)
(A+B)
.54.2
49.3
Scrubber B (one module)
Gas volume, m^/hr
520
840
820
370
810
890
Pressure, mm H20
280
70
7
75
2.5
150
158°C
55
87
84
085
13
10
39
13
107
134
54.2
49.3
600
900
1200
400
900
1200
430
175
28
145
7
95
158°C
56
79
77
7 to 22
6 to 16
67 to 91
0 to 14.5
91 to 95
90 to 114
43.3
39.4
26
-------
Table 9. OPERATING DATA
Representative operating data
(as of October 1972)
Output, kW 156.000
Coal: used, ton/day 1,900
heat content, Btu/lb 9,700
Sulfur content, % 1.9
Total boiler gas volume being
treated in scrubber 75% (385,000 Nra3/h4)(241,000 scfm)
S02 concentration, ppra
Boiler outlet 1,900
Scrubber outlet 180
Stack 610
Gas temperature, °F .
Boiler outlet 315
Scrubber outlet 130
Stack 212
Reheated gas 172
Ca(OH)2, stoichiometric requirement, % 105 to 110
Representative characteristics of coal
(as burned--80% slime and 20% cleaned)
Heat value, Btu/lb 9,700 to 10,000
Moisture, wt % 19.5
Ash content, wt %, dry basis 29.1
Volatile matter, wt %, dry basis 35.4
Fixed carbon, wt .%,, dry basis 36.9
Moisture content, wt %, dry basis 1
Sulfur, wt %, dry basis 1.9 to 2.4
Ash fusibility, °F at std pressure 2,330
Characteristics of carbide sludge
Moisture, % Consumption, ton/day Purity (air-dried basis), %
Wet carbide 60.6 ~80 88 as Ca(OH)2
Dry carbide 5.6 ~50 90 as Ca(OH)2
Total - ~110 (wet basis)
27
-------
OPERATING PARAMETERS AND PERFORMANCE DATA
Detailed information on the operating parameters and performance of the
Mitsui S02 scrubbing system has been presented in reports prepared by the power
station manager. During our inspection trip, a translated copy of performance
test data was obtained from Chemico Tokyo. These data, prepared by Mitsui Miike
Machinery Company, Ltd., are presented in Appendix A. Note that these data show
that better than 80 percent S0? removal can be obtained by lime scrubbing using
stoichiometric amounts of lime of 90 to 110 percent (as low as 86.7 to 87.3 percent
stoichiometry is shown with 84.6 to 85.0 percent S02 removal efficiency). Data
assembled by our inspection team from the Mitsui report and from our on-site
investigation are also presented in Appendix A.
Table 10, an analysis of the coal burned by Mitsui, is especially
significant relative to the operating costs for the scrubbing system.
The Miike power station had planned to use only 20 percent of the lower-grade,
cheaper "slime" coal, and the balance was to have been cleaned, pulverized coal;
however, the installation of the S02 scrubbing system made it possible for
the station to use about 80 percent slime coal. This fuel cost savings partially
pays for the scrubbers. Operating costs reported by Mr. Sakanishi (power plant
manager) are 0.1
-------
APPENDIX A.
DATA TABLES
-------
Table A-l. SCHEDULE OF SCRUBBER INSTALLATION AND OPERATION
AT MITSUI ALUMINUM COMPANY, OMUTA, JAPAN
Date
Comments
August 10, 1971
October 10, 1971
March 10, 1972
March 25, 1972
.March 29, 1972
April 10, 1972
April 25, 1972
May 8 to 13, 1973
October 17, 1972
November 7, 1972
March 3, 1973
June 12, 1973
July 6, 1973
Started foundation construction.
Started scrubber tower erection.
Started individual unit operation.
Started entire system (water operation).
Commercial operation began using B
scrubber to scrub 75 percent, of total
boiler flue- gas until April 10, 1972.
Changed over to scrubber A on 75
percent of gases until October 17, 1972.
Successful acceptance tests completed by Ministry
of International Trade and Industry (MITI).
Performance tests completed by Chemico.
Plant shut down for power plant maintenance.
Scrubbers inspected and cleaned.
Plant operation resumed using scrubber B on
75 percent of gases until March 3, 1973.
Changed over to scrubber A on 75 percent of
gases until June 12, 1973.
Started operation of both scrubbers in parallel
and scrubbed 90 percent of total boiler flue
gas (45 percent in each scrubber).
Plant shut down for boiler maintenance.
Scrubbers inspected on July 7, 1973.
A-l
-------
Table A-2. SUMMARY OF OPERATING DATA FOR 13 MONTHS
The following shows the monthly typical data for 13 months after the start-
Date recorded
Item recorded
April 10 May 18 June 16 July 15 Aug. 15
1972
Scrubber operated;A or B.
Gas stream
Flow to scrubber, NM3/Hr.W.G
% of total boiler gas flow
Temperature at;
Inlet/outlet of scrubber.
Stack
Press, drop at; mm.W.C
1st stage throat/mist elim.
2nd stage throat mist elim.
Slurry stream
Make-up slurry flow to: \. /Hr
1st stage/2nd stage
Recycle slurry flow to: M3/Hr
Tang, nozzles, lst/2nd
Bull nozzle, lst/2nH
Counter spraying,lst/2nd
A
A
349,000 384,060 302,000 267,000 267,000
68 75 59 52 52
146/50 153/53 1&7/54 155/55 162/55
99 106
145/9 176/9 125/10
110/3 15Q/5 10S/3
110/10
85/2
112
110/13*
85/3
.5/6.0 8.0/7.5 7.5/5,5 7.0/4.8 5.75/4.25
890/890 910/8UO 870/9.10 940/920 900/900
590/400 590/390 5.90/410 630/400 620/410
SOo concentration at:
Inlet of scrubber
Stack
PPM
12001 11 no/- 1180/
1200 1190 1200
1!>00 to 1800 to
2200 2009
770
1200/ .1220/
1200 1220
1800 to
2000
860
1800 to
2000
790
Note:(l) All above figures except gas flow are the indication i>n instruments/
analyzer. . (he readings of
Gas flows are calculated based on opening degree of damper located
at discharge side of B.U. Fan and gas flow from boiler being
constant at 512.000 NM3/Hr. wet gas basis.
(2) In operation during April, '72 thru. Oct. '72, the gas flow to scrubber
was varied in the ranges as indicated in the above table in order
to test the relationship between P.H. value of recycle slurry and
SO2 removal efficiency, etc. (Gas flow on design,basis of one
scrubber is 384.000 NM3/Hr. wet gas basis)
Reproduced as received from Mitsui Aluminum Company, Ltd.
A-2
-------
AT S09 CONTROL SYSTEM FOR MITSUI ALUMINUM
-up operation was performed in April, 1072
Unto: August, 19 73
Prepared by: Mitsui Aluminium
Sept. 12
A
384,000
75
161/55
104
173/13*
144/7
8.75/6.5
900/900
600/400
1200/
lino
1900 to
2350
Oct. 15
A
341,000
665
156/55
103
- 145/13*
122/5
10.5/8.0
900/900
600/420
1200/
1200
1900 to
2000
630
Nov. 16
B
384,000
75
154/54
104
150/12
155/6
7.5/7.25
900/900
590/400
1180/
1200
2000 to
~-2>300
630
Dec. 15
B
384,000
75
1-17/50
100
100/12
155/6
fl.75/8.5(
900/900
600/400
1200/
1200
2000 to
2300
650
1973
.Jan. 15
B
323,000
63
150/52
106
120'11--
122 '4
8.0/6.25
910/MO
610/420
1?00/
12:20
1800 to
2000
55(<
Feb. 15
B
384,000
75
157/51
104
155/13
150/5
8.25/6.25
920/900
610/420
1?00/
1220
2000 to
2300
535
Mar. 15
A
384,000
75
156/50
106
172/11
135/5
10.0/7.25
920/910
610/390
1?20/
1220
2000 to
2300
600
April 19
A
384,000
75
154/52
98
175/12
138/8
9.0/7.25
950/900
630/420
12?0/
1200
1900 to
2200
550
May 15
A
384,000
75
160/53
100
175/14
138/6
9.0/6.5
920/890
620/420
1200/
1200
2100 to
2300
700
(3) Pressure drops at 1st stage mist eliminator with mark * for July thru. Oct., 1972;
The instrument indicating^p at 1st stage M.E. showed 23 to 27 mm W.C. during operation
in July thru. Oct., 1972, but Mitsui Aluminium found out at the previous annual maintenance
shut, down in last October, that the element of this instrument was plugged by the scale and the
instrument did not indn-nte tho correct figures. Therefore, Mitsui Aluminium corrected these
figures to ones as indicated in the above table.
(4) | The plant was shut down once for annual-maintenance shut down starting from Oct. 17 to
Nov. 10, 1972 during the operation for 13 months.
A-3
-------
Table A-3. SUMMARY OF OPERATING DATA FOR TWO-SCRUBBER
so2 CONTROL SYSTEM AT MITSUI ALUMINUM a,b
Parameter
July 4, 1973
Scrubber
July 5, 1973
Scrubber
Gas stream
Gas flow, Nm3/hrc
Gas temp, at inlet/outlet, °C
--, at stack, °C
Pressure drop, mm W.G.
First throat
First mi§t eliminator
Second throat
Second mist eliminator
Slurry stream
Make-up slurry, m /hr
First stage for 2 scrubbers
Second stage for 2 scrubbers
Recycle slurry, m /hr
Bull nozzles, lst/2nd
Tang, nozzles, lst/2nd
Counter spray, 1st/2nd
S0» concentration, ppm
Inlet of scrubber
Outlet of scrubber
Stack
236,000 236,000
158/55
94
82 to 85 65 to 70
8 to 9 7 to 8
61 to 65 75 to 78
3 2.5 to 3.0
10.5 to 11.5 in total
7.5 to 8.5 in total
500/400 500/360
800/800 800/820
790/760 790/790
1900 to 2140
210 to 310
270 to 540
236,000 236,000
158/55
96
80. 'to 85
8 to 9
60 to 65
3
67 to 70
7
75
2-3
10.0 to 12.5 in total
8.0 to 9.6 in total
500/400 500/360
800/800 820/810
790/760 800/790
1980 to 2100
180 to 280
230 to 540
Prepared by Mitsui Aluminum, August 1973.
Two scrubbers in parallel were operated together on July 4 and July 5, 1973,
prior to shutdown of the plant. This summary presents operating data for the
two-scrubber operation.
Gas flow was not measured by using gas flow meter and was calculated based on
the opening degree of damper located at the discharge side of B.U. Fan. The
gas flow from boiler was fed to 2 scrubbers at same ratio of boiler gas flow
by controlling the opening degree of dampers located at inlet of each scrubber.
A-4
-------
Table A-4. MITSUI PERFORMANCE TEST DATA3
SO, concentration SO, analysis
No.
slurry
pH bleed flow MakeuP Reheat Corrected Statutory
r slurry furnace scrubber requirements, %
Inlet, Outlet, Inlet, Outlet, 1st 2nd 1st, 2nd, density, heavy oil outlet SQ2 Efficiency, %
Time Methodb ppm ppm pg^n ppm ra3/hr m3/hr g/cm3 cone., ppm Manual Analytical Manual Analytical
1
2
3
4
5
6
7
8
9
10:25
to
10:45
11:16
to
11:36
12:39
to
12:59
14:20
to
14:50
21:30
to
21:45
21:55
to
22:15
22:30
to
22:50
23:25
to
23:45
23:45
to
0:10
I
N
I
N
R
I
R
I
R
1402 230 1830 230
1754 305 1780 230
1411 270 1850 2*50
1815 307 1800 270
1599 204 1780 155
1369 155 1790 145
1687 193 1810 145
1415 174 1850 150
1663 141 1840 150
7.1 7.65 7.0 6.6 1.074 540 203
6.9 8.3 6.8 6.5 1.076 540 278
.
6.8 8.2, 6.9 6.7 1.076 540 243
6.75 7.9 6.9 6.75 1.076 540 280
7.65 7.8 10.7 5.8 1.081 540 177
7.7 7.9 10.7 S.8 1.080 540 128
7.6 7.8 10.7 5.8 1.080 541 166
7.6 7.95 10.7 5.8 1.080 540 146
7.6 7.8 10.7 5.8 1.080 540 114
85.5 87.4 106.8 81.9
84.1 86.7 89.3 90.5
82.8 86.5 111 84.8
84.6 85.0 86.7 87.3
88.9 91.3 125.5 112.7
90.7 91.7 146.5 112.2
90.2 92.0 119.2 111.0
89.6 91.9 141.2 108
93.1 91.8 122.5 109.2
>
I
aWitnessed by Chemico USA and Mitsui Aluminum and recorded by Mitsui Miike Machinery. These data of May 11, 1972, were accepted by Mitsui Aluminum
to satisfy Chemico system guarantee for S02 removal and sludge stoichiometry requirement.
Methods ; I = iodine by JIS; N = neutralization method; R = Reich. S02 analysis: inlet, electrical conductivity method; outlet, infrared method.
Analyses: S content of heavy oil = 0.026S; density of heavy oil = 0.95; purity of carbide sludge = 0.92.
-------
APPENDIX B.
DETAILED REPORT ON TROUBLE
IN ASH POND WATER PUMPS
-------
Date: August 1973
Prepared by: Mitsui Aluminum
Translated by Chemico Tokyo
The following is the detailed report on the trouble in ash pond water pumps
experienced in June 1972, which resulted in increase in gas pressure drop
at the mist eliminator of 1st stage. The water transferred by the ash pond
water pump is used for washing of mist eliminator.
June 9: At 17:00, a little vibration of pump was found out, but the
pump was in operation.
June 10: At 05:15, water flow from pump was not constant, but the
pump was still in operation.
June 11: At 01:15, it was found out that the pump was in operation
with water flow of zero and the pump was shut down at 01:30.
At 01:20, the supply source of the water was shifted to
industrial water from ash pond.
June 13: After the pump was repaired, the supply source of water was
shifted to ash pond from industrial water again at 07:30.
During operation of pump, the wire-netting with opening size
of 5mm x 5mm was installed at suction of pump.
At 19:05, it was found out that the water flow from pump
was being down and the pump was shut down. The supply
source of water was shifted to industrial water again.
June 14: The pump pit was cleaned.
June 15 through June 17:
The operation of pump was re-started at 12:00 on June 15, but
there was still trouble in pumps including pump for stand-by.
During these 3 days, it was tried again and again to operate
the pump after repairs, but the pumps were not workable. The
industrial water was used for washing of mist eliminator.
June 18: After repairs, the operation of pump was successfully restarted
and the ash pond water was supplied by the pump.
June 20: The washing procedure was modified to allow for a longer
intermittent washing of each section.
June 28: To prevent the further trouble in pumps, a new strainer was
installed on the self water bearing lines on [sic] located on pumps,
The following shows the change of water pressure and gas pressure drop in/
during the above operation with troubles in ash pond water pumps.
B-l
-------
1) Pressure of water used for washing of 1st stage mist eliminator
at inlet of scrubber;
Pressure of ash pond
water in normal operation:
Pressure of ash pond water
pump during operation with
trouble in pumps:
Pressure of industrial
water:
2.0kg/cm2G.
kg 2
Decreased to 1.4 6/cm G.
0.5 to 0.7 kg/cm2G.
2) Gas pressure drop at throats and mist eliminators;
scrubber operated: scrubber A.
Date
June 8
9
10
11
12
13
14
15 .
16
17
18
19
23
24
Gas flow in
% of boiler
gas flow
56.2
56.2
56.2
56.2
56.2
56.2
56.2
59.0
59.0
59.0
56.2
56.2
56.2
56.2
Gas pressure drop, mm W.G.
Ist-stage
throat
120
120
120
120
118
120
120
125
'125
125
120
120
120
120
Ist-stage
mist eliminator
8
8
9
8
9
9
9
10
11
11
10
10
10
10
2nd-stage
throat
98
95
100
95
100
98
100
105
105
105
95
100
100
100
2nd-stage
mist eliminator
2.5
3
3
3
3
3
3
3
4
4
3
3
3
3
As you may find in the above table, the pressure drop at mist eliminator of 1st
stage was increased to 10 mm W.G. from 8 mm W.G., whereas there was no change of
pressure drop at mist eliminator of 2nd stage.
B-2
-------
APPENDIX C
SUMMARY OF SIGNIFICANT LOAD VARIATIONS
-------
On seven specific occasions, the boiler was shut down or its output
decreased. A summary of variations in load for each of these occasions
is given below.
1. On October 16, 1972, the boiler was shut down for its annual
inspection and maintenance as follows:
Inlet S02 Stack S02
Total boiler flue concentration, concentration,
Time gas to scrubber, % ppm ppm
8:00 am
9:00 am
4:00 pm
5:00 pm
6:00 pm
73
50
30
--
0
2040
2130
2000
1850
1780
740
760
cut off
--
--
Because this was the first time the scrubber was shut down, it was done
cautiously over a 10-hour period. The pH was kept stable throughout the
shutdown by manual adjustment of the carbide sludge feed rate.
2. On November 11, 1972, the boiler was first brought up to its design
rate of 156 MW after completion of the annual maintenance and then
the scrubbing system was started. Several hours were taken to bring
the scrubber on line but specific details are not available.
3. On March 27, 1973, the boiler load was decreased from 156 MW to 95 MW
and then increased back up to 156 MW in order to repair a boiler super-
heater spray valve. During this time, the scrubber was adjusted as
follows:
Slurry makeup
Flue gas being rate. m-Vhr S02 concentration, ppm
Time
9:00 am
10:00 am
3:00 pm
treated, %
75
55
75
1st stage
10.75
7.25
10.25
2nd stage
8.25
5.25
7.50
Inlet
2100
2040
2080
Stack
380
380
370
Approximately 30 minutes were taken to decrease the amount of flue gas
being treated from 75 percent to 55 percent and about 30 minutes to in-
crease it from 55 percent to 75 percent.
C-l
-------
4. On May 30, 1973, the power generator load was reduced from 156 MW to
135 MW and then increased to 156 MW as follows in order to disconnect
a supply line between Mitsui Aluminum and Kyushu Electric:
a. At 8:30 the load began decreasing from 156 MW.
b. At 8:52 the load had been decreasing to 135 MW.
c. At 10:40 the load began increasing from 135 MW.
d. At 11:00 the full load of 156 MW was restored.
During this time, the only change that was made to the scrubbing
system was to decrease the pH by manually adjusting the makeup
feed rates of carbide sludge to the first and second stage of
the scrubber as follows:
Time
8:00 am
9:00 am
10:00 am
11:00 am
12:00 n
Fresh lime
m3/h
1st stage
9.30
6.50
6.80
7.75
9.30
feed rate,
r
2nd stage
7.50
4.50
5.30
6.75
8.40
SC>2 concentration, ppm
Inlet
2080
2070
2070
2060
2070
Stack
625
660
700
710
685
5. On May 31, 1973, the boiler load was reduced from 156 MW to 125 MW in
about 45 minutes and then brought back up to about 156 MW in stages over
a 2-hour period. No changes were made to the scrubbing system other
than to adjust the lime makeup rate manually as follows:
Makeup lime rate, m /hr S02 concentration, ppm
Time
6:00 pm
7:00 pm
8:00 pm
9:00 pm
10:00 pm
11:00 pm
1st stage
10.0
9.25
9.00
9.50
9.75
9.75
2nd stage
7.50
6.50
6.50
4.25
7.50
7.50
Inlet
2350
2150
2150
2200
2210
2250
btack
620
550
465
490
530
600
6. On June 12, 1973, module B was placed in service along with module A.
Prior to this time, scrubber A was handling 75 percent of the boiler
flue gases and scrubber B was handling 0 percent. After the change-
over, each module was handling about 46 percent of the boiler flue
gases. Details of the changeover are not available.
C-2
-------
7. On July 5, 1973, the boiler was shut down while maintenance could be
performed in the aluminum smelter. Prior to shutting down the boiler,
both modules were shut down as follows:
Gas volume being
Time treated, V
10:00 pm 92
10:10 pm 76 : '
10:20 pm 76
10:30 pm 55
10:40 pm 55
10:50 pm 30
11:00 pro 30
11:10 pm 15
11:14 pm Shut down reheat furnace
11:20 pm 15
11:27 pm 0
While the scrubbing modules were being shut down, the carbide sludge
makeup to the scrubbing system was manually decreased as follows:
Lime makeup, m3/hr S02 concentration, ppm
Time
9:50 pm
10:15 pm
10:35 pm
10:55 pm
11:15 pm
11:30 pm
1st stage
10.75
9.0
7.75,
6.25
5.0
stop
2nd stage
9.0
8.0
6.75
6.0
5.0
stop
Inlet
2005
1980
2000
2000
2020
--
Scrubber outlet
290
260
230
160
70
--
During the shutdown, the following observations were made:
a. The pressure drop across the scrubber gradually decreased.
b. The stack temperature gradually increased.
c. The scrubber outlet SCL concentration decreased.
d. The recycle slurry rate was constant.
e. The pH was controlled to within ±0.2 of the desired pH.
C-3
-------
The pressure drop across the scrubbing system at 9:50 pm just before the
shutdown was as follows:
Pressure drop, mm
Location Module A Module B
First-stage throat 90 66
First-stage demister 8 6.5
Second-stage throat 65 77
Second-stage demister 2 2
Total 165 151.5
C-4
-------
APPENDIX D.
RELATIONSHIP BETWEEN ASH POND WATER
AND LEAKAGE WATER
(Copy of Report Prepared by Mitsui Aluminum
with Translation by Chemico Tokyo)
-------
by
Co.
To
H5 tO 4-8^- 41-133
— 10:00
r i
pH8.15
SS 6PM
COD328
D-l
-------
AncJyi\C(Jl
-Hi*
2 tf/7°)l' !
Bo
6th
3. t > 7° j / 7"«. (Se»«fJli«i
7 lil^ 4 .?.»
(D
*->• 48* 2. 7
r.:3|
D-2
-------
4.
*^A<
*xJH
pH
W/w
Ca(ppw)
MI( " )
Fe C " )
OK - )
SQ4( - )
/^Vl V O*\9\
2/& 14'- 20'
83
4680
475 ,
68
o.54
880
1.306
A-t-tU -iro»4 •$
24 14'- 20'
.6.7
47.530
333
1.140
' I31
15,725
2.374
(3)**^"
2/ I4.°"20'
6.8
45:450
.326
1.094
OGS
14475
2,278
it »j p/xu>oci(
>e J-'T 3«v«et!
^ 8'- 45'
US^h .x
29000
358
HT: 665
0.55
8.575
1.593
®^VSI
2/7 |4'- 40'
8.°
48.610
360
1.16:3
I.32
16.600
2471 ;
5.
> 9"
D-3
-------
m ;
. 48. 1 , 6
To
©
D-4
-------
4
Ss<\
70
G
8
8.°- 82
5.140
46.750
43.800
45000
470
333
338
350-360
MjCv)
GO
061
883
ipoo~i?oo
0
38
O.o9
.2~ o.4-
0? ( v)
819
15600
12870
17.000 -I8MO
.576
2.350
2.173
5
/js* ,1, <
r^.«.s«>n
is ci*£ -to
PH
7"'J - K X)^
At
-to
-Hie ponoL
D-5
-------
2. ^>r/
3. t77''
13-4-5'
_
O. •*/• ^-
iP P
D-6
-------
4.
SA^
XU fKc -Trout o(4Ae t*c.io*y -Sx^ i^A^jt-r
l&c:
.
pH
5.V
'G.
e
4250
472
73
41 GOO
_ G.65
34950
45000^46000
—: :—.j
336
352
1.072
9/6
350-360
1.000^-1.200
O.
'2
(363
15.000
12.750
SQ*'( • ) 1.6 60
2387
2.216
2300^2400
L, T \
T,
a-
-4c kg, -tU
not -to W.
-b^e.
AS
D-7
-------
2.
f ;
4-7. /$. 2
D-8
-------
4
-*?S
,v
.
pri
5.
8
3090
3800
24000
dCpp-iii)
342
400
484
386.8
• Mg(Pfon)
530.
io
••CL.('Pfw)'
400
3825
475 '
6350
1737
1979
C JK
L
PH tf" £Jt;€fl
D-9
-------
Conclusion:
The leakage of water from the shore at low tide seems to be
the returning of the sea water which penetrated once into the shore
during high tide. This is a natural phenomenon usually observed
so often at the sea shore o~ reclaimed land. Therefore, the said
leakage must not be the water penetrated from the ash pond.
(For reference, the pond water keeps always the same level.).
In addition, the higher figures of pH value, conductivity, Mg, Cl,
etc. of Sample 2 and 3 compared with those of Sample 1 as shown
f
in the^above Table indicate that there is no leakge of the water from
the pond.
The reason why pH value of the leaked water in question af-the front
sea of the plywood factory is lower than pH value of the sea water
in Miike port might be because Sulfate ion contained in refuse coal
which were used for reclamation of the plywood factory area decrease
pH value of the sea water penetrated into the reclaimed land during
high tide in Ariake Bay.
D-10
-------
APPENDIX E.
CLIMATIC DATA FOR OMUTA AREA
-------
Table E-l. MONTHLY CLIMATIC DATA FOR THE CITY OF OMUTAa
Rainfall,
Date
Dec.
Jan.
Feb.
Mar.
Apr.
trt
M May
June
July
Aug.
Sept
Oct.
Nov.
1971
1972
1972
1972
1972
1972
1972
1972
1972
1972
1972
1972
Total
51
166
125
102
176
166
477
802
295
108 .
99
89
1st 1
5
58
68
17
31
55
54
564
44
39
28
64
3 2nd
32
16
35
61
74
79
136
204
14
63
43
19
3rd
14
92
22
24
71
26
287
35
237
6
28
6
mm
Temperature (max.), C
Temperature (min.),
C Relative humidity
(mean
value) , %
No., of rainy days
1 mm
8
13
13
10 .
15
15
15
16
11
10
10
15
10 mm
1,
4
7 •
4
6
6 •
9
9
8
4
6
2
_30 mm Avera£e
1
- 3
1
1
2
1
5
8
4
1
0
1
9.5
12.4
11.2
15.0
19.3
25.1
26. S
30.5
30.8
27.7
23.1
17.6
1st
8.5
12.2
12.3
10.9
16.0
C3. 5
25.8
28.4
31.8
30.6
25.6
21.6
2nd
12.8
1*2.2
12.2
15.7
21.4
22.3
J6.9
50.5
31.7
27.4
23. A
IS. 6
3rd
7.5
12.9
9.0
18.2
20. S
23.4
26.9
32/4
29.0
24.9
20.6
12.4
Average 1st
1.2
3.4
4.2
6.2
10.5
15.2
19.1
24.2
24.0
19.7
14.0
9.1
i-i
0.2
3.1
6.1
2.3
7.6
IS. 2
17.7
24.2
25.2
23.4
14.7
11.9
2nd
1.1
4.2
4.5
6.6
11.6
15.5
20.1
23.1
24. S
19.6
14.6
9.4
3rd
2.2
3.0
1.9
9.2
12.2
14.9
19.4
25. 3
22.4
16.2
12.9
6.0
Average
72
75
73
70
69
68
73
78
73
70
71
74
1st
74
74
76
73
70
71
70
80
71
74
65
78
2nd
69
77
75
71
70
68
74
77
70
73
75
74
3rd
73
74
69
69
69
65
77
69
79
64
73
71
a Report of the Edison Electric Institute Study Group on S02 Removal Processes in Japanese Plants, Edison Electric Institute, New York, New York,
July 1973. p. 48.
1st ten days ci the month.
2nd: 2nd ten days of the month.
3rd: The rest of the month.
-------
Table E-2. RAINFALL FOR JUNE 18 THROUGH JULY 7, 1973,
IN OMUTA
Date
June 18
19
20
21
22
23
24
25
26
27
Rainfall, inches Date
0.22 June 28
0.08 29
0.04 30
July 1
0.24 2
3
4
-- ' 5
2.30 6
2.89 7
Rainfall, inches
--
1.88
0.47
0.02
--
0.23
0.11^
0.56^
--
.
Fell during night of July 4, 1973.
E-2
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/2-73-007
2.
4. TITLE AND SUBTITLE
Operation and Performance
of the Lime Scrubbing System
at Mitsui Aluminum Company. Ltd.
7. AUTHOR(S)
James A. McCarthy and James F. Durham
9. PERFORMING ORG \NIZATION NAME AND ADDRESS
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Emission Standards and Engineering Division
Research Triangle Park, North Carolina 27711
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December 1973
6. PERFORMfNG ORGANIZATION
8. PERFORMING ORGANIZATION
CODE
REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Presented at National Enforcement Hearings, November 1973, Washington, D. C.
16. ABSTRACT
This is a report describing the^design and operation of the Mitsui Aluminum
Company's S02 scrubber at Omuta, Japan. The report is a factual presentation
of the information gathered by James A. McCarthy and James F. Durham of the
U. S. EPA, and Howard E. Hesketh of Southern Illinois University during their
visit to the plant in July of 1973.
17.
a. DESCRIPTORS
Lime Scrubbing
Air Pollution Control
Surfur Dioxide
Sulfur Dioxide Control
Emissions
Power Plants
Boilers
KEY WORDS AND DOCUMENT ANALYSIS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Flue-Gas Desul-
furization
18. DISTRIBUTION STATEMENT 19. SECURITY CLASS (This Report) 21. NO. OF PAGES
Release unlimited. Available from
Performing Organization, Item 9. 20. SECURITY CLAS
S(TMspage) 22. PRICE
EPA Form 2220-1 (9-73)
------- |