&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-245
November 1979
University of Washington
Electrostatic Scrubber
Tests: Combined
Particulate and S02
Control
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
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 sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses 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 environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-245
November 1979
University of Washington
Electrostatic Scrubber Tests:
Combined Paniculate and 862 Control
by
Michael J. Pilat
University of Washington
Department of Civil Engineering
Seattle, Washington 98195
Grant No. R806035
Program Element No. EHE624A
EPA Project Officer: Dale L. Harmon
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
Abstract
o
A 1700 am /hr (lOOOacfm) University of Washington Electrostatic Spray
Scrubber pilot plant was tested on the coal-fired boiler unit no. 2 at
Centralia Power Plant to demonstrate its effectiveness for controlling
fine particle and sulfur dioxide emissions. The multiple pass, portable
pilot plant operates by combining oppositely charged aerosol particles
and water droplets in two spray towers. Aerosol charging sections at
a negative polarity precede each spray tower. The pilot plant was
operated utilizing only one corona section and one spray tower. A liquor .
recycle system was constructed, giving the pilot plant the capability
to operate in an open-loop or closed-loop mode. All sulfur dioxide
tests were run in an open-loop operating mode using either water or
NapCO- solution as a scrubbing liquor.
Simultaneous inlet and outlet source tests using the UW mark 10 and
Mark 20 Cascade Impactors provided size dependent and overall mass basis
particle collection efficiency data. Measured overall particle collection
efficiencies ranged from 98.99% to 99.80% depending upon scrubbing
operating conditions. Particle mass concentrations measured at the scrubber
outlet ranged from 0.0074 gm/sdm (0.00324 gr/sdcf) to 0.0015 gm/sdm
(0.00065 gr/sdcf).
Sulfur dioxide concentrations at the inlet and outlet of the pilot plant
were .measured with a Thermo Electron Model 40 Sulfur Dioxide Analyzer.
Sulfur dioxide collection efficiencies ranged from 8.02% to 97.41%
depending on the scrubber operating conditions, inlet sulfur dioxide
concentration and the type of scrubbing liquor used.
-------
TABLE OF CONTENTS
Page
I. SUMMARY AND CONCLUSIONS 1
II. RECOMMENDATIONS 2
III. RESEARCH OBJECTIVES 3
IV. DESCRIPTION OF SOURCE 4
V. DESCRIPTION OF UW ELECTROSTATIC SCRUBBER APPARATUS 8
A. Review of Previous Work 8
B. Description of Overall System 8
C. Cooling Tower 10
D. Particle Charging Corona Section 10
E. Water Spray Towers 12
F. Mist Eliminator 15
G. Test Ducts 15
H. Fans 15
I. High Voltage Power Supplies 17
J. Liquor Control Panel in the Scrubber Trailer 17
K. Purge Air Heating System 18
VI. DESCRIPTION OF THE LIQUOR RECYCLE SYSTEM 20
A. Description of Overall System 20
B. Liquor Tanks 20
VII. EXPERIMENTAL PROCEDURES AND TEST EQUIPMENT 23
A. UW Mark 10-20 Cascade Impactor Trains 23
B. TECO Model 40 S02 Analyzer 23
VIII. PARTICULATE COLLECTION EFFICIENCY RESULTS 26
A. General Test Description 26
B. Particulate Collection Efficiency 26
IX. SULFUR DIOXIDE COLLECTION EFFICIENCY RESULTS 39
A. General Test Descriptions 39
B. S02 Collection Efficiency 39
1. Results Using Water as a Scrubbing Liquor 39
2. Results Using Na^CO., Solution as a Scrubbing Liquor 44
X. REFERENCES 50
IV
-------
LIST OF FIGURES
Page
IV-1 Photograph of UW Electrostatic Spray Scrubber Located at 5
Boiler Unit ND. 2, Centralia Power Plant
IV-2 Photograph of Liquor Recycle Trailer Location at Centralia 6
Power Plant
IV-3 Schematic of Ducting Arrangement to UW Electrostatic Spray 7
Scrubber Pilot Plant at Centralia Power Station
V-l General Layout of Electrostatic Scrubber Pilot Plant 9
V-2 Cooling Tower Schematic 11
V-3 Particle Charging Corona Section 13
V-4 Collection Plate Flushing System 14
V-5 Spray Header and Nozzle Arrangement Typical to Spray Towers 16
#1 and #2
V-6 Heated Purge Air System 19
VI-1 General Layout of the Liquor Recycle System 21
VII-1 UW Mark 10-20 Cascade Impactor Sampling System Schematic 25
VIII-1 Particle Collection Efficiency and Penetration vs. Particle 28
Size for Impactor Tests 1-4 and 7-8 (Log-Normal Approximation)
VIII-2 Particle Collection Efficiency and Penetration vs. Particle 29
Size for Impactor Tests 2 and 3 (Log-Normal Approximation)
VIII-3 Particle Collection Efficiency and Penetration vs. Particle 30
Size for Impactor Tests 2 and 3 (Parabolic curve fit)
VIII-4 Particle Mass Distribution vs. Particle Size for Impactor 32
Test 2
VIII-5 Particle Mass Distribution vs. Particle Size for Impactor 33
Test 3
VIII-6 Inlet and Outlet Particle Size Distributions for Impactor 34
Tests 1-4 (Log-Normal Approximation)
VIII-7 Inlet and Outlet Particle Size Distributions for Impactor 35
Tests 7-8 (Log-Normal Approximation)
-------
List of Figures (Con't.)
Page
VIII-8 Inlet and Outlet Particle Size Distributions for Impactor 36
Test 3 (Log-Normal Approximation and Actual Data)
VIII-9 Particle Mass Concentration for Particles Less Than Stated 37
Diameter for,Impactor Tests 1-4
VIII-10 Particle Mass Concentration for Particles Less Than Stated 38
Diameter for Impactor Tests 7-8
IX-1 Liquor-to-Gas Ratio vs. SCL Collection Efficiency for Tests 41
Using Water as a Scrubbing Liquor
IX-2 Inlet SCL Concentration vs. SCL Collection Efficiency for 42
Tests Using Water as a Scrubbing Liquor
IX-3 Change in SCL Collection Efficiency vs. Spray Voltage for 43
Tests Using water as a Scrubbing Liquor
IX-4 Liquor-to-Gas Ratio vs. SCL Collection Efficiency for Tests 47
Using NapCCL Solution as a Scrubbing Liquor
IX-5 Inlet S02 Concentration vs. SCL Collection Efficiency at 48
Constant Stoichiometric Ratios
IX-6 Stoichiometric Ratio vs. SCL Collection Efficiency at 49
varying L/G and Spray Voltage
-------
Table V-l
Table VII-1
Table VIII-1
Table IX-1
Table IX-2
LIST OF TABLES
High Voltage Power Supply Units
Source Test Parameters and Measurement Techniques
Results of Cascade Impactor Tests 1-4 and 7-8 at
Centralia Power Plant
Results of S02 Tests at Centralia Power Plant Using
Water as a Scrubbing Liquor
Results of S(L Tests at Centralia Power Plant Using
Na2C03 Solution as a Scrubbing Liquor
Page
17
24
27
40
45
vii
-------
ABBREVIATIONS AND SYMBOLS USED
acfm Actual cubic feet per minute
akm Actual 1,000 cubic meters
C Cunningham slip correction factor
dj-Q Particle aerodynamic cut diameter of cascade impactor jet stage
D. Diameter of jet in cascade impactor stage
\J
fps Feet per second
gm/sdm Grams per standard dry cubic meter
gpm Gallons per minute
KV Kilovolts
i Liters
L/G Liquid to gas flow rate ratio
MPa Pressure 1,000,000 pascals
psig Pounds per square inch
SCA Specific collection plate area
sm Standard cubic meter
SR Stoichiometric ratio (moles of sodium carbonate per mole of inlet
sulfur dioxide)
V, Gas velocity in jet
J
W Watts
GREEK SYMBOLS
y Gas viscosity
y Micron
TCQ Inertial impaction parameter at 50% collection efficiency for
particles of aerodynamic diameter dr«
vm
-------
ACKNOWLEDGEMENTS
The assistance, cooperation, and advice of our Project Officer, Dale L.
Harmon, Chemical Engineer, Particle Technology Branch, Industrial
Environmental Research Laboratory at the Environmental Protection Agency
is greatly appreciated. The cooperation of Ted Phillips, Tom White,
Pete Steinbrenner, and Bob Werner of the Pacific Power & Light Company
was very helpful. The assistance of University of Washington staff and
students including Terrell Gault, Gary Raemhild, Tracey Steig, Phil
Ayers, Steve Plotkin, Eric Piehl, and Jim Wilder is acknowledged.
Terrell Gault wrote his MSE thesis on sulfur dioxide collection with
the UW Electrostatic Scrubber. After receiving his MSE degree, Terrell
served as Project Engineer doing much of the combined sulfur dioxide
and particulate collection study at the Centralia Power Plant.
ix
-------
Section I
SUMMARY AND CONCLUSIONS
The University of Washington Electrostatic Spray Scrubber portable pilot
plant was tested at the Pacific Power and Light Steam Generating Plant
in Centralia, Washington on coal fired boiler no. 2 to demonstrate its
effectiveness in simultaneous control of particulate and sulfur dioxide
emissions. The pilot plant consists of a cooling tower, two corona sections
which charge the particles to a negative polarity, two spray towers into
which positively charged water droplets are sprayed and an electrostatic
mist eliminator. A liquor recycle system, that gives the capability of
open-loop or closed-loop modes has four tanks and two pumps.
The unit was operated using only one corona section and one spray tower
and the liquor system in an open loop mode. Overall particle collection
efficiencies ranged from 98.99% to 99.80% depending upon the scrubber
operating conditions. Using UW Mark 10 and Mark 20 Cascade Impactors,
particle size distributions from O.OSy to 30. y aerodynamic diameter were
measured at the inlet and outlet of the electrostatic scrubber.
Inlet and outlet sulfur dioxide concentrations were measured with a Thermo
Electron Model 40 S02 Analyzer. SO^ collection efficiencies ranged from
8.0% to 97.4% for various operating conditions. The efficiency was seen
to increase with an increase in spray voltage, liquor-to-gas ratio,
stoichiometric ratio, and inlet SOp concentration.
In conclusion it appears that the UW Electrostatic Spray Scrubber can
effectively collect particulate emissions from a coal fired boiler with
relatively low water usage and low corona section plate area. Further,
enhanced SO, collection efficiencies with electrostatic charging of the
scrubbing liquor implies that less alkaline material would be needed
with a charged scrubbing system for the same S02 collection efficiency.
-------
Section II
RECOMMENDATIONS
To better evaluate the effectiveness of the UW Electrostatic Scrubber for
control of participate and S02 emissions, it is recommended that additional
testing should be done with tne closed loop liquor recycle system using
lime scrubbing liquor. Extended field tests of 24 hours or more are
recommended to test the chemical and flow characteristics of the effluent
sludge, to check the capability of the pilot plant to electrostatically
charge liquor droplets that contain a high level of solids, and to optimize
operating conditions for minimum scaling.
After this year's testing of the scrubber at a coal-fired boiler, additional
testing at field sites is recommended to optimize the design and operating
parameters of the pilot plant for simultaneous control of particulate and
SOp emissions. A field source with a continuous high level of sulfur dioxide
would result in greater gas concentrations at the scrubber outlet. A higher
S02 concentration would allow a broader range of scrubber operating conditions
to be tested and increase the statistical reliability of the data.
-------
Section III
RESEARCH OBJECTIVES
The objectives of the research performed under the auspices of Environ-
mental Protection Agency Research Grant Number R803278-01 were to:
1. Demonstrate the effectiveness of the University of Washington
Electrostatic Scrubber for simultaneously controlling the
emissions of sulfur oxide and particulates emitted from coal-
fired power plants.
2. Determine the effect of electrostatic charging of the scrubbing
liquor on the sulfur oxide absorption rate and collection
efficiency.
3. Use the 1,700 m3/hr (1000 acfm) portable pilot plant of the UW
Electrostatic Scrubber to obtain the data needed to design
larger control systems.
-------
Section IV
DESCRIPTION OF THE SOURCE
The Centralia Steam-Electric Project located near Centralia, Washington
is a coal-fired electric power generating station owned by eight Northwest
utilities and operated by Pacific Power and Light Company personnel. The
plant contains two units which have a combined generating capacity of
1,330,000 kilowatts of electricity (665,000 kilowatts per unit). The two
boilers, manufactured by Combustion Engineering, are pulverized coal fired
type, each with a designed steam rate of 2,400,000 kilograms (5,200,000
pounds) per hour at 16.6 MPa (2400 psig) at the turbine inlet. The two
turbine generators were manufactured by Westinghouse Electric Corporation,
each with a guaranteed rating of 664,898 kilowatts. The primary fuel is
sub-bituminous C coal which comes from a strip mine adjacent to the plant.
The coal has a design heating value of 18,100 joules/gm.(8100 BTU's per
pound). Fly ash particulate emission is controlled by two electrostatic
precipitators.
In October 1977, the U of W Electrostatic Spray Scrubber was transported
to the Centralia Power Plant. The sample gas stream (approximately 1500
acfm) was tapped from the outlet of boiler unit number 2. A .3m (12 inch)
sampling scoop was installed (facing upstream) at the center point of the
transition duct between the air preheater and the precipitator. .25m (10
inch) diameter aluminum ducting connects the sampling scoop to the scrubber,
which is located approximately 18 m (60 feet) below on the ground level.
Due to the high negative static pressure in the main duct, a Dayton centri-
fugal blower was installed at the scrubber inlet to boost the air flow
capabilities. Fig.IV-1 shows the location of the scrubber trailer (on
the right) and the laboratory trailer (on the left). The liquor recycle
trailer is located at the back end of the scrubber trailer as shown in
Fig. IV-2. A schematic of the ducting arrangement showing the lengths of
duct from the sampling scoop to the inlet of the pilot plant is shown in
Fig. IV-3.
-------
Fig. IV-1. Photograph of UW Electrostatic Spray Scrubber Located
at Boiler Unit No. 2, Centralia Power Plant
-------
Fig. IV-2. Photograph of Liquor Recycle Trailer at Centralia
Power Plant
-------
ftlR.
m
18.0m
MATERIAL:
10" £>//}.
DUCT/^a
TOLERANCE:
.X ±
.XX ±
GAS FL0W
.XXX
UNIVERSITY OF WASHINGTON
DEPARTMENT OF CIVIL ENGINEERING
DRAWING NO.
DRAWING BY:
APPROVED BY:
SCALE:
DATE:
DATE:
Fig. IV-3. Schematic of Ducting Arrangement to UW Electrostatic Spray
Scrubber Pilot Plant at Centralia Power Station
-------
Section V
DESCRIPTION OF UW ELECTROSTATIC SCRUBBER APPARATUS
A. Review of Previous Work
Penney (1944) patented an electrified liquid spray test precipitator
involving particle charging by corona discharge and droplet charging
by either ion impaction or induction. Penney's system consisted of
a spray scrubber with electrostatically charged water droplets col-
lecting aerosol particles charged to the opposite polarity. Kraemer
and Johnstone (1955) reported theoretically calculated single droplet
(50 micron diameter droplet charged negatively to 5,000 volts) col-
lection efficiencies of 332,000% for 0.05 micron diameter particles
(4 electron unit positive charges per particle). Pilat, Jaasund, and
Sparks (1974) reported on theoretical calculation results and labora-
tory tests with an electrostatic spray scrubber apparatus. Pilat
(1975) reported on field testing during 1973-74 with a 1,700 am3/hr
(1000 acfm) UW Electrostatic Scrubber (Mark IP model) funded by the
Northwest Pulp and Paper Association. Pilat and Meyer (1976) reported
on the design and testing of a newer 1700 am3/hr (1000 acfm) UW Elec-
trostatic Scrubber (Mark 2P model) portable pilot plant. Pilat,
Raemhild, and Prem (1978) reported on tests of the UW Electrostatic
Scrubber at a steel plant. Pilat and Raemhild (1978) reported on
tests of the UW Electrostatic Scrubber at a coal-fired plant. The
UW Electrostatic Scrubber (patent pending) has been licensed to the
Pollution Control Systems Corporation (of Renton and Seattle, Washington)
for production and sales.
B. Description and Overall System
The major components of the pilot plant include a gas cooling tower,
an inlet and outlet test duct, two particle charging corona sections,
two charged water droplet spray towers, and a mist eliminator. Auxil-
iary equipment includes transition ductwork between major components
and a fan. The pilot plant is housed in a 12.2 m (40 feet) long
trailer and can be transported easily to different emission sources.
The general layout of the pilot plant is shown in Fig. V-l. Incoming
gases enter the top of the trailer to be treated in the. vertical gas
cooling tower and then turn vertically upward to enter the inlet test
duct. After moving down through the inlet test duct, the gases enter
the first of three horizontal passes.
The first pass contains both particle charging corona sections and the
first of two water spray towers. The two coronas are at either end of
this pass and are separated by spray tower #1. Spray tower #2 comprises
the entire second horizontal pass and the last (third) pass contains
the mist eliminator. Due to the extremely high collection efficiencies
when both coronas and towers were used, during this testing only corona
#2 and spray tower #2 were used.
8
-------
INCOMING
GASES
COOLING
TOWER
INLET TEST DUCT
SPRAY TOWER NO. 2
CORONA NO I
EXHAUST
GASES
SECTION A-A
CROSS SECTIONAL VIEW OF
THREE PASS HORIZONTAL SECTION
OUTLET TEST DUCT.
Mist
Eliminator
SPRAY TOWER NO. 2
SPRAY TOWER NO. I
CORONA NO. 2
ELEVATION VIEW
Fig. V-l. General Layout of Electrostatic Scrubber Pilot Plant
-------
At the outlet of the third horizontal pass, the gases enter the top of
the outlet test duct and are then directed to the fan before being
exhausted through the trailer roof.
C. Cooling Tower
The cooling tower is designed to lower the gas temperature to below
1210C (250°F) in order to maintain structural integrity of the system,
which is constructed of steel and fiberglass reinforced plastic. The
cooling tower, as shown in Fig. V-2, is .36 m (1 ft. 2 in.) in diameter
x 2.98 m (9 ft. 8 in.) in height and is constructed of 21 gage T. 304
stainless steel. Cooling water is introduced through four ports spaced
at .61 m (2 ft.) intervals on one side of the tower and is sprayed
vertically upward from the tower's center!ine. Four Bete Model W 10080 F
full cone stainless steel nozzles used for spraying are capable of
delivering up to 11.4 im (3.0 gpm) at .45 mPa (50 psig). A funnel built
into the bottom of the spray tower extends throught the trailer floor for
cooling water removal.
D. Particle Charging Corona Section
Only the corona section at the. downstream end of the first horizontal
gas passage (corona #2) was used. The corona shell is constructed from
4.8 mm (3/16 in.) wall thickness fiberglass reinforced plastic (FRP)
with interior dimensions of .61 m wide x 1.07 m high x 1.52 m long
(2 ft. x 3 ft. 6 in. x 5 ft.) in the direction of gas flow. Access
to a corona interior is through removable FRP end plates
which are normally bolted to 5.1 cm (2 in.) full perimeter 3.2 mm
(1/8 in.) thick face glanges on either end of a corona.
The corona is designed to operate in either a single or double lane1
gas passage mode. Switching from one to another requires rearrangement
of the adjustable collection plates and discharge frame(s). The width
of individual gas lane(s) for either mode is maintained at .3 m (1 ft.)
and the discharge frame to collection plate spacing is therefore .15m
(6 in.). Fig. V-3 shows a cutaway schematic of a corona set up for
single lane operation. The testing at Centralia Power Plant was
performed with a single lane corona section.
The overall dimensions of the discharge frame shown in Fig. V-3 are .70 m
high x 1.14 m long (27^ in. x 3 ft. 9 in.). The frame is constructed
of 6.4 mm x 1.91 cm (% in. x 3/4 in.) T. 304 stainless steel rectangular
bar stock members. Prior to the Centralia test program these frames were
modified by welding 3.2 mm (1/8 in.) diameter stainless steel rods in
4.45 cm (1-3/4 in.) lengths perpendicular to the vertical' members of
each discharge frame. The spikes have sharp points on both ends and
are welded at 5.0 cm (2 in.) intervals. This modification has decreased
the plate to frame spacing by 6.4 mm (1/4 in.).
The collection plates shown in Fig. V-3 are 1.05 m high x 1.5 m long .,
(41-1/4 in x 59 in.), giving each plate a cross sectional area of 1.58 m ,
which is used in the calculations for SCA. They are constructed from 11
gage T. 316 stainless steel. The plates serve as full chamber baffles
to keep the gases within the confines of the single lane passages.
10
-------
E
to
E
5
E
in
evi
3
o
UJ
V
.36m 0-12 go.
T. 304 S.S.
SPRAY NOZZLE
( I OF 4 SHOWN)
5.1 cm 0 x 5.1 cm
CLOSE NIPPLE
(4 LOCATIONS)
.30m 0
T. 304
- 12 go.
S.S.
Fig. V-2. Cooling Tower Schematic
11
-------
A negative corona is used to charge the particles negatively. This is
accomplished by maintaining the discharge frame(s) at a high negative
potential and the collection plates at a neutral or ground potential.
Field strengths generated in corona #2 is 6.15 KV/cm (15.61 KV/in.)
Corona power supplies are discussed in all other components inside the
corona. This isolation is provided by suspending the frame on two
2.5 cm (1 in.) diameter T. 303 stainless steel rods which are connected
to porcelain insulators. The Ceramaseal Model 902B1353-6 insulators are
housed in .3m diameter x .61 m long x 6.4 mm wall thickness (1 ft. x
2 ft. x 1/4 in.) plexiglass tubes which are centered 1.07 m (3 ft. 6 in.)
apart and are located on top of the corona shells. Two .30 m to .36 m x
7.6 cm (1 ft. to 1 ft. 2 in. x 3 in.) FRP reducing flanges are used to
join the plexiglass tubes to the corona top.
The insultators are continually flushed with a supply of heated purge
air. The temperature of the purge air is maintained at about 49°C
(120°F) and an even flow across a plexiglass tube section is obtained by
introducing the purge air through a distribution plate having approxi-
mately 10% hole area. The flushing face velocity of the purge air is
set at about .18 m/sec. (0.6 fps). This same purge air distribution flange
also serves as a support flange in that an insulator is bolted directly to
it. The high voltage lead-in to the discharge frame is through a feed-
through type insulator.
The collection plate and discharge frame flush system is shown schema-
tically in Fig. V-4. A continuous wall wash is supplied to the collection
plates through 2.5 cm (1 in.) FRP square tube which had 3.2 mm (1/8 in.)
diameter holes drilled diagonally into the corner adjacent to the collection
plate. .The discharge frame flush is an intermittent spray supplied by two
Bete 80° fan nozzles. The corona section and the mist eliminator are
equipped with this flushing system.
At the nominal gas flow rate of 1,700 am3/hr. (1000 acfm) the gas velo-
city in the corona is 1.45 m/sec (4.76 fps) for single lane operation
and .72 m/sec. (2.36 fps) for double lane operation. The corresponding
gas residence times are 1.05 and 2.10 seconds. By varying the volume
of air flow through the system, however, the gas residence time can
range from 0.70 seconds (single lane operation at 2550 am /hr. (1,500
acfm)) to 4.20 seconds (double lane operation at 850 anr/hr. (500 acfm)).
E. Water Spray Towers
The only spray tower used in the pilot plant during these tests comprises
the entire second horizontal gas passage. The spray tower is .91 m
in diameter x 4.8 mm wall thickness (3 ft. x 3/16 in.) and is constructed
from FRP. The length of the spray tower is 7.32 m [24 ft.). Gas velocity
in the spray tower at a nominal gas flow of 1700 arrH/hr. (1000 acfm ) is
0.72 m/sec. (2.36 fps). The corresponding gas residence time is 10.17
seconds.
Utilizing a single spray header in the tower, a maximum of 12 nozzles can
be used in the tower. The total number of nozzles used can vary depending
on the type of nozzle, the total water flow rate and the pressure head
12
-------
REMOVABLE GROUND PLATE (&S.)
FOR ONE LANE OPERATION
HIGH VOLTAGE
FEEDER CABLE
HIGH
VOLTAGE INSULATOR
DISCHARGE FRAME
SUSPENSION ROD
HIGH VOLTAGE
DISCHARGE FRAME
GAS INLET
AND OUTLET OPENING
PURGE AIR DUCT
FRP WALLS
S.S. GROUND PLATE (OUTER)
FOR TWO LANE OPERATION
Stainless Stee-1- Discharge Spikes
GROUND PLATE ALIGNMENT
AND SUPPORT BARS
F1g. V-3. Particle Charging Corona Section
-------
Collection Plate Flush
COLLECTION PLATES
(SINGLE LANE OPERATION)
PURGE AIR AND HIGH
VOLTAGE ENTRY(PURGE
AIR DUCTS, INSULATORS,
SUSPENSION RODS AND
DISCHARGE FRAME) OMITTED
FOR CLARITY
Discharge Frame Sprays
S
CORONA SHELL
Fig. Y-4. Collection Plate Flushing System
-------
desired. All nozzles spray in the direction of gas flow (co-currently).
A total of 5 to 7 nozzles were used in the Centralia tests, depending
on the desired flow rate. Bete TF6FCN full cone fog nozzle/header
arrangement typical for the spray tower is shown schematically in
Fig. V-5.
A positive charge is imparted to the water droplets by maintaining the
nozzles at a positive potential (direct charging). The nozzles are
electrically isolated from the spray tower walls by introducing heated
purge air through 7.6 cm diameter x 10.2 cm long (3 in. x 4 in.)
polyvinyl chloride (PVC) entry caps which are situated on top of the
two spray towers (see Fig. V-5). Both the water and the high voltage
lead-in cable enter through a 6.4 mm (1/4 in.) diameter street tee
fitting connected to the middle of each entry cap.
F. Mist Eliminator
The mist eliminator is situated in the middle of the third and last
horizontal pass and is used to remove entrained water droplets from the
airstream. The mist eliminator is identical to the corona sections,
with the exception of the discharge frame being maintained at a posi-
tive potential and the total height being 5.1 cm (2 in.) shorter
(necessitating an equivalent shortening of the discharge frame and
collection plates).
G. Test Ducts
The inlet and outlet test ducts are located immediately before the
first corona and immediately after the mist eliminator, respectively
(see Fig. V-l). Both test ducts are constructed from 4.8 mm wall
thickness (3/16 in.) FRP and are .30 m in diameter x 1.22 m long
(1 ft. x 4 ft.). Vertical gas flow in a downward direction is employed
because it allows the most convenient positioning of the particle sizing
source test equipment used and described in Section VI, "Particulate
Sampling Apparatus." The particle sizing source test equipment also
dictated the size of the test ports which are .15 m wide x .46 m high
(6 in. x 1 ft. 6 in.). The test ports are located three duct diameters
downstream and one duct diameter upstream from flow disturbances.
H. Fan
The fan used to induce the air flow (i.e., clean side) through the
pilot plant is a New York Blower Model RFE-12. The straight-bladed
fanwheel and housing are constructed from FRP. The fan is driven
through a split pulley belt drive by a Westinghouse 3.7 kW (5 hp), 208 volt,
3-phase motor turning at 1,800 rpm and is capable of delivering up to
2550 am3/hr. (1500 acfm) at 20.3 cm (8 in.) water colume (WC) static
pressure. The fan has a horizontal inlet and vertical outlet. A
4.8 mm (3/16 in.) FRP wall thickness x .30 m (1 ft.) diameter exhaust
duct containing an adjustable damper extends up through the trailer
roof.
15
-------
CT1
WATER HIGH
VOLTAGE
NOZZLE
Fig. V-5. Spray Header and Nozzle Arrangement Typical
to Spray Towers «1 and »2.
-------
As previously mentioned, a Dayton centrifugal blower was installed at
the scrubber inlet to increase the air flow capabilities at the
Centralia Power Plant.
I. High Voltage Power Supplies
Three high voltage power supply units used in the pilot plant serve the
corona, mist eliminator, and water droplet charging. All three units
operate off a 110 volt, 50 Hz, 1 0 supply and are equipped with multi-
range voltage and current meters on the high voltage output side. The
units are also equipped with overvoltage and overcurrent surge protection.
The power supplies used for the corona section
are equipped with spark rate controllers. The
which energizes the mist eliminator has a L.L.
voltage control and the NWL. unit for corona #2
rate controller. The three power supplies are
table.
and the mist eliminator
Universal Voltronics'unit
Little P-30 automatic
has an integrated NWL spark
described in the following
Table V-l. High Voltage Power Supply Units
Source
Corona #2
Mist Eliminator
Droplet Charging
Model
NWL
Universal Voltronics
Hipotronics #825-40
Polarity
Negative
Negative
Positive
Rated Peak
Output
KV
90
70
25
mA
30
25
40
J. Liquor Control Panel in the Scrubber Trailer
The water supply system for the cooling tower, spray tower and corona
flushing system is controlled at a single control panel situated near
the inlet sampling port. Two different sources supply water to the
control panel: charged water (either recycled or fresh) and uncharged
fresh water.
The charged liquor is used in the spray tower. It is pumped from a
sump tank in the liquor recycle trailer by a Gould centrifugal pump
model 3196ST. This pump provides the necessary liquor flow requirements
of .79 MPa (100 psig) and a maximum of 45.4 fc/min. (12 gpm) to the
spray tower. Since the spray charging is achieved by applying from 0
to 30 KV at the throat of each nozzle, the pump will also be at an
elevated potential. It is therefore electrically isolated on a micarda
base with a FRP cover.
17
-------
The uncharged fresh water is used in the gas cooling tower (approxi-
mately 11.4 £/min. (3 gpm) and the corona section flushing system (the
flow rate is unmonitored).
'.
K. Purge Air Heating System
The purge air heating system is schematically illustrated in Fig. V-6.
The system consists of both commercially available and custom built
components. The fan is a Barry Blower model BUF-90 Junior Fan em-
ploying a 248.3W (1/3 hp) motor with a maximum capacity of 850 cubic
meter per hour (500 acfm). The discharge air then passes through a
custom design Nelco Duct Heater. It is a 9 Kw heating unit with 4
stages to regulate the degree of heating required. The duct heater
operated on 208V, 30 power with a 110V control source which is external
of the heater. A custom designed distribution plenum follows the
heater and provides an adjustable purge air supply to the high voltage
access points on the corona section, the mist eliminator section and the
spray header in the tower.
18
-------
FRESH WATER
MAKE-UP
SPRAY
RECYCLE SUMP
CORONA NO. I
MIST
ELIMINATOR
DISTRIBUTION PLENUMAr
DUCT
HEATER
i
FILTER
CORONA N02
AfJ<>PURGE AIR
J INLET
ELEVATION VIEW
MIST ELIMINATOR
"PLAN VIEW"
F1g. V-6. Heated Purge Air System
-------
Section VI
DESCRIPTION OF THE LIQUOR RECYCLE SYSTEM
A. Description of Overall System
In order to use the Electrostatic Scrubber Pilot Plant to obtain data
needed to design larger control systems, the design of the liquor recycle
system was to enable the scrubber to operate in a "closed loop" mode.
The design was to be flexible so different scrubbing liquors could be used
with minimum modification to the system. The system has been built in a
moving van trailer so it is portable. The design also makes the recycle
system integrable with the present systemi i.e., the same electrical
isolation system is used.
The major components of the recycle system include a mix tank, two
alkaline slurry mixing tanks, two clarifying tanks, a sump tank and two
pumps. All plumbing between the equipment is done with PVC Schedule
80 pipe. The system is housed in a 12.2 m (40 ft.) long trailer and can
be easily transported to different emissions sources with the other parts
of the UW Electrostatic Scrubber Pilot Plant.
Fig. VI-1 shows the layout of the liquor recycle system. The effluent
of the scrubber goes first into a mix tank which is outside the back door
of the trailer. Alkaline slurry such as lime or Na2C03 is mixed in the
208 liter (55 gal.) drums above the mix tank and then added as a liquid
solution to this tank. From the mix tank the liquid passes a flow control
panel where it is divided into two streams.
One stream is directly recycled into the sump tank, and the other is
bled into the clarifying tanks. The latter stream then joins the
recycling stream before being pumped into the sump tank. From this tank
the liquor is pumped into the scrubber.
When operating in an "open-loop" liquor recycle mode the 7.6 cm (3 in.)
drain to the mix tank was disconnected and ducted to the power plant water
treatment system. Fresh water was then supplied to the mix tank at the
same consumption rate as supplied to the spray tower.
Water and Na?CO^ solution were the scrubbing liquors for open-loop
testing. J
B. Liquor Tanks
The mix tank is a 510 liter (135 gal.), .91 m (3 ft.) diameter, 1.22 m
(4 ft.) high tank that was purchased on a previous EPA grant. The alkaline
slurry mixing tanks are two teflon line 208. liter (55 gal.) drums. Flow
out of these tanks can be measured by sight glasses on the side of each tank.
From the mix tank the liquor is pumped through a Gould Model 3196 ST pump
to the flow control .panel.
20
-------
Alkaline Slurry Tanks
*
Control Rox
4
ro
Fig. VI-1. General Layout of the Liquor Recycle System
-------
The first clarifying tank is 2.13 m (7 ft.) high and 1.52 m (5 ft.) wide.
The liquor from the control box splits into two streams before entering the
tank. This helps minimize the disturbance of the settling material by the
incoming stream. A baffle plate is also suspended near these outlets for
the solids in the stream to impinge against. The diameter of the tank was
experimentally determined by the settling character!'sites of sludge composed
of flyash, lime and calcium carbonate. Knowing the solids flux as a function
of suspended solids concentration and assuming a maximum 38 liter influent
flowrate of 5% slurry into the tank, the tank must be at least 1.4 m (4.6 ft.)
in diameter. Its circular design is to prevent scale formation that could ;
happen in the corners of square tanks. The bottom of the tank has a 60
angle so the settled sludge can flow to the bottom drain without mechanical
assistance.
The second clarifying tank is 1.19 m (4 ft.) high and 1.19 m wide. The feed
from the first to the second settling tank is by gravity. Makeup water lost
by sludge removal from the first tank as well as Mq if additional alkalinity
is needed in the lime/limestone system. A 2.2 kW (3 hp), 3600 rpm, 110 v,
10 Deming centrifugal pump transfers the liquor from the second clarifying
tank to a point where it mixes with the liquor recycle stream and goes into
the sump tank. The sump tank is a .91 m (3 ft.) diameter 1.52 m (5 ft.)
high tank also purchased on a previous grant. As discussed in Section V-J
(page 17) the Gould Model 3196 St pump, housed in the main pilot plant trailer
sprays the liquor into the spray tower from this tank.
22
-------
Section VII
EXPERIMENTAL PROCEDURES AND TEST EQUIPMENT
The field test measurements for the UW Electrostatic Scrubber are listed in
Table VII-I.
A. UW Mark 10-20 Cascade Impactors
The Mark 10 and Mark 20 Cascade Impactors were used to measure both particle
size distribution and masis concentration at both the inlet and outlet test
ducts, respectively. The impactors provide this information by segregating
the aerosol sample into discrete size intervals. The weight on each plate
provides size distribution information and the total weight is used to
determine the mass concentration. The Mark 10 has been designed for high
particle concentrations characteristic of the scrubber's inlet and uses 27
stages plus one final filter to reduce overloading. The Mark 20 is designed
for low particle concentrations as exist at the outlet of the scrubber and
has 14 stages plus one final filter. Both impactors utilize reduced
absolute pressure in the last impactor stages to size particles as small as
.OByin diameter (aerodynamic).
The basic components of a sampling train utilizing a UW Cascade Impactor
as shown schematically in Fig. VII-1. The impingers in the condenser unit
are used to collect water vapor in the sample air stream and provide a basis
for calculating the moisture content of the gas stream which may be checked
against the wet and dry bulb determination. The dry gas meter is used to
determine the total sample volume. The absolute pressure gauge measures
the pressure on the last stage of the impactor.
By conducting simultaneous particles size distribution tests at both the
inlet and outlet test ducts, the size-dependent collection efficiency curve
of the pilot plant may be measured.
B. TECO Model 40 S02 Analyzer
A Thermo Electron Model 40 Fluorescent S02 Analyzer was used to measure
SOo levels in the gas stream at the inlet and outlet of the pilot plant. The
principle of operation of this monitor is based upon; the measurement of
the fluorescence of S02 produced by its absorption of ultraviolet radiation.
A sample gas conditioning unit was designed and fabricated at UW. The
unit samples the gases from the inlet or outlet ducts, passes the gases
through heated Teflon tubing, through a heated 19 x 90 mm glass fiber
thimble filter, through a Greenburg-Smith impinger containing sulfuric
acid (for removing excess water vapor), and then through heated Teflon
tubing into the TECO instrument.
A strip chart recorder was attached to the S02 analyzer. This allowed for
continuous monitoring of the S02 concentration being measured. Each data
. point in this report represents a 5 minute average as recorded on the strip
chart. By knowing the inlet and outlet S02 concentrations and gas flow .
rates, the S02 collection efficiency of the scrubber pilot plant could be
calculated.
23
-------
Table VII-1. Field Test Parameters and Measurement Methods
Parameter
Measurement Methods
1. Gas properties
- Sulfur dioxide concentration
- Velocity
- Volumetric flow rate
- Temperature
- Moisture
Continuous instrumentation,
batch source tests
Pi tot tube
Orifice flow meter
Thermometer, thermocouples
Wet-dry bulb, continuous
instrument
2. Particle properties
- Size distribution
- Mass concentration
UW Mark 10-20 Cascade Impactor
UW Mark 10-20 Cascade Impactor
3. Liquor properties
- PH
- Sulfite concentration
- Sulfate concentration
- Suspended solids
- Flow rate
- Pressure
pH meter
Titration with KI-KIO.
Gravemetric
Filtration
Rotameters
Pressure gauge
4. Scrubber conditions
- Particle charging voltage
- Particle charging current
- Liquor charging voltage
- Liquor charging current
- Mist eliminator voltage
- Mist eliminator current
- Gas pressure drop
Voltmeter on power supply
Ammeter on power supply
Voltmeter on power supply
Ammeter on power supply
Voltmeter on power supply
Ammeter on power supply
Static pressure taps
24
-------
MARK 10 OR MARK 20
CASCADE IMPACTOR
STACK WALL
SYSTEM 10/20 PORT
MOUNTING PLATE
SYSTEM 10/20
SAMPLE PROBE
ISOKINETIC
NOZZLE
ro
tn
GAS
FLOW
FLOW CONTROL VALVE
SYSTEM 10/20
INSTRUMENT BOX
SAMPLE LINE
CONDENSER UNIT VACUUM PUMP
CONNECTING CORD
DRY
GAS
METER
Fig. VII-1. UW Mark 10-20 Cascade Impactor Sampling System Schematic
-------
Section VIII
PARTICIPATE COLLECTION EFFICIENCY RESULTS
A. General Test Description
The first series of participate collection efficiency measurements on
the UW Electrostatic Spray Scrubber were performed with the unit in
the open-loop mode. The scrubbing liquor for these tests was water.
B. Particulate Collection Efficiency Measurements
Results of inlet-outlet impactor tests 1-4 and 7-8 are shown in Table
VIII-1. Scrubber operating parameters (corona voltage, mist eliminator
voltage and liquor flow rate) were held constant for tests 1-4 and 7-8
with a variation in spray voltage from 0 to 10 KV, A significant
variation in the inlet gas flowrate from 31.2 sdm /min. (1101 sdcfm
(1449 acfm)) to 36.2 sdm .min. (1281 sdcfm (1637 acfm)) was noted.
This variation was unavoidable. With both the scrubber I.D. fan and
booster fan operating at full capacity, a variation in the negative
static at the outlet of the air preheater (due to varying boiler
conditions) resulted in subsequent variations in inlet gas flow. This
created changes in the parameters L/G, SCA, and gas residence time.
The overall particle collection efficiency for tests 1-4 and 7-8
ranged from 98.99% to 99.80% (0.20% to 1.01% penetration). The
particle collection efficiency and penetration as a function of
particle size (aerodynamic cut diameter of cascade impactor stages,
d50) for these tests are shown in Fig. VIII-1. The symbols on the
carves of Figure VIII-1 are computer calculated collection efficiency
points and are on the graph only to identify the curves. The
aerodynamic cut diameter of the impactor stages is defined as the
diameter of the particle of unit density collection with 50% efficiency
and is calculated by:
18.
'
50 " CVj
where y is the gas viscosity, Dj the jet diameter, ₯,.« the inertial
impaction parameter at 50% collection efficiency, for particles of
diameter d, C the Cunningham correction factor, and Vj the gas velocity
in the jet aiameter. The collection efficiency particles greater than
.3 microns diameter was above 90% for each of these tests.
Figures VIII-2 and VIII-3 are the lognormal approximation and the
parabolic fit curve of the actual collection efficiency for test 2 and 3.
Test 2 was run with electrostatically charged liquor and test 3 was not.
Both tests had approximately the same overall collection efficiency
(99.55% for test 2 vs. 99.58% for test 3), and this is reflected by
the lognormal approximation. The actual collection efficiency curves
show the increase in collection of particles below .5 v diamaeter.
26
-------
Test
No.
1
2
3
4
7
8
Date
3/01/79
3/02/79
3/02/79
3/14/79
5/02/79
5/02/79
Inlet
Gas Flow
sdm /min
(sdcfm)
34.6
(1223)
34.5
(1219)
36.2
(1281)
36.0
(1270)
33.1
(1168)
31.2
(1101)
Outlet
Gas Flow
sdm3/min
(sdcfm)
51.7
(1825)
50.1
(1769)
50.7
(1791)
48.8
(1724)
44.0
(1553)
46.2
(1631)
Particle Mass Cone.
gm/sdm3
(grains/sdcf )
Inlet
1.0973
(0.47952)
0.9185
(0.40142)
1.0913
(0.47690)
1.2688
(0.55448)
1.0198
(0.44566)
1.1011
(0.48117)
Outlet
0.0074
(0.00324)
0.0029
(0.00126)
0.0033
(0.00146)
0.0049
(0.00214)
0.0016
(0.00069)
0.0015
(0.00065)
Overal 1
Coll. Eff.
(*)
98.99
99.55
99.58
99.48
99.79
99.80
Penetration
<*)
1.01
0.45
0.42
0.52
0.21
0.20
Total Liquor
Flow Rate
l/min. (gpm)
60.6 (16)
60.6 (16)
60.6 (16)
60.6 (16)
60.6 (16)
60.6 (16)
Corona
Voltage
(kV)
(-)90
90
90
90
90
90
Spray
Voltage
(kV)
0
(+)10
0
10
10
0
Demister
Vol tage
(kV)
(*)60
60
60
60
60
60
SCA
mZ/sm3/m1n
(ftZ/scfm)
0.091
: (0.028)
0.091
(0.028)
0.087
(0.026)
0.087
(0.027)
0.095
(0.029)
0.101
(0.031)
L/G
1/aKm3
(gal/1000 acf)
1.14 (10.6)
1.12 (10.4)
1.05 (9.8)
1.02 (9.5)
1.13 (10.5)
1.18 (11.0) '
Gas Residence Time (sec.)
Corona
Section
0.86
0.86
0.82
0.83
0.90
0.95
Spray
Tower
8.32
8.35
7.95
8.01
8.71
9.25
Mist
Eliminate*
0.82
0.82
0.78
0.79
0.86
0.91
ro
Table VIII-1. Results of Cascade Impactor Tests 1-4 and 7-8 at Centralia Power Plant.
-------
99.99
99.9
UJ
(_>
OH
LU
Q_
O
UJ
ii
(J
U_
UJ
O
UJ
O
O
LU
*-l
CJ
99.0
90.0
0.0
Tno-2
UW Electrostatic Scrubber
Central!a Power Plant
March-May 1979
Units
Overall Efficiency (%)
Overall Penetration (%)
SCA (m2/.(sm3/niin))
L/G (£/akm3)
Test
No.
1
2
3
4
7
8
Symbol
o
©
A
+
X
X
Overal 1
Eff.
98.99
99.55
99.58
99.48
99.79
99.80
Pen.
1.01
0.45
0.42
0.52
0.21
0.20
SCA
.091
.091
.087
.095
.095
.101
L/G
1.14
1.12
1.05
1.02
1.13
1.18
6-00 10"1 2 5 10° 2 5 101
PflRPICLE RERODYNflMIC DIRMETER. DSO(MICRONS)
Fig. VIII-1.
Particle Collection Efficiency and Penetration vs. Particle
Size for Impactor Tests 1-4 and 7-8(Log-Normal Approximation)
28
-------
).99
99.9
UJ
<£
UJ
Q.
u_
UJ
_
UJ
O
O
UJ
d
a:
a:
a.
99.0
90.0
0.0
UW Electrostatic Scrubber
Central ia Power Plant
March-May 1979
Test Eff. , SCA
No. (%):Cm2/(sm:Vmin))
L/G
2
3
99.55
99.58
0.091
0.087
1.12
1,05
Voltage(kV)
Corona Spray
C-)90 C+110
90 0
6.00
2 5 10° 2 5 101
PHRTICLE flERODYNRMIC DIflMETER. D50(MICRONS)
10-z
O
II
t
cc
a:
UJ
o_
Fig. VIII-2.
Particle- Collection Efficiency and Penetration vs. Particle
Size for Impactor Tests 2 and 3 (Log-Normal Approximation)
29
-------
99.9
CJ
a:
UJ
a.
UJ
u_
w 99.0
o
UJ
o
0
UJ
g 90.0
o.
o.o
1-110-2
UW Electrostatic Scrubber
Central!a Power Plant
March-May 1979
L/G
U/akm3)
Voltage(kV)
Corona Spray
2 9.9.55
3 99.58
0.091
0.091
u
6 UJ
Q_
8
CC
2 £
6.00 10-* 2 5 10° 2 5 101
PRRTICLE flERODYNHMIC DIflMETER. D50(MICRONS)
Fig. VIII-3.
Particle Collection Efficiency and Penetration vs. Particle
Size for Impactor Tests 2 and 3 (Parabolic Curve Fit)
30
-------
Particle mass distributions for tests 2 and 3 are shown in Figures VIII-4
and VIII-5. Oh these graphs the particle collection efficiency is the area
between the scrubber inlet and outlet mass distribution lines. As with
the actual efficiency curve, the maximum removal is measured above 10 y
diameter particles and the minimum with the dcn equal to approximately
.5 y. bu
The cumulative size distribution for tests 1-4 and 7-8 measured at the
inlet and outlet of the scrubber are shown in Figures VIII-6 and VIII-7.
The particle mass mean diameter (particle diameter at which 50% of the
particle mass is greater than this diameter and 50% less) was in the
22.2 to 65.6 micron diameter range at the scrubber inlet and in the
0.74 to 5.91 micron diameter range at the scrubber outlet. There is a
good correlation between the actual particle size distribution as measured
with the cascade impactors and the straight line (log-normal) approximation
as seen with test 3 in Fig. VIII-8, so this report uses the log normal
size distribution.
The particle mass concentrations (grains/sdcf) less than the stated particle
aerodynamic diameter, d5Q (microns) for these tests are shown in Fig. VIII-9
and VIII-10. The curves in these graphs illustrate the reduction in particle
size range at the inlet and outlet of the scrubber.
Particle mass concentrations measured at the scrubber outlet for tests 1-4
and 7-8 ranged from 0.0015 gn/sdm3 (.00065 grains/sdcf) to 0.0074 gm/sdm3
(.00324 grains/sdcf). The difference between the particulate grain'loading
at the scrubber inlet and outlet prevented simultaneous sampling. Despite
sampling times at the outlet as high as 1 hour, low weights on the impactor
substrates were experienced.
31
-------
UW Electrostatic Scrubber
Centralia Power Plant
March 2, 1979
Eff. Pen. 9 SCA, l/G, Voltage(kV)
(%) (%) (nf/(snr/min))f£/akm3) Corona Spray
99.55 0.45
10
2 5 10°
DIHMETER. MICRONS
VIII-4. Particle Mass Distribution vs. Particle Size for Impactor
Test 2
32
-------
UW Electrostatic Scrubber
Centralia Power Plant
March 2, 1979
Eff. Pen. . SCA, L/G, Voltage(kV)
(%) (%) (mz/(sm3/Piin))(l/akm3) Corona Spra^
10-4
10
,-2
2
10
2 5 10°
OIRMETER. MICRONS
Fig. VIII-5. Particle Mass Distribution vs. Particle Size for Impactor
Test 3
33
-------
100.0
80.0
60.0
50.0
40.0
30.0
20.0
10.0
8.0
1 6.0
es.o
1 4.0
£ 3.0
ui
§2.0
o
< 1.0
a.
.8
.6
.5
.4
.3
.2
0.1
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Inlet Size
Distribution
/' / / Outlet Size
/ / / Distribution
' ' '
J I__J I
i i / i i i i i i
001
Fig.
01 0.2 Q5 I 2 5 10 20 30 4O 50 60 70 BO 90 95
% of MASS LESS THAN STATED DIAMETER
99 998 99.9 99.99
VIII-6. Inlet and Outlet Particle Size Distributions for Impactor
Tests 1-4 (Log-Normal Approximation)
34
-------
100.0
80.0
60.0
50.0
40.0
30.0
20.0
10.0
8.0
le.o
|5.0
14.0
?(3.0
bl
52.0
kl
O
&
.8
.6
.5
.4
.3
.2
0.1
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Inlet Size
Distribution
Outlet Size
Distribution
0.01 0.1 0.2 Q5 I 2 5 10 20 30 4O 50 60 70 80 90 95
% Of MASS LESS THAN STATED DIAMETER
99 99S 99.9 99.99
Fig. VIII-7. Inlet and Outlet Particle Size Distributions for Impactor
Tests 7-8 (Log-Normal Approximation)
35
-------
CO
D
o:
LJ
UJ
UJ
CE
ii
a
OH
OC
O.
D
4
2
ioi
8
6
4
2
10°
8
6
4
2
10-1
8
6
4
2
n-7
-
/
/
'
/
j
/
est 3
X
/-"/
/
»
nlet
/
"7
^
*~
/
/
/
s
J
/
/
^
^X" X
x^' y
/
/
/
/
/
A
/
/
^
y
/
/
^
X
/
/
/
/
y
/
X
f
/
/
f
/
7
/
j
s
/
/
//
/
^
L(
/
//
/
/ X
1 /
' '
Test 3 Outlet
ctual
g Horma
kppn
ximati
)n
10 20304050607080 90
PERCENTflGE SMflLLER (BY WEIGHT)
95
Fig. VIII-8.
Inlet and Outlet Particle Size Distributions for Impactor
Test 3 (Log-Normal Approximation and Actual Data)
36
-------
a
CO
t»
cr
s
UJ
GC
t
O
UJ
u
s
I
CO
CO
to
10"
8
6
4
UJ
o
CO
o
I I I T I I
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Inlet Size
Distribution
Outlet Size
Distribution
Test 1
2 »
3 *
4 *-
3.00 5 10~* 2 5 10° 2 5 10*
PflRTICLE flERODYNRMIC DIRMETER. D50IMICRONS)
2 3.00
Fig. VIII-9. Particle Mass Concentration for Particle Less Than Stated
Diameter for Impactor Tests 1-4.
37
-------
;3.00
UJ
CJ
t
I
o:
cc
Q_
8
6
O
CO
CO
a:
UJ
UJ
cc
i
O
a:
cc
a.
a
UJ
cc
CO
CO
CO .,
UJ 10-3
-1 8
z
O 6
UJ
CJ
o
CJ
CO
CO
£ 8
ur«-
6
3.00
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Inlet Size
Distribution
8
-«Bt
7
or'
.J*~'
Outlet Size
Distribution
8
3.00 5 MT1 2 5 10° 2 S 101
PflRTICLE flERODYNflMIC DlflMETER. D50(MICRONS)
2 3.00
Fig. VIII-10. Particle Mass Concentration for Particles Less Than Stated
Diameter for Impactor Tests 7-8.
38
-------
Section IX
SULFUR DIOXIDE COLLECTION EFFICIENCY
A. General Test Description
S02 collection efficiency measurements on the UW Electrostatic Spray
Scrubber were performed with the unit in an open-loop operating mode.
In the open loop mode fresh water was added to the mix tank at the same
consumption rate as supplied to the tower, and the scrubber effluent
sent to the power plant water treatment system. Tests were run with fresh
water and with Na2C03 solution. Concentrated Na2CO, liquor was added to
the mix tank at a fixed rate during the .latter tests. S02 collection
efficiency measurements, described in Section VII, were performed on the
scrubber in both operating modes.
B. S02 Collection Efficiency
1. Results Using Water as a Scrubbing Liquor
initial SO, collection efficiency tests were performed with water
on March 16th. Also at the beginning of each test day with open-
loop system, several tests were run using water as a scrubbing liquor
The results of inlet/outlet S02 measurements are shown in Table IX-1
As with the particulate tests a significant variation in the inlet
gas flow rate from 32.9 sdm /min. (1163 sdcfm) to 41.5 sdm /min.
(1465 sdcfm). This created changes in gas residence time and L/G..
Inlet S02 concentrations varied from 270 ppm to 1300 ppm. Using
water as a scrubbing liquor, the overall S09 collection efficiency
ranged from 16.21% to 50.82%. *
Test series 2 demonstrated the effect of the corona charge and spray
voltage charge. With particle charging corona section off the
SO, removal increased from 17.56% to 20.09% by increasing the spray
voftage from 0 to 7.5 KV. With the corona on the collection efficiency
was increased to 25.48%. At this condition as with test series 1,
the spray voltage did not have a significant effect on the S02
collection efficiency.
Test Series 3 demonstrated the effect of L/G on S02 removal. With
the sprays off, S02 collection was 8.02% due to SO, absorption by-the
wet wall corona sections. As the L/G was increasea to 1.10 £/akm
(10.3 gal/1000 acf), the collection efficiency increased to 23.46%.
These results are illustrated in Fig. IX-1.
A comparison of test series 1, 4, 5 and 7 shows the effect of inlet
S0? concentration on S02 collection efficiency. This comparison is
plotted on Fig. IX-2. S02 removal ranged from 16.21% at 270 ppm
inlet SO, to 50.82% at 1300 ppm. The reason for this increased
39
-------
Table IX-1. Results of S02 Tests at Centralia Power Plant Using Water as a Scrubbing
Liquor.
Test Series
No. & Date
1
3-16-79
2
3-16-79
3
3-16-79
4
3-30-79
5
4-18-79
6
4-18-79
7
5-03-79
Inlet
Gas Flow
sdm /mln
(sdcfm)
37.2
(1314)
37.2
(1314)
37.2
(1314)
36.9
(1302)
41.5 .
(1465)
38.4
(1356)
32.9
(1163)
Outlet
Gas Flow
sdm /mi n
(sdcfm)
48.9
(1728)
48.9
(1728)
48.9
(1728)
46.2
(1632)
50.3
(1777)
49.8
(1757)
43.8
(1547)
S02
Concentration
(ppm)
Inlet
430
430
430
415
415
415
415
490
490
490
490
1300
615
615
500
500
270
270
Outlet
258
249
252
260
252
235
235
342.5
315
302.5
286
510
375
365
310'
290
170
160
Overal 1
Collection
Efficiency
(%)
21.05
23.80
22.88
17.56
20.09
25.48
25.48
8.02
15.41
18.76
23.46
50.82
26.04
28.01
19.67
24.85
16.21
21.14
Stoichiometric
Ratio
/ moles alkali «
\nole inlet S02'
0
0
0
L/G
fc/akm3
(gal/lOOOacf )
'1.10
(10.3)
1.10
(10.3)
0
0.53 (5.0)
0.80 (7.5)
1.10(10.3)
1.06
(9.9)
0.93
(8.6)
0.80
(7.5)
0.94
(8.8)
Corona
Charge
(kV)
(-)70
70
70
0
0
90
90
90
90
90
90
90
90
90
90
90
90
90
Spray
Vol tage
(kV)
0
( + )5
10
0
7.5
0
7.5
0
0
0
0
0
0
7.5
0
15
0
10
Demister
Vol tage
(kV)
(+)60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
-------
L/G (gal/1000 acf)
5.
10.
25
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
20
o
-------
60
I I I
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
50
40
01
r-
O
c
o
(J
O>
e\j
o
1/5
30
20
10
I
Test Series #1,4,5, and 7
Mater Liquor
{= = 0.93-1.10 -e/akm3
b
Spray Voltage = 0 kV
TF6FCN Fog Nozzle
Spray Pressure = .79 MPa
I I
200 400 600 800 1000
In (ppm average)
1200
1400
Fig. IX-2. S02 Collection Efficiency vs. Inlet SOo Concentration
for Tests Using Water as a Scrubbing Liquor
42
-------
0)
r-
O
CU
§ 3
o
IO
.c
O
I
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Test Series #1,5,6, & 7
Water Liquor
{jr = 0.74-1.10 £/akm3 _
TF6FCN Fog Nozzle
Pressure = .79 MPA
I I
10
Spray Voltage (kV)
15
fig. IX-3. Change in SO, Collection Efficiency vs. Spray
Voltage for Tests Using Water as a Scrubbing
Liquor
43
-------
removal is that a higher inlet S02 concentration raises the rate
of mass transfer at the gas-liquia interface.
The greatest effect of spray voltage upon S0« collection was seen
in test series 6. By increasing the spray voltage from 0 to
15KV,the efficiency was raised from 19.67% to 24.85%. Fig. IX-3
illustrates the change in S0? collection efficiency vs. the spray
voltage change in test series 1, 5, 6, and 7.
2. Tests Using Na2C03 Solution as a Scrubbing Liquor
S02 collection efficiency tests were run with Na?C(L liquor in March
through May 1979. For these tests, concentrated^Na«C03 solution
(10% by weight) was metered into the mix tank at a rate designed to
give a desired stoichiometric ratio. The pilot plant was run for
one half hour prior to each reading to allow the sodium carbonate
concentration to equilibrate throughout the entire liquor system.
As shown in Table IX-2 five sets of tests were conducted. The over-
all results of these tests showed the effect of spray voltage,
stoichiometric ratio (SR), liquor-to-gas ratio (L/G), and inlet S(L
efficiency varied from 38.38% to 97.41% using Na2C03 scrubbing liqoor.
During test series 1-N the inlet S0? concentration was 1200 ppm,
the highest encountered during these tests. Even with the low
stoichiometric ratio of .36, a collection efficiency of greater than
71% was realized. By increasing the spray voltage from 0 to 10 KV,
the SO,, efficiency was increased by 2.1% when the SR = .29, "and by
1.5% at SR = .36.
Test series 2-N and 3-N demonstrate the effect of L/G upon S02
collection efficiency as shown in Fig. IX-4. With inlet S02 Concen-
trations approximately the same, raising the L/G from 0.87 to
0.96 Vaknr increased the S02 removal from 8% to 16%. An increase
in collection efficiency was also seen with the application of
charge to the sprays, this difference being greater at lower stoich-
iometric ratios.
Inlet S02 concentrations of Runs 4 and 5 were only 265 ppm and 350 ppm,
respectively, which resulted in corresponding low S0? collection
efficiencies. Fig. IX-5 illustrates the effect of inlet S02 concen-
tration upon S0? efficiency with SR constant. Despite low Tnlet S02
levels test 5-N does show an increase in S0? removal from 45% to
78% by increasing the SR from 0.33 to 1.75. Collection efficiency
at all stoichiometric ratios was increased by application of a 10 KV
spray voltage as seen in Fig. IX-6.
44
-------
Table IX-2.
suits of SO, Tests at Centralia Power Plant Using Na2CO, Solution as
Scrubbing Liquor. J
Test Series
No. & Date
1-N
3-30-79
2-N
4-18-79
3-N
4-19-79
4-N
5-3-79
Inlet
Gas Flow
sdm /mln
(sdcfm)
36.9(1302)
36.9 1302)
36.9(1302)
36.9(1302)
41.5(1465)
41.5
38.4
38.4
1465
1356)
1356)
35.9(1268)
35.9(1268)
35.9(1268)
35.9(1268)
35.9(1268)
32.7
32.7
32.7
32.7
1156)
1156)
1156)
1156)
32.7(1156)
32.7
32.7
32.7
1156)
1156)
1156)
33.8(1192)
33.8(1192)
33.8(1192)
33.8(1192)
33.8(1192)
33.8(1192)
33.8(1192)
33.8(1192)
Outlet
Gas flow
sdirr /mln
(sdcfm)
46.2(1632)
46.2(1632)
46.2(1632)
46.2(1632)
50.3(1777
50.3(1777
49.8(1757
49.8(1757
48.0
48.0
48.0
48.0
1695
1695
1695
1695
48.0(1695)
43.8(1547)
43.8(1547)
43.8(1547)
43.8(1547)
43.8(1547)
43.8(1547)
43.8(1547)
43.8(1547)
46.4(1637)
46.
46.
46.
46.
46.
46.
46.
1637
1637
1637
1637
1637
1637
1637)
S02
Concentration
(ppm)
Inlet
1200
1200
1200
1200
590
590
500
500
600
600
540
540
480
255
255
255
255
255
265
265
265
265
265
265
265
265
265
275
275
Outlet
350
330
275
260
85
80
14
10
230
215
112
108
52
70
66
66
62
60
52
55
58
66
63
98
95
115
111
115
118
Overall
Collection
Efficiency
(%)
63.44
65.53
71.27
72.84
81.95
83.01
96.37
97.41
48.76
52.10
72.27
73.26
85.52
63.26
65.36
65.36
67.46
68.51
73.74
72.23
70.71
65.80
67.35
49.21
50.77
40.40
42.48
42.57
41.07
Stoichiometric
Ratio
moles Na0CO,
/ l .t \
lmole inlet S02'
.29
.29
.36
.36
.66
.66
1.55
1.55
.37
.37
.82
.82
1.90
1.07
.07
.07
.07
' .07
.21
.21
.21
.82
.82
.60
.60
.45
.45
.45
.45
L/G
fc/akm3
(gal/lOOOacf)
1.06
1.06
1.06
1.06
9.9)
9.9
9.9)
9.9)
0.93 (8.6)
0.93
0.98
8.6)
9.2)
0.98 (9.2)
0.87 (8.1)
0.87 (8.1)
0.87 (8.1)
0.87 (8.1)
0.87 (8.1)
0.94 (8.8)
0.94
0.94
0.94
0.94
0.94
0.94
0.94
8.8)
8.8
8.8)
8.8
8.8)
8.8)
8.8)
0.91 (8.5)
0.91 (8.5)
0.91 (8.5)
0.91
0.91
0.91
0.91
0.91
8.5)
8.5)
8.5)
8.5)
8.5)
Corona
Charge
(kV)
(-)90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
Spray
Voltage
(kV)
0
(+)io
0
10
0
7.5
0
7.5
0
7.5
0
7.5
0
0
7.5
10
0
7.5
0
7.5
0
0
7.5
0
10
0
7.5
15
0
Demister
Voltage
(kV)
(+)60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
01
-------
Table IX-2 (cont.)
Results of SO,, Tests at Centralia Power Plant Using
as a Scrubbing Liquor.
Solution
Test Series
No. & Date
5-N
5-17-79
Inlet
Gas Flow
sdm /min
(sdcfm)
32.3(1142)
32.3(1142)
32.3(1142)
32.3(1142)
32.3(1142)
36.8(1300)
36.8(1300)
36.8(1300)
36:8(1300)
36.8(1300)
36.8(1300)
Outlet
Gas Flow
sdm /min
(scdfm)
45.2(1595)
45.2(1595)
45.2(1595)
45.2(1595)
45.2(1595)
47.1(1662)
47.1(1662)
47.1(1662)
47.1(1662)
47.1(1662)
47.1(1662)
S02
Concentration
(pptn)
Inlet
340
340
355
355
355
338
338
330
330
360
360
Outlet
140
142.5
130
122
120
no
105
97
103
63
61
Overall
Collection
Efficiency
(V
42.49
41.46
48.85
52.00
52.79
58.39
60.28
62.42
60.10
77.63
78.34
Stoichiometric
Ratio
moles Na.CO.
/ £ J \
W>le inlet S02'
.33
.33
.42
.42
.42
.53
.53
.68
.68
1.25
1.25
L/G
a/a km3
(gal/lOOOacf)
0.88 (8.2)
0.88 (8.2)
0.88 (8.2)
0.88 (8.2)
0.88 (8.2)
0.77 (7.2)
0.77
0.77
0.77
7.2)
7.2)
7.2)
0.77 (7.2)
0.77 (7.2)
Corona
Charge
(kV)
90
90
90
90
90
90
90
90
90
90
90
Spray
Voltage
(kV)
10
0
0
10
14.5
0
10
10
0
0
10
Demister
Voltage
(kV)
60
60
60
60 .
60
60
60
60
60
60
60
en
-------
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
Test Series #2-N
Na2C03 Liquor
r = 0.96
545 ppm (average)
TF6FCN Fog Nozzle
Pressure = .79 MPa
20
.5 1.0
Stoichiometric Ratio
1.5
Fig- IX-4. Effect of L/G on S02 Collection Efficiency For
Tests Using Na2C03 Solution as a Scrubbing Liquor
47
-------
IQCf
90
80
70
u
c
a;
i
U
60
5C
U
QJ
O
(_>
evj
30
20
10
300
I I
UW Electrostatic Scrubber
Central!a Power Plant
March-May 1979
0.33
Na2C03 Liquor
| = 0.77 - 0.93 £/akm3
Spray Voltage = 0 kV
TF6FCN Fog Nozzle
Spray Pressure = .79 MPa
I I
400 500 600
S02 In (ppm average)
700
Fig. IX-5. S02 Collection Efficiency vs. Inlet SOo
Concentration at Varying Stoichiometric Ratios
48
-------
80
I I :
UW Electrostatic Scrubber
Centralia Power Plant
March-May 1979
70
o;
« 60
o
^
(->
o
o
o-
C\I
o
to
50
Na2C03 Liquor
{jr = .82 £/akm3
S02 In = 345 ppm (average)
TF6FCN Fog Nozzle
Pressure = .79 MPa
40
.25
1
I
1
,50 .75 1.00
Stoichiometric Ratio
1.25
1.50
Fig. IX-6. Effect of Stoichiometric Ratio and Spray
Voltage on S02 Collection Efficiency
49
-------
Section X
REFERENCES
1. Kraemer, H. F. and H. F. Johnstone (1955) "Collection of aerosol
particles in the presence of electric field," Ind. Engr.
Chem. 47 2426.
2. Penney, G. W. (1944) "Electrified liquid spray dust precipitator,"
U.S. Patent No. 2,357,354.
3. Pilat, M. J., S. A. Jaasund, and L. E. Sparks (1974) "Collection
of aerosol particles by electrostatic droplet spray scrubbers."
Envir. Sci. & Tech. £ 340-348.
4. Pilat, M. J. (1975) "Collection of aerosol particles by electro-
static droplet spray scrubber," APCA Journal 25_ 176-178.
5. Pilat, M. J. and D. F. Meyer (1976) "University of Washington
Electrostatic Spray Scrubber evaluation" Final Report on
Grant No. R-803278, EPA Report No. EPA-600/2-76-100 (NTIS
No. PB 2526537AS).
6. Pilat, M. J., G. A. Raemhild, and A. Prem (1978) "University of
Washington Electrostatic Scrubber tests at a steel plant,"
EPA Report No. EPA-600/7-78-177a.
7. Pilat, M. J. and G. A. Raemhild (1978) "University of Washington
Electrostatic Scrubber tests at a coal-fired power plant,"
EPA Report No. EPA-600/7-78-177b.
50
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-245
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
University of Washington Electrostatic Scrubber
Tests: Combined Particulate and SO2 Control
5. REPORT DATE
November 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
M.J.Pilat
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Washington
Department of Civil Engineering
Seattle, Washington 98195
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
Grant No. R806035
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: 6/78 - 8/79
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES IERL-RTP project officer is Dale L. Harmon, Mail Drop 61, 919/
541-2925. EPA-600/7-78-177a and -177b are earlier related reports.
16. ABSTRACT
The report gives results of tests of a 1700 a cu m/hr University of Wash-
ington electrostatic spray scrubber pilot plant on a coal-fired boiler to demonstrate
its effectiveness for controlling fine particle and SO2 emissions. The multiple-pass
portable pilot plant operates by combining oppositely charged aerosol particles and
water droplets in two spray towers. Aerosol charging sections at a negative polarity
precede each spray tower. For these tests, the pilot plant used only one charging
section and one spray tower. A liquor recycle system was constructed, permitting
the pilot plant to operate in an open- or closed-loop mode. All SO2 tests were run
in an open-loop mode using either water or Na2CO3 solution as the scrubbing liquor.
Simultaneous inlet and outlet source tests using cascade impactors provided size-
dependent and overall mass basis particle collection efficiency data. Measured over-
all particle collection efficiencies were 98.99%-99. 80%, depending on scrubbing
operating conditions. SO2 collection efficiencies were 8.02%-97.41%, depending on
the scrubber operating conditions, inlet SO2 concentration, and the type of scrubbing
liquor used.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Flue Gases
Scrubbers
Electrostatics
Desulfurization
Dust
Aerosols
Sodium Carbonates
Pollution Control
Stationary Sources
Particulate
13B
2 IB
07A,13I
20C
07D
11G
07B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
60
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
51
------- |