EPA-600/2-76-142
May 1976
Environmental Protection Technology Series
WET ELECTROSTATIC PRECIPITATOR
SYSTEM STUDY
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, 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-600/2-76-142
May 1976
WET
ELECTROSTATIC PRECIPITATOR
SYSTEM STUDY
by
John P. Gooch and Alan H. Dean
Southern Research Institute
2000 Ninth Avenue South
Birmingham, Alabama 35205
Contract No. 68-02-1313
ROAPNo. 21ACX-095
Program Element No. 1AB013
EPA Project Officer: Leslie E. Sparks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
This report describes a study of wet electrostatic precipitators
(WESP) through laboratory experiments, a field test of a full-
scale unit, interviews with manufacturers and users, and a
literature survey. Experiments were conducted with the objec-
tive of determining collection efficiency as a function of
particle size at varying current densities and specific collec-
tion areas. The results obtained were in reasonable agreement
with those predicted by simulating the experimental conditions
with a mathematical model. The feasibility of collecting S02
and particulate in a WESP was examined. As would be expected,
calculation of ion transport r^afij^Qi^cficate that only an in-
significant amount of SOa 'can be removed by selective ioniza-
tion of S02 molecules. Mass tr-aVisTer-'calculations indicated
that irrigated electrode surfaejers'^jpuld not provide sufficient
interfacial area for desired S02"removal levels. Based on
conventional geometry and stainless steel construction, a
WESP-scrubber device would be more costly than a scrubber-only
system because of high WESP capital costs. The effectiveness
of a WESP as a mist eliminator following a scrubber was shown
by calculation to be a function of the particle size distribu-
tion and the concentration of the mist to be collected. The
field test and literature survey indicated that WESPs have
been used effectively to control fine particle emissions in the
aluminum, iron and steel, chemical, and petrochemical industries.
11
-------
TABLE OF CONTENTS
Page
Abstract ii
List of Figures . iv
List of Tables ix
Sections
I Summary and Conclusions 1
II Introduction 3
III Laboratory Studies 4
IV Evaluation of Full-Scale Wet Electrostatic
Precipitator Installations 28
V Analysis of Potential Wet ESP Applications 53
VI Recommended Research and Development 69
VII Literature Study 72
VIII Summary of Status of WESP Technology and
Design Methods 168
IX Acknowledgments 170
X References 171
XI Appendix A 176
XII Appendix B 177
111
-------
FIGURES
No.
1 Plan and side views
2 Wet wall precipitator section at corona wire
3 Completed wet ESP test assembly
4 Gas burner assembly
5 Voltage-current relationship for section 2 of pilot
scale wet ESP 11
6 Inlet particle size distributions from Brink data 14
7 Experimentally measured and ideally calculated
fractional collection efficiencies in the labora-
tory precipitator 15
8 Experimentally measured, ideally calculated, and
sneakage-corrected fractional collection efficiencies
for a current density of 53.8 nA/cm2 and a gas ve-
locity of 0.72 m/sec in the laboratory precipitator 16
9 Experimentally measured, ideally calculated, and
sneakage-corrected fractional collection efficiencies
for a current density of 26.9 nA/cm2 and a gas ve-
locity of 0.72 m/sec in the laboratory precipitator 17
10 Experimentally measured, ideally calculated, and
sneakage-corrected migration velocities as a
function of particle size in the laboratory pre-
cipitator 19
11 Experimentally measured, ideally calculated, and
sneakage-corrected migration velocities as a
function of particle size for a current density
of 53.8 nA/cm2 and a gas velocity of 0.72 m/sec
in the laboratory precipitator 20
12 Experimentally measured, ideally calculated, and
sneakage-corrected migration velocities as a
function of particle size for a current density
of 26.9 nA/cm2 and a gas velocity of 0.72 m/sec
in the laboratory precipitator 22
IV
-------
FIGURES
(Continued)
13 Effective migration velocities from Climet and
Brink efficiency determination with DOP particu-
late at 50 nA/cm2 23
14 Effective migration velocities for SOa-DOP
particulate experiment (gas velocity = 0.76m/sec) 27
15 Schematic of scrubber-precipitator system and
sampling locations 29
16 Schematic of primary emission control system 30
17 Optical and diffusional sizing system 31
18 Andersen data on log probability co-ordinates 33
19 Cumulative size distributions on a number basis
for various industrial particulate sources as
measured by optical and diffusional methods 34
20 Measured fractional efficiencies for a wet electro-
static precipitator with the operating parameters
as indicated, installed downstream of a spray type
scrubber on an aluminum reduction pot line 35
21 Voltage current relationship (manual control) and
operating ranges (automatic control) 38
22 Discharge electrodes for wet ESP units at
British Steel 44
23 Lurgi discharge electrodes 48
24 Schematic diagram of scrubber-wet precipitator
system 54
25 Computer model projections for 50xlO~9 amps/cm2
and 50 KV 58
26 Computer model projections for 33xlO~9 amps/cm2
and 47.3 KV 59
27 Computer model projections for droplet size
distribution, 33x10 9 amps/cm2, 47.3 KV 66
28 Theoretical collection efficiencies as a
function of particle diameter 68
v
-------
FIGURES
(Continued)
29 A flow chart showing the steps in the steelmaking
process from the basic raw material and scrap in-
put to the finished product 73
30 Effect of burden improvement on dust rate from a
blast furnace to its dust collecting system 76
31 Typical wet type pipe precipitator cleaning blast
furnace gases 83
32 Flow diagram for wet cleaning iron blast furnace
gas with electrostatic precipitator 84
33 Blast furnace statistics for period 1920-1969 85
34 Distribution of precipitator inlet gas temperature
Blast furnace installations 86
35 Distribution of precipitator inlet dust loading
Blast furnace installations 87
36 Distribution of precipitator gas velocity
Blast furnace installations 88
37 Distribution of precipitator field strength
Blast furnace installations 89
38 Distribution of precipitator input power
Blast furnace installations 90
39 Design efficiency trends over the period 1939-1969
prorated on acfm basis for blast furnace instal-
lations 91
40 Relationship between collection efficiency and
specific collection area for electrostatic pre-
cipitators operating on blast furnace installations 92
41 Comparison of actual performance to design per-
formance on blast furnaces (Basis:ratio of mig-
ration velocities calculations using Deutsch
equation) 93
42 Dust output survey as a function of coke rate
and melt rate 106
43 Particle size studies made with dusts from hot
and cold blast cupolas 107
VI
-------
FIGURES
(Continued)
44 Hot-blast cupola plant (melting rate: 6m. ton/hr)
with wet electrostatic precipitator for cleaning
the waste gases 108
45 Waste and top gas cleaning with wet electrostatic
precipitator and venturi in series 110
46 Coke oven flow diagram 113
47 Typical pipe type electrostatic precipitator for
collection of tar 116
48 An integral tar collecting pipe type precipitator 117
49 Concentric ring detarrer 118
50 A process flow diagram of production of car-
buretted water gas 122
51 Oil shale retorting process 123
52 A single-stage vertical wire and pipe unit 124
53 Plate-type precipitator used for detarring 125
54 Typical flow chart for sulfur-burning contact
plant 130
55 Basic flow diagram of contact-process sulfuric
acid plant burning spent acid 132
56 Particle size distribution of sulfuric acid mist
from commercial contact plants 134
57 Wire-in-tube acid mist precipitator 135
58 Installed cost of sulfuric acid mist precipitators,
1965-1969 143
59 Flow diagram for typical thermal-process phosphoric
acid plant 146
60 Electrostatic precipitator used for PaOs mist
removal 148
61 Diagram of the Bayer Process 151
VI1
-------
FIGURES
(Continued)
62 Prebake Reduction Cell schematic arrangement 152
63 Vertical Stud Soderberg Cell schematic arrangement 153
64 Horizontal Stud Soderberg Cell schematic arrange-
ment 154
65 Particle size weight distribution, pot line
primary effluent 156
66 Composite particle size distribution by weight for
aluminum reduction cell air emissions (Kaiser
Aluminum and Chemical Corporation Plant at Tacoma,
Washington pot line 4) at the exit of the re-
duction cell 157
67 Anodic gas purification plant flow sheet 159
68 Cost data for wet precipitators 164
Vlll
-------
TABLES
No. Page
1 Typical inlet and outlet particle size data
for wet ESP experiments 13
2 SO2 removal experiments 25
3 Applied voltage in wet ESP test unit at
50 yA/ft2 current density 26
4 Mass train test results - 553 wet ESP 36
5 Precipitation rate parameters 41
6 Performance characteristics of wet precipitators
specified by British Steel 43
7 Mikropul wet electrostatic precipitator installations 50
8 Test results obtained by Research Cottrell on a
blast furnace wet precipitator installation 52
9 Power requirement comparison for particulate
controls in SOx removal systems 60
10 Capital cost estimates for particulate control
in particulate-SOx removal system 62
11 Operating cost estimates for particulate control
in particulate-SOx removal system $/yr 63
12 Size distribution of droplets at a spray scrubber
exit 65
13 Chemical analysis of blast furnace flue dusts 78
14 Weight percent composition of dust samples from
blast furnace gas cleaning plant 79
15 Size analysis of blast furnace flue dust 80
16 Electric arc steel furnace fume emission data 95
17 Electric furnace dust composition 96
18 Electric steel furnace fume particle size data 98
19 Fume cleaning for steelmaking processes 1969
(B SC) 99
IX
-------
TABLES
(Continued)
20 Chemical composition of cupola dust 104
21 Dust content of cupola waste gases 105
22 Cost comparison for various dust collectors for
cleaning waste gases of a hot blast cupola 111
23 Typical coke oven gas precipitator design para-
meters 115
24 Average design parameters for coke oven precipi-
tators 119
25 Precipitator economics for detarring of coke
oven gas (time period 1959-1969) 119
26 A summary of design data of electrostatic precipi-
tators on various applications for removing tar
and oil mist over period 1940 to 1963 128
27 Precipitator inlet mist loading
Summary of performance statistics sulfuric acid
mist precipitator (1945-1969) 137
28 Precipitator inlet gas temperature
Summary of performance statistics sulfuric acid
mist precipitator (1945-1969) 138
29 Precipitator input power
Summary of performance statistics sulfuric acid
mist precipitator (1945-1969) 139
30 Precipitator field strength
Summary of performance statistics sulfuric acid
mist precipitator 140
31 Precipitator gas velocity
Summary of performance statistics sulfuric acid
mist precipitator 141
32 Summary of sulfuric acid mist precipitator costs
1960-1969 142
33 Distribution of ratio (R) for sulfuric acid
mist precipitators (1945-1969) 144
-------
TABLES
(Continued)
34 Plant capacity for manufacturing primary
aluminum 149
35 Typical performance data of wet precipitator
at the Dalles 160
36 Control equipment considered for the primary
aluminum industry 161
37 Results of EPA source tests fluoride and par-
ticulate primary aluminum industry 162
38 Legend for tables and figures 163
39 Typical electrical conditions for wet pre-
cipitator at the Dalles 165
40 Emission measurements on electrostatic pre-
cipitator No. 1, Martin Marietta Plant at
Goldendale, Washington 166
41 Summary of measured emissions 167
XI
-------
SECTION I
SUMMARY AND CONCLUSIONS
A flexible laboratory scale electrostatic precipitator was
designed and constructed to (1) obtain performance data on
fine particle collection under idealized conditions and (2)
to evaluate the feasibility of incorporating a wet precipitator
scrubber-device in a gas cleaning system for coal-fired boilers.
Experiments were conducted with the objective of determining
collection efficiency as a function of particle size at varying
current densities and specific collection areas using dioctyl-
phthalate aerosol. The results obtained were in reasonable
agreement with those predicted by simulating the experimental
conditions with a mathematical model of electrostatic precipi-
tation.
Calculations were performed to determine the maximum amount of
S02 which could be transported to the collection electrode by
selective ionization of SOa molecules. As would be expected,
the results indicated that an insignificant fraction of the
SOa can be removed by this mechanism. Mass transfer calculations
were also performed to estimate the SOa which could be absorbed
by a basic solution irrigating the collection electrodes. It
was found that the interfacial area provided by the collecting
electrodes was insufficient to obtain significant S02 removal levels,
Experiments conducted in the laboratory test unit with NaOH
solution irrigating the collecting electrodes indicated that
no significant increase in SOa removal rates occurred as a re-
sult of energizing the precipitator power supplies. Simultaneous
collection of SO2 and OOP aerosol in the laboratory test unit
indicated a decrease in SOa removal, presumably caused by a
decrease in electrode wettability.
Cost estimates were made for a full-scale wet ESP-scrubber device
for collection of S02 and particulate in a power plant flue gas
cleaning system. The estimates indicated that, although the
WESP system would use less energy, a scrubber-only system is
more economical to operate because of the high capital costs
(using stainless steel construction and conventional electrode
geometries) of a WESP. Electrode scaling with the WESP is a
potentially serious problem which would require pilot plant
trials for an adequate evaluation.
The effectiveness of a wet electrostatic precipitator as a mist
eliminator following a scrubber was shown by calculation to be
a function of the particle size distribution and the concentration
of the mist to be collected. High droplet concentrations, such
as those existing at the outlet of a spray scrubber, may limit
the performance of a WESP through space charge suppression of
the corona current.
-------
A field test of a full scale wet electrostatic precipitator
was conducted, and an analysis of the results indicated that
the collection efficiency as a function of particle size re-
lationship is in fair agreement with the predictions of the
mathematical model over the particle diameter range 0.2 ym to
1.3 ym. For larger particles, the measured collection effi-
ciencies were significantly lower than the theoretical pre-
dictions, possibly because of liquor carryover from the
electrode irrigation system. The WESP was collecting fume
from an aluminum pot line. Overall mass collection efficiencies
on the fume, in which about 65% of the mass consisted of sub-micron
particles, ranged from 95.0 to 98.0% at a specific collecting area
of about 62 m2/(m3/sec). Corrosion and deposit formation in the
liquor supply system have occurred at this installation.
An evaluation of current WESP technology was conducted by a lit-
erature survey and by contacts with manufacturers and users. Both
plate and pipe collecting electrodes have been effectively utilized
in collecting fumes and mists from metallurgical and chemical pro-
cesses. The largest recent application of WESP has been in the
collection of fume from aluminum pot lines. Duct-type collecting
electrodes are generally used for the application in which rela-
tively large gas volume flows are to be cleaned. The advantages
of plate electrodes as opposed to pipe electrodes are less expen-
sive construction and the flexibility of allowing variations in
electrical sectionalization. Pipe electrodes can be more uniformly
irrigated than plates, and gas by-passage (sneakage) is not a
factor. Wire-pipe WESP's usually consist of only one electrical
section in the direction of gas flow.
-------
SECTION II
INTRODUCTION
This report describes a study of wet electrostatic precipi-
tator technology which was conducted with the following
objectives: (1) develop methods for calculating performance
of WESP's, (2) evaluate the potential of a wet electrostatic
precipitator as a device for collecting particulate and sulfur
oxides, or mists exiting a scrubber, in a gas cleaning system
for a utility power station, and (3) evaluate the status of
wet precipitator technology. Wet electrostatic precipitators,
for the purposes of this study, are defined as conventional
electrostatic precipitator geometries, consisting of either
wire-pipe or wire duct design, in which either a liquid or
solid phase particulate is collected with electrode irrigation,
or in which a liquid phase aerosol is collected without electrode
irrigation.
An electrostatic precipitator collecting a liquid aerosol under
laboratory conditions offers an opportunity to study performance
in the absence of unmodelled processes such as particle reentrain-
ment and back corona. Therefore, a laboratory-scale wet ESP
can be used to evaluate mathematical procedures for modelling
particle collection. The following sections describe experiments
performed with a laboratory-scale WESP, and compare the experi-
mental results with those obtained from a mathematical model
developed under other EPA contracts. The model is then used
to evaluate potential WESP applications in utility power
plant gas cleaning systems.
The status of existing wet ESP technology was evaluated by
conducting a performance test on a recently installed WESP in
the aluminum industry, and by conducting a literature survey.
Information was also obtained from WESP manufactures and users.
-------
SECTION III
LABORATORY STUDIES
DESIGN OF BENCH-SCALE TEST UNIT
A flexible test unit capable of treating about 0.0737 m3/sec
of synthetic flue gas (156 ft3/min)* was designed with the
following objectives in mind.
To obtain fundamental performance data for
particulate collection, especially in the
diameter range of 2.0 pm and smaller.
To evaluate the feasibility of incorporating
a wet precipitator scrubber device in a gas
cleaning system for coal-fired boilers.
The general arrangement of the test facility and a side view
with dimensions are given in Figure 1. Plate-to-plate spacing
is 12.7 cm (5 in.) and the collecting electrode area is 2.32 m2
(25 ft2). A gas velocity of 1.52 m/sec (5 ft/sec) gives a flow
rate of 0.0737 m3/sec which results in a Reynolds number of
about 16000 (well within the turbulent flow region) and an A/V
ratio of 31.5 m2/ (m3/sec) (160 ft2/1000 cfm) . The test unit
can also be assembled with plate-to-plate spacings of 18.75 cm
and 25.4 cm.
The test unit has sufficient space available to accommodate
an assembly consisting of alternating scrubbing and collecting
sections, with the spray sections comprising one-half the length
of the collecting sections. Under this contract, the spray
sections were not utilized. Figure 2 is a detailed cross-
section of the unit at a corona wire in a precipitation section.
The numbers in circles are call-out numbers which refer to de-
tails in detail drawings that are not shown.
The high-voltage insulators for the test unit are designed to
isolate the discharge electrodes in the presence of a wet en- .
vironment. This was a difficult design problem for a test unit
designed to obtain fundamental performance data, since bleed
air to dry the insulators and large open spaces around the in-
sulators are very undesirable from the standpoint of interpreting
experimental data. Originally, the plan was to isolate Teflon
feed-through insulators at the top of the assembly in a plexiglas
conduit purged with heated air. However, water drops from the
spray nozzles could deposit on the plexiglas, and the conduit
*Environmental Protection Agency policy is to express all measure-
ments in agency documents in metric units. When implementing this
practice will result in undue costs or lack of clarity, conversion
factors are provided for the non-metric units used in the report.
Conversion factors are given in Appendix A.
-------
r
-L o o 1 U ol* »«
** 1 "1 -H- J-
... ... - - , _,.
U* * »
. ~_r
~JTJ
lTTt
^
G
Figure 1. Plan and Side Views of Pilot ESP Test Facility
-------
could not be maintained at a temperature level sufficient to
insure quick evaporation. As a result of the above potential
hazard, an alternative concept of using a heated epoxy insulator
was designed.. The epoxy is a formulation which has a high thermal
conductivity and a low electrical conductivity. The heater
embedded in the epoxy was designed to maintain the temperature
of the insulator at about 250°F. Figure 2 shows a cross-
section of the cast epoxy insulator, which is used to jacket
the Teflon feed-through insulator surrounding the corona wire.
This design was only partially successful because of problems
with air bubbles in the casting and cracks forming in the
epoxy during operation of the internal heaters. The test unit
can at present be operated for short periods of time with
electrode irrigation with this design, but long-term operation
is not possible due to spray droplet accumulation on the in-
sulator surfaces. A solution to the problem would require a
design that would enable higher insulator surface temperature
to be maintained.
Although Figure 2 shows the plates irrigated.by sprays and
weirs, only the sprays were used due to non-uniform flow along
the length of the weirs. Experiments with the collecting elec-
trode spray assembly indicated the expected difficulty in main-
taining uniform sheet flow on the electrode surfaces. Polishing
the plate surface with emery cloth was found to be beneficial,
but repolishing was necessary immediately prior to spraying the
plates with water, apparently because of a change in the wet-
tability of the surfaces upon exposure to the atmosphere. In
an attempt to eliminate the time variation of the wettability,
a sample plate was sandblasted with medium grit sand. This
procedure resulted in much improved plate wetting, and the
wettability did not decrease upon exposure to the atmosphere.
All plates were therefore sandblasted prior to their installation
in the test unit.
Photographs of the completed installation are shown in Figures
3 and 4. Three electrical sections are employed, the first
two of which each consists of 0.581 m2 (6.25 ft2); the third
power supply feeds 1.162 m2 (12.50 ft2). Figure 4 shows a side
view of the test unit, including the heater assembly, spray
towers, a mixing chamber for studying condensation growth (con-
structed as part of another project), the water piping and
drainage system, the power supplies, inlet sampling ports, and
the electrified sections. Figure 4 shows a front close-up view
of the gas-fired burner assembly, including a hopper for feeding
solid particulate, and a steam venturi-ejector device which can
be used for redistribution of particulate. The burner assembly
has been equipped with a commercial hot water controller with
proportional band control on a pressure reducer in the gas supply
line. The pressure is controlled by a temperature bulb in the
warm water outlet of the scrubber. The controller also has a
thermostatically controlled pilot and main gas valve to shut
down in case of flame failure.
Voltage-current curves were obtained with and without water
-------
Figure 2. Wet Wall Precipitator
Section at Corona Wire
-------
-
-'
:
'
^.
I
H
(T
rt
n
C-
M
CO
rr
(0
ra
rt
CO
-------
Figure 4. Gas Burner Assembly
-------
irrigating the collecting electrodes, and with and without the
insulator heaters energized. Normal current-voltage relationships
were obtained with and without electrode irrigation when the
insulator heaters were connected to the AC power source through
an isolation transformer. Without the transformer, sparking
occurred to the epoxy insulator surfaces at relatively low
voltage. A typical voltage-current curve is shown in Figure 5.
Electrode irrigation causes the v-l curve to shift to the left.
Normally, increasing the water vapor content of an air-water
vapor mixture would be expected to shift the V-I curve to the
right. The shift toward higher current for a given applied volt-
age indicated for wet operation in Figure 5 may be caused by
leakage paths along the surface of the insulators due to moisture
accumulation, or an effective decrease in the wire to plate
spacing due to the presence of the electrode irrigation stream in
the upper portion of the inter-electrode region.
PARTICULATE COLLECTION EXPERIMENTS
Procedure
A series of particulate collection experiments was conducted for
the purpose of determining collection efficiency as a function
of particle size under idealized conditions. A liquid-phase
aerosol (dioctyl-phthalate or DOP) was chosen for the experiments
to eliminate the possibility of reentrainment. The particulate
generator consisted of a single atomizer which produced a stable
polydisperse aerosol. Electrode irrigation and humidification
were not used, and the collected oil droplets drained from the
collection electrodes by gravity. The carrier gas was ambient
air.
Particle size and concentration measurements at the inlet and
outlet of the test unit can be made with either an optical par-
ticle counter or with inertial impactors. In order to obtain
reliable particle size information with an optical particle
counter, however, it is necessary to obtain a calibration curve
for the instrument with the aerosol under investigation. An
attempt was made to obtain such a curve using a vibrating or-
ifice particle generator. However, due to problems with the
particle generator, we were unsuccessful in obtaining a useful
calibration curve over the particle size range of interest.
Therefore, it was decided to conduct the majority of the ex-
periments using impactors to determine particle size distributions
and fractional efficiencies. A small number of experiments were
conducted using a Climet-Ortec particle size analyzer and a
calibration curve obtained with polystyrene latex beads. These
measurements are compared with those made with the impactors
in a subsequent portion of this section.
10
-------
1000
900
800
700
i i i i i i i i
12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
400
Figure 5. Voltage-Current Relationship for Section 2
of Pilot Scale Wet ESP
11
-------
A five-stage Brink impactor was used at the precipitator inlet
and outlet. Aluminum foil substrates were employed to facilitate
weighing of the small masses collected on the individual stages,
and the mass determinations were made with a Cahn Electrobalance.
For the experiments conducted at the lower gas velocity, a
sampling time of six hours was required at the precipitator
outlet to obtain weighable quantities on the impactor substrates.
Six and four point traverses were conducted with the impactor
at the inlet and outlet, respectively, in order to insure that
a representative size distribution was obtained at each location.
Since the particulate from an electrostatic precipitator will
have an electrical charge, it was necessary to employ a charge
neutralizer at the inlet of the sampling line for all measure-
ments made at the precipitator outlet. Measurements with the
optical particle counter indicated that, if metal substrates
were used in the impactor, the charge acquired by the partic-
ulate substantially influenced the size distribution obtained
with the impactor. Since the impactor was calibrated with an
uncharged aerosol, the use of the charge neutralizer minimized
sizing errors due to electrostatic attraction.
Table 1 gives a typical set of inlet and outlet data obtained
from duplicate experiments at the indicated conditions. Good
reproducibility was obtained for all stages at the inlet, and
for stages 2 through 5 at the outlet. Isokinetic sampling was
not employed, since the effect of anisokinetic conditions on
particles with diameters less than 2.0 ym is expected to be
negligible. The effect of changing the gas flow rate through
the impactor on the indicated size distribution was checked by
reducing the flow rate from 56.6 cm3/sec to 28.3 cm3/sec. The
higher flow rate, which was used for all of the fractional
efficiency determinations, resulted in a gas velocity of 0.79
m/sec at the sample point. Figure 6 shows the inlet size dis-
tributions from Table 1 and the distribution obtained at the
same location with the reduced impactor flow rate. These data
indicate that the same size distribution is obtained at either
flow rate. The higher flow rate is more desirable because of
the smaller cut points and the increased amount of gas that may
be sampled per unit time.
Results
Fractional efficiency measurements were performed with the
Brink impactor and DOP aerosol at current densities of 26.9,
53.8, and 107.5 nA/cm2 at a gas velocity of 1.37 m/sec, and at
current densities of 26.9 and 53.8 nA/cm2 at a gas velocity
of 0.72 m/sec. The results obtained from these experiments,
expressed as collection efficiency as a function of particle di-
ameter, are presented in Figures 7, 8, and 9. Computed collection
efficiencies obtained with a mathematical model are also shown.l
In view of the difficulties in making these types of measurements,
the agreement between measured and calculated efficiencies shown
in Figure 7 is considered good. Figures 8 and 9 suggest that
12
-------
TABLE I
TYPICAL INLET AND OUTLET PARTICLE SIZE DATA FOR WET ESP EXPERIMENTS
(OOP Sprayer, Gas Velocity = 1.37 m/sec, Current Density = 55 na/cm2)
Mass Loading, mg/am3
Geometric
Test No.
Stage
Lower Diameter
Limit, yim
3.04
1.80
1.24
0.65
0.46
2ai
26.341
18.659
7.249
7.448
2.954
2bi
28.956
13.526
6.285
6.984
1.775
Inlet
2ci
24.596
14.128
7.238
6.743
2.514
2di
24.088
15.226
8.226
7.419
2.654
Average
25.995
15.385
7.250
7.148
2.474
2ao
0.0414
0.1494
0.2575
0.5724
0.3862
Outlet
2 bo
-\-o
0.1103
0.2115
0.4782
0.3770
Average
0.207
0.1299
0.2345
0.5253
0. 3816
D i ame te r , ]
ym
2.34
1.49
0.90
0.55
_oiiectior
Efficiency
%
99.16
96.77
92.65
84.58
-------
lAMETER, Micrometers
C
b c
PARTICLE D
O
O-
A
cfc
D-.N
0
AL
D
I A c
v ^
O e
f
A;
>FROM TABLE 1
REDUCED FLOW RATE
1
&
AC$
3
AH
>
^
/
D
O
|
01 O.I 1 10 50 9C
% SMALLER THAN INDICATED SIZE
Figure 6. Inlet Particle Size Distributions from
Brink Impactor Data
14
-------
qq q , _, , , , , , ,
*v-^ ' T ' ' i I I I ^ / A ^Tl ' ' ' '
99.8
99.5
8^ 99.0
o
z
UJ
g 98.0
u.
Ul
UJ 95.0
O
O
90.0
80.0
70.0
60.0
j = 26.9 no/cm2 , v = 1.37 m/sec
A j =53.8 no/cm^ , v =1.37m/sec
j = 107.5 na/cm2,v = 1.37 m/sec
O
^
D
{EXPERIMENTALLY )
MEASURED AT THE >
RESPECTIVE j and v J
I I I I I I I I I I I I I I I II
O.I 1.0 10.0
PARTICLE DIAMETER, Micrometers
Figure 7. Experimentally Measured and Ideally
Calculated Fractional Collection
Efficiencies in the Laboratory
Precipitator
15
-------
99.98
99.95
99.9
8*
x^
EFFICIENCY
CO
U)
'oo
z
o
H 99.5
o
LU
0
0
99.0
98.0
95.0
900
0
1 1 1 | 1 1 1 1 1 | . | | 1 1 II
THEORETICAL | '
l-Srcm/Sf / 'CORRECTED
/ / FOR 8 %
/ / SNEAKAGE
~ .//""
THEORETICAL / '
0 EXPERIMENTAL / /
/ .'0 EXPERIMENTAL
// -
/ /0
J /*
//'
// ~
Sf -
1 1 1 1 1 1 II 1 1 1 1 1 1 1 1
.1 1.0 I0.(
PARTICLE DIAMETER, Micrometers
Figure 8. Experimentally Measured, Ideally
Calculated, and Sneakage-Corrected .
Fractional Collection Efficiencies
for a Current Density of 53.8 nA/cm2
and a Gas Velocity of 0.72 m/sec in the
Laboratory Precipitator
16
-------
99.99
99.98
99.95
99.9
o
o
LL
U.
LJ
2
O
1-
o
LU
_J
O
O
99.8
99.5
99.0
98.0
95.0
90.0
O.I
j = 26.9 no/cm2
v =0.72 m/sec
THEORETICAL
OEXPERIMENTAL
I I I | IT TT
THEORETICAL
CORRECTED FOR
8% SNEAKAGE
EXPERIMENTAL
1.0
PARTICLE DIAMETER, Micrometers
10.0
Figure 9. Experimentally Measured, Ideally
Calculated, and Sneakage-Corrected
Fractional Collection Efficiencies for
a Current Density of 26.9 nA/cm2 and a
Gas Velocity of 0.72 m/sec in the
Laboratory Precipitator
17
-------
the agreement obtained between computed and measured results
worsens with decreasing current density. The assumption of 8%
gas sneakage over four stages improves agreement between computed
and measured results at a gas velocity of 0.72 m/sec. However,
the data obtained at this lower gas velocity with 26.9 nA/cm2
are considerably below the computer projections. Possible
causes for this lack of agreement are unmodelled effects such
as non-uniform current density and electric field, and particle
concentration gradients in the interelectrode space.
The performance of an electrostatic precipitator for monodisperse
particles may be expressed in terms of an effective migration
velocity, defined by
Q n / 100 ,
we = A ln (I00^7) '
where
w = effective migration velocity of the particle
size under consideration, cm/sec
= ratio of volume flow to plate area, m3/sec/m2
n- = collection fraction of the particle size under
consideration.
Figures 10, 11, and 12 give effective migration velocities ob-
tained from the efficiency data in Figures 7, 8, and 9.
Figure 10 indicates good agreement between the computed and
experimentally derived values of effective migration velocity.
The assumption of 8% sneakage over four stages improves agree-
ment between the computed and experimental results only for the
lowest current density. Figure 11 indicates that, at the lower
gas velocity of 0.72 m/sec, the correction for 8% by-passage
gives agreement with the experimental results. At 26.9 nA/cm2
and 0.72 m/sec, however, Figure 12) the experimentally derived
values for effective migration velocity are between 7.7 and
25.5% lower than the corrected computed values. Due to the
degree of uncertainty in the data, additional experiments are
needed for definite conclusions regarding confidence limits for
the model predictions. However, these comparisons suggest that,
although the model closely approximates the experimental results,
it tends to overpredict performance at lower values of current
density and gas velocity, and underpredict performance for the
highest value (107.5 nA/cm2) of current density.
Additional work is planned under EPA support to evaluate the
model predictions with varying electrode geometries and operating
conditions.
Figure 13 is a comparison of results obtained with the Brink
impactor and with a Climet-Ortec particle size analyzer. The
particle size information obtained with the Climet-Ortec is
based on a calibration curve obtained with polystyrene latex
beads (PSL). Since the measurements obtained with the Climet
18
-------
100.0
o
I
o
o
o
UJ 10.0
z
g
cr
1.0
I I
1 I I I
j=26.9na/cm2,v = l.37m/sec
A j = 53.8 no/cm2, v = 1.37 m/sec
j =107.5 na/cm2,v= 1.37 m/sec
O ( EXPERIMENTALLY
£ < MEASURED AT THE
D ( RESPECTIVE j and v
O.I
COMPUTED WITH 8% SNEAKAGE
ASSUMED OVER 4 STAGES
1.0
PARTICLE DIAMETER, Micrometers
10.0
Figure 10. Experimentally Measured, Ideally
Calculated, and Sneakage-Corrected
Migration Velocities as a Function of
Particle Size in the Laboratory
Precipitator
19
-------
100.0
o
0>
o
o
o
10.0
oc
I
1.0
O.I
i i i
i i i
\iii i i i_
j = 53.8 no /cm2, v =0.72 m/sec
THEORETICAL
THEORETICAL
O EXPERIMENTAL
EXPERIMENTAL
^_ CORRECTED FOR
"* 8% SNEAKAGE
i i i
i i l i i i i
i.o
PARTICLE DIAMETER,Micrometers
10.0
Figure 11,
Experimentally Measured, Ideally
Calculated, and Sneakage-Corrected
Migration Velocities as a Function
of Particle Size for a Current Density
of 53.8 nA/cm2 and a Gas Velocity of
0.72 m/sec in the Laboratory Precipitator
20
-------
ILOCITY, cm/sec
5 c
3 C
MIGRATION VE
C
3 C
0
1 1 1 | | 1 1 1 I | I | 1 1 1 1_
j =26.9 no/cm2, v=0.72m/sec ~
THEORETICAL
THEORETICAL '
0 EXPERIMENTAL /
<£*
S-s' CORRECTED FOR -
4^' 8% SNEAKAGE -
'* o
JtT*' - EXPERIMENTAL
J&
o*$ *
V&*'*
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
.1 .0 10
PARTICLE DIAMETER, Micrometers
Figure 12.
Experimentally Measured, Ideally
Calculated, and Sneakage-Corrected
Migration Velocities as a Function of
Particle Size for a Current Density
of 26.9 nA/cm2 and a Gas Velocity of
0.72 m/sec in the Laboratory Precipitator
21
-------
O
0)
m
E
O
O
O
d
z
O
o
5
UJ
o
UJ
LL.
U.
UJ
20
15
10
O.I
o o
D GAS VEL = 1.4 m/sec ,7.5% OOP SOLN
O GAS VEL = 0.73 m/sec, 7.5% OOP SOLN
£ GAS VEL = 0.73 m/sec , 100% OOP SOLN
" I'' =n73,7m/rC h«CTOR DATA
GAS VEL = 0.72 m/sec I
-CLIMET DATA
I
I
I I I
0.5
PARTICLE DIAMETER,
1.0
2.0
3.0
Figure 13. Effective Migration Velocities from Climet and
Brink Efficiency Determination with OOP Particu-
late at 50 na/cm2.
22
-------
were accomplished with the use of a diluter with an intake
several seconds in residence time downstream from the aerosol
generator, the concentration of DOP in the solution used for
the particle generator should not have an influence on the
indicated particle sizes. The comparison in Figure 13 in-
dicates fair agreement between effective migration velocities
derived from measured efficiencies obtained with the Climet-
Ortec and with the impactor. This suggests that the calibration
curve obtained for the optical instrument with PSL is also rea-
sonably accurate for a DOP aerosol.
SULFUR OXIDE AND PARTICULATE COLLECTION EXPERIMENTS
Since wet precipitator-scrubber systems have been proposed for
removing particulate and sulfur oxides from stack gases, it is
of interest to examine the simultaneous removal of DOP particulate
and S02 in the laboratory test unit. Section V will present
calculations which indicate that no significant additional S02
removal should be anticipated in a wet ESP due to the presence
of a corona current and an electric field. The experiments re-
ported in this section were limited in scope and were conducted
primarily for the purpose of determining the validity of this
conclusion.
The wet precipitator test unit was equipped with a positive dis-
placement metering pump and a static mixer for metering and
mixing a concentrated solution of sodium hydroxide into the
water used for irrigating the collection electrodes. Sulfur
dioxide was metered from a gas cylinder into the test unit up-
stream from the mixing chamber. Ambient air was used as the
carrier gas at a flow velocity of 0.76 m/sec and the S02 was
fed at a rate which resulted in a concentration of about 2000
parts per million by volume. The liquid to gas ratio was held
constant at 2.8 1/m3 (20.7 gal/1000 ft3)* and the residence time
of the gas in the irrigated and electrified region was about
four seconds.
The following groups of experiments were conducted:
1. The S02 removal efficiency of the device was determined
with only H20 irrigating the electrodes with the elec-
trified sets deenergized.
2. The S02 removal efficiency was determined with NaOH
solution irrigating the electrodes with the electrical
sets deenergized.
3. The S02 and particulate removal efficiency was determined
using DOP particulate with NaOH solution irrigating the
electrodes with the electrical sets energized.
23
-------
The S02 removal efficiencies obtained in these experiments are
given in Table 2. The molar ratio of NaOH to S02 entering the
test unit was 2.68 to 1 for experimental groups 2 and 3; there-
fore, the degree of removal in these experiments is governed by
mass transfer dynamics instead of chemical equilibria. Mass
transfer calculations that will be discussed in Section V in-
dicate that only a small portion of the absorption shown in
Table 2 can be accounted for by the area of the collecting elec-
trodes , and that electrical transport of S02 would be expected to
have a negligible effect on the overall SOa removal efficiency.
The major portion of the SOa absorption in these experiments was
apparently achieved by droplets generated by the electrode ir-
rigation manifolds. Table 2 shows that the SOa absorption de-
creased for the experiments conducted with DOP particulate under-
going collection. It is hypothesized that the indicated decrease
of S02 uptake was caused by a noticeable decrease in the wet-
tability of the collection electrodes resulting from the collection
of oil droplets on the electrodes.
These limited experiments tend to confirm the hypothesis that
electric fields and corona current do not significantly improve
the SOa uptake capability of a wet ESP simultaneously collecting
particulate and S02. Unfortunately, the conclusion is clouded
somewhat by the decrease in absorption area caused by the DOP
particulate. It may be argued, however, that plate wettability
is also likely to decrease under practical operating conditions
due to residual material remaining on the plates.
Particulate collection efficiencies were measured with the Climet-
Ortec particle size analyzer during the S02-DOP particulate ex-
periments. Effective migration velocities calculated from these
efficiencies are presented in Figure 14, and the applied voltages
for the "dry" and "wet" particulate collection experiments at a
current density of 50 yA/ft2 are given in Table 3. A comparison
of the data in Figures 13 and 14 indicate that the particulate
collection performance of the test unit was not affected, within
the uncertainty of the measurement techniques, by the presence of
S02 and the use of electrode irrigation.
24
-------
TABLE 2
S02 REMOVAL EXPERIMENTS
1. Average inlet concentration = 1979 ppm
1S02
SO2 Outlet Removal
Experiment Group Concentration Efficiency
No. (see text) ppm by volume %
(1) 1300 34.3
1379 30.3
1326 33.0
(2) 1015 48.7
960 51.5
1024 48.3
(3) 1172 40.8
1109 44.0
25
-------
TABLE 3
APPLIED VOLTAGE IN WET ESP , TEST UNIT AT
50 yA/ft2 CURRENT DENSITY
kV
Dry Walls, Wet Walls,
Power Supply Ambient Air, Ambient Air & S02/
No. OOP Particulate OOP Particulate
1 33 25
2 32 33
3 31.5 31
26
-------
o
e>
CO
£
o
^
H
0
O
UJ
z
g
i
5
tr
o
UJ
UJ
u.
li
U.
UJ
C\J
10
9
8
7
6
5
4
3
2
1 1 | 1 1 1 1 1 1
O
0
O
- o -
o -
- O
"~ ~"
1 1 1 1 1 1 1 1 1
0.2 0.3 0.4 0.5 0.6 0.8 1.0 2 3 ^
PARTICLE DIAMETER,
Figure 14. Effective Migration Velocities for
DOP Particulate Experiment (gas Velocity
0.76 m/sec).
27
-------
SECTION IV
EVALUATION OF FULL-SCALE WET
ELECTROSTATIC PRECIPITATOR INSTALLATIONS
FIELD TEST MEASUREMENTS
Introduction
Fractional and overall particulate collection efficiency mea-
surements were made on a plate-type wet electrostatic precipitator
collecting fume from an aluminum pot line consisting of horizontal
stud self-baking aluminum reduction cells. These measurements
were conducted under another contract and reported in detail
elsewhere.2 Comparisons were made between measured (with Andersen
impactors) and predicted collection efficiencies obtained from a
mathematical model of an electrostatic precipitator.
Figure 15 shows the arrangement of the unit on which the test
series was conducted, and Figure 16 is a schematic of the liquor
flow through the system as given by Bakke.3
Measurement Techniques
Particle Sizing -
Particle size and concentration measurements were conducted
using the following methods:
1. Diffusional techniques using condensation nuclei counters
and diffusion batteries for determining concentration
and size distribution on a number basis for particles
having diameters less than approximately 0.2 ym.
2. Optical techniques for determining concentrations and
size distributions for particles having diameters be-
tween approximately 0.3 ym and 1.5 ym. Figure 17
shows the optical and diffusional sizing system.
3. Inertial techniques using cascade impactors for de-
termining concentrations and size distributions on a
mass basis for particles having diameters between
approximately 0.25 ym and 5.0 ym. Andersen impactors
were used simultaneously at the precipitator inlet and
outlet. Extremely low mass loadings at the outlet made
necessary the operation of the impactors for approximately
16 hours to obtain weighable quantities of particulate.
Mass Loading Measurements -
A modified EPA sampling train with an in-stack filter holder was
28
-------
VO
PLATFORM,
VERT.
SAMPLING
SCRUBBERS
WET
ELECTROSTATIC
PRECIPITATOR
Figure 15. Schematic of scrubber-precipitator system and sampling locations
-------
STACK
u>
o
LIQUOR.
MAIN
3 T-R SETS
VALVE
POT GAS MANIFOLD
*a I&
WEP SPRAYS
INLET DUCT
RECEIVING
.TANK
SCRUBBER
SPRAYS
MAIN FAN
BOOSTER
PUMP
7/777
CYCLONIC
SCRUBBERS
(TWO)
LIQUOR
RETURN
Figure 16. Schematic of primary emission control system1
-------
Flowmeters
Cyclone Pump
Process
Exhaust
Line
.v-o
Charge
Neutral!zer
Cyclone
(Optional)
1J N Flowmeter
Particulate
Sample Line
Orifice
Diffusion
Battery
CN Counters
Aerosol
Photometer
Diffusional Dryer
(Optional)
Manometer
Recirculated
Clean Dilution
Air
Filter
Charge
Neutralizer Pressure
Balancing
Line
Pump
Bleed
Figure 17. Optical and diffusional sizing system
31
-------
used for the mass loading measurements, which were conducted
at the inlet and outlet simultaneously with the impactor runs.
An isokinetic traverse across the stack was conducted at both
the precipitator inlet and outlet through a single sampling
port at each location except for the last day, when a single
point mass determination was performed at the outlet.
Results
The mass median diameters of both inlet and outlet distributions
as determined by the Andersen impactors are less than 1.0 ym.
Figure 18 presents the Andersen data on log probability co-
ordinates.
Figure 19 gives the cumulative size distribution on a number
basis for this test series and various other industrial parti-
culate sources as measured by optical and diffusional methods.
Fractional efficiencies were computed from the optical and dif-
fusional data, based on inlet measurements conducted on August
20 and 21, and outlet measurements conducted on August 22 and 23,
1974. Figure 20 gives the results of these calculations, together
with the inertially determined fractional efficiencies. A
theoretically calculated fractional efficiency curve is also
given, and this will be discussed later.
Mass Loading Measurements -
Mass train measurements were performed by Guardian Systems, Inc.
of Birmingham, Alabama, under subcontract to Southern Research
Institute and these results are given in Table 4. Fair agree-
ment was obtained between the total mass loading with Andersen
impactors and the mass train at the inlet, but severe disagree-
ment occurred at the outlet.
The total mass obtained with a traverse using the mass train
at the outlet was greater than that collected with the impactors
by a ratio of approximately 5 to 1. When the mass filter was
operated near the center of the stack and near the sampling
location used for the impactors, the disagreement was reduced
to a ratio of about 3 to 1. A comparison of outlet loadings
however, indicates that the mass train results obtained during
this test series are in fair agreement with those obtained
previously by a local pollution control agency. Both the
Andersen data and the mass data showed good reproducibility.
It is our conclusion that the most probable cause of the mass
loading discrepancy is the collection of large water droplets
containing solids by the mass filter. Such droplets would be
subject to stratification in the stack, and this is qualitatively
indicated by the decrease in loading which occurred when the mass
train was operated at a single point. Additional work with a
32
-------
E
a.
g 1.0
u 0.8
5 0.6
0.4
0.2
0-1
INLET
OUTLET
12 5 10 20 30 40 50 80 90 95 98 99
% SMALLER THAN INDICATED SIZE
Figure 18.
Andersen data on log probability
co-ordinates
33
-------
108
107
o
UJ
O
o
UJ
p
-I
o
2
O
105
104
I03
102
0.
SUBMERGED
ARC FERRO-
ALLOY FURNACE
OPEN HEARTH
FURNACE
INLET TO WET ESP
AT ALUMINUM
PLANT
\ S02 BUBBLE
\CAPSCRUBQEB_
VPACKED BED
A S02 SCRUBBER
\
I
01
Figure 19,
O.I
1.0
5.0
PARTICLE DIAMETER ,um
Cumulative size distributions on
a number basis for various
industrial particulate sources
as measured by optical and
diffusional methods
34
-------
u>
Ul
s*
UJ
o
u.
u.
UJ
o
£5
H
8
99.98
99.9
99.8
99.5
99
98
95
90
60
30
0
1 ' ' / * ' '
/CALCULATED WITH COMPUTER MODEL
/
* / *
*/° * °
_ \^tT ° ^ *
4 *
MEASUREMENT METHOD:
A CASCADE IMPACTORS
0 OPTICAL PARTICLE COUNTERS
+ DIFFUSIONAL
- PRECIPITATOR CHARACTERISTICS:
TEMPERATURE -4I°C
SCA - 62 m2/(m3/sec)
CURRENT DENSITY - 30 nA/cm5
III II
^^
5~
A
"
-~
>
05 O.I 0.5 .0 5.0 I0.<
PARTICLE DIAMETER ,;im
Figure 20.
Measured fractional efficiencies for a wet electrostatic
precipitator with the operating parameters as indicated,
installed downstream of a spray type scrubber on an
aluminum reduction pot line.
-------
Table 4. MASS TRAIN TEST RESULTS - 553 WET ESP
u>
a\
Run No.
Date
Sampling Time, min.
H20, % by vol in gas.
Avg. Gas Temp., °C
JFlow, am'/sec
Vlow, DN m3/sec
mg/am3
gr/acf
gr/Dscf
Efficiency, %
Inlet
1
8/20
300
5.09
40.9
67.2
55.4
89.0
0.0389
0.0443
2
8/21
250
4.91
40.9
62.5
51.7
94.5
0.0413
0.0449
3
8/22
280
5.45
40.9
54.9
45.1
95.7
0.0418
0.0476
4
8/23
310
6.03
40.9
62.5
51.1
100.9
0.0441
0.0494
1
8/20
376
5.19
234.2
44.6
37.6
4.58
0.00200
0.00220
95.03
Outlet
2
8/21
375
4.96
38.1
43.5
36.3
4.26
0.00186
0.00209
95.34
3
8/22
375
5.22
38.1
44.4
36.9
3.57
0.00156
0.00166
96.51
4"
8/23
360
5.95
38.1
43.9
36.3
1.97
0.00086
0.00098
98.02
Notes:
1.
2.
3.
4.
Based on traverse across one sampling port and area of 3.05 m2(32.85 ft2)
Based on traverse across one sampling port and area of 3.54 m2(38.10 ft2)
0°C and 760 mm Hg.
Obtained at a single point near the center of the stack.
-------
traverse using a sampling device designed to provide sizing
information above 10 ym diameter would be required to resolve
the problem.
Analysis of Results
The wet electrostatic precipitator on which this test series
was conducted is a wire and plate design with three electrical
sections in series in the direction of gas flow. Plate-to-plate
spacing is 30.5 cm (1 ft), and each collecting electrode is
1.83 m long (6 ft) and 7.52 m high (25 ft). Thus, the total
parallel plate collecting electrode length is 5.48 m, or 18 ft.
Each electrical set powers 28 gas passages. The total parallel
plate collecting area is 2342 m2(25,200 ft2), and the "transverse
baffles", which are perpendicular to the gas flow, provide ad-
ditional collecting electrode area. The effective collecting
area provided by these baffles was estimated as 390 m2(4200 ft2),
resulting in a total collection area of 2732 m2(29,400 ft2).
Average specific collecting area during the test series was
therefore 62 m2/(m3/sec), or 315 ft2/(100 cfm).
Electrode irrigation is provided by sprays at the precipitator
inlet and above the collection plates. The sprays provide a
mist which is collected along with the particulates in the
flue gas, and the electrode cleaning is accomplished by the
coalescence and subsequent downward flow of the collected spray
droplets. The sprays are operated continuously, except for those
installed near the precipitator outlet, which are operated only
periodically. These spray nozzles were not in operation during
the test program. The irrigating fluid is a high pH sodium-
based liquid which is returned to clarifiers and a cryolite
recovery plant. Plant personnel reported that the cryolite
recovery system is essentially a closed liquid loop, which re-
sults in a solids content of about 5% by weight being returned
to the wet ESP - scrubber system. Liquor flow through the wet
ESP during the test program was constant at 31.5 I/sec (500
gal/min), which gives a liquid to gas ratio of about 0.7 1/m3
(5.3 gal/1000 ft ). Liquor temperature, based on measurements
reported by plant personnel, ranges from 90 to 104°F, and is
usually 94 to 95°F. No significant temperature drop has been
observed in the liquor loop across the precipitator.
Voltage and current readings were obtained from the panel meters
of the 553 precipitator periodically during the test program.
At the conclusion of the test program, voltage-current curves
were obtained for the unit with the spray system operating nor-
mally. The secondary voltage-current relationships are given
in Figure 21, along with the range of operation that was observed
for each electrical set during the test program. The difference
37
-------
550
500
450
400
350
300
250
200
150
100
Set No. 1
Set No. 2
Set No. 3
20 24 28 32 36 40 44 48 52 56
kV
Figure 21. Voltage Current Relationship (Manual Control)
and Operating Ranges (Automatic Control)
Shaded Areas Designate Observed Range of Operation
For Each Electrical Set During the Test Program
38
-------
between the voltage-current curves and the operating ranges is
a result of the fact that, in normal operation, the power supplies
are operating under automatic control with a certain spark rate,
whereas the V-I curves were obtained by manually increasing the
applied voltage until sparking occurred. The plant personnel
were operating the power supplies at a spark rate which was
believed to maximize the time-averaged electric field.
The V-I curve for the first electrical set is shifted toward
high voltages for a given current when compared with readings
from the other electrical sets. This behavior is often observed
and is a reflection of the higher space charge density contri-
buted by the higher particulate loadings which exist in the inlet
field. Although the third field operates at a relatively high
current, the average current density for all three sets was
only about 30 na/cm2. The resistivity of the particulate is
not a factor in the wet mode of operation.
Figure 20 presented fractional efficiencies and a predicted
curve obtained from a theoretically based computer model of an
electrostatic precipitator1. The mathematical model calculates
theoretically expected collection efficiencies for representa-
tive particle diameters as a function of precipitator operating
conditions. Predicted collection efficiencies for each particle
diameter are a function of the electric field, the charge on
the particle, and the ratio of collection area to gas volume
flow rate. Fair agreement is obtained between the measured
efficiencies over the particle diameter range 0.25-1.3 ym, but
the measured values depart drastically from the predictions and
from the trends indicated by the previously discussed laboratory
measurements, at diameters larger than 1.5 ym. This apparent
departure from the expected functional form may be caused by
the generation of particles within the device, possibly origi-
nating from the liquid sprays or from reentrained liquid that
is not captured by the outlet transverse baffles, which are
considered by the manufacturer to function as an electrostatically
augmented mist eliminator. It should be noted that the diameter
band 0.25-1.3 ym, based on the Andersen measurements, represents
54% of the mass at the inlet and 56% of the mass at the outlet.
Electrostatic precipitator performance is often described by an
empirical performance parameter termed the precipitation rate
parameter. The parameter is obtained by evaluating the Deutsch
equation using the overall mass efficiency and the ratio of
volume flow to plate area:
V , , 100 .
WP = A ln
39
-------
Evaluation of this relationship using the data in Table 4 gives
the results presented in Table 5. A predicted precipitation
rate parameter may be obtained from the computer model based
on the inlet size distribution obtained from the Andersen
impactor measurements. Based on the predicted efficiencies
indicated by Figure 20, numerical integration over the inlet
size distribution gives a total predicted penetration of 1.1%
(98.9% efficiency), and predicted precipitation rate parameter
of 7.3 cm/sec, which shows fair agreement with the data in
Table 5. Figure 20 shows, however, that the model underpredicts
fine particle collection efficiencies, and overpredicts col-
lection for particles larger than about 0.60 ym.
SUMMARIES OF TRIP REPORTS TO WET ESP MANUFACTURERS AND USERS
Dr. Charles E. Bates, head of the metallurgy section, has met
with representatives of Lurgi in Frankfurt, West Germany, and
British Steel Corporation in Moorgate, England. The following
paragraphs summarize the more important aspects of these
meetings.
Visit to British Steel Corporation
British Steel Corporation began applying wet electrostatic
precipitators to electric arc furnaces about ten years ago to
eliminate a fluorine problem in the effluent which, for the
case of two plants, resulted in the development of fluorosis
in cattle grazing nearby. Effluent particles from the electric
arc furnaces are predominantly under one micron in diameter.
Therefore, the only choices for emission control were electro-
static precipitators, bag filters, and venturi scrubbers.
The difficulties with bag filters were that the temperatures
of the gas streams were quite high, the particles quite small,
and the bags would tend to plug. The presence of moisture and
fluorine eliminated the use of fiberglass bags, and the operating
costs of venturi scrubbers were prohibitive. After considering
the possibilities, British Steel decided to install wet
precipitators. Initially, the design did not permit 100%
saturation of the gas before entering the precipitator, and
after four years of labor and much cost, it became apparent
that 100% saturation was essential at all times. This satura-
tion must be achieved in a conditioning chamber because of the
time required to fully saturate the gas stream. British Steel
found that if the gas entering the precipitator were not cooled
and completely saturated, the warm, moist iron oxide would
stick to the baffle, wires, and plates of the precipitator
like cement, but if the effluent were cooled and wet, the iron
oxide would not stick. Since water carry-over during flushing
caused sparkover and could possibly knock out a complete power
supply, individual power supplies were installed on each stage
of the precipitator to enable sections being cleaned to have the
40
-------
TABLE 5
PRECIPITATION RATE PARAMETERS
Run No. Gas Flow, m3/sec
1 44.6
2 43.5
3 44.4
4 43.9
Mass
Efficiency,
95.03
95.34
96.51
98.02
Precipitation
Rate
Parameter,
cm/sec
4.90
4.88
5.45
6.30
41
-------
power turned off. The first installation had two fields in series,
but while the plates in one section were being washed, only 50%
of the cleaning capacity was being utilized. In the second
installation, the arrangement was three fields, so that two fields
could operate while the third was being cleaned. All three fields
operate expect when washing and all precipitators are washed
intermittently for 3-4 minutes during each 30 minutes of operation
with high pressure water sprays. Constant irrigation proved un-
satisfactory because of irregularities in either the water flow
or plate shape.
The first precipitators installed by British Steel were
designed based on a precipitation rate parameter of 10 to 12
cm per second. Experience showed that better results were
obtained if the design precipitation rate parameters were
reduced to 7 to 8 cm/sec. Specifications required by British
Steel for wet precipitators recently constructed include those
given in Table 6.
Several types of discharge electrodes were tested prior to
installation of the electrostatic precipitators at British
Steel. Collection electrodes are of the plate type. Initially
a mild steel barbed wire was used, but dust bonded to the wire,
leading to corrosion. This same experience occurred when a
galvanized wire was used. Differently designed wires were found
to be satisfactory, but the most successful design proved to be
a stainless rod with stainless spikes drilled or welded into the
rod's surface. The stainless steel rod gave the necessary
rigidity, and the spikes provided the necessary field strength
(shown in Figure 22).
The maintenance costs of these wet precipitators are now
quite low and no unscheduled maintenance is experience. The
operating costs consist of costs for power, water supplies, and
maintenance, with about 1/3 of this cost due to maintenance.
The power supplies, with automatic voltage control, are rated
at 60 kV and 450 milliamps. Silicon rectifiers are employed and
the primary power supply is 415 volts, 50 cycles, + 6%. The
precipitators are supplied by W. C. Holmes of Huddersfield in
Yorkshire.
Visit to Lurgi Apparate - Technik GmbH, Frankfurt, Germany
Lurgi is of the opinion that wet precipitators are applicable
where the gas is already wet or the resistivity of the dust is so
high that water must be added to saturate the air, thus lowering
resistivity. If large amounts of water must be added, or if the
gas volumes are not too high, they prefer to use a scrubber and
avoid some of the corrosion and maintenance problems and high
capital costs associated with the wet precipitators. To capture
42
-------
TABLE 6
PERFORMANCE CHARACTERISTICS OF WET PRECIPITATORS
SPECIFIED BY BRITISH STEEL
Gas Flow
43 m3/sec at 60°C
Particle Size Distribution Particles range from 5 ym to . 1 ym
of Inlet Dust
Precipitation Rate
Parameter
Collection Efficiency
Inlet Dust Loading
Outlet Dust Loading
diameter, 95% less than 0.5 ym
8.5 cm/sec maximum
97.5%
4.6 g/m3 (2.0 gr/ft3) at N.T.P.
dry (maximum)
0.115 g/m3 (0.05 gr/ft3) at N.T.P.
dry (maximum)
43
-------
Studs 12.7 mm x 3.175 mm
dia are drilled or welded on
U
4*
H
O,
ro
-------
particulate and gases, one could use a scrubber to remove
chloride and sulfur oxides. Lurgi has never looked at wet
precipitators for power plant applications because they con-
sider them too costly and complicated. Lurgi has installed
wet precipitators on cupolas, arc furnaces, basic oxygen
furnaces, blast furnaces, and scarfers.
Basic Oxygen Furnaces -
Basic oxygen furnaces commonly use 40-60" pressure drop
scrubbers for particulate control, but Lurgi has built about
five wet electrostatic precipitators for this application.
These wet cleaning plants require considerable maintenance due
to the large amount of oxygen, fluorides, and sulfides in the
gas, which lead to corrosion of the internal parts of the control
system. Corrosion could be avoided by using stainless steel
but this would be prohibitively expensive. The market for wet
precipitators for this application does not appear favorable
to Lurgi.
Scarfers -
Scarfers commonly use water sprays to cool the grit produced
by the scarfing operation, so a wet precipitator becomes a
natural choice. Saturation of the gas before entering the
precipitator is essential as is maintaining an adequate flow
of water over the plates to prevent buildup. Fully saturated
effluent does not bond, and in fact, is easily washed off. The
gas volumes are relatively low, so one can afford stainless
steel internals, thereby minimizing maintenance costs. The
wet precipitator provides good performance, low operating cost,
and can be small in size when applied to scarfers. Lurgi has
sold seven or eight wet precipitators for effluent control of
scarfing operations over the past few years.
Blast Furnaces -
Electrostatic precipitators were very popular on older blast
furnaces where the top pressure was not particularly high. At
the present time, however, all blast furnaces being built have
top pressures of 15 to 20 psi and part of this pressure is
available to do work. With a high top pressure, sufficient
energy is available to simply pass the gas through a venturi
and allow the gas expansion to do much of the work required.
Therefore, high energy wet scrubbers, using this gas energy,
can be employed just as economically with a lower capital in-
vestment compared to precipitators. The wet precipitator does
work well on blast furnaces and corrosion is not a problem be-
cause of the very low oxygen partial pressure, and hence, low
corrosion potential, of the effluent gas. Lurgi has about 250
wet precipitators in operation and a partial list of these was
45
-------
furnished to Southern Research. Demand for wet precipitators
for blast furnace cleaning is not expected to increase except
in undeveloped countries where sophisticated furnaces will not
be used.
Cupolas -
Small gas volumes are involved in cupola operations compared
to most processing equipment, and wet precipitator usage is
possible. Lurgi has built three or four wet precipitators and
two or three dry precipitators in the last few years for foundries,
but because of the small profit margin involved for supplying
pollution equipment for small foundry operations, Lurgi does not
expect to pursue it.
Electric Arc Furnaces -
The most common control devices for arc furnace effluents are
bag houses and electrostatic precipitators. At this time Lurgi
leans toward dry precipitators since the dust resistivity is
reasonable when the gas is diluted with plant air and cooled
to about 70°F. Above a gas volume of about 500,000 cu. meters/
hr, the precipitator is less expensive than a bag house in total
operating costs.
Design Details -
Lurgi wet precipitators normally use plates and two types of
water supply. The continuous water sprays (about 45-50 psi)
employ nozzles to achieve as fine a mist as possible, using
clean water. The second system of sprays provides a high
volume of water to the plates every three or four hours to
give the plates a good cleansing. During this flushing oper-
ation the power is left on, but with voltage reduced to 10-15
kV instead of the normal 40 kV. The precipitator normally has
three fields, sometimes four, and one field is washed at a time.
When the gas going through the precipitators is incompletely
saturated, buildup and corrosion can be serious problems. Basic
water will cause buildup and acid must be added to get a release.
Acidic water will cause corrosion in a precipitator.
i
There are very few, if any, gravity water flows in use because
of difficulty in maintaining uniform water flow down the plates
(the weirs cannot easily be kept level and the plates or tubes
cannot easily be kept straight). Precipitation rate parameter
values of 7 to 10 cm/sec in wet precipitators and 3 to 5 cm/sec
in dry precipitators are obtained with the same distance between
wires and plates. The precipitation rate parameter in dry pre-
cipitators at 70 to 90°F using plant dilution air will sometimes
run as high as 10 cm/sec, but will vary widely with dust resistivity,
46
-------
On old blast furnaces, dust loads of 0.78 to 1.3 grams per
cubic meter are common in the raw effluent from the furnace.
A cyclone then drops the dust loading to 0.65 to 0.90 grams/m3
With a center charge going into the blast furnace, the dust
coming out is typically 0.5 to 0.65 grams/m3. The gas is then
cleaned by a wet scrubber which cleans the effluent to 30-100
mg/m3 , and then finally by a precipitator which cleans the gas
to about 5 mg/m3.
If the gas is to be sent to a turbine, the dust loading must
not be higher than 0.5-1.0 mg/m3, which can usually be achieved
with a secondary precipitator. The discharge electrodes of such
a precipitator are schematically illustrated in Figure 23. Com-
monly three fields are used. In the first field there will be
8 mm diameter stainless steel rod with welded studs spaced on
the rod. The rod in the second stage is the same size but with
square studs welded on it. In the third stage, a bare wire is
used with water flowing over it, the water stream itself pro-
viding enough asperities to cause corona for charging the dust.
Mikxopul Division of United States Filter Corporation
Mikropul has recently constructed large-scale wet precipitators
on aluminum pot lines, and the company has also been actively
pursuing wet precipitator usage in other application areas.
During the course of this contract, a trip was made to the
Mikropul Plant in Summit, New Jersey, and discussions were held
concerning Mikropul's experience and pilot plant test program.
At the time of the plant trip, the company was running pilot
plant tests of a wet precipitator designed for collection of
sulfur oxides and fly ash from a coal-fired power plant. The
pilot plant is equipped witTi" a propane burner to produce syn-
thetic flue gas. Collected fly ash from a coal-fired boiler
is injected with a blower, and S02 is added from a gas cylinder.
At the time the visit was made, SOz and particulate removal
trails were being performed with a calcium-based process. Liquid
to gas ratios were said to be of the order of 100 gal/1000 ft3,
and SC>2 removal efficiencies as a function of liquid to gas
ratio were reportedly typical of data reported by other investi-
gators. Scaling caused by precipitation of calcium salts on the
electrodes was viewed as the major problem area to be overcome
for power plant applications.
Mikropul's objective is to produce a single "box" for removing
both S02 and particulate. The pilot plant design essentially
consists of a scrubber, a particle precharger, a discharge
electrode and parallel plate collector, followed by an electro-
static demister section which can be periodically washed. In
addition to the calcium-based process, a sodium-based process
47
-------
8 nun
8 nun
4 mm diameter
stainless steel
studs attached
by either
welding or
drilling and
welding
4 mm
1st Stage Discharge
Electrode
2nd Stage Discharge
Electrode
Figure 23. Lurgi Discharge Electrodes
48
-------
is also under consideration for use with the wet precipitator
system.
Mikropul reported that they have obtained pilot plant data
from the following application areas:
Tire-cord curing ovens
Glass melting furnaces
Fiberglass forming lines - curing ovens, cooling section
Coke ovens
Sinter plant
Anode baking furnaces
Secondary brass melting furnaces
Phosphorus reduction furnace
Borax dust.
Table 7 gives a list, furnished by Mikropul, of their wet
precipitator installations.
Research Cottrell
A telephone conversation held with a staff member of Research
Cottrell concerning their current interests in the wet precip^
itator market. He reported that his company has no recently
constructed wet precipitator installations. Most of their wet
units have been used on blast furnaces, and many of these are
being replaced by high energy scrubbers.
Western Precipitator Division of Joy Manufacturing Company
Discussions were held with representatives of Western to de-
termine their opinions concerning the future application of
wet precipitators. Western manufactures wire-pipe wet ESP's,
typically with 96 ten-inch diameter, 15 ft long pipes in each
module. These modules have recently been used on scarfing
machines, with each unit treating about 25,000 cfm. No test
data was available for these units.
Although Western has considered the use of wire-pipe wet
ESP's for incorporation into an SOa removal system, they do
not consider it practical, because of corrosion and resultant
high materials cost, to use a wet ESP as the primary particle
remover. In trials of a scrubber-wet precipitator system that
have been conducted by Western, the wet ESP was said to function
as a mist eliminator following a low energy scrubber. Western
recommends "hot" precipitators for particulate collecting in
coal-fired power plants.
Blast Furnace Wet Precipitator Installation
A visit was made to a steel mill which uses wire and tube-type
49
-------
TABLE 7
MIKROPUL WET ELECTROSTATIC PRECIPITATOR INSTALLATIONS
Location
Description
Capacity,
cfm
Installation
Date
Aluminum Potlines
Martin.Marietta
Goldendale, Wash.. 20 units
Martin Marietta 7 units
The Dalles, Oregon 4 units -
Reynolds Metals Co.
Longview, Wash. pilot unit
Phase I
Phase II
Phase III
1 unit
4 units
4 units
12 units
10 units
7,500 cfm
12,000 cfm
6,000 cfm
50,000 cfm
50,000 cfm
100,000 cfm
50,000 cfm
100,000 cfm
100,000 cfm
1971
1972
Oct. 1971
June 1973
1974
Fiberglas Forming Lines
Certain-Teed Prod.
Kansas City,
Kansas 2 units
100,000 cfm
1973
Carbon Anode Baking Furnace
Airco Speer Carbon
Graphite
Niagara Falls, N.Y. 4 units
Reynolds Metals Co. 1 unit
9,000 cfm
26,500 cfm
1973
1973
Oil Mist
General Motors Corp.
Buick Division
Flint, Michigan 1 unit
15,000 cfm
1972
Sodium Sulfite Mist
Shattuck Chemical Co.
Denver, Colorado 1 unit
3,500 cfm
Phosphorus Rock Dust
W.R. Grace & Co.
Bartow, Florida 2 units
115,000 cfm
1972
1974
Coke Oven Emissions
Dominion Foundry
& Steel
Hamilton, Ontario 2 units
150,000 cfm
1974-1975
50
-------
wet precipitators for cleaning blast furnace gas. Two pre-
cipitators are employed, each of which has two units. Each
unit consists of 126 pipes 8 in. in diameter and 15 ft long,
giving a total of 504 tubes with a collecting area of 15833 ft2,
These units were installed in the 1941-1942 time period on a
large blast furnace. They are now being used on a smaller
blast furnace, and a high energy venturi has replaced the
wet precipitators on the large blast furnace. The current
plan is to phase out the precipitators in about two years,
and use high energy scrubbers exclusively.
Performance tests on the wet precipitators were conducted by
Research Cottrell in April of 1960, and Table 8 gives the
data reported to the user. At the time these tests were con-
ducted, the units were powered by mechanical rectifiers. These
have since been replaced with two tube-type rectifiers, each of
which powers one precipitator. During our visit, the panel
meters indicated a total DC current of 0.5 amps, which would
result in an average current density of about 30 ya/ft2. The
precipitators are preceded by wet scrubbers which cool the
furnace gas and remove the relatively large particles. The
gas flow and efficiency data in Table 8 result in precipitation
rate parameters between 9.5 and 11.5 cm/sec. These values are
approximately within the range of design values reported by
Oglesby and Nichols1* for blast furnace wet precipitators.
Although the management at the blast furnace wet ESP in-
stallation that we visited was planning to phase out the
units, their performance was said in general to be satis-
factory. The tubes require cleaning once or twice a year,
and the precipitators have been retubed at least once since
their original installation. The general reasoning seemed to
be that operating a blast furnace at relatively high pressure
was advantageous, and this pressure could be effectively
utilized in a scrubber for particulate removal.
51
-------
TABLE 8
TEST RESULTS OBTAINED BY RESEARCH COTTRELL ON A BLAST FURNACE
WET PRECIPITATOR INSTALLATION
en
Test No. *1
Date 4/22/60
Furnace Wind, cfm 73,000
Blast Pressure, psig 24.0
Top Gas Pressure, "H20 40
Hot Blast Temperature, °F 1050
Precipitator Inlet Gas Temp.,°F 88
Gas Volume, cfm 99,000
Precipitator Rating, % 90
Dust Concentrations, Grains per
cu. ft. Dry Gas @ 60°F & 30" Hg.
Inlet 0.133
Outlet ***
Precipitator Efficiency, % ***
*2 *3 **4 **5
4/25/60 4/25/60 4/26/60 4/26/60
71,000 71,000 64,000 64,000
22.0 22.0 19.0
40 30 43
1050 1050 950
90 90 90
103,500 103,500 86,000
94 94 156
0.0915
0.0051
94.4
18.5
44
890
90
86,000
156
0.128 0.0675 0.0698
0.00705 0.00861 0.0103
95.0
87.2
85.3
*Tests made with gas from No. 2 Furnace passing through both precipitators
**Tests made with gas from No. 2 Furnace passing through one precipitator.
***Contaminated outlet sample was discarded.
-------
SECTION V
ANALYSIS OF POTENTIAL WET ESP APPLICATIONS
SIMULTANEOUS REMOVAL OF PARTICULATE AND SULFUR OXIDES
The recent emphasis on reduction of sulfur oxide emissions
has resulted in a number of processes which involve wet
scrubbing of flue gas to remove both particulate and sulfur
oxides. High efficiency particulate removal in scrubbers
generally requires pressure drops of 25 to 50 cm of water or
more, which for utility power plants involves a considerable
energy loss. The removal of sulfur oxides, however, can be
accomplished with less energy in a scrubber which is designed
primarily for gas absorption. The electrostatic precipitation
process can collect particulate with less energy expenditure
than with wet scrubbers, but the sensitivity of dry precip-
itators to fuel composition changes and resultant dust resis-
tivity variations is a serious drawback. It is therefore of
interest to consider the use of a wet precipitator, which is
not sensitive to dust resistivity, as the primary particulate
collection device for small particle removal.
Figure 24 illustrates possible configurations of a combination
scrubber-wet precipitator. As indicated in the illustration,
the device may consist of a separate low energy scrubber and
wet-wall precipitator, or the scrubber and precipitator can
be combined into one unit with alternating scrubbing and
collecting plate sections. In either arrangement, it is ob-
vious that simultaneous particulate collection and gas absorption
will occur in the precipitator sections. Consideration will
therefore be given to the potential mechanisms of sulfur oxide
removal which exist in the electrical sections. These mechanisms
are:
1. Migration of ionized S02 molecules to the collecting
electrode under the influence of an electric field.
2. Diffusion and absorption of S02 into a basic solution
irrigating the collecting electrodes.
3. Absorption of S02 by droplets of basic solution which
may be sprayed into the interelectrode region.
An estimation of the maximum amount of S02 which might be removed
by electrical transport can be calculated by assuming a current
density and a plate area to volume flow ratio (specific collecting
area). As an example, we will consider a wet precipitator with a
specific collecting area of 59.1 m2/(m3/sec) (300 ft2/1000 cfm),
which operates with a current density of 100 nA/cm2. The corona
53
-------
ABSORBING SOLUTION
FLUB GAS
INLET
PRE-
CONDITIONING
CHAMBER
LON ENERGY
SCRUBBER
I
NET-HALL
PRECIPITATOR
HUB GAS
'TO TACK-
SPENT ABSORBENT AMD PLY ASH
TO DISPOSAL OR REGENERATION SYSTEM
OPTION A
ABSORBING SOLUTION
FLUE GAS
INLET
PRE-
CONDITIONING
CHAMBER
^
1 1
1
SPRAY COLLECTING) SPRAY COLLECTING! SPRAY COLLECTING
VOLUME PLATES j VOLUME PLATES (VOLUME PLATES
1
1
1
1 1
1 1
1 - I
1
1
1
|
1
FLUB GAS
'TO STACK
SPENT ABSORBENT AND FLY ASH
TO DISPOSAL OR REGENERATION SYSTEM
OPTION B
Figure 24.
Schematic Diagram of Scrubber-Wet
Precipitator System.
54
-------
current density is given by
coul nrt-3 coul
inn v in-9 amP v 10* cm x 1
100 x 10 _£ x 10 -p- x 1
If we assume that the current is carried by singly charged
ionized gas molecules, the rate at which such molecules are
transported to the collecting electrode is
10~3 coul 1 ion _ 6.25 x 1015 ions
m2-sec x 1.6 xlO~19coul ~ m2-sec
From the specific collecting area and the removal rate of ions,
it can be seen that the number of ions removed per cubic meter
of gas is given by
6.25xl015ions 59.1 m2 ions
m2-sec x m3/sec
At 120°F, it can be shown that the number of gas molecules in a
cubic meter of gas is equal to 2.28xl025. Thus, the fraction
of the total number of gas molecules which could be electrically
transported to the collecting electrodes is equal to
3.69xl017 ions/m3 , _._ ,rt_8 rt rtl <-«
2.28x10** molecules/m3 = 1-62x10 8, or 0.0162 ppm
It is therefore apparent that the corona current cannot be ex-
pected to directly contribute significantly to the amount of
sulfur oxides which might be removed in a wet precipitator .
The second potential mechanism is absorption of SOz into a basic
solution irrigating the collection electrode as a result of the
reduced partial pressure of S02 above such a solution. An esti-
mate of the gas . phase mass transfer coefficient applicable to
55
-------
these conditions may be obtained from the Sherwood equation,5
which was obtained for wetted wall columns with the fluid
descending in laminar flow and the gas ascending in well de-
veloped turbulent flow. The equation is
= 0.023 [dupl
Dm l~P~/
0.8!
0.33
where
d = equivalent diameter of gas passage
D ,D = molal and volumetric diffusivity of gas being
m v absorbed (S02)
u = gas velocity
p = density of gas phase
y = viscosity of gas phase
K = gas film coefficient
For a precipitator with 25.4 cm plate spacing and a gas velocity
of 1.52 m/sec (5 ft/sec), the above correlation gives Kg as
0.183 g mole/[m2 (sec) (mole frac.)]. If we make the assumptions
that the gas phase resistance is the rate controlling step and
that the partial pressure of S02 above the solution is negligible,
the absorption rate may be estimated from
N = Kg(A) ^in'
where
N = moles/sec of SO2 removed
A = absorption area
y, = log mean mole fraction of SO2 in gas phase
(Ii2-)
Uout/
For the previous calculation of the amount of S02 which could be
electrically transported, we assumed an A/V ratio of 59.1 m2/(m3/sec)
If we further assume a log mean SO2 concentration of 1000 ppm, the
S02 removal rate for 59.1 m2 is given by
56
-------
N = 0.183 5 gmole x 59>1 m2xlxlo-3 mole frac,
(m )(sec)(mole frac.)
= 0.0108 gmole/sec
The total gas flow passing each 59.1 m2 of plate area is
1 m3/sec, or 37.9 gmoles/sec at 120°F. The quantity of S02
removed, expressed as a fraction of the total gas, is therefore
0.0108 gmole (S02)/sec _ -6
37.9 gmole (gas)/sec
These calculations indicate that the solution irrigating the
walls will provide sufficient interfacial area for only a minor
portion of the required SOa removal, although it is possible
that turbulence created by the electric wind in the interelectrode
space may increase the value of Kg to some extent. The principal
SOa removal, however, must be accomplished by sprays in order to
increase the interfacial areas.
The use of a wet ESP in an SOx-particulate removal system must
therefore be decided upon the basis of the WESP's performance as
a particulate collection device. Figures 25 and 26 give computed
projection of particulate collection performance of a WESP on a
typical power station fly ash. These projections were obtained
from the SRI-EPA precipitator computer model,6 and the voltage-
current relationships are based upon the data contained in Figure 21,
A gas velocity distribution with a standard deviation of 25% of
the mean velocity was used in the model for both sets of indicated
electrical conditions. Gas by-passage was assumed to be 5% over
two and three stages. Figure 25 indicates the expected collection
efficiency relationship using what we consider to be optimistic
electrical operating conditions, whereas the efficiency relation-
ships in Figure 26 are based upon the average electrical conditions
which were obtained during the test program described in Section IV.
These projections suggest that a specific collecting area of
39.37 m2/(m3/sec), or 200 ft2/1000 cfm, would constitute a con-
servative estimate of the collection area required for achieving
99.6% or better collection efficiency of particulate.
Although in principal S02 removal can be accomplished by intro-
duction of spray droplets in the interelectrode region, it is our
57
-------
m2/(m3/sec)
c
J J. J^
99.9
QQ ft
N7C7.O
v^
o
z
LLJ
0
LL
U.
tl I
ZQQ *5
_ yy.o
o
i-
ai
_i
_i
O
99
QQ
yo
C
).{
,
y
/
n
-^
>0
5
r
(
i
f
A
i
t
y
j
f
t
?
i
-
nr
[
K
b
v
r
J
^
j
/
f
.._
J
f
J
'
f
f
P_
/
,
f
fl
J
f
A
/
-5
^
,
/
/
/
'
7
>
r
/
-
i--i
-4
19
i
/
t
f
-
1C
.7
.
/
A
*
-
X
^
J
)
-
/
1
i-
/
t?
H
V
t
^
^
1 1
/
H
--
t
/
-/
!9
f
jr
15
.6
Y
^-
V
'
^
\
5C
j
A
f
/
)
/
=
/
1
"1
=
^
/
f
'-
j
--
/N
/
j
^
i
-
^
>>y
<
r
i
f
f
H
i
V
2(
4
,\
V*
/
^
3(
/
3
/
j
.
f
*
(
/
LI
\
^
',
f
sc
V
^
^
r
I
S
-
,«
V
/
*
f
*>
-
1
4
N
^
^
,
t
9
Cj
k
/
c.
X1
3J
2
(
I/1
3
JC
^
\s
^\
O
=1
I1
)
0<
y/
'
X
0
/
X
V
^
^
fg
V4
^
V
v«
,'
#
<
LX1
J
>9
t
L^
[ )
1
59.
*
2 =
^^
^
--
--
rr
T .
30
SPECIFIC COLLECTING AREA, ft2/(1000ft3/min)
Figure 25. Collection Efficiencies Calculated With
Computer Model for 50 x 10~9 amps/cm2 and
50 KV
58
-------
9.8
19.7
m2 /(m3 /sec)
29.6 39.4
49.2
59.1
99.95
99.9
99.8
o
z
UJ
o
LU
z
o
o
ID
O
O
99.E
99
j
=1=
4=1
~f
i /
" r
pn
/
/
/
/
f
M
-
-
I
f
I
r
t
.
f
'
/
r
^L
j
/
i
f
i
/
/
/
r
/
y
/
y
y
/
^
y
f
/
r
-
V
/
^
>
/
y
^
^
/
y
v
/
^
/
y
>
/
?
f
7
r*
V
H
_
»
-*
V
f
-j
A
f
t
J
<
A
f
1?
j
J
X
V
0
^
V
^
^
*l
V
/
^
*-
J
A
o
S
,,*
c<
VJ
/
_.
X
y
i
^
^
j
n
=s
/
^
V
/
V
/
-
0
>
Nk
/I
J
,
^
^
kj
(
'
'-
ft
'
^
V5
\
'
M
k
V'
^
^
t
^
*
X
_
y
>
-3 ^
r
>^
^ *
50
100
150
200
250
300
SPECIFIC COLLECTING AREA, ft' /(1000ft3/min)
Figure 26.
Collection Efficiencies Calculated With
Computer Model for 33 x 10"9 amps/cm2 and
47.3 KV
59
-------
opinion that introduction of the sprays into the electrified
region in the required L/G ratios would limit electrical operating
conditions for full-scale units with practical values of plate
area for each T-R set. Therefore, S02 removal should be accom-
plished in a relatively inexpensive spray tower preceding a wet
ESP. Perhaps the most favorable circumstances for use of a WESP
would be in a sodium-based absorption system in which S02 removal
can be accomplished with moderate L/G ratios.
Table 9 gives a comparison of power requirements for two particulate
control methods in an SOx-particulate removal gas cleaning system.
Scrubber operating assumptions are based on data obtained at the
EPA Alkali Scrubbing Test Facility.7 Based on the indicated assump-
tions, the WESP system has lower power requirements, primarily as
a result of the low pressure drop through the system. Table 10
and 11 give capital and operating cost estimates. The WESP capital
costs are based on the assumption of stainless steel construction
with installed costs of $377/m2, or $35/ft2, and a specific col-
lecting area of 39.37 m2/(m3/sec). These comparisons indicate that
the WESP costs for particulate removal are about two times as
great as the scrubber due to the high capital costs. The wet ESP
system costs would have to be reduced by about 60% in order to
have the same annual operating costs as the TCA scrubber system.
Maintenance costs have not been considered quantitatively in this
analysis, but the possibility of scale formation on the electrodes
in the WESP is potentially a much more serious problem for the WESP
than is scale formation in a scrubber vessel.
Another consideration which is not reflected in these comparisons
is the performance of a dry ESP. Collection efficiencies of dry
ESP's under favorable operating conditions with high sulfur coals
have been measured as 99.7% with specific collecting areas less
than 49.2 m2/(m3/sec) , or 250 ft2/1000 cfm. Therefore, unlesns
WESP's can be demonstrated to achieve 99.5% or better collection
efficiency with SCA's considerably less than 39.4 m2/(m3/sec),
or 200 ft /1000 cfm, there is no economic justification for sub-
stituting a WESP for a conventional dry electrostatic precipitator
in an SOx-particulate removal system with flue gas produced from
high sulfur coals. For such instances, an economic analysis should
be conducted which compares the operating costs of a scrubber-only
system with those of a system consisting of a low energy scrubber
and a dry ESP.
COLLECTION OF HIGH RESISTIVITY DUST
Utilities in some parts of the United States are faced with re-
quirements for SOx removal from flue gas in which the dust resis-
tivity is sufficiently high to limit the electrical operating
conditions in the precipitator. Under these conditions, the
plate area required for 99.5% or greater particulate collection
60
-------
TABLE 9
POWER REQUIREMENT COMPARISON FOR PARTICULATE
CONTROL IN SOx REMOVAL SYSTEMS
kw/kacfm
Item Scrubber-Wet ESPScrubber Only (TCA)
Pump 1.080 2.03
Fan 0.468 2.30
Power for TR Sets 0.405
Insulator Power 0.180
Total 2.133 4.33
Assumptions:
1. Fan and pump efficiency 50%.
2. 5.08 cm HaO AP in gas for scrubber-WESP system.
3. Fluid head for pumping calc. = 28.1 m.
4. L/G for scrubber-wet ESP System
Scrubber (spray tower) = 3.4A/m3(25.8 gal/1000 ft3)
WESP = 0.7£/m3 (5 gal/1000 ft3)
5. L/G for scrubber only system = 7.8£/m3 (58.2 gal/1000 ft3).
6. 24.9 cm H20 AP in gas for scrubber-only system.
7. WESP has SCA of 39.37 m2/ (m3/sec).
8. TCA particulate collection efficiency = 99.8%.
9. WESP particulate collection efficiency = 99.75%.
10. S02 removal = 90%.
61
-------
TABLE 10
CAPITAL COST ESTIMATES FOR PARTICULATE CONTROL
IN PARTICULATE-SOx REMOVAL SYSTEM
150 MW Boiler, 183 m3/sec (394,000 acfm) at 50°C
Item (a) Scrubber-WESP Scrubber Only (TCA)
Wet ESP 2,760,000
Spray Scrubber 105,OOO10 85,000 (c)
TCA Vessel 608,OOO9
Scrubber Fans & Motors 255,000
Total 2,865,000 (b) 948,000 (b)
(a) Elements common to both systems not included.
(b) Cost scaled to February 1975 dollars with chemical plant cost index.
(c) Presaturator.
62
-------
TABLE 11
OPERATING COST ESTIMATES FOR PARTICULATE CONTROL
IN PARTICULATE-SOx REMOVAL SYSTEM, $/yr
Item (a) Scrubber WESP Scrubber Only
Cap charges at 15%/yr $430,000 $142,200
Energy cost (b) 40,300 81,900
$470,300 $224,100
(a) Elements common to both systems not included.
(b) Based on 6000 hr/yr $0.008/kwh
63
-------
efficiency can be 98.4 m2/(m3/sec), [500 ft2/1000 cfm], or
greater if a conventional ESP is used at operating temperature
in the 150°C range. The use of a WESP could lower the plate
area requirements by a factor of two or more, but presently
available information indicates that the increased materials
costs required by wet operation would partially or completely
offset the cost savings resulting from plate area reduction.
The possibility of electrode scaling in WESP operation is a factor
which would require pilot plant trials for an adequate evaluation.
COLLECTION OF MISTS FROM SCRUBBER SYSTEMS
The laboratory studies described in Section III indicate that
the mathematical model should provide a basis for evaluating the
potential effectiveness of WESP's for collecting mists from
scrubber systems if a size distribution of the mist is available.
Table 12 gives a size distribution, provided by EPA, of droplets
at a spray scrubber exit.8 The total droplet concentration is
11000 grams of liquid per cubic meter of gas, which is equivalent
to about 80 gal/1000 ft3. This extremely high concentration of
large droplets causes two problems in using the model to project
overall mass collection performance:
1. The model is based on the Deutsch equation for individual
particle sizes, and the Deutsch collection mechanism does
not apply to large particles (7-10 ym diam.).
2. The high droplet concentration results in high values of
particulate space charge, and the effects of this space
charge are not adequately represented in the model.
In view of these difficulties, results obtained from the model
with this size distribution should be considered as only very
rough approximations. Figure 27 shows overall collection efficiency
projected by the model with the size distribution of Table 12.
Wire-plate geometry was assumed, with a current density of 33xlO~9
amps/cm2 and an applied voltage of 47.3 kv. Model correction
parameters for the overall mass efficiency were:
1. A gas velocity distribution with a standard deviation of
25% of the mean.
2. Gas by-passage assumed to be equivalent to 5% over two
stages.
Figure 28 shows the theoretical collection efficiencies for the
same current density and applied voltage, but with space charge
effects due to large droplets ignored, as a function of particle
diameter.
64
-------
TABLE 12
SIZE DISTRIBUTION OF DROPLETS AT A SPRAY SCRUBBER EXIT
Average Droplet Cumulative Mass
Diameter/ cm Loading, grams/m3
1.5 1.6
2.6 8.8
6.6 26
18 590
36 1,400
100 11,000
65
-------
99.995
99.99
9.8
m2/(m3/sec)
19.7 29.6
39.4
t- - (--
IT.. . . I ..: . _.,
, . . I ~3..
H M- - -' ;
m
-H--
~^3iri£[^U
f^'-i-i.^f !:'--
^u^mmm-^
:Sb
K
. _£T^ 7
^T-H1'-' ,'^i J
=r.t=f 1.3
y^'-
^3~T=:
EiCMj
-t :_j.
50 100 150 200
SPECIFIC COLLECTING AREA, ft2 /(1000ft3 /min)
Figure 27. Collection Efficiency Calculated With
Droplet Size Distribution for 33 x 10~9
amps/cm2, 47.3 KV
a = 0.25, S = 0.05, N = 2
66
-------
These results indicate that a WESP would require about 39.4 m2/
(m3/sec) to achieve a theoretical collection efficiency of about
99.5% on 2.5 ym diameter particles (again ignoring space charge
due to the large drops), even though the overall mass collection
efficiency at this SCA is projected to exceed 99.99%. If the
large droplet space charge is allowed to retard collection effi-
ciency using an approximate method for reducing free ion densities,
the theoretical collection efficiencies at 39.4 m2/(m3/sec) for
the droplet size bands are reduced as shown in Figure 28. The
overall mass collection efficiency of 99.992% indicated in Figure 27
would result in an outlet mass loading of 0.88 grams/m3, which would
reduce to ^0.09 grams/m3 if 10% of the droplets consisted of non-
volatile solids. Thus, the model projections indicate that a WESP
would require an excessive plate area to collect the distribution given
in Table 12 with an efficiency such that the outlet loading from the
WESP would be less than .0458 g/m3 (.02 gr/acf). This is a result of
the excessive inlet loading of 11,000 grams/m3.
For other size distributions, it is recommended that the theoretical
fractional efficiencies curves of Figure 28 be used to obtain an es-
timate of overall collection efficiency. These curves are based
on wire-plate geometry and are valid only if particulate space
charge is not sufficient to disrupt the electrical conditions. EPA
is currently supporting studies to better define conditions under
which space charge suppression of corona current will occur.
67
-------
49.2
99.995
99.99
99.95
99.9
99.8
UJ
o
8
4X DROPLET COLLECTION EFFICIENCIES
1 WITH SPACE CHARGE EFFECT EST.
99.5
50 100 150 200
SPECIFIC COLLECTING AREA, ft2 /(1000ft3/min)
250
Figure 28. Theoretical Collection Efficiencies
as a Function of Particle Diameter
68
-------
SECTION VI
RECOMMENDED RESEARCH AND DEVELOPMENT
The laboratory studies conducted under this contract have shown
that wet electrostatic precipitators can be effective collectors
of fine particulate under idealized conditions. The effective-
ness of WESP's used under field conditions is below the poten-
tial performance due to (1) space charge effects, which limit
allowable current density, caused by incoming particulate
and the electrode irrigation system, (2) deposits forming on
the electrodes which cause sparkover at low applied voltage
levels, (3) corrosion problems peculiar to the application
area and the chosen materials of construction. Recommended
research and development programs for these problem areas are
given below.
1.
PROBLEM
Space charge limitation of
current
SUGGESTED EFFORT
a) Develop more accurate
methods of calculating
the effects of space on
collection efficiency.
b) Conduct laboratory studies
to better define conditions
under which corona quench-
ing will occur when high
concentrations of fine
particulate are present.
c) Use theoretical calcula-
tions and available data, and
attempt to determine re-
lationship between collec-
tion efficiency and sec-
tionalization under space
charge conditions. Also
investigate effect of
electrode geometry. Objec-
tive is to optimize fine
particulate collection.
Total Effort 2.0 man-years
69
-------
Deposit formation
Total Effort - 1.5 man-years
Corrosion
a) Survey types of electrode
irrigation systems used
by different manufacturers
in each application area.
b) Determine location and
nature of deposit.forma-
tion problems.
c) Develop recommendations
for implementing electrode
irrigation in major appli-
cation areas.
a) Survey corrosion problems
in each application area.
b) Establish causes of corro-
sion and recommend materials
of construction for the
various existing and poten-
tial applications.
Total Effort - 1.0 man-year
With regard to the possible use of a WESP in an S0x-particulate
removal system, the previous section has shown that, with
existing WESP technology, a scrubber system appears to be
more economical than a WESP-scrubber combination. In view of
the potential energy saving associated with the use of the
electrostatic precipitation process in the collection of
particulate, there is some justification for exploring means
of reducing the capital cost of a WESP for power plant applica-
tions. At present, a WESP appears to have potential value for
such an application only when SOX removal must be accomplished
in the presence of high resistivity dust which results in
excessive dry ESP plate areas. The following research and
development efforts are suggested.
1) Determine which SOX removal systems are feasible for use
in power plants burning relatively .low S Western coals
(0.5 man-year).
2) Determine whether the candidate processes have the
capability of providing a liquor stream which is likely
to be suitable for plate irrigation (0.5 man-year).
70
-------
3) If the above item gives a positive result, design a pilot-
plant WESP test program for the power plant application to
examine the scaling problem and to study performance under
various conditions (0.25 man-year). The pilot program
would also explore means of reducing materials costs.
71
-------
SECTION VII
LITERATURE STUDY
Wet electrostatic precipitators have been used in the metal-
lurgical and chemical industries for many years. Information
on the design and application of WESP's in the industries with
the highest usage was accumulated by SRI under Contract CPA 22-
69-73. This information has been updated and expanded where
appropriate, and is presented in the following sub-sections.
IRON AND STEEL INDUSTRY
Introduction1l'l2
On the basis of existing evidence, the deliberate smelting of
ore to produce iron began between 1350 B.C. and 1100 B.C. over
a wide geographic area. In America an iron works was established
in Virginia on the James River about 1619; this was destroyed in
an Indian raid in 1622 and never rebuilt. The Hammersmith (now
Saugus), Massachusetts iron works was begun in 1645 and was the
first successful iron works in this country. Iron was then pro-
duced by the reduction of iron ore in charcoal furnaces. The
blast furnace was introduced into England about 1500 A.D. Coke
was first used as a blast-furnace fuel in England in 1619. About
200 years later - again in England - the principle of heating the
air before it was blown into the furnace was introduced, hence the
term "hot blast". The Bessemer steelmaking process eventually
supplanted all techniques before it. Closely following the
Bessemer process was the development of the regenerative-type
furnace that, now known as the open-hearth furnace, became adapted
to steel-making and evolved into the principal means for producing
steel throughout the world. Although the Bessemer converter pro-
duction exceeded that of the open-hearth until 1908, the latter
grew rapidly until, in the 1950's, 90% of the steel produced in
this country was by the open-hearth process. A third process, the
electric furnance, is a relative newcomer to the field of steel-
making and is gradually finding more and more applications in the
quantity production of quality steels. The most recently developed
steelmaking method is a pneumatic process that involves blowing
high-purity oxygen onto the surface of a bath of molten pig iron,
a method known as the basic oxygen furnace (BOF), or LD process.
Estimates for 1980 show about 25-30% of the steelmaking will be
carried out in electric furnaces, 65-70% by the BOF process, and
5% by the open-hearth process. The Bessemer converter has es-
sentially disappeared as a steelmaking method. A typical steel
plant operation in which there are about 10 major areas requiring
dust or fume control equipment is shown as a flow chart in Figure 29,
Each of these areas will be discussed in detail in subsequent
72
-------
IRON ORE
FINES
SINTERING
HIT-
MATERIAL
PREPARATION
SCRAP
BLAST FURNACE
STEEL SCRAP
PIG AND CAST IRON
PRODUCTION
PIG IRON
CAST IRON
I. BASIC OXYGEN FURNACE
2. OPEN HEARTH FURNACE
3. ELECTRIC FURNACE
4. BESSEMER CONVERTER
STEEL PRODUCTION
STEEL INGOTS
SCARFING
SURFACE PREPARTION
_^ TO ROLLING MILL
OR PURCHASER
Figure 29.
A Flow Chart Showing the Steps in the Steelmaking
Process From the Basic Raw Material and Scrap
Input to the Finished Product.1*
73
-------
sections with emphasis on process parameters relating to the use
of wet electrostatic precipitators. Coke oven gas detarring will
be discussed in a section describing collection of mists.
Blcist Furnace
The blast furnace is a cylindrically-shaped structure made of steel
and lined with refractory brick, usually about 100 feet tall and
25-35 feet in diameter. Near the bottom are nozzles or "tuyeres"
through which preheated air is blown into the furnace under pressure,
The raw materials used in the production of pig iron are semi-
continuously charged into the blast furnace. The charge consists
of iron-bearing materials (iron ore, sinter, pellets, mill scale,
open-hearth or Bessemer slag, iron or steel scrap, etc.), fuel
(coke), and flux (limestone and/or dolomite), and is introduced
into the top of the furnace through two successive cone-shaped
valves, at least one of which is always closed. The shape and
location of the lower valve (or bell) is such that the charge is
deposited uniformly around the circumference of the furnace near
the walls. As the charge melts, or is burned in the lower region
of the furnace, the stock column gradually settles, thus leaving
room for additional charges at the top. The reducing atmosphere
necessary for reduction of iron ore is generated by a reaction be-
tween the precharged coke and preheated air (1000-1700°F). The air
may be enriched with natural gas, fuel oil, or oxygen. As the
blast rises through the burden, it reacts exothermically with the
coke to produce the high temperature reducing gases which react
with the iron oxide to produce molten pig iron. The reduction
reactions can be summarized as a single reversible reaction as
follows:
Fe203 + 3CO t 2Fe + SCOa
The presence of an excess of carbon monoxide keeps the equilibrium
shifted to the right. The molten iron and slag collect in two
layers at the bottom of the furnace, the less dense slag floating
on the iron and protecting it against oxidation. The slag is
occasionally withdrawn from the furnace, and several times a day,
the molten iron is withdrawn through a tap hole into a large ladle
lined with fire brick that holds several hundred tons of the hot
metal.
The gases which pass through the charge leave the top of the
furnace through four "off takes" arranged around the dome, and
pass through a brick-lined downcomer to the dustcatcher, which is
a large settling chamber. The coarser dust particles settle in the
low gas velocity regions of the settling chamber.
74
-------
The particulate material emitted during normal operation of the
blast furnace originates from several sources, including dirt and
other fines in the charged ore, the dust caused by the downward
abrasive action of the burden through the furnace, and coke and
limestone dust. These particulates are carried out of the furnace
by the gas stream, and dust loadings range from 7 to 17 gr/scf.
A troublesome source of blast furnace emissions is that occurring
during a burden "slip". Slips are sudden movements of the burden
in the furnace occurring when a portion of the burden forms a
bridge within the furnace. The underlying burden continues to
melt and settles under the bridge and the gas pressure increases.
When the bridge opens or slips, a large volume of gas suddenly
passes through the furnace system and carries with it great
quantities of the lighter portions of the stock.
Two emissions come from the top of a blast furnace; top gas and
the dust which it entrains. After passing through the charge
the gases which leave the furnace at a temperature of around 300°F,
have a composition typically as follows:
Constituent Volume %
CO 26.2
C02 13.0
H2 1.9
CHi, 0.2
N2 58.7
The particulate material emitted during normal operation of the
blast furnace comes from several sources, including dirt and other
fines in the charged ore, the dust developed by the downward abra-
sive action of the burden through the furnace, and coke and lime-
stone dust.13
The output of particulate material from blast furnace operations
is primarily dependent on the physical characteristics of the charge,
For example, the burden can range from unscreened ore to that con-
sisting mainly of sinter, pellets, or mixtures of sinter and pellets,
Different combinations of burden, fuel, and air can be used to
effect different output characteristics. Sinter and pelletized
ore, when used as a charge material, naturally effect a marked re-
duction in dust product as compared to either screened or unscreened
ore, because of the improved physical characteristics. A good
illustration of this is shown in Figure 30, where the amount of
dust from a furnace is shown as a function of the amount of sinter
in the charge . * "*
One of the major reasons industry cleans blast furnace gas is to
render it sufficiently clean for heating coke ovens, boilers,
75
-------
UJ
C/)
O o
LU j=
O
O
Q
300
5 20°
+_
§ S
H I 100
LU <»
tr o
a.
1
I
1
0 1000 2000 3000
AMOUNT OF SINTER PLUS PELLETS CHARGED,
pounds per net ton of hot meta I
(balance of burden is iron ore )
Figure 30.
Effect of Burden Improvement on Dust Rate From
A Blast Furnace to Its Dust Collecting System14
76
-------
stoves, soaking pits, and gas engines. The gas has a heating
value of about 100 Btu/cf, but must be cleaned before being used
successfully.
During normal blast furnace operations, the dust loading of the
gas leaving the furnace is in the 7-10 gr/scf range although
loadings as high as 17 gr/scf have been reported. For each 1000
tons of hot metal produced, about 100 tons of particulate is
expected. From a 1400 ton-per-day furnace, Bishop reports 68%
coarse flue dust and 32% fine.15 The coarse dust to which he
referred was removed by greatly reducing the velocity of the gas,
and at the same time, suddenly changing the direction of the gas
stream. The dust for the above mentioned plant contained around
30% iron, 15% carbon, 10% silicon dioxide, and small amounts of
aluminum oxide, manganese oxide, calcium oxide, and other materials.
A typical chemical composition of blast furnace effluent is diffi-
cult to supply since the composition varies so greatly as shown in
Table 13. However, the major constituents and the usual com-
position range is as follows: 35-50% iron, 8-13% silica, 2-5%
alumina, 3-4% calcium oxide, 3-10% carbon, and small amounts of
alkali elements, phosphorus, zinc, lead, sulfur, and other trace
elements. Some additional data on the chemical composition of
dust samples taken from the dust catchers, wet scrubbers, electro-
static precipitators, and dust leaving the precipitator are given
in Table 14. 7 The material collected in flue dust catchers and
washers has a high iron content, and after agglomeration, becomes
a suitable material for recharging. Precipitator dust contains
less iron and several objectionable impurities. Particle size
distributions of blast furnace dust also varies depending on the
type of ore being charged into the furnace and prior beneficiation.
Table 15 presents a range of observed particle size distributions.
As stated earlier, blast furnace gas must be cleaned prior to
utilizing its heating value in order to prevent clogging of burners,
gas mains, and other maintenance problems. In a typical installa-
tion, gas from a blast furnace is generally cleaned as follows.
Upon leaving the furnace, gas is passed through a dust catcher
which removes particles larger than a few hundred microns (50-70%
by weight), leaving a dust concentration in the gas of 3-6 gr/scf.
From the dust catcher, the gas is further cleaned in a two-stage
plant. The first stage consists of a primary cleaner to separate
the coarse fractions and a secondary cleaner to separate the fine
dust. A typical combination of a two-stage cleaner consists of a
wet scrubber and a wet electrostatic precipitator in series. Typical
dust concentrations for the gas at various points in a modern gas-
cleaning system are:
77
-------
Table 1316
Chemical Analysis of Blast Furnace Flue Dusts
j Components
Fe
FeO
SiO2
A1203
MgO
CaO
Na2O
K20
ZnO
P
S
Mn
C
Pb
Cu
Alkali
Zn
Weight Percent
Range for Several
U. S. Plants
36.5-50.3
n. a.
8.9-13.4
2.2- 5.3
0.9- 1.6
3.8-4.5
n. a.
n. a.
n. a.
0.1- 0.2
0.2- 0.4
0.5- 0.9
3.7-13.9
n. a.
n. a.
-
n. a.
Midwest
Plant
47.10
11.87
8.17
1.88
0.22
4.10
0.24
1.01
0.60
0.03
n. a.
0.70
n. a.
n. a.
n. a.
-
n. a.
Range for Several
European Plants
5.0-40.0
n. a.
9.0-30.0
4.0- 15.0
1.0- 5.0
7.0-28.0
-
-
-
0.3- 1.2
- 0.1
0.3- 1.5
5.0- 10.0
0- 15.0
trace
0-20.0
0-35.0
78
-------
Table 14i7
Weight Percent Composition of
Dust Samples from Blast Furnace Gas Cleaning Plant
Sample
Com-
ponents
Insol.
Fe20,
A l*03
MnO
CaO
MgO
P
S
Cl
Zn
Na2O
K2O
Loss on
ignition
Deposit from
primary dust
catcher
8.50
75.43
1.80
0.62
1.60
0.80
0.24
trace
nil
0.28
0.56
5.90
Deposit from
secondary dust
catcher
11.60
53.00
1.55
0.60
2.40
1. 09.
0.20
.
trace
nil
0.32
0.92
23.50
Deposit frorr
washers
9.40
61.30
4.35
0.63
3.56
Dust in
gas leaving
washers
13.68
14.40
6.91
2.16
6.36
1.67 ! 8.04
0.41 i 0.54
0.41 ! 1.89
nil i 1.32
nil
0.25
0.55
15.65
1.20
9.25
20.90
-
Deposit from
precipitator
22.60
19.30
15.58
1.10
7.12
9.90
0.71
1.33
0.20
0.90
1.55
2.70
15.95
Dust in
gas leaving
precipitator
11.44
8.64
6.70
1.49
4.68
9.28
0.57
1.37
3.44
1.20
9.75
22.40
-
Deposit in
power station
burners
3.52
2.80
3.04
0.14
0.36
29.22
0.70
5.80
36.15
9.06
-------
Table 1517
Size Analysis of Blast Furnace Flue Dust
Screen Size
Mesh
20
30
40
50
70
100
140
200
-200
Microns
833
589
414
295
208
147
104
74
-74
Dust Loading
Range, %
2.5-20.2
3.9- 10.6
7.0-11.7
10.7- 12.4
10.0- 15.0
10.2- 16.8
7.7- 12.5
5.3- 8,8
15.4-22.6
80
-------
at top of furnace 7-13 gr/acf
dust catcher outlet 3-6 gr/acf
after primary washer 0.05-0.7 gr/acf
after electrostatic
precipitator 0.005-0.01 gr/acf
Wet electrostatic precipitators are normally used for secondary
blast furnace gas cleaning since the gas delivered to the blast
furnace precipitator has been cooled to saturation. The type
most widely used is a vertical flow employing pipes of 8-inch
or 12-inch diameter as collecting electrodes. Plate-type col-
lecting electrodes with horizontal flow have also been used. A
rather unique plate-type with vertical flow example is located
in the Usinor-Dunkirk steel plant in France.18
In this plant the primary gas cleaning consists of dust catcher
with a diffusor cone where deposition occurs by gravity and two
cyclones in parallel with a rectangular and tangential gas outlet.
The wet secondary cleaning consists of three secondary cleaning
lines from the two cyclones. Each line carries a scrubber and
wet electrostatic precipitator. The scrubber is a simple tower
in which the atomized water travels in the opposite direction of
the gas stream and its essential task is to lower the gas temper-
ature from about 200°C to 30°C. Each precipitator consists of a
vertical cylinder, divided into two parts which can be isolated
in operation by a damper. Each section contains a group of
vertical plates at a distance of 25 cm and in the form of a
rectangle. Between each plate are suspended ionizing filaments
fixed at the top to heated high voltage insulators. The plates
are suspended at the top from tubes through which the water, which
streams down the plates, is supplied. Intensive periodic spraying
reinforces the cleaning operation. For the three lines, and with
maximum feed rate, the grain loading after cleaning is 0.0028
grains/cu. ft.
A fairly new innovation by Lurgi known as the "Venturion Electro-
Precipitator"19, combines venturi scrubbers and the wet precipitator
in a common housing to meet the demand for additional space economy.
Other advantages are less consumption of water resulting in lower
investment cost for the water treatment equipment, and elimination
of scrubbing tower foundations. Precipitators of this type have
been installed at Yawata's Kukioka (Japan) blast furnace and at
the Tobata (Japan) No. 3 blast furnace with excellent results.
With inlet gas concentrations between 2.51 and 5.02 grams/Mm3,
collection efficiencies averaged greater than 99.9%. Compared
in cost to the dry Cottrell and the normal wet-precipitator-spray
tower combination, the "Venturion-Precipitator" costs considerably
81
-------
less (about 30%) than the Cottrell and slightly less (about 10%)
than the normal arrangement.
The more typical wet type pipe electrostatic precipitator is shown
in Figure 31. In some designs the pipes are suspended from a
header sheet at the top, with the lower end unsupported, allowing
the gas to surround the pipes. In other designs, a lower header
sheet is used to prevent the gas from surrounding the pipes. Dis-
charge electrodes are usually vertically hung, twisted square bars
(3/16 - 1/4 in.) spanning the height of the collecting electrodes.
Rapping the electrodes to dislodge dust is not required since the
collectors are flushed with liquid. An overall flow diagram of
wet cleaning blast furnace gas with a wet precipitator is given in
Figure 32. Of the 60 blast furnaces at present on basic iron
production in the British Steel Corporation, all but seven use
this three-stage cleaning system.21
Trends in the application of electrostatic precipitators for
cleaning blast furnace gas are indicated by the number of instal-
lations from 1931 to 1968 as shown in Figure 33, which presents
the number of furnaces and the number of furnaces with electro-
static precipitators installed. Precipitator inlet gas temperatures
and inlet particulate loading are shown in Figures 34 and 35. Dis-
tribution of precipitator gas velocity, field strength, and power
are shown in Figures 36, 37, and 38, and design efficiency trends
are given in Figure 39.
Figure 40 is a plot of the efficiency of blast furnace precipitators
as a function of the collection plate area to gas volume ratio. The
range of design values is given by the curves, and test data are
plotted individually as points. The design data would give pre-
cipitation rate parameters ranging from about 9-11 cm/sec (0.3-0.37
ft/sec). Test data show a fairly wide scatter band. Design values
of precipitation rate parameter w for pipe-type blast furnace pre-
cipitators are usually in the range of 0.25 to 0.45 ft/sec, while
performance values of w tend to be somewhat higher. Figure 41
gives more detail on the comparison of design and actual per-
formance by using as a basis of comparison the ratio (R) of actual
performance w to the anticipated performance w. For instance, a
value of R = 1.20 means that the precipitator size is 20% larger
than required to meet the expected performance for that particular
unit.
Electric Arc Furnaces
Electric arc furnaces employed for melting and refining steel are
cylindrical, refractory-lined structures. They range in size from
a diameter of about seven feet, with a hot metal capacity of about
four tons, to a diameter of 24.5 feet with a capacity of 200 tons.
The refractory lining used inside the steel structure may be either
82
-------
Gas
Outlet
Insulator
Compartment
Steam
Coils
Discharge
Electrode Systerr
Water Pon
and Weirs
Gas
Deflector
Approximate Scale:
5 ft = 1 in.
Figure 31. Typical Wet Type Pipe Precipitator for Cleaning
Blast Furnace Gases.
1 7
83
-------
BLAST
FURNACE
SURPLUS GAS
AIR FOR
HOT BLAST
DUST
CATCHER
ELECTROSTATIC
PRECIPITATOR
STOVE
Figure 32.
Flow Diagram for Wet Cleaning Iron Blast
Furnace Gas with Electrostatic Precipitator2'
84
-------
500
400
CO
O
JS
CO
cc
UJ
00
300
200
100
.TOTAL NUMBER OF FURNACES
NUMBER OF
FURNACES WITH
PRECIPITATORS
1920 1930 1940 1950 I960 1970
YEAR
Figure 33. Blast Furnace Statistics for Period 1920-1969.38
85
-------
oo
34-
32-
30
28-
26-
24-
^0
- 22
0
uj 20
S 18-
u.
uj 16
£ 14-
UJ
* 12-
6
4
eff:*f#ft:-ff:V#:-
p::::S::x:;:S;:x:S
60 7(
oxjxixjxwxWx
I'X'X*XvX*X'X'X*X
:;X;X:X:X;:xt:;:-:;X::;:
D 8
liilllllll
yilxiSSiiHlxiis;:;
ix^^xS-x^^:
^s^^-^^
iSSw^HS:!^
:X:>Xx'X':':'x-x-:vC'
SS^SX^KX^S:
:::i:j$i:i:i:li:ip:i:i:
X*XvX*X*X*X*X*X*
*::x:::::S::::::xS::::
0 9
j:iii:igxp£i:;i;K:8
X;>:j;j;|:;X-:«X;>X;X;r
:x:x^^?S§S:
i:j:;:i:i:::i:j:j:::|:i:j:j:i:j:j:j
:;^i$K^Wx§
:::::>x₯x₯x::::x:W
^XX;X;X;X;XvX;X\
>0 l(
|:|x|xi:;:|x|x;x|: ::₯
::::::::i:::::::::::::::x::::x
)0 1
0 IJ
20 130
-50
-40
in
JO
-30 o
0
CO
c
O
UJ
-20 §
Ul
cr
u.
-10
PRECIPITATOR INLET GAS TEMPERATURE,°F
Figure 34. Distribution of Precipitator Inlet Gas Temperature
Blast Furnace Installations.38
-------
CO
32
30-
28-
26-
24-
22-
c
o
o 20
v L^\J
£ 18-
>-" 16-
o
z
o
UJ io
o: i *-
u.
10-
8-
6-
4-
2-
n
CBu
(1) NUMBERS ABOVE BARS ARE :
(|?) TOTAL ACFS IN THOUSANDS AND
15.3
(9)
Illlllll
X-IwX"x"X*XvX*X*
;X;"x*X'>>X"XvX'>
IX'X'X'X'i'I'X'I'Xv'X'
liiiJx'iSiiiiiif-JS-i^x'
§p|||:i|||;
₯x':::₯x'x'x-::X":::
^ii^ljiii
^^iiii
:?$:5:-:S:$:5:*:S^
X**X'X*X*X'XvX*X
;Xx*X*I%*X*X"X'X*I*
X*X*I*X'X*x"X*X*X
'XvX*X*X*XvX*I*X
'x*x-x*x"x"x-Xvi':
;x*x*x*;*rx$*x*x"x
*x-x*r*x'x"xwx*x
J-xj^-rjx-x-xixix:
*X*X*X*X***I***I*X*X
:x:x:x:i:::::::::::::::x::
!;XvXv"X'I*X':*X*X
?W:?:-:₯:?x?:S-:
Illlllli
'i-S'S'i'S'X'X'i-i-i'i-i
w:S5:W:-:?:₯:?
xiii?ilil^i
NUMBER OF PRECIPITATOK5 IN
PARENTHESIS
(2) TIME PERIOD COVERED IS
1931-1969
30.7
(20)
i^igj^^^j^
X'M'X'X'XvX'X'X*
>X;XvXvX;X;XxX
::::i"i:x'i:::::::':i'x'i:l:::i
^S^^^^S
?:W:₯:S:S?:?:S
X*XvX*X"X*X*X*X*
X'x'X'X'X-X'X'X'X
XvXvXvX;X;X;X^
S₯:?S?:-%₯:-:-:§!
S:::x|:|:₯S:w:|:J₯
x:x:x'x-x'x:x'x:x
?S-:-S?:?:?:?:₯:₯
i:i:i:i:x.i:i.::i:i:i::'i:i'x::
(a) ^^ 1931-1949
(b) | | 1950-1969
'
7.4
(3) 1-1
(2) 1.9 0.7
$;:;&&$ W^^^^m^my^^m^^^
-50
-40
s£
o
-30 u]
0
UJ
(T
U.
1 1 1
-20 <
« J
UJ
cr
-10
-0
O.I 0.2 0.3 0.4 . 0.5 0.6
PRECIPITATOR INLET DUST LOADING, grs/sdcf
0.7
Figure 35. Distribution of Precipitator Inlet Dust Loading.
Blast Furnace Installations.38
-------
00
00
34
32-
30-
-
28-
26-
_
24-
22-
in
c
° 20-
o
I 18-
in
>: i6-
o
| 14-
o
£ 12-
u.
10
8-
4
^ ^^
p
£.
n
3.2
(5)
!*X*"vXvX'X'X*!*X*I*
vX*I*M*X%*XvX*I*M
S:;:ig:;:;::::S::::::::::
"x*XvX*I*X*X*XvX
6O 4
Vx\y.^T
(34)
12.5
(16)
::::::::::::::::::::::::-:::::::::::
:i:5:i%?S§S::S₯S
?:;5:;:j:i:jSS$wiS:
iiiSiiiijii^Sj^:?*
>::.:.>::.;-: <: .;.;.X-MV
vX;X;X;X;X;I\vX;X
;:;X:x:::x:x:x':':'i-x':
*t*I'X*X*X*X*'*X'I*X*'*
vxX;':'X';';">x->x->
:?S:$:vi-:₯:?:₯:-J:
X*X'X'X'I*X^*«*X*X*X
:;:ix;:;:;x:S:::::;:::::ii::
I'X-X->X'X'x'X'X'x
I*X'X*X*I**vX*X"X*X
:S::S:::::?S::::::x:::
:SilxliSiP^:^
:j:i:;:₯:i:^Si:$SJ?x'
:?l?^?S?iS:^S
:X:::::::::X::::::::;::::::::::
iwjSSJigiixSg
:::::::::::i:::i:S:x':::::':$:
:g5$xi:j:xi$i^5
i*X'X-*'X-x-x-;*Xgx"i'
'X*X*XvX*X*X*X*X*!
'X'X'x*X*I*X'X*X"'*"*'
'X*i*i*r*x*x*x*x*i'****"*'
S$S?x-S$:?ivHi:
i:::?x::i^x?S:x=«i
JJ33
(1) NUMBERS ABOVE BARS ARE:
TOTAL ACFS IN THOUSANDS AND
NUMBER OF PRECIPITATORS IN
PARENTHESIS
(2) TIME PERIOD COVERED IS
1931 -1968
(a) {ill 1931 -1949
(b) | | 1950-1968
17.9
(M)
17.2
(9)
XvXvX-X'X'X'X'X X'XvXvXvXvX-X ' 3 .6
X'X'X*X-X-X"X"X' X*X"x*x"X'X'X'X-*'l
«iiSi^iS?§! S:i:i:x::i:SS§Sx:i '3' :
x^:S:S^x'5>5:?$ :i:i:::x::i:iS§i:iS?x>: ' -^ '
SSiiSSSSiiiw :«S:::iSwx₯x:S %: ix? Zf ::: :;= ( ' ) -
;:jx;:;:;:j:jx:j:::|x:x::: x-x-x-x-xxixxiiix: :j:j: :>:xj ::: :.:: :: i::::::::::::::::::-:-:-:^::-^
3.0
6.0
9.0 12.0 15.0 18.0
PRECIPITATOR GAS VELOCITY - fps
21.0
24.0
-40
-35
-25
-20
-15
-10
-5
o
o
UJ
oc
UJ
a:
Figure 36.
Distribution of Precipitator Gas Velocity.
Blast Furnace Installations.38
-------
00
16-
_
15-
-
14-
13
12-
§ II "
o
~ 10
2
in
S 9-
"
^»
O Q
UJ
^
O 7
UJ
o:
O~"
~
5-
4-
3-
2
^
1-
csa
( 1 ) NUMBERS ABOVE BARS ARE :
TOTAL ACFS IN THOUSANDS AND
NUMBER OF PRECIPITATORS IN
PARENTHESIS
(2) TIME PERIOD COVERED IS
1931-1969
16.1 14.1
(ID (10)
13.4 7.1
(6) (6)
14.8
(10)
13.2 3.1
(2) (2)
-70
-60
o
3s
-50 >
u
z
UJ
0
UJ
-40 £
UJ
H
-30 3
UJ
oa
-20
-n
9.0 10.0 11.0 12.0 13.0 14.0 15.0
PRECIPITATOR FIELD STRENGTH (AVG),kV/in.
16.0
17.0
Figure 37.
Distribution of Precipitator Field Strength.
Blast Furnace Installations.38
-------
VC
o
17-
1C
15-
14-
I -2
lo
12-
:'>x*X'X'X'X-X'
21.0
(13)
X*X*X*XvXvyyX
vX;X;X;X;X;XvX;
c
(
(
5.2
(6)
::::::::::::::;:::::::v:::::::::i
^^3
} NUMBERS ABOVE B/
TOTAL ACFS IN TH(
NUMBER OF PRECIP
PARENTHESIS
2) TIME PERIOD COVE
1931-1969
(a) [|i|| 1931 -194
(b) | | 1950-196
5.6
(7)
3.0
(5)
iSjiliiiSiilii-Slijisj:
50 100
PRECIPITATOR INPUT POWER , watts/lOOOcfm
Figure 38. Distribution of Precipitator Input Power.
Blast Furnace Installations.38
350
35
30
25
20
o
O
UJ
CC
-15 >
_
UJ
cr
-10
-5
-------
vo
99
97
96-
,o 95
cj 94
UJ
° 93
-t
UJ
CO
UJ
Q ~
OC. "
e 91-
51
o
UJ
oc
a.
90-
70-
50-
l«
»25 1930 1935 1940 1945 1950 1955 I960 1965 IS
70
YEAR
Figure 39
Design Efficiency Trends Over the Period 1939-1969
Prorated on acfm Basis for Blast Furnace Installations.38
-------
99.7
99.6
99
£ 98
_-<
u
£ 97
c
o
96
£90
80 --
70
60
0
o Design
A Test
Design Range
From Available Data
11.3 cm
9. 7 cm/sec
0.05 0.10 0.15 0.20
Collection Area to Volume Flow, ft2/cm
0.25
Figure 40.
Relationship Between Collection Efficiency
and Specific Collection Area for Electrostatic
Precipitators Operating on Blast Furnace
Installations. 1 7
92
-------
to
32-
30-
28-
26-
24-
2 22-
o
*
S 20-
o
i is-
^
>- 16-
y
UJ ,4-
o: '2-
u.
10-
8_
6-
-
2-
0-
UfiU
1
(1) NUMBERS ABOVE BARS ARE
TOTAL ACFS IN lOOOs
(2) TIME PERIOD COVERED IS
1931 - 1969
(a)Kmffi 1931-1944
(h)l 1 1950-1969
(6.9)
"X*X*X*X*X*X'
X'l-X'X'X'X'X-
X-XvX*X'X*X*l
(6.3)
X;X;Xv';'X'X*
i'x*x x'x-x-x
Si% Sj:i:j:j:i:j:
XvX* *X*X*I'X*
(24.3)
(9.6)
SwiSiviwS
ivX'I'X'X'X'X*
:::::::::::::
lilijliillii
ixi-i-iiiiS-i-SS
|:|:||i:i:::|:i:::
X'XvXvX'X'I'
X'lvXvX'X'X
ffff'sfffffff,
x:::x:::x::::::::::
:;::::: -xji-xj
xlx-i-ivi;:::::;:::
:::::::::::::x:x:::::
i:;:j:::::::::i:;:i:::;:i
x|xi:|xW:;>x:
XvX'x'XvX*
1111111
WioiWxixl:
::::::::::::::::::::::::::
X*XvX'X\\*X
XvX'X'rX'X1
(22.1)
i::S:::x-:x-x::::::::
#:*x'*x':x£<
:::::::::::::::::::::::::::
$:$:$:$:!:$::?
i$:S5:x:x:SS
iiiiijijiiiiiiiiiiijijiSi
XvX'X'X'X'X*
XvX*XvX*X*X
(II. 1)
x^^^
x*x*x*x"x"i*x
X-X-X-X'X'X*!
$:$:$:$:$:$:-;
XxX'X'XixS
IvXvXvX'Xv
vXvX'XvXv!
(4.2)
vX*X*X*X*X*X
X*XvX"X*X*X'
0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.
(6.4)
^x-iWxWxi
'&&38B8&
>X*X*XvX*X*X*
'X*X*X*XvXvX
40 1
I f\ f\\
(6.9)
Ij*X*X*XvX*X"
50 1.
(6.8)
:::x':Si*:::::5:£
W%8&8&
X»XvX*X*X*X
X*X*X*X*X*X*1*,
-45
-40
-35
>0
5*
-30 >""
o
z
UJ
-5>S °
tO jjj
oc
u.
UJ
-20 >
H
<
_l
UJ
-15 K
-10
-5
-0
60 1.70 1.80
RATIO (R) = PERFORMANCE W/DESI6N W
Figure 41.
Comparison of Actual Performance to Design
Performance on Blast Furnaces (Basis: ratio
of precipitation rate parameters using
Deutsch equation).3 8
-------
acid or basic, but basic refractories are more commonly used.
Because of the flexibility of operation, the electric arc
furnace is used to produce steels with a wide range of composi-
tion, including carbon, alloy, and stainless grades. In 1970,
15.3% of the steel produced in the U.S. was produced in electric
arc furnaces.22
In practice, steel scrap, and perhaps hot metal from a blast
furnace, are charged into the furnace, the electrodes lowered,
and the power turned on. Power for melting and super-heating
the charged material is supplied to the furnace through externally
supported carbon electrodes which are automatically raised or
lowered through holes in the furnace roof. Intense heat is pro-
duced by the current arcing between the electrodes and the metal
charge, and this, coupled with the resistance heating that occurs
as the current flows through the charge materials, results in
melting and super-heating of the charge.
Very little air is admitted to the electric arc furnace during
the meltdown period, and during this time the furnace operates
v/ith an atmosphere containing a small amount of carbon monoxide.23
The particulate emissions from the electric arc furnaces during
this meltdown period are composed primarily of volatile matter
from the charged scrap including oil, grease, and oxides of metals
v/ith high vapor pressures. Zinc oxide from galvanized scrap is
probably the most common metallic oxide evolved during this period.
When lancing occurs after meltdown, the oxygen combines with carbon,
silicon, and manganese in the melt, to produce a furnace atmosphere
containing primarily carbon monoxide (80%) , free oxygen, carbon
dioxide, and hydrogen. Reported exit temperatures vary from 1800-
2900°F. Since the flue gases contain large carbon monoxide con-
centrations when released from the electric furnace, there is an
explosion hazard and the gases must be burned or diluted with air
before entering dust cleaning devices. The flue gas volume to be
treated by a dust collecting device ranges from a minimum of 12
to perhaps 200 times the volume of oxygen lanced into the furnace,
depending on the method used to cool and collect the waste gases.
Wide variations of emissions generated during electric furnace
steelmaking are attributable to the physical nature of scrap used,
the cleanliness of the scrap, the nature of the melting operation,
and oxygen lancing. A compilation of data on total particulate
emissions from various electric arc furnaces is given in Table 16 . 2 "*
Just as the amount of particulate matter from the electric furnace
varies with the charge and operating conditions, so does the com-
position of the particulate. Table 17 presents reported compositions
of dust emissions.25'26 . .
94
-------
Table 16 2*
Electric Arc Steel Furnace Fume Emission Data
Case
A
B
C
D
E
F
Rated
Furnace
Size,
tons
50
75
75
501
501
751
3
3
6
10
10
22
3
3
3
6
6
18
6
3
Average
Melting
Rate,
tons/hr
18.3
23.5
23.5
14.4
13.6
21.9
1.5
1.1
3.1
6.6
5.4
1.52
1.9
1.6
1.9
2.6
3.0
5.4
4.1
1.8
Cycle
Time,
hr
4
4
4
4
4
4
2
2
2
2
2
2
2
2.3
2
3
1.2
1.8
Fume
emission/ton
Melted,
Ib/ton
9.33
18. 64
7.6
6.9
12.3
12.6
7.6
10.4
5.5
5.2
13.4
4.5
5.8
5.7
15.3
12.8
6.1
29.4
12.7
Furnace Process
Basic,
Basic,
Acid,
Acid,
Basic,
Basic,
Acid
Acid
Acid
Acid,
Acid,
single slag
single slag
oxygen blow
oxygen blow
oxygen blow
oxygen blow
single slag
single slag
Refer to same furnaces as Case A.
n
Two 2-ton furnaces operating in parallel.
Q
Average for one 50-ton and two 75-ton furnaces processing normal scrap.
4Average for one 50-ton and two 75-ton furnaces processing dirty, subquality
scrap.
95
-------
m , . -I -,2 5 . Z 6
Table 1'
Electric Furnace Dust Composition
Component Weight, %
ZnO 37
Fe 25
CaO 6
MnO 4
A12O3 3
SO3 3
SiO2 2
MgO 2
CuO 0.2
P205 0.2
96
-------
Dust from electric arc furnaces tends to be extremely fine as
indicated in Table 18. These data, from several sources, in-
dicate that 70 to 100% of the fume is below 5.0 microns in size.
Because of the small particle size of arc furnace fume, only three
types of cleaning equipment, bag filter, venturi scrubber, and
electrostatic precipitator, are considered. A British Steel
Corporation survey in 1969 indicates the distribution of utili-
zation of these collectors for arc furnace fume.29 As shown in
Table 19, there is a strong preference in the United Kingdom to
use wet precipitators on the larger arc furnaces. High energy
wet scrubbers are increasingly used for 20-30 ton furnaces while
bag filters are used mostly on small furnaces. A^ paper by Holland
and Whitwam30 deals with the design problems involved in providing
a suitable technique (in this case, wet precipitation) to over-
come the difficulties of providing direct roof extraction for a
large (100 ton) electric furnace using high rates of tonnage oxygen.
Pilot plant tests at the plant described in the paper left a
choice of two types of cleaning equipment as best for a large
plant - venturi scrubber and wet electrostatic precipitator. The
venturi scrubber was eliminated because of the large pressure drop
(30 in. wg) and the necessity for varying the throat to maintain
the pressure drop at the varying volumes. The operating conditions
which had to be met are listed below:
maximum oxygen usage 2300 cfm NTP
maximum-gas volume 35,000 cfm NTP
maximum dust burden 6.5 grains/cf NTP
particle size 90% < 1 ym
maximum gas temperature 1250°C during lancing
1000°C during refining and melting
volumes at NTP lancing 35,000 cfm
melting 15,000 cfm
refining 0-5000 cfm
The wet precipitator is of the horizontal gas-flow type with two
fields in series connected to a high-voltage supply from one trans-
former selenium rectifier equipment with transductor control. If
an electrical fault develops in the precipitator, either of the
two fields can be isolated from the high voltage supply, which en-
ables partial gas cleaning to be maintained until the fault is
remedied. Stainless steel was used for the collecting and dis-
charge electrodes, the casing was tile lined, and other mild steel
surfaces in contact with the gages were protected with acid-resistant
paint. Flushing of the precipitated materials is provided by means
97
-------
Table I8
Electric Steel Furnace Fume Particle'
Size Data
Percent by Weight below Given Particle Diameter
~67! 176 3~7o57o~
micron micron micron micron
901
95 1001
901
701
71. 92
67.92
72.52
*ESP Manual27
2Muhlrad, W., "Dust Extraction from the
Fume of Electric Arc Furnaces", Iron and
Coal 183, 669-675 (1961).28
98
-------
Table 1929
FUME CLEANING FOR STEELMAKING PROCESSES 1969 (BSC)
Electrostatic
Steelmaking Precipitators High Energy Bag
Process Dry Wet Wet Scrubbers Filters
LD, Kaldo,
Rotor Converters 17
Open Hearth
Oa Lanced 32 -
Electric Arc
Furnaces, O2
Lanced
up to 60 tons
capacity - 1
over 60 tons
capacity 4 14
99
-------
of oscillating water jets located above each field and their in-
troduction is automatically sequenced to occur during the fettling
and charging period on the furnace, when the precipitator is de-
energized. The wash water is introduced at the rate of 1000 gal/min
for a period of 7 minutes.
A fairly recent innovation in wet precipitator technology for
cleaning electric furnace fume is described by Fraunfelder.3l' 32
This system consists of a hydro-scrubber (marble-bed scrubber) with
a wet electrical precipitator arranged above it in the same housing.
The fumes are pre-cooled at the scrubber inlet by direct water sprays
and passed through a bed of marbles which is sprayed with water from
below. The fumes then pass through the precipitator where final
cleaning takes place. At predetermined intervals the plates are
flushed for about three minutes. The power, supplied by standard
selenium rectifiers, 55 kV, 750 mA, is maintained during flushing.
The pilot system having yielded good results, the ferro-silicon
works of Valmoesa (Fraubunden, Switzerland) placed an order for an
industrial collection system of this type to cope with a gas volume
of 108,000 m3/hr. With an initial dust loading ranging from 0.5 to
5.0 g/m3 NTP, cleaning efficiency was measured as 98.4%. According
to the findings of the Swiss Materials Testing Institute, the dust
content of the clean gas ranged from 30-60 mg/m3 NTP, averaging
37 mg/m3 NTP.
Water requirements are 1400 1/min continuously for the Hydro-washer
(scrubber) and an additional 1000 1/min during the three min per
hour the nozzles for the precipitator are used. This gives an
overall water consumption of 90 m3/hr or 0.83 1/m3 of gas treated.
Scarfing
Before steel can be rolled, surface defects in the bloom, ingot,
and billets must be removed by scarfing, in which the skin of the
steel slab is removed. In this operation, slabs from the slabbing
mill are conveyed to the stationary scarfing machine, where they
are preheated with oxygen and acetylene for about three seconds,
after which the scarfing operation begins. The slabs pass the
cutting torches at 80 to 120 fpm where a cut of about 1/16 inch
is made on two sides of the slab. The sparks and fume are blown
downward by compressed air toward a target plate which is contin-
uously sprayed with water.
During the cutting operation a coarse, high pressure spray of water
is introduced immediately after the gas nozzles to granulate the
larger chips and wet the fume. Because of the high temperatures
generated during the scarfing operation, some of the steel is
vaporized and subsequently oxidized. The iron oxide fume is usually
quite fine.
100
-------
There are many references in the literature to wet precipitator
usage for cleaning scarfing effluent, but little detail is pro-
vided concerning precipitator design and performance. A paper
by Elliott and Lafreniere33 is the best example of wet precipitator
usage uncovered thus far. The Steel Company of Canada, Limited,
Hamilton, Ontario, investigated various collecting equipment used
elsewhere for similar applications and selected a tubular-type wet
electrostatic precipitator. Since the gases entering the pre-
cipitator are saturated with water vapor, it was felt that the
collected material could be more readily removed by water sprays
than by mechanical shakers. The choice of a pipe-type rather than
the plate-type precipitator was made to insure adequate water dis-
tribution during the washing cycle. Distribution weirs were in-
stalled as a precautionary measure to insure that the tubes were
kept clean.
The precipitator installation consists of twin units in parallel
designed to handle 70,000 cfm at dust loadings of 1.0 gr. per cu.
ft. and maximum temperature of 150°F. Actual operating conditions
indicate a flow of 56,000 cfm and temperatures varying between 40°F
and 70°F. The precipitator consists of a group of 192, ten-inch
diameter collecting pipes and 0.1055 inch diameter discharge
electrode wires (#30 stainless) enclosed in two rectangular casings
of 5/16 inch steel plate. All interior surfaces of the structure
are epoxy coated and internal parts are made of stainless steel.
Each electrode is supported from a centering clip attached to the
electrode support framework, and the lower end of each discharge
electrode is held taut by a 15 Ib cast iron weight. A frame is
provided to hold the bottom end of the electrodes in position.
The upper framework is suspended from four insulators which are
located in the respective insulator compartments. The entire
framework including the electrodes is insulated from the grounded
shell by these support insulators. To prevent dust and moisture
from accumulating on the insulators, a heating and ventilating
system is provided which consists of a fan and heater and the
necessary ducts to the insulator compartments. Warm, dry air
is blown into the compartments, and a thermostat maintains the
temperature at about 200°F.
The collecting pipes are 10" in diameter by 15* long and constructed
of 12 gauge #304 stainless steel. Each pipe is suspended from a
stainless steel header sheet and held vertically in position by an
alignment grid located at the lower end of the pipes.
The washing system consists of eight stainless steel full cone
pattern nozzles located at the top of the casing. The power is
on the precipitator at all times. After twenty blooms have been
scarfed, the bloom scarf counter energizes a relay timer which
energizes a wash timer which opens the water valve to the spray
nozzles and closes the louvers at the entry to the exhaust fan. The
louvers remain closed during the washing to reduce the gas flow to the
precipitator and thus prevent carry-over of wash water out the stacks.
The complete wash cycle is accomplished between scarfs and takes
approximately 42 seconds to complete.
101
-------
Cupolas
3 if
The iron cupola is a moderately low cost, efficient method of
continuously providing molten iron at the desired temperature
and chemistry for foundry operations. The cupola is a cylindrical
furnace lined with fireclay or firestone refractories which re-
semble a miniature blast furnace. It differs primarily in that
pig iron and steel scrap replace iron ore in the charge. Since,
in most installations, no water cooling is utilized in the melting
zone, the lining has to be repaired between periods of operation.
The charging door is located on the side of the cupola near the
top. Legs support the cupola, thus permitting the use of a drop
bottom which facilitates the removal of the remaining burden after
the last charge has been tapped.
The charge is composed of coke, steel scrap, iron scrap, and pig
iron in alternate layers of metal and coke with sufficient lime-
stone being added to flux the ash from the coke to form slag. When
properly mixed with the above materials and with sufficient com-
bustion air, a self-sustaining exothermic reaction takes place due
to burning of the coke, which provides the heat necessary to melt
the charge. Blast air for combustion is blown in through tuyeres,
located near the bottom of the furnace. The combustion air can be
handled in one of two ways, depending upon the type of blast system
used:
1. In the cold blast system, the cupola operates with
preheating of the combustion air. Temperatures at
the cupola exit usually range from 1200 to 2000°F.
2. In the hot blast system, air preheaters preheat the
blast air for the cupola. Changes in production rate
can be obtained by altering the preheat temperature
and air blast volume.
Cupola operation is usually cyclic because of the limited life of
the refractory linings. Considerable quantities of effluent, both
gaseous and particulate, are evolved, and the amount is seldom
constant except for short periods of time. The effluent rate
varies with blast rate, coke consumption, physical properties of
coke, type and cleanliness of metal scrap in the charge, coke-to-
iron ratio, bed height, burden height, and preheat temperature.
The effluent rate also changes at intervals when the furnace is
charged with iron, steel scrap, coke, and flux.
Stack gases consist mainly of carbon dioxide, carbon monoxide,
sulfur dioxide, nitrogen, and oxygen. The sulfur dioxide is
probably the most objectionable, both from a standpoint of nuisance
odor and as a cause of corrosion to the gas cleaning plant by the
formation of sulfurous acid. Carbon monoxide is usually oxidized
to carbon dioxide, though explosion can occur if this oxidation is
not complete. Among the remaining constituents of cupola waste
102
-------
gas, only fluorine has some occasional importance. The fluorine
emission can be attributed to the addition of fluorspar which is
sometimes used in basic cupolas to create favorable conditions for
desulfurizing by decreasing the slag viscosity. Only a few cupolas
are operated with a basic slag.
Particulate material emitted from the cupola stack consists pri-
marily of the oxides of silicon, iron, calcium, aluminum, magnesium,
and manganese, but may also contain small amounts of coke dust,
coke ash, limestone, zinc oxide, and smoldering products from paint,
grease, oil, rubber, and other combustibles. In Table 20 is sum-
marized a mean range of composition of cupola dust as gathered by
H. Pacyna from available literature and reported by Engels. Engels
also summarizes work by W. Patterson, H. Siepman, and H. Pacyna
which gives particulate concentration ranges and scatter values
(Table 21).
The particle size distribution ranges between wide limits, depending
on melt rate, coke usage, scrap formulation, and furnace operating
variables. Older data indicate that the proportion of dust below
10 microns, as a rule, does not exceed 10%, but later studies in-
dicate that particle sizes under 1 y may constitute 40% and more
of the total weight.
There appears to be a relationship between the coke ratio (coke/scrap)
and the melting rate per unit area of furnace and the emissions from
a cupola stack, as illustrated in Figure 42. While a substantial
amount of scatter exists, increasing amounts of coke per unit of
iron melted in basic hot blast cupolas generally increase the amount
of particulate material. Higher melting rates tend to decrease the
amount of particulate when expressed in terms of weight per unit of
iron melted. Although the data indicate a wide range of dust
emission, most of the data points fall within the range of 4 to
12 Ib/t iron.
Particle size distribution data for both hot and cold blast cupolas
are presented in Figure 43.
Though the literature search has not revealed extensive utilization
of wet precipitators for collection of cupola dusts in the United
States, there are a number of such installations overseas, some
of the more recent of which are described by Engels and Weber, and
Ussleber.3S'36
One of the first wet-type electrostatic precipitators for dust
removal in a hot blast cupola went into operation in early 1964
in a foundry in Velbert, Germany. Figure 44 illustrates the hot
blast cupola plant. The comparatively small space available and
the fact that a settling tank with pump station was already on hand
were decisive in the selection of the wet-type precipitator. The
gases not needed in the recuperator are piped directly to the gas
103
-------
Table 2031»
CHEMICAL COMPOSITION OF CUPOLA DUST
Mean Range Scatter Values
Si02 20-40 % 10-45 %
CaO 3-6 % 2-18 %
A1203 2-4 % 0.5-25 %
MgO 1-3 % 0.5-5 %
FeO (Fe203,Fe) 12-16 % 5-26 %
MnO 1-2 % 0.5-9 %
Ignition Loss
(C, S, C02) 20-50% 10-64 %
104
-------
Table 213"
DUST CONTENT OF CUPOLA WASTE GASES
Main Scatter
Range Values
Cold blast cupola:
Top gas, undiluted, g/Nm3 6-11 2-15
Waste gas from cupola stack,
g/Nm1
Total dust emission, kg/t iron 5-10 2-12
g/Nm3 2-6 1-8
Hot blast cupola (acid)
Top gas, undiluted, g/Nm3 6-14 3-25
Waste gas from recuperator,
g/Nm* 3-7 1-10
Residual gas from cupola stack,
g/Nm3 0-5-3 0.2-7
Total dust emission at cupola
furnace, kg/t iron 8-12 2-20
105
-------
C
h
(D
to
o o
Ht d
en
O rt
O
X" O
(D d
rt
oi d
rt rt
(D
W
ro d
3
*Tt
H-
O
3
1 70
K
» 60
je
8 50
a. 40
P
& 30
i-
S 20
o ,o
c/>
2 n
CONFIR
o UNCONF
8
o
MED
IRMED
% . .
f
l«_o
.r *
O.05 0.10 0.15
COKE RATE, tons per ton of iron
0.20 2 4 6 8 10 12 14 16
MELTING RATE ,tons per sq. meter/hour
-------
100
m =*: 80
2 UJ"
o y
JZ CO
3
ffi u
"§
co ^J
Q CO
N O
W UJ
LU 00
60
40
20
O.I
TOP GAS DUST FROM
HOT BLAST CUPOLAS
WASTE GAS FROM
COLD BLAST CUPOLAS
10 100
PARTICLE SIZE./zm
1000
Figure 43.
Particle Size Studies Made with
Dusts from Hot and Cold Blast Cupolas.31*
107
-------
E I ectro-
static
'rec i p i -
:ator
Combustion
Chamber -
'13
ecuperator
-83 Gas
Cooler
1- Cupo I a Stack 11-
2- R i ng Ma i n for 12-
CooIi ng CupoI a 13-
3- Top Gas Offtake m-
4- Dust Pre-Arrester 15.
5- Gas Stop i6_
6- Gas Main 17-
7- Gas Main From 18-
Other Cupolas 19-
8&9- Corresponding Mains 20-
For Excess Gas
10- Combustion Air Fan 21-
Pr imi ng Burne r
Hot Gas Main
Cascade
Fan
Waste Gas Main
Induced Draft Fan
Waste-Gas Chimney
Chimney Damper
Waste Gas Return
Butterfly Valve For
Waste Gas Return
Fresh Ai r Fan
SIudqe Co Ilect i ng
and Disposal Pond
22- Cold Blast Main
23- Recuperator By-Pass
Pass Main
24- Hot Blast Main
25- Blow-Off VaIve
26- Bust le Pipe
27- Tuyeres
28- Fan For Dust
Extract ion
Figure 44.
Hot-Blast Cupola Plant (melting rate: 6 m.ton/hr)
with Wet Electrostatic Precipitator for Cleaning
the Waste Gases.36
108
-------
cleaning plant through a by-pass piping system.
values were measured:
Dust content of the mixed gas
prior to wet precipitator 2.8 g/Nm3
The following
Clean gas content behind wet
precipitator
Collection efficiency
Dust emission
0.0474 g/Nm3
98.3%
0.384 kg/h = 0.064 kg/t iron
The gas cleaning equipment consists principally of a cooling tower
(which cools gases to about 60°C), an induced draft ventilator,
the vanes of which are sprayed to prevent scaling, and the vertical
wet electrostatic precipitator. Water consumption is 1.5 m3/h,
consumption of neutralizing agent is 16.6 kg/h, and power consumption
is 23 KW.
Engels states that the withdrawal of top gases below the top opening
is advisable in newly constructed plants where the heat content of
the gases is not utilized in the recuperator. The externally-heated
hot blast stove is independent of the top gas supply. The top
gases are drawn off below the top opening without prior dilution
with outside air for cleaning and are first introduced into an
Imatra-Venturi scrubber which provides cooling, saturation, and
preliminary cleaning. The scrubber is in combination with a vertical
wet-type electrostatic precipitator to form an integrated unit. The
measured clean gas dust content of 0.084 g/Nm3 corresponds to a
collection efficiency of 98.85%.
A wet precipitator with an Imatra-Venturi scrubber as preliminary
filter and saturator is employed for cleaning a mixture of top gas
and recuperator waste gas in an acid-lined 12 ton hot blast cupola
(Figure 45). In this plant the top gases are completely drawn off
below the charge opening, and only part of the gases are intro-
duced into the combustion chamber. With a raw dust content of
4.65 g/Nm3 and a gas volume of 15,700 Nm3/h, the system cleans the
gases to a dust content of 0.140 g/Nm3 (97% collection efficiency).
Table 22 presents a cost comparison for various dust collectors for
cleaning waste gases of a hot blast cupola. The individual data
were taken from various sources and converted by Engels so that
costs for the venturi tube dust removal device were fixed at 1
and all other collectors were expressed in relation to the venturi
scrubber.
109
-------
TOP GAS
COLLECTING PIPE
COMBUSTION
CHAMBER
Figure 45. Waste and Top Gas Cleaning with
Wet Electrostatic Precipitator and
Venturi in Series. 3I»
110
-------
Table 22. Cost Comparison for Various Dust Collectors for Cleaning Waste Gases of a Hot Blast Cupola
(Index value 1 for venturi tube dust separation)
Waste gas quantity to be cleaned: 12500 Nm3/h Depreciation: 20% annum
Venturi Tube
Wet Dust Dry
Collector, Electro-precipitator
Wet Electro-
Fabric
Dust Collector
Plant Dust
Guaranteed clean
dust content:
kg/h
g/Nm3
Initial investment
costs (Index value)
Depreciation and
interest (Index value)
Power consumption
(Index value)
Collector
3.50
0.280
1.00*
1.00*
1.00
Simple Design
3.50
0.280
1.04*
1.04*
0.62
Offer 1
1.75
0.140
3.24
3.24
0.23
Offer 2
2.60
0.2
1.68
1.68
0.36
Precipitator
1.75
0.140
1.59*
1.59*
0.24
Offer 1
2.40
0.192
1.45
1.45
0.65
Offer 2
1.50
0.120
2.36
2.36
0.43
Maintenance, ser-
vicing, water
consumption, neu-
tralization
(Index value)
Costs per operation
hour (Index value)
1.00
1.00*
1.06
0.88*
0.62
1.67
0.536
1.01
0.83
0.96
0.55 0.54
1.14 1.50
* without facilities for water supply and settling tank
-------
MIST COLLECTION IN THE IRON AND STEEL AND IN THE PETROCHEMICAL
INDUSTRY
Electrostatic precipitators are used as collectors of tars and
oils in the iron and steel and petrochemical industries. These
types of units do not usually employ forced irrigation of the
collecting electrodes, but they are "wet" in the sense that the
collected material is a fluid, and no rapping is employed. The
literature survey has not revealed any significant new develop-
ments in this application area since the preparation of "A Manual
of Electrostatic Precipitator Technology, Part II" in 1970.37
Therefore, the following discussion is taken from the Manual and
from Research Cottrell reports submitted to SRI under Subcontract
No. H-6228.38'39'63
The use of mist precipitators is most prevalent in the iron
and steel industry for collection of coke oven emissions. Coke,
the chief fuel used in blast furnaces, is the residue after
distillation of certain grades of bituminous coal. The two
methods of producing metallurgical coke are the beehive and
by-product processes. The beehive process was used for coke
production in the U.S. until about 1917, and since there were
just a few beehive oven installations where the waste heat
from the products of combustion was used in steam generating
units, no attempts at gas cleaning were made. The by-product
process is presently used in the production of 98% of all coke,
primarily because of its favorable economic aspects. Gas
cleaning equipment is necessary in the by-product process since
means are provided for full utilization and recovery of the gas
and coal chemicals. In short, all the products of combustion
and destructive distillation of the coal involved in the beehive
process are vented directly to the atmosphere while those of the
by-product process are withdrawn from the oven and processed
to produce useful by-products. A coke oven flow diagram is
shown in Figure 46.
In either type of oven, the beehive or by-product, the
coke producing processing consists mainly of driving off
certain volatile matter, leaving in the residue a high per-
centage of carbon mixed with relatively small amounts of
impurities. The beehive oven, seldom used anymore, will not
be discussed in this report. In the by-product coking process,
the coal is heated in the absence of air, and the volatile
material is piped to equipment which extracts the valuable
ingredients. After the extraction process, some of the gas
(heating value, 550 Btu/ft3) returns to the ovens for use in
heating the coking chambers and other plant processes. These
ovens are rectangular in shape, usually 30 to 40 feet long,
6 to 14 feet high, and 11 to 22 inches wide. As many as 100
of them may be set together in a battery for ease in charging
112
-------
Coke Guide
Quenching
Car
Light
Oil
Scrubber
By-Product
Precipi.tator
Oven Fuel
Precipitator
Figure 46. Coke Oven Flow Diagram38
-------
and discharging the coal and coke. Ports at the top of the
oven can receive 16 to 20 tons of coal. The ports are then
sealed and coal begins to fuse, starting at the walls of the
oven, which may generate heat from 1600° to 2100°F. When
coking is finished (18 to 20 hours), doors at the ends of the
oven chamber are opened, and a pusher ram ,shoves the entire
charge of the coke into railway cars.
As stated above, the products of destructive distillation
in the by-product process are withdrawn from the oven and processed
to produce useful by-products. Carbonization of one one of coal
yields the following products:
coke 1200-1400 Ibs
coke breeze 100-200 Ibs
coke oven gas (550 Btu/cf) 9500-11,500 cf
crude light oil 2-4 gallons
tar 8-12 gallons
ammonium sulfate 20-28 Ibs
ammoniacal sulfate 13-35 gallons
All of the above products except the coke leave the oven
in association with the gas. The analysis of the gas coming
from the coke is approximately as follows:
CO 2
02
N2
CO
H2
CH,,
' 2 ^^ ^
I TT
2.2%
0.8%
8.1%
6.3%
46.5%
32.1%
3.5%
0.5%
The concentration of suspended matter at the precipitator
inlet varies from 1 to 15 grains/scf. Water comprises about
50% of total precipitate, the remainder being tar, oil, etc.
Control Technology
As early as 1916 pilot scale electrostatic precipitators
were demonstrating the technical feasibility of removing tar
from coke oven gas, but fear of explosions caused by sparking
caused reluctance on the part of industry. Successful operation
eventually eliminated resistance, and in .1927 commercial
installations in coke oven gas were first made and rapidly
became standard equipment. The arrangement of equipment
developed to remove and recover coke oven emissions is shown
in Figure 46. The gas leaves the ovens at 1100 to 1300°F,
after which it is cooled to about 95°F, thus condensing out a
considerable amount of tar and ammoniacal liquor. These condensed
products are processed further in other equipment for separation
and refinement. The gas then passes through the exhauster and
114
-------
enters the electrostatic precipi-tator where the suspended oil
and tar are collected with an efficiency normally between 95
and 99%. The gas then moves through reheaters, ammonia absorbers
or saturators, gas coolers, and light oil scrubbers, before
entering the ga's holder. If the gas is to be used for underfiring
the coke ovens,' the gas must be further cleaned in another
electrostatic precipitator referred to as a fuel gas precipitator,
which cleans the gas sufficiently to prevent the deposition of
tar in the very small apertures of the coke oven burners. Some
of this gas is also used to ventilate the insulator compartments
of the primary precipitator. The gas volume used for underfiring
is usually a small fraction of the total gas, and consequently,
only relatively small precipitators are required.
The typical electrostatic precipitator used for detarring
consists of a group of grounded pipes six inches to eight inches
in diameter and six feet to nine feet in length, suspended from
a header plate in a round shell. Figure 47 shows the typical
design for this type precipitator. The early coke oven precipi-
tators were energized by electrical equipment using mechanical
rectifiers, thus requiring a separate substation to house the
electrical controls, transformers, and rectifiers. The more
recent utilization of tube and silicon rectifiers permits the
location of the rectifiers and transformers in weatherproof
cabinets, thereby eliminating the necessity for substations.
Figure 48 shows a self-contained tar precipitator attached to
the shell of which is a compartment containing a 35 kV electrical
set with tube rectifier. The collecting pipes of this unit are
six inches in diameter and six feet long.
Another precipitator design used for cleaning coke oven
gas is shown in Figure 49. The discharge electrode wires are
centrally located in the annulus between the concentric
tube-collection electrodes. Both surfaces of the tubes serve
as collecting surface.
Based on data from a group of installations covering
a period of 1931 to 1961, design efficiencies are generally
about 95-99%. Typical design parameters for coke oven gas
precipitators are given in Table 23.
TABLE 2338
TYPICAL COKE OVEN GAS PRECIPITATOR DESIGN PARAMETERS
148 pipes, 8 in. dia. x 9 ft. long
115
-------
4"D/a. Tar Drain
Figure 47,.
Typical Pipe Type Electrostatic
Precipitator for Collection of Tar.38
116
-------
Gas Outlet c
Gas Inlet
Receiving or
Collecting
Electrodes
Control Board
Charging and
Precipitating
Electrodes
Steam
Connections
»-Electrical
Connections
Electronic
Tube Rectifier
Transformer
Drain
Figure 48.
An Integral Tar Collecting Pipe
Type Precipitator.38
117
-------
Insulator
Drain
Hanger Rod Shield
Insulator
Compartment
Discharge
Electrode-
Spreader Bar
Tar Drain
Distributio
Tubes,
Steam Connection
Tar Drain
Gas
Outlet
mn
Hanger
Rod
fTop
O Section
t
Middle
\ Section
Gas
Flow
t)
f.''-\
-*'R\ 1
~
j
*
Bottom
Section
Access Hole
I \
Figure 49. Concentric Ring Detarrer
3 8
118
-------
Gas flow = 16,500 cfm at 100°F
Inlet concentration = 0.68 gr/cf
Outlet concentration = 0.0068 gr/cf
Efficiency = 99%
Precipitation rate parameter =13.7 cm/sec
Corona power = 190 watts/1000 cfm
Table 24 gives an average of electrostatic precipitator
design parameters for detarring coke ovens. The ratio (R) is
the ratio of performance to design precipitation rate parameters.
TABLE 2438
AVERAGE DESIGN PARAMETERS FOR COKE OVEN PRECIPITATORS
Avg. Ppt.
No. of Avg. Gas Avg. Ppt. Field Input Power
Installations Vel.(Ft/sec) Strength (kV/in.) (Watts/1000 cfm)
12 8.3 11.2 100
Avg. Ppt. Gas Avg. Ppt. Inlet Average
Temp, °F Loading (gr/scfd) Ratio (R)
113 0.3 1.11
Average precipitator cost data both on an FOB and
erected basis are presented in terms of cost per ACFM for tne
efficiency and gas volume indicated in Table 25. Numbers in
parentheses are numbers of installations used in determining
the costs indicated.
TABLE 2538
PRECIPITATOR ECONOMICS FOR DETARRING OF COKE OVEN GAS
(Time Period 1959-1969)
119
-------
Avg. Cost in
Type Gas Volume Eff. Dollars/ACFM
Precipitator Range - 100's ACFM Range - % Erected FOB
Pipe-type 0-50 95 3.55(2) 2.52(3)
(vertical flow) 50-100 95 1.43(1) 1.10(1)
Because of the limited number of installations, the above data
is only a rough indication of costs.
Detarring in the petrochemical industry is primarily concerned
with the cleaning of water gas, oil gas, and producer gas. As
in the case of tar removal from coke oven gas, the precipitation
of tars from the gases of the petroleum industry is technically
straightforward inasmuch as the precipitated tars are usually
free flowing.
A relatively new application is the use of electrostatic
precipitators to remove tar, fine carbon, and oil mist from
the acetylene gas manufactured from naphtha or crude oil.
Another relatively new application in the petrochemical
field is the utilization of precipitators for the removal
of tars from oil shale distillation gases. This application
may become more important with the recent acceleration of
oil production from shale. The purpose of the precipitator
in this application is to clean the gases which are subsequently
condensed and treated to produce petrochemical products.
In the process for manufacturing carburetted water or
producer gas, the initial gas known as "blue gas" is made by
passing steam through a bed of incandescent carbon, which
may originate from either coke or anthracite coal. There are
two cycles, designated in plant parlance as "make" and "blow".
The coal or coke is fed to the gas generator and air is admitted
under the bed during the "blow" cycle. The air is shut off
and steam is admitted for the "make" cycle. The residue ash
is then withdrawn. Carburetted blue gas is a mixture of blue gas
(or water gas, as it is often termed), made as above, and oil
gas formed by the cracking of oil in a chamber through which
the blue gas passes.
The carburetting process enriches the blue gas to as
much as 700 Btu per cubic foot of gas, depending upon the
amount of oil used. With the present day use of natural gas,
many companies mix natural gas with the above gases, the
mixture being automatically controlled by means of calorimetric
equipment. Prior to entering the precipitator the gas is
passed through a direct contact water cooler where the
temperature is reduced to around 100°F and saturated with
water vapor. In the cooler some of the tars and oils contained
120
-------
in the gas are scrubbed out by the water sprays, and sulfur
compounds and other gaseous material are removed from the
system in purifiers located after the electrostatic precipitator.
A process flow diagram of production of carburetted water gas
is shown in Figure 50.
The process for separating oil from shale is shown in
the flow diagram/ Figure 51. The shale to be processed is fed
into a retort where heat is applied to drive off the various
components such as gas, oil, and carbonaceous residue. From
the retort the gas is passed through indirect water coolers
where most of the condensable material condenses out as a
submicron sized mist. The mist-laden gas in then introduced
to the precipitator where the condensed oil and carbonaceous
residue are separated from the gas.
Natural gas is used as the primary raw material in the
production of acetylene. The gas is burned in an oxygen-limited
atmosphere under controlled temperature and pressure. The
gaseous products consist of from 50 to 55% hydrogen, from 30 to
35% carbon monoxide, and from 7 to 9%'acetylene. Because of the
explosion hazards involved, the oxygen content is rigidly
controlled to a fraction of 1%. In addition to the gases,
submicron size particulate carbon is also produced and carried
along by the gas as a smoke. Prior to entering the precipitator,
the gas is passed through a direct-contact water cooler where
the temperature is reduced to about 100°F. The gas then enters
the wet-type precipitators where practically all of the remaining
carbon is removed from the gas.
The Cottrell precipitator for detarring manufactured gases was
first developed at the plant of the Ann Arbor Gas Company at
Ann Arbor, Michigan, about 1914, while the first commercial
application recorded was on producer gas in 1916 at the Minnesota
Steel Company at Duluth, Minnesota.1*0 In initial applications
the gas was usually treated following the wash boxes or scrubbers
at low temperature, so that the tar was collected in the presence
of water. Early experimental and semi-commercial operations
soon led to regular commercial installations. In 1924 there
were five such installations in the United States cleaning
about 70 million cubic feet per day. By 1962 there were 600
tar precipitators treating 5 million cfm of fuel gas. Since
1962 only about 25 additional tar precipitators have been
installed with some of these being replacements for old precipi-
tators .
The electrostatic precipitator most commonly utilized for detarring
and cleaning of gaseous products is a single-stage vertical wire
and pipe unit, as illustrated' in Figure 52- A plate type unit is
shown in Figure 53.1*1
121
-------
to
ro
To waste Heat Boiler
Air i
Air
OToum
Feed
I
_
Generotor
i
i
Water
4
To Consumer
Water
Wash Box
Water
Scrubber
Gas Holder
Purifier
Elec. Pptr. Exhauster
Relief Holder
Tar Sump
Figure 50. A Process Flow Diagram of Production of Carburetted Water Gas63
-------
Oil Shale
Fuel Air
Recuperative
Stoves
.etort Electrostatic
Precipitator
Gas
Gas
Blower
Flue Gas
Shale Oil
roduct Gas
Figure 51. Oil Shale Retorting Process63
123
-------
High Voltage
.Insulator
Support
Insulator
Steam Coil
High Tension
Support Fram
Collecting
Electrode Pipes
Shell
High Tension
Electrode
Electrode
Weight
Compartment
To Clean
-..Gas Main
BSftfai
Gas Deflector
Cone
~ Collected ^
DD , . ., CO
Liquid
Out
Figure 52. A Single-Stage Vertical Wire and Pipe Unit38
124
-------
Central electrode centering gear
Chamber filled with penetrol
Interlocked earthing switch
H.T. cable with styrenated
terminations
Steam connectio
Interlocked hand hold
Steam pipe to tra
Gas outlet branch
Light hole
Central electrode
(serrated)
Steam trap
Electrode centering
insulator
Main H.T. insulator
Thermometer
Steam coil
Gas baffel
Electrode flushing system
Top manhole
Flushing connection
Sight hole
^Compression springs
Bottom electrode framework
and gas distributor
Purge connection
Tar drain and
seal pot
Steel plate collecting
electrodes
Stainless steel wire
discharge electrodes
Gas inlet branch
Bottom manhole
Figure^. 53. Plate-Type Precipitator Used for Detarring1*1
125
-------
The precipitator universally used for the removal of
tars and oils from carburetted water gas is a single stage
vertical pipe type unit. The design consists of the collecting
pipes, supported from a top header and immersed in the
incoming gas. The high voltage support insulators are
mounted in heated turrets attached to the side of the unit,
which is cylindrical in shape. The precipitator is provided
with gas purge connections at all of the high points in the
system. This type of unit is often referred to as a
"low-volume tar" precipitator. The discharge electrode used
in the low-volume tar precipitator is known as the "stiff rod
and prong" electrode. The most widely used form of this
type electrode consists of pointed star-shaped washers spaced
about 1 1/2 inches apart along a rigid rod for its full
length. Collecting electrodes are of the pipe type, the sizes
of which vary from 6" to 15" diameter. The discharge
electrode assembly is supported directly on the insulators
which are mounted on brackets inside of the precipitator
directly in the gas stream.
The collected tars and oil are free flowing and are
removed from the vessel sump to an external tar pot through
a liquid trap. From the tar pot the liquid is pumped to
its final point of reuse or disposal.
The precipitator most commonly used for cleaning
acetylene gas is a single-stage, vertical flow type unit.
To withstand the gas pressure the configuration of the
shell is cylindrical. The collecting pipes are supported
from a top header. The outer surfaces of the collecting pipes
are exposed to the incoming gas, while the inside surfaces
are continuously flushed by a film of water which overflows
weirs attached to the top of the pipe. The normal overflow
is augmented by periodic spraying from sprays located under
the roof.
Customarily there is only one type of discharge electrode
used in the treatment of acetylene gas, the "weighted twisted
squares" or weighted bar electrodes. These electrodes consist
of vertically hung square bars spanning the full height of the
collecting electrodes. They are usually 3/16" or 1/4" square
and are twisted longitudinally, held by weights at the bottom.
Because of the moisture content of the gas treated in the
precipitator, the discharge electrode support system consists
of heated outboard compartments. In this arrangement the
discharge electrode frame is supported by a horizontal beam
running across the top of the precipitator. The beam is
supported by two insulators, one on either side, located
in heated compartments attached to the outside of the precipi-
tator shell.
126
-------
The precipitator used to collect the oil and carbonaceous
material from shale oil manufacture consists of a single stage
vertical flow unit using pipes for the collecting surface.
The vessel construction is cylindrical. Since the temperature
of the gas being treated approaches atmospheric, no thermal
insulation is used. The collected liquid accumulates in a sump
provided at the bottom of the shell. The discharge electrodes
for shale oil application are of the weighted wire and stiff
rod and prong types, and the support frame used is the heated
outboard compartments method. The use of pipe type collecting
electrodes is common for the small sized precipitators normally
used in shale oil production.
Table 26 presents a summary of design data for electrostatic
precipitator installations in fuel gas purification from 1940
to 1963.
COLLECTION OF ACID MISTS FROM SULFURIC AND PHOSPHORIC ACID
PRODUCTION
Sulfuric Acid Production
Basically, the production of sulfuric acid involves the
oxidation of generated sulfur dioxide to sulfur trioxide and
its hydration to sulfuric acid. Sulfuric acid capacity in
1970 amounted to 38 million short tons produced in 250 plants
of which 215 are of the contact type, the remaining 35 being
the older chamber type.
The chamber process uses the reduction of NOa. to NO as the
oxidizing mechanism to convert the SOz to SOs. The oxidation takes
place in lead-lined chambers. However, the process produces
sulfuric acid with a concentration of only about 78% and this
low concentration together with high operational costs and NOX
emissions, have caused an ever diminishing use of the chamber
process. For these reasons, this discussion will deal
exclusively with the contact process of sulfuric acid manu-
facture .
The chemistry of the contact process is very simple.
First, the sulfur in the feedstock is burned in air, giving
sulfur dioxide:
S + 02 * SO 2
Then the sulfur dioxide is catalytically oxidized to sulfur
trioxide:
2S02 + 02 -» 2S03
127
-------
to
00
TABLE 2663
A SUMMARY OF DESIGN DATA OF ELECTROSTATIC PRECIPITATORS ON VARIOUS APPLICATIONS
FOR REMOVING TAR AND OIL MIST OVER PERIOD 1940 to 1963
Collecting
Type of No. of Suspended Total Gas Temperature Efficiency
Applications Install. Matter Vol. ACFM °F %
Carburet ted
Water Gas
Oil Gas
Reformed Gas
Shale Oil Gas
Acetylene
55
3
3
2
1
Tar and
Mist
Tar and
Mist
Tar and
Mist
Tar and
Mist
Tar and
Oil
Oil
Oil
Oil
Oil
315
18
3
20
42
,800
,200
,200
,900
,100
70-110
80-100
80-100
100-200
100
95
95-98
95-98.
95-97.
99-92
5
5
Mist
-------
Finally, the sulfur trioxide is absorbed in a strong, aqueous
solution of sulfuric acid:
SO 3 + H20 -» HaSCK
Contact plants are often classified according to the raw
materials charged to them; e.g., elemental sulfur, spent acid,
and hydrogen sulfide, and sulfide ores and smelter gas. The
sulfur burning plants are sometimes called hot-gas purification
plants, while those plants which utilize sulfide ores are
called metallurgical or cold gas purification acid plants.
Plants burning hydrogen sulfide may be of the hot-gas or
true wet-gas purification type, the latter case being where
the sulfur dioxide gas is not dried, thus allowing the moisture
to pass through the conversion system.
Sulfur burning plants require no gas purification (except for
hot filtration) and are the most inexpensive of the contact
type plants. No heat is required for the S02 gases, only cooling,
so that all heat evolved may be recovered as relatively high pres-
sure steam. A flow chart for this system is shown in Figure 54,
and can be divided into the following sequences:1*2
Transportion of sulfur to the plant
Melting of sulfur
Pumping and atomizing of melted sulfur
Burning of sulfur
Recovery of heat from or cooling of hot SO2 gas
Purification of S02 gas (by hot filtration)
Oxidation of S02 to S03 in converters
Temperature control to secure good yields of SO3
Absorption of SO 3 in strong acid
Cooling of acid from absorbers
Pumping of acid over absorption towers
Two types of plants are used to process sulfuric acid produced
by burning hydrogen sulfide and spent-acid. In one, the sulfur
dioxide and the combustion products are passed through gas-cleaning
and mist removal equipment. Mist removal is usually accomplished
by electrostatic precipitation and moisture removal by absorption
in concentrated sulfuric acid. The gas stream then passes through
129
-------
CO
O
f^\ -Air intake
iCf\r
silencer
IT or filter
Turbo- Starting
blower fan
r
-f^r
J
VZfifc
'/S&i
VBiWZTi
vzvzza
_ Air
Air
Exit gos
slack
To boiler
I
ABSORBING
TOWER
S
I
U-i
I
^
I
WASTE HEAT
BOILER
HOT CAS
FILTER
HEAT
EXCHANGER
4-PASS
CONVERTER
ECONOMIZER
Absorbing ocidr-
cooler I
'I
£
OLEUM
TOWER'
+
m
Drying acid "f
cooler
Sulfur
Water
Electricity
Direct operating lobor_
Steam credit.
Repairs (labor and materials).
_0.337 tens
_4,400 gol.
10 kwh
. 0.16 mon-hr.
ZOOOIbs.
60*
:]iJ
I Oleum
I to storage
Per ton of 100% acid in plant
of 200 tons daily capacity
Oleum pump
Figure 54. Typical flowchart for sulfur-burning contact plant.1*2
-------
a drying tower from which a blower draws the gas and discharges
it to the sulfur trioxide converter. A schematic of the contact
process sulfuric acid plant burning spent acid is shown in Figure 55. **3
In the other type plant, known as a "wet-gas plant", the wet gases
from the combustion process are charged directly into the converter
after heat recovery with no intermediate treatment. A highly
efficient mist recovery system following the absorber is required
since absorption is not highly efficient due to the excess moisture
and acid mist content of the gas.
The configuration of sulfide ore and smelter gas plants
is very similar to that of a spent-acid plant except that
a roaster is used in place of the combustion furnace. The
raw material used in these plants is smelter gas available
from ore roasting, smelting, and refining operations associated
with nonferrous metals production. The sulfur dioxide in the
gas is contaminated with dust, acid mist, and gaseous impurities,
the removal of which is accomplished by cyclone dust collectors,
scrubbers and electrostatic precipitators (wet, dry, or mist
type). After the gases are cooled and cleaned and the excess
water vapor removed, they are scrubbed with 98 percent acid
in a drying tower. Beginning with the drying tower stage,
the process is nearly identical to that of sulfur burning
plants. When sulfide ores are the source of sulfur, the cost
may be as much as three times that of the sulfur burning plant.
Effluent from Contact Plants
Exit gas from the absorber accounts for the major portion
of emissions from contact sulfuric acid plants. This gas
contains unreacted sulfur dioxide, sulfuric acid spray and mist,
and unabsorbed sulfur trioxide. Tail gases containing SO3
hydrate form a finely divided mist upon contact with atmospheric
moisture. The particle size of these mists can range from
submicron to 10 microns and larger. Smaller particles have
greater light-scattering effect, so as particle size decreases,
the plume becomes more dense. Contact plant emissions range
from 7 to 95 wt. % less than 3 y . A report by G. R. Gillespie ** **
gives particle size distribution data of sulfuric acid from
the waste gases of a contact acid plant, and also of sulfuric
acid made from SO3 and water under laboratory conditions,
as measured with a jet impactor. The field work was conducted
at the Krummrich Contact Plant in Monsanto, Illinois, and the
gases were sampled from a point in the exit stack of two units
approximately three feet from the absorption towers. Two
samples were extracted from each of the two units by a sampling
train consisting of a cyclone, a four-stage impactor, and a fine
131
-------
'SPENT ACID
SULFUR
FUEL OIL
)H
A WATER
STEAM
/-s
FURNACE
DUST
COLLECTOR
WASTE HEAT
BOILER
GAS
SCRUBBER
s
»
5
\\M
\s
LbR
\\\\
\\v
WATE
1
-H
R
I
f\
s
i S
1 ! l
ELECTROSTATIC
PRECIPITATORS
'111 111
li»H 1
X_P 1
JU.
/ >
V\\\\\\>Y
CDRYINGS
ll\\\v\\
^
m^
s
T
1
\
/
\\\\
SO
TRIP
L\\\\
f
BLOWER
TO
ATMOS-
PHERE
k\\\\\\\v
ABSORPTION
TOWER
*
I
1ST
IN/
1
IT
»-
OR
* MM
HEAT
cvruAMPCDC
tAutlnNutnd
IIIIL^
1
_r
J
^^^
UAAA^AUUX
pSqTTtTTTtft,
CONVERTE
r-t
/
Sf>
Y
i
I
t
J
l
\
*
^>
/^^\
P
1
I
OLFUA
(
WATER|| j
_ 98* ACID
PUMP TANK
-^-ACID TRANSFER-*
a s
C
ACID COOLERS 7
PRODUCT
-PRODUCT
93% ACID
PUMP TANK"
Figure 55.
Basic flow diagram of contact-process sulfuric acid
plant burning spent acid. **3
132
-------
glass wool filter. The size fractionations of the impactor were
1.36, 0.77, 0.39, and 0.21 microns (diameter) and the size
distributions obtained are shown in Figure 56. The flow rates
(at 1 atmos. and 150°F) in the two stacks were 6800 cfm and
15,700 cfm with respective mist loadings of 0.73 mg/1 and 0.62
mg/1.
Precipitator Design for Sulfuric Acid Plants
Electrostatic precipitators are highly effective when used
for collection of acid mist. Typically, the design for an acid
mist precipitator is of the wire-in-tube type shown in Figure 57
though the wire-and-plate design is common in Europe. The wire-
in-tube electrostatic precipitator shown in the figure contains
lead alloy tubes mounted between lead-covered tube plates. The
discharge electrodes consist of lead-covered steel cable in which
the cross-section of the sheath forms a six-pointed star to give
sharp edges to enhance corona discharge. The ionizing electrodes
are suspended from metal bars attached to lead-covered cross-
pieces supported by the heavy high-tension leads. Because of
possible shorting-out of the high voltage system, the sealing
of the high tension leads is of paramount importance. This
can be accomplished by oil seals, but, because of the risk of
fire, these have been superseded in many cases by a closed
chamber at the top of the precipitator which houses the insulators
carrying the lead-in wire and electrode connection. Condensation
of moisture on the insulator surface is prevented by feeding
dried and purified air into the chamber. Steam coils with a
water-repellant silicone film can be fitted around the insulators
to accomplish the same purpose.
Gas flows for acid mist precipitators range from about 10,000
to 30,000 cfm at gas temperatures of 100°F to 180°F. To meet
these requirements, together with a high collection efficiency,
the following basic design parameters are typically used:
Basic Parameter
Precipitation rate
Specific collection surface
Pipe diameter
Gas velocity
Corona power
Corona power density
Rectifier voltage
Range
0.20-0.30 ft/sec
200-400 sq ft/1000 cfm
10 in.
3-6 ft/sec
100-500 watts/1000 cfm
0.5-2.0 watts/sq ft
75-100 kV peak
133
-------
00
o.u
2.0
1.5
>i
PARTICLE Dl
o o c
ro OJ 4
'
x*
I/
X
/
/
/
/
/
o
>j
/
X
/'X
/
X
/
X
<%
x
^
s
RUN NO. MIST LOADING
MG./LITER
M-IO O 0.73
M-13 0.62
i i iii
O.I 0.5 I
Figure 56
10 20 30 40 50 60 70 80 90
CUMULATIVE MASS PERCENT LESS THAN Dp
95
98 99 99.5 99.9
Particle size distribution of sulfuric acid mist from commercial
contact plants.* **
-------
INTERLOCKED
ACCESS DOOR
PORCELAIN
STYRENATED
TERMINATION
TENSION
WEIGHTS
INTERLOCKED
ACCESS DOOR
HIGH TENSION
CABLE
SUPPORT
INSULATORS
Figure 57. Wire-in-tube acid mist precipitator.
3 9
135
-------
Very little design data concerning plate-type mist precipitators
has been uncovered in the literature search. However, the
Air Pollution Engineering Manual describes briefly a lead-
lined, two-stage precipitator designed to handle 20,000 cfm
tail gas from a 300 ton/day contact sulfuric acid plant.ks Con-
ditioned gas flows to the precipitator ionizing section, which
consists of about 75 grounded curtain electrodes and 100 electrode
wire extensions. The gas then flows to the precipitation section
where there are twelve 14' x 14' lead plates and 375 electrode
wires supplied 75,000 volts DC by a battery of silicon rectifiers.
All structural material in contact with the acid mist is lead clad,
and electrical wires are stainless steel cores with lead cladding.
Performance Statistics and Cost Data for Sulfuric Acid Mist
Precipitator Installations
Research Cottrell has summarized critical process and electro-
static precipitator operating parameters, and these data are
given in the following graphs and tables.1*6 Tables 27 through 31
summarize for the years 1945-1969 the precipitator gas velocity,
inlet mist loading, inlet gas temperature, and input power and
field strength. The number of installations and precipitators,
and total gas volumes involved in the statistical analyses are
also presented.
Precipitator cost data on an installed basis are presented
in terms of cost versus gas flow rate for several efficiency
levels for the period 1960 to 1969. Limited data on FOB pre-
cipitator cost are included. The data are summarized in Table 32
and Figure 58. The spread in data is indicated by the arrows
on the figure. The scatter in cost data, at a given efficiency
and gas rate, can be attributed to many factors, the most
important ones being the size of the precipitator as related to
mist characteristics, primarily particle size, and gas conditions;
the geographical location of the installation (particularly for
erected cost since labor cost can vary considerably across the
country); the pricing-profit policy of various corporations
bidding on the job; whether the installation is a backfit or
upgrading of an existing installation which may require additional
improving to "shoehorn" the precipitator into the overall
installation; and the type and degree of sophistication of the
electrical system.
The performance of sulfuric acid precipitators has been reported
in terms of the ratio of the actual value to the design value of
precipitation rate parameter (w). The ratio (R) is defined as
the ratio of performance w to the design w and is therefore the
ratio of actual performance to the anticipated performance. For
instance, a value of R = 1.20 indicates that the precipitator
size is 20% larger than required to meet the guarantee efficiency
for that particular unit. Table 33 summarizes the data used in
136
-------
TABLE 2739
PRECIPITATOR INLET MIST LOADING
Summary of Performance Statistics
Sulfuric Acid Mist Precipitator
(1945 - 1969)
Pptr.
Inlet Load
gr/SCFD
0-0.19
0.2-0.39
0.4-0.59
0.6-0.79
0.8-0.99
1.0-1.19
1.2-1.39
1.4-1.59
1.6-1.79
1.8-1.99
2.0-2.19
2.2-2.39
2.4-2.59
4.0-4.19
No. of
Install.
1
3
4
3
3
0
0
3
1
1
1
0
2
1
No. of
Pptr.
1
4
8
5
6
0
0
4
2
2
2
0
2
1
Pptr. Capacity
Thousand ACFM
In Interval
16.5
31.9
122.3
88.5
79.6
0
0
57.9
36
28.1
33
0
27
10.2
% of
Install.
In
Loading
Interval
4.3
13.0
17.4
13.0
13.0
0
0
13.0
4.3
4.3
4.3
0
8.7
4.3
Totals
23
37
531
100
137
-------
TABLE 2839
Precipitator Inlet Gas Temperature
Summary of Performance Statistics
Sulfuric Acid Mist Precipitator
(1945-1969)
Pptr. Capacity
% of
Install.
In
Inlet Gas
Temp . °F
60-79.9
80-99.9
100-109.9
120-139-9
140-159.9
160-179.9
180-199.9
No. of
Install.
0
6
5
1
4
7
0
No. of
Pptr.
0
11
5
2
9
8
0
Thousand ACFM
In Interval
0
139
37.9
12.0
95.1
221.8
0
Temperatun
Interval
0
26
21
4
17
30
0
.1
.7
.3
.4
.4
Totals
23
35
505.8
100
138
-------
TABLE 2939
Precipitator Input Power
Summary of Performance Statistics
Sulfuric Acid Mist Precipitator
(1945 - 1969)
Input Power,
Watts/1000 ACFM
0-99.9
100-199.9
200-299.9
300-399.9
400-499.9
500-599.9
600-699.9
700-799.9
800-899.9
900-999.9
No. of
Install.
0
2
2
3
2
1
2
0
0
1
No. of
Pptrs .
0
2
4
5
3
1
2
0
0
1
Pptr. Capacity
Thousand ACFM
In Interval
0
20.7
41.0
101.0
40.1
10.2
47.8
0
0
8.0
% of
Install .
In Power
Interval
0
15.4
15.4
23.1
15.4
7.7
15.4
0
0
7.7
Totals
13
18
268.8
100
139
-------
TABLE 3039
PRECIPITATOR FIELD STRENGTH
Summary of Performance Statistics
Sulfuric Acid Mist Precipitator
(1945-1969)
Field
Strength
KV/inch
8-8.4
8.5-8.9
9-9.4
9.5-10.9
10-10.4
10.5-10.9
11-11.4
11.5-11.9
12-12.9
12.5-12.9
No. of
Install.
0
1
1
0
3
0
2
2
3
2
No. of
Pptrs.
0
2
3
0
6
0
3
4
4
3
Pptr. Capacity
Thousand ACFM
In Interval
0
17.0
42
-
127.8
-
25.8
61
59.1
43.5
% of Install.
In Interval
0
7.1
7.1
0
21.4
0
14.3
14.3
21.4
14.3
Totals
14
25
376.2
100
140
-------
TABLE 3139
Precipitator Gas Velocity
Summary of Performance Statistics
Sulfuric Acid Mist Precipitator
(1945-1969)
Velocity
Interval
FPS
1.5-1.9
2-2.4
2.5-2.9
3-3.4
3.5-3.9
4-4.4
4.5-4.9
5-5.4
5.5-5.9
6-6.4
6.5-6.9
7-7.4
7.5-7.9
No. of
Install.
0
1
1
6
2
2
5
0
1
1
1
0
1
No. of
Pptrs .
0
1
11
2
4
8
8
0
2
2
2
0
1
Pptr . Capacity
Thousand ACFM In
Velocity Interval
0
28.75
8
143.3
23.2
58.2
127.9
0
33
25
25
0
16.5
% of
Install.
in
Velocity
Interval
0
4.8
4.8
28.6
9.5
9.5
23.8
0
4.8
4.8
4.8
0
4.8
Totals
21
42
488.9
100
141
-------
TABLE 3239
Summary of Sulfuric Acid Mist
Precipitator Costs 1960-1969
Efficiency
Range
0 - 10,000 cfm
1965-1969
FOB* Erected*
1960-1965
FOB Erected
10,000 - 25,000 cfm
25,000 - 50,000 cfm
1965-1969
FOB Erected
1960-1965
FOB Erected
1965-1969
FOB Erected
1960-1965
FOB Erected
90 - 95
95 - 99
99+
14.5
(1)
6.00
(1)
3.77
(1)
5.84
(4)
2.21
(1)
4.65
(1)
3.94
(2)
H
rfk
10
Efficiency
Range
50,000 - 100,000 cfm
1965-1969
FOB Erected
1960-1965
FOB Erected
> 100,000 cfm
1965-1969
FOB Erected
1960-1965
FOB Erected
90 - 95
95 - 99
99+
6.22
(1)
3.71
(3)
2.93
(1)
2.53
(5)
*Note: Costs are S/acfm. Number in parentheses is number of installations on which
contract prices were averaged.
-------
1000
co
tr
o
o
u.
o
CO
o
<
co
13
O
X
h-
co
8 100
Q
LU
CO
CC
O
O.
O
U
o:
a.
10
i I I r
DESIGN EFFICIENCY RANGE
99 +
95-99
90-95
FOB COST
95-99% EFFICIENCY
II
INDICATES DATA
SPREAD
10
100
GAS VOLUME THROUGH PRECIPITATOR, THOUSANDS ACFM
1000
Figure 53. Installed Cost of Sulfuric Acid Mist Precipitators,
1965-1969.
3 9
143
-------
TABLE 3339
DISTRIBUTION OF RATIO (R) FOR SULFURIC
ACID MIST PRECIPITATORS
(1945 - 1969)
Pptr. Capacity
% of
R = WP/WD
Interval
0.5-0.59
0.6-0.69
0.7-0.79
0.8-0.89
0.9-0.99
1.0-1.09
1.1-1.19
1.2-1.29
1.3-1.39
1.4-1.49
1.5-1.59
No. of
Install.
1
2
2
1
2
4
2
3
4
1
0
No. of
Pptrs .
2
3
3
2
4
6
2
5
7
2
0
Thousand ACFM
In Interval
36
53.8
41
57.5
50.2
76.1
20.4
57.2
96.8
36.0
0
Installations
In Interval
4.5
9.1
9.1
4.5
9.1
18.2
9.1
13.6
18.2
4.5
0
TOTALS
22
36
525.0
100
144
-------
the statistical analysis in which a total of 22 installations
consisting of 36 acid mist precipitators were examined. The
ratio for all the installations was distributed between 0.5
and 1.50.
Phosphoric Acid Production
Phosphoric acid is manufactured by two processes: (1) the
thermal process, involving the burning of elemental phosphorus
to the pentoxide, followed by hydration, and (2) the wet pro-
cess, involving the treatment of phosphate rock with sulfuric
acid. Since the literature search thus far has revealed extensive
use of mist electrostatic precipitators for acid mist collection
for the thermal process only, the following discussion will not
include information on the wet process.
In its simplest form, manufacture of phosphoric acid by the thermal
process involves three steps:
1. Oxidizing liquid elemental phosphorus in a cylindrical
combustion chamber to produce phosphorus pentoxide,
Pi» + 502 ->
2. Hydrating the phosphorus pentoxide with dilute acid
or water to produce phosphoric acid liquid and mist,
P..OIO + 6H20 * 4H3POi,
3. Removal of the phosphoric acid mist from the gas
stream, usually accomplished by scrubber, mist
eliminators, or electrostatic precipitators.
A typical flow diagram for thermal process manufacture of phosphoric
acid is given in Figure 59-1*7 The phosphorus is transferred from
the liquid phosphorus feed tank to the burner tower when the phos-
phorus is mixed with air and oxidized. The resulting phosphorus
pentoxide vapor and excess air then pass into a hydration tower
from which the product acid discharges. The weak acid collected
in the abatement equipment is recycled within the process.
The principal emission from the manufacture of phosphoric acid
by the thermal process is acid mist in the absorber discharge
gas. The particle size of the acid mist ranges from 0.2 to 2.6
microns with a mass median diameter of 1.6 microns.1*8
Collection of Phosphoric Acid Mists
Packed and open tower scrubbers, venturi scrubbers, cyclonic
separators with wire mesh eliminators, fiber mist eliminators,
145
-------
H
*fc
a\
AIR
BLOWER
PHOSPHORUS
FEED TANK
^ Pv
IP
STACK
EFFLUENT
(AIR + H3PO4 MIST)
ACID TREATING PLANT
STACK EFFLUENT
(AIR+ H2S)
t i
BLOWER
HYDROGEN SULFIDE,
SODIUM HYDROSULFIDE,
OR SODIUM SULFIDE
ABATEMENT
EQUIPMENT
PHOSPHORUS
COMBUSTION
CHAMBER
BLOWER
PRODUCT
I ACID TO
I STORAGE
'PUMP
BURNING AND HYDRATION SECTION
RAW ACID TO STORAGE
ACID TREATING SECTION
(USED IN THE MANUFACTURE OF ACID
FOR FOOD AND SPECIAL USES)
Figure 59. Flow Diagram for Typical Thermal-Pro cess Phosphoric Acid Plant1* 7
-------
high energy wire mesh contactors, and mist electrostatic pre-
cipitators have all been used to collect phosphoric acid mist
from thermal-process phosphoric acid manufacture.
The typical design for a mist precipitator for phosphoric
acid collection consists of a single-stage, vertical flow,
pipe-type unit. Stainless steel is the conventional material
of construction instead of lead because of temperatures which
may fluctuate above the softening point of lead. The collecting
electrodes, normally stainless, are of variable diameter and
require no cleaning because of the free-flowing nature of the
precipitate. Inlet temperatures are usually below 200°F with
grain loadings, which vary considerably, typically being 10-60
gr/scf (H3POi» basis) . Typical voltage requirements are
about 75 kV supplied by various electrical energization systems.
TVA has used precipitators for many years to reduce phosphoric
acid emissions, and the construction details of a large precipi.-
tator operated by the TVA are shown in Figure 60v1*9 All
stainless steel used on this unit is American Iron and Steel
Institute type 316. Electric current is supplied from a
half-wave rectifier by electrical leads entering the precipi-
tator through lead-lined oil seals which contain insulators.
A stainless steel fan pulls 9.4 m3/sec of gases at 106°C
through the three parallel sections of the precipitator,
which collects 99% of the fine mist. The precipitator is
followed by a water scrubber.
Since high concentrations of phosphoric acid mist are
often present in process streams and since the particle size
of the mist is small, large precipitators are usually required.
Two mist electrostatic precipitators are often used in series
to meet the high efficiency requirements rather than the single
section as depicted in the figure. In general, phosphoric
acid losses from a mist precipitator are affected only
slightly by the rate of gas flow or by temperature so long
as the gas flow is below design rate. However, cleanliness
of the equipment and electrical conditions employed affect
such losses materially.
ALUMINUM INDUSTRY
Primary aluminum production, as used in this report, will
include the smelting and refining of aluminum as opposed to
secondary aluminum production which is primarily engaged in the
recovery of aluminum from scrap. There are 27 primary aluminum
smelters in the United States with plant capacities ranging from
80,000 to 275,000 tons per year, the locations, capacities,
and cell types of which are shown in Table 34.50 Aluminum, pro-
duced from alumina (A1203) by electrolytic reduction in fused
147
-------
CRUSHED CORE AND
ASPHALT FILL
51 mtn STANDARD STEEL
PIPE-LEAD COVERED
ACIDPROOF BRICK
230mm
HARDBURNED SHALE
/ BRICK 200mm
k
" ISO mm CHANNEL-
LEAD COVERED
ACID
COLLECTION.
COMPARTMENT
7Mmm
0-D. CARBON
TILE /COLLAR
2S4mml.D.« 330mm
i 365m LONG
FOUNDATION
COATED WITH
PITCH
ACID
DRAIN
Figure 60. Electrostatic precipitator used for P2OS mist
removal.1*9
148
-------
' TABLE 34
PLANT CAPACITY FOR MANUFACTURING PRIMARY ALUMINUM
State and City
Annual Capacity,
Tons
Company
Reduction Cell
Types*
Alabama
Sheffield
Scottsboro
Arkansas
221,000
210,000
Reynolds
Revere
HS
PB
Arkadelphia
Jones Mills
Indiana
Evansville
Kentucky
Hawesville
Louisiana
Chalmette
Lake Charles
Maryland
Frederick
Missouri
New Madrid
Montana
Columbia Falls
New York
Messena
North Carolina
Badin
Ohio
Hannibal
Oregon
The Dalles
Trontdale
63,000
122,000
175,000
45,000
260,000
35,000
90,000
110,000
175,000
128,000
125,000
100,000
240,000
87,000
100,000
Reynolds
Reynolds
Alcoa
National-
Southwire
Aluminum
Kaiser
Gulf Coast
Aluminum
Eastalco
Noranda
Anaconda
Reynolds
Alcoa
Alcoa
Ormet
Martin Marietta
Reynolds
HS
PB
PB
PB
HS
PB
PB
PB
VS
HS
PB
PB
PB
VS
PB
Tennessee
Alcoa 200,000
New Johnsonville 140,000
Texas
Corpus Christi 110,000
Point Comfort 175,000
Rockdale 275,000
Washington
Bellingham 265,000
Longview 190,000
Tacoma 81,000
Spokane 206,000
Vancouver 100,000
Wenatchee 175,000
Alcoa
Conalco
Reynolds
Alcoa
Alcoa
Reynolds
Kaiser
Alcoa
Alcoa
all
PB
HS
VS
PB
PB
HS
HS
PB
PB
PB
West Virginia
Ravenswood
163,000
Kaiser
PB
* PB = Prebaked Anode
HS = Horizontal Stud Soderberg
VS = Vertical Stud Soderberg
149
-------
cryolite (AlF3-3NaF), passes through three stages from ore to
metal: bauxite, alumina, and primary aluminum. In most cases
the Bayer process, shown in Figure 61/ is used to refine alumina
from bauxite.51 The alumina is next sent to a primary aluminum
processing plant where it is converted to the metal by electrolysis.
Process Description
The Hall-Heroult process is used by all domestic primary aluminum
producers for the electrolytic reduction of purified alumina
to aluminum metal. The process involves the dissolving of
alumina in a bath of molten cryolite. The bath is less dense
than molten aluminum and is maintained in a molten state in a
carbon crucible known as a "cell" or "pot". The carbon crucible
serves as the cathode, and a carbon block serves as the anode.
Electrolysis decomposes the alumina into aluminum and oxygen, and
because of its greater density, the aluminum sinks to the bottom
of the cell. Periodically the aluminum metal is withdrawn, and
fresh alumina feed and bath chemicals are added by piercing the
frozen crust of the bath. This "crust breaking" agitates the bath
and is a time of increased particulate and gaseous emissions.
Classes of Electrolytic Cells Employed in Hall-Heroult Process
The variations of cell construction are based on the method of
anode manufacture, and the effluent characteristics from each
cell differ in type and quantity. Reduction cells are of three
basic types: prebaked cells which use prebaked carbon anodes,
and two types of Soderberg cells which use large single anodes
continuously baked in place over the bath. In all cell types
the anode is composed of coke bound with pitch.
Prebaked or Niagara Cell (Illustrated in Figure 62)
s z
Prebaked potlines employ multiple blocks of preformed carbon,
supported by copper and aluminum hangers to the electrical bus-
bars overhead. The bus is adjustable to compensate for anode
consumption.
Vertical Stud Soderberg (Illustrated in Figure 63)
5 3
The vertical stud Soderberg cell, because of the placement of
the anode studs in a vertical position, allows gases liberated
by the cell to be hooded most easily of the cell types. Carbon
paste is added to the top of the cell and slowly carbonized as
it moves toward the bath. The vertically positioned metal studs
carry electric current to the carbonized portion through the top.
Horizontal Stud Soderberg (Illustrated in Figure 64) 5I* -
150
-------
Ul
Bauxite
NaOH or Na2C03
Lime (If used)
Steam
Recovered
steam
Expansion
Dilution
1
Grinding
I
Mixing
Reheating
I
Solution of
alumina
under
pressure
Separation of
red muds
Final
filtering
Wash water
Washing of
red muds
Red muds
to waste
Evaporation
Temperature
exchange
Precipitation
of Al (OH)3
Separation of
Al (OH)3
r
Gas or fuel oil
AL (OH).
Priming
Wash water
Calcination
T
Washing of
Al (OH)3
Calcined commercial
Alumina
Figure 61. Diagram of the Bayer Process51
-------
ANODE BUS
ALUMINA (ORE) BIN
CRUST BREAKER
RISER BUS TO
NEXT CELL
rr^
.! n^
CARBON
ANODE
J-.-J-.V
I
STEEL CRADLE
SIDE HOOD FOR
'VENT CONTROL
ALUMINA
CRUST
CRYOLITE BATH
MOLTEN
ALUMINUM
CATHODE
RING BUS
FLOOR
STEEL CATHODE
COLLECTOR BAR
ALUMINA INSULATION
Figure 62. Prebake Reduction Cell Schematic Arrangement52
152
-------
ANOOE ROD
ANODE BUS
TO EFFLUENT
COLLECTION
SYSTEM
FLUID PASTE
PARTIALLY BAKED
ANODE,/^xV
STEEL ANODE
STUD
ANODE CASING \ n
GAS COLLECTING
SKIRT
MOLTEN
ELECTROLYTE
MOLTEN ALUMINUM
CATHODE
CATHODE BUS
RAMMED
CARBON
STEEL CATHODE
COLLECTION BAR
CARBON
BLOCK
LINING
STEEL
CRADLE
Figure 63. Vertical Stud Soderberg Cell Schematic Arrangement53
153
-------
PASTE COMPARTMENT COVER
ALUMINA HOPPERS
REMOVABLE
CHANNELS
ALUMINA
CRUST
STEEL SHELL
INSULATION
CARBON LINING
PASTE COMPARTMENT
CASTING
POT ENCLOSURE
DOOR
ANODE STUDS
GAS AND FUME
' EVOLVING
CATHODE
COLLECTOR BAR
MOLTEN ALUMINUM
Figure 64. Horizontal Stud Soderberg Cell Schematic Arrangement1
154
-------
Hood requirements for the horizontal stud Soderberg are much more
difficult because of the location of the studs at the side of
the electrode rather than at the top. Since access to the cell is
necessary for electrode pin removal and replacement, burning of
the gases and tars is not possible, and tar-fouling becomes a
problem. Therefore, when electrostatic precipitator usage is
attempted, the plates require water flushing to prevent fouling
by tars.
Effluents originate at several processing operations in an
aluminum smelter. Since the most difficult to control come from
the potlines, this report will emphasize those effluents originating
from the prebaked and Soderberg cells. The airborne effluents
from the electrolytic process consist mainly of:
CO and C02, formed when the released oxygen reacts
with the carbon anode and escapes as CO or C02;
SOX/ from sulfur present in the petroleum coke and
pitch used to make the anodes;
Particulates of vaporized bath materials;
HF gas, resulting from the hydrolysis of fluoride
salts;
Alumina, cryolite, and aluminum fluoride dusts,
entrained from the bath crust;
Hydrocarbons, volatilized from the binders used to
make the anode.
The quantity and type of fume emitted from reduction cells
depend greatly on the cell design and operating techniques used.
Reported determinations of particle size distributions of the dust
and fume emissions collected from prebaked and horizontal Soderberg
cells are shown in Figure 65.5S Two plots are shown for prebake
potlines, one the average of four samples of pot emissions, the
other the average of five samples of electrostatic precipitator
intake. Another size distribution of horizontal spike Soderberg
cell emissions is presented in Figure 66.56 This typical size
distribution was determined with a cascade impactor at the exit
of the cell hood on potline 4 at the Kaiser Aluminum and Chemical
Corporation plant in Tacoma, Washington. Microscopic analyses were
performed to verify the theoretical predictions of the University
of Washington cascade impactor. The size distribution is the result
155
-------
Weight % Larger Particles
98 95 9O 80 70 60 50 40 30 20 10 5 2
100
50
10
M
35
5
o
a.
1.0
0.5
O.I
PB
\ X
PB
HSS Total Solids
HSS
PB Total Solids
1 i i
I I I
5 10 20 30 40 50 60 70 80
Weight % Smaller Particles
90 95 98
Figure 65. Particle Size Weight Distribution,
Potline Primary Effluent55
156
-------
100
50
10
b
E
_o
o
o
Q.
0.5
Aluminum reduction cell
air collection flowrate
= 1,600 -scfm
Collection air temperature
= 290°F
0.2 L
J_
I
I
J I
I
I
_L
10 20 30 40 50 60 70 80 90
Percentage of Particles Smaller by Weight
95
Figure 66.
Composite Particle Size Distribution
by Weight for Aluminum Reduction Cell
Air Emissions (Kaiser Aluminum and
Chemical Corporation Plant at Tacoma,
Wash., Potline 4) at the Exit of the
Reduction Cell.56
157
-------
of combining all the size distribution data (140 tests) and taking
into account the percentage of time each cell operation takes during
a day. Particle concentration and size distribution were found to
change significantly with changes in cell process operation.
Effective control of aluminum potline effluents involves
three types of collection systems which can be classed as: 1)
capture of all effluents by potline hooding, 2) subsequent
collection of these captured pollutants by existing control
devices, and 3) scrubbing of roof "monitor" emissions. This
report will deal with those collection devices utilized for
collection of potline effluents which have been hooded.
While the overall pollution potential of a primary
aluminum facility is not Significantly affected by the anode
configuration, the choice of the associated air pollution control
systems is closely related to the cell type. Therefore controls
will be discussed separately for the prebake and Soderberg cells.
Contrpl devices for vertical stud Soderberg cells have
included multicyclones and spray-type scrubbers. Exhausts may
be treated by bag filters coated with lime or alumina, or by
electrostatic precipitators, but the residual tar creates a
fouling problem in the collection system. A collection system for
Montecatini Edison, a plant in Milan which incorporates dry
cyclones, a dry electrostatic precipitator, and two scrubbers,
is shown in Figure 67.57
As stated earlier, hooding for the horizontal stud Soderberg
cell is inherently less complete and the large volumes of air
entrained during replacement and readjustment of studs creates a
dilute mixture of hydrocarbons which encourages a tar-fouling
problem in ducts and control equipment. Existing controls have
consisted of scrubbers, but where electrostatic precipitators are
attempted, the plates require water flushing to prevent fouling
by tars.
Particulate emissions from prebake pots contain none of the
tar found in the horizontal stud Soderberg; instead "dusting" of
the carbon anode produces carbon particles in addition to alumina,
etc. Current controls have consisted of dry-type cyclones or
electrostatic precipitators followed by wet scrubbers.
158
-------
vo
1. Gas suction skirt
2. Burners
3. Cyclones
4. Filter screen
5. Electrofilter
6. Scrubber
7. Chimney
Figure 67. Anodic Gas Purification Plant Flowsheet57
-------
Table 36 summarizes the removal equipment considered for
emission control of the different types of effluent streams
in the aluminum industry.58
Table 37 presents source sampling test results of a number
of plants which were selected on the basis of representation
of best control technology. The program included testing of
potline installations at two vertical stud plants, three pre-
bake plants-, and one horizontal stud plant. A legend is given
in Table 38 for the various abbreviations used.59
Cost data for control equipment in the primary aluminum industry
are sparse, but Singmaster and Breyer have presented data de-
rived from purchase costs reported by equipment manufacturers.
Of specific interest is cost data for wet electrostatic pre-
cipitators used for control of vertical and horizontal stud
cell types (shown in Figure 68).60 It can be seen that the
recent data of Bakke indicates much higher costs than the data
obtained by Singmaster and Breyer.61
Use of Wet Precipitators for Collection of Fume from Electrolytic
Reduction Cells
Martin Marietta Aluminum -
Several wet-wall precipitator units of the Mikropul design are
used to collect fume from vertical spike Soderberg cells at the
Dalles, Oregon. Two sizes are employed, which are capable of
treating 2.83 m3/sec (6,000 acfm) and 5.66 m3/sec (12000 acfm).
The precipitators have two plate sections in series with sprays
at the entrance of each section and above the plates.
A burner is installed in the stack, but it does not remove all of the
condensable hydrocarbons. A scrubber precedes the precipitator/
which reportedly removes about 95% of the gaseous contaminants.
Performance data from this installation, as reported by a representa-
tive of Martin Marietta to SRI personnel, is given in Table 35.
TABLE 356I»
TYPICAL PERFORMANCE DATA OF WET
PRECIPITATOR AT THE DALLES
Efficiency
gr/scf mg/m3 Range
Inlet 0.13-0.20 298-458
Outlet 0.0024-0.003 5.49-6.87 97.7-98.8
160
-------
TABLE 3658
CONTROL EQUIPMENT CONSIDERED FOR THE PRIMARY ALUMINUM INDUSTRY
Burner
Incinerator
Multiple Cyclone
Baghouse Filter
Fluid Bed Dry Scrubber
Coated Filter Dry Scrubber
Injected Alumina Dry Scrubber
Dry Electrostatic Precipitator
Wet Electrostatic Precipitator
o>, Spray Tower
I-1 Spray Screen
High Pressure Spray Screen
Wet Centrifugal Scrubber
Venturi
Chamber Scrubber
Wet Impingement Scrubber
Cross Flow Packed Bed
Floating Bed (Bouncing Ball)
Sieve Plate Tower
Self-Induced Spray (Bubbler)
Vertical Flow Packed Bed
PREBAKE POTLINES
Prim.Sec.Sec.
no with no
Sec. Prim. Prim.
A
A
A*
A
A
B
B
A
B
A
A
B
VS POTLINES
Prim.Sec.
no with
Sec. Prim.
A
A*
A*
A*
A
A
B
D
A
B
B
A*
A*
A
A
A*!/
HS POTLINES
Prim.Sec.
no with
Sec. Prim.
ANCILLARY PROCESSES
C
A
B
D
D
B
C
A
Bake
Plant
Dry
Mtls.
A
A
A
C
A
B
B
B
A
Paste
Mix
Cast
House
B
A
B
A
B
B
B
B
B
Prim. Primary collection system.
Sec. Secondary or potroom system.
A In current use in the United States.
A* In current use outside the United States.
A*l/ Used in one foreign plant. Not considered
economically feasible in the United States.
B Considered feasible but not known to be in use.
C In development stage.
D Superseded by other equipment.
Considered not feasible, economically and/or
technically.
-------
TABLE 3759
ISJ
RESULTS OF EPA SOURCE TESTS FLUORIDE AND PARTICULATE
PRIMARY ALUMINUM INDUSTRY
Ibs/Ton Aluminum *
PARTICULATES
Primary Collection
Secondary Collection
Primary Emission
Secondary Emission
Total Emission
Primary Efficiency(*)
Primary Collection
Secondary Collection
Primary Emission
Secondary Emission
Total Emission
Primary Efficiency(%)
ANODE BAKE PLANTS
Plant A
Cell Type VS
Primary Control Device BS-ST
Secondary Control Device SS
ion
ction
i
ion
ncy (%)
iency(%)
Efficiency (%)
articulates & Gaseous)
ion
ction
n
ion
ncy(%)
iency(%)
Efficiency(%)
Plant
Control
(5)
ssions
Emissions
NS
-
-
-
-
-
-
53.50
3.65
1.65,
1.651
3.30
96.91
54.79
94.22
G
ESP
1.56
0.88
Al
VS
BS-WESP
SS
91. 26(67. 60)2
27.12(26.68)
0.12 (0.11)
9.53(5.83)
9.65(5.94)
99.86(99.84)
64.85(78.15)
91.85(93.70)
37.69
3.02
0.01
0.81
0.82
99.97
73.17
97.98
H
ESP
3.96
1.25
A2 B BI
VS PB PB
BS-WESP FBDS FBDS
SS None None
NS 110.10 100.30
NS NS
13.79 1.80
NS NS
_ _ _
87.46 98.19
NC NC
_
NS NS 37.80(48.05)
NS - 1.20
0.02 - 0.14
2.05 - 1.20
2.07 - 1.34
99.94 - 99.62(99.54)
32.12 - NC
94.92 - 96.89(97.46)
C D D,
PB PB PB
ST FBDS FBDS
SS None None
NS 72.26 NS
NS
4.00
NS
_ _ -
94.43
NC -
"* ~* ~
28.10 61.40 NS
9.06, 1.17 NS
69. 643 0.87(0.49)4 0.35
7.30 1.17 1.17
76.94 2.04(1.66) 1.52
98.64(99.23) 99.11
19.42 NC NC
96.89(97.46) 94.95
E F
HS VS
ST-WESP ST-WESP
None None
81.80 38.20
NS NS
5.95 1.34
NS NS
NS NS
92.73 96.58
NC NC
_
46.53 NS
2.06
0.41
2.06
2.47
99.11
NC
94.95
'Refer to legend on following page.
-------
TABLE 38
LEGEND FOR TABLES AND FIGURES
VS - Vertical stud Sbderberg cell
PB - Prebake cell
HS - Horizontal stud Soderberg cell
BS - "Bubbler" scrubber
ST - Spray tower
WESP - Wet electrostatic precipitator
SS - Spray screen
FBDS - Fluid bed dry scrubber
NS - No sampling
NC - No Control
(1) - Average of two tests; one test suspected to be
contaminated and deleted.
(2) - Average of two tests; one test deleted due to
stud blow during test.
(3) - Samples suspected to be contaminated during
sampling; these plant data are suspect.
(4) - Data with two tests deleted due to suspected
control system malfunction.
(5) - Anode plant emission units are Ib/ton anode
produced.
163
-------
1000
0>
CO
o
o
o
o
o
CO
o
o
UJ
CO
<
o
(E
10
(a)
(a)51«vs
i i i i i i
i i i
Figure 68.
10 100
GAS VOLUME FLOW, l,000'« acfm
Cost Data for Wet Precipitators
1000
164
-------
The specific collecting area of these units, not including the
exit baffles is about 53.2 m2/(m3/sec), Or 270 ft2/1000 cfm.
Typical electrical operating conditions are given in Table 39
The liquid to gas ratio used in the precipitator sprays was
about 1.0 1 of liquid per m3 of gas (7.5 gal/1000 cf).
TABLE 396
-------
A similar wet precipitator installation is employed by
Martin Marietta at Goldendale, Washington. Table 40 gives outlet
data obtained by the State of Washington in October of 1972 at
this location on the No. 1 precipitator, and Table 41 compares
the wet precipitator emissions with other sources of emissions at
the plant.62
TABLE 4062
EMISSION MEASUREMENTS ON ELECTROSTATIC PRECIPITATOR
NO. 1, MARTIN MARIETTA PLANT AT GOLDENDALE, WASHINGTON
Mass Loading Sampling
Benzene Sol. Benzene Insol. Total Time
ft3 Sampled mg/m3 gr/ft3 mg/m3 gr/ft3 mg/m3 gr/ft3 Minutes
1437 1.00 0.00044 3.02 0.00132 4.03 0.00176 222
166
-------
TABLE 4162
SUMMARY OF MEASURED EMISSIONS
SOURCE
SAMPLING
LOCATION
POLLUTANT
MEASURED
CONCENTRATION
EMISSION
RATE
ALLOWABLE
EMISSION
EMISSION
FACTOR
METHOD
Pot room
Hoods around
anode bases
Pot room
Hoods around
anode bases
Pot room
Hoods around
anode bases
Hoods around
anode bases
Pot room
Hoods around
anode bases
Pot room
Hoods around
anode bases
Pot room
Roof monitor
Electrostatic
precipitator
Roof monitor
Electrostatic
Precipitator
Roof monitor
Electrostatic
Precipitator
Electrostatic
Precipitator
Roof monitor
Line B, Sec. 1
Electrostatic
Precipitator
No. 1
Roof monitor
Line B, Sec. 1
Electrostatic
Frecipitator
No. 1
Roof monitor
Particulate
Particulate
Extracted
Particulate
Extracted
Particulate
Fluorides
(as HF)
Fluorides
(as HF)
SO 2
CO
CO
Hydrocarbons
(as C»O
Hydrocarbons
3-4
Benzopyrene
.00063 gr/ft3
.00176 gr/ft3
.00026 gr/ft3
.00132 gr/ft3
203 ug/m3
98 ug/m3
26 ppm
24 ppm
12750 ppm
6 ppm
75 ppm
158 Nanograms/M3
69.1 Ib/hr
2.54 Ib/hr
28.5 Ib/hr
1.91 Ib/hr
10.15 Ib/hr
.62 Ib/hr
47.6 Ib/hr
1496 Ib/hr
9523 Ib/hr
214 Ib/hr
32 Ib/hr
.0083 Ib/hr
15 Ib/ton
total
To meet
V7AC 18-48
ambient
& forage
standard
5.38 Ib/ton
.20 Ib/ton
2.22 Ib/ton
.15 Ib/ton
.79 Ib/ton
.05 Ib/ton
3.65 Ib/ton
117 Ib/ton
742 Ib/ton
16.7 Ib/ton
2.5 Ib/ton
.00065 Ib/ton
Intermediate
Volume
Sampler
Intermediate
Volume
Sampler
(benzene
extract)
NaOH
absorption
& specific
ion electrode
NaOH
absorption &
turbidometric
Grab sample
& non -disper-
sive infra-
red
Grab sample
& flame
ionization
I.V. Sampler
+ Benzene
Extraction +
Adsorbtion
Chromatography
+ Color ime trie
* Based on average daily production of 308 tons aluminum
-------
SECTION VIII
SUMMARY OF STATUS OF WESP TECHNOLOGY AND DESIGN METHODS
Wet electrostatic precipitators have been effectively used
to collect particulate matter in the aluminum, iron and steel,
chemical, and petrochemical industries. Duct-type collecting
electrodes are generally used for the applications in which re-
latively large gas volumes are to be cleaned. The advantages of
plate electrodes as opposed to pipe electrodes are less expensive
construction and the flexibility of allowing variations in
electrical sectionalization. Pipe electrodes can be more uniformly
irrigated than plates, and gas by-passage (sneakage) is not a
factor. Saturation of the gas stream with water prior to entering
a WESP is desirable with all electrode geometries to prevent deposit
formation and to maximize effective utilization of the precipitation
process for removing particles.
The largest recent application of WESPs has been in the
collection of fume from aluminum pot lines. A field test of a
full scale duct-type wet ESP installed on an aluminum pot line
indicated that collection efficiencies of 95 to 98% were achieved
on a particle size distribution in which about 65% of the mass
consisted of sub-micron particles. The specific collection area
during the test period was approximately 62 m2/(m3/sec). Corrosion
and deposit formation in the liquor supply system have occurred
at this installation.
In the iron and steel industry, WESPs are used on electric
arc furnaces, blast furnaces, basic oxygen furnaces, scarfers, and
cupolas. In some of the electric arc furnace WESP installations,
intermittent washing of the plate-type electrodes was found to
be preferable to continuous irrigation. These units have
either two or three fields in series, and the field undergoing
washing is de-energized during the washing period. This is in
contrast to the aluminum pot line installations, which employ
continuous irrigation. Wet precipitators of the wire and pipe
design have been widely used for cleaning blast furnace gas,
but this is a declining application area because of the recent
trend toward the use of blast furnaces with high top pressures.
The energy available with the high pressures can be utilized in
high energy particulate scrubbers with a lower capital investment
compared to precipitators.
168
-------
Electrostatic precipitators are used as collectors of tars
and oils in the steel and petrochemical industry, and for col-
lection of acid mists from sulfuric and phosphoric acid production
in the chemical industry. These types of units do not usually
employ forced irrigation of the collecting electrodes, but they
are "wet" in the sense that collected material is a fluid, and
no electrode rapping is employed. Detarring ESPs are typically
of the wire and pipe design, with the pipes suspended from a
header plate in a round shell. Acid mist precipitators are also
typically of the wire in tube type, but the wire and plate design
is common in Europe.
Design methods for sizing wet electrostatic precipitators
have included pilot plant studies and design by analogy based on
the empirical application of the Deutsch equation with a "precipitation
rate parameter" used as an indicator of performance. For the blast
furnace and sulfuric acid mist applications, the data in the
literature concerning design versus installed performance is
reported as the term (R) which is the ratio of the precipitation
rate parameter obtained from test data to that used in designing
the unit. For example, a value of R of 1.20 means that the
precipitator plate area is 20% larger than required to meet the
design efficiency for that particular unit. The scatter in the
R values (typical range, .5 to 1.5) indicates the need for more
precise sizing methods.
The agreement obtained between the laboratory data and
the predictions of the mathematical model (Section III) suggests
that the mathematical model offers a tool which can be used to
estimate plate area requirements for various wet precipitator
application areas. The computer model in its present state of
development is useful for qualitatively predicting performance
trends caused by changes in specific collecting area, electrical
conditions, and particle size distributions. Current density,
applied voltage, and the particle size distribution are the most
important variables in the calculation of overall mass collection
efficiency for a given specific collection area. A detailed
discussion of the mathematical model is given in Reference 1.
169
-------
ACKNOWLEDGMENTS
The design of the wet precipitator was performed by Mr.
Norman H. Francis, Research Engineer, and the laboratory-experimental
work was performed by Mr. Leon B. Hill, Research Technician. The
European visits were conducted by Dr. Charles E. Bates, Head of the
Metallurgy Section.
170
-------
SECTION X
REFERENCES
1. Gooch, J., J. McDonald, and S. Oglesby, Jr. A Mathematical
Model Of Electrostatic Precipitation. Southern Research
Institute, Birmingham, Alabama. Contract No. 68-02-0265.
The Environmental Protection Agency, Research Triangle
Park, North Carolina. April 1975.
2. Gooch, J. and J. McCain. Particulate Collection Efficiency
Measurements On A Wet Electrostatic Precipitator. Southern
Research Institute, Birmingham, Alabama. Contract No.
68-02-1308. The Environmental Protection Agency, Research
Triangle Park, North Carolina. March 1975.
3. Bakke, Even. The Application of Wet Electrostatic
Precipitators For Control Of Fine Particulate Matter.
Paper presented at the Symposium on Control of Fine
Particulate Emissions from Industrial Sources for the
Joint U.S.-U.S.S.R. Working Group, Stationary Source Air
Pollution Technology. San Francisco, California,
January 15-18, 1974.
4. Oglesby, S. and G. Nichols. A Manual of Electrostatic
Precipitator Technology. Part II - Application Areas.
Prepared under Contract CPA 22-69-73 for NAPCA. Southern
Research Institute, Birmingham, Alabama. August 25, 1970.
5. Faust, et al. Principles of Unit Operations. New York,
John Wiley & Sons, 1960. p. 170.
6. Gooch, J. and J. McDonald, op. cit.
7. Epstein, Michael. EPA Alkali Scrubbing Test Facility:
Summary of Testing through October 1974. Report to U.S.
E.P.A., Contract No. PH 22-68-67. Bechtel Corporation.
San Francisco, California. June 1975.
8. Droplet Measurements at the Colbert Scrubber Pilot Plant
Performed by KLD Associates, Inc. KLD Technical Memorandum
No. 14. August 1974.
9. Ensor, et al. Evaluation Of A Particulate Scrubber On A
Coal-Fired Utility Boiler. Meteorology Research, Inc.,
Altadena, California. Contract No. 68-02-1802. The
Environmental Protection Agency, Research Triangle Park,
North Carolina. November 1975.
171
-------
10. Calvert, S., J. Goldshmid, D. Leith, and D. Mehta.
Scrubber Handbook. Ambient Purification Technology.
Riverside, California.
11. United States Steel. The Making, Shaping, and Treating
of Steel, McGannon, Harold E. (ed.). 8th Edition, 1964.
12. Oglesby, S. and G. Nichols, op. cit.
13. Uys, J. and J. Kirkpatrick. The Beneficiation of Raw
Material in the Steel Industry and Its Effect Upon Air
Pollution Control. J. Air Pollut. Control Assoc. 13;
20-27, January 1963.
14. Varga, J. and H. Lownie. Final Technological Report on
A System Analysis Study of the Integrated Iron and Steel
Industry, to Division of Process Control Engineering,
NAPCA, from Battelle Memorial Institute. May 15, 1969.
15. Bishop, C. A. Metallurgical Furnace Stacks. AIHA
Quarterly 1^:34-39, March 1950.
16. Oglesby, S. and G. Nichols, op. cit.
17. Oglesby, S. and G. Nichols, op. cit.
18. Spreux, M. and L. Gagnaire. The Dust Extraction Equip-
ment at the Usinor-Dunkirk (France) Steelplant. Trans-
lated from Revue de Metallurgi 6£(11):1103-10, 1963.
19. Scheidel, K. Electrical Gas Cleaning in Heavy Industry.
Metallyesellschaft Mitteilungen (Metallgesellschaft
Review of the Activities), no. 4, 33-44, 1962.
20. Otsubo, S., et al. Operation of Newly Installed Lurgi
Venturion Type Blast Furnace Gas Precipitator. Seitetsu
Kenkyu, no. 249, 148-155, December 1964.
21. Speight, G. Air Pollution Control in the British Iron
and Steel Industry. Steel Times 200:395-402,407, May 1972.
22. American Iron and Steel Institute, Annual Statistical
Report, 1970.
23. Walters, C. Air Pollution Control for the Metals Industry,
Republic Steel Corporation. June 4, 1968.
24. Oglesby, S. and G. Nichols, op cit.
25. Systems Analysis of the Integrated Iron and Steel Industry.
Report to Battelle Memorial Institute by Swindell-Dressier
Company, Pittsburgh, Pennsylvania. NAPCA Contract
PH 22-68-65. March 31, 1969.
172
-------
26. Oglesby, S. and G. Nichols, op. cit.
27. ibid.
28. Muhlrad, W. Dust Extraction from the Fume of Electric
Arc Furnaces. Iron and Coal T83;669~675, 1961.
29. Davis, C. Gas and Liquid Filtration in the Steel
Industry. Filtration Separation 8^(4) :418-424, July/
August 1971.
30. Holland, M. and K. Whitwam. Direct Fume Extraction for
Large Arc Furnaces, Fume Arrestment Spec. Report 53,
William Lea and Co. 1964. p. 150-159.
31. Fraunfelder, A. New Development in Cleaning Equipment
for Fumes Emanating from Ferrosilicon Electric Furnaces.
Tidsskrift for Kjerni Bergvesen Metallurgi 23;110-114,
May 1963.
32. Fraunfelder, A. Experience Gained with a New Scrubber-
Precipitator Combination. Krupp Technical Review 22 (3);
125-126, 1964.
33. Elliott, A. and A. Lafreniere. The Design and Operation
of a Wet Electrostatic Precipitator to Control Billet
Scarfing Emissions. Paper #71-159 presented at the 64th
Annual Meeting of the Air Pollution Control Association,
Atlantic City, N.J., June 27-July 2, 1971.
34. Engels, G. and E. Weber. Cupola Emission Control.
Translation, P. S. Cowen (ed.). Cleveland, Ohio,
Gray and Ductile Iron Founders' Society, 1967.
35. Engels, G. and E. Weber, op. cit.
36. Ussleber, K. Design of Hot Blast Cupola Plant With
a Wet Electrostatic Precipitator for Cleaning the Stack
Gas. Translated from Giesserei J52_(7) :194-197, 1965.
37. Oglesby, S. and G. Nichols, op. cit.
38. Brown, R. F. A Report on the Use of Electrostatic Pre-
cipitators in the Iron and Steel Industry. Research-
Cottrell, Inc., Bound Brook, N.J. February 10, 1970.
173
-------
39. Frisch, N. W. A Report on the Use of Electrostatic
Precipitators in the Chemical Process Industries.
Research-Cottrell, Inc., Bound Brook, N.J., February 24,
1970.
40. Cree, K. H. Cottrell Electrical Precipitation as Applied
to the Manufactured Gas Industry. Am. Gas J. 162;27,
March 1945.
41. Francombe, K. W. Electro-Detarring. Trans. Inst. Gas
Eng. £9:132-195, 1949-1950.
42. Shreve, R. Norris. Chemical Process Industries, 3rd
Edition. New York, McGraw-Hill Book Co., 1967. p. 329-
330.
43. Compilation of Air Pollution Emission Factors, 2nd
Edition. U.S. Environmental Protection Agency, Research
Triangle Park, N.C., April 1973.
44. Gillespie, G. R. Particle Size Distribution in Hygro-
scopic Aerosols. Engineering Experiment Station, Univ-
ersity of Illinois, Urbana, Illinois. May 1, 1953.
45. Air Pollution Engineering Manual, 2nd Edition. John A.
Danielson (ed.). Air Pollution Control District, County
of Los Angeles, May 1973.
46. Frisch, N. W., op. cit.
47. Atmospheric Emissions from Thermal-Process Phosphoric
Acid Manufacture, Cooperative Study Project, Manufacturing
Chemists' Association, Inc., and Public Health Service,
NAPCA Publication No. AP-48, October 1968, p. 11.
Superintendent of Documents, U. S. Government Printing
Office, Washington, D.C.
48. Brink, J. A. and C. E. Contant. Experiments on an
Industrial Venturi Scrubber. Ind. Eng. Chem. 50:1157-60,
August 1958.
49. Brink, J. A. Removal of Phosphoric Acid Mists. In:
Processes for Air Pollution Control, 2nd Edition, G.
Nonhebel (ed.). Cleveland, Ohio, CRC Press, 1972. p. 557.
50. Jones, H. R. Pollution Control in the Nonferrous Metals
Industry. Noyes Data Corp., Park Ridge, N.J., 1972. p. 12,14,
174
-------
51. MRI. Particulate Pollutant System Study. Vol. Ill -
Handbook of Emission Properties, May 1, 1971. p. 287.
52. Singmaster and Breyer. Air Pollution Control in the
Primary Aluminum Industry, Vol. I, Sec. 3. July 23, 1973,
p. 11.
53. Ibid, p. 15.
54. Ibid, p. 14.
55. Ibid, Sec. 4, p. 6.
56. Hanna, Thomas and Michael Pilat. Size Distribution of
Particulates Emitted from a Horizontal Spike Soderberg
Aluminum Reduction Cell. J. Air Pollut. Control Assoc.
Z2:536, July 1972.
57. Callaioli, G., U. Lecis, and R. Morea. Systems of Gas
Collection and Cleaning in Electrolytic Furnaces of
Montecatini Edison Aluminum Plants. Paper A70-23,
Institute for Iron and Steel Studies, 1970 Offshore
Technology Conference.
58. Singmaster and Breyer, op. cit., p. 13.
59. Ibid, p. 20-21.
60. Ibid, p. 5.
61. Op. cit.
62. Washington State Dept. of Ecology, Report No. 72-28,
Source Test, Summary of Emissions to Atmosphere.
November 21, 1972.
63. Sui, c.T. A Report on the Use of Electrostatic Precip-
itators in the Petroleum Refining Industry. Research-
Cottrell, Inc., Bound Brook, N.J., February 13, 1970.
64. Personal Communication with Martin Marietta.
175
-------
SECTION XI
APPENDIX A
CONVERTING UNITS OF MEASURE
To Convert From
Ibs
grains/cf
cfm
lbs/in.2
oF
ft2/1000 cfm
inches w.g.
gallon
ft
inches
tons
cubic inches
cubic feet
cubic feet
gallon/minute
square feet
square inches
gallon/1000 ft3
grams
To
kg
grams/m3
m /sec
kg/m2
°C (
m2/(m3/sec)
mm Hg
liter
m
m
kg
cubic centimeters
cubic meters
liters
liters/second
square meters
square centimeters
liters/m3
grains
Multiply By
0.454
2.29
0.000472
703
F - 32) x 5/9
0.197
1.868
3.785
0.3048
0.0254
908
16.39
0.028
28.32
0.0631
0.0929
6.452
0.135
15.43
176
-------
SECTION XII
APPENDIX B
LITERATURE STUDY BIBLIOGRAPHY
FUNDAMENTAL INFORMATION
Tuma, Jiri. Dust Properties Which Affect Separation. Staub.
26(11) :l-6, 1966.
Liu, Benjamin, and Yeh, Hsu-Chi. On the Theory of Charging of
Aerosol Particles in an Electric Field. Journal of Applied
Physics. 39(3):1396-1402, February 1968.
Glowiak, Bohdan, and Kabsch, Piotr. Method for Determination of
the Wettability of Dusts and Importance of the Investigation
Results for Practical (Jse. Staub. 32(1):12-15, January 1972.
Hanson, D. N. and Wilke, C. R. Electrostatic Precipitator
Analysis. Industrial and Engineering Chemistry, Process Design
and Development. 8(3):357-364, July 1969.
Masuda, S. I., Onishi, Toshio, and Saito, Hiroshi. Inlet-Gas
Humidification System for an Electrostatic Precipitator. Ind.
Eng. Chem. Process Design Develop. 5(2):135-45, April 1966.
Richards, Clyde. Distortion and Instability of Electrically
Stressed Water Drops Falling at Terminal Velocity. Dissertation
Abstracts-B. 32, June 1972.
Darby, K. The Use of Electrostatic Forces for the Separation of
Suspended Materials in Liquids and Gases. Lodge-Cottre11 Limited,
Birmingham, England.
Frauenfelder, A. Overcoming Special Problems in Electrical Pre-
cipitation. Filtration Society's Conference on What's New in
Dust Control and Air Cleaning at the Dust Control and Air Clean-
ing Exhibition, Olympia, London. September 25-27, 1973.
Schutz, Alfred. The Electrical Charging of Aerosols. Staub.
27(12):24-32, December 1967.
PATENTS
Revell, A. and Welch, W. Gas Filter Apparatus. U.S. Patent
#3,505,786. April 14, 1970.
Meston, A. F. Downdraft Wet Precipitator. Patent #1,239,844.
February 3, 1920.
Masuda, S. Apparatus for Electrostatic Precipitation of Dust.
U.W. Patent #3,444,668. May 20, 1969.
177
-------
PATENTS (Cont)
Schmidt, W. A. Process and Apparatus for Collecting Suspended
Material from Furnace Gases. Patent #1,413,877. April 25, 1922.
Cottrell, F. G. Purification of Gases. Patent #1,016,476.
February 6, 1912.
Wolcott, E. R. Method for the Removal of Suspended Material from
Gases. Patent #1,479,270. January 1, 1924.
Finney, J. A. Electrode Configuration in an Electrical Precipi-
tator. Patent #3,570,218. March 16, 1971.
Hodson, P. and Klemperer, H. Liquid Cleaned Precipitator. Patent
#2,615,530. October 28, 1952.
Lenehan, Joseph W. Collecting Electrode Structure. Patent
#2,853,150. September 23, 1958.
Gustafsson, S. and Ohnfeldt, P. Method for Cleaning the Electrodes
in Electro-filters. Patent #2,874,802. February 1959.
Hedberg, C. W. J. Collection of Suspended Particles. Patent
#2,677,439. May 4, 1954.
Bradley, L. Apparatus for Separating Tar from Gases. Patent
#1,473,800. November 13, 1923.
Vicard, P.G. Process and Apparatus for the Purification of Gases.
Patent #2,983,332. May 9, 1961.
Wiemer, J. Wet Electrostatic Precipitator. Patent #3,221,475.
December 7, 1965.
Ebert, P. Gas Cleaning Apparatus and More Particularly to an
Improved Electrical Precipitator. Patent #3,492,790. February
3, 1970.
Penney, G.W. Electrical Dust Precipitator Utilizing Liquid Sprays.
Patent #2,357,355. September 5, 1944.
Procedure for Electrical Cleaning of Gas Containing Dust with the
Aid of a Wet Electrostatic Precipitator Having Plate, Wire, or
Tube Electrodes. Patent #904,334 (France) Zschocke-Werke Company.
February 26, 1945.
Mullin, D. D. Electrostatic Precipitator. Patent #3,444,667.
May 20, 1969.
Watanabe, Y. and Nishikawa, T. Purification Device for Heated Gas.
(Japan) Patent Sho 48-11550. April 13, 1973.
178
-------
MIST COLLECTION
Adrain, Robert C. Two-Stage Electrical Precipitators. Air
Pollution Engineering Manual. Danielson, J.A. (comp. and ed.).
Cincinnati, Ohio, Public Health Service, National Center for
Air Pollution Control, PHS-PUB-999-AP-40, 1967. p. 156-166.
Donovan, J. R. and Stuber, P. J. Sulfuric Acid Production from
Ore Roaster Gases. Journal of Metals. November 1967. p. 45-50.
Semrau, Konrad T. Control of Sulfur Oxide Emissions from Primary
Copper, Lead, and Zinc Smelters - A Review. (Paper 70-97 pre-
sented at the 63rd Annual Meeting of the Air Pollution Control
Association. St. Louis, Missouri. June 14-18, 1970.)
Electrostatic Fume Collector. Staub. 34(3):102. March 1974.
Handte and Company.
Electronic Air Cleaners Whisk Away Texturing Plants' Oil Mist
Smoke. Textile World. 124(9):207. September 1974.
Tiesler, H. The Reduction of Odor Emissions Through Aerosol
Formation and Electrostatic Precipitation. Schriftenrihe Ver.
Wasser-Boden-Lufthyg (Berlin). (35):71-77. 1971.
Kubyshev, N., Krotuv, N., and Meysner, R. Production of Sulfuric
Acid in Lead Factories. Tsvetnye Metally. (Moscow) (10):22-25.
October 1972.
Dougill, G. Cleaning of Producer Gas. Transactions Institute
Chemical Engineers. 23:230-3nd. 1945.
Pexton, S., Dougill, G., and Ravald, L. Removal of Tar, Naphtha-
lene, and Amonia from Gas. Gas Journal. 254:539-544. June 9,
1948.
Beltran, M. Smoke Abatement for the Carpet Industry. Carpet and
Rug Industry. May 1973.
Beltran, M. Smoke Abatement for Textile Finishers. American
Dyestuff Reporter. August 1973. p. 26-31.
Willett, H. Cleaning, Cooling, and Drying of Gases. Manufacture
of Sulfuric Acid. J. Duecker. 1959.
Lilley, J. N. Handling of Acid Slurries at the Flash Smelting
Plant of the International Nickel Company of Canada. Canadian
Mining Mettalurgical Bulletin. 54(589), May 1961.
Niimura, M., Konada, T., and Kojima, R. Sulfur Recovery from
Green-Charged Reverberatory Off-Gas at Onahama Copper Smelter.
(TMS Paper No. A73-47, The Metallurgical Society of AIME, N.Y.
179
-------
MIST COLLECTION (Cont)
Hensinger, C.E., Wakefield, R.E. , and Glaus, K.E. Turning Pollu-
tion Gases into Profits. Eng. Mining J. 169:131-135, February
1968.
Hensinger, C.E. and Wakefield, R.E. New Roasters Spur Produc-
tion of Sulfuric Acid and Zinc Oxide Pellets. Chem. Engineering.
June 1968. p 70-73.
Seals, G.C. and Cadle, J.J. Copper and By-Product Sulfuric Acid
Production by Palabora Mining Company. Journal of Metals. July
1968. p 85-90.
Cree, K. H. Cottrell Electrical Precipitation as Applied to the
Manufactured Gas Industry. Amercian Gas Journal. 162:27-30,
March 1945.
Manufacturing Chem. Assoc. Inc. and Public Health Service, U.S.
Dept. of HEW. Atmospheric Emissions from Thermal-Process
Phosphoric Acid Manufacture, 1965.
Ross, W.H., Garothers, J. N., and Merz, A.R. Use of Cottrell
Precipitator in Recovering HsPOu Evolved in Volatilization
Method of Treating Phosphate Rock. Journal of Industrial Engineer-
ing Chemistry. 9(1):26-31, June 1917.
Allgood, R. W. Sulfuric Acid and Liquid Sulfur Dioxide Manufactured
from smelter Gases at Copper Cliff, Ontario. Canadian Mining
Metallurgical Bulletin. 45(497):153-155, March 1952.
Mills, John L., Kalian, Leo E., Lemke, Eric E., and MacKnight,
Robert J. Unique Applications of Air Pollution Control Devices.
(49th Annual Metting of APCA. Buffalo, New York. May 20-24,
1956.) p 11-1 - 11-20.
Apakhov, I. A., Inyushkin, N. V., Kravchenko, E. A., Bulycheva,
L. I., Shakhov, G. F., Goncharov, A. Ye., Chemenov, A. N., and
Usov, G. A. Scrubbing of Sulfuric Acid Mist with the Aid of
Industrial Electrofilters. Khim. Pro. (Moscow) 9:676-678,1972.
Staschik, Guenter. Separation of Plasticizer Mists in the Pro-
cessing of PVC Mixtures. Kunststoffe. 59(7):416-418, 1969.
Kitani, R. , Kubo, Y., and Otsuka, H. Anti-Acid Refractory Mist
Cottrell. Jap. Pat. SHO 47-28609, July 1972.
Roll, Kempton H. Controlling Corrosive Air Pollutants. Journal
Air Pollution Control Association. 1(6):6-10, May 1952.
180
-------
MIST COLLECTION (Cont)
Bronsard, Joseph A. and Toro, Richard F. Odor Control of Paper
Recycling. Pollut. Eng. 4(5):18-19/ August 1972.
Gotham, R.L. Electrostatic Precipitation of Sulfuric Acid Mists.
Proc. Clean Air Conf. University of New South Wales, Paper 20.
2, 1962.
Faith, W.L. Air Pollution Abatement...Survey of Current Practices
and Costs. Chemical Engineering Progress. 55(3):38-43, March 1959,
Thompson, R. and Butler, E. History of Sulfuric Acid Production
from Converter Gas at the Kennecott Copper Corporation's Utah
Smelter. Journal of Metals. 20(7):32-34, July 1968.
Plastics Industry Ponders its Recycling Problems. Chemical
Engineering. 78(12):56, June 1973.
Manufacturing Chemists Association, Inc., and Public Health
Service. Atmospheric Emissions from Sulfuric Acid Manufacturing
Processes. U.S. Dept. of Health, Education, and Welfare, 1965.
McBreen, J.P. Electric Detarrers for Gasworks. Metropolitan-
Vickers Gazette. 22:238-240, February 1948.
Quitter, Volker. Electrostatic Separation of SO3 Mists. Staub.
30(4):8010, April 1970.
Ochs, H. The Purification of Recirculated Air in Rolling Opera-
tions. Metall. 4:348-351, 1965.
Brink, J.A. Mist Removal - Removal of Phosphoric Acid Mists in
Processes Air Pollution Control, Nonhebel, G. (ed). Cleveland,
CRC Press, 1972. p 549-563.
Stastny, E.P. Electrostatic Precipitation. Chemical Engineering
Progress. 62(4):47-50, April 1966.
Sykes, W. and Broomhead, F. Problems of Electrical Precipitation
Reviewed. Gas World. 134(3494):98-104, August 1951.
Francombe, K.W. Electro-Detarring. Transactions Institute of
Gas Engineer. 99:132-195, 1949-50.
Jepoon, J. Electrostatic Precipitators for Reformed Gas.
Gas Journal. 317:224-227, February 1964.
Westh, H. C. Electrostatic Precipitators. Trans. Ingeni^ren
(Copenhagen). 58(l):36-48, 1949.
181
-------
MIST COLLECTIONS (Cont)
Goncharov, A. Y. E. and Fedotov, M. V. Intensified Process for
the Removal of Sulfuric Acid Mist from Gases by Electrostatic
Technique. Khim. Prom., No. 4. 1973. p 49-51.
Ochs, Hans-Joachim. Precipitation of Oil Mists with Electro-
static Precipitators. Energie Tech. 23(1):9012, January 1971.
Grootenhuis, B. J. Small-Scale Electrostatic Precipitator.
Chem. and Process Eng. 48(2) :101, February 1967.
Norman, G. H. C. Electrostatic Precipitation. Can. Mining
Journal. 91(10):69-71, October 1970.
Stopperka, Klaus. Electroprecipitation of Sulfuric Acid Mists
from the Waste Gas of a Sulfuric Acid Production Plant. Staub.
25(11) :70-4, November 1965.
Goldfield, J. and McAnlis, R. Low-Voltage Electrostatic Precipi-
tators to Collect Oil Mists from Roofing Felt Asphalt Saturators
and Stills. Amer. Ind. Hygiene J. 24(4):411-416, 1963.
Wet Electrostatic Precipitator Tames Oils, Gases and Fires.
Chem. Eng. July 1973. p 74-76.
Asarco Dedicates Hayden Acid Plant. Mining Congr. J. 58(3):28-29,
March 1972.
Fairs, Lowrie G. Removal of Acid Mists from Chemical Processes.
Processes Air Pollution Control (Eng.) 1972. p 520-49.
Rodolff, D. W. and Marble, E.R. Jr. Inspiration's New Look.
J Metals. 29(7):14-24, July 1972.
IRON AND STEEL INDUSTRY
Johnson, D. Air Pollution Practice in the UK Steelmaking Industry.
Iron and Steel (London). 44(6) :463-466, December 1971.
Dust Removal from Fumes Emitted by The Valmoesa Electrometallurgical
Plant (Switzerland). J. Four Elec. (3):75-81, March 1969.
Arimitsu, T. Dust Prevention at Iron Manufacturing Factory,
Proc. Assoc. Prev. Ind. Pub. Niusance Symp., 2nd. Japan,
October 3-7, 1972.
Lindau, L. and Schweitz, L. Air Pollution from Cupolas, Statens
Naturvardsverk (Stockholm), 1968.
Engels, G. Present Status of Dust Control for the Cupola in the
German Federal Republic; Results of the V.D.G. Questionnaire in
1963. Giesserei. 51:68-73, February, 1964.
182
-------
IRON AND STEEL INDUSTRY (Cont)
Restricting Dust Emission in Blast Furnace Operation. Society
of German Metallurgists (Dusseldorf). VDI No. 2099, February
1959.
O'Mara, R. F. Dust and Fume Problems in the Iron and Steel
Industry. Iron and Steel Engineer. October 1953. p 100-106.
Smith, Jack. Coast's Largest Steel Plant Keeps Air Clean. Air
Engineering. 1:21-24, November 1959.
Heinrich, R. F. Special Applications of the Electrical Gas
Cleaning Procedures. Ver. Deut. Ingr. (Duesseldorf). 98(28):
1633-1638, October 1956.
Guthmann, Von Kurt. Development of Blast Furnace Gas Cleaning
in the Last Ten Years. Stahl and Eisen. 73(5):283-291, February
1953.
Howell, G. A. Air Pollution Control in the Steel Industry. Iron
and Steel Engineer. October 1953. p 91-94.
Merkel, Christian, and Guelland, Joachim. Some Dust Collection
Problems in the Steel Industry. Neue Hiiette. 17 (5) :257-260,
May 1972.
Tanigushi, Akira and Kamoshida, Shuichi. A New Electric Dust
Collector. Jap. Pat. SHO 45-28478.
O'Mara, Richard and Flodin, Carl R. Electrostatic Precipitation
as Applied to the Cleaning of Gray Iron Cupola Gases. Air Repair.
3(2):105-108, November 1953.
Hashimoto, Kiyotaka. The Point of Planning and Its Effect on
Operation of an Electric Precipitator in Various Industry Smoke
Abatement (IV) - Treatment for Waste Gas of Iron and Steel
Industry. Journal of Pollution Control. 3(2):93-012, February
1967.
Punch, G. Gas Cleaning in the Iron and Steel Industry - Part:II:
Applications in Fume Arrestment. Iron and Steel Inst., London
(England). SR-83, 1963. p 10-23.
Renwanz, Roger A. and Specht, Schaeffer, E. Survey of Smoke
Control, Part I. Steel. 135(21), November 1954.
Ueda, Akiyoshi. Precipitation of Blast Furnace. Chem. Eng.
35(l):47-53, January 1971.
183
-------
IRON AND STEEL INDUSTRY (Cont)
Otsubo, S., Sanuki, Y., Yamamoto, T., Nakazono, I., and Haruta,
M. Operation of Newly Installed Lurgi Venturion Type Blast
Furnace Gas Precipitator. Seitetsu Kenkyu, No. 249. December
1964. p 148-155.
Guthmann, K. Cupola Gas Cleaning - Blast Furnace Gas. Giesserei.
42(19):519-524, 1955.
Guthmann, K. Emission of Dust and Gases in Iron Works and Iron
Foundries. VDI Berichte. 53:46-50, 1961.
Review of the International Fair - Dust Collectors. Staub.
28(10):33-36, October 1968.
Engels, G. German Foundry Experience in Cupola Dust Emission and
Emission Control. Transactions of the American Foundryman's
Society. 78:287-292, 1970.
Dust Collecting Devices for Cupolas. Air Conservation. 4(2):61-63,
1970.
Fraunfelder, A. Experience Gained with a New Scrubber-Precipitator
Combination. Krapp Technical Review. 22(3):125-126, 1964.
Mulrad, W. Dust Extraction from the Fume of Electric Arc Furnaces.
Iron and Coal. 183:669-675, 1961.
Jennings, R. F. Blast Furnace Gas Cleaning - An Analysis of Plant
Performance. Journal of the Iron and Steel Institute. 164:305-324,
March 1950.
SO2 Control Pilot Plant. Iron and Steel Engineer. November 1973.
p 100
Rice, Owen R. Blast Furnace Gas Conditioning. Iron and Steel
Engineer. 19(2):66-89, December 1942.
Burks, G. P. Modern Blast Furnace Gas Cleaning Practices. Journal
of Metals. 188:746-750, May 1950.
Meinshausen, G. Operation Experience with Oxygen Steelworks'
Fume Arrestment Equipment. Stahl and Eisen. 87(22):1304-1309,
November 1967.
Meister, W. Dedusting of the Brown Fumes of a Bessemer Converter
with a Wet Electrostatic Precipitator. Stahl and Eisen. 80(24):
1803-1805, November 1960.
184
-------
IRON AND STEEL INDUSTRY (Cont)
Guthmann, K. New Knowledge and Experience in the Purification
of Air in Metallurgical Plants. Radex-Rundschau. (3):139-162,
June 1966.
Meister, W. and Koglin, W. Clean Air and the Steel Industry.
Arbeitsschutz. 6:138-140, 1960.
Bothe, R. New Methods in Electrostatic Dedusting of Waste Gases
with High CO Content Coming from Steel Works Converters. Second
International Congress on the International Union of Air Pollu-
tion Prevention Association. Washington. Paper EN-34D.
December 6-11, 1970.
Werner, H. Questions of Dedusting in Foundries with Special
Attention to the Adding of Electrofilters. Klepzig Fachber.
72:392-396, October, 1964.
Hohle, Leonhard. Use of Disintegrators for Cupola Dust Removal.
Giesserei. 58:357-632, June 1971.
Speight, G. E. Air Pollution Control in the British Iron and
Steel Industry. Steel Times. 200:395-402, 407, May 1972.
Spreux, M. and Gagnaire, L. The Dust Extraction Equipment at the
Usinor-Dunkirk (France) Steelplant. Trans. Revue de Metallurgie.
60(11):1103-10, 1963.
Holland, M. and Whitman, K. Direct Fume Extraction for Large Arc
Furnaces. Fume Arrestment Spec. Report 83. William Lea & Co.
1964. p 150-159.
Frauenfelder, A. New Development in Cleaning Equipment for Fumes
Emanating from Ferrosilicon Electric Furnaces. Tidsskrift for
Kjemi Bergvesen Metallurgi. 23(5):110-114, May 1963.
Bruederle, E., Scheidel, C., and Werner, H. Electro-Precipitators
in the European Iron and Steel Industry. Blast Furnace and Steel
Plant. 38:1031-1037, October 1960.
Hipkin, A. S. Arc Furnace Fume Cleaning. Conference on the Con-
trol and Collection of Fume from Steel Foundry Melting Furnaces
by B.S.C.R.A. New York. April 1963. p 77-82.
Scheidel, K. Industrial Gas Cleaning by Electroprecipitation.
Metallgesellschaft Review of the Activities. (4):33-44, 1962.
O'Mara, R. F. Dust and Fume Problems in the Steel Industry.
Engineer's Digest. 14(12) :456-459, 470, December 1953.
Vacek, A. and Schertler, A. Waste Gas Cleaning Systems of Oxygen
and Steel Plants. Iron and Steel Institute. 71:82-89, 1957.
185
-------
IRON AND STEEL INDUSTRY (Cont)
Richardson, H. L. The Role of Electrical Precipitation in the
Steel Industry. Iron and Steel Engineer. 30(10):95-100, October
1953.
Shaw, F. M. Iron Foundry Air Pollution in the United Kingdom.
British Foundryman. 65(3):90-105, March 1972.
Organization for Economic Co-operation and Development. Air
Pollution in the Iron and Steel Industry. McGraw Hill, New York,
1963.
Poetschke, Herbert. Dust Collection and Combustion. Wasser Luft
Betrieb. 17(3), 1973.
Fume Control on the Electric Arc Furnace. British Steelmaker.
37(8):20-21, 23-24, August 1971.
Davis, C. M., and Coles, R. R. Gas and Liquid Filtration Separa-
tion. 8(4):418-424, July/August 1971.
Singhal, R. K. Fume Cleaning Systems Used in the Steel Industry
- Part I. Steel Times (London). 197 (8) .-531-538, August, 1969.
Esser, Karol. Design Requirements for Industrial Plants with
Air-Polluting Emission. Staub. 25(1):2-5, January 1965.
Engels, G. Some German Experiences with Cupola Dust Extraction
Plants Using Various Systems - Part II. Geisserei. 52(2)-.33-37,
January 1965.
Liesegang, Dietrich. Wet Separators for Cleaning Cupola Furnace
Gases. Geisserei. 58(12):1-15, June 1971.
Ussleber, K. Design of a Hot-Blast Cupola Plant with a Wet
Electrostatic Precipitator for Cleaning the Stack Gas. Giesserei.
52(7):194-197, 1965.
Misaki, Haruo. Removal of Sulfur from Flue Gases by the Double
Contact Process. Fuel and Combustion. 38(4):314-321, 1971.
Sennitzler, Herman. Further Tests with a New Electrostatic
Precipitator. Staub. 25(3):43-45, March 1965.
Elliott, A. C. and Lafreniere, A. J. The Design and Operation of
a Wet Electrostatic Precipitator to Control Billet Scarfing
Emission. (Presented at 64th Annual Meeting of the Air Pollution
Control Association. Atlantic City. June 27-July 2, 1971.)
Berg, Bengt R. Development of a New Horizontal-Flow Plate-Type
Precipitator for Blast Furnace Gas Cleaning. Iron and Steel
Engineer. October 1959. p 93-100.
186
-------
IRON AND STEEL INDUSTRY (Cont)
Weise, William H. Blast Furnace Flue Dust Treatment Facilities.
Sewage Ind. Wastes. 28(11):1398-1402, November 1956.
Ferrari, Renzo. Experiences in Developing an Effective Pollution
Control System for Submerged Arc Ferroalloy Furnace Operation.
Journal of Metals. April 1968. p 95-104.
Yocum, J. E. and Chapman, S. The Collection of Silica Fume with
an Electrostatic Precipitator. Journal of Air Pollution Control
Association. 8(l):45-52, May 1958.
Kolbe, F. Dry Lurgi-Type Electrical Dust Precipitation of the
LD Process. Trans. Berg-und Huttenmannische Monatshefte.
104(2):26-31, 1959.
Rasworschegg, H. Waste Heat Boiler Systems and Gas Cleaning
Plants Downstream of LD Vessels and Converters. Trans. Berg-
und Huttenmannische Monatschefte. 104(2):31-40, 1959.
Granville, R. A. The Capital Costs of Some Waste-Gas Cleaning
Plants for Use in Iron and Steel Works. Iron and Steel Inst.
Special Report No. 61. 1958. p 23-30.
Speight, G. F. Best Practicable Means in the Iron and Steel
Industry. Chem. Engr. (London). 9(217):132-139, March 1973.
COPPER INDUSTRY
Restriction of Emission Copper-Ore Mills. Verein Deutscher
Ingenieur 2101, September 1966.
Smiezak, A., Dalewski, A., and Peszat, S. Investigation of a
Dust Control System for Gases from Shaft Furnaces for the'Pro-
duction of Copper Matte. Proceedings of the Metallurgical
Institutes. (1-2):21-30, 1972.
ALUMINUM INDUSTRY
Konopka, A. Particulate Control Technology in Primary Nonferrous
Smelting. (Presented at American Institute of Chemical Engineers
and Institute Mexicano de Ingenieros Quimicos Joint Meeting, 3rd.
Denver. September 1970.)
McCabe, Louis C. Atmospheric Pollution. Industrial and Engineer-
ing Chemistry. 44(5):121-122A, May 1952.
Schiele, G. Electrical Cleaning of Waste Gases from Electrode
Furnaces. Aluminum. 43(3):171-174, March 1967.
187
-------
ALUMINUM INDUSTRY (Cont)
Bakke, Even. The Application of Wet Electrostatic Precipitators
for Control of Fine Particulate Matter. (Presented at EPA's Joint
U.S.-U.S.S.R. Symposium on Control of Fine Particulate Emissions
from Industrial Sources. San Francisco. January 15-18, 1974.)
Rossano, August T. Jr. and Pilat, Michael J. Recent Developments
in the Control of Air Pollution from Primary Aluminum Smelters
in the United States. (Paper EN-16F. Clean Air Congress, 2nd.
Washington. December 6-11, 1970.)
Report by Singmaster and Breyer. Emission Control Techniques:
Removal Equipment - Electrostatic Precipitators. Air Pollution
Control in the Primary Aluminum Industry. 1, Sec. 5. July 1973.
p 29-33.
PULP AND PAPER INDUSTRY
Electrostatic Precipitation: New Designs Spell Wider Use. Chem.
Eng. 77:32-33, July 1970.
GENERAL
Electrostatic Precipitator,. Fiber-Dyne Manufacturer's Litera-
ture .
Charged Droplet Scrubber. Manufacturer's Literature (Civiltech).
Ochi, Noritaka. Various Dust Collectors and Their Useful Appli-
cations. Japan Society of Mechanical Engineers, Tokyo.
(Presented at Seminar on Recent Technology on Public Pollution
Control, 370th. Fukuoka. December 11-12, 1972.)
Wase, Kjell. Luftfororeningar fran Kupolugnar Och Tekniska
Mojligheter Att Begransa Dessa. Information Fran Statens
Naturvardsverk, Volume I, 1968 (National Swedish Nature Conservancy
Office.)
Domanski, Ireneusz. Design and Operation of Wet Electrostatic
Dust Precipitators. Cospodarka Paliwami J. Energia. (1):1-16,
25-27.
Okamoto, Yasuo. Recent Dust Collectors and the Electrostatic
Precipitator. (Presented at Seminar on Plastics Used in Public
Nuisance Prevention Devices and Their Processing Techniques.
Osaka. February 19-20, 1973.)
Adachi, N., Kimura, M., and Hashimoto, S. Electric Filtration
of SOa. Taiki Osen Kenkyu (J. Japan Soc. Air Pollution).
2(1):98-100, 1967.
188
-------
GENERAL (Cont)
Adachi, T. Study on the Removal of SOa Gases by a Wet-Type
Electric Precipitator. Taiki Osen Kenkyu (J. Japan Soc. Air
Pollution). 4(2):188-193, November 1970.
Inyushkin, N. and Averbukh, Ya. Influence of Gas Flow Pressure
Conditions on Dust Collection in an Electrostatic Field. Soviet
J. Non-Ferrous Metals. 35:35-38, 1962.
Restricting Dust Emission in Blast-Furnace Operation. VDI
Kommission Reinhaltung Der Luft. (Duesseldorf). February 1959.
Robinson, M. Electrostatic Precipitation. Air Pollution Control.
Werner Strauss, Part 1, 275. John Wiley & Sons, New York, 1971.
Yurev, N. Purification of Industrial Gases Containing a Large
Amount of Arsenic. Tsvetnye Metally. 45(2):26-30, 1972.
Hydro-Precipitrol, Fluid-Ionics Systems. Manufacturer's
Literature.
Funke, G. Dust and Dust Collection Problems of Cement Shaft
Kilns. Zement-Kalk-Gipa. 13:137-144, April 1960.
Johnson, G. A. and Peterson, K. F. Air Pollution Prevention at
a Modern Zinc Smelter. Air Repair. 3(3):173-178, February 1964.
Ertl, D. W. Electrostatic Gas Cleaning. South African Mech.
Engr. 16(8):159-168, March 1967.
Bauer, Hans-Dieter and Bruckmann, Erich. Efficiency of Precipi-
tators - A Critical Examination Under Special Consideration of
the Particle Size Distribution and The Concentration of the Dust.
Staub. 33(3):126-129, March 1973.
Matsuyma, T. (illegible), Aoki, K., and Koya, N. A Chemical
Engineering Study of the A107 Electric Dust Collector. (Report
No. 2. 5th Fall Meeting of the Society of Chemical Engineers,
1971.
Weber, E. Applicability of Test Stand Results in Practice for
Precipitators. Staub. 33(3):129-133, March 1973.
Batel, Wilhelm. The Electrostatic Precipitator: Development
State and Trends. Staub. 33(3):133-140, March 1973.
Underwood, G. Removal of Sub-Micron Particles from Industrial
Gases, Particularly in the Steel and Electricity Industries.
International Journal of Air and Water Pollution. 6:229-263,
1962.
189
-------
GENERAL (Cont)
Stairmand, C. J. The Design and Performance of Modern Gas-Clean-
ing Equipment. Institute of Fuel. 20, February 1956.
Pottinger, J. F. The Collection of Difficult Materials by
Electrostatic Precipitation. Australian Chem. Process. Eng.
20(2):ll-23, February 1967.
Engels, G. Cupola Furnace Dust Removal. 127:132-167. Giesserei
Verlag, 1967.
Strindehag, 0. M. Liquid Surface Electrostatic Precipitator.
Review Scientific Instruments. 38:95-99, January 1967.
Gilbert, Nathan. Removal of Particulate Matter from Gaseous
Wastes - Wet Collectors. Department of Chemical Engineering.
University of Cincinnati. 1961. p 39-40.
Soo, S. L. Electro-aerodynamic Precipitator. U. S. Patent
No. 3,500,614. March 1970.
Parkington, J.W. and Laurie-Walker, S. Attainment of High Pre-
cipitation Efficiencies on Fine and Sub-Micron Dusts and Fumes.
Colloques Internationaux du Centre National de la Recherche
Scientifique. (Paris). (102) :321-362, 1961.
Guthmann, K. Air Cleaning in Foundries, Especially Steel, III,
Dedusting of Fumes from Electric Arc Furnaces. Radex Rundschau
(7):323-347, 1958.
Beaver, C. E. Cottrell Electrical Precipitation Equipment - Some
Technical and Engineering Features, Recent Development, and
Applications in the Chemical Field. Trans. Amer. Inst. Chem. Eng.
42:251-261, 1946.
Electric Chimney Attracts Smoke and Washes it Away. The Engineer.
33, Decmember 1970.
Stairmand, C- J- Selection of Gas Cleaning Equipment: A Study
of Basic Concepts. Filtration and Separation. 1:53-63,
January-February 1970.
Heinrich, R. F. and Anderson, J. R. Electro-Precipitators in the
Chemical Industry - Their Applications, Cost and Operation. Brit.
Chem. Eng. 2:75, February 1957.
Knott, D. R. J. The Heenan Aeropur Electrofilter. Public Cleansing
(London). 60 (6):274-276, June 1970.
Electro-Aerodynamic Precipitator. Steam Heating Engineer.
40(469):18-19, December 1970.
190
-------
GENERAL (Cont)
Rathgeber, Ferdinand. Electric Filters for Dry and Wet Separa-
tion. Wasser Luft Betrieb. 14(1):456-460, November 1970.
Yurev, N. V. Purification of Industrial Gases Containing a
Large Amount of Arsenic. Tsvetnye Metally. 45(2)126-30, 1972.
Device for Precipitation and Collection of Charged Particles from
a Gas. (To Aktiebolaget Atomergi) British Patent 1,176,425,
Nuclear Science Abstracts. 24(8), 1970.
Eisner, J. H. Possible Uses of Modern Electric Gas Cleaning
Plants. Wasser, Luft, and Betrib. 14(10):394-402, October 197
(digit missing).
Wet Precipitator Introduces New Concept. Paper Trade Journal.
July 1973.
Sargent, Gordan D. Gas/Solid Separations. Chemical Engineering.
February 1971.
Soo, S. L. and Rodgers, L. E. An Electro-Aerodynamic Precipitator.
(Paper No. G8-104 61st Annual Meeting, Air Pollution Control
Association. St. Paul. June 26, 1968.)
Soo, S. L. and Rodgers, L. W. Further Studies on the Electro-
Aerodynamic Precipitator. Powder Technology. 5(1), November
1971.
Wright, F. Dynamic Dust and Fume Precipitators. Iron and Steel
Inst. Special Report 61. 1958. p 103-108.
Bothe, R. Increasing the Efficiency of Existing Dust Removal
Plants for Flue Gas. Stahl und Eisen. 89:30-35, January 1969.
Guthmann, Kurt. Operation and Maintenance of Electrical Dust
Collectors. Staub. 26(4):13-15, April 1966.
Weber, E. Dust Extraction from Cupolas with Special Reference
to the Costs Involved. Geisserei. 49(6):136-142, 1962.
Koglin, Waldemar. The Load Dependence of Electrostatic Precipi-
tators. Staub. 28(4):10-15, October 1968.
Koglin, Waldemar. Separation of Dusts and Mists in Wet Electro-
static Precipitators. Staub. 30(4):6-8, April 1970.
Eyraud, G., Joubert, J., Morel, R., Henry, C., and Roumesy, B.
Investigations of a New Dust Collector with Electrostatic Water-
Spraying. International Convention on Air Pollution. Paper V/5.
(London) 1966.
191
-------
GENERAL (Cont)
Review of the Exhibition ' Reinhalting Der Luff. Staub. 30(2)
53-54, February 1970.
192
-------
TECHNICAL REPORT DATA
(Please read Inslructions on the rcrcnc before completing)
1. RtPORT NO.
EPA-600/2-76-142
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Wet Electrostatic Precipitator System Study
5. REPORT DATE
May 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John P. Gooch and Alan H. Dean
8. PERFORMING ORGANIZATION REPORT NO
SORI-EAS-75-651
3133
9. PERFORMING OR8ANIZATION NAME AND ADDRESS
Southern Research Institute
2000 Ninth Avenue South
Birmingham, Alabama 35205
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ACX-095
11. CONTRACT/GRANT NO.
68-02-1313
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; 5/73-4/75
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES ffiRL-RTP Project Officer for this report is L.E. Sparks, Mail
Drop 61, Ext 2925.
16. ABSTRACT
The report describes a study of wet electrostatic precipitators (WESP's)
through laboratory experiments, a field test of a full-scale unit, interviews with
manufacturers and users, and a literature survey. Experiments were aimed at
determining collection efficiency as a function of particle size at varying current
densities and specific collection areas. The results agreed reasonably with those
predicted by simulating the experimental conditions with a mathematical model. The
feasibility of collecting SO2 and particulate in a WESP was examined. As expected,
calculation of ion transport rates indicates that only an insignificant amount of SO2
can be removed by selective ionization of SO2 molecules. Mass transfer calculations
indicated that irrigated electrode surfaces would not provide sufficient interfacial
area for desired SO2 removal levels. Based on conventional geometry and stainless
steel construction, a WESP/scrubber device would be more costly than a scrubber-
only system because of high WESP capital costs. The effectiveness of a WESP as a
mist eliminator following a scrubber was calculated to be a function of the particle
size distribution and the concentration of the mist to be collected. The field test and
literature survey showed that WESP's have been used effectively to control fine par-
ticle emissions in the aluminum, iron and steel, chemical and petrochemical fields.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Electrostatic
Precipitators
Dust
Sulfur Dioxide
Aluminum Industry
Iron and Steel
Industry
Chemical Industry
Petrochemistry
b.lDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Wet Electrostatic Pre-
cipitators
Collection Efficiency
Particulates
c. COSATI Field/Group
13B
11G
07B
11F
07A
08G
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
204
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
193
------- |