United States Industrial Environmental Research EPA-600 7-79-078
Environmental Protection Laboratory March 1979
Agency Research Triangle Park NC 27711
Fugitive and Fine
Particle Control Using
Electrostatically Charged
Fog
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161.
-------
EPA-600/7-79-078
March 1979
Fugitive and Fine Particle
Control Using Electrostatically
Charged Fog
by
Stuart A. Hoenig
University of Arizona
Tucson. Arizona 85721
Grant No. R805228
Program Element No. EHE624
EPA Project Officer: Dennis C. Orehmel
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
-------
DISCLAIMER STATEMENT
This report has been reviewed by Dr. Stuart A. Hoenig, and the
Environmental Protection Agency tEPA), and has been approved for
publication. Approval does not signify the views and policies of the
Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
-------
ABSTRACT
We have demonstrated that most industrial pollutants acquire an
electrostatic charge as they are dispersed into the air. If this charged,
airborne material is exposed to an oppositely charged water fog there is
enhanced contact between the particulates and the fog droplets. After
contact is made the wetted particulates agglomerate rapidly and fall out of
the atmosphere.
This technique has been tested on a wide variety of industrial
pollutants ranging from silica flour to sulfur dioxide and fly ash. In
general, there has been, significant suppression of pollution with a minimum
of water fog. The system is therefore well suited to control of moving
fugitive dust sources where the usual duct and baghouse systems are
ineffective or too costly.
We have also been able to develop electrostatic hoods and screens
that can be used to push, or direct pollutants to the proper area for
collection. Another area of interest has been the control of diesel engine
particulates and modifications to cyclone systems for collection of fine
particulates. In both cases, improvements on the order of 7Q.% have been
achieved with relatively simple apparatus.
The charged fog systems are now being tested in various industrial
applications with generally good results. All of the work to date
including industrial applications that have been released by the companies
involved will be discussed in this report.
This paper covers- the work done under EPA grant on development of
new charged fog systems for control of fugitive dust? demonstration testing
of the systems in industrial locations; investigation of the use of charged
fog for control of sulfur dioxide; design and construction of a high
temperature stack simulator for fog gun testing; development of new
electrostatic hoods and screens for pollution control and a discussion of
new methods for improvement of dry cyclones.
This report was- submitted in fulfillment of Grant No. R80S228Q20 by
Professor Stuart A» Hbenig under the sponsorship of the U. S. Environmental
Protection Agency. This report covers a period from 1 September 1977
through. 3Q September 19JS and work was completed as of 3d November 1978.
iii
-------
TABLE OF CONTENTS
Page
LIST OF ILLUSTRATIONS vi
INTRODUCTION 1
CONCLUSIONS AND RECOMMENDATIONS 1
I. LABORATORY INVESTIGATIONS 1
A. IMPROVEMENT OF DRY CYCLONES 1
B. LABORATORY STUDIES OF CHARGED FOG FOR SOz AND
DUST CONTROL AT ELEVATED TEMPERATURES. . 4
C. TESTS WITH THE ANACONDA STACK SIMULATOR 5
D. NEW APPARATUS DEVELOPMENT 6
E. DEVELOPMENT OF A HAND GRINDER WITH AN INTEGRAL
DUST CONTROL SYSTEM 3
F. THE MECHANISMS OF PARTICULATE CHARGING B
G. MEASUREMENT OF THE QUANTITY OF WATER NEEDED TO
SUPPRESS A GIVEN DUST LEVEL. .... 9
H. DEVELOPMENT OF NEW NON-FOG DUST CONTROL SYSTEMS 10
1. Control of Diesel Engine Particulates by Means
of an Electrostatic Technique 10
2. The Electrostatic Hood and Curtain 11
I. IMPROVING THE PERFORMANCE OF ELECTROSTATIC
PRECIPITATORS 14
J. CONTROL OF DUST AND SMOKE DEPOSITION ON
OPTICAL SURFACES 15
K. DEVELOPMENT OF HIGH VOLTAGE INSULATORS FOR USE
IN INDUSTRIAL ENVIRONMENTS 16
-------
TABLE OF CONTENTS (Continued)
Page
L. APPLICATIONS OF CHARGED FOG TO THE SUPPRESSION
OF VARIOUS INDUSTRIAL PARTICULATES 16
II. INDUSTRIAL TESTS 18
III. DEVELOPMENT OF A HANDBOOK FOR USERS OF THE
CHARGED FOG DUST CONTROL SYSTEM 19
IV. DEVELOPMENT OF SPECIALIZED DUST TESTING EQUIPMENT 20
REFERENCES 21
-------
LIST OF ILLUSTRATIONS
1. Oust cyclone modified for improved fine particulate
collection 22
2. Fog generator for dust cyclone (scale:full) 23
3. Improvement of control of fine particle emissions for
dry cyclone operating on cotton trash. Air flow
12.2 mVmin; water flow 20 ml/min; dust flow 190 g/min 24
4. Improvement of cyclone efficiency with cone and charged
fog, operating on cotton trash 25
*W. Strauss, Industrial Gas Cleaning. Pergamon Press, 1975.
5. System for high temperature tests of the effect of charged
fog on sulfur dioxide and selected dust materials 26
6. Control of copper company fly ash by means of charged water
fog. Water flow 3 ml/min; air flow 0.58 m3/hr.; fog charg-
ing voltage +6000 volts. Each column represents the average
of at least eight runs 27
7. Control of SOz with positively charged water fog. Horizon-
tal shaded bars show levels of fly ash. SO2 flow 12 ml/min;
water flow 3 ml/min; air flow 0.7 m3/hr 28
8. Control of power plant fly ash with positively charged fog.
Water flow 3 ml/min; air flow 0.564 m3/hr 29
9. Control of power plant fly ash with positively charged water
fog and 1% MR compound (Johnson March Company). Water flow
1.5 ml/min. 30
10. Schematic drawing of stack simulator apparatus funded by
Anaconda Company 31
11. Control of smelter fly ash with charged water fog, tested
in Anaconda stack simulator. Water flow 100 ml/min; air
velocity 106.4 m/min; temperature 23°C; contact fr-i™*
2 seconds , 32
vi
-------
LIST OF ILLUSTRATIONS (Continued)
Page
12. Electrostatic fogger manufactured by Ritten Corporation,
Ltd 33
13. Schematic drawing of Transvector and Swirl Air fog nozzle
system for projection of charged fog 34
14. Charged fog projection by Transvector system 35
15. Experimental dust generation system 36
16. Effect of charged fog in reducing dust pollution shown in
Figure 15 37
17. Results of outdoor test of dust reduction by charged fog .... 38
18. Schematic drawing of special fog generator for dusty or
high-temperature environments (scale:full) 39
19. Modification of typical hand cup grinder to provide for
dust control by water addition 40
20. Reduction of dust from an air driven grinder by water fog.
ARO Corporation, Model 7025 KH5C at 5000 RPM grinding
cast iron 41
21. Effect of heating to 250°C on the charging behavior of
ultra-pure quartz 42
22. Control of iron foundry dust from cyclone area with
negatively charged water fog at various flow rates.
Dust flow to tunnel 4.6 g/min; air flow 5.1 m3/hr 43
23. Schematic drawing of a diesel particulate control system
with swirler to induce flow rotation 44
24. Reduction in particulate level in diesel smoke electro-
static precipitator. Above: electrostatic field off;
below: electrostatic field on 45
VI1
-------
LIST OF ILLUSTRATIONS Continued)
Page
25. Control of simulated dlesel smoke by means of a swirling
flow electrostatic precipitator. Air velocity 1100 ft/
min 0335.5 m/min); corona voltage -20,000; test timpi
1 hour 46
26. Schematic drawing of electrostatic hood system 47
27. Model of experimental electrostatic hcod in operation.
Left: With electrostatic field "off", the smoke rises
through tha cones. Right: With field "on", the smoke
is drawn into the space between the cones and carried
out the top of the system 43
28. "Pushing" uncharged aluminum chloride smoke with electro-
static curtain. Above: Curtain "off". Below: Curtain
"on"; smoke is pushed backwards 49
29. Schematic drawing of electrostatic curtain system with
grounded screens on both sides 50
30. Schematic drawing of electrostatic curtain system on copper
smelter converter 51
31. Smoke from burning sodium is drawn into electrostatic curtain
and deposited on brushes and grounded screen 52
32. System for control of simulated dust boil-up by charged fog
gun. ................ . 53
33. Dust boil-up created when additional dust drops down the
Pipe 54
34. Application of charged fog to dust boil-up shown in
Figure 33 55
35. Schematic drawing of electric wind system for cleaning
insulating shield; the "wind" generated by the phonograph
needles keeps the insulator from building up deposits 56
Vlll
-------
LIST OF ILLUSTRATIONS (Continued)
Page
36. Schematic drawing of electric wind system for cleaning
insulating shield; the "wind" generated by the phono-
graph needles keeps the insulator from building up
deposits 57
37. Schematic drawing of high voltage insulator for wet
environments (scale:full) 58
38. Control of grain dust with water fog. Water flow
10 ml/min; air flow 4.81 m3/hr 59
39. Control of cotton dust (from towel manufacturing area)
with water fog. Water flow 15 ml/min; air flow 150
SCFH 4.25 m /hr. Data taken with GCA Corporation
RDM-101 Beta Ray dust monitor 60
40. Control of trona dust with charged steam. Air flow 150
SCFH (4.25 m3/min) ; steam flow 15 ml/min 61
41. Control of sandblasting grit with charged fog. Water
flow 15 ml/min; air flow 210 SCFH (5.94 m3/hr.) 62
42. Control of dry red lead battery dust with charged fog.
Water flow 15 ml/min; air flow 150 SCFH (4.25 m3/hr. ..... 63
43. Control of aluminum oxide dust with charged fog. Water
flow 15 ml/min; air flow 150 SCFH (4.25 m3/hr.) 64
44. Control of respirable dust (turkey barn floor sweepings)
with charged fog. Water flow 15 ml/min; air flow 150 SCFH
(4.25 m3/hr.) 65
45. Control of cotton brack dust with charged fog. Water flow
10 ml/min; air flow 5.1 m3/hr. Reeve Angel 0.1 micron
glass filter 66
46. Control of carbon black dust with charged fog. Water flow
12 ml/min; air flow 6.2 m3/hr. Reeve Angel 0.1 micron
glass filter; sampling time 2 minutes 67
-------
LIST OF ILLUSTRATIONS (Continued)
Page
47. Control of bauxite dust with charged fog. Water flow
20 ml/min; air flow 5.1 m3/hr 68
48. Control of gypsum dust with charge fog. Water flow
20 ml/min; air flow 5.1 m3/hr 69
49. Control of calcium propionate dust with charged fog.
Water flow 20 ml/min; air flow 4.24 m3/hr 70
50. Schematic drawing of experimental test system for control
of welding smoke arid fumes 71
51. Reduction of metal welding fume with charged fog. Air
flow 2.33 m3/hr. Miller Electrical Manufacturing Company/
Appleton, Wisconsin, Model 35-S continuous wire welder .... 72
52. Reduction of metal welding fume with charged fog. Air
flow 2.83 m3/hr. Miller Electrical Manufacturing Company,
Appleton, Wisconsin, Model 35-S welder 73
53. Schematic drawing of dust control system for industrial
sander ............................ 74
54. Reduction of dust from sander with charged fog. Left:
Dust without charged fog. Right: With charged fog on .... 75
55. Sandblasting without charged fog ............... 76
56. Sandblasting with charged fog for dust control ........ 77
57. Effect of charged fog on dust at copper smelter furnace
clean-out operation. Above: Without fog. Below: Use of
charged fog has eliminated blowing dust ............ 78
58. Effect of charged fog on cotton dust in an operating cotton
gin. Water flow 50 ml/min; air flow 1.13 m /hr. , sampling
4 minutes ........................ 79
59. Effect of charged fog on cotton dust with fog gun at two
different locations in cotton gin. Water flow 100 ml/min;
air flow 4.5 m3/hr ...................... SO
-------
LIST OF ILLUSTRATIONS (.Continued)
Page
60. Fog gun system for dust control in duct 81
61. Fog gun for dust control in a drop box 82
62. Fog gun system for dust control on grinder 83
63. Fog gun system for dust control during grinding 84
64. Fog gun system for control of dust boil-up during rap-
ping in an electrostatic precipitator 85
65. Differences in pile-up of dry and wet dust in GCA
RDM-101 dust monitor 86
66. Calibration curve for converting optical data to actual
dust loading, assorted dust materials 87
-------
INTRODUCTION
In an earlier report [1] we discussed the first two-and-one-half
years of the Fugitive Dust Control program and reviewed the technology that
had been developed. At that time there were still many questions about the
use of charged fog in actual industrial applications and it was by no means
clear just what the effects of temperature and contact time would be.
Since that report was written the program has developed in a number of
directions that will be discussed below. In most cases the applications we
had looked for have developed and a number of new and interesting areas of
research have been opened.
CONCLUSIONS AND RECOMMENDATIONS
We suggest that the charged fog technique has been demonstrated as
an effective control technique for a number of fugitive dust problems. In
some cases it is possible to suppress sulfur dioxide and diesel exhaust
particulates by the same technique indicating that the charged fog system
will enjoy wide use in industry and transportation. We have also begun to
investigate specialized techniques for pushing or drawing industrial fumes
from the area where they are generated to locations where control
techniques can be applied. Again we feel that these systems will see wide
industrial use.
We recommend that the present program be continued and that some
thought be given to expansion so that more work can be done with industrial
organizations to demonstrate the control systems in-the-plant.
I. LABORATORY INVESTIGATIONS
A. IMPROVEMENT OF DRY CYCLONES
Cyclones are widely used in industry for control of dust and
fibrous particulates. They are simple, have no moving parts and require
relatively little energy to operate. The problem with these systems has
been their inability to "catch" the respirable particles in an *ir stream
without significant added complexity and pressure drop. An examination of
the literature [2] suggested that the problem was twofold. First, much of
the dust emitted from the "clean air" outlet consisted of "fines" that had
-------
been deposited in the dust box and then "picked up" again by the vortex.
The second problem was the escape of fine dust particles from the boundary
layer at the inside of the cyclone. Here the suggestion was that the
centrifugal forces on these small particles were not large enough to
overcome the aerodynamic forces causing boundary layer breakaway [2],
We felt that the dust box "pick up" problem might well be solved by
some sort of cone or cone and plate system that would allow dust to fall
into the box but would prevent the vortex from picking the fines back out
again. In Figure 1 we show a typical cyclone, of the type used for our
experiments, with a cone in place just above the dust box discharge.
Results with this system have been most satisfactory and will be shown in
Figures 3 and 4.
To solve the centrifugal problem we felt that the most appropriate
system would be a simple charged fog generator to induce agglomeration with
the hope that the agglomerates would be heavier and therefore more subject
to centrifugal effects. This may be an oversimplification of a complex
problem in that one author [3] has demonstrated that electrostatic effects
are important even in "normal" dry cyclone operations where the dust is
charged by triboelectric effects. It may well be that this phenomenon will
have to be investigated in some detail as part of our future program.
For the first series of experiments we purchased a twelve inch
•nominal diameter high efficiency cyclone. The charged fog generator was
located on the cyclone centerline as shown in Figure 1. The fogger was
brought in through the clean air outlet with the idea that if the system
were used in industry this would be the simplest type of retrofit system.
(We have found that industrial managers are quite willing to try new
devices as long as there is no requirement for modification of equipment
that is currently operating well.)
For experimental purposes the cyclone was driven at a flow rate of
410 SCFM (11.6 m3/min) by a centrifugal blower. The cotton trash was
poured into . the air stream from paper cups that were dumped into the air
intake of the blower every thirty seconds. This system is somewhat
inelegant but we were unable to find any sort of feeding system that would
work with cotton trash and, in fact, the paper cup system was suggested by
Mr. Marvis N. Gillium of the U. S. Department of Agriculture Cotton
Research Laboratory in Mesilla, New Mexico.
The fog generator for this system was designed to be simple and not
subject to clogging by fibrous particulates. The system used is shown in
Figure 2. The corona is generated between the grounded nozzle and the high
voltage wire; typical operating voltages have been some 17 kV. To date
this system has been most satisfactory. We have been able to operate the
fogger at various levels in the cyclone without trouble.
The fine emissions from the cyclone were measured by means of a GCA
Corporation RDM-101 Beta Ray system. Periodic tests with an Anderson-2000
-------
impaction sampler indicated that the emissions were entirely in the
respirable range (below ten micrometers aerodynamic diameter).
The results with this system are shown in Figure 3 where we show
the dust emission from the "clean air" outlet of the cyclone over some
twenty "runs" (each run represents a test with the GCA unit). As the dust
box filled up the emissions increased due to dust box pickup, as Stairmand-
has suggested [2]. The next lower curve, taken after the cone had been
installed, showed a reduction in emissions of about twenty percent
corresponding to Stairmand1s theory. Once again the emissions rose as the
dust box was filled, suggesting that some pickup was still occurring. The
lower curves show the effect of added water fog and it is apparent that the
fog performs two separate functions: first, it induces agglomeration and
thereby improves cyclone operation; second, it dampens the dust in the dust
box so that pickup is made more difficult. The effect of this "wetting"
was quite apparent when a window was installed in the dust box; without the
cone or the fog the cyclone vortex entered the box and created a storm of
dust particles.
The greatest reduction in dust (71%) occurred with (+) charged fog
and we feel that this is a most satisfactory result for such a limited
program. To appreciate how this will affect cyclone operations we have
plotted, in Figure 4, the usual cyclone efficiency curve and the University
of Arizona data. We feel that this technique has "pushed" the cyclone
system into a new area without the usual complexities and costs associated
with highly efficient cyclones. It should be noted that the quantity of
water involved was quite small (20 ml/min) and in no case were the walls of
the cyclone wetted. There was no apparent buildup of corrosion or deposits
on the walls of the cyclone after many hours of operation. We plan to
continue this work by developing some improved forms of the cone system
that will be easier to install in existing cyclones.
Another area of interest involves testing with other dusts. The
Donaldson Company of Minneapolis, Minnesota has provided a quantity of AC
fine dust (the standard material used for tests of industrial filters).
The whole question of triboelectric charging and its effect on cyclone
operation and the use of charged fog remains to be investigated. At the
moment we have data on just one material, cotton trash, whose
characteristics are completely undefined.
Another question relates to the industrial application of the
system on large 36" cyclones of the type used by the cotton industry. It
is well known that large cyclones are not as efficient as small ones and in
many cases manufacturers have been forced to use a multiplicity of small
cyclones rather than a single large unit. This results in higher costs,
complexity and frequent clogging on fibrous dusts, e.g., cotton lint. We
have received funding to install a charged fogger on a 40" cyclone handling
cotton lint at the Marana, Arizona gin of the Producers Cotton Oil Company.
We hope to demonstrate that large cyclones can operate efficiently if
certain modifications and additions are made. This industrial application
will have the added advantage of demonstrating the following system on a
large scale apparatus.
-------
B. LABORATORY STUDIES OF CHARGED FOG FOR SOz AND DUST
CONTROL AT ELEVATED TEMPERATURES
One of the questions raised about the charged fog dust control
system is the application in high temperature environments where some of
the fog might be lost to evaporation. In previous reports [1] we suggested
that most of the fog would be captured by dust and therefore be available
for agglomeration, even at high temperatures. At that time there was only
a limited amount of data to support this idea. Further data has been
obtained with the system shown in Figure 5 where temperatures up to 370°C
can be generated. The first results are shown in Figure 6 for a typical
copper smelter fly ash. It is clear that the fog was effective in
suppressing the respirable dust and that the presence of significant levels
of SOz did not affect the process.
In another series of tests, with the system of Figure 5*, we held
the initial level of dust (copper company fly ash) and SOz constant and
observed the effect of charged fog on the reduction of SOz and fly ash.
Typical results over a range of temperatures are shown in Figure 7. It is
clear that the charged fog was effective in reducing both the SOz and the
dust level. The effect of the charged fog on the dust did not vary
appreciably with temperature but maximum SOz reduction was at about 250°C.
This was observed consistently and we suggest that below 250°C most of the
SOz has not converted to SOa and is relatively insoluble in the fog. At
250°C the conversion was almost complete and the effect was large, while at
higher temperatures the SOa becomes less soluble in the fog and the effect
falls off. At present we have no apparatus for measurement of SOa versus
SOz so this is mere speculation. We hope to continue these tests and
develop apparatus that will allow continuous monitoring of both SOz and SOa
concentrations.
In future high temperature tests we plan to make use of the
Anaconda Stack Simulator to be discussed in Section C.
In another series of tests we have investigated the use of charged
fog in suppression of fly ash from a western power plant burning low sulfur
coal. The first results are shown in Figure 8 where we have plotted the
initial and final dust levels as a function of the operating temperature.
It is clear that there was a significant dust reduction, especially at 100°
and 150°C which is close to the operating temperatures of many power plant
electrostatic precipitators. This suggests that the charged fog system
might be used to pre-agglomerate the fly ash, thereby improving the
collection efficiency of the ESP. This might be particularly important
with older ESP units where the efficiency for fine (one micrometer)
particulates is low.
*It should be noted that the metal box shown in Figure 5 was quite
small and that the air velocity in the system was only a few feet per
second. This suggests that wall effects could be most important in this
system, and it was in recognition of this problem that we built the
Anaconda Stack Simulator.
-------
Tests at higher temperatures (200°C and 250°C) yielded somewhat
conflicting results and we suggest that this was due to desorption of water
from the fly ash. For future studies we will precondition the fly ash to
remove adsorbed water before running the experiments.
The initial studies were done without added SOz because of a
breakdown in SO2 system. Some very limited, high temperature SOz data is
shown in Figure 8 and we did not see the large reduction in SOz that was
observed with copper company B precipitator dust. We suggest that with the
metallic copper company dust there was significant conversion of SOz to SO3
and that the SOa reacted very quickly with the injected fog. In the case
of the nonmetallic fly ash this SOz to SO3 conversion did not occur.
SOz is not very soluble in water and we suggest that this is the
reason that little reduction in SOz was observed. We propose to repeat this
experiment with water that has been charged with Ca(OH>2 with the
expectation that the SOz will react rapidly to form CaSOj.
Another possible improvement involves the use of additives to
reduce surface tension and improve dust water contact. In Figure 9 we show
the effects of adding Johnson Marsh Company compound MR to the water; there
was a significant improvement and we propose to investigate this effect
further. In this connection it is worth noting that the addition of
detergent materials to water used for scrubbing or fogging is quite
controversial in that the lowering of surface tension will reduce the
maximum charge that the droplets can hold while at the same time a lower
surface tension might well enhance particulate wetting when the droplet
actually contacts one or more dust particles. Resolution of this question
will require a series of rather detailed experiments which we hope to
schedule as time is available.
In view- of the discussion above where we indicated that the
existing in-the-furnace test system is subject to a number of errors we
plan to do all future testing in the Anaconda system shown in Figure 10.
This will allow accurate measurements of contact times and the effects of
a heated environment.
C. TESTS WITH THE ANACONDA STACK SIMULATOR
In earlier reports [1] we discussed design and construction of the
Anaconda funded stack simulator. This unit provides for heat input,
addition of dust and gases, a controlled air flow, plus appropriate
fogging and sampling systems. To date all tests have made use of
precipitator dust provided by a local copper smelter.
Typical results at ambient temperature are shown in Figure 11.
The upper curve shows the typical dust density profile measured in some
three successive runs. We consider this to be excellent uniformity in a
practical system that can simulate industrial conditions. The lower
curves show the effect of added water fog; with (-) fog there was a
significant reduction in respirable dust. This is most encouraging
-------
especially in view of the fact that in this system the total contact time
between dust and fog was relatively short (1.5 seconds). It appears from
Figure 11 that successful dust-fog interaction did take place in times as
short as two seconds, suggesting that the system might well be used in
ducts leading to a precipitator or cyclone system. In this connection it
is worth noting that we have been contacted by a major manufacturer of ESP
systems to propose experiments that would demonstrate how the charged fog
system might be applied to improving ESP operation.
In the next contract period we expect to begin high temperature
tests with this system and to investigate the effects of added S02.
Previous laboratory studies have indicated that charged fog can be used to
induce absorption of S02 on dust and then serve as a binder for the
resultant agglomerates. As such this method offers the opportunity to
control both dust and SO2 in a single operation.
D. NEW APPARATUS DEVELOPMENT
One barrier to the wider use of charged fog is the limited size of
the units commercially available. The commercial Fogger I (Figure 12) was
first developed and sold by the Ransburg Corporation of Indianapolis,
Indiana. In November 1978 the fog gun line was purchased by the Ritten
Corporation, Ltd. of Ardmore, Pennsylvania. Ritten will be marketing the
Fogger I and II along with the other air pollution systems discussed
below.
Fogger I is inherently a rather small system with a limiting flow
rate of 1 gal/hr (3.78 L/hr) and is intended for small or highly
concentrated dust sources. In view of this limitation we appreciated the
need for larger systems and began the development of the units to be
marketed as Fogger II, III, etc.
The problem here is not simply generating more fog. The fog must
be charged and this may be difficult to dp with electrostatic induction
when the fog flow is very heavy. In a heavy fog flow only the outer layer
of fog, that is nearest the induction charging ring, will be charged. We
have found it possible to charge such flows directly by holding the nozzle
at a high voltage. This does present the problem of electrical leakage
down the water supply line, but we have found it possible to eliminate
this leakage by the use of plastic tubing and injecting air bubbles into
the water stream close to the nozzle. The bubbles provide a barrier to
the flow of electricity without interfering with the operation of the fog
nozzle.
Another type of charging system, where the nozzle is grounded and
the fog is forced to flow through a corona discharge, was shown in
Figure 2. We have found this system to be most reliable and expect to use
it on some of the new fogging systems. It has the added advantage that
there is no inherent voltage limitation as there is with inductive
charging. The electrostatic field and the corona discharge can be shaped
for most efficient charging of the fog pattern appropriate to the
-------
problem. We expect to use this system on the high temperature fogger
being designed and tested for the Anaconda Stack Simulator.
To generate large quantities of fog we have made use of the Swirl
Air nozzles sold by the Delevan Manufacturing Company of West Des Moines,
Iowa. These units are effective at high flow rates and have the added
advantage that there are no small orifices to be plugged if the water is
contaminated. (We have found that the industrial water provided for
spraying is frequently contaminated by rust and nonsoluble particulates;
clogging with conventional nozzles is a serious problem.) The Delavan
Company has provided data on the droplet size generated by their units.
Typical mean diameters of fifty micrometers can be expected with the
existing systems.
Another problem with large scale use of charged fog is the
"projection" of the fog some distance, i.e., as far as 25 feet (7.62 m).
If fog droplets are thrown into still air they will travel some three feet
(0.91 m) before being "stopped" by aerodynamic drag. The best way to
extend this range is to provide a "sheath" of moving air around the fog.
This moving air will carry the fog with it to the target. The Swirl Air
nozzles do provide some sheath air naturally but we have found it
practical to increase the effect with devices called Transvectors
manufactured by the Vortec Corporation of Cincinnati, Ohio.
These units make use of the Coanda effect to generate a large flow
of low pressure air with a small input of high pressure air. Typical
multiplication ratios, i.e., induced flow/input flow, are as large as 40
and there are no moving parts to require maintenance. We suggest that the
Swirl Air nozzles be mounted inside an appropriately sized Transvector, as
shown in Figure 13, in order to project the fog to the greatest possible
distance. A photograph of a system of this type in action is shown in
Figure 14. Typical fog water flows with this unit are 11 to IS gal/hr
(41.6 to 56.8 1/hr) and charge collection tests have indicated that we are
charging all of the fog rather than the outer shell as would be the case
with induction charging.
To demonstrate the effect of this dust control system on a "large"
dust source we set up the dust generation system shown in Figure 15.
Figure 16 shows the effect of the charged fog in reducing this pollution.
The dust in this case was gypsum which we have found to be a low cost
material for these large scale tests. Typical numerical data with a
system of this type is shown in Figure 17.
We have made some industrial tests of these large units and that
data will be discussed in Section II.
Another area of interest has been the development of special fog
guns for extremely dusty environments. There have been some cases where
dust build-up on the usual Fogger I units has resulted in short circuits
and burn up. One modification to prevent this is an air blown plastic
shield that fits around the grounded nozzle and prevents dust build-up. A
drawing of this modified Fogger I system (suggested by Mr. Werner
-------
Alchenberger) is shown in Figure 18. We have tested this unit on a Fogger
I in the Anaconda dust tunnel with excellent results. We anticipate that
fog guns used in very dusty locations will be modified accordingly.
The Anaconda test facility discussed above will allow us to test
the charged fog system at high temperatures. This will require a fog gun
that can operate at temperatures as high as 200°C, which precludes the use
of plastic insulators. The first system for this purpose is much like
that shown in Figure 2. Preliminary tests indicate that this unit will
allow us to generate charged fog at the high temperatures achieved in the
Anaconda system.
E. DEVELOPMENT OF A HAND GRINDER WITH AN INTEGRAL
DUST CONTROL SYSTEM
Industrial grinding, sanding and chipping are known to generate
copious amounts of dust. In many cases the dust contains large quantities
of silica or other hazardous materials and as such presents a problemm to
industries concerned with industrial health. There have been attempts to
attach various types of dust collectors to grinders but in most cases they
have failed because the high velocity of the grinding wheel or sanding
disc propels the dust out of the collection region. It is possible to
provide overhead collectors to prevent the dust from entering other parts
of the plant but this is of little help to the operator who handles the
grinder since he is exposed to the dust before it can get to the
collection hood.
Under these circumstances we felt it was important to develop a
dust control system that was an integral part of the grinder but did not
upset the normal weight, balance or appearance of the system. The final
technique is shown in Figure 19. Water is passed down the shaft of the
grinder and dispensed to the working area of the cup grinder by
centrifugal force. (This system is the subject of a patent application by
the Ransburg Corporation.) This added water not only suppresses dust, it
also serves as a coolant to prevent "burning" the metal by excessive
grinding pressure. There has been some observation that water cooled
grinding is faster than dry grinding but these are only auxiliary factors
to the major purpose of the study, which was control of dust.
Typical results with the dust controlled grinder are shown in
Figure 20. There was a significant reduction in respirable dust with a
minimal quantity of water, suggesting that this system may have important
industrial applications. The ARO Company of Bryan, Ohio is test marketing
this system at the present time and commercial versions should be
available in the early part of 1979.
F. THE MECHANISMS OF PARTICULATE CHARGING
In earlier reports [1] we suggested that the anomalous charging of
certain particulates might be due to the diffusion of ions to the surface
8
-------
during some sort of heating process. The candidate ion was sodium since
it is a common contaminant and is well known to have a rapid rate of
diffusion in solids. The idea was that when a material was heated sodium
would diffuse to the surface and produce a layer of positive charge that
would reduce or even reverse the normal negative charge found in
particulates.
To test this assumption a specimen of pure powdered quartz,
provided by the International Minerals and Chemical Corporation of
Libertyville, Illinois, was ground with mortar and pestle and then
vacuumed into the special Anderson 2000 sampler and Trek voltmeter system
discussed in our first report [1]. The charge was measured as a voltage
and is shown in the open bars on the histogram of Figure 21. All of the
dust was negatively charged but the larger particulates had the highest
level of charge.
We should note here that this type of graph may be deceptive.
What is plotted is effectively total charge for the particles size range
of interest. If the plates had been weighed after collection the ratio of
charge to weight of material collected would probably yield bars of
approximately equal height. The interest here was in the effect of
temperature and for these first experiments no weighing was done.
He plan to repeat this experiment with pure quartz, desert sand
and other particulates, at a series of increasing temperatures, to see if
the sodium diffusion effect can be confirmed.
G. MEASUREMENT OF THE QUANTITY OF WATER NEEDED TO
SUPPRESS A GIVEN DUST LEVEL
This is obviously important information for planning of a dust
control system but until recently our dust feeding apparatus could not
provide a constant dust level to permit comparative experiments over some
period of time. Recently, we have made use of a Vibra Screw feeder (Vibra
Screw Incorporated, Totowa, New Jersey 07511) to produce a relatively
constant level of dust in the 2' x 2' dust tunnel. This dust was exposed
to increasingly higher levels of (-) charged fog (this polarity had been
found to be most effective in earlier experiments) and the change in dust
level measured with the Anderson 2000 impaction sampler. The data is
shown in Figure 22. At a water flow rate of 40 ml/min the dust density at
the four micrometer size level was reduced by some 86% We suggest that
further increases in flow rate would be of little value. The flow rate
for optimum control was 3.7 ml/g. This value will be used for design
purposes in future studies.
i
We plan to extend these tests to other industrial particulates as
fr-ima permits. The work is tedious and there are a number of other topics
that are of more pressing interest.
-------
H. DEVELOPMENT OF NEW NON-FOG DUST CONTROL SYSTEMS
In discussions with industrial managers it has become clear that
control of fugitive dust will require more than a series of fog guns. In
many cases the fog guns cannot be used in the area where the dust and
smoke are generated. In some cases the pollutants must be collected and
moved into an area where appropriate control techniques can be used. For
some situations, i.e., diesel exhaust particulates, the control system
must be designed to fit on the vehicle itself, preferably as part of the
regular exhaust muffler unit.
All of these problems suggst that we look into techniques for
containing and moving fugitive dust or smoke so that proper control
techniques can be used. The results of this work to date will be
discussed in this section. In every case the necessary patent disclosures
have been filed with the University Research Office.
1. Control of Diesel Engine Particulates by Means of an
Electrostatic Technique
One major objection to the diesel engine is the visible plume of
carbon particulates. Another, and potentially more serious problem, is
the respirable particles that have been suspected of being mutagenic [4],
In view of the increasing use of diesels for automobiles and their
potential for underground application in mines, it seems important to
develop some method for control of exhaust particulates.
The particles are essentially carbon flakes some 0.1 to 10
micrometers in diameter; they cannot be removed by conventional filters
because of the associated flow of engine exhaust gases, while cyclones are
of little value because of the size and weight of the particles.
The book by Lawton and Weinberg [5] suggests that these particles
are electrostatically charged. In view of this charging we suggested that
electrostatic . forces be used to draw the carbon dust out of the gas flow.
A typical system is shown schematically in Figure 23, where we have added
a swirler system to induce flow rotation. The rotation was expected to
encourage large particles to move outward toward the wall. The diesel
smoke simulation system involved the partial combustion of CjHz with
limited air (personnel at the Donaldson Company in Minneapolis, Minnesota
suggested that this was the best method for generating diesel type
particulates). The only other element in the system was a corona wire on
the axis of the pipe. Typical operating conditions were: air velocity,
335.5 m/min (1100 ft/min); temperature, 23C; corona voltage -20kV. The
simulated diesel smoke was injected at the intake port of the driving
blower.
Initial results were observed visually and a photograph is shown
in Figure 24 with the electrostatic field "off" and "on". The reduction
in particulate level is very apparent. To obtain quantitative data the
experiment was repeated with an Anderson impaction sampler at the outlet
of the plastic tube. Typical results are shown in Figure 25; there was a
10
-------
significant reduction in fine particulates, and w« suggest that this nay
be a low cost method for controlling diesel engine emissions. After
continued operation one would expect emission of larger agglomerates but
these clumps of carbon would fall out quickly and be too large for
consideration as respirable particles.
We hope to improve this system fur-chsr to achieve higher
efficiency and possibly develop some way of permanently capturing the
collected particles to preclude reemission. Tests on an operating diesel
engine in the University Mechanical Engineering Department will be run as
soon as the necessary apparatus is finished.
2. The Electrostatic Hood and Curtain
One of the major problems with fugitive dust and fume systems is
that in many cases they cannot be installed at the point of generation
because of the need to bring in a crane or front loader. A copper
converter is a perfect example of this problem; the need to move cranes
and ladles in and out of the area precludes the installation of a
permanent hood at the most effective location. In seme cases hoods have
been installed above in back of the converter, in nhe hope of catching
some of the emission, but in most situations this has not been very
satisfactory and converters are a major source of smelter fugitive
emissions.
A somewhat similar problem exists in smelter matt-tap areas where
personnel movement and crane operations preclude the use of fixed hoods;
as a result, there is significant fugitive emission that is hard to
control once it has mixed with a large quantity cf air.
We have been developing two types of devices that we feel will
help alleviate these problems. The first is an electrostatic hood to
collect dust and smoke for ultimate disposal. The second is an
electrostatic curtain that might be used to' "push" dust and smoke sideways
toward a hood.
a. The Electrostatic Hood
This device is shown schematically in Figure 26. The objective
here was to allow the dust and fume from a hot area to move upward by
means of natural buoyancy, while at the same time containing and
compressing the smoke to prevent escape and allow the use of a small
overhead hood for eventual collection. The operational smelter system
would be based on chain mail mesh supported by appropriate metal rings.
The system would be light, easy to move or fold up when not needed and
relatively immune from damage by moving objects, i.e., the crane. The
outer shell would be held at ground potential and thus provide a
significant personnel safety factor.
11
-------
During operation the inner cone would be held at a high negative
potential (25 kV) to draw smoke, fume, etc., up between the inner and
outer cones. There is extensive experimental evidence that most dust and
smoke from "hot" sources are positively charged [5], but for uncharged
smoke we have designed a smoke charger.
A working model of the experimental hood is shown in Figure 27
with the voltage "off" on the left. The smoke is provided by partial
combustion of Czfo, a reaction that is known to produce positively charged
smoke [5]. With the field "off" the smoke rises through the cone in the
normal manner. With the field "on" (Figure 27, right side) the smoke is
drawn into the space between the cones and carried out the top of the
system. In actual practice there would be a small conventional hood in
this position to draw off and suppress this emission.
The smoke that does not emerge from the top of the electrostatic
hood system is deposited on the inner cone, and in practice we would
expect that the entire hood system would be washed or shaken periodically
to remove the deposited material. Our experiments have indicated that the
deposits are highly agglomerated and can be shaken off without breakup
into respirable dust particles. This should simplify the cleaning
process.
The smoke charging system was designed and tested after we tried
to use the electrostatic hood on uncharged oil vapor smoke and found that
it did not work. The unit is a simple modification of the improved
electrostatic curtain to be described below. The uncharged smoke is drawn
in at the bottom of the system and as it passes the corona points it is
charged. Some of the smoke is deposited on the grounded screen, but any
smoke that passes through the screen is drawn up to the electrostatic hood
in the normal manner. In either case the smoke is removed from the
ambient air and is no longer a pollutant. Periodically, the system may
have to be washed down or shaken to remove agglomerates, but this should
not be a severe problem.
We feel that the addition of a smoke charger will make the hood
system useful on a wide variety of pollution control problems. We hope to
test a similar system on lead fume where control by charged fog has not
yet been successful. We expect that lead fume will be intensely charged
by the corona ion current, and that the charged fume will be agglomerated
by oppositely charged fog.
b. The Electrostatic Curtain
The electrostatic curtain is a dual purpose device. On one hand
it provides a significant flow of molecular ions that can be used to
charge dust, smoke or fume. This might be especially appropriate for
dusts or fumes, e.g., lead, that are difficult to charge by ordinary means
because of back corona effects. A second advantage of this system is that
it produces an electric wind that can be used to "push" smoke, dust or
fume to the proper area for control. Electrostatic devices have a number
12
-------
of advantages over conventional fans or blowers; there are no moving parts
to corrode, a steady pushing effect can be produced over a wide area with
a relatively small expenditure of power and the systems are light and can
be flexible enough to roll up or swing out of the way. When contamination
does occur they can be simply shaken clean or even washed with a hose. If
inadvertent contact occur:-: the flexible format will minimize damage and
the design is such that no high voltage components are exposed to cause
injury.
Our designs have gone through several modifications over a period
of time. The first systems used steel phonograph needles inserted by hand
in a brass plate to demonstrate the effect. This system was quite
effective. In Figure 28 we show the application to uncharged aluminum
chloride smoke. In the upper photograph the electric curtain is "off"; in
the lower photograph the curtain is "on" and the smoke is pushed
backwards. Typical air velocities with this system are 1-2 m/sec, strong
enough to push smoke or dust rising by thermal convection.
We felt this design was inappropriate for large scale manufacture
and after some experimental studies settled on the system shown
schematically in Figure 29. Here the phonograph needles have been
replaced by wire brushes than ars low in cost and available in a wide
variety of sizes and shapes. For our application the brushes would be
mounted end to end with the insulators between every two or four brushes
as needed. (The initial batch of insulators was made to our design by
Professor M. K. Grossman of the University of Arizona Department of Art,
Ceramics Group. For large scale purchases we will go to a commercial
ceramics manufacturer.) In front of the brushes there will be a grounded
metal screen to serve as the corona counterelectrode.
Behind the brushes we propose to mount a porous plastic material
to permit airflow and another grounded screen as a safety element. With
this system every exposed surface is electrically grounded so that there
is no danger of personnel injury due to inadvertent contact. The whole
system can be swung out of the way or rolled up', like a window shade when
not needed.
One possible application for the system is shown in Figure 30 on a
copper smelter converter. During times of heavy fume emissions the
curtain would serve to "push" the smoke toward the hood for more efficient
collection.
These applications of the electrostatic curtain imply a
significant "scale up" in size over the laboratory systems tested to date
and it is appropriate to comment on what changes in performance might be
expected from this vast change in size. In general we suggest that the
"pushing velocity" of some 1 to 2 m/sec observed above would not change to
any significant degree with a larger curtain. The distance that this flow
will "push" the pollution is harder to estimate but since normal rates of
rise for heated smoke are less than 1 m/sec we would expect that the
curtain might push the smoke at least 2 or 2-1/2 meters in the horizontal
direction. Unfortunately the present laboratory space is not suitable for
13
-------
larger scale tests. We have discussed this problem with some of the local
copper companies and hope to make use of their experimental smelter
facilities in the Spring of 1979.
One interesting application of this system is control of smoke from
burning sodium. Sodium is converted to sodium oxide by burning in air and
to date it has not been practical to completely suppress the submicron
particulates by fogging or wet scrubbers. We suggested to the company
involved that the electrostatic curtain be used for this purpose and a
typical experiment where the smoke from burning sodium is drawn into the
curtain and deposited on the brushes and the grounded screen is shown in
Figure 31. For industrial application we would anticipate a series of
brushes and screens to insure that every particle of sodium oxide is
collected. When the brushes and screens are "full" the system can be
washed down with water to remove the deposits.
We feel that this system will solve the immediate problems and
improvements will allow the same type of unit to be used on a
semipermanent basis. A similar system may well be effective on lead fume
which again has been traditionally hard to collect with conventional ESP
systems.
I. IMPROVING THE PERFORMANCE OF ELECTROSTATIC PRECIPITATORS
Electrostatic precipitators CESPl are widely used in industry for
control of dust, fume and smoke. In general their operation is quite
satisfactory with three exceptions. First, charging of one micrometer
particles is relatively inefficient and the collection efficiency in this
size range may be rather low. A second problem with ESP units is boil-up
during rapping. As the dust falls into the hoppers there is air
entrainment and the fine particles are picked up by the air stream where
they can appear at the outlet. A last ESP problem, of particular concern
to the non-ferrous metal industry, is corrosion. This leads to rapid
failure of ESP wires and side plates with excessive inleakage (300%
inleakage is not unknown! that results in a loss of stack draft.
We have been discussing the idea of fogging ahead of a
precipitator with a large manufacturer of these units and hope to begin
tests- in the near future. The data of Figure 11 indicate that rapid
agglomeration can be achieved by means of charged fog, suggesting that
this technique can be used ahead of an ESP.
The boil-up problem has been simulated on campus with the system
shown in Figure 32. In Figure 33 we show the boil-up that occurs when the
dust drops down the pipe. In Figure 34 we show the effect of charged fog.
There was a significant decrease in dust, suggesting that this method
could be used with ESP hoppers especially in the last stages where any
re-entrained dust is very likely to be carried out of the system.
The ESP corrosion problem is particularly serious for industries
that have significant levels of HzO and SOz in the ESP feed gas. There
14
-------
will always be some SO3 present and this can bring the dew point of the
gas down from 350°F to as- low as 200°F. Under these conditions- any cold
spot, e.g., a hopper, can be a point of HzStK formation. The acid eats
away at side plates resulting in still more leakage, cooling, and acid
formation. In many cases the ESP units purchased by the copper industry
are "down" more than they are operating and corrosion is the major
problem.
We have begun looking into this problem with the hope of finding
an epoxy or ceramic coating that might be used in corrosion prone areas of
an ESP. The requirements are severe; SOOT and HzSO* exposure, and to
date it appears that flame sprayed ceramic coatings will be most effective
for this purpose. We have talked to two companies that handle flame
sprayed ceramics, and small scale tests of mild steel coupons coated with
appropriate materials may be done in the near future.
J. CONTROL OP DOST AND SMOKE DEPOSITION ON OPTICAL SURFACES
This is a secondary application of the electrostatic curtain
technique and as such presents some topics of general interest.
This study originated with a request from the National Aeronautics
and Space Administration to look at techniques that might be used to
reduce dust deposition on the optical surfaces of a comet survey
satellite. Calculations indicated that passage through the comet tail
would coat the optical surfaces within a few minutes.
The electrostatic curtain was one approach to a control technique
for this problem and present indications are that a system of this type
might well be used on mirrors and lenses exposed to atmospheric
contamination on earth. One such application involves the optical system
for the SHIVA nuclear fusion system being developed by Lawrence Livermore
Laboratory and discussions with, that group have begun.
Another approach, developed by Mr. Merwyn G. Utter, is shown in
Figure 35. Here the lens is surrounded by two metal rings mounted outside
the field of view. When the lower ring is driven at about -10 kV the
falling dust moves outward rather than falling on the lens. Similar
results were observed with cigarette smoke, suggesting that this system
could be used at least with small CIO cm diameter1 lenses. This system
has the advantage that there is no corona discharge to produce electrical
interference or stray light that might interfere with optical studies.
Another protection technique involves the use of a corona sprayer
to produce a layer of charge on the lens surface itself. If the lens is
charged negatively and the falling dust is also charged negatively by an
electric curtain, the dust will be repelled by the lens. We have observed
this effect in the laboratory but found it difficult to photograph. It
appears that a combined technique of this type would be most effective in
controlling dust fall onto optical surfaces.
15
-------
DEVELOPMENT OF HIGH VOLTAGE INSULATORS FOR USE
IN INDUSTRIAL ENVIRONMENTS *^iQ*S FOR USE
K.
became
by deposits anaf
high voltage environment where the
will actually "draw" dust to t£ area
^Particularly the case in a
? aCr°SS the
over inso * is «— to blow air
clean the insulators. We felt t£at thf P^10dlc shut*>wns to physically
for our application and Mr. ^t^SS^T*- M **«****
several insulator systems that would be TS? • ^ ^ *"* testing of
environment. One such desio^ is .SJ ? ^dependent of dust in the
wind generated by the phonocra^ nL^ *" PlgUre 36' Here the electric
the insulator clean evenfafoSof^ W3S qUit6 effective " keeping
produced by partial coSstion of 2a^ xSa" "T^" °f CarbOn S™°^
that the cleaning effect is nrodu^d h !K Sy m has ^ advantage
the curtain or scree" Produce<* by the same high voltage that drives
brack, carbon black, bauxite,
Figures 38 through 49.
. * , , cotton
' calcu™» propionate are shovm i
n
t
is of interest in tne
16
-------
to the University by T. M. Caid and Sons of Tucson, Arizona. This welder
was a bare wire, CQz shielded system and operated without flux. For this
reason there was no flux smoke and the fume was primarily finely divided
metal particles. To see the effects, if any, of added flux we placed the
flux CLinde Company type AWS-A5.17-691 in the welding area by hand and
sampled the emissions with glass fiber filters. The analytical data,
provided by the University Analytical Laboratory, are given in Table 1;
Blank 1 and Blank 2 were clean filters. The data indicates little
difference between the flux and no-flux results except for somewhat higher
levels of Ca, Sr, K, and Mg when flux was used. We do not consider these
differences significant in terms of pollution emitted during welding.
Table 1. Welding Fume Data With and Without Flux
(All data reported as mg/filterl
Sample 1.0.
Blank 1
Blank 2
With Flux
No Flux
Sample 1.0.
Fe
15.1
14.6
988
1820
Sr
Ma Cu
893 <.6
523 1.48
1610 26.3
1760 31.0
Li Rb K
Mn Zn
<.15 1.9
<.15 6.8
162 7.5
170 10.2
Mg • Ni Co
Ca
172
55.6
398
277
Be
Pb
2.23
2.23
2.43
3.54
Ba V
Cr
0.23
0.15
<.oa
0.27
Cd
Blank 1 4.24 0.3 0.68 60.6 52.4 <.03 <.13 <.025 <20 <.4 <.2S
Blank 2 <.5 0.08 0.78 141 9.08 <.03 <.13 <.02S <20 <.4 <.25
With Flux 3.63 0.8 1.18 152 108 <.03 <.13 <.025 <20 <.4 <.25
No Flux 5.25 0.62 0.84 115 77.4 <.03 <.13 <.025 <20 <.4 <.25
The first experimental results in terms of the reduction in metal
fume with, charged fog are shown in Figures 51 and 52.
In general there was about a 50% reduction in captured
particulates, but fcha data was subject to wide variations. An attempt was
made to run at two different water flows in the hope of demonstrating that
there would be an enhanced effect with lower water flows. In one case
(Figure 511 this effect was observed but in another series of tests
(Figure 521 just the opposite results were obtained. There were wide
variations in the data from day to day for reasons which are not clear at
17
-------
the moment. We are planning some tests with a heated source of lead or
zinc in the hope of gaining more understanding of the metal fume/water fog
interaction.
The variations in the experimental results may have been due to
changes in the welding process itself. The Miller unit was a small one
and our technician is not a trained welder. Another problem may have been
changes in the position of the welding head with respect to the stove pipe
used to draw off the smoke and fume. The welding head had to be moved
about to prevent the build up of a mound of weld material and this did
change the induction characteristics of the system. We tried adding a 10"
diameter funnel to the welding end of the system in order to improve the
"draw" of the stove pipe and provide a more uniform flow of fume but this
did not improve matters. We suggest that the problem was due primarily to
variations in the welding process itself. In spite of these difficulties
there was a general reduction in the fume from welding when charged fog
was used.
In this connection we might note that this test system does not
allow all of the fume produced by the welder to be exposed to the charged
fog. We estimate that some 20% of the fume was pulled into the pipe
(Figure 50) before the charged fog could get at it. This suggests that
under actual industrial conditions where all of the fume is exposed to
charged fog, the effectiveness of the system might rise from some 58% to
75%.
II. INDUSTRIAL TESTS
The available data from industry is limited to what we can obtain
at local mines, smelters, cement plants, etc., and most of the results in
1977 were discussed in our last report [1]. Since then tests have been
run at the Tucson plant of the Gates Learjet Corporation where the dust
source was a belt sander.
The physical layout is shown in Figure 53. The irregular nature
of the sander operation and the movement of workmen, lift trucks, etc., in
the area precluded numerical measurements of the dust level. It was
decided to look for "visual results" with charged fog. Typical before and
after photographs are shown in the attached Figure 54. It was very
apparent that the charged fog reduced visible dust generated by the sander
and Company management has made arrangements for installation of one or
more fog guns.
Other industrial results, at a local sandblasting operation, are
shown in Figures 55 and 56. Here again, the irregular nature of the
operation and the ambient winds precluded the use of conventional dust
samplers. The fact that the man running the sandblaster was protected by
a White Cap System and the need to keep the fog off sandblasted surfaces
made it impractical to use personnel samplers. We chose to photograph the
operation with and without the fog as shown in Figures 55 and 56. There
18
-------
was a significant reduction in visible dust and the company involved is
rebuilding their sandblast booth to accommodate a fog gun.
Other commercial installations of eight or more Fogger I units
have been reported by the Ritten Company with excellent results at a
Midwestern foundry and a West Coast cement plant.
In the Southwest we have been working with several copper smelters
on the application of charged fog units for control of dust and SOz. In
one case a Fogger I mounted above the silica belt has been most effective
in suppressing the dust from the operation.
In another area of the same plant we have made use of the
experimental Fogger IX to suppress dust from a furnace clean-out
operation. The dust normally blows from the unloading area into the
smelter as shown in the upper part of Figure 57. The effect of charged
fog is shown in the lower photograph of Figure 57; the offensive dust has
been entirely eliminated.
In connection with these industrial tests there is always the
question of how much of the observed dust reduction was due to the charged
fog versus the simple "blow away" effect of the air aspirated by the fog
gun itself. It is impossible to give a general rule for every case but we
have attempted to test for these effects by first operating the fog gun
with the air "on" and the water "off" to check on the blow-away effects.
Another technique involved actually testing downwind of the fog gun as was
done in connection with the test set up of Figure 17. In the sawdust
tests of Figure 54 the agglomerated dust was quite visible on the floor
under the fog gun. In the test of Figure 57 it was "impossible" to stand
down wind of the dust source with the fogger "off"; with the fogger "on"
the only noticable effect was an occasional spray of fine mist. We expect
to continue these industrial tests and to make arrangements for
quantitative testing whenever possible; however we have found that many
industrial managers are "only" interested in visible effects. If they
"see" the system work they do not want to waste *••*««» on a lot of
measurements unless they are required to "meet" an OSHA or EPA citation.
Another series of tests with the fogging units were done in a
local cotton gin where fibrous pa articulates were a problem. In this case
it was possible to obtain numerical data and in Figures 58 and 59 we show
data taken in three areas of an operating cotton gin. In every case there
was significant reduction of dust and we anticipate working more closely
with the cotton industry in the future.
III. DEVELOPMENT OF A HANDBOOK FOR USERS OF THE
CHARGED FOG DUST CONTROL SYSTEM
The application of charged fog to the control of fugitive dust is
quite new and we recognized the need for some sort of handbook that would
help engineers in applying this technique. We have been able to do some
19
-------
work in this area but new applications seem to occur so rapidly that it
has been difficult to stop and simply finish the material. Under these
circumstances it seems appropriate to include Figures 60 through 64
showing typical applications of the fog guns in industrial situations. In
general the first rule seems to be "suit the fog guns to the dust source."
If the dust source is very large or diffuse the small Fogger I will not be
effective unless many units are mounted in the same area. For large dust
sources it may be more effective to obtain a Fogger II from the Ritten
Corporation (contact Mr. F. A. Sando, President, The Ritten Corporation,
Ltd., 40 Rittenhouse Place, Ardmore, Pennsylvania 19003 (215) 896-0900) or
the author.
Dust testing is another problem. If the GCA Corporation RDM-101
unit is used there may be significant changes in the way the dust deposits
under "dry" and "moist" conditions. This is shown in Figure 65. In some
cases we have observed that a smaller quantity of dust will yield a larger
reading on the RDM-101 because of the "pile up" effect. In these cases we
suggest the use of filters for evaluation.
Another problem has been observed with personnel dust samplers.
They vary widely in the amount of dust they collect and users are urged to
take a large number of readings before drawing any conclusions.
One last problem that may arise is due to aspiration of ambient
air through the Fogger I. In some cases this may give indications that
the quantity of dust in the area has actually "increased" when the fog
guns were on. This is obviously impossible: the fog guns cannot "make"
dust. If aspiration is a problem or if aspirated dust causes short
circuits in the fogger, the author or Mr. F. A. Sando at the Ritten
Corporation should be contacted for appropriate modification systems.
IV. DEVELOPMENT OF SPECIALIZED DUST TESTING EQUIPMENT
In view of the difficulties in dust testing discussed above we
have developed a simple dust monitor that can be used to give rapid
qualitative estimates of the dust density in an area before and after
fogging. The system makes use of a hand operated pump (the Bachrach
Instrument Company, Mountain View, California 94043) and a Nuclepore 0.8
micrometer filter (Nuclepore Corporation, Pleasanton, California 94566).
Dust laden air is drawn through the filter and the dust density is
evaluated by a hand held optical reflection test system. This system is a
portable version of the tape samplers used in automatic smoke evaluation
systems.
The optical data can be converted to actual dust loading in mg by
means of a calibration curve such as that shown in Figure 66. For precise
dust measurements a calibration curve for each type of dust would be
needed. For demonstration purposes where only the relative decrease in
dust density with charged fog is required, no calibration curve is needed.
20
-------
It is worth noting that if a calibration curve is available, a
knowledge of the number of pump strokes and the air flow per stroke allows
the operator to calculate the dust density in mg/m .
Another mode of operation would involve the use of several
filters, of decreasing pore size, in series. This would allow the dust to
be separated into fractional sizes before the optical evaluation.
1. Hoenig, S. A., "Use of Electrostatically Charged Fog for
Control of Fugitive Dust Emissions," EPA-600/7-77-131, November 1977.
Available from NTIS, Springfield, Virginia 22161.
*»
2. Stairoand, C. J., "The Design and Performance of Cyclone
Separators," Trans. Institution Chem. Engrg. (GT. B), 29, 1951, pg. 356.
3. Giles, W. B., "Electrostatic Separation in Cyclones,"
unpublished paper from G. E. Research Laboratory, Schenectady, New York.
DOE Contract EX-76-C-01-2357.
4. Barth, 0. S. and S. M. Blacker, "EPA Program to Assess the
Public Health Significance of Diesel Emissions," Journal Air Poll. Cont.
Assoc., 28, No. 8_, 1978, pp. 769-771.
5. Lawton, J. and F. J. Weinberg, Electrical Aspects of
Combustion. Clarendon Press, Oxford: 1969, pp. 247.
21
-------
CLEAN AIR
OUTLET
CHARGED FOG
'GENERATOR
11.75"
(29.8 cm)
I
SPRAY
CONE
^
J/
^
3
^
4
4'-
( 122
' \— -
__) °
12 "0.
(30.5 cm)
-0"
! cm )
1
ADDED CONE
TO PREVENT
DUST PICKUP
AIR PLUS
DUST INLET
4'-Ort
( 122 cm )
DUST COLLECTION
BOX
Figure 1.
Dust cyclone modified for improved fine particulate
collection.
22
-------
AIR PLUS
WATER LINES
SPRAYING
SYSTEMS CO.
1/8 J8C NOZZLE
3/8" THREADED
'ROD SUPPORT
SHARP POINT
HIGH VOLTAGE
JOINT
PLASTIC SPRAY
SHIELD
1.6mm BRASS ROD
Figure 2. Fog generator for dust cyclone (scale: full)
23
-------
K)
.U
14
12
10
8
6
4
2
RESPIRABLE
DUST DENSITY
mg
INITIAL DUST LEVEL
WITH CONE IN PLACE
UNCHARGED FOG
CONE PLUS (-) FOG
CONE PLUS (+) FOG
10.52
71 % REDUCTION
NUMBER OF RUNS
-I 1-
-I 1-
-I 1-
10
15
20
Figure 3. Improvement of control of fine particle emissions for dry cyclone operating on cotton
trash. Air flow 12.2 m3/min» water flow 20 ml/min; dust flow 190 g/min.
-------
99.9
99.5
99.0
97
95
90
80
50
10
.CYCLONE
EFFICIENCY
BEST U. OF A.
RESULTS
TYPICAL HIGH
EFFICIENCY
CYCLONE DATA*
I 3 5 10
PARTICLE DIAMETER ( MICROMETERS)
Figure 4. Improvement of cyclone efficiency with cone and charged fog,
operating on cotton trash.
*W. Strauss, Industrial Gas Cleaning, Pergamon Press, 1975.
25
-------
ro
A
A A A,
A
A A
OVEN
S02 SAMPLER
DUST SAMPLER
CHARGED FOG
INLET
STEEL TEST TANK
I ft. 3 (0.0284 m3 )
THERMOCOUPLE
DUST, AIR , SO2
INLET
DUST, S02 REMOVAL
SYSTEM
AIR
ARRASTRA MILL
AIR LINE
Figure 5. System for high temperature tests of the effect of charged fog on sulfur dioxide and
selected dust materials.
-------
DUST DENSITY
16
14
to
400 PPM
1200 PPM
12
10
8
6
4
\ f
800 PPM \
S08 S^ ^
/
^
"
0 ' ' '
-
^->
r
/
^
^
*
',
^•*
JU
n
/
':
f
J
*J
i
0 100°
S02 ^
T\
FJ
/
'
/
/
/
/
/
i
-
/ i
4
71
i
t
t
t
/
/
t
/
A
a
^— '
A~ »
r
J
(
*f
i
',
i
\
•^— i
INITIAL DUST
/"""DENSITY
r
EFFECT OF
n^CHARGED FOG
>
t
A
' TP O ^^
. 1 ^^
200" 300* 4OO*
Figure 6. Control of copper company fly ash by means of charged water fog. Water flow 3
air flow 0.58 mj/hr.j fog charging voltage +6000 volts. Each column represents the
average of .at J^ast eight, runs,
-------
12
IO
to
co
O
X IO* PPM
S02 LEVEL
% SO2 REDUCTION
f
too
60%
50
75%
200
S02 LEVEL BEFORE FOGGING
S02 LEVEL AFTER
FOGGING
45%
3OO
350
COPPER COMPANY
FLY ASM TYPICAL
LEVEL , BEFORE
FOGGING
FLY ASH AFTER
FOGGING
TEMPERATURE "C
00
Figure 7. Control of SO2 with positively charged water fog. Horizontal shaded^bars show levels
of fly/ash. SO2 flow 12 ml/minj water flow 3 ml/min? air flow 0.7 m /hr.
-------
ro
10
16
H
12
10
a
FLY ASH
OEf
*SITY mg/m^ p INITIAL DUST LEVEL
Q m FOG
% REDUCTION
\
T \
\
/
I /
50V.
1
I
52%
[ BAR SHOWS SCATTER OVER
14 READINGS
T T
1
24%
.
T
\
S
\
y
\
\
1
35%
IOOO
PMM
1 800
600
400
v y%
V
V
^ 200
V
\
-N — o
SO2
SCALE
25%
PI
r
s
V
\
s
\
s
s
\
\
s
V
\
20%
100 "C
150 'C
200 °C
2OO
20O
20O
Figure 8. Control of power plant fly ash with positively charged fog.
Water flow 3 ml/mini air flow 0.564 mj/hr.
25O»C
-------
u>
O
6
5
4
3
2
1
O
mg/m 3
DATA SPREAD 35
-
-
_
-
-
—
r
12 RUNS
4
S
S
\
\
\
\
S
S
\
15%
55%
>
s
s
s
V
s
34%
\
S
\
\
S
\
S
A
1
i
-
30
25
20
5
\
v
s
y
V
V
i
8%
15
10
5
mg
-
- 0.5>u
_
_
-
—
77%
/*•
84°/
El
200 °C
GCA DATA
250° C
200° C
GLASS FIBER
0.5/A
67%
\
v
V
v
\
v
250°C
FILTERS
Fiyure 9. Control of power plant fly ash with positively charged water fog and
1% MR compound (Johnson March Co.) Water flow 1.5 ml/min .
-------
u>
FUME +
DUST
CONTROL
16
X
COOLING
WATER SPRAY
( IF NEEDED )
SAMPLE
PORT
I2"D. X I/8"
WELDED TUBE
(STEEL )
CHARGED
FOG
200,000
BTU / Hr
GAS BURNER
0-250 CFM
COOLING AIR
PERIPHERAL FLOW
Figure 10. Schematic drawing of stack simulator apparatus funded by Anaconda Company.
-------
12
10
8
0
DUST
DENSITY
mg
INITIAL
DUST
LEVEL
3 RUNS
O 12345678
PARTICLE DIAMETER (MICROMETERS)
Figure 11. Control of smelter fly ash with charged water fog, tested in Anaconda stack x.
Water flow 100 ml/minj air velocity 106.4 ra/min; temperature 23°C,- contact time 2 seconds,
-------
Fooger I does not m«diemicaily
captun and moor* the dust but
simply puti it into a Eonn that is easier
to dispose of. Whereas other methods
of pollution control work bMt on larger
parades atdust. theFoggsrlis
•specially •tt*cttr« with smaUsr dust
parade* — those mast injurious to
Ewoititi Titd most *J1ffV^ilt to •Uoinotv
with conventional emttpoUuttoa units.
Thus, the Fogg«r I Is roitabl* tor u*»
Qs ^^ sois QMC0S of controilino dust
and gas or as a supptansat to axiiting
pollutiOD control dmricM. to 'mop up'
dust portidw misssd by wasting
• Wa
i Kitten Qscaostatic Foggw I is
msatil*. •oonomicnl and «fia»nt. In
addition, it is •asy to instoil. dos* not
rwiuin hoods or ducts and USM Twy
lira* »n«rgy and water. Som»
instidlanaos raquin as littte as 1/2
gallon per hour.
Voter requirement*: up to 1/2 CPM
(2 y mini and SO PSI (4 leg/cm*)*
• Air requirements: up to 13 SCFM
(.3SnvVm> and 110 PS (8 kg/CB1)'
. Power 11SV. 80 Ht 30W (also
115/230V. 50 Hi)
* Shippiiicr wvio iitz czpprox. 45 Ib*.
(20kg)
•Depending on application
Rtttstt Ssclioetanc
Fogger I can help you *O!T»
your air pollution problems.
call or write:
POLLUTION CONTROL UNIT
FOR RESFIRABLE DUST AMD GAS.
«Rltteahoae« Place
Animate. PA 19003
(JtS ) 890-0900
Figure 12. Electrostatic fogger manufactured by Ritten Corporation,
Ltd.
33
-------
HIGH PRESSURE
DRIVING AIR
SHEATH AIR
SWIRL AIR
NOZZLE
ZZZZ2ZZ7ZZZZZZZZZHK
Figure 13. Schematic drawing of Transvector and Swirl Air fog nozzle system for
projection of charged fog.
-------
by Transvector system
-------
. 4.
.' ••> »
m •
•;'*?»•
**
-I
•-*
- ;
•*«*.
"
Figure 15. Experimental dust generation system.
f?»*fe*
.._.
•
-. • - -
-------
Figure 16. Effect of charged fog in reducing dust pollution shown in Figure 15.
f a
/ / f I
-------
FOG GUN
I WIND
8 km / hr
o
o
STATION
DUST LOADING
AVERAGE OF
4 RUNS
DUST LOADING
WITH (+) FOG
3 GAL./mln
o
SAMPLING STATION
E W S
2.23 mg/m3 2.0 mg/m3 3.48
0.2
NET REDUCTION 91 %
0.0
100 %
68.4 %
Figure 17. Results of outdoor test of dust reduction by charged fog.
38
-------
FOG
NOZZLE
10
HARGEO
FOG
METAL DUST
CAP
»
INDUCTION
RING
PORCELAIN
INSULATOR
AIR BLOW
TO KEEP
DUST OFF
METAL PLATE
t
AIR
1
AIR
HIGH VOLTAGE WIRE
Figure 10. Schematic drawing of special fog generator for dusty or high-temperature environments
(scale: full).
-------
WATER PATH
GUARD
EXISTING
BODY OF
GRINDER
INLET
GRINDING ROTATING SEAL
WHEEL
Figure 19. Modification of typical hand cup grinder to provide for dust
control by water addition.
40
-------
0
DUST DENSITY
FOG WATER FLOW
6O ml t mln
TO SAMPLER
- I.
•— 8O mm
PARTICLE DIAMETER I MICROMETERS)
I
6 9
Figure 20. Reduction of dust from an air driven grinder by water fog. ARO Corporation
Model 7025 KII5C at 5000 RPM grinding cast iron.
-------
60
50
40
30
20
10
-I VOLTAGE
VIA TREK UNIT
BAR SHOWS SCATTER
'OVER s RUNS
BEFORE HEATING
I. AV-3. SIZE II MICRONS
2. AVG. SIZE 5.8 MICRONS
3.AVG. SIZE 2.7 MICRONS
4.AVG. SIZE 0.875 MICRONS
AFTER HEATING
77?!
2.
3.
Figure 21. Effect of heating to 2SO°C on the charging behavior of ultra-
pure quartz.
42
-------
26
£>.
U>
20
16
12
a
4
o
DUST
DENSITY
mo / ro.**
INITIAL
DRY
DUST
LEVEL
0
(2.74m)
- 9' —
CHARGED FOG
'>•
,x»-
=E
t
SAMPLER
! 2 3
PARTICLE
45 6 7 8
DIAMETER (MICROMETERS I
Figure 22. Control of iron foundry dust from cyclone area with negatively charged water fog at
various flow rates. Dust flow to tunnel 4.6 g/minj air flow 5.1 mj/hr.
-------
73.7 cm
CLEAN
AIR
OUT
GROUNDED METAL SCREEN
PLASTIC PIPE
CORONA
'WIRE
AIR PLUS
CARBON
PARTICULATES
25 kV
SWIRL INDUCER
INSULATOR
Figure 23. Schematic drawing of a diesel particulate control system with swirler to induce flow rotation,
-------
-------
en
O
OUST DENSITY
mg/
DUST DENSITY WITH
FIELD "OFF"
DUST DENSITY
WITH FIELD "ON
PARTICLE DIAMETER (MICROMETERS)
O I 234567Q9
Figure 25. Control of simulated diesel smoke by means of a swirling flow electrostatic precipitator,
Air velocity 1100 ft/min (335.5 m/rain)» corona voltage -20,000; test time 1 hour.
-------
CORONA
POINTS
- — 16.5 0— -
~ •
-11.40-
LI t n 1 1 M nil '
rt
0-25 KV
+
CHAIN
MAIL
MESH
15 KV
j^
\CHARG1NG SCREEN
FOR UNCHARGED
SMOKE
SMOKE
PATHS
FIELD
POSITIVE
FIELD
"ON"
SMOKE)
"CFFU
Figure 26. Schematic drawing of electrostatic hood system.
47
-------
Model of experimental electrostatic hood in operation. Left: With electrostatic field
"off", the smoke rises through the cones. Right: With field "on", the smoke is drawn
into the space between the cones and carried out the top of the system.
-------
-
"Pushing" uncharged aluminum chloride smoke with electrostatic
curtain. Above: Curtain "off". Below: Curtain "on"; smoke
is pushed backwards.
49
-------
AIR
INPUT
GROUNDED SCREEN
BRUSH IONIZER
ELECTRIC WIND
BRUSH IONIZER
BACK SIDE
OF SYSTEM
CERAMIC INSULATOR
FLEXI BLE INSULATOR
( POROUS )
Figure 29.
Schematic drawing of electrostatic curtain system with
grounded screens on both sides. h
50
-------
HOOD
\
SMOKE PATH
FIELD.
\
"OFF*
CONVERTER
ELECTROSTATIC'
CURTAIN
SWING UP OF
v CURTAIN SUPPORT
\ FOR CRANE PASSAGE
Y
CURTAIN
SUPPORT
POWER SUPPLY ,
INSULATOR, AMD
CURTAIN ROLL-UP
SYSTEM
CHAIN MAIL
CORONA POINT
Figure 30. Schematic drawing of electrostatic curtain system on copper
smelter converter.
51
-------
-------
— -•—10.2 cm D.
152 cm
FOG GUN
CHARGED FOG
DUST BOIL-UP
Figure 32. System for control of simulated dust boil-up by charged fog
gun.
53
-------
in
-
s •
•!f
•' "' ' '*$£
; •>Jlf&iia
rff-31
•-V.,"
• •:•
-
Figure 33. Dust boil-up created when additional dust drops down the pipe.
54
-------
1
\**$r*m®.:
.• ' '.' ' -.-'^''
- - .:•• ,v -\... ' i> .- "-.- •
.... ,.; ';;51i
Figure 34. Application of charged fog to dust boil-up shown in Figure 33.
-------
BRASS RING
3.5 mm D.
FALLING
DUST
127 mm D.
-a.
10 kV
/
"7 \
106 mm D. —^
•100 mm D.-
35 mm
15 mm
Figure 35. Schematic drawing of optical dust control system.
56
-------
MOVEABLE COARSE
SCREEN
PHONOGRAPH NEEDLE (TYR)
SHIELD
GROUNDED
SIDE
AIR FLOW TO KEEP
INSULATOR CLEAN
HIGH VOLTAGE
SIDE
Figure 36. Schematic drawing of electric wind system for cleaning insulating shield; the "wind"
generated by the phonograph needles keeps the insulator from building up deposits.
-------
WATER DROPS
WATER
RUN-OFF
MATERIAL PLEXIGLAS
OR BORON NITRIDE
^r- 35 kV
Figure 37. Schematic drawing of high voltage insulator for wet environ-
ments (scale: full).
58
-------
7.0
6.0
50
Ul
10
3.O
2.0
1.0
O
OUST DENSITY
mg /
DRY DUST LEVEL
UNCHARGED FOG
-) FOG
(-H FOG
PARTICLE,
(M|CRONS)
O I
0
Figure 38. Control of grain dust with water fog. Water flow 10 ml/mini air flow 4.81 m3/hr.
-------
a\
O
ARBITRARY
UNITS
lOOr
80
60
20
O
INITIAL COTTON
OUST LEVEL
2.1 m/m3
m FOG
UNCHARGED
FOG T
NOTE CHARGED FOG DATA WAS
RUN TWICE WITH FOUR
MEASUREMENTS OF DUST
DENSITY PER RUN
M FOG
Figure 39. Control of cotton dust (from towel manufacturing area) with water fog. Water flow
15 ml/min; air flow 150 SCFH (4.25 mVhr). Data taken with GCA Corp. RDM-101 Beta Ray
dust monitor.
-------
O
DUST DENSITY
mq / m3
0
INITIAL
DUST
LEVEL
<-) STEAM AND
/UNCHARGED STEAM
PARTICLE SIZE (MICROMETERS
DUST
STEAM
( 065 X O 65 )
METER TUNNEL
Figure 40. Control of trona dust with charged ateam. Air flow ISO SCFH (4.25 ra'/rain)»
steam flow 15 ml/min.
-------
a\
to
4.0
3.0
2.0
1.0
0.0
DUST DENSITY
mg
INITIAL
DUST LEVEL
PARTICLE DIAMETER (MICROMETERS)
I
Figure 41. Control of sandblasting grit with charged fog. Water flow 15 ml/minj air flow
210 SCFH (5.94 m3/hr).
-------
2.0
a\
10
OS
EXHAUST ••—
DUST DENSITY
ni /
0 —
0
FOQ INLET
SAMPLING
DUST INLET
O'X 2'X 21
OUST TUNNEL
PARTICLE
3-15
DIAMETER I MICROMETERS I
Figure 42. Control of dry red lead battery dugt with charged fog,
air flow ISO SCFH (4.25 m'/hr).
Water flow 15 ml/mini
-------
20
0\
*>.
10
0
ing / m^
DUST DENSITY
INITIAL
DUST LEVEL
CHARGED FOG AND
UNCHARGED FO
L»- 2.9m —|
•DUST
•FOG
O.65 X O.65
METER TUNNEL
PARTICLE DIAMETER I MICROMETERS I
.1 i I « • ir
Figure 43. Control of aluminum oxide dust with charged fog. Water flow 15 rol/min,
air flow 150 SCFH (4.25 3
-------
a\
in
O
OUST DENSITY
TWO HUMS
PARTICLE OIAMETEH < MICHQMETEHS i
,. ^*-,- -
—«—
7
-4--
U
-i—
9
Figure 44. Control of resplrable dust (turkey barn floor sweepings) with charged fog,
Water flow 15 ml/mini air flow 150 SCFH (4.25 m'/h*).
-------
DUST LOADING
ON FILTER
mg
(Tv
[
8
7
6
5
A
3
2
1
n
-
-
•
•
.
•
54%
i
t
t
t
t
i
t
t
t
t
41%
Q INITIAL DUST LEVEL
p
^
•j
)
!
>
K
i
>
K
V
V
S
V
V
s
s
V
s
V
57%
@ UNCHARGED FOG
R(-) FOG
H(+) FOG
^
s
s
s
V
V
V
V
V
V
NET DUST
47% REDUCTION VALUES
Figure 45. Control of cotton brack dust with charged fog,
Reeve Angel 0.1 micron glass filter.
Water flow 10 ml/min; air flow 5.1 in /hr.
-------
8.0
7.0
DUST DENSITY
ON FILTER
-mg /
a\
6.0
5O
%^.«^
40
3.0
2.0
1.0
r\
-
—
-
.
-
n
)
L
;
k
;
>
>
)
s
f
>
n
^. ^v
38.3 % L_2.44m-r ^
REDUCTION ^SAMPLING
.»-, h» — 3.O5 m H
r
/
/
/
/
/
LOCATIONS
(] INITIAL DUST LEVEL
[3 UNCHARGED FOG
^ (4) FOG
0 (-) FOG
2.44 m
3.05
Figure 46.
Control of carbon black dust with charged fog. Water flow 12 ml/rainj air flow 6.2 m3/hr.
Reeve Angel 0.1 micron glass filter; sampling time 2 min.
-------
CO
B
7
6
5
4
2
*
DUST ON
FILTER
119 0.5 fc
.
i GLASS FIBER
FILTER.
<+)
(-)
1
INITIAL
OUST
LEVEL
83%
in
M
0
M
M
FOG AVG. 61 %
FOG AVG. 72.5%
62%
O • — • -" — •-*-•
(-) (-
62%
m
1
1
t
I
1
1
t
6(
H (-) <
)%
I.O^i
GLASS FIBER
FILTER
(•H FOG AVG. 51.5 %
-) FOG AVG.
83%
n
(
t
t
1
f
41%
f) (-) I
CHARGED
\
S
\
s
I
\
40%
y
i
t
t
i
i
t
i
i
i
/
/
) I-
61.5%
62 %
n
V
s
V
\
\
V
s
\
5.0 \t
NUCLEPORE
FILTER
(-*•» FOG AVG. 53.5 %
(
-) FOG AVG. 64.5 %
75%
(+) (
FOG POLARITY
f
f
/
t
-
61%
) (
y.
^1
^
yl
s)
54%
tr\
+ ) (-
r-
_
46%
n
V
V
N
\
y
i
y
y
\
-\ (M )
Figure 47.
Control of bauxite dust with charged fog. Water flow 20 ml/min, air flow 5.1 m3/hr
-------
ID
6
7
6
5
4
3
2
1
rt
1 l»- 1 L-M t-V»r»l^
mg
•
-
•
-
-
•
__ _
75%
S
•
•
IM
68
\J
°/
1
_.
r
o
I
66 "X
I
j
65 "y
I
1
75
^
X
D
I
• • .
52%
!
(-)FOG H)FOG
0.5 MICRON
GLASS FILTER
(-)FOG (+)FOG
1.0 MICRON
GLASS FILTER
HFOG (-f)FOG
5 MICRON
NUCLEPORE FILTER
Figure 46. Control of gypsum dust with charged fog. Water flow 20 ml/minj air flow 5.1 m3/hr
-------
14
12
10
8
DUST
DENSITY
INITIAL DUST
LEVEL
I 23456789
PARTICLE DIAMETER (MICROMETERS)
figure 49. Control of calcium propionate dust with charged fV.-r
4.24 m3hr.
v;ai:«ii i:)..-r 20 ml/min; air flow
-------
CHARGED FOG GUN
WELDER
4"D.(IOcm) X 24"(59cm)
STOVEPIPE
FAN
DUST / FUME DETECTOR
figure 50. Schematic drawing of experimental test system for control of welding smoke and fumes.
-------
-4
to
7
6
5
4
3
2
1
O
•
FILTER
LOADING
mg
n
FOG WATER FLOW
15 ml / m
56%
7
t.
/
/
X
/
in
(-) FOG
^ INITIAL L
% REOUCT
/AFTER FC
(
I
59
f~r~
r
'
/
,
j
'/
(
%
59%
p-
'.
',
/
n
O5(j GLASS
FIBER
FILTER
l(j GLASS
FIBER
FILTER
GCA
i
FOG WA
40 ml / i
5
n "
/
/
T
ni
Ef
n
•
\
i
15
r™
;
/
/
FLOW
-) FOG
.— .
4O%
ff
/
O.5M GLASS Ip GLASS GCA
FIBER FIBER
FILTER FILTER
Figure 51. Reduction of metal welding fume with charged fog. Air flow 2.83 ma/hr.
Miller Electrical Mfg. Co., Appleton, Wisconsin, Model 35-S continuous wire welder.
-------
0
FILTER
LOADING
.nig
FOG WATER FLOW
IS ml/mln (-> FOG
FOG WATER FLOW
10 ml/mln (-) FOG
67'
INITIAL LEVEL
% REDUCTION
'AFTER FOGGING
46%
71
47%
68%
fn
663%
71
63%
tn
0.5M GLASS
FIBER
FILTER
|(i GLASS
FIBER
FILTER
GCA
0.5(i GLASS
FIBER
FILTER
l(j GLASS
FIBER
FILTER
GCA
Figure 52. Reduction of metal welding fume with charged fog. Air flow 2.83 mj/hr,
Miller Electrical Mfg. Co., Appleton, Wisconsin, Model 35-S welder.
-------
8"
SANDING BELT
V
FOG GUN
DUST
GCA UNIT
SIDE VIEW
SANDIMG BELT
FOG GUN
Q
D
WORK
/GCA
/UNIT
O
DUST
TOP VIEW
Figure 53. Schematic drawing of dust control system for industrial sander.
-------
Reduction of dust from sander with charged fog. Left
without charged fog. Right: With charged fog on.
Figure 54.
-------
vR'sr
> -•< • \
*~'•*••* i *
^ vi
• . ^ ^K
"^i-Wrfwr t;
UIH«
111 in in
Figure 55. Sandblasting without charged
-------
Figure 56. Sandblasting with charged fog for dust control 7p^
-------
Figure 57
Effect of charged fog on dust at copper smelter furnace
clean-out operation. Above: Without fog. Below: Use
of charged fog has eliminated blowing dust.
-------
e
to
DUST LOADING
ON FILTER
mg
REEVE ANGEL 934 All
O.I MICRON GLASS FILTER
INITIAL DUST
LEVEL
,(+\ FOG
/<-)FOG
/ /NO CHARGE
/ //a 7. 3% DUST
/ II REDUCTION
••"•^
NUCLEPORE
8 MICRON FILTER
NUCLEPORE
0.6 MICRON FILTER
INITIAL DUST INITIAL DUST
' LEVEL /LEVEL
,< + ) FOG / .I-H FOG
If-) FOG V/ lf-\ FOG
//03.2V. DUST //83.3V. DUST
//REDUCTION //REDUCTION
Figure 58. Effect of charged fog on cotton dust in an operating cotton gin. Water flow 50 ml/minj
air flow 1.13 m3/hr» sampling time 4 minutes.
-------
00
o
l.b
1.0
0.5
FILTER DUST
LOADING n>g
SAMPLER AND FOG GUN AT TOP OF PRESS
REEVE
01 Ml
GLAS
INITIAL
DUST
LEVEL
^^xT^
ANGEL
CRON
FILTER
NUCL
a MI
FILTI
69.1% NET
REDUCTION
"\ /
EFFECT 01
n
— - J
II
NUCLEPORE
O.6 MICRON
FILTER
EPORE
CRON
R
64.6% NET
REDUCTION
--
(-) FOG 1 1
^^
^^
33
RE
1
% NET
DUCTION
SAMPLER AND FOG GUN AT CENTER OF PRESS
INITIAL
DUST
LEVEL
J 1 NUCLEPORE
NUC
8 M
MACHINE FIL1
MALFUNCTION
TEST STOPPED
EARLY
n
LEPORE °6 »
ICRON FILT
-£R
62% NET
REDUCTION
EFFECT OF
(-) FOG 1 1
MICRON
ER
70.0% NET
REDUCTION
Figure 59. Effect of charged fog on cotton dust with fog gun at two different locations in
cotton gin. Water flow 100 ml/mini air flow 4.5 m /hr.
-------
FOG GUN
SAMPLER
rOG GUN
J
c
J
FOG GUN
FOG GUN
SIDE VIEW
TOP VIEW
Figure 60. Fog gun system for dust control in duct.
-------
DROP 30X
BELT,
CANVAS COVER
FOG GUN
SAMPL£3
SIDE VIEW
FROM T VI EW
FOG GUN
CANVAS COVER USED
' FOR TEST CONTROL
SAMPLER
3-0 VIEW
Figure 61. Fog gun system for dust control in a drop box.
32
-------
TUBE USED FOR TEST CONTROL
00
GRINDER
FOG GUN
SAMPLER
FAN
Figure 62. Fog gun ayatern for dust control on grinder.
-------
GRINDING WHEEL
FOG GUN
GRINDING MATERIAL
SAMPLER
FOG GUN
SAMPLER
\
GRINDING WHEEL
-GRINDING MATERIAL
J L
SIDE VIEW
FRONT VIEW
Figure 63. Fog gun system for dust control during grinding,
-------
I RAPPER
FALLING
DUST
SHEET
CORONA
/"WIRE
CHARGED M
FOG
INJECTOR
OUST
COLLECTING
PLATE
J
DUST
HOPPER
U
\ / SCREW
CONVEYOR
Figure 64. Fog gun system for control of dust boil-up during rapping
an electrostatic precipitator.
35
-------
CD
a\
V
BETA RAY
SOURCE
DRY DUST
rr. •••••-f
^COLLECTING
PLATE
TUBE
V
BETA RAY
SOURCE
WET DUST
.COLLECTING
PLATE
GEIGER
TUBE
Figure 65. Differences in pile-up of dry and wet dust in GCA RDM-101 dust monitor.
-------
GO
08
0.7
O6
OS
O.4
0.3
0.2
01
OUST LOADING
ON FILTER
FLY ASH
( MAGMA I
NUCLEPORE CORP.
8 MICRON FILTERS
ft
LIMESTONE
COKING
COAL
RELATIVE OPTICAL DENSITY
*45 6 7 B~
IO
Figure 66. Calibration curve for converting optical data to actual dust loading, assorted dust
materials.
-------
TECHNICAL REPORT DATA |
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-600/7-79-078
4. TITLE AND SUBTITLE
Fugitive and Fine Particle Control Using
Electrostatically Charged Fog
7. AUTHOR(S)
Stuart A. Hoenig
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Arizona
Tucson, Arizona 85721
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION- NO.
March 1979
6. PERFORMING ORGANIZATION CODE
1
8. PERFORMING ORGANIZATION REPORT NO.
1
10. PROGRAM ELEMENT NO. 1
EHE624
11. CONTRACT/GRANT NO.
Grant R805228 '
13. TYPE OF REPORT AND PERIOD COVERED
Final; 10/77 - 12/78 i
14. SPONSORING AGENCY CODE '
EPA/600/13 (
^.SUPPLEMENTARY NOTES T£RL-RTP project officer is Dennis C. Drehmel, MD-61, 919/54]'
2925.
1
'PKa v9tmf\mt' /T-ITTOO vtaonltc f\f o o+tirlTT f\f fiirri+4ir*i o«/1 f^na i^OT»4*4/»la /»rk*s^i*/%l '
using electrostatically charged fog. Most industrial pollutants acquire an electro-
static charge as they are dispersed into the air. Exposing this charged airborne i
material to an oppositely charged water fog enhances contact between the particles
and the fog droplets. After contact, the wetted particles agglomerate rapidly and
fall out of the atmosphere. This technique has been tested on a wide variety of indus- j
trial pollutants ranging from silica flour to SO2 and fly ash. In general, there has
been significant suppression of pollution with a minimum of water fog. In addition,
electrostatic hoods and screens can be used to push or direct pollutants to the pro- |
per area.for collection. The system is therefore well suited to control of moving
fugitive dust sources where the usual duct and baghouse systems are ineffective or
too costly. The charged fog systems are now being tested in various industrial appli-|
cations with generally good results. All work to date, including industrial applica-
tions that have been released by the companies involved, are discussed in the re-
port. The report covers work on: developing new charged fog systems for controlling!
fugitive dust; demonstration testing of the systems in industrial locations; and de-
signing and constructing a high-temperature stack simulator for fog gun testing.
17. KEY WORDS AND DOCUMENT ANALYSIS I
a. DESCRIPTORS
Pollution Leakage
Dust Cyclone Separators
Aerosols
Fog
Electrostatics
Processing
18. DISTRIBUTION STATEMENT
Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Particulate
Fugitive Emissions
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATI Field/Group
13B 14B ,
11G 07A,13I 1
07D
04B ,
20C !
13H
99
22. PRICE
1
EPA Form 2220-1 (9-73)
88
------- |