EPA-650/2-75-043
February 1975
Environmental Protection Technology Series
INVEST
PARTICULATE MATTER
MONITORING USING
CONTACT ELECTRIFICATION
«*
53SZ
LU
o
PRO
-------
EPA-650/2-75-043
INVESTIGATION OF
PARTICULATE MATTER MONITORING
USING CONTACT ELECTRIFICATION
by
Dr. Walter John
California State College, Stanislaus
Turlock, California 95380
Grant No. 802726
ROAP No. 26AAM-65
Program Element No. 1AA010
EPA Project Officer: Mr. John Nader
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, D. C. 20460
February 1975
-------
EPA REVIEW NOTICE
This report has been reviewed by the National Environmental Research
Center - Research Triangle Park, Office of Research and Development.
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environ-
mental Protection Agency, have been grouped into series. These broad
categories were established to facilitate further development and applica-
tion of environmental technology. Elimination of traditional grouping was
consciously planned to foster technology transfer and maximum interface
in related fields. These 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
9. MISCELLANEOUS
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.
This document is available to the public for sale through the National
Technical Information Service, Springfield, Virginia 22161.
Publication No. EPA-650/2-75-043
11
-------
ABSTRACT
The theory of the charging of aerosol particles by contact electricity
is reviewed, as well as the development of monitors for particulate
matter based on this principle. Data on the performance of these
monitors is scarce and sometimes contradictory. In the present work,
laboratory tests were carried out with a variety of test dusts. The
results show that the dynamic response of the contact electricity
monitor tracks well with that of an optical detector. The total charge
correlates well with the gravimetric mass. Humidity has little effect
on the response, but droplets cause failure of the instrument. Pre-
charge on the particles did not influence the detector. Some evidence
was obtained showing that particle size does not have an important
effect on the response, although there is a cutoff for very small
particles. Results for the sensitivity of the instrument can be grouped
according to the electrical resistivity of the material. It is found
that the condition of the surface of the Inconel probe has a major
effect on the sensitivity. Additional work is necessary on this aspect
of the detector. The theory of Cheng and Soo for the charging of metal
particles is discussed. There is at present no theory applicable to
insulators.
iii
-------
CONTENTS
List of Figures vi
List of Tables vi
Acknowledgements vii
SECTIONS
I Conclusions 1
II Introduction 2
III Theory of Charging of Aerosol Particles
by Contact Electricity 4
IV Contact Electricity Monitors for
Particulate Matter 9
V Experimental Measurements 15
VI References 36
VII Bibliography 38
VIII Appendix - Theory of Contact Charging of
Aerosol Particles 41
-------
LIST OF FIGURES
Page
No.
1 Experimental Layout 16
2 Dust Generator 19
3 Dynamic Response 23
4 Charge vs. Mass for Aluminum Oxide 25
5 Charge vs. Mass for Aluminum 27
6 Sensitivity vs. Date 33
LIST OF TABLES
No.
1 Sample Characteristics 21
2 Sensitivity Measurements 34
vi
-------
ACKNOWLEDGEMENTS
This project was supported by the U.S. Environmental Protection Agency
under Grant No. R-802726-01. The project officer was Mr. John Nader,
National Environmental Research Center, Research Triangle Park, N.C.
I thank Mr. Nader for his encouragement of this work and for helpful
suggestions. I express my appreciation also to Dr. Carl Gatlin,
President, California State College, Stanislaus, and to the College
Foundation for their sponsorship of this project.
Mr. Joseph Zurlinden assisted with all phases of the experimental work.
Thanks are also due to Mr. Arnold H. Gruber and 1KOR, Inc. for supplying
valuable information on the IKOR AQM.
vii
-------
SECTION I
CONCLUSIONS
The pnisent experimental results confirm the promising findings reported
by others concerning the dynamic response of the contact electricity
monitor, as well as the good correlation of total charge to mass deter-
mined gravimetrically. No evidence was found for disturbing effects of
humidity (below the dew point) or precharge. Particle size was found
not tc have an important influence on the response, although a cutoff
at very small particle size exists.
The sensitivities for various substances can be grouped according to
electrical resistivity. This is predicted by the theory of Cheng and
Soo for metals and good semiconductors. There is no applicable theory
for insulators. Sensitivities ranged over a factor of 60, so that
some constituents of a mixture will be difficult to detect.
It was found that the condition of the surface of the stainless steel
probe had a major influence on the sensitivity. Additional work is
necessary before this surface effect can be understood, or at least
controlled.
-1-
-------
SECTION n
INTRODUCTION
When aerosol particles come into momentary contact with a surface, they
1 2
may acquire an electrostatic charge as a result. ' This phenomenon
3 8
has been used as the basis of a monitor for particulate matter. ' An
aerosol stream can be drawn past an insulated probe. Particles colliding
with the probe then transfer or carry away charge from the probe. The
electrical current from such a probe has been found to correlate
surprisingly well with the mass concentration of the aerosol. '
The contact electricity type of monitor offers some important advantages
by making possible in-situ measurements in real time. The monitor can
respond fairly rapidly to changes in mass concentration and is sensitive
over a wide range, particularly to high concentrations. The electrical
signal is convenient for data collection and the device is comparatively
simple in construction. The history of these instruments goes back more
than a decade, and at least two models have been available commercially.
•H/. R. Harper, "Contact and Frictional Electrification", Oxford U.
Press, Oxford, (1967).
2L.B. Loeb, "Static Electrification", Springer-Verlag. Berlin,
(1958).
,A. Schutz, Staub 24_, 359 (1964).
*A. Schutz, Staub J26, 18 (1966).
5A. Schutz, Staub-Reinhalt der Luft, Band 26 (1966) No. 5, Seite
198/201, 1-4.
6R. Prochazka, Staub 24_, 353 (1964).
7R. Prochazka, Staub 26_, 22 (1966).
8L. Cheng and S.L. Soo, J, Appl. Phys. 41, 585 (1970).
^H. Schnitzler, SchrReihe Ver. Wass.-Boden Lufthyg, Berlin-Dahlem,
V. 33, Stuttgart (1970).
^IKOR, Inc., unpublished reports, and private comm. from A.H.
Gruber.
-2-
-------
In spite of this, the contact electricity monitor has not come into
general use. This may be partially due to a feeling on the part of
potential users of the need for better understanding of the physical
principles involved in this type of monitor. * The present investi-
gation was undertaken to assess the current state of knowledge of these
principles. In addition, laboratory experiments vere devised to determine
the important parameters involved in the operation of these instruments.
•^G.J. Sem, et al, "Instrumentation for Measurement of Particulate
Emissions from Combustion Sources", Vol. I: Particulate Mass, Therrao-
Systems, Inc. APTD-0733, (1971).
-3-
-------
SECTION III
THEORY OF CHARGING OF AEROSOL PARTICLES BY CONTACT ELECTRICITY
General Theory
It is necessary to discuss first the terminology used. "Contact" elec-
tricity refers to charge transferred as a result of pure contact with
no sliding or rubbing. On the other hand, "triboelectricity" involves
1 2
rubbing (tribo is Greek meaning rubbing). ' The latter may include
the transfer of material and local heating. In the case of the
charging of aerosol particles, it is not clear which type of inter-
action takes place. Probably all types are involved to some degree.
In the present work, we shall use the term "contact" in discussing
the charging of particles, recognizing that other kinds of interactions
may well be present.
Pure contact is difficult to achieve in practice. Harper1 has succeeded
in some careful experiments with large metal spheres. His results agree
with the theory for contact between metals. When two metals touch, the
difference in the contact potentials of the two materials will cause a
flow of electrons to take place. The buildup of charge will continue
until the resulting electric potential equals the difference in the two
contact potentials.
-4-
-------
This charging of metals can also be described in terms of the quantum
mechanical energy band theory.1'12'13 In a metal, the energy levels
are filled up to a certain maximum energy, the Fermi level. The
additional energy from the Fermi energy up to zero energy is called the
work function. When two metals are brought into contact, the Fermi
levels will initially be at different heights (energies). Electrons
will then flow from the higher to the lower until the two Fermi levels
are equalized. At that time, the difference in work functions gives the
difference in potential of the two metals.
In the case of metals, some charge can flow even when there is a small
gap between the surfaces by the quantum mechanical tunnel effect.
A serious complication arises during the separation of the metals.
The air breaks down and an appreciable back-flow of charge takes place.
The net charge transferred as a result of the contact then depends on
the magnitude of the backflow.
The theory for metal-metal contact can be generalized to include semi-
conductors.1'12 Whereas for metals the charges reside on the surface,
for semiconductors there are fewer electrons available at the surface
and there will be a flow of charge from the interior. Since there is
now an appreciable resistivity, there will be an associated time delay
12"Static Electrification, 1971", Proc. of the Third Conf. on
Static electrification organized by the Static Electrification Group
of the Institute of Physics held in London, May 1971. Conf. Series,
No. 11, The Institute of Phys., London and Bristol.
13"Electrostatics and Its Applications", A.D. Moore, Ed., John
Wiley & Sons, N.Y., (1973).
-5-
-------
with a time constant in seconds given by 8.85-10 Kp, where K is the
dielectric constant and P the resistivity in ohm-cm. This implies
that the amount of charge transferred will depend on the duration of
contact. Even for large spheres, the time constant becomes too large
for appreciable flow of charge in the case of semiconductors having
high resistivity.
Insulators have an even higher resistivity so that there is no flow of
charge from the bulk material. Surface states become the important
source of charge for transfer. Such surface states are complicated,
depending on the detailed physical and chemical condition of the
surface. Adsorbed ions giving rise to electric double layers play
an important role. Harper states that the charge transferred
consists of ions rather than electrons. It may be mentioned also
that there is a class of insulators designated electrophobic which
apparently have no surface levels available for contact charging, if
the surfaces are clean. These materials include some common plastics
such as nylon, lucite and teflon.
In the case of insulators particularly, there are other charging mech-
anisms besides contact charging which may have to be considered, such
as electrolytic effects for moist surfaces and frictional sffects
1 2
resulting in transfer of material. '
The foregoing has been a brief review of the status of knowledge of
contact charging. The reader is referred to the bibliography for
further information. He can sumarize by saying that our knowledge
-6-
-------
is good or quantitative for metals but becomes progressively worse
as we make the transition to insulators. This applies to ideal
surfaces; in real cases, the charging mechanism may be drastically
altered by contamination of the surfaces or by the physical condition.
Charging of Aerosol Particles
When the theory of contact charging is applied to aerosol particles,
there are some additional considerations. We are concerned with the
impact of the particle with a solid surface, a metal in the practical
case of a probe. The charging is a dynamic process;** the duration
of contact is short, of the order of 10~9 to 10~10 seconds for our
present purposes. Therefore, there will be no time for flow of charge
from the bulk material for substances having resistivities greater
than about ID-* ohm-cm.
The surfaces of metal particles will invariably be oxidized. According
to Harper,t1' such oxide layers behave as a metal with a work function
of about 5.5eV. Most oxides are semiconductors; when the layer is
thick, it behaves as a semiconductor. If the layer is not too thick
then, charge can be transported either by conduction or by tunnelling.
A serious attempt to calculate the contact charging of metal particles
has been made by Soo and his collaborators.3'1^ Cheng and
•^S.L. Soo, "Dynamics of Charged Suspensions", Topics in Current
Aerosol Research, Vol. 2, International Reviews in Aerosol Physics and
Chemistry, Pergamon Press Ltd., Oxford, (1971).
-7-
-------
consider the impact of spherical particles in some detail. This
theory greatly overestimates the magnitude of the charging. Cheng
and Soo point out that the theory applies to clean surfaces; surface
contamination could have an important influence on the charge transfer.
In Appendix I, this theory is discussed in terms of the present
application.
In conclusion, some progress has been made in understanding the contact
charging of particles, but theoretical predictions are not yet quanti-
tative, and there is no theory available for insulators.
-8-
-------
SECTION IV
CONTACT ELECTRICITY MONITORS FOR PARTICULATE MATTER
Review of Instrument Development
An early version of an instrument employing a spherical metal probe was
o
described by Schiitz. A later version of his instrument utilized a
probe-in-nozzle technique. Prochazka » has described the develop-
ment of the Konitest, an instrument which was produced commercially for
a time. In the Konitest, the electrode is a tube of a semiconductor,
steatite (magnesium hydrosilicate). The gas is introduced radially
and the resulting helical path causes the particles to impinge on the
walls of the tube. In a second version, the tube is shaped as a Venturi
g
nozzle. Soo and his collaborators constructed instruments using a
spherical metal electrode and also a tubular electrode. In the USSR,
Kisler has described some contact electricity instruments including
a monitor using a wire probe. The IKOR Air Quality Monitor was apparently
independently developed and is now commercially available. It utilizes
a bullet-shaped Inconel probe in a pipe.
In all of the instruments mentioned above, the transfer of charge to the
probe by particles colliding with it results in a current which is con-
tinuously monitored with an electrometer. Some other types of instruments
have been developed to observe the electrical pulses produced by individual
15S. YA. Kisler, Mekh. Avtomat. Proizvod., 2£ (9): 27-28, (1972)
-9-
-------
particles. Min, et al. described such a device whereby particles
charged by wall collisions induce a voltage on an electrode. Detectors
for individual particles, however, appear to be applicable only to
large particles carrying substantial charges. Other types of monitors
have been developed which utilize electrostatic charging of particles
but not contact charging. Grindell 17 described a dust monitor which
charged the particles in a high voltage corona. The particles were
then precipitated on a collector electrode and the current determined.
18
In the instrument developed by Coenen, the particles acquired charge
by repeated collisions with the plates of a high-voltage capacitor.
Characteristics of the Contact Electricity Monitor
Some experimental investigations have been made of the characteristics of
the contact electricity monitor. These investigations are few in number;
most are under field rather than laboratory conditions. On some of the
most important characteristics, there are contradictory reports. The
status is summarized below by topic.
(1) Electrical current vs. mass concentration. In most cases the instru-
ments are to be used as an indirect measurement of the particulate mass
concentration in the gas monitored. It is most important to establish
the correlation between the electrical current from the probe and the
mass concentration measured gravimetrically. Schfltz reported the
16K. Min, B. T. Chao, and M.E. Wyman, Rev. Sci. Instr. 34, 529 (1963)
17D.H. Grindell, Proc. Inst. Elec. Engrs. (London) 107A; 353-365,
(Jan. 1960).
18W. Coenen, Staub 27., 32 (1967).
-10-
-------
following relationship for quartz dust:
I = akb
where I is the electrical current in amperes, k is the mass concentration
in mg/nr*, the exponent b ranges between 1.26 and 1.30, and a is inversely
proportional to the particle diameter.
On the other hand, Prochazka *> found the Konitest to give an accurately
linear indication for concentrations from 0 to 3g/m^. The dusts were
sampled at various industrial sites. Ito, et al. ^ also reported a
linear relationship for a Konitest monitoring cement kiln exhaust at
the exit of an electrostatic precipitator for concentrations up to
10g/m3. Schnitzler, et al. 9 found a correlation coefficient of 0.92
between the gravimetric mass and the Konitest reading for effluent from
a coal-fired plant. The performance of the Konitest compared very
favorably to the best response obtained from transmissometers and beta
radiation attenuation monitors at the same location.
Q
Cheng and Soo ° obtained plots of mass flow vs. probe current using coal
dust. The plots show small deviations from linearity, being concave
upward.
Field tests of the IKOR Air Quality Monitor show that the integrated
current (total charge) has a reproducible ratio to the gravimetric mass
for aluminum oxide. Good agreement was also obtained between the
19
*T. Ito, H. Saito, and N. Furuya, Proc. Japan Soc. of Air Poll.,
13th, (1972), p. 247.
-11-
-------
total charge and the mass concentration obtained with an EPA sampling
train. These tests did not span a wide range of concentrations.
(2) Dependence of the current on particle size. As pointed out above,
Schiltz found the constant a to be inversely proportional to particle
diameter. Thus, higher sensitivity is predicted for small particles.
o
On the other hand, Cheng and Soo predict a current independent of
particle size for one experimental configuration. They report that
this was verified by Min. Ito, et al. 19 report the Konitest to be
little affected by particle size when sampling cement dust.
(3) The effect of humidity. Water layers on surfaces usually contain
dissolved impurities. Some authors believe electrolytic ions are
2 1
important for contact charging. However, Harper does not, and
tried unsuccessfully to obtain an experimental correlation between
humidity and charging.
Schiitz ^ reported no effect on his instruments' performance for relative
humidity up to 99%. He did warn that droplets of water cause erroneous
indications. Cheng and Soo ® point out that the conductivity of
19
mineral particles increases with the relative humidity. Ito, et al.
state that humidity caused no problems in their tests of the Konitest.
Kisler 15 states that the humidity of the gas should be below the dew
point.
(4) Temperature effects. Schiitz4 stated that the contact electricity
monitor should not be operated above 70°C due to the presence of thermal
emf 's. However, a modified IKOR AQM has been operated at 593°C.
-12-
-------
(5) Probe material. With the exception of the Konitest, most instruments
have used metal probes. According to Schfltz, the use of metal for the
probe ensures that the charging will always be positive. It seems clear
that this is a dubious assumption; in fact in field tests of the IKOR the
sign of the current has been found to vary with material and the standard
IKOR AQM is equipped with an automatic polarity-reversing circuit to
maintain positive indication on the recorder.
(6) Composition of dust sampled. Although field tests have been conducted
with a variety of dusts, there has been no reported correlation of the
response according to some characteristic of the dust sampled. Schtltz
concluded that the influence of the properties of the material was far
19
less decisive than the particle size. Ito, et al. stated that the
current obtained for a given mass concentration depended on the material,
but that for cement dust the electrical resistance had little effect on
the output. Most users of the contact electricity monitors have simply
empirically calibrated the monitor for each type of dust sampled.
o
According t«j the theory of Cheng and Soo, the amount of charging
depends on the electrical conductivity, the density, and the elastic
properties of the materials of both particles and probe. (See Appendix I)
(7) Discussion. The above review indicates that the mass correlation
observed by several experimenters is surprisingly good. However, there
is need for additional work to clarify some of the conflicting reports
-13-
-------
in order to identify the real problems associated with this type of
instrument. There is also a need for tetter understanding of the
charging mechanism.
-14-
-------
SECTION V
EXPERIMENTAL MEASUREMENTS
Experimental Arrangement
Because of the Incomplete state of the theory of the contact electricity
type of monitor and because experimental information is sparse and
sometimes contradictory, a series of controlled laboratory tests were
devised to investigate the response of such a detector under various
conditions. The goal was to determine the important parameters affecting
the operation and to try to understand these in terms of the theory.
The general arrangement is shown schematically in Fig. 1. Room air is
drawn into a metal manifold 70 cm long x 3.2 cm inside diameter. Inlets
were provided for adding humidified air and aerosol generated from test
dusts. The manifold ended in the inlet of the IKOR Air Quality Monitor.*
Another inlet nearby led to an optical analyzer. The main aerosol stream
passed the IKOR probe, through the filter (or blank) and then was impelled
by the IKOR blower to the exhaust.
The airflow rate through the IKOR AQM was between 8 and 14 £/s (17-30 CFM) .
The room air temperature was maintained at 23°C by the laboratory air
conditioning and the relative humidity was stable over long periods of
time, being typically at about 50%. Even on the most sensitive scale,
no room dust could be detected by the IKOR AQM.
*IKOR, Inc. Burlington, Mass.
-15-
-------
Exhaust
t
Figure 1. Experimental Layout
Double lines denote pipes for aerosol, single lines denote electrical cable.
Hygrometer
n
Humidity
Generator
t
L
IKOR
Sensor-
Probe
Gravimetric
Filter
Dust
Generator
Source of
Compressed
Air
CLIMET
Optical
Analyzer
Air Filter
and Regulator
Chart
Recorder
IKOR
Integrator
IKOR
Control
Unit
Single
Channel
Analyzer
Multi-
Channel
Spectrum
Analyzer
Chart
Recorder
Counter
-------
The IKOR Air Quality Monitor
A Model 206 IKOR AQM was used for this work. This unit consists of a
stainless steel sampling pipe, 2.4 cm inside diameter x 135 cm long,
connected to a flexible Teflon-lined hose 318 cm long. The hose
conducts the aerosol stream to the sensor unit where it passes
through a stainless steel pipe 2.4 cm i.d. containing an Inconel 625
probe shaped like a bullet and centered in the pipe. At the widest
part the probe is 2.0 cm in diameter. The probe is insulated and
connected to an electrometer circuit which is of a modern operational
amplifier (integrated circuit) design. The electrical current is
read out on a panel meter and on a Rustrak recorder. The readout is
linear with a switch for 5 decades of sensitivity. During the present
work, the absolute current was determined with a Keithley 610 C
Electrometer; full scale on the IKOR meter ranges from 10~^ to 10~^ amp.
The IKOR unit was also equipped with a current integrator so that the
total charge accumulated over a period of time could be determined. The
IKOR electronics was very stable, maintaining its zero setting and
calibration over two months of operation.
The airflow rate of the IKOR AQM is determined by difference in pressure
sensed by pitot tubes. Thermocouples are provided for measurement of
gas temperature.
-17-
-------
The Optical Analyzer
A Climet CI-201 Optical Analyzer* was used to obtain an independent
measure of aerosol concentration. In this instrument, forward-scattered
light is sensed by a photomultiplier. The output of a count-rate meter
is displayed on a chart recorder with a logarithmic scale. The analog
pulses from individual particles were passed through a single channel
analyzer and then to a counter. The analog pulses were also processed
by a 1024 channel pulse-height analyzer to obtain particle size spectra.
The system was calibrated with monodisperse latex spheres. It was
determined that the spectrum analyzer operated between 1.0 and 5.5 microns.
The counter had a threshold set at 3.7 microns in order to reduce the
counting rate to -manageable levels.
Dust Generator
The dust generator devised for this work is shown in Fig. 2. Compressed
air is filtered, regulated, and passed through copper tubing to a
hypodermic needle (hole dia. 4-10" cm) positioned near the dust sample
in the bottom of the test tube. The jet stirs up the dust; the air
stream containing entrained particles then rises past the washers
(intended to intercept large particles) and leaves via the side arm.
The aerosol stream then emerges as a jet which Impinges on the bottom
of the second test tube. This was designed to break up aggregates and
to further disperse the dust.
*Climet Instruments Company, Redlands, California
-18-
-------
Figure 2
Dust Generator
Compressed Air
Hypodermic-
Needles
-19-
-------
The generator effectively discriminated against large particles or aggre-
gates. For example, very little output could be obtained with stainless
steel particles with diameter 24-48ynu Also, no appreciable output could
be obtained from a sample of carbon black particles 0.47ym diameter which
agglomerated strongly. For all samples used, the peak of the size dis-
tribution as obtained from the optical analyzer was below one micron.
This was irrespective of the size information provided by the supplier
of the sized samples. In the case of stainless steel samples, a bimodal
distribution was obtained; however, the preponderance of the particles
were in the less-than-one micron size group.
Samples used included polydisperse aluminum oxide and polydlsperse aluminum.
These materials were replaced often in order to avoid depletion of the
smaller particles. The remaining samples were sized in varying degrees.
Table 1 lists the characteristics of the samples.
A maximum flow rate of 11 £/min was achieved by the generator, with dust
•»
concentrations ranging up to 0.1 g/m . The maximum dust concentration
which could be generated depended on the type of sample. Also, the
stability of the concentration level depended on the sample characteristics.
For some samples, the concentration was stable for extended periods,
for others frequent adjustments were necessary. With this generator,
tests could be run for about one hour with 1 g of sample.
-20-
-------
Sample
Titanium Dioxide
Red Iron Oxide
Silicon Dioxide
Fly Ash
Aluminum Oxide
Glass Beads
Glass Beads
Molybdenum
Tungsten
Carbon Black
Stainless Steel
Silver
Copper
Aluminum
Aluminum
TABLE 1
SAMPLE CHARACTERISTICS
Origin
P.I.S.* No. 2-12
P.I.S, No. 25-4
P.I.S. No. 10-6
P.I.S. No. 14-1
Ma theson, Coleman
& Bell
P.I.S. No. 8-7
P.I.S. »o. 8-1
P.I.S. No.' 19-6
P.I.S. No. 19-1
P.I.S. No. 2-6
P.I.S. No. 27-1B
P.I.S. No. 27-7
P.I.S. No. 27-6
P.I.S. No. 27-2A
Unknown origin
Remarks
0.45pm av. dia.**
10.3m2/g surface area
0.3-0.Sum dia.,
spheroidal oxide
1-4.Sum dia.
Hum av. dia., 82wt.%
less than 44um
Reagent grade powder,
polydisperse
3-8um dia.,
soda-lime glass
0.5-3um dia.,
soda-lime glass
4.4um dia., 100% between
0 and lOum, nonspherical
metal
l.OSum av. dia., 100%
between 0 and Sum,
nonspherical metal
O.Olum av. dia., 97%
between 0.005 and 0.017,
surface area 1000m2/g
6-12um dia., spherical
metal
6-9ym dia.,
spherical metal
5-150um dia.,
spherical metal
spherical metal,
0.5-6um
polydisperse
*Particle Information Service, Grants Pass, Oregon.
**Size information from P.I.S., however, see discussion in Sec. IV concerning the
observed particle size spectra.
-21-
-------
Experimental Tests
Tests were performed to Investigate the dynamic response of the monitor,
the correlation of the total charge to the sampled mass, the effect of
humidity, precharge effects, particle size effects, the importance of the
surface condition of the probe, and the sensitivity for various substances.
These tests are described below.
(1) Dynamic response. The time constant (1/e) of the IKOR AQM is approxi-
mately 10 sec. Since the Climet optical analyzer is orders of magnitude
faster, an R-C network was added to increase the Climet time constant to
4 sec. Fig. 3 shows the superimposed chart recordings of the two instruments
for polydisperse aluminum oxide dust. Since the original IKOR record was
linear, it was necessary to transform it for the log plot. It has been
pointed out before that a logarithmic display is advantageous for a
contact electricity monitor.
Some of the differences in the records, particularly the more extreme highs
and lows of the Climet t:ace, can be attributed to the 2.5 times shorter
time constant of the Climet. Of course, there can be some real differences
in the instantaneous concentrations at the sampling inlets. Also, the
Climet sampled only the large end of the size distribution.
On the basis of many such runs, it is concluded that the correlation of
the current records is satisfactory. There is one difference which should
be mentioned. Sometimes at the very beginning of a run, the IKOR AQM is
-22-
-------
i'lgure
-.jsponse
Comparison of detector signals
for polydisperse aluminum oxide dust
IKOR AQM
OPTICAL ANALYZER
0
100
200
300 400
Time, Seconds
500
600
700
-23-
-------
much less sensitive. The sensitivity may increase by an order of magnitude
in the first minute or so. This may happen even if the instrument has
been previously running on the same dust, sometimes even within the
hour. This sensitizing effect will be discussed further below in Sec. (6).
(2) Correlation of total charge to sampled mass. A series of runs were
•nade with polydisperse aluminum oxide to compare the integrated current
from the IKOR AQM to the mass determined gravimetrically. Gelman type A,
142mm dia. fiberglass filters were weighed to 0.1 rag. Since the room
humidity was very stable, the filters were simply kept open to the room
throughout the operation. Reproducibility of the weights was found to
be within 0.1 mg in most cases. Runs were about 20 minutes in duration,
with the average concentration varied from run to run.
The results are plotted in Fig. 4. The line is drawn with a 45° slope
corresponding to true proportionality. The line was located by eye.
It is concluded tliat the proportionality is satisfactory over a factor
20 in concentration.
It is reasonable to expect a strictly proportional response. If each
particle independently transfers charge to the probe, that charge will
eventually be returned to ground through the electrometer. Since the
probe is maintained very near to ground potential by the circuit, there
should be no charging effects. One could imagine a buildup of particles
on the probe which would then alter the response for subsequent particles.*
This, however, would shift the entire response curve. This was not
observed during the aluminum oxide runs. However, a shift was seen during
-24-
-------
Figure
Charge vs. Mass for Aluminum Oxide
Plot of total charge from the IKOR AQM vs gravimetric
mass of aluminum oxide sample. The line is drawn at
45° (exact proportionality).
100 r
01
-M
•H
§
0) in
hd 10
6
o
o o
o
10
100
Mass, Mg.
-25-
-------
the succeeding aluminum measurements. Mass correlations were also run
for the aluminum dust. The results are graphed in Fig. 5. The numbers
are in sequential order of the runs. There was a very close propor-
tionality for a while, then the sensitivity apparently decreased to the
dashed line. Further discussion of this decreasing sensitivity will
follow in Sec. (6). It is concluded, however, that the response of
the IKOR is proportional to mass for both aluminum oxide and aluminum,
i.e., an insulator and a metal.
(3) Effect of humidity. Moisture was added to the air by passing com-
pressed air through warm, moist filter paper. The relative humidity was
measured by the wet bulb-dry bulb technique. Using aluminum oxide dust,
the relative humidity was varied from 45% to 69%. No effect on the IKOR
response was observed. The upper limii: to such an effect is estimated
at 20%. Incidentally, the IKOR AQM cannot tolerate the introduction of
wet steam. This evidently causes electrode leakage and complete instrument
failure until the probe structure is dry. Problems from droplets have
been mentioned by other authors. * The lack of humidity-produced
effect is an indication that the charge transfer process does not involve
surface electrolytic effects or that such effects are not important.
The present tests did not investigate possible effects from very low
humidity.
(4) Precharge effects. From published work, it is evident that the
separation of dust particles as in our dust generator results in electrical
20
charging of the particles. If the particles are charged before
Kunkel, J. Appl. Phys. 21, 820 (1950)
-26-
-------
1000 r-
M
•4->
•H
£
100
Q>
bfl
ft
10
Figure 5. Charge vs. Mass for Aluminum
Plot of total charge from the IKOR AQM vs gravimetric
mass of aluminum sample. The lines are drawn at 45°.
Data points are numbered sequentially.
I I
10
100
500
Mass, Mg,
-27-
-------
encountering the IKSOR probe, this might affect the response. We investi-
gated this possible effect by using a radioactive source to irradiate the
aerosol stream emerging from the dust generator. This should reduce the
charge through the interaction between charged gaseous ions and the
21
particles. A 1.2mCi Cd-109 source was placed within a centimeter
of the 6 mm air tube, which had a thin Al foil window to pass the
radiation. It is estimated that 2-5R/hr was incident on the aerosol
stream. The ratio of the IKOR current for aluminum oxide dust with
and without radiation was 0.97 ± 0.28. It is concluded that no effect
was produced.
This finding is reinforced by the observed lack of effect of humidity.
If significant precharging effects were present, one would expect humidity
to have an appreciable influence on it.
(5) Particle size effects. Direct observation of particle size effects
would require the measurement of the complete particle size distributions
of the aerosols actually sampled. Since only the > lym portion of the
distribution was measured in the present experiments, the information on
particle size effects is more indirect. However, there are several kinds
of evidence which are persuasive as to a surprising lack of dependence of
the IKOR AQM response to particle size. Our experimental evidence against
dependence on particle size includes:
21D. Keefe, P.J. Nolan, and T. Rich, Proc. R. Irish Acad. 60, 27 (1959)
-28-
-------
(a) Many samples of materials were used with various quoted particle
sizes. There were obvious variations as judged by eye. The
polydisperse Al and a sized (0.5-6 micron) Al sample gave closely
similar results for the sensitivity in yC/g. One glass bead sample
quoted as 0.5-3 micron diameter and another 3-8 micron diameter
yielded sensitivities of 0.46 ± 0.01 and 0.41 ± 0.03 yC/g.
(b) The IKOR AQM flow rate was varied to investigate the effect on
sensitivity. A variation of flow rate would affect the probability
of impaction on the probe. Also, the charging process could itself
be velocity dependent. Both of these effects involve particle size.
Over a variation of a factor of 1.7 in flow rate, the sensitivity
measured gravimetrically changed by less than 25% (standard deviation)
for Al203, and by less than 18% for Al.
(c) A crude test for response to very small particles was made with
cigarette smoke from filter cigarettes. The IKOR AQM barely
responded to levels which were off scale for the optical analyzer.
This shows there is an effective cutoff in the response to very small
particles, no doubt due to the fact that they tend to follow the
stream lines and hence fail to impact on the probe. This test shows
that the present instrument does not respond inordinately to very
small particles; in fact, for sufficiently small particles it does
not respond at all.
-29-
-------
In summary, these tests are not sufficient to rule out a dependence on
particle size, but they do show that such a dependence is not pronounced
above the cutoff. In Appendix I, it is shown that the theory of Cheng
and Soo predicts the charging to be dependent on particle size. The
overall response also depends on the probability of impaction on the
probe. This depends on the details of the instrument design. Evidently
the IKOR AQM is well designed in this respect.
(6) Surface condition of the probe. Direct evidence was obtained that
the condition of the surface of the probe of the IKOR instrument has a
very important effect on the sensitivity. It was mentioned above that
there is sometimes a sensitizing effect at the beginning of a run. It
was also mentioned that the sensitivity changed during the measurements
with Al dust. On July 2, 1974, a series of sensitivity runs was commenced
with various substances, beginning with A^C^. On July 30, upon returning
to A1203, it was found that the sensitivity had increased by a factor of 5.
The sensitivity for Al increased by almost a factor of 10 from July 10 to
August 1.
Between runs with different substances, the apparatus was thoroughly
cleaned. The probe was wiped clean with a dry cloth and replaced without
touching with hands. The probe has a polished surface with less than a
mirror finish. Some deep scratches can be seen. After a run, no dust
could be seen on the forward surface of the probe even under close
scrutiny. A light coating could be seen on the tail surface where the
-30-
-------
diameter Is reduced, probably owing to the slower stream velocity there.
The reassembled apparatus was then run with clean air until no indication
could be seen on the most sensitive scale.
Following the last measurement with Al dust which indicated increased
sensitivity, the probe was cleaned following the instructions suggested
by the manufacturer. This consisted of scrubbing the probe with an
abrasive cleaner (Ajax) and a nylon brush. This was followed by a
washing with detergent and a thorough rinse. Surprisingly, the result
was a lower sensitivity for Al, almost to the first value obtained on
July 10.
There would seem to be two possible explanations of the surface sensitizing
effect. One would be a cleaning effect or exposure of fresh surface
produced by the bombardment by particles, analogous to sand-blasting.
The other possibility is that a coating of particles builds up on the
surface, altering the charging conditions, since particles would then
encounter other particles rather than the surface itself. At present
the evidence is insufficient to allow one to decide between these two
possible explanations. The author leans slightly towards the sand-
blasting hypothesis.
(7) Sensitivity for various substances. A variety of materials were
sampled to investigate the sensitivity of the IKOR instrument. The
samples were chosen to cover a range of electrical resistivity, particle
size, particle shape, and chemical composition. Many of the materials
are found in stack effluent.
-31-
-------
In Fig. 6, the measured sensitivities have been graphed versus the date
of the measurement. Because of the change of sensitivity discussed above
in (6) , the first measurements of Al^Os and Al have been arbitrarily
excluded, i.e., the data from July 13-August 1 is used. When these
data are arranged in the order of increasing sensitivity, it is found
that they are naturally grouped according to electrical resistivity
(see Table 2).
The value for fly ash is consistent with its principal constituents,
Si02, Al203> and iron oxide. The resistivity of the glass beads was
10 ohm- cm, many orders of magnitude less than the other insulators.
Molybdenum and tungsten have about 3.3 times higher resistivity than
copper. Both were described as nonspherical particle samples and
both are brittle metals. Thus, the area of contact for these particles
may be smaller, thus reducing the charging. The other metals were
described as spherical particle samples.
-32-
-------
Figure 6. Sensitivity vs Date
Plot of sensitivity measurements for various substances
vs the date of the measurement.
3.0
2.0
CO
M
>>
•r-l
- rf
U
*fH
(A
c 1.0
(11
W
C/l
*
-
•
-
Ej
c
"E
^^
0)
0)
-U
en
0)
(0 M
0) Ol
1-1 p-l p.
C* W fii
•H > 0
« r-< O
4J -H y^
00 CO \£J
0 ©
o
•-I
PQ
C CO CU
o -o ••-•
,O CO Q)
t-i fl) CO
10 (0 C
U Ol 01 3
to T3 T3 H
/VJ » -H -H ^\
V«/ C9 X X V/
rH O O
O -H -r-t
_^ O Q 01
"w & C 3
< O 3 0 "x
•H U O
^s C2 4
3 ^^^
C
O)
-------
TABLE 2
SENSITIVITY MEASUREMENTS
I. Insulators vC/g
Titanium dioxide 0.051 ± 0.004*
Red iron oxide 0.06 ±0.01
Silicon dioxide 0.09 ±0.02
Fly ash 0.148 ± 0.005
Aluminum oxide 0.20 ± 0.01
Glass beads 3-8y 0.41 ± 0.03
Glass beads 0.5-3y 0.46 ± 0.01
II. Intermediate conductors, semiconductors
Molybdenum** 0.51 ±0.04
Tungsten** 0.67 ± 0.13
Carbon black 0.68 ±0.09
III. Metallic conductors
Stainless steel 2.0 ±0.5
Silver 2.0 ±0.2
Copper 2.1 ±0.2
Aluminum 2.9 ±0.4
*standard deviation
**nonspherical metal particles
-34-
-------
Carbon black is a semiconductor with a resistivity intermediate to
insulators and metals.
The ranking according to electrical resistivity can be understood in
terms of the characteristic time for charge transport. For good in-
sulators, this time is large so that only the area in the vicinity of
the contact is involved in charge transfer. The area is larger for
semiconductors. For metallic conductors the charge can probably be
transported from the entire particle. The oxide layers on the metal
particles are presumably thin enough so that they can be penetrated
either by electron tunnelling or by conduction with a small time
constant. The dependence on electrical resistivity is further discussed
from a theoretical viewpoint in Appendix I.
These data provide a basis for predicting the sensitivity for other
substances, at least for dry dusts. It is likely that they will fall
within the range spanned by the table. The wide range in the sensi-
tivities (factor of 60) has some implications for the detection of
components of a mixture. A small amount of an insulator would be
difficult to detect in the presence of metals or even semiconductors.
The charges per particle implied by the data of Table 2 are not
unreasonable. However, we must regard the sensitivities listed as
lower limits until the question of the probe surface is cleared up.
-35-
-------
SECTION VI
REFERENCES
1. W. R. Harper, "Contact and Frictional Electrification", Oxford U.
Press, Oxford, (1967).
2. L. B. Loeb, "Static Electrification", Springer-Verlag, Berlin,
(1958).
3. A. Schutz, Staub 24. 359 (1964).
4. A. Schutz, Staub £6, 18 (1966).
5. A. Schutz, Staub-Reinhalt der Luft, Band 26 (1966) No. 5, Seite
198/201, 1-4.
6. R. Prochazka, Staub 24, 353 (1964).
7. R. Prochazka, Staub 26, 22 (1966).
8. L. Cheng and S. L. Soo, J. Appl. Phys. 41, 585 (1970).
9. H. Schnitzler, SchrReihe Ver. Wass.-Boden Lufthyg, Berlin-Dahlem,
V. 33, Stuttgart (1970).
10. IKOR, Inc., unpublished reports, and private comm. from A. H.
Gruber.
11. G. J. Sem, et al, "Instrumentation for Measurement of Particulate
Emissions from Combustion Sources", Vol. I: Particulate Mass,
Thermo-Systems, Inc. APTD-0733, (1971).
12. "Static Electrification, 1971", Proc. of the Third Conf. on static
electrification organized by the Static Electrification Group of
the Institute of Physics held in London, May 1971. Conf. Series,
No. 11, The Institute of Phys., London and Bristol.
13. "Electrostatics and Its Applications", A. D. Moore, Ed., John Wiley
& Sons, N.Y., (1973).
14. S. L. Soo, "Dynamics of Charged Suspensions", Topics in Current
Aerosol Research, Vol. 2, International Reviews in Aerosol Physics
and Chemistry, Pergamon Press Ltd., Oxford, (1971).
15. S. YA. Kisler, Mekh. Avtomat. Proizvod., 2£ (9): 27-28, (1972).
16. K. Min, B. T. Chao, and M.E. Wyman, Rev. Sci. Instr. 34_, 529 (1963).
-36-
-------
17. D. H. Grindell, Proc. Inst. Elec. Engrs. (London) 107A; 353-365,
(Jan. 1960).
18. W. Coenen, Staub 2T_y 32 (1967).
19. T. Ito, H. Saito, and N. Furuya, Proc. Japan Soc. of Air Poll.,
13th, (1972), p. 247.
20. W. B. Kunkel, J. Appl. Phys. 2^, 820 (1950).
21. D. Keefe, P. J. Nolan, and T. Rich, Proc. R. Irish Acad. 60, 27
(1959). ~~
-37-
-------
SECTION VII
BIBLIOGRAPHY
I. Static and Contact Electricity
W.R. Harper, Proc. R. Soc. A205, 83 (1951). The Volta effect
as a cause of static electrification.
F.A. Vick, Suppl. No. 2, Brit. J. Appl. Phys. (London) (1952).
Theory of contact electrification.
"Static Electrification", L.B. Loeb, Springer-Verlag, Berlin,
1958. A classical reference, somewhat dated.
"The Electrical Behavior of Aerosols", K.T. Whitby and B.Y.H.
Lui, Chap. 3, "Aerosol Science", C.N. Davies, Ed., Acad. Press,
N.Y., (1966).
"Contact and Frictional Electrification", W.R. Harper, Oxford,
1967. The best single source of information on static electri-
city.
"Static Electrification", Proceedings of the Conference organ-
ized by the Institute of Physics and the Physical Society,
Static Electrification Group, London, May 1967. Inst. of Phys.
and Phys. Soc. Conference Series Number 4.
A. Schutz, Staub 27_, (12) 24 (1967). Electrical charging of
aerosols.
"Fluid Dynamics of Multiphase Systems", S.L. Soo, Blaisdell
Publ. Co., Waltham, Mass. (1967).
"Electrostatics", A.D. Moore, Doubleday & Co., Inc., Garden City,
N.Y., 1968. An entertaining account, particularly of electro-
static machines.
L. Cheng and S.L. Soo, J. Appl. Phys. 41. 585 (1970). Charging
of dust particles by Impact.
"Static Electrification, 1971", Proceedings of the Third Con-
ference on Static Electrification organized by the Static
Electrification Group of the Institute of Physics held in
London, May 1971. Conference Series, Number 11, The Institute
of Physics, London and Bristol. These proceedings and the
1967 proceedings listed below constitute excellent reviews of
the status of the field.
"Dynamics of Charged Suspensions", S.L. Soo. Topics in Current
Aerosol Research, Vol. 2, International Reviews in Aerosol
Physics and Chemistry, Pergamon Press Ltd., Oxford, 1971.
"Electrostatics and Its Applications", A.D. Moore, Ed., John Wiley
and Sons, N.Y., 1973. A very recent review of the field by
authoritative authors.
-38-
-------
II. Experimental Work and Contact Electricity Monitors
R.E. Volrath, Phys. Rev. 42, 298 (1932). Development of a
high voltage generator reaching 260 kV using blown dust.
W. B. Kunkel, J. Appl. Phys. 2^, 820 (1950), "The Static
Electrification of Dust Particles on Dispersion into a Cloud".
J.A. Medley, Br. J. Appl. Phys. Supp 2, S28 (1953a). Investi-
gated the effect of applied electric field on contact potential.
D. Keefe, P.T. Nolan and T. Rich, Proc. R. Irish Acad. 60_ (4),
27 (1959), Charge equilibrium in aerosols according to
Boltzmann's Law.
D.H. Grindell, Proc. Inst. Elec. Engrs. (London) IOTA: 353-
365, Jan. 1960. Particles charged in a high-voltage corona
were collected on an electrode.
K. Min, B.T. Chao, and M.E. Wyman, Rev. Sci. Instr. 34,
529 (1963). Measurement of charge on single particles in solid
gas suspension flow.
R. Prochazka, Staub 24, 353 (1964). Data on the mass cor-
relation of the Konitest current, early Konitest reference.
A. Schutz, Staub 24, 359 (1964). New arrangement for dust
measurement.
A. Schutz, Staub 26, 18 (1966). Principal reference for the
measurement by Schutz.
A. Schutz, Staub-Reinhalt der Luft, Band 26 (1966) No. 5,
Seite 198/201, 1-4, Registrierendes kontactelektrisches
Staubmessgerat mit logarithmischer Anzeige.
R. Prochazka, Staub 26, 22 (1966). Recording dust measure-
ment with the Konitest.
W. Coenen, Staub 2_7_, 32 (1967). A battery-powered device
charging particles by causing them to bounce between high-
voltage plates.
E.E. Weaver and Clyde Orr, Jr., "Atmospheric Contamination
and Triboelectrification", Georgia Inst. of Tech., Atlanta,
GIT-B-366, APTD-0629, 1 June 1968—31 May 1970. Investigated
triboelectrification between various substrates and platinum
in the presence of various gases and vapors.
W.J. Megaw and A.C. Wells, Nature 224. 689 (1969). "Pro-
duction of Monodisperse Submicron Aerosols of Which Each
Particle Carries a Specified Number of Electronic Charges".
-39-
-------
II. Experimental Work, and Contact Electricity Monitors (Continued)
J. Kolar, Tech, Ueberwach (Duesseldorf) 10 (6): 188-190,
June (1969). The functioning of a Konitest in heating plant.
N.A. Fuchs and F.I. Murashkevich, Staub 30. 1 (1970). Dust
generator with automatic feed and method for reducing particle
charging.
G.J. Sem, et al, "Instrumentation for Measurement of Particu-
late Emissions from Combustion Sources, Vol. I: Particulate
Mass, Thermo-Systems, Inc. Report No. APTD-0733, 1971. In-
cludes a critical discussion of the Konitest.
J.R. Melcher and K.S. Sachar, "Electrical Induction of
Particulate Agglomeration", Mass. Depart, of Tech., Cambridge,
APTD-0869, 10 Aug, 1971.
T. Ito, H. Saito, and N. Furuya, Proc. Symp. Japan Soc. Air
Pollution, 13th, 1972, p. 72. Continuous Dust Collection
Measurement Konitest. (in Japanese)
S. YA. Kisler, Mekh. Avtomat. Proizvod., 26_ (9): 27-28, 1972.
Monitoring of Concentration of Disperse Phase of Aerosol
Flows by Electric Contact Method. (in Russian)
A.K. Kamra, J. Appl. Phys. 44, 125 (1973). "Experimental
Study of the Electrification Produced by Dispersion of Dust
into the Air".
-40-
-------
SECTION VIII
APPENDIX
THEORY OF CONTACT CHARGING OF AEROSOL PARTICLES
Review of the Theory
The theory developed by Cheng and Soo^8' involves the transfer of
charge during a collision between two elastic spheres of radii a, and
a, . During the impact there will be a mean area of contact A^ last-
ing for At, both of these quantities depending on the kinematics of
the collision and the elastic properties of the materials. While the
spheres are in contact, the current density will be proportional to
the voltage difference between the two spheres. If the current is
integrated over the period of impact, the charge transferred is
obtained: ^ = _^f_ ^ . cpl} p x . ^orA^
where C,,2 are the electrical capacities of the spheres
<)>!,2 are the work functions of the materials
Ci+C2
and a = A2lh2l •
h2i, the charge transfer coefficient, is given by
CTl CT2
di
-------
The transfer lengths are defined by the equations
CTi CT2
where J2, is the current density and Vc is the potential at the point
of contact. Thus dj_ and da are the effective distances in the material
over which current is transported.
Now Cheng and Soo derive the following expressions for A21 and At
rt aj,a2
- - ' a
&! + aa i
2.9*+
-
i COS 9 "i
21
I* nns O / ~J-~*= 1 / ~1™2 \ ~ fv.j.V_
21,
„ u cos 9 . (ki+k
v
(note: ce is distinct from <* )
i
with r* the ratio of the rebound speed to the incoming speed
,, is the velocity of approach
0 is the angle of impact
nij 2 are the masses of the spheres
)
an<^ k, 2 are elast^c parameters given by
k = *
TT E
and x^ is Poisson's Ratio, E the modulus of elasticity.
Application to the Present Problem
It is instructive to evaluate numerically some of the above quantities
using reasonable values of the parameters. Take sphere 1 to be the aerosol
-42-
-------
particle and sphere 2 to be the probe. Then
Using k, = k, = 3.10~13 cm2 /dyne
0=0
p~! = 8g/cm3 (density)
aj = 10 cm
f\
AU2, = 1.5-10 cm/s (typical for the IKOR AQM)
r* = 1
we find that
3 • 10 cm
-10,
At - 6-10
'= Tra^j = l*10~i:Lcm2
The short duration of the impact implies that the theory applies only
f\
to metals or good semiconductors (p Z 10 ohm* cm) since only then will
there be appreciable current flow. The area of contact has a diameter
roughly 1/20 of the particle diameter.
Dependence of Q2i on the Parameters
The coefficient ex depends on,h21 . If we take di"=3 and d^ = a2 , the
remaining variable is o^ , since (T2 is fixed (metal probe).
When P,< 10 ohm-cm, l-e"aAt = !>
and
-43-
-------
This equation results because the resistivity is small enough to allow
complete charge transfer within the contact time At.
If pl > 10 ohm-cm, Q2 j is reduced by the factor (1-e a ). For example,
when Pt = 10 ohm»cm, (l-e"Q^ |t) = 0.05 and only 5% of the charge is trans
ferred within At. The theory, therefore, applies only to metals or very
good semiconductors.
Q21 is the charge transferred per particle. The total charge transf-
erred to or from the probe is given by
QT * N, Q21
where N, , the number of aerosol particles, is given in terms of the total
mass of particles, MI , by
M
•* 3 _
— IT a p
3 1 Kl
_2
For PI < 10 ohm-cm, QjX? ai • Thus the theory predicts that the total
charge transferred to the probe is inversely proportional to the par-
ticle radius squared. Also, for Pj < 10 ohm* on, the charge transferred
is predicted to be independent of the velocity of the particles rela-
tive to the probe.
Magnitude of the Charge Transferred
Consider metal particles with a radius of O.OSum. Assume that cp2-cp
1 volt. Then QJI = 10~17 C/particle. Furthermore, QT = 2000
-44-
-------
This is three orders of magnitude greater than observed (See Table 2).
The calculated charge could be brought into agreement with experiment
by assuming that the particles had an effective resistivity of 5.10
ohm-cm which is in the semiconductor range.
The oxide layer on the particle surfaces could easily increase the
effective resistance, not to mention contaminant layers on the surfaces
of the particles as well as on the probe surface.
Summary of the Theory
While the theory is not quantitative, the model may still be useful
in understanding the dependence of the charge on the physical para-
meters. However, the theory cannot as yet be regarded as verified by
experiment.
The above theory applies only to metals or good semiconductors. For
materials with f> > 10 ohm-cm, which includes most semiconductors and
all insulators the charge transferred is that already on the surface
in the area of contact. In the absence of detailed knowledge of the
physical processes involved, it is not possible to carry out the
analysis for the charging of insulating materials.
-45-
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO.
EPA-650/2-75-043
3. RECIPIENT'S ACCESSIOWNO.
4 TITLE AND SUBTITLE
Investigation of Participate Matter
Monitoring Using Contact Electrification
5. REPORT DATE
Feb ruary 1975
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Dr. Walter John
B. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG MMIZATION NAME AND ADDRESS
California State College, Stanislaus
Turlock, CA 95380
10. PROGRAM ELEMENT NO.
1AA010
11. CONTRACT/GRANT NO.
802726
12. SPONSORING AGENCY NAME AND ADDRESS
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park
North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The theory of the charging of aerosol particles by contact electricity is reviewed,
as well as the development of monitors for particulate matter "based on this
principle. Data on the performance of these monitors is scarce and sometimes
contradictory. In the present work, laboratory tests were carried out with a
variety of test dusts. The results show that the dynamic response of the contact
electricity monitor tracks well with that of an optical detector. The total charge
correlates well with the gravimetric mass. Humidity or precharge on the particles
did not influence the detector. Some evidence was obtained showing that particle
size does not have an important effect on the response, although there is a cutoff
for very small particles. Results for the sensitivity of the instrument can be
grouped according to the electrical resistivity of the material. It is found that
the condition of the surface of the Inconel probe has a major effect on the
sensitivity. Additional work is necessary on this aspect of the detector. The
theory of Cheng and Soo for the charging of metal particles is discussed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED-TERMS
COS AT I Field/Group
Monitors, Electrostatic charge, Dust,
Particles, Fly ash, Konimeters, Sampling
probes, Electrostatic measurement methods,
Electric charge, Triboelectric charging
Pollution measurement
methods, Continuous
monitoring methods,
Particulate sampling,
Contact electricity
1302, 2003,
1711, 2102
18. DISTRIBUTION STATEMENT
Release Unlimited
. SECURITY CLASS (ThisReport}
Unclassified
21. NO. OF PAGES
52
20 SECURITY CLASS fThis page)
Unclassified
22 PRICE
EPA Form 2220-1 (9-73)
46
------- |