TOFI
ORNLP
2436
f
-
-
.
.
'.
1
TT-
.
OS
2
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963
i
OPNL-P.2436
Conf-660904-1
SLP 2 2 1966
CHARACTERIZATION OF RADIOACTIVE PARTICULATE AEROSOLS BY THE FIBKOUS FII:TER
ANALYZER *
M. D. Silverman, J. Truitt, W. E. Browning, Jr., L. F. Franzen, and R. E. Adams
Reactor Chemistry Division, Oak Ridge National Laboratoryj: Oak Ridgear Tennessee
ABSTRACT
H.C. $ 1.00; MN.SO
The fibrous filter analyzer (FFA) has been developed for measuring
the characteristics of radioactive aerosols in terms of their response to
filtration processes by determining their distribution vs depth in a filter
under carefully controlled conditions. Moisture was shown to have no
significant effect on the performance of the FFA, although the test aerosol
itself was affected by moisture. The filtration efficiency data obtained
agreed well with the theoretical treatment of filtration developed by
Torgeson. The analyzer was calibrated against particles having diameters
from 150 to 1500 Angstroms, measured by electron microscopy.
INTRODUCTION
The fibrous filter analyzer (FFA) characterizes radioactive aerosols
by determining their distribution as a function of depth under carefully
controlled conditions. Although it was developed originally for measuring
the amounts of fission products associated with aerosols resulting from
simulated reactor accidents, the FFA can also be employed for monitoring
radioactive aerosols associated with normal operation of reactors or
processing plants. These measurements are needed because different kinds
of aerosols differ in their behavior during dispersal, transport, deposition,
and filtration; and meaningful interpretation of their behavior or reliable
design of filtration systems depends upon the ability to distinguish between
fission products associated with different kinds of aerosols. There is a
need for both field devices and for laboratory devices for reference
measurements. The field devices are needed for application in accident-
simulation experiments or in monitoring effluents from reactors or processing
plants. The laboratory reference measurements are needed for calibrating
the relatively simple field devices against more fundamental standards.

2
-41;
IN NUCLEAR SCIENCE ABSTRACTS
RELEASED FOR ANNOUNCEMENT
:
-
-
. -- *
M
TEL
-
TP
The ideal field characterization device should meet several require-
ments. It must discriminate, unambiguously, between the various forms of
aerosols carrying fission procucts. This discrimination between forms
should be accomplished by measurement of processes that are similar to
processes important in the behavior of fission products under the conditions
of interest so that application of the data will not involve extensive
extrapolation from the conditions of measurement. Analytical techniques
for the measurement of the materials deposited by these processes should
involve radioactivity so that only those processes that affect fission
product behavior will be emphasized. From a practical standpoint, besides
determining the amount of radioactivity involved with various aerosols, the
IS
.
1
14
*A@Jeucis aponied by the U. 3. Atino. Duero Comuo:lon umier cuartsect vitis
tie Union Carbue Couporation.

-
-.
-
.
-
+
- -
-
3
1.
device should be simple, reliable, inexpensive with respect to both
construction and analysis, and require a minimum of electrical and
mechanical accessories. Its performance should not be adversely
affected under testing conditions, particularly by moisture. Such a
device usually takes the form of an assembly that is passive in the
· sense that it simply sits in place but is dynamic in the sense that it
makes its measurement using the motion of gas passing through it to bring
into play processes like diffusion and inertial impaction. The information
developed by such a device is stored in it in the form of distribution of
radioactivity until the device is retrieved and radioassayed. This
distribution is then interpreted in terms of the amounts of fission
products having various characteristics that affect their behavior.
.
.
.
-
.
=
:
.
:
-
-
-
.
Some characterization devices have been developed that meet most of
these requirements to a greater or lesser degree. Some popular methods of
measurement of aerosols are quite useful but fail to satisfy these criteria
in important respects. For example, electron microscopy and some light .
scattering devices give explicit information about the sizes of particles
which are opaque to electrons or light, but make no distinction between
those which carry radioactivity and those which do not. Condensation
nuclei detectors of the Pollack or Aitken type select particles according
to whether they nucleate condensation rather than according to their
radioactivity. Certain electrostatic analyzers measure particles by the
charge they carry. There are a number of aerosol analyzers which measure
aerosol particles by their radioactivity and thus measure the ones
important in the behavior of radioactive materials. The cascade impactor
(1) and the Andersen sampler (2) both are suitable for measuring the
character of radioactive aerosols in terms of their behavior while
impinging on a flat surface. The Goetz aerosol. spectrometer (3) makes
similar measurements in a centrifugal field. Several investigators have
developed electrostatic classifiers suitable for detection using radio-
activity of the aerosol (4-8). "The FFA, in essence, is an "in situ"
analytical device which characterizes aerosols dynamically by particle
response to the major processes of filtration: diffusion, interception,
and inertial impaction.
--
-
The concept of the FFA originated from a note by Sisefsky (9) who
determined the penetration of "fall-out" material in a filter by radio-
assay of thin layers of filtering material peeled off with pressure-
sensitive cellophane tape. Although the technique yielded the desired
information, it was limited in accuracy because of the broad spectrum
of fiber sizes in the commercial filter employed and in the varying
thickness of the peeled-off layers. By contrast, the fibrous filter
analyzer is made from uniform diameter material formed into a layered
structure to facilitate separation of the fiber bed into discrete layers
for radioassay. Preiiminary results using this principle of aerosol
characterization were reported previously (10-12), especially with
reference to the diffusion flow regime. The present paper not only
supplies results under a much wider range of experimental conditions
including moisture but takes into consideration the mechanisms of inter-
ception and inertial impaction. "Preliminary calibration has been
LEGAL NOTICE
.
.
-
This report was propared us an account of Government sponsored work. Neither the nited
Statau, nor the Commission, nor any person acting on behalf of the Commission:
A. Makos may warranty or representation, exprouard or implied, with respect to the accu-
racy, completeness, or usefulness of the faformation contained in this report, or that the wine
of way information, appuntua, method, or process di lorod la the report may not Infringe
privately owned righto; or
B. Asnimas any liabilities with respoct to the use of, or for damages resulting from the
na of way laformation, apparatus, method, or proceu disclosed in this report,
As wood in the abov, "porno0 acthy on behalf of the Commission" tocludeo may on
ploys or contractor of the Couninslon, or employee of such contractor, to the extent that
much employw or contractor of the Commission, or employee of much contractor prepares,
denominatas, or provides accouto, way information purmuaat to No oployment or contract
with the Commission, or kjo employment with such contractor.
t

-3-
accomplished with reference to aerosol particle sizes obtained by
electron microscopy.
EXPERIMENTAL
The experimental arrangement was designed to facilitate separation of
the fiber bed into discrete layers for radioassay after exposure to the
aerosol. We chose fibers with a high degree of uniformity in order to
facilitate theoretical analysis. Raw Dacron-polyester staple fibers (99%
were 11.3 + 0.8 u) 3.8 cm long were separated into a uniform web by means
of a carding machine (13). Punched-out disks of this material, 3.8 cm in
diameter and 0.05 cm thick, were inserted in series between uniform photo-
etched stainless steel screens supported by cut-out metal washers, the
whole enclosed in a cylindrical polystyrene holder, and provided with
O-ring seals to give a compact filtering device (Fig. 3.). A test aerosol
containing 652n was produced by using a Tesla coil to generate a spark
between two pieces of zinc foil placed approximately 2 mm apart. Radio-
active ozn was chosen for the test aerosol because it possessed the
requisite properties of a strong y-emitter of long half-life which is
easily prepared by neutron irradiation and is also easily decontaminated
by aqueous chemical methods. A stream of air passed over the electrodes
and carried the aerosol through the system (Fig. 2) containing the
filters. Electron micrographs of samples of the aerosol collected on
membrane filters yielded information regarding the size of the particles.
Depending on the experimental conditions, aerosols have been prepared
over the size range 50 to 10,000A (0.05 to 1 m.).
Experiments were conducted over a wide velocity range, the linear flow
rạtes varying from 0.2 to 300 cm/sec. The moisture content of the inlet
air to the system was varied from < 1 to 100% relative humidity for
different experimerts. The data were analyzed by means of graphs (Fig. 3)
in which the log of the relative radioactivity collected per layer is
plotted against depth in the filter. Each layer contained from 2 to 8
x 109 of fiber per square centimeter of filter area, evaluated from disk
weights and. the density of the fiber. The abscissa in Fig. 3 has been
simplified to show filter depth in centimeters, but the very small non-
uniformity of the filter pads has been accounted for in plotting the data.
In the region of low flow where Brownian diffusion is the dominant
process for particle transport to the fiber, filtration efficiency
decreased with increasing velocity. In the high-flow region, where
inertial impaction is the principal process for particle transport, the
efficiency increased with increasing velocity. In the intermediate-flow
region, the interception range, where geometrical considerations of
particle size and fiber size are very important, the filtration efficiency
is largely independent of velocity. Thus, the velocity dependence of the
slope of these distribution curves identifies the mechanism of filtration,
and the magnitude of the slope indicates the intensity of the filtering
action.
.-..
-
-
•
•
REDUCTION OF DATA
RÁ
The intensity of the filtering action is a significant quality of an
aerosol, and it is useful to express this quality in terms that are
relatively independent of experimental conditions. The filtering action
of fiber mat is the cumulative effect of particle removal by individual
fibers. The collection efficiency of an individual fiber, n, is defined
as the ratio of the cross-sectional area of the aerosol stream from which
particles are removed to the projected area of the fiber in the direction
of flow and may be obtained from the slope of the distribution curve on
the semi-log graph. This slope may be substituted directly for (in N/N.)/L
in the expression derived by Langmuir (14) and confirmed by Davies (15)
and Chen (16) for the single-fiber efficiency:
n = - In Ñ
0 (1 - alam
T. -,
(1)
where No and N are the upstream and downstream particle concentrations,
respectively, a is the volume fraction occupied by the fibers, L is the
thickness of mat, and df the fiber diameter, all 'In consistent units.
The slopes of the distribution curves on a semilog plot such as Fig. 3 can
be shown to be identical to the expression (in N/N.)/L: the data plotted
is the amount of activity deposited in the fiber bed, that is the amount
removed from the gas stream, -an, per unit depth, dl. The logarithm of
(-aN/aL) is plotted vs L. When for an aerosol or for a single component
of an aerosol the plot yields a straight line, then
not intendaishironto mi
.
. d ln (-aN/aL)
L = - a,
dL
where -a is the slope of the plot. Integrating (2),
in (-aN/aL) = -aL + I
where I is the intercept at L = 0 on the plot.
Integrating (4)
Since the data yields information about only those species which
deposit in the filter, N - O as I + , and c = 0. For 1 = 0,
I
01 ರ
whence
(6)


2
-
5. !.
3
N
2
E P
and (in N/N.)/L = -a.
(7)
Comparison with (2) shows that the slope of the line on & semilog plot of
deposition may be substituted directly for (in N/N.)/L in Eq. (1). Thus
this expression, commonly know). as the Chen equation, has been used
extensively to calculate the single fiber efficiency, n, which is a
measure of the quality of an aerosol, i.e., how easily it can be trapped
in a fibrous filter. This information about the aerosol is useful per se,
but it also can be used to calculate the diameter of the aerosol particles.
OPTIMIZATION OF DESIGN OF A FIBROUS FILTER ANALYZER
The choice of materials, dimensions, and operating conditions for a
fibrous filter analyzer should be based on its application. We were
interested in aerosols which could carry radioactive fission products in
reactor accidents; simulated reactor accidents have yielded particles whose
diameters range from 50 to several thousand Angstroms. A computer study of
fundamental variables in filtration had indicated fiber diameter to be the
most important parameter. For particles in this size range the most
effective fibers in practical fibrous filters are those near the small
diameter end of the fiber size spectrum. The 11.3 micron diameter Dacron
fibers selected were the smallest ones available which were uniform in
diameter. A smaller diameter would be preferable. A material which would
resist higher temperatures would also be more desirable.
The packing density of the fibers must be controlled. Since fiber
filtration theory is based on the flow of the aerosol around isolated
fibers, low values of X, the volume fraction occupied by the fibers, are
desired. However, at the very low packing densities it is too difficult
to maintain reproducibility and stability. The data in Table 1 show how
the single fiber efficiency varied with a. For each gas velocity a value
of a of approximately 0.15 yielded an appreciably lower value of nithan
did the other values of a. However, for a's of around 0.04, the value of
n was relatively insensitive to a. For this reason, packing densities
of around 0.04 were adopted.
The full range of practical gas velocities has been explored
experimentally. For some particle size ranges expected low velocities
would be preferable; but for particles over about 1000 Angstroms in
diameter, the plot of n vs particle diameter is double-valued, as will be
noted in a later section; and if information in this range is needed, it
may be necessary to operate samplers at two different velocities.
EFFECTS OF MOISTURE
Some of the more important applications of aerosol analyzers are in
circumstances where moisture may be present. For example, following an
accident to a water-cooled reactor, fission products could be released to
the steam atmosphere resulting from evaporation of the coolant. It is
necessary to be able to measure aerosol materials under conditions
MAILI
-
'
C'
.35
.
T
---
LOVE
corresponding to such an accident. Accordingly, tests were carried out
to determine the possible effects of moisture on the performance of the
fibrous filter analyzer. Figures 4 and 5 show some of the data obtained.
The log of the amount of activity on each layer of the fibrous bed 18 .
plotted vs depth through the bed. The tests were carried out at room
temperature at relative humidities of less than 2% and at 4100%. There is
a very noticcable difference between the distribution curves obtained under
wet and dry conditions. Under dry conditions the initial steep slope is
more pronounced. Under wet conditions the overall reduction of activity
16 diminished. It is interesting to note, however, that the lower portion
or the curves have approximately the same slope. This means that the more
penetrating of the two species of particles is filtered with equal
efficiency under both wet and dry conditions. In other words, the inter-
action hetweer: this species of particle and the fibers is not affected by
moisture, and the fibrous filter analyzer 13 measuring particles
satisfactorily unaer moist conditions and dry conditions. The change in
the front end of the distribution curve under moist conditions can be
interpreted as an effect of the moisture on the aerosol itself. Evidently
an easily filtered species is more abundant under dry conditions and less
abundant under moist conditions. There are severel possible explanations
for this. The moisture may inhibit agglomeration or perhaps may affect
the formation or growth of particles which are easily filtered.
E
.PL
A series of experiments was conducted to measure the effect of humidity
on the filter collection efficiency of the more penetrating species. These
tests were carried out under two levels of relative humidity, less than 2
and 100%, over the velocity range 0.66 to 322 cm/sec. The single fiber
efficiencies, n, were calculated from the slopes of the lower portions of
the distribution curves and are presented in Table 2. Also tabulated in
Table 2 for each velocity are the ratios of the filter collection
efficiencies under wet vs dry conditions. The mean value of this ratio
for all velocities combined is 1.19 with a standard deviation of 0.36.
This indicates that the ratio did not differ significantly from 1 and
there was no significant effect of moisture on the filter collection
efficiency. This means that the fibrous filter analyzer gave the same
results for aerosol characteristics under both wet and dry conditions.
The analyzer indicated, however, that there was a difference in the
abundance of a less penetrating species in the aerosol.
THEORETICAL
$
The theoretical analysis of data obtained from fibrous filter analyzers
is of interest for a number of reasons. If the data can be correlated
with theory, greater credence can be placed in the conclusions, and the
theoretical analysis can be used for interpolation between calibration
points. In addition, the availability of data obtained with fibrous
material having uniform fiber diameter and for a large number of points
through the depth of the bed afford a unique opportunity to confront
theoretical expressions with experimental data.

2
The theoretical interpretation of fibrous filter analyzer data
uitained at relatively low linear velocities of 0.2 to 2.5 cra/sec has been
reported previously (10-12). It was found that in this range in the single
fiber collection efficiency, is proportional to 1/(velocity)1/2. This
relationship was interpreted to indicate that, in the low flow region,
diffusion is the primary mechanism of filtration.
Measurements were made of pressure drop through the fibrous material
at various flow rates and packing densities and some of these data are
presented in Fig. 6. The pressure drop data were used to compute the fiber
diameter using a calculation technique described by Whitby (17). The
effective fiber diameter was 10.8 microns which compared favorably with the
21.3 micron value observed microscopically. This indicates that the flow
of air through the fibrous bed is indeed of the type assumed in the
theoretical treatment which follows.
Theoretical calculations were made of fiber efficiencies for the
conditions of the fibrous filter analyzer over the entire range of
velocities and taking into consideration not only diffusion, but also the
processes of intercepti in and inertial impaction. The theoretical treat-
ment developed by Torgeson (18) was used. This theory is an adaptation of
Davies' (15) interception and impaction theory, with a new combined
interception and diffusion theory. The Torgeson treatment was selected by
Whitby (17) as that which agreed most closely with a variety of
experimental data. In the Torgeson theory,
My = 0.75 Mimp + FG nao
na was evaluated by
na = 0.7 (Re)0.4
(Pe)0.6
where Pe is the Peclet number,
Pe = vo df
C
D Bm
v. is the linear velocity and Dom is the diffusion coefficient,
where c is the Cunningham slip correction, k is the Boltzman constant,
T is the temperature in OK, u is the viscosity of air, and dp is the
particle size.
The (CpRe/2) is the drag coefficient which is
-
-
--
-
..
PS
M
C Re
que - Anime
:
-
--
E
G is interception correction factor, F is inertial correction factor given
-
--
by
-
-
-.
p=1 +0.025 , 0.6 P20.6 610.5 + 9.48 (Steno." (Fe)-0.6).
is the impaction parameter.
Theoretical non were calculated for a series of particle sizes
ranging from 67 to 13,400 Angstroms and for velocities ranging from 0.66
to 322 cm/sec. A three-dimensional plot of these DT vs velocity and
particle size is shown in Fig. 7. The region of the curves in which the
calibration points lie is almost, entirely near the bottom of the valley.
This is to be expected because of the method of selection of particles.
Only those particles which are more penetrating were available in that
portion of the fibrous bed where the calibration was carried out. By the
same token, these are the particles which are most important since they
are the ones which set a limit on filter efficiency. Therefore, they are
the ones we most want to measure. The sensitivity of this method of
calibration is limited by the spread of particle sizes observed by the
electron microscope. However, within these limits the experimental points
agree fairly well with the theoretical treatment. With this information,
it is possible to interpret fibrous filter analyzer data in terms of
particle diameter for particles from 150 to about 1500 Angstroms in
diameter.
CALIBRATION
The fibrous filter analyzer yields information which is valuable
without reference to absolute particle size; 15 distinguishes aerosol
particles according to their behavior and gives a quantitative measure of
this important quality. It would be desirable, however, to establish
a relationship between this measure of aerosol characteristic and other
measures, and the usual practice is to describe aerosol particles in
terms of their diameter. For this reason, comparison measurements are
being made to calibrate the fibrous filter analyzer against particles of
known size. Some of the more obvious procedures of generating particles
of known size labelled with radioactivity and passing them through a
fibrous filter analyzer have not yet been completed. One set of
measurements, however, has been made which permits this calibration
comparison to be made.
The heterodisperse aerosol previously described, labelled with ºzn
was passed through a fibrous filter analyzer. The fibrous bed eliminated
16
2.
,
.
all but the most penetrating aerosol species. That the separation was
nearly complete was indicated by the linearity of the distribution plots,
near the exit end of the fibrous fllter bed. The value of the single
fiber collection efficiency, 1), was evaluated for this linear portion of
the curve. The dimensions of the particles emerging from the fibrous bed
were measured by trapping them on a carbon-coated membrane: filter and
measuring them by electron microscopy. The results of these two
measurements are presented in Table 3. The results have also been plotted
on the three-dimensional graph in Fig. 7.
CONCLUSIONS
The fibrous filter analyzer (FFA) has been developed for measuring
the characteristics of radioactive aerosols in terms or their response to
filtration processes by determining their distribution vs depth in a filter
under carefully controlled conditions. Moisture was shown to have no
significant effect on the performance of the FFA, although the test
aerosol itself was affected by moisture. The filtration efficiency data
obtained agreed well with the theoretical treatment of filtration developed
by Torgeson. The analyzer was calibrated against particles having
diameters from 150 to 1500 Angstroms, measured by electron microscopy.
*
-
*5E
REFERENCES
2
-
1.
-3 *
. ,5
K. R. May, J. Sci. Instrum 22, 187 (1945).
A. A. Andersen, "New Sampler for the Collection, Sizing, and
Enumeration of Viable Airborne Particles," J. Bacteriology 76,
471-84 (1958).
2.
.
A
.
.. .
-
...
,
3.
A. Goetz and 0. Preining, "Bestimmung der Gröbenverteilung cines
Aerosols Mittels des Goetz'schen Aerosolspektrometers," Acta Phys.
Austriaca 14, 293-301 (1961).
M. H. Wilkening, "Natural Radioactivity as a Tracer in the Sorting
of Aerosols According to Mobility," Rev. Sci. Instr. 23(1), 13
(1952).
5.
V. Mohnen and K. Stierstadt, "Die Verteilung der Naturlichen Radio-
aktivitat auf das Grobenspektrum des Naturlichen Aerosols,"
Zeitschrift fur Physik 173, 273-293 (1963).
6
M. Gass, M. Saran and K. Stierstadt, "Das Grobenspektrum des
Naturlichen Atmospharischen Aerosols," Zeitschrift fur Physik 185,
269-277 (1965).
7.
R. S. Stuart, "The Analysis, Optimization, and Design of an Airborne
Radiation Sampling Electrostatic Precipitator, " Thesis, Master of
Science, School of Engineering of the Institute of Technology Air
University, May 1962.
K. T. Whitby and C. M. Peterson, "Electrical Neutralization and
Particle Size Measurement of Dye Aerosols," I and EC Fundamentals
4, 66 (February 1965).
9.
:
J. Sisefsky, "A Method for Determination of Particle Penetration
Depths in a Filter," Nature 182, 1438 (1.958).
10. M. D. Silverman and W. E. Browning, Jr., "Fibrous Filters as Particle
Size Analyzers," Science 143, 572-3 (1964).
11. W. E. Browning, Jr., R. D. Ackley, and M. D. Silverman,
"Characterization of Gas-borne Fission Products," pp. 155-162,
Eighth AEC Air Cleaning Conference held at Oak Ridge National
Laboratory, October 22-25, 1963, TID-7677.
12.
M. D, Silverman and W. E. Browning, Jr., "Fibrous Filters as Particle
Analyzers," Trans. Am. Nucl. Soc. 6(2) 401 (November 1963).
We thank the Cotton Spinning Laboratory of the University of
Tennessee, Knoxville, Tennessee, for carding the Dacron fiber.
14.
I. Langmuir, "Filtration of Aerosols and the Development of Filter
Materials," OSRD Report ivo. 865 (1942).
E
n
.
-11-
15. C. N. Davies, Proc. Inst. Mech. Engrs. London 1B, 185 (1952).
16. C. Chen, "Filtration of Aerosols by Fibrous Media," Chem. Rev. 55,
595 (1955).
17. K. T. Whitby, "Calculation of the Clean Fractional Efficiency of Low
Media Density Filters," ASHRAE Journal , pp. 56-65 (September 1965).
18. W. L. Torgeson, "The Theoretical Collection Efficiency of Fibrous
Filters due to the Combined Effects of Inertia, Diffusion, and
Interception," Paper No. J-1057, Applied Science Division, Litton
Systems, Inc., St. Paul, Minnesota.
FIGURE CAPTIONS
1.
CRNL-PHOTO-61430, Fibrous Filter Analyzer.
2.
ORNL-LR-DWG-76776, Apparatus for Characterization of Aerosols
Using Fibrous Filters.
3.
ORNL-DWG-63-521, Distribution of Zinc Activity vs Depth in Filter.
ORNL-DWG-66-591, Response of the Fibrous Filter Analyzer Under Wet
and Dry Conditions at a Linear Velocity of 3 fpm.
ORNL-DWG-66-592, Response of the Fibrous Filter Analyzer Under Wet
and Dry Conditions at a Linear Velocity of 11.5 1pm.
ORNL-DWG-66-6520, The Effect of Packing Density on Pressure Drop for
Varying Velocity.
-
1-
-
1.- *
ESTER
-
TV

Tiiu
-12-
Table I.
The Effect of Packing Denghty on the Single Fiber Efficiency
\
(cm/sec)
:
Filter Volume
Fraction Solids
Experimentat: Single
Fiber Efficiency
0.66
.oko8
ဝ.ook
0.154
0.0717
0.0o1
0.00
1
လို့ ထထထ ထထထ
o.o18
0.0778
0.131
0.028
0.025
0.0106
0.0375
0.0779
0.159
0.0355
0.02.1
O.O172
4.9
4.9
0.085
0.0760
O.155
0.036
0.015
0.0384
0.0769
0.154
0.15,
0.0261
0.0200
O.O1.1
O.O119
7.9
7.9
14.8
14.8
1.8
0.0368
0.0703
0.18
0.0597
0.ok12
O.O198
29.6
2.o55
0.0852
0.158
0.0571
0.0102
O.O124
29.6
5.
5.4
5.4
o.olx72
0.0897
o.15
0.035
၀.၀၀89
0.0016
.
:
1
f
(
i

*
*
-
1 -
-
•
-
-
-
7
"_
Y
-13-
Table II. The Effect of Humidity on the Single Fiber Efficiency
so
Relative Humidity
Ratio W/D
0.66
0.0776
0.66
100
0.121
1.56
2.0
2.0
100
0.0471
0.0476
1.01
0.0395
0.0476
100
10.8
10.8
0.0286
0.0334
1.16
15.1
0.0248
0.0332
15.1
100
1.34
oo
0.0255
0.0288
100
1.13
100
0.0421
0.0379
0.90
0.0587
0.124
100
2.11
EF mm
100
0.0311
0.0308
0.99
45.5
45.5
0.0785
0.0654
100
0.83
322
100
0.0389
0.0340
0.87
Average 1.19
ir
-
---
tas
-14-
2-13
--
2.
.
.
Table III. Comparison of Experimentally and Theoretically Determined
Particle Diameters
3
.
.
.
.&.
S
v
Relative Humidity
n
Experimental
Single Fiber
Efficiency
Particle Diameter Particle Diameter
Experimental Calculated
EM Photo, AO . Aº
I
DAV
2.0
0.0413
0.0466
2000
2895
1700
1600
2.0
0.0395
0.0476
1020
2200
710
10.8
10.8
0.0288
0.0332
300-600
43 - 410
15.1
0.0249
0.0332
680
150-1000
3000-5000
600
21.0
21.0
0.0255
0.0284
400-800
400-600
-
.
.
I
35.7
35.7
0.103
0.0587
70-500
1450-3600
LA
0.0311
0.0308
500-2000
500-5000
2
.
ta
0.0785
0.0654
300-3560
740-3850
. 6:19Mi
322
322
0:0389
0.0684
200-700
300-600
ic
...
F2
2
PESANTES
•
:

مهم مهمه .
. ..
. ..
. .. ..
. :: س.م
}
08CLASSIFICA
OAK RIDGE NATIOXAL LABORATORY
.ه واه . ::
.:C
..
...من... م
..
.. .. ... ..م
..
ه .همه سه ماه سه
ن
... .جب مطمن
.
.. ..
. . .
. . .
-
نننننننننمننننمنمسسمند
ممم ... ...
.. ... .. . . . .
تعمید
کنسليه
لنننغلضممضة... همممممسقط
.مي
. ..
. . .. ..
. .
.
.
.
. .
-- -- --
... ...
.
.
هسننممنمحمدی
،نعه مما دفعدن، مصممه دسسسسسمدسنسنسلخ
نما
.
Frigi. Fihrow. Filter Analygi.
.
:


اساس
har
To be replaced by
نجاست
اما مهم است
. .
.. جا
منہ اس
.
ا ، ا
سمی
/7
»
c
.
UNCLASSIFIED
ORAL-LR-OWO 76776

FLOWMETER
HOLDER FOR FIBER FILTERS
FILTERED AIR
PARTICLE
GENERATOR
ABSOLUTE
FILTER
HOOD
EXHAUST
PUMP
Ze FOIL SPARK GAP
FLOWMETER
TESLA
COIL
Fig. Mateo Apparatus for Characterizoion of Aerosols Using Fibrous Filters.
m
.-.
-
refor
<25
UNCLASSIFIED
ORNL - DWG 63-521

AIR VELOCITY
88 Il/min
- 2110 cmicm2 --
.
FRACTION OF ACTIMTY (RELATIVE COMCENTRATION/FILTER DEPTH)'
Q002
2001
-
0.0005
men det er interes
0.0002
0.0001
0.00005
FILTER DEPTH (FIBER LENGTH / FILTER AREA, cm/cm2)
Fig. Distribu!ion of Zinc Activity vs Depth in Filter.
ierten en met een
1.
AL-KO
ORNL-DWG 66-591

-SATURATED AIR
ننننننننننننننننننننننننننننصمتنه
RATED AIR
NIN, RATIO OF DOWNSTREAM TO UPSTREAM PARTICLE CONCENTRATIONS
DRY AIR
MORE PENETRATING
COMPONENTS
.
.
.
LESS PENETRATING
COMPONENTS
vanhastaminmaternitanici na sedeniew direnean erama
.........
2
-DRY AIR
10 a
5
·
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
3.5
FILTER NUMBER
-
.......
1
RESPONSE OF THE FIBROUS FILTER ANALYZER UNDER WET AND DRY
CONDITIONS AT A LINEAR VELOCITY OF 3 fpm
-
-
-
ORNL-DWG 66-592

SATURATED AIR
DRY AIR
NINO, RATIO OF DOWNSTREAM TO UPSTREAM PARTICLE CONCENTRATIONS
ATURATED AIR
MORE PENETRATING
COMPONENTS
LESS PENETRATING
COMPONENTS
•
DRY AIR
0
1
2
3
4
5
9
10
11
12
13
14
6 7 8
FILTER NUMBER
RESPONSE OF THE FIBROUS FILTER ANALYZER UNDER WET AND DRY
CONDITIONS AT A LINEAR VELOCITY OF 11.5 fpm
ple. 5

.
*****
*
nino
ir Vasyning Velocity :
• ORNL-DWG 66-6520
AP (mm H2O)
Oa~0.04
La ar 0.08
da~0.015
o 5th SERIES a0.04
..
5
10
15
20 25 30
FLOW RATE (cm/sec)
35
40
45
.
-
.
.!
1
.1.
LL
'
ALWA
A
:
1
WWWWWWWWWWWWWWWWWW
Nu
W
WWW
We
TW
V
Cri

d
hite
.
1
WATE
HITLUS
.
END
.
MATE---
-
-
.
,
0
C
.
DATE FILMED
10/ 21 / 66
RA
*
1
M
e
.
-
.
: HITI
HONI
M
W
w
,
S
.
NUL
BRY" M
:
M2
Min
MID
di
"
TT