CD A U.S. Environmental Protection Agency Industrial Environmental Research
mmm f\ Off ice of Research and Development Laboratory
Off ice of Research and Development Laboratory _ .
Research Triangle Park, North Carolina 27711 MaTCh
EPA-600/7-78-043
TECHNICAL MANUAL: A SURVEY OF
EQUIPMENT AND METHODS FOR
PARTICULATE SAMPLING IN
INDUSTRIAL PROCESS STREAMS
Interagency
Energy-Environment
Research and Development
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-600/7-78-043
March 1978
TECHNICAL MANUAL: A SURVEY OF
EQUIPMENT AND METHODS FOR
PARTICULATE SAMPLING IN INDUSTRIAL
PROCESS STREAMS
by
W. B. Smith, P. R. Cavanaugh, and R. R. Wilson
Southern Research Institute
2000 Ninth Avenue, South
Birmingham, Alabama 35205
Contract No. 68-02-2131
T. D. 10904A
Program Element No. EHE624
EPA Project Officer: D. Bruce Harris
Industrial Environmental Research Laboratory
Office of Energy, Minerals and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
Chicago,
77*»#*SeKS»
rh raeo. li- °uuu
-------
ABSTRACT
*nrf .Jhjs.technical manual lists and describes the instruments
and techniques that are available for measuring the concent?a-
F—n-^^-s i^srs
** °
cess streams.
strumpnfQ **vTH *8 t™* exPerCental method's'and'prototjpe in-
struments TO the extent that the information could be found
an evaluation of the performance" of each instrument is inSlCded.
The manual describes instruments and procedures for measur-
I?9aTsoSinc?S^rati°n^ °PaSity' and Part?cle size distribution
for contro? Sf ? Procedures for planning and implementing tests
b?bl?ography CS evaluatlon' a glossary, and an extensive
-------
CONTENTS
Page
Abstract ii
Figures yi
Tables -.'....!!!!!!!!!! xi
Acknowledgement xii
1. Introduction 1
2. Mass Concentration [ 2
Filtration 2
Introduction 2
EPA Test Method 5 . 3
Nozzle 4
Probe 7
Pitot Tube 7
Particulate Sample Collector 8
Gaseous Sample Collector 8
Sampling Box 9
Meter Box 10
Performance 10
ASTM - Test Method 14
ASME Performance Test Code 27 15
Isokinetic Sampling 17
High Volume Samplers 18
Filter Materials 20
Summary. 22
Process Monitors 23
Introduction 23
Beta Particle Attenuation Monitors 24
Instrument Development 26
Performance. .. 29
Summary. 33
Piezoelectric Mass Monitors 33
Performance 39
Temperature 40
Humidity... 40
Particle collection
characteristics 41
Linear response limit 42
Considerations for stack
application 43
Summary. 44
iii
-------
CONTENTS (CO.NT)
Charge Transfer 45
Instrument Development 46
Performance 48
Summary 50
Optical Methods 51
Conventional Transmissometers 51
Summary 58
Other Optical Methods.. 58
Multiple-wavelength transmis-
someters 58
Light scattering 62
Other Methods 78
3. Opac i ty _ g 0
4. Particle Size Distributions ................... 93
Established Techniques 93
Field Measurements 93
Aerodynamic Methods 93
Cascade impactors 94
Cyclones 109
Optical Particle Counters 118
Diffusion Batteries with Condensa-
tion Nuclei Counters 126
Electrical Mobility. 134
Laboratory Measurements 140
Sedimentation and Elutriation 140
Centrifuges 143
Microscopy.. 145
Sieves.. 151
Coulter Counter '. 153
New Techniques 153
Low Pressure Impactors. 153
Impactors with Beta Radiation Attenua-
tion Sensors 155
Cascade Impactors with Piezoelectric
Crystal Sensors 158
Virtual Impactors . , 160
Optical Measurement Techniques 163
Hot Wire Anemometry 170
Large Volume Samplers 171
5. Control Device Evaluation 173
Objectives of Control Device Tests 173
Type and Number of Tests Required 174
General Problems and Considerations 176
Plant Location 177
Laboratory Space... 177
Sampling Location and Accessibility..... 177
Power Requirements 178
IV
-------
CONTENTS (CONT)
Type of Ports 178
Flue Gas Velocity and Nozzle Sizes 178
Duct Size 179
Gas Temperature and Dew Point 179
Water Droplets and Corrosive Gases 180
Volatile Components 180
Process Cycles and Feedstock
Variations. 181
Long and Short Sampling Times 181
Planning a Field Test. 182
6. Summary 184
References 186
Glossary. . . 206
Bibliography 218
-------
FIGURES
Number
5 £or
7
8
9 wwa^iuy ut smoKe plumes containing particles of
fnn^rent ?izes.and refractive indexes as a
function of their mass concentration. After
Page
1 The EPA Method 5 particulate sampling train. . . ........ 5
2 ASTM type particulate sampling train ....... ........ ... 16
3 Schematic flow diagram of a typical RAC Automatic
Stack Monitor System installation. (Drawing not
scale.) Used by permission ..... ........ ... ......... 28
4 ™1'.^.??^.. 3S
6
' 8 a 36
Two types of particle collectors for piezoelectric
monitors A. Electrostatic precipitation
B. Impaction. After Daley and Lundgren.56 ......... 33
A £JS?Jble stack >mPling system using a proposed
double sampling diluter and a piezoelectric
microbalance sensor. After Semf ' et al.53 ........... 45
52
tion relationship of laboratory
A-** Z -"^.v. power plant emissions with
different particle sizes. After Connor.71.!";. 54
11 ^menis^rpfrticulatrm °pacity and mass measure-
• **'•••••*•• .DO
12
VI
-------
FIGURES (CONT)
Number Page
13 Opacity - mass concentration relationship for various
industrial sources. After Reisman, et al.78 59
14 Mean extinction coefficient as a function of the
phase shift parameter p . After Dobbins and
Jizmagian. 8 ° .v.s.* 60
15 Results of monochromatic vs. white-light optical
density measurements made on sludge incinerator
emissions. After Reisman, et al.78 63
16 Scattering function vs. particle radius for several
refractive indices. After Quensel.83 65
17 Optical assembly diagram of a nephelometer used in
stack monitoring. The scattering angle 9, for
any light ray from the source, is the angle be-
tween the ray and the horizontal line a. From
Ensor and Bevan.8 5 66
18 Mass correlation data taken with the Plant Process
Viscometer at the inlet and outlet of a particulate
scrubber on a coal-fired utility boiler. After
Ensor, et al.88.. 68
19 Effect of particle size distribution on particle
volume concentration/scattering coefficient.
From Ensor, et al.8 8 69
20 Optical diagram of the PILLS V instrument. From
Schmitt, etal.91 70
21 Theoretical response of PILLS V vs. particle size.
Calculations for log-normal size distributions with
geometric standard deviations of 1.65 and varying
number mean diameter. From Schmitt, et al.93 72
22 Schematic of Laser-TV Monitor. After Tipton.96 77
23 Schematic of a transmissometer showing projection
and viewing angles which must be no greater than
5° for EPA compliance 82
24 A typical double pass in situ transmissometer design.
After Nader.7 9 83
VII
-------
FIGURES (CONT)
Number Page
25 A single pass transmissometer design. After
Haville1 °3 85
26 Effluent transmittance vs. in stack transmittance for
varying ratios of stack exit diameter to in stack
path length: A = 1/4, B = 1/2, C = 3/4, D = 1,
E = 4/3, F=2, G = 4. After Nader.105 87
27 Particle extinction coefficients for various aerosols
over three scattering regions: Reyleigh, Mie, and
Geometric.... 90
28 Schematic diagram, operation of cascade impactor 95
29 Approximate relationship among jet diameter, number
of jets per stage, jet velocity, and stage cut
point for circular jet impactors. From Smith
and McCain.l23 f f 97
30 Design chart for round impactors. (D50 = aerodynamic
diameter at 50% cut point.) After Marple.112 ... 98
31 Schematics of five commercial cascade impactors 105
32 Calibration of an Anderson Mark III impactor. Collec-
tion efficiency vs. particle size for stages 1
through 8. After Gushing, et al.117 108
33 Hypothetical flow through typical reverse flow
cyclone 110
34 Comparison of cascade impactor stage with cyclone
collection efficiency curve 112
35 Series cyclone used in the U.S.S.R. for sizing flue
gas aerosol particles. From Rusanov.132 114
36 Schematic of the Southern Research Institute Three
Series Cyclone System 115
37 Comparison of Southern Research Institute Three Series
Cyclone System data with cascade impactor curve
After Gooding. 13'*. -f 116
38 The EPA/Southern Research Institute Five Series
Cyclone System ^ 117
viii
-------
FIGURES (CONT)
Number Page
39 Laboratory calibration of the EPA/Southern Research
Institute Five Series Cyclone System. (Flow rate
of 28.3 &/rain, temperature of 20°C, and particle
density of 1 g/cm3.).... 119
40 Schematic of the Acurex-Aerotherm Source Assessment
Sampling System (SASS) .. . 120
41 Schematic of an optical single particle counter 121
42 Experimental calibration curves for two optical
particle counters. After Willeke and Liu.136 123
43 Optical configurations for six commercial particle
counters 124
44 A rectangular channel diffusion battery 127
45 Screen type diffusion battery. The battery is 21 cm
long, 4 cm in diameter, and contains 55,635 mesh
stainless steel screens. After Sinclair . 1 "*5 ... 128
46 Diagram of a condensation nuclei counter. After
Haber 1 and Fusco.1 "** 130
47 Diffusion battery and condensation nuclei counter
layout for fine particle sizing... 132
48 Theoretical parallel plate diffusion battery penetra-
tion curves 133
49 Particle mobility as a function of diameter for
shellac aerosol particles charged in a positive
ion field. K is the dielectric constant of the
aerosol particles. After Cochet and Trillat.155 135
50 The electric mobility principle.. 137
51 Schematic of the Thermosystems Model 3030 Electrical
Aerosol Analyser. After Sem.158 138
52 The Roller elutriator. After Allen.160 142
53 The Bahco microparticle classifier 144
ix
-------
FIGURES (CONT)
Number Page
54 A cut-away sketch of the Goetz Aerosol Spectrometer
spiral centrifuge. In assembled form the vertical
axes (1) coincide and the horizontal arrows (2)
coincide. After Gerber. 16I* ... 145
55 Cross-sectional sketch of the Stober Centrifuge.
After Stober and Flachsbart. 16 5 .... 147
56 Cross-sectional sketch of a conifuge 148
57 Three diameters used to estimate particle size in
microscopic analysis 150
58 Operating principle of the Coulter counter. Courtesy
of Coulter Electronics .. 154
59 Cross section of prototype Mark IV University of
Washington Source Test Cascade Impactor 156
60 Sampling train utilizing a low pressure impactor.
After Pilat.176 157
61 Particle deposition and' ^-attenuation with a. a
continuously moving tape, and b. a stationary
tape that is stepped forward periodically 159
62 Virtual impactors (centripeters, dichotomus samplers,
stagnation impactors) a. impingement into a stag-
nant air space; b. opposed axisymmetric jets 161
63 Scattered light intensity versus scattering angle
for two spherical particles of equal diameter.
The solid curve is for a glassy, non-absorbing
sphere and the dashed curve is for an absorbing
sphere. After Gravatt. 1 9 l 165
64 A system for collecting large volume samples from
industrial process streams. After McFarland
and Bertch.1206 ._ 172
-------
TABLES
Number Page
1 Sampling Systems for Testing by EPA Method 5 6
2 Overall Comparison 20
3 Glass Fiber Filter Products 20
4 Typical Flue Gas Conditions and Operating Variables
for CPM Calibration (after Wostradowski95).... 74
5 Linear Regression Results of CPM Calibration Curves
for Each Mill (after Wostradowski95) 74
6 Comparison of Correlation Data with Particulate
Characteristics (after Wostradowski 9 5) 75
7 Relative PILLS V Response as a Function of Salt Con-
centration and Color of the Extracted Dust Samples
from Mill B (after Wostradowski95) 75
8 Field Test Results. (Results in Grains/Standard ft3.
After Tipton9 6) 78
9 Commercial Cascade Impactor Sampling Systems 99
10 Cascade Impactor Stage Parameters, Andersen Mark III
Stack Sampler 100
11 Cascade Impactor Stage Parameters, Modified Brink
Model B Cascade Impactor 101
12 Cascade Impactor Stage Parameters, MRI Model 1502
Inertial Cascade Impactors 102
13 Cascade Impactor Stage Parameters, Sierra Model 226
Source Sampler 103
14 Cascade Impactor Stage Parameters, University of
Washington Mark III Source Test Cascade Impactor 104
15 Characteristics of Commercial, Optical, Particle
Counters ^ -125
16 Comparison Table of Common Sieve Series 152
17 Particulate Control Device Tests 133
18 Status of Particulate Sampling Methods for Process
Streams
XI
185
-------
ACKNOWLEDGEMENT
Members of the Southern Research Institute staff who helped
to write and edit this report are William Farthing, Kenneth M.
Gushing, Joseph D. McCain, Charles Feazel, and James Ragland.
Don Davis and Michael Myers were the illustrators. Doris Thrower,
Ann Billingsley, and Anne Smith prepared the manuscript; and
Dorothy Reedy prepared the bibliography. The assistance and coop-
eration of all are appreciated.
xii
-------
SECTION I
INTRODUCTION
The purpose of this manual is to list and describe the in-
struments and techniques that are available for measuring the
concentration or size distribution of particles suspended in
process streams. The standard, or well established methods are
described as well as some experimental methods and prototype in-
struments. To the extent that the information could be found,
an evaluation of the performance of each instrument is included.
It is not within the scope of this document to train personnel
to make the measurements, but to provide a project leader with
enough information to select intelligently the methods and in-
struments to be used.
Section II contains descriptions of instruments and procedures
for measuring mass concentrations, Section III is devoted to
measurements of opacity, Section IV to particle-size measurements,
and Section V specifically to control device evaluation. Section
VI is a brief summary of the status of instruments that are in
use or under development for sampling particulate matter in process
streams. Also, a glossary and extensive bibliography are included.
-------
SECTION II
MASS CONCENTRATION
FILTRATION
Introduction
Particulate mass concentration measurement methods using
filtration as the means of sample collection can be classified
according to the sampling flow rate used and the location of
the filter in or out of the process stream. Low sampling flow
rate methods usually sample in the 14.2 £/min (1/2 ft3/min)
to 42.5 jt/min (1 1/2 ft'/min) range. High flow rate methods
usually operate above 142 £/min (5 ft3/min). Use of a filter
located outside the confines of the process stream is referred
to as an extractive method. Use of the filter located in the
process stream is referred to as an in situ method.
Various organizations have promulgated specific procedures
and sampling train designs for one or more of these methods.
The EPA Test Method 5' specifies the use of an extractive sampler
Sampling trains constructed to meet Method 5 specifications were
initially designed to operate at flow rates up to 28.3 £/min
(1 ftVmin); but, recently a 113 £/min (4 SCFM) extractive sam-
pler has been developed which (reportedly) complies with the
Method 5 specifications. The proposed EPA Test Method 17 speci-
fies the use of in situ sampling.2 The American Society for
Testing Materials (ASTM) specifies an in situ sampler.3 The
American Society of Mechanical Engineers (ASME) Performance
Test Code 27 specifies the use of either an in situ or extractive
sampler.* The ASME will soon be releasing a new Performance
Test Code 38 which will supercede the Performance Test Code 27.
-------
The Industrial Gas Cleaning Institute (IGCI Pub. No. 101) and
Western Precipitation Company (Bulletin WP50) have also suggested
sampling methods.
EPA Test Method 5
Official performance testing of stationary sources for
particulate emissions must be conducted with the EPA Test Method
5 "Determination of Particulate Emission from Stationary Sources".1
The stationary sources covered include new steam boilers, in-
cinerators, cement plants, pulp and paper mills, and the like.
All states require the use of some form of the Method 5 train
for compliance testing. Method 5 relies on the removal or extrac-
tion of a dust laden gaseous sample from the duct or stack fol-
lowed by the subsequent removal of the particles on a filter
with concurrent measurement of the sample volume to determine
particulate concentration.
With EPA Method 5, one obtains a measure of the average
particulate mass concentration for the cross-sectional area
of the duct during the time sampling takes place. However,
only the particles that are present at or below 120°C are detec-
ted. The particulate concentration is expressed in terms of
the dry component of the stack gas, excluding the component
contributed by water and other vapors. (Regulations in some
states include the wet component.) Finally, this concentration
is expressed as the concentration that would be present under
conditions of standard temperature and pressure.
A sample is removed from the duct by using a prescribed
traversing procedure which involves drawing portions of the
sample from different points within the duct. This procedure
yields, in effect, an approximate integration of collected mass
and sample volume over the cross-sectional area of the duct.
-------
Before sampling, however, it is necessary to determine the
number of sampling points. EPA Test Method 1 "Sample and Velo-
city Traverse for Stationary Sources"5 describes the computations
used to determine the number of sampling points for both the velo-
city traverse and mass sampling traverse. The number of points
will depend on the size and shape of the duct. The velocity tra-
verse is performed before the mass sampling traverse using EPA
Test Method 2 "Determination of Stack Gas Velocity and Volumetric
Flow Rate".6 The velocity information is used to select the proper
nozzle size and to determine the average stack velocity from which
the average volumetric flow rate is determined. The average volu-
metric flow rate along with the average participate concentration
obtained.by Method 5 is used to determine the average mass emis-
sion rate.
Method 5 requires that isokinetic sampling conditions be main-
tained. Thus, at each traverse point, the sample velocity at that
point is adjusted to equal the gas velocity in the duct.
The Federal Register gives detailed specification for the
apparatus comprising the sampling train which must be used to
properly conduct a Method 5 test. The sampling train consists
of a nozzle, probe, pitot tube, particulate sample collector,
gaseous sample collector, sampling box, and meter set; refer to
Figure 1. The user can either construct his own sampling train
by following the specifications7 or he can use one of the many
commercial models available (see Table I).
Nozzle—
The nozzle removes the sample from the gas stream and has
several restrictions to its use. It should disturb the gas flow
as little as possible, or the sample will not be representative.
A thin wall, sharp edged, nozzle disturbs the flow the least.
The nozzle should be of a size to permit isokinetic sampling and
-------
TEMPERATURE
SENSOR
PROBE
I
HEATED
AREA
FILTER HOLDER
REVERSE-TYPE
PITOT TUBE
I
PITOT
MANOMETER
THERMOM
ETERS
ORIFICE
IMPINGER TRAIN OPTIONAL:
MAY BE REPLACED BY AN
EQUIVALENT CONDENSER
THERMOMETER
CHECK
VALVE
BY-PASS
VALVE
"•' itXW-tX*
MAIN
VALVE
MANOMETER DRY TEST METER AIR TIGHT PUMP
Figure 1. The EPA Method 5 paniculate sampling train.
VACUUM LINE
0700-14.16
3630-201
-------
TABLE I. SAMPLING SYSTEMS FOR TESTING BY EPA METHOD 5
en
Company
Aerotherm-Acurex
Glass Innovations, Inc.
Joy Manufacturing Co.
Lear Siegler, Inc./
Environmental Tech-
nology Div.
Misco International
Chemicals, Inc.
Research Appliance
Company
Scientific Glass &
Instruments, Inc.
Address
485 Clyde Avenue,
Mountain View, CA 94042
P.O. Box B
Addison, NY 14801
Commerce Road
Montgomeryville, PA 18936
One Inverness Dr. East
Englewood, CA 80110
1021 S. Noel Avenue
Wheeling, IL 60090
Pioneer and Hardies Rd,
Gibsonia, PA 15044
7246 Wynnewood
Houston, TX 77001
Train Title
High Volume Stack Sampler
The Source Sampler
Emission Parameter Analyzer
PM100 Manual Stack Sampler
Stack Source Sampler
Staksamplr
Stack-0-Lator
Note: Most companies will supply filters for use with their trains upon request.
-------
to allow easy access. A buttonhook nozzle is usually used, since
it has the advantage of requiring a minimum porthole size of 6.4
cm (1 1/2 in.). A set of nozzles, 0.32 cm (1/8 in.) up to 1.27
cm (1/2 in.) inside diameter of 0.16 cm (1/16 in.) increments is
usually sufficient for routine stack sampling applications.
Probe—
The probe removes the sampled stream from the stack. The
major requirement is that it does not significantly alter the
sample from stack conditions. The sample temperature may be allowed
to fall below the gas temperature in the duct, and the filter box
can be no hotter than 120°C±14°C; except for fossil fueled power
plants, where the maximum is 160°C. If the aerosol temperature
falls too low, water and other condensible vapors in the sample
stream begin to condense, which quickly leads to clogging of the
filter media. Method 5 stipulates the use of heat resistant glass
sampling lines for probes less than 2.5 meter (8 ft.) in length.
Glass probe liners, as compared to metal probe liners, are de-
sirable because of the ease and completeness with which glass can
be cleaned and because glass is chemically inert. However, to
prevent breakage in lengths over 2.5 meters, Method 5 permits the
use of approved steel probes. New regulations require a thermo-
couple to be attached to the probe end for monitoring the stack
gas temperature.
Pitot Tube—
A pitot tube is attached to the probe to monitor the gas velo-
city. A pressure drop, which is generated by the gas velocity
in the duct, is monitored to insure isdkinetic sampling velocities.
Method 5 specifies a Stausscheibe type, also called the S-type
or reversed pitot. This pitot is used instead of a standard type
(Prandt) because of the clogging tendency of the standard type
-------
in heavy dust laden streams. The S-type has two main advantages:
it is easy to construct and, because of its design, it shows
a larger difference in pressure for the same gas velocity, than
the standard. At extremely low or high velocities, the pitot
method is inaccurate and unreliable. Calibration of a properly
constructed S-type pitot tube is not required.
Particulate Sample Collector—
The glass fiber filter should be at least 99.95% efficient
for collection of 0.3 micron dioctylpthalate smoke particles.
It must also be inert to chemical reactions. There are sometimes
problems with unknown efficiency characteristics of the glass
fiber filter. Method 5 requires a filter efficiency test to
be performed, unless the test data from the suppliers quality
control program indicates that the filter has sufficient collection
efficiency.
The cyclone, which is available with many commercial trains,
is not required by Method 5. Its use results in the removal
of larger particles, thus allowing for longer sampling times
by preventing early blocking of the filter.
Gaseous Sample Collector —
Gas absorbers are generally called impingers or bubblers.
There are four impingers in the Method 5 sample train. They
remove the water, gases, vapor and condensible particulate matter
and allow the moisture content of the gas stream to be determined.
Condensible particulate matter is defined as that which is formed
by either coalescing of ultrafine particles or by condensation
of gases upon temperature reduction. All impingers are of the
Greenberg-Smith type. The first impinger cools the hot gases
and provides for some gas absorption/condensation. The second
impinger is for fine particulate removal and final gas absorption.
8
-------
The third impinger is a dry type used to collect any carryover
from the previous impingers. The last impinger contains silica
gel for the final removal of all water vapor. By removing the
water vapor, the silica gel serves to protect the working parts
of the vaccuum pump and dry gas meter. There are two other items
in the impinger train for the protection of the metering system
and sample. The first is a thermometer so that the temperature
of the exit gas from the bubblers can be kept low enough to protect
the metering system. The second item is a check valve which
primarily protects the sample by preventing any backflow through
the sample. When the filter becomes loaded, a partial vacuum
is created between the filter and impingers. If a check valve
is not used, or if the valve sticks when the pump is turned off,
the water in the first bubbler can be drawn back up into the
filter holder. Also, if the stack is at a significant negative
pressure, the vacuum created between the filter and the stack
would add to the vacuum created by the loading effect and could
result in a ruptured filter.
The EPA and some states do not require the measurement of
the condensible particulate fraction and hence the impingers
are not specifically required. The fragile glass impinger train
may be replaced by a suitable condenser. The condenser may be
as simple as a piece of coiled tubing immersed in an ice bath.
The condenser should be followed by a silica gel drying tube
to collect the remaining moisture and protect the vacuum pump
and dry gas meter.
Sampling Box—
The sampling box serves to hold the probe, the filter holder,
the impinger train and its ice bath. The filter holder is con-
tained in a heated area of the sampling box and must be kept
below 120°C±14°C. These areas must be well insulated and water
tight. There are many choices of sample box supports available
-------
with commercial sampling trains. Several use an overhead rail-
type support while others use supports on which the box slides.
Where the condensable particulate fraction is not required by
state regulation or is of no interest, the sampling box can be
simplified.8 The impingers/condenser, then, can be located remotely
from the sampling box, with the connection to the filter made
by means of an umbilical cord.
Meter Box—
The meter or control box contains the following items:
a vacuum pump capable of maintaining isokinetic flows during
heavy filter loadings,• a control valve to vary the sample stream
flow rate; a vacuum gauge for measuring the sample stream pres-
sure; a dry gas meter equipped with temperature measuring de-
vices at inlet and outlet for determining the sample volume,
a calibrated orifice meter which is used to monitor the sample
stream flow rate; two pressure gauges, one to measure the pitot
tube pressure drop, and the other to measure the orifice meter
pressure drop; a variable voltage power supply to maintain the
probe and filter box at their respective temperature by means
of their individual heaters; and a pyrometer or a potentiometer
suitably calibrated for thermocouple measurements of the duct
and filter box temperature.
Calibration requirements are discussed in the EPA maintenance
procedures.' Critical laboratory calibrations include the orifice
meter, dry gas meter, and pitot tube. Calibration of the orifice
meter and dry gas meter requires the use of a wet gas meter.
Various other common laboratory instruments are required for
the maintenance and calibration of the other system components.
Performance—
An inherent limitation of the Method 5, indeed, of all
stack sampling systems, is the inability to obtain particulate
10
-------
matter in the same state as it exists when the plume mixes with
the atmosphere. This change to atmospheric conditions may re-
sult in particulate matter being formed in the plume that was
not present in the stack. In its original form, Method 5 also
captured the particulate matter that occurred after the sampled
gas was quenched in an ice bath and thus it attributed to the
plume, particulate matter that, it was expected, did not occur
in the actual plume. The present form does not include the im-
pinger catch. Tests were performed that compared EPA Method
5, including impingers with the commonly used alundum thimble
method, which followed ASME guidelines, and with the thimble
located out of stack. Simultaneous determination of particulate
mass concentration were made on the flue gas from a furnace burn-
ing about 34.0 kg/hr (75 Ib/hr) of low-sulfur lignite to yield
dust concentrations of 1.14 to 0.02 gm/m3 (0.5 to 0.09 grain/SCF).
The two methods agreed within 5% at the higher concentration
and the impinger residues were not significant. At the low con-
centration the EPA train yielded values from 50 to 200% greater
than the ASME train, and the impinger residues constituted a
substantial part of the total weight.10
The performance of the Method 5 has been compared with a
newly developed dilution source sampling system.11 This system
uses dry atmospheric air to withdraw a sample from a process
gas stream and simultaneously dilute and cool it with atmospheric
air to ambient temperature and pressure in ways that replicate
the way actual plumes reach equilibrium with the atmosphere.
Tests of the diluter and Method 5 were simultaneously conducted
on the exhaust of a coal-fired stoker furnace with over-fired
oil. Results from one test showed the dilution sampling system
recorded particulate concentration 7.2% larger than that recorded
by Method 5. A second test of the diluter and an in-stack impactor
showed the existence of submicron particles, that would be emitted
to the atmosphere, that were not present at stack gas temperature.
Since, for most sources, particularly coal-fired boilers, submicron
11
-------
particles comprise only a small fraction of the total mass, this
occurrence is insignificant when measuring mass concentration.
However, fine particles would be important for the characteri-
zation of plumes.
A possible problem, one that is just the opposite of the
first problem discussed, is the measurement of "false particles"
in the sampling process. For example, particles collected early
in the run and distributed along the probe walls and in the filter
provide a large surface area continually exposed to the fresh
sample. There is thus an opportunity for catalytic oxidation
of S02 and S03 , physi- and chemisorption of organic vapors on
particulate surfaces, or compound conversion; or if cooling of
the sample gas occurs, sulfuric acid can condense to an aerosol.
A study done on oil-fired boilers indicated that Method 5 proce-
dure does not lead to the formation of false particles for that
source, except when the S03 concentration and external filter
temperature permit condensation of sulfuric acid.12
A study performed on glass-fiber filter media did show a
reaction between the SO component in the process stream and
A
the filter media.13'11* However, this reaction was found to be
active primarily at high process stream temperatures. At 120°C,
the Method 5 sampling temperature, the weight gains due to SO
J\
reaction did not exceed l.mg for 47 mm diameter filters.
Another problem arises when the humidity of the process
stream is high. The sampling flow rate is measured by refer-
ring to the pressure drop across the orifice meter after the
water vapor has been removed by the impingers. Therefore, the
measured volumetric flow rate through the orifice is always less
than the actual volumetric flow rate through the nozzle. This
may produce an anisokinetic sampling condition. For example,
when sampling a stack whose moisture content during sampling
shows a 2% drop, from 10% to 8%, an error of approximately 2.2%
12
-------
will be reflected in the measured volumetric flow rate. But
when sampling a stack where moisture content was calculated to
be 90% and the average moisture content during sampling shows
a 2% drop to 88%, an error of approximately 16.7% will be re-
flected in the measured volumetric flow rate. Such an error
is enough to invalidate the test, as the percent isokineticity
will fall outside the 90 to 110% range. An experimental method
to overcome this problem is described by Patankar and Ott.15
Probe loss studies in the Method 5 probe have shown that
deposition of particles in the heated probe ranged from 10% for
a 2 pro MMD aerosol to 97% for a 17 \im MMD aerosol.16 Such high
probe deposition indicates the need for consistent, thorough
probe washing procedures.
Studies have been performed to determine the precision obtain-
able with the Method 5 and the clarity of its instructions. A
series of collaborative tests were performed by Hamil et al.
on fossil fuel-fired steam generators,17 Portland cement plants,18
and municipal incinerators.19'20 A collaborative test involves
several laboratories simultaneously conducting tests on a common
source. There tests were performed in a "real world" manner;
this involves no interaction between collaborating laboratories
or outside supervision, which would bias the results.
The results of the collaborative tests are expressed as stan-
dard deviations of three principal components:
a^ - the between-laboratory standard deviation. This repre-
sents the total variation in a result, composed of with-
in-laboratory and laboratory bias components. The
between-laboratory variance can be written as
"
13
-------
a - the within-laboratory standard deviation. This repre-
sents the ordinary sampling error in replicate runs made
at the same mass concentration by the same laboratory
te am.
aL - the laboratory bias standard deviation. This represents
the variation that can be expected between two independent
laboratory teams determining results at the same mass con-
centration. This variation is attributable to such fac-
tors as different operation, equipment, and analysis.
The most recent collaborative test,20 performed on an incinerator,
produced the following results:
ab = 12'1% a = 10.4% a = 6.1%
LI
These estimates were considerably lower than those obtained in
the first collaborative tests.17'18'19 The test study attributed
the improved results to the use of more experienced laboratory col-
laborators, the performance of more frequent calibration checks
and rechecks, the revision of the Method 5 sample handling and
recovery procedures, more reliable statistics, and the absence
of the "high value" effects found in the first collaborative
tests. The "high value" effects were thought to be due to the
accidental scraping of the particulate matter adhering to the
stack wall into the sampling probe tip during the probes' inser-
tion and removal through the sampling port.
ASTM - Test Method
Both the ASTM and the ASME provide specifications for in
situ samplers. The ASTM Method is similar to the EPA Test Method
5. The main difference is the use of an instack filter with
no restrictions on the sampling flow rate used. However, the
14
-------
sizes of the sampler components (tubing, filter holder, etc.)
usually place an upper limit on the flow rate. With the ASTM
arrangement, shown in Figure 2, a thimble-shaped filter is used
to sample high mass concentrations. The pitot tube and the other
parts are similar to the Method 5 sampler. External heating
of the filter by auxiliary equipment is usually not needed.
However, the filter should be preheated by locating the filter
in the process stream for at least 30 minutes to insure that
the temperature of the filter is in equilibrium with the tempera-
ture of the process stream. When inserting the filter for pre-
heating, the nozzle must be pointed in the downstream direction
of the gas flow. This orientation will prevent pretest collec-
tion of particulate matter. Also, when inserting the filter
into a duct not at ambient pressure, the sampling lines must
be closed in order to prevent undesirable gas flow through the
filter.
ASME Performance Test Code 27
The ASME Performance Test Code provides for the use of a vari-
ety of instruments and methods.4 Paragraph 55 of Section 4 of the
Code states "Testing experience has not been uniform enough to
permit standardized sampler design. This code, therefore,
merely gives limiting requirements which past experience has shown
desirable to avoid major sources of error". The Code is designed
as a source document which provides technically sound options
to be selected and agreed upon by the contractor and the contrac-
tee who performs the sampling. According to the Code, the sam-
pling device shall consist of a tube or nozzle for insertion
into the gas stream and through which the sample is drawn, and
a filter (thimble, flat dish, or bag type) for removing the par-
ticles. For the purpose of the Power Test Code, 99.0% collection
efficiency by weight is satisfactory, and the filter can be made
of cotton, wool, filter paper, glass wool, nylon, or orIon.
The filter can be an extractive or in situ filter, however, the
extractive filter is used much of the time.
15
-------
GLASS FIBER THIMBLE FILTER
SAMPLING
NOZZLE
(W^
V^u
U ;
--A , /
....'} — /. - 1 — '
i 1 ' >y \
pnwnPMCPo
no VCD
UK Y tn
CHECK
VALVE
i — ~^
^\
\
REVERSE-TYPE
PITOT TUBE
PITOT
MANOMETER-^"
THERMOMETERS^ fc
ORIFICE
MAIN VALVE
DRY TEST METER
AIR-TIGHT PUMP
070O-14.17
3630-202
Figure 2. ASTM type particulate sampling train.
16
-------
Isokinetic Sampling—
Also required is a means of checking the quality of the
velocity of the gas entering the nozzle and the velocity of the
gas in the flue at the point of sampling. This can be performed
by use of a pitot tube traverse before the sampling is performed.
Then the appropriate sampling flow rate can be calculated for
each point. If the process stream velocity is not constant,
pi tot tube measurements should be performed during sampling to
insure isokinetic sampling conditions. Or, in the case of varia-
ble process stream velocities, a null type nozzle can be used
in place of the pitot traverse. A null type nozzle operates
on the principle that isokinetic sampling exists when the static
pressure in the duct is equal to the static pressure inside the
sample stream.
A method to measure the quantity of gas sampled and a pump
to draw the sample is needed. Provisions should be made to heat
the filter when used extractively.
Both the EPA Method 5 and the ASTM Method comply with the
ASME Method.
Performance tests have compared the in situ ASME Method
with the EPA Test Method 5. These tests produced conflicting
results.21 Other comparative studies were conducted by the
National Council of the Paper Industry for Air and Stream Improve-
ments.22 A summary of a company's experience in the measurement
of fly ash collector efficiencies and particulate emissions from
coal-fired power plants using the ASME Method and modifications
to the ASME Method is given in Reference 21.
17
-------
High Volume Samplers
High volume samplers are designed to gather a relatively
large amount of particulate matter in a short period of time.
Two commercial versions, one made by Rader Pneumatics Inc.,
P.O. Box 20128, Portland, Oregon 97220, and the other by Acurex-
Aerotherm Corporation, 485 Clyde Avenue, Mountain View, California,
represent the state of the art. Rader manufactures both a manual
and an automatic sampler. The Aerotherm version meets EPA Test
Method 5 specifications and therefore can be used for compliance
testing (manufacturer's claim).
The Aerotherm Train includes three nozzles 6.35 mm, 12.7 mm,
19.8 mm (1/4, 1/2, and 3/4 inches) in diameter. The probe can be
rotated through 360° for sampling in off-angle ducts, and it
can be turned 90° for sampling horizontal ducts from the top.
The 142 mm diameter filter is fiberglass and is enclosed in a
heated box at the end of the probe. Isokinetic sampling is main-
tained manually with the aid of nomographs. The gauges for the
pitot tube, orifice, and temperature conditions are located in
a control unit. The nominal sample flow rate is 113 i/min
(4 SCFM). It has four impingers and can be operated at tempera-
tures up to 260°C (500°F).
The Rader unit, a portable, hand-held utility sampler, uses
an unheated 20.3 x 25.4 cm (8x10 inch) filter mounted on the
probe. Isokinetic sampling can be maintained either manually
or automatically. The automatic model has a microprocessor that
controls sampling functions, performs calculations, and displays
sampling parameter values measured by the sensors. The manual
model has the gauges for the pitot tube, orifice, and temperature
measurements mounted on the probe. Sampling flow rates up to
1.98 m3/min (70 ft3/min) can be achieved and the sampler can be
operated in temperatures up to 260°C (500°F). Neither model has
impingers or condensers to recover the water vapor in the sample
stream.
18
-------
The Rader unit, has been used to measure emissions from hog
fuel fired boilers, a bark burning boiler, a bark and wood fired
incinerator, wigwam burners, an asphalt batching plant, a seed
cleaning plant, and a filter-cyclone system handling wood fiber.23
The Rader Hi-Volume sampler was compared against the "standard"
Method 5 train, which included the impinger catch (back half) ,
and a "modified" Method 5 train which considered only the probe
and filter catch (front half).21* Throughout the test the back
half catch totaled from 1.8% to 23% of the total "standard"
Method 5 train. The comparison of the overall results is shown
in Table II. An overall mean, x, standard deviation, s, and
standard error, S-, (standard deviation of the mean) for the
J\
mass loading are computed from all test runs for each of the
sampling methods. The standard deviation and standard errors
for the high-volume method are both lower than for the other
methods. This might be at least partially attributed to the
number of samples (runs) per test collected by the different
methods. Both Method 5 results were determined from two samples
collected each day - one in the morning and one in the afternoon.
The high-volume method had a maximum of 8 and a minimum of 4
samples taken per day, making the statistics more reliable, and,
consequently, the standard deviation and standard error would
tend to be lower for the high-volume method.
A portable, lightweight, intermediate-volume sampler with a
filter area of 129 cm2 (20 sq. in.) has also been developed.
The instrument samples in the .09 to .68 m3/nun (4 to 29 CFM)
range with resulting nozzle sample velocity matching the 25-250
cm/sec (10-100 fps) velocities normally encountered in stacks.
A high temperature 0-ring seal control external leakage. Field
test measurements with the sampler demonstrated a 94% correla-
tion with standard methods.25
19
-------
TABLE II. OVERALL COMPARISON
High Volume
Method 5
Modified Method 5
X
(gr/SDCF)
.199
.202
.182
s
(gr/SDCF)
.064
.080
.071
s-
X
(gr/SDCF)
.024
.030
.027
Filter Materials
Filter materials for use in mass collection equipment are
available from scientific equipment supply houses in several
different shapes, sizes, and compositions. Although membrane,
teflon fiber, cellulose, metal-alloy, quartz, and ceramic filters
are available, the most widely used for stack sampling is the
glass fiber filter. A list of glass fiber filters commonly used
in air pollution studies is given in Table III. This list is
not exhaustive. For a particular test, a filter should be chosen
considering the objectives of the testing program and the char-
acteristics of the sampling environment and equipment. In some
sampling programs, a particular filter may be used to avoid spe-
cific problems. For example, reaction between the sulfur com-
pounds in a gas and alkaline sites in glass fibers often cause
unwanted weight gains. In such a situation, it would be desir-
able to use a filter that is known to have low sulfur sensitivity,
such as Reeve Angel 934'AH.26
TABLE III. GLASS FIBER FILTER PRODUCTS
Gelman Type A Gelman Instrument Company
Gelman Type AE 600 S. Wagner Road
Spectrograde Ann Arbor, MI 48106
MSA 1106-BH Mine Safety Appliance Co.
400 Penn Center Blvd.
Pittsburgh, PA 15235
20
-------
TABLE III (Continued)
GF/A Whatman, Inc.
GF/C 9 Bridewell Place
GF/D Clifton, NJ 07014
Reeve Angel 90OAF
Reeve Angel 934AH
Balston Microfibre Balston, Inc.
703 Massachusetts Ave,
Lexington, MA 02173
The EPA Method 5 specifies the use of glass fiber filters
having 99.95% collection efficiency for 0.3 urn dioctyl phthalate
(OOP) smoke particles. The gas velocity, sample conditions,
and the size of the filter are not specified, nor is a list of
commercial filters meeting the specification provided. However,
test data from the filter supplier's quality control program
can be substituted for efficiency tests results.1 Efficiencies
of some filters have been measured by Appel and Wesolowski,27
Elder et.al.,28 Mueller,29 Lundgren and Gunderson3°'31 and Staf-
ford and Ettinger.32'33'3"
Thimbles for in situ sampling probes are made by Schleicher
and Schuell, Nuclepore, and Flakt, and are available from BGI,
Inc., Waltham, MA; SF Air Products, Inc., Old Greenwich, CT;
and the Carborundum Company, Knoxville, TN; as well as from the
suppliers of ASTM and other sampling trains. A brief survey
of thimble holders is given by M. Ellis in the April 1976 issue
of Stack Sampling News.
21
-------
Summary
The major difference between the filtration methods is their
requirements concerning the use of in situ and/or extractive
sampling. The EPA Method 5 requires extractive; the proposed
EPA Method 17 requires in situ; the ASTM Method requires in-situ;
and the ASME Method permits both in situ and extractive. Because
of the large size of the high volume sampler's filter holders,
these devices are not designed for in situ use.
The main advantage of the in situ sampler over the extractive
sampler is the fact that substantially all of the particulate
matter is deposited directly in the filter; therefore, only a
small area needs to be washed. Because the filter is maintained
at the stack gas temperature, auxiliary heating of the filter
is not needed.
The main disadvantage of the in situ sampler over the extrac-
tive sampler is the fact that the in situ sampler is limited
to process streams where temperatures do not exceed the limit
of the filter medium and holder. In fact, thermal expansion of
the filter holder may create gas leakage problems. The instack
filter cannot yield data on condensable particulate matter in
the plume.
Another difference between the filtration methods is the
sampling flow rate(a) used in each method. Sampling trains con-
structed to meet EPA Method 5 specifications were initially de-
signed to operate at flow rates up to 28.3 l/min (1 ft3/min)-
recently, a 113 £/min (4 SCFM) sampler has been developed which
complies with EPA Method 5 specifications. ASTM and ASME Methods
do not define a flow rate range. Some high volume trains can
operate at flow rates up to 1.98 m3/min (70 ft3/min)
22
-------
The main advantage in the use of a high flow rate sampler
lies in the fact that the amount of time required to sample a
given volume of stack gas is small compared to a low flow rate
sampler. In a process stream where the mass concentration is
constant, the time required for sampling is markedly reduced.
In a process stream where the mass concentration is highly varia-
ble, a larger number of high volume runs would be required to
obtain a value representation of the same average mass concentra-
tion obtainable from one run of the low volume run. Statistically,
it is more desirable to obtain several samples of a value than
just one sample. For stable streams this will give additional
information revealing the precision with which the method has been
applied. When using high flow rate extractive samplers the high
ratio of sample gas flow rate to probe wall area minimizes errors
due to loss of particulate matter on the tubing walls between
the nozzle and the filter, minimizes heat losses, and thus helps
to prevent the condensation of vapors in the train. The high
ratio also can be a disadvantage when cooling of the sample gas
stream is required to protect the equipment since auxiliary cool-
ing equipment may be needed.
PROCESS MONITORS
Introduction
The ideal process stream mass monitor would have the follow-
ing features:
1. The sensing principle used to detect the particles in a gas
stream would be a direct measurement of the mass of the par-
Hr-j,-:-;.
2. The mass sensor would be insensitive to such factors as changes
in gas temperature and humidity, corrosive gases, and liquid
droplets.
3. The monitor would provide continuous, instantaneous ("real-
time") measurements of mass concentration.
23
-------
4. Since the mass concentration in a process stream often varies
within the cross-sectional area of the duct, the ideal monitor
would measure the average mass concentration across the entire
cross-sectional area of the duct.
5. A monitor with its sensor mounted directly within the gas
stream, called an in situ monitor, is generally preferred over
the extractive monitor, in which the sample may be altered sig-
nificantly prior to the measurement.
No monitor currently available has all the above qualifica-
tions. The development of process monitors has begun to gather
momentum only recently, and much of the performance data pertain-
ing to their operation at various sources and under various con-
ditions has been shown to be contradictory or of limited useful-
ness. Nevertheless, a process monitor may provide sufficient
accuracy for certain applications.
Beta Particle Attenuation Monitors
When beta particles impinge on matter, some are absorbed,
some are scattered, and some are transmitted. The reduction in
the incident beam intensity is known as beta radiation attenua-
tion. The attenuation is primarily a function of the beta particle
energy, the .amount of the attenuating matter in the radiation path,
and the electron density of the matter. The electron density
is the ratio of the atomic number (number of electrons per mole-
cule) to the atomic weight (mass of the molecule). This ratio
is essentially the same (between 0.45 and 0.51) for most elements
below the very heavy ones. Hydrogen is an exception, but its
presence does not usually cause a significant error. Because
the electron density for most of the elements is almost the same,
beta attenuation is practically independent of the chemical compo-
sition of the absorber. Beta attenuation Is considered by many
engineers and scientists to be a direct measure of mass.
24
-------
The relationship between the attenuation and the absorbing
mass is approximated by the exponential function
I/IQ = exp -UmX (1)
where I is the incident beam intensity, I is the transmitted beam
intensity, X is the absorber area density in mg/m2, and y is a
function of the maximum energy of the beta particle. For a given
absorber, the value of X is increased by increasing the thick-
ness of the absorber. The greater the thickness, the greater the
mass area density (mg/m2) and consequently the greater the elec-
tron area density.
Both gases and particulate matter attenuate beta radiation.
Therefore, in situ application of beta attenuation as a measure-
ment of mass concentration is not feasible because the mass of
gas molecules in a gas stream far outweigh the mass of the parti-
culate matter, and beta attenuation methods cannot discriminate
between the two. Consequently, beta attenuation monitoring methods
require an extractive sampling system.
The major component of the beta radiation attenuation moni-
tor, or simply beta monitor, is the beta sensor. The beta sensor
consists of a beta source and a beta detector. The particles
to be measured are placed between the source and the detector.
Usually, the particles are collected on a filter tape. The beta
radiation passing through the filter tape is measure before the
particles are deposited, for a zero measurement, and after par-
ticle deposition. The difference in these measurements is a mea-
sure of the mass of the particle deposit. Measurement of the
air volume from which the particles were collected results in
the determination of the particulate mass concentration. After
each measurement cycle, the filter tape is advanced to provide a
clean spot for particle deposition.
25
-------
Instrument Development—
Many specific instrument designs using various combinations
of beta sources, beta detectors, and particle collection devices
have been previously described and tested in industrial and ambient
environments. These combinations include a prototype model using
a C11* source, Geiger Mu'ller (GM) detector, and an electrostatic
precipitator collection device;35 a stack model36 and a vehicle
emissions model37 which was tested on a stack,38 using a C14 source,
GM detector and filter collector; an ambient model using a Cll+
source, proportional counter detector, and filter collector;39'4°
a vehicle emissions model using a C1'1* source, scintillation de-
tector and filter collector;1*1 an ambient model using a C11* source,
GM detector, and impactor particle collector; **2 a stack model using
a Pm11*7 source, solid state detector, and filter collector;1*3
a stack model using a Pm11*7 source, GM detector, and a combina-
tion cyclone-filter collector j1*1* and a model using a GM detector
and cyclone.1*5 The model that uses a cyclone-filter combination
is capable of sampling isokinetically on an automatic basis.1*1*
Lilienfeld has reviewed the application of 3-monitors to indus-
trial dust measurements, ambient monitoring, and particle size
analysis. His review contains discussions of the principle, prob-
lems, and errors associated with the 8-technique. **6
Recent 6-monitor systems designed for stack applications
include computerized beta monitoring systems. Two such systems
are manufactured by RAC Corp., Pioneer and Hardies Road, Gibsonia,
PA 15044, and Lear Siegler, Inc., One Inverness Drive East, Engle-
wood, CA 80110
The computerized monitoring systems are composed of the
following major units: (1) sampler, (2) particle measurement,
26
-------
(3) gas conditioner, (4) gas volume measurement, and (5) control
and computer units. The RAC unit is shown in Figure 3.
Sampler units consist of a nozzle of suitable size, and a
probe, which in the RAC system is fixed, and in the Lear Siegler
system is optionally stationary or is motorized for traversing
the stack. Probes are generally heated to keep the gas above
the dew point. A boundary layer diluter, where provided, condi-
tions gas streams containing high levels of temperature, humidity,
or particulate concentration. The diluter adds a controlled volume
of dry air into the probe. These systems automatically maintain
a constant flow rate; however, at present, they do not automatically
maintain an isokinetic flow rate.
The particle measurement unit consists of a heated nozzle
or air tight filter holder. The unit also contains the drive
and indexing system so that the filter tape can be moved forward
and backward for zero and sample counts and for sample collecting.
The tape movements are commanded by the computer, and data from
the beta counts are stored in the computer memory. The source
used is a C11* radionuclide with a half-life of 5730 years and a
radiating strength of 100 microcuries.
The gas conditioner contains a cooling and dehydrating module
to condition the gas before it enters the gas measurement unit.
Automatic, periodic back-flushing of effluent sample lines is
provided to clean the lines of any residual particles.
The control and computer unit is the main feature that dif-
ferentiates the computerized beta monitoring systems from their
predecessors. In the system, the minicomputer sequences all units
as well as individual component functions. It receives measure-
ments of the sample gas parameters, beta counts, and other per-
tinent data, such as diluter volume. The minicomputer uses this
27
-------
PROBE &OILUTER (Optional)
Sample Flow
Outlet
Orifice
Beta Radiation Gauge
C14 Radiation Source
Pressurizing Air Line
Dilution Air Line —P
DEHYDRATION MODULE
(Refrigerated Condenser)
Water/Condensate Discharge
lenoid Valve
Purge/Back-Flush Air Line
Control Station
can be located
up to 250' from
Sampling Module
MASTER CONTROL t MINI-COMPUTER MODULE
Beta Counter Volume Counter
Tape Printout
Electric
flow
Control
Valves
Flow
Rota-
meters:
#1
Sample
Stream
#2
Dilution Air
#3
Purge/Back-
Flush Air
3630-205
MINI-COMPUTER
CONTROL CONSOLE
* (Measures Sample Volume) ""-Exhaust
»* (Measures Dilution Air Volume)
Figure 3. Schematic flow diagram of a typical RAC Automatic
Stack Monitor System installation. (Drawing not to
scale.) Used by permission.
28
-------
information and equation 1 to compute the mass concentration for
conditions of standard temperature and pressure. The results
are furnished on a printed type readout. In the Lear Siegler
system, the computer only processes the beta counts to produce
a signal proportional to the collected mass. A strip chart re-
corder is used for the mass readout. Mass concentration must
be calculated from the measured flow conditions. Calibration
of the monitors is performed in the factory; however, the Lear
Siegler system has accommodations for gravimetric calibrations
in the field.
Performance—
First it should be noted that very few ^-monitors are in ser-
vice as process monitors, and very little quantitative information
is available on their performance.
Possible sources of errors are loss of sample in the probe,
variation in filter thickness, nonuniform deposition of dust,
losses in filter efficiency, statistics of radiation counting,
deviation from the attenuation law (Equation 1), and uncertain
flow measurement.
Beta attenuation is not solely dependent on the ratio of
atomic number to atomic weight, but is also partially dependent
on the atomic number. Because of this dependency and because
the atomic number to atomic weight ratio varies .between 0.45 and
0.51, the beta monitors are somewhat sensitive to the chemical
composition of the absorbing matter. One study compared the ab-
sorption characteristics of aluminum and membrane tape and found
them to be significantly different.39 On the other hand, a dif-
ferent study gave data on a variety of absorbers which are in
particle form. These included soot, fly ash, cement, dust,
29
-------
gypsum, and open burning coal. This data indicated that these
absorbers have very similar absorption characteristics.45 There
was a 10% uncertainty in the data and therefore this study cannot
be considered conclusive.
It has been shown that the geometry of the apparatus (e.g.,
the source window thickness, the distance from source to detector,
filter tape thickness) significantly affects the amount of beta
attenuation that takes place for a given absorber thickness.
This problem can be partially handled by calibration. One study
demonstrated that the variation in thickness of the filter ma-
terial has a significant effect on the calibration of the sen-
sor.47 This effect was attributed to the nonexponential character
of the beta radiation attenuation. It was calculated that an
instrument, using a C1" source, that does not take into account
the nonexponential character of beta radiation attenuation will
yield a mass error of about 8%, assuming that the unknown filter
variations are limited to 1 mg/cm2.
First generation models could be used only in process streams
which were less than 170°C, which is the operating temperature
of the glass fiber filter tape. The computerized models have
a range of up to 538°C (1000'F) with the use of the sample diluter
and conditioners.
The computerized models are best used in gas streams which
contain particle sizes predominently less than about 5 ym. This
is due to the inability of the instrument to sample isokineti-
cally on an automatic basis and probe losses. The monitor using
the cyclone-filter combination with automatic isokinetic sampling
ability1*.1* would not have these limitations.
30
-------
The response time consists of the time required for initial
filter tape count, transportation of the filter tape to and from
the particle collection location, particle collection, and final
filter tape count. The counting and transportation take around
3 minutes in all beta monitors. Particle collection using a
filter can take anywhere from 5 to 10 minutes. Particle collec-
tion time will depend on the particle concentration of the process
stream, minimum mass detection levels of the sensor, and the type
of particle collection employed. Monitors using an impactor would
require a relatively short collection time due to its ability
to deposit particles in a concentrated area on the impaction sur-
face. Models using a cyclone would have a short collection time
because of the ability to use high flow rates with a cyclone.
Early experience with first generation monitors showed an
accuracy of ±10%, not including the error due to losses in the
probe. It included only those errors generated by the instrument
alone; such as varying tape thickness, statistical variation in
count rate, particle collection efficiencies less than 100%, etc.1*5
Two first generation monitors were tested on a coal-fired
power plant.^ For these tests, the instrument reading was com-
pared with the manually measured gravimetric concentration for
concentrations ranging up to 250 mg/m3. Results yielded correla-
tion coefficients 0.94 for 24 sampling runs of the A monitor with
a 22% variation at a 150 mg/m3 concentration for a 95% confidence
level. The B monitor achieved a 0.81 coefficient for 43 sampling
runs and a 31% variation.
Two prototype beta monitors (those described in references
37 and 41) were extensively evaluated in a particle sampling fa-
cility attached to a coal-fired power plant.38 The ratio of the
mass measured by the instrument to the mass on the independent
31
-------
reference filter was obtained for many runs. One model gave a
ratio of 0.98 with a standard deviation of 0.04. The other model
gave a poorer performance with a ratio that varied from 0.43 to
0.61 and a standard deviation that varied from 0.54 to 0.073.
The explanation given for this factor of two difference in the
measured ratios was attributed to problems within the latter in-
strument and not in the sampling techniques.38
Recent studies on application of beta monitors include coal-
fired and lignite fuel power plants, cement plants, and ferro-
alloy plants.36 For the coal-fired power plant, the particle
concentration determined by the monitor was compared with the
particle concentration determined by weighing the mass deposited
on the filter tape. Results produced a confidence range of 30%
for the individual measurements and a confidence range of 5% for
the mean values. Tests at the lignite-fired power plant compared
the instrument's reading with the mass concentration measured
gravimetrically up to concentrations of 200 mg/m3 STP. Results
gave a standard deviation of up to 16% for the individual readings
It also was found that after an operating period of three months,
no dust could be detected in the probes.
Other data, however, on applications to a ferro-alloy plant,
and an oil-fired power plant1*9 have indicated probe losses ranging
from 30% to 86% of the total catch determined by Method 5. Cor-
relation between the EPA Method 5 and the beta monitor varied
from 0.14 to 0.80.
Initial experience with a computerized model that is now
installed in several locations has shown that it will function
with a variation of only 1 to 5% from values obtained with EPA
Method 5.50 This model offers, as an option, a sample diluter
probe consisting of a fixed probe which is filled with a porous
inner liner. The dilution air is forced through the walls of
32
-------
the inner liner and provides an effective boundary layer which
minimizes probe deposits and also conditions the sample. A bound-
ary layer interface system for extractive sampling has been developed,
which is similar to the diluter probe.51 There are plans to evaluate
this interface system with a beta monitor.
Summary—
Beta monitors of several designs have been tested on indus-
trial sources over the past ten years. Advantages include a sensing
principle that is very closely correlated to mass and independent
of particle composition, low sensitivity to particle and aerosol
parameters other than mass, and a movable filter tape which makes
it convenient for performing chemical analysis concurrent with
sampling.
Disadvantages include a response time longer than some other
monitors, the need for an extraction/dilution system, and a sensi-
tivity to variations in filter tape thickness.
Piezoelectric Mass Monitors
Piezolectricity is a property of certain crystals, such as
quartz, which involves the production of an electrical charge
on certain faces of the crystal when the crystal becomes mechani-
cally stressed. The converse process also occurs; that is, a
piezoelectric crystal becomes mechanically stressed where an elec-
trical charge is placed on certain faces. This two-way capability
is responsible for the ability of a piezoelectric crystal to cause
an oscillating electric circuit to resonate at the natural vibra-
tional frequency of the crystal.
When foreign material adheres to the surface of a vibrating
piezoelectric crystal, the natural frequency of vibration of the
crystal decreases. The magnitude of the frequency change is di-
33
-------
rectly proportional to the mass of the added material. This was
first shown theoretically and experimentally by Sauerbrey for
thin metallic films.52 This relationship also holds for any ma-
terial which sticks to the surface and vibrates with the crystal
The relationship is:
AM = - Af
where
AM = mass added to the electrode area (yg)
Af = change in natural vibrational frequency (Hz)
A = electrode area of the. crystal (cm2)
fo = natural vibrational frequency of the crystal (Hz)
A schematic of a quartz crystal transducer is shown in Fig
ure 4. The particles are deposited onto the crystal surfaces
with a collection device, such as a small electrostatic precipi
tator chamber or an impactor.
There are four common vibration modes for quartz crystals.
Most commonly used in piezoelectric monitors is the thickness-
shear mode, which is created by cutting the crystal so that the
crystal axes are oriented in such a way that the crystal will
vibrate as shown in Figure 5.
A typical piezoelectric monitor consists of the following
components as shown schematically in Figure 6.
1. Primary crystal used to sense deposited particles.
2. A particle collector.
3. A pump to draw the aerosol through the collector.
34
-------
PARTICLES
PARENT
COLLECTION
DEVICE
FORCE FIELD
GENERATED BY
PARENT DEVICE
lUARTZ CRYSTAL
METALLIC
ELECTRODES
PIEZOELECTRIC
OSCILLATOR
CIRCUIT
FREQUENCY
MONITOR
3630-206
Figure 4. Schematic system for transducing particle
quartz crystal oscillator. After Sem, et al.
using
35
-------
PARTICLES
T
3630-208
Figure 5. The thickness-shear mode of oscillation for a quartz
crystal. After Sem, et al 53
PRIMARY
CRYSTAL,
REFERENCE
CRYSTAL
PRIMARY^
OSCILLATOR
CIRCUIT
REFERENCE
OSCILLATOR
CIRCUIT
MIXER
FREQUENCY
COUNTER
UIKfcCT
MASS
CONCENTRATOR
INDICATOR
J
FREQUENCY
3630-207
Figure 6. Piezoelectric microbalance electronic block diagram.
After Sem, etal.53
36
-------
4. Reference crystal used to subtract out possible frequency
changes caused by changes in gas temperature and humidity,
5. Oscillator circuit for both primary and reference crystal,
6. Mixer circuitry to subtract the signals from the two
crystals.
7. Digital frequency counter to monitor the mixer output,
which is proportional to the total mass adhering to the
crystal.
and/or
8. Apparatus to compute and to provide direct readout of
mass concentration.
Both precipitators and impactors are used as particle col-
lectors; examples of each are shown in Figure 7. Equation 3 gives
the equation relating the mass concentration to the frequency
change.
r _ i. Af
C ' SQ At (3)
where:
C = mass concentration in yg/m3
Af = change in mixer frequency (Hz)
Q = sampled aerosol flow rate (m3/sec)
S = Af/AM theoretical mass sensitivity, Equation 2 (Hz/yg).
Thus, the mass concentration is proportional to the time
rate of change of the frequency. Direct readout of mass concentra-
tion can be achieved by converting the frequency change to a volt-
37
-------
1. AIR INLET
2. AIR OUTLET
3. TEST CRYSTAL
4. REFERENCE CRYSTAL
5. PRECIPITATOR ELECTRODE
6. AEROSOL-CORONA CONTACT ORIFICE
7. IMPACTION ORIFICE
8. BY-PASS AIR ORIFICES
9. THERMISTER POSITIONS
3630-209
Figure 7. Two types of particle collectors for piezoelectric monitors.
A. Electrostatic precipitation. B. Impaction. After Daley
and Lundgren.56
38
-------
age. Then, use of circuitry to differentiate the voltage with
respect to time will produce a signal proportional to the mass
concentration. This signal can be applied to a strip chart re-
corder.
At least two commercial instruments are available, based
on the piezoelectric principle, for monitoring particulate prop-
erties. One is a mass monitor manufactured by TSI Incorporated
(500 Cardigan Rd., St. Paul, MN 55165). Another instrument is
a cascade impactor, discussed in' Section 4 of this report.
Neither of these devices is designed to operate at mass concen-
trations as high as those typically found in process streams.
The TSI device, however, has been compared to filter methods for
measuring the concentration of several aerosol materials, and
shown to yield an accurate measure. The most important appli-
cation of the TSI device is monitoring worker environments for
respirable dust concentrations.
Performance—
Most of the work with piezoelectric monitors conducted prior
to 1970 focused on thin film measurement and gas concentration
measurement. King has noted a list of references which deal with
quartz crystal applications ranging from measuring dew points
to detection of hydrocarbons and sulfur compounds.55 These are
designed for ambient air monitoring.
Piezoelectric monitors designed for the purpose of monitoring
particulate mass concentration in industrial process streams or
stacks have not been developed, and. the development of an in situ
piezoelectric monitor is not forseeable based on the state of
current technology. However, it is likely that the ambient piezo-
electric monitors will be modified or adapted for stack use, per-
haps through the use of improved sample extraction/conditioning
39
-------
systems. Thus, information pertaining to the performance charac-
teristics of the ambient monitors would provide insight into the
problems that are likely to be encountered when extending the
ambient monitors to handle process streams. Tests have shown
ambient monitors to be sensitive to a number of factors other
than mass concentrations, such as, temperature, humidity, particle
size, and type of particle collector used. The monitor is not
sensitive, per se, to the composition of the participate matter.
Temperature-Quartz crystals are somewhat sensitive to tem-
perature changes. Because the mass concentration is a function
of the time rate of change of the frequency, a time rate of change
of the temperature will be measured as an apparent change in mass
concentration.
Tests on two monitors, one using an electrostatic precipi-
tator as the collection device and the other using an impactor,
showed that temperature changes of 0.3 to 1.0°C/min would produce
an error of 5 Mg/m=. This ls usually inslgni£lc£mt ^
with the relatively high mass concentration ranges sampled by
the monitors, it was also found that the use of a reference
crystal did not compensate for inlet temperature changes because
only the test crystal exposed to the temperature fluctuations
observed at the inlet.56
Humidity-There are two types of errors attributable to
changes in the relative humidity. One is associated with the
crystal and its electrode, and the other stems from the hygro-
scopic nature of the aerosol deposit.
It was found that the platinum electrode provided with the
electrostatic precipitator collector showed wide variation in
humidity response. This variation was due to the corona action
on the test electrode. For this unit the problems became severe
40
-------
for relative humidities above 30%. As with the temperature re-
sponse, a time rate of change of the relative humidity produced
a change in the apparent mass concentration. The hygroscopic
character of the aerosol deposit on the test crystal greatly af-
fects the ability of the reference crystal to compensate for rela-
tive humidity changes in the air stream. For many materials en-
countered in ambient aerosol sampling, it was found that relative
humidity changes of less than 0.2%/min can cause mass changes
that are greater than the change associated with the accumulating
aerosol particles.56 Condensed water vapor would also cause a
problem.
Particle collection characteristics—For spherical particles,
the mass sensing ability of 5 and 10 MHz AT-cut quartz crystals
begins to decrease at a particle diameter of approximately 2 ym
and reaches zero at 20 ym. However, good agreement between cal-
culated and experimental sensitivity values for a polydispersed
deposit with a 2.5 ym MMD indicated that irregular particles con-
siderably larger than 2 ym are probably sensed when present in
polydisperse deposits.56
The reason for this lack of sensitivity beyond 2 ym lies
in the fact that in order for a particle to be sensed, it must
adhere perfectly to the crystal and vibrate with the crystal.
As the particle size increases, the ratio of particle adhesive
force to inertial forces decrease. Thus, the mass of particles
above certain critical sizes will not be sensed with 100% effi-
ciency. The critical size is increased by depositing particles
uniformly onto the crystal surface and by minimizing crystal drive
level. Application of a thin coat of sticky material to the crys-
tal surface is another method. However, problems can arise with
the absorption of gases by the adhesive coatings.
41
-------
Mass sensitivity is not uniformly distributed over the active
area of the crystal. The mass sensitivity is greatest at the
center and decreases towards the edge of the active area. In
cases where the deposit area is larger than or conterminous with
the active area of the crystal, the crystal mass sensitivity (--
is constant. However, when the deposit area is less than the
active area, the crystal .mass sensitivity is a function of the
location and size of the deposit.
In tests of the electrostatic collector, it was found that
the diameter of the deposit area was 5.3 mm for 0.2 ym diameter
particles and decreased to 2 mm for 20 ym particles. This decrease
in deposit area size with increasing particle diameter is due
to the higher electrical mobility of the large particles.56
This creates different sensitivities for different particle size
deposits. The impactor version had a more constant deposit area
diameter for different particle size. The use of an impactor,
however, usually results in a large number of small particles
(typically <0.5 ym) being lost which are smaller than the cut
off point for the impactor.
Linear response limit—Sometim^ during sampling, a point
will be reached when additional mass deposited onto the crystal
surface will not be sensed at 100% efficiency and will result
in a non-linear response. This is due to the additional particles
no longer vibrating in unison with the crystal.
It was found that some material showed non-linearity almost
from the start of sampling; this may necessitate very short sampl-
ing times. The monitor with the impactor had one typical fre-
quency change vs. accumulated mass curve, whereas, the monitor
with the electrostatic collector showed a wide variety of fre-
quency change vs. accumulated mass curves for different particle
sizes. The reasons given for this variety of curves for the
42
-------
electrostatic collector were: agglomeration of small particles
on the crystal surface, shifts in the region of deposition due
to changes in the electric field caused by the initial annular
deposit, and evaporation of volatile compounds under the action
of the corona. Concentration-sampling time products which would
produce a 20% deviation from linearity were found to be from 100
to 6,000 min-yg/m3 and 5,000 to 60,000 min-yg/m3 for the impactor
collector and electrostatic collection respectively.56 It is
therefore important to determine the linear response limit by
calibrating the piezoelectric monitor with the aerosol to be
sampled.
Considerations for stack application—The excessively high
levels of temperature, mass loading, and humidity present in pro-
cess streams and stacks make the ambient monitor unusable for
such applications unless the sample is cooled and diluted. Two
ambient studies have been performed which tested the performance
of ambient monitors.56'57
High temperatures tend to increase the crystal temperature
coefficient to an unacceptable level. Typical operating tempera-
tures for piezoelectric monitors range up to 66°C (150°F), al-
though it may be possible to increase the operating temperature
by the selection of a different crystal cut type. Typical stack
temperatures are in the range of 149 to 232°C (300 to 450°F).
Problems with sensing large particles might be alleviated
by the use of a low frequency crystal or one with a different
vibration mode.
High mass loading results in unrealistically short sampling
times. Typical mass loading limits for proper operation of these
monitors can range from 2 to 20,000 yg/m3. Thus, the upper con-
centration limit of the monitors falls where the mass concentra-
tion range usually begins at the outlet of a high efficiency
43
-------
control device on a process stream. High mass loading could be
handled with the use of a proposed double sampling diluter, such
as the one shown in Figure 8. in the double sampling diluter,
the large sampling probe extracts a sample at isokinetic condi-
tions and at a high flow rate. The required flow rate (usually
around 1 liter/min) for the monitor is acquired by a second iso-
kinetic sampling of the first sample. Like all extractive me-
thods, this would suffer
High humidity and temperature conditions could be handled
with the use of sample conditioning, such as, the addition of
a measured amount of clean, dry air to the sample. An example
of such a conditioner is being used with a beta monitor discussed
in the section on Beta Attenuation Monitors.
Summary—
Piezoelectric monitors have had no applications in sampling
industrial process streams, nor are there any prototype monitors
known to be designed for this purpose. However, they have been
used for ambient and automobile emissions monitoring and show
promise as process stream monitors. Advantages include a sensing
Pnndple that relates directly to mass and which is independent
of particle composition, and yields continuous, instantaneous
("real-time") measurements.
Disadvantages include a need for an extraction/dilution sys-
tem, sensitivity to changes in gas temperature and humidity (de-
creases for particle sizes greater than 2 Mm), the necessity of
periodic cleaning to prevent non-linear sensor response, and the
need for calibration for the aerosol to be sampled to determine
the linear response limit.
44
-------
SAMPLING
PROBE
(I
SAMPLING
PROBE
n
SAMPLING
CONDITIONER
OSCILLATOR
CIRCUIT
FLUE
GAS
. DUCT
WALL
PARTICLE
COLLECTION-
REGION
QUARTZ
CRYSTAL
FLOWMETER
VAVLE)M
PUMP
APPROXIMATELY
150 LITERS/MINUTE
PUMP I
APPROXIMATELY
1 LITER/MINUTE
3630-210
Figure 8. A possible stack sampling system using a proposed double
sampling diluter and a piezoelectric microbalance sensor.
After Sem et al.53
45
-------
Charge Transfer
The phenomenon of transfer of a negative electric charge
is observed when two bodies of different composition come into
contact. The transfer can occur during either static contact
or triboelectric (rubbing) contact. The mechanism of transfer
in static contact is essentially the same for both metals and
semiconductors.58'59 Upon contact, a flow of electrons is ini-
tiated due to the different contact potentials of the materials.
This flow will continue until the build-up of charge produces
an electric potential that is equal and opposite to the difference
between these contact potentials. The theory for the static con-
tact charging of insulators is now well developed and the charge
transfer mechanism is thought to be different from that of metals
and semiconductors. In the charge transfer process, it is not
known which type of interaction (static or triboelectric) actually
takes place; probably both occur. In the case of insulators
particularly, triboelectric effects along with electrolytic ef-
fects on the moist surfaces of particles probably play large
roles.
Instrument Development—
In all charge-transfer instruments, the aerosol stream is
forced to collide with a sensor. When the particles in the aero-
sol stream contact the sensor a charge is transferred producing
a current that is continuously monitored with an electrometer.
The charge-transfer instruments that have been used for mea-
suring the concentrations of aerosols differ mainly in the shape
and construction of the sensor. Two instruments, one with a
spherical metal sensor, the other with a sensor-in-nozzle design
are described by Schutz.60 The Konitest design of a bullet shaped
sensor is discussed by Prochazka.61'62 Kony Company Ltd, a Japa-
nese concern, has just recently developed an automatic continuous,
46
-------
dust concentration meter, Konytest (patterned after the German
Konitest model) under a technical agreement with Dr. Prochazka
of West Germany.63 No performance data or specifications for
this installation are known at this time. An in-stack model has
been developed in the USSR.61* The IKOR Air Quality Monitors (IKOR
Inc.) P.O. Box 660, Blackburn Industrial Park, Gloucester, Mass.
01930, use a variety of sensor materials and configurations,
depending upon the nature of the particles to be monitored and
their gaseous environment. The IKOR instruments are the only
instruments commercially .available in this country and are avail-
able in three models. Models 206 and 207 are extractive; Model
2710 is the newly developed in situ monitor.
Model 206 consists of three components: a probe, a sensor
unit, and a control unit. In monitoring, the probe is inserted
into the stack, where it simultaneously extracts a continuous
sample and measures the stack gas conditions. The sample flow
rate is monitored by a venturi attached to the probe. Isokinetic
sampling conditions are maintained manually. The sample is con-
ducted through the probe and flexible hose which are both elec-
trically heated, and carried to the sensor unit. Collisions be-
tween particles in the gas stream and the sensor produce an elec-
tric current that is proportional to the particulate mass flow
past the sensor per unit time and is thus a measure of the in-
stantaneous mass concentration. The current is electronically
processed to produce an output voltage that is registered on the
(particulate) mass flow rate meter. A strip chart recorder pro-
vides a permanent record of the output signal with time. An in-
tegrator is used to automatically sum the area under the output
curve and thus produce a measure of the total transferred charge
during the specific period of operation. This total transferred
charge can be compared to the filter weight gain during the in-
itial period of operation for purposes of calibration.
47
-------
Model 207 is essentially the same, except for the Integrator/
Test Set, which is incorporated into the control unit. Once
calibrated on a particular source, Models 206 and 207 can be ope-
rated with or without the filter. Model 2710 consists of a probe
with a sensor attached at its end for in situ sampling, and a
control unit. No pump is used on the in situ system, thus, the
impact velocity of the particle is just the velocity in the dust.
Performance—
There are other factors, beside mass, that can affect the
amount of charge transferred to the sensor from particles in any
given process stream. Some of the possible factors are the chemi-
cal composition of the particles and the sensor material, condi-
tion of the sensor surface, particle size, and particle charge.
This sensitivity to factors other than mass can result in erro-
neous readings and frequent need for recalibration. The extent
to which these factors affect the instrument's response is dis-
cussed in a laboratory study of Model 206 conducted by Walter
John65 and in additional sources such as field test reports66"70
and company sales literature. The information applies to all
models unless stated otherwise.
The sensitivity of the instrument is defined as the amount
of charge transferred to the sensor per unit mass, expressed in
yCoul/g. John found the sensitivity to be constant for runs in
which the total sampled mass of aluminum oxide varied from 10
mg to 100 mg; similar results were obtained for aluminum dust.65
Field tests also have shown this constant sensitivity for aluminum
oxide. The range of mass concentrations employed for these tests
is not available. Company sales literature gives the mass con-
centration range for Models 207 and 206 as 0.000023 g/DNCM (0.00001
gr/SCF) to 23 g/DNCM (10 gr/SCF) and 0.000023 g/DNCM (0.00001
gr/SCF) to 230 g/DNCM (100 gr/SCF) respectively; the 2710 is given
48
-------
as 0.00023 g/DNCM (0.0001 gr/SCF) to 230 g/DNCM (100 gr/SCF).
These values are dependent on actual stack conditions. Actual
ranges obtained under typical stack conditions for coal-fired
boilers are typically from 0.012 g/DNCM (0,0005 gr/SCF) to 115
g/DNCM (50 gr/SCF). 66'67
The sensitivity varies with the composiion of the particles
and the sensor surface, requiring that the instrument be recali-
brated whenever the particle composition changes. John reported
the results of sensitivity measurements on 13 different compounds,
many of which are found in stack effluents. He found that metals
produced the highest sensitivities, followed by semiconductors,
then insulators. Metals have the lowest electrical resistivities,
semiconductors have higher resistivities, insulators have the
highest resistivities. This suggests that the sensitivity is
inversely proportional to the electrical resistivity of the par-
ticle. This result also can be derived from charge transfer
theory. The ratio of highest sensitivity to lowest sensitivity
is 60 to 1. This wide range makes the instrument unreliable in
process streams which contain variable mixtures of metals, semi-
conductors, and insulators. Liquid droplets also produce a signal
and therefore must be considered. Company literature gives the
particle diameter size range of their models as 0.1 to 100 ym,
depending on operating conditions.
6 6
Tests have shown high sensitivity to submicron particles.
John, however, found a complete lack of response to tobacco smoke.
He found similar results for a polydispersed and a 0.5-6 ym sample
of aluminum and a 0.5-3 ym and a 3-8 ym glass bead sample.
Corrosion will affect the response by changing the chemical
composition of the sensor surface, by producing electrochemical
reactions, and by breaking down the sensor insulation. The sen-
sors can be obtained in a variety of materials to minimize these
49
-------
effects. Liquid droplets can cause signal loss through leakage.
Smearing of the sensor surface by buildup of wet, oily, or waxy
particles will affect the reproducibility of the signal. This
can be minimized by keeping the sample temperature at or above
the stack temperature and by frequent cleaning of the sensor.
In normal operation with discrete, dry particles, a slight ac-
cumulation of dust on the sensor is observed, although, presum-
ably, an equilibrium condition is established.
If a particle is charged, as for example by passing through
an electrostatic precipitator field, it will transfer its charge
to the sensor upon contact, thus producing an erroneous reading.
Knapp reported that non-uniformity of charge distributed among
the particles in the stack was a problem for in situ devices.
In fact, one of the advantages reported for the extractive model
was the tendency of the hose to distribute this charge more evenly
through collisions of the particles with the hose wall. In one
instance, a situation was reported where the in situ model was
monitoring the outlet of an electrostatic precipitator on a coal-
fired boiler. The device showed a decrease in mass concentration
during rapping, instead of the expected increase recorded by the
transmissometers.68
Model 206 has been tested with an EPA Method 5 train.69
The results were favorable and indicated good agreement between
the two methods. Replicate runs performed at two different alumi-
num reduction baghouses produced calibration factors with standard
deviations of ±3% and ±7.6%.66 in a test performed on a dry
scrubber system, the results obtained with two identical 207
models were in good agreement.70
Summary—
Charge transfer monitors have been used on industrial sources
for over 14 years. Advantages include in situ or extractive sampl-
ing and continuous, instantaneous, real-time measurements. Dis-
50
-------
advantages include indirect measurement of mass; strong dependence
on chemical composition of the particles; sensor sensitivity to
particle size (suspected lower size limit), water droplets, corro-
sive gases, and particle charge; and degradation of sensor per-
formance when exposed to wet, waxy or sticky particles. This
last disadvantage would hamper usage at combustion systems fired
with residual oil. Sources with electrostatic precipitation pre-
sent precharging problems, as discussed. In conclusion, the IKOR
monitor performs best when applied to the situation where process
stream conditions are constant or change predictably, and which
contain dry, discrete, uncharged particles.
Optical Methods
Conventional Transmissometers—
The basic function of a conventional transmissometer is to
measure opacity. Although opacity alone is not a direct measure-
ment of mass concentration, it can be a good relative measurement
if the optical properties and size distribution of the aerosol
particles remain constant. In the following discussion, typical
results and the necessary considerations are delineated to relate
opacity to mass concentration. For a description of the hard-
ware and techniques involved in the measurement of opacity the
reader is referred to Section 3 entitled "Opacity".
Light scattering theory predicts a dependence of light attenu-
ation on not only mass concentration but also on particle size
and composition. Figure 9 shows the results of applying this
theory to calculate the effects of various particle sizes and
composition on the relationship between the opacity and mass
concentration of aerosols.71 At particle diameters above 3 or
4 ym the refractive index of the particle plays little role in
determining the opacity-mass concentration relationship. How-
ever, at particle diameters below 3 or 4 pm, the refractive index
plays a major role.
51
-------
D
_l
Q.
ac
ui
co
>
o
0
10
20
30
% 40
50
60
70
80
* \\
\f \
\
\
\
•
\
\
\
- \
o
'->
•p.
\
\\
\
\
V
-A
rn
O
•»
o
\03
r
\
V
\%
I
V
A
LOG NORMAL DISTRIBUTION
STANDARD DEVIATION CTg = 4
PARTICLE DENSITY = 2 gram/cm3
WAVELENGTH = 0.55 A<
REFRACTIVE INDEX
WHITE = 1.5
1.96 - 0.661
' ^
\^,
0.1 0.2 0.3
MASS CONCENTRATION, g/m3
\
0.4
0.10
0.20
0.30
0)
ui
O
LL
U.
Ill
O
O
z
o
<
D
z
LLI
0.40
0,
3630-211
0.50
5
Figure 9. Opacity of smoke plumes containing particles of different
sizes and refractive indexes as a function of their mass
concentration. After Connor.71
52
-------
The particle size dependency of the opacity-mass concentra-
tion relationship has been studied experimentally in the labora-
tory by Uthe and Lapple.72 Fly ash from a bituminous coal-fired
power plant was collected and then classified into a series of
size fractions. These size fractions were pneumatically injected
into an aerosol chamber at controlled concentrations which were
calculated from aerosol generation rates. The opacity-mass con-
centration relationship was measured with a transmissometer and
is shown for four different particle size ranges (see Figure 10).
Comparison of the fly ash tests with the theoretical calcula-
tions presented in Figure 9, shows that, for similar particle
sizes, the fly ash attenuated about 50% as much light as the theo-
retically calculated values for spherical particles with a refrac-
tive index of 1.5 and a density of 2 gram/cm3. This difference
can be attributed to differences in aerosol characteristics be-
tween the fly ash and the particle properties assumed in the theo-
retical calculations. The fly ash was reported to have a size
distribution with a = 1.5 and contained black, absorbing particles;
also, the density of 2 gram/cm3 used in the calculations is low
for fly ash.
For a transmissometer to be useful as a monitor of the mass
concentration, the properties (other than mass) of the particles
being monitored must remain fairly constant over the monitoring
period. Experimental data are available showing that good opacity-
mass concentration calibration can be obtained on some sources.
The sources evaluated include coal-fired plants;53'58'73 lignite-
fired power plants;71* cement plant;75 Kraft pulp mill recovery
furnace;76'77 petroleum refinery, asphaltic concrete plant, sewage
sludge incinerator, brass smelter, lead smelter, oil fired power
plant,78 and hog fuel boiler.77
53
-------
80
0.1
0.10
03
0.20
— 0.30
0.40
o
o
UJ
I-
0.50
°-2 0.3 0.4 0.5 0.6
MASS CONCENTRATION, g/m3
0.7 0.8
3630-212
Figure 10. Opacity-mass concentration relationship of laboratory
generated coal-fired power plant emissions with different
particle sizes. After Con not-.71
54
-------
The test results for a normally operating coal-fired power
plant gave correlation coefficients for two different instruments
at 0.63 and 0.87 for 300 measurement runs, with a measurement
tolerance of ±68% and ±39% respectively at a concentration of
150 mg/m3 and a confidence interval of 95%. The instrument showed
different correlation curves for conditions of minimum load opera-
tion and operation with and without soot blowing.lf8
Nader reported tests that were performed over one 3-month
interval and two 2-month intervals spanning a one-year period
representing different seasons of power plant operation.79 Emis-
sions were increased at various times by cutting off one or more
electrostatic precipitator stages. Correlation curves were es-
sentially the same for the three different time periods with co-
efficients of 0.93, 0.98, and 0.99. The coefficient for the com-
posite correlation curve for the data for all three time inter-
vals is 0.97 (see Figure 11). Mass concentration ranged from
55 to 360 mg/m3. No problem with window contamination occurred
with continuous operation of the transmissometer spanning the
one-year period.
Comparative measurements at lignite-fired power plants7 **
gave results which were not as good as those obtained at the coal-
fired power plant, and showed a different calibration at the be-
ginning and end of a 3-month test period.
Comparative measurements were conducted on emissions from
the clean gas duct of a rotary cement kiln with a suspension gas
preheater and a subsequent rotary dryer.75 The results showed
that all the operating conditions can be satisfactorily described
by the common straight regression line given in Figure 12. The
coefficient of correlation is 0.984 and a RMS deviation of cor-
relation of 17 mg/m3.
55
-------
0-1 0.2 0.3 0.4
MASS CONCENTRATION, gm/m2
3630-213
Figure 11. Correlation data between opacity and mass measurements
of particulate matter in emissions from a coal-burning
power plant. After Nader.79
-------
0.50
0.40
0.30
I
t-.
0.
03
in
o
0.30 —
u 0.20 —
0.10 —
0 0.1 0.2 0.3 0.4
MASS CONCENTRATION, g/m3
0.5
3630-214
Figure 12. Opacity-mass concentration relationship for particulate
emissions from a cement plant kiln. After Buhne and Duwel.75
57
-------
A study was performed downstream of an electrostatic precipi-
tator on a kraft pulp mill recovery furnace. A correlation coef-
ficient of 0.99 was produced for mass concentration of 0.7 to
5 g/m3.76
Opacity-mass concentration measurements78 are shown for an
asphaltic concrete plant, an incinerator plant, a refinery cata-
lytic cracker regenerator, a lead smelter, and a brass smelter,
Figure 13.
Summary—Conventional transmissometers are routinely used for
providing a qualitative measurement, i.e., where changes in opacity
are used as a general indicator of changes in mass concentration.
Generally, transmissometers are not relied upon to produce quan-
titative measurements, i.e., where actual values of mass concen-
tration are obtained. This is due to the uncertainty introduced
by the strong dependence of the sensing principle on the particle
size distribution and index of refraction. The transmissometer
does possess the advantage of being able to provide an in situ,
continuous, real-time, integrated measurement. In conclusion,
it is unlikely that conventional transmissometers will ever be
used for routine quantitative measurement of mass concentration.
The multiple-wavelength transmissometer, discussed in the next
subsection, is a better candidate because it eliminates the un-
certainties caused by variable particle size distribution.
Other Optical Methods—
Multiple-wavelength transmissometers—The general principle
underlying the multi-wavelength transmissometer can be seen by
referring to Figure 14. In this figure, the mean extinction co-
58
-------
s 0.30
MASS CONCENTRATION GIVEN
FOR ACTUAL STACK CONDITIONS
0.0 0.16 0.32 0.48 0.64 0.80
MASS CONCENTRATION, grams M'3
ASPHALTIC CONCRETE PLANT EMISSIONS
0.96
0.36
0.32
0.28 -7
0.24 2"
0.20 |
O
0.16 ^
0.12 £
•Z
0.08 ^
UJ
0.04
0.0
MASS CONCENTRATION GIVEN
FOR ACTUAL STACK CONDITIONS
0.0
0.0
0.04 0.08 0.12 0.16
MASS CONCENTRATION, grams NT3
REFINERY CATALYTIC CRACKER REGENERATOR EMISSIONS
o 0.10
• FULL PROBE
O NO CONE
D STACK HOOD ON
MASS CONCENTRATION GIVEN
FOR ACTUAL STACK CONDITIONS
0.0 0.01 0.02 0.03 0.04 0.05
MASS CONCENTRATION, grams M'3
INCINERATOR PLANT EMISSIONS
0.32
0.28
0.24
0.20
0.16
0.12
0.08
0.04
Jo.o
0.30
s
8 0.20
73
< 0.10
MASS CONCENTRATION GIVEN
FOR ACTUAL STACK CONDITIONS
0.0 0.02 0.04 0.06 0.08 0.10
MASS CONCENTRATION, grams M'3
LEAD SMELTER EMISSIONS
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.0
0.06
0.02
O.Q
MASS CONCENTRATION GIVEN
FOR ACTUAL STACK CONDITION -
0.02
0.01
0.01 0.02 0.03
MASS CONCENTRATION, grams M'3
BRASS SMELTER EMISSIONS
Figure 13. Opacity - mass concentration relationship for various
industrial sources. After Reisman, et al.78
59
-------
E 2
Figure 14. Mean extinction coefficient as a function of the phase
shift parameter p vs.. After Dobbins and Jizmagianio
60
-------
efficient (E) is shown as a function of the phase shaft param-
eter (p ) of a polydisperse aerosol.80 The phase shift param-
eter is defined in terms of the mean volume-surface diameter:
Pvs = 2
-------
from 0.4 ym to 12 ym. The filters have been chosen so as to avoid
the absorption bands associated with the common stack gases.
The output of the detectors are fed to phase locked amplifiers
which in turn drive multiple pen chart recorders. The system
records the transmission through the stack at each of the selected
wavelengths.
There are several complex computational methods whereby the
particle size distribution and mass concentration can be obtained
from the optical density measurements made at the different
lengths. These are discussed in detail by Kerker.82
wave-
The system described by Reisman, et al., uses a computer
to perform the necessary data reduction78 This system has been
tested on a petroleum refinery catalytic cracking unit catalyst
regenerator, a sewage treatment plant incinerator for sludge
burning, an asphaltic concrete plant, a secondary brass and lead
smelter, and an oil-fired power plant. An example is given for
data taken at a sludge incinerator (see Figure 15). Data were
taken at 0.4 ym, 1.25 ym, 2.2 ym, and 11.6 ym. The optical den-
sities at these wavelengths are shown versus the optical density
from white-light. The data show a steady decrease in scattering
efficiency with increase in the wavelength. At 1.25 ym the ef-
ficiency is about one-fourth of that at 0.4 ym and at 2.2 ym the
efficiency is one-half of that at 1.25 ym.
The use of multiple-wavelength transmissometers to monitor
mass emissions seems promising, but the systems are more compli-
cated than ordinary transmissometers, and an undesirable depen-
dence on the particle refractive index can introduce errors.
Light scattering—Suspended particles in an aerosol will
scatter (diffract, refract, and reflect), and absorb incident
light; the remaining portion is transmitted. Whereas transmissom-
62
-------
CO
O
JC
o
O
O
CO
111
Q
_j
CJ
0.
o
0.0
•10 .20
OPTICAL DENSITY, white-light
3630-217
Figure 15. Results of monochromatic vs. white-light optical density
measurements made on sludge incinerator emissions.
After Reisman, et al.78
63
-------
eters use this remaining portion of the incident light as a mea-
sure of the particulate mass concentration or of opacity; other
instruments use the scattered portions. Instruments that detect
the scattered light can be much more sensitive at low particulate
concentrations than transmissometers.
Light scattering instruments suffer from some of the same
problems as transmissometers when attempting to infer mass; i.e.,
sensitivity to particle size, shape, and chemical composition.
The functional dependence of the instrument response to these
factors is determined by the detection angles employed relative
to the incident beam. This point is illustrated by Figure 16
where the scattered intensity versus particle size is plotted
for two detection systems and the same white light illumination.
For ideal aerosol mass monitors (many particles in the beam) these
curves would fall on the same straight line with a slope of three.
The fact that these lines are not straight and do not have a slope
of three means the deduced mass depends upon the aerosol size
distribution. The variation with refractive index means that
the deduced mass also depends upon chemical composition. Further
discussion of scattering versus refractive index can be found
in Hodkinson and Greenfield's work.84 The effects of such be-
havior are accounted for in practice by calibrations of the in-
strument against another more direct mass measurement of the
aerosol of interest.
Nephelometers, devices that attempt to measure all of the
scattered light, have recently been applied to stack monitoring.
One such instrument called the Plant Process Visiometer (PPV)
has been developed by Meteorology Research, inc.85'86'87 It must
be pointed out here that this instrument is designed to measure
opacity and is not considered a mass monitor per se; however,
it is normally quite sensitive to mass changes. A diagram of
its optical assembly is shown in Figure 17. The sample, extracted
64
-------
SCATTERING FUNCTION IN RELATIVE UNITS
a\
-------
LIGHT
SOURCE
APERTURES
DETECTOR
OPAL GLASS
CALIBRATOR
LIGHT TRAP
3630-219
Figure 17. Optical assembly diagram of a nephelometer used in stack
monitoring. The scattering angle 0, for any fight ray from
the source, is the angle between the ray and the horizontal
line a. From Ensor and Bevan.85
6b
-------
through a probe with no dilution, is passed through the detector
view. The light source is diffuse so that light rays illuminate
different portions of the sample in a wide range of angles from
near 0° to near 180° with respect to the detector view. During
operation the detector signal is calibrated with an opal glass
calibrator which has been adjusted to give a scattering coeffi-
cient of 0.055 m"1 as determined with oil smoke using an inte-
grating nepnelometer and a transmissometer. This scattering
coefficient value corresponds to an opacity of 5.4 percent as-
suming no light absorption (see the section on opacity for further
discussion).
Figure 18 shows mass correlation data with the PPV at the
inlet and outlet of a particulate scrubber on a coal-fired utility
boiler. The mass was determined gravimetrically and the scat-
tering coefficient averaged over the time of each mass run.
Ensor88 found that about half of the fluctuation is explained
by the uncertainties in sampling, and part of the remaining varia-
tion by changes in the particle size distribution. Figure 19
illustrates the effect which size distribution variation has upon
the correlation between mass concentration and scattering coef-
ficient. The data from Figure 20 were used to obtain the ratio
of particle volume concentration to the scattering coefficient
assuming a particle density of 2 g/cm3. The geometric mass mean
radius was derived from inertial impactor runs. The solid curve
(from Ensor and Pilat89) is that calculated for log-normal size
distributions with a geometric standard deviation of 4 and a
constant particle refractive index of 1.50 with no light absorp-
tion .
An in situ monitor has been developed90 that is based on
the measurement of the backscattered light. It uses a laser as
the light source and is a single ended instrument, i.e., both
67
-------
o
o
o
0.1
0.8
0.6
cc
LU
I-
o
M 0.2
0.01
| • •
.
• *
rO
00
O INLET
• OUTLET
10
1
50 100
MASS CONCENTRATION, mg/Am3
00
500
3630-220
Figure 18.
Mass correlation data taken with the Plant Process
Viscometer at the inlet and outlet of a particulate
scrubber on a coal-fired utility boiler. After
Ensor, et al.88
68
-------
CO
o
co
E
u
CM 10.0
E
o
H
cc
<
a.
o
1.0
(MASS/SCATTERING COEFFICIENT) VS GEOMETRIC MASS MEAN RADIUS
THEORETICAL CURVE WITH GEOMETRTc~STANDARD DEVIATION = 4
PARTICLE DENSITY ASSUMED TO BE 2 g/cm3
0.1
0.1
uL
_L
1-0 10.0
GEOMETRIC MASS MEAN RADIUS, d50/2,
J
100.0
3630-221
Figure 19. Effect of particle size distribution on particle volume
concentration/scattering coefficient. From Ensor, et al.88
-------
BACKSCATTERED
BEAM
SAMPLING
VOLUME
LIGHT COLLECTION
LENS
EMITTED
BEAM
LIGHT OMITTING
DIODE
SIGNAL
DETECTOR
BEAM FORMING
LENS 3630-222
Figure 20. Optical diagram of the PILLS V instrument. From
Schmitt, et al.91
70
-------
the light source and detector are located within the same enclo-
sure. The instrument is called the PILLS V (see Figure 20).
It is a member of a family of Particulate Instrumentation by Laser
Light Scattering devices developed by Environmental Systems Cor-
poration. An interesting feature of the PILLS V is the way in
which the instrument determines mass concentration. The instru-
ment optically defines a sample of 12 cm3 (0.73 in3) at 10 cm
from the end of the probe within the process stream. Detection
of the scattered light at angles greater than 160° relative to
the beam produces an electrical signal that is proportional to
the mass contained within the sample volume. Since the sample
volume is a constant, the mass concentration is read directly
from an appropriately labeled scale on the instrument meter.
At present, the instrument does not possess the capability to
traverse large stacks in order to obtain multi-point measurements.
Since the particulate mass concentration is frequently not uniform
across the entire cross-sectional area of the stack, the use of
such a small sampling volume and the inability to traverse creates
a problem when trying to obtain data that is representative of
the actual total mass concentration present within the stack.
Some of the specifications92 for the model P-5A, an improved
version of PILLS V, include the following: a measurement range
of 0.001-10 gram/ACM; response that is proportional to particle
mass concentration and is relatively independent of the particle
size in the range of approximately 0.1-8 ym; a process gas pres-
sure limit of +5 inches of water from ambient (higher limits are
optional); a process gas temperature limit of 260°C (500°F) (nega-
tive pressure streams permit use at higher temperatures); an in-
strument response that is independent of gas velocity; an optional
automatic zero and span calibration at preset intervals without
removal from the stack; and, a light source consisting of a highly
collimated beam of monochromatic laser light whose wavelength
is 0.9 urn. Figure 21 illustrates the theoretical variation of
71
-------
CO
CO
cc
LU
a.
i-
a.
D
O
CO
0.01 —
0.02 0.05 0.1 0.2 0.5 1.0 2.0 5.0 10.0 20.0
NUMBER MEAN DIAMETER, micrometers 3630223
Figure 21. Theoretical response of PILLS V vs. particle size.
Calculations for log-normal size distributions with
geometric standard deviations of 1.65 and varying
number mean diameter. From Schmitt, et al.93
12
-------
the instrument response with particle size for two types of par-
ticles, Si02 and H20. These calculations were performed for an
assumed log-normal size distribution with geometric standard
deviation of 1.65. Note that the ordinate is output per unit
mass taking the density into account for these two types of par-
ticles. Although the response is about the same for these two
compositions calibrations using gravimetric measurements for
reference are necessary for each process stream.9 3 '9 "*'9 5 It
should be pointed out also that particle irregularities affect
this type of measurement more than opacity or total scattering.
Performance data on the PILLS V are available in which mass
correlation tests have been performed. Independent wind tunnel
correlations were performed where mass concentration was derived
from tunnel parameters (mass input rate, volumetric air flow rate,
air temperature, etc.). These tests show linearity correlations
within ±10%.91 In another series of tests performed by an ESP
manufacturer at a coal-fired power plant, correlations of the
PILLS V to EPA Method 5 measurements varied by within ±7% on 5
tests, +10% on one, and -17% on one.91 Two extensive series of
tests9 tt'95 have been performed in the pulp and paper industry
which illustrate its sensitivity to size distribution and par-
ticle composition. In one of those series95 the PILLS V was com-
pared with a sampling train similar to the EPA Method 17 at three
paper mills. Tables IV-VII summarize the conditions and results
of those tests. Table VI shows the improvement of the reliability
of the data, indicated by an increase in the correlation coeffi-
cient, when the per cent of particle sizes within the 0.1-10 ym
range is increased. Table IV shows the correlation of instrument
response to color and salt content. The higher salt content pro-
duced a higher instrument response as did the lighter color (higher
reflection). It is interesting, however, that the dark color
with high salt content produced a high instrument response. This
suggests that the salt content had a more profound effect than
the color.
73
-------
TABLE IV. TYPICAL FLUE GAS CONDITIONS AND OPERATING
VARIABLES FOR CPM* CALIBRATION (AFTER WOSTRADOWSKI95)
MILL
,„.
A
AV.'I ,|(J,-
C02 , %
02 , %
H20, %
Temp, °F
Dust Cone, gr/sdcf
Filter Mesh Size, pm
Boiler Fuel Type
Continuous Operation, months
Isokinetic Sampling
Installation
*Continuous Particulate Monitor
(PILLS V)
B
8
12.5
10-15
450
0,1-0.7
1.0
Hog/Gas
2.5
Yes
Stack
8.5
12
10
420
0.15-1.0
0.3
Hog/Oil
1.5
Yes
— Duct to
16
3
33
325
0.05-1.2
0.3 or 1.0
Black Liquor
2.5
Yes
Stack —
TABLE V. LINEAR REGRESSION RESULTS OF CPM* CALIBRATION
CURVES FOR EACH MILL (AFTER WOSTRADOWSKI§5)
MILL
A
B
Instrument Sensitivity**
Particulate Cone Data, acf:
Slope***
Standard Error
Correlation Coeff, r2
Particulate Cone Data, sdcf
Slope***
Standard Error
0.2
0.5
16.2
81.9
0.65
21.8
105.7
0.75
Correlation Coeff, r2
* Continuous particulate monitor (PILLS V).
1.0
9.6
51.7
0.68
15.5
67.6
0.79
12.1
39.1
0.89
32.0
109.0
0.89
** To compare the instrument response in each mill the instrument
sensitivity (range setting) has to be adjusted to the same value.
*** Linear regression Equation Y = MX+B where Y = particulate
Concentration, M = slope, X = instrument response and B = intercept
(Theoretically there should not be an intercept and this was
substantiated by experiment).
74
-------
TABLE VT. COMPARISON OF CORRELATION DATA WITH PARTICULATE
CHARACTERISTICS (AFTER WOSTRADOWSKI9 5)
MILL
Slope*
Correlation Coeff, r2
Particulate Size Outside
Range 0.1 to 10 pm, %
Particulate Color**
Salt Content, %
24.0
0.68
60
5-10
20
15.5
0.79
20
5-10
50
6.1
0.89
<5
0
30
* The instrument sensitivity (range setting) was adjusted (by
calculation) to that used at Mill B (0.5) to the other two mills.
This was done by multiplying the instrument response in Mill A by
2.5 and by dividing the instrument response in Mill C by 2.0.
** Baccarach color test (Petroleum Industry) is a darkness/lightness
rating based on 0 - white and 10 - black.
TABLE VII. RELATIVE PILLS V RESPONSE AS A FUNCTION OF
SALT CONCENTRATION AND COLOR OF THE EXTRACTED DUST
SAMPLES FROM MILL B (AFTER WOSTRADOWSKI
9 5 <
TEST
NO.
358
359
364
370
373
374
375
376
379
368
401
383
384
392
394
362
366
380
SALT >50%
OF DUST
X
X
X
X
X
X
X
X
SALT <50%
OF DUST
X
X
X
X
X
X
X
X
X
X
DARK
COLOR
X
X
X
X
X
X
X
X
. X
X
X
LIGHT
COLOR
X
X
X
X
X
X
X
. LOW
PILLS V
RESPONSE
X
X
X
X
X
X
X
"™
X
HIGH
PILLS V
RESPONSE
X
X
X
X
X
X
X
X
X
~
75
-------
Mass measurements have been performed and recommended -with
another backscattering instrument96 called an LTV monitor, al-
though a commercial model is not available. This device, illu-
strated in Figure 22 uses a high intensity argon or xenon laser
and a TV camera with telephoto lens. The camera optics image
the backscattered light at 175° from the focused view volume,
intersecting the laser beam. Particles that produce illumina-
tion above the sensitivity threshold can be resolved as distinct
flashes and the intensity of each can be measured. The size of
each particle is derived from the intensity. The particle mass
is then deduced from the size and an assumed density. The video
analysis circuitry is included in reference 96. The dynamic range
of the camera limits the size range to about a factor of 10.
This method has the capability of changing the position of the
view volume in the process stream. However, deduced particle
size from backscattering is very sensitive to refractive index
and errors are tripled in the calculation of the mass.
The results of tests of the LTV on coal-fired boilers are
quite good in comparison with EPA Method 5 and opacity measure-
ments given in Table VIII. The laser power was 800 watts and
the sample volume of 1 mm3 was set 40 cm into the duct. The
camera resolved this volume into 10" elements so that 103 par-
ticles per sample could be viewed.
In summary, two light scattering instruments, the MRI nephelo-
meter and the ESC PILLS V, are commercially available and being
used in routine measurements on process streams. As indicated
above these devices give acceptable measures of mass concentration
in real time if calibration is performed for each stream against
a direct mass technique and if the size distribution and composi-
tion of the aerosol remain nearly constant. Very little is known
at present about the limits of variation of these aerosol param-
eters for a given type of stream. At present studies are per-
76
-------
STACK GAS
WINDOW
PULSED ARGON OR
XENON LASER
TV CAMERA WITH
TELEPHOTO LENS
PARTICLE SIZE
ANALYZER
3630-224
Figure 22. Schematic of Laser-TV Monitor. After Tipton.96
77
-------
TABLE VIII. FIELD TEST RESULTS. (RESULTS IN GRAINS/STANDARD FT3
AFTER TIPTON 9 6) '
OPACITY METER
RESULT (%)
_ ^_
Coal-fired boiler
Oil-fired boiler
TEST
NO.
1
2
3
1
2
3
EPA TRAIN
TEST
0.295
0.104
0.143
0.035
0.0361
0.0558
LTV MONITOR 0]
TEST ]
— — _ —
0.297
0.100
0.112
0.0322
0.025
0.0575
40
27
25
formed on each process to test the correlation of these devices
to other mass measurements as in the studies described in ref-
erences 91 and 93-95.
Other Methods
Sem, et al.5'," have reviewed possible techniques for moni-
toring particulate emissions, and a number of interesting concepts
were discussed. Some of these are listed, in addition to methods
Sem did not mention, below:
Sem, et al.
Resonant Frequency
Gravimetric Weights
Rotating Masses
Impact Momentum
Capacitance - Di-
electric Change
Soiling Potential
Acoustical Attenuation
and Dispersion
Filter Pressure Drop
Gas Adsorption
Volume Measurement
Capacitance - Impact
Electrostatic Contact Charge;
Bounce
Ion Current Attenuation
Acoustical Particle Counter
Flame lonization (Altpeter,
et al.98)
Flame Photometry
Pressure Drop in Nozzle
78
-------
Additional
Guichard Apparatus - Charge Transfer (Knapp68)
Hot Wire Anemometry (Goldschmidt99)
Surface lonization (Extranuclear Laboratories, Inc.,
P.O. Box 11512, Pittsburgh, PA 15238)
Some of the techniques listed above may be applicable to
monitoring process streams, and some commercial instruments are
available for laboratory and ambient measurement. However, none
of these concepts are sufficiently proven to warrant more dis-
cussion in this manual.
79
-------
SECTION III
OPACITY
Suspended particles in a fluid medium will scatter and absorb
radiation; the remaining portion is transmitted. The transmit-
tance, T, of a fluid medium containing suspended particles is
defined as the ratio of transmitted radiation intensity to in-
cident radiation intensity. T is given by the Bouguer, or the
Beer-Lambert, law:100
T = exp (-EL)
(6)
where L is the thickness of the medium, and E, the extinction
coefficient of the medium. Sometimes the measured transmittance
is expressed in terms of optical density defined as
O.D. = Log (1/T)
(7)
instead of the transmittance. Consequently, instruments and
methods for aerosol measurement based upon light transmission
principles have been referred to as transmissometers, smoke den-
sity meters, photo-extinction measurements, turbidimetric mea-
surements, etc.
While transmittance is defined as the ratio of light trans-
mitted through the aerosol to the incident light, opacity is de-
fined as the ratio of the light attenuated from the beam by the
aerosol to the incident light (i.e., opacity = 1-T). Aerosols
which transmit all incident light are invisible, have a trans-
mittance of 100%, and an opacity of zero. Emissions which at-
tenuate all incident light are totally opaque, have an opacity
80
-------
of 100% and a transmittance of zero. By definition, opacity can
only be measured rigorously using transmittance, rather than light
scattering measurements, because the latter yield no measure of
the quantity of light that is absorbed.
Many versions of transmissometers, or smoke meters, are avail-
able as stack emission monitors. If the transmissometer is used
to measure in-stack opacity for purposes of compliance with federal
regulations, it must meet the EPA requirements for opacity mea-
surement systems as specified in the Federal Register of September
11, 1974.101 For instance, the use of visible light as a light
source is required. For other uses of the data, it may be pos-
sible to operate with nonvisible wavelengths. The angle of view
and the angle of projection are both specified, for compliance,
as no greater than 5° (see Figure 23).
A typical double pass in situ transmissometer design is il-
lustrated in Figure 24. The design shown employs a chopped, dual-
beam, optical system that automatically compensates for the ef-
1 fl 9
fects of temperature, voltage changes, and component aging.
The same source is optically divided into a measuring beam and
a reference beam. The measuring beam is reflected back by a
corner-cube retroreflector. The reference beam provides automatic
gain control to compensate for any changes in detector response
or source intensity. Zero calibration checks can be made by
inserting a test reflector in the light path outside the optical
window of the source/detector (transceiver). This test reflector
simulates a condition of zero opacity. The zero calibration check
is used to adjust for particulate accumulation on the optical
windows or drift due to the instrument circuitry. It is assumed
that the retroreflector windows soil at the same rate as the trans-
ceiver window.
81
-------
PROJECTION ANGLE ANGLE OF VIEW
SOURCE
SAMPLE VOLUME
APERTURE
SCHEMATIC OF A TYPICAL TRANSMISSOMETER SYSTEM
3630-225
Figure 23. Schematic of a transmissometer showing projection
and viewing angles which must be no greater than 50
for EPA compliance.
82
-------
CHOPPER FREQUENCY
MEASUREMENT
BEAM
F = 3.9 kHz
REFERENCE
CALIBRATION
FILTER
OBJECTIVE
FOCUSING
LENS
BEAM SPLINTER
RETROFELECTOR
00
ZERO
CALIBRATION
REFLECTOR
TUNGSTEN LAMP
(15W)
APERTURE
PLATE
ADJUSTABLE
IRIS COARSE
ZERO
CHOPPER
FREQUENCY
F = 2.5 kHz
ROTATING
CHOPPER
DISC
PURGE AIR BLOWER
AND FILTER
3630-226
SYNCHRONOUS
CHOPPER
MOTOR
Figure 24. A typical double pass in situ transmissometer design.
After Nader. 79
-------
Both the transceiver and retroreflector unit are specially
constructed with air purging attachments to keep the optical win-
dows free of particulate deposits, and can provide adequately
clean windows for three or more months of unattended operation.
A single pass design is shown in Figure 25. In this design,
the light source with the collimating lens and power supply are
placed on one side of the stack with the detector cell, electronics
and power supply on the opposite side. The beam makes only one
pass through the stack gas which eliminates the problems caused
by reflectivity or back scattering of the effluent being measured.
There are two photocells, one for detection and one used as a
reference. Because ambient temperature fluctuations affect the
sensitivity of silicon photocells, compensation of the detector
photocell is required. The reference photocell, which samples
the light from the source via a fiber optics assembly, is needed
to compensate for variations in light source intensity due to
aging, line voltage changes, and light source replacement. The
current output of the reference cell is proportional to the light
output of the source lamp. The design shown incorporates a moving
Mylar strip to protect the optical surfaces from dust accumula-
tion. This strip passes in front of the objective lens on both
the source and the sensor at a rate that effectively changes every
hour that the window is exposed to the stack.
Transmissometers usually contain an alarm or warning system
that alerts plant personnel when the opacity exceeds a preset
limit. An alarm and/or plant cut off switch can be automatically
activated when limit values are exceeded. One instrument pos-
sesses the ability to integrate the opacity measurements over
various time intervals. This permits automatic monitoring and
control of unacceptable emission or dust levels whch are present
for long periods of time, and not for just a brief moment. This
instrument was used to monitor cement factory emissions sources.104
84
-------
SOURCE UNIT
DETECTOR UNIT
r
POWER
SUPPLY
STACK
LIGHT SOURCE
WITH
COLLIMATING
LENS
CO
L
ZERO
ADJUST
APERATUREl
SAMPLE PATH
(THROUGH
STACK)
REFERENCE PATH
(FIBER OPTICS)
DETECTOR
CELL
REFERENCE
CELL
REGULATED
POWER
SUPPLY
AMPLIFIER
AND ANALOG
DIVIDER
STRIP-CHART
RECORDER
OPTIONAL
ALARM OR
CONTROL
RELAY
FUNCTIONAL BLOCK DIAGRAM
3630-227
Figure 25. A single pass transmissometer design. After Haville. 103
-------
Transmissometers can be used to measure the in-stack opacity
in order to obtain an estimate of the plume opacity for compliance
testing; performance specifications for these transmissometers
can be found in the Federal Register.101 Also, transmissometers
can be used to measure the in situ opacity for process control
or as an estimate of mass concentration.
When the required measurement is the opacity of the emissions
at the exit of the stack, a measurement at any other location
in the stack has to have its optical path length adjusted to the
exit diameter. The method of calculation for this adjustment
can be found in the Federal Register.101 Figure 26 shows the
relationship of effluent transmittance at the stack exit as a
function of in-stack transmittance for various ratios of stack
exit diameter to transmissometer optical path length.1Q5
Comparisons of transmissometer measurement with visual plume
opacity have been made. The in-stack measurement is usually com-
pared with an out of stack plume measurement performed by visual
observation by a trained observer or performed by telephotometry.
In one telephotometric study,IOG it was concluded that the
in-stack transmissometer can be used to monitor the opacity of
plumes emitted from steel plant basic oxygen furnaces and from
cement plants (over a small range of opacity) with reasonable
accuracy. However, use of the in-stack transmissometer for sul-
furic acid plants was found to be questionable.
Another study was performed to determine the influence of
transmissometer design on the correlation of in-stack opacity
measurements with the stack plume opacity as measured by tele-
photometry.107
86
-------
100
20 30 40 50 60
IN STACK TRANSIVHTTANCE, percent
70 80 90 100
3630-228
Figure 26.
Effluent transmittance vs. in stack transmittance for
varying ratios of stack exit diameter to in stack path
length: A = 1/4, B = 7/2, C = 3/4, D = 1, E = 4/3
F = 2, G = 4. After Nader. 105
87
-------
Ensor and Pilat have developed theoretical methods for cal-
culating plume opacity from the properties and concentration of
the particles in the plume.89 Use of constants that are deter-
mined experimentally from the effluent improves the reliability
of the theoretical method.108
The important parameters which affect transmissometer per-
formance in a given process stream are the particle size distri-
bution in the process stream, particle shape and refractive index,
the wavelength of the transmitted radiation, and the collimating
angles of the transmissometer.
The effect of the first four parameters are combined into
the extinctin coefficient, E, of the process stream.
E = TT I r2
/
rQ (a,m)N(r)dr
where
a = size parameter, 2r7r/X
r = particle radius
x = wavelength of the radiation
m = particle refractive index relative to the gas medium
N(r) = number size frequency distribution, i.e., the number
of particles of radius r per volume per Ar
QE = particle extinction coefficient
Thus, the extinction coefficient, E, is calculated by summing
the effects of all the individual particles in the process stream,
QE, the particle extinction coefficient, is defined as the
total light flux scattered and absorbed by a particle divided
88
-------
by the light flux incident on the particle. For spherical par-
ticle, with typical indices of refraction relative to air of 1.3
to 1.6, QE will vary from 0 to 4. For these particles in the
Rayleigh scattering region (diameter, d < 0.05 ym in white light)
QE is approximately 0. For this type of particle in the Mie scat-
tering region (0.05 < d < 2 ym in white light) Q varies from
0 to 4. For this type of particle in the geometric scattering
region (d > 2 ym in white light) QE approaches a theoretical limit
of 2 for very large particles. See Figure 27.
In practice, the particles in stack emissions are polydis-
persed and the light source is polychromatic. This results in
the smoothing out of the oscillatory behavior depicted in Figure
27. However, distribution of transparent particles which are
skewed to a narrow range of particle sizes in the Mie region can
result in opacity readings for this region similar to those found
for much higher mass concentrations of absorbing particles.
Aerosols with mean particle diameters above 2 ym, such as
those sometimes found in stacks, will generally have a mean par-
ticle extinction coefficient of 2 and a transmittance that is
not strongly dependent on wavelength. Measurements of the emis-
sions from a pulverized coal-fired power plant using an in-stack
transmissometer, were found to vary as a function of the wave-
length of monochromatic light in the visible range, there being
an increase in opacity of about 7% in going from red to blue
light.107 In a study using experimental white (oil) and black
(carbon) plumes, this effect was more pronounced with the trans-
parent white plume than with the absorbing black plume.109
To obtain true transmittance data the collimation angles
(angles of view and projection) for the transmitter and receiver
must be limited to reduce the sensitivity to stray light scatter
(see Figure 23). A zero degree angle is the ideal collimating
89
-------
A - TRANSPARENT MONODISPERSE SPHERES, m = 1 33
B - TRANSPARENT MONODISPERSE SPHERES m = 1 5
C - ABSORBING MONODISPERSE SPHERES, m = 1.59 - 0 66
1-0 1.5 2.0
PARTICLE DIAMETER, micrometers
MIE
GEOMETRIC
RAYLEIGH
Figure 27. Particle extinction coefficients for various aerosols over
three scatteringjegions: Rayleigh, Mie, and Geometric.
3630-229
90
-------
angle, whereas a finite angle will introduce a systematic error.
However, a compromise is necessary, since as a zero degree col-
limation is approached, instrument construction costs, operating
stability, and optical alignment problems increase. A transmis-
someter having a 5 degree collimating angle applied to the emis-
sions of a pulverized coal-fired stream generator gave an opacity
measurement that was about 5% low relative to the zero degree
value.
The error in the transmissometer measurement due to the use
of different light detection angles has been analyzed theoreti-
cally by Ensor and Pilat and was shown to be a function of de-
tection angle and particle size.110 They showed that, in general,
the error associated with a given detector viewing angle increases
with an increase in the particle mean diameter.
In situ transmissometers produce an instantaneous measure
of the average opacity created by the differing particle concentra-
tions that exist along the line of sight of the transmissometer
light beam. This eliminates the need for traversing the stack
as is done, for point measurements or extractive measurements.
However, the need for an extractive transmissometer arises when
measurement is to be made of process streams containing water
droplets. (Water vapor does not cause trouble but droplets do.)
In such cases, it is necessary to extract and then heat a sample
from the stream, thus vaporizing the water droplets.
As opacity, 1-T, approaches zero the relative error in its
measurement with a transmissometer becomes unavoidably large.
For example, a two per cent error in the transmittance measurement
gives a 50 per cent error in an opacity of four per cent. In
such cases a nephelometer as used by Ensor,86 may be a more ac-
curate measure of opacity, although it requires a probe and sampl-
ing traverses. This instrument when used as an opacity monitor
91
-------
attempts to determine E, the extinction coefficient, through a
measurement of the scattering coefficient alone where E = scatter-
inc coefficient + absorption coefficient. The errors in this
type of opacity measurement depend upon the variation of the ratio
of aerosol absorption coefficient to the scattering coefficient
and the errors associated with extractive sampling. The ratio
varies from zero for nonabsorbing particles to about one for
highly absorbing ones giving possible errors in opacity up to
100 per cent depending upon the calibration aerosol. However,
if a calibration aerosol is chosen judiciously (i.e., with optical
properties close to those of the sampled aerosol) and the opacity
is low, the nephelometer errors are much smaller than those ob-
tained with the transmissometer at low opacities. Operation of
the nephelometer is discussed further, including field test data,
in the section on mass monitors.
92
-------
SECTION IV
PARTICLE SIZE DISTRIBUTIONS
ESTABLISHED TECHNIQUES
Field Measurements
Aerodynamic Methods—•
In order to avoid unnecessary complications in data presenta-
tion, particles of different shapes may be assigned aerodynamic
diameters. The aerodynamic diameter of a particle is the diameter
of a unit density sphere that has the same settling velocity as
the particle of interest. The aerodynamic diameter is related
to the way that a particle will behave in the respiratory system
as well as in aerodynamic sizing devices.
Examples of aerodynamic particle sizing instruments are
centrifuges, cyclones, cascade impactors, and elutriators. Each
of these instruments employs the unique relationship between a
particle's diameter and mobility in gas or air to collect and
classify the particles by size. For pollution studies cyclones
and impactors, primarily the latter, are more useful because they
are rugged and compact enough for in situ sampling. As previously
explained, in situ sampling is preferred because the measured
size distribution may be seriously distorted if a probe is used
for sample extraction. In the following two subsections, methods
of using impactors and cyclones are discussed.
93
-------
Cascade impactors—The mechanism by which a cascade impactor
operates is illustrated in Figure 28. In each stage of an im-
pactor, the gas stream passes through an orifice and forms a jet
that is directed toward an impaction plate. For each stage there
is a characteristic particle diameter that has a 50% probability
of impaction. This characteristic diameter is called the D50
of the stage. Although single jets are shown in Figure 30 for
illustrative purposes, commercial impactors may have from one
to several hundred jets in a stage. Typically, an impactor has
five to ten stages.
The particle collection efficiency of a particular impactor
jet-plate combination is determined by 'such properties of the
aerosol such as; the particle shape and density, by the viscosity
of the gas, and by the design>of the impactor stage (that is the
shape of the jet, the diameter of the jet and the jet-to-plate
spacing).111'112'113'114'115 There is also a slight dependence
on the type of collection surface used (glass fiber, grease,
metal, etc.).116fl17 rl ia
Most modern impactor designs are based on the semi-empirical
theory of Ranz and Wong.119 Although more sophisticated theories
have been developed,120'121'122 these are more difficult to apply.
Since variations from ideal behavior in actual impactors dictate
that they be calibrated experimentally, the theory of Ranz and
Wong is generally satisfactory for the selection of jet diameters,
Cohen and Montan,111 Marple and Willeke,112 and Newton et al.,114
have published papers that summarize the important results from
theoretical and experimental studies to determine the most im-
portant factors in impactor performance:
1. The jet Reynolds number should be between 100 and 3000.
94
-------
\ PATH OF
\ SMALL PARTICLE
3630-230
Figure 28. Schematic diagram, operation of cascade impactor.
-------
2. The jet velocity should be 10 times greater than the
settling velocity of particles having the stage D50.
3. The jet velocity should be less than 110 m/sec.
4. The jet diameter should not be smaller than can be at-
tained by conventional machining technology.
5. The ratio of the jet-plate spacing and the jet diameter
or width (S/W) should lie between 1 and 3.
6. The ratio of the jet throat length to the jet diameter
(T/W) should be approximately equal to unity.
7. The jet entries should be streamlined or countersunk.
Smith and McCain123 have observed that the jet velocity for
optimum collection of dry particles may be as low as 10 m/sec,
which places a more stringent criterion on impactor design and
operation.
Figures 29 and 30 are charts that summarize the design cri-
teria for cascade impactors. It can be seen that it is almost
impossible to achieve D50's of 0.2-0.3 ym without violating some
of the recommended guidelines.
Table IX lists six commercially available cascade impactors
that are designed for instack use, and tables X through XIV show
some geometric and operating parameters for the commercial im-
pactors. Schematic diagrams of five commercially available impac-
tors are shown in Figure 31.
The impactors are all constructed of stainless steel for
corrosion resistance. All have round jets, except the Sierra
96
-------
10.0
u
DC
LLJ
0.01
Figure 29. Approximate relationship among jet diameter,
number of jets per stage, jet velocity, and stage
cut point for circular jet impactors. From
Smith and McCain. 123
97
-------
104
a
LU
O
_l
U_
W = Jet Diameter
Re = Reynolds Number
C = Cunningham Slip Correction
D50 = Particle Aerodynamic Dia.
at 50% Cut Point
10°
1 1 5
NUMBER OF ROUND JETS PER STAGE, n
50 100
500 1000
3630-232
Figure 30. Design chart for round impactors. (D^o = aerodynamic diameter
at 50% cut point.) After Marple. 112
98
-------
TABLE IX
COMMERCIAL CASCADE IMPACTOR SAMPLING SYSTEMS
Name
Andersen Stack Sampler
(Precollection Cyclone
Avail.)
Univ. of Washington
Mark III Source Test
Cascade Impactor
(Precollection Cyclone
Avail.)
Univ. of Washington
Mark V
Brink Cascade Impactor
(Precollection Cyclone
Avail.)
Sierra Source Cascade
Impactor - Model 226
(Precollection Cyclone
Avail.)
MRI Inertial Cascade
Impactor
Nominal Flow rate
(cm3/sec)
236
236
100
14.2
118
236
Substrates
Glass Fiber (Available from
manufacturer)
Stainless Steel Inserts,
Glass Fiber, Grease
Stainless Steel Inserts,
Glass Fiber, Grease
Glass Fiber, Aluminum,
Grease
Glass Fiber (Available
from manufacturer)
Stainless Steel, Alumn-
num, Mylar, Teflon.
Optional: Gold, Silver,
Nickel
Manufacturer
Andersen 2000, Inc.
P.O. Box 20769
Atlanta, GA 30320
Pollution Control
System Corp.
321 Evergreen Bldg.
Renton, WA 98055
Pollution Control
System Corp.
321 Evergreen Bldg.
Renton, WA 98055
Monsanto EviroChem
Systems, Inc.
St. Louis, MO 63166
Sierra Instruments, Inc
P.O. Box 909
Village Square
Carmel Valley, CA 9392<
Meteorology Research,
Inc.
Box 637
Altadena, CA 91001
-------
TABLE X.
CASCADE IMPACTOR STAGE PARAMETERS
ANDERSON MARK III STACK SAMPLER
Stage
M No. '
o
o
1
2
3
4
5
6
7
8
No . o f
Jets
264
264
264
264
264
264
264
156
D.-Jet
Diameter
(cm)
.1638
.1253
.0948
.0759
.0567
.0359
.0261
.0251
S-Jet
to Plate
Distance
(cm)
.254
.254
.254
.254
.254
.254
.254
.254
D
1
2
2
3
4
7
9
10
S
j
.55
.03
.68
.35
.48
.08
.73
.12
Reynolds
Number
45
59
78
98
131
206
284
500
Jet
Velocity
(m/sec)
0
0
1
2
3
9
17
31
•4
.8
.3
.0
.6
.0
.1
.5
Cumulative Frac-
tion of Impac-
tor Pressure Drop
at each stage
0
0
0
0
0
0
0
1
.0
.0
.0
.0
.0
.2
.3
.0
-------
TABLE XI.
CASCADE IMPACTOR STAGE PARAMETERS
MODIFIED BRINK MODEL B CASCADE IMPACTOR
o
Stage
No.
0
1
2
3
4
5
6
No. of
Jets
1
1
1
1
1
1
1
D.-Jet
Diameter
(cm)
.3598
.2439
.1755
.1375
.0930
.0726
.0573
S-Jet
to Plate
Distance
(cm)
1
0
0
0
0
0
0
.016
.749
.544
.424
.277
.213
.191
D
2.
3.
3.
3.
2.
2.
3.
S
j
82
07
10
08
98
93
33
Reynolds
Number
326
481
669
853
1263
1617
2049
Jet
Velocity
(m/sec)
1
3
6
9
21
35
58
.4
.0
.0
.7
.2
.3
.8
Cumulative Frac-
tion of Impac-
tor Pressure Drop
at each stage
0.
0.
0.
0.
0.
0.
1.
0
0
0
0
065
255
000
-------
TABLE XII.
CASCADE IMPACTOR STAGE PARAMETERS
MRI MODEL 1502 INERTIAL CASCADE IMPACTORS
Stage No., of
No. Jets
M
o
NJ
1 8
2 12
3 24
4 24
5 24
6 24
7 12
D.-Jet
Diameter
(cm)
0.870
0.476
0.205
0.118
0.084
0.052
0.052
S-Jet
to Plate
Distance
(cm)
0.767
0.419
0.191
0.191
0.191
0.191
0.191
S_
D .
D
.88
.88
.96
1.61
2.27
3.60
3.60
Reynolds
Number
281
341
411
684
973
1530
3059
Cumulative Frac-
Jet ' tion of Impac-
Velocity tor Pressure Drop
(m/jsec) at each stage
0.5
1.1
3.2
8.9-
18.2
45.9
102.3
0.0
o.o •
0.0
0.0
0.045
0.216
1.000
-------
TABLE XIII.
CASCADE IMPACTOR STAGE PARAMETERS
SIERRA MODEL 226 SOURCE SAMPLER
Stage
No.
1
M 2
o
3
4
5
6
W-Jet
Slit
Width
(cm)
0.
0.
0.
0.
0.
0.
3590
1988
1147
0627
0358
0288
Jet
Slit
Length
(cm)
5.156
5.152
3.882
3.844
3.869
2.301
S-Jet
to Plate
Distance
(cm)
0.
0.
0.
0.
0.
.0.
635
318
239
239
239
239
S
W (
1.77
1.60
2.08
3.81
6.68
8.30
Reynolds
Number
@14.16 1pm)
602
602
800
808
802
1348
Jet Cumulative Frac-
Velocity tion of Impac-
(m/sec) tor Pressure Drop
(@14.16 1pm) at each Stage
1
2
5
10
17
36
.3
.3
.4 •
.0
.4
.9
0
0
0
0
0
1
. 0
.0
.0
.154
.308
.000
-------
TABLE XIV
CASCADE IMPACTOR STAGE PARAMETERS
UNIVERSITY OF WASHINGTON MARK III SOURCE TEST CASCADE IMPACTOR
Stage
o No-
>£>
1
2
3
4
5
6
7
No. of
Jets
1
6
12
90
110
110
90
D.-Jet
Diameter
(cm)
1
0
0
0
0
0
0
.842
.577
.250
.0808
.0524
.0333
.0245
S-Jet
to Plate
Distance
(cm)
1
0
0
0
0
0
0
.422
.648
.318
.318
.318
.318
.318
S
°i
.78
1.12.
1.27
3.94
6.07
9.55
12.98
Reynolds
Number
1073
565
653
269
340
535
929
Cumulative Frac-
Jet tion of Impac-
Velocity tor Pressure Drop
(m/-sec) at each Stage
0
1
4
5
10
25
60
.9
.5
.1
.2
.2
.4
.0
0
0
0
0
0
0
1
.0
.0
•0
.019
.057
.189
.000
-------
INLET JET
STAGE NO. 1
FILTER
IMPACTOR BASE
PRECOLLECTION
CYCLONE
JET STAGE _
(7 TOTAL)
COLLECTION
PLATE
t
t
t
D
MRI MODEL 1502
MODIFIED BRINK
COLLECTION PLATE
COLLECTION
PLATE (7 TOTAL!
FILTER HOLDER
UNIVERSITY OF WASHINGTON MARK III
Figure 31. Schematics of five commercial cascade impactors (Sheet 1 of 2).
105
-------
INLET CONE
SIERRA MODEL 226
JET STAGE (9 TOTAL)
ANDERSON MARK III 3630233
Figure 31. Schematics of five commercial cascade impactors (Sheet 2 of 2).
106
-------
Model 226, which is a radial slit design,and all have stages with
multiple jets, except the Brink. It is customary to operate the
impactors at a constant flowrate during a tost so that the Dso's
will remain constant. The impactor flowrate is chosen, within
a fairly narrow allowable range, to give a certain sampling ve-
locity at the nozzle inlet. Streamlined nozzles of different
diameters are provided to allow the sample to be taken at a
velocity equal to that of the gas stream.
Since the impaction plates weigh a gram or more, and the
typical mass collected on a plate during a test is on the order
of 1-10 mg, it is often necessary to place a light weight collec-
tion substrate over the impaction plate to reduce the tare. These
substrates are usually glass fiber filter material or greased
aluminum foil. A second function of the substrates is to reduce
particle bounce.
Gushing, et al. have done extensive calibration studies of
the commercial, instack, cascade impactors.117 Figure 32 shows
results from calibration of the Andersen Mark III impactor that
are similar to the performance of the other types as well. Similar
results have been reported by Mercer and Stafford,122 Rao and
Whitby,116 and Calvert, et al.12* for impactors of different de-
sign. Notice that the collection efficiency increases, as particle
size increases, up to a maximum value that is less than 100%.
The decrease in collection efficiency for large particles repre-
sents bounce and can introduce serious errors in the calculated
particle-size distribution.
There bar, not been an extensive evaluation of cascade im-
pactors under field conditions, although some preliminary work
was reported by McCain, et al.125 It is difficult to judge from
existing data exactly how accurate impactors are, or how well
107
-------
0
.3 .4 .5 .6.7.8.91.0 2 3 4 5 6 7 8 9 10
PARTICLE DIAMETER, micrometers 3630-234
Figure 32. Calibration of an Anderson Mark III impactor.
Collection efficiency vs. particle size for stages
1 through 8. After Gushing, et al.11?
108
-------
the data taken by different groups or with different impactors
will correlate. Problems that are known to exist in the applica-
tion of impactors in the field are: substrate instability,122'26
the presence of charge on the aerosol particles,126 particle
bounce,116'122 and mechanical problems in the operation of the
impactor systems.
In the past, the reduction of data from an extensive field
test has been excessively tedious and time consuming. However,
a powerful computer program is now available that decreases the
effort required to reduce and analyze impactor data by approxi-
mately a factor of five.127
Research is continuing to improve the hardware and tech-
niques for making particle-size measurements with impactors.
At the present time, it is necessary that the operators be experi-
enced and that great care be exercised to avoid many potential
sources of errors in order to obtain reliable results.
Cyclones—Cyclones have been used for many years as devices
for cleaning dusty air and also to separate respirable and non-
respirable dusts in personnel exposure monitors. Strauss128 has
reviewed in detail the theory, design, and performance of indus-
trial cyclones, while Lippmann and Chan have performed several
experimental/theoretical studies of the small cyclones used as
personnel exposure monitors.129'130 in general it can be said
that the existing theories are not accurate enough to design
cyclones tor particle sizing, and thus such designs must be de-
veloped empirically.
Figure 33 illustrates a typical reverse flow cyclone. The
aerosol sample enters the cyclone through a tangential inlet and
forms a vortex flow pattern. Particles move outward toward the
cyclone wall with velocity that is determined by the geometry
and flowrate in the cyclone, and by their size. Large particles
109
-------
SAMPLE AIR FLOW
GAS EXIT TUBE
CAP
CYLINDER
CONE
COLLECTION CUP
3630-235
Figure 33. Hypothetical flow through typical reverse flow cyclone.
110
-------
reach the walls and are collected. Figure 34 compares the cali-
bration curve for a small cyclone with a typical impactor calibra-
tion curve. The cyclone can be seen to perform almost as well as t
impactor, and the problem of large particle bounce and reentrain-
ment is absent.
Several theories have been proposed in attempts to describe
the efficiency of particle collection by cyclones. The majority
of the theoretical equations, however, were developed for cyclone
design in industrial gas cleaning applications. Typically, the
theories include terms for the centripetal and aerodynamic drag
forces on the aerosol particles. As with impactors, cyclone per-
formance may be conveniently expressed in terms of a characteristic
D50, which is the diameter of particles that are collected with
50% efficiency. In their recent experiments with small cyclones,
Chan and Lippmann 1 3° have observed that most cyclone performance
data can be fitted by equations of the form
D50 = KQn (8)
where:
K is an empirical constant,
Q is the sample flow rate, and
n is an empirical constant.
Unfortunately, K and n are different for each cyclone geometry,
and apparently are impossible to predict. In their study, Chan
aad Lippmann found K to vary from 6.17 to 4591, and n from -0.636
to -2.13. A similar study by Smith and Wilson131 found K to vary
from 44 to 14, and n from -0.63 to -1.11 for five small cyclones.
In addition to the flowrate dependence indicated in equation
(8), cyclone D50's also are affected by temperature through the
viscosity of the gas. Smith and Wilson found this dependence
to be linear, but with a different slope for different cyclone
dimensions and flowrates.
Ill
-------
5?
it
O
Ui
Z
O
\-
o
O
O
IMPACTOR — I I— CYCLONE
1.0 1.5 2.0
PARTICLE DIAMETER / D50
Figure 34. Comparison of cascade impactor stage with cyclone collection
efficiency curve.
112
-------
A series of cyclones with progressively decreasing D5fl's
can be used instead of impactors to obtain particle size distri-
butions, with the advantages that larger samples are acquired,
particle bounce is not a problem, and no substrates are required.
Also, longer sampling times are possible with cyclones, which
can be an advantage at very dusty streams, or a disadvantage at
relatively clean streams.
Figure 35 shows a schematic diagram of a series cyclone sys-
tem that was described by Rusanov132 and is used in the Soviet
Union for obtaining particle size information. This device is
operated instack, but because of the rather large dimensions,
requires a 20 cm diameter port for entry.
Southern Research Institute, under EPA sponsorship, has de-
signed and built a prototype three-stage series cyclone system
for in-stack use.133 A sketch of this system is shown in Figure
36. it is designed to operate at 472 cm3/sec (1 ft3/min). The
Dso's for these cyclones are 3.0, 1.6, and 0.6 micrometer aero-
dynamic at 21°C. A 47 mm Gelman filter holder, (Gelman Instrument
Co., 600 South Wagner Road, Ann Arbor, MI 48106), is used as a
back up filter after the last cyclone. This series cyclone system
was designed for in-stack use and requires a 15 cm diameter sampling
port. Figure 37 shows a comparison of particle-size distributions
measured using the cyclone system and cascade impactors.:3^
Figure 38 illustrates a second generation EPA/Southern Re-
search series cyclone system now under development, containing
five cyclones and a back up filter. It is a compact system and
will fit through 10 cm diameter ports.. The initial prototype
was made of anodized aluminum with stainless steel connecting
hardware. A second prototype, for in-stack evaluation, is made
of titanium.
113
-------
INLET NOZZLE
CYCLONE 2
CYCLONE 1
3630-237
Figure 35. Series cyclone used in the U.S.S. R. for sizing flue gas aerosol
particles. From Rusanov. 132
114
-------
TO PUMP
BACKUP FILTER
CYCLONE 2
•CYCLONE 3
• NOZZLE
CYCLONE 1
3630-238
Figure 36. Schematic of the Southern Research Institute Three Series
Cyclone System.
-------
10°
CO
E
1
T3
o
•o
z
o
\-
00
cc
V)
Q
10"7
< 10-2
oc
LU
io-*
10-4
O SERIES CYCLONE
0-1 1.0 10
PARTICLE DIAMETER, Dgeo, //m
100
3630-239
Figure 37. Comparison of Southern Research Institute Three Series
Cyclone System data with cascade impactor curve
After Gcoding. 134
116
-------
CYCLONE 1
CYCLONE 4
CYCLONE 5
CYCLONE 2
CYCLONE 3
OUTLET
INLET NOZZLE
3630-240
Figure 38. The EPA/Southern Research Institute Five Series
Cyclone System.
117
-------
Figure 39 contains laboratory calibrations data for the five
cyclone prototype system. The D50's, at the test conditions,
are 0.32, 0.6, 1.3, 2.6, and 7.5 ym. A continuing research pro-
gram includes studies to investigate the dependence of the cyclone
cut points upon the sample flowrate and temperature so that the
behavior of the cyclones at stack conditions can be predicted
more accurately.131
The Acurex-Aerotherm Source Assessment Sampling System (SASS)
incorporates three cyclones and a back-up filter.135 Shown schemati-
cally in Figure 40, the SASS is designed to be operated at a flow-
rate of 3065 cm3/sec (6.5 ft3/min) with nominal cyclone D50's
of 10, 3, and 1 micrometer aerodynamic diameter at a gas tempera-
ture of 205°C. The cyclones, which are too large for in situ
sampling, are heated in an oven to keep the air stream from the
heated extractive probe at stack temperature or above the dew
point until the particulate is collected. Besides providing par-
ticle size distribution information, the cyclones collect gram
quantities of dust (due to the high flowrate) for later chemical
and biological analyses. The SASS train is available from Acurex-
Aerotherm, inc., 485 Clyde Ave., Mountain View, California 94042.
Small cyclone systems appear to be practical alternatives
to cascade impactors as instruments for measuring particle size
distributions in process streams under conditions where it is
appropriate to sample for longer periods and to obtain larger
samples.
Optical Particle Counters—
Figure 41 is a schematic diagram illustrating the principle
of operation for optical particle counters. A dilute aerosol
stream intersects the focus of a light beam to form an optical
"view volume." The photodetector is located so that no light
118
-------
COLLECTION EFFICIENCY,
V£>
f
O NO 5- f~-
-s po g gj
"•* co^. §•
«t» ^ » 5
: Oj
s-
q- Co ^-
111
££§
O O
-+> 3
I\O ^
^2
Q.
I
* 53
8-5?
o 8
-------
to
O
HEATER
CONTROLLER
STACK T.C. (
CONVECTION OVEN
FILTER
V"X-[ SS PROBE * I |
L.^ A
OVEN—"A A
T.C.
XAD-2
CARTRIDGE
CONDENSATE
COLLECTOR
DRY GAS METER
ORIFICE METERS
CENTRALIZED TEMPERATURE
AND PRESSURE READOUT
CONTROL MODULE
10 CFM VACUUM PUMP
GAS COOLER
GAS
TEMPERATURE
T.C.
IMP/COOLER
TRACE ELEMENT
COLLECTOR
\
IMPINGER
T.C.
3630-242
Figure 40. Schematic of the Acurex-Aerotherm Source Assessment
Sampling System (SASS).
-------
LIGHT TRAP
LAMP
SAMPLE AEROSOL
TO PUMP
PHOTOMULTIPLIER
3630-243
Figure 41. Schematic of an optical single particle coun ter.
121
-------
reaches its sensitive cathode except that scattered by particles
in the view volume. Thus, each particle that scatters light with
enough intensity will generate a current pulse at the photodetector,
and the amplitude of the pulse can be related to the particle
diameter. Optical particle counters yield real-time information
on particle size and concentration.
In an optical particle counter, the intensity of the scat-
tered light and the amplitude of the resulting current pulse depend
on the viewing angle, particle refractive index, particle shape,
and particle diameter. Different viewing angles and optical geome-
tires are chosen to optimize some aspect of the counter perform-
ance. For example, the use of near forward scattering will mini-
mize the dependence of the response on the particle refractive
index, but with a severe loss of resolution near 1 ym diameter.
The use of right angle scattering smooths out the response curve,
but the intensity is more dependent on the particle refractive
index. Figure 42 shows calibration data for near forward and
right angle scattering particle counters.136
Figure 43 illustrates some of the optical configurations
that are found in commercial particle counters. The pertinent
geometric and operating constants of the counters are summarized
in Table XV.
The commercial optical counters that are available now were
designed for laboratory work and have concentration limits of
a few hundred particles per cubic centimeter. The lower size
limit is nominally about 0.3 ym diameter. For use in studies
of industrial aerosols, dilution of the sample is required and
the useful upper limit in particle size has been limited by losses
in the dilution system to about 2.0 ym diameter.137 In addition,
the particle diameter that is measured is not aerodynamic, and
some assumptions must be made in order to compare optical with
122
-------
10
N5
UJ
O
O 0.5
CC
Ul
2
8
0.1
0.05
' ' '
O n = 1.6 (CARGILLE)
- • n = 1.6 (PSL)
A n = 1.4 (CARGILLE)
EXPERIMENTAL
n = 1.49 (DOP)
ROYCO PC 220
I Mill I I l I l i ill
0.5 1 5
PARTICLE DIAMETER, urn
a. Flight angle scattering.
10
10
. 11' MI
o
a.
O
1.0
I-
z
O 0.5
0.1
i—r
O n = 1.6 (CARGILLE)
• n = 1.6 (PSL)
A n = 1.4 (CARGILLE)
EXPERIMENTAL
n = 1.49 (DOP)
0.5
ROYCO PC 245
I I "I I i i
LL
1 5 10
PARTICLE DIAMETER, am
3630-244
b. Near forward scattering.
Figure 42. Experimental calibration curves for two optical particle
counters. After Willeke and Liu. 136
-------
SENSOR
CHAMBER. X
L^'
r=*^
\ \
X,\
PHOTOMULTIPLIER
T
lc
VIEW VOLUME
CALIBRATOR
CLIMET
SCATTERING )
PHOTODETECTOR \
MODULE I
CURVED MIRROR
90.9% REFLECTIVITY
'REFERENCE
PHOTOOETECTOR
MODULE
5 mm F.L.
PARABOLIC MIRROR
90% REFLECTIVITY
"0" RING SEAL
DUMP WINDOW
^ERODYNAMICALLY
FOCUSING INLET
SHEATH AIR
PMS LAS-200
W SAMPLE AIR
COLLECTION PUPIL
LIGHT LENS LENS
TRAP
PHOTOMULTIPLIER
REFLECTOR
DEFINING
OPERTURE
RELAY
LENS
AEROSOL
FLOW
PHOTOMULTIPLIER
TUBE
COLLECTING
LENSES
ROYCO 220
LAMP CONDENSER
.LENSES
ROYCO 225
PHOTOMULTIPLIER
TUBE
CONDENSER LENSES RELAY
LENS COLLECTING
| LE- —
SLIT
PHOTOMULTIPLIER
LIGHT
TRAP
VIEWING
VOLUME
ROYCO 245
BAND L40-1
Figure 43. Optical configurations for six commercial particle counters.
124
-------
TABLE XV.
CHARACTERISTICS OF COMMERCIAL, OPTICAL, PARTICLE COUNTERS
Bausch & Lomb Model 40-1
820 Linden Avel
Rochester, NY 14625
Climet Models 201, 208
Climet Inst. Co.
1620 W. Colton Ave.
Redlands, CA 92373
Illuminating Cone
Half Angle, y
13°
15
Particle Measuring Systems
1855 S. 57th Ct.
Boulder, CO 80301
Light Trap Half Collecting Aperture Inclination Between
Angle, o Half Angle, B Illuminating And Viewing
Collecting Cone Axis, i|) Volume
33°
35
53°
90
0°
0.5 mm3
0.5
Sampling
Rate
170 cm /min
7,080
N)
U1
Climet Model 150
Royco Model 218
Royco Inst.
41 Jefferson Dr.
Menlo Park, CA 94025
Royco Model 220
Royco Model 245
Royco Model 225
Tech Ecology Model 200
Tech Ecology, Inc.
645 N. Mary Ave.
Sunnyvalle, CA 94086
'. Tech Ecology Model 208
•Model LAS-200
12
5
24
5
5
5
5
\
18
11
' -• .
16
7
8
10
35
28
30
24
25
25
20
20
120
0
0
90
0
0
0
0
0
0.4
0.25
2.63
4.0
2.0
0.46
2.5
0.003
472
283
2,830
28,300
283 or 2,830
283
2,830
120 or 1,200
*632.8 mm laser ilium., all others are white light.
-------
aerodynamic data. (It is possible to "calibrate" an optical
counter, on a particulate source, to yield aerodynamic data.
This is done by using special calibration impactors,x38 or settling
chambers.139) Nevertheless, the ability to obtain real-time in-
formation can sometimes be very important and the special problems
in sampling with optical counters may be justified.
Diffusion Batteries with Condensation Nuclei Counters—
The classical technique for measuring the size distribution
of submicron particles employs the relationship between particle
diffusivity and diameter. In a diffusional sizing system, the
test aerosol is drawn, under conditions of laminar flow, through
a number of narrow, rectangular channels, a cluster of small bore
tubes, or a series of small mesh screens (diffusion batteries).
For a given particle diameter and diffusion battery geometry,
it is possible to predict the rate at which particles are lost
to the walls by diffusion, the rate being higher for smaller par-
ticles. The total number of particles penetrating the diffusion
battery is measured under several test conditions where the main
adjustable parameter is the aerosol retention time, and the par-
ticle-size distribution is calculated by means of suitable mathe-
matical deconvolution techniques. It is only necessary that the
particle detector that is used at the inlet and outlet of the
diffusion battery system respond to the total concentration,
by number, of the particles in the size range of interest. Fig-
ure 44 illustrates the geometry of a rectangular channel diffusion
battery, and Figure 45 a screen-type diffusion battery.
Condensation nuclei (CN) counters function on the principle
that particles act as nuclei for the condensation of water or
other condensable vapors in a supersaturated environment. This
process is used to detect and count particles in the 0.002 to
0.3 micron range (often referred to as condensation or Aitken
126
-------
CHANNEL DIMENSIONS
MULTI CHANNEL BATTERY
3630-246
Figure 44. A rectangular channel diffusion battery.
127
-------
10
SAMPLING
PORT (TYP)
SECTION CONTAINING
SCREENS (TYP)
3630-247
Figure 45. Screen type diffusion battery. The battery is 21 cm long,
4 cm in diameter, and contains 55, 635 mesh stainless
steel screens. After Sinclair. 145
128
-------
nuclei). In condensation nuclei detectors, a sample is withdrawn
from the gas stream, humidified, and brought to a supersaturated
condition by reducing the pressure. In this supersaturated con-
dition, condensation will be initiated on all particles larger
than a certain critical size and will continue as long as the
sample is supersaturated. This condensation process :forms a.
homogeneous aerosol, predominantly composed of the condensed vapor
containing one drop for each original particle whose size was
greater than the critical size appropriate to the degree of super-
saturation obtained; a greater degree of supersaturation is used
to initiate growth on smaller particles. The number of particles
that are formed is estimated from the light scattering properties
of the final aerosol.
Because of the nature of this process, measurements of very
high concentrations can be in error as a result of a lack of cor-
respondence between particle concentration and scattering or at-
tenuation of light. Additional errors can result from depletion
of the vapor available for condensation. Certain condensation
nuclei measuring techniques can also obtain information on the
size distribution of the nuclei; that is variations in the degree
of supersaturation will provide size discrimination by changing
the critical size for which condensation will occur. However,
the critical size for initiating condensation is also affected
by the volume fraction of water soluble material contained in
the original aerosol particle, so the critical size will be un-
certain unless the solubility of the aerosol particles is known. llf°
At very high degrees of supersaturation (about 400%) , solubility
effects are only minor and essentially all particles in the original
aerosol with diameters larger than 0.002 ym will initiate the
condensation process. Figure 46, after Haberl, illustrates the
condensation nuclei counter operating principle.1111
129
-------
PHOTO DETECTOR
RANGE
VACUUM
PUMP
INNER LIGHT STOP
OUTER LIGHT STOP
a
SOURCE LAMP
GEAR
MOTOR
3630-248
Figure 46. Diagram of a condensation nuclei counter. After
Haberl and Fusco. 141
130
-------
Figure 47 is a schematic diagram that illustrates an experi-
mental setup for measuring particle-size distributions by dif-
fusional means, and Figure 48 shows penetration curves for four
operating configurations.
Fuchs11*2 has reviewed diffusional sizing work up until 1956,
while'Sinclair,1*3'11""1*5 Breslin et al.,1 *6 Twomey, 1"9 Sansone
and Weyel,llf8 and Ragland, et al.,1*9 have reported more recent
work, both theoretical and experimental.
TSI Incorporated (500 Cardigan Rd., St. Paul, MN 55165) now
manufactures and sells screen-type diffusion batteries of Sinclair
design (Figure 45). These diffusion batteries are 21 cm long,
approximately 4 cm in diameter, weigh 0.9 kg, and contain 55
stainless steel screens of 635 mesh.
Four models of CN counters are now available commercially.
Two automatic, or motorized, types are the General Electric Model
CNC-2 (General Electric-Ordnance Systems, Electronics Systems
Division, Pittsfield, MA 01201) and the Environment-One Model
Rich 100 (Environment-One Corporation, Schenectady, NY 12301).
Small, manually operated, CN counters are also available from
Gardner Associates (Gardner Associates, Schenectady, NY 12301) ,
and Environment-One.
The General Electric CN counter has mechanically actuated
valves and is insensitive to moderate pressure variations at the
inlet. The aerosol concentration is measured by the detection
of scattered light from the test aerosol.
A disadvantage of the flow/valving arrangement in the General
Electric counter is the intermittent (I/sec) flow which introduces
severe pressure pulsations into the sampling system. This problem
131
-------
ANTI-PULSATION
DEVICE,
SAMPLE FROM
DILUTER
ANTI-
PULSATION
DEVICE
CN COUNTER
RETURN TO
DILUTER
CN COUNTER
D.B. 1
D.B.
D.B. 2
D.B. 2
D.B. 2
-•»• RETURN
TO DILUTER
3630-249
Figure 47. Diffusion battery and condensation nuclei counter layout
for fine particle sizing.
132
-------
55
Z*
o
0.01
0.02
0.03 0.04 0.05 0.1 0.2
PARTICLE DIAMETER, ,um
100
0.3 0.4 0.5
3630-250
Figure 48. Theoretical parallel plate diffusion battery penetration curves.
133
-------
has been minimized by the use of antipulsation devices consisting
of a rubber diaphram* •* "* or two metal cylinders connected by a
small orifice,it9 essentially pneumatic R-C networks.
The automatic Environment-One counter has some pneumatic
valves. A pressure of more than 5 cm of water at the inlet can
interrupt the operation. In the E-l, the aerosol concentration
is measured by light extinction. The sampling rate of the E-l
counter can be adjusted from about 0.6 to 4.2 £/min. Soderholm
has reported a modification to the E-l counter that replaces
pneumatic values with solenoidal ones.150
Because of the long retention time required for removal of
particles by diffusion, measurements with diffusion batteries and
CN counters are very time consuming. With the system described
by Ragland, et al., for example, approximately two hours are re-
quired to measure a particle-size distribution from 0.01 to 0.2
149
ym. Obviously, this method is best applied to stable aerosol
streams. It is possible that the new, smaller diffusion batteries
will allow much shorter sampling times, but pulsations in flow
may pose a serious problem for the low volume geometries.
Electrical Mobility—
An instrument that was developed for measuring laboratory
and ambient aerosols over the 0.003 to 1 ym range of diameters,
the electrical mobility analyzer, can also be applied to process
streams with a suitable sample dilution and cooling interface.
Figure 49 illustrates the relationship between the diameter
and electrical mobility of small aerosol particles. If particles
larger than those of minimum mobility are removed from the sample,
the remaining particles exhibit a monotonically decreasing mobil-
ity with increasing diameter. Several aerosol spectrometers,
134
-------
u
0)
E
>"
m
O
o
H
CC
10-
10"
O
o o
.o o
— D
O E = 5.0 x 105 V/m
Nt = 8.0 x 1011 sec/m3
a E = 1.5x105V/m
Nt = 3.2 x 1012 sec/m3
SHELLAC AEROSOL K = 3.2
0.0
0.2
0.4 0.6 0.8 1-0
PARTICLE DIAMETER, /um
1.2 1-4
3630-251
Figure 49. Particle mobility as a function of diameter for shellac aerosol
particles charged in a positive ion field. K is the dielectric
constant of the aerosol particles. After Cochet and Trillat. 155
135
-------
or mobility analyzers
«ameter-mobiiity relaptoc
t° their size'",>»,15, ,.tO classify particles according
°» which these devices operate19""5" J11"81"^ «e principle
conditions of homogeneous electric £ll, "" ^"^ Under
and then passed into the spectromL '"" ^ COnce"tration,
h of the device and a trll "' , "^ ^ "OWS d°™ "*
a knowled9e o£ the syBt™'!*r'e 8leCtric "eld is applied.
culated
by precipitation and o
of particles in a"alySeS °f the
oM
°' Metrical .erosol .nJ ^l^' *° ^ve!op a series
'he 0. of Minnesota devices is no ' °OJnmerc«l version of
« the Model 3030 (Pi9ure 51 " "I" ""*•** by M1, Tncorporated
«e size distribution of par :clhe.EAA la deSi^ to measure
1-0 urn diameter. since tL S1" the ra"9^ «rom 0.0032 to
»«» ^ ^ - 1000 ,g;:rd L0:::;"::8"0" ran9e f°r ^ —
gas aerosois. Uution.1B required for most industrial
»e EAA is operated in the foil. •
P»P draws the aerosol throuoh th ' ^"^ ' AS a vacu™
corona generated at a hiah , analyz« (see Pigure 51)
* the sarapl P9 s;: L?6"1" withi- ^
flOKS from the charter t ^r^31 Ch«9e- «» charged
cylinder of aerosol! *"' a"alyZer Sec«°" •• an
rod, to Whi rr°Undin9 3 ^ °f Clea" a^-
negative voltage can be applied,
136
-------
CHARGED PARTICLES
HV
CLEAN AIR
LAMINAR FLOW
\
\
SMALLER PARTICLES OF
HIGH ELECTRICAL MOBILITY
LARGER PARTICLES OF
LOW ELECTRICAL MOBILITY
3630-232
Figure 50. The electric mobility principle.
137
-------
CONTROL MODULE
ANALYZER OUTPUT SIGNAL -
DATA READ COMMAND - -
CYCLE START COMMAND -
CYCLE RESET COMMAND -
AEROSOL FLOWMETER READOUT
CHARTER CURRENT READOUT
-. CH&RGER VOLTAGE READOUT
AUTOMATIC HIGH VOLTAGE CONTROL AND READOUT
ELECTROMETER (ANALYZER CURRENT) READOUT
TOTAL FLCWMETER READOUT
U)
00
• EITERNAL
» DATA
•^ACOUISITIOtl
- SYSTEM
TO VACUUM PUUP
3630-253
Figure 51. Schematic of the Thermosystems Model 3030 Electrical
Aerosol Analyzer. After Sem. 158
-------
passes axially through the center of the analyzer tube. Particles
smaller than a certain size (with highest electrical mobility)
are drawn to the collecting rod when the voltage corresponding
to that size is on the rod. Larger particles pass through the
analyzer tube and are collected by a filter. The electrical charges
on these particles drain off through an electrometer, giving a
measure of current.
A step increase in rod voltage will cause particles of a
larger size to be collected by the rod with a resulting decrease
in electrometer current. This decrease in current is related
to the additional number of particles being collected. A total
of eleven voltage steps divide the 0.0032 to 1.0 micron size range
of the instrument into ten equal logarithmic size intervals.
Different size intervals can be programmed by means of an optional
plug-in memory card.
The electrical aerosol analyzer can be operated either auto-
matically or manually. In the automatic mode, the analyzer steps
through the entire size range. For size and concentration monitor-
ing over an extended period of time, the analyzer may be inter-
mittently triggered by an external timer. The standard readout
consists of a digital display within the control circuit module,
although a chart recorder output is available. It is almost
always advantageous to use a strip chart recorder to record the
data. This allows the operator to identify a stable reading that
may be superimposed on source variations and also gives a per-
manent record of the raw data.1'*8 The EAA requires only two
minutes to perform a complete size distribution analysis, which
generally makes it advantageous to use, especially on stable
sources.
139
-------
When the EAA is applied to fluctuating sources a peculiar
problem arises. The instrument reading is cumulative, and it
is impossible to tell whether variations in the reading reflect
changes in the distribution or concentration of particles; hence,
recordings that show rapid fluctuations in amplitude must be
interpreted with great care. The lack of sensitivity can also
be a problem at extremely clean sources.
Laboratory Measurements
Measurements of the size distribution of particles that have
been collected in the field and transported to a laboratory must
be interpreted with great caution, if not skepticism. It is dif-
ficult to collect representative samples in the first place, and
it is almost impossible to reconstruct the original size distri-
bution under laboratory conditions. For example, one cannot dis-
tinguish from laboratory measurements, whether or not some of
the particles existed in the process gas stream as agglomerates
of smaller particles. In spite of the limitations inherent in
laboratory methods, they must be used in some instances to deter-
mine particle size and to segregate particles for analysis of
their composition or other properties of interest. This section
contains a discussion of some of the "standard" techniques used
for particle size analysis of dust samples.
Sedimentation and Elutriation—
Elutriators and sedimentation devices separate particles
that are dispersed in a fluid according to their settling velo-
cities due to the acceleration of the earth's gravity. The set-
tling, or terminal, velocity of a particle in air is
gpCD2
v = -
140
-------
where
P = the particle density,
C = the slip correction factor,
y = the viscosity of air,
D = the particle Stokes diameter, and
g = the gravitational acceleration.
Large particles in a quiescent aerosol will settle to the
bottom region of the chamber more quickly than smaller particles
that have smaller settling velocities. This principle is used
in gravitational sedimentation and elutriation to obtain particle
size distributions of polydisperse aerosols. In elutriation,
the air is made to flow upward so that particles with settling
velocities equal to or less than the air velocity will have a
net velocity upward and particles which have settling velocities
greater than the air velocity will move downward.
There are a number of commercial devices and methods having
varying requirements of dust amounts and giving different ranges
of size distributions, with a minimum size usually no smaller
than two micrometers.159'160 An important disadvantage is the
inability of most sedimentation and elutriation devices to give
good size resolution. Another disadvantage is the length of time
(sometimes several hours) required to use some of the methods.
Popular methods of sedimentation sizing employ a pan balance
which weighs the amount of sediment falling on it from a suspen- '
sion, and the pipette, which collects the particles in a small
Pipette at the base of a large chamber. Cahn's electronic micro-
balance, (Cahn instrument Company, 7500 Jefferson St., Paramount,
CA 90723), has an attachment that permits it to function as a
settling chamber. Perhaps the most popular elutriator is the
Roller particle size analyzer illustrated in Figure 52 (the Roller
particle size analyzer is available from the American Standard
Instrument Co., inc., Silver Spring, MD).
141
-------
SEPARATOR TUBE
AIR SUPPLY
FLEXIBLE JOINT
POWDER
CIRCULATION
3630-254
Figure 52. The Roller efutriator. After Allen. 160
142
-------
An instrument that measures the size distribution of par-
ticles in a liquid suspension is the Xray Sedigraph, (Micromeritics
Instrument Corporation, 800 Goshen Springs Road, Norcross, GA
30071) . The sample is continually stirred until the sampling
period starts. The concentration of the particles is monitored
by means of the extinction of a collimated x~ ray'beam. Upon sampl-
ing, the x-ray beam is moved upward mechanically to shorten the
sampling time that is required. The particle-size distribution
is plotted automatically. The reported range of sensitivity of
the X-ray Sedigraph is from 0.1 to 100 ym.
Centrifuges—
Aerosol centrifuges provide a laboratory method of size-
classifying particles according to their aerodynamic diameters.
The advantage over elutriators is that the settling, or preci-
pitation, process is speeded up by the large centrifugal accelera-
tion and the effective sizing range of some instruments includes
much smaller particles. The sample dust is introduced in the
device as an aerosol and enters a chamber which contains a centri-
fugal force field.
In one type of aerosol centrifuge, the larger particles over-
come the viscous forces of the fluid and migrate to the wall of
the chamber, while the smaller particles remain suspended. After
the two size fractions are separated, one of them is reintroduced
into the device and is fractionated further, using a different
spin speed to give a slightly different centrifugal force. This
is repeated as many times as desired to give an adequate size
distribution. One of the more popular lab instruments using this
technique is the Bahco microparticle classifier, which is illus-
trated in Figure 53, and is available commercially from the Harry
W. Dietert Company, Detroit 4, Michigan. The cutoff size can
be varied from about two to fifty micrometers to give size dis-
tribution characterization of a 7 gm dust sample. A similar
143
-------
10 11 12 13
SCHEMATIC DIAGRAM
1. Electric Motor 9.
2. Threaded Spindle 10.
3. Symmetrical Disc 11.
4. Sifting Chamber 12.
5. Container 13.
6. Housing 14.
7. Top Edge 15.
8. Radial Vanes 16.
Feed Point
Feed Hole
Rotor
Rotary Duct
Feed Slot
Fan Wheel Outlet
Grading Member
Throttle
3630-255
Figure 53. The Bahco microparticle classifier.
144
-------
instrument is the B.C.U.K.A. (British
Association, Leatherhead, Surrey, U.K
which has a range of four to twenty-six micrometers.
Coal Utilization Research
) centrifugal elutriator
i e i
In the second type of centrifuge, the device is run continu-
ously, and the particle size distribution is determined from the
positions where the particles are deposited. Examples are spiral
centrifuges developed by Goetz, et al.,J 62'16 3'16** (Figure 54)
and by Stober and Flachsbart,l65 (Figure 55) that can classify
polydisperse dust samples with particles from a few hundredths
of a micron to approximately two micron in diameter. The conifuge,
first built by Sawyer and Walton166 and modified several times
since then,167'168 is useful in the study of aerodynamic shape
factor, but can also be used for the determination of size dis-
tributions especially for particles having aerodynamic diameters
smaller than twenty-five micrometers (see Figure 56). In contin-
uously operating centrifuges, the particles are generally deposited
onto a foil strip, where their position yields a measure of their
size, and their number is obtained by microscopy, radiation, or
by weighing segments of the foil.
Microscopy—
Microscopic analysis has long been regarded as the estab-
lished, fundamental technique of counting and sizing particles
that the human eye cannot comfortably see. Usually, the method
involves one person, a microscope, and a slide prepared with a
sample of the aerosol to be measured. A random selection of the
particles would then be measured and counted, with notable charac-
teristics of color, shape, transparency, or composition duly
recorded. The most difficult task, especially since the advent
of sophisticated computerized equipment has made counting and
sizing easier, is the preparation of a slide which contains a
representative sample of the aerosol.
145
-------
COLLECTING
FOIL
JET
ORIFICE
INLET TUBE
3630-256
Figure 54. A cut-away sketch of the Goetz Aerosol Spectrometer
spiral centrifuge. In assembled form the vertical axes
(1) coincide and the horizontal arrows (2) coincide.
After Gerber. 164
146
-------
THERMOCONTROLLED
WATER
AEROSOL
ENTRANCE
SPIRAL
DUCT
SUCTION PUMP
THERMOCONTROLLED
WATER
3630-257
Figure 55. Cross-sectional sketch of the Stober Centrifuge.
After Stober and Flachsbart.
147
-------
PARTICLE STREAM
CLEAN AIR.
LARGER PARTICLES
COLLECTED HERE
OUTER CONE
SMALLER PARTICLES
COLLECTED HERE
AXIS OF ROTATION
3630-258
Figure 56. Cross-sectional sketch of a conifuge.
148
-------
A careful technique is required to obtain a slide sample which
isn't biased toward large or small particles, does not contain
agglomerations which weren't present in the stack, does not break
up agglomerations which were present in the stack, is not too
dense or too sparse, and has not been contaminated in the process
of preparation. Different methods of slide preparation for optical
and photographic microscopy are discussed by Cadle159 and Allen.160
A particularly good discussion of particle analysis through micro-
scopy is given in Volume I of the McCrdne Particle Atlas.iss One
main disadvantage of microscopic analysis is the type of diameter
measured. Depending on the shape of the particles, several dif-
ferent types of diameter are used to characterize the size of
the particle. Three commonly used types of diameter are shown
in Figure 57 with their definitions. However, for most control
and standards work, the diameter of interest is the aerodynamic
diameter, which is based on the particle's behavior in air. In
these cases, the data from microscopic analysis is helpful only
insofar as it can be related to the particular need of the experi-
menter .
Particle sizes which can be easily studied on optical micro-
scopes range from about .2 to 100 micrometers. Electron micro-
scopes have decreased the smallest diameter of particles capable
of being analyzed by microscopy down to 0.001 micrometer. Both
scanning and transmission electron microscopes provide much informa-
tion on surface features, agglomeration, size, composition and
shape of particles in size ranges below that of optical micro-
scopes. Computerized scanning devices have increased the analyzing
ability of present day microscopes and simplified counting and
sizing.
Several commercial laboratories are equipped to provide physi-
cal and structural characterizations of dust samples quickly and
fairly inexpensively.
149
-------
F - Feret's diameter, the distance between two tangents on opposite
sides of the particle, parallel to a fixed direction.
M
Martin's diameter, the length of the line which bisects the image
of the particle, parallel to a fixed direction.
da - Diameter, of a circle having the same projected area as the particle in the
plane of the surface on which it rests.
3630-259
Figure 57. Three diameters used to estimate particle size in microscopic
analysis.
150
-------
Sieves—
Because of its relatively large lower particle size limit
(50-75 micrometers), sieving has a limited use for characterzing
most industrial sources today. However, for particles within
its workable size range, sieving can be a very accurate technique,
yielding adequate amounts of particles in each size range for
thorough chemical analysis.
Sieving, one of the oldest ways of sizing particles geomet-
rically, is the process by which a polydisperse powder is passed
through a series of screens with progressively smaller openings
until it is classified as desired. The lower size limit is set
by the size of the openings of the smallest available screen,
usually a woven wire cloth. Recently, micro etched screens have
become available. In the future, the lower size limit may be
lowered by using membrane filters which can be made with smaller
holes than woven fine wire cloth.
Sieves are available from several manufacturers in four
standard size series: Tyler, U.S., British, and German. See
Table XVI for a comparison of these series. Tyler screens are
manufactured by the W.S. Tyler Co., Cleveland, Ohio.
Other methods of size classification using sieving principles
are currently being studied and improved. Wet sieving is useful
for material originally suspended in a liquid or which forms
aggregates when dry-sieved. Air-jet sieving, where the particles
are "shaken" by a jet of air directed upward through a portion
oi the sieve, has been found to be quicker and more reproducible
than hand or machine sieving, although smaller amounts of powder
(5 to 10 g) are generally used. Felvation170 (using sieves in
conjunction with elutriation) and "sonic sifting"*71 (oscillation
of the air column in which the particles are suspended in a set
of sieves) are similar techniques that employ this principle.
151
-------
TABLE XVI
COMPARISON TABLE OF COMMON SIEVE SERIES
British
Tyler
Equiv.
Mesh
3.5
4
5
6
7
8
9
10
12
14
16
20
24
28
32
35
42
48
60
65
80
100
115
150
170
200
250
270
325
400
Tyler
b
U.S.
ASTM
c
Openings
in mm.
5.613
4.699
3.962
3.327
2.794
2.362
1.981
1.651
1.397
1.168
0.991
0.833
0.701
0.589
0.495
0 . 417
0.351
0.295
0.208
0.208
0.175
0.147
0.124
0.104
0.088
0.074
0.061
0.053
0.043
0.038
Mesh
No.
3.5
4
5
6
7
8
10
12
14
16
18
20
25
30
35
40
45
50
60
70
80
100
120
140
170
299
230
270
325
400
u.s.D
Openings
in mm.
5.66
4.76
4.00
3.36
2.83
2.38
2.00
1.68
1.41
1.19
1.00
0.84
0.71
0.59
0.50
0.42
0.35
0.297
0.250
0.210
0.177
0.149
0.125
0.105
0.088
0.974
0.062
0.053
0.044
0.037
Standard0
Mesh
No.
5
6
7
8
10
12
14
16
18
22
25
30
36
44
52
60
72
85
100
120
150
170
200
240
300
German
DINd
Openings DIN Mesh per Openings
in mm. No. sq. cm. in mm.
3.353
2.812
2.411
2.057
1.676
1.405
1.204
1.003
0.853
0.699
0.599
0.500
0.422
0.353
0.295
0.251
0.211
0.178
0.152
0.124
0.104
0.089
0.076
0.066
0.053
1
2
2.5
3
4
5
6
8
10
11
12
14
16
20
24
30
40
50
60
70
80
1
4
6.25
9
16
25
36
64
100
121
144
196
256
400
576
900
1600
2500
3600
4900
6400
100 10000
6.000
3.000
2.400
2.000
1.500
1.200
1.020
0.750
0.600
0.540
0.490
0.430
0.385
0.300
0.250
0.200
0.150
0.120
0.102
0.088
0.075
0.060
Standard Screen Scale Series.
Sieve Series
E-ll.
British Standard
d
(Fine
Sieve
Series) , National
Bureau of
Standards
Series, British Standards Institution,
LC-584
London
and
BS-410:1943.
German Standard Sieve Series, German Standard Specification DIN 1171.
152
-------
Coulter Counter--
Figure 58 illustrates the principle by which Coulter counters
(Coulter Electronics, Inc., 590 West 20th Street, Hialeah, FL
33010) operate. Particles suspended in an electrolyte are forced
through a small aperture in which an electric current has been
established. The particles passing through the aperture displace
the electrolyte, and if the conductivity of the particle is dif-
ferent from that of the electrolyte, an electrical pulse of ampli-
tude proportional to the particle-electrolyte interface volume
will be seen. A special pulse height analyzer is provided to
convert the electronic data into a size distribution. A biblio-
graphy of publications related to the operation of the Coulter
counter has been compiled by the manufacturer and is available
on request.
The requirement that the sample be suspended in an electro-
lyte may limit the application of the Coulter counter to very
inert particles, and those for which the response to a particular
electrolyte-particle suspension can be determined by calibration.
NEW TECHNIQUES
Several techniques are under development that may signifi-
cantly improve the technology of particulate sampling. Some of
the more promising methods are discussed briefly in this section.
It should be noted that most of these now require special skills
or knowledge in their application, and for many systems, only
one prototype has been fabricated.
Low Pressure Impactors
It is possible to extend the sizing capability of cascade
impactors to submicron particles by operating the devices at pres-
sures of 0.01 to 0.1 atmosphere. If all operating parameters
153
-------
THRESHOLD
3630-260
Figure 58. Operating principle of the Coulter counter.
Courtesy of Coulter Electronics.
154
-------
except the pressure are held constant, the cut point, or D50,
is inversely proportional to the square root of the slip correc-
tion factor:
D50 a C~h (10)
Since C increases rapidly with decreasing pressure, cut points
of 0.02 vim or less can be obtained.1 7 2 '* 7 3'l 7 4 Pilat175'176 has
developed and tested a low pressure impactor for sampling from
process streams. (See Figure 59).
Figure 60 shows the sampling train used by Pilat in his ex-
periments. Two impactors are operated in series. The first im-
pactor is of a conventional design with cut points from about
0.3 to 20 urn diameter. The second impactor is operated at reduced
pressure with cut points from about 0.03 to 0.2 ym diameter.
The sampling train contains a low pressure drop condenser; a
double vane, leakless, high vacuum pump; a control box with pres-
sure gauges, thermocouple pyrometers, and valves; and a dry gas
meter. A 90 mm diameter filter holder is used down stream from
the second impactor. The maximum flowrate is approximately 50
liters/minute. The main problems associated with this technique
are the large bulky equipment required, particle bounce, and the
very low mass collected on each stage.
Impactors with Beta Radiation Attenuation Sensors
Beta attenuation has some appeal as a detection mechanism
for cascade impactors in air pollution work because the impactor
separates the particles according to their aerodynamic behavior,
which is desirable, and the beta attenuation yields a direct mea-
sure of the amount of mass collected, which is also desirable.
155
-------
INLET NOZZLE
•JET STAGE
^^COLLECTION PLATE
TO VACUUM PUMP
TO PRESSURE GAUGE
3630-261
Figure 59. Cross section of prototype Mark IV University of
Washington Source Test Cascade Impactor.
156
-------
BCURA
CYCLONE
STAGE
PRESSURE
TAPS
DRY GAS
METER
3630-262
Figure 60. Samp/ing train utilizing a low pressure impactor.
After Pilat. 176
157
-------
Figure 61 illustrates how beta particle sources and detectors
can be used to monitor the deposition of dust under an impaction
jet. In one instance, the tape moves continuously and two de-
tector-source units are used to obtain a zero reading and to
measure the amount of mass collected. In the second example,
the tape is periodically stepped forward, and a zero reading is
taken from the initially clean area under the impaction jet.
A virtual impactor with beta attenuation detectors has been
developed for air pollution studies. The instrument is large,
and must be operated outside of the stack. The sample is trans-
ported to the instrument through a probe. A complete analysis
and evaluation of this system has not been published.177
An attempt was made to develop a seven-stage beta impactor
for in situ operation. The device was constructed, but never
operated in stack because suitable beta detectors were not avail-
able.178
Other problems are: selecting suitable tapes and greases
for compatability with the beta monitor and for good particle
retention, designing the impactor to give a uniform deposit, and
the mechanical problems associated with designing such a complex
system to be operated in a harsh, dirty environment.
In summary, it seems unlikely that multiple stage impactors
with beta attenuation as a detection mechanism can be made prac-
tical for in situ use in the foreseeable future.,
Cascade Impactors with Piezoelectric Crystal Sensors
Carpenter and Brenchly 179 and Chuan*8° have developed and
tested multiple-stage cascade impactors with piezoelectric crys-
tals on each stage to monitor the rate and amount of mass col-
lected. Chuan's impactor is now sold commercially by Berkeley
Controls, Inc (2700 Du Pont Drive, Irvine, CA 92714).
158
-------
IMPACTOR JET
D-0 PARTICLE DETECTOR
S-/3 PARTICLE DETECTOR
CLEAN TAPE
SOILED TAPE
TIME
IMPACTOR JET
CLEAN TAPE
TIME
DEPOSITION OF DUST
H P*~ TAPE STEPS FORWARD
3630-263
159
-------
Chuan's impactor has ten stages, with the cut points reported
to be from 0.05 to about 25 pm. Because of the extreme sensi-
tivity of the instrument (and upper limit on mass accumulation),
it is more suitable for ambient than stack work, where sample
extraction and dilution would be required. The impactor must
be disassembled, and each crystal cleaned before the limits of
the linear range are reached. In a typical urban atmosphere, this
limit is reached in about 2 hours. 57
The best application of piezoelectric impactors would seem
to be monitoring real time flucturations in fairly dilute aerosols.
(See also Piezoelectric Mass Monitors in Section 2.)
Virtual Impactors
Figure 62 illustrates the operating principle of virtual
impactors, sometimes called centripeters, dichotomous samplers,
or stagnation impactors. The aerosol jet is directed toward a
stagnant zone, or an opposing jet of clean gas, and a "virtual"
surface is formed at the boundary between the aerosol jet and
air space or opposing jet. The jet streamlines are diverted as
in a normal impactor. Particles of larger Stokes number impinge
on (and pass through) the virtual surface, while those having
smaller Stokes numbers follow the streamlines. The principal
advantage over conventional impactors is that both aerosol frac-
tions can be preserved, one containing the larger particles and
another containing the smaller particles. Also, uniform depo-
sition of the sized particles on filters is possible with virtual
impactors, whereas conventional impactors yield conical deposits
underneath the jets.
;>cv<.-r ,j I mult iplo-ntaye, virtual impactors have been developed,
all for the purpose of obtaining large quantities of sized par-
ticles, in uniform deposits, for subsequent analysis. Hounam
and Sherwood originated the concept of the virtual impactor, and
160
-------
D
DUSTY GAS
SMALLER PARTICLES
(MAJORITY OF FLOW)
SMALLER PARTICLES
LARGER
PARTICLES
VIRTUAL IMPACTION
SURFACE
CLEAN GAS
FILTER SMALL FRACTION
OF FLOW
3630-264
Figure 62. Virtual impactors (centripeters, dichotomus samplers, stagnation
impactors) a. impingemen t in to a stagnan t air space; b. opposed
axisymmetric jets.
161
-------
developed a four stage (three jet stages and a backup filter)
device for sampling radioactive aerosols.181 In laboratory cali-
bration studies with their impactor, Hounam and Sherwood found
that a substantial fraction of the sample aerosol was not col-
lected on the filters, but rather on the walls of the instrument.
A commercial version of the Hounam and Sherwood virtual impactor
is available from BGI, Inc., 58 Guinan St., Waltham, MA 02154.
Conner,182 almost simultaneously with Hounam and Sherwood,
developed a two-stage virtual impactor (one jet stage and a backup
filter) for collecting large samples of particles, one above and
one below one micron in diameter. Connor demonstrated that the
sharpness of cut for the virtual impactor was comparable with that
of conventional impactors, and also that there is a definite
requirement for a controlled flow of air into the collection
nozzle (through the filter) in order to establish optimum per-
formance.
• Peterson183 and Loo , et al.,181+ have developed virtual im-
pactors for sampling urban aerosols, and the Peterson version
is now sold by Sierra Instruments Co. (P.O. Box 909, Village Sq.,
Carmel Valley, CA 93924).
Schott and Ranz185 have developed and tested, in the labora-
tory, a "jet cone" impactor that is a hybrid design wherein the
aerosol jet is directed toward a conical surface. Any reentrained
particles are carried into a secondary collection zone by a por-
tion of the flow, as in virtual impactors. This concept would
seem to have the advantages of a virtual impactor, without the
disadvantage of an ill defined (virtual) impaction surface.
Research on the opposed-jet concept has been done by Luna,181
who originated the idea, Brooks,187 and Willeke.188 These instru-
ments appear to be quite sensitive to the geometry and alignment
of various components, and further research is needed to fully
evaluate their potential as useful tools in pollution analysis
and research.
162
-------
Virtual impactors have played a minor role in pollution
studies to date, with very little, if any, application to process
streams. The major advantage of these devices appears to be the
capability of using them to obtain uniformly deposited films of
dust for analysis by X-ray fluorescence, or any other technique
that requires similar sample preparation, and apparently less
particle reentrainment.
Optical Measurement Techniques
When light is incident upon a particle, some of the radiation
will be absorbed, some scattered, and some polarization will
occur. The exact nature and magnitude of the interaction depends
on the ratio of the particle diameter to the wavelength of the
radiation, and the shape and composition of the particle. Thus,
measurements can be envisioned that would yield information of
particle size, shape, concentration, and composition. It appears,
from the information now available, that optical methods offer
the greatest hope for a major advance in the technology of par-
ticulate sampling. Any successful instrument, however, must be
able to function in a harsh environment where extremes in tem-
perature, particle concentration, corrosion, etc. are found.,
Also, the parameter that is measured should ideally be related
to the aerodynamic diameter of the particles.
Although there are no proven commercial instruments available
for measuring particle size distributions in process streams,
a variety of methods have been proposed, and several prototype
instruments developed. This section contains brief descriptions
of some of the promising methods.
Hodkinson189 suggested a method of minimizing the dependence
on particle reiractive index in sizing measurements from a study
of the Fraunhofer diffraction formulation at small angles of
163
-------
forward scattering. The basis of this method involves measurement
of the intensity of light scattered by a single particle at two
small angles, and calculation of the ratio of the two intensities.
Figure 63 shows the scattered light intensity versus scatter-
ing angle for two spherical particles of equal diameter. One
particle is a glassy, nonabsorbing sphere with refractive index
equal to 1.55. The other is absorbing, such as carbon, and has
a refractive index of 1.96-0.661. For small angles, the intensity
of the scattered radiation is approximately the same for both
spheres, although at large angles there is a difference of orders
of magnitude.
Shofner et al.,19° Gravatt,191 and Chan192 have developed
prototype systems for particle sizing that are based on the in-
tensity ratio concept of Hodkinson. Shofner's system, the "PILLS-
IV", is designed for in situ operation. The intensities of the
scattered light pulses at the angles 6a and 62 are normalized
to the reference pulse at 6 = 0° for synchronization and to com-
pensate for fluctuations in intensity of the laser source. The
optics and sensors are kept clean and cool by the use of a purge
air system.
The laser used in the PILLS-IV system is a semiconductor
junction diode (X = 0.9 ym). The useful size range for particle
sizing is from 0.2 to 3.0 ym diameter. Shofner states that the
view volume of his system is approximately 2xlQ-7 cm3. The upper
concentration limit for single particle counters is determined
by the requirement that the probability of more than one particle
appearing in the view volume at a given time be much less than
unity. For Shofner's system this would set the concentration
limit at approximately 106 particles/cm3, a value much higher
than for conventional single particle counters.
164
-------
>-
CO
I-
I I
DIAMETER = 1.0 urn. \ = 514.5 nm
0.1
80 100 120 140 160
180
SCATTERING ANGLE, degrees
3630-265
Figure 63. Scattered light intensity versus scattering angle for two spherical
particles of equal diameter. The solid^curye is for a glassy, non-
absorbing sphere and the dashed curve is for an absorbing sphere.
After Gravatt. 191
165
-------
Gooding131* has tested the PILLS-IV prototype system simul-
taneously with several inertial sizing devices to measure particle
sizes. The PILLS-IV data did not agree well with impactor data
for this source, and further calibration may be required.
An optical particle sizing device developed by R.G. Knollen-
berg.193 for atmospheric application may be adaptable to industrial
emission measurements. The device is, in some respects, similar
to a conventional, near forward scattering, single particle coun-
ter, except that the sensing zone is contained within an open
cavity laser. This configuration yields very high illumination
levels, permitting the detection of particles smaller than those
sized by most light-scattering instruments. The method also has
potential for the measurement of particle velocities (across the
beam) through an optical heterodyne effect. The particle imaging
and detection system of the device permits the rejection of signals
from particles outside the nominal view volume so that the sample
gas stream flows in a fairly unrestricted manner through the
relatively large open cavity. However, substantial modifications
in terms of cooling and purge air for the optical components would
be required before the device could be used in-stack.
Systems employing optical fourier transforms to obtain par-
ticle-size distributions in the 5-100 ym diameter range have been
described by Cornillaut1 9* and McSweeny.195 In this technique
a moderately large diameter, collimated beam of spatially fil-
tered, coherent light is used to produce a diffraction pattern
from all particles in a known volume of space. The diffraction
pattern is imaged on a detector array having circular symmetry,
permitting a determination of the radial distribution of the
intensity of light in the superposed diffraction patterns of the
randomly distributed particles in the view volume. A numerical
inversion process, which can be adequately achieved by matrix
multiplication of the intensity data by an inversion matrix,
provides the required size distribution. The inversion process
166
-------
can be carried out in real time using a mini-computer. With the
proper selection of measurement points in the diffraction pattern
the size interval covered by the technique can be extended outside
the previously mentioned 5-100 jam range.
Wertheimer and Wilcock196 developed an approximate technique
based upon diffraction theory to determine the average size of
a distribution of particles (large particles, where a>20). The
technique utilizes three rotating masks of different shapes to
spatially filter the detected signal. With the r2 mask, the
detected signal from light scattered by a single particle is
proportional to the second power of the particle radius. Simi-
larly, with the r^r1*) mask, the detected signal is proportional
to the third (fourth) power of the particle radius. For many
particles in the field of view of the detector, the three detected
signals are proportional to the second, third, and fourth moments
of the distribution. These signals yield the volume mean radius,
the area mean radius, and the standard deviation of the area
distribution. Wertheimer and Wilcox demonstrated the usefulness
of this technique using particles from 4 to 83 ym in diameter.
Consideration of the principles on which this method is based
suggests that it is resistive to refractive index to the same
degree as forward scattering particle counter. The method has
been incorporated into a commercial instrument that is sold by
Leeds and Northrup Co. (Dickerson Road, North Wales, PA).
Imaging systems, either of a direct type or of a type using
reconstructed images from holograms, have not been widely used
for size distribution analysis in flue gases, but have been used
routinely for work with liquid aerosols-particularly to size
aerosols produced by various types of spray nozzles. In both
direct imaging systems and holographic systems, a short light
pulse of high intensity is used to illuminate the particles.
The pulse durations from available illuminators are short enough
to effectively eliminate blur due to particle motion for velo-
cities up to 300 m/sec.
167
-------
Flash Television Particle counters providing real time size
distributions have been described by Hotham 197 using pulsed ultra-
violet laser illumination, and by Simmons and Dominic198 using
xenon flash tubes for illuminators. The reported range for size
distribution determinations for the latter device is 0.3 to 10,000
ym. In Hotham1s system, the image size analysis can be performed
instantaneously on a basis of image height, length, perimeter
or projected area. Dynamic processes and particle motion can
be observed and studied by use of a video tape recorder. The
view volume in systems of this type is defined electronically
in width and height to exclude particles which are only partially
within the field of view while focus detection circuits are used
to define the depth of the view volume and exclude out-of-focas
particles. Because of cost and the practical difficulties in-
volved in the use of such a system in a flue gas environment,
applications of these systems will probably be limited to special
•research service.
Holography as a technique for investigating aerosols has
several advantages over most of the methods previously described.
The aerosol is not disturbed by the measurement process, a large
depth of field is possible and, as in the flash T.V. method, the
particles can be effectively "stopped" for examination at speeds
up to a few hundred meters per second. Typical system resolution
limits, however, result in a lower limit in sensitivity for par-
ticle sizing of about 5 ym. By double-pulsing the laser illumi-
nator one can obtain holograms which permit the determination
of particle velocities in three dimensions. Matthews and Kemp199
have described the use of a two-beam holographic system for de-
termining the spatial distribution of limestone particulate in-
jected into an operating 140 MW 24-foot-wide pulverized coal fired
steam boiler. The systems described by both Matthews and Kemp,
and Allen, et al., utilize low angle forward scattered light from
a pulsed ruby laser. By using pulsed ultraviolet laser illumi-
nation, some gain can probably be achieved in resolving smaller
168
-------
particles. Image Analyzing Computers, Inc., of Monsey, NY, offers
an automatic analyzer for reading out and analyzing aerosol data
from holograms, making it possible to eliminate manual analysis.
The analyzer was designed in collaboration with the Meteorological
Office and the Chemical Defense Establishment at Porton Down,
England. When used with holograms obtained with a pulsed ruby
laser the analyzer provides information on the size, shape, and
location of all particles having diameters larger than a few
micrometers in a sample volume up to one liter in size.
Laser Doppler Velocimeters (LDV) are routinely used for mea-
suring the velocity of gases, and these instruments can also be
used to obtain information of particle size. In an LDV, the laser
beam is split into two components which intersect at a small angle
at the point where the measurement is to be made. The beams form
interference fringes in the zone where they intersect, and if
a particle passes through this zone, light is scattered from the
bright fringes to a photomultiplier. The frequency of the a-c
component of the resulting pulse is related to the velocity with
which the particle intersects the fringes (and hence the velocity
of the gas). The ratio of the a-c component of the pulse to the
d-c component (seen at certain angles) is proportional to the
particle size. Farmer,200 Robinson et. al.,201 Adrian and Or-
loff,202 and Roberds203 have done experimental and theoretical
studies of LDV systems designed to enhance the sensitivity to
particle size. A commercial LDV particle spectrometer based on
Farmer's work is available from Spectron Development Laboratories,
Inc. (Tullahoma, TN 37388). To our knowledge, the instrument
has not been applied to in-stack measurements, and no report on
its performance is available. An advantage of LDV systems is
the potential for in situ sampling with little or no perturba-
tion of the sample. Disadvantages are the sensitivity to particle
refraction index and the complexity of the system.
169
-------
Wilson201* has described a technique where individual par-
ticles are accelerated through an orifice, and their velocity
measured by means of an LDV. The difference between the velocity
of the carrier gas and the particles is related to the aerodynamic
diameter of the particles. This technique is still in a labora-
tory development stage, and it is difficult to assess its potential
applicability to process streams at this time.
Hot Wire Anemometry
An electronic instrument has been developed by Medecki and
Magnus205 of KLD Associates, Inc. (Huntington, NY, USA) for sizing
liquid droplets, especially in scrubbers. The instrument operates
by inertial deposition of 1 ym to 600 jam spray droplets on a 5
pm diameter by 1 mm long platinum sensing element of the type
used in hot-wire anemometry. Droplets smaller than 1 ym can be
measured with a change in sensor geometry. The sensing element
is electrically heated to a predetermined temperature. Impinging
particles cool the sensing element, resulting in changes in resis-
tance which are related to the sizes of the impinging droplets.
The commercially available version of the device provides con-
centration outputs in six selectable size channels. Size cali-
brations for the channels are for water droplets; however, the
application of the method is not, in principle, limited to water.
Because the device is essentially a modification of a hot-wire
anemometer, it could also theoretically be used to measure flow
velocity and temperature, permitting impingement rates to be con-
verted to aerosol concentrations. Although commercial prototypes
are available now, this instrument is still under development
and detailed performance analyses are not available.
170
-------
Large Volume Samplers
McFarland and Bertch206 have developed a system for collect-
ing bulk samples of classified dust for subsequent use in health
related research (see Figure 64) . The system contains, in series,
two cyclones, a virtual impactor, and a bag filter. The D50's of
the cyclones are 10 and 7 ym, and that of the virtual impactor
is 5 ym at a sample flowrate of 850 1/min. The particulate col-
lection components are housed in an insulated enclosure that is
2.7 x 1 x 2 m. In sampling for 12 days at the outlet of an elec-
trostatic precipitator, McFarland collected 8.1 kg of dust: 5.4
kg in the large cyclone, 1.3 kg in the small cyclone, 0.6 kg in
the virtual impactor, and 0.8 kg in the filter. A new system,
designed to sample at a flowrate of 33 m3/ndn is now under de-
velopment. 207
171
-------
4.5 KW
ELECTRICAL
HEAT
DISCHARGE
INSULATED ENCLOSURE
FABRIC
FILTER
FOR SMALL
FRACTION
FIRST
CYCLONE
SECOND
CYCLONE
CONTROL
AND
PUMP
FABRIC
FILTER
FOR
LARGER SIZE
FRACTION
FI-OW
CONTROL
JDUST
[HOPPER
STACK
GAS FLOW
3630-266
Figure 64. A system for collecting large volume samples from industrial
process streams. After Me Far/and and Bertch.206
172
-------
SECTION V
CONTROL DEVICE EVALUATION
OBJECTIVES OF CONTROL DEVICE TESTS
Several reasons exist for performing control device evalua-
tions. These reasons may range from a verification of compliance
with emissions requirements, to programs related strictly to
research.
The majority of stationary air pollution sources need some
type of control device to satisfy the national, state, or local
air pollution regulations that limit the allowable emissions. In
order to determine whether the plant is in compliance with these
regulations, tests are performed to measure the amount of air
pollutant emissions from the control device in question. This
is one type of control device evaluation and it is usually the
simplest and least expensive.
Another reason for performing tests on a control device is to
optimize the performance of the installation. These tests might
be requested by the owners of the plant where the control device
is installed, or by the control device manufacturer. Usually
tests of both the inlet and outlet particulate mass concentration
are made resulting in a measure of the particulate collection
efficiency. In some instances the fractional efficiency (effic-
iency as a function of particle size) is desired and measurements
of the particle size distributions of the inlet and outlet dusts
are necessary.
173
-------
If a particular control device is performing poorly due to
poor maintenance, or poor design, etc., then tests might be required
in order to obtain data to be used in designing additional or re-
placement control device units.
To obtain data for purely research purposes is a fourth reason
for performing a control device evaluation. In each test the data
may be used to confirm existing theories of control device opera-
tion or to develop new theories for modelling and predicting control
device performance. Research tests may involve total systems studies
on the source/control device combination. These tests are usually
the most complicated and expensive because of the amount of data
that is desired.
TYPE AND NUMBER OF TESTS REQUIRED
As mentioned in the previous section, the type and number of
tests that are performed during a control device evaluation depend
on the reason for the tests and the amount of funding available.
In most cases the standard compliance test involves a
determination of the particulate mass concentration at the control
device outlet. Depending on the type of control device, some
measurements of the gaseous emissions may also be required. The
minimum number of tests to be performed during a compliance test
is usually set by Federal or State regulations.
In order to study the performance of a control device, measure-
ments of both the inlet and outlet mass concentration are performed.
These data are required for calculations of the particulate collec-
tion efficiency. If the collection efficiency is to be related
to particle size, then particle size measurements must be per-
formed at the inlet and outlet. If the source is stable, fewer tests
will be required than if the plant process is cyclic or variable
174
-------
over an indeterm!nant time period. If a control device appears
to be performing poorly, then other tests might be necessary, depend-
ing on the type of problem encountered. For example, on an elec-
trostatic precipitator installation, a measurement of the dust
resistivity could explain a poor collection efficiency. At a
fabric filter installation, the problem might be torn bags in one
baghouse compartment. This might require a special test strategy
to isolate this compartment. In a scrubber installation, exces-
sive liquid entrainment could cause poor performance and tests
might be required to measure the droplet concentration or size
distribution.
Data that are required for control device design are the
particulate mass concentration; the particle size distribution;
the physical, chemical, and electrical properties of the dust
to be collected; and the effluent gas temperature, pressure, and
composition. A fairly extensive testing program is necessary
in order to obtain these data. Tests should be performed during
all normal process cycles and with all types of expected feed-
stock to insure that the control device will not be designed
undersize.
If testing is to be performed on a control device for research
purposes only, then the tests that are made are dependent on the
information which is desired as well as the amount of funding. As
is true of all experimental type programs, the more data that are
obtained, the more reliable will be the conclusions based on those
data. Usually control device research programs are designed to
gather as much information as practical for the money available.
Generally, research studies concern the particulate mass concen-
trations at the inlet and outlet, the inlet and outlet particle
size distributions, gas analysis, the dust properties, the control
device operation parameters, plant process data, previous control
device maintenance data, and the economics of the particular control
device. Of course, the type of control device and plant will
175
-------
determine the specific tests which are conducted. For example,
measurement of the dust resistivity would not be required at a
baghouse or scrubber. Liquid feed rates, pressures, etc., how-
ever, would be required.
In some instances the type of tests which are conducted depend
on cooperation from the plant personnel. They may or may not be
willing to alter the feedstock or change the settings on the particle
collector controls, for example. Usually these problems are worked
out as the test plan is developed.
In summary, control device testing is not a routine operation
that has had all the problems worked out or specific procedures
developed. Each plant-control device combination is unique and
should be treated as such. Certain problems exist at one instal-
lation which might not be encountered at any other control device
installation. The number of specific tests which should be per-
formed will depend on the type of control device, the stability of
the source, the length of time allowed for testing, and the avail-
able funding. It is usually advisable to perform as many tests
as practical, because later it may be found that some tests must
be disqualified.
GENERAL PROBLEMS AND CONSIDERATIONS
It is rare that a control device evaluation program does not
encounter several problems in performing the required tests. These
problems can cover a wide range of circumstances and affect the
ability to complete the test program successfully. Although it
is impossible to anticipate every contingency, careful planning
can reduce the likelihood of complete failure of the test program.
A discussion of the more commonly encountered problems and situ-
ations is presented below.
176
-------
Plant Location
Plant location will generally not be a problem unless it
is a long distance to a city where acceptable accommodations and
supplies are available, or if the nearest airport is not conven-
ient for shipping equipment or for transportation of personnel.
Also, depending on the time of year, the local weather can force
postponement of testing, unusual working hours, or require the
construction of special shelters for test crew members required
to work out of doors.
Laboratory Space
Usually arrangements can be made to obtain the use of a por-
tion of the chemical laboratory normally found at most industrial
plants. However, the location of this laboratory may not be con-
venient to the sampling site. As part of the pre-test site sur-
vey, a decision should be made as to whether some type of tem-
porary, mobile lab or trailer would be more convenient than the
plant laboratory space.
Sampling Location and Accessibility
In most new plants the requirements of compliance testing
(ports, platforms, power, etc.) have been taken into account in
designing the facility. This is frequently not true of older
plants. At many sites, the equipment must be hand carried or
hoisted to the sampling location. Stack testing can be difficult
without a properly designed platform. Sometimes platforms and
scaffolding must be erected to allow direct access to the sampl-
ing location. It is recommended that a pre-test site survey be
conducted to determine if any platforms or shelters must be built
prior to actual testing. This is also a good time to inspect
the entire plant and establish contact with the plant employee
who will be responsible for liaison with the plant managers.
177
-------
Power Requirements
Depending on the amount of equipment operating at one time,
the accessible power outlets at most sampling locations may or
may not be adequate. In many instances long extension cords are
necessary, and in some cases a transformer is needed to change
the available power to 110 volts. Before testing, the power re-
quirements should be calculated and plant personnel contacted
if it appears that additional power outlets will be required.
Type of Ports
Almost all sampling of flue gases and dusts requires some
type of port. Before the tests begin it is advisable to know
the location, type, number, and size of the ports that are avail-
able (inlet, outlet, stack). For some types of test equipment,
the ports may be too small and require replacement with larger
diameter ports. The number of ports will also determine the flexi-
bility that one has in planning for traverses of the duct to
obtain representative samples. There may also be some difficul-
ties with the type of ports that are installed, whether flanged
or threaded internally or externally. Other problems commonly
encountered with sampling ports are the length of the port nipple,
rusting of port caps onto the nipples, and caking of dust inside
the ports which must be chiseled away before sampling can begin,
etc.
Flue Gas Velocity and Nozzle Sizes
Depending on the location of the sampling ports, the flue
gas velocity can sometimes be very high or very low. Isokinetic
sampling is highly desirable when sampling dusts. Gas velocities
are usually lowest in the transforms immediately upstream or
downstream from control devices. In these instances the nozzle
178
-------
sizes required for isokinetic sampling may be larger than standard
sizes. On the other hand high gas velocities can require imprac-
tically small nozzles, especially when sampling less than 14 liters
per minute. If the concern is with particles smaller than about
five micrometers diameter, errors from non-isokinetic sampling
are less significant.
Duct Size
The duct size will generally determine whether traversing
is feasible. Traversing twenty foot deep ducts is not a simple
matter, expecially if the probes must be heat traced. Special
hoists sometimes must be erected. Small circular ducts usually
cannot be effectively traversed, and, in some cases instruments
that are normally operated in situ must be operated in an oven
with special sampling probes for extracting the samples.
Gas Temperature and Dew Point
Under some circumstances the gas temperature can cause dif-
ficulties. Too high a temperature can cause galling, metal fatigue,
collection substrate problems, and poor vacuum sealing for in
situ sampling equipment. Low gas temperatures can be especially
troublesome when the slightest temperature drop can cause excur-
sions through dew points causing condensation within the probe
or on collection filters. In either case, probes and other sampl-
ing equipment may have to be insulated or heat traced to prevent
premature cooling of the gases. H-jSO,, condensation, or chemical
reaction can mask particulate weight gains on glass fiber col-
lection substrates. Usually heating to 17°C above the gas dew
point is recommended to avoid condensation. At some plants tem-
perature fluctuations can occur as a result of process variations
or excess air in boiler operations. A knowledge of this type
of activity is desirable before testing.
179
-------
Water Droplets and Corrosive Gases
At some types of scrubber or wet precipitator operations,
water droplet mists will exist at the outlet sampling location.
These droplets may be clean water or contain dissolved or sus-
pended solids. At some locations these droplets might be the
object of sampling; however, at most locations the droplets must
be evaporated in order to capture the particulate matter without
clogging or caking the filter mat and to obtain an accurate mea-
sure of the particle size distribution emitted to the atmosphere.
Corrosive gases can also be difficult to handle. S02 or S03 can
be a problem if the flue gas has a high moisture content or if
the gas temperature is near the dew point. Corrosion of equip-
ment can occur, as well as the masking of filter weight gains
by S02 uptake or H^SO,, mists.
At some locations, particle growth, such as that found in
H2SO, mists, can mask the true concentration of-fine particulate
matter. This can only be, alleviated by keeping the gas tempera-
ture in the sampling train sufficiently high or by. dilution with
clean dry air. Reevaporation of H2SO, mists requires very high
temperatures, and this problem can usually more easily be avoided
than corrected.
Volatile Components
In planning an effective sampling protocol, it is necessary
to consider whether or not volatile components make up a signi-
ficant part of the emissions. Smelting processes are a notable
example of sources where much of the mass emissions consist of
compounds that exist in vapor form at flue gas temperatures, but
condense to form solid particles upon cooling in the atmosphere.
For process streams such as these, the nature or quantity of the
sample is dependent on the temperature of the sampling train. it
180
-------
is usually advantageous to design a special train with several
stages kept at progressively lower temperatures, in order to fully
understand the nature of the emissions. Sulfuric acid is a good
example of a volatile pollutant that can present control and
sampling problems.
Process Cycles and Feedstock Variations
In many plants, such as iron and steel mills and smelting
operations, the effluent gas and dust characteristics vary dra-
matically over a single process cycle. If the test objective
is to obtain a good average of the emissions, the sampling time
is quite flexible. However, if the test objective is to isolate
emissions from a particular part or from each part of an average
cycle, the sampling time must be short, or the tests interrupted
periodically and run only during the part of interest. At some
plants the supply of fuel or feedstock can change. Normally a
plant will maintain logs of the important process parameters,
and this information should be obtained and correlated with the
test data. This can avoid costly repetition of test procedures
or invalidation of the test data.
Long and Short Sampling Times
In general, control devices are very efficient particle col-
lectors. High collection efficiencies mean that the inlet and
outlet dust concentrations can differ by factors up to 1000.
Particular problems arise when sampling requirements demand that
minimum (or maximum) amounts of dust be collected in order to
obtain valid results. At a control device inlet, high dust con-
centrations may necessitate undesirably short sampling durations.
Extremely short (less than five minutes) sampling times may not
allow an adequate period of integration over the plant process
cycles unless the cycles are very stable and long. On the other
181
-------
hand, low dust concentrations at control device outlets sometimes
require sampling times of 12 hours or longer hampering the study
of emissions from each part of a process cycle. It is helpful
to use low flow rate sampling devices at control device inlets
and high flow rate devices at outlets in order to obtain reason-
able sampling times.
Planning a Field Test
Table XVII indicates some of the considerations and problems
that must be dealt with in developing a test plan for control
device evaluations. Although this table is designed to serve
as a planning outline, the relative importance of the facets of
the plan, or considerations that are not listed, can only be
established from a good understanding of the plant-control device
system and the objectives of the test.
182
-------
TABLE XVII
PARTICULATE CONTROL DEVICE TESTS
Assure Compliance
with EPA
Objective of Tests Regulation
Tests Required
Control Device Data
Plant Process Data
Technical Considerations
(Decisions/Problems)
Adequate Space,
Condensible Vapors/
Volatile Particles
Mass Concentration/
Traverse Strategy
Aerosol Gas Velocity
Process/Emission
Variations
Select Particle Sizing
Methods
Filter Mass Stability
Sample Preservation
n
o
x
•j
X
p
0
p
c
o
Optimize Performance
of Control Device
n
r
o
X
D-
Y
P
p
o
P
c
Q
o
c
Q
Obtain Design
Data for
Control Device
i
1
1
1
1
p
I
1
x
D'
y
P
P
0
p
p
1 0
i n
Q
p
n
Obtain Data
for Modeling
Studies
i n
1,11
i n
'»u
i n
i n
i n
I,U
Y
i n
pep nni..
Y
x
[)
i n
Y
D"
y
p
P
n
p
p
i n
I,U
i n
Q
p
Systems Studies
Process and
Control Device
i n
I,U
i n
i n
i n
i n
Y
i n
i n*
x
Y
x
x
Q
i n
x
y
p
P
p
P
i n
i n
p
Key 0 Ontlnt
I Inlet
X - Required
D - Decision based on Specific site or test objectives
C - Must be considered
* vs. Particle Diameter
183
-------
SECTION VI
SUMMARY
It is disappointing to everyone involved in aerosol sampling
that more convenient and efficient methods are not available for
making measurements of particle size and concentration. When good
resolution and accuracy are needed, one must rely on manual tech-
niques such as filters for mass and cascade impactors for sizing
measurements. Nevertheless, progress is being made in the develop-
ment of more convenient methods that yield real-time information.
For some applications, such instruments already yield useful infor-
mation. Table XVIII summarizes the current status of particulate
sampling methods.
184
-------
TABLE XVIII
STATUS OF PARTICULATE SAMPLING METHODS FOR PROCESS STREAMS
MASS CONCENTRATION
Filters - C
3-Particle Attenuation
Charge Transfer — CP
Transmissometers - P
P Light Scattering - P, CP
Piezoelectric Microbalances
- R
OPACITY
Transmissometers - C
Nephelometers - CP
PARTICLE SIZE
Cascade Impactors - C, P
Cyclones - P, C
Light Scattering - P
Diffusion Batteries and
Condensation Nuclei Counters - P
Electrical Mobility - P
C - Commercial instruments in everyday use.
CP - Commercial instruments available, these may require
special adaptation or skills.
P - Prototype systems have been used. These require special
adaptation or skills.
R - Established measurement technique, but not applied to
process streams.
185
-------
REFERENCES
1. U.S. Environmental Protection Agency. Standards of Per-
formance for New Stationary Sources. Federal Register,
43(160), 1977. pp. 41776-41782.
2. U.S. Environmental Protection Agency, Standards of Perform-
ance for New Stationary Sources. Federal Register, 42(187),
1976. pp. 42020-42028.
3. American Society for Testing and Materials. Standard Method
for Sampling Stacks for Particulate Matter, Designation
D2928-71, In: Annual Book of ASTM Standards, Philadelphia,
Pennsylvania, 1977. pp. 592-618.
.4. American Society of Mechanical Engineers. Determining Dust
Concentrations in a Gas Stream, Power Test Code 27. New
York, New York, 1957. 25 pp.
i
5. U.S. Environmental Protection Agency. Standards of Perform-
ance for New Stationary Sources. Federal Register, 42(160),
1977. pp. 41755-41758.
6. U.S. Environmental Protection Agency. Standards of Perform-
ance for New Stationary Sources. Federal Register, 41(160),
1977. pp. 41758-41768.
7. Martin, R.M. Construction Details of Isokinetic Source-
Sampling Equipment. APTD-0581, U.S. Environmental Protec-
tion Agency, Research Triangle Park. North Carolina, 1971.
8. Brenchly, D.L., C.D. Turley, and R.F. Yarmac. Use of the
EPA Particulate Trains for Compliance Testing. In: Air
Pollution Control and Industrial Energy Production, Noll,
Davis, and Duncan, eds. Ann Arbor Science Publishers, Inc.,
Ann Arbor, Michigan, 1975. pp. 73-85.
9. Rom, J.J. Maintenance, Calibration, and Operation of Iso-
kinetic Source Sampling Equipment. APTD-0576, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, 1972. 35 pp.
10. Selle, S.H. and G.H. Gronhovd. Some Comparisons of Simul-
taneous Stack Gas Particulate Determinations Using the ASME
and EPA Methods. J. Eng. Power, 95(3):145-149, 1973.
186
-------
11. Heinsohn, R.J., J.G. Wehrman, J.W. Davis, G.W. Anderson.
A Comparison of the Particulate Matter Obtained Using a
Dilution Sampling System and a Method 5 Sampling System.
In: Proceedings of the 70th Annual Meeting of the Air
Pollution Control Association, Toronto, Ontario, Canada,
1977. Paper 77-12.1.
12. Peters, E.T.., P.L. Levins, A.W. Doyle, K.N. Werner, and
C.H. Summers. Interim Report: Evaluation of the EPA Par-
ticulate Measurement Method as applied to Oil-Fired Utility
Boilers. Arthur D. Little, Inc., Cambridge, Masschusetts,
1973. 32 pp. Prepared'under Contract No.68-02-0632 for
the U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, 1973.
13. Felix, L.G., G.I. Clinard, G.E. Lacey, and J.D. McCain.
Inertial Cascade Impactor Substrate Media for Flue Gas
Sampling. EPA-600/7-77-060, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1977. 89
pp.
14. Forrest, J. and L. Newman. Sampling and Analysis of Atmos-
pheric Sulfur Compounds for Isotope Ratio Studies. Atmos-
pheric Environment, 7(5):5l6-573, 1973.
15. Patankar, U. and W.T. Ott. An Experimental Method of Par-
ticulate Sampling of High Humidity Flue Gases. In: Pro-
ceedings of 70th Annual Meeting, Air Pollution Control Asso-
ciation, Toronto, Ontario, Canada, 1977. Paper 77-49.3.
16. Elder, J.C., M.I. Tillery, and H.J. Ettinger. Evaluation
of EPA Method 5 Probe Deposition and Filter Media Effici-
ency. In: Proceedings of 70th Annual Meeting, Air Pol-
lution Control Association, Toronto, Ontario, Canada, 1977.
Paper 77-12.3.
17. Hamil, H.F., D.E. Camann, and R.E. Thomas. The Collabora-
tive Study of EPA Methods 5, 6, and 7 in Fossil Fuel Fired
Stream Generators - Final Report. EPA-650/4-74-013. U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, 1974. 39 pp.
18. Hamil, H.F., and D.E. Camann. Collaborative Study of Method
for the Determination of Particulate Matter Emissions from
Stationary Sources (Portland Cement Plants). EPA-650/4-
74-029. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1974. 45 pp.
187
-------
19. Hamil, H.F., and R.E. Thomas. Collaborative Study of Method
for the Determination of Particulate Matter Emissions from
Stationary Sources (Municipal Incinerators). EPA-650/4-
74-022. U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1974. 37 pp.
20. Hamil, H.F., and R.E. Thomas. Collaborative Studies of
Particulate Emissions Measurements by EPA Methods, 2, 3,
and 5 Using Paired Particulate Sampling Trains (Municipal
Incinerator). EPA-600/4-76-014, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina,
1976. 91 pp.
21. Lytle, J.H. ASME Method for Measurement of Particulate
Emissions. In: Proceedings of the Workshop on Sampling,
Analysis, and Monitoring of Stack Emissions. Electric Power
Research Institute, Dallas, Texas, 1975. pp. 203-220.
22. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Comparisons of Source Particulate Emis-
sions Measurement Methods for Combination Fuel-Fired Boilers,
Atmospheric Quality Improvement Technical Bulletin No. 75.
New York, N.Y. 38 pp.
23. Boubel, R.W. A High Volume Stack Sampler. Journal of the
Air Pollution Control Association, 21(12):783-787, 1971.
24. Boubel, R.W., and J.M. Morford. Simultaneous EPA Method
5 and High Volume Sampling of Wood Fired Boilers. Presented
at the 70th Annual Meeting of the Air Pollution Control
Association, Toronto, Ontario, Canada, 1977. Paper 12-2.
25. Beckwith, G.S. Particulate Sampling with a Compact Inter-
mediate Volume Sampler. Preprint, Air Pollution Control
Association, Pacific Northwest International Section, Boise,
Idaho, 1974, 35 pp. (Presented at the 67th Annual Meeting
of the Air Pollution Control Association, Pacific Northwest
International Section, Boise, Idaho, 1974).
26. Felix, L.G., G.I. Clinard, G.E. Lacey, and J.D.McCain.
Inertial Cascade Impactor Substrate Media for Flue Gas
Sampling. EPA-600/7-77-060. U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, 1977.
89 pp.
27. Appel, B.R., and J.J. Wesolowski. Selection of Filter Media
for Particulate Sampling with a Lundgren Impactor. AIHL
Report No. 125, California Department of Public Health,
Berkley, California, 1972. 22 pp.
188
-------
28. Elder, J.C., M.I. Tillery, and H.J. Ettinger. Evaluation
of EPA Method 5 Probe Deposition and Filter Media Effici-
ency. In: Proceedings of the 70th Annual Meeting, Air
Pollution Control Association, Toronto, Ontario, Canada,
1977. Paper 77-12.3.
29. Mueller, P.K. Selection of Filter Media: An Annotated
Outline. Presented: 13th Conference on Methods in Air
Pollution and Industrial Hygiene Studies, University of
California, Berkley, California, 1972.
30. Lundgren, D.A. , and T.C. Gunderson. Efficiency and Load-
ing Characteristics of EPA's High-Temperature Quartz Fiber
Filter Media. Amer. Ind. Hyg. Assoc. J. 36(12):866-871,
1975.
31. Lundgren, D.A., and .T.C. Gunderson. Filtration Character-
istics of Glass Fiber Filter Media at Elevated Temperatures.
EPA-600/2-76-192. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, 1976. 95 pp.
32. Stafford, R.G., and H.J. Ettinger. Filter Efficiency as
a Function of Particle Size and Velocity. Atmospheric
Environment, 6:353-362, 1972.
33. Stafford, R.G., and H.J. Ettinger. Efficiency of IPA-1478
Filter Paper Against Polystyrene Latex and Dioctyl Phthalate
Aerosols, Amer. Ind. Hyg. Assoc. J., 32 (5):493-498, 1972.
34. Stafford, R.G., and H.J. Ettinger. Comparison of Filter
Media Against Liquid and Solid Aerosols. Amer. Ind. Hyg.
Assoc. J., 32(5):319-326, 1971.
35. Horn, W. Process for Continuous Gravimetric Determination
of the Concentration of Dustlike Emissions. Staub-Reinhalt,
Luft 28(9): 20-24, 1968.
36. Dresia, H., and F. Spohr. Experience with the Radiometric
Dust Measuring Unit Beta Staubmeter. Staub-Reinhalt, Luft,
31(6):19-27, 1971.
37. Lilienfeld, P., and J. Dulchinos, Vehicle Particulate Ex-
haust Mass Monitor. GCA-TR-72-11-G, GCA Corporation, Bed-
ford, Massachusetts, 1972. 41 pp. Prepared under Contract
No. 68-02-0209 for U.S. Environmental Protection Agency,
Durham, North Carolina, 1972.
38. Sem, G.J., and J.A. Borgos. State of the Art: 1971 In-
strumentation for Measurement of Particulate Emissions from
Combustion Sources Volume IV: Experiments and Final Report.
EPA-650/2-73-022, U.S. Environmental Protection Agency,
Washington, D.C., 1973. 55 pp.
189
-------
39. Nader, J.S. and D.R. Allen. A Mass Loading and Radio-
J? ^ % ^YZernf°r AtmosPheric Particulate. Presented
A™ < I Annual Meeting of the Air Pollution Control
Association, Los Angeles, California, 1959.
4°' ina^n^r" ^V' n^*' Registering, Radiometric Measur
o?9Dust £5eS * C°mbined Measurement of the Emissions
3 1Vy 1" Air> Staub-Reinhalt, Luft
J -
03 lo l i > -
.103-106, 1974. [English Translation] .
icnl';;'-an< I- Ch°- Devel°Pnient of Nucleonic Par-
ticulate Emission Gauge. EPA-APTD-1150, U.S. Environmental
ml? «npp?enCy' Research Triangle Park, NortS Ca?S?ina?
AniPrir^n Tn/iP"i. B^a-Absorption Impactor Aerosol Mass Monitor,
729^1970 Hygiene Association Journal, 31(6):722-
43. Jenarie, M., and P. Bodin. Mise au point et essais d'utili-
sation d un appareil continu pour la mesure de poussieres
155 31967P n ^ ray°ns betu' Rev- Pollution Atm 35:147-
ProtoveA M , and W. Romanek.
Prototype Fly Ash Monitor for Municipal Incinerator Stacks.
ence JSS? ed^ngs.of «* 1970 National Incinerator Confer-
ence, ASME, Cincinnati, Ohio, 1970. pp. 182-188.
45. Dresia, H., P. Tischotter, and G. Felden. Kontinuierliches
46. Lilienfeld, Pedro. Design and Operation of Dust Measuring
ut '
an
Measurement. Staub Reinhalt, Luft, 35 (1) :5-9, 1975:
Ki-K- - Jander« Messtand fur die
and Kalibrienng von Registrierenden Staub-und
inTTeinenr Steinkohlengefeuerten. Kraftwerk.
Vol.
49. Jackson, M.R. Evaluation of Field Prototype Beta Gauges
°aCt #68-°2-1216 ^ ^^E
190
-------
50. Giever, P.M. Development of Automatic Monitors; Light
Transmission and Beta Gaging. In: Proceedings of the
Specialty Conference on Continuous Monitoring of Stationary
Air Pollution Sources, Air Pollution Control Association,
Greater St. Louis Section, St. Louis, Missouri, 1975. pp.
51. Ranade, M.B. Sampling Interface for Quantitative Transport
of Aerosols. EPA-750/2-74-016, U.S. Environmental Protec-
tion Agency Research Triangle Park, North Carolina, 1973.
142 pp.
52. Sauerbrey, G. Verwendung von Schwingguarzen zur Wagung
dunner Schichten und Zur Milrowagung. Zeitschrift fur
Physik Vol 155:206-222, 1959.
53. .Sem, G.J., J.A. Borgos, J.G. Olin, J.P. Pilney, B.Y.H. Liu,
N. Barsic, K.T. Whitby, and F.D. Dorman. State of the Art,
1971 Instrumentation for Measurement of Particulate Emis-
sions from Combustion Sources. Vol. II: Particulate Mass-
Detail Report. APTD-0734, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1971. 223
PP-
54. Sem, G.J., K. Tsurubayashi, and K. Homma. Performance of
the Piezoelectric Mierobalance Respirable Aerosol Sensor.
Amer. Ind. Hyg. Assoc. J., 38:580-588, 1977.
55. King, W.H., Jr. The State of the Art in Piezoelectric
Sensors. In: Proceedings of the 25th Annual Symposium
on Frequency Control, U.S. Army Electronics Command, Ft.
Monmouth, New Jersey, 1971. 5 pp.
56. Daley, P.S., and D.A. Lundgren. The Performance of Piezo-
electric Crystal Sensors Used to Determine Aerosol Mass
Concentrations. American Industrial Hygiene Association
Journal 36 (7) : 518-532, 1975.
57. Lundgren, D.A. , L.D. Carter, and P.S. Daley. Aerosol Mass
Measurement Using Piezoelectric Crystal Sensors. In: Fine
Particles-Aerosol Generation, Measurement, Sampling, and
Analysis, B.Y.H. Liu, ed.f Academic Press, New York, New
York, 1976. pp. 485-519.
58. Harper, W.R. Contact and Frictional Electrification.
Oxford University Press, London, England, 1967. 369 pp.
59. Loeb, L.B. Static Electrification. Springer-Verlag, Ber-
lin, Germany, 1958. 240 pp.
191
-------
60. Schutz, A. Eine Anordnung zur registr ierenden Kontaktelek
Trischen Staubmessung. Staub-Reinhalt, Luft, 24(9) :359-
•36 3 i 1964 .
61
62
Prochazka, R. Neneste Entwicklung des auf Kontaktelektris-
cher Basis beruhenden Staubgehaltsme-Bgerates Koni Test
Staub-Reinhalt, Luft, 24 (9) :353-359, 1964.
Prochazka, R Recording Dust Measurement with the Konitest
Staub-Reinhalt, Luft, 26(5):22-28, 1966. *°nitest,
^S' IMn°ya' Research and Development in Japan on Fine
ticle Measurement and New Control Devices. In- Fine
??ne'ra"on' Measurement, Sampling, and
.New *>rk, New
64' ^Sl6r/ r'Y?" and B'S- Fishkin- TyPe PK-4 Automatic Im-
/Sn0 act Electric Dust Measurement Instrument.
^ Z ?*mW°fking Group on Stationary Source Air Pollu
Control Technology Project A-4. APTIC No. 75270.
65. John, w. Investigation of Particulate Matter Monitoring
Using Contact Electrification. EPA-650/2-75-043, U.S.
66. Gruber A.H., and E.K. Bastress. Application of the Tri-
P^n^CJf5^° £lnfa*U*ement °f Airborne Particles.
Joint Conference on Sensing of Environ-
rgf
the Specialty Conference on Continuous Monitoring of Sta-
tiSSarLeaJ^iUtri0n-S°2rCe?' Air Pollution Con?rol Associa-
1975.' pp? 135-154 Section, St. Louis, Missouri, March,
68. Knapp, K.T. New Techniques for Continuous Measurement of
Mass Emissions. In: Proceedings of the Workshop on Sampl-
ing, Analysis and Monitoring of Stack Emissions. Electric
Power Research Institute, Dallas, Texas, 1975. pp. 49-78.
69. IKOR, inc. Results of Parallel Testing with an IKOR Air
192
-------
70. Eastman, w.B. Use of an IKOR Model 207 Air Quanity Monitor
to Measure Particulate Emissions from Dry Scrubber Systems
Presented at the Meeting of the Metallurgical Society of
American Institute of Mining, Metallurgical, and Petroleum
Engineers, New York, N.Y. 1975. Paper A75-34.
71. Conner, W.D. Measurement of the Opacity and Mass Concentra-
™Q?io Particulate- Emissions by Transmissometry. EPA-650/2-
74-128, U.S. Environmental Protection Agency, Research Tri-
angle Park, North Carolina, 1974. 39 pp.
72. Uthe, E.E., and C.E. Lapple. Study of Laser Backscatter
by Particulates in Stack Emissions. EPA-R2-089, U.S. En-
vironmental Protection Agency, Research Triangle Park, North
Carolina, 1972. 518 pp.
73. Schneider, W.A. Opacity Monitoring of Stack Emissions:
A Design Tool with Promising Results. In: The 1974 Elec-
tric Utility-Generation Planbook, McGraw-Hill, New York,
New York, 1974. pp. 74-76.
74. Duwel, L. Latest State of Development of Control Instru-
ments for the Continuous Monitoring of Dust Emissions
Staub-Reinhalt, Luf t, 28 (3) :42-53, 1968.
75. Buhne, K.-W., and L. Duwel. Recording Dust Emission Mea-
surements in the Cement Industry with the RM4 Smoke Density
Meter Made by Messrs Sick, Staub-Reinhalt, Luft, 32(8) :19-
^•6 , 19 72.
76. Larssen, S. , D.S. Ensor, and M.J. Pilat. Relationship of
Plume Opacity to the Properties of Particulates Emitted
from Kraft Recovery Furnaces. Tappi, 55(l):88-92, 1972.
77. Cristello, J.C. Evaluation of Lear Siegler Transmissom-
Services' Camas> Washington,
78. Reisman, E., w.D. Gerber, and N.D. Potter. in Stack Trans-
missometer Measurement of Particulate Opacity and Mass Con-
centration. EPA-650/2-74-120, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1974 105
pp.
79. Nader, J.S. Source Monitoring. In: Air Pollution, 3rd
Ed., Vol III, Measuring, Monitoring, and Surveillance of
Air Pollution, A.C. Stern, ed. Academic Press, New York
New York, 1976. pp. 589-645.
80. Dobbins, R.A., and G.S. Jizmagian. Particle Size Measure-
ments^Based on Use of Mean Scattering Cross Sections. J
Opt. Soc. Am., 56(10) :1351-1354, 1966.
193
-------
81. Smith, w.B., and J.D. McCain. Particle Size Measurement
in Industrial Flue Gases. In: Air Pollution Control -
Vol. Ill, Werner Strauss, ed. John Wiley and Sons, New
York, New York (in press).
82. Kerker, M. The Scattering of Light and Other Electromag-
n 5™!?i?tl0n* Academic Press, New York, New York, 1969
83. Quenzel, H. Influence of Refractive Index on the Accuracy
of Size Determination of Aerosol Particles with Light-Scat
tering Aerosol Counters. Appl. Opt. 8 (1) :165-169, 1969.
84. Hodkinson, J.R., and J.R. Greenfield. Response Calcula-
counters and
85. Ensor, D.S., and L.D. Bevan. Application of Nephelometry
^ the Monitoring of Air Pollution Sources. Presented at
the 1977 Annual Meeting of the Air Pollution Control Assoc-
iation, Pacific Northwest International Section, Seattle,
Washington, 1973. Paper 73-AP-14.
86. Ensor, D.S. Plume Opacity Measurements. In: Proceedings
of the Symposium on Control of Fine-Particulate Emissions
from industrial Sources, Particulate Technical Sub-Group
or tne U.S. - U.S.S.R. Working Group on Stationary Source
1974 pp 641-672r01 Technology' San Francisco, California,
87. Ensor, D.S. , L.D. Bevan, and G. Markowski. Application
of Nephelometry to the Monitoring of Air Pollution Sources.
In: Proceedings of the Sixty-Seventh Annual Meeting, Air
Pollution Control Association, Denver, Colorado, 1974.
Paper 74-110.
vBcS- Jackson' S' Calvert, C. Lake, D.V. Wallon,
K.S. Campbell, T.A. Cahill, and R.G. Flucchini.
Evaluation of a Particulate Scrubber on a Coal-Fired Utility
Boiler. EPA-600/2-75-074, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1975. 200
pp.
89. Ensor, D.S., and M.J. Pilat. Calculation of Smoke Plume
Opacity from Particulate Air Pollutant Properties. J. Air
Poll. Control Assoc., 21 (8) : 496-501, 1971.
90. Shofner, P.M., G. Kreikebaum, and H.W. Schmitt. In-Situ
Continuous Measurement of Particulate Mass Concentration.
Presented at the 68th Annual Meeting and Exhibition of the
Poliutlon Control Association, Boston, Massachusetts,
Paper 75-41.1.
194
-------
91. Schmitt, H.W., R.J. Nuspliger, and G. Kreikebaum. Contin-
uous In-Situ Particulate Mass Concentration Measurement
of Industrial Discharges. Presented at the 70th Annual
Meeting of the Air Pollution Control Association, Toronto,
Ontario, Canada, 1977. Paper 77-27.4.
92. Environmental Systems Corporation. Sales Literature for
the Particulate Monitor Model P-5A, Knoxville, Tennessee.
93. Schmitt, H.W., G. Kreikebaum, and P.M. Shofner. New De-
velopments in Electro-Optical Monitoring of Particulates.
Environmental Systems Corporation, Knoxville, Tennessee,
1976. 33 pp. Presented at: Industrial Pollution Control
Instrumentation Conference, Newark, New Jersey, 1976.
94. Hood, K.T., and T. Briody. Evaluation of the Performance
and Applicability of a Laser Light Backscatter Measurement
Instrument to the Monitoring of Particulate Concentration
Prior to and Exiting a Kraft Recovery, Furnace Electrostatic
Precipitator. Special Report. National Council of the
Paper Industry for Air and Stream Improvement, New York,
New York,1977. 24 pp.
95. Wostradowski, R.A. Continuous Particulate Monitoring of
Hog-Fuel Fired Power Boilers and Kraft Recovery Boilers.
In: Preprints of the 1977 Environmental Improvement Con-
ference, Canadian Pulp & Paper Association, Moncton, New
Brunswick, Canada, 1977. pp. 87-94.
96. Tipton, D.F. A Particle Analyzer for Stack Emissions.
Powder Tech. 14:245-252. 1976.
97. Sem, G.J., J.A. Borgo, J.G. Olin, J.P. Pilney, B.Y.H. Liu,
N. Barsic, K.T. Whitby, and F.D. Dorman. State of the Art:
1971 Instrumentation for Measurement of Particulate Emis-
sions from Combustion Sources. Vol I: Particulate Mass-Sum-
mary Report. APTD-0733, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1971. 194
pp.
98. Altpeter, L.L., Jr., T.S. Hermann, J.P. Pilney, A.J. Senechal,
and D.L. Overland. Performance of Flame lonization Detec-
tor as Atmospheric Aerosol Monitor. I. Conventional Model.
Environ. Sci. Tech., 10(10):997-1002, 1976.
99. Goldschmidt, V.W. Measurement of Aerosol Concentrations
with a Hot Wire Anemoneter. J. of Colloid Sci., 20:617-
634, 1975.
195
-------
100. Hodkinson, J.R. The Optical Measurement of Aerosols. In-
Aerosol Science, C.N. Davis, ed. Academic Press, New York,
New York, 1966. Chapter 10, pp. 287-357.
101. U.S. Environmental Protection Agency. Appendix B - Perform-
ance Specifications, Performance Specification 1 - Perform-
ance Specifications and Specification Test Procedures for
Transmissometer Systems for Continuous Measurement of the
S?a?otY °f Stack Effluents- Federal Register, 39(177):32860-
32862, 1974.
102. Beutner, H.P. Measurement of Opacity and Particulate Emis-
sions with an On-Stack Transmissometer. J. Air Pollution
Control Association, 24(9):865-871, 1974.
103. Haville, D. A Single-Pass Photoelectric Opacity Measure-
ment System. In: Proceedings of the Specialty Conference
on Continuous Monitoring of Stationary Air Pollution Sources,
Air Pollution Control Association, St. Louis, Missouri,
1975. pp. 154-170.
104. Pfister, E. An Integration Instrument for Timewise Evalua-
tion of Emissions. Staub-Reinhalt, Luft, 34(2):53-56, 1974.
105. Nader, J.S., F. Jaye, and W. Conner. Performance Specifica-
tions for Stationary-Source Monitoring Systems for Gases
and Visible Emissions. EPA-650/2-74-013, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
1974. 74 pp.
106. Avetta, E.D. In Stack Transmissometer Evaluation and Ap-
?c1nntl°n t0 Particulate Opacity Measurement. EPA-650/2-
/5-008, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1975. 124 pp.
107. Peterson, C.M. In-Stack Transmissometer Techniques for
Measuring Opacities of Particulate Emissions from Stationary
Sources. EPA-R2-72-099, U.S., Environmental Protection
Agency, Research Triangle Park, North Carolina, 1972. 88
pp.
108. Crocker, B.B. Monitoring Effluent Particulate Emissions
by Stack Opacity. Preprint, American Institute of Chemical
Engineers, New York, New York, 1974. 15 pp. Presented
at the 78th National Meeting of the American Institute of
Chemical Engineers, Salt Lake City, Utah, 1974. Paper 46d.
109. Conner, W.D., and J.R. Hodkinson. Optical Properties and
Visual Effects of Smoke-Stack Plumes. EPA-AP-30. U.S.
Environmental Protection Agency, Washington, B.C., 1972.
196
-------
110. Ensor, D.S., and M.J. Pilat. The Effect of Particle Size
Distribution on Light Transmittance Measurement. American
Industrial Hygiene Association Journal, 32(5):287-292, 1971.
111. Cohen, J.J., and D.N. Montan. Theoretical Considerations,
Design, and Evaluation of a Cascade Impactor. Amer. Ind.
Hyg. Assoc. J., 1976. pp. 95-104.
112. Marple, V.A., and K. Willeke. Impactor Design, Atmos.
Environ., 10:891-896, 1976.
113. Mercer, T.T. On the Calibration of Cascade Impactors.
Ann. Occup. Hyg., 6:1-17, 1963.
114. Newton, G.J., O.G. Raabe, and B.V. Mokler. Cascade Impac-
tor Design and Performance. J. Aerosol Sci., 8:339-347,
1977.
115. Marple, V.A., and B.Y.H. Liu. Characteristics of Laminar
Jet Impactors. Environ. Sci. & Tech., 8 (7):648-654, 1974.
116. Rao, A.K., and K.T. Whitby. Nonideal Collection Charac-
teristics of Single Stage and Cascade Impactors. Amer.
Ind. Hyg. Assoc. J., 38:174-179, 1977.
117. Gushing, K.M., G.E. Lacey, J.D. McCain, and W.B. Smith.
Particulate Sizing Techniques for Control Device Evaluation:
Cascade Impactor Calibrations. EPA-600/2-76-280, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, 1976. 94 pp.
118. Lundgren, D.A. An Aerosol Sampler for Determination of
Particle Concentration as a Function of Size and Time.
J. Air Pollut. Contr. Assoc., 17(4):225-259, 1967.
119. Ranz, W.E., and J.B. Wong. Impaction of Dust and Smoke
Particles. Ind. Eng. Chem., 44(6):1371-1381, 1952.
120. Davies, C.N., and M. Aylward. The Trajectories of Heavy,
Solid Particles in a Two-Dimensional Jet of Ideal Fluid
Impinging Normally Upon a Plate. Proc. Phys. Soc., 64:889-
991, 1951.
121. Marple, V.A. A Fundamental Study of Inertial Impactors.
University Microfilms, Ann Arbor, Michigan, 1970. 243 pp.
122. Mercer, T.T., and R.G. Stafford. Impaction from Round Jets.
Ann. Occup. Hyg., 12:41-48, 1969.
197
-------
123. Smith, W.B., and J.D. McCain. Particle Size Measurement
in Industrial Flue Gases. In: Air Pollution Control, Part
III, Werner Strauss, ed. John Wiley & Sons, New York, New
York. (in press)
Calvert, S., C. Lake, and R. Parker. Cascade Impactor Cali-
bration Guidelines. EPA-600/2-76-118. U.S. Environmental
no?* i°n A9encv' Research Triangle Park, North Carolina,
J. y / b . 4 J pp .
124
125. McCain, J.D., K.M. Gushing, and A.N. Bird, Jr. Field Mea-
surements of Particle Size Distribution with Inertial Sizing
Devices. EPA-650/2-73-035. U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1973. 52
PP «
126. Brink, J.S Jr., E.D. Kennedy, and H.S. Yu. Particle Size
Measurements with Cascade Impactor s. In: Proceedings of
65th Annual Meeting, AIChE, New York, New York, 19721
n- Felix' and J- Johnson. Cascade Impactor
Data Reduction System: Procedures Manual. EPA Contract
Number 68-02-2131. U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, 1978. (in press)
128-
129. Lippmann M and T.L. Chan. Calibration of Dual-Inlet
sampiing-
of Air 4' ? M. Lippmann. Particle Collection Efficiencies
of Air Sampling Cyclones: An Empirical Theory. Environ.
Sci. Techno!., 11 (4) : 377-382 , 1977.
*'*' Wilson' Jr - Development and Labor a-
78 « TT o °f a Flve~stage Cyclone System. EPA-600/7
TrT i u;s-,Environniental Protection Agency, Research
Triangle Park, North Carolina, 1978.
132. Rusanov, A. A. Determination of the Basic Properties of
133. Smith, W.B., K.M. Gushing, G.E. Lacey, and J.D. McCain.
P?A ^n/^^81211"9 Techniques for Control Device Evaluation.
£PA-650/2-74-102A, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, 1975. 132 pp.
198
-------
134. Gooding, C.H. Wind Tunnel Evaluation of Particle Sizing
Instruments. EPA-600/2-76-073, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, 1976.
72 pp.
135. Hamersma, J.W., S. Reynolds, and R.F. Maddalone. Procedures
Manual for Level 1 Environmental Assessment. EPA-600/2-
76-160A, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1976. 220 pp.
136. Willeke, K., and B.Y.H. Liu. Single Particle Optical Coun-
ter: Principle and Application. In: Fine Particles, Aero-
sol Generation, Measurement, Sampling, and Analysis. Aca-
demic Press, B.Y.H. Liu, ed., 1976. pp. 698-725.
137. Smith, W.B., K.M. Gushing, and J.D. McCain. Procedures
Manual for Electrostatic Precipitator Evaluation. EPA-600/7-
77-059, U.S. Environmental Protection, Research Triangle
Park, North Carolina, 1977. 430 pp.
138. Marple, V.A. The Aerodynamic Size Calibration of Optical
Particle Counters by Inertial Impactors. Particle Tech.
Lab. Pub. 306, presented at Aerosol Measurement Workshop,
University of Florida, Gainesville, Florida, 1976. 13 pp.
139. McCain, J.D., K. Gushing, and W. Smith. Methods for De-
termining Particulate Mass and Size Properties: Laboratory
and Field Measurements. J. Air Pollut. Contr. Assoc.,
24(12):1172-1176, 1974.
140. Junge, C.E., and E. McLaren. Relationship of Cloud Nuclei
Spectra to Aerosol Size Distribution and Composition. J.
of Atmos. Sci., 28(3):382-390, 1971.
141. Haberl, J.B., and S.J. Fusco. Condensation Nuclei Coun-
ters: Theory and Principles of Operation. Prepared for
presentation at the llth Conference on Methods in Air Pol-
lution and Industrial Hygiene Studies at the University
of California, Berkeley, California, sponsored by California
Air Resources Board and California Department of Public
Health, 1970. 24 pp.
142. Fuchs, N.A., I.E. Stechkina, and V.I. Starosselskii. On
the Determination of Particle Size Distribution in Poly-
disperse Aerosols by the Diffusion Method. Brit. J. Appl.
Phys., 16:280-281, 1962.
143. Sinclair, D., R.J. Countess, B.Y.H. Liu, and D.Y.H. Pui.
Experimental Verification of Diffusion Battery Theory.
J. Air Pollut. Contr. Assoc., 26(7):661-663, 1976.
199
-------
144.
Sinclair, D. A Portable Diffusion Battery: Its Applica-
tion to Measuring Aerosol Size Characteristics. Amer Ind
Hyg. Assoc. J., 33 (11) : 729-735, 1972.
BatPrv T *' H°°PeS- A N°Vel F°rm °f uson
Battery. Amer. Ind. Hyg. Assoc. J., 36(l):39-42, 1975.
S'F> Gu99enheini' and A.C. George. Compact
147 " iMati°n °f Aer°SO1 Size Distributions
1963 Measurements- J- of Franklin Inst.,
148. Sansone, E.B., and D.A. Weyel. A Note on the Penetration
a anAeroso1 »ith a Log-Normal Size
J. Aerosol Sci., 2:413-415, 1971.
149.
arc n USable Prot°tYPe System for Sizing
ber 68 02 ^il n ***" °'5 m Di^eter. EPA Contract Num-
- Agency' Re-
15°* NSc!ei°C^nf; fdifjcation of a Commercial Condensation
" Sdy F10W' Atmos- ^viron, 10:659-
660, 197
-C- WellS* A Hi9h ^solution Charge and
152.
L. De-
Aeroso1
Proceed' no ;°'f f tended Electric Mobility Method. In:
' Minneapolis,
Impactor/Prp'- a?d.D' ^ ' Devel°P^nt of an In-Stack
ppPao? °r^ re"Pitator for Sizing Submicron Particles. EPRI
forn?a! El6CtriC Power Research Institute, Palo Alto, Cali-
?nCElectrlcaf?v ?A Trijl;t Charging of Submicron Particles
of Y Ionized Fields; Measurement of the Rate
200
-------
156. Whitby, K.T., and W.E. Clark. Electric Aerosol Particle
Counting and Size Distribution Measuring System for the
0.015 to 1 Micron Size Range. Tellus, 18:573-586, 1966.
157. Liu, B.Y.H., K.T. Whitby, and D.Y.H. .Pui. A Portable Elec-
trical Analyzer for Size Distribution Measurement of Sub-
Micron Aerosols. J. Air Pollut. Contr. Assoc., 24(11):1067-
1072, 1974.
158. Sem, G.J. Submicron Particle Sizing Experience on a Smoke
Stack Using the Electrical Aerosol Size Analyzer. EPA-600/2-
77-060, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1975. pp. 276-300.
159. Cadle, R.D. Particle Size Determination. Interscience
Publishers, Inc., New York, New York, 1955. 303 pp.
160. Allen, T. Particle Size Measurement. Chapman and Hall
Ltd., London, England, 1975. 454 pp.
161. Godridge, A.M., S. Badzioch, and P.G.W. Hawksley. A Par-
ticle Size Classifier for Preparing Graded Sub-Sieve Frac-
tions. J. Sci. Instrum. 39:611-613, 1962.
162. Goetz, A., and T. Kallai. Instrumentation for Determining
Size and Mass Distribution of Submicron Aerosols. APCA
J., 12:479-486, 1962.
163. Goetz, A., H.J.R. Stevenson, and 0. Preining. The Design
and Performance of the Aerosol Spectrometer. APCA J.,
10:378-838, 1960.
164. Gerber, H.E. On the Performance of the Goetz Aerosol Spec-
trometer. Atmos. Environ., 5:1009-1031, 1971.
165. Stober, W., and H. Flachsbart. Size-Separating Precipita-
tion of Aerosols in a Spinning Spiral Duct. Environ. Sci.
Technol., 3 (12) :1280-1296, 1969.
166. Swayer, K.F., and W.H. Walton. The "Conifuge" - A Size-
Separating Sampling Device for Airborne Particles. J. Sci.
Instrum., 27:272-276, 1950.
167. Keith, C.N., and J.C. Derrick. Measurement of the Particle
Size Distribution and Concentration of Cigarette Smoke by
the "Conifuge". J. Colloid Sci., 14:340-356, 1960.
168. Tillery, M.I. Design and Calibration of a Modified Coni-
fuge. Assessment of Airborne Radioactivity, IAEA, Vienna,
1967.
201
-------
169. McCrone, W.C., and J.G. Delly. The Particle Atlas, Edition
Two. Ann Arbor Science, Ann Arbor, Michigan, 1973. 319
pp.
170. Kaye, B.H. Symposium on Particle Size Analysis Society
for Analytical Chemistry, Loughborough, England, 1966.
171. Allen-Bradley Sonic Sifter. U.S. Patent 3,045,817.
172. Stern, S.C., H.W. Zeller, and A.I. Schekman. Collection
Efficiency of Jet I.mpactors at Reduced Pressures. Ind.
Eng. Chem. Fundam., 1(4):275-344, 1962.
173. Parker, G.W., and H. Buchholz. Size Classification of Sub-
micron Particles by a Low-Pressure Cascade Impactor. ORNL-
4226, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
1968. 64 pp.
174. McFarland, A.R., H.S. Nye, and-C.H. Erickson. Development
of a Low Pressure Impactor. EPA-650/2-74-014, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, 1973. 56 pp.
175. Pilat, M.J. Submicron Particle Sampling with Cascade Im-
pactor. In: Proceedings of the 66th Annual Meeting of
the Air Pollution Control Association, Chicago, Illinois,
1973. Paper 73-384.
176. Pilat, M.J., G.M. Fioretti, and E.B. Powell. Sizing of
0.02-20 Micron Diameter Particles Emitted from Coal-Fired
Power Boiler with Cascade Impactors. Paper presented APCA-
PNWIS Meeting, Vancouver, B.C., 1975.
177. Wagman, J., and C.M. Peterson. A Continuous Monitor for
Size Distribution of Airborne Particulate Emissions. In:
Proceedings of the 3rd International Clean Air Congress,
Dusseldorf, Germany, October 8-12, 1973. p. C6-8.
178. Lilienfeld, P., D.P. Anderson, and D.W. Cooper. Design,
Development, and Demonstration of a Fine Particulate Mea-
suring Device. EPA-600/2-77-077, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina,
1977.
179. Carpenter, T.E., and D.L. Brenchley. A Piezoelectric Cas-
cade Impactor for Aerosol Monitoring. Amer. Ind. Hyg.
Assoc. J., 503-510, 1972.
180. Chuan, R.L. An Active Cascade Impactor for Real Time Sizing
of Airborne Particles. Celesco Industries, Inc., Costa
Mesa, California, Bulletin AT-149. 3 pp.
202
-------
181. Hounam, R.F., and R.J. Sherwood. The Cascade Centripeter:
A Device for Determining the Concentration and Size Distri-
bution of Aerosols. Ind. Hyg. J., 122-131, 1965.
182. Connor, W.D. An Inertial-Type Particle Separator for Col-
lecting Large Samples. J. Amer. Pollut. Contr. Assoc.,
16(1) :33-38, 1966.
183. Loo, B.W., and J.M. Jaklevic. An Evaluation of the EEC
Virtual Impactor. Lawrence Berkeley Laboratory Report No.
LBL-2468, January, 1974.
184. Loos, B.W., J.M. Jaklevic, and F.S. Goulding. Dichotomous
Virtual Impactors for Large Scale Monitoring of Airborne
Particulate Matter. In: Fine Particles, Aerosol Genera-
tion, Measurement, Sampling, and Analysis. Academic Press,
B.Y.H. Liu, ed.7 1976. pp. 311-350.
185. Schott, J.H., and W.E. Ranz. Jet-Cone Impactors as Aerosol
Particle Separators. Environ. Sci. Tech. 10(13):1250-1256,
1976.
186. Luna, R. A Study of Impinging Axi-Symmetric Jets and Their
Applications. Dissertation, Univ. Microfilms, Ann Arbor,
Michigan, 1965. 117 pp.
187. Brooks, E.F., M.E. Taylor, T.E. Chamberlain, R.J. Golik,
and R. Watson. Development Study of a Novel Continuous
Flow Impactor, TRW Systems and Energy. Final Report on
EPA Contract No. 69-02-2165, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1977. 107
pp.
188. Willeke, K. Size Classification by Opposing Jets. Oral
Presentation at Tenth Aerosol Technology Meeting, Los Alamos,
New Mexico, September, 1977.
189. Hodkinson, J.R. The Optical Measurement of Aerosols. In:
Aerosol Sci., C.N. Davies, ed. Academic Press, New York,
New York, 1966. pp. 287-357.
190. Shofner, P.M., G. Kreikebaum, H.W. Schmitt, and B.E. Barn-
hart. In Situ, Continuous Measurement of Particulate Size
Distribution and Mass Concentration Using Electro-Optical
Instrumentation. In: Proceedings of Fifth Annual Indus-
trial Air Pollution Control Conference, Knoxville, Tennessee,
April, 1975.
191. Gravatt, C.C., Jr. Real Time Measurement of the Size Dis-
tribution of Particulate Matter by a Light Scattering Method.
J. Air Pollut. Contr. Assoc., 23(12):1035-1038, 1973.
203
-------
192. Chan, P.w. Optical Measurements of Smoke Particle Size
Generated by Electric Arcs. EPA-650/2-74-034, U.S. Environ-
mental Protection Agency, Washington, D.C., 1974. 49 pp.
SGatteri^ Aerosol Spectrometer.
Size AnalYzer . Appl. Opt., 11(2)
195. McSweeney, A. A Diffraction Technique to Measure Size Dis-
tribution of Large Airborne Particles. EPA-600/3-76-073,
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, 1976.
196. Wertheimer, A.L., and W.L. Wilcock. Light Scattering Mea-
1616-1620, °rtiCle °istr ibutions' ^P1- Opt - • 15(6):
197. Hotham, G.A. Size of Respirable Aerosols by Pulsing UV
Maryland? W4. ^5°^! Measurement Seminar, Gaithersburg,
198. Simmons, H., and J. Dominic. A High-Speed Spray Analyzer
Tnrh Turbine Fuel Nozzles. Oral Presentation, ASME Gas
Turbine Conference, Session 26, Cleveland, Ohio, March 12,
199. Mathews, B.J and R.F. Kemp. Holography of Light Scat-
Large Steam Boiler' 63rd Annual
Continuous Particulate Monitor-
200. Farmer w.M. Measurement of Particle Size, Number Density,
X£i? a'- Mi" 3nd W'P- Chu' Dif^action Analysis of
Doppler Signal Characteristics for a Cross-Beam Laser Dop-
pler Velocimeter. App. Opt., 14(9):2177-2183, 1975.
202. Adrian, R.j and K.L. Orloff. Laser Anemometer Signals:
Visibility Characteristics and Application to Particle
Sizing. App. Opt., 16(3):677-684, 1977.
203. Roberds, D.W Particle Sizing Using Laser Interferometry.
App. Opt., 16(7):1861-1868, 1977.
204. Wilson, J.c. Aerosol Particle Size Measurement by Laser
M?nn™ yelocimetry. Ph.D. Thesis, University of Minnesota,
Minneapolis, Minnesota, 1977.
204
-------
205. Medecki, H., M. Kaufman, and D.E. Magnus. Design, Develop-
ment and Field Test of a Droplet Measuring Device. EPA-
650/2-75-018, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, 1975. 56 pp.
206. McFarland, A.R., and R.W. Bertch. Fractionator for Size
Classification of Aerosolized Solid Particulate Matter.
Environ. Sci. Tech., 11(8):781-784, 1977.
207. McFarland, A.R. Private communication.
205
-------
GLOSSARY
Aerodynamic diameter, DA: The diameter of a sphere of unit den-
sity which has the same settling velocity in the gas as the
particle of interest. See also Stokes diameter.
Aerodynamic impaction diameter, DAI: The aerodynamic impaction
diameter of a particle is an indication of the way that a
particle behaves in an inertial impactor or in a control
device where inertial impaction is the primary mechanism
for collection. If the particle Stokes diameter, DS, is
known, the aerodynamic injection diameter is equal to:
where p is the particle density, gm/cm3, and
P
C is the slip correction factor.
See also Stokes diameter.
Aerosol: A suspension of solid or liquid particles in a qas.
Air centrifuge: A laboratory device which uses centrifugal force,
created by spinning part of the device, to separate particles
larger than a certain aerodynamic diameter from an aerosol.
Blank: A blank usually refers to a controlled cascade impactor
test run in which the particles are removed by a prefilter.
If the measured impactor stage weights are found to change
significantly and consistently, the actual test runs should
be corrected for this background.
Bounce: Bounce in this document refers to inadequate retention
of particles that strike the impaction surface in cascade
impactors. If the particle does not adhere, it is said to
bounce. See also Re-entrainment.
Cascade impactor: An instrument which uses impaction to separate
particles from an aerosol and deposit them on stages in order
of decreasing aerodynamic diameter.
206
-------
Centrifugal spectrometer: A laboratory device which uses cen-
trifugal force, created by spinning part of the device, to
separate and deposit particles from an aerosol in narrow
bands that represent discrete intervals of aerodynamic diam-
eter .
Condensation: The coalescence of vapors either into liquid par-
ticles in the gas stream or on sampling equipment walls.
Confidence limits: A range of values about an arithmetic mean.
If a large number of sets of samples are taken, the param-
eter being estimated will be within the confidence limits
of a specified percentage of the sets.
Conifuge: A centrifugal spectrometer in which the aerosol flows
between two coaxial rotating cones, the particles being de-
posited on the outer cone.
Coulter counter: A particle sizing instrument which measures
the change in current flow through a small orifice immersed
in an electrolyte as a particle passes through the orifice.
Critical orifice: An orifice through which the air flow rate is
kept constant by maintaining sonic velocity.
Cut-point: The cut-point of an impactor stage or cyclone is the
particle diameter for which all particles of equal or greater
diameter are captured and all particles with smaller diam-
eters are not captured. No real impactor or cyclone actually
has a sharp step function cut-point, but the theoretically
defined D50 of a stage is often called its cut-point.
Cyclone: A device that causes an aerosol to spiral around its
walls, thus separating particles by centrifugation, the large
particles being deposited on the wall and at one end of the
device, and the smaller particles passing out an opening
in the other end of the device.
D50: The D50 of an impactor stage or cyclone is the particle
diameter for which the device is 50 percent efficient. Fifty
percent of the particles of that diameter are captured and
50% are passed to the next stage. The theoretical expres-
sion for the D50 of an impactor stage is
[18 STK y R
CPPV:
207
-------
where Stk = Stokes number, determined by calibration for
50% collection efficiency, dimensionless,
y = gas viscosity, poise,
R = impactor jet radius (for slot impactors, the
slot half width), cm,
V. = gas velocity through impactor jet, cm/sec,
C = Slip correction factor, dimensionless,
p = particle density, g/cm3.
D50(AI), aerodynamic impaction diameter, is found by setting
C and p = 1.0.
DSO(A), the aerodynamic diameter, is found by setting p
= 1.0, and p
D50(S), Stokes diameter, is found by setting p = the actual
particle density. "
There is no accurate expression for the D50 of a cyclone.
Diffusion: The net movement of gas molecules or particles from
a high to a low concentration area due to Browian motion.
Diffusion battery: A device in which a number of small ducts
or cells are arranged in parallel so that significant dif-
fusion losses are possible at relatively large sampling flow
rates.
OOP: Dioctyl phthalate. An organic fluid of low vapor pressure,
frequently used in the generation of aerosols for calibra-
tion and testing of air pollution measuring devices.
Electric particle analyzer: An instrument which sizes and col-
lects particles on the basis of their electrical mobility.
Electrostatic precipitator: A device which collects aerosol par-
ticles electrically by charging them in a unipolar ion field
and then collecting them on a suitable surface by a combina-
tion of turbulent transport and a strong electric field.
Elutriation: The process by which particles in a moving stream
are separated aerodynamically by the pull of gravity.
208
-------
Equivalent polystyrene latex (PSL) diameter: The intensity of
light scattered by a particle at any given angle is dependent
upon the particle size, shape and index of refraction. It
is impractical to measure each of these parameters, however,
and the theory for light scattered by irregularly shaped
particles is not well developed. Sizes based on light scat-
tering by single particles are therefore usually estimated
by comparison of the intensity of scattered light from the
particle with the intensities due to a series of calibration
spheres of precisely known size. Although spinning disc
and vibrating orifice aerosol generators can be used to
generate monodisperse calibration aerosols of different physi-
cal properties, most manufacturers of optical particle sizing
instruments use polystyrene latex spheres to calibrate their
instruments. It is convenient to define an equivalent PSL
diameter as the diameter of a PSL sphere which evokes the
same response from a particular optical instrument as the
particle of interest.
Equivalent volume diameter: Certain instruments, such as the
Coulter Counter, have, as the measured size parameter, the
volumes of the individual particles. Size distributions
from such techniques are given in terms of the diameters
of spheres having the same volume as the particles of in-
terest.
Extinction coefficient: The extinction coefficient of an aerosol
is given by the following equation:
E =
r2 QT?(a,m) N(r) dr
£1
where « = size parameter, 2rir/X
r = particle radius
X = wavelength of the radiation
m = particle refractive index relative to the gas
medium
N(r) = number of size frequency distribution, i.e. the
number of particles of radius r per volume per
Ar.
QF = particle extinction coefficient, defined as the
total light flux scattered and absorbed by a
particle divided by the light flux incident on
the particle.
209
-------
Extractive sampling: Sampling of a particulate laden process
effluent stream by means of a probe inserted inside the pro-
cess stream duct to allow transport of the gas to some type
of sampling instrument located outside the process stream
duct.
Geometric mean diameter: The geometric mean diameter is the diam-
eter of a particle which has the logarithmic mean for the
size distribution. This can be expressed mathematically
as:
log GMD
log D, + log D_ +
= — H - 1 - ± - i
+ log
N
or as
GMD =
Geometric standard deviation: A measure of dispersion in a log-
normal distribution, given by:
N
X) fi (log D-j-log
• -i -J -J
log ag =
N-l
where
a is the geometric standard deviation,
f . is the relative mass, surface area, or number of
•^ particles in the interval,
D. is the diameter characteristic of the j
3 and
interval,
N is the total number of intervals.
Grease: In impactor terminology, grease is a substance that is
placed on an impactor stage or substrate to serve as a par-
ticle adhesive.
Horizontal elutriator: A device in which an aerosol passes in
laminar flow between two parallel horizontal surfaces and
particles are deposited onto the lower surfaces by sedimen-
tation. See also Elutriation.
Impaction: The separation of particles of sufficient inertia
from a flowing aerosol onto a surface as the surface deflects
the aerosol.
Impinger: A device which causes the separation of particles of
sufficient inertia from a flowing aerosol onto a surface
under water as the water deflects the aerosol stream.
210
-------
Index of refraction: The ratio of the velocities of light in
two adjacent media as the light passes from one medium to
the other. Usually one medium is assumed to be air.
Inertial impaction parameter, Y : The inertial impaction parameter
is similar to the Stokes number; however, the characteris-
tic dimension of the system is the diameter or width of the
jet, not the radius or half width. Thus
T = 2 C Vo Pp r2/9yD
where V is the particle velocity, cm/sec,
p is the particle density, g/cm3,
r is the radius of the particle, cm,
C is the slip correction factor,
y is the gas viscosity, poise, and
D is the diameter or width of the jet, cm.
See also Stokes number .
In situ sampling: Placement of a sampling device directly into
a process gas stream in order to sample the particles or
gas directly.
Isokinetic sampling: The method of sampling in which the velocity
of the aerosol flowing into the sampling inlet equals that
of the aerosol flowing past it. See also Stokes number.
Lognormal distribution: A distribution of frequencies which is
symmetric or bell shaped when plotted along a logarithmic
abcissa.
Mean free path of gas molecules: The average distance that mole-
cules travel between collisions. For practical purposes,
the mean free path is given with sufficient accuracy by the
following equation:
X = 2U (8.3 x 107T\
1.01 x 106P 3 MM
where y is the viscosity of the gas, poise,
P is the pressure of the gas, atm,
T is the temperature, °Kelvin, and
MM is the mean molecular weight.
211
-------
Mean of a distribution: The term "mean" is used to denote the
arithmetic mean, or average, of a distribution. In a par-
ticle size distribution the mass mean diameter is the diam-
eter of a particle which has the average mass for the entire
particle distribution.
Median of a distribution: The median divides the area under a
frequency curve in half. For example, the mass median diam-
eter (MMD) of a particle size distribution is the size at
which 50% of the mass consists of particles of larger diam-
eter, and 50% of the mass consists of particles having smaller
diameters.
Mode of a distribution: The mode represents the diameter which
occurs most commonly in a particle size distribution. The
mode is seldom used as a descriptive term in aerosol physics.
Monodisperse aerosol: An aerosol in which all the particles are
the same size.
Normal distribution: A distribution of frequencies which is sym-
metric or bell shaped when plotted along a linear abscissa.
Also called a Gaussian distribution.
Particle mobility: The ratio of the velocity of a particle to
the force causing steady motion is called the mobility, b.
/-*
b =
3-rryD '
where y is the gas viscosity, poise,
D is the particle diameter, cm, and
C is the slip correction factor.
The electric mobility of a particle is given by:
b = 3TryD
where q is the electric charge on the particle.
Particle size distribution: A mathematical relationship express-
ing the relative amount of particles in an incremental range
of sizes.
212
-------
Particle stopping distance, .*: The distance travelled by a par-
ticle as it accelerates from some initial velocity to the
velocity of the gas stream.
8, = TV
where T is the relaxation time (sec), and
V is the velocity of the gas stream (cm/sec).
Polydisperse aerosol (heterodisperse): An aerosol containing
particles of many sizes.
Preconditioning: Unwanted weight changes of impactor glass fiber
collection substrates may be reduced by placing a batch of
substrates inside the duct to be sampled, and pumping or
passing filtered flue gas through them for several hours.
Such a procedure is referred to as "preconditioning the
substrates.
Precutter or precollector: A collection device, often a cyclone,
which is put ahead of the impactor in order to reduce the
first stage loading. This is necessary because in some
streams the high loading of large particulate would overload
the first stage before an acceptable sample had been gathered
on the last stages.
Probe: A pipe used for the transport of process effluent gas
from the interior of the process ducting to a sampling in-
strument. Usually probes are insulated and heat traced and
have some type of nozzle attachment at the end to be inserted
in the gas stream for isokinetic sampling. In the case or,
in situ sampling, the probe is used to connect the sampling
instrument inside the duct to accessory equipment outside
the duct. If there is no accessory equipment, the probe
is used as a handle for inserting, transversing, securing,
and removing the sampling instrument.
Process stream: Any particulate laden gaseous effluent that is
an end product of a manufacturing or energy conversion pro-
cess.
Real-time monitor: Any sampling instrument which processes data
and gives an instantaneous display of information concern-
ing the process stream effluent under consideration.
Rebound: See Bounce.
Re-entrainment: The re-entrance of formerly collected particles
into the aerosol stream.
213
-------
Relaxation time, T: The time required for a particle to accele-
rate from some initial velocity to the velocity of the car-
rier gas.
T =
2V2c
where pp is the partible density, g/cm3,
r is the particle radius, cm,
C is the slip correction factor, and
y is the gas viscosity, poise.
Resistivity: The resistivity of a conductor is defined as the
ratio of the potential gradient across the conductor to the
current per unit cross-sectional area. The resistivity of
dusts is very important in the efficient operation of an
electrostatic precipitator used as a pollution control
device.
Reynolds number: A dimensionless parameter defined as
R =
where p = fluid density
V = fluid velocity
y = fluid viscosity, and
D = some characteristic dimension of the fluid flow
system
Generally, a Reynolds number less than 2000 indicates laminar
tlow, greater than 4000 indicates turbulent flow, and 2000-4000
is a transition region in which the flow can be laminar or tur-
bulent.
Sampling train: The components of a system used to remove or
sample dusts from process streams. A basic particulate mass
sampling train would consist of a nozzle, a probe, a filter,
one or more devices for monitoring gas flow, and a pump.
Pi tot tubes and thermocouples, used to monitor the gas velo-
city and temperature, are generally regarded as part of a
sampling train.
214
-------
Slip correction factor: A correction to Stokes' law made neces-
sary by the existence of a finite net gas velocity at a par-
ticle surface. Stokes1 law can be applied to submicron par-
ticles if a slip correction factor, C, is used.
C = ! .
1.23 + 0.41 exp
-0.44D
where
A is the mean free path of the gas molecules, ym,
and
D is the particle diameter, ym.
The constants in this equation were determined empirically
for air at standard temperature and pressure, and are thus only
approximate for stack conditions. If the exponential term is
neglected, the equation is referred to as the Cunningham correc-
tion factor.
Stage: A stage of an impactor is usually considered to be the
accelerating jet (or plate containing multiple jets) and
the surface on which the accelerated particles impact.
Stokes diameter, Dg: The diameter of a sphere having the same
density and which behaves the same aerodynamically as the
particle of interest. For spherical particles, the Stokes
diameter is equal to the diameter of the particle.
Stokes number, Stk: The ratio of the particle stopping distance
to some characteristic dimension of the sampling system.
For example, if the stopping distance for particles of a
given diameter is much smaller than the radius of a sampling
nozzle, (Stk « 1) the particles will be sampled accurately
in spite of flow disturbances due to the nozzle design or
sampling velocity. If the particle stopping distance is
comparable in magnitude to the nozzle diameter, however,
the particles may cross flow streamlines and either enter
or miss the nozzle in quantities which are not proportional
to the particle concentration in the duct. Thus, for Stk
on the order of 0.1 or greater, isokinetic sampling is re-
quired. (See isokinetic sampling.) In impaction theory,
the characteristic dimension of the system is the radius
or half width of the jet, R. Thus,
Stk =
2p r2CV(
9yR
215
-------
where V0 is the particle velocity, cm/sec,
p is the particle density, g/cm3,
P
r is the radius of the particle, cm,
C is the slip correction factor,
y is the gas viscosity, poise, and
R is the radius or half width of the jet, cm.
If the particle diameter D is substituted, this equation
becomes P
P D / CV
- P P
~
18 UR
See also Inertial impaction parameter.
Substrate: The removable, often disposable, surface on which
impacted particles are collected. Substrates are charac-
teristically light and can be weighed on a microbalance.
Temperature and Pressure Standards: Laboratory standard condi-
tions have generally been recognized for many years as 0°C
and 760 mm Hg. Recently the US EPA has set standard con-
ditions for all stationary source testing to be 20°C and
760 mm Hg. Engineering standards have been defined for some
time and are 70°F and 29.92 in. Hg. In order to avoid con-
fusion, the designation "normal", (N) , is used to denote
engineering standard conditions in metric units (21°C, 760
mm Hg) .
When denoting measures of gas volume, the letter "d",
or "D", (for "dry") is sometimes included to signify that
the volume measured contains no water vapor. In stationary
source testing, the letter "a", or "A", (for "actual") signi-
fies the volume of the gas at the actual stack conditions,
for example, the volume the gas would have at 200°C, 740
mm Hg, and 10% H20.
Examples of stationary source testing nomenclature:
s.d.c.f. (or DSCF or SDCF) —standard dry cubic feet—a gas
volume measured at 20°C, 760 mm Hg, and 0% H20.
a.c.f. (or ACF) — actual cubic feet — a gas volume measured
at conditions other than standard, usually given in the
text.
216
-------
ACM (or Am )—actual cubic meters
ACCM (or Acm3)—actual cubic centimeters
DNCM (or DNm3)—dry normal cubic meters—a gas volume mea-
sured at 21°C, 760 mm Hg, and 0% H20.
Traverse: A systematic sampling from various points inside a
process effluent duct in order to obtain a representative
sample. The number and position of the traverse points is
dependent on the size and shape of the ducting at the sampl-
ing location.
217
-------
BIBLIOGRAPHY
A literature search was made for articles, reports, and books
pertaining to particulate sampling from industrial process streams
with an emphasis on control device evaluation covering, in general,
the past two years. The bibliography was planned to be a supplement
to the list of references, naming some of the most recent publica-
tions and also those "classic" publications which are most often
cited by recent authors. The search included a subject search of
the Engineering Index, 1974-1976; Chemical Abstracts, 1976-1977;
Air Pollution Abstracts, July, 1974 - July 1976; The EPA Publica-
tions Bibliography, January-September, 1977; and other indices to
a lesser extent.
An extensive search was made of the references in the Environ-
mental Engineering Library of Southern Research Institute, which
contained a major portion of the references listed in the bibli-
ography. Publications over three years old were generally not
included unless they contained information that was not found or
not superceded in recent papers. The list of references cited
in Sections II - IV of the manual are not necessarily duplicated
in the bibliography; however, they should be consulted first for
information on particulate measurement.
The formats of the references generally fall into four main
groups:
1. Reports on government contracts: author(s), title, per-
forming organization or company, sponsoring government
agency, address of government agency, year of publication,
218
-------
number of pages, government report number (when appli-
cable or available), and National Technical Information
Service number (when applicable or available).
2. Books: author(s), title, publisher, publisher's address,
year of publication, and number of pages.
3. Journal articles: author(s), title, name of journal,
volume number, issue number (if applicable), page
numbers, and year of publication.
4. Papers and proceedings of technical meetings: author(s),
title, name, location, and year of meeting, page numbers
or paper number (when applicable).
Contents of the bibliography are arranged alphabetically by
author under the following headings:
1. General References
2. Sample Extraction
3. Filter Media
4. Mass Concentration
5. Particle Size Distribution
6. Opacity - Transmissometers - Nephelometers
7. Analytical Technique
8. Control Device Evaluation - Field Tests.
219
-------
1, GENERAL REFERENCES
AMERICAN SOCIETY FOR TESTING * MATERIALS
ANNUAL BOOK OF ASTM STANDARDS* STANDARD FOR METRIC PRACTICE
ASTM, PHILADELPHIA, PA,, PP.
-------
CAt,V£f?T, S.f J. GOLOSCHMjn, n. LEITH, AND D. M£HTA
WET SCRUBBE« SYSTEM STUDY, VOL. I, SCRUBBER HANDBOOK
A.P.T., INC.* EPA, RESEARCH TRIANGLE PARK, N.C., 1972, 82* PP
EPA.R2-72»nflA PR 213 OJ6
CALVFRT, S.. J. GOLDSHMJD, 0, LOTH, AND 0. MEHTA
WET SCRUBBER SYSTEM STUDY VOLUME II FINAL REPORT AND BIBLIOGRAPHY
A.P.T., INC., EPA, RESEARCH TRIANGLE PARK. N.C.. 1975, Iflt PP
EPA.R2.72-.lt8B PP 21* 017
CARVER, L. D.
PARTICLE SIZE ANALYSIS
INDUSTRIAL RESEARCH, PP*. 40-13, 1971
CASSATT, W. A., AND R'. S, HADDOCK, EDITORS
AEROSOL MEASUREMENTS
SEMINAR ON AEROSOL MEASUREMENTS, NATIONAL BUREAU OF
STANDARDS, WASHINGTON, 0, C., 197
-------
N. A.
THF. MECHANICS OF AtRQSOLS
THE MACMILLAN' CO,. N£W YORK, 1964. 408 PP.
GCA CORP
APPENDICES TO HANDBOOK OF FABRIC FILTER TECHNOLOGY, VOL, II
GCA CORP. FOR NAPC ADMIN, U.S. DEPT'. HE*, 1970, 208 PP.
PB 200 649
GCA CORP
BIBLIOGRAPHY. VOL. Ill, FABRIC FILTER SYSTEMS STUDY
GCA CORP. FOR NAPC ADMIN,, U.S. DfPT. HEW. 1970. 179 PP.
P8 200 650
GOETZ, A., AND T, KALLAT
INSTRUMENTATION FOR DETERMINING SIZE" AND MA88»DISt*lBUTlQW
Of SU8MICRON AEROSOLS
J. APCA, 12 (10), PP. 479-486. 1962
GREEN, H. L'.. AND W. R, LANE
PARTICIPATE CLOUDSl DUSTS, SMOKES AND MISTS
D. VAN NOSTRAND CO., INC., PRINCETON, 1964, 471 PP.
HELLER. W., AND M, NAKAGAKI
THEORETICAL INVESTIGATIONS ON THE LIGHT SCATTERING OF SPHERES.
xvii. ANGULAR & SPECTRAL LOCATION OF INTENSITY MAXIMA & MINIMA
J. OF CHEM. PHYSICS, 64 (12), PP. 4912««920, 1
-------
LOWE, H. J.. AMD n. H. LUCAS
THE PHYSICS OF ELECTROSTATIC PRECIPITATION
BRITISH J. OF APPL. PHY'., £«, SUPPLE ?, PP. S40«5«7, 195"*
MEPCER, T. T.
AEROSOL TECHNOLOGY IN HAZARD EVALUATION
ACADEMIC PRFSS. NEW YORK, N. V., 39« PP., 1*73
MORROW, N. I.., R. S. RRIFP, AND R*. R. BERTRAND
AIR SAMPLING AND ANALYSIS
CHEM. ENG/DESKBOOK ISSUE, PP. 125»iS2. 1972
OGLESBY, S.. JR.f AND G'. 8. NICHOLS
A -MANUAL OF ELECTROSTATIC PRECJPITATOR TECHNOLOGY
SOUTHERN RESEARCH INSTITUTE, NAPCA, CINCINNATI, OHIO
mo, 875 PP.
P8 196 3«0
PETFRS, E. T.. J. E. OBFRHQLTZiR, AND J. R. VALENTINE
DEVELOPMENT OF METHODS FOR SAMPLINg AND ANALYSIS OF PARTICULATE
AND GASEOUS FLUORIDES FROM STATIONARY SOURCES
ARTHUR D. LITTLF, EPA. DURHAM, Nt c'.» 1972, 129 PP.
EPA.R2-72-126 PB 213 313
RAGLAND, J. W.t K. M. CUSHINQ, J, o', MCCAIN, AND W. B. SMITH
HP-25 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUDIESI STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1977. 127 PP.
EPA-600/7-77-058
RAGLAND, J. W"., K. M, CUSHJNG, J. D'. MCCAIN, AND W. B. SMITH
HP-65 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUDIESI STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1976'. 122 PP.
RAO, A, K.
PARTICULATE REMOVAL FROM GAS STREAMS AT HIGH TEMPERATURE/HIGH
PRESSURE
MIDWEST RESEARCH INST'., EPA, RESEARCH TRIANGLE PARK, N. c.,
$3 PP., 1975
EPA-600/2-75-020 PR 245 858
223
-------
SEM, G. J.
STATE OF THF ARTJ 1971 INSTRUMENTATION FOR MEASUREMENT Of
PARTICUI ATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME J
THERMO.SYSTFMS, INC., EPA, RESEARCH TRIANGLE PARK, N.C.
1971, I9fl PP.
PB 202 665
SEM, 6. j'.
STATE OF THE ARTl 1971 INSTRUMENTATION FOR MEASUREMENT OF
PARTICULAR EMISSIONS FROM COMBUSTION SOURCES. VOLUME II
THERM0-SYSTFMS, INC., EPA, RESEARCH TRIANGLE PARK, N.C., 1971
225 PP.
PB 202 666
G. J.
STATE OF THE ART! 197! INSTRUMENTATION FOR MEASUREMENT OF
PARTICULATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME IIT
THERMO.SYSTEMS, INC., EPA, RESEARCH TRIANGLE PARK, N. C.
1972, 84 PP.
P8 233 391
SMITH, w. B., AND R. R. WILSON, JR,
DEVELOPMENT AND LABORATORY EVALUATION OF A FIVE-STAGE
CYCLONE SYSTEM
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N'.C,, 1978, 66 PP.
EPA-600/7-7S-008
SMITH, W. B*. , K. M. CUSHING, AND J. D'. MCCAIN
PROCEDURES MANUAL FOR ELECTROSTATIC PRECIPITATQR EVALUATION
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N, C., 1977. 430 PP.
EPA-600/7-77-059
SOUTHERN RESEARCH INSTITUTE
PROCEEDINGS OF THE WORKSHOP ON SAMPLING, ANALYSIS, AND MONITOR.
ING OF STACK EMISSIONS
SOUTHERN RESEARCH INSTITUTE,, ELECTRIC POWER RESEARCH INST.,
PALO ALTO. CALIFORNIA. 1975". 346 PP*.
WEITZMAN, L., AND j. c, REED
REVIEW OF PARTICIPATE STACK TEST REPORTS
67TH ANNUAL MEETING, APCA, DENVER, COLO,, 197«, PAPER 74-194
WHITE, H. J.
ELECTROSTATIC PRECIPITATION OF FLY ASH. PART I.
J. OF APCA, 27 (1), PP. 15-21, 1977
224
-------
WHITE, M, J'.
ELECTROSTATIC PRECIPITATION OF FLY ASH. PART IT.
J. OF APCA, 27 (2), PP. 114-120. 1977
WHITE, H. j.
ELECTROSTATIC PRECIPITATION OF FLY ASH. PART III.
J. OF APCA, £7 (3), PP. 206*?l7, 1977
WHITE, H. J.
ELECTROSTATIC PRECIPITATION qF FLY ASH, PART TfV
J. OF APCA, 27 f4), PP, 308-310, 1977
WHITE. H, J.
INDUSTRIAL ELECTROSTATIC PRECIPITATION
ADDISON*HESLEY PUBLISHING co.. INC., READING, ms, 376 PP.
2, SAMPLE EXTRACTION
AGARWAL, J. K'.
AEROSOL SAMPLING AND TRANSPORT
UNIVERSITY OF MINNESOTA, THESIS, 1975. 17S PP.
BROOKS, E. F.. AND R, L*. WILLIAMS
FLOW AND GAS SAMPLING MANUAL
TRW SYSTEMS GROUP, EPA, RESEARCH TRIANGLE PARK, N'.C,, 1976
100 PP.
EPA.600/2-76,203 PB 258 080
FUCHS, N. A*.
REVIEW PAPERS! SAMPLING OF AEROSOLS
ATMOS. ENVIRON., 9, PP, 697»707, 1975
, H, Fl, D. E. CAMANN, ANI5 R'. E, THOMAS
THE COLLABORATIVE STUDY OF EPA METHODS 5, 6, AND 7 IN FOSSIL FUEL
FIRED STEAM GENERATORS . FINAL REPORT
SOUTHWEST RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N.C., 19741, 39 PP.
EPA»650/a-7«»Ol3 PB 237 695
HANSON, H. A., R. J. DAVINJ, J. K'. MORGAN, AND A, A. IVERSEN
PARTICIPATE SAMPLING STRATEGIES FOR LARGE POWER PLANTS INCLUDING
NONUNIFORM FLOW
FLUIDYNE ENGINEERING CORP., EPA, RESEARCH TRIANGLE PARK,
N. C., 1976'. 371 PP.
EPA-60Q/2-76-170 PB 257 090
225
-------
t«!"^N2i B* J" J» L» HAU., A. *r. JOFNSEN, AND j. M. CARROLL
CORRECTION OF S-TVPE PITOT. STATIC TUBE COEFFICIENTS WHEN US
FOR TSOKINETjr SAMPLING FROM STATIONARY SOURCES "
ENVIRON. SCI. & TECH,. 11 (7). PP. 694.700, 1977
LOGAN, T, J., R. M. FELDER, AND j". K. FERRELt
EXPERIMENTAL INVESTIGATION OF ISOKINETIC AND ' ANISOKINETIT
SAMPLING OF PARTICULATES IN STACK GASFS
ANNUAL HEFTING AICHE, NEW YORK. N'.Y., 1972, 25 PP.
, D.f AND S8
AEROSOL SAMPLING WITH A SIDE PORT PROBE
. IND. HYG. ASSOC'. J., Pp. 20*.'ai5. 1967
, G. A.
BV
AFA SYMP. INSTRUMENTS & TECHNIQUES FOR ASSESSMENT OF AIRRORNF
RADIO ACTIVITY IN NUCLEAR OPERATIONS, VIENNA, 1967. PP.
SEL.OEN, M. G., JR.
ESTIMATES OF ERRORS IN ANISOKlNETIC SAMPLING OF PARTICULATE
j. OF APCA NOTEBOOK, 37 o). PP. ^35.?36, 1977
SEM, G. J. ..."
STATE OF THE ARTI 19T1 . INSTRUMENTATION FOR MEASUREMENT OF
PARTICULATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME i
THERMO-SYSTEMS, INC., EPA. RESEARCH TRIANGLE PARK. M C
1971, 194 PP.
PB 202 665
SEM, G. J.
STATE OF THE ART: l<*7l INSTRUMENTATION FOR MEASUREMENT OF
PARTICULATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME II
THERMO-SYSTEMS. INC., EPA. RESEARCH TRIANGLE PARK. N.C.. 1971
C. C T? fT *^ f ' • ' '
PR ?0? 666
SMITH, F. H'.
THE EFFECTS OF NOZZLE DESIGN AND SAMPLING TECHNIQUES ON AEROSOL
MEASUREMENTS
ARO, INC., EPA, RESEARCH TRIANGLE PARK, V. C.. 1974, 89 PP.
EPA-650/2-74.Q70 PR 2fl3 5fl8
226
-------
VITOIS, V.
THEORETICAL LIMITS OF ERRORS DUE TO ANISOKJNETIC SAMPLING OF
PARTICIPATE MATTER
J. OF APCA, 16 (2), PP. 79w«U, 1966
WALLIN, S. C.
SAMPLING OF PAPTICULATES EMITTED FROM STATIONARY SOURCES
ANN. OCCUP. HYG.» 16 (4), 353-571, 1973
WATSON, H. H.
ERRORS DUE TO ANI8DKINETJC SAMPLING OF AEROSOLS
AMER. IND. HYG. ASSOC. QUARTERLY 15 (13,
3, FILTER MEDIA
ADAMS, J., A. BENSON, AND E. PETERS
PROPERTIES OF VARIOUS FILTER MEDIA SUGGESTED FOR JN-sTACK
SAMPLING
ARTHUR D. LITTLE* INC'., NEW YORK, N*. Y., 1974, 20 PP.
BENSON, A, L.. Pt I. LEVINS, A. A'. MASSUCCO, AND
J. R. VALENTINE
DEVELOPMENT OF A HIGH-PURITY FILTER FOR HIGH TEMPERATURE
PARTICIPATE SAMPLING AND ANALYSIS
ARTHUR D. LITTLE, INC*., EPA, WASHINGTON, D. C., 1973, 80 PP.
EPA-650/2-70-032 PR 230 806
FELIX. L. G., G. I. CLINARD, 6..C'. LACEY, AND J. D. MCCAIN
INERTIAL CASCADE IMPACTOR SUBSTRATE MEDIA FOR FLUE GAS SAMPLING
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C,, 1977. S9 PP.
EPA-600/7-77-060
FORREST, J.. AND L. NEWMAN
SAMPLING AND ANALYSIS OF ATMOSPHERIC SULFUR COMPOUNDS FOR ISO-
TOPE RATIO STUDIES
ATMOS. ENVIROW.. 7, PP. 561*573, 1973
GELMAN, C., AND J. C. MARSHALL
HIGH PURITY FIBROUS AIR SAMPLING MEDIA
ANNUAL MEETING, AMER, IND'. HYG. ASSOC'., MIAMI, FLA. 1975
PP. 512-517
227
-------
MFMEON ASSOCIATES
ON THE FILTRATION EFFICIENCY OF ALUNPUM THIMBLES AND OTHF«
SAMPLING FILTERS
HEMFON ASSOC., PITTSBURGH, PA,, 1973, 8 PP.
LIU. B. Y. H.f AND K, W. LEE
EFFICIENCY OF MFMBRANE AND NUCLEPORI FILTERS FOR
SUBMICPOMETFR AEROSOLS
ENVIRON. SCI. AND TECH., 10 C«), Pp'. ia?«350f 1976
D. A,, AND K, T, WMlTBY
EFFECT OF PARTICLE ELECTROSTATIC CHARGE ON FILTRATION
BY FIBROUS FILTERS
IREC PROCESS DESIGN & DEVELOPMENT, fl f«), PP. 3«5-149, 1Q65
D. A., AND T, C. GUNOERSON
FILTRATION CHARACTERISTICS OF GLASS FIBER FILTER MEDIA
AT ELEVATED TEMPERATURES
UNIV. OF FLA., EPA, RESEARCH TRIANGLE PARK, N. C., 1976.
95 PP
EPA-feOO/2«76-192 PB 257 132
D. A., AND T. C.
EFFICIENCY AND LOADING CHARACTERISTICS OF EPA'S HIGH.
TEMPERATURE QUARTZ FIBER FILTER MEDIA
AMFR. IND. HYC. ASSOC'. J. 36 (12), PK 866-872, 1975
NEUSTADTER, H. E,, S. M". SIDK, AND R. 8. KING
THE USE OF WHATMAN«*I1 FILTERS FOR HIGH VOLUME AIR SAMPLING
ATMOS. ENVIRON. 9 (1). PP. 101*109, 1975
MASS CONCENTRATION
BEUTNER, H, P.
MEASUREMENT OF OPACITY AND PARTICIPATE EMISSIONS WITH AN
ON.STACK TRANSMISSOMETER
J. OF APCA, 2tt (91, PP, 865.871, I97a
BLANN, D. R".
MEASUREMENT METHODS AT HIGH TEMPERATURE AND PRESSURE
SYMP. PARTICIPATE CONTROL IN ENERGY PROCESSES. SAN
FRANCISCO, CALIF., 1974. 25 PP.
228
-------
Bt SANDRI
PARTICIPATE SAMPLING HAS GONE AUTOMATIC
fcftTH ANNUAL MFETING, APCA. BOSTON. MASS.,* 1975
PAPER 75-?4'.2
BRENCHLEY. D. I., C. D..TURLEV, AND R'. F,
USf[ OF THP FPA PARTICIPATE TRAIN FOR COMPLIANCE TESTING
AIP POLLUT. CONTROL & IMPOST, ENERGY PRODUCTION, CHAP, 4,
ANN ARBOR SCIENCE PUR., ANN ARBOR, MICH., 1975, PP. 73.85
BREUFR, H., J. GEBHART, K. POBOCK, AND U. TEICHE.RT
PHOTOELECTRIC MEASURING APPARATUS FOR DETERMINATION OF THF
FINE DUST CONCENTRATION'
SI AU8 REINHALTUN6 DER UUFT, IN ENGLISH, 33 («)
PP. 187-190, 1973
APPLICATION V AN OSCILLATING QUARTZ CRYSTAL TO MEASURE THE
MASS OF SUSPENDED PARTICULATE MATTER _
U5TH NAT. MEETING AM. CHEM. soc., 1973, DALLAS, TEXAS
1973, 30 PP.
FOR THE DIRECT MEASUREMENT OF PARTICIPATE MASS
AEROSOL SCI'., 1. PP. 1H"H«. 1970
PARTICULATE'MASS MEASUREMENT BY PIEZOELECTRIC CRVSTAL
PROCEEDINGS-SEMINAR ON AEROSOL MEASUREMENT, NAT'L BU». OF
STANDARDS, WASHINGTON, D. c>, 19?«, PP. I37*ias
RAPID* ASSESSMENT OF PARTICIPATE MASS CONCENTRATION IN THE
ATMOSPHERE WITH A PIEZOELECTRIC INSTRUMENT
ADV. INSTRUMENT. 30 (PT*. 2) ISA ANN. CONF., 1975, PAPER 620
CLARKE, A. G.. M. A, MOGHADASSI. AND A. WILLIAMS
A COMPARISON OF TECHNIQUES FOR AUTOMATIC AEROSOL MASS CONCENTRA'
TION MEASUREMENT
j. AEROSOL sci., s» PP. 73-fli. 1977
CONNER, W. D.
MEASUREMENT OF THE OPACITY AND MASS CONCENTRATION OF
PARTICIPATE EMISSIONS BY TRANSMISSQMETRY
EPA, RESEARCH TRIANGLE PARK, N. c., 197*. 39 PP.
EPA-650/2-7«.l?8 PR 2«1 25-1
229
-------
DALEY, P. S., AMD D. A,
THE PERFORMANCE OF PIEZOELECTRIC CRYSTAL SENSORS USED TO
DETERMINE AEROSOL MASS CONCENTRATIONS
AMEH. IND. HVG. ASSOC. J., PP. 51R.53?, 1975
DOBBINS, ». A.. AND 6. 8. J
OPTICAL SCATTERING CROSS SECTIONS FOR POLYOISPERSIQNS OF
DIELECTRIC SPHEPES
J. OPT. SOC. AMER.t 56 (13), PP. IjflS-lSSO, 1966
DOBBINS, R. A., AND G. S, JIZMA6IAN
PARTICLE SIZE MEASUREMENTS BASED ON USE OF MEAN SCATTERING
CROSS SECTIONS
J. OPT. SOC. OF AMER. 56 UO). PP. 13§1»135USTRlAl
SOURCES, U.S..U.S.S.R. WORKIN0 SROUP, SAN FRANCISCO, CALTF.,
1974. PAPER 26
DORSEY, J. A., AND J, 0*.
CONTINUOUS PARTICULATE MONITORING
63RD ANNUAL MEETING, AICHE, CHICAGO. ILL., 1970, PAPER 5A
FUNKHOUSER, J. T.
MANUAL METHODS FOR SAMPLING AND ANALYSIS OF PARTICUUTE
EMISSIONS FROM MUNICIPAL INCINERATORS
ARTHUR C, LITTLf, CO.. EPA, WASHINGTON. D*. C., 1973, 293 PP.
EPA-650/2-73-023 PR ?38 <*76
GRUBER, ARNOLD H,
IN. STACK CONTINUOUS PARTICULATE MONITORING USING THE CHARGE
TRANSFER PROCESS
APCA SPECIALTY CONFERENCE! CONTINUOUS MONITORING OF STATION.
ARY AIR POLLUTION SOURCES, 8T. LOUIS, MISSOURI, 1975, 20 PP.
GRUBER, A. H., AND E, K*. BASTRESS
APPLICATION OF THE TRIBOELECTRIC EFFECT TO THE MEASUREMENT OF
AIRBORNE PARTICLES
??ND JOINT CONF. SENSING OF ENVIRONMENTAL POLLUTANTS
INSTRUMENT SOC, OF AM'., 1973, pp, 161-160
GUNTHER, R.
AM OPTICAL SMOKE DENSITY M£T£R FOR DIRECT INDICATION OF THF,
AMOUNT OF SOLIDS PER CUBIC METER OF FLUE GAS
STAUB REINHALTUN6 DER LUFT, 33 (93, PP. 3^5-35^, 1973.
230
-------
HAMIL, H, F., D. F. CAMANN, AND P. F, THOMAS
THE COLLABOPAHVE STUDY OP FPA METHODS 5, 6, AND 7 IN FOSSIL FUEL
FIRED STEAM GENERATORS • FINAL REPORT
SOUTHWEST RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N.C., 197U, 39 PP.
EPA".650/4-7««Ol3 PB 237 695
HANSON, H. A., AND 0. P. SAARI
EFFECTIVE .SAMPLING TECHNIQUES FOR PARTICIPATE EMISSIONS. FROM
ATYPICAL STATIONARY SOURCES - INTERIM REPORT
FLUIDVNF ENGINEERING CORP., EPA, RESEARCH TRIANGLE PARK, N.C.
1977, 130 PP.
HANSON, H. A., R, J. DAVINJ, J. K. MORGAN, AND A. A. IVERSEN
PARTICULAR SAMPLING STRATEGIES FOR LARGE POWER PLANTS INCLUDING
NONUNIFORM FLOW
FLUIDYNE ENGINEERING CORP., EPA, RESEARCH TRIANGLE PARK,
N. C., 1976'. 371 PP.
EPA»600/a-76-170 P8 257 090
noon, K, T.
OPACITY AND PARTICIPATE EMISSION RELATIONSHIPS FOR PULP MILLS
NATIONAL COUNC. OF THE PAPER INO. FOP AIR AND STREAM
IMPROVEMENT, INC.. 1976
HOPTON, F'. J.i N, G. H, GUILFORD, j'm A. CRAIGMILE,
H. C. H. VEPGEER* AND L'. E. FRENCH
A TEST SYSTEM FOR EVALUATION OF SOURCE SAMPLING TECHNIQUES
AND CALIBRATION OF EQUIPMENT
NEWSLETTER, SOURCE EVALUATION SOCIETY, WESTON, CONN.
1 (3), 8 PR'., 1976
HUSAP,, R. 8.
ATMOSPHERIC PARTICIPATE MASS MONITORING WITH A BETA RADIATION
DETECTOR
ATMOS. ENVIRON., 8, PP. 183»t88, 197«
KENDALL, D. R.
RECOMMENDATIONS ON A PREFERRED PROCEDURE FOR THE DETERMINATION
OF PARTICIPATE IN GASEOUS EMISSIONS
J. OF APCA, 26 (9), PP. 871»«7a, 1976
KUTYNA, A. G,
COMPARISON OF SOURCE PARTICULATE EMISSION MEASUREMENT METHODS
FOR COMBINATION FUEL»FIRED BOILERS
TECHNICAL BULLETIN 75, NATIONAL COUNCIL OF THE PAPER INDUSTRY
FOR Al» AND STREAM IMPROVEMENT, INC., I97fl. 35 PP.
231
-------
LAPSON, w. F.f AND H, j'. DEHNE
DESIGN, DEVELOPMENT, AND FABRICATION OF A PROTOTYPE HIGH. VOLUME
PARTICULATE MASS SAMPLING TWAIN
ACUREX-AEROTHERM CORP., EPA, RESEARCH TRIANGLE PARK, N, C.,
1974. 37 PP.
EPA-fe5Q/2*7««067 PB 245 19A
LTLIENEELD, P.
DESIGN AND OPERATION OE DUST MEASURING INSTRUMENTATION BASED ON
THE BETA-RADIATION METHOD
STAUR REINHALTUNG DER LUFT, 35, PP. 458-465, 1975
LYTLE, J. H.
A$ME METHOD FOR MEASUREMENT OF PARTICULATE EMISSIONS
WORKSHOP. SAMPLING, ANALYSIS, AND MONITORING OF STACK
EMISSIONS, DALLAS, TEXAS, 1975, PP, 203-220
MAC I AS, E. S., AND R, B'. HUSAR
ATMOSPHERIC PARTICULATE MASS MEASUREMENT WITH BETA ATTENUATION
MASS MONITOR
ENVIRON. SCI, & TECH.. 10 (9), PP. 9Q4.9Q7*
HACIAS, E. S., AND R, B'. HUSAR
A REVIEW OF ATMOSPHERIC PARTlCULATg MASS MEASUREMENT VIA THE
BETA ATTENUATION TECHNIQUE
PROCEEDINGS! SYMPOSIUM ON FINE PARTICLES, MINNEAPOLIS, MINN.
1975, PP. 555-564
NADER, J. S.
CURRENT TECHNOLOGY FOR CONTINUOUS MONITORING OF PARTICUUTE
EMISSIONS
j'. OF APCA, 25 (8), PP, 81«»B2J, 1975
OLJN, J. G.. G. J. SEM, AND R. P. TRAUTNER
AIR»OUALITY MONITORING OF PARTICLE MASS CONCENTRATION WITH A
PIEZOELECTRIC PARTICLE M1CROBALANCE
6«TH ANNUAL MEETING, APCA, ATLANTIC CTTY, N.J.
1971, PAPER 71-1
PFISTER, E., AND F. E*. SICK
AN INTEGRATION INSTRUMENT FOR TIMEWISE EVALUATION OF
EMISSIONS
STAUR REINHALTUNG DER LUFT, (ENGLISH). 34 (2), PP. 53-56
1974
232
-------
PU.AT, M.
PLUME
ATMOS
J., AND D. S.
HPACITY AND PARTICULATE MASS CONCENTRATION
ENVIRON'., <». PP. 163-171, 1070
PROCHAZKA, R,
RECORDING DUST MEASUREMENT HITH THf KONITEST
PP. 22»2fl. DATE & SOURCE NOT DETERMINABLE
RAGLAND, J. W., K, M, GUSHING* J. D'. MCCAIN, AND W, B. SMITH
HP.25 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUDIES! STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1977. 127 PP.
EPA-60Q/7-77»058
RAGLAND, J. W., K. M. CUSHINf,, J. D. MCCAIN, AND W. B. SMITH
HP-65 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUDIESI STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1976. 12? PP.
EPA-600/8-76-002
REISMAN, E.. W. D. GER8ER, AND N. p'. POTTER
IN-STACK TRANSMISSOMETER MEASUREMENT OF PARTICIPATE OPACITY
AND MASS CONCENTRATIONS
PHILCO.FORD CORP., IPA, RESEARCH TRIANGLE PARK, N.C.,
115 PP.
EPA-65ft/2-7a-120 PR 239
MONITORING
SEANY, R. J., R. K, HALPIN, AND B, A. MAGUIRE
A PORTABLE RECORDING INSTRUMENT
-------
SEM, G, J.
STATE OF THF ART! 1971 INSTRUMENTATION FOR MEASUREMENT OF
PARTICULAR EMISSIONS FROM COMRIJSTlON SOURCES. VOLUME II
THERMO. SYSTEMS, INC., EPA, RESEARCH TRIANGLE PARK. N.C., 1971
225 PP.
PR iO? 666
6. J.
STATE OF THE ARTJ 1971 INSTRUMENTATION FOR MEASUREMENT OF
PARTICIPATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME III
THERMQ-SYSTFMS, INC., EPA, RESEARCH TRIANGLE PARK. N. c.
197?, 8fl PP.
PB J?33 393
G. J., K. TSURUBAYASHI, AND K, HQMMA
PERFORMANCE OF THE PIEZOELECTRIC MiCROBALANCE RESPIRABLE
AEROSOL SENSOR
AM, INO. HYG. ASSOC. J,. 38, 1977.. FP*. 580-588,
SHOFNER, F. M., G. KREIKEBAUM, AND H. W*. 3CHMJTT
IN SITU CONTINUOUS MEASUREMENT OF PARTICLE MASS CONCENTRATION
68TH ANNUAL MEETING, APCA. BOSTON, MASS., 1975, PAPER 75-41.1
SMITH, W. B"., K. M, CUSHING, AND J. D*. MCCAIN
PROCEDURES MANUAL FOR ELECTROSTATIC PRECIPITATOR EVALUATION
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1977. «30 PP.
EPA-600/7-77.059
SOUTHERN RESEARCH INSTITUTE
PROCEEDINGS OF THE WORKSHOP ON SAMPLING, ANALYSIS, AND MONITOR.
IMG OF STACK EMISSIONS
SOUTHERN RESEARCH INSTITUTE, ELECTRIC POWER RESEARCH IN$T.,
PALO ALTO, CALIFORNIA, 1975. 3«6 PP.
STEEN, B. '
A NEW SIMPLE ISOKINETIC SAMPLER FOR THE DETERMINATION OF
PARTICLE FLUX
ATMOS. ENVIRON., tl, PP". 623«6?7, 1977
ZALEIKO, N, S'., AND A. LICATA
PARTICLE MASS SOURCE MONITORING, MANUAL VS. INSTRUMENTAL
1975ANALYSIS INSTRUMENTATION
R. SOC. OF AM., 13, PP. H5«122, 1975
234
-------
5, PARTICLE-SIZE DISTRIBUTIONS
AGARWAI , J, K., AMP G*. J. 3EM
A CONTINUOUS FLOW CNC CAPABLE Of COUNTING SINGLE PARTICLES
UNPUBLISHED. THERMQ8YSTFMS, I*C'., TECHNICAL PAPER,
5 TYPED PAGES.
ALTPETER, L. L., JR., J'. P. PILNEY, L'. Wf RUST,
AND 0. L. OVERLAND
RECENT DEVELOPMENTS REGARDING THE USE OF A FLAME IONIZATION
DETECTOR AS AN AEROSOL MONITOR
PROCEEDINGS! SYMPOSIUM ON FINE PARTICLES, MINNEAPOLIS, MJNN.
PP. 6?5»6<»7, 1975
ANDERSEN, A. A.
A SAMPtER FOR RESPIRATORY HEALTH HAZARD ASSESSMENT
AMER. IND. HYG. ASSOC*. J,, PP. 160»16§. 1966
ANDERSEN, A. A.
NEW SAMPLER FOR THE COLLECTION, SIZING, AND ENUMERATION OF
VIABLE AIRBORNE PARTICLES
J. OF BACTERIOL. 76, PP*. 471. «84.
ANDERSON, D". P., AND P. LILIENFELD
DEVELOPMENT AND TESTING OF AN IN.STACK VIRTUAL IMPACTOR
70TH ANNUAL MEETING, APCA. TORONTO, 1977, PAPER 77-1?. 7
ANDERSON, P. L.
DEVELOPMENT OF A CENTRIFUGE FOR SOURCE SAMPLING TO DETERMINE
PARTICLE SIZE DISTRIBUTION
ENVIRON. SCI. & TECH. 10 (2), PP. IflSMSO, 1976
ANDERSON, W'. L.f AND R, E. BEISSNER
COUNTING AND CLASSIFYING SMALL OBJECTS BY FAR^FIELD LIGHT
SCATTERING
APPL. OPT., 10 (7), PP. 1501-1508, 1971
ARAGON, S. R. AND R, PECORA
THEORY OF DYNAMIC LIGHT SCATTERING FROM POLYDISPERSE SYSTF-MS
J. OF CHEM. PHYS., 6« (6), PP. 2395*2403* ' 1976
BAKHANOVA, R. A., AND L*. V. IVANCH£NKO
THE CALIBRATION CURVE OF PHOTOELECTRIC COUNTERS AND COMPUTATION
OF PARTICLE SIZE DISTRIBUTION WHEN THE RELATIONSHIP BETWEEN
PARTICLE SIZE AND ELECTRICAL PULSE AMPLITUDE is AMBIGUOUS
AEROSOL SCI'., «, PP.
-------
BARTH. W.
DESIGN AMD LAYOUT OF THE CYCLONE SEPARATOR ON THE BASIS OF
NFW INVESTIGATIONS
. WAERMF KRAFT, 8 (GERMAN), pp. 1-9, 1956
RELDEN, L. H,. AND C, M,
OPTICAL MEASUREMENT OF PARTICLE SIZE DISTRIBUTION AND
CONCENTRATION
TECHNICAL INFORMATION SERIES REPORT, GENERAL ELECTRIC CO..
, «5 PP.
BERNER, A.
PRACTICAL EXPERIENCE WITH 20-STAGE IMPACTOR
STAUB REINHALTLING DER LUFT, IN ENGLISH, 32 (85, PP, 1*8, 197?
BERNER, A,
A SIMPLE PROCEDURE FOR CORRECTION OF FREQUENCY DISTRIBUTIONS
MEASURED WITH A MULTI-STAGE IMPACTOR
STAUB REINHALTIJNG DER LUFT, IN ENGLISH, 33 U), PP. 190-19<1
BETHEA, R, M., AMD P. R. MQREY
A COMPARISON OF COTTON OUST SAMPLING TECHNIQUES
AMER, INO. HGY. ASSOC*. J., 37* PP. «.47*65«, 1976
BEXON, R. , G. 0. BISHOP, AND J, 8IGG3
AEROSOL SUING BY HOLOGRAPHY USING THE QUANTIMET
CAMBRIDGE INSTRUMENT COMPANY, INC'.; BROCHURE, DATE NOT
DETERMINABLE
BLACHMAN, M. W, , AND M,
PERFORMANCE CHARACTERISTICS OF THE MULTICYCLONE AEROSOL
SAMPLER
AMER. .INO. HYG. ASSOC'. J., PP. 311-326, 197«
BLACKER, 8. M.
EVALUATION OF THE ANDERSEN STACK SAMPLER FOR PARTICLE SIZE
DETERMINATION
APPLIED TECH. DIV., EPA. DURHAM, N. C., 19 PP., 1972
BLAKE, D.
OPERATING AND SERVICE MANUAL SOURCE ASSESSMENT SAMPLING SYSTEM
ACURE* CORP., EPA, RESEARCH TRIANGLE PARK. N.C.,
1977, 115 PP.
236
-------
BLANK', D. R.
MEASUREMENT METHODS AT HIGH TEMPERATURE AMD PRESSURE
SYMP. PARTICULATE CONTROL IN ENERGY PROCESSES, SAN
FRANCISCO, CALIF., i97«f 25 PP.
RRFSLIN, A. J., S. F. GUGGENHEIM, AMD A. C. GEORGE
COMPACT HIGH. EFFICIENCY DIFFUSION BATTERIES
STAUR REINHALTUNG DER LUFT, IN ENGLISH,
31 («), PP. 1-5, 1971
BRINK, J. A., JR. .....
CASCADE IMPACTOR FOR ADIABATIC MEASUREMENTS
IND. AND ENG. CHEM,, 50 (4), PP. 6Ǥ-648, 1958
K, J. A,, JR., E. D'. KENNEDY, AND H. S. YU
PARTICLE SIZE MEASUREMENTS WITH CASCADE IMPACTQRS
65TH ANNUAL MEETING, AICHE, NEW YORK, N. Y., 1972
BROOKMAN, R. S., J, F. PHIULIPPI, AND C. L. MA1SCH
SMALL-DIAMETER CYCLONES
CHEM. ENGIN. PROGRESS. 59 (in, PP. 66»69. 1963
8UCHHOLZ, H.
AN UNDERPRESSURE CASCADE IMPACTOR
STAU8 REINHALTUNG DER LUFT, IN ENGLISH, 30 («), PP. J7*20,
1970
BURKHOLZ, A.
INVESTIGATIONS ON A CASCADE IMRACTOR
STAUB REINHALTUNG DER LUFT, IN ENGLISH, 31 flO), PP, 381-385
1973
CADLE, R. D.
PARTICLE SIZE DETERMINATION
INTERSClENcr PUBLISHERS. INC., NEW YORK, 1955, 303 PP.
, s., c. LAKE, AND R. PARKER
CASCADE IMPACTOR CALIBRATION GUIDELINES
A.P.T., INC., EPA, RESEARCH TRIANGLE PARK, N.C., 1976,
-------
CAPI.AN, K. .!., I , J, DOEMENY, AND g. D. SORENSON
PERFORMANCE CHARACTERISTICS OF THE 10 MICRON RESPIRABLE
SAMPLER* PART II * COAL DUST STUDIES
AMER. INO. HVG. ASSOC. J., 38. PP. 162-173, 1977
CARPENTER. T. E.f AND D. L, BRENCHLEY
A PIEZOELECTRIC CASCADE IMPACTOR FOR AEROSOL MONITORING
AMER. IND. HYG. ASSOC'. J,. PP. 503-510, 1972
CHANG, H. C.
A PARALLEL MIJI. T ICYCLONE SIZE-SELECTIVE PARTICIPATE SAMPLING
TRAIN
IND. HYG. ASSOC. J., PP. 538-545, 1974
CHAN, P. W,
OPTICAL MEASUREMENTS OF SMOKE PARTICLE SIZE GENERATED
BY ELECTRIC ARCS
COLO. STATE UNIV., EPA, WASHINGTON, 0. Ct, 1974, 49 PP.
EPA-650/2-74-034 PB 236 580
CHAN, T., AND M. LIPPMANN
PARTICLE COLLECTION EFFICIEWCIES OF AIR SAMPLING CYCLONES*
AN EMPIRICAL THEORY
ENVIRON. SCI. & TECH, 11 (4), PP. 377-382, 1977
CLAUSEN, J... A. GRANT, 0. MOORE. AND 8. REYNOLDS
FIELD SAMPLING FOR CYTOTOXJ.CITY TEST SAMPLES USING A SERIES
CYCLONE SAMPLING TRAIN
TRW SYSTEMS GROUP, EPA, RESEARCH TRIANGLE PARK, N. C.
1975, 91 PP.
COHEN, J, J., AND D, N, MONTAN
THEORETICAL CONSIDERATIONS, DESIGN, AND EVALUATION OF A CASCADE
JMPACTQR
AMER. IND. HYG. ASSOC'. J,. PP. 95-104, 1976
COLLINS, E. A'., J. A, DAVIDSON, AND C'. A, DANIELS
REVIEW OF COMMON METHODS OF PARTICLE SIZE MEASUREMENT
J. PAINT. TECHNOL. 47 (604), pp. 35*56. 1975
COOPER, DOUGLAS W.
DATA INVERSION METHOD AND ERROR ESTIMATE FOR CASCADE
CENTRIPETERS
AM. IND, HYG. ASSOC. Jt, 37, PP. 622-627, 1976
238
-------
D. W,, AND J, W. DAVIS
CASCADE IMPACTORS FOR AEROSOLS} IMPROVED DATA ANALYSIS
AMER. IND. HYG. ASSOC'. J., 33. PP. 79-89, 1972
COOPER, 0. «.. AND L, A. SPIELMAN
A NEW PARTICLE SIZE CLASSIFIERi VARIABLE-SLIT IMPACTOR WITH
PHOTO-COUNTING
ATMOS. ENVIRON. 8.. PP. 321»232,
CORNTLLAULT, j.
PARTICLE SIZE ANALYZER
APPL, OPTICS, 11 (2), PP. 265*268. 1972
COUCHHAN, J. c.t AND H, N.
SIMPLIFIED METHOD FOR DETERMINING CASCADE IMPACTOR STAGE
EFFICIENCIES
AMER. IND. HYG. ASSOC*. J,, PP. 62*67, 1967
GUSHING, K. M.. G. E. LACEY, J. D, MCCAIN, AND W, B. SMITH
PARTICULATE SIZING TECHNIQUES FOR CONTROL DEVICE EVALUATION,
CASCADE IMPACTOR CALIBRATIONS
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK.
N. C., 1976, 9a PP.
EPA- 600/2-76-280
DAVIES, C, N,t AND M,
THE TRAJECTORIES OF HEAVY, SOLID PARTICLES IN A TWO-
DIMENSIONAL JET OF IDEAL FLUID IMPINGING NORMALLY UPON A PLATE
PROCEEDINGS. PHYSICAL SOC., 6«, PP. 889*991, 1951
DAVIES, R.
PARTICLE SIZE ANALYSIS
IND. AMD ENG. CHEM., 62 (12), PP, 87-9J, 1970
DINGLE. A. N.. AND B, M. JQSHl
AMMONIUM SUI.FATE CRYSTALLIZATION IN ANDERSEN CASCADE
IMPACTOR SAMPLES
ATMOS. ENVIRON. 8, PP. 1119*1130, 1974
DOBBINS, R. A., AND G'. S. JIZMAGIAN
PARTICLE SIZE MEASUREMENTS BASED ON USE OF MEAN SCATTERING
CROSS SECTIONS
J. OPT, SOC. OF AMER, 56 (10). pp'. 1351*1350, 1966
DROZIN, F. G., AND V, K. LAMER
THE DETERMINATION OF PARTICLE SIZE DISTRIBUTION OF AEROSOLS
BY PRECIPITATION OF CHARGED PARTICLES
J. COLLOID SCT., 14, PP'. 7a-90, 1959
239
-------
DZUBAY, T. G.. L, E. HlNFS, AND R. K. STEVENS
PARTICLE BOUNCE ERRORS TN CASCADE JMPACTORS
ENVIRON. io, PP, 2?9»?3«, 1974
Tubl?; !" G" 6> *' CANARD, G. E. LACEY, AND J. D. MCCAIN
INERTIAL CASCADE IMPACTOR SUBSTRATE MEDIA FOR. FLUE GAS SAMPl IN,S
SOUTHERN RESEARCH INSTITUTE, EPA. RESEARCH TRIANGLE PARK.
N. C., 1977, 89 PPf
FERNANDEZ, G.
EiSJ1-1, 3m DETERMINATION BY USE OF ELECTROSTATIC
TION PATTERNS AND RADIOACTIVE TRACERS
THESIS. AIR UNIVERSITY,
FLESCH, J. P.
CALIBRATION STUDIES OF A NEW SU8-MICRON AEROSOL SIZE
CLASSIFIER
J. OF COLL. AND INTERFACE SCI. 29 (3). PP. 502*509, 1969
FLESCH, j. P.. c, H, NORRIS, AND A. E'. NUGENT, JR.
CALIBRATING PARTICULATE AIR SAMPLERS WITH MONOOISPERSE AFROSOLSl
APPLICATION TO THE ANDERSEN CASCADE JMPACTOR
AMER. IND, HGY. ASSOC'.;J., PP. 507.516, 1967
FLOYD, J,, AND K, KNAPP
FINE PARTICLE MEASUREMENT IN STATIONARY SOURCES
69TH ANNUAL MEETING, APCA. PORTLAND, OREGON! 1976,
PAPER 7fe«30.10
FORNEY, L. J.
AEROSOL FRACTIONATOR FOR LARGE«SCALF SAMPLING
REV. SCI. INSTRUM., ^16 (10). PP. 1264.1269, 1976
FUCHS, N. A., I. 8, STECHKINA, AND V, I, STARQSSELSKI I
ON THE DETERMINATION OF PARTICLE SIZE DISTRIBUTION JN POL y-
DISPERSE AEROSOLS BY THF DIFFUSION METHOD
BRIT. j. APPL*. PHYS. 16. PP. 260*281 ,
GER8ER, H.E.
ON THE PERFORMANCE OF THE GOETZ AEROSOL SPECTROMETER
ATMOS. ENVIRON,, 5, PP. t009«lOSl. 1971
GOETZ, A., H. J. Rt STEVENSON, AND 0, PREINJNG
THE DESIGN AND PERFORMANCE OF THE AEROSOL SPECTROMETER
J. APCA, 10 (5). PP. 376*583, I960
240
-------
GOLDSCMNIDT, V. W.
OF AEROSOL CONCENTRATIONS WITH A HOT
J. OF COLLOID SCI., 20, PP. 617-634, 1965
GOOD1NG, C. H.
WIND TUNNEL EVALUATION OF PARTICLE SIZING INSTRUMENTS
RESEARCH TRIANGLE INST.. EPA, RESEARCH TRIANGLE PARK, N.r.
1976, 72 PP. •
EPA-600/2.76-073 P8 251 n%
GRAEDEL, T. E.
CHANNEL WIDTH DETERMINATION AND ELECTRONIC PULSE PROCESSING
LOSSES IN OPTICAL PARTICLE COUNTERS
AEROSOL SCI., 5, PP. 12S-131, 1974
GRASSLt H.
DETERMINATION OF CLOUD DROP SIZE DISTRIBUTIONS FROM SPECTRAL
TRANSMISSION MEASUREMENTS
BEITRAGE ZUR PHYSIC DER ATMOSPMARF UN GERMAN), 43.
PP. 255-284. 1970
GRASSL, H.
DETERMINATION OF AEROSOL SIZE DISTRIBUTIONS FROM SPECTRAL
ATTENUATION MEASUREMENTS
APPL. OPT. 10 (11), PP. 2554.2538, 1971
GRAVATT, C. C.
LIGHT SCATTERING METHODS FOR THE CHARACTERIZATION OF
PARTICULATE MATTER IN REAL TIME
PROCEEDINGS . SEMINAR ON AEROSOL MEASUREMENTS NAT'L. BUR OF
STANDARDS, WASHINGTON, 0. C., 197« " * '
GRAVATT, C. C., JR.
SJitrl^E MEASUREMfcNT OF THE SIZE DISTRIBUTION OF PARTICULATE
MATTER BY A LIGHT SCATTERING METHOD
J. OF APCA, 23 (123. PP. 1035-103S, 1973
GRAY, D. C,
SURVEY OF SAMPLING TECHNIQUES FOR DEFINING RESPIRABLE CONCFN.
TRATION AND/OR PARTICLE-SIZE CHARACTERISTICS OF AEROSOLS
LOS ALAMOS SCI. LAB., ERDA, WASHINGTON, D. C., H PP., 1976
SUCKER, F. T.. J. TUMA, H. M. LIN, C. M. HUANG, S. C. EMS,
AND T. R. MARSHALL
RAPID MEASUREMENT OF LIGHT-SCATTERING DIAGRAMS FROM SINGLE
™X!;LrS ^ AEROSOL STREAM & DETERM. 0F LATEX PARTICLE SIZE
AEROSOL SCI., 4, PP. 189»«04, 1973
241
-------
GUSSMAN, R. A., A. M, SACCO, AND M. M'.'MCMAMQN
DF8IGN AND CALIBRATION OF A HIGH. VOLUME CASCADE IMPACTOR
J. OF APCA, ;?3 (9), PP. 778*78?, 197?
HABERL, J. B.
A LINEAR SCALE AITKEN NUCLEI COUNTER WITH AUTOMATIC RANGE
SELECTION " &
J. OF APCA, 2, 3 PP., 1977
HARRIS, 0. R.
PROCEDURES FOR CASCADE IMPACTOR CALIBRATION AND OPERATION IN
PROCESS STREAMS
EPA, WASHINGTON, D. C'., 1977, l?l pp.
HEYDER, J,, AMD J. PORSTENDORFER
COMPARISON OF OPTICAL AMD CENTRIFUGAL AEROSOL SPECTROMETRYj
LIQUID AND NON»SPHERICAL PARTICLES
AEROSOL SCI, 5, PP. 387-400, 1974
HINDE, A. L., AND P. J. D. LLOYD
REAL-TIME PARTICLE SIZE ANALYSIS IN WfT CLOSED-CIRCUIT Mj| L
POWDER TECHNOL. 12 tn, PP. 57-so, 1975
HOCHSTRASSER, J. M.
THE INVESTIGATION AND DEVELOPMENT OF CYCLONE DUST COLLECTOR
THEORIES FOR APPLICATION TO MINIATURE CYCLONE PRESAMpLERS
DISSERTATION, UNIVERSITY OF CINCINMATI, 1976, 368 PP.
HOGAN, A, W.
CALIBRATION OF PHOTO ELECTRIC NUCLEUS COUNTERS
68TH ANNUAL MEETING, APCA, BOSTON, MASS., 1975, PAPER
75-62.2
HORVATH, H., AND A, T*. ROSSANO, JR.
TECHNIQUE FOR MEASURING DUST COLLECTOR EFFICIENCY AS A FUNCTION
OF PARTICLE SIZE
J. OF APCA, 20 f«), PP. 24<|-2«6, 1970
HOTHAM, 6. A..
SUE OF RESPJRABLE AEROSOLS BY PULSING UV LASER MACHINE
AEROSOL MEASUREMENT SEMINAR, FDA AND NBS, GAITHER8BURG, MD.,
ma, 55 PP.
HOUNAM, R. F.f AND R. J, SHERWOOD
THE CASCADE CENTRlPETERi A DEVICE FOR DETERMINING THE
CONCENTRATION AND SIZE DISTRIBUTION OF AEROSOLS
IND. HYG, J., PP. 122-Ht, 1965
242
-------
II NOVA, .«'. , AND S. YLI
ON SEPARATION OF MfcCHANJS* OF TWO-DIMFNTIONAL CASCADE
KAQAKU KOGAKII, 33, PP. 1265«l?7t. 19*9
NI, R, R., AND C, F, CALLIS
PARTICLE SIZE* MEASUREMENT, INTERPRETATION, AND APPLICATION
JOHN WILEY .& SONS., INC'., NEW YORK, 1963, 165 PP.
JACKSON, M. i,,
PARTICLE-MOLECULE COLLECTION BY SONIC FLOW IMPINGERS
J. OF APCA, ?U (6), PP, 569»57§, 1974
JACKSON, M. u, S. CHIMQNAS, AND R. G'. PATTERSON
SAMPLE COLLECTION OF SOLID AND LIQUID AEROSOLS BY SONIC-
FLOW IMPINGEMENT
76TH NATIONAL MEETING OF AICHE, MAR. 1*74
JAENICKE, R.
THE DOUBLE-STAGE 1MPACTOR, A FURTHER APPLICATION OF THE
IMPACTOR PRINCIPLE
STA«'B REINHALTUNG OER LUFT, 31 Cfe), PP. 1*10, 1971
JAENICKE, R'., AND H. J, KANTER
DIRECT CONDENSATION NUCLEI COUNTER WITH AUTOMATIC PHOTOGRAPHIC
RECORDING, AND GENERAL PROBLEMS OF "ABSOLUTE" COUNTERS
J. OF APPL. METEOROLOGY. 15 (6), Pp'. 620-632, 1976
KERKER, M, '
LIGHT SCATTERING BY SINGLE AEROSOL PARTICLES
PROCEEDINGS * SEMINAR ON AEROSOL MEASUREMENTS NAT'L BUR. OF
STANDARDS, WASHINGTON, D. C., 197fl
KNQLLENBERG, R, G.
THE OPTICAL ARRAYi AN ALTERNATIVE TO SCATTERING OR EXTINCTION
FOR AIRBORNE PARTICLE SIZE DETERMINATION
J. OF APPL. METEOROLOGY, 9, PP. 86.105, 1970
KNOLLENBERG. R. G.
ACTIVE SCATTERING AEROSOL SPECTROM£TRY
PROCEEDINGS * SEMINAR ON AEROSOL MEASUREMENT NAT«L. BUR. OF
STANDARDS, WASHINGTON, D, C., 1974
KNOLLENBERG, Rt G..
THREE NEW INSTRUMENTS FOR CLOUD MEASUREMENTS | THE 2-D SPECTRO-
METER, THE FORWARD SCATTERING SPECTROMETER PROBE, AND THE
ACTIVE SCATTERING AEROSOL SPECTROMETER
PREPRINT VOLUME INTERNATIONAL CONF. ON CLOUD PHYSICS.
1976
243
-------
KNUTSON, F. 0., AND K. T. WHTTBV
AEROSOL CLASSIFICATION BY ELECTRIC MOBILITYl APPARATUS,
AND APPLICATIONS
j. AEROSOL set.. *, PP. 443*451. 1975
KOPS, J., L. -HERMANS, AND J, F, VAN DE VATE
CALIBRATION OF A STOBER CENTRIFUGAL AEROSOL
AEROSOL SCI., S, PP. 379-386,
KRFIKEBALIM, G., AND F. M. SH.OFNER
DESIGN CONSIDERATIONS AMD FIELD PERFORMANCE FOR AN INSJTU,
CONTINUOUS FINE PARTICULATE MONITOR BASED ON RATIO-TYPE LASER
LIGHT SCATTERING
INTERNAL CONF. ENVIRON, SENSING AND ASSESSMENT, LAS VEGAS,
NEVADA, }97% 18 PP.
KUBIE, G,
A NOTE ON A TREATMENT OF IMPACTOR DATA FOR SOME AEROSOLS
AEROSOL SCIENCE, 2, PP. 23-30, 1971
LEHBETTER, J. Q,, AND 8'. R, FISH
THE JET FILTER - SINGLE-STAGE SIZE-SELECTIVE SAMPLER FOR
AIRBORNE PARTICIPATES
. IND. HYG. ASSOC. J.. PP. 90-93, 1972
UE, R. E., JR.
SIZE DISTRIBUTION OF SUSPENDED PARTICULATES IN AIR
RESEARCH AND DEVELOPMENT, PP. i»«ai, 1972
LEITH, D., AMD D. MEHTA
CVCLONE PERFORMANCE AND DESIGN
ATMOS. ENVIRON., 7, PP. 527-5«9, 1973
LEITH, D., AND M. W. FIRST
UNCERTAINTY IN PARTICLE COUNTING AND SIZING PROCEDURES
AMER. IND. HYG. ASSOC'. J., PP. 103-108, 1976
LEITH, D,, AND W. LICHT
THE COLLECTION EFFICIENCY OF CYCLONE TYPE PARTICLE COLLECTORS
A NEH THEORETICAL APPROACH
A.I.CH.E. SYMPOSIUM SERIES, V, 69, PP*. 196-206, 1971
LESCHONSKI, W. A,, AND R, KOGLIN
PARTICLE SIZE ANALYSIS, SEPARATION METHODS
a6 (195, ppt 831-824. (GERMAN), 197«
244
-------
ROYCO INSTRUMENTS PARTICLE COUNTERS! CAPABILITIES AND
LIMITATIOMS
AF.ROSOL MEASUREMENT WORKSHOP, GAINESVILLE, 1976. P. 1-6
LIPPMANN. J.
REVIEW OF CASCADE IMPACTORS FOR PARTICLE SIZE ANALYSIS AND NEW
CALIBRATION FOR THE CASELLA CASCADE JMPACTQR
IND, HYG. J., PP. «03««16. 1959
LIPPMANN, M.
SIZE-SELECTIVE SAMPLING FOR INHALATION HAZARD EVALUATIONS
PROCEEDINGSI SYMPOSIUM ON FINE PARTICLES, MINNEAPOLIS, MINN.
1975, PP. 267-310
LIPPMANN, M., AND T, L. CHAN
CALIBRATION OF DUAL-INLET CYCLONES FOR 'RESPIRABLE' MASS
SAMPLING
AMER. IND. HYG. ASSOC. J., PP. 189-206. 1974
LIU, 8. Y. H.
LABORATORY GENERATION OF PARTICIPATES WITH EMPHASIS ON SljRMTCRON
AEROSOLS
J. OF APCA, 2ti (12), PP*. 1173-11T2, 197«
LIU, 8. Y. H,, AND A, VERMA
A PULSE CHARGING, PULSE PRECIPITATING ELECTROSTATIC
AEROSOL SAMPLER
ANAL. CH£M. flO'(«), PP, 8aO-8«7, 1962
LIU, 8. Y. H., AMD D. Y'. H. PUT
ON THE PERFORMANCE OF THE ELECTRICAL AEROSOL ANALVZER
J. AERO. SCI.. 6* PPt 2«9-26«, 1975
LIU, B. Y. H.. K. T. WHITBY, AND D. Y'. H, PUI
A PORTABLE ELECTRICAL ANALYZER POR SIZE DISTRIBUTION MEASUREMENT
of SUB-MICRON AEROSOLS
J. OF APCA, a« fll), PP'. 1067-1072. 1974
LIU, 8. Y. H., R, N. BERGLUND, AND J. K. AGARWAL
EXPERIMENTAL STUDIES OF OPTICAL PARTICLE COUNTERS
ATMOS, ENVIR. 8. PP. 717-732, 197«
Liu* B. Y. H., v. A. MARPLE, K. T*. WHITBY, AND N. j, BARSIC
SIZE DISTRIBUTION MEASUREMENT OF AIRBORNE COAL OUST BY
OPTICAL PARTICLE COUNTERS
AMER. IND. HYG, ASSOC. J.. PP. «
-------
LOCHART. I.. AND P. L. PATTERSON, JR.
FIITFR PACK TECHNIQUE CLASSIFYING RADIOACTIVE AEROSOLS
BY PARTICLE SIZE
U.S. NAVAL RES. LAB., WASHINGTON, p'. C., 11 PP.. 1963
LOFFLER, F.
THE CALCULATION OF CENTRIFUGAL SEPARATORS
STAUB REINHALTUN6 OIR LUFT, IN ENGLISH, 30 (1?). PP. 105
1970
LOO, B. «., j*. M. JAKLEVIC, AND F. A, GOULOING
DICHOTOMOUS VIRTUAL IMPACTORS FOR LARGE SCALE MONITORING OF
AIRBORNE PARTICULAR MATTER
.PROCEEDINGS! SYMPOSIUM ON FINE PARTICLES, MINNEAPOLIS, MINN.
1975, PP. 311-350
LUDWIG, F. L.
BEHAVIOR OF A NUMERICAL ANALOG TO A CASCADE JMPACTQR
ENVIRON. SCI. g, TECH., 2 m, PP. 5«7*550, 1968
LUNA, R.
A STUDY OF IMPINGING AXI-SY^METRIC JETS AND THEIR
APPLICATIONS
DISSERTATION, PRINCETON UNIV., UNIV. MICROFILM, HIGH
WYCOMB, ENGLAND, 117 PP.
LUNDGREN, n. A.
AN AEROSOL SAMPLER FOR DETERMINATION OF PARTICLE CONCENTRATION
AS A FUNCTION OF SIZE AND TIME
J. OF APCA, 17 (4), PP. 225-559, 1967
LUNDGREN, 0. A,, AND A. R. MCFARLAND
APPLICATION OF A LIGHT-SCATTERING AEROSOL COUNTER AND A FOUR.
STAGE IMPACTOR TO INDUSTRIAL HYGIENE AIR SAMPLING
AM£R. IND. HYG. ASSOC'., 32, PP. 35. «2. 1971
MACWILLIAM G. L.. F, GUTIERREZ, AND A'. S. LEE
INVESTIGATION OF A PERFUSION IMPACTION DUST SEPARATOR
OAK RIDGE NATIONAL LABORATORY, (15 PP» 1975
j. P., AND G, MADELAINE
NEW METHODS FOR AEROSOL SIZE DISTRIBUTION DETERMINATION WITH
A DIFFUSION BATTERY
WATER AIR SOIL POLLUT., 3 C«), PP*. 527-535, 197«
246
-------
,1. P., P. Y. TURPTN, G. MODELAINE. AND J, BRXCAPO
NOUVELLF METHOOE DE DETERMINATION DE LA GRANULOMETRIF. D'UM
AEROSOL Ad MOYEN D'UNF BATTERIE DE DIFFUSION
AFROSOI, SCI. (FRENCH) 5, PP. 339*355, 1974
MALLOVE, E. F., AMD *'„ C. HINDS
AEROSOL MEASUREMENT BY COMBINED LIGHT SCATTERING AND
CENTRIFUGATION
j. AEROSOL sci., 7, PP,
MALTON1, G. G.* C. MELANORI, V. PROpI. 6, TARRONI,
A. DEZAlACOMn, G, F, 80MPANE, AND M. FORMIGNANI
AM IMPROVED PARALLEL PLATE MOBILITY ANALYZER FOR AEROSOL
PARTICLES
AEROSOL SCI., «. PP. ««7-«55, 1973
W. H., AMP R, l. TANNER
DIFFUSION SAMPLING METHOD FOR AMBIENT AEROSOL SIZE
TION WITH CHEMICAL COMPOSITION DETERMINATION
ANAL. CHEM. qg (13), PP". 1999*2001, 1976
MARLOW, W. H., P, C, REIST, AND G. A,
ASPECTS OF THE PERFORMANCE OF THE ELECTRICAL AEROSOL ANALYZER
UNDER NONIDEAL CONDITIONS
J. AEROSOL SCI.. 7, PP, 457*462. 1976
MARPCE, V. A.
THE AERODYNAMIC SIZE CALIBRATION OF OPTICAL PARTICLE COUNTERS
BY INERTIAL IMPACTORS
PARTICLE TECH. LAB. PUB*. #306* PRESENTED AT AEROSOL MEASURE*
MENT WORKSHOP, u. OF FLA, GAINESVILLE. 1976, is PP,
MARPLE, V. A.
A FUNDAMENTAL STUDY OF TNERTIAL IMPACTORS »
DISSERTATION, UNIV. OF MINN,, UNIVERSITY MICROFILMS, HIGH
WYCOMB, ENGLAND, 1970. 2«3 PP.
MARPtE, V. A., AND 8, Y'. H. LlU
CHARACTERISTICS OF LAMINAR JET IMPACTQRS
ENVIRON. SCI. & TECH.. 9 (7), PP, 6
-------
MAKPLE, V. A.. B. Y, H. LIU, AND K. ?'. WHITBY
FLUID MECHANICS OF THE LAMINAR FLOW AEROSOL IMPACTOR
AEROSOL SCI.. 5, PP. 1-16, 197/i
MARPLE, V. A.. N. J. BARSIC, AND X, T'. WHITBY
INSTRUMENTS AND TECHNIQUES FOR DYNAMIC PARTICLE SIZE
OF COAL DUST | FINAL REPORT
UNIV. OF MINN., DEPT. OF INTERIOR, WASHINGTON, D. C,,
J974, 131 PP.
MARTENS, A, E.
ERRORS IN MEASUREMENT AND COUNTING OF PARTICLES USINg LIGHT
SCATTERING
J. OF APCA, 15 (10), PP'. 661-663,
MARTENS, A. E.» AND D'. D, DOOMAN
COMMENTS ON| INFLUENCE OF REFRACTIVE INDEX ON THE ACCURACY OF
SIZE DETERMINATION OF AEROSOL PARTICLES WITH LIGHT-SCATTERING'. .
APPL. OPT. 9 (8), PP. 1930-1938, 1970
MATTESON, M. J., G. F'. BOSCOE, AND 0. PREINING
DESIGN THEORY AND CALIBRATION OF A FIELD TYPE AEROSOL
SPECTROMETER
AEROSOL sci'., 5, PP. 71-79, 197*1
MATTHEWS, 8. J.
DEVELOPMENT OF LASER INSTRUMENTATION FOR PARTICLE MEASUREMENT
TRW SYSTEMS GROUP, EPA, RESEARCH TRIANGLE PARK, N.C., 1971,
72 PP.
APTD*087Q Pfi ?OS 189
MATTHEWS, B. J., AND C. W. LEAR
APPLICATION OF HOLOGRAPHIC METHODS TO THE MEASUREMENT OF
FLAMES AND PARTICULATE, VOLUME II
TRW SYSTEMS GROUP, EPA, WASHINGTON, D*. C,. 1974. 123 PP.
EPA-650/2-74.031B PR 335 675
MATTHEWS, R. J., AND Rt F. KEMP
HOLOGRAPHY OF LIGHT SCATTERED BY PARTICULATE IN A LARGE
STEAM BOILER
63Rn ANNUAL MEETING, AICHE, SYMPOSIUM^ CONTINUOUS PARTICULATE
MONITORING, NOV. « DEC. 1973
MAY, K. R.
AEROSOL IMPACTOR JETS
J. OF AEROSOL SCI., 6, PP. 403-411* 1975
248
-------
MAY. K. R.
THE CASCADE JMPACTORj AN INSTRUMENT FOR SAMPLING COARSF
AfROSOLS
J. OF SCI. INSTRtf- S3, PP. 187- IPS, l
-------
MERCER. T. T.
ON THE CALIBRATION OF CASCADE
ANN. OCCUP. HYG. 6, PP. i-i7f 1963
MFRCF.R, T, T.
TH£ INTERPRETATION! OF CASCADE IMPACTQR DATA
INOUSTR. HYG. J., 36, PP. £36-2«l, 1
-------
DESIGN OF CYCLONE SEPARATORS IN TNf ENGINEERING PRACTICE
STAUR REINHALTUNG DE» LUFT, IN ENGLISH, 30 (5), PP. 1-12.
1970
MYERS, P. I., T, H. 8ARLAK, AND W". I . FITE
AN AUTOMATIC, REAL-TIME DETECTOR AND 8IZER FOR SUBMICRON
AIRBORNE PARTICIPATE MATTER
SUBMITTED TO RSI, 2« PP'.f 1975
NATUSCH, D. 6. S,, AND J. R. WALLACE
DETERMINATION OF AIRBORNE PARTICLE SHE DISTRIBUTIONS I CALCULA.
TION OF CROSS-SENSITIVITY AND DISCRETENESS EFFECTS IN CASCADE
IMPACTION
ATMOS. ENVIRON.. 10, PP. 315. 3ga, 1976
NEWTON, G. J., 0, G, RAABE, AND B'. V, MOKLER
CASCADE IMPACTOR DESIGN AND PERFORMANCE
j. AEROSOL sci., e, PP. 339-s«7. 1977
NOL.L, K. E.
A ROTARY INERTIAL IMPACTOR FOR SAMPLING GIANT PARTICLES IN THE
ATMOSPHERE
ATMOS. ENVIRON., «, PP, 9*19, 1970
NOLL, K. E.. AND M. J. PILAT
INERTIAL IMPACTION OF PARTICLES UPON RECTANGULAR BODIES
J. OF COLL. AND INTER. SCI., 33 C2), PP. 197-207, 1970
PARKER, G. w., AND H, BUCHHOLZ
SIZE CLASSIFICATION OF SUBMICRON PARTICLES BY A IOW«PR£SSURE
CASCADE IMPACTOR
ORNL-4326, 6« PP., 1968
PARKER, R.
CALIBRATION OF FINE PARTICIPATE SIZING DEVICES
AIR POLLUTION TECHNOLOGY, INC., 1976
EPA-600/2-76-116 PR 352 656
J. H*., AND R. H, PERRY
ENGINEERING MANUAL! A PRACTICAL REFERENCE OF DATA & METHODS IN
ARCH., CHEM., CIVIL, ELEC.,MECH.f & NUCLEAR ENGINEERING
MCGRAW-HILL ROOK CO., INC., NEW YORK, N.Y'., PP. 60-63,
PICH, J.
A NOTE ON THE DIFFUSIVE DEPOSITION OF AEROSOLS ON A
CYLINDER
AEROSOL SCI., 1, PP. 17-19, 1970
251
-------
PTCKNETT, R. G.
A MEW METHOD OF DETERMINING AEROSOL SIZE DISTRIBUTIONS EROM
MULTISTAGE SAMPLER DATA
AEROSOL SCI.f 3, PP. 185-198. 197?
PILAT, M. j'.
SU8MJCRON PARTICLE SAMPLING WITH CASCADE IMPACTOR
fefeTM ANNUAL MEETING, APCA, CHICAGO, ILL. 1973. PAPER 73-28U
PILAT, M. j;f 0. s. ENSOR. AND j. c'. BOSCH
CASCADE IMPACTOR FOR SIZING PARTICIPATES IN EMISSION SOURCES
AMER. TWO. MYG. ASSOC. J., 3£ (8). pp'. 508-511, 1971
PILAT, M. J., |v. S. ENSOR, AND J. c'. BOSCH
SOURCE TEST CASCADE IMPACTOR
ATMOS. ENVIRON.. «, PR, 671-679. 1970
P1LAT, M. J., G. Mt FIORETTI, AND E. ft. POWELL
SIZING OF 0.02-20 MICRON DIAMETER PARTICLES EMITTED FROM COAL
FIRED POWER BOILER WITH CASCADE IMPACTORS
PAPER PRESENTED APCA-PNWIS MEETING, VANCOUVER. B. C., 1975
j. M., R. i. MITCHELL, AND R'. E. THOMAS
THE CASCADE IMPACTOR FOR PARTICLE-SIZE ANALYSIS OF AEROSOLS
42MD ANNUAL MEETING CHEM. SPEC. MANUF! ASSOC., INC., NEW YORK
N. V,, 1955
PQLLAK, L. W.. T, C. n*cONNOR, AND A. L, METNIEKS
ON THE DETERMINATION OF THE DIFFUSION COEFFICIENT OF
HETEROGENEOUS AEROSOLS BY THE DYNAMIC METHOD
GEOFISICA PURA E APPLICATA, 3«, PP. 183-190, 1975
H.
INFLUENCE OF REFRACTIVE INDEX ON THE ACCURACY OF SIZE DETERMINA-
TION OF AEROSOL PARTICLES *ITH LIGHT-SCATTERING AEROSOL COUNTERS
APPLIED OPTICS. 8 (1). PP. 165-169. 1969
RAGLAND, J. w'., K; M, CUSHING. J, D'. MCCAIN, AND W. B. SMITH
HP-25 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUDIESJ STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANQLI PARK.
N. C., 1977. 127 PP.
EPA-600/7-77-058
RAGI.AND, J. W., K. M. CUSHING, J, 0. MCCAIN, AND *, B. SMITH
HP- 65 PROGRAMMABLE POCKET CALCULATOR APPLIED TO AIR POLLUTION
MEASUREMENT STUOXESI -.STATIONARY SOURCES
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK.
N. C., 1976*. 122 PP.
EPA-600/8-76-002
252
-------
RANZ, n. F.
PRINCIPLES OF JNERTTAL IMPACTJONi A SURVEY OF INFORMATION
APPLICABLE TO AN ANALYSIS OF MIST AND OUST COLLECTORS
U.S. PUBLIC HEALTH SERVICE, «5 PP., 1956
RANZ, w. F.. AND J. B. WONG
IMPACTION OF DUST AND SMOKE PARTICLES ON SURFACE AND BODY
COLLECTORS
INO. AND EMC. CHEM,, -«4 (6), Pp; 1371.1381, 195?
w, £,, AND J, 8. WONG
JET IMPACTORS FOR DETERMINING THE PARTICLE-SIZE DISTRIBUTIONS
OF AEROSOLS
IND. HYG. & OCCUP. MED.. PP. 46«-477, 1953
RAO, A. K.
AM EXPERIMENTAL STUDY OF IWERTJAL IMPACTORS
DISSERTATION, UNIV, OF MINNESOTA, 197«5, 194 PP.
RAO, A, K., AND K. T,
NONIDEAL COLLECTION CHARACTERISTICS OF SINGLE STAGE AND
CASCADE IMPACTpRS
AMER. IND. HYG. ASSOCl J« , 38, Pp'. 17««I79, 19?7
RATH, R., AND D. POHL
PARTICLE SIZE DETERMINATION WITH THE PARTICLE COUNTER
8TAUB REINHALTUNG DER LUFT, (ENGLISH*. 34 CO),
PP. 10-8-114, !97«
ROBTRDS, D. W., W. M. FARMER, AND A*. E. LENNERT
INTERFEROMETRIC INSTRUMENTATION FOR PARTICLE SIZE ANALYSIS
ARNOLD RESEARCH ORGANIZATION, INC., EPA. WASHINGTON, o. c.
1974, 56 PP.
EPA«650/2-73*03« PR 240 584
SCHOTT, J. H., AND W. E'.
JET-CONE IMPACTORS AS AEROSOL PARTICLE SEPARATORS
J. OF ENVIRON. SCI. & TECH., 10 (13), RP, 1250-1256, 1976
SCHUSTER, B., AND R. KNOLLENBERG
DETECTION AND SIZING OF SMALL PARTICLES IN AN OPEN CAVITY
SAS LASER
APPL. OPT., 11 (7), PP. 1515-1520, 1972
8, A.
AN EVALUATION OF A .HIGH-VOLUME pASCADE PARTICLE IMpAeTOR SYSTEM
2ND JOINT CONF. SENSING ENVIRON. POLLUT,, WASHINGTON. D.C.
PP. 109-115, 1973 . • • • » *.u.
253
-------
. G. J.
STATE OF THF ART* 1971 INSTRUMENTATION FOR MEASUREMENT OF
PARTICIPATE EMISSIONS FROM COMBUSTION SOURCES. VOLUME ni
THERMO. SYSTEMS, INC., EPA, RESEARCH TRIANGLE PARK, N. c.
1972, 850/2-7«-102 PB 2aO 670
SMITH, w. B., K, M, CUSHING, AND J, D'. MCCAIN
PROCEDURES MANUAL FOR ELECTROSTATIC PRECIPITATOR EVALUATION
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK,
N. C., 1977. 430 PP.
EPA«feOO/7"77«059
254
-------
SONKJN, t. S.
A MODIFIED CASCADE I^PACTORi A DPVJC'E FOR SAMPLING AND SIZING
AEROSOLS OF PARTICLES BF.LOW ONE MTCRHN IN DIAMETER
J. OF IND. HYR. & TOXICOLOGY* 2fl (6), PP. 269»;?72,
SODLE, B. w.
CONCERNING THE CALIBRATION CONSTANTS OF CASCADE IMPACTS.
SPECIAL REFERENCE TO THE CASEILA MK'. 2
AEROSOL SCI., ?. PP. 1-ta. 1071
SPERTELL, R. «„, AND H»
AIRBORNE DENSITY OF FERRIC OXIDE AGGREGATE HICROSPHERES
INC. HVB. ASSOC. J., PP. 73^.740, i«?71
8PROULL. K. T.
EFFECT OF OUST CONCENTRATION UPON THE CASHFLOW CAPACITY
OF A CYCLONIC COLLECTOR
J. OF APCA, 16 (8), PP'. 439»44t*
STAIRMAND, C. J.
THE DESIGN AND PERFORMANCE OF CYCLONE SEPARATORS
INST, OF CHEM. ENG,, 39, PP. 356«-383, 1951
STAIRMAND, C. J.
PRESSURE DROP IN CYCLONE SEPARATORS
ENGINEERING, 168, PP.
STF.NHOUSE. J. 1. T., AND P. J. LLOYD
SAMPLING ERRORS DUE TO INERTIAL CLASSIFICATION
65TH ANNUAL MEETING AICHE, N^W YORK, N. Y., 19?a, PAPER 30
STERN, S, C'., H. W. ZELLER, AND A, I. SCHEKMAN
COLLECTION EFFICIENCY OF JET IMPACTORS AT REDUCED PRESSURES
UEC FUNDAMENTALS* 1 <«> » PP. ' Zt5«S«4. 1962
SUNDELOF, L. 0.
ON THE ACCURATE CALCULATION OF PARTICLE'DISTRIBUTIONS IN
AEROSOLS FROM IMPACHON DATA
STAUR REINHALTUNG DEP LUF'T, IN ENGLISH, 27 (8),
PP. 22«?8, 1967
SVAROVSKY, L., AND J, SVAROVSKA
A NEW ANALOGUE DATA ANALYSER FOR PARTICLE SIZE DISTRIBUTION
MEASUREMENTS IN A DISC CENTRIFUGE
J. PHYSICS E» SCI. INSTR. 9, PP. 959-962* 1976
255
-------
S*APTZ, 0. R., M, B. OENTON AMH J. L. MOVERS
ON CALIBRATING OF CASCADE IMPACTORS
AMER. TWO. HVG ASSOC. j'.f PP. a29-«39. 1973
VAN OSDFLt , n. (COMPILER)
PROCEEDINGS! SEMINAR ON IN. STACK PARTICLE SIZING FOR PARTICIPATE
CONTROL DEVICE EVALUATION
RESEARCH TRIANGLE INST,. EPA, RESEARCH TRIANGLE PARK, N. C.
1975, PP. 135-142
EPA-6QO/2-77-060
K. T., AND 8. Y. H. LIU
GENERATION OF COUNTABLE PULSES BY HIGH CONCENTRATIONS OF SUB-
COUNTABLE SIZE PARTICLES IN THE SENSING VOLUME OF OPTICAL
COUNTERS
J*. OF COLL. AND INTER. SCI.. 25, PP. 537-546, 1967
K. T., AND R. A. VQMELA
RESPONSE OF SINGLE PARTICLE OPTICAL COUNTERS TO NONIDEAL
PARTICLES
ENVIRON. SCI. & TECH.. 1 (10). PP'. 801-614, 1967
WHITBY, K'. T., AND W. E. CLARK
ELECTRIC AEROSOL PARTICLE COUNTING AND SIZE DISTRIBUTION
MEASURING SYSTEM FOR THE o.ois TO i MICRON SIZE RANGE
TELLUS, 10, PP, S73«586, 1966
WILLEKE, K.
PERFORMANCE OF THE SLOTTED IMPACTOR
1<5TH AMER^ IND. HYG. CONF., MINNEAPOLIS. MINN., PARTICLE
TECH. LAB. PUB. 2*10, 32 PP,, 1965
K.
PERFORMANCE OF THf SLOTTED IMPACTOR
AMER, INO. HYG, ASSOC. J,, 16 (9), 683-691, 1975
WILLEKE, K., AND B, Y. H, LIU
SINGLE PARTICLE OPTICAL COUNTERl PRINCIPLE AND APPLICATION
1975SYMP. ON FINE PARTICLES, MINNEAPOLIS, MAY 28*30, 1975
FINE PARTICLES! AEROSOL GENERATION, MEASUREMENT, SAMPLING
AND ANALYSIS, PP, 698*729, 1976
K,. AND j, j. MCFETERS
THE INFLUENCE OF FLOW ENTRY AND COLLECTING SURFACE ON THE
IMPACTION EFFICIENCY OF JNERTIAL IMPACTORS
J. OF COLL. AMD INTER. sci,» 53 CD. PP. 121*127, 1975
256
-------
WILLIAMS, 1., AND A. R, MEDLEY
THE CHOICE, DESIGN AND PERFORMANCE OF A MULT ICHANNF,L AEROSOL
PARTICLE COUNTER
AEROSOL sci.. 3. PP. 363-375, 597?
WINKLER, P.
RELATIVE HUMIDITY AND ADHESION OF ATMOSPHERIC PARTICLES TO
PLATES OF IMPACTORS
AEROSOL SCI., 5. PP. 235«2ao, 197*
WITTEf A. R., AND 0. E, HAFLINSER
APPLICATION OF HOLOGRAPHIC METHODS TO THE MEASUREMENT OF
FLAMES AND PARTICULATE, VOL. i
TRW SYSTEMS GROUP, EPA, RESEARCH TRIANGLE PARK, N. C,
197«, 69 PP.
EPA.650/2*7«»03«A PB 536 580
YAMAMOTO, 6., AND M. TANAKA
DETERMINATION OF AEROSOL SIZE DISTRIBUTION FROM SPECTRAL
ATTENUATION MEASUREMENTS
APPL. OPT. fl C2)i PP. «a7«a53»
YUU, S«, AND K. 1 1 NO Y A
SEPARATION MECHANISM OF ROUND»NOZZLI CASCADE IMPACTOR«EFFECT
OF CLEARANCE
KAGOKU K06AKU (IN JAPANESE), 3«. PP*. «27*«12, 1970
ZEBEL, (3., AND D. HOCHRAINER
MEASUREMENT OF SIZE DISTRIBUTION OF FINE DUST WITH THE AID OF
AN IMPROVED SPECTRAL IMPACTOR
STAUB REINHALTUN6 OER LUFT, IN ENGLISH. 32 (3)
PP. 6-H, 1972
ZINKY» W. R.
A NEW TOOL FOR AIR POLLUTION CONTROL* THE AEROSOL PARTICLE
COUNTER
J. OF APGA, 12 C12), PP. S78-S80, 1962
6. OPACITY
AVETTA, E. D.
IN-STACK TRANSMISSOMETER EVALUATION AND APPLICATION TO
PARTICULATE OPACITY MEASUREMENT
P§ 243 402
257
-------
BEUTNER, H, P.
MEASUREMENT OF OPACITY AND PARTICIPATE EMISSIONS WJTH AN
QN-STACK TRAN8MISSOMETER
J. 'OF APCA, ?>J (9), PP, 865-871. 1974
COLLINS, K. £., AMD 0. J. STEELE
HIGH. SENSITIVITY RECORDING OPTICAL DENSITY METER
J. SCI. INSTR'., 38, PP, 186»190, 1961
«. n.
OF THE OPACITY AND MASS CONCENTRATION OF
PARTICIPATE EMISSIONS 8Y TRANSMISSOMETRY
EPA. RESEARCH TRIANGLE PARK, N. c., 1974. 39 PP.
EPA-650/2-74-128 PB 241 251
DOBBINS, R. A.. AND G". S. JIZMAGIAN
OPTICAL SCATTERING CROSS SECTIONS FOR POlYDlSPERSIONS OF
DIELECTRIC SPHERES
J. OPT. SOC. AMER., 56 (135, PP. 1345-1350, 1966
ENSOR, D. S.
PLUME OPACITY MEASUREMENT
U.S.-U.S.S,R. WORKING GRP, STATIONARY SOURCE AIR POLLUTION
CONTROL TECH. SYMP, SAN FRANCISCO, 1974, MR17«»PA«1H2
ENSOR, D. S.. AND M. J,
CALCULATION OF SMOKE f»LUME OPACITY FROM PARTICULATE AIR
POLLUTANT PROPERTIES
Jr. OF APCA, 21 (8), PP, 496»50l. 1971
ENSOR, D, S*., AND M. J,
THE EFFECT OF PARTICLE SIZE DISTRIBUTION ON LIGHT TRANSMlTTANCF.
MEASUREMENT
AMER. i NO'. HYG. ASSOC. J,, 32, PP. 287-292, 1971
ENSOR, D. Si. L. D, SEVAN, AND 6, MARKQW8KI
APPLICATION OF NEPHELOMfTTRY TO THE MONITORING OF AIR POLLUTION
SOURCES
67TH ANNUAL MEETING, APCA, DENVER, COLO., 1970, PAPER 7«-iio
ENVIRONMENTAL PROTECTION AGENCY
EPA REQUIREMENTS FOR SUBMITTAL OF IMPLEMENTATION PLANS,
STANDARDS FOR NEW STATIONARY SOURCES
FEDERAL REGISTER, «0 (194), PP. 46240»46271. 197S
258
-------
HERMANN, J,( AMD H. J.
THE INFLUENCE OF ' PARTICLE' SIZE IN EXTINCTION MEASUREMENTS
STAUB REINHALTUNG OER LU^T, TN ENGLISH. 34, PP, 123-129
1974
HINKLEY, E. o.f R. T..KU, K. w. NILL, AND j. r, BUTLER
LONG-PATH MONJTORINGI ADVANCED INSTRUMENTATION WITH A
TUNABLE DIODE LASER
APPL. OPT..J5 (7). PP'. 1653»J.655»
HOOD, K. T.
OPACITY AND PARTICULATE EMISSION RELATIONSHIPS FOR PULP MILLS
NATIONAL COLINC* OF THE PAPER IND, FOR AIR AND STREAM
IMPROVEMENT. INC, i 1976
KRAUSt F. J.
THE INFLUENCE OF FORWARD SCATTERING ON MEASUREMENTS OF THE
DEGREE OF TRANSMISSION OF AEROSOLS
STAUB REINHALTUNG DERLUFT, IN ENGLISH, 33 (9), PP, 3«l-345,
1973
LAMBIE, R.
IMPROVED SMOKE DENSITY RECORDER
J. OF SCI*. INSTR. 37, PP. 174M4.6/ I960
LARSEN. S., D. S, ENSOR, AND M, j', PILAT
RELATIONSHIP OF PLUME OPACITY TO THE PROPERTIES OF PARTICULATES
EMITTED FPOM KRAFT RECOVERY FURNACES
TAPPt, 55 (1), PP. 88-92, 1972
LITTLEWOOD, A.
MEASUREMENT OF THE OPTICAL DENSITY OF SMOKE IN A CHIMNEY
J. OF SCl'. INSTR. 33, PP. a95"499. 1956
MASON, R. D.
EVALUATING PARTICULATF EMISSIONS PROBLEMS THROUGH OPACITY
MONITORING
LEAR SIEGLER, INC., 20 PP., 1970
MCRANIE, R. D.
EVALUATION OF SAMPLE CONDITIONERS & CONTINUOUS STACK MONjTORS
FOR MEASUREMENT OF SULFUR DIOXIDE, NITROGEN OXIDES AND OPACITY
SOUTHERN COMPANY SERVICES. INC., 259 PP.* 1975
NADER, J. 8,
CURRENT TECHNOLOGY FOR CONTINUOUS MONITORING OF PARTICULATE
EMISSIONS
J. OF APCA, 25 (81, PP, 814-821. 1975
259
-------
NADER, J. S., F. JAYE. AND W. CONNER
PERFORMANCE SPECIFICATIONS FOR STATIONARY SOURCE MONITORING
SYSTEMS FOR GASES AND VISIBLE EMISSIONS
NATIONAL ENVIRON. RES. CTR, EPA. RESEARCH TRIANGLE PARK, N.£.
1974. 73 PP.
EPA.650/2-74-013 PB 230 934
PILAT, M. j.
PLUME OPACITY RELATED TO PARTICLE PROPERTIES
APCA-PNWIS MEETING, NOV>
PILAT, M, J. f AND D. S, ENSOR
PLUME OPACITY AND PARTICIPATE M*SS CONCENTRATION
ATMOS ENVIRON., 4, PP. !63*173, 1970
PEISMAN. E., W. D. SER8ER, AND N. D*. POTTER
IN. STACK TRANSMISSOMETER MEASUREMENT OF PARTICULATE OPACITY
AND MASS CONCENTRATIONS
PHILCO-FORD CORP., EPA, RESEARCH TRIANGLE PARK, N.C.,
US PP.
7fl»120 PB 239
SCHNEIDER, W. A.
OPACITY MONITORING OF STACK EMISSIONS! A DESIGN TOOL
WITH PROMISING RESULTS
GENERATION PLANBOOK, 3 PP., 197«
VINCENT, J. M.
EVALUATION OF A LIGHT TRANSMISSION TECHNIQUE FOR TESTING A
TWO-STAGE ELECTROSTATIC DUST PRECIPITATOR
J. OF PHY. Df APPL. PHYS. 4, PP. lft3S»i8«l, 1971
WOLF, P. C.
CONTINUOUS ACROSS-THE-STACK MEASUREMENT
ENVIRON. SCI. & TECH., 9 C3), PP. 221-225, 1975
7. ANALYTICAL TECHNIQUES
CAHILL, T. A.
CASCADE JMPACTQR DATA FOR ELEMENTAL ANALYSIS
SEMINAR-IN.STACK PARTICLE SIZING FOR PARTICULATE CONTROL
DEVICE IVAL., EPA, RESEARCH TRIANGLE PARK, N,C. 197B, PP.
260
-------
CAHILL, T. A., L. L. ASHRAUGH, J. B. BARONF, R. A.
p. J. FEENEY, AND G. W, WOLFF
ANALYSIS OF RFSPIRAHLE FRACTIONS IN ATMOSPHERIC PARTICIPATES
VIA SEQUENTIAL FILTRATION
J. OF APCA, 11 (7), PPf 675-678. 1977
HULETT, L. D., J. M. DALE, J, F. EMERY, W*. S, LYON, JR., AND
W
TF-CHNIQUES FOR CHARACTERIZATION! OF PARTICULAR MATTER! NEUTRON
ACTIVATION ANALYSTS, X-RAY PMOTOELECTRON 8PECTROSCOPY, SCANNING
ELECTRON MICROSCOPY
WORKSHOP. SAMPLING, ANALYSIS, AND MONITORING or STACK
EMISSIONS, fPRl SR-ttt. DALLAS, TEXAS, 1975. PP. 241-256
R. B., D. W, NEUENDORF, AND K. J. YOST
TRACE METAL SAMPLES COLLECTED IN THI FRONT AND BACK HALVES
OF THE EPA STACK SAMPLING TRAIN
J. OF APCA, 25 (10), PP'. 1058MOS9, 1975
JAKLEVIC, J. M., AND R. L. WALTER
COMPARISON OF MINIMUM DETECTABLE LIMITS AMONG X.RAY SPECTROMETERS
X-RAY FLUORESC. ANAL. ENVIRON. SAMPLES, PP. 68*75, 1977
JENSEN, B., AND J. w. NELSON
NOVEL AEROSOL SAMPLING APPARATUS FOR ELEMENTAL ANALYSIS
PROC. 2ND INTERNS CONF* "NUCLEAR METHODS IN ENVIRONMENTAL
RESEARCH", U. OF MISSOURI-COLUMBIA, 1974
MAHAR, H,
EVALUATION OF SELECTED METHODS FOR CHEMICAL AND BIOLOGICAL
TESTING OF INDUSTRIAL PARTICIPATE EMISSIONS
MITRE CORP., EPA, RESEARCH TRIANGLE PARK, N.C., 1975, sa PP.
EPA-600/2-76-137 PR 257 912
ROBERTS, N. J.
AEROSOL TRACE ELEMENT ANALYSIS USING NEUTRON ACTIVATION AND
X-RAY FLUORESCENCE
LAWRENCE LIVERMORE LAB., U.S. AEC, 135 PP., 197«
SOWINSKI, E'. J., AND I. H. SUFFETT
AN EXPERIMENTAL LABORATORY SYSTEM TO EVALUATE THE INDUSTRIAL
FATE OF TOXIC VOLATILE INORGANIC HYDRIDES
AMER. IND. HYG. ASSOC'. J., 38, PP*. 351-357, 1977
261
-------
fl, CONTROL DEVICE EVALUATION.FIELD TESTS
BOSCH, J, C.f M. J. PILAT, AMD B, F. HRUTFIORO
SIZF DISTRIBUTION OF AEROSOLS FROM A KRAFT Mitt, RECOVERY
FURNACE
TAPPI, 5H HI), PP. 1871-1875, 1971
BRADWAY, R, M,, AND R. W, CASS
FRACTIONAL EFFICIENCY OF A UTILITY BOILER BAGHOUSEl
NUCLA GENERATING PLANT
GCA CORP, EPA. RESEARCH TRIANGLE PARK* N, C., 1975, 148 PP,
75«031A P8
BRADY, J. D.^ F. N. HILL, AND K, M, GRAVES
USE OF INFRTIAL IMPACTOR DATA TO SELECT AIR POLLUTION CONTROL
EQUIPMENT
70TH ANNUAL MEETING, APCA, TORONTO, 1977, PAPER 77-42.4
BROOKS, D. A.
MEASUREMENT AND CHARACTERIZATION OF PARTICLES IN WET SCRUBBING
PROCESS FOR SOX CONTROL
TRW SYSTEMS GROUP, IPA, WASHINGTON, -pi C.. 1975, 125 PP.
EPA«650/2-73-024 PB
BROWN, R. F.
PARTICULATE COLLECTION STUDY, EPA/TVA FULL-SCALE DRY LIMESTONE
INJECTION TESTS
COTTRELL ENVIRON, SYSTEMS, INC., EPA, RESEARCH TRIANGLE PARK
N.C., 1974, 197 PP. .
EPA-65Q/2-74.053 PB 260 586
BURNS, E. A.
INSTRUMENTAL ANALYSES FOR WET SCRUBBING PROCESSES
TRW SYSTEMS GROUP FOR NAT'L ENVIRON'. RES, CENTER, 1974
EPA-650/2-7U.064 PR 240 616
BYFRS, R. L.
PROCEEDINGS. SEMINAR! IN.STACK PARTICLE SIZING FOR PARTICULATf
CONTROL DEVICE EVALUATION
EPA, RESEARCH TRIANGLE PARK, N. c'., 1975, PP, 135-147
EPA.600/2-77i»060
262
-------
C.ALVFRT, s., AND s, YUNG
EVALUATION OF ELECTROSTATIC SCRUBBER
A.P.T., INC., EPA. RESEARCH TRIANGLE PARK. N.C.. 1975
S,, C. JHAVERI. AND S. YUNG
FINE PARTICLE SCRUBBER PERFORMANCE TESTS
A>.T., IMC.. EPA, RESEARCH TRIAN6LF PARK, N. C'., 1974,
269 PP.
EPA«650/2»74»093 P8 240 325
CARP, R., w, PIULLE, AND" j, 'P. GOOCH
FABRIC FILTER AND ELECTROSTATIC PPf CIPXTATORf FINE PARTICLE
EMISSION COMPARISON
ELECTRIC POW£R RESEARCH INST., AMERICAN PO^ER CONF,f
CHICAGO, ILL.. 1977, 39 PP.
CASS, R. W., AND J, V. LANGLEY
•FRACTIONAL EFFICIENCY OF A STEEL MjLU 8AGHOUSE
GCA CORP., EPA
EPA
CASS, R. W., AND R. M. 8RADWAY
FRACTIONAL EFFICIENCY OF A UTILITY BOILER BAGHOUSEt SUN8URY
STEAM-ELECTRIC STATION
CCA/TECH,, EPA, RESEARCH TRIANGLE PARK, N.C,, 1976, 2^*4 PP.
EPA-600/2-76-077A P8 253 945
COOPER, 0. W.
DYNACTOR SCRUBBER EVALUATION
GCA CORP. FOR NATIONAL ENVIRONMENTAL RESEARCH CENTER, 1975
116 PP.
PB 2«3 365
DENNIS, R., AND J. WILDER
FABRIC FILTER CLEANING STUDIES
GCA/TECH,, EPA, RESEARCH TRIANGLE PARK, N. c,, 1975, «3» PP.
EPA-650/2-75-009 PB 240 372
OISMUKES, E. G.
CONDITIONING OF FLY ASH WITH SULFUR TRfOXIOE AND AMMQNIA
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK, N.C
1975, 169 PP.
EPA«600/2-7S«OlS PB 247 231
ENSOR, D, S.
FIELD EXPERIENCE WITH CASCADE IMPACTORSl QUALITY CONTROL OF
TEST RESULTS
SEMINAR, IN-STACK PARTICLE SUING FOR FARTICULATE CONTROL
DEVICE FVAL., EPA. RESEARCH TRIANGLE PARK, N.C. 1975, PP. 118-lSfl
263
-------
EH50R, D. S., H. S, JACKSON, S. CA|,VERT, C, LAKF,
n. v. wALtON, R. E. NTLAN, *. s. CAMPBELL, AMD T. A. CAHRI
EVALUATION OF A PARTICULAR SCRUBBER ON A COAL-FIRED UTILITY
BOILER
METEROLtiGY RES, IMC., EPA, RESEARCH TRIANGLE PARK, N.C.
1075
EPA»600/2-75-074 PB 249 56?
ENSOR, D. s'.. H. G, HOOPER, AND R, W, SCHECK
DETERMINATION OF THE FRACT, EFFIC., OPACITY CHARACTERISTICS,
ENG. ft ECON. ASPECTS OF FABRIC FILTER OPERATING ON UTILITY BOILE'
METEOROLOGY RESEARCH, INC., EPRI, PALO ALTO, CALIF
1176, 2?0 PP.
N, G. J.. R. S. SERENIUS, AND A, D. MCINTYRE
MEASUREMENT AND CHARACTERIZATION OF RECOVERY FURNACE
PARTICULATES. A STATUS REPORT
PULP AND PAPER MAGAZINE OF CANADA, 74 (12), PP. 98«104, J07?
FPANCONERI, P'., AND L. KAPLAN
DETERMINATION AND EVALUATION OF STACK EMISSIONS FROM
MUNICIPAL INCINERATORS
J. OF APCA, 26 (9), PP. 8B7-888, 1976
ERASER, D. A.
THE COLLECTION OF SURMJCRON PARTICLES' BY' ELECTROSTATIC
PRECIPITATION
IND. HYG. QUARTE«LV, PP'. 75-79, 1956
HALL, R. R., AND R. DENNIS
MOBILE FABRIC FILTER SYSTEM DESIGN AND FIELD TEST RESULTS
CCA/TECHNOLOGY, EPA, RESEARCH TRIANGLE PARK,
N.C., 1975, 136 PP.
EPA-65G/2-75-059 PB 249 514
HESKETH, H. E.
AEROSOL CAPTURE EFFICIENCY IN SCRUBBERS
68TH ANNUAL MEETING, APCA, SOUTHERN ILLINOIS UNIVERSITY AT
CARBQNDALE, J975
JACKO, R. B.f D. W. NEUENDORF, AND F. VAURE
FACTIONAL COLLECTION EFEICIENCY OF ELECTROSTATIC PRECIPITATOR FOI
OPEM HEARTH FURNACE TRACE METAL EMISSIONS
ENVIRON. SCI. & TECH., 10 (105, PP, tQQ2»1005, 1976
264
-------
JAMGOCHIAN. E. M.. M. f. MILLER, AND R. REALE
ll*r,l^r«lION °F CAT"OX HIGH EFFICIENCY ELECTROSTATIC
r " t|. C I r J T A T 0 R
THE MITRE CORP., EPA, RESEARCH TRIANGLE PARK, N.C., 1*75
1 U £ f*r g
EPA-.600/2.75.037 PR 246 647
JOHNSON, L, D., AND R'. M. STATNICK
L1°UID LEVFLS IN ^FLUENT GASES FROM
CONTROL SYSTEMS LAB., EPA, 'RESEARCH TRIANGLE PARK. N C .
1974, 17 PP. * *
EPA-650/2-71.050 P8 233 7J9
KUTYNA, A, G,
COMPARISON OF SOURCE PARTICULATF EMISSION MEASUREMENT METHODS
FOP COMBINATION FUEL'FIRED BOILERS
TECHNICAL BULLETIN 75. NATIONAL COUNCIL OF THE PAPER INDUSTRY
FOR AIR AND STREAM IMPROVEMENT, INC., 1974, 35 PP. .1'WL5TRT
MCCAIN, j. D.
EVALUATION OF A REXNORD GRAVEL REP FILTER
HE!^ ^?EARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK. N.C
, 53 PP *
76*16« P§ 255 095
MCCAIN, J. D.
EVALUATION OF ARQNETICS TWO-PHASF JET SCRUBBER
SOUTHERN RESEARCH INSTITUTE, EPA, 1974. 43 PP'
P8 239
MCCAIN, J. p.
EVALUATION OF CENTRIFIED SCRUBBER
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK.
N.C., 1975
EPA-650/2-74.J29A PR ga3 626
MCCAIN, J. 0.. AND W. B'. SMITH
AIR CLEANING SYSTEM EVALUATION
;-,: **w*.. ^.^^
EPA«650/2-7«-028 PR 232 «36
MCCAIN, j. D., j. P. GOOCH, AND w. B, SMITH
RESULTS OF FIELD MEASUREMENTS OF INDUSTRIAL PARTICIPATE SOURCES
AND ELECTROSTATIC PRECIPHATOR PERFORMANCE
J. OF APCA, 25 (2), PP. 117-121. 1975
265
-------
MCKF.NNA, j,
APPLYING FABRIC FILTRATION TO COAL FIRED INDUSTRIAL ROTI FRS
A PILOT SCALE INVESTIGATION '
FNVIRO-SYSTEMS AND RESEARCH INC., EPA. 1975, 201 PP
MCKENNA, j. D.
APPLYING FABRIC FILTRATION TO COAL FIRED INDUSTRIAL BQJLfRS
A PRELIMINARY PILOT SCALE INVESTIGATION
ENVIRO-SYSTEMS AND RESEARCH INC., NERD, 1974, 90 PP.
EPA..65Q/2.7fl-058 PB 237 117
NICHOLS, G. B., AND J*. D. MCCAIN
PARTJCULATE COLLECTION EFFICIENCY MEASUREMENTS ON THREE
ELECTROSTATIC PRECIPIT ATORS
SOUTHERN RESEARCH INSTITUTE, EPA, 1975
EPA«6QQ/2»75«056 PR 248 220
PIt,AT, M. J.. AND F. MEYER
UNIV. OP WASH. ELECTROSTATIC SPRAY SCRUBBER EVALUATION
UNIVERSITY. OF WASHINGTON, EPA, RESEARCH TRIANGLE PARK, N'C.
1976, 74 PP.
EPA-60Q/2-76-100 P8 ESI 655
PILAT, M. j;f e. A, RAEMHILO. D. L, HARMON
TEST OF UNIVERSITY OF WASHINGTON ELECTROSTATIC SCRUBBER AT AN
ELECTRIC ARC STEEL FURNACE
PROCEEOlNGS»SYMPj PARTICULATE COLLECTION PROBLEMS USING ELEC-
TROSTATIC PRECIP. IN METALLURGICAL INO., DENVER. 1977, 259 PP(
EPA-60Q/2-77-208
PILAT, M. J..
UNIV. OF WASH. ELECTROSTATIC SPRAY SCRUBBER EVALUATION
UNIV. OF WASHINGTON, EPA, RESEARCH TRIANGLE PARK, N.C.
1976, 77 PP. •••*'•
EPA-600/2-76-100 PB 252 653
PRAKASH, C, B., AND F. E. MURRAY
COLLECTION OF KRAFT MILL PARTICIPATES USING A CONDENSATION
MECHANISM
PULP & PAPER MAGAZINE OF CANADA, 7« (7), PP. 101-105, 1973
SCHWITZGEBEL, K,
DEVELOPMENT OF SAMPLING AND ANALYTICAL METHODS OF LIME/LIMESTONE
WET SCRUBBING TESTS
RADIAN CORP., EPA, RESEARCH TRIANGLE PARK, N.C,, 197*1, 7fe PR.
EPA»650/2«74«02r PB 236
266
-------
SEH> G. J.
SUBMICRON PARTICLE SIZE MEASUREMENT 0F STACK EMISSIONS Us IMP,
THE ELECTRICAL MOBILITY TECHNIQUE i^iv.9 J5i b
ELECTRJC POWER RESEARCH IN$T., WORKSHOP-SAMPLING. ANALYSTS
AND MONITORING OF STACK EMISSIONS, DALLAS. TEX. 1975. 16 PP.
SMITH, W, 8., K. M. GUSHING, AND J. D.I MCCAIN
PROCEDURES MANUAL FOR ELECTROSTATIC PRECIPITATOR EVALUATION
SOUTHERN RESEARCH INSTITUTE, EPA, RESEARCH TRIANGLE PARK.
N. C,, 1977, 430 PP.
EPA-600/7-77-059
SPENCER, H. u/., HI
RAPPING REENTRAINMENT IN -A NEARLY FULL SCALE PILOT
ELECTROSTATIC PRECIPITATOR
SOUTHERN RESEARCH INSTITUTE. EPA, RESEARCH TRIANGLE PARK.
N. C., 1976. 178 PP. "'
EPA»600/?«76»140 PR 255 964
SPROULL, W. T.
MINIMIZING RAPPING LOSS IN PRECIPITATORS AT A SODO-MEGAWATT
COAL"FIRED POWER STATION
J. OF APCA, 2? (3)f PP. 181*186,
STEELE, R. D., G, C, PAGE, AND G. t, MEENAGHAN
'
« OF HVGROSCOPIC PARTICULATES IN AN ENTRAINED WATER
ENVIRONMENT
J^. OF APCA, 25 C6), PP. 634-635, 1975
VINCENT, J. H.
EVALUATION OF A LIGHT TRANSMISSION TECHNIQUE FOR TESTING A
TWO.STAGE, ELECTROSTATIC DUST PRECIPITATOR
J. OF PHY. DI APPL. PHYS. 4, PP. 1835-1841, 1971
267
-------
TECHNICAL REPORT DATA
(I lease read Instructions on the reverse before completing)
EPA-600/7-78-043
4. TITLE AND SUBTITLE
Technical Manual: A Survey of Equipment and Methods
for Particulate Sampling in Industrial Process
Streams
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
March 1978
6. PERFORMING ORGANIZATION CODE
W.B. Smith, P.R. Cavanaugh, andR.R. Wilson
8. PERFORMING ORGANIZATION REPORT NO.
"bRFORMING ORGANIZATION NAME AND ADDRESS
Southern Research Institute
2000 Ninth Avenue, South
Birmingham, Alabama 35205
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
68-02-2131, T.D. 10904A
JG AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 2/77-1/78
14. SPONSORING AGENCY CODE
EPA/600/13
project officer to D.Bruce Harris, Mail Drop 62,
ABSTRACT rj,fre manuai jjgte ^ describes the instruments and techniques that are
available for measuring the concentration or size distribution of particles suspended
in process streams. The standard, official, well established methods are described
as well as some experimental methods and prototype instruments. To the extent
that the information could be found, an evaluation of the performance of each
instrument is included. The manual describes instruments and procedures for
measuring mass concentrations, opacity, and particle size distribution. It also
includes procedures for planning and implementing tests for control device evalua-
tion, a glossary, and an extensive bibliography.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Industrial Processes
Dust Control
Sampling
Measuring
Concentration
Opacity
Pollution Control
Stationary Sources
Particulates
13B
13H
14B
07D
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
280
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)-
268
------- |