v>EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-86-013
July 1986
Air
Generalized
Particle Size
Distributions
For Use In
Preparing
Size Specific
Particulate
Emission
Inventories
-------
EPA-450/4-86-013
Generalized Particle Size Distribution
For Use In Preparing Size Specific
Participate Emission Inventories
By
Keith D. Rosbury
PEi Associates, Inc.
Golden, Colorado 80401
EPA Contract No. 68-02-3512
EPA Project Officer: Arch MacQueen
Air Management Technology Branch
Monitoring and Data Analysis Division
.-...- • :-.• ~ protection Agency,
' 'V:>-':i;-y (5PL-16)
",..„,., V:,-:-eet, Boom 1670
IJi OC»'.:,04
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
July 1986
-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U.S. Environmental
Protection Agency, and approved for publication as received from the contractor. Approval does not
signify that the contents necessarily reflect the views and policies of the Agency, neither does mention of
trade names or commercial products constitute endorsement or recommendation for use.
-------
CONTENTS
Paqe
Contents ii
Figures and Tables iii
1. Rationale for Developing Generalized Particle
Size Distributions 1
2. Basic Approach 2
2.1 Basis for Generalized Categories 2
2.2 Theoretical or Measured Data 4
2.3 Reconciling Differences in the Data Base 4
2.4 Control Devices 5
3. Development of Generalized Particle Size
Distributions for Uncontrolled Sources 6
3.1 Identification of References 6
3.2 Variations in Data 7
3.3 Compilation of Data into Computerized
Data File ' 14
3.4 Development of Initial Classification
System 15
3.5 Development of Final Generalized
Categories and Corresponding Size
Distributions 16
4. Development of Generalized Particle Size
Distributions for Controlled Sources 23
4.1 Calculation of the Size Distribution for
a Controlled Source 23
5. How to Use the Generalized Particle Size
Distributions and Control Efficiency Data 25
5.1 Uncontrolled Sources 25
5.2 Controlled Sources 25
5.3 Example Calculations 26
References 28
Appendix A Generalized Particle Size
Distributions A-l
Appendix B Calculation Sheet B-l
-------
FIGURES
Number Page
5-1 Example Calculation for Determining 27
Uncontrolled and Controlled Particle
Size-Specific Emissions
TABLES
3-1 Particle Size Category by AP-42 Section 20
3-2 Final Generalized Particle Size 22
Distribution Categories
4-1 Average Collection Efficiencies of Various 24
Particulate Control Devices
-------
SECTION 1
RATIONALE FOR DEVELOPING GENERALIZED PARTICLE
SIZE DISTRIBUTIONS
A size-specific National Ambient Air Quality Standard for
particulate is being proposed. Implementation of this standard
will necessitate the preparation of particle size-specific
emission inventories. The U.S. Environmental Protection Agency
(EPA) has developed particle size-specific data for a limited
number of the processes that account for a large fraction of
total national emissions. These data are being incorporated into
the Compilation of Air pollutant Emission Factors (AP-42). Still
needed, however, is particle size information for many processes
that will be of local impact and concern. The purpose of this
assignment is to develop generalized particle size distributions
applicable to sources that have not been sampled adequately to
calculate a size distribution. Generalized size distributions
should only be used in the absence of source-specific particle
size distributions such as those found in the main text of AP-42.
Further, the data should be used for regional emission
inventories only, and should not be used for individual source
compliance purposes.
-------
SECTION 2
BASIC APPROACH
Several technical questions arose at the outset of this
attempt to develop generalized particle size distributions.
These questions and the report sections in which they are
addressed are as follows:
1. On what basis should the generalized categories be
created? (Section 2.1)
2. Should the generalized distributions be based on
theoretical data, measured data, or some combination?
(Section 2.2)
3. If the generalized distributions are based on measured
data, how can data measured by various sampling
methodologies and instrumentation be reconciled into
one data base? (Section 2.3)
4. How should the compounding influence of changes in
particle size distribution caused by control devices be
treated? (Section 2.4)
2.1 BASIS FOR GENERALIZED CATEGORIES
Particle size distribution can be categorized in two
different ways:
1. By the basic physical processes generating the
emissions (e.g., combustion, melting, grinding, wind
erosion).
2. By industry (e.g., metallurgical, mineral products,
iron and steel, phosphate fertilizers).
-------
Designing the category system according to the basic
physical processes generating the emission was believed to be the
more logical approach. Examination of the measured data,
however, indicated the need for adding a second dimension, i.e.,
the material being processed. For example, the emissions
generated by the handling of a fine powdery material differ from
those generated by the handling of a coarse aggregate.
Designing a category system according to basic industry was
found to be infeasible because the emission distributions
generated by the many diverse processes within each industry are
so dissimilar. For example, the iron and steel industry includes
the following basic operations: coke production; sinter
production; iron production; steel production? semifinished
product preparation; heat and electricity preparation; and
handling, transport, and storage of raw materials. Further
complicating this approach are the different processes,
equipment, and materials used within each of these operations.
Because accounting for these differences would necessitate
reverting to the basic process/material handled approach
discussed in the preceding paragraph, categorization of particle
size distribution by industry holds no advantage.
The development of generalized particle size distributions by
basic physical process and materials handled is described in
Section 3.
-------
2.2 THEORETICAL OR MEASURED DATA
A literature search was made of many chemistry, physics, and
engineering sources for theoretical approaches to the prediction
of particle size distribution. In addition, telephone interviews
were conducted with several individuals who are, well known for
their contributions in the field of particulate technology.
Results were very limited. This absence of viable theoretical
approaches made it mandatory to rely primarily on measured data.
2.3 RECONCILING DIFFERENCES IN THE DATA BASE
During compilation of the data base, more than 400 test
series were examined. The data produced by these tests vary
widely with respect to their quality because of such factors as
number of tests, source-operating conditions during the test
{percent capacity, representativeness, upset conditions, etc.),
test instrumentation (Anderson, Brinks, etc.), quality assurance,
method of calculation, physical/aerodynamic/Stokes diameter
questions, etc.
The problems created by these diversities were not
completely overcome. In the basic reference used, the Fine
Particulate Emission Inventory System (FPEIS), the information
for each test series was often not available to answer the
questions. Documentation in the original reports on which this
information was taken is often inadequate. Also, the sheer size
of the data set prohibited an in-depth investigation into each
test series within the scope of this study.
-------
A procedure for reconciling differences in the data base is
discussed in Section 3.
2.4 CONTROL DEVICES
The preceding discussion has centered on the particle size
distribution of an uncontrolled source. Control devices also
influence particle size distributions because each device has a
different control efficiency for different particle size ranges.
For example, a cyclone has a collection efficiency of about 30
percent for 1.0-pm particles and about 93 to 98 percent for
particles in the 20- to 44-ym size range. Also, the same basic
process/material can be controlled by different control devices
in different applications. Therefore, the number of
process/material/control type permutations becomes enormous.
To overcome these two problems, the data for uncontrolled
sources were analyzed separately from those from controlled
sources. The impact of the control device on the particle size
distribution was determined by applying control-device-specific
average collection efficiencies by particle size range to an
uncontrolled particle size distribution. Table A-2 (taken from
AP-42) provides the basis for this approach. These data were
updated by using the 1982 EPA publication, Control Techniques for
Particulate Emissions from Stationary Sources.
The procedure for accounting for the influence of control
devices on particle size distribution is described in detail in
Section 5.
-------
SECTION 3
DEVELOPMENT OF GENERALIZED PARTICLE SIZE
DISTRIBUTIONS FOR UNCONTROLLED SOURCES
The approach decided upon was to develop generalized particle
size distributions for uncontrolled sources with measured data
according to the basic process and material being handled. This
involved the following steps:
(1) Identification of references containing the results of
source testing that produced measured particle size
distributions.
(2) Development of a procedure to account for variations in
the data resulting from differences in sampling
methodology, instrumentation, etc.
(3) Compilation of the data base into a computerized data
file.
(4) Development of initial generalized particle size
classification system.
(5) Development of final generalized particle-size-distribution
categories and a size distribution for each category.
(6) Assignment of a generalized particle size distribution to
all particulate sources listed in AP-42 that did not
already show a particle size distribution.
3.1 IDENTIFICATION OF REFERENCES
The EPA has sponsored several studies for the compilation of
particle size data. Among these are:
(1) Fine Particle Emission Inventory System (FPEIS) . This
computerized system (EPA 1985) is maintained in EPA's
Industrial Environmental Research Laboratory at
Research Triangle Park, North Carolina. The system
-------
contains the results of EPA-sponsored source testing
and source testing sponsored by others. The results of
more than 300 testing programs are documented. From
this report, we obtained all data in the system as of
June 1983. For certain source categories (stationary
internal combustion engines, grain processing,
aggregate processing), additional data inserted to the
FPEIS between June 1983 and May 1985 were obtained.
(2) AP-42 Update for Selected Particle Size Data (EPA
1984). This 1984 report is a compilation of particle
size data derived from primary source testing documents
representing sources that do not have data of
sufficient quality to be presented in the main body of
AP-42 or for which there is no corresponding AP-42
section.
(3) Inhalable Particulate Program. The Office of Air Qual-
ity Planning and Standards (OAQPS) and the Office of
Research and Development (ORD) sponsored an extensive
multiyear source-testing program to develop particle
size data for selected source categories. These data
will greatly expand current knowledge related to
particle size distributions. Source test measurements
from this program were used in the data base. Final
data are expected to be released over the next several
months both by separate ORD report and as an integral
part of AP-42.
(4) Miscellaneous Data. Source testing data available from
sources other than those listed above consist of file
data from testing often sponsored by state or nongov-
ernmental groups. An effort was made to compile much
of these data from contractors' private files and other
miscellaneous sources.
Although all existing source test data reporting particle
size distributions were probably not obtained, the data base is
substantially complete. It should be noted, however, that all
data in the data base have not been peer-reviewed.
3.2 VARIATIONS IN DATA
Data gathered into the data base come from more than 400
testing programs. These data vary widely in quality, and likely
sources of error are discussed.
-------
3.2.1 Errors in Measuring Point Sources
Particulate matter emitted from point sources may be
measured to determine compliance with applicable emission
limitations, to evaluate control equipment performance, or to
establish emission factors. Many of the test methods, however,
may introduce biases that can influence the validity of the
results.
3.2.1.1 Mass Concentration Measurement—
The most precise method of determining the mass
concentration of particulate matter in a gas stream is to collect
the entire volume of gas and the particulate matter and to
determine the mass concentration from this sample. This
procedure, however, is feasible only with a few sources (those
that have very low volumetric flow rates). Various groups have
developed procedures for sampling small portions of a gas stream
to obtain a representative sample of the total gas stream.
Examples of these procedures are EPA Reference Methods 5 and 17,
American Society for Testing and materials (ASTM) Method
D2928-71, and the American Society of Mechanical Engineers (ASME)
Power Test Code 27. The predominant test procedure for
characterization of particulate matter is EPA Reference Method 5,
Determination of Particulate Emissions From Stationary Sources,
Appendix A, 40 CFR 60. The quality assurance checks specified in
Method 5 combined with the use of EPA Methods 1, 2, 3, and 4 help
to ensure the accuracy of mass concentration determinations
obtained by this procedure.
-------
Method 5 is based on extractive filtration. Gas is
extracted isokinetically; i.e., the velocity of the gas entering
the sampling nozzle is equal to the gas velocity passing by the
nozzle at that sampling point. The extraction is made through a
nozzle to an externally heated filter held at 120° ± 14°C. The
particulate matter is captured in the sampling probe and on the
filter, and the filtered gases are then sent through a series of
impingers to remove moisture and other components before they
pass through a dry gas meter. For a test to be valid, isokinetic
conditions must be maintained within ±10 percent of 100 percent.
In a gas stream with both large and small particles, sampling
rates lower than 100 percent isokinetic can bias the sample
toward larger particles, and can strongly bias the mass
concentration calculations. The reverse is true with sampling
rates above 100 percent isokinetic; in this case, the bias toward
smaller particles would result in an apparent mass concentration
that is lower than the actual emission rates.
Establishing isokinetic sampling rates depends on the
characteristics of the individual sampling train and on
determination of gas velocity, gas volumetric flow rate (EPA
Method 2), gas molecular weight (EPA Method 3), and gas moisture
content (EPA Method 4). Procedures outlined in EPA Method 1 are
used to determine the location and suitability of the sampling
site and the location of the sampling points to provide a
representative sample of the gas stream. Thus the use of EPA
Method 5 depends on the proper use of other EPA test methods,
-------
each of which affects whether the mass concentration data will be
representative of the actual emissions from a stationary source.
3.2.1.2 Particle Size Analysis—
The cascade inertial impactor is the device most commonly
used for particulate sizing. The sampling train consists of a
probe, a precutter (such as a cyclone), and the cascade impactor.
The cascade inertial impactor technique provides a
distribution of aerodynamic particle diameters. A cascade
impactor usually has 5 to 10 stages of decreasing orifice
diameters. It is usually assembled to give an alternating
pattern of orifice plates and collection plates. As the orifice
size decreases, the gas velocity through each orifice increases.
Larger particles cannot overcome the inertial force imparted to
them through the orifice and thus impact the collector plate.
Because smaller particles have less inertia, the gas stream
carries them to the next stage. The last stage is usually
followed by a filter for the capture of the smallest particles
that have escaped impaction. Gravimetric methods are used in the
analysis of each stage to determine particle size distribution,
geometric mass median diameter, and geometric standard deviation.
The results of cascade impactors are influenced by the deposition
of particulate in the probe. For example, one test indicated
that at a velocity of 15 m/s, 33 percent of the 10-um particles
were collected in the probe.
Cascade impactors are typically in situ (i.e., in-stack)
devices used with isokinetic sampling rates. When samples are
10
-------
obtained in situ at the stack temperature, the particle size
distribution should be representative of the actual particle size
distribution in the duct. Failure to sample isokinetically
results in a biased and unrepresentative particle size
distribution. A bias toward larger particle sizes occurs with
underisokinetic sampling (i.e., velocity entering nozzle is lower
than the localized gas velocity),.and bias toward smaller sizes
occurs with overisokinetic sampling. Cascade impactors are
provided in stages with nominal values for aerodynamic cut-size
diameters. Each impactor should be calibrated periodically to
determine the actual value of the cut-size diameter for each
stage.
Cascade impactors are susceptible to several problems.
First, in gas streams with high particulate loadings, quick
buildup of material on the stages may shorten the available
testing period. Second, particle reentrainment and bounce can
bias the particle size distribution toward smaller particles.
Finally, fracturing of the larger particles at the impaction
stage may lead to generation of fine particulate and to a
consequent bias toward small particle sizes.
Cyclones are also used for in situ and extractive
aerodynamic particle sizing, but to a lesser extent than cascade
impactors. The aerosol sample enters the cyclone through a
tangential inlet and follows a vortex flow pattern. Particles
that cannot follow the gas streamlines move outward toward the
cyclone wall and, depending on cyclone geometry, gas flow rate,
11
-------
and particle size, may reach the cyclone walls and be collected.
The use of a series of cyclones of different geometric dimensions
at a constant flow rate allows particles to be removed from a gas
stream according to size. The fractionating capability of
cyclones is not theoretically predictable to the degree of
accuracy possible with impactors. Cyclones have an advantage
over impactors in that large samples can be acquired and less
particle reentrainment occurs.
Size-distribution analysis of collected particulate samples
is often performed in the laboratory instead of by in situ
procedures. Errors are possible because the original flue gas
particle size distribution is almost impossible to reconstruct
under laboratory conditions. The gas-stream state of particles
or particle groups may be altered by additional agglomeration or
particle breakup during sample collection. Size distribution
results based on sedimentation and elutriation, centrifuging,
sieving, and electronic counting are meaningful only when the
effects of sample collection and redispersion are negligible or
clearly known.
Microscopic analysis is regarded as the fundamental
technique for counting and sizing particles. This procedure
involves manual or computerized microscopic examination of a
prepared slide containing a representative sample of the aerosol.
The slide must be prepared carefully so that the in-stack state
of the aerosol sample is not altered. Microscopic examination of
particulate matter does not yield size information in terms of
12
-------
aerodynamic diameters; instead, it yields information in terms of
physical diameters. Aerodynamic and physical diameter data are
not directly comparable.
3.2.2 Errors in Measuring Open Sources
Measurement of mass concentration and particle size
distribution at open sources is generally regarded as less
accurate than measurement of point sources. No EPA standard
methods exist for sampling open sources, and sampling
instrumentation, methods, and quality assurance procedures vary
widely. Compared with sampling point sources, sampling of open
sources is plagued by variations in source strength and
difficulties associated with obtaining a sample of a
representative portion of the plume.
3.2.3 Reconciling^ Di_f ference^^in the Data Base
The preceding discussion has indicated that several sources
of error are possible in source testing particulate data for both
point and open sources. Most of the data taken from the four
references listed in Section 3.1 have not undergone EPA's peer
review process.
Problems encountered in attempts to reconcile differences in
the data base were not completely overcome. The FPEIS (the basic
reference used) did not always include inforitfation describing
each test series. Also, documentation in the original reports
was often inadequate. The sheer size of- the data set and the
scope of this study prohibited an in-depth investigation into
each test series.
13
-------
3.4 DEVELOPMENT OF INITIAL CLASSIFICATION SYSTEM
Two alternate approaches were used to develop the initial
classification system accounting for the basic process and
material being processed:
(1) Development of classifications by use of the computer
and statistical programs relating mean values and
correlation analyses applied to the entire data set.
(2) A theoretical approach based on the use of engineering
judgment regarding basic processes and materials being
processed.
3.4.1 Computerized Statistical Approach
For all the test series documenting testing of uncontrolled
sources, the following were entered into a computerized data
base: a process description, cumulative mass at three or four
particle sizes (usually <2.5 ym, <6.0 pm, and <10 urn, but
reference-dependent), and FPEIS Test Series number.
The computerized data base was subjected to two statistical
approaches for development of the category system: 1) rank
ordering of test series by percent of particles less than 10 ym
and 2) correlation analysis by using the three or four cumulative
mass values. The results were the same in both cases. The test
series which were grouped together by statistical routines were
in no way related by process. For example, the size distribution
of emissions from an industrial boiler fired with low-sulfur coal
was found to be identical to that of open fugitive emissions from
an unpaved road in an iron and steel facility. Therefore, the
computerized statistical approach could not be used to develop
categories corresponding to basic process/material combinations.
15
-------
3.4.2 Theoretical Approach
In the theoretical approach, engineering judgment was used
to develop an initial category system according to basic
processes and material being processed. The procedure involved
reviewing each section of AP-42 to develop an initial list of
categories. The processes and materials in each section were
identified and then combined into a single list. Next, an effort
was made to combine similar process/material combinations as a
means of reducing the number of initial generalized categories.
This initial list of generalized categories based on engineering
judgment contained 33 entries.
•
3.5 DEVELOPMENT OF FINAL GENERALIZED CATEGORIES AND CORRESPONDING
SIZE DISTRIBUTIONS
3.5.1 Development of Particle Size Distributions
After the data were coded into the data base, they were
sorted according to the 33 generalized process categories. To
develop the average particle size distribution for each
generalized category, a replicate of the PADRE program for
combining data was used. Within each category, the percentages
of all particulates less than 2.5 ym in size were averaged to
produce a mean value. Similarly, the values of all particulates
under 6.0 ym and all under 10 ym were averaged. These three mean
values were then plotted and connected with a line to obtain the
particle size distribution for that category. To obtain
cumulative mass values for size fractions other than 2.5, 6.0,
16
-------
and 10.0 ym, a utility program that acknowledges the log-
probability format of the data was used.
Results of the initial curve fitting were mixed. For some
generalized categories the plots were satisfactory (<10 ym values
were within ± 15 percent), for some they indicated random data
scatter, and for others there were groupings of data (10-ym
values clustered around two or more percentages). In this
application, data scatter can be attributed to one or more of
three factors:
1. Test data assigned to improper category.
2. Category too broadly defined.
3. Test data not representative of category because of
unrepresentative source conditions or measurement
errors.
Using the following procedures, we critically examined the
plots for each generalized category for the possibility of
unacceptable data scatter resulting from any of these factors:
1. Categorization of all data was verified for
correctness. Potential sources of error were data
entry mistakes and improper judgment in category
assignment. This required going back to the original
data reference to obtain more information about
conditions during testing.
2. An attempt was made to arrive at a more restrictive
definition of each category, which resulted in the
creation of additional categories. This was usually
based on the material being processed. When the
categories were more restrictively defined, some test
series were reassigned to a different category.
3. Extreme values (high or low) were critically examined.
This entailed reexamination of the original references.
In some cases the extreme values could be attributed to
a special testing condition. In other cases the data
were obviously illogical and could only be attributed
to measurement or reporting error. Any data that were
17
-------
determined to be unrepresentative or in error were
removed from the data base.
Even after these procedures were applied, some categories
still had far more data scatter than others. This could be
attributed to real variations in source emissions or to
measurement or reporting errors.
The procedures described resulted in expansion of the number
of generalized categories from 33 to 43 categories.
3.5.2 Development of Final Categories
Forty-three categories were considered to be an undesirably
large number for the following reasons:
1. Fewer categories would be less cumbersome for local and
state agencies in developing SIP revisions.
2. These data probably do not justify the implied preci-
sion of 43 categories; e.g., differences of five per-
centage points in cumulative mass probably could be
just as attributable to data "noise" as to real source
differences.
An attempt was made to reduce the number of categories by
rank-ordering all categories by cumulative mass of particulates
less than 10 pm in size. When categories had cumulative mass
percentages that were representative of related process/materials
combinations, these categories were combined. In addition to the
rank ordering procedure, categories were also eliminated when
they represented source categories for which particle size
distributions were already in, or planned to be in AP-42. The
result was a total of nine categories.
18
-------
3.5.3 Assignment of Generic Categories to Particulate Sources
Listed in AP-42
The form of the data presentation was determined by the
intended end use of the data, i.e., emission inventories.
Because the basic reference for emission factors is AP-42, it was
decided to link the data presentation to AP-42 organization. A
tabular presentation was developed that lists the particular
process name and number, and the assigned generalized particle
size distribution. These data are shown in Table 3-1.
Table 3-2 lists the generalized particle size categories,
the percent cumulative mass of particles in the <2.5 ym, <6.0 ym
and <10 ym size categories. Data supporting each category, and a
particle size distribution for each category are shown in
Appendix A.
19
-------
TABLE 3-1. PARTICLE SIZE CATEGORY BY AP-42 SECTION
AP-42
Section
1.1
l.'Z
1.3
1.4
1.5
1.6
1.7
1.8
1.9
1.10
1.11
2.1
2.3
3.2
5.4
5.8
5.10
5.11
5.12
5.16
5.17
6.1
~
6.2
6.3
6.4
Source Category
External combustion
Bituminous coal combustion
Anthracite coal combustion
Fuel oil combustion
Utility, residual oil
Industrial, residual oil
Utility, distillate oil
Commercial, residual oil
Commercial, distillate
Residential, distillate
Natural gas combustion
Liquefied petroleum gas
Wood waste combustion in
boilers
Lignite, combustion
Bagasse Combustion
Residential fireplaces
Wood stoves
Waste oil combustion
Solid waste disposal
Refuse Incinerators
Conical burners (wood waste)
Internal combustion engine
j
Highway vehicles"
Off highway
Chemical process
Charcoal production
Hydrofluoric acid
Spar drying
Spar handling
Transfer
Paint
Phosphoric acid (thermal
process)
Phthalic anhydride
Sodium carbonate
Sulfuric acid
Food and agricultural
Alfalfa dehydrating
Primary cyclone
Meal collector cyclone
Pellet cooler cyclone
Pellet regrind cyclone
Coffee roasting
Cotton ginning
Feed and grain mills and
elevators
Unloading
Category
Numberc
b
a
a
2
b
2
a
1
9
3
3
3
4
a
9
a
b
b
7
7
7
6
b
b
AP-42
Section
6.5
6.7
6.8
6.10
6.10.3
6.11
5.14
6.16
6.17
6.18
7,1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
C
Source Category
Food and agricultural (cont.)
Grain elevators
Grain processing
Fermentation
Heat smokehouses
Ammonium nitrate fertilizers
Phosphate fertilizers
Ammonium phosphates
Reactor/ ammoniator-
granulator
Dryer/cooler
Starch manufacturing
Ure* manufacturing
Defoliation and harvesting
of cotton
Trailer loading
Transport
Harvesting of grain
Harvesting machine
Truck loading
Field transport
Ammonium sulfate manufacturing
Rotary dryer
Fluidized bed dryer
Metallurgical industry
Primary aluminum production
Bauxite grinding
Aluminum hydroxide calcining
Anode baking furnace
Prebake cell
Vertical Soderberg
Horizontal Soderberg
Coke manufacturing
Primary copper smelting
Ferroalloy production
Iron and steel production
Blast furnace
Slips
Cast house
Sintering
Windbox
Sinter discharge
Basic oxygen furnace
Electric arc furnace
Primary lead smelting
Z1nc smelting
Secondary aluminum
Sweating furnace
Smelting
Crucible furnace
Reverberatory furnace
Secondary copper smelting
and alloying
Gray iron foundries
ategory
Number
6
7
647
9
a
2
4
4
7
*
6
5
6
6
6
b
b
4
5
9
a
8
a
a
a
£
a
a
a
a
a
a
a
&
8
3
a
8
a
a. Categories with particle size data specific to process included in the main body of the text.
b. Categories with particle size data specific to process included 1n Appendix C.I.
c. Data for each numbered category are shown in Appendix A.
d. Highway vehicles data are reported in AP-42 Volume II: Mobile Sources.
20
-------
TABLE 3.1 (continued).
Af-42
Section
Source Category
Category
Hus*erc
AP-42
Section
Source Category
Category
Number
Metallurgical industry (cont.)
7.11 Secondary lead processing a
7.12 Secondary magnesium smelting 8
7.13 Sc««l foundries
naleing b
7.14 Secondary zinc saelting 3
7.15 Storage battery production b
7.13 Leadbesring ore crushing and
grinding 4
Mineral products
8.1 Asphaltic concrete plants
Process m
8.3 Bricks and related clay
products
Rav materials handling
Dryersi grinders, etc. b
Tunnel/periodic kilns
Gas find a
Oil fired a
Coal fired a
3.3 Caetable refractorias
Raw MI«rial dryer 3
Raw Material crushing and
screening 3
Electric arc nelting 8
Caring oven 3
8.6 Portland ceioent manufacturing
Cry process
Kilns a
Dryers, grinders, etc. 4
Wet process
Ulna a
Dryers,, grinde**. etc, *
8.7 Ceraaic clay manufacturing
Drying 3
Grinding 4
Storage 3
3.8 Clay and fly ash sintering
Fly ash sintering, crushing,
screening and yard storage 5
Clay nixed with coke
Crushing, screening, and
yard storage 3
8.9 Coal cleaning 3
8.10 Concrete batching 3
8,11 Glass fiber manufacturing
Unloading and conveying 3
Storage bins 3
Mixing and weighing 3
Class furnace - wool a
Glass furnace - textile a
9.13 Class manufacturing a
8.14 Gypsum manufacturing
Rotary ore dryer a
Roller Bill 4
Mineral products (cont.)
Impact mill 4
flaah calciner a
Continuous kettle calcinar a
8.15 Lime manufacturing a
3.16 Mineral wool manufacturing
Cupola 3
Reverberacory furnace 8
Blow chamber S
Curing oven 9
Cooler 9
3.18 Phosphate rock processing
Crying a
Calcining a
Grinding b
Transfer and storage 3
8.19.1 Sand and gravel processing
Continuous drop
Transfer station
Pile formation - stacker
Batch drop
Active storage piles
Vehicle traffic unsaved road
8.19.2 Crashed stone processing
Dry crushing
Primary crushing a
Secondary crushing
and screening a
Tertiary crushing
and screening 3
Eecruehlng and screening 4
fines Bill 4
Screening, conveying,
and handling a
8.22 Taeonite ore processing
Fine crushing 4
Waste gas a
Pellet handling 4
Grate discharge 5
Grate feed 4
Sentonite blending 4
Coarse crushing 3
Ore transfer 3
Bentonlee transfer 4
0npavad itoeda a
3.23 Metallic minerals processing a
8,24 Western surface coal ainiag s
Wood processing
10.1 Chemical wood pulping a
Miscellaneous sources
11.2
Fugitive dust
a. Categories with particle size data specific to process Included In the main body of the text.
b. Categories with particle size data specific to process included in Appendix C.I.
c. Data for each numbered category are shown in Appendix A,
21
-------
TABLE 3-2. FINAL GENERALIZED PARTICLE SIZE DISTRIBUTION CATEGORIES
(% cumulative mass)
Generic
Category
Number
1
2
3
4
5
6
7
8
9
Process
Stationary Internal
combustion engines
Combustion
Mechanically
generated
Mechanically
generated
Calcining and other
heat reaction
processes
Grain handling
Grain processing
Melting, smelting
refining
Condensation,
hydration,
absorption,
prilling and
distillation
Material
Gasoline and
diesel fuel
Mixed fuels
Aggregate,
unprocessed ores
Uranium,
processed ores
Aggregate,
unprocessed ores
Grain
Grain
Metals, except
aluminum
All
<2.5 urn
Mass
Less
Than
90
45
15
30
17
1
23
B2
78
Min
78
3?
3
1
3
0
17
63
59
Max
99
70
35
51
42
2
34
99
99
S.D.a
11
17
7
19
11
1
9
12
17
6.0 urn
Mass
Less
Than
93
70
34
62
35
7
43
89
91
Min
86
49
15
17
9
3
35
75
61
Max
99
84
65
83
74
12
48
99
99
S.O.
7
14
13
17
19
3
7
9
12
10.0 un
Mass
Less
Than
96
79
51
85
50
15
61
92
94
Min
92
56
23
70
14
6
56
80
71
Max
99
87
81
93
84
25
65
99
99
S.D.
4
12
14
7
19
7
5
7
9
N)
Standard Deviation
-------
SECTION 4
DEVELOPMENT OF GENERALIZED PARTICLE SIZE
DISTRIBUTIONS FOR CONTROLLED SOURCES
4.1 CALCULATION OF THE SIZE DISTRIBUTION FOR A CONTROLLED SOURCE
Section 3 presents detailed procedures for developing a
particle size distribution for uncontrolled sources. The purpose
of this section is to describe the development of a procedure to
allow calculation of a size distribution for a controlled source.
The large number of possible source/control combinations
prompted the use of generalized data by type of control device and
fractional control efficiency. This approach is based on Table
A-2 in AP-42. This table was updated to reflect recent
technology and is presented here as Table 4-1. The primary
reference for the update was EPA's Control Techniques for
Particulate Emissions from Stationary Sources (EPA 1980).
However, other references were also used (EPA 1977; Gushing
undated).
To use Table 4-1, the analyst must first develop the
uncontrolled size distribution according to the procedures given
in Section 4. The fractional control efficiencies are applied to
the uncontrolled size distribution to calculate the controlled
size distribution. This procedure is illustrated in Section 5.
23
-------
TABLE 4-1. AVERAGE COLLECTION EFFICIENCIES OF VARIOUS
PARTICULATE CONTROL DEVICES.
(percent)
Type of collector
Baffled settling chamber
Simple (high-throughput)
cyclone
High-efficiency and
multiple cyclones
Electrostatic precipitator
(ESP)
Packed-bed scrubber
Venturi scrubber
Wet-impingement scrubber
Fabric filter
Particle size, pm
Overall
—
80
90-99
99.5
90-95
96-97
90
99.3-99.9
0 - 2.5
NR
50-70
80-95
96.1-99.5
90-99.6
93-97
8-74
99.3-99.9
2.5 - 6
0-6
70-83
95-98
99.7
98-99.6
94.0-98.3
74-98
99.7-99.9
6-10
6-20
83-90
99
99.3-99.8
98-99.6
98.3-99.0
90-98
99.8-99.9
The data shown represent an average of actual efficiencies. The
efficiencies are representative of well-designed and well-operated
control equipment. Site-specific factors (e.g., type of particulate
being collected, varying pressure drops across scrubbers, maintenance of
equipment) will affect the collection efficiencies. The efficiencies
shown are intended to provide guidance for estimating control equipment
performance when site-specific data are not available.
NR Not reported.
24
-------
SECTION 5
HOW TO USE THE GENERALIZED PARTICLE SIZE
DISTRIBUTIONS AND CONTROL EFFICIENCY DATA
Appendix B contains a calculation sheet to assist the
analyst in preparing particle size specific emission estimates.
5.1 UNCONTROLLED SOURCES
The following instructions apply to each particulate
emission source for which a particle size distribution is desired
and for which no source specific particle size information is
give elsewhere in this AP-42:
1. Identify and review the AP-42 section dealing with the
source.
2. Obtain the uncontrolled emission factor from the main
text of AP-42 and calculate uncontrolled total
particulate emissions.
3. To develop the size distribution, for sources which do
not have source specific in this AP-42, obtain the
generalized particle size distribution category number from
Table 3-1.
4. Obtain the particle size distribution for the
appropriate category from Table 3-2. Apply the
particle size distribution to the uncontrolled
particulate emissions.
5.2 CONTROLLED SOURCES
To calculate the size distribution for a source with a
particulate control device the used should first calculate the
25
-------
uncontrolled size distributions. Next, the fractional control
efficiency for the control device should be estimated using Table
4-1. The Calculation Sheet (Appendix B) allows the user to
record the type of control device and the collection efficiency
from Table 4-1, the mass in the size range before and after
control, and the cumulative mass. The user should note that the
uncontrolled size data is expressed in cumulative fraction less
than the stated size. The control efficiency data applies only
to the size range indicated and is not cumulative.
5.3 EXAMPLE CALCULATION
An example calculation is shown on Figure 5-1. After
recording process identifiers, uncontrolled total particulate
emissions, uncontrolled size-specific emissions, and controlled
size specific emission are then calculated.
26
-------
FIGURE 5-1. EXAMPLE CALCULATION FOR DETERMINING UNCONTROLLED
AND CONTROLLED PARTICLE SIZE-SPECIFIC EMISSIONS.
SOURCE IDENTIFICATION
Source name and address: ABC Brick Manufacturing
Process description:
AP-42 category:
Uncontrolled AP-42
emission factor:
Activity parameter:
Uncontrolled emissions:
24 Dusty Way
Anywhere, USA
Dryers/Grinders
8.3 Bricks and Related Clay Products
96 Ibs/ton
63,700 tons/year
3057.6 tons/year
(units)
(units)
(units)
UNCONTROLLED SIZE DISTRIBUTION
Category name: Mechanically Generated/Aggregate, Unprocessed Ores
Category number: 3
Particle size, urn
< 2.5
< 6
Generic distribution, Cumulative
< 10
percent less than or equal to:
Mass in size range, (units = tons/year):
CONTROLLED SIZE DISTRIBUTION
Type of control device: Fabric Filter
Collection efficiency Table 4-1:
*
Mass in size range before control
(units=tons/year):
Mass in size range after control:
Cumulative mass:
15 34
458.6 1039.6
Particle size, ym
0-2.5 2.5-6
99.6 99.8
458.6 581.0
1.83 1.16
2.99
51
1559.4
6-10
99.9
519.8
0.52
3.51
* Note that uncontrolled size data is cumulative percent less than.
Control efficiency data applies only to size range and is not cumulative.
27
-------
REFERENCES
Gushing, K. M. Undated. Development of Horizontal Elutriators
for Sampling Inhalable Particulate Fugitive Emissions. Southern
Research Institute, Birmingham, Alabama.
Environmental Protection Agency. 1977. Operation and
Maintenance of Particulate Control Devices on Coal-Fired Utility
Boilers. EPA-600/2-77-129.
Environmental Protection Agency. 1982. Control Techniques for
Particulate Emissions From Stationary Sources—Volumes 1 and 2.
EPA-450/3-81-005a.
Environmental Protection Agency. 1984. AP-42 Update for
Selected Particle Size Data. Prepared by Engineering-Science,
Durham, NC for Air Management Technology Branch, Research
Triangle Park, NC 27711.
Environmental Protection Agency. 1985. Fine Particulate
Emission Inventory System. Air and Energy Engineering Research
Laboratory, Research Triangle Par, NC 27711.
28
-------
APPENDIX A
GENERALIZED PARTICLE SIZE DISTRIBUTIONS
-------
This appendix contains two sheets for each of the nine generalized
particle size categories. The first sheet presents category identifiers, a
plot of the size distribution, and a particle size summary. The second sheet
for each category lists the data that were used to develop the category
distribution.
A-l
-------
Category: 1
Process: Stationary Internal Combustion Engines
Material: Gasoline and Diesel Fuel
Category 1 describes emissions from stationary internal combustion
engines. The particulate emissions are generated from fuel combustion.
V
H~
Z
3"
j™
«t
73
98
95
90
80
70
60
50
an
1 — •
-
_
-
-
• — i 1 1 — '
^»— —•'
^*-*-*^
1 } 1 IT '
-
•"""
-
.
_
_
_
i i i i i i i i i
2 345
PARTICLE DIAMETER,
10
Cumulative %
less than or equal
Particle to stated size
size, ym (uncontrolled)
1.0* 82
2.0a 88
2.5a 90
3.0? 90
4.0? 92
5.0a 93
6.0 93
10.0 96
Minimum
Value
78
86
92
Maximum
Value
99
99
99
Standard
Deviation
11
7
4
a Value calculated from data reported at 2.5, 6.0, and 10.0 ym.
statistical parameters are given for the calculated value.
No
A-2
-------
Category: 2
Process: Combustion
Material: Mixed Fuels
Category 2 contains boilers firing a mixture of fuels regardless of the
fuel combination. The fuels include gas, coal, coke, and petroleum.
Particulate emissions are generated as the result of firing these
miscellaneous fuels.
o
at:
LU
a.
95
90
80
70
60
50
40
30
20
10
j i i
I i i i
2345 10
PARTICLE DIAMETER, \m
Cumulative %
less than or equal
Particle to stated size
size, ym (uncontrolled)
1.0? 23
2.0a 40
2.5a 45
3.0* 50
4.0* 58
5.0a 64
6.0 70
10.0 79
Minimum
Value
32
49
56
Maximum
Value
70
84
87
Standard
Deviation
17
14
12
Value calculated from data reported at 2.5, 6.0, and 10.0
statistical parameters are given for the calculated value.
No
A-4
-------
Category:
Process:
Material:
Combustion
Mixed Fuels
Source description
Ind, boiler-petroleum/coke
Util. boiler-80% coal/20% coke
Util. bdiler-75% coke/25% gas
Util. boiler-10% gas/90% coal
Util. boiler-petroleum/coke
Util. boiler-petroleum/coke
2.5
35
32
63
70
34
38
Cumulative percent less
than or equal to
stated size
6.0 ym
78
65
84
82
63
49
10.0
87
81
87
86
78
56
Ref.
1/163
1/73
1/108
1/82
1/75
1/100
A-5
-------
Category:
Process:
Material:
Mechanically Generated
Aggregate, Unprocessed Ores
Category 3 covers material handling and processing of aggregate and
unprocessed ore. This broad category includes emissions from milling,
grinding, crushing, screening, conveying, cooling, and drying of material.
Emissions are generated through either the movement of the material or the
interaction of the material with mechanical devices.
V
t—
z
=>
u
90
80
70
60
50
40
30
20
10
5
2
2345
PARTICLE DIAMETER,
10
Particle
size, ym
1.0*
2.0a
2.5
3.0*
4.0a
5.0a
6.0
10.0
Cumulative %
less than or equal
to stated size
(uncontrolled)
15
18
25
30
34
51
Minimum
Value
15
23
Maximum
Value
35
65
81
Standard
Deviation
13
14
Value calculated from data reported at 2.5, 6.0, and 10.0 ym.
statistical parameters are given for the calculated value.
No
A-6
-------
Category: 3
Process: Mechanically Generated
Material: Aggregate, Unprocessed Ore
Cumulative percent less
than or equal to
stated size
Source description 2.5 pm
Asphalt batch-dry/screen./mix. 15
Asphalt concrete-drum mix 21
Cement-clinker cooler 8
Clay aggregate-clinker cooler 16
Clay aggregate-clinker cooler 15
Copper ore-conveying 10
Copper ore-crushing 18
Copper ore-crushing 12
Copper ore-crushing 11
Copper ore-loadout 5
Copper ore-truck dump 14
Feldspar milling 11
Fluorspar processing-rotary drum 10
dryer
Gold-ore crushing/conveying/storage 16
Gypsum-rock dryer 10
Molybdenum-screening 21
Molybdenum-screening 27
Phosphate rock-dryer 20
Sodium carbonate-drying 22
Sodium carbonate-drying 10
Talc-grinding 18
Vanadium ore-dryer. 12
Vanadium ore-dryer 12
Vanadium ore-drying/grinding 13
Zinc ore-crushing 3
Zinc ore-crushing/screening/conveying 7
Zinc ore-dryer 35
Zinc ore-screening 26
Zinc ore-screw conveying ~ 7
6.0 ym
21
52
17
30
26
31
34
25
22
27
49
23
30
37
30
46
55
41
65
15
43
33
31
36
19
30
41
52
22
10.0 ym
44
66
32
40
38
53
42
50
43
43
81
37
48
62
39
70
72
60
69
23
60
44
60
58
38
48
62
64
29
Ref.
1/41
1/299
1/86
7
2
1/310
1/310
1/309
1/329
1/345
1/339
4
2
1/335
1/358
-360
1/334
1/333
1/94
1/376
1/378
4
1/290
1/337
1/338
l/344b
l/334a
1/343
l/344c
l/344d
A-7
-------
Category:
Process:
Material:
Mechanically Generated
Uranium, Processed Ores
Category 4 covers material handling and processing of uranium and
processed ores. While similar to Category 3, uranium and processed ores can
be expected to have a greater size consistency than unprocessed ores.
Particulate emissions are generated as a result of agitating the materials by
screening or transfer, during size reduction of the materials by crushing and
grinding, or by drying.
1/1
V
95
90
u, 80
fsj
£ 70
2 60
50
40
30
20
10
5
2
1
0.5
1 I I 1 r
i i i i i
2 345
PARTICLE DIAMETER, \*n
10
Cumulative %
less than or equal
Particle to stated size
size, ym (uncontrolled)
1.0? 6
2.0a 21
2.5a 30
3.Of 36
4.0! 48
5.0a 58
6.0 62
10.0 85
Minimum
Value
17
70
Maximum
Value
51
83
93
Standard
Deviation
19
17
7
Value calculated from data reported at 2.5, 6.0, and 10.0 ym. No
statistical parameters are given for the calculated value.
A-8
-------
Category: 4
Process: Mechanically Generated
Material: Uranium, Processed Ores
Cumulative percent less
than or equal to
stated size
Source description 2.5 pm
Ammonium sulfate-dryer 1
Ammonium sulfate-dryer 8
Clay-dryer 37
Clay mfg.-milling 5
Clay mfg.-milling 14
Clay mfg.-Raymond mill 50
Potassium chloride-dryer 22
Potassium chloride-dryer 19
Salt-dryer 49
Salt-dryer 36
Uranium ore-crusher, grizzly and 51
transfer points
Uranium ore-fine ore bin exhaust 51
Uranium ore-loading 45
6.0 pm
17
53
75
52
59
52
64
68
59
77
75
83
77
10.0 pm
70
83
90
85
86
85
85
89
69
92
87
93
88
Ref.
1/163
1/383
1/88
1/381
1/384
1/96
1/350
1/386
1/53
1/52
1/284
1/285
1/286
A-9
-------
Category:
Process:
Material:
«/
Calcining and other Heat Reaction Processes
Aggregate, Unprocessed Ores
Category 5 covers the use of calciners and kilns in processing a variety
of aggregates and unprocessed ores. Emissions are generated as a result of
these high temperature operations.
90
£ 80
o 70
UJ
H 60
1/1 50
V
£ 40
UJ
£ 30
UJ
°" 20
UJ
< 10
I 5
2
1
I I 1 I I I
2 345
PARTICLE DIAMETER, urn
10
Cumulative %
less than or equal
Particle to stated size
size, ym (uncontrolled)
1.0? 6
2.0a 13
2.5a 17
3.Of 20
4.0* 26
5.0a 31
6.0 35
10.0 50
Minimum
Value
9
14
Maximum
Value
42
74
84
Standard
Deviation
11
19
19
Value calculated from data reported at 2.5, 6.0, and 10.0 ym.
statistical parameters are given for the calculated value.
No
A-10
-------
Category:
Process:
Material:
Calcining and Other Heat Reaction Processes
Aggregate, Unprocessed Ore
Cumulative percent less
than or equal to
stated size
Source description 2.5 pm
Brick mfg.-kiln/dry 25
Brick mfg.-kiln/dry 21
Cement mfg,-kiln 42
Cement mfg.-rotary kiln 18
Clay aggregate-rotary kiln 14
Gypsum-flash calciners 23
Iron ore benefication-grate kiln 18
system
Lime mfg.-rotary kiln 3
Lime mfg.-rotary kiln 27
Lime mfg.-rotary kiln 3
Pulp/paper-lime recovery kiln 23
Shale aggregate plant-rotary kiln 3
Sodium carbonate-calcining 23
Sodium carbonate-calcining 19
Taconite proc.-preheat 4
Vanadium ore-kiln drying 3
6.0 urn
50
44
74
38
29
57
28
9
56
14
34
13
40
39
14
21
10.0 ym
70
62
84
57
42
75
35
14
67
35
49
25
53
50
45
43
Ref.
1/354
1/33
1/298
1/80
2
1/295
8
1/330
1/294
1/295
1/104
-107
2
1/375
1/377
1/348
1/289
A-ll
-------
Category:
Process:
Material:
Grain Handling
Grain
Category 6 contains various grain handling (versus grain processing)
operations. These processes could include material transfer, ginning and
other miscellaneous handling of grain. Emissions are generated by mechanical
agitation of the material.
tSI
o
30
20
10
5
2
' 1
0.5
0.2
0.1
0.05
0.01
1IIIIITT
2345 10
PARTICLE DIAMETER, ym
Cumulative %
less than or equal
Particle to stated size
size, yi (uncontrolled)
1.0a .07
2.0a .60
2.5a 1
3.0a 2
4.Of 3
5.0a 5
6.0 7
10.0 15
Minimum
Value
3
6
Maximum
Value
12
25
Standard
Deviation
3
7
Value calculated from data reported at 2.5, 6.0, and 10.0
statistical parameters are given for the calculated value.
No
A-12
-------
Category:
Process:
Material:
Grain Handling
Grain
Source description
Cotton ginning-roller gin, bale
press
Cotton ginning-roller gin, gin stand
Cotton ginning-saw gin, bale press
Cotton ginning-saw gin, gin stand
Rice-dryer
Cumulative percent less
than or equal to
stated size
2.5 pm 6.0 ym 10.0 wm Ref.
1
1
1
0
2
7
3
5
12
13
17
6
14
25
5
5
5
1/228
A-13
-------
Category:
Process:
Material:
Grain Processing
Grain
Category 7 includes grain processing operations such as drying,
screening, grinding and separation. The particulate emissions are generated
during forced-air flow, separation or size reduction.
|v
80
70
60
50
40
30
20
10
i iiit
2 345
PARTICLE DIAMETER,
10
Cumulative %
less than or equal
Particle to stated size
size, ym (uncontrolled)
1.0? 8
2.0a 18
2.5a 23
3.0? 27
4.0a 34
5.0a 40
6.0 43
10.0 61
Minimum
Value
17
35
56
Maximum
Value
34
48
65
Standard
Deviation
Value calculated from data reported at 2.5, 6.0, and 10.0 ,ym.
statistical parameters are given for the calculated value.
A-14
-------
Category: 7
Process: Grain Processing
Material: Grain
Source description
Agricultural feed-production
Cereal-dryer
Cotton gin-battery condenser
effluent
2.5
19
34
17
Cumulative percent less
than or equal to
stated size
6.0 urn
46
48
35
10.0
65
56
61
Ref.
1/154
2
1/27
A-15
-------
Category: 8
Process: Melting, Smelting, Refining
Material: Metals, except Aluminum
Category 8 includes the melting, smelting, and refining of metals
(including glass) other than aluminum. All primary and secondary production
processes for these materials which involve a physical or chemical change are
included in this category. Materials handling and transfer are not included.
Particulate emissions are generated as a result of high-temperature melting,
smelting, and refining.
o
cr
s«
98
95
90
80
70
60
50
40
i
-
_
I
-
<-
• t it1
^^^
*~~**^
1 ! i '
— - '
-
-
'
•
-
L
i ill
i ! I 1
2345 10
PARTICLE DIAMETER,
Cumulative %
less than or equal
Particle to stated size
size, urn (uncontrolled)
1.0* 72
2.CT 80
2.5a 82
3.0* 84
4.0? 86
5.0a 88
6.0 89
10.0 92
Minimum
Value
63
75
80
Maximum
Value
99
99
99
Standard
Deviation
12
9
7
Value calculated from data reported at 2.5, 6.0, and 10.0 pm,
statistical parameters are given for the calculated value.
No
A-16
-------
Category: 8
Process: Melting, Smelting, Refining
Material: Metals, except aluminum
Cumulative percent less
than or equal to
stated size
Source description
Borax-fusing furnace
Copper-smelter
FE. prod.-ferroscilicon
Ferroalloy-EAF
Glass-manufacturing
Gray iron-cupola
Gray iron-scrap cupola
Iron & steel prod.-iron cupola
Mineral wool-cupola
Steel foundry-EAF
Steel foundry-EAF
Steel foundry-EAF oxygen decarb.
Steel foundry-EAF oxygen decarb.
Steel foundry-open hearth
Steel foundry-open hearth
Steel foundry-open hearth
Zinc-fuming furnace
Zinc-retort furnace
Zinc-roaster
Zinc-smelter-sintering
Zinc-vert, retort
2.5 y
88
96
97
83
91
93
95
92
67
69
69
69
67
68
80
82
63
82
99
92
75
6.0 ym
98
99
99
84
93
98
99
96
82
79
84
79
76
86
83
88
75
97
99
99
77
10.0 ym
99
99
99
94
95
99
99
98
91
82
90
81
80
92
85
92
82
99
99
99
86
Ref.
1/90
1/2
1/51
1/280
1/219,
223,
224
1/54
1/55
1/42
1/123
1/308
1/76
2
2
1/83
1/233
1/45
2
1/44
1/1
1/3
1/43
A-17
-------
Category: 9
Process: Condensation, Hydration, Absorption, Prilling and Distillation
Material: All
Category 9 includes condensation, hydration, absorption, prilling, and
distillation of all materials. These processes involve the physical
separation or combination of a wide variety of materials such as sulfuric acid
and ammonium nitrate fertilizer. Coke ovens are included since they can be
considered a distillation process which separates the volatile matter from
coal to produce coke.
98
95
90
80
70
60
50
40
I i i i i i
2 345
PARTICLE DIAMETER,
Cumulative %
less than or equal
Particle to stated size
size, Pm (uncontrolled)
l.o! 60 •
2.CT 74
2.5a 78
3.0? 81
4.0* 85 .
5.0a 88
6.0 91
10.0 94
Minimum
Value
59
61
71
Maximum
Value
99
99
99
Standard
Deviation
17
12
9
a Value calculated from data reported at 2.5, 6.0, and 10.0 urn.
statistical parameters are given for the calculated value.
A-18
-------
Category:
Process:
Material:
Condensation,
All
Hydration, Absorption, Prilling, Distillation
Cumulative percent less
than or equal to
stated size
Source description
Amm. nit. fert.-rotary prilling
Amm. nit. fert.-urea prilling
Amm. nit. fert.-urea prilling
Amm. nit. fert.-urea prilling
Amm. nit. fert.-urea prilling
Iron & steel prod.-coke oven
Pulp mill-sulfate pulp
Sul. acid-absorb
Sul. acid-absorb. (20% 0)
Sul. acid-absorb. (32X 0)
2.5
83
70
73
97
47
77
77
59
97
99
6.0 m
89
89
89
99
61
96
87
98
99
99
10.0 wn
96
94
93
99
71
98
94
99
99
99
Ref.
1/336
1/362
1/355
1/48
1/372,
380
1/142
1/83-
84
3
3
3
A-19
-------
REFERENCES FOR APPENDIX A
1. Fine Particle Emission Inventory System, U.S. Environmental Protection
Agency, Office of Research and Development, Research Triangle Park, NC,
1965.
2. Confidential Test Data from Various Sources, PE! Associates, Inc.,
Cincinnati, OH, 1985.
3. Final Guideline Document: Control of SulfuricAcid Production Units,
EPA-450/2-77-019, U.S. Environmental ProtectionAgency, Research Triangle
Park, NC, 1977.
4. Air Pollution Emission Test, Bunge Corp., Destrehan, La., EMB-74-GRN-7,
U.S. Environmental Protection Agency, Research Triangle Park, NC, 1974.
5. I.W. Kirk, "Air Quality in Saw and Roller Gin Plants", Transactions of
the ASAE, Volume 20, No. 5, 1977.
6. Emission Test Report, Lightweight Aggregate Industry, Salite Corp.,
EMB-80-LWA-6, U.S. Environmental Protection Agency, Research Triangle
Park, NC, 1982.
7. Air Pollution Emission Test, Lightweight Aggregate Industry, Texas
Industries, Inc., EMB-8Q-LWA-3, U.S. Environmental Protection Agency,
Research Triangle Park, NC, 1981.
8. Air Pollution Emission Test, Empire Mining Company, Palmer, Michigan,
EMB-76-IOB-2, U.S. Environmental Protection Agency, Research Triangle
Park, NC, 1975.
9. H. Taback, et. al., Fine Particulate Emission from StationarySources in
the South Coast Air Basin, KVB. Inc.. Tustin. CA, 1979.
10. K. Rosbury, Generalized Particle Size Distributions for Use in Preparing
Particle Size Specific Emission Inventories, Contract No. 68-02-3890, PEI
Associates, Inc., Golden, CO, 1985.
A-20
-------
APPENDIX B
CALCULATION SHEET
-------
CALCULATION SHEET
SOURCE IDENTIFICATION
Source name and address:
Process description:
AP-42 category:
Uncontrolled AP-42
emission factor:
Activity parameter:
Uncontrolled emissions:
_(units)
.(units)
(units)
UNCONTROLLED SIZE DISTRIBUTION
Category name:
Category number:
Particle size, pm
< 2.5 < 6
< 10
Generic distribution, Cumulative
percent less than or equal to:
Mass in size range, (units = tons/year):
CONTROLLED SIZE DISTRIBUTION
Type of control device:
Particle size,
0-2.5 2.5-6
6-10
Collection efficiency Table 4-1:
*
Mass in size range before control
(units=tons/year):
Mass in size range after control:
Cumulative mass:
Note that uncontrolled size data is cumulative percent less than.
Control efficiency data applies only to size range and is not cumulative,
B-l
-------
TECHNICAL REPORT DATA
(Please rud Instructions un the rct'crse before completing/
1. REPORT NO. I 2.
EPA-450/4-86-013 |
4. TITLE AND SUBTITLE
Generalized Particle Size Distributions For Use In
Preoaring Size Specific Particulate Emission Inventories
7. AUTHOR(S)
PEI Associates, Inc.
Golden, CO 80401
9. PERFORMING ORGANIZATION NAME AND ADDRESS
12. SPONSORING AGENCY NAME AND ADDRESS
Air Management Technology Branch
Monitoring And Data Analysis Division
Office Of Air Quality Planning And Standards
Research Triangle, NC 27711
3. RECIPIENT'S ACCESSION NO.
5 REPORT DATE
July 1986
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO?
68-02-3512
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA Project Officer: A. A. MacQueen
16. ABSTRACT
This accumulation of particle size data is intended to be of use to State and
local air pollution control agencies in the development of emission inventories.
In light of a proposed National Ambient Air Quality Standard for particulate, this
document is expected to be of help in the ensuing work on State Implementation Plans,
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Emission Inventories
Particle Size Data
State Implementation Plans
Particle Size Distribution
18. DISTRIBUTION STATEMENT
b. IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report]
20 SECURITY CLASS (This page;
c. COSATI Field/Croup
21. NO, OF PAGES
60
22. PBICE
EPA Form 2220—1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |