United States Industrial Environmental Research EPA-600/2-79-103
Environmental Protection Laboratory May 1979
Agency Research Triangle Park NC 27711
Research and Development
Iron and Steel Plant
Open Source Fugitive
Emission Evaluation
-------
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 ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, 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-600/2-7i-103
May 1979
Iron and Steel Plant Open Source
Fugitive Emission Evaluation
by
Chatten Cowherd, Jr., Russel Bohn, and Thomas Cuscino, Jr.
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Contract No. 68-02-2609, W.A. 3
Program Element No. 1AB604
EPA Project Officer; Robert V. Hendriks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
This report was prepared for the Environmental Protection Agency's
Industrial Environmental Research Laboratory under EPA Contract No. 68-
02-2609, Work Assignment No. 3. Mr. Robert V. Hendriks, Metallurgical
Processes Branch, was the requestor of this work.
The work was performed in the Environmental and Materials Sciences
Division of Midwest Research Institute. Dr. Chatten Cowherd served as
task manager and was the principal author of this teport. Mr. Russel
Bonn was responsible for source testing activities utilizing the exposure
profiling technique. Mr. Thomas Cuscino coordinated the moisture sampling
study and assisted in the wind erosion measurements. Mrs. Mary Ann
Grelinger and Mrs. Christine Maxwell were responsible for emission data
reduction. Mr. Reed Hodgin performed data analysis for the moisture
parameter study.
Dr. Dennis Lane of Kansas University also contributed to the moisture
parameter study. Dr. Dale Gillette of the National Center for Atomspheric
Research directed the operation of the NCAR wind tunnel.
May 14, 1979
iii
-------
CONTENTS
Preface* .*.,.«**«*.*«***.•«..**«***•••*« iii
Figures • vi
Tables vii
Summary and Conclusion. ....***«..«..**.«•*... 1
1*0 Introduction .........*........**..... 6
2*0 Selection of Sources and Test Conditions ...**.«.**. 12
2*1 Emission Contributions of Open Dust Sources *.***.« 12
2*2 Distribution of Source Tests .*.****... 16
2*3 Source Test Conditions. ................. 22
3.0 Source Testing by Exposure Profiling ............. 27
3*1 Sampling Equipment* *...* * 27
3.2 Sample Handling and Analysis* .............. 30
3.3 Results for Vehicular Traffic on Unpaved Roads. * «... 31
3*4 Vehicular Traffic on Paved Roads. ***..***«*.. 39
3*5 Storage File Stacking ........... 46
4*0 Wind Erosion Testing .................«.*. 59
4*1 Sampling Equipment* .*«•...**..•****•** 59
4.2 Preliminary Testing 59
4.3 Emissions Testing Program **.* **..... 62
5*0 Refinement of Emission Factor Equations* .*.*.*.**.. 71
5*1 Vehicular Traffic on Unpaved Roads * 71
5*2 Vehicular Traffic on Paved Roads .**... 73
5.3 Storage Pile Formation by Continuous Load-in
(Stacking) 80
6.0 Development of Storage Pile Silt and Moisture Values 85
6.1 Testing Program ....***.* * 85
6*2 Test Results—Intensive Study *...*.....**.. 86
6*3 Test Results—Extended Study 90
-------
CONTENTS (concluded)
7.0 Additional Research Needs, 97
7.1 Emission Inventory Handbook • 97
7.2 Unpaved Road Dust Controls 97
7.3 Wind Erosion of Exposed Aggregate Materials 98
8.0 References 99
9.0 Glossary 100
10.0 English to Metric Unit Conversion Table 104
Appendices
A. Emission Factor Calculation Procedures. ........ A-l
B. Procedures for Surface Aggregate Sampling and Analysis* .....
vi
-------
FIGURES
Number
SC-1 Revised MRI emission factor equations ,.,,.. 3
1-1 Quality assurance (QA) rating scheme for emission factors . 10
3-1 MRI exposure profiler 29
3-2 Sampling equipment layout for Runs F-21 through F-26. ... 33
3-3 Sampling equipment layout for Runs G-27 through G-29. ... 34
3-4 Sampling equipment layout for Suns G-30 through G-32. ... 35
3-5 Sampling equipment layout for Runs F-13 through F-15. ... 43
3-6 Sampling equipment layout for Runs F-16 through F~18. ... 44
3-7 Sampling equipment layout for Runs H-10 through H-12. ... 51
3-8 Sampling equipment layout for Runs F-19 and F-20. ..... 52
4-1 Emissions sampling module for portable wind tunnel 60
4-2 Average emission factor versus cumulative erosion time. . . 69
5-1 Predictive emission factor equation for vehicular traffic
on unpaved roads. . 72
5-2 Comparison of predicted and actual emissions—untreated
roads 74
5-3 Fine particle fractions of TSP emissions. 76
5-4 Effectiveness of road dust suppressants .......... 77
5-5 Predictive emission factor equation for vehicular traffic
on paved roads. ............... . 79
5-6 Predictive emission factor equation for storage pile
formations by means of conveyor stacker ......... 82
6-1 Observed storage pile moisture versus time of day 88
6-2 Storage pile moisture normalization factor (= 1.0 for
1400 hr 1JDT) , 89
6-3 Correlation of normalized moisture with weighted
precipitation—coal piles ......... 95
6-4 Correlation of normalized moisture with weighted
precipitation—iron pellet piles 96
B-l Location of incremental sampling sites on an unpaved road . B-4
B-2 Sampling data form for unpaved roads. ........... B-5
B-3 Location of incremental sampling sites on a paved road. . , B-7
B-4 Sampling data form for paved roads B-8
B-5 Sampling data form for storage piles. B-10
vii
-------
TABLES
Number Pag€
1-1 Emission Factor Quality Assurance Limitations
(Effective March 1978) , 9
2-1 Fugitive Dust Emission Factors Experimentally
Determined by MRI—Nonmetric Units ...... 14
2-2 Generic Source Contributions to Suspended Particulate
Emissions 17
2-3 Estimated Precisions of Available Emission Eate Input
Parameters » . 19
2-4 Expected Precisions in Emission lates. .......... 21
2-5 Emission Factor Correction Parameters for Open Dust
Sources. ........ , ..... 23
2-6 Source Test Conditions 24
3-1 Field Measurements . 28
3-2 Exposure Profiling Test Site Parameters. 32
3-3 Plume Sampling Data—Unpaved Roads 36
3-4 Suspended Particulate Concentration and Exposure
Measurements—Unpaved Roads. .............. 38
3-5 Particle Size Data—Dnpaved Roads (Density » 3 g/cm3)... 40
3-6 Isokinetic Correction Parameters—Unpaved Roads. ..... 41
3-7 Emission Factors and Adjustment Parameters—-Unpaved Roads. 42
3-8 Plume Sampling Data—Paved Roads ..... 45
3-9 Suspended Particulate Concentration and Exposure
Measurements—Paved Roads. 47
3-10 Particle Size Data—Paved Roads (Density - 3 g/cm^).... 48
3-11 Isokinetic Correction Parameters—Paved Roads. ...... 49
3-12 Emission Factors and Adjustment Parameters—Paved Roads. . 50
3-13 Plume Sampling Data—Storage Pile Stacking 53
3-14 Suspended Particulate Concentration and Exposure
Measurements—Storage Pile Stacking. .... 55
3-15 Particle Size Data—Storage Pile Stacking. ........ 56
3-16 Isokinetic Correction Parameters—Storage Pile Stacking. . 57
3-17 Emission Factors and Adjustment Parameters—Storage
Pile Stacking 58
4-1 Observed Threshold Velocities (June 12, 1978) 61
4-2 Wind Erosion Test Site Parameters. , , . . , 64
4-3 Wind Erosion Sampling Parameters . 65
viii
-------
TABLES (continued)
Number Page
4-4 Particle Size Data 66
4-5 Properties of Loose Surface Material ..... 67
4-6 Wind Erosion Test Results. . , 68
5-1 Predicted Versus Actual Emissions (Unpaved Roads) 73
5-2 Predicted Versus Actual Emissions (Paved Roads) 81
5-3 Predicted Versus Actual Emissions (Load-In by Stacker) . . 84
6-1 Surface Moisture Variation in Dormant Piles at Armco
Middletown Works 87
6-2 Storage Pile Moisture and Precipitation/Evaporation
(Annco, Inc., Middletown, Ohio) "91
6-3 Storage Pile Moisture and Precipitation/Evaporation
(Bethlehem Steel, Bethlehem, Pennsylvania) 92
6-4 Storage Pile Moisture and Precipitation/Evaporation
(Inland Steel, East Chicago, Illinois) 93
A-l 50% Cutoff Diameters for Sierra Cyclone Preseparator and
Cascade Impactor Operated at 34 m^/hr (20 cfm) . . . . . A-4
A-2 Example Calculation for Run G-29—Unpaved Roads. ..... A-6
A-3 Example Calculation for Run F-18—Paved Roads A-7
A-4 Example Calculation for Run H-12—Storage Pile Stacking. . A-8
B-l Moisture Analysis Procedures ..... B-13
B-2 Silt Analysis Procedures , . , . » B-14
ix
-------
SUMMARY AND CONCLUSIONS
This report presents the results of a field testing program aimed
at increasing the reliability of emission factors for open dust sources
within the integrated iron and steel industry. The predominant factor
limiting the quality assurance ratings of the emission factor equations
for open dust sources that were previously developed by MRI was the
restricted number of test measurements in relation to the number of
correction parameters appearing in each equation. Based on statistical
analysis of source contributions and the reliability of previously
developed emission factors, a source testing plan was developed.
Specifically, the following tests were performed at three integrated
iron and steel plants:
* Eighteen tests of vehicular traffic on untreated, unpaved roads.
* Two tests of vehicular traffic on treated unpaved roads.
* Six tests of vehicular traffic on paved roads.
* Five tests of storage pile stacking.
The primary tool for quantification of dust emissions from the
above sources was the MRI exposure profiler. Other equipment used in
the testing included cascade impactors with cyclone precollectors for
particle sizing, high-volume air samplers for determining upwind and
downwind particulate concentrations, and recording wind instruments used
to determine mean wind speed and direction for adjusting the MRI exposure
profiler to isokinetic sampling conditions.
For all of the emission tests, samples of the emitting materials
were collected for laboratory analysis to determine properties which
affect the emission rates. Unpaved and paved roads were sampled by
removing loose material (by means of vacuuming and/or broom sweeping)
from lateral strips of road surface extending across the traveled portion.
Storage piles were sampled to a depth exceeding the size of the largest
aggregate pieces. Pertinent equipment parameters (vehicle weight and
speed, stacker drop distance) were also recorded during each test.
-------
In addition, 12 tests of wind erosion emissions were performed
utilizing a portable wind tunnel with a specially designed isokinetic
sampling system. Eight tests were performed on the upper flat surface
of an inactive coal storage pile—three tests of one section of undisturbed
(crusted) surface and five tests of a disturbed section. This was
followed by two tests of the flat ground surface (undisturbed) adjacent
to a dolomite storage pile and two tests of disturbed prairie soil in
the same area. Both mass emission rates and particle size distributions
were measured as a function of tunnel wind velocity.
This study also addressed a special problem related to the determination
of storage pile surface moisture for aggregate materials. Surface
moisture is known to affect the rate of wind erosion of exposed materials.
Because of the high degree of variability of surface moisture in response
to daily evaporation cycles, as well as to precipitation events and
mechanical disturbances, it is desirable to develop empirical relationships
for daily and seasonal average surface moisture values as a function of
meteorological conditions and properties of stored aggregate materials.
Figure SC-1 presents the revised emissions factor equations developed
in this study for traffic entrained dust from unpaved roads, traffic-
entrained dust from paved roads and storage pile formation by means of a
translating conveyor stacker. These factors describe emissions of
.particles smaller than 30 pm in Stokes diameter.
Based on an expanded data set of 24 tests, the revised MRI emission
factor equation for traffic-entrained dust from unpaved roads predicts
measured emission factors with precision factor* of 1.48 as compared to
a precision factor of 1.66 for the unrevised equation. The addition of
a correction term related to the average number of wheels per vehicle
reduced the mean prediction error, as suggested by the clear tendency
of the unrevised equation to underpredict measured emission factors when
the test road was traveled by a substantial portion of 10- and 18-wheel
vehicles rather than 4- and 6-wheel vehicles.
Approximately 35% of measured road dust emissions in the suspended
particulate size range (particles smaller than 30 pm in diameter) consist
of fine particles (particles smaller than 5 ym in diameter) which have
the potential for transport over distances greater than a few kilometers
from the source. This fine particle fraction appears to be independent
of average vehicle weight and road surface composition.
The precision factor (f) is defined such that the 95% confidence inter-
val for a predicted emission factor value (P) extends from P/f to Pf.
-------
OWN DUSf SOUICEs
QA RATING:
EMISSION FACTOR -
METRIC:
{MISSION FACTCW -
NON-MEINC;
SYMtOlS AND UNITS
Vehicular Traffic an Unpaved Rs0*fa
S for Dry Condition*
C Far Annual Average Condi lioni
(V/ S\/W ^'^/w^0-5/ 4 1
TfJlwl^"?/ t"<7 laJs'J ''•/"•''•'""
«,,» t±\l±\(*f-7l~f-s
lf *•* litJbo/U. l«J
s ™ flit content 0f road lurfsca mafwriaj
S s mravQgt vahtel* fp«ad
w ^fmrogB number of wheoli p«r vvtiicfe
W */***""'
% %
km/tir mph
lonne> >0ni
Vehicubr Irs I fie an P(w«d Read
@ far Nortnal Urban Traffic
C Far Inrfujtriul Fiur.i Traffic
0.7
EF 0'»M"J(wM280/t77j
IF * 0.090 1 t~\ |-jjj| (-|^g) (™f ) * Ib/wet^ipi
rntlrtc ^nChiv mtl ric
EF » tuiptrnded parti cu late emiitiont kg/veh-km (t/veh-mi
n * ow*nbar of traffic tarns* -
i = iij* content of rocd surface maNrfc! % %
L " lurfava duit loading on traveled porHon kfl/km Ifa/mf
of rOfflrf
W * sserage vehicle **%ig!*f wnnet too
Storage Ptl* Formation by Meant of
8
||)(_u_)(£)
(T)
fiVJiWJl^
EF'O.OOII J V ^A lb/«n
Iff
material trantlenvd
s * illt eOfWNmf Of aggregate "^
M * moiit\jr* content of aggregate
-------
Limited testing of chemical dust suppressants for industrial unpaved
roads indicates a high initial control efficiency (exceeding 90%) which
decreases by more than 10% with the passage of 200 to 300 vehicles.
Consistent with the emission factor equation, the lowering of emissions
is reflected by the reduced silt content of the road surface material •
after the application of chemical dust suppressants. Additional testing
is needed to better quantify the performance of road dust suppressants.
Testing is also needed to verify and/or refine the emission factor
adjustment term which accounts for climatic mitigation.
The expanded test data set for traffic-generated dust from paved
roads indicates that the unrevised MR1 emission factor equation consistently
underpredicts emissions (by up to a factor of 7) for industrial paved
roads. This is thought to be due to the additional dust generation from
unpaved areas adjacent to the paved surface. Incorporation of emission
factor correction terms which account for emissions from unpaved shoulders
and for the number of traffic lanes, improves the precision factor from
14.1 to 3.31.
Modification of the MRI emission factor equation for continuous
drop operations (translating conveyor stacker) by the addition of a
linear correction term involving drop distance aids in improvidng the
predictive capability of the equation. However, predictive errors
remain significant, which indicates effects of complex physical phenomena
not accounted for in the emission factor equation.
The results of the wind erosion testing indicate that natural surface
crusts are very effective in mitigating suspended dust emissions. In
addition, test data show that a given surface has a finite potential for
erosion prior to additional mechanical disturbance. Erosion rates increase
with wind velocity and decrease with erosion time.
Based on the results of the testing program to determine moisture levels
in storage pile surface aggregate, it was found that daily moisture decreases
from an early morning maximum to an afternoon minimum, the rate depending on
the prevailing evaporation forces and the amount of fines in the stored
aggregate. Daily average storage pile moisture was found to be strongly
related to weighted precipitation over the previous 4 days. Active piles
were less sensitive than dormant piles to precipitation because of the
turnover of stored material.
It is recommended that an emission inventory handbook be developed for
open dust sources that would provide guidance on determining emission factor
correction parameters, source extent values, and control efficiencies, both
natural and anthropogenic. This would aid in minimizing inventory errors
associated with all factors which enter into emission rate calculations.
-------
It is also recommended that road dust controls be studied. Although
the emission factor equation for this major source has been reliable in
dealing with a wide range of uncontrolled source conditions, little is
known about the mitigative effects of natural controls and common industrial
control practices such as watering. By studying watering effectiveness,
optimization parameters for this control measure could be developed.
Significant information on the dynamics of wind erosion from storage
piles and bare ground areas has been developed in this study. Observed
phenomena associated with protective natural crusting and erosion potential
have not been incorporated into emission factor expressions that are com-
monly used to estimate dust emissions from wind erosion. More investiga-
tion is needed to define these phenomena.
-------
SECTION 1.0
INTRODUCTION
Iron- and steel-making processes, which are characteristically
batch or semicontinuoiis operations, generate substantial quantities of
fugitive (nonducted) emissions at numerous points in the process cycle.
There are numerous materials handling steps in the storage and preparation
of raw materials and in the disposal of process wastes. Additionally,
fugitive emissions escape from reactor vessels during charging, process
heating, and tapping.
Fugitive emissions in the iron and steel industry can be generally
divided into two classes—process fugitive emissions and open dust
source fugitive emissions. Process fugitive emissions include uncaptured
particulates and gases that are generated by steel-making furnaces,
sinter machines, and metal forming and finishing equipment, and that are
discharged to the atmosphere through building ventilation systems. Open
dust sources of fugitive emissions include such sources as raw material
storage piles, from which emissions are generated by the forces of wind
and machinery acting on exposed aggregate materials.
In a recent study of fugitive emissions from integrated iron and
steel plants, Midwest Research Institute determined that open dust
sources (specifically, vehicular traffic on unpaved and paved roads and
storage pile activities) ranked with steel-making furnaces and sinter
machines as sources which emit the largest quantities of fine and suspended
particulate, taking into account typically applied control measures. ±J
It became evident that open dust sources should occupy a prime position
in control strategy development for fugitive particulate emissions
within integrated iron and steel plants. Moreover, preliminary analysis
of promising control options for both process sources of fugitive emissions
and open dust sources indicated that control of open dust sources has a
highly favorable cost-effectiveness ratio for particulate.
The technical soundness of these conclusions and the foundation for
more detailed investigation rest on the availability of reliable particu-
late emission factors and particle size distributions for the sources
under consideration. In turn, fugitive emissions are especially difficult
to characterize for the following reasons:
6
-------
1. Emission rates have a high degree of temporal variability.
2. Emissions are discharged from a wide variety of source con-
figurations.
3. Emissions are comprised of a wide range of particle sizes,
including coarse particles which deposit immediately adjacent to the
source.
The scheme for quantification of emission factors must effectively deal
with these complications.
Since 1972, KRI has been engaged in a series of field testing
programs to develop emission factors for open dust sources associated
with agriculture and industry. To provide for the requirement that the
emission factors would be applicable on a national basis, at the outset
MRI analyzed the physical principles of fugitive dust generation to
ascertain the parameters which would cause emissions to vary from one
location to another. These parameters were found to be grouped into
three categories:
1. Measures of source activity or energy expended (for example,
the speed and weight of a vehicle traveling on an unpaved road).
2. Properties of the material being disturbed (for example, the
content of silt in the surface material on an unpaved road).
3. Climatic parameters (for example, number of precipitation-free
days per year on which emissions tend to be at a maximum).
By constructing the emission factors as mathematical equations with
multiplicative correction terms, the factors developed by MRI became
applicable to a range of source conditions limited only by the extent of
experimental verification.—*—*—
The use of the silt content as a measure of the dust generation
potential of a material acted on by the forces of wind or machinery
was an important step in extending the applicability of the emission
factor equations to the wide variety of aggregate materials of industrial
importance. The upper size limit of silt particles (75 /xm in diameter)
is the smallest particle size for which size analysis by dry sieving is
practical, and this particle size is also a reasonable upper limit for
particulates which can become airborne. Analyses of atmospheric samples
of fugitive dust indicate a consistency in size distribution so that ,
particles in specific size ranges exhibit fairly constant mass ratios.-^3-'
-------
In order to quantify source-specific emission factors, MRI developed
the "exposure profiling" technique, which uses the isokinetic profiling
concept that is the basis for conventional source testing.A' Exposure
profiling consists of the direct measurement of the passage of airborne
pollutant immediately downwind of the source by means of simultaneous
multipoint sampling over the effective cross section of the fugitive
emissions plume. This technique uses a mass-balance calculation scheme
similar to EPA Method 5 stack testing rather than requiring indirect
calculation through the application of a generalized atmospheric disper-
sion model.
The emission factors developed by MRI have been made specific to
particles smaller than 30 ^im in Stokes diameter, so that emissions may
be related to ambient concentrations of total suspended particulate.
The upper size limit of 30 urn for suspended particulate is the approximate
effective cutoff diameter for capture of fugitive dust by a standard
high volume particulate sampler (based on a typical particle density of
2 to 2.5 g/cm) .— It should be noted, however, that analysis of parameters
affecting the atmospheric transport of fugitive dust indicates that only
the portion smaller than about 5 jum in diameter will be transported over
distances greater than 5 to 10 km from the source.—'
In 1977, as noted above, MRI performed field testing of open dust
sources at two integrated iron and steel plants (designated as Plants A
and E) in order to extend the applicability of the previously developed
emissions factor equations to open dust sources in the iron and steel
industry.— The sources tested were: (a) light-duty vehicular traffic on
unpaved roads; (b) heavy-duty vehicular traffic on unpaved roads; (c) mixed
vehicular traffic on paved roads; (d) mobile stacking of lump iron ore;
(e) mobile stacking of pelletized iron ore; and (f) load-out of processed
slag into a truck with a front-end loader. These sources involved
materials handling equipment of a scale significantly larger than had
been tested previously. Criteria used in choosing the above sources for
testing included the relative importance of sources as determined from
plant surveys, the amenability of sources to accurate testing, and the
accessibility of sources for testing within the selected iron and steel
plants.
This report presents the results of a follow-up investigation aimed
at increasing the reliability of emission factors for open dust sources
within integrated iron and steel plants. As indicated in Table 1-1, the
predominant factor limiting the quality assurance ratings (Figure 1-1)
of the emission factor equations previously developed by MRI for open
dust sources is the restricted number of test measurements in relation
to the number of correction parameters appearing in each equation.
-------
TABLE 1-1. EMISSION FACTOR QUALITY ASSURANCE LIMITATIONS
(Effective March 1978)
Source category
Quality
assurance
rating
Test data limitations
Vehicular Traffic on Unpaved B
Roads - Dry Conditions
Vehicular Traffic on Unpaved C
Roads - Annual Conditions
Vehicular Traffic on Paved B
Roads - Normal Urban Traffic
Insufficient number of tests
Insufficient number of tests;
limited to dry surfaces
Insufficient number of tests
Vehicular Traffic on Paved C
Roads - Industrial Plant
Traffic
Storage Pile Formation by Means B
of Translating Conveyor Stacker
Transfer of Aggregate from
Loader to Truck
B
Insufficient number of tests;
probable effect of dust
resuspension from underbodies
Insufficient number of tests
Insufficient number of tests
Vehicular Traffic Around
Storage Piles
Wind Erosion from Storage
Piles
Wind Erosion of Exposed
Areas
Insufficient number of tests;
questionable measurement
accuracy
Insufficient number of tests;
questionable measurement
accuracy
Insufficient number of tests;
limited to dry uncrusted
surfaces
-------
QUALITY ASSURANCE RATING SCHEME
A = FORMULATION BASED ON STATISTICALLY REPRESENTATIVE
NUMBER OF ACCURATE FIELD MEASUREMENTS (EMISSIONS,
METEOROLOGY AND PROCESS DATA) SPANNING EXPECTED
PARAMETER RANGES
B = FORMULATION BASED ON LIMITED NUMBER OF ACCURATE
FIELD MEASUREMENTS
C = FORMULATION OR SPECIFIC VALUE BASED ON LIMITED
NUMBER OF MEASUREMENTS OF UNDETERMINED ACCURACY
OR
EXTRAPOLATION OF B-RATED DATA FROM SIMILAR PROCESSES
D = ESTIMATE MADE BY KNOWLEDGEABLE PERSONNEL
E = ASSUMED VALUE
Figure 1-1. Quality assurance (QA) rating scheme for emission factors,
-------
In addition, this study addresses a special problem related to the
determination of aggregate material surface moisture which affects the
rate of wind erosion of exposed materials. Because of the high degree
of variability of storage pile surface moisture in response to daily
evaporation cycles, as well as to precipitation events and mechanical
disturbances, it is desirable to develop empirical relationships for
daily and seasonal average surface moisture values as a function of
meteorological conditions and materials properties.
This report is organized by subject area as follows:
Section 2 presents the statistical plan used to select source
types and conditions for testing.
. Section 3 describes source testing procedures and results
of exposure profiling of emissions from (a) vehicular
traffic on unpaved roads; (b) vehicular traffic on paved
roads, and (c) storage pile stacking.
Section 4 describes procedures and results of wind
erosion testing utilizing a portable wind tunnel rather
than the exposure profiling apparatus.
Section 5 addresses the refinement of previously developed
emission factor equations through the incorporation of
data presented in Sections 3 and 4, and assesses the
reliability of refined emission factors.
. Section 6 describes the procedures and results of field
data collection and analysis to develop empirical relation-
ships for unbound moisture in storage pile surface materials.
. Section 7 outlines additional research needs.
Metric units with some non-metric equivalents are used in this
report. The word ton always refers to short ton (abbreviated "T") ,
which is equivalent to 2,000 Ib. The word tonne always refers to the
metric tonne (abbreviated "t"), which is equivalent to 2,200 Ib. An
English-to-metric conversion table follows Section 9.
11
-------
SECTION 2.0
SELECTION 01 SOURCES AND TEST CONDITIONS
This section presents the statistical plan for selection of sources
and test conditions. The objective of the statistical analysis was to
find the optimal distribution of tests over generic source categories
such that maximum precision in the estimated total emissions from open
dust sources within integrated iron and steel plants would be achieved.
Weighting factors are assigned to each source category by balancing
(a) relative source contributions to total particulate emissions and
(b) precisions of the previously developed emission factor equations, as
described below.
2.1 EmissionContributions of Open Dust Sources
This section presents the methodology used to estimate nationwide
emission contributions of open dust sources within integrated iron and
steel plants. For this purpose, sources were grouped by similarity of
physical mechanisms for dust generation. The following generic source
categories were considered in this test plan:
Vehicular Traffic on Unpaved Surfaces
Unpaved roads
Storage pile maintenance
. Vehicular Traffic on Paved Surfaces
Batch Drop Operations
Loaders
Railcars
Trucks
Gantry/clamshell buckets
Continuous Drop Operations
Stackers
Conveyor transfer stations
Bucket wheel barge unloading
12
-------
. Wind Erosion
Storage piles
Exposed areas
Table 2-1 lists the predictive emission factor equations previously
developed for each source category.—
The following subsections detail the procedure used to derive the
emission estimate for each generic source category.
2.1.1 Vehicular Traffic
Emission factors for light, medium, and heavy duty traffic on
unpaved roads were calculated using the emission factor equations in
Table 2-1. The values of the correction parameters were based on
averages from four open dust source surveys previously performed by MRI.—'
The emission factors were then multiplied by the average source extent
(vehicle miles traveled) which were calculated from the open dust surveys.
Finally, it was assumed that there were 50 major plants in the nation
producing the emission rate calculated for the average plant.
The emission factor for paved roads was calculated as the average
of two tests performed by MRI at an iron and steel plant. —' The emission
factor was then multiplied by the average source extent (vehicle miles
traveled) calculated from the open dust surveys. Finally, the emission
rate for paved road traffic at the average plant was multiplied by 50,
in order to extrapolate to nationwide emissions.
The emission factor used for storage pile maintenance was developed
from the emission factors calculated for four plants previously surveyed.
Separate weighted emission factors were determined for pellets and coal.
The weighted emission factors were multiplied by the 1976 nationwide
tonnage of these materials received at iron and steel plants. Finally,
the summed emission rate for pellets and coal was linearly scaled by
the ratio of all aggregate materials received to the sum of coal and
pellets received. In this manner, the total nationwide emission rate
for pile maintenance and other traffic associated with storage of all
aggregate material was calculated.
2.1.2 Batch and Continuous Drop Operations
The following assumptions were used in calculating emissions for
the batch and continuous drop categories:
1. Fifty percent of the aggregate material received in the average
plant arrives by barge and 50% by rail.
13
-------
TABLE 2-1. FUGITIVE DUST EMISSION FACTORS EXPERIMENTALLY DETERMINED BY MRI—HONMETRIC
Source category
Measure of extent:
Emission factor^/
fib/unit of source extent)
Correc tlon parameters
f-»
•e-
1. Vehicular Traffic on Onpaved Buds Vehicle-Males Traveled
£« Vehicular Traffic on Paved Rosds Vehicle— Miles Traveled
3 . Batch Load-In Tons of Material l U'
sand and gravel operations)^
d = Kuraber of Dry Days per Year
f - Percentage at Tine Kind Speed Exceeds 1Z mph
at height of 1.0 It
D - Duration of Material Storage {days)
e - Surface Erodibility (tons/acre.'»ear)
-E - Thomthtiaitea Preclpltatlon-Evaporattan In^gx
&/ HRI emission factor equations uei:e formerly presented only in nonmetrlc units*
W Annual average anls«lons of dust pBrticles smaller than 30 urn in diameter baaed on particle density of 2.5
-------
2. The 50% arriving by rail is batch unloaded.
3. Of the 50% arriving by barge, half is batch unloaded (gantry/
clamshell) and half is continuously unloaded (bucket wheel).
4. All aggregate passes through two transfer stations in its
lifetime at the average iron and steel plant.
5, Railcars have 100 tons capacity and haul aggregate with an
average density of 2.5 g/cm .
3
6. The average clamshell is 20 yd in volume,
7. The average truck has 50 tons capacity and hauls aggregate
with an average density of 2.5 g/cm .
8. The averages of the silts and moistures measured during the
previous open dust surveys of four plants are representative nationwide
values.
9. The loading into storage piles of all aggregate is apportioned
as follows: 10% dropped by truck, 10% dropped by loader, and 80% dropped
by stacker.
The two aggregates selected as representative of all aggregate
materials were coal and iron-bearing pellets. These particular materials
were selected for two reasonss (a) they represent over 50% of the total
aggregate stored at iron and steel plants; and (b) more data are available
on the silt and moisture of these materials than other aggregate materials
stored at iron and steel plants.
The silt and moisture measurements obtained during the open dust
surveys of the four plants surveyed were averaged in an attempt to
obtain representative nationwide values. For coal, the average silt and
moisture percentages were 4.4 and 2.8, respectively; and for pellets,
the average silt and moisture percentages were 8.6 and 1.1, respectively.
Based on the above assumptions and the average silt and moisture
values, 1976 nationwide emission rates for coal and pellet batch and
continuous drop sources were calculated. The sum of these emission
rates was then scaled linearly by the ratio of total aggregate receipts
to the sum of coal and pellet receipts. In this fashion, the emission
rates for total aggregate batch drop and continuous drop were calculated.
15
-------
2.1.3 Wind Erosion
The emission factors for wind erosion from pellet and coal piles
were weighted averages of emission factors calculated for four previously
surveyed plants. The weighting of each plant emission factor reflected
the mass of the material located at each plant.
The emission rates for coal and pellets were calculated by multiply-
ing the emission factors by the 1976 nationwide receipts at iron and
steel plants. The total emission rate for wind erosion from all aggregate
piles was calculated by linearly scaling the sum of the emission rates
for coal and pellets by the ratio of the total aggregate receipts to the
sum of the coal and pellet receipts.
The emission factor for wind erosion of bare areas was calculated
as a weighted average of the emission factors for the four previously
surveyed plants. The plant emission factors were weighted by source
extent (acres exposed).
The emission rate for the average plant was calculated by multiplying
the weighted average emission factor by the arithmetic average source
extent observed at the four previously surveyed plants. Finally, the
nationwide emission rate was obtained by multiplying the emission rate
for the average plant by 50, which is the number of major plants estimated
to exist in the country.
Table 2-2 gives the uncontrolled and the controlled 1976 suspended
particulate emission rates for the open dust source categories. The
typical control efficiencies are estimates of current practice, as
presented in the previous MRI report. —' The right-hand column shows the
percent contribution of each generic category to the nationwide dust
emissions from open dust sources within integrated iron and steel plants.
2.2 Distribution of Source Tests
This section describes the statistical methodology used to determine
the optimal distribution of 42 source tests over the five generic source
categories of open dust sources such that maximum precision is achieved
in estimating the total emissions from open dust sources. It was estimated
that 42 tests (including laboratory tests of wind erosion) could be
performed within the funding limit of this program.
The total controlled emission rate or inventory (I) is the sum of
contributions from five sources, namely:
16
-------
TABLE 2-2. GENERIC SOURCE CONTEIBUTIONB TO SUSPENDED PARTICULATE EMISSIONS
Estimated
typical 1976 Nationwide suspended
control partlculate emission rate
efficiency Uncontrolled
Source (I)
Vehicular traffic on unpaved surfaces
Unpaved roads
Storage pile maintenance
* Vehicular traffic on paved surfaces
• Batch drop operations
Loaders
Railcars
Trucks
Gantry/clatnshell
* Continuous drop operations
Stackers
Conveyor transfer stations
Bucket wheels
* Wind erosion
Storage piles
Exposed areas
50
40
50
0
0
0
0
40
50
0
40
40
(tons/vr) ,
48,800
12,600
15,000
100
110
43
120
1,270
3,170
400
8,500
3,000
(tonnes/yr)
44,300
11,400
13,600
91
100
39
109
1,150
2,880
363
7,710
2,720
Percent of
Controlled total controlled
(tons/yr)
24,400
7,600
7,500
100
110
43
120
760
1,580
400
5,100
1,800
49,513
(tonnes/yr)
22,100
6,890
6,800
91
100
39
109
690
1,430
363
4,630
1,630
emissions
65
15
1
5
14
-------
XI = Vehicular traffic on unpaved surfaces; this source has an
approximate weight Wl = 0.65;
X2 « Vehicular traffic on paved surfaces, W2 = 0.15;
X3 * Batch drop operations, W3 = 0.01;
X4 = Continuous drop operations, W4 = 0.05;
X5 - Wind erosion, W5 = 0.14.
Also, each X is the product of three components: e = emission
factor, S = source extent, C = (complement of) control efficiency. All
three of these components have an uncertainty about them, but this
"sampling error" can be approximated from prior work,* as given in
Table 2-3.
However, only the emission factors are sampled in the testing
program; the errors in S and C are irreducible. This means that the
precision of I has a lower bound that cannot be reduced by increasing
the sample size of emission factor determinations.
Given a total sample size n, it must be determined how many tests
to execute on each source category, i.e., how to efficiently allocate
the sample. The objective, of course, is to maximize the precision in
total emission factor. This is a standard problem in sampling theory,
and the resulting rule is:
W; "/vor (X^) (the Neyman allocation)
5
2 i
i = i
It is assumed in these calculations that e and S are distributed log-
normally and C is distributed binomially. In other words, an e or S
precision is known as "a factor of " while a C precision is
known as "+ %,"
18
-------
2-3. OF AVAILABLE
XI.
X2.
X3.
X4.
X5.
Source Emission factor Source extent
category precision* precision*
Vehicular traffic on
unpaved surfaces Factor of 1.5 Factor of 1.5
Vehicular traffic on
paved surfaces Factor of 3 Factor of 1.5
Batch drop operations Factor of 2 Factor of 1,5
Continuous drop
operations Factor of 3 Factor of 1.5
Wind erosion Factor of 5 Factor of 1.5
Range of typical
control efficiency
Conservative Optimistic
50 + 251 50 + 12 .5%
50 + 25% 50 + 12.5%
0 (constant) 0 (constant)
50 + 25% 50 + 12.57,
40 + 20% 40 + 107.
* 95% confidence level.
(Fractional \
1-control I
efficiency /
-------
The Neyman allocation for the proposed experiment yields:
Source Ideal % of Total Sample Size Proposed Sample Size
XI 36.8% . 15
X2 23.0% 9
X3 1.0% 0
X4 7.7% 6
X5 31.5% 12
42
In the project we are constrained to run tests in sets of three, and
also to run a minimum of six tests per source.
The entire set of 42 tests thus allocated should estimate the
plantwide emission factor to within + 11% (with 95% confidence), i.e.,
an average emission factor will be determined to within a factor of
about 1.1.
Unfortunately, the uncertainty in I will still reflect the uncertain-
ties in S and C, even though the emission factor precision has been
markedly improved. The precision (standard deviation) of a triple
product X^ « e^S^C^ follows a rms rule* which means, loosely speaking,
that the precision of X is determined by its "weakest link."
Table 2-4 illustrates these considerations explicitly. In 4a, the
expected precision for each source and the total emission rate are shown
based on 42 emission factor tests. In 4b, comparable results are shown
under the assumption of perfect emission factor values, i.e., the irreduc-
ible uncertainty in I due to uncertainties in S and C. For illustrative
purposes, 4c displays hypothetical precisions attainable if S and C were
known constants.
It is clear from Table 2-4 (by comparison of 4a and 4b) that increas-
ing the number of emission factor tests to infinity will allow only a
slight improvement in the source-specific and overall emission rate
precisions. This indicates the necessity for improving techniques
used to determine source extent and for quantifying actual control
efficiencies of commonly used emission control techniques. Current
uncertainty in control efficiency estimates is the limiting factor in
developing precise emission rate values. The precision values presented
in 4c are those achievable after the current testing program, based on
perfectly known source extents and control efficiencies.
_ 3
rel var (X1-X2-X3) ~ £ rel var (X.^ iff Z± independent.
20
-------
TABLE 2-4. EXPECTED PRECISIONS IN EMISSION RATES
a/
4a. Given n=42 tests for emission factors
Source
Precision a
Precision b
XI
X2
X3
X4
X5
Total
+ 67.7%
+ 97.5%
^
+ 110. 3%^
+ 113. 1%^'
+ 80.7%
+ 52.1%
± 87.3%
_
+ 101. 5%-/
+ 104.5%-
+ 67.2%
4b. Given perfectly known emission factors (n~»)
Source
Precision a
Precision b
XI
X2
X3
X4
X5
Total
64.4%
64.4%
40.5%
64.4%
64.4%
64.2%
+ 47.6%
+ 47.6%
+ 40.5%
+ 47.6%
+ 47.6%
+ 47.5%
4c. Given perfectly known source extents and control efficiencies (n.=k2)
Source
Precision^
XI
X2
X3
X4
X5
Total
+ 10.5%
+ 36.6%
+ 44.8%
+ 46.5%
+ 11.0%
Factor of 1.11
1.44
1.57
1.59
ji/ 95% confidence interval as a % of the estimated value (4- 2 cv). Pre-
cision "a" is with a conservative estimate of C uncertainty, while
Precision "b" uses a more optimistic guess*
t>/ Of course the minimum possible value physically is zero.
£/ The errors in particular emission factors are asymmetric, e,g., "a factor
of 2" rather than + 100%, etc. However, the total (or average) of error
is asymptotically symmetric, because it arises as a sum of five sources.
21
-------
2.3 Source Test Conditions
This section describes the rationale for selection of conditions
under which open dust sources were to be tested. Table 2-5 lists the
physical parameters which enter into the mechanisms for generation of
airborne dust from each generic source category. These parameters may
be used to adjust emissions estimates to properties of the materials
being disturbed, the equipment involved, and the meteorological conditions,
Quantified parameters are those which appear in the previously developed
emission factor equations, as given in Table 2-1.
Table 2-6 lists the proposed conditions under which sources were to
be tested according to the distribution of tests given on page 10. Test
conditions were selected based on the following:
1. Known or suspected importance of each parameter.
2, Parameter controllability.
3, Normal range of variation in each parameter across the steel
industry.
4. Typical values of parameters which are not highly variable
across the industry.
In short, parameter selection was made to maximize the applicability
of the emission factor equations to source conditions which are represen-
tative of the industry.
Although a total of 44 tests were performed pursuant to the plan
outlined in this section, adjustments in the mix of test conditions were
necessary as dictated by the availability of sources amenable to testing
within the industry. In addition, it was not always possible to test
under ideal wind and moisture conditions. A detailed explanation of
actual test conditions is presented in Sections 3 and 4.
22
-------
TABLE 2-5. EMISSION FACTOR CORRECTION PARAMETERS FOR OFEN DUST SOURCES
Correction parameters
Source category
Quantified
Unqualified
Vehicular traffic on
unpaved surfaces
Vehicular traffic on
paved surfaces
Batch drop operations
Continuous drop
operations
Wind erosion of
storage piles
and exposed areas
Surface silt content
Vehicle speed!/
Vehicle weights/
Surface silt content^./
ft/
Surface loading!
Silt content!*/
Moisture content
Wind speed
Loader eapacityS/
Silt content-
Moisture content
Wind speed
Surface erodibilityk/
Surface silt content^./
Wind speed
Surface moisture content
Surface loading
Surface moisture content
Particle density—/
Vehicle weights/
Vehicle speed3/
Surface moisture content
Particle density—/
Particle density!*/
Drop distance3-'
Belt width3./
Particle density-
Particle density—'
a_/ Controllable through equipment selection.
b/ Controllable through material selection.
£/ May be tied to vehicle weight and/or speed.
23
-------
TABLE 2-6. SOURCE TEST CONDITIONS
I. Vehicular Traffic onUnpaved Roads (Exposure Profiling^
A. Sampling locations (three roads)
(1) Slag haulage road with heavy-duty traffic
(2) Representative unpaved road with mixed traffic
(dirt-surfaced)
(3) Representative unpaved road with mixed traffic
(slag-surfaced)
B. Distribution of tests (15 total)
(1) Three tests - 15 passes of heavy-duty trucks on slag
haulage road
(2a) Three tests - 15 passes of vehicle mix on representative
road
(2b) Three tests - 15 passes of heavy-duty vehicles on
representative road
(2c) Three tests - 15 passes of medium-duty vehicles on
representative road
(3) Three tests - 15 passes of vehicle mix on second
representative road ,
C. Special conditions
(1) Use normal plant traffic and vehicle mix
(2) Test only dry, untreated road surfaces
(3) Restrict testing to periods of moderate winds
(2 to 7 m/see) of constant mean direction
(4) Locate sampling equipment about 5 m from edge
of road
II. Vehicular Traffic on Paved Roads (Exposure Profiling)
A. Sampling locations (three roads)
(1) Representative paved road with light surface dust
loading (< 300 kg/km)
(2) Representative paved road with medium surface dust
loading (300 to 1,500 kg/km)
(3) Representative paved road with heavy surface dust
loading (> 1,500 kg/km)
24
-------
TABLE 2-6. (Continued)
B. Distribution of tests (nine total)
Nine tests - 100 passes of vehicle mix on each test
road (three tests per road)
C. Special conditions
(1) Use normal plant traffic and vehicle mix
(2) Test only dry road surfaces that have not been
cleaned for at least one week
(3) Restrict testing to periods of moderate winds
(2 to 7 m/sec) of constant mean direction
(4) Locate sampling equipment about 5 m from edge
of road
III. Continuous Drop Operation - Storage Pile Stacking
A. Sampling locations (ore bedding area)
(1) Translating stacker for iron ore pellets
(2) Translating stacker for lump iron ore
(3) Alternative: translating stacker for coal
B. Distribution of test (six total)
(1) Three tests - 15 passes of pellet stacker
(2) Three tests - 15 passes of lump ore stacker
C. Special conditions
(1) Tests to begin when new pile is being formed
(2) Test only dry materials
(3) Restrict testing to periods of moderate winds
(2 to 7 m/sec) of constant mean direction
(4) Locate sampling equipment along edge of pile
area
IV, Wind Erosion
A. Sampling locations
Field site (ore bedding area and coal storage area) to
which NCAR* portable wind tunnel may be transported
* National Center for Atmospheric Research (Dr» Dale Gillette),
25
-------
TABLE 2-6. (Continued)
B. Distribution of tests (12 total)
(1) Six tests - iron ore erosion at three wind speeds
(two tests per speed)
(2) Six tests - coal erosion at three wind speeds (two
tests per speed)
C. Special conditions
(1) Horizontal surface of test materials required to
facilitate wind tunnel testing
(2) Testing under dry surface conditions (< 1% moisture
in aggregate)
26
-------
SECTION 3.0
SOURCE TESTING BY EXPOSURE PROFILING
This section describes the program of field testing using the
exposure profiling method to develop additional emission factor data for
open dust sources. Specifically, the following field tests were performed
at three integrated iron and steel plants:
Eighteen tests of vehicular traffic on untreated, unpaved
roads.
Two tests of vehicular traffic on treated unpaved roads.
Six tests of vehicular traffic on paved roads.
Five tests of storage pile stacking.
Table 3-1 specifies the kinds and frequencies of field measurements
that were conducted during each run. "Composite" samples denote a set
of single samples taken from several locations in the area; "integrated"
samples are those taken at one location for the duration of the run.
3.1 Sampling Equipment
The primary tool for quantification of emission rate was the MRI .
exposure profiler, which was developed under EPA Contract No. 68-02-0619.
The profiler (Figure 3-1) consists of a portable tower (4 to 6 m height)
supporting an array of sampling heads. Each sampling head is operated
as an isokinetic exposure sampler directing passage of the flow stream
through a settling chamber (trapping particles larger than about 50 fj,m
in diameter) and then upward through a standard 8 in. by 10 in. glass
fiber filter positioned horizontally. Sampling intakes were pointed
into the wind, and sampling velocity of each intake was adjusted to
match the local mean wind speed, as determined prior to each test.
Throughout each test, wind speed was monitored by recording anemometers
at two heights, and the vertical wind speed profile was determined by
assuming a logarithmic distribution.
27
-------
3-1. FIELD
Test Parameter
Quits
Sampling Mode
Me«»uremef>t Method
1. Meteorology
a. Wind speed
b. Wind direction
e. Cloud cover
d. Temperature
t. Relative humidity
2. Storage PI las
a.. Material type
bf Moisture concent
e. Dust texture
d. Material throughput
m/sec
deg
'%
"C
%
—
% moisture
X silt
tonnes
Continuous
Continuous
Sing It
Single
Stogie
Composite
Single
Composite
..
Recording instrument at "background"
station; sensors *t reference height
Visual observation
Sling psychrometer
Sling psychrometer
Determined by plant personnel
Oven drying
Dry sieving
Determined by plant personnel
3. Ro«d Surfaces
t. Pavement type -- Composite
b. Surface condition -- Composite
c. Dust loading g/m2 Multiple
d. Dust texture % silt Multiple
4. Vehicular Traffic
a. Mix — Multiple
b. Count " Cumulative
5. Suspended Oust
a. Exposure (versus height) mg/cm2 Integrated
b. Mass size distribution wn Integrated
c. Downwind concentration ug/m Integrated
d. Background concentration ug/gr Integrated
•. Duration af sampling »in Cumulative
Observation (photographs)
Observation
Dry vacuuming
Dry sieving
Observation (ear, truck, number of
•xles, etc,)
Automatic counters
Isokinetic high-volume filtration
(MBI method)
High-volume cascade impaction
High-volume filtration (EPA method)
High-volume filtration (IPA method)
Timing
28
-------
Figure 3-1. MRI exposure profiler.
29
-------
Sampling time was sufficient to provide sufficient particulate mass
and to average over several units of cyclic fluctuation in the emission
rate, e.g., vehicle passes on an unpaved road. The first condition was
easily met because of the proximity of the sampling grid to the source.
In addition to airborne dust passage (exposure), fugitive dust
parameters that were measured included suspended dust concentration and
particle size distribution. Conventional high-volume filtration units
were operated upwind and downwind of the test source.
A Sierra Instruments high-volume parallel-slot cascade impactor
with a 34 m /hr (20 cfm) flow controller was used to measure particle
size distribution along side of the exposure profiler. The impactor
unit was equipped with a Sierra cyclone preseparator to remove coarse
particles which otherwise would tend to bounce off the glass fiber
impaction substrates, causing fine particle measurement bias. The
cyclone sampling intake was directed into the wind, resulting in iso-
kinetic sampling for a wind speed of 5 mph.
In order to determine the properties of aggregate materials being
disturbed by the action of machinery or wind, representative samples of
the materials were obtained for analysis in the laboratory. Unpave'd and
paved roads were sampled by vacuuming and broom sweeping to remove loose
material from lateral strips of road surface extending across the
traveled portion. Storage piles were sampled to a depth exceeding the
size of the largest aggregate pieces.
3.2 Sample Handling and Analysis
To prevent dust losses, the collected samples of dust emissions
were carefully transferred at the end of each run to protective containers
within the MRI instrument van. High-volume filters from 'the MRI exposure
profiler and from standard high-volume units, and impaction substrates
were folded and placed in individual envelopes. Dust that collected on
the interior surfaces of each exposure probe was rinsed with distilled
water into separate glass jars. Dust was transferred from the cyclone
precollector in a similar manner.
Dust samples from the field tests were returned to MRI and analyzed
gravimetrically in the laboratory. Glass fiber filters and impaction
substrates were conditioned at constant temperature and relative humidity
for 24 hr prior to weighing, the same conditioning procedure used before
taring. Water washes from the exposure profiler intakes and the cyclone
precollectors were filtered after which the tared filters were dried,
conditioned at constant humidity, and reweighed.
30
-------
Samples of road dust and storage pile materials were dried to
determine moisture content and screened to determine the weight fraction
passing a 200-mesh screen, which gives the silt content. A conventional
shaker was used for this purpose. That portion of the material passing
through the 200-mesh screen was analyzed to determine the density of
potentially suspendable particles.
Table 3-2 gives the site parameters for the field tests conducted.
The following section describes the locations of the sampling instruments
at each test site and presents the results of the testing.
3.3 Results for Vehicular Traffic on Unpaved Roads
As indicated in Table 3-2, 21 tests of dust emissions from vehicular
traffic on unpaved roads were performed—18 tests of untreated roads and
3 tests of a 100-m road segment treated with Coherex® applied at 10%
strength in water. Arrangements for the application of Coherex®were
made by plant personnel as part of an internal study of dust suppressants.
The results of Test Series G-l to G-9 will not be reported because
unanticipated static charge problems created unreliable tare weights for
the high-purity glass fiber filters used for those tests. Figures 3-2
through 3-4 show the locations of sampling instruments relative to the
test road segments.
In addition to the silt content of the road surface material, the
emission factor equation (Figure 2-1) requires data on vehicle speed and
weight, averaged over the vehicle passes (approximately 50) accumulated
during a test. During each test, the speeds of vehicles passing the
sampling station were estimated by timing vehicles over a known travel
distance. Estimates of vehicle weights were obtained from plant personnel.
In some tests, the vehicle passes sampled were dominated by controlled
test vehicles traveling at preselected speeds.
Table 3-3 lists, for each run, the individual point values of exposure
exposure (net mass per sampling intake area) within the fugitive dust
plume as measured by the exposure profiling equipment. Also given are
the point values of filter exposure consisting only of particulate
collected by the filter following the settling chamber. Finally, the
integrated exposure value is given for each run.
Table 3-4 compares particulate concentrations measured by the
upwind hi-vol and by three types of downwind samplers (exposure profiling
head, standard hi-vol, and high-volume cascade impactor) located 5 m from
the test road and near the vertical center of the plume at a height of
2 m above ground. For the interpolated profiler concentrations, both
nonisokinetic and isokinetic values are given.; Also indicated are hi-
vol concentrations measured at distances further downwind.
31
-------
TABLE 3-2. EXPOSURE PROFILING TEST SITE PAMMETEES
Source Run
, G-l
Unpaved Roads
(crushed slag) <
G-2
G-3
G-4
G-5
G-6
C-7
C-8
G-9
Unpaved loads 1 F— 21
(dirt/crushed | F-22
slag) j F-23
Unpaved Roads
(Coherex
treated dirt;/
F-24
F-25
U iJjS
crushed slag)
!G-27
G-28
5-29
0-30
0-31
0-32
Paved Koads
(light surface
dust loading)
Paved Roads
(moderate
surface dust
F-13
F-14
F-1S
F-16
F-17
F-18
loading)
Iron Pellet
Stacking
H-10
M-ll
1-12
Coal Stacking 1 ^-19,
tF-20-
Date
6/20/78
6/20/78
6/20/78
6/21/78
6/21/78
6/21/78
6/22/78
6/22/78
6/22/78
S/ 1/78
S/ 1/78
8/ 1/78
8/ 2/78
8/ 2/78
8/ 2/78
8/ 7/78
8/ 7/78
8/ 7/78
8/ 8/78
8/ 8/78
8/ 8/78
7/18/78
7/18/78
7/18/78
7/19/78
7/19/78
7/19/78
6/29/78
6/29/78
6/29/78
7/20/78
7/20/78
Start
time
1126
1252
1417
1107
1515
1604
1348
1440
1507
1028
1117
1142
1045
1305
1257
1409
1444
1016
1112
1316
1159
1350
1509
1344
1438
1531
1011
1125
1226
1250
1407
Exposure
sampling
duration
ton)
44
42
43
25
36
17
14
14
14
32
12
33
30
23
60
25
27
40
38
40
83
60
51
36
34
82
17
14
8
57
15
No. of
vehicle
passes
or weight
transferred
50
40
30
42
41
177
26
32
28
40
50
68
123
100
74
52
78
46
57
68
88
123
47
66
61
96
672 t (741 T)
183 t (202 T)
374 t (412 T)
121 t (133 T)
0 t (0 T)
Meteorology
Ambient air
82
86
86
71
77
77
77
79
78
79
87
87
83
89
80
75
76
78
84
89
85
92
88
90
90
93
82
86
86
93
96
Mean wind
(m/sec) <
___
2
1
-
2
2
4
4.5
3
2
1
2
2
2
2
2
2
6.3
4.5
5.4
2
2
1
1
1
1
0.5-1
2
3
1
1
tnph)
.
4
3
-
5
5
9
10
7
4
3
4
4
4
5
5
5
14
10
12
4
4
2
3
3
3
1-2
4
6
3
3
a/ Measured at 1«5 and 4.5 m above ground.
t>/ Background test; no coal stacked.
32
-------
F-24, F-25 ^£=
F-21, F-22
F-23
Section of Road \
Treated with Coherex x
for Runs F-24 and
F-25
SAMPLING INSTRUMENTATION
Symbol
fl
0
A
A
*
Instrument
Profiler
Cascade Impactor
Downwind Hi-Vol
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2.0
2.0
4.0
10
Wind
Direction
20 30 40 50
Figure 3-2. Sampling equipment layout for Runs F-21 through F-26.
33
-------
SAMPLING INSTRUMENTATION
Symbol
a
o
A
A
*
Instrument
Profiler
Cascade Impactor
Downwind Ht-Vol
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2.0
2.0
4.0
10
20 30 40 50
Meters
Wind
Direction
Figure 3-3, Sampling equipment layout for Runs G-27 through G-29,
34
-------
SAMPLING INSTRUMENTATION
Symbol
a
0
A
A
*
Instrument
Profiler
Cascade Impactor
Downwind Hi-Vol
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2.0
2.0
4.0
10
20 30
40
50
i
Meters
Wind
Direction
Figure 3-4. Sampling equipment layout for Runs G-30 through G-32.
35
-------
TABLE 3-3. PLUME SAMPLING DATA—UNPAVID ROADS
Bun
F-21
F-22
f-23
F-24
P-25
G-27
028
Sampling
height
Cm)
1.5
3.0
4.5-
6.0
1.5
3.0
4.5
6.0
1.5
3.0
4.5
6»0
1.5
3.0
4.5
6.0
1.5
3.0
4.5
6.0
1.5
3.0
4.5
6.0
1.5
3.0
4.5
6*0
Sampling
rate
(mVfar)
27
29
27
26
27
29
27
26
27
29
27
26
27
29
27
26
27
29
27
26
27
27
NA
26
27
27
NA
26
(cfia)
16
17
16
15
16
17
16
15
16
17
16
15
16
17
16
15
16
17
16
15
16
16
NA
15
16
16
NA
15
Total
exposure
(nig/ em2)
1.36
0.69
0.32
0.38
1.02
0.79
0.45
0,37
1.59
1.14
0.77
0.61
0.067
0.041
0.043
0.047
0.34
0.26
0.11
0.24
3.66
9.29
HA
1.84
4.48
3.02
NA
1.24
Filter
exposure
(mg/aa2)
0.79
0.39
0.17
0.13
0.60
0.47
0.24
0.17
1.01
0.72
0.45
0.34
0.048
0.039
0.032
0.031
0.20
0.13
0.08
0.12
4.69
3.12
NA
1.07
2.26
1.60
NA
0.77
Integrated
filter
exposure
(kg/tan) (lb/ni)
34.7 123
27.7 98.3
47.4 168
3.35 11.9
10.0 35.3
234 901
110 391
36
-------
TABLE 3-3 (Continued)
Run
Sampling
height
(m)
Sampling
rate
-------
TABLE 3-4. PARTICULATE CONCENTM.TIOH AND EXPOSURE
MEASUREMENTS--UNPAVED ROADS
00
Particulate concentration (Mg/m3) at
Run
F-21
1-22
1-23
F-24
F-25
G-27
G-28
G-29
G-30
G-31
G-32
Upwind
background
925
925
925
498
433
50
50
50
1,165
1,165
764
Profiler**]^
Non-isokinetlc
2,920
5,330
3,320
537
1,175
8,470
9,180
10,900
5,930
4,720
10,100
Isokinetic
3,590
7,410
3,980
618
1,330
9,740
10,470
13,600
5,210
4,860
10,100
2 n above ground
Downwind
Cascade
impactor—
2,990
4,820
£/
866
1,200
12,300
6,980
11,400
6,400
5,270
12,300
Standard hi-rol
5m 20 m 50 m
2,140 1,330
4,390 1,800
3,260 1,610
629 452
782 641
10,100
7,040
9,570
3,840
4,060
8,190
a/ Interpolated from 1.5 m and 3.0 m concentrations.
to/ Positioned at 5 m downwind.
£/ Invalid data; improperly sequenced stages.
-------
Table 3-5 summarizes the particle sizing data for the unpaved road
tests. Particle size is expressed as Stokes diameter based on actual
density of silt-size particles. In addition to data from the cascade
impactor measurements, Table 3-5 also gives the average percent of the
exposure measurement consisting of filter catch weighted by the individual
exposure values for each run.
Table 3-6 gives the wind speed and intake velocity used to calculate
the average isokinetic ratio for each run. Also presented are isokinetic
correction factors for exposure and concentration, calculated from the
particle size data and isokinetic ratio values for each run according to
the procedure delineated in Appendix A.
Table 3-7 presents the isokinetic emission factors for suspended
particulates, particles smaller than 30 pm in Stokes diameter, and for
fine particulates, particles smaller than 5 fim in Stokes diameter. Also
indicated in Table 3-7 are vehicle and site parameters which are believed
to have a significant effect on observed emission rates.
An example emission factor calculation based on data for Run G-29
is given in Appendix A.
3.4 Vehicular Traffic on Paved Roads
As indicated in Table 3-2, six tests of dust emissions from vehicular
traffic on paved roads were performed. Figures 3-5 and 3-6 show the
locations of sampling instruments relative to the test road segments.
In addition to the silt loading on the road surface material, the
emission factor equation (Table 2-1) requires data on the number of
traffic lanes and the vehicle weight averaged over the vehicle passes
(approximately 50) accumulated during a test. Estimates of vehicle
weights were obtained from plant personnel. In some tests, the vehicle
passes sampled were dominated by controlled test vehicles traveling at
preselected speeds.
Table 3-8 lists the individual point values of exposure (net mass
per sampling intake area) within the fugitive dust plume as measured by
the exposure profiling equipment for each run. Also given are the point
values of filter exposure consisting only of particulate collected by
the filter following the settling chamber. Finally, the integrated
exposure value is given for each run.
39
-------
3-5. DATA— ROADS (Density = 3 g/emj)—
Cascade impactor
Mass median Percent Percent
Run diameter (fim) <30 fim < 5 pm Ratio^-
F-21
F-22
F-23^'
F-24
F-25^'
G-27
G-28
G-29
G-30
G-31
G-32
10.0
11
8.2
17
22
20
38
26
21
74
72
83
61
56
59
45 -
53
59
34
31
36
26
33
21
17
22
21
0.46
0.43
0.44
0.43
0.59
0.36
0.38
0.42
0.36
Profiler
Percent Weighted average 7.
>50 /xm captured on the filter
17
18.5
10
29
34
29
45
37
30
55
56
56
74
57
55
53
57
39
44
49
a! Based upon previous MRI testing..!/
b/ Percent < 5 urn -r- percent < 30 /urn.
c/ Data invalid} improperly sequenced stages.
d/ Data invalid; insufficient substrate loadings.
-------
TABLE 3-6. ISOKINETIC CORRECTION PARAMETERS—UNPAVED ROADS
Run
F..-21
F-22
F-23
F-24
F-25
G-27
G-28
G-29
G-30
G-31
G-32
Ht =
(cm/
sec)
134
99.1
144
142
170
198
210
183
541
408
452
Wind
1.5 m
(fpm)
263
195
283
279
335
389
414
360
1,065
803
890
speed
Ht =
(cm/
sec)
210
185
226
202
197
260
266
238
702
514
597
Intake velocity
4.5 m
(fpm)
414
365
445
398
387
512
524
468
1,381
1,012
1,175
Ht =
(cm/
sec)
293
293
293
293
293
293
293
293
420
420
420
1.5 m
(fpm)
576
576
576
576
576
576
576
576
827
827
827
Ht =
(cm/
sec)
293
293
293
293
293
293
d/
i/
d/
d/
i/
4.5 m
(fpm)
576
576
576
576
576
576
-
-
-
-
—
Isokinetic
ratioS/
1.79
2.27
1.66
1.76
1.60
1.30
1.39
1.60
0.777
1.03
0.929
Isokinetic
correction factor
Exposure
0.725
0.679
0.760b/
0.677
0.7072/
0.896
0.892
0,874
1.05
0.981
1.0
Concen-
tration
1.23 •
1.39
1 . 20£/
1.15
1.15
1.14
1.25
0.878
1.03
1.0
a/ Intake velocity i wind speed.
b/ Based on averages of particle size data from Runs F-21 and F-22.
c/ Based on particle size data from Run F-24.
d/ Flow controller not functioning properly.
-------
3-7. AID -
Road surface
material
Run
¥-21}
F-22j
F-23,
F-24]
F-25
F-26]
G-27]
G-28
G-29
G-30
G-31
G-32
Silt
Type (7.)
(Dirt/crushed 9
slag 9
9
Coheres 0
treated dirt/ 0
crushed slag
5
5
5
> Crushed slag 4
4
4
.0
.0
.0
.03
.02
.3
.3
.3
.3
.3
.3
Mean
vehic le
speed
(km/
hr)
24
24
24
24
24
35
37
39
40
47
35
(mph)
15
15
15
15
15
22
23
24
25
29
22
Mean Mean number Suspended
vehicle of wheels particulate
weight per vehicle emission factor
(tonnesXtons)
3 3
3 3
4 4
3 3
3 3
15 17
11 12
8 9
13 14
7 8
27 30
pass
4
4
4
4
4
11
10
8
9
6
13
(kg/VKT)
0
0
0
0
0
3
2
1
2
1
4
.85
.48
.65
.021
.101
.4
.0
.6
.5
.4
.5
(Ib/VMf)
3.0
1.7
2.3
0.073
0.36
12.0
7.2
5.6
8.7
5.1
16.0
Fine
particulate
emission factor
(kg/VKT) (Ib/VMT)
0.29
0.15
0.21§/
0.0076
Q.036^/
0.88
0.66
0.34
0.43
0.31
0.95
1.0
0.53
0.755/
0.026
3.1
2.4
1.2
1.5
1.1
3.4
a/ Based on averages of particle size data from Runs F-21 and F-22.
b/ Based on particle size data from Run F-24.
-------
Paved
Slag Hauling
Road
SAMPUNG INSTRUMENTATION
Symbol
&
O
A
A
*
Instrument1
Profiler
Cascade Impactor
Downwind Hi-Vol
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.0, 2.0,
3.0, 4.0
2.0
2.0
2.0
4.0
Unpaved
Parking
Lot
0
r
10 20 30 40
i i i i
50
i.
Meters
N
Wind
Direction
Figure 3-5. Sampling equipment layout for Runs F-13 through F-15.
43
-------
Unpdved Road
'•^
Storage Area
SAMPLING INSTRUMENTATION
Symbol
fl
0
A
A
*
Instrument
Profiler
Cascade Impactor
Downwind Hi-Vol
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2.0
2.0
4.0
Wind
Direction
10
20 30 40 50
Meters
Figure 3-6, Sampling equipment layout for Runs F-16 through F-18,
44
-------
TABLE 3-8. PLUME SAMPLING DATA—PAVED ROADS
Bun
F-13
f-14
F-15
P- 16
w-n
F-18
Sampling
height
(m)
1.0
2.0
3.0
4.0
1.0
2.0
3.0
* 4.0
1.0
2.0
3.0
4.0
1.5
3.0
4.S
6.0
l.S
3.0
4.5
6.0
1.5
3.0
4.5
6.0
Sampling
race
(m3/hr) (cfn)
24
27
24
22
22
27
24
22
22
27
24
22
27
29
27
26
27
29
27
26
27
29
27
26
14
16
14
13
13
16
14
13
13
16
14
13
16
17
16
IS
16
17
16
15
16
17
16
15
Total
exposure
(mg/cm2)
0*60
0.36
0.33
0.32
0.31
0.20
0.13
0.11
0.20
0.05
0.02
0.15
2.19
1.60
0.99
0.42
1.70
1.44
0.87
0.37
0.44
0.40
0*31
0.28
Filter
exposure
(rag/ en )
0,24
0,17
0.16
0.16
0.18
0.12
0.10
0.08
0,12
0.04
0.02
0.07
1.57
1,09
0.66
0.33
1,29
0,98
0.60
0.27
0.30
0.29
0.23
0.20
Integrated
filter
exposure
Ocg/ta) (Ib/mi)
10.5 37.2
6.14 21.8
4.14 14.7
68,8 244
58.9 209
18,9 67.0
45
-------
Table 3-9 compares particulate concentrations measured by the
upwind hi-vol and by three types of downwind samplers (exposure profiling
head, standard hi-vol, and high-volume cascade impactor) located 5 m
from the test road and near the vertical center of the plume at a height
of 2 m above ground. For the interpolated profiler concentrations,, both
nonisokinetic and isokinetic values are given,! .Also indicated are hi-
vol concentrations measured at distances further downwind.
Table 3-10 summarizes the particle sizing data for the paved road
tests. Particle size is expressed as Stokes diameter based oa actual
density of silt-size particles. In addition to data from the cascade
impactor measurements. Table 3-10 also gives for each run the average
percent of the exposure measurement consisting of filter catch, weighted
by the individual exposure values.
Table 3-11 gives the wind speed and intake velocity used to calculate
the average isokinetic ratio for each run. Also presented are isokinetic
correction factors for exposure and concentration, calculated from the
particle size data and isokinetic ratio values for each run according to
the procedure delineated in Appendix A.
Table 3-12 presents the isokinetic emission factors for suspended
particulates,, particles smaller than 30 /itm in Stokes diameter, and for
fine particulates, particles smaller than 5 /Ltm in Stokes diameter. Also
indicated in Table 3-12 are and site parameters which are believed to
have a significant effect on observed emission rates.
An example emission factor calculation based on data for Run F-18
is given in Appendix A.
3.5 Storage Pile Stacking
As indicated in Table 3-2, four tests of dust emissions from storage
pile formation by means of a mobile conveyor stacker were performed,
three tests of iron pellet stacking and one test of coal stacking. For
each test, the stacking arm was passed back and forth in front of the
profiler so that the sampler configuration was the same as that used for
roads (moving point source configuration). Figures 3-7 and 3-8 show the
locations of sampling instruments relative to the stacking strips.
Table 3-13 lists the individual point values of exposure (net mass
per sampling intake area) within the fugitive dust plume as measured by
the exposure profiling equipment for each run. Also given are the point
values of filter exposure consisting only of particulate collected by
the filter following the settling chamber. Finally, the integrated
exposure value is given for each run.
-------
TABLE 3-9. SUSPENDED PARTICULATE CONCENTRATION AND EXPOSURE
Particulate Concentration (ug/nH) at
Run
F-13
F-14
F-15
*•
•^ F-16
F-17
F-18
Upwind
Background
134
134
134
1,520
1,520
920
Profiler*/
Non-Isokinetic
383
327
195
4,620^7
4,170^
1,170^
Isokinetic
433
32?
310
6,840
7,590
1,540
2 m Above Ground
Downwind
Cascade
Impactor 5 m
429 201
323 279
360 243
3,900 2,850^7
8,130 3,760^
1,180 722^7
Standard Hi-vol
20 m
288
234
81
1,700
1,470
-
50 m
211
176
181
-
-
-
aj Interpolated from 1.5 m and 3.0 in concentrations.
b/ 3m downwind.
-------
TABLE 3-10. PARTICLE SIZE DATA—PAVED ROADS (Density = 3 g/cra3)5'
Cascade impact or
Run
F-13
F-14
F-15*/
F-16
F-17
F-18
Mass median
diameter (pro)
14
4.3
_
12.5
50
9.0
Percent
<30 jam
69
97
_
68
41
78
Percent
<5 nn
27
56
—
31
14
36
Percent
Ratlot/ >50 pm
0. 39 21
0.58 0.9
_ —
0.46 23
0.34 50
0.46 15
Profiler
Weighted average %
captured on the filter
44
65
66
71
72
72
aj Based upon previous MRI test ing .A'
jb/ Percent < 5 jum ~ percent < 30 jum.
c/ Data invalid; insufficient substrate loadings.
-------
TABLE 3-11. ISOKI1ETIC
\o
Wind speed
Run
F-13
F-14
F-15
F-16
F-17
F-18
Ht = 1.5 m
(cm/
sec) (fpm)
174k/ 343l/
148k/ 29lk/
62.sk/ 123k/
113 223
116 228
119 234
Ht = 4.5 m
(cm/
sec) (fpm)
222S/ 43?£-/
219— / 43 2£/
65.5— / 129— /
160 315
170 334
163 320
Intake velocity
Ht = 1.5 m
(cm/
sec)
(fpm)
Ht = 4.5 m
(cm/
sec)
256k/ 503k/ 256£/
(fpm)
5032./
237k/ 46?k/ 256S./ 503£/
237k/
293
293
293
467k/
576
576
576
256£/
293
293
293
503£/
576
576
576
Isokinetlc
ratio*-/
1.
1.
3.
2.
2.
2.
31
38
85
21
13
13
Isokinetic
cor rec t ion f ac tor
Concen-
Exposure tration
0
0
0
1
0
0
.873
.744
.4I2*/
.16
.887
.633
1.13
1.00
1.59i/
1.48
1.82
1.32
a/ Intake velocity 4 wind speed,
b/ At 1.0-m height.
cj At 3.0-m height.
d/ Based on averages of particle size data from Runs F-13 and F-14.
-------
TABLE 3-12. EMISSION FACTORS AND ADJUSTMENT PARAMETERS—PAVED ROADS
Road surface
material
Loading
Run
F-13
g F-14
F-15
F-16
r-17
P-18
Slit (1)
13.2
13.2
13.2
6,8
6.8
6.8
(kg/tan) (lb/mt)
57.2
57.2
57.2
627
627
627
203
203
203
2,225
2,225
2,225
Mean Mean Mean number Suspended
vehicle vehicle of wheels part ten late
a peed weight per vehicle emission factor
(roph ) ( tonnea ) { ton a )
7
» 5
5
12
11
5
8
5
5
13
12
5
pass (kg/VKT)
0.16
0.056
0.045
0.70
0.48
0.14
(Ib/WI)
0.58
0.20
0.16
2.5
1.7
0.48
Fine
parttculate
eraleston
(fcg/VKT)
0.043
0.031
0.0191/
0.22
0.067
0.050
factor
Clb/VMT)
0.16
0.11
0.0665/
0.78
0.24
0.17
a/ Baaed on averages of particle size data from Runs F-13 and F-14.
-------
X
Run H-12
Instrumentation
SAMPLING INSTRUMENTATION
Symbol
a
O
A
A
*
Instrument
Profiler
Cascade Impactor
Downwind HI-VoI
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2,0
2.0
4.0
Stacker Boom Length = 67 m
Run 10 Drop Height - 9 m
Run 11 Drop Height = 11 m
Run 12 Drop Height * 12 m
10
20
30
i
40 50
Meters
Run H-10
Instrumentation
Wind
Direction
RunH-11
Instrumentation
Run 10
Figure 3-7, Sampling equipment layout for Runs H-10 through H-12.
51
-------
Stacker Length ~10 m
Drop Height • 5 m
SAMPUNG INSTRUMENTATION
Symbol
fi
O
A
A
*
Instrument
Profiler
Cascade Impacfror
Downwind Hi-VoI
Upwind Hi-Vol
Wind Station
Sampling
Height (m)
1.5, 3.0,
4.5, 6.0
2.0
2.0
2.0
4.0
0 10 20 30
i'ii
40 51
Meters
N
Wind
Direction
Figure 3-8. Sampling equipment layout for Runs F-19 and F-20.
52
-------
TABLE 3-13. PLUME SAMPLING DATA—STORAGE PILE STACKING
Sun
H-10
H-ll
H-12
F-19
F-20S/
Sampling
height
(a)
i,5
3.0
4.5
6.0
1,5 '
3.0
4.5
6.0
1.5
3.0
4.5
6.0
1.3
3.0
4.5
6.0
1.3
3,0
4.5
6.0
Sampling
race
26
24
24
22
26
24
24
22
26
24
24
22
27
31
27
26
29
29
27
26
» (cfin)
15
14
14
13
15
14
14
13
15
14
14
13
16
18
16
IS
17
17
16
15
Total
exposure
(iag/ on2)
12.1
5*88
3.18
4.13
0.92
0.74
0.50
0.10
3,45
1.15
1.11
1.82
0.82
0.34
0,35
0.27
0.23
0,19
0.24
0.21
Filter
exposure
(«g/eni2>
2,43
1.43
0.39
2.56
0.42
0.62
0.46
0.09
1.88
0.35
0.80
1.59
0.42
0.21
0.19
0.15
0.09S
0.084
0.062
0.062
Integrated
filter
exposure
(kg/to) (ib/ai)
89.9 319
25.6 90,8
56.9 202
18.5 65. &
5.44 19.3 '
£/ Background run only*
53
-------
Table 3-14 compares particulate concentrations measured by the
upwind hi-vol and by three types of downwind samplers (exposure profiling
head, standard hi-vol, and high-volume cascade impactor) located 5 m
from the test road and near the vertical center of the plume at a height
of 2 m above ground. For the interpolated profiler concentrations, both
nonisokinetic and isokinetic values are given. Also indicated are hi-
vol concentrations measured at distances further downwind. . ; ,
Table 3-15 summarizes the particle sizing data for the storage pile
stacking tests. Particle size is expressed as Stokes diameter based on
actual density of silt-size particles. In addition to data from the
cascade impactor measurements, Table 3-15 also gives the average percent
of the exposure measurement consisting of filter catch, weighted by the
individual exposure values for each run.
Table 3-16 gives wind speed and intake velocity used to calculate
the average isokinetic ratio for each run. Also presented are isokinetic
correction factors for exposure and concentration, calculated from the
particle size data and isokinetic ratio values for each run according to
the procedure delineated in Appendix A.
Table 3-17 presents the isokinetic emission factors for suspended
particulates, particles smaller than 30 fj.m in Stokes diameter, and for
fine particulates, particles smaller than 5 /im in Stokes diameter. Also
indicated in Table 3-17 are vehicle site parameters which are believed
to have a significant effect on observed emission rates.
An example emission factor calculation based on data for Run H-12
is given in Appendix A.
54
-------
Ul
Ul
TABLE 3-14. SUSPENDED PARTICULATE CONCENTRATION AND EXPOSURE
MEASUREMENTS—STORAGE PILE STACKING
Farticulate concentration (M8/m ) at 2 m above ground
Run
H-10
H-ll
H-12
F-19
F-20£/
Downwind
Upwind Profilezl^ Cascade
background Non -isokinetic Isokinetic impactor
670 36,900 150, 000-' 39,700
700 4,580 5,360 4,860
800 21,600 25,300 13,400
630 1,280 2,620 2,500
630 1,400
Standard Hi-vol
5m 20m 50m
11,400
3,990
8,560
452 636 500
1,110 606 528
a/ Interpolated from 1.5 m and 3,0 m concentrations.
b/ Suspect because of large isokinetic ratio.
c/ Background run only.
-------
3-15. PARTICLE DATA--SIOMG1 PILE STACKING
Ul
Particle Cascade Imjjaetor
Run
H-10
H-ll^
H-12
F-19
O./
J&mm ? jfl=-
a./ Based
Density Median Percent Percent Percent
(g/cm3)3/ Diameter (/urn) < 30 #01 < 5 ^m Ratio-' >50 ftm
4.9 96 21 5 0.24 72
4.9 _
4.9 11.4 75 30 0.40 15.5
1.4 63 35 8.2 0.23 55
1.4 17 69 14 0.20 17.5
on coal densities from Handbook of Chemistry and Physics, 53rd Ed.,
Profiler
Weighted Average %
Captured on the Filter
29
71
61
55
35
CRC Press, 1972-1973
*
b_/ Percent < 5 j^m -f- percent < 30 jLtm.
c/ Data invalid; insufficient substrate loadings.
d/ Background run only.
-------
TABLE 3-16, ISOKINEfIC CORRECTION PARAMETERS—STORAGE PILE STACKING
U1
Hind speed
Run
H-10
H-ll
H-12
F-19
F-20
Ht -
(on/
sec)
58.9
183
221
71.6
77.2
1.5 IB
(fpm)
116
360
436
141
152
Ht =
(cm/
sec)
58.9
157
302
194
172
4,5 m
(fpm)
116
310
594
382
339
Intake velocity
Ht =
(cm/
sec)
274
274
274
293
311
1.5 m
(fpm)
540
540
540
576
612
Ht *
(cm/
sec)
238
238
238
293
293
4.5 m
(fpm)
468
468
468
576
576
Is o kinetic
ratio-^
4.34
1.50
1.01
2.80
2.86
Isokinetic
correction factor
Exposure
0.937
0.77?!/
0.777
0.928
-
Concen-
tration
4.07
1.17*/
1.17
2.05
-
a/ Intake velocity 4. wind speed.
b/ Based on particle size data from Run H-12.
-------
TABLE 3-17. EMISSION FACTORS AND ADJUSTMEIT PA1AMETE1S—STORAGE PILE
Mean
Number of Stacker Stacking Drop wind
Aggregate material stacker velocity rate distance speed
Run Type Silt (7.) Moisture (2) paases Cm/sec) (raph) (tonnes/hr) (tons/hr) (n) Wsec) (raph)
H-10 Iron 1.4 2.6 1 0.2 0.5 5,000 5,500 9 0.67 1.5
pellet
M-il 1.8 3.5 11 0.9 2.0 5,000 5,500 11 1.8 4.0
in
00 H-12 1.7 3.4 lot/ 0.2 0.5 5,000 5,500 12 2.7 6.0
F-19 Co«l 5.9 4.8 30 0.2 0.5 1,100 1,200 5 1.3 3.0
Suspended
participate
emission factor
(g/ tonne) (Ib/ton)
1.2 0.0023
1.5 0.0029
1.2 0.0023
0.070 0.00014
Fine
par tic ul ate
emission factor
(g/tonne) fib/ton)
0.060 0.00012
0.45i/ 0.00087£'
0.35 0.00069
0.0056 0.000011
aj Baaed on particle size data fron Ron H-12.
h/ Estimate.
-------
SECTION 4.0
WIND IEOSION TESTING
4.1 Sampling Equipment
For the measurement of dust emissions generated by wind erosion of
storage piles, a portable wind tunnel developed by Dr. Dale Gillette was
used.— The open-floored test section of the tunnel was placed directly
on the surface to be tested (15 cm x 2.4 m)» and the tunnel air flow was
adjusted to predetermined velocities up to a nominal 27 m/sec (60 mph)
as measured by a pitot tube at the downstream end of the test section.
An emissions sampling module was designed and fabricated by MRI for
use with the pull-through wind tunnel in measuring particulate emissions
and particle size distributions generated by wind erosion. As shown in
Figure 4-1, the sampling module was located between the tunnel outlet
hose and the fan inlet. The sampling train, which was operated at
34 m^/hr (20 cfm), consisted of a tapered probe, cyclone precollector,
parallel-slot cascade impactor, back-up filter, and high-volume motor.
Interchangeable probe tips were sized for isokinetic sampling at cross-
sectional average velocities of 7, 12, 17, and 27 m/sec within the
tunnel test section.
4.2 Preliminary Testing
Prior to the development of the emissions sampling module, preliminary
tests were conducted on crusted and disturbed surfaces of an inactive
coal storage pile and nearby prairie soil within an integrated and steel
plant. A test surface was disturbed, i.e., the thin crust was broken,
by walking over it repeatedly with a twisting action.
The purposes of the preliminary tests were to determine the threshold
velocities for wind erosion (minimum velocities at which wind erosion is
initiated) and to gather other data needed for the design of the sampling
module. The threshold velocity for a particular surface was determined
by observing the onset of surface particle movement as the wind velocity
was gradually increased. As indicated in Table 4-1, the surface crusts,
especially for soil, were found to be very effective in protecting
against wind erosion.
59
-------
NCAR WIND TUNNEL MODIFICATION
Cascade Impactor
a*
o
Cyclone Precol lector
Sin. Schedule 40 Pipe
20cm LD.
(Aluminum)
Flexible Hose Connecting
to Wind Tunnel
TOP VIEW
Figure 4-1. Emissions sampling module for portable wind tunnel.
-------
TABLE 4-1, OBSERVED TH1ESHDID VELOCITIES
(June 12, 1978)
Threshold friction
Approximate threshold
tunnel centerline
velocity^'
Surface
Coal pile
Undisturbed
Disturbed
Prairie soil
Undisturbed
Disturbed
velocity
Test 1
128
96
b/
~25-
(cm/sec)
Test 2
137
93
b/
~25£
(ra/sec)
13
8.9
> 27
4.5
(raph)
30
20
> 60
10
j|/ Calculated assuming a roughness height = 0«1 cm,
b/ Unobserved within tunnel flow range.
£/ Only slight deflection on Pitot tube pressure gauge
61
-------
It was also observed during the preliminary tests that, at wind
velocities substantially exceeding the threshold value for a test surface,
the erosion rate decayed rapidly with time.
4.3 Emissions Testing Program
In the emissions testing program, 12 tests were performed. A total
of eight tests were performed on the upper flat surface of an inactive
coal storage pile—three tests of one section of undisturbed (crusted)
surface, three tests of a disturbed section, and two tests of a second
disturbed section. This was followed by two tests of the flat ground
surface (undisturbed) adjacent to a dolomite storage pile and two tests
of disturbed prairie soil in the same area where the preliminary tests
were conducted.
In order to determine the quantity and textural properties of each
material being eroded, samples of the materials were removed from an
area adjacent to the test surface before each test and from the test
surface subsequent to each test. The samples were obtained by manually
sweeping the surface with a small broom. In the case of both of the
disturbed surfaces (coal and soil), a consolidated sublayer was found
within a depth of 1 to 2 cm below the original surface.
To prevent dust losses, the collected samples of dust emissions
were carefully transferred at the end of each run, to protective con-
tainers within the MRI instrument van. High-volume filters and impac-
tion substrates were folded and placed in individual envelopes. Dust
that collected on the interior surfaces of the sampling probe was rinsed
with distilled water into separate glass jars. Dust was transferred
from the cyclone precollector in a similar manner.
Dust samples from the field tests were returned to MRI and analyzed
gravimetrically in the laboratory. Glass fiber filters and impaction
substrates were conditioned at constant temperature and relative humidity
for 24 hr prior to weighing (the same conditioning procedure used before
taring). Water washes from the sampling probe and cyclone precollector
were filtered after which the tared filters were dried, conditioned at
constant humidity, and reweighed.
Samples of surface materials were dried to determine moisture
content and screened to determine the weight fraction passing a 200-mesh
screen, which gives the silt content. A conventional shaker was used
for this purpose.
62
-------
Table 4-2 gives the wind erosion test site parameters. Note that
at tunnel locations B, C, and D on the coal storage pile, experiments
were conducted in succession at the same velocity to measure the decay
in erosion rate.
Table 4-3 lists the sampling parameters for the wind erosion tests.
For Runs C-l, C-2» and C-3, the incorrect probe tip size was used resulting
in a low isokinetic ratio. For Run Oil, rapid clogging of the screen
at the end of the diffuser section prevented the maintenance of the
desired tunnel flow rate.
Table 4-4 summarizes the particle size data for the wind erosion
tests. The small portion of material collected on the interior surface
of the probe tip was disregarded in the particle size analysis. For
runs having isokinetic ratio values less than 0.8, particle size distri-
butions were adjusted according to the procedure outlined in Appendix A.
Table 4-5 presents data on the surface properties which are believed
to have a significant effect on emission rate. Table 4-6 summarizes the
wind erosion test results.
Figure 4-2 shows the dependence of the average erosion rate on cumu-
lative erosion time for the coal pile tests. Each data point is labeled
with the appropriate tunnel centerline wind velocity. As expected for a
given erosion time, the average erosion rate is highly dependent on wind
speed. It is also evident that the naturally formed surface dust was
effective in reducing wind erosion.
Figure 4-2 also shows the decay of emission rate with cumulative
erosion time for test surface B (undisturbed) and C (disturbed) at the
indicated wind velocities. The areas under the lines shown represent
the total quantity of suspended particulate generated as a function of
erosion time. It should be noted that the tunnel centerline wind
velocities used in these tests substantially exceeded the threshold
values corresponding to the onset of wind erosion for uncrusted and
crusted coal surfaces.
The results of the wind erosion testing indicate that natural
surface crusts are very effective in mitigating suspended dust emissions.
In addition, test data show that a given surface has a. finite potential
for erosion prior to mechanical disturbance. Erosion rates increase
with wind velocity and decrease with erosion time.
63
-------
TABLE 4-2. WIND E10SION SITE PARAMETERS
Cross -sectional Ambient meteorology
Surface
Type
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Dolomite
Dolomite
Prairie soil
Prairie soil
Condition
Undisturbed
Ifiidlsturbed
Undisturbed
Disturbed
Disturbed
Disturbed
Bia turned
Disturbed
Undisturbed
Undisturbed
Disturbed
Dia turbed
Run
C-l
C-2
C-3
e-4
C-5
C-6
C-7
C-8
C-9
C-10
C-ll
C-12
Uflt <°F) (7.)
35 22 71 45
56 - -
56 -
19 8.9 48 36
36 -
3fi 14 57 41
**' -
35*' -.- -
23 17 63 47
33 -
22 - - '
24 -
m
0
-
-
100
-
70
50
. -
0
-
-
-
a/ Estimated average; velocity fell from initial value of 24 m/sec {54 raph) due to plugging of tunnel screen.
b_/ Estimated value; plugging of tunnel screen prevented higher velocity.
-------
TABLE 4-3. WIND EROSION SAMPLING PARAMETERS
Wind tunnel test section
Run
C-l
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-ll
C-12
Cross -sectional
average velocity
(m/sec) (mph)
15.6
25.0
25.0
8.49
16.1
16.1
19. 2-^
15.6*'
10.3
14.8
9.83
10.7
35
56
56
19
36
36
43^
35^
23
33
22
24
Flow
Probe tip
rate Diameter
(m3/hr) (cm)
1300
2090
2090
716
1320
1320
1600^
1300^
853
1230
806
893
3.8J
5.08
5.08
3.81
3.18
3.18
2.54
2.54
3.81
3.18
5.08
3.81
Area
(cm2)
11.4
20.3
20.3
11.4
7.92
7.92
5.07
5.07
11.4
7.92
20.3
11.4
Sampling module
Velocity
Approach
(cm/sec)
1400
2240
2240
767
1420
1420
172Q*-/
1400-'
913
1320
863
956
Inlet
(cm/ sec)
828
465
465
828
1190
1190
1860
1860
828
1190
465
828
Isokinetic
ratio
0.594
0.208
0.208
1.08
0.841
0.841
1.08S/
1.33S-'
0.907
0.904
0.539
0.866
Volume
sampled
(m3)
5.66
2.83
5.23
5.66
1.13
3.40
0.850
0.378
5.66
5.66
5.66
1.08
Total mass
collected
(s)
0.8680
2.7565
0.2176
0.2995
2.4418
0.4106
2.6867
3.0931
0.3773
4.2370
0.6034
8.1764
a/ Estimated value.
-------
TABLE 4-4. PARTICLE SIZE DATA
a*
CT.
Run
C-l
C-2
G-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-ll
C-12
Surface Particle density Mass median Percent
type (B/cmr)£/ diameter (urn) < 30 pm
Undisturbed coal
Undisturbed coal
Undisturbed coal
Disturbed coal
Disturbed coal
Disturbed coal
Disturbed coal
Disturbed coal
Undisturbed dolomite
Undisturbed dolomite
Disturbed prairie soil
Disturbed prairie soil
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
2.5
2.5
1.8
1.8
> 100
> 100
>100
> 100
> 100
>100
85
86
95
> 100
97
> 100
9.0
5.5
13
16
12
18
30
30
32
9,0
29
12
Percent
<5 pm
2.7
1.4
2.0
3.6
3.0
4.5
7.6
7.7
11.5
2.8
7.8
3.0
Ratio£/
0.30
0.25
0.15
0.23
0.25
0.25
0.25
0.26
0.36
0,31
0.27
0.25
Percent
>50 urn
88
92
81
77
85
76
61
61
60
88
62
83
a_/ Estimated values.
b/ Percent < 5 pm -i- percent < 30 pm.
-------
TABLE 4-5. PROPERTIES OF LOOSE SURFACE MATERIAL
Run
01
02
03
04
C-5
06
07
08
09
010
Oil
012
Surface type
Undisturbed coal ^
Undisturbed coal
Undisturbed coal
>
Disturbed coal
Disturbed coal
Disturbed coal
Disturbed coal
Disturbed coal
Before erosion After erosion
Loading Silt Moisture Loading Silt
(ke/m2 ) fit) O, ) (kfi/m2 ) fit)
-
1,050 3.3 0.1
19.4
-
2,180 6.5 0.9 2,220 5.2
_
1,790 7.4
Undisturbed dolomite \ 73111. 6
} 7,080 12.4 0.2
Undisturbed dolomite j 2,550 8.8
Disturbed prairie soil J -
| 53,100 25.7 4.5
Disturbed prairie soil J -
Moisture
-
-
-
-
-
0.6
-
0.5
1.0
0.5
-
-
-
-------
4-6. WIND EEOSION TEST RESULTS
00
Run
0-1 •
C-2
C-3
0-4
G-5
C-6
0-1
C-8
C-9
G-W
C-U
012
Cross-sectional
average velocity Friction Roughness
Surface Silt Moisture in test section velocity height
type (1) {%) (mph) (em/see) (cm/ see) (cm)
Undisturbed 3.3 0.1 35 15.fi 109 0.01
coal
Undisturbed 3.3 0.1 56 25.0 145 0.004
coal
Undisturbed 3.3 0.1 56 25.0 145 0.004
coal
Disturbed 5.8 0.7 19 8,49 53 0.003
eoai
Disturbed 5.8 0.7 36 16.0 63 0.001
coal
Disturbed 5.8 0.7 36 16.1 63 0.001
coal
Distorted 6.9 0.7 43 19,2 of «/
coal
Disturbed 6.9 0.7 35 15.6 af oj
coal
Undisturbed 12.0 0.6 23 10.3 53 0.003
dolomite
Ondilturbed 10,6 0.4 33 14,8 186 0,lf
dolomite
Disturbed 23.7 4.5 22 9.83 48 0.03
prairie soil
Disturbed 25.7 4.5 24 10,7 33 0.0003
prairie sol 1
Cumulative
erosion Suspended partlculate Fine particutate
time emission factor emission factor
(rain) (lb/ see/acre)
-------
10.0
I I I I
I I I ( I
1= 1.0
I
a
•f
-a
a
1
0,1
0,01
O(D.I9)
O(C,I6)
•(A. 16}
* Undisturbed Coal Surface
0 Disturbed Coal Surface
(Tunnel Location, Wind Velocity - m/sec)
O(C.8.S)
I I I I I I I I I I
«(8.25)
I I I I I I I
8 10 12
Cumulative Srosion Time (Min, )
14
16
18
20
Figure 4-2. Average emission factor vsrsus cumulative erosion time,
69
-------
Additional test data are needed to define the relationship of dust
emissions generated by wind erosion to the influencing parameters.
These relationships, coupled with an analysis of wind flow patterns
around basic storage pile configurations, would form the basis for
improvement of existing emission factors.
70
-------
SECTION 5.0
REFINEMENT OF EMISSION FACTOR EQUATIONS
This section presents refined emission factor equations for:
(a) vehicular traffic on unpaved roads; (b) vehicular traffic on paved
roads; (c) storage pile formation by continuous load-in or stacking; and
(d) wind erosion of storage piles and bare ground areas. Refinements to
previously developed equations have been adopted as necessary to extend
the predictive capability of the equations to the expanded test data
bases without loss in precision. In this way, the quality assurance
(QA) ratings, as given in Figure 1-1, may be improved.
5.1 Vehicular Traffic on Unpaved Roads
Figure 5-1 shows the predictive emission factor equation for vehicular
traffic on unpaved roads, as derived by multiple regression analysis of
the test data shown in Table 5-1. The coefficient and the first two
correction terms in Figure 5-1 are identical to the expression given in
AP-42 as follows:
0.6 (0.81 s)
which describes the emissions of particles smaller than 30 /urn in Stokes
diameter generated by light duty vehicles traveling on unpaved roads.
The weight correction term in Figure 5-1 was developed on the basis of
prior testing; however, the term was formerly raised to the 0.8 power.
Table 5-1 compares measured emissions with predicted emissions as
calculated from the equation given in Figure 5-1. In addition to the
test results presented in Section 3, the results of testing of traffic
on haul roads at a taconite mine (Test Series I), which was performed as
part of another study, have also been added to the data base. §/ AS
shown in Figure 5-2, in the tests conducted on a previously inactive
road (Runs 1-1 through 1-5), emissions approached the predicted values
with successive tests. The test truck was loaded between Runs 1-3 and
1-4. Also,measured emissions for Runs 1-7 and 1-8 were significantly
lower than predicted, presumably because of the considerable rainfall on
the days prior to testing.
71
-------
OPEN DUST SOURCES Vehicular Traffic on Unpaved Roads
QA RATING; I for Dry Conditions
C for Annual Average Conditions
*~ ' .* I iff JI Jto
W
2.7
°-7/w\°*5/ d \
IT/ \365j k9/veh-km
EF = 5 9 I-— I -—
tr -jfi? tlOfi *3A
l , i
I
Determined by profiling
of emissions from light-
duty vehicles on gravel
and dirt roads under
dry conditions.
Estimated factor to
account for mitigating
effects of precipitation
over period of one
year.
Determined by profiling of emissions from
medium-and heavy-duty vehicles on gravel
and dirt roads under dry conditions.
EF - suspended particulate emissions
s = silt content of road surface material
S ~ average vehicle speed
w = average number of wheels per vehicle
W = average vehicle weight
d = dry days per year
metric
non-metric
kg/veh-km Ib/veh-mi
km/hr mph
tonnes tons
Figure 5-1, Predictive emission factor equation for vehicular traffic
on unpaved roads.
72
-------
TABLE 5-1. PREDICTED VERSUS ACTUAL EMISSIONS (UNPAVED ROADS)
Head surface
Average
Silt vehicle speed
Run
R-l)
R-2
8-3 1
8-8 )
8-10
1-13
A- 14 '
A- 15
B-l)
1.2
e-3)
F-21
F.22
F-23
F-24
F-25
\
G-27
G-28
0-29
G-30
G-31
G-32
l.l\
1-2 |
1-3 J
1-4 I
1-5)
1-7 |
1-8 1
1-9
1-10
I-ll
type
Crushed
Limestone
Dire
Crushed
Dirt
Dirt/
crushed
slag
Dirt/ slag
(Coherers/
Crushed
slag
Crushed
rock and
•/ glacial
till
Crushed
U rock
(eaeonite/
waste)
Crushed
rock
(TREX)!/
(%) (ta/hr) (fnph)
12
13
13
20
5
68
4.8
4.8
8,7
8.7
8.7
9,0
9.0
9.0
0.03
0.02
5.3
5.3
5.3
4.3
4.3
4.3
4.7
4.7
4.7
4.7
4.7
6.1
6.1
1-3 ,
l.Sfi'
1.8
46
48
64
48
64
48
48
48
23
26
26
24
24
24
24
24
35
37
39
40
47
35
24
24
24
24
24
22
22
21
21
23
30
30
40
30
40
30
30
30
14
16
16
15
15
15
15
15.
22
23
24
25
29
22
15
15
15
15
15
13.5
13.5
13
13
24
Average
Emission factor-'
vehicle wei({ht Average Mb. of Predicted^/
Actual
(tonnes Xtons) vehicle wheels (iegARDdb/VMlXkg/VKTXlb/VHI)
3
3
3
3
3
3
64
64
31
31
21
3
3
4
3
3
15
11
8
13
7
27
61
61
61
142
142
107
106
100
102
115
3
3
3
3
3
3
70
70
34
34
23
3
3
4
3
3
17
12
9
14
8
30
67
67
67
157
157
118
117
110
112
127
4.0
4.0
4.0
4.5
4.0
4.0
4.0
4,0
9.4
8.3
6.4
4,0
4.0
4.1
4.0
4.0
11. 0
9.5
7.S
8.5
6.2
13.0
6.0
6.0
6.0
6.0
6.0
6,0
6,0
6,0
6.0
6,0
1,7
l.S
2.4
2.9
0.93
9,3
6.0
6,0
4.7
5,1
3.4
0.62
0.62
0.76
*l
^
3.0
2.3
1.8
2.1
1.4
3.9
. 3.5
3.5
3.5
6.4
6.4
6,1
6,1
y
d/
5.9
6.4
8.5
10.4
3.3
33.0
21.4
21.4
16.7
18.0
12,0
2.2
2.2
2.7
«/
£/
10.7
6,1
6.3
7.5
6.1
14,0
12.4
12.4
12.4
22.6
22.6
21.6
21.5
d/
I/
1.7
1.9
2.2
2.3
1.1
9.0
6.0
6.5
3.8
3.4
4.1
0.84
0.48
0.65
0.021
0.10
3.4
2.0
1.6
2,4
1.4
4.5
1.0
2.1
4.1
5.1
7.0
3.3
3.3
0.56
0.6S
1.0
6.0
6.8
7.9
B. I
3.9
32.0
21,5
23.0
13,6
12,2
14,5
3.0
1.7
2.3
0.073
0.36
12.0
7,2
5.6.
8,7
5.1
16.0
3.7
7,5
14.5
18.1
25.0
11,6
11.6
2,0
2.3
3,6
Predicted
-r actual •
0.98
0.94 ;
1.08
1.29
0.85
1.03
1,00
0,93
1.23
1.47
0.83
0.73
1.29
1.19
.
-
0.89
1.13
1.12
0.87
0.99
0.88
3.36
1.66
0.86
1.25
0.90
1.86
1.85
-
a/ Particles smaller than 30 fim in Stokes diameter, based on actual density of silt particles.
by Based on revised Hfil emission facto* equation.
£/ Tests performed on treated road (see text).
d_/ Equation not applicable,
•/ test Series 1-1 througi.I-5 performed on previously inactive road.
J/ Tests performed on day following 2 days of rain totaling 1.13 In,
£/ Assumed value.
73
-------
TEST SITE - ROAD SURFACE MATERIAL
A Rural Kansas - Crushed Limestone
Rural Kansas - Dirt
O Iron and Steel Plant - Crushed Slag
C Iron and Steel Plant- Dirt/Gushed Slag
• Iron and Steel Plant - Dirt
D Taconite Mine - Crushed Rock (Taconire/Waste}
• TaconTte Mine - Crushed Rock/Glacial Till
I Inch Rain on Previous Evening
(2 Data Points)
Successive Tests on Previously Inactive Road
(Ib/VMT)
12 (kg/VKT)
Figure 5-2.
PREDICTED EMISSION FACTOR
Comparison of predicted and actual emissions—untreated roads.
-------
The wheel correction term appears in the emission factor equation
for the first time. The need for this term was indicated by the fact
that for Test Series E and G, the emission factor equation without a
wheel correction term consistently underpredicted the measured factors.
This appeared to be due to the effect of 10- and 18-wheel trucks, which
comprised a substantial number of the passes in those tests. In all
other test series, the vehicle mix was dominated by four- and six-wheel
vehicles.
Excluding Test Series I except for Run Nos. 1-3 and 1-5, the revised
emission factor equation presented in Figure 5-1 predicts actual test
results with a precision factor of 1.48. By comparison, the precision
factor for the unrevised equation from Table 2-1 is 1.66.
As indicated in Figure 5-3, there is no apparent relation between
the fraction of the emissions consisting of fine particles and the
average vehicle weight or the road surface composition. The average
value is approximately 35% by weight.
As stated above, limited testing of the effects of a chemical dust
suppressant was also conducted. Coherex®(a petroleum-based emulsion)
was used to treat a dirt/slag surfaced service road traveled by light-
and medium-duty vehicles at an integrated iron and steel plant. Coherex®
was applied at 10% strength in water.
Figure 5-4 shows a plot of measured dust control efficiency as a
function of the number of vehicle passes following application of the
road dust suppressant. Control efficiency was calculated by comparing
controlled emissions with uncontrolled emissions measured prior to road
surface treatment. As indicated, the effectiveness of the road dust
suppressant was initially high but began to decay with road usage. It
should also be noted that the apparent performance of Coherex was
negatively affected by tracking of material from the untreated road
surface connected to the 100-ft treated segment.
Figure 5-4 also shows the results obtained from the similar testing
of another chemical dust suppressant at a taconite mine. TREX (ammonium
lignin sulfonate—a water soluble by-product of papermaking) was applied
to the waste rock aggregate comprising the surface of a haul road. A 20
to 25% solution of TREX in water was sprayed on the road at a rate of
0.08 gal./sq yard of road surface.
Once again the effectiveness of the dust suppressant was found to
be initially high, but decayed with road usage. According to taconite
mine personnel, the binding effect of TREX can be partially restored by
the addition of water to the road surface.
75
-------
ON
I.UU
*""~">,
E E
I-J* °-80
.______-,*«'
'•'nun *^
Z
o
p 0.60
3
LU
t/l
* 0.40
u
LJJ
Z 0.20
U-
0
(
SURFACE MATERIAL
A Crushed
A Dirt
a Crushed
Limestone
Rock (Taconit
— O Crushed Slag
• Dirt/Crushed Slag
• Crushed Rock/Glacial
-
> o
-* ° 0 A D
£ 0 A D
4°
_A
20 40 60 80 100 120
i i i i i i i i l i l i
> 20 40 60 80 100
140
1 l
120
e/Wastei
Till
m
U
-i I
140 T<
AVIRAGE VEHICLE WEIGHT
Figure 5-3. Fine particle fractions of TSP emissions,
160 Tons
-------
EFFECTIVENESS OF ROAD DUST SUPPRESSANTS
o
z
UJ
u
o
e£.
I—
Z
O
u
90 -
80
70
60
50
VEHICLE TYPE
DUST SUPPRESSANT
O Haul Truck
A Light Duty Vehicles
Lignin Sulfonote
Coherex
40
TOO 120 140 160 180 200 220 240 260
VEHICLE PASSES FOLLOWING TREATMENT
280 300
Figure 5-4. Effectiveness of road dust suppressants.
-------
With regard to the effects of natural mitigation of road dust
emissions, the final term in the emission factor equation for traffic on
unpaved roads (Figure 5-1) is used to reduce emissions from dry condi-
tions to annual average conditions. The simple assumption is made that
emissions are negligible on days with measurable precipitation and are
at a maximum on the rest of the days. Obviously, neither assumption is
defendable alone; but there is a reasonable balancing effect. On the
one hand, 0.01 in, of rain would have a negligible effect in reducing
emissions on an otherwise dry, sunny day. On the other hand, even on
dry days, emissions during early morning hours are reduced because of
overnight condensation and upward migration of subsurface moisture; and
on cloudy, humid days, road surface material tends to retain moisture.
Further natural mitigation occurs because of snow cover and frozen
surface conditions. In any case, further experimentation is needed to
verify and refine this factor.
5.2 Vehicular Traffic on Paved Roads
Figure 5-5 shows the predictive emission factor formula for vehicular
traffic on paved roads. As indicated, the coefficient and the first- two
correction terms were determined by field testing of emissions from
traffic consisting primarily of light-duty vehicles on urban arterial
roadways and on a test strip that was artificially loaded with surface
dust in excess of normal levels. The vehicle weight correction term was
added by analogy to the experimentally determined factor for unpaved
roadways, and more testing is needed to confirm the validity of this
correction term. The number of lanes comprising the traveled portion of
the road and over which the surface dust loading is distributed was
added as a correction term to account for the fact that emissions increase
in proportion to surface dust loading.
The industrial road correction factor was added to the emission
factor equation because measured emissions from medium-duty and heavy-
duty vehicles traveling on paved roadways at both Plant E (tested previously)
and plant F were substantially in excess of the predicted levels without
such a correction term. There are several plausible explanations for
the increase in dust emissions from paved roads within integrated iron
and steel plants as compared to urban roads. Paved roads within inte-
grated iron and steel plants are typically bordered by unpaved surfaces
and there are no curbings to prevent traffic from traveling on these
surfaces. Therefore, additional dust generation may result from;
1. Resuspension from vehicle underbodies of dust accumulated
during travel over unpaved surfaces.
78
-------
OPEN DUST SOURCE; Vehicular Traffic on Paved Roads
QA RATINGS B For Normal Urban Traffic
C for Industrial Plant Traffic
*-«-H4)(*)ter)(ft)
kg/veh-km
UI
Determined by profiling of
emissions from traffic (mostly
light-duty) on arterial road-
ways with values for s and L
assumed.
Determined by profiling of emissions
from industrial plant traffic yielding
higher than predicted emissions,
presumably due to resuspension of
dust from vehicle underbodies and
from unpaved road shoulders.
s \/ I \/W\0.7 .
TO JI iooo AT) lb/veh-m'
ii i
Assumed by analogy
to experimentally
determined factor
for unpaved roads,
Determined by profiling of emissions from
light-duty vehicles on roadway which was
artificially loaded with known quantities
of gravel fines and pulverized topsoil.
EF = suspended particulate emissions
I = industrial road augmentation factor (see text)
n = number of traffic lanes
s = silt content of road surface material
L = surface dust loading on traveled portion of road
W = average vehicle weight
metric non-metric
kg/veh-km Ib/veh-mi
kg/km
tonnes
Ib/mi
tons
Figure 5-5. Predictive emission factor equation for vehicular traffic
on paved roads.
79
-------
2. Emissions from unpaved shoulders generated by the wakes of
large vehicles.
3, Emissions from unpaved shoulders during passage of two large
vehicles.
Also, there nay be a wheel effect similar to that indicated for
unpaved roads. Resuspension of dust from vehicle underbodies was visually
evident at Plant E as the heavy-duty vehicles traveled from an unpaved
area onto the paved roadway.
Quantification of these phenomena would require substantial additional
testing with detailed analysis of site conditions and traffic patterns
at test sites. For now, it is suggested that a multiplier of 7 be used
with the emission factor equation when Item 1 above is readily observed
and that a multiplier of 3.5 be used when the paved road (usually without
curbs) is bordered by unpaved and unvegetated shoulders. These factors
were determined by regression analysis of the test data for Plants E. and
F.
Table 5-2 compares measured emissions with predicted emissions as
calculated from the equations given in Figure 5-5. The revised emission
factor equation predicts actual test results with a precision factor of
3.31. This is a marked improvement over the precision factor of 14.1
associated with the unrevised equation from Table 2-1.
It should be noted that the emission factor for re-entrained dust
from paved roadways contains no correction term for precipitation.
Although emissions from wet pavement are reduced, increased carryover of
surface material by vehicles occurs during wet periods, and emissions
reach a maximum when the pavement dries. More testing would be helpful
in analyzing the net effects of precipitation on re-entrained dust
emissions.
5.3 Storage Pile Formation by Continuous Load-in (Stacking)
Figure 5-6 gives the predictive emission factor equation for storage
pile formation (load-in) by means of a translating conveyor stacker.
The equation was originally developed from the results of field testing
of emissions from the stacking of pelletized and lump iron ore at Plant A.—
The effect of wind speed on emissions occurs presumably because of the
increased atmospheric exposure of suspendable particles during the drop
from the stacker to the pile.
80
-------
TABLE 5-2. PREDICTED VERSUS ACTUAL EMISSIONS (PAVED ROADS)
Road surface dust
Run
P-9
P-10
P-14
E-7
E-8
P-3,
P-5,
P-6
P-15,
P-16
F-13 ^
F-14
F-15 ;
F-16 ^
F-17
F-18 f
Type
(Pulver-
ized
topaoll-'
Gravel^
I (Iron and
steel)
Plant E
Urban
arterial
site l£f
Urban
arterial
site 2£/
Iron and
Bteel
plant
Iron and
steel
plant
Loading
excluding curbs!/
(kg/km) (lb Anile)
1,990
809
1,890
225
225
45. 1^'
42.0^
57.2
57.2
57.2
629
629
629
7,060
2,870
6,700
800
800
1601'
149^
203
203
203
2,230
2,230
2,230
No. of
traffic
lane a
4
4
4
2
2
4
4
2
2
2
2
2
2
Silt
m
45
92
23
5.1
5.1
uP
101'
13.2
13.2
13.2
6.8
6.8
6.8
I
(industrial
multiplier)
1
1
I
7
7
1
1
1
1
1
3.5
3.5
1
Average
vchlc IP
uelftht
(tonnes )
3
3
3
6
7
3
3
7
5
5
12
11
5
(tons)
3
3
3
7
B
3
3
8
5
5
13
12
5
Emission
Predicted^/
(kg/VKT)
0.82
0.6B
0.39
0.26
0.29
0.0039
0.0037
0.096
0.068
0.068
0.76
0.70
0.11
(lb/VMT)
2.9
2.4
1.4
0.93
1.02
0.014
0.013
0.34
0.24
0.24
2.7
2.5
0.39
factors^
Actual
(kg/VKt)
1.0
0.59
0.13
0.21
0.28
0.0042
0.0037
0.16
0.056
0.045
0.70
0.48
0.14
(Ib/WTB
3.7
2.1
0.46
0.76
1.0
0.015
0.0130
0.5B
0.20
0.16
2.5
1.7
0.48
Predicted
-i- Actual
0.78
1.14
3.04
1.22
1.02
0.93
1. 00
0.59
1.20
1.50
I. Off
1.47
0.81
a/ Loading distributed over traveled portion of road, i.e., traffic lanes.
b/ Particles smaller than 30 ftm in Stokea diameter based on actual density of silt particles.
c/ Based on revised MR1 emission factor equation.
d/ Four-lane test roadway artificially loaded.
e/ Four-lane roadway with traffic count of about 10,000 vehicles per day, mostly light-duty.
f/ Estimated value.
-------
OPEN DUST SOURCE: Storage Pile Formation by Means of
Conveyor Stacker
QA RATING: B
EF= 0.00090
Mv2
1~/
kg/ tonne
EF= 0.0018
I
Ib/ton
Determined by profiling of emissions
from pile stacking of pelletized and
lump iron ore and coal.
metric
EF = suspended particulate emissions kg/tonne of
material transferred
s = silt content of aggregate %
M = moisture content of aggregate %
U = mean wind speed m/sec
H = drop height m
non-metric
Ib/ton of
material transferred
mph
ft
Figure 5-6, Predictive emission factor equation for storage pile
formations by means of conveyor stacker.
82
-------
An additional adjustment term containing drop distance has been
added to the emission factor equation. It is assumed that emissions are
proportional to drop distance, accounting for the additional energy
released on impact and the greater time of exposure during the drop.
Table 5-3 compares measured emissions with predicted emissions as
calculated from the equation given in Figure 5-6. The revised emission
factor equation predicts actual test results with an improved precision*
However, the sample size remains too small for meaningful statistical de-
termination of the precision factor.
Addition of the drop distance correction term aids significantly in
predicting the results of Runs H-10 through H-12 although a large discrepancy
remains for the first two of these runs. This may be due to lack of
representativeness of the pellet moisture values for these runs. The
pellets stacked during these runs comprised the last portion of a barge
shipment, and moisture variations may have been substantial if water had
collected in the bottom of the ship hold. The pellets were observed to
be unusually wet when the samples were taken.
83
-------
TABLE 5-3. PREDICTED VERSUS ACTUAL EMISSIONS (LOAD-IN BY STACKER)
Emission factor^' x
Aggregate
Run
A-8 \
\
A- 10 )
A-ll ^
A-12 }
j
A- 13 J
00 \
*- H-10 1
(
H-ll /
1
B-12 /
F-19
Type
Iron
ore
pellets
I
Lump
> iron
ore
i
(Iron
ore
pellets
Coal
Silt
(%)
4,8
4.8
2.8
11.9
19.1
1.4
1.8
1.7
5,9
Moisture
«)
0.64
0.64
2.0£/
4.3
4.3
2.6
3.5
3.4
4.8
Drop Wind
distance speed
Cm)
3.0
1.5
4.5
3.0
3.5
9.0
11.0
12.0
5.0
(in/sec
1.0
2.0
0.8
0.8
1.0
0.7
1.8
2.7 .
1.3
)(mph)
2.3
4.5
1.8
1.8
2.2
1.5
4.0
6.0
3.0
Predicted^/
(kg/
tonne )
3.9
3.7
0.27
0.16
0.38
0.13
0.31
0.50
0.18
(lb/
ton)
7.8
7.5
0.54
0.33
0.76
0.27
0.62
1.0
0.37
103
Actual
(kg/
tonne )
l.l
25.0
0.26
0.19
0.12
1.1
1.4
l.l
0.070
(lb/
ton)
2.3
5.0
0.53
0.38
0.25
2.3
2.9
2.3
0.14
Predicted
-i- Actual
3.39
1.50
1.02
0.87
3.04
0.12
0.21
0.43
2.64
a/ Particles smaller than 30 fJLm in Stokes diameter based on an adjusted density of 2.5 g/cm^;
*""" T
multiply emission factor values by 10 to obtain units given.
b/ Based on revised MRI emission factor equation.
c/ Estimated value.
-------
SECTION 6.0
DEVELOPMENT OF STORAGE PILE SILT
AND MOISTURE VALUES
This section describes a field study of the physical properties of
aggregate materials which are known to affect the atmospheric dust
emissions generated by exposed materials handling operations associated
with adding material to or removing material from an open storage pile
and by wind erosion of the exposed surface of the pile. Aggregate
materials of interest are those which are stored in significant quantities
within integrated iron and steel plants, specifically iron-bearing pellets,
coal, iron ore, limestone .and slag. The properties of concern are
moisture content and texture (silt content and cloddiness).
The testing program focused on the moisture content of storage pile
surface material because of the strong dependence (inverse square) of
wind-generated dust emissions on moisture, and because of the highly
variable nature of this parameter. Temporal variations in surface
moisture content are a function of precipitation and evaporation rates
during the time of exposure. Because available emission factors are
based on field tests generally performed with dry materials, seasonal
and annual emission estimates must be adjusted to higher moisture values
reflective of various climatic and exposure conditions.
6.1 Testing Program
The field testing program was divided into two segments: an intensive
short-term program entailing daily collection of one to three samples of
dormant coal and iron pellet storage piles; and a longer term program of
weekly sampling of coal and iron pellet storage piles, both dormant and
active.
The 1-week program of intensive sampling was conducted by MRI at
Armco, Inc., Middletown, Ohio. The purpose of the intensive program was
to determine the diurnal variation of storage pile surface moisture.
85
-------
The program of weekly sampling and analysis of coal and Iron pellet-
storage piles, extending over a. period of 2 to 3 months, was conducted
by personnel at three cooperating plants! Anaco, Inc., Middletown,
Ohio; Bethlehem Steel, Bethlehem, Pennsylvania; and Inland Steel, last
Chicago, Indiana. The purpose of this extended sampling program was to
gather data for use in developing a relationship between daily storage
pile surface moisture, after normalization to remove the daytime portion
of the diurnal moisture cycle, and precipitation/evaporation parameters.
The specific sampling program for each plant was formulated during
a presurvey, taking into account: the materials stored at the plant,
both live (active) storage and dead (inactive) storage; and the accessibility
of the material for sampling, including the load-in and load-out streams.
The materials of greatest interest were pellets and coal although iron
ore, limestone, and other materials were also considered for sampling.
The procedures developed by MR! for sampling of aggregate storage
piles and for silt and moisture analysis of collected samples are reproduced
in Appendix B. Appropriate meteorological data for the sampling locations
and periods were obtained by MRI from area weather stations.
6.2 Test Results—Intensive Study
Table 6-1 lists the results of the 1-week intensive field study
conducted by MRI at Armeo's Middletown works. No rainfall occurred
during or within 4 days previous to the sampling period, and the sampling
days exhibited similar meteorology.
The data in Table 6-1 may be used to determine the diurnal variation
in surface moisture content for a precipitation-free period. During the
summer months, an increase in pile surface moisture during nighttime
hours may be expected due to condensation and/or diffusion of moisture
from wetter material within the pile; however, during daytime hours,
surface moisture normally decreases because of increased evaporation.
By averaging the moisture values for the morning, mid-day, and
afternoon sampling times, the curves in Figure 6-1 may be constructed.
The fact that the curve for Armco coal lies below the curve for Armco
pellets indicates that coal has a greater capability than pellets for
moisture retention. This is consistent with the substantially larger
quantity of fines in crushed coal as compared to iron-bearing pellets.
The curves shown in Figure 6-2 for coal and pellets have been
normalized to unit moisture at 1400 hours (2 p.m.). In this way,
moisture values for sampling times between about 0930 hours and 1400
hours may be adjusted to the equivalent 1400 value. This allows po-
tential correlation of "daily" moistures with precipitation events.
86
-------
TABLE 6-1, SURFACE MOISTURE VARIATION IN DORMANT
PILES AT AHMCO MIDDLETOWN WORKS
Material
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Coal
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Pellets
Sampling
date
7/17/78
7/18/78
7/18/78
7/18/78
7/19/78
7/19/78
7/19/78
7/20/78
7/20/78
7/20/78
7/21/78
7/17/78
7/18/78
7/18/78
7/18/78
7/19/78
7/19/78
7/19/78
7/20/78
7/20/78
7/20/78
7/21/78
Time
(EOT)
1415
0930
1130
1415
0930
1100
1400
0930
1115
1315
0900
1415
1000
1130
1445
0945
1115
1415
0945
1130
1330
0915
Moisture
content
(%)
1.51
2.89
2.11
1.62
1.55
1.67
1.57
1.27
1.63
1.12
1.50
-
0.95
0.40
0.21
1.26
0.21
0.43
0.19
0.26
0.05
0.37
Relative
Humidity Temperature
(%) (CG) (°F)
-
-
55
72
54
48
72
53
48
84
-
-
-
55
72
54
48
72
53
48
84
-
-
30
26
30
33
27
32
36
27
„
-
-
30
26
30
33
27
32
36
27
-
-
86
78
86
92
81
90
97
81
„
-
-
86
78
86
92
81
90
97
81
Cloud
cover
(%)
0
0
0
20
0
0
0
0
0
5
100
0
0
0
30
0
0
0
0
0
20
100
87
-------
00
03
3.0i-
§>
O
(5
1.0
Erie Pellets
Armco Pellets
I
I
0800 0900 1000 1100 1200 1300
Time of Day (Hours)
1400
1500
Figure 6-1. Observed storage pile moisture versus time of day.
-------
3.0
00
,. 2.0
o
o
o
I
"5
o
1.0
lArmco Pellets
^^.^ Erie Pellets
^"
Armco Coal
I
I
0800 0900 1000 1100 1200
Time of Day (Hours)
1300
1400
1500
Figure 6-2. Storage pile moisture normalization factor
(-1.0 for 1400 hr LOT).
-------
It should be emphasized that the normalization curves derived from
the intensive study apply only to geographical locations and times of
year which exhibit similar evaporative conditions. For example, the
lower curve for Erie iron pellets in Figure 6-2 was derived from moisture
measurements corresponding to a. significantly lower daytime evaporation
rate; thus, that curve shows a much smaller moisture decay rate,—'
6.3 Test Results—Extended Study
Tables 6-2 through 6-4 present the results of weekly storage pile
sampling conducted by Armco, Inc., Bethlehem Steel and Inland Steel,
respectively. Precipitation data for the 4 days previous to the day of
sampling were obtained from nearby weather stations. However, the
nearest evaporation observation sites were 20 or more miles from the
storage piles.
It should be noted that the 24-hour periods preceding precipitation
observation ended at 0800 hours at the Middletown station and at 0900
hours at the Gary station. Therefore, the precipitation "day" preceding
sampling at these two locations extended to the morning of the day on
which sampling occurred. Fortunately, with few exceptions, no precipita-
tion occurred on the "day" of sampling. Because hourly precipitation
data were available at Bethlehem, the day preceding sampling was taken
to be the 24-hour period ending at 1400 hours on the sampling day.
A number of correlations of daytime surface moisture levels to
precipitation and evaporation data were attempted for various site-
specific data sets. The following conclusions were derived froa this
effort:
1. Correlations were improved, as expected, by treating coal and
pellets separately and by separating data from active and dormant storage
piles.
2. The strongest correlation was found to exist between weighted
precipitation for the 4 days prior to sampling (as described below), and
normalized storage pile surface moisture.
3. No correlation of storage pile surface moisture with evaporation
as a separate variable was found.
4. The data from Inland Steel were not amenable to correlation,
possibly because of inconsistency between Inland's standard sampling and
analysis methods and those recommended by MEI and adopted by the other
two plants.
90
-------
6-2. PILE AID PRECIPITATION/EVAPORATION
INC., OHIO)
\o
Aggregate
Active
coal
Dormant
coal
Active
pellets
Dormant
pellets
Sampling
1978
date Time
7/24
8/1
8/7
8/18
8/21
8/28
7/24
8/1
8/7
8/21
8/28
7/18
7/28
8/4
8/11
8/18
8/25
8/31
7/18
7/28
8/4
8/11
0930
1000
0945
0935
1010
1010
1000
1025
1015
1040
1025
1420
0830
0830
0930
0910
1200
1200
1445
0900
0900
1000
Precipitation (nun)S/
Day previous
1
20.3
0
1.8
2.5
0
19.0
20.3
0
1.8
0
19.0
0
0
0
0
2,5
0
41.9
0
0
0
0
2
0
27.9
0.8
0,5
3,8
0.8
0
27.9
0.8
3.8
0.8
0
0
3.9
13. S
0,5
0
1.3
0
0
3.9
13.5
3
14.0
11.4
Q
0
-
0
14.0
11.4
0
-
0
0
0
0.5
0
0
0
5.1
0
0
0.5
Q
4
0
0
0
0
2.5
0
0
0
0
2.5
0
0
20.6
0
0
0
Q
19.0
0
20.6
0
0
Weighted
value
22.4
11.9
2.0
2.8
1.5
19.3
22.4
11.9
2.0
1.5
19.3
0
1.0
4.6
5.1
2.8
0
43.9
0
1.0
4.6
5.1
Evaporation (mro)]>/
Day previous
1
16.8
4.8
3.8
2.3
3.S
3,8
16.8
4.8
3.8
3.8
3.8
7.4
5.8
3,0
3.8
2.3
H.6
4.8
7.4
5.8
3.0
3.8
2
7.6
3.6
0.8
4.3
1.4
5.1
J.6
3.6
0.8
7.4
5.1
8.4
7.4
3,6
S.I
4.3
2.5
2.5
6.4
7.4
3.6
5.1
3
7.9
5.1
10.4
6.4
8.6
1.8
7,9
5.1
10.4
8.6
1.8
7.9
4.8
3.6
4.8
6.4
3.8
5.6
7.9
4.8
3.6
4.8
4
5.8
6,6
1.0
4.8
2.3
8.6
5.8
6.6
3.0
2.3
8.6
7.9
7.6
4.8
4.1
4.8
6.9
3.8
7.9
7.6
4.8
4.1
Weighted
value
21.1
7.1
5.6
5.1
7.9
6.4
21.1
7.1
5.6
7,9
6.*
11, f
9.7
5.1
6.6
5.1
10.4
6.9
11.9
9.7
5.1
6,6
Moisture content of
surface aggregate (%)
Observed
5.8
2.8
3.8
1.3
1.5
5.0
8.0
4.4
4.2
1.1
7,6
2.8
1,3
2.5
2.4
4.4
0.66
5.3
0.21
1.2
4.7
4.4
Normalized
4.7
2.3
3.1
1.1
1.2
4.0
6.5
3.5
3.4
0,9
6,1
2.8
0.4
0.8
0.8
1.5
0.6
4.7
0.2
0.4
1.6
1.8
a/ From uonofficial MiddleCown rain gauge data*
b/ Fcorn Deer Creek Lake, Ohio, evaporation a tat Ion.
-------
10
to
TABLE 6-3. STORAGE PILE MOISTURE MD P1ICIPIIATION/EVAPORATION
PENNSYLVANIA)
Sampling
Aggregate
Dormant
coal
Dormant
pellets
1978
date
9/14
9/25
9/29
11/10
9/14
9/25
9/29
11/10
Time
1430
1430
1400
1430
1400
1500
1430
1400
1
0
0
0
0
0
0
0
0
Precipitation (inn)"" Evauoration (nm)~
Day
2
0
0
0
0
0
0
0
0
Previous
3 4
0 0,5
12.7 4.3
0 0
o a
0 0.5
1.3 4.3
0 0
0 0
Weighted Day Previous
value I 2 34
0 2.8 5.3 2.8 1.3
0.5 2.5 2.0 4.3 1.5
0 2.3 3.0 2.5 3.3
0 -
0 2.8 5.3 2.8 1.3
0.5 2.5 1.3 4.3 1.5
0 2.3 3.0 2.5 3.3
o - - - -
Mbisture
content of
Wbighted surface apareEate (1)
value Observed
5.8 5.9
4.3 5.2
4.3 4.8
- 4.6
5.8 2.6
4.3 2.2 -
4.3 1.8
1.3
normalized
6.3
5.6
4.8
4.9
2.6
2.2
1.8
1.3
aj From Allentown - Bethlehem National Heather Service office.
_b/ Front Landisville 2NH, Pennsylvania) evaporation station.
-------
TABLE 6-4. STORAGE PILE MOISTURE AND PRECIPITATION/EVAPORATION
(INLAND STEEL, EAST CHICAGO, ILLINOIS)
vo
Sampling
Aggregate
Iron
pellets
Coal
1978
date
7/26
8/4
8/9
8/17
8/23
8/30
9/6
9/13
9/20
9/27
10/5
10/11
7/26
8/4
8/9
8/17
8/23
8/30
9/6
9/13
9/20
9/27
10/5
10/11
Precipitation (mm)5
Day Previous
Time
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1100
1
0
23.4
0
17.8
0
0
0
2
0
0
0
0
0
2.0
0
0.51 0
8.6
0
-
-
0
23.4
0
17.8
0
0
0
0.51
8.6
0
_
""
25.7
0
_
-
0
0
0
0
0
2.0
0
0
25.7
0
_
~
3
17.8
5.1
0
0
0
1.8
0
0
0
0
_
-
17.8
5.1
0
0
0
1.8
0
0
0
0
-
~
4
5.6
0
0
1.3
1.5
0
0
0
0
0
_
-
5.6
0
0
1.3
1.5
0
0
0
0
0
-
~
Weight erf
value
1.8
1.5.2
0
11.2
0
0.76
0
0.25
11.4
0
-
-
1.8
15.2
0
11.2
0
0.76
0
0.25
11.4
0
_
"•
Evaporation (ran)-
Dav Previous
1
2.8
3.0
4.6
8.9
2.0
3.8
2.0
-
0.8
0.8
0.3
1.5
2.8
3.0
4.6
8.9
2.0
3.8
2.0
-
0.8
0.8
0.3
1.5
2
2.3
1.3
2.8
3.0
1.3
1.3
3.3
-
4.1
2.0
2.8
1.8
2.3
1.3
2.8
3.0
1.3
1.3
3.3
_
4.1
2.0
2.8
1.8
3
2.5
2.0
2.8
2.8
3.6
2.8
4.3
-
2.5
1.0
1.8
1.0
2.5
2.0
2.8
2.8
3.6
2.8
4.3
_
2.5
1.0
1.8
1.0
4
4.8
5.1
6.4
0.5
4 1
2.3
2.3
3.0
2.0
1.5
1.8
0.8
4.8
5.1
6.4
0.5
4.1
2.3
2.1
3.0
2.0
1.5
1.8
0.8
Weighted
val ne
4.1
4.1
6.4
10.4
1.3
4.8
4.1
-
2.8
1.8
1.5
-
4.3
4.1
6.4
10.4
3.3
4.8
4.1
0
2.8
1.8
1.5
—
Moisture
surface ac
Observed
4.5
3.5
4.6
3.3
1.3
3.9
1.9
4.85
4.38
0.2
4.8
1.6
6.0
3.8
5.7
8.9
0.82
1.32
1.33
9.22
3.47
1.19
5.37
4.56
content of
ERrecate (%)
Normalized
3.0
2.3
3.1
2.2
0.9
2.6
1.3
3.2
2.9
0.1
3.2
1.1
4.8
3.0
4.6
7.1
0.7
1.1
1.1
7.4
2.8
1.0
4.3
3.6
a/ From Gary| Indiana, official rain guage.
J>/ From Valparaiso, Indiana, water works.
-------
The weighted precipitation (P) value takes into account that the
more recent the precipitation, the stronger its effect on the observed
storage pile moisture. It is calculated as follows;
4 days
^w" E P« 6XP Hn-0.5)]
n = 1
Thus, the residual effect of precipitation decreases exponentally and
is neglected after 4 days.
As shown in Figures 6-3 and 6-4, a high degree of correlation
between storage pile surface moisture and weighted precipitation was
found for the Armco data. For both coal and pellets, the surface
moisture levels of active piles were less sensitive to precipitation
than the dormant piles. This is because the surfaces of the active
piles are disturbed on a daily basis. Unfortunately, all of the
Bethlehem samples were collected on days with PW = 0» so that correla-
tion of Bethlehem moisture values with weighted precipitation was
meaningless.
The question might be raised as to why storage pile surface mois-
ture correlated well with weighted precipitation but very poorly with
evaporation as a separate variable. There are several possible explana-
tions for this finding:
1. The evaporation in data were obtained from weather stations
located several miles from the test piles,
2. Pan evaporation measured under full exposure conditions does
not reflect microclimate effects around storage piles resulting from
shading, wind channeling, etc.
3, Moisture transfer between the interior of a pile and the pile
surface may contribute substantially to the surface moisture balance.
The regression equations given in Figures 6-3 and 6-4 may be used
to determine monthly, seasonal or annual values of surface moisture for
coal or pellet piles. This would be accomplished by substitution of
weighted precipitation values calculated from daily precipitation data
for the geographical area being considered. It would also be necessary
to relate average normalized moisture for 1400 hours to the value for
the time of day which represents the daily average.
94
-------
VO
6.0
5.0
#4.0
3
tn
3.0
2.0
Dormant Coal
M=Q.22P-H.6
R=0.91
O
Active Cool
M=0.13P+1.4T
R = 0.84
1.0
0
1
1
0
10
20 30
Weighted Precipitation (mm)
40
50
Figure 6-3. Correlation of normalized moisture with weighted precipitation--coal piles.
-------
5.0
4.0
Dormant Pellets
M = 0.32 P+0.14
Active Pellets
M = 0.08 P+ 0.99
1=0.83
10
20 30
Weighted Precipitation (mm)
40
50
Figure 6-4. Correlation of normalized moisture with weighted precipitation—iron pellet piles,
-------
SECTION 7.0
ADDITIONAL RESEARCH NEEDS
Listed below are suggestions for future work on open dust sources
found within the iron and steel industry as well as other industries
which involve extensive materials handling. These suggestions reflect
our assessment of the highest priority research needs, based in part on
the nature of frequent requests for information.
7,1 Emission Inventory Handbook
The MRI predictive emission factor equations are receiving wide-
spread application in connection with requirements for State Implementation
Plan revisions. Many requests for guidance on the selection of correction
parameter values and on the determination of source extent values for
specific Industrial applications are being received. The preparation of
a handbook for emission inventory of fugitive dust sources would provide
a much needed resource for work in this area. The handbook would describe
schemes for calculating correction parameters and source extent values
for various source categories. In addition, typical correction parameter
values would be provided for common types of emitting surfaces (road
surface materials, stored aggregates, etc.).
7.2 Unpaved Road Dust Controls
As shown in Section 2, unpaved roads constitute the major source of
fugitive dust within industries which handle large quantities of aggregate
materials. Currently, the reliability of unpaved road emission estimates
is limited primarily by lack of data on road dust control measures. The
most common control practice is watering. Very little data exist on the
effectiveness of watering as a function of road surface, traffic, and
meteorological conditions. Clearly there is a need for accurate quanti-
fication of the time-dependent effectiveness of typical watering programs
used in industry. The MRI Exposure Profiler is ideal for this application.
Undoubtedly, this testing would shed substantial light on the effects of
parameters (droplet size, coverage, intensity of application, etc.)
which can be used to optimize watering. The information in Section 6 on
storage pile moisture cycles would provide information pertinent to the
proposed study. As part of this study, some testing of chemical dust
suppressants might also be conducted.
97
-------
7.3 Wind Erosion of Exposed Aggregate Materials
The information presented in Section 4.0 is useful in defining the
complex principles underlying Che phenomena of wind erosion of exposed
aggregate materials. This work has involved in situ testing with a
portable wind tunnel, which provides the distinct advantage of sampling
under controlled conditions without prior disturbance of the natural
surface condition. Although wind erosion of active (disturbed) materials
is a major source of fugitive dust, presently available emissions data
are far too limited to characterize this source. Additional fundamental
investigation would provide valuable information as to physical parameters
which enter into the wind erosion process, and direct means for minimising
emissions without the need for added controls.
98
-------
SECTION 8.0
REFERENCES
1. Bohn, R., T. Cuscino, Jr., and C. Cowherd, Jr. Fugitive Emissions From
Integrated Iron and Steel Plants, Final Report, Midwest lesearch Insti-
tute for U.S. Environmental Protection Agency, Publication No. EPA-600/
2-78-050, March 1978.
2. Cowherd, C., Jr., K. Axetell, Jr., C. M. Guenther, and G. Jutze. Develop-
ment of Emission Factors for Fugitive Dust Sources, Final Report, Midwest
Research Institute for U.S. Environmental Protection Agency, Publication
No, EPA-450/3-74-037, N1IS No. PB 238262/AS, June 1974.
3. Cowherd, C,, Jr., C. M. Maxwell, and D, W, Nelson. Quantification of
Dust Entrainment From Paved Roadways. Final Report, Midwest Research
Institute for U.S. Environmental Protection Agency, Publication No. EPA-
450/3-77-027, July 1977.
4. Mann, C. 0., and C. Cowherd, Jr. Fugitive Dust Sources, Compilation of
Air Pollution Emission Factors. Section 11.2, U.S. Environmental Protec-
tion Agency, Publication Af-42, OTIS No, PB 254274/AS, December 1975.
5, Cowherd, C., Jr. Measurement of Fugitive Participate. Second Symposium
on Fugitive Emissions Measurement and Control, Houston, Texas, May 1977.
6. Cowherd, C., Jr., and C. M. Guenther. Development of a Methodology and
Emission Inventory for Fugitive Dust for the Regional Air Pollution
Study. Final Report, Midwest Research Institute for U.S. Environmental
Protection Agency, Publication No, IPA-450/3-76-003, January 1976,
7. Gillette, Dale. Tests with a Portable Wind Tunnel for Determining Wind
Erosion Threshold Velocities. Atmospheric Environment, 12:2309, 1978,
8. Guscino, T., Jr. Taconite Mining Fugitive Emissions Study. Final
Report, Midwest Research Institute for Minnesota Pollution Control
Agency, June 7, 1979.
99
-------
SECTION 9.0
GLOSSARY
Activity Factor - Measure of the intensity of aggregate material disturbance
by mechanical forces in relation to reference activity level defined as
unity.
Cloddiness - The mass percentage of an aggregate sample smaller than 0.84
mm in diameter as determined by dry sieving.
Cost-Effectiveness - The cost of control per pound of reduced fine particle
emissions..
Dry Day - Day without measurable (0.01 in, or more) precipitation.
Dry Sieving - The sieving of oven-dried aggregate by passing it through a
series of screens of descending opening size.
Duration of Storage - The average time that a unit of aggregate material
remains in open storage, or the average pile turnover time.
Dust Suppressant - Water or chemical solution which, when applied to an
aggregate material, binds suspendable particulate to larger particles.
Exposed Area, Effective - The total exposed area reduced by an amount which
reflects the sheltering effect of buildings and other objects that retard
the wind.
Exposed Area, Total - Outdoor ground area subject to the action of wind
and protected by little or no vegetation.
Exposure - The point value of the flux (mass/area-time) of airborne particu-
late passing through the atmosphere, integrated over the time of measure-
ment.
Exposure, Filter - Exposure determined from filter catch within an exposure
sampler.
100
-------
Exposure, Integrated - The result of mathematical integration of spatially
distributed measurements of airborne particulate exposure downwind of a
fugitive emissions source.
Exposure, Total - Exposure calculated from both filter catch and settling
chamber catch within an exposure sampler.
Exposure Profiling - Direct measurement of the total passage of airborne
particulate Immediately downwind of the source by means of simultaneous
multipoint isokinetic sampling over the effective cross-section of the
fugitive emissions plume.
Exposure Sampler - Directional particulate sampler with settling chamber
and backup filter, having variable flow control (5 to 20 cfm) to provide
for isokinetic sampling at wind speeds of 4 to 15 mph.
Friction Velocity - A measure of wind shear stress on an exposed surface as
determined from the slope of the logarithmic velocity profile near the
surface.
Fugitive Emissions, Total - All particles from either open dust or process
fugitive sources as measured immediately adjacent to the source.
Fugitive Emissions - Emissions not originating from a stack, duct, or
flue.
Load-in - The addition of material to a storage pile.
Load-out - The removal of material from a storage pile.
Materials Handling - The receiving and transport of raw, intermediate and
waste materials, including barge/railcar unloading, conveyor transport
and associated conveyor transfer and screening stations.
Moisture Content - The mass portion of an aggregate sample consisting of
unbound moisture as determined from weight loss in oven drying with
correction for the estimated difference from total unbound moisture.
Particle Diameter, Aerodynamic - The diameter of a hypothetical sphere of
unit density (1 g/cm) having the same terminal settling velocity as the
particle in question, regardless of its geometric size, shape and true
density.
Particle Diameter, Stokes - The diameter of a hypothetical sphere having
the same density and terminal settling velocity as the particle in
question, regardless of its geometric size and shape.
101
-------
Particle Drift Distance - Horizontal distance from point of particle injec-
tion into the atmosphere to point of removal by contact with the ground
surface.
Particulate, Fine - Airborne particulate smaller than 5 Jim in Stokes diam-
eter.
Particulate, Suspended - Airborne particulate smaller than 30 micrometers,
in Stokes diameter, the approximate cut-off diameter for the capture of
particulate matter by a standard high-volume sampler, based on a particle
density of 2 to 2.5 g/cm ,
Precipitation-Evaporation Index - A climatic factor equal to ten times the
sum of 12 consecutive monthly ratios of precipitation in inches over
evaporation in inches, which is used as a measure of the annual average
moisture of exposed material on a flat surface of compacted aggregate.
Precision Factor - The precision factor (f) for an emission factor equa-
tion is defined such that the 951 confidence interval for a predicted
emission factor value (P) extends from P/f to Pf; the precision factor
is determined by exponentiating twice the standard deviation of the
differences between the natural logarithms of the predicted and observed
emission factors.
Road, Paved - A roadway constructed of rigid surface materials, such as
asphalt, cement, concrete and brick.
Road, Unpaved - A roadway constructed of non-rigid surface materials such
as dirt, gravel (crushed stone or slag), and oil and chip surfaces.
Road Surface Dust Loading - The mass of loose surface dust on a paved road-
way, per length of roadway, as determined by dry vacuuming.
Road Surface Material - Loose material present on the surface of an unpaved
road.
Roughness Height - A measure of the roughness of an exposed surface as
determined from the y-intercept of the logarithmic velocity profile
near the surface.
Source, Open Dust - Any source from which emissions are generated by the
forces of wind and machinery acting on exposed aggregate materials.
Source, Process Fugitive Emissions - An unducted source of emissions in-
volving a process step which alters the chemical or physical charac-
teristics of a material, frequently occurring within a building.
102
-------
Silt Content - The mass portion of an aggregate sample smaller than 75 microm-
eters in diameter as determined by dry sieving.
Spray System - A device for applying a liquid dust suppressant in the form
of droplets to an aggregate material for the purposes of controlling the
generation of dust.
Storage Pile Activities - Processes associated with aggregate storage; piles,
specifically, load-in, vehicular traffic around storage piles, wind erosion
from storage piles, and load-out,
Surface Erodibility - Potential for wind erosion losses from an unsheltered
area, based on the percentage of erodible particles (smaller than 0,85
mm in diameter) in the surface material.
Surface Stabilization - The formation of a resistive crust on an exposed
aggregate surface through the action of a dust suppressant, which sup-
presses the release of otherwise suspendable particles.
Vehicle, Heavy Duty - A motor vehicle with a gross vehicle traveling weight
exceeding 30 tons.
Vehicle, Light Duty - A motor vehicle with a gross vehicle traveling weight
is less than or equal to 3 tons.
Vehicle, Medium Duty - A motor vehicle with a gross vehicle traveling weight
is greater than 3 tons, but less than 30 tons.
Windbreak - A natural or man-made object which reduces the ambient wind
speed in the immediate locality.
103
-------
SECTION 10.0
ENGLISH TO METRIC UNIT CONVERSION TABLE
English unit Multiplied by Metric unit
Ib/T
Ib/vehicle mile
Ib/acre yr
Ik
T
raph
mile
ft
acre
0,500
0.282
112
0.454
0.907
0.447
1.61
0.305
0.00405
kg/t
kg /vehicle km
kg /km2 yr
kg
t
m/s
km
m
km2
104
-------
APPENDIX A
EMISSIOH FACTOR CALCULATION
A-l
-------
This appendix summarizes the calculation procedures used to derive
the emission factors presented in this report* Example calculations are
presented for each source category.
1.0 Emission Rate
The passage of airborne particulate, i.e., the quantity of emissions
per unit of source activity, is obtained by spatial integration (over
the effective cross-section of the plume) of distributed measurements of
exposure (mass/area). The exposure is the point value of the flux
(mass/area^time) of airborne particulate integrated over the time of
measurement.
Mathematically stated, the total mass emission rate (R) is given
by;
1 =± ! /•/•m(h.w)
t / I a
A
where m = dust catch by exposure sampler after subtraction of
background
a = intake area of sampler
t = sampling time
h «• vertical distance coordinate
w "• lateral distance coordinate
A = effective cross-sectional area of plume
In the case of a line source or moving point source with an emission
height near ground level, the mass emission rate per source length unit
being sampled is given by:
H
E - 1 f EJh) dh
t / a
o
where W = width of the sampling intake
H = effective extent of the plume above ground
In order to obtain an accurate measurement of airborne particulate
exposure, sampling must be conducted isokinetically; e.g., flow stream-
lines enter the sampler rectilinearly. This means that the sampling
intake must be aimed directly into the wind and, to the extent possible,
the sampling velocity must equal the local wind speed. The first condition
is by far the more critical.
A-2
-------
2.0 Isokinetie Corrections
If it is necessary to sample at a nonisokinetie flow rate (e.g., to
obtain sufficient sample under light wind conditions), the following
multiplicative factors should be used to correct measured exposures and
concentrations to corresponding isokinetic values:
Fine Particles Coarse Particles
(d < 5 Mm) (d > 50 urn)
Exposure Multiplier U/u 1
Concentration Multiplier 1 u/U
where: u = sampling intake velocity at a given elevation
U = wind velocity at same elevation as u
d = aerodynamic (equivalent sphere) particle diameter
For a particle-size distribution containing a mixture of fine,
intermediate, and coarse particles, the isokinetic correction factor is
an average of the above factors weighted by the relative proportion of
coarse and fine particles, for example, if the mass of fine particles
in the distribution equals twice the mass of the coarse particles, the
weighted isokinetic correction for exposure would bei
1/3 [2 (U/u) + 1]
3.0 Particle Size Distribution
As stated above, a cyclone preseparator (Sierra Instruments Model
230-CP) was used in conjunction with a high-volume cascade impactor
(Sierra Instruments Model 235) to measure airborne particle size distri-
bution. The purpose of the preseparator was to remove coarse particles
which otherwise would tend to bounce through the impactor to the back-up
filter, thereby causing fine particle measurement bias, fable A-l gives
the 50% cutoff diameters for the cyclone precollector and the impaetion
stages.
Based on laboratory calibration with monodisperse spheres of unit
density, the cyclone was found to have a 50% cutoff diameter of 5.5 Mm
for a flowrate of 40 cfm. The manufacturer recommends that the value of
11 wm be used for the cutoff diameter at 20 cfm, reflecting an inverse
proportion between the cutoff diameter and the flow rate. However, while
some data have been compiled to support this dependence for small cyclones,
which is presumed to be the result of turbulence effects, other data for
lower inlet velocities seem to indicate that an inverse dependence of
A-3
-------
TABLE A-l. 50% CUTOFF DIAMETERS FOR SIERRA CYCLONE PRESEPARATOR
AND CASCADE IMPACTOR OPERATED AT 34 m3/hr (20 cfm)
. __ ___ _
Particle density 1 g/cm3 2 g/cnP 2.5 g/crn.3 3 g/cm3 4 g/cm3 5 g/ca?
Cyclone 11 7*8 7*0 6.3 5.5 4.9
Stage 1
Stage 2
Stage 3
Stage 4
Stage 5
10.2
4.2
2.1
1.4
0.73
7.2
3.0
1.5
0.99
0.52
6.4
2.7
1.3
0.88
0.46
5.9
2.4
1.2
0.81
0.42
5.1
2.1
1.0
0.7
0.36
4.6
1.9
0.94
0.63
0.33
cutoff diameter on the square root of flow rate may apply, as dictated by
traditional cyclone performance theory .^li/ Nevertheless, the manufacturer's
recommendation was followed in this study.
As indicated by the simultaneous measurement of airborne particle-
size distribution, one impactor being used with a precollector and a
second without a precollector, the cyclone precollector is very effective
in reducing fine particle measurement bias. However, the following
observations indicate that correction for residual coarse particle
bounce is needed:
1. There is a monotonic decrease in collected particulate weight
on each successive impaction stage followed by a several-fold increase
in weight collected by the back-up filter.
2. Because the assumed value (0.2 £tm)* for the effective cutoff
diameter of the glass fiber back-up filter fits the progression of
cutoff diameters for the impaction stages, the weight collected on the
back-up filter should follow the particulate weight progression on the
impactor stages.
The excess particulate on the back-up filter is postulated to
consist of coarse particles that penetrated the cyclone (with small
probability) and bounced through the impactor.
Average of 0.3 £im, for which a high percentage of particulate is
known to be removed by filtration, and 0,1 p,n, which is frequently
cited as the lower limit of particle removal for glass fiber filters.
A-4
-------
To correct the measured particle size distribution for the effects
of residual particle bounce, the following procedure was used:
1. The calibrated cutoff diameter for the cyclone preseparator
was used to fix the upper end of the particle-size distribution.
2. At the lower end of the particle-size distribution, the partieu-
late weight on the back-up filter was reduced to fit the particulate
weight distribution of the impactor stages, thereby extending the monotonic
decrease in particulate weight observed on the impactor stages.
The log-normal distribution determined in this manner is extrapo-
lated to larger particle sizes as required for the calculations.
4.0 Adjustment of Imission Factors toParticle Size Cutoffs
In the body of this report, emission factors are presented for
suspended particulates (particles smaller than 30 /urn in Stokes diameter,
based on a particle density of 2.5 g/cnr) and for fine particulates
(particles smaller than 5 pm in Stokes diameter, based on a particle
density of 2.5 g/cm^>. These values are determined by multiplying the
total emission factor by appropriate weight percentage values from the
particle size distribution corrected to a particle density of 2.5 g/cm .
In order to find emission factors corresponding to other particle
size cutoffs, the following steps must be taken:
1. For a given test, construct a straight-line particle size
distribution on log-probability graph paper using the values for weight
per cents smaller than 30 and 5 jura.
2. Determine the value for weight percent smaller than the desired
diameter (D ).
3. Calculate the emission factor for particles smaller than D
using the following expression:
EF = IF
-------
TABLE A-2. EXAMPLE CALCULATION FOR KJN G-29—UNPAVID iDADS
Result
Metric
Nonmetric
D.
Plot filter exposure versus sampler
height.
Graphically integrate to determine
the area under the vertical ex-
posure profile.
Divide B by the number of vehicle
passes (78) to arrive at the
integrated filter exposure.
Correct C to isokinetic conditions
following the procedure given
in Appendix A*
Multiply D by the ratio of the
percent < 30 j*m (50%) over
the percent captured on the
filter (57%) to obtain the
emission factor for particles
smaller than 30 /xm.
135 kg/tan
1.7 kg/vehicle
km
1*5 kg/vehicle
km
1.6 kg/vehicle
km
480 Ib/mile
6»1 lb/vehicle
mile
5*4 lb/vehicle
mile
5.6 lb/vehicle
mile
A-6
-------
TABLE A-3. EXAMPLE CALCULATION FOR RUN F-18—PAVED IOADS
Result
Metric
Nonmetric
A* Plot filter exposure versus sampler
height*
B* Graphically integrate to determine
the area under the vertical ex-
posure profile*
C« Divide B by the number of vehicle
passes (96) to arrive at the
integrated filter exposure*
D« Correct C to isokinetie conditions
following the procedure given
in Appendix A*
E« Multiply D by the ratio of the per-
cent < 30 fj.m (78%) over the per-
cent captured on the filter (72%)
to obtain the emission factor for
particles smaller than 30 jim.
19 kg/km
67 Ib/mile
0.20 kg/vehicle 0.70 Ib/vehicle
km mile
0.12 kg/vehicle
km
0.14 kg/vehicle
km
0.44 lb/vehicle
mile
0.48 lb/vehicle
mile
A-7
-------
TABLE A-4* EXAMPLE CALCULATION FOR RUN H-12—STORAGE PILE STACKING
Result
Metric
Nonmetric
A. Plot filter exposure versus sampler
height*
B* Graphically integrate to determine
the area under the exposure sur-
face*
C» Divide B by the number of stacker
passes*
D« Multiply C by the stacker velocity
(mph or Wsec) and the inverted
stacking rate (hr/ton or hr/toime)
to arrive at the integrated fil-
ter exposure*
E« Correct D to isokinetic conditions
following the procedure given
in Appendix A*
P* Multiply E by the ratio of the per-
cent < 21.4 jum (67%) over the
percent captured on the filter
(61%) to obtain the emission
factor for particles smaller
than 21.4 /urn. (This correction
simulates what the sampling
equipment "sees" as particles
< 30 im when density is 4.9
g/ern3.)
91 kg
5.6 kg/stacker
km
0.0010 kg/t
200 Ib
20 Lb/stacker-
mile
0.0020 Ib/T
0,0010 kg/t
0.0011 kg/t
0.0021 Ib/T
0.0023 Ib/T
A-8
-------
REFERENCE
A-l. Chan, T.» and M« Lippman. Particle Collection Efficiencies of Air
Sampling Cyclones! An Empirical Theory. Environmental Science and
Technology, Il(4)s377» 1977.
A-9
-------
APPENDIX B
PROCEDURES FOR SURFACE AGGREGATE
SAMPLING AND ANALYSIS
B-l
-------
The predictive emission factor equations presented in this report
require data on the properties of the dust-emitting aggregate materials
being disturbed by the action of wind or machinery. This appendix
presents recommended procedures for collection, preparation and labora-
tory moisture analysis of representative samples of loose aggregate
materials from the surfaces of: (a) unpaved roads; (b) paved roads;
(e) storage piles; and (d) exposed areas.
The starting point for development of the recommended procedures
was a review of American Society of Testing and Materials (ASTM) Standards
in search of standard methodologies applicable to the specific sampling
and analysis problems.
When it was practicable, the recommended procedures were structured
identically to ASTM standard procedures. When this was not possible,
the attempt was made to develop the procedure in a manner consistent
with the intent of the majority of pertinent ASTM Standards.
1.0 Number and Size of Incremental and Gross Samples
ASTM Standards generally suggest that (a) the number of gross
samples to be taken is one per 900 tonnes (1,000 tons) of material;
(b) the minimum size of a gross sample should range from 14 kg (30 Ib)
to 230 kg (500 Ib) depending on the type and size distribution of the
material; and (c) the number of incremental samples should range from 3
to 50. These general requirements apply to aggregate materials but not
necessarily roadway surface materials.
The recommendations presented below are based on a desire to approach
representative sampling yet remain within reasonable constraints of
manpower and time. It is recommended that 23-kg (50-lb) gross samples
be collected in a number of increments ranging from 10 for storage piles
to 4 for unpaved roads. Paved road samples, while normally consisting
of less than 23 kg (50 Ib), will comprise a number of increments.
For a typical unpaved road, 9.1 m (30 ft) in width and having
5/8 cm (1/4 in.) of loose surface material (1.5 g/cm bulk density),
there are approximately 140,000 kg (300,000 Ib) or 140 tonnes (150 tons)
of material in 1 mile. Consequently, one gross sample of at least 23 kg
(50-lb) weight taken in at least four increments from a 16 km (10-mile)
section of roadway (having similar surface material) would satisfy
general ASTM criteria.
For a four-lane paved road of 15-m (50-ft) width, there is typically
230 kg (500 Ib) of surface dust per road mile. To satisfy ASTM criteria,
approximately 150 m (500 ft) of road length in increments stretching
B-2
-------
over 1,600 km (1,000 miles) would have to be sampled to achieve 23 kg
(50 Ib) of sample. Since this would involve an excessive commitment of
time and manpower, a number of incremental samples of small size over a
road segment not exceeding 40 km (25 miles) (having similar surface
conditions) is recommended in order to reasonably approach representative
sampling of paved roadway surfaces.
Because aggregate storage piles typically contain several thousand
tons of material, it is recommended that one 23-kg (50-lb) gross sample
consisting of 10 increments be collected from each pile. For very large
piles exceeding 90,000 tonnes (100,000 tons), more than one gross sample
should be taken.
Assuming that an exposed ground area is covered by a 5/8-cm
in.) thick layer of loose sand, soil, or crushed stone (1.5 g/cm? bulk
density), 0.004 sq km (1 acre) would have 39,000 kg (85,000 Ib) or 39 tonnes
(43 tons) of surface material. Thus, one gross sample for every 0.1 sq km
(25 acres) of exposed area would be consistent with ASTM Standards.
Where there are large acreages of exposed area, it is recommended that
one 23-kg (50-lb) gross sample be collected for every major exposed
surface type (e.g., tailings, glacial drift, etc.).
2.0 Collection of^ incremental and Gross Samples
This section will discuss the appropriate sample collection technique
for each source type.
2.1 Unpaved Roads
The incremental samples from unpaved roads should be distributed
over the road segment, as shown in Figure B-l. At least four incremental
samples should be collected. If the surface condition of the road
varies significantly, it must be broken into smaller sampling segments,
each having a relatively uniform condition.
The loose surface material is removed from the hard road base with
a whisk broom and a dustpan. Figure B-2 presents a data form to be used
for the sampling of unpaved roads.
2.2 Paved Roads
Ideally, for a given road type (residential, commercial, industrial,
etc.), one gross sample per every 40 km (25 miles) should be collected.
This gross sample should consist of at least two separate increments per
travel lane. Thus, the gross sample collected from a four-lane roadway
would consist of eight sample increments.
B-3
-------
td
K
L= 16km (IQMi.)
-H
o
CO
-Sample Strip 20cm (Sin.) Wide
L = 5.3km (3.3 Mi.)
Figure B-l. Location of incremental sampling sites on an unpaved road.
-------
MRI Project
No
MIDWEST RESEARCH INSTITUTE
Sampling Data
Unpaved Roads
Date .
Recorded by.
Type of Material Sampled:.
Site of Sampling:———
SAMPLING METHOD
1. Sampling device: whisk broom and dust pan
2. Sampling depth; loose surface material
3. Sample container; metal or plastic bucket with sealed poly liner
4, Gross sample specifications;
(a) 1 sample of 23kg (50 Ib.) minimum for every 16km (lOmi.) sampled
(b) composite of 4 increments^ lateral strips of 20cm (Sin.) width extending over traveled
portion of roadway half
Indicate deviations from above method;
SAMPUNG DATA
Sample
No.
Time
Location
Surface
Area
Depth
Quantity
of Sample
DIAGRAM
4/78
Figure B-2. Sampling data form for unpaved roads,
B-5
-------
Figure B-3 presents a diagram for the above sampling situation.
Each incremental sample should consist of a lateral strip 0.3 to 3 m (1
to 10 ft) in width across a travel lane. The exact width is dependent
on the amount of loose surface material on the paved roadway. For a
visually dirty road, a width of 0.3 m (1 ft) is sufficient; but for a
visually clean road, a width of 3 m (10 ft) is needed to obtain adequate
sample.
The above sampling procedure may be considered as the preferred
method of collecting surface dust from paved roadways. In many instances,
however, the collection of eight sample increments may not be feasible
due to manpower, equipment, and traffic/hazard limitations. Samples of
questionable representativeness can be obtained from a single increment
(curb to curb) on a given roadway. When it is necessary to resort to
this sampling strategy, care must be taken to select sites that have
dust loading and traffic characteristics typical of the entire roadway
segment of interest. In this situation, sampling from a strip 3 m to
9 m (10 to 30 ft) in width is suggested. From this width, sufficient
sample can be collected, and a step forward representativeness in sample
acquisition will be accomplished.
Samples are removed from the road surface by vacuuming, preceded .by
broom sweeping if large aggregate is present. Figure B-4 presents a
data form to be used for the sampling of paved roads.
As indicated previously, values for the dust loading on the traveled
portion of the roadway are needed for inclusion in the emission factor
equation. Information pertaining to dust loading on curb and parking
areas is useful in estimating carry-on potential or to justify the need
for roadway cleaning.
2.3 Storage Piles
In sampling the surface of a pile to determine representative
properties for use in the wind erosion equation, a gross sample made up
of top, middle, and bottom incremental samples should ideally be obtained
since the wind disturbs the entire surface of the pile. However, it is
impractical to climb to the top or even middle of most industrial storage
piles because of the large size.
The most practical approach in sampling from large piles is to
minimize the bias by sampling as near to the middle of the pile as
practical and by selecting sampling locations in a random fashion.
Incremental samples should be obtained along the entire perimeter of the
pile. The spacing between the samples should be such that the entire
pile perimeter is traversed with approximately equidistant incremental
samples. If small piles are sampled, incremental samples should be
collected from the top, middle, and bottom.
B-6
-------
40km (25mi.) of similar road type
8
Increment I
Figure B-3. Location of incremental sampling sites on a paved road.
-------
MIDWEST RESEARCH INSTITUTE
MRI Project
No
Sampling Data
Paved Roads
Date
Recorded by
Type of Material Sampled:,
Site of Sampling:
Type of Pavement: ___^_
.Surface Condition
SAMPLING METHOD
1. Sampling device: Portable vacuum cleaner (broom sweep first if loading is heavy)
2, Sampling depth: loose surface material
3. Sample container: metal or plastic bucket with sealed poly liner for coarse particles,
vacuum cleaner bag for Fine particles
4, Gross sample specifications:
(a) 1 sample for significant road segment with given surface characteristics -
not to exceed 40km (25 mi.)
(b) composite of 8 increments: lateral strips of 0.3 to 3m (1 to 10ft.) width, extending
over traveled portion of roadway half
Indicate deviations from above method: __
SAMPLING DATA
Sample
No.
Vac
Bag
Time
Surface
Area
Quantity
of Sample
Sample
No.
Vac
Bag
Time
Surface
Area
Quantity
of Sample
DIAGRAM: C « curb P - parking or travel lane T = travel lane
-C2
-P2
-T2
-C4
-P4
-T4
-PI
-Cl
-T3
~P3
-C3
4/78
Figure B-4. Sampling data form for paved roads.
-------
An incremental sample (e.g., one shovelful) is collected by skimming
the surface of the pile in a direction upward along the face. Every
effort must be made by the person obtaining the sample not to purposely
avoid sampling larger pieces of raw material, figure B-5 presents a
data form to be used for the sampling of storage piles.
In obtaining a gross sample for the purpose of characterizing a
load-in or load-out process, incremental samples should be taken from
the portion of the storage pile surface (a) which has been formed by
the addition of aggregate material or (b) from which aggregate material •
is being reclaimed. Usually, it is not feasible to sample the aggregate
material before or after it is in place in the pile.
2.4 Exposed Areas
The selection of incremental sampling locations for exposed areas
should be done prior to obtaining samples. The exposed areas must be
identified, preferably on a map; and the sites selected so that 10
incremental sampling sites cover the major acreage of similar surface
type as equally spaced as possible.
At each incremental sampling site, a 0.3-m (1-ft) square section
should be selected in a random manner within the area previously designated.
If the surface is smooth, as a tailings basin might be, the 0.3-m (1-ft)
square can be swept down to hardpan with a dustpan and a whisk broom.
If 2.3 kg (5 Ib) of material is not collected, the size of the square
should be expanded until at least 2.3 kg (5 Ib) is gathered. If the
surface is rough (e.g., a plowed field), a thin layer of the surface
must be removed with a straight-edged shovel from the entire 0.3-m (1-
ft) square. The size of this square should be increased until 2.3 kg
(5 Ib) is gathered.
3.0 Sample Preparation
Once the 23-kg (50-lb) gross sample is brought to the laboratory,
it must be prepared for silt and moisture analysis. This entails
dividing the sample to a workable size.
A 23-kg (50-lb) gross sample can be divided by usings (a) mechanical
devices; (b) alternate shovel method; (c) riffle; or (d) coning and
quartering method. Mechanical division devices are not discussed in
this section since they are not found in many laboratories. The alter-
nate shovel method is actually only necessary for samples weighing
hundreds of pounds. Therefore, this report discusses only the use of
the riffle and the coning and quartering method.
B-9
-------
MRI Project
No._
MIDWEST RESEARCH INSTITUTE
Sampling Data
Storage Piles
Date
Recorded by.
Type of Material Sampled:.
Site Of
SAMPLING METHOD
1. Sampling device: pointed shovel
2. Sampling depth: 10-15cm (4-6 inches)
3. Sample container: metal or plastic bucket with sealed poly liner
4, Gross sample specifications;
(a) 1 sample of 23kg (50 Ib.) minimum for every pile sampled
(b) composite of 10 increments
5. Minimum portion of stored material (at one site) to be sampled: 25%
Indicate deviations from above method:
t:
SAMPLING DATA
Sample
No.
Time
Location (Refer to map)
Surface
Area
Depth
Quantity
of Sample
4/78
Figure B-5. Sampling data form for storage piles.
-------
ASTM Standards describe the selection of the correct riffle size
and the correct use of the riffle. Riffle slot widths should be at
least three times the size of the material being divided. The following
M™."| /
quote describes the use of the riffle: - '
Divide the gross sample by using a riffle, liffles properly
used will reduce sample variability but cannot eliminate it.
Riffles are shown in Figure B-6, (a) and (b) . Pass the material
through the riffle f rom .a feed scoop, feed bucket, or riffle pan
having a lip or opening the full length of the riffle. When using
any of the above containers to feed the riffle, spread the material
evenly in the container, raise the container, and hold it with its
front edge resting on top of the feed chute, then slowly tilt it so
that the material flows in a uniform stream through the hopper
straight down over the center of the riffle into all the slots,
thence into the riffle pans, one half of the sample being collected
in a pan. Under no circumstances shovel the sample into the riffle,
or dribble into the riffle from a small-mouthed container. Do not
allow the material to build up in or above the riffle slots. If it
does not flow freely through the slots, shake or vibrate the riffle
to facilitate even
The procedure for coning and quartering is best illustrated in
Figure B-7. The following is a description of the procedure:
(1) Mix the material and shovel it into a neat cone;
(2) flatten the cone by pressing the top without further
mixing; (3) divide the flat circular pile into equal quarters
by cutting or scraping out two diameters at right angles;
(4) discard two opposite quarters; (5) thoroughly mix the two
remaining quarters, shovel them into a cone, and repeat the
quartering and discarding procedures until the sample has been
reduced to 0.9 to 1.8 kg (2 to 4 Ib). Samples likely to be
affected by moisture or drying must be handled rapidly, preferably
in an area with a controlled atmosphere, and sealed in a container
to prevent further changes during transportation and storage. Care
must be taken that the material is not contaminated by anything
on the floor or that a portion is not lost through cracks or holes.
Preferably, the coning and quartering operation should be conducted
on a floor covered with clean paper. Coning and quartering is a
simple procedure which is applicable to all powdered materials and
to sample sizes ranging from a few grams to several hundred pounds.
The size of the laboratory sample is important — too little sample
will not be representative and too much sample will be unwield. Ideally,
one would like to analyze the entire gross sample in batches, but this
B-ll
-------
is not practical. While all AS1M Standards acknowledge this impraetieality,
they disagree on the exact size, as indicated by the range of recommended
samples, extending from 0.05 to 27 kg (0.1 to 60 Ib),
The main principle in sizing the laboratory sample is to have
sufficient coarse and fine portions to be representative of the material
and to allow sufficient mass on each sieve so that the weighing is
accurate. A recommended rule of thumb is to have twice as much coarse
sample as fine sample. A laboratory sample of 800 to 1,600 g is recom-
mended since that is the largest quantity that can be handled by the
scales normally available (1,600-g capacity).
4.0 Laboratory Analysis of jjamples
Laboratory analysis of the samples to determine silt and moisture
contents will be identical whether the samples originate from storage
piles, roads, or exposed areas. Minor differences will occur for drying
materials with chemically bound moisture.
4,1 Moisture Analysis
The basic recommended procedure for moisture analysis is determination
of weight loss on oven drying. Table B-l presents a step-by-step procedure
for determining moisture content. Exceptions to this general procedure
are made for any material composed of hydrated minerals or organic
materials. Because of the danger of measuring chemically bound moisture
for these materials if they are over-dried, the drying time should be
lowered to only 1-1/2 hr. Coal and soil are examples of materials that
should be analyzed by this latter procedure.
4.2 Silt Analysis
The basic recommended procedure for silt analysis is mechanical,
dry sieving. A step-by-step procedure is given in Table B-2. The
sieving time is variable; sieving should be continued until the net
sample weight collected in the pan increases by less than 3.0% of the
previous net sample weight collected in the pan. A minor variation of
3.0% is allowed since some grinding will occur, and consequently, the
weight will continue to increase. When the change reduces to 3.0%, it
is hoped that the natural silt, has been passed through the No. 200 sieve
screen and that any additional increase is due to grinding.
B-12
-------
TABLE B-l. MOISTURE ANALYSIS PROCEDURES
1. Preheat the oven to approximately 110°C (230°F). Record oven temperature.
2. Tare the laboratory sample containers which will be placed in the oven.
Tare the containers with the lids on if they have lids. Record the tare
weight(s). Check zero before weighing.
3. Record the make, capacity, smallest division, and accuracy (if displayed)
of the scale.
4. Weigh the laboratory sample in the container(s). Record the combined
weight(s). Check zero before weighing.
5. Place sample in oven and dry overnight.—
6. Remove sample container from oven and (a) weigh immediately if uncovered,
being careful of the hot container; or (b) place tight-fitting lid on the
container and let cool before weighing. Record the combined sample and
container weight(s). Check zero before weighing.
7. Calculate the moisture as the initial weight of the sample and container
minus the oven-dried weight of the sample and container divided by the
initial weight of the sample alone. Record the value.
8. Calculate the sample weight as the oven-dried weight of the sample and
container minus the weight of the container. Record the value.
a/ Dry materials composed of hydrated minerals or organic materials like coal
and certain soils for only 1-1/2 hr.
B-13
-------
TABLE B-2. SILT ANALYSIS PROCEDURES
1. Select the appropriate 8-in. diameter, 2-in, deep sieve sizes. Recommended
U.S. Standard Series sizes are: 3/8 in., No. 4, No* 20, No* 40, No* 100,
No* 140, Mo* 200} and a pan* Comparable Tyler Series sizes can also be utilized.
The No, 20 and the No. 200 are mandatory. The others can be varied if the
recommended sieves are not available or if buildup on one particular sieve
during sieving indicates that an intermediate sieve should be inserted.
2. Obtain a mechanical sieving device such as a vibratory shaker or a Roto-*Tap*
3. Clean the sieves with compressed air and/or a soft brush. Material lodged
in the sieve openings or adhering to the sides of the sieve should be re-
moved (if possible) without handling the screen roughly.
4. Attain a scale (capacity of at least 1,600 g) and record make, capacity,
smallest division, date of last calibration, '-and accuracy (if available).
5, Tare sieves and pan. Check the zero before every weighing. Record weights.
6. After nesting the sieves in decreasing order with pan at the bottom, dump
dried laboratory sample (probably immediately after moisture analysis) into
the top sieve. Brush fine material adhering to the sides of the container
into the top sieve and cover the top sieve with a special lid normally pur-
chased with the pan.
7. Place nested sieves into the mechanical device and sieve for 20 min. Remove
pan containing minus No. 200 and weigh. Replace pan beneath the sieves and
sieve for another 10 min. Remove pan and weigh. When the difference between
two successive pan sample weighings (where the tare of the pan has been sub-
tracted) is less than 3.0%, the sieving is complete.
8. Weigh each sieve and its contents and record the weight. Check the zero
before every weighing.
9. Collect the laboratory sample and place the sample in a separate container
if further analysis is expected.
B-14
-------
REFERENCES
1-1 . D2013-72. Standard Method of Preparing Coal Samples for Analysis.
Annual Book of ASTM Standards, 1977.
B-2. Silver-man, Leslie, Charles I. Billings, and Melvin W. First. Particle
Size Analysis in Industrial Hygiene. Academic Press, New York, New York,
pp. 69-70, 1971.
B-15
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-103
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Iron and Steel Plant Open Source Fugitive Emission
Evaluation
5. REPORT DATE
May 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Chatten Cowherd, Jr., Russel Bohn, and
Thomas Cuscino, Jr.
8. PERFORMING ORGANIZATION REPORT NO
9, PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
1AB604
11. CONTRACT/GRANT NO.
68-02-2609, W.A. 3
12. SPONSORING 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; 2-12/78
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES i£RL.RTp project officer is Robert V. Hendriks, MD-62, 919/-
16. ABSTRACT
The report gives results of field tests aimed at increasing the reliability
of equations used to predict emission factors for open fugitive emission sources at
iron and steel plants. The accuracy of previously developed equations is limited by
the restricted number of actual test measurements used as their basis. Results of
18 emission tests of traffic-entrained dust from unpaved roads show that the pre-
viously developed equation for this source predicts measured emission factors with-
in a mean prediction error of 16% of the mean measured value. An adjustment factor
for the number of vehicle wheels should reduce the prediction error. Limited tests
of chemical dust suppressants for unpaved roads indicate a high initial control effic-
iency which decays with passage of vehicles. Six tests of traffic on paved roads indi-
cate that the previous equation consistently underpredicts emissions for paved roads
in industrial plants. This is probably due to the effect of unpaved areas adjacent to
the paved surfaces. Five tests of storage pile stacking show that the previous equa-
tion for continuous drop operations can be improved by adding a correction factor
for drop distance. Twelve tests of wind erosion emissions, using a portable wind
tunnel, yielded sufficient information on the dynamics of wind erosion from storage
piles and bare ground areas.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Pollution Dust
Iron and Steel Industry
Leakage Dust Control
Evaluation Roads
Mathematical Models
Tests Storage
Wind Erosion
b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Fugitive Emissions
Particulate
c. COS AT I Field/Group
13B
11F
14B
12A
02 C
11G
15E
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
139
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220<1 (9-73)
------- |