A/450/3-78/003 rev
Jd States Office of Air Quality EPA-450/3-78-003
•onmental Protection Planning and Standards (Revised)
icy Research Triangle Park NC 27711 August 1978
&EB& A Method ^.N. a 2/711
for Characterization and
Quantification of Fugitive
Lead Emissions from
Secondary Lead Smelters,
Ferroalloy Plants and
Gray Iron Foundries
(Revised)
-------
A Method for Characterization
> and Quantification of Fugitive Lead
I Emissions from Secondary Lead Smelters,
^ Ferroalloy Plants and Gray Iron Foundries
1 (Revised)
Vj
John M. Zoller, George A. Jutze, and Larry A. Elfers
V,
^ PEDCo Environmental, Inc.
i 1 1 499 Chester Road
> Cincinnati, Ohio 45246
Contract No. 68-02-2515, Task No. 7
and Contract No. 68-02-2585, Task No. 10
^ EPA Task Officer: Charles C. Masser
x> Prepared for
^ U.S. ENVIRONMENTAL PROTECTION AGENCY
^ Office of Air, Noise, and Radiation
<^ Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
August 1978
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35) , U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22161.
This report was furnished to the Environmental Protection Agency by
PEDCo Environmental, Inc. , Cincinnati, Ohio 45246, in fulfillment
of Contract No. 68-02-2515. The contents of this report are reproduced
herein as received from PEDCo Environmental, Inc. The opinions, findings,
and conclusions expressed are those of the author and not necessarily
those of the Environmental Protection Agency. Mention of company or
product names is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication No. EPA-450/3-78-003
11
-------
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1-1
2.0 STATUS OF CURRENTLY AVAILABLE INFORMATION 2-1
2.1 Secondary Lead Smelters 2-1
2.2 Ferroalloy Plants 2-5
2.3 Gray Iron Foundries 2-9
3.0 APPLICATION OF FUGITIVE LEAD EMISSION FACTORS 3-1
DEVELOPED FOR OTHER SOURCE CATEGORIES
3.1 Secondary Lead Smelters 3-1
3.2 Ferroalloy Plants 3-2
3.3 Gray Iron Foundries 3-3
4.0 APPLICABILITY OF FUGITIVE LEAD FACTORS DEVELOPED 4-1
FROM A FIELD STUDY
4.1 Secondary Lead Smelters 4-1
4.2 Ferroalloy Plants 4-5
4.3 Gray Iron Foundries 4-9
5.0 STATE OF THE ART FOR DETERMINATION OF INPLANT 5-1
FUGITIVE LEAD EMISSIONS
5.1 General Approaches for Monitoring and 5-2
Analyses
5.2 Sampling Approaches 5-11
5.3 Manpower Estimates 5-23
111
-------
TABLE OF CONTENTS (continued)
Page
6.0 STATE OF THE ART FOR DETERMINATION OF A PLANT 6-1
EMISSION FACTOR FROM AMBIENT SAMPLING
6.1 General Approach - Monitoring and Analysis 6-2
6.2 Specific Approach for Fugitive Lead Sampling 6-15
6.3 Manpower Estimates 6-24
6.4 Calculation of Plant Emission Factor 6-26
7.0 CONCLUSIONS AND FACTOR DEVELOPMENT CONSIDERATIONS 7-1
7.1 Conclusions 7-1
7.2 Factor Development Considerations 7-4
IV
-------
LIST OF FIGURES
Figure Page
4-1 Secondary Lead Smelter Processes 4-3
4-2 Submerged-Arc Ferroalloy Production Process 4-7
4-3 Gray Iron Foundry Process 4-11
5-1 Location of Velometer Measurement Points 5-14
to Achieve Representative Velocity from a
Vane Axel Fan
5-2 Diagram of Traversing System 5-18
5-3 Spacial Diagram of Traversing Systems 5-19
6-1 Maximum Downwind Sampler Distances 6-9
6-2 Maximum Crosswind Sampler Distances 6-11
6-3 Upwind/Downwind Sampler Locations for 6-17
Cement Plant
6-4 Test Program Schedule 6-22
6-5 Field Test Schedule 6-23
-------
LIST OF TABLES
Table
2-1 Probability of Exceeding Lead Concentration 2-4
Levels Near Selected Industries
2-2 Lead Concentration Study Near Two Secondary 2-6
Lead Smelters
4-1 Lead Concent of Various Ferroalloy Ores 4-8
4-2 U.S. Ferroalloy Production in 1975 4-8
4-3 Estimated Amount of Material Charged in the 4-9
Various Types of Foundry Melting Furnaces
in 1975
5-1 Required Fugitive Emission Sampling Equip- 5-6
ment
5-2 Estimated Manpower Requirements with Respect 5-24
to Tasks and Manpower Categories (Manhours)
6-1 Pre-Test Survey Information to be Obtained 6-6
for Application of Fugitive Emission
Sampling Methods
6-2 Atmospheric Stability Categories 6-10
6-3 Estimated Manpower Requirements with 6-25
Respect to Tasks and Manpower Categories
(Manhours)
VI
-------
1.0 INTRODUCTION
Currently, there is insufficient information to deter-
mine if fugitive lead emissions from secondary lead smelt-
ers, ferroalloy plants, and gray iron foundries are sig-
nificant enough to cause violations of a proposed national
ambient air quality standard for lead. Limited data do
show, however, that such a possibility exists.
The purpose of this task is to summarize current in-
formation relative to fugitive lead emissions from these
sources, investigate the application of fugitive lead emis-
sion factors developed for other source categories, and
report the applicability of fugitive lead factors developed
from a field study. Current state of the art techniques are
developed for source measurements of fugitive emissions
(i.e. inplant) and ambient measurements of fugitive emis-
sions (i.e. upwind/downwind). A comparison of both methods
is presented. This report will aid in determining if field
studies are worthwhile and, if so, recommend the types of
studies to be followed.
1-1
-------
2.0 STATUS OF CURRENTLY AVAILABLE INFORMATION
There are only limited data currently available re-
garding quantitative estimates of fugitive lead emissions
from secondary lead smelters, ferroalloy plants, and gray
iron foundries. Also, little information is available
concerning the impact on ambient air quality and possible
violations of an ambient lead standard due to these three
source categories. The extent of this information is sum-
marized for each source category in the following sections.
2.1 SECONDARY LEAD SMELTERS
Data indicate that particulate emissions from secondary
lead blast furnaces contain approximately 23 percent
lead. '2'3'4'5'6' It is unknown if this composition is
also characteristic of fugitive emissions that occur during
charging or slag and lead tapping. Only total particulate
*
fugitive emission factors are available for secondary lead
reverberatory, blast, or pot furnaces; and, these factors
are based on an engineering estimate that fugitive emissions
equal 5 percent of the uncontrolled stack emissions.
The casting emission factor was based on limited tests of
the roof monitor over casting operations of a primary
smelter.
2-1
-------
Using the total particulate fugitive emission estimates
in conjunction with the assumption that the emissions have
the same characteristics as the furnace emissions (i.e., ap-
proximately 23 percent lead), would result in fugitive
emission factors that are based entirely on estimates.
These would be at the very best, "order of magnitude" esti-
mates and are as follows:
Source
Lead and iron
scrap burning
Fugitive emission estimates
Total particulate(7)
0.5-1.0 g/kg scrap
(1.0-2.0 Ib/ton)
Lead only
0.1-0.2 g/kg scrap
(0.2-0.5 Ib/ton)
Sweating furnace
Reverberatory or
blast furnace
Pot furnace
Casting
0.8-1.75 g/kg charge
(1.6-3.5 Ib/ton)
1.4-7.85 g/kg charge
(2.8-15.7 Ib/ton)
0.02 g/kg charge
(0.04 Ib/ton)
0.44 g/kg lead cast
(0.88 Ib/ton)
0.2-0.4 g/kg charge
(0.4-0.8 Ib/ton)
0.3-1.8 g/kg charge
(0.6-3.6 Ib/ton)
0.005 g/kg charge
(0.009 Ib/ton)
0.1 g/kg lead cast
(0.2 Ib/ton)
Estimates of the air quality impact of fugitive lead
emissions from this industry, based on such highly sub-
jective factors, would certainly be subject to criticism if
utilized for proposed standards or for health effects
analyses. Collection of sampling data is necessary to
support development of emission factors.
Limited data were found on ambient air lead concentra-
tions around secondary lead smelters.
(8)
Measurements by
2-2
-------
the Texas Air Control Board show 24-hour lead concentrations
at 76 meters (250 ft) from one secondary lead smelter to be
in the range of 3.3 to 13.7 yg/m , and at 91 meters (300 ft)
from another secondary lead smelter to be in the range of
3 (8 9)
24.8 to 111.6 yg/m . These concentrations were mea-
sured relatively close to the plant, indicating that fugi-
tive emissions could very well be the major contributing
emission source. However, there was a lack of detailed
(Q\
information about the emission sources. It was not
reported whether fugitive sources, or poorly controlled or
uncontrolled nonfugitive sources were believed to be causing
the lead impact.
These limited data show in Table 2-1 that it is likely
that 100 percent of the monthly average lead concentrations
3 (8)
near secondary lead smelters will exceed 5.0 yg/m . As a
result, it appears that the ambient lead concentrations
surrounding secondary lead smelters can be expected to
exceed 1.5 yg/m , 90-day average, on a regular basis.
In another study using samplers at various distances
from two secondary lead smelters, 24-hour lead concentra-
3
tions were 0.5 to 26.5 yg/m when measured 60 meters (197
ft) from one smelter, and 0.2 to 74.0 yg/m when measured 90
meters (295 ft) from the other smelter. ' ' The range of
2-3
------- An error occurred while trying to OCR this image.
-------
concentrations at each receptor are shown in Table 2-2. The
samplers showing the highest lead impact were located near-
est the two plants. Other samplers, located farther from
the sources, showed a decrease in lead concentrations with
distance from the smelters. This again indicated that
fugitive sources (characteristically with low emission
release heights) were the primary contributor to the nearby
measured concentrations. ' The geometric mean ambient
lead concentration near the two smelters was 3.0 yg/m while
in an urban control area away from the smelters the mean was
0.8 yg/m . This is another indication that an assumed
ambient lead standard of 1.5 yg/rn , 90-day average, may be
exceeded around secondary lead smelters.
2.2 FERROALLOY PLANTS
Information on fugitive lead emissions from the ferro-
alloy industry is also lacking. Electric submerged arc
furnaces are the type most widely used for the production of
ferroalloys. The little information that is available on
lead emissions from ferroalloy plants pertains to this
furnace type.
Open electric arc ferroalloy furnaces are considered a
fugitive emission source since the emissions emanate from a
nonconfined area. Emissions from this source are captured
2-5
-------
Table 2-2. LEAD CONCENTRATION STUDY NEAR TWO
SECONDARY LEAD SMELTERS '
Source
and
receptor
Smelter A
Receptor 1
Receptor 2
Receptor 3
Smelter B
Receptor 4
Receptor 5
Receptor 6
Receptor 7
Lead
emissions,
tons/day
(Mg/day)
0.045
(0.041)
0.09
(0.082)
Distance
from
source,
meters
_
60
100
220
_
90
120
130
195
Number
of
samples
_
96
57
94
_
101
64
73
101
2 4 -hour
concentration
range
yg/m
_
0.5-26.5
0.5-9.7
0.2-13.7
_
0.2-74.0
0.4-6.0
0.3-27.5
0.3-5.0
Source: Roberts, T.M., T.C. Hutchinson, J. Paciga, A.
Chattopadhyay, R.E. Jervis, and J. Van Loom.
December 20, 1974. "Lead Contamination Around
Secondary Smelters: Estimation of Dispersal and
Accumulation by Humans," Science, Volume 186 (4169),
pp. 1,120-1,123. (Reference 11, as reported in
Reference 8.)
2-6
-------
by hooding or enclosing the furnace, then removed from the
gas stream by fabric filters, venturi scrubbers, or (less
common) electrostatic precipitators.
The uncontrolled lead emission factor for electric arc
furnaces producing manganese alloys (FeMn and SiMn) was
developed knowing that the lead content of manganese ore
ranges from 0.002 to 0.01 percent by weight, with an average
lead content of 0.005 percent. ' Manganese ore is im-
ported because there is no domestic production of this ore.
Since about 75 percent of the lead in the ore is released in
(12)
the furnace fume, this results in an uncontrolled lead
emission factor of 0.038 g/kg product (0.075 Ib/ton) for
manganese alloys. However, note that there may be wide
variations in the lead emission rates, depending on the ore
used.(13)
The lead emission factor for silicon alloy production
(FeSi, silicon metal, and CaSi) is calculated using the
reported lead content of particulate from a furnace pro-
ducing silicon alloy (0.02% lead)(12/13) and the total
particulate emission factor for silicon alloy furnaces
(weighted by product mix) of 275 g/kg product (550 lb/
ton). ' This results in an uncontrolled lead emission
factor of 0.05 g/kg product (0.1 Ib/ton).(1/12)
2-7
-------
The lead emission factor for production of chromium
alloys is estimated using a reported lead content of 0.001
percent ' in the fume from a furnace producing FeCrSi,
and an uncontrolled total particulate emission factor for
chromium alloy furnaces (weighted by product mix) of 260
g/kg product (520 Ib/ton). ' This results in an uncon-
trolled lead emission factor of 0.0025 g/kg product (0.005
Ib/ton).(12>
These emission rates represent uncontrolled emissions.
In practice, the ferroalloy furnaces are hooded or enclosed
and emissions ducted to control devices. However, there are
likely to be some fugitive emissions from the furnace that
escape capture. Quantitative estimates of emission rates
for leaks due to ineffective capture have not been made,
possibly due to the very site-specific nature of these
emission levels. Emission factors for tapping and casting
emissions also have not been developed. Therefore lead
emission rates from these operations are unknown.
Ferroalloy ore handling is a minor source of fugitive
lead emissions. Total fugitive emissions from ore and raw
materials handling and preparation have been estimated at 5
g/kg alloy produced (10 Ib/ton). ' However, lead con-
tent of FeMn, SiMn, and FeCrSi ores have been reported at
0.01 percent and less. ' If the percent lead in the
2-8
-------
emissions is assumed to be the same as the percent lead in
the raw materials, the lead emission factor for material
handling would be 0.0005 g/kg alloy produced (0.001 Ib/ton).
Assuming this factor valid for the entire 1975 industry
production of 2.0 x 10 Mg (2.2 x 10 tons), the indus-
try-wide fugitive lead emissions from materials handling
would be 1 Mg/yr (1.1 tons/yr).
Two ambient air samples at a ferroalloy plant show 24-
hour lead concentrations at 61 meters (200 ft) from the
3 (8 9)
source to be 2.5 and 4.5 yg/m . These are the only .
data found in the literature and are very little information
on which to estimate the industry-wide fugitive lead impact
from ferroalloy production. However, from these data as
shown in Table 2-1, it appears that there may be an ambient
lead concentration problem surrounding ferroalloy plants.
2.3 GRAY IRON FOUNDRIES
Cupolas are the most widely used furnace for the pro-
duction of gray iron. Other furnace types include electric
arc, electric induction, and reverberatory. Fugitive emis-
sions from the cupola and reverberatory furnaces occur
during charging and tapping. The furnace emissions from
cupolas and reverberatories are not considered fugitive
since they always emanate from a stack. The electric arc
2-9
-------
and induction furnace emissions are considered fugitive for
the charging, tapping, and furnace operations.
The only fugitive emission factors for gray iron
foundry cupola/reverberatory furnaces are for total particu-
late fugitive emissions; and, these factors are based on an
engineering estimate that fugitive emissions equal 5 percent
of the uncontrolled stack emissions reported in AP-42. '
Tests on several cupolas indicate that emissions of
lead and particulate vary considerably, depending on the
quality of the scrap charged, cupola blast velocity, tem-
perature of the melt zone, and lead content. One study
reported a range of 0.5 to 2.0 percent lead in cupola par-
ticulate emissions, with an average of 1.2 percent.
Another study reported a concentration of 2.6 to 3.4 percent
(18)
lead in the emissions. One investigator indicated a
lead content of 1.2 to 5.7 percent, with an average of 4.3
show
(20)
(19)
percent. Tests on a Los Angeles cupola operation showed
that 17 percent of the particulate emissions was lead,
probably attributable to a high percentage of scrap metal in
the charge. Thus the average lead content of emissions from
foundry cupola furnaces is approximately 3 percent.
[Note that increased use of scrap could raise the lead
content of the emissions.] Applying the 3 percent lead to
2-10
-------
the total particulate fugitive emission estimates would
result in the following estimates:
Furnace type
Fugitive emission estimates
per unit weight of metal charged
Total
particulate
Lead only
Cupola
Reverberatory
0.05-1 g/kg
(0.1-2 Ib/ton)
0.05 g/kg
(0.1 Ib/ton)
0.0015-0.03 g/kg
(0.003-0.06 Ib/ton)
0.0015 g/kg
(0.003 Ib/ton)
The emission factor for electric induction furnaces
published in AP-42 for total particulate emissions is
0.75 g/kg of metal charged (1.5 Ib/ton). Total particulate
emission rates for electric arc furnaces have been estimated
at 7.75 g/kg of iron produced (15.5 Ib/ton).
(16)
Of this
total, 0.75 g/kg of iron produced (1.5 Ib/ton) are for
charging and tapping emissions. Charging represents the
bulk (90 to 95%) of the combined charging and tapping emis-
sions .
Since there are no data on the percent lead in particu-
late emissions from electric or reverberatory furnaces,
these emissions may also be assumed to contain 3 percent
lead. Applying the 3 percent lead to the electric induction
and electric arc furnace total particulate emission factors
would result as follows:
2-11
-------
Furnace type
Fugitive emission estimates,
per unit weight of metal
Total Mr. ,,.
particulate(15'16)
Lead only
Electric induction
Electric arc
0.75 g/kg charged
(1.5 Ib/ton)
7.75 g/kg produced
(15.5 Ib/ton)
0.02 g/kg charged
(0.04 Ib/ton)
0.23 g/kg produced
(0.46 Ib/ton)
Again, these estimates would be at very best "order of
magnitude estimates." Factors developed in such a manner
are inadequate to assess the air quality impact of fugitive
lead emissions from these furnace types in the gray iron
industry; at least for the purposes of proposed standards.
Sampling data are necessary in order to develop the needed
emission factors.
Ambient air lead concentrations were measured around
seven gray iron foundries in Texas and are summarized in
Table 2-1. ' The measurements were taken 76 to 168
meters (250 to 550 ft) from the source. The reported 24-
hour lead concentrations ranged from "not detectable" to
50.9 ug/m .
Based on these limited tests (a total of only 16
samples), it can be seen (Table 2-1) that the ambient lead
concentrations surrounding gray iron foundries will gen-
erally exceed 1.5 yg/m .
2-12
-------
REFERENCES FOR SECTION 2.0
1. Control Techniques for Lead Air Emissions. PEDCo
Environmental, Inc. Cincinnati, Ohio. Prepared for
the U.S. Environmental Protection Agency, National
Environmental Research Center under Contract No. 68-
02-1375, Task Order No. 32. January 1977.
2. EPA Test No. 71-C1-27 at American Smelting and Refining
Co. Engineering Science, Inc. for U.S. Environmental
Protection Agency. Research Triangle Park, North
Carolina. February 1972.
3. EPA Test No. 71-C1-30 at West Coast Smelting and Re-
fining Company. Engineering Science, Inc. for U.S.
Environmental Protection Agency. Research Triangle
Park, North Carolina. March 1972.
4. EPA Test No. 71-C1-76 at R.L. Lavin and Sons, Inc.
Engineering Science, Inc. for U.S. Environmental Pro-
tection Agency. Research Triangle Park, North Carolina.
March 1972.
5. EPA Test No. 74-SLD-l. Preliminary Report. Emission
Testing Branch. Environmental Protection Agency.
Research Triangle Park, North Carolina. Contract No.
68-02-0225. Task No. 22, July 1974.
6. Tests No. 72-Cl-7,8,29 and 33. Emission Testing Branch.
Environmental Protection Agency, Research Triangle
Park, North Carolina. Contract No. 68-02-0230. August
1972.
7. Technical Guidance for Control of Industrial Process
Fugitive Particulate Emissions. PEDCo Environmental,
Inc. Cincinnati, Ohio. Prepared for the U.S. Environ-
mental Protection Agency, Office of Air Quality Plan-
ning and Standards under Contract No. 68-02-1375, Task
Order No. 33. Publication No. EPA-450/3-77-010. March
1977.
2-13
-------
8. Standard Support and Environmental Impact Statement:
National Ambient Air Quality Standard for Lead. Chap-
ter 5, Lead Emission Sources and Air Quality Data.
METREK, a Division of Mitre Corporation. March 23,
1977. Draft.
9. A Report of Typical Element Emissions from Texas Smelt-
ers. Texas Air Control Board. Austin, Texas. April
1974a.
10. A Report of Typical Element Emissions from Texas Found-
ries. Texas Air Control Board. Austin, Texas. April
1974b.
11. Roberts, T.M., T.C. Hutchinson, J. Paciga, A. Chat-
top adhy ay , R.E. Jervis, and J. Van Loon. Lead Contami-
nation Around Secondary Smelters: Estimation of Dis-
persal and Accumulation by Humans. Science. Volume
186 (4169): pp. 1,120-1,123. December 20, 1974.
12. Statement by the Ferroalloys Association to National
Air Pollution Control Techniques Advisory Committee
(NAPCTAC) Meeting on Atmospheric Lead Emissions -
Ferroalloy Production. The Ferroalloys Association.
Washington, D.C. March 1977.
13. Engineering and Cost Study of the Ferroalloy Industry.
U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards. Research Triangle
Park, North Carolina. Publication No. EPA-450/2-74-008.
May 1974.
14. Trace Pollutant Emissions from the Processing of Metal-
lic Ores. PEDCo Environmental, Inc. Cincinnati, Ohio.
Prepared for the U.S. Environmental Protection Agency
under Contract No. 68-02-1321, Task Order No. 4. 1974.
15. Compilation of Air Pollutant Emission Factors, Second
Edition. U.S. Environmental Protection Agency, Office
of Air Quality Planning and Standards. Research Tri-
angle Park, North Carolina. Publication No. AP-42.
February 1976.
16. Standards Support and Environmental Impact Statement:
An Investigation of the Best Systems of Emission Reduc-
tion for Electric Arc Furnaces in the Gray Iron Foundry
Industry. "U.S. Environmental Protection Agency, Emis-
sion Standards and Engineering Division. Research
Triangle Park, North Carolina. November 1975. Draft.
2-14
-------
17. Davis, W.E. Emission Study of Industrial Sources of
Lead Pollutants. 1970. W.E. Davis and Associates.
Leawood, Kansas. U.S. Environmental Protection Agency.
EPA Contract No. 68-02-0271. April 1973. 123 p.
18. Kistler, J. Two Modern Methods for Abating Air Pollu-
tion in Foundries and Iron and Steel Works. Giesserei.
Dusseldorf. 43 (13) :333-340. June 1956. Text in
German.
19. Drake, J.F. et.al. Iron Age. 163 (12). 1949. pp.
88-92.
20. Danielson, J.A. (ed.). Air Pollution Engineering
Manual. Second Edition. Air Pollution Control Dis-
trict of Los Angeles. For U.S. EPA. Research Triangle
Park, North Carolina. May 1973. 987 p. Publication
No. AP-40.
2-15
-------
3.0 APPLICATION OF FUGITIVE LEAD EMISSION FACTORS
DEVELOPED FOR OTHER SOURCE CATEGORIES
Because of the lack of information on fugitive lead
emissions from secondary lead smelters, ferroalloy plants,
and gray iron foundries, the application of fugitive lead
emission factors from other source categories was investi-
gated. The major difficulty in applying fugitive lead
emission factors developed for one source to another source
is the variation in emissions caused by differences in raw
materials. While the same general type of equipment may be
used, the difference in raw materials causes different
emission rates and characteristics. The lead emission rates
will likely differ significantly due to differences in lead
contents of the raw materials. Therefore when applying
fugitive lead emission factors developed for one source to
another source category, the total particulate emission
rates and lead contents of the raw materials must be com-
pared.
3.1 SECONDARY LEAD SMELTERS
Although there is more emission information available
for primary lead smelters than secondary lead smelters, the
3-1
-------
emission factors for these two source categories are not
comparable. The total particulate emission rate from a
secondary lead smelter blast furnace is approximately half
the emission rate from a primary lead smelter blast fur-
nace. This may be because ore is used at primary smelt-
ers while scrap is used at secondary smelters. The dif-
ferences in raw materials cause a significant difference in
emission rates and characteristics (i.e., percent lead).
In addition, the fugitive emissions from primary lead smelt-
er dross reverberatory furnaces can not be expected to
approximate the fugitive emissions from a secondary smelter
reverberatory furnace. Total particulate emissions from the
secondary lead smelter reverberatory furnace are more than
seven times the primary smelter dross reverberatory fur-
nace. Lead casting emissions for primary and secondary
lead smelters may be comparable; however, some measured
data or at least a materials-balance analysis would be
required to determine if a valid comparison exists.
3.2 FERROALLOY PLANTS
Although many industries use electric arc furnaces,
emissions generated by electric arc furnaces for the produc-
tion of ferroalloys are unique for the following reason. For
the production of ferroalloys, total particulate emissions
3-2
-------
from open electric arc furnaces range from 97.5 to 312.5
g/kg produced (195 to 625 Ib/ton) depending on the type of
product. By comparison, particulate emission factors for
electric arc furnaces used in steel foundries range from 2
to 20 g/kg metal charged (4 to 40 Ib/ton).^ This large
disparity in emission rates occurs because of the difference
in raw materials. Raw materials used in steel foundries
include scrap steel, pig iron, ferroalloy, and limestone;
while alloy ore (e.g., chrome ore for production of FeCr),
limestone, quartz, coal and wood chips, and scrap iron are
used for the production of ferroalloys. Lead emissions from
tapping and casting ferroalloys cannot be compared to other
industries (such as foundries), primarily because of the
anticipated differences in lead content of the products and
differences in total emission rates. It seems apparent that
emissions, especially lead emissions, generated by the
ferroalloy furnaces are unique to that industry.
3.3 GRAY IRON FOUNDRIES
It is difficult to apply fugitive lead emission factors
developed for other source categories to the gray iron
foundry industry. This difficulty is largely due to the
wide variation that is found in lead content of the scrap
and other raw materials. This variation occurs within the
gray iron foundry industry as well as between that industry
3-3
-------
and other related industries. For example, the lead content
of metals charged in a gray iron electric arc furnace may be
less than or greater than the lead content of metals charged
to a steel foundry electric arc furnace. The specifications
of the finished product influence the quality control of the
scrap metals and the lead content of the charge. There are
also wide variations in total particulate emission rates
between the gray iron foundry industry and other related
industries.
Cupolas are used in foundries primarily for the produc-
tion of gray iron, but may also be used for brass and bronze
production. The total particulate emission factor for a
brass/bronze production cupola is 36.5 g/kg metal charged
(73 Ib/ton) while the gray iron cupola total particulate
emission factor is 8.5 g/kg metal charged (17 Ib/ton).
This difference in emission rates plus an anticipated dif-
ference in lead content of the raw materials, make unlikely
a good agreement of fugitive lead emission factors between
these two sources.
The total particulate emission factor for an electric
induction furnace producing gray iron is 0.75 g/kg metal
charged (1.5 Ib/ton). This is 15 times the particulate
emission factor for an induction furnace in a steel foundry,
but similar to the brass/bronze induction furnace emission
3-4
-------
factor of 1 g/kg metal charges (2 Ib/ton). ' However, it
is not anticipated that the lead emissions for the gray iron
and brass/bronze induction furnaces would be as close as
their respective total particulate emission rates because of
differences in the lead content of the metals usually
charged.
The total particulate emission rate from a gray iron
(2)
electric arc furnace, 7.75 g/kg iron produced (15.5 lb/
ton), is similar to the particulate emission rate from a
steel foundry, 6.5 g/kg metal charged (13 Ib/ton). As
was pointed out, however, the lead contents of the charge
materials may not be the same; in fact, they are often quite
different. Therefore while an operational relationship may
be found between these two source categories, data developed
specifically for gray iron electric arc furnaces would be
much more accurate for that source category.
Reverberatory furnaces producing gray iron have a
particulate emission factor of 1 g/kg metal charged (2
Ib/ton) while steel foundry open hearth furnaces have a
particulate emission factor of 5.5 g/kg metal charged (11
Ib/ton) and brass/bronze reverberatory furnaces have a
particulate emission factor of 35 g/kg metal charged (70
Ib/ton). Again, because of the large disparity among
3-5
-------
these particulate emission factors plus an anticipated
variation of the percent lead in the charge, it is unlikely
that fugitive lead emission factors for the steel and brass/
bronze furnaces would approximate fugitive lead emission
factors for the gray iron reverberatory furnace.
3-6
-------
REFERENCES FOR SECTION 3.0
1. Compilation of Air Pollutant Emission Factors, Second
Edition. U.S. Environmental Protection Agency. Office
of Air Quality Planning and Standards. Research Tri-
angle Park, North Carolina. Publication No. AP-42.
February 1976.
2. Standards Support and Environmental Impact Statement:
An Investigation of the Best Systems of Emission
Reduction for Electric Arc Furnaces in the Gray Iron
Foundry Industry. U.S. Environmental Protection
Agency, Emission Standards and Engineering Division.
Research Triangle Park, North Carolina. November 1975.
Draft.
3-7
-------
4.0 APPLICABILITY OF FUGITIVE LEAD FACTORS
DEVELOPED FROM A FIELD STUDY
Before conducting a field study to measure fugitive
lead emissions and develop emission factors, it must be
determined if such a study at one or several representative
plants would be reasonably applicable for estimation of
emissions from other individual plants in the same source
category.
4.1 SECONDARY LEAD SMELTERS
In 1975, over 548,500 Mg of lead(1) (604,600 tons) was
(2)
produced at approximately 90 secondary lead smelters in the
United States. Two-thirds of the production from the
secondary lead industry is processed in blast furnaces (or
cupolas), with the remaining done in reverberatory furnaces
and pot furnaces.
The three most common grades of lead are soft, semi-
soft, and hard. Soft lead is approximately 99.9 percent
lead and is produced by the pot furnace. Semisoft lead is
approximately 99.6 percent lead and is produced by the
reverberatory furnace. Hard lead is typically between 88
and 95 percent lead and is produced by the blast furnace. '
4-1
-------
Figure 4-1 shows secondary lead smelter processes and
emission points. Blast furnaces generally produce 18 to 73
(4)
Mg per day of lead (20 to 80 ton/day). The furnace is a
vertical production unit which is charged through a door
near the top while blast air is blown in through tuyeres
near the bottom. The process is semicontinuous in that the
charge is added over a period of 1 or 2 days c-md product is
withdrawn nearly continuously during that period. The
charge stock consists of oxidized lead and lead scrap to be
reduced, plus coke for combustion, limestone, scrap iron,
it of
(3,6)
and rerun slag. Approximately 70 percent of the molten
charge material is tapped off as hard lead.
A reverberatory furnace is merely a device for heating
the charge stock by direct contact with the products of
combustion of oil and/or gas burners and by radiation from
the hot walls of the furnace. The charges may be a mixture
of lead scrap, battery plates, oxides, drosses, and lead
residues. These are put into the furnace at regular inter-
vals as the mass of the charge becomes fluid. Molten metal
is tapped off as the level of metal rises.
A typical reverberatory furnace produces 45 Mg/day of
(4)
lead ingot (50 ton/day). About 47 percent of the charge
stock is recovered as metal, 46 percent is recovered as
slag, and 7 percent leaves as smoke and fumes. '
4-2
------- An error occurred while trying to OCR this image.
-------
Pot furnaces generally produce from 0.9 to 45 Mg of
lead per day (1 to 50 ton/day) and are used primarily for
remelting, alloying, and refining processes. In general,
since pot furnaces are indirectly fired, their pollution
potential is much lower than that of blast or reverberatory
furnaces. During melting and holding operations, uncon-
trolled emissions are low because the vapor pressure of lead
is low at the melting temperature. During dross skimming
and refining, however, emissions increase substantially. '
Since approximately two-thirds of the production from
secondary lead smelters is from blast furnaces, testing of
this furnace type would cover the majority of the industry.
Also, plant to plant variations in the operation of blast
furnaces, reverberatory furnaces, and pot furnaces could not
be identified from literature sources. The major difference
that may occur between plants is the variation in quality of
scrap materials used. The effect this variation may have on
fugitive lead emissions is not known but is something that
could be determined as a result of a test program. The
fugitive lead emission rates may be found to relate to the
lead content of the materials charged to the furnace.
Testing of blast/reverberatory furnaces for fugitive
lead will cover the large sources in the secondary lead
4-4
-------
industry. Since blast furnaces produce the majority of
secondary lead, this furnace type warrants emphasis in the
test program. Additional testing of pot furnaces will
complete the sources of fugitive lead emissions in this
industry. All of these sources may possibly be found at one
smelter, but testing of several representative plants is
suggested so that results will be more representative of
industry-wide emission rates. This will also minimize bias
of the results by a single plant and will permit an analysis
relating emission levels to lead content of the raw mate-
rials.
4.2 FERROALLOY PLANTS
The United States produced 2,009,000 Mg (2,215,000
(8) (2)
tons) of ferroalloys in 1975 at 50 plants utilizing
one of six types of operations: electric furnace, vacuum
furnace, induction furnace, blast furnace, electrolytic
process, and exothermic (aluminothermic) process. ' ' '
Of these, over 90 percent are electric submerged—arc fur-
naces. ' There are only six electrolytic processes, six
aluminothermic processes, fewer than five vacuum and induc-
tion furnaces, and two blast furnaces producing ferroalloys
in the United States. ' Since the submerged-arc elec-
tric furnace is by far the most common furnace type, it is
4-5
-------
recommended that any sampling program of the ferroalloy
industry should concentrate on this furnace.
The basic design and operation of all ferroalloy
producing submerged-arc electric furnaces are essentially
the same. Figure 4-2 shows a process flow diagram for
submerged-arc electric furnace ferroalloy production. The
charge consists of raw ore with a reducing agent, such as
alumina, coal and/or coke, and slagging materials such as
silica or gravel. Lead is a naturally occurring trace
element of variable concentrations in the raw materials.
The zone of intense heat (2200 to 2800«C, or 4000 to 5000°F)
around the carbon electrodes is responsible for carbon
reduction of the metallic oxides present. The various
impurities are trapped in the slag, and the molten ferro-
alloy is tapped from the bottom of the furnace and cast. '
Although there is uniform industry-wide operation of
electric arc ferroalloy furnaces, there are major variations
in particulate emission rates depending on the type of
(12)
ferroalloy produced. Also, the lead emission rates
vary depending on the product because the different ores
used have various lead contents. Table 4-1 shows typical
lead analyses that have been reported.
4-6
-------
(fi
en
0)
o
o
0
-H
-p
u
(0
o
OJ
o
M
(0
i
T)
Q)
Cn
M
«•
0)
tn
-H
4-7
-------
Table 4-1. LEAD CONTENT OF VARIOUS FERROALLOY ORES
Ore
FeMn
FeCrSi
SiMn
Percent lead
0.002 - 0.01C
<0.01b
<0.01b
Reference 13.
k
Reference 11.
The production distribution by ferroalloy type is shown
in Table 4-2.
Table 4-2. U.S. FERROALLOY PRODUCTION IN 1975a
Production
Product
Ferro-manganese (FeMn)
Silico-manganese (SiMn)
Ferro- silicon
(FeSi, silvery
iron, silicon metal)
Ferro- Chromiums (FeCr)
Other ferroalloys (FeP,
FeCo, FeTi, etc.)
Total estimated produc-
tion
Mg
551,300
202,100
907,200
222,200
127,000
2,009,800
Tons
607,697
222,772
1,000,000
244,938
140,000
2,215,000
Obtained from Mr. Thomas Jones, U.S. Bureau of Mines,
(6)
Ferroalloys Division, Washington, D.C. July 20, 1976.
This product distribution indicates that representative
results will not be obtained by testing only one plant which
4-8
-------
produces just one type of ferroalloy. Tests must include
the major ferroalloy products; ferro-silicon and ferro-
manganese, followed in importance by ferro-chromiums and
silicomanganese. Tests possibly could be completed at one
facility, but would likely require several plants to com-
plete a full test program.
The test program should also include analysis of the
raw materials (ores) for lead content. This will provide
the necessary data to determine if the fugitive lead emis-
sion factors are related to type of product, lead content of
the ore, or both.
4.3 GRAY IRON FOUNDRIES
Gray iron is produced from cupola, electric, or rever-
beratory furnaces in approximately 1500 foundries in the
United States. ' The estimated throughputs of raw
materials in 1975 are shown in Table 4-3.
Table 4-3. ESTIMATED AMOUNT OF MATERIAL CHARGED IN THE
VARIOUS TYPES OF FOUNDRY MELTING FURNACES IN 19753
Type of
melting
furnace
Cupola
Electric
Reverberatory
Material charged to furnace
Mg
12,400,000
4,260,000
1,090,000
tons
13,700,000
4,700,000
1,200,000
Percent
of total
70
24
6
Source: Communication with Mr. Don Dussy. U.S. Bureau of
Mines. Washington, D.C. July 12, 1976.
4-9
-------
(12)
Approximately 85 percent of the charge is metal; there-
fore a total of 15,100,000 Mg of gray iron (16,700,000 tons)
was produced in 1975. ' Approximately 70 percent of the
electric furnace production is from electric arc furnaces.
(14)
The remainder is from electric induction furnaces. ' The
iron foundry processes are shown in Figure 4-3. [Note: not
shown in this figure is the reverberatory furnace, since it
is a minor part of the gray iron industry."]
The cupola utilized is similar to the blast furnace
used in the iron and steel industry. The furnace is a
firebrick-lined vertical cylindrical steel shell, approxi-
mately 0.7 to 2.9 meters in diameter (27 to 108 in), sup-
ported on structural steel legs. Air is supplied near the
base of the cupola through a windbox and tuyeres.
The cupola is prepared for melting by securing the
bottom drop door, placing a layer of sand over the door to
prevent heat damage, closing the tap and slag holes, and
charging coke for the bed. The bed is ignited and allowed
to burn through. The charges of coke, flux (limestone,
fluorspar and soda ash), and metal (pig iron, scrap and
steel) are placed in alternate layers up to the charge door
which is 4.9 to 6.7 meters above the bottom (16 to 22 ft).
The blast air is then turned on, and melting begins. Charg-
4-10
-------
< O
Q Z O 1-1
Z Qi Z t—
I/) --' U- Z 1-1 O —i O
U.
~— •
-C3
z
: o ^1-1
GO
5!
o
<
U- O I— O 1-1
>-.ai
3«t<:cicrD
_l_lh-_Jl
ooot-Jo
Q.Q.Q_UJZ
ZDO^ — 1>— i
CJ O O LU — -L
_
O^^«C O • — -
Q:u.3:z
i-i- — IT) < O
i-tO
OIDOOOZ
.^o
-------
ing continues until the desired quantity has been melted,
after which the blast is shut off and the furnace bottom is
dropped, allowing the remaining excess charge to fall to the
foundry floor or into a charging box. This material is
recharged during the next operating cycle. '
Operating factors are broken down into two distinct
groups: 1) methods of operations, such as blast rate and
temperature, type of lining, operating variables of the
afterburners; and 2) the quality of charge materials,
including metal to coke ratios, use of oxygen or natural
gas, and the use of coke briquettes. ' '
Electric arc furnaces are commonly used in the second-
ary melting of iron where special alloys are to be made.
These furnaces may be either the direct or indirect arc
type. Pig iron and scrap iron are charged to the furnace
and melted, and alloying elements and fluxes are added at
specified intervals. These furnaces have the advantage of
(3 fi^
rapid and accurate heat control.
Since no gases are used in the heating process, some
undesirable effects on the metal are eliminated. Since arc
furnaces in the iron industry are virtually always used to
prepare special alloy irons, the quality of the material
charged is closely controlled. The charging of greasy
4-12
-------
scrap, which would emit combustible air contaminants, would
only needlessly complicate the alloying procedure. '
Channel and coreless types of electric induction
furnaces are used for melting cast iron. In this type of
furnace, alternating current is passed through a primary
coil with a solid iron core or hollow barcoil. The molten
iron contained within a loop that surrounds the primary coil
acts as the secondary coil. The alternating current (from
60 to 1000 hz) that flows through the primary coil induces a
current in the loop. The electrical resistance of the
molten metal creates the heat for melting. The heated metal
in the channel type circulates to the main furnace chamber
and is replaced by cooler metal. This circulation results
/ -a c \
in uniform metal temperature and alloy composition. '
Use of induction melting has grown during the last
decade, principally because of its potential for air pollu-
tion control. No fossil fuels are used, no significant
metal oxidation takes place during melting, and contamina-
tion of the charge is minimal.
A reverberatory furnace operates by radiating heat from
the burner flame, roof, and walls onto the charged material.
The reverberatory furnace usually consists of a shallow
rectangular refractory hearth for holding the metal charge.
4-13
-------
The furnace is enclosed by vertical side walls and covered
with a low, arched, refractory-lined roof. Combustion of
fuel occurs directly above the molten bath; the walls and
roof receive radiant heat from the hot combustion products
and reradiate this heat to the surface of the bath. These
furnaces are being phased out of production due to curtail-
ments of natural gas and oil supplies. '
In addition to expected plant-to-plant differences in
the types of furnaces used, different types of scraps and
other raw materials are used in various foundries depending
on end product specifications. Tests have shown that
foundry furnace emissions range from 0.5 to 17 percent
lead. Foundries using high quality scrap and/or ingots
would likely have lower fugitive lead emissions than found-
ries using lead contaminated scrap. The operations of iron
innoculation (where done), pouring into molds, casting,
shakeout, cleaning and finishing are relatively independent
of the furnace type used to melt the metal. These opera-
tions are conducted in a consistent manner throughout the
industry.
The melting furnaces are the major sources of fugitive
lead emission found in the gray iron foundry. To obtain
representative test results, the cupola must be given the
highest priority for sampling since it produces 70 percent
4-14
-------
of all gray iron. Electric arc furnaces should have the
next highest priority in the sampling program. These two
furnace types account for over 85 percent of all gray iron
production. Since electric induction and reverberatory
furnaces are a minor part of the gray iron industry, these
furnaces would have low priority in a sampling program.
The quantities of lead emissions expected from iron
innoculation, pouring into molds, casting, shakeout, clean-
ing and finishing are unknown. However these lead emissions
are expected to be small since it is likely that the major-
ity of lead contained in the furnace charge material is
emitted from the furnace as fume or removed in the slag.
Tests are needed to determine the actual lead emission
levels from these sources.
4-15
-------
REFERENCES FOR SECTION 4.0
1. U.S. Department of Interior, Bureau of Mines. Washing-
ton, D.C. 1975.
2. Standard Support and Environmental Impact Statement:
National Ambient Air Quality Standard for Lead.
Appendices. METREK, a Division of Mitre Corporation.
Draft.
3. Danielson, J.A. (ed.). Air Pollution Engineering
Manual. U.S. Environmental Protection Agency. Office
of Air Quality Planning and Standards. Research
Triangle Park, N.C. Publication No. AP-40. May 1973.
4. Background Information for Proposed New Source Perform-
ance Standards: Secondary Lead Smelters and Refin-
eries. Volume I, Main Text. U.S. Environmental Pro-
tection Agency. Office of Air Quality Planning and
Standards. Research Triangle Park, North Carolina.
Publication No. APTD-1352a. June 1973.
5. Hardinson, L.C. Study of Technical and Cost Informa-
tion for Gas Cleaning Equipment in the Lime and Second-
ary Non-Ferrous Metallurgical Industries. National
Technical Information Service. Stamford, Connecticut.
PB-198-137. December 1970.
6. Control Techniques for Lead Air Emissions. PEDCo
Environmental, Inc., Cincinnati, Ohio. Prepared for
the U.S. Environmental Protection Agency. National
Environmental Research Center under Contract No.
68-02-1375, Task Order No. 32. January 1977.
7. Nance, J.T. and K.D. Luedtke, Lead Refining In: Air
Pollution Engineering Manual, U.S. DHEW, NCAPC. PHS
Publication No. 999-AP-40. Cincinnati, Ohio. 1967.
p. 302.
8. Matthews, N.A. Ferroalloys. Preprint from the 1974
Bureau of Mines Mineral Yearbook. U.S. Department of
Interior. Bureau of Mines. Washington, D.C. 1974.
4-16
-------
9. Sansom, R.L. Development Document for Proposed Ef-
fluent Limitations, Guidelines and New Source Per-
formance Standards for the Smelter and Slag Processing
Segment of the Ferroalloy Manufacturing Point Source
Category. Environmental Protection Agency, Contract
No. 440/1-73/ 008, August 1973.
10. Background Information for Standards of Performance:
Electric Submerged-Arc Furnaces for Production of
Ferroalloys, Volume 1: Proposed Standards. U.S.
Environmental Protection Agency, Emission Standards and
Engineering Division. Research Triangle Park, North
Carolina. Publication No. EPA-450/2-74-018a. October
1974.
11. Engineering and Cost Study of the Ferroalloy Industry.
U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards. Research Triangle
Park, North Carolina. Publication No. EPA-450/2-74-
008. May 1974.
12. Compilation of Air Pollutant Emission Factors, Second
Edition. U.S. Environmental Protection Agency. Office
of Air Quality Planning and Standards. Research
Triangle Park, North Carolina. Publication No. AP-42.
February 1976.
13. Statement by the Ferroalloys Association to National
Air Pollution Control Techniques Advisory Committee
(NAPCTAC) Meeting on Atmospheric Lead Emissions -
Ferroalloy Production. The Ferroalloys Association.
Washington, D.C. March, 1977.
14. Standards Support and Environmental Impact Statement:
An Investigation of the Best Systems of Emission
Reduction for Electric Arc Furnaces in the Gray Iron
Foundry Industry. U.S. Environmental Protection
Agency, Emission Standards and Engineering Division.
Research Triangle Park, North Carolina. Draft.
November 1975.
15. Weisburg, M.I. Field Operations and Enforcement Manual
for Air Pollution Control. Vol. III. Pacific Environ-
mental Services, Inc. Santa Monica, California. For
U.S. Environmental Protection Agency. EPA 70-122.
August 1972.
4-17
-------
5.0 STATE OF THE ART FOR DETERMINATION OF
INPLANT FUGITIVE LEAD EMISSIONS
The three industries, namely secondary lead smelting,
ferroalloy production and gray iron foundries, have similar
fugitive emission sources. The fugitive lead emission
sources include: secondary lead smelting - blast furnace
charging/tapping, reverberatory furnace charging/tapping,
and pot furnace charging/melting/tapping; ferroalloy pro-
duction - materials handling, and electric arc furnace
charging/smelting/tapping; and gray iron foundry - cupola
charging/tapping, and electric arc or induction furnace
chargi ng/melting/tapping.
These operations normally occur inside of buildings
with the fugitive emissions escaping to the atmosphere
through power vents (wall or roof mounted) or natural draft
vents (common in the roofs of large buildings). Open
sources, such as materials handling, also occur. Since the
three industries have these common types of fugitive emis-
sion points, the sampling approaches would be similar. The
remainder of this section discusses the sampling techniques
for power vents, natural draft vents, and materials handling
5-1
-------
operations. Also included is the guasi-stack method, which
requires the temporary enclosure or hooding of the source.
The quasi-stack technique may be used for open sources or
sources within a building, the limiting factor being the
ability to enclose or hood the emission source. Estimates
pertaining to man power requirements for the performance of
these tests are also presented.
5.1 GENERAL APPROACHES FOR MONITORING AND ANALYSES
5.1.1 Determination of Sources for Test
The initial task in a test program would be to conduct
a presurvey. A two man team, consisting of an engineer and
the senior field monitoring technician, would visit the
plant and obtain the following information:
0 Source Isolability. Can the emissions be measured
separately from emissions from other sources? Can
the source be enclosed?
0 Source Location. Is the source indoors or out?
Does location permit access of measuring equip-
ment?
0 Meteorological Conditions. Will wind conditions
or precipitation interfere with measurements?
Will rain or snow on ground effect dust levels?
0 Number and Size of Sources. Are emissions from a
single, well defined location or many scattered
locations? Is source small enough to hood?
0 Homogeneity of Emissions. Are emissions the same
type everywhere at the site? Are reactive effects
between different emissions involved?
5-2
-------
0 Continuity of Process. Will emissions be produced
long enough to obtain meaningful samples?
0 Effects of Measurements. Will installation of
measuring equipment alter the process or the emis-
sions? Will measurements interfere with produc-
tion?
0 Nature of Emissions. Are measurements of parti-
cles, gases, liquids required? Are emissions
hazardous?
0 Emission Generation Rate. Are enough emissions
produced to provide measurable sampling time?
0 Emission Dilution. Will transport air reduce
emission concentration below measurable levels?
Physical Plant Parameters
The following list of physical parameters should be
obtained:
Building layout and plan view of potential study areas
Building site elevations to identify obstructions and
structure available to support test setup
Work flow diagrams
Locations of suitable sampling sites
Physical layout measurements to supplement drawings
Work space required at potential sampling sites
Process Description
The following information pertaining to the process
should be obtained:
Process flow diagram with fugitive emission points
identified
General description of process chemistry
General description of process operations including
initial estimate of fugitive emissions
Drawings of equipment or segments of processes where
fugitive emissions are to be measured
Photographs (if permitted) of process area where
fugitive emissions are to be measured.
5-3
-------
° Inplant Support and Assistance
Items listed below are of extreme value since they will
enable the test team to perform these functions more
efficiently.
Location of available services (power outlets, main
tenance and plant engineering personnel, laboratories,
etc.)
Local vendors who can fabricate and supply test system
components
Shift schedules
Location of Operations Records (combine with process
operation information)
Health and safety considerations
Names, extensions, locations of process foremen and
supervisors where tests are to be conducted
Access routes to the areas where test equipment/instru-
mentation will be located
Names, extensions, locations of plant security and
safety supervisors
Regional meteorological summaries
5.1.2 Selection of Monitoring Equipment
In general, three basic types of fugitive emission
sources are normally encountered, they are: power vents
which can be roof or wall mounted, natural ventilators such
as doors, windows, and large open roof vents, and open
sources such as material storage and transfer operations and
furnace charging operations which are not controlled.
Particulate sampling equipment best suited for these
sampling operations are high volume filtration devices which
can be employed with a various sampling media. For gross
particulate emission measurements, glass fiber filter media
is preferred. The use of size fractionation devices such as
5-4
-------
the virtual impactor on Dichotomas sampler has gained wide
usage for the determination of respirable particulates.
Other filtration devices, employing sampling rates in the 1
to 10 1/min range and membrane filter substrates can be used
for particle size distribution data employing microscopic
techniques.
Beyond the physical collection of the sample other
equipment to measure volumetric gaseous emission rates,
emission gas temperature and barometric pressure are re-
quired.
In addition to the collection and measurement devices,
in many cases, it will be necessary to have duct flanges,
extentions etc. fabricated and in some instances custom
fabricated traverse systems to which the sampling equipment
may be attached such that traverse sampling can be con-
ducted.
In summary, the minimum equipment required for con-
ducting fugitive emission sampling is presented in Table
5-1. Additional discussion of meteorological instrumenta-
tion is contained in Sections 6.1 and 6.2.
5-5
-------
Table 5-1. REQUIRED FUGITIVE EMISSION SAMPLING EQUIPMENT
Equipment
Directional high volume
air samplers
Custom fabricated high
volume traversing air
samplers
Recording meteorological
sy s tern
Respirable dust*
samplers
Rotating vane
anemometers
Hot-film wedge sensors
and data system
Thermocouples and
records
Barometers
Expendable support
equipment and filter
media
Number required
4
7
1
2
3
7
2
1
™
Estimated cost
$ 2,400
2,100
3,000
6,000
450
5,000
2,400
180
8,000
* Optional
5.1.3 The Field Test Plan
The U.S. EPA "Technical Manual for the Measurement of
Fugitive Emissions: Roof Monitor Sampling Method for
Industrial Fugitive Emissions" presents an indepth dis-
cussion pertaining to this topic. The plan as presented,
and extracted from this document is presented below:
5-6
-------
"Measurement programs are very demanding in terms of
the scheduling and completion of many preparatory tasks,
observations at sometimes widely separated locations,
instrument checks to verify measurement validity, etc. It
is therefore essential that all of the experiment design and
planning be done prior to the start of the measurement pro-
gram in the form of a detailed test plan. The preparation
of such a plan enables the investigator to "pre-think"
effectively and cross-check all of the details of the
design and operation of a measurement program prior to the
commitment of manpower and resources. The plan then also
serves as the guide for the actual performance of the work.
The test plan provides a formal specification of the equip-
ment and procedures required to satisfy the objectives of
the measurement program. It is based on the information
collected in the informal pretest survey report and des-
cribes the most effective sampling equipment, procedures,
and timetables consistent with the program objectives and
site characteristics."
5.1.3.1 Outline of Field Test Plan - The following outline
was abstracted from the EPA document "Technical Manual for
the Measurement of Fugitive Emissions: Roof Monitor Sampling
Method for Industrial Fugitive Emissions."
5-7
-------
0 Background
The introductory paragraph containing the pertinent
information leading to the need to conduct the measure-
ment program and a short description of the information
required to answer that need.
° Objective
A concise statement of the problem addressed by the
test program and a brief description of the program's
planned method for its solution.
0 Approach
A description of the measurement scheme and data reduc-
tion methodology employed in the program with a discus-
sion of how each will answer the needs identified in
the background statement.
0 Instrumentation/Equipment/Facilities
A description of the instrumentation arrays to be used
to collect the samples and meteorological data identi-
fied in the approach description. The number and
frequency of samples to be taken and the sampling array
resolution should be described.
A detailed description of the equipment to be employed
and its purpose.
A description of the facilities required to operate the
measurement program, including work space, electrical
power, support from plant personnel, special construc-
tion , etc.
° Schedule
A detailed chronology of a typical set of measurements,
or a test, and the overall schedule of events from the
planning stage through the completion of the test
program report.
0 Limitations
A definition of the conditions under which the measure-
ment project is to be conducted. If, for example,
5-8
-------
successful tests can be conducted only during occur-
rences of certain source operations, those favorable
limits should be stated.
0 Analysis Method
A description of the methods which will be used to
analyze the samples collected and the resultant data,
e.g., statistical or case analysis, and critical
aspects of that method.
0 Report Requirements
A draft outline of the report on the analysis of the
data to be collected along with definitions indicating
the purpose of the report and the audience it is to be
directed to.
° Quality Assurance
The test plan should address itself to the development
of a quality assurance program. This QA program should
be an integral part of the measurement program and be
incorporated as a portion of the test plan either
directly or by reference.
0 Responsibilities
A list of persons who are responsible for each phase of
the measurement program, as defined in the schedule,
both for the testing organization and for the plant
site.
5.1.4 Production Rate Determination
To determine fugitive lead emission rates from each
fugitive emission source, the production rate pertaining to
the specific operation must be measured during the sampling
period. The method for determination of the production rate
will vary with each operation. For example, the tapping
operation of a blast furnace may be based on the quantity of
5-9
-------
molten lead processed per cast, if a "Quasi-stack" sampling
(2)
method is used for the determination of the fugitive
emissions. Should the fugitive measurement be based on a
roof or wall mounted power vent, then measurements of the
total production rate, which is represented by the emissions
through the power vent(s), would be required. In most cases
measurements of this production rate will not be a function
of the emission test team. The plant production engineer
will have this data available. Therefore a strong line of
communication must be established between both parties such
that exact production rates can be obtained during the test
period.
5.1.5 Common Analytical Requirements
Several common analytical requirements exist with
respect to the determination of fugitive lead emission rates
from the three processes. They can be summarized as fol-
lows :
0 Filter media for use with the various sample col-
lection systems. This media must be preweighed
following the published U.S. EPA QA guidelines.
0 Calibration equipment for field sampling equip-
ment .
0 Analytical services for gravimetric analyses.
0 Analytical services for lead analyses.
0 Analytical services for particle size distribution
employing microscopic techniques.
5-10
-------
0 Quality Control Audit functions for both field and
laboratory operations.
The EPA tentative reference method for lead analysis is pre-
sented in Reference 3. This method can be employed with
minor alterations, as the sample type may dictate, for the
determinants of lead content of samples collected from
fugitive lead emission studies.
5.2 SAMPLING APPROACHES
As previously stated, several approaches are available
for sampling particulate emissions from fugitive dust
sources. The selection of which approach to use will depend
on the physical configuration of the emission source. The
sampling approaches which are presented in this section
cover the majority of the types of emission sources most
commonly encountered.
5.2.1 Power Vents
Power ventilators are wall or roof mounted. The sampl-
ing techniques employed for this type of source are identi-
cal. In order to sample this type of source, the following
data must be obtained:
0 Area of power vent
0 Average air velocity through the power vent
0 Barometric pressure and temperature at the time of
sampling
5-11
-------
0 Sampling rate of the collection device
0 Total participate loading and lead content during
sampling
0 Process production rate for the sampling time
period
5.2.1.1 Example Test for Power Vents - A typical example
for sampling power vents is presented below.
As a result of the presurvey, a power ventilator was
identifyed as a fugitive emission source. The vent is
located approximately three stories above ground level
within the plant building and is wall mounted. The total
height of the building is approximately six stories. Scaf-
folding was erected into place to serve as a working plat-
form. The diameter of the vent was measured to be 1.8 m (6
ft). A circular sheet metal flange, 0.3 mx 1.8 m (1 ft x 6
ft) was fabricated and installed around the vent to elimi-
nate velocity contour effects on the velocity measurements.
The velocity of the fan was then measured employing a
rotating vane anemometer. The following procedure is recom-
mended for this measurement, it is based on measuring the
velocity at several points which properly represent equal
areas based on aerodynamic fan characteristics.
0 Procedure for Fan Velocity Measurements
Since fan characteristics are such that the maximum
velocity is observed at the tips of the blades, and as a
5-12
-------
result velocity contour effects are present in the areas
around the vent opening. To eliminate these effects, which
would result in negative biased velocity error, a circular
sheet metal flange should be installed in front of the power
vent. Measurements of the fan velocity are then made
several equal areas which are representative of the fan
characteristics. At each point a one minute rotating vane
velometer measurement is made. The location of the points
at which the measurements should be made are presented in
Figure 5-1.
The 13 one minute velocity measurements are averaged to
obtain the average fan velocity.
The volumetric air flow of the fan was then determined
by multiplying the fan velocity by the area of the flange
around the power vent. This volumetric air flow rate was
then corrected to standard conditions of 760 mm Hg and 25°C.
A calibrated high volume air sampler was then located ap-
proximately 1.8 m (6 ft) from the power vent. Sampling of
the fugitive emissions entering the power vent was conducted
for three one-hour periods during one eight-hour work shift.
The collection media employed was a 20 cm x 25 cm (8 in. x
10 in.) glass fiber filter. The total fugitive emissions
and the lead content was determined employing gravimetric
5-13
-------
03
••-•
CD
co
cu
0)
f-i
Q)
QJ
o
03
CO
£
Q)
CO
03
0)
S
I
i
•3
>
<*-4
O
CO
fi
O
CJ
CSJ
03
g
O
2
0)
>
o •
Q-i— i
i— CM CO CM ^3"
Q O CJ Q O
» A •« CO «*»••»
03
CJ
O
1-1
Q OO
o in co o r^ LO o
i— CM oo in vo r^ en
I
CD
5-14
-------
and atomic absorption spectroscopy. The concentration of
the total suspended participates and the lead content were
expressed in yg/m at standard conditions, 760 mm Hg and
25°C. During the three test periods the process production
rate data was obtained from the plant process engineer.
From the above data the emission rates for each test
period was calculated by multiplying the concentrations from
the sampler by the average volumetric flow rate of the power
vent. With this data and the process production rate, an
emission factor was calculated.
5.2.2 Natural Draft Vents
Natural draft ventilators are most commonly installed
on the roof of large buildings which house many batch type
process operations. These vents are often referred to as
the cupola on the roof of the building. This term should
not be confused with the cupola furnace. The same data must
be obtained as described for power ventilators, the major
difference is that the air velocity through the cupola is
very low and the opening is normally quite large. It is not
uncommon for a cupola to be 90 m (300 ft) long by 8 m (25
o 2
ft) wide which results in an area of 720 m (7500 ft ).
Emissions rise as a result of the operations below which
produce hot gases. The major sampling problems associated
5-15
-------
with sampling this type of vent are centered around the low
gaseous flow rate. The determination of the velocity can
become a problem since it is low, normally less than 60
m/min (200 ft/min), and can vary with meteorological con-
ditions while sampling. In addition to the problem of
measuring a low velocity, it is not homogenous throughout
the area of the ventilator. Hot wire anemometers and hot-
film wedge sensors are the instrument of choice since they
have good accuracy and precision and provide the capability
(4)
for remote read out of the velocity.
5.2.2.1 Example Test for Natural Draft Vents - A typical
example for sampling natural draft ventilators is presented
below.
As a result of the presurvey a "cupola" was identified
as a source of fugitive emissions. The cupola was above a
lead casting operation where twelve casting pots and molds
were operating on a batch basis. Cupola emissions varied in
intensity as well as physical location as a result of the
batch type operation. The inside dimensions of the cupola
were 30 m by 8 m or 240 m2 (100 ft by 25 ft or 2500 ft2).
As a result of the above information, sampling was performed
in the following manner:
0 Seven constant speed traversing lines were fabri-
cated, each system was designed to traverse at a
rate of 4.2 cm/min (1.66 in./min) to provide a
three hour sampling period over the 8m (25 ft)
vent width.
5-16
-------
The traversing systems were installed at the base
of the cupola. Figure 5-2 illustrates this con-
figuration.
To each traversing system an inverted high volume
air sampler and a hot film wedge sensor was in-
stalled.
The traversing systems were spaced across the
cupola such that two systems were 1.5 m (5 ft)
from each end and the other five were 4.6m (15
ft) apart, see Figure 5-3.
The electrical outputs from each hot film wedge
were connected to a multichannel read out system
where the velocity was recorded.
Each of the high volume samplers were calculated
at standard conditions, 760 mm Hg and 25°C, and
glass fiber filter media was used as the collector
substrate.
The seven systems were activated simultaneously
and sampled for three hours, which comprised one
test.
The process production rate was obtained from the
plant production engineer for the period during
the test. In addition to the production rate, the
number of batch pours were also obtained.
The average velocity for each sensor, each minute
was determined and from these measurements and the
area of the vent on average volumetric flow rate
was determined.
The total suspended particulate and lead content
was determined from each sample and expressed in
terms of lag/nP. From the seven samples an average
concentrate for both total suspended particles and
lead was determined.
The total emissions for TSP and lead were then
calculated using the average volumetric flow rate
(m^/min) and the average concentrations
5-17
-------
g
1
CO
(D
>
(fl
H
-P
^
O
p
•
(N
I
in
(D
5-18
-------
Q)
-P
W
>1
CO
d
•H
W
H
Q)
(0
(0
H
tr>
(0
•H
rH
(0
•H
O
(0
cu
i
m
o>
-H
fc
5-19
-------
From the emission data and the process data, a fugitive
lead emission factor was calculated for the production rate
and for each batch casting.
5.2.3 Material Handling Operations
Emissions of fugitive particulate material are often a
result of open conveyor belt, dumping and storage operations
of materials and products. The magnitude of the emission
depends on the prevailing meteorological conditions and the
type of material handled. One approach for the determina-
tion of emission factors from these sources requires an
extended sampling program employing directional operated
high volume sampling and meteorological equipment.
5.2.3.1 Example Test for Sampling Materials Handling
Operations - The fugitive particulate emission source was
identified as a dross transfer operation. This material was
being transferred back to an electric arc furnace for addi-
tional recovery of product. The sampling of this source
consisted of the placement of four directional operated high
volume air samplers and a meteorological system in the near
proximity of the material transfer operation. Three of the
samplers were placed 3 m (10 ft) apart at a location which
was determined to be in the predominate down wind director
approximately 3 m (10 ft) from the operation. The other
sampler was placed approximately 8 m (25 ft) up wind of the
5-20
-------
operation to serve as a background site. All directional
samplers were set to operate over a 180° sector such that
they would sample when the wind direction was perpendicular
to the transfer belt. Each sampling period consisted of 24
hours, at the end of each period the glass fiber filters
were changed and the average wind speed for the period that
the samplers were operational was calculated. This sampling
program was continued until three sets of samples were
collected for each of the following average wind speed
conditions; less than 2.2 m/sec (5 mph), 2.2 to 4.5 m/sec (5
to 10 mph), and greater than 4.5 m/sec (10 mph). The total
suspended particulate and lead concentrations, less the
background levels, were determined for each set of wind
speed conditions. From the emission data, and an estimate
of the emission area, an emission rate was determined for
each of the three average wind speed conditions. These
emission rates were then employed with the average process
rate to obtain emission factors for the operation under
varying wind speed conditions.
5.2.4 Quasi-Stack Method(1'2)
This method requires that the source of emissions be
isolable and that an enclosure can be installed capable of
capturing emissions without interference with plant opera-
tions. These types of sources are normally single opera-
5-21
-------
tions with emissions which are not confined within a build-
ing and as a result other techniques as described above
cannot be applied. The techniques consist of constructing a
temporary hood or enclosure over the source. The enclosure
is then vented to an exhaust duct or stack of regular cross-
sectional area. Emissions are then measured in the exhaust
duct using standard stack sampling or similar well recog-
nized methods. This approach is necessarily restricted to
those sources of emissions that are isolable and physically
arranged so as to permit the installation of a temporary
hood or enclosure that will not interfere with plant opera-
tions or alter the character of the process of the emis-
sions .
5.2.4.1 Example Test for Quasi-Stack Sampling - As a result
of a presurvey a major fugitive particulate emission source
was identified at a ferroalloy production process. The
source was a result of emissions from an electric arc fur-
nace. A large hood with a measurement duct and blower was
constructed over the source. The volumetric air flow within
the duct was adjusted such that no visible particulate
emissions were observed escaping the hood. At this point
the particulate emissions were sampled employing the EPA
source sampling method 5. Three sampling periods of 60
5-22
-------
minutes each were conducted during a normal work shift. The
total suspended particulate and lead emissions were deter-
mined from these samples. From this data and the production
rate data an emission factor was determined.
5.3 MANPOWER ESTIMATES
Table 5-2 presents estimates of manpower requirements
for each of the example test programs as presented in the
above sections. The total manhour requirements for con-
ducting four test programs as outlined above is estimated at
2770 hours. This level of effort does not include the
fabrication of sheet metal ducts, hoods, etc. which would
require the services of a subcontractor. The cost of this
element is estimated at $5000 to $7000. Therefore, assuming
no extraordinary services/resources are required, the cost
of a series of power vent, natural draft vent, materials
handling, and quasi-stack tests can be estimated as follows:
5-23
------- An error occurred while trying to OCR this image.
-------
Item Estimated cost*
0 Burdened labor, 2700 hours @ $22/hr $ 59,400
0 Transportation and Per Diem
(1) Pre-survey, 2 men for 2 days @ $150/ 600
day (including transportation)
(2) Field test, 6 men for 20 days @ $70/ 8,400
day (including transportation)
0 Sampling Site Preparation
(1) Four sites @ $100 ea 400
(2) Fabrication of sheet metal ducts, 6,000
hoods, etc.
0 Expendable supplies; estimated @ $40/ 800
day for 20 days
Total Estimated Cost $ 75,600
* This estimate is based on the measurement and determination
of emission factors from the four emission sources (power
vent, natural draft vent, materials handling, and quasi-
stack) at one plant.
5-25
-------
REFERENCES FOR SECTION 5.0
1. Technical Manual for the Measurement of Fugitive
Emissions: Roof Monitor Sampling Method for Industrial
Fugitive Emissions. U.S. EPA - 600/2-76-089b. May
1976.
2. Technical Manual for the Measurement of Fugitive
Emissions: Quasi-Stack Sampling Method for Industrial
Fugitive Emissions. U.S. EPA - 600/2-76-089c. May
1976.
3. Tentative Reference Method for the Determination of
Lead in Suspended Particulate Matter Collected from
Ambient Air. U.S. EPA, Environmental Monitoring and
Support Laboratory. July 1977.
4. Vollaro, R.F. A Survey of Commercially Available
Instrumentation for the Measurement of Low-Range Gas
Velocities. U.S. EPA, Emission Measurement Branch,
January 3, 1977.
5. Federal Register, Vol. 36, No. 247, December 23, 1971.
5-26
-------
6.0 STATE OF THE ART FOR DETERMINATION OF A
PLANT EMISSION FACTOR FROM AMBIENT SAMPLING
As noted in Section 5.0, the three industries under
consideration, namely secondary lead smelting, ferroalloy
production and grey iron foundries, have similar fugitive
emission sources. For example, the major sources of fugi-
tive lead emissions from a secondary lead smelter (typically
producing 45 Mg/day or 50 ton/day) all result from blast
furnace operations - charging, slag tapping, and lead tap-
ping/casting (- 33 g/sec or 4.3 Ibs/hr). For a similarly-
sized ferroalloy facility, again the principle sources of
fugitive lead particulates are furnace charging and tapping
operations (~ 76 g/sec or 10.0 Ibs/hr). And, for a gray
iron foundry with a cupola and electric arc furnace each
producing 9 Mg/hr (10 ton/hr) with an induction unit oper-
ating at 1.8 Mg/hr (2 ton/hr), the combined charging and
tapping operations will produce the bulk of the fugitive
lead particulates (- 38 g/sec or 5.0 Ibs/hr).
Due to their inherent commonalities relative to fugi-
tive lead sources, the elements of a generalized sampling
approach which is applicable to each industry has been
6-1
-------
developed. The plant emission factor then may be developed
from the sampling data.
6.1 GENERAL APPROACH - MONITORING AND ANALYSIS
In the case of the three industrial source-types under
consideration, it appears that the best currently-available
technique for measuring their ambient impact is by judi-
ciously measuring the ambient particulate concentrations
upwind and downwind of "typical" facilities with high volume
samplers. The contribution of emissions from nonfugitive
sources can be estimated by dispersion modeling. This
contribution can then be subtracted from the estimated
impact for the entire plant. The remainder is then inferred
to be the impact of the fugitives.
This technique is not without sources of error. Pre-
cautions should be taken in the interpretation of all mea-
sured and inferred concentrations. However, the approach is
nevertheless a reasonable one for estimating the magnitude
of a suspected fugitive emission problem and for identifying
possible corrective actions.
The short-term, "worst case" impact of fugitive emis-
sions at points near a plant is generally of chief concern.
Accordingly, the downwind sampler(s) should be located so as
to be sensitive to this impact. The number of samplers to
6-2
-------
be used to determine downwind concentrations should reflect
the number, strength, and size of suspected fugitive lead
sources at the facility. The field sampling program should
be conducted over a sufficiently long period to assure that
conditions reasonably approximating the worst case are
observed.
The upwind sampler(s) should be located so as to be
representative of the same air mass as that being sampled by
the downwind sampler(s). It is important to ensure that no
extraneous sources are affecting the sampler(s). Generally,
the downwind sampler(s) should be located as near as pos-
sible to the emission sources in question. Because many of
these sources are near ground level, their air quality
impact is highest near the source and decreases rather
quickly with downwind distance. However, for some elevated
sources, such as roof monitors, the maximum air quality
impact may be at an appreciable distance from the plant
boundary. The selection of locations for samplers must be
made with this in mind.
In order to document the relationship between measured
air quality and suspected fugitive lead particulate sources,
simultaneous meteorological measurements must also be made.
Meteorological parameters to be measured include, at a
minimum, wind speed and wind direction. The availability of
6-3
-------
on-site meteorological data is also important in situations
where dispersion modeling is applied to determine the impact
of nonfugitive stack emissions. For this purpose, hourly
observations of temperature and cloud cover are also needed.
Cloud cover may be best obtained from a nearby National
Weather Service station.
Prior to conducting field sampling, three important
functions must be completed:
0 Field pre-survey and report.
0 Sampling system design.
0 Test plan and field protocol development.
These are discussed in the following sections.
6.1.1 Pretest Survey
In order to design an effective test plan, a pretest
survey of the facility and its surroundings is required. A
good general knowledge is required of the plant layout,
process chemistry and flow, surrounding environment, and
prevailing meteorological conditions. Specific operational
characteristics relative to the products involved, the space
and manpower skills available, emission control equipment
installed, and the safety and health procedures observed,
will also influence the sampling plan. Work flow patterns
and production schedules that may result in periodic changes
in the nature or quantity of emissions or that indicate
6-4
-------
periods for the most effective and least disruptive sampling
must also be considered. Most of this information can only
be obtained by a site survey and follow-on discussions with
plant management. Table 6-1 notes some of the most useful
data to be obtained. Additional information will be sug-
gested by considerations of the particular on-site situa-
tion.
6.1.2 Sampling System Design
The number and location of the samples is extremely
important to the successful conduct of an upwind-downwind
sampling program. The design of the sampling system is
influenced by factors such as source complexity and size,
available sampling locations, topography, and prevailing
meteorological conditions which govern the distribution of
the pollutant plume in the ambient atmosphere. Most situa-
tions will in general fit into some combination of the
following parameters:
0 Source - Sources may be either homogeneous, emit-
ting a single type of mixture of pollutants from
each and every emission location, or heteroge-
neous, emitting different types or mixtures of
pollutants from different locations. The physical
size of a source will determine the extent of the
pollutant plume and may influence its homogeneity,
the proximity of different emissions to each other
largely influencing the degree of mixing in the
plume for a given downwind distance.
6-5
-------
Table 6-1. PRE-TEST SURVEY INFORMATION TO BE OBTAINED FOR
APPLICATION OF FUGITIVE EMISSION SAMPLING METHODS
Plant
Layout
Drawings:
Building Layout and Plan View of Potential Study Areas
Building Side Elevations to Identify Obstructions and
Structure Available to Support Test Setup
Work Flow Diagrams
Locations of Suitable Sampling Sites
Physical Layout Measurements to Supplement Drawings
Work Space Required at Potential Sampling Sites
Process
Process Flow Diagram with Fugitive Emission Points
Identified
General Description of Process Chemistry
General Description of Process Operations Including
Initial Estimate of Fugitive Emissions
Drawings of Equipment or Segments of Processes Where
Fugitive Emissions are to be Measured
Photographs (if permitted) of Process Area Where
Fugitive Emissions are to be Measured
Names, Extensions, Locations of Process Foremen and
Supervisors Where Tests are to be Conducted
Operations
Location of Available Services (Power Outlets, Main-
tenance and Plant Engineering Personnel, Labora-
tories, etc.)
Local Vendors Who Can Fabricate and Supply Test System
Components
Shift Schedules
Location of Operations Records (combine with process
operation information)
Health and Safety Considerations
Other
Access routes to the areas Where Test Equipment/Instru-
mentation Will Be Located
Names, Extensions, Locations of Plant Security and
Safety Supervisors
6-6
-------
0 Site - Sites should be located on level terrain
with free access of ambient air from all sides,
not obstructed by hills or buildings that inter-
fere with or influence the ambient air flow either
up- or downwind, or located in a valley between
hills or large buildings that influence the air
flow both up- and downwind. Each type of topog-
raphy will influence the extent and homogeneity of
the pollutant plume depending on the direction of
the wind flow relative to the obstructions.
0 Wind Direction - The direction of the prevailing
wind determines the basic location of upwind and
downwind samplers. It will influence the pol-
lutant plume in every instance except that of a
homogeneous cloud at an open level site. In other
instances, the wind may be directed generally
across or parallel to obstructing hills or valleys
which may result in channeling, lofting, or swirl-
ing of the air flow across the site that will
distort the pollutant plume.
° Localized - The homogeneity of the ambient air
approaching the measurement site may affect the
composition and distribution of different pol-
lutants within the pollutant plume. Contributions
from sources upwind of the site may result in
variations in the pollutant concentrations in the
ambient air passing over the site and thus in the
pollutant plume as well.
0 Wind Speed - Wind speed, which can affect the
plume's size and distribution, need not be con-
sidered as a governing design factor since it is
controllable by scheduling to avoid sampling dur-
ing periods of either excessive wind velocity or
calm conditions. Wind speeds within normal limits
are taken into consideration in data reduction
calculations.
The location of the samplers, especially those down-
wind, is critical in order to ensure that samples are taken
at points known to be within the pollutant plume at measura-
ble concentrations. An estimate of acceptable downwind
6-7
-------
(2)
sampler locations may be made utilizing the equation for
the diffusion of gases and particulates in the atmosphere
along the plume centerline from a ground level source: x =
Q/TrKu, where
X = pollutant concentrations representative of sampling
times up to one hour at receptor point, gm/m3
Q = source emission rate, gm/sec
K = product of standard deviations of vertical and
horizontal pollutant distribution, m2
u = wind speed, m/sec
This equation assumes a Gaussian distribution of pollutants
in both the vertical and horizontal directions and no depo-
sition or reaction of pollutants at the earth's surface. By
rearranging terms, the product of the standard deviations
(K), which are functions of the downwind distance (x) of
the receptor from the source, may be determined as a func-
tion of easily estimated or measured parameters in:
K = Q/TTXU, where
Q is estimated from published emission factors,
X is set equal to a selected hourly value related to
the sampling method detection limit, and
u is measured at the site.
The maximum downwind sampler distance from the source along
the axis of the wind direction (x) may then be determined
from the curves of Figure 6-1, which relate K and x for
various atmospheric stability categories. For distances
6-8
-------
106
5x 104
2x104
5x 103
§2x103
? 103
» 5x102
2x102
102
50
20
10
1 f
Atmospheric
stability
category
100 200
400
600
800
1000
Maximum downwind sampler distance from a ground level
source along wind direction axis (x) - maters
Figure 6-1. Maximum downwind sampler distances
6-9
-------
less than 100 meters the curves have been extrapolated from
(2)
original data and should be used with caution espcially
when comparing measured and estimated concentrations. The
stability categories are listed and explained in Table
(1)
6-2.
Table 6-2. ATMOSPHERIC STABILITY CATEGORIES
(1)
Wind speed
m/sec
< 2
2-3
3-5
5-6
> 6
Day*
Solar altitude"1"
> 60°
A
A-B
B
C
C
35°-60°
A-B
B
B-C
C-D
D
15°-35°
B
C
C
D
D
Night
Thin Overcast or
> 50% Clouds
-
E
D
D
D
< 50% Clouds
-
F
E
D
D
* Day is one hour after sunrise to one hour before sunset.
t Solar altitude may be determined from Table 170, Solar
Altitude and Azimuth, Smithsonian Meteorological Tables.
Use neutral class D for overcast conditions at any time.
Partial cloud cover (60 percent to 85 percent)
will reduce effective solar altitude one division (e.g.,
from > 60° to 35°-60°) for middle clouds and two divi-
sions (e.g., from > 60° to 15°-35°) for low clouds.
When suitable x-distances have been selected, which
may be any distance less than the maximum determined from
Figure 6-1, cross-wind distances (y) perpendicular to the x-
axis must be determined that will ensure that samples are
taken within the limits of the plume. Maximum cross-wind
distances are plotted as a function of x in the curves of
Figure 6-2 for the same atmospheric stability categories
used in determining x.
6-10
-------
300
200
fj 100
0)
u
10
Q.
s
•a
o
o
E
3
'i
50
10 —
Atmospheric
stability
category
100 200 400 600
Downwind sampler distance (x) - meters
800
1000
Figure 6-2. Maximum crosswind sampler distances.^1'
6-11
-------
To illustrate the application of the equations and
curves presented in this section, assume a source emitting
particulates into a four m/sec wind, u (expected at site),
at an estimated emission rate, Q, 30 g/sec (from emission
factors and plant production rate) and a sampler with a
lower detection limit, M, of 0.001 g and flow rate, F, 1.4
m /min (50 cfm). For an expected sampling time, T, of 60
minutes, the required pollutant concentration, x/ at the
sampler is x = M/FT, where
M = 0.001 g,
F = 1.4 m /min x 1.5 adjustment factor (to compensate
for likely inaccuracies in estimates of concentra
tion) = 2.1 m^/min, and
T = 60 minutes,
thus x = 8 x 10~6 g/m3
The product of the pollutant plume's standard deviations, K,
is found in the equation K = Q/TTXU, where
Q = 30 g/sec
X = 8 x 10 g/m , and
u = 4 m/sec,
thus K = 3 x 105 m2
As you can see from this K and Figure 6-1, our maxi-
mum downwind distances for sampler locations for any atmos-
pheric stability category will exceed 1000 meters.
Now as an additional example, to measure the plant
emissions under a worse case condition let us assume this
6-12
-------
occurs during very early morning with clear skies and a 2
m/sec wind speed resulting in a stability category F. We
are able to locate a sampler 600 meters downwind from the
plant and obtain a one hour sample. Our resulting (measured)
-3 3
X, after we have adjusted for background, is 4 x 10 g/m .
2
From Figure 6-1, K is 200 m . Therefore, our empirical
plant emission rate is 5 grams per second.
Upwind samplers should ideally be located on the wind
direction axis just far enough upwind to prevent sampling
the backwash of the pollutant plume. A minimum upwind
distance of roughly twice the height of the building
housing the source will usually be sufficient.
6.1.3 Field Test Plan
The field test plan provides a formal specification of
the equipment and procedures required to meet the objectives
of the measurement program. It is based on the information
collected in the pretest survey and defines the required
sampling equipment, procedures, and schedules consistent
with the program objectives in light of the constraints
imposed by the specific source being evaluated. A sug-
gested outline for the field test plan is abstracted as
follows:
Background - The introductory paragraph containing
the pertinent information leading to the need to
conduct the measurement program and a short de-
scription of the information required to answer
that need.
6-13
-------
Objective - A concise statement of the problem
addressed by the test program and a brief descrip-
tion of the program's planned method for its
solution.
Approach - A description of the measurement scheme
and data reduction methodology employed in the
program with a discussion of how each will answer
the needs identified in the background statement.
Instrumentation/Equipment/Facilities - A descrip-
tion of the instrumentation arrays to be used to
collect the samples and meteorological data iden-
tified in the approach description. The number
and frequency of samples to be taken and the
sampling array resolution should be described.
A detailed description of the equipment to be
employed and its purpose.
A description of the facilities required to oper-
ate the measurement program, including work space,
electrical power, support from plant personnel,
special construction, etc.
Schedule - A detailed chronology of a typical set
of measurements, or a test, and the overall sched-
ule of events from the planning stage through the
completion of the test program report.
Limitations - A definition of the conditions under
which the measurement project is to be conducted.
If, for example, successful tests can be conducted
only during occurrences of certain source opera-
tions, those favorable limits should be stated.
Analysis Method - A description of the methods
which will be used to analyze the samples col-
lected and the resultant data, e.g., statistical
or case analysis, and critical aspects of that
method.
Report Requirements - A draft outline of the
report on the analysis of the data to be collected
along with definitions indicating the purpose of
the report and the audience it is to be directed
to.
6-14
-------
0 Quality Assurance - The test plan should address
itself to the development of a quality assurance
program. This QA program should be an integral
part of the measurement program and be incor-
porated as a portion of the test plan either
directly or by reference.
0 Responsibilities - A list of persons who are
responsible for each phase of the measurement
program, as defined in the schedule, both for the
testing organization and for the plant site.
6.2 SPECIFIC APPROACH FOR FUGITIVE LEAD SAMPLING
As previously noted, fugitive lead emissions from all
three industries under consideration eminate from furnace
charging, tapping, and affiliate process operations as well
as reentrainment of settled dust by wind and vehicles.
Therefore, recognizing that for a specific industrial
facility there may well be a series of constraining factors
(operational, topographical, physical, etc.) which will
influence the actual sampling program required, an approach
has been developed which presents a reasonable plan for a
typical source evaluation. The approach presented is some-
what idealized in that only minor constraints have been
placed on the resources and time allocated. In actual
application, practical considerations may well dictate or
permit reductions in both resources (personnel and equip-
ment) and sampling period.
Based on the plant configuration and the frequency
distribution of wind directions in the area, a single high-
6-15
-------
volume sampler should be positioned upwind of the plant
fence line and four samplers along the downwind fence line.
For illustrative purposes, Figure 6-3 depicts the proximate
location of these samplers, as well as the physical plant
layout, location of paved roads, and local topographic
features as they would be alligned for upwind/downwind
sampling at a typical cement plant assuming prevailing wind
(4)
direction is westerly.
Sampling periods should be selected so as to be repre-
sentative of both routine and maximum operating conditions
if at all possible. Plant management should agree to keep a
record of in-plant operating activity during the periods
selected for sampling. By correlating the measured particu-
late lead levels with source-related activities, the agency
will be able to determine if it is possible to develop
meaningful emission or air quality impact factors by the
upwind/downwind technique.
6.2.1 Sampling Protocol
The upwind/downwind monitor configuration should con-
sist of a monitoring site at the upwind plant fence line,
including one high-volume air sampler and a recording
meteorological system. On the downwind side of the source,
near the property line, four high-volume air samplers will
be located parallel to the fence line at intervals of ap-
6-16
-------
-p
c
-p
-------
proximately 60 meters (200 ft). The upwind site will employ
a small generator to power the single sampler and the
meteorological station; the other sites will require two
larger motor generators for electrical power. One generator
is necessary for each pair of downwind samplers. The gen-
erator is placed between the pair of samplers with 30 meter
(100 ft) power lines to each sampler. Three field tech-
nicians will perform the tests, one at each generator loca-
tion. Whenever the wind direction is verified to be from
the upwind direction (+_ about 20 degrees), the technician at
the meteorological control site will contact the other tech-
nicians via radio and initiate sampling. Each sampling
period will be as long as practicable, during which the wind
direction will be monitored to determine whether it persists
from the desired directional sector. A decrease in flow
rates of more than 10 percent on the downwind samplers
indicates that sufficient sample has been collected, and the
test period will be terminated. The above program will be
repeated several times (approximately three to six) with the
objective of observing optimum conditions of speed and
directional persistence of the wind. Also, it is imperative
that various source operational conditions be observed.
6.2.2 Monitoring Instrumentation
The two basic measurements required for this task are
the measurement of wind speed/direction and the total sus-
pended particulates in the ambient air.
6-18
-------
Meteorological Monitoring - Wind speed and direction
should be monitored using instrumentation meeting the
specifications in EPA's Ambient Monitoring Guidelines for
Prevention of Significant Deterioration (PSD), as follows: '
0 Systems should exhibit a starting threshold of
less than 0.5 m/sec wind speed (at 10 degrees de-
flection for direction vanes). Wind speed systems
should be accurate above the starting threshold to
within 0.25 m/sec at speeds equal to or less than
5 m/sec. At higher speeds, the error should not
exceed 5 percent of the observed speed (maximum
error not to exceed 2.5 m/sec).
0 The damping ratio of the wind vane should be
between 0.4 and 0.65.
0 Wind direction system errors should not exceed 3
degrees from true 10-minute or greater averages,
including sensor orientation errors.
0 The standard exposure of wind instruments is 10
meters above level, open ground. Where this is
not possible, the anemometer should be installed
such that its indications are reasonably unaf-
fected by local obstructions.
Total Suspended Particulate Monitoring - Total sus-
pended particulates will be collected with a high-volume air
sampler, as described in the Code of Federal Regulation 40,
Part 50.11, Appendix B, July 1, 1975, Pages 12 through 16.
Specifically, five samplers are required, which will be
equipped with the following ancillary equipment:
0 Dixon flow recorders
0 Running time meters
0 Quick-change filter cartridges
6-19
-------
Monitoring Support Equipment
Support equipment to be used is listed below:
0 High-volume air sampler calibration kit
0 Three motor generators:
1) Two 3-KVA, 5-hp gasoline
2) One 1.5-KVA, 3-hp gasoline
0 Sampling van with covered bed
0 Power extension cords; five 30 m (100 ft) cords of
No. 12 wire with ground and exterior duplex re-
ceptacles, two 15 m (50 ft) cords of No. 14 wire
with ground, and exterior extension cords
0 Preweighed fiberglass filter media consisting of
100 filters, folders, envelopes, and data sheets
0 Three portable citizen-band radios
6.2.3 Analytical Requirements
Various analytical requirements exist with respect to
the determination of fugitive lead emissions on ambient air
quality. They can be summarized as follows:
0 Filter media for use with the various sample
collection systems. This media must be preweighed
following the published U.S. EPA QA guidelines.
0 Calibration equipment for field sampling equip-
ment.
0 Analytical services for gravimetric analyses.
0 Analytical services for lead analyses.
0 Quality control audit functions for both field and
laboratory operations.
The EPA tentative reference method for lead analysis was
discussed in Section 5 and is presented in Reference 6.
6-20
-------
This method can be employed, with minor alterations as the
sample type may dictate, for the determination of lead
content of samples collected.
6.2.4 Scheduling
The test program schedule is presented in Figure 6-4.
Ten specific tasks are defined and programmed over an antic-
ipated period of performance of 25 days. This schedule
allows ten working days for field monitoring. This period
may turn out to be much shorter if the desired meteoro-
logical conditions are observed relatively early in the
program. However, the desired conditions may not be ob-
served at all and therefore the time period may have to be
extended. Additionally, Figure 6-5 presents a test activity
schedule to be used by the supervisory field technician in
conducting each field test in the series. Eight specific
activities are defined over a projected 9-hour test period.
6.2.5 Laboratory Procedures and Data Reporting
The suspended particulate loading will be determined
according to the method described in the Code of Federal
Regulations 40, part 50.11, Appendix B, July 1, 1975, pages
12 through 16. The lead fraction will be determined by the
methodology referenced in 6.2.3. The particulate material
on the exposed filter will be equilibrated under the same
temperature and humidity conditions as experienced in weigh-
6-21
------- An error occurred while trying to OCR this image.
-------
Activity
Set up of downwind monitoring
equipment
Calibration check of downwind
samplers
Set up of upwind monitoring
equipment
Calibration check of upwind
monitoring equipment
Sampling
Removal of samples and
meteorological strip chart
Removal of sampling equipment
Preventive maintenance of
equipment
Time (hours)
1
-
mm
2
mm
-
3
-
4
5
6
7
-
8
—
-
9
limn
Figure 6-5. Field test schedule.
6-23
-------
ing the unexposed filters. The weight of the particulate
material collected on exposed filters will be determined
gravimetrically. The calculated TSP and lead concentrations
are based on the net weight of collected material and the
sample air volume corrected to standard conditions (760 mm
Hg and 25°C). These data will be reported on a separate
data sheet for each test in the series. This data sheet
will include the following information:
0 Date/time of sampling
0 Test series identification
0 Identification of the specific location of each
sampler
0 TSP concentration expressed in yg/m
0 Lead concentration expressed in yg/m
In addition, the meteorological data for the specific study
period should be reduced and reported in SAROAD format.
Quality control procedures to be followed throughout
this test series will be consistent with those defined in
"Quality Control Practices in Processing Air Pollution
Samples," U.S. Environmental Protection Agency Publication
No. APTD-1132. After all data have been reduced to SAROAD
format, an independent audit will be performed on 7 percent
of all values reported.
6.3 MANPOWER ESTIMATES
Table 6-3 presents an estimate of manpower requirements
for a typical 10-day test program as described in the above
6-24
-------
EH
U
W
CO
w
ffi
EH
H
CO
EH
W
s
H
P
O
w
w
o
Q
w
EH
H
EH
CO
W
ro
I
n3
EH
CO
ID
O
a
CO
w
H
w
EH
u
w
o
CO
CO
o
EH
en
O
X
1
C
ffl
a
•
rH
0)
C
C
O
tn
i_i
0)
cw
^
0
H
X
U
Q)
EH
,
CO
4-)
tn
-rH
e
•& O CO
rH CN
4-1
c in
0) 4J 4J
g 3 M
a o o
•H ^ Du
3 U 0)
Cr1 <1) in in
c o) m x 03 4J
a) -HO Q) 4-> M
rH cn 'O tn 6 m o
•H EC ftf ' — 1 ^4 V( Q)
4-J M c o
O [ ^ Q) p>) O fd M [ 1 rH
•n -H 4J D M fii 03 Hj
O C rH CU T) 5H d
SH O -H 4J C 4-> 4-> O 'd 4-> 4-)
222rmeuijta3£
(T3 (d (t5 -rH Tj -H (fl fO
OiCliai4->rH (OrH Old,
-------
sections. A total of approximately 528 man-hours/plant
evaluation is projected. This level of efforr does not
include any specialized site preparation, such as grading,
which would require the services of a subcontractor. There-
fore, assuming no extraordinary services/resources are
required, the cost per upwind/downwind test series can be
estimated as follows:
Item Estimated Cost
0 Burdened Labor; 528 hours @ $ 11,600
$22/hr
0 Transportation and Per Diem
(1) Pre-survey, 2 men for 2 days @ 600
$150/day (including transporta-
tion)
(2) Field test, 3 men for 10 days @ 2,100
$70/day (including transporta-
tion)
° Sampling Site Preparation; 5 sites @ 500
$100 ea.
0 Expendable Supplies; estimated @ 400
$40/day for 10 days
Total Estimated Cost $ 15,200
6.4 CALCULATION OF PLANT EMISSION FACTOR
The upwind/downwind sampling program provides an in-
dication of the total plant impact - conventional point
sources as well as fugitive emission sources. In order to
determine the impact of only the fugitive emission sources,
6-26
-------
it is necessary to subtract the impact of the conventional
point sources. Subtracting the upwind concentrations and
predicted pollutant concentrations contributed by con-
ventional point sources at the downwind measuring site
from the actual downwind measured values yields the fugi-
tive impact.
For monitoring to be reliable, the data collection
program must be comprehensive in scope and subject to
strict quality control. Also, the interpretation of moni-
toring results is not always straightforward. Neighboring
sources and/or high background concentrations often present
complications. This is especially true for urban or indus-
trialized areas. High background concentrations or severe
influence from other sources can make it very difficult, if
not impossible, to isolate and separate the actual plant
contribution to measured air quality. Even when it is
possible to isolate the impact of the plant of concern, it
is often difficult to relate the total impact to the indi-
vidual contributing sources in the plant complex. It is
even difficult in some cases to adequately distinguish the
impact of fugitive sources from that of the stack emissions.
An estimate of the contribution of the conventional
point sources under the meteorological conditions that
occurred during a given sampling interval may be determined
6-27
-------
by the application of a dispersion model. The major dis-
advantage is the uncertainty associated with model esti-
mates. The major sources of error in dispersion modeling are
as follows:
0 Inadequacies in the simulation of physical phenomena
by models
0 Inadequacies in the input data to models
0 Lack of expertise in applying models and in
interpreting the results.
To the extent practicable, emissions from nonfugitive
sources should be quantified and characterized for the
specific days of interest. For example, it is entirely
likely that a source whose controls normally operate at a
specified collection efficiency may, on a particular day,
actually be emitting at a rate much higher than normal. If
such were the case, the point source modeling results based
upon the normal emission rate would greatly underestimate
the contribution of nonfugitive sources and thereby exag-
gerate the inferred impact of the fugitive sources.
Once the ambient fugitive emission impact has been
determined a dispersion model can again be applied to back-
calculate the estimated fugitive emission source strength or
emission rate. The fugitive emission rate divided by the
plant production rate will result in the plant fugitive
emission factor. The results from several plants are nec-
6-28
-------
essary if an industry-wide plant fugitive emission factor is
to be developed. This factor may also include such param-
eters as plant capacity and lead content of the raw mate-
rials.
6-29
-------
REFERENCES FOR SECTION 6.0
1. "Technical Manual for the Measurement of Fugitive
Emissions: Upwind-Downwind Sampling Method for Indus-
trial Fugitive Emissions." Industrial Environmental
Research Laboratory, U.S. EPA, Research Triangle Park,
North Carolina. April 1976. Publication No. EPA-600/
2-76-089a.
2. Turner, D. Bruce, "Workbook of Atmospheric Dispersion
Estimates," U.S. Department of Health, Education and
Welfare, Public Health Service Publication No. 999-
AP-26, Revised 1969.
3. "Technical Manual for the Measurement of Fugitive
Emissions: Roof Monitoring Sampling Method for Indus-
trial Fugitive Emissions." U.S. EPA. May 1976.
Publication No. EPA 600/2-76-089b.
4. "Technical Guidance for Control of Industrial Process
Fugitive Particulate Emissions." U.S. EPA, OAQPS,
Research Triangle Park, North Carolina. March 1977.
Publication No. EPA-450/3-77-010.
5. "Ambient Monitoring Guidelines for Prevention of
Significant Deterioration (PSD)." U.S. EPA. May 1978
Publication No. EPA-450/2-78-019.
6. "Tentative Reference Method for the Determination
of Lead in Suspended Particulate Matter Collected
from Ambient Air." U.S. EPA, Environmental Moni-
toring and Support Laboratory. July 1977.
6-30
-------
7 .0 CONCLUSIONS AND FACTOR DEVELOPMENT CONSIDERATIONS
In Sections 2.0 through 6.0 of this report, the current
status of information relative to fugitive lead emissions
from the three source categories in question, plus both
inplant emissions and ambient impact quantification tech-
niques, has been summarized.
In this section we present our evaluation of these
factors, resulting in conclusions and formulation of recom-
mendations dealing with the development of fugitive lead
emission factors from data gathered by a field measurement
program.
7.1 CONCLUSIONS
Based on our evaluation of information available in
existing literature, the following conclusions are apparent:
0 There are insufficient data from which to develop
reliable industry-wide fugitive lead emission
factors for all three categorical sources.
0 Evaluation of very limited ambient lead sampling
data, in light of proportional estimated impact
from identified fugitive lead sources, indicates
that all three process sources examined have a
realistic potential for emitting fugitive lead-
bearing materials which will cause local ambient
lead measurements to exceed 1.5 yg/m^ on a regular
basis.
7-1
-------
0 It is not possible to apply fugitive lead emission
data/factors developed for several other similar
process sources to the three categories under
investigation, basically due to significant dif-
ferences in raw materials being processed and
techniques used.
0 It is not feasible to develop an industry wide
fugitive lead emission factor based on plant
production using the upwind-downwind ambient
monitoring technique due to the following reasons:
1. The use of this ambient monitoring technique
coupled with dispersion model estimates of
the nonfugitive impact is open to significant
and variable sources of error.
2. There is a variation in lead content in the
raw materials used in these industries. In
addition, each process within a facility may
undergo variations in throughput.
3. The number and type of individual processes
per plant vary within the industry. Also,
the levels of control at each plant are not
the same.
0 It is possible to develop a short-term, "worst
case" plant wide lead emission factor using the
upwind-downwind technique. This would be espe-
cially applicable for enforcement or compliance
purposes.
0 In-plant sampling of fugitive lead emission points
such as draft and fan vents, roof monitors, mate-
rials handling operations, hood exhausts, and
furnace/cupola charging and tapping, can be accom-
plished using available sample collectors and
state-of-the-art techniques such as "quasi-stack."
While the measurement methods are not universally
accepted or standardized, they are capable of
providing relative and reproducible results appli-
cable to the development of reliable fugitive lead
emission factors.
Finally, as pointed out previously, the three indus-
tries (secondary lead smelting, ferroalloy production and
7-2
-------
gray iron foundries) have similar fugitive emission sources
such as: blast furnace charging/tapping, pot furnace charg-
ing/melting/tapping, materials handling, electric arc fur-
nace charging/smelting/tapping, and cupola charging/tapping.
Most of these operations normally occur inside plant build-
ings with the fugitive emissions escaping to the atmosphere
through vents, monitors, leakage, or due to poor house-
keeping. Open sources, such as materials handling, also
occur. Since the three industries have these common types
of fugitive emission points, their sampling approaches would
be similar.
All of the above leads to a final conclusion that it is
technically feasible (within state-of-the-art limitations)
to conduct an inplant fugitive lead emissions measurement
program leading to the development of reliable lead emission
factors applicable to sources common to the industry opera-
tions evaluated in this report. Further, considering the
real or potential impact these emissions have on atmospheric
lead levels, it would appear there is a need to have these
factors available for future use by agencies as an aid in
determining the possible compliance status of those cate-
gorical sources within their jurisdiction.
-------
7.2 FACTOR DEVELOPMENT CONSIDERATIONS
If U.S. EPA wishes to develop fugitive lead emission
factors for the three industrial categories in question,
several items must be considered. Following, in general
terms, are these major considerations.
7.2.1 Number of Plants and Operations to be Tested
0 Secondary Lead Smelters - Since approximately two-
thirds of the production from secondary lead smelters is
from blast furnaces, testing of this furnace type would
cover the majority of the industry. However, plant-to-plant
variations in the operation of furnaces could not be iden-
tified from literature sources. The effect this variation
may have on fugitive lead emissions is not known but is
something that could be determined as a result of the test
program. Testing of blast/reverberatory furnace operations
will cover the large sources in the industry. Additional
testing of pot furnaces will complete the sources of fugi-
tive lead emissions in this industry. All of these sources
may possibly be found at one smelter, but testing at more
than one plant is suggested so that results will be more
representative of industry-wide emission rates. This will
also minimize bias of the results by a single plant and will
permit an analysis relating emission levels to lead content
of the raw materials.
7-4
-------
0 Ferroalloy Plants - Based on product distribution,
representative results will not be obtained by testing only
one plant which produces just one type of ferroalloy. Tests
must include the major ferroalloy products; ferro-silicon
and ferro-manganese, followed in importance by ferro-chro-
miums and silicomanganese. Tests possibly could be com-
pleted at one facility, but would likely require several
plants to complete a full test program. The test program
should also include analysis of the raw materials (ores) for
lead content. This will provide the necessary data to
determine how the fugitive lead emission factors are related
to type of product, lead content of the ore, or both.
0 Gray Iron Foundries - The melting furnaces are the
major source of fugitive lead emission found in the gray
iron foundry. To obtain representative test results, the
cupola must be given the highest priority for sampling since
it produces 70 percent of all gray iron. Electric arc
furnaces should have the next highest priority in the sam-
pling program since together, they account for over 85
percent of all gray iron production. Since electric induc-
tion and reverberatory furnaces are a minor part of the gray
iron industry, these furnaces could reasonably be omitted
from the sampling program. Additionally, while the quan-
tities of lead emissions expected from iron innoculation,
7-5
-------
pouring into molds, casting, shakeout, cleaning and fin-
ishing are unknown, emissions are expected to be small since
it is likely that the majority of lead contained in the
furnace charge material is emitted from the furnace as fume
or removed in the slag.
In summary, a minimal, but reasonable, inplant sampling
program for the three industry types is tabulated as fol-
lows :
Plant Type and
Number Tested
Secondary Lead
Smelters - 2
Ferroalloy
Plants - 2
Gray Iron
Foundries - 2
Source
Blast Furnace
Pot Furnace
Electric-arc
Furnace
Cupola
Electric-arc
Furnace
Operations
Charging/
tapping
Charging/
melting/
tapping
Charging/
smelting/
tapping
Charging/
tapping
Charging/
melting/
tapping
In addition, at all plants the handling and transfer
operations should be sampled. Also, raw material samples
should certainly be taken for comparative analysis with
fugitive samples.
7-6
-------
7.2.2 General Cost Estimate/Schedule
A cost estimate was presented in Section 5.0 for emis-
sion factor development for individual processes within one
facility. Since there are process variations between
facilities within the three categories studied, a suggested
test program with cost estimate (4^15%) is developed herein
for a six-plant, two per category, test series. There are
considerable savings of resources due to efficiencies gained
in a multiple plant test program. The following assumptions
have been made:
0 Each field test/plant will require an average of
12 days (10 test days plus 2 travel days) instead
of 20. This reduction in the number of sampling
days will occur as a result of a multiple test
series. Repeated tests will increase the famil-
iarity with the process and sampling techniques
which will result in greater efficiency.
0 Pre-survey travel and man-hour requirements can be
reduced by 40 percent through planning and dual-
purpose trips.
0 Fabrication costs for specialized field sampling
ducts, temporary hoods, etc., can be reduced by 50
percent through multiple use in various plants.
Therefore the estimated cost/test series/plant pres-
ented on Page 5-25 is modified as follows:
7-7
-------
Estimated
Item cost*
0 Burdened labor, 1640 hours @ $22/hr $36,080
0 Transportation and Per Diem
(1) Pre-survey, 2 men for 1.2 days 360
@ $150/day (including trans-
portation)
(2) Field test, 6 men for 12 days @ 5,040
$70/day (including transporta-
tion)
0 Sampling Site Preparation
(I) Four sites @ $100 ea 400
(2) Fabrication of sheet metal ducts, 3,000
hoods, etc.
0 Expendable supplies; estimated @ $40/ 480
day for 12 days
0 Analysis of raw materials 400
Total Estimated Cost ' $45,360
* This estimate is based on the measurement and
determination of emission factors from the
four emission sources (vents, common sources,
materials handling, and quasi-stack) at one
plant.
Therefore, if the six-plant series were to be imple-
mented, U.S. EPA should anticipate an expenditure in the
range of $230,000 to $310,000. Additionally, the entire
test series, plus data analysis and reporting, should be
able to be accomplished over an eight to nine month period.
7-8
-------
TECHNICAL REPORT DATA
(Plcast read Instructions on the rcicrst- if/ore completing!
1 REPORT NO. 2
EPA-450/3-78-003
4 TITLE ANDSUBTITLE A Method for Cha rac te r i z a t ion
and Quantification of Fugitive Lead Emissions
"rom Secondary Lead Smelters, Ferroalloy
Plants and Gray Iron Foundries (Revised)
7 AUTHOR(S) _ , „ -, -, „
John M. Zoller, George A. Jutze,
Larry A. Elfers
P PERFORMING ORG \NIZATION NAME AND ADDRESS
PEDCo Environmental, Inc.
11499 Chester Road
Cincinnati, Ohio 45246
12 SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Monitoring and Data Analysis Division
Research Triangle Park, North Carolina 27711
3 RECIPIENT'S ACCESSION>NO
5 REPORT DATE Revised
January J97R August- 1 978
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
3264-G, 3327-J
10. PROGRAM ELEMENT NO.
11 CONTRACT/GRANT NO.
68-02-2515, 68-02-2585,
Task No. 7 Task No. '10
13. TYPE OF REPORT AND PERIOD COVERED
Fi nal
14 SPONSORING AGENCY CODE
200/04
15 SUPPLEMENTARY NOTES
EPA Task Officer - Charles C. Masser
16. ABSTRACT
This report summarizes current information relative to fugitive lead emissions from
secondary lead smelters , ferroalloy plants , and gray iron foundries . Also included
are an investigation of the application of fugitive lead emission factors from other
source categories to the three subject industries, and a report on the applicability
oi fugitive lead factors that could be developed from a field study . Current state of
the art techniques for source measurements of fugitive emissions (i.e. inplant) and
ambient measurements of fugitive emissions (i.e. up wind/ downwind) are presented
and compared . This report aids in determining if field studies are worthwhile and
recommends the types of studies to be followed.
17. KEY WORDS AND DOCUMENT ANALYSIS
a DESCRIPTORS
Air Pollution
Dust
Metallurgy
Lead
18. DISTRIBUTION STATEMENT
Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
Emission Factors
Fugitive Dust
Secondary Lead
Smelters, Ferro-
alloy Plants,
Foundries
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATl Field/Group
13B
11G
11F, 13H
21. NO. OF PAGES
110
22. PRICE
EPA Form 2220-1 (9-73)
7-9
------- |