&EFA
United States
Environmental Protection
Agency
Office of Air Quality
Plann;ng and Standards
Research Triangle Park NC 27711
EMB Report 84SLD3
March 1984
Air
Secondary Lead
Smelter Test Of
Area Source
Fugitive Emissions
For Arsenic,
Cadmium, And
Lead
Chloride Metals
Tampa, Florida
Volume 1
-------
CORPORATION
DCN 84-222-078-12-01
EMISSION TEST REPORT:
CHLORIDE METALS SECONDARY LEAD SMELTER
3507 S. 50th Street
Tampa, Florida
Final Report
Volume 1
Prepared for:
Mr. Frank Clay
U. S. Environmental Protection Agency
ESED/EMB (MD-13)
Research Triangle Park, North Carolina 27711
EPA Contract No. 68-02-3850
Work Assignments 04 and 12
ESE/Project No. 83/19
Prepared by:
M. R. Fuchs
M. J. Krall
G. D. Rives
Radian Corporation
7 December 1984
8501 Mo-Pac Blvd./P.O. Box 9948 / Austin, Texas 78766 / (512)454-4797
-------
TABLE OF CONTENTS
Section Page
1 Introduction 1-1
1.1 Program Background 1-1
1.2 Program Objectives 1-2
1.3 Brief Process Description 1-3
1.4 Emission Test Program 1-4
1.5 Descriptions of Report Sections 1-6
2 Presentation of Results 2-1
2.1 Fugitive Emission Results 2-1
2.1.1 Area Sources Fugitive Emission Results 2-3
2.1.2 Discussion of Results 2-7
2.1.2.1 Effects on Area Source Emission
Rates 2-7
2.1.2.2 Data Outliers 2-22
2.1.2.3 Evaluation of Control Approach 2-23
2.1.3 Smelter Complex Emission Rates 2-25
2.1.4 Particle Size Distribution Results 2-33
2.2 Smelter Building Concentration Results 2-36
2.3 Point Source Emission Results 2-39
2.4 Recommendations for Testing Activities on Future
Programs 2-39
2.4.1 Ambient Monitoring Network 2-39
2.4.2 Isokinetic Sample Collection 2-41
2.4.3 Smelter Building Concentration Sampling 2-42
3 Process Description 3-1
3.1 General Plant Information 3-1
3.2 Process Information 3-3
3.3 Production 3-5
3.4 Area Fugitive Emissions and Vehicle Activity
Patterns 3-9
3.4.1 Blast Furnace Operation 3-9
3.4.2 Battery Breaking 3-10
3.4.3 Lead Casting 3-10
3.4.4 Refining Activities 3-17
3.5 Test Areas 3-17
3.5.1 Battery Breaking Area 3-20
3.5.2 Raw Material Storage Area 3-22
3.5.3 Smelter Building 3-24
3.5.4 Slag and Dross Storage Area 3-27
3.5.5 Roadway (Major Vehicle Pathway) 3-30
3.5.6 Sanitary Baghouse and Scrubber Testing 3-33
4 Sampling Locations 4-1
4.1 Area Sources 4-1
4.1.1 Battery Breaking Area (Area 1) 4-1
4.1.2 Raw Materials Storage Area (Area 2) 4-3
-------
COBPORjmOf
TABLE OF CONTENTS (Continued)
Section Page
4.1.3 Slag/Dross Storage Area (Area 4) 4-3
4.1.4 Roadway 4-6
4.2 Smelter Building 4-6
4.3 Stationary Sources 4-9
4.3.1 Slag Tap Baghouse 4-9
4.3.2 Kinpactor Scrubber 4-12
5 Sampling and Analysis 5-1
5.1 Sampling Techniques 5-1
5.1.1 Fugitive Emissions Sampling Techniques 5-1
5.1.1.1 Exposure Profiling Technique 5-1
5.1.1.2 Smelter Building Emissions
Measurements 5-11
5.1.2 Smelter Building Concentrations Method 5-13
5.1.3 EPA Method 108 5-15
5.2 Analytical Methods 5-16
5.2.1 Arsenic, Lead, Cadmium Analysis 5-16
5.2.1.1 Dissolution Technique 5-16
5.2.1.2 Instrumental Analysis 5-17
5.2.2 Filter Weighings 5-18
5.2.2.1 Filter Conditioning 5-18
5.2.2.2 Instrument 5-18
6 Quality Assurance 6-1
6.1 Sampling Quality Assurance 6-1
6.1.1 Equipment Calibration 6-2
6.1.2 Sampling Protocols 6-6
6.1.3 Sample Handling Techniques 6-7
6.2 Analytical Quality Control 6-8
6.2.1 Radian Laboratories Quality Control 6-8
6.2.2 Project-Specific Quality Control 6-9
6.3 Data Analysis Quality Assurance 6-14
References
ii
-------
TABLES
Number Page
1 Average Area Source Emission Rates •. 2-4
2 Comparison of Emission Rates to Measured Concentrations 2-6
3 Smelter Building Emission Rates 2-8
4 Effect of Influential Factors on Emission Rates 2-21
5 Comparison of Emission Rates 2-24
6 Roadway Emissions Reduction by Wetting 2-26
7 Smelter Complex Emission Rates Determined by Application
of the Ventilation Model 2-31
8 Smelter Complex Emission Rates 2-32
9 Chemical Concentrations in Particle Size Fractions 2-34
10 Smelter Building Concentrations 2-37
11 Stationary Sources Emission Rates 2-40
12 General Vehicle Activity Patterns Associated with
Blast Furnace Operation 3-11
13 General Vehicle Activity Patterns Associated with
the Battery Breaking Operations 3-13
14 General Vehicle Activity Patterns Associated with
Ingot Casting 3-15
15 General Vehicle Activity Patterns Associated with
Refining Kettle Operations 3-18
16 Battery Breaking Area. Summary of General Conditions and
Process-Related Activities During the Sampling Period 3-21
17 Raw Materials Storage Area. Summary of General Conditions
and Process-Related Activities During the Sampling Period.... 3-23
18 Smelter Building. Summary of General Conditions and
Process-Related Activities During the Sampling Period 3-25
19 Summary of General Activities During Site-Specific Low
Volume Sampling Periods Inside the Smelter Building 3-26
ill
-------
TABLES (Continued)
Number Page
20 Slag and Dross Storage Area. Summary of General Conditions
and Process-Related Activities during the Sampling Period.... 3-29
21 Roadway. Summary of General Conditions and Process-Related
Activities During the Sampling Period 3-31
22 Process Conditions During the Sanitary Baghouse Outlet
Tests 3-34
23 Kettle Activities During the Scrubber Outlet Tests 3-35
24 Sierra Model 235 Five-Stage Impactor Particle Size
Cut-Offs (Microns) 5-7
25 Summary of Calibrated Equipment Used in Sampling at
Chloride Metals 6-3
26 Hi-Volume Sampler Calibration Checks 6-5
27 Results of Analysis of Lead Quality Control Standard 6-10
28 Results of Analysis of Duplicate Filter Samples 6-12
29 Comparison of Blank Filter Weights 6-13
iv
-------
CORPORATION
FIGURES
Number Page
1 The layout of the Chloride Metals secondary lead smelter
indicating process area sources 1-5
2 Smelter building data graphs. Emission rates are not
corrected for background (upwind) concentrations 2-10
3 Raw materials storage area data graphs. Emission rates are
not corrected for background (upwind) concentrations 2-12
4 Roadway (dry and wet) data graphs. Emission rates are
not corrected for background (upwind) concentrations 2-14
5 Slag/Dross storage area data graphs. Emission rates are
not corrected for background (upwind) concentrations 2-16
6 Battery breaking area data graphs. Emission rates are
not corrected for background (upwind) concentrations 2-18
7 Sample poles location on March 14 2-28
8 Sample poles location on March 15 2-29
9 Sample poles location on March 16 2-30
10 Plot plan of Chloride Metals plant in Tampa, Florida
illustrating the five fugitive sampling areas 3-2
11 Blast furance lead production February 20 through March 21,
1984. Daily lead production (tons) prior to and during
the testing period 3-6
12 Refined metal production February 20 through March 21,
1984. Daily refined lead production (tons) prior to and
during the testing period 3-8
13 Major vehicle activity patterns associated with blast
furnace operation 3-12
14 Major vehicle activity patterns associated with the
battery blasting operation 3-14
15 Major vehicle activity patterns associated with the
casting or pouring of refined metal 3-16
16 Major vehicle activity patterns associated with the
refining operations 3-19
v
-------
FIGURES (Continued)
Number Page
17 Plot plan of Chloride Metals Secondary Lead Smelter
identifying area sources 4-2
18 Battery breaking area looking southwest. Raw materials
storage area and smelter building in background 4-4
19 Raw materials storage area looking north. Battery
breaking area behind concrete wall 4-5
20 Slag/dross storage area looking west. Smelter buiding
to the right in picture 4-7
21 Roadway looking west. Smelter building in background—
process baghouse on the left in picture 4-8
22 Smelter building dimensions and effective emission
areas of openings 4-10
23 Exposure profiling pole sampler with three total suspended
particulate samplers 5-4
24 Ventilation model 5-9
25 Determination of ventilation base dimensions 5-10
26 Normal distribution of measured species concentrations 5-12
27 Critical orifice sampler with filter cassette 5-14
vi
-------
comKMumOM
SECTION 1
INTRODUCTION
In January 1981, the Environmental Protection Agency (EPA) listed
arsenic as a hazardous air pollutant. In response to the listing of
arsenic, the EPA is charged with the responsibility of determining those
industries which pose a risk to the health of the population and to the
environment. EPA is presently assessing the health and environmental risk
posed by arsenic emissions from the secondary lead smelting industry. This
report presents the results of an emissions test study at a secondary lead
smelter to develop an arsenic emissions data base to assess the health and
environmental risks presented by that industry.
1.1 PROGRAM BACKGROUND
The EPA is presently assessing the potential for arsenic emissions from
secondary lead smelters. Available data indicate that the secondary lead
industry is a source of arsenic, and modeling studies indicate that fugitive
sources account for the majority of arsenic emissions from secondary lead
smelters. The existing data base for fugitive arsenic emissions from secon-
dary lead smelters is insufficient to accurately quantify the magnitude of
arsenic fugitive emissions from the industry. Additionally, no empirical
data are available to assess the emissions reduction potential of various
fugitive emissions control alternatives. The purpose of this project is to
develop relative magnitudes of fugitive arsenic emissions from the secondary
lead industry and to provide data to assess the reduction potential of
fugitive emissions control techniques. The test data will serve as inputs
to the Human Exposure Model (HEM) to assess the risk to the population
associated with fugitive arsenic emissions from the secondary lead industry.
1-1
-------
1.2 PROGRAM OBJECTIVES
The primary objectives of this program are:
• to estimate fugitive arsenic emissions from a secondary
lead smelter;
• to determine relative contributions of emissions from
various sources at the smelter; and
• to provide data to assess the reduction potential of
fugitive emissions practices typically applied in the
secondary lead smelting industry.
In addition to fugitive arsenic emissions, fugitive emissions of lead,
cadmium, and particulate matter were also investigated. Lead emissions were
determined to allow a comparison of arsenic emissions to lead emissions.
Lead fugitive emissions data for secondary lead smelters are available
through other studies and relative emissions rates of arsenic to lead deter-
mined concurrently in this study may be transferable to other more limited
data sets.
EPA is studying the health effects of cadmium and is considering
listing cadmium as a hazardous air pollutant. Emission rates for cadmium
were determined to be used in future evaluations of the environmental risk
of cadmium emissions from the secondary lead industry.
The emissions test for fugitive arsenic, lead, and cadmium emissions
utilized test methods developed for particulate matter. The collected
particulate was analyzed for arsenic, lead, and cadmium to determine emis-
sions for those parameters. Particulate mass concentrations were determined
in this study and particulate emission rates were calculated to allow a
comparison of the relative concentrations of arsenic, lead, and cadmium to
the particulate loading.
1-2
-------
1.3 BRIEF PROCESS DESCRIPTION
Secondary lead smelters process used batteries and lead scrap to pro-
duce lead and lead alloys. Typical processing steps at a secondary lead
smelter are recovering the lead and lead parts from used batteries, storing
that material prior to charging the material into the furnace, smelting the
scrap lead to produce crude lead, and refining the lead to produce a high
purity lead or a lead alloy. Slag and process dusts are typically retained
on the plant site for a period of time prior to reuse or disposal.
The processing steps described above are generally performed in recog-
nizable, definable areas of the plant. Each of these processing areas
presents a potential for arsenic emissions.
Testing was conducted at the Chloride Metals secondary lead smelter
located in Tampa, Florida. The Chloride Metals plant was selected as the
test site because:
• the plant employs emissions controls typical of the indus-
try;
• lead alloys typical of the industry are produced;
• relative to the secondary lead insutry as a whole, the
physical layout of the facility was judged to be more condu-
cive to the measurement of emissions from areas considered
to be the major sources of fugitive emissions; and
• the expected weather during the proposed time of testing
was conducive to the performance of the test program.
The Chloride Metals plant processes used batteries almost solely and
produces lead and lead alloys. The processing steps are typical of the
1-3
-------
RADIAN
secondary lead industry. The layout of the Chloride Metals secondary lead
smelter is shown in Figure 1.
1.4 EMISSION TEST PROGRAM
The primary objective of the emission test program is to estimate
fugitive emission rates at the Chloride Metals secondary lead smelter. Five
process area sources were identified and tested for fugitive emissions at
Chloride Metals:
• smelter building,
• raw materials storage,
• battery breaking area,
• slag/dross storage, and
• an intraplant vehicle roadway, i.e., roadway.
The locations of those areas are shown in Figure 1.
Fugitive emission rates for arsenic, lead, cadmium, and particulate
were determined for the five sources listed above using an adaptation of the
exposure profiling technique. By this method ambient concentrations are
measured in the emission plume of the source, and the emission rate is
calculated by ventilating the area of the plume by the average wind speed
during the sampling period.
Another objective of the program is to assess the reduction potential
of fugitive emissions control practices typically applied at secondary lead
smelters. Chloride Metals wets most of the smelter complex to control
fugitive emissions. This practice is common in the industry. Automatic
sprinklers wet the roadway, battery breaking area, and dross bins. The
roadway was selected as the site to be tested to assess the reduction
potential for controlling fugitive emissions by wetting. The roadway area
was tested wet (controlled) as is typically the case at Chloride Metals and
1-4
-------
Columbia Street
Natural Gas
Storage
BATTERY
BREAKING
AREA\ \
RAW MATERIALS
STORAGE
AREA
ROADWAY
\\
SMELTER
BUILDING
Slag Tap
Bagnouse
SMELTER
BUILDING
EXTENSION
Refining
Kettles
Liad
0»ide
Plint
BLAST SLAG
ff /STORAGE
•V /AREA/
GE'
Kinpactor
Scrubber
'DRO
Raleigh Street
AREA 1 92 ft. x 121 ft. = 11,132 ft2
AREA 2 59 ft. x 86 ft. = 5,074 ft2
AREA 3 144ft. x 78ft. = 11,232ft2
AREA 4 73 ft. x 118 ft. = 8,614 ft2
AREA 5 48 ft. x 140 ft. = 6,720 ft2
Figure 1. The layout of the Chloride Metals secondary
lead smelter indicating process area sources.
1-5
-------
dry (uncontrolled) achieved by turning off the sprinklers and allowing the
area to dry.
Chloride Metals also employs other fugitive emissions controls typical
of the secondary lead industry. The slag tap and lead well of the blast
furnace are hooded and vented through a baghouse. And the refining kettles
and associated drossing enclosures are hooded and vented through a high
energy wet scrubber. Those two sources were tested to determine arsenic,
lead, cadmium, and particulate emissions from the control devices. Concen-
trations of lead, arsenic, and cadmium were measured in the smelter building
to assess fugitive emissions escaping the hooding systems.
1.5 DESCRIPTIONS OF REPORT SECTIONS
The Presentation of Results is in Section 2. Section 3 presents the
Process Description. The Sampling Locations are depicted in Section 4.
Sampling and Analysis are discussed in Section 5. Quality Assurance activi-
ties for the program are described in Section 6. Specific and individual
sampling and analytical measurements, calculations, preliminary results, and
other pertinent information associated with the program are included in
appendices to this report.
1-6
-------
CORPORATION
SECTION 2
PRESENTATION OF RESULTS
The major objectives of the emissions test at Chloride Metals are to
estimate fugitive emissions from process areas at the smelter and to rank
the sources of fugitive emissions according to their relative magnitude.
The following sections present the results of testing at Chloride Metals.
Emission rates for arsenic, lead, cadmium, and particulate are determined
and presented.
2.1 FUGITIVE EMISSION RESULTS
Data available prior to the performance of this program indicate that
fugitive emissions at secondary lead smelters are the primary source of
emissions of the smelter complex. Based on those results, the emphasis of
this program was placed on the determination of fugitive emissions. Fugi-
tive emissions from four area sources and the smelter building were mea-
sured. The four area sources are:
• slag/dross storage area,
• battery breaking area,
• raw materials storage, and
• the major vehicle pathway or roadway.
Another aspect of this program was to provide data to evaluate typical
practices used to control fugitive emissions at secondary lead smelters.
One of the most common control practices is the wetting of yard areas and
material storage piles to suppress the entrainment of particulate matter.
At secondary lead smelters, process areas are usually open areas in which
one or possibly more processing steps are performed. At Chloride Metals,
2-1
-------
the four area sources listed above and the smelter building are the major
process areas. There are other areas at Chloride Metals which contribute to
smelter complex emissions, but the emissions of other areas, such as the
rolling mill building, are considered minor relative to the four area
sources just discussed and the smelter building.
Chloride Metals has an automatic sprinkler system which periodically
wets the roadway and battery breaking area. Dross bins are continuously
wetted by sprinklers, and the smelter building is occasionally wetted manu-
ally. Water used to wet these areas is collected in an underground collec-
tion system, and the water is treated and reused.
Because the roadway is paved and dries quickly, it was selected as the
test area to evaluate fugitive emissions control by wetting. The roadway
was tested wet (controlled) and dry (uncontrolled) to evaluate the effec-
tiveness of wetting for the suppression of fugitive emissions.
Two stationary sources (non-fugitive) were also tested to determine
emission rates for arsenic, lead, cadmium, and particulate. These two
sources, the slag tap baghouse and a high energy wet scrubber, were tested
because they control sources which may be fugitive at other plants. The
slag tap baghouse removes particulate matter collected by a hooding system
over the slag tap and lead well at the blast furnace. The high energy wet
scrubber (the Kinpactor scrubber) controls emissions collected by a hooding
system over four refinery kettles and the associated dross bins also located
in the smelter building.
Section 2.1.1 presents emission rate results and associated data for
the four area sources and the smelter building. Section 2.1.2 discusses the
area emission rate results. The data gathered during this program also
allowed the determination of emissions from the smelter complex. Those
results are presented in Section 2.1.3. Section 2.1.4 presents particle
size distribution results. Section 2.2 presents results of arsenic, lead,
and cadmium concentration measurements in the smelter building. The results
2-2
-------
of stationary (point) source emissions are presented in Section 2.3. Recom-
mendations for future testing programs are given in Section 2.4.
2.1.1 Area Sources Fugitive Emission Results
Arsenic, lead, cadmium, and particulate emission rates were determined
for four area sources and the smelter building. The area sources are:
• slag/dross storage,
• battery breaking area,
• raw materials storage, and
• roadway.
As proposed in the test plan for this project, each source was to be
tested three times with the exception of the roadway which was to be tested
six times—three times wet and three times dry. During the test program,
the following number of measurements were performed on the area sources:
• slag/dross storage - 3,
• battery breaking area - 4,
• raw materials storage - 5,
• roadway (dry) - 3,
• roadway (wet) - 3, and
• smelter building - 7.
The extra measurements on the smelter building, raw materials storage,
and battery breaking area were conducted to develop more data on those
sources which were visually observed during the testing program to be the
most significant sources of emissions by either seeing particulate and/or
fume emissions or by relative loadings of the particulate filters.
The average fugitive emission rates of arsenic, lead, cadmium, and
particulate for the area sources are presented in Table 1. The range of
emission rates measured for each area source are also presented.
2-3
-------
TABLE 1. AVERAGE AREA SOURCE EMISSION RATES
Emission Rates (Range) (g/hr)a
Area
Smelter Building
Raw Materials Storage
Roadway (Dry)
Slag/Dross Storage
Battery Breaking Area
Roadway (Wet)
As
0.44
(0.10-0.92)
0.41
(0.027-1.7)
0.26
(0.034-0.66)
0.21
(0.18-0.25)
0.062
(0.027-0.12)
0.039
(0.020-0.072)
Pb
80
(32-160)
67
(12-240)
82
(8.3-210)
30
(12-56)
13
(9.3-16)
12
(3.6-19)
Cd
0.093
(0.017-0.23)
0.057
(0.015-0.18)
0.17
(0.043-0.39)
0.049
(0.034-0.063)
0.030
(0.010-0.060)
0.022
(0.0061-0.049)
Particulate
280
(110-730)
270
(100-740)
390
(180-790)
180
(84-270)
200
(84-310)
170
(170-180)
90
D
i
Number of
Measurements
7
5
3
3
4
3
aEmission rates are not corrected for background (upwind) concentrations.
-------
The area sources listed in Table 1 are ranked by the magnitude of the
arsenic emission rate, the element of primary focus in this program. The
results indicate that the smelter building is the single largest contributor
of arsenic, and ranks second in lead, cadmium, and particulate emissions.
These data are consistent with visual observations during the testing pro-
gram. Note that the roadway (dry) is the third-ranked emission source
because of arsenic emissions, but the roadway (dry) has the highest lead,
cadmium, and particulate emissions. These results seem inconsistent and are
the result of extraordinarily high emissions on one sampling day, March 20,
as seen in Table 2. Section 2.1.2.2 discusses this point further. However,
it should be pointed out that the roadway is kept wet at Chloride Metals and
the roadway (dry) data are not typical of day-to-day operations.
Table 2 presents the individual measurements and average results of
emission rates for the four area sources and smelter building. The data are
presented with wind speed to allow a quick correlation of wind speed to the
magnitude of emissions. During the testing period, wind speed varied from
2.8 to 7.7 miles per hour.
As is discussed in Section 5.1.1.1, the testing of an area source
included measuring ambient concentrations of arsenic, lead, cadmium, and
particulate just downwind of the area source and also just upwind to evalu-
ate the magnitude of upwind emissions into the area source being tested.
Table 2 presents the average upwind and downwind concentrations measured
during each run for the determination of area sources emission rates. These
data allow an evaluation of the effect of upwind concentrations on the
magnitude of downwind concentrations and the emission rate. The average
concentrations were determined from results measured at vertical elevations
of 4 feet, 8 feet, and 12 feet and for all downwind samplers run concur-
rently during a test for a given source.
Upwind concentrations were not used to determine the amount of arsenic,
lead, cadmium, and particulate entering the four area sources and thus
contributing to the downwind concentrations and emission rates. Smelter
2-5
-------
TABLE 2. COMPARISON OF EMISSION RATES TO MEASURED CONCENTRATIONS
Area
Smelter
Bui Id lag
Haw
Materials
Storage
Slag/
Dross
Storage
Roadway
(dry)
Battery
Breaking
Yard
Headway
(wet)
Date
3/13
3/14
3/15
3/16
3/19
3/19
3/20
3/13
3/15
3/16
3/19
3/20
3/6
3/7
3/8
3/12
3/19
3/20
3/13
3/14
3/15
3/16
3/7
3/8
3/9
Time
1025-1652
0919-1607
0851-1617
0845-1603
(AM) 1005-1315
(PM) 1348-1552
0901-1333
1040-1657
0847-1614
0850-1550
0959-1553
0921-1319
1016-1715
1108-1709
1143-1702
0850-1548
0953-1550
0910-1337
1107-1654
0925-1615
0856-1611
0852-1556
1130-1755
1122-1717
0922-1612
Wind
Speed
(mph)
4.1
3.8
5.5
3.1
4.7
4.7
7.7
3.7
4.5
2.8
4.3
7.7
7.7
4.8
4.0
6.0
4.3
7.3
3.7
3.9
4.5
2.8
4.8
4.4
4.4
Emission Bate*
(s/hr)
As
0.78
0.20
0.17
0.66
0.24
0.10
0.92
0.027
0.038
0.23
0.056
1.7
0.25
0.18
0.20
0.034
0.086
0.66
0.049
0.027
0.051
0.12
0.072
0.025
0.020
Pb
120
47
41
43
120
32
160
15
12
47
23
240
12
21
56
8.3
29
210
11
14
16
9.3
19
3.6
14
Cd
0.070
0.11
0.090
0.097
0.017
0.036
0.23
0.032
0.018
0.040
0.015
0.18
0.049
0.063
0.034
0.066
0.043
0.39
0.060
0.030
0.019
0.010
0.011
0.049
0.0061
Particulate
340
180
150
160
300
110
730
100
140
240
150
740
270
200
84
210
180
790
310
190
220
84
180
170
170
Average Downwind
Concent rat ion (ua/m )
As Pb
0.542 84.1
0.212 52.2
0.236 55.6
1.76 126
0.312 175
0.228 71.0
0.750 134
0.0595 20.9
0.0512 15.9
0.189 27.8
0.0676 27.2
0.460 82.6
0.160 9.98
0.150 17.5
0.190 21.0
0.0460 11.2
0.0480 18.4
0.280 48.9
0.0255 8.36
0.0221 12.1
0.0266 12.8
0.0498 8.12
0.0477 10.7
<0.0114 2.81
<0. 00962 3.71
Cd
0.0474
0.109
0.108
0.255
0.0304
0.0804
0.190
0.0312
0.0196
0.0318
0.0188
0.131
0.0162
0.0377
0.0367
0.0238
0.0251
0.0938
0.0192
<0.0218
0.0120
0.0132
0.00718
0.00722
<0. 00273
Particulate
234
195
192
364
393
238
594
133
135
148
181
391
121
118
139
173
113
295
123
141
142
95.2
112
97.4
121
As
0.0343
0.0319
0.0340
0.0435
0.0572
0.0572
0.460
0.628
0.0148
0.0142
0.0749
0.554
0.173
0.271
0.0281
0.0209
0.0126
0.132
0.0282
0.0155
0.0272
0.0118
0.0399
Average Upwind
Concentration (ua/ur)
Pb
4.31
10.2
9.82
9.54
6.48
6.48
36.8
110
8.75
15.9
31.7
114
13.8
50.0
<0.516
4.59
2.78
18.4
8.16
5.79
9.17
6.40
21.9
<0. 00208 1.22
0.0172
24.3
Cd
0.0119
0.0344
0.0349
0.0604
0.0213
0.0213
0.0914
0.0676
0.0131
0.00752
0.0334
0.191
0.00936
0.0938
0.00580
0.0279
0.0116
0.0756
0.0304
0.0180
0.0153
0.0128
0.00804
<0. 00105
0.00413
Particulate
92.8
99.4
108
85.9
79.1
79.1
275
271
117
119
150
588
114
267
59.2
123
80.3
214
162
116
146
112
149
62.6
230
m
\
aEmission rates are not corrected for background (upwind) concentrations.
-------
building emission rates were determined without the background contribution
and also by subtracting the background contribution (without and with back-
ground correction). Emission rates with and without background correction
were determined for the smelter building because of less difficulty in the
data reduction methodology in comparison to the ventilation model approach
(described in Section 5.1.1.1) used for the data reduction of the four area
sources. Smelter building emission rates with and without background cor-
rection are presented in Table 3. The arsenic emission rate of the smelter
building is reduced by 42 percent by subtracting the background contribu-
tion.
2.1.2 Discussion of Results
The area source and smelter building emissions were evaluated for the
effect by three factors: upwind concentration, wind speed, and smelter
building emission correction. The results were also reviewed for consis-
tency. Roadway data allowed an evaluation of the degree of control on
fugitive emissions achievable by wetting.
2.1.2.1 Effects on Area Source Emission Rates—
In the evaluation of the emission rates developed for the area sources,
several questions evolved over what may affect fugitive emission rates at
the Chloride Metals secondary lead smelter. Those questions include:
• To what extent are area emission rates affected by measured
upwind concentrations?
• What effect does wind speed have on fugitive emission rates?
and
• Since visual observations during the test program indi-
cated that the smelter building was the predominant source
of emissions and calculated smelter building emission rates
were highest, what effect does the location of the smelter
2-7
-------
TABLE 3. SMELTER BUILDING EMISSION RATES
i
00
Emission Rate (g/hr) without
Background Correction
Date
3/13/84
3/14/84
3/15/84
3/16/84
3/19/84 (AM)
3/19/84 (PM)
3/20/84
Average
Time Speed
(mph)
1025-1652 4.1
0919-1607 3.8
0851-1617 5.5
0845-1603 3.1
1005-1315 4.7
1348-1552 4.7
0901-1333 7.7
As
0.78
0.20
0.17
0.66
0.24
0.10
0.92
0.44
Pb
120
47
41
43
120
32
160
80
Cd
0.070
0.11
0.090
0.097
0.017
0.036
0.23
0.093
Particulate
340
180
150
160
300
110
730
280
5H
!g
Emission Rate (g/hr) with ff
Background Correction Z
As
0.73
0.14
0.10
0.63
0.17
0.016
B
0.26
Pb
120
28
21
36
110
22
86
60
Cd
0.052
0.036
B
0.055
B
0.0037
0.037
0.026
Particulate
200
B
B
98
200
B
140
91
B = background contribution greater than total emission rate.
-------
building relative to the location of the area source and
wind direction have on fugitive emissions?
One of the concerns that became apparent upon reviewing the test data
was the relative concentrations measured at upwind and downwind samplers.
For about one-third of the area source fugitive emission tests, average
upwind concentrations were greater than average downwind concentrations.
Upwind concentration results were not used to determine background contribu-
tions to the source during the testing periods. Obviously, the flow of a
species into an area source will affect the emission rate. To evaluate the
effect of measured upwind concentrations upon emission rates, graphs were
drawn to compare arsenic, lead, and particulate emission rates to upwind
concentrations of those respective species. The graphs are shown in Figures
2 through 6. A review of those graphs indicates that arsenic emissions from
the roadway demonstrate a dependence upon the upwind concentration of arse-
nic. If the data generated on March 20 are not included, then that apparent
influence is negated.
In a similar fashion, downwind concentrations of arsenic measured at the
roadway were plotted against upwind concentrations of arsenic. The purpose
of this graph was to investigate if the calculation of emission rates in some
way dampened the apparent effect of upwind concentrations. The relationship
of the two plots is very similar and indicate that arsenic emissions may be
positively influenced by the upwind concentration of arsenic.
Wind speed certainly has an effect upon fugitive emissions of suspended
particulate due to the force available to entrain particulate matter. For
process operations in obstructed, turbulent wind pattern areas, the effect
is less defined. To evaluate the effect of wind speed upon area emission
rates, wind speed was plotted against arsenic emission rates. These plots
are also shown in Figures 2 through 6. The raw materials storage area and
roadway appear to be influenced by wind speed, but that apparent effect is
principally the result of data points for emission rates measured on March
20.
2-9
-------
1.0
0.73
a
3
0.50
0.2S
0.10 0.20 0.30 0.40
Upwind Concentration at Arsenic ifi?/ m^)
0.50
170
150
£
2
i 100
«
.2
6
SO
10 20 30
Upwind Concentration of Lead l|i3/mj
40
Figure 2. Smelter building data graphs.
Emission rates are not corrected
for background (upwind) concentrations.
2-10
-------
i.o r
0.50
0.2S
• Wind Speed (MPH)
ISO-
'S-
4 S
270' . 360'
•W- -N-
789
750
600
a
3
2
400
200
100 200 300
Upwind Concentration of Particular (fialm*)
Figure 2. (Continued)
2-11
-------
1.80
2
3 1.20
I
0.60
0.10 0.20 0.30 0.40 0.50 0.80 0.70
Upwind Concentration of Arsenic (/ig/m3)
250
200
2
3
190
100
SO
20 40
Upwind Concentration o
30
120
Figure 3. Raw materials storage area data graphs.
Emission rates are not corrected for
background (upwind) concentrations.
2-12
-------
1.80
I
3
1.20
0.60
• DegrMS from
/ North
• Wind SpMd (MPM)
Smaller Building • South
0'
• -N-
• 0
90-
•£•
I
ISO-
'S-
4 3
270'
•W-
MO'
-N-
9
800
Z
3 600
i
400
200
200 400 800
Upwind Conctntritlon of PvtlculM* i(Ltg/m3)
Figure 3. (Continued)
2-13
-------
0.70
£ 0.60
i
8
£
0.40
0.20
0'
• -H-
• 0
• Wind Spnd (MPH)
Smelter Building • West'Soutrtweat
90'
•6-
ISO-
'S-
4 i
270' 380'
•W- -N-
789
800
I
3 800
400
200
100 200 250
Upwind Concentration of Paniculate l/jg/mJ)
Figure 4. Roadway (dry and wet) data graphs.
Emission rates are not corrected for
background (upwind) concentrations.
Note: © ® Roadway (dry}
• • Roadway (wet)
2-14
-------
0.30
: 0.20
0.10
o.ro
0.80
£
3
I 0.40
$
I
0.20
0.05 0.10
Upwind Concentration of Arsenic
0.15
200
ISO
100
50
5 10 15 20
Upwind Concentration of Lead
25
Figure 4. (Continued)
2-15
-------
0.39
0.30
2
3
8
0.20
0.10
0.10 0.20 0.30 0.33
Upwind Concentration of Arsenic (jig/m-l)
60
Z
3
-------
0.39
0.30
S
3
I 0.20
0.10
• Oegren from
North
• Wind SpMO (MPHI
300
Smilttr Building • North/Northwest
0'
•N-
0
90'
•6-
2 3
180'
•S-
4 S
270'
•W-
360'
•N-
9
S
3
200
100
100 200 300
Upwind Concentration of Paniculate (/ng/m^)
Figure 5. (Continued)
2-17
-------
0.15
0.10
0.05
0.01 0.02
Upwind Concentration of Arsenic l/iig/m3)
0.03
20
£
3 15
10
2.5 5 7.5
Upwind Concentration of Lead (/ig/m*)
Figure 6. Battery breaking area data graphs.
Emission rates are not corrected for
background (upwind) concentrations.
2-18
-------
0.13
3 0.10
o.os
• Oeqrws from
North
• Wind Speed
Smelter Building • Soutn
90* 180* 270' 380'
.£. -s. -w- -N-
2 34 58789
300
200
I
£
100
60 120 180
Uowind Concentration of Panieuiat* l^ig/mJ)
Figure 6. (Continued)
2-19
-------
As stated earlier, the smelter building was visually observed to be the
greatest emitting source at Chloride Metals during the testing program. A
series of plots were developed to evaluate the influence of smelter building
emissions on area source emission rates. This effect would be evident if
the smelter building is the predominant source of emissions at Chloride
Metals. Arsenic emission rates for the area sources and smelter building
were plotted against wind direction during the individual test periods.
These plots are also shown in Figures 2 through 6. The plots of the four
area sources also include the direction of the smelter building from the
area source. If the smelter building positively influences area source
emissions, then higher area source emission rates would be expected when the
wind is from the direction of the smelter building. The raw materials
storage area and the roadway indicate that smelter building emissions
apparently affect those area emission rates.
The series of plots discussed were developed to evaluate the following
influences on area source emission rates:
• upwind concentration of arsenic,
• wind speed, and
• smelter building emissions.
Table 4 presents a subjective evaluation of the effects of those factors on
area source emission rates.
The roadway is evaluated to be the only area source for which the
arsenic emission rate is positively affected by the upwind concentration of
arsenic. However, the apparent positive effect may be due to the very high
emissions measured on March 20. The smelter building and raw materials
storage area were also sampled on March 20 and of those two sources, the
smelter building may possibly indicate a positive effect, and the raw mate-
rials storage demonstrates no effect. The deduction of these results would
be that arsenic emissions from the roadway are positively affected by upwind
concentrations. But since the roadway was the only source of the five
2-20
-------
TABLE 4. EFFECT OF INFLUENTIAL FACTORS ON EMISSION RATES
Influential Factor
Source
Smelter Building
Raw Materials Storage
Roadway (dry and wet)
Slag/Dross Stoage
Battery Breaking Area
Upwind
Concentration
(Arsenic)
Possible Effect
No Effect
Positive3
No Effect
Negative
Wind
Speed
No Effect
Positive3
Positive3
No Effect
Negative
Smelter Building
Emissions
—
Positive
Positive
No Effect
Possible Effect
^Effect may be overestimated due to high emissions measured on March 20
2-21
-------
RADIAN
COOPORjmOM
tested to demonstrate such an effect, as a general rule, upwind concentra-
tions of arsenic did not affect the arsenic emission rates measured in this
study.
The plots of wind speed versus arsenic emission rates indicate that the
raw materials storage and roadway emissions were positively affected by wind
speed. Once again, these apparent effects appear to be due to March 20
data. Arsenic emission rates would show no dependence on wind speed if the
March 20 data were omitted.
Wind erosion of exposed areas is associated with wind speeds greater
than 12 mph, the threshold erosion velocity. An emission factor for deter-
mining fugitive particulate emissions by wind erosion of an exposed area
includes a term for the percentage of time the wind speed exceeds 12 mph
(1). There are no exponentials associated with that term. These references
would seem to indicate that it is unlikely that an increase in wind speed
from 3.5 to 7.5 mph would increase emissions by as much as several orders of
magnitude.
Two area sources indicate a positive influence on arsenic emissions by
smelter building emissions. These sources are raw materials storage and the
roadway. The battery breaking area is possibly affected. It should be
pointed out that the two sources positively affected by the smelter building
(raw materials storage and roadway) are the two area sources closest to the
smelter building.
2.1.2.2 Data Outliers—
The average emission rates of the area sources are presented in Table
1, and the sources are ranked according to arsenic emission rate. Table 2
presents the individual measurements of area source emission rates. Review-
ing the data presented in Table 2, the smelter building, raw materials
storage, and dry roadway were sampled March 20. Emission rates measured
that day were much higher for all three sources tested for all parameters
determined as compared to results from other test periods. Since all three
2-22
-------
measured sources had significantly higher emissions on March 20 for all test
species, it is not likely that the sampling and analysis procedures are
responsible for the increases. On March 20 the wind speed was about 7 1/2
miles per hour which is about twice the average of other testing periods.
It is not likely that the increase in wind speed alone accounts for the
increase in measured emission rates. Plant lead production on March 20 was
typical of the production rates during the sampling program. Particle size
distributions measured on March 20 were typical of measured results for
other test periods. The addition of alloying agents and removal of drosses
from the refining kettles was greater on March 20 than other test dates.
But since all three sources had greater emissions for all four measured
species, it is unlikely that this factor alone accounts for the entire
increase in emissions on March 20. There is no identified, defensible,
single explanation for the increased emissions measured on March 20. It is
most probable the increased emissions measured that day were due to a combi-
nation of influences. With a limited data base, it is difficult to make an
assessment as to whether the March 20 data are within the range of emissions
representative of the area sources.
Since emission rates measured on March 20 are so much higher than
emission rates determined on other days (as much as three orders of magni-
tude above the average of other runs) and since only three of the six
sources were tested on that day, average emission rates were determined for
the sources omitting the March 20 data. Those emission rates are shown in
Table 5. As seen in Table 5, the ranking of the area sources changes
considerably, although the smelter building is still the greatest source of
emissions and the wet roadway the lowest.
2.1.2.3 Evaluation of Control Approach—
An aspect of this program is to evaluate fugitive emission control
practices generally applied at secondary lead smelters. Chloride Metals
operates an automatic, sequential sprinkler system over several high traffic
areas and/or areas of heavy dust concentration.
2-23
-------
NJ
TABLE 5. COMPARISON OF EMISSION RATES
1.
2.
3.
4.
5.
6.
Area
Smelter Building^
Raw Materials (V)
Storage
Roadway (Dry) (V)
Slag/Dross (2J
Storage
Battery BreakingOi)
Area
Broadway (Wet) (T)
Smelter Complex0
Emission
As
0.44
(0.10-0.92)
0.41
(0.027-1.7)
0.26
(0.034-0.66)
0.21
(0.18-0. 25)
0.062
(0.027-0.12)
0.039
(0.020-0.072)
1.2
(0.35-3.1)
Rate (Range) Average of All Measurements
Pb
80
(32-160)
67
(12-240)
82
(8.3-210)
30
(12-56)
13
(9.3-16)
12
(3.6-19)
200
(69-490)
(0
(0
(0
(0
(0
(0
(0
Cd
0.093
.017-0.23)
0.057
.015-0.18)
0.17
.043-0.39)
0.049
.034-0.063)
0.030
.010-0.060)
0.022
.0061-0.049)
0.25
.082-0.58)
Participate
280
(110-730)
270
(100-740)
390
(180-790)
180
(84-270)
200
(84-310)
170
(170-180)
1100
(550-2200)
(0
(0
(0
(0
(0
(0
(0
Emission Rate (Range) Average Excluding March 20 Data
As
0.36
.10-0.78)
0.088
.027-0.23)
0.060
.034-0.086)
0.21
.18-0.25)
0.062
.027-0.12)
0.039
.020-0.072)
0.76
.35-1.5)
Pb
67
(32-120)
24
(12-47)
19
(8.3-29)
30
(12-56)
13
(9.3-16)
12
(3.6-19)
150
(69-260)
Cd
0.070
(0.017-0.11)
0.026
(0.015-0.040)
0.054
(0.043-0.066)
0.049
(0.034-0.063)
0.030
(0.010-0.060)
0.022
(0.0061-0.049)
0.20
(0.082-0.32)
Particulate
210
(110-340)
160
(100-240)
200
(180-210)
180
(84-270)
200
(84-310)
170
(170-180)
920
(550-1300)
aEmisaion rates are not corrected for background (upwind) concentrations.
''Relative ranking based on arsenic emission rate.
C8melter complex emission rates include vet roadway emission rates and do not include dry roadway emission rates.
-------
The major vehicle pathway or roadway is wetted by sprinklers 3 out of
every 15 minutes. The water sprayed on the roadway is collected at a drain
near the center of the roadway. The roadway was selected as the area source
to be tested for an evaluation of the degree of control achievable by
wetting. The reason the roadway was selected is that the roadway is paved
and drains well, and the area becomes completely dry in one day if the
sprinklers are turned off and it does not rain.
The roadway was tested three times wet and three time dry. The average
emission rates for the roadway wet and dry are shown in Table 6 along with
the percent of emissions controlled by wetting. The dry roadway was tested
on March 20 and the emission rates for that day are considerably higher than
other data. The dry roadway emission rates shown in Table 6 include aver-
ages including and excluding the data collected on March 20.
Because dry roadway emission rates were calculated using two sets of
data, the percentage reduction in roadway emissions achieved by wetting
demonstrates a good deal of variability. Reviewing the primary set of data
(all measurements), the wetting of a paved roadway can achieve an emissions
reduction of 56 to 87 percent.
2.1.3 Smelter Complex Emission Rates
The test data developed at Chloride Metals not only allowed determina-
tion of emission rates for the area sources and smelter building, but also
allowed the determination of emission rates for the smelter complex by two
approaches. Smelter complex emission rates for arsenic, lead, cadmium, and
particulate can be determined by summing the respective emission rates of
the four area sources and smelter building. This approach assumes that
these five sources are the predominant contributors to the smelter complex
emissions.
The second approach is to apply the ventilation model to concentration
data which can be related to the entire smelter complex. On March 14, 15,
2-25
-------
CORPORATION
TABLE 6. ROADWAY EMISSIONS REDUCTION BY WETTING
Emission Rate (g/hr)
As
Pb
Cd
Particulate
Roadway (Dry) - Average of all
Measurements3
Roadway (Wet)
Reduction
0.26
82
0.17
0.039 12 0.022
85% 85% 87%
390
170
56%
Roadway (Dry) - Average
excluding March 20 data3
Roadway (Wet)
Reduction
0.060 19
0.054
0.039 12 0.022
35% 37% 59%
200
170
15%
aEmission rates are not corrected for background (upwind) concentrations.
2-26
-------
and 16, winds were from the eastnortheast to east and pole samplers were
positioned along the length of the western edge of the smelter complex.
Figures 7, 8, and 9 indicate the locations of pole samplers on those days.
Data for the pole samplers located away from the smelter building (pole
samplers 1, 2, 4, 6, and 7) were reduced using the methodology described in
Section 5.1.1.1. The data of the pole samplers at the smelter building
(pole samplers 3 and 5) were reduced in a very similar manner according to
the approach described by Schwitzgebel (2) for well-mixed emission sources.
The pole sampler concentration results were averaged (as opposed to integra-
ting under a curve of varying concentration versus height) because the data
indicate that emissions were well-mixed. The concentration data were venti-
lated through an area defined by the height of the smelter building and 110
percent of the width of the smelter building and at the average wind speed
measured during the respective testing periods. The average flux rates for
pole samplers away from the smelter building were ventilated across the
distance remaining after the smelter building ventilation base was sub-
tracted from the length of the western boundary of the smelter complex (see
table in Appendix C). The emission rate for the smelter complex was, there-
fore, the sum of emission rates across the ventilation base of the smelter
building and across the remaining distance of the western boundary of the
smelter complex. Emission rates for arsenic, lead, cadmium, and particulate
for the smelter complex determined by this approach are presented in
Table 7.
Smelter complex emission rates determined by the two approaches are
shown in Table 8. The smelter complex emission rates determined by summing
the area source emission rates include the wet roadway emission rates and
not the dry roadway. On March 14, 15, and 16, the automatic sprinkler
system at the roadway was operating.
The correlation between smelter complex emission rates determined by
summing area source emission rates and by application of the ventilation
model is very good. It would be expected that the ventilation model
2-27
-------
RADIAN
3/14/84
Average Wind Direction (Degrees clockwise from north:
North = 0°): Battery Breaking Area - NE (41°);
Smelter-ENE (62°)
Average Wind Speed (mph) = BB - 3.9; S - 3.8
Columbia Street
Coke
Storage
O>ide
PUnt
i
n
BATTERY
BREAKING
ARE^a
5N MATERIALS -s—
STORAGE \
AREA V.
c
o
C
Natural Gas •
Storage '
Ǥ
CO CC
Plant
Office
Sewage P
Treatment I
SMELTER
BUILDING
EXTENSION
—I
•I*
Kinpactor
Scrubber
DROSS STORAGE
0|_y AREA O
BLAST SLAG
STORAGE
AREA
D
Macnme Shoo
Metals
Engng
Raleigh Street
Key:
through
numbered sample poles
sample poles used to determine smelter
complex emission rates
meteorological station
approximate wind direction
Figure 7. Sample poles locations on March 14.
2-28
-------
RADIAN
3/15/84
Average Wind Direction (Degrees clockwise from north;
North = 0°): Battery Breaking Area - ENE (64°);
Smelter - E (99°)
Average Wind Speed (mph) = BB - 4.5; S - 5.5
Columbia Street
Coke
^
1
1
<
—
l;ad
0>ide
Plint
O
o
l>
0
"3?
w
8 ys
w/ra
< >
•
<§>
BAnERY
n BREAKING
AREAJSJ
_T__ x^ w w <,
1 1 RAW MATERIALS ^ —
*=t&-sr 8i
Kinpactor \ J '
• Iscrubber
XTENSION
-AST SLAG
>TORAGE
AREA
U
Plant
Office
Sewage
Treatment
j
Metals
Engng
. S-J. DROSS STORAGE
r
) M AREA
U_J ,, — ^^ (/^
o o
•Z £
-------
RADIAN
3/16/84
Average Wind Direction (Degrees clockwise from north;
North = 0°): Battery Breaking Area - ENE (73°);
Smelter - ENE (73°)
Average Wind Speed (mph) = BB - 2.8; S - 3.1
Columbia Street
Coke
h
1
1
Storage
L;ad
O>ide
Pltnt
O
O
O
<
• /
&
V BATTERY
n BREAKING
AREA!
" i . X.
S/RN! ' RAW MATERIALS ,^—
gsJZU L-K STORAGE \
|N^i_^aL^ AREA v
* h
A
a.
o
5. o
c
<
~
«1 >V
C8./V
Js|P
f
c
i
Natural Gas •
Storage '
L -
E Loading
r
h
a
(Q\
Process
Baghouse r— i
3 r-
Q Slag Tap
1 s
O)
1*
= 1
1
1
I
-* P
m =
a §
a tr
Plant
Office
ewaqe 1
atment)
^27 | tooooool
0&0 BUILDING Metals
IKiV O
Kinpactor X^ -
& 1 Scrubber
XTENSION 1 I gngng
^ST SLAG | ( *^
TORAGE
AREA
Q DROSS STORAGE
) 1 1 1 AHEA O
L-t-J k-V^ t/^N
O
i
!
Raleign Street
Key:
through
numbered sample poles
sample poles used to determine smelter
complex emission rates
meteorological station
approximate wind direction
Figure 9. Sample poles locations on March 16.
2-30
-------
TABLE 7. SMELTER COMPLEX EMISSION RATES DETERMINED BY
APPLICATION OF THE VENTILATION MODEL
Date
3/14
3/15
3/16
As
0.52
0.81
3.4
Emission
Pb
140
200
280
Rate (g/hr)
Cd
0.29
0.41
0.50
Particulate
710
960
980
Average 1.6 210 0.40 880
2-31
-------
RADIAN
TABLE 8. SMELTER COMPLEX EMISSION RATES
As
Emission Rate Range (g/hr)
Pb
Cd
Particulate
Smelter Complex Emission 1.2
Rates Determined by (0.35-3.1)
Summing Area Sources
Emission Rates3'
200 0.25 1100
(69-490) (0.082-0.58) (550-2200)
Smelter Complex Emission 1.6 210
Rates Determined by (0.52-3.4) (140-280)
Application of the
Ventilation Model0»d
0.40
(0.29-0.50)
880
(710-980)
aArea sources emission rates taken from Table 1.
Smelter complex emission rates include wet roadway emission rates and do
not include dry roadway emission rates.
cSmelter complex average emission rates taken from Table 7.
Smelter complex emission rates determined using data produced on March 14,
15, and 16.
2-32
-------
approach would produce higher emission rates, and it does for arsenic, lead,
and cadmium. In the application of the ventilation model, concentrations
measured at the smelter building were ventilated over a distance equivalent
to 110 percent of the length of the smelter building or a distance of 158
feet. The smelter building is 144 feet long and the west side of the
building has only two openings, each 12 feet wide. Therefore, the smelter
complex emission rates determined in this manner should be a "worst case"
situation.
2.1.4 Particle Size Distribution Results
Particle size distribution measurements were performed during the
testing of the four area sources and the smelter building. The measurements
were performed using a Sierra Model 235 ambient cascade impactor. One
representative sample, including six collection substrates, collected at
each of the sources was analyzed for arsenic, lead, and cadmium.
The particle size and analytical results for the representative samples
selected are presented in Table 9. Fugitive particulate emissions were very
small and since the emissions of arsenic, lead, and cadmium are the result
of the transport of particulate emissions, arsenic, lead, and cadmium emis-
sions are associated with very small aerosols. Typically 50 to 80 percent
of the total mass of particulate emissions were less than 0.5 microns. All
of the particle size distribution results are included in Appendix G.
The particle size distribution (PSD) samples were collected at a height
of 4 feet and simultaneously with total suspended particulate filters.
There are some discrepancies between total particulate and total arsenic
concentrations measured by the particle size distribution sampler and by the
total suspended particulate filter at a height of 4 feet. Total particulate
and total arsenic results determined from the PSD data are determined by the
summing of either total mass gain or arsenic analytical results for the six
PSD filters. For that reason, the determination of total particulate and
total arsenic concentrations from the PSD data has inherently more error
2-33
-------
TABLE 9. CHEMICAL CONCENTRATIONS IN PARTICLE SIZE FRACTIONS
10
Impact or Conditions
Area Date
Slag/Dross 3/7
Storage
Roadway 3/9
(wet)
Roadway 3/12
(dry)
Raw 3/20
Materials
Storage
Wind
Time Speed Stage
(mph)
1110-1706 4.8 1
2
3
4
5
Back-up
0923-1604 4.4 1
2
3
4
5
Back-up
0859-1544 6.0 1
2
3
4
5
Back-up
0926-1319 7.7 1
2
3
4
5
Back-up
Dp a
P50
(ym)
7.16
2.96
1.45
0.94
0.49
7.34
3.04
1.49
0.96
0.50
7.18
2.97
1.46
0.94
0.49
7.43
3.07
1.51
0.98
0.51
Percentage
by Weight
-------
TABLE 9. (Continued)
to
Ol
Imp actor Conditions
Area
Wind
Date Tine Speed Stage
(mph)
Percentage
D a by Weight
50
-------
than the determination of those results from the single total suspended
particulate filter.
2.2 SMELTER BUILDING CONCENTRATION RESULTS
Concentrations of arsenic, lead, and cadmium were measured in several
areas of the smelter building. The samples were collected on filter cas-
settes with orifice controlled sampling flow rates. The filters were stand
mounted at heights of 2 feet, 4 feet, and 6 feet and were located in one of
the three following areas in the smelter building:
•
• refining kettles,
• blast furnace slag tap, or
• blast furnace lead well.
Samples collected near the refining kettles were taken during a variety
of operating conditions including:
• hard lead production,
• soft lead production,
• dressing, and
• casting.
The sample collection data sheets included in Appendix E describe process
operations during the time of sample collection.
Arsenic, lead, and cadmium concentrations measured in the smelter
building are presented in Table 10. All analytical results for arsenic and
cadmium were below the instrumental detection limit, but arsenic and cadmium
•3
results consistently show concentrations to be less than 0.4 yg/m . Lead
«j O
concentrations ranged from less than 7 yg/m to 257 yg/m with most values
o
in the range of 20 to 70 yg/m . The measured lead levels in the smelter
building are below the lead level estimated by Chloride Metals personnel.
2-36
-------
TABLE 10. SMELTER BUILDING CONCENTRATIONS
Sampler
Location
Kettle 1-2
Kettle 2
Kettle 1-2
Kettle 2
Kettles 3-4
Kettles 3-4
Kettle 4
Kettle 3
Kettle 4
Kettle 3
Furnace
(lead
well)
Furnace
(slag
tap)
Furnace
(lead
well)
Date
3/7
3/7
3/8
3/8
3/9
3/9
3/12
3/12
3/13
3/13
3/16
3/16
3/19
Time
0933-1717
0934-1717
0814-1347
0814-1347
0839-1426
0839-1426
0910-1510
0910-1510
0902-1500
0902-1220
1000-1513
1002-1513
1248-1630
Sampler
Height
(ft)
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
2
4
6
Concentrations (yg/m )
As
<0.168
<0.117
<0.158
<0.158
<0.167
<0.159
<0.233
<0.163
<0.220
<0.219
<0.232
<0.221
<0.224
<0.156
<0.211
<0.210
<0.222
<0.212
<0.216
<0.150
<0.204
<0.203
<0.214
<0.204
<0.217
<0.151
<0.205
<0.369
<0.390
<0.371
<0.248
<0.173
<0.234
<0.235
<0.248
<0.236
<0.350
<0.244
<0.330
Pb
67.0
58.3
58.0
78.8
66.6
52.9
257
75.9
72.0
27.0
28.6
30.9
48.5
27.1
<7.05
<7.01
<7.41
18.4
71.9
135
115
115
114
88.5
<7.24
<5.04
<6.83
<12.3
<13.0
<12.4
<8.28
11.5
30.5
102
99.2
55.2
<11.7
24.4
33.0
Cd
<0.105
<0.077
<0.112
<0.105
<0.111
<0.106
<0.156
<0.108
<0.147
<0.147
<0.154
<0.146
<0.149
<0.104
<0.141
<0.140
<0.148
<0.141
<0.144
<0.100
<0.136
<0.135
<0.143
<0.136
<0.145
<0.101
<0.137
<0.246
<0.260
<0.248
<0.166
<0.115
<0.156
<0.156
<0.165
<0.158
<0.233
<0.163
<0.220
(Continued)
2-37
-------
TABLE 10. (Continued)
Concentrations (yg/m )
Sampler
Location
Furnace
(slag
tap)
Kettle 4
Kettle 1
Furnace
(slag
tap)
Furnace
(lead
well)
Date
3/19
3/20
3/20
3/21
3/21
Sampler
Time Height
(ft)
1248-1630 2
4
6
1015-1511 2
4
6
1022-1514 2
4
6
0946-1318 2
4
6
0946-1318 2
4
6
As
<0.329
<0.347
<0.331
<0.247
<0.261
<0.248
<0.266
<0.185
<0.251
<0.367
<0.256
<0.346
<0.344
<0.364
<0.347
Pb
32.9
<11.6
<11.0
41.1
43.4
<8.28
<8.87
<6.18
<8.38
<12.2
<8.52
<11.5
<11.5
<12.1
34.7
Cd
<0.219
<0.231
<0.221
<0.164
<0.174
<0.166
<0.177
<0.124
<0.168
<0.244
<0.170
<0.231
<0.230
<0.242
<0.231
2-38
-------
2.3 POINT SOURCE EMISSION RESULTS
Two stationary sources were tested at Chloride Metals, the slag tap
baghouse and the high energy wet scrubber. The slag tap baghouse controls
emissions collected by a hooding system over the slag tap area and lead well
of the blast furnace. The high energy wet scrubber, or Kinpactor scrubber,
controls emissions collected by a hooding system over the refining kettles
and associated dressing enclosures.
Each source was sampled three times and emission rates for arsenic,
lead, cadmium, and particulate were determined. The emission rates of the
two stationary sources are presented in Table 11. Emissions from these two
sources are considerably less than fugitive emissions from the area sources.
2.4 RECOMMENDATIONS FOR TESTING ACTIVITIES ON FUTURE PROGRAMS
There are several recommendations which result from the Chloride Metals
emission test program for future testing activities for industrial fugitive
emissions. Those recommendations are:
• to implement a concurrent plant boundary ambient monitoring
program;
• to collect samples isokinetically for fugitive particulate
(or particulate contained) emissions; and
• if concentrations of arsenic, lead, and cadmium are measured
in a secondary lead smelter work place; the total volume of
gas sampled should be in excess of 5 standard cubic meters.
2.4.1 Ambient Monitoring Network
The inclusion of an ambient monitoring network in the scope of work for
a fugitive emissions study will provide additional information for the
2-39
-------
TABLE 11. STATIONARY SOURCES EMISSION RATES
Emission Rate (g/hr)
Source
Kinpactor
Scrubber
Date
3/13
3/14
3/20
Time
1530-1730
1450-1650
1115-1315
As
<0.041
<0.040
<0.040
Pb Cd
<1.4 <0.0084
<1.3 <0.0083
1.6 <0.0083
Particulate
130
13
67
Slag Tap
Baghouse
3/16 1038-1232 <0.021
3/19 1352-1604 <0.017
3/21 0954-1154 <0.018
2.1 0.0038
1.7 <0.0037
1.8 0.0043
29
32
41
2-40
-------
evaluation of fugitive emissions results. An ambient monitoring network
should be operated near plant boundaries concurrently with fugitive (and if
necessary point source) measurement activities within the plant. The data
would be used for correlation to modeled ambient contributions from fugitive
and point sources within the plant. The correlation of results would pro-
vide a check of fugitive (and point source) emission rate determinations.
The results of area source emissions developed in the Chloride Metals emis-
sions test were extrapolated to the total smelter emissions. This approach
neglected other potential sources. Ambient results would be an aid in the
identification of the contribution of other sources as well as a check on
the accuracy of measured emissions. An ambient monitoring network was not
included in this study because the location of the Chloride Metals plant was
not amenable to the necessary array of samplers.
2.4.2 Isokinetic Sample Collection
Due to schedule and budgetary constraints, fugitive particulate samples
were not collected isokinetically during exposure profiling testing at
Chloride Metals. By not sampling isokinetically, it is possible that re-
sults are not representative of actual conditions. The concept of isoki-
netic sampling is applicable to fugitive particulate measurements in the
same manner as applied to point source emissions testing (EPA Reference
Methods 5 and 17).
Concerning the results at Chloride Metals, samples were typically
collected at about 30 percent of isokinetic. At subisokientic sampling
rates, the collected particulate sample may be biased towards the larger
particulate, but the extent of the bias is dependent upon the particle size
distribution. However, measured particulate mass mean diameters at Chloride
Metals were typically less than 0.5 microns. The expected bias effect of
subisokinetic sampling for fugitive particulate emissions at Chloride Metals
is minimal.
2-41
-------
2.4.3 Smelter Building Concentration Sampling
During the study at Chloride Metals, airborne particulate samples were
collected in the smelter building to evaluate the source of arsenic, lead,
and cadmium emissions. Samples were typically collected for 4 to 6 hours at
a rate of 4.9 liters per minute for a total sample volume of about 1.5 cubic
meters. All analyses for arsenic and cadmium were below analytical detec-
tion limits (0.003 ppm for arsenic and 0.002 ppm for cadmium), and most
analyses for lead were below the analytical detection limit (0.1 ppm). If
similar samples are collected in future programs, the total sample volume
should be increased significantly (in excess of 5 cubic meters) to allow for
accurate measurement of arsenic, lead, and cadmium concentrations.
2-42
-------
SECTION 3
PROCESS DESCRIPTION
3.1 GENERAL PLANT INFORMATION
Chloride Metals is a small to medium sized secondary lead smelter
located in Tampa, Florida. This plant was selected as a fugitive emissions
test site because it is considered to be representative of a typical secon-
dary lead smelter. The operational practices employed, the raw materials
used, and the products produced are all representative of the secondary lead
industry as a whole. The plant operates 5 days per week and has a rated
lead production capacity of 12,000 tons of lead per year. During the test
period, the plant operated only one of its two blast furnaces. Conse-
quently, lead production at the plant was 50 percent of capacity. The major
lead bearing raw materials are recycled automobile batteries and battery
plant scrap. The major finished product is a hard lead alloy with an
arsenic content of 0.17 weight percent. Hard lead production typically
requires the addition of alloying agents such as antimony, tin, and arsenic.
The arsenic content of the blast furnace metal is increased in the refining
process by the addition of a high arsenic containing master alloy (10 per-
cent arsenic by weight).
The physical layout of the plant is shown in Figure 10. The smelting
plant consists of a battery breaking facility, an open raw materials storage
area, an open-air smelter building that houses a blast furnace, three 50-ton
refining kettles, one 20-ton refining kettle, a lead casting machine, a
building extension that houses an inoperative blast furnace and 55-gallon
drums containing refining drosses, an open blast furnace slag storage area,
and an open refining dross storage area. Also associated with the process
activities at the plant are one process baghouse for controlling the
3-1
-------
Columbia Street
N
A
Coke
Storage
Lead
Oxide
Plant
n
BATTERY
BREAKING
ARPA
Loading
RAW MATERIALS
STORAGE
AREA
CO
Q.
O
Sttorage
i
• i
L.J
Ingot
Build
Blast
Furnace
BUILDING
Refining
Kettles
I
i
i
rig
I
I
Brec
Roor
U-
Coolers
Process
Baahouse
Office
Tap
Baghouse
Sewage
Treatmen :
o
£
55
BLAST SLAG
'STORAGE
AREA
Machine Shop
Metals
Engng
A Scrubber
-**- DROSS STORAGE
E
t/-M
Raleigh Street
Figure 10. Plot plan of Chloride Metals plant in Tampa, Florida
illustrating the five fugitive sampling areas.
(Scale: 1 inch equals approximately 50 feet)
3-2
-------
offgases from the blast furnace and the slag tap baghouse for control of
emissions from the lead well and furnace slag tapping. A Kinpactor venturi
scrubber controls the combined offgases from the refining kettle ventilation
system. An enclosed screw conveyor is used for the direct recycle of bag-
house dust to the blast furnace.
Area fugitive emissions are controlled at the plant by the use of
automatic sprinkler systems coupled with paving of the smelter grounds. The
sprinkler systems are controlled by an automatic timer that turns the water
on every 15 to 20 minutes. Approximately 1,500 gallons of fresh water are
sprayed daily. Water is applied manually in the smelter building (200
gallons/day) and in the battery breaking area (600 gallons/day), in addition
to the sprinkler system at the battery breaking area.
In addition to smelter operations, this plant manufactures lead oxide
and operates a rolling mill where lead product fabrication occurs. Both of
these activities as well as vehicular traffic from a nearby highway contri-
bute to lead levels measured near the plant. The only potential source of
arsenic emissions in the area is a coal-fired power plant located roughly
one-half mile southwest of the smelter.
3.2 PROCESS INFORMATION
A detailed process description can be found in the NSS contractor trip
report dated November 4, 1983 (Appendix K). In general, materials flow
through the plant from north to south (refer to the plot plan, Figure 10).
Scrap batteries are brought to the north end of the plant by local scrap
dealers. The batteries are unloaded and transported to the battery saw
where they are separated from the cases, and the lead contents are dumped
into the raw material storage area. The battery covers are crushed, and a
flotation separator system is used to separate plastic, lead terminals, lead
oxide paste, and rubber. The lead scrap is then transported from the raw
material storage area to the charge preparation area inside the smelter
building. Feed materials are manually shoveled into a skip hoist bucket and
3-3
-------
are subsequently charged to the furnace. The furnace feed consists of
battery group material, battery paste, refining kettle drosses, battery
plant scrap, lead-bearing water treatment sludge, coke, limestone, scrap
iron, rerun blast furnace slag, and sand. These materials are fed to the
furnace approximately 8 to 12 times per hour. In addition to the prepared
feed materials, dust captured by the process baghouse is automatically screw
conveyed back to the blast furnace on a continuous basis. Molten lead is
tapped continuously from the blast furnace into one of four lead button
molds. Slag is intermittently tapped approximately twice an hour from the
opposite side of the furnace. The slag is allowed to cool and is trans-
ported to an open area beyond the southeast corner of the smelter building.
It is eventually either disposed of in an approved landfill or rerun to the
blast furnace. Rerun blast furnace slag is manually selected based on a
visual assessment of its silica content. High silica content slag is
charged back to the blast furnace primarily as a flux material serving as a
lead-bearing substitute for sand.
Blast furnace lead is alloyed in one of four gas-fired refining ket-
tles. The two kettles (one 50-ton and one 20-ton) at the northwest corner
of the smelter building are used for hard lead refining only. Hard lead
refining consists of an initial light dross removal step followed by removal
of other drosses and/or addition of alloying agents, depending on the speci-
fications of the product metal. The light dross consists mainly of lead
oxide and appears as a dusty black powder when cooled. Other drosses poten-
tially removed in hard lead refining are copper dross and tin dross. Anti-
mony and arsenic are among the alloying agents commonly added to the kettles
in the production of hard lead. The hard lead product is cast into ingots
using the casting machine.
The two 50-ton kettles at the southwest corner of the smelter building
are used for soft lead refining. The production of soft lead typically
involves five refining steps. The temperature of the molten lead is in-
creased during the first four steps and then is decreased for the final
step. In the first step, a "light dross" consisting primarily of lead oxide
3-4
-------
and other impurities forms at the surface of the molten lead bath as the
refining kettle is heated. The light dross, which appears as a very dusty
black powder, is manually skimmed off the surface and placed in a hopper.
This light dross will eventually be recharged to the furnace. The second
step involves the addition of sulfur to cause a copper containing dross to
form. The copper dross is manually skimmed off the surface and stored in a
drum. In the third refining step, the molten lead temperature is increased
until a yellow tin-containing dross forms. Both the copper and tin drosses
generated during these refining steps are charged to the furnace when high
copper and tin specifications are required in the blast furnace metal. The
fourth step involves the addition of sodium nitrate under conditions of
increasing temperature to form an antimony dross. The antimony dross is
skimmed off the surface and placed in a hopper. This dross material is
stored in 3-sided open air bins located south of the smelter building. This
dross is combined with other furnace charge materials when a high antimony
blast furnace metal is desired. In the final refining step, the temperature
of the molten lead is reduced and sulfur or caustic is added to the kettle
to provide a final cleaning. Upon completion of this final step, the now-
soft lead is cast into ingots.
3.3 PRODUCTION
Daily production data were provided by the plant for the period between
February 20 and March 21, 1984. This represents the 11 operating days
before sampling began as well as the 12 days when sampling actually
occurred. These data are presented in Appendix J.
The general trends in blast furnace metal production for this period
are presented in Figure 11. As seen in the figure, the blast furnace lead
production ranged from 13-42 tons per day during periods in which the fur-
nace was operating. For the period from February 20 to March 21, the
average daily furnace production was 33.5 tons per day (excluding weekends
and the 2 days that the furnace was not operated). During actual sampling
3-5
-------
40
30
u>
I
TJ
O)
U
3
_J
V)
10 »t-
c
O
20
10
190
»
I I I I i
I I I I I
M T W T F
2/20 - 2/24
M T W T F
2/27 - 3/2
I I I I i
M T W T F
I I I I
I I I
3/5 - 3/9
M T W T F
3/12 - 3/16
M T W
3/19 - 3/21
Figure 11. Blast furnace lead production February 20 through March 21, 1984. Daily lead
production (tons) prior to and during the testing period.
-------
RADIAN
periods and when the furnace was operating, the average daily lead produc-
tion was approximately 37 tons per day.
Except for the 2 days in which the blast furnace was not operated,
testing at this facility was conducted during periods of typical lead pro-
duction.
The daily rates of refined metal production are presented in Figure 12.
A total of 664 tons of refined metal were either cast into ingots or blocks
between February 20 and March 21, 1984. Approximately 39 percent of this
was soft lead while the remaining 61 percent was one of several hard lead
alloys. A very similar ratio of hard to soft lead was produced during the
actual testing period (March 6 - March 21).
Smelter operations and associated activities were monitored during the
entire test program. Several short-term episodes of equipment failure
occurred (i.e., battery breaking saw, case crusher, skip hoist bucket, screw
conveyor, etc.); however, these were considered representative of smelter
operation in general. None of these failures resulted in visible fugitive
emissions. During the early stages of testing, the water circulation jacket
on the blast furnace ruptured and consequently resulted in two days' loss in
lead production (3/9/84 and 3/12/84). According to plant personnel, this
incident represented a major equipment failure and should not be considered -
representative of typical smelter operations. Despite the inoperable fur-
nace, the area being tested (the road area between the smelter building and
the ingot storage building) was judged to demonstrate typical vehicle acti-
vities during the 6-8 hour sampling period.
Since the smelting facility operates 24 hours per day, and the sample
collection took place only during daytime hours, it was determined that
certain activities were not represented during the 6—8 hour sampling
periods. It was observed that some of the potential dust generating activi-
ties are routinely performed during night shifts. Activities such as
loading of slag into dumpsters and the dumping of dross materials were not
3-7
-------
m
b
70
OJ
00
•a
a>
o
3
-a
o
a.
-a
ID
oc.
\n
c
O
60
50
40
30
20
10
I I I
I I
M T W T F
2/20 - 2/24
M T W T F
2/27 - 3/2
I
I I I
M T W T F
3/5 - 3/9
T W T F
3/12 - 3/16
M T W
3/19 - 3/21
Figure 12. Refined metal production February 20 through March 21, 1984. Daily refined
lead production (tons) prior to and during the testing period.
-------
observed during actual testing periods. However, it is assumed that the
majority of the activities associated with smelter operation were observed
during the sampling periods, and the fugitive emissions measured in this
test program represent typical emissions at this smelter.
3.4 AREA FUGITIVE EMISSIONS AND VEHICLE ACTIVITY PATTERNS
Five potential area sources of arsenic emissions were identified at the
smelter. These are the battery breaking area, the raw material storage
area, the smelter building, the slag and dross storage area, and the intra-
plant vehicle roadways. The major area fugitive control technique practiced
at the smelter is wet suppression. This technique can be practiced effec-
tively at this location since all of the smelter grounds are paved. An
automatic sprinkler system is used to wet the battery breaking area, the raw
material storage area, the dross storage area, and the access roadway north
of the process baghouse and west of the ingot storage building.
Fugitive particulate emission rates are influenced by the amount of
wind, rain, and overall vehicular activity. Based on test results obtained
from fugitive studies on other types of outdoor storage, two-thirds of the
fugitive particulate emissions may be attributed to loading and unloading of
storage piles or reentrainment from vehicular activity.
Each of the factors influencing fugitive emissions were monitored
during the entire test period. Vehicle activities were observed, and gene-
ral traffic patterns associated with smelter operation were characterized.
Intraplant vehicle movement is primarily associated with one of four activi-
ties. These are battery breaking, blast furnace operation, refining activi-
ties, and lead casting.
3.4.1 Blast Furnace Operation
The majority of the vehicle activity at the plant is directly asso-
ciated with the operation of the blast furnace. These general activity
3-9
-------
patterns are listed in Table 12 and illustrated in Figure 13. Much of the
activity associated with blast furnace operation is due to front end loaders
supplying raw materials to the charge preparation area (Pattern A). This
activity occurs every 45 minutes during typical furnace operations. Addi-
tional activity is due to forklifts removing lead products and slag from the
furnace area (Patterns G and H). Slag buttons are taken to the slag pile,
and blast furnace buttons are transported to storage at one and a half hour
intervals.
3.4.2 Battery Breaking
The general vehicle activity patterns occurring in the battery breaking
area are listed in Table 13 and illustrated in Figure 14. These activities
consist of trucks delivering junk batteries to be recycled (Pattern L) and
forklifts supplying pallets of batteries to the sawing operation (Pat-
tern J). Additional forklift activity in this area is attributed to removal
of empty pallets (Pattern K) and the transport of lead collected in the
float sink separation system to the area material storage area (Pattern M).
During battery breaking operations, activity patterns J and K (supplying
batteries and removing pallets) represent the majority of the vehicle acti-
vities. These pathways are used every 10-15 minutes for the duration of the
battery breaking campaign. Junk battery deliveries (Pattern L) were very
infrequent during the sampling period. On the average, only 3 to 4 small
trucks delivered junk batteries each day.
3.4.3 Lead Casting
The general activity patterns associated with casting lead are listed
in Table 14 and illustrated in Figure 15. Transporting stacks of lead
ingots from the end of the casting line to the ingot storage building
(Pattern N) constitutes the single most frequent vehicle activity associated
with lead casting. During periods in which lead is being cast, this pathway
is used continuously. On the average, forklifts transport a stack of lead
every 4 minutes. Depending on the amount of available storage area in the
3-10
-------
TABLE 12. GENERAL VEHICLE ACTIVITY PATTERNS ASSOCIATED
WITH BLAST FURNACE OPERATION
Vehicle
Pattern
(Figure 13)
Description of Vehicle Activity
Pathway
Usage/hr
A Front end loader transporting lead bearing raw 1.33
materials from storage area to charge prepara-
tion area
B Front end loader transporting coke from storage 0.67
area to charge preparation area
C Front end loader transporting scrap iron from 0.50
storage area to charge preparation area
D* Forklift transporting drums containing battery ND
plant scrap to charge preparation area
E* Forklift taking empty drums to storage ND
F Forklift removing slag pot from the slag tap hood 0.67
and transporting to the corner of the smelter
building where the slag is allowed to cool
G Forklift transporting blast furnace slag to 0.67
storage pile
H Forklift transporting lead buttons from the 0.67
lead well to the blast metal storage area
I Miscellaneous front end loader activity trans- ND
porting limestone and sand to the charge
preparation area
*These vehicle patterns are only observed when battery plant scrap is
charged to the furnace.
ND = Not Determined.
3-11
-------
Figure 13. Major vehicle activity patterns associated with blast
furnace operation. Refer to Table 11 for descriptions.
3-12
-------
TABLE 13. GENERAL VEHICLE ACTIVITY PATTERNS ASSOCIATED WITH THE
BATTERY BREAKING OPERATIONS
Vehicle
Pattern Pathway
(Figure 14) Description of Vehicle Activity Usage/hr
J Forklift transporting pallets of junk batteries 5
from storage area to battery saw
K Forklift taking empty pallets to storage 5
L Small trucks carrying scrap batteries 0.5
M Forklift transporting lead scrap collected in ND
the float-sink operation to the raw material
storage area
ND = Not Determined.
3-13
-------
N
A
i
, ^^
p
is
m
•
+
rt
.
9T*
a
,4C
L r>
oooooa
D
Figure 14. Major vehicle activity patterns associated with the
battery breaking operation. Refer to Table 12 for
descriptions.
3-14
-------
TABLE 14. GENERAL VEHICLE ACTIVITY PATTERNS ASSOCIATED WITH
INGOT CASTING
Vehicle
Pattern Pathway
(Figure 15) Description of Vehicle Activity Usage/hr
N Forklift transporting stacks of lead ingots from 0.9
casting area to scales and then to storage area
0 Forklift transporting stacks of lead ingots from 0.9
the end of the casting line to temporary
storage
P Forklifts transporting soft lead blocks from 1
refining kettles to the lead oxide building
Q Forklifts transporting stacks of lead ingots 1
from the casting area to the dock on the north
side of the lead oxide building
3-15
-------
Figure 15. Major vehicle activity patterns associated with the
casting or pouring of refined metal. Refer to
Table 13 for descriptions.
3-16
-------
ingot building, the lead may be temporarily stored in the smelter building
(Pattern 0). Soft lead is poured into large blocks directly from the re-
fining kettle. These soft lead blocks are transported to the lead oxide
building where they are stored (Pattern P).
3.4.4 Refining Activities
The general activity patterns associated wtih the refining process are
listed in Table 15 and illustrated in Figure 16. Initially, forklifts
transport blast furnace metal from the storage area to the kettles for
melting (Pattern R). The largest kettles have a 50-ton capacity; and,
consequently, 50 trips are eventually required to fill the kettles. This is
most likely the largest single vehicle activity associated with the refining
process. Forklifts also carry alloying additives from the ingot storage
building to the refining area (Pattern S), and transport refining drosses to
storage area (Pattern T). Although several tons of refinery drosses were
generated during the test period, these were primarily removed from the
kettles and stored during second and third shifts. Consequently, this
vehicle activity (Pattern T) was at a minimum during actual sampling
periods.
3.5 TEST AREAS
Five areas which were identified as potential sources of fugitive
arsenic emissions were tested using a modified exposure profiling sampling
technique. Figure 10 shows the location of the five test areas by number.
These are the battery breaking area (Area 1), the raw material storage area
(Area 2), the smelter building (Area 3), the slag and dross storage area
(Area 4), and the intraplant vehicle roadway (Area 5). A brief physical
description of each of these areas along with a summary of the process
related activities occurring during the sampling periods will be described
in the following sections.
3-17
-------
RADIAN
TABLE 15. GENERAL VEHICLE ACTIVITY PATTERNS ASSOCIATED WITH
REFINING KETTLE OPERATIONS
Vehicle
Pattern
(Figure 16) Description of Vehicle Activity
R Forklift transporting blast furnace metal from storage area
to refining kettle
S Forklift transporting alloying additives from ingot storage
building to kettles
T Forklift transporting refining kettle drosses to storage
3-18
-------
RADIAN
D
\
-v
p
C
<:
N
l I
Figure 16. Major vehicle activity patterns associated with
refining operations. Refer to Table 14 for
descriptions.
3-19
-------
3.5.1 Battery Breaking Area (Area 1)
The battery breaking area is located at the north end of the plant and
consists of a shed-like structure surrounded by open pavement where pallets
of junk batteries are stored. The battery saw facility consists of an open-
air building containing an automatic feed conveyor system and a slow speed
saw for removing the covers from junk batteries. The facility also houses a
hammer mill for breaking battery cases and a float/sink separation system
for separating plastic, lead terminals, lead oxide paste, and rubber.
Sprinkler heads mounted on the west side of the rolling mill building
serve to keep the battery storage area and the area surrounding the battery
cutting facility wet. Manual wetting of the saw equipment and pavement area
underneath the saw is practiced routinely.
Table 16 presents a summary of the test conditions and the process
related activities recorded during periods in which sampling was conducted
in the battery breaking area. Fugitive sampling in this area was performed
in conjunction with sampling in the raw material storage area and the smel-
ter building. Samples were collected for 6-8 hours each day for 4 consecu-
tive days.
As seen in Table 16, with the exception of March 13, the vehicle
activity counts recorded and the vehicle patterns used were very similar
during each of the 3-4 days. Day-to-day differences in test conditions were
primarily due to variations in wind direction.
The sprinkler system in the battery breaking area was not operated
during the sampling periods due to the positioning of the sampling equip-
ment. Under normal conditions, the sprinklers in this area are operated on
a cyclic basis spraying 3 out of every 15 minutes. Manual wetting of the
saw and equipment was observed. Manual wetting of the pavement surrounding
the building was not observed. The tests conducted under these conditions
3-20
-------
Date
TABLE 16. BATTERY BREAKING AREA. SUMMARY OF GENERAL CONDITIONS AND PROCESS-RELATED
ACTIVITIES DURING THE SAMPLING PERIOD
Wind
Direction
Vehicle*
Patterns Vehicle
that Affect Activity
the Test Area Counts
Number of
Batteries Conditions
Broken Wet/Dry
General Observations
N)
3/13
3/14
3/15
3/16
W-SW
NE
E-NE
E
J, K, L
J, K, L
J, K, L
J, K, L
13
1,440
Dry
45
54
47
2,162
2,287
609
Dry
Dry
Dry
-Initial electrical problems
resulted in intermittent
operation of sampling
equipment
-Unloading of coke from a
train car represented a
potential source of upwind
particulate matter
-A localized whirlwind
stirred up substantial
amount of dust at one point
during the sampling period
aVehicle patterns are illustrated in Figure 14.
J = Forklifts transporting pallets of junk batteries from storage area to battery saw.
K = Forklifts transporting empty pallets to storage.
L = Small trucks delivering loads of scrap batteries.
The automatic sprinkler system was not operated in this area during the sampling period due to the
positioning of the samplers. Manual wetting was observed near the battery saw approximately once
per day.
-------
are considered "dry" since the automatic suppression system was not in use
and no area wetting was observed.
3.5.2 Raw Material Storage Area
The raw material storage area is located just south of the battery
breaking facility and is used for storing battery group materials and other
lead bearing raw materials. This area consists of an uncovered, three-sided
concrete structure with walls approximately 8 feet high. The dimensions of
the raw material storage area are approximately 85 feet by 60 feet. The
south side of this area is open to the smelter building. Battery plates and
lead oxide paste from scrap batteries constitute the majority of the stored
materials. In addition, miscellaneous lead scrap is periodically dumped in
this area by local scrap dealers. The battery materials are generally kept
wet by battery acid from the broken batteries and by water used in the
cutting operation. In addition, wastewaters from the cutting operation are
continuously pumped into this area. The material piles are located so that
sludge containing waters will slowly filter through the scrap battery
plates. This process aids in increasing the lead recovery from the waste-
water and simultaneously provides a partial means of wet suppression for the
storage piles. Front end loaders are used to transport the lead bearing raw
materials from the storage piles into the smelter building where the blast
furnace charge is prepared.
Table 17 presents a summary of the test conditions and the process
related activities recorded during periods in which sampling was conducted
in the raw material storage area. Samples were collected for 6-8 hours each
day for 5 test days. This testing was performed in conjunction with sam-
pling in the battery breaking area, the smelter building, and the roadway.
As seen in Table 17, vehicle activity counts are relatively low as
compared to those recorded for the battery breaking area (Table 16). Front
end loaders remove a load of battery group material from the raw material
storage area approximately every 45 minutes. Additional loader activity in
3-22
-------
TABLE 17. RAW MATERIALS STORAGE AREA. SUMMARY OF GENERAL CONDITIONS AND PROCESS-RELATED
ACTIVITIES DURING THE SAMPLING PERIOD
S3
U>
Date
3/13
3/15
3/16
3/19
3/20
Wind Process8
Direction Activities
W-SW BF
BB
C
RF
E-NE BF
BB
C
RF
E BF
BB
C
SE BF
BB
C
RF
S BF
RF
C
Vehicle Battery
Affect Plates Fed to
Test Area Furnace (Ibs) Conditions0 General Observations
7 19,890 Dry — Initial electrical problems
resulted in intermittent
operation of sampling
equipment
— Unloading of coke from a
train car represented a
potential source of upwind
particulate matter
10 73,620 Dry
8 65,600 Dry
14 68,004 Dry — Front end loader had a flat
tire and consequently
charge materials consisted
of battery plant scrap from
55-gallon drums
d 28,044 Dry — Battery breaking facility
was not operating due to
electrical problems
aProcess activities during the actual sampling period.
BF = blast furnace operation, BB = battery breaking, C = casting, RF = refining.
Data provided by the facility.
cThe automatic sprinkler system was not operated in this area during the sampling period due to the
positioning of the samplers. Manual rinsing of the front end loader was performed each time the
loader left the raw materials storage area.
Data not available.
-------
RADIAN
the storage area relates to the battery breaking operations. The battery
contents dumped from the battery breaking facility are cleaned up periodi-
cally during the day and redistributed among the existing group storage
piles.
The sprinkler system which wets the raw material storage area was not
operated due to the positioning of the sampling equipment. Under normal
conditions at this plant, this area is intermittently wetted by the automa-
tic sprinkler system. This practice also aids in keeping the raw material
storage area wet.
3.5.3 Smelter Building
The dimensions of the smelter building are approximately 90 feet by 145
feet and the roof is 20 to 30 feet high. It is completely open on the
north, east, and south sides and is enclosed on the west except for two
doorways. The building houses a blast furnace, four refining kettles, and a
casting machine. An extension to the smelter building at the southeast
houses a currently inoperative furnace and several 55-gallon drums con-
taining refining drosses.
Table 18 presents a summary of the test conditions and the process
related activities recorded during periods in which emissions from the
smelter building were sampled. Fugitive samples were collected on 4 conse-
cutive days using the high volume pole sampling equipment. Additional low
volume, site specific fugitive sampling was conducted within the smelter
building near the refining kettles, and near the lead and slag tapping wells
of the blast furnace. These samples were collected in order to qualita-
tively assess the contribution of the process fugitive emissions from these
sources to the total area fugitive emissions from the smelter building.
Table 19 summarizes the general activities represented during each low
volume sampling period.
3-24
-------
TABLE 18. SMELTER BUILDING. SUMMARY OF GENERAL CONDITIONS AND PROCESS-RELATED ACTIVITIES
DURING THE SAMPLING PERIOD
Date
Wind
Direction
Process8
Activities
Vehicle6
Patterns
that Affect
the Test Area
Vehicle
Activity
Counts
Lead
Produced
(Tons)
Lead Pumped
from Kettles
(Tons)
Conditions0
Wet/Dry
General Observations
I
3/13 W-SW
3/14 HE
3/15 E-NE
BF A.B.C.F.G.H.I
BB
C N
RF
29
BF A.B.C.F.G.H.I 91
BB
C H
RF
BF A,B,C.F,G,H,I 100
38
37
63
19.5
60
Dry
Dry
Dry
—Initial electrical problems
resulted in intermittent
—Unloading of coke from a train
car represented a potential
source of upwind particulate
matter
—Skip hoist conveyor system broke
down, no furnace charging for
2.5 hours
—Substantial floor vetting in the
smelter area while the skip
hoist was inoperable
U>
N)
t-n
3/16 E
3/19 S-SE
3/20 a
C
RF
BF
BB
C
BF
BB
C
RF
BF
RF
C
N
A.B.C.F.G.H.I 81 . 36 41.5
N
A.B.C.D.F.G.B.I 91 34 20
O.Q
A.B.C.F.G.H.I d 35 42
H
Dry
Dry — Front end loader had a flat tire
and consequently charge materials
consisted of battery plant scrap
from 55-gallon drums
Dry — Sb and As were added to Kettle *4
which was directly upwind of
Sampler 5
— Battery breaking facility was not
operating due to electrical prob-
lems
8Process activities during the actual sampling period.
BF - blast furnace operation, BB " battery breaking. C • casting, RF - refining.
Vehicle patterns are illustrated in Figure 13.
A - Front end loaders transporting lead bearing raw materials from storage area to charge preparation area.
B - Front end loaders transporting coke from storage area to charge preparation area.
C - Front end loaders transporting scrap iron from storage area to charge preparation area.
D • Forklift removing the slag pot from the slag tap hood and transporting to the corner of the smelter building where the slag ia allowed to cool.
G - Forklift transporting blast furnace slag to the storage pile.
H - Forklift transporting lead buttons from the lead tapping area to the blast metal storage area.
I - Miscellaneous front end loader activity transporting limestone and sand to the charge preparation area.
N - Forklift transporting stacks of lead ingots from the casting area to scales and then to storage.
"•The smelter building floor is completely hosed down once per shift.
Data not available.
-------
TABLE 19. SUMMARY OF GENERAL ACTIVITIES DURING SITE-SPECIFIC LOW VOLUME SAMPLING PERIODS
INSIDE THE SMELTER BUILDING
Dates
U)
I
KJ
3/7
3/8
3/9
3/12
3/13
3/16
3/19
3/20
3/21
Sampler Location
General Activities During Sampling Periods
0
5
Refining kettle #2
Refining kettle #2
Refining kettle #3 and #4
Refining kettle #3 and #4
Refining kettle #3 and #4
Blast furnace*; lead and
slag tap area
Blast furnace*; lead and
slag tap area
Refining kettle #1 and #4
Blast furnace*; lead and
slag tap area
— Soft lead kettle (dressing began after sampling)
— Caustic skim (last clean-up)
— Pouring of soft lead
— Forklift activity
— Casting hard lead (pumped from kettle #3)
— Forklift activity
— Melting lead in both kettles
Dressing kettle #3
Casting hard lead (pumped from kettle
Forklift activity
Melting lead in #4
Produced 10.5 tons of Pb
Produced 4.5 buttons of slag
Produced 10 tons of Pb
Produced 4.5 buttons of slag
Mixing kettle #1
Addition of arsenic master alloy (684 Ibs of
10% As)
Produced 8.5 tons of Pb
Produced 4.5 buttons of slag
*Lead and slag production estimates are based on 7 hour sampling periods and daily blast furnace
production data reported by the plant.
-------
Process activities during the four high volume sampling periods were
very similar, as were the related vehicle activity counts. Production rates
of blast furnace lead were also comparable. Day-to-day differences in test
conditions were primarily due to variations in wind direction.
As seen by the vehicle counts recorded in Table 18, the smelter
building supports the majority of traffic recorded at this smelter. The
smelter building is the central location for all process related activities
and potentially represents the major source of fugitive emissions. All of
the furnace feed materials are transported into the smelter building by
front end loaders. Almost all of the furnace products, as well as the
refining and casting products, are handled by forklifts. Table 18 includes
a listing of the vehicle patterns observed during each sampling period.
These patterns were discussed earlier and illustrated in Figures 13
through 16.
3.5.4 Slag and Dross Storage Area
Slag is tapped from the blast furnace into a crucible which is located
in the slag tap hood enclosure. After approximately three tappings (one and
one-half hours), the slag is removed from the hood and taken to the north-
east corner of the smelter building where it is allowed to cool for an
additional one and one-half hours. During this cooling period, visible
fumes are emitted from the slag. The crucible is then transported to the
slag pile where the glass-like aggregate is dumped. Initially, the slag
material fumes but later cools and breaks into smaller pieces. Little
activity was observed in this area during testing, and the slag pile re-
mained dry during all sample collections.
Two types of slag are generated at this smelter. These are (1) slag
which according to the EP toxicity test is not classified as hazardous
material and may be landfilled in a normal landfill, and (2) slag which by
the EP toxicity test is classified as hazardous and must be disposed of in
an approved landfill. The latter type of slag is generated when caustic
3-27
-------
drosses and sodium containing materials are charged to the furnace. During
the testing period, no hazardous slag was generated.
Little or no slag was rerun in the furnace during the test period. All
slag pile clean-up and loading into dumpsters occurred during the second and
third shifts; and, consequently, was not captured during the sampling
periods.
Dross materials removed from the refining kettles are finely divided
and dusty. Some drosses are stored in drums and covered with plastic.
Light dross, antimony slags, and caustic skims are all wetted and deposited
in the open 3-sided bins along the southern wall of the plant property.
Sprinkler heads mounted on the dross bins serve to keep the dross storage
area and the access pathway near the bins wet.
Table 20 presents a summary of the test conditions and the process
related activities recorded during periods in which emissions from this area
were sampled. As seen from this table, vehicle activity counts are rela-
tively low compared to those observed in the smelter building (Table 18) or
battery breaking area (Table 16). The majority of the vehicle activity
results from the storage of furnace products. It was observed that both
lead and slag were transported into this area every one and one-half hour.
Although refining activities were conducted, no dross dumping occurred
during the actual sampling periods.
The sprinkler system was operated during 2 of the 3 test days. The use
of the sprinkler on the last day was prohibited due to exposed electrical
connections which were needed to supply power to the sampling equipment.
Although no additional water was sprayed during this "dry" test, the area
immediately around the dross storage bins remained wet and muddy during the
entire sampling period. The slag pile and the area where blast furnace
metal is stored remained dry during each of the 3 test days.
3-28
-------
TABLE 20. SLAG AND DROSS STORAGE AREA. SUMMARY OF GENERAL CONDITIONS AND PROCESS-RELATED
ACTIVITIES DURING THE SAMPLING PERIOD
Wind
Date Direction
3/6 SB
Vehicle1*
Patterns
Process* that Affect
Activities the Test Area
BF H.G
BB
BF
Drosses
Vehicle Lead Removed
Activity Produced from Kettles Conditions0
Counts (Tons) (Tons) Wet/Dry General Observations
7 42 2.6 Wet — No drosses were dumped during
sampling period
— No slag was broken or loaded
into dumpsters during the
sampling period.
— Slag pile was dry/dross area wet
and muddy.
3/7
NW
u>
3/8
HE
BF
BB
C
BF
BB
C
H.G
41
6.4
G,E
12
41
2.7
Wet —No drosses were dumped during
the sampling period
—No slag was broken or loaded
into dumpsters during the
sampling period
—Slag pile was dry/dross area waa
wet and muddy.
Dry —No drosses were dumped during
the sampling period.
—No slag wae broken or loaded
into dumpsters during the
sampling period
—Vehicle activity associated with
blast furnace metal storage was
not captured due to sampler
position
—The automatic sprinkler system
waa not operated due to open
electrical connections required
to operate the samplers.
However, the test area waa atill
muddy during the entire sampling
period. The slag pile waa dry.
aProcess activities during the actual sampling period.
BF - blast furnace operation, BB - battery breaking, C - casting, BF > refining.
Vehicle patterns are illustrated in Figure 13.
E - Forklift taking empty drums to storage.
G » Forklift transporting blast furnace slag to storage pile.
H - Forklift transporting lead buttons from the lead well to the blast metal storage area.
cWet conditions correspond to periods in which the automatic sprinkler system was operated in the dross storage area. These sprinklers did not wet
the slag pile or the area where the blast furnace metal is stored. Dry conditions indicate that the automatic sprinkler was not operated.
-------
3.5.5 Roadway (Major Vehicle Pathway)
The smelter access road is a paved area located between the ingot
storage building and the smelter building (Area 5 in Figure 10). In gene-
ral, the roadway appeared to be the cleanest area in the facility since
little dust was observed on the road surface. This roadway is primarily
used by forklifts to transport lead from the smelter building to storage or
to the rolling mill for fabrication. In addition, this roadway is used as
an access to the raw material storage area by local dealers who sell scrap
lead to the smelter.
Half circle impact sprinkler heads are mounted on the north side of the
process baghouse in order to reduce dust entrainment from vehicle traffic.
These sprinklers are operated on a timed cycle operating 3 out of every 15
minutes. Similar manually operated impact sprinkler heads are located on
the southwest corner of the rolling mill and the west side of the ingot
storage building.
Table 21 presents a summary of the test conditions and the process
related activities recorded during the 6 days in which emissions from the
area were sampled.
The major vehicle roadway was sampled under both wet and dry conditions
in order to assess the fugitive emission reduction potential of wet suppres-
sion. The vehicle activity counts recorded in Table 21 are primarily in-
fluenced by the casting operations. After casting, if the lead is taken to
the ingot storage building (as was on 3/7), the vehicle activity counts in
the roadway area are high. If the lead ingots are taken to the loading dock
north of the lead oxide building or temporarily stored in the smelter
building, the vehicle activity in the roadway is not influenced by the
casting operation.
The blast furnace was not operated during 2 of the 6 test days (3/9 and
3/12) due to a broken water jacket in the furnace shaft. The inoperable
3-30
-------
TABLE 21. ROADWAY. SUMMARY OF GENERAL CONDITIONS AND PROCESS-RELATED ACTIVITIES
DURING THE SAMPLING PERIOD
Date
Hind
Direction
Process*
Activities
Vehicle11
Patterns
that affect
the Test Area
Vehicle
Activity
Counts
Refined
Metal
Produced
(Tons)
Conditions'
Met/Dry
General Observations
|
3/7
3/8
3/9
NH
HW
BF
BB
C
BF
C
BB
C
73
13
17
25
44
18
Wet
Wet
Wet
OJ
3/12
14
Dry
3/19
SB
BF
BB
C
RF
17
20
Dry
—Unloading of lead scrap from a truck created a
lot of dust
—Soft lead vas cast and taken directly to the lead
oxide building
—Blast furnace was not operating due to a broken
vater jacket
—Maintenance personnel were using a jackhammer in
order to remove lead and slag from the blast
furnace crucible. This furnace repair activity
created dust which was not typical of normal
operation
—The ingots produced from the casting machine were
taken to loading dock at the lead oxide building.
Consequently, the road access area showed less
vehicle activity than typically expected during
caating operations
—Vehicle activity was due .to maintenance related
activities
—Blast furnace was undergoing repair
—Extensive manual wetting in the raw material
storage area
—Open burning upwind of the plant resulted in
deposition of visible paniculate matter in the
sampling srea
—Vehicle activity was due to maintenance related
activities
—Ingots from the casting operation were
temporarily stored inside the smelter building.
Consequently, the road access area showed less
vehicle activity than typically expected during
casting operations
(Continued)
-------
TABLE 21. (Continued)
W
I
K>
Date
3/20
Vehicle1* Refined
Patterns Vehicle Metal
Wind Process8 that Affect Activity Produced
Direction Activities the Teat Area Counts (Tons)
SE BF 11 42
C P
RF
Conditions0
Wet/Dry General Observations
Dry — Soft lead was cast and taken directly to the
oxide building
lead
'Process activities during the actual sampling period.
BF - blast furnace operation, BB - battery breaking, C
casting, RF - refining.
''Vehicle patterns are illustrated in Figure 16.
D - Fork lifta transporting drums containing battery plant scrap to charge preparation area.
M - Forklifts transporting lead scrap collected in the float/sink operation to the raw material storage area.
N = Forklifts transporting stacks of lead ingots from the casting areas to scales and then to atorage.
0 • Forklifts transporting stacks of lead ingots from the end of the casting line to temporary storage inside the smelter building
P ** Forklifts transporting soft lead blocka from refining kettles to the lead oxide building.
Q * Forklifts tranaporting stacks of lead ingots from the casting area to the dock on the north side of the lead oxide building.
cWet conditions correspond to the cyclic operation of the sprinkler syste
automatic sprinkler was not operated.
ounted on the process baghouse. Dry conditions indicate that the
-------
blast furnace did not significantly affect the vehicle patterns in the
roadway area since no vehicle activity in this area is directly associated
with supplying feed materials-to or storing products from the blast furnace.
During the period that the blast furnace was down, crude lead was being
refined. The transportation of lead pigs was through the intrapIant road-
way.
3.5.6 Sanitary Baghouse and Scrubber Testing
The fugitive control devices used at this plant are a sanitary baghouse
to control emissions captured by the lead and slag tap hood and a wet
scrubber to control emissions captured by the refining kettle and dross
hoods. In general, these control devices have short stacks and discharge at
low temperatures. As a result of these conditions, stack emissions from the
process fugitive control devices have the potential to contribute to or
interfere with the smelter area fugitive measurements.
Three stack gas tests were performed on both the scrubber and the
sanitary baghouse outlets using EPA Method 108. The lead and slag tap
activities during each of the sanitary baghouse stack tests are listed in
Table 22. The refining kettle and dressing activities during each of the
scrubber outlet tests are listed in Table 23.
3-33
-------
TABLE 22. PROCESS CONDITIONS DURING THE SANITARY BAGHOUSE OUTLET TESTS
Buttons of Number of
Date Time Lead Tapped Slag Taps3
3/16 1038-1232 3
3/19 1352-1604 3
3/21 0954-1154 2.5
aThe number of times that the damper was opened on the slag tap hood
ventilation system. This was determined by the change in stack gas
velocity.
3-34
-------
TABLE 23. KETTLE ACTIVITIES DURING THE SCRUBBER OUTLET TESTS
Date
Kettle Activities
3/13 Kettle #1: Empty
Kettle #2: Mixing (forming antimony/arsenic dross)
Kettle #3: Holding molten lead - no activity
Kettle #4: Melting blast furnace metal
3/14 Kettle #1: Empty
Kettle #2: Holding molten lead - no activity
Kettle #3: Casting
Kettle #4: Holding molten lead - no activity
3/15 Kettle #1: Mixing (forming antimony/arsenic dross)
Kettle #2: Empty
Kettle #3: Empty
Kettle #4: Addition of arsenic master alloy
3-35
-------
SECTION 4
SAMPLING LOCATIONS
Four area sources, one contained source (the smelter building), and two
point sources were tested for arsenic, lead, cadmium, and particulate emis-
sions. The following sections describe the sources that were sampled.
4.1 AREA SOURCES
Following are descriptions of the four area sources tested. Figure 17
identifies the area sources and defines their dimensions.
4.1.1 Battery Breaking Area (Area 1)
Scrap batteries delivered to the plant are received in the battery
breaking area and are stored there on pallets until recycled. To recover
the scrap lead, scrap bearing parts, and sludges in the batteries, the
batteries are manually fed through a slow speed saw and cut in half. The
battery internals, the plates primarily, are manually dropped from the
battery cases and fall into the raw materials storage area. The cases are
then conveyed to a hammer mill, crushed, and fed to a flotation system to
separate lead parts, lead bearing sludges, plastic, and rubber. The lead
parts and sludges are fed to the blast furnace, the plastic is sold, and the
rubber is landfilled.
The battery breaking area is located on the north edge of the plant,
due north and adjacent to the charge storage area. For emission rate deter-
minations, the dimensions of the area are defined to be 92 feet by 121 feet.
On the eastern boundary is the rolling mill building which does contribute
lead fumes. Areas directly north and west of the battery breaking area are
4-1
-------
RADIAN
Columbia Street
N
w
«
o
I
o
BATTERY
BREAKING
N \I\AREA\
•Y/YfrM
RAW MATERIALS .>-
STORAGE/ A/
AREA
?
1
Natural Gas,
Storage
L1
I I Loading
\ £=4 DocK
Q
JJ
\\V\\
ROADWAY
\ \ \ \ -Pf0cess
\ \ \ \ Bagnouse
SMELTER
BUILDING
EXTENSION I I
////n
O
S
(O
Raleigh Street
AREA 1 92 ft. x 121 ft. = 11,132 ft*
AREA 2 59 ft. x 86 ft. = 5,074 ft2
AREA 3 144ft. x 78ft. = 11,232ft2
AREA 4 73ft. x 118 tt. = 8,614 ft2
AREA 5 48 ft. X 140 ft. = 6,720 ft*
Figure 17. Plot plan of Chloride Metals Secondary Lead Smelter
identifying area sources.
4-2
-------
open. The battery saw is located about midway on the southern edge of the
battery breaking area which is also the northern edge of the raw materials
storage area. The flotation system is to the west and adjacent to the
battery saw. The battery breaking area is seen in the picture in Figure 18.
Chloride Metals does operate a series of sequential sprinklers located
along the rolling mill building for wetting the battery breaking area.
During testing, the sprinklers were not operated to prevent the wetting of
samplers.
4.1.2 Raw Materials Storage Area (Area 2)
The raw materials storage area is located between the battery breaking
area and smelter building. An eight-foot high concrete wall bounds all but
about one-half of the boundary at the smelter building for front-end loader
access. Lead plates from the battery saw fall directly into the raw mate-
rials storage area. The plates and other lead bearing materials are occa-
sionally organized in the area using a front-end loader. Additionally, the
sludge containing waters from the flotation system are filtered through the
charge materials to remove and collect the lead bearing sludge.
The dimensions of the raw materials storage area are defined to be 59
feet by 86 feet. The northern boundary of the raw materials storage area is
the southern boundary of the battery breaking area. The southern boundary
of the raw materials storage area is at the edge of the smelter building.
On the eastern side of the raw materials storage area is the rolling mill
building. Figure 19 is a picture of the raw materials storage area.
4.1.3 Slag/Dross Storage Area (Area 4)
When slag is tapped, the slag kettle typically remains under the
hooding at the slag tap until the next tapping period to control emissions
from the hot slag containing kettle. When necessary to remove the full
kettle to prepare for the next slag tap, the kettle is taken by a forklift
4-3
-------
-.;-. -y. .•.•• = -. -
--
Figure 18. Battery breaking area looking southwest,
Raw materials storage area and smelter
building in background.
4-4
-------
Figure 19. Raw materials storage area looking north.
Battery breaking area behind concrete wall,
4-5
-------
RADIAN
to the slag storage area just southeast of the smelter building to cool
further and complete solidification of the slag. The solid slag is even-
tually dumped from the kettle into the slag storage area. As necessary,
slag is removed from the storage area by front-end loader and emptied into a
truck trailer for shipment and landfilling.
Drosses are stored adjacent to the slag storage area in bins east of
the smelter building and at the property line. Due to the proximity of the
slag storage area and dross storage bins, the two were sampled as a single
source, the slag/dross storage area. The dross storage bins are wetted by
sprinklers while the slag storage area is not.
The boundaries of the slag/dross storage area are not easily recogni-
zable, but the dimensions of the area were defined to be 73 feet by 118 feet
for emission rate calculations. The slag/dross storage area is seen in
Figure 20.
4.1.4 Roadway
The roadway, or major vehicle pathway, is a paved area located between
the smelter building and ingot building. Along the southern boundary is the
process baghouse and along the northern edge is the rolling mill building.
Automated sprinklers located on the process baghouse wet the roadway. To
evaluate the degree of control achievable by wetting of the roadway, the
area was tested wet and dry.
The dimensions of the roadway are defined to be 59 feet by 140 feet for
use in determining the area emission rates. The roadway viewing towards the
smelter building is shown in Figure 21.
4.2 SMELTER BUILDING
Smelting and refining activities are conducted in the smelter building.
Although the plant has two blast furnaces, only one is presently operating.
4-6
-------
•
Figure 20. Slag/dross storage area looking west.
Smelter building to the right in picture,
4-7
-------
Figure 21. Roadway looking west. Smelter building in background-
process baghouse on the left in picture.
4-8
-------
There are four refining kettles in the smelter building. The dimensions of
the smelter building are shown in Figure 22.
Flue gas, or process gas, emissions are vented through the process
baghouse which was not tested in this program. The slag tap and slag kettle
positioned at the slag tap are hooded and vented through the slag tap
baghouse. The lead well is also hooded and vented through the slag tap
baghouse. Slag tap baghouse emissions were tested.
The four refining kettles and dressing enclosures located directly
adjacent to each kettle are hooded and are vented through a high energy wet
scrubber, the Kinpactor scrubber. The Kinpactor scrubber was sampled.
The smelter building is the center of activities at the Chloride Metals
plant and would be assumed to be the primary source of emissions due to
vehicular movement and process operations. Smelter building emission rates
for arsenic, lead, cadmium, and particulate were determined but in addition
concentrations of arsenic, lead, and cadmium were determined in the smelter
building near the refining kettles and near the slag tap and lead well at
the blast furnace.
4.3 STATIONARY SOURCES
Two stationary sources, the slag tap baghouse and the Kinpactor scrub-
ber, were sampled at Chloride Metals.
4.3.1 Slag Tap Baghouse
Fugitive emissions collected by hooding systems on the slag tap and
lead well at the blast furnace are vented through the slag tap baghouse.
The baghouse is located just east of the smelter building. The exhaust fan
with a short vent stack is mounted on the side of the baghouse.
4-9
-------
I
M
O
Total Area = 1458 sq.ft.
1
12'
East
Effective Area
50%
196 sq. ft.
< 20 \'i ' »•
Effective
Area
50%
298 sq. ft.
« 3V >
Effective
Area
50%
149 sq. ft.
< 15'/j' »
Effective
Area
0%
Smelter Area
422 sq. ft.
• ?°' ».
Roadway
393 sq. ft.
m in Vi ' fe
126 sq. ft.
lO'/i'
West
2
12'
138 sq. ft.
23'
29'
JO
s
19.2'
68'
Figure 22. Smelter building dimensions and effective emission areas of openings.
-------
•e-
i
South
78'
North
a
»
o
Figure 22. (Continued)
-------
One 3-inch port in the vent stack was accessible from the top of the
baghouse. The inside diameter of the circular stack was 19 1/2 inches.
4.3.2 Kinpactor Scrubber
The refining kettles and dross bins are hooded and vented through a
high energy wet scrubber, the Kinpactor scrubber. Scrubber emissions are
vented through a stack mounted on top of the scrubber.
One 3-inch port was available for testing on the stack. The port had
no access and scaffolding had to be erected to reach the sampling port. The
diameter of the circular stack was 22 inches.
4-12
-------
SECTION 5
SAMPLING AND ANALYSIS
The following sections describe the sampling and analytical techniques
used to determine arsenic, lead, cadmium, and particulate emissions at the
Chloride Metals secondary lead smelter. Fugitive emission rates were deter-
mined for four area sources and the smelter building. Area sources are
defined to be open, unconfined definable process operations areas. Point
source emissions were determined on two stacks. Concentrations of arsenic,
lead, and cadmium were determined in the smelter building.
5.1 SAMPLING TECHNIQUES
Sampling techniques for determining fugitive particulate emission rates
and point source emissions are described in the following sections:
5.1.1 Fugitive Emissions Sampling Techniques
The following sections describe the fugitive emissions measurement
techniques.
5.1.1.1 Exposure Profiling Technique—
Area source fugitive emissions were measured using the exposure pro-
filing technique.
Sampling Approach—The exposure profiling technique was used to measure fugitive
particulate emissions from the following area sources:
• slag/dross storage,
• raw materials storage,
5-1
-------
caopoaimoi
• battery breaking area, and
• roadway
The exposure profiling method was originally developed by MRI (1) to measure
particulate emissions from specific open sources. As developed, the method
utilized the isokinetic profiling concept which is the basis for conven-
tional source testing. For measurement of nonbuoyant fugitive emissions,
samplers are distributed over a vertical plane positioned just downwind from
the source to measure particulate concentrations in the plume from that
source. Sample intakes are pointed into the wind, and the sampling velocity
is adjusted to match the local wind speed, as monitored by an anemometer in
close proximity to the source and samplers. A vertical line grid of sam-
plers is sufficient for measurement of emissions from line or moving point
sources while a two-dimensional array (vertical and horizontal) of samplers
is required for measurement of area source emissions.
During this program, area sources were tested which necessitated the
use of multiple samplers to measure particulate concentrations in both the
vertical and horizontal dimensions of the plume from the source. Also, due
to budgetary limitations, isokinetic sampling was not attempted. The most
important criteria for performance of the exposure profiling method are
locating the samplers downwind of the source and having the samplers face
into the wind in-line with the source. Those criteria were met in this
program by monitoring wind direction and wind speed within the source area
and adjusting the location of the samplers accordingly.
During testing at Chloride Metals, the wind speed ranged from 2.8 to
7.7 miles per hour. Sampler flow rates typically ranged from 40 to 60 cubic
feet per minute. At those sampling rates, the filter face velocity is 1.0
to 1.3 miles per hour. As discussed previously, because of the small size
of particulate emissions, the bias of subisokinetic sampling is minimal.
The exposure profiling technique is designed to measure particulate
emissions. In this program, arsenic, lead, and cadmium emissions were also
5-2
-------
determined by analyzing the filters for those three elements. The elemental
concentrations were handled in the same manner as particulate concentrations
to determine emission rates for arsenic, lead, and cadmium.
As the application of the exposure profiling was conceived, samplers
were to be distributed over a sufficiently large portion of the plume from
an area source to define vertical and lateral boundaries of the plume by
spatial extrapolation of exposure measurements. During the performance of
the testing effort, the location of samplers was primarily dictated by
buildings and/or other obstructions at or near the boundary of the area
source and by traffic patterns in the area. As a general rule, samplers
were located on or very near the defined boundary of the area source being
tested. The samplers were positioned as far from obstructions as possible
to reduce the influence of wake effects at the sampler.
Samplers used to determine particulate concentrations during exposure
profile method testing were hi-volume ambient samplers marketed by General
Metal Works. Three hi-volume filter samplers including a filter plate,
motor, and housing with orifice plate were positioned at heights of 4 feet,
8 feet, and 12 feet. The three samplers were attached to a structure of
angle iron mounted on a wooden pallet. The timer and flow rate monitors for
the hi-volume samplers were attached to the base of the angle iron struc-
ture. The structure mounted with the hi-volume samplers and associated
timers and flow meters is termed a pole sampler and is depicted in Fig-
ure 23.
During the testing of an area source, at least two and usually three
pole samplers were positioned downwind of the area source and one upwind of
the source. Prior to sampling, a wind direction sensor and an anemometer
were stationed near the sampling area, and the wind direction monitored for
approximately 30 minutes to establish the wind direction prior to locating
the pole samplers. Wind speed and wind direction were monitored over the
entire testing period. The wind direction sensor and anemometer were posi-
tioned at a height of 10 feet.
5-3
-------
RADIAN
Hi-Volume Filter
Holder and Motor Housing
with Orifice
Chart Recorders for
Sampler Orif iceAP
and Sampling Time
Figure 23. Exposure profiling pole sampler with three
total suspended particulate samplers.
5-4
-------
Once the pole samplers were positioned, a disk chart was loaded into the
hi-volume samplers timer and flow rate monitors, and the hi-volume samplers
were loaded with filters. To facilitate filter loading and unloading, the
8-inch by 10-inch filters were contained in filter cartridges. During
sampling, the pole samplers and area source were constantly inspected to
spot operating problems with the samplers and to document operations within
the area being tested. Each hi-volume sampler required 6 to 7 amps of 110
volt electrical power so each pole required about 20 amps, 110V. Consider-
able problems were experienced with overloading of electrical circuits.
These problems were solved over the course of the testing effort.
Typically, the sampling time for determination of area source fugitive
emissions was about 6 hours. However, sampling times varied from about 4
hours to about 8 hours. The length of time necesary to collect a represen-
tative sample is the result of collecting enough particulate matter to be
accurately weighed and sampling over a sufficient period of time to average
process fluctuations or process area activities.
Other considerations are the conditions that had been specified that
would negate measurements. Those criteria are:
• rainfall in sufficient quantity to wash or knock particulate
matter from the filter (only a mist could be tolerated); and
• a change in wind direction of 90° from the wind direction
established at the start of sampling. Short-term wind
shifts (variability) were not allowed to negate results;
only a consistent, well-defined shift in the wind direction
of 90° of more.
Fortunately, no tests were negated due to these criteria. There was no rain
over the three-week test period. On one or possibly two occasions, wind
shifts did cause sampling to be terminated, but the sampling time was ade-
quate for representative collection.
5-5
-------
Eight pole samplers were constructed to perform the testing effort at
Chloride Metals. Usually two, but at times three, area sources were tested
simultaneously. Two met (meteorological) stations were used during the
program to allow concurrent measurement of wind speed and wind direction in
two areas.
During the exposure profiling testing for fugitive particulate emis-
sions from area sources, the size distribution of suspended particulate
matter was measured. Particle size distribution measurements were performed
using a Sierra Model 235 five-stage cascade impactor designed for ambient
applications. The impactor attaches to a standard hi-volume sampler in like
manner to fitting a total particulate filter to a hi-volume sampler. The
Sierra Model 235, as all cascade impactors, aerodynamically sizes suspended
particulate into discrete size ranges. The particle size cut-points for the
Model 235 ambient cascade impactor are presented in Table 24. During this
program, the samplers were operated near 40 cubic feet per minute. Sized
particulate was collected on glass fiber filter substrates to facilitate
sample recovery and provide for more accurate weighing of the mass weight
gain.
Three of the eight pole samplers were outfitted with ambient cascade
impactors for particle size distribution measurements at the four foot
level. Metal cross members were attached horizontally and hi-volume sam-
plers were mounted on each end of the cross members. The samples were
located four feet from ground level and eight feet apart.
Particle size distribution measurements were performed at both upwind
and downwind sampling sites for each area source. Since there were only
three cascade impactors and usually two area sources were sampled concur-
rently, upwind and downwind samples were not collected simultaneously during
all test periods.
Data Reduction—Data reduction to determine area emission rates was per-
formed using an adaptation of the methodology described by Schwitzgebel (2).
5-6
-------
TABLE 24. SIERKA MODEL 235 FIVE-STAGE IMPACTOR PARTICLE SIZE
CUT-OFFS (MICRONS)*
Flow Rate
Stage Number
1
2
3
4
5
40 cfm
7.2
3.0
1.5
0.95
0.49
20 cfm
10.2
4.2
2.1
1.4
0.73
aCut-offs at 50% collection efficiency for spherical particles with unity
mass density @25°C and 760 mm Hg.
5-7
-------
The approach is a ventilation model in which area source emission rates are
determined as a function of the height of emissions, the width of the
ventilated area, and the wind speed. As depicted in Figure 24, emissions
per unit time are assumed to equal the concentration contained in a rec-
tangle defined by a ventilation base (V^) on the x-axis, the wind speed (v)
on the y-axis, and the height (hmax) on the z-axis, which is selected in
such a way that the vertical concentration approaches zero.
The ventilation base is the width of the effective area source defined
by the average wind direction during the sampling period. The dimension of
the ventilation base was determined graphically as shown in Figure 25. The
wind vector is the average wind speed during the sampling period and mea-
sured in or very near the area source. The lead concentration at any height
is defined as a function of that height and decreases with increasing
height.
With these definitions, the contribution of a horizontal segment at
height h and the thickness H can be described as follows:
AEh,h+ Ah ' C ' Ah ' v ' Vb (1)
where, AE^ ^ + Ah = area emissions by the segment between h and
h + Ah
v = wind speed
Vjj = ventilation base
C(h) = concentration at height h
h = vertical dimension of ventilated segment.
Schwitzgebel forced the concentration results as a function of height
n
to be a linear function to define the concentration at ground level ( hQ)
and the height at which the vertical concentration is zero (bmax)• With
these factors, the area emission rate would be determined by:
AE - 1/2 Ch0 • hmax ' v • Vb (2)
5-8
-------
RADIAN
hmax
c(hmax) = 0
Figure 24. Ventilation model.
5-9
-------
co
C
o
•H
CO
0)
cu
co
ta
,0
B
o
•H
4-1
tO
c
o
0)
c
4J
cu
a
-------
In this program, the measured concentration data for each pole sampler
were fitted to a normal curve (actually one-half of a normal curve) as shown
in Figure 26. The area of the curve was determined by integrating under the
curve to determine the factor [C(h) * Ah] in Equation 1, i.e., Area (yg/m ) =
O
concentration (yg/m ) * height (m). The curve areas for all downwind sam-
pling locations during a sampling period at a given source were averaged and
multiplied by the average wind speed in the area source during the sampling
period to determine the flux rate (yg/m-sec). The flux rate is then multi-
plied by the ventilation base to determine the emission rate.
5.1.1.2 Smelter Building Emissions Measurements—
Arsenic, lead, cadmium, and particulate emissions from the smelter
building were measured by applying the pole samplers described in the pre-
vious section. Initially wind flow patterns through the smelter building
were observed to determine in and out wind flow of the building. One pole
sampler was located just outside of the smelter building at a point of flow
into the building, i.e., upwind. Two to three pole samplers were located at
the edge of the building at locations of wind flow from the building, i.e.,
downwind.
During the testing period, the air velocity was measured at a series of
matrix points at building openings. A Rurz Model 441 M air velocity meter
was used to measure air velocity. The meter utilizes a hot-wire anemometer
which consists of a velocity sensor and a temperature sensor. The velocity
sensor is heated and operated as a constant-temperature thermal anemometer
which responds to velocity by sensing the cooling effect of the air as it
passes over the heated velocity sensor. The wind direction, i.e., in or
out, was also noted. The measured air velocities were related to respective
building opening areas to determine the air flow rates into and out of the
smelter building. Smelter building dimensions and effective emission areas
of building openings used to calculate air flow rates are shown in Figure
22. The smelter building air flow rates are presented in Appendix D.
5-11
-------
The concentrations of arsenic, lead, cadmium, and particulate measured
at three heights at smelter "downwind" sites typically did not demonstrate a
decrease in concentration with an increase in height. More often concentra-
tions increased as height increased. This is understandable due to the
heating of building air by the blast furnace and refining kettles in the
smelter building. For these reasons the concentrations of building emis-
sions were not integrated as a function of height, but instead, an average
concentration was calculated for each of the four species.
The smelter building emission rates were calculated by relating the
concentration of each species to the air flow rate at the building opening.
In some cases, concentration results were not available for all building
openings and data from a near location measured at the smelter building were
applied to that opening.
5.1.2 Smelter Building Concentrations Method
Concentrations of arsenic, lead, and cadmium were measured in the
smelter building in the vicinity of the blast furnace and the refining
kettles. Measurements were taken using a membrane filter and an orifice
sampler to control and measure flow.
The filters used were 37 mm Millipore MF-type membrane filters having a
0.8 pore size. The filters were contained in a plastic cassette which
affixed directly to the orifice sampler shown in Figure 27. The orifice
sampler included a calibrated, critical orifice. The critical orifices are
interchangeable. During this program, critical orifices had a flow rate of
4.9 liters per minute at a pressure drop of 16 inches of mercury vacuum.
The vacuum drop across the critical orifice was maintained by a double-
diaphragm pump equipped with a vacuum gauge to monitor the line vacuum.
Heavy-walled rubber vacuum hose was used to connect the sampler to the pump.
Samples were collected concurrently at heights of 2 feet, 4 feet, and 6
feet. The samplers were attached to a single vertical standard affixed to a
5-13
-------
Vaccuum Hose
37mm
Filter Cassette
Filter Aaaoter
Flow Limiting Orifice
Figure 27. Critical orifice sampler with filter cassette.
5-14
-------
RADIAN
board base. All three samplers were manifolded to a single pump to maintain
the pressure drop of the critical orifices.
Measurements were taken near the refining kettles during the following
process steps:
• production of hard lead (high arsenic),
• production of soft lead,
• lead casting, and
• dressing.
Measurements were also taken near the slag tap and lead well of the blast
furnace.
5.1.3 EPA Method 108
EPA Method 108 is an Appendix B test method developed to determine
particulate and gaseous arsenic emissions from stationary sources. The
method was used to quantify arsenic emissions from the slag tap baghouse
(slag tap and lead tap hoods) and the high energy wet scrubber through which
off-gases from the refining kettle and dressing hoods are vented.
The method was applied without deviation as printed in the Federal
Register, Vol. 48, No. 140, Pages 33166-33172, July 20, 1983. Sample gas
was drawn isokinetically through a gooseneck nozzle, a heat-traced, glass-
lined probe, and a glass fiber filter maintained at a temperature of 230° to
275°F for particulate collection. The sample then passed through a chilled
impinger train for gaseous arsenic and SC^ collection. The first two im-
pinge rs contained water for gaseous arsenic collection. The third, fourth,
and fifth impingers contained 10% l^C^ for S02 collection. The filter, the
probe wash, and the water impinger solutions were analyzed for arsenic.
The samples collected from the wet scrubber and slag tap baghouse were
also analyzed for lead and cadmium. Method 108 may produce low gaseous lead
5-15
-------
and cadmium results because of inefficient collection of lead and cadmium in
the water impingers. The solution for arsenic collection in Method 108 is
water while EPA Method 12 calls for 0.1 N HN03 for collection of gaseous
lead. There is no protocol procedure for cadmium.
5.2 ANALYTICAL METHODS
All total particulate loading filters and a representative portion of
size distribution filter substrates collected during fugitive emissions
testing were analyzed for arsenic, lead, and cadmium. In addition, the
membrane filters collected in the smelter building and Method 108 samples
were analyzed for the three metals.
5.2.1 Arsenic. Lead. Cadmium Analysis
Following sections describe the dissolution technique and instrumental
methods followed for the analysis of arsenic, lead, and cadmium.
5.2.1.1 Dissolution Technique—
As stated in Section 5.1.1.1, the 8-inch by 10-inch total suspended
particulate filters used for the exposure profiling and smelter building
emissions were loaded into filter cartridges which in turn were attached to
the hi-volume sampler. After sample collection, the filter cartridge was
returned to the mobile laboratory for sample retrieval. As the filter was
removed, it was folded in half, face-to-face and inserted in an envelope.
The filters were conditioned and weighed on site. At Radian laboratories
the exposed area of the filters was quartered, and one-quarter of the filter
was analyzed for arsenic, lead, and cadmium. There is a possibility that
one-quarter of the filter is not representative of the whole filter if
particulate was not homogeneously distributed over the filter. To address
that potential problem, two quarters of some filters were analyzed indivi-
dually.
5-16
-------
For the analysis of the membrane cassette filters, particle size dis-
tribution filter substrates, and the Method 108 filter, the whole filter was
digested for analysis. The Method 108 impinger solutions were analyzed
directly.
The filter samples for analysis were first digested to get the elements
into solution for instrumental analysis. The filter was placed in a beaker
and 5 milliliters of concentrated nitric acid and 5 milliliters of concen-
trated hydrochloric acid were added to the beaker. The beaker was then
heated for one hour. Water was then added to the beaker and heated for
another hour. The solute was then transferred to a volumetric flask and
taken to volume. Prior to instrumental analysis, the sample solute is
filtered.
5.2.1.2 Instrumental Analysis—
Filter dissolution samples and impinger solutions were analyzed for
arsenic, lead, and cadmium using atomic absorption spectroscopy. Arsenic
was analyzed using a hydride generation attachment for the atomic absorption
spectrophotometer. An aliquot of the sample is introduced into the system
and acidified. Sodium borohydride is added to reduce the arsenic to the
hydride, and the generated gas is purged into a hydrogen-argon flame for
analysis. The detection limit for arsenic by the hydride generation tech-
nique is 0.003 yg/ml (ppmv).
Samples were analyzed for lead using flame atomic absorption spectro-
scopy. An air-acetylene flame was used as the source for excitation. The
detection limit for lead by the flame atomic absorption spectroscopy method
is 0.1 yg/ml.
The analysis of cadmium was performed by graphite furnace atomic ab-
sorption spectroscopy. The detection limit for cadmium by the graphite
furnace atomic absorption spectroscopy technique is 0.002 yg/ml.
5-17
-------
RADIAN
5.2.2 Filter Weighings
All filters including 8-inch by 10-inch hi-volume filters used in the
exposure profiling technique, glass fiber substrates used for ambient par-
ticle size distribution measurements, and the filters used in the Method 108
testing, were weighed before and after testing to determine particulate
weight gain. The 37 mm cassette filters were not weighed, and the particu-
late weight gain was not determined.
5.2.2.1 Filter Conditioning—
The 8-inch x 10-inch hi-volume filters and the glass fiber substrates
for particle size measurements were conditioned prior to all weighings
following the protocol of EPA ambient air monitoring guidelines (4). Prior
to weighing, the filters were equilibrated for 24 hours in a conditioning
environment controlled by a constant humidity chamber. The chamber con-
trolled temperatures to 20 to 25°C +3°C and maintained the relative humidity
below 50 percent. The ambient air monitoring guidelines require that the
relative humidity vary by no more than 5 percent between weighings.
Filters used in the Method 108 testing were baked at 260°C and then
dessicated and weighed to a constant weight (+0.5 mg).
5.2.2.2 Instrument—
Filter weights were determined using a Mettler AE 163 electronic analy-
tical balance. Weighings were made to the nearest 0.1 milligram. Weighings
were performed at the plant site in a mobile laboratory.
5-18
-------
SECTION 6
QUALITY ASSURANCE
Emission test program quality assurance/quality control guidelines are
developed to ensure the production of acceptable and representative data.
The performance of a successful quality assurance program provides the
following:
• the detection of problems in the sampling and analytical
tasks;
• the determination of estimates of precision and accuracy;
• the proper documentation of procedures and methods used in
the emissions test program.
Following sections describe the quality assurance activities performed for
the Chloride Metals emission test program.
6.1 SAMPLING QUALITY ASSURANCE
Radian implemented quality assurance procedures during all sampling
activities. Quality assurance procedures were performed for the three
sampling techniques followed during the Chloride Metals emissions test
program:
• exposure profiling technique for area fugitive emissions;
• EPA Method 108 for stationary sources emissions; and
6-1
-------
RADIAN
• workplace sampling for smelter building concentrations.
The quality control procedures implemented during the sampling program
prescribed guidelines for:
• equipment calibration;
• sampling protocol; and
• sample handling techniques.
The major emphasis on quality control for sampling was strict calibra-
tion of gas metering systems, wind monitors, temperature control devices,
leak testing, and documentation.
Of primary concern was that the sampling equipment be in proper opera-
ting condition prior to and during sampling. In order to achieve this,
equipment was inspected and cleaned thoroughly prior to the field effort,
monitoring devices were checked and calibrated, and volume measurement
devices were calibrated prior to sampling. All calibrations were properly
documented and retained. During sampling, equipment was monitored con-
tinuously for proper operation.
6.1.1 Equipment Calibration
All of the sampling techniques performed at Chloride Metals required
flow monitoring and temperature sensing equipment. Table 25 presents a
summary of the equipment which was calibrated for the emissions test. The
accurate measurement of sample volumes is one of the most critical criteria
for assuring proper sample collection. During the Chloride Metals emission
test program, sample volume metering devices used included hi-volume sam-
plers, critical orifices, and dry gas meters.
The hi-volume samplers used in this program are identical to ambient
air monitoring systems except that the samplers are not contained in an all-
weather housing. The flow monitoring system, AF transducer, and timer are
6-2
-------
TABLE 25. SUMMARY OF CALIBRATED EQUIPMENT USED IN SAMPLING AT CHLORIDE METALS
Parameter
Volumetric Gas
Flow Rate
Molecular
Weight
Moisture
Partlculate &
Gaseous
Arsenic
Particulate
Particulate
Wind Speed
Wind Direction
Method
EPA-1,
EPA-2
EPA-3
EPA-4
EPA- 108
Exposure
Profiling
Critical
Orifice
Sampler
Exposure
Profiling
Exposure
Profiling
Calibrated Equipment Used in Measuring Parameters
Type-S Differential Temperature Gas Anemometer/ Wind
Pitot Pressure Measuring Metering Sampling Vane or Direction
Tube Gauge Device System Nozzle Fyrlte Hot-Wlre Sensor
* * *
*
* * *
* * * * *
* *
*
*
*
-------
identical. Each of the hi-volume gas metering systems (a total of 27 were
operated simultaneously) was calibrated at Radian prior to the emissions
test program according to methods published in Quality Assurance Handbook
for Air Pollution Measurement Systems Volume II - Ambient Air Specific
Methods (3). The timers were not calibrated over a 24-hour period because
sampling periods were 8 hours or less and because start and stop times were
recorded. Calibration data were recorded and the data retained. Following
the emissions test program, a single-point calibration check was performed
on the hi-volume samplers. The test plan had established an acceptable
calibration check to be within 5% of the initial multi-point calibration
data. The posttest calibration results were within acceptable limits for
all but three of the hi-volume samplers. For those samplers which had post-
test calibrations outside of the 5% limit, the pretest and posttest calibra-
tion results were averaged, and that value was used in sampling volume
determinations. The posttest results and respective pretest results are
presented in Table 26.
Critical orifice samplers were used to measure smelter building concen-
trations. The six critical orifices used during this program were cali-
brated against an NBS traceable-calibrated wet test meter. The calibration
data were dutifully recorded and the records retained.
The dry gas meter used in the EPA Method 108 train was calibrated in
conformance to EPA's publication Quality Assurance Handbook for Air
Pollution Measurement Systems Volume III - Stationary Source Specific
Methods (4). Calibration was performed over a range of flow rates as sug-
gested in the EPA publication and calibration correction factors for the gas
meter and orifice meter in the sampling unit were calculated according to
suggested procedures. A posttest calibration was performed at the com-
pletion of the program. All calibration results were documented using EPA
prescribed forms and the results retained.
6-4
-------
TABLE 26. HI-VOLUME SAMPLER CALIBRATION CHECKS
IB
1M
IT
2B
2M
2T
3B
3M
3T
4L
4R
4M
4T
5L
5R
5M
5T
6B
6M
6T
7B
7M
7T
8L
8R
8M
8T
Serial
No.
12516
9711
12635
13241
12531
13229
8905
9709
13631
9737
8911
12638
12893
14592
13629
12887
12581
9705
9693
12507
9736
9702
9699
13106
12888
15981
12886
SCFM
Pretest
56.9
59.7
46.7
57.2
59.7
57.5
56.4
59.7
59.9
39.9
59.1
58.1
58.3
41.5
58.9
55.5
51.5
56.1
55.8
57.8
57.2
58.3
58.3
43.4
58.3
56.9
59.1
Plate
No.
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(7)
(13)
(13)
(13)
(7)
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(13)
(7)
(13)
(13)
(13)
SCFM
Posttest
56.9
59.4
47.7
58.8
59.4
57.8
56.9
58.1
59.1
43.1
57.2
58.3
56.9
43.1
58.9
58.9
59.1
57.5
56.4
58.0
58.1
59.1
58.9
43.8
59.1
59.7
59.9
% Difference
0.0
0.5
2.1
2.8
0.5
0.5
0.9
2.7
1.3
8.0
3.2
0.3
2.4
3.9
0.0
6.1
12.6
2.5
1.1
0.3
1.6
1.4
1.0
0.9
1.4
4.9
1.4
6-5
-------
6.1.2 Sampling Protocols
Sampling techniques used during this program were EPA referenced or
"state-of-the-art" methods, some with modifications to be more applicable.
Sample collection was done in accordance with the methods prescribed in the
Chloride Metals Test Plan (5).
QA procedure checks to be performed during exposure profiling measure-
ments included:
• use of data forms to record start time, initial AP and stop
time, final AP;
• performance of visual inspections of sampling systems;
• performance of visual monitoring of wind direction to assure
proper alignment of samplers; and
• performance of a single point posttest calibration of each
hi-volume sampler.
During sample collection with the critical orifice samplers, the fol-
lowing QA procedures were followed:
• use of data forms to record start time, stop time, and
initial, intermittent, and final sampling system pressure
drop;
• routine inspection and monitoring of sampling system
pressure drop; and
• performance of visual inspections of sampling systems.
6-6
-------
The following QA procedure checks were followed during Method 108
sampling:
• use of standard data forms and source sampling data sheet
checklist;
• performance of visual inspections of sampling systems,
• performance of system leak checks before and after sampling;
• performance of heating system checks;
• performance of impinger ice checks;
• performance of isokinetic sampling rate checks; and
• performance of daily data review and calculation checks.
6.1.3 Sample Handling Techniques
The objectives of a quality assurance program regarding sample handling
techniques are to ensure that the integrity of collected samples is not
diminished prior to analysis.
To aid in the retrieval of filter samples collected during the exposure
profiling sampling, the filters were contained in filter cartridges. Fil-
ters were loaded in the cartridges in the mobile laboratory on site. After
sample collection, the filter cartridges were returned to the mobile labora-
tory for retrieval of samples in an enclosed, controlled environment, having
no wind currents.
After samples have been properly obtained in the field, their subsequent
handling during transfer to the analytical laboratories becomes an important
factor in the successful performance of an emissions characterization
6-7
-------
program. All collected samples were labeled at the time of sample collection
with adequate descriptions of the samples to prevent confusion among multiple
samples, and the sample numbers were entered into a logbook. Samples were
inventoried against logbook records prior to shipment.
Samples submitted to Radian Analytical Services Laboratories included a
chain-of-custody report to document the receipt of samples and to have a
mechanism for tracking the samples through the analyses. The completed
chain-of-custody reports used in this project are included in Appendix I.
6.2 ANALYTICAL QUALITY CONTROL
Analytical quality control was performed by ongoing quality control
activities in Radian Laboratories and also by project specific quality
control activities.
6.2.1 Radian Laboratories Quality Control
A standard regime of analytical quality control is followed routinely
by Radian laboratories. The procedures use various checks in order to
determine the validity of analyses. These include:
• calibration standards,
• certified standards,
• in-lab standards,
• blanks,
• spikes, and
• replicates.
Quality assurance begins with the sample log and carries through the
reporting of data.
The unique identifying number assigned in the field and recorded in the
sample log facilitates tracking and identification and prevents mix-ups
6-8
-------
during the analysis process. A chain-of-custody report is used to monitor
the samples through analytical laboratories.
Chemical characterization of samples was performed using standard
atomic absorption spectrophotometry (AAS) and digestion techniques.
The accuracy and precision of these analyses in Radian's laboratories
are documented through ongoing QA-QC programs. Accuracy is evaluated by
analyzing standards and blind, spiked samples. Precision is monitored by
replication of analysis on 5 to 15 percent of samples to establish back-
ground concentration and potential interferences.
6.2.2 Project-Specific Quality Control
Project-specific analytical quality control activities involved two
major criteria. First, the samples were submitted in three sets. In this
project, there were 376 filters and 21 EPA Method 108 samples analyzed for
arsenic, lead, and cadmium. To facilitate monitoring of results and identi-
fication of any analytical problems, three groups of samples were submitted
to the laboratories, with only one group of filters being in the lab at a
given time. Analytical results were reviewed before another group of fil-
ters was submitted to the lab.
Second, with each group of samples, NBS certified standards for lead
were submitted for analysis. The lead standard (SRM 2674) is a set of three
filter strips with lead loadings of 100, 303, and 1505 yg/filter. In addi-
tion, each group of filters submitted for analysis included blank filters
for background correction. Table 27 presents the results of analysis of the
NBS lead standard filters submitted with the three sets of filters. The
accuracy of the analysis of the first set of filters was rather poor.
However, the results of the last two sets of filters were quite good. The
quality control results of the first set of filters were below acceptable
limits for lead analysis by atomic absorption spectroscopy. Due to budge-
tary constraints and because the objectives of the program were to develop
6-9
-------
TABLE 27. RESULTS OF ANALYSIS OF LEAD QUALITY CONTROL STANDARD3
Lead Content (yg/filter)
Average
Value
Tolerance
Limits
Measured
Value
Relative
Accuracy
1st Set
100
303
1505
97-103
294-312
1477-1533
75
190
1010
75
63
67
2nd Set
100
303
1505
97-103
294-312
1477-1533
47
290
1500
47
96
99+
3rd Set
303
1505
294-312
1477-1533
290
1500
116
99+
6-10
-------
em
qualitative more than quantitative emission rates, the analyses were not
repeated.
In preparation for analysis, the exposed area of the 8-inch x 10-inch
exposure profiling filters were cut into four equal sections. One-quarter
of the filter was submitted for analysis. In response to the concern that
one-quarter of the filter would not provide a representative sample because
of non-homogeneous distribution of particulate on the filter, analyses for
arsenic, lead, and cadmium were performed on two quarters of 24 filters.
The results of analysis of these filters are presented in Table 28.
The relative difference in the analysis of lead for 24 duplicate fil-
«j
ters was +12.6% with an average concentration of 28.7 yg/m . That dif-
ference is based upon the ambient sample concentration and is more correla-
table to emission rate results. The precision of filter analysis for lead
is quite good. The relative difference between analysis of filters for
arsenic and cadmium is +24.6% and +.64.5%, respectively, with average concen-
trations of 0.180 yg As/m3 and 0.042 yg Cd/m3. While the precision of
filter analyses for arsenic and cadmium is considerably less than for lead,
the results indicate that precision is directly related to the concentration
of the element on the filter.
Mass weight gains were determined on all 8-inch x 10-inch hi-volume
filters and the impactor substrates used in the exposure profiling sampling.
These filters were conditioned for 24-hours prior to weighing in a constant
humidity chamber. The filters were conditioned at a constant humidity
maintained below 50 percent, and the humidity could not vary by more than 5
percent between initial and final weighings.
To document that proper weighing procedures were being followed and
that the methodology was precise, blank filters were weighed on two occa-
sions after conditioning for 24-hour periods. The results of those
weighings are presented in Table 29. The repeatability of measurements is
excellent. Filter particulate weight gains were typically in the range of
6-11
-------
TABLE 28. -RESULTS OF ANALYSIS OF DUPLICATE FILTER SAMPLES
Filter Number
1st Set 028251
028214
028219
028217
028227
028205
131149
028203
Average
2nd Set 131119
131138
131118
131113
131196
5189665
5189638
5189785
Average
3rd Set 5189622
5189617
5189633
5189751
5189773
5189782
5189792
5189642
Average
Average (All)
Pb
Primary
Sample
22.4
22.8
16.7
45.1
30.4
1.37
3.66
1.36
18.0
9.05
15.1
16.3
54.3
40.6
13.8
2.14
28.2
22.4
8.15
3.11
4.91
75.1
158
82.2
7.70
27.3
45.8
28.7
Concentration (pg/cr)
Duplicate
16.8
24.0
17.3
52.9
29.6
1.43
3.45
1.16
9.05
13.3
16.5
53.6
44.9
13.8
1.49
26.3
8.74
3.59
3.40
84.5
188
100
9.80
34.9
Relative
Difference
Z
-25.0
5.3
3.6
17.3
-2.6
4.4
-5.7
-14.7
±9.8
0
-11.9
1.2
-1.3
10.6
0
30.4
-6.7
±7.8
7.3
15.4
-30.8
12.5
19.0
21.7
27.3
27.8
±20.2
±12.6
At Concentration (pg/m3)
Primary
Sample
0.242
0.196
0.107
0.258
0.084
HD
0.014
HD
0.150
0.040
0.070
0.004
0.400
0.229
0.039
0.010
0.197
0.124
0.0105
0.0153
0.0360
0.450
0.988
0.516
0.031
0.084
0.266
0.180
Duplicate
0.218
0.215
0.113
0.271
0.076
HD
0.010
HD
0.035
0.064
0.007
0.289
0.178
0.035
0.007
0.178
0.0297
0.0134
0.0265
0.479
0.988
0.569
0.043
0.106
Relative
Difference
Z
-9.9
9.7
5.6
5.0
-9.5
-28.6
±11.4
-12.5
-8.6
75.0
-27.8
-22.3
-10.3
-30.0
-9.6
±24.5
183
-12.4
-26.4
6.4
0
10.3
38.7
26.2
±37.9
±24.6
Cd Concentration (Pg/m3)
Primary
Sample
0.010
0.013
0.059
0.103
0.014
0.001
0.006
0.001
0.026
0.004
0.023
0.019
0.032
0.119
0.022
0.011
0.038
0.033
0.0204
0.0091
0.0159
0.122
0.178
0.137
0.0175
0.0137
0.064
0.041
Duplicate
0.021
0.021
0.030
0.023
0.008
0.001
0.005
0.001
0.005
0.005
0.005
0.010
0.130
0.002
0.003
HD
0.0157
0.0115
0.0166
0.0563
0.109
0.078
0.0322
0.0737
Relative
Difference
Z
110
61.5
-49.2
-77.7
-42.9
0
-16.7
0
±44.8
25.0
-78.3
-73.7
-68.8
9.2
-90.9
-72.7
±59.8
-23.0
26.4
4.4
-53.9
-38.8
-43.1
84.0
438
±89.0
±64.5
-------
TABLE 29. COMPARISON OF BLANK FILTER WEIGHTS
Filter No.
131139
131140
028225
028224
5189799
5189800
5189699
5189700
000072
000188
956024
956027
Initial
Weight
(g>
4.0524
4.0519
3.5318
3.5959
4.1410
4.1579
4.2685
4.2253
1.4508
1.4182
1.4621
1.4633
Final
Weight
(g)
4.0513
4.0517
3.5326
3.5963
4.1405
4.1577
4.2694
4.2259
1.4510
1.4182
1.4624
1.4631
Difference
(g)
0.0011
0.0002
0.0008
0.0004
0.0005
0.0002
0.0009
0.0006
0.0002
0.0000
0.0003
0.0002
Relative
Difference
(%)
0.027
0.005
0.023
0.011
0.012
0.005
0.021
0.014
0.014
0
0.021
0.014
Average
0.0004
0.014
6-13
-------
0.10 to 0.05 grains, and the average variability of duplicate filter
weighings was 0.0004 grains. The variability of weighings should have no
effect on particulate weight gain measurements.
6.3 DATA ANALYSIS QUALITY ASSURANCE
Quality assurance procedures for data processing activities centered on
validating sampling data and using computerized data reduction systems.
Exposure profiling measurement data reduction required the following
information:
• the configuration of samples relative to the area source and
wind direction to determine the ventilation base of the area
source and to ensure that samplers were located directly
downwind of the source;
• the start time, stop time, average AP, and filter
location and identification for each sampler to determine
the volume of gas sampled; and
• the wind speed.
These data were recorded during sampling on preformatted forms presented in
the test plan. In addition to the above data, wind velocities into and out
of the smelter building were necessary to determine emission rates for that
source.
Flux rate calculations for each pole sampler were determined by com-
puter. Program inputs were keypunched and verified. Computer printouts of
flux rates are in Appendix C.
6-14
-------
Sampling data for the concentration measurements in the smelter
building were also recorded on preformatted forms included in the test plan.
The following data were recorded:
• sample location,
• filter number and height,
• start and stop times, and
• critical orifice sampler line vacuum to ensure calibration
flow rates.
During Method 108 sampling of the slag tap baghouse and Kinpactor
scrubber, data were recorded on field data sheets formatted for coding
directly into Radian's Source Sampling Data Reduction Program. The data
entered are uniquely identified by sampling location and sample identifica-
tion number and serve as inputs to the data reduction program. The compu-
terized data reduction program produces an input verification report to
review the input data. Gas velocity data were also reduced by the Source
Sampling Data Reduction Program. The computerized data reduction results
for the slag tap baghouse and Kinpactor scrubber are in Appendix H.
6-15
-------
REFERENCES
(1) Bohn, R., T. Cuscino, Jr. and C. Cowherd, Jr., Fugitive Emissions from
Integrated Iron and Steel Plants. Midwest Research Institute for U. S.
EPA, EPA-600/2-78-050, March 1978.
(2) Schwitzgebel, Klaus, et al., Fugitive Emissions at a Secondary Lead
Smelter. Radian Corporation, Austin, Texas, December 1981.
(3) U. S. EPA, Quality Assurance Handbook for Air Pollution Measurement
Systems Volume II Ambient Air Specific Methods. U. S. EPA, EPA-600/
4-77-027a, May 1977.
(4) U. S. EPA, Quality Assurance Handbook for Air Pollution Measurement
Systems Volume III Stationary Source Specific Methods. U. S. EPA, EPA-
600/4-77-027b, August 1977.
(5) Fuchs, Michael R., "Test Plan for Chloride Metals Secondary Lead
Smelter in Tampa, Florida," Radian Corporation, Austin, Texas, February
1984.
------- |