OF AM AIR CURTAIN
HOODING SYSTEM FOR A PRIMARY
COPPER CONVERTER
ASARCO, Inc.
Tacoma, Washington
PEDCo ENVIRONMENTAL
-------
A-80-40
II-A-40
EVALUATION OF AN AIR CURTAIN
HOODING SYSTEM FOR A PRIMARY
COPPER CONVERTER
ASARCO, Inc.
Tacoma, Washington
by
PEDCo Environmental, Inc.
Cincinnati, Ohio 45246
Contract No. 68-03-2924
Work Directive 9
PN 3490-9
and
Contract No. 68-02-3546
Task Assignment No. 12
PN 3530-12
Project Officers
John 0. Burckle
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45220
and
Alfred Vervaert and Frank Clay
Emission Standards and Engineering Division
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
December 1983
-------
DISCLAIMER
This draft report was prepared for the U.S. Environmental
Protection Agency (EPA) by PEDCo Environmental, Inc., Cincinnati,
Ohio, under Contract No. 68-03-2924, Work Directive No. 9 and
Contract No. 68-02-3456, Task Assignment 12. The contents of
this document dosnot necessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
FOREWORD
Hi
-------
ABSTRACT
This report presents the results of tests conducted to
evaluate the effectiveness of a full-scale air curtain capture
system installed on a primary copper smelter for capture of low
level fugitive particulate, including trace metals, and sulfur
dioxide. The test work was performed onsite at ASARCO's Tacoma
Smelter on the first domestic full-scale prototype system, re-
sulting in the first published evaluation of a full-scale fugi-
tive capture system based upon the air curtain approaches ap-
plied to a primary copper converter.
The installation of the air curtain hooding system has per-
mitted a quantitative approach to the direct measurement of the
fugitive emissions for the first time. In this program, the
fugitives captured by the air curtain were measured at a down-
stream sampling point in the exhaust side of the air curtain
system during the various portions of the converter cycle.
Emission factors were established for sulfur dioxide, filterable
particulate (Method 5), inhalable particulate, and selected trace
elements.
-------
CONTENTS
Page
Foreword iii
Abstract iv
Figures vii
Tables ix
Acknowledgment xiv
1. Introduction 1
2. Process Description 4
2.1 Converter operation 4
2.2 Converter emissions 8
2.3 Air curtain hooding system 9
3. Process Operation and Test Log 13
4. Air Curtain Capture Efficiency 21
4.1 Tracer gas mass balance 21
4.2 Opacity 51
4.3 Visual observations 54
5. Emission Factor Development 84
5.1 Sulfur dioxide (S02) 85
5.2 Filterable particulate emissions 97
5.3 Filterable and gaseous arsenic 102
5.4 Particle size results 107
5.5 Trace metals—antimony, bismuth, cadmium, lead,
and selenium 152
5.6 Process samples 154
6. Quality Assurance 157
7. Sampling and Analytical Plan 170
7.1 Sample location 170
7.2 Velocity and gas temperature 170
7.3 Molecular weight 173
-------
CONTENTS (continued)
7.4 Particulate/arsenic
7.5 Particle size distribution
7.6 Sulfur dioxide manual method
7.7 Continuous monitoring for sulfur dioxide
7.8 Hood capture efficiency using a tracer gas
7.9 Opacity
Appendices
A Computer Printouts and Example Calculations A-l
B Field Data B-l
C Laboratory Results C-l
D Sampling and Analytical Procedures D-l
E Calibration Procedures and Results E-l
F Quality Assurance Project Plan F-l
G Project Participants G-l
H Visual Observation Logs H-l
I Interlaboratory Comparison Study of Proposed
Method 108 1-1
VI
-------
FIGURES
Number Page
1 Typical Peirce-Smith Primary Copper Converter 5
2 Converter with Primary Hood 6
3 Copper Converter Operation 6
4 Converter Air Curtain/Secondary Hooding
System (No Scale) 11
5 Air Curtain Control System 12
6 SF, Concentration Profile 29
O
7 Graphical Presentation of Air Curtain Sample
Ports 31
8 Injection Point Matrix Used for Preliminary
Tracer Gas Tests 32
9 Velocity Profile for Low Flow Condition 38
10 Velocity Profile for High Flow Condition 39
11 SF, Injection Locations 40
12 Tracer Injection Matrix 41
13 Comparison of Hood Collection Efficiency
and Matrix Port Injection 46
14 Comparison of Hood Collection Efficiency
and Matrix Point Injection 47
15 Average Opacity vs. Converter Operation 53
16 Individual Particle Size Distributions for
the Charging Mode Sample Runs 114
17 Individual Particle Size Distributions for
the Skimming Mode Sample Runs 115
VII
-------
FIGURES (continued)
Number Page
18 Individual Particle Size Distributions for
the Blowing Mode Sample Runs 116
19 Average Particle Size Distribution for the
Charging Mode 118
20 Average Particle Size Distribution for the
Skimming Mode 119
21 Average Particle Size Distribution for the
Blowing Mode 120
22 Comparison of Elemental Concentrations for
the Charging Mode (Run Nos. PSMC-1, -2, -3) 133
23 Comparison of Elemental Concentrations for
the Charging Mode (Run Nos. PSMC-4, -5) 134
24 Comparison of Elemental Concentrations for
the Skimming Mode 142
25 Comparison of Elemental Concentrations for
the Blowing Mode 150
26 Audit Report Dry Gas Meter (Meter Box No. FB-4) 161
27 Audit Report Dry Gas Meter (Meter Box No. FB-5) 162
28 Audit Report Dry Gas Meter (Meter Box No. FB-7) 163
29 Audit Report Dry Gas Meter (Meter Box No. FB-9) 164
30 Audit Report S02 Analysis 165
31 No. 4 Converter Air Curtain Exhaust Duct
Sample Site 172
32 Transmissometer Installation on Secondary
Hood 179
Vlll
-------
TABLES
Number Page
1 Converter Cycle and Tests Conducted 14
2 Sample Matrix 15
3 Test Log 16
4 Summary of Preliminary Tracer Recovery Efficiency
Data 69
5 Summary of Single Point Tracer Recovery Efficiency
Tests 70
6 Summary of Tracer Recovery Efficiency at Air
Curtain Suction Inlet 71
7 Summary of Tracer Recovery Test Data on
Exhaust Side 72
8 Summary of Tracer Recovery Test Data on
Jet Side 74
9 Summary of Matrix Area Tracer Recovery Efficiency
Data 75
10 Collection Efficiency of SF, Within the Air
Curtain Matrix Area (December 1982) 34
11 Summary of Volumetric Flow Data 36
12 Tracer Collection Efficiency Within the Air Curtain
Control (Matrix) Area (January 14, 1983) 43
13 Summary of Matrix Injection Test Data, 1/14/83 76
14 SFfi Collection Efficiency Within the Control
TMatrix) Area (January 17, 18, 19, 1983) 45
15 Summary of Matrix Injection Test Data, 1/17/83 78
16 Summary of Matrix Injection Test Data, 1/18/83 79
17 Summary of Matrix Injection Test Data, 1/19/83 80
IX
-------
TABLES (continued)
Number Page
18 Summary of Tracer Recovery Efficiency for
Upper Control Volume 48
19 Summary of Tracer Recovery Efficiency for
Lower Control Volume 49
20 Summary of Sample and Analytical Data for
Special Injection Point Tests 81
21 Summary of Opacity of Emissions Escaping Air
Curtain 52
22 Summary of Visual Observations of Hood Capture
Effectiveness by Converter Event 55
23 Summary of Visual Observations Logs by Event 56
24 S02 Emission Summary 87
25 S0_ Emission During Matte Charges 88
26 SO Emission During Slag Skims 89
27 SO Emission During Copper Pour 91
28 SO Emission During Cold Additions 92
29 S0_ Emission During Standby, Blow, and Idle
Modes 93
30 S0_ Emission During Converter Rolls 95
31 S0« Emissions During Upset Conditions 96
32 Comparison of S0_ Emission During Normal
and Upset Primary Hood Operation 98
33 Summary of Filterable Particulate Emissions
Data 99
34 Particulate Emission Factor Development 102
35 Summary of Filterable and Gaseous Arsenic
Emission Data 103
36 Development of Arsenic Emission Factors 106
37 Particulate Loading and Impactor Flow Rate
Data for the Particle Size Runs 108
-------
TABLES (continued)
Number Page
38 Summary of Filterable Particulate Concentra-
tions for the Particle Size Runs 110
39 Summary of Filterable Particulate Emission
Rates for the Particle Size Runs 111
40 Summary of Inhalable Particulate Concentra-
tions During the Particle Size Runs 121
41 Summary of Inhalable Particulate Emission
Rates During the Particle Size Runs 122
42 Summary of Arsenic Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 126
43 Summary of Selenium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 127
44 Summary of Cadmium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 128
45 Summary of Antimony Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 129
46 Summary of Lead Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 130
47 Summary of Bismuth Concentration and Mass
Emission Rate for the Particle Size Runs
for the Charging Mode 131
48 Summary of Arsenic Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 135
49 Summary of Selenium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 136
50 Summary of Cadmium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 137
XI
-------
TABLES (continued)
Number
51 Summary of Antimony Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 138
52 Summary of Lead Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 139
53 Summary of Bismuth Concentration and Mass
Emission Rate for the Particle Size Runs
for the Skimming Mode 140
54 Summary of Arsenic Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 144
55 Summary of Selenium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 145
56 Summary of Cadmium Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 146
57 Summary of Antimony Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 147
58 Summary of Lead Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 148
59 Summary of Bismuth Concentration and Mass
Emission Rate for the Particle Size Runs
for the Blowing Mode 149
60 Comparison of the Total Particulate Concen-
tration Measured by the Particles Size Runs
to the Elemental Concentrations 151
61 Summary of Trace Metal Emission Results 153
62 Arsenic and Lead in Process Samples 155
63 Field Equipment Calibration 159
64 Example Blank Filter and Reagent Analysis 160
XII
-------
TABLES (continued)
Number Page
65 Arsenic and Selenium QA Audit 160
66 Comparison of Manual and CEM S0_ Results 167
67 Summary of SFr Background Analysis 168
b
68 Sample Matrix 171
Kill
-------
ACKNOWLEDGMENT
Overall project coordination and process observation were
performed by Messrs. John Burckle and Al Vervaert of the EPA's
Industrial Environmental Research Laboratory (Cincinnati) and
Office of Air Quality Planning and Standards. Mr. Frank Clay,
also of EPA, coordinated the transmissometer portion of this
project, and Messrs. Melvin Belich and Robert Budd represented
ASARCO during the test program and coordinated the scheduling and
process operations. Mr. James Nolan of the Puget Sound Air
Pollution Control Authority also observed the test program.
Messrs. Chuck Bruffey and Thomas Clark were the PEDCo project
managers. Principal authors were Messrs. Chuck Bruffey, Paul
Clarke, Thomas Clark, and Mark Phillips.
xiv
-------
SECTION 1
INTRODUCTION
During the period of January 14 through 22, 1983, personnel
from PEDCo Environmental, Inc., conducted an emission sampling
program designed to evaluate the effectiveness of the capture of
fugitive emissions by an air curtain hooding system on the No. 4
primary copper converter at the ASARCO smelter in Tacoma, Wash-
ington.
The primary objectives of the test program were:
1. To estimate the air curtain capture efficiency (overall
and during specific converter operational modes).
2. To estimate fugitive emission factors for the overall
converter cycle and specific operational modes for:
0 sulfur dioxide (S02)
0 filterable particulate
0 filterable and gaseous arsenic
0 selected trace elements (arsenic, cadmium,
lead, bismuth, antimony, and selenium)
0 Particle size distribution (inhalable par-
ticulate)
Three separate converter cycles were evaluated during the
test period. Emphasis was placed on obtaining representative
emissions data during the periods of heaviest generation of
fugitive emissions; i.e., matte charge and cold additions, slag
skimming, and copper pouring.
-------
The fugitive emissions collection efficiency of the air
curtain hooding system (secondary hood) was estimated by per-
forming a tracer gas study during the test program. The tracer,
sulfur hexafluoride (SFfi), was injected into the fugitive emis-
sion plume, and a gas chromatograph equipped with an electron
capture detector was used to measure the amount of tracer cap-
tured by the collection system. The tracer recovery efficiency
was then calculated on the basis of the material balance of inlet
and outlet tracer gas mass flow rates.
Visual observations were also performed by two observers
during the test program. The duration, location, and opacity of
visible emissions escaping the secondary hood were recorded. The
observers made estimates of the overall capture effectiveness
achieved during each mode of converter operation.
A Lear-Siegler Model RM4 transmissometer was installed at
the top of the air curtain hooding system to evaluate the overall
performance of the air curtain system in the control of visible
fugitive emissions. This performance evaluation was based on the
measurement and documentation of the opacity of fugitive emis-
sions escaping the air curtain during the converter operating
cycle. This evaluation was conducted simultaneously with the
other test programs conducted on the air curtain system. The RM4
transmissometer was chosen for this task because of its insensi-
tivity to environmental lighting conditions, which may bias
visible emission observers who are trained to read opacity in
sunlight against a contrasting background.
Filterable particulate and arsenic emissions were measured
by U.S. Environmental Protection Agency (EPA) Methods 5* and
108.** Two sampling trains were used to obtain the filterable
particulate and arsenic samples. Sampling was performed for the
duration of each converter cycle tested and during specific
converter roll-out modes—matte charge, slag skim, cold addi-
tions, and copper pouring.
*
40 CFR 60, Appendix A, Reference Method 5, July 1, 1982.
**
Method 108 has not been proposed and is in a draft form.
-------
Particle size distribution and subsequent inhalable particu-
late (IP) data were collected according to the guidelines in the
Procedures Manual for Inhalable Particulate Sampler Operation.*
Size distribution samples were collected separately during con-
verter blowing, charging, and skimming modes.
Trace element analyses were performed on particle size frac-
tions by atomic absorption analytical techniques. The trace
elements for which analyses were performed are arsenic, cadmium,
lead, antimony, selenium, and bismuth.
Sulfur dioxide captured by the air curtain system was con-
tinuously measured by a pulsed fluorescence monitor and also by
manual sampling.
Data collected during the test program indicate a 90 percent
or better fugitive emission collection efficiency is achievable
for the overall converter cycle and specific operating modes.
The capture efficiency for converter roll-in, roll-out, and slag
skimming operations showed more variability than other operating
modes. Visual observations indicate converter and crane opera-
tion are significant variables in the generation and capture of
fugitive emissions during those events.
Emission data collected at the secondary hood exhaust sample
location showed that fugitive emissions were generated primarily
during converter roll-in and roll-out activities and that arsenic
and lead were the major trace elements in these emissions. The
data also show the primary hooding system is very effective in
controlling emissions during the converter blowing mode.
Southern Research Institute. Procedures Manual for Inhalable
Particulate Sampler Operation. Prepared for EPA under Con-
tract No. 68-02-3118. November 1979.
-------
SECTION 2
PROCESS DESCRIPTION
Copper converting is a batch operation conducted in two
stages. Its purpose is to convert copper matte produced by a
smelting furnace [a complex mixture of copper sulfide (Cu_S),
ferrous sulfide (FeS), and trace elements] into blister copper.
In the first stage, iron removal is accomplished by blowing air
through the molten matte to oxidize the ferrous sulfide to fer-
rous oxides and eliminate sulfur as S0_. These oxides combine
with the silica flux which is added to the molten bath to produce
a molten iron silicate slag that is periodically skimmed from the
furnace. In the second stage, the air blowing is continued to
oxidize the copper sulfides to form blister copper and S0_. This
blister copper contains about 98.5 to 99.3 percent copper, 0.3
percent sulfur, and some dissolved oxygen and trace metal impuri-
ties.
2.1 CONVERTER OPERATION
The Peirce-Smith converter tested (designated No. 4) is a
horizontal, refractory-lined steel cylinder [4x9 meters (13 x
30 feet)] with an opening in the center (called the converter
mouth). The converter is mounted on rollers so it can rotate
through an arc of about 120 degrees from the vertical. Com-
pressed air or oxygen-enriched air is supplied through a header
along the back of the converter and passes through a horizontal
row of openings (tuyeres) in the shell into the interior.
Figure 1 is a schematic of the typical Peirce-Smith copper
converter. Figure 2 shows a typical converter and primary hood
section.
-------
TUYERE
PIPES
CONVERTER
MOUTH
PNEUMATIC
PUNCHERS
Figure 1. Typical Peirce-Smith primary copper converter.
-------
HOOD GATE
APRON
HOOD GATE
APRON
FLAPPER
TUYERES
RIDING RING
FLAPPER
TUYERES
RIDING RING
CHARGING
BLOWING
Figure 2. Converter with primary hood.
PRIMARY
HOOD
CHARGING
BLOWING
SKIMMING
Figure 3. Copper converter operation.
-------
Molten matte is charged through the converter mouth by
ladles carried by overhead cranes. During charging, the con-
verter is rotated to bring the converter mouth to an angle of
about 60 degrees from the vertical (as shown in Figure 3). The
ladles of matte, which are poured through the mouth of the con-
verter, fill it about halfway. Silica fluxing materials are also
charged through chutes and conveyor belts into the side of the
converter. When the converter is in the charging position, the
tuyeres are above the level of the matte. After charging, air or
oxygen-enriched air is supplied under pressure to the tuyere
line, and blowing commences. The converter is then rotated,
which swings the mouth to the vertical and submerges the tuyeres
to a depth of 6 to 24 inches below the surface of the matte. The
primary hood is then lowered over the mouth of the converter.
Emissions generated during blowing are captured by the primary
hood and are then routed to the SO recovery plant. The primary
hood operates only while air is being blown through the converter
tuyeres.
The blowing continues until a substantial layer of slag has
been formed in the converter. The primary hood is raised and the
converter is again rotated, which swings the mouth through an arc
of about 120 degrees from the vertical and raises the tuyere line
above the surface of the molten bath. The air supply to the
tuyere line is shut off, and the blowing is discontinued. Slag
is skimmed or poured into a ladle, and is recycled to the rever-
beratory furnace for recovery of entrained copper matte. The
converter is then returned to the charging position, and fresh
matte, fluxing materials, and cold supplements (such as smelter
reverts) are added to bring the converter charge back to the
working level. The converter is rotated again to the vertical
position, the primary hood lowered, and blowing is resumed.
This process is repeated until a charge of copper sulfide
(white metal) accumulates in the converter and fills it to the
working level. At this stage, the copper blow or finish blow
begins. During this portion of the converter operating cycle,
-------
the copper sulfide is oxidized and forms S0_ and blister copper.
On completion of the copper blow, the converter contains only
metallic copper (blister). The converter is rotated to the
pouring or skimming position and the blister copper is poured
into ladles for transfer to the anode furnace. The emptied
converter is then charged with fresh matte and fluxing materials,
and the converting cycle is repeated.
The copper converting process is autogenous; sufficient heat
is produced by the oxidation of sulfur to maintain the necessary
operating temperature of 1200° to 1260°C (2200° to 2300°F).
Consequently, no fuel or other source of heat energy is required.
More heat is released within the converter during the slag blow
than during the copper blow. The oxidation of one pound of
ferrous sulfide according to the following reaction releases
about 2742 kJ (2600 Btu):
2FeS + 302 + Si02 •»• 2FeO • Si02 + 2S02
while the oxidation of one pound of cupric sulfide according to
the following reaction releases only about 633 kJ (600 Btu):
Cu2S + 0_ -»• 2Cu + S0_
Thus, the amount of heat released during a slag blow is more than
sufficient to keep the bath in a molten state and compensate for
heat losses. Converter operators must control the converter
temperature to prevent damage to the refractory lining during the
slag blow. Smelter reverts and copper scrap are charged to the
converters to lower the converter temperature generated by the
excessive heat being released.
2.2 CONVERTER EMISSIONS
As gases laden with dust and fumes escape from the converter
mouth, they are captured by a large gas collection system (pri-
mary hooding system). Room air enters this system through a gap
between the hood and converter and water is sprayed into the duct
to cool the gases. The total process gas stream passes through a
-------
series of cyclones and two fans and is then divided and sent to
one of two gas cleaning systems consisting of electrostatic
precipitators, spray chambers, and packed and open scrubbing
towers. The gases are then ducted either to an SO recovery
plant or to a single-contact sulfuric acid plant.
When the converter is rolled out for skimming and charging,
the gate on the primary hood (see Figure 2) is moved up and away
from the mouth to provide clearance for the overhead crane and
ladle. The injection of blowing air continues until the molten
bath level is below the tuyeres to prevent the matte from enter-
ing the tuyeres and causing them to freeze. Significant amounts
of off-gases from the hot bath escape the primary hood system
during this period. (When the blowing air is turned off, the
primary hood draft also shuts off.) These emissions (fugitive
emissions) usually are greater when cold material such as scrap
is charged. Before the converter is rolled back to the blowing
position, the air is turned on again, which causes the resumption
of significant emissions.
The most significant fugitive emissions from a converter are
generated during the converter roll-out and roll-in modes, charg-
ing (including cold additions), slag skimming, and blister copper
pouring. These emissions consist primarily of particulate matter
and sulfur dioxide (S0?). Fumes present in the gases include
particulate oxides of arsenic, antimony, and lead, metallic
sulfates, and sulfur trioxide (SO..). At the ASARCO plant, the
fugitive emissions are captured by an air curtain hooding system
which is described below.
2.3 AIR CURTAIN HOODING SYSTEM
An air curtain is a horizontal moving "sheet" of air that
extends across an open space and decreases the passage of gases
on one side of the curtain to the other side. The air curtain is
created by blowing compressed air through a slot or nozzle. The
objective is to achieve absolutely uniform and parallel movement
of the air sheet with little or no turbulence. On the opposite
-------
side of the space, the moving air sheet is captured by a suction
plenum and air exhaust system.
Figures 4 and 5 illustrate the air curtain as applied to the
No. 4 Converter. The curtain directs fumes rising from the
source into the suction plenum. Room air is also pulled into the
curtain from both above and below. Because most fumes cannot
penetrate the curtain, they are collected by the suction plenum.
The collected emissions then pass through a suction fan, an
electrostatic precipitator for particulate removal, and finally
are released to the atmosphere through the main stack.
Air volume control for the system is regulated automatically
by dampers in the air curtain jet, the 1.5-m (5-ft) exhaust duct,
and the induced-draft fan. The dampers are manually set for a
predetermined exhaust-side flow and, when placed in the automatic
control position, are activated by movement of the primary hood
and converter. When the primary hood is lifted and the converter
is rolled out, the system switches to a high flow condition to
control the heavy amount of fugitive emissions generated during
roll-out activities (matte charging, slag skimming, cold addi-
tions, and copper pouring). At the completion of the converter
roll-out activities, the converter is rolled in and the primary
hood is retracted over the converter mouth. At this point, the
system switches to a lower flow condition because fugitive emis-
sions are much lower during the blowing and idling modes.
10
-------
FROM AIR JET FAN
AIR CURTAIN
PRIMARY HOOD
LOCATION
TO I.D. FAN
SEAL
Figure 4. Converter air curtain/secondary hooding
system (no scale).
11
-------
JET SIDE
EXHAUST SIDE
AIR
CURTAIN
JET
t
NO. 4 CONVERTER
(FUME SOURCE)
BAFFLE
WALL
TO SUCTION FAN AND
HOOD SAMPLE LOCATION
Figure 5. Air curtain control system.
12
-------
SECTION 3
PROCESS OPERATION AND TEST LOG
This section summarizes converter operations relative to
specific testing activities conducted from January 18 through 22,
1983.
The Test Plan for this project called for the evaluation of
a total of four complete converter cycles and certain segments of
each cycle. Production curtailments due to poor meteorological
conditions and operational problems with the reverberatory fur-
nace and sulfur dioxide recovery plant precluded obtaining data
for four complete cycles. Instead, three separate cycle segments
were evaluated, as shown in Table 1. Tables 2 and 3 summarize
specific sample events and the segments of each converter cycle
sampled.
Table 1 shows sequential converter operations from the
initial matte charge through the copper finish blow. A complete
cycle (from the initial matte charge through the copper pour)
normally takes about 12 hours. Because of the previously men-
tioned production curtailments, however, converter cycles were
frequently interrupted and seldom completed within the 12-hour
time frame during the period of testing activities.
As Table 1 shows, each converter cycle tested followed the
same basic format: each cycle began with the matte being charged,
followed by a series of slag blows, slag skims, and material
additions. This pattern was continuous through the cleanup blow.
The only difference between individual converter cycles was the
variation in the quantity and type (quality) of the materials
charged to the converter. This variation affected the amount of
slag produced, which is a function of the purity of the material
charged. It also altered the number of blows required to remove
13
-------
TABLE 1. CONVERTER CYCLE AND TESTS CONDUCTED
Cycle test
no. and
date (1983)
1
Jan. 18 (—)
Jan. 19 (««•)
Charge
1o 79
2
Charge
No. 80
3
I-. 99
Charge
No. 81
Converter
operation
sequence
SO - CEM
Paniculate/
Particle size
Hood capture
efficiency
(SF I
Initial
matte
charge
Tracer
No. 1
Skim -» charge
-*• additions
b
b
b
b
b
gas tests not |
No. 2
Skim • charge
•• additions
c
c
c
c
c
>erformed during
Slag blow
No. 3
Skim * charge
• additions
this cycle.
No. 4
Skim •• charge
• additions
No. 5
Sk in -* charge
* cold additions
NO 5 SLAG BLOW
IMS ELIMINATED IN
THIS CONVERTER
rvn F
Cleanup blow
Skim • cold
additions
Finish blow
Cold additions -
copper slag skim »
blister copper pour
'Start Cycle Test No. 1 on January 18. 1983.
bEnd Cycle Test No. 1 on January 18, 1983.
cStart Cycle Test No. 1 on January 19, 1983.
''End Cycle Test No. 1 on January 19, 1983.
-------
TABLE 2. SAMPLE MATRIX
$•»!• trtln type
P«rtlc»lite/«r»e«.lc
(continuous snvllno,
triln)
•*rtl«ltt*/iwnlc
specific Bode »•»-
pllno. trtln)
Mlrttn Nirt III
1*Vtctor (chir^lnf
•0*)
Mtrtn Mr* III
IXMCtor (>t1*Hnf
«*)
MtrM* P»rt III
tapKtor vltn !$-•»
•ode)
10, CmttnWM
•OTltor
Sccointonr hood
cipture efficiency
if. tricer 9as
Oncltjr gjlm t
trintaliioMter
DOT Kg.
P*TC-1
MTC-?
P»tC-3
nUH-l
WSH-?
P«JM-3
PSHC-i
KMC-?
PSK-]
KMC-4
nnc-s
«SS-1
nn-t
n$s-i
fSS-l
rw-?
rs§-j
ra-4
Crtle
le.t
•o.
1
t
3
1
t
3
1
1
;
2
)
1
t
3
1
t
I
3
Oil
fn-
llBl-
mrlei
•nd 1-?
1-3
Mtt
(IW3)
I/IB
1/19
l/?0
l/»
1/18
1/19
1/20
im
I/IB
1/19
I/I*
i/ro
1/20
1/22
I/IB
1/19
1/20
1/22
I/IB
1/19
1/20
1/20
1/22
l/ll-
1/22
1/1*
1/17
I/IB-
1/20
1/18-
1/22
5e9»«t» <«*>ted durlnq the cm>«erter cycle
Nittt clMrfi.
do. of Udlei
tested
14
14
4
14
14
0
14
2
10
4
4
I
1
Cold •ddltlont.
to. of lidlet
tetted
B
10«
1."
;
ioa
0
1
7
4
6«
M"
I
I
Anode end copper
•fair of the cold addition! "ere blocU of blister.
bSe»en of tlw cold iMItloni nere blocks of blister.
-------
TABLE 3. TEST LOG
Cycle
Test
No.
1
1
Dtte
(1983)
1/18
1/19
T1«e
(24-h)
0802
0810
0813-0907
0908-0916
0923-1046
1047-1110
1111-1212
1237-1250
1254-1259
1301-1302
1306
1315-1327
1341
1342-1510
1511
1513-1545
1545-1558
1616-1622
1627
0726-0729
0730-0809
0810-0833
0834-0838
0839-0844
0845-0915
0916-0934
0935-0937
0938-0941
0942-0944
0945-1009
1010-1013
1014-1029
1030-1036
1037-1126
1127-1135
1143-1229
1230-1235
1236-1316
1317-1321
1322-1400
Converter
event
Copper finish blow
Cold addition (ladle)
Copper finish blow
Copper slag skin
(11 ladles)
Idle
Copper finish blow
Blister copper pour
(10 ladles)
Natte charge (3 ladles)
Copper slag addition
(1 ladle)
Natte charge (1 ladle)
Anode slag addition
(1 ladle)
Natte charge (2 ladles)
Copper slag addition
(1 ladle)
Idle
Cold addition, shell slag
(1 ladle)
No. 1 slag blow
Slag skin (3 ladles)
Natte charge (3 ladles)
Converter on Idle until
1/19/83
Cold addition (1 ladle)
No. 2 slag blow
Slag ski* (3 ladles)
Natte charge (2 ladles)
Cold addition, shell slag
(2 ladles)
No. 3 slag blow
Slag skin (2 ladles)
Natte charge (1 ladle)
Cold addition (1 ladle)
Hatte charge (1 ladle)
No. 4 slag blow
Slag skin (1 ladle)
Cleanup blow
Slag skin (1 ladle)
Idle
Cold addition, scrap
(2 ladles)
Copper finish blow
Cold addition (2 ladles)
Copper finish blow
Cold addition, scrap
(2 ladles)
Copper finish blow
End Test No. 1
50,-
CEN
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Partlculat*/
arsenic PATC
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Partlculate/
arsenic PASN
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
six
Blowing
X
X
X
X
X
X
X
X
X
X
Particle
e dtstrlbu
Charging
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
tlon
Skinning
X
X
X
X
X
X
X
Hood
capture effi-
ciency (SFR)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Opacity
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
-------
TABLE 3 (continued)
Cycle
Test
No.
2
3
Date
(1983)
1/20
1/22
TIM
(24-h)
0820-0947
0948-1001
1002-1003
1014-1049
1050-1108
1111-1116
1117-1119
1120-1125
1126-1350
1351-1424
1425-1447
1448-1454
1455-1459
1500-1536
1537-1553
1554-1614
1615-1654
1655-1707
1708-1732
1733-1755
1756-1809
1810-1814
1815-1835
1836-1922
1923-2001
0907-0911
0912-0951
0952-1009
1010-1042
1043-1049
1050-1055
1056-1131
1132-1136
1137-1204
1205-1209
1210-1601
1602-1621
1622-1630
1631-1700
1701-1715
1716-1725
1726-1734
1735-1824
1827-1830
Converter
event
Matte charge (6 ladles)
Cold addition, shell slag
(2 ladles)
Matte charge (1 ladle)
No. 1 slag blow
Slag ski* (21 ladles)
Matte charge (3 ladles)
Cold addition, shell slag
(2 ladles)
No. 2 slag blow
Idle
No. 2 slag blow (continued)
Slag skim (3 ladles)
Matte charge (2 ladles)
Cold addition, shell slag
(2 ladles)
No. 3 slag blow
Slag skim (2 ladles)
Matte charge (2 ladles)
No. 4 slag blow
Slag skim (2 ladles)
Idle
Cleanup blow
Slag skim (1 ladle)
Cold addition, blister
copper (4 blocks)
Copper finish blow
Idle
Copper finish blow
End Test No. 2
Cold addition, shell slag
(2 ladles)
No. 3 slag blow
Slag skim (2» ladles)
Idle
Matte charge (2 ladles)
Cold addition, shell slag
(2 ladles)
No. 4 slag blow
Idle
Slag skim (3 ladles)
Matte charge (2 ladles)
Idle
No. 5 slag blow
Slag skim (2 ladles)
Idle
Cleanup blow
Slag blow (1 ladle)
Cold addition, blister
copper (5 blocks)
Copper finish blow
Cold addition, scrap wood
(1 ladle)
CEM
X
X
X
X
X
Participate/
arsenic PATC
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Partlculate/
arsenic PASM
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Particle
size distribution
Blowing
X
X
X
X
X
X
X
X
X
X
Charging
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Skimming
X
X
X
X
X
X
X
X
Hood
capture effi-
ciency (SFfi)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Opacity
X
X
X
-------
TABLE 3 (continued)
Cycle
Test
No.
Date
(1983)
TIM
(24-h)
1831-1834
1835-1900
1901-1930
1931-2001
2002-2032
2033-2034
2035-2109
2110-2114
2115-2116
2117-2207
2208-2214
2215-2302
2303-2338
Converter
event
Cold addition, blister
copper (2 blocks)
Idle
Copper finish blow
Idle
Copper finish blow
Cold addition
Copper finish blow
Slag skim
Cold addition, metal
powder and wood
(1 ladle)
Copper finish blow
Cu slag skim
Idle
Blister copper pour
(9 ladles)
End of Test No. 3
SO,-
CEM
X
X
X
X
X
X
X
X
X
X
X
X
X
Participate/
arsenic PATC
X
X
X
X
X
X
X
X
X
X
Paniculate/
arsenic PASH
Particle
size distribution
Blowing
X
X
X
X
Charging
X
X
X
Skimming
Hood
capture effi-
ciency (SFR)
Opacity
X
X
CD
-------
the slag and, consequently, the length of the converter cycle.
During the first and third converter cycles tested, five slag
blows were required to make the material in the converter pure
enough for the cleanup blow; whereas it took only four slag blows
during the second converter cycle tested.
The sulfur dioxide continuous emissions monitoring system
(SO -CEM) was in operation throughout each of the three cycles
evaluated. In fact, the SO--CEM commenced sample data collection
on January 14 and provided a continuous SO- emission character-
ization for the No. 4 Converter through January 22. The manual
emission tests (particulate/arsenic, particle size distribution,
tracer capture efficiency, transmissometer and visual observa-
tions) were performed during specific converter operations, as
shown in Table 1.
Tests for particle size distribution were performed with
Andersen Mark III in-stack particle size impactors. Size distri-
bution samples were collected during the converter charging mode
(matte and other additions), skimming mode (slag and blister
pour), and blowing mode (slag, cleanup, and finish). Table 2
shows the variation in the number of charges, pours, or blows
tested during each particle size run. Particle size samples were
obtained by compositing samples over several separate charge,
pour, or blow periods in order to obtain sufficient loadings on
the individual impactor stages.
The preliminary tracer recovery efficiency tests were per-
formed on January 14 and 17. The preliminary runs were designed
to provide data relative to tracer recovery efficiency as a
function of tracer injection location. These data are summarized
in Section 4 of this report. As shown in Tables 1 and 3, the
tracer recovery efficiency tests were conducted during most of
the operating modes of the first two converter cycles tested. No
tracer tests were performed on January 22. On January 20 between
.10:15 and 11:30 a.m., the primary hood damper system malfunc-
tioned. During this period, massive emissions escaped the pri-
mary hood during the blow mode on several occasions. Opacity
19
-------
measurements were obtained with a transmissometer during the
manual emission tests, as shown. Observations of visible emis-
sions were made by the two observers throughout the test period.
Tests for filterable particulate and arsenic were performed
with two separate, yet identical, sample trains. One train
(designated PATC) was run for the duration of each cycle tested
to provide a composite of all emissions over the entire cycle,
and the second train (designated PASM) was run only during oper-
ations when the converter was in a rolled out, or open, position
to provide a composite of emissions occurring during such activ-
ities as charging, slag skimming, and copper pouring. For cycle
Test No. 3, the PASM particulate/arsenic sample train was run
during slag skimming only in an effort to further characterize
skimming emissions.
20
-------
SECTION 4
AIR CURTAIN CAPTURE EFFICIENCY
The primary objective of the test program was to obtain an
estimate of the overall capture or collection efficiency of the
air curtain control system and its efficiency during specific
converter operational modes.
Because no absolute measurement procedure was available for
quantification of hood collection efficiency, several techniques
were used to estimate this efficiency. These techniques included
1) a tracer mass balance, 2) observation of visible emissions,
and 3) opacity measurements with a transmissometer. This section
describes the techniques used to quantify the air curtain collec-
tion efficiency and presents the test results. Example calcula-
tions are given in Appendix A. Field and analytical data are
presented in Appendix B. A detailed description of the sample
and analytical techniques is presented in Appendix D and the logs
of visible emissions observations are presented in Appendix H.
4.1 TRACER GAS MASS BALANCE
4.1.1 Principle
A suitable tracer was quantitatively injected at various
points within the air curtain control area. By combining the
measurements of the tracer concentration at the air curtain
exhaust sampling point with flow rate measurements, it was pos-
sible to calculate the amount of tracer passing the sampling
point (i.e., the amount collected by the air curtain and suction
plenum). The tracer recovery efficiency was then calculated from
the amount injected and the amount captured on a mass-flow basis.
21
-------
Sulfur hexafluoride (SFr) was used as the tracer. This
b
colorless, odorless, tasteless gas is not flammable, is nontoxic,
and is extremely inert. It is also stable up to a temperature of
500°C (932°F) . The minimum detection limit for SF,. (using a gas
b
chromatograph with electron capture detector) is 5 parts per
trillion. The SF. background level in ambient air is in the
-14
10 range.*
The SF.. was injected into the control of the air curtain at
b
a constant rate. Maintenance of a constant pressure on a limit-
ing orifice ensured that a constant injection flow rate was
achieved. The limiting orifice was calibrated before and after
each sustained injection by using a 0 to 10 cubic centimeter (cc)
scale bubble meter. Injection rates were 30 to 50 cc/min. Also,
the temperature at each injection point was monitored during each
SF- injection to ensure that decomposition did not occur.
b
Single point samples of the air curtain hood flue gas were
collected at a downstream sampling location by pulling a constant
rate at a point of average velocity in the exhaust duct into a
leak-free, 15-liter Tedlar bag. The sample bags were taken to
the onsite laboratory for immediate analysis by a Perkin-Elmer
Model 3920 gas chromatograph equipped with a Ni-63 electron
capture detector and a Valco gas sampling valve with a 1-ml
sampling loop.
For each injection calculation, the average of the SFg
injection rate in cc/min determined before and after each sus-
tained injection was used. The concentration of SF at the air
curtain hood sample location combined with the average gas flow
measurements in cc/min units yields the SF.. mass flow rate at the
b
sample location. The air curtain capture efficiency for each
injection point was then calculated as follows:
Q. Qm
E = l - ( 1 " m )
Qi
J. E. Lovelock, Nature, pp. 230-379 (1971).
22
-------
where:
E = SF.. capture efficiency
D
Q. = SF injection rate into air curtain control area in
1 cc/min
Q = SFfi mass flow rate at sample location, cc/min; where
Q = SF, concentration (part/part volume/volume) x
average volumetric flow rate (cc/min)
In an effort to define the accuracy of this method, the
following procedure was developed for determining relative error,
The definition for the term "percent error" as used here is
defined as follows. An associated or assumed error of ±10 per-
cent represents a standard deviation of ±3 a's (standard devia-
tion units).
The equation for calculating the tracer recovery efficiency
(E) is:
Q. Q
E = [1 - <-^-—-)] 100 (1)
where:
Q. = amount of the tracer substance injected into the
1 air stream (cc/min)
Q = amount of the tracer substance measured (cc/min)
Equation 1 can also be expressed as:
O 2
E = (—) x 10 (2)
The relative error of the recovery efficiency [RE(E)] is ex-
pressed as:
RE(E) .
23
-------
where:
v
Var(E) = variance of E and
Var(E) = -i^- [-yVar(Q.) +Var(Qj] (4)
Q. Q.
i i
Var(Q.) Var(0 )
RE(E) = [ 5-i- + j5i—] (5)
Qi Qm
From Equation 1,
Q. = the amount of SF, injected (6)
and Qm = C F (7)
C = ppb, concentration of tracer measured at down-
stream sample location
F = cc/min, mass flow rate at collection point
Using proper calibration and sample techniques, it is as-
sumed that the error in the SF, injection rate is ±5 percent. It
is further assumed that the error in the SF, concentration mea-
b
surement and in the gas flow measurement (F) (using EPA Method
2*) is ±10 percent.
Based on average data values obtained during the tracer
study, typical values used in this calculation would be:
Q. = 45.00 cc/min SF,
l b
RE(Q.) = ±5 percent
and
F = 126,230 acfm = 3.58 x 10 cc/min
C = 1.2 x 10~8 part/part SFg
Q^ = 42.96 cc/min
RE(c) = ±10 percent
RE(f) = ±10 percent
*
The EPA Program for the Standardization of Stationary Source
Emission Test Methodology, A Review, EPA-600/4-76-044.
24
-------
Equations 4 and 5 become
a RE(Qi) 2
Var(Q.) = Q. [ —] (8)
1 1 3
Var(Qm) =C2 [
m 3 3
For the typical values used here, the relative error of the
recovery efficiency is determined as follows:
Var(Qi) = (45) * (^^-} *
= 0.56
Var(Qm) . (1.2}' [I0.10M35.8))' + (35.8, * [(0.10H1.21 ,'
= 4.10
RF/EI - r°-56 + ( 4-io n %
RE(E) - [(45)2+ <(35.8)2n
= ±0.059
= ±5.9%
Because the error of the measurements is assumed to repre-
sent the interval equivalent to 3 standard deviations, the rela-
tive error in tracer recovery efficiency measurements would be
approximately ±18 percent (i.e., 3 x 5.9%).* Tracer recovery
data reported in the following sections are subject to the ±18
percent error limits.
4.1.2 Preliminary Method Evaluation
Before tracer gas tests could be performed on the air cur-
tain hooding system, it was necessary to evaluate available
sample and analytical methodologies relative to this program.
Literature searches were made to ascertain the compatibility of
*
Bennett, C.A., and N.L. Franklin, Statistical Analysis in
Chemistry and the Chemical Industry, John Wiley & Sons, Inc.,
1954.
25
-------
the use of sulfur hexafluoride as a tracer in previous studies
with the planned application on the air curtain hooding system.
The GC-ECD was assembled specifically for SF.. analysis
6
according to referenced procedures.* A series of calibration
runs was made to evaluate instrument response and accuracy.
These preliminary runs compared favorably with available ana-
lytical data.* An evaluation also was performed to determine if
tracer losses occurred in the Tedlar bags used to collect the
actual flue gas samples. A known concentration of sulfur hexa-
fluoride was placed in sample bags and analyzed after a 24-hour
period. No loss of sulfur hexafluoride was observed. The sample
bags were then purged with nitrogen and analyzed for residual
sulfur hexafluoride to evaluate the effects of reusing sample
bags for multiple sample collection. In each case, the sulfur
hexafluoride tracer concentration was below the minimum detection
-12
limit (5 x 10 ) of the analytical method.
4.1.3 Tracer Method Feasibility Tests
In December 1982 PEDCo performed a series of tracer gas
tests on the converter air curtain secondary hood to evaluate the
feasibility of the tracer gas method relative to establishing a
reproducible data base for tracer recovery efficiency and to
evaluate the most practical means of obtaining the desired data.
Initially, a series of tests were performed in which SF, was
quantitatively injected directly into the secondary hood exhaust
duct downstream of the suction plenum. Flue gas samples were
collected at the secondary hood test location and the combined
tracer concentration measured at the sample location and flue gas
flow measurements yielded the mass flow of tracer at the sample
location. Tracer recovery efficiencies were then calculated as
described in Subsection 4.1.1. These data, summarized in Tables
4 through 6 (see pages 69 through 71), were used to quantify
*
Improvements in the Determination of Sulfur Hexafluoride for
Use as a Meteorological Tracer. Analytical Chemistry, Vol. 44,
No. 4, April 1972; and SF6 Tracer Gas Analysis of Mine Ventila-
tion Systems, Technology News, No. 143, May 1982.
26
-------
inherent measurement biases and establish the most practical
means of obtaining the data.
At the completion of these tests, the tracer was quanti-
tatively injected directly into the area above the No. 4 con-
verter to establish the effective hood control area and to eval-
uate measurement techniques for the planned manual test program
in January 1983. These data are summarized in Tables 7 through 9
(see pages 72 through 75).
In summary, the December 1982 tracer test data indicated
that the methodology as applied to the air curtain hood system
provided a feasible means of estimating air curtain capture
efficiency. The suitability of sulfur hexafluoride as the tracer
gas was also verified. The following sections detail the December
1982 test activities.
To establish the tracer recovery efficiency and to account
for any inherent measurement biases, the tracer was quantitative-
ly injected directly into the air curtain secondary hood exhaust
duct downstream of the secondary hood suction plenum. Flue gas
samples were collected at the secondary hood exhaust sample
location approximately 76 meters (250 feet) downstream from the
injection point. The combined tracer concentration measured at
the sample location and the flue gas flow measurements yielded
the mass flow rate of tracer at the sample location. Collection
efficiencies were then calculated by comparing the amount of
tracer injected per minute and the mass flow rate of the tracer
passing the sampling location. Tables 4 through 6 summarize the
results of the preliminary tracer recovery tests.
Test Runs 1 through 4 (Table 4) were trial runs designed to
establish the injection rate and subsequent tracer concentration
measurable by the GC-ECD and an exponential dilution calibration
curve. As shown in Table 4, the tracer injection rate was in-
creased for each of the first four tests until the measured
concentration was in the proper range. For these four tests, two
samples were collected simultaneously at two points of average
27
-------
velocity in the duct. For tests 5T through 14T integrated sam-
ples were collected by traversing the cross-sectional area of the
duct. Gas flow rates reported for these runs are actual flow
rates measured during the collection of each sample. Recovery
efficiencies ranged from 74 to 94 percent for the integrated
samples, and the average tracer recovery efficiency was 84 per-
cent.
Table 5 summarizes the results of samples collected at a
single point in the secondary hood exhaust duct. During these
tests, the tracer was injected into the air curtain exhaust duct
downstream of the secondary hood suction plenum. Single point
sampling was desirable because more samples could be collected in
a given amount of time. Also, the amount of test equipment
required at the sample location to make other manual tests pro-
hibited the collection of integrated gas samples. The tracer
recovery efficiency ranged from 79 to 97 percent, and the average
recovery efficiency was 89 percent. The gas flow rates reported
in Table 5 and subsequent calculations represent average values
obtained for each flow mode as described below.
Air volume for the air curtain control system is regulated
by a series of dampers activated by limiting switches during
movement of the converter and primary hood. For the December
1982 tests, three separate damper settings and subsequent exhaust
gas volumes were used in the operation of the air curtain system.
Table 5 presents the average volumetric flow rates measured by
EPA Reference Methods 1 and 2* at the air curtain hood exhaust
test location. Velocity measurements were made for each flow
mode to determine an average volumetric flow rate for each set-
ting. Volumetric flows were calculated in actual cubic meters
per minute and cubic feet per minute and then converted to cubic
centimeters per minute for the tracer mass flow calculations.
Stack gas moisture content was less than 1 percent by volume as
determined using wet bulb-dry bulb techniques.
*
Stack gas composition was ambient; 20.9% oxygen, 0.0% carbon
dioxide, with the balance being nitrogen.
28
-------
Before the single point recovery tests were performed, flue
gas samples were collected at each of the 12 sample points (EPA
Reference Method 1) to define the SF., concentration profile in
b
the exhaust duct. Figure 6 presents the SFr concentration pro-
fa
file.
ID'8
»-
DC
«
o.
>.
i—
oc.
2
1,0"
2
z
Ul
<_)
to
^^
«" 1 «
ID'10
(
O fi A A A
SECONDARY EXHAUST DUCT
tr^P OA PORT
A CROSS - SECTION
A
TRAVERSE POINT
123 H 5 6
111 * i I i * i * T
) 10 20 30 40 50 60
TRAVERSE POINT LOCATION. INCHES
(FROM INSIDE PORT MALL)
Figure 6. SFg concentration profile.
Negative pressure at the sample location causes air inleak-
age and dilution near the sample port. This dilution caused
distortion of the SF.. concentration at Test Points 1 and 2 (both
b
sample ports), which explains the lower recovery efficiencies on
the integrated samples (Table 4). The remaining points showed
uniform SF,. concentration distribution.
b
Table 6 presents a summary of tracer recovery efficiency at
the air curtain suction plenum. Injections were made directly
into the suction plenum, and six gas samples were collected at a
single point in the duct. Recovery efficiencies ranged from 91
to 97 percent and averaged 94 percent.
29
-------
The tracer recovery efficiency data presented in Tables 4
through 6 indicated that collection efficiencies were repro-
ducible and comparable regardless of the sample collection tech-
nique (single point versus multi-point sample collection) used.
As noted previously, single point sample collection was desirable
based on the number of samples which could be collected in a
short amount of time. Average recovery efficiencies ranging from
84 to 94 percent indicate a measurement bias (to the low side)
was prevalent for these samples. However, considering the ±18
percent relative error associated with the measurement technique,
the data do provide a good estimation of tracer collection effic-
iency.
Based on the recovery test data presented in Tables 4
through 6, preparations were made to begin injections into the
area immediately above the No. 4 converter. As shown in Figures
7 and 8, sample ports were located in access doors on both sides
of the air curtain wall so that a sample matrix could be construc-
ted immediately above the No. 4 converter. Starting on the
curtain exhaust side, injections were made at four sample points,
placed at approximately 0.6-m (2-foot) intervals, in each of the
ports designated in Figure 8. Sustained injections were made in
each port, and bubble meter calibrations of the injection system
were performed before and after each injection. At each injec-
tion point the temperature was recorded to preclude possible
tracer decomposition. Temperatures within the measurement area
(Figure 7) ranged from 13°C (55°F) to 315°C (600°F) and averaged
value of 37°C (96°F). Based on the temperatures measured in the
control area, thermal decomposition of the tracer did not occur.
A single sample was collected at the exhaust duct sample
location for each injection point. The velocity head was re-
corded for each sample collected, and based on the converter
operational mode and fan damper setting, the average gas flow
rate determined for that damper setting was used in the collec-
tion efficiency calculations. A similar procedure was used on
the jet side of the curtain, except that two points (Nos. 5 and
6) were used to complete a six-point cross-sectional traverse.
30
-------
EXHAUST SIDE
JET SIDE
CONVERTER
CONVERTER
ArSLE FLOOR
Figure 7. Graphical presentation of air curtain sample ports.
31
-------
CONVERTER
" AISLE
AIR CURTAIN
EXHAUST
LIGHT [~|
A
C-l C-2 C-3
XXX
C-4 C-5 C-6
XXX
-L
B-l B-2
X X
B-3 B-4
X X
A
/
D-3' D-4
X X
f
PRIMARY
HOOD
\
\
\
\
AIR CURTAIN JET
B-2
X
B-4
X
K
B-l
X
B-3
X
LIGHT
A
PRIMARY
HOOD
\
D-2
x \
D-4
X
D-l
X
*D-3
X
\
C-3 C-2 C-l
XXX
C-6 C-5 C-4
XXX
CONVERTER
AISLE
Figure 8. Injection point matrix used for preliminary
tracer gas tests.
32
-------
Tables 7 and 8 summarize the results of the air curtain sample
matrix test results.
Table 9 presents a summary of the tracer recovery effi-
ciencies and corresponding converter operation modes. No attempt
was made to evaluate recovery efficiencies at the same sample
point for the various converter modes. Also, several matrix
points (Port D4, Points 1-6) were not evaluated because they were
inaccessible. Matrix points Cl-2, C5-2, C5-5, and Dl-5 were
evaluated during a change in converter operation, which resulted
in fluctuating volumetric flow rates and inaccurate tracer mass
flow calculations. These data are not included with the reported
results.
A total of 76 SF.. injections were made in the three-dimen-
o
sional matrix previously described. Table 10 summarizes the
results by sample port, relative position within the matrix area,
and converter operating mode. Tracer recovery efficiencies
ranged from 73 to 113 percent, and the overall average was 87
percent. After a review of these data, it was decided that the
number of injection points in the matrix could be reduced for the
January 1983 test effort.
4.1.4 Estimated Air Curtain Hood Capture Efficiency Using a
Tracer Gas
Based on the December 1982 field work, the following con-
clusions were drawn relative to achieving an estimation of air
curtain capture efficiency.
1. The tracer gas methodology provided a feasible means of
estimating air curtain capture efficiency,
2. The suitability of sulfur hexafluoride as the tracer
gas was verified, and
3. The air curtain control volume was established.
This section summarizes results from the January 1983 test
program in which additional tracer recovery tests were conducted
in the upper and lower portions of the air curtain control volume,
The objective in performing tracer recovery tests in the lower
33
-------
TABLE 10. COLLECTION EFFICIENCY OF SF, WITHIN
THE AIR CURTAIN MATRIX AREA (DECEMBER°1982)
Port
B-l
B-2
B-3
B-4
Average
C-l
C-2
C-3
C-4
C-5
C-6
Average
D-l
D-2
D-3
Average
Grand
average
Position
1
Exhaust
83
84
82
88
84.2
90
76
75
84
110
104
89.8
84
78
82
81.3
86.2
2
83
80
83
90
84.0
92
77
73
86
99
91
86.3
91
80
82
84.3
85.2
3
83
81
82
99
86.2
84
76
95
87
90
98
88.3
79
78
83
79.7
85.2
4
81
82
79
83
81.2
85
77
76
92
95
90
85.8
74
91
83
82.7
83.7
5
91
94
83
99
91.8
89
95
93
91
96
94
93.0
110
.
94
102
94.1
6
Jet
80
88
97
84
87.2
90
75
73
78
83
82
80.2
96
_
113
104.5
86.8
Average
83.5
84.8
84.3
90.5
85.8
88.3
79.3
80.3
80.8
95.5
93.2
87.2
89.0
81.8
89.5
86.7
86.8
Note:
SFg recovery efficiencies are subject to a ±18 percent relative
error as described in Subsection 4.1.2.
34
-------
portion of the control volume was to evaluate the effectiveness
of the secondary hood in controlling emissions below the converter
mouth and near the front edge of the baffle walls. Also, this
data would indicate if thermal buoyancy or lift increased collec-
tion efficiency during specific converter roll-out modes. These
data, in conjunction with visual observations and the transmis-
someter opacity data, were then used to estimate hood capture
efficiency; overall and for specific converter operating modes.
Prior to the start of the test program, air flow measure-
ments were performed in the air curtain hood exhaust duct at the
sampling plane according to procedures described in EPA Methods 1
and 2. These measurements were used to establish baseline volu-
metric flow data for the automatic damper system associated with
the air curtain control system.
Before the evaluation test program was started, ASARCO per-
sonnel made adjustments to the damper settings in an effort to
optimize volumetric flow for the test program. The flow patterns
established during the December 1982 preliminary tests were
changed as follows: 1) the high flow mode was reduced to 3540
actual m3/min (126,230 acfm), 2) the medium flow mode was elimi-
nated, and 3) the low flow mode was set at 2124 actual m3/min
(75,500 acfm). In practice, the low flow condition was used for
blowing and idling, and the high flow condition was used for
converter roll-out activities, i.e., charging and skimming.
Table 11 summarizes the volumetric flow data obtained at the
air curtain exhaust sample location. No further adjustments were
made to the suction fan setting during the sampling period. The
average values for each flow condition were used in all tracer
gas mass balance calculations, depending on converter operational
mode at the time of tracer sample collection. The average vol-
umetric flow rate for the low flow condition was 2124 actual
m3/min (75,505 acfm) with an average static pressure of -5.3 imnHg
(-2.86 in.HaO). The average volumetric flow rate for the high
flow condition was 3552 actual m3/min (126,230 acfm) with an
average static pressure of -15.6 mmHg (-8.38 in.H20). Based on
35
-------
TABLE 11. SUMMARY OF VOLUMETRIC FLOW DATA
Run No.
¥-6
¥-7
¥-8
¥-9
¥-10
¥-11
¥-17
¥-18
¥-19
¥-20
¥-21
¥-22
¥-23
¥-24
Date
(1983)
1/12
1/12
1/12
1/12
1/12
1/13
1/13
1/13
1/13
1/13
1/13
1/14
1/18
1/19
Average
¥-12
¥-13
¥-14
¥-15
¥-16
¥-17
1/13
1/13
1/13
1/13
1/13
1/13
Average
Flow
condition
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
High
High
High
High
High
¥o1umetric flow rate6
Actual
m'/min (acfm)
2115 (75,182)
2075 (73,748)
2122 (75,427)
2179 (77,452)
2144 (76,194)
2138 (75,977)
2121 (75,410)
2084 (74,406)
2105 (74,797)
2084 (74,051)
2155 (76,589)
2188 (77,768)
2099 (74,616)
2123 (75,446)
2124 (75,505)
3530 (125,475)
3577 (127,123)
3521 (125,148)
3621 (128,685)
3526 (125,322)
3535 (125,624)
3552 (126,230)
Standard
dNtn'/min (dscfm)
2125 (75,527)
2083 (74,040)
2132 (75,757)
2187 (77,716)
2150 (76,429)
2175 (77,311)
2171 (77,156)
2141 (76,086)
2153 (76,504)
2130 (75,686)
2202 (78,274)
2244 (79,752)
2097 (74,535)
2089 (74,249)
2149 (76,359)
3577 (127,124)
3620 (128,658)
3569 (126,855)
3671 (130,482)
3577 (127,134)
3413 (121,288)
3571 (126,924)
Static
pressure ,
mmHg
(in.H,0)
- 5.6 (-3.00)
- 5.6 (-3.00)
- 5.0 (-2.70)
- 5.0 (-2.70)
- 5.0 (-2.70)
- 5.6 (-3.00)
- 5.2 (-2.80)
- 5.4 (-2.90)
- 5.6 (-3.00)
- 5.4 (-2.90)
- 5.2 (-2.80)
- 5.6 (-3.00)
- 4.7 (-2.50)
- 5.6 (-3.00)
- 5.3 (-2.86)
-15.7 (-8.40)
-15.5 (-8.30)
-15.3 (-8.20)
-15.7 (-8.40)
-15.3 (-8.20)
-16.4 (-8.80)
-15.6 (-8.38)
Flue
gas tempera-
ture, °C (°F)
14 (57)
14 (57)
14 (57)
14.4 (58)
14.4 (58)
15 (59)
13 (56)
13 (56)
13 (56)
14 (57)
14 (57)
11 (52)
11.7 (53)
16 (61)
14 (57)
12 (54)
12.8 (55)
12 (54)
12 (54)
12 (54)
17 (63)
13 (56)
G«s d
composition
0,
20.9
20.9
20.9
20.9
C02
0.0
0.0
0.0
0.0
Moisture
content, %
(% by volume)
-1.0 (0.76)
<1.0
-1.0 (0.76)
<1.0
Static pressure
monitored during
manual emission tests,
mmHg (in.HjO)
- 4.5 (-2.4)
- 5.2 (-2.8)
- 4.7 (-2.5)
-
- 5.6 (-3.0)
-
- 5.4 (-2.9)
- 5.6 (-3.0)
- 5.0 (-2.7)
-
- 5.2 (-2.8)
- 4.7 (-2.5)
-
- 5.10 (-2.73)
-15.9 (-8.5)
-16.8 (-9.0)
-15.7 (-8.4)
-15.5 (-8.3)
-15.7 (-8.4)
-15.7 (-8.4)
-15.9 (-8.5)
U>
CTv
'Flow condition related to converter mode. Low - all blowing modes, converter hold and idle; High - matte charge, slag skim, cold additions, copper pouring,
and occasionally for converter Idle.
b¥olumetric flow rate in actual cubic meters per minute (actual m'/min) and actual cubic feet per minute (acfm) at stack conditions using 1 percent moisture
by volume and 28.84 dry molecular weight. Also, dry normal cubic meters per minute (dNmVnrin) and dry standard cubic feet per minute (dscfm) at 20°C (68°F)
and 760 mmHg (29.92 in.Hg) and zero percent moisture.
cStatic pressure obtained using a 0-36 in.HjO water manometer.
Gas compostIon as determined using a Fyrite gas analyzer and an Orsat gas analyzer as described in EPA Reference Method 3.
'Moisture content as determined using wet bulb-dry bulb techniques. Value in parentheses 1s average measured percent moisture from the manual emission
tests for particulate. Average wet bulb temperature: 12°C (54°F); average dry bulb temperature: 14°C (58°F).
fSt.attc pressure as monitored during SF, sample collection and manual emission tests, 1/18 through 1/22, 1983. Data represent values recorded on the SFfi
sample collection data sheets - Appendix B.
-------
previous data, moisture content of the gas stream was estimated
to be 1 percent and the dry molecular weight was estimated to be
28.84 (ambient) for calculation purposes. This value was veri-
fied by wet bulb/dry bulb measurements and an Orsat gas analyzer
for gas composition (oxygen and carbon dioxide). The flue gas
static pressure was measured using a 0-36 in. water manometer.
The static pressure was closely monitored throughout the test
series to ensure that the volumetric flow did not change signif-
icantly. As the data in Table 11 show, the static pressures
monitored during the manual field tests compared favorably with
those measured during the volumetric flow determinations. The
data also show that no significant change occurred in the vol-
umetric flow during the test series.
Figures 9 and 10 present profiles of the air curtain exhaust
duct velocity for each flow mode. Average values (by traverse
point) for measurements made during a given flow mode were used
to construct these profiles. The tracer samples were collected
at Sample Point 3 in both ports. A PEDCo representative (in the
converter building) coordinated the sample activities and the
monitoring of velocity pressures at the sample collection point.
Figure 11 shows the tracer injection locations used during
this manual test program. Tracer injection locations were varied
to establish tracer recovery efficiency by characterizing the
effective capture volume of the system. A sample matrix (de-
scribed in Subsection 4.1.3) was used to locate multiple injec-
tion points within the upper control volume. Sample ports in
adjacent access doors on both sides of the converter baffle walls
were used to construct the sample injection matrix (Figure 12).
The results of this sampling were used to characterize the tracer
recovery efficiency of the air curtain hood system in the area
immediately above the converter. Several single point injections
were performed in the lower portion of the control volume (Figure
11) to further characterize the effective capture volume of the
hooding system, particularly during converter roll-out activi-
ties. The results from these injections would indicate if the
37
-------
1.6
.
Sac
^ 1 3
>- o» ' • J
OJ
00
l.l
l.o
1.45
1.39
1719
j'3 AVERAGE Ap (inches H20) ALL POINTS: 1.33
2 3 4 5 6
0 2.6 8.8 17.8 42.2 51.2 57.4 60
SAMPLING POINT LOCATION, inches
Figure 9. Velocity profile for low flow condition.
-------
OJ
4.5F
o
CM
v, 4.0
3.5
3.0
4J5
3.10 AVERAGE p (Inches
^3.07
123
1 _L J
ALL POINTS: 3.69
0 2.6 8.8 17.8
4.23
PORT C
PORT D
5
1
3.38
6
I
42.2 51.2 57.4 60
SAMPLING POINT LOCATION, Inches
Figure 10. Velocity profile for high flow condition.
-------
LEGEND:
AREA SAMPLED USING
MATRIX TRAVERSE
INJECTION LOCATIONS
SAMPLE I.D.
V SP1 & 2
D SP3 - 5
• SP7 - 12
O SP13 - 71
TOP VIEW
JET SIDE
AIR
CURTAIN
JET
GRADE
BAFFLE
WALL
EXHAUST SIDE
AIR CURTAIN
NO. 4 CONVERTER
(FUME SOURCE)
BAFFLE
WALL
TO SUCTION FAN
ELEVATION
Figure 11. SFg injection locations.
40
-------
CONVERTER AISLE FLOOR
O MATRIX INJECTION POINTS
Figure 12. Tracer injection matrix.
41
-------
secondary hood was effective in controlling the area below the
converter mouth and near the edge of the baffle walls. Also this
data would indicate if thermal lift increased the collection
efficiency during converter roll-out.
The planned matrix tests called for injection of tracer at
16 points in a three-dimensional configuration above the top of
the converter. The matrix injections were to be made through
Ports B-2, C-l, C-6, and D-l at four positions each. The matrix
design provided for three injections at each injection point on
each of two days, for a total of 96 test points. The resulting
data were to be evaluated on site. Based on this evaluation, the
number of injection points within the sample matrix were to be
increased or reduced according to the statistical differences
between injection location and converter operational mode for
similar injection points. As noted previously, process opera-
tional problems made it impossible to complete the test series as
planned, and several changes had to be made to compensate for the
inconsistent process operation.
On January 14, 1983, it was possible to determine recovery
efficiencies for 45 of the 48 planned tests. Recovery efficien-
cies are presented in Table 12, and pertinent test data for each
sample injection are summarized in Table 13 (see pages 76 and
77). Average recovery efficiencies for individual releases
varied from 69 to 119 percent, and the overall average efficiency
for the 45 tests was 94 percent. The port through which the
releases of the tracer substance were made did not have any
effect on the average recovery efficiency. The average recovery
efficiency of all releases made through a given port ranged from
93.0 percent for Port C-6 to 95.4 percent for Port C-l. This
difference was not statistically significant. The variability
between the average recovery efficiency of the replicates made at
a given position (between the jet side and the exhaust side) was
statistically significant. The greatest difference occurred at
Port D-l, where the average recovery efficiency ranged from 83.3
42
-------
TABLE 12. TRACER RECOVERY EFFICIENCY WITHIN THE AIR
CURTAIN CONTROL (MATRIX) AREA (JANUARY 14. 1983)
Port
B-2
B-2
B-2
Average
C-l
C-l
C-l
Average
C-6
C-6
C-6
Average
D-l
D-l
D-l
Average
Position
Exhaust
1
97
89
94
93.3
97
95
97
96.3
95
101
94
96.7
91
89
93
91.0
2
102
95
97
98.0
93
105
95
97.7
94
93
97
94.7
105
119
93
105.7
Jet
3
94
89
79
87.3
95
97
96.0
96
89
92.5
98
98
98.0
4
97
96
94
85.7
94
91
90
91.7
93
81
90
88.0
91
90
69
83.3
Grand
average
Average
97.5
92.2
91.0
93.6
94.7
96.5
94.8
95.4
94.0
92.8
92.5
93.0
96.2
99.3
88.2
94.2
94.0
Mote: SFs recovery efficiencies are subject to a +18 percent relative error
as described in Subsection 4.1.2.
43
-------
to 105.7 percent. The recovery efficiencies for Positions 1 and
2 (near the exhaust side) were approximately 96.6 percent and
were generally higher than those for Positions 3 and 4 (near the
jet side) which were approximately 91.6 percent.
Average tracer recovery efficiencies for the various con-
verter-operating modes are also presented. With the exception of
cold additions, the average recovery efficiency was not affected
by the operating mode of the converter; averages varied from 92.8
during blowing to 95.0 during slag skimming.
The remaining test series of 48 injections were performed on
January 17, 18, and 19. The results of these tests are sum-
marized in Table 14, and pertinent test data for each sample
injection are summarized in Tables 15 through 17 (see pages 78
through 80). For this series of tests, the tracer recovery
efficiency varied from 65 to 119 percent and the overall average
was 96.0 percent. The port through which the tracer releases
were made had no effect on the average tracer recovery efficiency
of the air curtain hood. The average efficiency varied from 94.5
percent at Port C-6 to 98.0 percent at Port B-2. For positions
within the matrix, the average recovery efficiency varied from
80.7 percent at Position 4, Port D-l, to 106 percent at Position
2, Port D-l. Recovery efficiencies were consistently higher for
Positions 1 and 2 (near the exhaust side) than for Positions 3
and 4 (near the jet side). Figures 13 and 14 present these data
graphically.
Again, operating mode had no adverse effect on the recovery
efficiency measured for the air curtain hood by the tracer method.
Average collection efficiencies varied from 88.5 percent during
copper pouring to 96.7 during cold additions.
In summary, the results of the tracer recovery efficiency
data within the upper portion of the air curtain control area
indicate that, on the average, at least 90 percent of the gases
and particulate matter entering this area is likely to be cap-
tured by the air curtain hooding system. Both the December and
44
-------
rT«LE:.14. SFfi COLLECTION EFFICIENCY WITHIN THE CONTROL
(MATRIX? AREA (JANUARY 17, 18, 19, 1983)
Port
B-2
B-2
B-2
Test
date
(1983)
1/17
1/18
1/19
Average
C-l
C-l
C-l
1/17
1/18
1/19
Average
C-6
C-6
C-6
1/17
1/18
1/19
Average
D-l
D-l
D-l
1/17
1/18
1/19
Average
Position
Exhaust
1
103
95
101
99.7
105
95
99
99.7
104
93
96
97.7
106
101
106
104.3
2
99
107.5
103
103.2
103
96
103
100.7
102
106.5
99
102.5
106
111
97
106.0
Jet
3
95
81
119
98.3
99
84
104
95.7
101
95
93
96.3
96
92
89
95.0
4
93
87
92
90.7
93
91
69
84.3
81
79
95
81.7
65
88
89
80.7
Overall
average
Average
97.5
92.6
103.8
98.0
100.0
91.5
93.8
95.1
97.0
93.4
93.2
94.5
93.2
98.0
98.2
96.5
96.0
Note: SF5 recovery efficiencies are subject to a +18 percent relative error
as described in Subsection 4.1.2.
45
-------
JET
no
EXHAUST
100
U-l -. _
~ 90
80
70
3 2
POSITION
Figure 13. Comparison of hood collection efficiency
and matrix port injection.
46
-------
no
100
90
8 80
70
EXHAUST
JET
POSITION 2
POSITION f
POSITION 3
^•7 POSITION 4
^^
— — — -T7*" "**
j
B-2
PRIMARY HOOD-
D-l C-6
PORTS
C-1
-CFRONT
Figure 14. Comparison of hood collection efficiency
and matrix point injection.
47
-------
January matrix test results verify this conclusion. As indi-
cated, the converter operating mode had no adverse effect on the
SFg control area collection efficiency. Table 18 shows a summary
of the tracer recovery efficiency for the upper control volume.
TABLE 18. SUMMARY OF TRACER RECOVERY EFFICIENCY FOR UPPER CONTROL VOLUME
Converter
cycle
Number of
injections
Mean, %
Standard
deviation, %
Blow
19
27
92.8
96.7
8.47
10.06
Slag
skim
9
7
95.0
94.3
4.87
8.08
Matte
charge
7
6
93.1
94.2
3.13
9.54
Cold
add
3
3
102.0
96.7
14.7
20.4
Idle
7
4
93.4
100.0
3.6
11.4
Copper
pour
4
88.5
9.6
The first number in each pair is the result of Test 1 and the second of Test 2.
Since injections were made within the upper portion of the air
curtain control area, these data would not account for spillage
outside the upper control area during the injection and would not
necessarily correspond to an overall visual assessment of hood
capture effectiveness. Data relating visual assessment of hood
capture effectiveness to tracer recovery efficiencies are pre-
sented in Subsection 4.3 of this report.
In an effort to characterize the effective capture area of
the air curtain hooding system, several tests were performed in
the lower portion of the air curtain control area (Figure 11).
Sustained injections of tracer were made on January 18, 19, and
20 at the designated locations. The results of these tests are
summarized in Table 19. Pertinent sample and analytical data are
presented in Table 20 (see pages 81 through 83).
48
-------
TABLE 19. SUMMARY OF TRACER RECOVERY EFFICIENCY FOR LOWER CONTROL VOLUME
Test
Converter mode
Blowing
Matte charge
Slag skimming
Cold addition
Idle
Copper pour
Number
of releases
6
17
28
6
8
4
Collection efficiency, %
Mean
33.0
61.8
84.0
61.5
53.8
80.8
Standard
deviation
5.0
27.6
18.4
18.3
22.7
16.9
Range
27-42
35-91
52-128
49-76
30-95
61-98
Tests SP-1 and SP-2, which were performed on January 18,
involved the release of the tracer material at a point along the
baffle wall on the jet side of the air curtain hood. These tests
were conducted during slag skimming and the average recovery
efficiency measured was 94.5 percent, which is comparable to that
reported for the releases on the three-dimensional matrix in the
space above the converter.
Tests SP-3 to SP-5, also performed on January 18, involved
the release of the tracer material at a location slightly above
the ladle near the jet side of the hood. The average recovery
efficiency was 64.3 percent.
On January 19, Tests SP-7 through SP-12 were performed with
release of the tracer material at a location slightly above the
ladle and very close to the wall on the exhaust side. Tests SP-7
through SP-10 were conducted during slag skimming; the recovery
efficiency measured for the four tests ranged from 52 to 79
percent and the overall average recovery efficiency was 63.5
percent. Tests SP-11 and SP-12 were run during matte charging
and the average recovery efficiency was 68.5 percent.
Tests SP-13 through SP-19 also were performed on January 19.
In these tests the tracer material was also released at a loca-
tion slightly above the ladle, but farther from the wall on the
exhaust side. The recovery efficiency measured for the seven
tests ranged from 30 to 89 percent, and the overall average was
49
-------
58.7 percent. It should be noted that the samples for Tests
SP-14, SP-15, and SP-16 (which had collection efficiencies of 32,
33, and 30 percent, respectively) were collected during the
blowing mode. These values would be expected because the hooding
system was in the low flow mode and there was no thermal lift to
enhance the collection efficiency.
The final series of tests (SP-20 through SP-73) were per-
formed on January 20. In these tests the tracer was released
very near the ladle on the exhaust side of the hooding system.
Recovery efficiencies, which were determined for 53 releases of
the tracer material, ranged from 27 to 128 percent, with an
overall average of 70 percent. The efficiencies varied from 38
percent for the 6 tests performed during blowing to 84 percent
for the 28 tests performed during slag skimming. The difference
between average efficiencies for the several operating modes is
statistically significant.
The special injection tests were designed to characterize
hood capture effectiveness in the lower portion of the air cur-
tain control volume. The data clearly show the effects of in-
creased exhaust side air volume and thermal lift from hot gases
during converter roll-out modes. Also, the data indicate that
determination of tracer recovery from the lower portion of the
control volume is heavily dependent on injection locations and
converter operation. For example, tests performed during the
blowing mode, which is conducted with the primary hood closed and
little heat escape, ranged from 27 to 42 percent, compared with
84 percent during slag skimming, 61.8 percent during matte charg-
ing, 61.5 percent for cold additions, and 81 percent for blister
copper pouring. The tracer recovery efficiencies measured during
the slag skimming and copper pour activities are higher because
of the position of the SF.. injection probe and the additional
b
thermal lift. The ladles used during these operating modes were
placed on the ground immediately in front of the converter while
slag or blister copper was being discharged from the converter,
and the injection probe was located above and to the left of the
50
-------
ladle. During the matte charging and cold addition operating
modes, the probe was actually below the ladle because an overhead
crane has to raise the filled ladles to charge the material to
the converter.
In summary, the special point injection tests indicate that
thermal lift plays a significant role in increased collection
efficiencies in the lower portion of the control volume. Also,
the lower tracer collection efficiencies for the various con-
verter roll-out modes could be indicative of the observed fume
spillage outside of the lower control volume.
4.2 OPACITY
The opacity of emissions escaping the air curtain were
monitored and recorded by use of a double-pass transmissometer
coupled with a strip-chart recorder. A detailed description of
the instrumentation and test methodology are presented in Section
7 and Appendix D. The instrument was placed in operation at the
start of each test period and operated continuously throughout
the converter cycle. Table 21 summarizes the opacity of emis-
sions escaping the air curtain slot during the test program, and
Figure 15 shows the average opacity of emissions from each oper-
ation during the cycle when the converter was rolled out. A
total of 86 converter operations that generate fugitive emissions
were monitored during the test program. Average opacities ranged
from 9 percent during blister copper pouring to 21 percent during
cold addition charging. Emissions from converter operations con-
trolled by the primary hood (slag blowing and finish blowing)
showed zero percent opacity from the air curtain. Because the
air curtain was in operation during the entire testing program,
there was no opportunity to evaluate fugitive emissions when the
air curtain was not operating.
51
-------
TABLE 21. SUMMARY OF OPACITY OF EMISSIONS ESCAPING AIR CURTAIN
Converter operation
Cold addition charge
(reverts and cold dope)
Matte charge
Scrap copper charge
Blister copper charge
(imperfect anodes)
Slag skimming
Blister copper pour
Total
Number of
operations
observed
14
25
5
4
31
9
88
Opacity, %
Average
21
14
18
9
18
9
Low
5
5
9
9
2
5
High
54
34
28
9
50
17
Two major problems were encountered during opacity moni-
toring. The first problem involved the strip-chart recorder,
which was equipped with a capillary inking system. This system
did not ink properly in the dusty environment of the smelter.
Recorder chart speed was also restricted to 1 inch per hour,
which resulted in poor resolution and the inability to separate
events that occurred within short time intervals. Every effort
was made to minimize the effect of these problems on data qual-
ity. To provide better resolution and to separate close-occurring
events, the operator manually advance the strip chart. On Satur-
day, January 22, 1983, a dual-speed strip chart recorder was
found and used during that test run. This recorder had chart
speeds of 6 and 30 inches per hour. The faster speeds improved
resolution and allowed events to be recorded on a real-time
basis. The capillary ink system on the recorder was cleaned and
flushed as frequently as practical.
The second problem involved blinding of the transmissometer
by the crane block and cables as the crane moved in and out while
servicing the converter. The instrument was mounted at the top
of the secondary hood in a position that appeared to minimize the
crane cables' interference with the measurement beam. This
52
-------
en
Ui
100
90
80
9)
£ 70
-
a.
o
60
50
40
30
20
10|
0
14
21
18
18
MATTE COLD* SCRAP IMPERFECT SLAG BLISTER
CHARGING ADD COPPER ANODES SKIMMING COPPER
x / POOR
CHARGE
CONVERTER OPERATION
'REVERTS OR COLD DOPE
Figure 15. Average opacity vs. converter operation.
-------
mounting arrangement proved to be satisfactory during most of the
test program; however, on January 22 the second shift crane
operator positioned the crane block so that the transmissometer
beam was blocked during most of the charging and slagging opera-
tions that occurred between 3:30 p.m. and 9:30 p.m.
4.3 VISUAL OBSERVATIONS
Throughout the test program, Mr. Alfred Vervaert of the U.S.
EPA and Mr. James Nolan of the Puget Sound Air Pollution Control
Authority (PSAPCA) visually observed the hood's performance. In
general, hood performance was characterized by estimating overall
hood capture effectiveness and the location, approximate opacity,
duration, and significance of any visible emissions observed. As
shown in the observation logs, lighting and background conditions
within the converter building made absolute opacity readings very
difficult. Typed copies of the observation logs are in Appendix
H. Table 22 summarizes visual estimation of hood capture effici-
encies by converter event and Table 23 presents the observation
logs in terms of the location, duration, and approximate opacity
of escaping emissions and estimated hood capture efficiency for
the three cycles evaluated.
The estimated secondary hood capture efficiency as visually
determined by two independent observers are comparable for each
converter event. The largest discrepancy occurred during addi-
tions (other than copper matte) where average estimated capture
efficiencies were 95 percent (Observer No. 1) and 85 percent
(Observer No. 2). Average estimated capture efficiencies were
equal to or greater than 85 percent for the following converter
events: blow-hold, matte charging, copper pouring, and other
additions (copper scrap, blister copper, etc.). These data
compare favorably with the tracer recovery efficiencies measured
in the control area for similar operating modes. Converter
roll-in and roll-out events were characterized by average esti-
mated capture efficiencies of 77 percent (Observer No. 1) and 76
54
-------
TABLE 22. SUMMARY OF VISUAL OBSERVATIONS OF HOOD CAPTURE
EFFECTIVENESS BY CONVERTER EVENT
Converter
event
Roll-in
Roll -out
Blow-hold
Matte-charge
Slag skim
Copper pour
Other addi-
tions
a
Observer No. 1
No. of events
observed
21
14
42
34
12
42
Estimated
hood capture
efficiency, %
77
96
94
78
92
95
b
Observer No. 2
No. of events
observed
20
5
33
30
12
26
Estimated
hood captuce
efficiency , %
76
90
91
82
85
85
Observer No. 1 was Mr. A.E. Vervaert of the U.S. EPA.
b
Observer No. 2 was Mr. J. Nolan of the PSAPCA.
cFrom observation logs (Appendix H). Represents number of recorded obser-
vations per event.
Represents average value for the number of recorded events.
55
-------
TABLE 23. SUMMARY OF VISUAL OBSERVATIONS LOGS BY EVENT
Event
Cu pour
Blow and
hold
Date
(1983)
1/18
1/18
1/19
Approx-
imate
time, h
0915
1109
1112
1115
1116
1122
1128
1131
1137
1142
1146
1200
1204
1210
0805
0835
0900,
0940*
1100
1510
0722*
0729
0744*
0745
0757,
0823*
0842
0854
0856
0900
0902
0907
0940
0945
1000
1238
1330
1356
1358
Observer 1
Location of
escaping
emissions
Pour
Pour
1" hood
Out front
Upper right
front of
hood
Front right
Front right
Front
None
None
None
None
None
None
None
None
Mouth
None
1° hood
1° hood
None
None
1° hood
None
None
None
1" hood
1' hood
Appoxlmate
duration of
emissions
1 mln.
4 m1n.
1 mln.
5 sec.
15 sec.
10 sec.
5 sec.
Slight
puffing
Approximate opac-
ity of escaping
emissions
Moderate
Moderate
Moderate-heavy; 10
Mode rate -heavy; 20
Heavy; 30
Moderate-heavy
30
Moderate
Light -moderate
Light -moderate
Liqht -moderate
Little or none
Light
Very light
Light
Light
Moderate
Approximate
hood capture
efficiency
195
>95
>95
100
>95
90
80
80
80-90
>95
100
>95
100
100
100
100
Observer 2
Location of
escaping
emissions
Out front under
sheet metal
Top right
Front
Out front below
sheet metal
Top right
Slot/right front
Front
Right front
Front
Left front
Front
Bottom front
Slot
None
None
None
None
None
2" hood-left rear
2" hood-left rear
None
2° hood-left rear
None
None
None
2" hood-left rear
2° hood-left rear
Approximate
duration of
emissions
Small puffs
3 mln.
15 sec.
1 mln.
4 mln.
2 mln.
22 mln.
> Z5 mln.
Approximate opac-
ity of escaping
emissions
10-20
5-10
5-10
5-10
10-20 cont; 35
large puffs
5-10/40-50
60/10
10-20
<5
15-20
50
20
10
Very little
5-10
5-10
20-40
20-40
Approximate
hood capture
efficiency
>90
>90
>90
-90
80
70
'80
80-90
>90
80-90
90
70-80
Ul
CD
(continued)
-------
TABLE 23 (continued)
Event
Blow and
hold
(cont'd)
Slag
skim
Date
(1983)
1/20
1/22
1/18
1/19
1/20
Approx-
imate
time, h
1000
1003
1012
1015
1018
1043
1500
1730
1815
0910
0911
1544
1547
1552
1554
1558
0807
0815
0823
0912
0922
1007
1028
1045
1048
1052
Observer 1
Location of
escaping
emissions
hood
hood
hood
hood
hood
All part
1" hood
and con-
verter
Upper
middle of
opening
Front
Front
Mouth
Pour
Front
Front
Mouth
Front
Front
Front right
Front
Front
Front
Appoxlmate
duration of
emissions
2 m1n.
<5 m1n.
5 sec.
10-15 sec.
5 sec.
30 sec.
60 sec.
2) mln.
3 mln.
1 mln.
3) mln.
7 mln.
3 mln.
5-10 sec.
2 mln.
Approximate opac-
ity of escaping
emissions
Heavy
Moderate-heavy
Very dense, 80
Heavy
Heavy
Extremely heavy
Heavy; 20
Moderate
Heavy; 20-30
Moderate
Moderate-heavy
Heavy; 60
>30
Light
Moderate-heavy; 20
Approximate
hood capture
efficiency
95
95
>95
95
95
>95
95
100
90
90
95
<60
<60
<60
<70
100
85-90
<80-90
Heavy-v. heavy; 20-30
Light 100
Moderate-heavy; 20-30 <80-90
Very heavy
Heavy; 20-30
Moderate-heavy; 20
Moderate; 30
< 50-80
<80
90
70-80;
overall
80
Observer 2
Location of
escaping
emissions
Slot
Slot
510, charge door
1° hood
None
None
Top
Back of con-
verter
Front under
sheet metal
Front of 2° hood
Right part 2°
hood
Front of hood
Front/some puff
through slot
Slot/front right
--/Front right
Front and slot
Front
Front/slot
Slot/front
20* of front of
hood (lower
right)
Top right right
(5-90* of hood
area)
Approximate
duration of
emissions
30 sec.
22 mln.
Small puffs
1 mln.
2 mln.
Small puffs
-/puffs
1 n>1n.
Puffs/puffs
5 mln.
15 sec.
1} mln-puffs
2 mln.
Approximate opac-
ity of escaping
emissions
30
5-10
Quite a bit
100
Small amount
10-20
100
60-80
80
Heavy ;60 -80
10-20/20
Moderate/30
30-40
30-40
40/10-40
20/60
40
40
20
Approximate
hood capture
efficiency
90
90
90
90
90
>90
50-70
90
90
50-70
50-70
90
>90
70
80-90
90
50-90; avg. -80
70-90
90
90
Ul
(continued)
-------
TABLE 23 (continued)
Event
Slag
Skim
(cont'd)
Sl«g
Skin
(cont'd)
Date
(1983)
1/20
1/22
Other additions
Blister
Cu-ladle,
Cu slag
1/18
Approx-
imate
time, h
1058
1424
1432
1438
1537
1545
1655
1703
0959
1002
1012
1137
1140
1145
1153
1622
1628
1630
1718
0921
1243
Observer 1
location of
escaping
emissions
Front
Front
Front
Front
Right front
Front
Front
Right front
Mouth
Behind 2°
hood; S10,
chg. chute
Appoxlmate
duration of
emissions
2 mln.
<2 mln.
It mln.
2 mln.
45 sec.
50 sec.
54 sec.
45 sec.
1 mln.
45 sec.
45 sec.
2 mln.
43 sec.
3 mln.
53 sec.
2 mln
2 mln.
10 sec.
3 mln.
35 sec.
15 mln.
Approximate opac-
ity of escaping
emissions
20
Approximate
hood capture
efficiency
80-90;
overall
85
70
Moderate-heavy; 20-30 70
Heavy; 10-20
Heavy
Light
Heavy; 30-40
Moderate-heavy; 20
Heavy; 20-30
Moderate-heavy
20; moderate-heavy
Moderate-heavy; 20
Heavy
Heavy; <20
Moderate-heavy; <20
Moderate-heavy;
10-20-30
Heavy; 20
Moderate
Heavy; 20-40
Heavy; 20
Moderate-heavy
40-50
70-80
70
100
70
80-90
80
95
95
90
70-80
70-90
80-95;
overall
90
80-95;
overall
90
80
90
70
80-90;
overall
85
>95
>90
Observer 2
Location of
escaping
emissions
Lower right front
(30% of 2" hood
face)
Front (25% of
face)/slot
Front (25«)/slot
Right front face
(201}
Front (25J)
Front (25%)
Slot/face (20%)
Slot/ face (50%)
Right front
Slot
Slot/front
Front/slot
Front
Front
Front
Top of hood
Approximate
duration of
emissions
11 mln.
45 sec.
1} mln.
3 mln.
1 mln.
2 mtn.
It mln. puffs
31 mln.
50 sec. puffs
1 mln. 45
sec.
2 mln. 45
sec. (slight
puffs)
30 sec. puffs
1) mln. puffs
3 mln. puffs
4 mln. puffs
Small puffs
Approximate opac-
ity of escaping
emissions
20-40
40-60/20
50-60/10
20-30
40-50
30-40
40/60-80
20/40
40
10-20
40-50/40
20/10
20
40
80
Approximate
hood capture
efficiency
80
70
80
80-90
70-80
70-80
80
80-90
>90
>90
>90
80-90
90
>90
>90
80-90
CO
(continued)
-------
TABLE 23 (continued)
Event
Anode
slag
—
Cu slag
--
Scrap
Shell
slag
S10,
S10,
S10,
—
—
--
—
--
Shell
slag
S10,
—
Date
(1983)
1/19
Approx-
imate
time, h
1305
1311
1344
1348
1509
1511
1514
1515
1517
1520
1521
1522
1524
1525
1530
0725
0731
0744
Observer 1
Location of
escaping
emissions
1° hood
1° hood-
mld-pt.
1° hood
Behind
2° hood
1° hood
1" hood
1° hood
None
Appoxlntate
duration of
emissions
3 mln.
1 mln.
5 sec.
2 sec.
Approximate opac-
ity of escaping
emissions
10-20
Moderate-light; <5
Light
30
20
Light
Approximate
hood capture
efficiency
90->95
90
95
100
>95
100
100
100
Observer 2
Location of
escaping
emissions
Slot
Slot
Front and top
of hood
1° hood
Top left rear
of 2° hood
Left top rear
of 2° hood
Left rear of
of 2° hood
Left rear of
of 2° hood
Left rear of
of 2° hood
Approximate
duration of
emissions
Short du-
ration
i mln.
Approximate opac-
ity of escaping
emissions
20-30 ;upto 80
30-40
100
100
60-70
10-20
5-10
20-30
10-20
5-10
Approximate
hood capture
efficiency
>90
100
>90
-25
-25
Effectively
controlled
Ul
vo
(continued)
-------
TABLE 23 (continued)
Event
S10,
--
Shell
slag
Shell
slag
Shell
slag
t scrap
eharqes
Scrap
~
Scrap
Scrap
Scrap-
large
piece
Scrap
Shell
slag
Shell
slag
510,
Date
(1983)
1/20
Approx-
imate
time, h
0755
0757
0836
0840
0936
1127-
1135
1229
1235
1316
1317
1410
0947
0952
1008
Observer 1
Location of
escaping
emissions
None
Right front
Slot
Rear right
1° hood
Appoxlmate
duration of
emissions
2 sec
30 sec.
15-20 sec.
1 mln.
28 sec.
Approximate opac-
ity of escaping
emissions
20
60
60-70
50-60
20
40
60
Heavy
Approximate
hood capture
efficiency
85-85;
overall
90
100
40-95;
overall
85
70-80
90-95
>90
>90
95
>95
even-
tually
95
Observer 2
Location of
escaping
emissions
Rear left of
2° hood
Front
Slot and top
of front
Slot
Slot
Slot
Slot
Slot
Rear of slot/
top right
front of 2°
hood
Right side of
slot
1° hood
Approximate
duration of
emissions
Puffs
1 puff
1 puff
5 sec.
1 m1n.
5 sec.
1 mln.
Puff
40 sec/
puffs
30 sec.
Approximate opac-
ity of escaping
emissions
5-10
40-60
80-100
40
30-40
60
5-10
60
5-10
40-100/10
40-60
Heavy
Approximate
hood capture
efficiency
>90
>95
50-60
>90
>90
>90
>90
>90
80
80
>90
(continued)
-------
TABLE 23 (continued)
Evtnt
S10,
Shell
slag
Shell
slag
Shell
slag
Shell
slag
S10,
Blister
Cu block
Blister
Cu block
Blister
Cu block
Blister
Cu block
Shell
slag
Shell
slag
Shell
slag
Shell
slag
S10,
Date
(1983)
1/22
Approx-
imate
time, h
1011
1117
1118
1454
1457
1505
1808
1810
1812
1814
0907
0911
1051
1055
1605
Observer 1
Location of
escaping
emissions
1° hood
Upper right
front
Upper right
rear
Mn_ —
nvnc
Appoxlmate
duration of
emissions
20 sec.
20 sec.
3 sec.
2 sec.
10 sec.
Approximate opac-
ity of escaping
emissions
80
30-40
30
Approximate
hood capture
efficiency
>95
100
100
>95
>95
95
>95
>95
100
90-95;
overall
95
100
>95
100
Observer 2
Location of
escaping
emissions
1" hood
Slot
Right front
face/slot
Approximate
duration of
emissions
20 sec.
15 sec.
Puffs
1 puff/puffs
20 sec.
15 sec.
Approximate opac-
ity of escaping
emissions
Very heavy
20
60/50
Approximate
hood capture
efficiency
>90
>90
>90
>90
100
>90
100
100
100
(continued)
-------
TABLE 23 (continued)
Event
510,
Blister
Cu block
Blister
Cu block
Blister
Cu block
Blister
Cu block
Blister
Cu block
Cu spills
Blister
Cu block
Blister
Bu block
Powdered
scrap
Powdered
scrap
Matte
charge
Date
(1983)
1/18
Approx-
imate
time, h
1705
1726
1738
1731
1733
1735
1827
1832
1834
2033
2116
1237
1240
1300
1313
1325*
Observer 1
Location of
escaping
emissions
None
Top
Ladle
Top of hood
1° hood
Mouth
Appoxlmate
duration of
emissions
5 sec.
5 sec.
5 sec.
5 sec.
5 sec.
5 mln.
30 sec.
10 sec.
15 sec.
Approximate opac-
ity of escaping
emissions
Heavy- v. heavy
V. light
Heavy-v. heavy
V. heavy; 30
Moderate; <10
Light
Approximate
hood capture
efficiency
100
100
100
100
100
95
90
95
80 prior
to fire
Undeter-
minable
>95
100
>95
>95
95
>95
Observer 2
Location of
escaping
emissions
Front
Front/ top
Top
Slot/front
Approximate
duration of
emissions
Small puffs
--/few puffs
Approximate opac-
ity of escaping
emissions
5
5/5-10
5-10
20/10-20
Approximate
hood capture
efficiency
>90
>»5
>90
>90
to
(continued)
-------
TABLE 23 (continued)
Event
Matte
charge
(cont'd)
Date
(1983)
1/18
1/19
1/20
Approx-
imate
time, h
1327
1335*
1340
1616
1618
1621
0830
0834
0932
0939
0822
0833
0840
0854
0855
0909
Observer 1
Location of
escaping
emissions
1° hood
1° hood
(m1d-pt)
Top
Mouth
Front
Mouth
Mouth
Mouth
Above air
Jet
Appoxlmate
duration of
emissions
30 sec.
30 sec.
10 sec.
1-2 sec.
30 sec.
30 sec.
15 sec.
18 sec.
45 sec.
20 sec.
20 sec.
5 sec.
Approximate opac-
ity of escaping
emissions
V. heavy; <20
Heavy
Heavy
20
Heavy; 20-30
Heavy
<20
Light
Heavy
Moderate
Moderate
15
Moderate-heavy
>20
Approximate
hood capture
efficiency
>95
>95
Nearly
100
>95
95
>90
100
>80
>95
>95
95
100
>95
<50
90S
100
100
>90
95
<90;
range
85-95
Observer 2
Location of
escaping
emissions
Slot
Slot
Top right front
Right front of
hood
Top lOf of face/
slot
Slot
Slot
Approximate
duration of
emissions
20 sec.
30 sec.
30 sec.
30 sec.
10 sec.
15 sec.
Puff/H
m1n
Puff
15 sec.
15 sec.
Approximate opac-
ity of escaping
emissions
10-20
15-20
40
40-60
4-60/50
10-20
10-20
Approximate
hood capture
efficiency
>95
>90
80-90
>90
80-90
>95
>90
>90
>95
80
>95
>90
MO
>*
10
U)
(continued)
-------
TABLE 23 (continued)
Event
Matte
charge
(cont'd)
Roll-In
Roll -out
In
Out
In
Out
Date
(1983)
1/30
1/22
1/18
Approx-
imate
time, h
0911
0915
0926
0930
0944
0958
1110
1111
1114
1449
1452
1554
1608
1046
1047
1206
1209
0805
0906
1045
1109
Observer 1
Location of
escaping
emissions
Right rear
of hood
Right rear
Right rear
Right rear
Ladle
Appoxlmate
duration of
emissions
Large puffs
Large puffs
Brief puff-
Ing
5 sec.
25 sec.
20 sec.
15 sec.
20 sec.
20 sec.
20 sec.
15 sec.
IS sec.
12 sec.
Approximate opac-
ity of escaping
emissions
30
Moderate
20
20
Light
30
Approximate
hood capture
efficiency
>95
>90
95
95
95
>95
>95
>95
90-95
95
>95
>95
95
95
100
>95
75
100
100
70
Observer 2
Location of
escaping
emissions
Slot
Slot/front
Slot
Slot
Slot
Slot
None
Top
Approximate
duration of
emissions
Puffs
17 sec.
Small puffs
15 sec.
25 sec.
(puffs)
Small puffs
15 sec.
20 sec.
I m1n.
10 sec.
Approximate opac-
ity of escaping
emissions
10-20
5- 10/20
20-40
<10
20-30
0
20-30
Approximate
hood capture
efficiency
>95
>90
>95
>90
>90
>90
>90
>95
>95
>90
>90
>90
>90
^70-tD
>90
(continued)
-------
TABLE 23 (continued)
Event
Roll-In
Roll -out
Out
In
Out
In
Out
In
Out
In
Out
Out
Out
In
Out
In
Out
In/Out
In/Out
In/Out
Out
Out
In
Out
In
Date
(1983)
1/19
1/20
Approx-
imate
time, h
1544
0729
0807
0842
0912
0940
1007
1028
1228
1312
1319
1408
1119
1129
1139
1142
1214
1424
1537
1615
1756
1815
Observer 1
Location of
escaping
emissions
Hood
Appoxlmate
duration of
emissions
5 sec.
5-10 sec.
2 sec.
30 sec.
5 sec.
2 sec.
Approximate opac-
ity of escaping
emissions
60
40-50
Moderate-heavy
30-40
-20
30
100
20-30
Approximate
hood capture
efficiency
<60
<50
>95
<50
90
<50
>90
>95
Undeter-
minable
100
90
Observer 2
Location of
escaping
emissions
Slot (some
from front)
Slot and top
front of hood
Front
Slot
Slot
1° hood
Slot
Top of hood and
2° hood slot
Top
Approximate
duration of
emissions
1 mln.
30 sec.
30 sec.
15 sec.
10 sec.
10 sec.
20 sec.
10 sec.
-45 sec.
Approximate opac-
ity of escaping
emissions
100
50-60
-20
30
40
60
Approximate
hood capture
efficiency
50-70
70
50
50 nax.
50
>90
70-80
70
>90
>90
70-80
50-60
(continued)
-------
TABLE 23 (continued)
Event
(toll-In
Roll -out
In
Out
In
Out
In
Out
Out
In
Out
Out
Out
Date
(1983)
1/22
Approx-
imate
time, h
1915
1959
0913
0952
1057
1138
1716
1736
1823
1931
2115
Observer 1
Location of
escaping
emissions
Over top
of hood
Right
front
Appoxlmate
duration of
emissions
10 sec.
5 sec.
10 sec.
5 sec.
Approximate opac-
ity of escaping
emissions
Light
40-60
20 (pen. 40-
60)
40
Approximate
hood capture
efficiency
90
60-70
70
<50
100
60
60
Observer 2
Location of
escaping
emissions
Fop
Approximate
duration of
emissions
Large puffs
Approximate opac-
ity of escaping
emissions
Approximate
hood capture
efficiency
99
>90
90
>90
80
-------
percent (Observer No. 2) . Slag skimming operations showed ave-
rage capture efficiencies of 78 percent (Observer No. 1) and 82
percent (Observer No. 2).
Visual emissions observation revealed that converter and
crane operations introduce significant variability in overall
hood capture efficiency particularly for skimming operations.
For example, as visually determined, hood capture effectiveness
increased considerably (greater than 90 percent) during skimming
operations when the overhead crane operator held the receiving
ladle next to the converter while the converter was slowly ro-
tated to the discharge position. In contrast, when the receiving
ladle was placed on the ground during skimming operations and the
slag discharge rate was rapid, considerable fumes spilled into
the converter aisle. Converter and overhead crane operations
were inconsistent throughout the entire test program.
Observations of visual emissions to estimate hood capture
efficiency were made by assessing the overall capture effective-
ness during specific operating modes. The duration and intensity
of fugitive emissions generated were highly variable often last-
ing only seconds as recorded by the observers. Since these
observations were overall assessments, that is, the entire con-
verter-secondary hood area, they would not necessarily correspond
to tracer gas recovery data for a given converter mode. For
example and as noted previously, when the receiving ladle was
placed on the ground during skimming operations and the slag
discharge rate was rapid, considerable fumes spilled into the
converter aisle. Tracer recovery tests performed in the upper
control volume during such an event would not account for this
spillage and would probably show a greater recovery efficiency
than an overall visual assessment. For this reason, visual
observation must be used in conjunction with the tracer to quan-
tify capture effectiveness.
Data from the test program substantiates this fact as the
average tracer recovery efficiency in the upper control area
averaged over 90 percent during skimming operations compared to
67
-------
average visual assessments of 78 and 82 percent. In contrast,
tracer recovery efficiencies from the special injection point
tests during skimming operations averaged 84 percent.
In summary, the visual observation and tracer recovery data
indicate that the fugitive emission capture effectiveness of the
secondary hood is greater than 90 percent. The capture effec-
tiveness during converter roll-in and roll-out and slag skimming
operations is more variable than other converter modes since
fugitive emissions generated during these events are more de-
pendent upon converter and crane operations. It was observed
that careful operation of the converter and ladle during the
discharge of slag and blister copper could minimize the occur-
rence of fume "spillage" and provide capture efficiencies of 90
percent or greater.
Thermal lift plays a significant role in increased collec-
tion efficiencies for fumes generated in the lower portion of the
control area. Also, the lower tracer recovery efficiencies for
the various converter roll-out modes are indicative of fume
spillage outside the control area.
It is believed that no practical correlation can be made
between opacities recorded by the observers and the transmissom-
eter. The transmissometer was mounted perpendicular to the
longitudinal axis of the slot, whereas position of the visual
observers was such that their view was parallel to the longitu-
dinal axis of the slot, which resulted in a considerably longer
path length through the escaping emissions. The apparent opacity
increases as the path length through the emissions increases.
Also, when positioned in front of the converter, the overhead
crane interfered with visual observations above the slot area.
68
-------
TABLE 4. SUMMARY OF PRELIMINARY TRACER RECOVERY EFFICIENCY DATA
Sample
I.D.
1A&B
2A&B
3A&B
4A&B
PR-5T
PR-6T
PR-7T
PR-8T
PT-9T
PT-10T
PT-11T
PT-12T
PT-13T
PT-14T
Date
(1982)
12/9
12/9
12/9
12/9
12/10
12/10
12/10
12/10
12/10
12/10
12/10
12/10
12/11
12/11
Measured
gas flow
rate, cc/min
2.92 x 10s
2.92 x 109
2.92 x 109
2.92 x 109
2.91 x 109
2.89 x 109
3.02 x 109
3.03 x 109
3.01 x 109
2.96 x 109
2.96 x 109
2.94 x 109
2.65 x 109
2.60 x 109
SF6D
concentra-
tion, V/V
-
-
.
3.09 x 10*9
8.02 x Kf9
6.98 x 10"9
6.88 x 10"9
6.99 x 10"9
9.12 x 10*9
8.95 x 10'9
9.34 x 10*9
8.84 x 10"9
8.81 x 10"9
8.55 x 10"9
Average SFJ.
Injection0
rate, cc/m1n
2.87
3.47
10.07
10.14
24.91
24.91
24.91
24.91
30.91
30.91
30.91
30.91
30.21
30.21
SF6d
mass flow
rate, cc/min
.
.
.
9.02
23.34
20.17
20.78
21.18
27.45
26.49
27.65
25.99
23.35
22.23
Recovery6
effi-
ciency, %
.
.
_
89
94
81
83
85
89
86
89
84
77
74
Runs 1 through 4 A&B were collected at two separate points of average velocity in the duct. For
Runs 1 through 3 the SF, concentration was at the minimum detectable limit. Runs PR-5T through
PR-14T were collected as integrated samples by traversing the cross-sectional area of the
stack.
Measured gas flow rate (secondary exhaust duct): acfm x 28,320 = cc/min.
SF, concentration (parts per part; volume per volume)
CSF, injection rate determined by bubble meter calibration before and after each injection. The
difference ranged from 0.06 to 0.14 cc/min for Tests 1 through 4 and 0.06 to 2.58 cc/min for
Tests 5T through 14T.
SFg mass flow rate = gas flow rate (cc/min) x SF, concentration (v/v) = cc/min SF,.
eRecovery efficiency =
100 - (^F6 inJected ' SF, measured>
SF, injected
v
x 100
Note: Tracer recovery efficiencies are subject to a +18 percent relative error as described
In Subsection 4.1.2.
69
-------
TABLE 5. SUMMARY OF SINGLE POINT TRACER RECOVERY EFFICIENCY TESTS
Sample
I.D.
PRSP-13
PRSP-29
PRSP-3a
PRSP-43
PRSP-5b
PRSP-6b
PRSP-7b
PRSP-8C
PRSP-9C
PRSP-10C
Date
(1982)
12/11
12/11
12/11
12/11
12/11
12/11
12/11
12/11
12/11
12/11
Average
gas flow
rate, cc/min
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
4.12 x 109
4.12 x 109
2.05 x 109
2.05 x 109
2.05 x 109
SFs.
concentra-
tion, v/v
9.89 x 10"9
1.00 x 10"8
9.67 x 10"9
9.74 x 10"9
5.55 x 10"9
5.77 x 10"9
5.75 x 10"9
1.24 x 10'8
1.25 x 10'8
1.26 x 10"8
Total
Average
SF6
injection
rate, cc/min
30.21
30.21
30.21
30.21
28.94
28.94
28.94
28.94
28.94
28.94
294.48
29.45
SF,
mass flow
rate, cc/min
28.89
29.20
28.24
28.44
22.87
23.77
23.69
25.42
25.63
25.83
261.98
26.20
Recovery
effi-
ciency, %
96
97
93
94
79
82
82
88
89
89
89
a
A medium air flow.
5A high air flow.
"A low air flow.
Note: SFfi recovery efficiencies are subject to a ±18 percent relative
error as described in Subsection 4.1.2.
-------
TABLE 6. SUMMARY OF TRACER RECOVERY EFFICICNCY AT AIR CURTAIN SUCTION INLET
Sample
I.D.
AC-1
AC-2
AC-3
AC-4
AC-5
AC-6
Date
(1982)
12/11
12/11
12/11
12/11
12/11
12/11
Average
gas flow
rate, cc/min
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
SF6
concentra-
tion, v/v
9.95 x 10"9
9.68 x 10"9
9.93 x 10"9
1.00 x 10"8
1.03 x 10"8
1.01 x 10"8
Average
. .SF6.
injection
rate, cc/min
30.92
30.92
30.92
30.92
30.92
30.92
30.92
SF6
mass flow
rate, cc/min
29.05
28.27
28.99
29.20
30.08
29.49
29.18
Recovery
effi-
ciency, %
94
91
94
94
97
95
94
Average gas flow rate at medium flow setting.
Note: SFg recovery efficiencies are subject to a ±18 percent relative
error as described in Subsection 4.1.2.
-------
TABLE 7. SUMMARY OF TRACER RECOVERY TEST DATA ON EXHAUST SIDE
Sample
1.0.
A (Light)
81-1
Bl-2
Bl-3
Bl-4
B2-1
B2-2
B?-3
B2-4
B3-1
B3-2
B3-3
B3-4
B4-1
B4-2
B4-3
B4-4
Cl-1
.Cl-3
Cl-4
C2-1
C2-2
C2-3
C2-4
C3-1
C3-2
C3-3
C3-4
C4-1
C4-2
C4-3
C4-4
C5-1
Date
(1962)
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
Avenge,
gas flow*
rate, cc/min
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
2.92 x 109
2.92 x 109
4.12 x 109
4.12 x 109
4.12 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
2.92 x 109
4.12 x 109
SF,
concentra-
tion, v/v
9.67 x 10'9
9.12 x 10'9
9.17 x 10'9
9.14 x 10'9
8.90 x 10"9
6.49 x 10~9
6.17 x 10'9
6.26 x 10"9
6.33 x 10'9
6.24 x 10"9
6.35 x 1(T9
6.29 x 10'9
6.04 x 10'9
9.33 x 10"9
9.62 x 10"9
7.46 x 10"9
6.26 x 10~9
7.43 x 10'9
9.73 x 10"9
9.63 x 10"9
B.50 x 10"9
8.53 x 10~9
8.45 x 10"9
8.58 x 10"9
8.26 x 10"9
8.09 x 10"9
7.47 x 10"9
5.94 x 10"9
6.46 x 10"9
6.55 x 10"9
6.68 x 1C'9
9.92 x 10'9
8.22 x 10'9
SF,
Injection
rate, cc/min
30.72
32.17
32.17
32.17
32.17
31.96
31.96
31.96
31.96
31.43
31.43
31.43
31.43
31.06
31. OB
31.08
31.08
34.01
34.01
34.01
32.47
32.47
32.47
32.47
32.33
32.33
32.33
32.33
31.55
31.55
31.55
31.55
30.68
SF. MSS flow
rate, cc/min
28.24
26.63
26.78
26.69
25.99
26.74
25.42
25.79
26.08
25.71
26.16
25.91
24.88
27.24
28.09
30.74
25.79
30.61
28.41
28.82
24. 82
24.91
24.67
25.05
24.12
23.62
30.78
24.47
26.62
26.99
27.52
28.97
33.87
Recovery
efficiency, %
92
83
83
83
81
84
80
81
82
82
83
82
79
88
90
99
83
90
84
85
76
77
76
77
75
73
95
76
84
86
87
92
110
(continued)
72
-------
TABLE 7 (continued)
Sample
I.D.
C5-3
C5-4
C6-1
C6-2
C6-3
C6-4
Dl-1
Dl-2
Dl-3
Dl-4
D2-1
D2-2
D2-3
D2-4
D3-1
D3-2
D3-3
D3-4
Date
(1982)
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
12/12
Average
gas flow
rate, cc/nrin
2.92 x 109
2.92 x 109
4.12 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
4.12 x 109
SF,
concentra-
tion, v/v
9.48 x 10"9
9.96 x 10"9
7.76 x 10"9
9.60 x 10"9
1.03 x 10'8
9.46 x 10"9
9.20 x 10"9
7.06 x 10"9
8.62 x 10"9
8.15 x 10"9
8.55 x 10~9
8.79 x 10"9
8.57 x 10"9
7.04 x 10"9
6.29 x 10'9
6.23 x 10"9
6.32 x 10'9
6.32 x 10"9
SF,
Injection
rate, cc/min
30.68
30.68
30.72
30.72
30.72
30.72
32.00
32.00
32.00
32.00
31.94
31.94
31.94
31.94
31.47
31.47
31.47
31.47
SF, mass flow
rate, cc/min
27.68
29.08
31.97
28.03
30.08
27.62
26.86
29.09
25.17
23.80
24.97
25.67
25.02
29.00
25.91
25.67
26.04
26.04
Recovery
efficiency, %
90
95
104
91
98
90 *
84
91
79
74
78
80
78
91
82
82
83
83
Secondary exhaust duct gas flow rate in cubic centimeters per minute. Values represent
average measured flow rates (EPA Methods 1 and 2) for each of three separate damper
settings.
Note: Tracer recovery efficiencies are subject to a +18 percent relative error as
described In Subsection 4.1.2.
73
-------
TABLE 8. SUMMARY OF TRACER RECOVERY TEST DATA ON JET SIDE
Sample
I.D.
A (Light)
Cl-5
Cl-6
C2-5
C2-6
C3-5
C3-6
C4-5
C4-6
C5-6
C6-5
C6-6
Dl-6
D3-5
03 -6
Bl-5
Bl-6
B2-5
B2-6
B3-5
B3-6
B4-5
B4-6
Date
(1982)
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
12/13
Average
gas flow
rate, cc/»1n
2.92 x 109
4.12 x 109
4.12 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.05 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
4.12 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.92 x 109
2.05 x 109
2.92 x 109
2.92 x 109
2.92 x 109
SF,
concentra-
tion, v/v
1.05 x 10'8
6.46 x 10"9
6.51 x 10'9
9.89 x 10"9
7.78 x 10'9
9.70 x 10"9
1.08 x 10"8
9.52 x 10"9
8.12 x 10"9
8.63 x 10"9
9.75 x 10"9
8.51 x 10'9
1.00 x 10"8
9.86 x 10'9
8.37 x 10'9
9.68 x 10"9
8.50 x 10"9
1.00 x 10"8
9.42 x 10"9
1.26 x 10"8
1.03 x 10"8
1.06 x 10'8
8.90 x 10"9
SF,
Injection
rate, cc/min
31.10
29.89
29.89
30.38
30.38
30.45
30.45
30.50
30.50
30.54
30.41
30.41
30.30
30.56
30.56
31.04
31.04
31.10
31.10
31.01
31.01
31.10
31.10
Unconnected
SF, mass flow
rate, cc/min
30.66
26.62
26.82
28.88
22.72
28.32
22.14
27.80
23.71
25.20
28.47
24.85
29.20
28.79
34.48
28.27
24.82
29.20
27.51
25.83
30.08
30.95
25.99
Recovery
efficiency, %
99
89
90
95
75
93
73
91
78
83
94
82
96
94
113
91
80
94
88
83
97
99
84
Note: Tracer recovery efficiencies are subject to a +18 percent relative error as
described in Subsection 4.1.2.
74
-------
TABLE 9. SUMMARY OF MATRIX AREA TRACER RECOVERY EFFICIENCY DATA
01
Simple I.D.
A (light)
Bl-1
Bl-2
Bl-3
Bl-4
Bl-5
Bl-6
B2-1
82-2
B2-3
BZ-4
82 -5
B2-6
B3-1
B3-2
B3-3
83-4
B3-S
B3-6
84-1
84-2
B4-3
B4-4
84-5
B4-6
Cl-1
Cl-3
Cl-4
Cl-5
Cl-6
C2-1
C2-2
C2-3
C2-4
C2-5
C2-6
Recovery
efficiency, J
92
99
83
63
83
Bl
91
80
84
80
81
82
94
88
82
83
82
79
83
97
88
90
99
83
99
84
90
84
85
89
90
76
77
76
77
95
75
Converter mode
Blowing
Blowing
Blowing
Blowing
Blowing
Blowing
Blowing
Blowing
Slag skinning
Idle
Idle
Idle
Blowing
Blowing
Idle
Idle
Idle
Blister copper pour
Silica charge
Blowing
Blowing
Blowing
Copper pour
Primary hood up
Blowing
Blowing
Slag skim
Blowing
Blowing
Matte charge
Cold dope charge
Blowing
Blowing
Blowing
Blowing
Blowing
Blowing
Sample I.D.
C3-1
C3-2
C3-3
C3-4
C3-5
C3-6
Dl-2
Dl-2
Dl-3
01 -4
01-5
Dl-6
D2-1
D2-2
02-3
02-4
C4-1
C4-2
C4-3
C4-4
C4-5
C4-6
C5-1
C5-3
C5-4
C5-5
C5-6
C6-1
C6-2
C6-3
C6-4
C6-5
C6-6
D3-1
03-2
D3-3
D3-4
D3-5
D3-6
Recovery
efficiency, %
75
73
95
76
93
73
84
91
79
74
110
96
78
80
78
91
84
86
87
92
91
78
no
90
95
96
83
104
91
98
90
94
82
82
82
83
83
94
113
Converter mode
Blowing
Blowing
Blowing and skinning slag
Skimming slag
Blowing
Silica charge
Blowing
Skinning slag
Blowing
Blowing
Skinning slag
Blowing
Blowing
Blowing
Blowing
Skimming slag
Matte charge
Matte charge
Matte charge
Blowing
Blowing
Blowing
Matte charge
Blowing
Blowing
Matte charge
Blowing
Matte charge
Blowing
Blowing
Blowing
Blowing
Blowing
Idle
Idle
Idle
Idle
Blowing
Skimming slag
Note: Tracer recovery efficiencies are subject to a +18 percent relative
error as described in Subsection 4.1.2.
-------
TABLE 13. SUMMARY OF MATRIX INJECTION TEST DATA, 1/14/83
Saaple 1.0.
and time
C6-4
16:05
C6-3
16:15
Cl-3
16:25
B2-3
16:33
Dl-3
16:44
B2-4
16:55
Dl-4
17:02
Cl-4
17:11
C6-3
17:18
Cl-4
17:27
C6-4
17:35
Dl-3
17:47
B2-3
17:58
Cl-3
18:10
Dl-4
18:16
B2-4
18:23
C6-3
18:30
C6-4
18:36
Dl-3
18:45
Cl-3
18:53
B2-3
19:01
Cl-4
19:07
B2-4
19:15
Dl-4
19:22
Date
(1983)
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
Average gas
flow rate,,
cc/min x 10
3.58
(cold addition)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(slag sMn)
3.58
(matte charge)
2.14
(idle)
3.58
(cold addition)
3.58
(matte charge)
3.58
(matte charge)
2.14
(blow)
2.14
(blow)
3.58
(slag skim)
3.58
(slag skim)
3.58
(matte charge)
3.58
(matte charge)
3.58
(matte charge)
2.14
(Idle)
2.14
(Idle)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
SF, concen-
tration 1n v/v
1.20 x 10'8
-
-
2.03 x 10'8
1.26 x 10'8
1.24 x 10"8
1.91 x 10'8
1.17 x 10'8
1.18 x 10'8
1.12 x 10~8
1.71 x 10'8
-
1.08 x 10"8
1.26 x 10"8
1.26 x 10'8
1.36 x 10"8
1.27 x 10"8
2.21 x 10'8
2.30 x 10"8
2.27 x 10'8
1.84 x 10"8
2.12 x 10"8
2.21 x 10'8
1.62 x 10"8
SF, Injection
rite, cc/m1n
46.02
45.24
45.62
46.34
46.16
45.86
45.07
44.52
44.16
44.27
45.14
45.28
43.42
47.72
49.88
50.59
51.30
52.50
49.82
49.94
50.02
50.34
50.46
50.36
SF, nss
flow, cc/min
42.96
-
-
43.44
45.11
44.39
40.87
41.89
42.24
40.10
36.59
-
38.66
45.11
45.11
48.69
45.47
47.29
49.22
48.58
39.38
45.37
47.29
34.67
Recovery
efficiency. 1
93
Sample void
Sample void
94
98
97
91
94
96
91
81
Sample void
89
95
90
96
89
90
98
97
79
90
94
69
(continued)
76
-------
TABLE 13 (continued)
Sinple I.D.
tnd tiM
B2-1
19:50
C6-2
19:56
Dl-1
20:02
Cl-1
20:10
B2-2
20:18
Dl-2
20:25
C6-1
21:01
Cl-2
21:10
B2-2
21:18
Dl-2
21:25
C6-1
21:31
Cl-2
21:37
Cl-1
21:45
62-1
21:51
C6-2
21:56
Dl-1
22:01
Cl-1
22:10
B2-2
22:16
Cl-2
22:22
B2-1
22:27
C6-2
22:33
Dl-1
22:46
Dl-2
22:54
C6-1
22:59
Dm
(1983)
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
1/14
Average gas
flow rite,.
cc/n1n x 10s
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(slag sk1«)
3.58
(slag skim)
3.58
(slag skim)
3.58
(matte charge)
3.58
(cold addition)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(slag skim)
3.58
(slag skim)
2.14
(idle)
2.14
(Idle)
3.58
(slag skim)
3. 58
(slag sk1»)
3.58
(matte charge)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
SF, concen-
tnt'on 1n v/v
2.18 x ID*8
2.14 x 10"8
2.04 x 10"8
2.18 x 10"8
2.29 x 10*8
1.41 x 1CT8
1.26 x 10"8
1.25 x 10"8
1.29 x 10'8
1.61 x 10'8
2.28 x 10'8
2.37 x 10'8
2.15 x 10*8
1.20 x 10'8
1.25 x 10"8
2.00 x 10'8
2.11 x 10"8
1.27 x 10'8
1.24 x 10"8
1.22 x 10"8
2.12 x 10"8
2.04 x 10'8
2.04 x 10*8
2.06 x 10~8
$F, Injection
rate, cc/n1n
48.02
48.02
48.02
48.02
48.04
48.04
48.04
48.04
48.37
48.37
48.37
48.37
48.26
48.26
48.26
48.26
46.69
46.69
46.69
46.69
46.95
46.95
46.95
46.95
SF, MSS
flow, cc/m1n
46.65
45.80
43.66
46.65
49.01
50.48
45.11
44.75
46.18
57.64
48.79
50.72
46.01
42.96
44.75
42.80
45.15
45.47
44.39
43.68
45.37
43.66
43.66
44.08
Recovery
efficiency, 1
97
95
91
97
102
105
94
93
95
119
101
105
95
89
93
89
97
97
95
94
97
93
93
94
Note:
SF, recovery
tl8n 4.1.2.
efficiencies ire subject to i t!8 percent relative error is described 1n Subsec-
77
-------
TABLE 15. SUMMARY OF MATRIX INJEXTION TEST DATA, 1/17/83
Simple I.D.
and Time
B2-1
10:15
B2-2
10:17
Cl-1
10:20
Cl-2
10:23
C6-1
10:31
C6-2
10:33
01-1
10:36
01-2
10:40
B2-3
11:13
B2-4
11:15
Cl-3
11:18
Cl-4 .
11:21
C6-3
11:30
C6-4
11:32
01-3
11:34
Dl-4
11:36
Date
(1983)
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
1/17
Average gas
flow rate.,
cc/eln x 10*
(converter node)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
2.14
(line blow)
3. SB
(matte charge)
3.56
(matte charge)
2.14
(line blow)
2.14
(line blow)
SF, concen-
tration in v/v
2.13 x 10'8
2.27 x 10'8
2.39 x 10"8
2.35 x 10'8
2.39 x 10"8
2.33 x 10"8
2.42 x 10~8
2.42 x 10*8
2.08 x 10'8
2.05 x 10'8
2.17 x 10"8
2.05 x 10'8
1.32 x 10'8
1.06 x 10'8
2.10 x 10"8
1.41 x 10"8
SFfi injection
rite, cc/nin
48.77
48.77
48.77
48.77
48.96
48.96
48.96
48.96
46.97
46.97
46.97
46.97
46.68
46.68
46.68
46.68
SF, mass
flow, cc/m1n
50.08
48.58
51.15
50.29
51.15
49.86
51.79
51.79
44.51
43.87
46.44
43.87
47.26
37.95
44.94
30.17
Recovery
efficiency, %
103
99
105
103
104
102
106
106
95
93
99
93
101
81
96
65
78
-------
TABLE 16. SUMMARY OF MATRIX INJECTION TEST DATA, 1/18/83
Sample 1.0.
and Time
B2-3
09:14
B2-4
09:18
Cl-3
09:21
Cl-4
11:12
C6-3
11:15
C6-4
11:18
Dl-3
11:21
Dl-4
11:24
B2-1
12:54
B2-2
12:56
Cl-1
12:59
Cl-2
13:00
C6-1
13:04
C6-2
13:09
Dl-1
13:30
01-2
13:46
B2-1
15:37
B2-2
15:40
Cl-1
15:42
Cl-2
15:59
C6-1
16:04
C6-2
16:07
Dl-1
16:11
01-2
16:14
Date
(1983)
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
1/18
Average gas
flow rate,-
cc/min x 10
(converter mode)
3.58
(slag skim)
3.58
(slag sMm)
2.14
(idle)
3.58
(slag skim)
3.58
(copper pour)
3.58
(copper pour)
3.58
(copper pour)
3.58
(copper pour)
2.14
(copper slag
charge)
3.58
(copper slag
charge)
3.58
(copper slag
charge)
2.14
(matte charge)
2.14
(matte charge)
3.58
(matte charge)
3.58
(matte charge)
3.58
(matte charge)
2.14
(Idle)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(matte charge)
3.58
(natte charge)
3.58
(Mtte charge)
SF, concen-
tration 1n v/v
1.07 x 10"8
1.15 x 10*8
1.85 x 10*8
1.20 x 10"8
1.26 x 10'8
1.05 x 10"8
1.22 x 10"8
1.17 x 10"8
2.12 x 10~8
1.54 x 10'8
1.27 x 10'8
2.00 x 10'8
2.10 x 10'8
1.44 x 10'8
1.96 x 10'8
1.88 x 10"8
1.33 x 10"8
1.41 x 10'8
1.33 x 10"8
1.43 x 10*8
1.30 x 10'8
1.47 x 10'8
1.37 x 10*8
1.51 x 10*8
SFfi Injection
rite, cc/min
47.12
47.12
47.12
47.42
47.42
47.42
47.42
47.42
49.23
49.23
49.23
49.23
49.23
49.23
49.23
49.23
48.82
48.82
48.82
48.82
48.82
48.82
48.82
48.82
SF, mass
flow, cc/m1n
38.31
41.17
39.59
42.96
45.11
37.59
43.68
41.89
45.37
55.13
45.47
42.80
44.94
51.55
70.17
67.30
47.61
50.48
47.61
51.19
46.54
52.63
49.05
54.06
Recovery
efficiency, t
81
87
84
91
95
79
92
88
92
112
92
87
91
105
143
137
98
103
98
105
95
108
101
111
79
-------
TABLE 17. SUMMARY OF MATRIX INJECTION TEST DATA, 1/19/83
Sample I.D.
and Time
Dl-1
12:24
Dl-2
12:27
C6-1
12:20
C6-2
12:22
Cl-1
12:11
Cl-2
12:15
B2-1
12:06
B2-2
12:06
B2-3
13:18
B2-4
13:19
Cl-3
13:21
Cl-4
13:23
C6-3
13:28
C6-4
18:30
Dl-3
13:33
Dl-4
13:25
Date
(1983)
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
Average gas
flow rate,,
cc/min x 10
(converter node)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(cold addition)
3.58
(cold addition)
3.58
(hood up - idle)
2.14
(blow)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(hood up)
SF, concen-
tration in v/v
2.39 x 10'8
2.26 x 10'8
2.17 x 10'8
2.22 x 10'8
2.23 x 10'8
2.32 x 10'8
2.28 x 10~8
2.32 x 10'8
1.58 x 10"8
1.23 x 10'8
1.39 x 10"8
1.53 x 10"8
2.07 x 10"8
1.89 x 10"8
2.18 x 10*8
1.19 x 10"8
SF, injection
rite, cc/min
48.16
48.16
48.16
48.16
48.35
48.35
48.35
48.35
47.70
47.70
47.70
47.70
47.86
47.86
47.86
47.86
SF, mass
flow, cc/min
51.15
48.36
46.44
47.51
47.72
49.65
48.79
49.65
56.56
44.03
49.76
32.74
44.30
40.45
46.65
42.60
Recovery
efficiency, 1
106
101
96
99
99
103
101
103
119
92
104
69
93
85
97
89
80
-------
TABLE 20. SUMMARY OF SAMPLE
DATA FOR SPECIAL INJECTION
AND ANALYTICAL
POINT TESTS
SMple l.D.
•nd TIM
Sf-1
11:30
SP-2
11:34
SP-3
11:37
SP-4
11:40
SP-5
12:00
SP-7
09:18
SP-8
09:20
SP-9
09:23
SP-10
09:27
SP-11
09:30
SP-12
09:32
SP-1 3
09:43
SP-1 4
10:22
SP-1 5
10:24
SP-16
10:27
SP-1 7
10:31
SP-18
10:33
SP-19
10:35
SP-20
08:26
SP-21
08:28
SP-22
08:30
SP-23
08:32
SP-24
08:34
SP-25
08:36
SP-26
08:38
Date
(1983)
1/18
1/18
1/18
1/18
1/18
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/19
1/20
1/20
1/20
1/20
1/20
1/20
1/20
Average
g*i flM r«t|,
cc/«1n x 10'
(converter aode)
3.58
(copper pour)
3.58
(copper pour)
3.S8
(copper pour)
3.58
(copper pour)
2.14
(Idle)
3.58
(slag skin)
3.58
(slag skin)
3.58
(slag skin)
3.58
(slag skin)
3.58
(Mtte charge)
3.58
(Mtte charge)
3.58
(Mtte charge)
2.14
(blow)
2.14
(blow)
2.14
(blow)
3.58
(slag skin)
3.58
(slag skim)
3.58
(slag skim)
3.58
(Mtte charge)
3.58
(Mtte charge)
3.58
(utte charge)
3.58
(Mtte charge)
3.58
(idle)
3.58
(Idle)
3.58
(Mtte charge)
SF, concen-
tration In v/v
1.19 x Uf8
1.29 x UT8
9.58 x 10'9
7.94 x 10~9
1.36 x 10"8
7.14 x 10'9
8.22 x 10'9
8.68 x 10"9
1.08 x 10"8
8.56 x 10'9
1.03 x 10'8
9.52 x 10'9
7.17 x 10"9
7.45 x 10'9
6.85 x 10~9
1.20 x 10"8
9.94 x 10'9
1.10 x 10'9
6.3 xlO-10
5.32 x 10'9
7.20 x 10'9
6.85 x 10'9
6.16 x 10'9
7.04 x 10'9
6.77 x 10"9
SF, Injection
rite, cc/nin
46.92
46.92
46.92
46.92
49.06
49.03
49.03
49.03
49.03
49.03
49.03
48.18
48.18
48.18
48.18
48.18
48.18
48.18
48.25
48.25
48.25
48.25
48.25
48.25
48.25
SF, MSS
flow, cc/nln
42.60
46.18
34.30
28.43
29.10
25.56
29.43
31.07
38.66
30.64
36.87
34.08
15.34
15.94
14.66
42.96
35.59
39.38
2.26
19.05
25.78
24.52
22.05
25.20
24.24
Recovery
efficiency, t
91
98
73
61
59
52
60
63
79
62
75
71
32
33
30
89
74
82
5a
40
53
51
46
52
50
81
-------
TABLE 20 (continued)
Sanple I.D.
tnd T1«e
SP-27
09:28
SP-28
09:49
SP-29
09:53
SP-30
10:02
SP-31
10:33
SP-32
10:36
SP-33
10:36
SP-34
10:46
SP-35
10:55
SP-36
11:00
SP-37
11:02
SP-38
11:04
SP-39
11:06
SP-40
11:09
SP-41
11:11
SP-42
11:14
SP-43
11:16
SP-44
14:25
SP-46
14:33
SP-47
14:35
SP-48
14:37
SP-49
14:39
SP-50
14:42
SP-51
14:45
SP-52
14:47
Date
(1983)
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
Avenge
gas flow rate,
cc/«in x 10s
(converter Bode)
3.58
(Mtte charge)
3.58
(cold addition)
3.58
(cold addition)
3.58
(•atte charge)
2.14
(bio*)
2.14
(blow)
2.14
(bio-)
3.58
(slag skin)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(idle)
3.58
(idle)
3.58
(Idle)
3.58
(matte charge)
3.58
(Mtte charge)
3.58
(cold addition)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skioi)
3.58
(slag skim)
3.58
(Mtte charge)
SF, concen-
tration In v/v
1.07 x 10"8
6.87 x 10~9
6.59 x 10"9
4.68 x 10"9
7.57 x 10'9
6.08 x 10'9
9.44 x 10"9
9.25 x 10'9
8.28 x 10"9
1.16 x 10'8
1.21 x 10"8
5.28 x 10"9
4.18 x 10*9
4.06 x 10'9
9.58 x 10'9
9.86 x 10"9
7.29 x 10'9
1.17 x 10'8
1.69 x 10'8
1.11 x 10'8
1.02 x 10'8
1.23 x 10'8
1.54 x 10'8
9.76 x 10'9
1.20 x UT8
SF, Injection
rite, cc/Hln
46.25
48.25
48.25
48.25
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.62
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
SF, nass
flow, cc/min
38.31
24.59
23.59
16.75
16.20
13.01
20.20
33.20
29.64
41.53
43.32
18.90
14.96
14.53
34.30
35.30
26.03
41.89
60.50
39.74
36.52
44.03
55.13
34.94
42.96
Recovery
efficiency, I
79
51
49
35
34
27
42
70
62
87
91
40
31
30
72
74
55
89
128
84
77
93
117
74
91
82
-------
TABLE 20 (continued)
Staple I.D.
and TIM
SP-53
14:46
SP-54
14: SO
SP-55
14:52
SP-56
14:55
SP-57
14:56
SP-58
14:58
SP-59
15:00
SP-60
15:38
SP-61
15:40
SP-62
15:42
SP-63
15:44
SP-64
15:46
SP-65
15:47
SP-66
15:48
SP-67
15:50
SP-68
15:52
SP-69
15:54
SP-70
16:55
SP-71
16:57
SP-72
17:03
SP-73
17:05
Date
(1983)
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
1/20
Average
gas flow rau,
cc/«1n x 10s
(converter node)
3.58
(matte charge)
3.58
(matte charge)
3.58
(matte charge)
3.58
(cold addition)
3.58
(cold addition)
3.56
(cold addition)
3.58
(roll In)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(Idle)
3.58
(•atte charge)
3.58
(matte charge)
3.58
(slag skim)
3.58
(slag skim)
3.58
(slag skim)
3.58
(Idle)
SFfi concen-
tration In v/v
1.20 x 10*8
1.01 x 10"8
1.09 x 10'8
9.19 x 10~9
1.0 x 1Q'8
8.94 x 10"9
9.10 x 10"9
1.26 x 10"8
1.02 x 10'8
9.16 x 10~9
1.07 x 10'8
1.34 x 10'8
9.55 x 10"9
9.31 x 10"9
1.02 x 10'8
1.02 x 10"8
1.04 x 10'8
1.15 x 10'8
1.36 x 10"8
1.62 x 10"8
1.25 x 10'8
SF, injection
rite, cc/nln
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47*17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
47.17
SF, nass
flOK. CC/Dln
42.96
36.16
39.02
32.90
35.80
32.00
32.58
45.11
36.52
32.79
38.31
47.97
34.19
33.33
36.52
36.52
37.23
41.17
48.69
57.99
44.75
Recovery
efficiency, 1
91
77
83
70
76
68
69
96
77
70
Bl
102
72
71
77
77
79
87
103
123
95
"Excluded from average.
83
-------
SECTION 5
EMISSION FACTOR DEVELOPMENT
This section summarizes emission test results for sulfur
dioxide, filterable particulate, filterable and gaseous arsenic,
particle size distribution (inhalable particulate), and trace
metals in the exhaust stream of the secondary hood or air curtain
system.* Results for each pollutant are reported separately, and
where applicable, emission factors have been developed for over-
all converter operation and specific operational modes (charging,
skimming, and blowing). As noted previously, converter produc-
tion curtailments were caused by stagnant air conditions and
process operational problems resulted in frequently interrupted
test activity. For these reasons, only three converter cycle
segments (Section 3) were evaluated instead of four complete
cycles as originally planned. Except for the basic cycl/; opera-
tions normally encountered during the converter process (i.e.,
matte charges, slag skimming, etc.) specific process operational
details, such as the number of ladles of matte charged, the
number of ladles skimmed, the number and type of other charges
(scrap, copper slag, anode slag, and blister copper), and event
duration, varied greatly. Consequently, between-cycle data
comparisons are likely to show a degree of variability attributed
primarily to process operation.
Appendix A contains example calculations and computer print-
outs of all the emission results. Appendices B and C present
field and laboratory data. Appendix D details the sample and
analytical procedures used during this test program, and Appendix
E summarizes equipment calibration procedures and results.
No measurements were made in the primary hood exhaust stream
which contains the bulk of the converter emissions.
84
-------
5.1 SULFUR DIOXIDE (SC>2)
The S0_ continuous emission monitor (CEM) was set up, cali-
brated, and allowed to stabilize for 24 hours before data col-
lection was begun. The CEM recorded the better part of six
converter cycles between January 14 and January 22, 1983. The
initial matte charge for Charge 76 took place shortly before the
CEM system began sampling. The finish blow and copper pour
emissions at the end of Charge 76 were discarded because of the
failure of a manometer in the CEM sample interface. The CEM
system was returned to service in time to record over 90 percent
of the emissions from Charge 77. The hood SO,, emissions for
Charges 78 through 81 were continuously recorded with only minor
gaps for daily calibration and maintenance. High concentrations
recorded during standby which occurred on January 17 during Cycle
78 and high concentrations which occurred as a result of a pri-
mary hood malfunction on January 22 during Charge 80 were not
included in the summary of charge or cycle emissions.
At the end of data collection, 200 hours of strip chart
recordings were evaluated. The PEDCo operator's logs and nota-
tions on the SO- strip chart identified individual emission
episodes and operating modes. The ASARCO operator's log provided
the number of ladles charged and skimmed and the tons of copper
produced per charge. Under normal operating conditions, the
measured S0_ emission data appeared on the strip chart as a
series of well-defined peaks. Each individual peak was treated
as a separate emission event. The maximum SO,, concentration was
calculated from total peak height. The average S0_ concentration
was calculated by dividing peak area by peak duration. Individ-
ual SO- emission rates were calculated using the average S0_
concentration of each peak and the exhaust flow rates measured at
the air curtain hood test location. These individual emission
rates were multiplied by event duration to yield the mass of SO-
emitted per event.
85
-------
More than 470 individual data points were reduced and used
in the calculation of the overall emission factors presented in
Table 24. The results for Charge 76 are not representative
because a large portion of the cycle could not be included. The
results for Charge 78 include SO- emissions for matte charges
during the early morning hours of January 18 that are much higher
than any other cycle. Based only on the data gathered for Charges
77, 79, 80, and 81, the SO. emission factor for one converter
cycle is estimated to be approximately 7 pounds of S0_ per ton of
copper produced. The average concentration of S02 emitted during
normal operation for all six charges was approximately 90 ppm.
The cycle and mode SO- averages are the sum of the average indi-
vidual event concentrations times the individual event duration
divided by the total duration of events.
Since each peak was integrated from baseline to baseline,
the periods between peaks, defined as zero emissions, were not
included in the average values. The maximum and minimum emis-
sions throughout each cycle varied greatly, as shown in Table 24,
from peaks which exceeded 7000 ppm S0_ to low level emissions
that averaged only 1 ppm S0_.
The S02 emissions within each cycle were evaluated in rela-
tion to specific operating modes. Table 25 summarizes the S0_
emissions that occurred during matte charges. If the data from
Charges 76 and 77 are disregarded as incomplete, the average
number of matte charges per cycle is 14, and S0_ emissions aver-
aged 0.53 Kg/Mg (1.16 Ib per ton) copper produced. The S02
emissions for individual matte charges also varied significantly
from maximum concentrations of more than 1530 ppm to a minimum
concentration of 30 ppm. The average concentration for all matte
charges was approximately 120 ppm S02. The S02 emissions from
slag skimming are summarized in Table 26. The average number of
slag skims for each cycle was 10, and the average S02 emission
rate was 0.36 Kg/Mg (0.8 Ib/ton). The maximum slag skim SO2
emission concentrations did not exceed 572 ppm while the minimum
86
-------
TABLE 24. S02 EMISSION SUMMARY
S02, Kg (Ib)
Copper, Megagrams
(tons)
Emission factor,
Kg/Megagram
(Ib/ton)
Maximum SOp-Concen-
tration, ' ppm
Minimum SOp concen-
tration,9 ppm
Average SOpuConcen-
tration, ppm
Charge No.
76a
174 (384)
163 (180)
.97 (2.13)
>983
4.0
68
77b
390 (859)
100 (110)
3.54 (7.81)
>1330
4.0
83
78C
849 (1872)
113 (125)
6.80 (15.0)
6690
29
112
79
553 (1220)
136 (150)
3.69 (8.13)
>7000
1.0
108
80d
370 (815)
109 (120)
3.08 (6.79)
1780
2.1
89
81
366 (795)
136 (150)
2.40 (5.30)
1165
30
77
CD
Incomplete cycle; initial matte charge, finish blow, and copper pour not included.
Incomplete cycle; initial matte charge not included.
GHigh S09 emissions during standby; overnight 1/16-1/17 not included.
j f.
High S09 emissions during primary hood malfunction not included.
eS09 concentration measured by peak height.
* £•
Parts per million, ppm, dry basis.
9SO, concentration measured by peak height, periods with zero emissions not included.
L C.
Time weighted average of operating mode average concentrations.
-------
TABLE 25.
S02 EMISSION DURING MATTE CHARGES
Number of events
Average duration, min.
Total S02, Kg (Ib)
Average 50,,/event,
Kg (Ib) i
S02/ton Cu, Kg (Ib)
Maximum S0? concen-
tration, ppm
Minimum SOp concen-
tration, ppm
Average S0? concen-
tration, ppm
Charge No.
76
8
4
27 (59.0)
3.4 (7.4)
0.15 (0.33)
981
88
100
77
7
6
17 (38.8)
2.5 (5.5)
0.16 (0.35)
283
56
53
78
13
6
119 (261.3)
9.1 (20.1)
0.95 (2.09)
1210
154
180
79
14
4
39 (85.9)
2.8 (6.1)
0.26 (0.57)
805
106
94
80
13
3
52 (115.5)
4.0 (8.9)
0.45 (1.0)
>1530
122
168
81
16
3
66 (145.1)
4.1 (9.07)
0.44 (0.97)
728
30
124
00
00
aPeriods of zero emissions not included.
-------
TABLE 26. S02 EMISSION DURING SLAG SKIMS
Number of events
Average duration, min.
Total S02, Kg (Ib)
Average S0,/event,
Kg (Ib) z ,
S02/ton Cu, Kg (Ib)
Maximum S0? concen-
tration, ppm
Minimum SCLaconcen-
tration, ppm
Average S02 concen-
tration, ppm
Charge No.
76
10
6
35.7 (78.7)
3.6 (7.9)
0.20 (0.44)
431
46
81
77
7
7
61.3 (135.2)
8.8 (19.3)
0.56 (1.23)
515
166
141
78
13
4
54.7 (120.6)
4.2 (9.3)
0.44 (0.96)
572
86
148
79
9
4
31.8 (70.2)
3.5 (7.8)
0.21 (0.47)
261
97
124
80
10
3
42.5 (93.8)
4.3 (9.4)
0.35 (0.78)
266
112
148
81
11
4
60.9 (134
5.5 (12.
0.40 (0.
358
75
153
00
vo
'Periods of zero emissions not included.
-------
emission concentrations were 46 ppm S0_. The average concen-
tration during all slag skims was approximately 133 ppm. Table
27 summarizes average SO emissions during copper pours. The
emission episodes from this process were the most uniform, with
an average emission factor of 0.30 Kg/Mg (0.67 Ib S0_/ton). The
copper pour emissions generally began with one large peak fol-
lowed by peaks of steadily decreasing size. The maximum peak
concentration was 729 ppm SO and the minimum peak concentration
was 30 ppm S02. The average SO. concentration during all copper
pour was 75 ppm.
The cold addition mode included all ladles of material
charged to the converter except copper matte. Emissions from
these charges, which included shell reverts and scrap copper,
varied significantly throughout the cycles. An occasional ladle
of anode furnace slag was also included in this group, even
though it was in a molten state when charged to the converter.
These anode slag additions generated the most intense SO. emis-
sions episodes, and several peaks were measured at more than 6000
ppm SO . The average emission rates shown in Table 28 reflect
the wide divergence within this group. The overall cold addition
average was 1.2 Kg/Mg (2.65 Ib SO^/ton) copper. The peak emis-
sions concentration during cold additions exceeded the measure-
ment capacity of the S0_ analyzer, which was approximately 7000
ppm. The minimum peak concentrations were 30 ppm S02. The
overall average emission concentrations during cold addition was
approximately 300 ppm.
Table 29 lists S0_ emissions during standby, idle, or blow
modes. Standby mode included those periods when the converter
was on hold due to meteorological production curtailments. Idle
mode included periods of maintenance downtime during a production
cycle or periods of converter inactivity between production
cycles. Emission data collected when the tuyere lines were
blowing are included in this summary. Periods when the converter
was on hold for more than one hour were considered process upsets
and are not included in these data. This group is characterized
90
-------
TABLE 27. S02 EMISSION DURING COPPER POUR
Number of events
Average duration, min.
Total S02, Kg (Ib)
Average S07/event,
Kg (Ib) i
S02/ton Cu, Kg (Ib)
Maximum SOp concen-
tration, ppm
Minimum S02aconcen-
t rat ion, ppm
Average SOpuConcen-
tration, ppm
Charge No.a
77
12
4
26.2 (57.8)
2.2 (4.81)
0.24 (0.53)
374
30
42
78
14
4
37.6 (82.9)
15.2 (6.91)
0.30 (0.66)
729
55
78
79
11
3
52.1 (114.9)
3.7 (8.21)
0.35 (0.77)
496
62
116
80
10
4
56.1 (123.7)
5.1 (11.25)
0.47 (1.03)
644
154
141
81
11
4
25.9 (57.0)
2.6 (5.7)
0.17 (0.38)
243
90
73
vo
No available data from Charge No. 76 copper pour.
'Periods of zero emissions not included.
-------
TABLE 28.
S02 EMISSION DURING COLD ADDITIONS
Number of events
Average duration,
min.
Total S02, Kg (Ib)
Average S09/event,
Kg (Ib) i
S02/ton Cu, Kg (Ib)
Maximum S0? concen-
tration, ppm
Minimum S02 concen-
tration, ppm
Average S02aconcen-
tration, ppm
Charge No.
76
10
4
45.8 (101.0)
4.6 (10.1)
0.25 (0.56)
987
82
131
77
10
7
167.7 (369.6)
16.8 (37.0)
1.5 (3.36)
>1330
126
296
78
9
4
281.1 (619.8)
31.3 (68.9)
2.2 (4.96)
6690
86
536
79
14
5
337.6 (744.3)
24.1 (53.2)
2.2 (4.96)
>7000
36
500
80
10
3
79.2 (174.6)
7.9 (17.5)
0.66 (1.46)
1470
140
274
81
14
3
40.4 (89.0)
2.9 (6.4)
0.27 (0.59)
486
30
99
VO
aPeriods of zero emissions not included.
-------
TABLE 29. S02 EMISSION DURING STANDBY, BLOW, AND IDLE MODES
Number of events
Average duration,
min.
Total S02, Kg (Ib)
Average S09/event,
Kg (Ib) i
S02/ton Cu, Kg (Ib)
Maximum S0? concen-
tration, ppm
Minimum S0? concen-
tration, ppm
Average S0?.concen-
tration, ppm
Charge No.
76
10
18
5.10 (11.0)
0.50 (1.1)
0.03 (0.06)
75
4.0
11
77
14
24
17.6 (38.7)
1.3 (2.8)
0.16 (0.35)
107
4.0
14
78
15
26
43.5 (96.0)
2.9 (6.4)
0.35 (0.77)
257
29
41
79
11
31
1.4 (3.04)
0.14 (0.3)
.009 (0.02)
20
1.0
2.0
80
11
28
8.3 (18.4)
0.77 (1.7)
0.07 (0.15)
260
2.1
20
81
10
20
38.2 (84.16)
3.8 (8.4)
0.25 (0.56)
169
118
20
vo
u>
Periods of zero emissions not included.
'Periods of zero emissions greater than one hour in duration are not included.
-------
by periods of very low S0_ emissions that occur when the primary
hood is down. Most of the data collected during the blow modes
show no detectable S0_. This is due in part to the high instru-
ment range used to detect the intense emission episodes. The
lower detectable limit on the high- and mid-ranges of the S0_
analyzer were 16 ppm and 2.5 ppm, respectively. On one of the
few occasions when the S02 analyzer was switched to the low range
it showed readings of approximately 1 and 2 ppm SO- during the
converter blows. This low level could have resulted from resi-
dual S02 in the system. The distinct peaks with concentrations
over 200 ppm that occurred during the blow mode could represent
fugitive emissions captured from other nearby converters. The
overall average concentrations during the blow mode was less than
20 ppm S02>
The emissions recorded during the converter roll-out mode
(summarized in Table 30) include all the peaks that immediately
precede or follow a converter blow. These emissions are gen-
erated when the primary hood is raised or lowered while air is
blowing through the tuyere lines. The converter operators are
responsible for controlling this aspect; as a result, the con-
verter roll-out emissions range from zero to more than 2000 ppm
S02. The overall average during this mode was 1.05 Kg S02/Mg
(2.32 Ib SO~/ton) copper. The overall average concentration
during the converter roll-out is approximately 330 ppm S02. The
data segments that were not included in the data summaries just
discussed are listed in Table 31. These include all idle periods
of one hour or more in duration, as well as two significant
periods of unusually high or sustained emissions that qualified
as process upsets.
One of the high emission periods was due to a loss in pri-
mary hood draft as a result of operating problems in the chemical
plant. This upset which occurred during Charge 80 was character-
ized by frequent releases of smoke from the primary hood during
the blow mode. The S02 emissions were very intense and irregu-
lar, reaching maximum concentrations of 3420 ppm. The minimum
94
-------
TABLE 30.
S02 EMISSION DURING CONVERTER ROLLS
Number of events
Average duration,
min.
Total S02, Kg (Ib)
Average S09/event,
Kg (Ib) i
S02/ton Cu, Kg (Ib)
Maximum S0? concen-
tration, ppm
Minimum S0« concen-
tration, ppm
Average S02 concen-
tration, ppm
Charge No.
76
1 n
L\J
4
61 (134.4)
6.1 (13.4)
0.34 (0.75)
>983
86
212
77
7
4
127.8 (281.7)
14.2 (31.2)
0.90 (1.99)
>727
143
399
78
19
3
313.8 (691.8)
16.5 (36.4)
2.5 (5.53)
2300
62
472
79
1 O
± L.
3
92.4 (203.7)
7.7 (17.0)
0.62 (1.36)
1010
77
321
80
15
3
130.9 (288.5)
8.7 (19.2)
1.09 (2.4)
1780
18
320
81
16
5
129.5 (285.4)
8.1 (17.8)
0.86 (1.9)
1170
154
255
vo
U1
aPeriods of zero emissions not included.
-------
TABLE 31. S02 EMISSIONS DURING UPSET CONDITIONS'
Mode
Duration, min.
Total S02, Kg (Ib)
Maximum SO,, con-
centration,
ppm
Minimum S02 con-
centration,
ppm
Average S02 con-
centration,
ppm
Charge No.
77
04--.— JU..
o lanuujr
583
0.51 (1.12)
30.0
-
<1.0
78
Standby
568
176.7 (389.6)
83
40
59
78
Standby
1032
0.0
0.0
-
0.0
79
C+^v^kw
o u u 1 1 u u y
904
0.0
0.0
-
0.0
80
Standby
92
0.0
0.0
-
0.0
80
D v* -J ma v*\/
i i i MIU i jr
hood
leak
82
174 (383.7)
3420
204
889
81
C + anrlKt/
\J VU 1 IVJfc/^J
1902
0.0
0.0
-
0.0
vo
These values are not included in previous data summaries.
-------
S0_ concentrations fell to 204 ppm between puffs but the overall
average concentration was close to 1000 ppm. The other high
emission period occurred during a standby mode during Charge 78.
The emissions were characterized by consistent low level emis-
sions. One S02 peak did occur during this time reaching a maxi-
mum concentration of 128 ppm. The average concentration was
approximately 50 ppm and the emission slowly tapered off toward
the end of this period to reach the minimum value of 31 ppm. The
emission episodes described above occurred during apparent upset
conditions. A comparison was made between these two upsets and
similar operations which were conducted during the same charge
cycle after the upset was corrected. This data is presented in
Table 32. The normal operation data chosen for this table was
collected on the low range of the analyzer. It is characterized
by flat traces which taper towards zero. As seen on Table 29,
some peaks did occur during the blow, standby, and idle modes.
In general, the periods of operation when the primary hood was
down had very low SO emissions which would indicate efficient
primary hood operation.
A detailed summary of emissions is given in Appendix B,
along with examples of the strip charts. The data reduction
methods are included in Appendix D.
5.2 FILTERABLE PARTICULATE EMISSIONS
Table 33 summarizes the flue gas conditions and particulate
emissions data collected during the particulate/arsenic test
runs. Volumetric flow rates are expressed in cubic meters per
second and actual cubic feet per minute at stack conditions.
Flow rates corrected to standard conditions [20°C and 760 mm Hg
(68°F and 29.92 in.Hg) and zero percent moisture] are expressed
as dry normal cubic meters per minute and dry standard cubic feet
per minute. Particulate concentrations are reported in milli-
grams per dry normal cubic meter and grains per dry standard
cubic foot. Emission rates are expressed in kilograms per hour
and pounds per hour. The product of the concentration and the
97
-------
TABU: 32. COMPARISON OF so9 EMISSION DURING NORMAL
AND UPSET PRIMARY HOOD OPERATION
Primary hood
operation
Converter mode
Duration, min.
Maximum S0? con-
centration, ppm
Minimum S02 con-
centration, ppm
Average SOp con-
centration, ppm
Total S02, Ib
Charge No.
78
Upset
Standby
558
128
31
50
352
79
Normal
Finish blow
and standby
92
1.7
0.6
1.2
1.36
80
Upset
Blow
82
3420
204
889
383
80
Normal
Blow
62
3.0
1.0
2.0
1.55
98
-------
TABLE 33. SUMMARY OF FILTERABLE PARTICULATE EMISSIONS DATA
Run No.
(Charqt NO.)
Date
(1983)
Stapling
period
Sampling
time,
•In.
Sample voluM
dN»> ] dscf
Filterable
parttculate
mass, mg
Filterable panicu-
late concentration
ng/dNrn1 | gr/dscf
Filterable
partlculate
emission rate
kg/h 1 Ib/h
Flow rate
Standard
dNm'/mln 1 dscfm
Actual
m'/mln
acfw
Tenper-
ature
•C | °F
HolS-
ture, I
CO.,
Y
Isokl-
netlcs, 1
Total cycle
»*TC-1
(79)
MTC-2
(80)
r*TC-3
mn .
1/18
1/19
1/20
1/22
0806-1604
074S-1403
0820-2003
0910-2343
650.0
470.0
453.0
11.218
6.657
7.474
396.173
235.103
263.952
364.9
307.5
153.6
Average
32.5
46.2
20.6
33.1
0.014
0.020
0.009
0.014
5.2
7.3
3.2
5.2
11.5
16.1
7.1
11.6
2,680
2.630
2.610
2.640
94,700
92,900
92,100
93.200
2,780
2.690
2.660
2.710
98,000
94,900
94,100
95.700
20
21
17
19
68
69
63
67
0.68
0.77
0.84
0.76
0.0
0.0
0.0
0.0
20.9
20.9
20.9
20.9
103.0
86.2
101.3
-
Specific mode
PASM-1
MSM-2
(80)
MSM-3
(UK-
1/18
1/19
1/20
1/22
0909-1625
0830-1322
0820-1832
0955-1930
233.5
131.6
40.4
3.475
1.885
0.638
122.709
66.565
22.536
227.8
194.9
62.4
Average
65.6
103.4
97.8
88.9
0.029
0.045
0.043
0.039
14.3
22.2
21.2
19.2
31.6
49.0
46.8
42.5
3,571
3.571
3.571
3.571
126.924
126.924
126.924
126.924
3,552
3.552
3,552
3,522
126,238
126.230
126.230
126,230
27
23
22
24
80
74
72
75
0.71
0.37
1.16
0.75
0.0
0.0
0.0
0:0
20.9
20.9
20.9
20.9
100.0
100.8
102.2
-
-------
volumetric flow rate is the mass emission rate. For Tests PATC
1-3 (converter Cycle Nos. 79, 80, and 81, respectively), the
measured flow rate obtained from traversing the exhaust flue was
used in the calculation of mass emission rates. For Tests
PASM 1-3, however, the average flow rate obtained from the vol-
umetric flow evaluation of the high flow mode [3571 dNm3/min
(126,924 dscfm)] was used because testing was performed at a
single point in the exhaust flue. These data are summarized in
Table 11 (Section 4 pg. 36). The filterable particulate fraction
represents material collected in the sample probe and on the
filter, both of which were heated to approximately 121°C (250°F).
During total cycle testing, the volumetric flow rate (cor-
rected to standard conditions) averaged 2640 dNm3/min (93,200
dscfm), whereas the actual flow rate averaged 2710 m3/min (95,700
acfm). The flue gas temperature averaged 19°C (66°F) and the
moisture content averaged 0.76 percent; carbon dioxide and oxygen
contents were 0.0 and 20.9 percent, respectively. The filterable
particulate concentration ranged from 20.6 mg/dNm3 (0.009 gr/dscf)
to 46.2 mg/dNm3 (0.02 gr/dscf).
For the specific mode tests, the filterable particulate
concentration ranged from 65.6 mg/dNm3 (0.029 gr/dscf) to 103.4
mg/dNm3 (0.045 gr/dscf). Test PASM-3 was performed only during
slag skimming and the measured concentration was 97.8 mg/dNm3
(0.043 gr/dscf).
The filterable particulate concentration during Test No.
PATC-2 was higher than the concentrations measured during the
other two tests. At the conclusion of the test, a leak rate of
0.08 ft3/min was found. The metered sample volume was corrected
for this leak rate by multiplying the total sample time in min-
utes by the leak rate and subtracting the resulting value from
the metered volume. Using the leak corrected sample volume, a
particulate concentration of 46.2 mg/dNm3 (0.02 gr/dscf) was
calculated and reported in Table 33. Because the leak is be-
lieved to have occurred at the port change, correcting the sample
100
-------
volume for only the second port sample would result in a correc-
ted particulate concentration of 0.018 gr/dscf. Particle size
distribution results reported in Section 5.4 show the majority of
particles to be less than 10 micrometers in diameter; therefore,
the nonisokinetic sample condition calculated as a result of the
leak correction (Table 33) would not have a significant impact on
test results. Also, the loss of primary hood draft resulting
from operating problems at the chemical plant occurred during
this test run. As noted in Subsection 5.1, this upset was char-
acterized by frequent releases of smoke and fumes from the pri-
mary hood during the blow mode. The full cycle sample train was
run until the smoke releases totally overwhelmed the secondary
hood at which time all sampling was terminated until repairs were
completed. Since this condition did not occur during Tests
PATC-1 or PATC-3, an increase in emission results for PATC-2
would be expected.
With regard to the development of emission factors, process
curtailments and operational inconsistencies make it impractical
to develop representative emission factors for the overall cycle
on any basis other than a pound particulate per ton copper pro-
duced. Table 34 summarizes filterable particulate emission
factors for the full-cycle (PATC) and specific-mode (PASM) tests.
As discussed previously, Test PATC-2 is probably biased high
because of an excessive post-test leak rate and subsequent sample
volume correction and the loss of primary hood draft resulting in
increased emissions during the blow mode. Also, Test PASM-3 was
performed during slag skimming operations only, A total of 7.25
ladles of slag were skimmed from the converter during this test.
Based on information supplied by ASARCO, each ladle contains 12
to 15 tons of slag. Therefore, during this test, between 87 and
109 tons of slag were skimmed, which yields an emission factor
ranging from 0.29 Ib of particulate per ton of slag skimmed to
0.36 Ib/ton.
101
-------
TABLE 34. PARTICULATE EMISSION FACTOR DEVELOPMENT
Converter
cycle No.
1
2
3
Test
I.D.
PATC-1
PASM-1
PATC-2
PASM-2
PATC-3
PASM-3C
Date
(1983)
1/18-19
1/20
1/22
Tons copper
produced
150
120
150
Total
particu-
late, Ib
124.6
123.0
126.0
107.5
53.6
31. 5C
Emission factor,
Ib participate/
ton copper
produced
0.83
0.82
1.05
0.90
0.36
0.21C
Information obtained from ASARCO converter operation logs.
Product of particulate mass emission rate and time of test.
c Slag skimming only.
The particulate emission results indicate that the majority
of emissions are generated when the primary hood is raised, i.e.,
charging and skimming. This is substantiated by visual observa-
tions and the relative equality of the full cycle and specific
mode emission results; specifically the total particulate and
emission factor data presented in Table 34. Also, the results
from the third converter cycle test (PATC-3 and PASM-3) clearly
show the effects of variable converter cycle operation since the
number of matte charges, slag skims, and cold additions were
significantly less than the other converter cycle tests (see
Section 3).
5.3 FILTERABLE AND GASEOUS ARSENIC
Table 35 summarizes the filterable and gaseous arsenic
emissions data for tests conducted by EPA Reference Methods 5 and
108.* Two sampling trains were used to obtain the particulate
40 CFR 60, Appendix A, Reference Method 5, July 1, 1982.
108 has been proposed and is in draft form.
Method
102
-------
TABLE 35. SUMMARY OF FILTERABLE AND GASEOUS ARSENIC EMISSION DATA
Cycle
Test
No.
1
2
3
Run
No.
PATC-1
PASM-1
PATC-2
PASM-2
PATC-3r
PASM-3C
Date
(1983)
1/18-19
1/20
1/22
Sampling
penoa
1/18 909-1625
1/19 830-1322
820-2003
910-2342
Concentration
Filterable
mg/aiNm-' ^gr/ascr;
2.18 (0.0009)
4.98 (0.002)
3.89 (0.0017)
9.01 (0.004)
1.35 (0.0006)
5.86 (0.003)c
Gaseous
mg/dNm3 (gr/dscf)
0.28 (0.0001)
0.86 (0.0004)
5.02 (0.002)
4.72 (0.002)
0.44 (0.0002)
0.24 (0.0001)c
Mass emission rate
Filterable
Kg/n (\o/n)
0.33 (0.73)
0.99 (2.18)
0.61 (1.35)
1.97 (4.35)
0.21 (0.47)
1.48 (3.26)c
Gaseous
Kg/n v iD/n;
0.04 (0.08)
0.20 (0.44)
0.72 (1.59)
0.99 (2.18)
0.07 (0.16)
0.05 (0.11)c
o
oo
aFilterable and gaseous arsenic concentration in milligrams per dry normal cubic meter (mg/dNm3) and
grains per dry standard cubic foot (gr/dscf). Standard conditions: 760 mmHg (29.92 in.Hg), 20°C
(68°F), and 0 percent moisture.
Mass emission rate in kilograms per hour (kg/h) and pounds per hour (Ib/h) calculated using measured
concentrations and volumetric flow rates reported in Table 5-9.
cSlag skim only.
-------
and arsenic samples. Sampling was performed for the duration of
each converter cycle tested and during specific converter rollout
modes: matte charge, slag skim, cold addition, and copper pour-
ing. Analysis for filterable and gaseous arsenic was performed
at the completion of the gravimetric particulate determination.
Arsenic concentrations are reported in milligrams per dry
normal cubic meter and grains per dry standard cubic foot.
Emission rates are expressed in kilograms per hour and pounds per
hour. The product of the concentration and the volumetric flow
rate is the mass emission rate. For the total cycle tests (de-
signated PATC), the measured flow rate obtained from the sample
traverse was used in the calculations. For tests conducted
during converter rollout activities (designated PASM), the aver-
age flow rate obtained from the volumetric flow evaluation of the
high-flow mode was used because these tests were performed at a
single point in the duct. Volumetric flow data were summarized
earlier in Sections 4 and 5.2.
The filterable arsenic fraction represents material col-
lected in the sample probe and on the filter, both of which were
heated to approximately 121°C (250°F). The gaseous arsenic
fraction represents material that passed through the heated
filter and condensed or was trapped in the impinger section of
the sample train, which was maintained at a temperature of 20°C
(68°F) or less.
During the total cycle tests, the filterable arsenic concen-
tration ranged from 1.35 mg/dNm3 (0.0006 gr/dscf) to 3.89 mg/dNm3
(0.0017 gr/dscf), and the corresponding mass emission rates
ranged from 0.21 kg/h (0.47 Ib/h) to 0.61 kg/h (1.36 Ib/h).
Gaseous arsenic concentrations during Tests PATC-1 and PATC-3
were 0.28 mg/dNm3 (0.0001 gr/dscf) and 0.44 mg/dNm3 (0.0002
gr/dscf), respectively.
During Test PATC-2, the gaseous arsenic concentration was
5.02 mg/dNm3 (0.002 gr/dscf). As noted in Subsections 5.1 and
5.2, the loss of draft by the primary hood caused by operational
problems at the chemical plant resulted in frequent releases of
104
-------
smoke and fumes from the primary hood. During this period,
particularly in the converter blow mode, heavy volumes of smoke
escaped the primary hood system, and some of these emissions were
captured by the secondary hood. Sampling continued throughout
these intermediate upsets, but was finally stopped when the air
curtain control system became overwhelmed by continuous and heavy
emission discharge from the primary hood. Therefore, it is
reasonable to conclude that fugitive emissions generated by the
malfunctioning primary hood draft contributed to the higher
arsenic concentrations observed during the second cycle test.
During the specific mode tests, filterable arsenic concen-
trations ranged from 4.98 mg/dNm3 (0.002 gr/dscf) to 9.01 mg/dNm3
(0.004 gr/dscf),, and corresponding emission rates ranged from
0.99 kg/h (2.18 Ib/h) to 1.98 kg/h (4.35 Ib/h). Gaseous arsenic
concentrations ranged from 0.24 mg/dNm3 (0.0001 gr/dscf) to 4.72
mg/dNm3 (0.002 gr/dscf) and the corresponding emission rates
ranged from 0.05 kg/h (0.11 Ib/h) to 0.99 kg/h (2.18 Ib/h).
Table 36 presents total arsenic emission factors on the
basis of a pound of arsenic per ton of copper produced. The
total arsenic value for each run was calculated by adding the
filterable and gaseous fractions (in milligrams), calculating the
concentration arid mass emission rate (in pounds per hour) , and
multiplying the mass emission rate by the time of the test (in
hours).
Arsenic emission factors for the total cycle tests ranged
from 0.03 Ib/tori to 0.20 Ib/ton.
Arsenic emission factors for specific mode Tests PASM-1 and
2 were 0.07 Ib/ton and 0.12 Ib/ton, respectively.
For Test PASM-3, which was run only during slag skimming
operations, the arsenic emission factor was 0.02 Ib/ton of copper
produced. During this test, a total of 7.25 ladles of slag were
skimmed from the converter. Based on information supplied by
ASARCO, each ladle contains 12 to 15 tons of slag. Therefore,
between 87 to 109 tons of slag were skimmed, which yields a
skimming emission factor of about 0.025 Ib of arsenic per ton of
slag skimmed.
105
-------
TABLE: 36. DEVELOPMENT OF ARSENIC EMISSION FACTORS
Converter
cycle No.
1
Charge No. 79
2
Charge No. 80
3
Charge No. 81
Test
I.D.
PATC-1
PASM-1
PATC-2
PASM-2
PATC-3
PASM-3C
Date
(1983)
1/18-19
1/20
1/22
Tons copper
produced
150
120
150
K
Total0
arsenic, Ib
9.48
10.80
24.27
14.32
4.78
2.61C
Emission factor,
Ib arsenic/ton
copper produced
0.06
0.07
0.20
0.12
0.03
0.02C
Information obtained from ASARCO converter operation logs.
''Total arsenic obtained by adding the filterable and gaseous fractions (in
milligrams), calculating the concentration and subsequent mass emission
rate (in pounds per hour), and multiplying the mass emission rate by the
time of test (in hours).
"Slag skimming only.
In summary, the arsenic emission data show that the majority
of these emissions are generated when the primary hood is in the
raised position. This is substantiated by the relative equality
of the total arsenic and emission factor data presented in Table
36; specifically converter cycle Tests 1 and 3. Results from
converter cycle Test 2 show the impact of the primary hood mal-
function on arsenic emissions. The results from the full cycle
sample train, run primarily during the blowing mode, are signifi-
cantly higher than the specific mode (roll-out activity) sample
train. Both the particulate and arsenic emission data indicate
that, when functioning properly, the primary hood on the No. 4
converter is very effective in controlling emissions during the
blowing mode.
106
-------
5.4 PARTICLE SIZE RESULTS
Particle size distribution tests were conducted during three
different operating modes. One segment of the converter cycle
that was tested was the charging mode, which consisted of all
additions to the converter (matte charge, cold additions, anode
additions, and copper slag additions). Two other segments were
also tested—the skimming mode, which consisted of slag skims and
the copper pour, and the blowing mode, which included the slag,
cleanup, and finish blows.
The particle size distribution tests were conducted at
points of average velocity in the duct that exhausts emissions
collected by the air curtain hooding system. These tests were
conducted simultaneously with the particulate/arsenic tests (at
separate sampling points). The particle size distribution re-
sults are presented as the actual measured results during each
sample run. Inhalable particulate fractions, percentage less
than 10 to 2.5 pm, are also presented.
5.4.1 Measured Particle Size Emission Results
Table 37 presents the particulate loading, impactor flow
data, and particle size cut points for each particle size run.
During the charging mode sample runs, the particulate loading
measured on the first filter stage, in the impactor rinse, and on
the last four filter stages accounted for most of the particulate
collected by the impactor. The sampling rates in all of the
particle size runs conducted during the charging mode were within
±4 percent of the isokinetic value. (The isokinetic rate is the
ratio of the velocity of the sample gas stream entering the
nozzle to the velocity of the stack gas, expressed as a percent-
age.) Actual impactor performance limited the aerodynamic size
ranges that could be determined during the charging mode particle
size runs to a maximum diameter of 9.4 ym and a minimum diameter
of 0.3 ym.
During the skimming mode, the particulate loading for the
three sample runs on the individual impactor filter stages did
107
-------
TABLE 37. PARTICIPATE LOADING AND IMPACTOR FLOW RATE DATA FOR THE PARTICLE SIZE RUNS
Run No.
Date
(1983)
Sampling
duration
Impactor
flow rate,
a cms (acfm)
Isokln-
ettcs,
I
Average stack
temperature,
°C (°F)
Paniculate loading, rag
Stage No.
Rinse
+ 0
1
2
3
4
5
6
7
Backup
Charging
PSNC-1
cutpolnt urn
PSMC-2
cutpolnt urn
PSMC-3
cutpolnt urn
PSMC-4
cutpotnt urn
PSMC-5
cutpoint u«
1/1B-
1/19
1/19
1/20
1/20
1/22
io6-i62j
0835-0845
0936-1322
0820-1121
1450-1838
1047-2120
1.18 (0.693)
1.16 (0.685)
1.09 (0.643)
1.09 (0.644)
1.05 (0.615)
!M
102
100
102
96
29 (85}
25 (77)
28 (83)
27 (81)
21 (69)
28.8
8.9
9.6
8.9
13.0
9.2
12.7
9.2
20.4
9.4
8.8
7.9
3.2
7.8
5.3
8.2
6.0
8.1
5.4
8.3
6.5
5.1
1.6
5.0
3.2
5.3
11.4
5.2
4.2
5.4
3.5
3.3
1.3
3.2
2.0
3.4
4.4
3.3
2.4
3.5
5.3
1.8
1.3
1.7
2.1
1.9
3.8
1.8
3.8
1.9
13.4
0.9
2.5
0.8
9.7
1.0
3.0
0.9
8.9
1.0
19.4
0.7
4.3
0.6
10.8
0.7
4.0
0.6
7.2
0.7
17.6
0.4
7.2
0.3
10.5
0.4
4.5
0.3
4.6
0.4
12.6
<0.4
7.3
<0.3
12.2
<0.4
4.4
<0.3
4.5
<0.4
o
00
Skinning mode
PSSS-1
cutpolnt inn
PSSS-2
cutpolnt urn
PSSS-3
cutpoint w*i
1/18-
1/19
1/20
1/22
0907-1606
0812-1039
1045-1708
0955-1730
1.33 (0.785)
1.20 (0.709)
1.22 (0.717)
112
111
111
26 (78)
23 (74)
24 (75)
16.3
8.3
9.3
8.7
7.9
8.7
8.7
7.3
5.6
7.7
2.8
7.7
8.3
4.7
8.2
4.9
0.0
5.0
4.2
3.0
2.2
3.2
1.5
3.2
5.4
1.6
6.0
1.7
1.8
1.8
6.6
0.8
7.8
0.8
5.0
0.9
16.5
0.5
16.0
0.6
13.0
0.7
18.4
0.3
12.3
0.3
10.4
0.4
26.7
<0.3
6.9
<0.3
5.7
<0.4
Blowing node
PS8-1
cutpolnt urn
PSB-2
cutpolnt urn
PSB-3
cutpolnt un
PSB-4
cutpolnt pit
1/18-
1/19
1/20
1/20
1/22
0813-1530
0732-1401
1004-1418
1616-2000
1607-2211
0.760 (0.447)
0.822 (0.484)
0.826 (0.486)
0.829 (0.488)
89
100
105
99
18 (64)
17 (63)
18 (65)
17 (63)
67.0
10.9
45.7
10.4
58.2
10.4
31.4
10.4
0.3
9.6
0.6
9.2
0.2
9.2
0.0
9.2
0.3
6.2
0.8
6.0
0.0
6.0
0.5
5.9
0.7
4.0
1.3
3.8
0.0
3.8
0.3
3.8
0.6
2.2
2.6
2.1
0.2
2.1
0.2
2.1
1.2
1.1
8.8
1.0
0.3
1.0
0.3
1.0
2.2
0.7
9.7
0.7
0.2
0.7
0.8
0.7
2.6
0.4
7.5
0.4
0.2
0.4
0.8
0.4
2.3
<0.4
3.0
<0.4
0.6
<0.4
0.8
<0.4
-------
not follow a distinguishable pattern. The sampling rates of all
of the particle size runs conducted during the skimming mode were
within 12 percent of the isokinetic value. The aerodynamic size
ranges determined for these particle size runs varied from a
maximum diameter of 8.7 ym to a minimum diameter of 0.3 ym.
During the blowing mode sample runs, the particulate loading
measured on the first filter stage and the impactor rinse (which
included the material collected in the cyclone precutter) ac-
counted for an average of 83 percent of the total particulate
loading during the four sample runs. The sampling rates of all
of the particle size runs conducted during the blowing mode were
within ±11 percent of the isokinetic value. Because of the
expected low grain loading at the sampling location during the
blowing mode, the sampling train was operated at the maximum
constant flow rate through the impactor consistent with isoki-
netic sampling. For this reason the 15-ym cyclone precutter did
not provide a 15-ym cut point. The aerodynamic size ranges
determined for the particle size runs conducted during the blow-
ing mode ranged from a maximum diameter of 10.9 ym to a minimum
of 0.4 ym.
The cumulative and fractional particulate concentrations and
the emission rates for each particle size run are presented in
Tables 38 and 39.
The particulate concentrations and emission rates for the
eight size ranges were calculated for each sample run by multi-
plying the total measured value by the cumulative weight percent
less than each cut point, as determined from the individual
particle size distribution curves.
The particulate concentrations for the particle size runs
are reported in milligrams per dry normal cubic meters (mg/dNm3)
and grains per dry standard cubic feet (gr/dscf). Emission rates
are expressed in kilograms per hour and pounds per hour. The
product of the concentration measured by the particle size runs
and the average volumetric flow rate measured by the preliminary
velocity traverses is the mass emission rate.
109
-------
TABLE 38. SUMMARY OF FILTERABLE PARTICULATE CONCENTRATIONS FOR THE PARTICLE SIZE RUNS
Kun No.
rwe-i
cutpotnt unt
PSHC-2
cutpolnt uin
cutpolnt no
PSHC-4
cutpolnt u»
PSC-S
cutpolnt am
Oil*
(1983)
1/IS-
1/19
1/19
!/2B
1/20
1/22
t(tf
TgToW*
165
94
9S
13?
157
ll .
gr/Jscr
0.072
0.041
n nai
O.OSS
o.osa
Stage 0
nq/dNm'l gr/dscf
124
8.9
70
8.9
•0
9.2
101
9.2
IDS
9.4
0.054
0.031
0.035
0.044
0.046
Stan 1
mg/dtaii' j gr/d«f
HI
7.8
63
7.8
n
6.2
a;
B.I
91
8.3
0.049
0.027
0.03!
0.038
0.040
Cumulative lett thin Indicated flit
Stl« 2
•q/dlttn^ gr/dtcf
Charging mode
102
S.I
59
5.0
ft?
5.3
59
5.2
80
5.4
0.045
0.026
n,n?9
0.026
0.035
Stas
mg/dlfcn'
97
3.3
55
3.2
M
3.4
48
3.3
74
3.5
r }
gr/dscf
0.042
0.024
n,n?R
0.021
0.032
Stage 4
ing/dkni'j gr/<)scF
M»9« 5
•g^o*"1! gr/ifacf
SU««
•oVdNin^ grH,cf
StM
•1/O.W1
89
1.8
52
1.7
ft?
1.9
39
1.8
64
1.9
0.039
0.023
0,077
0.017
0.028
70
0.9
46
0.8
411
1.0
32
0.9
42
1.0
0.031
0.020
0,071
0.014
0.018
43
0.7
M
0.6
12
0.7
22
0.6
23
0.7
0.019
0.016
0.014
0.009
0.010
IB
0.4
IB
0.3
17
0.4
11
0.3
12
0.4
f '
jrJKcr
0.008
0.008
0.008
0.005
O.OOS
SklMlng I
PSSS-I
cutpolnt (m
PSSS-2
cutpofnt u*
PSSS-3
cutpolnt u«
1/18-
1/19
1/20
1/22
40
77
64
0.018
0.033
0.028
35
8.3
67
a.;
54
a.;
0.015
0.029
0.023
31
7.3
61
7.7
50
7.7
0.014
0.027
0.022
28
4.7
53
4.9
50
5.0
0.012
0.023
0.022
27
3.0
51
3.2
4B
3.2
0.012
0.022
0.021
25
1.6
44
1.7
45
l.B
0.011
0.019
0.020
22
O.B
36
O.B
31
0.9
0.010
0.016
0.017
16
0.5
»
0.6
21
0.7
0.007
0.001
o.oo*
10
0.3
7
0.3
B
0.4
0.004
0.00]
0.001
Blowing i
cse-i
cutpolnt UM
PSB-2
cutpolnt in
rSB-3
cutpolnt iff
PSB-4
cutpolnt w
1/18-
1/19
1/20
1/20
1/22
25
107
45
1]
0.011
0.047
0.020
0.006
3
10.9
46
10.4
1
10.4
1
10.4
0.001
0.020
0.0006
0.0006
3
9.6
45
9.2
,
9.2
1
9.2
0.001
0.020
0.0005
0.0006
3
6.2
44
6.0
,
6.0
1
5.9
0.001
0.019
C.0005
C.0005
3
4.0
42
3.8
1
3.8
1
3.8
0.001
o.oie
0.0005
0.0005
3
2.2
39
2.1
1
2.1
1
2.1
0.001
0.017
0.0004
0.0004
2
1.1
27
1-0
1
1.0
1
1.0
0.001
0.012
0.0003
0.0001
2
0.7
14
0.7
1
0.7
1
0.7
0.0007
0.006
0.0003
0.0003
|
0.4
4
0.4
0.0e
0.4
o.oe
0.4
0.000]
o.on
0.0002
0.0001
w per dry nomal cubic utter.
Grain? per dry standard cubic foot.
-------
TABLE 39. SUMMARY OF FILTERABLE PARTICULATE EMISSION RATES FOR THE PARTICLE SIZE RUNS3
Run No.
Date
(1983)
Cumulative less than Indicated size
Total
kg/hb|.1b/hc
Stage 0
kg/h [ Ib/h
Staae 1
kg/h | Ib/h
Stage 2
kg/h I Ib/h
Staae 3
kg/h Ib/h
Stage 4
kg/h f Ib/h
Stage 5
kg/h Ib/h
Stage 6
kg/h j Ib/h
Stage 7
kg/h | Ib/h
Charging mode
PSNC-1
cutpolnt um
PSHC-2
cutpolnt um
PSMC-3
cutpolnt um
PSHC-4
cutpolnt um
PSHC-5
cutpolnt um
1/18-
1/19
1/19
1/ZO
1/20
\m
35.4
20.1
21. 0
28.3
33.6
78.3
44.6
46.8
63.1
74.0
26.6
8.9
15.0
8.9
17.1
9.2
21.6
9.2
22.5
9.4
58.7
33.7
38.1
47.9
50.0
23.8
7.8
13.5
7.8
15.4
8.2
18.6
8.1
19.5
8.3
53.3
29.4
33.7
41.3
43.5
21.9
5.1
12.6
5.0
14.4
5.3
12.6
5.2
17.1
5.4
49.0
28.3
31.5
28.3
38.1
20.8
3.3
11.9
3.2
13.9
3.4
10.3
3.3
15.9
3.5
45.7
26.1
30.5
22.8
34.8
19.1
1.8
30.2
1.7
13.3
1.9
8.4
1.8
13.7
1.9
42.4
25.0
29.4
18.5
30.5
15.0
0.9
9.9
0.8
10.3
1.0
6.9
0.9
9.0
1.0
33.7
21.8
22.8
15.2
19.6
9.2
0.7
7.7
0.6
6.9
0.7
4.7
0.6
4.9
0.7
20.7
17.4
15.2
9.8
10.9
3.9
0.4
3.9
0.3
3.6
0.4
2.4
0.3
2.6
0.4
8.7
8.7
8.7
5.4
5.4
Skinning mode
PSSS-1
cutpolnt um
PSSS-2
cutpolnt um
PSSS-3
cutpolnt um
1/18-
1/19
1/20
1/22
8.6
16.5
13.7
19.6
35.9
30.5
7.5
8.3
14.4
8.7
11.6
8.7
16.3
31.5
25.0
6.6
7.3
13.1
7.7
10.7
7.7
15.2
29.4
23.9
6.0
4.7
11.4
4.9
10.7
5.0
13.1
25.0
23.9
5.8
3.0
10.9
3.2
10.3
3.2
13.1
23.9
22.8
5.4
1.6
9.4
1.7
?-6
1.8
12.0
20.7
21.8
4.7
0.8
7.7
0.8
8.4
0.9
10.9
17.4
18.5
3.4
0.5
4.3
0.6
4.5
0.7
7.6
9.8
9.8
2.1
0.3
1.5
0.3
1.7
0.4
4.4
3.3
3.3
Blowing mode
PSB-1
cutpolnt um
PSB-2
cutpolnt um
PSB-3
cutpolnt um
PSB-4
cutpolnt um
1/18-
1/19
1/20
1/20
1/22
3.2
13.8
5.8
1.7
7.2
30.8
13.1
3.9
0.4
10.9
5.9
10.4
0.1
10.4
0.1
10.4
0.7
13.1
0.4
0.4
0.4
9.6
5.8
9.2
0.1
9.2
0.1
9.2
0.7
13.1
0.3
0.4
0.4
6.2
5.7
6.0
0.1
6.0
0.1
5.9
0.7
12.4
0.3
0.3
0.4
4.0
5.4
3.8
0.1
3.8
0.1
3.8
0.7
11.8
0.3
0.3
0.4
2.2
5.0
2.1
0.1
2.1
0.1
2.1
0.7
11.1
0.3
0.3
0.3
1.1
3.5
1.0
0.1
1.0
0.1
1.0
0.7
7.9
0.2
0.3
0.3
0.7
1.8
0.7
0.1
0.7
0.1
0.7
0.5
3.9
0.2
0.2
0.1
0.4
0.5
0.4
0.0d
0.4
0.0d
0.4
0.2
1.3
0.1
0.07
'Emission rate based on average volumetric flow rates of 2149 dHm'/raln (76,359 dscfm) during low flow (blow mode) and 3571 dlta'/mln (126,924 dscfm) durtttg
high flow (charging and slaglng modes)
Kilograms per hour.
GPounds per hour.
-------
The moisture content measured by the corresponding particu-
late/arsenic test was used to calculate the particle size param-
eters for each sample run.
During the charging mode, the total particulate concentra-
tion measured by the particle size runs ranged between 94 mg/dNm3
(0.041 gr/dscf) and 165 mg/dNm3 (0.072 gr/dscf). One reason the
particulate concentration varied during the charging mode is that
the different materials charged into the converter during each
tested cycle caused significant variations. The visible emis-
sions varied during the charging modes. Another reason is that
the converter was not operated consistently throughout the charg-
ing mode. During some of the charging segments, the converter
was rolled out cind the primary hood was raised up and left that
way from one charging of a ladle of material to the next. Visual
emissions decreased significantly in the periods between each
charge. On other occasions the converter was rolled out for the
charging mode and the primary hood was raised, but the hood was
lowered again over the converter mouth after each ladle of mate-
rial was charged. Particle size sampling was conducted while the
primary hood was left up between ladles of material being charged
(high flow mode), as this was considered to be part of the charg-
ing mode. The reduced particulate emissions at these times
caused the overall sample to show a lower concentration.
The total particulate emission rate as determined from the
particle size distribution tests during the charging mode ranged
between 20.1 kg/h (44.6 Ib/h) and 35.4 kg/h (78.3 Ib/h). These
values show the same variations as the concentration results
because an average flue gas flow rate of 3571 dNm3 (126,924
dscfm) as measured during the specific mode tests was used for
all of the emission rate calculations.
During the skimming mode, the particle size runs showed a
total particulate concentration ranging between 40 mg/dNm3 (0.018
gr/dscf) and 77 mg/dNm3 (0.033 gr/dscf). The same operating
procedures were followed during the charging mode; sometimes the
primary hood was left up between pourings of ladles of slag or
112
-------
copper and sometimes it was lowered. This resulted in the same
dilution effects; in the measured concentrations during the par-
ticle size runs., The visible emissions decreased significantly
between ladles of slag or copper being poured. During both the
charging and skimming modes, the time between ladles of material
being charged or skimmed varied throughout the testing period,
depending on process operations.
The total particulate emission rate during the skimming mode
ranged between 8.6 kg/h (19.6 Ib/h) and 16.5 kg/h (35.9 Ib/h).
The average high-flow-mode exhaust gas flow rate of 3571 dNm3/m
(126,924 dscfm) was also used for all of the skimming emission
rate calculations.
During the blowing mode, the total particulate concentra-
tions measured by the particle sizing runs ranged between 13
mg/dNm3 (0.006 gr/dscf) and 107 mg/dNm3 (0.047 gr/dscf). This
large variation in particulate concentrations resulted from a
very high concentration during one run (Run No. PSB-2), during
which the primary hood draft malfunctioned and emissions normally
captured by the primary hood escaped and were collected by the
air curtain exhaust system. Visual observations of the process
during the blowing mode indicates that the visible emissions were
very low (not visible) while the primary hood was operating
correctly. This explains why the particulate concentrations
measured by the particle size runs were much lower during the
blowing mode thcin during the charging and skimming modes.
The total particulate emission rate during the blowing mode
ranged between 1.7 kg/h (3.9 Ib/h) and 13.8 kg/h (30.8 Ib/h).
The average low-flow-mode flue gas flow rate of 2149 dNm3/min
(76,359 dscfm) was used to calculate all of the particulate
emission rates for the blowing mode.
Figures 16 through 18 present particle size distribution
curves for each of the three converter modes tested. These
graphs show that the size of the particulate generated during
each of the three modes differ.
113
-------
-+»
-*«
PARTICLE SIZE, micrometers
figure 16. Individual particle size distributions for the charging mode sample runs.
-------
• '.*'• » ». i i •
U_ Ml .1. _—J--jJ. II II I
PARTICLE SIZE, micrometers
figure 17, Individual particle size distributions for the skiirening mode sample runs.
-------
• PSB-1
• PSB-2
PSB-3
© PSB-4
—*
PARTICLE SIZE, micrometers
Figure 18. Individual particle size distributions for the blowing mode sample runs.
-------
The particle size distribution curves, however, for each
individual sample run conducted during the charging and skimming
modes compare closely with each other for each specific mode
tested. This indicates that the size of the particulate emitted
during the charging and skimming modes was fairly consistent
during the testing period. Consequently, the individual particle
size distribution curves for the sample runs conducted during the
blowing mode do not compare closely with each other. The par-
ticle size results for Run PSB-2 show a much larger percentage of
particles less than 10 ym in size than the other three sample
runs. The large variation in particle sizes during Run PSB-2
resulted from an upset in the primary hood, which caused particu-
late normally captured by the primary hood to escape and be
collected in the air curtain exhaust system.
Figures 19 through 21 present average particle size distri-
bution curves for each of three converter operating modes tested
(charging, skimming, and blowing). The highest and lowest mea-
sured values are shown for each indicated cut point.
The average particle size curves for the charging and skim-
ming modes show that most of the particulate emissions captured
by the air curtain exhaust system are in the inhalable particu-
late (IP) range (<10 vim) , and that a large percentage of the
emissions are in the fine particulate (FP) range (<2.5 ym) .
The average particle size distribution curves for the blow-
ing mode shows that most of the particulate emissions captured by
the secondary exhaust system are greater than 10 ym and not in
the inhalable particulate range.
5.4.2 Inhalable Particulate Emission Data
Tables 40 and 41 present the IP and FP concentration and
emission rate results for each particle size run conducted.
Inhalable particulate (<10 ym) and fine particulate (<2.5 ym)
concentrations and emission rates are calculated for each specif-
ic operating mode tested by multiplying the total measured value
by the cumulative weight percent less than the stated value as
117
-------
00
99.9
£90.0
o
o
o
1/1
50.0
«c
UI
10.0
1.0
0.1
i i i i i i 111
1.0 10.0
PARTICLE SIZE, micrometers
i i i i i
i i i i i
100.0
Figure 19. Average particle size distribution for the charging mode.
-------
" •'
T T I T T I I I I T I I T T T |
90.0
o
o
50.0
10.0
1.0
0.1
I 1 1
I
I I I I I I
1.0 10.0
PARTICLE SIZE, micrometers
Figure 20. Average particle size distribution for the skimming mode.
100.0
-------
ro
O
99.9
2 90.0
o
o
12 50.0
10.0
1.0
0.1
I I I
I I
I I I I I I 11 1
III
I I I I I I 1
0.1 10.0
PARTICLE SIZE, micrometers
Figure 21. Average particle size distribution for the blowina mode.
100.0
-------
TABLE 40. SUMMARY OF INHALABLE PARTICULATE CONCENTRATIONS
DURING THE PARTICLE SIZE RUNS
ro
Dim Nn.
Date
(1983)
Cumulative less than Indicated size
T9ta1 h
mg/dNm'"! gr/dsef
10 urn > X
mg/dNm3 [ gr/dscf
5 vm > X
mq/dNm3 j yr/dscf
2.5
mg/dNm3
{m > X
gr/dscf
Charging mode0
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
1/18-19
1/19
1/20
1/20
1/22
165
94
98
132
157
0.072
0.041
0.043
0.058
0.068
129
72
82
110
110
0.056
0.032
0.036
0.048
0.048
102
59
67
57
79
0.045
0.026
0.029
0.025
0.034
94
54
64
42
71
0.041
0.023
0.027
0.019
0.031
Skinning modec
Blowing mode
'Milligrams per dry normal cubic meter.
Grains per dry standard cubic foot.
cThe 10-uin cutpoint is extrapolated from the particle size distribution curves.
PSSS-1
PSSS-2
PSSS-3
1/18-19
1/20
1/22
40
77
64
0.018
0.033
0.028
36
71
55
0.016
0.030
0.024
28
53
50
0.013
0.023
0.022
26
50
47
0.012
0.021
0.020
PSB-1
PSB-2
PSB-3
PSB-4
1/18-19
1/20
1/20
1/22
25
107
45
13
0.011
0.047
0.020
0.006
3
45
1
1
0.001
0.020
0.0005
0.0006
3
43
1
1
0.001
0.019
0.0005
0.0005
3
40
1
1
0.001
0.017
0.0004
0.0004
-------
TABLE 41. SUMMARY OF INHALABLE PARTICULATE EMISSION RATES
DURING THE PARTICLE SIZE RUNS3
Run No.
Date
(1983)
Cumulative less than indicated size
Total
kg/h° | lb/hc
10 pm > X
kg/h | Ib/h
5 vm
kg/h
> X
Ib/h
2.5 ym > X
kg/h | Ib/h
Charging mode
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
1/18-
1/19
1/19
1/20
1/20
1/22
35.4
20.1
21.0
28.3
33.6
78.3
44.6
46.8
63.1
74.0
27.6
15.4
17.6
23.6
23.6
60.9
34.8
39.2
52.2
52.2
21.9
12.6
14.4
12.2
16.9
49.0
28.3
31.5
27.2
37.0
20.1
11.6
13.7
8.6
15.2
44.6
25.0
29.4
18.5
33.7
Skimming mode
PSSS-1
PSSS-2
PSSS-3
1/18-
1/19
1/20
1/22
8.6
16.5
13.7
19.6
35.9
30.5
7.7
15.2
11.8
17.4
32.6
26.1
6.0
11.4
10.7
14.1
25.0
23.9
5.6
10.7
10.1
13.1
22.8
21.8
Blowing mode
PSB-1
PSB-2
PSB-3
PSB-4
1/18-
1/19
1/20
1/20
1/22
3.2
13.8
5.8
1.7
7.2
30.8
13.1
3.9
0.4
5.8
0.1
0.1
0.7
13.1
0.3
0.4
0.4
5.5
0.1
0.1
0.7
12.4
0.3
0.3
0.4
5.2
0.1
0.1
0.7
11.1
,0.3
0.3
Emission rate based on the average volumetric flow rates of 2149 dNm3/min
(76,359 dscfm) during low flow (blowing mode) and 3571 dNm3/min (126,924
dscfm) during the high flow (charging and skimming modes).
Kilograms per hour.
cPounds per hour.
The 10-ym cutpoint is extrapolated from the particle size distribution
curves.
122
-------
determined from the individual particle size distribution curves.
During the charging and skimming modes, the desired 10 ym diam-
eter was determined from extrapolated portions of the best-fit
curves generated for each individual sample run. The results of
the particle size distribution tests conducted during the charg-
ing and skimming modes indicate that most of the particulate
emissions collected by the air curtain exhaust system are in the
IP range.
During the charging mode, the amount of particulate in the
IP range varied between 70 and 84 percent of the total particu-
late emissions, whereas the amount of particulate in the FP range
varied between 32 and 65 percent. The results of the particle
size runs conducted during the charging mode show that the IP
concentration varied between 72 mg/dNm3 (0.032 gr/dscf) and 129
mg/dNm3 (0.056 gr/dscf), while the FP concentration varied be-
tween 42 mg/dNm3 (0.019 gr/dscf) and 94 mg/dNm3 (0.041 gr/dscf).
The IP emission rate during the charging mode ranged between 15.4
kg/h (34.8 Ib/h) and 27.6 kg/h (60.9 Ib/h), and the FP emission
rate averaged between 9.0 kg/h (20.7 Ib/h) and 20.1 kg/h (44.6
Ib/h).
During the skimming mode, the amount of particulate in the
IP range varied between 86 and 92 percent of the total particu-
late emissions, whereas the amount in the FP range varied between
65 and 73 percent. The results of the particle size runs con-
ducted during the skimming mode show that the IP concentration
varied between 36 mg/dNm3 (0.016 gr/dscf) and 71 mg/dNm3 (0.030
gr/dscf), and the FP concentration varied between 26 mg/dNm3
(0.012 gr/dscf) and 50 mg/dNm3 (0.021 gr/dscf). The IP emission
rate during the skimming mode ranged between 7.7 kg/h (17.4 Ib/h)
and 15.2 kg/h (32.6 Ib/h), and the FP emission rate ranged be-
tween 5.6 kg/h (13.1 Ib/h) and 10.7 kg/h (22.8 Ib/h).
During the blowing mode, the amount of particulate measured
in the IP and FP ranges varied between 2 and 12 percent, ex-
cluding the results of Run PSB-2, which was not considered to be
representative of normal operating conditions because it was
conducted durng an upset in the primary hood system.
123
-------
A reasonable explanation of why particle sizes measured
during the blowing mode were larger than those measured during
the charging and skimming modes is that the primary hood operates
only during the blowing mode and thus captures most of the
process emissions generated. The particulate matter collected by
the air curtain exhaust system during the blowing mode represents
fugitive emissions from inside the converter building. These
fugitive emissions, which were visually observed in the converter
building during the test series, are attributable to the general
operation of the plant (i.e., crane operation, housekeeping
practices, etc.). Uncontrolled fugitive emissions also emanated
from the operations of the other two converters during the test
series.
The results of the particle size runs conducted during the
charging mode (excluding Run PSB-2) show that the IP and FP
concentrations varied between 1 mg/dNm3 (0.0004 gr/dscf) and 3
mg/dNm3 (0.001 gr/dscf). The IP and FP emission rates during the
blowing mode varied between 0.1 kg/h (0.7 Ib/h) and 0.4 kg/h (0.7
Ib/h). Because of the small percentage of particles in the IP
range and the low IP emission rate, the particulate matter col-
lected by the air curtain hooding system during the blowing mode
should not be considered significant.
5.4.3 Elemental Analysis of the Particle Size Runs
Selected individual particle size runs were analyzed by
Atomic Absorption Spectrophotometry for six elements (arsenic,
selenium, cadmium, antimony, lead, and bismuth). The runs se-
lected for elemental analysis had measurable particulate concen-
trations throughout the particle size distribution. Each par-
ticle size sample run selected for elemental analysis was sub-
divided into three size ranges. The criterion for the selection
of the three size ranges was to choose the actual cut points
closest to the inhalable particulate (IP) cut point of 10 ym and
fine particulate (FP) cut point of 2.5 ym. This would ideally
yield size ranges of greater than 10 ym, between 10 ym and 2.5
ym, and less than 2.5 ym.
124
-------
The results of the elemental analysis is presented sepa-
rately for the three different operating modes tested (charging,
blowing, and skimming).
Charging Mode--
Tables 42 through 47 present the concentration and mass
emission rates for the elemental analysis performed on the five
particle size runs conducted during the charging mode. The data
are presented for each element on a total basis for the entire
particle size run and individually for each of the three size
ranges selected.
The total concentration of arsenic measured from the par-
ticle size runs; ranged between 4.0 mg/dNm3 (0.002 gr/dscf) and
28.0 mg/dNm3 (0.012 gr/dscf). This corresponds to a mass emis-
sion rate ranging between 0.9 kg/h (2.2 Ib/h) and 6.0 kg/h (13.1
Ib/h).
The total concentration of selenium measured from the par-
ticle size runs ranged from less than 0.05 mg/dNm3 (0.00002
gr/dscf) to 0.1 mg/dNm3 (0.00006 gr/dscf). This corresponds to a
mass emission rate of less than 0.05 kg/h (0.1 Ib/h) for all of
the sample runs.
The total concentration of cadmium measured from the par-
ticle size runs ranged between 0.05 mg/dNm3 (0.00002 gr/dscf) and
0.9 mg/dNm3 (0.0004 gr/dscf). For one sample run the mass emis-
sion rate was 0.2 kg/h (0.4 Ib/h); all of the other sample runs
had measurement rates of less than 0.05 kg/h (0.1 Ib/h).
The total concentration of antimony measured from the par-
ticle size runs ranged between 0.2 mg/dNm3 (0.0007 gr/dscf) and
1.8 mg/dNm3 (0.0008 gr/dscf). This corresponds to a mass emis-
sion rate ranging than 0.05 kg/h (0.1 Ib/h) to 0.4 kg/h (0.9
Ib/h).
The total concentration of lead measured from the particle
size runs ranged between 6.2 mg/dNm3 (0.003 gr/dscf) and 30.4
mg/dNm3 (0.013 gr/dscf). This corresponds to a mass emission
rate ranging between 1.5 kg/h (3.3 Ib/h) and 6.4 kg/h (14.1
Ib/h).
125
-------
TABLE 42. SUMMARY OF ARSENIC CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Concentration
Total
mg/dNm3t)
20.0
4.0
28.0
10.5
9.4
gr/dscfc
n nno
\j . \j\j j
0.002
0.012
0.005
0.004
x > 9.1 pm
mg/dNm3
1 C
•I. U
0.5
2.0
1.5
1.0
gr/dscf
r\ nnm
u. \HJ\I /
0.0002
0.0009
0.0006
0.0005
9.1 urn > * > 3.4 urn
mg/dNm3
1.4
0.4
1.9
1.5
1.4
gr/dscf
0.0006
0.0001
0.0008
0.0007
0. 0006
x < 3.4 pro
mg/dNm3
17.0
3.1
24.0
7.5
7.0
gr/dscf
n nn-t
u. uu/
0.001
0.010
0.003
0.003
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/hd
4.3
0.9
6.0
2.2
2.0
lb/he
9.8
2.2
13.1
5.4
4.4
x > 9.1 urn
kg/h
0.3
0.1
0.4
0.4
2.5
Ib/h
0.8
0.2
1.0
0.7
0.5
9.1 urn > x > 3.4 urn
kg/h
0.3
0.05
0.4
0.3
0.3
Ib/h
0.7
0.1
0.9
0.8
0.7
x < 3.4 urn
kg/h
3.6
0.5
5.1
1.6
1.5
Ib/h
7.6
1.1
10.9
3.3
3.3
ro
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
Pounds per hour.
-------
TABLE 43. SUMMARY OF SELENIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE3
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Concentration
Total
mg/dNm3b
0.1
0.04
0.05
0.03
0.05
gr/dscfc
0.00006
0.00002
0.00002
0.00001
0.00002
x > 9.1 |jm
mg/dNm3
0.02
0.006
0.005
0.008
0.01
gr/dscf
0.000008
0.000003
0.000002
0.000004
0.000006
9.1 pm > x > 3.4 pm
mg/dNm3
0.01
0.003
0.003
0.008
0.007
gr/dscf
0.000005
0.000001
0.000001
0.000003
0.000003
x < 3.4 pm
mg/dNm3
0.2
0.03
0.05
0.02
0.03
gr/dscf
0.00005
0.00001
0.00002
0.000007
0.00001
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/hd
0.03
0.01
0.01
0.005
0.01
lb/he
0.07
0.002
0.002
0.01
0.02
x > 9.1 urn
kg/h
0.004
0.001
0.001
0.002
0.003
Ib/h
0.009
0.003
0.002
0.004
0.007
9.1 urn > x > 3.4 pm
kg/h
0.002
0.005
0.001
0.001
0.001
Ib/h
0.055
0.001
0.01
0.003
0.003
x < 3.4 pm
kg/h
0.02
0.005
0.01
0.003
0.005
Ib/h
0.05
0.01
0.02
0.008
0.01
ro
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76, 359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 44. SUMMARY OF CADMIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Concentration
Total
mg/dNm3b
0.9
0.2
0.2
0.07
0.05
gr/dscfc
0.0004
0.00009
0.00008
0.00003
0.00002
x > 9.1 |jm
mg/dNm3
0.5
0.01
0.01
0.02
0.02
gr/dscf
0.00002
0.000005
0.000005
0.000007
0.000008
9.1 urn > x > 3.4 Mm
mg/dNm3
0.06
0.01
0.01
0.02
0.01
gr/dscf
0.00002
0.000005
0.000005
0.000007
0.000004
x < 3.4 |jm
mg/dNm3
0.8
0.2
0.2
0.04
0.02
gr/dscf
0.0004
0.00007
0.00007
0.00002
0.00001
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/h°
0.2
0.04
0.04
0.01
0.01
lb/he
0.4
0.1
0.09
0.03
0.02
x > 9.1 urn
kg/h
0.01
0.002
0.002
0.003
0.004
Ib/h
0.02
0.005
0.005
0.008
0.009
9.1 pm > x > 3.4 \jan
kg/h
0.01
0.002
0.002
0.003
0.002
Ib/h
0.02
0.005
0.005
0.008
0.004
x < 3.4 urn
kg/h
0.2
0.03
0.03
0.01
0.005
Ib/h
0.4
0.08
0.08
0.02
0.01
ro
oo.
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow (blowing
mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
Grains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 45. SUMMARY OF ANTIMONY CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE
a
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/20
Concentration
Total
mg/dNm3b
1.6
0.2
1.8
0.7
0.6
gr/dscfc
0.0007
0.0001
0.0008
0.0003
0.0002
x > 9.1 urn
mg/dNm3
0.2
0.06
0.1
0.1
0.1
gr/dscf
0.00009
0.00003
0.00005
0.00006
0.00006
9.1 \Jirn > x > 3.4 |jm
mg/dNm3
0.2
0.03
0.09
0.2
0.1
gr/dscf
0.00008
0.00001
0.00004
0.00009
- 0.00004
x < 3.4 urn
mg/dNm3
1.2
0.1
1.6
0.4
0.3
gr/dscf
0.0005
0.00005
0.0007
0.0002
0.0001
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/ha
0.3
0.05
0.4
0.1
0.1
lb/he
0.8
0.1
0.9
0.3
0.2
x > 9.1 urn
kg/h
0.4
0.01
0.02
0.03
0.03
Ib/h
0.1
0.03
0.05
0.07
0.07
9.1 urn > x > 3.4 urn
kg/h
0.04
0.005
0.02
0.04
0.02
Ib/h
0.09
0.01
0.04
0.1
0.04
x < 3.4 pm
kg/h
0.2
0.02
0.3
0.1
0.05
Ib/h
0.5
0.05
0.8
0.2
0.1
vo
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 46. SUMMARY OF LEAD CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Concentration
Total
mg/dNm3D
30.4
6.2
8.6
8.0
9.9
gr/dscfc
0.013
0.0026
0.0036
0.0035
0.0043
x > 9.1 urn
mg/dNm3
2.7
0.7
0.8
1.6
1.7
gr/dscf
0.001
0.0003
0.0003
0.0007
0.0007
9.1 urn > x > 3.4 urn
mg/dNm3
2.7
0.6
0.6
1.8
1.3
gr/dscf
0.001
0.0003
0.0003
0.0008
0.0006
x < 3.4 urn
mg/dNm3
24.9
4.9
7.3
4.6
6.9
gr/dscf
0.011
0.002
0.003
0.002
0.003
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/hd
6.4
1.5
2.0
1.5
2.0
lb/he
14.1
3.3
4.4
3.3
4.4
x > 9.1 pm
kg/h
0.5
0.1
0.1
0.3
0.3
Ib/h
1.1
0.3
0.3
0.8
0.8
9.1 urn > x > 3.4 urn
kg/h
0.5
0.1
0.1
0.4
0.3
Ib/h
1.1
0.3
0.3
0.9
0.7
x < 3.4 pm
kg/h
5.4
1.0
1.5
1.0
1.5
Ib/h
12.0
2.2
3.3
2.2
3.3
CO
o
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 47 SUMMARY OF BISMUTH CONCENTRATION AND MASS EMISSION RATE
'FOR THE PARTICLE SIZE RUNS FOR THE CHARGING MODE
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Concentration
Total
mg/dNm3b
0.6
0.3
0.2
0.2
0.1
gr/dscfc
0.0024
0.00009
0.0001
0.00008
0.00006
x > 9.1 urn
mg/dNm3
0.05
0.02
0.02
0.04
0.04
gr/dscf
0.00002
0.00001
0.00001
0.00002
0.00002
9.1 urn > x > 3.4 |jm
mg/dNm3
0.05
0.02
0.02
0.05
0.03
gr/dscf
0.00002
0.00001
0.000008
0.00002
0.00001
x < 3.4 urn
mg/dNm3
0.5
0.1
0.2
0.09
0.07
gr/dscf
0.002
0.00007
0.00008
0.00004
0.00003
Run No.
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
Date
(1983)
1/18-19
1/19
1/20
1/20
1/22
Emission rate
Total
kg/hd
0.1
0.04
0.05
0.04
0.03
lb/he
0.3
0.1
0.1
0.09
0.07
x > 9.1 urn
kg/h
0.01
0.005
0.005
0.01
0.01
Ib/h
0.02
0.01
0.01
0.02
0.02
9.1 urn > x > 3.4 urn
kg/h
0.01
0.005
0.004
0.01
0.005
Ib/h
0.02
0.01
0.009
0.02
0.01
x < 3.4 urn
kg/h
0.1
0.03
0.04
0.02
0.01
Ib/h
0.2
0.08
0.09
0.04
0.03
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
The total concentration of bismuth measured from the par-
ticle size runs ranged between 0.1 mg/dNm3 (0.00006 gr/dscf) and
0.6 mg/dNm3 (0.00024 gr/dscf). For two sample runs the mass
emission rates were 0.1 and 0.05 kg/h (0.3 and 0.1 Ib/h); all of
the other sample runs had mass emission rates of less than 0.05
kg/h (0.1 Ib/h).
The data presented in Tables 42 through 47 indicate that
arsenic and lead comprised the majority of the measurable ele-
mental concentration for the particle size runs conducted during
the charging mode. The concentrations of selenium, cadmium,
antimony, and bismuth measured in the particle size runs conduct-
ed during the charging mode are also presented in Tables 42
through 47.
The variations in the concentrations of arsenic and lead
during the testing period can be attributed to the fact that the
material being charged to the converter during the particle size
runs was not consistent. A comparison of the material charged to
the measured elemental concentration indicates that the greater
the number of matte charges sampled by a particle size run, the
higher the measured concentrations of arsenic and lead.
Figures 22 and 23 present a breakdown of the elemental
concentrations for the three size ranges selected for analysis
from the size runs conducted during the charging mode. These
figures indicate that for all six elements measured, the majority
of elemental concentrations are in the fine particulate range.
Skimming Mode—
Tables 48 through 53 present the concentration and mass
emission rates for the elemental analysis performed on the three
particle size runs conducted during the skimming mode. The data
are presented in the same format as that used for the charging
mode results.
The total concentration of arsenic measured from the par-
ticle size runs ranged between 2.0 mg/dNm3 (0.0009 gr/dscf) and
5.6 mg/dNm3 (0..002 gr/dscf). This corresponds to a mass emission
rate ranging between 0.4 kg/h (1.0 Ib/h) and 1.0 kg/h (2.2 Ib/h).
132
-------
too-
n
JO-
SO--
40 —
10...
9...
g...
6...
$....
1...
1 —
.9
.8
.7'
.5-
.4
.3.
.1 -
.09-
.08-
.07-
.K
.04
.03
.01-1-
sm MMI
9.1 to 3.4ut,
< 1.4iP
n
_j_
13
te
14
St
48
CO
SI
St
u
13
(1
33
At
K
St
46
Co
SI K
Sb Pb
>3
ti
13 34
As S»
51
m w.
MM M. FSNC-?
«TOH1C WWER/U.EICHT
CO
MM NO. PSMC-3
B: 63
Pt Bi
Figure 22. Comparison of elemental concentrations for the charging mode
(Run Nos. PSMC-1, -2, -3.)
133
-------
*o--
so —•
10 —
JO— •
JO---.
10...
8...
7 • —
6-..
5....
4 —
SIZE IMGE
9.1 to 3.4JT.
r
E
i
E
l —
.9"
.6-
.7-
.«•
.5-
.08-
.07-
.06'
.06-
.04-
.03-
.02-1-
33
As
34
s.
46
co
SI
st
82
PC
83
B1
33
At
34
Si
4E SI
CD Sb
B3
Si
IIM NO. KMC-4
UN NO. PSHC-S
ATONIC
Figure 23. Comparison of elemental concentrations for the charging mode
(Run Nos. PSMC-4, -5).
134
-------
TABLE 48. SUMMARY OF ARSENIC CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE
a
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Concentration
Total
mg/dNm3b
2.0
5.6
4.8
gr/dscfc
0.0009
0.0024
0.0022
x ^ 8.6 pro
mg/dNm3
0.1
0.4
0.2
gr/dscf
0.00004
0.0002
0.0001
8.6 |jm > x > 3.1 pm
mg/dNm3
0.1
0.5
0.3 .
gr/dscf
0.00005
0.0002
0.0001
x < 3.1 pm
mg/dNm3
1.8
4.7
4.3
gr/dscf
0.0008
0.002
0.002
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Emission rate
Total
kg/hd
0.4
1.0
1.0
lb/he
1.0
2.2
2.2
x ^ 8.6 pm
kg/h
0.02
0.1
0.05
Ib/h
0.04
0.2
0.1
8.6 pm > x > 3.1 pm
kg/h
0.02
0.1..
0.05
Ib/h
0.05
0.2
0.1
x < 3.1 pm
kg/h
0.4
1.0
1.0
Ib/h
0.9
2.2
2.2
to
en
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
GGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 49. SUMMARY OF SELENIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE3
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Concentration
Total
mg/dNm3D
2.1
0.04
0.05
gr/dscfc
0.0009
0.00002
0.00002
x > 8.6 um
mg/dNm3
0.005
0.005
0.003
gr/dscf
0.000002
0.000002
0.000001
8.6 um > x > 3. 1 um
mg/dNm3
0.01
0.006
0.002
gr/dscf
0.000006
0.000002
0. 0000009
x < 3.1 um
mg/dNm3
2.10
0.03
0.05
gr/dscf
0.0009
0.00001
0.00002
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Emission rate
Total
kg/hd
0.4
0.01
0.01
lb/he
1.0
0.02
0.02
x >_ 8.6 urn
kg/h
0.001
0.001
0.0005
Ib/h
0.002
0.002
0.001
8.6 um > x > 3.1 um
kg/h
0.003
0.001
0.0004
Ib/h
0.007
0.002
0.001
x < 3.1 Mm
kg/h
0.4
0.005
0.01
Ib/h
1.0
0.01
0.02
(A)
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNm3/min (126, 924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 50. SUMMARY OF CADMIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE3
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Concentration
Total
mg/dNm3b
0.02
0.1
0.07
gr/dscfc
0.000008
0.000067
0.000033
x >_ 8.6 urn
mg/dNm3
0.002
0.007
0.003
gr/dscf
0.0000008
0.000003
0.000001
8.6 urn > x > 3.1 urn
mg/dNm3
0.002
0.01
0.004
gr/dscf
U.UUUUUl
0.000004
0.000002
x < 3.1 pm
mg/dNm3
0.01
0.1
0.07
gr/dscf
0. 000006
0.00006
0.00003
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Emission rate
Total
kg/hd
0.004
0.03
0.01
lb/he
0.009
0.07
0.03
x ^ 8.6 urn
kg/h
0.0004
0.001
0.0005
Ib/h
0.0009
0.003
0.001
8. 6 |jm > x > 3. 1 urn
kg/h
0.0005
0.002
0.001
Ib/h
0.001
0.004
0.002
x < 3.1 urn
kg/h
0.003
0.03
0.01
Ib/h
0.007
0.07
0.03
OJ
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNnrVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
Grains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 51. SUMMARY OF ANTIMONY CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE3
Run No.
PSSS-i
PSSS-2
PSSS-3
Date
(1983)
1/18- 19
1/20
1/22
Concentration
Total
mg/dNm3b
0.18
0.80
0.57
gr/dscfc
0.00008
0.00036
0.00022
x > 8.6 um
mg/dNm3
0.03
0.05
0.03
gr/dscf
0.00001
0.00002
0.00001
8.6 pm > x > 3.1 urn
mg/dNm3
0.03
0.08
0.03
gr/dscf
u. uuuui
0.00004
0.00001
x < 3.1 urn
mg/dNm3
0. 12
0.67
0.52
gr/dscf
U. UUUUD
0.0003
0.0002
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Emission rate
Total
kg/hd
0.04
0.1
0.1
lb/he
0.09
0.3
0.3
x > 8.6 pm
kg/h
0.005
0.01
0.005
Ib/h
0.01
0.02
0.01
8.6 urn > x > 3.1 urn
kg/h
0.005
0.02
0.005
Ib/h
0.01
0.04
0.01
x < 3.1 urn
kg/h
0.02
0.1
0.1
Ib/h
0.05
0.03
0.2
CA>
00
aEmission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
(blowing mode) and 3571 dNm3/min (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
Grains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 52. SUMMARY OF LEAD CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE3
Date
Run No. (1983)
PSSS-1 1/18-19
PSSS-2 1/20
PSSS-3 1/22
Concentration
Total
mg/dNm3t)
4.9
18.4
16.6
gr/dscfc
0.002
0.008
0.0075
x ^ 8.6 urn
mg/dNm3
0.2
0.8
0.4
gr/dscf
0.0001
0.0003
0.0002
8. 6 pm > x > 3. 1 urn
mg/dNm3
0.3
1.1
0.6
gr/dscf
0.0001
0.0005
0.0003
x < 3.1 urn
mg/dNm3
4.3
16.5
15.6
gr/dscf
0.002
0.007
0.007
Date
Run No. (1983)
PSSS-1 1/18-19
PSSS-2 1/20
PSSS-3 1.22
Emission rate
Total
kg/hd
1.0
3.9
3.5
lb/he
2.2
8.7
7.6
x > 8.6 urn
kg/h
0.05
0.1
0.1
Ib/h
0.1
0.3
0.2
8. 6 pm > x > 3. 1 |jm
kg/h
0.05
0.2
0.1
Ib/h
0.1
0.5
0.3
x < 3.1 urn
kg/h
1.0
3.5
3.5
Ib/h
2.2
7.6
7.6
CO
aEmission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
GGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 53. SUMMARY OF BISMUTH CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE SKIMMING MODE
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Concentration
Total
mg/dNm3t)
0.17
0.54
0.42
gr/dscfc
0.00007
0.00022
0.0002
x > 8.6 pm
mg/dNm3
0.005
0.02
0.01
gr/dscf
0.000002
0.000009
0.000005
8.6 urn > x > 3.1 pm
mg/dNm3
0.007
0.03
0.01
gr/dscf
0.000003
0.00001
0.000005
x < 3.1 pm
mg/dNm3
0.16
0.49
0.40
gr/dscf
0.00007
0.0002
0.0002
Run No.
PSSS-1
PSSS-2
PSSS-3
Date
(1983)
1/18-19
1/20
1/22
Emission rate
Total
kg/hd
0.03
0.1
0.1
lb/he
0.08
0.2
0.2
x > 8.6 urn
kg/h
0.001
0.004
0.002
Ib/h
0.002
0.01
0.005
8.6 pm > x > 3.1 pm
kg/h
0.001
0.005
0.002
Ib/h
0.003
0.01
0.005
x < 3.1 pm
kg/h
0.03
0.1
0.1
Ib/h
0.08
0.2
0.2
aEmission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
GGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
The total concentrations of selenium measured from the
particle size runs ranged from less than 0.05 mg/dNm3 (0.00002
gr/dscf) to 2.1 mg/dNm3 (0.0009 gr/dscf). This corresponds to a
mass emission rate ranging from less than 0.05 kg/h (0.1 Ib/h) to
0.4 kg/h (1.0 Ib/h). The reason for the large concentration of
selenium measured in Run No. PSSS-1 could not be determined from
the converter's observation log and is therefore considered
suspect.
The total concentration of cadmium measured from the par-
ticle size runs ranged from less than 0.05 mg/dNm3 (0.00002
gr/dscf) to 0.15 mg/dNm3 (0.00006 gr/dscf). This corresponds to
a mass emission rate of less than 0.05 kg/h (0.1 Ib/h).
The total concentration of antimony measured from the par-
ticle size runs ranged between 0.18 and 0.57 mg/dNm3 (0.00008 and
0.003 gr/dscf). This corresponds to a mass emission rate ranging
from less than 0.05 kg/h to 0.1 kg/h (0.00002 to 0.3 Ib/h).
The total concentration of lead measured from the particle
size runs ranged between 4.9 mg/dNm3 (0.002 gr/dscf) and 18.4
mg/dNm3 (0.008 gr/dscf). This corresponds to a mass emission
rate ranging between 1.0 kg/h (2.2 Ib/h) and 3.9 kg/h (8.7 Ib/h).
The total concentration of bismuth measured from the par-
ticulate size runs ranged between 0.17 and 0.54 mg/dNm3 (0.00007
and 0.0002 gr/dscf). This corresponds to a mass emission rate
ranging from less than 0.05 kg/h to 0.1 kg/h (0.1 to 0.2 Ib/h).
Figure 24 presents a breakdown of the elemental concentra-
tion for the three size ranges selected for analysis from the
particle size runs conducted during the skimming mode. This
figure indicates that for all six elements measured the majority
of the elemental concentrations are in the fine particulate
range.
The concentrations for five of the six elements appear to
follow a pattern; Run No. PSSS-1 had the lowest concentration,
Run No. PSSS-2 the highest concentration, and Run No. PSSS-3 a
concentration in between the two previous runs. The total par-
ticulate concentrations for the particle size runs also follow
this same pattern.
141
-------
IB-
ID
ro-
w-
so-
40
io-
n-
ic...
8--
t...
<,....
4 —
3 —
1
r ?-
g
i
8 'i
.7'
.S-
.4 •
.3.
.03- -
.01-
S1ZC WMf
6.6 to 3.1ur.
D
33
As
M
W
48
U
SI
it,
K
83
11
33
At
M
Si
46
CO
SI K
Sb »b
83
13
•e
CO
5)
Sb
63
61
Ml M. PSSS-1
10. nss-z
OMCR/tUICKT
(UN »0. »SSS-3
Figure 24. Comparison of elemental concentrations for the skimming mode.
142
-------
Blowing Mode—
Tables 54 through 59 present the concentration and mass
emission rates for the elemental analysis performed on two of the
particle size runs conducted during the blowing mode. These data
are presented in the same format as that used for the charging
mode results.
The total concentrations of arsenic, antimony, lead, and
bismuth measured for the two particle size runs varied signifi-
cantly. The reason for this large variation is that during Run
No. PSB-2, the primary hood malfunctioned and emissions normally
captured by the primary hood escaped and were collected by the
air curtain system. This indicates that the elemental concentra-
tion measured by Run No. PSB-2 is biased high.
The elemental analyses for selenium and cadmium for the two
particle size runs show that the concentrations of these elements
are less than the detectable level.
Figure 25 presents a breakdown of the elemental concentra-
tion for the three size ranges selected for analysis from the two
chosen particle size runs conducted during the blowing mode.
This figure indicates that a higher percentage of the elemental
concentration is comprised of larger particulate compared to the
charging and skimming modes.
Table 60 presents a comparison of the total particulate
concentration with the elemental concentrations measured for the
particle size runs.
For the charging mode the particulate concentration deter-
mined by adding the individual concentrations from the elemental
analysis comprised between 12.2 and 39.5 percent of the total
particulate concentration measured by the particle size runs.
The concentration of arsenic in these runs ranged between 4.6 and
28.2 percent of the total particle size concentration, while the
concentration of lead ranged between 5.6 and 18.2 percent. The
concentration of the remaining four elements (selenium, cadmium,
antimony, and bismuth) comprised between 0.6 and 2.3 percent of
the total particle size concentration.
143
-------
TABLE 54. SUMMARY OF ARSENIC CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE3
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm3b
0.37
23.7
gr/dscfc
0.0002
0.01
x > 9.9 pro
mg/dNm3
0.13
1.5
gr/dscf
0.00006
0.0007
9.9 urn > x > 2.1 jjm
mg/dNm3
0.03
2.2
gr/dscf
0.00001
0.0009
x < 2.1 urn
mg/dNm3
0.21
20.1
gr/dscf
0.00009
0.009
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate
Total
kg/ha
0.01
3.0
lb/he
0.1
6.5
x > 9.9 pm
kg/h
0.02
0.2
Ib/h
0.04
0.5
9.9 [im > x > 2.1 M"i
kg/h
0.003
0.3
Ib/h
0.07
0.6
x < 2.1 urn
kg/h
0.03
2.7
Ib/h
0.06
5.9
Emission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 55. SUMMARY OF SELENIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE
a
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm3b
0.007
0.02
gr/dscfc
0.000003
0.000008
x > 9.9 M"i
mg/dNm3
0.005
0.01
gr/dscf
0.0000002
0.000006
9.9 pm > x > 2.1 urn
mg/dNm3
<0.0004
<0.002
gr/dscf
<0. 0000002
<0. 0000007
x < 2.1 pin
mg/dNm3
0.001
0.003
gr/dscf
0.0000006
0.000001
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate
Total
kg/hd
0.001
0.002
1b/he
0.002
0.005
x > 9.9 urn
kg/h
0.0006
0.002
Ib/h
0.001
0.004
9.9 |jm > x > 2.1 pm
kg/h
<0. 00006
<0.0002
Ib/h
<0.0001
<0.0005
x < 2.1 pm
kg/h
0.0002
0.0003
Ib/h
0.0004
0.0007
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard' cubic foot.
Kilograms per hour.
p
Pounds per hour.
-------
TABLE 56. SUMMARY OF CADMIUM CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE3
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm3b
0.006
0.03
gr/dscfc
0.000003
0.00001
x :> 9.9 pin
mg/dNm3
0.002
0.02
gr/dscf
0.000001
0.000007
9.9 urn > x > 2.1 urn
mg/dNm3
<0.0004
0.01
gr/dscf
<0. 0000002
0.000004
x < 2.1 \im
mg/dNm3
0.004
0.008
gr/dscf
0.000002
0.000003
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate6
Total
kg/hd
0.001
0.003
lb/he
0.002
0.007
x > 9.9 |jm
kg/h
0.0003
0.002
Ib/h
0.0007
0.005
9.9 urn > x > 2.1 urn
kg/h
0.00006
0.001
Ib/h
0.0001
0.003
x < 2.1 urn
kg/h
0.0006
0.0009
Ib/h
0.001
0.002
aEmission rate based on average volumetric flow rates of 2149 dNnrVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 57. SUMMARY OF ANTIMONY CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE3
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm3b
0.09
1.75
gr/dscfc
0.00004
0.0008
x > 9.9 pro
mg/dNm3
0.07
0.27
gr/dscf
0.00003
0.0001
9.9 pm > x > 2.1 pm
mg/dNm3
0.001
0.10
gr/dscf
0.0000006
0.00004
x < 2.1 urn
mg/dNm3
0.01
1.38
gr/dscf
0.000006
0.0006
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate6
Total
kg/hd
0.01
0.24
lb/he
0.03
0.5
x > 9.9 urn
kg/h
0.009
0.03
Ib/h
0.02
0.07
9.9 |jm > x > 2.1 pm
kg/h
0.0002
0.01
Ib/h
0.0004
0.03
x < 2.1 pm
kg/h
0.002
0.2
Ib/h
0.004
0.4
aEmission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
(blowing mode) and 3571 dNm3/min (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
Pounds per hour.
-------
TABLE 58. SUMMARY OF LEAD CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE3
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm30
0.71
8.97
gr/dscfc
0.0003
0.004
x > 9.9 (jm
mg/dNm3
0.45
2.0
gr/dscf
0.0002
0.0009
9.9 pm > x > 2.1 \jm
mg/dNm3
0.03
0.70
gr/dscf
0.00001
0.0003
x < 2.1 |jm
mg/dNm3
0.23
6.3
gr/dscf
0.0001
0.003
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate
Total
kg/ha
0.09
1.2
lb/he
0.2
2.6
x > 9.9 pm
kg/h
0.06
0.3
Ib/h
0.1
0.6
9.9 pm > x > 2.1 pm
kg/h
0.003
0.09
Ib/h
0.007
0.2
x < 2.1 pm
kg/h
0.03
0.9
Ib/h
0.07
2.0
00
aEmission rate based on average volumetric flow rates of 2149 dNmVmin (76,359 dscfm) during low flow
(blowing mode) and 3571 dNm3/min (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
cGrains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
TABLE 59. SUMMARY OF BISMUTH CONCENTRATION AND MASS EMISSION RATE
FOR THE PARTICLE SIZE RUNS FOR THE BLOWING MODE
a
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Concentration
Total
mg/dNm3b
0.01
0.30
gr/dscfc
0.000005
0.0001
x > 9.9 pro
mg/dNm3
0.007
0.05
gr/dscf
0.000003
0.000002
9.9 urn > x > 2.1 M">
mg/dNm3
0.0008
0.03
gr/dscf
0.0000004
0.00001
x < 2.1 urn
mg/dNm3
0.005
0.2
gr/dscf
0.000002
0.0001
Run No.
PSB-1
PSB-2
Date
(1983)
1/18
1/20
Emission rate6
Total
kg/hd
0.001
0.03
lb/he
0.003
0.07
x 2l 9.9 urn
kg/h
0.0009
0.006
Ib/h
0.002
0.01
9.9 urn > x > 2.1 urn
kg/h
0.0001
0.003
Ib/h
0.0003
0.007
x < 2.1 urn
kg/h
0.0006
0.03
Ib/h
0.001
0.07
-p*
10
aEmission rate based on average volumetric flow rates of 2149 dNm3/min (76,359 dscfm) during low flow
(blowing mode) and 3571 dNmVmin (126,924 dscfm) during high flow (charging and skimming modes).
Milligrams per dry normal cubic meter.
Grains per dry standard cubic foot.
Kilograms per hour.
ePounds per hour.
-------
100--
•C-"
70-
•o —
«o -
K> —
»•-
10...
6-.
7 —
4...
5 —
4 —
3 —
2 —
i
.9
.6
.7-
.6
.S-
.06
.07
.06
.K
.CM
.03
.02- -
.01
SIZt MNCt
> 9.»un<
9.9 to Mui-
n
J3 14 «e SI 12 63
As S* CO St »b K
•M n. PS*-)
33 X 46 51 8? B3
»J S« Co Sb fb B(
NO. nt-i
33 M »( SI
»s &c CO St
MM HO.
•TOMIC
Figure 25. Comparison of elemental concentrations for the blowing mode.
150
-------
TABLE 60. COMPARISON OF THE TOTAL PARTICULATE CONCENTRATION MEASURED BY THE
PARTICLES SIZE RUNS TO THE ELEMENTAL CONCENTRATIONS
Run No.
Total particulate
concentration,
mg/dNm3 (gr/dscf)
Percent of the total
particulate concentration
determined from the total
elemental concentration
Percent of the total particulate
concentration for each element measured
As
Se
Cd
Sb
Pb
Bi
Charging mode
PSMC-1
PSMC-2
PSMC-3
PSMC-4
PSMC-5
165 (0.072)
94 (0.041)
98 (0.043)
132 (0.058)
157 (0.068)
32.5
12.2
39.5
14.7
12.6
12.3
4.6
28.2
8.3
6.0
0.07
0.04
0.05
0.02
0.03
0.5
0.2
0.2
0.05
0.03
1.0
0.2
1.8
0.5
0.4
18.2
7.0
9.0
5.6
6.1
0.4
0.2
0.2
0.2
0.08
Blowing mode
PSB-1
PSB-2
25 (0.011)
107 (0.047)
4.8
32.7
1.5
22.1
0.02
0.02
0.02
0.02
0.4
1.6
2.8
8.4
0.04
0.3
Skimming mode
PSSS-1
PSSS-2
PSSS-3
40 (0.018)
77 (0.033)
64 (0.028)
23.2
32.8
26.2
5.3
6.7
7.3
5.3
0.06
0.08
0.04
0.2
0.1
0.5
1.0
0.9
11.7
24.1
25.5
0.4
0.7
0.7
-------
For the blowing mode the particulate concentration deter-
mined by adding the individual concentrations from the elemental
analysis for the two sample runs comprised 4.7 and 32.4 percent
of the total particulate concentration measured by the particle
size runs.
The concentrations of arsenic in these two runs were 1.5 and
22.1 percent, while the concentrations of antimony were 0.4 and
1.6 percent and lead 2.8 and 8.4 percent of the total particle
size concentration. The remaining three elements (selenium,
cadmium, and bismuth) comprised less than 1 percent of the total
particle size concentration.
For the skimming mode the particulate concentration deter-
mined by adding the individual concentrations from the elemental
analysis comprised between 23.2 and 32.7 percent of the total
particulate concentration measured by the particle size runs.
The concentration of arsenic in these runs ranged between 5.3 and
7.3 percent of the total particle size concentration. The con-
centration of selenium was between 0.0 and 5.3 percent and lead
between 11.7 and 25.5 percent of the total particle size concen-
tration. The remaining three elements (cadmium, antimony, and
bismuth) comprised between 0.9 and 1.9 percent of the total
particle size concentration.
5.5 TRACE METALS—ANTIMONY, BISMUTH, CADMIUM, LEAD, AND SELENIUM
Table 61 summarizes the concentrations of antimony (Sb),
bismuth (Bi), cadmium (Cd), lead (Pb), and selenium (Se). Ali-
quots of samples from tests PATC and PASM were analyzed by atomic
absorption techniques at the completion of the particulate and
arsenic analysis. The reported concentrations represent the
combined filterable and gaseous fractions of each metal. Mass
emission rates were calculated by use of the volumetric flows
presented in Sections 5.2, Table 33. In the total cycle tests,
the concentrations of antimony, bismuth, cadmium, and selenium
were less than 1.0 mg/dNm3 (0.0004 gr/dscf), and the lead concen-
trations ranged from 3.44 mg/dNm3 (0.0015 gr/dscf) to 8.79 mg/dNm3
(0.004 gr/dscf).
152
-------
TABLE 61. SUMMARY OF TRACE METAL EMISSION RESULTS
Cycle
Test
No.
1
1
2
2
3
3
Run
No.
PATC-1
PASM-1
PATC-2
PASM-2
PATC-3
PASM-3C
Date
(1983)
1/18-19
1/18-19
1/20
1/20
1/22
1/22
Total concentration3, mg/dNm3 (gr/dscf)
Sb
0.23
(0.0001)
0.51
(0.0002)
0.81
(0.0004)
1.44
(0.0006)
0.15
(0.00007)
0.75
(0.0003)
Bi
0.12
(0.00005)
0.17
(0.00007)
0.29
(0.0001)
0.38
(0.0002)
0.10
(0.00004)
1.09
(0.0005)
Cd
0.08
(0.00004)
0.19
(0.00008)
0.09
(0.00004)
0.18
(0.00008)
0.02
(0.00001)
0.15
(0.0004)
Pb
4.79
(0.002)
9.49
(0.004)
8.79
(0.004)
18.9
(0.008)
3.44
(0.0015)
21.7
(0.009)
Se
0.38
(0.0002)
1.0
(0.0004)
0.05
(0.00002)
0.07
(0.00003)
0.23
(0.0001)
0.10
(0.00004)
Mass emission rate
Sb
0.04
(0.08)
0.10
(0.22)
0.15
(0.32)
0.29
(0.65)
0.03
(0.06)
0.15
(0.33)
Bi
0.02
(0.04)
0.04
(0.08)
0.04
(0.08)
0.10
(0.22)
0.01
(0.03)
0.24
(0.54)
Cd
0.01
(0.03)
0.04
(0.09)
0.01
(0.03)
0.04
(0.09)
0.004
(0.008)
0.02
(0.04)
kg/h (Ib/h)
Pb
0.73
(1.6)
2.0
(4.4)
1.45
(3.2)
3.95
(8.7)
0.54
(1.2)
4.4
(9.8)
Se
0.07
(0.16)
0.20
(0.44)
0.01
(0.02)
0.01
(0.03)
0.04
(0.08)
0.02
(0.04)
en
CO
aTotal concentration (filterable and gaseous fractions) of antimony (Sb), bismuth (Bi), cadmium (Cd), lead (Pb),
and selenium (Se).
Calculations of mass emission rate were based on measured concentration and volumetric flows reported in
Table 5-9.
°Slag skimming only.
-------
During specific mode Tests PASM-1 and 2, antimony concentra-
tions ranged from 0.51 mg/dNm3 (0.0002 gr/dscf) to 1.44 mg/dNm3
(0.0006 gr/dscf). Bismuth and cadmium concentrations were less
than 0.5 mg/dNm3 (0.0002 gr/dscf) and selenium concentrations
ranged from 0.07 mg/dNm3 (0.00003 gr/dscf) during Test PASM-2 to
1.0 mg/dNM3 (0.004 gr/dscf) during Test PASM-1. Lead concentra-
tions during these tests were 9.49 mg/dNm3 (0.004 gr/dscf) and
18.9 mg/dNm3 (0.008 gr/dscf), respectively.
Test Run PASM-3, which was conducted only during slag skim-
ming operations, showed a lead concentration of 21.7 mg/dNm3
(0.009 gr/dscf).
The trace metal emission data support conclusions drawn in
Subsections 5.1 through 5.3 that the majority of fugitive emis-
sions from the No. 4 converter are generated during converter
roll-out modes. Data from Test PASM-3, conducted only during
slag skim operations, indicate that lead and bismuth emissions
are generated primarily during the converter skimming mode.
Emission data from Test PATC-2, during which the malfunction in
the primary hood system occurred, show higher levels of antimony
and bismuth than the other PATC test runs.
5.6 PROCESS SAMPLES
Table 62 summarizes analytical results for arsenic and lead
in the process samples collected by ASARCO personnel during the
test program. Results for each element are expressed as percent
by weight. Samples were digested according to procedures de-
scribed in draft EPA Method 108 and analyzed by atomic absorp-
tion. Although lead analysis was not required for these samples,
the results are reported to validate the lead results obtained
during the particulate and particle size distribution tests.
The El Indio and Le Panto concentrates are high in arsenic
(~11%). All the other charge materials are comparatively low in
arsenic (generally less than 0.2%). Also presented in Table 62
are the percentages of concentrate ore (by weight) charged during
154
-------
TABLE 62. ARSENIC AND LEAD IN PROCESS SAMPLES
Sample description
Test No. 1 (1/19) - Charge 79 -
Test No. 1 (1/19) - Charge 79 -
Test No. 1 (1/19) - Charge 79 -
Test No. 1 (1/19) - Charge 79 -
Test No. 2 (1/20) - Charge 80 -
Test No. 2 (1/20) - Charge 80 -
Test No. 2 (1/20) - Charge 80 -
Test No. 2 (1/20) - Charge 80 -
Test No. 2 (1/20) - Charge 80 -
Test No. 3 (1/22) - Charge 81 -
Test No. 3 (1/22) - Charge 81 -
Test No. 3 (1/22) - Charge 81 -
Test No. 3 (1/22) - Charge 81 -
Reverberatory matte - composite
from ASARCO
Gibraltor concentrates
Lornex concentrates
El Indio concentrates
Le Panto concentrates
St. Joe
Troy
cleanup blow
2nd blow
3rd blow
4th blow
cleanup
1st blow
2nd blow
3rd blow
4th blow
cleanup
3rd skim
4th skim
5th skim
as received
Arsenic
(As), %
0.09
0.15
0.13
0.18
0.07
0.17
0.22
0.15
0.13
0.19
0.12
0.14
0.24
0.70
0.008
0.08
10.8
11.1
NSR
NSR
Lead
(Pb), %
4.8
3.5
4.5
2.9
2.9
1.1
1.6
2.8
2.3
0.7
3.6
2.2
1.9
3.0
0.05
0.03
0.09
0.13
-
-
Total
Concentrate
charge - 1/83
% by weight
9.99
39.01
-
19.88
7.74
3.12
79.74
NSR = No sample received.
155
-------
the January test period. As shown, the concentrate charge ac-
counted for 79.74 percent of the total charge (the remainder of
20.26 percent represents other inert materials) and of this
total, 19.88 percent or approximately one fourth of the total ore
charge was high arsenic Le Panto concentrate.
156
-------
SECTION 6
QUALITY ASSURANCE
Because the end product of testing is to produce representa-
tive emission results, quality assurance is one of the main
facets of stack sampling. Quality assurance guidelines provide
the detailed procedures and actions necessary for defining and
producing acceptable data. Five such documents were used in this
test program to ensure the collection of acceptable data and to
provide a definition of unacceptable data. The following docu-
ments comprise the source-specific test plan prepared by PEDCo
and reviewed by EPA-IERL: the source-specific Quality Assurance
plan prepared by PEDCo and reviewed by EPA-IERL; the EPA Quality
Assurance Handbook Volume III, EPA-600/4-77-027; the PEDCo Envi-
ronmental Emission Test Quality Assurance Plan; and the PEDCo
Environmental Laboratory Quality Assurance Plan. The last two,
which are PEDCo's general guideline manuals, define the company's
standard operating procedures and are followed by the emission
testing groups and the laboratory groups.
For this test program, the following steps were taken to
ensure that the testing and analytical procedures produced qual-
ity data.
0 Calibration of all field sampling equipment. (Appendix
E describes calibration guidelines in more detail.)
0 Checks on train configuration and calculations.
0 Onsite quality assurance checks (i.e., sampling train,
pitot tube, and Orsat line leak checks) and quality
assurance checks of all test equipment prior to use.
0 Use of designated analytical equipment and sampling
reagents.
157
-------
0 Internal and external audits to ensure accuracy in
sampling and analysis.
Table 63 lists the sampling equipment used to perform the
particulate/arsenic, particle size, and SO tests as well as the
calibration guidelines and limits. In addition to the pre-test
and post-test calibrations, a field audit was performed on the
metering system used for the particulate/arsenic, particle size,
and SO sample runs. Figures 26 through 29 show an example audit
run for each dry gas meter used for particulate/arsenic, particle
size, and S0_ tests.
As a check on the reliability of the method used to analyze
the filters for the particulate tests, sets of filters that had
been preweighed in the lab were resubmitted for replicate anal-
ysis. Table 64 summarizes the results of a blank filter and
reagent analysis.
An impactor run was made to determine if the gaseous emis-
sions generated by the process biased the particle size results.
Two absolute filters were placed prior to the impactor to remove
filterable particulate emissions. The impactor's filter stages
were recovered and analyzed in the same manner as during the test
runs. As the data in Table 64 show, the filtered stack gas did
not react with the filter media.
The atomic absorption (AA) spectrophotometer used for the
trace metal analysis was calibrated for the specific analysis
each time a batch of samples was analyzed by use of NBS trace-
able 1000-ppm solutions of the individual metals. Table 65
presents results of the QA audit performed on samples supplied by
EMSL/QAD.
Audit solutions prepared by the EPA were used to check the
analytical procedures and reagent used for the manual S0_ sam-
pling and onsite analysis. Figure 30 presents the results of
these analytical audits. The audit results show that the ana-
lytical techniques were good.
The simultaneously performed manual S0_ test results were
compared with the monitor data to assure proper CEM operations.
158
-------
TABLE 63. FIELD EQUIPMENT CALIBRATION
Equipment
Meter box
PI tot tube
Thermocouple
Digital In-
dicator
Implnger
thermometer
Dry gas
Mter ther-
Trip balance
Barometer
Probe nozzle
Test
No.
FB-4
FB-S
FB-7
FB-9
015
032
278
284
133
147
202
203
124
221
FB-4 Inlet
FB-4 Outlet
FB-5 Inlet
FB-S Outlet
FB-7 Inlet
FB-7 Outlet
FB-9 Inlet
FB-9 Outlet
270
229
PATC 2-111
PATC 3-116
PASM 2-221
PSNCP
PSSSP
PEW
Calibrated
against
Uet test meter
Geometric speci-
fications
ASTN-3F
Millivolt signals
ASTM-3F
ASTM-2 or 3F
Type S weights
NBS traceable
barometer
Callper
Allowable
deviation
(Y 10.05 Y pre-test)
AH 3 10.15
(Y tO.05 Y post-test)
See Appendix E
1.5S
0.5%
i2°F
t5°F
tO. 5 g
+0.10 In.Hg
(0.20 post-test)
Dn 10.004 In.
Actual
deviation
-0.003
-0.06
-0.006
-0.008
-0.11
-0.001
-0.019
-0.03
*0.008
-0.022
-0.03
-0.004
0.046
0.075
0.017
-0.501
0.191
-1.07S
-0.611
-0.411
-0.06S
3.0-F
l.3"F
4.0°F
2.5-F
3.0°F
2.5°F
1.0'F
1.0'F
0.0 g
0.01 In.
0.00 1n.
0.00 In.
0.00 In.
0.002 In.
0.003 In.
0.001 In.
0.002 In.
Within
allowable
Units
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Comments
Used for partlculate/arsenlc tests
Used for partlculate/arsenlc and
S02 tests
Used for particle size tests
Used for particle size tests
Cp • 0.84
Ul
-------
TABLE 64. EXAMPLE BLANK FILTER AND REAGENT ANALYSIS
Sample type
Particulate
87-mm Reeve Angel
934 AH
No. 0002439
Particle size
64-mm Reeve Angel
934 AH
BG-05
BG-14
BG-91
BG-36
BG-01
BF-52
BG-27
BF-82
Backup 0000005
Acetone blank
(780 ml volume)
Original tare
weight, mg
370.4
160.0
142.7
161.0
143.9
161.4
148.4
168.6
143.0
217.8
103,264.7
Blank
weight, mg
370.9
160.3
142.7
161.0
143.8
161.9
148.2
168.6
143.0
219.0
103,272.9
Net
weight, mg
+0.5
+0.3
0.0
0.0
-0.1
+0.5
-0.2
0.0
0.0
+1.2
+8.9 =
0.028 mg/ga
0.01 mg/g used in calculations.
TABLE 65. ARSENIC AND SELENIUM QA AUDIT
Audit9
Sample
No.
1
2
3
Arsenic
Measured
value,
mg/ liter
0.027
0.194
0.072
True
value,
mg/ liter
0.024
0.182
0.061
95%
confidence
interval ,
mg/ liter
0.017-0.032
0.129-0.230
0.043-0.078
Selenium
Measured
value,
mg/liter
<0.012
0.046
0.014
True
value,
mg/liter
0.0087
0.048
0.016
95%
confidence
interval ,
mg/liter
none
listed
Received from Mr. Guy Simes, EMSL/QAD Cincinnati.
160
-------
DATE:
AUDIT REPORT DRY GAS METER
CLIENT:
BAROMETRIC'
ORIFICE NO.
PRESSURE (Pu-J: ^/0/in. Hg METER BOX NO. /3*~/
vf
IJr PRETEST
Y: Ii9&
ORIFICE K FACTOR: */, f7/ V/Z?"/ AUDITOR: 3>Z&
Orifice
manometer
reading
AH
in H20
W
Dry gas
meter
reading
Vtff
ft3
/T700D
V£/>ttf
Temperatures
Ambient
Tai/Taf
rT
io
Dry qas meter
Inlet
VTif
°F
^/
Ir?
Outlet
Toi/Tof
°F
r?
$T
Duration
of
run
0
min
"KII?-
Dry gas
meter
volume
Vm
ft3
Moo
Average temperatures
Ambient
Ta
°F
s?
Dry gas
meter
Tm
°F
Vmstd Xct
ft3 ft3
tOrf JO&Z'f //tOD^
Audit
Y
- I.O&
Y
deviation
-0&
/w
Vmstd
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)
VVt
(1203)( 0 )( K )(Pbar)
(T + 460)1/2
a
Audit Y
"•» „
a
-------
DATE:
AUDIT REPORT DRY GAS METER
CLIENT:
BAROMETR/C PRESSURE (P^'.ftjf in. Hg METER
ORIFICE NO. /# PRETES
BOX NO. f*8~S^
,T Y: ,9/T
ORIFICE K FACTOR: *fr79hr/0~V AUDITOR: UR0
Orifice
manometer
reading
AH
in H20
Iff
Dry gas
meter
reading
ft3'
£ ll.OOO
fftnw
Temperatures
Ambient
°F
67
(^
Dry gas meter
Inlet
°F
n
11
Outlet
Toi/Tof
°F
10
10
Duration
of
run
9
min
/f,#7
Dry gas
meter
volume
Vm
ft3
HIM
Average temperatures
Ambient Dry gas
meter
T T
a m
op op
V V
m in
mstd ma
ft3 f
££ 7£«T" II^O I\M
Audit
ct
Y
t3
fc? .WO
Y
deviation
-of
i
Vmstd
07.647M Vm )(Pbar + AH/13.6)
(Tm + 460)
Audit Y
v™ ,
act.
Vmstd
\ct
(1203)( 0 )( K )(P. )
(T + 460)1/2
a
Y deviation, %
(Y audit - Y pre-test)(100%)
(Y audit)
Audit Y must be in the range, pre-test Y +.0.05 r
Figure 27. Audit report dry gas meter
(Meter Box No. FB-5).
162
-------
DATE: _
BAROMETRIC'PRESSURE (Pbar):
ORIFICE NO.
AUDIT REPORT DRY GAS METER
CLIENT:
ORIFICE K FACTOR:
in. Hg METER BOX NO.
PRETEST Y: _
AUDITOR:
Orifice
manometer
reading
AH
in H20
Z'P-|
Dry gas
meter
reading
Yvf
TT;
W.000
/6(,,2DO
Temperatures
Ambient
T ./T *
°F
/ /I
ij i
LX"
Dry gas meter
Inlet
VTif
°F
7?
77
Outlet
Toi/Tof
°F
£?
10
Duration
of
run
min
frtrt.r
/J7/7T
Dry gas
meter
vol ume
Vm
ft3
fttfft)
Average temperatures
Ambient
Ta
°F
u
Dry gas
meter
Tm
°F
7i7T
V
std
ft3
I2.I2&
v_
mact
ft3
/£#£
Audit
Y
W2
Y
deviation
%
-13
n.i
\td
(17.647)( Vm )(Pbar + AH/13.6)
(TB + 460)
\ct
(1203)( 0 )( K )<
Pbar^
a
Audit Y
mapt
V|T1std
Y deviation, %
(Y audit - Y pre-test)(100%)
(Y audit)
Audit Y must be in the range, pre-test Y ±0.05 y
Figure 28. AUdit report dry gas meter
(Meter Box No. FB-7).
163
-------
DATE:
AUDIT REPORT DRY GAS METER
CLIENT:
BAROMETRIC PRESSURE (Pbar): 30 Pfr4n. Hg METER BOX NO.
ORIFICE NO.
ORIFICE K FACTOR:
PRETEST Y:
AUDITOR:
Orifice
manometer
reading
AH
in H20
£tt
Dry gas
meter
reading
Vvf
ft3
lfS.ooo
Ki. /co
Temperatures
Ambient
W
°F
ft
Si
Dry gas meter
Inlet
w
°F
52
^
Outlet
Toi/Tof
°F
&
sry
Duration
of
run
0
min
tffr*
ir.nt.
Dry gas
meter
volume
»•
n3
li.Ut>
Average temperatures
Ambient
Ta
°F
fi,r
Dry gas
meter
Tm
°F
ST3-
"•"std
ft3
/3.617
Xct
ft3
/5.CT7
Audit
Y
.K?
Y
deviation
%
~te
Vmstd
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)
Audit Y
v" ,
&£t
Vmstd
\ct
(1203)( 0 )( K )(P. )
(Ta + 460)1/2
a
Y deviation, %
(Y audit - Y pre- test) (100*)
(Y audit)
Audit Y must be in the range, pre-test Y ±0.05 '
Figure 29. Audit report dry gas meter
(Meter Box No. FB-9).
164
-------
PLANT AsAR
-------
As shown in Table 66, the manual S02 and CEM S0_ emission data
compare closely over the segments tested.
Standard quality assurance guidelines were followed through-
out the tracer study. All sample equipment was calibrated ac-
cording to applied EPA criteria prior to its use. The Perkin-
Elmer Model 3920 gas chromatograph was calibrated and checked
daily prior to sample analysis (as described in Appendix D).
Also, the limiting orifice associated with the SF- injection
b
system was calibrated prior to and after each sustained injection
by use of a 0-10 cc scale bubble meter and stop watch.
Background tests to determine the presence of SF., were
b
performed daily prior to the hood capture efficiency tests.
Background samples were obtained by collecting flue gas samples
from the exhaust duct and from the air curtain control area and
analyzed for SF . Table 67 summarizes the background analytical
results.
-12
Reported SF- concentrations were less than 5 x 10 , which
was below the working range of the GC calibration curves. Sam-
ples BR-A and B were designed to preclude the possibility of SFfi
recirculating within the air curtain control area. The SFg was
injected in the control area for a sustained period of time.
After the SF.. probe was removed, approximately 3 minutes were
b
allowed to elapse before samples were collected at the secondary
exhaust sample location and analyzed for SFC. As shown, the
b
values obtained were comparable to standard background data.
Before the transmissometer was shipped to ASARCO, the in-
strument was inspected and serviced by a Lear Siegler field
service engineer. The instrument was assembled in the laboratory
during this service inspection, and PEDCo personnel were in-
structed in the proper optical and electronic alignment proce-
dures. After the transmissometer arrived on site and before its
installation on the air curtain system, it was set up in a smoke-
free environment at a distance equal to the measurement distance,
and the optical and electronic alignment was checked. A general
operational check was also conducted. The instrument was then
166
-------
TABLE 66. COMPARISON
OF MANUAL AND CEM S02 RESULTS
Date
(1983)
1/14
1/14
1/14
1/14
1/14
1/14
1/14
Run
No.
PS02-1
PS02-2
PS02-3
PS02-4
PS02-5
PS02-6
PS02-7
Manual M-6
S02 concen-
tration, ppm
10.6
35.5
39.9
16.0
67.0
69.0
1.7
Average CEM
S02 concen-
tration, ppm
7.2
54.2
40.8
15.5
61.6
60.2
1.0
167
-------
TABLE 67. SUMMARY OF SFg BACKGROUND ANALYSIS
Sample
No.
BRA
BRB
B-l
B-2
B-3
B-4
B-5
B-6
B-7
B-8
B-9
B-10
B-ll
B-12
B-13
Sample
location
Air curtain control area
Air curtain control area
Exhaust duct
Air curtain control area
Exhaust duct
Ambient
Exhaust duct
Air curtain control area
Ambient
Ambient
Air curtain control area
Exhaust duct
Air curtain control area
Air curtain control area
Air curtain control area
Date
(1983)
1/13
1/13
1/13
1/13
1/14
1/14
1/17
1/17
1/17
1/18
1/18
1/19
1/19
1/20
1/20
SFg concentration - v/va
(< indicated values)
5 x ID'12
5 x ID'12
5 x ID'12
5 x 1C'12
5 x ID'12
5 x ID'12
5 x ID'12
5 x 1C'12
5 x ID'12
5 x 10"12
5 x 1C'12
5 x 10"12
5 x ID'12
5 x ID'12
5 x ID'12
'Reported concentrations are out of the working range of the ECD calibration.
168
-------
installed on the air curtain system while the No. 4 converter was
down. After the installation and while the No. 4 converter was
still down, alignment and calibration checks were conducted and
necessary adjustments were made.
Before their use, the strip chart recorders used to record
transmissometer output were calibrated with a multimeter cali-
brator. Each day before the start of testing, the optical align-
ment of the transmissometer was checked and zero and span cali-
bration checks were conducted.
169
-------
SECTION 7
SAMPLING AND ANALYTICAL PLAN
This section describes the sampling sites and the test
methods used to characterize the emissions captured by the air
curtain system on ASARCO's No. 4 Copper Converter.
Table 68 presents a sample matrix outlining the number of
segments tested during the particulate/arsenic and particle size
sample runs.
A PEDCo representative stationed in the converter building
coordinated all tests with a plant representative and with the
personnel operating the sampling trains to assure that sampling
was conducted during the proper segments of the converter cycle.
7.1 SAMPLE LOCATION
The manual testing, continuous monitoring, and tracer study
tests were conducted in the air curtain exhaust duct, as shown in
Figure 31.
Four 15-2-cm (6-in.) inside diameter (I.D.) sample ports,
located 8 stack diameters downstream and 2 diameters upstream
from the nearest flow disturbance, were used to collect the
particulate/arsenic, S0_, particle size, and tracer gas samples
in the 1.5-m (5 ft) I.D. round duct.
The S0_ CEM samples were obtained in a 6.4-cm (2^-in.) port
located ~112-cm ("44 in.) below the four sampling ports. These
sampling sites met all criteria specified in EPA Method 1.*
7.2 VELOCITY AND GAS TEMPERATURE
A Type S pitot tube and an inclined draft gauge manometer
were used to measure the gas velocity pressures at the manual
170
-------
TABLE 68. SAMPLE MATRIX
Staple triln type
Partlculate/arsenlc
continuous sampling
train
Partlculate/arsenic
specific segment
sampling train
Andersen Nark III
Impactor (charging
mode)
Andersen Nark III
Impactor (skinning
mode)
Andersen Nark III
Impactor with IS-mn
precutter (blowing
mode)
Run No.
PATC-1
PATC-2
PATC-3
PASM-1
PASM-2
PASM-3
PSNCP-1
PSMC-1
PSMC-2
PSHC-3
PSttC-4
PSHC-5
PSSSP-1
PSSS-1
PSSS-2
PSSS-3
PSBP-1
PSB-1
PSB-2
PSB-3
PSB-4
Date
(1983)
1/18
1/19
1/20
1/23
1/18
1/19
1/20
1/22
1/14
1/18
1/19
1/19
1/20
1/20
1/22
1/14
1/18
1/19
1/20
1/22
1/14
1/18
1/19
1/20
1/20
1/22
Segments sampled during the converter cycle
Matte charge.
No. of ladles
tested
14
14
4
14
14
9
14
2
10
4
4
Cold additions,
No. of ladles
tested
8
10»
lib
7
«0a
4
1
7
4
6a
,2b
Slag addition.
No. of ladles
tested
4.50
4.50
4.50
Slag skim.
No. of ladles
tested
10.75
9.75
7.25
10.75
9.75
7.25
10.75
10.75
9.25
7.25
No. of
blows
tested
4
4
3
4
4
2
1
1
Cleanup
blows
tested
1 •
1
1
1
1
1
Finish
blows
tested
1
1
1
1
1
1
Copper pour.
No. of ladles
tested
10
9
10
10
Four of the cold additions were blocks of blister.
Seven of the cold additions were blocks of blister.
-------
CROSS SECTION
DUCT I.D. - 60 in.
NIPPLE I.D. « 6 1n.
NIPPLE LENGTH - 2.5 in.
STACK DIAMETER
DOWNSTREAM » 8
STACK DIAMETER
UPSTREAM = 2
6-1n. I.D. PORTS
L METHODS
ING PORTS C 4"
n. I.D.) A
1
36 1/2 in.
I
ff, , , -,/.-•>-, ffS, , tf((> ,(f«
np PPORF IT . .
^~
1
f
FLOW
O,.,
©„
n,
TRAVERSE
POINT
1
2
3
4
"1 5
6
X
Bf
DISTANCE FROM
OUTSIDE OF
NIPPLE, in.
5.1
11.3
20.3
44.8
53.8
59.9
\R GRATE PLATE FORM
I
TO I.D. FAN
Figure 31. No. 4 converter air curtain
exhaust duct sample site.
172
-------
methods test site. Velocity pressures were measured at each
sampling point across the duct to determine an average value.
Measurements were taken in accordance with procedures outlined in
Method 2 of the Federal Register.* The temperature at each
sampling point was measured with a thermocouple and digital
readout.
7.3 MOLECULAR WEIGHT
Flue gas composition was determined in accordance with
procedures described in Method 3.* Grab samples were collected
at the manual test site during the preliminary impactor runs, and
an Orsat Gas Analyzer was used to analyze the bag contents for
oxygen and carbon dioxide. Because these results verified that
the gas streams were essentially air, additional samples were not
collected.
7.4 PARTICULATE/ARSENIC
Methods 5* and 108** (as described in the Federal Register*)
were used to measure particulate and arsenic concentrations. All
tests were conducted isokinetically by regulating the sample flow
rate relative to the gas velocity in the duct (as measured by the
pitot tube and thermocouple attached to the sampling probe). The
continuous sampling train traversed the cross-sectional area of
the duct, whereas the specific mode train was run at a single
point of average velocity. The sampling train consisted of a
heated glass-lined probe, a heated 7.6-cm (3-in.) diameter glass
fiber (Whatman Reeve Angel 934AH) filter, and a series of six
Greenburg-Smith impingers followed by a vacuum line, vacuum
gauge, leak-free vacuum pump, dry gas meter, thermometers, and a
calibrated orifice.
*
40 CFR 60, Appendix A, Reference Methods 1, 2, 3 and 5, July 1,
1981.
**
Method 108 is a proposed method.
173
-------
An acetone rinse of the nozzle, probe, and filter holder
portions of the sampling train was made at the end of each test.
Upon completion of the acetone rinse, an additional rinse with
0.1 N NaOH was performed. The acetone rinse and particulate
caught on the filter media were dried at room temperature, desic-
cated to a constant weight, and weighed on an analytical balance.
Total filterable particulate matter was determined by adding
these two values. The volume of water collected in the impinger
section of the sampling train was measured at the end of each
sample run to determine the moisture content of the flue gas.
The contents of the first two impingers were transferred to a
polyethylene container. The impingers and all connecting glass-
ware, including the back half of the filter holder, were rinsed
with 0.1 N NaOH and the rinse was added to the container. The
contents of the third, fourth, and fifth impingers were trans-
ferred to a polyethylene container. The impingers and all con-
necting glassware were rinsed with distilled water and added to
the container. Upon completion of the gravimetric analysis, the
filter acetone rinse and solids contained in the 0.1 N NaOH rinse
of the front half of the sampling train were prepped, combined,
and analyzed for arsenic (by atomic absorption). The contents of
the first two impingers and 0.1 N NaOH rinse also were analyzed
for arsenic by atomic absorption. The contents of the third,
fourth, and fifth impingers and distilled water rinse were ti-
trated with NaOH to determine S0_ concentrations.
7.5 PARTICLE SIZE DISTRIBUTION
Particle size samples at the manual methods test site were
collected with an Andersen Mark III cascade impactor. The Mark
III is an in-stack, multistage cascade impactor that yields a
total of eight particle cut sizes normally ranging from 0.5 to
15 ym. Substrates for this impactor were 64-mm-diameter glass
fiber filters.
A cyclone precutter was attached to the front of the impac-
tor used to sample during the blowing portions of the converter
174
-------
cycle. This removed the larger particles and avoided the need to
use buttonhook nozzles. Because the sampling rate could not be
adjusted to obtain the 15-ym cut point of the cyclone precutter,
the weight of particulate collected by the cyclone was added to
the weight in the first stage of the respective impactor.
All particle size samples were collected at a point of
average velocity near the centroid of the duct. The isokinetic
sampling rate was based on initial measurements of velocity,
pressure, and temperature. Constant cut-point characteristics
were maintained during sampling, but velocity pressures and
temperatures were measured periodically at the sampling point to
evaluate the actual variation in isokinetic rate. Nozzles were
selected to keep sampling rates in the recommended range of 0.3
to 0.75 acfm. Each filter was recovered, desiccated, and weighed
on an analytical balance. The inlet chamber and nozzle were
brushed and rinsed with acetone, and the rinse was evaporated,
desiccated, and weighed. This weight was added to the first
stage of the impactor.
Upon completion of the gravimetric analysis, the samples
were analyzed by atomic absorption spectrometry to determine the
concentrations of selected trace metals (arsenic, cadmium, lead,
antimony, selenium, and bismuth).
7.6 SULFUR DIOXIDE MANUAL METHOD
The EPA Method 6 sampling procedure described in the Federal
Register* was used to measure the sulfur dioxide concentration
for comparison with the CEM measurements. A single sampling
point located approximately in the center of the duct was sampled
at a constant rate for 20 minutes.
The EPA Method 6 sampling train consisted of a heated glass-
lined probe (containing a plug of glass wool at the probe tip), a
series of Greenburg-Smith impingers, a vacuum line, a dry gas
meter, and a leak-free diaphragm vacuum pump.
*
40 CFR 60, Appendix A, Reference Method 6, July 1, 1981.
175
-------
Upon completion of each sample run, the sampling train was
leak-checked and then purged with ambient air for 15 minutes.
The samples were recovered and analyzed for sulfur dioxide on
site by the barium-thorin titration method.
7.7 CONTINUOUS MONITORING FOR SULFUR DIOXIDE
A Thermo-Electron Model 40 S02 analyzer operating on the
principle of pulsed fluorescence was used for continual measuring
of the S0_ concentration in the gas stream. During the testing
period, a continuous sample was extracted from the duct at a
representative sampling point by use of an unheated stainless
steel probe. This gas then passed through an unheated coalescing
filter to remove particulate matter and moisture droplets, and
then into the Teflon sample line, which transported it to the
analyzer. The sample line pressure was indicated on a manometer
placed just prior to the analyzer.
Triplicate injections of each standard gas were made into
the analyzer during the initial startup period. The analyzer was
also calibrated daily by use of appropriate mid-range or span
calibration gases. A sample line integrity check also was con-
ducted daily during the testing period.
The CEM data collected during the test series were manually
reduced to a maximum S0_ concentration and average S0_ concentra-
tion in units of parts per million for each discernible event
(i.e., one S0_ peak resulting from one ladle of matte charged was
counted as one event, but two emission peaks that merged as a
result of two ladles of copper blister poured in rapid succession
also were treated as one event). The average SO_ concentration,
the duration of the event, and the average stack velocity were
used to calculate the pounds of S0? captured per event. The
manual S02 results were compared with the corresponding CEM data
to verify the CEM results.
176
-------
7.8 HOOD CAPTURE EFFICIENCY USING A TRACER GAS
7.8.1 Tracer Gas
Measured quantities of SF,. tracer gas were injected into the
b
air curtain control area. Measurements of the tracer concentra-
tion at the sampling point, combined with flow rate measurements,
permitted the calculation of the amount passing the sampling
point (i.e., the amount collected by the air curtain and suction
plenum). The collection efficiency was then calculated from the
amount injected and the amount captured on a mass flow basis.
Sulfur hexafluoride (SF,,) , which was used as the tracer, is
b
a colorless, odorless, tasteless gas that is nonflammable and
completely nontoxic. It is also stable up to a temperature of
500°C (923°F), and the minimum detection limit by the GC electron
capture analytical technique is 5 parts per trillion.
The SF^ was injected into the controlled area of the air
b
curtain at a constant rate. A constant pressure was maintained
on a limiting orifice to ensure a constant injection flow rate.
The injection system was calibrated by a bubble meter before and
after each sustained injection. The temperature at each injec-
tion point was monitored during the injection of the SFf to avoid
b
decomposition. The tracer was injected over a selected time
period or operation mode, as required.
Single point samples of the secondary hood flue gas were
collected at the downstream sampling location by pulling them at
a constant rate into a leak-free Tedlar 15-liter bag. Samples
were collected over a selected time period or operational mode,
as required.
SF, analysis was performed by using a Perkin-Elmer Model
b
3920 gas chromatograph equipped with a Ni-63 electron capture
detector and a Valco gas sampling valve with a 1-ml sampling
loop. An exponential dilution system was used to construct
calibration curves and the electron capture detector's response
to SF,. was determined with a conventional strip chart recorder.
D
Peak heights were measured to determine response and were com-
pared with calibration curves prepared by use of the exponential
177
-------
dilution system just before analysis to determine the actual
concentrations of SF...
D
7.9 OPACITY
The opacity of emissions escaping capture by the air curtain
and passing through the slot was monitored by use of a Lear
Siegler Model RM4 double-pass transmissometer. The RM4 was
chosen for this task because it has a design feature that makes
it insensitive to ambient light.
The instrument was installed at the top of the secondary
hood below the crane vail. Figure 32 presents a diagram of the
secondary hood and transmissometer installation. Visual obser-
vation from a position at the top of the secondary hood revealed
that emissions escaping the air curtain and exiting through the
slot were fairly uniform along the length of the slot. As a
result of this observation, the instrument was located near the
center of the slot in a position that appeared to allow minimum
interference to the measurement beam from the crane cables. In
this location the optical path length for the instrument was 5.89
m (19.33 feet) (flange to flange). A Leeds Northrup Speedomax
strip chart recorder with a 0 to 20 mA input was coupled to the
transmissometer output to record opacity data.
178
-------
BALOON
AIR CURTAIN
p
L
TRA
t
|(["7"f RETROREFLECTOR |
19-ft. 4-1n.
276-1n.
HTFivrn
i
Jr'
-u=
^ 123-1n. ^
187-1n. _
U L-J
SOITTH SIDE
NORTH SIDE
TOP view
NORTH SIDE
SOUTH SIDE
V4-1n.
UANSCEIVER
RETROREFLECTOR
AIR CURTAIN
PIAN VIEW
Figure 32. Transmissometer installation on secondary hood.
179
------- |