United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 80-SAL-2
August 1981
Air
IvEPA Secondary Aluminum
Emission Test Report
Vista Metals Corporation
Fontana, California
-------
9104.00
11/25/81
SECONDARY ALUMINUM SMELTING
EMISSION TEST REPORT
VISTA METALS CORPORATION
FONTANA, CALIFORNIA
/
Contract No. 68-02-3541
Work Assignment 1
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
EMISSIONS MEASUREMENT BRANCH
RESEARCH TRIANGLE PARK, NORTH CAROLINA
SEPTEMBER 1981
Submitted by
ENGINEERING-S CIENCE
125 West Huntington Drive
Arcadia, California 91006
-------
TABLE OF CONTENTS
List of Figures
List of Tables
Preface
SECTION 1 INTRODUCTION
1.1 Background
1.2 Brief Process Description
1.3 Emission Measurement Program
1.4 Description of Report Sections
SECTION 2 SUMMARY OF RESULTS
2.1 Reverberatory Furnace - Chlorination
System
2.2 Reverberatory Furnace Charging Well
Emissions
2.3 Reverberatory Furnace Combustion Stack
Emissions
2.4 Borings Dryer Emissions
2.5 Audit Sample Results
2.6 Cleanup Evaluation Results
SECTION 3 PROCESS DESCRIPTION AND OPERATIONS
3.1 General Process Description
3.2 Reverberatory Furnace Description
3.3 Emission Control Equipment
3.4 Process Operations During Testing
3.5 Conclusions
SECTION 4 LOCATION OF SAMPLING POINTS
4.1 Reverberatory Furnace Chlorination
Process Control Equipment
4.2 Borings Dryer
4.3 Particle Size Test Locations
4.4 Visible Emission Observation Locations
4.5 Fugitive Emission Observation Loctions
4.6 Scrubber Liquor Sampling Locations
4.7 Pressure Drop Measurement
4.8 Stack Gas Molecular Weight Sampling
Locations
1-1
1-1
1-1
1-2
1-6
2-1
2-1
2-31
2-31
2-31
2-36
2-40
3-1
3-2
3-4
3-6
3-13
4-1
4-1
4-6
4-6
4-6
4-9
4-9
4-9
4-9
4-9
-------
TABLE OF CONTENTS (Continued)
Page
SECTION 5 SAMPLING AND ANALYSIS METHODS 5-1
5.1 EPA Reference Methods Used in this Program 5-1
5.2 Particulate, Chlorine, and Chloride
Sampling and Analysis 5-2
5.3 Particulate, Condensible Hydrocarbon, and
Non-Condensible Hydrocarbon Sampling and
Analysis 5-12
5.4 Particle Size Distribution Tests 5-14
5.5 Visible Emissions Observations 5-20
5.6 Fugitive Emissions 5-20
5.7 Scrubber Liquor Sampling and pH Analysis 5-21
5.8 Scrubber Pressure Drop Measurement 5-21
5.9 Cleanup Evaluation Test Procedure 5-22
-------
LIST OF APPENDICES*
APPENDIX
A Computer Summary of Chlorination
B Example Calculations for Particulate,
Chlorine, and Chloride Emissions
C Field Data Sheets
C.I Reverberatory Furnace Chlorination
Scrubber (Inlet, Outlet)
C.2 Borings Dryer
D Sampling Logs
D.I Daily Summary Logs-
D.2 Sampling Task Log
D.3 Sample Handling Log
D.4 Sample Identification Log
E Particle Size Tests
E.I Computer Data Reduction Results
E.2 Field Data Sheets
E.3 ' Analytical Data Sheets
F Visible Emissions Observations
F.I Observer Certification Certificates
EPA Method 9 Guidelines
F.2 Chlorination Scrubber VEO Sheets
F.3 Furnace Charging Well VEO Sheets
G Fugitive Emissions
G.I EPA Method 22 Guidelines
G.2 Reverberatory Furnace Charging Well
G.3 Borings Dryer
H Scrubber Liquor and Pressure Drop
H.I Reverberatory Furnace Chlorination
Scrubber Pressure Drop Sheets
H.2 Reverberatory Furnace Chlorination Scrubber
Data Sheets for Scrubber Liquor pH
I Sample Train Calibration Data
I.I Orifice Calibration Data
1.2 Nozzle Calibration Data
1.3 Pitot Tube Calibration Data
1.4 Temperature Device Calibration Data Gas
Certification Data
* The List of Appendices is included for information only. Appendices
are not contained in this copy of the report.
-------
LIST OF APPENDICES
APPENDIX
J Sampling and Analytical Procedures
J.I Particulate, Chlorine, and Total Chloride
J.2 Particulate, Condensible Hydrocarbons, and
Non-condensible Hydrocarbons
K Analysis Data
L Audit Sample Analysis
L.I Chlorine Audit Sample Data
L.2 Chloride Audit Sample Data
M Cleanup Evaluation Results
N Project Participants
-------
LIST OF FIGURES
/
Figure Title
1.1 Schematic of Secondary Aluminum Smelting Process
at Vista Metals Corporation, Fontana, California 1-2
2.1 Andersen 2000 6-Stage Impactor Particle Size
Results: Particle Size Versus Percent Weight
Less/.Greater than Stated Sized - Reverberatory
Furnace Chlorination Scrubber Inlet 2-13
2.2a Andersen 2000 5-Stage Impactor Particle Size
Results: Differential Mass Loading (dM/d logD)
Versus Particle Diameter Reverberatory Furnace
Chlorination Scrubber Inlet 2-14
2.2b ' Andersen 2000 5-Stage Impactor Particle Results:
Differential Mass Loading (dM/d logD) Versus
Particle Diameter Reverberatory Furnace
Chlorination Scrubber Inlet 2-15
2.3a Visible Emissions Observations at the Reverberatory
Furnace Chlorination Scrubber Outlet, May 19, 1981 2-19
2.3b Visible Emissions Observations at the Reverberatory
Furnace Chlorination Scrubber Outlet, May 20, 1981 2-20
2.3c Visible Emissions Observations at the Reverberatory
Furnace Chlorination Scrubber Outlet, May 20, 1981 2-21
2.4 Visible Emissions Observations at the Charging Well
Afterburner Outlet, May 28, 1981 2~24
2.5 Andersen 2000 6-Stage Impactor Particle Size
Results: Particle Size Versus Percent Weight
Less/Greater Than Stated Size - Borings Dryer
Uncontrolled 2-34
2.6 Andersen 2000 5-Stage Impactor Particle Size
Results: Differential Mass Loading Versus
Particle Diameter Borings Dryer Uncontrolled 2-35
3.1 Side View of Reverberatory Furnace 3-3
3.2 Diagram of Wet Scrubber for Controlling
Demagging Emissions 3-5
4.1 Overhead View of Plant Layout Vista Metals
Corporation, Fontana, California 4-2
4.2 Reverberatory Furnace No. 2 Chlorination Chamber
and Scrubber system at Vista Metals Corporation,
Fontana, California 4-3
4.3 Reverberatory Furnace Chlorination System
Scrubber Inlet Sampling Location, Vista Metals
Corporation, Fontana, California 4-5
4.4 Reverberatory Furnace Chlorination System
Scrubber Outlet Sampling Location, Vista Metals
Corporation, Fontana, California 4-7
4.5 Borings Dryer Emissions Sampling Location, Vista
Metals, Corporation, Fontana, California 4-8
-------
LIST OF FIGURES—Continued
Figure Title Page
4.6 Overhead View of Emission Sources and Observer
Locations for Conduct of Visible Emission
and Fugitive Emission Observations at Vista
Metals Corporation, Fontana, California 4-10
4.7 Chlorination Scrubber Liquor Sampling Location
and Pressure Drop Measurement Locations,
Vista Metals Corporation, Fontana, California 4-11
5.1 Particulate, Chlorine, and Total Chloride
Sampling Train 5-3
5.2 Method 5A/THCA Sampling Train 5-13
5.3 Schematic of the Andersen Impactor Sampling Train 5-15
5.4 Schematic of the Andersen Impactor 5-16
-------
LIST OF TABLES
Table Title
2-la Summary of Controlled and Uncontrolled Total
Particulate, Chlorine, and Total Chloride
From the Reverberatory Furnace Chlorination
Scrubber 2-3
2-lb Summary of Controlled and Uncontrolled Total
Particulate, Chlorine, and Total Chloride From
the Reverberatory Furnace Chlorination Scrubber 2-4
2-2 Summary of Particulate, Chlorine, and Total
Chloride Measurements on Gases Entering the
Reverberatory Furnace Chlorination Scrubber 2-8
2-3 Summary of Particulate, Chlorine, and Total
Chloride Measurements Exiting at the Reverbera-
tory Furnace Chlorination Scrubber 2-9
2-4 Particle Size Results at the Reverberatory Furnace
Chlorination Scrubber Inlet 2-11
2-5 Visible Emissions Observations at the Reverberatory
Furnace Chlorination Scrubber Outlet 2-17
2-6 Visible Emissions Observations at the Reverberatory
Furnace Charging Well Outlet 2-23
2-7 Fugitive Emissions Observations at the Reverberatory
Furnace Charging Well 2-25
2-8 Fugitive Emissions Observations in the Borings
Dryer Charging Area 2-26
2-9 Fugitive Emissions Observations in the Borings
Dryer Central Area 2-28
2-10 Fugitive Emissions Observations in the Borings
Dryer Unloading Area 2-28
2-11 Pressure Drop Across Chlorination Scrubber and
Temperature and pH of Scrubber Liquor 2-30
2-12 Particle Size Results of Uncontrolled Borings
Dryer 2-33
2-13 Flame lonization Detector (FID) Data Summary on
Uncontrolled Gases at the Borings Dryer 2-37
2-14 Flame lonization Detector (FID) Data Summary on
Controlled Gases Exiting From the Borings
Dryer Afterburner 2-38
2-15 Vista Metals Audit Sample Results 2-39
2-16 Cleanup Evaluation Results 2-41
-------
PREFACE
The work described in this report was conducted by personnel from
Engineering-Science, Inc. (ES), TRW Environmental Engineering Division,
Vista Metals Corporation in Fontana, California, and the U.S. Environ-
mental Protection Agency (EPA).
The scope of work was initially issued under Task Orders 44 and 46
of EPA Contract No. 68-02-2815 and continued under Work Assignment: 1
of EPA Contract No. 68-03-3541. Engineering-Science personnel assigned
to the project included Mr. George Weant as Project Manager, Mr. Donald
R. Holtz, as Task Manager, and Mr. Larry Cottone as Test Team Leader
for the Vista Metals test. Mr. Cottone was also responsible for
summarizing data in this report.
Mr. Robert Newman of TRW, under contract to the Office of Air
Quality Planning and Standards, Industrial Studies Branch of the EPA, was
responsible for monitoring process operations during the test program
and for preparing Section 3.0, Process Description and Operations, of
this report. Mr. Lester Samstag and Mr. Harold Jochai of Vista Metals
contributed significantly to the success of the test program through
their cooperation and assistance.
Mr. Clyde E. Riley and Mr. Gary McAlister, Office of Air Quality
Planning and Standards, Emissions Measurement Branch of the EPA, were
the EMB Task Managers. Mr. James A. Eddinger, Office of Air Quality
Planning and Standards, Industrial Studies Branch, EPA, served as
5
Project Lead Engineer and was responsible for coordinating the process
operation monitoring.
-------
SECTION 1
INTRODUCTION
-------
1.0 INTRODUCTION
1.1 Background
The United States Environmental Protection Agency (EPA) is in the
process of developing the Standards of Performance for New Stationary
Sources (SPNSS) in the Secondary Aluminum Industry. When promulgated,
these standards will reflect the degree of emission limitation achiev-
able through application of the best demonstrated emission control
technology available. In developing these standards, EPA utilizes!
emission data obtained from existing sources in the aluminum industry
that appear to be well controlled. EPA engaged Engineering-Science to
conduct emission tests on secondary aluminum industry sources to obtain
these data and to develop and evaluate emission test methods for the
industry.
EPA's Office of Air Quality Planning and Standards (OAQPS) selected
the Vista Metals Corporation secondary aluminum smelter in Fontansi,
California, as a site for standards development testing. This report
summarizes the test program conducted at Vista Metals.
1.2 Brief Process Description
Figure 1-1 shows a simplified flow diagram of that portion of the
Vista Metals Corporation secondary aluminum smelting process pertaiining
to these tests. The following briefly describes the process:
Secondary aluminum smelting consists of converting various types
of aluminum scrap into aluminum alloy ingots. Selected scrap* and
alloys are blended and melted in natural gas or fuel oil firesd
reverberatory furnaces. The magnesium content of the molten metal
is reduced to a desirable level by injection of chlorine, the
chlorine combining with the magnesium to form magnesium chloride.
The magnesium chloride floats to the top of the melt and is removed
as dross. Although some chlorine escapes the melt and emits to
the control system during most of the chlorination period, the
chlorine emission rate probably increases significantly near
the end of the cycle when little magnesium remains for reaction.
1-1
-------
Schematic of Secondary Aluminum Smelting Process
at Vista Metals Corporation, Fontana, California
i
to
ALUMINUM
BORINGS AND
TURNINGS
ATMOSPHERE
1
AFTERBURNER
m
z
G>
z
m
m
3J
2
O
CO
O
fn
z
O
m
BORINGS
DRYER
SCRAP ALUMINUM-
ATMOSPHERE
T
AFTERBURNER
ATMOSPHERE
k CHARGING
7 WELL
COMBUSTION
ZONE
CHLORINATION
CHAMBER
MOLTEN ALUMINUM
REVERBERATORY FURNACE
ATMOSPHERE
INGOT
MACHINE
STORE & SHIP
-------
Following chlorination the metal is poured into ingot molds. The
process is a batch operation and schedules vary depending on the
type and magnesium content of the scrap charged to the furnace and
the specifications for the product. Support facilities, such as
the borings dryer and sweat furnace, operate as needed.
The Vista Metals Corporation plant in Fontana, California receives
a portion of the aluminum scrap in the form of borings and turnings
from machining of aluminum. Because the cutting oils associated
with these borings and turnings can interfere with operations; if
charged directly to the furnace, the borings and turnings are first
passed through a borings dryer. A natural gas fired afterburner
controls borings dryer emissions.
Dried borings and other scrap are loaded into the furnaces at the
charging well and melted by immersion in molten aluminum. Heat to
the process comes from the natural gas burned in the reverberatory
furnace combustion chamber. The molten aluminum after being brought
to temperature is purged with chlorine to remove magnesium impuri-
ties before being poured into molds. Emissions from the reverbera-
tory furnace discharge through three individual stacks; one each
for the charging well, the gas burner, and the chlorination process.
Charging well emissions pass through an afterburner before discharge;
combustion chamber emissions discharge directly to the atmosphere,
and chlorination chamber emissions pass through a packed bed scrubber
before discharge.
1.3 Emission Measurement Program
Engineering-Science conducted an emission measurement program at
Vista Metals Corporation, Fontana, California, during the period from
May 18 through May 22, 1981. The goals of the test program were to
characterize and quantify controlled and uncontrolled emissions from the
chlorination process and the borings dryer, determine control equipment
efficiencies and evaluate visible and fugitive emissions from the borings
dryer and all of the reverberatory furnace sources.
During the test program a representative of TRW, the NPNSS con-
tractor, recorded process data for the reverberatory furnace operation.
The chronology of the emission tests is contained in Daily Sampling Logs
located in Appendix D. The components of the measuring program were as
follows.
1-3
-------
1.3.1 Reverberatory Furnace Chlorination Emissions
Total Participate, Chlorine, and Chlorides in Gas Streams
Three concurrent test runs were performed at scrubber inlet and
outlet locations. Test runs planned for the settling chamber inlet
were dropped because of very low velocities and plugging of test equip-
ment with acid and particulate. Scrubber inlet and outlet tests were
scheduled to coincide as much as possible with the end of the chlorina-
tion cycles so that the expected higher chlorine emission rates during
that time could be measured.
Particle Size Distribution in Gas Stream at Scrubber Inlet
Six particle size runs were performed at the scrubber inlet.
Visible Emissions at Scrubber Outlet (EPA Method 9)
During each particulate-chlorine-chloride sample run, an observer
recorded opacities from the start of chlorination until darkness.
Gas Analysis of Gas Streams
Two Orsat runs were conducted at the scrubber inlet.
Scrubber Solution Collection
Samples of the scrubber liquor were collected periodically during
the test runs, and the pH and temperature recorded.
Pressure Drop Across Scrubber
The gas pressure drop across the scrubber was measured periodically
during each of the test runs.
1.3.2 Reverberatory Furnace Charging Well Emissions
Fugitive Emissions in Furnace Area (EPA Method 22)
Observations were conducted but simultaneous process data was not
documented.
1-4
-------
Visible Emissions at Charging Well Stacks (EPA Method 9)
Observations were conducted but simultaneous process data was not
documented.
1.3.3 Reverberatory Furnace Combustion Stack Emissions
Visible Emissions at Stack Outlet (EPA Method 9)
No observations were conducted.
1.3.4 Borings Dryer Emissions
Total Particulate, Condensible Hydrocarbons, and Non-Condens:Lble
Hydrocarbon Sampling in Gas Streams
One partial test run yielding marginal total particulate and
condensable hydrocarbon information was conducted on uncontrolled
emissions. A separate test run was conducted for non-condensable hydro-
carbons. For the controlled emissions, no particulate and condensible
hydrocarbon data were obtained, but a short non-condensable hydrocarbon
test run was completed.
Particle Sizing in Uncontrolled Gas Stream
One particle size run was conducted.
Fugitive Emissions in Dryer Area (EPA Method 22)
Emission occurrences were recorded during the test run.
Visible Emissions at Borings Dryer Stack (EPA Method 9)
Because emission testing was unsuccessful, these observations were
not conducted.
Gas Analysis of Gas Streams
Orsat grab samples were taken and analyzed for both controlled and
uncontrolled gas streams.
1-5
-------
1.3.5 Cleanup Evaluation
Prior to emission testing, each sample train to be used was
assembled and charged as if ready to perform a test for either chlorine/
chloride or condensable hydrocarbons. The unexposed impinger contents
and wash were then recovered, prepared, and analyzed according to pro-
cedure. The purpose of the cleanup was to establish blank values for
the sampling trains and also to familiarize the cleanup and analytical
personnel with the procedure.
Audit samples for both chlorine and chloride were prepared by EPA
and analyzed by Engineering-Science prior to the analysis of actual field
samples. The audit sample results were given immediately to EPA to
assess the accuracy of the analysis procedure.
1.4 Description of Report Sections
The remaining sections of this report present the Summary of
Results (Section 2.0), Process Description and Operations (Section 3.0),
Location of Sampling Points (Section 4.0), and Sampling and Analytical
Methods (Section 5.0). Descriptions of methods and procedures, field and
laboratory data, and calculations are presented in the various appendices,
as noted in the Table of Contents. Appendix L contains the results of
audit sample analyses, and Appendix M contains the results of the clean-
up evaluations performed on the sampling train equipment.
1-6
-------
SECTION 2
SUMMARY OF RESULTS
-------
2.0 SUMMARY OF RESULTS
2.1 Reverberatory Furnace - Chlorination System
The ES site test work plan for this investigation of particulaite,
chlorine and chloride loadings in the reverberatory furnace chlorination
chamber ventilation system included simultaneous measurements at the
settling chamber inlet, the settling chamber outlet/scrubber inlet and
at the exhaust stack. As predicted by ES and Vista Metals Corporation
engineers from a site visit the previous week, velocity measurements
at the settling chamber inlet site were found to be below the measure-
ment range of a standard inclined manometer or micromanometer. The
design and operation of the chlorination system limits gas flow from
the chamber to that amount resulting from displacement by chlorine gas
injection, from thermal expansion, and from some vaporization of metals.
Since the velocity of gas into the settling chamber was below the usable
range of available instrumentation, testing could not be conducted at
isokinetic conditions. Also, during the settling chamber inlet velocity
traverse the test team found that the pitot tube soon became plugged
with a green sticky substance, judged possibly to be hydrochloric acid
and aluminum oxide or other oxides and chlorides of aluminum and mag-
nesium. Even if isokinetic sampling could be achieved at this location,
the consistency of this substance would likely prevent completion of a
test run. The settling chamber appears to collect most of this material,
and plant operators said the settling chamber required frequent emptying.
Simultaneous testing was conducted for particulate, chlorine and
chlorides at the scrubber inlet and scrubber outlet locations. Single
2-1
-------
test runs were conducted the evenings of May 19, 20, and 21, 1981.
Process operations were monitored by a representative of TRW who also
coordinated actual periods of testing to insure samples were collected
under normal process operating conditions.
Table 2-la (English Units) and 2-lb (Metric Units) summarize the
results of particulate, chlorine and chloride sampling performed on
the inlet (uncontrolled) and outlet (controlled) sides of the chlorination
scrubber. The format of Table 2-la and 2-lb allows a quick evaluation
of inlet and outlet loadings during each test run as well as control
efficiencies for the different pollutants sampled.
2.1.1 Total Particulate Loading Results
Total particulate includes only the filter catch and particulate
washed from the probe and filter holder front-half. Inlet particulate
loadings from the test series ranged from 0.179 to 0.364 grains per
dry standard cubic foot (DSCF) with an average of 0.283 grains per
DSCF. Mass rates into the scrubber ranged from 2.12 to 4.80 pounds per
hour with an average of 3.57 pounds per hour. Corresponding scrubber
outlet values were 0.009 to 0.029 grains per DSCF with an average of
0.016 grains per DSCF, mass rates ranged from 0.109 to 0.337 pounds
per hour with an average of 0.193 pounds per hour. Particulate control
efficiencies ranged from 93 to 97.1 percent with an average value of
94.6 percent. Particulate loading results appear to be accurate and
should be acceptable for reference in Standards of Performance for New
Stationary Source (SPNSS) development.
2-2
-------
TABLE 2-la (English Units)
SUMMARY OF CONTROLLED AND UNCONTROLLED TOTAL PARTICULATE,
CHLORINE, AND TOTAL CHLORIDE FROM THE
REVERBERATORY FURNACE CHLORINATION SCRUBBER
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Ul
Run Number
Date
Description
REVERBERATORY FURNACE
CHLORINATION SCRUBBER3
TOTAL PARTICULATE EMISSIONSb
Gralns/DSCF (Probe & Filter)0
Pounds/Hour
Collection Efficiency (Z)
CHLORINE EMISSIONS
pproV (average)
Crains/DSCF
Pounds/Hour
Collection Efficiency (X)
CHLORIDE EMISSIONS
Front-Half (Probe & Filter)
Gralns/DSCF
Pounds/Hour
Collection Efficiency (2)
Back-Half (Implngers)f
Gralns/DSCF
Pounds/Hour
Collection Efficiency (X)
Total f
Gralns/DSCF
Pounds/Hour
Collection Efficiency (Z)
Run 1
May 19, 1981
Uncontrolled Controlled
Run 2
May 20, 1981
Uncontrolled Controlled
Run 3
May 21, 1981
Uncontrolled Controlled
Average
Uncontrolled Controlled
0.179
2.120
93.8
> 968
> 1.246
>14.788
0.086
1.018
0.011
0.132
18
0.024
0.283
98.1
83.1
0.2396
2.837e
89.2
0.325
3.855
87.6
0.014
0.173
0.026
0.306
0.040
0.479
0.364
4.806
93.0
> 6283
> 8.082
M06.651
0.107
1.415
11.622
153.354
0.029
0.337
144
0.186
2.14
98.0
96.0
99.7
11.729
154.769
99.6
0.005
0.057
0.043
0.493
0.048
0.550
0.305
3.782
97.1
> 1595d
> 2.03d
>25.147d
0.009
0.109
26
0.034
0.392
98.4
0.098
1.216
7.148
88.402
98.4
99.8
7.246
89.618
99.8
0.002
0.020
0.017
0.196
0.019
0.216
aScrubber Uncontrolled = Inlet values from Table 2-6.
Scrubber Controlled = Outlet values from Table 2-7.
bFllter catch and wash from probe and filter holder front half.
^Grains per Dry Standard Cubic Foot @ 68°F, 29.92 In Hg.
"Values based on run second half emissions equal to the first. Second half analysis results were Invalid.
eP.eanalysis of this casplo using a specific tori eleci.i icle indicates these numbers may be low by a factor of 13.
^Includes chlorine
0.283
3.569
94.6
> 2949
> 3.786
>48.862
98.1
0.097
1.202
93.3
6.336
81.531
99.6
6.433
82.747
99.5
0.016
0.193
63
0.081
0.938
0.007
0.081
0.029
0.332
0.039
0.415
-------
TABLE 2-lb (Metric Units)
SUMMARY OF CONTROLLED AND UNCONTROLLED TOTAL PARTICULATE,
CHLORINE, AND TOTAL CHLORIDE FROM THE
REVERBERATOR? FURNACE CHLORINATION SCRUBBER
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Run Number
Date
Description
REVERBERATORY FURNACE
CHLORINATION SCRUBBER3
Run 1
May 19, 1981
Uncontrolled Controlled
Run 2
May 20, 1981
Uncontrolled Controlled
Run 3
May 21, 1981
Uncontrolled Controlled
Average
Uncontrolled Controlled
Is*
TOTAL PARTICULATE EMISSIONS"
c
Mg/DNm3
Kg/Hour
Collection Efficiency (X)
CHLORINE EMISSIONS
ppmV (average)
Mg/DNm3
Kg/Hour
Collection Efficiency (Z)
CHLORIDE EMISSIONS
Front-Half (Probe & Filter)
Mg/DNm3
Kg/Hour
Collection Efficiency (%)
Back-Half (Implngers)f
Mg/DNm3
Kg/Hour
Collection Efficiency (Z)
Total*
Mg/DNm3
Kg/Hour
Collection Efficiency (%)
408.81
0.962
> 968
> 2852
>6.708
93.8
98.1
25.21
0.060
18
55
1.128
197 32
0.462 0.078
83.1
547e 60
1.2876 0.139
89.2
744 92
1.749 0.217
87.6
833.43
2.180
6283
18499.7
48.378
67.02
0.153
93.0
144
425.754
0.971
98.0
245 11
0.642 0.026
96.0
26603 98
69.560 0.224
99.7
26848 110
70.202 0.249
99.6
698.76
1.715
97.1
1595d
4647d
98.4
21.34
0.049
26
78
0.178
224 5
0.552 0.009
98.4
16362 39
40.098 0.089
99.8
16586 43
40.650 0.098
98.8
aScrubber Uncontrolled = Inlet values from Table 2-6.
Scrubber Controlled = Outlet values from Table 2-7.
"Filter catch and wash from probe and filter holder front half.
^Milligrams per Dry Normal Cubic Meter @ 20°C, 760 mm Hg.
"Values based on run second half emissions equal to the first. Second half analysis results were Invalid.
eReanalysls of this sample using a specific ion electricle indicates these numbers may be low by a factor of 13.
fIncludes chlorine
647.0 37.857
1.619 0.087
94.6
2949 63
8666 186.3
22.165 0.387
98.1
222 16
0.552 0.037
93.3
14504 65.67
36.982 0.151
99.6
14726 81.67
37.534 0.188
99.5
-------
2.1.2 Chlorine Loading Results
Inlet chlorine measurements ranged from a low of 968 pptnV (1.246
grains/DSCF) to a high of 6283 ppmV (8.082 grains/DSCF) with an average
concentration of 2949 ppmV (3.786 grains/DSCF). The corresponding.
outlet values were a range of 18 to 144 ppmV (0.024 to 0.186 grains/DSCF)
with an average of 63 ppmV (0.081 grains/DSCF). Chlorine gas removal
efficiencies for the scrubber system ranged from a low of 98.0% to a
high of 98.4% with an average value of 98.1%.
During Run No. 2 a process upset resulting in high chlorine
concentrations caused sampling solutions at the inlet test location to
became saturated with chlorine. Also, during transfer of these samples
from the test site to the ES Laboratory, pressure built up in the
inlet sample bottle, apparently due to chlorine gas released from
solution. These conditions indicate that actual chlorine levels at
the inlet location were higher than measured.
Chlorine loadings exceeded total chloride values for test Runs 1,
uncontrolled, 2, controlled, and 3, controlled, but it is suspected
this inconsistency resulted from chlorine loss during sample handling
and storage or from interferences in chloride analysis, and not from
problems with chlorine measurement or analysis. Chlorine measurement
results, at least at the outlet location, should, therefore, be acceptable
for SPNSS reference purposes. Inlet location chlorine concentrations,
except for run number 2 which is lower than actual, should also be
acceptable. On Run No. 3 inlet, on the second set of impingers that
served the latter half of the run, the analyst failed to achieve acceptable
chlorine titrations. To make the data from this third run usable, it was
2-5
-------
assumed the chlorine mass during the second half of the run equaled
that of the first. Section 5 further discusses measurements and
analysis.
2.1.3 Chloride Loading Results
Particulate chlorides were collected in the front half of the sample
train, and gaseous chlorides (including chlorine) were collected in the
back half. Particulate chloride loadings at the scrubber inlet ranged
from 0.086 to 0.107 grains/DSCF (1.018 to 1.415 pounds/ hour) and averaged
0.097 grains/DSCF (1.202 pounds/hour). Particulate chloride concentrations
at the scrubber outlet ranged from 0.002 to 0.014 grains/DSCF (0.020 to
0.173 pounds/hour) with an average of 0.007 grains/DSCF (0.081 pounds/hour).
Particulate chloride removal efficiency ranged from 83.1% to 98.4% with
an average control efficiency of 93.3%.
Gaseous chlorides, including chlorine, ranged from 0.239 to 11.628
grains/DSCF with an average of 6.336 grains/DSCF at the inlet site.
The 0.239 value, however, may be incorrect as subsequent chloride
analysis using the specific ion electrical method rather than the
the mercuric nitrate method indicated the number should be 3.14. Com-
paritive analysis of the other samples showed general agreement
between the two methods. The range at the outlet site was 0.017 to
0.043 grains/DSCF with an average of 0.032 grains/DSCF. As mentioned
previously, there is an obvious inconsistency in the data because some
of the chlorine loadings exceeding the gaseous chloride loadings
for the same sample. This may be due to chlorine loss from the sample
between the chlorine and chloride analysis.
2-6
-------
Total chloride loadings at the scrubber inlet ranged from 0.325 to
(
11.729 grains/DSCF with an average loading of 6.433 grains/DSCF. The
corresponding values at the outlet site are a range of 0.019 to 0.048
with an average value of 0.039 grains/DSCF. Because chloride concentra-
tions measured lower in some instances than theoretically possible,,
these values do not appear suitable for SPNSS reference.
2.1.4 Summary of Particulate, Chlorine and Total Chloride Tests
Tables 2-2 and 2-3 summarize parameters measured during the
particulate, chlorine and total chloride tests conducted on the inlet
and outlet of the chlorination scrubber at the Vista Metals Corporation,
Fontana, California. All tests were accomplished within the specified
isokinetic rate of 100 + 10%.
Gas flow rates measured at the two sites were fairly constant.
The inlet values were consistently higher than the outlet values. The
higher inlet values are possibly the result of turbulence caused by
bends and dilution near the inlet port. The test crew experienced
some plugging of the pitot tube at the inlet site by particulate material
and frequently used a compressor to clear the lines.
Gas measurements were made at the inlet site on May 20 and 21,. As
expected, due to the high dilution factor, oxygen and CC>2 values were
similar to ambient air. The oxygen values exceeded 20.9% (i.e. 21.3%)
because the chlorine gas was absorbed as oxygen by the Orsat analyser.
During the May 20th chlorination, the plant operators expressed
concern that magnesium was not being removed from the molten metal as
fast as expected, and the greenish gas observed at the air-bleed-in
2-7
-------
TABLE 2-2
Run #1
5/19/81
Inlet
57.980
1385
3.5
0.13
20.9
< 0.1
63.3
104.0
10.5
Run #2
5/20/81
Inlet
95.378
1540
3.6
0.13
20.9
< 0.1
65.0
102.5
11.0
Run #3
5/21/81
Inlet
61.472
1445
3.51
0.13
20.9
< 0.1
67.8
105.6
11.3
Average
Inlet
71.61
1457
3.5
0.13
20.9
< 0.1
65.4
104.1
11.3
SUMMARY OF PARTICULATE, CHLORINE, AND TOTAL CHLORIDE MEASUREMENTS
ON GASES ENTERING THE REVERBERATORY FURNACE CHLORINATION SCRUBBER
AT VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Run Number
Date
Location
Volume Gas Sampled (DSCF)a
Volumetric Flowrate (DSCFM)b
% Moisture (Runs 1 & 3 assumed
same as Run 2)
% C02
% 02
% CO
Stack Temperature (°F)
% Isokinetic
Scrubber Solution pH
TOTAL PARTICULATE EMISSIONS (Probe &
Filter)
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
CHLORINE EMISSIONS
Average ppmV
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
CHLORIDE EMISSIONS
Front-Half (Probe & Filter)
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
Back-Half (Impingers)d
Total Sample Weight (milligrams)d
Grains/DSCFd
Pounds/Hourd
Total
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
672.6
0.179
2.12
2255.8
0.364
4.806
1218.9
0.305
3.782
1382.4
0.283
3.569
> 968
> 4,692
> 1.246
M4.788
> 6283
> 50,060
> 8.012
>106.651
> 1595C
> 8,184^
> 1.03°
>25.147C
> 2949
>20,979
> 3.786
>48.862
323
0.086
1.018
900e
0.239e
2.837e
1223
0.325
3.855
664
0.101
1.415
71,978
11.622
153.354
72,641
11.729
154.769
392
0.097
1.216
28,491
7.148
88.402
28,883
7.246
89.618
460
0.097
1.202
33,790
6.336
81.531
34,249
6.433
82.747
a) Dry Standard Cubic Feet @ 68°F, 29.92 inches Hg.
b) Dry Standard Cubic Feet per minute.
c) Chlorine analysis for only the first one-half of this run are valid. These
values are based on the assumption that second half emissions equal the first.
d) These values include chlorine as chloride, and may be suspect because of possible
chlorine loss to the atmosphere or problems with chloride analysis.
e) Reanalysis of this sample using a specific ion electride indicates these numbers
may be low by a factor of 13.
2-8
-------
TABLE 2-3
SUMMARY OF PARTICULATE, CHLORINE, AND TOTAL CHLORIDE MEASUREMENTS
EXITING THE REVERBERATORY FURNACE CHLORINATION SCRUBBER
AT VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Run Number
Date
Location
Volume Gas Sampled (DSCF)a
Volumetric Flowrate (DSCFM)b
% Moisture
% C02
% 02
% CO
Stack Temperature (°F)
% Isokinetic
Opacity (%)
Pressure Drop (inches ^0)
TOTAL PARTICULATE EMISSIONS (Probe &
Filter)
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
CHLORINE EMISSIONS
Run #1
5/19/81
Outlet
72.273
1395
1.6
< 0.1
20.9
< 0.1
65.3
97.5
12.4
2.9
Run #2
5/20/81
Outlet
117.518
1341
2.5
< 0.1
20.9
< 0.1
73.5
98.2
8.4
2.8
Run #3
5/21/81
Outlet
63.260
1362
2.8
< 0.1
20.9
< 0.1
74.6
91.2
5.5
2.9
Average
Outlet
84.350
1366
2.3
< 0.1
20.9
< 0.1
71.1
95.6
8.8
2.9
51.7
0.011
0.132
223.5
0.029
0.337
38.3
0.009
0.109
104.5
0.016
0.193
Average ppmV
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
CHLORIDE EMISSIONS
Front-Half (Probe & Filter)
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
Back-Half (Impingers)c
Total Sample Weight (milligrams)0
Grains/DSCFc
Pounds/Hour0
Total
Total Sample Weight (milligrams)
Grains/DSCF
Pounds/Hour
18
111
0.024
0.283
144
1417
0.186
2.14
26
138
0.034
0.392
63
555.3
0.081
0.938
68
0.014
0.173
120
0.026
0.306
188
0.040
0.479
0
38
.005
0.057
327
0.043
0.493
365
0.048
0.552
7
0.002
0.020
69
0.017
0.196
76
0.019
0.217
38
0.007
0.081
172
0.029
0.332
210
0.039
0.415
a) Dry Standard Cubic Feet @ 68°F, 29.92 inches Hg.
b) Dry Standard Cubic Feet per minute.
c) These values include chlorine as chloride, and are suspect because of apparent
chlorine loss to the atmosphere or problems with chloride analysis.
2-9
-------
location indicated that chlorine was not reacting well with the magnesium.
Mr. Jochai of Vista Metals indicated that trace metals in the melt may
be inhibiting the reaction. As previously mentioned, high inlet and
outlet chlorine/chloride concentrations were measured during this
chlorination.
2.1.5. Particle Size Tests
2.1.5.1 Reverberatory Furnace Chlorination Scrubber Inlet
Particle size test results of the scrubber inlet are summarized in
Table 2-4 and Figures 2.1., 2.2a, and 2.2b. Test Runs 1 and 2 were
conducted on May 18, Run 3 on May 19, Runs 4 and 5 on May 20, and Run 6
on May 21. Test Runs 1 and 2 were conducted during the third quarter
of the chlorination period. Run No. 3 was conducted at the end of the
chlorination cycle, Run No. 4 within the first quarter of the cycle,
and Runs 5 and 6 in approximately the middle of the cycles.
Grain loadings during the scrubber inlet particle sizing runs ranged
from a low of 0.789 gr/DSCF for Run No. SI-PS-4, to a high of 5.928 gr/DSCF
for Run No. SI-PS-1, and averaged 2.227 gr/DSCF. Run No. SI-PS-1 was
conducted during a period of process malfunction due to a broken chlorine
Injection probe. The chlorine gas supply was shut off, and the broken
probe removed at 2015 hours, only 2 minutes after the three minute par--
ticle sizing run was terminated. The broken chlorine probe resulted
in a process upset condition which could possibly account for the extreme
skew in the size distribution toward large particles, and the highest
grain loading for the particle size tests performed. In all cases the
fraction of particles exceeding a Dp50 cut point of 11 microns was greater
than 73%.
2-10
-------
TABLE 2-4
PARTICULE SIZE RESULTS3 AT THE REVERBERATORY FURNACE
CHLORINATION SCRUBBER INLET AT
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Test Date
Time, and
Run No.
5/18/81
2010-2013
SI-PS-1
5/18/81
2045-2048
SI-PS- 2
5/19/81
2108-2111
SI-PS-3
(l)Sampling Duration Stage
(2)Impactor Flow Rate Index
(3)Stack Temp (°F) No.
(1)
(2)
(3)
(1)
(2)
(3)
(1)
(2)
(3)
3.0 minutes
0.647 ACFM
75°F
3.0 minutes
0.635 ACFM
75°F
3.0 minutes
0.649 ACFM
62°F
SOD
SI
S2
S3
S4
S5
Back-up
filter
S0b
SI
S2
S3
S4
S5
Back-up
filter
S0b
SI
S2
S3
S4
S5
Back-up
filter
Delta
Weight
(mg)
692.55
1.56
2.96
3.43
2.93
0.0
0.0
110.75
2.86
5.67
6.23
5.88
1.99
2.26
129.13
2.40
8.81
1.98
0.03
0.78
0.0
% In
Size
Range
98.45
0.22
0.42
0.49
—
—
—
81.65
2.11
4.18
4.59
4.34
1.46
1.67
90.22
1.68
6.16
1.38
0.02
0.54
—
Cumulative
%less than
Size Range
1
1
0
0
18
16
12
7
3
1
9
8
1
0
0
.55
.32
.90
.42
—
—
—
.35
.24
.06
.47
.13
.67
—
.78
.10
.95
.57
.54
—
—
Size
Range
(microns)
>11
7.11-11
3.32- 7
2.16- 3
1.11- 2
0.54- 1
0.0C- 0
>11
7.18-11
3.35- 7
2.18- 3
1.12- 2
0.54- 1
0.0C- 0
>11
7.04-11
3.29- 7
2.13- 3
1.10- 2
0.53- 1
0.0C- 0
.35
.35
.11
.32
.16
.11
.54
.46
.46
.18
.35
.18
.12
.54
.23
.23
.04
.29
.13
.10
.53
Effective
Cut Diameter
(Dn50-microns)
11.
7.
3.
2.
1.
0.
-
11.
7.
3.
2.
1.
0.
-
11.
7.
3.
2.
1.
0.
-
35
11
32
16
11
54
-
46
18
35
18
12
54
-
23
04
29
13
10
53
-
a. particle sizing determinations employed a 6-stage Anderson Mark III impactor.
b. Nozzle, pre-cutter, inlet cone, and zero stage wash weight added to stage 1 weight (index No. SO)
c. Back-up filter has an actual 0.3 micron retention (DOP).
-------
TABLE 2-4 (Cont'd)
PARTICULE SIZE RESULTS3 AT THE REVERBERATORY FURNACE
CHLORINATION SCRUBBER INLET AT
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
NJ
I
Test Date
Time, and
Run No.
5/20/81
1811-1823
SI-PS-4
5/20/81
2024-2037
SI-PS-5
5/21/81
2028-2031
SI-PS-6
(l)Sampling Duration Stage
(2)Impactor Flow Rate Index
(3)Stack Temp (°F) No.
(1)12.0 minutes
(2)
(3)
0.668 ACFM
64° F
(1)13.0 minutes
(2)
(3)
(1)
(2)
(3)
0.655 ACFM
64°F
3.0 minutes
0.643 ACFM
69°F
SOD
SI
S2
S3
S4
S5
Back-up
filter
S0b
SI
S2
S3
S4
S5
Back-up
filter
S0b
SI
S2
S3
S4
S5
Back-up
filter
Delta
Weight
(mg)
288.78
17.13
20.87
15.01
15.89
13.19
24.04
891 .96
35.05
38.54
26.06
27.21
27.39
104.57
241.20
3.31
4.25
2.49
2.53
2.03
1.68
% In
Size
Range
73.13
4.34
5.28
3.80
4.02
3.34
6.09
77.50
3.05
3.35
2.26
2.36
2.39
9.09
93.67
1.29
1.65
0.97
0.98
0.79
0.65
Cumulative
%less than
Size Range
26
22
17
13
9
6
22
19
16
13
11
9
6
5
3
2
1
0
.87
.54
.25
.45
.43
.09
.50
.46
.11
.84
.48
.09
.33
.04
.39
.42
.44
.65
—
Size
Range
(microns)
>11 .08
6.94-11.08
3.24-
2.11-
1.08-
0.52-
0.0C-
>]
6.94
3.24
2.11
1.08
0.52
LI. 19
7.01-11.19
3.28-
2.13-
1.09-
0.53-
0.0C-
>]
7.01
3.28
2.13
1.09
0.53
11.34
7.10-11.34
3.32-
2.15-
1.11-
0.54-
0.0C-
7.10
3.32
2.15
1.11
0.54
Effective
Cut Diameter
(Dp50-microns)
6
3
2
1
0
11
7
3
2
1
0
11
7
3
2
1
0
.08
.94
.24
.11
.08
.52
—
.19
.01
.28
.13
.09
.53
—
.34
.10
.32
.15
.11
.54
—
a. particle sizing determinations employed a 6-stage Anderson Mark III impactor.
b. Nozzle, pre-cutter, inlet cone, and zero stage wash weight added to stage 1 weight (index No. SO)
c. Back-up filter has an actual 0.3 micron retention (DOP).
-------
FIGURE 2.1
Dso CUTPOINT, microns
~~ *• w * «* » =H b w - N w *«•»-«•«• O O O O O O O O O O _JLJ
Particle Size Results
Andersen 6-Stage Mark III Impactor
everberatory Furnace Chlorination Scrubber Inlet
Vista Metals Corporation, Fontana, CA
Assumed Density =1.0
"IF" = Impactor Flowrate
!
! 1
i
i
\
! 1 1
i ' '
©
(7\
1 ^^
i
G
j
i JA
! A
(1) Al
y H
} &
fl
H*y
P)
R
^,
i i ! '
i
i
to
! i
! 1 i !
J I
| 1
! t ; i
j :
i i
i i
1
. (jj
r»iA
»— *c
rjTj^s
^y
^s
^
•
i
U, i-
^^J
1
1
|
i
!
© SI-PS-1
IF =
E SI-PS-2 IF =
^ SI-PS-:
' 0 bl-PS-4
, <$> SI-PS-C
0 SI-PS-6
t
i
IF =
IF =
IF =
IF =
0.647
0.635
0.649
0.668
0.655
0.643
i i
; i
i
i '
i
—
i
j j
1
!
i
i
\
\
i
i
i
I
1 ,
j ; '
'
l
0.01 0.04 0.1 0.2 JJ I 2 i 10 » 30 «> 30 60 "0 30 30 38 98 99 J9.3 399 99.99
CUMULATIVE % LESS THAN STATED SIZE
2-13
ENGINEERING-SCIENCE
-------
FIGURE 2.2a
Reve
100000
10000
2
o
CO
o
O)
J
Q 1000
o»
0
•o
2
•o
o
c
9 100
CO
o
_J
CO
(0
ea
10
1
.]
Particle Size Results
Andersen 6-Stage Mark III Impactor
rberatory Furnace Chlorination Scrubber Inlet
Vista Metals Corporation, Fontana, CA
O RUN SI-PS-1
IF = 0.647
E3 RUN SI-PS-2
IF = 0.635
A RUN SI-PS-3
IF = 0.649
"IF" = Impactor Flowrate
n
1
j
A
i 1— .;_
I
i
1
i
!
i
i
i
i
11 1 i '
1 PI
F1
L-J
0
©
A
st
&
m
j
!
/if
® $
j 1 * — i — T™ — r*-
1
i i
i ' i
!
!
i
|
i
• j
1
.
I
i
i
i
i
j
i i
o i
©
j
! i
: i ''
i
j
I!
i
; i
• : i
^1 j
j i
i 1
1 i
-\
i ' i
i ' !
\
! i
• i |
I i
: i \
i i i — ' ' ; — t-r-
; i • i
1 ! i !
i
i
1 i i !
i
1 \
I
MM
; * • i '
i i . . ., •
i i
: i i
I
1 !
i
j i i 1 j i
!
i
t 1 10
Geometric Mean of Particle Diameter (microns)
100
2-14
ENGINEERING-SCIENCE
-------
FIGURE 2.2b
Reve
100000
10000
<••»
2
o
CO
o
«».
0)
E
Q 1000
a
0
J
•o
*••
2
•o
at
c
T3 100
a
o
-I
CO
CO
(0
2
10
1
^
Particle Size Results
Andersen 6-Stage Mark III Impactor
rberatory Furnace Chlorination Scrubber Inlet
Vista Metals Corporation, Fontana, CA
© RUN SI-PS-4
IF = 0.668
H RUN SI-PS-5
IF = 0.655
A RUN SI-PS-6
IF = 0.643
"IF" = Impactor Flowrate
i
I
i
]
i
-j-
j
<
•
k
£
i
I
&
i
!
i
ai
A
&
(•)
I*
_.
(i
1
1
!
i
-^—3:
i
•
Hi
jK.
V
1 '
\
\ '
' 1 '
! i i
i
1
i !
i 1 j — i — . — i — | i |
• ! 1
i
, •
i
!
!
i
i
1
I
I
i
—
]
A
i-r (
©
j
; 1 [
' 1
j
1
|
! ! i
i i
i
1
i
i
> i i
1
' '
L 1 10
Geometric Mean of Particle Diameter (micro is)
100
0
2-15
ENGINEERING-SCIENCE
-------
If results from Test Run 1 can be considered outliers due to a
process malfunction, then it would appear that the middle, not the end
of the chlorination cycle exhibits a larger percentage of the emissions
from the damaging operation. Chlorination cycle emissions were highly
variable however, and because particle size test runs were relatively
short there was little opportunity to dampen out the variations.
2.1.6 Visible Emissions Observations
Table 2-5 summarizes visible emission observations made of the
reverberatory furnace chlorination scrubber exhaust. Figures 2.3a, 2.3b,
and 2.3c graphically illustrate these observations. Observations are
presented in six minute averages for each test run. Observations made
on May 19, 1981 had the highest (24.0%) and lowest (0%) six minute
average opacities during the test program. Of the three sampling
days, the visible emissions observer expressed the most confidence in
readings on the last day, May 21. Observations on all three days were
difficult due to intermittent steam emissions and scattering of the
plume by wind, but on the last day the lower wind speed allowed for
more confident readings. On the first day, May 19, a 45 second period
of high recorded readings was discarded because the observer read the
steam plume opacity. This was the first observation where the plume
appeared to contain steam.
Visible emissions observations were made on the reverberatory
furnace charging well afterburner exhaust on May 28, 1981. No process
data was collected during this period. Six minute average observations
2-16
-------
TABLE 2-5
VISIBLE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHLORINATION SCRUBBER OUTLET
AT VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Date
5/19/81
Run
Number
Six-Minute
Time Period
1741:00
1747:00
1753:00
1800:00
1806:00
1812:00
1815:00
1821:00
1827:00
1833:00
1839:00
1845:00
1851:00
1857:00
1903:00
1909:00
1915:00
1921:00
1927:00
1933:00
1939:00
1945:00
1951:00
1746:45
1752:45
1756:10
1805:45
1811:45
1813:45
1820:45
1826:45
1832:45
1838:45
1844:45
1850:45
1856:45
1902:45
1908:45
1914:45
1920:45
1926:45
1932:45
1938:45
1944:45
1950:45
1952:45
Average Opacity
(Percent)
1.0
11.2*
11.9
24.0
21.0
0
12.1
18.5
16.5
22.5
Average
19.6
17.7
12.5
1.3
0
2.9
13.1
15.4
12.9
9.4
10.8
15.2
15.6
12.4
Observer
Location
150 ft. west of stack
60 ft. north of stack
on plant floor
150 ft. west of stack
(1817 - 150 ft. NW of
stack on roof line)
Comments; Gusty winds
and steam in the plume
made observations
difficult.
5/20/81 ;
I 1732:00
1738:00
1744:00
1750:00
1756:00
1802:00
1808:00
1814:00
1820:00
1826:00
1832:00
1838:00
1844:00
1737:45
1743:45
1749:45
1755:45
1801:45
1807:45
1813:45
1819:45
1825:45
1831:45
1837:45
1843:45
1849:45
7.5
12.9
12.1
11.5
15.0
11.9
13.1
11.5
9.8
11.7
11.3
9.6
9.0
100 ft. NW of discharge
Comments: Wind changes
and steam in the plume
interfered with accurate
observations .
*Due to high bias, data for 1750:15-1750:45 were discarded.
2-17
-------
TABLE 2-5 continued
VISIBLE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHLORINATION SCRUBBER OUTLET
AT VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Run Six-Minute Average Opacity Observer
Date Number Time Period (Percent) Location
5/20/81 2 1850:00
(cont'd.) 1856:00
1902:00
1908:00
1914:00
1920:00
1926:00
1932:00
1938:00
1944:00
1950:00
Average
5/21/81 3 1855:00
1901:00
1907:00
1913:00
1919:00
1925:00
1931:00
1937:00
1943:00
1949:00
1955:00
2001:00
2007:00
2013:00
2019:00
2025:00
2031:00
2037:00
2043:00
Average
1855:45
1901:45
1907:45
1913:45
1919:45
1925:45
1931:45
1937:45
1943:45
1949:45
1951:45
1900:45
1906:45
1912:45
1918:45
1924:45
1930:45
1936:45
1942:45
1948:45
1954:45
2000:45
2006:45
2012:45
2018:45
2024:45
2030:45
2036:45
2042:45
2048:45
8.5
8.3
5.6
2.3
4.0
9.0
5.0
5.8
4.8
1.0
0.2
8.4
2.3 NW of discharge
0
6.7 Comments: Because of
10.2 improved conditions,
' 15.0 the observer had more
13.1 confidence in these
3.8 readings than those
3.1 on May 19 and 20.
9.6
8.5
3.3
0.6
3.3
5.6
1.5
5.4
4.8
4.6
3.1
5.5
2-18
-------
25
20
L
o
cc.
15
a.
o
a:
10
1730
1800
1830
1900
1930
2000
2030
2100
VISIBLE EMISSIONS OBSERVATIONS AT THE REVERBERATORY FURNACE CHLORINATION SCRUBBER OUTLET
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA - MAY 19, 1981
en
33
-------
25
20
15
i
ts>
O
UJ
Q-
Ci
•a:
Q-
C3
C£.
UJ
m
z
z
m
m
3)
z
m
z
m
LO.
1730
1800
1830
1900
1930
2000
2030
2100
VISIBLE EMISSIONS OBSERVATIONS AT THE REVERBERATORY FURNACE CHLORINATION SCRUBBER OUTLET
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA - MAY 20, 1981
-------
to
I
to
rn
z
g
z
m
m
JU
z
D
l
0)
rn
z
o
m
25
20
Ul
o
Ul
a.
E15
o
-------
ranged'from 0 to 2.4 percent opacity. Table 2-6 lists these six-minute
averages. Figure 2.4 is a graphical representation. Visible emissions
at Vista Metals, with some reservations because of the steam plume from the
chlorination scrubber, should be acceptable for SPNSS reference purposes.
2.1.7 Fugitive Emissions Observations
Fugitive emission observations were made at the reverberatory
furnace charging well on May 28 according to EPA reference Method 22.
Emission frequencies ranged from 12.6 to 86.6 percent during the three
hours and 10 minutes of observation. Table 2-7 shows these observations.
Fugitive emissions observations were made at the borings dryer
charging, central and discharge areas on May 22, 1981. The results of
the observations are summarized in Tables 2-8 through 2-10. Emission
frequencies were 63% and 75% of the observation periods at the charging
area, 96% and 100% at the central area and 100% at the discharge area.
2.1.8 Scrubber Liquor Analysis
Scrubber liquor samples were collected during conductance of
particulate, chlorine, chloride tests on the scrubber inlet and outlet
on May 19, 20, and 21. The temperature of the liquor was measured
immediately upon sample collection. The pH of the liquor was measured
approximately one hour after collection of the samples. This allowed
the temperatures of the samples to stabilize.
The temperature of the liquor ranged from a low average of 66°F
on May 19 to a high average of 77°F on May 21. The low average pH was
10.4 on May 19 and the high average pH was 11.3 on May 21. These data
are summarized in Table 2-11.
2-22
-------
TABLE 2-6
VISIBLE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHARGING WELL OUTLET AT
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Run Six-Minute Average Opacity
Date Number Time Period (Percent)
5/28/81 1 1015:00
1021:00
1027:00
1033:00
1039:00
1045:00
1051:00
1057:00
1103:00
1109:00
1227:00
1233:00
1239:00
1245:00
1251:00
1257:00
1303:00
1309:00
1315:00
1321:00
1335:00
1341:00
1347:00
1353:00
1359:00
1405:00
1411:00
1417:00
1421:00
1427:00
1020:45
1026:45
1032:45
1038:45
1044:45
1050:45
1056:45
1102:45
1108:45
1114:45
1232:45
1238:45
1244:45
1250:45
1256:45
1302:45
1308:45
1314:45
1320:45
1326:45
1340:45
1346:45
1352:45
1358:45
1404:45
1410:45
1416:45
1420:45
1426:45
1432:45
1.5
1.9
1.5
0.0
2.1
0.2
0.8
1.0
0.0
1.0
0.0
0.0
0.0
2.5
0.8
1.0
0.8
1.5
0.2
0.2
0.0
0.0
0.2
0.0
0.6
0.0
0.6
0.4
0.2
2.1
Observer
Location
East-southeast (150 ft)
from stack
Comments; This was
a brown plume when
visible, and against
a blue sky.
Aver age
0.7
2-23
-------
to
1
to
.p-
rn
z
a
z
m
rn
2
z
G)
l
to
O
m
z
o
m
25
20
i
r^
•z.
LU
CJ
LU
a.
gl5
o
<^
a.
o
LU
C£.
LU
$10
5
I i i i l i i i
—
—
_ _
~
i— t r~~i
n n
, M L-TTl n 1 L-^TI— 1 _ rt-, n-i n-u_ l , ,
1000 11 00 1200 1300 1400 1500 1600 1700
VISIBLE EMISSIONS OBSERVATIONS AT THE CHARGING WELL'.AFTERBURNER OUTLET
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA - MAY 28, 1981
8
I
.p-
-------
TABLE 2-7
FUGITIVE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHARGING WELL AT
VISTA METALS CORPORATION, MAY 28, 1981
Clock
Time
(20 min.
intervals)
1010-1030
1045-1105
1110-1130
1225-1255
1255-1315
Duration
(min:sec)
00:32
01:57
00:10
00:16
01:53
00:03
00:52
03:51
01:12
00:39
00:15
00:34
01:45
01:30
00:23
04:29
00:31
00:01
00:42
01:39
01:21
00:31
00:07
01:22
00:16
17:15
00:09
00:06
00:47
01:01
00:26
00:15
01:00
01:17
00:55
00:08
01:31
Accumulated Emission
Emission Time Frequency
(min:sec) (%)
00:32
02:29
02:39
02:55
04:48
04:51
05:43
09:34
10:46
11 : 25
11:40
12:14
13:59
15:29 76.5
00:23
04:52
05:23 26.2
00:01
00:43
02:08
03:29
04:00
04:07
05:29
00:16
17:31 86.6
00:09
00:15
01:02
02:03
02:29
02:44
02:44
05:01
05:56
06:04
07:35
2-25
-------
TABLE 2-7 continued
FUGITIVE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHARGING WELL AT
VISTA METALS CORPORATION, MAY 28, 1981
Clock
Time
(20 min.
intervals)
1255-1315 (cont1
1315-1335
1335-1355
1355-1415
1415-1435
Duration
(min: sec)
d.) 00:02
00:07
00:04
00:03
00:29
00:12
01:01
04:12
00:39
00:02
00:12
01:15
00:07
00:07
00:02
00:04
00:12
00:50
00:50
00:01
00:07
00:03
00:03
02:37
01:01
00:16
00:06
00:26
00:44
00:54
00:37
00:06
00:12
00:08
00:07
00:09
Accumulated
Emission Time
(min:sec)
07:37
07:44
07:48
07:51
08:20
08:32
01:01
05:13
05:52
00:02
00:14
01:29
01:36
01:43
01:45
01:49
02:01
02:51
00:50
00:51
00:58
01:01
01:04
02:41
03:42
03:58
04:04
04:30
00:44
01:38
02:15
02:21
02:33
02:41
02:48
02:57
Emission
Frequency
(%)
41.6
27.6
12.6
21.5
2-26
-------
TABLE 2-7 continued
FUGITIVE EMISSIONS OBSERVATIONS AT THE
REVERBERATORY FURNACE CHARGING WELL AT
VISTA METALS CORPORATION, MAY 28, 1981
Clock
Time
(20 min.
intervals)
Duration
(mintsec)
Accumulated
Emission Time
(min;sec)
Emission
Frequency
1415-1435 (Cont'd.)
00:33
00:47
00:06
00:17
00:26
00:26
00:07
01:25
00:21
00:42
00:07
00:12
00:32
03:30
04:17
04:23
04:40
05:06
05:32
05:39
07:04
07:25
08:07
08:14
08:26
08:58
42.9
2-27
-------
TABLE 2-8
FUGITIVE EMISSIONS OBSERVATIONS IN THE
BORINGS DRYER CHARGING AREA AT
VISTA METALS CORPORATION, MAY 22, 1981
Clock
Time
(20 min.
intervals)
Duration
(minisec)
Accumulated
Emission Time
(min:sec)
Emission
Frequency
1049-1110
1340-1400
13:15
13:50
01:10
13:15
13:50
15:00
63.1
75.0
TABLE 2-9
FUGITIVE EMISSIONS OBSERVATIONS IN THE
BORINGS DRYER CENTRAL AREA AT
VISTA METALS CORPORATION, MAY 22, 1981
Clock
Time
(20 min.
intervals)
1049-1110
1340-1400
Duration
(min:sec)
21:00
19:15
Accumulated
Emission Time
(min:sec)
21:00
19:15
Emission
Frequency
100
96.3
TABLE 2-10
FUGITIVE EMISSIONS OBSERVATIONS IN THE
BORINGS DRYER UNLOADING AREA
VISTA METALS CORPORATION, MAY 22, 1981
Clock
Time
(20 min.
intervals)
Duration
(minjsec)
Accumulated
Emission Time
(min;sec)
Emission
Frequency
1049-1110
1340-1400
21:00
20:00
21:00
20:00
100
100
2-28
-------
2.1.9 Pressure Drop Determinations across Reverberatory Furaace
Chlorination Scrubber
Pressure drop across the scrubber system was monitored during the
three test runs on the inlet and outlet of that unit. The pressure
drop measurements are summarized in Table 2-11. Average pressure
drops for each of the three evenings were 2.9, 2.8, and 2.9 inches of
water for May 19, 20, and 21 respectively.
2.1.10 Stack Gas Molecular Weight Determinations
Stack gas molecular weight determinations were made based on Orsat
analyses and moisture determinations summarized in sections 2.1 and 2.4,
As mentioned previously, the chlorine gas at the scrubber inlet was
absorbed in the oxygen burrett of the Orsat analyzer. The chlorine
absorption resulted in apparent oxygen concentrations as high as 21.3%.
Molecular weight determinations were made using 20.9 as the assumed
oxygen concentration at the scrubber inlet and outlet sites. Molecular
weight determinations are presented in Tables 2.2 and 2.3 and included
in computer summaries in Appendix A for scrubber inlet and outlet
sites.
Molecular weight determinations were made at the borings dryer
outlet with the afterburner not operating. The molecular weight (wet)
was determined to be 28.38. The uncontrolled dry gas molecular weight
was 29.41. No moisture determination was made with the borings dryer
afterburner operating, therefore no wet molecular weight was calculated,
The dry gas molecular weight was 30.08 with the afterburner operating.
2-29
-------
TABLE 2-11
PRESSURE DROP ACROSS CHLORINATION SCRUBBER
AND TEMPERATURE AND pH OF SCRUBBER LIQUOR
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Particulate/Chlorine/ Test
Chloride Run Number Date
M5/C1 - 1 5/19/81
M5/C1 - 2 5/20/81
M5/C1 - 3 5/21/81
Time
1755
1855
1907
1927
1937
1947
2007
2015
2035
2105
Average
1740
1746
1812
1830
1850
1855
1910
1935
2015
2045
2115
2120
2145
Average
1905
1910
1933
1935
2010
2026
2045
2110
2140
Average
Scrubber
Ap (in. H?0)
2.0
3.1
2.9
2.9
3.2
3.5
2.7
2.5
2.9
2.7
3.1
2.9
2.9
2.8
2.8
2.6
2.8
2.6
2.8
2.7
2.8
2.9
3.2
3.1
2.9
Liquor
Temp.(F°)
68
70
66
64
62
66
58
64
67
76
80
86
85
74
76
78
83
76
70
77
Liquor
PH
10.5
10.5
10.0
10.5
10.5
10.4
11
11
11
11
11
11
11
11
11.5
11.5
11.0
11.5
11.0
11.3
2-30
-------
2.2 Reverberatory Furnace Charging Well Emissions
Visible emissions observations (VEO's) were made on the reverberatory
furnace charging well stack on May 28, 1981; however, no record of pro-
cess operations were made during that observation period. The results
are presented in Section 2.6. No other tests were conducted on stack
emissions from this source.
2.3 Reverberatory Furnace Combustion Stack Emissions
No testing was conducted on the reverberatory furnace combustion
stack during this test program.
2.4 Borings Dryer Emissions
The borings dryer exhaust stack was found to have extremely .low
flow velocities, most pronounced during uncontrolled operation, and
only marginally within the usable range of available instrumentation.
With the afterburner in operation the flow velocities increased bul: the
stack gas temperature was high, exceeding 2000°F. Due to these
conditions, no comprehensive testing was conducted on either controlled
or uncontrolled borings dryer emissions. Some preliminary tests were
conducted and are discussed below in Sections 2.4.1 and 2.4.2; however,
no data summaries are presented for these tests.
2.4.1 Borings Dryer Uncontrolled
A preliminary velocity traverse was conducted on the borings dryer
with the afterburner not operating on May 22, 1981. The velocity head
ranged from approximately 0.01 to 0.002 inches of water, well below
the usable range of the micromanometer. An attempt was made to collect
a particulate sample by EPA Reference Method 5 but the filter became
2-31
-------
plugged after 1 minute of sampling. The stack temperature was 426°F,
and the moisture, determined from the impinger volume change and silica
gel, was 9.06%. The particulate concentration was calculated to be
3.354 grains/DSCF (26.103 pounds/hour). The condensible hydrocarbon
concentration was determined to be 0.589 grains/DSCF (4.580 pounds/hour).
The stack gas flow rate was calculated to be 907 DSCFM. The average
oxygen content was 14.8%, carbon dioxide, 5.0%, and carbon monoxide,
0.7%. It. is stressed that these data were not collected under acceptable
test conditions and are therefore presented here for information pur-
poses only in anticipation that they may be useful for any future
testing of this or a similar unit. Field data sheets presenting the
borings dryer testing can be found in Appendix C.2.
Noncondensible hydrocarbon testing was also conducted on uncontrolled
emissions from the borings dryer. Noncondensible hydrocarbons ranged
from 425 to 660 ppmV as hexane. These data are presented in Section 2.13.
2.4.1.1 Particle Size
A single particle size distribution test was conducted on the
borings dryer uncontrolled emissions on May 22, 1981. No attempt was
made to sample the stream isokinetically due to the extremely low gas
flows in the stack and the resulting low gas velocities in the impactor.
Extremely low flows through a cascade impactor result in unpredictable
sizing characteristics by the jets. Table 2-12 and Figures 2.5 and
2-6 illustrate the results of this test. The grain loading determined
during this run was 3.356 grains/DSCF.
2-32
-------
TABLE 2-12
PARTICULE SIZE RESULTS3 OF UNCONTROLLED BORING
DRYER AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA
N>
I
10
U)
Test Date (1) Sampling Duration Stage
Time, and (2)Impactor Flow Rate Index
Run No. No.
5/22/81 (1) 0.8 minutes
1316:10-1317 (2) 1.539 ACFM
DU-PS-1 (3) 426 .0°F
S0b
SI
S2
S3
S4
S5
Back-up
filter
Delta Effective % In
Weight Cut Diameter Size
(mg) (DPSO-microns) Range
79.40
3.09
4.80
2.67
21.46
24.81
8.42
8.69
5.45
2.54
1.65
0.85
0.40
—
54.89
2.14
3.32
1.84
14.84
17.15
5.82
Cumulative
%less than
Size Range
45.11
42.97
39.65
37.81
22.97
5.82
—
Size
Range
(microns)
>8.69
5.45-8.69
2.54-5.45
1.65-2.54
0.85-1.65
0.40-0.85
0.0C-0.40
a. Particle sizing determinations employed at-stage Anderson Mark III impactor.
b. Nozzle, pre-cutter, inlet cone, and zero stage wash weight added to Stage No. 1 weight (index No. SO)
c. Back-up filter has an actual 0.3 micron retention.
-------
FIGURE 2.5_
loo
9O
M
SO
3O
4O
10
zo
10
s •
O T
w
0 *
I 3
m
H- 4
Z
5 ,
flu
1-
3
O *
O
IO
0
1
.1
.7
.8
.3
.4
.3
.2
I
Particle Size Results
Andersen 6-Stage Mark III Impactor
Borings Dryer Uncontrolled
Vista Metals Corporation Fontana, CA
Impactor Flowrate = 1.539
}
i
i
1
\ \
<
I
i
J
i
i
i
i j
i [
1
i
i
i
I
t
1 ;
I
1
J j
i
|
\zs
i
i
j | i
i
i
i
1
1
(•)
@
i (<
(•}
a
[
\
'
i
1
i
-
RUN
DU-PS-1
i
i
i i
j
i i
-
i
!
I
i
i
'
i
i
!
i
i !
1
3.01 0.05 0.1 O.J 15 L 2 3 10 M 3) *0 30 50 '0 30 50 » 98 99 99.3 999 39.99
CUMULATIVE % LESS THAN STATED SIZE
2-34
ENGJNEEHING-SCJENCE
-------
FIGURE 2.6
Mass Loading dM/d Log D (mg/DSCM)
h-*
t-' O
. i- O O O O O
Particle Size Results
Andersen 6-Stage Mark III Impactor
Borings Dryer Uncontrolled
Vista Metals Corporation, Fontana, CA
RUN DU-PS-1
Impactor Flowrate = 1.539
(a
£
"i
IT,
im
(i
— j—
i
_
ij
1
j
!
1
1
— i —
i
j
i
1
L 1 10 100
Geometric Mean of Particle Diameter (microns)
2-35
ENGINEERING-SCIENCE
-------
2.4.1.2 Flame lonization Detector Results - Uncontrolled Non-
condensible Hydrocarbon Emissions from the Borings Dryer
Table 2-13 summarizes noncondensible hydrocarbon measurements
made on emissions of the uncontrolled borings dryer. Hydrocarbon
concentrations are reported as hexane and ranged from a low of 400 to
685 ppmV during normal operation of the dryer. Emission rates were
determined based on very rough estimates of the stack gas flow rate.
2.4.2 Borings Dryer Controlled
Controlled emissions from the borings dryer were tested for non-
condensible hydrocarbons and Orsat analysis. Noncondensible hydrocarbon
concentrations ranged from 2.4 to 3.7 ppmV as hexane. Oxygen content
averaged 2.0%; carbon dioxide averaged 8.5% and carbon monoxide averaged
0.5%. Noncondensible hydrocarbons concentrations are presented in
Table 2.14 and the Orsat results in Appendix C.2.2.
2.5 Audit Sample Results
The results of analyses of audit samples provided by EPA and
analyzed by ES prior to analysis of Vista Metals Corporation test
samples are summarized in Table 2-15. The audit samples were analyzed
by EPA in September, 1980. During method development work performed by
ES in December 1980, several of the audit samples were analyzed for
chlorine using Methods 409-D and 409-E (Standard Methods for the
Examination of Water and Wastewater, Fourteenth Edition). The results
of those analyses averaged 26% below the reported EPA values. Since
chlorine is unstable, sample degradation is believed to be the cause
for the discrepancy. The December results were used for comparison
2-36
-------
TABLE 2-13
FLAME IONIZATION DETECTOR (FID) DATA SUMMARY ON
UNCONTROLLED GASES AT THE BORINGS DRYER
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
NJ
LO
Gaseous Hydrocarbon
Run Traverse Time
Date No. Points Start
5-22-81 1 Stack Center 1234
Point 1235
1236
1237
1238
1240
1241
1242
1243
1244
1245
1246
1247a
1248s
I249a
I250a
I251a
I252a
I253a
I254a
I255a
(min.)
End
1235
1236
1237
1238
1239
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
Minimum**
ppm (V)
400
445
490
490
510
579
630
600
578
554
500
460
425
386
342
305
280
255
240
220
202
Maximum**
ppm (V)
470
489
512
532
619
675
685
645
620
600
550
510
465
430
385
342
310
280
214
240
218
Concentration
Point
ppm (V)
440
473
495
510
560
635
652
610
598
571
528
482
445
408
365
326
295
270
250
230
211
Average
gr/DSCFc
0.699
0.751
0.786
0.810
0.889
1.008
1.035
0.978
0.949
0.906
0.828
0.765
0.706
0.648
0.579
0.518
0.468
0.429
0.397
0.365
0.335
Volumetric
Flow
DSCFM
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
907.6
Pollutant
Mass Ratec
Ibs/hr
5.439
5.844
6.116
6.303
6.918
7.844
8.054
7.611
7.385
7.050
6.521
5.953
5.494
5.043
4.506
4.031
3.642
3.338
3.090
2.840
2.607
a. THCA operator was informed that the borings dryer charging conveyor was down. Test aborted at 1256.
Data not included in emission calculations summarized in text.
b. As hexane.
c. A hexane molecular weight of 86 was used to calculate hydrocarbon concentrations and mass flow rate.
-------
TABLE 2-14
FLAME IONIZATION DETECTOR (FID) DATA SUMMARY ON
CONTROLLED GASES EXITING THE BORINGS DRYER AFTERBURNER
AT THE VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Gaseous Hydrocarbon Concentration
Run
Date No .
5-22-81 1
a. As hexane
b . A hexane
Traverse Time
Points Start
Stack Center 1124
Point 1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
•
molecular weight of 86
(min.)
End
1125
1126
11,27
1128
1129
1130
1131
1132
1133
1134
1135
1136
Minimum Maximum Point Average
ppm
2.5
Chart
2.4
2.4
2.6
2.9
3.0
3.1
3.5
3.7
3.8
3.6
was used to
(V)a ppm (V)a ppm (V)a gr/DSCFb
2.8 2.5 0.004
spike-no data reduction attempted
2.4 2.4 0.004
2.6 2.5 0.004
2.9 2.0 0.004
3.0 3.0 0.005
3.1 3.1 0.005
3.5 3.2 0.005
3.8 3.7 0.006
3.8 3.7 0.006
3.7 3.7 0.006
3.7 3.6 0.006
calculate hydrocarbon concentrations.
2-38
-------
TABLE 2-15
VISTA METALS AUDIT SAMPLE RESULTS
Audit
Sample
Number
4075
5255
3095
1014
1023
2268
2289
3076
3096
4068
4078
5263
5280
2235
1230
3015
4016
5241
EPA Results
September
1980 (mg/1)
357.6
458.6
254.8
50.96
50.96
152.9
152.9
254.8
254.8
356.7
356.7
458.6
458.6
3000.0
1000.0
5000 .0
7000.0
9000.0
ES Results
December
1980 (mg 1)
268.0
354.0
183.0
38.0
38.0
108.0
Component
Analyzera
C12
C12
C12
C12
C12
ci2
ci2
C12
C12
C12
C12
C12
C12
Tot Cl
Tot Cl
Tot Cl
Tot Cl
Tot Cl
Method1*
A
A
D
D
D
D
D
D
D
D
D
D
M
M
M
M
M
ES Results
May
1981 mg/1
224
414
Sample Spilled
29
35
150d
102
175
180
250
255
320
320
3149.0
949.7
4841.0
6797.9
8697.3
%
Error0
-16.4
+ 16.9
-23.7
-7.9
+38. 9d
-33.3
-31.3
-29.4
-29.9
-28.5
-30.2
-30.2
+5.0
-5.0
-3.2
-2.9
-3.4
a. C12 = combined chlorine; Tot Cl = Total Chlorides
b. A = arsenite; D = DPD; M = Mecuric Nitrate
c. % Error = ES Results - EPA Results x 100
EPA Results
d. The first titration of this sample required slightly more than the recommended
amount titrant, but yielded 93 mg/1 and a -13.9% error. The figure listed
resulted from a second titration of a smaller aliquot. The first titration
results, however, appear more reasonable.
Note: ES results of December 1980 used to determine percent error when those
analyses were available. Sample degradation is cited as the possible reason.
reason for the consistent negative error.
2-39
-------
with the Vista audit results determined in May 1981 when analyses had
been conducted.
Arsenite titration of two chlorine audit samples resulted in
relative errors of +17 and -16 percent. These errors were.determined
by comparison with ES December analyses. Audit sample concentrations
below the accuracy of the method may have caused these errors, parti-
cularly the positive error. In this particular sample, the difference
in the amount of titrant used between the blank and the sample was
only 3%. Some chlorine sample degradation would be expected between
December and May.
Chlorine results of three samples analyzed by the DPD method and
compared to ES December results indicated relative errors from -24 to
+39%. Another titration of the one positive error sample showed
a negative error. The negative error appears more likely. Seven
additional DPD titrations for chlorine were compared to EPA September
1980 results and showed -28.5 to -33.3% error. The error in chlorine
audit analyses is probably the result of the C10~ ion reducing to
Cl~, or possibly escaping out of solution.
2.6 Cleanup Evaluation Results
Cleanup evaluation results are presented in Table 2-16. The
scrubber inlet and outlet trains were charged with reagents prior to
the first test run and these blanks recovered according to normal sample
recovery procedures.
2-40
-------
TABLE 2-16
CLEANUP EVALUATION RESULTS
AT VISTA METALS CORPORATION, FONTANA, CALIFORNIA
Sample Description
Inlet train - front-half rinse
prior to 1st test run
Inlet train - back-half
recovery prior to 1st test run
Outlet train - front-half rinse
prior to 1st test run
Outlet train - back-half 1st
impinger prior to first test run
Outlet train - 2nd and 3rd
impinger prior to 1st test run
Inlet train - front-half rinse
after Run 1
Inlet train - impinger rinse
after Run 2
Filter blank
Distilled water blank
Chlorine
(mg)
N/A
0.0
N/A
0.0
0.0
N/A
0.0
Not analyzed
Not analyzed
Chloride
(mg)
0.0
0.0
0.0
0.0
0.0
16.99
Not analyzed
0.0
Not analyzed
Parti-
culate
(mg)
1.2
N/A
1.7
N/A
N/A
Not
Analyzed
Not
Analyzed
Not
Analyzed
4.9/1
rinse solvent and diluent
for all scrubber runs)
Acetone blank residue
(DU-M5/4-1)
Methylene chloride blank
residue (DU-M5/4-1)
N/A
N/A
N/A
N/A
1.5/1
0.0
N/A = Not applicable.
2-41
-------
The inlet train front half runs showed no chloride, and showed a
particulate residue of 1.2 mg. The inlet train back half rinse showed
no chlorine or chloride. The procedure did not require front half
chlorine analysis or back half particulate residue. The outlet train
front half cleanup evaluation showed no chloride, and showed 1.7 mg
particulate residue. The outlet train back half anlaysis, both the
first impinger and the combined second and third impinger, showed no
chlorine or chloride. The analyst on the chlorine analysis, because
the chlorine values were zero on the outlet, did not understand the
necessity of recording the zero results on lab data sheets, and there-
fore failed to do so. An additional water rinse of the front half of
the inlet train was done after sample recovery of Run 1 to evaluate the
efficiency of cleanup procedures. Similarly, an additional rinse of
the inlet train impingers was made following Run 2. The latter evaluation
resulted in no residual chlorine in the impingers, but the front half
rinse recovered 17 milligrams of chloride. A filter blank analysis
showed no chloride.
Blank values were taken of the distilled water used for cleanup
and dilution and showed 4.9 mg/1 of residue. Distilled water was not
analyzed directly for chlorine or chloride as this was essentiall
accomplished in the cleanup evaluation sample recovery analysis.
Blank residue values also were taken of acetone and raethylene
chloride. The acetone showed 1.5 mg/1 residue, but none was evident
from the methylene chloride.
2-42
-------
Blank chlorine values of the sodium arsenite solution used in the
chlorination scrubber inlet train were taken before each titration and
blank chloride values were taken during chloride analysis. Blank
potassium hydroxide chlorine and chloride values were also taken during
analysis.
2.13.2 Controlled Noncondensible Hydrocarbon Emissions from the
Borings Dryer
Table 2-16 summarizes noncondensible hydrocarbon concentrations
measured with the borings dryer afterburner operating. Hydrocarbon
concentrations as hexane ranged from 2.4 to 3.8 ppmV.
2-43
-------
SECTION 3
PROCESS DESCRIPTION AND OPERATION
-------
3.0 PROCESS OPERATIONS
3.1 General Process Operations
The Vista Metals facility was constructed in 1968. The plant
ioperates 24 hours per day, 5.5 days per week, 50 weeks per year. The
plant has a maximum production capacity of 54.4 gigagrams (60,000 tons)
of aluminum product per year. Actual production of aluminum product is
approximately 33.7 gigagrams (37,000 tons) per year. The amount of
aluminum scrap processed by the facility is 37.4 gigagrams (41,300 tons)
per year. The plant operates a borings (chip) dryer, a sweat furnace,
and six reverberatory furnaces.
The borings dryer processes 315 grams per second (2,500 pounds per
hour) of aluminum and operates 18 hours per day, 6 days per week. The
dryer operates at a temperature of 477.4 K (400°F). The dryer processes
borings which may have up to 20 percent by weight of oil. The feed is
controlled depending on the percentage of oil in the borings. The dried
borings are passed into a magnetic separator to remove ferrous material.
Emissions from the borings dryer are controlled by an afterburner.
The sweat furnace is used to separate aluminum from scrap metal
containing significant quantities of iron. The sweat furnace can
process a maximum of 252 grams per second (2,000 pounds per hour) of
aluminum scrap. The furnace operates 16 hours per day, 4 days per week,
40 weeks per year. The furnace operates at a temperature of 1,088 K
(1,500°F) and processes scrap which has between 50 to 90 percent aluminum
by weight. Emissions from the sweat furnace are controlled by an
afterburner.
The plant has 6 reverberatory furnaces which range in capacity from
27.2 to 40.8 megagrams (60,000 to 90,000 pounds).
Four 40.8 megagram (90,000 pound) furnaces are used to produce
aluminum billets. These furnaces process clean aluminum scrap and
3-1
-------
supply molten aluminum to level pour direct chill billet casting
machines. No air pollution control equipment is utilized and emissions
are vented directly to the atmosphere.
The Vista Metals plant has two reverberatory furnaces that are used
to process scrap associated with the secondary aluminum smelting operation.
The two furnaces have capacities of 27.2 megagrams and 31.7 megagrams
(60,000 pounds and 70,000 pounds). Both furnaces produce aluminum ingots.
These furnaces have a 24-hour heat cycle which consists of 16 hours of
charging, 4 hours of demagging and 4 hours of tapping. Only the
31.7 megagram (70,000 pound) furnace was in operation during the week
of source testing.
3.2 Reverberatory Furnace Description
There are three sections to each furnace: the charging well, the
combustion chamber, and the chlorination chamber. A diagram of the
reverberatory furnace is provided in Figure 3.1. No control exists for
air emissions from the furnace combustion chamber because emissions
apparently consist only of products of combustion. Emissions from the
charging well are controlled by the use of an afterburner. The after-
burners operate at a temperature of approximately 1,199 K (1,700°F).
The purpose of the source test was to evaluate emissions produced during
the demagging operation. Demagging is conducted in the furnace chlorination
chambers.
The purpose of the demagging process is to reduce the magnesium
content of the molten aluminum. During chlorine demagging operations,
chlorine is injected into the melt and reacts with magnesium to form
magnesium chloride:
Mg + C12 + MgCl2
Magnesium chloride is a liquid at the molten metal temperature and can
be skimmed off after demagging is completed.
The reverberatory furnace chlorination chamber at Vista is
approximately 1.2 meters (4 feet) wide and 3.03 meters (10 feet) long,
and is located to the rear of the furnace. An archway beneath the
molten metal level in the common wall between the furnace and the
chamber, permits the flow of metal.
3-2
-------
SIDE VIEW OF REVERBERATORY FURNACE
To
Wet
Scrubber
To
Afterburner
To
Atmosphere
CO
I
OJ
Level of
Molten Aluminum
Chlorination
Chamber
Combustion
Chamber
Charging
Well
m
GO
-------
During demagging, chlorine gas is sent under pressure through a
porcelain-coated iron tube and is bubbled up through the molten aluminum.
The end of the tube is placed approximately 0.15 meters (6 inches) from
the bottom of the chamber.
There are approximately 8 alloys that are routinely produced at the
plant. The allowable magnesium concentration varies based on the type
of alloy being produced.
During demagging, chlorine is added so rapidly that large quantities
of both aluminum chloride and magnesium chloride are formed and not all
of the chlorine reacts with the metals. As a result, a large quantity
of aluminum chloride is discharged along with some chlorine gas and some
entrained magnesium chloride. Aluminum chloride sublimes at 454 K
(357°F), so that it is vaporous at the temperature of molten aluminum.
As the vapors cool in the atmosphere, submicron fumes are formed.
Aluminum chloride is extremely hygroscopic and absorbs moisture from
the air, with which it reacts to form hydrogen chloride.
3.3 Emission Control Equipment
Each of the furnace chlorination chambers at Vista are controlled
by separate wet scrubbers. A diagram of the scrubber tested is provided
in Figure 3.2. The scrubbers are packed tower units that were designed
by plant personnel. The principle of design is that the contaminant-
laden stream is passed through beds of a fiberglass collection material,
and a liquid is passed over the collecting surface to keep it clean
and prevent reentrainment of deposited materials. Collection of the
contaminant depends upon the length of contact time of the gas stream
on the collecting surfaces.
A settling chamber is located prior to each of the wet scrubbers.
The settling chamber is necessary because of the high loadings of parti -
culate matter produced during the demagging operations. According to
plant personnel the settling chambers are cleaned out once every two
weeks.
An air-bleed-in port was located downstream of the settling chamber;
prior to the wet scrubber. This port enabled plant workers to observe
the density of emissions which would indicate the efficiency of the
demagging process.
3-4
-------
DIAGRAM OF WET SCRUBBER FOR CONTROLLING DEMAGGING EMISSIONS
Caustic
Wet
Scrubber
Scrubbing Liquid
Recycling Tank
\
Air-bleed-in
o
Outlet Sampling Port
Fan
\
run.
O
o
Inlet
Sampling
nn«4-
f\JI O
Settling
Chamber
i
Chlorination
Chamber
Exhaust
-------
Because of the acidic nature of the demagging emissions, both
scrubbers at Vista use caustic scrubbing solutions. Scrubbing solutions
average between 5 percent to 10 percent caustic (sodium hydroxide). The
caustic scrubbing liquids are recycled and pH monitored to insure proper
alkalinity is maintained.
3.4 Process Qperatibns During Testing
The demagging tests were run between Sunday, May 17 through Thursday,
May 21, 1981. The test on May 17 consisted only of a velocity traverse
on the scrubber settling chamber inlet and there was no requirement for
the process operation to be monitored. During the week of testing,
3 sets of simultaneous inlet and outlet test runs were conducted on the
chlorination scrubber. Six particle size runs were also performed on
the scrubber inlet. Visible emission observations were made at the
scrubber outlet during the chlorination periods.
During each heat cycle, approximately 27.9 megagrams (62,000 pounds)
of aluminum ingot was produced. This figure assumes that 3.6 megagrams
(8,000 pounds) of "heel" remained in the furnace after each tapping was
completed.
All scrap charged during the week of testing had been pretreated
in the borings dryer in order to remove the majority of organic
contamination.
The following is a description of the process operation during the
week of testing.
Monday. May 18. 1981
A S-14 alloy was being produced during this reverberatory furnace
heat cycle. Charging of the furnace was initiated at 5:00 a.m. and was
completed by 3:30 p.m. The type of aluminum scrap charged consisted
entirely of aluminum turnings and borings. Approximately 0.9 megagram
(1 ton) of flux material was added to the furnace during charging
operations.
The demagging operation was commenced at 5:35 p.m. The maximum
allowable magnesium concentration for the S-14 alloy was .40 percent.
5:35 p.m.
Demagging started. Initial magnesium concentration of aluminum
measured to be .95 percent. Initial pH of scrubbing solution
3-6
-------
measured to be 13. Line pressure of chlorine flow was
207 kilopascals (30 pounds per square inch).
6:00 p.m.
Magnesium concentration at .79 percent.
6:30 p.m.
Chlorine gas turned off and porcelain tube replaced.
6:35 p.m.
Chlorine back on.
7:15 p.m.
Some problem experienced in keeping pH levels of caustic up.
Additional sodium hydroxide added to scrubbing solution. Greenish
material was observed in the air-bleed-in port downstream of the
settling chamber. Supervisor speculated that this may be due to
the presence of unreacted chlorine in demagging exhaust.
7:50 p.m.
Magnesium concentration at .59 percent.
8:10 p.m.
Particle size probe inserted in inlet stack. Line pressure of
chlorine flow at 207 kilopascals (30 pounds per square inch).
8:13 p.m.
Particle size probe removed.
8:15 p.m.
Chlorine gas turned off and porcelain tube replaced. Two holes
found in old porcelain tube.
8:20 p.m.
Chlorine back on.
8:25 p.m.
Magnesium concentration at .59 percent. Green emissions still
observed in air-bleed-in port.
8:42 p.m.
Second particle size probe inserted.
8:45 p.m.
Particle size probe removed.
3-7
-------
8:47 p.m.
Inlet emissions still greenish in appearance. Supervisor decided
to install new tank of chlorine in case tank in use was contaminated.
Chlorine turned off and tank replaced.
9:00 p.m.
Emissions observed in port appear white.
10:00 p.m.
Demagging ended. Final magnesium concentration at .40 percent.
Because tanks were changed during demagging, chlorine feed rate
could not be determined. Furnace temperature during demagging
operation was 1033 K (1400°F).
Tuesday. May 19. 1981
A 380 alloy was being produced during the reverberatory furnace
heat cycle. Charging of the furnace was initiated at 3:00 a.m. and was
completed at approximately 4:00 p.m. The type of aluminum scrap processed
during this period consisted of aluminum borings and turnings.
The demagging operation was started at 5:35 p.m. The maximum
allowable magnesium concentration for the 380 alloy was .30 percent.
5:35 p.m.
Demagging started. Initial magnesium concentration of aluminum
measured to be .76 percent. Initial pH of scrubbing solution
measured to be 13. Line pressure of chlorine flow was 276
kilopascals (40 pounds per square inch).
6:20 p.m.
Chlorine runs out. New tank installed. Chlorine used in old tank
was 100 kilograms (220 pounds).
6:30 p.m.
Chlorine turned back on. Line pressure 276 kilopascals (40 pounds
per square inch).
6:35 p.m.
Magnesium concentration at .70 percent.
7:00 p.m.
Testing started at scrubber inlet and outlet.
7:15 p.m.
Magnesium concentration at .61 percent.
3-8
-------
7:20 p.m.
Chlorine gas turned off and porcelain tube replaced.
7:23 p.m.
Chlorine gas back on. Line pressure of chlorine flow at 276
kilopascals (40 pounds per square inch).
7:55 p.m.
Magnesium concentration at .55 percent.
8:20 p.m.
Magnesium concentration at .46 percent.
8:50 p.m.
Magnesium concentration at .38 percent.
9:05 p.m.
Test stopped.
9:05 p.m.
Particle size probe inserted.
9:12 p.m.
Particle size probe removed.
9:13 p.m.
Demagging ended. Final magnesium concentration at .29 percent.
Amount of chlorine used in second tank was 549 kilograms (1210
pounds). Total chlorine used during demagging was 649 kilograms
(1430 pounds). This is equivalent to a process rate of 48 grams
per second (381 pounds per hour).
Chlorine pressure remained constant at 276 kilopascals (40 pounds;
per square inch). Furnace temperature during demagging operation
was 1033 K (1400°F).
Wednesday. May 20. 1981
A 380 alloy was being produced during the reverberatory furnace
heat cycle. Charging of the furnace was initiated at approximately
5:00 a.m. and completed at 4:00 p.m. The type of aluminum scrap
processed during this period consisted of aluminum bor.ings and turnings.
The demagging operation was started at 5:30 p.m. The maximum
allowable magnesium concentration for the 380 alloy was .30 percent.
3-9
-------
5:30 p.m.
Demagging started. Initial magnesium concentration measured to be
.90 percent. Line pressure of chlorine flow was 276 kilopascals
(40 pounds per square inch). Initial pH of scrubbing solution
measured to be 13.
6:10 p.m.
Particle size probe inserted.
6:20 p.m.
Particle size probe removed. Chlorine runs out. New tank installed.
6:30 p.m.
Chlorine turned back on. Line pressure at 276 kilopascals (40 pounds
per square inch).
6:35 p.m.
Testing started at scrubber inlet and outlet.
6:40 p.m.
Magnesium concentration at .80 percent.
7:25 p.m.
Magnesium concentration at .72 percent.
7:40 p.m.
Batch of copper radiators added to the charging well.
7:50 p.m.
Chlorine gas turned off and porcelain tube replaced.
7:53 p.m.
Chlorine gas back on. Line pressure of chlorine flow was 276
kilopascals (40 pounds per square inch).
8:05 p.m.
Magnesium concentration at .67 percent.
8:20 p.m.
Test stopped.
8:25 p.m.
Particle size probe inserted.
8:35 p.m.
Magnesium concentration at .62 percent.
8:35 p.m.
Particle size probe removed.
3-10
-------
8:45 p.m.
Second test run started at scrubber inlet and outlet.
9:10 p.m.
Magnesium concentration at .59 percent.
9:27 p.m.
Chlorine gas turned off and porcelain tube replaced.
9:34 p.m.
Chlorine gas turned back on. Line pressure at 276 kilopascals
(40 pounds per square inch).
9:35 p.m.
Test stopped.
9:40 p.m.
Magnesium concentration at .54 percent.
11:05 p.m.
Demagging ended. Final magnesium concentration at .30 percent.
Amount of chlorine used in first tank could not be accurately
determined. Amount of chlorine used in second tank was 886
kilograms (1970 pounds). Based on the time period this tank was
in use, the chlorine process weight would be equivalent to 55 grams
per second (437 pounds per hour).
Chlorine pressure remained constant at 276 kilopascals (40 pounds;
per square inch) throughout demagging operation. Furnace temperature
remained constant at 1033 K (1400°F).
During this demagging period, the magnesium concentrations were
dropping off slowly. This would indicate the emissions would
increase. This contention is supported by the fact that more
caustic than usual had to be added to the scrubber during the
demagging period.
Thursday. May 21. 1981
A A108Z alloy was being produced during the reverberatory furnace
heat cycle. Charging of the furnace was started at approximately
5:00 a.m. and completed at 4:00 p.m. The type of aluminum scrap
processed during this period consisted of aluminum borings and turnings.
3-11
-------
Demagging was started at 5:50 p.m. The maximum allowable magnesium
concentration for the A108Z alloy was .10 percent.
5:50 p.m.
Demagging started. Initial magnesium concentration at .74 percent.
Line pressure of chlorine flow was 276 kilopascals (40 pounds per
square inch). Initial pH of scrubbing solution measured to be 13.
6:00 p.m.
Chlorine runs out. Amount of chlorine used in first tank was
4.5 kilograms (10 pounds).
6:05 p.m.
New tank installed.- Chlorine turned back on. Line pressure of
chlorine flow was 276 kilopascals (40 pounds per square inch).
6:50 p.m.
Magnesium concentration at .70 percent.
7:20 p.m.
Magnesium concentration at .65 percent.
7:30 p.m.
Source test run started at scrubber inlet and outlet.
7:50 p.m.
Magnesium concentration at .60 percent.
8:20 p.m.
First traverse completed at scrubber inlet.
8:20 p.m.
Magnesium concentration at .58 percent.
8:25 p.m.
Particle size probe inserted.
8:32 p.m.
Chlorine gas turned off and porcelain tube replaced.
8:35 p.m.
Particle size probe removed.
8:40 p.m.
Chlorine gas turned back on. Line pressure at 276 kilopascals
(40 pounds per square inch).
8:45 p.m.'
Second traverse started at scrubber inlet.
3-12
-------
9:10 p.m.
Batch of copper radiators added to charging well.
9:25 p.m.
Magnesium concentration at .37 percent.
9:40 p.m.
Test stopped.
9:55 p.m.
Magnesium concentration at .32 percent.
10:30 p.m.
Magnesium concentration at .27 percent.
11:00 p.m.
Magnesium concentration at .17 percent.
11:30 p.m.
Demagging ended. Final magnesium concentration at .10 percent.
Total amount of chlorine used during demagging was 895 kilograms
(1990 pounds). Chlorine process rate was equivalent to 44 grams
per second (352 pounds per hour).
Chlorine pressure remained constant at 276 kllopascals (40 pound:;
per square inch). Furnace temperature remained constant at 1033 K
(1400°F).
3.5 Conclusions
According to discussions with plant personnel, during the demagging
test runs, the chlorination process was operating within the range of
normal conditions. The test run done on Wednesday, May 20, 1981, was
conducted during conditions which would be representative of "worst-
case" emissions. This is because magnesium concentrations dropped
slowly resulting in a greater emission rate of molecular chlorine and
aluminum chloride. The presence of molecular chlorine may have con-
tributed to the "greenish" appearance of emissions as viewed in the
inlet port to the wet scrubber.
On several occasions during testing the chlorine flow was turned
off for short periods of time. This would be necessary when a porcelain
tube was being replaced or a new tank of chlorine installed. Porcelain
tubes were replaced when a hole would occur in the lance.
3-13
-------
The brief interruptions in chlorine flow never lasted more than a few
minutes and should not affect the results of the stack tests. It should
be noted that these interruptions are a normal part of the plant's
demagging process.
3-14
-------
\
SECTION 4
LOCATION OF SAMPLING POINTS
-------
4.0 LOCATION OF SAMPLING POINTS
The borings dryer and the Number 2 reverberatory furnace were
tested at the Vista Metals Corporation, Fontana, California, facility.
Figure 4-1 shows the plant layout with respect to these two processes.
4.1 Reverberatory Furnace Chlorination Process Control Equipment
The reverberatory furnace chlorination process is a sealed system.
The only gases vented from the system are apparently the result of
thermal expansion, including vaporization of metal, and displacement by
the introduction of chlorine gas at the rate of approximately 20 standard
cubic feet per minute. As a result, the gas flow from the process is
quite low and is estimated to be less than 1 CFM, since a significant
amount of the chlorine is bound during the demagging process.
The vented gas passes through a 12-inch diameter duct and into a
settling chamber where particulate matter falls out of the gas stream.
At the outlet of the settling chamber is a one-way flapper valve which
inhibits flow back into the chlorination system. Immediately downsstream
of the flapper valve is an opening in the side of the duct which allows
approximately 1300-1400 SCFm of ambient air to be drawn into the system
by the blower located at the base of the stack. The result is thai:
the exhaust gas stream is greatly diluted, and in effect is actually
drawn into the scrubber system by a hooding arrangement, rather than
an induced draft on the chlorination exhaust stream.
The gases then pass into the base of the packed bed scrubber and
flow upward through the countercurrent liquor flow in the scrubber.
The scrubbed gases pass through a horizontal, 12-inch diameter duct,
through the blower and out the stack. Figure 4-2 illustrates the
chlorination gas handling system.
4-1
-------
FIGURE 4.1
Overhead View of Plant Layout
Vista Metals Corporation, Fontana, California
SCRUBBER STACK
AFTERBURNER
STACK
CHLORINATION
SCRUBBER
i I
CHLORINATION
i CHAMBER J
i
-------
FIGURE 4.2
REVERBERATORY FURNACE NO. 2 CHLORINATION CHAMBER
AND SCRUBBER SYSTEM AT
VISTA METALS CORP. FONTANA CALIFORNIA
6'
BLOWER
10.75 ID
— 12"
ROOF LINE
19*
fe±
3" SAMPLING PORT
SCRUBBER
ONE WAY
FLAPPER VALVE \ **BIENT
AIR INTAKE
T
u
MOLTEN
ALUMINUM
_ CHLORINATION
CHAMBER
n........^i .
54"|
r>
3"SAMf
PORTS
\
REVERBERATORY
FURNACE 2
SETTLING
CHAMBER
4-3
ENGINEERING-SCIENCE
-------
4.1.1 Settling Chamber Inlet
Approximately 17 feet downstream of the last flow disturbance
(a 90° elbow) and approximately 48 inches upstream of the settling
chamber inlet, a single port was installed in the settling chamber
inlet duct. Due to the potential for explosion should air leak into
the chlorination chamber, Vista Metals installed a gate valve to allow
the port to be closed during testing. Due to the extremely low gas
flows ( Ap of approximately 0.015) and extremely high grain loadings
(the S-type pitot tube was plugged within 1 minute of insertion into
the gas stream), no testing was conducted at this site. This is explained
in more detail in Section 2.1.
4.1.2 Scrubber Inlet Test Site
The scrubber inlet test site was located in an 18.25 inch internal
diameter duct. The two access ports were located 54 inches downstream
of a 90° horizontal to vertical elbow and 9 inches upstream of a 90°
vertical to horizontal elbow. The test location is illustrated in Figure
4-3. One of the two sampling ports was a 3-inch diameter hole cut in
the duct. The second port was a 10-inch by 4-inch slot. Both ports
were closed to leakage using tape while testing was in progess. A
second slot and a condensation drain were sealed with tape to minimize
leakage into the system downstream of the test location. Sampling points
employed during the test program are indicated in Figure 4-3.
4.1.3 Scrubber Outlet Test Site
The scrubber outlet test site was located in the 10.75-inch diameter
stack. Two 3-inch test ports were located at 90 degrees to each other
4-4
-------
REVERBERATORY FURNACE CHLORINATION SYSTEM
SCRUBBER INLET SAMPLING LOCATION
VISTA METALS CORPORATION, FONTANA. CALIFORNIA
CROSS SECTION THROUGH SCRUBBER
INLET TEST LOCATION
PORT A
10" SLOT BLOCKED DURING TESTING
• / TO MINIMIZE LEAKAGE
APPROXIMATELY
ItTxIO"VENTILATION DOOR
OPEN TO AMBIENT AIR „
i
Ul
c/j
CJ
TRAVERSE POINT
I
2
3
4
5
6 "
7
8
9
10
II
12
13
14
15
16
PORT
DISTANCE FROM POINT
TO PORT OPENING
(INCHES)
1.0
1.0
1.6
2.3
3.1
4.0
5.2
B.B
11.4
13.1
14.2
15.2
16.0
16.7
17.2
17.2
SCRUBBER
54"
i
ONE WAY
/ FLAPPER VALVE
40 ACFM
FLOW FROM
SETTLING CHAMBER
1300-1500 ACFM
DILUTION AIR
SAMPLEPORT AT 90'TO EACH OTHER
J
-------
approximately 44 inches downstream of the fan and approximately
11 inches upstream of the stack outlet. Figure 4-4 illustrates the
sampling location and provides the location of the sampling points.
4.2 Borings Dryer
Ventilation of the borings dryer is by natural draft. The product
discharge point is also where combustion air enters the process. Flue
gases from the dryer burner pass into the afterburner chamber and up
the 47-inch internal diameter stack. Figure 4-5 illustrates the sampling
location and sampling points. The stack gas velocity is quite low (A p
was approximately 0.001) and, with the afterburner in operation,
stack gas temperatures were in excess of 2000°F. Only preliminary
tests were performed on this unit as explained in Section 2.4.
4.3 Particle Size Test Locations
Particle size distribution determinations were made at the rever-
beratory furnace chlorination scrubber inlet and at the borings dryer.
Both test sites were sampled using a straight nozzle rather than the
button-hook design. The same test port location was used at the borings
dryer site as that described in Section 4.2. At the reverberatory
furnace chlorination scrubber, there existed a second slot, 4 inches
by 10 inches, downstream of the slot used for particulate and velocity
testing described in Section 4.1. Insertion of the Andersen Impactor
through this downstream location allowed sampling in the same plane
used for particulate and velocity determinations. A single point of
average velocity was used for particle size sample collection.
4-6
-------
REVERBERATORY FURNACE CHLORINATION
SCRUBBER OUTLET SAMPLING LOCATION
VISTA METALS CORPORATION, FONTANA, CALIFORNIA
SOUTH PORT (B)
WEST PORT (A)
TRAVERSE POINT
1
2
3
4
5
6
7
9
10
11
!2
DISTANCE OF POINT
FROM OUTSIDE OF PORT
(INCHES)
13*
135$
127/8
12
II
77/8
65/8
57/8
5*
5
K
3"SAMPLE PORTS
10.75 ID
10'
42'
FROM SCRUBBER
FAN
-------
PORT B
BORINGS DRYER EMISSIONS SAMPLING LOCATION
VISTA METALS CORPORATION. FONTANA, CALIFORNIA;
B"
TRAVERSE POINT
1
2
3
4
5
6
7
8
9
10
PORT A DISTANCE OF POINT
FROM OUTSIDE OF PORT
(INCHES)
8.2
10.9
13.9
17.6
23.1
37.9
43.4
47.1
50.2
52.8
111"
0
« 47'!-
ROTARY BORINGS DRYER
t
AFTERBURNER
J
z
era
=0
z
tn
VI
o
z
CJ
en
-p.
CJI
-------
4.4 Visible Emission Observation Locations
The observer locations, wind direction, sun locations, and source
locations recorded during visible emission observations made at Vista
Metals Corporation are illustrated in Figure 4-6. Observation points
were selected to meet EPA Method 9 criteria as closely as practicable.
Observations on May 19, 20, and 21 were made from rooftop level while
those made on May 28 were conducted from the ground.
4.5 Fugitive Emission Observation Locations
Figure 4-6 indicates the ground level location of the observer for
fugitive emissions from the furnace charging well and the borings dryer
processes. All observations were made from ground level.
4.6 Scrubber Liquor Sampling Locations
Figure 4-7 illustrates the Vista Metals Corporation chlorination
scrubber system. Scrubber liquor samples were collected at the scrubber
discharge into the caustic mix tank.
4.7 Pressure Drop Measurement Locations
The location of the taps used for monitoring pressure drop across
the scrubber are illustrated in Figure 4-7. The inlet tap was a slot
near the base of the scrubber. Both tubes were inserted several inches
into the ducts.
4.8 Stack Gas Molecular Weight Sampling Locations
Samples for Orsat analysis were taken at the chlorination scrubber
inlet and outlet test ports and at the borings dryer test ports.
Chlorination scrubber inlet and outlet Orsat samples were taken during
testing from the unused port. Borings dryer Orsat samples were taken
during both controlled and uncontrolled operation.
4-9
-------
FIGURE 4.6
Overhead View of Emission Sources
and Observer Locations for Conduct of Visible Emission
and Fugitive Emission Observations
at Vista Metals Corporation, Fontana, California
DIRECTION OF SUN
MAY 28, 1981
A MAY 28, 1981
SCRUBBER DISCHARGE
STACK
FURNACE CHARGING
J J€LL STACK
LEGEND:
A Visible Emissions
Observer Locations
©Fugitive Emissions
Observer Locations
.Direction
of Wind
O
DIRECTION OF SUN
MAY 19, 1981
MAY 20, 1981
MAY 21, 1981
BORINGS DRYER BUILDING
, (
i
Al
O O
MAY 19,1981
D
j
O
A
MAY 20, 1981
MAY 21, 1981
4-10
ENGINEERING-SCIENCE
-------
FIGURE 4.7
CHLORINATION SCRUBBER LIQUOR
SAMPLING LOCATION AND PRESSURE DROP
MEASUREMENT LOCATIONS
VISTA METALS CORPORATION, FONTANA. CALIFORNIA
TO STACK
GAS FLOW
PRESSURE TAP
PRESSURE
DROP
MANOMETER
T T
LIQUOR
FLOW
••M*/™ ™""1
PRESSURE TAP
(IN PORT)
GAS
FLOW
SCRUBBER LIQUOR
SAMPLE COLLECTION POINT
T
CAUS
MIX T
TIC
ANK
INTERN ITTANT LKIUOR DISCHARGE
4-11
ENGINEERING-SCIENCE
-------
SECTION 5
SAMPLING AND ANALYSIS METHODS
-------
5.0 SAMPLING AND ANALYSIS METHODS
This section presents general descriptions of sampling and analysis
procedures employed during the emissions testing program conducted at the
Vista Metals Corporation secondary aluminum smelting facility in Fontana,
California, the week of May 18, 1981. Details of sampling and analysis
procedures are contained in Appendices J and K.
5.1 EPA Reference Methods Used in This Program
The following EPA Reference Methods were used during this emission
testing program. These methods are taken from "Standards of Performance
for New Stationary Sources", Appendix A, 40 CFR Part 60.
Method 1 - Sampling and Velocity Traverses for Stationary Sources
This method specifies the number and location of sampling points
within a duct, taking into account duct size and shape and local
flow disturbances.
Method 2 - Determination of Stack Gas Velocity and Volumetric
Flowrate
This method specifies the measurement of gas velocity and flowrate
using a pltot tube, manometer, and temperature sensor. The physical
dimensions of the pitot tube and its spatial relationship to the
temperature sensor and any sample probe are also specified.
Method 3 - Gas Analysis for Carbon Dioxide, Oxygen, Excess Air,
and Dry Molecular Weight
This method describes the collection of both grab and integrated
samples and the analysis for carbon dioxide, oxygen, and carbon
monoxide. It also describes the calculations to determine percent
excess air and dry molecular weight.
Method 4 - Determination of Moisture Content in Stack Gases
This method describes the extraction of a gas sample from a stack
and the removal and measurement of the moisture in that sample by
condensation impingers. The assembly and operation of the required
sampling train is specified.
5-1
-------
Method 9 - Visual Determination of the Opacity of Emissions from
Stationary Sources
This method describes how trained observers are to determine the
opacity of emissions. The duration and frequency of observations,
the orientation of the observer with respect to the source, the sun
and the background, the methods of data recording and calculation,
and the qualifications of observers are specified.
Proposed EPA Method 22, which describes a method for determining
the frequency of visible fugitive emissions from material processing
sources was also employed in the testing.
The emission tests for total particulate, chlorine, and chloride
were conducted using a modification of EPA Reference Method 5 with an
absorbing solution in the impingers to capture the chlorine and chloride,
and an analysis procedure for chlorine and chloride. Total particulate
and condensible hydrocarbon measurement also employed EPA Method 5
techniques, with the major departure being the extraction and drydown
of the impinger catch to measure the noncondensible hydrocarbons. The
following paragraph describes the sampling methods and analysis used
in the testing.
5.2 Particulate, Chlorine, and Chloride Sampling and Analysis
5.2.1 Sampling Methods
5.2.1.1 Reverberatory Furnace Scrubber Inlet Sampling Methods
The test team used EPA Method 1 to determine the location and
number of sampling points, and EPA Method 2 to measure gas velocities.
The pitot tubes were of the S type design, constructed in accordance
with EPA Method 2.
A standard EPA Method 5 train was used to collect total particulate
matter, particulate chlorides, and gaseous chlorine compounds. A
schematic of this sample train is shown in Figure 5-1. Field data for
5-2
-------
Particulate, Chlorine and Total Chloride Sampling Train
THERMOCOUPLE
POINT A
PI TOT TUBE
PROME
THERMOCOUPLE
PROBE
PI TOT TUBE
STACK WALL
PI TOT MANOMETER
HEATED AREA
THERMOMETERS
ORIFICE
DRY HAS MUTER
TILTER HOLDER
THERMOCOUPLE
IMPINGER TRAIN
THERMOCOUPLE
CHECK VALVE
1MPINGKR
ICE RAIH
VACUUM LINE
VACUUM GIIAfiE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
the test were recorded on standard Method 5 type data sheets and are
contained in Appendix C. Control device inlet samples were collected
in an alkaline arsenite impinger solution of 2.5 N potassium hydroxide
and 0.5 N sodium arsenite. The analytical method including instructions
for preparation of the absorbing solution are contained in Appendix J.
The method offers both a high and a low option on solution concentration.
The stronger solution was chosen for the scrubber inlet because of the
expected high chlorine concentrations. The impinger train consisted
of 4 impingers, the first two each containing 200 ml of absorbing
solution and the third containing 100 ml of absorbing solution. The
fourth impinger contained silica gel for final moisture removal. It
was deemed more important to have the additional absorbing capacity of
the third impinger than to have an empty knock out impinger.
The sample train used Reeve Angel type 934 AH glass fiber filters.
This type of filter absorbs lesser quantities of acid gases than other
filters and thus allows for more accurate particulate measurement.
Sample train operation was identical to the EPA Method 5 procedure
with the exception of filter temperature control and probe heat.
Effluent particulates are temperature sensitive and therefore, to prevent
thermal degradation, the filter and probe were maintained at approximately
20°F above the stack gas temperature. This temperature was considered
adequate to prevent condensation. No cyclone was used in the train.
Tests runs were scheduled to coincide with the latter portion of
the chlorination period because with most of the magnesium content of
the melt removed, free chlorine was expected to pass through the melt
into the exhaust duct. The demagging operation on May 19 lasted from
1735 until 2113, with the source test occurring from 1911 to 2104. On
5-4
-------
May 20, demagging lasted from 1730 to 2305 and testing started at 1835
and stopped at 2135. The plant was experiencing upset conditions with
high chlorine concentrations during the test period. The high chlorine
concentration was evidenced by the green gas visible at the air bleed-in
location. The test team traversed each inlet port once, and with another
set of impingers traversed one of the ports. Solutions were saturated
in both sets of impingers indicating some chlorine probably passed through
the train. On May 21 demagging occurred from 1750 to 2130. The source
test run started at 1930 and ended at 2140. A separate train was used for
each port on this third run to avoid the solution saturation that occurred
during the previous test runs.
5.2.1.2 Reverberatory Furnace Chlorination Scrubber Outlet Sampling
Methods
Scrubber outlet sampling followed the same procedure as the scrubber
inlet, except that the first three impingers were each filled with 100 ml
of 0.1 N potassium hydroxide. Outlet testing was conducted simultaneously
with inlet testing. Test time variations were scheduled into the start
of the test runs so that tests could finish at the same time, and thus be
operating simultaneously at the time of the expected high chlorine emissions.
Unlike the inlet train, only one set of impingers was used for each test.
5.2.2 Sample Recovery and Preparation
ES leased a large covered truck for train preparations, sample
recovery, cleanup, and equipment storage. Chlorine analysis was conducted
at the ES-Arcadia laboratory, approximately 40 miles from the test site.
Following each test run, separate liquid fractions were recovered
from components of each sample train. For the scrubber inlet train,
one sample was collected for the probe and front half rinse and the
5-5
-------
other for the combined impinger contents and back half rinse. For the
outlet samples, one sample was collected for the probe and front half
wash, one for the first impinger and wash, and one for the combined
second and third impingers and wash. The first impinger was separated
from the other two in the event that chlorine concentrations were so low
that dilution with the contents of the second and third impingers would
reduce chlorine and chloride concentrations below the detection range.
The cleanup person on the first outlet run inadvertently deviated from
the procedure by combining the contents of all three impingers and the
wash prior to analysis. Subsequent analysis, however, showed sufficient
concentration for analysis. On the second and third test runs the proper
procedure was followed. Filters were carefully removed from both inlet
and outlet trains and placed in petri dishes for transport to the lab
for dessication, weighing, and analysis.
For the cleanup procedure on both the inlet and outlet trains, the
probe was rinsed three times with deionized distilled water and this
rinse was combined with the filter housing front half rinse to form the
front half sample. Prior to any back half rinsing, the impinger contents
were transferred to a 250 ml graduated cylinder for moisture content
determination. Following this determination on the inlet sampler, all
back half glassware and impingers were rinsed with deionized distilled
water and combined in a 1000 ml volumetric flask and brought to a 1000 ml
volume. On the outlet samples, as previously mentioned, the first
impinger contents and rinse were kept separate from the second and
third impinger contents and rinses. On the outlet train the samples
were not brought to volume. Sections C.I.I and C.I.2 of Appendix C
contain the sample recovery data sheets for inlet and outlet samples.
5-6
-------
Standard pH paper was used to check impinger solution pH following
each run. On Run No. 1, the first impinger of the inlet train was neutral
to acidic; and on the outlet train all were basic. On Run No. 2, both
inlet trains were saturated, i.e., neutral in all impingers. On the
outlet side, the first impinger had a pH of 7, with remaining impingers
basic. On Run No. 3 both inlet trains had neutral first impingers, and
the outlet train was all basic. On the saturated Run No. 2 inlet, precipi-
tate was observable in the impingers, but partly disappeared as the
impinger solution warmed to ambient temperature.
5.2.3 Sample Analysis
Particulate/chlorine/chloride samples were brought to the ES Arcadia
Laboratory as soon as possible after cleaning, and the samples immediately
prepared and analyzed for chlorine. Chlorine analysis occurred from 1-1/2
to 4 hours after completion of each test run. Inlet samples were analyzed
by the alkaline arsenite procedure and outlet samples were analyzed by
DPD Ferrous Titrimetric Method 409E. Appendix J contains the analysis
methods used.
Chlorine
For chlorine analyses at the scrubber inlet using the arsenite
procedure, 25 ml aliquots were taken for titration from the 1000 ml
sample volumes. In the arsenite procedure, chlorine is determined by
measurement of the quantity of unreacted arsenite. On Run No. 2 both the
inlet sample sets were saturated and all the arsenite had been tied to
chlorine. Because the analyst could get no titration end point, he reduced
the aliquot size to 10 ml and added 1 ml of arsenite to reduce any un-
combined chlorine. The sample was then titrated and an end point reached.
Blanks were treated accordingly. There is some question with this procedure
5-7
-------
and results should be viewed accordingly. Because it is certain the
samples were saturated, as evidenced by the neutral pH and the chlorine
release upon opening of the sample jars in the laboratory, the chlorine
values obtained should be considered minimums. Inlet mass chlorine
amounts could be double those measured as one set of impingers operated
twice as long as the other and both of the impinger solutions were saturated
and unable to tie up additional chlorine. On the second set of impingers
of Run No. 3 inlet, the analyst could not detect any chlorine in the sample.
The sample obviously contained chlorine because subsequent chloride
analysis (after addition of l^C^ for CIO" to Cl~) showed significant
quantities. For purposes of this analysis it was assumed that the second
half chlorine mass was identical to that in the first set of Run No. 3 inlet
impingers.
For scrubber outlet chlorine concentrations using the DPD Ferrous
Titrimetric procedure, small aliquots (1 to 2 ml) were brought to 100 ml,
.the pH adjusted, and titrated. These major dilutions were necessary to
keep the titer volume within the 4 ml specified in the "Standard Methods
for Water and Wastewater" procedure. To minimize error in these dilutions,
however, the analyst should have started with a larger aliquot and con-
ducted successive dilutions to achieve the same end. The site test plan
specified a 10 ml aliquot, which required in all cases titer volume in
excess of 4 ml. Rather than successive dilutions of the first aliquot,
the analyst, in all but one sample, chose to take additional smaller aliquots
for direct dilution. This latter procedure is more subject to error
because it requires the metering of a small aliquot, where in successive
dilution larger samples are metered. Successive dilutions were conducted
5-8
-------
on Run No. 2, however, but starting with a 2 ml aliquot. The first im-
pingers showed a neutral pH indicating saturation of the solution, and
a 2 ml aliquot was brought to 100 ml, and 10 ml of that volume sub-
sequently diluted to 100 ml. This amounted to a 500:1 dilution. On
analysis of the second and third impingers, a 2 ml aliquot was similarly
diluted and titrated. Also on this second and third impinger, a 1 ml
aliquot was diluted directly to 100 ml. The results of these two
dilution methods compared favorably.
Chlorides and Particulate Matter
Chloride analysis was conducted in the ES McLean Laboratory after
completion of the field test. Total chlorides were determined by the
mercuric nitrate titration procedure. For the control device inlet
samples, chlorides were determined by titrating a sample aliquot directly.
Since chlorine reacts in the arsenite absorbing solution to form two moles
of chloride for each mole of chlorine absorbed, it was necessary to cal-
culate the chlorine contribution to the total chloride concentration.
Subsequently, this contribution can be subtracted from the total chloride
concentration to determine the chloride concentration originally present
in the sample gas stream. Prior to analysis of the outlet samples, where
the Method 409E KOH absorbing solution had been used, it was necessary
to pretreat the impinger sample solution with hydrogen peroxide to destroy
any residual chlorine remaining in the sample. Residual chlorine gradu-
ally decomposes to form chlorides, so to eliminate questions concerning
the percentage conversion, all outlet field samples required this
pretreatment prior to chloride analysis. At the time the test plan was
prepared, the HnO* treatment seemed necessary only for Cl analysis
5-9
-------
purposes and not to prevent Cl2 loss. Therefore, the time of treatment
was not specified. For reasons that became apparent from the test
results and explained in the following paragraph, the outlet samples
should have been treated for chlorine to chloride conversion immediately
following chlorine analysis.
The results of the chloride analysis were lower than expected, and
as mentioned in Section 2, were often less than the total chlorine
measured. Even with no chloride caught in the impingers as chloride,
the conversion of the chlorine to chloride should result in like quan-
tities. Possible explanations include the following:
1. Loss of chlorine from the sample between the time of chlorine
analysis in the ES Arcadia Lab to the time of chloride analysis
at the ES McLean Lab. The solutions containing chlorine were
not pretreated and therefore were unstable, and losses could
have occurred before sample jars were closed for shipment, or
what is more likely, after jars were opened for chloride
analysis. Gases may have released from solution and escaped
when the cap was removed. The ES McLean Laboratory staff
reported strong chlorine odors when opening the jars. For
the samples in arsenite, there should not have been any C10~
ion available to release, except in Run No. 2 where the solution
was saturated. For the DPD samples, some of this loss could
have been prevented by converting the chlorine in hypochlorite
form to chloride with hydrogen peroxide before shipment or
storage of samples, not after.
2. Interferences in the mercuric nitrate chloride analyses from
chrornate or ferrous ions. The mercuric nitrate method lists
these ions as possible interferents. The scrap may contain
some of these contaminants such as iron, and in the outlet
samples the scrubber water may contain these ions. It is
intended to rerun the chloride samples using a specific ion
electrode.
3. A third possibility is that the C10~ to Cl~ conversion by H202
is not complete, but this has not been investigated.
For particulate and particulate chloride, the probe wash and filter
fractions were analyzed for total chlorides and then dessicated to dryness
prior to gravimetric analysis. The probe rinse volume was determined
5-10
-------
by weighing the sample bottle before and after the contents had been
transferred to tared beakers for dry down. A standard gravimetric
analysis was completed for all filters. Because some of the sample volume
had been taken for chlorine and chloride analysis, the entire volume
was not available for dry down. Particulate matter results were adjusted
by multiplying the drydown results by the ratio of the original total
sample volume to that which was used in the drydown. For chlorine/
chloride analysis, the analyst attempted to take representative aliquots
to minimize error from settling of solids.
After the final weight had been recorded, the water soluable
chloride fraction was determined by extracting the filter with 100 ml
of distilled water followed by a mercuric titration of chlorides.
5.2.4 Analysis of Audit Samples
The chlorine audit samples were analyzed just prior to start of
the field sample analysis at the ES Arcadia Laboratory. The same pro-
cedures were employed as with the field samples. Section 2.5 discusses
the results, which are tabulated in Table 2-13. Chlorine concentrations
in the alkaline arsenite audit samples were in the 200 to 400 mg/liter
range, generally below what is considered to be the range of accuracy
of this analysis method. These low concentrations may account for the
deviations from the audit sample values. For the DPD analysis results,
the error can probably be attributed to sample degradation since the
time of the previous analysis in September and December of 1980.
The ES McLean Laboratory analyzed the chloride samples. These
results also are discussed in Section 2.5. Appendix L contains the
laboratory data and the letter report to EPA.
5-11
-------
5.3 Particulate, Condensible Hydrocarbon, and Non-Condensible Hydro-
Carbon Sampling and Analysis
5.3.1 Borings Dryer Emissions Sampling
Figure 5.2 illustrates the sampling train used on the borings
dryer emissions. As explained in Section 2.4, very little information
was obtained because of the test conditions. One very brief test was
conducted on uncontrolled emissions. The test was not isokinetic, and
the Total Hydrocarbon Analyzer was disconnected because the flow volume
required for the THCA did not leave a flow adequate to allow operation
of the meter box. The short particulate/condensible HC run followed
the proposed Method 5A procedure in Appendix J-3.
The flame ionization total hydrocarbon analyzer (Scott Model 215)
was calibrated before being delivered to the field and also at the test
site. Ultrapure air, containing less than 0.1 ppm total hydrocarbon,
was used to "zero" the instrument. A known concentration of hexane
(688 ppm) was then fed into the analyzer and the response of the analyzer
adjusted to reflect the actual concentration. Two other known
concentrations of hexane (5.2 and 72 ppm) were then introduced into the
analyzer to verify that the analyzer's response was linear. Also, at
the conclusion of the test, the zero and span calibration gases were
reintroduced into the THCA to check for instrument drift. The zero and
hexane span gases were purchased from Scott Specialty Gases, San
Bernardino, California, and were qualified at Hh 2% accuracy.
Non-condensible hydrocarbons were measured for both uncontrolled
and controlled conditions by pulling the sample first through the filter
and impingers prior to analysis with the THCA. The Scott Environmental
Systems THCA was used to determine the quantity (in ppm) of the
5-12
-------
ZERO & SPAN
CALIBRATION
GASES
Method 5A/THCA Sampling Train
— TEMP. GUAGE
PRESSURE GUAGE (SLACK TUBE MANOMETER)
SHUT-OFF VALVE
TOTAL HYDROCARBON ANALYZER W/
FLAME IONIZATION DETECTOR
ril.TER HOLDER
THERMOCOUPLE
IMPINGER TRAIN
THERMOCOUPLE
CHECK VALVE
AIR-TIGHT
FLUOROCARBON LIME
THERMOCOUPLE
PROBE
PI TOT TUBE
STACK WALL
PI TO! MANOMETER
HEATED AREA
THERMOMETERS
ORIFICE
ORY HAS MEIER
IHP INKER
ICE BAIII
VACUUM LINE
VACUUM GIIAfiE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
hydrocarbons that failed to condense after passing through the aqueous
impinger solutions. The THCA pump was used to pull the sample.
Strip chart records were retained for both sample analysis and
calibration procedures. An Esterline Angus Model L-1102-S Strip Chart
Recorder was employed to receive the output of the THCA in the zero to
100 millivolt range.
5.4 Particle Size Distribution Tests
The total particle size distribution of uncontrolled emissions were
determined with an Andersen Cascade Impactor. Resolution across a range
of particle sizes from about 0.5 to 11.0 micrometers (ym) were obtained.
Two standard cascade impactors were employed for the Vista test program.
5.4.1 Particle Size Sampling and Analytical Equipment Description
The Andersen in-stack cascade sampler is a multi-stage, multi-jet
impactor for in situ particle sizing. It aerodynamically classifies
particulates into six size ranges and accounts for size, shape, and
density. Particle sizing at Vista Metals was conducted on the uncon-
trolled borings dryer emissions. A preimpactor was used in order to
collect the larger particles (>10 ym) upstream of the multistage impactor.
Due to high loading from the uncontrolled dryer, only a 50 second test
was conducted. The sample was not collected isokinetically due to
low stack gas flow rates and a resulting low flow rate in the impactor
which would, in turn, result in unreliable size separation.
Figure 5.3 shows a schematic of the Andersen Impactor Sampling Train
and Figure 5.4 shows a schematic of the Andersen Impactor. Impactor
sampling is subject to spurious filter weight changes, either positive
or negative. Some filter media is susceptible to artifact particulate
5-14
-------
GAS FLOW
DUCT WALL
NOZZLE
PREIHPACTOR
CASCADE IHPACTOR
PROBE
THERMOMETERS
SCHEMATIC OF THE ANDERSEN
IMPACTOR SAMPLING TRAIN
VACUUM LINE
BY-PASS
VALVE
SILICA GEL PACKED
DRYING TUBE
VACUUM GAGE
HI-
AIR-TIGHT PUMP
-------
FIGURE 5.4
SCHEMATIC OF THE ANDERSEN IMPACTOR
CASCADE
IMPACTOR
TRAJECTORY OF
PARTICLE TOO
SMALL TO IMPACT
L\\\\\\\\
\\\\\\\1
PARTICLE SIZING ACTION OF
CASCADE IMPACTOR
JET EXIT
TRAJECTORY OF
IMPACTED PARTICLE
5-16
ENGINEERING-SCIENCE
-------
formation on the filter itself by reaction with acid gases, resulting
in anomolous weight gains. Super-drying of filters in hot: air streams can
cause weight losses. Various filter compositions and/or pre-treatment
procedures have been employed to improve the accuracy of impactor
sampling.
Based upon a review of pre-treatment procedures, Engineering-Science
employed Whatman Type QMA quartz/borosilicate glass media. Quartz media
was selected because of its chemical inertness to artifact particulate
formation, and this particular media has been strengthened with a small
amount of borosilicate glass to improve its mechanical integrity. The
filters were temperature conditioned by heating a muffle furnace to the
expected site stack temperature for a period of two hours., Subsequently,
the tare weight of each filter was measured. All filters were tared in
numbered aluminum foil wrappers. The tared filters were grouped into
sets and placed in petri dishes for storage and handling.
The impactor trains were assembled as shown in Figure 5.3. The
pump/meter assembly are standard EPA Method 5 equipment. The use of a
silica gel cartridge only for moisture removal was a deviation from
the site test plan, which showed an impinger train for not: only moisture
removal but moisture determination. In these tests moisture content was
determined from adjacent Method 5 tests. Procedure and acceptance
criteria as used for standard Method 5 sampling was followed by the
impactor sampling.
5.4.2 Equipment Calibration
Sampling equipment was calibrated in accordance with EPA Method 5,
APTD-0576, and the "Quality Assurance Handbook for Air Pollution
5-17
-------
Measurement Systems", Vol. III. Complete records of all calibrations
are maintained at ES and are presented in Appendix I.
5.4.3 Determination of Sampling Points
ES sampled at one sample point for each test run. The sample
point was located at a point of average velocity.
5.4.4 Determination of Sampling Rate and Nozzle Size
Determination of sampling rate and nozzle size were such that the
sample rate was within the design range of 0.50 to 0.75 acfm with
further consideration given to adherence of the ± 10% isokinetic criteria
where possible.
Gas stream parameters - velocity, temperature, moisture, and static
pressure - were obtained from prior tests or from the Method 5 tests.
5.4.5 Andersen Impactor Test Procedure
All chloride scrubber inlet tests complied with the minimum 3
minute sample time criteria. The one test on the borings dryer, however,
ran for only 50 seconds and also was not isokenetic. Other criteria
were generally followed. The following data items were recorded at
three minute intervals: stack temperature, gas meter volume, silica
gel impinger temperature, orifice pressure differential, dry gas meter
inlet and outlet temperature, and pump vacuum. The data were recorded
on a standard particulate field data sheet.
5.4.6 Sample Recovery and Analysis
Sample recovery was performed in a van parked on Vista Metals
property. Acetone was used as necessary for final clean up of the
nozzle and pre-impactor. The combined rinse was placed in a sample
5-18
-------
bottle for subsequent dry down. Before reloading the impactor, it was
rinsed a second time with acetone. This second rinse was discarded.
The cascade impactor for particle sizing was rinsed and brushed
with acetone followed by a methylene wash as part of the cleanup and
prior to each sample run. Acetone was used first because of its non-
reactivity with brushes and also because the cleanup person preferred
to limit exposure to methylene chloride. A set of pre-weighed substrates
was removed from their petri dish and aluminum foil wrappers and loaded
in accordance with the impactor manufacturer's procedures. At the
conclusion of each run, the substrates were returned to the proper
foil wrappers and petri dish for subsequent post-test weighing in the
ES McLean Laboratory.
Analysis of samples was gravimetric, employing a Cahn electronic
Balance Model Number 21. The balance has an accuracy of 0.05 mg and a
sensitivity of 0.01 mg. The substrates were weighed to a constant
weight of + 0.05 mg. The gravimetric analyses were recorded on special
laboratory notebook forms, copies of which are included in Appendix F.
The analysis was completed approximately four weeks after the samples
were received in the laboratory. Data reduction was done using the EPA
Program Number 600/778-072. One modification was employed which involved
the use of the square root of Stokes Number at 50% collection efficiency
( / 4* 50 dimensionless inertial impaction parameter).
The above mentioned program as written by Southern Research Institute
specifies assigning a Stokes Number to each stage as it is calibrated.
For experiments performed in a lab environment this is accepted procedure.
5-19
-------
The difference in the square root of Stokes Number between stages becomes
insignificant when viewed in light of inherent errors associated with
field sampling. Consequently the Stokes Number for round jets, found
in Anderson's impactors, is averaged and set equal to 0.374 for all stages.
5.5 Visible Emissions Observations
EPA Method 9 visible emission observations were conducted concurrently
with the individual outlet emission runs. Readings were recorded by a
certified observer as outlined in "Guidelines for Evaluation of Visible
Emissions", EPA-340/I-75-007. There were no modifications or deviations
from Method 9. The chlorination scrubber plume appeared to contain
water vapor although the stack temperature was approximately ambient. The
distinction between water vapor and visible particulate was difficult.
At the time reverberatory furnace charging well VEO's were taken, no
qualified process observer was present.
5.6 Fugitive Emissions
EPA Method 22 (without modification or deviations) was used to
determine fugitive emissions at the borings dryer charge, discharge
and outer area, and the reverberatory furnace charging well. The
method does not require that the opacity of emissions be determined.
It determines the amount of time that any visible emissions occur
during the observation period; i.e., the accumulated emissions time.
A qualified process observer was not present during any of the observa-
tions.
5-20
-------
5.7 Scrubber Liquor Sampling and pH Analysis
Scrubber liquor samples were collected from the closed-loop recycle
tank at approximately 30-minute intervals during the entire chlorination
period. All liquor samples were collected in 125 ml sample bottles
with Teflon™ cap liners. Room temperature pH was determined using pH
paper. Liquor samples were stored separately and not composited. The
samples will be held for 90 days in the event that EPA or the NPNSS
contractor decide that additional tests are needed. The temperature of
each sample was recorded at the time of collection.
5.8 Scrubber Pressure Drop Measurement
Pressure drop was measured between the inlet and outlet of the
scrubber control device. These measurements with a 3-foot U-tube
manometer using colored water as liquid were taken at approximately 30
minute intervals throughout the test period.
5.9 Cleanup Evaluation Test Procedure
Cleanup evaluations and blank analyses were conducted on inlet
trains and outlet trains for chlorine/chloride and particulate. A
particulate hydrocarbon train cleanup was conducted but not analyzed.
Impingers were charged with solution as if for testing, and then cleaned
as if containing sample. The alkaline aresenite samples were brought
to volume for transfer to the lab, and the DPD samples were transferred
as impinger solution and wash. After respective analysis in the Arcadia
Lab, the containers containing the blank solutions were shipped to the
ES McLean Lab for chloride analysis by the mercuric nitrate method,
and for particulate analysis. Particulate analysis was conducted only
5-21
-------
on the front half wash, as the methods did not call for back half
particulate. Appendix M contains further discussion of the cleanup
procedure.
5.10 Stack Gas Molecular Weight Determinations
Dry molecular weight was determined by a Burrell-type Orsat analyzer
with the measurement tube graduated in 0.2% concentration intervals.
One-quarter inch diameter stainless steel tubing was used to collect
the samples from the stack. EPA Method 3 procedures using a squeeze
bulb to obtain instantaneous samples were employed. Although a Fyrite
was listed as an option in the work plan for use on the chlorination
system inlets and outlets, the Orsat was chosen because of its accuracy
and because it was needed for the borings dryer emissions. The system
was leak checked in accordance with EPA Method 3 procedures and tests
conducted accordingly.
5-22
------- |