v>EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-79-119
June 1979
Research and Development
Performance
Evaluation of an
Electrostatic
Precipitator
Installed on a
Copper Smelter
Reverberatory
Furnace
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-119
June 1979
PERFORMANCE EVALUATION OF AN ELE'CTROSTATIC PRECIPITATOR
INSTALLED ON A COPPER SMELTER REVERBERATORY FURNACE
by
Southern Research Institute
Birmingham, Alabama 35205
Grant No. R-804955
Project Officer
John O. Burckle
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environ-
mental Research Laboratory, U. S. Environmental Protection
Agency, and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies of
the U. S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory -
Cincinnati (lERL-Ci) assists in developing and demonstrating new and
improved methodologies that will meet these needs both efficiently and
economically.
This report presents the findings of an investigation of air pollutant
emissions from the reverberatory furnace pollution control system at a
primary copper smelter. The study was performed to assess the degree of
particulate emissions control and control problems associated with the
application of electrostatic precipitators in the nonferrous metals pro-
duction industry. The results are being used within the Agency's Office of
Research and Development as part of a larger effort to define the potential
environmental impact of emissions from this industry segment and the need
for improved controls. The findings will also be useful to other Agency
components and the industry in dealing with environmental control problems.
The Metals and Inorganic Chemicals Branch of the Industrial Pollution
Control Division should be contacted for any additional information desired
concerning this program.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
ill
-------
ABSTRACT
This report describes tests performed during the period
January 11-16, 1977, on the electrostatic precipitator installed
on the copper reverberatory furnace at the Kennecott Copper
Corporation smelter at Hayden, Arizona. These tests provided
data on the chemical characterization of particulates, noncon-
densables, and gases in addition to operating and performance
measurements of electrical parameters, particle size, voltage-
current distribution, and resistivity. Efforts were also made
to develop computer simulations of ESP performance and to
evaluate overall performance of the control device.
The operating condition of the reverberatory furnace was
normal (with one major exception); however, the operation of the
ESP was erratic, and other unavoidable restraints on the sampling
program prevented acquisition of reliable data to evaluate repre-
sentative performance of the ESP. Nevertheless, analysis of the
data shows the types of information which can be obtained in
evaluation of control devices if the test locations can be
utilized to obtain data representative of normal "on stream"
operations.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables viii
Acknowledgments ix
1. Summary and Conclusions 1
2. Process Description 4
3. Test Facility 10
4. Test Program 18
5. Chemical Analysis 20
6. Particle Size Distribution Measurements 39
7. Electrostatic Precipitator Electrical Conditions 73
8. Mathematical Modeling 82
Appendices
A. Smelter Charts 91
B. Chemical Sampling-Radian 113
C. Chemical Sampling-Southern Research Institute 128
D. Impactor Data 140
v
-------
FIGURES
Number Page
1 Simplified flow diagram of copper smelter 5
2 Reverberatory furnace schematic 9
3 Layout of the precipitator showing sampling
locations H
4 Modified Brink BMS-11 cascade impactor!!!!!!!!!!!!!! 40
5 University of Washington Mark III source test
cascade impactor 4^
6 Inlet cumulative mass loading . ] 47
7 Inlet differential mass concentration (dM/dlogD).... 48
8 Outlet cumulative mass loading, 1/12/77 52
9 Outlet differential mass concentration (dM/dlogD),
10 Outlet cumulative mass loading, 1/14/77 '. 54
11 Outlet differential mass concentration (dM/dloqD),
1/14/77 ....:..;.. 55
12 Outlet cumulative mass loading, 1/15/77 56
13 Outlet differential mass concentration (dM/dlogD),
1/15/77 ....;..:.. 57
14 Outlet cumulative mass loading, 1/16/77 58
15 Outlet differential mass concentration (dM/dloqD),
1/16/77 .......... 59
16 Penetration-efficiency, 1/12/77 60
17 Penetration-efficiency, 1/14/77 [ 61
18 Penetration-efficiency, 1/15/77 62
19 Penetration-efficiency, 1/16/77 63
20 Diagrammatic representation of the thermosystems
Model 3030 Electrical Aerosol Analyzer (EAA) 65
21 The Sample Extraction and Dilution System (SEDS)
designed by Southern Research Institute 66
22 Inlet differential number concentration (dN/dlogD),
average 67
23 Outlet differential number concentration (dN/dlogD),
1/15/77 69
24 Outlet differential number concentration (dN/dlogD),
1/16/77 70
25 Penetration efficiency versus Stokes particle
diameter, 1/15/77 71
26 Penetration efficiency versus Stokes particle
diameter, 1/16/77 72
27 Electrostatic precipitator configuration 74
VI
-------
FIGURES (continued)
Number
28
29
30
31
32
33
Voltage-current curves, 1/13/77
Voltage-current curves, 1/14/77
Voltage-current curves,
Voltage-current curves,
Voltage-current curves,
1/15/77,
1/16/77 A.M.
1/16/77 P.M.
Theoretical and experimental fractional efficiencies
76
77
78
79
80
90
VII
-------
TABLES
Number Page
1 Smelter operating data 8
2 Electrostatic precipitator descriptive parameters
modified Koppers installation on Kennecott's
Hayden Reverb , 12
3 Power supply log, Kennecott reverberatory furnace
precipitator 14
4 Sample port utilization for test program. 19
5 Total concentrations of trace elements at the ESP
inlet duct 24
6 Total concentrations of trace elements at the ESP
outlet duct and stack 25
7 Average values of total elemental concentrations at
the ESP inlet and outlet 27
8 Fractions of total elemental concentrations appear-
ing as vapors at the ESP inlet 29
9 Total and vapor concentrations of elements at the
ESP outlet , 3,
10 Fractions of total elemental concentrations appear-
ing as condensable matter at the ESP inlet 32
11 Concentrations of miscellaneous gases observed at
the ESP inlet and outlet 34
12 Results of SSMS analysis of reverberatory feed and
sand sg
13 Results of analyses of a composite of hopper-dust
samples 37
14 Results of analyses of individual hopper samples.... 38
15 Inlet modified Brink impactor blank stage mass gains 43
16 Outlet University of Washington impactor blank mass
gains 44
17 Calculated impactor mass loadings for the inlet.... 49
18 Calculated impactor mass loadings for the outlet... 51
19 Computer data input (Kennecott) 89
Vlll
-------
ACKNOWLEDGMENTS
The cooperation of the personnel at Kennecott is gratefully
acknowledged in the conduct of this research test program.
Specifically we appreciate the assistance of Mr. K. H. Matheson,
Jr., General Manager, and Messrs. J. S. Nebeker, Clint Fitch,
Mike Kearney, and Joe Mortimer. We also acknowledge the assis-
tance of personnel from Radian Corporation, particularly Messrs.
Klaus Schwitzgebel and Bob Collins.
IX
-------
SECTION 1
SUMMARY AND CONCLUSIONS
During the period of January 11-16, 1977, Southern Research
Institute, and Radian Corporation (as a subcontractor) conducted
tests to evaluate the performance of the electrostatic precipi-
tator installed on the copper reverberatory furnace at the
Kennecott Copper Corporation's Ray Mines smelter located at
Hayden, Arizona. Southern Research Institute and Radian Corpora-
tion collaborated' on collection of data to characterize the
chemical species in the inlet and outlet to the precipitator
with respect to particulates, noncondensables, and gases.
Southern Research Institute also collected smelter and precipi-
tator operating data including electrical operating data for the
precipitator, voltage-current characteristics, resistivity
measurements, and particle size measurements including character-
ization of the overall and particle-size control efficiency.
Computer simulations of the performance of the electrostatic pre-
cipitator were also made using the computer systems model devel-
oped at Southern Research Institute under the sponsorship of the
U. S. Environmental Protection Agency, Industrial Environmental
Research Laboratory, Research Triangle Park, N. C.
Sections 2, 3 and 4 of this report describe the overall
process, test facility, and test program. Section 5 outlines
the results of the various chemical analyses, and Sections 6 and
7 present information on the overall performance of the precipi-
tator. Section 8 describes the computer simulation efforts.
Supporting data with detailed reverberatory operating data,
analytical procedures, and cascade impactor data are presented
in the Appendices.
The primary objectives of the test program were to evaluate
the "on stream" performance of an electrostatic precipitator
associated with operation of a copper reverberatory furnace and
to evaluate methods for determining the chemical species in the
particulate, condensable, and non-condensable emissions from the
reverb and the precipitator.
Operation of the reverberatory furnace during the test
period of January 11-16, 1977, was considered "normal" except
for occasional minor interruptions in converter operations and a
major reverb shut-down on January 13. The feed to the reverb was
-------
primarily a partially roasted calcine containing recycled con-
verter slags and dusts. During the test period, the reverb was
fired with No. 2 fuel oil. Typically, the reverb was charged at
fifteen-minute intervals, slag and matte were tapped periodi-
cally, and fettling was added intermittently. All of these
cyclic and intermittent operations cause excursions in both the
reverb temperature and the effluent that affect the gas flow and
particulate distribution into the precipitator. However, these
variations are normal for a typical copper reverberatory furnace.
To evaluate the performance of the precipitator, gas and
particulate test measurements were made at the inlet and outlet
of the precipitator and at the stack, using isokinetic methods
for collection of representative particulate, condensable, and
non-condensable samples.
The electrical condition and operation of the precipitator
power supplies were erratic during the test period with at least
a part of the power being off during the entire test period.
Further, the voltage-current data indicated anomalous and less
than optimal behavior on several occasions, so that any data
relative to normal operation of the ESP are highly questionable
and are not necessarily representative of typical operation of
this precipitator.
In addition to the electrical malfunctions which precluded
collection of reliable and representative electrical operating
data, sampling in both the inlet and outlet ducts presented
major difficulties that affected the reliability of the particu-
late and gas measurements and analyses. For example, the sam-
pling ports at the inlet were located approximately one meter
from both upstream and downstream flow disturbances. (Note: The
location was dictated by the physical configuration and access-
ibility) . Also, except for measurements in the stack, only
single point readings (no traverses) were made at either the
inlet or outlet sample locations. These restraints necessarily
affect the quality of the data collected and prevent any conclu-
sive judgement of normal precipitator performance. Further,
there was apparently an excessive in-leakage of air to the
precipitator (about 40% of the inlet flow) which further compli-
cated any significant analysis of performance data.
In spite of the preceding limitations, the test data have
been analyzed and the results are presented in this report. Some
of the observations based on the tests are summarized below:
1. Most of the major chemical elements and many of the
minor elements present in the gas stream were removed
at efficiencies in excess of 90%, based on chemical
analyses. Elements removed at greater than 90% were
copper, iron, aluminum, lead, zinc, and molydenum
(present at the inlet at concentrations greater than
-------
10 mg/m3); and cadmium, chromium, nickel, antimony,
and selenium (present at less than 10 mg/m3).
2. Three elements (arsenic, mercury, and fluorine)
were not effectively removed by the precipitator
at, normal operating temperatures (260°C or 500°F).
No significant increase in particulate collection /'
of these elements was observed by decreasing the ^
temperature (of an out-of-stack filter) to 120°C
(250°F).
3. Electrical conditions were highly variable during
the test period, and current densities in specific
fields varied from day to day.
4. Particulate resistivity was not limiting the operating
characteristics of the precipitator.
5. Collection efficiency varied from one test to another;
however, most of the data showed collection efficien-
cies greater than 95%.
-------
SECTION 2
PROCESS DESCRIPTION
A simplified flow diagram of the smelting system for the
Kennecott smelter at Hayden, Arizona is shown in Figure 1. Copper
concentrates (containing recycled converter slag and dust), pre-
cipitates, and lime-silica flux are blended together as a dry
feed to a single fluid-bed roaster operated at about 1150°F. In
the roaster (or reactor), the concentrates are partially roasted
to obtain calcines for feed to the reverberatory furnace and an
S02-rich gas for feed to the sulfuric acid plant.
The roaster underflow calcine, the roaster overflow cal-
cines from the cyclones, and various dusts from gas coolers,
waste-heat boilers, and precipitators are combined to provide the
charge to the reverberatory furnace. Fettling (miscellaneous
slags or sand) are also added to the reverb as needed for special
cooling. The nominal operating temperature of the reverb is
about 2800°F, but some variations result from intermittent charg-
ing (about every fifteen minutes), slag skimming, and matte tap-
ping. The reverb at Hayden may be either gas or oil fired.
The primary functions of the reverberatory furnace are
melting of the mineral charge and separation of the molten charge
into a slag waste and a copper-bearing matte. The slag is peri-
odically skimmed out of the reverb for disposal, and the molten
matte is periodically tapped from the bottom of the reverb and
transferred by ladles to one of the three Fierce-Smith convert-
ers. Off-gases from the reverb are sent to two waste-heat boil-
ers and then to the electrostatic precipitator before being
exhausted to the atmosphere through a 600-ft stack.
The matte from the reverb is transferred to one of the con-
verters where silica flux is added, and the resultant mixture is
air blown to obtain a copper-rich iron-silica slag which is re-
moved from the converter, slow cooled, and then returned to the
crushing/flotation/concentrating circuit. After removal of slag
from the converter, the remaining molten mass is air blown to
convert the residual copper sulfide or "white metal" to molten
copper or blister copper which is then transferred to the anode
furnace for casting. Off gases from the converter are cooled,
and dusts are removed in a Peabody scrubber before the S02-rich
gases are sent to the acid plant. Recovered converter dusts are
recycled to the concentrate plant.
-------
CONCENTRATES
PRECIPITATES
FLUX
600 FT
STACK
AIR
FLUOSOLIDS
ROASTER
1150°F
1
CALCINE
^-
GAS
COOLER
DUSTS
CALCINE AND DUST
ELECTROSTATIC
PRECIPITATORS
AIR AND FUEL
WASTE-HEAT
BOILERS
(2)
REVERBERATORY
FURNACE
2800°F
REVERB SLAG
DISPOSAL
MATTE
SILICA
FLUX
BLISTER COPPER
TO ANODE FURNACE
PIERCE-SMITH
CONVERTERS
(3)
\CE ^
w
GAS
CONVERTER SLAG
I ^ (FOR RECYCLE TO
CONCENTRATOR)
DUST
PEABODY
I
DUSTS TO
S02-RICH
1 GASES TO
ACID PLANT
CONCENTRATOR
Figure 1. Simplified Flow Diagram of Copper Smelter.
-------
During the test period of January 11-16, 1977, the roaster,
reverberatory furnace, and converters were operated under moder-
ately stable conditions, except for a major reverb shut-down on
January 13 and minor interruptions in converter operations.
Operational data for the test period are summarized in Table 1
and indicate that the overall smelter operation was reasonably
consistent during the period, recognizing the normal cyclic dis-
continuities of reverb and converter operations.
The reverberatory furnace was fired exclusively with No. 2
fuel oil; no natural gas was used. The nominal temperature of
the reverberatory furnace was reported to be 2800°F, however,
actual temperatures measured near the feed-end of the reverb
generally ranged from 2500 to 2800°F with typical intermittent
decreases of 200 to 300°F with each charge cycle. Temperature
profiles for the test period are shown in the Appendix, Figures
A-13 to A-18.
A schematic diagram of the reverberatory furnace system is
shown in Figure 2. Typical average operating conditions for the
reactor (roaster) and reverberatory furnace during the January
12-16 tests were:
Reactor (roaster);
Dry feed = 60 tons/hr
Air feed = 15,000 CFM
Temperature = 1150°F
Reverberatory furnace;
Fuel oil = 30 gal/min
Calcine charge = 15 tons/charge @ 15 minute intervals
Fettling added = 0.5-1.0 ton/hr (intermittent)
Slag tapped = 11-12 tons/hr (intermittent)
Matte tapped = 2 ladles/hr (intermittent)
Temperature = 2500-2800°F
The discontinuities in operation of the reverberatory furnace are
normal for copper smelting, and it is recognized that surges are
encountered which affect downstream operations such as the waste
heat boilers and the electrostatic precipitator. Removal of slag
and tapping of matte cause relatively minor downstream perturba-
tions; however, the periodic charging of feed to the reverb
results in significant downstream surges in both gas temperature
and composition (including particulate loading). The results of
these reverb surges are illustrated in Figures A-13 to A-18 which
show reverb temperatures and are also confirmed in Figure A-21
which gives a typical waste-heat boiler profile. For example,
the steam temperature shows a fluctuation of 20-30°F that is
coincident with the periodic charging of the reverb at fifteen
minute intervals. Similar perturbations must necessarily be
-------
reflected in the electrostatic precipitator; however, the effects
would probably be lessened by damping from the waste-heat boilers
and the balloon-flue leading to the precipitator.
Based on the data obtained from the Hayden smelter, it
appears that the operation of the reverberatory furnace during
the period January 11-16, 1977, was typical of normal operations
(except for the failure on January 13), and variations encounter-
ed downstream were typical of normal "on stream" reverb opera-
tion.
-------
TABLE 1. SMELTER OPERATING DATA*
1. Dry Feed
Date
1/11
1/12
1/14
1/15
1/16
to Reactor
A
464
471
460
478
480
(tons/shift)
B
484
477
492
445
497
C
455
477
448
480
503
Avg.
467
475
467
468
493
Average feed rate = 60 tons/hr
Flux Added to Concentrate 3. Fettling to Reverb
Date
Tons
1/11 25
1/12 41
1/14 30
1/15 28
1/16 30
Avg. 30.8
Slag Removed (tons)
Date Reverb Converter
Date
1/11
1/12
1/14
1/15
1/16
Avg.
Ladles
Tons
18
20
10
12
26
17.2
of Matte Produced
1/11
1/12
1/14
1/15
1/16
Avg.
245
212
318
206
373
271
498**
620
647
629
599
1/11
1/12
1/14
1/15
1/16
Avg.
48
51
47
48
43
47,
6. Fuel Oil Burned (gal/day)
1/11
1/12
1/14
1/15
1/16
Avg.
42,904
41,261
43,606
43,461
39,881
42,223
7,
8,
9,
10,
Roaster (Reactor) Temperature = 1150 + 100°F
Reverberatory Furnace Temperature = Nominal 2800°F
Reverberatory Gas Flow = 13,500 CFM (estimated)
Roaster (Reactor) Air Flow = Nominal 15,000 CFM
**
See Figures A-l to A-20 for daily operating charts.
No. 3 converter not operating.
-------
ROASTED
CHARGE
COMBUSTION
AIR
FUEL
OIL
MATTE SLAG
TAP TAP
HOLES HOLES
ELECTROSTATIC
PRECIPITATOR
(4 CHAMBERS)
Figure 2. Reverberatory Furnace Schematic
-------
SECTION 3
TEST FACILITY
Figure 3 is a schematic which shows the precipitator layout
and the sampling locations used during the test program. The
inlet sampling ports (Test Point No. 1) were located in the inlet
plenum adjacent to the entry to the No. 1 precipitator chamber;
the outlet sampling ports (Test Point No. 2) were located in the
duct between the precipitator and the stack; and the stack sam-
pling ports (Test Point No. 3) were located at the 70 meter (230
ft) level with the four ports at 90° angles. Test Point No. 1
was approximately 1 m from both upstream and downstream flow dis-
turbances and therefore was not regarded as a desirable sampling
location.
The precipitator on which the tests were conducted is an
old Koppers unit which has recently been upgraded for improved
performance. Physically, the precipitator is divided into four
chambers. Each chamber is equipped with dampers on the inlet
and outlet. The dampers on the outlet side are closed during
rapping to reduce reentrainment. Each chamber is closed for
rapping for about five minutes every hour. During the rapping
periods, all gas flow is handled by the three "on stream"
chambers so that intermittent periodic flow surges (other than
those resulting from reverb operations) of 33% are normal for
the precipitator operation.
The total collection electrode area is 5,044 m2 (54,300
ft2) and was designed for a total gas flow of about 140 m3/sec
(300,000 ft3/min)at 500°F. From these data, the design SCA is
estimated to be 36 m2/(m3/sec) or 181 ft2/(1000 ft3/min.) Actual
inlet flows during the test period averaged 73.4 m3/sec (155,500
ft3/min) at 473°F and ranged from 68 m3/sec (143,850 ft3/min) to
79 m3/sec (168,000 ft3/min) which would give an actual SCA of
of about 70 m2/(m3/sec) or 350 ft2/(1000 ft3/min). Other
descriptive parameters for the precipitator are shown in Table
2.
Electrically, the precipitator is divided into twelve sec-
tions which are energized by six power supplies as shown in
Figure 3. As a part of recent modifications, the power supplies
were rewired for full wave operation, and transformer secondary
current meters were installed, in addition to the current and
10
-------
TEST POINT NO. 2 (OUTLET)
T-
cr
LL
CC
<
T
C
V
1
/o
?
(
(
(
JVF
_T/F
JVF
~\
v^
8V.
01
^ 1
P.
3 3 oc
— v m
) CQ
<
~\ 2 o
[=)
^ /
T/F
^
Ir2
<
T
" T/F
s
^ /
^ 6
*
^ 4 (T
s m
J OQ
<
R 5 0
P
^ /
3T^
OUTLET FIELD
CENTER FIELD
INLET FIELD
TEST POINT NO. 3
TEST POINT NO. 1 (INLET)
Figure 3. Layout of the precipitator showing sampling locations.
11
-------
TABLE 2. ELECTROSTATIC PRECIPITATOR DESCRIPTIVE PARAMETERS
MODIFIED KOPPERS INSTALLATION ON KENNECOTT'S HAYDEN REVERB
Item
Metric
English
Collection electrode area
(4 chambers)
Each T/R set area (6 sets)
No. of fields per chamber
Collection electrode spacing
Collection electrode dimensions
Corona Electrode dimensions
(square wire)
Corona electrode spacing
(parallel to flow)
Number of gas passages (total
Gas passage length (active)
Volume flow rate (design)
Actual flow rate (average)
Design temperature
Design efficiency
Design specific collection
electrode area (SCA)
Actual specific collection
electrode area (SCA)
5,044 m2
840 m2
3
0.254 m
2.29 x 6.10 m
0.397 cm
0.152 m
60
6.86 m
140 m3/sec
73.4 m3/sec
260°C
95%
36 m2/(m3/sec)
70 m2/(m3/sec)
54,300 ft2
9,050 ft2
3
10 in.
7.5 x 20 ft
0.15625 in.
6 in.
60
22.5 ft
300,000 ft3/min
155,500 ft3/min
500°C
95%
181 ft2/(1000
ftVmin)
350 ft2/(1000
ft3/min)
12
-------
voltage meters which are on the primary side of the power supply
transformers. During the test period, voltage dividers were
installed to allow secondary voltage measurements in addition
to available primary voltage and primary and secondary current
data. Representative power supply data obtained during the test
period are shown in Table 3.
The power supply data in Table 3 illustrate that the elec-
trical operation of the precipitator was very erratic. For the
entire test period, TR No. 3 serving chambers 1 and 2 was par-
tially or totally out of service. While a partial loss of one
out of six TR sets would not materially affect the efficiency
of the electrostatic precipitator, the erratic behavior of at
least three of the remaining five would severely limit the value
of any data collected regarding typical performance of the pre-
cipitator.
Therefore, data in this report should be used only with
full recognition of the conditions existing when the tests were
made. In brief, while the operation of the reverberatory_fur-
nace may be regarded as normal, the test locations (especially
Test Point No. 1) were not ideal, and the electrical performance
of the precipitator was inadequate to support a meaningful or
significant evaluation of normal operation of expected efficiency
for this installation.
13
-------
TABLE 3. POWER SUPPLY LOG
KENNECOTT REVERBERATORY FURNACE PRECIPITATOR
HAYDEN, ARIZONA
Date Time
1/12/77 0950
1/12/77 1300
1/12/77 1800
1/12/77 1910
1/14/77 1030
T/R
set
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
33
4
5
6
Primary
voltage
(V)
-
-
280
280
205
300
135
315
285
230
210
220-260
<100
325
300
24
-
250
100
350
Primary
current
(A)
-
-
70
40
26
58
60
67
65
30
25
20-38
62
62
64
30
-
27
60
55
Primary
power
(kW)
-
-
19.6
11.2
5.3
17.4
8.1
21.1
18.5
6.9
5.3
-
<6.2
20.2
19.2
0.7
-
6.8
6.0
19.3
Secondary
voltage
(kV)
-
-
-
-
-
-
-
-
49.3
55.7
43.6
51.3
-
-
44.3
49.6
-
47.7
-
-
Secondary
current
(mA)
250
801
601
1602
250
250
240
120
70
0
250
250
240
145
82
215
230
245
240
95
87
50-140
260
245
250
160
-
80
250
220
Spark
Power rate
(kW) (SPM)
-
- -
0
0
t)
10
0
0
11.8 0
5.3 60
3.8 10
15
0
0
11.1 10
7.9 0
-
3.8 0
0
0
- continued -
-------
TABLE 3.
(continued)
Ul
Date Time
1/14/77 1245
1/14/77 1445
1/14/77 1740
1/15/776 1015
1/15/776 1355
T/R
set
1
2
33
4
5
6
I
2
3*
4
5
6s
1
2
3"
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
Primary
voltage
(V)
300
240
-
260
-
370
280
226
170
220
-
310
270
250
150
210
-
270
170
160
160
170
-
220
170
180
160
200
-
250
Primary
current
(A)
64
20
-
24
62
62
60
17
10
17
60
45
52
25
10
15
44
32
14
^5
^5
12
58-62
16
14
10
10
14
58-62
30
Primary
Power
(kW)
19.2
4.8
-
6.2
-
22.9
16.8
3.8
1.7
3.7
-
14.0
14.0
6.3
1.5
3.2
_
8.6
2.4
0.8
0.8
2.0
-
3.5
2.4
1.8
1.6
2.8
-
7.5
Secondary
voltage
(kV)
44.8
48.2
46.2
-
-
44.4
46.4
37.3
43.5
—
-
44.3
47.3
38.0
45.9
_
-
37.0
42.8
38.3
41.4-45.0
-
-
37.1
41.8-45.6
39.1
46.1
-
-
Secondary
current
(mA)
240
60
-
70
250
250
220
40
30
50
250
160
190
80
30
40
180
100
30
10
20
15-20
240-260
50
30
10
30
20
250
100
Power
(kW)
10.8
2.9
-
3.2
-
-
V
9.8
1.9
1.1
2.2
-
-
8.4
3.8
1.1
1.8
-
-
1.1
0.4
0.8
•v-0.7
-
-
1.1
0.4
1.2
0.9
-
-
Spark
rate
(SPM)
0
0
0
0
0
0
100
50
0
10
0
0
125
25
100
10
75
0
0
Unstable
0
10-25
Unstable
Unstable
-
^50
0
0-50
0
0-25
- continued -
-------
TABLE 3. (continued)
O-i
Date Time
1/15/776 1510
1/15/776 1740
1/16/77 7 1000
1/16/777 1200
T/R
set
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
Primary
voltage
(V)
150
150
160
160
70 ?
285
130
130
150
170
80 ?
210
172
165
160
220
^90
320
180
190
150
240
80 ?
332
Primary
current
(A)
11
<10
10
10
62
40
'vS
<5
'vS
12
58-62
18
22
13
12
27.5
60-62
62.5
23
14
10
29
60-62
62.5
Primary
Power
(kW)
1.7
1.5
1.6
1.6
4.3
11.4
1.0
0.7
0.8
2.0
4.8
3.8
3.8
2.1
1.9
6.1
5.5
20.0
4.1
2.7
1.5
7.0
0.5
20.8
Secondary
voltage
(kV)
34.9
35.2-41.8
37.9
35.1-4J..9
-
-
33.2
41.5
37.3
42.5
-
-
33.7
36.3
34.7
41.0
-
-
35.8
40.3
34.4
44.4
-
-
Secondary
current
(mA)
20
10
20
10
260
135
10
^5
20
20
260
50
60
33
30
85
250-260
240
65
40
25
90
250-260
245
Power
(kW)
0.7
0.4
0.8
0.4
-
-
0.3
0.2
0.7
0.9
-
-
2.0
1.2
1.0
3.5
-
-
2.3
1.6
0.9
4.0
-
-
Spark
rate
(SPM)
0
25-50
25-50
25-50
0
20
0
25-50
0
0-25
Unstable
0-25
0
0
0
0
0
0
0
0
0
0-10
0
0
- continued -
-------
TABLE 3. (continued)
Date Time
1/16/777 1345
1/16/777 1615
1.
2.
3.
4.
5.
6.
7.
8.
# 5
Spark limited in
Spark limited in
No. 3 TR off.
Center field of
No. 6 TR placed
Primary Primary Primary Secondary Secondary Spark
T/R voltage current Power voltage current Power rate
set (V) (A) (kW) (kV) (mA) (kw) (SPM)
1 190 24 4.6 35.4 75 2.7 0-25
2 190 14 2.7 - 40 - 0
3 150 10 1.5 - 30 - 0
4 217 39 8.5 45.0 135 6.1 0
5s 215 19.5 4.2 42.5-43.2 80 3.4 0
6 320 62 19.8 44.1 240 10.6 0
1 200 26 5.2 38.3 75 2.9 0
2 190 15 2.9 - 40 0
3 150 8 1.2 - 20 - 0
4 260 38 9.9 45.2 120 5.4 0
5 120 48-52 0.6 <9.5 210 2.0 Unstable
6 310 49 15.2 45.5 180 8.2 0-25
in manual 182 6 1.1 40.9 40 1.7 0
manual operation.
automatic operation.
Chamber No. 2 off.
in manual due to arcing in bus duct.
All TR's in manual, half of No. 3 off.
TR's 5 and 6 in
Voltage dividers
auto, others manual, half of No. 3 off.
moved from 2 and 3 to 5 and 6 .
-------
SECTION 4
TEST PROGRAM
The test program conducted at Kennecott's copper smelter at
Hayden, Arizona on January 11-16, 1977, was primarily concerned
with evaluation of the performance of the electrostatic precipi-
tator following the reverbatory furnace.
The test program was carried out by Southern Research
Institute and Radian Corporation as part of EPA Research Grant
R80955, administered by U.S. EPA's Industrial Environmental
Research Laboratory, Cincinnati, Ohio.
The primary objectives of the SoRI-Radian program were:
1. Characterization of the chemical species and phases at
the inlet and outlet of the precipitator
a. Particulates - Radian and SoRI
b. Noncondensables - Radian and SoRI
c. Gas analysis (Orsat) - SoRI
2. Evaluation of precipitator operation
a. Electrical operating data - SoRI
b. Particle size measurements - SoRI
c. Voltage-current characterization - SoRI
d. Resistivity measurements - SoRI
e. Computer model simulation - SoRI
The sampling locations were shown earlier in Figure 3.
The sampling schedule for the test period is shown in Table 4.
Since the reverbatory furnace was not operational on January 13,
no tests could be made on that date; however, the original test
plan was extended an additional day to conduct all tests ori-
ginally scheduled.
18
-------
TABLE 4. SAMPLE PORT UTILIZATION FOR TEST PROGRAM
Precipitator Inlet
Date
1/11/77
1/12/77
1/13/77
1/14/77
1/15/77
1/16/77
Port
5
1-2
3
4
4
5
6
Time
AM-PM
PM
PM
AM
PM
AM-PM
AM-PM
Organization
Radian
SoRI
SoRI
SoRI
SoRI
Radian
SoRI
Reverberatory furnace not
2
2-3
4
4
1-2
3
4
4
5
6
1-2
3
4
4
6
AM-PM
AM-PM
AM
PM
AM-PM
PM
AM
PM
AM-PM
AM-PM
PM
PM
AM
PM
AM-PM
SoRI
SoRI
SoRI
SoRI
SoRI
SoRI
SoRI
SoRI
Radian
SoRI
SoRI
SoRI
SoRI
SoRI
SoRI
Test
WEP
Impactor
Hi-Vol
Resistivity
Orsat
WEP
Ultrafine
Port
6
1
1
2
3
4
6
operational. No tests
Dltrafine
Impactor
Resistivity
Orsat
Impactor/
Ultrafine
Impactor/Hi Vol
Resistivity
Orsat
WEP
Trace element
Impactor
Impactor
Resistivity
Orsat
Trace Element
1
2
3
4
6
1
2
3
4
5
1
2
3
4
Precipitator Outlet
Time
AM-PM
AM
PM
AM-PM
AM-PM
AM-PM
AM-PM
Organization
Radian
SoRI
SoRI
SoRI
SoRI
Radian
Radian
Stack
Test Port Time Organization Test
Vapor Train
Orsat
Trace element
Impactor
Impactor
WEP
Vapor train
conducted.
AM
AM-PM
AM-PM
AM-PM
AM-PM
PM
AM-PM
AM-PM
AM-PM
AM-PM
AM-PM
AM-PM
AM-PM
AM-PM
SoRI
SoRI
SoRI
SoRI
Radian
SoRI
SoRI
SoRI
SoRI
Radian
SoRI
SoRI
SoRI
SoRI
Orsat
Trace element South AM-PM Radian WEP
Impactor
Impactor
Vapor Train
Orsat
Ultrafine North AM-PM Radian WEP
Impactor
Impactor
WEP
Orsat
Ultrafine
Impactor
Impactor
-------
SECTION 5
CHEMICAL ANALYSIS
TRACE ELEMENTS IN REVERBERATORY-FURNACE OFF-GASES
Sampling Methods
The task of sampling trace elements from gas streams enter-
ing and leaving the ESP was shared by Radian Corporation and
Southern Research Institute. The objective in having two organ-
izations participate in trace-element sampling was to obtain data
by different procedures for comparative purposes. The sampling
procedures employed by Radian had been used in characterizing
emissions from various sources in the past; the procedures
employed by the Institute, on the other hand, had had little
previous utilization but offered the potential advantage of
fractionating the sampled material as solids, condensables, and
vapors.
Radian used three types of sampling trains, as described
briefly below and in detail in Appendix B:
• One train was referred to as the "WEP" train, where
the acronym stands for wet electrostatic precipitator.
Sample gas was drawn through a Pyrex sampling nozzle,
then through the WEP, and finally through a series of
solution-filled impingers. Particulates and some
vapors were collected in the WEP. Vapors escaping the
WEP were caught in the impingers.
• Another train was specific for mercury vapor. Sample
gas was drawn through aqueous hydrogen peroxide to
remove sulfur dioxide and then through a plug of fine
gold wire, which retained mercury by amalgamation.
• The third train was applicable for vapors in general.
A glass-fiber filter inserted in the sampling duct
removed particulates. Vapors passing through the
filter were removed in a series of external impingers
having the same contents as the impingers in the WEP
train.
20
-------
Southern Research Institute also used three types of sam-
pling trains. The components of these trains are described
briefly below and in detail in Appendix C:
• One train was used at the ESP inlet for collecting
particulates and vapors. Three cyclone samplers in
series for particulates were inserted in the sampling
duct with a backup glass-fiber thimble. A glass-fiber
filter was then mounted in an oven at 120°C (250°F)
outside the duct to collect condensable vapors, and
it was followed by a series of impingers to collect
residual vapors. This sampling train is subsequently
referred to as a "CFI" train (the letters designate
cyclones, filters, and impingers, respectively).
• Another train—basically a simplification of the
first—was used at the ESP outlet. Only one cyclone
was inserted in the sampling duct, and it was followed
by a backup glass-fiber filter disc instead of a glass-
fiber thimble. This train is also referred to as a
"CFI" train.
• The third train was for mercury vapor. Sample gas
was drawn through a sodium carbonate bubbler to remove
sulfur dioxide and then through acidified potassium
permanganate to absorb the mercury. This train
yielded results that were much lower than those
obtained with the CFI trains. Later work in the
laboratory showed that the absorption efficiency of
the vapor train was poor. Thus, all of the data given
in this report for mercury were based on the results
obtained with the CFI trains.
Both Radian and the Institute collected samples at the inlet
and outlet ducts of the ESP, at locations previously shown in
Figure 3. In addition, Radian collected samples from the stack
at the 70-m (230-ft) level. Composite samples of particulates
and vapors were collected at the inlet and outlet of the ESP by
both Radian (with the WEP train) and the Institute (with the CFI
trains). Samples were collected at the stack location only by
Radian, using the WEP train. Radian used the vapor train only
at the ESP outlet, but Radian and the Institute collected mercury
in the special trains for this element at both the inlet and
outlet.
Because of the limited number of sampling ports available in
the inlet and outlet ducts, both organizations sampled from
single points at these locations. Sampling by traversing over a
number of points at the inlet and outlet was not practical
because of the need for operating other equipment in the limited
number of sampling ports available. Especially at the inlet,
the sampling location was not ideal, and the samples collected
21
-------
may not have been representative. The first plenum entering the
ESP is located immediately following the inlet sampling points,
and there was considerable uncertainty about the sampling
velocity needed for isokinetic conditions. Also, the low gas
velocity introduced the possibility of particulate stratification
in the plane of the inlet sampling ports. The sampling points
in the stack, on the other hand, were more nearly ideal. Here,
Radian sampled at a series of points along the diameter of the
stack.
Analytical Methods
Both Radian and the Institute determined the concentrations
of 19 elements in the samples removed from the gas streams enter-
ing and leaving the ESP. These 19 elements were as follows:
Silver, Ag Cadmium, Cd Iron, Fe Antimony, Sb
Aluminum, Al Cobalt, Co Mercury, Hg Selenium, Se
Arsenic, As Chromium, Cr Molybdenum, Mo Vanadium, V
Gold, Au Copper, Cu Nickel, Ni Zinc, Zn
Barium, Ba Fluorine, F Lead, Pb
In general, sample analysis required two types of opera-
tions: (1) dissolution of materials collected in the solid
state and (2) processing of solutions by one or more steps as
required to obtain quantitative determinations of the individual
elements. The procedures used by Radian and the Institute
differed in a number of respects. Details are given in Appendix
B for the methods used by Radian and in Appendix C for the
methods used by the Institute. Some of the highlights of the
experimental procedures are presented in the following para-
graphs, however.
Procedures Used by Radian—
Solid matter collected in the WEP was treated by either of
two methods. Perchloric acid digestion was used to dissolve
samples to be analyzed for most of the elements at low concentra-
tions. Lithium borate fusion was used as an alternative method
for most of the elements at higher concentrations; the cooled
melt was dissolved in hydrochloric acid and hot water.
Liquid samples from the WEP train (dissolved solids and
original liquor) were combined prior to analytical processing.
Similarly, the different kinds of solutions used in the impingers
were combined prior to analysis.
Atomic absorption spectrophotometry was the most commonly
used analytical tool. Some elements were determined by direct
injection into a flame or graphite furnace. Others were first
concentrated by extraction prior to determination by atomic
absorption. Fluorine was one element not determined by atomic
22
-------
absorption; this element was determined with a fluoride ion-
selective electrode. Selenium was another exception, being de-
termined by a fluorometric procedure.
Procedures Used by the Institute—
Solid samples collected in cyclones were routinely dissolved
by a procedure that involved a series of steps involving
additions of nitric, sulfuric, and perchloric acids. A few
samples in more abundant supply were treated by an alternative
procedure involving attack first by aqua regia and then per-
chloric acid. Sodium hydroxide fusion was used to prepare sam-
ples for fluoride determination.
Solids collected on filters presented a special problem,
since insufficient material was present for separation from the
filter media. These samples were extracted from the filters by
refluxing with aqua regia.
As a rule, the contents of the impingers were combined
prior to further processing. Most of the elements were deter-
mined by atomic absorption. Fluorine was an exception, being
determined with an ion-selective electrode as in Radian's
procedure. Selenium was another exception, being determined
fluorometrically as in Radian's work.
Analytical Results
Complete tabulations of the analytical results obtained by
Radian and the Institute are given in Appendices B and C.
Information abstracted from those tabulations is presented and
discussed in the sections of this report immediately following.
Total Elemental Concentrations—
The total concentrations of the trace elements occurring as
particulate and vapor that were found at the inlet and outlet
sides of the ESP are given in Tables 5 and 6, respectively.
These data are based on the results obtained by the individual
organizations for samples collected on the indicated dates.
The data obtained by both Radian and the Institute show
wide variations from day to day. These variations may be due
to deficiencies in sampling or analysis; however, there is no
way to assess the relative importance of such deficiencies and
actual changes in emission rates resulting from process varia-
tions, since the nature of the reverberatory furnace and pre-
cipitator operations were such that significant changes in
emission rates would be expected.
23
-------
TABLE 5. TOTAL CONCENTRATIONS OF TRACE ELEMENTS
AT THE ESP INLET DUCT
Date
Sampling train
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Jan. 12
WEP
Jan. 14
WEP
Jan. 15
WEP
Concentration , a
0.46
30
12
0.001
<7.0
4.3
0.064
0.33
250
86
280
_b
28
0.22
25
2.8
0.64
0.32
27
0.18
53
28
0.005
<0.7
13
0.47
1.0
590
82
710
0.0011°
43
1.6
36
3.4
2.0
0.43
5.7
0.45
16
12
0.001
<0.1
2.3
0.071
0.55
270
57
270
0.016C
13
0.28
14
1.8
0.76
0.47
22
Jan. 15
CFI
mg/m3
0.13
26
13
<0.001
0.070
1.2
0.05
0.52
330
34
250
0.011
13
0.49
4.4
1.0
1.0
0.07
23
Jan. 16
CFI
0.32
82
23
<0.001
0.38
9.4
0.08
0.88
960
60
840
0.021
31
1.5
25
3.5
3.0
0.17
54
a. Calculated for 21°C (70°F) and 760 mmHg.
b. Not determined.
c. Determined with Radian's special train for mercury vapor.
24
-------
TABLE 6. TOTAL CONCENTRATIONS OF TRACE ELEMENTS
AT THE ESP OUTLET DUCT AND STACK
Date
Sampling train
a.
b.
c.
Location
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Calculated
Jan. 12
WEP
Duct
Jan. 12
CFI
Duct
Concentration,
0.093
0.91
8.4
<0. 00004
<0.05
0.20
0.0020
0.069
3.6
69
2.3
_b
0.80
0.086
0.13
0.16
0.011
0.38
1.5
for 21°C (70
0.01
0.87
6.8
<0.001
<0.01
0.14
<0.02
0.03
8.1
21
3.2
0.008
0.61
0.04
0.35
0.15
0.07
<0.01
2.4
Jan. 14, 15
WEP
Stack
a mg/m3
0.14
0.57
8.5
<0. 00022
<0.05
0.23
0.0034
0.041
7.3
47
3.4
O.*0018c
1.1
<0.003
0.91
0.18
0.0027
0.43
1.9
°F) and 760 mmHg.
Not determined.
Determined
with Radian1
s special
train for
mercury vapor
25
-------
Samples were collected simultaneously by Radian and the
Institute at the inlet of the ESP on January 15 and at the out-
let on January 12. A comparison of the individual results
obtained for these samples shows reasonable agreement only in
rare instances. The differences may be due to sampling or
analytical deficiencies, or they may also reflect actual dif-
ferences in concentrations at the points used for sampling.
To facilitate observations from the data, average concen-
trations at the inlet and outlet were calculated and are pre-
sented in the second and third columns of Table 7. Averaging
cannot be entirely justified, for equal weight cannot be placed
on the significance of the individual sets of analytical data.
This point is illustrated, for example, by the fact that data
for the outlet duct were obtained at single points on 1 day
whereas the data for the stack were obtained by a traversing
procedure on 2 days. Even so, the following conclusions appear
to be valid:
• Inlet concentrations
- Copper and iron were the elements having the highest
concentrations, around 500 mg/m3.
- Six elements ranked next (aluminum, arsenic, fluorine,
molybdenum, lead, and zinc) with concentrations
between 10 and 100 mg/m3.
- Three elements were the next most abundant (cadmium,
antimony, and selenium) with concentrations between
1 and 10 mg/m3.
- Seven of the other eight elements (silver, gold,
cobalt, chromium, mercury, nickel, and vanadium)
occurred at concentrations below 1 mg/m3. The
remaining element—barium—probably was also below this
level.
• Outlet concentrations
- Fluorine was the most abundant element, at a concentra-
tion of about 50 mg/m3 or about 70% of the value at the
inlet.
- Arsenic increased to the second most abundant element
at the outlet from the eighth rank at the inlet.
- Copper and iron were reduced in relative abundance
from first and second at the inlet to third and fourth,
respectively, at the outlet.
26
-------
TABLE 7. COMPARISON OF TOTAL ELEMENTAL CONCENTRATIONS
AT THE ESP INLET AND OUTLET
Concentration,3 '
mg/m3
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Eg
MO
Ni
Pb
Sb
Se
V
Zn
Inlet
0.31
41
18
<0.002
-c
6.0
0.15
0.67
480
64
470
0.010
25
0.81
20
2.5
1.5
0.29
26
Outlet o
0.08
0.78
7.9
<0.001
<0.05
0.20
<0.02
0.046
6.3
45
3.0
0.005
0.85
<0.10
0.47
0.16
0.026
-c
1.9
Concentration
ratio,
utlet to inlet
0.26
0.02
0.44
-C
-C
0.03
<0.13
0.07
0.01
0.70
0.01
0.50
0.03
<0.12
0.02
0.06
0.02
_c
0.07
Estimated
ESP
ef ficiencyb
64
97
38
_c
_c
96
>82
90
98
2
98
30
96
>83
97
92
97
-c
90
a. Averages calculated from the data in Tables 5 and 6
Expressed for 21°C (70°F) and 760 mmHg.
b. Based on the data in the preceding column and an
estimated 40% in-leakage of air (based on data
in Table 11).
c. Not determined.
27
-------
The observations about relative concentrations at the
inlet and outlet of the ESP give a qualitative indication of
the effect of the ESP in controlling emission rates of the
various elements. A more nearly quantitative indication of the
ESP efficiencies for controlling the various elements is given
by the data in the last column of Table 7. These data are based
on the ratios of the outlet and inlet concentrations and an
estimated value of the in-leakage of air between the ESP inlet
and outlet. The estimated value of the in-leakage of air is
40% of the reverberatory off-gases; this figure is based on
analyses of flue gases at the ESP inlet and outlet, which are
summarized later in Table 11. From the estimated ESP efficien-
cies for the individual elements, the following conclusions may
be reached:
• Most of the elements were removed from the gas stream
by efficiencies in excess of 90%.
• Three elements (arsenic, fluorine, and mercury) were
exceptions to this rule, being removed by efficiencies
less than 50%. A reasonable inference from this
observation is that these three elements occurred to
a large degree as vapors; this conclusion is supported
by other data discussed in the paragraphs immediately
following.
• A fourth element (silver) was another exception.
There is no basis, however, for believing that this
element occurred as a vapor to any marked degree.
Apparent Vapor-Phase Concentrations—
ESP inlet— An analysis of the data obtained for the Insti-
tute's fractionated samples in the CFI train gives a basis for
estimating the fraction of each element that occurred in the
vapor phase at the ESP inlet. The data in question are given in
the tables in Appendix C. The analysis of these data consisted
of calculating the ratio of material collected outside the duct
(on the filter and in the impingers) to the total amount of
material collected.
The values of the ratios thus obtained are presented in
Table 8. They permit the identification of the following
elements as substances occurring to an appreciable degree in the
vapor phase: arsenic (42 to 66%) , fluorine (97 to 98%),
28
-------
TABLE 8. FRACTIONS OF TOTAL ELEMENTAL CONCENTRATIONS
APPEARING AS VAPORS AT THE ESP INLET
Date
Sampling train
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Jan. 15
CFI
Concentration ,
mg/m3
Totaia
0.13
26
13
<0.001
0.070
1.2
0.05
0.52
330
34
250
0.011
13
0.49
4.4
4.0
1.0
0.07
23
Vapor"
<0.02
0.22
8.6
<0.001
<0.04
<0.01
<0.03
<0.02
0.087
33
0.125
0.011
<0.06
<0.02
<0.03
<0.04
0.081
<0.06
0.084
Vapor
fraction
<0.15
<0.01
0.66
_c
-C
<0 .01
-C
<0.04
<0.01
0.97
<0.01
1.00
<0.01
<0 .04
<0.01
<0.01
0.08
-C
<0.01
Jan. 16
CFI
Concentration ,
mg/m 3
Totaia
0.32
82
23
<0.001
0.38
9.4
0.08
0.88
960
60
840
0.021
31
1.5
25
3.5
3.0
0.17
54
Vapor0
<0.02
1.8
9.6
<0.001
<0.04
<0.01
<0.03
<0.02
0.14
59
0.21
0.021
<0.06
0.064
<0.03
<0.03
0.20
<0.06
1.6
Vapor
fraction
<0.06
0.02
0.42
_c
-C
<0.01
-c
<0.02
<0.01
0.98
<0.01
1.00
<0.01
0.04
<0.01
<0.01
0.07
_c
0.03
a. Based on data in Table 5 (originally compiled from data in
Tables C-4 and C-5 in Appendix C). Calculated for 21°C
(70°F) and 760 mmHg.
b. Compiled from data in Tables C-4 and C-5 in Appendix C.
Also calculated for 21°C (70°F and 760 mmHg.
c. Not determined.
29
-------
mercury (100%), and selenium (7 to 8%). Each of these elements
except selenium was also found to be collected by the ESP with
low efficiency (see Table 7). On the other hand, silver cannot
be rated as a vaporous element despite its low collection effi-
ciency in the ESP.
ESP outlet—In principle, the data obtained by the Institute
at the ESP outlet could be treated in the same manner as the data
for the inlet. In actuality, such a treatment would not be
fruitful because of large relative uncertainties in the amounts
of material collected in the impingers at the outlet. However,
Radian collected a sample at the outlet under conditions designed
to collect vapors selectively (by use of the vapor train
described on page 20). The results of the analysis of this
sample are given in Table 9. For comparison, the average values
computed for the total elemental concentrations at the outlet
(from Table 7) are also given in Table 9, along with the ratios
of apparent vapor concentrations to total concentrations.
The information in Table 9 indicates that three elements
found as vapors at the inlet—arsenic, fluorine, and mercury—
occurred almost entirely as vapors at the outlet. This informa-
tion also suggests that some of the other elements—aluminum,
for example—occurred to a significant degree as vapors at the
outlet. Such a conclusion does not seem plausible, however.
Probably a better interpretation is that such elements occurred
as fine particulates that were collected along with the vaporous
elements.
Apparent Concentrations of Condensables—
An analysis of data for the Institute's samples from the CFI
train permits not only an estimate of the fractions of the ele-
ments in the vapor phase, as discussed above, but an estimate
of the fractions that condensed between the sampling duct and the
external filter at 120°C. The results of the data analysis for
the ESP inlet are given in Table 10. They show that for most of
the elements, less than 1% appeared as condensable matter and
that the maximum for any of the elements was 6% for arsenic in
one of the two samples.
Discussion
This discussion of the trace-element data is addressed to
two questions:
• What is the existing efficiency versus element of the ESP
at the Hayden reverberatory furnace?
• Could the efficiency be significantly improved in theory
if the option of operating the ESP at a lower gas tempera-
ture were exercised?
30
-------
TABLE 9. TOTAL AND VAPOR CONCENTRATIONS
OF ELEMENTS AT THE ESP OUTLET
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Concentrations,3 mg/m3
Totaio
0.08
0.78
7.9
<0.001
<0.05
0.20
<0.02
0.046
6.3
45
3.0
0.005
0.85
<0.10
0.47
0.16
0.026
-e
1.9
Vaporc
0.015
0.30
10
<0.0001
<0.02
0.00038
<0.0002
0.019
0.088
45
0.12
0.0044
0.0061
0.0028
0.0044
0.0027
0.00027
0.091
0.0048
Ratio , vapor
to total
0.17
0.38
1.29^
-e
-e
<0.01
-e
0.41
0.01
1.0
0.04
0.88
<0.01
-e
0.01
0.02
<0.01
-e
<0.01
a. At 21°C (70°F) and 760 mmHg.
b. From the third column of Table 7.
c. From the results of sampling by Radian with
the vapor train•
d. The anomaly of this ratio stems from the com-
parison of data from different sources.
e. Not determined.
31
-------
TABLE 10. FRACTIONS OF TOTAL ELEMENTAL CONCENTRATIONS
APPEARING AS CONDENSABLE MATTER AT THE ESP INLET
OJ
Date
Sampling train
Jan. 15
CFI
Concentration, mg/m3
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Total*
0.13
26
13
<0.001
0.070
1.2
0.05
0.52
330
34
250
0.011
13
0.49
4.4
1.0
1.0
0.07
23
CondensablesD
<0.003
<0.03
0.082
_c
— C
<0.001
<0.001
<0.001
0.011
_c
0.015
<0.001
<0.005
0.005
<0.005
<0.02
<0.01
<0.001
0.054
Condensable
fraction
<0.02
<0.01
0.01
-c
-c
<0.01
<0.02
<0.01
<0.01
-c
<0.01
<0.09
<0.01
<0.01
<0.01
<0.02
<0.01
<0.02
<0.01
Jan. 16
CFI
Concentration, mg/m3
Totala
0.32
82
23
<0.001
0.38
9.4
0.08
0.88
960
60
840
0.021
31
1.5
25
3.5
3.0
0.17
54
CondensablesD
<0.003
1.5
1.4
-C
-c
<0.001
<0.001
<0.002
0.015.
-c
0.046
<0.001
<0.005
0.064
<0.005
<0.03
<0.01
<0.001
1.6
Condensable
fraction
<0.01
0.02
0.06
-c
_c
<0.01
<0.01
<0.01
<0.01
-c
<0.01
<0.05
<0.01
0.04
<0.01
<0.01
<0.01
<0.01
0.03
a. Based on data in Table 5 (originally compiled from data in Tables C-4 and C-5 in
Appendix C) . Calculated for 21°C (70°F) and 760 rnmHg.
b. Compiled from data in Tables C-4 and C-5 in Appendix C. Also calculated for 21°C
(70°F) and 760 mmHg.
c. Not determined.
-------
With respect to the first of these questions, the analytical
data indicate that all elements except three identified as vapor-
ous elements (arsenic, fluorine, and mercury) are collected with
efficiencies exceeding 90% and that the elements accounting for
most of the inlet mass (copper and iron) are collected with effi-
ciencies around 98 to 99%. There may be some error in the quan-
titative aspects of these data owing to the uncertainty in the
air in-leakage factor, but the conclusions are qualitatively
correct.
With respect to the second question, it should be pointed
out that the gas stream entering the ESP averaged around 260°C
(500°F), whereas theoretically it might be cooled to a lower
temperature—say 120°C (250°F). The comparison of quantities of
condensed material and total material at the ESP inlet indicates
that the condensables represented only a small fraction of the
total elemental concentrations. Hence, it would appear that
lowering the ESP temperature would not give a substantial
increase in efficiency. On the other hand, a comparison of
condensed material and total material at the ESP outlet might
suggest that the absolute emission rate could be substantially
reduced by cooling the gas stream to 120°C. Unfortunately, the
poor accuracy of analytical data for the outlet precludes any
test of this point. It must be emphasized that this discussion
of condensables is based entirely on analyses of selected
elements, not including sulfur, for example. The condensation
of sulfur trioxide as sulfuric acid could increase the quantity
of condensables significantly. Also, other considerations—such
as loss of stack draft at a lower gas temperature—could weigh
against any marked reduction in ESP temperature.
ANALYSIS OF OTHER MATERIALS IN THE REVERBERATORY OFF-GASES
Southern Research Institute was responsible for determining
the concentrations of sulfur oxides, carbon dioxide, oxygen and
water vapor in the reverberatory furnace off-gases. Methods
used for this work are described in Appendix G. The results
are given in Table 11. These data are mainly of interest in
connection with the following points:
• Evidence of air in-leakage between the inlet and out-
let of the ESP. The concentrations of carbon dioxide
and water vapor were lower at the outlet than at the
inlet, whereas the reverse was true for oxygen. These
concentration differences are consistent with air in-
leakage at a rate of about 40% of the stack gases.
• Levels of sulfur oxides. Most of the data for sulfur
dioxide and sulfur trioxide are for the ESP outlet.
Here, the averages of the concentrations observed are
0.53% (5300 ppm) for sulfur dioxide and 0.0028% (28 ppm)
for sulfur trioxide. The relative concentration of
33
-------
TABLE 11. CONCENTRATIONS OF MISCELLANEOUS GASES
OBSERVED AT THE ESP INLET AND OUTLET
Concentration, volume %, found on
Gas
CO 2
02
Location
Inlet
Outlet
Inlet
Outlet
Jan. 12
™
6.9
7.1
_
9.6
7.7
Jan. 14 Jan. 15 Jan. 16
10.4 9.4
6.5
7.0
7.1
7.8
5.2 7.3
9.5
9.5
8.7
8.2
H20
SO;
SO;
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
-
7.5
_
0.62
0.56
0.36
_
0.0033
0.0037
0.0022
-
7.8
_
0.52
0.57
_
0.0025
0.0026
9.3
0.83
9.7
0.0036
34
-------
sulfur trioxide is surprisingly low in comparison with
relative concentrations observed in coal-fired electric
power plants.
ANALYSIS OF OTHER MATERIALS ASSOCIATED WITH THE REVERBERATORY
FURNACE
Other materials analyzed besides those in the reverberatory
off-gases were feed material for the reverberatory furnace and
dust removed from the ESP hoppers. The data are given as part
of the experimental record, but no interpretation or discussion
of these data seems to be needed in connection with the primary
analytical task of describing emissions from the reverberatory
furnace and the electrostatic precipitator.
Feed Material
A sample of reverberatory feed material was prepared as a
composite of individual samples collected on January 12, 14, 15,
and 16 and submitted for analysis by spark-source mass spectrom-
etry (SSMS) at Accu-Labs Research, Inc., of Wheat Ridge,
Colorado. The results submitted by Accu-Labs are given in
Table 12.
Hopper Dust
A sample of hopper dust was prepared as a composite of
individual samples that were deposited in the ESP on January 12,
14, 15, and 16. Portions of the composite were submitted for
analysis by Accu-Labs (SSMS), Radian (see Appendix B for
methods), and the Institute (see Appendix C). The results
obtained by the three organizations are compared in Table 13.
The results of analyses of the individual samples of hopper
dust, which were made by the Institute, are given in Table 14.
35
-------
TABLE 12. RESULTS OF SSMS ANALYSIS OF
REVERBERATORY FEED
Element
*Ag
*A1
As
*Au
B
*Ba
Be
Bi
Br
Ca
*Cd
Ce
Cl
*Co
*Cr
Cs
*Cu
Dy
Er
Eu
*F
*Fe
Ga
Gd
Ge
Concn, mg/g
0.015
>10
0.15
<0.0001
0.0017
0.14
<0. 00024
0.035
0.0034
2.4
0.014
0.020
0.070
0.39
0.047
0.0011
>10
0.0012
0.0023
0.0012
0.54
>10
0.00087
0.00096
0.0033
Element
Hf
*Hg
Ho
I
In
Ir
K
La
Li
Lu
Mg
Mn
*Mo
Na
Nb
Nd
*Ni
Os
P
*Pb
Pd
Pr
Pt
Re
Rb
Concn, mg/g
0.0028
Not reported
0.00035
Int. Std
—
<0.0001
>10
0.017
0.0029
0.00033
>5
0.17
0.60
0.14
0.0036
0.013
0.083
<0.0001
0.40
0.25
<0.0001
0.0045
<0.0001
Int. Std
0.016
Element
Rh
Ru
S
*Sb
Sc
*Se
Si
Sm
Sn
Sr
Ta
Tb
Te
Th
Ti
Tl
Tm
U
*v
W
Y
Yb
Zr
*Zn
Concn, mg/g
<0.0001
<0.0001
>10
0.048
0.0048
0.078
>10
0.0017
0.210
0.049
0.0019
0.00038
0.0033
0.0096
1.7
0.0027
0.00051
0.0058
0.038
0.014
0.0076
0.0016
0.085
0.60
* These elements were determined analytically in the reverbera-
tory off-gases.
36
-------
TABLE 13. RESULTS OF SSMS ANALYSIS OF A COMPOSITE OF HOPPER-DUST SAMPLES
u>
Element
Ag
Al
As
Au
B
Ba
Be
Bi
Br
Ca
Cd
Ce
Cl
Co
Cr
Cs
Cu
Dy
Er
Eu
F
Fe
Ga
Gd
Ge
Hf
Hg
Ho
I
In
Ir
K
La
Li
Lu
Mg
Mn
Concentration, mg/g
Accu-Labs
0.082
>10
4.1
0.00025
0.0017
0.180
0.0016
5
0.060
>5
2.0
0.053
0.049
0.39
0.37
0.011
>10
0.0032
0.0035
0.0013
1.2
>10
0.0065
0.048
0.16
0.0055
Not reported
0.0010
0.330
Int. Std.
<0.0001
>10
0.060
0.014
0.00050
>5
0.64
Radian
0.095
12.5
2.7
<0.0001
0.0035
3.1
0.53
0.24
217
0.39
300
0.00055
SRI
0.090
19.3
3.6
<0.0005
0.61
2.0
0.27
0.30
246
0.44
197
0.001
Element
Mo
Na
Nb
Nd
Ni
Os
P
Pb
Pd
Pr
Pt
Re
Rb
Rh
Ru
S
Sb
Sc
Se
Si
Sm
Sn
Sr
Ta
Tb
Te
Th
Ti
Tl
Tm
U
V
W
y
Yb
Zr
Zn
Concentration, mg/g
Accu-Labs Radian
>10 11
1.4
0.0036
0.050
0.055 0.18
<0.0001
2.4
>10 8.3
<0.0001
0.014
<0.0001
Int. Std.
0.061
<0.0001
<0.0001
>10
1.7 0.93
0.0027
0.93 0.25
>10
0.0083
>10
0.057
0.00087
0.00081
0.033
0.0064
>10
0.0094
0.00077
0.0058
0.070 0.098
0.12
0.026
0.0028
0.11
>10 12
SRI
8.2
0.17
6.6
1.6
0.96
0.047
16
-------
TABLE 14. RESULTS OF CHEMICAL ANALYSES OF
INDIVIDUAL HOPPER SAMPLES
Concentration, mg/g, on
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
MO
Ni
Pb
Sb
Se
V
Zn
Jan. 12
0.080
18.5
4.8
<0.0005
0.67
2.9
0.22
0.28
245
0.50
179
<0.001
9.4
0.22
6.8
1.7
0.62
0.042
20
Jan. 14
0.090
17.8
5.0
<0.0005
0.56
3.0
0.23
0.29
237
0.43
192
<0.001
9.2
0.22
9.4
1.7
0.72
0.046
19
Jan. 15
0.060
19.7
2.4
<0.0005
0.64
1.3
0.29
0.29
243
0.42
185
<0.001
7.1
0.22
5.2
1.1
1.1
0.050
13
Jan. 16
0.080
19.0
3.5
<0.0005
0.57
2.5
0.26
0.32
243
0.55
213
<0.001
9.1
0.16
7.8
1.3
0.64
0.042
18
38
-------
SECTION 6
PARTICLE SIZE DISTRIBUTION MEASUREMENTS
Inertial, optical and electrical mobility methods were used
to measure particle size distributions from 0.01 ym to 8.0 ym
diameter. Measurements were made at the inlet and outlet of the
electrostatic precipitator in vertical sampling ports. Both at
the inlet and outlet, hoppers were located directly beneath the
sampling points, and at both sampling locations, the ductwork
was approximately 20 feet deep. Because of flow disturbances
caused by the hoppers, it was not feasible to obtain a repre-
sentative sample for particle diameters greater than about 5.0
ym. Therefore, no traverses were made.
Cascade impactors were used for in_ situ sampling at the
inlet and outlet. At the inlet, modified Brink BMS-11 impactors
were used, and at the outlet, measurements were made with
University of Washington Mark III Source Test Cascade Impactors
(U of W). These two impactors are shown in Figures 4 and 5.
The collection stages of these impactors are usually too massive
to be accomodated in any type of field-portable balance sensitive
enough to detect weighing differences of 0.01 mg. Therefore,
light weight inserts were fabricated for these collection stages.
Glass-fiber filter mats as well as properly formed metal shim-
stock can be used. During preliminary tests conducted in
December, 1976, particulate matter samples obtained with cascade
impactors at the inlet and outlet were found to adhere well to
bare metal impaction surfaces. Therefore, during the test in
January lightweight metal collection stage inserts were used in
both inlet and outlet impactors. These inserts were fabricated
from 300 series stainless steel shimstock and were cleaned prior
to use in an ultrasonic cleaner with benzene, acetone, and
distilled water. Next, the inserts or "substrates" were baked
at 343°C (650°F) for one hour. After initial weighings the
substrates were dessicated until use. After use, the substrates
were dessicated for a minimum of 12 hours before final weighing.
Back-up filters used in the Brink and U of W impactors were acid
washed, in situ conditioned Reeve Angel 934 AH glass-fiber filter
material. These substrates were prepared according to the pro-
cedures given in E.P.A. report 600/7-77-060. After in situ
conditioning, the back-up filters were dessicated at least 12
hours before initial weighings were made. After use, the glass-
fiber filter material was again dessicated for at least 12 hours
39
-------
NOZZLE
t
PRECOLLECTION
CYCLONE
JET STAGE
(7 TOTAL)
COLLECTION
PLATE
SPRING
0700-14.1
Figure 4. Modified Brink BMS-11 cascade impactor used for
inlet particle size distribution measurements.
40
-------
JET STAGE O-RING
COLLECTION PLATE
INLET
\
FILTER HOLDER
COLLECTION
PLATE (7 TOTAL)
JET STAGE
(7 TOTAL)
0700-14.2
Figure 5. University of Washington Mark III Source Test Cascade
Impactor used for outlet particle size distribution
measurements.
41
-------
before final weighings were made. All inlet impactor substrates
and back-up filters were weighed on a Cahn model G-2 electro-
balance. All outlet impactor substrates and back-up filters were
weighed on a Cahn model 4100 electrobalance equipped with a
specially designed "weigh-below" chamber, built to accomodate the
75 mm diameter U of W impactor substrates. The weighing pre-
cision with these balances as used under field conditions is
approximately + 0.05 mg.
Since the particulate matter was cohesive, no particle
"bounce" problems were encountered. If the particles are not
cohesive and do not adhere well to the collection substrates,
it is possible for large particles to be re-entrained or "bounce"
from one stage to another, finally being caught by the back-up
filter. In such an instance, the back-up filter catch is not
comprised of particles whose diameter is generally smaller than
the D50 of the last impactor stage, and the filter catch is
meaningless for data reduction purposes. Here the back-up
filter catch masses are meaningful and these were included in
the calibrations for cumulative mass.
During each day of sampling, one impactor at each sampling
location was run as a "blank". The impactor was prepared as if
a normal sample were to be taken, but a filter was placed in
front of the impactor so that only clean flue gas could enter
and pass through. The impactor was placed in the duct, allowed
to warm up, and an impactor "run" was made at a flowrate and run
time commensurate with those of the other impactor runs made at
the same sampling location on that day. Afterwards, the sub-
strates were weighed and compared with their initial weights.
In this manner, background measurements were made to allow for
the effect of the flue gases on the impactor substrates and back-
up filters on a daily basis. Tables 15 and 16 show the results
of these blank impactor runs made at the inlet and outlet of
the precipitator. Averages were calculated each day to find a
correction which would be applied to the stage weights for that
day. Only one back-up filter blank was obtained at each sampling
location daily as opposed to several blanks for the impactor
stages. This single value was judged to be too uncertain for use
as a correction so a pooled correction value based on all back-up
filter blanks was used rather than daily averages as were used
for two stages. Also, since the first collection stage of the
U of W impactor (SI) is different in design from that of the
following collection stages (S2-S7),the four SI blank substrate
weight changes were averaged to obtain a pooled correction for
SI which was used for all outlet runs.
Helium pycnometer density measurements made on ash collected
at the inlet indicated a density of 3.13 gm/cm3. At the outlet,
the density was measured to be 1.84 gm/cm . This creates some
difficulty in interpreting the impactor data because the char-
acteristic stage cutoff diameters of the impactors are density
42
-------
TABLE 15 .
INLET MODIFIED BRINK IMPACTOR BLANK STAGE MASS GAINS
Date
Substrate Set #
Sample Time (min)
Temperature (°C)
Cyclone (mg)
SO (mg)
w SI (mg)
S2 (mg)
S3 (mg)
S4 (mg)
S5 (mg)
SF (mg)
SO-S5 avg.
(mg)
SF avg.
1-2-77
3N
30
243
NA
0.08
-0.04
0. 36
0.12
0.00
-0.08
-4.04*
0.0710.16
0.27±0.02
1-14-77
6N
30
232
NA
0.38
0. 38
0.32
0.44
0.40
0.36
0.28
0.38±0.04
1-15-77
9N
30
246
NA
0.18
0.28
0.11
0.02
0.18
0.06
0.24
0.15±0.09
1-16-77
13N
30
243
NA
0.24
0.28
0.46
0.36
0.16
0.24
0.28
0.29±0.11
*Excluded from SF average
-------
TABLE 16.
OUTLET UNIVERSITY OF WASHINGTON IMPACTOR BLANK MASS GAINS
Date
Substrate Set #
Sample Time (min)
Temperature (°C)
SI (mg)
S2 (mg)
S3 (mg)
S4 (mg)
S5 (mg)
S6 (mg)
S7 (mg)
SF (mg)
S2-S7 avg.
(mg)
1-12-77
96
120
177
1.18
0.28
0.33
0.39
0.29
0.31
0. 38
-1.28
0.33±0.05
1-14-77
93
135
168
-0.03
0.32
0.25
0.21
0.32
0.22
0.30
-1.48
0.2710.05
1-15-77
88
120
185
-0.07
0.23
0.25
0.20
0.24
0.21
0.34
-0.94
0.25+0.05
1-16-77
86
120
174
1.90
1.41
1.82
1.01
0.75
0.55
0.79
-1.53
1.06+0.48
SI avg. = 0. 36±0.74
SF avg. = -1.31+0.27
-------
dependent. It was concluded that the outlet density was the
appropriate density to use for the fine particles as the outlet
aerosol was comprised predominantly of these fine particles.
The inlet density was assumed to be more representative of large
particles, these large particles dominate the inlet aerosol on
a total mass basis and are much more effectively collected by
the ESP than the fine particle fraction of the inlet aerosol.
Because the interest in this program centered on the ESP col-
lection characteristics of the more difficult to remove, fine
particle fraction of the aerosol, the particle density obtained
from the outlet sample was used in making the cut size calcula-
tions for the impactors. This procedure will introduce errors
in the fractional efficiencies for the large (say >5 y diameter)
particles but probably represents the best method for treating
the data on fine particle emissions. (The tacit assumption is
made that a particle does not change in size or density in pas-
sing through the ESP). The magnitude of the size shift resulting
from the density assumption can be seen in a later discussion of
the ultra fine particulate data. Any particle formation or
growth as a result of cooling of volatile components of the gas
stream is also ignored in the analysis of these data.
The particle size distributions described in this portion
of the report are impactor measurements which have undergone
treatments by a complex computer data reduction procedure. The
computer programs involved were developed by Southern Research
Institute under E.P.A. Contract 68-02-2131*. .This set of com-
puter programs performs fits to cumulative mass distribution
data from each impactor run. Then differential size distributions,
which provide measures of concentrations within small size
Intervals, are derived by differentiation of the fitted cumula-
tive distribution curves. A set of statistical programs averages
groups of these data, determines confidence intervals, and plots
the results on a computer controlled plotting system.
In Appendix D, computer generated data sheets are repro-
duced. These data sheets give the corrected stage weights, run
conditions, stage by stage cumulative masses, and differential
size distributions calculated from the data for each inlet and
outlet impactor run.
At the inlet the Brink impactors were used to obtain single
point samples in port 2. Flowrates were chosen so that nozzle
inlet velocities were isokinetic at the average flue gas
velocity. The sampling point was 6 feet deep into a 20-foot
flue with a 14-foot deep hopper underneath the duct. Inlet
blank runs were usually run in the mornings in ports 1 or 3.
* "A Data Reduction System for Cascade Impactors", McCain,
Clinard, Felix, and Johnson, EPA-600/7-78-132a, July, 1978.
45
-------
The inlet "real" runs were usually made in the afternoon because
other equipment was using all of the available impactor ports
in the mornings. A total of three "real" runs were made each
day along with the one blank run described above. Typical inlet
sampling times were about 30 minutes.
The inlet impactor runs show a slight trend toward increas-
ing cumulative mass loading below 8ym diameter, with calendar
progession. Differences in the differential distribution curves
are slight, and there was no statistically significant difference
in these curves for the first three days of testing. However,
for 1/16/77 there does appear to be a significant, but small,
increase in emissions in the 1.0-2.0ym size range. The day to
day variations were small and,within the scatter of the data,
not significant. Therefore, for the purposes of penetration-
efficiency calculations, all the inlet runs were pooled and
averaged. Figure 6 shows the averaged cumulative mass curve
obtained from this average. As for all the particle size distri-
bution curves presented in this report, 50% confidence intervals
(probable error of the mean) are shown. Figure 7 shows the
differential size distribution for the pooled inlet data. All
particle diameters are based on an ash density of 1.84 gm/cm3,
and are Stokes' diameters. Examination .of the dM/dlogD curve,
Figure 7, shows a minimum in the mass emissions occurring near
l.Oym with another decrease indicated for particle diameters less
than 0.25ym. The calculated inlet mass loadings for each impac-
tor run are shown in Table 17. Note that these mass loadings
cannot be compared with other data since the larger particle
sizes (>5.0ym) could not be representatively sampled by the
impactors because of isokinetic sampling limitations and inabili-
ty to obtain complete traverses of the duct. For presentation
purposes data plotting was cut off at lOym; however, the data
are probably not accurate above 5ym. A significant portion of
the particulate was contained in particles larger than lOym.
Thirty (30) to sixty (60) percent of the inlet particulate was
included in particles larger than 8ym even near the top of the
run of deep, low velocity ducting.
At the outlet, U of W impactors were used to obtain single
point samples in ports 4 and 5. Impactor flow rates were chosen
to be isokinetic at the average duct velocity at the sampling
location. The sampling points were 10 feet down in a 23-foot
deep duct, with a 20-foot deep hopper underneath the duct. A
total of three real runs was made each day along with the one
blank run described previously. The average sampling time was
120 minutes.
The outlet size distributions show poor agreement on a day
by day basis. These differences are most likely tied to the
operation of the ESP, especially the electrical conditions ob-
served in the ESP TR sets. These differences make it unreason-
able to average the impactor runs over the four days of testing
46
-------
"3
H-
CUMULATIVE MASS LOADING (MG/ACM)
n
CTl
9,
O P-
Hi P)
P- 0
Cb ft
fD O
3 t-i
O
c
P- 3
3 en
rt
0 «^
hj fD
< H
pi en
i i £*
en en
PJ en
H ft
fD O
en fD
zr en
O
SI t3
3 PJ
' 3-
p-
n
fD
QJ
P-
P)
3
fD
ft
l-i
•
<_n
O
O
H /D M-
M H
» n .
1 fi :
c
^ D , % •
PI t~~j i— *• •
ft *~ • /— > "
p- ^. V.J
^ ^^* c?
fD ^.
P^
3 L
en IT
en -r-
i — i
O
in /-•>
^>, ^^ •
* H o:
uQ 1 1 Wv
•* *•« r^
91 n
j
- H o c
Ml 1 1
ru »->•
p- f~M IL ji M A "T"T\ lf~ i j A r^f"* i f~\ A I~»TK ii~< / r"*r^ x A f~*r~ N
fD
ft
-------
§
SMELTER TEST GRAM) AVERAGE JMMff 1977
HC = 1.84
1
I
1
I
» i i i mil 1—i i i mil 1 i i i mil
10"1 10° 1O1
PARTICLE DIAMETER (MICROMETERS)
Figure 7. Inlet differential mass concentration (dM/dlogD),
averaged over all inlet impactor runs versus Stokes
particle diameter. 50% confidence intervals are shown.
48
-------
TABLE 17.
CALCULATED IMPACTOR MASS LOADINGS FOR THE INLET
SAMPLING LOCATION. NUMBERS IN PARENTHESES
REFER TO THE POWER OF 10 MULTIPLIER
Run Code
KCCI-1
KCCI-3
KCCI-4
£ KCCI-5
KCCI-7
KCCI-8
KCCI-10
KCCI-11
KCCI-12
KCCI-14
KCCI-15
KCCI-16
Date
1/12/77
1/12/77
1/12/77
1/14/77
1/14/77
1/14/77
1/15/77
1/15/77
1/15/77
1/16/77
1/16/77
1/16/77
Time Port 'No.
1257
1423
1500
1005
1252
1409
1315
1357
1448
1255
1341
1440
2
2
2
2
2
2
2
2
2
2
2
2
Grains/ACF
2
2
1
1
3
1
3
4
1
4
2
5
.0202
.5793
.5953
.6866
.6300
.6031
.1218
.4437
.4994
.6350
.9486
.5007
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
(-1)
<-D
Grains/DNCF*
4.
5.
3.
3.
7.
3.
6.
9.
3.
9.
6.
1.
0639 (-1)
1887(-1)
2092 (-1)
4039(-1)
1684(-1)
1647(-1)
2645(-l)
1754(-1)
0960(-1)
6129(-1)
0825(-1)
1347(0)
mg/ACM
4
5
3
3
8
3
7
1
3
1
6
1
.6229(2)
.9024(2)
.6506(2)
.8595(2)
.3067(2)
.6685(2)
.1438(2)
.0169(3)
.4312(2)
.0606(3)
.7473(2)
.2588(3)
mg/DNCM*
9.2996(2)
1.1873(3)
7.3436(2)
7.7893(2)
1.6404(3)
7.2419(2)
1.4335(3)
2.0996(3)
7.0848(2)
2.1997(3)
1.3919(3)
2.5967(3)
*Normal, or Engineering Standard conditions are defined to be 21°C (70°F) and
760 mm (29.92in) Hg atmospheric pressure.
-------
so individual averages were made for each day. Figures 8-15
show the cumulative and differential distribution curves for each
day of testing on a Stokes diameter basis (based on a density of
1.84 gm/cm ). Qualitatively, the curves show similar behavior
for particulate emissions as a function of particle diameter.
Data from January 12, 14 and 15 all show that the emissions
increase with increasing particle diameter in the 0.5 to O.Sym
size range, reaching a maximum near O.Sym and then decreasing
to a minimum near 2ym. Similar behavior also is seen in the
results of January 16 except that the sizes at which the maximum
and minimum occur are shifted toward slightly smaller values.
A similar behavior was observed in the inlet distributions except
the maximum in that case was located at about O.Sym and the
minimum near lym. Particle growth by sublimation or condensation
could result in this difference in the inlet and outlet behavior.
The calculated outlet mass loadings for each impactor run are
shown in Table 18. Again, these values are probably not repre-
sentative of the entire duct since vertical concentration
stratification was likely to have been severe for large
particles.
Penetration-efficiency estimates were derived from the
impactor data using the grand average 'for the inlet and daily
averages for the outlet. Figures 16 through 19 show these
penetration-efficiency curves. Fifty percent confidence inter-
vals (probable errors) are shown about each efficiency point.
The structure of these curves differs from day to day. Since
the inlet particle size distribution appeared to be fairly
stable from day to day, these differences are presumed to result
from the daily variation in the behavior of the precipitator.
Inlet and outlet ultrafine particle size distribution meas-
urements were made with both electrical mobility and optical
methods. Two types of mobility analyses were considered for
this test: (1) diffusional methods and (2) electrical mobility
methods. Because of its compactness and short measurement time
(29.5 kg and 2 minutes) as compared with the diffusional method
(136 kg and 2 hours), the electrical mobility method was selected
for use on this test. The instrument used was a Thermosystems
model 3030 Electrical Aerosol Analyzer*. The electrical mobility
method operates by placing a known charge on the particles and
precipitating these particles under precisely controlled condi-
tions. Size selectivity is obtained by varying the electric
field in the precipitator section of the mobility analyzer.
Charged particle mobility is monotonically related to particle
* Liu, B.Y.H., K. T. Whitby, and D.Y.H. Pui. "A Portable Electri-
cal Aerosol Analyzer for Size Distribution Measurements of Sub-
micron Aerosols". Presented at the 66th Annual Meeting of the Air
Pollution Control Association, Paper No. 73-383, June, 1973.
50
-------
TABLE 18.
CALCULATED IMPACTOR MASS LOADINGS FOR THE OUTLET
SAMPLING LOCATION. NUMBERS IN PARENTHESES
REFER TO THE POWER OF 10 MULTIPLIER
Run Code
KCCO-1
KCCO-2
KCCO- 3
KCCO-6
KCCO- 7
KCCO- 8
KCCO- 9
KCCO- 10
KCCO-11
KCCO- 13
KCCO- 15
KCCO- 16
Date
1/12/77
1/12/77
1/12/77
1/14/77
1/14/77
1/14/77
1/15/77
1/15/77
1/15/77
1/16/77
1/16/77
1/16/77
Time
1015
1015
1435
1016
1145
1145
1000
1000
1410
0915
1345
1345
Port No.
5
4
5
4
5
4
4
5
4
4
4
5
Grain s/ACF
9.
5.
6.
9.
8.
6.
1.
1.
1.
1.
4.
5.
6698(-3)
5090(-3)
641K-3)
474K-3)
3099(-3)
3370 (-3)
0477(-2)
3739 (-2)
2456(-2)
5550(-2)
2665(-3)
1644 (-3)
Grains/DNCF*
1.
9.
1.
1.
1.
1.
1.
2.
2.
2.
7.
9.
6627(-2)
4724(-3)
1635(02)
6302)-2)
4659(-2)
1179(-2)
8425(-2)
416K-2)
2450(-2)
7467(-2)
5832(-3)
1791 (-3)
mg/ACM
2.
1.
1.
2.
1.
1.
2.
3.
2.
3.
9.
1.
2128(1)
2606(1)
5197(1)
1680(1)
9016(1)
4501(1)
3976(1)
1439(1)
8504(1)
5583(1)
7633(0)
1818(1)
mg/DNCM*
3.
2.
2.
3.
3.
2.
4.
5.
5.
6.
1.
2.
8048(1)
1676(1)
6624(1)
7305(1)
3545(1)
5581(1)
2164(1)
5289(1)
1374(1)
2854(1)
7353(1)
1005(1)
*Normal, or Engineering Standard conditions are defined to be 21°C (70°F) and
760 mm (29.92 in) Hg atmospheric pressure.
-------
-icr1
DUILET 9ELTER TEST -1 Mfr77
W) = 1.8431/0:
101-
10°
h
<
-"-ID"3
H 1 1 I M I H
1 ! I I I I H
10"
icf
101
PARTICLE DIAMETER (MICROMETERS)
Figure 8. Outlet cumulative mass loading, averaged over the
first day of testing, versus Stokes particle diameter,
50% confidence intervals are shown.
52
-------
SB.IH1 IBT -
no = 1^4 am
101--
10P-
H
I
Hill1
10
rl
H 1—I MINI
10P
H 1—I I I II
101
PARTICLE DIAMETER (MICROMETERS)
Figure 9. Outlet differential mass concentration (dM/dlogD),
averaged over the first day of testing, versus Stokes
particle diameter. 50% confidence intervals are shown.
53
-------
OUTLET 3B.TER TEST - 2 M4-77
a
101--
uf
I . .
J-icr3
10
1
H 1—I I I I I II
10°
H 1—I I I I I H
101
PARTICLE DIAMETER (MICRDKCTERS)
Figure 10. Outlet cumulative mass loading, averaged over the
second day of testing, versus Stokes particle diameter.
50% confidence intervals are shown.
54
-------
DUItET SaiER TEST - 2 i-H-77
no = 1.84 awr
101--
i
H 1—II M Ml
1 1—I I II I H
icr1 10° lo1
PARTICLE DIAMETER (MICROMETERS)
Figure 11. Outlet differential mass concentration (dM/dlogD),
averaged over the second day of testing, versus Stokes
particle diameter. 50% confidence intervals are shown,
55
-------
-icr1
omii aara TEST -1 1-15-77
M
a
in
101--
10°-
H 1—I MINI
1—I I I I I l|
10"1 10° 1O1
PARTICLE DIAMETER (MICROMETERS)
Figure 12. Outlet cumulative mass loading, averaged over the third
day of testing, versus Stokes particle diameter. 50%
confidence intervals are shown.
56
-------
amn sera IB! - 3
101-
d
±cP
I
) '
I
H 1 1 I I I I l|
ICT1 10° 101
PARTICLE DIAMETER (MICROMETERS)
Figure 13. Outlet differential mass concentration (dM/dlogD),
averaged over the third day of testing, versus Stokes
diameter. 50% confidence intervals are shown.
57
-------
IWRETaaTHlEST - 4 Mfi-77
00 = 1.0401/0:
10E-r
10°-
i
H 1 1 I I I I M
—-i n~i
b
\
tD
-•-icr3
—I—I—I I I 11II
10"1 10°
PARTICLE DIAMETER (MICROMETERS)
Figure 14.
Outlet cumulative mass loading, averaged over the fourth
day of testing, versus Stokes particle diameter. 50%
confidence intervals are shown.
58
-------
OUTLET aCLTEP TEST - 4 1-16-77
WO : 1.84 QMT
- 101--
1 1 I I M
icr1 10°
PARTICLE DIAMETER (MICROMET
101
Figure 15. Outlet differential mass concentration (dM/dlogD),
averaged over the fourth day of testing, versus Stokes
particle diameter. 50% confidence intervals are shown.
59
-------
PERCENT EFFICIENCY
CTi
O
2 U) Nl CD LDLD LD ID • • »
•2 -r, M.° OO O Ul CD ID Ul CD CD
C T] "^
fD ^^ ^-1
o> -H
• 1 1
l~1
0) Mi T) 1 —
H 0 fD !_
fD n 3 n
fO
co rt rt
3" 3" K Q
O fD 0) (__|
3 Ml P- ^>
• P- O ^
w i m
rt (D |
O.S m £:
&* H* ^n v j
•< o -^J XJ
O fD
Ml 3 __
o 2
£^ H
cn < fj
rt fD •TI
p. h( XU
3 cn Q
iQ C ^
' w m
CO 1
ui rt III
o O I*
* fo H3 "
o cn \J] . . -
O \~s }-*• •
3 1) n-
Mi 0) VV
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Illllllltlll 1 ' ' '
i i i i i i i i i | i i i i r r r r i | i i i r | i i | 1 1 1 r | 1 1 1 1 | 1 1 | 1 | n m
ff P
5 a
?
a **" T
• t^™^*™^"^ P^
^^ La
h fr
*^ ••"
. 1" * 1 *** "T
| 9 | ^
i • i H
• _9 i H
H»H 9
~ • — ^ • 1
m
*^ Hj
K^l "Tl
HH H
TL n
1*1 ii
n
MM £H
MM 2
i-!^7 ^
• 1 9 1
1 * '
1 * '
• | • |
1 1 1 i 1 1 1 1 1 1 1 1 M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
i i i i | i i i i i r i r r t'l i F i [ i' i' r i1 | i r | r r i i | r r r i | i i | i |
Sir! ^wru K ui rup-ooo
n>p- OOOO o..
3 £ ui m p
fD fD
£& PERCENT PENETRATION
rt 0)
fD 3
i-! fD
< ft
(a fD
-------
PERCENT EFFICIENCY
H-
i m M HI
mNm m CD U tfl • • •
^ — p.O O O O CJ1 CD 03 01 CD CD
£ °
h( O (D ~H
fD h( 3 M
w rt rt Q
3* tT* H |
O (D (ii rn
Z rt FI
3 to H-
• fD O i— i
0 3 LJ
o i M
3 fD >J
CL Ml £
O.H- ?
(DO r*
0 S F
^kQ ^
i
^
rt ~
fD < /-%
CO fD 9
rt i-i ^
H- CO f~J
3 C J-
lQ CQ I
• Tr
w p
ui o 2
OX »
CO —
O IT
O t3 M
H- rt CD
•
•
1 V
Oi H*' \^y r^^ *
(DO O -
3 H Si
0 fD "^h
M. a
fc S
* 9
_ i
nj •»
" l™1^*1— H pi
l«nAM-Mi '^^ *••
^~^^^"^ 1 ^ 4m
\ t \ Jf* ™
^ ^J -—
5""
I • i H
'- ':::7^l |
H-»H U
~n
i^^ ™n
M H
C
M m
*** n
!•( JJ
M
M
1*1
h*H
w fil ni LA ^n ni i v /~*t <^> x—«
CD ^SXI^ f^ Ul IUI—N^XJV^;
aOOOO OB.
ft 3
fD (D
ft)
en
RJ
PERCENT PENETRATION
-------
PERCENT EFFICIENCY
CTi
N)
H-
| CD tn tr
fl>
?
^ I &
OJ Mi Jg) ^Q |i
(D K 3 H
w rt rt .—,
tr tr n • U
O (D ft) |
Sct£ m
.3-0
n i D
H> >
O fl> — 1
M,3 rS (-* .
0) <
ft (D x-^
3'S 2Z
^g n
o o Q
3 ^ F ^
&ct ff
(D H- UI it*
3 n ^ r~ •
O M O "
(D fl> ^±
H-CL ?t
3 H- V.
SNCD CD CO CDU3- ••
OO O U\ CDUJCnODtl
> * i
*T
• - -
"^ ^MM
i-JlT^
* 1-^—1
M
: ^
^•UJry K en RJPOOO
) O O O • • •
$»
s
ui
ru
0)
PERCENT PENETRATION
-------
PERCENT EFFICIENCY
CO
c LD ID CD
M rn rn m
mNm m CD mtD-??
E -TJ p.° O O O Cn CD CD Ul CD CD
3n °'~
Oi Hi ^ 3 **
M 0 (D H
(D hf 3 H
(D ,__
CO ft ft [ |
3-3- H H
O (D 0) I—
s: rt Fl
3 Hi H-
• O O ,—,
C 3 D
H 1 L_J
n- ro r"1
tr M» >
^ ^^
pf H- aqfe
ji» o n
n> m U-X
s,| g S-
rt
ro < .^
CO (D ^«
H- CO I_J
3 C pj
w 2
rt t 1
Ul O ^
O tf db
dp fD | '
CO _J
° m
3o> ~n
HI i-i /y
H- rt U) . . •
o, i-*- ^^^ n^ •
(DO pi -
3 I-1 S^T
»
* N^^^
H«H
I j |
' l"*^
W^Mj
» ^^^
1 • 1
1 • 1
_
^"*~" "^
^
1^ 4
l—^"-"" 4
,
_' ^
^"~"*~~~H
H""*'""^
1 •
, ^
• * i
^^M^MMM^
S" &WW K m rupooo
aOOOO ...
3 •» cn ru ^
rt 3
(D (D
3 If PERCENT PENETRATION
0) if
i
CO
-------
diameter in the operating range of the instrument (0.015 to 0.3
ym particle diameter). A diagrammatic representation of this
device is shown in Figure 20. A Royco PC 225 mainframe and
PC 241 optical sensing unit were used to make the optical single
particle light scattering measurements. The useful particle
size range of this instrument is approximately 0.3 to 2.0ym
(PSL). Particle diameters reported here for the optical counter
are not based on equivalent polystyrene latex particle size cali-
brations as is the usual practice. Rather, a horizonital elu-
triator was used with the optical counter to determine particle
sizes based on gravitational settling. These "settling" diame-
ters are the diameters reported.
Neither of the instruments used in ultrafine particle dis-
tribution determinations can tolerate raw flue gases as sample
streams nor can they cope with the particle concentrations found
in flue gases. Thus, ultrafine measurements are based on ex-
tractive sampling with a metered sample being diluted with clean
dry air, to both condition the sample and reduce the particle
concentrations to levels within the operating limits of the
instruments. The required dilution typically ranges from 10:1
to >1000:1 depending on the particle source and the location of
the sampling port (i.e., upstream or downstream of the control
device). A diagrammatic representation of the Southern Research
Institute Sample Extraction and Dilution System (SEDS) in shown
in Figure 21.
Inlet ultrafine measurements were made on January 12 and
14, 1977. A total of approximately eight hours of usable data
was taken during these two days. The SEDS was used in port 2
and sampling was not done concurrently with the impactors. The
sample was extracted from a depth of 3 to 4 feet. No traverses
were made.
Figure 22 shows differential size distributions on a number
basis for the inlet for January 12 and 14. These results repre-
sent a grand average for this period. Similar data obtained
with impactors are also shown for comparison purposes. As can
be seen, the ultrafine and impactor data do not compare well at
the inlet. There could be several reasons for this. First, the
samples were single point samples taken at different depths. If
there were large concentration gradients or severe stratifica-
tion, then a poor comparison might result; this, however, is
unlikely. Second, the ultrafine and impactor samples were not
taken at the same time. Third, there may be cyclic variations in
the operation of the reverberatory furnace which have a time
period longer than the 30 minute sampling time for the impactors.
Fourth, unrecognized sublimation/condensation effects from As202
and/or H2S04 may have influenced the data through phase changes
resulting from the sampling and measurement processes. Day to
day reproducibility in the ultrafine inlet data appeared to be
good.
64
-------
CONTROL MODULE
ANALYZER OUTPUT SIGNAL
DATA READ COMMAND
CYCLE START COMMAND
CYCLE RESET COMMAND
AEROSOL FLOWMETER READOUT
CHARTER CURRENT READOUT
CHARGER VOLTAGE READOUT
AUTOMATIC HIGH VOLTAGE CONTROL AND READOUT
ELECTROMETER (ANALYZER CURRENT) READOUT
. TOTAL FLOWMCTER READOUT
-fc EXTERNAL
-to DATA
—I ACQUISITION
TO VACUUM PUMP
3630-043
Figure 20.
Diagrammatic representation of the T.hermosystems
Model 3030 Electrical Aerosol Analyzer (EAA).
-------
TIME
AVERAGING
CHAMBER
CTi
CHARGE NEUTRALIZER
PROCESS EXHAUST LINE
CHARGE NEUTRALIZER
CYCLONE
ORIFICE WITH BALL AND SOCKET
JOINTS FOR QUICK RELEASE
SOX ABSORBERS (OPTIONAL)
HEATED INSULATED BOX
RECIRCULATED CLEAN, DRY, DILUTION AIR
o
FILTER BLEED NO. 2
COOLING COIL
PRESSURE
BALANCING
LINE
DRYER
—-JX3 BLEED NO. 1
MANOMETER
Figure 21. The Sample Extraction and Dilution System (SEDS) designed
by Southern Research Institute. Shown in diagrammatic form.
-------
DIET DWCniHtTWfBE OMWBDH
Ml = 141401/0:
1014-
r f
DC
lO10^
109,
ICf,
105-
A\«
9Impactor
o Electrical Mobility
A Optical
.. Impactor data for inlet particle
density of 3.13 gm/cm3
ic
s
» I I I Hill 1 Mill
1O
"1
JJCP
i i i mi
101
i i i HUH
PARTICLE DIAMETER (MICROMETERS)
Figure 22. Inlet differential number concentration (dN/dlogD), averaged over
inlet tests conducted on January 12 and 14, versus Stokes particle
diameter. Data taken by Cascade impactors, optical counters and
electrical mobility measurements are shown for comparison. 50%
confidence intervals are shown.
67
-------
Outlet ultrafine particle size distribution measurements
were made in port 2 on January 15 and 16/ 1977. A total of
approximately seven hours of usable data was taken. The sampling
point was 3 to 4 feet deep into the duct. As with the impactor
data, in contrast to the behavior at the inlet, the outlet
ultrafine particle size distribution data differed greatly from
day to day, therefore, the agreement between impactor and ultra-
fine data was examined on a daily basis. Figures 23 and 24 show
the outlet dN/dlogD impactor and ultrafine measurements compared
for each day's outlet testing. In Figure 20, the agreement is
much better than in Figure 24. Any agreement may be fortuitous
since the same sampling problems occurred at the inlet. In
addition, the impactors and the ultrafine system were using
different sampling ports.
Penetration-efficiency estimates in the ultrafine particle
size range were calculated by averaging the inlet data taken on
January 12 and 14 and comparing this with the outlet data taken
on January 15 and 16. This assumes that the inlet concentrations
and size distribution remains constant throughout the test
series. This assumption appears to be generally valid in this
case based on the reproducibility of the ultrafine data during
the time it was taken and on the relative invariance of the
impactor data. The lack of two instrumentation systems for ob-
taining ultrafine data necessitated obtaining the data sequen-
tially as was done here. Figures 25 and 26 show the calculated
efficiencies versus particle size. Also plotted on Figures 25
and 26 are impactor-based efficiencies for the purposes of com-
parison.
68
-------
amiT DfKIDHLIMFBC WHOS*. 1-15-77
M) = 1.84 OUT
o
.10"Nr
107-
10*
• Irapactor
o Electrical mobility
A Optical
b
\
10
"2
> M in
1O
'1
i I I Ml
10°
I
101
Iff
PARTICLE DIAMETER (MICRDMET
Figure 23. Outlet differential number concentration (dN/dlogD), for January 15,
versus Stokes particle diameter. Data taken by Cascade impactors,
optical counters, and electrical mobility measurements are shown for
comparison. 50% confidence intervals are shown.
69
-------
OUTLET WNnaHJLTMFINE CBHBSDH, Mfi-77
00 = 1JM QMI
1013.
tfl
u
106
• Impactor
o Electrical mobility
A (Optical
I I M
1O"
10
-i
I M II
10P
i i i IIIIH—ill mill
101
PARTICLE DIAMETER CMICROMET
Figure 24. Outlet differential number concentration (dtt/dlogD), for January 16,
versus Stokes particle diameter. Data taken by Cascade impactors,
optical counters, and electrical mobility measurements are shown for
comparison. 50% confidence intervals are shown.
70
-------
PENETRATION-EFFICIENCY
BPOBHJLTROC OWIRISDN. 1-15-77
Hfrl.84
§
u
M
i-,
\-
tt
CIDaD'
99 :
•
9B:
•
95-
90:
«
i
*
80^
70:
60:1
"I
4O +
-
•
^ • Impactor
c Ultrafine T
; i \\
i
TT
[ il1!
- I I :
ITT^ ;
! I j
' I!' '
*d =
-, • ' ' ' i iii| f — I I 1 1 Mil 1 — r 1 1 1 1 1 1
;0.5
L±
-5 M
<
^ g
•10 y
• h-
'ao u
3O
4O
5O
PD
10- ID'1 10° 101™
PADTTn C nT A i tr— rr— ri n j-r<-~i-»— .. ..— .-.— _^_ .
Figure 25.
Penetration efficiency versus Stokes particle diameter
for January 15. Data taken by Cascade impactors and
electrical mobility methods are shown. Inlet data used
in these calculations were taken on January 12 & 14.
50% confidence intervals are shown.
71
-------
PE1SETRATIDN-EFFICIENCY
OHWEBU 1-16-77
MJ=U4
99.5-
99 :
9B:
& 95-
M
U
C 90:
fa ;
h- :
U 8°:
oi :
7O:
60:
50:
/in J
B ««
•
i r , * I
: $ '|T :
1 1 ' p
: i I -
B •
• «
1 i
, •
i •
i •
', 0 Impactor J
f *
i «
o Ultrafine
i •
• -
> «
» -
. •
. «
•> •
. •
. «
. *
• •
> *
. •
l i i i i llU 1 1 1 1 1 III* 1 1 1 1 1 lit
-0.5
-1
-S
-5 M
H
; 1—
-1O ^
Q!
: h-
•30 H
' tf
;3O
;4O
;5O
-cn
1O"5 1O"1 10° 1O1
PARTICLE DIAMETER (MICRDMET
Figure 26.
Penetration efficiency versus Stokes particle diameter for
January 16. Data taken by Cascade impactors and electrical
mobility methods are shown. Inlet data used in these
calculations were taken on January 12 & 14. 50% confidence
intervals are shown.
72
-------
SECTION 7
ELECTROSTATIC PRECIPITATOR ELECTRICAL CONDITIONS
The collection efficiency of an electrostatic precipitator
is directly related to the electrical condition that exists
within the interelectrode space. The electrical conditions are
governed by the power supplies that serve to energize the pre-
cipitator, by the general mechanical alignment of the various
sections, by the electrical resistivity of the particulate, and
by the dust buildup conditions on the corona wires and collec-
tion electrodes.
The power supplies installed in this precipitator are de-
signed to deliver approximately thirty nanoamperes per square
centimeter (^30ya/ft ) and a maximum voltage of approximately
sixty kilovolts. These current limits are somewhat low. In
those cases where other factors do not limit the performance,
current densities of perhaps sixty to seventy nanoamperes per
square centimeter may be expected. The electrical energization
of the electrostatic precipitator is shown in Figure 27.
The general mechnanical alignment of the electrostatic pre-
cipitator could not be evaluated in this installation because
this requires a physical inspection of the internals. We were
not able to gain access to these internals during the test
period since no plant shutdown was scheduled. After a review
of the data and considerations of the overall value of that
test program to the aims of the research grant, we chose not to
make an additional inspection trip to conduct this inspection.
If the electrostatic precipitator had been operating acceptably,
this inspection would have been performed.
In general, the current density in an operating precipitator
increases from inlet to outlet while the secondary voltage de-
creases from inlet to outlet. This phenomenon is generally
caused by the presence of a reasonably high electrical space
charge from the .fine particles suspended in the inlet gas stream.
This suppresses the current and increases the voltage on the
inlet fields. As the gas stream proceeds through the ESP the
particles are collected, reducing the space charge, allowing the
current to increase while the voltage decreases.
73
-------
EXIT PLENUM
OUTLET FIELD
CENTER FIELD
INLET FIELD
Figure 27. Electrostatic Precipitator Configuration.
74
-------
This condition was approximately obtained on January 13
through January 15, 1977. However, there was a significant
variation in the electrical conditions within this period. (See
Figures 28, 29, and 30). The general shape of the curves are
correct, but there exists the possibility of dust buildup on
the electrodes and/or mechanical misalignment within the field.
An anomaly was noted on January 15. All sets exhibited an
unusually high corona start voltage on this day. This suggests
a possible dust buildup on the corona electrode, leading to an
increase in the effective corona wire diameter.
The voltage vs. current curves for the morning and evening
of January 16 are shown in Figures 31 and 32. The voltage-
current curves in the morning (Figure 31) had a normal configu-
ration; however, the inlet curve in the evening (Figure 32)
shows a higher current than either the middle or outlet field,
in contrast to what is expected. A possible explanation for
this increase in current in the inlet field is a buildup of
material on the insulators that support and act as guides for the
corona frame. This buildup provided a parallel resistive leakage
path to ground for this field.
The electrical conditions were highly variable during the
test period. Current densities in specific fields tended to
change from day to day. These changes suggest a significant
variation in particulate and gas loadings and dust buildup
conditions over the test period. See Table 3 on page 14 for the
electrical operating log during the test period.
The electrical resistivity was measured with an E.P.A. high
temperature in situ probe. This probe has been modified for use
in temperatures encountered in smelter operations and for hot
side electrostatic precipitators in the Electric Utility Indus-
try. During this test period, fuel oil was fired in the rever-
beratory furnace. The values obtained from these measurements
are given below.
Temp Resistivity Cell Depth
Date Time °C -ohm*cm cm
1/11/77 1600 238 8.0 x 109 0.06
1/12/77 0900 238 1.5 x 1010 0.11
1/12/77 1100 227 5.5 x 109 0.09
1/12/77 1450 246 7.6 x 107 0.16
1/14/77 1130 241 7.4 x 108 0.06
1/15/77 1015 254 5.1 x 1010 0.12
1/16/77 0830 243 1.7 x 109 0.06
75
-------
21.0
18.0
15.0
CM
Q.
§ 12.0
LLJ
o
9-0
6.0
3.0
0.0
A INLET FIELD
D CENTER FIELD
O OUTLET FIELD
16
I
I
24 32
VOLTAGE, kV
40
48
Figure 28. Voltage-current curves, January 13, 1977.
76
-------
21.0
18.0
15.0
a.
§ 12.0
Z
LLJ
D
o
9.0
6.0
1 I I I I f
A INLET FIELD
D CENTER FIELD
O OUTLET FIELD
24 32
VOLTAGE, kV
40 48
Figure 29. Voltage-current curves, January 14, 1977.
77
-------
7.00
6.00
5.00
a. 4.00
1
o
1
cc
cc
D
O
3.00
2.00
1.00
0.00
A INLET FIELD
D CENTER FIELD
O OUTLET FIELD
J I
I
16
24 32
VOLTAGE, kV
48
Figure 30. Voltage-current curves, January 15, 1977
78
-------
14.0
INLET FIELD
D CENTER FIELD
O OUTLET FIELD
24 32
VOLTAGE, kV
Figure 31. Voltage-current curves, January 16, 1977 AM.
79
-------
28.0
A INLET FIELD
D CENTER FIELD
O OUTLET FIELD
0.0
24 32
VOLTAGE, kV
Figure 32. Voltage-current curves, January 16, 1977 PM.
80
-------
The range of resistivities are such that resistivity does not
limit the performance of this electrostatic precipitator when
firing oil.
Some measurements were made at an earlier time when gas was
fired. These values (shown below) are somewhat higher and may
tend to cause electrical sparkover during normal operation.
Temp Resistivity CellDepth
Date Time, ^C -ohm-cm cm
12/8/76 0845 No Test
12/8/76 1050 279 8.1 x 10ll 0.06
12/8/76 1255 307 7.0 x 10x° 0.07
12/8/76 1555 316 1.6 x 1012 0.08
12/9/76 0815 338 1.6 x 101 x 0.05
12/9/76 - - 6.3 x 1010 0.09
12/9/76 1145 338 7.6 x 1010 0.07
81
-------
SECTION 8
MATHEMATICAL MODELING
UTILIZATION OF TEST DATA
One objective of the test program was to acquire repre-
sentative data concerning the gas stream and the operation of
the precipitator in order to test the extent of applicability
of the existing mathematical model of electrostatic precipita-
tion. If the predictions of the model show good agreement with
the test data under the measured conditions, then the model can
be used to predict precipitator performance under other specified
operating conditions in order to obtain effective precipitator
sizing. If the predictions of the existing model fail to give
satisfactory agreement with the test data, then the test data
can be utilized to establish where the model is unsatisfactory
and to develop a more appropriate model for smelter precipitators.
The data acquired during the test program have been utilized
in two ways in conjunction with the mathematical model of elec-
trostatic precipitation. First, certain test data have been used
as input parameters for the model. These measured input para-
meters include the inlet mass loading and particle size distribu-
tion, electrical operating conditions, the gas volume flow rate,
temperature, and pressure, and the electrical resistivity of the
particulate. The mathematical model can predict precipitator
performance based on these measured input parameters and the
known precipitator geometry. Secondly, certain test data have
been utilized for making comparisons with the predictions of
the mathematical model. These data include the inlet and outlet
mass loadings and particle size distributions from which overall
mass and fractional collection efficiencies can be determined.
DESCRIPTION OF THE MODEL
Under the sponsorship of the Environmental Protection
Agency, Southern Research Institute has developed a theoretically
based mathematical model of the electrostatic precipitation
process.* The model was developed from coal-fired utility data.
*J. P. Gooch, J. R. McDonald, and S. Oglesby, Jr. A
Mathematical Model of Electrostatic Precipitation.
Environmental Protection Technology Series, Publication
No. EPA-650/2-75-037 (April, 1975).
82
-------
The mathematical operations and calculations in the model are
performed by a computer program. The most important input data
to the program are (1) the measured particle size distributions
of the inlet dust, (2) the current and voltage to each series
field, (3) precipitator geometry, and (4) gas flow. The computer
program divides the precipitator into length increments in the
direction of gas flow and calculates the electric field distri-
bution^ and the particle charge, particle migration velocity, and
collection efficiency as a function of particle diameter for each
length increment.
Electric Field Calculation
Since the particle migration velocity is a function of the
electric field at the plate, it is necessary to calculate the
electric field adjacent to the collection electrode. The method
employed for this calculation is a numerical technique introduced
by Leutert and Bohlen.* The equations which must be solved are
written in discrete form in two dimensions as
= -£ , and
£o
n2 = en (*V M + AV A_p
p ° ^Ax Ax + Ay AyJ
where
p = space charge, coul/m3
.y = distance parallel to gas flow from wire to wire, m.
x = distance perpendicular to gas flow from wire to
plate, m
°= permittivity of free space, cou!2/(N-m2) and
V = potential, volts.
The computer model uses a subroutine which iterates on a
grid of electric field and space charge density until conver-
gency is obtained. Following convergence, an electric field
profile is available which is applicable to the voltage and
current existing in the length increment under consideration, and
which is consistent with the assumptions and boundary conditions
used in the solution.
*G. Leutert and B. Bohlen, "The Spatial Trend of Electric Field
Strength and Space Charge Density in Plate-Type Electrostatic
Precipitators. " Staub 3_2 (7):27 (July, 1972).
83
-------
Charging Rate Calculation
Calculation of particle charge is accomplished by using a
model developed by Smith and McDonald.* The model in differen-
tial form is
ne2 (r0-a)
_
J exp l(
_
dt ~ 4e o n 2 47re0kTar0
S 90
4. [3ar02 - ro3 (K+2) + a3(K-l)] eEo'cose,, . Q,fi
+ - kTroMK+2) - )] sinede
(_ne2/4lTe QakT)
where
q = charge, coulombs
t = time, seconds
NO - undisturbed ion concentration, #/m3
b. = ion mobility
e = electronic charge, coulombs
n = number of charges on particle
i
n — number of charges on particle at saturation due to
field charging mechanism
ns ' (1 + 2lff7> Eoa2/e
a = particle radius, m
K = particle dielectric constant
E0 - external field, Volts/m
*W, B. Smith and J. R. McDonald, "Development of a Theory
for the Charging of Particles by Unipolar Ions." J. Aero-
sol Sci., Vol. 7, pp. 151-166 (1976).
84
-------
k = Boltzmann's constant, joules/°K
T = absolute temperature, °K
ro = point at which radial component of electric field is
zero
%
v = mean thermal speed of ions, m/sec.
This equation is integrated numerically using the quartic
Runge-Kutta method. The model reduces to the classical diffusion
charging equation in the absence of an applied electric field and
approaches the results obtained from the classical field charging
equation for large particles and large values of electric fields.
This calculation is also performed by a subroutine of the com-
puter model. When the calculation is complete for each length
increment, values of particle charge are available for all parti-
cle sizes represented in the input particle size distribution
histogram for use in the subsequent collection calculations.
Particle Collection Calculation
The next step in calculating theoretical collection effi-
ciency is the calculation of the electrical drift velocity or
migration velocity resulting from the coulomb and viscous drag
forces acting upon a suspended particle. For particle sizes and
electrical conditions of practical interest, the time required
for the particle to achieve the steady state value of velocity
is negligible, and the migration velocity is given by:
6iray
where
w = migration velocity of a particle of radius a, m/sec
q = charge on particle, coul
Ep = electric field near the collection electrode, volt/m
a = particle radius, m
U = gas viscosity, kg/m-sec
C = Cunningham correction factor
= (1 + AA/a)
85
-------
where
A = 1.257 + 0.400 exp(-1.10a/X)
and
X = mean free path of gas molecules?: m.
The computer model uses the Deutsch equation to predict the
collection fraction riij for the ith particle size in the jth
incremental length of the precipitator. Thus, the Deutsch
equation is applied in the form
. . = 1 - exp(-w A /Q) ,
^j J- / J J
where
w. . = migration velocity of the ith particle size in the
lr^ jth increment
A. = collection plate area in the jth increment
Q = volumetric flow rate.
The fractional efficiency r\^ for a given particle size
over the entire length of the precipitator is determined from
N. ,
l/l
where N* -: is the number of particles of the ith particle size
per cubic meter of gas entering the jth increment. The quantity
NJ_ .; can be written in the form
N. . = N. . . - [1 -
i/j ifj-l J-/J"-1- j"-1- -"-/j -•-
where N^ ]_ = NifO/ the number of particles of the ith particle
size per'cubic meter of gas in the inlet size distribution. The
overall collection efficiency ri for the entire polydisperse aero-
sol is obtained from
86
-------
where P± is the precentage by mass of the ith particle size in
the inlet size distribution.
After convergence on the overall mass efficiency has been
obtained, the program calculates effective or length averaged
migration velocities for the different particle sizes from the
Deutsch equation:
- 8;
where
w
= affm/sece migrati°n vel°city of particle of radius
= total collecting area
n = collecting fraction for given particle size over
total length.
The model contains empirical correction factors which are
used as multipliers on the ideally-calculated effective migration
velocities in order to account for the nonideal effects of gas
velocity distribution, gas sneakage, and particle reentrainment.
These correction factors are based either on very simplifying
assumptions or limited data and are intended only to give a rough
estimate of the degradation in precipitator performance caused by
these nonideal conditions.
MODELING OF KENNECOTT PRECIPITATOR
For several reasons, the field test data obtained from
measurements on the Kennecott precipitator were not suitable for
use in the existing precipitator performance model or for use in
developing a model which can be applied to the smelter industry.
The Kennecott installation was not representative of a well-
functioning precipitator and data which are vital to the modeling
effort were not obtained due to constraints placed on the meas-
urements which could be made.
From the standpoint of a modeling effort, the following
problems were associated with the evaluation of precipitator
performance at the Kennecott installation: (1) all inlet size
distribution measurements were obtained at the same location
and this location should produce a measured size distribution
which .is prejudiced towards the finer size range due to physical
constraints at the sampling site; (2) considerable uncertainty
87
-------
exists in the flow rate measurements; (3) the precipitator had
considerable inleakage of air (estimated to be 40% based on data
in Table 11) and a large temperature drop from the inlet to the
outlet of well over 100°F; (4) during all of the testing, sig-
nificant fractions of the precipitator suffered electrical
outage; (5) no determination of the gas velocity distribution was
made at the inlet of the precipitator, again due to physical
constraints at the sampling site; and (6) no data were obtained
to gauge the effect of rapping losses on total emissions.
Due to the serious problems mentioned above, it does not
appear to be sensible to attempt to model the performance of the
Kennecott precipitator during the test period. Instead, it seems
to be more reasonable to attempt to project how the precipitator
would perform if it were operating properly. This type of pro-
jection has been made using the E.P.A.-S.R.I. mathematical model
of electrostatic precipitation which was developed in conjunction
with studies of emissions from coal-fired boilers.
Figure 33 shows the experimental fractional efficiency
curve representative of the entire test period and two theoreti-
cal projections. The input data which were used in the model
are given in Table 19. The inlet size distribution is the one
which was measured and may lead the model to predict a lower
overall mass efficiency than it should. The operating voltages
and currents were taken to be representative of those which
could be maintained in the precipitator based on measurements
made during the test period (average of upper limit of V-I
curves measured on 1/12/77). The values for the flow rate and
temperature are based on data at the inlet.
In both theoretical projections shown in Figure 33, the
normalized standard deviation of the velocity distribution was
assumed to be 0.25 and it was assumed that 5% gas sneakage
occurred over three stages (actual gas sneakage was much greater
than this, however). Losses in efficiency due to rapping re-
entrainment were determined by a procedure developed for the
Electric Power Research Institute based on rap and no-rap studies
of precipitators collecting fly ash. The existing model does
not account for certain effects such as particle charging near
corona wires, particle concentration gradients, and flow field
phenomena which should enhance particle collection. Empirical
correction factors for the calculated migration velocities have
been developed based on comparisons of measured no-rap migration
velocities and those based on data from precipitators which were
at least 27 feet in electrical length and were collecting fly
ash. Thus, the extent of applicability, if any, to the Kennecott
precipitator is unknown. Theoretical curve I contains these
correction factors whereas curve II does not.
88
-------
TABLE 19.
COMPUTER DATA INPUT (KENNECOTT)
SCA = 349 ft2/1000 acfm
Volume Flow - 1.555 x 105 acfm (Inlet)
Inlet Grain Loading - 0.20 gr/acf
Total Collecting Length - 22.5 ft
Gas Velocity - 2.59 ft/sec
Estimated Efficiency - 95.0%
Dust Density - 3130 kg/m3
Resistivity - 1.9 x 1010 ohm-cm
Gas Temperature - 473°F
Atmospheric Pressure - 710 mm of Hg
No. of Electrical Fields - 3
Field No. 123
Length of Electrical Field (ft) 7.5 7.5 7.5
Area of Electrical Field (ft2) 18,100 18,100 18,100
Average Applied Voltage (kV) 52.0 44.1 46.1
Average Current (A) 0.158 0.425 0.489
No. of Wires/Linear Section 12 12 12
1/2 Wire-to-Wire Spacing (in) 3.0 3.0 3.0
Total Wire Length (ft) 13,215 13,215 13,215
Corona Wire Radius (in) 0.078125 0.078125 0.078125
Wire-to-Plate Spacing (in) 5.0 5.0 5.0
89
-------
ss
0
o
o
LL
LL
LU
99.9
99.8
99.0
98
95
90
80
0
1 1 1 1 1
• EXPERIMENTAL
• THEORETICAL 1
A THEORETICAL II
— " —
• • A •_
• ••• •£*£*•• *~
^ A ^
. A
_ A A _
I ill i
1 1.0 10.
PARTICLE DIAMETER, jum
Figure 33. Theoretical and experimental
fractional efficiencies.
90
-------
APPENDIX A
SMELTER OPERATING CHARTS
This Appendix contains photocopies of instrument charts
indicative of the smelter operating conditions during the test
period. The charts give the twenty four hour operating data for
the reactor feed rate and the reactor and reverberatory furnace
temperatures. In addition, example charts for the reverberatory
furnace typical flow rate, reactor air flow rate, and waste heat
boiler profile are presented.
Operation Date Page
Reactor (Roaster) Feed Rate Jan 11 92
12 93
13 94
14 95
15 96
16 97
Reactor (Roaster) Temperature 11 98
12 99
13 100
14 101
15 102
16 103
Reverberatory Furnace Temperature 11 104
12 105
13 106
14 107
15 108
16 109
Typical Reverberatory Gas Flow Chart - 110
Typical Reactor (Roaster) Air Flow
Chart - 111
Typical Waste-Heat Boiler Profile - 112
91
-------
Figure A-l: Approximate Reactor (Roaster) Feed Rate, Tons/hr
(January 11, 1977)
92
-------
" ThVr. ' ii ; Lt—«•«••••,• :-!r:nr?.'f/,.7
Figure A-2: Approximate Reactor (Roaster) Feed Rate, Tons/hr
(January 12, 1977)
93
-------
Figure A-3: Approximate Reactor (Roaster) Peed Rate, Tons/hr
(Janaury 13, 1977)
94
-------
Figure A-4: Approximate Reactor (Roaster) Feed Rate, Tons/hr
(January 14, 1977)
95
-------
Figure A-5: Approximate Reactor (Roaster) Feed Rate, Tons/hr
(January 15, 1977)
96
-------
Figure A-6: Approximate Reactor (Roaster) Feed Rate, Tons/hr
(January 16, 1977)
97
-------
00
s?
0>
o
rt
O
^ H
c-i
pj -^
3 »
C O
PJ (U
h( 0)
^< rt
n>
H3
(D
0)
rt
C
Hi
(D
NOON
-------
16 AM
Wd 9
Figure A-8: Reactor (Roaster) Temperature, °F
(January 12, 1977)
99
-------
!6 AM
IWd 9
Figure A-9: Reactor (Roaster) Temperature, °F
(January 13, 1977)
100
-------
!6 AM
Figure A-10: Reactor (Roaster) Temperature, °F
(January 14, 1977)
101
-------
Wd 9
Figure A-ll: Reactor (Roaster) Temperature, °F
(January 15, 1977)
102
-------
!6 AM
|Wd 9
Figure A-12: Reactor (Roaster) Temperature, °F
(January 16, 1977)
103
-------
12 NOOM
I_ \ V~ \~~" T——^ 1
Figure A-13: Reverberatory Furnace Temperature, °F
(January 11, 1977)
104
-------
Figure A-14: Reverberatory Furnace Temperature, °F
(January 12, 1977)
105
-------
12 NOON
9,
<&/
«^feStr^^^
Figure A-15: Reverberatory Furnace Temperature, °F
(January 13, 1977)
106
-------
ll
\9
<&,
Figure A-16: Reverberatory Furnace Temperature, °F
(January 14, 1977)
107
-------
Figure A-17: Reverberatory Furnace Temperature, °F
(January 15, 1977)
108
-------
Figure A-18: Reverberatory Furnace Temperature, °F
(January 16, 1977)
109
-------
NOON
Figure A-19: Typical Reverberatory Gas Flow Chart
(Not Calibrated)
110
-------
!6 AM
Figure A-20: Typical Reactor (Roaster) Air
Flow Chart
111
-------
NOON
Figure A-21: Typical Waste-Heat Boiler Profile
112
-------
APPENDIX B
CHEMICAL SAMPLING AND ANALYSIS BY RADIAN CORPORATION
SAMPLING TRAINS
Trace Elements; Gaseous and Particulate
The samples for analysis of elements in gaseous compounds
and particulates were collected from the reverberatory off-gases
at the ESP inlet and outlet and at the stack. The samples were
collected isokinetically from single points in the ESP inlet and
outlet ducts. The stack was sampled isokinetically over a 10
point traverse along one diameter.
The train used is shown in Figure B-l. Samples were drawn
through a Pyrex sampling nozzle and probe and then through a wet
electrostatic precipitator (WEP) by means of Teflon tubing. In
the WEP, shown in greater detail in Figure B-2, gas bubbled
through the circulating electrolytic reservoir and then passed
up through a cylindrical chamber, the walls of which were wetted
with the electrolyte (5% nitric acid). Collection of particulate
and vapors was achieved in this area by applying a 12-kV poten-
tial across the center platinum electrode and the wetted outer
wall, thus creating a corona discharge at the platinum wire.
The particulate-free gas stream exited at the top of the WEP.
The sample was retained in the electrolyte as a mixture of
suspended solids and dissolved species.
Gaseous compounds escaping the WEP were collected by a
series of eight impingers. The contents of each impinger is
given in Table B-l.
113
-------
PYREX PYREX LINED PROBE
NOZZLE
ACID IMPINGERS
CAUSTIC IMPINGERS
HYDROGEN PEROXIDE
IMPINGER
WET ELECTROSTATIC
PRECIPITATOR (WEP)
i E'v >.•«. « V|VL ti*\M *• "/ft "i IT" *
A.VIL- £ .\»"t vV. V ^ t ;W<> -". -V* J
ICE BATH
DRY IMPINGERS
SILCA GEL
IMPINGER
DRY
GAS
METER
FINE
ADJUSTMENT VALVE
COARSE
ADJUSTMENT VALVE
PUMP
Figure B-1. Schematic diagram of the Integral WEP Sampling Train.
114
-------
SAMPLE
OUTLET'
PLATINUM ELECTRODE
-WEP Body
-SAMPLE
"INLET
PERISTALTIC
PUMP
RESERVOIR OF
CIRCULATING
ELECTROLYTE (5% HNO3)
Figure B-2. Wet Electrostatic Precipitator.
115
-------
TABLE B-l. IMPINGER SOLUTIONS USED FOR
TRACE ELEMENT COLLECTION
Impinger
Nuinber
Solutions
1,2 1:1:1 sulfuric acid, nitric acid, deionized water
in a Greenburg-Smith impinger
3,7 dry modified Greenburg-Smith impingers
4,5 20% potassium hydroxide in a Greenburg-Smith
impinger
6 hydrogen peroxide in a Greenburg-Smith impinger
8 preweighed silica gel in a modified Greenburg-
Smith impinger
Mercury Vapor
Mercury vapor was collected by a gold amalgamation tech-
nique. The gas was first passed through an in-stack glass-fiber
filter to remove particulates, then through 6% hydrogen peroxide
to remove S02 interference, and finally through a quartz tube
containing a plug of very fine gold wire. Mercury vapor was
collected on the gold surface through amalgamation. The mercury
was later thermally desorbed and analyzed with a flameless
atomic absorption technique.
Vapors of Other Trace Elements
The collection of compounds in the vapor state was accom-
plished with a series of impingers preceded by a glass-fiber
in-stack filter to remove particulates. The sampling train is
shown in Figure B-3. The contents of the impingers were iden-
tical to those listed in Table B-l.
Sampling Schedule
The schedule of sampling with the WEP train and the vapor
train is given in Table B-2. In general, sampling with the
mercury vapor train was performed as part of each operation with
the WEP train or the vapor train.
116
-------
ACID IMPINGERS
CAUSTIC IMPINGERS
HYDROGEN fEROXIDE
IMPlNGER
TEFLON TUBING
FILTER HOLDER
4
PYREX LINED PROBE
NOZZLE
ICE BATH
DRY SILCAGEL
IMPINGERS IMPlNGER
• FINE
ADJUSTMENT VALVE
COARSE
ADJUSTMENT VALVE
PUMP
Figure B-3. Schematic diagram of the Vapor-Phase Element Samp/ing Train.
117
-------
TABLE B-2. SCHEDULE OF SAMPLING
BY RADIAN CORPORATION
Date
Location
Train
Time, hr
Jan . 11
Jan . 12
Jan. 14
Jan. 15
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Stack
Inlet
Stack
Vapor
WEP
WEP
Vapor
WEP
Vapor
WEP
WEP
WEP
1120
1128
1052
0917
0843
0917
0950
0943
0940
- 1630
- 1613
- 1730
- 1705
- 1500
- 1705
- 1630
- 1343
- 1625
ANALYTICAL METHODS
Trace Elements
The analysis of samples from the WEP train consisted of two
major steps: dissolution and chemical analysis. Sample disso-
lution techniques included perchloric acid digestion and lithium
borate fusion. The analytical techniques were based on atomic
absorption spectrophotometry (AAS), ion-selective electrometry,
and fluorometry.
Figures B-4, B-5, and B-6 summarize the dissolution and
analytical procedures. The remainder of this section describes
the procedures in greater detail.
Sample Dissolution
The dissolution of solid samples (from the WEP train or the
ESP hoppers) was achieved with the following techniques:
• Percholoric acid digestion - The sample was first treated
with nitric and hydrofluoric acids. Perchloric acid was
added for final oxidation of the sample. A small amount
of hydrochloric acid was finally added to ensure complete
dissolution.
118
-------
FLUE
GAS •
. Au. Amal..
. FAA.
.Hg
Mo
Sb, Cd, Ni,
Pb, Co
AA: ATOMIC ABSORPTION, FLAME
Dl: DIRECT INJECTION
DCS: DOUBLE CAPILLARY SYSTEM
FAA: FLAMELESS ATOMIC ABSORPTION
HGA-AA: HEATED GRAPHITE ANALYZER OF THE
ATOMIC ABSORPTION SPECTROPHOTOMETER
ICE: INOFiGANIC COMPLEX EXTRACTION
OE: ORGANIC EXTRACTION
PAD: PERCHLORIC ACID DIGESTION
SA: STANDARD ADDITIONS
SIE: SPECIFIC ION ELECTRODE
Figure B'-4. Analytical Scheme for WEP Liquor.
119
-------
.IMPINGER SOLUTION
LIQUOR
SAMPLE
.KMn04 OXIDATION.
-AA.
FAA.
Mo
Co, Sb, Cd, Ni, Pb
• Hg
Al, Fe, Cr, Cu, Zn
H202 IMPINGER SOLUITON
AA: ATOMIC ABSORPTION, FLAME
Dl: DIRECT INJECTION
DCS: DOUBLE CAPILLARY SYSTEM
FAA: FLAMELESS ATOMIC ABSORPTION
HGA-AA: HEATED GRAPHITE ANALYZER OF AA
ICE: INORGANIC COMPLEX EXTRACTION
OE: ORGANIC EXTRACTION
PAD: PERCHLORIC ACID DIGESTION
SA: STANDARD ADDITIONS
SIE: SPECIFIC ION ELECTRODE
ACID IMPINGER: 1:1:1: c.H2S04, c.HNOo, D.I WATER
BASIC IMPINGER: 20% KOH
H202 IMPINGER: 5% H202
Figure B-5. Analytical Scheme for Impinger Liquor.
120
-------
Dust
.Sample'
.Au Amal..
.FAA.
.Hg
.PAD.
-OE.
. AA.
.Mo
.SA/DCS.
. HGA-AA Sb, Cd, Co,
Fe, Ni, PB
flfl fli. Cr, Cu, Zn
.ICE.
.HGA-AA.
-As
.SA.
.HGA-AA.
.Ag, V
.SA/DCS
.EMISSION.
-Ba
.ACID Dig..
.HGA-AA.
.Au
-ACID BOMB .
-OF.
.FLUOROMETRIC — Se
.SOLID DISSOLUTION -ION EXCHANGE TITRIMETRY .
.FUSION.
.S1E.
AA: ATOMIC ABSORPTION, FLAME
Dl: DIRECT INJECTION
DCS: DOUBLE CAPILLARY SYSTEM
HGA-AA: HEATED GRAPHITE ANALYZER OF THE
ATOMIC ABSORPTION SPECTROPHOTOMETER
ICE: INORGANIC COMPLEX EXTRACTION
OE: ORGANIC EXTRACTION
PAD: PERCHLORIC ACID DIGESTION
SA: STANDARD ADDITIONS
SIE: SPECIFIC ION ELECTRODE
Figure B-6. Analytical Scheme for Precipitator Dust.
121
-------
Lithium borate fusion - A small amount of sample was
fused with lithium borate. The cooled melt was
dissolved in hydrochloric acid and hot water.
Most elements present in higher concentrations were analyzed
from the lithium borate fusion. Elements present in trace con-
centrations were, in general, determined from solutions obtained
from the perchloric acid digestion scheme.
There is some concern that volatile elements, such as
arsenic, may have been partially lost by evaporation or sublima-
tion during sample treatments with hot acids. However, it has
been reported (C. Feldman, Anal. Chem. 49, 826, 1977) that no
appreciable loss of arsenic occurs during" the treatment of fly
ash with acids at temperatures exceeding 200°C. Thus, it is
assumed that losses of volatile elements were not of significant
magnitude.
Elemental Determinations
The analytical procedures used were originally developed for
the determination of trace elements in coal, coal ashes, sludges,
and plant and animal tissues. The drastic change in the matrix
observed in samples collected at the copper smelter necessitated
screening of the procedures for accuracy and reliability. This
task was accomplished by the method of standard addition and
interference studies.
Silver and Vanadium—
Silver and vanadium were determined by the method of
standard additions with the graphite furnace of the atomic
absorption spectrophotometer. Samples used were the WEP liquor,
the impinger solutions, and the perchloric acid solution of
solids. There was no sample preconcentration needed for the
determination.
Aluminum—
Aluminum was determined in the WEP liquor, the impinger
solutions, and the perchloric acid solution of solids by AAS with
the nitrous oxide-acetylene flame.
Arsenic—
The WEP liquor, the impinger solutions, and the perchloric
acid solution of solids were used for the arsenic determination.
Arsenic was complexed in an acidic medium as the heteropoly acid
of molybdenum. The aqueous complex was injected into the heated
graphite analyzer of the atomic absorption spectrophotometer.
A charring temperature of 1200°C was used.
122
-------
Gold-
Gold in solid samples was determined by direct injection
into the heated graphite analyzer after the solids had been
digested in a solution of hydrochloric and nitric acids. The
WEP liquor and the impinger solutions were injected with no
pretreatment.
Barium--
Samples from the WEP liquor, the impinger solutions, and the
perchloric acid solution of solids were aspirated into the
nitrous oxide-acetylene flame of the atomic absorption spectro-
photometer in the emission mode. The method of standard addi-
tions with the double capillary system was used. Potassium
salts were added to enhance the signal and mask interfering
species.
Cadmium and Lead—
The double complexing agent of ammonium pyrrolidine dithio-
carbamate and diethylammonium diethyldithiocarbamate was used to
chelate lead and cadmium in aqueous solution. Lead and cadmium
were extracted simultaneously with methyl isobutyl ketone (MIBK)
from the WEP liquor, the impinger solutions, and the perchloric
acid solution of solids. The extracted sample was injected into
the graphite furnace attachment to the atomic absorption spec-
trophotometer.
Cobalt and Nickel—
Cobalt and nickel were chelated with diethyldithiocarbamate
and then extracted with MIBK from the WEP liquor, the impinger
solutions, or the perchloric acid solution of solids. The
extracted sample was injected into the graphite furnace and
determined by AAS.
Chromium, Copper, Iron and Zinc—
The four metals were determined by AAS with the air-
acetylene flame. The samples were analyzed by the method of
standard additions with a double capillary system. The WEP
liquor, the impinger solutions, and the perchloric acid solution
of solids were analyzed by this method.
Fluorine—
Solid samples were fused with sodium carbonate, and the
melt was dissolved in deionized water. WEP liquor and impinger
solutions were run directly. Final determination was done with
a fluoride-specific ion electrode by the method of known addi-
tions to remove the effects of any interfering ions.
123
-------
Mercury--
In the amalgamation method, gas samples were drawn through
a plug of gold wool. Deamalgamation was accomplished by heating
the gold wool. The released mercury was purged through the
absorption cell of the atomic absorption spectrophotometer.
Solids were analyzed for mercury by weighing a sample into a
platinum boat and heating the sample slowly in a chamber. The
off-gases containing elemental mercury were purged through a gold
plug. Deamalgamation and determination by AAS followed the same
procedure as described above.
Liquid samples were acidified. Mercury was oxidized next
with potassium permaganate. Hydroxylamine hydrochloride and
stannous chloride were next used to reduce the mercury to the
metallic state. Air was bubbled through the solution and passed
through the absorption cell of the atomic absorption spectro-
photometer .
Molybdenum—
Molybdenum was complexed as the thiocyanate, extracted into
MIBK, and aspirated into the nitrous oxide-acetylene flame of the
atomic absorption spectrophotometer. Ascorbic acid and sodium
fluoride were used to mask interference from iron and titanium.
Antimony—
Antimony was extracted as the iodide into a mixture of
tributylphosphate and MIBK. Extraction was performed.on the
WEP liquor, the impinger solutions, and the perchloric acid
solution of solids. Sulfamic acid was added to the acid impinger
solution prior to extraction to remove the nitrates. Peroxide
impinger solutions were boiled prior to extraction to decompose
the hydrogen peroxide. The extracted solution was injected into
a graphite tube of the atomic absorption spectrophotometer, which
had been coated with ammonium molybdate.
Selenium—
Solid samples were digested in a Teflon bomb with nitric
acid and perchloric acid. Following the digestion, the sample
was heated with dilute hydrochloric acid. WEP liquors and
impinger solutions were also heated with HC1. Extraction pro-
cedures for all samples were the same from this point forward.
Following stabilization with formic acid, hydroxylamine, and
EDTA, the samples were complexed with 2,3-diaminonaphthalene.
The selenium complex was extracted into cyclohexane and measured
on a fluorometer.
124
-------
ANALYTICAL RESULTS
The analytical data for the liquors from the WEP train (the
combination of dissolved solids and original liquor, on one hand,
and the combination of impinger solutions) are given in Table
B-3. Also given in this table are data for the impingers used
in the vapor sampling train. Calculated concentrations of the
elements in the units lb/106 scf are given in Table B-4; values
recalculated in the units mg/m3 are given in the body of this
report.
125
-------
TABLE B-3'. ELEMENTAL CONCENTRATIONS (yg/g^IN SEPARATE FRACTIONS COLLECTED IN THE WEP AND VAPOR SAMPLING TRAINS
N)
ESP INLET
Date
Radian No
Weight (grams)
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
WEP
liquor
Jan. 12
947
1384.03
0.94
66.0
26.0
0.002
<0.6
9.0
0.13
0.59
530.0
190.0
590.0
<0.01
60.0
0.47
51.0
5.8
1320 ppb
0.25
54.0
WEP -
impingers
Jan. 12
948
1287.57
<0.01
0.82
0.098
<0.0005
<0.8
0.006
<0.001
0.10
0.20
3.2
0.28
<0.01
0.041
<0.01
0.06
0.003
9.7 ppb
0.43
0.14
WEP
liquor
Jan. 14
953
1416.49
0.22
110.0
56.0
0.0005
<0.6
25.0
0.92
1.90
1114.0
170.0
1400.0
0.027
82.0
3.1
72.0
6.9
3960 ppb
0.36
11.0
WEP
impingers
Jan. 14
954
1208.59
0.13
0.66
0.29
0.002
<0.9
0.086
<0.001
0.12
1.7
1.1
0.20
<0.01
0.28
<0.01
0.10
0.012
15 ppb
0.54
0.54
WEP
liquor
Jan. 15
951
1275.00
0.81
32.0
26.0
O.D005
<0.1
5.0
0,13
1.1
560.0
120.0
550,0
<0.01
28.0
0.60
30.0
3.6
940 ppb
0.15
45.0
WEP
impingers
Jan. 15
959
1209.34
0.15
<0.1
0.22
<0.0005
<0.1
0.019
0.015
0.11
1.4
1.5
0.22
<0.01
0.098
<0.01
0.070
0.015
8.6 ppb
0.87
0.18
WEP
liquor
Jan. 12
949
628.19
0.039
4.1
48.0
<0.0005
<0.1
1.2
0.029
0.12
21.0
420.0
14.0
<0.01
4.6
0.48
0.77
0.88
43 ppb
0.42
8.6
ESP OUTLET
WEP
impingers
Jan. 12 .
950
1389.62
0.23
0.5
0.059
<0.0005
<0.3
0.002
<0.001
0.128
0.11
2.5
0.10
<0.01
0.012
0.011
<0.05
0.012
11 ppb
0.81
0.030
Vapor train
impingers
Jan. 11, 12, 14
957
1659.90
0.080
1.6
54.0
<0.0005
<0.1
0.002
<0,001
0.10
0.47
240.0
0.66
<0.01
0.033
0.015
0.024
0.014
1.4 ppb
0.49
0.026
STACK
WEP
liquor
Jan. 14, 15
955
1192.10
0.62
3.8
56.0
0.0008
<0.3
1.6
0.018
0.19
49.0
320.0
23.0
<0.01
7.6
<0.01
6.0
1.1
17 ppb
0.010
13.0
WEP
impingers
Jan. 14, 15
956
1411.11
0.25
0.1
0.050
0.0005
<0.1
0.001
<0.001
0.071
0.23
3.3
0.24
<0.01
0.046
<0.01
<0.05
0.010
0.51 ppb
2.5
0.23
a. Fraction of the sample weight shown in the third row of data. This value is of primary interest in showing the distribution
of elements in different sampling devices. It is related to the stack gas concentration through an unspecified volume of gas sampled.
-------
TABLE B-4. TOTAL ELEMENTAL CONCENTRATIONS (.lb/106 scf)3 CALCULATED FROM
ANALYSES OF SAMPLES COLLECTED IN THE WEP AND VAPOR SAMPLING TRAINS
ESP INLET
Date
Sampling Method
Sample Volume (scf)
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Jan. 12
WEP
100.08
0
1
0
6
<0
0
0
0
16
5
18
1
0
1
0
0
0
1
.029
.9
.79
.lxlO~5
.4
.270
.0041
.021
.0
.5
.8
.014
.6
.18
.041
.020
.7
Jan. 14
WEP
96.54
0
3
1
0
<0
0
'0
0
37
5
45
7
2
0
2
0
0
0
0
.011
.4
.8
.00027
.04
.82
.030
.066
.0
.2
.lxlO~5b
.7
.099
.3
.22
.13
.027
.360
Jan. 15
WEP
92. 7&
0
1
0
6
<0
0
0
0
17
3
17
0
0
0
0
0
0
0
1
.029.
.0
.78
.2xlO~s
.006
.150
.0045
.035
.0
.6
.001b
.850
.018
.910
.110
.048
.030
.4
ESP
Jan. 12
WEP
126.64
0
0
.006
.058
0.530
2xlO~6
<0
0
0
0
0
4
0
0
0
0
0
0
0
0
.003
.013
.00013
.0044
.230
.4
.15
.051
.0055
.0084
.0099
.00073
.024
.095
OUTLET
Jan. 11, 12, 14
Vapor train
309.34
0.00095
0.019
0.64
6xlO~6
<0.001
2. 4xlO~ 5
-------
APPENDIX C
CHEMICAL SAMPLING AND ANALYSIS BY
SOUTHERN RESEARCH INSTITUTE
SAMPLING TRAINS
Trace Elements; Gaseous and Particulate
ESP inlet—
The sampling train for trace elements in the gaseous and
particulate states consisted of components located inside and
outside the flue-gas duct, as shown schematically in Figure C-l.
The internal portion of the train consisted of a series of three
cyclones with nominal cut points of 10, 3 and 1 ym; a glass-fiber
thimble; and a 6-ft stainless steel probe. Outside of the stack,
the probe was connected to a flexible Teflon transfer line main-
tained at 200°C (400°F), which, in turn, was attached to a
glass-fiber filter in a 120°C (250°F) oven. A 4-in. length of
stainless steel tubing was placed between the end of the Teflon
line and the entrance to the oven. This tubing acted as a heat
exchanger in lowering gas temperatures to 120°C at the face of
the filter. A small thermocouple in the filter holder was used
to monitor gas temperatures. Trace-element vapors passing
through the filter were drawn into a series of seven Greenburg-
Smith impingers in an ice bath. The first two impingers con-
tained about 75 and 160 ml, respectively, of a 1:1:1 mixture
of sulfuric acid, nitric acid, and water. The third and seventh
impingers were dry. The fourth impinger contained 250 ml of a
30% solution of hydrogen peroxide, and the fifth and sixth
impingers were filled with 200 ml of a 7% solution of potassium
hydroxide. The remainder of the train consisted of a condensate
collector, a calibrated orifice with a water manometer, a
mercury manometer, a dry gas meter, and a vacuum pump.
ESP outlet—
The sampling train used at the precipitator outlet was
essentially the same as described above except a single cyclone
(1 ym) was substituted for the three-cyclone set and a 47-mm
glass-fiber filter disc was used in place of the glass-fiber
thimble.
128
-------
K)
FLEXIBLE
HEATED
SAMPLING
LINE
CYCL
OVEN
,, 01
A
ONE FILTER
r
1
4
'/////// DUCT
THIMBLE
CYCLONES
ORIFICE
IMPINGERS
DRIERITE
TOWER
H2O
X
PUMP
MANOMETER
DRY
GAS
METER
Hg
MANOMETER
Figure C-1. Schematic diagram of cyclone-filter impinger sampling train.
-------
Mercury Vapor
Mercury vapor was sampled by drawing flue gas through, in
sequence, a glass-lined heated probe with a quartz-wool filter
plug, a saturated sodium carbonate bubbler (to remove sulfur
oxides), and two bubblers containing a solution of 3% potassium
permanganate in 5% nitric acid. The carbonate scrubber proved
to be effective in removing the sulfur oxides and showed no
tendency to absorb mercury. At a sampling rate of about 1 £/min
the collection efficiency of the first permanganate bubbler
averaged only 76% in the field test, as opposed to 99% in
laboratory experiments. This comparison suggests that the
mercury concentration in the flue gas, as determined by this
method, quite possibly is low. Unless otherwise stated, the
results given for mercury are based on samples obtained with the
cyclone-filter-impinger train rather than the permanganate
bubbler.
Sulfur Oxides
Sulfur oxides were collected by the controlled condensation
method for sulfur trioxide and an absorption method for sulfur
dioxide. This procedure consisted of drawing flue gas through
a heated glass-lined sampling probe, with a quartz-wool filter
plug to remove particulate, into a condenser. The condenser was
maintained at 60 to 90°C, a temperature below the dew point of
sulfur trioxide as sulfuric acid but above the dew point of
water vapor. Thus, sulfuric acid was retained in the condenser
while sulfur dioxide was swept into a 3% hydrogen peroxide
bubbler to be converted to sulfuric acid. The remainder of the
train consisted of a drying tower, a dry test meter, and a
vacuum pump. The total sulfate content of the condenser and
bubbler samples was determined by a barium perchlorate titration
with Thorin indicator.
Other Gases
Carbon dioxide and oxygen were determined with an Orsat
apparatus. Water vapor was determined by collection in a tared
Drierite absorption tube.
Sampling Schedule
Sampling runs with the trace-element train were conducted
at the outlet of the electrostatic precipitator on January 12
and 14, 1977, and at the inlet to the precipitator on January
15 and 16, 1977. Simultaneous samples were collected by
Radian Corporation using the wet electrostatic precipitator
sampling system. Because of space limitations, the probes for
the two trains could not be inserted into the same sampling port;
instead, they were inserted into adjacent sampling ports in both
test locations. Sulfur oxide, mercury, and Orsat samples were
130
T
-------
obtained from an additional port. The schedule for sample
collection is shown in Table C-l.
METHODS FOR TRACE ELEMENTS
Sample Preparation
Samples returned to the laboratory for analysis included
cyclone, thimble, and filter solids, and impinger solutions from
both the trace-element and mercury sampling trains. The acid,
base, and peroxide impingers from the trace-element train were
combined on all except the January 14 run. For this set of
samples, the acid, the base, the peroxide impingers were kept
separate in order to determine the distribution of the trace
metals in the various types of collection media.
Acid attack—
Several acids and combinations of acids were screened for
effectiveness in dissolving the hopper sample. The more
promising procedures were then modified into a method subse-
quently referred to as the standard acid attack. This method,
described below, was used to process all the cyclone and hopper
samples collected during the test program.
A 0.500 g sample, or a lesser weight depending on the
availability of material, was added to a Teflon beaker containing
10 ml of concentrated sulfuric acid and 5 ml of concentrated
nitric acid. The mixture was slowly taken to fumes of sulfur
trioxide and cooled, and an additional 10 ml of concentrated
nitric acid was added. The sample was again taken to fumes of
sulfur trioxide and cooled, and 2 ml of 72% perchloric acid was
added. After being taken to fumes of perchloric acid, the
residue-was digested with distilled water and filtered through
a 0.45 ym membrane filter.
The filter was transferred to a platinum crucible and ---___
attacked with 3 ml of 40% hydrofluoric acid. Then, 3 ml of con-
centrated nitric acid was added, and the sample was digested.
Next, an addition of 2 ml of 72% perchloric acid was made and
the sample was further digested. Finally, the sample was diluted
with water, again digested, and combined with the original fil-
trate and diluted to a volume of 250-ml.
With those samples in abundant supply, a second type of
acid attack was used for a comparatively study. This method con-
sisted of digesting a 0.500 g sample of 10 ml of aqua regia and
1 ml of 72% perchloric acid in a Teflon beaker. After being
taken to fumes of perchloric acid, the sample was cooled, diluted
with water, and filtered. The filter was discarded, and the
filtrate diluted to 250 ml.
131
-------
TABLE C-l. SCHEDULE FOR SAMPLING BY SOUTHERN RESEARCH INSTITUTE
Date Location
Jan 12 Outlet
Jan 14 Outlet
Jan 15 Inlet
Jan 16 Inlet
Train
Trace element
Sulfur oxides
Mercury
Trace element
Sulfur oxides
Mercury
Trace element
Sulfur oxides
Mercury
Trace element
Sulfur oxides
Mercury
Time, hr
1100-1730
1000-1100
1130-1230
1700-1720
1320-1630
930-1430
1000-1100
1135-1315
1615-1700
1350-1535
945-1345
1345-1500
1530-1700
955-1440
1200-1300
1330-1630
Gas temp ,
°C
179
171
174
188
182
177
177
177
179
177
241
266
271
246
243
243
Volume
sampled, m3
5.0
0.037
0.069
0.043
0.164
4.2
0.079
0.124
0.055
0.091
2.4
0.063
0.079
1.9
0.018
0.084
132
-------
Neither the in-stack thimble and filter nor the out-of-stack
filter collected sufficient particulate to enable a sample to be
separated from the glass-fiber matrix. Therefore, the entire
filter or thimble was extracted by an overnight reflux with 20-30
ml of aqua regia. Each sample was then filtered through a
0.45 ym membrane filter and diluted to 100 ml with deionized
water.
A question of possible loss of volatile elements during hot
acid treatment was addressed in the discussion of Radian's
sample dissolution methods. The loss is assumed to be negligible
for reasons already cited in Appendix B.
Fusion attack—
An attempt was made to dissolve the hopper samples in
several types of fluxes including potassium pyrosulfate, sodium
tetraborate, lithium metaborate, sodium carbonate, and sodium
hydroxide. Based on the quantity of insoluble material remain-
ing after the fusion, none of these procedures were considered
satisfactory for preparing samples for a complete analysis. The
sodium hydroxide attack, however, was used to solubilize particu-
lates for fluorine determination.
Elemental Determinations
Most of the analyses were conducted by use of a Perkin-
Elmer Model 603 atomic absorption spectrophotometer, which could
be used when appropriate with an HGA-2100 graphite furnace. The
method of addition was used in each procedure as a means of re-
ducing potential interferences.
Silver, aluminum, cadmium, chromium, copper, iron, nickel, and
zinc—
Aluminum was determined by atomic absorption spectrometry
(AAS) with a nitrous oxide-acetylene flame. A 0.2% KC1 solution
was added to the samples to control ionization. The remaining
metals were also determined by AAS but with an air-acetylene
flame and a 4-inch single-shot burner head. Scale expansion was
used to increase the sensitivity of the instrument when neces-
sary.
Arsenic—
Arsenic in.the impinger, filter and thimble samples was
determined by AAS with air-acetylene flame, a D2 background
corrector, and an electrodeless discharge lamp.
Gold--
The gold content of the impinger, cyclone, and hopper
133
-------
samples was determined by the method of Strong and Murray-Smith
(Talanta 21, 1253 (1974)). This procedure was based on an ex-
traction of" the gold with MIBK from a sample made 20% in HC1.
The sensitivity of the method was increased by the use of the
graphite furnace instead of flame atomic absorption. The filter
and thimble samples were not analyzed for gold since an excessive
amount of the limited volume of sample would have been expended.
Barium—
The graphite furnace was used to determine the barium con-
centration in the impinger, hopper, and cyclone samples. The
thimble and filter samples were not run because the entire volume
of these solutions had been used for other analyses.
Cobalt—
The impinger, hopper, and cyclone samples were analyzed with
the graphite furnace. The filter and thimble samples were
analyzed by AAS with an air-acetylene flame.
Fluorine—
All determinations of this element as fluoride were made by
means of an Orion Model 94-09A fluoride-specific ion electrode,
a conventional saturated potassium chloride reference electrode,
and a Corning Model 12 expanded-scale pH meter. The impinger
solutions were adjusted to pH 6; then 1 M sodium citrate was
added to act both as a buffer and a complexing agent for alum-
inum and iron. The hopper and cyclone samples were fused with
sodium hydroxide in a nickel crucible, digested overnight with
water, and then dissolved in 1 M hydrochloric acid prior to
analysis. No attempt was made to analyze the filter or thimble
samples, and only the cyclone catches at the precipitator inlet
contained enough material for this type of analysis.
Mercury—
The impinger samples from the mercury and trace-element
sampling trains were analyzed for mercury by the cold vapor
technique of Hatch and Ott (Anal. Chem. 40, 2085 (1968))
Because of interference problems', this method was unsatisfactory
for the cyclone, hopper, thimble and filter samples; therefore,
a procedure of Tindall (Atomic Absorption Newsletter 16. U) ,
38 (1977)) was adopted. This method was based on the conversion
of mercury to a tetraiodomercurate ion in an ammoniacal state
followed by the extraction of the complex into MIBK. The extract
was then analyzed by AAS with an air-acetylene flame.
134
-------
Molybdenum—
The molybdenum content of the impinger solutions was deter-
mined with the graphite furnace, while the remaining samples
were analyzed by AAS with a nitrous oxide-acetylene flame.
Lead and Antimony—
Lead and antimony contents of the impinger solutions were
determined by the procedure of Burke (Analyst 97, 19 (1972)).
Both of these elements were quantitatively extracted from a 10%
hydrochloric acid solution of the sample containing 2% ascorbic
acid and 6% of potassium iodide in a single extraction with a 5%
solution of trioctylphosphine oxide in MIBK. The extract was
nebulized directly into the burner of the atomic absorption
spectrophotometer. The remaining samples were analyzed without
resorting to the extraction procedure.
Selenium—
The graphite furnace was used to analyze the hopper,
cyclone, filter, and thimble samples for selenium. Nickel was
added to the samples to prevent the loss of selenium during the
charring cycle. Since the high sulfate content in the impinger
solutions caused severe interference problems with this method,
these samples were analyzed by the diaminonaphthalene procedure
of Lott, et a.1 (Anal. Chem. 35, (1963)).
Vanadium—
The vanadium content of all samples was determined by AAS
with the graphite furnace.
ANALYTICAL RESULTS FOR TRACE ELEMENTS
The results of individual samples collected at the ESP
inlet and outlet in the cyclone-filter-impinger sampling trains
are given in Tables C-2, C-3, C-4, and C-5.
135
-------
TABLE C-2. ELEMENTAL CONCENTRATIONS CALCULATED
FOR SAMPLES COLLECTED AT THE OUTLET DUCT
ON JANUARY 12, 1977
Concentration,3 mg/m3
E lement
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Cyclone
in duct
0.006
0.32
0.265
<0.001
<0.003
0.12
<0.005
0.022
7.0 ,
_b
3.0
<0.001
0.48
0.037
0.20
0.12
0.027
<0.002
1.6
Filter
in duct
<0.002
0.19
0.082,
_D
b"~
0.020
<0.001
0.003
1.1 ,
b
0.10
<0.001
0.13
<0.001
0.15
0.030
0.009
<0.001
0.54
Filter
outside duct
<0.003
0.22
0.10 .
D
b~
<0.001
<0.001
<0.001
0.007,
b
0.014
<0.001
<0.002
<0.001
0.002
<0.01
<0.004
<0.001
0.24
Impinge r
outside duct
<0.01
0.15
6.3
<0.001
<0.02
<0.01
<0.015
<0.01
0.046
21
0.09
0.008
<0.03
<0.01
<0.01
<0.01
0.027
<0.03
0.006
a. At 21°C (70°F) and 760 mmHg.
b. Not determined.
136
-------
TABLE C-3. ELEMENTAL CONCENTRATIONS CALCULATED
FOR SAMPLES COLLECTED AT THE OUTLET DUCT
ON JANUARY 14, 1977
Concentration,3 mg/m3
Element
Ag
Al
As
Au.
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
a. At
b. Sam
Cyclone
in duct
0.004
0.20
0.039
<0.001
<0.003
0.088
<0.005
0.017
4.6
-
2.3
<0.001
0.31
0.024
0.23
0.048
0.014
<0.002
0.80
Filterb
outside duct
<0.002
0.20
0.18
_c
c
0.011
<0.001
0.005
1.0
c
0.064
<0.001
0.10
<0.001
0.075
<0.01
<0.005
<0.001
0.45
Impingers
outside duct
acid Peroxide
<0.01
0.12
7.0
<0.001
<0.025
<0.005
<0.005
<0.01
0.028
23
0.039
0.009
<0.01
<0.01
<0.005
<0.015
0.031
<0.01
0.008
<0.01
<0.015
0.056
<0.001
<0.025
<0.005
<0.005
<0.01
0.023
0.009
0.006
<0.001
<0.01
<0.01
<0.005
<0.015
<0.005
<0.01
0.008
base
<0.01
<0.015
<0.01
<0.001
<0.025
<0.005
<0.005
<0.01
<0.016
12
<0.006
<0.001
<0.01
<0.01
<0.005
<0.015
<0.01
<0.01
0.005
21°C (70°F) and 760 mmHg.
pie on filter
inside duct was
lost; because of this,
c.
the data in this table are not included in the body of
this report.
Not determined.
137
-------
TABLE C-4. ELEMENTAL CONCENTRATIONS CALCULATED
FOR SAMPLES COLLECTED AT THE INLET DUCT
ON JANUARY 15, 1977
•a _
Concentration, mg/m
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Cyclone
in duct
0.079
25.9
2.0
0.001
0.070
0.62
0.044
0.40
271
0.57
242
<0.001
8.5
0.47
3.3
0.73
0.82
0.062
12
Filter
in duct
0.046,
o
2.1 ,
D
_b
0.58
<0.001
0.12
50 ,
-
3.5
<0.001
4.4
0.024
1.1
0.33
0.16
0.007
11
Filter
outside duct
<0.003
<0.03
0.082,
D
b~»
_
<0.001
<0.001
<0.001
0.011,
-
0.015
<0.001
<0.005
0.005
<0.005
<0.02
<0.01
<0.001
0.054
Impinger
outside duct
<0.02
0.22
8.5
<0.001
<0.04
<0.01
<0.03
<0.02
0.076
33
0.11
0.011
<0.06
<0.02
<0.03
<0.025
0.081
<0.06
0.031
a. At 21°C (70°F) and 760 mmHg.
b. Not determined.
138
-------
TABLE C-5. ELEMENTAL CONCENTRATIONS CALCULATED
FOR SAMPLES COLLECTED AT THE INLET DUCT
ON JANUARY 16, 1977
Concentration,9 mg/m3
Element
Ag
Al
As
Au
Ba
Cd
Co
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
V
Zn
Cyclone
in duct
0.28
80
9.1
<0.001
0.38
6.2
0.083
0.78
930
1.9
833
<0.001
27
1.42
19
3.2
2.8
0.17
42
Filter
in duct
0.036.
D
4.7 .
D
b—
3.2
<0.001
0.10
26 b
*J
2.9
<0.001
3.9
0.022
6.3
0.32
0.034
0.004
10
Filter
outside duct
<0.003
1.5
X'4 b
u
~K
u
<0.001
<0.001
<0.002
0.015b
—
0.046
<0.001
<0.005
0.064
<0.005
<0.03
<0.01
<0.001
1.6
Impinger
outside duct
<0.02
0.34
8.2
<0.001
<0.04
<0.01
<0.03
<0.02
0.12
59
0.16
0.021
<0.06
<0.02
<0.03
<0.025
0.20
<0.06
0.015
a. At 21°C (70°F) and 760 mmHg.
b. Not determined.
139
-------
APPENDIX D
INLET AND OUTLET CASCADE IMPACTOR
COMPUTER DATA SHEETS
The data for the size distribution samples taken by cascade
impactors is given in this Appendix. Sampling time ranged from
30 to 40 minutes at the inlet side of the control device and
from 75 to 135 minutes at the outlet.
Inlet Samples
Date Time Part
1-12-77
1-12-77
1-12-77
1-14-77
1-14-77
1-14-77
1-15-77
1-15-77
1-15-77
1-16-77
1-16-77
1-16-77
1257
1423
1500
1005
1252
1409
1315
1357
1448
1255
1341
1440
2
2
2
2
2
2
2
2
2
2
2
2
Outlet Samples
Page
141
142
143
144
145
146
147
148
149
150
151
152
Date
1-12-77
1-12-77
1-12-77
1-14-77
1-14-77
1-14-77
1-15-77
1-15-77
1-15-77
1-16-77
1-16-77
1-16-77
Time
1015
1015
1435
1016
1145
1145
1000
1000
1410
0915
1345
1345
Part
5
4
5
4
5
4
4
5
4
4
4
5
Page
153
154
155
156
157
158
159
160
161
162
163
164
140
-------
KCCI-1 1-12-77 pnpr-2 1?57
IMPACTOR FLnWRATF = n,092 ACFM
IMPACTOR PRESSURE DROP = a'. 7 IN, OF HG
ASSUMED PARTICLE DENSITY s l'.80 GM/CU.CM.
INLET SAMPLF MODIFIED BRINK CASCADE IMPACTOR NUMBER - c
1MPACTOR TEMPERATURE = 070.0 F = 203.3 C SAMPLING DURATION = 30,00 MIN
STACK TEMPERATURE = 070,0 F = 202
6.71E+18
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE 21 DEG C AND 760MM HG.
SOUARF ROOTS OF PSI BY STAGE 0.322 0.322 0.351 0.38H 0.330 0.350 0.273
HOLE OIAMETFRS BY STAGE (CENT I METERS) 0,3658 o'.B460 0.1720 0.1360 0.0896 0.0719 0.0589
-------
H-
J^
to
KCCI-3 1-12-77 PORT-? 1423
IMPACTOP FLo«RATE = 0,078 ACFM
IMPACTOR PRESSURE DROP = 2.0 IN. OF HG
ASSUMED PARTICLE DENSITY = l'.H4 GM/CU.CM.
GAS COMPOSITION (PERCENT) co2 a
CALC. MASS LOADING = 2.5793E-01 GR/ACF
IMPACTHR STAGE CYC
STAGE INDEX NUMBER 1
050 (MICROMETERS) 8.46
MASS (MILLIGRAMS) 15,76
MG/DNCM/STAGE
CUM, PERCENT OF MASS SMALLER THAN 050 59,70
CUM, (MG/ACM) SMALIFR THAN D50
CUM, (MG/DNCM) SMAt.Lf.R THAN D50
CUM, (GR/ACF) SMALLER THAN 050
CUM. (GR/DNCF) SMALLER THAN D50
GEO, MEAN DIA. (MICROMETERS)
DM/DLOGD (MG/DNCM)
DN/DLOGD (NO, PARTKLES/DNCM)
INLET SAMPLE MODIFIED BRINK CASCADE IMPACTOR NUMBER - A
IMPACTOR TEMPERATURE s 470.0 F = 203.3 C SAMPLING DURATION s 30,00 HIM
STACK TEMPERATURE = 470.0 F = 243.3 c
STACK PRESSURE = 27,86 IN, OF HG
6.47 CO a 0.00 N
5.1887E-01 GR/DNCF
SO SI S2
2 3 a
4.86 2.67 1,72
5,63 5.29 2.15
MAX. PARTICLE DIAMETER = 7H.lt MICROMETERS
8,05
S5
7
*Z = 77,98 02
5.9024E+03 MG/ACM
S3 S4
5 6
1.13 0.36 0.30
2.77 0.13 3.79
1.73E+02 1.63E+02 6.62F+01 8,52E*01 4.00E+00
45,31 31.78 26.28 19.20 18.87 9.18
3.52E+02 2.67E+02 1.8BE+02 1.55E+02 1.13E+02 l.llEt02 5.42E+01
7.09E-I-02 5.38E + 02 3.77E + 02 3.12Ft02 2.2BE + 02 2.24E + 02 1.09Et02
l,5aE-01 1.17E-01 8.20F-02 fe'.r8E-02 4.95E-02 4.87E-02 2.37F-02
3.10E-01 2.35E-01 1.65E-01 1.36F-01 9.96E-02 9.79E-02 4.76F-02
2,51E*01 6.41E+00 3.60F400 2.14E*00 1.39E+00 6.38E-01 3.26E-01
5.1UE+02 7.21E+02 6.26E+02 3.45E+02 4.72E+02 8.02E+00 1.39E+03
3.3fiEt07 2.84E+09 1.39E+10 3.64E+10 1.81F+11 3.21E+10 4.17E+13
H20 « 7,50
1.1873E+03 MG/DNCM
FILTFR
8
3.59
r.lOE + 02
6,«6E-02
3.67Et02
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE 21 DEC C AND T60MM HG.
SQUARE ROOTS OF PSI BY STAGE 0,322 0.322 0,338 0.345 0.258 0,317 0,229
HOLE DIAMETERS BY STAGE (CENTIMETERS) 0,355a 0.2022 0,1779 0,136(1 O.OSatt 0.0705 0,0556
-------
U)
KCCI-a 1-1P-77 PORT.a 1500
IMPACTOR FI..OWRATF = o.oss ACFM
IMPACTOR PRESSURE DROP = 2.J IN, OF HG
ASSUMED PARTICLE DENSITY = r.a-o GM/CU.CM.
INLET SAMPLE MODIFIED BRINK CASCADE IMPACTOR NUMBER • o
IMPACTOR TEMPERATURF = «70,o F = 2«3.3 c SAMPLING DURATION * 30,00
STACK TEMPERATURE = «70,0 f = 2a3.3 C
STACK PRESSURE = 27,86 IN. OF HG MAX. PARTICLE DIAMETER = 74.a MICROMETERS
CO a 0.00
3.2092E-01 GR/ONCF
SO SI S2
Z 1 0
a,66 2.62 1,68
3,07 2.93 1,61
MG/DNCM/STAGF. S.O«E*02 8.67E+01 8.27E+01 .«2 jo,3i 26,ao 11.4* 0,00
GAS COMPOSITION (.PERCENT) C02 = 6.«7
CALC. MASS LOA13ING = 1.5953E-01 GR/ACF
IMPACTOR STAGE CYC
STAGE INDEX NUMBER i
D50 {MICROMETERS) 8,10
MASS (MILLIGRAMS) 10,76
N2 = 77,98 02 = 8,05
3.6506E+02 MG/ACM
S3 sa ss
5 6 7
1.1? O.afe 0.33
1,03 3,95 3,03
CUM, {MG/ACM) SMALLER THAN DSO
CUM. (MG/ONCM) SMALLER THAN 050
CUM. (GR/ACF) SMAHF.R THAN 050
CUM, (GR/ONCF) SMALLER THAN 050
GEO, MEAN OIA, (MICROMETERS)
OM/DLOGn (MG/DNCM)
ON/OLOGO CNO. PARTICLES/ONCMJ
?.16E+02 1.7UE+02 1.33E+02 1.11E+02 9,6aE+01 «,20E+01 O.OOE-01
a.35E+02 3.19E+02 2.67E+02 2.23E+02 1.9«E+02 «,««E+01 O.OOE-01
9,aaE-02 7.58E*02 5.81E-02 a,8«E-02 a.2lE-02 1.83F-Q2 O.OOE-01
1.90F-01 1.53F.-01 1.17E-01 9.73E-02 8.a7E-02 3.69C-02 O.OOE-01
?,a5E*01 6.15E'*00 3.a9E + 00 2.10E + 00 1.37E + 00 7,1*>E-01 3.92F.-01
3.15E + 02 3.61E + 02 3.30E + 02 2.36E + 02 1.6aE + 02 2.87E+02 6,2aF. + 02
2.2lEt07 1,62E*09 8.oaE+09 2,65£*10 6.63E+10 8.13E*11 l.OBF+13
H20 i 7.SO
7,3a36E*02 MG/DNCM
FILTER
a
0,00
O.OOE-01
5.55F-02
O.OOE-01
O.OOE-01 .
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE 21 DEG C AND 760MM HG,
SQUARE ROOTS Of PSI BY STAGE 0.322 0.322 0.3«6 0.35a 0.297 0,337 0.22*
HOLE DIAMETERS BY STAGE (CENTIMETERS) 0.3560 0.2«61 0,1778 0.1368 0.0937 0,0739 0,0550
-------
KCCI-5 1-1U-77 PORT-? 1005
IMPACTOR FLClrtRATF = 0.1 OH ACFM
1MPACTOR PRFSSUPF DROP = 3.8 IN. OF HG
ASSUMED PARTICLE DENSITY = l.Rtt KM/CU.CM.
INLET SAMPLE MODIFIED. HRINK CASCADE IMPACTOR NUMBER - D
IMPACTOR TEMPFRATURE = 170.0 F = ?<43,3 C SAMPLING DURATION = 30.00
STACK TEMPERATURE * "70.0 F = 2«3.3 C
STACK PRESSURE = 27.86 IN. OF HG MAX, PARTICLE DIAMETER * 7«.
-------
I-1
*>
U1
CO = 0.00
7.16B«F-01 GR/DNf.F
KCCI-7 1-14-77 PORT-? 1?52
IMPACTOR FLOIPATF = Q.OS3 ACFW
IMPACTOR PRESSURE DROP = ?.3 IN, OF HG
ASSUMED PARTICLE OFNSITY = 1.84 GM/CU.CM.
GAS COMPOSITION (PERCENT) C02 = 6.5S
CALC. MASS LOADING = J.*300E-01 GR/ACF
IMPACTOR STAGE CYC
STAGE INDEX NUMBER 1
050 (MICROMETERS) 6.15
MASS (MILLIGRAMS) 2».90
MG/DNCM/STAGE B,20Et02 2.07E+02 2.40E+02 9.03C+01
CUM. PERCENT OF MASS SMALLER THAN D50 50.66 38.19 23.75 18.32 14.43
TNLEt SAMPLE MODIFIED BRINK CASCADE IMPACTOR NUMBER - A
IMPACTOR TEMPERATURE = 450.0 E = ?3?.? C SAMPLING DURATION = 50,00 HIN
STACK TEMPERATURE = 45Q.O F = ??2.? C
STACK PRESSURE = 27.86 IN. OF HG MAX. PARTICLE DIAMETER *
N2 = 77,36 n;
8.3067E+02 MG/ACM
= 8.30
SO
p
4,67
7,30
SI
3
2.57
8.46
S?
4
1.65
3,18
S3
5
1,09
2.28
34
6
0.34
3.18
Sb
7
0,28
3,1*
9.03E+01 8,97Et01
9,00 3,60
CUM, (MG/ACM) SMALLER THAN 050
CUM, (MG/ONCM) SMALLER THAN 050
CUM, (GR/ACF) SMALLER THAN oso
CUM, (GR/oNcF) SMALLER THAN 050
GEO, MEAN DIA. CMICRQMFTERS)
DM/OLOGD (MG/D^CM)
ON/OLOGD (NO. PART1CIES/ONCM)
fl.?lF. + 02 3.17E + 02 1 .97F + 02 1.52E + 0? 1.20E + 02 7.07E + 01 2.99E + 01
8.31E+02 6.27E+02 3.90E+02 3.01E+02 ?.37E+02 1.48Et02 5.91F+01
1.B4F-01 J.39E-01 8.62E-02 6.65F-02 5.24E-02 3,?7E-02 1.31F-0?
3.63E-01 2.74E-01 1.70E-01 1.31E-01 1.03E-01 6.45E-02 2.58E-02
2.U6E+01 6.16E+00 3.16F+00 2.06E+00 1.34E+00 6.11E-01 3.I2F-01
8.53E+02 8,61Et02 9.23E+0? 4.70E+02 3.58E+0? 1.80E+0? 1.06E+03
5,95E*07 3.82E + 09 2.31F. + 10 5.60E + 10 1.55E+11 8.20E-H1 3.63E + 13
74.4 MICROMETERS
H20 = 7,80
1.6404F.+03 MG/DMC^
FILTER
fl
2.11
5.99E+01
6,03t-0?
1.99E+02
9.40E+14
NORMAL (ENGJNFERING RTANOARD) r.ONMTIONS ARE 21 DEG C AMD 760MM MC.
SQUARE ROOTS OF PST BY STAGE 0.3M « *™
HOLE. DIAMETERS HY STAGE (C£Nr TMETf:RS) 0,355. 0.«22
«•"»
0.1779
0.345
0 . t J6«
0.25«
0.0«B4
0.317
O.n705
0.229
0.055*
-------
.t*.
CTi
KCXI-B l-l/J-77 PORT-i> 1109
IMPACTOR FLOWPAIF = 0,073 AC1FM
IHPACTOH PRESSURE PROP = l.fi IN. OF HG
ASSUMED PARTICLE DENSITY = 1.81 GM/CU.CM. STACK PRESSURE = ?7,87 IN. OF HG
GAS COMPOSITION (PERCENT) C02 = ft.55 CO s 0.00
TMIET SAMPLE MODIFIED BRINK CASCADE IMPACTQR NUMBER - C
IMPACTOH TFMPERATURF = «so.o r = 23?.2 r. SAMPLING DURATION = 50,00
STACK TEMPERATURE = 150.0 F = ?3?.2 C
CA(_C. MASS LOADING e 1.6031F-01 GR/AtF
IKPACTOR STAGE
STAGE INDEX NUMBER
CYC
3.1h47£-0t
SO SI
S?
B,A7
MASS (MILLIGRAMS) 7,90
MG/DNCM/STAGE ;
CUM, PERCENT OF MASS SMALLER THAN D50 65,27
5.22 2.81
1,18 2.30
3.81F+Q1 '
60,09 19.98
1.7?
3,06
5
1,32
1,08
«AX. PARTICLE DIAMETER a
77.36 02 = 8,3C
3.66B5F+02 MG/ACM
S3 51 SS
6 7
0,37
36,53
CUM. rMfi/AC«M SMALLER THAN 050
CUM., (MG/DNCM) SMALL EP THAN 050
CUM, (GR/ACF) SMALLER THAN DSO
CUM, CGR/D^CF) SMAiLEf THAN DSO
RED, MEAN 01*. (MICROMETERS)
OM/DLOGD tMG/(5NC.M)
(NO. PARTICLES/DNCM)
2.39E + 02 2.20F + 02 1.S3E + 02
0.53
1.76
3.18E+01 '
31.78 21.01 11,30
1.I7E+02 B.S2F+01 4.11E+01
1.73F+02 1.35E+0? 3.62E+02 2.65F+02 2.30E+02 1.71E+02 B.lflEtOl
1.05E-01 9.63E-02 8.01E-02 5'.86F-02 5.09E-02 3.85E-02 1.81E-02
2.07F-01 1.90E-01 1.58F-01 1.16F-01 1.01E-01 7.61E-02 3.58E-0?
2.5«E+01 6.73E+00 3.83E*00 2.22E+00 1.52E+00 8.37E-01 1.13F-01
2.73E + 02 1.73E + 02 ?.77E*02 1.79E + 02 2.88E + 02 1,13F*02 6.03F. + 02
1.73E+07 S.89E+08 5.11E+09 1.55E+10 8.18E+10 2.52E+11 7.21E+12
71.1 MICROMETERS
H20 = 7.80
FILTER
B
2.57
fi.29E+01
1.09E-01
2.7SF*02
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE 21 DEC C AND 760MM HG.
SQUARE ROOTS OF PSI HY STAGE 0,322 0.322 0.351 0.388 0.330 0,350 0,273
HOLE DIAMETERS BY STAGE (CENTIMETERS) 0.36S8 0.2fl60 0,1721 0,1360 0.0896 0,0719 0,0589
-------
KCr.I-10 1-1S-77 POPT-;> J315
IMPACT-OR f'LOWRATF = 0.083 ACFM
IMPACTOR PRESSURE DROP = 2.3 IN. OF HG
ASSUMED PARTICLE DENSITY = t'.ea GM/CU.CM'.
IMLFT SAMPIE MoniFIED BRINK CASCADE IMPACTOR NUMBER - C
IMPACTOR TEMPERATURE = 450,0 F = 21?,? C SAMPLING DURATION = 30.00 MIN
STACK TEMPEHATURf = 450,0 F = 232.2 C
STACK PRESSURE = 37,87 IN, OF HG MAX. PARTICLE DIAMETER i
GAS COMPOSITION (PERCENT) , C02 = 9.43
CALC. MASS LOADING = 3.1218E-01 GR/ACF
IMPACTOR STAGE CYC
STAGE INDEX NUMBER 1
D50 (MICROMETERS) 8.08
MASS (MILLIGRAMS) 23,76
MG/DNCM/STAGE
CUM, PERCENT OF MASS SMALLER THAN DSO 52.83
CUM. ("t;/ACM) SMALLER THAN 050
CUM. (MG/DNCM) SMAU.F.R THAN 050
CUM, (GR/ACF) SMALLER THAN 050
CUM, (GR/oNcF) SMALIER THAN 050
GEO, MEAN DIA. (MICROMETERS)
DM/DLOGD (MG/ONCM)
DN/OLOGO (NO, PARTICLES/DNCM)
4
5
6.26'
SO
?
.86
.58
CO = 0.00
i5E-oi GR/DNC'F
S!
3
2.62
2.86
S2
4
1,62
3.70
N2 = 76.55 02 = 4.72
7.1438£t02 MG/ACM
S3 S4 85
567
1.23 0.49 0,34
2.60 5.30 3,74
1.07E+0? 7.50F+01 1.53E+02 1.08E+02
41.75 36.07 28,73 23.57 13.04 -5.62
3.77E + 02 2.9RF + 02 2.5RE + 02 2.Q5F. + 02 1.68F + 02 9.32E + 01 4.01E + 01
7.57F + 02 5,99Ft02 5.17E + 02 4.12E + 02 3.38E402 1.87E + 02 8.05F. + 01
1.65E-01 1.30E-01 1.13E-01 8.97E-02 7.36E-02 4.07E-02 1.75E-02
3.31E-01 2.62E-01 2.26E-01 1.80E-01 1.48E.01 8.17F-02 3.52F-02
2.45E + 01 6.26E + 00 3.56E + 00 2.06E + 00 1.41E-I-00 7.72E-01 4.05F-OJ
7.11F. + 02 7.28E + 02 3.07E + 02 5.16E + 02 6.17E + 02 3.80E + 02 6.81E + 02
5.01F+07 3.07E+09 7.0UE+09 6.11F+10 2.28E+11 8.56E+11 1.06E+13
74.4 MICROMETERS
H20 = 9.30
1.4335E+03 MG/DMCM
FILTER
8
2.83
B.16E+01
9.47E-02
2.71E+02
3.31E+14
NORMAL (FNGINFERING STANOAR!)) CONDITIONS ARE 21 DEC C AND 760MM HG.
SQUARE ROOTS OF PSI BY STAGE. 0.322 0,322 0.35! 0,388 0.330
HOLE OTAMETF'PS BY STAGF (CENTIMETERS) 0.3658 0.2460 0.1724 0,1360 0.0896
0,350
0.0719
0.273
0.05R9
-------
00
KCCI-ft t-f5-77 PDRT-2 1157
tHPACTrm FLOiiRATF a O.Ottlj ALFM
IMPAC.TOP PRESSUPE DKOP = 2.4 IN. OF HG
A&SUMEO PARTICLE DENSITY = 1.6a 6M/CU.CM.
INLET SAMPI F MODIFIED BRINK CASCADE JHPACTOR NUMBER - o
IMPACTUH TEMPERATURE 3 476,0 F = 216.7 c SAMPLING DURATION » 30,00 MIN
STACK TEMPERATURE s 476.0 F * ?46.7 C
STACK PRESSURE = 27,86 IN. OF HG MAX. PARTICLE DIAMETER = 74.« MICROMETERS
CO = 0.00
9.175«E-01 GR/ONCF
GAS COMPOSITION (PERCENT) CO? = 9.43
CALC, MASS LOADING = 4.4437E-01 GR/ACF
IMPACTt'R STAGE CYC
STAGE INDEX NUMBER i
MASS (MILLIGRAMS) 41,98
MG/DNCM/STAGF 1.20E+03 1.89E+02
CUM, PERCENT OF MASS SMALLER THAN 050 43,49 34,61 ?2.fl2
4,42Et02 3.52E+02
N2 B 76.55
SO
?
«,62
6.60
S!
3
2.59
8.76
S2
4
1,66
2.84
S3
5
1.11
1.94
CUM. (MG/ACM) SMAL.I EW THAN 050
CUM, (MG/ONCM) SMALLER THAN DSO
CUM, (GR/ACF) SMALtER THAN 050
CUM, (GP/nNcF) SMAlLER THAW 050
GEO, MEAN DIA. (MICROMETERS)
DM/DUOGD (MG/ONCM)
NO, PARTTCLES/ONCM)
02 = 4,72
MG/ACM
34 S5
6 7
0.45 0,33
5.84 3,46
8.14E+01 5,56E*01 1.6TE+02 9.91E+01
18,99 16.3R 8.52 3,86
1.93Ft02 1.67F+-02 8.66E + 01 3,93E»01
9.13E+02 7.2TE+02 4.79E+Q2 3.99E+02 3.44E+02 1.79F+02 8.11E+01
1.9JE-01 1.54E-01 1.01E-01 8.44E-02 7.28E-02 3.79E-02 1.72F.-02
3.99E-01 3.1SE-01 2.00E-01 1.74F-01 1.50E-01 7.82E-02 3.54E-02
2,41E*01 6,09EtOO 3.46E+00 2.08E+00 1.36E+00 7.08E-01 3.87E-01
1.P4E+03 7.88E+02 l.OOE+03 4,22F*02 3.13E+02 4.33E+02 7,22Et02
8,85E*07 3,63E*09 2.51E+10 4.89E+10 1.31E+11 1.26E+12 1.29E+13
H?0 i 9.30
2.0996FtOS MG/ONCM
FILTER
8
2,87
8.22E+01
5.46F-02
2.73E+02
1.74E+15
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE ?t DEC C AND 760MM HG.
SQUARE ROOTS OF PSI BY STAGE 0.322 0.322 0.346 0.354 0.297 0.337 0.2?6
'HOLE DIAMETERS RY STAGE (CENTIMETERS) 0.3560 0.2461 0.1778 0.1368 0.0937 0,0739 0.0550
-------
VD
KCCI-1? 1-15-77
IMPACTOR FI O^RATF = o.o«i ACFM
IMPACTdR PRESSURE DROP = 2.1 I*1'. OF HG
ASSUMED PAHTlCIF DENSITY = 1.84 GM/CU.CM.
INLET SAMPLE MODIFIED BRINK CASCADE IMPACTOR NUMBER - A
IMPACTOR TEMPERATURE .= a?6.o F s 2«6.7 c SAMPLING DURATION = 50.00 MIN
STACK TEMPERATURE = 476.0 F = ?4h,7 C
STACK PRESSURE = 27,86 IN, OF HG MAX. PARTICLE DIAMETER =
GAS COMPOSITION (PERCENT)
CALC, MASS IOAOING = 1
IMPACTOR STAGE
STAGE INDEX NUMBER
050 (MICROMETERS)
MASS (MILLIG^A^S)
MG/DNCM/STAGE 5,47Et01 1.05E+02 1.72E+02 S.96E+01 6.27E+01
CUM, PERCENT OF MASS S«ALLFR THAN 050 92.38 7?.13 48.24 39.94 31.22 20.20
C02 s Q.13
GR/ACF
CYC
t
8,P7
1,80
3
«,
a.
CO =
0,00
.0960E-01 GR/ONCF
SO
2
75
78
2
5
SI S2
3 4
.61 1,68
.64 1,96
N2 = 76,55
3.4312E+02
S3
5
1.11 0.
2,06 2.
MG
54
6
35
60
02 a 4,
/ACM
85
7
0,29
2,78
72
CUM, (MG/ACM) SMAI.IEP THAN oso
CUM, (MG/D^CM) SMALLER THAN 050
CUM, (GR/ACF) SMALLER THAN 050
CUM, (GR/DNCF) SMALLER THAN 050
GEO, MEAN DIA. (MICROMETERS)
DM/DLOGD CMG/DNC")
DN/DLOGD (NO. PARTICLES/DNCM)
8,43
2.89Et01
3.17E + 02 2,'47E + 02 1.6ftE*0? 1.37E + 02 1.07E + 02 6.93E +
6.5UE+02 5.11E+02 3.«?E+0? 2.P3E+02 2.21E+02 l.a3F*02 5.97E+01
1.39F-01 1.08E-01 7.23E-02 5.99F-02 4.68F-02 3.03E-02 1.26E-02
2.R6E-01 2.23E-01 l.«9E-01 1.2UE-01 9.A6E-02 6,2faF-02 2.61E-02
2,asEt01 6,27EtOO 3.52E+00 2.09E+00 1.36E+flO 6.21E-01 3.17F-01
5,7aE+01 A,0«E+02 6.S9E+0? 3.10E+02 3.46E+02 1.58E+02 l.OOF+03
3.90E + 06 2.55E + 09 1.S7E + 10 3,^2E'flO l.«3E + l.l 6.8«F+11 3.27F + 13
.ti MICROMFTERS
H20 a 9.30
MG/ONCM
FILTER
8
1.99
6.05E+01
6.18^-02
2.01E+02
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE 21 OEG C AND 760MM HG,
SQUARE ROOTS OF PS! BY STAGE 0,322 0.322 0,338 0,345 0.258 0,317 0,229
HOLE DIAMETERS NY STAGE (CENTIMETERS) 0.3SS4 0.2422 0.1779 0.13M 0.0884 0,0705 0.05S6
-------
Ul
o
KCCI-14 1-16-77 pnpT-2 1755
JMPACTnp FLOKRATF' = 9.966 ACFM
IMPACTOR PRESSISRE DROP = 1.5 IN. OF HG
ASSUMED PARTICIE DENSITY = l.B« GM/CU.CM,
INLET SAMPLE MODIFIED BRINK CASCADE IMPACTOR NUMBER - C
TMPACTOR TEMPERATURE = '476,0 F = ?46.7 C SAMPLING DURATION r 30,00 WIN
STACK TEMPERATURE = 176.0 F r 216,7 C
STACK PRESSURE e 27,S6 IN. OF HG
MAX. PARTICLE DIAMETER a
GAS COMPOSITION fPF.RCENT) CO? = fl.«9
CALC. MASS LOADING = 4.63SOE-01 GR/ACF
IMPACTOR STAGE: CYC
STAGE INDEX NUMBER 1
050 (MICROMETERS) 9,03"
MASS (MILLIGRAMS) 3?.71
N2 a 75.2? 02 = 6.59
1.0606F+03 MGXACM
S3 SI S5
5 fc 7
1.38 0.55 0.39
3.tO 5.65 5.01
CO = 0.00
9.6!?9O01 GR/DNCF
SO SI 32
2 3 a
5,a,bi 36,ea 25.«6 23,29 17,71 8.52 o,3«
CUM, (MG/ACM) SMALLER THAN 050
CUM, (MG/DNCM) SMAll_Ef THAN D50
CUM. (GR/ACF1 SMALLER THAN 050
CUM. (GR/DNCF) SMALUFP- THAN 050
GEO, MEAN DIA. (MICROMETERS)
DM/DLOGD (MG/DNCM)
DN/OLOGO (NO. PARTICLES/DNCMI
'J.94E + 02 3.91E402 ?.70Ft02 ?,07 2.98E+09 1.5UE+10 1.96F+10 P.65E+11 8.01E+11 1.P3E+13
U.ft MICROMETERS
H?0 s 9.70
2.1997E + 03
FILTER
8
.6af +00
1.16F-01
2.54E+01
1.69Ftl3
NORMAL (ENGINEERING STANDARD) CONDITIOMS ARE 21 DEC C AND 760MM HG.
SQUARE ROOTS OF PSI BY STAGE ft.3?? 0.32? 0.351 0,388 0.330 0,350 0.273
HOLE DlA.MFTf.RS BY STAGE (CENTIMETERS) 0,3658 0.2U60 0,1721 0.1360 0.0896 0,0719 0,0589
-------
Ul
KCCI-IS 1-16-77 POMT-2 13»1
IMPACTOR FLOWRATE - 0,086 ACFM
PRESSURE nwnp = 2.« IN. OF KG
PARTICLF OFNSJTY = 1.«U GM/CU.CM.
GAS COMPOSITION- (PERCENT) C02 = 8
CALC. «ASS LOADING = 2.9U86F-01 GR/ACF
IMPACTOR STAGE CYC
STAGE INDE* NUMBER t
050 CMICROMfT£RS) P.01
MASS (MILLIGRAMS) 25,40
INLET SAMPl E
TEMPFPATURE a 470,0 F =
STACK TEMPERATURE r 470.0 F = ?4
STACK PRESSURE = 27,S3 IN. OF HG
49 CO = 0.00
01 GR/DMCF
SI S?
3 a
2.53 1.62
10.03 2.75
SO
2
fl, 60
7.61
CUM. PERCENT OF MASS SMALLER THAN 050 59.31
«6.o8
?8.6«
23, s&
CUM. (KG/ACM) SMALLER THAN 050
CUM. (MG/DMcM) SMALLER THAN DSO
CUM, (GR/ACFT SMALLER THAN P50
CUK. fGP/DNCF) SMALLFH THAM 050
5EO. MFAN DIA. (MICROMETERS)
DM/DLOGD CMG/DNCM1
ON/DLOGO (NO. PARTULESYDNC>n
«.nOE+02 3.1JE+02 1.93E+02 1.61E+02
B.26E+02 6.S1F+02 3.99E+0? 3.3PE+0?
1.75E-OI 1.36E-01 8.UOE-02 T.03E-02
3.61E-01 ?.HOF.-(H 1.7«E-OJ J.4SF-01
2.a«E+01 6.07E+00 3.U1E+00 P.02E+00
5.93E + 02 7.75E + 02 9.U5E-t02 3.51F + 02
O.P3E+07 S.S^F+Og 2.4BF+10 fl.39E+10
MODIFIED BRINK CASCADE IMPACTOR N.UMBFR - A
243.3 C SAMPLING DURATION = 35,00 HIM
5.^ C
MAX. PARTICLE DIAMETER = 7«,4 MICROMETERS
M? = 75,22 02 = 6,59 H2Q = 9.70
6.74-73E + 02 MG/ACM 1.3919F + 03 MG/DNCM
S3 S4 S5 FILTER
567 8
1.07 0.33 0,27
1.41 5,57 a,57 2.17
3.4feE+01 1.37F+02 1.12E+0? b.32E+0>
21,«0 11.72 3,77
1.4UF+0? 7.91F+OI 2.55E+01
2.98E + 02 1.63E + 02 5,-?5E-fOl
6..3IE-02 3.46E-02 1.11E-02
1.30E-01 7.13F-02 2.30F-02
1.52E+00 5.98E'01 3.03E-01 5.76F-0?
I.9JE+0? P.71E+02 1.31E+03 1.77E+02
H.69F + 10 U32E + 1? ft.87E+13 9.63F + 14
NORMAL CENGINEFfUNG STANDARD) COMDITIOMS ARE 21 DEC C AMD 760MM HG.
SQUARE ROOTS OF PST BY STAGE 0.32? 0.32? 0.358 0.3U5 0.258 O.J17 0.229
HOLE DIAMETERS BY STAGE (CEMTIMETf-RS) 0.355M O.?tt?2 0,1779 0.1364 0.08fl4 0,0705 0,0556
-------
rc
z: ^
Z3 O
t—
o: -
tn
2.
C
^
or
fl.
C
ul
5^
CO
(/}
<
•FKCEIMT) C
U-
0
t~
*— i
a
i
C:
(_>
CO
W5
<
21
U»
45
<
cn
x
i
tj
a1.
o
X
LD
s:
IT
0
Z
—
'2
(_.
o:
U-1
Q..
£
ID
U
0
in
o.
,_.
t_!
O
Lfl
O
^
<
X
»—
a
UJ
<
2:
tc
u_
CJ
X
a:
d
— •
TL
r?
u
o
IT)
Ci
a:
-------
U1
U)
KCCO-1 1-12-77 PdKT-S I01S
1MPACTOH FLtiwRATE = 0.^93 4CFM
IMPACTOR PRfTSSHKh [>«OP = 0.6 TM. OF HG
ASSUME!) PARTICLF Df.NSlTV s 1'.81 CM/CU.CM
GAS coMposninw CPFRCEN!!) ctv
CAtC. MASS LOADING = <5.6h98E-03 GH/ACF
IMPACTOR STAGE
STAGE INDEX NUMBER
D50 (MICROMETERS)
MASS (MILl IPRiMS)
MG/DSCM/STAfiE
CUM. PERCENT OF MASS SMALLER THAN oso
CUM, (MR/AC!*) SMAI.I t'R THAN D50
CUM, f-G/ONCM) SMALLER THAN D50
CUM. fGR/ACF) SMALLER THAN D50
CUM, (GR/nNCF) SMALLER THAN D50
GEO, MEAN OIA. (MICROMETERS)
OUTLET SAMPLE U, OF W. MARK III SOURCE TF.ST IMPACTOR NO, - C
IMPAC.TOR TEMPERATURE = 335,0 F = if>8.3 c SAMPLING DURATION = 120,00 MI
STACK IFMPf.RATlIRE = 335,0 F = tb8.3 C
STACK PRESSURE s 27,86 IN, OF HG MAX. PARTICtE PIAME.TER s 11.0 MICROMETERS
N2 = 77,98 02 = fl,05
2.2128E+-01 MG/ACM
S7
7
0.26
5.78
DN/DLOGD (NO. PARTICLES/DNCM)
= 6.17 C 0 = 0.00
1.6627E-02 GR/DNCF
SI S2 S3 SI S5 S6
123156
10.82 11.11 1,72 2,03 1,05 0.73
7.82 1.17 1.08 1,88 1.14 1,92
1.02E + 01 1.92E + 00 5.32E + 00 2.15E + 00 1.88E + 00 2,51EtOO 7,'ilE + OO 6.73E+00
73.51 68,56 SI.75 48.39 13.52 37,02 17,16
1.63E+01 1.52E+01 1.21F. + 01 1.07E+01 9.63E+00 8.19E+00 3.66E+00
2.80E + 01 ?.61Et01 2.08E + 01 l.SIEt-01 1.66E + 01 1.41E + 01 6.61E + 00
7.11E-03 6.63E-03 5.29F.-03 4.68FJ-03 1.21F-03 3.58E-03 1.69E-03
1.22E-02 \.11E-02 9.10E-03 8.05F-03 7.2UE-03 6.16E-03 2.90E-03
1.23E+01 1.10E+01 7.24E+00 3.09E+00 1.16E+00 8.79E-01 1,10E-ni 1.87F-0!
9.11E + 01 -1.70E + 0? 1.13E + 01 6.70 (r + 00 6.61E + 00 1.60F + 01 1.70E + OJ 2,24Ft01
5.08E+07 -1.33E+08 3.92E+07 2.35E+08 2.20E+09 2.11E+10 2.07E+11 3.58E+12
H?0 x 7.50
3.8oi8F+oi MG/DNCM
FILTER
a
5.16
NORMAL (ENGINEERING STANflAHU) CONDITInMS 4RF. 21 DEC C AND 760>1M HG.
SHUAPF POO'S OF PSI BY STAKE 0.111 0.330 0.371 0.320 0.295 0,363 0,312
HOLE PlAMfTFCS HY STAGE (CENT JMFTi'RS) 1.8237 O.SB7a 0.2fl59 0,0807 O.OS52 0,0376 0.02*0
-------
KCCD-2 1-12-77 PtlRT-a 1015
ft Fl nt'RATF r 0,"J9.I ACFM
PKESSL(RF HKOP = 1.0 IN. OF HG
ASSUMED PARTICLE PFNSITY s I'.Bfl GM/CU.CM
GAS COMpnsTTI(.U. fPFRCFMH C02
CALC. MASS LOADING = 5.5090E-03 GR/ACF
IMPACTQR STAGE
STAGE INDEX NUWBFR
050
MASS
MG/I3SCM/STAGF
CUM. PfRCENT OF MASS SMALtER THAM 050
CUM, (^G/ACM) SMALLER THAN 050
CUM, CMG/DNCK) S«Al LFR THAN D50
CUM. CGR/ACF1 SMALLER THAN 050
CUM, (GR/DNCF) SMALLER THAN 050
GEP, MFAN DIA. (MICROMETERS)
DM/OLORD (MG/ONCM)
(NO, PARTICLES/ONCM)
OUTLET SAMPLE u, OF w, MARK m SOURCE TEST IMPACTOR NO, - A
IMP-ACTOR TEMPERATURE = 335.0 F = 168.3 c SAMPLING DURATION = 120,00 MIN
STACK TFMPERATURE = 335,0 F c 168,3 C
STACK PRFSSURF = 27.86 IN. OF HG MAX. PARTICLE DIAMETER = 14.0 MICROMETERS
SI
1
9.67
a.59
CO s O.Of)
9.4724F-03 GR/DNCF
S2 S3
2 3
9,65 a.32
0.28 0.47
s 6.47 CO s O.Of) N2 a 77.98 02 = 8.05 H20
I.2606F+01 MG/ACM 2.1676E+01
54 SS S6 S? FILTER
4 5 6 7 8
1.51 0.95 0,53 0.23
3.22 2.35 0.87 2.18 7,09
4.79E+00 2.92E-01 4.91E-01 3.36E+00 2.45E+00 9.08E-01 2.28EtOO 7,4'OE + OO
7R.19 76.86 74.63 59,33 48,17 44,04 33.68
9.B6E + 00 «>,69Etf)0 9,«1E + 00 7,a8F + 00 6.07E + 00 5.55E400 4.25E + 00
1.69E+01 1.67E+01 1.62E+01 1.29F+01 1.04F+01 9.55E+00 7.30E+00
4.31E-03 4.23E-03 4.11E-03 3.27F-03 2.65E-03 2.43E-03 1.86E-03
7.41E-03 7.28E-03 7.07E-03 5.62F-03 4.56F-03 4.17E-03 3.19E-03
1.16E + 01 9.66E + 00 6.45E*00 2.55F + 00 1.20F + 00 7.12E-01 3.49F.-01 1.62F-01
?.98E+01 4.29E+02 1.40E+00 7.37E+00 1.2?Et01 3.60E+00 6.22E+00 2.46E+01
1.96E + 07 'J,94E + 08 5.42E + 06 4.60F. + 08 7,35Et09 1.04E+10 I.51E + 11 6.00F + 12
7.50
NORMAL (KNGINEtRI'MR STANDARD) CONOtTIONS APE 21 HEG C AND 760MM HG.
SQUARE ROOTS nF PST BY STACK 0.1«1 0,330 0,371 0,271 0,308 0,373 0,3«9
HOLE OTAMETERS BY ?TAGE (CFNTI MET(-RK) i.S?37 0,^768 0,2501 O.OR06 O.OS24 0,0333 0.02«5
-------
U1
Ul
KCCO-J l_l?-77
IMPACIOR ftov-RATE, = o.42« ACFM
IMPACTOP PRrSSURF DROP = 0.7 IN. OF HG
ASSUMED PARTICLE DENSITY = 1.84 GM/CU.CM.
HUTLFT SAMPI f U. OF W. MARK III SOURCE TFST IMPACTOR NO, " B
IMPACTOR TEMPERATURE = 350.0 F = 176,7 C SAMPLING DURATION s 120.00 fI
STACK TEMPERATURE = 350.0 F = 176.7 c
STACK PRESSURE c 27.86 IN, OF HG MAX. PARTICLE OIAMETEft = 14.0 MICROMETERS
SI
1
10.49
0.73
GAS COMPOSITION (PERCENT) C02 = 6.47
CALC, MASS Lf*niNG = 6.64HE-03 GP/ACF
IMPACTHR STAGE
STAGE INDEX NUMBER
050 CMIcRnMgTERS)
MASS (MILLIGRAMS)
MG/DSCM/STAGF:
CUM. PEWCFNT OF MASS SMALLER THAN 050
CUM. (MG/ACM) SI-UILER THAM DSO
CUM, (MG/DNCM) SMALLER THAN nso
CUM, (GR/ACF) SMALLER TH*N D50
CUM, (GR/DNCF) SMALLER THAN 050
GEO, MEAN OIA, (MJCROHETEHS)
DM/DtOGD (MG/DNCM)
DN/DLOGD (NO. PARTTCLF.S/DNCM)
N2 = 77.98 02 =
1.5197E+01 HG/ACM
8.05
S4
4
1.95
4.27
S5
5
0.99
2.04
S6
6
0,54
3.79
S7
7
0.43
0.52
FILTER
8
1.47
CO = 0.00
1.163SE-02 GR/0NCF
S2 S3
2 3
9.65 '4,56
1.52 7.54
9.00E-01 1.87E+00 9.30E+00 5.27F+00 2.52E+00 4.67E+00 6.41.E..01 1.81E+00
96.66 89,7? 55.26 3S.74 26.42 9,10 6.72
1.47E+01 1.36E+01 8-.40F + 00 5,43EtOO 4.01F+00 1.38E+00 1.02E+00
2.57E+01 2.39E+01 1.47E+OJ 9.52F+00 7.03E+00 2.42E+00 1.79F. + 00
6.42E-03 S.96E.O-J 3.67F-03 2.'57E-03 l.TSf-05 6,0«E-Ofl a.«6E»0«
I.12F-02 l.OflF-05 6,«F-0* a.!6E-03 3.07E-03 1.06E-03 7,«2E-04
1.?.1F + 01 t.OlEtOl 6.63F + 00 2.99E + 00 1.39E + 00 7.35K-01 «.B3E-01 3.03E-01
7.19E+00 5.1«E+01 ?.86E*0! 1.S3E+01 8.53F+00 J.flOE+ni 6.14E+00 6.02E+00
U.19Et06 5.24F + 07 1.02F + 08 5.S7F + OB 3.29E + G9 4.71E + 10
H20 s 7.50
NORMAL (ENGINEERING STANDARD) CONDITIONS ARE ?1 DEC C AND 760MM HG.
SQUARE ROOTS OF PSI 8V STAGE O.l«fl 0.300 0.371 0.322 Q.313 0.340 0,337
HOtE DIAMETERS RY STAGE (CF. NTIMETF RS) 1.8237 O.SB22 0.2458 0,0802 O.OS04 0.0340 0.0323
-------
U1
KCCO-6 1-121-77 pnpT-41 1016
T M P A C T 01-' F I. 'V,ii R A 1 F_ a 0 . '4 b 1 A C F M
TriPACTOk PKF 'Ssu^F DROP = 0.8 I'*, OF HR
ASSimeo PARTICLE DFNSITY = T.84 KM/CD.CM,
GAS COMPOSITION (PFPCt.MT) CO? :
CAI.C. "ASS LOADING s 9./J741E-03 GR/ACF
IMPACIOM STAGE
STAGE INDEX NUMHFR
OUTLET SAMPLE u, OF «, MARK TII SOURCE TEST JMPACTOR NO, - n
IMPACTOP TFMpFptruRE - 33b.o r = 168.3 c SAMPLING DURATION r 135,00 M
STACK TFMPERATURF - 335.0 F = 168,3 C
STACK PRESSURE = 27.93 TN. OF HG MAX. PARTICLE OIAMETtR =
MASS CHILL
MG/DSCM/STAGE
CUM, PERCF.NT OF »-ASS SMALI E« THAN nso
CUM, (MG/ACH) S"AI l.f'R TH4M D50
CUM, (MG/nNCM) SMAI.LF" THAN 050
CUM. fGR/ACF) SMALLER THAN 050
CUM, fGP/pNf.F) SMAILER THAN D50
GEO, MEAN DJA. (MICROMETERS)
N2 = 77,36 02 = 8.50
2.16HOE+01 MG/ACM
S4
4
1.81
1 .86
SS
S
0,94
1.27
S6
6
0,57
S79
S7
7
0.21
0.76
FTLTFR
8
2.56
(NO, PA"TTCLES/ONCM)
: fc.'iS CO I 0.00
1.6302F-0? GR/DIMCF
SI S 2 Si
t 2 3
9.9H 9,90 4,49
15.39 4.85 5.76
1.52F+01 4.80E+00 5.70FfOO l,B4EtOO 1.26E+00 5,73f+00 7.52E-01 2.53F+00
59.75 «7.07 32.01 27,14 23.82 8,68 6.69
1.30Et01 I.02E+01 6.94F+00 5.88F+00 5.16E+00 1.88F+00 1.45E+00
2.23F + 01 1.76F + 01 1.19F + 01 1.01E + 01 8.B9F + 00 3.2'4E + 00 2.50E + 00
5.66F-03 4.46E-03 3.03E-03 2.57E-03 2.26E-03 B.23E-04 6.34E-04
9.74F-03 7.67E-03 5.22F-03 4.43F-03 3.SHF-03 1.42E-03 1.09E-03
l.lflE+01 9.94E+00 6.66E+00 2.85E+00 1.30F+00 7.34F-01 3.49E-01 1.50E-01
1.03E + 03 1,35E-(03 1.66E + Q1 4.65F + 00 4,4?E + 00 2.68F + 01 1.7«E + 00 8.41E + 00
6.50E+07 1.43E+09 5.82E+07 2.09E+08 P.OBE+09 7,05FtlO a.25F+10 2.57E+12
i'4.0 MICROMfTERS
H20 = 7.60
3.7305E+01
tf MGINFFHl^G STANDARD) CONDITIOMS ARE 21 DEG r A'-)D 760MH MR.
SQUARE ROPTS fit PS! BV STAGE 0.114 0,330 0.371 0,319 0,321 0,389
HOLE DtiMplHvs HY RTAliE (CFNTI MF TEHS) 1.8237 0.57«3 0.2S12 0,0793 0,0495 0,0330
0,354
0.0229
-------
KCCO-7 1-14-77 PnH7-5 1145 OUTLET SAMPLE U. OF W, MARK III SOURCE TEST IMPACTOR NO, - C
IMPACrnw FLfiWRATF r 0.46? ACFM IMPACTOR TEMPERATURE = 3S5.0 F = 179.4 C SAMPLING DURATION s 120,00 MIN
IMPACTOR PWFSSUKK nROP = 0.8 IN. nF HG STACK TEMPERA TURF = 35S.O F = 179.4 C
ASSUMFD PARTICLf DFNSITV = 1.84 CM/CD.CM. STACK PRESSURE = 27.93 IN. OF HG MAX. PARTICLE DIAMETER = U.O MICROMETERS
GAS COMPOSITION (PERCENT) C02 r 6.b5 CO = 0.00 N2 « 77,36 02 = 8.30 H20 = 7.80
CAtC. MASS LOADING = 8.3009E-03 GR/ATE 1.46W-02 GR/DNCF 1.9016Ef01 MG/ACM -5,3545E*01 MG/DNCM
IMPACTOP STAGF SI S2 S3 54 S5 S6 S7 FILTER
STAGE INDEX NUMBER ! 2 5 a 5 67 fl
050 (MICROMfTERSJ 10.06 10,33 4,38 1.88 0.97 0.67 0.23
MASS (MILLIGRAMS) 2.97 0.92 4,74 1,77 1.63 2.20 5,00 10.61
MG/DSC"/STAGE 3.38E+00 1,05F*00 S.40E+00 2.02E+00 1.86E+00 2.51E+00 5.70E+00 1.21E+01
^1 CUM, PERCENT OF MASS SMALLER THAN DBO 90.05 86.96 71.oe .65,15 S9.68 52.31 35.56
CUM, (MG/ACM) SMALLER THAN D50 1.71E+01 1.6SE+01 1.35E+01 1.24E+01 1.13K+01 9.95E+00 6,76f+00
CUM, (MG/t'NCM) SMALLER THAN DSD 3.02F + 01 2.92E + 01 ?,38F. + 01 ?,19E + 01 2.00E + 01 1.75E + 01 1.19E + 01
CUM. (GR/ACF) SMALLER THAN D<50 7.48E-03 7.23E-03 5.91E-03 5.41F-03 4.96E-03 4.35E-03 2.9SE-03
CUM, (GR/DNCF) SMALLER THAN D50 1.32E-02 1.27E-02 1.0«E-02 9.55F-03 8.75E-03 7.67E-03 5.21E-03
GEO, MEAN DIA, (MICROMETERS) t.19Et01 l.fHF+01 6.72E+00 2.87F+00 1.35E+00 8.06E-01 3.95F-01 1.65E-01
DM/DLOGD (MG/DNCM) 2.36E+01 -9.25E+01 1.45E+01 5.48F+00 6.47E+00 1.56E+01 1.24E+01 4.02E+01
DN/nLOGD (NO. PAPTtCUES/ONCM) 1.46F+07 -9.06F+07 4.95E+07 2.42F+08 2.74E+09 3,10EtlO 2.09E+11 9.28E+12
NORMAL tFNGJK'FERING STANDARD) CONDITIONS ARE 21 t)EG C AND 760MM HG.
SQUARE ROOTS OF PST «V STAGE 0.1M 0.330 0.371 0.320 0.295 0,363 0.312
HOLE DIAMETERS BY STAGF (CENTIMETERS) 1.8237 O.Sa7a 0.2459 0,0807 0,0532 0.0376 0.0260
-------
I-1
Ul
00
KCCO-H 1-1U-77 POHT-a ll<15
TMpAClnR FLototfAIE = 0,752 Af.fH
IMPACTOR PRESSUW.r DROP = 2.2 IM, OF HG
ASSUMFD PARTICLE DFNSHY = I'.sa GM/CU.O
GAS CHMpOSlTIflM (PFHCEMT) C02
CALC, MASS LOADING s 6.5370f-03 CR/ACF
IMPACTDH STAGE
STAGE INDEX NUMBER
MASS (MILLIGRAMS)
MG/OSCM/STAGE
CUM, PERCENT OF MASS SMALLER THAN 050
C'JM. (MG/ACM3 SMALLf.R THAU 050
COM, (MG/DNCM) S^AIL€R THAN 050
CUM. (RR/ACF) SMALLER THAN 050
CU». (RR/ONCF) S^AllFR THAN 050
GEO. HEAN DIA. (MICPHMETERS3
ES/ONCM)
OUTLET SAMPLE U. OF W, MARK III SOURCE TEST IMPACTOH NO, - A
IMPACTOR TEMPFRATMRF. s 355.0 f - i7P,a c SAMPLING DURATION s IPO.OO MI
STACK TEMPEHATlJWf = 355.0 F = 179.14 C
STACK PRESSURE = 27,93 IN, OF HG
6.55 CO = 0,00 K
!.1179f-02 GR/ONCF
S2 S3
2 3
7.8« 3.S9
1,57 0.61
SI
1
7.H6
sa
a
1,21
2,27
ON/OLOGD (NO.
MAX, PARTICLF DIAMETER =
= 77,36 02 = H.30
i.asoiF+oi MG/ACM
S5 S6 S7
5 A 7
0.75 O.UO 0,16
o.9a 2,30 a,20
9.45E+00 l.IOE+00 3.22E+00 1.59E+00 6.57E-01 1,61E*00 2,9«E*00 5,36F*00
b3.5« 59.30 a6.87 aO,75 38,21 32,01 20,69
9.21E+00 R.60E+00 6.80F+00 5.91F+00 5.5aE*OC a,baF+00 3.00E+00
1.63E+01 1.52E+01 1.20F+01 l.oaF+01 9,76FtOO 8,19EtOO 5.29F+00
4.03E-03 3.76E-03 2.97F-03 2.58E-03 2.U2E-03 2.03F-03 1.31E-03
7.10E-03 6.63E-03 5,2aF-03 a,56£-03 a,27E-03 3.58E-03 2.31F.-03
1.05E+01 7.85E+00 5.23E+00 2.05F+00 9.52E-01 5,»8F-01 2.50F-01 1.10E-01
3.77E*01 1.61E+03 9,18E*00 3,«aE+00 3.18E+00 5.89E+00 7,19EtOO 1,7«E+01
3,83E*09 3,72Ft)0 a,77Etll 1.37E+13
la.o MICROHM
H20 = 7.80
2.5581E+01
FILTER
a
7.67
NORMAL (FNGTMF.EPING STA'-'DA^O) CONOtTIONS ARE 21 OEG C AND 760MM HG.
SQUARF ROOTS OF PST R* SfAGF 0.1«« 0.530 0,371 0,?71 0.308 0,373 0.3«9
HOLE DlAMETpRS BY STAGE (f.F M TMf; TE»S) 1.8237 D.57b8 0.2501 O.OS08 O.OS2a 0,0333 0.02tS
-------
U1
KCco-9 1-15-77 pp.RT-a loan
rl.'nvijATE = ft.fifc'i AC.FM
PRFSSl-'Wf" DROP = 4.0 TM. OF HG
PARTICLE DENSITY = j'.efl
PAS COMPOSITION (PERCENT)
CA1 C. MASS LIUPING
jfriPACTOR STAGE
STAGE INDEX NUMBER
050
MASS
MG/OSCM/STAGE
CUM.-PERCENT OF ^ASS SMALLER THAN D50
CUM, {Mfi/AC*!'} SMALLER THAN 050
Ct'M, f M£/jy^'C M ) SMALLER THAN D50
CUM. (GR/ACF) St»Al. JER THAN 050
CUM. fGR/o"CFj SMALLER THA»J 050
GEO. «EAM OIA.
OUTLET SAMPLF U, OF W. MARK TIJ SOURCE TFST IMPACTOH NO, - C
IMPACTOR TEMPERA TURF = 315*0 F = 173,9 c
STACK TEMPERATURE = 3'(5.0 F * 173,9 C
STACK PRESSURE = 28,13 IN. OF HG MAX. PARTICLE DIAMETER =
C02 = 9.fl3 fO = 0.00 ' M2 = 7fe.55 02 = 4.7;
DURATION
82.00 MIN
14.0 MICROMETERS
M20 = 9.30
I.,fl.9« 99,21 89,96 83,20
2.3976E+01 MC/ACM
S5
5
0,67
5, '10
S6
6
o,a«
10.35
S/
7
0.13
1«.9«
FILTER
8
9,29
4.80E+00 9,2QE»00 1.S3E+01 8«25F.»00
71, 9B 50,86 19,31
?,2IQE»01 2,38E*01
a,2HE*01 a,18t+OJ
l.n'^F-02 1..MF-02 9.HJE-03 «.72F-Oi
1.8«E-»)2 1.W3E-02 1.A6E-02 1 .53F-02
1,9-9F*01 1.73F*01 U-21F + OI
S.29t-(S3 2.TJ2E-03
9.35F-OJ ^.SfeF-OS
q,32E-0? -
7,7flF*00 1.62E-HU
DN/DLOGO
, PARTICLES/15NCM)
3.A7EH3
MORMAl {F*lGJWEFPTNfi SIAWOA^B) C
-------
CTl
O
KCCn-10 1-15-77 PPRT-5 1000
I M P A T T 0 R F t 0 H p. A T F: s 0 . 5 (15 A C F M
IMPACTOR PRESSURE DROP = i.u IN, OF HG
ASSUMED PARTICLE DENSITY = 1.841 GM/CU.CM
GAS COMPOSITION (PERCENT) CO? :
CALC. MASS (.('AOING = 1.3739E-02 GR/ACF
IMPACTOR STAGE
STAGE INDEX NUMBER
MASS (MILLIGRAMS)
MG/DSCM/STAfiE
CUM. PERCENT OF MASS SMALLER THAN 050
CUM. (MG/ACH) SMALLER THAN 050
CUM. (MG/D^CM) SMALLFR THAN D50
CUM, (GR/ACF) SMALLER THAN D50
CUM, (GR/DNCF) SMALLER THAN 050
GEO, MEAN DIA, (MICROMETERS)
DN/DLOGD (NO. PARTICLES/DMCM)
OUTLET SAMPLE U, OF W. MARK III SOURCE TEST IMPACTOH NO, •> D
IMPACTOR TFMPf PATURE = 345.0 F - 173.9 C SAMPLING DURATION : 75.00 MINI
STACK TEMPERATURE = 34S.O F = 173.9 C
STACK PRESSURE = 28,13 IN, OF HG MAX, PARTICLE DIAMETFR =
N2 s 76.SS 02 = 4,7?
3.1439E+01 MG/ACM
S5 S6 S7
567
0.82 0,49 0.17
2.98 ia,42 3,56
5.88E-01 6.66E+OQ a.lOF+00 8.61E-02 «.2RE+00 2.07F+01 5.11E+00 1.45E+01
9R.95 H7.07 79.75 79,60 71.97 35.05 25.93
3.11E+01 ?.7«FtOl 2.51F+01 2.50F+01 2.26F+01 1.10E+01 S.lSEtOO
5.«7F+01 4.P1E+01 4.41F+0] 4.40E+01 3.98E+01 1.94E+01 1.43E+01
1.36F-02 1.20E-02 1.10F.-02 1.09F-02 9.89E-03 4.82E-03 3.56E-05
2.39E-02 2.10E-02 1.93E-02 1.92F-02 1.74E-0? S.«7E-03 6.27E-0.5
1.11E+01 S.79F+00 5.B9F+00 2.51F+00 1.14EtOO 6.35E-01 2.93F-01 1.23E-01
1.88E + 03 1.19E + 0] 2.17F-01 1.49E-t-01 9,35E*01 1.13E + 01 4.83F + 01
2.86E+09 6.05F+07 1.42F+07 l.OUE+10 3.79E+11 4.69E+11 2.69E+13
9.43
SI
1
8.83
0.41
CO
2.4161E-C
52
2
8,76
4.64
= 0.00
>? GR/ONCF
S3
3
3.96
2.86
S4
4
1.59
0,06
ia.o MICROMETERS
H20 a 9.30
FILTER
8
10.1?
NORMAL (ENGINEERING STANDARD) mNDlTTDNS ARE 21 OEG C AND 7&OMM HG,
SOUARF ROtiTS OF PSI BY STAGE n.i'ia 0.330 0,371
HOLF OIAMFTERS HV STAGE f CENT IMFTf-;Rs) 1.8237 0.5743 0.2512
0.319
0.0793
0.321
0.0495
0.3S9
0.0330
0.35«
0,0229
-------
J-1S-/7 RDRT-a I'JIO
PR F i n h H A r F - n. m Q A c F M
PRF -SSURF OROP = 1.0 IM. OF HR
ASSUMED PARTTCIE DFMSITY = 1.8U GM/CU.O
PAS COMPOSITION (PERCENT) C02
CAlC. MjSS LOAHTl'iG a l.?'J56F-0? GP/ACF
IMPACTPR SURF
STAGE IftOF-X NUMBER
050 (MICROSigRS)
MASS (HIH. if.KAMS)
MG/OSCM/STAGE
CUM, PERCENT OF MASS SMALLER THAN 050
CUM, (MG/ACI-I) SMALLER THAN 050
CUM. f"!f./D^CM) SMALLER THAN 050
CUM, tGR/ACF 3 SM4LI F. R THAN 0*50
CUM,. (Gp/DWCF) SMALLE8 THAN D50
GLO, MEAN DIA. (MJCPOMETERS)
DM/01 Ofin f«G/i>'"C'")
0^/DLHGr {NO,
OUH.FT SAMPLF 1). OF H, MARK III SOURCE TFST IMPACTOR NO, - A
IMPATTflR TF«PEPA7URF s .56S.O F = 185,0 C. SAMPLING DURATION = 120.00 MlN
STACK TEMPLRATURF = 3hS.O F z 185,0 C
STACK PRESSURE = 28.13 IN. OF HG MAX. PARTICLE DIAMETER =
» 9.43 CO s 0.00
?.?(10
1.05F-0? 1.02F-02 8.73E-03 7,(J3F-03 6.53E-nj «,81F-03 2.2RE-03
1.89E-02 1.80E-02 1,57F-02 1.3«E-02 1.18E-02 8.67F-03 «,11E-03
1.15F + 01 9,'l6FtOO 6,32E*00 2.50E + 00 1.17E + OQ 6.92E-01 3.37E-01 1.55E-01
a.ShEtOl !.67tt03 1.77Et01 1.20E + 01 1.86Et01 ?,82fi + 01 2.85E*01 3.17E + Q1
3.3«E*07 2.05E + 09 7.26F.t07 7.97H08 1,?1£ + 10 B.SSF + IO 7.73E + 1J J
RMAL {FNfiJMFf BTr-C S7ANOAR01 CONDITIONS
SOUARf pnoTp rif PST flV STftRF
VOI.f olAHfTrfS HY KI/Gf f CFMTI "f TF«S1
21 OEG C AND TfeOMM HfJ,
n.l<*1 0.330 0,371 O.P7! 0.308 0,373
(.8237 n,t,768 O,?1.?)! 0,0808 0.nS?a 0,0333
fl»3a<>
-------
CTl
KCCO-13 1-16-77 POHT-'4 0915
TMPACTOR FLOW?*TE = 0.515 AtFM
IMPACTOP PHFSSURF DROP = O.a IN. DF HG
4SSUMED PARTICIE DENSITY = 1.84 GM/Cll.CM
GAS COHPOSITinN (PFRC'.MT) C02
C4LC. MASS HHPING s 1.5550E-0? GR/ACF
IMPACTOH STAGF
STAGE INDEX NUMBER
050 (MirwoMerTERS)
MASS (MILLIGRAMS)
MG/DSCM/STAGE
CUM. PERCENT or MASS SMALLER THAN 050
CUM, (MG/ACM) SMALL!R THAN 050
CUM, (MG/DNr.M) SMAIlff THAN D50
CUM. (GR/ACF) SMALLER THAN 050
CUM, (GR/DNCF) SMALLER THAN 050
GEO, MEAN DIA. (MICROMETERS)
OM/DLOGD tMU/Ok)CM)
fNU. PARTtClES
OUTLET SAMPLE u. OF w, MARK in SOURCE TEST IMPACTOR NO, . A
IMPACTOR TEMPERATURE = 345.0 F = 173,9 c SAMPLING DURATION = 120,00 MIN
STACK TEMPERATURE = 345.0 F = 173.9 c
STACK PRESSURE = 28,13 IN, OF HG MAX. PARTICLE DIAMETER
= 8,49 CO = 0,00
2.7467F-02 GR/DNCF
SI S? S3
1 2 3
9.11 9.09 4,02
P2.67 3,79 14,89
N2 = 7,97 02 s 6,59
3.558JE+01 MG/ACM
S4 S5 S6 S7
4567
ltt.0 MICROMETERS
H20 = 9.70
6,?85«f>0i M6/DNCH
FILTER
8
1,36 0.83 0,44 0,17
6,30 1.74 3,83 5,69 3,3?
P.32E+01 3.88E+00 1,52E*01 6.49E+00 t,78E+00 3.92E+00 5.8PE+00 3.40E+00
63.59 57.51 33.60 23.41 20.6? 1«,47 5,33
?.2hF+01 2.05E+01. 1.20E+01 8.33F+00 7.34E+00 5.15E+00 1.90E+00
4.00E + 01 3.61E + 01 2.11E + 01 .1.47E + 01 1.30F + 01 9.09E + 00 3.35E + 00
9.89E-03 8.94F-03 5.22E-03 3.64E.Q3 3.21E-03 2.25E-03 8.29E-04
1.75E-02 1.58E-0? 9.23E-03 6.43F-03 5.66E-03 3.97F-03 1.46F-03
1.13E+01 9,10EtOQ 6.05E+00 2.34E+00 1.06E+00 6.04E-01 2.71E-01 1.18E-01
1.24E+02 5.63E+03 4.30E+01 1.38F+01 H.30E+00 1.41E+01 1.39E+01 1.13E+01
«.96E*07 7.76E + 09 2.02E + 08 1.12E + 09 7.16E + 09 6.67E + 10 7,27E*11 7.HE+12
NORMAL IEMGINEFTNG STAMnARni TOMOITIONS ARE 21'OEG C AND 760MM HG.
SQUARE ROOTS OF PS! BY ST4GE 0.1'l« 0.330 0.371 0,271 0.30H 0.373 0.3«9
HOLE OlAM£TfRS BY STAGE (CENTIMETERS) 1.8237 0.5768 0.2501 0.0808 0.0524 ^1,0333 0.02«5
-------
H
CTl
(jO
KCCO-15 1-16-77 PdRT-u JV45
IMPACT-DR Fl OXR-ATE = 0.5C-1) ACF".
I^PACTOR PPF.SSURF DHCP = 0.3 IN. OF HG
ASSUMED PARTICLE DENSITY = l.Ba GM/CU.CM.
GAS COMPOStTTOM (PERCENT! C02 :
CAtC. "ASS LflAOIMG = a.2665F-03 GR/ACF
OUTLET SAMPLE U. OF W. MARK III SOURCF TEST IMPACTOR NO. ' t
IMPACTOR TtKPERATURF . 350.0 F . 176.7 C SAMPLING DURATION = 120.00 HIN
STACK TFMPfRATURF = 350,0 F = 176.7 C
STACK PRESSURE = 2B.13 IN. OF HG MAX. PARTICLE DIAMETER = 10.0 MICROMETERS
STAGE INDFX NIJM[(f;R
p50 (MICROMETERS)
MASS (MILLIGHAMS)
MG/DSCM/STAGE
CUM. PERCENT OF MASS SMALLER THAN 050
CUM. (MR/ACM)- SMALLER TH4Ni 050
CUM, (MG/DMCM5 SMAL LF." THAN D?io
COM. (GR/ACF) SMALLER THAN D50
CUM, (GR/ONCF) SMALLER THAM D50
SEO. MEAN DIA. (MICROMETERS)
DM/DLOGO (MG/nNCM)
DN/DLOGB (NO. PARTICl F-S/ONCM)
N2 = 7.97 02 = 6,59
9.7633E+00 MG/ACM
SU
a
1.65
0,90
55
S
O.S2
0.50
Sfc
6
0.55
0.95
S7
7
0.16
a. is
H20 s 9.70
1.7353E+01 MG/DNCM
FILTER
8
t.17
B.<49 ro = o .00
7.5R32E-03 GR/OMCF
SI S2 S3
1 2 3
P.20 9.a« 3.96
a.B2 0.73 3.60
5.03E+00 7.62E-01 3.76E+00 9.39E-OJ 5.22F-01 9.92E-01 a.36EtOO 1.22E+00
71.39 67.06 «5.70 00.36 37.39 31.75 6.94
h.97E+00 6.55F+00 a.W + OO 3.94E + 00 3.65F+00 3.10E+00 6.7BE-01
1.3UF.+01 1.16E+OJ 7.93F+00 7.00F+00 6.U9E+00 5.51E*00 1.20E + 00
3.05E-03 ?.R6E-03 1.95E-03 1.72E-03 1.60E-OJ 1.35E-03 2.96E-Ofl
5.ajE-03 S.09E-03 3.«7E-03 3.06F-03 2.8aF-03 2.aiE-03 5.27F-Oa
1.13E+01 9.32F+00 6.11E+00 ?.S6F+00 1.16E+00 6.69E-01 2.98E-01 1.1SE-OJ
P.76E + 01 -6.67EfO! 9.95F + 00 2.4BF + 00 1.71ET+00 5.66E + 00 B.2BE*00 a.06E + 00
1.96E+07 -R.56E+07 a.5?E+07 l.S«F+OB 1.13E+09 1.96EtlO 3.25E+11 2.7BE+12
NORMAL (EWGINEF.RTNB STANDARD) CONDITIONS ARE 2t OEG C AND 760MM HG.
SQUARE ROOTS OF PSI BY STAGE 0.»" ".330 0.37, 0.320 0.295 0.36J
HOLE OIAHETFRS BY SUCF f CENTIMETERS) l.«37 O.S87a «.?«5P 0.0807 0.0532 0.0376
0.0260
-------
X
o
C3 ffl
3L TO
V O
o •
3C H S
:;S TJ -H -i >—
'-• *->0
OOO
c: >
a; -i
m c
!*
--4 -J
oooo
CDOOU4
QOO
o
c
ru
-o
iJl —I t-»
» » •
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-119
I. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Performance Evaluation of an Electrostatic
Precipitator Installed on a Copper Smelter
Reverberatory Furnace
5. REPORT DATE
June 1979 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Anonymous
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southern Research Institute
2000 Ninth Avenue
Birmingham, Alabama 35205
10. PROGRAM ELEMENT NO.
1AB604
11. CONTRACT/GRANT NO.
R-804955
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmetal Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report describes tests performed on the electrostatic
precipitator installed on the copper reverberatory furnace at the
Kennecott Copper Corporation smelter at Hayden, Arizona. These tests
provided data on the chemical characterization of particulates, noncon-
densables, and gases in addition to operating and performance measure-
ments of electrical parameters, particle size, voltage-current
distribution, and resistivity. Efforts were also made to develop
computer simulations of ESP performance and to evaluate overall perfor-
mance of the control device. The operation of the ESP was erratic, and
other unavoidable restraints on the sampling program prevented
acquisition of reliable data to evaluate representative performance of
the ESP. Nevertheless, analysis of the data shows the types of
information which can be obtained in evaluation of control devices if
the test locations can be utilized to obtain data representative of
normal "on stream" operations.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Exhaust emissions
Smelting
Trace elements
Pollution
Electrostatic precipitators
Particle size distribution
Air pollutant emis-
sions
Air pollutant control
Copper smelter
63 A
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
175
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
165
U. S. GOVERNMENT PRINTING OFFICE: 1979 — 657-060/5314
------- |