U.S. Environmental Protection Agency Industrial Environmental Research EPA-600/7-76-027
Office of Research and Development Laboratory
Research Triangle Park, North Carolina 27711 OCtODef 1976
EFFECT OF A
FLYASH CONDITIONING AGENT
ON POWER PLANT EMISSIONS
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven 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
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure t e rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Government, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
EPA-600/7-76-027
October 1976
EFFECT OF A
FLYASH CONDITIONING AGENT
ON POWER PLANT EMISSIONS
by
Leslie E. Sparks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Program Element No. EHE624
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
CONTENTS
Figures v
Tables vi
Acknowledgment vii
1 Introduction . 1
Conclusions 2
2 Recommendations 4
3 Emissions Tests 5
Description of the Plant 5
Test Schedule 5
Test Methods 7
Ammonia 7
Sulfur Oxides 7
Organics 7
Particulate Size Measurement 8
Flyash Resistivity 8
Electrical Measurements 8
4 Results and Discussion 10
Gaseous Emission Tests 10
Ammonia Data 10
Sulfur Trioxide Data 10
Sulfur Dioxide Data 13
Organic Data 13
Particle Size Data 14
Dust Resistivity Data 14
Electrical Data 18
Particle Collection Efficiency 29
5 Recommendations for Future Tests 38
-------
CONTENTS (cont'd)
6 References 40
Appendix A Effect of Flyash Conditioning Agent on Coal-Fired Power
Plant Emissions 41
Appendix B Precipitator Voltage—Current Data 108
Appendix C Voltage Vs Current Data 116
Appendix D S02 Removal in an Electrostatic Precipitator 122
Appendix E Calculation of Efficiency as a Function of Particle Diameter 124
IV
-------
FIGURES
Number Page
1 Differential particle size distributions at outlet of ESP,
Montour, December 10, 11, 16, 17, and 18, 1975 15
2 Mean differential particle size distribution at inlet of ESP,
Montour Plant, December 10, 11, 16, 17, and 18, 1975 16
3 Flyash resistivity measurements, Montour Power Plant 17
4 Secondary V-J Characteristics of Power Set 1A-2A 19
5 Secondary V-J Characteristics of Power Set 1B-2B 20
6 Secondary V-J Characteristics of Power Set 1C-2C 21
7 Secondary V-J Characteristics of Power Set 1D-2D 22
8 Primary V-I Characteristics, Power Set 1A-2A 23
9 Primary V-I Characteristics, Power Set 1B-2B 24
10 Primary V-I Characteristics, Power Set 1C-2C 25
11 Primary V-I Characteristics, Power Set 1D-2D 26
12 Efficiency versus particle diameter for December 10, 1975 (low
sulfur coal conditioned with LPA 402A) 30
13 Efficiency versus particle diameter for December 11 (low sulfur
coal conditioned with LPA 402A) 31
14 Efficiency versus particle diameter for December 16 (high sul-
fur coal) 32
15 Efficiency versus particle diameter for December 17 (high sul-
fur coal) 33
16 Efficiency versus particle diameter for December 18 (low sulfur
coal with water injection) 34
17 Efficiency versus particle diameter (all tests with ± 90% con-
fidence) 35
18 Efficiency predicted by ESP computer model as a function of
particle diameter for Montour 37
-------
TABLES
Number Page.
1 Test Schedule 6
2 Test Plan 6
3 Summary of Gaseous Emission Test Data at Montour Using All
Data 11
4 Revised Summary of Gaseous Emission Test Data at Montour with
Data Points Outside 90 Percent Confidence Limits Deleted . . 12
5 Current Densities at Break in Secondary V-J Characteristics . 28
6 Average Operating Power Densities 28
VI
-------
ACKNOWLEDGMENT
The author wishes to acknowledge the work of data collection and
reduction done by the Process Measurements Branch, IERL-RTP. Assistance
provided by Southern Research Institute personnel (particularly by
Dr. Herbert Spencer, who provided the discussion on electrical readings)
was extremely helpful. The cooperation of Pennsylvania Power and Light
Company, which made the test work possible, is gratefully acknowledged.
Editorial assistance was provided by the Research Triangle Institute.
vn
-------
SECTION 1
INTRODUCTION
The combustion of moderate to low sulfur coal to meet sulfur oxide
emission standards increases the difficulty of particle collection in electro-
static precipitators (the most common particulate collection device for coal-
fired boilers). The difficulty is caused by an increase in dust electrical
resistivity at normal ESP temperatures when moderate to low sulfur coal is
burned. The increase in dust resistivity is generally believed to be due
to low concentrations of sulfur trioxide (SO^) in flue gas from combustion
of moderate to low sulfur coals. Because of the increase in resistivity
when low sulfur coal is burned, switching a boiler/ESP system designed for
high sulfur coal to low sulfur coal will result in increased particulate
emissions unless action is taken to either increase the specific collector
area of the electrostatic precipitator or reduce the resistivity of the
flyash.
Resistivity modification by injecting a chemical conditioning agent
into the flue gas is one method of reducing the impact of coal switching.
Data on the effectiveness and environmental impact of conditioning agents are
rare.
This study was undertaken as a preliminary program to add to the limited
data base on emissions of conditioning agents and on the effectiveness of
conditioning agents for restoring electrostatic precipitator performance
after a switch from high sulfur to low sulfur coal. The tests were conducted
at Pennsylvania Power and Light Company's Montour Station where a proprietary
conditioning agents LPA 402A (Apollo Chemical Company), was being used
to improve electrostatic precipitator performance.
The objectives of the program were:
1. To obtain data on overall emissions (gaseous and particulate)
when LPA 402A was injected.
-------
2. To obtain data on the effectiveness of the electrostatic
precipitator in collecting the conditioning agent.
3. To obtain data on electrostatic precipitator participate col-
lection when high sulfur coal and low sulfur coal with condi-
tioning and low sulfur coal without conditioning were burned.
Primary emphasis of the program was placed on objectives 1 and 2. Much
of the data for objective 3 are questionable because insufficient time was
available for the electrostatic precipitator to reach steady-state operation
when low sulfur coal without conditioning was used.
CONCLUSIONS
The data reported indicate that:
1. The electrostatic precipitator at Montour operated efficiently
when high sulfur coal was burned and could meet the required
particulate emission standards.
2. The combustion of low sulfur coal produced a high resistivity
flyash with the expected deterioration of ESP electrical
conditions.
3. The injection of Apollo Chemical's conditioning agent LPA 402A
reduced the ash resistivity from 10'' ohm-cm to 4 x 10^° ohm-cm.
4. The injection of LPA 402A may have resulted in increased emissions
of sulfur trioxide and ammonia compared with unconditioned low
sulfur coal data.
5. It is unlikely that the ESP can meet particulate emission standards
when low sulfur coal is burned even when the flue gas is conditioned
with LPA 402A under conditions similar to those during the tests.
6. The techniques for measuring flue gas concentrations of sulfur tri-
oxide lack the precision necessary to determine, with a high statistical
confidence, differences in sulfur trioxide emissions with and without
conditioning.
7. The in-situ flyash resistivity data provide the best evidence of
the effects of conditioning with LPA 402A.
8. The techniques for measuring particle size distribution do not have
the precision necessary to determine, with a high statistical con-
fidence, differences in particle collection efficiency as a function
of particle diameter due to changes in operating conditions.
9. Electrostatic precipitators generally respond slowly to changes in
flyash resistivity.
10. There is no significant difference between ESP inlet and outlet
SOo concentrations.
-------
11. The data provide useful background information for planning future
tests.
12. Additional testing at Montour is warranted.
The data taken during these tests do not show an improvement in ESP per-
formance when LPA 402A was used to condition the high resistivity flyash. This
is probably due to the fact that insufficient time was available to allow the
ESP to respond to the change in ash resistivity when the boiler was switched
from high sulfur to moderate sulfur coal. Data reported by Cragle (1976)>
showing an improvement in ESP performance when LPA 402A was used as a condi-
tioning agent * are consistent with the resistivity data and ESP computer
modeling calculations in this report.
-------
SECTION 2
RECOMMENDATIONS
Additional data are necessary on the impact of flue gas conditioning
on overall pollutant emissions from coal-fired power plants. Such data
are required in the near future to achieve a rational approach to simultaneous
compliance with sulfur oxide and particulate emission standards.
Companies desiring to use conditioning to improve ESP performance should
conduct careful studies to determine the impact of the conditioning agent on
all air pollutant emissions. Studies should also be conducted to ensure
that use of a conditioning agent does not create water pollution problems
due, for example, to leaching of the agent from flyash disposal piles.
Improvements are necessary in measurement techniques for sulfur trioxide,
ammonia, and particle size distributions. Available techniques provide use-
ful answers; but their precision under field use is inadequate to detect
differences in emissions under various conditions with a high degree of statis-
tical confidence. The need for improved instruments cannot be over<-emphasized.
Future tests should be carefully designed to determine the effectiveness
and environmental impact of conditioning agents to improve ESP performance.
Simultaneous inlet and outlet measurements should be made whenever possible.
If simultaneous or near-simultaneous measurements are impossible, randomi-
zation of sampling location for each sample should be used to avoid bias
caused, for example, by changing boiler load. Secondary voltage-current data
should be obtained for each electrical section for each testing day. Several
days operation between tests should be allowed to enable the ESP to reach
steady-state when flyash resistivity is changed. Secondary voltage-current
data can be used to determine when steady-state is reached.
Research is needed to determine the mechanisms that cause conditioning
to change flyash resistivity.
-------
SECTION 3
EMISSIONS TESTS
DESCRIPTION OF THE PLANT
The tests were conducted at Pennsylvania Power and Light Company's
Montour Station. All tests were performed on unit number 2, a 750 megawatt
coal-fired boiler.
Particulate emissions are controlled by an electrostatic precipitator
2 2
in a Chevron configuration. The total plate area is 42809.7 m (460,800 ft ).
2
There are 16 transformer-rectifier (TR) sets and each TR set supplies 2675.6 m
2
(28,800 ft ) of plate area. The design efficiency of the precipitator is
99.5 percent when 1.5 percent sulfur coal is burned (Cragle, 1976).
The conditioning agent is injected before the air heater where the flue
gas temperature is ~ 510°C (950°F) at full load (Cragle, 1976). The normal
injection rate is 95 liters per hour (25 gallons per hour).
TEST SCHEDULE
The Montour test program was conducted as described in Table 1. At the
onset of the program, December 10 and 11, low sulfur coal was burned and the
LPA 402A was injected at 95 liters per hour (25 GPH). From December 16
through December 17, high sulfur coal was burned without injection of the
additive. On December 18, 1975, the plant was switched back to low sulfur
coal, but the additive was not injected. Water was injected at 95 liters per
hour (25 GPH) during this low sulfur run.
Table 2 presents the test plan for Montour. The 5 cfm cyclone train
was not used due to damaged connector threads. The remainder of the plan
was followed. Additional details of the test plan are contained in Appendix A.
-------
TABLE 1. TEST SCHEDULE
December 8 and 9
December 10 and 11
December 12
December 13 and 14
December 15
December 16
December 17
December 18
Arrive at plant and set up.
Full load tests using low sulfur coal,
conditioner LPA 402A injected.,
No tests; crew returns to Research
Triangle Park, N.C.
Weekend, no tests.
Crew returns to plant-
Full load tests using high sulfur coal,
no additive injected.
Full load tests using high sulfur coal,
no additive injected.
Full load tests using low sulfur coal,
water injected.
TABLE 2. TEST PLAN
Test
Dust Resistivity SRI Probe
Particle Size Distribution
Location
ESP Inlet
ESP Inlet
No. Per Day
3
4
Time Per Test
2-3,, hrs
1/2-1 hr
using Brink Impactors
Particle Size Distribution
using Andersen Impactors
and 5 cfm Cyclone Train
S02/S03 and NH3
Organics
Opacity
Opacity
ESP Outlet
ESP Inlet
& Outlet
ESP Inlet
& Outlet
ESP Inlet
EPA Instrument
ESP Outlet
PP&L Instrument
2
2
1-2 hrs
4
4
Continuous
Continuous
-------
TEST METHODS
The test methods used during the tests are fully discussed in Appendix
A. An abbreviated discussion of the test methods is presented here.
Ammonia
The ammonia concentration in the flue gas during the Montour test was
determined by a modified Kjeldahl-Titrimetric Procedure which was developed
by EPA personnel. It is based on the absorption of the ammonia in impingers
which contained a sulfuric acid solution. After the sample is complete,
free ammonia is distilled into a flask and a titration procedure followed to
determine the ammonia concentration. EPA personnel have determined that this
method of ammonia concentration determination has acceptable accuracy above
about 10 ppm.
Sulfur Oxides
Sulfur oxides were collected by a sample train based on absorption of
sulfur trioxide in an isopropanol (IPA) solution and absorption of sulfur
dioxide in a tiJ^2 solution. Collection of the SO was followed by titration
using the IPA-Thorin technique. This method of sulfur oxides determination
has been widely used and found to be reliable.
Comparison of the Montour sulfur oxides concentration data with similar
data taken at other plants indicates that the Montour data are reasonable.
The variability in the S02 measurements at Montour is consistent with that
found at other locations while the coefficients of variation of the SO.,
measurements are somewhat larger than the norm.
Organics
The composition of organic vapors which might be present in the flue gas
was of interest because of possible undesirable decomposition products from
the injected additive. The organic vapors were absorbed onto a polymer medium,
Tenax GC, and later desorbed into a smaller volume for analysis. With the
Montour tests, about 1415 liters of gas was passed through the absorbent
and the organics which were collected were desorbed into about 120 ml of pen-
tane. The extract was analyzed in two different ways. Fourier Transform
-------
infrared spectroscopy was used to provide a qualitative analysis. A linked
GC/mass spectroscopy analysis was used to detect diethylnitrosamine, which
was thought to be a likely decomposition product.
This technique for analysis of nitrosamine in the flue gas is extremely
sensitive. Its detection limit under the conditions used during this test
is less than 5 parts per trillion.
Particulate Size Measurement
Cascade impactors (inertia! sizing devices) were used to determine
particulate size distributions at Montour. The Brink Impactor, a low sample
rate device, was used on the inlet to the electrostatic precipitators (ESP);
the Andersen impactor, a high sample rate impactor, was used on the outlet.
Isokinetic sampling procedures were followed with both impactors. The
Andersen impactor used the factory precut glass fiber substrates. Foil
substrates, coated with polyglycol grease, were used as substrates in the
Brink impactor.
The fiberglass substrates used with the Andersen impactor were pre-
conditioned by in-situ exposure to the flue gas using procedures developed by
the Southern Research Institute. Blanks were also run each day with the
Andersen impactor. The average weight gain on a blank was used as a correction
on that day's Andersen stage weights. Blanks were not run with the Brink
impactor.
Flyash Resistivity
The resistivity of the flyash was measured using an in-situ point-to-
plane resistivity probe and the reported results were obtained using the
parallel disc method. The point-to-plane probe simulates dust collection as
it occurs in an ESP, and is thought to be more relevant to precipitator per-
formance than is data derived from mechanically collected dust. The resis-
tivity is determined from the electrical characteristics of the dust just
prior to sparkover. Flyash resistivity measurements as made at Montour have
a precision of around ± 30 percent.
Electrical Measurements
During the test, primary voltages, primary currents, and spark
rates for the 16 transformer-rectifier sets on the number 1 unit
8
-------
were recorded once per shift. The results of these measurements are con-
tained in Appendix B. Secondary current meters were not available during
the test. Secondary voltage readings were obtained during testing for opera-
tion with high sulfur coal, with a lower sulfur coal with Apollo Chemical
Company additive LPA 402A and with a low sulfur coal with the addition of
water.
The secondary voltage readings were made with a voltage divider probe
consisting of a 120 MQ resistor and a 12ko resistor connected in series. The
probe was calibrated against a commercial high voltage meter with a 1 per-
3
cent accuracy prior to the test. (A calibration factor of 9.40 x 10 was
determined for voltage read across the 12kn resistor.) Secondary voltages
read during the test are estimated to be accurate to ±0.5 kV. Secondary vol-
tage-current characteristics were generated for each test condition for the
1-2 chamber of the precipitator. The results of these measurements are con-
tained in Appendix C. These V-I characteristics were generated by turning off
the power set, installing the voltage divider, than manually turning up the
power. The corresponding primary voltages, currents, and secondary voltages
were then recorded as the primary voltage input was manually decreased. Esti-
mates of the secondary currents were than calculated by assuming that the
secondary power output was 60 percent of the primary power input. Typically,
transformer-rectifier sets have conversion factors from 40 percent to 80
percent. By assuming a value of 60 percent, the estimated possible error in
the calculated secondary currents is on the order of 33 percent. The average
current densities to the precipitator plates for each power set were calcula-
ted using the above assumption of 60 percent power conversion and the following
procedure.
The primary voltage and current were multiplied to obtain primary power.
The primary power was then multiplied by 0.6 and divided by the secondary
voltage to obtain secondary current, which was in turn divided by 2.68 x 10 cm.
the plate area supplied by each power set.
-------
SECTION 4
RESULTS AND DISCUSSION
GASEOUS EMISSION TESTS
The results of the gaseous emission tests using all the data are
summarized in Table 3. The data taken at Montour show considerable scatter
as is shown by the large standard deviations in Table 3. Harris, in Appendix
A, has examined the data taken at Montour and eliminated some of the data
points. Table 4 is a summary of the data with the data points outside the
90 percent confidence limits deleted. See Appendix A for the details of
how the points were deleted.
Ammonia Data
The data clearly show that injection of LPA 402A caused emission of
ammonia. The exact concentration of the ammonia emissions is uncertain due
to the scatter in the data. The measured ammonia concentrations are below
the useful detection limits of the ammonia test method; i.e., the accuracy
and precision of the method at these concentrations is low.
Sulfur Trioxide Data
There is no statistically significant difference between the inlet and
outlet sulfur trioxide data for the three test conditions. This is in
agreement with SO^ data taken by others at other power plants; e.g., Dis-
mukes (1975). Therefore, the inlet and outlet SO, data for each test condi-
tion were combined and averaged in order to make comparisons between test
conditions.
The flue gas S03 concentration with conditioned low sulfur coal is the
same as the flue gas S03 concentration with high sulfur coal.
10
-------
TABLE 3. SUMMARY OF GASEOUS EMISSION TEST DATA AT MONTOUR USING ALL DATA
<;
*•
Coal and
Condition
High Sulfur
No Conditioning
Low Sulfur with
Injection of
LPA 402A
Low Sulfur with
Injection of Water
iOp ppm
ESP
Inlet
1340.2
(84.6)
929.7
(97.1)
680.1
(182.1)
SOo ppm
ESP
Outlet
942
(136.6)
842
(75.6)
644
(60.7)
Combined
SO 2 ppm
1140.9
(237.6)
881.9
(93)
663.4
(121)
SOo ppm
ESP
Inlet
28.8
(20.9)
27.2
(8.4)
29.5
(20.8)
SOg ppm
ESP
Outlet
25.6
(10.3)
30
(19.8)
16
(6.7)
Combined
SO, ppm
27.2
(15.3)
28.8
(15)
22.8
(17.3)
NH3 ppm NHg ppm
ESP ESP
Inlet Outlet
Oa
3.0 8.1
(2.3) (9.1)
0 0
Combined
NHg ppm
Oa
5.1
(6.8)
0
Standard deviation of measurements shown in parentheses.
^ne measurement of 25.4 ppm, 6 measurements of 0.00 ppm of
-------
ro
TABLE 4. REVISED SUMMARY OF GASEOUS EMISSION TEST DATA AT MONTOUR WITH DATA POINTS OUTSIDE 90
PERCENT CONFIDENCE LIMITS DELETED
Coal and
Condition
High Sulfur
No Conditioning
Low Sulfur with
Injection of
LPA 402A
Low Sulfur with
Injection of
Water
SOp ppm
ESP
Inlet
1340
(84.6)
890
(46.2)
683
(178.4)
S02 ppm
ESP
Outlet
942
(136.6)
870.4
(34,2)
644
(60.7)
Combined
SOp ppm
1141
(237)
-879
(38.6)
663
(121)
SO g ppm
ESP
Inlet
19-. 5
(11-5)
27.2
(8.35)
16.2
(6.7)
S03 ppm
ESP
Outlet
25.6
(10.3)
18.4
(10,0)
16
(5)
Combined
SOo ppm
2.3
(10.4)
23.3
(9.7)
16
(5.3)
NHo ppm NHg ppm
ESP ESP Combined
Inlet Outlet NH3 ppm
.000
3 8.1 5.5
(2.3) (9.1) (6.8)
000
Standard deviation shown in parentheses.
-------
The flue gas SOg concentration with conditioned low sulfur coal is
significantly higher than the flue gas SOo concentration with unconditioned
low sulfur coal. Based on chemical analysis of LPA 402A, the measured in-
crease in flue gas sulfur trioxide concentrations is reasonable. However, it
is not clear that all the increase in sulfur trioxide emissions is due to
injection of LPA 402A. The sulfur dioxide data indicate that the coal sul-
fur content may have changed. Thus, some of the increase in sulfur trioxide
emissions may be due to a change in coal sulfur content.
Sulfur Dioxide Data
The sulfur dioxide data for the high sulfur coal tests show a significant
difference between inlet and outlet sulfur dioxide concentrations. The inlet
and outlet data were taken on separate days so a reasonable explanation for
the change in sulfur dioxide concentration is a change in coal sulfur con-
tent.
There is no reason to expect significant removal of sulfur dioxide in
an electrostatic precipitator. It is true that ionized sulfur dioxide mole-
cules could be removed by the electrostatic precipitator. However, for the
current density and sulfur dioxide concentrations at Montour, the possible
number of sulfur dioxide molecules that could be removed by ionization and
collection is insignificant. Calculations supporting this statement are
included in Appendix D.
The sulfur dioxide concentrations during the tests where LPA 402A was
injected are significantly higher than the sulfur dioxide concentrations
for low sulfur coal without injection. Injection of LPA 402A could not
increase the sulfur dioxide concentration to the level indicated. Thus, the
increase in sulfur dioxide concentration is probably due to an increase in
coal sulfur content.
Organic Data
Qualitative analysis by infrared spectroscopy identified three compounds
which might be decomposition products of the LPA 402A. These were a carboxylic
acid, a salt of the carboxylic acid, and a possible alkanol amine. The
quantitative analysis for diethylnitrosamine showed that the nitrosamine
concentration was less than the detection limits of the analysis method (less
than 5 parts per trillion).
13
-------
PARTICLE SIZE DATA
Particle size distributions at both the inlet and the outlet of the ESP
were determined for all three types of firing. The inlet data were fairly
consistent throughout the test. Coal type and boiler variables had little
apparent effect on the inlet size distribution. There was a great deal more
variation in the outlet particulate data, both within a given data set and
from day to day.
Average inlet and outlet particle size distributions by mass are presented
in Figures 1 and 2. Note that the outlet particle size distribution by mass
for unconditioned low sulfur coal shows lower mass concentrations for all
particle diameters than shown by the size distribution curves for the other
two test conditions. As explained in the section on electrostatic precipi-
tator efficiency, this is probably due to failure of the electrostatic
precipitator to reach steady-state operation after the switch from high to low
sulfur coal. Cragle (1976) reports that several days operation is required
for the electrostatic precipitator at Montour to reach steady-state following
a change in coal type.
Inlet and outlet size distributions for all test conditions are presented
in Appendix A. An examination of the data in Appendix A shows that the data
for particles with diameters less than about 2 microns are generally more un-
certain than the data for larger particle diameters.
DUST RESISTIVITY DATA
The results of the ash resistivity tests are presented in Figure 3, The
injection of LPA 402A reduced the resistivity of the low sulfur coal flyash
by about 60 percent.
Extrapolation of the ash resistivity versus temperature data for the
LPA 402A conditioned flyash indicates that electrostatic precipitator operation
at flue gas temperatures in the range of 137 to 140°C (roughly equivalent to
the temperatures when high sulfur coal was burned) would produce ash resistivity
of approximately 10 ohm-cm. Such a change in operating temperature should
improve the particle collection efficiency of the electrostatic precipitator.
14
-------
103P
I02
10
z
*x
a*
E
c*
5 I01
<
x.
O
O.I
Figure 1
l r
TT
Low sulfur, _
LPA 402A -
injected
sx \- High sulfur -
^« unrnnriitinnpH
\
\
unconditioned
Low sulfur
water
injected
\
LEGEND
o DECEMBER 10
• DECEMBER II
D DECEMBER 16
" DECEMBER 17
A DECEMBER 18
ASSUMED PARTICLE DENSITY = 2.27 g/fcc
i i i i i I I i i i i i i i i i i
\
\w High sulfur '
unconditioned
1.0 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Differential particle size distributions at outlet of ESP, Montour,
December 10, 11, 16, 17, and 18, 1975.
15
-------
MEAN, WITH 90%
CONFIDENCE LIMITS
ASSUMED PARTICLE DENSITY = 2.27g/cc
I0f
10
e
z
x.
o>
E
<
*v
O
10'
0.1
I I I I I I II
l.O 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Figure 2 . Mean differential particle size distribution at inlet to ESP,
Montour Plant, December 10, 11, 16, 17, and 18, 1975.
16
-------
10"
=E
>o
• LPA - 402A= 95 |/hr
LOW SULFUR COAL, 750 mw AVG.
V H20' 95 l/hr
LOW SULFUR COAL,750mw
• NO INJECTION
HIGH SULFUR COAL, 755 mw
tr
•JO
280
285
290 295 300
TEMPERATURE,°F
305
310
Figure 3. Flyash resistivity measurements, Montour Power Plant.
-------
ELECTRICAL DATA
Voltage-current (V-I) characteristics supply information on the electrical
operation of a precipitator and can be used to determine the effect of an
additive on precipitator performance. V-I characteristics are sensitive
to changes in resistivity, particle size distribution, and grain loadings of
the ash entering the unit.
Figures 4 through 7 are plots of the secondary voltage (V)-current density
(J) characteristics for the 1A-2A, 1B-2B, 1C-2C, and 1D-2D power sets for each
test condition. Figures 8 through 11 are plots of the primary voltage-
current characteristics for each of the 1-2 power sets and for each test condi-
tion.
The secondary V-J characteristics obtained when burning high sulfur coal
with a low resistivity ash appear to have no significant irregularities or
unusual behavior, except for power set 1C-2C where secondary voltage jumped
17 kV to 58 kV when input power was increased. The measurements for the 1C-
2C unit are probably in error. With the ash from the high sulfur coal, maxi-
mum voltages on the order of 39 kV were obtained. The maximum current densi-
p 2
ties ranged from 14 nA/cm at the inlet to 34 nA/cirr at the outlet.
The secondary V-J characteristics obtained with low sulfur coal show pos-
sible signs of back corona. Higher current densities were obtained in the
inlet for a given voltage with the low sulfur coals than with high sulfur coal.
Since the low sulfur coal ash had a higher resistivity than the high sulfur
coal ash, the V-J characteristics in Figure 5 show either the presence of back
corona or a significant change in the inlet grain loading or particle size
distribution. Measurement with Brink impaetors at the inlet of the precipita-
tor showed changes in the 0.4 ym to 20 ym particle diameter range that would
probably not affect the V-J characteristics. Data below 0.4 ym were not
obtained. If back corona was occurring, the V-J characteristics indicated
that less back corona was being obtained when the additive was added.
The V-J characteristics in Figure 6 for the low sulfur coal test had
abnormal curvatures and the measurements for the high sulfur coal were off
scale and are probably in error. There is a possibility that the 1C-2C
power supply was not operating properly. The current densities at which
sparking occurred indicated a decrease in ash resistivity with the additive.
18
-------
40
36
32
28
24
UJ
IT
£E
O 20
o:
2
CM
O
O
LU
cn
16
12
8
4
LEGEND
A—Low Sulfur Coal with 951/hr
LPA 402A (12-18-75,1215)
O-High Sulfur Coal (12-16-75)
-Low Sulfur Coal with 951/hr
H20 (12-18-75, 1330)
_L
JL
30
38
32 34 36
SECONDARY VOLTAGE , kV
Figure 4. Secondary V-J Characteristics of Power Set 1A-2A.
19
40
-------
LEGEND
O-High Sulfur Cool (12-16-75)
D
— Low Sulfur Coal with 951/hr
LPA 402A (12-18-75,1215)
-Low Sulfur Coal with 951/
hr H20 (12-18-75, 1330)
30
32 34 36
SECONDARY VOLTAGE , kV
38
40
Figure 5. Secondary V-J Characteristics of Power Set 1B-2B.
20
-------
40,
CM
E
o
UJ
CO
36
32
28
24
UJ
£T
I 20
16
8
28
To 32 kV: 49.6 £sS
LEGEND
A-Low Sulfur Coal with 95 l/hr
LPA 402A (12-18-75, 1130)
O-Low Sulfur Coal with 95 l/hr
H20 (12-18-76,1400)
30
32 34 36
SECONDARY VOLTAGE , kV
38
40
Figure 6. Secondary V-J Characteristics of Power Set 1C-2C.
21
-------
CM
U
h-
LU
or
o
O-High Sulfur Coal (12-16-75)
A -Low Sulfur Coal with 95 l/hr
LPA 402A (12-18-75,1100)
O- Low Sulfur Coal with 95 l/hr
H20 (12-18-75,1415)
32 34 36 38
SECONDARY VOLTAGE , KV
Figure 7. Secondary V-J Characteristics of Power Set
1D-2D.
22
-------
250
200
£
2
* 150
LU
CC
CE
3
cr
<
1 100
50
LEGEND
O-High Sulfur Coal (12-16-75)
Sulfur Coal with 95 l/hr
LPA 402A (12-18-75,1215)
O- Low Sulfur Coal with 95 l/hr H20
(12-18-75, 1330)
200
Figure 8.
OPERATING
POINTS
L
j I
250 300 350
PRIMARY VOLTAGE , VOLTS
Primary V-I Characteristics, Power Set 1A-2A.
23
-------
250
200
LEGEND
O-High Sulfur Coal (12-16-75)
A —Low Sulfur Coal with 951/hr
LPA402A (12-18-75,1200)
_ O-Low Sulfur Coal with 951/hr H20
(12-18-75,1345)
o?
I I50h
H
UJ
or
£T
o
O
I 100
50
200
** .OPERATING
POINTS
FLUCTUATING
I
250 300
PRIMARY VOLTAGE , VOLTS
350
Figure 9. Primary V-I Characteristics, Power Set 1B-2B.
24
-------
250
200
1
UJ
EC
at
r>
o
cr
a.
150
100
50
LEGEND
O-High Sulfur Coal
(12-16-75)
A—Low Sulfur Coal with
LPA 402A (12-18-75)
_ O—Low Sulfur Coal
with H20 (12-18-75,
1400)
J I
I
OPERATING
POINTS
200
250 300
PRIMARY VOLTAGE , VOLTS
350
Figure 10. Primary V-I Characteristics, Power Set 1C-2C.
25
-------
250
200
o
>
cr
(T
Q.
150
100
50
200
PERATIN&
POINTS
LEGEND
O-High Sulfur Coal
(12-16-75)
A-Low Sulfur Coal with
951/hr LPA 402A
(12-18-75, 11=00)
a-
I
Low Sulfur Coal with
951/hr H20
(12-18-75, 1415)
I
250 300
PRIMARY VOLTAGE , VOLTS
350
Figure 11. Primary V-I Characteristics, Power Set 1D-2D.
26
-------
The V-J characteristics for the outlet power set are shown in Figure
7. The low sulfur coal curves with water injection and with the additive
injection have the same behavior. Both low sulfur V-J characteristics initially
lie below the high sulfur coal V-J characteristics then turn up to cross the
high sulfur coal characteristics. This type of behavior indicates the develop-
ment of back corona for the low sulfur coal ash. The small difference between
the low sulfur coal characteristics with water injection and with an additive
injection indicates that the additive was not affecting the resistivity of
the ash collected in the last section. However, a definite conclusion cannot
be made since only an hour was allowed for stabilization between measurements.
The current densities at the break (point at which heavy sparking
probably started) in the secondary V-J characteristics are tabulated in Table
5. Significantly higher current densities were obtained with the high sulfur
coal. A significant increase in obtainable current density was also observed
from the inlet to the outlet of the precipitator. No consistent differences
were obtained for the two low sulfur coal conditions; again, the short time
between recording of the V-J characteristics is probably responsible for this.
The average primary operating power densities for one chamber are tabulated
in Table 6. The high sulfur coal with lower resistivity ash had the highest
power densities, as expected.
27
-------
TABLE 5. CURRENT DENSITIES AT BREAK IN SECONDARY
V-J CHARACTERISTICS
High sulfur coal
Low sulfur coal, LPA 402A d
Low sulfur coal, H20
In-Situ
Resistivity
ft-cm
l.BxlO10
4.0 x 1010
9.9 x 1010
Current
1A-2A
nA/cm2
14
13.5
14.0
1B-2B
nA/cm2
18
17
10
Density
1C-2C
nA/cm2
b
13.5-21.6
13
1D-2D
nA/cm2
34
d9-21
9-21
At break in V-J characteristics or maximum obtained at all current densities
±33 percent
Abnormal V-J characteristics
C0btained on a December 10, 1975. V-J measured with LPA 402A injection December
18, 1975.
A sharp break in V-J characteristics is not evident; however, back corona
probably occurred at a current density less than 21 nA/cm2.
TABLE 6. AVERAGE OPERATING POWER DENSITIES
High sulfur coal
Low sulfur coal, LPA 402A
Low sulfur coal, H20
In-Situ
Resistivity
fi-cm
1.5xl010
4.0 x 1010
9.9 x 1010
Average ?
Primary Power Density, mW/cm
1A-2A
1.03
0.95
0.86
1B-2B
1.04
0.94
1.05
1C-2C
1.73
0.70
0.93
1D-2D
2.64
1.54
2.24
One chamber selected to correspond with V-J data.
28
-------
PARTICLE COLLECTION EFFICIENCY
The particle collection efficiency as a function of particle diameter
was calculated as described in Appendix E. The results of these calculations
are shown in Figures 12, 13, 14, 15, and 16. The 90 percent confidence inter-
vals for each curve are also shown. The confidence intervals were estimated
as described in Appendix E.
The confidence intervals for most of the particle collection efficiency
versus particle diameter curves are rather broad. As can be seen from Figure
17, much of the data for all five days falls within the 90 percent confidence
interval for the December 16, 1975 data. Such broad confidence bands make
the statistical interpretation of the data difficult if one desires a high
degree of statistical confidence.
The broad confidence interval is probably caused by source; i.e., boiler/
electrostatic precipitator system variation and cascade impactor measurement
errors. Of these, cascade impactor errors are probably more important.
In future tests the confidence intervals can be reduced by better (i.e.,
more precise) cascade impactors and/or by taking more data. Because state-
of-the-art impactors and techniques for using impactors were used for these
tests, it is unlikely that better impactors will be available soon. Thus
more data should be taken in future tests.
The efficiency versus particle diameter curves for the high sulfur coal
and conditioned low sulfur coal are consistent with the dust resistivity and
electrostatic precipitator operating data discussed earlier. The efficiency
versus particle diameter curves for the low sulfur unconditioned tests are not
consistent with the dust resistivity and electrostatic precipitator operating
data.
The data used to calculate the efficiency curve for unconditioned low
sulfur coal were taken less than 24 hours after the boiler was switched from
high sulfur to low sulfur coal. Such a short time is inadequate to allow the
electrostatic precipitator to adjust to the change in dust resistivity. The per-
formance of an electrostatic precipitator is governed, in part, by the electri-
cal resistivity of the dust layers on the collecting electrodes. Several days
operating time is required for the resistivity of the collected dust layer to
fully respond to a change in particle resistivity.
29
-------
99.9 h
0.1
H99.9
0.1
0.2 03 0.5 0.7 1.0 2 3 4 5 6 7 8 9 10
PARTICLE DIAMETER, microns
Figure 12. Efficiency versus particle diameter for December 10, 1975,
(Low sulfur coal conditioned with LPA 402A).
30
-------
0.1
5
10
I 2°
o
I 3°
| 40
< 50
tr
I- 60
g 70
a.
80
90
95
98
99
99.9
t-90%
Mean
-90%
I I 1—I I I,,,,I,,
99.9
99
98
95
90
80
70
60
40 y
o
30 t
UJ
20
10
5
2
I
0.1
0.1 0.2 03 0.5 0.7 1.0 2
PARTICLE DIAMETER, microns
4 5 6 7 8 9 10
Figure 13. Efficiency versus particle diameter for December 11
(low sulfur coal conditioned with LPA 402A).
31
-------
O.I
5
10
c 20
f
40 y
y
30 it
UJ
20
10
5
2
0.1
0.1 Q2 03 0.5 0.7 1.0 2 3
PARTICLE DIAMETER, microns
4 5 6 7 8 9 10
Figure 14. Efficiency versus particle diameter for December 16 (high
sulfur coal).
32
-------
"1—I—I—I I I
O.I
5
10
20
30
40
5°
60
TO
a.
80
90
95
98
99
99.9
0.1
+ 90%
Mean
-90%
99.9
99
98
95
90
80
70 |
60 !
50 >
z
40 y
o
30 fc
IU
20
10
5
2
I
0.1
0.2
0.5 0.7 1.0
PARTICLE DIAMETER, microns
5 6 7 8 9 10
Figure 15. Efficiency versus particle diameter for December 17
(high sulfur coal).
33
-------
O.I -
5
10
30
40
£60
UJ
g 70
a.
80
90
95
98
99
99.9
0.1
-90%
J L
99.9
99
98
95
90
80
70
60
50
40
30
20
10
5
2
I
0.1
0.2 03 0.5 07 1,0
4 5 6 7 8 9 10
PARTICLE DIAMETER, microns
Figure 16. Efficiency versus particle diameter for December 18
(low sulfur coal with water injection).
34
-------
O.I
5
10
20
| 'O
< S0
£60
UJ
5 70
a.
80
90
95
98
99
99.9
0.1
-90%
+90% •
DEC. 15
LEGEND
DECEMBER 16, HIGH SULFUR COAL
DECEMBER 10; LOW SULFUR
COAL-CONDITIONED
DECEMBER II, LOW SULFUR
COAL-CONDITIONED
DECEMBER 17, HFGH SULFUR COAL
DECEMBER 18, LOW SULFUR
COAL-CONDITIONED
J L
99.9
99
98
95
90
80
70
60
40
o
30 fc
20
10
5
2
I
0.1
0.2
03
0.5 0.7
1.0
5 6 7 8 9 10
PARTICLE DIAMETER, microns
Figure 17.
Efficiency versus particle diameter (all tests with-
+ 90% confidence).
35
-------
Cragle (1976) reports that the use of LPA 402A at Montour reduced parti-
culate emissions by a factor of about 5. Such a reduction in emissions is con-
sistent with the change in dust resistivity reported here when LPA 402A was
injected.
The performance of the electrostatic precipitator at Montour was modeled
using the electrostatic precipitator computer model described by Gooch et al.
(1975). A correlation developed by Hall (1971) was used to calculate the
allowable current density for the measured resistivity. Measured current
densities were not used because of their uncertainty (± 33 percent) and the
inertia of the electrostatic precipitator. The results of the computer model
calculations are shown in Figure 18.
The agreement between computer calculations and measured efficiencies is
good for the high sulfur coal and conditioned low sulfur coal tests. The
computer calculations are consistent with the results reported by Cragle (1976)
for conditioned and unconditioned low sulfur coal.
36
-------
O.I
5
10
1 2°
O
| 30
§ 40
t 50
60
70
80
90
95
98
99
99.9
0.1
LEGEND
HIGH SULFUR COAL
LOW SULFUR COAL CONDITIONED
WITH LPA402A
LOW SULFUR COAL,UNCONDITIONED ~
_LJL
99.9
99
98
95
90
80
70
60
50
40
30
20
10
5
2
I
0.1
0.2
03
0.5
0.7
1.0
5 6 7 8 9 10
PARTICLE DIAMETER, microns
Figure 18
Efficiency predicted by ESP computer model as a
function of particle diameter for Montour.
37
-------
SECTION 5
RECOMMENDATIONS FOR FUTURE TESTS
The data taken during the tests at Montour demonstrate the difficulties
of conducting experiments at full scale installations where complete control
and/or knowledge of all important parameters is impossible. The experimental
problems are compounded by the slow response of the electrostatic precipitator
to changes in ash resistivity and by the low precision of many of the measure-
ment techniques. The following discussion is offered as guidance for future
experiments similar to those conducted at Montour.
The fact that electrostatic precipitators respond slowly to changes in
ash resistivity must be taken into account in planning experiments involving
changes in ash resistivity; e.g., change from high to low sulfur coal (or use
of conditioning agents). The schedule for the experiments must allow several
days of electrostatic precipitator operation after conditions are changed
before testing can begin. Secondary voltage-current data and/or continuous
in-stack light transmittance data are useful for determining when to start
testing.
The variability of plant operating conditions (due to factors such as
change in boiler load, change in coal chemical composition, and ambient
temperature changes) is extremely difficult to deal with. The inertia
of the system and the immobility of some of the instruments used to make
measurements make it impractical to use randomized experimental design.
Once equipment is set up to make inlet measurements for example, half
a day to a day must be spent obtaining inlet data. The danger of such a
procedure is demonstrated by the results of the sulfur dioxide measurements
made at Montour when high sulfur coal was burned. It appears that the sul-
fur content of the coal on the day the outlet measurements were made was slightly
lower than the sulfur content of the coal on the day the inlet measurements
were made. The result of the change in coal sulfur content is an apparent
removal of sulfur dioxide by the electrostatic precipitator.
38
-------
A similar difference in coal sulfur content for the two low sulfur
coals burned at Montour may be responsible for part of the difference on the
sulfur trioxide data for the conditioned and unconditioned tests.
Simultaneous inlet and outlet measurements may help overcome some of the
problems due to source variability.
For studies of conditioning agents it may be necessary to conduct one
set of experiments to determine the effect of the conditioning agent on the
electrostatic precipitator and a separate set of experiments to determine the
impact of conditioning on gaseous emissions. An experimental design using
randomized treatments might be feasible for the experiments on gaseous
emissions.
The electrostatic precipitator computer model, Gooch et al. (1975), is
a useful tool for dealing with some of the problems caused by source variability
and system inertia. For example, the computer model can be used to estimate
what will happen if sufficient time is available for the electrostatic
precipitator to reach steady-state operation.
Secondary voltage-current data for all TR sets are essential for electro-
static precipitator evaluations.
The measurement techniques for many of the emissions of interest—for
example, sulfur trioxide, ammonia, and particle size distribution—lack
adequate precision when used in the field. Large amounts of data are required
to statistically examine field test data. This fact must be kept in mind
when a test plan is developed. The amount of data required for each test
condition should be estimated from a knowledge of the precision of the measure-
ment techniques and an estimate of the change in the quantity being measured
due to change in operating conditions. If the time required for obtaining the
required data is excessive, low levels of statistical confidence may have
to be accepted.
39
-------
SECTION 6
REFERENCES
1. Cragle, S., "Operating Experience with ESP Conditioning in Relation to
an Electrostatic Precipitator Upgrading Program", paper presented at
Conference on Particulate Collection Problems in Converting to Low Sulfur
Coals. Research Triangle Park, North Carolina, 1976.
2. Dismukes, E. B., Conditioning of Flyash with Sulfur Trioxide and
Ammonia. EPA Report EPA-60Q/2-75-015, NTIS PB 247-231/AS. 1975.
3. Gooch, J. P., McDonald, J.R., and Oglesby, S. A Mathematical Model of
Electrostatic Precipitation. EPA Report EPA-650/2-75-037, NTIS PB 246-
188/AS. 1975
4. Hall, H.J., "Trends in Electrical Energization of Electrostatic Precipi-
tators". Proceedings of the Electrostatic Precipitator Symposium, February
23-25, 1971. Birmingham, Alabama.
40
-------
July 1976
APPENDIX A
EFFECT OF FLYASH CONDITIONING AGENT
ON COAL-FIRED POWER PLANT EMISSIONS
by
D.B. Harris
Process Measurements Branch
Industrial Environmental Research Laboratory
Office of Research and Development
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
41
-------
DISCLAIMER
This report has been reviewed by the Process Measurements Branch,
Industrial Processes Division, U.S. Environmental Protection Agency, and
approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
42
-------
ABSTRACT
The gaseous and particulate emissions from a coal fired power plant
were studied to investigate the environmental effects of a chemical flyash
conditioning agent. Low sulfur coal was burned, and the emission sampled,
with and without chemical addition. High sulfur coal was burned without
chemical addition. Sulfur oxides, ammonia, and organics concentrations
were determined during the test program, and the flyash resistivity and
precipitator removal efficiency were measured.
Mean SC^ concentrations varied from 663 ppm for low sulfur, unconditioned
(LS, UN) ash to 1141 ppm for the high sulfur coal (HS, UN). Low sulfur coal
with conditioner (LS, C) yielded a mean of 879 ppm SOo. The SO., concentration
for LS, C and HS, UN cases averaged 29 ppm, while the LS, UN coal produced
a mean SO- concentration of 16 ppm.
Ammonia was detected only during the LS, C run, and was apparently a
decomposition product of the conditioner.
The organic samples were quantitatively analyzed only for diethylnitros-
amine, and none was found by a procedure with a lower detection limit of about
5 ppt. A Level I analysis was also conducted.
Flyash resistivity measurements were made and, at constant temperature,
the LS, C ash had a resistivity 60 percent lower than did the LS, UN ash,
and generally comparable to the HS, UN ash resistivity. Particulate removal
efficiency measurements, however, indicated that the LS, UN and HS, UN ashes
were removed to about the same high efficiency, while the LS, C ash was less
efficiently removed. A precipitator "inertia" effect, improving the LS, UN
results, is suggested as a possible reason for this anomaly.
43
-------
CONTENTS
Abstract 43
Figures 46
Tables 47
Acknowledgments 48
A-l. Introduction , 49
A-2. Experimental 50
2.1 Test Schedule 50
2.2 Sulfur Oxides 51
Description of Test Method 51
Quality of Measurement 54
2.3 Ammonia 54
Description of Test Method 54
Discussion of Titration Method 56
Quality of Measurement 56
2.4 Organic Vapors 56
Quality of Measurement 58
2.5 Particulate Size Measurement 58
Description of Test Method 58
Quality of Measurement 60
2.6 Flyash Resistivity 61
Measurement Technique 61
Quality of Measurement 61
A-3. Results 63
3.1 Sulfur Oxides 63
Montour Data 63
Comparison of Montour to Other Power Plants 67
3.2 Ammonia Concentrations 67
3.3 Orgam'cs 67
44
-------
3.4 Particle Size 71
Particle Size at Inlet to ESP 71
Outlet Particle Size Distributions 71
Fractional Efficiency 71
3.5 Flyash Resistivity 71
3.6 Boiler Operation Parameters 83
A-4 References 85
Appendix A-A Analysis of Monteurville Power Plant Effluent 86
Appendix A-B Impactor Data 94
45
-------
FIGURES
Number Page
A-l Sulfur oxides sample train 53
A-2 Ammonia sampling apparatus ... 55
A-3 Organic sampling apparatus 57
A-4 Impactor sampling train 59
A-5 Point-to-plane flyash resistivity probe 62
A-6 Results of Level I organic analysis at Montour 70
A-7 Differential particle size distribution at inlet to ESP, Mon-
tour Plant, December 10 and 11, 1975; low sulfur coal, condi-
tioner injected at 95 1/hr 72
A-8 Differential particle size distribution at inlet to ESP, Mon-
tour Plant, December 16 and 17, 1975; high sulfur coal, no
conditioner 73
A-9 Differential particle size distribution at inlet to ESP, Mon-
tour Plant, December 18, 1975; low sulfur coal, water injected
at 95 1/hr 74
A-10 Mean differential size distribution at inlet to ESP, Montour
Plant, December 10, 11, 16, 17, and 18, 1975 75
A-ll Differential particle size distributions at outlet of ESP,
Montour Plant, December 10 and 11, 1975 76
A-12 Differential particle size distributions at outlet of ESP,
Montour, December 16 and 17, 1975 77
A-13 Differential particle size distributions at outlet of ESP,
Montour, December 18, 1975 78
A-14 Differential particle size distributions at outlet of ESP,
Montour, December 10, 11, 16, 17, and 18, 1975 79
A-l5 Fractional removal efficiency of Montour ESP 80
A-16 Flyash resistivity measurements, Montour Power Plant .... 81
46
-------
TABLES
Number
A-l
A-2
A-3
A-4
A-5
A-6
A-7
A-8
A-9
A-A-1
A-B-1
A-B-2
A-B-3
A-B-4
A-B-5
A-B-6
A-B-7
A-B-8
A-B-9
A-B-10
Test Schedule
Test Plan . .
Sulfur Oxides Concentration at Montour Steam Power Plant Low
Sulfur Coal, Flyash Conditioned with 95 1/hr LPA 402A ....
Sulfur Oxides Concentration at Montour Steam Plant High Sulfur
Coal, No Additive Injection
Sulfur Oxides Concentration at Montour Steam Plant Low Sulfur
Coal, 95 1/hr Water Injection
Comparison of Sulfur Oxides Data From Various Power Plants .
Ammonia Concentration at Montour Power Plant
Flyash Resistivity
Boiler Operation Parameters, Montour Test
Digest of Data Sheets from Tenax Organic Runs. Montour Steam
Power Plant, Danville, Pa
Inlet Impactor Data,
Inlet Impactor Data,
Inlet Impactor Data,
Inlet Impactor Data,
Inlet Impactor Data,
Outlet Impactor Data,
Outlet 'Impactor Data,
Outlet Impactor Data;
PP&L, Montour Plant (1-3 thru -6) .
PP&L, Montour Plant (1-7 thru -10) .
PP&L, Montour Plant (1-11 thru -16)
PP&L, Montour Plant (1-17 thru -19)
PP&L, Montour Plant (1-20 thru -25)
PP&L, Montour Plant (0-1 thru -5) .
PP&L, Montour Plant (0-6 thru -10)
PP&L, Montour Plant (0-11 thru -15)
Outlet Impactor Data, PP&L, Montour Plant (0-16 thru -19)
Outlet Impactor Data, PP&L, Montour Plant (0-20 thru -23)
Paqe
50
51
64
65
66
68
69
82
84
93
98
99
100
101
102
103
104
105
106
107
47
-------
ACKNOWLEDGMENTS
Pennsylvania Power and Light and the Montour Power Plant staff were
very helpful throughout this test program. Their assistance was invaluable.
Appreciation is expressed to Dr. Wallace Smith, Dr. Herbert Spenser,
and Dr. Edward Dismukes of Southern Research Institute for their assis-
tance in gathering and interpreting the information presented in this
report.
Battelle Columbus Laboratories performed the organic vapor analyses
for the Montour tests.
Douglas VanOsdell of the Research Triangle Institute provided
editorial assistance in the preparation of this report.
48
-------
SECTION A-l
INTRODUCTION
The Particulate Technology Branch (PATB) of IERL-RTP requested in
November 1975 that the Prpcess Measurements Branch (PMB) conduct a study of
the effects of a chemical flyash conditioning agent upon the effluent
characteristics of an ESP-controlled coal-fired power plant. Arrangements
were made with the Pennsylvania Power and Light Company to test such a
scheme at their Montour Station during December 1975.
The goals of the program as agreed to by PATB and PMB were:
1) Ascertain the gaseous effluent content for both additive
and non-additive conditions. This was to include S02 and S0o»
NHg, and organics (specifically nitrosamines).
2) Measure the flyash resistivity.
3) Determine the flyash particle size distribution.
Because of the manpower limitations, no effort was to be made to obtain
mass loading information. Through a PATB contract, Southern Research
Institute personnel were provided to assist in gathering the data.
49
-------
SECTION A-2
EXPERIMENTAL
2.1 TEST SCHEDULE
The Montour test program was conducted as described in Table A-l. At
the onset of the program, December 10 and 11, low sulfur coal was burned
and the LPA 402A was injected at 95 1/hr. During the weekend, and through
December 17, high sulfur coal was burned without injection of any conditioner.
On December 18, 1975, the plant was switched back to low sulfur coal, but
the additive was not injected. Water was injected as a conditioner at 95
1/hr during this low sulfur run. The original test plan called for continuous
low sulfur coal burning from the beginning of the test until after the low
sulfur coal tests were completed. A low sulfur coal train was derailed,
however, and the Montour plant was forced to go to high sulfur coal over
the weekend and then back to low sulfur coal to complete the test.
TABLE A-l. TEST SCHEDULE (1975)
December 8 and 9
December 10 and 11
December 12
December 13 and 14
December 15
December 16
December 17
December 18
Arrive at plant and set up-
Full load tests using low sulfur coal,
conditioner LPA 402A injected.
No tests; crew returns to Research
Triangle Park, N.C.
Weekend, no tests.
Crew returns to plant-
Full load tests using high sulfur coal.
No additive injection.
Full load tests using high sulfur coal.
No additive.
Full load tests using low sulfur coal,
water injected.
50
-------
The preclpitator controlling a single boiler was tested on both the inlet
and outlet sides. The additive was injected upstream of the inlet port. The
precipitator had a total plate area of 42,808.3 m2 (460,800 ft2), with
2675.5 m2 (28,800 ft2) supplied by each power supply.
Table A-2 presents the test plan for Montour. The 5 cfm cyclone train
was not used due to damaged connector threads. The remainder of the plan
was followed.
TABLE A-2. TEST PLAN
Test
Location
No. Per Day
Opacity
Opac i ty
ESP Inlet
EPA Instrument
ESP Outlet
PP&L Instrument
Time Per Test
Dust Resistivity SRI Probe
Particle Size Distribution
using Brink Impactors
Particle Size Distribution
using Andersen Impactors
and 5 cfm Cyclone Train
S02/S03 and NH3
Organic s
ESP Inlet
ESP Inlet
ESP Outlet
ESP Inlet
& Outlet
ESP Inlet
& Outlet
3
4
4
2
2
2-3 hrs
1/2-1 hr
1-2 hrs
4
4
Continuous
Continuous
2.2 SULFUR OXIDES
Description of Test Method
Sulfur oxides concentrations during the Montour test program were
determined by the IPA-Thorin titration method. The method was originally
developed by Shell Development Co. and has been adapted by EPA. It is based
on absorption of the sulfur oxides in 80 percent isopropanol (IPA) and 3 per-
cent hydrogen peroxide (1^02), and subsequent titration for sulfate ion.
51
-------
Figure A-l is a schematic of the apparatus used at Montour. Glass wool
inside the duct filters out particulate. The temperature of the probe was
kept at approximately 121°C (250°F) in order to prevent condensation within
the probe. The first absorber contained 30 milliliters of 80 percent IPA.
The alcohol was used to extract $03 and prevent oxidation of the S02. The
absorbers used were modified Lamp sulfur absorbers called Shell bubblers.
1 ij
Alyea and Backstrom , and Flint have confirmed the effectiveness of IPA as
an inhibitor of S02 oxidation. Gaseous SOo is imperfectly extracted in the
IPA and forms a fine sulfuric acid mist. A disengagement device must be
used to prevent carryover of the mist into the next absorber. A bulbous
disengagement volume was used in the Montour test program. The works
of Flint2, Shell3, and Matty and Diehl4 have confirmed that S03 can be
quantitatively extracted using this procedure.
Gaseous S02 is absorbed in the IPA to some extent, and it is necessary to
purge air or nitrogen through the IPA solution to strip the S02 out of the
IPA. Shell has shown that 15 minutes is adequate to strip out all of the
dissolved S02 when using Lamp sulfur absorbers. Approximately 15 minutes of
air purge was used during the Montour tests.
The second and third bubblers in the train contained a 3 percent aqueous
solution of hydrogen peroxide to extract the S02 from the gas stream. Shell
o
Development Co. has shown that 100 percent of the S02 can be recovered by
using two H202 absorbers in series.
Downstream of the second H202 absorber a drying column was used to dry
the gas before it entered the dry gas meter. Temperature and pressure measure-
ments were made in conjunction with the flowrate measurement.
After the test program was completed, the concentration of sulfate in the
absorbers was determined by titration with a barium perchlorate solution to
the Thorin endpoint. The first absorber solution and the disengagement
impinger were washed together in order to collect all of the S03 sulfate,
then titrated. The S02 concentration was determined by titration of the combined
solution from the last two absorbers after addition of some IPA. The Thorin
indicator is active only in an alcohol solution. A blank was carried throughout
the titration procedure each day. The IPA and H202 solutions were prepared
fresh daily.
52
-------
SHELL BUBBLERS
PUMP WITH
BY PASS
HEATING TAPE AND
TEMPERATURE CONTROL
DUCT
WALL
202
SOLUTION
GAS
METER
Figure A-"I. Sulfur oxides sample train.
-------
Quality of Measurement
Shell Development determined that acceptable sensitivity could be obtained
over the range of 25 to 6000 parts per million of total sulfur oxides using
3
the IPA-Thorin technique. This estimate was based on a 42 liter gas sample.
Errors in the IPA-Thorin Titration have been determined to be:
± 0.4 % for 25 yg S03/ml
± 4 % for 2.5 yg S
In other words, the titration is accurate to within ± 0.1 yg S03/ml . For 30
ppm of S03 in a 60 liter gas sample with the absorbent solution diluted to
250 ml and titrated, the solution would contain about 25 yg S03/ml , for an
error of about 0.4 percent. The S03 concentration in the H202-IPA solution
will be higher, and the error less. Absolute accuracy of the method has
not been determined. However, comparison of the IPA-Thorin with other tech-
niques (Seidman and Fritz and Yamamura') has shown no significant differences.
The IPA-Thorin procedure is subject to some limitations. Errors may occur
in high fluoride gas streams (fluoride concentration > sulfide concentration),
if the filter is not used and particulate penetrates the probe or if conden-
sation occurs in the probe. Ammonia and certain ammonia compounds are also
said to interfere with the SO, determination, although this has not been
8
quantified.
2.3 AMMONIA
Description of Test Method
Ammonia concentration during thq Montour test was determined by a
modified Kjeldahl-Titrimetric Procedure. The procedure was developed by
EPA personnel. It is based on the absorption of the ammonia in sulfuric
acid followed by titration.
A schematic of the apparatus used at Montour to collect ammonia is shown
in Figure A-2. Glass wool was again used to prevent particulate from entering
the probe. Heating tape was used to control the temperature of the probe
and the absorbers are midget impingers which contain 15 ml of 0.1 normal sul-
furic acid. Two empty impingers were used as entrainment separators. A silica
gel column was used to dry the gas and temperature and pressure measurements
were made concurrently with the flow rate measurement.
54
-------
GLASS
.WOOL
en
en
DRYING COLUMN
HEATING TAPE AND
TEMPERATURE CONTROL
PUMP WITH
BYPASS
ICE BATH
3 IMPINGERS CONTAINING 15 ML
O.I N H2S04, 2 EMPTY IMPINGERS
DRY GAS METER
Figure A-2. Ammonia sampling apparatus.
-------
Discussion of Titration Method
After the gas sample had been passed through the absorbent solution,
the contents of the impingers were combined. Free ammonia was distilled over
into a flask containing boric acid and indicator solution (2:1 mixture of methyl
red and methylene blue in 95 percent ethanol), where ammonia borate was formed.
Titration with 0.02 N H2S04 yielded the acid equivalent to the distilled
ammonia. The indicator change was from green to blue.
Quality of Measurement
EPA personnel have determined that this method of ammonia concentration
determination has an accuracy of ± 2 percent between 10 and 500 ppm for a 20
liter sample. The precision is about the same. The procedure has not been
used above 500 ppm, and its upper limit is unknown.
A blank was used to check on the titration and correction was applied
to the results if needed. A blank was run each day, as the solutions were
made up fresh daily.
2.4 ORGANIC VAPORS
The composition of organic vapors which might be present in the flue gas
was of interest because of possible undesirable decomposition products from
the injected additive. Organic vapors frequently occur in flue gases in very
low concentrations and it is often necessary to use concentrating techniques
to make accurate measurements. Absorption of the organics from a large gas
volume onto a polymer medium, followed by desorption into a smaller volume,
is a commonly used method. During the Montour tests, about 1415 liters (50
scf) of gas was passed through the absorbent and the organics were desorbed
into about 120 ml of pentane. The sample train used in the Montour tests is
shown as Figure A-3. The gas sample was collected by a heated probe which
was kept constant at about 70°C. The cyclone and filter were used to remove
particulate from the stream, and the oven and heat exchanger maintained a
constant gas temperature of about 60°C. The particulate was not analyzed.
The prepackaged cartridge contained Tenax GC [a polymer of 2,6-(diphenylpara-
phenylene)oxide] as the absorbent. The remainder of the train consisted of a
standard mass sampling train. Sulfur oxides were removed and the gas was
dried prior to flowrate measurement.
56
-------
PROBE
en
OVEN
TENAX GC
ABSORPTION
COLUMN
PUMP
LEAR-SIEGLER TRAIN
HEAT EXCHANGER
Figure A-3. Organic sampling apparatus.
-------
The analysis of the organic samples was conducted as described in
g
"Technical Manual for Analysis of Organic Materials in Process Streams."
The absorbed organics were recovered in a 24 hour continuous pentane extrac-
tion. Level I (semi-quantitative) analysis was performed on half of each
extracted sample. Each sample was fractionated by liquid chromatography
into eight fractions. The general classes of compounds expected in each frac-
Q
tion were known from previous work. Fourier Transform Infrared Spectroscopy
(FTIR) was used to achieve semi-quantitative data for each of the 40
individual fractions.
Further analysis was carried out on the unfractionated samples. Diethyl-
amine was chosen as a likely decomposition product, and each of the samples
was analyzed by gas chromatography/mass spectroscopy (6C/MS) for the amine.
The analysis procedures are elaborated upon in a letter report by
P.W. Jones to Dr. L.D. Johnson, which is attached as Appendix A-A.
Quality of Measurement
The sensitivity of the FTIR analysis varies for the different fractions
of the Level I analysis. Fraction 1 and 2 compounds are detected at around
1-10 yg while those in Fractions 3 through 8 are detected at 0.1 to 1 yg levels,
The GC/MS analysis used to detect and quantify diethylnitrosamine is both
sensitive and precise. Recoveries of nitrosamine in processing were estimated
at 80 percent. As explained in Appendix A-A, the detection limit for nitros-
amine was around 5 yg, and for a 1400 1 (50 scf) sample this is about 3
parts per trillion (ppt).
2.5 PARTICIPATE SIZE MEASUREMENT
Description of Test Method
Cascade impactors, inertia! sizing devices, were used to determine
particulate size distributions at Montour, The Brink Impactor, a low sample
rate device, was used on the inlet to the electrostatic precipitators and the
Andersen Impactor, a high sample rate impactor, was used on the outlet. Nor-
mal isokinetic sampling procedures were followed with both impactors and they
were operated to minimize scouring and overloading.
A particulate sampling train representative of that used with the Ander-
sen and the Brink is shown in Figure A-4. Each impactor was operated within
the stack and the temperature of the gas in the impactor was measured by
58
-------
GAS FLOW
IMPACTOR
HEATING
JACKET
DRYING
COLUMN
CONDENSERS
TEMPERATURE
CONTROLLER
HEAT
EXCHANGER
ICE BATH
BLEED
CALIBRATED
ORIFICE
•VENT
\ ^ ,1.
1 ™- • 1^ ' '
V
\^
C
! j
)RYGA
METEF
If tor
VACUUM
PUMP
J
{
MANOMETERS
(OR DIFF. PRESSURE GAUGES)
Figure A-4. Impactor sampling train.
59
-------
thermocouple. A heat exchanger was used to reduce temperature fluctuations
during the sample period. The drying column of indicating silica gel dried
the gas. Pressure and temperature measurements were made concurrently with
the flowrate measurements. Dual flowrate instruments were used: an orifice
meter with a manometer for flowrate measurements and a dry gas meter for
volumetric flow measurements.
The impactors were disassembled and the mass of particulate on each
stage was determined gravimetrically. Factory precut glass fiber substrates
were used in the Andersen impactor. Foil substrates, coated with polyglycol
grease, were used as substrates in the Brink impactor. A straightforward data
reduction was used to determine the differential size distribution on both
the inlet and outlet measurements.
Quality of Measurement
Several steps were taken to increase the reliability of this data. The
data were examined and unreliable data were eliminated as described in Appendix
A-B. The weights of particulated collected on each stage during the impactor
runs have also been included in Appendix A-B. The fiberglass substrates used
with the Andersen impactor were preconditioned by in-situ exposure to the flue
gas using procedures developed by the Southern Research Institute. This was
to reduce the problem of anomalous weight gains which have been observed in
some cases in SOp containing flue gases. Blanks were also run each day with
the Andersen impactor. The stage weights were determined after operating
the impactor on filtered flue gas, and the weight change on each stage was
measured. If weight gains were observed on the stages, the average weight
gain was determined and subtracted from the Andersen stage weights obtained
during the day's runs. More specific information on the blank runs is pre-
sented in Appendix A-B. Blanks were not run with the Brink impactor, as the
impactor was upright during the test and experience indicated they would
be unnecessary.
The data reduction used the Southern Research Institute computer program
which is based on a fixed impaction parameter of 0.145 for round jet impactors.
60
-------
2.6 FLYASH RESISTIVITY
Measurement Technique
The resistivity of the flyash was measured using an in-situ point-to-
plane resistivity probe, and the reported results were obtained using the
parallel disc method. Figure A-5 shows a probe of this type. The point-to-
plane in-situ probe simulates dust collection as it occurs in an ESP. The
data derived from point-to-plane probes is, therefore, thought to be more
relevant to precipitator performance than is data derived from mechanically
collected dust.
The probe was inserted in the duct and allowed to come to thermal equili-
brium prior to taking a reading. The parallel disc method requires that a
layer of dust (usually about 1 mm thick) be collected on the collection sur-
face. The thickness of this layer was determined in-situ by lowering the
upper disc until it just contacted the dust. With the upper disc in position,
the voltage drop across the two surfaces was increased until the dust layer
broke down electrically and sparkover occurred. From knowledge of the
geometry of the surfaces and the voltage and current just prior to spark-
over, the flyash resistivity was calculated.
Quality of Measurement
Flyash resistivity measurements as made at Montour have a precision of
around ± 30 percent. This has been the precision achieved by Southern Research
Institute personnel on similar tests and during this test program. The
absolute accuracy of the data is not known, if indeed there is an absolute
measurement. The measurement must be made indirectly, and is affected by the
method of depositing the dust, of determining the thickness of the layer, by
coal, by temperature, and boiler variables. The usefulness of the data is
in its application to ESP design and performance, based on experience with
similar tests.
61
-------
HIGH VOLTAGE
CONNECTION
DIAL INDICATOR
PICOAMMETER
CONNECTION
MOVABLE
SHAFT
STATIONARY
POINT
GROUNDED
RING
Figure A-5. Point-to-plane flyash resistivity probe.11
62
-------
SECTION A-3
RESULTS
3.1 SULFUR OXIDES
Montour Data
The sulfur oxides data taken during the LPA 402A injection are presented
in Table A-3. The average S02 concentration was 879 ppm and the mean SO,
concentration was 29 ppm. Low sulfur coal was being burned in the boiler
while 95 1/hr of the additive was being injected. Data taken while burning
high sulfur coal are presented in Table A-4. The mean concentrations for
S02 and SOg were 1141 ppm and 23 ppm, respectively. Nothing was injected
during this period. Table A-5 presents data on low sulfur coal with water
injection.
Coal from the same seam was burned during both low sulfur coal runs,
and it may be significant that the flue gas during water injection is con-
siderably lower in sulfur oxides concentration. The plant was operating
at about 8 percent greater output during the water injection runs.
As indicated in Tables A-3, A-4, and A-5, several of the SOV data points
A
were rejected as spurious values. A data point was rejected if it fell out-
side of the 90 percent confidence limits of the remainder of the data.
The confidence limits were calculated about the mean as a multiple of the
standard deviation. Based on the number of data points, "t-values" were
12
used as the multiplying factor for the standard deviation. This procedure
is somewhat misleading, as it implicitly assumes that all the variation is
random measurement error. In the case of power plant flue gas, boiler var-
iation may impose additional variability on the data which is not error.
With this limitation in mind, the procedure is still useful during periods
of fairly constant boiler operation, as was apparently the case during the
Montour tests. For the same reason, caution must be exercised when evalua-
ting all of the statistical treatments of these data.
63
-------
TABLE A-3. SULFUR OXIDES CONCENTRATION AT MONTOUR STEAM POWER PLANT LOW
SULFUR COAL. FLY ASH CONDITIONED WITH 95. 1/hr'LPA 402A..
Sample
1
2
3
4
5
6
OUTLET
17
18
19
20
21
INLET
OVERALL
Location
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Inlet
Inlet
Inlet
Inlet
Inlet
AVERAGE
Date
&
Time
12/10/75
1000
12/10/75
1030
12/10/75
1035
12/10/75
1050
12/10/75
1100
12/10/75
1120
12/11/75
1105
12/11/75
1120
12/11/75
1130
12/11/75
1140
12/11/75
PPM
so2
856.0
873.4
875.5
826.5
700. 8a
920.4
X" = 870.4
a = 34.2
CV = 4%
957.0
857.1
1088.13
862.5
883.9
X" = 890.1
a = 46.1
CV = 5.2%
879
PPM
so3
28.2
6.1
50.0
24.5
14.6
56.8
X" = 30.0
o = 19.8
CV = 66 %
38.1
29.0
31.3
18.2
19.5
X" = 27.2
o = 8.35
CV = 30.7%
28.8
PPM
Total
S0x
884
879
926
851
977
903
995
886
881
903
916
909
Value rejected as spurious. Outside of 90% confidence limit. Mean (X),
standard deviation (a), and coefficient of variation (CV) do not include
this point.
64
-------
TABLE A-4. SULFUR OXIDES CONCENTRATION AT MONTOUR STEAM PLANT HIGH SULFUR
COAL, NO ADDITIVE INJECTION
Sample
25
26
27
28
INLET
29
30
31
32
OUTLET
OVERALL
Location
Inlet
Inlet
Inlet
Inlet
Outlet
Outlet
Outlet
Outlet
AVERAGE
Date
&
Time
12/16/75
1030
12/16/75
1050
12/16/75
1105
12/16/75
1120
12/16/75
1400
12/16/75
1420
12/16/75
1430
12/16/75
1440
PPM
so2
1261.9
1444.2
1373.3
1281.4
I = 1 340
o = 84.6
CV = 6.3%
926.1
970.9
769.7
1100.8
X" = 942
a = 136.6
CV = 14.5%
1141
PPM
so3
13.3
12.4
32.7
56. S3
X" = 19.5
a = 11.5
CV = 59%
10.7
28.2
34.5
28.9
X" = 25.6
a = 10.3
CV = 40.3%
23
PPM
Total
S0x
1275
1457
1406
—
1379
937
999
804
1130
968
1164
a Value rejected as spurious. Outside of 90% confidence limit. Mean (X),
standard deviation (a), and coefficient of variation (CV) do not include
this point.
65
-------
TABLE A-5. SULFUR OXIDES CONCENTRATION AT MONTOUR STEAM PLANT LOW SULFUR
COAL, 95 1/hr WATER INJECTION
Sample Location
41 Outlet
42 Outlet
43 Outlet
OUTLET
48 Inlet
49 Inlet
50 Inlet
INLET
OVERALL AVERAGE
Date
&
Time
12/18/75
1345
12/18/75
1400
12/18/75
1420
12/18/75
1540
12/18/75
1615
12/18/75
1630
PPM
so2
574.2
674.4
683.6
1= 644
a = 60.7
CV = 9.4%
872.8
656.5
518.9
X" = 683
a = 178.4
CV = 26.1%
663
PPM
so3
8.8
17.4
21.9
X" = 16.2
a = 6.7
CV = 41.5%
19.5
56. 7a
12.4
X"= 16.0
a = 5.0
CV = 31.5%
16
PPM
Total
S0x
583
692
706
660
893
—
531
712
679
a Value rejected as spurious. Outside of 90% confidence limits. Mean (X),
standard deviation (a), and coefficient of variation (CV) do not include
this point.
66
-------
Comparison of Montour to Other Power Plants
Table A-6 presents the results of a comparison of the Montour sulfur
oxides data with that taken at other power plants. The Montour data
consists of from three to six point averages, while the other data are
averaged from at least eight points. The variability in the SC^ measure-
ments at Montour is consistent with that found at other locations. The
coefficients of variation of the S03 measurements are, in some cases, larger
than the norm. The small number of measurements in the Montour sample pro-
bably accounts for some of this variation.
3.2 AMMONIA CONCENTRATIONS
The results of the ammonia concentration measurements are presented
in Table A-7. Ammonia was detected only during the injection of LPA 402A.
The reliability limit of the test procedure has been established as 10 ppm,
and five of the samples in which ammonia was detected are below this limit.
The conditioning agent apparently contained nitrogen which could decompose
to ammonia.
3.3 ORGANICS
The organic samples collected on the Tenax cartridges were extracted,
then fractionated according to the methods presented as Level I analysis in
9
"Technical Manual for Analysis of Organic Materials in Process Streams."
The five samples were each divided into two parts. One part was reserved
for further analysis, while the other was fractionated by liquid chromato-
graphy into eight fractions. The weight of material separated into each
fraction is presented below.
Sample
Jl
J2
J3
J4
SW1
1
Fraction Height (g)
345
8
0.00054 0.0068 0.00082 0.0057 0.0014 0.00006
0.00054 0.0091 0.00084 0.0065 0.00009 0.00003
0.00023 0.0060 0.00018 0.00048 0.00004 0.00008
0.00018 0.000077 0.00013 0.0039 0.00021 0.00027 0.00003 0.00001
0.0014 0.0038 0.00047 0.0010 0.00047 0.0077 0:00001 0.000005
0.00023 0.0011
0.0024 0.00044
0.00023 0.00018
67
-------
TABLE A-6. COMPARISON OF SULFUR OXIDES DATA FROM VARIOUS POWER PLANTS
Power Plant
PPL (Shemokan)
March 1975
PPL (Shemokan)
March 1975
IPC (East Alton)
WEPCO
Shawnee
Wai den Research
Averages
PPL (Montour)
Low S, water injection
Inlet
Outlet
Low S, LPA 402A
injection
Inlet
Outlet
High S, no injection
Inlet
Outlet
S02
(ppm)
1700
652
2339
1550
--
—
—
683
644
890
870
1340
942
Cvso2
6%
13%
13%
13%
--
--
11%
26%
10%
5%
4%
6%
14%
so3
(ppm)
54.7
40.2
11.0
16.3
25.0
29.0
--
16
16
27
30
20
26
cvso3
26%
26%
16%
28%
45%
28%
28%
32%
42%
31%
66%
59%
40%
68
-------
TABLE A-7. AMMONIA CONCENTRATION AT MONTOUR POWER PLANT
Sample
Time
Date
Location
Ammonia
(ppm)
Low Sulfur Coal, LPA 402A Injected at 95 1/hr
7
8
9
10
11
12
13
14
15
16
1535
1545
1600
1610
1620
0915
0930
0940
0950
1000
12/10/75
12/10/75
12/10/75
12/10/75
12/10/75
12/11/75
12/11/75
12/11/75
12/11/75
12/11/75
High Sulfur Coal, No Injection
Six samples on 12/16/75 and 12/17/75
Low Sulfur Coal, Water Injected
Eight samples on 12/18/75
Outlet
Outlet
Outlet
Outlet
Outlet
Inlet
Inlet
Inlet
Inlet
Inlet
18.2
17.5
None Detected
4.7
None Detected
2.6
4.5
5.8
1.9
None Detected
None Detected
None Detected
Each of the 40 individual fractions was further analyzed by Fourier
Transform Infrared Spectroscopy (FTIR). This analysis detected certain material
classes which were present in comparatively large amounts (< around 10-30 yg).
Figure A-6 presents these results. The compounds or classes of compounds which
are identified are those which were found by FTIR, and are not the only materials
present in that fraction.
A quantitative analysis for nitrosamine was carried out on half of each of
the original organic samples, Diethylnitrosamine was chosen as a likely
decomposition product of the injected conditioner. No diethylnitrosamine was
found in any of the samples; for the gas volumes which were collected, the
nitrosamine composition must have been below about 2-5 parts per trillion.
Further discussion of the results is attached in. Appendix A-A.
69
-------
Jl
LOW SULFUR COAL
LPA 402A INJECTED
-AROMATIC HC (fused) ALIPHATIC KETONE-, CARBOXYLiC ACID (soap)
-SILICONE AROMATIC KETONE/,./ ALCOHOL (aliphatic) or
HC QUINONE ,jy ALKANAL AMINE •
ALIPHATIC ESTER ///CARBOXYLIC
: s # ACID **•
I
4
JSL
6
ZO 40 60 80
WEIGHT PERCENT OF NUMBERED FRACTION
100
J2
LOW SULFUR COAL
LPA 402A INJECTED
-AROMATIC HC (fused)
SILICONE
-AROMATIC HC
-ALIPHATIC ESTER
.AROMATIC
ALIPHATIC KETONE
CARBOXYLIC ACID SALT (soap]^
ACID
*CARBC
6
30 4O 60 80
WEIGHT PERCENT OF NUMBERED FRACTION
J3
HIGH SULFUR COAL
NO INJECTION
.SILICONE
-AROMATIC HC
-AROMATIC KETONE/QUINONE
DIFFERENT AROMATIC
QUINONE
ALIPHATIC KETONE
ZO 40 60 80
WEIGHT PERCENT OF NUMBERED FRACTION
100
BWI
LOW SULFUR COAL
H20 INJECTED
•SILICONE
-AROMATIC KETONE/GUI NONE
SUBSTITUTED PLENOL
/-SECOND PLENOL
6
20 40 SO 80 100
WEIGHT PERCENT OF NUMBERED FRACTION
Notes: M SAMPLE J4 VERY SIMILAR TO SAMPLE J3. THE IDENTIFIED COMPOUNDS ARE NOT THE
ONLY COMPOUNDS PRESENT IN A GIVEN FRACTION.
Figure A-6. Results of Level I organic analysis at Montour.
70
-------
3.4 PARTICLE SIZE
Particle Size at Inlet to ESP
The differential size distributions calculated from Brink impactor data
taken on the inlet side of the Montour ESP are presented in Figures A-7
through A-10. Figure A-10 is an average of the five days of testing, and the
other figures are means for the indicated days. The 90 percent confidence
limits for the data are also presented on these Figures. The 90 percent
confidence range is small for the combined five days' data indicating
consistency over a number of tests. Coal type and boiler variables have
little apparent effect on the inlet particle size distributions.
Outlet Particle Size Distributions
The particulate data from the outlet side of the ESP, taken with an
Andersen impactor, are presented in Figures A-ll through -13. The 90 percent
confidence limits for the data have also been included on these figures.
The scatter of the outlet data was considerable both within a given data
set and from day to day. The size of the confidence ranges is an indica-
tion of the scatter within a data set, and Figure A-14 shows the day-to-day
variation.
Fractional Efficiency
Estimates of fractional removal efficiency on each of the five days
are presented in Figure A-15. The highest efficiency of removal was achieved
on one day of high sulfur coal. Contrary to expectations, the low sulfur
coal with water injection was controlled to nearly the same efficiency as
the high sulfur coal, and considerably better than the conditioned ash
was controlled. This result is inconsistent with the lowered flyash resis-
tivity of the conditioned ash, which one would expect to lead to improved
collection.
3.5 FLYASH RESISTIVITY
The flyash resistivity data taken at Montour is tabulated in Table A-8
and presented as a function of temperature and firing condition in Figure A-
16. The data are consistent for resistivity measurements, even when ignoring
the effect of temperature. Standard deviations and coefficients of variation
for the data are also presented in Table A-8. Based on this data, the LPA
71
-------
K)
e
o»
e
o>
o
I03
10'
O.I
—I 1—I I I I 111 1 1—I—I I I 11
MEAN, WITH 90% CONFIDENCE LIMITS
£ DECEMBER 10
^DECEMBER II
ASSUMED PARTICLE DENSITY = 2.27 g/cc
J L
10
10
"i—i—r
A
rq
J L
100
GEOMETRIC MEAN PARTICLE DIAMETER
Figure A-7. Differential particle size distribution at inlet to ESP, Montour
Plant, December 10 and 11, 1975; low sulfur coal, conditioner
injected at 95 1/hr.
72
-------
I05
I04
CO
o»
E
o«
_o
< I03
o
itf
0.1
I \ | ! I | I | I - 1 - 1 - 1 |
MEAN, WITH 90% CONFIDENCE LIMITS
DECEMBER 16
1—I I I I I LI
DECEMBER 17
ASSUMED PARTICLE DENSITY = 2.27 g/cc
J L
X
1.0 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Fiaure A-8 Differential particle size distribution at inlet to ESP,
' Montour Plant, December 16 and 17, 1975; high sulfur coal, no
conditioner.
73
-------
I05
10
E
z
•V
o>
E
o>
o
10 3
10'
I I 1 I II 111 I
MEAN WITH 90%
CONFIDENCE LIMITS
DECEMBER 18,1975
_ ASSUMED PARTICLE DENSITY 2.27g/cc
1 I I I I I I
0.1
E
E
J 1 1
J L
1.0 10
GEOMETRIC MEAN RETICLE DIAMETER
100
Figure A-9. Differential particle size distribution at inlet to ESP, Montour
Plant, December 18, 1975; low sulfur coal, water injected at
95 1/hr.
74
-------
in
E
z
o»
0
o»
0
io-
i r i i i i ii i n i i i i i i i
MEAN, WITH 90%
CONFIDENCE LIMITS
ASSUMED PARTICLE DENSITY = 2.27 g/cc
10
I0
10
0.1
E
i
1.0 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Figure A-10. Mean differential particle size distribution at inlet to ESP,
Montour Plant, December 10, 11, 16, 17, and 18, 1975.
75
-------
10;
to
IO
e
z
o>
o>
o
< IO1 b
O
O.I
§
1 1 I 1 1 L
MEAN WITH 90% CONFIDENCE ~
LIMITS
E
3
DECEMBER 10
DECEMBER II
ASSUMED PARTICLE
DENSITY - 2.27 g/cc
i i i i i
1.0
GEOMETRIC MEAN PARTICLE DIAMETER
100
Figure A-ll. Differential particle size distributions at outlet of ESP, Mon-
tour Plant, December 10 and 11, 1975.
76
-------
Z
o«
o
I I I 1 I I I II 1 1—TT
MEAN WITH 90% CONFIDENCE LIMITS
| DECEMBER 16, 1975
IDECEMBER IT, 1975
I
i 1
I
0.1 1-0 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Figure A-12. Differential particle size distributions at outlet of ESP, Mon-
tour, December 16 and 17, 1975.
77
-------
IU
I02
"I
z
o>
E.
Q
o> ,
o I01
i
I
• 1 I"1 1 1 1 1 | 1 II
MEAN
1 1 Mill T 1 1 — 1 1 M L
, WITH 90% CONFIDENCE LIMITS -
~ DECEMBER 18, 1975
ASSUMED PARTICLE DENSITY" 2.27g/cc
-
1
0
m
\ \
0
1 1
B
. E
i i i i ,,1, ,i
»
1
B =
—
E :
i i ' i-J...i i i i i i i i i
O.I
1.0 10
GEOMETRIC MEAN PARTICLE DIAMETER(
100
Figure A-13. Differential particle size distributions at outlet of ESP Mon-
tour Plant, December 18, 1975.
78
-------
10'
IO
E
Z
o>
e.
Q
o>
O
0,1
Low sulfur, _
LPA 402A -
injected
xs \_ High sulfur ~
^•_ ™ itnr-fmHitinnoH
\
\
unconditioned
h
NA Low sulfur
water
injected
\
LEGEND
o DECEMBER 10
• DECEMBER II
D DECEMBER 16
• DECEMBER 17
A DECEMBER 18
ASSUMED PARTICLE DENSITY = 2.27 g/bc
i i i i i I l i i i i I l I I I I i
\
\
High sulfur
unconditioned
J I I I_L
1.0 10
GEOMETRIC MEAN PARTICLE DIAMETER
100
Figure A-14. Differential particle size distributions at outlet of ESP, Mon-
tour, December 10, 11, 16, 17, and 18, 1975.
79
-------
oo
o
o.oi
O.I
I
2
5
10
2P
40
. 50
O 60
UJ
o
ec
80
UJ
UJ 90
a.
95
98
99
99.9
99.99
i—r
0.1
fir*
HIGH SULFUR,
NO INJECTION
LOW SULFUR A.I__A._ir #',
WATER INJECTED-* & ""
^---a
HIGH SULFUR X
NO INJECTION-*0
LOW SULFUR
LPA 4O2A INJECTED
O DECEMBER 10, 1975
• DECEMBER 11,1975
D DECEMBER 16,1975
• DECEMBER 17,1975
A DECEMBER 18,1975
I I II | | 99.9
99.9
99
98
95
90 m
~n
80 g
70 2
60 Q
50"
40 S
30m
20 5
10
5
2
I
O.I
J I
I | I I I
00.1
1.0
100
PARTICLE DIAMETER,
Figure A-15. Fractional removal efficiency of Montour ESP.
-------
CO
10"
E
>0
a
LPA - 402A= 95 l/hr
LOW SULFUR COAL, 750 mw AV6.
H20 = 95 l/hr
LOW SULFUR COAL,750mw
NO INJECTION
HIGH SULFUR COAL,755 mw
10
280
285
290 295 300
TEMPERATURE, °F
305
310
Figure A-16. Flyash resistivity measurements, Montour Power Plant.
-------
TABLE A-8. FLYASH RESISTIVITY
Test
Temp, °F
DATE
LOW SULFUR COAL,
1A
IB
1C
2A
2B
2C
308
300
304
300
295
295
12/10/75
12/10/75
12/10/75
12/11/75
12/11/75
12/11/75
TIME
LPA 402A
1000
1200
1430
0915
1110
1400
CONDITION
INJECTION AT 95
732 MW
635 MW
730 MW
735 MW
680 MW
710 MW
703 MW
RESISTIVITY
1/hr
5.3 x 1010
4.0 x 1010
5.1 x 1010
4.0 x 1010
3.1 x 1010
2.7 x 1010
4.0 x 1010
1.05 x 1010
26% .
ft-cm
= X"
= 0
= cv
HIGH SULFUR COAL, NO INJECTION
3A
3B
3C
4A
4B
5A
5B
284
282
285
281
285
LOW
300
300
12/16/75
12/16/75
12/16/75
12/17/75
12/17/75
SULFUR COAL
12/18/75
12/19/75
0830
1030
1300
0830
1600
, WATER
1500
1645
755 MW
755 MW
755 MW
755 MW
755 MW
755 MW
INJECTION AT 95 1
750 MW
750 MW
750 MW
1.8X1010
1.4 xlO10
1,5 x 1010
1.7 xlO10
1.4 xlO10
1.5x 1010
0.18 x 1010
12%
/hr
10 x 1010
9.8 x 1010
9.9 x 1010
0.14 x 1010
1.4%
= X
= a
= CV
= x"
= 0
= cv
X = Mean
o = Standard Deviation
CV = Coefficient of Variation = (a x 100%)/I
82
-------
402A injection lowered the low sulfur coal flyash resistivity by about
60 percent as compared to water injection.
Lowered flyash resistivity leads to expectations of improved precipita-
tor performance, but this was not the case at Montour. An empirical explana-
tion may be that the low sulfur, unconditioned ash collection data was
influenced by the high sulfur coal run the day before. This "inertia" effect
has been observed at other test sites. Some indication of its presence
may be found in the significant difference between the collection efficiencies
on the two days of high sulfur coal. Boiler variables were fairly constant,
as was flyash resistivity, but collection improved on the second day.
3.6 BOILER OPERATION PARAMETERS
The available information on boiler operation during the Montour
test is presented in Table A-9. Significant load variations occurred while
testing was going on only at 1200-1400 on December 10. The measurements
which were in progress at about this time were outlet sulfur oxides, outlet
ammonia, organics, particulates, and flyash resistivity.
83
-------
TABLE A-9. BOILER OPERATION PARAMETERS, MONTOUR TEST
•03
.fa.
12/10/75
LOAD
Gas Outlet Temp
(°F)
Oxygen Cone. (%)
12/11/75
LOAD
Gas Outlet Temp
(°F)
Oxygen Cone. (%)
12/16/75
LOAD
Gas Outlet Temp
(°F)
Oxygen Cone. (%)
12/17/75
LOAD
Gas Outlet Temp
(°F)
Oxygen Cone. {%)
12/18/75
LOAD
Gas Outlet Temp
(°F)
Oxygen Cone. (%)
(MW)
A
B
(MW)
A
B
(MW)
A
B
(MW)
A
B
(MW)
A
B
0200
749
298
286
5.3
468
259
248
5.2
488
268
251
5.1
756
289
285
5.1
750
292
288
5.0
0400
752
299
290
5.2
471
257
250
5.5
489
264
259
5.2
—
288
285
5.2
746
294
288
5.0
0600
750
299
291
5.2
710
277
278
5.0
570
272
270
5.4
271
268
5.2
750
292
289
5.3
0800
722
300
287
5.2
735
286
288
5.2
746
294
292
5.0
728
269
272
5.2
746
290
288
5.3
1000
720
295
294
5.2
735
287
290
5.1
742
297
294
5.4
735
276
276
5.4
745
289
288
5.2
1200
642
289
290
5.1
710
284
290
5.1
748
300
298
5.3
739
278
278
5.1
749
291
286
5.1
1400
577
285
275
5.7
700
283
290
4.9
745
300
299
5.6
729
279
280
5.3
743
288
292
5.2
1600
725
287
291
5.1
712
289
289
5.0
756
298
290
5.6
724
286
281
5.3
746
288
290
5.1
1800
734
287
292
5.2
748
291
292
5.2
755
299
288
5.4
759
301
289
5.1
737
284
290
5.2
2000
737
284
295
5.1
751
292
293
5.2
755
289
286
5.5
758
298
290
5.0
740
285
290
5.0
2200
706
283
292
5.3
737
290
292
5.0
756
290
286
5.5
758
297
289
5.0
760
286
290
5.0
2400
647
276
280
5.1
751
289
288
5.0
745
289
286
5.1
754
295
289
5.0
750
285
290
5.0
-------
SECTION A-4
REFERENCES
1. Alyea, H.J. and H.N.A. Backstrom. The Inhibitive Action of Alcohols
on the Oxidation of Sodium Sulfite. J. Amer. Chem. Soc., 51: 90-109,
1929.
2. Flint, D. Determination of Small Concentrations of $03 in the Presence
of Larger Concentrations of S02- J. Soc. Chem. Ind., 67: 2-5, 1948.
3. Shell Development Company, Analytical Department. "Determination of
Sulfur Dioxide and Sulfur Trioxide in Stack Gases," Emeryville Method
Series, 4516/59A.
4. Matty, R.E. and E.K. Diehl. "Measuring Flue-Gas S0? and SOV" Power,
101, 94-97, November 1957.
5. "Improved Chemical Methods for Sampling and Analysis of Gaseous
Pollutants From the Combustion of Fossil Fuels, Volume I, Sulfur Oxides,"
PB 209-267, APTD 1106, Walden Research, June 1971.
6. Seidman, E.B. "Determination of Sulfur Oxides in Stack Gases." Anal.
Chem., 30: 1680-2, 1958.
7. Fritz, J.S. and S.S. Yamamura. "Rapid Microtitration of Sulfate."
Anal. Chem., 27: 1461-4, 1955.
8. Source Sampling, prepared by Institute for Air Pollution Training,
Office of Manpower Development, EPA.
9. P.W. Jones, et al. "Technical Manual for Analysis of Organic Materials
in Process Streams." PB 259-299/AS, EPA-600/2-76-072, Battelle, March 1976,
10. Harris, D.B. "Tentative Procedures for Particle Sizing in Process
Streams, Cascade Impactors." PB 250-375/AS, EPA-600/2-76-023, February
1976.
11. Nichols, G.B. "Techniques for Measuring Fly Ash Resistivity," PB 244-140/
AS, EPA-650/2-74-079, Southern Research Institute, August 1974.
12. Salzberg, H.W., J.I. Morrow, S.R. Cohen, M.E. Green. Physical Chemistry.
Academic Press, Inc. New York, 1969, pp 14-5.
85
-------
APPENDIX A-A
April 15, 1976
llBaiteiie
Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Telephone (614) 424-6424
Telex 24-5454
Dr. Larry D. Johnson
Process Measurement Branch
Industrial Energy Research Laboratory
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina
27711
Dear Larry:
Analysis of Monteurville Power Plant Effluent
(Contract No. 68-02-1409, Task 42)
Five organic vapor samples of Monteurville power plant effluent were collected
using Battelle Adsorbent Samplers. No filters or probe rinses were provided.
The sample description and assigned sample numbers are given below
Sample
#1 Monteurville power plant effluent
#2 Monteurville power plant effluent
#3 Monteurville power plant effluent
#4 Monteurville power plant effluent
(Unlabelled sampler, contained blue liquid)
BCL Sample Number
Jl
J2
J3
J4
BW1
Level I Analyses
Level I analyses on the five samples were carried out according to the methods
defined in 'Technical Manual on Measurement of Organic Materials in Process
Streams* prepared for EPA by BCL. The weight of material in each fraction is
given in the following table.
Sample
Jl
J2
J3
J4
BW1
Fraction Weight (g)
1
0.0041
0.0024
0.00023
0.00018
0.0014
0.
0.
0.
0.
0.
2
0011
00044
00018
000077
0038
3
0.00054
0.00054
0.00023
0.00013
0.00047
0
0
0
0
0
4
.0068
.0091
.0060
.0039
.0010
0.
0.
0.
0.
0.
5
00082
00084
00018
00021
00047
0
0
0
0
0
6
.0057
.0065
.00048
.00027
.0077
7
0.0014
0.00009
0.00004
0.00003
0.00001
8
0.00006
0.00003
0.00008
0.00001
0.000005
Infrared analysis of the 40 individual fractions was carried out by Fourier
Transform infrared spectroscopy (FTIR); the results are shown in the attached
Table of Level I analytical data. The original FTIR spectra will be supplied
upon request.
86
-------
Dr. Larry D. Johnson 2 April 15, 1976
Nitrosamine Analysis
This Task specifically requested quantitative Level II analysis for nitrosamines
in the collected samples. Following discussion with EPA personnel, it was agreed
that the nitrosamine of concern was diethylnitrosamine, in view of the amine
additives used in this power plant.
Analysis was carried out by GC-MS specific ion current integration, in a similar
manner to that successfully employed for POM analyses in these laboratories.
Aniline was selected as an appropriate internal standard, in view of its similar
volatility to the nitrosamine (boiling points 184 and 177 C, respectively). Gas
chromatographic separation of diethylnitrosamine and this internal standard were
satisfactorily achieved on a 6-foot Silar 10CP column. Mass spectrometric
analysis was accomplished by methane chemical ionization, using a Finnigan 3200
quadrupole mass spectrometer with an interfaced PDP8 computer for data handling.
The analytical technique is illustrated by the analysis of a standard mixture
containing 1000 ng each of the nitrosamine and the internal standard, as shown
in the attached reconstructed gas chromatogram (RGC). In this attached RGC,
the black trace depicts the total ion current for the GC-MS analysis of a
mixture of nitrosamine and standard. The red trace is an ion overlay for the
ion m/e = 103 (nitrosamine) and the green trace is the ion overlay for the ion
m/e = 94 (standard). The two attached mass spectra (22-18) and (90-83) are the
methane chemical ionization mass spectra for diethylnitrosamine and aniline,
respectively; the ions groups at m/e = 101, 102, 103, 104 and 92, 93, 94,
95 were used for the quantitative analysis, as described in the 'Technical
Manual for Measurement of Organic Materials in Process Streams' (prepared for
EPA by BCL). It is clearly evident that highly selective nitrosamine analysis
is provided by this technique.
Prior to actual sample analysis, extensive studies were carried out to determine
the optimum method for concentrating the pentane extracts of the Adsorbent
Samplers, and also to determine the .typical nitrosamine recoveries and quanti-
tation calibration factor. The entire pentane extract of each sampler (about
120 ml) was subsequently concentrated by Kuderna-Danish evaporation to about
100 ul; the average nitrosamine recoveries were previously estimated to be
greater than 80% by this technique. Calibration factors for quantitation (the
ratio of specific ion current response for nitrosamine and standard) determined
over the 1000 to 10,000 ng range, showed no significant variation, and thus, a
mean value was chosen for the subsequent analysis.
The detection limit for diethylnitrosamine was shown to lie between 1 and 10 ng,
by the analytical procedure-described above; thus a reasonable detection limit
of 5 ng was assumed.
Quantitative analysis of the five samples Jl, J2, J3, J4, and BW1 failed to
detect the presence of diethylnitrosamine in any sample. Thus we must assume
that any diethylnitrosamine present was at levels lower than 5 ng in each sample.
This laboratory was not supplied with details relating to the volume of stack
gas sampled, but assuming a usual volume of 30 cu ft for the Battelle Adsorbent
Samplers, this means that the upper limit for diethylnitrosamine content of the
Monteurville power plant stack gas effluent was 5 parts per trillion (ppt).
87
-------
Dr. Larry Johnson 3 April 15, 1976
Conclusion to Level II Nitrosamine Analysis
This program has clearly demonstrated the utility of the BCL-developed ion
current integration techniques for the analysis of hazardous species such as
nitrosamines in combustion effluents. In this instance diethylnitrosamine was
not detected, but its upper limit in combustion effluent was reasonably esti-
mated to be 5 ppt. Use of the developing EPA SASS train would lower this
detection limit to about 0.2 ppt, on account of the larger gas volume which
may be sampled by this system.
If you have any comments or questions with regard to the above results, please
do not hesitate to contact me at Extension 1158, or Paul Strup at Extension 1710.
Sincere regards,
Peter W. Jones
Associate Manager
Organic and Structural
Chemistry Section
PWJ:pb
Enclosures
88
-------
FRACTION
SAMPLE
J 1 ,
J2 '
J3
J4
BW 1
1
AROMAT
$%Sty
2
IC^HYDROCARBOfS
t
SILICONE
ATIC
HYDROCARBON
1 ESTER,
• f
SILICONE (TRACE)
"
1
AROMATIC HC
MffiKMHRMnOBBBBMBBBBE!
(TRACE)
SILICONE ^
^ JSjUgONE
(TRACE)
AROMATIC
HYDROCARBON
>
AF
3
1 * *
ALIPHATIC
ESTER
4
5
AROMATIC*2' <
KETONE OR
QUINO
•
^f ALIPHAT
KETONE
AROMATIC*2' KETONE
TRACE
OR QUINONE
ALIP
KETC
AROMATIC KETONE*2'/OUINONE
>
2nd AROM
KETONE/QUINONE
,
AROMATIC KETONE*2'/OUINONE ^
SOMATIC KETONE*
2'/QUINONE
2nd AROW
6
CARBOXYLIC
ACID
1
'1C
(S)
4
CARBOXYLIC
ACID
i ATIC
>NE(S)
ALIPHATIC
^ KETONE
ALIPHATIC „
HHHMBHHHHfiUBBHEHHHBSu
KETONE
KETONE/QUINONE
A PHEN91-'4'
SUBSTITUTED)
k 2nd*5' PHENOL
7
>
CARBOXYLIC
ACID (SOAPSJ '
ALCOHOL*3' ,
|M^gKMUHMWMKB««H«eBIH«H
^^^^^^^M^^^B^^^^^^^^M^^^HB
, CARBOXYLIC
ACID SALT
(SOAP) ^
)
/
', )
\
/
j
\
(
(1) AROMATIC HYDROCARBON, PROBABLY MULTIPLE OR FUSED RING
(2) SAME AROMATIC KETONE/QUINONE RUNS THROUGH ALL OF SERIES OF 5 SAMPLES
(3) ALIPHATIC ALCOHOL, POLYALCOHOL OR ALKANOL AMINE
(4) A PHENOLIC COMPOUND, PROBABLY PHENYL SUBSTITUTED
(5) PROBABLY A SECOND PHENOL, HIGHLY SUBSTITUTED, POSSIBLY POLYHYDROXY COMPOUND
LEVEL I ORGANIC ANALYSIS OF MONTEURVILLE POWER PLANT EFFLUENT
(CONTRACT 68-02-1409, TASK 42)
-------
N-STTD
o
0 10203010SS607D803(3
SPECTHUH NUMBER
1DC5 110 120 130
-------
SPECTRUM 22-18 DVeTHNLNn"«oSf\V\MM t
N-STD
jilifjcn.
M-H
70 80
u/*r
163 110 120
It® 1S0
91
-------
SPECTRUM 20 -
tt-STD
ftNtVMMe
€3 _J
70 80 30
we
:»., , .It - -
Tllll If*V>Mfi*|i'K*lllf
100 110 120 130 HO
92
-------
TABLE A-A-1. DIGEST OF DATA SHEETS FROM TENAX ORGANIC RUNS. MONTOUR STEAM POWER PLANT, DANVILLE, PA.
<£>
CO
Orifice Pump Stack Heater Avg. Meti
Run Number AP AP Temp. Box T. Temp.
Field 1 Battelle Date Time ("Hg) ("Hg) (°F) (°F) (°F)
1 Jl 12/10/75 1430-1645 1.25 7.5 240 -- 94
2 J2 12/11/75 0940-1220 1.25 7.5 230 — 93
3 03 12/16/75 1000-1222 1.25 7.8 260 250 71
4 J4 12/16/75 1420-1635 1.25 10.0 260 250- 80
300
5 BW1 12/18/75 1515-1730 1.25 10.0 240 250- 75
300
2r Sample
Volume Condenser
(liters/set) Number
1511.7/53.4 1
1491.6/52.7 20
1555.2/54.9 19
1338.6/47.3 18
1389.4/49.1 15
NOTES:
Run #2
(1) Heater box went out for 15 minutes during run.
(2) Water condensing at outlet of Tenax condenser causing pressure drop.
Run #3
(1) Water condensation at outlet of condenser.
-------
APPENDIX A-B
Soutnern Research Institute
2OOO NINTH AVENUE SOUTH
BIRMINGHAM.ALABAMA 352O5
TELEPHONE 205-323-6592
February 19, 1976
Mr. D. Bruce Harris
Project Officer
Environmental Protection Agency
Industrial Environmental Research
Laboratory
Mail Stop MD-62
Research Triangle Park, NC 27711
Dear Bruce:
Here are the reduced data from Danville. We have included
a description of procedures used to handle the blank weights,
etc. If you need anything else, such as the computer printouts,
please let me know.
Sincerely,
Wallace B. Smith
Head, Physics Section
WBSrmlm
Enclosures
94
-------
Danville Tests and Impactor Data
Impactor data, both inlet and outlet, were handled according
to the best professional skill of SRI personnel. Nevertheless,
data taken on the following days' runs have been modified or de-
leted for the reasons given below. Also, particulate stage weights
were compensated for background by subtracting stage weights obtained
from, the blank.runs.
Outlet
11 December 75 - 1st run. Stage 8 deleted due to negative weight.
18 December 75,- 2nd run. Deleted completely due to extreme over-
loading of impactor.
Inlet
All stage 6 data were deleted due to the unreliability of the
weights gathered (most lost weight).
9 December 75(2 runs) and 10 December 75(1st run) - deleted completely
due to high flow rates and unreliability of numbers.
10 December 75 - 2nd run - Stage 5 deleted due to negative weight.
11 December 75 - 2nd run - Stages 1, 3, and filter deleted due to
unreliability of weight or negative weight.
16 December 75 - 1st run - Stages 3, and 5 deleted due to negative
weight.
16 December 75 - 2nd run - Stage 5 deleted due to negative weight.
16 December 75 - 3rd run - Stages 4, and 5 deleted due to negative
weight.
95
-------
-2-
Danville Tests-
18 December 75 - 2nd run - Stage 5 deleted due to negative weight.
18 December 75 - 5th run - Stage 5 deleted due to unreliability of
number.
18 December 75 - 6th run - Deleted completely due to nozzle pointing
downstream.
96
-------
Table of Blank Data
Blank Runs
Date 12/10/75 12/11/75 12/16/75 12/17/75 12/18/75
SO
SI
S2
S3
S4
S5
S6
S7
S8
SF
-0.42 mg
-0.08
-0.57
-0.09
-0.30
-0.64
-0.55 "
-0.09
-1.02
-1.89 mg
-1.05
-0.94
-0.77
-0.86
0.91
-1.14
-0.65
-0.36
3.12 mg
0.55
0.74
0.68
0.13
0.13
0.07
0.32
0.28
4.19 mg*
0.25
0.16
0.01
-0.10
0.14
-0.19
0.06
0.37
0.45 mg
0.43
0.49
0.38
0.36
0.33
0.32
0.39
0.70
x=0.36
x=0.12
w/o nega-
tives
x=0.43
The blanks for 10 December and 11 December had negative
weight gains because of severe sticking of the substrates.
Thus, no modification of the particulate-gathering runs was
made. Otherwise, Stages SI thru S8 were averaged (x shown)
and these were subtracted from particulate-gethering runs for
that day. SO was subtracted from SF for each day. If sub-
traction produced zero or negative weight gains, then zero
was entered.
*O-ring left out. Thus, SI acted as a filter, and this run
was treated as a blank with SI weight ignored.
97
-------
TABLE A-B-1. INLET IMPACTOR DATA, PP&L, HQNTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Brink Impactor
Stage Weights (nig)
CYC
SO
SI
S2
S3
S4
S5
S6b
SF
PPLI -3
12/10/75
29.25
0.6
285
35
0.038
—
5N
0920
c
176.57
9.84
15.58
6.86
1.95
1.00
0.71
—
0.69
PPLI-4
12/10/75
29.25
0.2
285
15
0.022
164
4N
1100
199.92
12.38
7.62
4.85
1.39
0.71
-0.07 a
—
0.24
PPLI-5
12/10/75
29.25
0.9
285
10
0.022
157
6N
1540
128.88
7.42
6.21
3.92
1.03
0.46
0.07
-0.16 a
0.10
PPLI-6
12/10/75
29.25
0.2
285
20
0.022
—
7N
1620
183.43
15.17
8.09
3.08
0.92
0.33
0.24
___
0.10
NOTES:
a Negative stage weights treated as zero weight.
b All stage 6 data deleted as unreliable. Most lost weight.
c Run deleted. High flow rate.
98
-------
TABLE A-B-2. INLET IMPACTOR DATA, PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp. (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Brink Impactor
Stage Weights (mg)
CYC
SO
SI
S2
S3
S4
S5
S6 b
SF
PPLI-7
12/11/75
29.52
0.2
285
10
0.022
164
8N
0950
146.29
9.68
10.71
3.79
1.06
0.37
0.18
—
0.21
PPLI-8
12/11/75
29.52
0.9
285
10
0.022
157
ION
1045
305.54
13.86
7.41 c
2.65
- 0.80 a
0.14
0.17
- 0.14 a
- 0.28 a
PPLI-9
12/11/75
29.52
0.2
285
10
0.022
—
9N
1345
205.68
12.67
13.66
3.98
1.83
0.47
0.25
—
0.43
PPLI-10
12/11/75
29.52
0.2
285
10
0.022
164
UN
1445
182.41
11.46
14.25
4.57
1.81
0.36
0.14
—
0.40
NOTES:
a Negative stage weights treated as zero weight.
b All stage 6 data deleted as unreliable. Most lost weight.
c Deleted as unreliable weight.
99
-------
TABLE A-B-3. INLET IMPACTOR DATA. PP&L, MONTQUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Brink Impactor
Stage Weights (mg)
CYC
SO
SI
S2
S3
S4
S5
S6b
SF
PPLI-11 PPLI-12 PPLI-13 PPLI-14
12/16/75 12/16/75 '
29.52
0.2
285
10
0.022
164
12N
0920
57.0
2.56
2.24
1.09
-0.57a
0.21
-0.49a
—
0.16
29.52
0.9
285
10
0.022
157
13N
1022
109.3
11.36
9.94
3.04
0.34
0.39
-0.13a
-0.08a
0.14
12/16/75 12/16/75
29.52
0.2
285
10
0.022
—
14N
1001
67.5
12.82
7.71
2.70
0.67
-0.08a
0.00
—
0.07
29.52
0.2
285
10
0.022
164
15N
1350
84.6
6.87
7.47
3.16
1.14
0.55
0.13
—
0.30
PPLI-15
12/16/75
29.52
0.2
285
10
0.022
—
16N
1440
73.0
8.39
8.63
2.99
1.11
0.70
0.41
—
0.29
PPLI-16
12/16/75
29.52
0.9
285
10
0.022
157
17N
1520
86.0
10.35
7.21
2.85
1.10
0.48
0.21
0.95
0.24
NOTES:
a Negative stage weights treated as zero weight
b All stage 6 data deleted as unreliable. Most lost weight,
100
-------
TABLE A-B-4. INLET IMPACTOR DATA. PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Brink Impactor
Stage Weights (mg)
CYC
SO
SI
S2
S3
S4
S5
S6a
SF
PPLI-17
12/17/75
29.60
0.9
281
10
0.022
157
18N
0941
56.7
5.97
3.93
2.18
0.82
0.44
0.12
0.07
0.27
PPLI-18
12/17/75
29.60
0.2
281
14
0.022
—
19N
1030
67.5
11.01
9,26
3.64
1.09
0.34
0,18
,-- -
0.32
PPLI-19
12/17/75
29.60
0.2
281
10
0.022
164
20N
1355
64,0
5.04
7,70
2.87
0,87
0,69
0.15
,--~
0,40
NOTES:
a All stage 6 data deleted as unreliable. Most lost weight.
101
-------
TABLE A-B-5. INLET IMPACTOR DATA, PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Brink Impactor
Stage Weights (rug)
CYC
SO
SI
S2
S3
S4
S5
S6b
SF
PPLI-20
12/18/75
29.40
0.9
290
10
0.022
157
21 N
1230
133.4
14.50
12.55
5.97
1.85
0.82
0.30
0.08
0.36
PPLI-21
12/18/75
29.40
0.2
290
10
0.022
164
22N
1315
104.2
10.36
7.75
5.61
8.89
0.45
0.00
—
0.45
PPLI-22
12/18/75
29.40
0.2
290
10
0.022
None
23N
1400
103.0
10.53
10.32
4.93
1.65
0.83
0.25
—
0.23
PPLI-23
12/18/75
29.40
0.2
290
10
0.022
157
24N
1445
108.8
10.07
15.28
3.15
1.62
0.62
0.42
—
0.64
PPLI-24
12/18/75
29.40
0.2
290
10
0.022
164
25N
1530
72.7
9.02
13.26
4.28
1.32
0.36
0.05C
—
1.32
PPLI-25
12/18/75
29.40
0.2
290
10
0.022
None
26N
1615
a
12. la
1.42
0.90
0.46
0.08
0.18
0.04
—
0.15
NOTES:
a Nozzle turned downstream on this run. Run not used in calculations.
b All stage 6 data deleted as unreliable. Most lost weight.
c Deleted as unreliable weight.
102
-------
TABLE A-B-6. OUTLET IMPACTOR DATA, PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Andersen Impactor
Stage Weights (rug)
SI
S2
S3
S4
S5
S6
S7
S8
SF
E
PPLO-1
12/9/75
29.20
0.324
280
60
0.5
507
A- 28
1545
9.27
2.62
3.11
3.19
4.31
6.25
3.57
2.00
2.84
37.16
PPLO-2
12/10/75
29.25
0.324
280
20
0.5
522
A-29
0935
20.72
7.53
8.59
10.67
11.71
9.80
6.00
2.05
8.40
85.47
PPLO-3
12/10/75
29.25
0.324
280
20
0.5
507
A-30
1100
17.57
7.85
10.82
9.61
11.19
10.69
5.07
0.53
3.44
75.77
PPLO-4
12/10/75
29.25
0.324
280
20
0.5
506
A-26
1323
-0.42
-0.08
-0.57
-0.09
-0.30
-0.64
-0.55
-0.09
-1.02
BLANK3
RUN
PPLO-5
12/10/75
29.25
0.324
280
20
0.5
128
A-27
1545
11.45
5.80
5.86
7.61
7.70
6.85
3.69
1.70
5.21
55.87
NOTES:
a Negative weight blank was not used. Severe substrate sticking problems,
103
-------
TABLE A-B-7. OUTLET IMPACTOR DATA, PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Andersen Impactor
Stage Weights (mg)
SI
S2
S3
S4
S5
S6
S7
S8
SF
z
PPLO-6
12/11/75
29.52
0.324
280
20
0.5
522
A-24
0930
13.19
5.03
6.55
11.45
10.05
6.42
7.02
- 0.31a
3.59
58.30
PPLO-7
12/11/75
29.52
0.324
280
20
0.5
128
A-25
1100
12.31
4.76
6.93
8.83
11.50
8.52
4.45
1.80
1.42
60.52
PPLO-8
12/11/75
29.52
0.324
280
20
0.5
506
A-23
1310
-1.05
-0.94
-0.77
-0.86
0.91
-1.14
-0.65
-0.36
-1.89
b
PPLO-9
12/11/75
29.52
0.324
280
20
0.5
507
A-40
1415
4.22
5.04
8.63
8.63
8.27
7.09
3.69
1.25
3.15
50.97
BLANK RUN
PPLO-10
12/11/75
29.52
0.324
280
20
0.5
522
A-41
1530
14.35
7.01
12.94
13.36
11.37
12.30
3.45
1.15
3.19
79.12
NOTES:
a Negative stage weight treated as zero weight.
b Negative weight blank was not used. Severe substrate sticking problems,
104
-------
TABLE A-B-8. .OUTLET IMPACTOR DATA. PP&L. MONTOUR PLANT
Run Code'
Date
Amb. Press.
("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time
(mi n )
Flow Rate
(acfm)
Impactor Ident.
Set Ident.
Start Time
PPLO-11
12/16/75
29.52
0.324
280
30
0.5
522
A-42
0915
PPLO-12
12/16/75
29.52
0.324
280
30
0.5
128
A-43
1045
PPLO-13
12/15/75
29.52
0.324
280
30
0.5
507
A-44
1256
PPLO-14
12/16/75
29.52
0.324
280
30
0.5
506
A-45
1410
PPLO-15
12/16/75
29.52
0.324
280
30
0.5
522
A-46
1545
Andersen Impactor
Stage Weights3
(mg)
SI
S2
S3
S4
S5
S6
S7
S8
SF
£
NOTES:
9.45 (9.09)
4.33 (3.97)
4.40 (4.04)
5.24 (4.88)
5.26 (4.90)
5.69 (5.33)
4.67 (4.31)
4.17 (3.81)
2.90 (0.00)
(40.33)
a Weights in parentheses
15.00 (14.64)
5.15 (4.79)
5.57 (5.21)
6.118 (5.82)
6.85 (6.49)
6.38 (6.02)
5.22 (4.86)
2.10 (1.74)
3.62 (0.50)
(50.07)
are actual wei
9.56 (9.20)
6.36 (6.00)
6.81 (6.45)
7.99 (7.63)
7.17 (6.81)
7.93 (7.57)
3.43 (3.07)
2.08 (1,72)
7.00 (3.88)
(43.13)
ghts reduced
3.12b
0.55
0.74
0.68
0.13
0.13
0.07
0.32
0.28°
I = 0.36d
a = 0.26
BLANK RUN
6.17 (5.81)
4.05 (3.69)
4.82 (4.46)
4.94 (4.58)
5.65 (5.29)
5.76 (5.40)
2.87 (2.51)
1.63 (1.27)
3.08 (0.00)
(33.01)
by mean gain of stages
1 ._•_..!» _ .- 1- T _ „ 1. f 1"
S1-S8 during blank runs. Stage SF weight reduced by weight on blank SF.
b Weight gain of stage SO, which was same as that of final stage (SF),
c Weight on stage S8 in blank run configuration,
d Y represents mean; a represents standard deviation,
105
-------
TABLE A-B-9. OUTLET IHPACTOR DATA. PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Andersen Impactor
Stage Weights a (mg)
SI
S2
S3
S4
S5
S6
S7
S8
SF
E
PPLO-16
12/17/75
29.60
0.324
280
30
0.5
507
A-47
0900
4.19b
0.25
0.16
0.01
-0.10
0.14
-0.19
0.06
0.37
X~ = 0.12C
a = 0.09
BLANK RUN
PPLO-17
12/17/75
29.60
0.324
280
30
0.5
128
A-48
1045
1.74 (1.62)
1.43 (1.31)
1.48 (1.36)
1.61 (1.49)
1.86 (1.74)
1.72 (1.60)
1.02 (0.90)
0.44 (0.32)
1.59 (1.22)
(11.56)
PPLO-18
12/17/75
29.60
0.324
280
30
0.5
507
A-49
1250
1.81 (1.69)
1.21 (1.09)
1.13 (1.01)
1.15 (1.03)
0.75 (0.63)
0.87 (0.75)
1.15 (1.03)
0.61 (0.49)
2.04 (1.67)
PPLO-19
12/17/75
29.60
0.324
280
30
0.5
128
A-50
1405
1.68 (1.56)
1.21 (1.09)
1.60 (1.48)
1.50 (1.38)
1.55 (1.43)
1.95 (1.83)
0.90 (0.78)
1.79 (1.67)
1.45 (1.08)
NOTES:
a Weights in parentheses are actual weights reduced by mean gain of stages
S1-S8 during blank runs. Stage SF weight reduced by weight on blank SF.
b Neglected as spurious point. 0-ring left out and SI served as a filter.
c Mean without including negative weights.
106
-------
TABLE A-B-10. OUTLET IMPACTOR DATA. PP&L, MONTOUR PLANT
Run Code
Date
Amb. Press. ("Hg)
AP ("Hg)
Stack Temp (°F)
Sampling Time (min)
Flow Rate (acfm)
Impactor Ident.
Set Ident.
Start Time
Andersen Impactor
Stage Weightsa (mg)
SI
S2
S3
S4
S5
S6
S7
S8
SF
£
PPLO-20
12/18/75
29.40
0.324
280
20
0.5
128
A-51
1232
5.15 (4.72)
2.21 (1.78)
2.25 (1.82)
2.25 (1.82)
4.23 (3.80)
5.00 (4.57)
3.57 (3.14)
2.33 (1.90)
1.90 (1.45)
(25.00)
PPLO-21
12/18/75
29.40
0.324
280
40
0.5
507
A-52
1420
b
110.75 (110.32)
49.07 (48.64)
82.85 (82.42)
59.96 (56.53)
30.00 (29.57)
18.07 (17.64)
7.59 (7.16)
1.53 (1.10)
2.39 (1.94)
(355.32)
PPLO-22
12/18/75
29.40
0.324
280
20
0.5
?
A-53
1545
4.62 (4.19)
2.24 (1.81)
3.30 (2.87)
2.89 (2.46)
4.26 (3.83)
4.42 (3.99)
1.80 (1.37)
0.88 (0.45)
1.62 (1.17)
(22.14)
PPLO-23
12/18/75
29.40
0.324
280
30
0.5
522
A- 54
1620
0.43
0.49
0.38
0.36
0.33
0.32
0.39
0.70
0.45
X" = 0.43
a = 0.12
BLANK RUN
NOTES:
a Weights in parentheses are actual weights reduced by mean gain of stages
S1-S8 during blank runs. Stage SF weight reduced by weight on blank SF.
b Run deleted due to severe overloading of impactor.
107
-------
APPENDIX B
TABLE B-l. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 8:15 AM EXIT GAS
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 10 December 1975
TEMP: A-2910
Primary
Voltage
310
280
280
290
290
290
270
280
270
290
275
300
320
300
290
300
F; B-296QF
Primary
Current
125
120
130
135
120
120
75
115
90
85
90
130
115
100
75
150
LOAD: 720 MW
EXCESS AIR: 5.3%
Estimate*!
Sparks
Per Win.
60
24
6
10
96
30
36
42
36
36
60
42
66
60
36
12
108
-------
TABLE B-2. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 8:45 PM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 11 December 1975
GAS TEMP: 2A-2910
Primary
Voltage
280-310
270-280
270-280
280-290
280-290
290-300
270-280
270-290
270-280
280-290
230-270
290-300
300-320
290-320
300-310
290-310
F; 2B-291°F
Primary
Current
105-135
110-120
95-130
95-135
100-125
110-115
10-100
75-110
95-105
70-100
0-150
100-150
100-130
75-125
50-125
125-150
LOAD: 749 MW
EXCESS AIR: 5.2%
Estimated
Sparks
Per Min.
120
96
60
48
120
60
72
96
60
60
120
96
120
144
96
48
109
-------
TABLE B-3. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 11 AM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE:
GAS TEMP:
Primary
Voltage
300
232
240
290
295
295
235
230
235
300
235
300
310
300
300
295
11 December 1975
288- 291 °F
Primary
Current
125
122
128
135
122
no
60
85
105
90
55
160
125
105
100
125
LOAD: 738 MW
EXCESS AIR: 5.1%
Estimated
Sparks
Per Min.
90
60
72
60
120
90
60
72
48
66
72
60
90
90
80
48
110
-------
TABLE B-4. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 5:11 PM EXIT GAS
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 16 December 1975
TEMP: A-301Q
Primary
Voltage
310-340
280-300
310
290-310
300-330
270-300
280-300
280-300
290-310
280-290
310-330
300-310
300-350
290-310
290-310
290-300
Fj B-2889F
Primary
Current
75-125
75-100
130
115-140
100-130
75-100
75-125
115-135
85-115
105-110
150-225
215-225
75-105
50-100
75-100
125-150
LOAD: 751 MW
EXCESS AIR: 5.0%
Estimated
Sparks
Per Min.
36
60
0
6
16
48
21
36
12
3
24
24
40
72
60
16
111
-------
TABLE B-5. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 7:15 PM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 16 December 1975
GAS TEMP: A-289°F
Primary
Voltage
290-320
290-300
290-310
290-310
300-320
290-310
290-310
290-310
280-300
270-280
280-320
300-320
300-330
290-320
290-310
300-310
; B-286°F
Primary
Current
50-100
50-100
105-135
115-140
75-120
80-110
50-125
100-125
85-110
90-100
50-200
215-230
50-125
50-100
50-200
150-165
LOAD: 747 MW
EXCESS AIR: 5.4%
Estimated
Sparks
Per Min,
64
72
8
6
40
54
24
22
12
4
24
20
75
64
12
6
112
-------
TABLE B~6. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 11:35PM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 16 December 1975
GAS TEMP: A-2890
Primary
Voltage
290-310
270-280
290-300
300-310
310-320
300-310
280-300
290-300
300-310
290-300
300-330
310-320
310-340
310-320
280-320
310-315
F; B-285°F
Primary
Current
50-65
50-65
90-115
135-145
95-110
90-100
60-105
105-120
80-95
95-107
160-190
220-240
75-100
55-90
50-125
170-185
LOAD: 747 MW
EXCESS AIR: 5.3%
s
Estimated
Sparks
Per Min.
60
48
36
6
42
72
48
24
36
6
60
6
60
60
36
6
113
-------
TABLE B-7. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 12: 30 AM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 18 December 1975
GAS TEMP: A- 295°
Primary
Voltage
290-320
280-300
300-310
270-290
310-320
305-315
290-300
290-300
290-310
300-310
310-320
300-310
300-320
290-310
280-300
280-290
Fj B-289°F
Primary
Current
55-75
70-90
85-125
60-105
100-120
110-120
95-120
120-125
70-95
105-115
130-170
215-225
70-85
55-80
55-100
105-120
LOAD: 746 MW
EXCESS AIR: 5.0%
Estimated
Sparks
Per Min.
72
60
24
30
36
30
36
48
36
12
60
36
72
72
60
48
114
-------
TABLE B-8. PRECIPITATOR VOLTAGE - CURRENT DATA
UNIT: Montour #2
TIME: 4:40 PM EXIT
Control #
5A/6A
5B/6B
5C/6C
5D/6D
7A/8A
7B/8B
7C/8C
7D/8D
1A/2A
1B/2B
1C/2C
1D/2D
3A/4A
3B/4B
3C/4C
3D/4D
DATE: 18 December 1975
GAS TEMP: A-2860
Primary
Voltage
250
280
290
295
280
290
250
280
220
280
285
300
280
260
280
300
F; B-290°F
Primary
Current
55
85
105
125
95
90
50
120
85
100
90
200
100
50
50
115
LOAD: 754 MW
EXCESS AIR: 5.0%
Estimated
Sparks
Per Min.
60
72
24
12
66
78
30
30
20
12
48
60
72
72
72
20
115
-------
APPENDIX C
TABLE C-l. VOLTAGE VS CURRENT DATA
DATE: 16
TIME: 6:1
Primary
Current
Amps
85
75
50
25
December 1975
5 PM
Primary
Voltage
Volts
280
275
250
222
LOCATION:
CONDITION:
Primary
Power
kW
23.8
20.6
12.5
5.6
Montour
High Sulfur
Secondary
Current
Amps
0.371
0.329
0.22
0.105
POWER SET:
1A/2A
Coal - No Injection
Secondary
Voltage
kV
38.54
37.6
33.84
31.96
Current
Density
nA/cm2
14
12
8.2
3.9
TABLE
C-2. VOLTAGE VS CURRENT
DATA
DATE: 16
December 1975
TIME: Not Available
Primary
Current
Amps
no
100
87.5
75
50
25
Primary
Voltage
Volts
285
280
275
260
250
200
LOCATION:
CONDITION:
Primary
Power
kW
31.35
28.0
24.0
19.5
12.5
5.0
Montour
High Sulfur
Secondary
Current
Amps
0.480
0.45
0.39
0.34
0.23
0.11
POWER SET:
1B/2B
Coal - No Injection
Secondary
Voltage
kV
38.5
37.6
36.7
34.8
32.9
27.3
Current
Density
nA/cm2
18
17
15
13
8.6
4.1
116
-------
TABLE C-3. VOLTAGE VS CURRENT DATA
DATE: 16
TIME: Not
Primary
Current
Amps
225
125
100
75
50
December 1975
Available
Primary
Voltage
Volts
250
300
285
275
250
LOCATION:
CONDITION:
Primary
Power
kW
56.25
37.5
28.5
20.62
12.5
Montour
High Sulfur
Secondary
Current
Amps
0.651
POWER SET:
1C/2C
Coal - No Injection
Secondary
Voltage
kV
58,0
17.8
16.0
Current
Density
nA/cm
24.3
TABLE
C-4. VOLTAGE VS CURRENT
DATA
DATE: 16
TIME: Not
Primary
Current
Amps
225
200
175
150
125
100
75
50
December 1975
Available
Primary
Voltage
Volts
300
300
290
280
265
250
235
210
LOCATION:
CONDITION:
Primary
Power
kW
67.5
60.0
50.75
42.0
33.1
25.0
17.62
10.5
Montour
High Sulfur
Secondary
Current
Amps
1.05
0.91
0.81
0.69
0.57
0.46
0.34
0.28
POWER SET:
1D/2D
Coal - No Injection
Secondary
Voltage
kV
38.54
39.48
37.6
36.7
34.78
32.9
31.02
22.56
Current
Density
nA/cm
39.2
34.0
30.3
25.8
21.3
17.2
12.7
10.5
117
-------
TABLE C-5. VOLTAGE VS CURRENT DATA
DATE: 18
TIME: 12:
Primary
Current
Amps
165
85
70
60
45
December 1975
15 PM
Primary
Voltage
Volts
280
270
260
250
240
LOCATION:
CONDITION:
Primary
Power
kW
46
23
18.2
15
10.8
Montour
POWER SET:
1A/2A
Low Sulfur + 95 1/hr LPA 402A + H20
Secondary
Current
Amps
0.87
0.36
0.30
0.257
Secondary
Voltage
kV
32
37.6
35.7
34.8
33.8
Current
Density
nA/cm2
32
13.5
11.3
9.6
7.1
TABLE C-6. VOLTAGE VS CURRENT
DATA
DATE: 18
TIME: 12:
Primary
Current
Amps
170
105
90
75
65
55
45
December 1975
00 N
Primary
Voltage
Volts
270-300
280
270
260
250
240
230
LOCATION:
CONDITION:
Primary
Power
kW
48.5
29.4
24.3
19.5
11.3
13.2
10.4
Montour
Low Sulfur
Secondary
Current
Amps
0.884
0.458
0.388
0.328
0.28
0.234
0,194
POWER SET:
Coal + 95 1/hr
Secondary
Voltage
kV
32.9
38.5
37.6
35.7
34.8
33.8
32.0
1B/2B
LPA 402A + H20
Current
Dens i ty
2
nA/cm
33.0
17.1
14.5
12.2
10.5
8.76
7.25
118
-------
TABLE C-7. VOLTAGE VS CURRENT DATA
DATE: 18
TIME: 11:
Primary
Current
Amps
240
130
125
80
75
60
45
December 1975
30 AM
Primary
Voltage
Volts
290-300
280-290
280
270
260
250
240
LOCATION:
CONDITION:
Primary
Power
kW
70.8
37.1
35
21.6
19.5
15.0
10.8
Montour
Low Sulfur +
Secondary
Current
Amps
1.33
0.577
0.559
0.363
0.336
0.374
0.203
POWER SET
95 1/hr LPA
Secondary
Voltage
kV
32
38.5
37.6
35.7
34.8
32.9
32.0
: 1C/2C
402A + H20
Current
Density
2
nA/cm
49.6
21.6
20.9
13.6
12.6
10.2
7.57
TABLE
C-8. VOLTAGE VS CURRENT
DATA
DATE: 18
TIME: 11:
Primary
Current
Amps
230
170
140
125
105
90
75
60
45
December 1975
00 AM
Primary
Voltage
Volts
300
280
270
260
250
240
230
220
210
LOCATION:
CONDITION:
Primary
Power
kW
69
47.6
37.8
32.5
26.3
21.6
17.3
13.2
9.45
Montour
Low Sulfur +
Secondary
Current
Amps
1.1
1.27
1.06
0.561
0.465
0.394
0.324
0.255
0.195
POWER SET
95 1/hr LPA
Secondary
Voltage
kV
37.6
37.6
35.7
34.8
33.8
32.9
32
31
29.1
: 1D/2D
402A + H20
Current
Density
nA/cm2
41.2
47.5
39.6
21.0
17.4
14.7
12.1
9.54
7.27
119
-------
TABLE C-9. VOLTAGE VS CURRENT DATA
DATE: 18 December 1975
TIME: 1:30
Primary
Current
Amps
70
105
85
70
60
50
PM
Primary
Voltage
Volts
260-300
280
270
260
250
240
LOCATION:
CONDITION:
Primary
Power
kW
Montour
Low Sulfur
Secondary*
Current
Amps
0.36
0.46
0.37
0.31
0.26
0.22
POWER SET:
Coal + H20
Secondary
Voltage
kV
32.9
37.6
36.7
34.8
33.8
32.9
1A/2A
Current
Density
nA/cm2
13
17
14
12
9.9
8.2
*
Calculated
Current. Assume 60% Power
TABLE
Conversion
C-10. VOLTAGE VS CURRENT
DATA
DATE: 18 December 1975
TIME: 1:45
Primary
Current
Amps
70
60
60
50
30
PM
Primary
Voltage
Volts
260-300
270
260
250
220
LOCATION:
CONDITION:
Primary
Power
kW
47.6
Montour
Low Sulfur
Secondary
Current
Amps
0.37
0.26
0.26
0.21
0.13
POWER SET:
Coal + H20
Secondary
Voltage
kV
32
37
36
35
30
1B/2B
Current
Density
nA/cm2
33
9.8
9.7
8.0
4.9
120
-------
TABLE C-11. VOLTAGE VS CURRENT DATA
DATE:
TIME:
Primary
Current
Amps
70
50
30
20
18 December 1975
2:00 PM
Primary
Voltage
Volts
250-320
240
230
220
TABLE
LOCATION:
CONDITION:
Primary
Power
kW
20
12
Montour
Low Sulfur
Secondary
Current
Amps
0.35
0.13
0.091
C-12. VOLTAGE VS CURRENT
POWER SET:
1C/2C
Coal + H20 Injection
Secondary
Voltage
kV
33.8
32.0
31.0
29.1
DATA
Current
Density
nA/cm2
13
8.4
5.0
3.4
DATE:
TIME:
Primary
Current
Amps
225
125
100
80
70
60
50
20
18 December 1975
2:15 PM
Primary
Voltage
Volts
280-300
260
250
240
230
220
210
200
LOCATION:
CONDITION:
Primary
Power
kW
Montour
Low Sulfur
Secondary
Current
Amps
1.1
0.56
0.44
0.36
0.31
0.26
0.21
0.084
POWER SET:
1D/2D
Coal + ^0 Injection
Secondary
Voltage
kV
35.7
34.8
33.8
32.0
31.0
30.1
30.1
28.2
Current
Density
nA/cm2
41
21
16
13
12
9.7
7.8
3.1
121
-------
APPENDIX D
S02 REMOVAL IN AN ELECTROSTATIC PRECIPITATOR
G. B. Nichols, Southern Research Institute
At frequent intervals, the discussion of S02 removal by electrostatic
precipitation arises. This technique will work in principle because of the
low ionization potential of the SOn molecule, but the method is not feasible
in terms of conventional electrostatic precipitators, as shown by the
following analysis.
We will determine the amount of time required to remove the S02 from a
flue gas with an SO? concentration of 2000 ppm with an electrostatic precipi-
-92
tator operating with an average current density (j) of 100 x 10 A/cm .
The SOo concentration corresponds to plants burning coal with a sulfur
content of about 3 percent and the current density is on the high side of
the average for flyash precipitators.
The number of molecules of SOo per cubic centimeter of gas is determined
for Avogadro's number and the gas concentration for S02- Avogadro's number is
the number of molecules present in 22.4 liters of a gas at standard conditions,
NQ = 6.024 x 1023 molecules/22.4 liters
= 2.7 x 1019 molecules/cm3
For an S02 concentration of 2000 ppm, the number of SOo molecules/cm is;
N$0 = NQ x 2 x 10"3 = 5.4 x 1016 molecules's02/cm3.
^ 2
The number of electrons reaching a 1 cm area of the plate per second
is determined from the current density.
N = 100 x ID'9 m x ] co^b/sec 1 electron
- -
6 en/ ' amp 1.6 x 10" Iy coulombs
e = 6.25 x 1011 electrons/sec
122
-------
If each electron is attached to an SO? molecule, then there will be
11
6.25 x 10 S02 molecules transported to each square centimeter of collection
electrode per second.
For a pi ate-to-corona-wire spacing of 10 cm, the number of SO? molecules
2
contained within a 1 cm area between the wire and plate is the product of the
S09 concentration (N™ ) and the volume (10 cm^) or Ncn within the 10 cm space
f- -,-j oUp OUp
5.4 x 10" S09 moleculis; i
11
The time required to remove the S02 molecules at a rate of 6.25 x 10
molecules per second is the ratio of the concentration to the rate.
5.4 x 10 molecules
6.25 x 1011 molecules/sec
5
Time = 8.65 x 10 seconds = 240 hours
Thus we see that a precipitator must retain the gas stream for a period of 240
hours to remove 2000 ppm of S02- If the gas velocity through the precipitator
were 1.5 m/sec, the precipitator length required would be 366 m!
123
-------
APPENDIX E
CALCULATION OF EFFICIENCY AS A FUNCTION OF PARTICLE DIAMETER
The particle collection efficiency as a function of particle diameter
s calculated as follows:
1. An "eyeball" curve was drawn through the inlet and outlet
AC/A log d data contained in Appendix A for each day of testing.
2. Points were picked from the "eyeball" curves for calculation
of the penetration of particles of diameter d, Pt(d).
3. Pt(d) was calculated from:
Pt(d) = (AC/A log d)outlet * (AC/A log d)inlet.
4. Pt(d) data were plotted and curves drawn through the points.
The 90 percent confidence limits of the penetration versus particle
ameter curve were estimated as follows:
1. An "eyeball" curve was drawn through the +90 percent confidence
limit and -90 percent confidence limit of the inlet and outlet
AC/A log d data for each day's testing. These curves were drawn,
as nearly as possible, parallel to the curves through the mean
of the AC/A log d points.
2. Points were picked from these curves for calculation of +90 percent
Pt(d) and -90 percent Pt(d).
3. +90 percent Pt(d) was calculated from:
+90% Pt(d) = -90% (AC/A log d)outlet * +90% (AC/A log d)inlet.
4. -90 percent Pt(d) was calculated from:
-90% Pt(d) = +90% (AC/A log djoutlet * -90% (AC/A log d)inlet.
The points used to calculate Pt(d), +90 percent Pt(d), and -90 percent
(d) for the December 16 test data are shown in Table E-l.
The Pt(d) curve with estimated ± 90 percent confidence limits is shown
Figure E-l.
124
-------
It is recognized that this procedure lacks statistical rigor. However,
it does provide useful results. Work is underway to develop more rigorous
methods of performing the calculations. This procedure does take into
account the different confidence intervals for the various stages of impactors.
In general, the lower stages, corresponding to diameters less than 2 microns,
have wider confidence limits than do the upper stages. Thus the penetration
versus particle diameter curves have very wide confidence bands for diameters
less than 2 microns.
TABLE E-l. PENETRATION CALCULATIONS FOR DECEMBER 16, 1975
AC
A log d
d
micron
0.35
0.5
0.8
1.0
2.0
4.0
8.0
10.0
20.0
30
inlet
900
230
310
350
800
2800
8800
13000
20000
25000
outlet
52
64
78
84
110
120
95
78
50
30
Pt(d)
27.4
27.8
25
24
13.8
4.3
1.07
0.6
0.25
0.12
AC
A log d
+90%
inlet
250
350
520
600
1300
3800
12000
19000
27000
29000
-90%
outlet
24
31
42
46
60
65
50
47
25
20
-90%Pt(d)
9.6
8.86
8.1
7.7
4.62
1.71
4.2
0.25
0.93
0.07
AC
A log d
+90%
inlet
100
130
180
200
500
1800
5500
7000
14000
21000
-90%
outlet
80
97
115
122
160
175
140
no
75
45
+90%Pt(d)
80
75
64
61
32
9.72
2.55
1.6
0.54
0.22
125
-------
O.I
5
10
1 2°
y
I 30
| 40
H 50
tr.
l- 60
UJ
5 70
a.
80
90
95
98
99
99.9
I I I I I I
0.1
-90%
i I i i
99.9
99
98
95
90
80
70 |
60 »
50 >"
40 2
o
u.
ill
20
10
5
2
I
0.1
0.2 03 0.5 0.7 1.0 2 3
PARTICLE DIAMETER, microns
5 6 7 8 9 10
Figure E-l. Efficiency versus particle diameter for December 16
high sulfur coal.
126
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-76-027
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Effect of a Flyash Conditioning Agent on Power
Plant Emissions
5. REPORT DATE
October 1976
6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
L.E. Sparks
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Participate Technology Branch
Industrial Environmental Research Laboratory/EPA
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
NA (In-House Report)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
In-House (Final): 12/75-6/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT,
The report gives results of a study undertaken as a preliminary program
to provide data on the environmental effects of a chemical flyash conditioning agent
(Apollo Chemicals conditioner LPA 402A). Both the emissions due to the chemical
and its effect on electrostatic precipitator (ESP) performance were investigated. The
tests were conducted over a 10-day period at Pennsylvania Power and Light Co. 's
Montour Plant with the plant operating on high sulfur coal (without conditioner) and on
low sulfur coal (with and without conditioner). Sulfur oxides (SOx), ammonia, orga-
nics, particulates, flyash resistivity, and ESP power supply values were measured
during each test period. During conditioner injection, the low sulfur coal flyash
resistivity was reduced about 60%, although the ESPs responded slowly to this change
and its effect was not clearly evident during the test period. The results of the SOx,
ammonia, and particulate measurements were inconclusive due both to insufficient
precision for the number of field tests and to the effect of boiler transients. It is
unlikely that the ESP will meet particulate standards when low sulfur coal is burned
even if the conditioner is used under test conditions. The test provided useful back-
ground information for planning. More thorough testing at Montour seems warranted.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Fly Ash
Treatment
Dust
Electric Power Plants
Flue Gases
Electrostatic Precip-
itators
13. DISTRIBUTION STATEMENT
Unlimited
b.IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Chemical Conditioning
Particulates
19. SECURITY CLASS (Tills Report)
Unclassified
20. SECURITY CLASS (This page}
Unclassified
COSATI Field/Group
13 B
2 IB
11G
10B
21D
!1. NO. OF PAGES
134
22. PRICE
EPA Form 2220-1 (9-73)
127
------- |