&EPA
TVA
United Slates
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-80-171
October 1980
Tennessee Valley
Authority
Office of Power
Energy Demonstrations
and Technology
Chattanooga TN 37401
EDT-116
Field Study to Obtain
Trace Element Mass
Balances at a Coal-fired
Utility
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2, Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the 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 sys-
tems The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects- assessments of, and development of, control technologies for energy
system's; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessanly reflect
the views and policies of the Government, nor does ment.on of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-80-171
October 1980
Field Study to Obtain Trace
Element Mass Balances at a
Coal-fired Utility Boiler
by
Robert Evers, V.E. Vandergriff, and R.L. Zielke
TVA, Division of Energy Demonstrations and Technology
1140 Chestnut Street, Tower II
Chattanooga, Tennessee 37401
Interagency Agreement No. D5-E721
Program Element No. 1NE624A
EPA Project Officer: Julian W. Jones
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, IMC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
DISCLAIMER
This report was prepared by the Tennessee Valley Authority and has
been reviewed by the Office of Energy, Minerals, and Industry, United
States Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the Tennessee Valley Authority or the United States
Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
11
-------
ABSTRACT
This program was to identify the mass flow rates of the minor and
trace elements from the various streams from a coal-fired utility boiler.
This information was used to obtain a mass balance for 25 elements. The
mass balances used the inlet and outlet flows associated with the three
major pieces of equipment; i.e., the pulverizer, boiler, and electrostatic
precipitator. This provided a mass balance for each element for the vari-
ous parts of this system. Along with the trace elements which were being
measured, organic samples were obtained and analyzed from various streams
for polychlorinated biphenyls (PCB's) and polynuclear organic matter
(POM's) by Monsanto Research Corporation and by GCA/Technology Division.
Thus, the mass balance presented reflected a fairly complete picture of
unit no. 1 (Colbert Steam Plant) under a normal operating condition. The
mass balances show that sampling techniques need to be improved. First,
the analysis of the vapor phase samples reported all concentrations below
the detection limit for each element. Second, the mass balances of only
ten elements (represented 61 percent of the total ash flow) closed within
±10 percent for at least two of the three major pieces of this system.
111
-------
CONTENTS
Disclaimer
Abstract
List of Figures
List of Tables
.......
Acknowledgments ............ ' ............ V11
Table of Abbreviations ........... .......... *
Conversion Table ....... '.'.'.
Executive Summary ... ' •
J
...
..................... xin
1. Introduction ............ ,
2. Conclusions .......... ........ ~
3. Recommendations ............ ....... c
4. Experiment .............. .......... g
Boiler Description and Operation ...... '.'.'.'.'. 6
Sampling Procedures .................. g
Coal ....................... g
Whole Coal ................. g
Pulverized Coal ............... 11
Pyrites ..................... H
Bottom Ash .................... n
Boiler Outlet Duct ................ 14
Mechanical Collector Fly Ash ........... 14
Electrostatic Precipitator Hopper Ash ...... 16
Electrostatic Precipitator Inlet and
Outlet Ducts .................. i£
Raw Sluice Water ............... jg
Methods of Chemical Analysis ........... 17
5. Results and Discussion ....... . ....... ' -,0
Boiler Operating Conditions for Test ....... 18
Coal Analysis ................ ig
Characterization of Waste Streams ......... 22
Coal and Ash ............... 2?
Ash Sluice Water ............. ....
Fine Particulate Characterization ...'.'.'.'.'. 25
Physical Characterization ....... '. '. ' 25
Particulate Sizing by Aerodynamic
Cascade Impactors ........... OQ
Brink Cascade Impactor ......... '.'.''' 30
Andersen Cascade Impactor ......... ' oo
Size Distribution of Fly Ash by
Light Transmission Microscope. ... 44
Density of Fly Ash by Pycometer. ..'.'.'.' 44
Particle Counters ............ [ 44
Chemical Characterization of Fine
Particulates ............... 44
Emissions Samplings for Organics and Sulfates '. '. 48
Organic and other Analyses of Ash Sluice
Samples ........... ci
IV
-------
CONTENTS
(continued)
Mass Balance 51
Determination of Flow Rates 51
Material and Flow Analysis - System and
Sample Points 51
Material Flow Analysis - By Trace Element .... 66
Special Considerations and Assumptions 72
The Mechanical Collector 74
Trace Element Flows 74
Vapor Phase Data 74
Flow Rate Estimation and Associated
Problems 74
Estimate of the Mass Balance 78
References 81
Appendixes
A Description of Sampling Trains 82
Train 1 - Total Participate 83
Train 2 - Vapor-Phase Trace Elements 83
Train 3 - Particulate Interference 83
Train 4 - Mercury/Backup Vapor-Phase
Trace Elements 87
Brink Cascade Impactor 87
Andersen Cascade Impactor 87
Optical and Diffusional Sizing System 87
Polycyclic Organic Material (POM)
Sampling System 91
B Methods of Chemical Analysis 93
-------
LIST OF FIGURES
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Title
Colbert Unit No. 1
Coal and Ash Flow Diagram .
Ash Sluicing Test Schedule
Bottom Ash Sampling Flume
Lognormal Distribution Plot of Runs 2, 3, 6,
10, and 14
Weibull Distribution Plot of Runs 2, 3, 6,
10, and 14
Lognormal Distribution Plots of Runs 1, 4, 7,
9, 11, 12, 13, 15, 16, and 17
Weibull Distribution Plot of Runs 1, 4, 7, 9,
11, 12, 13, 15, 16, and 17
Lognormal Distribution Plot for Andersen Impactor .
Weibull Distribution Plot for Andersen Impactor . ,
Largest and Smallest Total Flows
Errors in the Material Balance By Trace Element
Around the Pulverizer
Errors in the Material Balance By Trace Element
Around the Boiler
Errors in the Material Balance By Trace Element
Around the ESP
o; — MfioA SwRtpm Schematic
Page
7
9
12
13
38
- 39
40
41
42
43
68
. 69
70
71
75
15
-------
LIST OF TABLES
Table Title Page
1 Elements With Good Material Balances ........ xiii
2 Summary of Samples and Constituents Determined ... 10
3 Major and Trace Elements in Coal Under Consideration
in This Study ................... ^
4 Procedure for Preservation of Water Samples ..... 15
5 Boiler Operating Conditions for Test ........ 19
6 Proximate Analysis of Coal ............. 20
7 Ultimate Analysis of Pulverized Coal ......... 20
8 Major Elements (Given as Oxides) of Ash Produced
in the Laboratory by Firing Coal at Approximately
800°C ....................... 21
9 Chemical Balance Around the Pulverized Coal
Using Trace Element Analysis and Ultimate
Analysis of Coal .................. 21
10 Trace Element Concentration in Coal, Pyrites,
Bottom Ash, and Mechanical Collector Ash ...... 23
11 Trace Element Concentration in ESP Inlet
Particulate, ESP Hopper Ash and ESP
Outlet Particulate ................. 24
12 Trace Element Concentration in Bottom Ash
and ESP Sluice Water Samples ............ 26
13 Trace Element Concentration in Inflow
Water Samples ................... 27
14 Field Analysis of Ash Sluice Water Samples ..... 28
15 Test Summary of Brink Sampling Parameters ...... 31
16 Comparison of the Lognormal and Weibull
Distributions Fitted to the Brink Impactor
Data From Colbert ................. 32
17 Summary Statistics of Curve Fits to the Brink
Impactor Data From Colbert Grouped Runs ...... 32
18 Estimated Median Particle Diameter in Microns
Based on Individual Runs of Brink Impactor Data . . 34
vii
-------
LIST OF TABLES
(Continued)
Table Title Page
19 Estimated Median Particle Diameter in Microns
Based on Grouped Runs of Brink Impactor Data ... 35
20 Test Summary of Andersen Sampling Parameters 36
21 Comparison of the Lognormal and Wiebull
Distributions Fitted to the Andersen
Impactor Data From Colbert 37
22 Estimated Median Particle Diameter in Microns
Based on Individual and Grouped Runs of
Andersen Impactor Data 37
23 The Concentration of Condensation Nuclei
Size Particles (Inlet Runs) 45
24 The Concentration of Condensation Nuclei Size
Particles (Outlet Runs) 46
25 Compositing Schedule for Chemical Analysis of
Particulate Samples From the Brink and
Andersen Impactors 47
26 Chemical Analysis of ESP Inlet Fine
Particulate—Brink Impactor 49
27 Chemical Analysis of ESP Outlet Fine
Particulate—Andersen Impactor 50
28 Polycyclic Organic Materials 52
29 Trace Metal Concentrations in Filtrate of Ash
Slurry Samples 55
30 Mass Flow Rate for Various Flows in System 56
31 Summary of Element Flow Rates 57
32 Total Material Flows of the Trace Elements and
Estimated Errors Around the Pulverizer, Boiler,
and Electrostatic Precipitator 65
33 Total Material Flows of Trace Elements By Test. ... 65
34 Estimated Errors Around the Pulverizer, Boiler,
and Electrostatic Precipitator By Test 65
35 Mean and Standard Deviation of Daily Flows of
Trace Elements (in Ibs/hr) By Sample Point .... 67
Vlll
-------
LIST OF TABLES
(Continued)
Table Title Page
36 Summation of Material Flows (Ib/h) for Trace Elements
in Whole Coal 67
37 Errors in Material Balances for Trace Elements
With Large Total Flows 72
38 Trace Elements With Good Material Balances 73
39 Analysis of Large Flow Trace Elements With
Large Errors Days 1 and 8 Flows Removed 73
40 Trace Element with Minimum Detectable Limits 76
41 Flow Rates and Associated Errors 77
42 Estimate Mass Balance of Major and Trace Elements
(Summation of 25 Elements) by Sample Point 79
43 Estimated Mass Balance by Element for Each Sample
Point 80
ix
-------
ACKNOWLEDGMENTS
This study was initiated by TVA as part of the projects entitled "Fly
Ash Characterization and Disposal" and "Characterization of Effluents from
Coal-Fired Utility Boiler," and is supported under Federal Interagency
Energy/Environment Research and Development Program between TVA and EPA.
Thanks are extended to EPA Project Officers Michael C. Osborne and Dr. Ron A.
Venezia, and TVA Project Director Dr. Hollis B. Flora II. Appreciation is
also extended to James R. Crooks, Robert L. Frank, Ronald A. Hiltunen,
Dr. Lyman Howe, Dr. Chao-Ming Huang, Frank G. Parker, Shirley S. Ray,
R. J. Ruane, and Randall L. Snipes for their aid in this effort.
-------
TABLE OF ABBREVIATIONS
Alk Alkalinity
A&MF Ash and Moisture Free
BA Bottom Ash
ESP Electrostatic Precipitator
MC Mechanical Collector
ORP Oxidation Reduction Potential
PCB Polychlorinated Biphenols
POM Polycyclic Organic Material
XI
-------
CONVERSION TABLE
A list of conversion factors for British units used in this report is as
follows:
British Metric
1 micron 10 6 meters
1 inch 2.54 centimeters
1 foot 0.3049 meter
1 mile 1.609 kilometers
1 pound 0.454 kilogram
1 ton (short) 0.9072 metric tons
1 gallon 3.785 liters
-------
EXECUTIVE SUMMARY
This study established sampling points around each of the following
major pieces of equipment in a fossil-fired steam plant—pulverizer,
boiler, and electrostatic precipitator. At these sampling points, the
mass flow rate and the concentrations of 25 elements were measured. The
measurements included vapor samples from the flue gas duct although vapor
phase flows were not used in the mass balance calculations because all
concentrations were below detection limits. If only the trace elements
are considered, the mass imbalance for the entire system ranged from -6.4
to +12. This range of closure was acceptable for the total trace element
material balances. The mass imbalance for each element was quite different.
Only ten elements (Table 1) closed within ±10 percent for two or more mass
balances around the major pieces of equipment. These ten elements represent
approximately 61 percent of the total mass flowrate in the fly and bottom
ashes.
Table 1
Elements With Good Material Balances
Si V
Al Pb
Mg Cu
Ti Sb
Ba Be
Since the material balance for the total system closed and the analy-
tical methods were adequate for most samples (except vapor samples), it
was expected that the balances for each element would close. A problem
was encountered with some major elements (i.e., Fe, Ca, K, and Na), and
these elements exceeded ±10 percent imbalance in two or more segments of
this system; therefore, some doubt is cast on the rest of the trace element
material balances.
The size distributions of fly ash were also considered. Cascade
impactors were used to evaluate the size distributions of the larger par-
ticles, while optical or condensation nuclei counters were used for the
submicron particles. Samples from the cascade impactors were combined to
allow chemical analysis for three ranges of particle sizes—<1 pm, 1 to 3
pm and >3 pm. Samples were analyzed by TVA and Accu-Lab. These samples
allowed verification of elements which tended to concentrate in the smaller
particle ranges. The halogens and volatile elements were predominate. The
alkali metals (except Ca) also seemed to be concentrated in the smaller
particles, while the refractory oxides generally showed no trends relating
to particle size. Cr, Cu, and Fe tended to concentrate in the smaller
particle sizes of the Brink cascade impactor but showed no trend from the
Andersen cascade impactor. (These are listed only as trends because of
the large amount of scatter in the data.)
Xlll
-------
The water side of this project involved measuring the bottom ash and
fly ash flow rates from a flume that attached to the inlet of the ash
pond. This sampling procedure also required that the inflow sluice water
be analyzed as well as the transport water to provide adequate background
water quality data for the associated mass flows of each element in the
water side of this program. The problem encountered with these measure-
ments was that some of this data indicated a lower bottom ash/fly ash
split than the data from the air side. The data from the air side was
used in the mass balance when it conflicted with the water side because
the air side data was obtained by standard methods. The ESP Hopper
measurements from the water side agreed closely with the air side
measurement.
Organic sampling for POM's and PCB's was accomplished for various
types of samples. GCA/Technology Division collected vapor samples from
the inlet and outlet of the ESP and determined the concentration of sul-
fate in the flue gas from the outlet of the ESP. GCA found the concen-
tration of benzo(a) pyrene to be 0.3 M8/m3 aru* an upper limit to PCB
emissions of 1.7 |Jg/m3. Sulfate emissions were 6.47 mg/m3. The solid
and slurry samples (coal, pyrite, and ash) were sent to Monsanto Research
Corporation for analysis. No PCB's were detected in any of the samples.
POM's were found in all but one sample; however, only two samples con-
tained POM's in appreciable quantities (>1 (Jg).
This report completes a summary of all sampling that was undertaken
at Colbert Steam Plant (unit 1). It was a complete mass balance study
intended to help characterize the various waste streams associated with
a coal-fired utility boiler. The conclusions that were drawn are sup-
ported by our data along with trends that were noted. Some of our conclu-
sions conflict with other reports, especially in the fine particulate
analysis. We have only reported what we could support from the data in
this report.
xiv
-------
SECTION 1
INTRODUCTION
Interest in coal combustion research has increased in recent years
due, partly at least, to concern for the environmental effects of coal
combustion products. Two tasks of the EPA Interagency Energy/Environment
Research and Development Projects, "Characterization of Effluents from
Coal-Fired Utility Boilers" and "Fly Ash Characterization and Disposal,"
comprised this research effort. The "Characterization of Gaseous Emis-
sion" task was concerned with atmospheric emission of trace constituents
in coal. The "Fly Ash Characterization" task consisted of investigating
the chemical constituents of liquid and solid effluents from the unit.
Because the combustion of fossil fuels is known to generate a multitude
of chemical species that may be discharged into the environment as com-
bustion products, and because the use of coal is expected to increase
during the next two or three decades, attention was given to obtaining
information on the contribution of power plants to the environmental
loading of these chemical substances. This study may in turn lead to a
more complete understanding for the distribution of these species and
ultimately to the development of better means to control their discharge
to the environment.
Studies (1-6) have been reported which undertook to determine the
trace element composition in coal and in fly ash collected at various
locations along the flue gas stream and to delineate the mass balance of
trace elements about the power plant systems. These studies have suc-
ceeded in demonstrating some aspects for the flow of trace elements in
the flue gas and the distribution of these elements in various fractions
contained in the fly ash particles. However, the fates of many of the
trace elements in the vapor phase were not well known primarily due to
inadequate sample collection techniques. Additionally, there were no
previous studies focused on the water side. This study utilized the
water side data to provide a more complete understanding for the fate of
trace elements in effluents (whether liquid, solid, or gaseous).
In view of the concern with toxic trace elements that may be released
from fossil-fuel combustion, TVA conducted a research program to quantify
and characterize such combustion products from all effluent streams from
one of its coal-fired power plants — specifically, Colbert Steam Plant in
northwestern Alabama.
-------
SECTION 2
CONCLUSIONS
The following conclusions are based on the results from an 8-day
field study conducted on a TVA steam plant (Colbert). The test boiler
was a wall-fired utility boiler rated at 200 MW, and the coal was mined
from western Kentucky.
1. The mass balance around the pulverizer accounts for approxi-
mately 94 percent of the substance in coal. After making
appropriate allowances for oxygen and silica content, the
sum of the 27 trace elements considered in this study
closely approximates the ash content of coal.
2. The high variability in the data at the electrostatic precipi-
tator (ESP) outlet, compared to other sampling points, suggests
that difficulty in analyzing small sample sizes was primarily
responsible for spurious results or that the sample was not
representative.
3. Fly ash has a much more pronounced effect on raw sluice water
quality than does bottom ash. Ca, B, Se, Cr, Al, F, and 804
concentrations and pH, alkalinity, and conductivity experi-
enced relatively large increases, especially in the fly ash
sluice water. Zn and Cu appear to have been partially removed
from the dissolved portion of the sluice water during bottom
ash sluicing; and Cu, Mg, and Si appear to have been reduced
from the solution during fly ash sluicing.
4. Since there were no essential differences in the levels of Ca,
Mg, S04, Si, and Hg for the 45-minute and 14-hour supernatant
samples of fly ash or bottom ash sluice water, it is suggested
that most chemical reactions between ash and water occurred
within 45 minutes from the start of the sluicing.
5. Based on the participate sizing and optical particle counting
results, it may be concluded that the ESP removed a large per-
centage of particles greater than 3.5 (Jm in diameter. One half
the total mass of particles leaving the system in stack gases
were smaller than 3.5 M">- Because particles with a diameter
of 0.4 (Jm comprise 80 percent of the total particle count at
both the ESP inlet and outlet. It is suggested that the ele-
mental mass composition of particles <3.5 pm should be studied
further.
6. Our results of the fine particulate chemical analysis indicate
that elements generally considered volatile tended to concen-
trate in fine particulate smaller than 3 pm. The alkaline
earth and alkali metal elements also tended to concentrate
in the fine particulate smaller than 3 |Jm. Most of the
remaining elements showed no discernable trends or preference
for particle size.
-------
7. No polychlorinated biphenyls were detected in any of the ash
sluice samples; only two out of eight samples contained poly-
cyclic organic matter in appreciable quantities (>1 (Jg).
8. The aggregate material balance for the trace element flows
over the entire 8-day sampling period had the smallest error
around the pulverizer (+3.7 percent error), a +5.8 percent
error around the ESP, and the largest error around the boiler
(-6.8 percent). On an aggregate basis this would indicate
adequate sampling and analytical procedures, but detailed
examination of individual trace elements indicated this error
was sufficiently large to mask their behavior.
9. Especially noticeable between the whole coal and pulverized
coal flow rates, the system flow rates were extremely different
and inconsistent throughout the system on days 1 and 8. This
difference may be due to a flow rate estimation problem and/or
a detection limit problem.
10. Fifteen of the twenty-five elements had mass imbalances in
excess of ±10 percent for two of the three loops in the sys-
tem. Surprisingly, iron, calcium, potassium, and sodium fell
into this category. Because these four elements are relatively
abundant in coal, these results were unexpected. Removal of
tests 1 and 8 did not improve the results. Ten elements
(approximately 61 percent of the total trace element flow) had
a material balance within ±10 percent for at least two of the
three loops in the system.
11. Of the pulverizer, boiler, and ESP loops, the largest varia-
bility was associated with the ESP and appeared to be both a
problem of day-to-day variability and the difficulty of col-
lecting a sample and analyzing the vapor phase and fine par-
ticulate data. Large variability was consistently seen in the
ESP outlet data, the pyrite data, and the bottom ash data.
12. Two groups of elements, which were distinctly identifiable
through the system, followed similar trends in their distri-
bution in the waste streams. The first group consisted of
lead, iron, manganese, potassium, silicon, and cobalt. The
second group consisted of beryllium, vanadium, chromium,
magnesium, copper, titanium, and zinc. Not all expected
groupings could be identified due to the coarseness of the
data.
13. The vapor phase sampling and analytical procedures were inade-
quate. Lower limits of detectability in the vapor phase con-
sistently exceeded the total flow rate of the element in the
whole coal. Chlorine, fluorine, and selenium were three ele-
ments whose mass balances around the boiler and ESP could be
drastically improved through better vapor phase data and more
flow rate sampling around the boiler and ESP.
-------
14. Contamination from stainless steel particles may have occurred
during the entire sampling period as chromium, nickel, and
manganese exhibited large gains through the system.
-------
SECTION 3
RECOMMENDATIONS
1. Since the product formed by ashing coal in the laboratory could
be qualitatively and quantitatively different from that formed
in a boiler, false estimates of total ash flow rates may result.
Therefore, it is recommended that work be done to determine if
ash produced in the laboratory is substantially different from
ash produced in a boiler.
2. Sampling procedures need to be improved for these elements--F,
Cl, Sb, As, Be, B, Hg, Se, S, and Pb. Large amounts of these
elements are unaccounted for in the mass balance.
3. It is strongly recommended that sulfur be included with Fe and
Al as a monitored element in all flows. The instrumentation,
chemical behavior, and combustion behavior of sulfur would serve
as an important check on sampling procedures, system behavior,
flow rates, and accuracy.
4. Variability of power plant operating conditions is a factor
that must be recognized. In order to identify the behavior of
trace elements under these various conditions, a longer samp-
ling program is recommended.
5. The addition of measured flow rates around the boiler and ESP
could drastically improve the mass balance estimates and, hence,
the understanding of the behavior of trace elements through
the system.
-------
SECTION 4
EXPERIMENT
This study began in August 1976, at the Tennessee Valley Authority's
Colbert Steam Plant located in northwest Alabama on the Tennessee River.
Because of manpower limitations, a contractor experienced in stack sampling,
Midwest Research Institute (MRI), was selected to conduct the sampling
program for trace elements in the boiler flue gases. The GCA Corporation,
under contract to the Environmental Protection Agency (EPA), sampled the
adjacent electrostatic precipitator (ESP) outlet duct "A" on the same
unit for polycyclic organic matter. TVA personnel conducted all other
sampling efforts during this study.
All tests were conducted with the unit at steady operating condi-
tions near full load (approximately 180 MW) throughout each daily test
period. The unit was brought to test conditions and allowed to stabilize
for a minimum of one hour before testing began. Boiler soot blowing was
done each morning at least one hour before testing began, and air pre-
heater soot blowing was done on the midnight shift. Flue gas tempera-
ture was near the design temperature of 340°F for the flue gas entering
the ESP.
BOILER DESCRIPTION AND OPERATION
The Colbert unit 1 (Figure 1), which was selected for this study,
is a split-wall, pulverized coal-fed, wall-fired, dry-bottom boiler
manufactured by Babcock and Wilcox Company. Eighteen burners are arranged
in three horizontal rows of 6 burners each. The unit has a rated capacity
of 1,280,000 pounds of steam per hour and a rated 200-MW capacity at a
coal fuel input of approximately 76 tons per hour. The steam generator
is served with a Ljungstrum regenerative single-pass air preheater and
by a Lodge-Cottrell double-chamber electrostatic precipitator with three
fields, each nine feet deep. The nameplate design data for the precipi-
tator specifies a collection efficiency of 97 percent with a gas flow of
906,000 ft3/min at 340°F flue gas temperature.
Coal is fed continuously from the coal bunker, across a coal-scale
system, to the pulverizer, where it is ground to a consistency resembling
that of talcum powder. As the coal is pulverized, hard pyritic material
is rejected automatically by the mill and is collected in hoppers adja-
cent to the pulverizers. The pulverized coal is then transported to the
furnace burners by the hot primary combustion air, where it is burned
instantaneously at temperatures in the range of 2700°F-3200°F.
The solid byproducts from coal combustion can be generally classi-
fied as bottom ash, fly ash, and slag. The bottom ash and slag [collec-
tively referred to as bottom ash (BA)] fall to the bottom of the furnace,
where they are collected and removed by high-pressure water jets. The
fly ash is carried on through the boiler in the hot gas stream and passes
through the air preheater where the gases are cooled to approximately
400°F. The gas stream, with the entrained fly ash, continues on through
-------
Figure I, Colbert unit I.
-------
mechanical fly ash collectors (MC) that have had the collection vanes
removed, where a small percentage of fly ash is removed by gravitational
settling. Upon exit from the mechanical collectors, the gas stream enters
the ESP which removes most of the remaining fly ash (97 percent) and col-
lects it in hoppers. The gases and any remaining fly ash exit the system
through the stack.
Bottom ash, pyritic material, and the small amount of fly ash col-
lected in the mechanical collectors are sluiced to a common sump. This
slurry then passes through clinker grinders and is pumped to the ash
settling pond. Fly ash collected in the ESP hoppers is removed by a
water-operated exhauster and sluiced to the ash settling pond as a slurry.
SAMPLING PROCEDURES
Basically two different sampling efforts were carried out simul-
taneously during this test. One effort focused on the gaseous and
particulate emissions from the boiler. The second effort was directed
at characterizing the various ashes that were removed from the process
at various points in the system. Figure 2 is a block diagram of the
process with the various sampling points identified. Table 2 is a sum-
mary of the samples taken and the chemical constituents or parameters
for which they were analyzed. Table 3 is a listing of the trace elements
investigated. The two efforts overlapped at the ESP, where two samples
of the hopper ash were taken.
Because of financial constraints, the gaseous emissions study was
conducted on only one-half the unit—the "B" side; and hopper ash samples
were, therefore, taken on only that half of the ESP. The ash characteri-
zation effort took a sample across the entire unit. The following material
is a description of the sampling procedures used at each sampling point
identified in Figure 2.
Coal
Whole Coal--
Coal flowing from the coal bunker to the test boiler passes across
one of six coal scales serving the boiler. These scales are designed
for a 500-pound-per-dump capacity and are equipped with counters for
recording the number of dumps. The scales also have a bypass arrangement
for feeding coal directly to the pulverizers in the event of a scale mal-
function. Coal flow rates for the test periods were based on these scale
counters, and estimates were made for those periods when the bypass
arrangement was in use-.
The whole coal samples were taken from each of the six coal scales
serving the test boiler. A sampling ladle was used to collect a grab
sample at 15-minute intervals from each of the six coal scales. All the
grab samples for a given test were composited in a large plastic con-
tainer and, at the end of each test, the total sample was ground to No. 4
mesh and then riffled into a quart-size proportional homogeneous sample.
This sample was put into a labeled glass jar.
-------
COAL SCALES
COAL PULV.
PYRITE HOPPER
FLY ASH
BOTTOM ASH
HYDROVACTOR
CLINKER
GRINDER
MECHANICAL
COLLECTOR
VAPOR
SOLIDS
OVERFLOW
(IS) VAPOR
SOLIDS
HYDROVACTOR
AIR SEPARATOR
ASH POND
Figure 2. Coal and ash flow diagram.
-------
TABLE 2. SUMMARY OF SAMPLES AND CONSTITUENTS DETERMINED
Sample - Location
Constituents Analyzed
Coal - 1, 2
Pyrites - 3
Bottom Ash, Dry Ash - 6
,45 Minute Supernatant - 6
,14 Hour Supernatant - 6
Boiler Outler - 4
Mechanical Collector Ash - 5
ESP Ash, Dry Hopper Ash - 9
,45 Minute Supernatant - 7
,14 Hour Supernatant - 7
ESP Inlet Duct, Total Particulate
,Vapor Phase - 8
^articulate Sizing - 8
ESP Outlet Duct, Total Particulate
,Vapor Phase - 10
,Particulate Sizing - 10
Raw Sluice Water, Total
,Dissolved
- 8
Trace elements,* proximate and ultimate
analysis
Trace elements
Trace elements, major oxides (Al,
Ca, Pe, K, Mg, Na, S, Si, and Ti)
Ca, Mg, Hg, S04, Si, PH, A1K, ORP,
Conductivity
Trace elements, S04, pH, A1K, ORP,
Conductivity
CO, C02, 02
Trace elements, major oxides
Trace elements
Ca, Mg, Hg, S04, Si, pH, A1K, ORP,
Conductivity
Trace elements, S04, pH, A1K, ORP,
Conductivity
Trace elements (except Na, B, S, Si)
Trace elements (except Na, B, S, Si)
Trace elements (except Na, B, S, Si)
- 10 Trace elements (except Na, B, S, Si)
Trace elements (except K, Na, B, S, Si)
Trace elements (except B and S)
Trace elements, S04, SS, pH, A1K,
ORP, Conductivity
Trace elements, S04, TDS
"'Trace elements are those listed in Table 3.
10
-------
TABLE 3. MAJOR AND TRACE ELEMENTS IN
UNDER CONSIDERATION IN THIS STUDY
Aluminum (Al)
Antimony (Sb)
Arsenic (As)
Barium (Ba)
Beryllium (Be)
Boron (B)
Cadmium (Cd)
Calcium (Ca)
Chlorine (Cl)
Chromium (Cr)
Cobalt (Co)
Copper (Cu)
Fluorine (F)
Iron (Fe)
Lead (Pb)
Magnesium (Mg)
Manganese (Mn)
Mercury (Hg)
Nickel (Ni)
Potassium (K)
Selenium (Se)
Silicon (Si)
Sodium (Na)
Sulfur (S)
Titanium (Ti)
Vanadium (V)
Zinc (Zn)
Pulverized Coal--
Pulverized coal samples were taken hourly using a cyclone sampler
to traverse one of three coal transport ducts from each of the six pul-
verizers. All the samples for a given daily test were composited and,
at the end of the day, each composite sample was mixed, reduced to a
quart-size sample, and sealed in a dry glass jar.
Pyrites^
At the start and end of a test, the pyrite hoppers were emptied.
Figure 3 shows the sluicing schedule that was used during this test for
pyrites, mechanical collector, bottom ash, and the ESP. At the end of
the test, a grab sample was collected from each of the six pyrite hop-
pers and a measurement of the depth of the pyrites in each of the hop-
pers was made. These measurements were used to calculate the total
volume of pyrites collected during the daily test period. At the end of
each day, the pyrite grab samples for a given test were composited,
ground to No. 4 mesh, reduced to a quart-size daily sample, and sealed
in a glass jar.
Bottom Ash
Bottom ash was removed from the boiler, sluiced to the ash pond
prior to each test, and then allowed to accumulate in the boiler during
the test. At the end of the test, the bottom ash was again sluiced to
the ash pond unless operational constraints required bottom ash sluicing
prior to the end of the test. Collection of a daily representative grab
sample from the bottom of the furnace was infeasible; therefore, all bot-
tom ash samples were taken as the ash was being sluiced to the ash pond
(Figure 2). A sampling flume (Figure 4), designed to provide a homogene-
ous sidestream from which representative samples could be taken, was
installed on the end of the transport pipe. The sidestream flow of
approximately 360 gallons per minute represented one-tenth of the total
sluice flow. Bottom ash slurry samples were taken at 30-second intervals
during sluicing, composited in 30-gallon plastic containers, and allowed
to stand undisturbed.
11
-------
^•MM
•M
•Ml^HM
••Ml
«^
V^^M
•«••
•••••
M»
^BMBH
•M^M
•m
«MM
—
—
••Ml
—
—
—
•
m^m
^•M
M
M
_.
••MB
•» I
^Hi^
•VI
H^^^B
^ •
^ia
•
—
—
•
tv
-
-•
—
^w <
--
»*•
• ^B
•^
-•
-•
-•
--
—
—
-
•»•
-
—
~
1--
-J
• •»
—
—
-
1—
L—
—
lm
-
-
—
—
-
ESP
MC TCCT .
BA TEST '
PYRITE
ESP
MC XC-CT o
BA TEST 2
PYRITE
ESP
MC
BA TEST 3
DV O 1 T C
r Y K II t
ESP
MC TFQT 4
BA TEST 4
PYRITE
ESP
MC TPQT c:
BA TEST 5
PYRITE
ESP
MC TCCT c
BA TEST 6
PYRITE
ESP
MC
TIT f T ~7
BA TEST 7
PYRITE
ESP
TF^T ft
BA TEST 8
PYRITE
MN I 23456789 10
N
23456
PRETEST SLUICING PERIOD
•-*• SLUICING AND SAMPLING
DURING TEST PERIOD
Figure 3. Ash sluicing test schedule.
-------
EXISTING ASH SLUICE PIPE
MIXING OF ASH a SLUICE WATER PROPORTION SAMPLE SPLITTER EXTENSION
i 4
i
[ \ FLOV
T L
1* 20"
!"
4"
11.75 -D.
TOP
14
22
SIDE
9
*
20'- 0"
13-0"
8'-0"
^S
OBLIQUE VIEW
SAMPLING POINT
5
Figure 4. Bottom osh sampling flume.
-------
After a 45-minute settling period, a water sample was siphoned from
the surface of the composite sample and analyzed in the field for pH,
alkalinity, Oxidation Reduction Potential (ORP), and conductivity. A
portion of this supernatant was filtered through a 0.45-pm millipore
filter and preserved for laboratory analysis. After the large composite
sample had remained quiescent for 14 hours, all the water was siphoned
from the settled solids and the decanted water volume was measured.
Samples of this 14-hour supernatant underwent the same field handling
and field analysis as the previous 45-minute supernatant sample. Table 4
gives the liquid sample preservation procedures that were used in the
field when taking all of the liquid samples. All of the settled solids
were placed in a plastic container and sent to the laboratory for analysis.
Composite samples of the first four days and of the last four days were
also taken.
Boiler Outlet Duct
The major gaseous components of the flue gas were measured at the
boiler outlet with instruments installed on a mobile TVA van. The
instruments installed included: a DuPont Model 400 nondispersive UV
spectrophotometer for S02, a Thermo Electron model 10A chemiluminescent
monitor for NO and NO , a Beckman Model 864 IR analyzer for CO and another
for C02, a Beckman Moael 742 polarographic sensor for 02, and a Beckman
Model 400 flame ionization analyzer for hydrocarbon. The hydrocarbon
analyzer was not used in this study because of technical difficulties.
The C02 analyzer malfunctioned, and no data on this constituent were
obtained. However, ORSAT measurements of CO, C02, and 02 were made at
the ESP inlet duct; these measurements compared favorably with the van
instrumentation measurements of CO and 02.
Five sampling probes to a common sample header were used to traverse
the duct at the entrance to the air heater through five ports at three
traverse points (depth) to draw representative samples. The samples were
conveyed through Teflon tubing to the instruments located in the van for
measurements of S02, NO-NO , CO, and 02. Gas monitoring at the boiler
outlet was done simultaneously with the sampling efforts at the ESP inlet
and outlet ducts.
Mechanical Collector Fly Ash
The mechanical collector hoppers were emptied at the start of each
test and at the end of a test run. Dry fly ash samples were taken from
a sampling port located in the bottom of each of the eight mechanical
collector hoppers. A tubular thief sampler was used to collect the sam-
ples at 30-second intervals as each of the eight hoppers were emptied.
Samples were composited for each hopper in plastic bags. The ash samples
representing each hopper were composited into a daily sample, based on
the percentage of the total mechanical ash collected in a given hopper,
by taking weighed portions from each bag. The percentage of total ash
in a hopper was estimated by the percentage of the total sluicing time
required to empty a given hopper. These composite samples were sealed
in a glass jar.
14
-------
TABLE 4. PROCEDURE FOR PRESERVATION OF WATER SAMPLES
Constituent
Sample Container
Sample Preparation
and Preservative
Total species
Al, Sb, Ba, Be, Cd, Ca, Cr,
Co, Cu, Fe, Pb, Mg, Mn, Ni,
B, Cl, F, Na, S04, SS
500-ml glass bottle
(HC1 clean)
1-quart plastic cubitainer
(new)
Sample was put in container that
had been predosed with 2-ml of
1+1 HN03.
Stored at 4°C.
B. Dissolved species
Al, Sb, As, Ba, Be, Cd, Ca,
Cr, Co, Cu, Fe, Pb, Mg, Mn,
Hg, Ni, K, Se, Ti, V, Zn
B, Cl, F, Na, S04, Si02, TDS
500-ml glass bottle
(HC1 clean)
125-ml plastic bottle
(HC1 clean)
Sample was filtered through
0.45 |Jm millipore filter pad and
then put in the sample container
that had been predosed with 2-ml
of 1+1 HN03.
Sample was filtered through
0.45 pro millipore filter pad and
then put in the sample
container.
-------
Also, composite samples of the first four days and of the last four
days were taken.
Electrostatic Precipitator Hopper Ash
Prior to each test the ESP hoppers were emptied. The large volume
of fly ash collected during the test necessitated emptying the hoppers
at 2-hour intervals during the daily test period. Dry fly ash samples
were collected from sampling ports located in the bottom of each of the
12 hoppers. A tubular thief sampler was used to collect the samples at
a rate of one per minute for each hopper as they were emptied. These
samples were composited in a manner identical to the mechanical collec-
tor samples so that there were four composite samples (one for each
sampling period) at the end of an 8-hour day. As mentioned previously,
two composite samples were taken at the hoppers — one over the entire ESP,
and the other over only the "B" side of the ESP.
In addition to the dry ESP fly ash samples, a slurry sample was col-
lected at the ash pond. Two 100 ml samples were taken at 15-second inter-
vals as the fly ash slurry flowed from the end of the sluice pipe. The
samples were composited during each sluicing period in 30-gallon plastic
containers and allowed to settle. After 45 minutes of settling, a water
sample was siphoned from the top of the container and analyzed in the
field for pH, alkalinity, ORP, and conductivity. Also, a portion of the
sample was filtered through a 0.45-fJm millipore filter and preserved for
laboratory analysis. After 14 hours of settling, the supernatant was
siphoned from the settled solids and the total volume was measured. Sam-
ples of this 14-hour supernatant underwent the same field analysis and
handling as the 45-minute supernatant did. The settled solids were placed
in a plastic bag and stored for future analysis.
Electrostatic Precipitator Inlet and Outlet Ducts
A number of different sampling techniques were utilized in sampling
the ESP inlet and outlet ducts. A description of the equipment used can
be found in Appendix A.
Raw Sluice Water
Raw sluice water samples were collected for each daily test by set-
ting a valve, located in the sluice water supply line, to drip at a rate
sufficient to fill a 5-gallon cubitainer during each daily test period.
A sample of the inflow water was analyzed in the field for pH, alkalinity,
ORP, and conductivity. Two additional samples, one for suspended solids
and the other for dissolved solids, were taken and sealed in sample jars.
The dissolved solids sample was filtered through a 0.45-Um millipore
filter before being sealed in the sample jar.
16
-------
METHODS OF CHEMICAL ANALYSIS
A number of different chemical analyses were performed on the vari-
ous samples taken during this test. The techniques used for the quanti-
tative determinations were based on:
1. Atomic Emission Spectroscopy
2. Atomic Absorption
3. Gravimetry
4. Potentiometric Titration
5. Ion Chromatography
6. Specific Ion Electrode
7. Differential Pulse Anodic Stripping Voltametry
8. Colorimetry
9. Titrimetric Analysis
10. Turbidmetric Analysis
The procedures used are described in detail in Appendix B.
As a check on the chemical analysis procedures, certified NSB Stand-
ards 1632 coal and 1633 fly ash were analyzed in parallel with the test
samples. (These data are presented in Appendix B.) Appropriate blanks
were run on the various collection substrates to correct for trace ele-
ments contained in the substrates. An aliquot of each liquid reagent
used in each sampling train was analyzed for trace elements. In each
case the value obtained for an element in the blank was subtracted from
the corresponding element in the test sample. If the sample value was
not at least twice the value of the blank, the difference was reported
as a "less than" number for that element. Semiquantitative analysis,
based on spark source mass spectrometry for trace elements in some of
the fine particulate impactor samples, was performed by Accu-Labs, Inc.,
so that a comparison of methods and results could be made.
17
-------
SECTION 5
RESULTS AND DISCUSSION
The results obtained from these studies are presented in the following
terms:
Boiler Operating Conditions for Test
Coal Analysis
Characterization of Waste Streams
Mass Balance
BOILER OPERATING CONDITIONS FOR TEST
During each test, all data on the unit operating conditions were
obtained from the control room instrumentation on an hourly basis and
were averaged for each parameter (Table 5). It can be seen that the unit
was operated at essentially constant conditions with relatively minor
variations in operating conditions. Initial study plans called for con-
ducting the tests under various boiler operating conditions; but cracks
in the turbine spindle, discovered during annual outage, necessitated
operating the boiler at "near full" load during the entire 2-week
testing period.
COAL ANALYSIS
Since the composition of coal varies greatly from source to source
and even within a single source (coal bed), the composition of ash and
stack emissions can also vary greatly. Coal burned during the test
period was mined from the coal fields of western Kentucky. The coal
consisted of carbonaceous matter and a mixture of various minerals
(shales, clays, sulfides, and carbonates).
Proximate analysis of the whole coal and pulverized coal samples is
summarized in Table 6. The results were essentially the same for both
coal samples. Ultimate analysis of the pulverized coal is summarized in
Table 7. In the ultimate analysis, oxygen is calculated by subtracting
the sum of the percentages of moisture, ash content, carbon, hydrogen,
and nitrogen from 100 percent.
The ash from the proximate analysis was analyzed for nine major
elements and the results are given as oxides of the elements in Table 8.
The values for all oxides, except Si02, were obtained using standard ana-
lytical procedures. It was assumed that the nine oxides in Table 8 com-
prised 99.5 percent of the total residue. Since Si02 is extremely dif-
ficult to measure chemically, SiO£ was determined by subtracting the sum
of the other eight oxides from 99.5 percent.
The coal samples were analyzed for the trace element listed in Table 3.
These data were combined with the proximate and ultimate analysis to provide
an additional check on the balance of elements in the coal. The sum of the
18
-------
TABLE 5. BOILER OPERATING CONDITIONS FOR TEST
(AVERAGE VALUES)
Test
Load (MW)
Bar. Pressure (in. of Hg)
Air Flow 106 (Ib/h)
Steam Flow 106 (Ib/h)
Pulverizer A (Amps)
B
C
D
E
F
FD Fan A (Amps)
ID Fan A
FD Fan B
ID Fan B
Furnace Draft A (in. of H20)
Furnace Draft B
Air Heater Diff. A (in. of H20)
B
Gas Leaving Air Heater A (°F)
Gas Leaving Air Heater B (°F)
1
183
29.71
1.28
1.28
49
40
0
47
44
45
55
152
60
161
-.43
-.45
3.06
3.04
342
343
2
179
29.79
1.27
1.13
47
32*
32*
32*
43
48
56
148
60
161
-.59
-.63
3.1
3.0
344
345
3
188
29.76
1.35
1.34
12.3*
32*
45
50
46
46
57
159
60
170
-.65
-.55
3.3
3.2
345
340
4
181
29.63
1.25
1.28
49
0*
46
47
45
44
55
153
58
161
-.46
-.46
3.0
3.0
341
352
5
182
29.68
1.27
1.30
50
0*
45
44
49
48
55
155
58
158
-.54
-.54
3.1
3.0
344
348
6
184
29.73
1.27
1.31
46
29*
44
41
49
42
56
154
58
155
-.51
-.50
3.0
2.8
344
345
7
188
29.70
1.30
1.32
48
41
48
46
0*
45
57
158
60
163
-.54
-.54
3.3
3.4
344
344
8
183
29.74
1.27
1.29
49
0*
48
41
45
46
56
156
60
161
-.58
-.58
3.1
3.3
343
340
*Indicates pulverizer was off for one or more sampling periods.
19
-------
TABLE 6. PROXIMATE ANALYSIS OF WHOLE COAL
Test
Avg.
°/0 Moisture
% Ash
% Volatile
% Carbon
% Sulfur
Btu/lb as
Btu/lb Dry
Btu/lb A &
Matter
Received
MF
7.9
15.5
37.4
47.1
4.2
11349
12323
14584
5.6
16.4
36.8
46.8
4.0
11590
12278
14686
5.5
14.3
38.5
47.2
3.8
11922
12616
14719
5.4
15.0
37.6
47.4
4.0
11813
12487
14685
6.4
16.4
37.0
46.6
5.2
11488
12273
14677
7.0
15.5
37.5
47.0
4.5
11536
12404
14675
7.4
15.6
37.8
46.6
4.2
11456
12372
14659
5.8
15.7
37.7
46.6
3.9
11636
12352
14651
6.4
15.6
37.5
46.9
4.2
11599
12388
14667
PROXIMATE ANALYSIS OF PULVERIZED COAL
Test
8
Avg.
% Moisture
I Ash
I Volatile
% Carbon
% Sulfur
Matter
1.1
15.8
37.2
47.0
3.8
2.6
15.4
37.7
46.9
3.9
2.3
13.8
38.4
47.8
3.6
2.1
14.8
37.8
47.4
3.9
2.8
15.1
37.6
47.3
4.1
2.6
14.8
37.4
47.8
3.9
2.6
16.6
37.7
45.7
3.9
2.3
16.7
38.1
45.2
3.9
2.3
15.4
37.7
46.9
3.9
Btu/lb as Received
Btu/lb Dry
Btu/lb A &
MF
12296
14602
12300
14539
12550
14562
12472
14640
12488
14712
12521
14693
12150
14574
12142
14574
12365
14612
Test
TABLE 7. ULTIMATE ANALYSIS OF PULVERIZED COAL
Avg.
% Carbon
% Hydrogen
% Nitrogen
% Oxygen*
67.7
4.8
1.5
6.4
68.2
4.8
1.5
6.2
69.5
4.8
1.6
6.6
68.9
4.8
1.5
6.1
68.9
4.8
1.5
5.6
69.3
4.8
1.5
5.7
67.1
4.7
1.5
6.2
67.1
4.7
1.5
6.1
68.3
4.8
1.5
6.1
*Note: Percent oxygen is obtained by difference and was not directly analyzed.
20
-------
Test
TABLE 8. MAJOR ELEMENTS (GIVEN AS OXIDES) OF ASH PRODUCED IN
THE LABORATORY BY FIRING COAL AT APPROXIMATELY 800°F
(Percent)
SiOc
A1203 Fe203
CaO
MgO
SO,
Na20 K20 Ti02 Total
1
2
3
4
5
6
7
8
48.7
48.8
49.7
49.1
49.1
49.1
50.8
50.3
17.9
17.6
18.3
17.8
17.1
17.1
17.7
17.3
20.0
20.9
20.3
21.4
21.8
21.4
19.5
20.2
4.8
4.5
4.0
3.9
4.2
4.3
4.3
4.5
1.1
1.1
1.0
1.1
1.1
1.1
1.1
1.1
2.9
2.6
2.3
2.2
2.4
2.7
1.9
2.0
0.6
0.6
0.5
0.6
0.6
0.6
0.6
0.5
2.6
2.5
2.5
2.5
2.4
2.4
2.7
2.7
0.9
0.9
0.9
0.9
0.8
0.8
0.9
0.9
99.5
99.5
99.5
99.5
99.5
99.5
99.5
99.5
TABLE 9. CHEMICAL BALANCE MOUND THE PULVERIZED COAL USING TRACE
ELEMENT ANALYSIS AND ULTIMATE ANALYSIS OF COAL
Test
1234567
Sum of 27 Element
Concentrations (%) in
Pulverized Coal* 12.14
Carbon-«lf
Hydrogen**
Nitrogen**
Oxygen**
Moisture**
Total
10.96 10.26 10.42
11.25 10.99
67.7
4.8
1.5
6.4
1.1
68.2
4.8
1.5
6.2
2.6
69.5
4.8
1.6
6.6
2.3
68.9
4.8
1.5
6.1
2.1
68.9
4.8
1.5
5.6
3.8
69.3
4.8
1.5
5.7
2.6
67.1
4.7
1.5
6.2
2.6
67.1
4.7
1.5
6.1
2.3
93.64 94.26 95.06 93.82
93.35 92.69
*Sum of 27 element concentrations in pulverized coal samples.
**From ultimate coal analyses (Table 7).
21
-------
trace elements, carbon, hydrogen, nitrogen, oxygen, and moisture is shown
to represent 94 percent of the total coal (Table 9).
The results of the trace element analysis of coal are reported in
Table 10. The three predominate elements in coal, excluding sulfur, are
silicon, aluminum, and iron; these 3 elements comprise approximately 86
percent of the total mass of elements. Calcium, magnesium, potassium,
sodium, chlorine, oxygen, and titanium comprise an additional 13.2 per-
cent. The remaining trace elements provide less than 1 percent of the
total mass of elements.
CHARACTERIZATION OF WASTE STREAMS
The characteristics of each trace element in coal cause each ele-
ment to be dispersed to the various ash streams or to volatilize and be
released to the atmosphere in the stack gas following combustion. The
primary objective of this effort was to simply characterize the waste
streams from a utility boiler. For convenience in presenting the data
and discussion of the results, the elements have been divided into the
following five groups:
Group A - Alkali Earth and Alkali Metals (Ba, Ca, Mg, K, Na)
Group B - Halogens (Cl, F)
Group C - Volatiles (those elements most likely to form volatile
combustion products) (Sb, As, Be, B, Hg, Se, S)
Group D - Refractories (those elements most likely to form highly
refractory oxides on combustion) (Al, Cd, Co, Cr, Cu, Fe,
Mn, Ni, Si, Ti, V, Zn)
Group E - Heavy Metals (Pb)
Coal and Ash
The concentration of a number of trace elements was investigated in
each of the daily test samples of pulverized coal, pyrites, and the ash
samples. The results are reported in Tables 10 and 11. As one might
anticipate, the major constituents of the ash were iron, aluminum, and
silicon. Following these three major constituents in order of abundance
were calcium, potassium, magnesium, sulfur, titanium, and sodium.
Ash Sluice Water
To provide background water quality data, inflow water samples were
collected during each daily test. All bottom ash samples and eight ESP
fly ash samples were collected as slurry as they were being sluiced to
the ash pond. After a 45-minute settling period, a water sample was
siphoned from the top of the samples. Following the 14-hour settling
22
-------
TABLE 10. TRACE ELEMENT CONCENTRATION IN COAL, PYRITES, BOTTOM ASH, AND MECHANICAL COLLECTOR ASH ((Jg/g)
ro
co
?st
1
2
3
4
5
6
7
8
Ba
120
100
100
<50
<50
<50
100
270
Group A
Ca
5040
6420
4600
4120
4090
4570
3990
5380
Mg
735
937
655
780
751
773
794
998
K
1800
2300
1600
1900
2100
2100
2000
1700
Na
600
650
500
650
750
650
600
650
Grou]
Cl
1400
1800
1600
1800
1800
1800
1300
1500
3 B
F
50
60
60
10
100
80
160
200
Group C
Sb
1.9
2.3
2.4
2.5
2.1
1.9
2.1
1.8
As
15
15
8
10
17
11
9
9
Be
1.1
1.3
1.1
1.3
1.3
1.2
1.6
1.3
B
.
-
-
-
-
-
-
-
Hg
0.32
0.27
0.25
0.27
0.22
0.23
0.23
0.28
Se
2.9
3.3
2.7
3.4
2.8
2.9
3.0
3.4
S
WHOLE
_
-
-
-
-
-
-
-
Al
COAL
16000
16000
14000
14000
14000
11000
14000
16000
Cd Co
<5 <5
<5 <5
<5 <5
<5 <5
<5 <5
<5 <5
<5 <5
<5 <5
Cr
17.4
20.2
14.7
19.8
19.1
18.4
20.6
18.9
Cu
8.7
10.3
8.0
9.6
9.9
5.9
5.4
6.7
Fe
17900
22800
16600
18200
24000
20700
20200
21100
Group D
Mn
41.0
53.8
34.8
38.8
42.7
46.4
34.2
41.4
Ni
8.5
9.4
7.3
8.1
11.2
12.0
8.7
11.0
Si
30900
31100
25300
27100
30200
25200
29800
33900
Ti
638
835
658
700
703
707
711
677
V
52.6
64.0
59.0
75.5
74.2
68.2
94.1
66.3
Zn
2045
3065
3872
685
1707
2167
4693
1970
Group E
Pb
17.7
21.8
15.4
18.0
24.1
28.0
21.6
28.9
PULVERIZED COAL
1
2
3
4
5
6
7
8
180
95
54
70
85
100
40
69
4220
3370
2870
2640
2890
2810
4160
4040
941
748
676
708
661
620
1060
1050
3530
3000
2640
2910
2810
2460
3600
3770
417
375
319
373
425
275
275
319
1800
1900
1600
2000
2000
1800
1600
1600
41
10
36
<5
46
16
22
40
2.0
2.1
1.9
2.9
2.4
2.3
2.5
2.2
12
16
24
20
19
16
19
21
1.40
1.01
1.30
1.15
1.17
0.94
1.59
1.43
90
80
100
90
90
90
90
90
0.15
<0.10
<0.10
0.10
0.12
0.12
0.10
0.13
4.0
4.7
3.7
4.7
4.4
4.5
5.5
3.1
35000
34000
34000
36000
37000
35000
35000
37000
17200
15000
14700
14500
14000
14500
15000
15200
1 25
<1 36
1 <3
<1 18
<1 18
<1 10
<1 17
<1 15
25.7
22.9
21.6
20.5
22.5
16.3
29.6
24.3
18.6
12.5
11.0
10.8
11.6
10.1
13.0
12.8
22500
23000
20600
21100
24000
27100
24000
22500
55.1
47.1
37.7
38.3
47.2
41.4
49.8
46.2
18
17
14
15
17
15
14
15
34300
27000
24000
22800
32000
25500
26500
23000
731
641
595
630
601
490
728
719
76.2
69.1
68.1
69.5
75.6
57.2
76.1
77.9
54.0
45.1
38.5
41.1
45.4
39.1
54.4
36.8
18
19
14
16
19
20
17
43
PYRITE
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
985
735
465
180
185
605
270
200
935
1140
715
780
710
1050
855
550
9270
8280
6650
4670
6520
8370
9240
8700
30900
26300
19900
31700
22500
20200
14700
18700
1400
1810
1500
1090
1470
1740
1850
1730
6670
5580
4050
5650
5000
4740
3410
4360
3960
5530
4260
3350
3960
5090
5260
5600
21200
18200
15700
16500
16700
16200
11900
18500
466
457
275
325
325
310
564
838
2620
2720
2580
2430
2760
2480
2920
2920
53
10
45
37
45
62
67
34
17
52
26
75
39
17
28
26
56
28
56
38
50
50
56
80
<5
<5
<5
<5
<5
<5
<5
<5
2.6
3.0
1.5
3.4
1.8
2.7
2.3
2.2
4.2
3.2
4.3
5.0
4.5
3.7
5.2
3.0
120
140
110
110
77
72
55
70
4
5
8
15
12
7
8
8
1.09
1.88
1.01
0.89
1.12
1.72
1.33
1.18
8.00
6.96
6.37
6.88
6.61
6.14
3.66
5.42
50
50
50
50
40
50
50
50
230
220
250
180
250
200
220
250
0.68
0.77
0.95
0.75
0.85
0.58
1.0
1.1
<0.10
<0.10
<0.10
<0.10
0.10
<0.10
<0.10
<0.10
7.1
6.5
7.4
10.0
10.0
9.1
21.0
17.0
0.5
0.4
0.9
1.0
0.7
0.7
0.5
0.6
262000
164000
404000
102000
301000
341000
149000
318000
BOTTOM
3600
2000
5200
3600
5600
3200
2200
3100
MECHANICAL
1
2
5
6
7
8
960
1230
1060
1350
935
1480
33000
33900
16200
24600
36800
37500
4610
4650
2140
4810
4800
4850
14400
13900
10000
13100
14100
15400
2020
2220
2020
2180
2120
2080
46
38
37
40
36
36
<5
<5
<5
<5
<5
<5
6.5
4.3
2.7
3.3
5.2
4.3
14
18
14
15
16
12
5.29
5.69
4.10
6.24
5.70
5.48
180
210
200
210
210
220
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
2.0
2.4
2.5
2.1
2.0
2.8
5200
5600
4700
4400
4100
4300
19100
18800
19500
18000
17500
15500
19100
29400
ASH
95000
98000
90000
76800
81800
83300
85000
88600
<1 49
<1 52
<1 33
<1 38
2 44
2 35
<1 44
<1 44
<1 40
<1 49
<1 35
<1 51
<1 41
<1 52
<1 39
<1 45
38.9
60.2
36.9
53.3
53.1
57.4
41.5
55.1
164
152
141
139
154
206
120
115
29.7
41.4
40.5
26.4
36.8
31.3
28.2
32.6
73.3
67.4
58.3
69.9
68.8
64.3
38.4
55.1
185000
240000
265000
210000
251000
215000
230000
205000
190000
195000
200000
183000
192000
227000
195000
172000
136
142
120
95.5
155
397
182
152
597
352
255
311
373
342
181
224
28
84
71
51
93
57
46
63
100
100
94
100
110
93
100
114
29900
34600
30500
31000
31500
27500
40700
65700
159000
136000
134000
159000
151000
132000
152000
142000
735
1070
857
686
801
1050
975
994
6960
4000
3610
3850
4020
3550
2250
3640
66.6
141
72.6
77.1
74.9
114
88.3
66.0
463
428
377
429
437
426
214
316
279
255
142
85.6
101
193
267
126
258
213
138
234
224
177
107
122
102
166
222
129
199
177
162
156
95
92
73
111
101
93
84
81
COLLECTOR
70000
75000
75000
75000
75000
75000
<1 42
<1 66
<1 56
<1 51
<1 56
<1 62
123
135
79.9
155
128
130
70.6
71.2
43.1
78.7
63.5
71.3
260000
260000
265000
265000
290000
265000
377
366
237
647
420
354
107
108
103
115
108
84
106000
112000
124000
124000
112000
123000
3250
3350
2590
3390
3290
3470
400
395
297
428
383
374
160
200
136
232
163
170
97
118
96
102
98
95
-------
TABLE 11. TRACE ELEMENT CONCENTRATION IN ESP INLET PARTICULATE, ESP HOPPER ASH AND ESP OUTLET PARTICULATE (klg/1)
Group A
Test
I
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Ba
1300
1010
710
810
980
980
590
550
1100
1045
670
660
1000
980
830
720
1180
1080
640
885
935
1000
640
635
168
51
84
40
21
79
26
88
Ca
30400
26700
22800
23100
26500
24700
18900
24100
40800
40950
33500
35600
32600
35400
31200
37800
26900
27900
25700
21000
20500
31300
25250
24000
7750
2815
5930
2890
2610
6240
6500
6200
Mg K
5970 16100
5725 17800
5150 20300
5170 10400
5950 22100
5670 19800
4810 17800
5840 23600
6990 17100
6390 17450
5820 17900
6190 18300
5970 19100
5640 15100
5600 12300
7020 18200
6160 22200
6230 22300
5760 21200
5700 21200
4950 20500
6690 20800
5880 21500
5870 23200
2100 2750
783 2280
1200 1590
802 1190
352 978
1620 5120
1220 3500
1530 3940
Na
4600
4500
4600
4400
4800
4900
4200
4500
3800
3750
4000
4200
3900
4400
4100
4100
2820
3420
3320
3520
3680
3280
3480
3480
.
-
_
_
-
-
_
-
Group B
Cl
6900
4550
2800
2800
5100
2900
2500
3800
100
200
100
100
100
100
100
100
37
47
22
50
43
37
27
96
2715
3418
29567
14083
3339
4493
36223
3749
F
420
400
260
245
310
245
230
300
70
75
100
70
10
70
50
20
<5
40
<5
<5
<5
<5
<5
<5
155
160
527
159
148
257
289
214
Sb
18
24
17
12
19
31
15
19
10
13
15
18
10
15
14
16
11
20
16
12
12
14
7.
12
60
106
65
90
114
119
101
40
As
83
90
99
82
100
100
79
81
25
32
13
78
33
34
180
64
45
57
78
60
80
75
.3 64
57
366
726
550
1295
534
1005
660
938
Group C
Be B Hg Se S Al Cd Co
ELECTROSTATIC PRECIPITATOR INLET
8.9 - 0.38 19 - 100000 15 44
10.2 - 0.63 14 - 100000 14 44
10.4 - 0.75 17 - 100000 13 41
9.4 - 0.38 20 - 100000 17 46
10.7 - 0.58 19 - 97000 13 47
10.7 - 0.52 19 - 100000 13 42
10.3 - 0.60 15 - 100000 11 46
10.4 - 0.67 19 - 100000 14 46
ELECTROSTATIC PRECIPITATOR HOPPER ("B" SIDE)
11.4 - 0.22 4.0 - 91000 9 24
9.0 - 0.69 3.8 - 96500 7 28
9.4 - 0.10 6.0 - 96000 5 33
8.6 - 0.10 3.3 - 99000 5 35
9.4 - 0.90 4.0 - 100000 6 60
9.4 - 0.48 5.0 - 94000 5 43
10.3 - 0.10 4.7 - 96000 11 56
10.6 - 0.88 6.3 - 96000 5 62
ELECTROSTATIC PRECIPITATOR HOPPER (ENTIRE UNIT)
8.48 500 <0.10 3.1 5700 100000 2 15
8.80 480 <0.10 3.4 5600 105000 3 44
9.52 480 <0.10 4.2 4900 90000 4 48
8.42 470 <0.10 3.5 4500 100000 4 50
7.66 500 <0.10 4.1 5100 100000 8 44
9.75 450 0.17 4.2 5300 90000 2 36
9.06 450 <0.10 3.7 4600 100000 4 30
8.07 440 0.17 -4.5 4900 95000 2 50
ELECTROSTATIC PRECIPITATOR OUTLET
2.5 - 1.16 3-8 - 13600 53 829
1.4 - 5.74 1.8 - 4670 37 1100
1.8 - 16.0 1.5 - 7400 86 491
1.3 - 0.54 5.5 - 4430 41 1200
1.1 - 0.85 7.8 - 1430 30 1000
3.0 - 0.78 6.4 - 10200 99 608
3.2 - 17.0 22.9 - 11200 56 445
2.4 - 13.6 22.5 - 12500 45 413
Group D
Cr
567
739
475
366
507
515
606
722
152
144
139
135
142
200
94
141
173
167
151
163
150
179
151
144
28200
39050
22400
31600
47400
39200
35600
12900
Cu
82
100
92
88
99
92
83
86
79
76
74
71
73
68
46
73
90.9
105
86.6
86.4
79.5
101
83.0
80.0
269
228
643
300
390
1200
1600
495
Fe
111000
116500
113000
108000
113000
108000
98100
91300
151000
148000
150000
155000
153000
149000
109000
149000
130000
135000
150000
150000
155000
169000
150000
150000
190000
220000
92000
210000
200000
100000
77000
77000
Mn
468
414
343
340
470
421
276
313
424
423
350
372
384
433
328
342
338
356
304
290
341
439
295
264
3600
5025
2460
5340
5430
2080
1630
1120
Ni
137
196
204
138
168
167
189
221
85
83
80
76
84
72
53
78
100
110
106
97
104
107
88
109
8540
8890
9660
13400
21100
17840
19300
6190
Si
188000
190500
191000
191000
192000
192000
195000
204000
201000
191000
189000
195000
196000
201000
195000
199000
138000
141000
144000
150000
138000
164000
150000
138000
-
-
-
-
-
-
-
-
Ti
5340
5490
5280
4950
5650
5240
5300
5580
5990
5640
5430
5480
5520
4950
3020
6320
4770
4750
4400
4760
4260
4800
4410
4460
628
227
266
146
66
471
514
618
V
477
548
557
526
575
612
483
524
541
494
489
481
580
531
328
522
472
505
508
508
449
578
459
441
171
122
121
125
123
210
203
152
Group E
Zn
579
632
564
633
733
666
449
545
556
515
428
443
478
506
381
454
367
410
462
373
337
542
376
316
476
597
781
692
960
1540
959
810
Pb
84
78
61
100
113
96
45
36
75
82
70
62
75
75
66
81
140
140
135
120
150
152
114
127
41
42
74
56
38
95
85
59
-------
period all the water was siphoned from the settled solids, the decanted
water volume was measured, and a sample was retained for analysis.
Results of the field and laboratory water analyses are reported in
Tables 12, 13, and 14.
This sampling scheme was used to evaluate the leaching character-
istics of the major and minor trace elements in ash. Bottom ash
(Table 12) had little effect on the raw sluice watgr trace element con-
centration (Table 13). Increases in the Ca and 864 content were most
noticeable with minimum increases in Cl, F, As, B, Al, and Si. During
the bottom ash sluicing, Zn and Cu appeared to have been partially removed
from the dissolved portion of the sluice water.
The effect of ESP ash (Table 13) on the raw sluice water dissolved
trace elements was generally the same as the effect of the bottom ash±
Constituents showing the largest increases were Ca, K, Na, Al, and 804,
Cl, F, As, B, Se, Al, and Cr showed slight increases. Mg, Cu, and Si
appear to have been removed from the dissolved portion of the sluice
water during the fly ash sluicing.
It should be noted that there were essentially no differences between
the respective 45-minute and 14-hour supernatant samples for the concentra-
tions of Ca, Mg, 804, Si, and Hg, suggesting that any chemical reactions
between the ash and the dissolved portion of the 14-hour sluice water
sample occurred during the sluicing operation (~ 7-minute flow time) for
the elements measured (see Table 12).
The field analysis of sluice waters (Table 14) shows similar results
in that the respective 45-minute and 14-hour samples displayed essentially
the same characteristics. The ESP ash had a much more pronounced effect
on the sluice water quality parameters determined in the field than did
the bottom ash.
Fine Particulate Characterization
Physical Characterization
The size distribution, number, and mass concentration of the fine
particulates were determined for particles between 0.005 and 10 pro by
using four different instruments. Aerodynamic cascade impactors measured
particles with aerodynamic diameters between 0.5 to 10 |Jm. Particles
having diameters between 0.3 to 1.5 pm were counted with an optical
counter. Smaller particles having nuclei-size particles were counted
using the diffusion principle with a condensation nuclei counter. Size
fraction measurement, particle color, and aspect ratio from the aero-
dynamic cascade impactors were confirmed by Walter C. McCrone Associates,
Inc., with the use of a light transmission microscope and a scanning
electron microscope.
25
-------
Group A
TABLE 12. TRACE ELEMENT CONCENTRATION IN BOTTOM ASH AND ELECTROSTATIC PRECIPITATION SLUICE WATER SAMPLES (mg/1)
Group C
st
Ba Ca
Mg K
Na
Cl F
Sb As
Be
B
BOTTOM
1
2
3
4
5
6
7
8
<0.1 37
<0.1 38
<0. 75
<0. 65
<0. 28
<0. 34
<0. 33
<0. 29
2.9 1.9
3.4 1.8
3.5 1.9
3.6 1.8
3.6 1.8
3.3 1.8
3.2 1.7
3.0 1.9
5.4
5.5
5.4
5.6
5.9
5.6
5.7
6.2
13 0.10
7 0.08
8 0.13
10 0.14
13 0.09
11 0.08
14 0.09
11 0.07
<0.1 0.016
<0.1 0.013
<0.1 0.009
<0.1 0.009
<0.1 0.009
<0.1 0.009
<0.1 0.009
<0.1 0.011
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.30
0.20
0.20
0.30
0.16
0.15
0.25
0.28
BOTTOM
1
2
3
4
5
6
7
8
33
35
64
57
26
32
32
27
2.9
3.4
3.5
3.6
3.6
3.3
3.2
3.1
Hg
ASH SLUICE
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
ASH SLUICE
0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
ELECTROSTATIC PRECIPITATOR
1
2
3
4
5
6
7
8
0.3 660
0.3 520
<0.1 440
<0.1 410
0.3 460
<0.1 420
0.2 490
0.4 510
0.02 23
0.02 21
0.02 23
0.03 21
0.02 21
0.03 20
0.02 21
0.02 23
20
18
19
19
20
19
18
19
13 0.53
11 0.46
11 0.44
12 0.42
12 0.40
13 0.49
10 0.53
10 0.61
<0.1 0.006
<0.1 C0.002
<0.1 0.005
<0.1 <0.002
<0.1 0.009
<0.1 <0.002
0.2 0.002
0.2 0.008
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
8.2
8.2
8.0
7.0
7.0
5.5
4.0
7.0
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
Se S04 Al Cd Co
WATER (DISSOLVED) - SETTLING TIME
<0.001 71 0.5 <0.001 <0.005
<0.001 57 0.3 <0.001 <0.005
0.002 130 0.3 <0.001 <0.005
0.003 120 0.3 <0.001 <0.005
<0.001 43 0.2 <0.001 <0.005
<0.001 76 0.3 <0.001 <0.005
<0.001 76 0.4 <0.001 <0.005
<0.001 82 0.3 <0.001 <0.005
WATER (DISSOLVED) - SETTLING TIME
74
51
110
74
40
63
82
68
Cr
- 14 H
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
- 45 MIN
SLUICE WATER (DISSOLVED) - SETTLING TIME
0.033 490 0.4 <0.001 <0 . 005
0.040 580 1.2 <0.001 <0.005
0.040 490 3.6 <0.001 <0 . 005
0.028 440 4.0 <0.001 <0.005
0.023 560 2.8 <0.001 <0.005
0.050 520 3.9 <0.001 <0.005
0.042 620 1.8 <0.001 <0.005
0.058 740 1.4 <0.001 <0.005
0.098
0.060
0.063
0.072
0.075
0.079
0.068
0.060
ELECTROSTATIC PRECIPITATOR SLUICE WATER (DISSOLVED) - SETTLING TIME
1
2
3
4
5
6
7
8
640
530
430
390
450
400
480
510
0.01
0.02
0.02
0.03
0.02
0.04
0.03
0.03
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
<0.0002
590
460
500
420
640
550
500
480
£. ~
Cu
<0.03
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
- 14
0.01
0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
- 45
Fe
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
H
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
MIN
Mn Ni
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
<0.01 <0.05
Si Ti V An
3.6 <1 0.4 0.01
3.1 <1 0.2 0.01
2.9 <1 <0.1 0.01
2.7 <1 <0.1 0.01
2.6 <1 <0.1 0.02
3.1 <1 0.2 0.01
3.2 <1 <0.1 0.01
3.2 <1 <0.1 0.01
3.5
2.9
2.7
2.6
2.4
3.0
2.9
3.0
0.56 <1 <0.1 0.06
0.71 <1 <0.1 0.03
0.84 <1 0.5 0.03
0.98 <1 0.1 0.02
0.75 <1 <0.1 0.03
0.98 <1 0.4 0.02
0.75 <1 0.2 0.02
0.93 <1 0.2 0.02
0.71
0.84
0.79
0.89
0.71
0.71
0.47
0.61
Pb
<0.010
<0.010
<0.010
<0.010
<0.010
0.010
<0.010
<0.010
0.012
0.015
0.011
<0.010
<0.010
<0.010
<0.010
<0.010
-------
TABLE 13. TRACE ELEMENT CONCENTRATION IN INFLOW WATER SAMPLES (mg/1)
to
Test
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Ba
<0.002
<0.002
CO. 002
<0.002
<0.002
<0.002
CO. 002
<0.002
CO. 1
CO . 1
cO.l
CO . 1
<0. 1
<0 - 1
<0 - 1
<0 . 1
Ca
0.14
0. 13
0.18
0.11
0.10
0.15
0.18
0. 15
19
18
18
17
17
17
17
17
Mg
0.04
0.06
0.08
0.05
0.04
0.08
0.05
0.06
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
0
0
0
0
0
0
0
0
1 .
1 .
1.
1.
1.
1.
1.
1
K
.04
.04
.06
.05
.03
.05
.06
.04
4
3
3
3
4
3
2
.3
1
0
0
0
0
0
0
0
0
5.
5.
5.
5.
5
5
5
5
Ma
.01
.01
.02
.01
.01
.04
.06
.06
3
0
1
1
.6
.3
.3
.5
Cl
7* 0
7* 0
8* 0
7* 0
7* 0
8* 0
8* 0
7* 0
7 0
7 0
8 0
7 0
7 0
8 0
8 0
7 0
F
.08*
. 06*
.06*
. 06*
. 06*
. 06*
.06*
.06*
.06
.06
.06
.06
.06
.06
.06
.06
i
0
0
0
0
0
0
<0
0
CO
<0
<0
CO
<0
<0
<0
>b
.002
.005
.005
.004
.002
.004
.002
.005
.1
. 1
.1
. 1
.1
. 1
.1
. 1
<0.002
<0.002
<0.002
<0.002
CO.002
<0.002
<0.002
<0.002
Group C
~~ Be B ~
<0.001
<0.001
<0.001
<0.001
CO. 001
<0.001
<0.001
<0.001
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.04*
0 . 04*
0 . 06*
0.09*
0.06*
0.11*
0 . 08*
0 . 04*
0.03
0.04
0.06
0.06
0.06
0.11
0.08
0.04
~Tg Se SO^
<0.
<0.
-------
TABLE 14. FIELD ANALYSIS OF ASH SLUICE WATER SAMPLES
INFLOW DAILY COMPOSITE
Alkalinity as
Test
1
2
3
4
5
6
7
8
Avg.
(units)
7.60
7.35
7.80
7.7
7.9
8.3
8.0
7.9
7.8
Phen.
(mg/D
0
0
0
0
0
0
0
0
Total
(mg/D
48
45
45
45
45
45
45
45
45
ORP Conductivity
(mV) (pmhos)
265 172
260 175
240 155
280 163
340 165
260 170
280 170
320 160
281 166
1
2
3
4
5
6
7
8
9.3
9.0
8.9
8.8
8.8
9.2
9.6
9.4
Avg.
9.1
BOTTOM ASH SLUICE WATER
(Settling Time - 45 min)
14
9
11
10
6
10
12
12
10
44
60
65
67
53
53
47
42
215
160
120
120
245
220
190
210
54
185
285
230
425
350
220
260
290
215
284
(Settling Time - 14 h)
1
2
3
4
5
6
7
8
Avg.
9.5
9.0
8.9
8.9
8.9
9.2
9.3
9.6
9.2
16
9
0
6
8
10
10
8
45
72
56
41
52
50
43
35
49
150
180
160
335
240
245
236
245
224
250
300
272
490
215
260
256
230
284
28
-------
TABLE 14. (Continued)
ELECTROSTATIC PRECIPITATOR SLUICE WATER
Test
1
2
3
4
5
6
7
8
Avg.
(units)
11.6
11.5
11.6
11.3
11.6
11.8
11.8
11.8
11.5
(Settling Time - 45 min)
Alkalinity as CaC03
Phen.
(mg/1)
1020
772
521
456
581
460
676
714
Total
(mg/1)
1084
805
613
506
633
509
725
765
650
705
Conductivity
((jmhos)
4100
15 4000
25 2600
40 2750
95 3250
85 2700
90 3600
85 3400
62 3300
(Settling Time - 14 h)
1
2
3
4
5
6
7
8
11.9
11.9
11.8
11.5
12.0
12.2
12.4
12.6
Avg.
12.0
1064
750
545
607
530
674
757
704
1101
792
595
670
573
722
801
751
10
10
5
90
85
55
60
45
4000
4000
2600
3450
3100
3200
3450
3400
29
-------
Particulate Sizing by Aerodynamic Cascade Impactors
Two types of cascade impactors were employed to collect and aero-
dynamically size fine particulates. The Brink Cascade Impactor (BMS II)
was used to sample the inlet of the ESP. The Andersen Sampler (Mark III)
was used to sample the outlet of the ESP.
Several runs were conducted using the impactors, and four of these
were compiled runs. A compiled run was performed by sampling several
constant velocity points and changing the plate (or plates) in the impac-
tor as it became loaded with particulate matter. The compiled runs
(No. 6 and 7 Brink and Cmp 1 and 2 Andersen) were conducted to accumulate
a sufficient amount of particulate for further analysis.
Brink Cascade Impactor
Seventeen runs using the Brink Cascade Impactor were performed on
the ESP inlet to size particles; but due to power losses on the fifth and
eighth runs, only fifteen runs produced data. Particles in the inlet flue
gas differed greatly in their size distributions depending on whether the
inlet grain loading was "high" or "low." Runs 2, 3, 6, 10, and 14 were
classified as a group characterized by inlet grain loadings of 1.11 grains/
scf or less and identified as having "low" grain loadings. Since runs 1,
4, 7, 9, 11, 12, 13, 15, 16, and 17 had grain loadings of 1.18 grains/scf
and higher, they were considered to have "high" inlet grain loadings
(Table 15).
Both groups of data were fitted individually and as groups to log-
normal and Weibull distributions to determine which distribution better
described the particle size distributions observed. The Weibull distri-
bution was a better fit to the individual runs of the low inlet grain
loading group; but when the runs were combined and analyzed as a group, the
lognormal and Weibull distributions were equally good (Tables 16 and 17).
Similar behavior was noted in the high inlet grain loading groups.
The lognormal and Weibull distributions were compared on the basis of
the R-square statistic, the error mean square (EMS), the F-value, and a
graphical evaluation of the distribution fit to the data. The R-square
statistic indicates the fraction of the total variation explained by the
fitted distribution. For example, run 1 in Table 16 had the lognormal
distribution to explain approximately 85.84 percent of the total varia-
tion, while the Weibull distribution fit explained 89.09 percent. The
EMS is the residual variation left over after fitting the distribution
to the data and is an estimate of the random error in the data. If a
fitted distribution is inappropriate for the data, the EMS is inflated.
The last column o-f Table 16 is the ratio of the lognormal EMS to the
Weibull EMS. The ratio of 1.2976 for run 1 indicates that using the
Weibull distribution reduced the EMS and slightly raised the R-square
statistic. For 14 of the 15 runs, the Weibull fits were better than the
lognormal fits.
Overall, the Weibull distribution seemed to characterize and describe
particle size data better for this set of data. Estimates of the median
30
-------
TABLE 15. TEST SUMMARY OF BRINK SAMPLING PARAMETERS
Start Duration
Run1
1
2
3
4
6
Compi 1 ed
7
Compi 1 ed
9
10
11
12
13
14
15
16
17
Date
8-10-
8-10-
8-11-
8-11-
8-12-
8-13-
76
76
76
76
76
76
8-16-76
8-16-
8-16-
8-17-
8-17-
76
76
76
76
8-17-76
8-18-76
8-18-
8-18-
76
76
Time
12'
2
10:
1
12
1
1
8
9
10
12
13
9
11
12
8
10
13:
10.
12
13
:25
:48
:54
:30
:31
:22
:52
:10
:17
:22
:15
:15
:01
:00
:41
:42
:07
:50
:00
:30
:05
(min)
7
7
7
7
5
2.
2.
5
2.
2.
5
5
5
5
5
5
5
5
5
5
5
^
5
.5
.5
.5
rlO 1 (fCU 1 d I D Ld Ll\ udi-UiiriTLi-J-v ,ji_cii-.i-v
Sampling Stack Gas Composition % Weight Temp Pressure Pressure
. - ---. ~ ~^^ r;TI ^ n." it,,.- fCiT"! ("-in Hn ^ (in W „ f> "l
Inlet to ESP 11.4 7.4 0 81.2 9.0 30.12 29.03 350
Inlet to ESP 11.4 7.4 0 81.2 9.0 30.12 29.03 350
Inlet to ESP 12.3 6.3 0 81.4 8.4 30.22 29.19 350
Inlet to ESP 11.8 6.2 0 82.0 8.5 30.14 29.10 350
Inlet to ESP 11.4 6.8 0 81.8 8.6 30.10 29.06 350
Inlet to ESP 11.8 7.1 0 81.1 8.4 30.17 29.15 350
Inlet to ESP 11.7 6.1 0 82.2 10.0 30.12 28.90 350
Inlet to ESP 11.7 6.1 0 82.2 10.0 30.12 28.90 370
Inlet to ESP 11.7 6.1 0 82.2 10.0 30.12 28.90 370
Inlet to ESP 11.5 7.4 0 81.1 9.6 30.14 28.97 370
Inlet to ESP 11.5 7.4 0 81.1 9.6 30.14 28.97 370
Inlet to ESP 11.5 7.4 0 81.1 9.6 30.14 28.97 370
Inlet to ESP 10.8 7.5 0 81.7 9.2 30.03 28.93 370
Inlet to ESP 10.8 7.5 0 81.7 9.2 30.03 28.93 370
Inlet to ESP 10.8 7.5 0 81.7 9.2 30.03 28.93 370
rometric
ressure
in. HE)
29.
29.
29.
29.
29.
29.
29.
29.
29.
29
29
29
29
29
29
70
70
80
80
76
58
.61
61
.61
.68
.68
.68
.66
.66
.66
Static
Pressure
(in. H2Q)
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
-8.0
Brink Sample
AP Vo 1 ume
(in. Hg)2 (acf)3
1.8 0.
1.8 0.
1.8 0.
1 .8 0.
1.7 0.
1.7 1.
1.7 0.
1.7 0
1.7 0
1.7 0
1.7 0
1.7 0
1.7 0
1.7 0
1.7 0
455
455
455
455
650
.300
.325
.325
.325
.325
.325
.325
.358
.325
.325
Sample Nozzle
Flow Diameter
(acfm)4 (mm)
0.
0.
0.
0.
0.
0.
0.
0
0
0
0
0
0
0
0
065
065
065
065
065
.065
.065
.065
.065
.065
.065
.065
.065
.065
.065
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Grain
Loading
(Grain/scf)
1
1
0
1
0
1
1
0
2.
1 .
1.
0.
1 .
2.
2.
.5591
. 1130
.3779
.5575
.7626
.9066
.1867
.4900
2111
9202
8610
5346
2894
5630
9457
'Runs No. 5 and 8 were destroyed in the field due to power loss.
2AP = pressure difference across the Brinks.
3acf = actual cubic feet.
4acfm = actual cubic feet per minute.
5Grain/scf = grain per standard cubic feet.
-------
TABLE 16. COMPARISON OF THE LOGNORMAL AND WEIBULL DISTRIBUTIONS
FITTED TO THE BRINK IMPACTOR DATA FROM COLBERT
Run No.
1
2
3
4
6
7
9
10
11
12
13
14
15
16
17
Lognormal
R- Square
0.8584
0.9072
0.9457
0.8456
0.8617
0.9036
0.9642
0.8978
0.9406
0.9203
0.8779
0.8945
0.9116
0.9461
0.8706
Error Mean
Square (EMS)
0.1718
0.1126
0.0659
0.1874
0.1678
0.1171
0.0435
0.1226
0.0713
0.0956
0.1465
0.1266
0.1061
0.0647
0.1552
Individual
Runs
Weibull
R-Square
0.8909
0.9518
0.9680
0.8831
0.9173
0.9411
0.9526
0.9436
0.9649
0.9523
0.9184
0.9294
0.9452
0.9779
0.8903
Error Mean
Square (EMS)
0.1324
0.0586
0.0388
0.1419
0.1004
0.0714
0.0576
0.0677
0.0421
0.0573
0.0980
0.0848
0.0658
0.0266
0.1316
Ratio of
Lognormal EMS
To Weibull EMS
1.2976
1.9215
1.6985
1.3206
1.6713
1.6401
0.7552
1.8109
1.6936
1.6684
1.4949
1.4929
1.6125
2.4323
1.1793
TABLE 17. SUMMARY STATISTICS OF CURVE FITS TO THE BRINK IMPACTOR
DATA FROM COLBERT GROUPED RUNS
Inlet Grain Error Mean Error Mean
Loading Square R-Square F-Value Square R-Square F-Value
Low1
High2
0.1427
0.2185
0.8346
0.7373
141.28
162.76
0.1425
0.2204
0.8349
0.7350
141.61
160.84
!Runs 2, 3, 6, 10, and 14.
2Runs 1, 4, 7, 9, 11, 12, 13, 15, 16, 17.
32
-------
particle size for the individual runs and the low and high inlet grain
loading groups are summarized in Tables 18 and 19. Figures 5 through 8
show the sample data and the fitted cumulative distributions for both
the low and high inlet grain loading data.
The median particle diameter is the particle diameter at which the
cumulative mass concentration is 50 percent. Thus, 50 percent of the
mass occurs above this diameter and 50 percent occurs below. After fit-
ting a particle size distribution to the data from the impactor, the
median particle diameter and a 95-percent confidence interval are
estimated.
The median particle size diameters, estimated from the fitted distri-
butions, must be used with caution since the fitted distributions included
the cyclone with its accumulated mass loading. Normally, the cyclone is
omitted from the analysis; but for this particular set of Brink data, a
large fraction of the mass was caught in the cyclone. When the cyclone
was omitted from the analysis, the estimate median particle diameter fell
outside the range of the sample data. Inclusion of the cyclone data
allowed for the estimation, within the range of sample data, of a median
particle size for the low inlet grain loading. The high inlet grain load-
ing group still had the median particle diameter fall outside of the sample
data range, but the estimate becomes more reasonable.
For the low inlet grain loading group, the lognormal distribution
estimate of the median particle size diameter was 4.85 pro, while the
Weibull estimate was 4.88 (Jm. The high inlet grain loading groups which
had the lognormal fit gave an estimated median diameter of 12.87 |Jm and the
Weibull fit gave an estimate of 9.20 pm.
Andersen Cascade Impactor
Seven runs using the Andersen Impactor were analyzed from the outlet
of the ESP to size particles. A summary of sampling parameters is found
in Table 20. The lognormal and Weibull distributions were fitted to the
data to determine which distribution best described the particle size
distributions observed. The lognormal distribution fit the individual
runs and the grouping of the individual runs better than the Weibull dis-
tribution (as shown in Table 21). The median particle diameter estimated
from the lognormal distribution was 2.77 |Jm while the Weibull estimate
was 3.08 [Jm. A significant reduction in the mass and size of the particles
which entered the ESP due to the variation of the inlet grain loadings
was noted. For example, the low grain loading groups from the inlet had
27.39 percent of the particulate less than 2.77 (Jm as compared to the
high grain loading group from the inlet which had only 10.75 percent
below 2.77 [Jm (based on the Weibull fit).
Table 22 summarizes the individual and group estimates of the median
particle diameter for both distributions. Figures 9 and 10 show the sample
data and fitted cumulative distributions for the Andersen data.
33
-------
TABLE 18. ESTIMATED MEDIAN PARTICLE DIAMETER IN MICRONS BASED
ON INDIVIDUAL RUNS OF BRINK IMPACTOR DATA
Lognormal Weibull
Inlet Grain Median Particle Median Particle
Loading Diameter (|Jm) Diameter (|Jm)
Low
Run 2
Run 2 5.69 5.91
Run 3 5.62 5.89
Run 6 4.73 5.30
Run 10 4.50 4.92
Run 14 6.02 6.19
High
Run 1 13.45 11.37
Run 4 13.18 11.05
Run 7 27.16 17.98
Run 9 16.45 11.89
Run 11 26.65 16.74
Run 12 18.12 12.83
Run 13 19.03 14.48
Run 15 18.96 14.26
Run 16 20.20 13.34
Run 17 48.44 26.28
34
-------
TABLE 19. ESTIMATED MEDIAN PARTICLE DIAMETER IN MICRONS
BASED ON GROUPED RUNS OF BRINK IMPACTOR DATA
Lognormal Veibull
Inlet Grain Median Particle 95% Confidence Median Particle 95% Confidence
Loading Diameter (pm) Interval Diameter ((Jm) Interval
Low 4.85 3.86-6.09 4.88 3.88-6.13
High 12.87 9.17-18.05 9.20 6.87-12.32
-------
TABLE 20. TEST SUMMARY OF ANDERSEN SAMPLING PARAMETERS
ON
Run
1
2
3
5
6
7
8
Composite
1
Composite
2
Date
8/10/76
8/11/76
8/11/76
8/12/76
8/13/76
8/13/76
8/16/76
8/16/76
8/17/76
Start
Time
14
09
13:
13:
07:
10:
08:
11:
11:
:37
:30
:30
:35
:30
:45
:05
30
40
Duratioi
(rain)
37
60
70
70
120
120
120
360
360
i Stack Gas
C02
11.7
12.1
11.8
12.4
11.8
11.8
11.7
11.7
11.5
°2.
7.2
6.4
6.2
6.8
7.1
7.1
6.1
6.1
7.4
Composition (%)
CO
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
__N2
81.1
81.5
82.0
80.8
81.1
81.1
82.2
82.2
Rl 1
H20
8.7
7.7
9.7
4.6
8.4
6.2
8.9
8.9
in H
Mole
We
Dry
30.16
30.19
30.14
30.26
30.17
30.17
30.13
30.13
•}n i /.
cular
ight
Wet
29.11
29.25
28.95
29.69
29.42
29.42
29.06
29.06
TO 00
Stack
Temp
325
325
325
330
345
345
350
350
Barometric
Pressure
(in. Hg)
29
29
29
29
29
29.
29,
29.
.70
.80
.80
.76
.58
.58
,61
61
Stack
Pressure
(in. He)
28
29
29
28
28
28,
28.
28.
.90
.00
.00
.96
,78
,78
.81
81
Sample
Volume
(rf)
13
22
27
27
46
44,
45.
140.
.9000
.480
.000
.005
.380
.920
,900
415
Sample
Flow Rate
0.58823
0,
0,
0,
0.
0.
0.
.57880
.59521
55903
59342
56497
61093
0.60395
Sample
Volume
(rl^rfl
13.050
21
24
24
42
40
42
131
.094
.793
.462
.601
.558
.376
. 709
Nozzle
Diameter
(
0
0
0
0.
0.
0.
0.
0.
in. ;
.25
.25
.25
.25
.25
25
25
25
Grain
Loading
Grains/scf
0.0599
0.0523
0.0610
0.0339
0.0304
0.0313
0.0261
0.0275
133.755 0.57248 125.462 0.25
-------
TABLE 21. COMPARISON OF THE IOGNORMAL AND WEIBULL DISTRIBUTIONS
FITTED TO THE ANDERSEN IMPACTOR DATA FROM COLBERT
Lognormal
Weibull
Run No. R-Square
1
2
3
5
6
7
All Runs
Grouped
0.9937
0.9908
0.9923
0.9948
0.9939
0.9924
0.9903
0.9788
Error Mean
Square (EMS)
0.0082
0.0119
0.0101
0.0068
0.0080
0.0099
0.0126
0.0206
R-Square
0.9845
0.9867
0.9911
0.9729
0.9799
0.9841
0.9776
0.9661
Error Mean
Square (EMS)
0.0201
0.0174
0.0116
0.0352
0.0263
0.0206
0.0291
0.0329
Ratio of
Lognormal EMS
To Weibull EMS
0.4080
0.6839
0.8707
0.1932
0.3042
0.4806
0.4330
0.6261
TABLE 22. ESTIMATED MEDIAN PARTICLE DIAMETER IN MICRONS BASED ON
INDIVIDUAL AND GROUPED RUNS OF ANDERSEN IMPACTOR DATA
Run No.
1
2
3
5
6
7
8
All Runs
Grouped
95% Confidence
Interval
Lognormal
Median Particle
Diameter (|Jm)
3.
2.
01
68
3.23
2.
2,
2.
79
32
50
3.02
2.77
2.67-2.87
Weibull
Median Particle
Diameter (pm)
,34
.01
,52
.11
,62
,81
3.32
3.08
2.94-3.23
37
-------
LOGNORMAL DISTRIBUTION
100
80 -
o
V
3*
Ul
60 -
40 -
O
20 -
0
COLBERT INLET
BRINK LOW GRAIN LOADINGS
RUNS 2, 3, 6, 10, 14
I
6 8 10
DIAMETER (MICRONS)
12
14
16
Figure 5. Lognormal distribution plot of runs 2, 3, 6, 10 and 14.
38
-------
WEIBULL DISTRIBUTION
100
80
o
V
LoJ
>
60
20
COLBERT INLET
BRINK LOW GRAIN LOADINGS
RUNS 2, 3, 6, 10, 14
_L
4 6 8 10
DIAMETER (MICRONS)
14 16
Figure 6. Weibull distribution plot of runs 2, 3, 6, 10 and 14.
39
-------
100
LOGNORMAL DISTRIBUTION
80
Q
V
55
LL)
60
40
20
0
050=12.87
COLBERT INLET
BRINK HIGH GRAIN LOADINGS
RUNS I, 4, 7, 9, II, 12', 13, 15,
16, 17
6 8 10 12
DIAMETER (MICRONS)
14
16
Figure 7. Lognormal distribution of
runs I, 4, 7, 9, II, 12, 13, 15, 16 and 17.
40
-------
WEIBULL DISTRIBUTION
100
80
Q
V 60
£
UJ
050 = 9.20
40
ID
o
20
COLBERT INLET
BRINK HIGH GRAIN LOADINGS
RUNS I, 4, 7, 9, II, 12, 13, 15,
16, 17
_L
0
6 8 10 12 14
DIAMETER (MICRONS)
Figure 8. Weibull distribution of
runs 1,4, 7, 9, II, 12, 13, 15, 16 and 17.
41
-------
LOGNORMAL DISTRIBUTION
100
80
Q
V
85
UJ
60
40
O
050 = 2.77.7 •
COLBERT OUTLET
ANDERSEN
RUNS I, 2, 3, 5, 6, 7, 8
J I I
J L
2 46 8 10 12 14 16
DIAMETER (MICRONS)
Figure 9. Lognormal distribution plot for Andersen impactor.
42
-------
WEIBULL DISTRIBUTION
100
COLBERT OUTLET
ANDERSEN
RUNS I, 2, 3, 5, 6, 7, 8
J I
I
I
4 6 8 10 12
DIAMETER (MICRONS)
14 16
Figure 10. Weibull distribution plot for Andersen impactor.
43
-------
Size Distribution of Fly Ash by Light Transmission Microscope
Observation of the fly ash particles, as viewed through a light trans-
mission microscope at a low power (I0-25x), showed the overall color to be
light brown (which was confirmed visually). Approximately 95 percent of
the particles appeared spherical in shape, and 80 percent of these spheres
had an aspect ratio (ratio of average length to average diameter) of 1.0
to 1.5. The remainder of the particulate sample was composed of black,
irregular flakes appearing to be unburned carbon.
Density of Fly Ash by Pycometer
The density of the fly ash sample as determined by a pycometer was
found to be 2.263 g/cc.
Particle Counters
The Climet Model 0208A Optical Counter and the General Electric Model
112L428 Gl Condensation Nuclei Counter were used to count the submicron
particles. The condensate nuclei counters only count the total number
of particles present in a stream; therefore, the Thermo Systems, Inc.,
Model 3040 Diffusion Battery was used to classify the particle sizes
through use of the diffusion principle. Four inlet and four outlet runs
were conducted using the Climet Optical Counter, and seven inlet and seven
outlet runs were performed using the Diffusion Battery Condensation Nuclei
Counter system.
Optical counters are designed to measure low (ambient level) concen-
trations of particles smaller than 1.5 pm. A dynamic dilution system,
designed and built by MRI and consisting of a cyclone with a cutoff diame-
ter of 2.65 (Jm, was incorporated in front of the counters to dilute and
condition the sample continuously. As a result, only particles with
diameters smaller than 2.65 pm were allowed to enter the optical instru-
ments. It was found that particles with a diameter of 0.4 (Jm comprise
80 percent of the total count of particles in the flue gas at both the
inlet and outlet.
A graphical stripping technique was used to reduce the collected data.
A summary of the inlet data is given in Table 23, and a summary of the
outlet data is given in Table 24.
Chemical Characterization of Fine Particulates
Composite samples of a number of runs at both the ESP inlet and outlet
were prepared in order to have a sample large enough to chemically analyze.
Runs were composited based on the coal analysis; i.e., for the tests in
which coal was most similar, the corresponding impactor runs were composited.
Table 25 gives a breakdown of the compositing schedule.
Rather than chemically analyzing each stage from the impactors of the
various runs conducted, the stages were composited according to particle
44
-------
TABLE 23. THE CONCENTRATION OF CONDENSATION
NUCLEI SIZE PARTICLES (INLET RUNS)
% Particle Larger/Smaller
than Stated Diameter (Corresponding Number of Particle/cm3) x 106
Run No. 1 ([jm) Counter Conditions Stack Conditions
> 0.13 19.888 13.647
9% > 0.085 2.034 1.396
3% < 0.085 0.678 0.466
96% > 0.14 16.032 11.310
4% > 0.05 0.668 0.471
94% > 0.14 17.39 12.043
6% > 0.088 1.11 0.769
92% > 0.14 14.72 10.194
8% > 0.06 1.28 0.886
90% > 0.140 13.41 9.288
10% > 0.070 1.49 1.032
90% > 0.175 13.68 9.562
10% > 0.070 1.52 1.062
95% > 0.130 14.725 10.104
5% > 0.085 0.775 1.129
xRun No. 1 was a dry run.
45
-------
TABLE 24. THE CONCENTRATION OF CONDENSATION NUCLEI
SIZE PARTICLES (OUTLET RUNS)
% Particle Larger/Smaller
Run
No.1
2
3
4
5
6
7
8
than Stated Diameter
(pm)
0.200 3
0.040 5
0.015 3
0.010 3
0.050 3
0.010 3
0.055 3
0.040 3
0.015 3
0.010 3
0.055 3
0.025 :
0.010 :
0.060 :
0.015 :
0.060 3
0.010 :
0.005 3
0.060 3
0.015 3
0.005 :
14% 5
- 13% 5
> 61% -
> 9% 5
> 3%
9% >
> 33% ^
> 58% 3
22% 5
> 6% ;
> 19% J
> 13% )
> 40%
10% ;
> n 3
> 24% 3
> 59%
5% 3
> 10% 3
> 85% 3
8% 3
> 39% 3
> 37% 3
> 16%
oo/ >
*H '
> 13% 3
> 4% 3
> 75%
- 0.200
• 0.040
> 0.015
> 0.010
- 0.050
• 0.010
' 0.005
> 0.055
> 0.040
> 0.015
> 0.010
> 0.055
> 0.025
» 0.010
- 0.060
- 0.015
> 0.005
> 0.060
- 0.010
> 0.005
> 0.060
> 0.015
> 0.005
(Corresponding Number of Particle/cm3) x 106
Counter Conditions
1.358
1.261
5.917
0.873
0.291
24.525
89.925
158.050
20.328
5.544
17.556
12.012
36.960
35 . 460
24.822
85.104
209.214
42.030
84.060
714.510
29.904
145.782
138.306
59.808
28.464
46.254
14.232
266.850
Stack Conditions
0.949
0.881
4.136
0.610
0.203
17.144
62.791
110.482
14.468
3.945
12.495
8.549
26.306
24.784
17.349
59.483
146.228
29.649
59.297
504.026
20.710
100.963
95.785
41.421
20.078
32.627
10.039
188.235
1Run No. 1 was a dry run.
46
-------
TABLE 25. COMPOSITING SCHEDULE FOR CHEMICAL ANALYSIS OF PARTICIPATE
SAMPLES FROM THE BRINK AND ANDERSEN IMPACTORS
Inlet Duct - Brink Impactor
Test
Runs
(1, 10, 11
15, 16, 17)
(2)*
(3, 12
13, 14)
(4)*
(9)
Outlet Duct
(1, 5)
(6)*
(2, 3)
(7)
(8)
Total
Sample
Weight, g
0.11766
0.02069
0.00186
0.01850
0.00437
0.00022
0.05146
0.00809
0.00235
0.02420
0.00197
0.00018
2.18
1.91
1.17
- Andersen
0.07790
0.05327
0.02963
0.0327
0.02116
0.03019
0.09556
0.07895
0.03690
0.03601
0.01630
0.03087
0.03509
0.01787
0.01879
Stages
Composited
Oc, 1
2, 3
4, 5
Oc, 1
2, 3
4, 5
Oc, 1
2, 3
4, 5
Oc, 1
2, 3
4, 5
Oc, 1
2, 3
4, 5
Impactor
Oc, 1, 2, 3
4, 5
4, 7, 8
Oc, 1, 2, 3
4, 5
6, 7, 8
Oc, 1, 2, 3
4, 5
6, 7, 8
Oc, 1, 2, 3
4, 5
6, 7, 8
Oc, 1, 2, 3
4, 5
6, 7, 8
Approx.
Particle
Size, pm
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
> 3
1-3
< 1
''Analyzed for Cl and F only.
47
-------
size ranges--3 |Jm--for chemical analysis. Tables
26 and 27 give the results of the fine particulate chemical analysis. The
analytical results, reported as Test-3 for the inlet and Test-4 for the
outlet, were provided by Accu-Lab using Spark Source Mass Spectrometry.
Previous studies7 indicate that Fe and Al are generally predominate
in larger particle sizes; Pb, Sb, Se, Hg, and Zn are concentrated in the
smaller particle sizes; and the other elements show either no preference
to size or are placed in either category depending on which paper is being
reviewed.
After a visual observation of the study results (presented in Tables 25
and 26) it can only be concluded that chemical analysis of such samples are
quite difficult and lead to results with a large amount of scatter (orders
of magnitude). Error can be introduced at any number of points — sample
collection, sample weighing, incomplete recovery of small samples during
the preparation for chemical analysis, contamination of samples during
physical handling or by the collection substrate, use of two different
microchemical analytical techniques, small samples near the limit of
sensitivity of the instrument or during the analysis of the blanks.
The large amount of scatter, due to all of the possible sources of
error in dealing with extremely small samples, prevents firm conclusions
regarding much of the chemical analysis data. There are, however, some
trends that might be pointed out for the reader's consideration. The
elements in Groups B and C, halogen and volatiles, tended to predominate
in the smaller particle size ranges. The elements listed in Group A,
except for Ca, tended to predominate in the smaller particle sizes
(Brink impactor data only). A majority of the elements in Group D,
refractories, showed no trend relating to particle size. Cr, Cu, and
Fe tended to concentrate in larger particles and Mn, V, and Zn tended to
concentrate in smaller particles (Brink impactor data only).
Emissions Sampling for Organics and Sulfates
During this study GCA/Technology Division, under contract to the U.S.
EPA, sampled emissions for organics from the A side of the furnace at the
ESP inlet and outlet gas ducts and provided quantitative analysis for sul-
fate, polychlorinated biphenyls (PCB), and polycyclic organic matter (POM).
A traversing gas stream was not employed; instead, a single point represen-
tative of the gas stream was sampled. Two samples were obtained at the ESP
outlet duct, and one sample was collected at the ESP inlet duct. The
results from this study are summarized below.
GCA reports that the ESP grain loading was 3.08 grains/scf, while
the average grain loading after the ESP was 0.013 grains/scf. This data
indicates a particulate collection efficiency of 99.58 percent for the
ESP (A side) on unit 1.
The chemical analysis of the samples taken at the ESP outlet indi-
cated the presence of several organic species, including: alphatic hydro-
carbons, aromatic hydrocarbons, carbonyl groups, and alcohols or phenols.
The concentration of benzo(a)pyrene in the ESP outlet duct was 0.3 [Jg/m3
48
-------
TABLE 26. CHEMICAL ANALYSIS OF ESP INLET FINE PARTICULATE - BRINK IMPACTOR
VO
D
Particles <1 Mm,
Element
Ba
Ca
Mg
K
Na
Cl
Fl
Sb
As
R*»
Hg
Se
Al
Cd
Co
Cr
Cu
Fe
Mn
Ni
Si
Ti
V
Zn
1
565
35502
6146
14857
2996
843
324
<42
1275
9 1
2.3
2.5
74508
39.4
<1275
566
207
134586
413
231
4842
425
847
2
718
31706
5944
16562
3314
1100
260
<97
2914
9.8
0.73
5.8
75723
12.6
<2914
616
118
134460
416
208
4899
517
739
••"» r&i &
3*
2150
26700
2900
18800
1700
280
350
19.4
41.2
8.8
None
Detected
4.7
33339
7.9
70.7
810
160
72000
440
140
R4000
28130
1710
1300
1
760
15400
5140
15700
2930
2425
617
120
3624
10.5
0.75
<7.3
58700
13.94
3624
288
91
50400
261
67
3750
559
1070
2
1094
19800
6530
20400
4045
2740
964
309
9270
14.0
1.67
<18.5
78200
86.46
9270
247
110
67300
376
74
5300
855
1500
3*
990
33830
3695
24830
1110
143
230
33.9
20.2
8.0
None
Detected
3.8
59025
6.6
38.6
850
72.9
47124
386
318
79160
11070
1244
1130
1
2326
39200
11470
19500
6580
20000
<2273
1344
40322
16.1
3.23
<80.6
74500
36.48
40322
1530
285
75700
2340
402
7660
1240
2030
2
1550
16300
11600
21900
5400
24400
<2778
1063
31914
10.6
1.91
<63.8
77900
11.03
31914
351
135
121200
5070
74
3780
660
1160
3*
1300
10735
2400
10826
965
1045
177
11.6
36.0
.2
None
Detected
7.8
39790
3.2
47.1
282
111
17140
217
65.4
25740
1584
795
2770
Pb
160.1
89.6
88.2
249.0
<93
155
<404
<320
160
-Accu-Lab Analytical Results
-------
TABLE 27. CHEMICAL ANALYSIS OF ESP OUTET FINE PARTICULATE - ANDERSEN IMPACTOR
B
S
D
Element
Ba
Ca
Mg
K
Na
Cl
Fl
Sb
As
Be
Hg
Se
Al
Cd
Co
Cr
Cu
Fe
Mn
Ni
Si
Ti
V
Zn
Pb
Par
1
<540
24000
10200
3890
17100
11226
363
<32
<963
0.23
0.34
3.0
13500
3.08
<963
20.6
15.0
1770
17.7
17.5
-
241
9.2
<357
40.5
rticles >
2
<440
24100
9310
5940
11700
-
-
<26
<784
1.92
2.31
8.7
22800
2.63
<784
75.4
49.2
16100
76.8
35.6
-
950
90.9
<291
32.6
>3 (Jm, MJ
3
<1168
26300
10600
4800
31000
_
-
<69
<2082
0.22
2.81
27.5
14000
8.22
<2082
54.5
20.2
4430
37.1
24.3
-
330.1
13.0
<773
57.0
iJg Particles 1-3 |Jm, Mg/g
4*
191.1
22200
9455
10386
46486
157.4
2663
2.9
4.3
<0.34
None
Detected
58.6
18750
<0.80
7.19
34.8
12.9
6750
22.9
5.32
139510
272.3
28.2
329.5
15.04
1
<789
28500
11800
4970
20900
6857
561
<47
<1407
0.39
0.65
2.8
17900
1.99
<1409
26.5
15.3
2170
25.4
19.7
-
296
19.4
<523
24.2
2
<533
28100
11600
5940
14200
_
-
<31
<949
1.08
0.25
8.2
21100
3.1
<949
32.8
17.2
5000
39.1
20.5
-
547
55.8
352
23.6
3
<2579
21200
8700
3750
54800
—
-
<153
<4601
<0.12
0.34
21.2
11800
<1 .9
<4601
18.6
<15
1230
<13.6
19.7
-
329.3
8.3
<1709
<22
4*
280.4
63370
13100
8151
65890
252
3000
1.8
1.6
0.33
None
Detected
17.8
16743
1.0
1.48
18.5
7.09
854.8
12.9
3.08
84000
176.9
30.8
554.8
3.8
Particles
1
<1411
25200
10300
5900
37700
8980
<394
<84
<2531
0.71
0.2
6.7
16800
3.60
<2531
24.6
11.2
2600
23.7
26.1
-
350
47.6
<940
24.7
2
<1139
21200
9100
4380
31400
w
-
<67
<2032
0.58
2.09
16.9
14300
1.67
<2032
23.9
9.0
2570
24.5
19.7
-
414
42.7
<755
16.4
<1 |Jm> Mg/g
3
<1362
23800
9740
4950
37600
_
-
<81
<2429
0.66
2.98
21.1
15100
1.87
<2429
26.1
<8
2680
30.3
19.4
-
408.9
50.8
<902
12.8
4*
428
31070
15147
8720
53115
133
4679
3.1
2.6
0.84
None
Detected
19.7
14444
0.97
4.0
20.7
12.9
3550
26.5
8.7
79645
747.1
121
412
19
'rAccu-Lab Analytical Results
-------
(0.4 g/hr) , which is about 15 times the MATE Level of 0.02 [Jg/m3- Also,
the upper limit to PCB emissions (1.7 pg/m3) was on the order of the back-
ground concentration of PCB's over the ocean. Finally, the concentration
of sulfate in the flue gas after the ESP was 6.47 mg/m^ (8,630 g/hr).
There is no sulfate MATE value for comparison.
Organic and Other Analyses of Ash Sluice Samples
During this study a number of liquid and dry ash samples were taken
for analysis of organic constituents. Monsanto Research Corporation, under
contract to EPA, analyzed these samples. Ash slurry samples were filtered
and analyzed as both a solid portion and a liquid portion.
Samples were analyzed by GC/MS for polycyclic organic matter (POM)
and polychlorinated biphenyls (PCB). No PCB's were detected in any of
the samples. POM's were found in all but one of the samples; however,
only two samples contained POM's in appreciable quantities (>1 [Jg)
(Table 28).
In addition to organic analysis, the samples were analyzed for total
chlorine, ionic chlorine, total organic carbon, and sulfur. The liquid
samples were also analyzed with the Jarrell Ash Atomcomp (ICAP) for trace
elements. Chlorine concentration ranged in general from 10 to 40 ppm.
The elements found in the liquid filtrates at levels in excess of 1 ppm
were Ba, B, Ca, Mg, Mo, Na, and S. Tables 28 and 298 are taken from the
final report. Table 29 also agrees with other data from Table 13.
MASS BALANCE
Determination of Flow Rates
The mass flow rates for the inflow coal and outflow ash streams are
presented in Table 30. Coal flow rates for the test periods were based on
data from the coal scales and counters; estimates were made for those
periods when there was a scale malfunction. The pulverized coal flow
rates were determined by subtracting the weight of the measured pyrites
and the coal moisture from the weight of the whole coal, provided by the
coal scale measurements. The bottom ash flows were determined by sub-
tracting pyrites and the fly ash flows from the total ash of the whole
coal flow. The ESP hopper ash flow was derived from the difference between
the total particulate trains operated before and after the ESP. All other
flows were measured directly. Table 31 contains a summary of all flows
for each element.
Material and Flow Analysis - System and Sample Points
The total material flow analysis summed up the individual trace
element flows over the eight days of sampling. For the pulverizer, boiler,
and ESP, the material flow error was calculated as the sum of the outputs
over the total input flow and was expressed as a percentage. For the pul-
the material flow error was the sum of all pyrite flows plus all
51
-------
TABLE 28. POLYCYCLIC ORGANIC MATERIALS - GC/MSa
Component
Dry ESP composite,
8/19/76 - Mg/g
Fraction
234
Colbert pyrite,
8/18/76 - Mg/g
Fraction
234
Colbert pulverizer,
8/18/76 - Mg/g
234
Dry ESP composite,
8/18/76 - MS/8
Fraction
234
Dibenzothiophene
Anthracene/phenanthrene
Methyl-anthracenes/-phenanthrenes
9 - Methyl-anthracene
Dimethyl-anthracenes/-phenanthrenes
Fluoranthene
Pyrene
Methyl-fluoranthenes/-pyrenes
Benz(c)phenanthrene
Chrysene/benz(a)anthracene
7, 12 - Dimethyl-benz(a)anthracenes (or isomers)
Benzofluoranthene(s)
Benzopyrene(s) (and perylene)
3 - Methylcholanthrene
Indeno(l, 2, 3 - cd)pyrene
Dibenz(a, h)anthracene
Dibenz(c, g)carbazole
Dibenzopyrenes
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. 0.421 0.521
N.D. 9.037 6.502
N.D. 14.545 13.076
N.D. N.D. N.D.
N.D. 17.236 14.967
N.D. 0.364 1.223
N.D. 1.758 1.474
N.D. 4.497 10.236
N.D. 0.417 1.351
N.D. 1.857 7.298
N.D. 17.658 18.844
N.D. 0.350 1.715
N.D. 0.426 1.502
N.D. N.D. 1.754
N.D. N.D. 1.270
N.D. N.D. 1.038
N.D. N.D. N.D.
N.D. N.D. 3.436
N.D. 1.422 0.466
N.D. 24.932 8.966
N.D. 48.004 16.432
N.D. N.D. N.D.
N.D. 53.903 19.315
N.D. 1.012 1.427
N.D. 4.116 1.612
N.D. 16.349 18.830
N.D. 0.611 1.039
N.D. 4.132 15.024
N.D. 20.908 24.485
N.D. 1.136 3.713
N.D. 1.553 4.757
N.D. 0.684 1.796
N.D. N.D. 1.985
N.D. N.D. 2.835
N.D. N.D. N.D.
N.D. 0.606 3.189
N.D. N.D. N.D.
N.D. 0.0231 0.0118
N.D. 0.045 N.D.
N.D. N.D. N.D.
N.D. 0.0434 N.D.
N.D. N.D. N.D.
N.D. 0.0231 N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. 0.0127 N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
N.D. N.D. N.D.
aDetect10n limit for Colbert pyrite and pulverizer is 0.5 M8; for all other samples, limit is 0.025 M8-
N.D. - Not detected.
-------
Table 28. (continued)
ESP fly ash, 8/18/76
Component
Dibenzothiophene
Anthracene/phenanthrene
Methyl-anthracenes/-phenanthrenes
9 - Methyl-anthracene
Dimethyl-anthracenes/-phenanthrenes
Fluoranthene
Pyrene
Methyl-fluoranthenes/-pyrenes
Benz(c)phenanthrene
Ch.rysene/benz( a) anthracene
7, 12 - Dimethyl-benz(a)anthracenes (or isomers)
Benzofluoranthene(s)
Benzopyrene(s) (and perylene)
3 - Methylcholanthrene
Indeno(], 2, 3 - cd)pyrene
Dibenz(a, hjanthracene
Dihenz(c, g)carbazole
Dibenzopyrenes
Liquid phase
extract, )Jg/l
2
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
somers) N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
3
N.D.
0.056
0.119
N.D.
0.161
0.088
0.174
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Solids portion,
Mg/g
Fraction
4 2
N.D.
0. 161
0.305
N.D.
0.393
0.084
0.160
0.123
N.D.
0.047
N.D.
N.D.
N.D.
N.D.
N.D.
0.066
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D,
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
3
N.D.
0.002
0.002
N.D.
0.003
<0.002
<0.002
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
4
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
S.D.
i * :
liquid phase
extract, pg/1
2
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
3
N.D.
0.845
1.01
N.D.
0.804
0.092
0.193
N.D.
N.D.
0.093
N.D.
0.071
0.062
0.248
N.D.
N.D.
N.D.
N.D.
Solids portion,
MK/R
Fraction
4 2
N.D.
0.331
0.553
N.D.
0.431
0.128
0.160
N.D.
N.D.
0.047
N.D.
N.D.
0.035
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
3
N.D.
0.002
0.005
N.D.
0.005
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.E.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
4
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D..
N.D.
N.D.
N.D. - Not detected.
-------
Table 28. (continued)
Bottom ash, 8/18/76
Bottom ash, 8/19/76
Ul
Liquid phase
extract, pg/1
Solids portion,
Liquid phase
extract, pg/1
Solids portion,
Mg/g
Component
Fraction
4 2
Fraction
A 2
Dibenzothiophene
Anthracene/phenanthrene
Methyl-anthracenes/-phenanthrenes
9 - Methyl-anthracene
Dimethyl-anthracenes/-phenanthrenes
Fluoranthene
Pyrene
Methyl-fluoranthenes/-pyrenes
Benz(c)phenanthrene
Chrysene/benz(a)anthracene
7, 12 - Dimethyl-benz(a)anthracenes (or isomers)
Benzofluoranthene(s)
Benzopyrene(s) (and perylene)
3 - Methylcholanthrene
Indeno(l, 2, 3 - cd)pyrene
Dibenz(a, (i)anthracene
Dibenz(c, g)carbazole
Dibenzopyrenes
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
0.299
0.315
N.D.
0.097
0.073
0.382
0.017
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
0.064
0.125
N.D.
0.068
N.D.
0.127
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
0.010
0.017
N.D.
0.017
0.003
0.003
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D. N.D. N.D. 0.007 0.04 0.007
N.D. 0.498 0.430 0.02 0.03 0.041
N.D. 0.581 0.466 0.03 0.78 0.068
N.D. N.D. N.D. <0.003 0.003 <0.003
N.D. 0.459 0.439 0.01 0.78 0.058
<0.003 0.03 0.007
<0.003 0.061 0.010
0.045 0.296 <0.003 0.24 0.047
N.D. 0.071 <0.003 0.003 <0.003
N.D. 1.04 <0.003 0.085 0.041
N.D. N.D. <0.003 0.580 0.021
N.D. N.D. <0.003 0.02 0.02
N.D. 0.270 <0.003 0.02 0.034
N.D. N.D. <0.003 0.027 0.014
N.D. N.D. <0.003 0.034 0.044
N.D. N.D. <0.003 0.03 0.047
N.D. 0.095 1.29
N.D. 0.141 1.07
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D. N.D. N.D. <0.003 <0.003 0.027
N.D. N.D. N.D. N.D. N.D. N.D.
N.D. - Not detected.
-------
TABLE 29. TRACE METAL CONCENTRATIONS IN FILTRATE OF ASH SLURRY SAMPLES
Trace metals,
Jarre]] ash
Atomcomp metals
(ppm)
Ag
Al
Ba
B
Ca
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Na
Ni
Pb
Sb
Si
Sn
Sr
Ti
V
Zn
P
Be
Zr
As
Bottom ash,
8/18/76, 3:27-3:55 p.m.
Sample - filtrates tested
Bottom ash, ESP fly ash,
8/19/76 8/18/76, 2:46-3:58 p.m.
ESP fly ash
8/19/76, 12:29-1:20 p.m.
0.1
28.9
<0. 1
<0. 1
0.2
3.1
5.7
<0. 1
2.9
0.1
0. 1
0.1
0.3
96.1
<0. 1
0.3
6.8
4.6
0.1
<0. 1
<0. 1
<0. 1
0.8
5.0
483.0
<0. 1
0. 1
1.2
16.7
0.1
1.6
0.7
1 .2
9.4
555.0
0.1
1.8
22.3
0.1
0.7
0.9
55
-------
TABLE 30. MASS FLOW MTE FOR VARIOUS FLOWS IN SYSTEM
Pulverized
Coal Bottom Hopper Stack Ash _—
Rav Coal Burned Pyrites Ash Ash Emission Bottom Fly~~Ash
Test (t/h) Ct/h) ,. _(t/h) iWh)_ (t/h) (t/h) Ash % _ %__
1 64.4 64.2 .20 1.70 7.92 .25 17 2 82.8
2 65.9 65.6 .30 2.19 7.16 58 22 1 77.9
3 72.1 71.9 .20 2.36 6.81 .38 24 7 75.3
4 66.8 66.6 .20 2.44 6.92 43 24 9 75 1
5 69.7 69.4 .30 2.54 6.06 1.38 26'.2 73.8
6 66.9 66.6 .30 2.14 6.90 .19 23 2 76 8
7 69.9 69.7 .20 3.14 8.26 .16 27 2 72.8
8 66.2 66.0 .20 2.61 8.16 .19 23.8 76.2
Average 67.7 67.5 .24 2.39 7.27 .45 23.7 76.3
o 2.6 2.6 .05 .42 0.77 .40 3.1 3.1
56
-------
TABLE 31. SUMMARY OF ELEMENT FLOW RATES (Ib/h)
WHOLE COAL (lb/h)
TEST A! - B3
Ca C, Cl Co Cr O, F ^ ».
Ol
1
2
3
4
5
6
7
8
r^J-
2060
1968
2019
1870
1952
1472
1957
2118
1.93
1.70
1.15
1.34
2.37
1.47
1.26
1.19
15.5
13.15
14.4
1.3
1.6
2.4
14.0
35.7
0.142
0.164
0.159
0.174
0.181
0.161
0.224
0.172
649
675
663
550
570
611
558
712
0.37
0.23
0.51
0.14
0.17
0.66
0.30
0.28
180
238
231
240
251
241
182
199
0.31
0.59
0.10
0.61
0.30
0.48
0.19
0.30
2.24
2.35
2.12
2.65
2.66
2.46
2.88
2.50
1.120
1.21
1.150
1.280
1.380
0.789
0.755
0.887
6.44
7.90
8.65
1.34
13.90
10.70
22.37
26.480
2306
2627
2394
2432
3346
2689
2824
2794
0.041
0.040
0.036
0.036
0.031
0.031
0.032
0.037
TEST K Mg Mn Na NX Pb Sb
U J.
1
2
3
4
5
6
7
8
232
275
231
254
293
281
280
225
<->
94.7
108.0
94.5
104
105
103
111
132
5.28
6.31
5.02
5.18
5.95
6.21
4.78
5.48
77.3
75.2
72.1
86.8
105.0
87.0
83.9
86.1
1.090
1.16
1.05
1.08
1.56
1.61
1.22
1.46
2.29
2.54
2.22
2.40
3.36
3.75
3.02
3.83
0.245
0.269
0.346
0.333
0.293
0.254
0.294
0.238
0.374
0.393
0.389
0.454
0.390
0.388
0.419
0.450
3980
3727
3648
3621
4210
3372
4166
4488
82.2
96.3
94.9
93.5
98.0
94.6
99.4
89.6
6.77
7.94
8.51
10.10
10.30
9.13
13.2
8.78
4.51
5.06
8.54
1.51
3.76
4.78
10.4
4.34
-------
TABLE 31. (Continued)
00
TEST
1
2
3
4
5
6
7
8
Al
7.64
11.30
7.80
7.20
10.50
9.30
7.64
11.80
As
0.048
0.084
0.044
0.044
0.046
0.043
0.022
0.028
Ba
0.394
0.441
0.186
0.072
0.111
0.363
0.108
0.080
Be
0.0
0.001
0.0
0.0
0.001
0.001
0.0
0.0
Ca
3.71
4.97
2.66
1.87
3.91
5.02
3.70
3.48
PYRITES I
Cd
0.0
0.001
0.0
0.0
0.001
0.001
0.0
0.0
Clb/h)
Cl
0.021
0.006
0.018
0.014
0.027
0.037
0.027
0.014
Co
0.020
0.031
0.013
0.015
0.026
0.021
0.018
0.018
Cr
0.016
0.036
0.015
0.021
0.032
0.034
0.017
0.022
Cu
0.012
0.025
0.016
0.011
0.022
0.019
0.011
0.013
F
0.022
0.017
0.022
0.015
0.030
0.030
0.022
0.032
Fe
74
144
106
84
151
129
92
82
Hg
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
TEST
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
Zn
1
2
3
4
5
6
7
8
1.58
3.32
1.70
1.34
2.38
3.05
2.10
2.24
0.560
1.090
0.600
0.436
0.882
1.040
0.740
0.692
0.054
0.085
0.048
0.038
0.093
0.238
0.073
0.061
0.186
0.274
0.110
0.130
0.195
0.186
0.226
0.343
0.011
0.050
0.028
0.020
0.056
0.034
0.018
0.025
0.041
0.100
0.089
0.052
0.119
0.106
0.065
0.062
0.001
0.002
0.001
0.001
0.001
0.002
0.001
0.001
0.003
0.004
0.003
0.004
0.006
0.005
0.008
0.007
12.0
20.8
12.2
12.4
18.9
16.5
16.3
26.3
0.294
0.642
0.343
0.274
0.481
0.630
0.390
0.398
0.027
0.086
0.029
0.031
0.045
0.068
0.035
0.026
0.112
0.153
0.057
0.034
0.061
0.116
0.107
0.050
-------
TABLE 31. (Continued)
TEST
TEST
AL
As
Ba
Be
Ca
PULVERIZED COAL (Ib/h)
Cd Cl Co
Cr
Cu
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
Fe
Zn
1
2
3
4
5
6
7
8
453
394
380
388
390
328
502
498
121.0
98.3
97.2
94.3
91.7
82.6
148.0
139.0
7.07
6.18
5.42
5.10
6.55
5.51
6.94
6.10
53.5
49.2
45.9
49.7
59.0
36.6
38.3
42.1
2.311
2.229
2.013
1.998
2.360
1.998
1.952
1.980
2.311
2.491
2.013
2.131
2.637
2.664
2.370
5.676
0.26
0.27
0.27
0.37
0.33
0.31
0.35
0.29
0.514
0.616
0.532
0.626
0.611
0.599
0.767
0.409
4404
3540
3451
3037
4442
3397
3694
3036
93.9
84.0
85.6
83.9
83.4
65.3
102.0
94.9
9.78
9.06
9.79
9.26
10.50
7.62
10.60
10.30
6.93
5.91
5.54
5.47
6.30
5.21
7.58
4.86
Hg
1
2
3
4
5
6
7
8
2208
1967
2114
1931
1943
1931
2091
2006
1.54
2.10
3.45
2.26
2.64
2.13
2.65
2.77
23.10
12.50
7.80
9.30
11.80
13.30
5.60
9.10
0.180
0.133
0.187
0.153
0.162
0.125
0.222
0.189
542
442
413
352
401
374
580
533
0.130
0.040
0.140
0.040
0.050
0.120
0.130
0.060
231
249
230
266
278
240
223
211
3.210
4.720
0.431
2.400
2.500
1.330
2.370
1.980
3.30
3.07
3.11
2.73
3.12
2.17
4.13
3.21
2.39
1.64
1.58
1.44
1.61
1.35
1.81
1.69
5.260
1 . 320
5.180
0.666
6.380
2.130
3.070
5.280
2889
3016
2962
2811
3331
3610
3346
2970
0.019
0.013
0.014
0.013
0.017
0.016
0.014
0.017
-------
TABLE 31. (Continued)
TEST
TEST
Al
As
Ba
Be
Ca
BOTTOM ASH (Ib/h)
Cd Cl Co
Cr
Cu
Fe
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
Zn
1
2
3
4
5
6
7
8
72.0
79.8
74.1
80.4
84.9
69.5
74.8
96.6
22.70
24.50
19.10
27.50
25.40
20.30
21.40
22.80
2.03
1.54
1.20
1.52
1.90
1.47
1.14
1.17
8.9
11.9
12.2
11.8
14.0
10.6
18.3
15.3
0.340
0.439
0.444
0.487
0.559
0.399
0.628
0.595
0.323
0.403
0.335
0.541
0.513
0.399
0.528
0.423
0.014
0.014
0.020
0.024
0.023
0.016
0.033
0.016
0.002
0.002
0.004
0.005
0.004
0.003
0.003
0.003
540
597
632
775
767
566
955
742
23.7
17.6
17.0
18.8
20.4
15.2
14.1
19.0
1.57
1.88
1.78
2.09
2.22
1.82
1.34
1.65
0.877
0.934
0.651
1.140
1.140
0.759
0.672
0.637
Hg
1
2
3
4
5
6
7
8
323
430
425
374
416
357
534
463
0.014
0.022
0.038
0.073
0.061
0.030
0.050
0.042
3.18
5.00
3.37
3-80
3.61
4.50
5.37
2.87
0.027
0.031
0.030
0.034
0.034
0.026
0.023
0.028
105.0
115.0
93.9
155.0
114.0
86.6
92.4
97.7
0.001
0.003
0.0
0.004
0.003
0.0
0.004
0.002
0.06
0.23
0.12
0.17
0.20
0.07
0.180
0.14
0.136
0.215
0.165
0.249
0.208
0.223
0.245
0.235
0.557
0.667
0.666
0.677
0.782
0.884
0.754
0.601
0.249
0.296
0.275
0.341
0.350
0.276
0.241
0.288
0.017
0.022
0.024
0.024
0.010
0.0
0.031
0.0
646
855
944
892
976
974
1225
898
0.0
0.0
0.0
0.0
0.0
0.0
0.001
0.0
-------
TABLE 31. (Continued)
TEST
TEST
Al
As
Ba
Be
Ca
ESP INLET (Ib/h)
Cd Cl Co
Cr
Cu
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
Fe
Zn
1
2
3
4
5
6
7
8
263
275
292
153
329
281
300
394
97.6
88.6
74.0
75.9
88.5
80.5
81.0
97.6
7.65
6.41
4.93
4.99
6.99
5.97
4.65
5.23
75.2
69.7
66.1
64.6
71.4
69.5
70.7
75.2
2.24
3.03
2.93
2.03
2.50
2.37
3.18
3.69
1.380
1.210
0.881
1.460
1.680
1.370
0.756
0.602
0.294
0.381
0.244
0.176
0.282
0.440
0.253
0.318
0.311
0.223
0.244
0.294
0.282
0.270
0.253
0.318
3073
2950
2746
2805
2854
2725
3285
3409
87.3
85.0
75.9
72.7
84.0
74.4
89.3
93.3
7.80
8.47
8.01
7.73
8.55
8.69
8.14
8.76
9.460
9.780
8.110
9.300
10.900
9.450
7.560
9.110
Hg
1
2
3
4
5
6
7
8
1635
1548
1438
1469
1442
1419
1684
1671
1.36
1.39
1.42
1.20
1.49
1.42
1.33
1.35
21.30
15.60
10.20
11.90
14.60
13.90
9.94
9.19
0.145
0.157
0.150
0.138
0.159
0.152
0.173
0.174
497
413
328
339
394
351
318
403
0.245
0.209
0.187
0.250
0.193
0.184
0.185
0.234
113.0
70.7
40.3
41.1
75.8
41.2
42.1
63.5
0.719
0.691
0.589
0.676
0.699
0.596
0.775
0.769
9.27
11.40
6.83
5.38
7.54
7.31
10.20
12.10
1.33
1.54
1.32
1.30
1.47
1.30
1.40
1.440
6.87
6.19
3.74
3.60
4.61
3.48
3.87
5.01
1814
1808
1624
1586
1680
1533
1652
1526
0.452
0.226
0.040
0.026
0.050
0.030
0.020
0.017
-------
TABLE 31. (Continued)
ON
fo
TEST
TEST
Al
As
Ba
Be
Ca
ESP HOPPER (Ib/h)
Cd Cl Co
Cr
Cu
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
Fe
Zn
1
2
3
4
5
6
7
8
271
250
244
253
231
208
203
297
111.0
91.3
79.2
85.6
72.3
77.9
92.5
115.0
6.720
6.040
4.760
5.140
4.650
5.980
5.420
5.580
60.2
53.7
54.5
58.1
47.2
60.7
67.8
66.9
1.340
1.180
1.080
1.050
1.020
0.988
0.878
1.280
1.190
1.170
0.950
0.863
0.912
1.030
1.100
1.330
0.151
0.187
0.204
0.249
0.121
0.207
0.231
0.261
0.063
0.055
0.082
0.046
0.048
0.069
0.078
0.103
3185
2735
2573
2697
2375
2775
3223
3249
94.9
80.6
73.9
75.8
66.9
68.3
49.9
103.0
8.58
7.07
6.66
6.65
7.03
7.33
5.42
8.52
8.81
7.35
5.83
6. 13
5. 79
6.99
6.30
7.41
Hg
1
2
3
5
6
7
8
1442
1384
1307
1369
1212
1298
1587
1567
0.396
0.453
0.177
1.080
0.400
0.469
2.980
1.050
17.40
14.90
9.12
9.13
12.10
13.50
13.70
11.80
0.181
0.130
0.128
0.119
0.114
0.130
0.113
0.173
646
584
456
492
395
489
516
617
0.143
0.064
0.004
0.004
0.073
0.002
0.182
0.002
1.59
2.87
0.05
1.25
1.15
0.83
0.04
1.18
0.380
0.397
0.449
0.484
0.727
0.594
0.925
1.010
2.41
2.06
1.89
1.87
1.72
2.76
1.55
2.30
1.250
1.090
1.010
0.985
0.889
0.941
0.767
1.200
1.110
1.060
1.360
0.968
0.121
0.966
0.826
0.327
2393
2115
2042
2144
1854
2057
1801
2433
0.003
0.010
0.001
0.001
0.011
0.007
0.002
0.014
-------
TABLE 31. (Continued)
TEST
1
2
3
4
5
6
TEST
1
2
3
4
5
6
7
Al
As
Ba
Be
Ca
6.81
5.37
5.64
3.79
3.94
3.95
3.56
4.83
K
1.38
2.89
1.21
1.02
2.69
1.98
1.11
1.52
0.091
0.24
0.183
0.101
0.292
0.369
0.121
0.312
Mg
1.050
0.895
0.914
0.687
0.969
0.627
0.388
0.591
0.084
0.060
0.064
0.034
0.058
0.031
0.006
0.034
Mn
1.800
5.800
1.870
4.570
14.900
0.805
0.518
0.432
0.001
0.002
0.001
0.001
0.003
0.001
0.001
0.001
Na
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
3.88
3.23
4.51
2.47
7.18
2.41
2.07
2.39
Ni
4.28
11.10
7.36
11.50
58.10
6.90
6.14
2.39
ESP
Cd
0.026
0.046
0.066
0.035
0.084
0.038
0.018
0.017
Pb
0.020
0.049
0.056
0.048
0.103
0.037
0.027
0.023
OUTLET (Ib/h)
Cl
1.36
4.13
22.50
12.10
9.19
1.37
11.50
0.62
Sb
0.030
0.122
0.050
0.077
0.315
0.046
0.032
0.016
Co
0.415
1.260
0.374
1.030
2.750
0.235
0.142
0.159
Se
0.001
0.002
0.001
0.003
0.016
0.001
0.007
0.009
Cr
14.10
45.00
17.10
27.10
130.00
15.20
11.30
4.98
Si
0.0
0.0
0.0
0.0
0,0
0.0
0.0
0.0
Cu
0.135
0.259
0.490
0.257
1.070
0.464
0.509
0.191
Ti
0.315
0.263
0.203
0.125
0.181
0.182
0.163
0.239
F
0.011
0.075
0.032
0.117
0.311
0.062
0.003
0.040
V
0.086
0.140
0.092
0.107
0.338
0.081
0.065
0.059
Fe
Hg
95.2
251.0
70.1
180.0
550.0
38.7
24.5
29.7
Zn
0.238
0.690
0.595
0.592
2.640
0.596
0.305
0.313
0.003
0.012
0.016
0.002
0.010
0.004
0.015
0.011
-------
pulverized coal flows divided by the total whole coal flows. The result-
ing error around the pulverizer was +3.7 percent. Similarly, the boiler
material flow error was the total bottom ash flows plus the ESP inlet
flows divided by the total pulverized coal flow. The error around the
boiler was -6.8 percent. Estimating the error around the ESP as the sum
of the outlet and hopper flows divided by the total inlet flow yielded a
+5.8 percent error. Table 32 summarizes the material flows and the
estimates of the material flow errors.
The material flows were also analyzed to identify significant shifts
in flow from test-to-test. Table 33 shows the material flows by test
for whole coal, pyrites, pulverized coal, bottom ash, and the inlet, hopper,
and outlet of the ESP. Tests 1 and 8, upon closer examination of the
data, revealed large differences between whole coal and pulverized coal
flows, and between the ESP inlet and ESP hopper. The material flows com-
pared here should match closely (within a couple of hundred Ib/h on the
average) when the system is operating. Since the flows matched well for
the other tests, a possible error occurred on tests 1 and 8. Further
investigation did not identify the source of error, but these tests were
noted as possibly being in error.
Material flows around the pulverizer, boiler, and ESP were calculated
on a daily basis in order to identify tests that may have been impacted
by a malfunction. The error percentages are summarized in Table 34.
Again, tests 1 and 8 stand out as either having large errors compared to
the other tests or having an error in the opposite direction from the
other tests. For example, the largest error on the pulverizer occurred
on test 1; and although most of the errors are positive, an extremely
large negative error occurred on test 8. The loop around the boiler had
consistently negative errors (the largest occurring on test 1) while the
only positive error occurred on test 8. The loop around the ESP did not
show the same pattern of errors in the material balance. Tests 2 and 7
showed extremely good and consistent material balances throughout the
system.
Examination of the whole coal and pulverized coal daily averages
and standard deviations showed expected behavior. The pyrites do not
constitute a significant proportion removed from the whole coal material
flow. The standard deviations for the whole coal and pulverized coal
are not significantly different.
The standard deviation of the daily pyrite flows relative to the
mean indicated substantial variation in the daily pyrite flows. Sometimes
this is an indicator of an erroneous or atypical value inflating the esti-
mate of the variance. Examination of daily pyrite flows in Table 33 indi-
cated two flows around 100 Ib/h, three flows near 130 Ib/h, and two flows
near 190 Ib/h. Thus, the relatively large standard deviation was not
due to one extremely high or low daily material flow.
The other flows which stood out as having relatively large standard
deviations were the bottom ash and ESP outlet flows. For the daily bottom
a(sh flows, two values stood out as being relatively different—test 1
had a low flow and test 7 had a high flow. These were noted for further
-------
TABLE 32. TOTAL MATERIAL FLOWS OF TRACE ELEMENTS AND ESTIMATED ERRORS
AROUND THE PULVERIZER, BOILER, AND ELECTROSTATIC PRECIPITATOR
I of Tests 1 through 8
Sample
Point
Whole Coal
Pyrites
Pulverized Coal
Bottom Ash
ESP Inlet
ESP Hopper
ESP Outlet
Total
Flow (Ib/h)
79,579.93
1,134.36
81,385.05
18,314.00
57,569.63
59,107.50
1,813.75
Pulverizer
Boiler
ESP
Error °/j
+3.7
-6.8
+5.8
TABLE 33. TOTAL MATERIAL FLOWS OF TRACE ELEMENTS
BY TEST (ROUNDED TO NEAREST Ib/h)
Sample
Point
Whole Coal
Pyrites
Pulverized
Coal
Bottom Ash
ESP Inlet
ESP Hopper
ESP Outlet
Test
1
9,710
101
11,064
1,751
7,726
8,255
131
2
9,840
188
9,892
2,143
7,375
7,340
333
3
9,502
132
9,826
2,226
6,734
6,863
133
4
9,281
108
9,057
2,346
6,657
7,211
246
5
10,978
189
11,077
2,429
7,081
6,290
785
6
8,996
166
10,112
2,110
6,632
7,077
74
7
10,337
124
10,775
2,946
7,575
7,581
63
8
10,936
128
9,584
2,363
7,791
8,491
49
TABLE 34. ESTIMATED ERRORS AROUND THE PULVERIZER, BOILER,
AND ELECTROSTATIC PRECIPITATOR BY TEST
Test
1
2
3
4
5
6
7
8
Total
Pulverizer
+ 15.
+ 2.
+ 4.
- 1.
+ 2,
+14,
+ 5,
-11,
4%
8%
3%
6%
3%
4%
2%
Boiler
-14.3%
- 3.8%
- 8.8%
- 0.6%
-14.1%
-13.5%
+ 3.7%
- 2
+ 6
- 6.;
.4%
.0%
ESP
+ 8.5
+ 4.0
+ 3.9
+ 12.0
- 0.1
+ 7.
+ 0.
.8
.9
+ 9.6
+ 5.1
65
-------
investigation, but no explanation was found for these values. The ESP
outlet figures showed extremely large variation consistently throughout
the daily flows. Again, further investigation did not reveal any
apparent errors or explanation for the large variation.
The average daily flow and the standard deviation of the daily flows
by sample point were examined to determine which sample points experienced
the largest day-to-day variation. Table 35 summarizes the comparison of
all 8 tests and tests 2-7. The daily flows for the ESP outlet show an
extremely large variation in both cases. Using tests 2-7 reduced the
variance of pulverized coal about 3 percent, bottom ash 17 percent, ESP
inlet 32 percent, ESP hopper 61 percent, and increased the variance at
the ESP outlet by 22 percent. The table indicates that there was no sig-
nificant difference in the average daily flows of the whole coal and pul-
verized coal. After removing tests 1 and 8 from the data for the ESP,
the inlet and hopper data behaved as expected; but there was still sig-
nificant variation in the outlet data. The amount of variation in all
eight test flows, when compared to the magnitude of the average test
flows, indicated that the sample points having the largest variability
were the ESP outlet, the pyrites, and the bottom ash. The remaining
sample points had a standard deviation less than 10 percent of the mean.
When looking at tests 2-7, the results were the same; relative to the
size of the average daily flow, the largest test-to-test variability
occurred at the ESP outlet, the pyrites, and the bottom ash. The source
of the variability does not appear to be the variability of the whole coal
put into the system alone. One source of variation might have been
operating variability in the pulverizer and boiler. The operating varia-
bility of the ESP depends on many factors (such as control settings); but
after day 5, the large change at the ESP outlet for small changes at the
inlet strongly suggest that the collection efficiency was improved through
some maintenance operation. Other possibilities for the variability were
the use of an inappropriate sampling technique or a change in the coal.
Material Flow Analysis - By Trace Element
The total material flows over all eight tests for each of the 25
trace elements in the whole coal, the percentage of the total whole
coal flow, and identification of large and small flows are presented in
Table 36. The five elements with the largest flows and the six elements
with the smallest flows are shown in Figure 11. The sum of five elements
accounted for 94.4 percent of the total 79,580 Ib/h of trace elements in
the whole coal. Figure 11 also shows which six elements had extremely
small flows. These estimates must be viewed cautiously since they border
on the limits of instrument detectability, and/or since they may be sub-
ject to large errors in the flow rate. Difficulty in estimating flow
rates is detailed elsewhere in this report.
Material flows around the pulverizer, boiler, and ESP were calcu-
lated for each trace element for the eight days of testing combined.
Figures 12, 13, and 14 present the results; and Table 37 summarizes those
elements with flows of 100 Ib/h or more as those elements account for
99.4 percent of the total flow. Examination of Figures 12, 13, and 14
allowed identification of elements which had errors in excess of ±10
66
-------
TABLE 35. MEAN AND STANDARD DEVIATION OF DAILY FLOWS OF TRACE
ELEMENTS (in Ibs/h) BY SAMPLE POINT
Point
Whole Coal
Pyrites
Pulverized Coal
Bottom Ash
ESP Inlet
ESP Hopper
ESP Outlet
All 8 Tests
Mean
Standard
Deviation
9,947.5
141.8
10,173.2
2,289.3
7,196.2
7,388.4
226.7
737.42
34.52
733.70
339.65
484.96
718.80
245.85
Test
Mean
9,822.4
151.0
10,122.9
2,366.7
7,008.8
7,060.2
272.3
2-7
Standard
Deviation
732.04
34.50
722.88
308.56
400.30
447.98
272.01
TABLE 36. SUMMATION OF MATERIAL FLOWS (Ib/h) FOR TRACE ELEMENTS
IN WHOLE COAL
Element Total Flow
Al
As
Ba
Be
Ca
Cd
Cl
Co
Cr
Cu
F
Fe
Hg
K
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
V
Zn
15,416.0
12.4
98.0
1.4
4,988.0
2.7
1,762.0
2.9
19.9
8.6
97.8
21,412.0
0.3
2,071.0
852.2
44.2
673.4
10.2
23.4
2.3
3.3
31,212.0
748.5
74.7
42.9
% of Whole
Coal Flow
19.37
0.02
0.12
0.00
6.27
0.00
2.21
0.00
0.03
0.01
0.12
26.91
Five Largest
Flows Ranked
Six Smallest
Flows Ranked
0.00
2,
1.
,60
.07
0.06
0.85
0.01
0.03
0.00
0.00
39.22
0.94
0.09
0.05
2
4
5
2
5
3
6
67
-------
oo
100,000.0
cc
x
or
o
10,000.0 -
1,000.0
100.0
10.0
1.0
O.I
_L
Si Fe Al Co K ' Se Co Cd
LARGER (FLOWS) SMALLER
ELEMENT
Sb Be Hg
Figure II. Largest and smallest total flows.
-------
2.0
1.0
o
c
ON
-1-0
-2.0
Figure 12. Errors in the material balance by trace element around the pulverizer.
-------
2.0
1.0
\- 0
o
c
-1.0
-2.0
Figure 13. Errors in the material balance by trace element around the boiler.
-------
2.0
1.0
15
O
-1.0
or
O
tr
CE
LJ
-2.0
638
569
505
447
395
348
305
267
232
200
172
146
122
101
82
65
49
35
22
10
0
-10
o -18
" -26
-33
-39
-45
-50
-55
-59
-63
-67
-70
-73
-75
-78
-80
-82
-83
-85
-86
Cr
Co
Zn
Se
Figure 14. Errors in the material balance by trace element around the ESP.
-------
TABLE 37. ERRORS IN MATERIAL BALANCES FOR TRACE ELEMENTS
WITH LARGE TOTAL FLOWS1
Element
Si
Al
Ca***
Mg
Ti
Na***
Total Flow
31,212.0
21,412.0
15,416.0
4,988.0
2,071.0
852.2
748.5
Pulverizer
Error
-6.6
.5
.5
673.4
+20.
+5,
-26.5
+61.8
+3.0
-7.0
-44.2
% Boiler
Error
+ 1.4
-17.3
-3.5
+7.3
-12.4
-0.5
+16.6
+77.8
% ESP
Error
-4.4
+36.7
-9.0
+38.8
-13.8
+6.9
-7.1
-16.6
lSummed over the eight days.
***Errors in excess of ±10% in at least two loops.
percent for two or more of the loops around the pulverizer, boiler, or ESP.
The troublesome elements are identified by three asterisks. Because only
14 of the 25 trace elements met this criterion, there were serious problems
with the data. The data of Figures 12, 13, and 14 also allowed identifica-
tion of elements which were consistently good except for one error exceed-
ing ±10 percent. A possible error in the flow estimation, the trace element
concentration, data recording, etc., could be suspected. Table 37 data
indicate that serious material imbalances were present for iron, calcium,
potassium, and sodium. This data was extremely surprising due to the
quantity of these elements present and the fact that these elements were
routinely analyzed in fly ash samples.
The elements which have material balances with +10 percent for two
of the three loops are summarized in Table 38 with their associated total
flows. Together, they account for 60.9 percent of the total flow. There-
fore, approximately 61 percent of the trace element flow had a material
balance with ±10 percent for at least two and possibly three of the loops.
Further analysis was done on iron, calcium, potassium, and sodium in an
attempt to see if there was a higher percentage of trace element flow
estimable. Since the total flows for tests one and eight were previously
identified as suspect, the material balances for these four trace elements
were calculated for tests 2 through 7 to see if the fluctuations of tests
1 and 8 were the major problem. Table 39 summaries the results; and, as
can be seen by comparison with Table 37, tests 1 and 8 flows were not the
problem with these four elements. This comparison suggests a possible
problem with flow estimation in general.
Special Considerations and Assumptions
Several topics and problem areas which occurred during the test pro-
gram and must be identified and considered when interpreting the results,
72
-------
TABLE 38. TRACE ELEMENTS WITH GOOD1 MATERIAL BALANCES
Total Flow I Pulverizer % Boiler % ESP
Element (Ib/h) Error Error Error
Si 31,212.0 -6.6 +1.4 -4.4
Al 15,416.0 +5.5 -3.5 -9.0
Mg 852.2 +3.0 -0.5 +6.9
Ti 748.5 -7.0 +16.6 -7.1
Ba 98.0 -3.8 +49.5 -4.3
V 74.7 +3.4 +4.7 -12.0
Pb 23.4 -2.1 -42.6 -4.6
Cu 8.6 +59.1 -0.7 +3.7
Sb 2.3 +8.5 +3.7 -4.7
Be 1.4 -1.7 +9.6 -11.9
^Error within +10% for two of three loops.
TABLE 39. ANALYSIS OF LARGE FLOW TRACE ELEMENTS WITH LARGE ERRORS
DAYS 1 AND 8 FLOWS REMOVED
Element % Pulverizer Error % Boiler Error % ESP Error
Fe +21.3 -17.4 +32.8
Ca -28.7 +9.3 +37.8
K +48.4 -9.1 -14.1
Na -45.1 +76.1 -17.0
73
-------
Figure 15 is a schematic of the system from the entrance point of the
whole coal to the exit point of the fly ash from the ESP. The legend on
the graph indicates where the flow rate and concentration was measured
during the 8-day program.
The Mechanical Collector
At the time of this study the mechanical collector was in place,
but all of the vanes were removed. By examining Figure 15, it is seen
that the fly ash exiting the boiler and entering the mechanical collector
was not sampled. It was assumed that the mechanical collector minus the
vanes would remove an insignificant amount of material. A later analysis
indicated that less than one-half of 1 percent of the daily flow through
the mechanical collector was being collected.
Trace Element Flows
Some of the trace elements also had minimum detectability problems.
Since the flows were small and measurable at most sample points, the
following procedure was used to generate flows in order that a material
balance could be approximated throughout the system: (1) the minimum
detectable limit of the trace element was used as a maximum possible
value for the interval from zero to that value; and (2) a random number
was selected from that interval in such a way that every number in the
interval had an equal chance of being selected. Table 40 lists the
elements, the sample points, the tests where minimum detectable limits
were encountered, and the "generated" value from the uniform random
number generator.
Vapor Phase Data
Sampling took place at the inlet and outlet of the ESP. Due to the
minute amount of material in the vapor phase, the collection took place
over the first week (identified as test 1) and the second week (identi-
fied as test 2). All of the element concentrations were below the
detectable or measurable limits. Some of the upper limits were higher
than the amount of element in the whole coal. For example, in the test 1
inlet data, chromium, nickel, and lead are indicated as having flows of
10.7 Ib/h, 16.24 Ib/h, and 10.4 lb/h, respectively, or less. The average
flows of these elements in the whole coal were 2.48 lb/h, 1.28 lb/h, and
2.93 lb/h, respectively. This data indicates that another approach to
sampling the vapor phase needs to be examined.
Flow Rate Estimation and Associated Problems
The material flow rate was measured at the whole coal, the pyrites
were measured from the pulverizer, and the solids were measured from the
ESP (hopper and gas) outlets. The remaining flow rates were estimated.
Table 41 summarizes the total material flows and contains an empirical
74
-------
WHOLE
COAL
PULVERIZER
PULVERIZED
COAL
BOILER
FLY ASH I
MECHANICAL
COLLECTOR
FLY ASH 2
ESP
ESP
OUTLET
PYRITES
BOTTOM ASH
MECHANICAL
COLLECTOR ASH
ESP
HOPPER ASH
SAMPLE POINT
WHOLE COAL
PYRITES
PULVERIZED COAL
BOTTOM ASH
FLY ASH I
MECHANICAL COLLECTOR ASH
FLY ASH 2
ESP HOPPER ASH
ESP OUTLET
CONCENTRATION
MEASURED
MEASURED
MEASURED
MEASURED
N2
N
MEASURED
MEASURED
MEASURED
FLOWRATE
MEASURED
MEASURED
E1
E
N
N
E
E
MEASURED
ESTIMATED
2NOT MEASURED OR ESTIMATED AT TIME OF STUDY
Figure 15. Simplified system schematic.
-------
TABLE 40. TRACE ELEMENTS WITH MINIMUM DETECTABLE LIMITS
Element Sample Point Test Limit (Ib/h) Generated Value
Ba Whole Coal
Ba Whole Coal
Ba Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Whole Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Pulverized Coal
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Cd Bottom Ash
Co Whole Coal
Co Whole Coal
Co Whole Coal
Co Whole Coal
Co Whole Coal
Co Whole Coal
Co Whole Coal
Co Whole Coal
F Bottom Ash
F Bottom Ash
F Bottom Ash
F Bottom Ash
F Bottom Ash
F Bottom Ash
F Bottom Ash
F Bottom Ash
4
5
6
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
6.60
6.90
6.60
0.64
0.66
0.72
0.67
0.70
0.67
0.70
0.66
0.13
0.13
0.14
0.13
0.14
0.13
0.14
0.13
0.003
0.004
0.005
0.005
0.005
0.004
0.006
0.005
0.64
0.66
0.72
0.67
0.70
0.67
0.70
0.66
0.017
0.022
0.024
0.024
0.025
0.021
0.031
0.026
1.33
1.68
2.47
0.37
0.23
0.51
0.14
0.17
0.66
0.30
0.28
0.13
0.04
0.14
0.04
0.05
0.12
0.13
0.06
0.001
0.003
0.002
0.004
0.003
0.001
0.004
0.002
0.31
0.59
0.10
0.61
0.36
0.48
0.19
0.30
0.010
0.021
0.018
0.013
0.007
0.002
0.029
0.003
76
-------
TABLE 41. FLOW RATES AND ASSOCIATED ERRORS
Sample
Point
Whole Coal
Pyrites
ESP Outlet
Test
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Flow Rate
64.4
65.9
72.1
66.8
69.
66.
69.
66.
400
600
400
400
600
600
400
400
501
1,168
762
856
2,752
387
318
386
Units of
Flow Rate
t/h
t/h
t/h
t/h
t/h
t/h
t/h
t/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Ib/h
Estimated
Error
±1
±1
±1
±1
±1
±1.5
±1.5
±1.5
,5
.5
.5
.5
.5
t/h
t/h
t/h
t/h
t/h
t/h
t/h
t/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±15 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
±20 Ib/h
77
-------
estimate of the error associated with the flows. Since two of the three
flows around the pulverizer were measured, the mass balance around the
pulverizer was the best. The lack of measured flow rates resulted in
the boiler and ESP having larger mass imbalances. The trace elements
constitute only a small part of the total mass flow: of the 68 tons of
whole coal per day on the average, only about 5 tons are trace elements;
of approximately 500 Ib/day in the pyrites, 140 Ib (28 per-cent) are
trace elements; of the 890 Ib/day at the ESP outlet, 225 Ib (25.3 percent)
are trace elements. After taking the natural variability of the coal and
the detection limits of the instruments used to measure trace elements,
the variability of the resulting data was understandable. The factors that
can reduce variability are more frequently measured flow rates around the
boiler and ESP and, when possible, the use of better instruments and
sampling procedures.
In examining the trace element data, one omission stood out—the
lack of sulfur data. It is strongly recommended that sulfur be included
with Fe and Al as an element to be monitored since the chemistry and
behavior characteristics of these elements are available as checks on
total system behavior, flow rates, and accuracy.
ESTIMATE OF THE MASS BALANCE
In estimating the mass balance of the trace elements, several assump-
tions were made. First, vapor phase estimates were not made for all
elements due to the lack of reliable and quantitative information on the
vapor phase. While the total mass of the elements in the vapor phase is
small when compared to the total mass balance, it is recognized that the
vapor phase is an area of concern. Second, for ease of computation,
the total trace element flow was set to a standard of 10,000 Ib/h in the
whole coal. Third, for each individual element the most reasonable
flows were chosen or estimated for consistency throughout the system.
While this was an empirical judgment, it represented the best estimate
possible.
Table 42 displays the estimated mass balance by sample points summed
over all 25 elements and shows the percentage breakdown of the flows.
Table 43 presents the estimated mass balance by element at each sample point.
Particularly difficult elements to estimate are identified.
78
-------
TABLE 42. ESTIMATED MASS BALANCE OF MAJOR AND TRACE
ELEMENTS (SUMMATION OF 25 ELEMENTS) BY SAMPLE POINTS
Flow (Ib/h) % of Loop Input
Input
Whole Coal
Pyrites
Pulverized Coal
Bottom Ash
Vapor Phase1
ESP Inlet
ESP Hopper
ESP Outlet
10,000.000
153.730
9,846.27
2,302.317
219.10
7,324.853
7,049.853
275.000
1.5373
98.4627
23.3826
2.2252
74.3922
96.2457
3.7543
Whole Coal
Whole Coal
Pulverized Coal
Pulverized Coal
Pulverized Coal
ESP Inlet
ESP Inlet
1The vapor phase was approximated by adding the chlorine and
fluorine material loss exiting the boiler.
79
-------
TABLE 43. ESTIMATED MASS BALANCE BY ELEMENT FOR EACH SAMPLE POINT1
Element
Al
As
Ba
Be
Ca
Cd
Cl2
Co
Cr
Cu
F2
Fe
Hg
K
Mg
Mn
Na
Ni
Pb
Sb
Se
Si
Ti
V
Zn
Whole
Coal
(lb/h)
1960.000
1.500
14.000
.170
625.000
.330
220.000
.360
2.500
.800
12.000
2645.000
.040
330.000
110.000
5.500
85.000
1.500
3.000
.280
.410
3875.000
90.000
9.810
7.800
Pyrites
(lb/h)
10.000
.050
.200
.001
3.500
.001
.040
.020
.030
.020
.030
115.711
0.000
2.200
0.800
.090
.250
.030
.080
.001
.006
20.000
0.500
.070
.100
Pulverized
Coal
(lb/h)
1950.000
1.450
13.800
.1690
621.500
.329
219.960
.340
2.470
.780
11.970
2529.289
.040
327.800
109.200
5.410
84.750
1.470
2.920
.279
.404
3855.000
89.500
9.740
7.700
Bottom
Ash
(lb/h)
430.000
.050
4.000
.029
69.000
.002
.150
.230
.670
.280
.030
947.780
0.000
75.000
22.200
1.410
14.750
.470
1.890
.020
.003
713.103
18.750
1.800
.700
Inlet
(lb/h)
1520.000
1.400
9.800
.1400
552.500
.327
8.650
.110
1.800
.500
4.000
1581.509
.040
252.800
87.000
4.000
70.000
1.000
1.030
.259
.401
3141.897
70.750
7.940
7.000
ESP
Hopper
(lb/h)
1473.250
1.100
9.750
.139
539.000
.287
1.150
.100
1.770
.300
3.000
1518.259
.010
250.800
84.400
3.400
68.000
.850
1.000
.199
.396
3007.843
70.500
7.850
6.500
Outlet
(lb/h)
46.750
.300
.050
.001
13.500
.040
7.500
.010
.030
.200
1.000
63,250
.03
2.000
2.600
0.600
2.000
.150
.030
.060
.00
134.05
.250
.090
.500
1The vapor phase is assumed to be 219.10 lb/h.
2Element's mass balance particularly difficult to estimate based on data because large
amounts are known to escape in the vapor phase.
80
-------
REFERENCES
1. Bolton, N. E., et al. "Trace Element Measurements at the Coal-Fired
Allen Steam Plant." Progress report, February 1973 - July 1973.
2. Klein, D. H., et al. "Pathways of Thirty-Eight Trace Elements Through
a Coal-Fired Power Plant." Paper submitted for publication, 1975.
3. Kaakinen, J. W., R. M. Jordan, M. H. Lawasani, and R. E. West. "Trace
Element Behavoir in Coal-Fired Power Plant." Env. Sci. Technol.,
9(9), 862-869 (September 1975).
4. Lee, R. E., H. L. Crist, A. E. Riley, and K. E. MacLeod. "Concen-
tration and Size of Trace Metal Emissions from a Power Plant, a Steel
Plant, and a Cotton Gin." Env. Sci. Technol., 9(7), 643-647 (July
1975).
5. Kaakinen, J. W., and R. M. Jordan. "Determination of a Trace Element
Mass Balance for a Coal-Fired Power Plant." Paper presented at the
First Annual NSF Conference on Trace Contaminants, Oak Ridge, Tennessee,
August 7-10, 1973.
6. Andren, A. W., D. H. Klein, and T. Talmi. "Selenium in Coal-Fired
Steam Plant Emissions." Env. Sci. Technol., 9(9), 856-858 (September
1975).
7. Natush, D.F.S., J. R. Wallace, C. A. Evans, Jr. Science. 183(4121),
202-4 (1974).
8. Haynes, W. M., R. B. Reznik, D. G. DeAngelis, G. W. Buttler. "Special
Project Report—Analysis of Colbert Station Samples." Monsanto Research
Corporation, MRC-DA-653, March 1977.
81
-------
Appendix A
DESCRIPTION OF SAMPLING TRAINS
82
-------
DESCRIPTION OF SAMPLING TRAINS
Train 1 - Total Particulate
A schematic of this train is depicted in Figure A-l. Basically,
this train is operated in the same manner as the EPA method 5 train. It
traverses the gas stream and samples isokinetically. In this, as well
as all other trains, 3.5-inch filter holders were used; and because of
delay in the delivery of filter substrates from the vendor, Gelman AE-
type filters were used instead of the intended Spectro Grade filters.
Simultaneous measurements of total particulate concentrations were
carried out at the inlet and outlet of the ESP. One sampling run of 1-4
hours was made each day at each of these locations. A total of eight
samples were taken at each location for the entire study.
Train 2 - Vapor-Phase Trace Elements
Figure A-2 shows a schematic of this train, which also served as a
backup train for Hg. In contrast to train 1, train 2 operates at a pro-
portional flow rate and at a fixed point in the flue duct. The K2C03
solution was used to collect Se, HC1, and HF, as well as to neutralize
S02. The KOH solution was used to capture Sb and As, and to scrub-out
S02, The strong oxidizing solutions containing mixtures of RzOz, HN03,
and AgN03 served to absorb Sb and As as well as the remaining elements
listed above. The lower ends of the bubbler tubes in impingers 1, 2,
and 3 were made of fritted glass with pore sizes of 100 |Jm to increase
the gas-liquid contact and to enhance the collection efficiencies of the
impingers.
During the sampling, the strengths of the K2C03 in impinger 1 and
of the HgC^ solutions in impingers 5 and 6 were checked by running paral-
lel sampling with identical impingers containing K2CC>3 and mixtures of
H202, HN03, and AgN03, respectively. The strength of K2C03 was deter-
mined by titration against H2S04 to the phenolphthalein end point, and
that of H202 by titration against KMn04. The absorbing solutions were
replenished whenever necessary to maintain at least a 50-percent excess.
Sampling was done simultaneously at the inlet and outlet of the ESP.
Preliminary studies indicated there were small concentrations of trace
elements present in the gases. The original plans for collecting 4-hour
samples in the morning and the afternoon were abandoned in favor of run-
ning the trains continuously for four or five days. A total of two
samples were collected at the inlet and outlet of the ESP, respectively.
Train 3 - Particulate Interference
This train, as depicted in Figure A-3, is a modification of the
original train which had been intended for use in determining the inter-
ference of fine particle penetration on the vapor-phase trace elements
collected by trains 2 and 4. The original train contained two particulate
filters; one was a 0.3-|Jm fiberglass filter in front of the condenser-ice
83
-------
oo
PROBE
REVERSE-TYPE
PiTOTTUBE
HEATED
THERMOMETER -\ CHECK
\_ VALVE
STACK
WALL
P1TOT
MANOMETER
VACUUM
LINE
THERMOMETER
oo
(|) - 100 ml H20
-EMPTY
-200g. SILICA GEL
TEST \
METER 1
ICE BATH
BY-PASS
VALVE
.
( )
AIR
TIGHT
V-Y PUMP
Figure A-L Total particulate train.
-------
CO
PROBE
FLOW
FILTER HOLDER
(0.3AC FIBERGLASS FILTER)
STACK
WALL
0.3>u. FILTER-
500 ml IMPINGERS
THERMOMETER
CHECK
VALVE
HEATED-
COMPARTMENT
1
LONE b
> a
D d
D ^
D d
D
-------
PROBE
FLOW
CONDENSER
STACK FT
WALL
ICE
BATH
CYCLONE
_ j
WATER TRAP
HEATED AREA
THERMOMETER
1
CHECK
VALVE
VACUUM
LINE
00
ON
THERMOMETER
BY-PASS
VALVE
VACUUM
GAUGE
CIRY TEST
METER
MAIN VALVE
( AAIR TIGHT
V JPUMP
V—*s
Figure A-3. Particulate interference sampler train.
-------
bath. As a result of the difficulties encountered with these filters
during the preliminary tests, the decision was made to eliminate the
membrane filter entirely and to replace the fiberglass filter with a
cyclone collector. Glass wools were packed in the outlet portion of the
cyclone to capture Si, Na, and K that may have penetrated the cyclone
collector. This train was run at the inlet duct.
Train 4 - Mercury/Backup Vapor-Phase Trace Elements
This train is shown schematically in Figure A-4. The strong oxidiz-
ing solutions of KMn04 are effective absorbers for Hg. Impinger 1 is
identical to impinger 1 used in train 2. Simultaneous sampling was done
at the inlet and outlet of the ESP. Two runs were made each day: one
in the morning and one in the afternoon. A total of eight samples were
collected at each of the two sampling locations.
Three separate solutions were retained for samples collected from
this train. The first was the catch in the K^COa solution from impinger 1.
The second was the combined catch from the KMn04 and HNOa solutions in
impingers 2-5. The third was the solution obtained from rinsing impingers
2-5 with 10-percent oxalic acid solution.
Brink Cascade Impactor
Figure A-5 is a schematic illustration of the BSM11, 5-stage Brink
impactor for particle size classification. Because of its lower gas sam-
pling flow rate and larger capacity, the Brink impactor was used at the
ESP inlet. Teflon and aluminum substrates were used with the stages,
but Teflon was found to be a very poor collection substrate during pre-
liminary tests. Final tests were conducted using aluminum substrates.
Runs of 2.5 to 7 minutes were made with the impactor in the morning
and again in the afternoon during each sampling day. The impactor was
inserted to a depth of five feet into the duct, and the sampling rate
was adjusted to achieve isokinetic sampling from an experimentally pre-
determined flue gas velocity at that point.
Andersen Cascade Impactor
Figure A-6 shows a schematic diagram of the Andersen system. Because
of its higher volume flow rate, this impactor was used at the ESP outlet
to obtain size distribution information. Fiberglass filters were used
as substrate. The sampling procedure used was identical to that followed
for the Brink Cascade Impactor. Test runs lasted from 37 to 360 minutes.
Optical and Diffusional Sizing System
An optical particle counter (Climet) and a diffusion battery (Thermo
Systems, Inc.), coupled with a condensation nuclei counter (General Elec-
tric), were used to measure the number, concentration, and size distribu-
tion of particulates in the size range from 0.02 pm up to 20 pm. A special
87
-------
PROBE
STACK WALL
REVERSE
TYPE PITOT
TUBE
oo
OO
PITOT
MANOMETER
$*•
-VJ,
ICE BATH
THERMOMETER
DRY GAS
METER
^ ITZ^
ol
\ / Sr
NEEDLE VALVE
tJ
ROTAMETER
PUMP
T)-200 ml, 10% K2C03 (WEIGHT/VOLUME)
20 ml,(3%KMn04 + 14.3%
-EMPTY
SILICA GEL
DRYING TUBE
Figure A-4. Mercury
-------
CYCLONE
n^p*^
A TCH
t ^
FLOW
IMPACTOR
1 1 1 1
^
-Jn 1
STACK
WALL
LJ
PROBE
^^TO Hg
MANOMETER
X
THERMOMETER
CONDENSER
00
BY-PASS
VALVE
VACUUM
GAUGE
CHECK
VALVE
MAIN VALVE
OAIR TIGHT
PUMP
Figure A-5. Brink cascade impactor sampling train.
-------
IMPACTOR
FLOW
STACK
WALL
THERMOMETER
CONDENSER
n
WATER TRAP
THERMOMETER
DRY TEST
METER
VACUUM
GAUGE
/ AAIR TIGHT
I JPUMP
V—\
VACUUM,
LINE;
Figure A-6.Andersen impactor sampling train.
-------
instack sample dilution system was designed by MRI for use with this par-
ticle measurement system. The sample dilution was affected by the injec-
tion of high-pressure clean air into one end of the dilution chamber which
was inserted in the stack. The vacuum created by the high-pressure air
jet drew flue gas into the chamber through a nozzle into a cyclone; and
the flue gas was diluted 50-fold in the chamber by the high-pressure air
before it could reach the sample probe of the optical and diffusional
sizing system. Figure 11 is a schematic of this sampling system.
Measurements were made alternately at the inlet and outlet ducts of
the ESP. Malfunction of the vacuum pump for the optical particle counter
reduced the number of runs that could be made with this instrument. Four
inlet and four outlet runs were conducted using the Climet optical counter.
Seven inlet and seven outlet runs were conducted using the diffusion
battery-condensation nuclei counter system.
Polycyclic Organic Material (POM) Sampling System
The POM system consists of a standard EPA Method 5 train modified
to include a Tenax adsorbent sampling system between the filter and the
impingers (Figure A-7). In operation, the stact gas is sampled isokineti-
cally by a sampling probe and passed through a heated filter. It then
passes into the cooling coil of the Tenax adsorbent sampler, up through
a Pyrex frit, and into the Tenax column. The cooling coil and Tenax
adsorbent are maintained at about 60°C by means of a thermostatically
controlled bath. The gas leaving the sampler goes through an aqueous
impinger, a Drierite trap, and a dry gas meter. At the conclusion of
the run, the adsorbent sampler is sealed and stored in darkness for
subsequent solvent extraction of the POM's.
Because of physical limitations, POM's were sampled through ports
on the adjacent inlet and outlet ducts of the ESP instead of the same
ducts being used for other samples. A total of three complete runs was
made—two at the outlet and one at the inlet of the ESP. Each sampling
run lasted for about 8 to 10 hours.
91
-------
DISC
PLUG
JACKET
FRITTED DISC
JOINT
A-?,
-------
Appendix B
METHODS OF CHEMICAL ANALYSIS
93
-------
METHODS OF CHEMICAL ANALYSIS
1. Analytical Procedures for Coal
l.a Be, Ca, Cr, Cu, Fe, K, Mg, Mn, Ni, Pb, Ti, V, and Zn in coal
l.b Al, Ba, Cd, Co, Na, and Si in coal
l.c As in coal
l.d Se and Sb in coal
l.e Cl in coal
l.f F in coal
l.g Hg in coal
2. Analytical Procedures for Particulate and Impinger Water from
Train 1 Inlet to ESP
2.a Be, Ca, Cr, Cu, Fe, K, Mg, Mn, Ni, Pb, Ti, V, and Zn in particulate
2.b Al, Ba, Cd, Co, and Na in particulate
2.c As in particulate
2.d Se and Sb in particulate
2.e Cl in particulate
2.f F in particulate
2.g Hg in particulate
2.h Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Ti, V, and
Zn in impinger water
2.i As in impinger water
2.j Cl in impinger water
2.k F in impinger water
2.1 Se and Sb in impinger water
2.m Si in impinger water
94
-------
3. Analytical Procedures for Particulate and Impiuger Water from
Train 1 Outlet to ESP
3.a Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Ni, Pb,
Sb, Se, Ti, V, and Zn in participate
3.b Cl and F in particulate
3.c Impinger water
4. Analytical Procedures for Vaporous Phase Trace Elements from Train 2
4. a Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Ti, V, and
Zn in K2C03 or KOH matrix
4.b As and Sb in K2C03 or KOH matrix
4.c Cl and F in K2C03 or KOH matrix
4.d Se in K2C03 or KOH matrix
4.e Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Ti, V, and
Zn in H202 matrix
4.f As and Sb in H202 matrix
4.g Cl and F in H202 matrix
6
4.h Se in H202 matrix
4.i Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ti, V, and Zn in
matrix
4.j Pb and Ni in H2S04 matrix
4.k As in H2S04 matrix
4.1 Cl and F in H2S04 matrix
4.m Se and Sb in H2S04 matrix
5. Analytical Procedures for Mercury from Train 4
5.a Hg in K2C03 matrix
5.b Hg in KMn04 matrix
5.c Hg in oxalic acid rinse
95
-------
6. Analytical Procedures for Particulate from the Brink Impactor
6.a Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Na, Ni,
Pb, Sb, Se, Ti, V, and Zn in particulate
6.b Cl and F in particulate
7. Analytical Procedures for Particulate from the Andejrseji Impactor
7.a Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Na, Ni,
Pb, Sb, Se, Ti, V, and Zn in particulate
7.b Cl and F in particulate
8. Analytical Procedures for ESP Hopper Ash
96
-------
METHODS OF CHEMICAL ANALYSIS FOR TEST SAMPLES
Eight analytical instrumental methods, as shown in Table B-l, were
used for the chemical analysis of the collected samples. The procedures
for analyzing samples were selected by TVA's laboratory according to sample
size and sample matrix. One test run of fine particulate sample collected
with the Brink and Andersen Impactors was analyzed by Accu-Lab, Inc., by
Spark Source Mass Spectrometry.
Coal and particulate (total particulate train at the precipitator
inlet, ESP hopper ash) samples had to be converted to solution for chemi-
cal analysis. They were subjected to acid or base dissolution, lithium
metaborate fusion, or Eschka digestion followed by acid dissolution.
Exceptions were the Train 1 inlet, Andersen and Brink impactor samples,
in which ethanol was used to dislodge the elements from the substrates
prior to dissolution by a mixture of hydrofluoric and perchloric acids.
Identical treatment of the substrate was made to obtain a background
correction.
Because of the extremely small quantity of particulate collected in
the probe of the total particulate train at the precipitator outlet, col-
lected on the Brink impactor pans, and collected on the post-ESP filter
substrate associated with the total particulate train and Andersen impactor,
microchemical techniques were employed in analyzing these samples.
Liquid samples (such as KOH, I^COa, KMn04, and impinger water) were
treated with either acid or oxidizing reagents before chemical analysis.
Other liquid samples (such as ^02 , ^804 or theoxalic acid rinse), were
diluted with reagent water and analyzed, directly analyzed, acidified,
or treated with oxidants prior to chemical analysis.
1 . Analysis Procedure for Coal
The coal was removed from the sealed container, oven-dried at 40°C
for 16 hours, and air-dried for 2 hours. The sample was then ground to
pass through a No. 60 mesh before being ashed or digested.2
Figure B-l depicts the analytical procedure for the determination
of trace elements in coal. As shown, the coal was ashed and then digested
for the determination of Be, Cr, Cu, K, V, Zn, Ca, Fe, Mg, Mn, Ti, Ni,
and Pb. The coal was digested for the determination of Si, Al, Ba, Cd,
Co, Na, As, Se, Sb, Cl, F, and Hg.
Five grams of the coal sample were accurately weighed into a porce-
lain dish and placed in a cold muffle furnace. The furnace was heated
to a temperature of 300°C for one hour and then increased to 500°C for
two hours. The ash was stirred and heated until all carbonaceous mate-
rial disappeared. The sample was cooled and ground in an agate mortar.
The coal was reignited at 500°C for one hour, cooled rapidly, and weighed
immediately. 2
97
-------
TABLE B-l
METHODS OF CHEMICAL ANALYSIS
00
Train 1 Mercury
Impunger Train 2
Water Vapor Phase Train Oxalic Acid 75 ml
Type Train 1 Train 1 Inlet & K2C03 or KM 04 10 ml
Sample Coal Inlet Outlet Outlet KOH H202 H2S04 H26o3 100 ml
Sample Size
Required 10 Grams 3.5 Grams 0.1 Grams 105 mis 200 mis 200 mis 200 mis
Inductively Be,Ca,Cr Be,Ca,Cr Al,As,Ba,Be Al.Ba.Be Al,Ba,Be Al,Ba,Be,Ca Al,Ba,Be
Coupled
Argon Cu,Fe,K Cu,Fe,K Ca,Cu,Co,Cr Ca,Cd,Co,Cr Ca,Cd,Co,Cr Cd,Co,Cr Ca,Cd,Co
Plasma
Atomic Mg,Mn,Ni Mg,Mn,Ni Fe,K,Mg,Hn Fe,Mg,Mn,Ni Cu,Fe,Mg Cu,Fe,Hg Cr,Cu,Fe
Emission Pb,Si,Ti Pb,Si,Ti Ni,Sb,Ti,V Pb,Ti,V,Zn Mn,Ni,Pb Mn,Ni,Pb Hg,Mn,Ti
Spectroscopy V,Zn V,Zn Zn Ti,V,Zn Ti.V.Zn V,Zn
Atomic Al,As,Ba Al,As,Ba As,Na,K
Absorption Cd,Co,Hg Cd.Co.Hg Hg,Se Se,Sb,Se As,Sb,Se As.Sb.Se As,Ni Hg
Na,Se,Sb Na,Se,Sb Pb,Sb,Se
Brink Andersen ESP Hopper
Impactor Impactor Ash
1 mg 10 mg
Al,As,Ba,Be Al,As,Ba,Be Be.Ca.Cr
Ca,Co,Cr,Cu Ca,Co,Cr,Cu Cu,Fe,K
Fe,K,Mg,Mn Fe,K,Mg,Mn Mg,Mn,Ni
Na,Ni,Sb Na,Ni,Sb Pb,Si,Ti
Ti,V,Zn Ti,V,Zn V,Zn
Al,As,Ba
Hg,Se Hg.Se Cd,Co,Hg
Na,Sb,Se
Gravimetry Cl
Potentiometric
Tritration Cl Cl Cl Cl Cl
Ion
Chroma tography CL,F
Speci fie
Ion F F F F F F
Electrode
Differential
Pulse
Anodic
Stripping Cd,Pb
Voltametry
Cl
CL,F C1,F
F
Cd,Pb Cd.Pb
Colorimetry Si
-------
Ashed, Digested with HF and HCIC>4
Atomic Emission
Digested with Lithium Metaborate
Atomic Emission
Be, Ca, Cr, Cu,
Fe, K, Mg.Mn, Ni,
Pb, Ti, V, Zn
Si
Digested with Lithium Metaborate
Digested with HMOs and
Digested with Eschka Mixture
Digested with Eschka Mixture
Atomic Absorption
Atomic Absorption
Atomic Absorption
Potentiometric Titration
Al, Ba, Cd,
Co, Na
As
Se, Sb
Cl
Digested with NaOH
Specific Ion Electrode
^Digested with Aqua Regia
Atomic Absorption
Hg
Figure B-l. Analytical procedures for trace element analysis of coal
99
-------
1-a Be, Ca, Cr, Cu, Fe, K, Mg, Mn, Ni. Pb. Ti, V, and Zn in Coal
A ground and weighed ash sample (approximately 0.05 grams) was
placed in a 1-ml platinum crucible; and 0.5-ml Ultrex grade HF and 0.05-ml
Ultrex grade HC104 were added. The mixture was heated to dryness at a
temperature of 250°C. The evaporation was repeated twice with new portions
of the acids. The residue was cooled and moistened with 0.05-ml Ultrex
HC1 and 0.5-ml deionized water. The mixture was covered and heated, with-
out boiling, in a sand bath approximately 5 minutes until all salts were
brought into solution. The solution was cooled, transferred to a volu-
metric flask, and made up to 5.0 ml with 5 percent Ultrex grade HC1.2 A
1.5-2.0-ml aliquot of this solution was used to determine the concentra-
tion of Be, Cr, Cu, K, V, Zn, Ni, and Pb by atomic emission. A 0.05-ml
aliquot was diluted to 1.05 ml (21X) with five percent Ultrex HC1 and
used to determine the concentration of Ca, Fe, Mg, Mn, and Ti by atomic
emission.1
l.b Al, Ba, Cd, Co, Na, and Si in Coal
A ground and weighed coal sample (0.5 grams) was mixed with 2.0 grams
of reagent grade lithium metaborate in a platinum crucible and heated in
a muffle furnace at 900°C until a clear melt was obtained. The mixture
was cooled and placed in a beaker containing 8 ml of reagent grade HN03
and 150 ml of deioni2ed water. The mixture was stirred vigorously until
all material dissolved, and tartaric acid was added at a concentration
of 1 percent. The solution was made up with deionized water in a volu-
metric flask5 to a final volume of 250 ml. One aliquot of this solution
was used to determine the concentration of Si by atomic emission.1 A
second aliquot of this solution was used to determine the concentrations
of Al, Ba, Cd, Co, and Na by atomic absorption.4
1.c As in Coal
For the determination of As, an accurately weighed 0.5 g sample of
ground coal was placed in a generating flask and mixed with 7.0 ml rea-
gent grade (1 + 1) H2S04 and 5.0 ml concentrated redistilled HN03, and
was evaporated to S03 fumes.11 The mixture was cooled, 25.0 ml of deio-
nized water were added, and the evaporation procedure was repeated to
expel oxides of nitrogen. The mixture was diluted to 100.0 ml with
deionized water; and the arsenic was converted to the hydride by adding
KI, SnCl2 and powdered zinc.12 The gaseous arsenic hydride was swept
into an argon-diluted, air-entrained hydrogen flame of an atomic absorp-
tion spectrophotometer.
l.d Se and Sb in Coal
A ground and weighed coal (1.0 gram) was mixed with 1.5 to 2.0 grams
of Eschka mixture (2 parts by weight of light calcined magnesium oxide to
one part of anhydrous sodium carbonate) in a porcelain crucible. This
mixture was covered with an additional 1.5 grams of Eschka mixture. The
crucible was placed in a cold muffle furnace and heated to 500°C for one
100
-------
hour; the temperature was then increased to 750°C for 2-1/2 hours. It
was necessary to maintain oxidizing conditions while ashing; therefore,
air was forced at one litre/minute through a 1/4-inch stainless steel
tubing located in the front of the furnace. Selenium was totally lost
from coal in the absence of oxidizing conditions. Exhausted gases passed
through a 1/4-inch stainless steel tube in the back of the furnace.
The crucible was cooled and the contents stirred with a platinum rod.
If any unburned coal remained, the crucible was returned to the furnace
and heated until all coal was ashed. The ash was transferred to a 150-ml
beaker which contained 20 ml hot deionized water and 3 grams of tartaric
acid. The crucible was rinsed carefully with 5 ml of concentrated reagent
grade HC1 and mixed with the contents in the beaker.
Two additional 5-ml aliquots of concentrated HC1 were added to the
mixture which was stirred after each addition. If any ash remained, addi-
tional acid was added to bring all material into solution. The solution
was then diluted with deionized water to a final volume of 100 ml. This
solution was used to determine the concentrations of Se and Sb.4
For Se, an aliquot of this solution was mixed with an equal amount
of concentrated reagent grade HC1 and placed in a water bath for at least
20 minutes to reduce Se6+ to Se4+. The selenium was converted to the
hydride by adding SnCl2.10 The gaseous selenium hydride was swept into
an argon-diluted, air-entrained hydrogen flame of an atomic absorption
spectrophotometer.4
For Sb, a 15-ml aliquot of the final solution was added to 5 ml of
reagent grade concentrated HC1. KI and SnCl2 were added to the solution
and set aside for 20 minutes. Sodium borohydride was added to convert
antimony to the hydride.9 The gaseous antimony hydride was swept into
an argon-diluted, air-entrained hydrogen flame of an atomic absorption
spectrophotometer.4'9
l.e Cl in Coal
A ground and weighed coal sample (1.0 gram) was mixed with three
grams of Eschka mixture in a procelain crucible. This mixture was covered
with an additional two grams of Eschka mixture. The crucible was placed
in a cold muffle furnace and gradually heated to 675 ± 25°C within one
hour and was maintained at this temperature for an hour thereafter. The
mixture was stirred with a platinum rod; and if any unburned coal remained,
the crucible was returned to the furnace and heated until all coal was
ashed. The ash was dissolved in 50 percent (V/V) nitric acid and the
chloride was determined by potentiometric titration with a standard
solution of silver nitrate.
l.f F in Coal
A ground coal sample (0.1 g) was weighed into a 50-ml nickel cruci-
ble. Two grams of NaOH (three milliliters of a 0.67 g/ml NaOH solution)
were added and oven-dried for 3-1/2 hours at 150°C. The crucible was
placed in a muffle furnace at 300°C, heated to 600°C, and ashed for three
101
-------
hours The ash was dissolved in a small amount of deionized water and
transferred to a distillation flask containing silver perchlorate solu-
tion. After rinsing the crucible and adding the rinse solution to the
flask the fluoride was isolated by steam distillation and determined
by the selective ion electrode.8
l.g Hg in Coal
A ground and weighed coal sample (0.2 g) was placed in a glass bottle.
Five milliliters of deionized water and 5 ml of aqua regia (three volumes
of concentrated reagent grade HC1 to one volume of concentrated reagent
erade HMOs) were added, and the solution was heated two minutes in a water
bath at 95°C. After the solution was cooled, 50 ml of deionized water
and 15 ml KMn04 were added.
The solution was heated over a water bath for 30 minutes at 95°C
and then cooled to room temperature. Six milliliters of sodium chloride-
hydro-xylamine sulfate were added to reduce excess permanganate. Fifty-
five milliliters of deionized water and 5 ml of stannous sulfate were
added to the solution. The mercury was determined by cold vapor atomic
absorption.4
2 Analysis Procedures for Particulate and Impinger Water from Train 1
Inlej^to ESP
The particulate collected by the train 1 inlet was impinged on 90
mm spectrograde glass fiber filters. The filters were packaged in jars
and sealed with plastic lids. A 250-ml Pyrex beaker covered with wax
paper contained the free, dry fine particulate ash sample trapped by the
cvclone (dry catch). Also a moist, charcoal-like particulate sample was
collected by scrubbing the probe with acetone (probe rinse) that was
nackaeed separately in a 250-ml Pyrex beaker covered with wax paper.
The impinger water was stored in a glass sample bottle. Figure B-2
represents the analytical procedures for train 1 particulate and
impinger water.
The filter was removed from the jar with platinum-tipped forceps.
As much particulate matter as possible was knocked off with a platinum
rod and composited with the dry catch in a tarred plastic cup. The weight
of the particulate was determined by the difference—sufficient (>4 grams)
for ordinary macrochemical analysis.
2 a Be (x_^J^J^J^ ME. Mn, Ni, Pb, Ti, V, and Zn in Particulate
About 50 mg of sample were weighed accurately into a 1-ml platinum
•hie and 0 5-ml Ultrex HF and 0.05-ml Ultrex HC104 was added. The
mixture was heated to dryness at a temperature below 250°C. The evap-
+ -nn w«<; repeated two more times with new portions of the acids. The
due was cooled and moistened with 0.05 ml Ultrex HC1 and 0.5 ml deio-
nized water. The mixture was covered and heated, without boiling, in a
102
-------
Digested with HF and HCKXj
Atomic Emission
Digested with Lithium Metaborate
Atomic Emission
Digested with Lithium Metaborate
Digested with HN03 and
Digested with Eschka mixture
Digested with Eschka mixture
Digested with NaOH
Digested with Aqua Regia
Direct Determination
Treated with H2S04 and HN03
Titrated with AgN03
Adjusted to pH 5
Atomic Absorption
Atomic Absorption
Atomic Absorption
Potentiometric Titration
Specific Ion Electrode
Atomic Absorption
Atomic Emission
Atomic Absorption
Potentiometric Titration
Specific Ion Electrode
Direct Determination
Atomic Absorption
^Digested with NaHC03
Colorimetry
Be, Ca, Cr, Cu,
Fe, K, Mg, Mn,
Ni.Pb.Ti , V,Zn
Si
Al, Ba, Cd,
Co, Na
As
Se, Sb
Cl
Hg
Al, Bo, Be, Co,
Cd,Co,Cr,Cu,
Fe, Mg,Mn,Ni,
Pb.Ti.V, Zn
As, Sb, Se
Cl
Na, K
Si
Figure B-2. Analytical procedures for
trace element analysis of train I inlet to ESP.
103
-------
sand bath until all salts and particulate matter were brought into solu-
tion. The cooled solution was transferred to a volumetric flask and made
up to a final volume of 5.0 ml with 5 percent Ultrex HC1.
A 1.5-2.0-ml aliquot of this solution was used to determine the con-
centration of Be, Cr, Cu, K, V, Zn, Ni, and Pb by atomic emission.1
A 50 pi (0.05-ml) aliquot of the solution was diluted to 1.05 ml
(21X) with 5 percent Ultrex HC1 and used to determine the concentration
of Ca, Fe, Mg, Mn, and Ti by atomic emission.1
2.b Al, Ba, Cd, Co, and Na in Particulate
An accurately weighed 0.5 g ground particulate sample was mixed with
2.0 g of reagent grade lithium metaborate in a platinum crucible and heated
in a muffle furnace at 900°C until a clear melt was obtained. The mixture
was cooled and placed in a beaker containing 8 ml of reagent grade HN03
and 150 ml of deionized water. The mixture was stirred vigorously until
all material dissolved; tartaric acid was then added at a concentration
of one percent. The solution was made up to a final volume of 250 ml
with deionized water in a volumetric flask.5
An aliquot of this solution was used to determine the concentration
of Si by atomic emission.1 A second aliquot of this solution was used
to determine the concentration of Al, Ba, Cd, Co, and Na by atomic
absorption.4
2.c As in Particulate
An accurately weighed 0.5 g sample of ground particulate was placed
in a flask, was mixed with 7.0 ml reagent (1 + 1) H2S04 and 5.0 ml concen-
trated redistilled HN03, and was evaporated to S03 fumes. The mixture
was cooled, 25.0 ml of deionized water was added, and the evaporation
procedure was repeated to expel oxides of nitrogen. The mixture was
diluted to 100.0 ml with deionized water, and the arsenic was converted
to the hydride by adding reducing reagents.12
The gaseous arsenic hydride was swept into an argon-diluted, air-
entrained hydrogen flame of an atomic absorption spectrophotometer.12'4
2.d Se and Sb in Particulate
An accurately weighed ground particulate sample (about 0.6 g) was
mixed with 1.5 to 2.0 g of Eschka mixture in a porcelain crucible. This
mixture was covered with an additional 1.5 g of Eschka mixture. The cru-
cible was placed in a cold muffle furnace and heated to 500°C for one
hour; the temperature was then increased to 750°C for 2-1/2 hours. It
was necessary to maintain oxidizing conditions while ashing; therefore,
air was forced at one litre/minute through a 1/4-inch stainless steel
tubing located in front of the furnace. Selenium is partially lost from
ash in the absence of oxidizing conditions. Exhausted gases passed
104
-------
through a 1/4-inch stainless steel tubing in the rear of the furnace.6
The crucible was cooled and the contents were stirred with a platinum
rod.
The mixture was transferred to a 150-ml beaker which contained 20 ml
of hot deionized water and three grams of tartaric acid. The crucible
was rinsed carefully with 5 ml of concentrated reagent grade HC1 and mixed
with the contents in the beaker.
Two additional 5-ml aliquots of concentrated HC1 were added to the
mixture which was stirred after each addition. If any particulate
remained, additional acid was added to bring all material into solution.
The solution was then diluted to a final volume of 100 ml with dionized
water. This solution was used for the determination of Se and Sb.4
For Se, an aliquot of this solution was mixed with an equal amount
of concentrated reagent grade HC1 and placed in a water bath for at least
20 minutes to reduce Se6+ to Se4+. The selenium was converted to the
hydride by adding SnCl2-4 The gaseous selenium hydride was swept into
an argon-diluted hydrogen flame of an atomic absorption spectrophotometer.6
For Sb, a 15-ml aliquot of the final solution was added to 5 ml of
reagent grade concentrated HC1. Two milliliters of 20 percent KI were
added to the solution and set aside for 20 minutes. SnCl2 was added to
convert antimony to the hydride.10 The gaseous antimony hydride was
swept into an argon-diluted hydrogen flame of an atomic absorption
spectrophotometer.6
2.e Cl in Particulate
A 1.0 g ground particulate sample was mixed with 1.5-2.0 g of Eschka
mixture in a porcelain crucible. This mixture was covered with an addi-
tional 1.5 g of Eschka. The crucible was placed in a cold muffle furnace
and heated to 500°C for 1 hour, and then the temperature was increased
to 750°C for 2-1/2 hours.
The particulate was transferred to a 150-ml beaker which contained
20 ml of hot deionized water. Forty milliliters of reagent grade HNOs
(1 + 1) were then added. The beaker was covered with a watch glass and
the contents stirred occasionally with a platinum rod to expedite solution.
An aliquot of this solution was titrated with 0.25N AgN03. Silver
and silver-silver chloride electrodes were used, and the endpoint was
determined by a potentiometer.7
2.f F in Particulate
For the determination of F, a 1.0 g ground particulate sample was
weighed into a 50-ml nickle crucible. Three milliliters of 2N NaOH solu-
tion were added and oven-dried for 3-1/2 hours at 150°C. The crucible
was placed in a muffle furnace at 300°C and then heated to 600°C for 3
hours.
105
-------
The particulate was dissolved in a small amount of deionized water
and 4 to 5 drops of 30 percent H202 were added to oxidize sulfites to
sulfates, and then boiled to destroy excess peroxide. Fluoride was iso-
lated by steam distillation and the concentration was determined by the
specific ion electrode method.8
2.g Hg in Particulate
To determine Hg, an 0.2 g ground particulate sample was placed in a
glass bottle. Five milliliters of deionized water and 5 ml of aqua regia
were added, and the solution was heated 2 minutes in a water bath at 95°C.
The solution was cooled and 50 ml of deionized water and 15 ml KMn04 were
added. The solution was heated over a water bath for 30 minutes at 95°C
and then cooled to room temperature. Six milliliters of sodium chloride-
hydroxylamine sulfate were added to reduce excess permanaganate. Fifty-
five milliliters of deionized water and 5 ml of stannous sulfate were
added; then the concentration of mercury was determined by atomic
absoption.
2.h Al, Ba, Be, Ca, Cd, Cu, Cr, Fe, Mg, Mn, Ni, Pg, Ti, V, and Zn
in Impinger Water
A 1.5-2.0-ml aliquot of the impinger water sample was used for the
direct determination of Al, Ba, Be, Ca, Cd, Cr, Cu, Fe, Mg, Mn, Ni, Pb,
Ti, V, and Zn.1
2.i As in Impinger Water
For the determination of As, a 25.0-ml aliquot was mixed with 7.0
ml reagent grade (1 + 1) H2S04 and 5.0 ml of concentrated redistilled
HN03, and was evaporated to SOs fumes. The mixture was cooled, and 25.0
ml of deionized water were added; the evaporation procedure was repeated
to expel oxides of nitrogen.11 The mixture was diluted to 100.0 ml with
deionized water and the arsenic was converted to the hydride by adding
SnCl2.10 The gaseous arsenic hydride was swept into an argon-diluted
hydrogen flame of an atomic absorption spectrophetometer.4'12
2.j Cl in Impinger Water
For Cl a 25.0-ml aliquot was pipetted into a 250-ml beaker. After
acidifying with sulfuric acid, the solution was boiled with peroxide to
destroy sulfite. After boiling in basic media to remove peroxide, the
solution was acidified with nitric acid and potentiometrically titrated
with standard silver nitrate.11
2.k F in Impinger Water
A 25.0-ml aliquot was mixed with 25.0 ml of buffer (pH 5.0-5.5) in
a Pyrex beaker. The concentration of fluoride was determined potentio-
metrically using a selective ion fluroide electrode.4
106
-------
2.1 Se and Sb in Impinger Water
These elements were determined after digestion by gaseous hydride
generation with an tomic absorption analytical finish.9'10
2.m Si in Impinger Water
The silica was determined by the automated molybdosilicate method.11
Automation by Technicon Autoanalyzer improved the detection limit from 1
to 0.1 mg/1.
3. Analytical Procedures for Par_t_icula_te___and Impinger Water from
Train 1 Outlet to ESP
Figure B-3 represents the analytical procedures used for Train 1
outlet particulate and impinger water.
The fine particulate collected by the Train 1 outlet was impinged
on 90 mm spectrograde glass fiber filters. The filters were packaged in
glass jars having plastic caps, [unlike the inlet, a separate free, dry
ash (dry catch) sample was absent]; but there was a moist, charcoal-like
particulate sample, collected by scrubbing the probe with acetone (probe
rinse) that was packaged separately in a 250-ml Pyrex beaker that was
sealed with wax paper. Because of the extremely small sample size, suf-
ficient particulate for chemical analysis could not be mechanically dis-
lodged from the filter or scraped from the beaker. Therefore, the
particulate was extracted with absolute ethanol by sonifying the material
on the filter and in the beaker. Because of the high background of these
substrates, it was thought the concentration of Na and Si were not valid;
so these results were not reported.
One-fourth of the filter was reserved for the analysis of chloride
and flouride. The glass filter was placed on a sheet of Teflon, and a
Teflon wedge one-fourth the size of the filter was used as a guide in
cutting a one-quarter section of the filter A surgical grade steel
scapel was used to cut the filter. Any particulate sticking to the
Telfon wedge was brushed onto the quartered filter with a camel hair
brush. The one-quarter section of the filter was used to determine F
and Cl in the particulate. The remaining three-quarter filter section
was used for the determination of Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu,
Fe, Hg, K, Na, Mg, Mn, Ni, Pb, Sb, Se, Ti, V, and Zn in particulate.
The probe rinse material received in the 250-ml Pyrex beaker was
extracted by Bonification with a sufficient amount of absolute ethanol
to cover all the particulate matter. This rinse material was transferred
quantitatively with additional ethanol to a tarred 100-ml Teflon beaker.
The Teflon beaker was dried under an infrared lamp for 10 minutes and
desiccated for 30 minutes over silica gel before it was tarred.
The three-fourth filter portion was dried on aluminum foil under an
infrared lamp for 10 minutes and desiccated over silica gel for 30 minutes
The tar weight was measured to the nearest one-hundredth of a milligram.
107
-------
Digested with HF and HCI04
Atomic Emission
Digested with HF and
Digested with HF and HCI04
Atomic Absorption
Voltammetry
Extracted with Na2C03 and NaHCOj
Direct Determination
Treated with H2SC>4 and HN03
Titrated with AgN03
Adjusted to pH 5
Direct Determination
IDigested with NaHC03
Ion Chromatogrophy
Atomic Emission
Atomic Absorption
Potentiometric Titration
Specific Ion Electrode
Atomic Absorption
Colorimetry
Al, As, Ba, Be,
Ca, Co, Cr, Cu,
Fe, K, Mg, Mn,
Ni, Sb, Ti, V, Zn
Se, Hg
Cd, Pb
Cl, F
Al, Ba, Be, Ca,
Cd, Co, Cr, Cu,
Fe, Mg, Mn, Ni,
Pb, Ti, V, Zn
As.Sb, Se
Cl
Na, K
Si
Figure B-3. Analytical procedures for
trace element analysis of train I outlet from ESP.
108
-------
The filter was removed from the foil with platinum-tipped forceps and
placed in the 100-ml Teflon beaker containing the ethanol slurry from
the probe rinse. Additional ethanol was added to cover the filter. The
particulate was removed from the filter by placing the beaker briefly in
an ultrasonic bath. The extracted filter was removed from the beaker
with platinum-tipped forceps, returned to the aluminim foil, and saved
for possible future examination. The ethanol was evaporated without
boiling and dried an additional 10 minutes under an infrared lamp. After
desiccation over silica gel for 30 minutes, the residue in the Teflon
beaker was reweighed and the weight extracted from three-fourths of the
filter; probe rinse material was determined by difference. The Teflon
beaker containing this combined residue was stored over silica gel for
chemical analysis of 22 elements (Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu,
Fe, Hg, K, Na, Mg, Mn, Ni, Pb, Sb, Se, Ti, V, and Zn).2
Each composited particulate residue (see Table B-2) was digested
with 5 ml Ultrex HF and one drop (0.05 ml) of Ultrex HC104. The mixture
was evaporated slowly, without boiling, to dryness in a sand bath at a
temperature not exceeding 220°C. The evaporation was repeated a second
time with 3 ml of Ultrex HF and one drop Ultrex HC104, and a third time
with 1 ml of Ultrex HF and one drop Ultrex HC104. The salts formed by this
triple evaporation were dissolved and diluted according to sample weight.
Referring to Table B-2, the salts formed by the triple evaporation
for the blank (0.49 mg), test 8 (113.57 mg), and test 4 (136.29 mg) were
mixed separately with 0.25 ml of concentrated hydrochloric acid (Baker
Ultrex reagent). After allowing a few minutes for dissolution, the mix-
ture was diluted with 2.5 ml of reagent water. The solution was covered
with a watch glass and heated just below boiling for about five minutes
or longer, as necessary, to dissolve the salts and large particulate clumps.
When the solution had cooled, the watch glass and beaker were washed with
1 ml of 5 percent (V/V) Ultrex hydrochloric acid; and the solution was
poured through a plastic funnel into a 5.0-ml volumetric flask. After
rinsing the beaker dropwise with an additional 10 drops (0.5 ml) of 5 per-
cent (V/V) Ultrex hydrochloric acid and combining this rinsing solution
with the solution in the volumetric flask, the solution was diluted to
exactly 5 ml with five percent (V/V) Ultrex hydrochloric acid. Immediately
the solution was poured into a 10-ml Teflon test tube and capped with a
Nalgene No. 16 polyethylene closure.
The salts formed by evaporation with hydrofluoric acid for tests 7
(149.24 mg), 9 (159.86 mg), and 1 (235.96 mg) were mixed separately with
0.5 ml Ultrex HC1 for partial dissolution and then 5.0 ml deionized water
with heating to dissolve the remaining particulate. After cooling to
room temperature, the solutions were transferred to separate 10.0-ml
volumetric flasks. The beakers were rinsed with 10 drops (0.5 ml) of 5
percent (V/V) Ultrex HC1 and added to the solution in the flasks. The
solutions were diluted to 10.0 ml with 5 percent (V/V) Ultrex HC1 and
stored in capped Teflon test tubes for chemcial analysis.
The salt residues for tests 5 (289.54 mg), 3 (443.11 mg), and 2
(618.88) were treated separately with 1.25 ml Ultrex HC1 and 12.5 ml
deionized water and heated until all salts and particulate matter had
109
-------
TABLE B-2. EXTRACTED WEIGHTS OF TRAIN 1 OUTLET PARTICIPATE1
Test No. Lab Extracted Weight (mg)
1
2
3
4
5
6
7
8
9
Blank
235.96
618.88
443.11
136.29
289.54
1403.72
149.24
113.57
159.86
0.492
Weight recovered by extracting and compositing three-fourths of the
material on the filter with all the probe material.
2Weight determined by extracting three-fourths of the blank filter
received with the samples.
dissolved. After cooling to room temperature, the solutions were trans-
ferred to separate 10.0-ml volumetric flasks. The beakers were rinsed
with 10 drops (0.5 ml) of 5 percent (V/V) Ultrex HC1 and added to the
solution in flasks. The solutions were diluted to 25.0 ml with 5 percent
(V/V) Ultrex HC1 and stored in Teflon bottles for chemical analysis.
The residue for test 6 (1403.72 mg) was mixed with 2.5 ml Ultrex
HC1 and 25.0 ml deioni2ed water and heated until all salts and particu-
late matter had dissolved. After cooling to room temperature, the solu-
tion was transferred to a 50.0-ml volumetric flask. The beaker was rinsed
with 10 drops (0.5 ml) of 5 percent (V/V) Ultrex HC1 and added to the
solution in the flask. The solution was then diluted with 5 percent (V/V)
Ultrex HC1 to a final volume of 50.0 ml and stored in a Teflon bottle
for chemical analysis.
3.a Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Ni, Pb,
Sb, Se, Ti, V, and Zn in Particulate
A 1.5-2.0-ml aliquot of each of the diluted solutions was used to
determine the concentration of Ba, Be, Cu, K, Ti, V, Zn, Co, As, Sb, Al,
Ca, Cr, Fe, Mg, Mn, and Ni by atomic emission.1 A 1-ml aliquot was used
110
-------
for the atomic absorption determination of Hg by the micro-cold vapor
technique. A 0.5-ml aliquot was used for the atomic absorption determi-
nation of Se by micro-gaseous hydride technique.3'4 A 1-ml aliquot was
treated with 2.0 ml of redistrilled HN03, and the concentrations of Cd
and Pb were determined by differential pulse anodic stripping voltammetry.13
3.b Cl and F in Particulate
The one-fourth section of the glass fiber filter containing the
particulate was folded and placed in a 15-ml graduated centrifuge tube.
This section of the filter was treated with 4.0 ml of a solution contain-
ing 0.0024 M Na2C03 and 0.003 M NaHC03. The centrifuge tube was sealed
with aluminum foil and sonified in a water bath and heated for an hour to
extract the soluble chloride and fluoride. The solution was cooled, and
a 250-^1 aliquot was used for the determination of Cl and F by ion chroma-
tography.14 The results for fluoride are for the free portion only because
the fluorosilicates were probably not extracted. Calculations for concen-
tration in ug/g are based on weights determined in the field (see Table B-3)
TABLE B-3. EXTRACTED WEIGHT OF TRAIN 1 OUTLET PARTICULATE
FOR Cl AND F1
Test No.
1
2
3
4
5
6
7
8
9
Field Weight (mg)
36.14
34.85
35.49
10.62
35.22
37.74
21.81
19.38
26.14
10ne-fourth of a filter was used for Cl and F
chemical analysis.
Ill
-------
3.c Impinger Water
The train 1 outlet impinger water was analyzed according to the
same procedures as the train 1 inlet.
4. Analytical Procedures for Vapor Phase Trace Elements from Train 2
The reagents used for the collection of vaporous trace elements were
K2C03, KOH, H202, and H2S04. The K2C03 reagent matrix was 10 percent
(W/V) K2C03 in water; the KOH reagent matrix was 1 M KOH in water; the
Hg09 reagent matrix was a mixture of 3 percent (1.8 M) H202, 0.14 M HN03,
and 0.02 M AgN03 in water; the H2S04 reagent matrix was concentrated sul-
furic acid. All reagents except K2C03 were reagent grade. The K2C03
was Ultrex grade from Baker Chemical Co. The anlytical procedures depicted
in Figure B-4 were used for the analysis of trace elements. As shown,
the K2C03 and HOH samples were analyzed by the same procedures.
4.a Al, Ba, Be, Ca, Cd, Co, Cr. Cu, Fe, Mg. Mn. Ni, Pb. Ti, V, and
Zn in the K2C03 and KOH Matrix
For the K2C03 and KOH samples, a 5.0-ml aliquot was placed in a 100-ml
volumetric flask and acidified with 1.0 ml redistilled HN03. The sample
was diluted with deionized water to a final volume of 100.0 ml. A 1.5-2.0-
ml aliquot of the final solution was used to determine the concentration
of Al, Ba, Be, Ca, Cd, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Ti, V, and Zn by atomic
emission.x
4.b As and Sb in K2C03 and KOH Matrix
A 15-ml aliquot of the sample was transferred to a 50-ml volumetric
flask. The sample was treated with 2.0 ml concentrated reagent grade
HC1 and diluted to a final volume of 50.0 ml with deionized water. As
and Sb were determined without digestion by gaseous hydride generation
with an atomic absorption analytical finish. 'l2
4.c Cl and F in K2C03 and KOH Matrix
For Cl a 25-ml aliquot was acidified with H2S04 and digested; the
Cl was determined by potentiometric titration with standard silver
nitrate.11 (Refer to section 2.j Cl in Impinger Water.) For F, a 25-ml
aliquot was acidified and the fluoride was determined by selective ion
electrode.4 (Refer to section 2.k F in Impinger Water.)
4.d Se in K2C03 and KOH Matrix
For Se, concentrated reagent grade HN03 was added to 30.0 ml to
achieve a pH<2. The sample was treated with 5 percent KM04 to maintain
a purple tint for at least 30 minutes. An equal volume of concentrated
reagent grade HC1 was added, and the sample was digested in a steam bath
112
-------
KgC03
or KOH
Matrix
Treated with HN03
Treated with HCI
'Digestion with Acidic
\Adjusted to pH 5
^Digestion with Acidic
Atomic Emission
Atomic Absorption
Potentiometric Titration
Specific Ion Electrode
Atomic Absorption
Al, Ba,Be,Ca,
led, Co, Cr, Cu, Fe,
I Mg, Mn.Ni, Pb,
Ti, V. Zn
I As, Sb
Cl
Se
Vapor
Phase
Train 2
H202
Matrix
H2SC>4
Matrix
Digestion with HN03
Diluted with Deionized Water
Adjusted to pH 5
Digestion with Acidic KMn04
Diluted with Deionized Water
Digestion with HN03
Titrated with AgNOj
Adjusted to pH 5
, Treated with HCI
Atomic Emission
Atomic Absorption
Gravimetric
-I Specific Ion Electrode
Atomic Absorption
Atomic Emission
Atomic Absorption
Potentiomelric Titration
Atomic Absorption
Al, Bo, Be, Co,
Cd, Co, Cr, Cu,
Fe, Mg, Mn, Ni,-
Pb, Ti, V, Zn
As, Sb
Cl
Se
Al, Bo, Be, Co,
ICd, Co, Cr,Cu,
| Fe, Mg, Mn, Ti,
V, Zn
Pb, Ni
Cl
~| Specific Ion Electrode I F
Sb, Se, As
Figure 6-4. Analytical procedures for
vaporous trace elements analysis of train 2
113
-------
at 95°C for at least 20 minutes. The sample was cooled and diluted with
6 M HC1 to a known volume and analyzed for selenium by atomic absorption
using the gaseous hydride technique.10
4.e Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn. Ni, Pb, Ti, V, and
Zn in H2C>2 Matrix
A 1.5-2.0-ml aliquot was used to determine the concentration of Al,
Ba, Be, Cu, Cd, Co, Cr, Ca, Fe, Mg, Mn, Ni, Pb, Ti, V, and Zn by atomic
emission.u) The concentrations were also verified by atomic absorption.4
4.f As and Sb in H202 Matrix
A 20.0-ml aliquot of the H^C^ matrix was digested to 863 fumes with
5.0 ml of concentrated redistilled HN03 and 5.0 ml reagent grade (1 + 1)
H2S04. The solution was cooled and 10 ml of distilled water and 10 ml
of concentrated redistilled HN03 were added. The solution was fumed again,
and the concentration of arsenic and antimony was determined by atomic
absorption using the gaseous hydride technique.9'12
4.g Cl and F in H202 Matrix
For the determination of Cl, a 10.0-ml aliquot of the H2C>2 matrix
was diluted to 100 ml with deionized water and filtered through a pre-
weighed glass fiber filter. The filter was rinsed with deionized water,
dried at 110°C, and reweighed. The difference in weight was the chlorine
capture.15 To determine F, an aliquot of the H^C^ matrix was mixed with
an equal volume of buffer (pH 5.0-5.5) in a Pyrex beaker. The concen-
tration of fluoride was determined potentiometrically using a selective
ion fluoride electrode.4
4.h Se in H202 Matrix
For the determination of Se, a 30.0-ml aliquot of H202 matrix was
treated with a sufficient amount of reagent grade 5 percent KMn04 (about
7 ml) to maintain a purple tint for at least 30 minutes. An equal volume
of concentrated reagent grade HC1 was added and the sample was digested
in a steam bath at 95°C for at least 20 minutes. The solution was cooled
and diluted with 6 M HC1 to 100 ml in a volumetric flask. The concentra-
tion of selenium was determined by atomic absorption using the gaseous
hydride technique.10
4.i Al, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ti, V, and Zn in
H2S04 Matrix
For the H2S04 samples, a 5.0-ml aliquot was transferred to a volu-
metric flask and diluted to 100.0 ml with deionized water. A 1.5-20.0-ml
aliquot was used to determine the concentration of Al, Ba, Be, Ca, Cd,
Co, Cr, Cu, Fe, Mg, Mn, Ti, V, and Zn by atomic emission.1
114
-------
4.j Pb and Ni in H2S04 Matrix
A 5.0-ml aliquot of the H2S04 matrix was transferred to a volumetric
flask, diluted to 100.0 ml with deionized water, and filtered through a
Gelman 0.45 Jjm membrane filter. The filter containing the insoluble mate-
rial was transferred to a 250-ml beaker and 3.0 ml of concentrated redis-
tilled HN03 was added. The beaker was covered with a watch glass and
heated gently to dissolve the filter. The temperature was increased to
digest the material and to evaporate the acid. The mixture was cooled
and 3.0 ml of concentrated redistilled HN03 were added. The beaker was
covered and heated until the digestion was completed. A 2.0-ml aliquot
of redistilled (1 + 1) HC1 was added and the residue heated gently to
dissolve the material. A small portion of deionized water was used to
wash down the sides of the beaker. The mixture was transferred to a
volumetric flask and diluted with deionized water to a final volume of
25 ml. An aliquot of the final solution was used to determine the con-
centration of Pb and Ni by atomic absorption.4
4.k As in H2S04 Matrix
A 15.0-ml aliquot of the H2S04 matrix was pipetted into 15.0 ml of
deionized water and then diluted to 50.0 ml with deionized water. The
concentration of arsenic was determined by atomic absorption using the
gaseous hydride technique.12
4.1 Cl and F in H2S04 Matrix
An aliquot was diluted. The chloride and fluoride were determined
according to sections 2.j and 2.k for these species in impinger water.
4.m Se and Sb in H2S04 Matrix
For Sb, a 15.0-ml sample of the H2S04 matrix was pipetted into a
50.0 ml volumetric and was diluted to 50.0 ml with deionized water. The
concentration of antimony was determined by atomic absorption using the
gaseous hydride technique.9 For Se, a 5.0-ml sample was added to 25 ml
of deionized water and 5 percent KM04 was added until a purple tint was
maintained for 30 minutes. An equal volume of concentrated reagent grade
HC1 was added and the sample was digested in a steam bath at 95°C for at
least 20 minutes. The sample was cooled and diluted with 6 M HC1 to a
known volume and analyzed for selenium by atomic absorption using the
gaseous hydride technique.10
5. Analytical Procedures for Mercury from Train 4
The reagents used for the collection of mercury vapor were I^COs
and KMn04> in HN03. There were three types of samples from the mercury
train: K2C03, KMn04 in HN03, and an oxalic acid rinse of the impingers
that contained the KMn04 solution. The K2C03 was 10 percent (W/V) K2C03
in water; the KMn04 in HN03 was three percent (W/V) KMn04 in 14.2 percent
115
-------
(V/V) HN03; the oxalic acid used for rinsing was 10 percent (W/V) oxalic
acid. All reagents except K2C03 were reagent grade. The K2C03 was Ultrex
grade from Baker Chemical Co. The analytical flow scheme is given by
Figure B-5.
5.a Hg in K2C03 Matrix
The mercury concentration in the J^COs matrix was determined by acidi-
fying 100 ml of sample with 5.0 ml of concentrated reagent grade h^SC^.
Thereafter, the solution was treated for mercury by the cold vapor atomic
absorption technique beginning with section 8.1 on page 124 of footnoted
reference.4
5.b Hg in KMn04 Matrix
A 10.0-ml sample of the KMn04 matrix was diluted with deionized water
to 100.0 ml. Thereafter, the solution was treated for mercury by the
cold vapor atomic absorption technique beginning with section 8.1 on page
124 of the footnoted reference.4 Hydroxylamine was added until the
permanganate color just dissipated.
5.c Hg in Oxalic Acid Rinse
A 75.0-ml sample of the oxalic acid rinse was diluted to 100.0 ml
with deionized water. Five milliliters of concentrated reagent grade
H2So4 and 2.5 ml concentrated redistilled HN03 were added. Hydroxylamine
and stannous chloride were added to generate atomic mercury for determi-
nation by the cold vapor atomic absorption technique.4 Calibration was
achieved in oxalic acid having the same concentration as that employed
for the diluted sample. Permanganate was not used to digest the sample
because permanganate is consumed endlessly by the oxalic acid.
6. Analytical Procedures for Analysis of Fine Particulate from Brink
Impactor
The fine particulate collected by the Brink Impactor was impinged
on aluminum pans and stored in a polyethylene cup. Figure B-6 represents
the analytical procedures used for trace element determination of the
fine particulate.
In order to obtain sufficient sample weight for analysis, it was
necessary to composit samples by tests and plates. Tables B-4 and B-5
represent the compositing scheme of the Brink Impactor samples.
6.a Al, As, Ba, Be. Ca. Cd, Co, Cr, Cu. Fe, Hg, K, Mg, Mn. Na, Ni,
Pb, Sb, Se, Ti, V, and Zn in Particulate
After removing the sample number from the polyethylene cup with a
Kim-Wipe moistened with absolute ethanol, the cup and the aluminum pan
116
-------
Atomic Absorption
Hg
Mercury
Train 4
1
KMn04
in HN03
Matrix
Oxalic Acid
Rinse
Atomic Absorption
Hg
Figure B-5. Analytical procedures for
mercury from train 4
117
-------
Ash
Digested with HF and HCI04
Atomic Emission
Digested with HF and HCI04
Atomic Absorption
Digested with HF and HCI04
Vol tammetry
Extracted with Na2C03 8 NaHCOj
Ion Chromatography
Al, As, Ba, Be.Ca,
Co, Cr, Cu, Fe, K,
Mg, Mn, Na, Ni,
Sb, Ti, V, Zn
Hg, Se
Cd, Pb
Cl, F
Figure B-6. Analytical procedures for
trace element analysis of Brink Impactor fine particulate
118
-------
TABLE B-4. COMPOSITING SCHEME FOR BRINK IMPACTOR FINE PARTICULATE FOR
DETERMINING 22 ELEMENTS
Composite of
Field Tests
1+10+11
15+16+17
3+12
13+14
1+10+11+15+16+17
3+12+13+14
1+10+11+15+16+17
3+12+13+14
Blank
Composite of
Plates
Oc+1
Oc+1
Oc+1
Oc+1
2+3
2+3
4+5
4+5
_
Approximate
Particle Size ((Jm)
>3
>3
>3
>3
1-3
1-3
0-1
0-1
_
Extracted Weight
(mg)
48.36
69.30
24.43
27.03
20.69
8.09
1.86
2.35
-
TABLE B-5. COMPOSITING SCHEME FOR BRINK IMPACTOR FINE PARTICULATE FOR
DETERMINING Cl AND F
Composite of
Field Tests
2
4
2
4
2
4
Blank
Composite of
Plates
Oc+1
Oc+1
2+3
2+3
4+5
4+5
_
Approximate
Particle Size (|Jm)
>3
>3
1-3
1-3
0-1
0-1
-
Extracted Weight
(mg)
18.50
24.20
4.37
1.97
0.22
0.18
0.02
119
-------
containing the fine particulate were dried for 10 minutes under an infra-
red lamp and desiccated for 30 minutes over silica gel. The tar weight
was measured to the nearest one-hundredth of a milligram. The aluminum
pan was placed in a 100-ml Teflon beaker with platinum-tipped forceps,
and any particulate matter remaining in the polyethylene cup was removed
with absolute ethanol and poured into the beakers. Additional ethanol
was added to the beaker to cover the aluminum pan.
The particulate matter was removed from the aluminum pan by briefly
placing the beaker in an ultrasonic bath. The pan was removed from the
beaker with platinum-tipped forceps and returned to the polyethylene cup
to be dried under an infrared lamp for 10 minutes. After desiccation
over silica gel for 30 minutes, the pan and cup were reweighed. The
weight extracted was determined by difference.
The next sample to be composited was extracted, dried, and weighed
as in the above procedure. This sample was combined with the material
in the beaker from the previous sample, and the added weight was deter-
mined by difference between empty and full sample vessels. This process
was repeated for each sample to yield the composited samples listed in
Table B-4. After compositing the samples, the ethanol was evaporated to
dryness under an infrared lamp and the beaker was stored in a desiccator.
The extraction and compositing scheme is similar to that used by Accu-Labs.16
The composited fine particulate (2-50 mg) in the 100-ml Teflon beaker
was treated with 1.5 ml of Ultrex HF and one drop (0.05 ml) of Ultrex
HC104. The mixture was evaporated slowly, without boiling, to dryness in
a sand bath at a temperature not exceeding 220°C. The evaporation was
repeated twice with 1.0 ml Ultrex HF and one drop (0.05 ml) Ultrex HC104.
The residue was cooled and moistened with 0.25 ml Ultrex HC1. After a
few minutes allowed for dissolution, the mixture was diluted with 2.5 ml
of deionized water, covered with a watch glass, and heated just below
boiling until all salts and particulate matter had dissolved. When the
solution had cooled, the watch glass and beaker were washed with 1 ml of
5 percent (V/V) Ultrex HC1 and combined with the solution in the volumetric
flask. The solution was diluted to 5.0 ml with 5 percent (V/V) Ultrex
HC1, poured into a 10-ml Teflon test tube, and capped with a polyethylene
closure.2
A 1.5-2.0-ml aliquot of the solution was used to determine the con-
centration of Ba, Be, Cu, K, V, Ti, Zn, Al, Ca, Cr, Fe, Mg, Mn, Ni, Co,
As, Sb, and Na by atomic emission.1 Another aliquot was used to determine
the concentration of Hg3 and Se10 by atomic absorption with cold vapor
and hydride generation, respectively. A microvapor vessel and microcell
were used for generating the atomic mercury in a 1-ml aliquot, and the
microvapor vessel was used to form gaseous selenium hydride in a 0.5-ml
aliquot.3 A 1-ml aliquot of the solution was treated with 2.0 ml of
redistilled HN03, and the Cd and Pb were determined by differential pulse
anodic stripping voltammetry beginning with section 8.4 in the footnoted
reference.13
The extraction and digestion procedures, repeated for an unexposed
aluminum pan, served as a blank for the chemical analysis.
120
-------
6.b Cl and F in Particulate
Two tests were reserved for the determination of Cl and F. Because
of small sample size, it was necessary to composite by plates. Table
B-5 shows the compositing scheme used for the fine particulate.
The fine particulate impinged on the aluminum pan in the polyethylene
cup (after removing the sample number from the cup with a Kim-Wipe moistened
with absolute ethanol) was dried for 10 minutes under an infrared lamp
and desiccated for 30 minutes over silica gel. The tar weight was measured
to the nearest one-hundredth of a milligram. The aluminum pan was placed
in a 30-ml Pyrex beaker with platinum-tipped forceps, and any particulate
matter remaining in the polyethylene cup was removed with absolute ethanol
and poured into the beaker. Additional ethanol was added to the beaker
to cover the aluminum pan.
The particulate matter was removed from the aluminum pan by briefly
placing the beaker in an ultrasonic bath. The pan was removed from the
beaker with platinum-tipped forceps and returned to the polyethylene cup
to be dried under an infrared lamp for 10 minutes. After desiccation
over silica gel for 30 minutes, the pan and cup were reweighed. The
weight extracted was determined by difference.
The next sample to be composited was extracted, dried, and weighed
as in the above procedure. This sample was combined with the material
in the beaker from the previous sample and the added weight was determined
by difference between empty and full sample vessels. This process was
repeated for each sample to yield the composited samples listed in Table
B-4. After compositing the samples, the ethanol was evaporated to dryness
under an infrared lamp and the beaker was stored in a desiccator.16
Depending on sample size (0.1-30 mg), 1-4 ml of a solution containing
0.0024 M Na2C03 and 0.003 M NaHC03 was pipetted into the beaker containing
the composited sample. This mixture was covered with aluminum foil and
sonified in a water bath for an hour and heated to extract the soluble
fluoride and chloride. The supernatant was decanted and centrifuged;
and about a 0.3-ml aliquot was withdrawn for determination of Cl and F
by ion chromatography.14 The results for fluoride are for the free
portion only because the fluorosilicates were probably not extracted.
7. Analytical Procedures for Particulate from the Andersen Impactor
The fine particulate ash collected by the Andersen Impactor was
impinged on stage filters. Gelman nonspectrograde AE glass fiber stage
filters having a diameter of 63.5 mm were used for collecting fine par-
ticulate. The filters were wrapped in aluminum foil for storage. Figure
B-7 represents the analysis procedures used to determine trace elements
in the fine particulate. In order to obtain sufficient sample size for
analysis, it was necessary to composite samples by tests and plates.
Tables B-6 and B-7 give the compositing scheme of the Andersen Impactor
samples.
121
-------
Ash
Digested with HF and HCI04
Atomic Emission
Digested with HF and HCI04
Atomic Absorption
Digested with HF and HCI04
Vol tammetry
Extracted with
8 NaHC03
Ion Chromatography
Al, As, Ba, Be.Ca,
Co, Cr, Cu, Fe,K,
Mg, Mn, Na, Ni,
Sb, Ti, V, Zn
Hg, Se
Cd, Pb
Cl, F
Figure B-7 Analytical procedures for
trace element analysis of Andersen Impactor fine participate
122
-------
TABLE B-6. COMPOSITING SCHEME FOR ANDERSEN IMPACTOR FINE PARTICULATE FOR
DETERMING 22 ELEMENTS
Composite of
Field Tests
1+5
2+3
7
1+5
2+3
7
1+5
2+3
7
Blank
TABLE B-7.
Composite of
Field Tests
6
6
6
Composite of Approximate Extracted Weight
Plates Particle Size (|Jm) (mg)
Oc+1+2+3
Oc+1+2+3
Oc+1+2+3
4+5
4+5
4+5
6+7+8
6+7+8
6+7+8
-
>3
>3
>3
1-3
1-3
1-3
0-1
0-1
0-1
-
COMPOSITING SCHEME FOR ANDERSEN IMPACTOR
DETERMINING F AND Cl
Composite of
Plates
Oc+1+2+3
4+5
6+7+8
Approximate
Particle Size i
>3
1-3
0-1
77.90
95.56
36.01
53.27
78.95
16.30
29.63
36.90
30.87
1.38
FINE PARTICULATE FOR
Field Weight
(Urn) (mg)
32.7
21.16
30.19
123
-------
7.a Al, As, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, Hg, K, Mg, Mn, Na, Ni , Pb,
Sb, Se, Ti, V, arid Zn in Particulate
The foil containing the filter was dried under an infrared lamp for
10 minutes and desiccated for 30 minutes over silica gel. The tar weight
was measured to the nearest one-hundredth of a milligram. The filter
was removed from the foil with platinum-tipped forceps and placed in a
100-ml Teflon beaker; any particulate clinging to the foil was removed
with absolute ethanol and added to the beaker. Additional ethanol was
added to the beaker to cover the filter. The particulate matter was
removed from the filter by briefly placing the beaker in an ultrasonic
bath. The filter was removed from the beaker with platinum-tipped for-
ceps and returned to the aluminum foil to be dried under an infrared
lamp for 10 minutes. After desiccation over silica gel for 30 minutes,
the foil and filter were reweighed and the weight extracted was deter-
mined by difference.
The next sample to be composited was extracted, dried, and weighed
as in the above procedure. This sample was combined with the material
in the beaker from the previous sample and the added weight was deter-
mined by difference between empty and full sample vessels. The process
was repeated for each sample to yield the composited samples listed in
Table B-6. After compositing the samples, the ethanol was evaporated to
dryness under an infrared lamp and the beaker was stored in a desiccator.16
The composited fine particulate (1-100 mg) was digested with 1.5 ml
of Ultrex HF and one drop of Ultrex HC104. The mixture was evaporated
slowly, without boiling, to dryness in a sand bath at a temperature not
exceeding 220°C. The evaporation was repeated twice with 1.0 ml Ultrex
HF and one drop (0.05 ml) Ultrex HC104. The residue was cooled and
moistened with 0.25 ml Ultrex HC1. After a few minutes allowed for dis-
solution, the mixture was diluted with 2.5 ml of deionized water. The
solution was covered with a watch glass and heated just below boiling
until all salts and particulate matter had dissolved. When the solution
had cooled, the beaker was washed with 1 ml of 5 percent (V/V) Ultrex
HC1, and the solution was transferred to a 5.0-ml volumetric flask. The
beaker was rinsed with an additional 10 drops (0.5 ml) of 5 percent (V/V)
Ultrex HC1 and added to the solution in the flask. The solution was
diluted to 5.0 ml with 5 percent (V/V) Ultrex HC1, poured into a 10-ml
Teflon test tube, and capped with a polyethylene closure.2
A 1.5-2.0-ml aliquot of the solution was used to determine the con-
centration of Ba, Be, Cu, K, V, Ti, Zn, Al, Ca, Cr, Fe, Mg, Mn, Ni, Co,
As, and Sb by atomic emission.1 Another aliquot was used to determine
the concentration of Hg3 and Se10 by atomic absorption with cold vapor
and hydride generating, respectively. A microvapor vessel and microcell
were used for generating the atomic mercury, and the microvapor vessel
was used to form gaseous selenium hydride. A 1-ml sample was used for
Se and a 0.5-ral sample was used for Hg. A 1-ml aliquot of the solution
was treated with 2.0 ml of redistilled HN03, and the concentrations of
Cd and Pb were determined by differential pulse anodic stripping vol-
tammetry beginning with section 8.4 in the footnoted reference.
13
124
-------
The extraction and digestion procedures, repeated for an unused non-
spectrograde filter, served as a blank in the analysis.
7.b Cl and F in Particulate
Chloride and fluoride were determined on the particulate captured
on the filters from test 6. The stage filters were folded and composited
according to Table B-7 in a 15-ml graduated centrifuge tube. The com-
posited filters were treated with 4.0 ml of a solution containing 0.0024
M Na2C03 and 0.003 M NaHC03. The centrifuge tube was sealed with aluminum
foil, sonified in a water bath, and heated for an hour to extract the
soluble fluoride and chloride. The solution was cooled and about 0.3 ml
was withdrawn for the determination of Cl and F by ion chromotography.14
The results for fluoride are for the free portion only because the fluo-
rosilicates are probably not extracted. Field weights given in Table
B-7 were used to determine concentrations of fluoride and chloride in
the particulate.
8. Analytical Procedures for ESP Hopper Ash
The anlytical procedures for the determination of trace elements in
ESP Hopper Ash are shown in Figure B-8. The ash was removed from sealed
containers without further preparation for chemical analysis by the pro-
cedures used to determine the 25 elements in the particulate from train
1 inlet to the ESP. (Refer to sections 2.a-g to learn the chemical
techniques.)
125
-------
Digested with HF and HCI04
Atomic Emission
Digested with Lithium Me tabor ate
Atomic Emission
Digested with Lithium Metaborate
Digested with HNOj and
Digested with Eschka Mixture
Digested with Eschka Mixture
.Digested with NaOH
[Digested with Aqua Regia
Atomic Absorption
Atomic Absorption
Atomic Absorption
Potentiometric Titration
Specific Ion Electrode
Atomic Absorption
Be, Ca, Cr, Cu,
Fe, K, Mg, Mn, Ni,
Pb, Ti, V, Zn
Si
Al, Ba, Cd,
Co, No.
As
Se.Sb
Cl
Hg
Figure B-8. Analytical procedures for trace element
analysis of ESP hopper ash
126
-------
REFERENCES
1. Jarrell-Ash Co. Instruction Manuals for Model 750 AtomComp and the
Inductively Coupled Argon Plasma Model 96-975 AtomComp., Waltham,
MA, December 1976.
2. Rose, John L., Jr. Personal communication on preparation of coal
and coal ash for trace element determinations under jurisdiction of
ASTM D-5 committee. TVA Transmission and Test Branch, Chattanooga,
TN, April 1976.
3. Hawley, J. E. and J. D. Ingle, Jr. Improvements in Cold Vapor Atomic
Absorption Determination of Mercury. Anal. Chem. 47; pp. 719-723,
1975.
4. U.S. Environmental Protection Agency, "Methods for Chemical Analysis
of Water and Wastes," Publication No. EPA-625-/6-74-003a. Cincinnati,
OH, 1976. pp. 92-98, 101-102, 107-108, 118-126, 134-138, 145, 147-148.
5. Boar, P. L. and L. K. Ingram. "The Comprehensive Analysis of Coal
Ash and Silicate Rocks by Atomic Absorption Spectrometry by a Fusion
Technique." Analyst 95_: 124-130, 1970.
6. Bosshart, Robert E. Personal communication on preparation of coal
and coal ash for trace selenium, antimony, and arsenic determinations
by gaseous hydride and atomic absorption. Bituminous Coal Research,
Inc., Monroeville, PA, January 1977.
7. American Society for Testing and Materials. "Standard Test Method
for Chlorine in Coal." D2361-66 (Reapproved 1972). Annual Book of
Standards, Part 26, 1976. pp. 315-317.
8. American Society for Testing and Materials. "Tentative Methods for
Analysis for Fluoride Content of the Atmosphere and Plant Tissues
(Manual Procedures)." D3269-73T. Annual Book of Standards, Part 26,
1975. pp. 706, 709.
9. Fishman, Marvin. Personal Communication on Procedure for Antimony.
USGS Analytical Methods Research, Denver Federal Center, Denver,
CO, 1977.
10. Lansford, M., E. M. McPherson and M. J. Fishman. "Determination of
Selenium in Water." Atomic Absorption Newsletter 13(4): 103-105,
1974.
11. American Public Health Association. Standard Methods for the Exami-
nation of Water and Waste Water, 14th Edition. New York, Amercian
Public Health Association, Publishers, 1975. pp. 285, 306-309,
487-490.
12. Howe, Lyman H. "Trace Analysis of Arsenic by Colorimetry, Atomic
Absorption, and Polarography." Tennessee Valley Authority, Division
of Environmental Planning, Publication Number TVA-E-EP-77-3/EPA-
600/7-77-036, April, 1977. pp. 7-8.
127
-------
13. American Society for Testing and Materials. "Proposed Method for
Determination of Cadmium and Lead by Differential Pulse Anodic
Stripping Voltammetry." Annual Book of Standards, Part 31, 1977.
pp. 1052-1058.
14. Dionex Corporation. Instruction Manual for Model 10 Ion Chromato-
graph, Sunnyvale, CA, September 1976.
15. Furman, V. Howell. Editor. "Standard Methods of Chemical Analysis"
Volume One - The Elements, Sixth Edition, Princeton, NJ, D. Van
Nostrand, Publishers, 1968. pp. 979-980.
16. Gilgren, William R. Personal Communication on preparation of very
small samples of fine particulate. CDM/Accu-Labs, Wheat Ridge, CO,
July 1977.
128
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1 REPORT NO.
EPA- 600/7-80-171
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE ANDSUBTITLE
?ield Study to Obtain Trace Element Mass Balances at a
Coal-fired Utility Boiler
5. REPORT DATE
October 1980
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert Evers, V.E. Vandergriff, and R.L, Zielke
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
TVA, Division of Energy Demonstrations and Technology
1140 Chestnut Street, Tower II
Chattanooga, Tennessee 37401
10. PROGRAM ELEMENT NO.
1NE624A
11. CONTRACT/GRANT NO.
EPA TAR-D5-E721
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
Task Final; 5/75-9/80
14. SPONSORING AGENCY CODE
EPA/600/13
xcoccxi >~ii J. i. irttigic jreiin., I'NV-. & i i 11 JCj IrA/OUU/1O
i. SUPPLEMENTARY NOTES IERL-RTP project officer is Julian W. Jones, Mail Drop 61, 919/541
i89.
15
2489.
16. ABSTRACT . ,
The report gives results of a study to identify mass flow rates or minor and
trace elements from streams of a coal-fired utility boiler (Colbert Steam Plant Unit
No. 1). This information was used to obtain a mass balance for 25 elements. The mass
balances used inlet and outlet flows associated with three major pieces of equipment:
the pulverizer, boiler, and electrostatic precipitator. This provided a mass balance
for each element for the various parts of the system. Along with the trace elements
which were being measured, organic samples were obtained and analyzed from various
streams for polychlorinated biphenyls (PCBs) and polynuclear organic matter (POMs).
Thus, the mass balance reflected a fairly complete picture of the boiler under normal
operating conditions. The mass balances show that sampling techniques need to be im-
proved. First, the analysis of the vapor-phase samples reported all concentrations
below the detection limit for each element. Second, the mass balances of only 10 ele-
ments (representing 61% of the total ash flow) closed within + or - 10% for at least
two of the three major pieces of the system.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTOF1S
Pollution
Mass Flow
Trace Elements
Chemical Analysis
Boilers
Coal
Combustion
Utilities
Chlorine Aromatic
Compounds
Biphenyl
Polycyclic Compounds
Organic Compounds
Sampling
b.IDENTIFIERS/OPEN. ENDED TERMS
Pollution Control
Stationary Sources
Mass Balances
COSATl Field/Group
13B
20D
06A
07D
13A
2 ID
21B
07C
14B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
143
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |