United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/7 79-222
October 1979
Research and Development
Evaluation of the
Ames Solid Waste
Recovery System
Part III
Environmental
Emissions of the
Stoker Fired Steam
Generators
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
systems, and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
US EPA-NEIC LIBRARY
'Denver Federal Center
Building 25, Ent. E-3
0 O Box 25227
Denver, CO 80225-0227
EPA-600/7-79-222
October 1979
EVALUATION OF THE AMES SOLID WASTE RECOVERY SYSTEM
PART III: ENVIRONMENTAL EMISSIONS OF THE STOKER FIRED STEAM GENERATORS
by
J. L. Hall, A. W. Joensen, D. Van Meter, R. Wehage, G. Severns, R. Reece
Iowa State University, Ames, Iowa
H. R. Shanks, Ames Laboratory, ERDA (DOE), Ames, Iowa
D. E. Fiscus, R. W. White, Midwest Research Institute, Kansas City, Missouri
EPA Grant No. R803903-01-0 to the City of Ames, Iowa ' *
ERDA (DOE) Contract No. W-7405 ENG-82
EPA Project Officers
Carlton C. Wiles
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
Robert A. Olexsey
Energy Systems Environmental Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
ERDA (DOE) Project Officer
Robert L. Butenhoff
Pollutant Characterization and Safety Research Division
Office of Health and Environmental Research
U. S. Department of Energy
Washington, D.C. 20535
This study was conducted in cooperation with the City of Ames, Iowa; Ames
Laboratory, DOE, Ames, Iowa; Engineering Research Institute and Mechanical
Engineering Department, Iowa State University, Ames, Iowa; Midwest Research
Institute, Kansas City, Missouri; and the Iowa Department of Environmental
Quality, Des Moines, Iowa
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on
our health often require that new and increasingly more efficient pollution
control methods be used. The Industrial Environmental Research Laboratory-
Cincinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
On August 30, 1975, the first continuous full-scale, solid waste recovery
system for the processing and burning of municipal solid waste as a supple-
mentary fuel for power generation commenced operation in the City of Ames,
Iowa. This report provides the results from the study of environmental emis-
sions from the stoker fired steam generators at the City of Ames. It provides
basic data on the environmental emissions when burning coal only and mixtures
of coal plus refuse derived fuel. The results and/or conclusions of this
report may be utilized to determine what changes occur in emissions when
changing from burning coal only to coal plus refuse derived fuel. The infor-
mation contained herein will be of interest to those designers or users who
are contemplating or working with a system similar to Ames. Requests for
further information regarding emissions of stoker fired steam generators
utilizing refuse derived fuel should be directed to the Fuels Technology
Branch, IERL, Cincinnati.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
The project entitled "Evaluation of the Ames Solid Waste Recovery System"
encompasses such a large scope of work and has generated such a large amount
of data that the annual report on year 01 is divided into three parts.
Part I, entitled "Summary of Environmental Emissions. Equipment, Facil-
ities, and Economic Evaluations" provides a summary of the environmental emis-
sions and boiler performance of stoker fired boilers burning refuse derived
fuel (RDF) and coal; characterization of the RDF produced by the processing
plant; processing plant and equipment performance evaluations; and an economic
analysis of the processing plant.
Part II, entitled "Performance of the Stoker Fired Steam Generators"
evaluates the thermodynamic and mechanical performance of the stoker boilers
while burning RDF'as a supplemental fuel with coal.
Part-III, entitled ."Environmental Emissions of the Stoker Fired Steam
Generators" describes the environmental impact of the stoker boiler cofiring
operation. The report includes sample analysis of the input and output
streams associated-with the operation of the stoker fired boilers while
burning coal only and coal plus RDF; characterization of the fuel (coal and
RDF), ash and stack effluents; and statistical analysis of the data.
The portion of the project covering environmental emissions from the
stoker boilers is jointly funded by the Environmental Protection Agency (EPA)
and the Department of Energy (DOE). These results are published jointly by
both agencies in Part III.
The balance of the project is funded by the EPA and these results are
published in Part I and Part II.
IV
-------
CONTENTS
Foreword
Abstract iv
Figures • vi
Tables x
Acknowledgments xiii
1. Introduction* ............... 1
2. Objectives 2
3. Summary of Results* ............ .. 3
4. Description of the Steam Generating Units 5
5. Experimental Design .......... .... 10
6. Sampling Methods 15
General 15
Input Streams 16
Exit Streams 18
7. Results 37
General 37
Input Streams 38
Exit Streams 43
References. ............................ 121
Appendix A - Statistical Analysis of Variance of Results ]_22
-------
FIGURES
Number Page
1 Gross sectional view of boiler unit No. 5. •••••••••• 7
2 Elevator view of boiler unit No. 6 and associated power
plant apparatus 8
3 Cross sectional view of boiler unit No. 6 ...... 9
4 Boiler unit No. 5 and unit No. 6 sampling locations 14
5 Sampling locations for isokinetic sampling of boiler unit
Nos. 5 and 6 exhaust gas stack ............... 19
6 Sampling locations for isokinetic sampling of boiler unit
Nos. 5 and 6 before the particulate collector 20
7 Photographs of sampling equipment 22
8 Block diagram of particulate emissions data flow 23
9 Block diagram of gaseous emissions data flow 24
10 Block diagram of trace element from filters data flow 25
11 Placement of Andersen Sampling Head in an EPA Method 5
sampling train 31
12 Stack particulate emission rate of boiler unit Nos. 5 and 6
as a function of refuse derived fuel heat input. ...... 49
13 Stack particulate emission rate of boiler unit Nos. 5 and 6
as a function of steam load 50
14 Particulate flow rate of boiler unit Nos. 5 and 6 before-
the-particulate-collector as a function of refuse derived
fuel heat input 51
VI
-------
FIGURES (Continued)
15 Particulate flow rate of boiler unit Nos. 5 and 6 before-
the-particulate-collector as a function of steam load. . • . 52
16 Particulate collector efficiency of boiler unit Nos. 5 and
6 as a function of refuse derived fuel heat input 53
17 Particulate collector efficiency of boiler unit Nos. 5 and 6
as a function of steam load 54
18 Percent excess air of boiler unit Nos. 5 and 6 as a function
of refuse derived fuel heat input. ... ..... 55
19 Percent excess air of boiler unit Nos» 5 and 6 as a function
of steam load 56
20 Overall cumulative particle size distribution of boiler
unit No. 5 stack particulate emissions ........... 63
21 Overall cumulative particle size distribution of boiler unit
No. 6 stack particulate emissions. ........ 64
22 Cumulative particle size distribution of boiler unit Nos.
5 and 6 stack particulate emissions for differing loads. . . 65
23 Cumulative particle size distribution of boiler unit stack
particulate emissions for differing percent RDF. .;.... 67
24 Cumulative particle size distribution of boiler unit No. 6
stack particulate emissions for differing percent RDF. ... 68
25 Corrected cumulative particle size percent less than D^Q of
boiler unit No. 5 stack particulate emissions as a function
of RDF and 60% load 69
26 Corrected cumulative particle size percent less than 050 of
boiler unit No. 5 stack particulate emissions as a function
of RDF and 80% load 70
27 Corrected cumulative particle size percent less than D^Q of
boiler unit No. 5 stack particulate emissions as a function
of RDF and 100% load 71
28 Corrected cumulative particle size percent less than D^Q of
boiler unit No. 6 stack particulate emissions as a function
of RDF and 80% load 72
vii
-------
FIGURES (Continued)
29 Corrected cumulative particle size percent less than D50 of
boiler unit No. 5 stack particulate emissions as a function
of steam load and 0% RDF .................. 73
30 Corrected cumulative particle size percent less than D5Q of
boiler unit No. 5 stack particulate emissions as a function
of steam load and 20% RDF .................. 7^
31 Corrected cumulative particle size percent less than 059 of
boiler unit No. 5 stack particulate emissions as a function
of steam load and 50% RDF .................. 75
32 NOV emissions from boiler unit Nos. 5 and 6 as a function of
j*i.
RDF heat input ....................... 85
33 NOX emissions from boiler unit Nos. 5 and 6 as a function of
boiler steam load ................ ...... 86
34 NOX emissions from boiler unit Nos. 5 and 6 as a function of
excess air ......................... 87
35 Sulfur emissions from boiler unit Nos. 5 and 6 as a function
of RDF heat input ...................... 90
36 Sulfur emissions from boiler unit Nos. 5 and 6 as a function
of boiler steam load ... ................. 91
37 Formaldehyde emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input ................. 95
38 Formaldehyde emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load ....... . ........ 96
39 Cyanide emissions from boiler unit Nos. 5 and 6 as a function
of RDF heat input ...................... 100
40 Cyanide emissions from boiler unit Nos. 5 and 6 as a function
of boiler steam load ................. ... 101
41 Phosphate emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input ................. 104
42 Phosphate emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load ................ 105
V2.ll
-------
FIGURES (Concluded)
43 Chloride emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input as determined by the SOX
train Ill
44 Chloride emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load as determined by the SOX
train 112
45 Chloride emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input as determined by the organic
acids train 113
46 Chloride emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load as determined by the organic
acids train 114
-------
TABLES
Number ~Page_
1 Characteristics of Ames Municipal Power Plant Steam
Generators "
2 Summary of Factorial Experimental Design 11
3 Test Summary of Environmental Sampling 12
4 Elements Determined and Analytical Method Used for Analysis
of Coal, Refuse Derived Fuel, and Ash Samples 17
5 Impinger Solutions for Sample Trains 26
6 Impinger Solutions for Mercury Vapor Sample Trains 27
7 Detection Limits of Elements Scanned by X-Ray Fluorescence
(XRF) 29
8 Detection Limits of Elements Scanned by Inductively
Coupled Plasma 30
9 Summary of Coal, RDF, and Ash Characteristics for Boiler
Unit No. 5 39
10 Summary of Coal, RDF, and Ash Characteristics for Boiler
Unit No. 6 41
11 Summary of Air, Feedwater, and Steam Characteristics for
Boiler Unit Nos. 5 and 6 42
12 Data Matrix of Unit No. 5 Stack Particulate Emissions .... 45
13 Data Matrix of Unit No. 6 Stack Particulate Emissions .... 46
14 Data Matrix of Unit No. 5 Particulate Flow Rate Before
the Particulate Collector 47
-------
TABLES (Continued)
15 Data Matrix of Unit No. 6 Particulate Flow Rate Before
the Particulate Collector 48
16 Data Matrix of Boiler Unit No. 5 Particulate Collector
Efficiency 57
17 Data Matrix of Boiler Unit No. 6 Particulate Collector
Efficiency 58
18 Example Sizing Data Reduction Sheet and Data Set 61
19 Data Matrices of Average Stack Orsat Data for Boiler
Unit No. 5 76
20 Data Matrices of Average Stack Orsat Data for Boiler
Unit No. 6 77
21 Data Matrices of Average Orsat Data Before the Particulate
Collector for Boiler Unit No. 5 78
22 Data Matrices of Average Orsat Data Before the Particulate
for Boiler Unit No. 6 79
23 Data Matrices of Average Orsat Data From All Sample
Locations on Unit No. 5 ' 80
24 Data Matrices of Average Orsat Data From All Sample
Locations on Unit No. 6 81
25 Data Matrix of NOX Emissions for Boiler Unit No. 5 83
26 Data Matrix of NOX Emissions for Boiler Unit No. 6 84
27 Data Matrix of SOX Emissions for Boiler Unit No. 5 88
28 Data Matrix of SOX Emissions for Boiler Unit No. 6 89
29 Data Matrix of Formaldehyde Emissions From Boiler
Unit No. 5 93
30 Data Matrix of Formaldehyde Emissions From Boiler
Unit No. 6 94
xi
-------
TABLES (Concluded)
31 Data Matrix of Cyanide Emissions From Boiler Unit No. 5 ... 98
32 Data Matrix of Cyanide Emissions From Boiler Unit No. 6 ... 99
33 Data Matrix of Phosphate Emissions From Boiler Unit
No. 5 102
34 Data Matrix of Phosphate Emissions From Boiler Unit
No. 6 103
35 Data Matrix of Chloride Emissions From Boiler Unit
No. 5 as Determined by the SOX Sampling Train 106
36 Data Matrix of Chloride Emissions From Boiler Unit
No. 6 as Determined by the SC> Sampling Train 107
A
37 Data Matrix of Chloride Emissions From Boiler Unit
No. 5 as Determined by the Organic Acids Train 108
38 Data Matrix of Chloride Emissions From Boiler Unit
No. 6 as Determined by the Organic Acids Train 109
39 Organic Compounds in Stack Emissions 116
40 Comparison of High and Low Temperature Ashing of
Solid Waste 120
Xll
-------
ACKNOWLEDGMENTS
This report was prepared for the Environmental Protection Agency under
Grant No. R803903-01-0 to the City of Ames, IA, and for the Energy Research and
Development Administration (ERDA) under Contract No. W-7405 ENG-82. During the
conduct of the study, ERDA became the U. S. Department of Energy (DOE). Here-
after in this report, the agency which is now DOE will be referred to as ERDA.
The EPA sponsored portion of the program was directed by Mr. Carlton C.
Wiles of the Municipal Environmental Research Laboratory, Solid and Hazardous
Waste Research Division, Office of Research and Development and Mr. Robert
Olexsey, Industrial Environmental Research Laboratory, Cincinnati, Ohio, Office
of Energy, Minerals, and Industry. The ERDA sponsored portion of this study
was directed by Mr. Robert L. Butenhoff, Physical and Technical Programs,
Division of Biomedical and Environmental Research.
The Engineering Research Institute Of Iowa State University was responsible
for the environmental sampling, and the preparation of this report. The Ames
Laboratory, ERDA was responsible for the major portion of the sample analysis.
Midwest Research Institute, was a consultant to the project and aided in devel-
opment of work plans and methods.
The principal authors of this report were: Dr. Jerry L. Hall, Professor
Alfred W. Joensen, Professor Delmar Van Meter, Mr. Roger Wehage, Mr. Gary
Severns, and Mr. Ronald Reece of the Mechanical Engineering Department, Iowa
State University; Mr. Howard R. Shanks of Ames Laboratory, ERDA; and Mr. Douglas
E. Fiscus and Mr. Robert W. White of Midwest Research Institute.
Many individuals have contributed to the sampling, analysis, data reduc-
tion and report preparation of this study. Their efforts have been greatly
appreciated, and the project could not have been accomplished without their
assistance. A partial token of our appreciation is given by listing below
those individuals who contributed.
The Engineering Research Institute, Iowa State University personnel are as
follows: Dr. Paul Peterson, Director, Engineering Research Institute;
Dr. Jordan Larson, Professor, Mr. William Bathie, Associate Professor,
Dr. Howard Shapiro, Assistant Professor, Mr. John Carroll, Research Assistant,
Mr. Ralph Schilling, Research Assistant, Mr. Larry Scheier, Research Assistant,
Mr. Panayotes Costidis, Research Assistant, Mr. Tom Fries, Research Assistant,
Mr. Don Erickson, Laboratory Assistant, Mr. Tom Hay, Laboratory Assistant,
Mr. Don Young, Laboratory Assistant, Mr. Doug Ryan, Laboratory Assistant,
xiii
-------
Ms. Betsy Morgan, Laboratory Assistant, Mr. Randy Hulsebus, Laboratory Assis-
tant, Mr. Richard Cool, Laboratory Assistant, Mr. David MacAninch, Laboratory
Assistant, Mr. Ed Kibalo, Laboratory Assistant, Mr. Erv Mussman, Laboratory
Assistant, Mr. Mike Lind, Laboratory Assistant.
Personnel of the Ames Laboratory, ERDA are: Dr. Velmer A. Fassel, Deputy
Director, Ames Laboratory, Dr. Richard Kniseley, Senior Chemist, Dr. Robert
Bachman, Associate Chemist, Mr. Robert Hofer, Associate Chemist, Mr. Edward
Dekalb, Associate Chemist, Mr. John Richard, Associate Chemist, Mr. Walter
Sutherland, Assistant Chemist, Mr. Raymond Vick, Assistant Chemist, Mr. Bill
Diedrichs, Assistant Chemist, Ms. Stephanie Syslo, Assistant Chemist, Mr. Gary
Austin, Assistant Chemist, Ms. Vera Peterson, Assistant Chemist, Mr. Bruce
Bear, Assistant Chemist, Mr. Michael Avery, Assistant Chemist, Ms. Mary Wilkes,
Research Assistant, Mr. Mark Parmentier, Research Assistant, Ms. Barbara Royer,
Technician, Ms. Marge Foddy, Research Assistant, Ms. Barbara Brown, Research
Assistant, and Mr. Jim Junko, Research Assistant.
In addition, the following personnel at the City of Ames, Iowa, Municipal
Electric Utility have aided significantly in accomplishment of the testing
program: Mr. Arnold Chantland, Director, Public Works, Mr. J. Keith Sedore,
Director, Electric Utility, Mr. Merlin Hove, Assistant Director, Electric
Utility, Mr. Don Riggs, Power Plant Supervisor, and Mr. Carl Baker, Power Plant
Supervisor.
Also, the following personnel of Midwest Research Institute have assisted
in the development and execution of the testing program: Mr. Paul Gorman,
Principal Chemical Engineer, Mr. Paul Constant, Head, Environmental Measure-
ments Section, Mr. Emile Baladi, Senior Environmental Engineer, and Dr. Mark
Marcus, Senior Chemist.
xiv
-------
SECTION 1
INTRODUCTION
The Ames Solid Waste Recovery System is a continuously operating system
that is processing municipal solid waste (MSW) for use as a supplemental fuel
in the existing steam generators of the Ames Municipal Power Plant. The pur-
pose of this report is to present results of the investigations of the envi-
ronmental effects of using solid waste as a supplemental fuel. This evalua-
tion is a major research program funded by the Environmental Protection Agency
(EPA), the Energy Research and Development Administration (ERDA, now DOE) with
earlier additional participation by the American Public Power Association (APPA)
This project is being performed jointly by the City of Ames, Iowa, Engineering
Research Institute and Mechanical Engineering Department of Iowa State Univer-
sity, Ames Laboratory/ERDA, and Midwest Research Institute (MRI).
This report presents the results and conclusions of the first-year environ-
mental emissions investigation on the two stoker-fired steam generator units
when using coal and refuse derived fuel (RDF).
Following sections of this report present a statement of the objectives,
a summary of results, a description of the boilers, a description of the ex-
perimental design and the sampling methods used, and a presentation of the
results. The results of the environmental investigations were analyzed sta-
tistically. This statistical study is presented in the Appendix.
-------
SECTION 2
OBJECTIVES
The purpose of this study was to characterize the environmental emissions
from the stoker fired boilers when a mixture of coal plus RDF was used as
fuel compared to coal only. Studies of the emissions were accomplished by
measuring the inputs and outputs from the boilers. A brief summary of the en-
vironmental investigations is given below.
1. Sample and characterize the fuel used in the steam generators
• Coal
• Processed solid waste
2. Sample and characterize the effluents from the steam generating units
a. Solids
• Study of the particulates in the flue gas for quantity, size
distribution, and composition. Testing conducted both be-
fore and after the cyclone dust collector.
• Analysis of bottom ash.
• Analysis for possible carcinogens in particulates.
b. Gases
• Analysis for SOX, NOX, chlorides, organic acids, aldehydes
and ketones, cyanide, phosphate, mercury, arsenic, selenium,
beryllium, and HC.
• Identification and analysis of heavy organics in the stack
gases.
-------
SECTION 3
SUMMARY OF RESULTS
In coal plus RDF firing applications, there are a variety of air pollution
control devices that could be used, such as multiple cyclones, scrubbers and
electrostatic precipitators, and therefore, the change in emissions due to
burning RDF is of great interest to the designer of future refuse burning in-
stallations. The following discussion presents a summary of the emissions
due to burning coal only and to burning coal plus RDF.
Both uncontrolled particulate emissions before the particulate collector
and stack particulate emissions to the atmosphere did not have clear overall
trends as a function of RDF heat input. Particulate emissions either in-
creased or stayed the same with percent RDF, depending on boiler unit and
boiler load. Particulate collector efficiency initially increased with in-
creasing percent RDF, then decreased with additional RDF input. The particu-
late collectors were cyclone separators (multiclones) whose efficiency is a
function of such factors as particulate concentration, gas flow rate, and par-
ticle size and density. No particulate sizing was done before the particulate
collector and the reasons for the changes in collector efficiency are not com-
pletely understood at this time. Future tests are planned to include particu-
late sizing before as well as after the collector, which will allow a better
understanding of the changes in collector efficiency.
NOx and sulfur emissions both have trends of decreasing emissions with in-
creased percent RDF. During these tests, boiler No. 5 used Iowa coal only.
Boiler No. 6 used a mixture of one-half Iowa and one-half Wyoming coal.
Wyoming coal is lower in sulfur content than Iowa coal, and thus, sulfur emis-
sions for boiler No. 6 are lower, and the effect of RDF is not as pronounced.
Chloride emissions increased with increasing percent RDF for all boiler
loads. Chloride emissions were substantially lower for coal only than for
coal plus RDF, and therefore, these emissions appear to be a function of the
chlorine in the RDF.
-------
Formaldehyde, cyanide, and phosphate emissions were quite variable, with
no clear trends of emissions as a function of percent RDF. Emissions at 20%
RDF were either lower, or only slightly higher than the coal only emissions.
The major exception being cyanide emissions from boiler No. 6 which showed a
relatively sharper increase in emissions at 20% RDF than for the other test
conditions. However, at 50% RDF, the increases and decreases from the 20%
RDF test condition were variable enough to make it difficult to establish
a trend based on percent RDF.
No significant hydrocarbon emissions in the GI to C$ range have been found
to date. Results from the organic acids analysis and the mercury train are
not yet available. Many of the heavy organic compounds in the stack emissions
analyzed were below the laboratory detection level and the majority of the
organics found were in the stack gases and not in particulate form. These
data are the results of only two stack samples performed to assess the poten-
tial presence of such compounds. Therefore, no comparisons of emissions as a
function of percent RDF can be made at this time.
RDF in combination with coal was successfully fired in the stoker boilers
with some difficulty but with no major problems. The maximum RDF firing rate
was 50% of the heat input to the boiler supplied by RDF. There was no signif-
icant direct effect of burning RDF on the measured boiler thermal efficiency.
There was no significant difference in the percent of the heat input leaving
as combustibles in the ash, the average being approximately 570 for both coal
and RDF. The RDF pneumatic feeders and the additional overfire air required
to burn RDF increased the secondary air (excess air) supplied. The increase
in excess air required to burn RDF reduced the boiler thermal efficiency.
There was general consensus among the boiler operators that more combustion
air through the grate is necessary when firing RDF to prevent slagging and
to maintain a proper fire bed.
Ultimate fouling of the superheater section of boiler No. 5 was experi-
enced. Calculation of the fuel fouling index correlates with this behavior.
The most significant influence is the higher sodium content of RDF which has
a detrimental effect on the fouling index. Soot blowers will be installed
to reduce this fouling behavior. In addition, an alternate method of RDF in-
jection might reduce this effect.
At most boiler loads, bottom ash tended to increase somewhat and fly ash
tended to decrease with increasing percent RDF.
Ash fusion temperatures of RDF are typically 60 to 100°C lower than for
coal. However, no specific correlation of boiler performance to ash fusion
temperatures has been determined.
-------
SECTION 4
DESCRIPTION OF THE STEAM GENERATING UNITS
At the City of Ames Municipal Power Plant the two existing spreader return
traveling grate stoker fired boilers (Nos. 5 and 6) have been modified to burn
RDF as a supplemental fuel with coal. The RDF is prepared in a nominal 136 Mg/
day refuse processing plant adjacent to the power plant. The processing plant
incorporates two stages of shredding, ferrous and nonferrous metal recovery,
and an air density separator. RDF is pneumatically conveyed from the process-
ing plant to a 454-Mg storage bin. RDF is conveyed to the boilers via a pneu-
matic transport system.
The characteristics of the stoker boilers are summarized in Table 1.
These boilers were installed in the 1950s, and use cyclone collectors (multi-
clones) for particulate removal from the exhaust gas to the atmosphere. Both
are traveling grate spreader stokers. Boiler No. 5 was built by Riley and
boiler No. 6 was built by Union Iron Works. Figures 1, 2, and 3 are cross-
sectional views of the boilers. In both boilers coal is discharged into the
boiler by a coal distributor. RDF is fed into the boiler by a pneumatic con-
veying system (not shown). The RDF entry point is adjacent to the coal dis-
tributor.
-------
TABLE 1. CHARACTERISTICS OF AMES MUNICIPAL POWER
PLANT STEAM GENERATORS
Unit
Manufacturer
Installation Date
Pre s sure/Temperature
kPa/°C
(PSI/°F)
Nominal Steam Output Capacity
kg/hr
(Ib/hr)
Coal Firing Equipment
Furnace Pressure
Dust Collection Equipment
Stack Height, m (ft)
Heat Input at Nominal Capacity
MJ/hr
(BTU x 106/hr)
Riley
1951
4,895/441
(710/825)
43,091
(95,000)
Spreader
Stoker
Traveling
Grate
Balanced
Draft
Western
Multiple
Cyclone
61 (200)
154
(146)
Union Iron Works
1958
4,998/441
(725/825)
56,699
(125,000)
Spreader
Stoker
Traveling
Grate
Balanced
Draft
American
Multiple
Cyclone
61 (200)
202
(191)
-------
(O Spreader Stoker O)
Western
Dust
Collector
Figure 1. Gross sectional view of boiler unit No. 5.
-------
00
Q-IU,
X Sub Bosem't Elov. 25
Figure 2. Elevator view of boiler unit No. 6 and associated power plant apparatus.
-------
Steam
Tubes .
o
v
Boiler
Coo\ Trajectory
I / I I
' ' '
/ I /
/ / /
/ j
ill Moving Grate /
_
Siftings
Coal
Distributor
1 'V ' ' t' ' 'f ' ' t T' ' ' V' ' ' f
? ??? r
Drive
Sprocket
Figure 3. Cross sectional view of boiler unit No. 6.
-------
SECTION 5
EXPERIMENTAL DESIGN
In this study, it was determined that two major factors could be con-
trolled at various levels. These factors were the load based on steam flow
and the amount of RDF based on heat energy input to the boiler. The levels
of these factors were chosen to be 60, 80, and 100% nominal load and 0, 20
and 50% RDF. To obtain sufficient data for statistical analysis, a factorial
experimental design with three replications was devised for each boiler as
summarized in Table 2. Thus, for boiler unit No. 5 the statistical design
is a 3 x 3 x 3 full factorial experiment with 27 runs needed to fill the data
matrix of this experiment. In addition, testing of two different size (and
design) traveling grate stoker fired boilers (unit Nos. 5 and 6) was accom-
plished at one load setting (80%) to obtain a relative size comparison for
all emission data at a given fixed load. The tests accomplished to date are
shown in the data matrix format of Table 3.
To satisfy the objectives of the environmental emissions study, all ap-
propriate input and output streams associated with the operation of steam
generating unit Nos. 5 and 6 were sampled. A block diagram showing the sam-
ple locations in both entering and leaving streams is given in Figure 4.
Sampling, physical characterization, and chemical element analysis were
performed on input fuels, ash, and particulate effluent. Size distributions
of the particulate effluent were also determined. In addition, gaseous ef-
fluents were sampled and characterized. These gaseous species include oxides
of nitrogen (NOX), oxides of sulfur (SOx), excess air or oxygen, carbon di-
oxide, aldehydes and ketones, chlorides, organic acids, mercury, and several
others. Each of these items was sampled on a regular basis according to the
test summary of environmental sampling shown in Table 3. Heavy organic spe-
cies, unburned hydrocarbons, and nitrous oxide were sampled on an intermit-
tent basis as man power, instrumentation, and equipment would allow.
10
-------
TABLE 2. SUMMARY OF FACTORIAL EXPERIMENTAL DESIGN
Stoker Boiler No. 5
Coal Used: Iowa
N\% Load
% RDF^v
20
50
60
4A,4B
20
21 S/
36
8
9A.9B
33
1
34
35
80
5
16
17
6
12
13
2
10
15
100
11*
32S/
7
14
19
3
18
c/
Stoker Boiler No. 6
Coal Used: Mixture of
50% Iowa, 50% Wyoming
\% Load
O/ D P^ C ^\
/O i\ ^J | ^v
0
20
50
80^
24
29
30
25
26
27
22
23
28
Q/ Test 21 conducted while pulling ash from boiler to determine any change in
performance and/or emissions due to ash removal.
b/ Bottom ash not weighed because of ash removal difficulties (slagging in boiler
and clinkering of ash).
c/ Boiler No. 5 cannot operate at 100% steam load and 50% RDF without severe
ash problems due to lack of excess air. Therefore, the third test in series
was not conducted.
d/ Load was changed from the originally planned 100% to 80% steam load to be
more typical of capability of boiler and air supply for refuse burning. This
change was essential from experience gained during testing of Boiler No. 5.
11
-------
TABLE 3. TEST SUMMARY OF ENVIRONMENTAL SAMPLING
(Check Marks Indicate That an Appro-
priate Sample Has Been Obtained)
Date
6- B-76
6-10-76
6-15-76
6-17-76
6-21-76
6-23-76
6-25-76
6-28-76
6-30-76
7- 2-76
7- 6-76
7- 8-76
7- 8-76
7-16-76
7-17-70
7-19-76
7-19-76
7-23-76
7-2V-76
8- 2-76
8- 2-76
7-26-76
Unit
05
#5
#5
#5fl-Bd/
l?5
#5
#5
#5
#5B-Bd/
#5
#5
#5B-Bd/
#5B-Bd/
05
#5
rj5B-H
/
/
/
/
/
/
/
/
/
/
/
/
-
Aldehydes
Ketanes
BPC
/
/
//
TEST
TEST
)
/
/
/
/
/
//
/
/
/
/
/
/
•
/
/
BORT
/
/
JORT1
Chlorides
Org. Acids
BPC
/
/
//
D - BO
D -• RA
i
/
/
/
/
/
//
/
J
/
/
/
/
/
/
•
ass
/
/
01
Hg.Be.Se
BPC
/
/
/
OULD
/
//
/
/
/
/
//
/
J
/
/
/
/
/
/
/
IOT
/
/
—
Heavy
Organics
BPC
31D L
/
/
)AD
HC S t%0
BPC
s
/
/
/
/
/
/
—
-------
TABLE 3. (Concluded)
Date
8- 5-76
8- 6-76
8- 9-76
8-10-76
8-11-76
8-12-76
8-13-76
8-16-76
8-18-76
8-24-76
8-24-76
8-25-76
8-26-76
8-26-76
8-27-76
Unit
lib
06
06
06
06
It
06
06
II 6
#5B-Bd/
#5B-Bd/
05
#5B-Bd/
#5B-Bd/
05
ILoad
80Z
80Z
80Z
80Z
80Z
80Z
80Z
80Z
80Z
100Z
100Z
60Z
60Z
60Z
60Z
Fuel.3/
C+RDF^
C+RDF
C
C+RDF
C+RDF
C+RDF
C+RDF
C
C
c
c
C+RDF
C+RDF
C+RDF
C
ZRefuse
50Z
50Z
OZ
20Z
20Z
20Z
50Z
OZ
OZ
OZ
OZ
20Z
50Z
50Z
OZ
Test
Designation
EPA 22
EPA 23
EPA 24
EPA 25
EPA 26
EPA 27
EPA 28
EPA 29
EPA 30
EPA 31
EPA 32
EPA 33
EPA 34
EPA 35*
EPA 36
Bottom
Ashk/
/
/
/
/
/
/
/
/
/
,sJ
J&l
J
J
J
J
Collector
Ash
/ wt
/ wt
/ wt
/ wt
/ wt
/ wt
/ wt
/ wt
/ wt
/ wt
J wt
/ wt
/ wt
/ wt
/ wt
ParticularpC/
BPC
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Sizing
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Orsat
BPC
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
NO
X
BPC
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
SO
X
BPC
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Aldehydes
Ketones
BPC
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Chlorides
Org. Acids
BPC
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Hg.Be.Se
BPC
S
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Heavy
Organlcs
BPC
S
HC i N^O
BPC
S
/
/
/
/
/
/
/
/
/
a/ Coal for tests on Boiler No. 5 is Iowa coal; on No. 6, 507. Wyoming and 507. Iowa coal.
b/ Bottom Ash and Collector Ash weighed together after completion of test and removal of ash to ash silo except where special weights are noted.
c/ BPC - Before Participate Collector; S = Stack; Sizing = Sizing on stack.
<]/ B-B indicates back-to-back testing.
e/ Boiler load dropped and test terminated early.
tl Test conducted while pulling ash to determine if boiler performance and emissions change when ash is pulled.
&/ Bottom ash not weighed due to ash removal difficulties.
-------
FLOW RATE
ULTIMATE ANALYSIS
HEATING VALUE
CHEMICAL ANALYSIS & TRACE ELEMENTS
ASH SOFTENING TEMPERATURE
FILTER PARTICULATE TRACE ELEMENTS
IMPINGER WATER TRACE ELEMENTS
EMISSION RATES OF PARTICULATE
PARTICULATE TRACE ELEMENTSI
IMPINGER WATER TRACE ELEMENTS
EMISSION RATES OF PARTICULATE
AND GASEOUS SPECIES
PARTICULATE SIZING
HUMIDITY
BAROMETER
INTAKE
TEMPERATURE
VOLUME FLOW
DENSITY
ULTIMATE ANALYSIS
HEATING VALUE
CHEMICAL ANALYSIS &
TRACE ELEMENTS
ASH SOFTENING
TEMPERATURE
COAL
•AIR
RDF
TEMPERATURE
FLOW RATE
1
FLOW RATE
CHEMICAL ANALYSIS &
TRACE ELEMENTS
SOFTENING TEMPERATURE
FLOWRATE
TEMPERATURE
PRESSURE
FLOW RATE
CHEMICAL ANALYSIS &
TRACE ELEMENTS
SOFTENING TEMPERATURE
Figure 4. Boiler unit No. 5 and unit No. 6 sampling locations.
-------
SECTION 6
SAMPLING METHODS
GENERAL
During a given test, samples were obtained at all points shown in Figure
4 on a given unit. The input fuel and boiler load were held as constant as
possible at their preselected nominal values during a given test. Each test
was typically 4 to 5 hr in length if no difficulties were encountered. On
some occasions, the test lasted for more than 8 hr. On other occasions,
back-to-back tests were accomplished during a given day. The back-to-back
tests were also indicated in Table 2.
The input fuel and grate ash were sampled at regular 1-hr intervals
throughout the testing period on any given day. These samples were then mixed
to yield a composite sample for the given test.
The stack effluents were generally sampled according to EPA prescribed
techniques. The preparation of each sampling train included: (a) cleaning
the sample train glassware with an appropriate acid wash followed by several
distilled water rinses, (b) preparation of chemicals and loading each sample
train impinger with the appropriate absorbing solution, (c) weighing and
labeling each impinger for each sample train, (d) weighing and loading parti-
culate train filters, (e) checking sample box, control box, and sampling probe
for proper operation. Once these checks were accomplished, the sampling train
was transported to the test site and prepared for operation as desired.
Leak checks were performed on each sample train from the sampling probe
tip both before and after each sample run. No sampling train was operated
until it had passed an appropriate leak check from the nozzle tip in the field
location. The maximum allowable leak rate was 0.6 liters/min at a sampling
train vacuum of 50.8 kPa. A leak rate of 0.6 liters/min was also allowed at
a lower vacuum as long as this vacuum level was not exceeded during the test.
All sample train leak checks during this study complied with this leak rate
criterion. After each experimental run, a leak check was also performed to
insure that no leak had developed during the sampling period. The samples
were obtained either isokinetically or proportionally as required but only
after the calibrations for the dry gas volume flow meter, flow orifice, Pitot
tube, sampling nozzle, and temperature sensors on the sampling train had been
previously completed.
15
-------
After sampling, the sampling train impingers and filters were weighed
and/or prepared for analysis in an appropriate manner. The impinger chemi-
cals were transferred to clean reagent bottles, the filters were desiccated
to dryness before weighing, and the coal, RDF, and ash samples were mixed
and ground. All of the samples were then transferred to the various analyti-
cal groups of the Ames Laboratory-ERDA for analysis.
The appropriate details of sampling and analysis for the experimental
runs of this study are described in the following paragraphs.
INPUT STREAMS
Coal
Coal samples were collected at 1-hr intervals during the experimental
runs. These samples were then combined and mixed to form a single test run
sample composite for analysis. The samples were prepared for analysis by
grinding in a hammer mill. Heating value, ultimate analysis, chemical analy-
sis for elements, and ash softening temperatures were then determined by the
techniques indicated in Table 4.
Refuse Derived Fuel (RDF)
Solid waste, RDF, samples were collected at 1-hr intervals during the
experimental runs. These samples were then combined and mixed to form a
single test run sample composite for analysis. The RDF samples were weighed,
heated to 110°C for sterilization, then reweighed to determine the moisture
content of the original sample. Samples were prepared for analysis by grinding
in a Wiley Mill. Heating value, ultimate analysis, chemical analysis for
elements, and ash softening temperatures were then determined by techniques
indicated in Table 4.
Air
The air stream entering the boiler unit for purposes of combustion was
characterized by measuring the wet bulb temperature, the dry bulb tempera-
ture, and the barometric pressure. The wet and dry bulb temperature read-
ings were used to determine the relative humidity by the standard psychometric
method.
16
-------
TABLE 4. ELEMENTS DETERMINED AND ANALYTICAL METHOD USED FOR ANALYSIS
OF COAL, REFUSE DERIVED FUEL, AND ASH SAMPLES
Element
X-ray fluorescence"
b/
ASTM Method
Aluminum
Arsenic
Calcium
Carbon
Chromium
Copper
Gallium
Germanium
Hydrogen
Iron
Lead .
a/
Magnesium"
Manganese
Nickel ,
a/
Phosphorus"
a/
Potassium"
Rubidium
Selenium
Silicon
Sodium^'
Strontium
Sulfur
Titanium
Vanadium
Zinc
Ash
Moisture
Higher heating value
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
D3178
D3178
D3177
D3174
D3173
D2015
_a/ Samples were also sent to ACCU-Labs Research, Inc. for analysis
by ASTM Method D2795 for these elements.
_b/ Additional elements were scanned but not detected by this tech-
nique. Detection limits for all elements scanned are given
in Table 7.
17
-------
EXIT STREAMS
Ash
Grate (Bottom) Ash—
Grate ash samples were collected at 1-hr intervals during the experi-
mental runs on boiler unit Nos. 5 or 6. These samples were then combined and
mixed to form a single sample composite for analysis. The grate ash sample
was prepared for analysis by grinding in a Braun plate-type pulverizer. The
resulting powdered samples were analyzed except for heating values, by the
techniques indicated in Table 4.
Hopper (Fly) Ash—
Hopper ash samples were collected from the ash hoppers of the mechanical
particulate collectors associated with each of the two boiler units. These
samples were obtained at the end of each experimental run. Several samples
were taken to obtain adequate representation of the hopper ash collected
during an experimental run. These samples were then combined and mixed to
form a single sample composite for analysis. The hopper ash was prepared for
analysis by grinding in a Braun plate-type pulverizer. The resulting powdered
samples were analyzed, except for heating values, by the techniques indicated
in Table 4.
Steam
The boiler load was determined by measuring the steam flow rate generated
by the boiler unit. The steam flow integrator reading, the steam temperature
and steam pressure were recorded along with feedwater flow rate and feedwater
temperature.
Stack Effluents
The stack effluents from boiler unit Nos. 5 and 6 have been obtained by
methods prescribed by EPA,-!^' Carotti and Kaiser,^' and Shannon.Ji/
Effluents, such as particulate matter, requiring isokinetic sampling
were obtained at 48 prescribed points in the boiler stack cross section as
shown in Figure 5. Particulate samples obtained before the mechanical parti-
culate collectors in unit Nos. 5 or 6 were obtained at 52 and 48 prescribed
points, respectively, in the direct cross section as shown in Figure 6. Both
boiler units exhaust from the individual particulate collectors into the same
smoke stack. Thus, when boiler unit No. 5 was being tested, unit No. 6 was
not operating and vice versa.
18
-------
PORT 2
PORT 4
POINT
A
B
C
D
E
F
RADIUS
CM
32.5
56.4
72.6
86.0
97.5
108.0
in.
12.79
22.20
28.59
33.86
38.37
42.51
POINT
G
H
1
J
K
L
RADIUS
CM
117.2
125.8
134.1
141.7
149 1
155.8
in.
46.15
49.53
52.79
55.80
58.69
61.32
Figure 5. Sampling locations for isokinetic sampling of
boiler unit Nos. 5 and 6 exhaust gas stack.
19
-------
UNIT 5
o o
0 0
0 0
O 0
A |«-B-»|*-B-
O
O
o
o
-H
o
o
o
0
~"r*
o
0
0
o
-B-»j*-B
O
o
o
o
~4*~
0
o
0
o
»-K
o
o
o
o
-B »(« B-
0
0
o
o
-*H
o
o
o
o
»|< E
O
o
o
o
»[< B
0 0
o o
o o
0 0
»[« B— »| A
if'
b' C
b''
a 1
^
DIMENSIONS
A
B
C
CM
20.5
41.0
533.4
in.
8.08
16.125
210.0
a
b
c
CM
12.0
24.1
96.5
in .
4.75
9.50
38.0
UNIT 6
0000
o o o o
0 0 0 O
o o o o
A U-B H« B >H B »H E
O O O O
O O O O
O O O O
O O O O
1 1 1 1
C
0 0 0 O
0 0 O O
O O O 0
O O O 0
»L B »[* B 4« B— H A
a
b
b
b
a '
T
«-
A
B
C
CM
16.5
33.0
396
in.
6.5
13.0
156.0
a
b
c
CM
14.6
29.2
116.8
in.
5.75
11.50
46.0
Figure 6. Sampling locations for isokinetic sampling of boiler
unit Nos. 5 and 6 before the particulate collector.
20
-------
Effluents requiring proportional sampling, such as the gaseous species,
were obtained at a single point in the boiler stack cross section. This point
usually corresponded to the position where the sampling probe was inserted to
its maximum length into the stack. The proportional sampling was accomplished
simultaneously with the particulate sampling during any experimental run.
Normally, three different sampling trains were operating simultaneously in the
stack along with another particulate sampling train located before the parti-
culate collector. The remaining stack porthole was used to obtain grab samples
for the Orsat analyses, nitrogen oxides, unburned hydrocarbons, and other mis-
cellaneous gaseous species.
The sampling train configuration was generally that of EPA Method 5, or
a modification thereof, as described in the Federal Register^"-^/ and shown in
Figure 7. The sampling train consists of a sampling probe, sampling box,
umbilical cord, and control box as shown in Figure la. The sample probe con-
sists of a heated glass liner, a sampling nozzle or probe tip, and an s-type
Pitot tube (Figure 7b). The sampling box consists of an oven heated filter
and cyclone (Figure 7c) and a set of impingers (Figure 7d) placed in an ice
bath and loaded with a variety of chemicals for proper absorption or condensa-
tion of the specific interest. The umbilical cord provides vacuum lines and
electrical cables between the sample box and control box. The gas flow rate
through the train can be metered, and its flow rate adjusted at the control
box. It provides electrical energy to the sample box and probe for heating
purposes, and it provides for measurement of a variety of temperatures in the
sampling system as well as the stack gas velocity. Block diagrams illustrat-
ing the various steps and/or measurements necessary to ascertain the particu-
late emissions, gaseous emissions, and elemental analysis of particulates
collected on the filter in the EPA Method 5 sampling train are shown, respec-
tively, in Figures 8, 9, and 10. The various gaseous samples have been col-
lected using a modified EPA Method 5 sampling train. The modifications include
a glass wool plug in the tip of the sampling probe to prevent particulate
matter from entering, and a section of glass tubing called a "bypass" which
replaces the filter and cyclone assembly in the oven of the sampling train.
A summary of the chemicals in each sample train impinger as well as the
flow rates used in this study are given in Tables 5 and 6.
Particulates--
The particulate effluent has been collected by the EPA Method 5 sampling
train previously described and as outlined in the Federal Register.—>—' The
particulate matter was collected on a quartz fiber filter and in a glass
cyclone preceding the filter. The particulate trapped in the sampling probe
during an experimental run was also added to the cyclone and filter catch to
yield a measurement of the particulate effluent leaving the smoke stack.
21
-------
(e)
EPA Method 5 sampling
train (a), nozzle (b),
oven (c), impinger box
(d), Orsat apparatus (e),
and NOx sampling train
(f).
Figure 7. Photographs of sampling equipment.
22
-------
PARTICULATE EMISSIONS
Sample Train
I
Filter Weights
Probe & Cyclone
Washings
Stack Volume
Flow, m3/Hr
Fuel Flow
Rate, kg/Hr
Fuel Heating
Value, kJ/kg
Flow Rate
Sample Time
Sample Conditions
Ambient Conditions
Corrected Volume
at Standard
Conditions
Concentration
of Particulate,
Ib/ft3 or g/m3
Particulate
Emission,
kg/106 kJ
Calibration
Corrections
Figure 8. Block diagram of particulate emissions data flow.
23
-------
GASEOUS EMISSIONS
Sample Train
Laboratory
Analysis
Quantity
of Specie,
g or g/ml
Stack Volume
Flow, m3/Hr
Fuel Flow
Rate, kg/Hr
Fuel Heating
Value,
Flow Rate
Sample Time
Sample Conditions
Ambient Conditions
Corrected Volume
at Standard
Conditions
Calibration
Corrections
Concentration
of Specie,
Ib/ft3 or g/m3
Gaseous
Emission,
kg/106 kJ
Figure 9. Block diagram of gaseous emissions data flow.
24
-------
TRACE ELEMENTS FROM FILTERS
Sample Train
Filter
Laboratory
Analysis
Quantity
of Element,
,2
cm'
Stack Volume
Flow, m3/Hr
Fuel Flow
Rate, kg/Hr
Fuel Heating
Value, kJ/kg
Filter Area
J.
Flow Rate
Sample Time
Sample Conditions
Ambient Conditions
Corrected Volume
at Standard
Conditions
Concentration
of Specie,
Ib/ft3 or g/rr,3
Trace Element
Emission,
kg/106kJ
Calibration
Corrections
Figure 10. Block diagram of trace element from filters data flow.
25
-------
TABLE 5. IMPINGER SOLUTIONS FOR SAMPLE TRAINS
to
Sample train
Particulate
(EPA method No. 5)
Oxides of sulfur
(EPA method No. 6)
Aldehydes and
Key tone siL'
Chlorides and .
. , b/
organic acids^
Andersen"
Mercury vapor""
"midget"
Solution in each impineer
Impinger No.
1 234 Flow rate
H-OS/ H^O3' Dry Drierite Isokinetic
Proportional
Isopropanol 3% HO 3% HO Drierite at 8 £/min
Proportional
Dry NaHSO NaHSO Drierite at 3 4/min
. Proportional
H Cr 1.5N NaOH 1.5N NaOH Drierite at 20 j£/min
a/
H 0~ Dry Drierite — Isokinetic
Proportional
10% K GO IG1 ICl Drierite at 1 £/min
_a/ Double distilled water.
b/ EPA method No. 5 modified.
-------
TABLE 6. IMPINGER SOLUTIONS FOR MERCURY VAPOR SAMPLE TRAINS
Solution in each impinger
EPA
test No.
1
2
3^
3^7
4A
4B
5
6
7
ftfl/
$a/
9A
9B
10
11
12
13
14
15
16&
36
Impineer No.
1
0.5N HN03
0.5N HN03
KMnO
0.5N HN03
ICl
ICl
ICl
ICl
ICl
K2C03
ICl
K2C03
K2C03
K2C03
K2C03
K2C03
K2C03
K2C03
K2C03
Dry
Dry
2
0.5N KOH
0.5N KOH
0.5N KOH
0.5N KOH
ICl
ICl
ICl
ICl
ICl
K2C03
ICl
ICl
ICl
ICl
ICl
ICl
ICl
ICl
ICl
HN03
HN03
3
Dry
Dry
Dry
Dry
ICl
ICl
ICl
ICl
ICl
ICl
ICl
ICl
K2C03
ICl
ICl
ICl
ICl
ICl
ICl
KOH
KOH
4
Drierite
Drierite
Drierite
Drierite
Drierite
Drierite
Drierite
Drierite
ICl
Drierite
ICl
ICl
ICl
ICl
ICl
ICl
ICl
ICl
Drierite
Dr i e r i t e
567
Ascarite
Ascarite
Ascarite
Ascarite
Ascarite Ascarite
ICl Drierite
Ascarite
Drierite
Drierite/ Ascarite
Ascarite
Drierite Drierite/
Ascarite
Drierite Drierite
Drierite Drierite
Drierite Drierite
Drierite Drierite
Drierite Drierite
_a/ Two sampling trains were used for Test Nos. 3 and 8.
b/ Same solution was in all impingers for Test Nos. 16 through 36.
Flow Rate: Proportional at 1
-------
The sample train flow rate was adjusted to a value that would allow iso-
kinetic sampling at each of the 48 points in the stack cross section. Sampling
was generally performed for a time period of 3 min at each of the 48 points
although some experimental runs were accomplished at 5 min/point. A pretest
velocity measurement (presurvey) was usually performed before any test run
so that isokinetic sampling could be adjusted on the sample train.
In addition to the determination of the mass of particulate effluent, the
particulates on the filters were analyzed for elemental composition by the
x-ray fluorescence technique. The list of elements scanned by this technique
is included in Table 7.
The loaded filters from the EPA Method 5 sampling train were found to con-
tain large amounts of sulfuric acid (H2S04). In order to analyze trace element
concentration on the filter, it was necessary to first fix the H2S04 as
(^4)2304 by pretreating the filter with Nt^OH fumes. Elemental composition
was then determined by x-ray fluorescence.
Trace elements in the particulate effluent sampling train impingers were
determined by an Inductively Coupled Plasma Analytical System (ICP).^' Trace
element analysis of impinger solutions from other sampling trains was also
accomplished by this technique. The list of elements scanned by this tech-
nique is included in Table 8.
Particulate Sizing--
Sizing of stack particulate emissions for boiler unit Nos. 5 and 6 was per-
formed in situ using an Andersen 2000 Inc. Mark III Stack Sampling Head. The
Andersen sizer is a cascade impactor that uses inertial separation to separate
particles by aerodynamic size. The Andersen sampler could not be inserted in
the stack because the stack sampling ports were too small. Instead, the
Andersen sampler was used in conjunction with an EPA Method 5 sampling train.
The probe was heated to 121 C and the sampler was secured in the oven of the
sampling box in place of the Method 5 filter and cyclone assembly. The oven
temperature was maintained at 121 C — 14 C. Specifically, a Misco sampling
box and control unit was used. Figure 11 shows the placement of the sampler
in the oven. Stack port 3, the southwest port, was used for all sizing tests.
Sampling of particles necessitates isokinetic sampling, therefore, the
use of an EPA Method 5 probe was felt to be the best way to achieve this.
Measurements of stack temperature and dynamic pressure as well as dry gas
meter readings and meter temperatures were taken throughout the test duration
and adjustments in flow made to maintain an isokinetic flow rate. Since all
sampling was performed at one location for a total time of not more than 20
min, very little variation in flow rate resulted. Short sampling times were
used because of the high particulate loading of the sampler stages.
28
-------
TABLE 7. DETECTION LIMITS OF ELEMENTS SCANNED BY X-RAY FLUORESCENCE (XRF)
Elements
Aluminum
Arsenic
Beryllium
Calcium
Chlorine
Chromium
Cobalt
Copper
Gallium
Germanium
Iron
Lead
Manganese
Nickel
Phosphorus
Potassium
Rubidium
Scandium
Selenium
Silicon
Strontium
Sulfur
Titanium
Vanadium
Yttrium
Zinc
Zirconium
Ash
samples
WPPM!/
10,000
10
5
800
3,000
100
30
20
10
10
50
25
70
20
8000
1,000
5
600
5
8,000
5
5,000
500
200
5
20
5
Coal
samples
¥PPM
5,000
2
2
400
1,000
20
20
10
2
2
20
10
20
10
4,000
500
1
300
2
4,000
1
3,000
200
50
1
5
1
Stack
filters
^g/cm2
3003/
0.2
0.2
40
100
0.5
0.4-k/
0.2
0.4-0./
0.4
0.4
1.0
0.5
0.3
250
50
0.1
20
0.2-S/
250
0.1
200
5^
14/
0.1
0.2
0.1
Refuse
derived
fuel
WPPM
61,000
61
30.5
4,880
18,300
610
183
122
61
61
305
152
427
122
48,800
6,100
30.5
3,660
30.5
48,800
30.5
30,500
3,050
1,220
30.5
122
30.5
_a/ Lead interferes at any concentration
Jb/ Iron interferes at
_c/ Lead interferes at
_d/ Iron interferes at
moderate concentration.
high concentration.
high concentration.
e/ Titanium interferes at moderate concentration.
_f/ Parts per million
on a weight
basis.
29
-------
TABLE 8. DETECTION LIMITS OF ELEMENTS SCANNED BY
INDUCTIVELY COUPLED PLASMA
Element
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Cadmium
Calcium
Chromium
Cobalt
Copper
Gallium
Germanium
Iron
Detection
limit
M-g/ml
0.009
0.03
0.04
0.001
0.0002
0.02
0.002
0.001
0.008
0.002
0.002
0.0009
0.02
0.1
0.005
Element
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Phosphorus
Selenium
Silicon
Strontium
Thallium
Tin
Titanium
Vanadium
Yttrium
Detection
limit
|^g/ml
0.02
0.004
0.0002
0.03
0.03
0.009
0.5
0.02
0.07
0.008
0.06
0.08
0.006
0.002
0.007
Detection
limit
Element Jig/ml
Zinc 0.0006
30
-------
ff
Oven Wall
Probe
Insulation
Cyclone
Preseparator
L
Andersen
Sizer
i J
To Impinger Train
Figure 11. Placement of Andersen Sampling Head in an EPA Method 5 sampling train.
31
-------
The probable presence of particles larger than 15 u suggested the
necessity of a cyclone preseparator. An Andersen preseparator was used. The
manufacturer supplied a calibration curve for the preseparator used.
Each test was started with a clean probe and glassware. At the conclu-
sion of each test the probe, nozzle, and connecting glassware were washed out
with acetone and the rinsings were collected in a weighed beaker for weighing
after evaporation of the acetone. During back-to-back testing, the washing
was done in the field.
Each stage of the Mark III sampler with the exception of the back-up
filter stage consists of a plate, a gasket, a filter, and a filter hold-down
cross. Each stage was assembled on a small piece of aluminum foil which was
marked for later use. This entire assembly, including the foil, was weighed
on a precision balance to the nearest tenth milligram and the weight recorded.
The foil was stored in a plastic bag for later use.
After completion of sampling and return to the laboratory, the probe,
cylone, and glassware were rinsed of any particulate that may have been col-
lected. The rinsings were placed under a chemical fume hood to evaporate the
acetone used for rinsing.
The Andersen sampler was disassembled in the laboratory and each stage
assembly then placed on the foil originally used for weighing. Each assembly
was weighed and the weight was recorded. After evaporation of the acetone, the
probe and cyclone washing beakers were weighed. Net weights were computed for
the probe, cyclone, and collection stages. All particulates collected in the
nozzle, probe, and glassware before the cyclone were considered probe collection.
Everything collected from the entrance of the cyclone to the first stage of the
sampler was considered the cyclone collection.
When two tests were conducted consecutively on the same day, the Andersen
sampler was removed from the oven and returned to the laboratory for weighing
of the loaded stages, cleaning, and preparation for the second test.
Orsat--
Orsat analyzers were used for the determination of oxygen, carbon dioxide,
and carbon monoxide in the stack gas. A photograph of an Orsat analyzer is
included in Figure 7e. The sample was obtained by hand pumping an appropriate
volume of stack gas into the analyzer. The sample was then successively bub-
bled through chemicals in each of the Orsat pipettes to absorb C02, 02, and CO
in the stack gas. Nitrogen was determined by difference from the values of
C02, 02, and CO. These measurements determine the effective molecular weight
of the stack gas as well as allow an estimate of the excess air in the effluent
stream to be determined.
32
-------
Oxides of Nitrogen (NOX)--
The determination of nitrogen oxide (NOX) emission was basically measured
by EPA Method 7. This method consists of collecting a sample of stack gas in
an evacuated flask containing a dilute sulfuric acid-hydrogen peroxide absorb-
ing solution. A photograph of a NOX sampling train is included in Figure 7f.
The nitrogen oxides except for nitrous oxide (^0), are measured colorimetri-
cally using a phenoldisulfonic acid (PDS) procedure.—'
Preliminary values of NOX emission using the EPA recommended absorbing
solution were found to be suspect. These values always appeared low relative
to samples obtained simultaneously using higher percentages of absorbing solu-
tion. The difficulty appeared to be caused by the sulfur content of the stack
gas and its reaction with the absorbing solution. Experiments were performed
to determine if the percentage of hydrogen peroxide solution was statistically
significant. At the 95% confidence level it was ascertained that a significant
difference did exist. Further sampling and analysis yielded no significant
differences at a 957» confidence level for absorbing solution from 1 to 970
hydrogen peroxide. Consequently, the ASTMZ/3% hydrogen peroxide absorbing
solution was selected as the appropriate value for the NOX measurements of
the experimental runs of this study.
Oxides of Sulfur (SOX)--
The determination of sulfur oxides (SOX) emissions was determined by EPA
Method 6 and reported as sulfur dioxide. This method consists of collecting
the sample of stack gas by an EPA Method 5 sampling train modified with a
glass wool particulate filter in the probe tip and a bypass section replacing
the filter in the sample train oven. The sample train flow rate was adjusted
to approximately 8 liters/min (0.3 cfm) and operated in the proportional sample
mode at a single point in the exhaust stack cross section. The sample was
collected over a period of approximately 30 min although it was eventually
determined that the amount collected during this time span was more than
sufficient for a reliable sample to be obtained. The sulfur dioxide, once
collected, was measured by a barium-thorin titration method.—
Aldehydes and Ketones--
The determination of aldehyde and ketone emissions was accomplished by
the method of Carotti and Kaiser.—' This method consists of collecting the
sample of stack gas by an EPA Method 5 sampling train modified with a glass
wool particulate filter in the probe tip and a bypass section replacing the
filter in the sample train oven. The sample train flow rate was adjusted
to approximately 3 liters/min (0.1 cfm) and operated in the proportional
sampling mode at a single point in the exhaust stack cross section. The
sample was usually collected during a minimum sampling period of 30 min,
33
-------
although some collection times were as long as 70 min. The sample was
absorbed in 1% aqueous sodium bisulfite (NaHS03) contained in the sample
train impingers and quantitative determination was made as formaldehyde.
Organic Acids, Cyanide, and Phosphate--
The determination of organic acids, cyanide (CN), and phosphate (PO^-3)
emissions was accomplished by the method of Garotti and Kaiser.^' This method
consists of collecting the sample of stack gas by an EPA Method 5 sampling
train modified with a glass wool particulate filter in the probe tip and a
bypass section replacing the filter in the sample train oven. The sample
train flow rate was adjusted to approximately 20 liters/min (0.7 cfm) and
operated in the proportional sampling mode at a single point in the exhaust
stack cross section. The sample was usually collected during a minimum samp-
ling period of 30 min although some collection times were as long as 75 min.
The sample was absorbed in a 1.5N solution of sodium hydroxide (NaOH) con-
tained in the sample train impingers.
CN~ was determined by ion selective electrode after adding ClCO^ to the
impinger solutions to remove sulfide interference. Phosphorus was determined
as PO^ by spectrophotometer after acidifying the impinger solution with I^SO^
and adding ammonium persulfate to release all organic phosphorus. Chloride
was also determined from this train by a colorimetric method.
A number of methods were used to obtain results for organic acids. An
ion chromatograph was found most satisfactory.
Chlorides--
The chloride emissions have been determined from the samples collected
in two of the previously described sample train. Part of the impinger solu-
tion from the EPA Method 6 sulfur dioxide sampling train was analyzed for
hydrochloric acid (HCl). The chlorides from this sample train were determined
by a colorimetric method.
The organic acid sampling train was also used to determine chlorides,
as reported by Carotti and Kaiser,.^' from an impinger in the sampling train
which contained a 1.5N NaOH solution. The chlorides from this sample train
were determined spectrophotometrically.
34
-------
Mercury, Arsenic, Intimony, and Beryllium--
Mercury vapor in the stack effluent was generally collected by the
Shannon— modification of the EPA Method 5 sampling train. The first impinger
contained 0.5N HNC>3 and the second impinger contained 0.5N KOH. The solutions
after sampling were analyzed by inductively coupled plasma for the presence of
mercury, arsenic, antimony, and beryllium. A second mercury train with K2C03
and IGl solutions was also used.—' These solutions were also analyzed, after
sampling, with the inductively coupled plasma unit for the above particular
elements. In addition, a number of other elements were also quantified from
these trains.
Hydrocarbons--
Total hydrocarbons in the C^ to €5 range were determined by a gas chromat-
ograph. The method consists of collecting a sample of stack gas in a dry
evacuated flask. The sample is then passed through a packing material in the
column of the gas chromatograph. The packing material such as carboseive B
successively retains the C^ through Cr species such that the chromatograph
output signal indicates their presence.
Organics--
Stack gases were drawn through a heated probe and across a column of
macroreticular resin (Tenex-GC) where the organics are collected. The organics
including polychlorinated biphenols (PCB) and polynuclear aromatics (PNA, PAH,
PCH, or POM) were determined quantitatively by a gas chromatograph, mass spec-
trometer system.
The sampling methods and classification schemes described in the litera-
ture^ as well as those developed here and described above have been shown to
be inadequate. However, some selected methods are clearly superior based on
interpretation of bench tests and these better methods have been used to study
stack gases and particulates. The progress to date with these samples is
discussed here.
Gas phase samples--S sampling procedure was developed for the accumulation
of vapor phase polynuclear aromatics (PA, PAH, PCH, or POM) on porous polymers
and the subsequent removal from the polymer followed by preseparation with final
separation and measurement by gas chromatography. Acenaphthylene, fluorene,
anthracene, phenanthrene, and perylene present in 5-to 295-liter air samples
at prepared concentrations of from 10~5 to 10~" g/liter were tested. Recovery
efficiencies of 90 to 100% were demonstrated. For other organic components
recovery efficiencies of 40 to 100% were related to component volatility and
air volume. These results on various designs of probes and various porous
polymers (XAD-2, XAD-4, XE-340, and Tenax-GC) were used to design and fabricate
35
-------
a probe which was deemed suitable for both ambient air and hot stack gas
sampling.
Particulate samples—A limited effort was expended in devising new schemes
for the collection of particulate samples. However, the treatment of particu-
lates was extensively tested for effectiveness in removal of absorbed organic
matter. Four schemes—thermal desorption, column elution, Soxhlet extraction,
and extraction by sonic vibration were tested. The latter two showed the most
promise for isolating organic matter in a form most suitable for separation
and measurement. Seven extraction solvents, diethylether, chloroform, hexane,
cyclohexane, benzene, methylene chloride, and acetone were tested critically
using sonic vibration and Soxhlet distillation. These tests included compari-
sons of collected particulates and comparisons of particulates which had been
fortified with known amounts of PNA's. Test results have not been completely
conclusive but some data show definitely that current extraction procedures may
be inadequate. Positive results reported in the literature for some procedures
are probably due to improper fortification procedures.
Water samples—The techniques for the measurement of PNA's, acenaphthylene,
phenanthrene, fluoranthene, pyrene, anthracene, chrysene, perylene, and benzo-
pyrene in water at the 20 to 500 g/liter level were established.
36
-------
SECTION 7
RESULTS
GENERAL
Most of the results are reported in a data matrix format for each boiler
unit. The independently controlled boiler parameters of percentage nominal
RDF heat energy input (0, 20, or 50) and percentage nominal load (60, 80, or
100) are the major headings in the data matrix. The measured variable for
each experimental run is tabulated in each cell of the data matrix along with
its test number designation (EPA No.)- Also for each cell, the cell average,
cell standard deviation, and coefficient of variation (standard deviation
divided by the mean) are tabulated.
Following each data matrix are plots of the measured variables as func-
tions of both RDF heat energy input and boiler steam load. Each data point
on the plot is the cell average, and the vertical bar through the data point
represents the cell standard deviation. Thus, each data point usually repre-
sents the average of three experimental runs. It should be noted that two
standard deviations on each side of the cell average would approximately
correspond to the 95% statistical confidence band. This means that there
would be about 95% probability that the actual value of the measured variable
would be within a range of the two standard deviations on each side of the
cell average.
Curves and lines drawn through the data points of these plots are meant
to show general trends in the data. Such curves or lines are not meant to
imply any particular mathematical expression which may (or may not) govern
the behavior of a particular measured variable.
37
-------
INPUT STREAMS
Coal
The heating value and ultimate analysis constituents for coal used in
boiler unit Nos. 5 and 6 are tabulated respectively in Tables 9 and 10.
The values of the average, the maximum, the minimum, the standard deviation,
the coefficient of variation, and the sample size used for these determina-
tions are given in these tables. The 95% confidence interval is also given
based on the t statistic distribution.
Chemical analysis for the elements contained in the coal samples are
presented by element alphabetically in the data matrix format in Appendix A.
The same analysis listed by EPA run number for the tests performed on boiler
unit Nos. 5 and 6 are contained in Appendix B. Based on measured coal flow rates
the predicted flow rate of each element into the boiler is also tabulated in
Appendix B.
Refuse Derived Fuel
The heating value and ultimate analysis constituents for RDF used in
boiler unit Nos. 5 and 6 are tabulated respectively in Tables 9 and 10.
The values of the average, the maximum, the minimum, the standard deviation,
the coefficient of variation, and the sample size used for these determination
are given in these tables. The 95% confidence interval is also given based
on the t statistic distribution.
Chemical analysis for the elements contained in the RDF are presented by
element alphabetically in the data matrix format in Appendix A. The same
analysis listed by EPA run number for the tests performed on boiler unit No. 5
and 6 are contained in Appendix C. Based on measured RDF flow rates the
predicted flow rate of each element into the boiler is also tabulated in
Appendix C.
Air
Dry bulb temperature, wet bulb temperature, barometric pressure, and
humidity are tabulated in Table 11. The average values of these measurements
as well as the standard deviation is given for each test cell of given load
and given amount of RDF input in this table.
38
-------
TABLE 9. SUMMARY OF COAL, RDF, AND ASH CHARACTERISTICS FOR BOILER UNIT NO. 5
Coal:
Iowa
k&
kg
RDF
k&
kg
7. Load/7. RDF
60
0
Bottom Ash
60
20
60
50
7. Load/ 7. RDF
80
0
80
20
Bottom Ash
80
50
100
0
100
20
100
50
Quantity
HHV (kj/kg)
FM
ASH
C
H
S
Cl
0
p(kg/m3)
HHV (kJ/kg)
FM
ASH
C
H
S
Cl
0
c x
H 7.
S 7.
Mineral 7.
C
H
S
Mineral
C
H
S
Mineral
C 7.
H 7.
S 7.
Mineral 7.
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
Mean
21,848
0.1081
0.2024
0.5171
0.0396
0.0666
0.0004
0.0658
125.36
13,063
0.2177
0.1660
0.3067
0.0231
0.0040
0.0025
0.2703
8.59
0.87
3.49
87.06
7.50
0.86
3.18
88.46
7.01
0.90
2.31
89.77
8.81
0.67
2.92
87.60
13.01
0.92
3.54
82.54
6.84
0.77
2.24
90.15
9.44
0.77
2.87
86.91
9.64
0.63
3.75
85.96
11.16
0.70
3.12
85.04
Maximum
23,057
0.1359
0.2296
0.5461
0.0476
0.0876
0.0013
0.0948
140.24
15,989
0.3012
0.2419
0.3726
0.370
0.0090
0.0052
0.3540
12.25
1.42
4.19
89.41
10.9
1.02
3.51
90.48
8.13
0.94
3.25
91.78
9.69
0.76
3.82
88.07
14.21
1.01
4.13
84.07
2.88
0.97
8.03
92.03
12.46
0.94
2.98
88.53
11.30
0.74
4.09
87.22
13.32
1.01
3.24
87.39
Minimum
20,804
0.0772
0.1818
0.4872
0.0316
0.0532
0
0.0424
111.33
11,128
0.1461
0.1133
0.2786
0.0166
0.0019
0.0018
0.2110
5.97
0.54
3.13
83.85
5.78
0.70
2.72
84.71
5.66
0.84
1.63
87.78
8.35
0.54
2.16
87.12
12.59
0.79
2.75
81.05
1.71
0.66
5.60
88.12
7.81
0.51
2.80
84.05
8.68
0.45
3.13
84.16
8.99
0.38
2.99
82.68
Standard
deviation
--
0.0150
0.0118
0.0139
0.0046
0.0072
0.0003
0.0116
8.123
1,092
0.0324
0.0320
0.0239
0.0056
0.0020
0.0008
0.0450
2.79
0.38
0.49
2.69
2.37
0.13
0.34
2.61
1.25
0.06
0.84
2.0
0.77
0.12
0.84
0.48
1.05
0.11
0.71
1.51
1.22
0.17
0.59
1.96
2.62
0.23
0.09
2.49
1.44
0.16
0.54
1.60
3.06
0.45
0.18
3.33
Coefficient
of
variation
0.0516
0.1389
0.0581
0.0269
0.1162
0.1080
0.9348
0.1763
0.0648
0.0836
0.1486
0.1928
0.0780
0.2424
0.5035
0.3346
0.1665
0.325
0.473
0.140
0.031
0.315
0.151
0.107
0.030
0.178
0.067
0.364
0.022
0.087
0.179
0.288
0.005
0.081
0.120
0.201
0.018
0.178
0.221
0.263
0.022
0.278
0.299
0.031
0 029
0.149
0.254
0.144
0.019
0.274
0.643
0.058
0.039
Sample
size
28
28
28
28
28
28
28
28
18
18
18
18
18
18
18
18
18
4
4
4
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
957.
Confidence
interval
±
482
0.0308
0.0241
0.0285
0.0094
0.0148
0.0007
0.0238
1.0788
991
0.0683
0.0675
0.0505
0.0118
0.0042
0.0017
0.0950
8.078
1.209
1.559
8.560
7.541
0.414
1.082
8.305
5.379
0.258
3.615
8.606
3.313
0.516
3.615
2.065
4.518
0.473
3.055
6.498
5.250
0.732
2.539
8.434
11.274
0.990
0.387
10.714
6.196
0.688
2.324
6.885
38.880
5.718
2.287
42.311
t
Statistic
2.052
2.110
3.182
3.182
4.303
4.303
4.303
4.303
4.303
4.303
12.706
(continued)
39
-------
TABLE 9. (Concluded)
% Load/%RDF
Collector Ash
1, Load/% RDF
Collector Ash
OVERALL
Bottom Ash
Collector Ash
60
0
60
20
60
50
80
0
80
20
80
50
100
0
100
20
100
50
Quantity
C %
H 1.
S 7.
Mineral '/.
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C %
H 7.
S %
Mineral 7.
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
-Mean
12.83
0.65
1.82
84.71
7.70
0.60
2.14
89.57
3.37
0.51
1.03
95.10
12.98
0.61
1.98
84.44
7.61
0.43
3.66
88.29
8.14
0.67
2.33
88.85
17.97
0.53
2.04
79.47
8.74
0.56
2.61
88.10
8.60
0.81
2.53
88.06
8.963
0.795
3.07
87.18
9.723
0.590
2.210
87.274
Maximum
20.15
0.82
2.16
88.90
5.80
0.79
2.33
91.53
4.57
0.67
1.14
96.26
11.05
0.65
2.07
86.47
12.5
0.44
4.22
90.17
13.7
0.85
2.61
93.48
26.71
0.68
2.14
85.38
11.30
00.59
2.85
91.56
9.57
0.90
2.81
89.23
14.21
1.42
4.19
92.03
26.71
0.90
4.22
96.26
Minimum
8.39
0.55
1.64
77.18
8.71
0.43
1.85
88.52
2.16
0.34
0.91
93.95
15.21
0.55
1.93
82.09
5.17
0.41
2.55
84.54
3.73
0.38
1.97
82.90
12.16
0.32
1.90
70.71
5.29
0.51
2.33
85.78
7.63
0.72
2.24
86.90
5.6
0.38
1.63
81.05
2.16
0.32
1.03
70.71
Standard
deviation
5.08
0.12
0.25
5.16
1.64
0.18
0.26
1.70
1.21
0.17
0.12
1.16
2.10
0.05
0.08
2.21
4.23
0.02
0.96
3.25
5.08
0.26
0.33
5.41
7.71
0.19
0.12
7.74
3.10
0.04
0.26
3.05
1.37
0.13
0.40
1.65
2.524
0.214
0.7
2.88
5.388
0.156
0.756
5.427
Coefficient
of
variation
0.396
0.185
0.137
0.061
0.213
0.300
0.121
0.019
0.359
0.333
0.117
0.012
0.162
0.082
0.040
0.026
0.556
0.047
0.262
0.037
0.624
0.388
0.142
0.061
0.429
0.358
0.059
0.097
0.355
0.071
0.100
0.035
0.159
0.160
0.158
0.019
0.212
0.269
0.227
0.033
0.554
0.265
0.342
0.062
Sample
size
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
2
2
2
28
28
28
28
27
27
27
27
Confidence
interval
±
16.165
0.382
0.796
16.419
7.057
0.775
1.119
7.315
5.207
7.32
0.516
4.991
9.036
0.215
0.344
9.510
18.202
0.086
4.131
13.985
21.859
1.119
1.420
23.279
33.176
0.818
0.516
33.305
13.339
0.172
1.119
13.124
17.407
1.652
5.082
20.965
5.179
0.439
1.436
5.910
11.078
0.545
0.703
0.127
t
Statistic
3.182
4.303
4.303
4.303
4.303
4.303
4.303
4.303
12.706
2.052
2.056
40
-------
TABLE 10. SUMMARY OF COAL, RDF, AND ASH CHARACTERISTICS FOR BOILER
UNIT NO. 6
Coalt
Iowa /Wyoming
k&
kg
RDF
kg.
kg
% Load/% RDF
Bottom Ash
% Load/% RDF
Collector Ash
OVERALL
Bottom Ash
Collector Ash
80
0
80
20
80
50
80
0
80
20
80
50
Quantity
HHV(kJ/kg)
FM
Ash
C
H
S
Cl
0
P(kg/m3)
HHV(kJ/kg)
FM
Ash
C
H
S
Cl
0
C %
H %
S %
Mineral 7.
C
H
S
Mineral
C
H
S
Mineral
C 7.
H 7.
S •'.
Mineral %
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
C
H
S
Mineral
Mean
22,262
0.1563
0.1162
0.5408
0.0416
0.0346
0.0005
0.1100
126.87
12.532
0.2523
0.1496
0.2957
0.0150
0.0031
0.0031
0.28n
3.77
0.81
1.95
93.48
2.84
0.71
1.65
94.79
2.31
0.51
1.09
96.09
12.93
0.64
1.68
84.75
9.51
0.58
1.94
87.97
10.17
0.85
1.79
87.19
2.974
0.677
1.563
94.786
10.869
0.690
1.804
86.637
Maximum
22,860
0.1666
0.1336
0.5572
0.0529
0.0476
0.0007
0.1260
136.96
13.179
0.3112
0.1696
0.3244
0.0217
0.0039
0.0067
0.3160
7.50
0.94
2.18
96.20
3.34
0.93
1.87
95.46
2.54
0.60
1.30
96.43
15.18
0.88
1.80
87.55
10.11
0.73
2.22
88.09
15.36
1.16
2.35
90.46
7.50
0.94
2.18
96.43
15.36
1.16
2.35
90.46
Minimum
21,439
0.1344
0.1006
0.5269
0.0371
0.0261
0.0003
0.0951
111.49
11.514
0.2016
0.1309
0.2737
0.064
0.0023
0.0020
0.2600
1.32
0.66
1.65
89.66
2.56
0.53
1.45
94.09
2.12
0.33
0.85
95.83
10.38
0.42
1.58
82.36
9.20
0.48
1.62
87.73
6.98
0.69
1.40
81.60
1.32
0.33
0.85
89.66
6.98
0.42
1.40
81.60
Standard
deviation
176
0.0108
0.0119
0.0092
0.0048
0.0064
0.0001
0.0097
78.87
1,872
0.0393
0.0150
0.0175
0.0052
0.0007
0.0018
0.0196
3.28
0.14
0.27
3.40
0.43
0.20
0.21
0.69
0.21
0.15
0.23
0.31
2.41
0.23
0.11
2.62
0.52
0.13
0.30
0.21
4.54
0.27
0.50
4.87
1.778
0.198
0.429
2.078
3.025
0.224
0.318
3.124
Coefficient
of
variation
0.0184
0.0690
0.1024
0.0177
0.1154
0.1848
0.2669
0.0882
0.0785
0.1494
0.1559
0.1006
0.0592
0.3467
0.2231
0.5890
0.0683
0.870
0.173
0.138
0.036
0.151
0.282
0.127
0.007
0.091
0.294
0.211
0.003
0.186
0.359
0.065
0.031
0.055
0.224
0.155
0.002
0.446
0.318
0.279
0.056
0.598
0.292
0.275
0 022
0.273
0.325
0.176
0.036
Sample
size
9
9
9
9
9
9
9
9
6
6
6
6
6
6
6
6
6
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
9
9
9
9
9
9
9
9
95%
Confidence
interval
±
407
0.0249
0.0274
0.0213
0.0111
0.0148
0.0003
0.0224
1.5985
684
0.1011
0.0387
0.0450
0.1645
0.0017
0.0046
0.0504
14.114
0.602
1.162
14.630
1.850
0.861
0.904
2.969
0.904
0.645
0.990
1.334
10.370
0.990
0.473
11.274
2.238
0.559
1.291
0.246
19.536
1.162
2.152
20.956
4.100
0.457
0.989
4.792
6.976
0.519
0.733
7.204
t
Statistic
2.306
2.571
4.303
4.303
4.303
4.303
4.303
4.303
2.306
2.306
41
-------
TABLE 11. SUMMARY OF AIR, FEEDWATER, AND STEAM CHARACTERISTICS FOR
RDF
Load I
Unit No. 5
60 0
60 20
60 50
80 0
80 20
80 50
100 0
100 20
100 50
Unit No. 6
80 0
80 20
80 50
EPA
4A
4B
20
21
36
Ave
SD
8
9A
9B
33
Ave
SD
1
34
35
Ave
SD
5
16
17
Ave
SD
6
12
13
Ave
SD
2
10
15
Ave
SD
11
31
32
Ave
SD
7
14
19
Ave
SD
3
18
Ave
SD
24
29
30
Ave
SD
25
26
27
Ave
SD
22
23
28
Ave
SD
Steam
load
60.00
60.00
59.47
59.47
58.95
59.59
0.44
60.00
60.00
60.00
60.00
60.00
0.0
61.05
59.47
60.53
60.35
0.81
80.00
80.53
81.05
80.53
0.53
78.95
78.95
81.05
79.65
1.21
80.00
79.47
77.89
79.12
1.10
97.37
97.89
93.16
96.14
2.59
97.89
94.74
92.63
95.09
2.65
92.63
96.84
94.74
2.98
80.40
79.20
81.20
80.27
1.01
80.80
81.20
79.20
80.40
1.06
79.20
79.20
78.40
78.93
0.46
RDF heat
(7.)
0
0
0
0
0
0
0
21.7
24.1
24.2
16.0
21.50
3.84
53.9
54.9
67.7
58.83
7.70
0
0
0
0
0
22.2
31.6
27.7
27.17
4.72
46.5
49.2
39.5
45.07
5.01
0
0
0
0
0
20.8
26.6
29.5
25.63
4.43
40.9
44.3
42.6
2.40
0
0
0
0
0
19.5
21.2
21.5
20.73
1.08
47.8
54.6
32.2
44.87
11.48
tempera
Wet bulb
°C
20
20
20
18
21
20
1.10
18.1
13
16
23
17.5
4.21
21
22
23
22.0
1.0
17.5
21
24
20.8
3.25
17
21
23
20.3
3.06
21
16
20
19
2.65
20
19
23
21
2.08
20
18
20
19
1.15
19
24
21.5
3.54
19
16
26
20.3
5.13
23
26
24
24
1.53
23
18
21
20.7
2.52
ture
Dry bulb
°C
32
32
32
24
31
30.2
3.49
31
18
24
31
26
6.27
30
30
28
29.3
1.15
30.3
27
31
29.4
2.14
25
29
33
29.0
4.0
28
24
34.7
28.9
5.41
29
23
28
27
3.21
31
31
26
29
2.89
23
28
25.5
3.54
26
19
30
25.0
5.57
29
30
30
29.7
0.58
30
23
30
27.7
4.04
Relative
humidity
(7.)
32
32
32
54
39
45.8
17.2
25
56
38
48
41.8
13.4
41
50
62
51
10.5
26
60
58
48
19.1
42
46
42
43.3
2.3
54
40
24
39.3
15.0
43
69
69
60.3
15.0
32
25
59
38.7
18.0
66
76
71
7.1
49
66
70
61.7
11.2
63
70
63
65.3
4.0
52
64
45
53.7
9.6
Barometric
pressure
kPa
97.02
97.02
98.71
98.78
97.32
97.77
0.90
97.93
98.14
98.07
98.21
98.09
0.12
98.00
97.66
97.32
97.66
0.34
98.00
97.93
97.83
97.92
0.085
97.63
97.90
97.70
97.84
0.13
97.26
98.00
98.21
97.82
0.50
98.00
98.34
98.27
98.20
0.18
97.36
98.27
98.48
98.04
0.60
96.95
98.04
97.50
0.77
97.80
98.51
98.27
98.19
0.36
97.66
97.90
97.93
97.83
0.15
97.49
98.24
97.66
97.80
0.39
Feedwater Steam
Flow Temp. temp.
kg/hr °C °C
25,220
25,220
22,312
23,088
22,203
23,610
1,509
23,224
23,496
22,226
20,911
22,464
1,171
23,995
21,686
21,546
22,408
1,376
32,205
30,663
30,164
31,011
1,064
30,980
29,484
31,116
30,527
906
30,844
29,438
29,084
29,789
931
37,920
38,873
37,195
37,996
842
37,195
34,745
37,784
36,575
1,612
38,011
37,966
37,989
31.8
45,772
45,391
46,412
45,858
516
46,126
44,225
45,382
45 , 244
958
46,266
46.706
44,375
45,782
1,238
152 449
152 449
151 439
153 440
129 435
147.4 442
10.31 6.31
152 444
150 437
149 438
134 436
146.3 439
8.26 3.6
147 442
135 449
135 449
139 447
6.93 4.0
159 457
138 453
139 457
145.3 456
11.85 2.31
159 446
158 452
158 450
158.3 449
0.58 3.06
159 455
156 460
140 451
141.7 455
10.21 4.51
161 442
147 456
145 444
151 447
8.72 7.57
164 439
138 454
144 433
148.7 442
13.6 10.8
167 481
143 456
155 468.5
17.0 17.7
168 446
168 448
167 445
167.7 446
0.58 1.53
166.7 449
167 446
167 448
167 448
0.17 1.53
167 444
169 445
166 447
167 445
1.53 1.53
kPa
4,254
4,254
4,309
4,323
4,334
4,297
41
4,309
4,275
4,268
4,323
4,294
26
4,289
4,357
4,433
4,360
72
4,243
4,302
4,323
4,293
35
4,220
4,289
4,213
4,241
42
4,140
4,164
4,213
4,172
37
4,351
4,309
4,323
4,328
21
4,337
4,289
4,282
4,303
30
4,302
4,337
4,320
25
4,502
4,557
4,599
4,553
49
4,557
4,585
4,551
4,564
18
4,633
4,551
4,544
4,576
49
/ Flow
kg/hr
28,236
28,236
26,331
22,861
26,331
26,399
2,195
26,807
27,079
25,719
25,174
26,195
899
26,399
26,399
26,195
26,331
118
35,584
34,768
35,040
35,131
415
34,836
34,087
35,108
34,677
529
34,087
34,292
33,407
33,929
463
43,069
41,232
40,075
41,459
1,510
42,456
39.599
41,504
41,186
1,455
56,404
42,048
49,226
10,151
49,464
47,423
49,736
48,874
1,264
50,281
50,145
49,056
49,827
,671
48.321
48,920
48,444
48,562
316
_a/ Absolute pressure.
42
-------
EXIT STREAMS
Ash
Grate (Bottom) Ash—
The bottom (grate) ash characteristics including the combustible frac-
tion of carbon, hydrogen, and sulfur are reported in Tables 9 and 10. The
data reported include average values, maximum, minimum, standard deviation,
coefficient of variation, and sample size. The 95% confidence interval is
also given based on the t statistic distribution.
The results of the chemical analysis of the ash are given in Appendix A
in the data matrix format. The sample analysis listed by EPA run number is
given in Appendix D. Based on measured flow rates of grate ash, the predicted
flow rate of each element out of the boiler is also tabulated in Appendix D.
Hopper (Fly) Ash—
The hopper ash characteristics are presented in Table 9 and Table 10.
The combustible fractions of carbon, hydrogen, and sulfur are reported in
these tables. The data presented includes average values, maximum, minimum,
standard deviation, coefficient of variation, and sample size.
A chemical analysis for the elements in the hopper ash has also been
performed and the results are given in the Appendix A in the data matrix
format. The same analysis listed by EPA run number is given in Appendix E.
Based on measured flow rates of hopper ash, predicted flow rate of each ele-
ment out of the boiler is also tabulated in Appendix E.
Steam
The steam flow rate, steam temperature, steam pressure, feedwater flow
rate, and feedwater temperature are tabulated in Table 11.
The averages of these quantities and their standard deviations are given
for each test cell of given load and given amount of RDF input.
The feedwater flow rate was monitored as a check on the steam flow inte-
grator reading. However, as can be seen from Table 11, the feedwater flow
rate measurement was consistently below the steam integrator reading of each
boiler. Checks on the calibrations of the steam flow measurement devices
have revealed the steam flow integrator reading to be the appropriate flow
rate. All steam loads of this study were consequently obtained from the
steam flow integrator reading.
43
-------
Stack Effluents
Particulates—•
The results for the particulate matter, measured as a portion of the
stack effluents, are summarized in the data matrix format of Tables 12 and 13
for boiler units Nos. 5 and 6, respectively. The results for the particu-
late effluent, measured before the particulate collector, are summarized in
the data matrix format of Tables 14 and 15. The grams of particulate emit-
ted per megajoule of heat energy input to the boiler are shown in Figures 12
and 13, respectively, as a function of actual RDF heat energy input to the
boiler and actual steam load of the boiler. Figures 14 and 15 are similar
plots except that the particulate matter has been sampled from ports located
before the particulate collector. Figures 16 and 17 are plots showing the
collector efficiency as determined from the particulate measurements ob-
tained from locations both before (BPC) and after (stack) the particulate
collectors.
Figures 16 and 17 show that the particulate emissions from the stack of
boiler unit No. 5 decrease with RDF heat energy input at both the 80 and
100% steam load conditions. However, for the 60% load condition, the par-
ticulate emissions increase with increases in RDF heat energy input. For
boiler unit No. 6, the particulate emissions also increase with increases
in RDF heat energy input. Thus, for boiler unit Nos. 5 and 6, with 80%
load condition, the trends in particulate emission rates appear to be op-
posite. However, it should be noted that some factors not controlled in
the experiment may explain the apparent reversal in trends between the two
traveling grate stoker units. The major factor in this category is excess
air. Figures 18 and 19 are plots showing the variation in excess air with
RDF heat energy input and steam load, respectively. These plots show that
the amount of excess air at 80% steam load generally increases for boiler
unit No. 6 and, except for the 80% load condition, appear to also increase
for boiler unit No. 5 with RDF heat energy input. However, the trends of
particulate effluent with RDF are not always the same as the trends of ex-
cess air with RDF.
Except for the 80% load tests on boiler unit No. 5, the excess air
supplied to the boiler generally increases with corresponding increases in
RDF heat energy input. However, the excess air generally decreases with
increases in boiler load as shown in Figure 19.
The particulate effluent measured before the particulate collector ap-
pears to have less consistent trends and much more variation that corres-
ponding particulate effluent measured in the stack. Results of the col-
lector efficiency are presented in Tables 16 and 17.
44
-------
TABLE 12. DATA MATRIX OF UNIT NO. 5 STACK PARTICULATE EMISSIONS
EMISSION RATE, GRAMS PEk MEGAJJULE. STACK,CONC
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
*****************
* *
* *
* PERCENT *
* *******
* REFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
20
*******
*
*
*
*
*
*
* 50
*
*
*
*******
E *
*
*
******* ***
*
*
*
*
*
*
*
*
*
*
* ** **
*
*
*
*
*
*
*
*
*
*
EPA#
04A
04B
20
21
36
AVE
SD
cv
** ***
EPA*
08
0 ^A
09B
3 3
AVE
SD
CV
***** ** ***
*
*
*
*
*
*
*
*
*
*
*****
EPA*
0 1
34
3b
AVE
SD
cv
*****
***
***
60
***
( G
7
9
1
7
1
9
4
5
***
*********
*********
*********
M/MEGJ )
. 33E-01
. 64E-01
. 93E-01
. 64E-01
. 6
5
***
. 97E-01
.54F 00
. 10E-01
.54E 00
.05E 00
. 7 3 F - 0 1
.46E-01
*** ******
M/MEGJ )
. 141--01
.35E 00
.89E UO
.2bE 00
. 9 3 F - o i
.54E-01
*********
********************
PERCENT LOAD
********************
*
*
*
********
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
**
*
*
*
**
EPAW
05
1 7
AVE
SO
CV
******
EPA#
06
12
13
AVE
SD
cv
******
EPA*
02
I 0
15
AVfc
SD
cv
30
************
(GM/MEGJ )
4 .05E-01
1 .06E 00
7 . 3 1 E - 0 1
4 .61E-01
6 .31E-01
************
(GM/MEGJ )
3.89E-01
3.25E-01
3 .94E-0 1
3. 69E-01
3.84F-02
1 .04E-0 1
************
( GM/MEGJ )
2 .70E-01
3 . 4 3E - 0 1
4 .OOE-01
3. 3 BE- 01
6.51E-02
1 .93E-01
******************
****
****
*
*
*
****
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
****
******************
*
*
*
******************
100
************
EPA#
1 1
31
32
AVE
Su
CV
****
EPA*
07
14
19
AVE
SD
CV
****
EPA*
03S
1 8
AVE
SD
CV
*
*
*
******
(GM/MEGJ ) *
3.23E-
1 .65E
2. 02E
1 .336
8.94E-
6 .71E-
********
01 *
00 *
00 *
*
*
*
00 *
01 *
01 *
******
(GM/MEGJ ) *
4 .OOfc-
2.96E-
3.55E-
3 .50h-
5. 22E-
1 .49E-
********
01 *
01 *
01 *
*
*
*
01 *
0? *
01 *
******
(GM/MEGJ) *
? . 75E-
4.52E-
3 .64E-
1 .25E-
3.44E-
************
01 *
01 *
<:
*
*
*
01 *
01 *
01 *
******
-------
TABLE 13. DATA MATRIX OF UNIT NO. 6 STACK PARTICULATE EMISSIONS
EMISSION RATE. GRAMS PER MEGAJOULE, STACK,CONC
BOILER NUMBER 6
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
***********************************************************************
*
* PERCE
*
* REFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
"i
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
*
*
NT *
*******
E *
*
*
***** *****
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
PERCENT LOAD
**************
*
60 *
*
*
*
*
****************************************
****** ***************
(GM/MEGJ) *
*
*
*
*
*
*
*
*
*
EPA«
24-
29
30
AVE
SO
CV
80
*** ********
( GM/MEGJ )
6. 02E-01
5.25E-01
3.39E-01
4.88E-01
1 .35E-01
2. 77E-01
********************************
(GM/MEGJ) *
*
*
*
*
*
a
*
*
*
**** **********
(GM/MEGJ) *
*
*
*
*
#
*
*
*
*
EPA*
25
26
27
AVE
50
CV
(GM/MEGJ )
5. 05E-01
6.46E-01
8. 75E-01
6. 76E-01
1 .87E-01
2. 76E-01
******************
EPA*
22
23
23
AVE
SD
CV
(GM/MEGJ )
1 . 02E 00
1 . 01E 00
7.59E-01
9. 30E-01
1 .48E-01
1 .59E-01
*
*
*
**#*
*
*
*
*
*
*
*
*
*
*
****
#
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
*
100 *
*
******************
EPA* (GM/MEGJ) *
*
*
*
*
*
*
*
*
*
******************
EPA* ( GM/MEGJ ) *
*
*
*
*
*
*
*
*
*
******************
EPA* ( GM/MEGJ ) *
*
*
*
*
*
*
*
*
*
****************************#****:*******]!<************************ ******
-------
TABLE 14. DATA MATRIX OF UNIT NO. 5 PARTICIPATE FLOW RATE
BEFORE THE PARTICIPATE COLLECTOR
EMI SSI
BO ILER
PEWC'-'N
*******
*
*
* PERCE
*
* REFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*
*
*
*
*
*
» 50
*
*
*
*******
ON RATc. GntAMS PER MtG
NUMdER 5
T REFUSE AND LOAD ARE
**********************
*
*
NT *
* *
E *
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*
*
*
*.
*
*
*
*
*
*****
>ft if. 4( A ift
*** *
FPA*
04A
04i3
20
2 1
3 b
AVE
30
CV
** * **
EPA*
03
09 A
0->Q
3 3
AVE
30
C V
EPA*
0 1
34
35
A Vb
SO
CV
** * **
on
( GM/MEGJ )
3 . ObE 00
3 .0 IE 00
3 . 1 dE 00
2 . 6 3 E 00
3 . <; 1 E 00
3 . 1 SE 00
4 . 70 f--Q\
1 .4TL-01
************
(GM/MEGJ )
3. 96E 00
5 . 1 3E 00
4 .d6F 00
3 . d5E 00
4 . 4 b F 00
•.:> . 4 4 E - 0 1
1 .45E-01
( (jM/Mt GJ )
1 . tMF 00
3 . Si&tr. 00
4 . 0 3 E 00
3.4 6E 00
1 ,44F 00
4 . 1 6E-01
************
AJOULE ,
NOM I NAL
********
*
*
*
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
**
PER
rfr * * A ik *
^ W- ^ ~ V T
EPA*
OS
1 6
1 7
AVE
SO
CV
******
EPA*
Ob
1 2
13
AVE
so
CV
EPA*
1 0
1 5
A VE
SO
CV
******
BPC . CONC
VALUES
**** ********
CENT LOAD
HO
( GM/MEGJ )
3 . 8 IE 00
2. 87E 00
4 . 01E 00
3.56E 00
6. C7E-01
1 . 70E-01
************
( GM/MEGJ )
4.04E 00
4.32E 00
3.84E 00
4 . 07E 00
~> .4 It- 0 1
5 .^2E-02
( GM/MEGJ )
3.b6E 00
3. 30E 00
3.43E 00
1 .d7E-01
5 .46E-02
************
****
*
*
*
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
* * * *
** **
FPA*
1 1
31
32
AVE
SO
CV
** **
EPA *
07
14
l y
AVE
SO
CV
EPA #
03S
lb
AVt
SO
cv
* * * *
*******
1 00
( GM/ME
2 . 8bE
4 . d6E
4 . b4E
4 . 12E
1 . 1 OL
2 .67E
*******
(GM/ME
3. 18E
1 . 54E
1 . bbE
2 .23E-
8.45E
3 .80E
*****
*****
**
*
*
*
* *
*
*
* ******
G J ) *
00
00
00
00
00
-01
* *** *
GJ )
00
00
00
00
-01
-01
( GM/MEGJ )
4. 01E
2. 24F
3. 1 3E
1 .25E
4 .OOE
*******
00
00
00
00
-01
* * * * *
*
*
*
*
*
*
*
*
•if
**
*
*
*
*
*
*
*
*
*
*
* *
*
*
*
*
*
*
*
*
*
*
*• f
-------
TABLE 15. DATA MATRIX OF UNIT NO. 6 PARTICIPATE FLOW RATE
BEFORE THE PARTICULATE COLLECTOR
EMISSION KATE, GRAMS PER MEGAJOULE, OPC . COMC
BOILER NUMBER 6
PtWCENT REFUSE AND LQAC ARE NOMINAL VALUES
oo
*******
*
*
* PERCE
*
***** *****
*
*
NT *
** *****
* REFUSE *
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA#
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
********************************
PERCENT LOAD
********************
*
60 *
*
**************
(GM/MEGJ) *
*
*
*
*
*
*
*
*
*
**************
(GM/MEGJ) *
*
*
*
*
*
*
ft
*
*
* ** ***
EPA*
24
29
30
Ave
SD
cv
****
****************
80
**** ********
{ GM/MEGJ )
2.69E 00
2.S9E 00
1 . 74E 00
2.47E 00
6.54E-01
2.64E-01
******************
EPA*
25
26
27
AVE
SD
CV
********************
(GM/MEGJ) *
*
*
*
*
*
*
*
*
*
EPA*
22
23
28
AVE
SD
CV
(GM/MEGJ )
3.52E 00
3 . 7 OF 00
3 .1 9E 00
3.47E 00
?. 57FE-01
7 .41E-02
************
(GM/MEGJ )
5.24E 00
4.36E 00
3.53E 00
4.38E 00
8. 53E-01
1 .95E-01
*
*
*
******************
*
*
*
******************
*
100 *
*
**********************
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
EPA* (GM/MEGJ) *
*
*
*
*
*
*
*
*
*
******************
EPA# (GM/MEGJ) *
*
*
*
*
*
*
*
*
*
**********************
*
*
*
*
*
#
*
*
*
*
EPA# (GM/MEGJ) *
*
*
*
*
*
*
*
*
*
***********************************************************************
-------
Stack Emissions
Ames Power Plant
Boilers 5 and 6
1.4
1.2
1.0
O
. 0.8
LLJ
i
z
g
12 0.6
0.4
0.2
Symbol Load (% NOM)
O 60
a 80
^ 100
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
I
I
10
20 30 40
RDF HEAT INPUT, PCT
50
60
Figure 12. Stack particulate emission rate of boiler unit Nos. 5 and
6 as a function of refuse derived fuel heat input.
49
-------
1.6
1.4
1.2
8 i.
. 0.8
£ 0.6
0.4
0.2
Stack Emissions
Ames Power Plant
Boilers 5 and 6
o~ 0% RDF
a~ 20% RDF
50% RDF
Symbol
Open
Shaded
I
Confidence Interval at
95% Confidence Level
I
50
60 70 80 90
STEAM LOAD, PCT RATED OUPTUT
Figure 13. Stack particulate emission rate of boiler unit Nos. 5 and
6 as a function of steam load.
50
-------
5.0
4.8
4.4
4.0
LU
_J
O
—1
o
2 3.6
O
O
3.2
2.8
2.4
2.0 -
Emission Rotes
Before Particulafe Collector
Ames Power Plant
Boi lers 5 and 6
Unit Symbol
5 Open
6 Shaded
Confidence Interval
at 95 c- Confidence
Level
To
1.39
I
10
I
J_
50
60
Figure 14
20 30 40
RDF HEAT INPUT, PCT
Particulate flow rate of boiler unit Nos. 5 and 6 before-
the-particulate-collector as a function of refuse derived
fuel heat input.
51
-------
o
O
z
o
4.8
4.4
4.0
3.6
3.2
2.8
2.4 -
2.0 -
Emission Rates
Before Participate
Ames Power Plant
Boilers 5 and 6
Confidence Interval at
95% Confidence Level
I
I
60 70 80 90
STEAM LOAD, PCT RATED OUTPUT
100
Figure 15. Particulate flow rate of boiler unit Nos. 5 and 6
before-the-particulate-collector as a function of
steam load.
52
-------
100
90
80
(j 70
Z
O 60
I—
u
O
u
50
40
>
10 -
Collector Efficiency
Ames Power Plant
Boilers 5 and 6
10
O 60 % Load
a 80 % Load
A 100 % Load
Unit Symbol
5 Open
6 Shaded
Confidence Interval af
95% Confidence Level
I
I
I
I
50
60
Figure 16.
20 30 40
RDF HEAT INPUT, PCT
Particulate collector efficiency of boiler unit Nos. 5 and 6
as a function of refuse derived fuel heat input.
53
-------
100 r-
90
80
70
y
LU-
LL.
LU
an
O
H-
(J
UJ
O
u
60
50
40-
10 f-
50
Collector Efficiency
Ames Power Plant
Boilers 5 and 6
I
O 0 % RDF
a 20 % RDF
o 50 % RDF
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
I
I
90
100
Figure 17.
60 70 80
STEAM LOAD, PCT RATED OUTPUT
Particulate collector efficiency of boiler unit Nos. 5 and 6
as a function of steam load.
54
-------
0£
LU
Q.
a:
<
U
X
200
180
160
140
120
80
60
a ~
60 % Load
80 % Load
100 % Load
Unit Symbol
5 Open
6 Shaded
40
Confidence Interval at
95% Confidence Level
Figure 18,
I
I
10
20 30 40
PERCENT RDF HEAT INPUT
50
60
Percent excess air of boiler unit Nos. 5 and 6 as a
function of refuse derived fuel heat input.
55
-------
200 r-
180
160
140
U
Oi
120
U
* 100
80
60
40
4
0
u
Figure 19,
1
o~ 0% RDF
o~ 20% RDF
*~ 50% RDF
Unit Symbol
5 Open
6 Shaded
J_
I
I
I
I
i
Confidence Interval at
95% Confidence Level
I
60 70 80 90
STEAM LOAD, PCT RATED OUTPUT
100
Percent excess air of boiler unit Nos. 5 and 6
as a function of steam load.
56
-------
TABLE 16. DATA MATRIX OF BOILER UNIT NO. 5 PARTICULATE COLLECTOR EFFICIENCY
COLLECTOR EFFICIENCY
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
***********************************************************************
*
*
* PERCEi
* REFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
Xe
*
*
*
*
NT *
4: Tk
E *
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
* ****
*
*
*
*
*
*
*
*
*
*
** * **
EPA*
04A
0*3
20
2 I
3'i
AVE
S 0
cv
** ***
EPA*
0 si
0 9A
0 JH
3 3
AVE
SO
cv
** * **
EPA*
0 1
34
3 5
AViI
3 0
C V
60
*******
7 .
-------
TABLE 17. DATA MATRIX OF BOILER UNIT NO. 6 PARTICULATE COLLECTOR EFFICIENCY
00
LULLtCTOR EFFICIENCY
BOILER NUMBER 6
PERCENT REF
************
* *
* *
* PERCENT *
* **
* REFUSE *
* *
* *
************
* *
* *
* *
* *
* *
* *
* 0 *
* *
* *
* *
**** * ** *****
* *
* *
* *
* *
* *
* *
* 20 *
* *
* *
* *
******* *****
* *
* *
* *
* *
* *
* *
* 50 *
* *
* *
>v "^
USE AND LOAC ARE NOMINAL VALUES
*************************************************
PERCENT LOAD
*************************************************
* *
60 * 80 * 100
* *
***************** ********************************
EPA0 * EPA0 * EPA#
* 24 7.76E-01 *
* 29 3.25E-01 *
* 30 8.05E-01 *
* *
* *
* *
* AVE 8 . 02E-01 *
* SO 2.46E-02 *
* CV 3.07E-02 *
**********
*
*
*
**** ******
*
*
*
**** ******
*
*
*
*
*
*
*
*
*
*
***************>* *******************************************
CPAtf * EPA# * EPA#
* 25 8.56E-01 *
* 26 8.25E-01 *
* 27 7.26E-01 *
* *
* *
* *
* AVE 8.02E-01 *
* SD C- . 79E-02 *
* CV 8.46E-02 *
*
*
*
*
*
*
*
*
*
*
***********************************************************
EPA# * EPA# * EPA*
* 22 8.04E-01 *
* 23 7.69E-01 *
* 28 7.85E-01 *
* *
* *
* *
* AVE 7.86E-01 *
* SO 1 . 75E-02 *
* CV 2.23E-02 *
*
*
*
*
*
*
*
*
*
*
***************************** **************************** **************
-------
Regardless of the variation and trend reversals sometimes noted in the
measured particulates before the collector and in the stack, the trends in
collector efficiency are very consistent. Figures 16 and 17 indicate that
the collector efficiency increases for a given load as the RDF heat energy
input increases. For each given percentage of RDF, it is clear that an opti-
mum collector efficiency is reached approximately in the vicinity of 80% steam
load. However, cyclone collector efficiency, for a given cyclone design, is
a function of such factors as particle concentration, gas flow rate, particle
size, and particle density. Both boiler load and percent RDF may affect some
or all of these factors. Future studies are planned to determine the particle
size distribution of the particulates entering the collector which will aid
in the understanding of collector efficiency.
Results of the analysis for the chemical element composition of the par-
ticulates collected on the quartz fiber filters of the EPA Method 5 sampling
train are given in the data matrix format in Appendix F. The chemical element
composition of the very small particulates passing through the filter, but
captured in the sample train impinger water, are also tabulated in the data
matrix format of Appendix F. In Appendix F the listing is alphabetical by
e lament with boiler unit No. 5 data preceding boiler unit No. 6 data. Data for the
before collector particulates, as well as the stack particulates, are included.
The chemical element composition of the particulates is presented by
EPA test sequence in Appendix G for the stack particulates, and Appendix H
for the BCP. Based on the measured flow rates of particulates and the
chemical analysis of the particulates collected on the filter, predicted
effluent rates of the chemical elements have been estimated and tabulated
in these appendices. Such element effluent rates have been estimated based
both on the filter catch and the total particulate catch of filter, cyclone,
and sampling probe.
The chemical element composition of the particulates collected in
the impinger water is presented by EPA test sequence in Appendix I for the
stack particulates and Appendix J for the BCP. Based on the measured flow
rates of particulates and the chemical analysis of the particulates col-
lected in the impinger water, predicted effluent rates of the chemical
elements have been estimated and tabulated in these appendices. Such element
effluent rates have been estimated based on the amount of particulate trapped
in the impinger water.
Particulate Sizing—
Particulate size distributions were determined from the net weight of
particles collected on each stage of the Andersen sampler. The manufacturer's
calibration was used to estimate the size collection for each stage. The
amount of particulate passing through the probe tip but not reaching the
cyclone was weighed as was the amount of particulate collected in the cyclone.
The size distribution of the particles remaining in the probe was estimated
59
-------
from data supplied by Dr. A. K. Rao.— The size distribution of the cyclone
collection was estimated from manufacturer's information. These data was used
to correct the size distribution from that available at the sampler to the
actual distribution in the stack.
This technique provides for an estimate of the size distribution in the
stack, but no real measure of accuracy is known. To determine the accuracy
of the size distribution a complete calibration of the Andersen sampler with
known size distributions and flow rates would have to be made. Since it
would be nearly impossible to simulate the exact (and unknown) size distribu-
tion in the stack, it was decided to use the sampler only to show how the
size distribution changes when using the various amounts of RDF with coal.
Thus, the results of using the Andersen sampler in these tests show the
relative changes in size distribution with RDF compared to coal, but yield
only approximate results for absolute size distributions.
Table 18 is a sample data reduction form and a step-by-step procedure
for the data reduction follows.
1. List the net weights for the probe, the cyclone, and the sampler
stages. Determine the sum of these weights.
2. List the 50% cutoff diameters, D,-Q» for each stage at the flow rate
for the test.
3. Divide the individual net weights by the total weight and multiply
by 100 to yield the uncorrected percent collected, per stage. (For cases
where the probe weight is not available, a modified procedure is used, as
explained later.)
4. Sum the percent collected per stage cumulatively starting with the
backup filter (B) and enter next to the previous stage.
5. Find the weight corrected for the cyclone losses by dividing the
weight collected on each stage by the percent passing the cyclone and enter
next to that stage.
6. To find the weight corrected for the probe losses, divide the weight
corrected for the cyclone for each stage by the percent passing the probe and
enter next to that stage.
7. Calculate the corrected percent collected per stage by dividing the
weight corrected for probe losses by the total weight from step one and multi-
ply by 100.
60
-------
TABLE 18. EXAMPLE SIZING DATA REDUCTION SHEET AND DATA SET
UNIT 5
LOAD 80
RDF 0
StagfT
Weight
collected
gm
b/
microrr^
Uncorrected
collection
%/ stage
Uncorrected
accumulation
7. < D50
Cyclone
(1)
Weight
corrected Probe
for cyclone pass—
gm (%)
Weight
corrected
for probe
gm
Corrected
collection—'
%/stage
Corrected
accumulation
% < D50
Q = 11.02 N l/min
Vms = 220 N I
Probe
Cyclone
0
1
2
3
4
5
6
7
F
Total
0.3338
0.0179
0.0023
0. 0043
0.0177
0.0187
0.0103
0.0107
0.0111
0.0395
0.0714
0.5377
16.8
10.8
6.8
7.8
3.05
1.55
0.94
0.67
0
62.07
3.33
0.43
0.80
3.29
3.48
1.92
1.99
2.06
7.35
13.28
37.93
34.60
34.17
33.37
30.08
26.60
24.68
22.69
20.63
13.28
0
0.3338
5
18
71
88
97
98
98
98
98
0.0460
0.0239
0.0249
0.0213
0.0106
0.0109
0.0113
0.0403
0.0729
10
33
61
76
90
97
> 99
> 99
> 99
0.2478
0.0724
0.0408
0.0280
0.0118
0.0112
0.0114
0.0407
0.0736
46.09
13.46
7.59
5.21
2.19
2.08
2.12
7.57
13.69
10.79 N 4/min
= 213 N 1
53.91
40.45
32.86
27.65
25.46
23.38
21.26
13.69
0
Probe
Cyclone
0
1
2
3
4
5
6
7
F
Total
(0.3075)*
0.0019
0.0039
0.0036
0.0075
0.0146
0.0149
0.0106
0.0107
0.0306
0.0601
(0.4659)
17.0
11.0
7.0
4.93
3.13
1.6
0.96
0.69
0
(66.0)
0.41
0.84
0.77
1 61
3.13
3.20
2.28
2.30
6.57
12.90
34.01
33.60
32.76
31.99
30.38
27.25
24.05
21.77
19.47
12.90
0
5
18
70
89
97
98
98
98
98
0.2376
-
0.0780
0.0200
0.0107
0.0164
0.0154
0.0108
0.0109
0.0312
0.0613
6
30
58
78
90
97
> 99
> 99
> 99
0.2270
0.0667
0.0185
0.0210
0.0171
0.0112
0.0110
0.0315
0.619
48.72
14.32
3.97
4.51
3.67
2.40
2.36
6.76
13.29
51.28
36.96
32.99
28.48
24.81
22.41
20.05
13.29
0
_a/ Andersen stages and plates are not coincident. The probe includes everything ahead of the cyclone. The cyclone
includes everything from the entrance to the cyclone up to and including the "0" plate. Stage "0" corresponds
to plate "I", stage "1" to plate "2", etc., and "B" indicates the back-up filter stage.
_b/ Aerodynamic 50% cutoff diameter (DCQ) from manufacturer's supplied calibration based on flow rate.
_c/ As collected weight percent each stage based on 1007. equals sura of collection for all stages plus the probe and
cyclone*
d/ Percent of DCQ which will pass through the cyclone. From manufacturer supplied data.
_e/ Percent of D^Q which will pass through the probe into the cyclone. From reference.
_f/ Based on 100% equals the sum of the actual weights collected on all stages, the probe, and the cyclone.
* No probe weight. Probe weight set at 66% total.
61
-------
8. Calculate the corrected cumulative percent less than D50 for any
stage by summing the corrected percent per plate cumulatively starting with
the backup filter and entering next to the previous stage.
In cases where the probe weight was unavailable or incorrect, the aver-
age probe weight percent was used to compute a probable probe weight. Parti-
culate collected in the probe may be lost during handling, transportation from
the site to the laboratory, and when rinsing out the particulate. All valid
probe weights were averaged and yielded a value of 66% collection in the probe.
An overall average was used to accommodate all possible trends for individual
load and RDF combinations. The weight of the probe collection was calculated
from this value. The weight of the stage collection plus the cyclone collec-
tion was 34% of the total. The total weight is then the stage collection plus
the cyclone divided by 0.34. The probe collection is 0.66 times this total.
This total based on 66% collection in the probe was used for all further
calculations for these special cases.
Appendix K contains the tabulation of results from the Andersen sampler
for the experimental runs of this study.
Figures 20 and 21 show the average particle size distributions for all
tests performed on boiler unit Nos. 5 and 6, respectively. The horizontal bars
represent variations of one standard deviation on each side of the average
cumulative percent less than D5Q. From these plots, the average size distri-
bution on unit No. 6 at 80% load is shifted toward the larger sizes (lesser
amount of smaller sizes accumulated) than the average size distribution on
unit No. 5 at the combined 60, 80, and 100% loads.
Figure 22 shows the average size distributions for all tests performed
on boiler unit Nos. 5 and 6 at the nominal values of rated steam load of 60
80, and 100%. Each data point on this plot generally represents the average
of three experimental runs. This plot shows the size distribution shifting
on unit No. 5 first to smaller particles as the boiler load increases from 60
to 80%, then shifting to larger particles as the boiler load increases from
80 to 100%. The 100% load size distribution is shifted to the larger sizes
as compared to the 60% load condition. This variation in size distribution
appears to be in concert with the variation in efficiency of the multiclone
particulate collectors. Collector efficiency increases (e.g., recall Figures
16 and 17). At or near the 80% boiler load, more of the larger particles are
separated from the flow. The collector efficiency drops in moving from the
approximate 80% load condition to either the 60 or 100% load condition, thus,
allowing more of the larger particles through the collector. A comparison
of size distributions at the 80% load condition for boiler unit Nos. 5 and
6 is also shown in Figure 22. Boiler unit No. 6 and its collector have a
size distribution containing less of the smaller sizes than boiler unit No. 5.
This may be due to the differences in coal used in the two boilers and it may
be due to slightly different configurations (manufacturer) of the collector
units. ,"
-------
100
50
Particle Sizing Unit 5 Overall
Limits of Horizontal Bars Equal
One Standard Deviation
VI
O
u
if
a
«
N
10
c
—u
1.0
0.5
I
I
I
I
10 15 20 30 40 50 60 70
Cumulative Percent Less Than D«JQ
80
90
Figure 20. Overall cumulative particle size distribution of boiler
unit No. 5 stack particulate emissions.
63
-------
100
50
Particle Sizing Unit 6 Overall
Limit of Horizontal Bars Equal
One Standard Deviation
10
1.0
0.5
_L
J L
_L
-i.
_L
J_
10 15 20 30 40 50 60 70
Corrected Cumulative Percent Less Than 050
80
90
Figure 21. Overall cumulative particle size distribution of
boiler unit No. 6 stack particulate emissions.
64
-------
100,
50
Particle Sizing Unit 5 and 6
All RDF Average
o 60% RDF Load
a 80% RDF Load
^ 100% RDF Load
Open Symbol Unit 5
Solid Symbol Unit 6
2
o
o
o
>o
o
10
1.0
o a
0.5
I
10 15 20 30 40 50 60
Cumulative Percent Less Than 050
70
80
90
Figure 22. Cumulative particle size distribution of boiler
unit Nos. 5 and 6 stack particulate emissions for
differing loads.
65
-------
Figures 23 and 24 show the size distribution changes as a function of RDF
heat energy input to boiler unit Nos. 5 and 6, respectively. For both boiler
unit Nos. 5 and 6 the size distribution shifts to a lesser percent of smaller
samples as the percent RDF increases. The largest effect in this regard is at
a size of 1 u while the smallest effect is at a size of 10 u. Figures 25, 26,
and 27 illustrate the information contained in Figures 23 and 24 as a function
of particle size. The straight lines on this plot represent "best" linear
fits through three points by the method of least squares. At 60% load, the
size distribution increases (lesser percent of small particles accumulated)
with percent RDF. At 807<, load, the shift in size distribution is small and
not as dramatic as the shift at 60% load. At 100% load, the shift occurs in
the opposite direction.
Figure 28 shows the size distribution at 80% load for boiler unit No. 6
as the percent RDF increases. The trend is the same as that noted with boiler
unit No. 5, and indicates the size distribution shifting slightly to the larger
particles as the percent RDF increases.
Figures 29, 30, and 31 show the size distribution shifting to the larger
sizes as the load increases or decreases on either side of the 80% load data
points. At the 07o RDF level, the shift is larger in going from 80 to 100%
than in going from 80 to 60%. The opposite is true at the 50% RDF level
while the shift is about the same at the 20% RDF level.
From the above mentioned data and plots, it is evident that both boiler
load and percent RDF affect the particle size distribution. Increases in load
cause the distributions to increase in particle size at small percentages of
RDF, but to decrease in particle size at large percentages of RDF. Increases
in percent RDF cause the distributions to increase in particle size at low
boiler loads, but to decrease in particle size at large boiler loads. Which
effect predominates, boiler load or percent RDF, is not yet known but is
being studied further.
Orsat—
The Orsat analyzer previously shown in Figure 7 was used to obtain consti-
tuents of C02, 02, and N£ in the stack and before the particulate collector.
The averages of the corrected results of several samples taken during each
experimental run are presented in the data matrix format in Tables 19 through
24. These results were used to determine molecular weight of the stack gas
as well as to estimate excess air supplied to the boiler. Plots of excess
air as a function of percent RDF heat energy input and boiler load have been
previously included as Figures 18 and 19 respectively.
66
-------
100
50
VI
O
O
«o
a
0)
N
10
1.0
Particle Sizing Unit 5
All Loads Average
O 0% RDF
a 20% RDF
A 50% RDF
0.5
2
I
I
I
10 15 20 30 40 50 60 70
Cumulative Percent Less Than 050
80
90
Figure 23. Cumulative particle size distribution of boiler unit
stack particulate emissions for differing percent RDF.
67
-------
100
50
Particle Sizing Unit 6 80% Load
o 0% RDF
a 20% RDF
A 50% RDF
v>
I
O
8
a
.3
in
10
1.0
n o
0.5
I
j L
I
I
L
10 15 20 30 40 50 60 70
Corrected Cumulative Percent Less Than DC
80
90
Figure 24. Cumulative particle size distribution of boiler
unit No. 6 stack particulate emissions for
differing percent RDF.
68
-------
50
o
LO
Q
c
o
U
O_
0)
D
E
D
U
13
QJ
"(5
0)
L_
L-
O
U
40
30
O
Particulate Sizing Unit 5 60% Load
O—10 Microns
O— 7 Microns
D— 5 Microns
A~2.5 Microns
O— 1 Micron
20
10
0
0
10
Figure 25.
20
30
40
50
RDF Energy Input PCT Total
60
70
Corrected cumulative particle size percent less than
DCQ of boiler unit No. 5 stack particulate emissions
as a function of RDF and 60% load.
-------
50 r
Particulate Sizing Unit 5 80% Load
o
LO
a
c
o
U
a.
E
D
u
O
0)
L_
w
o
U
40
30
20
O—10 Microns
O— 7 Microns
D— 5 Microns
A~2.5 Microns
O
O
10
10
Figure 26.
20
30
40
50
60
70
RDF Energy Input PCT Total
Corrected cumulative particle size percent less than
DcjQ of boiler unit No. 5 stack particulate emissions
as a function of RDF and 807. load.
-------
50
o
u-)
Q
c
o
40
30
D
E
D
u
(J
cu
k_
k_
o
U
20
10
0
Participate Sizing Unit 5 100% Load
O —10 Microns
O— 7 Microns
D 5 Microns
A~2.5 Microns
O— 1 Micron
O
I
o
0
Figure 27.
10
20
30
40
50
60
70
RDF Energy Input PCT Total
Corrected cumulative particle size percent less than
boiler unit No. 5 stack particulate emissions as a
of RDF and 100% load.
Q of
function
-------
70 r
60
Jo 50
a
J
40
Particulate Sizing
Unit 6
80% Load
O^ 10 Microns
D~ 7 Microns
O~ 5 Microns
A~ 2.5 Microns
~ 1.0 Microns
10 20 30 40
RDF Heat Input, AVE PCT Total
50
Figure 28. Corrected cumulative particle size percent less than
D5Q of boiler unit No. 6 stack particulate emissions
as a function of RDF and 80% load.
72
-------
Particulate Sizing Unit 5 0%RDF
50
o
"•O
D
c
o
0)
U
a.
a>
40
-o
0)
"o
0)
L_
O
(J
30
20
O
10
O — 10 Microns
O — 7 Microns
D — 5 Microns
A —2.5 Microns
O — 1 Micron
O
10 Microns
7 Microns
5 Microns
2.5 Microns
1 Microns
_L
_L
25
30 35
Steam Load 103 kg/hr
40
45
Figure 29. Corrected cumulative particle size percent less than
050 of boiler unit No. 5 stack particulate emissions
as a function of steam load and 0% RDF.
73
-------
60
Particulate Sizing Unit 5 20% RDF
o
LO
a
c
D
50
40
u
a.
30
_o
3
D
U
£ 20
i_
o
U
O — 10 Microns
O — 7 Microns
D — 5 Microns
A —2.5 Microns
O — 1 Micron
7 Microns
5 Microns
10 Microns
1 Microns
2.5 Microns
10 -
0
_L
25
30 35
Steam Load 103 kg/hr
40
45
Figure 30. Corrected cumulative particle size percent less than
D^Q of boiler unit No. 5 stack particulate emissions
as a function of steam load and 20% RDF.
74
-------
60
U
a.
a>
50
o
10
Q
o 40
-------
TABLE 19. DATA MATRICES OF AVERAGE STACK ORSAT
DATA FOR BOILER UNIT NO. 5
MEASURED OHSAT VALUES XC02 STACK
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARt NOMINAL VALUES
PERCENT
REFUSE
0
20
50
0
20
50
0
20
50
60
EPA* XCU2
04A 6.07F. 00
040 7.37E 00
20 -5.66E 00
21 -3 ,O6E 00
36 7 . 91 E 00
AVE 7.73E 00
3D I . OHc 00
CV 1 .3JT-01
EPA* XC02
Od 6 .56t 00
0*4 A 7 . OOE 00
0913 T . OdE 00
31 7.6BE 00
AVE 7 . oaE oo
SO 4.61E-OI
CV 6.5IF-02
EPA» XC02
01 7 . 20F 00
34 7 . 74E 00
35 7.dlF 00
AVfc r.fjrtE 00
SO 3 . 34E-01
CV 4 . 4 1 F" -02
EPA* XO2
04A H.<*0f_ 00
04B 9 . 5 IE 00
20 1 . 05" 01
21 1 .05E 01
3o 1 .06E 01
AVF 9 . 4 1 F 00
SO } . 4 9 E - 0 1
CV J.rj7E-02
EPA* XU2
Od J.b9C 00
09A :1.'44F 00
OJ3 1 .20L 01
33 1 . 1 IE 01
AVE 1 . 0 3E 01
SO 1 ,r>OC 00
CV 1 . 1351-01
EPA» X02
01 1 . 1 4C 01
34 1 . OhE 01
3D 1 . 0 d '_ 0 1
AVE 1 . 10L 01
SO 4.26C-01
CV l.ddE-02
EPA» 3. 081- 01
3h 1. 15F 01
AVE R.24t 01
SO 2 . OOC 00
CV 2.43C-02
EPA« XN2
0-1 1 . 3 ') E 0 1
OyA i.4ot" 01
O^B a . 09£ 01
31 f). 1 ?C 01
AVE 3 . 6^ 01
50 1 . ftt 00
CV 2. SC-0,?
EP A# N 2
01 d . 4 1 01
34 -1. of. 01
11 3 . 4 C 0 1
AVE 1. 14L 01
SO 1 .47I--01
CV 1 .BOt -03
PERCENT LOAD
SO
EPA* XCG2
05 7.91F 00
10 7.50E 00
17 d.OSe 00
AVE 7.d3E 00
SO 3.02E-01
CV 3.86E-02
EPA* XCO2
Ofr 7 .69C 00
12 8.69E 00
1 J 9 . C 6E 00
AVE d • 4dE 00
SO 7.1 OF- 01
CV a.36b-0?
EPA* XC02
02 8. SHE 00
10 7.81E 00
15 7.66E 00
AVE 6. 12fc 00
SO 6.&5E-01
CV d.!9E-02
EPA* X02
05 a. 44fc 00
16 1 . 18E 01
1 7 1 .08E 01
A VE 1 . 03F 01
SU 1 .73F 00
CV 1 . 68E-0 1
EPA* XO2
00 d . 63E 00
12 9.Q3E 00
1 3 .t>4E-03
FPA* XN2
03 S 8.32E 01
• 1 b H . 19E 0 1
AVE B.26C 01
SO 9.14E-01
C V 1.1 1E-02
76
-------
TABLE 20. DATA MATRICES OF AVERAGE STACK ORSAT
DATA FOR BOILER UNIT NO. 6
•* ASUREO JBiAF VALUES XC02 STACK
BOILER NUMUE« 6
PtRCEST REFUSE AND LJAC AWE NOMINAL VALUES
PEhCEM
SEFUSE
0
zo
50
0
20
50
0
20
so
60
£•**• XC02
e^A* xc.is
ci^A* XC.j2
tPA* %J2
C->A* *J2
EPA* XJ2
tPA* 4N2
tVA* XN2
£PA» XN3
PfcMCENT LOAO
80
EPA» XC02
?4 d.S6E 00
2-J 9.67E 00
30 i . ise 01
AVC 1 .DIE 01
iD 1 .33E 00
CV 1.33F-01
rPAt XCQ2
»•> ».»2F 00
2*- a.ioe oo
27 7. 75E 00
Avr i.lOE 00
SD .' . 16E-01
CV 3. 90C-02
e^** xco2
2*? 9.44C OC
2 J 9.31E 00
28 d.Blt 00
AVC ;>.19t" 00
5D 3.JIE-01
CV 3.61E-02
F°A« XO2
2» y.63E 00
2» 9. COE 00
30 <>.66E OO
A vt a .4 TE oc
S3 l.SbC 00
CV 1 .85E-OJ
EPA« *32
25 1 . 02E 01
26 1 . 08E 01
27 1 . OOE 01
XC02
FP«« XC02
EPA. XC02
EPA« »u?
FPA» X02
FPA« XO2
£PA« XN2
EPA. XN2
EPA> XN2
77
-------
TABLE 21. DATA MATRICES OF AVERAGE ORSAT DATA
BEFORE THE PARTICULATE COLLECTOR
FOR BOILER UNIT NO. 5
MtASURED DRSAT VALUbS XC02 13PC
BOILEH NUMBER 5
PERCENT rtEhUSe ANU LOAD AHC NOMINAL VALUES
REFUSE
0
20
50
0
20
50
0
20
50
60
tPA* XC02
04A 7.20C 00
043 7 . 15E 00
20 9. 13E 00
21 1 . 0811 01
36 8.23t£ 00
AVE 4.51F 00
SU 1 .53E 00
£PA* XC02
09A 7.03E 00
OJB 5.84E 00
33 7 . 35h 00
AVE . 1 IE-01
CV B.81E-02
EPA* XCU2
06 tj. 6 7E 00
12 H.03F 00
1J 8.86E 00
AVE 7. 8t5E 00
SO 1 . 1 IE 00
CV 1.41E-01
EPA« XC02
02 8.51E 0«
10 9. 1 3E 00
15 7.77E 00
AVF d.47f 00
SU 6 . 79E-0 1
CV 8.01F-02
EPA» XU2
05 1 . 04E 0 1
16 1 . l«E 01
17 1 .2 BE 01
AVE I . 15t£ 01
SO 1 . 1 6E 00
CV 1 .OOE-01
tPAtf X02
06 1 .21F 01
12 1 . 0 5E 0 1
13 9 . 3 7E 0 0
AVE 1 .07E 01
SLt ! . 4 OF 00
CV 1 .31E-01
EPA* X02
02 0.49E 00
10 9. 94E 00
15 1 . 12E 01
AVE 1 .02F 01
SO S.74E-01
CV rt . 57E-02
EPA* XN2
05 H.28E 01
1 t> rt . 09E 0 1
17 8.09E 01
AVF H . 1 5t 0 1
SO 1 ,07e 00
CV 1.32E-02
EPA» XN2
06 8. 12F 01
12 8. 14E 01
13 8. 18E 01
AVF d • 1 5K 01
SO 3. OOE-01
CV 3.CPE-03
EPA* XN2
02 8 . 2 Of 01
10 8 • 09E 0 1
15 a. 1 IE 01
AVE 8 . 1 3E 01
SO 5.BOE-01
CV 7.21E-03
100
EPA* XCCJ2
11 1.25E 01
31 1.23E 01
J2 9.60E 00
AVE 1 . 15E 01
SD 1.64E 00
C V I .43fc -01
EPA* XCO2
07 8.85E 00
14 8.60E 00
19 1.02E 01
AVE 9.Z9E 00
SO 8.12E-01
CV b.73t-02
EPA* XC02
1 8 8 .66F 00
AVE 8.66E 00
SD 0. OOE-01
CV 0. OOE-01
EPA* XO2
11 6.00E 00
31 5.56E 00
32 9.49E 00
AVE 7.01E 00
SD 2.15E 00
CV 3.07E-01
EPA* XQ2
07 9.31E 00
14 1 .05E 01
19 8.05E 00
AV E 9.27E 00
SO 1.20F. 00
CV 1 .30E-01
EPA* XO2
18 1 . 1 IE 01
AVF 1 . 1 It 01
SD 0 . OOF -01
CV 0. OOE-01
EPA* XN2
11 H. 15F 01
31 8 . 2 1 E 01
32 8.09E 01
AVE a. 1 5K 01
SO 6.03E-01
C V 7 . 40F-03
EPA* XN2
07 8 . 1 9E 01
14 8.08E 01
19 8. 17E 01
AVE «. 1 4F. 01
SD h.Olfc-01
CV 7.38E-03
FPA* XN2
1H 8.03F 01
AVE 6.03E 01
SD 0 .OOE-01
CV 0. OOE-01
78
-------
TABLE 22. DATA MATRICES OF AVERAGE ORSAT DATA
BEFORE THE PARTICULATE FOR BOILER
UNIT NO. 6
MEASURED 0
BUlLE« NU
PERCENT 0
PERCENT
REFUSE
0
20
50
0
20
50
0
30
50
RSAT VALJES XC03 BPC
8Ea 6
FUit «NO LOAC A*E NOMINAL VALUES
PERCENT LOAD
f 0
EPA» XCU2
cPA« 5.CO?
FJA* XC(;2
EPA. XU2
EPA. XJ2
EPA. x j 2
F p A. XN2
FP A* *N 2
EPA« XN?
80
tPA. XCU2
2« rt.dSE 00
? 9 7 . 5 9F. 00
30 1 . J3E 01
AVc -y.57fc 00
b.T ?.a?E 00
CV 2. 53t-OI
tPA« xcoa
l1* 1.J3F 00
26 9. <;6E 00
? 7 1 . 0»E 01
AUC J.S(.t 00
50 T . a ?f - 0 1
CV 7.7tt-02
EPA* XCO?
22 J. 'iAE 00
23 n. 1 IE 00
2H 8. 1 5E 00
A VE ^. Obt- 00
50 8.^4fc-01
CV 9.86E-02
EPA« XG2
2 A 1 . CAE 0 1
29 9.61t 00
30 5.88E 00
AVE d.63E 00
SO 2.41E 00
Cv 2.80E-01
EPA» X02
25 SI . 8 5E 00
26 1 . 02E 01
27 7.75E 00
AVE $ . 27E 00
So ' .29F 00
C V 1 . 3 5E - 0 1
EPA* XD2
22 a. 7ie oc
23 4.63E 00
2H 1 . C3E 01
AVE 9 . 6Pfc 00
SO 8. 21fc-01
Cv M.546-02
tPA* XN2
2A rt . 08E 0 1
2 '-> 3 . 2 8E 01
30 6 . 1 JE 0 1
AVF B.1BC 01
Su 1 . 0 3t On
Cv 1 . 2*t-02
t PA* XN2
25 * . CBE 01
2*i 3 . C9E 01
27 b. 19E 01
A VL d . 1 2E 01
50 C-.61E-01
CV 6 . 91E-03
tPA* XN2
22 0 . 1 atf 0 1
23 4. 1 IE 01
2-1 8. 15E 01
AVE 3.13E 01
SO 2.42E-01
CV 2.98E-03
100
EPA* XCQ2
tPA» XC02
CPA. XC02
EPA. XD2
EPA« XC2
FPA« XC2
EPA. XN2
EPA* XN2
EPA* XN2
79
-------
TABLE 23. DATA MATRICES OF AVERAGE ORSAT DATA
FROM ALL SAMPLE LOCATIONS ON UNIT
NO. 5
AVERAGE STACK AND 3PC OHSAT VALUES XC02
BOILER NUMBER 5
PERCENT REFUSt AND LOAD AH£ NOMINAL VALUES
PERCENT
REFUSE
0
20
50
0
20
50
0
20
50
60
EPA* XCO?
04A 6.63F 00
04B 7.P6E 00
20 d.99k 00
21 3.74F 00
36 3 . 07F. 00
AVE 8.14E 00
3D 1 .26E 00
CV 1 .55E-01
EPA« XCU2
Od rJ.56h 00
09A 7.02E 00
096 6.4(it 00
33 7.51C 00
Avr 5.a9c oo
SO 4.34C-01
CV 7. 03L-02
bPA* XCO?
01 o.SoE 00
34 7 • 4 1 E 00
35 7 . 77E 00
AVE 7.25E 00
aO 6 . 1 db-01
CV 4.b3E-02
EPA* J402
04A "3.23E 00
04B J.31b 00
20 1 .041: 01
21 '.61600
36 1 . 0 7E 01
AVE ^.d5b 00
SO 'J.64F-01
CV 6.74t-02
EPA* XO?
0* ^.59E 00
0 VA r . OOE 01
09U 1 .27E 01
33 1 . 1 3E 01
AVE 1 . 09E 01
50 1 . 39E 00
CV 1 .2HE-01
EPA* X02
01 1 . 1 5b 01
34 1 . 12E 01
35 1 .12fc 01
AVE 1 • 1 3t- 01
SU 2 • 1 dt-01
C V 1 . 94E-02
EPA* XN2
04A f3.41E 01
04 8 y . 34 E 01
20 rt.OOe 01
21 d . O'jr 01
3f> 5.12E 01
AVE 3 ,?OE 01
SO 1 . 65E 00
CV 2.02E-02
fcPA* XN2
Od ^.j9L 01
09A 4.30E 01
09L) t* . 09b 01
J J A. 12U 01
AVT -1.22E 01
SO 1 . 4 3E 00
CV 1 . 74E-02
tPA» XN2
01 •). 19F 01
34 ^.14e: 01
35 d. 1 IF 01
AVcl J.lt)C 01
50 4 . 1 2E-01
CV 'j.OoE-OJ
PEKCENT LOAD
80
EPA« XCO2
05 7.35E 00
It 7.55E 00
17 7.24E 00
AVE 7.38E 00
SD 1 .55E-01
CV 2.09E-02
EPA* XCO2
00 7. 1HE 00
12 d.36F 00
13 d.96C 00
AVE «.17F 00
SO >.OdE-01
CV 1.1 lt-01
EPA« XC02
02 8.70t 00
10 8.47E 00
15 7 . 72C 00
AVE 8.30C 00
SO 5.13E-01
CV 6 . 1 8E-02
EPA» X02
05 9.446 00
16 1 . 1 fcE 0 1
17 1 .186 0 1
AVE 1 . 09E 0 1
SD 1 ,30E 00
CV 1 .19E-01
EPA« XO2
06 1 .046 01
12 1.02E 01
13 9.&2E 00
AVE 1 . 01B 01
SD •-. . 08F-0 1
CV 4.0CE-0?
EPA» XO2
02 9.40fc 00
10 9.69E 00
15 1 .01E 01
AVE 9.74E 00
SU 3.68E-01
CV 3.78E-02
EPA* «N2
05 3.32E 01
16 8.08E 01
17 8. 10E 01
AVE d.!7E 01
SO 1 .33E 00
CV 1.6.1E-02
tPA* XN2
06 S.24F 01
12 8. 14E 01
13 8. 14E 01
AVE d.lSE 01
SD ^.dSE-01
CV 7.19E-03
EPA* XN2
02 8.19E! 01
10 8 . 18E 0 1
15 H.21E 01
AVe 8.20t: 01
SO 1 .71E-01
CV 2.09E-03
100
EPA* XC02
11 1 . 15F 01
31 1 . 1 IE 01
32 9.78E 00
AVE 1.08E 01
SD 9.03E-01
CV «.37E-02
EPA* XC02
07 9.10F 00
14 9.78E 00
19 1 . 1 Ib 01
AVC 9.«9E 00
SD 1 .OOE 00
CV 1 .01E-01
EPA* XC02
03 S 7.95F 00
IB 1 . 01E 01
AVF 9.01E 00
SD 1.50E 00
CV 1 .66E-0 1
EPA* XO2
11 6.74E 00
31 7.01E 00
32 9.12E 00
AVE 7.1.2E 00
SO 1 .30E 00
CV 1 .71E-01
CPA* XO?
07 B.2JJE 00
14 d.41E 00
19 7. HE 00
AVF 7.92E 00
SD 7.03E-01
CV 6.88E-02
EPA* XU2
03S H.bSE 00
18 8.85E 00
AVE 8.85E 00
SD 1.03E-02
CV 1 . 17E-03
EPA* XN2
11 8. 17F 01
31 8*19b 01
32 8.1 IE 01
AVE 8.16E 01
SO 4.38E-01
CV 5.36E-03
EPA* XN2
07 8.27E 01
14 8.18E 01
19 8. 1SE 01
AVE 8.21b 01
SD 4.98E-01
CV (S.07E-03
EPA* XN2
03S 8.32E 01
18 a. 1 IE 01
AVE 8.21E 01
SD 1.50E 00
CV 1.B3E-02
80
-------
TABLE 24. DATA MATRICES OF AVERAGE ORSAT DATA
FROM ALL SAMPLE LOCATIONS ON UNIT NO.
AVERAGE STACK ANi: UPC CBSAT VALUES XC02
BOILER MJMbER 6
PERCENT REFUSE ANO LJAC ARE NOMINAL VALUES
PERCENT
fiEFUSt
0
20
50
0
20
50
0
20
50
60
EPA» *CG2
EPA* XCG2
FPA* XC02
EPA* XO2
EPA* XU2
EPA* X02
EPA* XN2
£PA* XN2
EPA* %N 2
PERCENT LOAD
ao
EP4* XC02
21 8.90E 00
29 4 . 63E 00
30 1 . 19E 0 1
AVt ^.B\t 00
SO 1 . 81E 00
cv i .ast-o i
EPA* XCD2
25 rt.nat 00
26 B.52E 00
? 7 9. C9E 00
AVt S.B3E 00
SO J.MC-Ol
CV 3 . 30E- 02
EPA* XC02
22 9.6'3E 00
23 9.21E 00
28 a. 51F 00
AVE 9 . 1 3E 00
SO 5.S2E-01
C V 6 . 48E- 02
EPA* X02
21 1 . OOE 01
29 9. 31F 00
30 6.27E 00
AVE H.54E 00
SO 2. OOE 00
CV 2.34E-01
EPA* XO2
25 1 .OOE 01
26 1.05F 01
27 8.90E 00
AVE 9.79F 00
SO _>.02F.-01
CV 8.19E-0?
EPA* t02
22 a.82£ 00
23 9.HOE 00
2H 9.82E 00
AVE 9.48E 00
SO 5.70E-01
CV 6.0 IE -02
EPA* XN2
34 a. 1 IE 01
29 8. 21E 01
30 a . 1 8E 0 1
AVt -t . 1 6E 01
SD 5.27E-01
CV C..45E-03
EPA* XN2
25 d. 1 IE 01
26 8. 10E 01
27 8.20E 01
AVE 3 . I 4E 01
SO 5.45E-01
CV 6.70E-03
EPA* XN2
22 d. 1 5E 01
33 8. 10E 01
28 8. 17E 01
AVE 8 . 1 4E 01
SO 3.54E-01
CV 4.34E-03
100
EPA* XC02
EPA* XC02
EPA* XC02
EPA* XO2
EPA* XO2
EPA* XCJ2
EPA* XN2
EPA* XN2
EPA* XN2
81
-------
Oxides of Nitrogen (NOX)--
The results of these tests are summarized in the data matrix format of
Tables 25 and 26, respectively, for boiler unit Nos. 5 and 6. The grams of
NOX emissions per megajoule of heat energy input to the boiler are plotted
in Figures 32, 33, and 34, respectively, as functions of nominal RDF, nominal
steam load, and excess air of the boiler. Each data point is the average
of runs listed in each cell of the data matrix and usually represents the
average of three experimental runs (replications). From Figure 32 it is
apparent that NOX decreases with increases above 10 to 15% nominal RDF heat
energy input. This is believed to be due to decreased temperatures in the
combustions chamber caused by introduction of the RDF as well as additional
combustion air into the furnace section of the boiler. Figure 34 also shows
that at any given load, the NOX emissions generally decrease as the excess
air increases. This figure also shows that as RDF increases NOX decreases
as already shown in Figure 32. At a particular percentage of RDF the excess
air and NOx emissions generally decrease as the load increases. The only
exception to these trends is a result of the NOX emissions at the 80% load,
07o RDF condition. Statistical analysis discussed in a following section
reveals that the excess air effect predominates. Thus, the decrease in NOX
appears to be due to the additional excess air. Figure 33 indicates that the
NOX emissions generally decrease with increasing load. The trend is most
dramatic for the experimental runs at the 50% nominal value of RDF.
It should also be noted that the NOX emissions from unit No. 6 are always
larger than those from unit No. 5. At 807o nominal steam load, the NO
X
emissions from unit No. 6 are about 1.67 times larger on the average than the
NOX emissions from unit No. 5. The size factor between unit Nos. 5 and 6,
which is about 1.31, may account for part of this difference. However, the
largest factor is currently believed to be due to the different amounts of
excess air available to the boilers through the RDF input lines as well as the
the overfire and underfire air supplies. NOV emissions from power plants
1 0 1 1 /
are affected by excess air. ' Therefore, differences in NOX may be due
to the differences in excess air occurring when RDF is burned. Adjustment
or normalization of the NOX emissions for percent excess air were beyond
the scope of this report.
Oxides of Sulfur (SOX)--
The results of these tests are summarized in the data matrix format of
Tables 27 and 28, respectively, for boiler unit Nos. 5 and 6. The grams of
SOX emission per megajoule of heat energy input to the boilers are plotted
in Figures 35 and 36, respectively, as a function of nominal RDF heat energy
input and nominal steam load of the boilers. Each plotted point of these
figures represents the average of the runs listed in each cell of the data
matrix and usually represents the average of three experimental runs (rep-
lications) .
82
-------
TABLE 25. DATA MATRIX OF NOY EMISSIONS FOR BOILER UNIT NO. 5
NUX FLOW MEASUUtU AT THE STACK
bOILEK NUMoEK b
PEnCENT REFUSE AND EGAD AhiE NOMINAL VALUES
00
*******
*
*
* Pbf-cCE
*
* PbFUS
*
JJC** * * **
*
*
*
*
V
* 0
*
*
***** *>x
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*
*
20
50
NT
E
**
**
*
**
*
*
*
***
*
*
*
*
*
*
*
*
*
*
***
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*****
*****
EPA*
04 A
0 4b
20
2 1
3o
AVE
SO
CV
** ***
E^Afr
0-3
0 'JA
0 3iJ
33
AV E
:>O
i_ V
** * * *
Er> A#
0 1
-34
3 D
AVE
SO
CV
*************************** J^w*******
PLPCc-NT LUAU
It** X< *********
00
.*******************
*
************
60
v***********
(MG/MFGJ)
1 .01 b
1 . Olb
8 . 47E
9 . 3 1 E
1 . 1 6E
02
02
01
01
02
01
1.15E 01
1 . 16E-01
A***********
( MG/MEGJ )
9.80t 01
1 . 1 5E 02
1.03E 02
1.01E 02
1.04E 02
7 . 51E 00
7 .21E-02
*******
(MG/MFGJ)
•r.> . b 2 b 01
M . 5 1 E 01
-) . 3 3 b 01
7.82b 01
1.9bE 01
2 . 4 S> b - 0 1
>}cj)cj)cjjcj)e3(tjp3{t: i(t
*
*
*
*********
* EPA*
* 05
* 1 t>
* 1 7
*
«
*
* AVE
* SO
* CV
** ****** v
* b P A A
* Ob
* 1 2
* 1 3
******** X< * <.
(MG/MEGJ)
1.07b 02
b . Sbt 01
7.37E 01
****** *».*
* 1
.*****>* * .
* EPA*
* 1 1
* 31
****** K*^
00
. * * * * V * ¥ * X<
(MO/ME GJ
o.2t.E 01
9.o1b 01
8.43E 01
7 . 9 9 b 01
2.blE 01
3 . 1 4E-01
****** ^****
(MG/MEGJ)
a.04E 0 1
H.35E 01
b.48E 01
* AV E
* SO
* C V
*•»*******
* bP Aff
* 07
* 1 4
* 1 9
*
8. 1 0 E 01
1 . 70E 01
2.1Ob-01
*********
(MG/MEGJ
7 . 40E 01
t. . 1 1 E 01
{> . 4 2 b 01
* *
>
**
*
*
*
*
*
**
*
*
*
A
*
*
*
*
*
*
AVE
:>D
CV
*******
EPA tt
02
1 0
15
AVE
SO
CV
7
1
1
* *
(
s
b
CJ
6
1
2
.o.b
. OOE
.3 1E-
<<****
M (j / M E
,52b
. 4 b E
. 38b
. 4bi-
. b 7b
* * ^**
0 1
0 1
01
* * f * >
GJ )
0 1
01
01
0 1
0 1
****«
*
*
*
t * 1
*
*
*
*
f
*
A
*
*
*:
1: *1
AV
E
7
. b4b
SO 1 . e>7E
C
* * * *
V
yf *
ePA*
03
1 8
AV
S
C
S
E
I)
V
2
* * *
(
b
4
4
7
1
. 1 8b-
*****
MG/ML
.53E
.43E
. 98E
. 7bE
. bbE-
01
01
0 1
* * * *
GJ )
01
01
01
00
01
*
* * * *
B
*
** **•
) *
*
*
*
*
*
*
*
*
* * **
) *
*
*
*
*
If
*
t
*
*
**
*
*
*
*
*
*
*
Jf
V
-------
00
-P-
TABLE 26. DATA MATRIX OF NOY EMISSIONS FOR BOILER UNIT NO. 6
NOX FLU* MEASURED AT THE STACK
BCILER NUMBER 6
PERCENT REFUSE AND LUAO APE NOMINAL VALUES
:****************** x< ************************************************
* * *
* * PERCENT LOAD *
* PERCENT * *
* *************************************************************
* REFUSE * * * *
* * 60 * SO * 100 *
X< * * * *
***********************************************************************
*
*
*
*
*
*
* 0
*
*
*
***** **
*
*
*
*
*
*
* 20
*
*
*
***** * *
*
*
*
*
*
*
f 50
#
*
*
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
* ****
*
*
*
*
*
*
*
*
*
*
EPA* (MG/MEGJ) *
*
*
*
*
*
*
*
*
*
***** **************
EPA# (MG/MEGJ) ¥
*
*
*
*
*
*
*
*
*
*******************
EPA# ( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
EPA*
24
29
30
A VE
SO
CV
( MG/MEGJ )
1.2 BE 02
1 .34E 02
1 .39E 02
1 . 33E 02
5 . b IE 00
4 . 20E-02
******************
EPA#
25
27
A VE
SO
CV
******
F PA4
22
23
28
AVF
SO
CV
( MG/MEGJ )
1 . 27E 02
1 .34E 02
1 . 31E 02
4 . 8 IE 00
3.68E-02
* EPA# (MG/MEGJ )
SK
*
*
*
*
*
*
*
*
********************
* EPA* ( MG/MEGJ )
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
******************************#***
( MG/MEGJ )
1 . 07E 02
1 . 07E 02
1 . 03E 02
1 . OfcE 02
2.72E 00
2.56E-02
* EPA* ( MG/MEGJ )
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
***********************************************************************
-------
O
O
0.15
0.13
0.12
0.11
0.10
0.09
^f-*
Z 0.08
0.07
0.06 -
0.05>-
O 60% Load
a 80% Load
* 100% Load
• 80% Load
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
10
20 30 40
ACTUAL RDF HEAT, PERCENT
50
60
Figure 32. NOX emissions from boiler unit Nos. 5 and 6
as a function of RDF heat input.
85
-------
o.isr
0.13 -
3
n
• oO% RDF
jj. n 20 % RDF
f * 50 % RDF
Unit Symbol
5 Open
6 Shaded
— Confidence Interval at
95% Confidence Level
60 80
ACTUAL LOAD, PERCENT STEAM OUTPUT
100
Figure 33. NOX emissions from boiler unit Nos. 5 and 6
as a function of boiler steam load.
86
-------
oo
O
~"
•o
o
I-
O
LU
x
0.14
0.12
0.10
°'08
0.06
0.04
60
•
•
\
*
--0.
_L
J_
_L
80
Figure 34.
100 120 140
EXCESS AIR, PERCENT
160
O 0% RDF
D 20% RDF
A 50% RDF
60% Load
80% Load
100% Load
Open Symbol Unit 5
Solid Symbol Unit 6
180
NOX emissions from boiler unit Nos. 5 and 6
as a function of excess air.
200
-------
00
00
TABLE 27. DATA MATRIX OF SOY EMISSIONS FOR BOILER UNIT NO. 5
SUX FLOW MEASURED AT THE STACK
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
***********************************************************************
*
*
* PEKCE
#
* REFUS
*
*
***** **
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
*
*
NT *
**
E *
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
** ***
EPA#
04A
04-B
20
36
AVE
SD
CV
*****
EPA#
08
0 9 A
09B
33
AVE
SO
CV
*****
EPA#
01
34
35
AVE
SD
CV
PERCciNT LOAD
********************************
*
t>O * 80
*
*** * ********
( MG/MEGJ )
3.36E 02
5.29E 02
1 . 99E 03
2.36E 03
1 .JOE 03
1 .02E 03
7.32E-01
******* *****
( MG/MEGJ )
2 .2 IE 03
2 . 98E 03
1 . 88E 03
2.23E 03
2.34F 03
4.59E 02
1 . S.oE-01
*** *********
( MG/MCGJ )
2 . 93E 02
9.27E 02
1 . 06E 03
7.62E 02
4 . 12E O2
5.41E-01
***
*
*
*
*
*
*
*
*
*
*
*****
EPA*
05
16
1 7
AVt
su
CV
** ******
*
*
*
*
*
*
*
*
*
*
EPA*
06
12
1 3
AV,_
SO
CV
********
*
*
*
*
*
*
*
*
*
*
EPA*
02
10
15
AVE
SO
CV
********
( MG/ME
2. 33E
2. 25E
2.28E
2 .29E
3.86E
1 . 69E-
* * **
GJ )
03
03
03
03
01
02
************
( MG/MEGJ )
1 . 69t
1 . B3E
2 . 07E
1 . 86E
1 . 94E
1 . 04E-
**** ****
< MG/ME
1 . 5bE
1 . 47E
1 . 47E
1 . 51d
£>.27E
4. 16E-
03
03
03
03
02
01
* ** *
GJ )
03
03
03
03
01
02
****
*
*
*
****************
100
****** **
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
EPA*
1 1
AVE
SD
C V
**** ********
{ MG/MEGJ )
2.39E 03
2.39E 03
0 .OOE-01
0. OOE-01
** *i.*K**********
EPA#
07
14
19
AVE
SD
CV
****
EPA*
03^
1 8
AVE
SD
CV
( MG/MEGJ )
2.20F 03
1 .65E 03
2.08E 03
1 . 98E 03
2.90E 02
1 .47E-01
************
( MG/MEGJ )
1 . 97E 03
1 . 50E 03
1 . 74E 03
J.27E 02
1 .896-01
*
*
*
**
*
*
*
**
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
***********************************************************************
-------
TABLE 28. DATA MATRIX OF SO EMISSIONS FOR BOILER UNIT NO. 6
SUX FLO* MEASURED AT THE STACK
HOILER NUMbFR 6
PERCENT REFUSE: AND LOAD ARE NOMINAL
VALUES
00
*
*
* PLPCE NT
* REFUSE
*
*
*
*
*
*
*
+
* 0
*
*
*
*
*
*
*
*
* 20
*
*
*
*
*
*
*
*
*
* 50
*
*
*
***************
* PFPCEN
*
+ A A ^i ft & &
v ^ *f *^ T f
*
*
*
* L PA*
*
*
*
*
*
*
*
*
*
* (- PA*
*
* EPA*
*
*
*
*
*
*
*
*
*
*•
C,0 *
*
( MG/MEGJ ) *
*
*
*
*
*
*
+
*
*
(MG/MEGJ) *
*
*
*
*
*
*
*
*
(MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
bPA*
24
29
30
A VE
SO
CV
E PA*
25
27
A VE
SO
C V
L PA*
2-3
2H
A vr
su
C V
80
(
1
4
4
7
5
*********************,
T LOAD
M G/Ml
. 40E
.67E
. .-iBE
. 69F
.49E
7. 1 4E-
( MG/NU
I
3
J
?
^
(
/
1
4
4
. O'jE
.47E
. 26C
. ^7t
. 1 1E-
MG/ME
. 22E
. 4t>E
. 34E
. 07E
. J6E-
-------
o
•o
o
O
l/l
O
Z)
u_
_j
Z)
2.8
2.4
2.0
1.6
1.2
0.8
0.4
_L
o 60% Load
D 80% Load
" 100% Load
• 80% Load
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
I
I
10
20 30 40
ACTUAL RDF HEAT, PCT
50
60
Figure 35. Sulfur emissions from boiler unit Nos. 5 and 6
as a function of RDF heat input.
90
-------
2.8
oo
LU
_l
ID
o
o
z
o
00
1/1
i
UJ
ID
2.4
2.0
1.6
1.2
0.8
0.4
0
Confidence Interval at
95% Confidence Level
0~ 0% RDF
o~ 20% RDF
^~ 50% RDF
Unit Symbol
5 Open
6 Shaded
60 70 80 90 100
ACTUAL LOAD. PCT RATED STEAM OUTPUT
Figure 36. Sulfur emissions from boiler unit Nos.
5 and 6 as a function of boiler steam load.
91
-------
From Figure 35 it appears that the general trend in sulfur dioxide
emissions is to decrease with an increased amount of nominal RDF heat energy
input to the boilers. The only exception to this is a single point at 60%
nominal load and 20% nominal RDF heat energy input. No reason is known for
this trend reversal, but it may be a peculiar operating condition of the
boiler. However, from this figure one can also conclude that as load increases
the sulfur dioxide emissions also increase.
It should be noted that the emissions of unit No. 5 are higher than
those of unit No. 6. This is because the coal selected for unit No. 6 was
a 50/50% mixture of Iowa (high sulfur) coal and Wyoming (low sulfur) coal.
This coal mixture has a smaller content of sulfur than the Iowa coal used
for the tests of unit No. 5. Iowa coal was selected for the tests of unit
No. 5 mainly because of its high sulfur content. This was done so that
the effect on sulfur emissions, due to burning RDF, which is low in sulfur
content, could more readily be observed.
With the exception of the 60% load/20% RDF data point, Figure 36 shows
that the sulfur dioxide emissions increase with increasing load as one
would normally expect. This figure also shows that the sulfur dioxide emis-
sions consistently decrease with corresponding increases in RDF heat energy
input, as would be expected. Thus, mixing RDF with coals of relatively
high sulfur content may allow some sulfur dioxide standards to be met while
still using the high sulfur coal.
Aldehydes and Ketones--
The results of these tests are summarized in the data matrix format of
Tables 29 and 30, respectively, for boiler unit Nos. 5 and 6. The milli-
grams of formaldehyde emission per megajoule of heat energy input to the
boilers are plotted in Figures 37 and 38, respectively, as a function of
nominal RDF heat energy input and nominal steam load of the boilers. Each
plotted point on these figures represents the average of three experimental
runs.
Figure 37 shows that the formaldehyde emissions generally increase with
increases in the RDF heat energy input to the boiler. This is consistent with
the fact that as the amount of RDF increases the material containing cellu-
lose fiber, such as wood chips and paper, increases, thus providing the
additional formaldehyde emissions. The only exception to this generalization
is the data for 100% boiler load on unit No. 5. The data at 100% boiler
load and 50% RDF are believed to be unreliable because of the difficulties
in operating the boiler at this condition.
92
-------
TABLE 29. DATA MATRIX OF FORMALDEHYDE EMISSIONS FROM BOILER UNIT NO. 5
ALDEHYDES AND KEYTONES FLOW MEASURED AT THE STACK
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
***********************************************************************
* * *
* * PEK'CENT LOAD *
* PERCENT * *
* ******************************************************* *** ***
* REFUS
*
*
*******
*
*
*
1=
*
*
* 0
*
4
*
*******
*
*
*
*
*
*
* 20
*
*
#
*******
*
*
*
*
*
*
* 50
*
*
*
E *
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
ik
#
*
*
*****
EPA*
04A
04B
20
21
36
AVE
SO
CV
*****
EPAW
Od
0 ->A
0-JD
33
AVE
SO
CV
* * *** **** *
*
*
*
*
*
#
*
*
*
*
EPA*
01
34
35
AVE
SO
CV
60
************
( MG/MEGJ )
2.50E 00
3.24E 00
2 .26fc-01
4 .24E-01
6.92E 00
2.ooE 00
2 .71 E 00
1 .026 00
************
( MG/MEGJ )
7.42E 00
3 .60E-02
4 . 04 E 00
1 . 02E 00
3 . 1 3E 00
3.33E 00
1 . 06F 00
************
(MG/MEGJ >
1 . 1 6L~ 00
1 .55E 01
2.01E 00
6.22E 00
•3 .05E 00
1 .29E 00
*
*
*
**
*
*
*
*
*
*
*
*
*
*
******
EPA#
05
16
17
AVE
SO
CV
********
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
EPA#
06
12
1 3
AVE
SU
CV
******
EPA*
02
1 0
Ib
AVE
SO
CV
80
************
( MG/MEGJ )
5. 75E-01
1 . 28E-0 1
2 .22C-02
2.4PE-01
2 . 93E-01
1 . 2 It 00
***** *******
{ MG/MEGJ )
3 . 2 bE - 0 1
1 . 96t-01
1 . 3 1 E - 0 1
2. 17E-0 1
•J. 87E- 02
4 . 54E- 01
************
( MG/MEbJ )
8 . b 7E 00
3.36E 00
3.H4E-01
<4 . 2 7b 00
4. 1 IE 00
<•>. C.2E- 01
*
*
*
*=***
*
A
V
A
*
A
*
*
A
*
****
*
*
¥
j!c
*
*
*
*
*
*
* ***
*
*
*
*
*
A
*
*
it
JK
100
*
*
*
******************
EPA*
1 1
31
32
AVE
SO
C V
******
EPA#
07
14
Avr
so
C V
(MG/MEGJ )
2.57E 00
1 .69E-01
7.21b 00
3.32E 00
3.58E 00
1 . 08E 00
*********
( MG/MEGJ )
3.93E 00
3. 20E-02
1 . 98E 00
2 . 7bE 00
1 .39E 00
***************
tP A#
03 S
1 8
AVE
SO
C V
( MG/MEGJ )
7. 80E-02
2. 37E-01
1 . b7E-01
1 . 12fc-01
7. 14E-01
*
*
*
*
*
»
*
*
*
*
***
*
*
*
*
*
*
x-.
*
*
*
* **-
*
*
*
*
*
4
*
*
*
*
************************************************************************
-------
TABLE 30. DATA MATRIX OF FORMALDEHYDE EMISSIONS FROM BOILER UNIT NO 6
ALDEHYDES AND KEYTONLS FLO* MEASURED AT THE STACK
fJOILER NUMBER 6
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
******* ************v************************X< **************************
* * *
* * PERCENT LOAD *
* PERCENT * *
* ******************************************************* ******
* REFUSE * * * *
* * 60 * 80 * 100 *
* * * * *
***********************************************************************
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
* **** **
*
*
*
*
*
*
* 50
*
*
*
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
(MG/MEGJ) *
*
*
*
*
*
*
*
*
*
**************
(MG/MEGJ) *
*
*
*
*
*
*
*
*
*
EPA*
24
29
30
AVE
SD
CV
******
EPA*
25
2b
27
AVE
SD
CV
********************
( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
EPA*
22
23
28
AVE
SO
CV
( MG/MEGo )
1 .62E 01
5. 36E-0 1
0. 50F-01
5.80E 00
9.02E 00
1.56E 00
************
( MG/MEGJ )
7.45E-01
5.41E-01
5. 17E-01
6.01E-01
1 .25E-01
2. 08E-01
**** ********
( MG/MEGJ )
1 . 71 E OU
1 .44E 00
1 . b2E 00
1 .56t 00
1 . 42E-01
9 . ObE-02
*
*
*
*
*
*
*
*
*
*
***
*
*
*
*
*
*
*
*
*
*
***
*
*
*
*
*
*
*
*
*
*
EPA* (MG/MEGJ) *
*
*
*
*
*
*
*
*
*
*******************
EPA* (MG/MEGJ) *
*
*
*
*
*
*
*
*
*
*******************
EPA* ( MG/ME (jj) *
*
*
*
*
*
*
*
*
*
**********************************************************************$
-------
7.Or
O 60% Load
0 80% Load
A 100% Load
Open Symbol Unit 5
Solid Symbol Unit 6
20 30 40
RDF HEAT INPUT, PERCENT
Figure 37- Formaldehyde emissions from boiler unit Nos.
5 and 6 as a function of RDF heat input.
95
-------
7.0
LO
OO
LU
Q
O
u_
6.0
^ 5.0
O
—>
t>
&
E_ 4.0
oo
z
O
3.0
2.0
1.0
50
I
I
I
O 0% RDF
D 20% RDF
A 50% RDF
Open Symbol Unit 5
Solid Symbol Unit 6
60 70 80 90
LOAD, PERCENT RATED OUTPUT
100
Figure 38. Formaldehyde emissions from boiler unit Nos. 5
and 6 as a function of boiler steam load.
96
-------
Figure 38 shows that formaldehyde emissions first decrease with load,
then increase after passing through a minimum near the 807=, boiler load.
From other observations and calculations boiler unit No. 5 seems to have its
best operation near the 80%, boiler load. The boiler operators have the
most experience in firing this unit near this load. Thus it is believed
that the entire system is operating in a more efficient manner at this level
of load which, in turn, helps decrease the formaldehyde emissions. Combus-
tion chamber temperatures were not measured during these experimental runs
so correlations of emissions with such temperatures are not obtainable.
Organic Acids, Cyanide, and Phosphate--
Results from the organic acid analysis are not yet available due to
difficulties in analysis techniques. Results from the analysis of the
chemicals in the organic acid train have yielded both cyanide and phosphate.
The cyanide emissions are tabulated in the data matrix format of Tables
31 and 32, respectively, for boiler unit Nos. 5 and 6. Figures 39 and 40
show the grams of cyanide per megajoule of heat energy input to the boilers
plotted, respectively, as a function of percent RDF heat energy input to the
boiler and boiler load.
From Figure 39 it is clear that the cyanide emissions increase with
increases in RDF for boiler unit No. 5. The trend at 80% load for boiler
unit No. 6 is somewhat reversed from that for boiler unit No. 5. Figure 40
indicates a decrease in cyanide emissions with an increase in boiler load.
The phosphate emissions are tabulated in the data matrix format of
Tables 33 and 34, respectively, for boiler unit Nos. 5 and 6. Figures 41
and 42 show the grams of phosphate per megajoule of heat energy input to
the boilers plotted, respectively, as a function of percent RDF heat energy
input to the boiler and boiler load.
From Figure 41 it appears that phosphate emissions generally increase
with increases in RDF. However, an exception to the general trend is the
80% load data on boiler unit No. 5. Figure 42 indicates relatively small
increases in phosphate emissions as the boiler load increases. Thus, RDF
appears to have a larger effect on phosphate emissions than do boiler loads.
Chlorides--
The chloride emissions have been measured by using appropriate analysis
of the sulfur dioxide sampling train and the organic acid sampling train.
Tables 35 and 36 are tabulations of the results from the sulfur dioxide
sampling train for boiler unit Nos. 5 and 6. Tables 37 and 38 are tabulations
of the results from the organic acids sampling train for boiler unit Nos.
5 and 6.
97
-------
TABLE 31. DATA MATRIX OF CYANIDE EMISSIONS FROM BOILER UNIT NO. 5
CYANIDE EMISSIONS, MICRCGRAMS PER MEGAJOULE
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
00
******* *****
* *
* *
* PERCENT *
* **
* REFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*** ** **
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
E *
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*****
*****
EPA*
04A
04H
20
21
36
AVE
SD
CV
*****
EPA*
08
09A
09B
J3
AVE
SD
CV
*****
EPA*
01
34
35
AVE
SD
CV
******************************************************
*
PERCENT LOAD *
*
******************************************************
60
*
*
*
80
*
*
*
************************************
( UG/MEGJ )
2.05E 02
2.09E 02
1 . 39E 02
1 . S5E 02
1 .82K 01
1 .53E 02
8.06E 01
5. 26E-01
*** *********
{ UG/MEGJ )
^.68E 01
?.15E 02
1.60E 02
2.82E 02
1 .86E 02
8.29E 01
4.46E-01
*
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
EPA*
05
16
1 7
AVE
SD
CV
(UG/MEGJ >
1 . 1 8E 02
3.65E 02
1 .68E 02
2. 17E 02
1 .31E 02
6.02F-01
*
*
*
*
*
*
*
*
*
*
**********************
EPA*
06
12
13
AVE
SD
CV
(UG/MEGJ )
8.29E 01
2.07E 02
2.17E 02
1 .69E 02
7.47E 01
4.42E-01
********************************
( UG/MEGJ )
1.88E 02
3.22E 02
4.26E 02
3 . 12E 02
1 .20E 02
3.83E-01
*
*
*
*
*
*
*
*
*
*
EPA*
02
1 0
1 5
AVE
SD
CV
(UG/MEGJ )
2.12E 02
9.64E 01
2.57E 02
1.89E 02
8.29E 01
4.40E-01
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
1
00
*
*
*
******************
EPA*
1 1
31
32
AVE
SD
CV
*****
EPA*
07
14
AVE
SD
CV
(UG/MEGJ
1.0 IE 02
3.09E 01
1 .50E 02
9.42E 01
6.00E 01
6.37E-01
) *
*
*
*
*
*
*
*
*
*
*************
(UG/MEGJ
1 .90E 02
1 . C8E 02
1.49E 02
5.80E 01
3.90E-01
**************
EPA*
03 S
1 8
AVE
SO
CV
(UG/MEGJ
1 . 77E 02
1.67E 02
1 . 72E 02
6.58E 00
3.83E-02
) *
*
*
*
*
*
*
*
*
*
****
) *
*
*
*
*
*
*
*
*
*
***********************************************************************
-------
TABLE 32. DATA MATRIX OF CYANIDE EMISSIONS FROM BOILER UNIT NO. 6
CYAN
BOIL
PERC
*** * *
*
*
* PER
*
* REF
*
*
**** *
*
*
*
*
*
*
* 0
*
*
*
* ****
*
*
*
*
*
*
* 20
*
*
*
*
*
*
*
*
*
* 50
*
*
*
**** *
IDE EM I SS t ON
ER NUMBER 6
EN T REFUSE A
************
*
S, MICRCGRAMS
PER ME
KD LOAD ARE NOMINAL
**************
******
GAJOULE
VALUES
************
** **
* PERCENT LOAD
CENT *
*******
USE *
*
*
************
* EPA*
*
*
*
*
*
*
*
*
*
************
* EPA*
*
*
*
*
*
*
*
*
*
* E PA u
*
*
*
*
*
*
*
*
*
************
**************
*
bO *
*
**************
( UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
**************
( UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
( UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
**************
******
******
EPA*
24
29
30
AVE
SD
CV
******
EPA*
25
26
27
AVE
SD
CV
EPA*
22
23
28
AVE
SD
CV
******
************
80
************
(UG/MEGJ )
1 . 71E 01
1 . lyE 02
1 . 53E 02
9 . 6 1 E 01
7. C6E 01
7 . 35E-0 1
************
(UG/MEGJ )
2.62E 02
1 . 92E 02
1 . 95F 02
2 . 16E 02
3. <53E 01
1 . 32E-0 1
(UG/MEGJ >
1 . 1 5E 02
2 .2 IE 02
1 . 44E 02
1 .60E 02
5.46F 01
3. 42F-0 1
************
****
*
*
*
* ***
*
*
*
*
*
*
*
*
*
*
* * **
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
** **
******************
*
*
*
******************
*
1 00 *
*
******************
EHA* (UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
******************
EPA* ( UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
EPA* ( UG/MEGJ ) *
*
*
*
*
*
*
*
*
*
******************
-------
360 •-
320
280
-o
i
O
O
L/l
o
Q
z
>
u
200
120
80
40
o 0% RDF
a 20% RDF
50% RDF
Confidence Interval at
95% Confidence Level
o1-
I
I
10 20 30 40 50
ACTUAL RDF HEAT INPUT, PERCENT
60
Figure 39. Cyanide emissions from boiler unit Nos.
5 and 6 as a function of RDF heat input.
100
-------
360
320
280
- 240
x
•o, 200
\
O
n 160
Q
Z 120
u
80
40 -
Confidence Interval at
95% Confidence Level
I
I
I
50
Figure 40.
o 0% RDF
a 20% RDF
50% RDF
Symbol
Open
Shaded
60 70 80
STEAM LOAD, PCT RATED OUTPUT
90
100
Cyanide emissions from boiler unit Nos.
5 and 6 as a function of boiler steam load,
101
-------
TABLE 33. DATA MATRIX OF PHOSPHATE EMISSIONS FROM BOILER UNIT NO. 5
PO4 I-LQW MEASURED AT THE STACK
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARt NOMINAL VALUES
***********************************************************************
* * *
* * PERCENT LOAD *
* PERCENT * *
* *************************************************************
* REFUSE * * * *
* * 60 * 60 * 100 *
* * * * *
*********************************************** J|:****V ******************
*
*
*
*
*
*
* 0
*
*
*
*
*
*
*
*
*
*
*
*
*
EPA#
04A
04B
20
21
36
AVE
SD
CV
******* **********
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
EPA*
08
09A
09B
33
AVE
SD
CV
** ** *
EPAtf
01
34
35
AVE
SO
CV
( MG/MEGJ
8.20E-02
3 . 40E-02
0 .OOE-01
0 . OOE-01
0. OOE-01
3 .32E-02
4.55E-02
1 .37E 00
) *
*
*
*
*
*
*
*
*
*
EPAtf
Ob
1 6
17
AVE
SD
CV
********************
( MG/MEGJ
2 . 17E-01
1 . 72E-01
4 . 90E-02
1 .90E-01
1 .57E-01
7.43E-02
4 .73E-01
**********
( MG/MEGJ
1 .40E 00
0. OOE-01
0 . OOE-01
4 .66E-01
3.07E-01
1 .73E 00
) *
*
*
*
*
*
*
*
*
*
EPA#
06
12
1 3
AVE
SD
CV
{ MG/MEGJ )
1 .06E-01
0 .OOE-01
1 . 99E 00
t> .97E-01
1 . 12E 00
1 .60E 00
******* *****
{ MG/MEGJ >
2 . 49E-01
0. OOE-01
0 .OOL-0 1
8 . 30E-02
1 . 4 4E - 0 1
1.73E 00
***#**:****** *****X<****
) *
&
*
*
*
*
*
*
*
*
EPA#
02
10
15
AVE
SD
CV
( MG/MEGJ )
0 .OOE-01
3.90E-02
0 .OOE-01
1 .30E-02
2.25E-02
1.73E 00
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
*
*
*
****
*
*
*
*
*
*
*
V
*
*
EPA*
1 1
31
32
AVE
SD
CV
( MG/MEGJ
3.00E-02
1 . 9OE-01
1 . 10E-01
1 . 10E-01
8 . OOE-02
7.27E-01
**************
EPA#
07
14
19
AVE
SD
CV
(MG/MEGJ
1 .52E-01
0 .OOE-01
0. OOE-01
5.07E-02
rt .7BE-02
1 .73E 00
) *
*
*
*
*
*
*
*
*
*
****
) *
*
*
*
*
*
*
*
*
*
******************
EPAtf
03S
1 b
AVE
SD
CV
( MG/MEGJ
1 .90E-01
1 .77E-01
1 .83E-01
9. 19E-03
5. Olfc-02
) *
*
*
*
*
*
*
*
*
*
******* ************* ***************************************************
-------
TABLE 34. DATA MATRIX OF PHOSPHATE EMISSIONS FROM BOILER UNIT NO. 6
PU4 FLO* MEASURED AT THE STACK
BOILtR NUMBER 6
PERCENT REFUSE AND LOAD ARE NOMINAL
V ALUr. S
V ** * JK **
*
*
* P E R L. E
*
* KEFUS
*
*
*******
*
*
*
*
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
**********
*
*
NT *
* * ** jf * *
E *
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
**********
* EPA*
*
*
*
*
*
*
*
*
*
***************
***************
*
60 *
*
***************
( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
***************
(, MG/MEGJ) *
*
*
*
*
*
*
*
*
*
***************
(MG/MEGJ) *
*
*
*
*
*
*
*
*
*
*****
PE^:
*****
*****
EPA*
24
29
30
AVE
SO
CV
# jfr Jf # *;
EPA*
2b
26
27
AVF
SO
CV
*****
EPA*
22
2_>
2o
AVE
SO
CV
*vtt*******
CENT LOAD
**********
80
**********
( MG/MEGJ
0 . OOE-0 1
0 . 06C- 0 1
0 . OOE-0 1
0 . OOt -0 1
0. OOE-01
1 . OOE 20
X^^^JvC^jfcj^JjciJt^:
( M G/MEG J
0. OOE-01
0 . OOE-01
0 . OOE-0 1
0 . OOE-0 1
0 . OOE-0 1
1 . OOE 20
**********
( Mb/MEGJ
? . 95E-01
1 . 23t 00
O . OOE-0 1
0 . 74E-0 1
h . 2 2 E - 0 1
9. 23E-01
L********* ************
*********************
*
* 100
*
) * EPA* ( MG/MEGJ )
*
*
*
*
ik
*
*
*
*
,
) * tPA* ( MG/ME GJ )
*
*
*
*
*
*
*
«.
*
*********************,
) * EPA* ( MG/MEGJ )
*
*
*
*
*
*
*
*
*
***
*
*
***
*
*
^
* * *
*
^
*
*
*
*
*
*
*
*
*
*
*
*
*
^
*
*
*
c **
*
*
*
*
*
*
*
*
*
-------
1.4r
O 60% Load
Q 80% Load
A 100% Load
Open Symbol Unit 5
Solid Symbol Unit 6
10
20 30 40
RDF HEAT INPUT, PERCENT
50
60
Figure 41. Phosphate emissions from boiler unit Nos.
as a function of RDF heat input.
5 and 6
104
-------
1.4
1/1
LU
O
O
E
o
oo
CO
CO
o
1.2
1.0
0.8
0.6
0.4
0.2
O 0% RDF
D 20% RDF
A 50% RDF
Open Symbol Unit 5
Solid Symbol Unit 6
oi 1 -
50 60 70 80 90
LOAD, PERCENT RATED OUTPUT
Figure 42. Phosphate emissions from boiler unit Nos. 5 and 6
as a function of boiler steam load.
100
105
-------
TABLE 35. DATA MATRIX OF CHLORIDE EMISSIONS FROM BOILER UNIT NO. 5 AS
DETERMINED BY THE SO^ SAMPLING TRAIN
J4_
CHLORINE FLO* USING SOX METHOD MEASURED AT
BCILER NUMBER 5
PERCENT REFUSt AND LOAD AtxE NOMINAL VALUES
THE STACK
**** ***
* MERGE
*
* HEFUS
*
* * * * * * *
*
*
*
*
*
*
* 0
*
*** ** **
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
A
* 50
*
*
*
*******
*****
*
*
NT *
**
E *
*
*
** ***
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
************************ * ********************
PERCENT LOAD
*****************
60
*****************
EPA* (MG/MEGJ)
04A 4.OOE 01
046 1.80E 01
20 1.50E 01
36 1.60E 01
AVE 2.23E 01
SD 1 . 1 91£ 01
CV 5. 35E-01
*****************
EPA* (MG/MEGJ)
08 4.OOE 01
09A /'.OOE 01
093 7.50E 01
33 4 . 70E 01
AVE 5.30E 01
S D 1 . 7 1 E 01
CV 2.95E-01
*****************
EPA* (MG/MEGJ)
01 3.90E 01
34 1.17E 02
35 1.44E 02
1.OOE 02
SD 5.45E 01
CV 5.45E-01
*****************
*****:*******
*
* 80
*
*****:£** * * Jjc*
* EPA* (M
* 05 1 .
* 16 1 .
* 1 7 1 .
*
*
*
* AVE 1.
* SD 2.
* CV 2 .
************
* EPA* (M
* 06 6 .
* 12 7 .
* 13 7.
*
*
*
* AVE 6.
* SD 7.
* CV 1 .
************
* EPA* (M
* 02 5.
* 1 0 1 .
* Ib 1 .
*
*
*
* AVE 9.
* SD 4.
* CV 4 .
************
************
*
*
*
************
G/MEGJ ) *
10E 01 *
1OE 01 *
60E 01 *
*
*
*
27E 01 *
d9E 00 *
28E-01 *
**** ********
G/MEGJ ) *
OOE Ol *
OOE 01 *
50E 01 *
*
*
*
83E 01 *
64E 00 *
12E-01 *
******** ****
G/MEGJ ) *
1OE 01 *
37E 02 *
03E 02 *
*
*
*
70E 01 *
33E 01 *
47E-01 *
************
** **
** **
EPA*
1 1
AVE
SD
CV
****
EPA*
07
1 4
1 9
AVE
SD
CV
****
EPA*
03S
18
AVE
SD
CV
** **
**************
*
*
*
**************
*
100 *
*
**************
(MG/MEGJ) *
7.OOE 00 *
*
*
*
*
*
7.OOE 00 *
O.OOE-01 *
0.OOE-01 *
**** **********
( MG/MEGJ ) *
3.90E 01 *
6.40E 01 *
8.30E 01 *
*
*
*
6.20E 01 *
2.21E 01 *
3.56E-01 *
**************
( MG/MEGJ ) *
1.16E 02 *
8.60E 01 *
*
*
*
*
1 . 01E 02 *
2.12E 01 *
2.1OE-01 *
**************
-------
TABLE 36. DATA MATRIX OF CHLORIDE EMISSIONS FROM BOILER UNIT NO. 6 AS
DETERMINED BY THE SO,, SAMPLING TRAIN
CHLOKINE FLO* U5ING SJX METHOD MtAbUKED AT THt STACK
GOILER NUMBER b
PERCENT REFUSE AND LOAD AhJE NOMINAL VALUtS
*******
*********^*************v********X<************ **********]
*
*
PERCENT LOAD
*
* H 6 FUSE
*
*
*** * * ** **
*
*
*
*
*
*
* 0
*
*
#
*********
*
*
*
*
*
*
* 20
*
*
*
*********
*
*
*
*
*
*
* 50
*
*
*
*******
*
*
*
a*******
* r yj A iu
*
*
*
*
*
*
*
*
*
*** *****
* EPA *
*
*
*
*
*
*
*
*
*
*** ** ***
* EPA *
*
*
*
*
*
*
*
*
*
**************
*
to *
*
**************
( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
**************
( MG/MEGJ ) *
*
*
*
V
*
*
*
*
*
**************
( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
******
******
EPA*
24
2 y
30
AVE
SO
CV
******
EPA*
25
27
AVE
SD
CV
******
EPA*
23
28
AVE
SU
CV
************
80
**** ******* V
( MG/MEGJ >
1 . OOfc 01
5.0 Ob 00
4.00E 00
6.33E 00
3.216 00
5. 08E-01
*** *********
( MG/Mb GJ )
4.70E 01
4 . OOF 01
4. 35E 01
4 . 95E 00
1 . 1 4E-01
*** * * ** * * * * *
( MG/MEuJ )
1 . 1 1 E 02
b ,40b 01
3. 7bP 01
3. 32E 01
3 . 806-01
****
*
*
*
**x< *
*
*
*
*
*
If
*
*
*
*
***x=
*
*
*
*
*c
*
*
*
*
*
** * *
*
*
*
*
Jfc
Me
*
*
*
*
*
*
*
******** **********
*
100 *
** ****************
EPA* (MG/MEGJ) *
*
*
*
*
*
*
******************
6PA* {MG/M6GJ) *
*
*
*
*
*
*
*
*•
******************
EPA* (MG/MEGJ} *
*
*
*
*
*
*
*
***********************************************************************
-------
O
00
TABLE 37. DATA MATRIX OF CHLORIDE EMISSIONS FROM BOILER UNIT NO. 5 AS
DETERMINED BY THE ORGANIC ACIDS TRAIN
CHLORINE FLOW MEASURED AT THE STACK
USING CHLORIDES AND ORGANIC ACIDS METHOD
BOILER NUMBER 5
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
***** ************ ***********************# ******************************
* *
* *
* PERCENT *
* **
* REFUSE *
* *
*******
*
*
*
A.
*
*
* 0
*
*
*
*******
*
*
*
*
*
*
* 20
*
*
*
*******
*
*
*
*
*
*
* 50
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
PEP CENT LOAD
***************** **3JC***** Jf********V**
60 * 80
*
*****
EPA#
04A
04B
20
21
36
AVE
3D
CV
*****
EPA#
08
09A
093
33
AVE
SO
CV
************
( MG/MEGJ )
7.73E 00
1 .47E 01
2.56E 01
2.41E 01
2.24E 01
1 . 89E 01
7.50E 00
3.97E-01
************
( MG/MEGJ)
8. 16E 01
7.26E 01
4.88E 01
7.32E 01
6. 91E 01
1 .41E 01
2 . O4E-01
*****************
EPA#
01
34
35
AVE
SD
CV
( MG/MEGJ )
1 . 16E 02
5.93E 01
1 .51E 02
1 . 09E 02
4.62E 01
4 . 25E-01
**
*
*
*
*
*
*
*
#
*
*
******
EPA#
05
16
1 7
AVE
SO
CV
********
«
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
*
*
*
EPA*
06
12
13
AVE
3D
CV
****
*
*
*
*
*
*
******************
*
100 *
*
******************************
( MG/MEGJ )
1 . 83E 01
o . 91E 00
5. 12E 00
I . 0 1 E 01
7 . 1 6E 00
7.08E-01
*
*
*
*
*
*
*
*
*
*
****************
(MG/MEGJ )
1 .21E 02
1 . 18E 02
1 . 17E 02
1 . 1 8E 02
2.22E 00
1 .d7E-02
******************
EPA*
02
1 0
15
AVE
3D
CV
(MG/MEGJ )
7.70E 01
9 .33E 01
9.36E 01
8.80E 01
9.54E 00
1 . 08E-01
*
*
»
*
*
*
*
*
*
*
EPA#
1 1
31
32
AVE
SD
CV
{ MG/MEGJ
3. 75E 01
7.5OE 00
1 . 15E 01
1.88E 01
1.63E 01
8.66E-01
****
) *
*
*
*
*
*
*
*
*
*
******************
EP A ft
O7
14
AVF
SD
CV
****** ***
*
*
*
*
*
*
*
*
*
*
EPA*
03 S
18
AVE
SD
CV
{ MG/MEGJ
0.96fc 01
9.51E 01
8.23E 01
1 .60E 01
2. 19E-01
******* **
(MG/MhGJ
5.39E 01
1 . 06E 02
7.96E 01
3.66E 01
4.59E-01
) *
*
*
*
*
*
*
*
*
*
****
) *
*
*
*
*
*
*
*
*
*
***************** *******************************************v**jjc*jj;*4: + ;(:^
-------
TABLE 38. DATA MATRIX OF CHLORIDE EMISSIONS FROM BOILER UNIT NO. 6 AS
DETERMINED BY THE ORGANIC ACIDS TRAIN
CHLORINE FLO* MEASURED AT THE STACK
USING CHLORIDES AND ORGANIC ACIDS METHOD
BCILER NUMBER 6
PERCENT REFUSE AND LOAD ARE NOMINAL VALUES
*****************************************************************
* * *
* * PERCENT LOAD *
* PERCENT * *
* ******* *****ic**************w**,K*c**¥***** *********************
* REFUSE * * * *
* * 60 * 80 * 1 00 »
* * * * «
***********************************************************************
*
*
*
*
*
*
* 0
*
*
*
*** ****
*
*
*
*
*
*
* 20
*
*
*
***** **
*
*
*
*
*
*
* 50
*
*
*
*
*
*
*
*
*
*
*
*
*
*****
*
*
*
*
*
*
*
*
*
*
* ****
*
*
*
*
*
*
*
*
*
*
EPA# ( MG/MEGJ ) *
*
*
*
*
*
*
*
*
*
***** **************
tPA# (MG/MEGJ) *
*
*
*
*
*
*
*
*
*
*******************
EPA* ( MG/MFGJ ) *
*
*
*
*
*
*
*
*
*
EPA*
24
29
30
AVE
SO
CV
******
EPA*
2b
26
27
AVE
SO
CV
******
EPA*
22
23
28
AVE
SD
CV
(MG/MEGJ )
2.32E 01
3. OOF 00
1.0 5E 01
1 . 39E 01
d . 1 SE 00
5 .86E-01
**** ******K
( MG/Mt E 0 1
4 . 43E-0 1
* EPA*
*
*
*
*
*
*
*
*
*
*********
* EPA*
*
*
*
*
*
*
*
*
*
*********
* EPA*
*
*
*
*
*
*
*
*
*
( MG/MEGJ ) *
*
*
*
*
*
*
*
w
*
**** **** ******
( MG/MEGJ ) *
*
*
L
*
*
&
*
*
*
**************
(MG/MEGJ) *
*
*
V
*
*
B
-
*
*
*****************************9*****«***********1t:************]I(*******tt«'«-
-------
Figures 43 and 44 give the chloride emissions from the sulfur dioxide sampling
train as a function of nominal RDF and nominal load, respectively. Figures
45 and 46 give the chloride emissions from the organic acids sampling train
as a function of nominal RDF and nominal load respectively.
Figures 43 and 45 show the chloride emissions to increase significantly
with increases in RDF heat energy input to the boiler. Figures 44 and 46
show relatively small changes in chloride emissions with load. The chloride
analysis from the two different sampling trains are in good agreement con-
cerning variation with RDF and variation with load. Thus, the chloride
emissions appear to be mainly related to the RDF input to the boilers.
Mercury, Arsenic, Antimony, and Beryllium Under Effluent Flow Rates Under
Chlorides—
The results from the mercury sampling trains are not yet ready for
presentation. However, analysis of the mercury train impinger chemicals
for the elements has been accomplished. This data is tabulated in Appendix
L in the data matrix format and alphabetical by element. The tabulation
includes analysis of impingers containing iodine monochloride (ICl), nitric
acid (HNO-j), and potassium hydroxide (KOH).
Appendices M, N, and 0 are tabulations of the ICl analysis, the HNC>3
analysis, and the KOH analysis, respectively, by EPA test sequence. These
appendices predict the effluents of the elements from the stack based on
the flow rate measurements and the chemical element analysis. The predicted
effluents of chemical elements from this analysis may be different from the
previous predictions from the particulate filters and particulate water
because of the sampling rates, and impinger chemicals used in the various
sampling trains.
Hydrocarbons--
No significant hydrocarbon emissions (>0.25 ppm) in the C, to Cc range
have been found in the stack emissions.
Organics--
Results of sampling for organics including polychlorinated biphenyls
(PCB) and polynuclear aromatics are summarized below.
PGBs—The usual procedure for the recovery and determination of PCB
from various sample materials is being followed. This procedure involves
extraction with a suitable solvent, cleanup of the extract using column
chromatography and detection, and determination using electron capture gas
chromatography (ECGC). With this procedure, no PCBs could be detected in
the coal samples, in vapor samples collected from the stack on macroreticu-
lar resin, in particulates from the dust collector or in particulates from
the stack.
110
-------
O.I6f-
0.14
0.12
o o.io
o
o
O
.
O
I
u
0.08
0.06
0.04
0.02 -
L I
o 0 % RDF
a 20 % RDF
a 50 % RDF
Unit Symbol
5 Open
6 Shaded
10
Confidence Interval at
95% Confidence Level
20 30 40
ACTUAL RDF HEAT, PERCENT
50
60
Figure 43. Chloride emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input as determined by the SOX
train.
Ill
-------
1/1
a
O
0.14
0.12
0.10
0.08
Q
£ 0.06
O
x
u
0.04
0.02
o 0 % RDF
a 20 % RDF
A 50 % RDF
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
I
I
60 70 80 90
ACTUAL LOAD, PERCENT STEAM OUTPUT
100
Figure 44. Chloride emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load as determined by the S0}
train.
112
-------
o
-°o
5
O
O
"6
0.16
0. 14
0.12
0.10
Z
^ 0.08
0.06
Q
a:
O
5 0.04
0.02 -
O1-
O ~ 60% Load
a ~ 80% Load
- ~ 100% Load
• ~ 80% Load
Unit Symbol
5 Open
6 Shaded
10
Confidence interval at
95% Confidence Level
1
20 30 40
ACTUAL RDF HEAT INPUT, PCT
50
60
Figure 45. Chloride emissions from boiler unit Nos. 5 and 6 as a
function of RDF heat input as determined by the organic
acids train.
113
-------
O
z
s
0.16 r-
0.14 -
0.12
0.10
0.08
O
od
U 0.06
U 0.04
0.02
o~ 0%RDF
a~ 20% RDF
*~ 50% RDF
Unit Symbol
5 Open
6 Shaded
Confidence Interval at
95% Confidence Level
50
60 70 80 90
ACTUAL LOAD, PCT STEAM OUTPUT
100
Figure 46. Chloride emissions from boiler unit Nos. 5 and 6 as a
function of boiler steam load as determined by the
organic acids train.
114
-------
The detection limit for this method is approximately 10 ppb. The limit of
the method can be increased to approximately 10 ppt by the perchlorination of
the PCBs. Negative results were also obtained for the various samples
using this perchlorination procedure. A recovery study was run on a dust
collector sample deliberately spiked with PCBs at the 1 ppb level. A 95%
recovery was obtained from the spiked sample.
In the case of the solid waste fuel, the initial experiment on per-
chlorination of the extract obtained by Soxhlet extracting a quantity of
ground processed refuse was positive. That is, a large peak was obtained
for decachlorobiphenyl. This peak could be caused by degradation of some
component of the refuse by the SbCl5, used as the perchlorinating reagent
to give biphenyl which in turn is chlorinated to the decachlorobiphenyl.
This possibility was investigated by following the standard, but longer
procedure for isolation and determination of PCBs. This involves extraction
cleanup, and separation using the procedure of Armos and Burke,-^ The
retention times of the major peaks of the extracts from the refuse matched
the major peaks of the arachlox 1254 and 1248 standards. The approximated
concentration of PCBs in the solid waste was 0.5 ppm.
Polynuclear aromatics (PNA)—
Gas phase results--Twenty-eight gas phase samples ranging in volume
from 40 to 1,500 liters were taken from the power plant stack and subjected to
separation, isolation, detection, and quantitation schemes. From 0.1 to 43
fim/liters were measured with no correlation to firing conditions or the amount
of solid waste being burned. Quantified compounds found in the stack emissions
are given in Table 39. Additional identified constituents were phenol, o-cresol
tolueme, diphenylamine, di-(2 -ethylhexyl)-phthalate, and triphenylphosphate.
Of this group the common flame retardent, triphenylphosphate, is the only
constituent of immediate interest and could be caused by either treated coal
or a portion of the refuse. Ambient air samples were also collected and
analyzed. The amount of organic material found was shown to be a function
of vehicular traffic rather than stack emissions.
Particulate results—Most of the particulate samples were collected
from inside the stack at the power plant. Twenty-four samples ranging in
amounts from 0.03 to 110 g were collected. Complete analysis of these samples
showed that the organic content was highly variable and complex. The dominant
constituents were saturated hydrocarbons with lesser amounts of unknown polar
materials. The upper limit for PNA content was 6.8 ng/g. This designation
of total PNA is based solely on the classification scheme and relative gas
chromatography (GC) retention times. Insufficient material was available for
confirmation using the available gas chromatography/mass spectrometry (GC/MS)
instrumentation. An upper limit of 0.3 ng/g for perylene has been established
using the identification scheme explained above.
115
-------
TABLE 39. ORGANIC COMPOUNDS IN STACK EMISSIONS
Compound
Naphthalene
Acenaphthalene
Fluorene
Anthracene
Fluoranthene
Pyrene
Benzof luorenes (1,2 and 2,3)
1,2 Benzanthracene
a and e Benzpyrenes and Perylene
20 Methylcholanthrene
Dibenzanthracenes (1,2-3,4 and A, H)
Dibenzanthracene (2,3-6,7)
Coronene and 3,4-9,10 Dibenzopyrene
Aliphatic hydrocarbons
Detection Limits
Stack Gases
(Hg/l,000 m3)
BDL-/
BDL
36.5
BDL
119
36.5
54.7
BDL
72.9
BDL
BDL
BDL
BDL
31,700
35 ng/1,000 m3
Particula tes
(ng/g)
BDL
BDL
BDL
BDL
BDL
BDL
BDL
BDL
0.42
BDL
BDL
BDL
BDL
340
0.35 ng/g
a/ BDL = Below Detection Limit.
116
-------
Impinger water results—Sixteen samples of impinger water collected
in a standard EPA Method 5 sampling train have been assayed. The organic con-
tent of these waters are in the mg/liter range and specific contaminants are
xylene, tetramethylbenzene, diethylphthalate, dimethylphenol, phthalic
anhydride, t-butylphenol, t-butylcresol, and di-t-butylethylphenol. A
series of silicones were identified in three water samples but these were
apparently artifacts caused by contamination from one of several connectors
used in the sampling train. Consistency in the contamination amounts has
not yet been observed and no definitive significance has been placed on
these impinger water results.
Miscellaneous--
Elemental sulfur in coal and particulates—Elemental sulfur was determined
by combining electron capture (EC) detection with cyclohexane extractions of
coal and particulates. The sensitivity for sulfur permitted its determination
in the samples at subparts per billion levels. The extraction procedures
allowed for a minimum of cleanup prior to the gas chromatography (GC) which is
both rapid and selective.
Particulate samples were collected from the 3-in. sampling ports located
approximately halfway up the stack of the unit No. 5 boiler. Three types of
samples were collected. Particulates No. 1 were from the accumulation in the
ports. Particulates No. 2 were collected by drawing the vapor from inside the
stack through a glass tube containing a glass wool plug. Particulates No. 3
represented that portion which settled onto horizontal trays placed inside the
stack.
Ten grams of particulates were Soxhlet extracted for 24 hr in 25 x 85 mm
glass thimbles using approximately 50 ml of cyclohexane. The cyclohexane
was transferred to volumetric flasks and diluted to 50 ml with cyclohexane.
Five micoliters aliquots of this solution were gas chromatographed without
further cleanup.
Elemental sulfur content of particulates--
Sample
No.
No.
No.
1
2
3
No.
Samples
3
2
10
Range
fcg/R )
150-250
65-87
4-15
Average
fcg/g )
198
76
9
The variation in the results obtained for the three types of sampling
suggests some correlation with particulate size. Because of the large
117
-------
amount of sample required (10 g) a determination as a fraction of size
has not been accomplished.
Four samples of Iowa coal, one of Wyoming coal, and one of Illinois
coal were analyzed for elemental sulfur. The coal was crushed and 1-g samples
which had passed a 60-mesh sieve were Soxhlet extracted for 24 hr. Large
35 x 90-mm glass thimbles were used to prevent plugging of the Soxhlet device
by the fines from the coal samples. The 90 ml of cyclohexane used for the
extraction was then quantitatively transferred to volumetric flasks and
diluted to 100 ml. Five-micoliters aliquots of this cyclohexane solution was
gas chromatographed directly without further cleanup.
Elemental sulfur content of coals--
Coal Source ug/g
Iowa Sample No. 1 1,000
Iowa Sample No. 2 2,080
Iowa Sample No. 3 280
Iowa Sample No. 4 205
Wyoming 160
Illinois 398
Comparison of low and high temperature ashing of solid waste from the
Ames solid waste recycling plant--
Sample preparation—A solid waste sample was dried overnight at
105°C to provide partial sterilization and then passed through a Wiley mill
with a 2-mm screen. The resulting material was a medium grey colored, very
lightweight mixture of fibers. Other small, lighter colored particles were
fairly evenly distributed throughout.
Several grams were oxidized using the International Plasma Corporation
plasma ashing apparatus. Contrary to original expectations, this was a
lengthy process which required about 10 days to obtain constant weight.
The ash was found to be 15.92% of the dried, milled starting material. Small
pieces of glass and stone were present in the ash.
An additional 10 g of the sample were ashed at 800°C for 1 hr. The
ash was found to be 11.23% of the starting material. Small pieces of glass
and other noncombustibles were also apparent in this ash.
X-ray fluorescence analysis—The two ash samples were analyzed
using an energy dispersive X-ray fluorescence analyzer. A molybdenum target
X-ray tube was used to 12 kV to excite low atomic weight elements and at
118
-------
30 kV to excite heavier elements. The ashed samples were placed in a chemplex
holder with a thin mylar film supporting the ash. The samples were not
compressed. Equal volumes of the two samples were used. The major difference
in the 12 kV spectra of the two samples was the absence of chlorine in the
sample ashed at 800°C. At 30 kV the main difference in the two spectra was
the absence of bromine in the sample ashed at 800°C. Spectral interferences
from Br and Pb interfere with the detection of Ga, As, and Se. The ratio
of incoherent to coherent scattered Mo radiation was found to be higher in
the plasma ashed sample, which indicates a higher concentration of low atomic
weight elements (such as H, C, N, 0). This observation is in agreement with
the larger amount of ash in the plasma ashed sample. Even though the ashing
had been continued for 10 days to reach constant weight, it appears that
some organic material remained unoxidized.
Numerical comparisons—The integrated background corrected counts
for many of the spectral features are presented in Table 40. The ratios of
counts for the two ash samples are also given. Due to the variable nature
of the sample, the ratios differ considerably from sample to sample. However,
ratios significantly less than 1.00 probably indicate loss of that element
when the refuse is ashed at 800°C. It should be noted that there could also
have been losses when the ashing was performed in the plasma asher. Elements
which 'lowed low ratios were S and K. It is clear that Cl and Br were also
lost when ashing at 800°C. Due to spectral interferences from Br and Pb, it
was not possible to determine whether Ga, As, and Se were found. It would
appear, however, that these elements were not present in significant amounts
(compared to the amount of lead, for instance). Germanium was not detected
in either sample.
Semiquantitative analysis—There was not enough ash from the plasma
ashed sample for a semiquantitative analysis. The high temperature ashed
sample was compressed into a planchet and the X-ray spectra taken, along
with a standard sample of SRM 1633 trace elements in coal ash. The results
of this analysis are included in Table 40.
Energy dispersive X-ray fluorescence was used for the elemental analysis
of the coal, bottom ash, collector ash, solid waste, and stack particulates.
A molybdenum target X-ray tube was used at 12 kV for low atomic weight elements
and at 30 kV for heavier elements. The detection limits in ppm by weight
are listed in Table 7. The detection limited for ash samples applies for the
bottom ash, collector ash, and solid waste.
119
-------
TABLE 40. COMPARISON OF HIGH AND LOW TEMPERATURE ASHING OF SOLID WASTE
12 kV Al
Si
S
K
Ca
Ti
Cr
Mn
Fe
30 Kv K
Ca
Ti
V
Cr
Mn
Fe
Ni
Cu
Zn
Ga
Ge
As
Se
Rb
Sr
Pb
Low temp, ash
Counts
3,605
25,341
35,710
83,036
666,744
174,100
5,908
35,789
726,925
12,133
105,453
34,414
6,383
1,937
15,691
431,129
2,131
11,114
228,404
10,455
816
2,407
1,758
8,189
92,717
267,867
High temp, ash
Counts
4,592
28,787
27,477
72,192
793,353
215,125
6,307
43,230
754,532
10,688
125,107
39,932
7,437
2,403
18,213
464,745
2,038
12,135
242,250
10,632
753
0
1,717
7,930
107,213
275,405
Concentration
~7.87o
15.1%
3.2%
1.8%
12.8%
1.58%
240 ppm
2,000 ppm
3.14%
1.8%
12.8%
1.58%
< 1,100 ppm Ti interference
240 ppm
2,000 ppm
3.14%
83 ppm
420 ppm
5,450 ppm
< 170 ppm Pb interference
< 16 ppm Zn interference
< 20 ppm Pb 7 and Br interference
< 10 ppm Pb interference
37 ppm
423 ppm
4,460 ppm
High temp.
Pntio
Low temp.
1.27
1.14
0.77
0.87
1.19
1.24
1.07
1.21
1.04
0.88
1.19
1.16
1.24
1.16
1.08
0.96
1.09
1.06
0.97
1.16
1.03
-------
REFERENCES
1. Standards of Performance for New Stationary Sources. Federal Register,
36(247), Part 11, December 23, 1971.
2. Standards of Performance for New Stationary Sources. Amendments to
Reference Methods. Federal Register, 36(111), Part 11, June 8, 1976.
3. Standards of Performance for New Stationary Sources. Federal Register,
38(66), Part 11, April 6, 1973.
4. Carotti, A. A., and E. R. Kaiser. J. for the Air Pollution Control
Association. 22(249), 1972.
5. Shannon, L. J., M. P. Schrag, F. I. Honea, and D. Bendersky. St. Louis/
Union Electric Refuse Firing Demonstration Air Pollution Test Report.
EPA-650/2-74-073, Control Systems Laboratory, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, August 1974.
6. Fassel, V. A., and R. N. Kniseley. Anal. Chem. Vol. 46, 1110A4, 1155A,
1974.
7. American Society for Testing and Materials, Method D1608.
8. Neker, M. B., and P. W. Jones. In Situ Decomposition Product Isolated
from Tenax-GC While Sampling Stack Gases. Anal. Chem. 49(3), March
1977.
9. Rao, A. K. Sampling Line Losses for the Andersen Cascade Impactor.
Midwest Research Institute, Kansas City, Missouri, October 8, 1975.
10. Gaffe, S. J., and R. W. Serstle, "Emissions from Coal Fired Power Plants:
A Comprehensive Summary." U.S. Department of Health, Education, and
Welfare, Public Health Service Publication No. 999-AP-35, 1967.
11. Selker, A. P. Program for Reduction of NOX from Togential Coal Fired
Boilers, Phase II. EPA report prepared by Combustion Engineering, Inc.
Windsor, Connecticut, EPA 650/2-73-005-a. June 1975.
12. J.A.O.A.C. 53, 761, 1970.
121
-------
APPENDIX A
STATISTICAL ANALYSIS OF VARIANCE OF RESULTS
The analysis of the variability in the results with respect to the con-
trolled parameters of boiler load and percent RDF in the fuel is given in
this section. Standard statistical techniques have been used.JJi' In order
for the reader to obtain an understanding of this section, a few paragraphs
of background, including terminology and explanation, are felt to be essential.
BACKGROUND
Terminology
Variables in experimenter controls are called factors in an experimental
design. The number of forms or categories of a factor appearing in an experi-
ment is termed the levels of that factor. A particular combination with one
level from each factor is called a treatment. If all possible treatments, or
a definite portion of them is of interest, the experiment is called a factorial
experiment. It is a complete factorial if all possible treatments are included.
A factor can be quantitative (e.g., different temperatures) or qualita-
tive (e.g , different methods of testing). There is usually no natural order
established among different levels of a qualitative factor while such order
does exist with levels of a quantitative factor.
In a two factor arrangement, if factor A has n levels, factor B has m
levels, there are nm treatments. If repeated measurements (or observations)
are made for each treatment there is said to be replication.
If the levels of all quantitative or qualitative factors in an experiment
are set at fixed values, one has a so called "fixed model" (or Model I) experi-
ment. This study is a fixed model experiment as all of the levels of each
factor are specified as indicated in the following paragraph.
122
-------
Factors and Levels of Factors for This Study
For this study, it was determined that two major factors could be con-
trolled at various levels. These factors were the load based on steam flow
and the amount of RDF based on heat energy input to the boiler. The levels
of these factors were chosen to be 60, 80, and 100% nominal load, and 0, 20
and 50% RDF. To obtain sufficient data for statistical analysis, a factorial
experimental design with three replications was devised for each boiler as
summarized previously in Table 3. Thus, for boiler unit No. 5, the statisti-
cal design is a 3 x 3 x 3 full factorial experiment with 27 runs necessary to
fill the data matrix of this experiment. In addition, testing of two differ-
ent size (and design) traveling grate stoker fired boilers (unit Nos. 5 and
6) was accomplished at one load setting (8070) to obtain a relative size com-
parison for all emission data at a given fixed load. Certainly there are
many other secondary and uncontrolled factors that might enter into this
study. In order to minimize and possibly eliminate the effect of these sec-
ondary and uncontrolled factors it isrusually necessary to perform what is
called randomization.
Randomization
Randomization is an important requirement of a properly designed experi-
ment so that an unbiased estimate of the error can be obtained. Any unmea-
sured, unknown, or undesirable influence upon the variation of the responses
may contribute to erroneous conclusions related to a set of data. When such
influences follow a specific pattern, they are referred to as bias in the
data. Biased influences may be caused by uncontrolled differences in time,
environment, procedure, etc. These biases can be removed from an experiment
if they are recognized during the planning phase and proper procedures taken
to control them. Often, however, it is difficult to recognize such biases or,
even if they are recognized, to control such biases. Thus, an alternative is
to randomize the experiment in such a way that the order of taking data (re-
sponses) is predetermined so that each possible response is given an equal
opportunity for occurrence. Randomization was not completely possible in
these experiments because weather conditions, availability of refuse, and
availability of a given boiler unit dictated the order in which tests were
scheduled and conducted. However, there was some randomness in each of the
above constraints, and this randomness was therefore transferred to the test
sequence.
Statistical Model of This Study
In order to test the significance of the different factors controlled in
these experiments, established statistical methods were used. For purposes
of efficiency and avoidance of incorrect conclusions, such methods were felt
to be essential for this study. The technique employed was to test the factors
known or assumed to control the variability of a certain measurement (such as
123
-------
SOX emissions) according to a prescribed model. The variability of each fac-
tor was determined in relation to the variability caused by measurement error
(replication). If the factor variability was significantly larger than the
measurement-error variability, the factor had a profound effect on the total
variability of the measured item. In this way, factors significantly affect-
ing the measured variables can be identified, isolated, and analyzed in greater
detail. Inherent to this type of testing is the so called "Analysis of Vari-
ance" procedures^- for fixed level experimental designs. Such procedures
depend on the desired factors with known and stated levels being tested rela-
tive to the measured variable (the response). For example, if boiler load
and percent RDF in the fuel are tested relative to SOX emissions, one would
have a two factor experiment. One factor would be boiler load with three
fixed levels (60, 80, and 100% of rated load), and the other factor would
be percent RDF in the fuel with three fixed levels (0, 20, and 50% of fuel
heat energy input). For these factors and associated levels of the factors,
the data of this study were analyzed according to the following statistical
model:
Yijk = p, + Li + Rj + (LR)ij + Eijk
where Yijk = measured variable (SOX or NOX for example),
p- = true value of the variable being measured with no effect
due to any of the following treatments,
Li = effect or treatment caused by the percent load,
Rj = effect or treatment caused by the percent RDF,
(LR)ij = interaction effect of percent load and percent RDF, and
Eijk = measurement error or effect for which one cannot
account.
A 95% confidence level was selected for stating conclusions in view of
the number of data and the measurement error. A 95% confidence level means
that one would be stating conclusions with 95% probability or 20 to 1 odds
of having the correct conclusion. Pertinent conclusions at the 95% confidence
level are contained in the following sections.
124
-------
TABULATION OF ANALYSIS OF VARIANCE (ANOV) RESULTS
The ANOV of the data was performed in terms of both emission flow rates
(kg/hr) and normalized emission rates (kg/MJ) based on heat energy input. The
normalized emission rates were found to be inappropriate in the ANOV proce-
dures because the normalizing parameter (heat energy input) was directly relat-
ed to both of the controlled parameters in the experiment. This rendered the
ANOV procedures ineffective. For this reason, only the ANOV results using
emission flow rates (kg/hr) are meaningful. Hence, they are presented and
discussed in the following sections of this report. Tables A-l, A-2, A-3,
and A-4 summarize the results of the ANOV procedures.
Element flow rates (kg/hr) in the coal, RDF, grate ash, and hopper ash
were affected by the boiler load of unit No. 5 as shown in the following table.
A positive (+) sign indicates a statistically significant (95% confidence
level) increase in the element flow rate with increase in boiler load. A
negative (-) sign indicates a statistically significant (95% confidence
level) decrease in the element flow rate with increase in boiler load. A
blank ( ) represents either no significant effect, or the inability to detect
or resolve a significant effect at the 95% confidence level. No analysis for
boiler unit No. 6 can be included in this tabulation since only one boiler load
(80%) was used for the experiments on this boiler.
Element flow rates (kg/hr) in the coal, RDF, grate ash, and hopper ash
were affected by the percent RDF in the fuel input to boiler Nos. 5 and 6 as
shown in Table A-2. As before, a (+) indicates a statistically significant
(95% confidence level) increase in the element flow rate with increase in
RDF. A (-) indicates a statistically significant (95% confidence level)
decrease in the element flow rate with increase in RDF. A ( ) represents
either no significant effect, or the inability to detect or resolve a signifi-
cant effect at the 95% confidence level.
COAL ELEMENT FLOW RATES
Analysis of the variability of the element flow rates in the coal show
the following:
1. As boiler load increases, the flow rate of most elements in the
coal, shown in Table A-l, also increases.
2. As the percent RDF increases for a given boiler load, the flow
rate of most elements in the coal shown in Table A-2 decreases. These conclu-
sions are valid for the data from both boiler unit Nos-. 5 and 6.
125
-------
TABLE A-l. ANOV OF ELEMENTS IN COAL, RDF, GRATE ASH, AND HOPPER ASH
WITH RESPECT TO BOILER LOAD FOR GIVEN LEVELS OF RDF INPUT
Boiler Collecter—
Element Coal RDF grate ash hopper ash
Al + + + +
As + + +
Ca + + + +
Cr + + + +
Cu +
Ga + +
Ge + + + +
Fe + + + +
Pb + +
Mg + + + +
Mn + + + +
Ni +
P +
K + + + +
Rb + + + +
Se + + +
Si + + + +
Na + +
Sr + + +
Ti + + +
V + +
Zn +
—' Hopper ash elements show an increase from 60 to 80% load but & decrease
from 80 to 100% load. The overall or general trend is an increase.
126
-------
TABLE A-2. ANOV ELEMENTS IN COAL, RDF, GRATE ASH, AND HOPPER ASH
WITH RESPECT TO PERCENT RDF INPUT IN THE FUEL FOR GIVEN
LEVELS OF BOILER LOAD
Boiler Collecter-/
Coal RDF grate ash hopper ash
Element 5 6k/ 5 6 56 56
Al - - + + +
As _-++ ++ +
Ca - - + + +
Cr - - + +
Cu - + + +
Ga - - + + ++ +
Ge - - + -i- + +
Fe - - + + +
Pb - - + + + +
Mg - + + + + +
Mn - - + + + +
Ni
p + + + +
K - - + +
Rb - - + + + +
Se - - + + + +
Si -- ++ ++ +
Na - + + + + +
Sr - - + +
Ti __++ + + +
V - - + + + +
Zn + + + + +
— Hopper ash elements show an increase from 60 to 80% load but a
decrease from 80 to 100% load. The overall or general trend in an
increase.
_b/ 5 and 6 refer to boiler unit Nos. 5 and 6.
127
-------
TABLE A-3. ANOV OF EFFLUENTS WITH BOILER LOAD FOR
GIVEN LEVELS OF RDF IN THE INPUT FUEL
Effluent (emission specie) Unit no. 5
Particulate Matter (Uncontrolled)^/ +
Particulate Matter (Controlled)^
Oxides of Nitrogen
Oxides of Sulfur +
Aldehydes and Ketones
Phosphate
Cyanide
Chlorides +
Carbon Dioxide (Orsat) +
Oxygen (Orsat)
Excess Air (Orsat)
_a/ Uncontrolled particulate emissions were measured between the
boiler and the mechanical particulate collector.
_b/ The boiler and the mechanical particulate collector controlled
particulate emissions were measured in the exhaust stack after
the mechanical particulate collector.
128
-------
TABLE A-4. ANOV WITH PERCENT RDF IN THE
INPUT FUEL FOR GIVEN LEVELS OF BOILER LOAD
Effluent (emission specie)
Unit no. 5
Unit no. 6
Particulate Matter (Uncontrolled!/)
Particulate Matter (Controlled^/
Oxides of Nitrogen
Oxides of Sulfur
Aldehydes and Ketones
Phosphate
Cyanide
Chlorides
Carbon Dioxide (Orsat)
Oxygen (Orsat)
Excess Air (Orsat)
a_/ Uncontrolled particulate emissions were measured between the boiler
and the mechanical particulate collector..
b/ The boiler and the mechanical particulate collector controlled
particulate emissions were measured in the exhaust stack after
the mechanical particulate collector.
129
-------
While these conclusions seem obvious based on "common sense" they are,
in fact, based on standard analysis of variance techniques.-i' Such techniques,
if valid, must be consistent with what one knows to be physically true, as
they are in this case. It should also be noted that when the variability is
large, the ANOV procedures may not detect significant trends in elemental
flow rates. In addition, one's intuition or what seems to be common sense
is often unreliable. Because of this, the ANOV procedures are felt to be
essential for interpretation of data of these experiments.
RDF ELEMENT FLOW RATES
Analysis of the variability of the element flow rates in the RDF show
the following:
1. As the boiler load increases for a given percentage of RDF, on
a heat energy input basis, the flow rate of most elements in the RDF as shown
in Table A-l increases.
2. As the percent RDF increases for a given boiler load, the flow
rate of most elements in the RDF as shown in Table A-2 increases. This con-
clusion is valid for both boiler unit Nos. 5 and 6.
Again these conclusions from ANOV are consistent with what one knows to
be physically true. However, the variability is sufficiently large in some
of the elemental constituents of the refuse that the ANOV procedure cannot
detect a significant change in elemental flow rate at the 95% confidence
level.
ASH ELEMENT FLOW RATES
Grate (Bottom) Ash
Analysis of variability of the element flow rate in the boiler grate
ash shows the following:
1. As the boiler load increases for a given percentage of RDF (on
a heat energy input basis) the flow rate of most elements, as shown in Table
A-l, in the boiler grate ash increase.
2. As the percent RDF increases for a given boiler load, the flow
rate of most elements in the boiler grate ash are shown in Table A-2. This
is generally true for boiler unit No. 5 and to a lesser extent on boiler unit
No. 6. The elements of Ge, Mn, P, Rb, and V on unit No. 6 did not exhibit
the increasing trend at the 95% confidence level as found on unit No. 5.
130
-------
Statistical ANOV has shown a correlation of some elements in the ash with
the percent RDF in the fuel. These elements include As, Gu, Ga, Pb, Mg, Mn,
Si, Na, V, and Zn.
Hopper (Fly) Ash
Analysis of variability of the element flow rates in the hopper ash show
the following:
1. As the boiler load increases for a given percentage of RDF (on
a heat energy input basis), the flow rate of many elements, as shown in Table
A-l, increases for a 60 to 80% load change, but decreases for an 80 to 100%
load change. It is natural to expect only increases in these flow rates as
load increases. This effect is most likely to be the result of a significant
decrease in particulate collector efficiency for loads above 85 to 90%.
2. As the percent RDF increases for a given boiler load, the flow
rate of most elements in the hopper ash follows a similar trend as above.
It was noted that a significant interaction effect between percent load and
percent RDF existed for most elements on boiler unit No. 5. This would be
expected since increases in percent RDF resulted in significant increases in
percent excess air (see the section under Effluent Flow Rates on Chlorides),
which in turn resulted in increased gas flow in the particulate collector
(an effect similar to increasing the load). No significant changes in ele-
ment flow rates were detected on boiler unit No. 6 with percent RDF at the
80% load condition.
EFFLUENT FLOW RATES
ANOV of the various emission flow rates (kg/hr) from the stack for boiler
unit Nos. 5 and 6 were performed with respect to both boiler load and percent
RDF in the fuel. Tables A-3 and A-4 summarize these results.
Effluent (emission) flow rates (kg/hr) from boiler unit No. 5 were
affected by boiler load as shown in Table A-3. Effluent (emission) flow
rates (kg/hr) from the stack serving boiler unit Nos. 5 and 6 were affected
by the percentage RDF in the fuel input, as shown in Table A-4. A (+)
in this table indicates a statistically significant (95% confidence level)
increase in the effluent flow rate with increase in boiler load. A (-)
indicates a statistically significant (95% confidence level) decrease in
effluent flow rate with increase in boiler load. A ( ) represents either no
significant effect, or the inability to detect or resolve a significant effect
at the 95% confidence level. No analysis for boiler unit No. 6 can be included
in this tabulation since only one boiler load (80%) was used for the experi-
ments on this boiler.
131
-------
Particulates
ANOV of the controlled (stack) particulate emissions with load on unit
No. 5, as shown in Table A-3, was inconclusive in that neither an increase
nor decrease in particulate flow rate (kg/hr) could be resolved. This means
that either a confidence level lower than 95% should be chosen or that more
data are needed in order to resolve the effect at the 95% confidence level.
Only one load (80%) was used on boiler unit No. 6 so ANOV could not be per-
formed on that data.
The ANOV of the controlled (stack) particulates with percent RDF, as
shown in Table A-4, indicates that boiler unit Nos. 5 and 6 have opposite
trends. The controlled particulate emissions generally decrease on unit
No. 5, but increase on unit No. 6, with an increase in percent RDF for a
given boiler load. The reason for the reverse trend on unit No. 6 is
believed to be from the additional carry over of "fines" from the input
fuel. Boiler unit No. 6 is not much larger physically than unit No. 5, yet
the rated loads are 12.5 Mw compared to 7.5 mw. This significant difference
in loads means there is also a significant increase in volume flow rate of
flue gas and effluents.
Thus, the largest flow rate of combustion or flue gas and particulates
will be in the unit No. 6 boiler. Since the two boilers have about the same
physical size, unit No. 6, with the higher volume flow, will then have the
higher flow velocity of gases and particulates. This, in turn, is believed
to be the cause of the additional amounts of fine particles carried through
this boiler to the stack.
ANOV of the uncontrolled particulates as sampled before the particulate
collectors, shows an increase with boiler load for given levels of RDF input
on unit No. 5. No significant increase or decrease was detected on either
unit Nos. 5 or 6 as percent RDF increased for given levels of boiler load.
Statistical analysis of variance for the data in megagrams per megajoules
units has revealed that some of the chemical elements in the stack particulate
effluent correlate well with RDF in the fuel input to the boiler. For example,
on boiler unit No. 5 using Iowa coal mixed with RDF, iron, copper, titanium,
nickel, zinc, gallium, germanium, selenium, and lead effluents from the stack
are significantly related to the RDF input. These elements are based on the
analysis of particulate matter caught on the quartz fiber filters in the sam-
pling train. Analysis of the particulates trapped in the impinger water
reveal significant correlation of arsenic, selenium, and lead with RDF input
to boiler unit No. 5.
132
-------
For boiler unit No. 6 using a 50/50 mixture of Iowa and Wyoming coal
combined with RDF, the significant elements in the stack effluents correla-
ting with RDF, in the input fuel are calcium, chromium, copper, gallium,
manganese, selenium, titanium, vanadium, and zinc. These elements are based
on the analysis of the particulate matter caught on the quartz fiber filters.
Analysis of the particulates trapped in the impinger water reveal no correla-
tion of the elements found with RDF in the input fuel for boiler unit No. 6.
Oxides of Nitrogen (NOX)
ANOV as shown in Tables A-3 and A-4 indicates a significant decrease in
NOX emission rate with increases in percent RDF for given levels of boiler
load. There also was no increase in percent RDF for given levels of boiler
load. No increase or decrease in NOX emission rate could be detected at the
95% confidence level with increases in boiler load. If there is an effect
with load, one must either lower the confidence level or obtain more data
in order to detect the trend.
The decrease in the NOX levels indicates either a decreased combustion
zone temperature or a decreased resident time in the combustion zone of the
appropriate chemical species.
Oxides of Sulfur (SOX)
ANOV as shown in Tables A-3 and A-4 indicates a significant increase in
SOX emission flow rate (kg/hr) as boiler load increases. A significant
decrease occurs as percent RDF is increased for given levels of boiler load.
These results were conclusive for boiler unit No. 5 and indicate the value
of using RDF to decrease the SOX emissions from such a stationary source of
effluent. These results also indicate that one could use a higher sulfur
content coal in combination with RDF and still achieve existing SOX emission
standards•
Aldehydes and Ketones (Formaldehyde)
ANOV as shown in Tables A-3 and A-4 indicates neither a detectable
increase nor decrease in the emission flow rate of the aldehydes and ketones
with variations in either load or percent RDF. If there is an effect with
either load or percent RDF, one must either lower the confidence level or
obtain more data in order to detect the effect.
The variability in the aldehyde and ketone data was relatively high com-
pared to the other data of this study. One reason for this is that the alde-
hyde and ketone emissions are quite dependent on the cellulose fiber in the
input fuel. Much of such cellulose fiber would be contained in the RDF as
wood chips, leaves, and lawn clippings. The amount of these items in the
RDF is quite variable on a day-to-day basis. This variability extends to the
133
-------
aldehyde and ketone emissions from the stack which, in turn, masks any trends
in the emissions from being detected in the ANOV procedure.
Organic Acids, Cyanide, and Phosphate
ANOV as shown in Tables A-3 and A-4 indicates neither a detectable
increase nor decrease in the emission flow rates of both cyanide and phos-
phate with variation of either load or percent RDF. As stated before, if
there is an effect with either load or percent RDF, one must either lower the
confidence level or obtain more data in order to detect the effect.
The variation in cyanide emission flow rates were quite high in similar
fashion to the aldehydes and ketones emission flow rates. This high varia-
bility marks any trends from being detected by the ANOV procedure. The phos-
phate emissions were usually quite close to the detection limit of the analy-
tical instrumentation and as such may be somewhat unreliable. For this reason
the ANOV results for the phosphate may be unreliable.
Chlorides
ANOV of the chloride flow rates, as shown in Table A-3 and A-4, shows a
significant increase with increases in both boiler load (for a fixed level
of percent RDF in the input fuel) and a percent RDF in the input fuel (for
a fixed level of boiler load).
The ANOV also shows that the increase in chlorides flow rate is more
predominant with increases in RDF than with increases in boiler load. This
would be natural to expect based on the constituents of RDF.
Carbon Dioxide, Oxygen and Excess Air from Orsat Measurements
ANOV of the Orsat data as shown in Tables A-3 and A-4 indicates the
following:
1. Carbon dioxide increases significantly, oxygen decreases signi-
ficantly, and percent excess air decreases significantly with increases in
boiler load for given levels of RDF in the input fuel.
2. Oxygen and excess air both increase significantly with increases
in the percentage of RDF included in the input fuel for given levels of boiler
load. This trend is a direct result of the boiler operators bringing more
"overfire" air into the boilers as the percent RDF is increased to the boiler
for a given boiler load.
No significant increase or decrease was detected for the carbon dioxide
emission with increase in RDF. If there is an effect, one must either lower
the confidence level,or obtain more data in order to detect the effect.
134
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-79-222
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE Evaluation of the Ames Solid Waste
Recovery System Part III: Environmental Emissions of
the Stoker Fired Steam Generators
5. REPORT DATE
October 1979 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J.L. Hall, A.W. Joensen, D. Van Meter, H.R. Shanks,
G. Severns, R. Wehage, R.Reece, D.E. Fiscus, R.W. Whits
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Engineering Research Institute
Iowa State University, Ames, Iowa 50011
Ames Laboratory, ERDA, Ames, Iowa
10. PROGRAM ELEMENT NO.
1NE-624-B
11. CONTRACT/GRANT NO.
Grant No. R803903-01-0
12. SPONSORING AGENpY NAME AND ADDRESS
Industrial Environmental Research Laboratory-Cinn., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Interim Feb. 5,1Q76-Fph.4 1 Q77
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
Project Officers: Carlton C. Wiles, (513) 684-7881; Robert A. Olexsey (513) 684-4363
16. ABSTRACT
This report describes the results of the :ests of environmental emissions from
the traveling grate stoker boilers at Ames, Iowa. All input and output streams asso-
ciated with operation of these steam generating units have been sampled. Characteriza-
tion of fuel (coal and process refuse), ash and stack effluents has been accomplished.
Statistical analysis of the data is also included.
Particulate emissions either increased or remained the same with increasing RDF
depending on boiler load. NOX and sulfur emissions decreased and chloride emissions
increased with increasing percent RDF. Formaldehyde, cyanide, and phosphate emissions
were quite variable with no clear trends of emissions as a function of percent RDF
burned. Many of the organic compounds were below the laboratory detection levels, and
no comparison of emissions can be made at this time. For the trace elements in the
uncontrolled particulates, there were increases due to burning RDF for only copper and
lead.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Coal
Refuse
Evaluation
Combustion
Air pollution
Particulate
Gas
Trace elements
Municipal wastes
Particulate emissions
Stationary sources
Emission sampling
Boilers
13B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
149
20. SECURITY CLASS (Thil page)
Unclassified
22. PRICE
EPA Form 222O-1 (9-73)
135
.US GOVEKSMEST « •,'",! OFFiCt '960-b 17-Ufe / 5498
------- |