xvEPA
United States
Environmental Protection
Agency
Office of Air Quality EMB Report No. 87-MIN-04
Planning and Standards Volume I
Research Triangle Park, NC 27711 September 1988
Air
Municipal Waste Combustion
Multipollutant Study
Characterization Emission Test Report
Marion County
Solid Waste-to-Energy Facility
Ogden Martin Systems of Marion, Inc.
Brooks, Oregon
-------
DCN No. 88-222-124-12-07 EMB Report No. 87-MIN-4
CHARACTERIZATION TEST REPORT
MARION COUNTY
SOLID WASTE-TO-ENERGY FACILITY, INC.
OGDEN MARTIN SYSTEMS OF MARION
BROOKS, OREGON
VOLUME I: SUMMARY OF RESULTS
ESED Project No. 86/19
EPA Contract No. 68-02-4338
Work Assignment 17
Prepared for:
Clyde E. Riley, Task Manager
Emissions Measurement Branch
Emission Standards and Engineering Division
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared by:
Carol L. Anderson
Michael A. Vancil
J. William Mayhew
Donna J. Holder
Radian Corporation
Post Office Box 13000
Research Triangle Park, NC 27709
September 1988
lmo/036
-------
DISCLAIMER
This report has been reviewed by the
Emission Standards Division of the Office of Air
Quality Planning and Standards, EPA, and
approved for publication. Mention of trade
names or commercial products is not intended to
constitute endorsement or recommendation for
use. Copies of this report are available
through the Library Services Office (MD-35),
U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina 27711.
lmo/036
-------
Acknowledgements
The work reported herein was performed by personnel
from Radian Corporation, Midwest Research Insitute (MRI),
Entropy Environmentalists, Inc., Ogden Projects, Inc., and
the U. S. Environmental Protection Agency (EPA).
Radian's Task Director, Winton Kelly, directed the
field sampling and analytical effort and was responsible
for summarizing the test and analytical data presented in
this report. Sample analyses were performed by Radian
Corporation in Research Triangle Park, North Carolina, and
by Triangle Laboratories, Inc., Research Triangle Park,
North Carolina. Entropy Environmentalists, Inc. conducted
the continuous HC1 monitoring.
Mr. Peter Schindler, Office of Air Quality Planning
and Standards, Industrial Studies Branch, EPA, served as
Project Lead Engineer and was responsible for coordinating
the process operations monitoring in conjunction with
Dr. Ted Brna and Mr. Jim Kilgroe, who served as the Air
and Energy Engineering Research Laboratory (AEERL) Lead
Engineers.
Mr. Clyde E. Riley, Office of Air Quality Planning
and Standards, Emission Measurements Branch, EPA, served
as Project Task Manager and was responsible for overall
test program coordination.
The Office of Air Quality Planning and Standards,
EPA, would like to thank the following individuals for
their cooperation and assistance in the execution of the
test program:
Ogden Martin Systems of Marion. Inc.
Mr. Fred Engelhardt, Facility Manager
Mr. Russel Johnston, Chief Engineer
Mr. Don Penrose, Maintenance Supervisor
Martin GmbH
Mr. Johannes Martin, Director of Engineering
and Design
Mr. Joachim Horn, Process Engineer
Ogden Projects. Inc.
Mr. David Sussman, Vice President -
Environmental Affairs
Mr. Jeffrey Hahn, Vice President -
Environmental Engineering
Mr. Henry Von Demfange, Manager -
Environmental Testing
The efforts of these individuals and members of their
staff are greatly appreciated.
-------
FOREWORD
The data contained in this report represent
the operating conditions of the facility at the
time of the test program. Since the completion
of the test program, however, a program of
screening the waste received at the facility and
removing materials which resulted in high S09
emissions has been implemented. Additionally,
the lime feed now operates at a higher rate than
during the test program. Because of these
actions, S0_ emissions are believed to have
decreased from the values reported here.
lmo/036
-------
RADIAN REPORT CERTIFICATION
This report has been reviewed by the following Radian personnel and is a
true representation of the results obtained from the sampling program at
Marion County Solid Waste-to-Energy Facility, Inc., Ogden Martin Systems of
Marion, Brooks, Oregon. The sampling and analytical methods were performed in
accordance with procedures outlined in the "Field Test Plan for the
Characterization Test Program" dated June 2, 1987. The sampling and
analytical plan was reviewed and accepted by the EPA/EMB Task Manager,
Clyde E. Riley.
APPROVALS
Project Director: A**— • <*, Date:
Winton E. Kelly1
Program Manager: // k^^ ' V ' ^Y (^ Date'
QA Officer: /y^ /IX/U^C^ / ' /^" -^-~ "- Date:
Donna J. Holder
-------
TABLE OF CONTENTS
Section Page
VOLUME I
List of Figures xiii
List of Tables x
1.0 INTRODUCTION 1-1
1.1 PURPOSE AND OBJECTIVES 1-1
1.2 BRIEF PROCESS DESCRIPTION 1-3
1.3 CHARACTERIZATION TEST PROGRAM 1-5
1.3.1 Sampling Matrix 1-5
1.3.2 Sampling and Analytical Procedures 1-7
1.4 ORGANIZATION 1-7
1.5 QUALITY ASSURANCE/QUALITY CONTROL (QA/QC) 1-14
1.6 DESCRIPTION OF REPORT SECTIONS 1-14
2.0 SUMMARY OF RESULTS 2-1
2.1 BASELINE EMISSIONS
2.1.1 Baseline Acid Gas Emissions 2-4
2.1.2 Temperature Profile for Baseline Conditions 2-6
2.1.3 Combustion Parameters and Combustion Efficiency. . . 2-7
2.1.4 Fixed Gases (CO, CO and 0_) 2-11
2.1.5 Additional Pollutants of Interest (NO and THC). . . 2-11
2.1.6 CDD/CDF Concentrations in Ash. . . . * 2-14
2.2 COMBUSTOR VARIATIONS 2-16
2.2.1 Temperature Profile during Combustor Variations. .- . 2-16
2.2.2 Combustion Parameters during Combustor Variations. . 2-23
2.2.3 Fixed Gases (CO, CO and 0 ) 2-23
2.2.4 Additional Pollutants of Interest (NO and THC). . . 2-35
2.2.5 Acid Gas Emissions X 2-37
2.2.6 CDD/CDF Concentration in the Ash 2-37
2.3 EFFECT OF OFF-DESIGN TEMPERATURES IN EMISSION CONTROL SYSTEM
2.3.1 Acid Gas Emissions during Control Device Variations. 2-41
2.3.2 Temperature Profile during Control Device Variations 2-48
2.3.3 Fixed Gases (CO, CO., and 0.) and Additional
Pollutants of Interest (NO and THC) 2-55
2.3.4 CDD/CDF Concentrations in Asn during Control Device
Variations 2-55
3.0 CONCLUSIONS 3-1
4.0 PROCESS DESCRIPTION AND OPERATION 4-1
4.1 PROCESS DESCRIPTION 4-1
4.1.1 Combustor Description 4-1
4.1.2 Emission Control System 4-2
4.2 TESTING GOALS 4-4
4.3 TESTING MATRIX 4-5
4.3.1 Combustor Evaluation 4-5
4.3.2 Control Device Evaluation 4-7
-------
TABLE OF CONTENTS
(Continued)
Section
5.0 SAMPLE POINT LOCATIONS
5.1 FLUE GAS
5.1.1 Boiler Outlet (Control Device Inlet) Sampling
Location
5.1.2 Midpoint Sampling Location
5.1.3 Breeching to the Outlet Stack
5.1.4 Outlet Stack Sampling Location
5.2 ASH AND PROCESS SAMPLES
5.2.1 Superheater Ash Sampling Location
5.2.2 Economizer Ash Sampling Location
5.2.3 Baghouse Ash and Cyclone Ash Sampling Locations.
5.2.4 Lime Slurry Sampling Location
5.2.5 Tesisorb Sampling Location
6.0 SAMPLING AND ANALYTICAL PROCEDURES
6.1 CONTINUOUS EMISSION MONITORS (CEMs)
6.1.1 Sampling at the Midpoint Location
6.1.2 Stratification Check
6.1.3 Averaging Method
6.2 MANUAL METHODS
6.2.1 HC1 Determination
6.2.1.1 Manual HC1 Sampling
6.2.1.2 HC1 Analysis
6.2.2 Volumetric Flowrate Determination
6.2.3 Moisture Determination
6.2.4 Fixed Gases Determination
6.2.5 SO- Determination
6.2.6 Asfi Sampling
7.0 INTERNAL QUALITY ASSURANCE/QUALITY CONTROL
7.1 QUALITY ASSSURANCE OVERVIEW OF THE MARION COUNTY
TEST PROGRAM
7.2 QA/QC OBJECTIVES AND RESULTS
7.3 QA/QC RESULTS
7.3.1 Ash CDD/CDF Sampling and Analysis ,
7.3.1.1 Internal Standard and Surrogate Recoveries.
7.3.1.2 Duplicate Analyses
7.3.1.3 Sample Blanks
7.3.2 HC1 Flue Gas Sampling and Analysis Quality Control .
7.3.3 Continuous Emission Monitor (GEM) Quality Control. ,
7.3.3.1 Daily Calibrations and Drift Checks ...
7.3.2.2 System Bias Checks
7.3.3.3 Response Times
7.3.3.4 Daily QC Checks
7.3.3.5 Multipoint Linearity Checks
7.3.3.6 Relative Accuracy
ii
-------
TABLE OF CONTENTS
(Continued)
Section Page
7.0 INTERNAL QUALITY ASSURANCE/QUALITY CONTROL (Continued)
7.3.4 Manual Sampling ................... 7-35
7.3.5 Validation of Fixed Gases Results .......... 7-35
7.3.6 EPA Method 6 SO Quality Control .......... 7-39
7.3.7 CEM Stratification Check .............. 7-47
7.3.8 Sulfur Dioxide (S02) Quenching Study ........ 7-47
8.0 REFERENCES ........................... 8-1
9.0 METRIC -TO -ENGLISH CONVERSION TABLE ............... 9-1
VOLUME II
APPENDIX A - SUMMARY OF CHARACTERIZATION TEST RESULTS
A.I Combustion Evaluatiion .................... A-l
A. 1.1 - Field Results ..................... A-2
A. 1.2 - CDD/CDF Ash, Lime Slurry and Tesisorb Results ..... A-8
A. 2 Control Device Evaluation ................... A-21
A. 2.1 - Field Results ..................... A- 22
A. 2. 2 - CDD/CDF Ash, Lime Slurry and Tesisorb Results ..... A-28
A. 3 Plots of Test Results ..................... A- 34
A. 3.1 - SO- Concentrations at the Inlet, Midpoint, and Outlet. A-35
A. 3. 2 - HCI Concentrations at the Inlet, Midpoint, and Outlet. A- 57
A. 3. 3 - Oxygen Concentrations at the Inlet, Midpoint
and Outlet ..................... A-71
A. 3. 4 - Inlet CO and 0- Concentrations ............ A-90
A. 3. 5 - Inlet NO Concentrations ............... A- 114
A. 3. 6 - Overall ci and S02 Removal Efficiencies ....... A- 129
A. 3. 7 - Quench Reactor Removal Efficiencies .......... A-151
A. 3. 8 - HCL Concentrations: Manual Methods vs. Instrument. . . A-157
A. 3. 9 - Combustion Air Flow .................. A- 161
A. 3. 10 - Overfire Air Differential Pressure .......... A- 167
A. 3. 11 - Furnace Temperatures ................. A-173
A. 3. 12 - Control Device Temperature Profile .......... A-179
A. 3. 13 - Moisture at the Inlet, Midpoint, and Outlet ..... A- 188
iii
-------
TABLE OF CONTENTS
(Continued)
Section Page
VOLUME III
APPENDIX B - CONTINUOUS EMISSIONS MONITORING
B.I Summary of Average GEM Results for Each Parameter B-3
B.2 GEM Hourly and One Minute Averages [adjusted for
drift and SO,, quenching] B-13
B.3 Calibration Data B-79
B.3.1 Calibration Summaries B-81
B.3.2 Daily Calibration Printouts B-99
B.4 On-line GEM data [not adjusted for drift and SO quenching] . B-140
APPENDIX C - MANUAL METHODS TEST RESULTS C-l
C.I Combustion Evaluation Test Results C-3
C.2 Control Device Evaluation Test Results C-17
APPENDIX D - PROCESS DATA: HOURLY AND ONE MINUTE AVERAGES D-l
VOLUME IV
APPENDIX E - FIELD DATA SHEETS
E.I HCL Field Data Sheets E-l
E.I.I Combustion Evaluation Data Sheets E-3
E.I.2 Control Device Evaluation Data Sheets E-135
E.2 Velocity Traverse Field Data Sheets E-173
E.2.1 - Combustion Evaluation Data Sheets E-175
E.2.2 - Control Device Evaluation Data Sheets E-234
E.3 Ash Sampling Field Data Sheets E-251
E.3.1 - Baghouse Ash Sample Sheets E-253
E.3.2 - Cyclone Ash Sample Sheets E-271
E.3.3 - Economizer Ash Sample Sheets E-287
E.3.4 - Superheater Ash Sample Sheets E-301
E.4 Process Sample Field Data Sheets E-313
E.4.1 - Tesisorb Sample Sheets E-315
E.4.2 - Lime Slurry Sample Sheets E-319
E.5 Preliminary Field Sampling Sheets .... E-323
E.5.1 - Cyclonic Flow Checks and
Preliminary Velocity Traverses E-325
E.5.2 - Traverse Point Location E-329
iv
-------
TABLE OF CONTENTS
(Continued)
Section Page
VOLUME V
APPENDIX F - ANALYTICAL DATA AND RESULTS F-l
F.I HC1 Analysis by SIE F-3
F.I.I - HC1 Analysis Summary Sheets F-5
F.I.2 - Specific Ion Electrode Chloride Analysis
Data Reports F-25
F.I.3 - Calibration Data F-43
F.2 ORSAT Analysis Data Sheets F-51
F.3 CDD/CDF Ash Analysis F-95
F.3.1 - Ash CDD/CDF Results F-97
F.3.2 - Lime Slurry and Tesisorb CDD/CDF Results F-113
F.3.3 - Internal Standard and Surrogate Recoveries for
Ash, Lime Slurry and Tesisorb CDD/CDF Analyses. . . F-117
F.3.4 - Triangle Laboratories Analytical Reports F-121
F.4 SO- Analysis by EPA Method 6 F-213
F.4.1 - S02 Analysis Summary Sheet F-215
F.4.2 - SO. Titration Data Sheets F-219
APPENDIX G - CHARACTERIZATION TEST LOGS G-l
G.I Summary of Sample Logs and Test Crew Chief Notes G-3
G.2 CEM Logbook G-ll
G.2.1 - CEM Log Summaries G-13
G.2.2 - CEM Log G-27
G.3 Analytical Logbook G-99
G.3.1 - Wet Chemistry Analytical Log G-101
G.3.2 - Sample Logbook G-131
G.3.3 - Sample Identification Logs G-173
G.4 Process Notes, Data Sheets, and Stripcharts G-193
VOLUME VI
APPENDIX H - QUALITY ASSURANCE INFORMATION H-3
H.I Ash CDD/CDF QA/QC Results H-5
H.I.I - Internal Standard and Surrogate Recoveries H-9
H.I.2 - Duplicate Analyses H-13
H.I.3 - Method Blank Results H-17
H.2 HC1 by SIE QA/QC Results H-19
H.2.1 - Field Blank Results H-23
H.2.2 - Method of Additions H-27
-------
TABLE OF CONTENTS
(Continued)
Section Page
H.3 GEM Quality Control Results H-27
H.3.1 - Daily Calibrations and Drift Checks H-29
H.3.2 - System Bias Check H-47
H.3. 3 - Response Times H-51
H.3.4 - Daily QC Checks H-71
H.3.5 - Multipoint Linearity Checks H-81
H.3.6 - Interference Checks H-93
H.3.7 - Standard Gas Certification Sheets H-97
H.3.8 - Validation of Fixed Gases (Orsat & GEM) H-131
H.3.9 - S0_ Quench Factor Adjustment H-141
H.3.10 - NO Stratification Check H-145
H.4 Manual Method QC Results H-151
H.4.1 - Train Leakchecks H-153
H.4.2 - Calibration Results H-159
H.4.2.1 Meterboxes H-161
H.4.2.2 Temperature Sensors H-183
H.4.2.3 Pitots H-187
H.4.2.4 Top Loader Balance H-203
APPENDIX I - SUMMARY OF EQUIPMENT USED 1-3
APPENDIX J - SAMPLE CALCULATIONS J-l
APPENDIX K - SAMPLING AND ANALYTICAL PROTOCOLS K-l
K.I - Summary of EPA Reference Methods K-3
K.2 - Ash, Lime Slurry and Tesisorb Sampling and
Analytical Procedures K-9
K.3 - ASME/EPA Protocol to Assay Stack Effluent Samples and
Residual Combustion Products for Polychlorinated
Dibenzo-p-dioxins (PCDD) and Polychlorinated
Dibenzofurans (PCDF). (December 31, 1984 Draft) K-15
APPENDIX L - PERTINENT CORRESPONDENCE L-l
L.I Test Program Summary Letter Report L-3
L.2 S02 Quenching Study Letter Report L-17
L.3 Sample Custody Letters L-45
L.3.1 - Sample Custody Letters for CDD/CDF
Analysis of Ash Samples L-47
L.3.2 - Sample Custody Letters for HC1
Analysis of Flue Gas Samples L-55
L.3.3 - Sample Custody Letters for Audit Samples L-71
L.4 Letter Reports for Ash CDD/CDF Analyses L-77
VI
-------
TABLE OF CONTENTS
(Continued)
Section Page
L.5 Telecon regarding CDD/CDF Confirmation
and Screening Analyses L-127
L.6 Letter Report for the CDD/CDF Audit Samples L-131
APPENDIX M - PROJECT PARTICIPANTS M-l
vii
-------
LIST OF FIGURES
Figure Page
1-1 Marion County Process Line 1-4
1-2 Marion County Characterization Test Program Line of
Communication 1-13
2-1 Location of Temperature Indicators for the Marion County MWC. . 2-8
2-2 Baseline Congener Distribution for Ash 2-18
2-3 Temperature Profile for Low Load Combustor
Evaluation Conditions 2-21
2-4 Temperature Profile for Air Distribution and Excess Air
Combustor Evaluation Conditions 2-22
2-5 Variation of Steam Load during the Combustor Evaluation .... 2-26
2-6 Variation of Excess Air during the Combustor Evaluation .... 2-27
2-7 Variation of Volumetric Flowrate during the
Combustor Evaluation 2-28
2-8 Fixed Gas Concentration Histories during the Combustion
Evaluation 2-31
2-9 Ash CDD Congener Distributions 2-46
2-10 Ash CDF Congener Distributions 2-47
2-11 Effect of Acid Gas Concentration on Control Efficiency 2-51
2-12 Temperature Profile for After the Quench Reactor for Control
Device Evaluation Conditions 2-54
2-13 CDD/CDF Congener Distributions for Baghouse Ash During the
Control Device Evaluation 2-60
5-1 Marion County MWC Process Line with Sampling Locations 5-2
5-2 Top View of Boiler Outlet and Midpoint Sampling Locations
at Marion County MWC 5-3
5-3 Side View of Boiler Outlet Sampling Location at Marion
County MWC 5-4
5-4 Traverse Point Location Diagram for Boiler Outlet Location
at Marion County MWC 5-5
Vlll
-------
LIST OF FIGURES
(Continued)
Figure Page
5-5 Side View of Midpoint Sampling Location at Marion County MWC. . 5-7
5-6 Velocity Traverse Point Location Diagram for the Midpoint
Location at Marion County MWC 5-8
5-7 Breeching to the Stack Sampling Location at Marion County MWC . 5-10
5-8 Stratification Point Location Diagram for the Breeching
Location at Marion County MWC 5-11
5-9 Outlet Stack Sampling Location at Marion County MWC 5-13
5-10 Side View of Outlet Stack Sampling Location at Marion
County MWC 5-14
5-11 Velocity Traverse Point Location Diagram for the Outlet Stack
Location at Marion County MWC 5-15
5-12 Side View of Superheater Ash Sampling Location at Marion
County MWC 5-17
5-13 Top View of Superheater Ash Sampling Location at Marion
County MWC 5-18
5-14 Side View of Economizer Ash Sampling Location at Marion
County MWC 5-19
5-15 Baghouse and Cyclone Ash Sampling Locations at Marion
County MWC 5-20
5-16 Tesisorb Sampling Location at Marion County MWC 5-22
6-1 GEM Sampling and Analysis Scheme for the Midpoint Sampling
7-1
7-2
7-3
7-4
7-5
7-6
Location for the Characterization Testing at Marion County. .
Validation of Fixed Gas Analysis for the Inlet CEM Results. . .
Validation of Fixed Gas Analysis for the Inlet Orsat Results. .
Validation of Fixed Gas Analysis for the Midpoint CEM Results .
Validation of Fixed Gas Analysis for the Midpoint
Orsat Results .
Validation of Fixed Gas Analysis for the Outlet CEM Results . .
Validation of Fixed Gas Analysis for the Outlet Orsat Results .
6-3
7-40
7-41
7-42
7-43
7-44
7-45
IX
-------
LIST OF TABLES
Table
1-1
1-2
1-3
1-4
1-5
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
2-11
2-12
2-13
2-14
Marion County Characterization Test Sampling Matrix
Target and Actual Values of Combustion and Control Device
Parameters Varied During the Characterization Test Program. .
Summary of the Sampling Intervals for Characterization
Test Program at the Marion County MWC
Summary of Sampling and Analytical Procedures
CDD/CDF Congeners Analyzed for the Marion Test Program
Summary of Baseline and Combustor Variation Results for the . .
Marion County MWC
Summary of Baseline and Control Device Variation Results for
for the Marion County MWC
Summary of Baseline Acid Gases and Control Efficiencies ....
Baseline Temperature Profile for the Marion County MWC
Baseline Combustion Parameters for the Marion County MWC. . . .
Baseline Fixed Gases Concentrations (CO, C00 , 0_)
NO and THC Emissions for Baseline Conditions
X
CDD and CDF Concentrations and 2378 -TCDD Toxic Equivalencies
for Ash from Baseline Conditions at Marion County MWC ....
CDD and CDF Concentrations for Ash at Baseline Conditions
at Marion County MWC
Combustor Variation Temperature Profile for the Marion
County MWC
Difference from Baseline for the Combustor Evaluation for the
Marion County MWC . . .
Combustion Parameters during the Combustor Evaluation
Difference from Baseline for Combustion Parameters during the
Combustor Evaluation . ..
Fixed Gases Concentrations during the Combustor Evaluation. . .
Page
1-6
1-8
1-9
1-11
1-12
2-2
2-3
2-5
2-9
2-10
2-12
2-13
2-15
2-17
2-19
2-20
2-24
2-25
2-29
-------
LIST OF TABLES
(Continued)
Table Page
2-15 NO and THC Emissions for the Combustor Evaluation Conditions . 2-36
X
2-16 Summary of Acid Gas Concentrations during the Combustor
Evaluation 2-38
2-17 Control Device Removal Efficiencies during the Combustor
Evaluation 2-39
2-18 CDD and CDF Concentrations and 2378-TCDD Toxic Equivalencies
for Ash from Combustor Evaluation Conditions at Marion
County MWC 2-40
2-19 CDD and CDF Results for Superheater Ash at Combustor
Evaluation Conditions 2-42
2-20 CDD and CDF Results for Economizer Ash at Combustor
Evaluation Conditions 2-43
2-21 CDD and CDF Results for Cyclone Ash at Combustor
Evaluation Conditions 2-44
2-22 CDD and CDF Results for Baghouse Ash at Combustor
Evaluation Conditions 2-45
2-23 Acid Gas Behavior for the Control Device Evaluation Testing . . 2-49
2-24 Temperature Profile and Difference from Baseline 2-53
2-25 CDD and CDF Concentrations and 2378-TCDD Toxic Equivalencies
for Ash from Control Device Evaluation Conditions at Marion
County MWC \ . 2-56
2-26 CDD and CDF Results for Cyclone Ash at Control Device
Evaluation Conditions 2-58
2-27 CDD and CDF Results for Baghouse Ash at Control Device
Evaluation Conditions 2-59
4-1 Combustor Evaluation Test Matrix 4-6
4-2 Process Operating Parameters Recorded during Marion County
Testing 4-8
4-3 Tested Operating Range of Primary Operating Variables 4-9
7-1 Summary of Estimated and Achieved Precision, Accuracy, and
Completeness Objectives 7-4
xi
-------
LIST OF TABLES
(Continued)
Table Page
7-2 Summary of Acceptance Criteria, Control Limits and
Corrective Action Followed for Marion County 7-6
7-3 Internal Standards Recovery Results for Marion County
CDD/CDF Ash Analyses 7-11
7-4 Surrogate Recoveries for Marion County Ash CDD/CDF Analyses. . . 7-13
7-5 Duplicate Results for Marion County CDD/CDF Ash Analyses .... 7-14
7-6 Analytical Method Blank Results for Marion County CDD/CDF
Ash Analyses 7-17
7-7 Relative Percent Differences Between SIE Direct Reading and
Known Addition Results for Chloride Concentrations 7-19
7-8 Summary of GEM Drift Checks, Marion County, Inlet 7-22
7-9 Summary of GEM Drift Checks, Marion County, Midpoint 7-23
7-10 Summary of GEM Drift Checks, Marion County, Outlet 7-24
7-11 Summary of GEM High Range S02 Drift Checks for Marion County . . 7-25
7-12 GEM System Bias Test for Marion County SO,, and CO. Analyzers. . 7-26
7-13 Responses Times (95%) for Marion County Midrange GEM QC Gases. . 7-27
7-14 Daily QG Checks for the Marion County CEMs 7-29
7-15 Comparison of Measured Method 3 and GEM 0 and CO-
Results for Marion County, Inlet 7-30
7-16 Comparison of Measured Method 3 and CEM 0 and C02
Results for Marion County, Midpoint 7-31
7-17 Comparison of Measured Method 3 and CEM 0 and C02 Results,
for Marion County, Outlet 7-32
7-18 Comparison of EPA Method 6 and CEM SO Results for Marion County 7-33
7-19 Comparison of HCL Manual Result (SIE) and CEM Result for
Marion County 7-34
7-20 Leakcheck Summary for the Marion County HC1 Sampling Trains. . . 7-36
7-21 Duplicate Results for Marion County Method 6 SO- Titrations. . . 7-46
xn
-------
LIST OF TABLES
(Continued)
Table Page
7-22 GEM Stratification Check for the Marion County Inlet
Sampling Location 7-48
7-23 GEM Stratification Check for the Marion County Midpoint
Sampling Location 7-49
7-24 GEM Stratification Check for the Marion County Outlet
Sampling Location 7-50
7-25 Comparison of Manufacturer's and Derived Quench Equations for
for Marion County TECO 40 (#79) SO- Analyzer 7-52
xiv
-------
1.0 INTRODUCTION
The U.S. Environmental Protection Agency (EPA) published an advance
notice of proposed rulemaking in the Federal Register (52 FR 25399) which
describes upcoming emission standards development for new municipal waste
combustors (MWC) under Section III of the Clean Air Act and for existing MWC
under Section III(d) of the Act. The Federal Register notice follows more
than a year's work of development of the technical and health related
documents which compose EPA's Report to Congress on MWC. The Report to
Congress was a joint effort involving the Offices of Air Quality Planning and
Standards (OAQPS), Solid Waste (OSW), and Research and Development (ORD).
The Emission Standards and Engineering Division (ESED) of OAQPS, through
its Industrial Studies Branch (ISB) and Emissions Measurements Branch (EMB),
is responsible for reviewing the existing air emission data base and gathering
additional data where necessary. As a result of this review, several MWC
emission tests were performed and several more are in the planning stages to
support the current standards development work. Of particular importance is a
more complete data base on emerging air pollution control technologies for
MWC.
The emissions that are being studied by EPA are the criteria
pollutants--particulate matter (PM), sulfur oxides, (SO,,), nitrogen oxides
(NO ), carbon monoxide (CO) and total hydrocarbons (THC); other acid gases,
2t
such as hydrogen chloride (HC1); chlorinated organics including chlorinated
dibenzo-p-dioxins (CDD) and chlorinated dibenzofurans (CDF); and specific
metals including arsenic (As), cadmium (Cd), chromium (Cr), mercury (Hg),
nickel (Ni), lead (Pb) and beryllium (Be).
1.1 PURPOSE AND OBJECTIVES
A data gap was identified by ESED in the area of quench reactor/fabric
filter (QR/FF) controlled emissions. Although QR/FF data were collected
during studies at Quebec City in the National Incinerator Testing and
lmo/036 1-1
-------
Evaluation Program (NITEP) Studies , additional data were required because the
unit tested was a pilot scale unit and the testing did not evaluate the effect
of combustion variation on control system performance. Thus, a parametric
test program was designed to supplement the QR/FF data base. The site
selected for the parametric test program was the Marion County Solid
Waste-to-Energy Facility in Brooks, Oregon. The principal objectives of the
parametric test program were:
1. To evaluate the control efficiency of the QR/FF system on organic
emissions (CDD/CDF) during combustor shutdown and startup
procedures.
2. To evaluate the variation in QR/FF acid gas control as a function of
control device operating temperature and lime stoichiometric ratio.
3. To evaluate the control efficiency of the QR/FF system over the
normal operating range of the combustor.
The parametric test program was conducted in two phases: the
characterization test program and the performance test program. The overall
objective of the characterization test program, which took place in June 1987
at the Marion County facility, was to evaluate the performance of the
combustor and the emissions control system over the range of operation allowed
by the facility's air quality permit. The results of the characterization
test program are the subject of this report.
The specific objectives of the characterization phase of the test program were
to:
1. Determine values for the baseline combustion parameters (combustion
efficiency, CO, C02> 02> S02, N0x> THC, HC1, and combustor
temperature profile) when the steam load, excess air, and air
distribution are set at normal or design conditions.
lmo/036 1-2
-------
2. Determine the baseline performance of the flue gas cleaning system
for SCL/HC1 removal when the temperature and reagent ratio
(stoichiometric ratio) are set at normal or design conditions.
3. Determine the effect of load, excess air and air distribution on CO
emissions at baseline emission control system operating conditions.
4. Determine SO?/HC1 removal efficiency and reagent ratios for
off-design temperatures in the emission control system during
baseline combustor operating conditions.
The evaluation was conducted primarily with continuous emission monitors
(CEMs) and plant instrumentation. During each of the process conditions,
Radian Corporation conducted continuous emission monitoring for S09, NO , 09>
£- X. £*
CO, C09 and THC at the inlet to the control devices and at the outlet of the
control devices. Also, SO,., 09 and CO,, were continuously monitored at a
midpoint between the quench reactor and the baghouse. Radian conducted
simultaneous manual sampling for HC1 at these three locations throughout the
test program. Entropy Environmentalists, Inc., conducted continuous
2
measurements of HCl at the three locations.
In addition to the measurements described in the previous paragraph,
CDD/CDF sampling at the inlet and outlet to the control devices was conducted
during the startup and shutdown test conditions. These results are reported
3
in a separate document. Baseline CDD/CDF emissions data were collected by
EPA during previous emissions tests conducted at the facility in September
19864 and February 1987.5
1.2 BRIEF PROCESS DESCRIPTION
Figure 1-1 presents a process diagram of the two identical combustor
systems at the Marion County facility. Unit No. 1 was tested during the
characterization test program. The combustor is a reciprocating grate,
mass-burning type with a waterwall boiler that produces superheated steam.
lmo/036 1-3
-------
Combustor ^^Boller Superheater Economizer
Quench Reactor/ Teslsorb
Acid Gas f*o<*
Scrubber Hopper
Quench
Pit
-Distributor
To Atmosphere
Figure 1-1. Marion County Process Line
-------
The flue gas passes from the combustor into convection, superheater, and
economizer sections before acid gas and particulate emissions are controlled
by a quench reactor and fabric filter emissions control system.
The refuse is typical residential and commercial solid waste. No sorting
or shredding is performed prior to incineration. The refuse is brought to the
enclosed tipping area by truck and unloaded into the receiving pit. A
manually operated overhead crane transfers the refuse from the receiving pit
to the incinerator charging chute. An inclined grate and ash discharge system
designed by Martin GmbH is used at the Marion County facility.
1.3 CHARACTERIZATION TEST PROGRAM
1.3.1 Sampling Matrix
The Characterization Test Program was performed from June 2 through
June 16, 1987. Table 1-1 presents the overall characterization test matrix
that was planned and performed by EPA in conjunction with Ogden Martin. The
first two test runs established baseline emissions. Combustor and control
device operating conditions were varied during the next 12 test runs.
During the test program, several procedures were modified and additional
tasks were added based on initial results. They are discussed below:
1. Superheater ash was added to the list of process samples taken. For
the superheater and economizer ash the collection technique was
changed to inserting a tube across the hopper. The draft through
the access hole was low enough so that the collected ash was not
re-entrained when the sample was removed from the port. This
technique proved appropriate for the various combustor ash
collection points.
2. An empty modified tip impinger was inserted as the first impinger in
the HC1 train. Evaluations of the collection efficiency of the HC1
sampling train and effect of the midpoint gas conditioning system
were also added to the test program. These modifications and
problems with the on-site specific ion electrode analyses more than
tripled the number of analyses performed.
lmo/036 1-5
-------
TABLE 1-1. MARION COUNTY CHARACTER! ZATIOH TEST SAMPLING MATRIX
Sample
1 2
Base- Base- 3(a) 3(b)
line line
6(a) 6(b)
10
ll(a) 1Kb)
H H
N L
H N
Process Operating Conditions
Combustors
Load (Ib/hr steam) H
Excess Air H
Overfire/Underflre Air N
Distribution
Control Device
Spray Dryer Outlet N
Temperature (°F)
Continuous Monitoring
02 (CEM)
CO (CEM)
CO2 (CEM)
S02 (CEM)
NO (CEM)
THC (CEM)
HC1 (CEM)
Manual Sampling
0 (Orsat) Radian
CO (Orsat) Radian
HC1 (manual) Radian
Fre-and Post-Test Radian
Velocity Traverse
Ash Saaoles
Superheater Radian
Economizer Radian
Cyclone Radian
Baghouse Radian
Emission Control Reagents
Line Slurry Radian
Tesisorb Radian
Process Monitoring (Control Room)
L
N
N
L
H
N
L
L
N
L
N
L
L
N
H
Radian
Radian
Radian
Radian
Radian
Radian
«
Radian
Radian
Plant Strip Chart
Recorders
Manual Recorded
Refuse feed rate
Radian •• - - -
Operator
H » normal
L - low
H - high
T » Test condition to be determined on-site.
b.
NO , TEC, CO were measured at the inlet and outlet
Conducted simultaneously at the inlet, midpoint and breeching.
only.
CConducted at the inlet, midpoint and breeching. The velocity traverses vere conducted at the inlet, midpoint and
outlet stack. For Runs 1 and 2, the outlet stack velocity traverses were conducted by Ogden Martin due to space
limitations on the platform.
1-6
-------
3. The approach to the stratification testing at the breeching was
modified to include a reference measurement and both SO., and NO
y X
were used as the flue gas indicators.
4. The test conditions for Runs 7b, 12 and 13 were to be determined
on-site. However, additional test conditions were not identified
and these runs were deleted. Shutdown and startup testing became
Runs 12 and 13, respectively.
The target and actual values of the combustion and control device
parameters that were varied during the characterization test program are
summarized in Table 1-2. Where applicable, the values are an average of all
tests at that condition.
The sampling intervals and samples collected for the Characterization
Test Program are summarized in Table 1-3. In general, each run was conducted
over a 3-hour period. Problems that occurred during each test run are also
noted.
1.3.2 Sampling and Analytical Procedures
Sampling at the control device inlet, midpoint and outlet were performed
simultaneously following similar protocols. A summary of the sampling and
analytical procedures used is presented in Table 1-4. The target CDD/CDF
congeners for the ash analyses are listed in Table 1-5.
1.4 ORGANIZATION
In order to describe the many interests in the test program, a
communication scheme is shown in Figure 1-2. Mr. Pete Schindler was the
EPA/ISB lead engineer. He was assisted by Mr. Steve Schliesser of Midwest
Research Institute. Dr. Ted Brna and Mr. James Kilgroe were the Air and
Energy Engineering Research Laboratory (AEERL) lead engineers. Mr. Schindler,
Mr. Schliesser, Mr. Kilgroe and Dr. Brna were responsible for coordinating the
overall test program with the plant officials and the Oregon Department of
Environmental Quality (ODEQ), and for ensuring that the process and control
lmo/036 1-7
-------
TABLE 1-2. TARGET AND ACTUAL VALUES OF COMBUSTION AND CONTROL DEVICE
PARAMETERS VARIED DURING THE CHARACTERIZATION TEST PROGRAM
Normal
(baseline)
Excess Air
Target Actual
70% 74.7
(8.5% 0 wet) 7.4
Steam Load
(Ib/hr)
Target Actual
66,400 67,100
Overfire Air
Distribution
(% of total air)
Target Actual
25 NT
Inlet Temperature
to Fabric Filter (°F)
Target Actual
285-300 300
I
00
High
Low
110%
(9.5% 0- wet)
44%
(6.5% 02 wet)
99.5
8.7
36.2 50,550 50,550
4.6
30 NT 360 330, 360
0 NT 260 262
aExcess air is calculated based on conditions at the inlet sampling location.
A high steam load test condition was not tested.
°Run 11A - 330°F, Run 11B - 360°F. During Run 11A, the average fabric filter inlet temperature was
330°F, although the target temperature was 360°F. During Run 11B, the target temperature of 360 F was
achieved. Therefore, the two averages are reported separately.
NT = Not available at this time.
-------
TABLE 1-3. SUKMARY OF THE SAMPLING INTERVALS FOR CHARACTERIZATION TEST PROGRAM AT THE MARION COUNTY MWC
(24-Hour Clock Basil)
Date
06/04/87
06/05/87
Ash
Run Inlet Midpoint Outlet CEMS* Baghouse Cyclone
1 1300-1600 1300-1600 1300-1600 1300-1600 1400, 1550 1330,1430,1530
2 1100-1400 1100-1400 1100-1400 1100-1400 1200,1257,1402 1130,1230,1330
Super-
heater
NS
NS
Lime Slurry
or
Economizer Teslsorb
NS NS
NS LS-1152
TS-1152
Comments
Outlet HC1 manual
results invalidated.
SO spikes occurred.
SO spikes occurred.
06/06/87 3a 1000-1300 1000-1300 1000-1300 1000-1300 1100,1205,1305
06/08/87
1030-1300
06/06/87 3b 1430-1521 1430-1730 1430-1730 1430-1730 1530,1640,1730 1500,1600,1700
NS
NS
1105,1120,1205
Sample
collected,
but log sheet
lost. Sample
times unknown.
1300-1600 1300-1600 1300-1600 1300-1600 1400,1500,1600 1330,1430,1530 1437,1455 1520,1550
NS
NS
NS
06/09/87 5 1000-1300 1000-1300 1000-1300 1000-1300 1100,1200,1300 1030,1130,1230 1030-1325 1031-1325 NS
06/10/87 6a 1000-1300 1000-1300 1000-1300 1000-1300 1100,1200,1300 1030-1300 1030-1311 1029-1310 NS
Inlet HC1 manual
results Invalidated.
CO spikes occurred
due to a blockage on
the feed table.
Difficulty in main-
taining high excess
air conditions.
Furnace draft was
unsteady and went
positive at times.
Some wet fuel was
burned.
No problems occurred.
Plugging of the
slaker strainer
caused erratic lime
slurry feedrates.
lmo/037
-------
TABLE 1-3. SUMMARY OF THE SAMPLING INTERVALS FOR CHARACTERIZATION TEST PROGRAM AT THE MARION COUNTY MWC (Continued)
(24-hour Clock Bails)
Ash
Manual Sampling
Data Run Inlet Midpoint Outlet
CEMS
Baghouse
Cyclone
Super-
heater
Economizer
Lime Slurry
or
Teslsorb
Comments
06/10/87 6b 1500-1734 1500-1734 1500-1734 1500-1734 1600, 1700 1530,1630,1730 1529-1730 1530-1729
I
K»
o
06/11/87 7 1400-1700 1400-1700 1400-1700 1400-1700 1500,1600,1700 1430-1700 1430-1701 1431-1700
06/12/87 8 1000-1300 1000-1300 1000-1300 1000-1300 1100,1200,1300 1030-1300 1030-1301 1030-1302
06/15/87 9 1430-1800 1430-1800 1430-1800 1430-1800 1530,1630,1730 1300,1400,1500 1459-1759 1500-1800
06/15/87 10 1230-1600 1230-1600 1230-1600 1230-1600 1330,1430,1530 1300,1400,1500 1200-1559 1259-1600
06/16/87 11» 1000-1300 1000-1300 1000-1300 1000-1300 1130,1230,1300 1030,1130,1230 1029-1258 1030-1259
06/16/87 lib 1430-1730 1430-1730 1430-1730 1430-1730 1530b,1630,1730 1500,1600,1700 1459-1730 1459-1729
NS Quench pit seal
broke causing CO
spike. Testing
aborted 1/2 hour
early due to
baghouse bypassing.
NS Difficulty in
maintaining quench
reactor outlet
temperature and
negative furnace
draft due to low
flue gas flowrates.
NS No problems
occurred.
LS-1605 Plant 02 data is
TS-1600 suspect due to
calibration
problems.
NS Inlet HC1 manual
results invalidated.
NS No problems
occurred.
NS No problems
occurred.
9A11 locations were sampled simultaneously.
Collected 35 gallon baghouse-ash sample for Ted Brna.
NS = sample not collected.
LS = lime slurry.
TS - Teslsorb.
lmo/037
-------
TABLE 1-4. SUMMARY OF SAMPLING AND ANALYTICAL PROCEDURES
Parameter
Sampling Method
Analytical Method
0 Inlet and Midpoint
0 Outlet
CO Inlet and Outlet
SO- Inlet
S0_ Midpoint and
Outlet
CO- Inlet, Midpoint
and Outlet
NO Inlet and Outlet
x
THC Inlet and Outlet
Baghouse Ash,
Cyclone Ash,
Economizer Ash, and
Superheater Ash
Tesisorb and
lime slurry
HC1
Moisture
Volumetric Flow Rate
Fixed gases (0_,
EPA Method 3A
EPA Method 3A
EPA Method 10
EPA Method 6C
EPA Method 6C
EPA Method 3A
EPA Method 7E
EPA Method 25A
Composited Grab Sample
Grab sample
EPA Method 5
(Modified)
EPA Method 4
EPA Methods 1 and 2
EPA Method 3
Thermox
Paramagnetic
Non-Dispersive Infrared (NDIR)
Spectrophotometric (UV range)
Pulsed Fluorescence
NDIR
Chemiluminescent
Flame lonization Detector (FID)
High resolution GC/MS
for CDD/CDF following
EPA/ASME Protocol
(Dec. 1984 draft)
High resolution GC/MS
for CDD/CDF following
EPA/ASME Protocol
(Dec. 1984 draft)
Specific Ion Electrode (SIE)
and Ion Chromatography (1C)
Orsat
1-11
-------
TABLE 1-5. CDD/CDF CONGENERS ANALYZED FOR
THE MARION COUNTY TEST PROGRAM
DIOXINS
Monochloro dibenzo-p-dioxin (MCDD)
Total dichlorinated dibenzo-p-dioxins (DCDD)
Total Trichlorinated dibenzo-p-dioxins (TrCDD)
2,3,7,8 Tetrachlorodibenzo-p-dioxin (2,3,7,8 TCDD)
Total Tetrachlorinated dibenzo-p-dioxins (TCDD)
1,2,3,7,8 Pentachlorodibenzo-p-dioxin (1,2,3,7,8 PCDD)
Total Pentachlorinated dibenzo-p-dioxins (PCDD)
1,2,3,4,7,8 Hexachlorodibenzo-p-dioxin (1,2,3,4,7,8 HxCDD)
1,2,3,6,7,8 Hexachlorodibenzo-p-dioxin (1,2,3,6,7,8 HxCDD)
1,2,3,7,8,9 Hexachlorodibenzo-p-dioxin (1,2,3,7,8,9 HxCDD)
Total Hexachlorinated dibenzo-p-dioxins (HxCDD)
1,2,3,4,6,7,8 Heptachlorodibenzo-p-dioxin (1,2,3,4,6,7,8 HpCDD)
Total Heptachlorinated dibenzo-p-dioxins (HpCDD)
Total Octachlorinated dibenzo-p-dioxins (OCDD)
FURANS
Monochloro dibenzofuran (MCDF)
Total dichlorinated dibenzofurans (DCDF)
Total Trichlorinated dibenzofurans (TrCDF)
2,3,7,8 Tetrachlorodibenzofurans (2,3,7,8 TCDF)
Total Tetrachlorinated dibenzofurans (TCDF)
1,2,3,7,8 Pentachlorodibenzofuran (1,2,3,7,8 PCDF)
2,3,4,7,8 Pentachlorodibenzofuran (2,3,4,7,8 PCDF)
Total Pentachlorinated dibenzofurans (PCDF)
1,2,3,4,7,8 Hexachlorodibenzofuran (1,2,3,4,7,8 HxCDF)
1,2,3,6,7,8 Hexachlorodibenzofuran (1,2,3,6,7,8 HxCDF)
1,2,3,7,8,9 Hexachlorodibenzofuran (1,2,3,7,8,9 HxCDF)
2,3,4,6,7,8 Hexachlorodibenzofuran (2,3,4,6,7,8 HxCDF)
Total Hexachlorinated dibenzofurans (HxCDF)
1,2,3,4,6,7,8 Heptachlorodibenzofuran (1,2,3,4,6,7,8 HpCDF)
1,2,3,4,7,8,9 Heptachlorodibenzofuran (1,2,3,4,7,8,9 HpCDF)
Total Heptachlorinated dibenzofurans (HpCDF)
Total Octachlorinated dibenzofurans (OCDF)
1-12
lmo/037
-------
EPA Advisor Functions
Ray Klicius
Bob McCaig
EPA Project Coordinators
Ted Brna*
Peter Schindler*
James Kilgroe*
MRI
Process Monitoring
Steve Schliesser*
OAQPS Coordinator
Peter Schindler*
AEERL Coordinator
Ted Brna*
QA Plan
nsberger*
OAQPS Task Manager
Gene
Riley
Radian Test Crew
Emission Testing
Winton
Kelly*
Entropy 1
HCIT
Phil Ju
Odgen Martin Coordinators
Jeffrey Hahn - — -
David Sussman
Fred Engelhardt
Ogden Martin
Program Coord nator
Jeffrey Hahn
rew Ogden Martin
1 Test Crew
* Henry Von Demfange*
ODEQ
Wendy Sims
AEERL - Air and Energy Engineering Research Laboratory
OAQPS - Office of Air Quality Planning and Standards
ODEQ - Oregon Department of Enviromental Quality
MRI - Midwest Research Institute
* - On-Site
Figure 1-2. Marion County Characterization Test Program
Line of Communication
oc
in
m
en
-------
equipment operating conditions were suitable for testing. While on-site, any
changes or problems were discussed between EPA, Oregon DEQ and Ogden Martin
and agreed upon (with input from the test crew chiefs) before a change was
made to the test program protocol.
1.5 QUALITY ASSURANCE/QUALITY CONTROL (QA/QC)
The test program was designed and executed with emphasis on completeness
and data quality. A comprehensive internal quality assurance (QA) and quality
control (QC) program was an integral part of Radian's test program. The goal
of the QA/QC effort was to ensure that the data collected were of known
precision and accuracy and that they were complete, representative and
comparable. Data comparability was achieved by using standard units of
measure as specified in the methods.
In addition to Radian's internal QC program, an independent performance
and systems audit was conducted by Entropy Environmentalists, Inc., and is
reported separately. The independent audit was conducted during 2 days prior
to the start of the test program, during 1 day at the middle of the test
program and during 1 day at the conclusion of the test program, as well as
periodically during the testing.
1.6 DESCRIPTION OF REPORT SECTIONS
The remaining sections of this volume are organized as follows:
Section 2.0 Summary of Results
Section 3.0 Conclusions and Recommendations
Section 4.0 Description of Process Operation
Section 5.0 Sampling Locations
Section 6.0 Sampling and Analytical Procedures
Section 7.0 Internal Quality Assurance/Quality Control
Section 8.0 References
Section 9.0 Metric-to-English Conversion Table
lmo/036 1-14
-------
The supporting data and calculations for the results presented in Volume I are
included in Volumes II to VI. Volume II contains a summary of the test
results, which includes 1-minute plots of selected variables. Volume III
includes the printouts of 1-minute averages for the GEM parameters and the
Method 5 results for the manual HC1 sampling train. Volume IV includes copies
of all the field data sheets. The analytical reports and test logs are
included in Appendix V. Appendix VI includes the QA/QC results, the summary
of equipment used, sample calculations, sampling and analytical protocols,
pertinent correspondence and project participants.
lmo/036 1-15
-------
2.0 SUMMARY OF RESULTS
The results of the characterization test program conducted at the Marion
County Solid Waste-to-Energy Facility are presented in this section. The
baseline, combustor variation and control device variation results are
summarized in Tables 2-1 and 2-2. These tables include the results of the
greatest interest. The results represent an average value of a parameter over
a sampling period. The HC1 reduction efficiency based on GEM data across the
total emission control system (cyclone, quench reactor, and fabric filter)
ranged from 75.8 to 98.4 percent. Controlled HC1 emissions ranged from 11.5
to 214 ppmv, dry, normalized to 12 percent CO-. Control system removal
efficiencies for SO- ranged from zero to 92.5 percent with controlled S0?
emissions ranging from 9.9 to 484 ppmv, dry, normalized to 12 percent C09.
The NO emissions ranged from 184 to 310, ppmv, dry, normalized to 12 percent
X
CO- and NO removal was not observed across the control system.
^ X
Non-condensible THC emissions ranged from 0.6 to 2.4 ppmv, as propane, dry, .
normalized to 12 percent CO,, and also were not reduced by the control system.
The CO emissions ranged from 2.2 to 17 ppmv, dry normalized to 12 percent CO^.
Excess air ranged from 36.2 to 144 percent.
Additional results and discussion are provided in the following sections.
Baseline results are presented first in Section 2.1. Then, the results of the
combustor evaluation and off-design temperature control system evaluation are
presented in Sections 2.2 and 2.3, respectively. The results are presented in
each subsection according to the following scheme: acid gases that include HC1
and SO-; temperature profile of the system; combustion parameters that include
steam load, excess air, and combustion efficiency; fixed gases that include
CO, C00 and 0 • additional pollutants that include NO and THC; and CDD/CDF
^ z~ x
concentrations in the superheater, economizer, cyclone and baghouse ash. The
supporting data and example calculations for the results presented are
included in the appendices.
English and metric units are used to present the results. Typically,
results of the sampling parameters (such as volumetric flowrate) are presented
in English units and concentrations of pollutants are reported in metric
units. Metric units are preferable for reporting the relatively low
lmo/036
2-1
-------
TABLE 2-1. SUMMARY OF BASELINE AND COMBUSTOR VARIATION RESULTS FOR THE MARION COUNTY MWC
NJ
I
M
TEST CONDITION
HCl REDUCTION EFFICIENCY (X)*
QUENCH REACTOR
TOTAL SYSTEM
SO REDUCTION EFFICIENCY (X)
QUENCH REACTOR
TOTAL SYSTEM
STOICHIOMETRIC RATIO
HCl EMISSIONS (ppmv at 12 X C02)*'b
INLET
MIDPOINT
OUTLET
SO EMISSIONS (pprav »t 12 X C02)b
INLET
MIDPOINT
OUTLET
NOx EMISSIONS (pprav «t 12 X CO2)b'°
THC EMISSIONS (ppmv »t 12 X C02)b/d
CO EMISSIONS (ppmv »t 12 X CO2)b
EXCESS AIR 0«>f
STEAM LOAD (Ib/hr)
TEMPERATURES (deg. F)8
MIDDLE OF COMBUSTOR, FIRST PASS
TOP OF COMBUSTOR, FIRST PASS
QUENCH REACTOR OUTLET
STACK OPACITY (X)
TOTAL CDD/CDF ASH RESULTS (ng/g)
SUPERHEATER
ECONOMIZER
CYCLONE
BAGHOUSE
1
BASELINE
64.4
85.9
17.5
25.3
1.08
646.2
224.7
83.7
558.9
449.7
383.1
305.1
0.9
11.5
73.3
67180
1666
1665
300
0
NS*
NS
NA
NA
2
BASELINE
70.2
94.9
55.9
69.2
1,33
631.0
183.3
35.0
298.9
128.4
99.5
285.4
0.6
11.2
71.1
67240
1708
1688
300
0
NS6
NS
NA
NA
3A
LOW LOAD
LOW AIR
61.6
90.2
26.0
57.6
1.26
495.6
176.6
49.9
428.4
294.5
185. 5
199.5
0.7
5.1
36.2
63990
1895
1771
300
2.9
NS"
NA
NA
NA
3B
HIGH AIR
73.9
92.5
23.1
49.7
1.07
703.8
160.4
47.7
522.7
351.2
236.9
310.2
0.6
17.0
99.5
63940
1572
1561
299
1.1
NS"
1.47
2.76
12.2
4
LOW 0/F
76.1
98.4
74.5
92.5
2.22
647.8
161.5
11.5
120.2
31.9
9.9
221.4
0.6
13.2
70.1
65460
1731
1694
301
1.1
3.70
46.5
2.60
12.9
5
HIGH O/F
84.8
93.7
41. 7
62.4
1.14
728.8
110.2
45.5
425.0
246.6
157.8
274.2
NR"
7.9
68.9
68970
1808
1734
299
1.0
NA"
NA
NA
NA
6A
LOW LOAD
72.1
91.2
39.1
52.4
1.40
693.1
225.0
69.6
339.6
240.5
184.6
256.4
1.9
2.0
70.1
51230
1767
1675
302
3.7
16.4
5.68
2.34
13.1
6B
LOW LOAD
HIGH AIR
84.8
95.6
55.7
80.7
2.24
624.8
92.5
27.4
275.2
118.8
52.5
233.3
1.5
16.9
144.1
47960
1490
1417
300
10.9
3.77
1.75
1.54
13.4
7
LOW LOAD
LOW AIR
67.2
90.2
29.8
53.0
1.62
652.8
224.7
67.2
281.1
206.8
139.2
190.6
NR8
2.2
57.9
51590
1885
1766
288
1.0
•
NA
NA
NA
NA
8
LOW LOAD
LOW O/F
67.2
93.0
24.7
58.2
2.50
568.2
185.9
39.5
210.1
157.6
87.5
183.6
1.6
6.9
85.8
49900
1733
1668
298
1.0
•
NA
NA
NA
NA
9
LOW LOAD
HIGH O/F
70.0
96.9
77.3
87.1
2.36
641.7
203.5
19.7
167.5
40.3
21.3
276.1
1.5
10.7
90.8
52090
1639
1578
299
1.2
*
NA
NA
NA
NA
*HCl reduction efficiencies and concentrations are based on CEM data.
All flue gas emissions are reported on a dry basis.
CNO data are reported for the Inlet only, since emissions were not affected by the control device.
, x
THC results are reported for the outlet for Runs 1-4 and for the inlet for Runs 6A, 6B, 8, and 9. The data were Invalidated due to instrument
malfunctions for both the inlet and outlet instruments for Runs 5 and 7.
*NA - Sample was collected but not analyzed. NR - Data set Invalidated due to Instrument malfunction. NS = Sample was not collected.
Percent excess air Is based on Inlet CEM data. ,
ps wee*1 measured
alibrated t he rinocoup J.e s . Accuracy of reported temperatures Is uncertain.
-------
TABLE 2-2. SUMMARY OF BASELINE AND CONTROL DEVICE VARIATION RESULTS FOR THE MARION COUNTY MWC
TEST CONDITION: 1
BASELINE
a
HCL REDUCTION EFFICIENCY (Z)
QUENCH REACTOR
TOTAL SYSTEM
S02 REDUCTION EFFICIENCY (Z)
QUENCH REACTOR
TOTAL SYSTEM
STOICHIOMETRIC RATIO
a,b
HCL EMISSIONS (pptnV at 12 Z C02)
INLET
MIDPOINT
OUTLET
b
S02 EMISSIONS (ppmV at 12 Z C02)
INLET
MIDPOINT
OUTLET
b,d
NOx EMISSIONS ( ppmV at 12 Z C02)
b,e
THC EMISSIONS (ppmV at 12 Z C02)
b
CO EMISSIONS (ppmV at 12 Z C02)
f
EXCESS AIR (Z)
STEAM LOAD (Ib/hr)
g
TEMPERATURES (deg. F)
MIDDLE OF COMBUSTOR
TOP OF COMBUSTOR, FIRST PASS
QUENCH REACTOR OUTLET
STACK OPACITY (Z)
TOTAL CDD/CDF ASH RESULTS (ng/g)
SUPERHEATER
ECONOMIZER
CYCLONE
BAGHOUSE
64.4
85.9
17.5
25.3
1.08
646.2
224.7
83.7
558.9
449.7
383.1
305.1
0.9
11.5
73.3
67180
1666
1665
300
0
h
NS
NS
NA
NA
2 10
BASELINE LOW TEMP
70.2
94.9
55.9
69.2
1.33
631.0
183.3
35.0
298.9
128.4
99.5
285.4
0.6
11.2
71.1
67240
1708
1688
300
0
h
NS
NS
NA
NA
78.8
97.6
18.2
72.9
1.14
814.4
180.0
20.4
382.6
325.8
108.0
265.2
2.3
10.8
79.7
67120
1784
1618
262
1.1
h
NA
NA
NA
5.11
11A 11B
HIGH TEMP HIGH TEMP
57.6
78.9
c
-14.6
0.9
1.06
718.2
295.4
157.7
470.2
522.5
484.4
247.0
2.4
8.9
72.1
67770
1827
1775
330
1.0
h
NA
NA
NA
6.69
61.7
75.8
c
-37.9
-18. 3C
1.59
750.0
313.1
213.8
118.0
177.5
164.5
261.9
1.7
14.5
77.5
66100
1745
1719
360
1.0
3.71
7.69
2.14
10.2
HC1 reduction efficiencies and concentrations are based on CEM data.
b
All flue gas emissions are reported on a dry basis.
c
Considering the accuracy of the instruments (calibrated at a large span but measuring low
concentrations), the results are considered equivalent. The data indicate that no real removal
of S02 occurred. The apparent negative removal efficiencies can be considered equivalent to zero.
d
NOx data are reported for the inlet only, since emissions were not affected by the control device.
e
THC results are for the outlet for Runs 1-2 and the inlet for Runs 10-11B.
f
Excess air based on inlet CEM data.
g
Combustor temperatures were measured using uncalibrated thermocouples. Accuracy of reported
temperatures is uncertain.
h
NA * Sample was collected but not analyzed. NS = Sample was not collected.
g
2-3
-------
concentrations that were measured. For the reader's ease, a Metrie-to-
English conversion table is included in Section 9.0.
2.1 BASELINE EMISSIONS
2.1.1 Baseline Acid Gas Emissions
The primary acid gases of interest for the characterization test program
were HC1 and SCL. Baseline acid gas concentrations and control efficiencies
are presented in Table 2-3.
Baseline uncontrolled SO., concentrations were measured during Runs 1 and
2, as well as Runs 10, 11A and 11B. The average uncontrolled SO,,
concentration for baseline operation was 366 ppmV, dry, normalized to 12
percent CO,, with a relative standard deviation of 46 percent. The
uncontrolled SO- concentrations ranged from 118 to 559 pptnV, dry, normalized
to 12 percent C0?. The significant variations in uncontrolled SO,, emissions
are a result of changes in fuel composition, since combustor conditions were
equivalent for these runs.
Baseline S0» concentrations after the quench reactor were 450 ppmV, dry
normalized to 12 percent CO- for Run 1 and 128 ppmV, dry, normalized to
12 percent CO,,. The average SO., concentration after the quench reactor was
289 ppmV, dry, normalized to 12 percent CO,,.
Baseline controlled SO,, concentrations ranged from 383 ppmV, dry
normalized to 12 percent CO,, for Run 1 to 99.5 ppmV, dry, normalized to
12 percent CO- for Run 2. The average controlled S0? concentration was
241 ppmV dry normalized to 12 percent C0«.
Baseline removal efficiency for S0? across the control device system was
25.3 percent during Run 1 and 69.2 percent during Run 2. The average baseline
removal efficiency across the control device was 47.2 percent. The quench
reactor removal efficiency for Run 1 was 17.5 percent and 55.9 percent for
Run 2. The average baseline quench reactor removal efficiency was
36.7 percent. The fabric filter reduced the SO- mass flowrate an additional
9.4 percent during Run 1 and 30.1 percent during Run 2.
lmo/036
2-4
-------
TABLE 2-3. SUMMARY OF BASELINE ACID CASES ADD COHTROL EFFICIENCIES
TEST CONDITIOH
INLET SO2< ppnw, dry
INLET SO2> Ib/hr
INLET HCl, ppow, dry*
INLET HCl, Ib/hr*
STOICHIOMETRIC RATIO
INLET S02, ppow S12X CO2
MIDPOINT SO2> ppow «12X C02
OUTLET S02< ppow 812X O02
INLET HCl, MANUAL, ppow 812X C02
MIDPOINT HCl, MANUAL, ppow 812X CO2
OUTLET HCL, MANUAL, ppow S12X C»2
INLET HCl, CEM, ppow 812X CO
MIDPOINT HCl, CEM, ppow 812X CO2
OUTLET HCl, CEM, ppow 812X C02
QUENCH REACTOR EFFICIENCY
PERCENT SO REDaCTIOH
PERCENT HCl REDUCTION, CEM
PERCENT HCl REDUCTION, MANUAL
FABRIC FILTER EFFICIENCY
PERCENT S02 REDUCTION
PERCENT HCl REDUCTION, CEM
PERCENT HCl REDUCTION, MANUAL
OVERALL SYSTEM EFFICIENCY
PERCENT SO REDUCTION
PERCENT HCl REDUCTION, CEM
PERCENT HCl REDUCTION, MANUAL
1
BASE-
LINE
484
147
480
83.0
1.08
559
450
383
462
177
NR
646
225
83.7
17.5
64.4
60.7
9.4
60.4
NR
25.3
85.9
NR
2
BASE-
LINE
274
87.4
519
94.3
1.33
299
128
99.5
502
222
37.6
631
183
35.0
55.9
70.2
54.5
30.1
82.8
84.7
69.2
94.9
93.1
10
LOU OR
OUT T.
328
99.9
699
60.5
1.14
383
326
108
NR
229
23.4
814
180
20.4
18.2
78.8
NR
66.8
88.6
89.8
72.9
97.6
NR
11A
HIGH OR
OUT T.
415
125
646
111
1.06
470
523
484
745
408
172
718
295
158
-14.6"
57.6
43.6
13.5
50.2
60.6
0.9
78.9
77.8
11B
HIGH QR
OUT T.
108
35.5
695
130
1.59
118
178
165
767
545
228
750
313
214
-37.9b
61.7
34.8
14.2
36.8
61.3
-18.3b
75.8
74.8
BASELINE
AVERAGE
379
117
500
88.7
1.21
429
289
241
482
200
18.8
639
204
59.3
36.7
67.3
57.6
19.7
71.6
42.4
47.2
90.4
46.5
OVERALL
AVERAGE
322
99
608
95.7
NA
366
NA
NA
619
NA
NA
712
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
PERCENT
RSD
44.8
42.8
16.8
27.6
NA
46
NA
NA
26
NA
NA
11
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Note: All values are reported on a dry basis.
NA - Not applicable.
NR - Not reported due to Invalidation.
Average of CEM and manual results.
b
Instruttent Inaccuracies because of measuring low concentrations while calibrated with a large span and differences between
Individual analyzers are responsible for the differences In SO concentration at the three locations. These values should be
considered equivalent and Indicate that no significant removal of SO took place during these runs.
2-5
-------
Baseline uncontrolled HC1 concentrations were also measured during
Runs 1, 2, 10, 11A and 11B. The average uncontrolled HC1 concentration for
baseline operation was 712 ppmV, dry, normalized to 12 percent C09 based on
CEM data and 619 ppmV, dry, normalized to 12 percent CO. based on the manual
method data. The manual method uncontrolled value for Run 10 was invalidated
because of a low moisture value and is not included in the average.
Baseline HC1 concentrations after the quench reactor ranged from 225 to
183 ppmV, dry, normalized to 12 percent CO,, for Runs 1 and 2, respectively,
based on CEM data. The average HC1 concentration after the quench reactor was
204 ppmV, dry, normalized to 12 percent CO,,.
The average controlled HCl concentration for baseline conditions was
59.3 ppmV, dry normalized to 12 percent C0_ based on CEM data. The HCl
concentration ranged from 83.7 to 35.0 ppmV, dry, normalized to 12 percent C0»
for Runs 1 and 2, respectively.
Average baseline control device HCl removal efficiencies were
90.4 percent for the continuous monitoring testing and 93.1 percent
(Run 2 only) for the manual methods tests. Quench reactor HCl removal
efficiency averaged 67.3 percent by CEM measurement and 57.6 percent by manual
measurement. Fabric filter HCl removal efficiency baseline averages were
71.6 percent by CEM and 84.7 percent for manual method Test 2.
The average stoichiometric ratio for the baseline conditions was 1.21.
The stoichiometric ratio for Run 1 was 1.08 due to the high S0_ concentrations
and 1.33 for Run 2, which had lower S0_ concentrations. Stoichiometric ratio
is the molar ratio of the actual calcium supplied by the quench reactor to the
theoretical calcium required to react with the inlet SO. and HCl.
2.1.2 Temperature Profile for Baseline Conditions _
The temperature of the flue gas was monitored at eleven points in the MWC
system beginning with the combustion air and ending at the breeching to the
lmo/036
2-6
-------
outlet stack. The thermocouples used to measure the furnace temperature were
uncalibrated, making the accuracy of the reported values uncertain. The
points are shown in Figure 2-1. The results for the baseline conditions are
summarized in Table 2-4. In addition to Runs 1 and 2, Runs 10, 11A and 11B
are considered baseline for temperatures through the quench reactor inlet. An
average baseline value and standard deviation were calculated for each
location. The standard deviation at each location was less than five percent
of the average indicating that conditions were similar during the baseline
tests.
The combustion air was preheated to an average temperature of 236°F. The
temperature achieved at the middle of the first pass of the combustor was
1746°F. At the economizer outlet the temperature was reduced to 423°F. The
quench reactor reduced the temperature to an average of 300 F.
2.1.3 Combustion Parameters and Combustion Efficiency
The primary indicators of combustion conditions are discussed in this
section and include steam load, excess air, combustion efficiency, CO
concentration and volumetric flowrate. Additional parameters are reported in
Appendix A. The results for the combustor baseline test conditions are
summarized in Table 2-5. In addition to Runs 1 and 2, Runs 10, 11A and 11B
are included as combustor baseline test runs.
During the baseline test runs, the average steam flowrate was 67082 Ib/hr
and the relative standard deviation was 0.9 percent. The target steam load
was 66,400 Ib/hr. The average excess air during baseline testing was 74.7
percent with a relative standard deviation of 5 percent. The target excess
air level was 70 percent. Excess air was measured at the combustor outlet
(control device inlet).
Combustion efficiency was calculated based on the ratio of moles of CO to
moles of CO and C09 measured at the combustor outlet. During the baseline
test runs, the combustion efficiency ranged from 99.88 to 99.93 percent with
an average of 99.9 percent. The CO concentration ranged from 13.3 to
7.9 ppmv, dry, respectively.
lmo/036
2-7
-------
To Atmosphere
I . I Quench Reactor'
Acid Gas
Scrubber
Tealsorb
Feed
Hopper
i
NJ
00
Quench
Pit
—Dlatrlbutor
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Combustion air
Middle of furnace, 1st pass
Top of furnace, 1 st pass
Economizer outlet
Inlet sampling location
Quench reactor inlet
Midpoint sampling location
Quench reactor outlet
Baghouse outlet
I.D. fan inlet
Breeching to outlet stack
Figure 2-1. Location of Temperature Indicators for the Marion County MWC
CO
tn
-------
TABLE 2-4. BASELINE TEMPERATURE PROFILE FOR THE MARION COUNTY MWC
LOCATION
CODE
1
2
3
4
5
6
7
8
9
10
11
TEMPERATURES (deg. F)
COMBUSTION AIR
MIDDLE OF FURNACE, FIRST PASS
TOP OF FURNACE, FIRST PASS
ECONOMIZER FLUE GAS OUTLET
INLET SAMPLING LOCATION
QUENCH REACTOR INLET
MIDPOINT SAMPLING LOCATION
QUENCH REACTOR OUTLET
BAGHOUSE OUTLET
ID FAN INLET
BREECHING TO OUTLET STACK
1
COMBUSTOR
BASELINE
245
1666
1665
417
420
432
298
300
280
285
291
2
COMBUSTOR
BASELINE
241
1708
1688
409
434
430
296
300
280
283
277
10
COMBUSTOR
BASELINE
230
1784
1618
424
441
436
281
262
252
254
260
11A
COMBUSTOR
BASELINE
232
1827
1775
422
433
434
310
330
302
305
303
11B
COMBUSTOR
BASELINE
230
1745
1719
445
465
461
362a
360
334
334
352
BASELINE
AVERAGE
236
1746
1693
423
439
439
297b
300
280
284
284
STANDARD
DEVIATION
(*)
2.9
3.6
3.5
3.2
3.8
2.9
0.7C
0.0
0.0
0.7
4.9
t-o
RUN 10 WAS A CONTROL DEVICE VARIATION RUN WITH A LOW QUENCH REACTOR TEMPERATURE.
RUNS 11A AND 11B WERE CONTROL DEVICE VARIATION RUNS WITH A HIGH QUENCH REACTOR TEMPERATURE.
RUNS 10, 11A AND 11B ARE CONSIDERED BASELINE FOR THE COMBUSTOR TEMPERATURES.
3
ONLY RUNS 1 AND 2 ARE USED TO CALCULATE THE COMBUSTOR BASELINE AVERAGE BELOW THIS POINT.
c
DIFFERENCE IS CALCULATED BASED ON RUNS 1 AND 2 ONLY.
DIFFERENCE - [RUN 1 - RUN 2]/[(0.5)*(RUN 1 + RUN 2)1* 100%
-------
TABLE 2-5. BASELINE COMBUSTION PARAMETERS FOR THE MARION COUNTY MWC
COMBUSTION PARAMETER
STEAM FLOW (Ib/hr)
a
EXCESS AIR (PERCENT)
a
CO CONCENTRATION (ppmv, DRY)
a
C02 CONCENTRATION (% by vol. DRY)
a
02 CONCENTRATION (% by vol. DRY)
a,b
COMBUSTION EFFICIENCY (%)
c
VOLUMETRIC FLOWRATE (ACFM)
1
COMBUSTOR
BASELINE
67180
73.3
10.0
10.4
9.0
99.90
57150
2
COMBUSTOR
BASELINE
67240
71.1
10.3
11.0
8.8
99.91
60920
10
COMBUSTOR
BASELINE
67120
79.7
9.3
10.3.
9.4
99.91
59100
11A
COMBUSTOR
BASELINE
67770
72.1
7.9
10.6
8.9
99.93
57430
11B
COMBUSTOR
BASELINE
66100
77.5
13.3
11
9.2
99.88
65620
.........
BASELINE
AVERAGE
67082
74.7
10.2
10.7
9.1
99.90
60044
STANDARD
DEVIATION
0.9
4.9
19.5
3.1
2.7
0.02
3463
N)
I
Measured at the inlet sampling location baaed on GEM data.
3
Combustion efficiency » moles of CO/[moles of CO * moles of C02]*100Z
Measured at the inlet sampling location.
-------
Volumetric flowrate as measured at the combustor inlet was an average of
60,044 acfm. The flowrate ranged from 57,150 to 65,620 acfm with relative
standard deviation of 6 percent.
2.1.4 Fixed Gases (CO. CO 0 )
Fixed gas concentrations at baseline are presented in Table 2-6. During
baseline conditions (Runs 1 and 2) carbon dioxide concentration in the boiler
outlet (control device inlet) flue gas averaged 10.1 percent by volume, dry,
by EPA Method 3 (Orsat analysis) and 10.7 percent by volume, dry, by GEM
instrument analysis for a pooled average of 10.4 percent by volume, dry.
Oxygen analyses yielded an average concentration of 9.2 percent by volume,
dry, by Orsat and 8.9 percent by volume, dry, by GEM, with a pooled average of
9.1 percent by volume, dry. Control device evaluation test runs can also be
considered as baseline test conditions for the boiler outlet since no furnace
parameters were abnormal. A pooled result incorporating ORSAT and CEM values
for Runs 1, 2, 10, 11A and 11B yielded an average of 9.1 percent by volume 0_,
dry, (RSD - 2.7 percent) and 10.4 percent by volume C0~, dry, (RSD -
4.0 percent). The relative standard deviations for these pooled averages are
excellent and indicate consistency in the process operation and the sample
analyses.
Carbon monoxide concentrations at the boiler outlet during baseline test
conditions were 10.0 ppmv, dry, (Run 1) and 10.3 ppmv, dry, (Run 2) for an
average baseline concentration of 10.2 ppmv, dry. The average baseline boiler
outlet CO concentration including the control device evaluation runs was
10.2 ppmv, dry, with a standard deviation of 1.98 (RSD = 19.5 percent).
Although there was some variability in carbon monoxide concentration, the CO
levels were consistently below 20 ppmV, dry basis.
i 2.1.5 Additional Pollutants of Interest (NO and THC)
! Baseline concentrations for NO and THC are presented in Table 2-7.
x r
I Baseline uncontrolled NO concentrations were 305 ppm normalized to 12 percent
j x
I CO for Run 1 and 285 ppm normalized to 12 percent CO for Run 2. This gives
i
x
lmo/036
an average N0_ baseline concentration of 295 ppm normalized to 12 percent CO..
2-11
-------
TABLE 2-6. BASELINE FIXED GASES CONCENTRATIONS (CO, CO-, 0?)
N5
I
1
TEST CONDITION BASELINE
INLET
o2,
co2,
ORSAT
%v,DRY
%v,DRY
PERCENT EXCESS AIR
Foa
INLET
0 ,
co2,
CO,
CEM
%v, DRY
%v,DRY
ppmv.DRY
PERCENT EXCESS AIR
Foa
INLET
co'
AVERAGEb
%V, DRY
, %V, DRY
PERCENT EXCESS AIR
Foa
9.0
10.0
72.7
1.19
9.0
10.4
10.0
73.3
1.14
9.0
10.2
73.0
1.17
2
BASELINE
9.4
10.1
78.4
1.15
8.8
11.0
10.3
71.1
1.10
9.1
10.6
74.8
1.13
10
LOW QR
OUT T.
8.8
10.2
69.9
1.19
9.4
10.3
9.3
79.7
1.12
9.1
10.3
75.8
1.16
11A
HIGH QR
OUT T.
9.0
10.2
73.7
1.17
8.9
10.6
7.9
72.1
1.13
9.0
10.4
72.9
1.15
11B
HIGH QR
OUT T.
9.4
9.7
78.6
1.19
9.2
11.0
13.3
77.5
1.06
9.3
10.4
78.1
1.13
BASELINE
AVERAGE
9.2
10.1
75.6
1.17
8.9
10.7
10.2
72.2
1.12
9.1
10.4
73.9
1.15
OVERALL
AVERAGE
9.1
10.0
74.7
1.18
9.1
10.7
10.2
74.7
1.11
9.1
10.4
74.7
1.14
PERCENT
RSD
2.9
2.1
5.1
1.52
2.7
3.1
19.5
4.9
2.83
2.7C
4.0C
4.7°
3.8C
aFo = (20.9 - %0 dry)/(%CO dry).
*
Average of Orsat and CEM values.
°Relative standard deviation based on all data points (CEM and Orsat).
-------
TABLE 2-7. NOx AND THC EMISSIONS FOR BASELINE CONDITIONS
to
i
TEST CONDITION:
1
BASE-
LINE
2 10 11A 11B
BASE- LOW QR HIGH QR HIGH QR BASELINE OVERALL PERCENT
LINE OUT T. OUT T. OUT T. AVERAGE AVERAGE RSD
INLET
NOx, pprav, DRY
NOx, ppmv @12* C02
NOx, Ib/hr
OUTLET
NOx, ppmv, DRY
NOx, ppmv @12% C02
NOx, Ib/hr
264
305
57.7
204
306
63.0
262
285
59.9
205
304
59.2
228
265
49.7
194
287
51.8
218
247
47.3
165
250
46.1
240
262
56.5
196
302
55.2
263
295
59
205
305
61
242
273
54
193
290
55
8.4
8.3
10.0
8.5
8.1
11.9
INLET
THC, ppmv as propane, DRY
THC, ppmv as propane @12% C02
THC, Ib/hr as propane
OUTLET
THC, ppmv as propane, DRY
THC, ppmv as propane @12% C02
THC, Ib/hr as propane
NR
NR
NR
0.6
0.9
0.2
0.7
0.8
0.2
0.4
0.6
0.1
2.0
2.3
0.4
NR
NR
NR
2.1
2.4
0.4
NR
NR
NR
1.6
1.7
0.4
NR
NR
NR
0.7
0.8
0.2
0.5
0.7
0.1
1.6
1.8
0.3
NA
NA
NA
39.9
41.6
37.9
NA
NA
NA
NR = Not reported due to invalidation or reading not not taken.
NA » Not applicable.
-------
Controlled baseline NO concentrations were 306 ppm normalized to
X
12 percent C0_ and 304 ppm normalized to 12 percent CO, for Run 1 and Run 2,
respectively. The average control device outlet NO concentrations was
X
305 ppm normalized to 12 percent CO-. Outlet mass flowrates for Runs 1 and 2
were 63.0 Ib/hr and 59.2 Ib/hr, respectively.
The consistency of normalized NO concentrations from inlet to outlet
indicates that the control device did not reduce the concentrations of
nitrogen oxides significantly. Observed differences between inlet and outlet
NO concentrations are within expected instrument variability.
X
Baseline total non-condensible hydrocarbon emissions for the Marion
County Solid Waste-to-Energy Facility were less than 1 ppm as propane, which
is a concentration close to the instrument's detection limit. The outlet
concentrations of the THC were 0.9 ppmV for Run 1 and 0.6 ppmV for Run 2. The
inlet THC monitor detected an average of 0.8 ppm (dry) for Run 2. (All values
are normalized to 12 percent CO-.) The control device does not appear to
affect the normalized concentration of THC.
2.1.6 CDD/CDF Concentrations in Ash
Baseline CDD/CDF concentrations in superheater, economizer, cyclone and
baghouse ashes are shown in Table 2-8. The total CDD, total CDF, and total
2378-TCDD toxic equivalent concentrations are given for the four ash sampling
locations. The baseline superheater ash and economizer ash results are from
Run 11B samples. Baseline results for the cyclone ash and baghouse ash are
7 8
averages from previous studies in February 1987 and September 1986 at the
Marion County MWC. There are no simultaneously collected baseline results for
the four ash sampling locations because there were difficulties with the
sampling technique for the economizer ash and superheater ash when the system
was at baseline conditions (Runs 1 and 2). During Run 11B, only the combustor
was at baseline conditions, so only the superheater ash and economizer ash
were indicative of baseline. In the previous studies at the Marion County
MWC, the system was at baseline, but economizer ash and superheater ash
samples were not collected.
lmo/036
-------
TABLE 2-8. CDD AND CDF CONCENTRATIONS AND 2378-TCDD TOXIC EQUIVALENCIES
FOR ASH FROM BASELINE CONDITIONS AT MARION COUNTY MWC
Ash Type
Superheater Ash
Economizer Ash
Cyclone Ash
Baghouse Ash
Superheater Ash
Economizer Ash
Cyclone Ash
Baghouse Ash
Run 11B
0.400
0.710
NB
NB
3.31
6.98
NB
NB
a b
Method Study Emission Test
Average Average
TOTAL CDD CONCENTRATION (ng/g)
NCd NC
NC NC
1.65 3.58
4.63 4.56
TOTAL CDF CONCENTRATION (ng/g)
NC NC
NC NC
2.88 1.22
11.4 3.38
2378-TCDD TOXIC EQUIVALENT CONCENTRATION
Baseline
Average
0.400
0.710
2.81
4.59
3.31
6.98
1.88-
6.58
(ng/g)
Superheater Ash
Economizer Ash
Cyclone Ash
Baghouse Ash
0.030 NC NC 0.030
0.085 NC NC 0.085
NB 0.060 0.082 0.074
NB 0.159 0.141 0.148
Average ash concentration from Runs 1 and 5 of February 1987 Method Study at
Marion County MWC.
b
Average ash concentration from Runs 1-3 of September 1986 Emission Test at
Marion County MWC.
c
Baseline average -is Run 11B results for superheater ash and economizer ash
and average of Method Study and Emission Test results for cyclone ash
and baghouse ash
d
NC =• Not collected. Sample not collected during this test.
NB = Not baseline. Quench reactor outlet temperature was varied in Run 11B,
so ash collected at the cyclone and baghouse was not from baseline conditions.
2-15
-------
The baseline total CDD average concentration ranged from 0.40 ng/g for
the superheater ash to 4.59 ng/g for the baghouse ash. The baseline total CDF
average concentration ranged from 3.31 ng/g for superheater ash to 6.98 ng/g
for economizer ash. The baseline 2378-TCDD toxic equivalency concentration
ranged from 0.03 ng/g for the superheater ash to 0.15 ng/g for the baghouse
ash.
The concentrations of the individual CDD/CDF species are presented in
Table 2-9. The baseline concentrations for the superheater ash and economizer
ash are from Run 11B. The concentrations include the confirmation results for
2378-TCDF and, if less interference occurred, for 2378-TCDD. The
concentrations for the cyclone ash and baghouse ash are averages of the
Emission Test and Method Study results. The results for each run of the
Emission Test and Method Study are given in References 7 and 8, respectively.
In Figure 2-2, the homologue distributions for CDD, and CDF at baseline
conditions are shown. This is based on Run 11B distributions for economizer
ash and superheater and an average distribution from the previous studies for
the superheater ash and cyclone ash. The distribution does not appear to
change significantly across the different sampling locations.
2.2 COMBUSTOR VARIATIONS
2.2.1 Temperature Profile During Combustor Variations
The temperature profile results for the combustor variation conditions
are presented in Table 2-10. The difference from the baseline average for
each of the combustor variation conditions is presented in Table 2-11 and
graphically in Figures 2-3 and 2-4. Figure 2-3 presents the differences for
all the low load conditions and Figure 2-4 presents the difference for the air
distribution and excess air conditions.
The temperatures monitored varied less than 20 percent from baseline.
Low load, high excess air (Run 6B) and low load, low excess air (Run 7) showed
the most change in the temperature profile.
lmo/036 2-16
-------
TABLE 2-9.
CDD AND CDF CONCENTRATIONS FOR ASH AT BASELINE
CONDITIONS AT MARION COUNTY MHC
CDD/CDF CONCENTRATION (ng/g)
Ash Type
Isomer
DIOXINS
Mono -CDD
Dl-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono-CDF
DL-CDF
Trl-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD/CDF
Superheater
(0.001)°
[0.009]
[0.054]
0.008
0.049
[0.006]
0.017
[0.003]
[0.004]
[0.011]
0.025
0.061
0.049
0.191
0.400
(0.001)
[0.240]
1.04
0.070
0.952
0.066
0.052
0.309
0.063
0.040
0.045
0.004
0.159
0.112
0.041
0.099
0.259
3.31
3.71
b
Economizer
(0.003)
(0.003)
0.030
0.013
0.099
0.023
0.100
0.011
[0.009]
0.016
0.052
0.081
0.080
0.208
0.710
(0.003)
[0.475]
1.91
0.220
2.16
0.144
0.153
0.856
0.178
0.075
0.099
[0.007]
0.375
0.356
0.037
0.189
0.225
6.98
7.69
Cyclone
(0.001)
0.014
0.056
0.014
0.164
0.040
0.388
0.026
0.084
0.076
0.764
0.383
0.341
0.456
2.81
0.004
(0.444)
0.712
0.160
0.378
0.032
0.070
0.202
0.042
0.020
0.041
0.002
0.089
0.065
0.002
0.017
0.044
1.88
4.69
Baghouse
(0.001)
0.025
0.142
0.020
0.270
0.055
0.550
0.052
0.099
0.090
1.10
0.622
0.605
0.959
4.59
0.010
0.276
2.59
0.516
1.42
0.086
0.189
0.700
0.094
0.048
0.157
(0.048)
0.192
0.148
0.024
0.046
0.033
6.58
11.2
Superheater ash and economizer ash results are from Run 11B. Cyclone
ash and baghouse ash results are average of results from Emission Test
and Method Study runs at Marion County MWC.
Concentration for the economizer ash is the average of duplicate analyses.
°Not detected. Detection limit given in parentheses; estimated maximum
possible concentration (EMPC) given in brackets. Values of detection
limits or EMPCs are not included in totals.
Lmo/038
2-17
-------
ODD
o
2
Eh
O
a
0.7
0.6 -
0.5 -
0.4 -
0.3 -
Code
A
8
C
0
E
F
Q
H
I
J
K
L
M
N
KEY
Congener
Dtoxini
Mono-CDt
Di-CDD
Tri-CDO
2378 TCC
Other TCC
12378 PC;
Other PCC'
123478 K
123678H
123789k
Other HxC
1234678-
Other HpC
Octa-CDO
ABCDEFGHI
M
CDF
0.7
0.6 -
0.5 -
0.4 -
I
b.
O
T U V Tf X Y ZAAABACADAE
0.3 -
Furans
Mono-CDf
Di-CDF
Tri-CDF
2378 TCD
Other TCi
12378 PC:
23478 PC:
Other PCC
123478h
123678h
234678K
123789h
Other HxC
1234678-
1234783-
Other HpC
Octa-CDF
Superheater IWSI Economizer V /A Cyclone
Bagbouae
Figure 2-2. Baseline Congener Distribution for Ash
2-18
-------
TABLE 2-10. COMBUSTOR VARIATION TEMPERATURE PROFILE FOR THE MARION COUNTY MWC
NJ
I
VO
LOCATION
CODE
1
2
3
4
5
6
7
8
9
10
11
TEMPERATURES (deg. F)
COMBUSTION AIR
MIDDLE OF FURNACE, FIRST PASS
TOP OF FURNACE, FIRST PASS
ECONOMIZER FLUE GAS OUTLET
INLET SAMPLING LOCATION
QUENCH REACTOR INLET
MIDPOINT SAMPLING LOCATION
QUENCH REACTOR OUTLET
BAGHOUSE OUTLET
ID FAN INLET
BREECHING TO OUTLET STACK
3A
LOW AIR
240
1895
1771
366
381
374
301
300
279
283
279
LOW AIR -
HIGH OF »
3B
HIGH LOW
EXCESS AIR A
245
1572
1561
432
452
440
296
299
280
285
279
Low excess air
High overfire
4
OF
IR
235
1731
1694
416
430
430
300
301
278
283
280
air
5
HIGH OF
AIR
239
1808
1734
416
428
433
298
299
279
282
280
distribution
6A
LOW
LOAD
254
1767
1675
364
408
369
302
302
277
281
277
6B
LOW LOAD
HIGH AIR
238
1490
1417
427
443
439
299
300
276
283
289
7
LOW LOAD
LOW AIR
240
1885
1766
364
386
363
300
288
271
278
282
8
LOW LOAD
LOW OF
234
1733
1668
377
387
383
302
298
279
284
282
9
LOW LOAD
HIGH OF
238
1639
1578
413
417
428
302
299
278
285
280
-------
TABLE 2-11. DIFFERENCE FROM BASELINE FOR COMBUSTOR EVALUATION FOR THE MARION COUNTY MWC
NJ
I
NJ
O
LOCATION
CODE
I
2
3
4
5
6
7
8
9
10
11
TEMPERATURE SENSOR LOCATIONS
COMBUSTION AIR
MIDDLE OF FURNACE, FIRST PASS
TOP OF FURNACE, FIRST PASS
ECONOMIZER FLUE CAS OUTLET
INLET SAMPLING LOCATION
QUENCH REACTOR INLET
MIDPOINT SAMPLING LOCATION
QUENCH REACTOR OUTLET
BAGHOUSE OUTLET
ID FAN INLET
BREECHING TO OUTLET STACK
3A
LOW AIR
2
9
5
-14
-13
-15
1
0
0
0
-2
3B
HIGH
EXCESS AIR
4
-10
-8
2
3
0
0
0
0
0
-2
4
LOW OF
AIR
0
-1
0
-2
-2
-2
1
0
-1
0
-1
5
HIGH OF
AIR
1
" 4
Z
-2
-2
-1
0
0
0
-1
-1
6A
LOW
LOAD
8
1
— 1
-14
-7
-16
2
1
-1
-1
-2
6B
LOW LOAD
HIGH AIR
1
-15
-16
1
1
0
1
0
-1
0
2
7
LOW LOAD
LOW AIR
2
8
4
-14
-12
-17
1
-4
-3
-2
-1
8
LOW LOAD
LOW OF
-1
-1
-1
-11
-12
-13
2
-1
0
0
-1
9
LOW LOAD
HIGH OF
1
-6
-7
-2
-5
-2
2
0
-1
0
-1
Difference (percent) = (Run value - baseline)/baseline * 100%. The baseline average is used.
-------
_
u
U)
u
u
111
ft
AFTER QUENCH
REACTOR
AFTER HEAT RECOVERY
-10 -
-15 -
-20
6A
TEMPERATURE LOCATION CODE
68 07 A 8
1. Combustion air
2. Middle of furnace, 1 st pass
3. Top of furnace, 1 st pass
4. Economizer outlet
5. Inlet Sampling location
6. Quench reactor inlet
7. Midpoint sampling location
8. Quench reactor outlet
9. Baghouse outlet
10. I.D. fan inlet
11. Breeching to outlet stack
Figure 2-3. Temperature Profile for Low Load Combustor
Evaluation Conditions
2-21
-------
K
x.f'
kJ
Z
d
IE
k.
o
UL
til
AFTER HEAT
RECOVERY
AFTER QUENCH
REACTOR
3A
TEMPERATURE LOCATION CODE
+ 3B 04
1. Combustion air
2. Middle of furnace, 1st pass
3. Top of furnace, 1st pass
4. Economizer outlet
5. Inlet Sampling location
6. Quench reactor inlet
7. Midpoint sampling location
8. Quench reactor outlet
9. Baghouse outlet
10. I.D. fan inlet
11. Breeching to outlet stack
Figure 2-4. Temperature Profile for Air Distribution and Excess Air
Combustor Evaluation Conditions
2-22
-------
2.2.2 Combustion Parameters during Combustor Variations
The results for the primary indicators of combustion conditions during
the combustor variations are summarized in Table 2-12. The difference from
baseline expressed as a percent for each parameter is summarized in
Table 2-13. Steam load was within 5 percent of baseline for Runs 3A, 3B, 4
and 5 and was reduced to approximately 75 percent for Runs 6A, 6B, 7, 8 and 9.
The variation in the steam load is shown graphically in Figure 2-5.
Excess air ranged from a low of 36.2 percent during Run 3A (low excess
air) to a high of 144 percent during Run 6B (low load, high excess air). The
variation in excess air is shown graphically in Figure 2-6.
Combustion efficiency varied from 99.86 to 99.98 percent. Correspon-
dingly, the CO concentration ranged from 12.6 to 1.7 ppmv dry. The CO values
reported are averages for each test run. The variations during each test run.
are discussed in Section 2.2.3.
To aid in the evaluation of the temperature profile, the variation in the
volumetric flowrate is shown graphically in Figure 2-7. The volumetric
flowrate of the flue gas was lowest during Run 7 (low load, low excess air)
and highest during Run 3B (high excess air). The volumetric flowrate did not
change more than 30 percent from baseline.
2.2.3 Fixed Gases (CO. CO.,. 0,)
2~2
Results of the fixed gases are presented on a dry basis in Table 2-14.
The average values discussed in this section are the averages of the CEM and
Orsat results. Over the course of the combustor variations, average oxygen
levels ranged from 6.0 percent by volume to 12.4 percent by volume and the
average carbon dioxide levels ranged from 7.4 percent by volume to 12.9
percent by volume. The highest oxygen concentrations were observed during the
high excess air test conditions 6B (low load, high excess air) and 3B (high
excess air). Oxygen content during these runs averaged 12.4 percent by volume
lmo/036
2-23
-------
TABLE 2-12. COMBUSTION PARAMETERS DURING THE COMBUSTOR EVALUATION
3A 3B
COMBUSTION PARAMETER LOW AIR HIGH
EXCESS AIR
to
S3
-P-
STEAM FLOW (Ib/hr)
a
EXCESS AIR (PERCENT)
a
CO CONCENTRATION (ppmv, DRV)
i a
C02 CONCENTRATION (X by vol, DRY)
a
02 CONCENTRATION (% by vol, DRY)
a,b
COMBUSTION EFFICIENCY (X)
C
VOLUMETRIC FLOWRATE (ACFM)
63990
36.2
5.6
13.1
5.7
99.96
46980
63940
99.5
12.6
8.9
10.6
99.86
67270
4 5
LOW 0/F HIGH 0/F
AIR AIR
65460
70.1
11.3
10.3
8.8
99.89
60600
68970
68.9
6.9
10.5
8.7
99.93
58590
6A 6B 7 8 9
LOW LOW LOAD LOW LOAD LOW LOAD LOW LOAD
LOAD HIGH AIR LOW AIR LOW 0/F HIGH 0/F
51230
70.1
1.7
10.3
8.8
99.98
52310
47960
144.1
10.3
7.3
12.5
99.86
59640
51590
57.9
2.1
11.6
7.8
99.98
42280
49900
85.8
5.6
9.8
9.8
99.94
42350
52090
90.8
8.5
9.5
10.1
99.91
49360
Measured at the inlet sampling location based on CEM data.
3
Combustion efficiency • moles of CO/[moles of CO + moles of C02]*100Z
Measured at the inlet sampling location.
-------
TABLE 2-13. DIFFERENCE FROM BASELINE FOR COMBUSTION PARAMETERS DURING THE COMBUSTOR EVALUATION
COMBUSTION PARAMETER
STEAM FLOW ( Ib/hr)
EXCESS AIR (PERCENT)
CO CONCENTRATION (ppmv, DRY)
N> C02 CONCENTRATION (Z by vol. DRY)
i
01 02 CONCENTRATION (Z by vol, DRY)
COMBUSTION EFFICIENCY (Z)
VOLUMETRIC FLOWRATE (ACFM)
3A
LOW AIR
95
-52
-45
23
-37
0
-22
3B
HIGH
EXCESS AIR
95
33
24
-17
17
0
12
4
LOW 0/F
AIR
98
-6
11
-3
-3
0
1
5
HIGH 0/F
AIR
103
-8
-32
-2
-4
0
-2
6A
LOW
LOAD
76
-6
-83
-3
-3
0
-13
6B
LOW LOAD
HIGH AIR
71
93
1
-32
38
0
-1
7
LOW LOAD
LOW AIR
77
-23
-79
9
-14
0
-30
8
LOW LOAD
LOW 0/F
74
15
-45
-8
8
0
-29
9
LOW LOAD
HIGH 0/F
78
21
-16
-11
11
0
-18
Difference (percent) - (Run value - baseline)/baseline * 100%. The baseline average is used.
-------
STEAM LOAD
DURING COMBUSTION VARIATIONS
38
5 6A
TEST CONDITION
68
Figure 2-5. Variation of Steam Loading During
the Combustor Evaluation
2-26
-------
EXCESS AIR
DURING COMBUSTON VARIATIONS
r.
n
a
o
x
u
160
150 -
140 -
130 -
120 -
110 -
1OO -
90-
80-
70 -
60 -
SO -
30 -
20 -
10 -
O
3A
38
6 €A
TEST CONDITION
68
Figure 2-6. Variation of Excess Air During
the Combustor Evaluation
2-27
-------
K
\*r
u
z
a
2
u
u
c
VOLUMETRIC FLOWRATE
DUWNO COMBUSTION VARIATIONS
5 6A
TEST CONDmON
68
Figure 2-7. Variation of Volumetric Flowrate During
the Combustor Evaluation
2-28
-------
TABLE 2-14. FIXED GASES (CO, CO , 0 ) AT COMBUSTOR EVALUATION TEST CONDITIONS
to
I
S3
VO
TEST CONDITION
INLET ORSAT
02, Xv, DRY
C02, Xv, DRY
PERCENT EXCESS AIR
Foa
INLET GEM
02, Xv, DRY
CO , Xv, DRY
CO, ppmv, DRY
PERCENT EXCESS AIR
Fo»
INLET AVERAGE*3
0 , Xv, DRY
CO , Xv, DRY
CO, ppmv, DRY adjusted
to 12 percent CO
PERCENT EXCESS AIR
Fo*
1
BASE-
LINE
9.0
10.0
72.7
1.19
9.0
10. A
10.0
73.3
1.14
9.0
10.2
11.8
73.0
1.17
2
BASE-
LINE
9.4
10.1
78.4
1.15
8.8
11.0
10.3
71.1
1.10
9.1
10.6
11.7
74.8
1.13
3A
LOW XS
AIR
6.2
12.6
40.2
1.17
5.7
13.1
5.6
36.2
1.16
6.0
12.9
5.2
38.2
1.17
3B
HIGH XS
AIR
10.9
8.9
106
1.13
10.6
8.9
12.6
99.5
1.16
10.8
8.9
17.0
103
1.15
4
LOW 0/F
AIR
8.9
9.7
70.6
1.24
8.8
10.3
11.3
70.1
1.17
8.9
10.0
13.6
70.4
1.21
5
HIGH 0/F
AIR
8.9
10.0
71.1
1.20
8.7
10.5
6.9
68.9
1.16
8.8
10.3
8.0
70.0
1.18
6A
LOW
LOAD
9.1
10.7
75.4
1.10
8.8
10.3
1.7
70.1
1.17
9.0
10.5
1.9
72.8
1.14
6B
LOW LOAD
HI XS AIR
12.2
7.5
136
1.16
12.5
7.3
10.3
144
1.15
12.4
7.4
16.7
140
1.16
7
LOW LOAD
Lo XS AIR
7.8
11.6
57.9
1.13
7.8
11.6
2.1
57.9
1.13
7.8
11.6
2.2
57.9
1.13
8
LOW LOAD
LOW 0/F
9.7
9.5
84.0
1.17
9.8
9.8
5.6
85.8
1.13
9.8
9.7
6.9
84.9
1.15
9
LOW LOAD
HIGH 0/F
10.0
9.4
88.7
1.16
10.1
9.5
8.5
90.8
1.14
10.1
9.5
10.7
89.8
1.15
*Fo - (20.9 -X 0 , dry)/(X C02< dry)
Average of Inlet CEM and Orsat values.
-------
for Run 6B and 10.8 percent by volume for Run 3B. Carbon dioxide
concentration was lowest during these runs with an average C00 concentration
for Run 6B of 7.4 percent by volume and an average CO,, concentration for Run
3B of 8.9 percent by volume. This is primarily due to dilution from excess
air. Excess air was 140 percent and 103 percent for Run 6B and 3B,
respectively.
Test conditions 3A and 7 had the lowest oxygen concentrations and the
highest CCL concentrations. These were both low excess air test conditions in
which the excess air was 38.2 percent for Run 3A and 57.9 percent for Run 7.
Carbon monoxide concentrations at the boiler outlet were low across the
spectrum of combustor variations made at the Marion County MWC. CO concen-
tration (corrected to 12 percent C0_) was lowest during Run 6A (low load) at
1.9 ppmv. The highest CO concentration observed was during Run 3B (low excess
air), at 17.0 ppmv. Comparison of combustor temperatures with CO levels
reveals that the five test conditions (6A, 7, 3A, 8, 5) exhibiting the lowest
CO concentrations (corrected to 12 percent C0_) corresponded to the five
highest middle furnace temperatures. These temperatures ranged from 1733 F
(Run 8) to 1895 F (Run 3A). All other mid-furnace temperatures ranged from
1490°F (Run 6B) to 1731°F (Run 4).
Figure 2-8 compares plots of CO, 0? and CO- concentration at the boiler
outlet with combustion air flow/steam load plots for Runs 2-9. Run 2 was
chosen as a more representative baseline than Run 1 due to more stable process
conditions and more consistent analyses. CO was particularly erratic during
Run 3B (high excess air) and Run 4 (low overfire/underfire air distribution)
and was also erratic during Run 2 (baseline), Run 6B (low load, high excess
air) and Run 8 (low load, low overfire/underfire air distribution).
High CO spikes were observed near the end of Run 3A (low excess air) and
at the middle and end of Run 6B (low load, high excess air). In Run 6B the CO
spikes are attributed to a quench pit seal that broke during the test. Run 3A
maintained steady low CO emissions through most of the test. The CO spikes in
lmo/036 2-30
-------
Run 2
Baseline
Run 3A
Low Excess Air
Run 3B
High Excess Air
Run 4
Low OF Air Distribution
Run 5
High OF Air Distribution
CO Concentration
ppmV
O2 Concentration
%V
CO2 Concentration
%V
Combustion
Air Flow D
(103 Ib/hr)
Steam Flow +
(103 Ib/hr)
f^f^J^
ty^f*^^
<^fa^(^^
^WV^^VH^
30
11:00 12:00 13:00 10:00 11:00 12:00
T T
14:30 15:30 16:30
Time
13:00 14:00
15:00 10:00
T r
11:00 12:00
13:00
Figure 2-8. Fixed Gas Concentration Histories during the Combustion Evaluation
-------
20.
CO Concentration
ppmV
15.
ID-
S'
20-
02 Concentration
%V
15-
10-
5
20
15-
C02 Concentration
%V
ID-
S'
120'
Combustion
Air Flow D
(103 Ib/hr)
Steam Flow +
(103 Ib/hr)
90-
60-
30-
10:00
Run 6A
Low Load
Run 6B Run 7 Run 8 Run 9
Low Load/High Excess Air Low Load/Low Excess Air Low Load/Low OF Air Distribution Low Load/High OF Air Distribution
k
—I 1
11:00 12:00 15:00
Full Scale Equals 90
On This Graph
frit}f^^W^
T
T
T
16:00
17:00 14:00 15:00 16:00 10:00 11:00 12:00 13:00 15:00 16:00 17:00
Time
Figure 2-8. Fixed Gas Concentration Histories during the Combustion Evaluation
(continued)
-------
3A were caused by a blockage on the feed table. The CO spikes exhibited in
Run 3A were also more pronounced than in Run 6B; typical CO concentrations
during Run 3A were approximately 4.5 ppmv but the peaks reached as high as 95
ppmv, whereas during Run 6B there was a great deal of fluctuation. During the
period of CO perturbation in Run 3A, oxygen concentration also showed some
unusually low troughs and CO- exhibited several high peaks. Combustion air
became more erratic during the Run 3A CO peaks and was generally lower than
for the rest of the run. Steam load also seemed to be more unstable and
decreased slightly during this period.
Oxygen and carbon dioxide levels varied from run to run due to the
different air distributions. Generally, these concentrations were relatively
stable, varying only one or two percent by volume. The greatest consistent
fluctuations were seen in Test conditions 3B, 4, 6B, 8 and 7. With the
exception of Run 7, these tests also exhibited erratic CO concentrations. The
fluctuations in 0_ and C0« during Run 7 were caused by erratic ID fan behavior
due to the low gas flowrates.
2.2.4 Additional Pollutants of Interest (NO and THC)
X
NO and THC concentrations during the combustor variations are presented
X
in Table 2-15. During the combustor variations testing, NO mass flowrates
jt
decreased with lower load conditions. The average control device inlet NO
X
for the low load conditions (Runs 6A, 6B, 7, 8, 9) was 35.0 Ib/hr (RSD = 20.4
percent). During normal load conditions (Runs 1, 2, 3A, 3B, 4, 5) NO mass
X
flowrates averaged 51.8 Ib/hr (RSD - 16.0 percent), while the baseline
conditions (Runs 1 & 2) averaged 58.8 Ib/hr.
Non-condensible THC emissions during all of the test conditions were
close to instrument detection limits. THC emissions for all runs were on the
order of magnitude of a tenth of a pound per hour as propane. The combustor
variations did not seem to affect the rate of hydrocarbon emissions at these
levels.
lmo/036 2-35
-------
TABLE 2-15. NOx AND THC EMISSIONS FOR THE COMBUSTOR EVALUATION CONDITIONS
TEST CONDITION:
INLET
NOx, ppmv, DRY
NOx, ppmv @12Z C02
NOx, Ib/hr
OUTLET
NOx, ppmv, DRY
NOx, ppmv @12X C02
NOx. Ib/hr
^ INLET
THC, ppmv aa propane, DRY
THC, ppmv as propane @12I C02
THC, Ib/hr aa propane
OUTLET
THC, ppmv as propane, DRY
THC, ppmv as propane @12Z C02
THC, Ib/hr as propane
1
BASE-
LINE
264
305
57.7
204
306
63.0
NR
NR
NR
0.6
0.9
0.2
2
BASE-
LINE
262
285
59.9
205
304
59.2
0.7
0.8
0.2
0.4
0.6
0.1
3A
LOW XS
AIR
218
200
40.4
165
195
38.6
NR
NR
NR
0.6
0.7
0.1
3B
HIGH XS
AIR
230
310
56.1
188
304
61.1
NR
NR
NR
0.4
0.6
0.1
4
LOW 0/F
AIR
190
221
43.3
149
239
42.4
0.5
0.6
0.1
0.4
0.6
0.1
5
HIGH 0/F
AIR
240
274
53.6
187
274
54.2
NR
NR
NR
0.4
0.0
0.0
6A
LOW LOAD
220
256
44.6
170
243
37.0
1.6
1.9
0.3
NR
0.0
0.0
6B
LOW LOAD
Hi XS AIR
142
233
32.3
126
270
37.8
0.9
1.5
0.2
NR
0.0
0.0
7
LOW LOAD
Lo XS AIR
184
191
30.8
148
202
30.9
NR
NR
NR
NR
NR
NR
8
LOW LOAD
LOW 0/F
150
184
25.7
112
177
24.9
1.3
1.6
0.2
NR
NR
NR
9
LOW LOAD
HIGH 0/F
219
276
42.0
170
283
43.5
1.2
1.5
0.2
NR
NR
NR
NR » Not reported due to invalidation or reading not taken.
-------
2.2.5 Acid Gas Emissions
Acid gas concentrations during the combustor evaluation are presented in
Tables 2-16 and 2-17. The average uncontrolled mass flowrates during low load
conditions for S02 and HC1 were 55.0 Ib/hr SO- with a relative standard
deviation of 34.0 percent and 73.6 Ib/hr HC1 with a relative standard
deviation of 19.2 percent. (HC1 average is based on combined GEM and manual
method results.) During normal load conditions uncontrolled SO- mass
flowrates averaged 105.9 Ib/hr with a relative standard deviation of 38.6
percent while GEM and manual HC1 tests gave an average uncontrolled HC1 mass
flowrate of 89.0 Ib/hr with a standard deviation of 13.6 percent.
The control efficiencies and stoichiometric ratios during the combustor
evaluation conditions are presented in Table 2-17. Combined S0? and HC1
stoichiometric ratios (molar ratio of calcium supplied by the quench reactor
to the theoretical calcium to react with inlet SO. and HC1) ranged from 1.07
L.
to 2.50. The HC1 control efficiencies for the overall control system ranged
from 85.9 to 98.4 percent. SO- removal efficiency ranged from 25.3 percent to
92.5 percent.
2.2.6 CDD/GDF Concentration in the Ash
In Table 2-18, the CDD/CDF concentration and 2378-TCDD toxic
equivalencies are presented for ash under combustor evaluation conditions.
Ash samples were taken at the superheater, economizer, cyclone, and baghouse.
The average results for ash at baseline conditions are also shown in
Table 2-18 for comparison.
Total CDD concentrations during combustor variations were significantly
different from baseline in at least one run for the superheater ash,
economizer ash, and cyclone ash. The baghouse ash CDD concentrations were not
significantly different than the baseline results for all the runs. The
economizer ash was significantly higher at 37 ng/g of total CDD than baseline
at 0.71 ng/g for Run 4. The superheater ash CDD concentrations were
significantly higher than baseline in Runs 4 and 6A. In the cyclone ash, the
CDD results for Runs 4, 6A and 6B were significantly lower than baseline.
lmo/036
2-37
-------
TABLE 2-16. SUMMARY OF ACID GAS CONCENTRATIONS DURING THE COMBUSTOR EVALUATION
(0
1
10
00
TEST CONDITION:
INLET SO , pprav 912X CO
MIDPOINT S02> ppmv (J12X O>2
OUTLET SO , ppmv 812 X CO
INLET HC1, MANUAL, ppmv Q12X CO
MIDPOINT HCl, MANUAL, ppmv @12X CO
OUTLET HCL, MANUAL, ppmv g!2X CO
INLET HCl, CEM, ppmv Q12X CO
MIDPOINT HCl, CEM, ppmv 812X CO
OUTLET HCl, CEM, ppmv @12X C02
1
BASE-
LINE
559
448
383
462
177
NR
646
225
83.7
2
BASE-
LINE
299
128
99.5
502
222
37.6
631
183
35.0
3A
LOW XS
AIR
428
295
186
385
229
53.8
496
177
49.9
3B
HIGH XS
AIR
523
351
237
NR
208
65.4
704
160
47.7
4
LOW 0/F
AIR
120
31.9
9.9
420
189
13.1
648
162
11.5
5
HIGH 0/F
AIR
425
247
158
598
325
66.3
729
110
45.5
6A
LOW LOAD
HI XS AIR
340
241
185
652
484
72.7
693
225
69.6
6B
LOW LOAD
Lo XS AIR
275
119
52.5
475
294
49.3
625
92.5
27.4
7
LOW LOAD
LOW 0/F
281
207
139
648
413
80.9
653
225
67.2
8
LOW LOAD
HIGH 0/F
210
158
87.5
530
266
50.5
568
186
39.5
9
LOW LOAD
HIGH 0/F
168
40.3
21.3
539
280
30.2
642
204
19.7
Note: All values reported are normalized to 12X CO .
NR - Not reported due to Invalidation.
-------
TABLE 2-17. CONTROL DEVICE REMOVAL EFFICIENCIES DURING THE COMBUSTION EVALUATION
N>
I
v£>
TEST CONDITION:
INLET SO , ppmv, dry
INLET SOj, Ib/hr
INLET HCI, ppmV, dry*
INLET HCI, Ib/hr*
STOICHIOMETRIC RATIO
QUENCH REACTOR EFFICIENCY
PERCENT SO REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
FABRIC FILTER EFFICIENCY
PERCENT SO REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
OVERALL SYSTEM EFFICIENCY
PERCENT SO REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
1
BASE-
LINE
484
147
480
83.0
1.08
17.5
64.4
60.7
9.4
60.4
NR
25.3
85.9
NR
2
BASE-
LINE
274
87.4
519
94.3
1.33
55.9
70.2
54.5
30.1
82.8
84.7
69.2
94.9
93.1
3A
LOU XS
AIR
468
121
481
70.6
1.26
26.0
61.6
35.8
42.8
74.4
78.7
57.6
90.2
86.3
3B
HIGH XS
AIR
388
132
522
101
1.07
23.1
73.9
NR
34.6
71.2
69.6
49.7
92.5
NR
4
LOU 0/F
AIR
103
32.8
458
82.8
2.22
74.5
76.1
56.8
70.7
93.3
93.4
92.5
98.4
97.2
5
HIGH 0/F
AIR
372
116
581
103
1.14
41.7
84.8
45.4
35.6
58.4
79.5
62.4
93.7
88.8
6A
LOU
LOAD
292
82.3
577
92.7
1.40
39.1
72.1
36.2
21.9
68.6
84.7
52.5
91.2
90.2
6B
LOU LOAD
HI XS AIR
167
53.1
335
60.4
2.24
55.7
84.8
36.5
56.5
70.8
83.5
80.7
95.6
89.5
7
LOU LOAD
Lo XS AIR
272
63.2
629
83.1
1.62
29.8
67.2
39.1
33.1
70.3
80.6
53.0
90.2
88.2
B
LOU LOAD
LOU 0/F
172
41.0
448
60.9
2.50
24.7
67.2
49.7
44.4
78.7
81.0
58.2
93.0
90.4
9
LOU LOAD
HIGH 0/F
133
35.4
467
71.1
2.36
77.3
70.0
50.8
43.4
89.7
88.5
87.1
96.9
94.3
NR - Not reported due to invalidation.
Average of CEM and manual results.
lmo/038
-------
TABLE 2-18. CDD AND CDF CONCENTRATIONS AND 2378-TCDD TOXIC EQUIVALENCIES FOR
ASH FROM COMBUSTOR EVALUATION CONDITIONS AT MARION COUNTY MWC
Ash Type
Superheater Ash
Economizer Ash
Cyc lone Ash
Baghouse Ash
Superheater Ash
Economizer Ash
Cyclone Ash
Baghouse Ash
Run 3B
High EA
NC
0.277
1.11
1.74
NC
1.19
1.65
10.5
Run 4
Low
OF Air
TOTAL CDD
1.78
37.1
0.522
1.84
TOTAL CDF
1.92
9.34
2.08
11.1
2378-TCDD TOXIC
Superheater Ash
Economizer Ash
Cyclone Ash
Baghouse Ash
NC
0.015
0.024
0.119
0.022
0.589
0.029
0.141
Run 6A Run 6B
Low Load
Low Load High EA
CONCENTRATION
6.91
0.634
0.625
2.18
CONCENTRATION
9.46
5.04
1.71
10.9
(ng/g)
0.926
0.520
0.681
2.33
(ng/g)
2.84
1.23
0.863
11.1
EQUIVALENT CONCENTRATION
0.176
0.047
0.031
0.113
0.036
0.009
0.017
0.115
Baseline
Average
0.400
0.710
2.81
4.59
3.31
6.98
1.88
6.58
(ng/g)
0.030
0.085
0.074
0.148
EA = Excess air.
OF Air = Overfire air distribution.
NC = Not collected. Sample not collected during this run.
2-40
-------
There were few significant variations from baseline for total CDF
concentrations. Only in the economizer ash samples for Runs 3B and 6B did the
concentrations significantly differ from baseline. These results were both
lower than baseline. The 2378-TCDD toxic equivalent concentrations for the
four types of ash were usually lower than the baseline results. Only for the
economizer ash of Run 4 and the superheater ash of Run 6A was the toxic
equivalency higher than baseline. For both ash samples, the toxic equivalency
was approximately six times the baseline value. The baghouse ash results were
similar to baseline results for all the runs, but were consistently lower.
The concentrations of the specific CDD/CDF congeners in the ash are shown
in Tables 2-19 to 2-22. Each table shows the results for all the samples
taken at a single sampling location. Congener distributions are presented
graphically in Figures 2-9 and 2-10. Baseline congener distributions are
distinguished by entirely shaded areas in Figures 2-9 and 2-10. Tables of the
distributions are presented in Appendix A.1.2. Most of the CDD and CDF
homologue distributions are fairly similar to baseline. For the economizer
ash, the CDD homologue distributions for Runs 4, 5B and 6B are different from
baseline. In Run 4 there is a greater fraction of lower chlorinated
homologues than baseline. For Runs 3B and 6B, there are greater fractions of
higher chlorinated homologues. The differences for Runs 3B and 6B may be from
low homologue concentrations, however, with many congeners not detected.
2.3 EFFECT OF OFF-DESIGN TEMPERATURES IN THE EMISSION CONTROL SYSTEM
2.3.1 Acid Gas Emissions during Control Device Variations
During the control device evaluation portion of the characterization
tests, the quench reactor outlet temperature was varied. The lime slurry feed
rate is controlled based on the flue gas temperature at the quench reactor
outlet. The quench reactor has two purposes: to reduce the temperature of the
flue gas before entering the baghouse and to reduce HCl and SO- emissions.
The stoichiometric ratio (molar ratio of calcium supplied by the quench
reactor to the theoretical calcium to react with the inlet SO^ and HCl) and
quench reactor outlet temperature cannot be independently controlled.
lmo/036 _
-------
TABLE 2-19. CDD AND CDF RESULTS FOR SUPERHEATER ASH AT COMBUSTOR EVALUATION CONDITIONS
TEST CONDITIONS 3B 4
Combustor
Load (Ib/hr steam) Normal Normal
Excess Air High Normal
Overf ire Air
Distribution Normal Low
Control Device
Quench Reactor Outlet
Temperature Normal Normal
Run Number
6A 6B
Low Low
Normal High
Normal Normal
Normal Normal
CDD /CDF CONCENTRATION (ng/g)
Run Number
Isomer 3B 4 6A 6B
DIOXINS
Mono -CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
(0.003)
[0.012]
0.056
[0.006]
8.174
0.011
0.330
0.010
[0.019]
0.033
0.537
0.138
Other HpCDD 0.190
Octa-CDD H 0.300
Total CDD ° 1.78
FURANS C
Mono-CDF 0 (0.001)
Di-CDF L [0.056]
Tri-CDF L 0.334
2378 TCDF E 0.040
Other TCDF C 0.685
12378 PCDF T [0.046]
23478 PCDF E 0.044
Other PCDF D 0.312
123478 HxCDF 0.077
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD/CDF
0.030
0.026
(0.003)
0.129
0.108
0.011
0.053
0.075
1.92
3.70
(0.001)
[0.023]
0.318
0.021
0.810
0.118
1.14
0.108
0.162
0.273
1.30
0.765
0.820
1.07
6.91
(0.001)
0.091
1.78
0.120
2.54
0.146
0.199
1.56
0.458
0.198
0.137
(0.003)
0.914
0.833
O.Q46
0.234
0.201
9.46
16.4
(0.001)
[0.022]
0.091
0.011
0.161
[0.025]
0.155
0.014
0.019
0.041
0.098
0.083
0.093
0.160
0.926
(0.001)
0.028
0.535
0.040
0.719
0.051
0.064
0.548
0.136
0.061
0.044
[0.012]
0.245
0.228
[0.014]
0.076
0.068
2.84
3.77
Baseline
Normal
Normal
Normal
Baseline
(0.001)
[0.009]
[0.054]
0.008
0.049
[0.006]
0.017
[0.003]
[0.004]
[0.011]
0.025
0.061
0.049
0.191
0.400
(0.001)
[0.240]
1.04
0.070
0.952
0.066
0.052
0.309
0.063
0,040
0.045
0.004
0.159
0.112
0.041
0.099
0.259
3.31
3.71
"Baseline is Run 11B results.
Not detected. Detection limit given in parentheses;
concentration (EMPC) given in brackets.
estimated maximum possible
2-42
-------
TABLE 2-20. CDD AND CDF RESULTS FOR ECONOMIZER ASH AT COMBUSTOR EVALUATION CONDITIONS
Run Nunber
TEST CONDITIONS
Ccmbustor
Load (Lb/hr steam)
Excels Air
Overfire Air
Distribution
Control Device
Quench Reactor Outlet
Temperature
3B
Normal
High
Normal
Normal
4
Normal
Normal
Low
Normal
CDD /CDF
6A
Low
Normal
Normal
Normal
6B
Low
High
Normal
Normal
Baseline
Normal
Normal
Normal
CONCENTRATION (ng/g)
Run Number
Isomer
DIOXINS
Mono -CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD /CDF
3B
(0.003)b
(0.003)
[0.023J
[0.037]
0.024
0.006
0.000
(0.003)
(0.003)
(0.005)
[0.052]
0.025
0.021
0.201
0.277
(0.003)
(0.008)
0.495
0.090
0.403
[0.021]
0.018
0.101
0.025
0.011
(0.003)
(0.003)
0.005
0.040
(0.003)
0.004
[0.023]
1.19
1.47
4
(0.003)
1.15
6.36
0.260
8.16
0.198
7.78
0.179
0.278
0.731
5.75
1.40
2.21
2.66
37.1
[0.050]
[0.605]
1.50
0.160
2.51
0.106
0.185
1.44
0.359
0.120
0.242
0.095
0.579
0.705
0.099
0.452
0.791
9.34
46.5
6A
(0.003)
[0.017]
[0.075]
0.014
0.087
[0.017]
0.043
[0.011]
[0.010]
[0.029]
0.070
0.039
0.082
0.249
0.634
[0.012]
0.135
1.64
0.100
1.72
0.095
0.081
0.561
0.110
0.052
0.042
[0.007]
0.194
0.150
[0.015]
0.067
0.090
5.04
5.68
6B
(0.003)
(0.003)
(0.003)
[0.003]
0.015
0.005
0.000
[0.003]
[0.005]
0.016
0.027
0.085
0.068
0.304
0.520
(0.001)
(0.005)
0.104
[0.04]
0.257
0.017
0.027
0.123
0.054
0.023
0.039
(0.003)
0.056
0.122
0.033
0.090
0.285
1.23
1.75
Baseline*
(0.003)
(0.003)
0.030
0.013
0.099
0.023
0.100
0.011
[0.009]
0.016
0.052
0.081
0.080
0.208
0.710
(0.003)
[0.475]
1.91
0.220
2.16
0.144
0.153
0.856
0.178
0.075
0.099
[0.007]
0.375
0.356
0.037
0.189
0.225
6.98
7.69
"Concentration of baseline ash is the average of duplicate analyses for Run 11B.
bNot detected. Detection limit given in parentheses: estimated maximum possible
concentration (EMPC) given in brackets.
2-43
-------
TABLE 2-21. CDD AND CDF RESULTS FOR CYCLONE ASH AT COMBUSTOR EVALUATION CONDITIONS
Run Number
TEST CONDITIONS
Combustor
Load (Ib/hr steam)
Excess Air
Overf Ire Air
Distribution
Control Device
Quench Reactor Outlet
Temperature
3B
Normal
High
Normal
Normal
4
Normal
Normal
Low
Normal
CDD/CDF
6A
Low
Normal
Normal
Normal
6B
Low
High
Normal
Normal
Baseline
Normal
Normal
Normal
Normal
CONCENTRATION (ng/g)
Run Number
Isomer
DIOXINS
Mono -CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono -CDF
Dl-CDF
Trl-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD/CDF
3B
b
(0.001)
[0.002]
[0.043]
0.012
0.126
[0.021]
0.217
0.012
0.036
0.046
0.154
0.168
0.139
0.201
1.11
(0.001)
[0.025]
0.331
[0.06]
0.639
[0.042]
0.037
0.273
0.064
0.035
0.026
(0.001)
0.104
0.102
[0.004]
0.020
0.021
1.65
2.76
4
(0.001)
(0.001)
[0.050]
[0.010]
0.064
0.011
0.077
0.006
[0.018]
0.021
0.084
0.086
0.066
0.107
0.522
(0.001)
[0.036]
0.645
0.100
0.798
0.051
0.038
0.229
0.044
0.022
0.015
(0.001)
0.066
0.057
[0.003]
0.015
[0.013]
2.08
2.60
6A
(0.001)
(0.001)
0.012
0.008
0.031
0.011
0.103
0.007
0.021
0.026
0.096
0.097
0.080
0.133
0.625
[0.001]
(0.003)
0.486
0.060
0.623
0.030
0.035
0.216
0.048
0.021
0.017
(0.001)
0.071
0.090
[0.003]
0.015
[0.015]
1.71
2.34
6B
(0.001)
(0.003)
0.011
0.005
0.024
0.007
0.088
0.007
0.021
0.027
0.095
0.122
0.089
0.185
0.681
(0.001)
[0.007]
0.225
0.020
0.241
0.020
[0.023]
0.129
0.041
0.020
[0.015]
(0.001)
0.056
0.083
[0.004]
0.010
0.018
0.863
1.54
Baseline
(0.001)
0.014
0.056
0.014
0.164
0.040
0.388
0.026
0.084
0.076
0.764
0.383
0.341
0.456
2.81
0.004
(0.444)
0.712
0.160
0.378
0.032
0.070
0.202
0.042
0.020
0.041
0.002
0.089
0.065
0.002
0.017
0.044
1.88
4.69
Baseline la average of Emission Test and Method Study results.
Not detected. Detection limit given in parentheses: estimated maximum possible
concentration (EMPC) given in brackets.
2-44
-------
TABLE 2-22. CDD AND CDF RESULTS FOR BAGHOOSE ASH AT COMBUSTOR EVALUATION CONDITIONS
TEST CONDITIONS
Combustor
Load (Ib/hr steam)
Excess Atr
Overflre Air
Distribution
Control Device
Quench Reactor Outlet
Temperature
Isomer
DIQXINS
Mono-CDD
Di-CDD
Trl-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD /CDF
3B
Normal
High
Normal
Normal
3Ba
(0.005)°
[0.059]
0.142
0.028
0.268
0.037
0.227
0.011
0.045
0.073
0.201
0.227
0.196
0.288
1.74
[0.026]
0.037
3.30
0.200
4.16
0.186
0.151
1.41
0.214
0.099
0.038
(0.010)
0.398
0.273
0.005
0.030
[0.134]
10.5
12.2
4
Normal
Normal
Low
Normal
CDD/CDF
4
(0.001)
[0.056]
[0.244]
0.035
0.330
0.042
0.311
0.022
0.054
0.082
0.165
0.255
0.201
0.347
1.84
[0.006]
0.104
3.60
0.280
4.33
0.197
0.160
1.47
0.168
0.096
0.066
(0.003)
0.392
0.164
0.013
0.017
0.034
11.1
12.9
Run Number
6A
Low
Normal
Normal
Normal
6B
Low
High
Normal
Normal
Baseline
Normal
Normal
Normal
Normal
CONCENTRATION (ng/g)
Run Number
6A
(0.001)
[0.081]
0.203
0.029
0.282
0.036
0.302
0.022
0.047
0.070
0.384
0.256
0.215
0.333
2.18
0.014
0.021
4.47
0.230
3.76
0.118
0.123
1.21
0.148
0.077
0.060
0.010
0.332
0.188
0.014
0.050
0.044
10.9
13.1
6B
[0.103]
0.035
0.203
(0.003)
0.282
0.040
0.261
0.024
0.046
0.074
0.364
0.290
0.278
0.437
2.33
0.019
2.330
4.25
0.520
2.27
0.172
0.145
0.786
0.156
0.076
[0.066]
(0.003)
0.064
0.236
[0.018]
0.078
[0.064]
11.1
13.4
Baseline
(0.001)
0.025
0.142
0.020
0.270
0.055
0.550
0.052
0.099
0.090
1.10
0.622
0.605
0.959
4.59
0.010
0.276
2.59
0.516
1.42
0.086
0.189
0.700
0.094
0.048
0.157
(0.048)
0.192
0.148
0.024
0.046
0.083
6.58
11.2
Concentration for Run 3B ash is the average of duplicate analyses.
bBaseline is average of Emission Test and Method Study results.
°Not detected. Detection limit given in parentheses; estimated maximum possible
concentration (EMPC) given in brackets.
2-45
-------
Sup«rh«at«r Ash
r-o
i
cr*
0 1
l^nJ
_4L
•fM [ft
i
Cyclon* Ash
L£L
.ift. Btfl ^.-..Bffi
KEY
Code Congener
Dioxins
Mono-CDD
DiCDD
TriCDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
A
B
C
D
E
F
G
H
J 123789 HxCDD
K Other HxCDD
L 1234678 HpCDD as
M Other HpCDD
N Octa CDD 0 5
^^^^ a
H Baseline § 04
^^^^ o
£
tXX^ Run 3B • 03
¥72 Run 4 02
Run 6A
Run 6B
Economizer Ash
JL. ,,, iafl I
CDErcHiJKt
_ . A .
Ba9hou««A8h
Figure 2-9. Ash CDD Congener Distributions
-------
Sup«fh»Ur Alh
Economizer Ath
ir /
0 0
05
a
3 04
I
o 03
a
0 2
01
0
_
111!
o p g R s
'
!
ft] 1>-
41) -iu .» W3 rfSI Wfl fer
"> t f" t r I r 1 "
KEY
Code Congener
Furans
O Mono-CDF
P DiCDF
Q TriCDF
R 2378 TCDF
S Other TCDF
T 12378 PCDF
U 23478 PCDF
V Other PCDF
W 123478 HxCDF
X 123678 HxCDF
() 7
0 ft
o r»
a
3 04
tj
£
| o.»
o ;;
0 1
n
Cyclon* Alh
k ifi
1 1
to!
ill
I
Bl
jflukiM
y
AiUi^ MJflL^mi
Y 234678 HxCDF
Z 123789 HxCDF
AA Other HxCDF
AB 1234678 HpCDF
AC 1234789 HpCDF
AD Other HpCDF
AE OctaCDF
m Baseline
I
£S2 Run 3B
V~77i Run *
f\X^ Run 6A
\/\A Run 6B
^UiKJll
-vM
JLl
I' Q R S T II V W X V 7
Baghoui* A*h
«Jutl/uIPtf
OPQRSTIIVVXYZ
All »( Al> AK
Figure 2-10. Ash CDF Congener Distributions
-------
During baseline conditions (Runs 1 and 2) the quench reactor outlet
temperature was 300 F. Test condition 10 had a lower than design quench
reactor outlet temperature of 262°F, which corresponds to an increased rate of
lime slurry injection. Runs 11A and 11B had increased quench reactor outlet
temperatures of 330 F and 360 F respectively. These higher temperatures are
due to decreased lime slurry injection, with Run 11B having the lowest rate.
Two parameters which may affect acid gas removal efficiency are: inlet
acid gas concentration and the injection rate of lime. These two variables
may be combined to form the stoichiometric ratio (molar ratio of supplied
calcium to acid gas) which is a major influence on acid gas reduction
efficiency. HCl and SO. concentrations during the control device evaluations
are presented in Table 2-23. The acid gas stoichiometric ratios and control
efficiencies are also summarized in Table 2-23. The stoichiometric ratios for
Runs 10, 11A, and 11B were 1.14, 1.06, and 1.59 respectively.
Figure 2-11 compares HCl and SO- inlet gas concentration to reduction
efficiencies. Only data from Run 2 are used to show baseline because inlet
S0_ concentrations for Run 1 are atypically high, causing a lower
stoichiometric ratio than normally used. These plots indicate that as acid
gas concentration increases, efficiency decreases. Peaks in SO- and HCl
correspond to lows in the respective efficiency plots. SO- concentration
seems to more dramatically affect efficiency than HCl; however, S0_
concentration varies more than HCl. These plots also indicate that the
control device removes HCl more effectively than S0«. Efficiency increased
for both HCl and S0_ during Run 10. HCl and S02 reduction efficiencies were
decreased during Runs 11A and 11B.
2.3.2 Temperature Profile during Control Device Variations
The temperature profile results for off-design temperatures in the
control system are presented in Table 2-24. Also included in Table 2-24 is
the difference from the baseline average, which is shown graphically in
Figure 2-12.
lmo/036 2-48
-------
TABLE 2-23. ACID GAS BEHAVIOR FOR THE CONTROL DEVICE EVALUATION TESTING
TEST CONDITION
INLET S02> ppmv, dry
INLET S02, Ib/hr
INLET HC1, ppnrv, dry3
INLET HC1, lb/hra
STOICHIOMETRIC RATIO
INLET S02 , ppnrv @12% CO
MIDPOINT SO., ppmv (§12% CO
OUTLET S02, ppnrv (§12% C02
INLET HC1, MANUAL, ppnrv @12% CO-
MIDPOINT HC1, MANUAL, ppnrv @12% CO
OUTLET HCL, MANUAL, ppmv (§12% CO
INLET HC1, CEM, ppmv @12% C02
MIDPOINT HC1, CEM, ppmv (§12% CO-
OUTLET HC1, CEM, ppmv (§12% C02
QUENCH REACTOR EFFICIENCY
PERCENT SO,, REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
FABRIC FILTER EFFICIENCY
PERCENT SO, REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
OVERALL SYSTEM EFFICIENCY
PERCENT SO. REDUCTION
PERCENT HCI REDUCTION, CEM
PERCENT HCI REDUCTION, MANUAL
1
BASE-
LINE
484
147
480
83.0
1.08
559
450
383
462
177
NR
646
225
83.7
17.5
64.4
60.7
9.4
60.4
NR
25.3
85.9
NR
2
BASE-
LINE
274
87.4
519
94.3
1.33
299
128
99.5
502
222
37.6
631
183
35.0
55.9
70.2
54.5
30.1
82.8
84.7
69.2
94.9
93.1
10
LOW QR
OUT T.
328
99.9
699
60.5
1.14
383
326
108
NR
229
23.4
814
180
20.4
18.2
78.8
NR
66.8
88.6
89.8
72.9
97.6
. NR
11A
HIGH QR
OUT T.
415
125
646
111
1.06
470
523
484
745
408
172
718
295
158
-14. 6b
57.6
43.6
13.5
50.2
60.6
0.9
78.9
77.8
11B
HIGH QR
OUT T.
108
35.5
695
130
1.59
118
178
165
767
545
228
750
313 -
214
-37. 9b
61.7
34.8
14.2
36.8
61.3
-18. 3b
75.8
74.8
Note: All values are reported on a dry basis.
NR - Not reported due to invalidation.
aAverage of CEM and manual results.
Instrument inaccuracies because of measuring low concentrations while calibrated with
a large span and differences between individual analyzers are responsible for the
differences in S09 concentration at the three locations. These valued should be
considered equivalent and indicate that no significant removal of S02 took place
during these runs.
2-49
-------
100
Control
Efficiency
D HCI Efficiency so •
+ SO2 Efficiency
25-1
1.0
0.75-
HCI Concentration 0.5
ppmV
(thousands)
0.25 H
1.0-
0.75-
SO2 Concentration
ppmV
(thousands)
0.5-
0.25-
0-
11:00
Test 2
Baseline
Test 10
Low QR Outlet Temp.
Test11A
High QR Outlet Temp.
1
12:00
1
13:00
12:30
13:30
1 —
14:30
Test 11B
High QR Outlet Temp.
10:00
—I
11:00
12:00
14:30
15:30
—I
16:30
17:30
Sampling Times
Figure 2-11. Effect of Acid Gas Concentration on Control Efficiency
cc
at
o
N.
§
-------
TABLE 2-24. TEMPERATURE PROFILE AND DIFFERENCE FROM BASELINE
NJ
Ul
OJ
LOCATION
CODE
7
8
9
10
11
TEMPERATURES, deg. F:
MIDPOINT SAMPLING LOCATION
QUENCH REACTOR OUTLET
BAGHOUSE OUTLET
ID FAN INLET
BREECHING TO OUTLET STACK
10
LOW QUENCH
TEMPERATURE
281
262
252
254
260
11A
HIGH QUENCH
TEMPERATURE
310
330
302
305
303
11B
HIGH QUENCH
TEMPERATURE
362
360
334
334
352
10
LOW QUENCH
TEMPERATURE
-5
-13
-10
-11
-8
11A
HIGH QUENCH
TEMPERATURE
4
10
8
7
7
11B
HIGH QUENCH
TEMPERATURE
22
20
19
18
24
Difference (percent) = (run value - baseline)/baseline * 100%.
-------
OFF TEMPERATURE PROFILE
30
20 -
UJ
r
10 -
O
U
QL
-10 -
-20 -
-30
l
10
11
TEMPERATURE LOCATION CODE
10 -I- 11A O
118
7. Midpoint sampling location
8. Quench reactor outlet
9. Baghouse outlet
10. I.D. fan inlet
11. Breeching to outlet stack
Figure 2-12. Temperature Profile for After the Quench Reactor for Control
Device Evaluation Conditions
2-54
-------
During the low temperature condition (Run 10), the quench reactor outlet
temperature was 13 percent below baseline. The moderately high temperature
condition (Run 11A) was 10 percent above baseline and the high temperature
condition (Run 11B) was 20 percent above baseline.
2.3.3 Fixed Gases (CO. CO 0-) and Additional Pollutants of Interest
(NO and THC)
x
During the control device evaluation runs (10, 11A, and 11B) no furnace
or combustion parameters were altered. Since the control device does not
significantly affect CO, 0-, C0_ NO or THC levels in the flue gas, other than
£. £- 2t
by leakage, the control device evaluation runs may be considered baseline
runs for fixed gas emissions at the boiler outlet. Therefore C0«, 0« and CO
concentrations for these tests are discussed in Section 2.1.4 and NO and THC
x
are discussed in Section 2.1.5 as baseline conditions.
2.3.4 CDD/CDF Concentrations in Ash during Control Device Variations
In Table 2-25, the CDD/CDF concentrations and 2378-TCDD toxic
equivalencies are presented for the ash under control device evaluation
conditions. Ash samples were taken at the baghouse for Runs 10, 11A, and 11B,
and at the cyclone for Run 11B. The ash samples from the superheater and
economizer were not analyzed, since these would not be affected by off-design
temperatures in the emission control system. The results for ash at baseline
conditions are shown also in Table 2-25.
The total CDD concentrations were 0.90 ng/g for Run 11B cyclone ash and
ranged from 1.22 ng/g for Run 11A baghouse ash to 1.9 ng/g for Run 11B
baghouse ash. The total CDF concentrations were 1.24 ng/g for Run 11B cyclone
ash and ranged from 3.80 ng/g for Run 10 baghouse ash to 8.33 ng/g for Run 11B
baghouse ash. The total CDF concentrations were not significantly different
from baseline for any of the test conditions or sampling locations. The total
CDD concentrations were significantly lower than baseline for all samples
except for Run 11B baghouse ash. Similar to the total CDD concentrations, the
lmo/036
2-55
-------
TABLE 2-25. CDD AND CDF CONCENTRATIONS AND 2378-TCDD TOXIC EQUIVALENCIES FOR
ASH FROM CONTROL DEVICE EVALUATION CONDITIONS AT MARION COUNTY MWC
Run 10 Run 11A Run 11B
Low QR High QR High QR Baseline
Ash Type Outlet Temp. Outlet Temp. Outlet Temp. Average
TOTAL CDD CONCENTRATION (ng/g)
Cyclone Ash NA NA 0.903 2.81
Baghouse Ash 1.30 1.22 1.90 4.59
TOTAL CDF CONCENTRATION (ng/g)
Cyclone Ash NA NA 1.24 1.88
Baghouse Ash 3.80 5.47 8.33 6.58
2378-TCDD TOXIC EQUIVALENT CONCENTRATION (ng/g)
Cyclone Ash NA NA 0.015 0.074
Baghouse Ash 0.019 0.036 0.077 0.148
QR = Quench reactor.
NA = Not analyzed. These samples were collected but not analyzed.
2-56
-------
2378-TCDD toxic equivalencies were significantly lower than baseline for all
samples except Run 11B baghouse ash. This sample yielded a toxic equivalency
lower than the baseline, although it was not significantly different. The
2378-TCDD equivalencies were all below 0.10 ng/g.
The concentrations of the specific CDD/CDF congeners in the ash are shown
in Tables 2-26 and 2-27. Each table shows the results for all the samples
taken at a single sampling location. Congener distributions are presented in
Figure 2-13. Tables of the distributions are in Appendix A. In Figure 2-13,
the baseline congener distributions are distinguished by entirely shaded
areas. All the distributions appear very similar to baseline.
lmo/036 2-57
-------
TABLE 2-26. CDD AND CDF RESULTS FOR CYCLONE ASH AT CONTROL DEVICE EVALUATION CONDITIONS
TEST CONDITIONS
Combustor
Load (Ib/hr steam)
Excess Air
Overf ire Air
Distribution
Control Device
Quench Reactor Outlet
Temperature
Run
11B
Normal
Normal
Normal
High
Number
Baseline
Normal
Normal
Normal
Normal
CDD/CDF CONCENTRATION (ng/g)
Isomer
DIOXINS
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD/CDF
Run
11B
b
(0.001)
(0.003)
[0.020]
0.005
0.055
[0.012J
0.124
0.013
0.032
0.030
0.142
0.168
0.164
0.170
0.903
[0.022]
[0.138]
0.413
[0.04]
0.430
0.022
0.026
0.168
0.041
0.019
[0.011]
(0.001)
0.049
0.061
(0.003)
0.007
[0.009]
1.24
2.14
Number
Baseline
(0.001)
0.014
0.056
0.014
0.164
0.040
0.388
0.026
0.084
0.076
0.764
0.383
0.341
0.456
2.81
0.004
(0.444)
0.712
0.160
0.378
0.032
0.070
0.202
0.042
0.020
0.041
0.002
0.089
0.065
0.002
0.017
0.044
1.88
4.69
aBaseline is average of Emission Test and Method Study results.
Not detected. Detection limit given in parentheses; estimated maximum possible
concentration (EMPC) given in brackets.
2-58
-------
TABLE 2-27. CDD AND CDF RESULTS FOR BAGHOUSE ASH AT CONTROL DEVICE EVALUATION CONDITIONS
Run Number
TEST CONDITIONS
Combustor
Load (Ib/hx steam)
Excess Air
Overfire Air
Distribution
Control Device
Quench Reactor Outlet
Temperature
10
Normal
Normal
Normal
Low
11A
Normal
Normal
Normal
High
11B
Normal
Normal
Normal
High
CDD/CDF CONCENTRATION
Isomer
DIOXINS
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Total CDD
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Total CDF
Total CDD/CDF
10
[0.025]
[0.022]
0.106
(0.001)
0.126
0.013
0.117
0.012
0.030
0.038
0.262
0.189
0.149
0.260
1.30
(0.001)
0.033
1.80
[0.11]
1.26
[0.044]
0.047
0.407
0.053
0.022
[0.021]
(0.003)
0.102
0.071
(0.003)
0.015
[0.013]
3.80
5.11
Run
11A
(0.003)
[0.029]
0.024
0.008
0.158
0.017
0.151
[0.013]
0.030
0.032
0.129
0.200
0.170
0.299
1.22
(0.003)
[0.574]
2.50
[0.11]
1.99
0.047
0.056
0.494
0.064
0.028
0.027
(0.003)
0.138
0.085
(0.003)
0.021
0.023
5.47
6.69
Number
11B
(0.003)
[0.054]
0.194
0.018
0.200
[0.025]
0.262
[0.017]
0.040
0.060
0.182
0.282
0.243
0.422
1.90
0.017
[0.950]
3.54
0.260
2.99
0.102
0.099
0.881
0.088
0.046
0.036
(0.003)
0.154
0.103
[0.006]
0.020
[0.018]
8.33
10.2
Baseline
Normal
Normal
Normal
Normal
(ng/g)
Baseline
(0.001)
0.025
0.142
0.020
0.270
0.055
0.550
0.052
0.099
0.090
1.10
0.622
0.605
0.959
4.59
0.010
0.276
2.59
0.516
1.42
0.086
0.189
0.700
0.094
0.048
0.157
(0.048)
0.192
0.148
0.024
0.046
0.083
6.58
11.2
aBaseline is the average of Emission Test and Method Study results.
Not detected. Detection limit given in parentheses; estimated maximum possible
concentration (EMPC) given in brackets.
2-59
-------
CDD
e
<3
3
CJ
a
o
a
c
a.
v
"o
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
•1
A B
K
,
,
"
,
I
o P q
Baseline
T
S p
p< ^ K
r *> m^
J? s E
E " v E
• * ' S M r-f~* ~^
C D E
' '
N F^
> '
)
bj '
;:
R S
E3
-
'
/
,
,
/
/
/
:
T
I
,
^c*n
F
E
i
1^1
U \
lun 10
•
1
k
p
B
I
i<
c
^
^
/
/
/
'
/
r^
;i
s
.
s
v
s
D
^^rf-, J~M^
^^n ^^..Ki i ^^S^^pJ
H I J
F
r«rw _-, — • _ B*
W X Y Z A/
rTTl Run 11A
v
' ^
' V
s
K
ft
i
^
f>
m
|
i
K
K
I
I?TTI
AB
C
__
x
/
/
/
/
/
/
^
^.
•^
s
s
s
s
^
f
S,
•
I
1
1
N
kc
3 t
_
'
/
/
i
f
?i
i—
\
•s
S
v
S
II
J
M
K
•)
Hv
HQ
V
K
K
^\
p
) ;
Q 1.
7]
1
IE
IB
KEY
Code Congener
Dtoxiiu
A Mono-CDD
B Di-CDD
C Tri-CDD
D 2378 TCDD
E Other TCDD
F 1 2378 PCD[
G Other PCDD
H 1 23478 HxCI
I 1 23678 HxCl
J1 TnQO .J rl
IZo/oB nxCl
K Other HxCD[|
L 1 234678 Hp]
M Other HpCDC
N Orta-PDD
O Mono-CDF
P Di-CDF
Q Tri-CDF
R 2378 TCDF
S Other TCDF
T 12378 PCDF
U 23478 PCDF
V Other PCDF
W 1 23478 HxC]
X 123678 HxC
Y 234678 HxC
Z 1 23789 HxCl
AA Other HxCDF
AB 1 234678 Hp:
AC 1 234789 Hp:
AD Other HpCDF
AE Octa-CDF
Figure 2-13. CDD/CDF Congener Distributions for Baghouse Ash
During the Control Device Evaluation
2-60
-------
3.0 CONCLUSIONS
The specific objectives of the characterization phase of the test
program, as discussed in Section 1.1, were achieved. Combustion parameters
and acid gas removal efficiencies were characterized for baseline operation.
The effects of load, excess air, and overfire air distribution were
determined to be minimal. CO emissions from baseline to worst case conditions
ranged from 11 ppmV, dry, normalized to 12 percent CO,, to 17 ppmV, dry,
normalized to 12 percent C0«.
S0? and HC1 removal efficiencies for the quench reactor/fabric filter
(QR/FF) emission control system were determined to be a function of quench
reactor outlet temperature and stoichiometric ratio. The effect of quench
reactor outlet temperature and stoichiometric ratio were not determined
separately due to the configuration of the quench reactor system. The HC1
removal efficiencies during the control device evaluation ranged from 97.6
percent at the lowest temperature condition (262 F) to 75.8 percent at the
highest temperature condition (360 F).
Removal efficiencies for S0« during the control device evaluation ranged
from zero at the highest temperature condition (360 F) to 72.9 percent at the
lowest temperature condition (262 F). However, S09 removal efficiencies were
highly variable due to the variability of the uncontrolled combustor
emissions. Also, SO,, was less effectively removed than HCl by the QR/FF
control system.
The quality assurance objectives for precision, accuracy and completeness
were met.
lmo/036
-------
4.0 PROCESS DESCRIPTION AND OPERATION
4.1 PROCESS DESCRIPTION
Ogden Martin operates two mass-burn waterwall combustors at the Marion
County Solid Waste-to-Energy Facility. Each unit has a design capacity of
250 Mg/day (275 tpd) of municipal solid waste. The furnaces are equipped
with Martin reverse-reciprocating stoker grate systems. The combustion
chambers are refractory-lined to a level of 9 m (30 ft) above the stoker.
Refuse is trucked to the facility and dumped into an enclosed receiving
pit. It is subsequently transferred to each combustor by overhead cranes.
Then, the solid waste passes downward through the feed chute and is pushed
onto the stoker grate by a hydraulically operated ram feeder.
4.1.1 Combustor Description
The combustor system is designed to operate at 90 percent excess air.
During baseline testing conditions, the combustor operated at about 70
percent excess air. Underfire air is supplied via five air plenums and
controlled by the pressure drop across the grate bars. Overfire combustion
air, which is typically 25 to 30 percent of the total air, is injected
through three rows of nozzles above the stoker at the front and rear walls of
the combustor at design pressures exceeding 4980 Pa (20 in. W.C.).
The combustion chamber is designed to sustain a flue gas temperature of
980°C (1800°F) for 2 seconds when solid waste is present on the stoker,
including startup and shutdown. To ensure that these time and temperature
specifications are maintained, each combustor is equipped with natural gas
auxiliary burners with an individual capacity of 13 MW (45 million Btu/hr)
located above the combustion chamber refractory lining.
The boiler system is a multi-pass design with a gas-tight membrane
waterwall design. From the top of the combustion chamber, the flue gas flows
downward through an open radiation pass before entering the evaporator tubes
4-1
-------
in the two-drum, boiler convection section. Superheater and economizer
sections follow, each in its own pass. Each combustion unit generates a
maximum continuous steam output of 30,000 kg/hr (66,400 Ib/hr) at a pressure
of 4520 kPa (655 psig) and a temperature of 370°C (700°F). The steam is
delivered to a 13.1 megawatt (45 million Btu/hr) turbine generator. The
electricity produced flows into the Portland General Electric Company (PGE)
grid.
The Martin combustion system consists of an oxygen (0?) controller that
controls the feeder and the grate speed, and a steam load controller that
controls the underfire air dampers. When the 0_ level is above a given set
point, waste feeding begins, and when the ()„ level is low, feeding stops. As
the feed rate increases, steam flow increases and the underfire air dampers
gradually close, reducing the flow of 0 ' As the 0 level is lowered, the
feeding rate slows. This system is self-modulating and is representative of
state-of-the-art combustion controls.
Bottom ash and grate siftings are discharged into a water-quenched
residue system. The ash disposal system consists of vibrating conveyors and
belt conveyors, which transport the residue to an enclosed storage area where
it is eventually trucked to a sanitary landfill for final disposal. Ash from
the cyclone and fabric filter is collected separately and conveyed to the ash
removal system to be handled and disposed of together with the bottom ash.
4.1.2 Emission Control System
The air pollution control system at the Marion County Solid Waste-to-
Energy Facility consists of a cyclone, quench reactor (spray dryer), a dry
venturi, and a fabric filter (baghouse). The flue gases leave the economizer
section at temperatures between 199°C to 270°C (390°F to 515°F) and enter the
bottom of the quench reactor through a cyclonic inlet where removal of
3
oversize particles takes place. Gas flowrates vary between 1636 m /min
(57,750 acfm) at 199°C (390°F) and 1885 m3/min (66,560 acfm) at 270°C
(515°F). Slaked pebble lime slurry is injected through an array of five
two-fluid nozzles near the bottom of the reactor vessel. The slurry water
4-2
lmo/036
-------
3
feed rate is approximately 0.05 to 0.07 m /min (12.8 to 18.2 gpm). The feed
rate is varied to maintain the quench reactor outlet temperature within an
operating range of 125-149°C (258-300°F). The stoichiometric ratio of lime
to HC1 is maintained at approximately 2 to 2.5 to ensure that upset peaks are
sufficiently controlled. The system is designed so that the stoichiometric
ratio cannot be changed independently of the quench reactor outlet
temperature, but rather is dependent on both the temperature and inlet acid
gas concentration.
The lime concentration in the slurry is held nearly constant.
Therefore, as the slurry feedrate increases so does the dry lime feedrate.
Dry lime is fed by screw feeder to the slurry mixing tank every five minutes.
The screw feeder is turned on until sufficient lime has been fed to the tank
to yield the desired lime concentration in the slurry. The dry lime feed
rate varies between 57 and 193 kg/hr (125-425 Ib/hr).
After the lime slurry is mixed, it is screened to remove large solids,
thereby maintaining a relatively stable specific gravity. The slurry is
pumped to a distribution loop where a portion of it is distributed to the
five nozzles and the remainder is recycled back to the slaker.
A low pressure drop dry venturi is located between the quench reactor
and the baghouse. Tesisorb is injected into the venturi at a design rate of
24 kg/hr (53 Ib/hr).
An Amerthem* reverse air baghouse is installed downstream of the dry
venturi for particulate matter (PM) collection. Each unit consists of six
compartments with 120 bags in each. The fabric filter has a gross air-to-
cloth ratio of 1.69:1 (net 2.31:1). The filter bags are made of a fiberglass
material suitable for flue gas temperatures up to 268°C (515°F). The PM,
lime, and Tesisorb cake on the fabric and must be cleaned off every 60 to 70
minutes. Unspent lime in the filter cake acts as an additional
neutralization mechanism for acid gas collection. PM and Oregon DEQ
condensible emissions are required to be controlled to a level of 69 mg/dscm
(0.03 gr/dscf) at 12 percent CO
4-3
lmo/036
-------
4.2 TESTING GOALS
The purpose of this characterization test was to evaluate the operation
and performance of the MWC system in order to determine:
1. The normal operating envelope of the combustor and resulting quench
reactor/fabric filter (QR/FF) performance over this operating
envelope.
2. The variation in performance of the QR/FF in the control of acid
gases at different control device operating temperatures.
3. The performance of the QR/FF in the control of organic emissions
(CDD/CDF) during combustor shutdown and startup conditions.
Each of these goals was met in a separate phase of the testing program.
During the combustor evaluation phase, the QR/FF control device was operated
at baseline conditions while combustor parameters were varied. During the
control device evaluation phase, the combustor was operated at baseline
conditions while quench reactor operating temperatures were varied.
The process shutdown and startup composed a separate evaluation outside the
characterization testing.
The results of the characterization testing will be used to determine
which of the combustor and control device operating conditions require
additional evaluation in a performance test. While the characterization test
consisted of only flue gas GEM measurements, manual HC1 sampling, and ash
sampling, performance testing will include CDD/CDF, metals, or other flue gas
measurements in addition to CEMs, and a more extensive ash sampling program.
The characterization testing is intended to provide EPA with clues concerning
which operating conditions could potentially result in episodes of higher air
pollution emissions, and to what extent the QR/FF controls these emissions.
4-4
lmo/036
-------
4.3 TESTING MATRIX
4.3.1 Combustor Evaluation
Three primary combustor operating variables were selected for
evaluation. These variables were:
1. Steam load,
2. Excess air, and
3. Overfire air distribution
A matrix of combustor evaluation test conditions is presented in Table 4-1.
Baseline conditions were evaluated during the first two days (Runs 1 and 2).
After establishing baseline conditions, five test conditions were evaluated.
The conditions were low excess air (Run 3A), high excess air (Run 3B), low
overfire air distribution (Run 4), high overfire air distribution (Run 5),
and low steam load (Run 6A). The low steam load operating condition was
maintained for Run 6B through Run 9 and the excess air and overfire air
distribution were varied again. The resulting operating conditions were, in
addition to low steam load for each test, low excess air (Run 7), high excess
air (Run 6B), low overfire air distribution (Run 8), and high overfire air
distribution (Run 9). It should be noted that during each of the runs (1-9),
the quench reactor/fabric filter was operating at baseline conditions with a
quench reactor outlet temperature set point of 300 F. With one exception,
which will be discussed later, the 300 F temperature was generally maintained
within ±5 F throughout the test runs.
Baseline operating conditions were established in meetings with the
facility owner/operator prior to the testing. For each of the three primary
combustor variables baseline conditions were reported to be:
Steam load - 66,400 Ib/hr
Excess air - 70 percent
Overfire air - 25 percent of total air
4-5
-------
TABLE 4-1. COMBUSTOR EVALUATION TEST MATRIX
Run #a
1
2
3A
3B
4
5
6A
6B
7
8
9
Description Steam Load
Baseline
Baseline
Low EAd
High EA
Low OF distribution
High OF distribution
Low load
Low load/high EA
Low load/low EA
Low load/low OF air
Low load/high OF air
N°
N
N
N
N
N
L
L
L
L
L
Excess Air
N
N
L6
Hf
N
N
N
H
L
N
N
OFb Air
Distribution
N
N
N
N
L
H
N
N
N
L
H
Runs 3A and 3B and Runs 6A and 6B were labelled to distinguish separate runs
performed on one given test day.
b.
OF - Overfire air distribution
N = normal
EA = Excess Air
L — low
fH = high
4-6
-------
Target values for the combustor variables during characterization testing
were established also. The low steam load target was established at 75
percent of normal, or approximately 50,550 Ib/hr. The low and high excess
air target values were 44 percent and 110 percent, respectively. The low and
high overfire air distribution target values were approximately 0 and
30 percent, respectively. With the exception of low steam load conditions,
the target amount of variation from baseline for each of the primary
combustor operating parameters was established for specific test runs by
evaluating flame patterns in the fire box and determining the resulting
operating conditions.
Steam load (Ib/hr), total combustion air flow (10 Ib/hr) and percent 09
at the boiler outlet were monitored directly from readouts in the control
room. The overfire air flowrate is monitored indirectly by a pressure
setting (in. WC) in each of the supply headers prior to being injected into
the furnace. There are three rows of overfire air nozzles (front, upper
rear, and lower rear).
A list of the process parameters which were recorded during each of the
test runs is provided in Table 4-2. These were generally recorded at
15-minute intervals with the exception of lime slurry specific gravity, which
was a field measurement reported by plant personnel every hour. Strip charts
were copied for those process parameters which were recorded in the control
room. Table 4-3 details the range in primary operating variables that was
measured during each of the Phase I runs.
4.3.2 Control Device Evaluation
The primary control device variable under evaluation was the quench
reactor (spray dryer) outlet temperature. Studies by Environment Canada
indicate that the performance of acid gas control equipment in the removal of
organic and acid gas emissions can be highly temperature dependent. One
Environment Canada study involved temperature variations with a
humidification/dry injection system that provided conclusive results on the
9
removal of these pollutants. The characterization program attempted to
4-7
lmo/036
-------
TABLE 4-2. PROCESS OPERATING PARAMETERS RECORDED
DURING MARION COUNTY TESTING
Parameters Units
Refuse feed rate (Crane weight scale) Ib
Steam flow Ib/hr
Steam pressure psig
Steam temperarture F
Combustion air flow 10 Ib/hr
Combustion air temperature F
Overfire air nozzle pressure in W.C.
- Front
- Upper rear
- Lower rear
0? concentration (boiler exit) % vol. (wet)
Temperatures F
- Middle of furnace 1st pass
- Top of furnace 1st pass
- Economizer outlet
- Quench Reactor inlet
- Quench reactor outlet
- I.D. fan inlet
- Baghouse outlet
Quench reactor inlet pressure in W.C.
Dry lime feed rate (Lime totalizer) Ib
Lime slurry specific gravity
Dry venturi AP in W.C.
Baghouse AP in W.C.
Baghouse cleaning cycle min
Stack opacity %
Furnace draft in W.C.
4-8
-------
TABLE 4-3. TESTED OPERATING RANGE OF PRIMARY OPERATING VARIABLES
Run #
1
2
3A
3B
4
5'
6A
6B
7
8
9
Steam Load
(Ib/hr)
67180
67240
63990
63940
65460
68970
51230
47960
51590
49900
52090
Excess Air
(percent)
72.7
78.4
40.2
106.0
70.6
71.7
75.4
135.6
57.9
84.0
88.7
Flue Gas
Flow Rate
(acfm)
57150
60920
46980
67270
60600
58590
52310
59640
42280
42350
49360
Over fire Air (in. WC)
Front
15.2
15.3
7.0
17.8
4.5
19.3
7.1
13.0
4.6
1.7
10.2
Upper
Rear
6.0
4.9
0.6
9.8
0.4
10.6
0.6
6.6
0
0.2
11.0
Lower
Rear
15.5
15.6
3.7
17.5
0.9
17.5
4.8
11.4
1.2
0.4
12.8
Measured at economizer outlet.
4-9
-------
verify the effect of temperature on control of organic and acid gas emissions
by a commercial scale quench reactor and fabric filter. However, the
operating temperature of the Marion County QR/FF could not be varied
independently at a given stoichiometric ratio as was done in the Environment
Canada study. The flow of lime slurry to the Marion County quench reactor is
adjusted based on the flue gas operating temperature measured at the quench
reactor outlet. As the flow of lime slurry is adjusted, the stoichiometric
ratio varies, since the lime content per unit mass of slurry is constant.
The baseline operating temperature was 300 F. The low operating temperature
target (Run 10) was 260 F, and the higher operating temperature targets
(Runs 11A and 11B) were 330°F and 360°F, respectively. To the extent
possible, normal steady-state combustion conditions were maintained at full
steam load during each of these runs.
The following sections describe the process operations that were
experienced during each of the runs in the characterization testing at Marion
County, with an attempt to highlight any process upsets or unusual operating
conditions that took place.
Run 1 - Baseline
Process operations were very stable during the test run. There was an
observed SO,, spike reported from the GEM instrument trailer that did not
correspond to any specific process variation (such as increased operating
temperature). Observations of the pit led to speculation that the source of
sulfur may have been the large quantities of gypsum sheet rock from
demolition wastes received that day.
Run 2 - Baseline
Process operations were very stable with no upsets.
4-10
lmo/036
-------
Run 3A - Low excess air
Excess air was dropped to a target value of 5 percent oxygen on a wet
basis at the economizer outlet. Steam flow was maintained at design levels
and the overfire air distribution was adjusted in an attempt to maintain the
baseline value of 25 percent total air. Between 1200 and 1230 there was a
blockage on the feed table which caused a secondary fire and unsteady air
conditions. As a result, several CO spikes were reported from the instrument
trailer. Operating at low excess air values resulted in increased furnace
operating temperatures. There was a delay in starting the test because of
excessive fuel bed thickness. This was caused when the ash discharge seal
broke resulting in increased 0_ values which caused the controller to
increase the feeder speed.
Run 3B - High excess air
After making adjustments to primary operating variables to establish the
high excess air operating conditions, the furnace temperatures dropped as
expected. There was some difficulty maintaining the high excess air (0?)
operating conditions early after the transition, and design steam loads had a
tendency to slip about 5-10 percent. The furnace draft was very unsteady,
and the furnace pressure went positive often during the testing period.
Run 4 - Low overfire air distribution
With the exception of 5" W.C. pressure on the front wall nozzles, the
overfire air flows were near zero. The 5" W.C. was necessary in order to
protect the nozzles from flames (provide cooling). Two drops in temperature
without corresponding drops in steam load or 0_ indicated wet fuel.
Run 5 - High overfire air distribution
Relatively stable operating conditions were experienced throughout the
test. No problems or major process excursions were observed.
4-11
lmo/036
-------
Run 6A - Low load
The run was completed without combustion upsets. Erratic lime slurry
flows were experienced due to frequent plugging of the slaker strainer.
Run 6B - Low load, high excess air
The quench pit seal was broken resulting in a spike in CO concentration.
Instrument air was lost twice, resulting in baghouse bypass. The test was
aborted one half hour early when this problem could not be resolved.
Run 7 - Low load, low excess air
The extremely low air flows associated with this operating condition
caused the I.D. fan to have difficulty in regulating itself. The furnace
draft was positive during several episodes, further reducing flue gas
flowrates. The quench reactor did not adjust slurry injection rates and the
quench reactor outlet temperature plummeted from the design target of 300 F to
232 F. This can be observed in the process data, and it resulted in an
increased removal of HCl and SO,- during the episode. Chunks of lime approxi-
mately 1 inch in diameter dropped down into the cyclone ash. The problem was
resolved by slightly increasing gas flowrates and steam load and maintaining
negative draft on the I.D. fan.
Run 8 - Low load/low overfire air
An electrical fault temporarily caused the fuel feeding to stop and the
feeder went to maximum stroke. It was corrected and did not affect testing.
Run 9 - Low load/high overfire air
The plant had problems with the daily 0~ calibration during testing.
Because of disagreement with Radian CEMs, the plant 0. data are considered
suspect. No process upsets were experienced.
4-12
lmo/036
-------
Run 10 - Low QR temperature
No process problems occurred.
Runs 11A/11B - High QR temperature
No process problems occurred.
4-13
-------
5.0 SAMPLE POINT LOCATIONS
The sampling locations are shown on the process line schematic in
Figure 5-1. Each sampling location is discussed in the following sections.
5.1 FLUE GAS
5.1.1 Boiler Outlet (Control Device Inlet) Sampling Location
The parameters that were measured at the boiler outlet (control device
inlet) sampling location include volumetric flowrate, moisture, S0_, HC1, 0
CO, C0_, NO and THC. A top view and side view of the boiler outlet sampling
£-. 3C
location are shown in Figures 5-2 and 5-3, respectively. The sampling
location has three 6-inch ID ports located in a circular duct 6 ft. 10 in. in
diameter. Two of the ports (Ports A and B) are located in the same plane, 90
apart. These ports were used for the manual test methods. The third port
(Port C) is located about two feet downstream on a different axis. This port
was used to extract a fixed point sample for the continuous emission monitors
(CEMs). All the ports have 8-inch-long nipples and are accessible from the
same platform.
EPA Method 1 was used to select the number and location of the traverse
points for Ports A and B. The ports are located approximately 4 equivalent
duct diameters (28'6") downstream of a 90 bend in the duct and approximately
1.9 equivalent duct diameters (13'1") upstream of a 90 bend in the duct.
Following EPA Method 1 procedures, a minimum of 24 traverse points was
required. The traverse point location diagram is presented in Figure 5-4.
A cyclonic flow check of the location was conducted according to EPA
Method 1 and the average degree of rotation was determined to be 5 . EPA
Method 1 specifies that the average degree of rotation should be equal to or
less than 10 . A stratification check was also conducted using NO as an
X
indicator. The difference across the duct was less than 2.5 percent of the
lmo/036 5_1
-------
To Atmosphere
Combustor
Boiler Superheater Economizer
Quench Reactor/ Teslsorb
Acid Gas Feed
Scrubber Hopper
ui
I
CO
02
ID. Fan A stack
O-1
Quench
Pit
VVVV "-—Distributor
I = Inlet location prior to the first control device
M = Midpoint location after quench reactor prior to the baghouse
O-1 = Outlet location in the breeching prior to the stack location
O-2 = Outlet location in the stack
A-1 = Baghouse ash
A-2 = Cyclone ash
A-3 = Economizer ash
A-4 = Superheater ash
Figure 5-1. Marion County MWC Process Line with Sampling Locations
-------
Flue Gas
From Economizer
Sampling
Platform
Inlet
Location
9 PortB
CEM Port C
Port A ! PortA
CEM Port C
Cyclone and Acid
Gas Scrubber
Midpoint
Location
Top View
To
Baghouse
cr
in
o
o
r-~
CO
Figure 5-2. Top View of Boiler Outlet and Midpoint Sampling
Locations at Marion County MWC
5-3
-------
From Cyclone and
Acid Gas Scrubber
28'-6"
Elevation
Above Grade
Direction of
Gas Flow
V
Port B
OEM
Port
2'2'
Side View
Port A
Platform
•To Baghouse
-$>•
Figure 5-3. Side View of Boiler Outlet Sampling Location at
Marion County MWC
tr
3
o
5-4
-------
Port A
17 7/8".
225/8"«
283/4"«
8" nipple
Measurement from the outside of the nipple for probe marking
b Traverse points are located as specified in EPA Method 1
eg
o
Figure 5-4. Traverse Point Location Diagram for Boiler Outlet
Location at Marion County MWC
5-5
-------
reference point, indicating that stratification was not significant at this
location.
The average volumetric flowrate through the duct was 29,400 dry standard
cubic feet per minute (dscfm) at an average temperature of 423°F. The
velocity head reading from the pitot tubes ranged from 0.07 to 0.2 in. H90 in
previous tests, which is in the low range for the manometers that are standard
equipment in Radian meter boxes. Thus, an inclined manometer with a zero to
one inch of water range was used. The velocity head reading remained in that
range during this test program. Static pressure draft at this point in the
system averaged negative 2.3 inches of H~0.
5.1.2 Midpoint Sampling Location
The parameters that were measured at the midpoint sampling location
include volumetric flowrate, moisture, HC1, SO., 0?, and C0_. A top view of
the midpoint sampling location was shown previously in Figure 5-2. A side
view of the midpoint sampling location is shown in Figure 5-5.
The midpoint sampling locations has three six-inch I.D. ports located in
a circular duct 51" in diameter. Two of the ports (Ports A and B) are located
in the same place, 90 apart. The third port (Port C) is located about two
feet downstream on a different axis. All the ports have 8-inch-long nipples.
Port C was used to extract the fixed point sample. Ports A and B were capped
except during pre- and post-test velocity traverses.
EPA Method 1 was used to select the number and location of the traverse
points for Ports A and B. The ports are located approximately 6 duct
diameters (28'6") downstream of a 90 F bend in the duct and approximately 5
equivalent duct diameters (25') upstream of 90 bend in the duct. Following
EPA Method 1, a minimum of 12 traverse points were required for the velocity
traverses. However, to coordinate sampling with the inlet, midpoint and
outlet, 24 traverse points were used. The traverse point location diagram is
presented in Figure 5-6.
lmo/036 5_6
-------
From Cyclone and
Acid Gas Scrubber
28'-6"
Elevation
Above Grade
•39'-10'
~25'
Direction of
Gas Flow
V
-4'3"
Port B
OEM
Port
2'2"
Side View
Port A
Platform
•To Baghouse—I-
tr
O)
(O
CM
O
r-.
co
a>
Figure 5-5. Side View of Midpoint Sampling Location at
Marion County MWC
5-7
-------
55 1/2" • 11
57 7/8" • 12
" nipple
•Measurement from the outside of the nipple for probe marking
'Traverse points are located as specified in EPA Method 1
§
r^
8
Figure 5-6. Velocity Traverse Point Location Diagram for the
Midpoint Location at Marion County MWC
5-8
-------
A cyclonic flow check of the location was conducted according to EPA
Method 1 and the average degree of rotation was 5 . EPA Method 1 specifies
that the average degree of rotation should be determined to be equal to or
less than 10 . A stratification check was also conducted using NO as an
X
indicator. The difference across the duct was less than 9 percent of the
reference point, indicating that stratification was not significant at this
location.
The average volumetric flowrate of the duct was 34,800 dscfm at an
average temperature of 303 F. Static pressure draft at this point in the
system averaged negative 4.9 inches of water.
5.1.3 Breeching to the Outlet Stack
The parameters that were measured at the breeching to the outlet stack
include SO-, HC1, 0 CO, CO NO , and THC. A side view of the breeching to
the outlet stack sampling location is shown in Figure 5-7.
The breeching sampling location has three four-inch ID ports located in a
rectangular duct 7 ft. 4 in. high by 3 ft. deep. All of the ports have
4-inch-long nipples. The ports were accessed by temporary scaffolding.
The ports are located approximately 18 inches upstream of dampers in the
ducting and therefore the location does not qualify as an EPA Method 1
location. However, only fixed point gaseous samples were extracted from the
breeching. A stratification check was performed using NO as an indicator
using the point location diagram shown in Figure 5-8. Since the HC1 probe was
fixed permanently in Port B, the stratification check was performed using only
Ports A and C. The difference across the duct was less than 2 percent of the
reference point, indicating that stratification was not significant at this
location. A cyclonic flowcheck conducted according to EPA Method 1, indicated
that the average degree of rotation was 2 . EPA Method 1 specifies that the
average degree of rotation should be equal to or less than 10
lmo/036 5-9
-------
^Diameter = 4"
® Port A
-13ft
Side View
cc
in
g
co
CD
Figure 5-7. Breeching to the Stack Sampling Location at
Marion County MWC
5-10
-------
J
3
i
<
4
nir
a
L
O7 "
O I
0-1 It
OR "
6"
10 " ..
IU
-1O"
7 " -.
'
I"
)ole
I I I
I ' I I ' I I I I
Port C " Port B " Port A
^ ... .nn» — — — fc-
Measurement from the outside of the nipple for probe marking
Figure 5-8. Stratification Point Location for the Breeching Location
at Marion County MWC
DC
00
CM
O
CO
cn
-------
The average volumetric flowrate of the duct was 37,400 dscfm at an
average temperature of 287 F. Static pressure at this point was 0.35 inches
WC for Runs 1 to 6B, 8.0 inches WC for Runs 6 to 10 and 0.50 inches WC for
Runs 11A and 11B.
5.1.4 Outlet Stack Sampling Location
The parameter that was measured at the outlet stack sampling location was
volumetric flowrate. A top view and side view of the outlet stack sampling
location are shown in Figures 5-9 and 5-10, respectively.
The outlet stack sampling location has three 4-inch ID ports located in a
circular duct 48" in diameter. Two of the ports (Ports A and B) are located
in the same plane, 90 apart. The third port (Port C) is located about two
feet downstream on a different axis. All the ports have 4-inch-long nipples.
Ports A and B were used, but Port C was capped since no fixed point
sampling was conducted at this location. EPA Method 1 was used to select the
number and location of the traverse points for Ports A and B. The ports are
located approximately 13 equivalent duct diameters (60') downstream of the
breeching and approximately 36 equivalent duct diameters (170') upstream of
the top of the stack. Following EPA Method 1, a minimum of 12 traverse points
were required. The traverse point location diagram is presented in
Figure 5-11.
A cyclonic flow check was conducted and the average degree of rotation
was confirmed to be less than 10° as specified by EPA Method 1. A stratifi-
cation check was not performed at this location since only velocity traverses
were collected.
lmo/036 5-12
-------
Ul
I
Top View
of Stack
27" wide
Door
Ladder
Door
30" wide
Figure 5-9. Outlet Stack Sampling Location at Marion County MWC
DC
s
s
r-
co
o>
-------
To
Atmosphere
250' L
Stack
Plant
CEM Port
Port A
Port C
Breeching
7'4"
f '/'/'/
4'3"
170'
Port B
1 _ 4 Inch Ports
< L 5 Inch Nipple
60'
20'
Ground Level
Figure 5-10. Side View of Outlet Stack Sampling Location
at Marion County MWC
cc
to
03
CM
5-14
-------
'Measurement from the outside of the nipple for probe marking
^Traverse points are located as specified in EPA Method 1
Figure 5-11. Velocity Traverse Point Location Diagram for the
Outlet Stack Location at Marion County MWC
5-15
-------
5.2 ASH AND PROCESS SAMPLES
5.2.1 Superheater Ash Sampling Location
The superheater ash was collected from the ash hopper before the ash
dropped on the conveyor to the quench pit. At this point in the system, the
draft is negative and a special sampling apparatus was required. A galvanized
metal trier was inserted into the base of the hopper to collect the falling
ash. The trier was withdrawn periodically to empty the ash. The side and
top views of the superheater ash sampling location and sampling apparatus are
shown in Figures 5-12 and 5-13, respectively.
5.2.2 Economizer Ash Sampling Location
The economizer ash sampling location was very similar to the superheater
ash sampling location. The ash was collected from the ash hopper using the
same type of sampling apparatus as used for the superheater ash. The side
view of the economizer ash sampling device and location is shown in
Figure 5-14 and the top view is the same as was shown in Figure 5-13.
5.2.3 Baghouse Ash and Cyclone Ash Sampling Locations
The sampling locations for the Unit No. 1 baghouse ash and cyclone ash
are shown in Figure 5-15. The baghouse ash was collected from a screw
conveyor at an intermediate transfer point before mixing with the cyclone ash.
A hole was cut in an access plate and a sliding cover was bolted over the hole
for easy access.
The cyclone ash was collected before mixing with the baghouse ash. A
sliding cover was also made for the cyclone ash access plate.
lmo/036 5-16
-------
Superheater
Ash Hopper
<5
Flue Gas to
Economizer
Access Port
Ash Sampling
Trier
Rotary Valve
Rotary Valve
Screw Conveyor to Quench Pit
cr
03
§
Figure 5-12. Side View of Superheater Ash Sampling Location
at Marion County MWC
5-17
-------
Ash Hopper Walls
Ash Sampling Trier
Ash Hopper
cr
CO
Figure 5-13. Top View of Superheater Ash Sampling Location
at Marion County MWC
5-18
-------
Flue Gas from
Superheater
<1
Economizer
Access Port
Ash Sampling
Trier
Rotary Valve
Rotary Valve
Screw Conveyor to Quench Pit
o
r-.
8
Figure 5-14. Side View of Economizer Ash Sampling Location
at Marion County MWC
5-19
-------
Flue Gas
to I.D. Fan
Baghouse
Ul
I
Quench Reactor/
Acid Gas
Scrubber
Incinerator
Building
Flue Gas from Economizer
Screw Conveyor
Screw
Conveyor
Baghouse Ash
Sampling Location
Cyclone Ash
Sampling Location
Hole in Plate
Pivot Point
Handle
Bottom View of Ash
Sampling Locations
Figure 5-15. Baghouse Ash and Cyclone Ash Sampling Locations at
Marion County MWC
DC
TJ
O
r-
O
r«-
oo
en
-------
5.2.4 Lime Slurry Sampling Location
The lime slurry samples were collected from the recycle hose on the lime
slurry mixing tank. The mixing tank is accessible from the second floor of
the area housing the lime slurry injection system.
5.2.5 Tesisorb Sampling Location
The Tesisorb samples were collected from the feed hopper to the injection
system. A small plate was removed on the hopper to collect the samples. The
sampling location is shown in Figure 5-16.
5-21
-------
Tesisorb from
Main Hopper
Tesisorb Sampling
Port
Lift off cover
and scoop out
Tesisorb
Feed Hopper
for Unit #2
Feed Hopper
for Unit #1
Figure 5-16. Tesisorb Sampling Location at Marion County
5-22
-------
6.0 SAMPLING AND ANALYTICAL PROCEDURES
The sampling methods used for the Marion County Characterization Test
were based on accepted EPA protocols. Modifications were made to suit the
needs of the test program. The sampling methods and pertinent modifications
are discussed below. Additional details of the sampling and analytical
procedures are included in the test plan.
6.1 CONTINUOUS EMISSION MONITORS (CEMs)
An extractive system was used to obtain flue gas samples for the CEM
systems. The sample was withdrawn continuously at a single point from the
stack and transferred to the CEM trailer through heat-traced teflon line. The
flue gas was conditioned (temperature lowered and moisture and particulate
removed) before the flue gas stream was split using a manifold to the various
analyzers.
CEMs were used to analyze flue gas from three locations: the control
device inlet (boiler outlet), the midpoint location (quench reactor outlet)
and the control device outlet (stack breeching). The flue gas was analyzed
for C0_, 0-, and SO- at each location. CO, NO and THC were monitored at the
^ ^ £. X
inlet and outlet only. HC1 concentrations were also monitored continuously by
Entropy Environmentalists, Inc., at the inlet, midpoint and outlet but are not
discussed in this report. t
The CEM equipment and sampling locations were standard systems, except
that modifications were made to the midpoint sampling location.
Stratification checks of the flue gas were also made. These site-specific
modifications are discussed below. Refer to Sections 3.5 and 4.1 of Reference
11 for more details on the sampling methods.
lmo/036 6"1
-------
6.1.1. Sampling at the Midpoint Location
The control device midpoint sampling location at the Marion County Solid
Waste-to-Energy Facility is situated downstream from the quench reactor and
prior to the Tesisorb injection system. Thus, flue gas extracted from the
midpoint is unusual from a sampling point of view in several respects:
1. Reactions between the injected lime and acid gases (primarily HCl
and S0_) in the flue gas occur in the zone between the quench
reactor and the baghouse. Due to turbulent flow and changing acid
gas concentrations, conditions at the midpoint are non-steady state
with respect to the reactions of interest.
2. Unreacted lime in the gas stream tends to adhere to the walls of the
sample vessel and may react with acid gases in the sample, creating
a bias.
3. There is an increased moisture content in this area due to the
injection of slaked lime. Condensate in the sample path could cause
undesired reaction of acid gases.
4. The use of a filter in the sample line is undesirable because acid
gas scrubbing would occur if a lime filter cake built up in sample
path.
In order to minimize these problems, a specially designed gas
conditioning system was used. The midpoint sample probe, particulate
reduction system, and moisture reduction system are shown in Figure 6-1.
Particulate is reduced in the extracted sample in two ways. First, the sample
probe intake is positioned away from the gas stream flow. Second, the
extracted sample passes through two cyclones. Following these particulate
reduction steps, the sample is divided by a manifold to the manual HCl
sampling train, the continuous emissions monitors and to the Entropy HCl
continuous monitor. The Radian continuous monitoring system then uses a
system of upright condensers and knockout impingers in an ice bath to reduce
moisture with minimal contact of the gas and condensate.
Overall, the system worked well. The residence time through the system
was not significantly increased due to the sample conditioning set-up. Leak
lmo/036 6"2
-------
t
HCI
^nntinnni
Heate
Sample I
\
Gas
Conditione
Q
d
Jne
^
-tr-
r Met£
rallhr
Dilution Air— i
II
Entropy's HCI
Dilution Probe
»red Orlface —M_
atlnn f5aQ — **
Sintered
Metal Frit
\
' -.
Heated
Sample Lines
Direction of
Flue Gas
Monitor
CO
SQ2 Analyzers
CO2 Analyzers
Ot Analyzers
Overflow
Heated
Sample
Probe
T
Method 5
Filter
(Heated)
Temperature
Controller
\
Heating Element
§
eg
o
r-
Figure 6-1. GEM Sampling and Analysis Scheme for the Midpoint Sampling
Location for the Marion County MWC
-------
problems were seldom encountered and were easy to correct when they occurred.
Acid and fixed gas concentrations were in the expected ranges and compared
logically to inlet and outlet concentrations. Oxygen concentration
consistently increased from inlet to midpoint to outlet and pollutant gas
concentrations consistently decreased. Also GEM SO- concentrations compared
favorably with manual method 6 runs during the interference tests. Orsat and
GEM values for CO- and 0? compared closely as well. The main problems
encountered with the midpoint sampling system are listed below:
1. S09 system bias checks at the midpoint show an average system bias
for SO. of 16 percent. This bias was probably due to SO- reaction
with adsorbed lime and/or leakage.
2. Fine particulate which passed through the cyclones caused the pump
for the continuous monitoring system to fail. The pumps were
replaced and rebuilt between runs to ensure uninterrupted sampling.
3. The manual method HCl train filter housing was installed backwards
due to the configuration of the conditioning system. An evaluation
of the bias for the manual method filter indicated that the negative
bias was 22 percent.
For future sampling at the Marion County facility, modifications to the
midpoint sampling system should be considered. One suggestion would be to
improve the initial particulate reduction system. Possibilities include: a
smaller cyclone in series with the existing ones, an improved probe design, or
perhaps even electrostatic methods. Additionally, frequent cleaning of the
system would avoid undesirable buildup in the system. The manifold should
also be modified to accommodate the filter for the HCl train.
6.1.2 Stratification Check
As an indication of stratification (incomplete mixing of the flue gas)
the inlet, midpoint, and breeching sampling locations were traversed using the
GEM probes. The test plan originally specified SO- as the indicator of
stratification. However, during the initial attempts the SO- concentrations
varied significantly with feed causing too much variation at each traverse
point. The indicator was then switched to NO .
lmo/036 6"4
-------
Two probes were used during a stratification check. The first probe was
located at a fixed point and was the reference probe. The second probe was
traversed across the duct collecting approximately 5 minutes of data at each
point. An average was calculated at each point for each probe. Relative
differences between each probe at each point should be less than 10 percent.
6.1.3 Averaging Method
GEM data were reported as approximately 1-minute averages. The Radian
data acquisition system used for this test program scanned each channel 1700
times per minute and then stored a 1700-scan average in memory. Depending on
the available space in memory, storing the data took a variable amount of time
varying by a few seconds. Thus, the 1700-scan averages were stored
approximately every minute, rather than exactly on a minute interval.
The 1-minute averages were averaged every hour to generate three to four
hourly averages per test run. Each hour interval was 90 percent complete (54
of 60 readings) to be considered valid and acceptable.
In the event that an hour interval was determined to be unacceptable for
a critical parameter (SO. and HC1 at all locations, CO and CL at inlet) the
test run was extended for additional hour intervals until a minimum of two
acceptable intervals were collected.
During the transition period between test conditions, monitoring was
continued. Therefore, the GEM analyzers were calibrated at the beginning and
end of each test day rather than for each test condition.
6.2 MANUAL METHODS
6.2.1 HC1 Determination
HC1 sampling was based on EPA Reference Method 5 with modifications that
allowed collection of HCl in the back half of the sampling train. Further
development of this method is currently underway. Thus, the method chosen was
the current consensus of the sampling community. The method is described in
Section 4.2.2 of Reference 11.
lmo/036 6-5
-------
6.2.1.1 Manual HC1 Sampling. The following program-specific changes
were required for manual HC1 sampling at all locations for the
characterization test program:
1. The sampling rate was between 0.2 to 0.3 acfm.
2. Sampling was not isokinetic.
3. Sampling was fixed point.
4. Particulates were not quantified in the HC1 trains.
5. A glasswool plug was placed in the glass probe liner for the outlet
train. No filter was used.
6. For the inlet train, a filter was used with no glasswool in the
probe liner.
7. The front half of the sampling trains was not recovered. The
glassware was rinsed with distilled water to remove particulate and
the rinses discarded.
8. Buttonhook nozzles were not used.
9. Sampling was conducted for 3 hours.
10. The pitots at the inlet location were blown back every 15 minutes
due to the high particulate loading.
6.2.1.2 HCl Analysis. Both on-site and laboratory analyses of the HC1
samples were performed for this test program. Aliquots of the samples were
analyzed by specific ion electrode (SIE) on-site. The analyzed aliquots were
saved and later reanalyzed by ion chromatography (1C) in the laboratory. The
samples were evaluated for matrix interference by the method of additions
using SIE.
6.2.2 Volumetric Flowrate Determination
The volumetric flowrate of flue gas was measured according to EPA
Method 2. The flowrate was determined at the inlet, midpoint, and outlet
sampling locations both prior to and at the completion of each test run.
lmo/036 6"6
-------
6.2.3 Moisture Determination
The average flue gas moisture content was determined according to EPA
Method 4. This is discussed in more detailed in Section 4.2.5 of
Reference 11.
6.2.4 Fixed Gases Determination
The molecular weight and CCL and CL content of the flue gas were
determined according to EPA Method 3 using ORSAT values. This is discussed in
more detail in Section 4.2.7 of Reference 11.
6.2.5 SO. Determination
Manual sampling and analyses for SO,, in the flue gas followed EPA
Method 6. The method was modified to use full-size impingers. This method
is presented more fully in Section 4.2.8 of Reference 11.
6.2.6 Ash Sampling
The sampling methods for the baghouse ash and cyclone ash are described
in Sections 3.4.5 and 4.2.3 of Reference 11. The sampling method for the
economizer ash and superheater ash was modified from those methods because of
negative draft at the sampling locations. The economizer ash and superheater
ash were sampled by placing a galvanized metal trier trough in the ash hopper.
This collected a continuous-grab sample of the falling ash. The trier was
emptied periodically and repositioned back in the hopper. Vacuum suction
sampling methods were attempted earlier at these sampling locations but
insufficient ash was collected. The collected grab samples were composited in
the same manner as for the baghouse ash and cyclone ash.
The analytical methods for determining CDD and CDF are described in
Section 5.5.1 of Reference 12. Both screening and confirmation analyses were
performed. The confirmation results for 2378-TCDF were used for each sample.
The confirmation results for 2378-TCDD were used only if less interference was
present. This was determined by comparing the screening and confirmation
results and selecting the lower value.
lmo/036 6'7
-------
7.0 INTERNAL QUALITY ASSURANCE/QUALITY CONTROL (QA/QC)
Internal and external quality assurance and quality control procedures
were strictly adhered to during this test program to ensure the production of
useful and valid data throughout the course of the project. Internal QA/QC
checks and procedures represent an integral part of the overall sampling
scheme. The results of Radian's internal quality assurance/quality control
program are presented in this section and in Appendix H. The results of the
external QA performed by Entropy Environmentalists, Inc., are presented in a
separate report.
7.1 QUALITY ASSURANCE OVERVIEW OF THE MARION COUNTY TEST PROGRAM
The Marion County Characterization test program was organized such that
the quality assurance function allowed complete independence in program
review. Radian's Quality Assurance Officer reports directly to the Radian
Program Manager for internal QA and Entropy Environmentalists, Inc., reported
directly to the EPA/EMB Task Manager for external QA. The primary QA/QC
program objective was to provide data of known quality with respect to
accuracy, precision, representativeness, and completeness. The QA/QC approach
focussed heavily upon controlling measurement data within established
acceptance criteria.
Internal QA conducted by Radian personnel centered around well-documented
methodologies which included detailed procedures for sampling and analysis,
calibrations, labeling sample containers, preparation and cleaning of sample
containers, sample preservation and storage, quality assurance, and quality
control samples. In order to maximize comparability of measurement data,
standard reference methods, including EPA and ASTM methods, were used whenever
possible. A chain-of-custody system was established which provides a
documented history of each sample and provides assurance that the integrity of
the samples was maintained throughout the course of sample collection,
handling, and analysis.
lmo/038 7-1
-------
The various data reduction, validation, and reporting tasks were defined
during initial project organization in order to meet the objectives of the
program. Specific responsibilities were assigned to various members of the
project team. In general, the Task Leaders were assigned primary responsi-
bility for data reduction, validation, and reporting requirements for their
respective tasks, and the Lead Technical Coordinator provided overall review
and coordination of the reporting efforts. Following initial data reduction,
daily data summaries were prepared and submitted to the EPA Task Manager.
These data summaries were used as input to the final report.
External quality assurance (QA) played a key role in the Marion County
Test Program. Entropy Environmentalist, Inc., provided an independent assess-
ment of the critical measurement systems by conducting performance evaluations
using apparatus and/or standards that were different from those used to
calibrate or collect the measurement data. The goal of the external audits
was to evaluate the potential of the measurement systems to produce data of
adequate quality to satisfy the objective of the test program. Upon
completion of each performance audit, the auditor(s) discussed any specific
weaknesses with the project team and made recommendations for corrective
action. An audit report was subsequently prepared and distributed to the
EPA/EMB Task Manager. The audit report outlines the audit approach and
presents a. summary of results and recommendations.
7.2 QA/QC OBJECTIVES AND RESULTS
The overall quality assurance/quality control (QA/QC) objective was to
ensure precision, accuracy, completeness, and representativeness for each
parameter measured in this test program. These data characteristics are
defined as follows:
o Precision - A measure of mutual agreement among individual measure-
ments of the same property, usually under prescribed similar
conditions. Precision is best expressed in terms of the standard
deviation (or the relative standard deviation). Various measures of
precision exist depending upon the prescribed conditions.
lmo/038 7-2
-------
o Accuracy - The degree of agreement of a measurement (or an average
of measurements of the same thing) , X, with an accepted or true
value, T, usually expressed as the difference between two values,
X-T, or the difference as percentage of the reference or true
value, 100 (X-T)/T, and sometimes expresses as a ratio, X/T.
Accuracy is a measure of the bias in a system.
o Completeness - A measure of the amount of valid data obtained from
a measurement system compared with the amount that was expected to
be obtained under the prescribed test conditions.
o Comparability - A measure of the confidence with which one data
set can be compared with another.
o Representativeness - The degree to which data accurately and
precisely represent a characteristic of population, variation of a
parameter at a sampling point, or an environmental condition.
A summary of the estimated and achieved precision, accuracy, and
completeness objectives is presented in Table 7-1. A more detailed
discussion can be found throughout this section of the report.
In general, the precision and accuracy of the continuous emission
monitors was well within the QC criterion shown in Table 7-1. In fact, the
day-to-day precision, expressed as the percent coefficient of variation
(Standard deviation/mean), was less than 3 percent for all analyzers except
S09 midpoint and THC outlet. The accuracy of the CEMs was within the QC
objective of ±10 percent for all monitors (0.9 - 5.8%). The accuracy of the
chloride analyses was also acceptable with a mean absolute relative error of
2.7 percent.
Table 7-2 is a summary of the QC checks and corresponding acceptance
criteria, control limits, and corrective actions that were followed during
this program. The criterion are based on the methods and the data used to
calculate the achieved values can be found in the appendices of this report.
lmo/038 7-3
-------
TABLE 7-1. SUMMARY OF ESTIMATED AND ACHIEVED PRECISION,
ACCURACY, AND COMPLETENESS OBJECTIVES3
Precision
Parameter
CDDb
CDFb
Estimated
±40%
±40%
Achieved
±28%
±22%
Accuracy
Estimated
±50%
±50%
Achieved
NA°
NA°
Completeness
Estimated
90%
90%
Achieved
100%
100%
d e
Continuous Emission Monitors '
Inlet:
0
C09
CO
THC
NO
soj
High Range
Midpoint:
0
CO
S°2
Outlet:
CO
S°2
NO
THC"
CO
°2
Velocity/
Volumetric
Flowrate
±10%
±10%
±10%
±10%
±10%
±10%
SO ±10%
±10%
±10%
±10%
±10%
±10%
±10%
±10%
±1Q%
ND
±6%
0.65%
2.56%
1.10%
2.44%
1.98%
1.81%
0.64%
1.48%
1.76%
3.86%
1.98%
0.88%
1.39%
7.20%
0.32%
NC
±10%
+10%
+10%
±10%
±10%
+10%
±10%
+10%
+10%
±10%
+10%
±10%
±10%
±10%
±10%
±10%
2.3%
6.6%
3.07%
4.6%
2.7%
2.1%
NC6
0.9%
5.8%
3.5%
2.3%
4.3%
4.5%
NC
5.3%
1.8%
1.4%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
90%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
lmo/038
7-4
-------
TABLE 7-1.. SUMMARY OF ESTIMATED AND ACHIEVED PRECISION,
ACCURACY, AND COMPLETENESS OBJECTIVES3 (continued)
Precision Accuracy Completeness
Parameter Estimated Achieved Estimated Achieved Estimated Achieved
Fixed Gases/
Molecular
Weight5 ±10% NC +20% NC 90% 100%
Flue Gas ,
K h
Temperature6'
HC11
±2°F
NE
NC +5°F
2.53% NE
0.36%
2.7%
90%
90%
100%
93%
The reference for the estimated precision, accuracy, completeness objectives
is previous experience with these methods as well as EPA Methods 1-5 and the
EPA/ASME protocol.
The values for precision represent the mean absolute differences for two
identical analyses of the same sample for the same isomers.
Q
The accuracy of the CDD/CDF analyses was evaluated by EPA prepared performance
audit samples. These results are not yet available. The accuracy objective
was measured value to within +50% of the true value for each isomer spiked.
Precision of the CEMs is expressed as the % coefficient of variation (CV)
determined from daily analyses of a QC standard, where
% CV = (Standard deviation/Mean) x 100
The accuracy of the CEMs is expressed as the absolute relative error as
determined from independent audit standards.
ND = Not determined for this parameter.
%C = No performance audit or QC analyses performed for this parameter.
tlelative accuracy expressed as the mean absolute relative error from ASTM
thermometer.
LPrecision (%CV) and accuracy (absolute relative error) based on analysis of
chloride QA audit sample.
-"Accuracy expressed as mean % absolute relative error from an EPA critical
orifice.
The accuracy of the outlet SO- monitor is based on the revised quench factor
equation discussed in Section 7.3.8.
lmo/038 7-5
-------
TABLE 7-2. SUMMARY OF ACCEPTANCE CRITERIA, CONTROL
LIMITS AND CORRECTIVE ACTION FOLLOWED FOR
MARION COUNTY
Criteria
Control Limit
Corrective
Action
Manual Sampling
Final Leakrate
(after each port)
Dry Gas Meter
Calibration
Individual Correction
Factors (J)
Average Correction
Factor
< 0.02 acfm or
4 percent of sampling rate
whichever is less
Post average
factor y agree ±5%
of prefactor
Agree within 2%
of average factor
1.00 ± 1%
Adjust sample
volume for port
Adjust sample
volumes using
the y that gives
smallest volume
Recalculate correction
factor
Adjust the dry
gas meter and
recalibrate
Intermediate
Dry Gas Meter
Calibrated every
six months against
EPA standard
Analytical Balance
(top loader)
0.1 mg of NBS
Class S Weights
Repair balance
and recalibrate
GEM Measurements
Linearity Multipoint
Calibration (four points)
Daily Drift
(zero and span)
R < 0.9950
a) ± 5%
b) 75% of data < 20%
c) > 20%
d) 2 days with drift
greater than 10 percent
Adjust instrument,
recalibrate
Data not adjusted
Adjust data assuming
linear drift over
testing period.
Reject data
Perform Instrument
maintenance
7-6
-------
TABLE 7-2. SUMMARY OF ACCEPTANCE CRITERIA, CONTROL
LIMITS, AND CORRECTIVE ACTION FOLLOWED FOR
MARION COUNTY (continued)
Criteria
Control Limit
Corrective
Action
GEM Measurements (continued)
Sampling System Bias
Daily QC Check
(mid-range)
Instrument Response
time
Interference Check
Line Leakcheck
Manifold Leakcheck
+ 5% of span
+ 10 percent of
certified concentration.
less than one-minute
+ 7% of manual result
> 0.5% 0,
> 0.5% 00
CDD/CDF Analytical Results
Internal Standard Recoveries 100 + 50%
Check heat tracing
and/or clean
sample line
Redo initial
calibration
Increase sample
flowrate or adjust
instrument
Repeat
interference
check to verify.
If verified, clean
sample lines and
check calibration
Locate and repair
leak, recheck
Locate and repair
leak, recheck
Re-extract and
re-analyze if
below 20% or
greater than
180%
Surrogate Recoveries
Verification of Identification
1) Ratio of M+ to M+2
or M+2 to M+4
100 + 50%
Within 20% of
theoretical value,
except for tetrachloro
which are taken within 13%
No action
Re-evaluate
peak
identification
7-7
-------
TABLE 7-2. SUMMARY OF ACCEPTANCE CRITERIA, CONTROL
LIMITS, AND CORRECTIVE ACTION FOLLOWED FOR
MARION COUNTY (continued)
Criteria
Control Limit
Corrective
Action
CDD/CDF Analytical Results (continued)
2) Retention Time
3) Signal-to-Noise Ratio
Duplicates
Within 3 seconds of the
corresponding or nearest
13C internal standard
or surrogate standard
(with reference to
continuing calibration)
greater than 2.5
Percent Difference < 50%
HC1 Analytical Results (specific ion electrode method)
Duplicate Percent Difference < 10%
Internal Audit Sample
Audit Blank
Interference Check by
Method of Additions
Linearity of
Calibration Curve
Relative Error ± 10%
of audit sample
+ 10% of true value
R > 0.995
Re-evaluate
peak
identification
Reconsider
peak
identification
Check data
manipulations
Reanalyze
Analyze by ion
chromatography
Analyze by ion
chromato graphy
Analyze by ion
chromato graphy
Re-do calibration
or use method
of additions
7-8
-------
7.3 QA/QC RESULTS
Sections 7.3.1 through 7.3.8 present the quality control (QC) procedures
specific to each sampling and/or analytical method. These sections contain
only a brief summary of results. The raw sampling and analytical QA/QC data
can be found in Appendix H.
7.3.1 Ash CDD/CDF Sampling and Analysis
Quality control for the ash sampling included procedures for contamin-
ation control as well as measurement integrity. Equal size increments were
collected at regularly scheduled intervals. Only sample containers and tools
that had been thoroughly and properly cleaned were used for sample collection.
Immediately after any compositing, all samples were properly transferred to
appropriate storage containers.
For the CDD/CDF ash analyses, the positive identification criteria
achieved for the characterization of polychlorinated dibenzodioxins and
dibenzofurans can be found in the Analytical report in Appendix F.4 and are
summarized below:
1. The integrated ion abundance ratio must be within 15 percent of
the theoretical value,
2. The retention time for an analyses must be within 3 standard
deviation intervals of the corr
standard or surrogate standard,
deviation intervals of the corresponding C-labeled internal
3. The monitored ions for an analyte must maximize within 3
standard deviations intervals,
4. The signal-to-noise ratio (S/N) for all monitored ions must be
greater than 2.5, and
lmo/038 7-9
-------
5. The measured response factors (RFs) for both labeled and
unlabeled compounds, obtained during a continuing calibration
run must be within 20 percent for tetra through
heptachlorinated compounds and within 25 percent for
octachlorinated compounds, of the mean values established
during the initial calibration.
7.3.1.1 Internal Standard and Surrogate Recoveries. CDD/CDF ash
samples were spiked with known amounts of internal standards and surrogates
prior to extraction. The internal standards were added during the soxhlet
extraction step. The internal standards recoveries were used by Triangle
Laboratories to adjust the results of the native species reported. The
surrogate recoveries were not used to adjust results but were used to
provide additional information on the extraction efficiency of the method.
The internal standard recoveries are summarized in Table 7-3. The QC
objective as required by the ASME/EPA protocol is ±50 percent recovery for
internal standards and surrogates. The internal standard recoveries for the
economizer and cyclone ash were all within the acceptable range. Recoveries
for superheater ash were all within the QC criterion except for the
13
recovery of CL^-CCDD which ranged from 30 to 48 percent. Good recoveries
of the other internal standards indicate that the lower recoveries reported
for CL--OCDD are not systematic analytical laboratory errors and are
probably caused by a sample matrix effect on the column cleanup and possibly
retention on carbonaceous ash.
Surrogate recoveries are summarized in Table 7-4. All ash surrogate
recoveries were well within the QC criterion of +50 percent, ranging from 76
to 135 percent.
7.3.1.2 Duplicate Analyses. Two of the Marion County ash samples were
analyzed in duplicate and these results are present in Table 7-5. The
purpose of the duplicates was to evaluate the reproducibility (precision) of
the combined sample preparation and analytical methodology. The QC criteria
for analysis of field duplicates is agreement to within ±50 percent.
lmo/038 7-10
-------
TABLE 7-3. INTERNAL STANDARDS RECOVERY RESULTS FOR
MARION COUNTY CDD/CDF ASH ANALYSES
Sample
2378
Superheater
Run 4
Run 6A
Run 6B
Run 11B
Recovery (%)
-13C12-TCDD
Ash
100
76
87
91
13C12-PCDD
89
76
82
88
13C12-HxCDD
79
74
77
83
13C12-HpCDD
68
51
60
63
13C12-OCDD
48
29
42
45
Economizer Ash
Run 3B
Run 4
Run 6A
Run 6B
Run 11B
Run 11B
(Duplicate)
Cyclone Ash
Run 3B
Run 4
Run 6A
Run 6B
Run 11B
96
97
87
97
90
102
88
84
88
79
80
92
95
78
96
83
96
91
86
94
82
84
78
79
65
79
73
79
85
85
91
78
81
72
75
59
74
63
67
67
71
73
63
60
61
62
47
65
48
49
55
52
60
52
46
lmo/038
7-11
-------
TABLE 7-3. INTERNAL STANDARDS RECOVERY RESULTS FOR
MARION COUNTY CDD/CDF ASH ANALYSES
(Continued)
Sample
Recovery (%)
2378-13C12-TCDD
13
C12-PCDD
13
C12-HxCDD
13
C12-HpCDD
Baghouse Ash
Run 3B 93
Run 3B (Dup.) 97
Run 4
Run 6A
Run 6B
Run 10
Run 11A
Run 11B
92
99
78
95
96
102
87
97
94
103
73
90
92
91
92
86
80
88
71
78
84
83
92
89
83
83
60
70
71
80
79
76
70
64
59
58
53
72
lmo/038
7-12
-------
TABLE 7-4. SURROGATE RECOVERIES FOR MARION COUNTY
ASH CDD/CDF ANALYSES
Sample
Superheater Ash
Run 4
Run 6A
Run 6B
Run 11B
Economizer Ash
Run 3B
Run 4
Run 6A
Run 6B
Run 11B
Run 11B (Dup
Cyclone Ash
Run 3B
Run 4
Run 6A
Run 6B
Run 11B
Baghouse Ash
Run 3B
Run 3B (Dup.
Run 4
Run 6A
Run 6B
Run 10
Run 11A
Run 11B
13C12-TCDF
99
92
98
92
99
95
102
97
90
-) 97
91
82
92
89
95
135
) 122
122
124
88
123
95
95
Recovery (%)
37C1-TCDD
100
98
95
95
102
103
99
103
99
101
76
97
96
95
94
105
96
94
98
82
94
96
97
13C12-HxCDF
99
94
94
96
96
97
99
95
97
103
95
91
93
93
93
113
111
107
103
85
100
104
99
lmo/038
7-13
-------
TABLE 7-5. DUPLICATE RESULTS FOR MARION COUNTY CDD/CDF ASH ANALYSES
a b
Duplicates '
Analyte Baghouse Ash
Dioxins
Mono - CDD
Di-CDD
Tri-CDD '
2378-TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
Octa-CDD
Result
#1
(0.005)
(0.010)
0.088
0.026
0.148
0.036
0.177
[0.030]
0.041
0.076
0.174
0.212
0.187
0.247
(3B)
Result
#2
(0.003)
[0.059]
0.196
0.030
0.389
0.038
0.276
0.022
0.048
0.070
0.228
0.242
0.205
0.328
Economizer Ash (11B)
Average
c
0.142
0.028
0.269
0.037
0.227
0.045
0.073
0.201
0.227
0.196
0.288
Absolute
Percent
Difference
--
76.1
14.3
89.6
5.40
43.6
--
15.6
8.22
26.9
13.2
9.18
28.1
Result
#1
(0.003)
(0.003)
0.021
0.013
0.083
0.020
0.080
0.010
[0.009]
0.031
0.050
0.069
0.070
0.188
Result
#2
(0.003)
(0.003)
0.038
0.013
0.114
0.025
0.121
0.011
[0.009]
[0.035]
0.053
0.093
0.090
0.227
Average
--
0.030
0.013
0.099
0.023
0.101
0.011
--
0.052
0.081
0.080
0.208
Absolute
Percent
Difference
56.7
0.00
31.3
21.7
40.6
9.10
5.77
29.6
25.0
18.8
Average
31.9
23.5
All results reported in ppb. ND - Not detected at the method detection limit shown in
parentheses. The estimated maximum possible concentration is given in brackets.
% Difference calculated as: % Diff - [ (X-j^ - X2)/X] x 100, where
X.. - result #1, X- - result #2 , X - (X- + X9)/2
JL £* «L £t
Dash indicates not applicable.
-------
TABLE 7-5. DUPLICATE RESULTS FOR MARION COUNTY CDD/CDF ASH ANALYSES (Continued)
a b
Duplicates '
Analyte
Furans
Mono -CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa-CDF
Baghouse
Result
#1
(0.005)
(0.015)
2.949
2.400
3.385
0.167
0.152
1.149
0.240
0.088
[0.059]
(0.010)
0.392
0.351
(0.008)
0.039
[0.134]
Result
#2
[0.
0.
3.
2.
4.
0.
0.
1.
0.
0.
0.
(0.
0.
0.
0.
0.
[0.
026]
074
652
000
108
204
150
666
187
109
075
003)
403
195
009
021
035]
Ash (3B)
Average
c
3.301
2.200
3.746
0.186
0.151
1.408
0.214
0.099
--
0.398
0.273
--
0.030
"• ~
Economizer Ash (11B)
Absolute
Percent Result
Difference #1
--
21.3
18.2
19.3
19.9
1.32
36.7
24.8
21.2
--
--
2.76
57.1
60.0
~ ~
(0.001)
[0.351]
1.596
1.900
1.777
0.134
0.140
0.794
0.158
0.063
0.093
[0.007]
0.325
0.317
0.030
0.158
0.195
Result
#2
(0.003)
[0.475]
2.224
2.100
2.036
0.153
0.166
0.918
0.198
0.086
0.105
(0.003)
0.425
0.395
0.043
0.220
0.254
Average
1
2
1
0
0
0
0
0
0
0
0
0
0
0
--
.910
.000
.907
.144
.153
.856
.178
.075
.099
.375
.356
.037
.189
.225
Absolute
Percent
Difference
--
32.
10.
13.
13.
16.
14.
22.
30.
12.
26.
21.
35.
32.
26.
9
0
6
2
9
5
5
7
1
7
9
1
8
2
Average 23.6 21.0
SA11 results reported in ppb. ND - Not detected at the method detection limit shown in
parentheses. The estimated maximum possible concentration is given in brackets.
b% Difference calculated as: % Diff - [(^ - X2)/X] x 100, where
XT - result #1, X = result #2, X
'Dash indicates not applicable.
X2)/2
-------
The reproducibility or precision of the sample preparation and analytical
methodology was well within the QC criterion for nearly every isomer. The
average differences for the duplicate analyses were 28 percent and 22 percent
for CDDs and CDFs, respectively.
7.3.1.3 Sample Blanks. Analytical method blanks were analyzed as part
of the QC program. These results are presented in Table 7-6. Insignificant
quantities of the target analytes were found in two of the method blanks.
Only OCDD, at a concentration of 0.006 ppb, was found in method blank #2.
Trace amounts of 2378-TCDF, 123478-HxCDF, and 1234678-HpCDF were found in
method blank #3. These concentrations were very close to the method detection
limit or in the noise range (10 times the detection limit), and therefore, are
not considered significant.
7.3.2 HC1 Flue Gas Sampling and Analysis Quality Control
HC1 sampling was based on EPA Reference Method 5 with modifications which
allowed the collection of HCl in the back half of the sampling train.
Sampling quality control followed standard Method 5 procedures.
The specific ion electrode (SIE) detection method was used for on-site measure-
ment of free chloride ions in aqueous solutions.
The on-site chloride analysis was audited externally daily with QA
samples prepared by EPA/Cincinnati QC samples. Two concentration levels were
prepared; 103 ug/ml and a 25.8 ug/ml samples. All of the audit results for
103 ug/ml QA samples were acceptable. The results for the 25.8 ug/ml audit
samples were variable depending on the amount of dilution used to prepare the
audit sample for analysis. With a five-fold dilution, the audit sample
results were acceptable (true value within ±10 percent) for two out of three
analysis.
lmo/038 7-16
-------
TABLE 7-6. ANALYTICAL METHOD BLANK RESULTS FOR
MARION COUNTY CDD/CDF ASH ANALYSES3'
Analyte
Total MCDD
Total DCDD
Total TriCDD
2378-TCDD
Total TCDD
12378-PCDD
Total PCDD
123478 -HxCDD
123678 -HxCDD
123789-HxCDD
Total HxCDD
1234678-HpCDD
Total HpCDD
OCDD
Total MCDF
Total DCDF
Total TriCDF
2378-TCDF
Total TCDF
12378-PCDF
23478-PCDF
Total-PCDF
123478 -HxCDF
123678-HxCDF
234678 -HxCDF
123789-HxCDF
Total HxCDF
1234678 -HpCDF
1234789 -HpCDF
Total HpCDF
OCDF
Method Blank 1
ND (0.003)
ND [0.007]
ND (0.005)
ND (0.003)
ND [0.004]
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND [0.011]
ND (0.003)
ND (0.003)
ND (0.005)
ND (0.003)
ND (0.005)
ND (0.005)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.005)
Method Blank 2
ND (0.001)
ND [0.006]
ND (0.005)
ND (0.003)
ND [0.003]
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND [0.007]
ND (0.003)
ND (0.003)
0.006
ND (0.001)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.001)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
Method Blank 3
ND (0.003)
ND (0.003)
ND (0.008)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.003)
ND (0.005)
ND (0.003)
ND (0.005)
ND (0.005)
0.005
0.005
ND (0.003)
ND (0.003)
ND (0.003)
0.009
ND (0.003)
ND (0.003)
ND (0.003)
0.010
0.025
ND (0.003)
0.028
ND (0.005)
aND= not detected at the method detection limit shown in parentheses.
Estimated maximum possible concentration reported in brackets.
Method detection limit and/or measured concentrations reported in ppb.
lmo/038
7-17
-------
Additional quality control included daily calibrations, analysis of
blanks, and using the method of known additions to determine whether
interferences were present in the sample matrix. The method of additions
results are presented in Table 7-7. The QC criterion was agreement between
the known addition measurement and direct reading measurement within +10
percent. All 25 ml sample aliquots were saved for analysis by ion
chromatograpy. These results were used to support the SIE data and can
be found in Appendix H.2.2.
As seen from Table 7-7, several of the samples had relative percent
differences outside the QC criterion of +10 percent difference. In
particular, the samples analyzed on 6-4-87 exceeded the +10 percent. Also,
the samples analyzed on 6-8-87 were, for the most part, outside the +10
percent criteria. These samples are considered to be incorrect due to some
type of erratic, electrode interference possibly due to a faulty electrode
or temperature fluctuations in the room. After instrument adjustments were
made, the 6-4-87 field samples were reanalyzed.
Reagent blanks were also analyzed as part of the SIE quality control
procedures. One HCl HPLC lUO blank of water used in the Phase II Modified
Method 5 trains was analyzed and showed less than 1 ppm chloride, or less
than the method detection limit. One 0.1N NaOH blank of the sodium
hydroxide used in the HCL train impingers was analyzed and was shown to
contain 150 ppm of chloride. The high ionic strength due to the hydroxide
solution may have increased the solution conductivity and resulted in high
chloride concentration measurement.
An additional QC step was performed as part of the HCl analyses. Two
cleaned MM5 sample bottles and two cleaned HCl sample bottles were analyzed
for background contamination. 100 ml of DI water was added to each sample
bottle and then 1 ml of Ionic Strength Adjuster (ISA), sodium nitrate, was
added. These solutions were then analyzed by SIE and all showed less
than 1 ppm ( 0.43 ppm) of chloride.
lmo/038 7-18
-------
TABLE 7-7. RELATIVE PERCENT DIFFERENCES BETWEEN SIE DIRECT READING AND
KNOWN ADDITION RESULTS FOR CHLORIDE CONCENTRATIONS
Sample ID
Concentration of
HC1 Obtained from
Known Addition
Measurement
(ppm)
Concentration of
HC1 Obtained from
Direct Reading
Measurement
(ppm)
Relative %
Difference
Based On
Direct
Reading
Measurement
b,d
MAR0604-HC11N1-I1A
MAR0604-HC11N1-I2A
MAR-0604-HC1-MID1-I1A
MAR-0604-HC1-MID1-I2A
MAR-0604-HC1-OUT1-I1A
MAR-0604-HC1-OUT1-I2A
MAR- 0605 -HC1-IN2-I1A
MAR- 0605 -HC1-IN2-I2A
MAR-0605-HC1-MID-2-I1A
MAR-0605-HC1-MID-2-I2A
MAR-0605-HC1-OUT-2-I1A
MAR-0605-HC1-OUT-2-I2A
MAR-0608-HC1-IN-4-I1A
MAR-0608-HC1-IN-4-I2A
MAR-0608-HC1-MID-4-I1A
MAR-0608-HC1-MID-4-12A
MAR-0608-HC1-OUT-4-I1A
MAR-0608-HC1-OUT-4-I2A
QA5
3000
800
1000
Offscale
250
15.0
4400
720
1200
52.0
220.0
10.0
3300
525
1040
Offscale
54.0
2.0
105.0
3900°
720°
1300°
230C
260C
24.0°
4200
700
1200
56.0
220.0
10.5
2800
430
860
16.0
63.0
1.90
100.0
-23.1
11.1
-23.1
. .
-3.8
-37.5
4.8
2.9
0.0
-7.1
0.0
-4.8
17. 9e
22. 16
20. 9e
-14. 3e
5.3
5.0
A difference between the two measurements (known addition and direct
reading) greater than 10 percent was the QC criteria that was used to
indicate the possibility of a complexing agent in the sample.
bRelative percent difference calculated as [(A - B)/B] x 100, where A is the
concentration of chloride obtained from the known addition measurement and
B is the concentration of chloride obtained from the direct reading
measurement.
CThese values were determined to be incorrect. Instrument adjustments were
subsequently performed to eliminate any electrode interferences. All field
samples that were analyzed during the same time period were reanalyzed
following all instrument adjustments. The method of additions samples were
not reanalyzed.
The specific meter used during the analysis is designed to automatically
determine method of additions concentrations. The instrument compensates
for the change in concentration because of the standard addition and gives a
reading for a new concentration which is compared directly to the
concentration in the original sample (direct reading measurement).
eThese values are considered to be outside the QC criteria, but samples were
not reanalyzed during this time period.
lmo/038
7-19
-------
7.3.3 Continuous Emission Monitor (GEM) Quality Control
CEMs were used to analyze flue gas from three locations: the control
device inlet (boiler outlet), the midpoint location (spray dryer outlet) and
the breeching to the stack. The flue gas was analyzed for C0?, CO, 0-, SO.,
NO , and THC at the inlet and outlet, and C00, 0., and S00 were monitored at
X £. 4- £.
the midpoint location.
The limited availability of GEM instruments required that instruments
based on different principles of operation be used to measure SO,, at the three
sampling locations. Two Thermo Electron Corporation (TECO) Model 40 SO,-
analyzers were used to monitor the midpoint and fabric filter outlet
locations. The midpoint TECO 40 SO- analyzer was used for the inlet location
for Runs 10, 11A, and 11B. The TECO works on the principle of pulsed
fluorescence. A pulsed source of ultraviolet radiation electronically excites
the S0« molecules in the sample cell. The excited molecules then decay back
to their ground state by fluorescence, emitting a photon. However, CO. and 0_
molecules also present in the sample will absorb the emitted photons causing
the S0« concentration to be lower than the true value. The results can be
adjusted using a quench factor which is discussed in Section 7.3.8.
A Western Model 721A SO- analyzer was used at the inlet location for
Runs 1-9. For Runs 10, 11A, and 11B the Western SO- analyzer was used for the
midpoint location. The Western instrument is essentially a continuous
spectrophotometer in the ultraviolet range. SO- selectively absorbs ultra-
violet (UV) light at a wavelength of 202.5 nm and measures the absorbance (A)
of the radiation through the sample cell by the decrease in intensity. This
type of analyzer is not affected by CO- and 0- concentrations.
The instruments used for CO were both Beckman Model 865, non-dispersive
infrared analyzers. Non-dispersive infrared analyzers emit a specific
wavelength of infrared radiation through the sample cell which is
selectively absorbed by CO- molecules. The CO instrument was offset at the
outlet location to compensate for interferences caused by the presence of CO-
in the flue gas since the CO levels in the flue gas were low (20-40 ppm).
Since this offset was significant for the outlet analyzer, the CO data for
this location were discarded.
lmo/038 7-20
-------
7.3.3.1 Daily Calibrations and Drift Checks. All CEM analyzers were
calibrated daily with a zero gas (generally N9), and a high-range span gas.
Calibrations were performed in the morning prior to and at the completion of
testing each day. Daily calibrations and drifts are summarized in Tables 7-8
through 7-11. Daily drift requirements for both zero and span were ±5 percent
for each run. For the 12 days of sampling with 15 analyzers (approximately
180 data points), the instrument drift was routinely within the 5 percent QC
criteria. The only instrument showing consistently high drifts was the outlet
NO analyzer. This drift was probably caused by some type of instrument
X
malfunction. However, since the data are drift corrected, the CEM data
quality is not affected.
7.3.3.2 System Bias Checks. During the course of the testing
program, bias checks of the CEM sampling systems were performed for the S02
and C09 analyzers. The checks were used to assess the potential measurement
bias caused by the sampling lines and gas conditioning system. This check
assesses the bias imparted to the sample by the sample lines and gas
conditioning system. The high bias observed at the midpoint location was
caused by the high lime/moisture content in the gas. Bias check results are
presented in Table 7-12. The QC criteria was sampling system bias ±5 percent
of span. The bias for C0_ and SO- analyzers changed from 0.5 to -5.2 and was
not considered enough to justify adjusting the data.
7.3.3.3 Response Times. Response times for the analyzers were
determined as part of the CEM QC procedures. These results are presented in
Table 7-13. The 95 percent response times for the fifteen analyzers ranged
from 0.31 to 2.0 minutes. Since all results are reduced to 1-hour averages
from the 1-minute data, the variation in response times is insignificant.
Also, the response times are small compared to the one or three hour averaging
intervals. The response times for obtaining 95 percent of the midrange QC gas
concentration from zero concentration were approximately 0.80 minutes. Since
the CEM/computer interface reads 1-minute averages during 3-hour tests, this
lag will have an insignificant effect on the CEM data quality.
lmo/038 7-21
-------
TABLE 7-8. SUMMARY OF CEM DRIFT CHECKS FOR MARION COUNTY, INLET3>b>C
Date
6-04-87
6-05-87
6-06-87
6-08-87
6-09-87
6-10-87
6-11-87
6-12-87
6-15-87
6-16-87
6-21-87
6-22-87
Test
Condition
1
2
3A,3B
4
5
6A.6B
7
8,9
10
11A,11B
12
13
°2
Drift
(%V)
-0.129
-0.575
-0.124
0.562
0.680
0.534
-0.255
0.158
1.775
1.347
-0.026
-0.810
co2
Drift
(%V)
9.281d
2.931
0.883
0.051
2.205
9.255d
16.945d
10.931d
4.359
-2.496
8.982d
-0.152
CO
Drift
(ppm)
0.098
0.108
1.133
1.388
2.253
1.764
0.853
1.074
1.341
0.848
-0.187
-0.060
so2
Drift
(ppm)
1.916
-2.326
0.589
-0.045
2.237
-0.542
2.007
3.529
-2.713
3.274
-0.829
2.473
NO
X
Drift
(ppm)
-1.130
-2.847
-1.947
0.551
1.814
-1.676
-2.747
-4.831
-0.696
2.050
0.700
-0.669
THC
Drift
(ppm)
11.891d
6.021d
-0.288
16.158d
31.854e
3.188
2.077
1.977
1.643
2.963
2.641
10.873d
Drifts expressed as [(Final Response Factor - Initial Response Factor)/
Initial Factor] x 100.
QC criterion is percent drift within +5 percent.
°The instrument spans were as follows: 0-25%V for 0-, 0-20%V for C02>
0-100 ppm for CO, 0-500 ppm for SO- runs 1-9, 0-1000 ppm for S02 runs
10-13, 0-1000 ppm for NO , and 0-100 ppm for THC runs 1-5, 12, 13 and
0-10 ppm for runs 6-11.
Exceeds 5% QC criteria.
eExceeds 20% rejection criteria. Drift correction applied and data
retained conditionally.
lmo/038 7-22
-------
TABLE 7-9. SUMMARY OF CEM DRIFT CHECKS FOR MARION COUNTY, MIDPOINT a'b'c
Date
6-04-87
6-05-87
6-06-87
6-08-87
6-09-87
6-10-87
6-11-87
6-12-87
6-15-87
6-16-87
6-21-87
6-22-87
Test
Condition
1
2
3A.3B
4
5
6A.6B
7
8,9
10
11A.11B
12
13
°2
Drift
(%V)
0.239
0.266
-0.287
0.298
1.297
2.927
1.495
1.408
1.252
1.413
0.074
0.353
co2
Drift
(%V)
6.880
4.203d
15.297d
0.409
-2.236
4.062
8.421d
13.740d
-0.460
-5.252d
3.506
2.600
so2
Drift
(ppm)
8.118d
22.772e
8.850d
12.403d
16.115d
3.075
2.924
8.608d
2.144
0.255
0.853
NR
aDrifts expressed as [(Final Response Factor - Initial Response Factor)/
Initial Factor] x 100.
QC criterion is percent drift within +5.
Instrument spans were 0-25%V for 02, 0-25% V for C02, and 0-500 ppm for
so2.
Exceeds 5% QC criteria.
eExceeds 20% rejection criteria. Drift correction applied and data retained
conditionally.
NR - not reported for this parameter on this day.
lmo/038
7-23
-------
TABLE 7-10. SUMMARY OF GEM DRIFT CHECKS FOR MARION COUNTY, OUTLETa>b>C
Date
6-04-87
6-05-87
6-06-87
6-08-87
6-09-87
6-10-87
6-11-87
6-12-87
6-15-87
6-16-87
6-21-87
6-22-87
Test °2
Condition Drift
(%V)
1
2
3A.3B
4
5
6A,6B
7
8,9
10
11A.11B
12
13
0.972
0.067
0.147
-1.128
-1.447
-1.011
-1.037
-0.909
1.120
-1.164
-1.352
-3.147
co2
Drift
(%V)
-0.566
0.372
9.473d
-9.837d
0.108
1.491
0.062
0.596
-0.411
0.392
0.551
5.020d
so2
Drift
(ppm)
-0.570
-0.185
1.161
-2.466
4.320
-1.339
-0.641
4.186
-1.628
0.915
0.949
-1.829
CO
Drift
(ppm)
0.910
-0.145
0.786
-0.112
0.231
1.075
1.355
1.008
0.290
-0.036
0.995
0.729
THC
Drift
(ppm)
0.797
-0.750
0.298
0.015
0.786
NR
NR
NR
NR
NR
0.329
1.379
NO
X
Drift
(ppm)
12.653d
8.305d
5.877
2.466
6.609d
9.874d
15.807d
24.983e
9.658d
18.786d
9.190d
6.614d
&Drifts expressed as [(Final Response Factor - Initial Response Factor)/
Initial Factor] x 100.
QC criterion is percent drift within ±5.
CThe instrument spans were 0-25% V for 0^, 0-20% V for C02> 0-500 ppm for
S02, 0-500 ppm for CO, and 0-1000 ppm for NO^ and 0_1Q ppm f
-------
TABLE 7-11. SUMMARY OF GEM HIGH RANGE SO DRIFT
CHECKS FOR MARION COUNTY*'
Date
6-04-87
6-05-87
6-06-87
6-08-87
6-09-87
6-10-87
6-11-87
6-12-87
6-15-87
6-16-87
Test
Condition
1
2
3A.3B
4
5
6A.6B
7
8,9
10
11A.11B
Location
Inlet
Inlet
Inlet
Inlet
Inlet
Inlet
Inlet
Inlet
Inlet
Midpoint
Percent
Drift
4.498
1.362
3.716
-0.136
0.317
1.910
1.504
3.352
4.667
1.409
Instrument
Range
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
0-5000 ppm
aDrifts expressed as [(Final Response Factor - Initial Response Factor)/
Initial Factor] x 100.
QC criterion is percent drift within +5.
lmo/038
7-25
-------
TABLE 7-12. GEM SYSTEM BIAS TEST FOR MARION COUNTY S02 AND C02 ANALYZERS
Date
Concentrations
Sampling Analyte Range Certified Gas To Gas To System Percent
System (Units) Manifold System Bias of Span
a,b
6/02/87 Inlet S02 0-500 219
(ppmV)
SO- 0-500
(ppmV)
CO
0-20
228 225.6 -2.4 -0.48
13.1 12.5 13.3 0.8 0.16
13.1 12.4 12.5 0.1 0.50
Outlet S02 0-1000 219
(ppmV)
225.9 213.2 -12.8 -1.28
S02 0-1000 13.1 15.6 16.8
(ppmV)
CO
0-20
13.1 13.5 13.5
1.2
0.0
6/04/87 Midpoint S02 0-1000 219
(ppmV)
aPercent of span calculated as: Percent of span - (system bias/span) x 100
b.
QC criteria is sampling system bias within ±5% of span.
0.12
0.00
222.7 171.1 -51.6 -5.16
6/09/87
6/17/87
co_
(%V)
Inlet SO.
(ppmV)
Midpoint S02
(ppmV)
Outlet S02
(ppmV)
Inlet S02
(ppmV)
Outlet S0«
(ppaV)
0-20
0-5000
0-1000
0-1000
0-1000
0-1000
13.1
839.4
82.1
82.1
219
219
13.1
815.9
88.7
88.7
219.1
230.1
12.7
825.2
72.9
82.9
192.9
217.6
-0.4
9.3
-15.8
-5.8
-26.2
-12.5
-2.00
0.19
-1.58
-0.58
-2.62
-1.25
7-26
-------
TABLE 7-13. RESPONSE TIMES (95%) FOR MARION COUNTY MIDRANGE CEM QC GASES'
Location
Instrument Inlet Midpoint Outlet
Response Gas Response
Time Concentration Time
(Minutes) (Minutes)
02 1.00 10.02% 0.88b
CO 1.15 202.0 ppmV NA
C02 0.85 7.99% 0.67
S02 0.50 214.9 ppmV 1.10
NO 0.67 380.3 ppmV NA •
X
THC 0.46 4.30 ppmV NA
Gas Response Gas
Concentration Time Concentration
(Minutes)
20.0% 2.00 20.0%
NA 0.45 202.0 ppmV
7.99 ppmV 0.31 7.99%
214.9 ppmV 0.75 214.9 ppmV
NA 0.50 380.3 ppmV
NA 0.73 42.7 ppmV
QC criteria is response time of less than one minute.
These response times were determined using high-range calibration gases.
NA - Not applicable. These locations were not sampled for these parameters.
7-27
-------
7.3.3.4 Daily QC Checks. After the morning calibrations, midrange
gases for all instruments were analyzed, with no adjustment, as a quality
control check of daily calibrations and to provide day-to-day precision
estimates for each instrument. The calibration was considered acceptable if
the quality control concentration was within +10 percent of the certified
concentration. If this QC check was unacceptable, another calibration was
performed and linearization was performed if deemed necessary. The daily
GEM QC checks are presented in Table 7-14. These results indicate that the
day-to-day precision of the instruments was well within the QC criteria of
+10 percent coefficient of variation (CV). The percentage CV was less than 6
percent for all analyzers. The calibration of the CEMs was also shown to be
consistent with mean percent difference within 10 percent.
7.3.3.5 Multipoint Linearity Checks. All GEM instruments were
calibrated on a multipoint basis each week on-site at the Marion County
facility. Multipoint calibrations were performed with four certified gases:
zero gas, a low scale gas concentration, a midrange concentration, and a
high scale concentration (span gas). The QC criterion for acceptable
2
linearity was a correlation coefficient (R ) of greater than or equal to
0.9950, where the independent variable was the cylinder gas concentration and
the dependent variable was the instrument response. All GEM linearity checks
2
were within the QC criteria of R greater than 0.9950, indicating that
linearity for all of the instruments was excellent.
7.3.3.6 Relative Accuracy. Interference checks were performed for
CO., 0-, and SO.. CO. and 0^ were checked using Manual EPA Method 3. CO,
THC, and NO were not checked. These results are presented in Tables 7-15
through 7-19. For Oy and CO,,, the QC criteria was absolute difference
between Orsat and GEM value within 1 percent. For S0?, the same criteria
applies for the difference between the Method 6 and CEM results. For the
inlet location, only the absolute difference between the Orsat and CEM
value for Run 11B slightly exceeded 1 percent; it was 1.3 percent. At the
midpoint, only 3 values exceeded the QC criteria; 1.8 (Run 6A, 02>, 1.9 (Run
6A, CO-) and 1.5 (Run 11A, CO ). At the outlet, 3 values exceeded the QC
criteria; 1.2 (Run 6A, 02), 1.8 (Run 5, C02), and 1.1 (Run 11A, C02). These
exceedences are most likely due to a leak in the CEM sampling system.
lmo/038 7-28
-------
TABLE 7-14. DAILY QUALITY CONTROL CHECKS FOR THE MARION COUNTY CEMs'
Number
of
Points
Inlet
1
10
11
1
10
9
2
10
5
6
6
3
2
Midpoint
1
10
1
10
7
4
Outlet
1
10
11
9
2
5
1
1
10
Parameter
0_ %V
Oj %V
CO ppmV
CO, %V
CO^ %V
SO, ppmV
SO, ppmV
NO ppmV
THC ppmC
THC ppmC .
SO ppmV
SO^ ppmV
S02 ppmV
0, %V
0, %V
CO, %V
col1 %v
SO, ppmV
SO, ppmV
CO %V
CO, %V
SO, ppmV
NO ppmV
NOX ppmV
TH& ppmC
CO, %V
CO ppmV
CO ppmV
^ean Percent difference
Certified
Concentration
5.02
10.02
39.8
4.00
7.99
214.9
442.6
380.3
42.7
4.3
1911.3
839.4
412.3
5.02
10.2
4.00
7.99
214.9
214.2
4.00
7.99
214.9
380.3
380.4
4.30
4.00
81.0
202.0
determined from
Mean
Measured
Concentration
4.8
9.78
39.0
4.00
8.26
224.3
442 . 6
378.4
42.6
4.5
1981.5
772.8
395.6
5.2
10.1
4.10
8.14
209.3
217.7
3.6
8.2
221.7
375.3
372.6
4.5
3.6
81.4
200.9
the data included
Mean
Percent
Difference0
-(4.38)b
-2.40
-2.01
(0.00)
3.38
4.37
0.00
-.50
-.23
4.65
3.67
-7.93
-4.05
(3.59)
0.99
(2.50)
1.88
-2.61
1.63
-(10.00)
2.63
3.16
-1.31
-2.05
4.65
-(10.00)
0.49
-.55
in Appendix
Percent
Coefficient
of
Variation
d
0.62
1.10
--
2.56
1.77
--
1.73
1.56
3.32
1.46
0.33
0.14
--
1.48
--
1.76
2.19
5.54
--
1.98
0.88
1.39
0.00
7.20
--
--
0.32
H and
calculated as:
[Measured Concentration - Certified Concentrationl x 100
Certified Concentration
Percent difference in parentheses is based on a single measurement.
CQC criteria was percent coefficient of variation and mean percent difference
within 10 percent.
Dash indicates %CV not applicable.
7-29
-------
TABLE 7-15. COMPARISON OF MEASURED METHOD 3 AND GEM 0
AND C02 RESULTS FOR MARION COUNTY. INLET
a,b
Test
Condition
1
2
3A
3B
4
5
6A
6B
7
8
9
10
11A
11B
°2
Method 3
9.0
9.4
6.2
10.9
8.9
8.9
9.1
12.2
7.8
9.7
10.0
8.8
9.0
9.4
CEM
9.0
8.8
5.7
10.6
8.8
8.7
8.8
12.5
7.8
9.8
10.1
9.4
8.9
9.2
Absolute
Difference
0.0
0.6
0.5
0.3
0.1
0.2
0.3
-0.3
0.0
-0.1
-0.1
-0.6
0.1
0.2
co2
Method 3
10.0
10.1
12.6
8.9
9.7
10.0
10.7
7.5
11.6
9.5
9.4
10.2
10.2
9.7
CEM
10.4
11.0
13.1
8.9
10.3
10.5
10.3
7.3
11.6
9.8
9.5
10.3
10.6
11.0
Absolute
Difference
-0.4
-0.9
-0.5
0.0
-0.6
-0.5
0.4
0.2
0.0
-0.3
-0.1
-0.1
-0.4
-1.3
All values expressed in percent, calculated as [Method 3 value - CEM value].
QC criteria is absolute difference between Orsat and CEM value within 1 percent.
7-30
-------
TABLE 7-16. COMPARISON OF MEASURED METHOD 3 AND CEM 0
AND C02 RESULTS FOR MARION COUNTY, MIDPOINT a>
Test
Condition
1
2
3A
3B
4
5
6A
6B
7
8
9
10
11A
11B
°2
Method 3
10.8
10.8
9.1
11.9
10.6
10.8
8.9
13.8
10.2
11.7
12.0
10.6
12.6
11.2
CEM
10.9
10.7
8.5
11.9
9.9
10.2
10.7
13.5
9.9
11.4
11.4
10.6
10.9
11.0
Absolute
Difference
-0.1
0.1
0.6
0.0
0.7
0.6
-1.8
0.3
0.3
0.3
0.6
0.0
1.7
0.2
co2
Method 3
8.7
8.5
10.3
7.6
8.8
8.8
10.7
6.2
9.3
7.9
7.5
8.7
7.3
8.4
CEM
8.6
9.0
11.1
8.5
8.8
8.9
8.8
6.2
9.4
8.2
7.9
8.8
8.8
8.6
Absolute
Difference
0.1
-0.5
-0.8
-0.9
0.0
-0.1
1.9
0.0
-0.1
-0.3
-0.4
-0.1
-1.5
-0.2
aAll values expressed in percent and calculated as [Method 3 value - CEM value].
3QC criteria is absolute difference between Orsat and CEM value within 1 percent.
7-31
-------
TABLE 7-17. COMPARISON OF MEASURED METHOD 3 AND CEM 0
AND C02 RESULTS FOR MARION COUNTY, OUTLET
2a,b
Test
Condition
I
2
3A
3B
4
5
6A
6B
7
8
9
10
11A
11B
°2
Method 3
11.9
12.0
9.6
13.4
11.5
12.8
10.6
14.1
11.1
12.4
12.4
11.8
12.8
12.2
CEM
11.7
11.9
10.1
13.2
12.2
12.0
11.8
15.0
11.3
12.8
13.1
11.7
11.9
12.0
Absolute
Difference
0.2
0.1
-0.5
0.2
-0.7
0.8
-1.2
-0.9
-0.2
-0.4
-0.7
0.1
0.9
0.2
co2
Method 3
7.8
7.8
9.8
6.4
7.9
6.4
8.5
5.7
8.7
7.4
7.2
7.6
6.8
7.4
CEM
8.0
8.1
10.2
7.4
7.5
8.2
8.4
5.6
8.8
7.6
7.2
8.1
7.9
7.8
Absolute
Difference
-0.2
-0.3
-0.4
-1.0
0.4
-1.8
0.1
0.1
-0.1
-0.2
0.0
-0.5
-1.1
-0.4
aAll values expressed in percent and calculated as [Method 3 value - CEM value].
QC criteria is absolute difference between Orsat and CEM value within 1 percent.
7-32
-------
TABLE 7-18. COMPARISON OF EPA METHOD 6 AND GEM SO- RESULTS FOR MARION COUNTY
a,b
U)
Test
Condition
1
2
3
Relative
Accuracy
Method
6
(ppmV)
519.5
274.8
376.3
Inlet
GEM
(ppraV)
433.0
225.9
346.5
Midpoint
Relative
Difference
(%) C
-16.7
-17.1
-7.91
-14.1%
Method
6
(ppmV)
325.8
138.1
250.4
GEM
(ppmV)
351.6
157.4
223.2
Relative
Difference
(%)C
7.91
13.9
-10.9
+ 2.5%
Method
6
(ppmV)
115.7
29.5
96.5
Outlet
GEM
(ppmV)
121.8
32.9
107.3
Relative
Difference
(%)C
5.27
11.5
11.2
+8.3%
Two minutes of the CEM sampling time were not included in the average concentration reported
because the data acquisition system exceeded the full range (voltage) during this 2-minute period.
The relative percent difference was calculated as:
[(GEM value - Method 6 value)/Method 6 value] x 100.
Q
A reasonable QC criteria was relative difference within +20 percent.
Relative Accuracy calculated according to Appendix F, Relative Accuracy Audit Procedure, 40CFR
Part 60.
-------
TABLE 7-19. COMPARISON OF HCl MANUAL RESULT (SIE) AND GEM RESULT FOR MARION COUNTY
a,b
Inlet
Test
Condition
1
2
3A
3B
4
5
6A
65
7
8
9
10
11A
11B
Average
GEM
(ppm)
560
579
541
522
556
638
595
380
631
464
508
699
634
688
SIE
(ppm)
400
460
420
...
361
524
559
289
626
433
427
...
635
703
Relative
Percent
Difference
40.0
25.7
28.7
...
54.2
21.8
6.40
31.4
0.82
7.26
19.1
...
-0.15
-2.23
19.8
CEM
(ppm)
161
138
163
114
118
81.7
165
47.8
176
127
134
132
217
224
Midpoint
SIE
(ppm)
127
168
212
148
139
241
355
152
324
182
185
168
299
391
Relative
Percent
Difference
27.1
-18.1
-23.0
-23.0
-14.6
-66.1
-53.5
-68.6
-45.7
-30.0
-27.4
-21.5
-27.5
-42.5
34.9
CEM
(ppm)
55.8
23.6
42.4
29.4
7.2
31.1
48.7
12.8
49.3
25.0
11.8
13.8
104
139
Outlet
SIE
(ppm)
c
25.4
45.7
40.3
8.2
45.3
50.9
23.0
59.3
32.0
18.1
15.8
113
148
Relative
Percent
Difference
-7.10
7.30
-27.0
-12.5
-31.4
-4.40
-44.3
-16.9
-22.0
-34.7
-12.8
-8.47
-6.23
17.6
Relative percent difference calculated as [(CEM value - SIE value)/SIE value] x 100.
bThere is currently no CEM or manual reference method for HCl. Therefore, there is no QC criteria
for the relative percent differences and both values are reported in this test report.
Dashes indicate run was invalidated. For more discussion, see Section 2 of this report.
-------
For the EPA Method 6 and GEM SO comparison, the relative differences
between the two values averaged -14.1, 3.63, and 9.3 percent, respectively,
for the inlet, midpoint, and outlet location. These differences were all
within the QC criterion for all locations.
As seen from Table 7-19, when comparing the manual versus GEM HC1
results, the average relative percent differences, based on the HC1 manual
result determined from SIE, were 20, 35, and 18, for the inlet, midpoint,
and outlet, respectively. However, since there is currently no HC1 GEM or
manual reference method and there is no QC criteria for the absolute relative
differences, these values are reported for informational purposes only. The
high positive differences (GEM values generally lower than manual results)
observed for the midpoint location are most likely due to a reaction of HCl in
the GEM sampling system (interface). The differences for the inlet and outlet
locations were variable, and averaged 19.8 and 17.6 percent, respectively. An
extra dry impinger was added to the HCl sampling train which resulted in
greater liquid contact and improved relative accuracy.
7.3.4 Manual Sampling
HCl sampling was based on EPA Reference Method 5 with modifications
which allowed the collection of HCl in the back half of the sampling train.
Calibrations and/or inspections were made on all equipment prior to sampling.
Sample train glassware and high-density polyethylene sample bottles were
precleaned as previously described. All cleaned glassware was then sealed
with glass plugs or parafilm to prevent contamination. Table 7-20 summarizes
the leakchecks for the HCl trains, which were all within the QC criteria of
0.02 cfm.
7.3.5 Validation of Fixed Gases Results
The validity of Orsat and CEM 0? and CO,, analysis results was confirmed
based on a combustion stoichiometry method. Normally, the ultimate CCL
concentrations were calculated based on an ultimate analysis of the fuel.
However, since ultimate analyses were not performed on the refuse
lmo/038 7-35
-------
TABLE 7-20. LEAKCHECK SUMMARY FOR THE MARION COUNTY
HC1 SAMPLING TRAINS
Date Test
Condition
6-04-87 1
1
1
6-05-87 2
2
2
6-06-87 3A
3A
3A
3B
3B
3B
Sampling
Location
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Leak
Check
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Leak,
Rate*'0
(ft3/min)
0.016
0.005
0.015
0.007
0.010
0.020
0.009
0.006
0.013
0.015
0.010
0.012
0.005
0.003
0.014
0.005
0.009
0.004
0.008
0.013
0.008
0.003
0.012
0.010
Pressure
(in. H20)
10
4
15
3
10
--
15
8
13
4
10
7
25
10
15
5
8
8
12
20
16
5
8
5
Locations sampled are relative positions in the air pollution control
system.
Leak rates are expressed in actual cubic feet of gas over a two minute
period.
CQC criteria is <0.02 acfm or 4 percent of sampling rate, whichever is less.
--Dash indicates leakrate was less than or equal to 0.02 acfm, but not
recorded.
lmo/038
7-36
-------
TABLE 7-20. LEAKCHECK SUMMARY FOR THE MARION COUNTY
HC1 SAMPLING TRAINS (Continued)
Date
6-08-87
6-09-87
6-10-87
6-11-87
Test
Condition
4
4
4
5
5
5
6A
6A
6A
6B
6B
6B
7
7
7
Sampling
Location
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Leak
Check
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Leak,
Rate15'0
(ft3/min)
0.007
0.008
0.010
0.003
0.012
0.005
0.012
0.002
0.008
0.008
0.007
0.007
0.007
0.006
0.009
0.003
0.012
0.003
0.014
0.009
0.003
0.002
0.012
0.007
0.010
0.005
0.005
0.003
0.005
0.017
Pressure
(in. H20)
15
10
12
5
6
4
14
5
12
5
6
5
15
5
10
4
8
4
14
5
10
5
5
4
15
6
10
5
5
4
aLocations sampled are relative positions in the air pollution control
system.
Leak rates are expressed in actual cubic feet of gas over a two minute
period.
°QC criteria is <0.02 acfm or 4 percent of sampling rate, whichever is less.
--Dash indicates leakrate was less than or equal to 0.02 acfm, but not recorded.
lmo/038
7-37
-------
TABLE 7-20. LEAKCHECK SUMMARY FOR THE MARION COUNTY
HC1 SAMPLING TRAINS (Continued)
Date Test
Condition
6-12-87 8
8
8
9
9
9
6-15-87 10
10
10
6-16-87 11A
11A
11A
11B
11B
11B
Sampling
Location
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Inlet
Midpoint
Outlet
Leak
Check
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Initial
Final
Leak,
Rat?'0
(ft3/min)
0.005
0.012
0.004
0.001
0.005
0.010
0.011
0.020
0.011
0.006
0.005
0.007
0.012
0.004
0.008
0.001
0.012
0.010
0.008
0.006
0.006
0.005
0.012
0.013
0.008
0.010
0.008
0.003
0.007
0.007
Pressure
(in. H20)
15
15
10
12
5
4
18
4
10
4
6
5
15
5
15
5
5
4
15
6
5
7
15
4
15
4
15
9
6
4
Locations sampled are relative positions in the air pollution control
system.
Leak rates are expressed in actual cubic feet of gas over a two minute
period.
°QC criteria is <0.02 acfm or 4 percent of sampling rate, whichever is less.
--Dash indicates leakrate was less than or equal to 0.02 acfm, but not recorded.
lmo/038
7-38
-------
from this site an average based on the individual analyses was used. This
approach assumes that the majority of the analyses are correct and intends to
identify individual poor analyses. Plots of 0 versus CO- were made for both
GEM and Orsat analyses at the inlet, midpoint, and outlet. An F was
calculated for each point using the equation:
FQ = (20.9 - %02, dry)/(% C02> dry)
Manipulation of this equation yields a straight line with slope equal to F .
The intercepts of this line are 20.9 percent for the 0^ axis and ultimate CO
for the CO- axis. The ultimate CO is the theoretical CO concentration at
20 9
zero percent excess air. (CO ult. — ' ).
2 F
o
An average FQ and ultimate CO was calculated for each type of CO /O
analysis for each location. Plots of the lines determined from these
parameters are shown in Figures 7-1 through 7-6. In each case, all of the
points adhered well to the lines indicating precision in the measurements.
It should be noted, however, that leaks are not detected by this method
since the points would only move along the given lines due to a leak. This
method ensures the integrity of the analysis, not the sampling. Comparison of
GEM and Orsat plots will, however, give some insight into sampling integrity.
These plots generally compare well. Test 6A at the GEM midpoint may have had
some leakage judging from the position of the point on the line compared to
the Orsat analysis, although the evidence is not conclusive. A degree of
accuracy (bias) may be determined by comparing the ultimate CO- analysis for
each point. With the exception of the GEM outlet system, all ultimate CO,,
values were near 18 percent CO-. The GEM outlet ultimate CO- was slightly
less than 19 percent.
7.3.6 EPA Method 6 SO- Quality Control
2
Sampling and analysis for SO- followed EPA Method 6 except that the
train was modified to use full-sized impingers. Quality control for the S02
analysis included duplicate titrations and analysis of a blank. These
results are presented in Table 7-21.
lmo/038 7-39
-------
i
*-
o
fc?
z
o
tt
z
u
o
z
o
u
z
lit.
u
X
o
0
VALIDATION OF FIXED GAS ANALYSIS
INLET CEM OXYGEN AND CARBON DIOXIDE
CARBON DIOXIDE CONCENIRATION
Figure 7-1. Validation of fixed gas analysis for the inlet CEM results
-------
K
x
2
o
U
u
2
O
u
U
X
o
0
VALIDATION OF FIXED GAS ANALYSIS
INLET ORSAT OXYGEN AND CARBON DIOXIDE
CARBON DIOXIDC CONCLN[RATION (dry.XV)
Figure 7-2. Validation of fixed gas analysis for the inlet Orsat results
-------
«vl
-p-
K
2
O
5
a
H
2
Ul
u
2
O
u
2
U)
u
x
O
0
VALIDATION OF FIXED GAS ANALYSIS
MIDPOINT CEM OXYGEN AND CARBON DIOXIDE
CARBON DIOXIDE CONCENTRATION
Figure 7-3. Validation of fixed gas analysis for the midpoint CEM results
-------
•^1
-p-
z
o
a
z
U
u
z
o
u
z
u
o
X
o
VALIDATION OF FIXED GAS ANALYSIS
MIDPT. ORSAT OXYGEN AND CARBON DIOXIDE
CARBON DIOXIDE CONCENTRATION (diy.%V)
Figure 7-4. Validation of fixed gas analysis for the midpoint Orsat results
-------
2
O
a
UJ
u
2
O
u
z
Ui
O
X
O
VALIDATION OF FIXED GAS ANALYSIS
OUTLET CEM OXYGEN AND CARBON DIOXIDE
CARBON DIOXIDE CONCENTRATION (dry.%V)
Figure 7-5. Validation of fixed gas analysis for the outlet CEM results
-------
i
-P~
Ul
K
z
o
a
LJ
u
z
o
u
z
Ul
u
X
o
0
VALIDATION OF FIXED GAS ANALYSIS
OUTLET ORSAT OXYGEN AND CARBON DIOXIDE
12 H
CARBON DIOXIDE CONCENTRATION (dry.%V)
16
18
20
Figure 7-6. Validation of fixed gas analysis for the outlet Orsat results
-------
TABLE 7-21. DUPLICATE RESULTS FOR MARION COUNTY
METHOD 6 S02 TITRATIONSa>b
Sample/
Run No.
EPAQA9237
EPAQA4175
EPAQA8339
EPAQA2003
EPAQA7243
INLET 1
INLET 2
INLET 3
MIDPOINT 1
MIDPOINT 2
MIDPOINT 3
OUTLET 1
OUTLET 2
OUTLET 3
Result #1
2.95
2.60
16.5
9.35
17.8
74.7
20.1
25.6
37.9
16.0
40.1
17.2
4.00
14.1
Result #2
3.00
2.60
16.4
9.30
17.5
74.1
20.0
25.7
38.2
15.9
39.1
17.3
3.95
14.0
Average
2.98
2.60
16.5
9.33
17.7
74.4
20.1
25.7
38.1
16.0
39.6
17.3
3.98
14.1
Q
% Difference
1.68
0.00
0.61
0.54
1.70
0.81
0.50
-0.39
-0.79
0.63
2.53
-0.58
1.26
0.71
All values reported in milliliters of barium perchlorate titrant.
Analytical method detection limit was 1.3 parts per million (ppm) or
1.3 ug/ml.
Percent difference calculated as [(X - X2)/X] x 100, where X- =
result #1, X2 = result #2, X
+ X2>/2.
lmo/038
7-46
-------
The fourteen duplicate titrations all agreed within the QC criteria
( ± 1 percent difference between duplicates) except for EPAQA9237,
EPAQA7243, midpoint Run 3, and outlet Run 2, which were outside the QC
criteria (1.68, 1.70, 2.53, and 1.26 percent, respectively).
During analysis of the Method 6 S02 samples, an analytical sample blank
was analyzed concurrent with the field samples. This blank consisted of the
barium perchlorate titrant. The blank was analyzed with each batch of field
samples. For all three analyses, the blank showed less than 0.05
milliliters of titrant required or nondetectable quantities of sulfur
dioxide (analytical method detection limit was 1.3 ppm).
7.3.7 GEM Stratification Check
Stratification checks were initially performed with SO- but the
significant variability in the SO- concentrations with time for a given
point required that a less variable parameter such as NO and a reference
point be used. QC checks for flue gas stratification in the CEM sampling duct
were therefore performed on June 6, 1987, using NO . The NO concentration
X 3C
measured by a fixed reference probe (located at the CEM probe location) was
compared to the NO concentration measured by traversing the duct with the
X
sampling probe for approximately 5 minutes. These results are presented in
Tables 7-22, 7-23, and 7-24, for the inlet, midpoint, and outlet sampling
locations. As seen from the tables, the average relative percent differences
between the fixed probe and the traverse readings were -0.48, -5.39, and 0.77
for the inlet, midpoint, and outlet, respectively, indicating that
stratification was insignificant at the spray dryer inlet and the baghouse
outlet. Stratification at the midpoint was higher but still within the
10 percent acceptance criteria.
7.3.8 Sulfur Dioxide (SO ) Quenching Study
^
External performance audits were conducted on Radian's CEMs on June 2,
3, 4, and 24. The performance evaluation audit of the S02 CEMs revealed a
potential problem with the outlet S02 analyzer, which showed a high bias of
lmo/038 7-47
-------
TABLE 7-22. GEM STRATIFICATION CHECK FOR THE MARION
COUNTY INLET SAMPLING LOCATION
Traverse
Point
Co-located
Al
A3
A5
A7
A9
All
B2
B4
B6
B8
BIO
B12
NO Concentration
X
Fixed
Reference
Probe
228
198
219
194
188
195
258
225
244
224
213
216
238
(ppmV)
Traverse
Probe
229
200
220
198
193
197
252
225
244
223
216
219
238
Relative
Percent ,
Difference '
0.44
1.01
0.45
2.06
0.03
0.01
-0.02
0.00
0.00
-0.45
1.4
1.38
0.00
tlelative percent differences calculated as:
[(Traverse Probe - Fixed Reference Probe)/Fixed Reference Probe] x 100.
QC criteria was relative percent difference within ±10 percent.
lmo/038
7-48
-------
TABLE 7-23. GEM STRATIFICATION CHECK FOR THE MARION
COUNTY MIDPOINT SAMPLING LOCATION
Traverse
Point
Co -located
Al
A3
A5
A7
A9
All
Bl
B3
B5
B7
B9
fill
NO Concentration
X
Fixed
Reference
Probe
156
200
213
221
215
169
174
105
109
130
150
146
143
(ppmV)
Traverse
Probe
200
201
213
226
214
174
177
114
117
138
157
153
148
Relative
Percent ,
Difference '
28.2
0.50
0.00
2.26
-0.46
2.95
1.72
8.57
7.33
6.15
4.66
4.79
3.49
Relative percent differences calculated as:
[(Traverse Probe - Fixed Reference Probe)/Fixed Reference Probe] x 100.
QC criteria was relative percent difference within +10 percent.
lmo/038
7-49
-------
TABLE 7-24. CEM STRATIFICATION CHECK FOR THE MARION
COUNTY OUTLET SAMPLING LOCATION
NO Concentration (ppmV)
X
Traverse
Point
Co -located
Al
A2
A3
Cl
C2
C3
Fixed Reference
Probe
165
163
163
162
152
163
171
Traverse
Probe
164
162
161
160
152
162
169
Relative
Percent ,
Difference '
-0.60
-0.61
-1.22
-1.23
-0.00
-0.61
-1.16
Relative percent differences calculated as:
[(Traverse Probe - Fixed Reference Probe)/Fixed Reference Probe] x 100,
QC criteria was relative percent difference within ±10 percent.
lmo/038
7-50
-------
12.3 percent and 15.8 percent when challenged with an SO /CO audit gas
mixture on June 2 and June 17. The analyzer appeared to be calibrated
correctly when checked with SO calibration gas. This type of analyzer
requires a correction for the quenching caused by CO- and 0 The uncorrected
reading was very close to the audit cylinder value.
As a result, a study was initiated in-house to determine if the supplied
manufacturer's quench correction factor equations used to correct for an
interference caused by the presence of CO and 0 were valid for the two TECO
40 SO^ analyzers used at the Marion County characterization test. A detailed
report of this study can be found in Appendix H.
Two TECO 40 SO- analyzers and one Western SO- analyzer were used for
the Marion County testing. All of the SO- analyzers consistently passed
internal QC checks and linearity checks using certified gases containing
only SO in nitrogen. However, as previously mentioned, the TECO 40
instruments exhibited poor accuracy in analyzing audit gases containing both
S02 and CO-. One of the TECO 40 instruments typically responded low, but
within the required limits of +10 percent of the gas SO- concentration. The
other TECO 40 tended to respond high and slightly outside the QC limits. All
analyzers were thoroughly checked out and no apparent malfunctions were found.
Therefore, a post-test study was performed to determine whether revised quench
factors could be used to correct the data.
The SO- study on the two TECO 40 analyzers revealed that the TECO 40
#79 (used primarily for the outlet sampling location) required a revised
quench factor, while the manufacturer's equation was deemed suitable for the
TECO 40 #99 (used primarily for the midpoint sampling location). Table 7-25
reflects the average improved accuracy of 13 percent for concentrations
determined using the revised quench equation. Using only the revised
equation, only two samples in Table 7-25 did not meet the acceptance criteria
( ±10 relative percent difference for audit gases and ±20 relative percent for
Method 6 SO- concentrations).
lmo/038
7-51
-------
TABLE 7-25. COMPARISON OF MANUFACTURER'S AND DERIVED QUENCH
EQUATIONS FOR MARION COUNTY TECO 40 (179) S02 ANALYZER
Ul
to
S0_ Reference
Date
6/2/87
6/17/87
6/18/87
6/18/87
6/18/87
Sample Concentration
Audit Gas
Western Analyzer
Audit Gas
Western Analyzer
Method 6, Run 1
Method 6, Run 2
Method 6, Run 3
(ppmV)
219.0
228.7
219.0
235.9
115.7
29.5
96.5
Equation lla
SO, Concentration
TECO 179 (ppmV)
251.0
251.0
253.7
253.7
135.1
43.2
144.1
Relative Percent
Difference from
the Reference
Concentration0
-14.61
-9.73
-15.83
-7.53
-16.77
-46.3
-49.3
Equation 12
SO- Concentration
TECO 179 (ppmV)
227.1
227.1
229.5
229.5
115.9
38.2
123.5
Relative Percent
Difference from
the Reference
Concentration0
3.69
-0.73
4.80
-2.71
0.17
29.4
27.9
Manufacturer's quench factor equation.
Derived quench factor equation.
c Relative percent difference calculated as: [(Equation II or 12) -"(S02 Reference Concentrat1on)/S02 Reference
Equation] x 100.
-------
8.0 REFERENCES
1- The National Incinerator Testing and Evaluation Program. Air Pollution
Control Technology: Summary Report. Flakt Canada Ltd. and Environmental
Canada. September 1986. Report EPS 3/UP/2.
2. Letter Report from Phil Juneau and J. Ron Jernigan, Entropy Environmen-
talists, Inc., to Clyde E. Riley, U.S. Environmental Protection Agency,
Emissions Measurement Branch, Letter Report/Interim Test Report for HC1
Monitoring Conducted Under Contract No. 68-02-4336. Work Assignment
No. 11. July 10, 1987.
3. Anderson, Carol L., Dennis Knisley, Butch Stackhouse, Michael Vancil and
Donna Holder (Radian Corporation) Shutdown and Startup Emission Test
Report for the Marion County MWC. Prepared for the U.S. Environmental
Protection Agency. Research Triangle Park, North Carolina.
September 1988 - EMB Report No. 87-MIN-4A.
4. Anderson, Carol L., William P. Gergen, J. William Mayhew and Phyllis
O'Hara (Radian Corporation). Emissions Test Report for CDD/CDF. Metals.
HC1. and SO and Particulate Testing at the Marion County MWC. Prepared
for the U.S. Environmental Protection Agency. Research Triangle Park,
North Carolina. September 1987. Radian DCN 87-222-124-06-16. EPA EMB
Report No. 86-MIN-3.
5. Letter Report from Michael A. Vancil, Radian Corporation to C.E. Riley,
EMB Task Manager, U.S. Environmental Protection Agency. Emission Test
Results for the PCDD/PCDF Internal Standards Recovery Study Field Test:
Runs 1. 2. 3. 5. 13 and 14. July 24, 1987.
6. Steinsberger, S., B. DeWees and R. Segall (Entropy Environmentalists,
Inc.) QA/QC Evaluation Report for Characterization Test Program at the
Marion County Solid Waste-to-Energv Facility. Prepared for the U.S.
Environmental Protection Agency. Research Triangle Park, North Carolina.
November 17, 1987.
7. Reference 5.
8. Reference 4.
9. Reference 1.
10. Test Methods for Evaluating Solid Waste. Volume II, Third edition, 1986.
SW 846.
lmo/036
-------
11. Hartman, Michael W., Winton E. Kelly, Donna J. Holder, Carol L.
Jarogochian, J. William Mayhew and Mary Jo Caldwell (Radian Corporation)
Field Test Plan for the Characterization Test Program at the Marion
County MWC. Prepared for the U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina. June 2, 1987. Radian
DCN 87-222-124-09-01.
12. Reference 4.
13. Reference 6.
lmo/036 8-2
-------
9.0 METRIC-TO-ENGLISH CONVERSION TABLE
Metric
English
0.028317 dscm
0.028317 dscmm
0.45359 kg/hr
1 ng/dscm
1 mg/dscm
°F
101325 Pa
1 ng/kg
1 dscf
1 dscfm
1 Ib/hr
.-10
4.3699 x 10 grains/dscf
4.3699 x 10"4 grains/dscf
(°C x 9/5) + 32°F
1 atm
-9
6.9998 x 10 grains/lb
6.9998 x 10"6 grains/lb
6.9998 x 10"3 grains/lb
lmo/036
9-1
------- |