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

-------