United States Office of Air Quality EMB Report No. 86-MIN-03
Environmental Protection Planning and Standards September 1987
Agency Research Triangle Park NC 27711
Air
Municipal Waste Combustion
Multipollutant Study
Emission Test Report
Marion County
Solid Waste-To-Energy Facility
Ogden Martin Systems of Marion, Inc.
Brooks, Oregon
Volume I: Summary of Results
-------
DCN No. 87-222-124-06-16 EMB Report No. 86-MIN-3
EMISSION TEST REPORT
CDD/CDF, METALS, HCl, S0
AND PARTICULATE TESTING
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 Assignments 2 and 6
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
William P- Gergen
J. William Mayhew
Phyllis O'Hara
Radian Corporation
Post Office Box 13000
Research Triangle Park, NC 27709
September 1987
-------
DISCLAIMER
This report has been reviewed by the
Emission Standards and Engineering 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, NC 27711.
-------
Acknowledgements
The work reported herein was performed by personnel
from Radian Corporation, Midwest Research Insitute (MRI),
and the U. S. Environmental Protection Agency (EPA).
Radian's Task Director, Carol Jamgochian, directed
the field sampling and analytical effort and was respon-
sible for summarizing the test and analytical data
presented in this report. Sample analyses were performed
by Radian Corporation in Research Triangle Park, North
Carolina, Triangle Laboratories, Inc., Research Triangle
Park, North Carolina and North Carolina State University
Nuclear Services Laboratory, Raleigh, North Carolina.
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.
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. Jim Barlow, Start-up Engineer
Mr. Russel Johnston, Chief Engineer
Mr. Don Penrose, Maintenance Supervisor
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.
-------
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 "Revised Sampling and Analytical
Plan for the Marion County Solid Waste-to-Energy Facility Boiler Outlet" dated
September 16, 1986. The sampling and analytical plan was reviewed and
accepted by the EPA/EMB Task Manager, Clyde E. Riley.
APPROVALS
Project Director: f jUiUs**^ <--K
-------
TABLE OF CONTENTS
Section Page
VOLUME I
1.0 INTRODUCTION 1-1
1.1 Background 1-2
1.2 Objectives 1-3
1.3 Brief Process Operation and Description 1-3
1.4 Emissions Measurement Program 1-5
1.4.1 Test Matrix 1-5
1.4.2 Sampling 1-5
1.4.3 Laboratory Analysis 1-7
1.5 Quality Assurance/Quality Control (QA/QC) 1-11
1.6 Description of Report Sections 1-11
2.0 SUMMARY OF RESULTS 2-1
2.1 CDD/CDF Emissions Results 2-2
2.1.1 2378-CDD Toxic Equivalency for Flue Gas Results. . 2-6
2.1.2 Cyclone Ash CDD/CDF Results 2-6
2.1.3 Baghouse Ash CDD/CDF Results 2-9
2.1.4 CDD/CDF Congener Distributions 2-9
2.2 Particulate Results 2-16
2.3 Specific Metals Results 2-16
2.3.1 Specific Metals-to-Particulate Ratios 2-19
2.3.2 Metals Results for the Baghouse Ash and
Cyclone Ash 2-19
2.3.3 Metals Results for the Lime Slurry and Tesisorb. . 2-22
2.4 Acid Gas Concentrations 2-24
2.4.1 SO- Results 2-24
2.4.2 HCI Results 2-29
2.4.3 Molar Ratios of Lime to HCI and SO 2-29
3.0 PROCESS DESCRIPTION AND OPERATION 3-1
3.1 Process Description 3-1
3.2 Air Pollution Control System 3-3
3.3 Combustor and Air Pollution Control System
Operating Conditions 3-4
4.0 SAMPLING LOCATIONS 4-1
4.1 Boiler Outlet Sampling Location 4-1
4.2 Baghouse Ash and Cyclone Ash Sampling Locations 4-6
4.3 Lime Slurry Sampling Location 4-6
4.4 Tesisorb Sampling Location 4-6
5.0 SAMPLING AND ANALYTICAL PROCEDURES 5-1
5.1 CDD/CDF Sampling and Analysis 5-1
5.1.1 Equipment and Sampling Preparation 5-3
5.1.2 Sampling Operations 5.4
-------
TABLE OF CONTENTS
(continued)
Section Page
5.1 CDD/CDF Sampling and Analysis (cont'd.)
5.1.3 Sample Recovery
5.1.4 Analysis 5'6
5.1.5 Data Reduction for CDD/CDF Results 5-10
5.2 HCl/Particulate Flue Gas Sampling and Analysis 5-11
5.2.1 Equipment and Sampling Preparation 5-11
5.2.2 Sampling Operations ^-1^-
5.2.3 Sample Recovery 5"13
5.2.4 Sample Analysis 5-13
5.2.5 Data Reduction 5'16
5.3 Lead/Cadmium/Particulate Determination 5-17
5.3.1 Equipment and Sampling Preparation 5-17
5.3.2 Sampling Operations 5-17
5.3.3 Sample Recovery 5-19
5.3.4 Analysis 5-19
5.3.5 Data Reduction 5-19
5.4 Hexavalent and Total Chromium and Nickel Sampling and
Analysis 5-22
5.4.1 Equipment and Sampling Preparation 5-22
5.4.2 Sampling Operations 5-22
5.4.3 Sample Recovery 5-24
5.4.4 Sample Analysis 5-24
5.4.5 Data Reduction 5-26
5.5 Baghouse Ash, Cyclone Ash, Lime Slurry and
Tesisorb Evaluations 5-26
5.5.1 Baghouse Ash and Cyclone Ash 5-26
5.5.2 Lime Slurry and Tesisorb Sampling and Analysis . . 5-29
5.6 SO Sampling and Analysis 5-31
5.6.1 Equipment and Sampling Preparation 5-31
5.6.2 Sampling Operations 5-31
5.6.3 Data Reduction 5-33
5.7 Molecular Weight by EPA Method 3 5-33
5.7.1 Sampling Operations 5-33
5.7.2 Analysis 5-35
5.8 Volumetric Flowrate by EPA Method 2 5-35
5.8.1 Sampling and Equipment Preparation 5-35
5.8.2 Sampling Operations 5-35
5.9 Moisture by EPA Method 4 5-36
6.0 QUALITY ASSURANCE AND QUALITY CONTROL 6-1
6.1 Standard Quality Assurance Procedures 6-1
6.1.1 Sampling Equipment Preparation 6-1
6.1.2 General Sampling QC Procedures 6-3
6.1.3 Sample Recovery 6-4
6.1.4 Sample Custody 6-5
11
-------
TABLE OF CONTENTS
(continued)
Section
6.1 Standard Quality Control Procedures (cont'd.)
6.1.5 Preparation of Samples for Analysis 6-5
6.1.6 Sample Analysis 6-6
6.1.7 Data Documentation and Verification 6-6
6.2 Method-Specific Sampling QC Procedures 6-6
6.2.1 Quality Control Procedures for Velocity/Volumetric
Flow Rate Determination 6-7
6.2.2 Quality Control Procedures for Molecular Weight
Determinations 6-7
6.2.3 Quality Control Procedures for Moisture
Determination 6-8
6.2.4 Quality Control for CDD/CDF Testing 6-8
6.2.5 Quality Control for Particulate Testing 6-22
6.2.6 Quality Control for HCl/Particulate Testing. ... 6-25
6.2.7 Quality Control for Pb/Cd/Particulate 6-30
6.2.8 Quality Control for Cr/Ni Testing 6-38
6.2.9 Quality Control Procedures for Continuous SO-
Determinations 6-44
6.2.10 Quality Control Procedures for Process Sampling. . 6-51
7.0 References 7-1
8.0 English to Metric Conversion Table 8-1
VOLUME II
APPENDIX A - SUMMARY OF PARTICULATE, DIOXIN, HC1, AND
TRACE METALS TEST DATA
A.I Dioxin Test Data Summaries - Boiler Outlet A-3
A. 2 HCl/Particulate Test Data Summaries - Boiler Outlet A-7
A.3 Pb/Cd/Particulate Test Data Summaries - Boiler Outlet A-13
A.4 Cr/Ni Test Data Summaries - Boiler Outlet A-19
A.5 Sample Calculations A-23
A. 5.1 CDD/CDF Data Reduction Equations A-25
A. 5. 2 Method 5 Data Reduction Equations A-27
A. 5. 3 Trace Metals Data Reduction Equations A-32
A.5.4 Analyte to Particulate Ratio A-33
A. 5. 5 GEM Data Reduction Equations A-34
A. 5. 6 HCl Data Reduction Equations A-36
A. 5. 7 Ratio of Actual Lime to Stoichiometric Lime A-36
A. 5. 8 Calculating Emission Factor A-37
A. 5. 9 Outlier Analysis of Uncontrolled PM Data A-37
A. 5.10 Refuse Feed Rate Calculations A-38
A.5.11 Calculation of Detection Limit for Hexavalent Chromium . . . A-40
111
-------
TABLE OF CONTENTS
(continued)
Section
APPENDIX B SUMMARY OF GEM TEST DATA
B.I CEM Test Data Summary B'3
B.2 Printout Data: One Minute Averages B-7
B.3 CEM Calibration Summary B'25
CEM Calibration Computer Printouts B-29
B.4 CEM Data Reduction Equations B-33
B.5 CEM Data Stripcharts B'37
B.6 CEM Daily Log B'43
APPENDIX C FIELD DATA SHEETS
C.I MM5 Dioxin Field Data Sheets Boiler Outlet C-3
C.2 M5 HCl/Particulate Field Data Sheets - Boiler Outlet C-23
C.3 Pb/Cd/Particulate Field Data Sheets Boiler Outlet C-43
C.4 Cr/Ni Field Data Sheets - Boiler Outlet C-63
C.5 K-Factor/Nomograph Field Data Sheets C-83
C.6 Preliminary Sampling and Velocity Traverse Sheets C-87
C.7 Orsat Analysis Field Data Sheets C-93
C.8 Baghouse Ash Field Data Sheets C-101
C.9 Cyclone Ash Field Data Sheets C-109
C.10 Lime Slurry Field Data Sheet C-117
C.ll Tesisorb Field Data Sheet C-123
APPENDIX D - LABORATORY ANALYTICAL RESULTS
D.I CDD/CDF Analytical Data D-3
CDD/CDF Data Summaries ..... D-5
Triangle Labs Analytical Report D-28
D.2 HC1 Analytical Data D-67
HC1 Data Summaries D-69
Radian/RTP Analytical Report D-73
D.3 Pb/Cd Analytical Data D-83
Pb/Cd Data Summaries D-85
Radian/RTP Analytical Report D-89
D.4 Cr/Ni Analytical Data D-97
Cr/Ni Data Summaries D-99
Radian/RTP Analytical Report D-103
Notebook #13540 D-lll
Notebook #13888 D-137
D.5 Particulate Analytical Data D-161
Particulate Data Summaries D-163
Radian/RTP Analytical Data Sheets D-166
D.6 Hexavalent and Total Chromium Analysis by AA of
the Baghouse and Cyclone Ash Analytical Data D-169
D.7 Analytical Results for AA Analyses of Sodium Matrix
Study Samples D-183
-------
TABLE OF CONTENTS
(continued)
Section Page
VOLUME III
APPENDIX E - CALIBRATION DATA
E.I Summary of Equipment Used During Test Program E-3
E.2 Equipment Calibration Sheets E-9
APPENDIX F - PERTINENT CORRESPONDENCE
F.I Performance of Marion Test Program; Letter Report F-3
F.2 Dioxin Analysis F-9
F.3 Metals Analysis F-25
APPENDIX G - COMBUSTOR AND CONTROL DEVICE DATA
G.I Process Data Logsheets G-3
G.2 Process Feed Sheets G-39
G.3 Baghouse Data Log Sheets G-65
APPENDIX H - FIELD TEST LOGS
H.I Daily Sampling Log Summary H-3
H.2 Crew Chief Field Operations Log H-7
H.3 Field Laboratory Log H-47
H.4 GEM Operators Log H-65
H.5 Sample Identification Log H-77
APPENDIX I - QUALITY ASSURANCE INFORMATION
I.I CDD/CDF Quality Control Results 1-3
1.2 CDD/CDF XAD and Filter Preparation Information 1-25
1.3 Pb/Cd Quality Control Results 1-33
1.4 Cr/Ni Quality Control Results 1-41
1.5 Particulate Quality Control Results 1-47
1.6 HCl Quality Control Results 1-51
1.7 CEM Quality Control Results 1-55
1.7.1 Summary of Daily Calibrations 1-57
1.7.2 System Bias Check 1-61
1.7.3 Response Times 1-65
1.7.4 Daily QC Check 1-69
1.7.5 Daily Drift Check 1-73
1.7.6 Multi-point Calibration 1-77
1.7.7 CEM Standard Gas Certification Sheets 1-81
1.8 Sampling Isokinetics and Leak Checks 1-87
APPENDIX J - SAMPLING AND ANALYTICAL PROTOCOLS
J.I Summary of EPA Reference Methods Used During Test Program . . . J-3
J.2 Baghouse Ash, Cyclone Ash, Lime Slurry, and Tesisorb Sampling
and Analytical Procedure J-9
J.3 ASME/EPA Protocol for the Determination of Chlorinated Organic
Compounds in Stack Emissions (December, 1984 Draft) J-17
J.4 Draft EPA Method for the Determination of Cadmium from
Stationary Sources j-49
J.5 Draft EPA Protocol for Emissions Sampling for Both Hexavalent
and Total Chromium (February 22, 1985) J-63
APPENDIX K - PROJECT PARTICIPANTS K-3
v
-------
LIST OF FIGURES
Figure Page
1-1 Marion County Process Line with Sampling Locations 1-4
2-1 CDD Homologue Distribution at Marion County 2-11
2-2 CDF Homologue Distribution at Marion County 2-12
2-3 Uncontrolled S0_ Concentration History at Marion County 2-27
2-4 0. Concentration History at Marion County 2-28
3-1 Marion County Process Line 3-2
4-1 Marion County Process Line With Sampling Locations 4-2
4-2 Side View of Boiler Outlet Sampling Location at Marion County. . 4-3
4-3 Top View of Boiler Outlet Sampling Location at Marion County . . 4-4
4-4 Traverse Point Location Diagram for Boiler Outlet Location at
Marion County 4-5
4-5 Baghouse Ash and Cyclone Ash Sampling Locations at Marion County 4-7
4-6 Tesisorb Sampling Locations at Marion County 4-8
5-1 CDD/CDF Sampling Train Configuration Used at Marion County ... 5-4
5-2 CDD/CDF Field Recovery Scheme Used at Marion County 5-6
5-3 CDD/CDF Analytical Scheme Used at Marion County 5-9
5-4 Particulate/HCL Sampling Train Configurations Used at
Marion County 5-12
5-5 Marion County Particulate/HCl Field Recovery and
Analytical Recovery 5-14
5-6 Particulate/Pb/Cd Sampling Train Configuration 5-18
5-7 Marion County Particulate/Pb/Cd Field Recovery and Analytical
Protocol 5-20
5-8 Cr/Ni Sampling Train Configuration Used at Marion County .... 5-23
5-9 Marion County Cr/Ni Field Recovery and Analytical Protocol ... 5-25
VI1
-------
Figure
LIST OF FIGURES
(Continued)
5-10 Marion County Baghouse Ash and Cyclone Ash CDD/CDF
Analytical Protocol ..................... 5-27
5-11 Marion County Baghouse Ash and Cyclone Ash Metals
Analysis Protocol ...................... 5-28
5-12 Marion County Metals Analytical Protocol for Tesisorb and
Lime Slurry Samples ..................... 5-30
5-13 SO GEM Sampling and Analysis Scheme for the Marion County
Test Program ......................... 5-32
5-14 Method 3 Integrated Bag Sampling Train ............. 5-34
6-1 Validation of Fixed Gas Analysis for Marion County 0_ and
CO Data ........................... 6-9
6-2 Multipoint GEM Calibrations for Marion County .......... 6-46
6-3 Daily Quality Control Check ................... 6-50
VI11
-------
LIST OF TABLES
Table Page
1-1 Overall Sampling Matrix for Marion County 1-6
1-2 Summary of Sampling Log for EPA Testing at Marion County .... 1-8
1-3 CDD/CDF Congeners Analyzed for the Marion County Test Program. . 1-10
2-1 Summary of Uncontrolled CDD/CDF Emissions for Marion County. . . 2-3
2-2 Uncontrolled CDD/CDF Emissions at Marion County MWC 2-4
2-3 Summary of Uncontrolled CDD/CDF Emissions for Marion County MWC
(Front, Back, and Total Fraction Results) 2-5
2-4 Uncontrolled CDD/CDF Concentrations Expressed as 2378-TCDD
Toxic Equivalents for Marion County MWC 2-6
2-5 Marion County MWC CDD/CDF Cyclone Ash Concentration and
2378-TCDD Toxic Equivalency 2-8
2-6 Marion County MWC CDD/CDF Baghouse Ash Concentration and
2378-TCDD Toxic Equivalency 2-10
2-7 Uncontrolled Flue Gas CDD/CDF Congener Distribution at
Marion County MWC 2-13
2-8 Cyclone Ash CDD/CDF Congener Distribution at Marion County MWC . 2-14
2-9 Baghouse Ash CDD/CDF Congener Distribution at Marion County MWC. 2-15
2-10 Summary of Uncontrolled Particulate Emissions
for the Marion County MWC 2-17
2-11 Summary of Uncontrolled Specific Metals Emissions for the
Marion County MWC 2-18
2-12 Uncontrolled Specific Metals Mass Emission Rates for the
Marion County MWC 2-20
2-13 Ratio of Uncontrolled Metals to Particulate Mass for the
Marion County MWC 2-21
2-14 Metals Results for Baghouse Ash and Cyclone Ash at
Marion County MWC 2-23
2-15 Metals Results for Lime Slurry and Tesisorb 2-24
2-16 Uncontrolled Marion County MWC S02 Results 2-26
IX
-------
LIST OF TABLES
(Continued)
Table
Page
2-17 Uncontrolled Flue Gas Chloride Concentration at the
Marion County MWC 2-30
2-18 Molar Ratio of Actual Lime to Stoichiometric Lime for
HC1 and SO 2"31
3-1 Air Pollution Control System Design Specifications 3-5
3-2 Average Process Data for Marion County Incinerator Test 3-6
3-3 Average Dry Scrubber/Fabric Filter Operating Data for the
Marion County MWC 3-7
5-1 Sampling Methods and Analytical Procedures Used for the
Marion County Test Program 5-2
5-2 CDD/CDF Sampling Train Components for Marion County
Shipped to Analytical Laboratory 5-8
5-3 HCl/Particulate Sample Components Collected at
Marion County MWC 5-15
5-4 Sample Components for Pb/Cd/Particulate Train Used at the
Marion County MWC 5-21
6-1 Summary of Equipment Used in Performing Source Sampling at the
Marion County MWC 6-2
6-2 CDD/CDF Testing Isokinetics and Leak Check Summary Boiler
Outlet at Marion County 6-11
6-3 Internal Standards Recovery Results for Marion County
CDD/CDF Analyses 6-13
6-4 Factors Adjusting Responses for Extraction Efficiency and
Variable Instrument Performance 6-15
6-5 Surrogate Recoveries for Marion County CDD/CDF Analyses 6-16
6-6 Marion County CDD/CDF Audit Samples 6-18
6-7 Summary of Dioxin and Furan Analyses Recovery Efficiency and
Lab Proof Blanks 6-20
6-8 Recovery Efficiency Blank Dioxin/Furan Data for Marion County
MM5 Samples 6-21
x
-------
LIST OF TABLES
(Continued)
Table Page
6-9 Summary of CDD/CDF Soxhlet and Aqueous Blank Results 6-23
6-10 CDD/CDF Duplicate Analyses for Marion County MWC 6-24
6-11 HCl/Particulate Testing Isokinetics and Leak Check Summary
Boiler Outlet at Marion County 6-26
6-12 Pb/Cd/Particulate Testing Isokinetics and Leak Check Summary -
Boiler Outlet at Marion County 6-27
6-13 Summary of Uncontrolled Particulate Blank Results 6-28
6-14 Summary of Uncontrolled HC1 Blank Results for the
Marion County MWC 6-29
6-15 Summary of HC1 Audit Sample Results Marion County Solid
Waste-to-Energy Facility, Boiler Outlet 6-32
6-16 Labproof and Recovery Efficiency Blanks for Pb/Cd Train at
Marion County 6-35
6-17 Results of Pb/Cd Audit Sample Analysis for Marion County
Test Program 6-36
6-18 Internal Laboratory QA Samples For Pb/Cd
Analyses by Atomic Absorption 6-37
6-19 Cr/Ni Testing Isokinetics and Leak Check Summary -
Boiler Outlet at Marion County 6-39
6-20 Laboratory Proof Blank, Recovery Efficiency Blank and
Reagent Blank Results for Cr/Ni Train at Marion County. ... 6-41
6-21 Results of Cr/Ni Audit Sample Analysis for Marion County
Test Program 6-42
6-22 Internal Laboratory QA Samples For Cr/Ni
Analyses by Atomic Absorption 6-43
6-23 Leak Check Summary for Marion County S02 GEM System 6-45
6-24 Summary of S02 and 02 Instrument Calibrations - Marion County. . 6-48
6-25 Summary of S02 and 02 Drift Check Results for Marion County. . . 6-49
6-26 GEM Response Times at Marion County 6-52
XI
-------
1.0 INTRODUCTION
1.0 INTRODUCTION
The United States Environmental Protection Agency (EPA) has published in
the Federal Register (July 7, 1987) an advanced notice of proposed rulemaking
which describes upcoming emissions standards development for new municipal
waste combustors (MWC's) under Section 111 of the Clean Air Act and for
existing MWC's under Section lll(d) of the Act. This Federal Register notice
culminates more than a year's work of development of the technical and health
related documents which comprise EPA's Report to Congress on MWC's. The
Report to Congress was a joint effort involving the efforts of 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 Measurement Branch (EMB), is
responsible for reviewing the existing air emissions data base and gathering
additional data where necessary. As a result of this review, several MWC
emission tests were performed and are in planning stages to support the
emissions standards development which is underway. Of particular importance
is a more complete data base on emerging air pollution control technologies
for MWC's.
The emissions that are being studied in this assessment are the criteria
pollutants--particulate matter (PM), sulfur oxides (SO ), nitrogen oxides
(NO ), carbon monoxide (CO) and total hydrocarbons (THC); other acid gases,
X
such as hydrochloric acid (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
-------
1.1 BACKGROUND
Ogden Martin Systems of Marion, Inc. was required by the Oregon
Department of Environmental Quality (ODEQ) to conduct a compliance testing
program to measure controlled particulate, CDD/CDF, metals, HC1 and S02
emissions at their Marion County Solid Waste-to-Energy Facility in Brooks,
Oregon. Process data were also collected as part of the compliance testing
program.
In order to provide additional data to evaluate the CDD/CDF and metals
removal effectiveness of emissions reduction systems, Ogden Martin Systems and
EPA agreed to jointly sponsor an expanded program during the ODEQ-required
tests. Ogden Martin Systems sponsored measurements of the controlled
emissions from the facility EPA sponsored measurements of the flue gas
at the boiler outlet prior to the emission control system, as well as sampling
of the cyclone flyash, baghouse flyash, Tesisorb, and lime slurry. Radian
Corporation, under contract to the Emissions Measurement Branch, performed the
EPA-sponsored testing. The Ogden Martin Systems-sponsored testing was
performed by Ogden Projects, Inc.
This report presents the results of the EPA-sponsored testing. The
results of the test program sponsored by Ogden Martin Systems have been
123
reported separately ' '
In addition to this report, a summary report is being prepared by Radian
Corporation that will present the complete set of data that was collected
during the joint sampling program. The summary report will discuss and
analyze the uncontrolled and controlled data to show removal efficiency for
CDD/CDF, metals, HC1, SO and particulates. The baghouse ash, bottom ash,
lime slurry and Tesisorb results will also be included, along with a more
detailed analysis of the continuous monitoring, combustor, and control
device operating data.
1-2
-------
1.2 OBJECTIVES
The main objective of the EPA-sponsored test program was to obtain
uncontrolled flue gas emissions data that could be compared with the Ogden
Martin-sponsored controlled flue gas emissions data. Specifically, the
EPA-sponsored test program was designed to:
Coordinate sampling at the control device inlet (EPA-sponsored) with
sampling at the control device outlet (Ogden Martin-sponsored) to
obtain data on efficiency of the control device for comparable
CDD/CDF, trace metals, S09, HC1 and particulate emissions.
Determine the uncontrolled and controlled emission characteristics
and inter-relationship of the particulate matter, CDD/CDF and trace
metals flue gas concentrations.
Determine the CDD/CDF and trace metal content of the baghouse ash
and cyclone ash. Determine the trace metal content only of the lime
slurry and Tesisorb.
Characterize process operations for each test run by recording
combustor and control device operating data as well as feed rates.
The Marion County facility was selected by EPA because the facility
is a well-designed and operated mass burn, waterwall, resource recovery system
with a state-of-the-art emission control system. The results from the Marion
County Facility will be incorporated into a data base in support of future
regulatory development for the MWC source category.
1.3 BRIEF PROCESS OPERATION AND DESCRIPTION
Figure 1-1 presents a process diagram of one of the two identical
combustor systems at the Marion County Facility. Unit No. 1, tested during
this program, is a reciprocating grate, mass-burning type combustor with a
waterwall boiler that produces superheated steam. The flue gas passes from
the combustor into convection, superheater, and economizer sections before
being treated by a quench reactor and baghouse 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
1-3
-------
To Atmosphere
Furnace
Boiler Superheater Economizer
Quench Reactor/ Teslsorb
Acid Gas Feed
Scrubber Hopper
Lime Slurry Dry
Mixing Tank Venturl
Quench
Pit
Grizzly
Scalper
Distributor
I.D. Fan stack
Figure 1-1. Marion County Process Line
-------
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 combustor charging chute. A Martin inclined grate and ash discharge
system is used at the Marion County facility.
The process description and operation discussion (Section 3) in this
report were prepared by ISB/EPA. Also, the process data recorded during the
test program are summarized in Section 3.0.
1.4 EMISSIONS MEASUREMENT PROGRAM
1.4.1 Test Matrix
The emissions measurement program at the Marion County facility was
conducted from September 22 to 30, 1986. Table 1-1 presents the overall test
matrix that was planned by EPA and Ogden Martin Systems for the program and
the organization that sponsored each type of sample. Sampling at the boiler
outlet and stack was conducted simultaneously for parameters measured during
the test program. Ogden Projects and Radian followed the same sampling
protocols and shared reagents when necessary to maintain and assure data
comparability. Test Runs 1, 2, and 3 were designed to measure CDD/CDF and
HCl/particulate matter concurrently. Test Runs 4, 5, and 6 were designed to
measure particulate matter, lead, cadmium, nickel, total chromium and
hexavalent chromium concurrently. Additional test runs were conducted by
Ogden Projects to measure the controlled emissions of NO , SO., CO, beryllium
X £.
mercury, HC1, and particulate matter. The results of these tests can be
obtained in References 4 to 6.
1.4.2 Sampling
In order to obtain as much data as possible during the sampling effort,
the sampling trains for the parameters studied were combined when possible.
The sampling trains were combined only if the precision and accuracy of the
data obtained for each parameter would not be affected. For this test
1-5
-------
TABLE 1-1. OVERALL SAMPLING MATRIX FOR THE MARION COUNTY MWC TEST PROGRAM
i
o\
Sample
CDD/CDF
Boiler Outlet
Outlet Stack
Participate Loading
Boiler Outlet
Outlet Stack
Pb/Cd
Boiler Outlet
Outlet Stack
Cr(VI)/Cr(III)/N1
Boiler Outlet
Outlet Stack
Hydrochloric Add
Boiler Outlet
Outlet Stack
Sulfur Dioxide
Boiler Outlet
Outlet Stack
Cyclone Ash
Baghouse Ash
Lime SI urry
Teslsorb
Run 1
EPA
Ogden Martin
EPA
Ogden Martin
—
—
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPAa
EPAa
—
--
Run 2
EPA
Ogden Martin
EPA
Ogden Martin
—
—
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPAa
EPAa
--
—
Run 3
EPA
Ogden Martin
EPA
Ogden Martin
—
—
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPAa
EPAa
--
—
Run 4
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPA
Ogden Martin
—
—
Ogden Martin
EPAb
EPAb
EPAC
EPAC
Run 5
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPA
Ogden Martin
—
—
Ogden Martin
EPAb
EPAb
EPAC
EPAC
Run 6
—
—
EPA
Ogden Martin
EPA
Ogden Martin
EPA
Ogden Martin
—
—
Ogden Martin
EPAb
EPAb
EPAC
EPAC
CDD/CDF analysis
Analysis for total chromium by AA, hexavalent chromium by colorimetric method, and arsenic, total chromium,
cadmium and nickel by NAA.
"Analysis for arsenic, total chromium, cadmium, and nickel by NAA.
-------
program, HC1 and particulate; lead, cadmium and particulate; and chromium and
nickel were collected in combined sampling trains. Particulate samples were
collected in the HC1 train for Runs 1, 2, and 3 and in the Pb/Cd train for
Runs 4, 5, 6.
During Runs 1, 2 and 3, CDD/CDF and HC1/PM sampling was conducted by
manual methods. SO. was measured by continuous emissions monitoring. Also,
baghouse ash and cyclone ash samples were collected for CDD/CDF analysis. The
CDD/CDF sampling was conducted according to the December 1984 draft of the
Environmental Standards Workshop protocol for sampling chlorinated organic
compounds. The protocol was developed jointly by EPA and the American
Society of Mechanical Engineers (ASME). HC1/PM sampling were conducted
according to EPA Reference Method 5 with modifications as described in
Section 5.2. SO- was measured according to EPA Method 6C.
During Runs 4, 5, and 6, Pb/Cd/PM and Cr/Ni sampling was conducted.
Baghouse ash, cyclone ash, lime slurry and Tesisorb samples were collected for
metals analysis. The draft EPA cadmium protocol was followed for the Pb/Cd/PM
Q
sampling. Cr/Ni sampling was conducted according to the draft EPA protocol
9
for emissions sampling of hexavalent and total chromium. Detailed sampling
procedures are available in Section 5 and Appendix J.
A summary of the sampling log for the EPA-sponsored test program is
presented in Table 1-2. The summary identifies the samples collected and the
sampling times for the EPA-sponsored tests as well as any problems that
occurred. A detailed log is provided in Appendix H.
1.4.3 Laboratory Analysis
The laboratory analyses were performed by three organizations: Triangle
Laboratories, Inc., North Carolina State University and Radian Corporation.
The CDD/CDF analyses were performed by Triangle Laboratories, Inc., Research
Triangle Park, N.C. The metals analyses were performed by the Radian
Inorganic Laboratory in Research Triangle Park, North Carolina and Nuclear
1-7
-------
TABLE 1-2. SUMMARY OF SAMPLING LOG FOR EPA TESTING AT MARION COUNTY
September 22-30, 1986
Date
Run
Samples
Collected
Sampling
Period
Notes
9/22/86 1
9/23/86 2
9/24/86 3
9/26/86 4
9/29/86 5
CDD/CDF
HC1/PM, S02,
Baghouse and
Cyclone Ash
CDD/CDF
HC1/PM, S02
Baghouse and
Cyclone ash
CDD/CDF
HC1/PM, SO
Baghouse and
Cyclone Ash
Pb/Cd/PM
Cr/Ni,
Baghouse and
Cyclone Ash
Tesisorb and
Lime
Pb/Cd/PM
Cr/Ni,
Baghouse and
Cyclone Ash
Tesisorb and
Lime
13:16 17:36
13:51 18:26
10:13 - 14:49
7:15 13:35
10:40 - 20:25
SO- data was not collected
for 25 percent of sampling
period due to a malfunction
in the gas conditioner. No
incinerator operating
problems occurred during
sampling period.
No sampling or incinerator
operating problems occurred
during sampling period.
No sampling or incinerator
operating problems occurred
during sampling period.
Pb/Cd/PM train collected
sample for 350 minutes
instead of 360 minutes due to
high vacuum in train. No
incinerator operating
problems occurred during
sampling period.
Crane malfunctions from
10:42 11:27 and 12:25 -
13:00. Stopped testing
from 10:50 14:00. No
sampling problems occurred.
9/30/86
Pb/Cd/PM
Cr/Ni,
Baghouse and
Cyclone Ash
Tesisorb and
Lime
9:30 16:20
Sample volume of Pb/Cd/PM
train was corrected for a
final leakrate above
0.02 ft /min. Correction
was 0.4 percent of total
sample volume. No inciner-
ator operating problems
occurred during the sampling
period.
The sampling period includes time for port changes and other down time
periods that occurred during sampling.
l-i
-------
Energy Services Department of Nuclear Engineering, North Carolina University.
The particulate samples were analyzed by the Radian Field Sampling Laboratory
also located in Research Triangle Park, North Carolina.
The CDD/CDF samples were analyzed by high resolution gas chromatography
and high resolution mass spectrometry (GC/MS) according to the December 1984
draft of the Environmental Standards Workshop Protocol. The congeners that
are reported are listed in Table 1-3. The total mono- through
octa-chlorinated homologues are reported, along with all the individual
2378-substituted CDD/CDF isomers such as 2378-TCDD. Detailed analytical
procedures are given in Section 5, Section 6 (QC procedures), and in
Appendix J-
For the flue gas samples collected for metals analysis, atomic absorption
(AA) was used for lead, cadmium, nickel, and total chromium analysis. For
hexavalent chromium, the diphenylcarbizide colorimetric technique was used.
The precision for both analysis methods is generally +10 percent. The Pb/Cd
particulate samples were prepared for AA analysis by a hydrofluoric/nitric
acid digestion in a Parr bomb apparatus. The Pb/Cd impinger samples were
concentrated and dissolved with hydrofluoric nitric acid. For the Cr/Ni
samples, the entire sample was concentrated prior to digestion in an alkaline
solution. The resulting alkaline solution was filtered through a paper
filter. An aliquot of the filtrate was removed and acidified for nickel
analysis by AA. A second aliquot was removed for hexavalent chromium
determination. The paper filter and residue were digested in a
sulfuric/hydrofluoric acid solution prior to analysis for total chromium and
nickel by AA. The baghouse and cyclone ash samples were analyzed for total
chromium and hexavalent chromium, only, using the methods described above.
As part of an EPA in-house study, aliquots of the digested trace metal
samples as well as samples of lime slurry and Tesisorb were sent to the
Nuclear Services Laboratory at North Carolina State University for neutron
activation analysis (NAA). Since NAA is not an EPA reference method, the
results from the NAA analyses are reported separately.
1-9
-------
TABLE 1-3. 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)
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)
1,2,3,7,8,9 Hexachlorodibenzofuran (1,2,3,7,8,9 HxCDF)
1,2,3,6,7,8 Hexachlorodibenzofuran (1,2,3,6,7,8 HxCDF)
Total Hexachlorinated dibenzofurans (HxCDF)
Total Heptachlorinated dibenzofurans (HpCDF)
Total Octachlorinated dibenzofurans (OCDF)
1-10
-------
1.5 QUALITY ASSURANCE/QUALITY CONTROL (QA/QC)
Completeness and data quality were emphasized during the test program at
the Marion County facility. QA/QC measures were incorporated into each
sampling or analytical task. The QA/QC program and results were reviewed by
the Radian Quality Assurance (QA) Officer. Section 6 addresses the QA/QC
techniques implemented during testing and analysis and the quality control
results.
1.6 DESCRIPTION OF REPORT SECTIONS
The emissions report is presented in three volumes. Volume I includes
the Summary of Results (Section 2.0), Process Description and Operation
(Section 3.0), Sampling Locations (Section 4.0), Sampling and Analytical
Procedures (Section 5.0) and Quality Assurance/Quality Control (Section 6.0).
The supporting data and calculations for the results presented in
Volume I are included in Volumes II and III. Volume II contains summaries of
all the emissions data, sample calculations, CEM one-minute averages, GEM
stripcharts, CEM calibration data, field data sheets and laboratory reports.
Volume III contains equipment calibration data, correspondence, combustor and
control device operating data, field test logs, quality assurance information,
sampling and analytical protocols, and a list of the project participants.
1-11
-------
2.0 SUMMARY OF RESULTS
The results of the emission test program conducted at the Marion County
Facility from September 22-30, 1986 are summarized in this section. In
addition to the presentation of the results, variabilities and outliers in the
data are discussed. Combustor operating abnormalities are also discussed in
relation to the results.
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
concentrations that were measured. .For the reader's ease, an
English-to-Metric conversion table is included in Section 8.0.
The results are normalized to an equivalent CO,- concentration basis to
allow comparison of the results on a standard basis. The EPA/MWC data set is
also normalized to 12 percent C0_, since many state regulations are based on a
12 percent CO- basis. The 12 percent CO- basis is appropriate for the MSW
source category because the ultimate analysis of refuse on a combustible
12
fraction basis has a reasonably constant carbon content.
The supporting data for the results presented in this section are
included in the appendices of this report. The equations used to calculate
the results are also included in the appendices.
2-1
-------
2.1 CDD/CDF EMISSIONS RESULTS
The uncontrolled CDD/CDF results are summarized in Table 2-1. However,
these results must be used as an estimate only. The internal standard
recoveries of the front half fractions of the flue gas samples were extremely
low which have limited the level of confidence of the CDD/CDF data set.
However, the results from Run 2 are considered by Triangle Laboratories to be
usable. Therefore, the Run 2 results are used in this section to estimate the
CDD/CDF emissions, 2378-toxic equivalencies and congener distributions.
For Run 2, the total CDD concentration was 26.5 ng/dscm @ 12 percent CC>2
and the total CDF concentration was 44.3 ng/dscm at 12 percent CO^. For the
CDDs, 23.5 ng/dscm @ 12 percent C0? were measured in the front half and
2.88 ng/dscm @ 12 percent CO- were measured in the back half for Run 2. For
the CDFs, 37.1 ng/dscm @ 12 percent C0_ were measured in the front half and
7.03 ng/dscm @ 12 percent CO, were measured in the back half for Run 2.
The reported CDD/CDF results have been adjusted for blanks and internal
standard recoveries. The results are reported on a dry basis, at standard
conditions (68 F and 1 atm) and are adjusted for excess air dilution (12% CO-
basis).
The results for each congener are summarized in Table 2-2. In this
table, Runs 1 and 3 include only the back half fraction results while Run 2
includes both front and back half fractions. For Run 2, the average
concentration of 2378-TCDD and 2378-TCDF was 0.41 ng/dscm @ 12% CO and
5.15 ng/dscm @ 12% CO respectively.
The CDD/CDF results are presented for the front half and back half
fractions in Table 2-3. The results show that the majority of the CDD/CDFs
were captured in the front half (probe rinse and filter) of the train. For
Run 2, 85 percent of the CDD/CDFs were collected in the front half of the
train.
2-2
-------
TABLE 2-1. SUMMARY OF UNCONTROLLED CDD/CDF EMISSIONS FOR THE MARION COUNTY MWC
Run No.
Sampling Parameters
Volume gas sampled (dscf)
Flue gas flow rate (dscfm)
Flue gas temperature ( F)
Percent moisture by volume
Percent 1sok1net1c
C0_ (percent by volume , dry.)
0_ (percent by volume, dry)
Process operations
Steam load (103 Ib/br)
Refuse Feed Rate (1(T Ib/hr)
D1ox1n Results6
Total CDD (ng/dscm)
Total CDD (corrected
to 12% C02, ng/dscm)
Furan Results6
Total CDF (ng/dscm)
Total CDF (corrected
to 1258 CO., ng/dscm)
D1ox1n-Furan Results9
Total CDD-CDF (ng/dscm)
Total CDD-CDF (corrected
to 12% C02, ng/dscm)
Emission Factors (ng/kg refuse)
Total CDD
Total CDF
Total CDD-CDF
Run 1
(9-22-86)
147
29,200
403
16.3
104.5
10.3
9.0
64
20.3
Front Back
Hal f Hal f Total
c 3.29 d
c 3.83 d
c 4.96 d
c 5.77 d
c 8.24 d
c 9.61 d
^—
—
~™
Run 2
(9-23-86)
144
29,100
395
17.9
103.0
9.8
9.8
65
25.3
Front Back
Hal f Hal f Total
19.3 2.36 21.7
23.5 2.88 26.5
30.4 5.76 36.1
37.1 7.03 44.3
49.7 8.13 57.8
60.6 9.92 70.8
93
156
249
Run 3
(9-24-86)
133
28,000
384
16.1
99.0
10.0
9.6
65
20.7
Front Back
Half Half
c 1.43
c 1.71
c 2.99
c 3.59
c 4.42
c 5.31
—
-—
Total
d
d
d
d
d
d
aEngl 1sh-to-metr1c conversion factors are Included 1n Section 8.0.
These values are averages of data taken over the sampling period. Samples were collected according
to EPA Method 3 with Orsat analysis.
°The front half results are not reportable due to extremely low Internal standard recoveries.
Total values are not calculated because the front half fraction results are not reported.
6CDD/CDF results are adjusted for Internal standard recoveries and sample blank results.
2-3
-------
TABLE 2-2. UNCONTROLLED CDD/CDF EMISSIONS AT
MARION COUNTY MWC
CONCENTRATION3'13'0
(Normalized to 12% C02)
ISOMER
Run
Run 2
Run 3C
DIOXINS
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hepta-CDD
Oc ta-CDD
Total CDDs
[0
[0
[0
[0
.0003 1
0.15
0 .06
0 .04
0 .07
.0034]
1 .10
.0034]
.0031 ]
0 .69
0 .83
0 .48
0 .41
[0 .0057 ]
0 .08
0 .11
0 .41
0 .57
[0 .048]
0 .
[0 .06
0 .
[0 .01
9.
7 .
7 .
49
1]
03
2]
04
86
94
[0.
[0
[0
[0
[0
001
0 .
0 .
0 .
0 .
.01
0 .
.01
.01
.01
0 .
0 .
0.
3]
10
10
04
00
2]
38
4]
3]
5]
32
32
45
3 .83
26 .53
1 .71
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Oc ta-CDF
0.08
[2.18]
1 .69
0 .59
0 .54
[0 .0017 ]
0 .45
1 .65
[0 .0017 ]
[0 .0017 ]
[0 .0017 I
0 .71
[0 .0006 ]
0 .05
1 .82
0.55
26 .12
5.15
4.65
[0 .024]
2 .52
1 .61
[0 .031 ]
[0 .033 ]
[0 .040]
1 .76
[0 .070]
0 .06
0 .05
[1 .43]
1 .88
0 .46
0 .37
[0 .0060]
0 .29
0 .35
[0 .0070]
[0 .0073 ]
[0 .0089]
0 .20
[0 .0041 ]
tO .0060]
Total CDFs
5 .77
44.25
3 .59
Tot. CDD+CDF
9.61
70 .78
5.31
C02 norm. ratios
1 .17
1 .22
1 .20
a CDD/CDF results are adjusted for internal standard
recoveries and sample blank results.
b Standard conditions are 68 F and 1 atm.
c Values in brackets are minimum detection limits (MDL) of
not-detectable congeners. Each not-detectable result has
a unique MDL that is a function of the sample matrix
and specific congener analyzed.
d Runs 1 and 3 include results from back half fractions
only. The front half results are not reportable due
to extremely low internal standard recoveries.
e C02 norm, ratios = normalization ratios for adjusting
concentration data to 12% C02. Ratio = 12/C02
measured.
2-U
-------
TABLE 2-3. UNCONTROLLED CDD/CDF EMISSIONS FOR MARION COUNTY MWC
i
vn
ISOMER
DIOXINS
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hep ta-CDD
Oc ta-CDD
Total CDDs
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Oc t a-CDF
Total CDFs
Tot. CDD+CDF
CONCENTRATION
Run lb Run lc Run 2
Front Back Total Front
Half Half Half
[0 .0002]
0.12
0 .05
0 .04
0 .06
[0 .0029]
0 .94
[0 .0029]
[0 .0026 ]
0 .59
0 .71
0 .41
0 .35
3 .29
0 .07
[1 .86]
1 .45
0 .51
0 .47
[0 .0014]
0.39
1 .42
[0. 0014]
[0 .0014]
[0 .0014]
0 .61
[0 .0005]
0 .05
4.96
8.24
[0 .0039]
[0 .012]
0 .00
0 .28
to.
to.
to.
to.
to.
1
[17
1
to.
to.
to.
to.
to.
to
0 .45
026]
053]
038]
036]
084]
6 .54
5 .99
6 .05
9.30
1 .40
.13]
9.29
3 .64
3 .27
013 ]
1 .69
0.15
019]
020]
025]
0 .93
056]
.33]
30 .4
49.7
(ng/dscm, as measured)
Run 2 Run 3b
Backd Total Front
Half Half
to.
to
to
to
to
to
to
to
to
0006 ]
0 .07
0 .09
0 .05
0 .01
.082]
0 .40
.010 ]
0 .03
.011 ]
0 .85
0 .43
0 .43
2 .36
0.09
0 .45
2.04
0 .57
0.53
.005]
0 .37
1 .16
.005]
.005]
.007 ]
0.51
.007 ]
0 .05
5 .76
8.13
[0 .0045]
0 .07
0 .09
0 .33
0 .46
[0 .108]
0 .40
[0 .048]
0 .03
[0 .095]
7 .38
6 .42
6 .48
21 .67
1 .48
0.45
21 .33
4.21
3 .80
[0 .018]
2 .06
1 .31
[0.024]
[0 .025]
[0 .032]
1 .44
[0 .063 ]
0 .05
36 .1
57 .8
[0.
[0
[0
[0
[0
Run 3C
Back Total
Half
0011 ]
0 .08
0 .08
0 .04
0 .00
.010]
0 .32
.011 1
.011 ]
.013 ]
0 .27
0 .27
0 .38
1 .43
0 .04
[1.19]
[0.
[0.
[0.
[0.
[0.
[0.
1 .57
0 .38
0.30
0050]
0 .24
0.29
0058]
0061 ]
0074]
0 .17
0034]
0050]
2 .99
4 .42
aValues in brackets are minimum detection limits (MDL) of not-detectable congeners.
Each not-detectable result has a unique MDL that is a function of the sample matrix
and specific congener analyzed.
Due to extremely low internal standard recoveries, results are not reported for the
front half fractions of Runs 1 and 3.
""Totals are not calculated for runs without front half data.
Run 2 back half fraction is the average of duplicate analyses.
-------
In order to quantify the internal standards recoveries of the front and
back half fractions of the CDD/CDF train, the front half and back half
fractions were analyzed separately. The front half of the sampling train
includes the probe liner rinse and filter, and the back half includes the
condenser rinse, XAD trap and impingers. The front half results are not
reportable for Runs 1 and 3 due to extremely low internal standard recoveries.
Matrix effects of the front half fractions are believed to have caused the low
recoveries. The front half results of Run 2 were determined to be usable by
Triangle Laboratories.
2.1.1 2378-CDD Toxic Equivalency for Flue Gas Results
The CDD/CDF results for Run 2 are expressed in terms of 2378-TCDD
toxicity equivalents normalized to 12 percent C0« in Table 2-4. Each congener
13
has a 2378-TCDD toxicity equivalency factor (also presented in Table 2-4)
that ranks the toxicity of the isomer relative to the toxicity of 2378-TCDD.
The equivalency factors were developed by EPA. In terms of 2378-TCDD
equivalents, the average CDD toxic equivalent concentration was 0.43 ng/dscm
and the average CDF toxic equivalent concentration was 0.76 ng/dscm. The
average total CDD/CDF 2378-TCDD toxic equivalent concentration was
1.20 ng/dscm.
2.1.2 Cyclone Ash CDD/CDF Results
The CDD/CDF results for the cyclone ash samples are summarized in
Table 2-5. The average total CDD concentration was 3.58 ng/g and the average
total CDF concentration was 1.22 ng/g. A nanogram per gram (ng/g) is
equivalent to parts per billion (ppb) by weight. The cyclone ash results are
consistent between runs considering the analytical accuracy of the CDD/CDF
method (+ 50 percent).
The CDD/CDF results as 2378-TCDD equivalents are also included in
Table 2-5. The average 2378-TCDD equivalent concentration was 0.08 ng/g.
2-6
-------
TABLE 2-4
UNCONTROLLED CDD/CDF CONCENTRATIONS
EXPRESSED AS 2378-TCDD TOXIC EQUIVALENTS
FOR MARION COUNTY MWC
ISOMER
DIOXINS
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hepta-CDD
Oc ta-CDD
FURANS
2378-TCDD
Equ iv .
Factors
CONCENTRATION3
(ng/g, normalized to 12% C02)
Run 2 Run 2 Toxic
Actual Equiv.
0 .0000
0 .0000
0 .0000
1 .0000
0 .0100
0 .5000
0 .0050
0 .0400
0 .0400
0 .0400
0 .0004
0 .0010
0 .0000
[0 .0057 ]
0 .08
0 .11
0.41
0 .57
.048]
0 .49
.061 ]
0 .03
.012]
9.04
7 .86
7 .94
[0
[0
[0
0 .00
0 .00
0 .00
0.41
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .01
0 .00
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hep ta-CDF
Oc ta-CDF
0 .0000
0 .0000
0 .0000
0 .1000
0 .0010
0 .1000
0 .1000
0 .0010
0 .0100
0 .0100
0 .0100
0 .0001
0 .0010
0 .0000
Total
CDD+CDF
Cone en .
1 .82
0 .55
26 .12
5.15
4.65
[0 .024]
2.52
1 .61
[0 .031 ]
[0 .033 ]
[0 .040]
1 .76
[0.070]
0 .06
Total
Toxic
.70.78 Equiv.
0 .00
0 .00
0 .00
0 .51
0 .00
0 .00
0 .25
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
1 .20
Values in brackets are minimum detection limits (MDL)
of not-detectable congeners. Each not-detectable result has
a unique MDL that is a function of the sample matrix and
specific congener analyzed.
2-7
-------
TABLE 2-5. CDD/CDF CONCENTRATIONS AND
ASH FROM THE MARION COUNTY
2378-TCDD EQUIVALENTS FOR THE CYCLONE
MWC
ro
i
03
Concentration (ng/g=ppb)
ISOMER
===============
DIOXINS
Mono-CDD
Di-CDD
Tr i-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hep t a-CDD
Oc ta-CDD
Total CDDs
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Octa-CDF
Total CDFs
Tot. CDD+CDF
Run 1
========
(0 .001)
0 .02
0 .02
0 .02
0 .05
0 .04
0 .46
(0 .002)
0 .08
0 .20
0 .85
0 .82
0 .46
3 .02
(0 .001)
(0 .333)
0 .35
0.15
0.12
0 .02
0 .08
0 .10
(0 .001)
(0 .001)
(0 .001)
0 .07
(0 .001)
0 .05
0.94
3 .96
Values in parentheses
are c on s id er ed
Each not-detec
and specific c
Run 2
= 1= = = = = = = = = = =
(0.001) (
0 .01
0 .04
0 .02
0.15
0 .05
0 .53
0 .05
0 .13
(0.001) (
1 .35
1 .14
0 .81
4.28
(0 .001)
(0.130) (
0 .34
0 .16
0.18
0 .03
0 .08
0 .11
(0 .001)
(0.001) (
0 .01 (
0 .10
0 .02
0 .08
1 .11
5 .39
are minimum
as zeroes in calcu
t ab 1 e re
ong ene r
suit has a
ana ly zed .
Run 3
= = = = = = = =
0 .001)
0 .04
0.09
0 .02
0 .22
0 .04
0 .37
0 .04
0 .10
0 .001)
1 .20
0 .83
0 .49
3 .44
0 .02
0 .444)
0.69
0.19
0 .33
0 .04
0 .07
0 .06
0 .03
0 .002)
0.001)
0 .06
0 .06
0 .05
1 .60
5 .04
detect
la t ing
un i qu e
Ave rage
=========
0 .00
0 .02
0 .05
0 .02
0.14
0 .04
0 .45
0 .03
0.10
0 .07
1 .13
0 .93
0.59
3 .58
0 .01
0 .00
0 .46
0.17
0 .21
0 .03
0.08
0 .09
0 .01
0 .00
0 .00
0 .08
0 .03
0 .06
1 .22
4.80
ion 1 imi
ave rages
MDL that
Tox ic
Equ i v .
Factor
==============
0 .0000
0 .0000
0 .0000
1 .0000
0 .0100
0 .5000
0 .0050
0 .0400
0 .0400
0 .0400
0 .0004
0 .0010
0 .0000
0 .0000
0 .0000
0 .0000
0 .1000
0 .0010
0 .1000
0 .1000
0 .0010
0 .0100
0 .0100
0 .0100
0 .0001
0 .0010
0 .0000
Total
2378-TCDD
Equiva lent
ts (MDL) of no
and t ox ic equ
is a f unc t ion
2378-TCDD Toxic
Concentration (
Run 1
= = = = = = =
0 .00
0 .00
0 .00
0 .02
0 .00
0 .02
0 .00
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .02
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .08
t-de t ec
iva lenc
of the
Run 2
==========
0 .00
0 .00
0 .00
0 .02
0 .00
0 .03
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .02
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .09
t ab le c ong
ie s .
s amp 1 e ma
Equivalent
ng/g=ppb)
Run 3
=======
0 .00
0 .00
0 .00
0.02
0 .00
0 .02
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .02
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0.08
ene r B .
t r ix
Ave rage
=========
0 .00
0 .00
0 .00
0 .02
0 .00
0 .02
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .02
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0.08
These
-------
2.1.3 Baghouse Ash CDD/CDF Results
The CDD/CDF results of the baghouse ash samples are summarized in
Table 2-6. These results are reported as measured and are not adjusted for
dilution by lime and Tesisorb which are expected to contain no CDD/CDFs. The
average CDD concentration (as measured) was 4.56 ng/g and the average CDF
concentration (as measured) was 3.38 ng/g. The baghouse ash results were
consistent between runs considering the analytical accuracy of the CDD/CDF
method (+ 50 percent).
The CDD/CDF results as 2378-TCDD equivalents are also included in
Table 2-6. The average 2378-TCDD equivalent concentration was 0.14 ng/g.
2.1.4 CDD/CDF Congener Distributions
The distributions of the CDD and CDF congeners in the uncontrolled flue
gas, cyclone ash and baghouse ash on a mole fraction basis are presented in
graphical form in Figures 2-1 and 2-2. Only Run 2 results are shown for the
uncontrolled flue gas, due to the low internal standard recoveries reported
for Runs 1 and 3. The distributions are presented in tabular form in
Tables 2-7, 2-8, and 2-9.
A visual comparison of the uncontrolled flue gas, cyclone ash, and
baghouse ash congener distributions indicate similar distributions for the
three process streams. For the CDD homologues, hexa-CDD and hepta-CDD were
the most prevalent at about 35 mole percent and 25 mole percent, respectively
For the CDF homologues, tri-CDF was the most prevalent at about 50 mole
percent.
2-9
-------
IX)
I
o
TABLE 2-6. MARION COUNTY MWC CDD/CDF BAGHOUSE ASH CONCENTRATION AND
2378-TCDD TOXIC EQUIVALENCY
ISOMER
D I OX INS
Mono-CDD
Di-CDD
Tr i-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hep ta-CDD
Oc ta-CDD
Tot a 1 CDDs
FURANS
Mono-CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Oc ta-CDF
Total CDFs
Concentration
Run 1 Run 2
(0 .001 )
(0 .001)
0 .26
0 .03
0 .34
0 .07
0 .94
0 .07
0.12
(0 .002)
1 .52
1 .05
0 .58
4.98
0 .04
(0 .470)
1 .20
0 .59
0 .57
0 .08
0 .22
0 .29
(0 .001)
(0 .001)
(0 .001)
0.19
0 .03
0 .05
3 .26
(0 .001 )
0.09
0.12
(0 .001)
0 .20
0 .06
0.72
0 .07
0.15
(0 .003)
1 .81
1 .45
0 .99
5 .66
0.01
0.79
1 .49
0.62
0 .58
0 .06
0 .23
0 .23
(0 .001)
(0 .001)
(0 .048)
0 .23
(0 .001)
0.11
4.35
( ng/g =ppb )
Run 3 Ave r ag e
(0 .001)
0 .04
0.11
0 .02
0 .20
0 .04
0 .34
0 .03
0.07
0.14
0.74
0.77
0 .56
3 .04
(0 .001 )
(1.317)
1 .18
0 .38
0 .45
0 .07
0.12
0.17
0.01
(0 .002)
(0 .017)
0 .06
0 .03
0 .06
2 .52
0 .00
0 .04
0.16
0 .02
0.25
0 .06
0 .67
0 .06
0.11
0.05
1 .36
1 .09
0 .71
4 .56
0.02
0 .26
1 .29
0 .53
0 .53
0 .07
0.19
0 .23
0 .00
0 .00
0 .00
0.16
0 .02
0 .07
3.38
Tox ic
Eq u i v .
Factor
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Tota
.0000
.0000
.0000
.0000
.0100
.5000
.0050
.0400
.0400
.0400
.0004
.0010
.0000
.0000
.0000
.0000
.1000
.0010
.1000
.1000
.0010
.0100
.0100
.0100
.0001
.0010
.0000
1
2378-TCDD Toxic
Concentration (
Run 1 Run 2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.00
.00
.00
.03
.00
.04
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.06
.00
.01
.02
.00
.00
.00
.00
.00
.00
.00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .03
0 .00
0 .00
0.01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .06
0 .00
0 .01
0 .02
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
Equivalent
^ng/g=PPb)
Run 3 Aver ag e
0 .00
0 .00
0 .00
0 .02
0 .00
0 .02
0 .00
0 .00
0 .00
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .04
0 .00
0 .01
0 .01
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0.02
0 .00
0 .03
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .05
0 .00
0 .01
0.02
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
0 .00
2378-TCDD
Tot. CDD+CDF
8.24
10 .01
5.55
7 .93
Equ iva lent
0
.17
0.14
0 .11
0.14
aValues in parentheses are minimum detection limits (MDL) of not-detectable congeners
are considered as zeroes in calculating averages and toxic equivalencies.
Each not-detectable result has a unique MDL that is a function of the sample matrix
and specific congener analyzed.
Run 3 is the average of duplicate analyses.
These
-------
0.34
0.32
03
0.28
026
0.24
0.22
0.2
0.18
0.16
0.14
0.12
0.1
O.OD
0 06
0.04
0.02
0
Uncontrolled Flue Gas
DIOXIN ISOUERS
Cyclone Ash
KEY
Code Congener
0.35 -
0.3 -
0.25 -
0.2 -
0.15 -
0 1 -
0.05 -
PT.J3 rTt-3 n-ra KT
A B C D E
/
-;
F G H 1
DIOXIN ISOMEWS
7
/
/
'
rn
/ .
a
- 7
Dioxins
A = Mono-CDD
B = Di-CDD
C = Tri-CDD
D = 2378-TCDD
E = Other TCDD
F = 12378 PCDD
G = Other PCDD
H = 123478 HxCDD
I = 123678 HxCDD
J = 123789 HxCDD
K = Other HxCDD
L = Heota-CDD
j K L u M = Octa-CDD
RUN |3
Baghouse Ash
B C
V7\ RUN f 1
XIN ISOMERS
EZ! RUN *3
Figure 2-1. CDD Congener Distribution at Marion County
2-11
-------
Uncontrolled Flue Gas
0.5 •
0 45
0.3
025
0 15
0 1
HOPQRsruvwxr
ruRAN isoueRs
r .' J HUN |2
Cyclone Ash
KEY
Code Congener
'_
H
N 0 F
'',
;
FJ7
' '
2 -
Q
a ;;
$ •;
^ /.
F
^
^
^ FT-,
1 prattfe
S T
1
HP. SK r0rfi
U V W X f Z M
N =
O =
P =
Q =
R =
C —
T =
U =
V =
W =
X =
Y =
=
AA =
Furans
Mono-CDF
Di-CDF
Tri-CDF
2378-TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Octa-CDF
E3 "UN 1'.
RUN |3
Baghouse Ash
FURAM ISOMERS
T"j HUH |2
RUN |3
Figure 2-2. CDF Congener Distribution at Marion County
2-12
-------
TABLE 2-7. UNCONTROLLED FLUE GAS CDD/CDF CONGENER
DISTRIBUTION AT MARION COUNTY MWC
MOLE FRACTION a'b
ISOMER Run #lc Run #2d Run #3C
MOL. WT.
DIOXINS
Mono-CDD 218 ND
Di-CDD 252 0.005
Tri-CDD 286 0.006
2378 TCDD 320 0.020
Other TCDD 320 0.027
12378 PCDD 354 ND
Other PCDD 354 0.026
123478 HxCDD 388 ND
123678 HxCDD 388 ND
123789 HxCDD 388 ND
Other HxCDD 388 0.358
Hepta-CDD 422 0.288
Octa-CDD 456 0.269
FURANS
Mono-CDF 202 0.056
Di-CDF 236 0.029
Tri-CDF 270 0.606
2378 TCDF 304 0.105
Other TCDF 304 0.095
12378 PCDF 338 ND
23478 PCDF 338 0.047
Other PCDF 338 0.030
123478 HxCDF 372 ND
123678 HxCDF 372 ND
123789 HxCDF 372 ND
Other HxCDF 372 0.031
Hepta-CDF 406 ND
Octa-CDF 440 0.001
a MOLE FRACTION is the mole fraction of each homologue
based on the mono- thru octa- homologues.
CDD fractions are based on total CDD and
CDF fractions are based on total CDF.
b Mole fractions are based on the total train results.
c Mole fractions are not reported for Runs 1 and 3
due to extremely low internal standard recoveries.
d ND = not detected.
2-13
-------
TABLE 2-8 . CYCLONE ASH CDD/CDF CONGENER DISTRIBUTION
AT MARION COUNTY MWC
MOLE FRACTIONa'b'C
ISOMER Run #1 Run #2 Run #3 MEAN
MOL. WT.
DI OX INS
Mono-CDD 218 ND ND ND 0.00
Di-CDD 252 0.010 0.004 0.018 0.01
Tri-CDD 286 0.009 0.013 0.035 0.02
2378 TCDD 320 0.008 0.006 0.007 0.01
Other TCDD 320 0.020 0.044 0.077 0.05
12378 PCDD 354 0.015 0.013 0.013 0.01
Other PCDD 354 0.170 0.139 0.118 0.14
123478 HxCDD 388 ND 0.012 0.012 0.01
123678 HxCDD 388 0.027 0.031 0.029 0.03
123789 HxCDD 388 0.067 ND ND 0.02
Other HxCDD 388 0.287 0.323 0.348 0.32
Hepta-CDD 422 0.254 0.251 0.222 0.24
Octa-CDD 456 0.132 0.165 0.121 0.14
FURANS
Mono-CDF 202 ND ND 0.018 0.01
Di-CDF 236 ND ND ND 0.00
Tri-CDF 270 0.421 0.354 0.476 0.42
2378 TCDF 304 0.160 0.148 0.116 0.14
Other TCDF 304 0.128 0.167 0.202 0.17
12378 PCDF 338 0.019 0.025 0.022 0.02
23478 PCDF 338 0.077 0.067 0.039 0.06
Other PCDF 338 0.096 0.092 0.033 0.07
123478 HxCDF 372 ND ND 0.015 0.01
123678 HxCDF 372 ND ND ND 0.00
123789 HxCDF 372 ND 0.008 ND 0.00
Other HxCDF 372 0.061 0.076 0.030 0.06
Hepta-CDF 406 ND 0.014 0.028 0.01
Octa-CDF 440 0.037 0.051 0.021 0.04
a MOLE FRACTION is the mole fraction of each homologue
based on the mono- thru octa- homologues.
CDD fractions are based on total CDD and
CDF fractions are based on total CDF.
b Mole fractions are based on the total train results.
c ND = not detected.
2-lU
-------
TABLE 2-9 . BAGHOUSE ASH CDD/CDF CONGENER DISTRIBUTION
AT MARION COUNTY MWC
MOLE FRACTIONSa'b'c
ISOMER Run #1 Run #2 Run #3 MEAN
MOL. WT.
DIOXINS
Mono-CDD 218 ND ND ND 0.00
Di-CDD 252 ND 0.025 0.015 0.01
Tri-CDD 286 0.069 0.029 0.039 0.05
2378 TCDD 320 0.007 ND 0.008 0.00
Other TCDD 320 0.081 0.043 0.074 0.07
12378 PCDD 354 0.015 0.012 0.014 0.01
Other PCDD 354 0.203 0.141 0.126 0.16
123478 HxCDD 388 0.014 0.012 0.013 0.01
123678 HxCDD 388 0.024 0.027 0.026 0.03
123789 HxCDD 388 ND ND 0.048 0.02
Other HxCDD 388 0.299 0.323 0.250 0.29
Hepta-CDD 422 0.190 0.238 0.227 0.22
Octa-CDD 456 0.097 0.150 0.161 0.14
FURANS
Mono-CDF 202 0.018 0.003 ND 0.01
Di-CDF 236 ND 0.219 ND 0.07
Tri-CDF 270 0.408 0.361 0.472 0.41
2378 TCDF 304 0.178 0.134 0.162 0.16
Other TCDF 304 0.172 0.125 0.188 0.16
12378 PCDF 338 0.022 0.012 0.027 0.02
23478 PCDF 338 0.060 0.045 0.046 0.05
Other PCDF 338 0.079 0.045 0.038 0.05
123478 HxCDF 372 ND ND 0.003 0.00
123678 HxCDF 372 ND ND ND 0.00
123789 HxCDF 372 ND ND ND 0.00
Other HxCDF 372 0.047 0.040 0.038 0.04
Hepta-CDF 406 0.007 ND 0.010 0.01
Octa-CDF 440 0.010 0.016 0.015 0.01
a MOLE FRACTION is the mole fraction of each homologue
based on the mono- thru octa- homologues.
CDD fractions are based on total CDD and
CDF fractions are based on total CDF.
b Mole fractions are based on the total train results.
c ND = not detected.
2-15
-------
2.2 PARTICULATE RESULTS
The uncontrolled flue gas particulate loading results are summarized in
Table 2-10. The average uncontrolled particulate loading was
0.8807 grains/dscf @ 12% CO with a range of 0.74 1.12 gr/dscf at 12% C02.
The particulate loading was measured for six runs which were four hours long
for Runs 1 3 and six hours long for Runs 4-6. The results have been adjusted
for blanks, and are reported as concentrations and mass emission rates.
14
All of the six data points passed the Dixon criterion for rejecting
extreme observations in either direction (outliers) and are therefore
considered valid data points (see appendices for calculations). The
particulate loading results have a relative standard deviation of
approximately 20 percent, which is considered a normal variation of the system
at normal operating conditions.
2.3 SPECIFIC METALS RESULTS
The uncontrolled flue gas was sampled to determine the concentrations of
the following metals: lead, cadmium, total chromium, hexavalent chromium and
nickel. The results are summarized in Table 2-11. The most prevalent metal
was lead at an average concentration of 20,000 ug/dscm at 12 percent C0_.
Cadmium was the second most prevalent metal at an average concentration of
1,090 ug/dscm at 12 percent CO.. The average total chromium concentration was
412 ug/dscm at 12 percent CO., and the average nickel concentration was
12.1 ug/dscm at 12 percent C0_ . Hexavalent chromium was not detected in any
of the three runs. The minimum detection limit for hexavalent chromium in the
flue gas was determined to be 8.0 ug/dscm at 12% CO (see appendices for
calculation). The results have been adjusted using the laboratory proof blank
for background metals and are reported on a dry basis.
The lead and cadmium results were very consistent with percent relative
standard deviations of less than 20 percent which is within the analytical
error of the method. At present, the methods for chromium and nickel are
2-16
-------
TABLE 2-10. SUMMARY OF UNCONTROLLED PARTICULATE EMISSIONS FOR THE MARION COUNTY MWC
IX)
i
Run No.
Date
Sampling Parameters3
Volume gas sampled (dscf)
Flue gas flow rate (dscfm)
Flue gas temperature (°F)
Moisture (percent by volume)
Isoklnetlcs (percent)
£®2 (percent by volume, dry)
®2 (percent by volume, dry)
Process operations
Steam load (103 Ib/hr)
Refuse Feedrate (103 Ib/hr)
Paniculate Results'?
Front Half Catch
(Probe, cyclone, and filter)
mg - mass
gr/dscf
gr/dscf (corrected to 12% C0_)
mg/dscm
mg/dscm (corrected to 12% C0?)
Ibs/hr
Kg/hr
Run 1
9-22-86
143
29,700
404
16.8
100.0
10.3
9.0
64
20.3
8303.2
0.8956
1.048
2049
2397
228
103
Run 2
9-23-86
147
30,000
399
18.5
101.6
9.80
9.8
65
25.3
8739.2
0.9160
1.118
2096
2557
236
107
Run 3
9-24-86
132
27,600
388
16.6
99.3
10.0
9.6
65
20.7
5287.1
0.6168
0.7402
1412
1694
146
66.3
Run 4
9-26-86
190
27,400
390
17.1
98.8
9.95
9.6
66
21.4
8630.0
0.7009
0.8481
1604
1941
164
74.6
Run 5
9-29-86
200
28,300
399
18.1
97.8
10.0
9.6
65
24.9
8553.4
0.6588
0.7906
1507
1808
160
72.6
Run 6
9-30-86
207
29,200
398
17.4
98.2
10.0
9.6
66
23.5
8270.1
0.6162
0.7394
1410
1692
154
69.9
Average
—
28,700
396
17.4
—
10.0
9.5
65
22.7
0.7341
0.8807
1680
2015
181
82.2
aEngl 1sh-to-metr1c conversion factors are Included 1n Section 8.0.
Standard conditions are 68°F (20°C) and 1 atm (1.01325 x 10S Pa)
Results are adjusted for blanks.
-------
TABLE 2-11. SUMMARY OF UNCONTROLLED SPECIFIC METALS FOR THE MARION COUNTY MWC
rv>
i
oo
Run No. Run 4 Run 5 Run 6
Date 09-26-86 09-29-86 09-30-86
Pb/Cd Cr/Ni Pb/Cd Cr/N1 Pb/Cd Cr/N1
Type Emissions train train train train train train
Sampling Parameters3
Volume gas sampled (dscf) 190 213 200 213 207 214
Flue gas flow rate (dscfm) 27,400 29,000 28,300 29,200 29,200 29,100
Flue gas temperature (°F) 390 396 399 404 398 401
Moisture (percent by volume) 17.1 17.6 18.1 18.4 17.4 17.3
Isok1net1cs (percent) 98.8 101.6 97.8 101.0 98.2 101.6
C02 (percent by volume, dry) 9.95 10.0 10.0
C*2 (percent by volume, dry) 9.6 9.6 9.6
Process Operations
Steam load (103 Ib/hr) 66 65 66
Refuse Feed Rate (103 Ib/hr) 21.4 24.9 23.5
Specific Metals Resultsb'c
(ug/dscm corrected to 12% C02)
Lead 16,400 19,400 24,200
Cadmium 1,150 1,180 950
Chromium (+VI)d ND ND ND
Total chromium 453 262 520
Nickel 16.5 4.49 15.3
Average
—
28,700
398
17.7
—
10.0
9.6
66
23.3
20,000
1,090
ND
412
12.1
Engl1sh-to-metr1c conversion factors are included in Section 8.0.
Standard conditions are 68°F (20°C) and 1 atm (1.01325 x 105 Pa)
Adjusted for blank results.
"Total train results.
A
ND = not detected. The minimum detection limit for Cr (+VI) In the flue gas was 8 ug/dscm at 12% CO
-------
state-of-the-art and have not been validated. Thus, precision has not been
determined. Considering that no criterion has been established, the total
chromium and nickel results are considered to be reasonably consistent with
percent relative standard deviations of 30 and 50 percent, respectively.
The uncontrolled mass emission rates and concentrations (not normalized
to 12 percent C0?) for these metals are summarized in Table 2-12. The mass
emission rates of lead, cadmium, total chromium and nickel were 1.77, 0.096,
0.037 and 0.0011 Ib/hr, respectively.
2.3.1 Specific Metals-To-Particulate Ratios
The ratio of the uncontrolled flue gas metals-to-particulate
concentrations are summarized in Table 2-13. The ratios are calculated by
dividing the metal concentration (ug/dscm) determined from total train results
by the particulate loading (mg/dscm) from the front half of the train.
The ratios were on the same order of magnitude for the three runs. For
lead, cadmium, total chromium, and nickel, the average ratio of metals to
particulate was 11.2, 0.602, 0.229 and 0.0067 mg/g, respectively.
Chromium (+VI) was not detected.
2.3.2 Metals Results for the Baghouse Ash and Cyclone Ash
The baghouse ash and cyclone ash were analyzed by three methods: NAA, AA
and colorimetry. First, the ashes were extracted and analyzed for hexavalent
chromium by colorimetry and total chromium by AA. However, the hexavalent
chromium results are not reported due to very low spike recoveries which are
believed to be caused by complexing with compounds in the matrix. Complete
details on these analyses are included in Appendix F.3.
Then, aliquots of the hexavalent and total chromium extracts were
submitted for analysis by NAA. The additional metals obtained include
arsenic, cadmium and nickel.
2-19
-------
TABLE 2-12. UNCONTROLLED SPECIFIC METALS MASS EMISSION RATES FOR THE MARION COUNTY MWCa'b
(Unadjusted)
Element
Lead
Cadmium
Chromium (+VI)e
Total Chromium
Nickel
Run
ug/dscm
13,600
952
ND
376
13.7
4
lb/hrd
1.39
0.098
0
0.041
0.0015
Run
ug/dscm
16,200
984
ND
218
3.7
5
lb/hrd
1.72
0.104
0
0.024
0.0004
Run
ug/dscm
20,100
788
ND
433
12.8
6
lb/hrd
2.20
0.086
0
0.047
0.0014
Average
ug/dscm Ib/hr
16,600
908
ND
342
10.1
1.77
0.096
0
0.037
0.0011
iv) Adjusted for laboratory proof blanks.
o Standard conditions are defined as 68°F and 1 atm.
GTotal train results.
d
Mass emission rate (Ib/hr) is the product of the concentration (ug/dscm) and the volumetric flow
rate of the flue gas (dscmm) and a conversion factor (1.32 x 10~7).
Q
ND = not detected: Minimum detection limit in the flue gas for Cr(+VI) was 6.7 ng/dscm
(unadjusted to 12 percent CO ).
-------
TABLE 2-13. RATIO OF UNCONTROLLED METALS TO PARTICULATE
MASS FOR THE MARION COUNTY MWC
Metal
Lead
Cadmium
Chromium (+VI)
Total Chromium
Nickel
mg metal/per gram of particulate
Run 4 Run 5 Run 6 Average
8.47 10.7 14.3
0.594 0.653 0.559
ND° ND° ND°
0.234 0.145 0.307
0.0085 0.0025 0.009
11.2
0.602
ND°
0.229
0.0067
Ratios are calculated with total train results by AA for the metals
and front half train results for the particulates.
The ratio (mg/g) is calculated by dividing the concentration
(ug/dscm) by the particulate loading (mg/dscm).
Q
ND = not detected. The minimum detection limit was 0.004 mg metal
per gram of particulate for chromium (+VI).
2-21
-------
The results for the cyclone ash and baghouse ash are presented in
Table 2-14. In the cyclone ash, hexavalent chromium by NAA was not detected
with a minimum detection limit of 0.00160 mg/g. The average total chromium
result was 0.174 mg/g by AA and 0.186 mg/g by NAA. The two results had a
percent difference of 7 percent. Arsenic, cadmium and nickel were not
detected by NAA in either the hexavalent or total chromium extracts. The
minimum detection limit for the cyclone ash was on the order of 0.02 mg/g for
arsenic, 0.05 mg/g for cadmium, and 0.1 mg/g for nickel.
For the baghouse ash, hexavalent chromium by NAA was not detected with a
minimum detection limit of 0.0020 mg/g. The average total chromium result was
0.0185 mg/g by AA and 0.0213 mg/g by NAA. The two results had a percent
difference of 14 percent.
The average arsenic concentration in the baghouse ash was 0.0629 mg/g
based on the total chromium extract analyzed by NAA and was not detected
except for Run 5 in the hexavalent extract analyzed by NAA. The arsenic
concentration was measured at 0.181 mg/g for Run 5.
Cadmium and nickel were not detected by NAA in either the hexavalent or
total chromium extracts. The minimum detection limit for the baghouse ash was
on the order of 0.08 mg/g for cadmium and 0.04 mg/g for nickel. The baghouse
results are not adjusted for lime and Tesisorb dilution.
2.3.3 Metals Results for the Lime Slurry and Tesisorb
The lime slurry and Tesisorb were analyzed by NAA. The samples analyzed
were composited over Runs 4 to 6. The results for total chromium, arsenic,
cadmium and nickel are reported in Table 2-15.
For the lime slurry, total chromium was measured at 0.00172 mg/g and
arsenic at 0.000323 mg/g. Cadmium and nickel were not detected with minimum
detection limits of 0.00282 mg/g for cadmium and 0.00180 mg/g for nickel.
2-22
-------
TABLE 2-14. METALS RESULTS FOR BAGHOUSE AND CYCLONE ASH FROM THE MARION COUNTY MWC
ro
i
u>
Metals Concentration (mg/g)a
Run 4
Ash
Cyclone Ash
Hexavalent Chromium
Total Chromium
Arsenic
Cadmium
Nickel
Baghouse Ash
Hexavalent Chromium
Total Chromium
Arsenic
Cadmium
Nickel
AA
0.234 0
CO
[0
[0
0.0183 0
0
0
[0
NAA
A
.280
.00467]
.0248]
.0910]
.0193
.0871
.123
.0718]
B
0.00152
CO. 0130]
CO. 06061
[0.02bbj
CO. 00193]
CO. 0125]
CO. 0717]
CO. 0280]
Run 5
AA
0.151 0
CO
CO
0
0.0177 0
0
CO
CO
NAA
A
.126
.0202]
.0524]
.125
.0188
.0632
.0391]
.0837]
B
CO. 00142]
CO. 0140]
CO. 0676]
CO. 0223]
0.002200
0.181
CO. 136]
CO. 0180]
AA
0.138 0
0
0
CO
0.0194 0
0
CO
0
Run 6
NAA
A
.152
.0551
.0389
.105]
.0260
.0385
.0381]
.0135
B
CO. 00185]
0.0407
CO. 0634]
CO. 0292]
CO. 00189]
CO. 0167]
CO. 0722]
CO. 0248]
Average
AA NAA
A
0.174 0.186
CO. 0267]
CO. 0387]
CO. 107]
0.0185 0.0213
0.0629
CO. 0667]
CO. 0563]
B
CO. 00160]
CO. 0226]
CO. 0639]
CO. 0260]
CO. 0020]
CO. 0700]
CO. 0933]
CO. 0236]
AA = Atom absorption analysis results. NAA - neutron activation analysis results.
A = Chromium (III) fraction of processed ash sample. B = chromium (VI) fraction of processed ash sample.
Total chromium results by AA Include chromium (III) only. NAA results Include hexavalent and trlvalent chromium.
°Brackets Indicate metal not detected. Detection limit given.
-------
TABLE 2-15. METALS RESULTS FOR LIME SLURRY AND TESISORB
Metal
Total Chromium
Arsenic
Cadmium
Nickel
Concentration
Lime Slurry
0.000172
0.000323
[0.000282]
[0.00180]
(mg/g)'1
Tesisorb
0.0475
[0.00414]b
[0.0214]
[0.150]
Kesults from NAA of composite sample. Samples composited over
Runs 4 to 6.
Not detected. Detection limit in brackets.
2-2U
-------
Total chromium was measured at 0.0475 mg/g in the Tesisorb. Arsenic,
cadmium and nickel were not detected with minimum detection limis of
0.00414 ng/g for cadmium, and 0.150 mg/g for nickel.
2.4 ACID GAS CONCENTRATIONS
The uncontrolled flue gas was analyzed for HC1 and SO-. HC1 was measured
using EPA Method 5 with modifications, which is a manual method, and SO.- was
measured using EPA Method 6C, which is a continuous monitoring method. Thus,
the variation of S0_ with time was also recorded.
2.4.1 SO, Results
z
A summary of the uncontrolled SO. concentration results are presented in
Table 2-16. Oxygen was also measured continuously in order to calculate the
SQj emission factors and to aid in leakchecking the GEM system. The 0_
results are also summarized in Table 2-16.
The average S0_ concentration was 155.5 ppmv. This average represents
the mean of the averages for the three test runs. The averages for the three
test runs are the mean of one-minute averages recorded during the flue gas
sampling period. The results have been adjusted for daily instrument drift
and for SO,- response quenching by CO- and 0-. The relationship for the
quenching adjustment is supplied by the instrument manufacturer.
The S09 and 0? concentrations are plotted versus time in Figures 2-3 and
2-4, respectively. The S00 concentration varied from about 50 to 200 ppmv as
a baseline with an occasional excursion to about 500 ppmv which lasted from
one to two hours. Since these large S0_ peaks did not have corresponding 0~
variations indicating dilution, they are considered to be caused by a change
in the composition of the refuse.
2-25
-------
TABLE 2-16.
UNCONTROLLED MARION COUNTY MWC SO RESULTS
Number of
Date Run Observations
9/22/86 1
9/23/86 2
9/24/86 3
Average 1,2,3
149
149
211
211
224
224
195
195
Parameter
S02 (ppmV)
0 (% vol)
S02 (ppmV)
0 (% vol)
S02 (ppmV)
0 (% vol)
S02 (ppmV)
0 (% vol)
a,b
Average
Value
96.1
9.1
157.1
8.8
213.4
9.1
155.5
9.0
Standard
Deviation
28.
0
139,
0
155,
0
107.
0,
.2
.8
.0
.7
.7
.7
.6
.7
Results are reported on a dry basis. The results have been adjusted for
daily instrument drift and SO- quenching effect.
i ^
Test run averages are the mean of one-minute averages. The data aquisition
system scans each channel 180 time per minute and saves a one-minute average
on floppy disk.
2-26
-------
Run 1
14,4 14 8 15,2 156 16 164
24 HOUR CLOCK
Run 2
155 16.5
24 HOUR CLOCK
Run 3
12
24 HOUR CLOCK
Mean = 96.1 ppmv
Std Dev = 28.2 ppmv
Obs* = 149
Mean = 157.1 ppmv
Std Dev = 139.0 ppmv
Obs* = 211
Mean = 213.4 ppmv
Std Dev = 155.7 ppmv
Obs* = 224
* Obs = Number of one-minute
average observations
Figure 2 - 3. Uncontrolled S02 Concentration History at Marion County
2-27
-------
132 136
Run 1
4 148 15.2
24 HOUR CLOCK
15.6 16
Run 2
15.5 16.5
24 HOUR CLOCK
Run 3
24 HOUR CLOCK
Mean = 9.1 % vol
Std Dev = 0.8 % vol
Obs* = 149
Mean = 8.8 % vol
Std Dev = 0.7 % vol
Obs* = 211
Mean = 9.1 % vol
Std Dev = 0.7 % vol
Obs* = 224
* Obs = Number of one-minute
average observations
Figure 2-4. O2 Concentration History at Marion County
2-28
-------
2.4.2 HCl (as total chlorides) Results
The HCl concentration in the flue gas is based on the amount of chloride
ions measured in the back half of the sampling train. The method assumes that
all the chloride ions in the back half fraction are from the disassociation of
HCl molecules. The chloride concentration is also measured for the front half
fraction, but these chloride ions are assumed to be from salts collected on
the filter. The chloride concentration for both the front half and back half
fractions are reported on a dry basis as mg/dscm in Table 2-17.
To calculate the HCl concentration on a parts per million basis by volume
(ppmv) the mass concentration (mg/dscm) of the back half fraction is converted
to ppmv using the molecular weight (see Appendix A.5 for actual calculation).
The HCl concentration on a ppmv basis is also presented in Table 2-18.
The average concentration of HCl as total chlorides was 489 ppmv. The
concentrations ranged from 471 to 512 ppmv dry. An average of 721 mg/dscm of
total chlorides were measured in the back half fraction and an average of 133
mg/dscm were measured in the front half fraction.
2.4.3 Molar Ratios of Lime to HCl and SO
The molar ratios of actual lime used to HCl and 0^ concentrations in the
uncontrolled flue gas are presented in Table 2-18. The calculation is based
on the lime feed rate and the molar flow rates of S02 and HCl into the quench
reactor and is included in Appendix A.5.
The average lime-to-HCl ratio was 3.54 and the average lime-to-HCl and
SO. ratio was 2.20. The reported design molar ratio of lime-to-HCl was 2.5.
2-29
-------
TABLE 2-17. UNCONTROLLED FLUE GAS CHLORIDE
CONCENTRATION AT MARION COUNTY MWCa
Chloride Concentration
(mg/dscm. dry)
Run Front Half
1 138
2 156
3 104
Average 133
Back Half
695
713
756
721
Concentration of HC1
as chloride (ppmv, dry)
471
483
512
489
Results are corrected for the laboratory proof blank.
The front half fraction is considered to be salts and is not
included in the HC1 result. All of the chlorides measured in the
back half fraction are assumed to be from HC1 collected from the
flue gas.
2-30
-------
TABLE 2-18. MOLAR RATIO OF ACTUAL LIME TO
STOICHIOMETRIC LIME FOR HC1 AND
Run
1
2
3
Average
Actual Lime
Stoichiometric
Lime for HC1
3.63
3.53
3.46
3.54
Actual Lime
Stoichiometric Lime
for HC1 and SO-
2.58
2.14
1.89
2.20
Q
Based on average CaO feedrate, average SO- and HC1 concentration,
average flue gas flow rate, and Stoichiometric requirements of one
(1) mole CaO per two (2) moles HC1 and one (1) mole CaO per one (1)
mole SO .
2-31
-------
3.0 PROCESS DESCRIPTION AND OPERATION
3.1 PROCESS DESCRIPTION
Ogden Martin operates two mass burn waterwall combustor 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.
Figure 3-1 presents a cross-section of Marion County process line.
Refuse is trucked to the facility and dumped into an enclosed receiving pit.
It is subsequently transferred to each combustor by overhead cranes. The
solid waste passes downward through the feed chute and is pushed onto the
stoker grate by a hydraulically operated ram feeder. The system is designed
to operate at 90 percent excess air. Underfire air is supplied via five air
plenums and controlled by the pressure drop across the grate bars. Overfire
combustion air, 25 to 30 percent of the total air, is injected through rows of
nozzles above the stoker at the front and rear walls of the combustor at
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/hr (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
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
3-1
-------
To Atmosphere
I
ro
Furnace
Boiler Superheater Economizer
Quench Reactor/ Teslsorb
Acid Gas Feed
Scrubber Hopper
Lime Slurry Dry
Mixing Tank Venturi
Quench
Pit
Grizzly
Scalper
Distributor
I.D. Fan stack
Figure 3-1. Marion County Process Line
-------
of 4520 kP (655 psig) and temperature of 370°C (700°F). The steam is
delivered to a 13.1 megawatt (45 Btu/hr) turbine generator. The electricity
produced flows into the Portland General Electric Company (PGE) grid.
Bottom ash and grate sittings are discharged into a water quenched
residue system. The ash disposal system includes 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
the baghouse and cyclone is collected separately and conveyed to the ash
removal system to be handled and disposed together with the bottom ash.
3.2 AIR POLLUTION CONTROL SYSTEM
The air pollution control system at the Marion County Solid
Waste-to-Energy Facility consists of a quench reactor, a dry venturi scrubber,
and baghouse. The flue gases leave the boiler economizer section at
temperatures between 225°C to 270°C (440°F to 515°F) and enter the bottom of
the dry scrubber vessel through a cyclonic inlet where removal of oversize
3
particles takes place. Gas flow rates vary between 1740 m /min (61,440 acfm)
at 225°C (440°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 quench water feed rate is approximately
3
0.05 to 0.07 m /min (12.8 to 18.2 gpm) based on the inlet flue gas
temperature. A constant operating temperature of 124 -149 C (256 - 300 F) is
maintained at the outlet of the quench reactor. The lime concentration in the
slurry is adjusted to compensate for changes in the acid gas concentration and
the dry lime feed rate will vary between 57 and 193 kg/hr (125-425 Ib/hr).
The stoichiometric ratio of lime/HCl is maintained at 2.5 to ensure that upset
peaks are sufficiently controlled.
A low pressure drop dry venturi is located between the quench reactor and
the baghouse. Tesisorb is injected into the venturi at a rate of 24 kg/hr
(53 Ib/hr).
3-3
-------
TM
An Amerthem fabric filter is installed downstream of the dry venturi
for particulate matter 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 unspent reagent
in the filter cake also acts as a further neutralization mechanism for acid
gas collection.
PM emissions including ODEQ condensibles are required to be controlled to
a level of 69 mg/dscm (0.03 gr/dscf) at 12% C0?. Design data for the air
pollution control system are listed in Table 3-1.
3.3 COMBUSTOR AND AIR POLLUTION CONTROL SYSTEM OPERATING CONDITIONS
Combustor and air pollution control system operating data were monitored
during the sampling periods by plant personnel. The following combustor
process parameters were recorded every 15 minutes:steam load; steam drum
pressure; steam temperature; middle of furnace temperature, first pass; top of
furnace temperature, first pass; economizer (econ) outlet temperature; primary
air temperature and flow; secondary air pressure (front, upper and lower rear
walls); I.D. fan inlet temperature; and stack opacity. Table 3-2 presents the
average values of the recorded process data during testing.
The following air pollution control system operating parameters were
recorded every 15 minutes: quench reactor inlet and outlet pressures and
outlet temperature; lime feed rate; slurry inlet and outlet velocity; dry
venturi differential pressure; Tesisorb feed rate; and baghouse differential
pressure, outlet temperature and cleaning cycle. Table 3-3 lists the average
process data for the air pollution control system recorded during testing
runs. Appendix G contains copies of the combustor and air pollution control
system operating data.
No combustor operation problems were experienced during testing. Several
combustion chamber temperature excursions were observed due to the variability
of the raw waste feed. However, these excursions were within the range of
-------
TABLE 3-1. AIR POLLUTION CONTROL SYSTEM DESIGN SPECIFICATIONS
Item
Design Specifications
Effective Removal
Particulate Matter
Outlet Emissions, mg/dscm
HC1, percent removal
SO., percent removal
Quench Reactor
Quench Water, m /min
Lime feed rate, kg/hr
Flue gas flow rate
3
Inlet, m /min
HC1, inlet, ppmv, dry
SO.., inlet, ppmv, dry
Particulates, inlet,
mg/scm @ 20°C
Outlet temperature, C
69
90
70
.05 .07
57 (normal), 193 (max.)
1740 @ 225°C
1885 @ 270°C
234 (normal), 700 (max.)
130 (normal), 385 (max.)
2.2 2.5
125
Dry Venturi
Tesisorb feed rate, kg/hr
24
Fabric Filter
Cleaning System
A/C Ratio
Reverse air
1.69:1 (gross)
2.31:1 (net)
3-5
-------
TABLE 3-2. AVERAGE PROCESS DATA FOR THE MARION COUNTY MWC
Parameter
Refuse Feed Rate, (1Q3 Ib/hr)
Steam Load, 10 Ib/hr
Steam Drum Pressure, psig
o
Steam Temperature, F
Middle of Furnace Temperature
1st pass, F
Top of Furnace Temperature
1st pass, F
o
Econ. Outlet Temperature, F
Primary Air
Temperature,
Flow, Ib/hr x 1Q3
Secondary Air
Pressure, in. W.C.
Combustor Front
Combustor Upper Rear
Combustor Lower Rear
I.D. Fan Inlet Temperature, F
Stack Opacity, X
Run 1
9/22/86
20. 3
64
620
702
1715
1567
403
299
69
14.6
0.0
13.4
255
0.0
Run 2
9/23/86
25.3
65
601
702
1726
1563
392
301
65
10.9
0.3
11
254
0.2
Run 3
9/24/86
20.7
65
600
699
1782
1607
380
305
53
11.3
0.3
10.8
255
0.0
Run 4
9/26/86
21.4
66
605
702
1759
1604
387
307
70
11.7
0.2
10.7
256
0.1
Run 5
9/29/86
24 .9
65
600
701
1706
1583
398
294
72
10.8
0.0
10.1
257
0.0
Run 6
9/30/86
23.5
66
610
701
1729
1607
397
291
71
11.5
0.1
10.3
258
0.0
Average
22.7
65
606
701
1736
1589
393
300
67
11.8
0.2
11.0
256
0.1
-------
TABLE 3-3. AVERAGE DRY SCRUBBER/QUENCH REACTOR OPERATING DATA FOR THE MARION COUNTY MWC
Quench Reactor
U)
— i
Run
1
2
3
4
5
6
Inlet
Pressure
Date In W.C.
09/22/86 -3.0
09/23/86 -3.0
09/24/86 -2.6
09/26/86 -3.2
09/29/86 -3.0
09/30/86 -3.0
Average -3,0
Outlet
Pressure
in W.C.
-5.5
-5.0
-4.9
-5.1
-5.2
-5.2
-5.2
Outlet
Temp.
°F
270
272
272
271
272
272
272
Lime Feed
Rate
Ib/hr
220
220
215
215
225
225
220
Dry
Diff.
Pressure
In w.c.
1.3
1.2
1.0
1.0
1.2
1.1
1.1
Venturl
Testsorb
Feed Rate
Ib/hr
26.5
26.5
26.5
26.5
26.5
26.5
26.5
Baghouse
Dlff. Outlet
Pressure Temp.
In w . c . F
5.4 258
5.2 255
4.6 260
5.0 260
5.8 260
6.1 260
5.4 260
Cleaning
Cycle
Mln.
60
60
75
60
60
60
60
-------
normal combustor operations. On 09/26/86, the Unit Number 1 baghouse was
by-passed at approximately 10:45 a.m. for less than 10 minutes. A delivery
truck had tapped into the facility's compressed air lines thus reducing the
air pressure in the control system instrument panel. As a result, the
baghouses were bypassed. This event did not affect testing at the inlet to
the control system.
-------
4.0 SAMPLING LOCATIONS
The sampling locations are shown on the process line schematic in
Figure 4-1. Each sampling location is discussed in the following sections.
4.1 BOILER OUTLET SAMPLING LOCATION
The parameters that were measured at the boiler outlet sampling location
included CDD/CDF, S0?, HC1, particulate, lead, cadmium, chromium, and nickel.
A side view and top view of the boiler outlet sampling location are shown in
Figures 4-2 and 4-3, respectively. The sampling location had three six-inch
I.D. ports located in a circular duct 6'10" in diameter. Two of the ports
(Ports A and B) were located in the same plane, 90 apart. These ports were
used for the manual test methods to obtain a full traverse. The third port
(Port C) was located about four feet downstream on a different axis. This
port was used to extract a fixed point sample for the continuous emission
monitors. All the ports had eight inch long nipples and were 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 were located approximately 4 duct
diameters (28'6") downstream of a 90 bend in the duct and 1.9 duct diameters
(13'1") upstream of a 90° bend in the duct. Following EPA Method 1
procedures, 24 traverse points were required. The traverse point location
diagram is presented in Figure 4-4.
A cyclonic flow check of the location was conducted according to EPA
Method 1. The average degree of rotation was 5 . EPA Method 1 specifies that
the average degree of rotation should be equal to or less than 10 .
The average volumetric flowrate of the duct was 29,000 dscfm at an
average temperature of 400°F. The velocity head reading from the pitot tube
ranged from 0.07 to 0.2 in H^O which is in the low range for the manometers
U-l
-------
Furnace
Boiler Outlet Tesisorb
Sampling Location Quench Reactor/ Tesisorb Sampling
\ ™.S?. .I8.6.!. Location
Boiler Superheater Economizer
To Atmosphere
Acid Gas
Scrubber Hopper
Quench
Pit
Grizzly
Scalper
7-1 Lime Slurry \ /\ /\
-^Mixing Tank If I I 1
Lime Slurry
y—Distributor Baghouse Ash
Sampling Location
I
Outlet Stack
Sampling Location
I.D. Fan stack
tr
o
h-
OO
Figure 4-1. Marion County Process Line with Sampling Locations
-------
-^> Boiler Outlet —>
28'-6"
-• EL39'-10"
Direction of
Gas Flow
Port
GEM
Port"
Port
4'
JL
Platform
To Cyclone and
'Acid Gas Scrubber
Side View
-------
Flue Gas
From Economizer
Sampling
Platform
Cyclone and Acid
Gas Scrubber
Top View
To
Baghouse
to
o
Figure 4 - 3. Top View of Boiler Outlet Sampling Location at Marion County
H-U
-------
Port A
PortB
—T- 8" nipple
oc
I
o
f"-
Measurement from the outside of the nipple for probe marking
b Traverse points are located as specified in EPA Method 1
Figure 4 - 4. Traverse Point Location Diagram for Boiler Outlet Location at
Marion County
-------
that are standard equipment in Radian meter boxes. A more sensitive manometer
for this range was used. Static pressure draft at this point in the system
averaged -2.0 inches of H.O during the test program.
4.2 BAGHOUSE ASH AND CYCLONE ASH SAMPLING LOCATIONS
The sampling locations for the Unit No. 1 baghouse ash and cyclone ash
are shown in Figure 4-5. 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.
4.3 LIME SLURRY SAMPLING LOCATION
The lime slurry samples were collected from the recycle hose on the lime
slurry mixing tank. The mixing tank was accessible from the second floor of
the area housing the lime slurry injection system.
4.4 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 4-6.
-------
Flue Gas
to I.D. Fan
Baghouse
\ /\
3uench Reactor/
Acid Gas
Scrubber
Incinerator
Building
Flue Gas from Economizer
X- >
'
'
Screw Conveyor
Screw
Conveyor
Baghouse Ash
Sampling Location
Cyclone Ash
Sampling Location
Hole in Plate
Pivot Point
Handle
Bottom View of Ash
Sampling Locations
cc
o
o
r^
8
Figure 4 - 5. Baghouse Ash and Cyclone Ash Sampling Locations at
Marion County
-------
Tesisorb from
Main Hopper
Feed Hopper
for Unit #2
Feed Hopper
for Unit #1
Tesisorb Sampling
Port
Lift off cover
and scoop out
Tesisorb
Figure 4 - 6. Tesisorb Sampling Location at Marion County
U-8
-------
5.0 SAMPLING AND ANALYTICAL PROCEDURES
The sampling and analytical procedures used for the Marion County test
program were the most recent revisions of the published methods. In some
cases, the methods were modified to incorporate the most recent developments
which have been accepted by the sampling community. In this section, brief
descriptions of each sampling and analytical method are provided. The
detailed sampling and analytical methods with modifications are included in
Appendix J. A summary of the sampling and analytical methods used for each
parameter is presented in Table 5-1.
5.1 CDD/CDF SAMPLING AND ANALYSIS
CDD/CDF sampling followed the December 1984 draft protocol for the
determination of chlorinated organic compounds in stack emissions which is
included in Appendix J.3. The protocol was developed by the Environmental
Standards workshop sponsored by the American Society of Mechanical Engineers
(ASME) and EPA. The method is based on EPA Reference Method 5.
The CDD/CDF analysis followed the ASME/EPA analytical procedures to assay
stack effluent samples and residual combustion products for CDDs and CDFs,
also dated December 1984 which is included in Appendix J.3. The sampling and
analytical procedures with modifications are discussed in the following
sections.
The samples were analyzed separately as front half and back half
fractions. This was a modification made by Radian at the request of EPA.
The purpose of this modification was to verify the internal standards recovery
of the front and back half fractions and not to divide filterable and gaseous
fractions. The different matrices of the fractions may effect the recovery of
the internal standards.
5-1
-------
TABLE 5-1. SAMPLING METHODS AND ANALYTICAL PROCEDURES
USED FOR THE MARION COUNTY TEST PROGRAM
Parameters
Sampling Method
Analytical Method
CDD/CDF
HC1/PM
Pb/Cd/PM
SO,
Baghouse ash and
Cyclone ash
Tesisorb and Lime
Slurry
Molecular weight
Moisture
Velocity
Temperature
ASME/EPA Environmental
Standards Workshop
(Dec. 1984)
EPA Method 5
with modifications
Draft EPA Protocol for
hexavalent and total
chromium (Feb. 22, 1985)
Draft EPA protocol for
cadmium
EPA Method 6C
Composite of grab sample
Grab samples
EPA Method 3
EPA Method 4
Method 2
Type K thermocouple
High resolution GC/MS
HC1: Ion chromatography
PM: Gravimetric
Cr(+VI): Colorimetry
Cr(+III): AA
Ni: AA
Pb and Cd: AA
PM: gravimetric
Non-dispersive, infrared
continuous analyzer
CDD/CDF: ASME/EPA protocol
Metals:
Hexavalent and total
chromium by colorimetry
and AA, respectively
Metals: Neutron Activation
Analysis
Orsat apparatus
Gravimetric
Method 2
5-2
-------
5.1.1 Equipment and Sampling Preparation
In addition to the standard EPA Method 5 requirements, the CDD/CDF
sampling method included several unique preparation steps which ensured that
the sampling train components were not contaminated with organics that may
interfere with analysis. The glassware, glass fiber filters and XAD resins
were cleaned and checked for residuals before being packed.
The protocol was modified by replacing hexane with methylene chloride
during all aspects of sampling preparation and recovery. Methylene chloride
was shown to recover the higher chlorinated CDD/CDFs better than hexane during
the EPA National Dioxin Strategy Tier 4 program.
The glassware was washed in soapy water, rinsed, baked and then rinsed
with acetone and methylene chloride. Once in the field, the train glassware
was assembled and rinsed with methylene chloride. The rinses were analyzed
for residual CDD/CDFs to generate the laboratory proof blank results.
The XAD resin and glass fiber filters were extracted in HPLC grade water,
methyl alcohol, methylene chloride and hexane, sequentially. At the
conclusion of the soxhlet extractions, a 500 ml aliquot of the final solvent
rinse (hexane) was concentrated to 5 ml. A 500 ml aliquot of fresh hexane of
the same lot was also concentrated as a blank. Both aliquots were analyzed by
GC/FID for determination of Total Chromatographable Organics (TCO). A
standard mixture of n-hydrocarbons that cover the TCO range of boiling points
which is from 100°C (C - bp 98°C) to 300°C (C^ - bp 303°C) was used to
quantify the TCO results in the aliquot. The results were reported as
milligrams of TCO per resin weight prepared (mg/g), corrected for the blank.
The results for the XAD resin used at Marion County are included in
Appendix 1.2. The XAD resin was packed in glass traps and the filters in
glass petri dishes for transport to the field.
The remaining preparation included calibration and leakchecking of all
the train equipment. This included meterboxes, thermocouples, nozzles, pitot
tubes, and umbilicals.
5-3
-------
5.1.2 Sampling Operations
The CDD/CDF sampling method used the sampling train shown in Figure 5-1.
Radian has modified the protocol configuration to include a horizontal
condenser rather than a vertical condenser. The horizontal condenser lowers
the profile of the train and reduces breakage. The XAD trap following the
condenser is still maintained in a vertical position.
The CDD/CDF trains were leakchecked at the start and finish of sampling
as required in EPA Method 5 as well as during the port change. In the event
that the leakrate was found to be above the minimum acceptable rate (0.02
ft /min) upon removal from a port, the sample volume was corrected for that
interval as specified in the sampling protocol. This calculation is shown in
more detail in Appendix A.5, but basically, the excessive leakrate was reduced
by the minimum acceptable rate and then multiplied by the sampling time of the
port.
At the boiler outlet, the duct was traversed following EPA Method 1. The
traverse diagram was shown previously in Figure 4-4. Samples were collected
for 10 minutes per point for a total of 240 minutes. Pertinent data were
recorded every five minutes. The flowrate of the flue gases and the nozzle
diameter required the sampling rate to be an average of 0.59 dscfm to insure
isokinetic sampling. Sampling was conducted simultaneously with the outlet
stack.
Other sampling operations that were unique to CDD/CDF sampling include
maintaining the gas temperature entering the XAD trap below 68 F. The gas was
cooled by the condenser and, the XAD trap, itself, had a water jacket in which
ice water was circulated. Otherwise, sampling followed EPA Method 5
specifications.
5.1.3 Sample Recovery
To facilitate transfer from the sampling location to the recovery
trailer, the sampling train was disassembled into four sections: the probe
-------
Thermocouple
•S" Type Pltot ^
/ Filter Holder
Thermocouple
Probe
vn
I
• Thermocouple Thermocouple
/ Check Valve
Stack Wall /
Pltot
Manometer
Reclrculatlon Pump
XAD • 2 Trap
Heated Zone
If
Silica Gel
(300 grams)
Water Knockout 100ml H PLC Water
Implnger
, ^/^\s~\ By-Pass
Thermocouples -^*( ) ( ) valve
Main Valve
Air-Tight
Pump
Vacuum Line
Figure 5-1. CDD/CDF Sampling Train Configuration Used at Marion County
-------
liner, the XAD trap and condenser, filter holder, and the impingers in their
bucket. Each of these sections were capped with methylene chloride-rinsed
aluminum foil before removal to the recovery trailer. Once in the trailer,
field recovery followed the scheme shown in Figure 5-2. The liquid fractions
were recovered into amber bottles that served to protect the samples from
light degradation. Field recovery resulted in the sample components listed in
Table 5-2. The samples were shipped as these components to the analytical
laboratory by truck.
5.1.4 Analysis
The CDD/CDF analyses for the Marion County samples were performed by
Triangle Laboratories, Research Triangle Park, North Carolina. The sample
components were extracted, combined and concentrated according to the scheme
shown in Figure 5-3. Isotopically-labelled internal standards and surrogate
compounds were added to the samples before the extraction process began. The
internal standards included 2378-13C -TCDD, 13C -PCDD, 13C -HpCDD,
13 13
C -HxCDD and C -OCDD. Once added to the samples, the internal standards
went through the entire extraction process and were measured on the GC/MS.
The recoveries of the internal standards were determined and the results of
the native species were adjusted according to the internal standard
recoveries. The results contained in Triangle Laboratories analytical report
are already adjusted for internal standard recoveries. (See Table 6-3,
13 37
Section 6.) The surrogate compounds which included C^-TCDF, Cl. -TCDD,
13
and C -HxCDF were added in a similar manner, but the surrogate recoveries
were not used to adjust the results of the native species. Internal standards
and surrogates are discussed in more detail in Section 6.
The samples were analyzed separately as front half and back half
fractions. This was a modification made by Radian at the request of EPA.
The purpose of this modification was to study the internal standards recovery
of the front and back half fractions and not to divide filterable and gaseous
5-6
-------
VJ1
I
—J
Filter Support
and Back Half
Front Half of Filter
Probe Liner Cyclone Filter Housing Filter Housing
1
1 \ \ \ \
Attach 250ml Brush and Brush and Carefully Rinse with
flask lo ball rinse with rinse with remove filter acetone
joint acetone until acetone from support (3x)
participate Is (3 x) with tweezers
Rinse with removed Rinse with
acetone, Rinse with Brush loose methylene
empty flask _ . meth
Into 950ml 3^3 "?h chlo
hnttlo J uniea wun ,-
bollle methylene <3
Brush liner chlorlde
and rinse with
3allquots
of acetone
Check liner
to see If
partlculate Is
removed, If
not repeat 3
Rinse with
3allquot8"
of methylene
chlo
ride
ylene partlculate chlo
ride onto filter (3
x)
Seal In petrl
dish
ride
x)
5th Implnger
Condenser X AD Trap 1st Implnger 2nd Implnger 3rd Implnger 4th Implnger (silica gel)
I 1 (knockout) 1 1 1 1
Rinse with Secure XAD * Weigh Weigh Weigh Weigh
acetone trap openings Weigh Implnger Implnger Implnger Implnger
(3x) with glass Implnger
balls and Empty Empty Empty Discard
Rinse with clamps Empty contents Into contents Into contents Into silica gel
methylene contents Into bottle bollle botlle
chic
0
ride Place In bottle
x) cooler for Rinse with Rinse with Rinse with
storage Rinse with acetone acetone acetone
acetone (3X) <3x) (3x)
13 x»
Rinse with Rinse with Rinse with
Rinse with melhylene methylene methylene
methylene chlorlde chlorlde chlorlde
chlorlde (3x) (3 x ) (3 x )
(3x1
PR
CR
SM
CD
IR
* Aliquot equals about 100ml. Empty all rinses into one 950ml bottle.
3 x = three times
Figure 5-2. CDD/CDF Field Recovery Scheme Used at Marion County
-------
TABLE 5-2. CDD/CDF SAMPLING TRAIN COMPONENTS FOR MARION
COUNTY SHIPPED TO ANALYTICAL LABORATORY
Container/Component Code
Glassware
1. Component Number 1 F
2. Component Number 2 PR
3. Component Number 3 CR
Filter(s)
o
Rinses of nozzle, probe, transfer line (if
used), cylcone, and front half of filter
holder
Rinses of back half of filter holder, filter
support and condenser
4. Component Number 4 CD First impinger contents and rinses
5. Component Number 5 IR
6. Component Number 6 SM XAD-2 resin
Second, third and fourth impinger
contents and rinses
R
a,,.
Rinses include acetone and methylene chloride combined in same container.
-------
PR
Acetone/MeCI, flrnses of
Probe Liner. Nozzle. Fronthalt
ol Filter Housing
IR
2nd. 3rd. 4tn impmger
Contents Acetone/MeCi, Rinses
CD
SM
CR
1st Impinger Contents
and
AcBtone/MeCl, Rinses
Add Surrogates
shake with
Hexane;
Combine Hexane
Fractions
From this point the
same procedure is
'oilwed for each of
the 1mL extracts
regardless of origin.
Concentrate to
1 ml
QC
3
o
N.
S
•For large volumes, organic solutions are concentrated using a rotary evaporator apparatus
For small volumes, samples are concentrated under nitrogen
Figure 5-3. CDD/CDF Analytical Scheme Used at Marion County
5-9
-------
fractions. The different matrices of the fractions may effect the recovery of
the internal standards.
The front half of the sample train consisted of the probe rinse, cyclone,
front half filter holder rinse, and the filter. The back half included the
j^
back half filter holder rinse, coil condenser rinse, XAD trap, and impinger
contents and rinses.
The samples were analyzed by high resolution gas chromatography, followed
by high resolution mass spectrometry (GC/MS). The instruments were calibrated
daily with external standards. The GC/MS results were recorded and stored
onto a computer file. The computer was used to reduce the data and to
summarize the results such as amount detected, detection limit, retention time
and internal and surrogate recoveries. The analytical report from Triangle
Laboratories is included in Appendix D.I.
5.1.5 Data Reduction for CDD/CDF Results
The data reduction for CDD/CDF results began with correcting for the
reagent blank results. Then the concentrations of CDD/CDFs in the flue gas
were calculated by dividing the nanograms of analyte by the volume of flue gas
collected. The volume of flue gas has been corrected to EPA standard
conditions and is reported on a dry standard cubic meter basis.
To normalize the concentrations to 12 percent C09, a ratio (12 * C09
observed) was calculated for each run based on EPA Method 3 results. Then the
concentrations from each run were multiplied by the respective normalization
ratio. The normalized concentrations were multiplied by 2378-TCDD toxic
equivalency factors to obtain 2378-TCDD toxic equivalents.
Mole fractions of each isomer were calculated by dividing the
concentration of each isomer by its molecular weight. The moles of each
isomer were summed for each run, and the fraction was obtained by dividing the
moles for each isomer by the total moles. All of the CDD/CDF calculations
5-10
-------
rniyr
described above were done on a Lotus 1-2-3 spreadsheet and were also
verified by hand calculation. The congener distribution plots (Figures 2-1
and 2-2) were also prepared using the spreadsheet. The calculations are
included in Appendix A.5.
5.2 HC1/PARTICULATE FLUE GAS SAMPLING AND ANALYSIS
HCl/particulate sampling followed the procedures specified in EPA
Reference Method 5. The combination HCl/particulate train was used in order
to reduce the number of separate tests required since the HC1 and particulate
results are not affected by using a combined train. The method was modified to
allow the collection of HC1 samples in the back half of the train. These
modifications included:
1) Replacing the water in the impingers with 0.1 N NaOH.
2) Pre-cleaning the glassware.
The sampling and analytical procedures with modifications are discussed
in the following sections.
5.2.1 Equipment and Sampling Preparation
Method 5 requires that the sample train glassware be cleaned with soap
and water and dried with acetone to minimize particulates on the glassware.
In addition, the glassware was rinsed with 0.1 N NaOH. The remaining
preparation included calibration and leakchecking of all the train equipment.
This included meterboxes, thermocouples, nozzles, pitot tubes and umbilicals.
5.2.2 Sampling Operations
The HCl/particulate samples were collected using the sampling train in
Figure 5-4. Sampling train data were recorded at 5 minute intervals. Train
leakrate determination, traversing, sampling time and sampling rate were the
same as described in Section 5.1.2 for the CDD/CDF trains.
5-11
-------
Filter Holder
Thermocouple
'S" Type Pltot
A
pie L
^—
A/
i nermocoupie — y—
<*Sk. ,-1 L-
Probe
^1
^i
L-
I
Stack Wall /
Pltot
^=
i
r
H-P—
i i
s
J
-
1
n
— —
.
j:
•*»
_
i
q
• — *^~~
.
\
t
**r
.
\
-(
~^ — »
En
J
/Check Valve
J^-l>^
Lr
-N^~^
i a
fpty
6^
- *
-^
Silica Gel \
" (300 grams)
-Ice Bath
Manometer Extra Condensate 100ml 0.1 N NaOH
Impinger
f~\s~\ BY-Pass
Thermocouples •**\) Q) Valve
orifice r:Or
Air-Tight
Pump
Vacuum Line
Figure 5-4. Particulate/HCI Sampling Train Configuration Used at Marion County
-------
5.2.3 Sample Recovery
To facilitate transfer from the sampling location to the recovery
trailer, the sampling train was disassembled into three sections: The probe
with liner, the filter holder, and the impingers in their bucket. Each
section was capped with aluminum foil before removal to the recovery trailer.
Once in the trailer, field recovery followed the scheme shown in the top half
of Figure 5-5. The sample fractions were recovered by first rinsing and
brushing the probe, cyclone and front half of filter housing with acetone.
Then, these components were recovered with 0.1 N NaOH. The back half of the
train (the impingers and back half of filter housing) were recovered
separately with 0.1 N NaOH. The filter was sealed in a petri dish. The NaOH
rinses were placed in high-density polyethylene sample bottles. The acetone
rinses were placed in amber borosilicate glass bottles.
Sample recovery resulted in the sample components listed in Table 5-3.
These samples were sent to the Radian analytical laboratories by truck.
5.2.4 Sample Analysis
Sample analysis followed the scheme shown in the bottom half of
Figure 5-5. The particulate analysis of the front half of the train was
performed at Radian's Field Testing Laboratory. When the particulate analyses
were complete, the particulates were re-suspended in the NaOH front half
fraction in preparation for HC1 analysis.
The front half and back half fractions were then submitted to Radian's
Inorganic Laboratory where they were analyzed for total chlorides. The
assumption is made that the chloride ions found in the back half of the train
resulted from gaseous HC1 in the flue gas. These results are used to
calculate the HC1 concentration. The front half fraction is also analyzed but
the chloride ions are considered to be from the various salts collected on the
filter. The samples were analyzed by ion chromatography. Each sample was
analyzed twice, and an average reported.
5-13
-------
-p-
| Front Half Sample Recovery Fractions |
Nozzle, Probe, Cyclone,
Brush/Rinse
Evaporate;
Dessicate; Weigh;
Partlculate (mg)
Dissolve residue
with 0.1 NNaOH
Extract with
0.1 NNaOH
Combine residue, 0.1 N NaOH Rinse, and extract
Filter Solution using Whatman
No. 40 filter media
Adjust solution to known volume
Analyze for total
Chlorides by Ion
Chromatography
Field Recovery
* Laboratory I
Analysis T
| Back Half Sample Recovery Fractions }
1st, 2nd, 3rd, 4th impingers
Contents and Rinses
O.tNNaOH Rinse
Adjust solution
to known volume
Analyze for total
Chlorides by Ion
Chromatography
DC
OJ
o
o
f-
co
Total Chloride (mg)
Total Chloride (mg)
Figure 5-5. Marion County Partculate/HCI Field Recovery and
Analytical Protocol
-------
TABLE 5-3. HC1/PARTICULATE SAMPLE COMPONENTS
COLLECTED AT THE MARION COUNTY MWC
Component Code Glassware
PR Acet Acetone rinses of probe liner, cyclone, front
half of filter housing
PR - NaOH 0.1 NaOH rinses of probe liner, cyclone, front
half of filter housing
- F Filter
- IR 0.1 NaOH rinses and contents of impingers and
back half of filter housing.
5-15
-------
5.2.5 Data Reduction
The particulate loading was calculated using the following equation.
(M -M, )_... + (M -M, ) .
C = p 13 filter p D rinses
P V
m(std)
C = particulate loading (mg/dscm)
M = mass of particulates (mg)
K = mass of the blank (mg)
V . ,. = volume of gas sampled at standard conditions
m(std) °
(dscm @ 1 atm and 68°F)
The HC1 concentrations were calculated based on the amount of the
chloride ions measured in the back half of the sampling train. The results
were corrected for the laboratory proof blank.
HC1 Concentration (mg/dscm)
M
C (mg/dscm) = c_ x mg
Vstd 100° Ug
where
M = amount of chloride ions (ug)
V = volume of gas sampled corrected to standard conditions (dscm)
HC1 Concentration (ppmv) as chloride
MW of Cl = 35.45
ppmv = C x 10' g x g mole x 0.024 dscm x 10
^ -
mg J5.45 g g mole
5-16
-------
5.3 LEAD/CADMIUM/PARTICULATE DETERMINATION
Gas sampling and analysis for lead, cadmium and particulates was
performed according to the draft EPA protocol for the determination of
cadmium which is based on EPA Reference Method 5.
5.3.1 Equipment and Sampling Preparation
Equipment and sampling preparation for Pb/Cd sampling followed the
specifications for EPA Method 5 with some additional glassware preparation.
The train glassware was washed in soapy water, rinsed with tap water, rinsed
with distilled water, rinsed with 0.1 N nitric acid, rinsed with distilled
water, and then dried with acetone. The glassware was then capped with
Parafilm . Before sampling, the laboratory proof blanks were collected by
recovering the glassware with 0.1 N nitric acid, and the rinses were analyzed.
5.3.2 Sampling Operations
The draft cadmium protocol is a variation of EPA Method 5 where 0.1 N
nitric acid replaces water in the impingers, and filter media with a low lead
background is used. The Pb/Cd/particulate sampling train used at Marion
County is shown in Figure 5-6. Due to the high moisture content of the flue
gas and the relatively long sampling period, an empty impinger was added as
the first impinger in the train to collect the extra volume of moisture.
Sampling was conducted for 5 minutes per traverse point. Each diameter
was traversed three times for a total of 360 minutes of sampling time. The
traverse points were previously shown in Figure 4-4. The flue gas was sampled
at an average rate of 0.56 dscfm to maintain isokinetic sampling. Sampling
was conducted simultaneously with the outlet stack. Train leakrate
determination followed the same procedure as described in Section 5.1.2 for
the CDD/CDF trains.
5-17
-------
vn
I
Co
Thermocouple
'S" Type Pitot ^
/ Filter Holder
Thermocouple —f-
^K.i-r
Probe
Thermocouple
/ Check Valve
Stack Wall /
Pltot
Manometer Extra Condensate 100ml 0.1N HNO3
Impinger
Thermocouples ""^Q Q Valve
Orifice
Silica Gel
(300 grams)
Main Valve
Air-Tight
Pump
Vacuum Line
Figure 5 - 6. Particulate/Pb/Cd Sampling Train Configuration used at
Marion County
-------
5.3.3. Sample Recovery
To facilitate transfer from the sampling location to the recovery
trailer, the sampling train was disassembled into three sections; the probe
assembly, the filter holder and the impingers in their bucket. Each section
D
was capped with Parafilm before being transferred to the recovery trailer.
Once in the trailer, the field recovery followed the scheme shown in the top
half of Figure 5-7. Basically, the sample fractions were recovered first as a
particulate train and then as a Pb/Cd train. The acetone rinses were placed
in borosilicate glass bottles and the nitric acid rinses were placed in
high-density polyethylene bottles.
The desiccation and evaporation of the particulate sample was performed
at Radian's Field testing laboratory. Therefore, in the field, the samples
were recovered into the sample components listed in Table 5-4. Although the
nozzle was recovered for particulate analysis, the rinse was kept separate,
and was used in particulate analysis only. The nozzle was not rinsed with
nitric acid and was not included in the Pb/Cd analyses because of possible
contamination.
5.3.4 Analysis
Once the particulate analyses were completed, the particulates were
resuspended in the front half nitric rinse. Then, the front half samples and
back half fractions were transferred to Radian's Inorganic Laboratory where
they were analyzed for lead and cadmium. The front half fractions were
digested in acid and Parr bombed before analysis by AA with an air acetylene
flame. Back half fractions (which were not Parr bombed) were digested in acid
and analyzed by AA. Aliquots of the digested samples were submitted for NAA
analysis.
5.3.5 Data Reduction
The particulate loading was calculated as described in Section 5.2.5.
Lead and cadmium concentrations in the flue gas were determined as ug/dscm
5-19
-------
Front Half Sample Recovery Fractions
Back Half Sample Recovery Fractions
1
Brush/Rinse Probe' Cvclone Brush/Rinse
1 1 1
Acetone No. 1 Acetone No.20.1N No.SAcelone
Rinse Rinse HNO, Rinse Discard
1 1
i \
Evaporate; Evaporate;
Dessicate, Weigh Desslcate; Weigh
1
Paniculate (mg) Partlculate(mg)
1
Dissolve Residue
In HNO, Rinse
V71
1
w
o
Filter
Desslcate;
Weigh
Partlcul
ale(mg)
Combine Residues with Filter
Concentrate and add HF
Parr Bomb Solution at
1 50 °C for 5 hours
I
Cool Solution, Transfer and
Dilute with 0.1N HNO,
Adjust to Known
Volume (50-100mL)
•Analyze for Cd and
other Metals by NAA
Cdljjg)
Adjust Aliquot for
AA Linearity Curve
1
Analyze for Cd by AA
Air Acetylene Flame
Analyze for Pb by AA
Air Acetylene Flame
1 1
Cd (j*g) Pb (MO)
1st, 2nd, 3rd, 4lh Impingers
Contents and Rinses
0.1NHNO,
Rinse
Heat/Evaporate
to Near Dryness
Dissolve Residue
with Cone. HNO,
Adjust to Known
Volume (50-100mL)
Adjust Aliquot for
AA Linearity Curve
h
Analyze for Pb by AA 1
Air Acetylene Flame |
5th Implnger
(Silica Gel)
1
Weigh and
Discard | PUM „ ory j
1 Laboratory i
" Analysis "
1
Analyze for Cd and
other Metals by NAA
Cd(pg)
Analyze for Cd by AA
Air Acetylene Flame
Cdf^g)
I
DC
CM
O
(^
CD
* NAA results are reported in a separate appendix.
Figure 5-7. Marion County Particulate/Pb/Cd Field Recovery and
Analytical Protocol
-------
TABLE 5-4. SAMPLE COMPONENTS FOR Pb/Cd/PARTICULATE
TRAIN USED AT THE MARION COUNTY MWC
Component Code Glassware
PR - Acet Acetone rinses of probe liner cyclone or cyclone
bypass, and front half of filter housing
PR - HNO_ 0.1 N Nitric rinses of probe liner, cyclone,
and front half of filter housing
F Filter
IR 0.1 N Nitric rinses and contents of impingers,
also rinses of back half of filter housing,
filter support, and zigzag connector.
5 NZ Acetone rinse of nozzle
a
A zigzag is the female-to-female glass connector that joins the exit of the
filter holder to the first impinger.
5-21
-------
with the following equation:
C ., = (C° metal - C. )
metal ~u
m(std)
where:
C° = amount of metal detected in sample (ug)
C, = amount of metal detected in the blank (ug)
b
V . = volume of gas sampled at standard
m(std) & o
conditions (dscm @ 1 atm, 68 F)
The metals concentrations were normalized using the same procedure
described for CDD/CDFs in Section 5.1.5.
5.4 HEXAVALENT AND TOTAL CHROMIUM AND NICKEL SAMPLING AND ANALYSIS
The chromium and nickel flue gas samples were collected according to the
draft EPA Protocol for Hexavalent and Total Chromium Emissions
(February 22, 1985). The draft protocol was modified by EMB. The
modifications are described in the following sections where appropriate.
5.4.1 Equipment and Sampling Preparation
Equipment preparation included the calibration and leakchecking of all
train equipment. This included meterboxes, thermocouples, nozzles, pitot
tubes and umbilicals.
The train glassware was cleaned according to the same procedure used for
the HC1 train. After soap and water washing, the glassware was rinsed with
NaOH and water followed by acetone to dry.
5.4.2 Sampling Operations
The Cr(+VI)/Cr(+III)/Nickel samples were collected at Marion County using
the sampling train shown in Figure 5-8. The sampling train was modified by
eliminating the glass fiber filter section. The filter section was excluded
5-22
-------
I
(V)
U)
Thermocouple
'S"TypePltot^
Filter Bypass Thermocouple
~h\ Glass Wool /
f I Plug f
Check Valve
Silica Gel
(300 grams)
100ml O.SNNaOH
Thermocouples
Orifice
Stack Wall /
Pltot
Manometer
Air-Tight
Pump
Vacuum Line
Figure 5 - 8. Cr/Ni Sampling Train Configuration used at Marion County
-------
in order to prevent the reduction of Cr(+VI) to Cr(+III) on the filter by
reducing agents in the flue gas. Particulate was collected in the impingers
which contained 0.5 N NaOH.
The sampling time, sampling rate, traversing and train leakrate
determination followed the same procedures described in Section 5.3.2 for the
Pb/Cd/particulate train.
5.4.3 Sample Recovery
Sample recovery in the field followed the procedure shown in the top half
of Figure 5-9. The front half of the train (probe liner and filter bypass)
was rinsed and brushed with distilled water to remove gross particulate, and
then rinsed with 200 ml of 0.1 N NaOH. The impingers were recovered with
distilled water. The nozzle was not recovered for metals analysis due to
possible contamination (see Figure 5-9).
When sample recovery was completed, the sample fractions were mixed and
adjusted with 10 N NaOH to a pH of 8. The alkaline matrix was intended to
prevent the reduction of the hexavalent chromium in the sample. The
alkalinity was checked until it stabilized and then the samples were placed in
high-density polyethylene containers for shipment to the analytical
laboratory.
5.4.4 Sample Analysis
The samples were analyzed according to the scheme shown in the bottom
half of the Figure 5-9. The samples were digested in an alkaline solution,
filtered, and then split into two aliquots. The particulate matter was
digested in acid and analyzed for trivalent chromium and nickel by AA. The
filtrate aliquot was analyzed for hexavalent chromium by diphenylcarbazide
colorimetry and for nickel by AA.
-------
Sample Recovery Fractions
4 Field Recovery 4
i Laboratory Analysis 1
Nozzle
Brush/Rinse
0.1 N NaOH
Rinse
Retain In Archive
\_n
I
ro
VJl
Remove Aliquot; Add Acid
to Convert to Acidic Matrix
Perform Acid Digestion
Analyze for Nl by AA
Air Acetylene Flame
T
Probe, Cyclone, Front Hall Brush/Rinse
1
No. 1 OINNaOH I
[ 200mL Max. Rinse |
No. 2 Acetone
rinse to dry
then discard
{Combine Rinse Solutions and |
Ad|usl to pH of 8 with 10N NaOH |
Heal/Evaporate to
Near Dryness
Combine 40mL 2% NaOH.
3% Na.CO, Solution
Heal w/constani stirring
for 30 minutes
Cool; Filter using 47mm
Teflon Filler Media
Collect Filtrate and Dilute
to 100ml Volume wiPlsl. H,O
Mix; Remove Allquol; Adjust
Volume to 80 ml w/H,O
Adjust pH to 2.0 with 10% H,SO,,
Dilute to IQOmLw/Dlsl H,O
Add Dlphenylcarbazlde Color
Reagent; Adjust Volume to
100mL with Delonlzed H,O
Allow 10 minutes for Color
Development; Analyze
Colorlmetrlcally for Cr"
Cr-lng)
Digest Paniculate Matter
on Filter w/3ml oM:1
H,SO, and 10ml cone. HF
Evaporate w/Water Bath;
Add Additional HF until
Filler Media Is Dissolved
tat/Evaporate to near
Dryness
1st, 2nd, 3rd, 4th Implngers
Contents and Rinses
I "
I Delonlzed. Distilled H,O I
I Rinse I
5th Implnger
(Silica Gel)
Weigh and discard
Combine Solution with
30mL9MH,SO,
Reflux for 4 Hours
at 150 'C
Cool; Filler usin
Teflon Filler Media
Dilute Filtrate lo 100mL
Volume
Analyze for Cr" by AA
Air Acetylene Flame
Remove Aliquot; Analyze
for Nl by AA Air
Acetylene Flame
*Remove Allquol;Analyze
for Cr"& Other
Metals by NAA
* NAA results are reported in a separate appendix.
oo
xt
Figure 5-9. Marion County Cr/Ni Field Recovery and Analytical Protocol
-------
5.4.5 Data Reduction
Chromium and nickel concentrations in the flue gas were calculated using
the same equations in Section 5.3.5 for Pb/Cd.
5.5 BAGHOUSE ASH, CYCLONE ASH, LIME SLURRY AND TESISORB EVALUATIONS
5.5.1 Baghouse Ash and Cyclone Ash
The baghouse and cyclone ash were sampled and analyzed according to the
same procedure. The ash samples were analyzed for CDD/CDFs for Runs 1 through
3 and for hexavalent and total chromium by colorimetry and AA, respectively,
for Runs 4 through 6. Aliquots of the hexavalent and total chromium extracts
were also analyzed by NAA to obtain arsenic, cadmium, nickel and chromium
results.
Grab samples of baghouse ash and cyclone ash were collected every 60
minutes starting 30 minutes after the start of flue gas sampling. The
30-minute delay allowed for ash hold-up in the ash handling system so that the
flue gas and ash sampling would be as simultaneous as possible. Approximately
1 kg of sample was collected in each grab.
At the end of each test run, the grab samples were mixed well and
quartered. Three 500 gram aliquots were placed in individual bottles. One
aliquot was analyzed for CDD/CDF or hexavalent chromium, total chromium
arsenic, cadmium, chromium and nickel. One aliquot was given to Ogden
Projects, and one aliquot was archived.
The CDD/CDF sampling and analysis scheme is shown in Figure 5-10. The
four hour flue gas sampling period allowed for a minimum of four grab samples
to be collected. A 10 gram aliquot was used for analysis.
The metals sampling and analytical protocol scheme is shown in
Figure 5-11. The six hour flue gas sampling period allowed for a minimum of
5-26
-------
Grab no. 1 Equal
Volume Sample
Grab No. 2 Equal
Volume Sample
Grab no. 3 Equal
Volume Sample
Grab no. 4 Equal
Volume Sample
Field Recovery
Combine Grab Samples
for Composite
1 Laboratory 1
T Analysis T
Remove 10 gram Aliquot
Extract with Toluene
in Soxhlet
Concentrate Toluene
Extract to 40ml w/N2
I
Concentrate to 1ml
Extract
i
Analyze for Dioxin/
FuransbyGC/MS
CDD/CDF (ng/g)
tr
r-
o
T—
s
co
Figure 5-10. Marion County Baghouse Ash and Cyclone Ash
CDD/CDF Analytical Protocol
5-27
-------
Grab No. 1
Equal
Volume
Sample
Grab No 2
Equal
Volume
Sample
Grab No. 3
Equal
Volume
Sample
Grab No. 4
Equal
Volume
Sample
Grab No. 5
Equal
Volume
Sample
Combine Grab Samples
for Composite
Grab No. 6
Equal
Volume
Sample
I Field Re
10 grams aliquot of ash
combined with 100mL of
2% NaOH and 3% Na,CO,
solution
f Laboratory i
Analysis *
Combine 40ml 2% NaOH,
3% NaCO, Solution
Heatw/constant stirring
for 30 minutes
Cool: Filter using 47mm
Teflon Filter Media
Collect Filtrate and Dilute
to 100mL Volume w/Olst. H,0
•Remove Aliquot:
Analyze for Cr' 4
Other Metals by NAA
Mix; Remove Aliquot; Adjust
Volume to SOmLw/Dlst. H,0
Adjust pH to 2.0 with 10%
H.SO.; Dilute to 100ml w/H.O
Add Dlphenylcarbazlde Color
Reagent; Adjust Volume to
100mL with Delonlzed H,0
Allow 10 minutes for Color
Development; Analyze
Colorlmetrlcally for Cr*
Cr-lyg/g)
I
Digest Paniculate Matter
on Filter w/3mlof1:1
H,SO. and 10ml cone. HF
-L
Evaporate w/Water Bath;
Add Additional HF until
Filter Media Is Dissolved
Heat evaporate to near
Dryness
Combine Solution with
30mL9MH,SO,
Reflux for 4 Hours
at 150 'C
Cool; Filter using 3-M m
Teflon Filter Media
Dilute Filtrate to 100ml
Volume
Analyze for Cr1 by AA
Air Acetylene Flame
1
Cr' (yg/g)
"Remove Aliquot;
Analyze for Cr* 4
Other Metals by NAA
1
o
r^
CO
NAA results are reported in a separate appendix.
Figure 5-11. Marion County Baghouse Ash and Cyclone Ash Metals
Analytical Protocol
5-28
-------
six grab samples to be collected. The ash samples were digested in an
alkaline solution and the digested sample was split for Cr (+VI) analysis by
colorimetry and Cr (+III) analysis by atomic absorption. Aliquots of the
extracts were submitted for NAA analysis in order to obtain nickel, cadmium,
chromium and arsenic results.
5.5.2 Lime Slurry and Tesisorb Sampling and Analysis
Lime slurry and Tesisorb were sampled and analyzed according to a similar
procedure. The lime slurry and Tesisorb samples were analyzed for background
metals which included cadmium, chromium, nickel and arsenic for Runs 4-6 by
neutron activation analysis (NAA).
Grab samples of the lime slurry and Tesisorb were collected once during
each test run. The samples from Runs 4, 5 and 6 were composited into one
single sample before being submitted for NAA. An aliquot of each was analyzed
for metals by NAA. Identical aliquots were also given to Ogden Projects. The
sampling and analytical scheme is shown in Figure 5-12.
The NAA analysis was performed by the Nuclear Services Laboratory at
North Carolina State University in Raleigh, N.C. NAA can be used to analyze
for all the specific metals except lead and beryllium; however, the method
cannot differentiate between different valence states or compounds of a metal
such as Cr(+III) or Cr(+VI).
During the NAA procedure, the samples are exposed to neutrons. The
neutrons excite the metal atoms causing them to emit gamma rays. The density
and wavelength of the gamma rays are measured and the information is logged by
a computer. Calibration standards with known amounts of metal standards are
also included in the sample batch, and by comparing the results from these
calibration standards to those of the samples, the type and concentration of
each metal detected is determined. However, since lead and beryllium do not
emit gamma rays, NAA cannot measure these metals.
5-29
-------
Run#1
Grab Sample
Run #2
Grab Sample
Run #3
Grab Sample
Composite Sample
of Runs 1,2, &3
Analyze for Trace
Metals by NAA
Metals (^g/g)
Figure 5-12. Marion County Metals Analytical Protocol for Tesisorb
and Lime Slurry Samples
5-30
-------
5.6 S02 SAMPLING AND ANALYSIS
S0~ was measured according to EPA Method 6C. EPA Method 6C measures SO-
using a continuous emissions monitor. 0~ was also monitored in order to
calculate SO^ emission factors and to aid in leakchecking the sampling system.
5.6.1 Equipment and Sampling Preparation
Each component of the GEM system was cleaned, leakchecked and calibrated
before going into the field. The components of the system include the probe,
the heated umbilical, the gas conditioner and pump, the manifold, the
analyzers, the computer data logger, and the strip chart recorder.
The probe was checked for leaks, and the filter was inspected. The
umbilical consists of teflon tubing which has been wrapped in heat tracing and
a protective sheath. The umbilical was cleaned with hot water, and then dried
and blanked with nitrogen. The umbilical was also leakchecked. The gas
conditioner pump, and manifold were leakchecked, cleaned, and the cooling
operation was checked.
A three point calibration was performed on each of the analyzers. The
calibrations also tested the operation of the computer data logger and
stripchart recorder. A correlation coefficient was calculated from the three
points to check the linearity of the response of the analyzer. (Acceptance
criterion is discussed in Section 6). Once the system was prepared, it was
disassembled and packed carefully for transporting to the site.
5.6.2 Sampling Operations
A schematic of the SO, GEM system used at the Marion County MWC is shown
in Figure 5-13. Before each run, the system was leakchecked and a system
blank was analyzed. Each of the analyzers was calibrated with a commercially
prepared and certified zero and span gas and the response factor was
5-31
-------
I
U)
ro
Flue Gas
from Stack
Instack Probe
with Filter
Heat Traced Teflon Umbilical
(50ftlong)(>120°C)
is
Gas Conditioner
and Pump (0 - 5 °C)
Exhaust
Overflow
Electronic Signal I
Located In Instrument Trailer
Figure 5-13. SO2 GEM Sampling and Analysis Scheme for the Marion County
Test Program
-------
determined. The calibration gases were introduced at the manifold. The
calibration results were transmitted to the data logger which also recorded
the data during the sampling run. The data logger scanned each analyzer 180
times per minute, and recorded one-minute averages. The data logger stored
the one-minute average instrument response from each analyzer on disk, and
also converted the response to a concentration using the calibration data
which was printed out as a hard copy. The instrument responses were also
recorded on a stripchart as a back-up to the data logger. At the end of the
sampling run, another zero and span calibration was performed. The final
calibration was used to determine the daily drift of the instrument. The
listings of the one-minute averages, stripcharts, and calibrations are
included in Appendix B.
5.6.3 Data Reduction
The data reduction for the GEM results was performed on computer. The
exact equations used are included in Appendix A.5 and are discussed in general
in this section. At the end of the sampling run, the daily drift of each
instrument was evaluated by comparing the initial and final calibrations.
Then, each one minute average was adjusted by assuming the drift was linear
throughout the day. Also, any invalid sections of the data were deleted
before the averages and standard deviations were calculated for each
parameter. The SO- values were also adjusted for the quenching effect of C0_
and 09. The relationship used to adjust the values was provided by the
instrument manufacturer.
5.7 MOLECULAR WEIGHT BY EPA METHOD 3
5.7.1 Sampling Operations
The molecular weight of the flue gas was determined using EPA Method 3.
During the flue gas sampling an integrated bag sample was extracted at a
single point from the sampling location at a rate of 0.5 ml/min. The sampling
train used is shown in Figure 5-14. The moisture knockout trap allowed the
5-33
-------
1/4" Stainless Steel
Probe
vn
I
OJ
Tubing
Pump
Ice Water Cooled
Moisture Knockout Trap
Tedlar Bag
Figure 5-14. Method 3 Integrated Bag Sampling Train
-------
analysis to be on a dry basis. Total sample volume was approximately 1.5
cubic feet.
5.7.2 Analysis
TM
The integrated bag samples were analyzed with an Orsat analyzer for CCL
and 0_. Nitrogen was determined by difference. The analysis was repeated
three times on each bag, and an average was used as the input for the Method 5
calculations. The absorbing solutions used in the analyzer were potassium
hydroxide (45% by volume) for carbon dioxide and pyrogallate (pyrogallol in
potassium hydroxide) for oxygen. The Orsat data sheets are included in
Appendix C.7.
5.8 VOLUMETRIC FLOWRATE BY EPA METHOD 2
Volumetric flowrate was measured according to EPA Method 2. A type K
thermocouple and S-type pitot tube were used to measure flue gas temperature
and velocity, respectively.
5.8.1 Sampling and Equipment Preparation.
For EPA Method 2, the pitot tubes were calibrated before use following
the directions in the method. Also, the pitots were leakchecked before and
after each run.
5.8.2 Sampling Operations
The volumetric flowrate data were recorded simultaneously with the other
sampling train data. The parameters that were measured included the pressure
drop across the pitots, stack temperature, stack draft and ambient pressure.
These parameters were measured at each traverse point which were previously
discussed in Section 4.0. A computer program was used to calculate the
average velocity during the sampling period. The calculations are included in
Appendix A.5.
5-35
-------
5.9 MOISTURE BY EPA METHOD 4
The average flue gas moisture content was determined according to EPA
Method 4. Before sampling, the initial weight of the impingers were recorded.
When sampling was completed, the final weights of the impingers were recorded,
and the weight gained was calculated. The weight gained and the volume of gas
sampled were used to calculate the percent by moisture of the flue gas. The
calculations are performed by computer, and a sample calculation is included
in Appendix A.5.
5-36
-------
6.0 QUALITY ASSURANCE AND QUALITY CONTROL (QA/QC)
Quality assurance/quality control guidelines outline pertinent steps
followed during the production of emission data to ensure and quantify the
acceptability and reliability of the data generated. The measures outlined in
this section were followed to ensure the production of high quality data from
the sampling and analytical efforts.
6.1 STANDARD QUALITY ASSURANCE PROCEDURES
QA/QC procedures were followed during sampling and analysis to ensure
that the data generated were of acceptable quality. The following quality
control and quality assurance procedures were used during EPA reference method
sampling and/or analysis.
6.1.1 Sampling Equipment Preparation
The checkout and calibration of source sampling equipment is vital to
maintaining data quality. Referenced calibration procedures were strictly
adhered to when available, and all results were documented and retained.
Table 6-1 summarizes the parameters of interest and the types of sampling
equipment that were used to measure each parameter. Prior to sampling, all
equipment was cleaned and checked to ensure operability. Equipment requiring
pretest calibration (Table 6-1) was calibrated in accordance with "Quality
Assurance Handbook of Air Pollution Measurements Systems, Volume III,
Stationary Source Specific Methods," (EPA 600 4-77-027b).
Pre-test calibration data for type "S" pitot tubes, temperature measuring
devices, and dry gas meters can be found in Appendix E.2. Balance calibration
data are located in Appendices D.4 (Particulate Operations Log) and H.4 (Field
Laboratory Log).
6-1
-------
TABLE 6-1. SUMMARY OF EQUIPMENT USED IN PERFORMING SOURCE SAMPLING AT THE MARION COUNTY MWC
o\
Parameter
Volumetric Flue Gas
Flow Rate
Gas Phase Composition
Moisture
Molecular Weight
CDD/CDF
Pb/Cd/Particulates
Method
EPA 1 & 2
EPA 4
EPA 3
ASME/EPA
Protocol
Draft
Cadmium
Protocol
Calibrated Equipment Used to Measure Parameters
Type "S" Temperature
Pitot Measuring Dry Gas
Tube Manometer Device Orsat Nozzles Balances Meter
XXX
XXX XX
X
X X X X X X X
X X X X X X X
Cr/Ni
Draft EPA
Protocol for
sampling
hexavalent and
total chromium
HC1/Particulates
EPA Method 5
with modifications
-------
6.1.2 General Sampling QC Procedures
The following QC checks were conducted for each of the EPA Methods 2, 3,
4, 5, and CDD/CDF, Pb/Cd/Particulate, Cr/Ni and HCl/Particulate sampling:
Prior To Sampling
Sampling equipment was inspected for possible damage from
shipment, and thoroughly checked to ensure operable components.
Duct size was determined.
Preliminary velocity, temperature, and moisture was determined to
aid in conducting isokinetic sampling.
Cyclonic or turbulent flow checks were performed.
The proper sampling nozzle size was determined.
Prior to Testing Each Day
The train was assembled in an environment free from uncontrolled
dust.
The number and location of the sampling points were checked before
taking measurements.
The S-type pitot tube was visually inspected.
Each leg of the S-type pitot tube was leak-checked.
The entire sampling train was leak-checked.
The oil manometer used to measure pressure across the S-type pitot
tube was leveled and zeroed.
The temperature measurement system was checked for operability by
measuring the ambient temperature prior to each traverse.
The heating and cooling systems were checked to ensure proper
operation, and cooling systems were stocked with ice.
Each sampling train was visually inspected for proper assembly.
Sampling ports were sealed to help prevent possible air inleakage.
Data requirements were reviewed prior to each sampling run.
During Testing Each Day
The roll and pitch axis of the S-type pitot tube and sampling
nozzle were properly maintained.
The train was leak-checked before and after any move from one
sampling port to another during a run or if a filter change took
place.
The probe, filter, sorbent trap, and impinger outlet were maintained
at the proper temperatures.
Ice was maintained in the impinger bath at all times.
Proper readings of the dry gas meter, pressures, temperature, and
pump vacuum were made during sampling at each traverse point.
Isokinetic sampling velocity was maintained within + 10 percent of
the duct velocity.
Any unusual occurrences involving the sampling train were noted on
the appropriate data form.
6-3
-------
After Testing Each Day
The entire sampling train was leak-checked.
The sampling nozzle was visually inspected.
The S-type pitot tube was visually inspected.
Each leg of the S-type pitot tube was leak-checked.
The sorbent trap, probe, filter, and impingers were immediately
recapped as the train was disassembled.
Any necessary sample train blanks were collected.
Field data sheets were checked for completion.
Field data sheets were given to sampling team leader for review.
6.1.3 Sample Recovery
To ensure sample integrity, careful recovery techniques were adhered to
by knowledgeable analysts. This section outlines the quality control
procedures followed to ensure sample integrity. These included:
Filters were handled out of drafts and transferred with tweezers.
Sample trains were disassembled and the samples recovered in clean
areas to prevent contamination.
The nozzle was capped prior to and following recovery.
The samples were transferred to appropriate storage containers and
clearly labeled.
Reagent dispenser bottles were clearly labeled.
The toploading field balance was rezeroed after every ten weighings
or less.
Sampling glassware was routinely rinsed three times with each
reagent to remove all of the sample.
Reagent lot numbers were recorded.
All sampling and recovery glassware was capped when not in use.
Probe and nozzle brushes, tweezers, and scrapers were rinsed before
use with the proper reagent(s) to prevent sample contamination.
Probe liners were recovered in an enclosed area. First, the nozzle was
removed, and wrapped in foil until recovery (if applicable). Then, the probe
liner was carefully removed from the sheath. The probe liners were recovered
by attaching a 250 ml flask on the male end of the probe. The flask ensured
6-h
-------
that all the rinses were collected. Reagents were rinsed into the nozzle end
of the probe and collected in the flask. For the HC1/PM, CDD/CDF, and
Pb/Cd/PM trains, the first reagent rinse was acetone. The probe liner was
rinsed and brushed at least three times until the liner was qualitatively
cleaned of particulate. For the Cr/Ni train the probe liner was brushed with
distilled water. Following the brushing, the probe liner was rinsed with the
appropriate second reagent.
6.1.4 Sample Custody
Each sample container was tared before use and weighed with the sample to
determine the sample weight. The liquid level was also marked on the label.
The sample label contained an alphanumeric code for identifying the sample as
well as an unique identifying code. The sampling date and weights were also
recorded on the label. Once the label was completed using pencil, the label
was sealed with solvent resistant tape. The pencil and tape would resist
smearing in the event the bottle leaked. The bottle lids were sealed with
teflon tape and an integrity seal was placed over the lid. Then, the sample
was logged into the master logbook which contained all the information on the
label as well as the sample's destination. Finally, the samples were wrapped
in bubblewrap and plastic baggies and packed upright in coolers.
Chain-of-custody forms were completed, signed and enclosed in each box of
samples. Also accompanying each shipment of samples was a letter detailing
the type of analyses required and the identification codes for the sample
bottles.
6.1.5 Preparation of Samples for Analysis
Upon receipt at the laboratory, the sample bottles were checked for signs
of leakage (liquid levels and weights verified) and the contents of the
shipment checked against the chain-of-custody form.
Prior to analysis, each sample was properly prepared for the appropriate
analytical method. This section outlines the quality control procedures used
to ensure proper sample preparation. Included are:
6-5
-------
Each sample identification code was crosschecked for accuracy
against the sample logbook.
The analytical requirements of each sample were reviewed.
The sample containers were checked for leakage or damage and any
anomalies were noted.
6.1.6 Sample Analysis
The quality assurance/quality control procedures followed during the
analysis task were dependent on the specific analysis being performed. One or
more of the following steps were taken:
Duplicate analyses were performed on 1 out of every 10 CDD/CDF
samples.
Blanks were analyzed to correct for background and/or matrix
interferences.
Blind QC samples were submitted to the analytical laboratories along
with the field samples.
For the CDD/CDF analyses, the samples were spiked with known
additions of the species of interest and recoveries were calculated.
6.1.7 Data Documentation and Verification
Several measures were taken to verify the completeness and accuracy of
the data generated. These included:
All sampling data were recorded on preformatted data sheets.
Data tables were prepared and reviewed for completeness and
accuracy.
All data that appeared to be outside expected ranges were carefully
scrutinized for process upsets and reanalyzed as necessary.
6.2. METHOD-SPECIFIC SAMPLING QC PROCEDURES
In addition to the general QC procedures listed in Sections 6.1.1 through
6.1.7, QC procedures specific to each sampling method were also incorporated
into the sampling scheme. These method specific procedures are discussed
below.
6-6
-------
6.2.1 Quality Control Procedures for Velocity/Volumetric Flow Rate
Determination
Data required to determine the volumetric gas flow rate were collected
using the methodology specified in EPA Method 2. Quality control procedures
followed were:
The S-type pitot tube was visually inspected before and after
sampling.
Both legs of the pitot tube were leak-checked before and after
sampling.
The number and location of the sampling traverse points were checked
before taking measurements.
Proper orientation of the S-type pitot tube was maintained while
making measurements.
The oil manometer was leveled and zeroed and the proper pressures
and temperatures were recorded.
6.2.2 Quality Control Procedures for Molecular Weight Determinations
Sampling
Samples used for determination of stack gas molecular weight were
collected using the integrated sampling technique specified in EPA Method 3.
Quality control for the Method 3 sampling focused on the following procedures:
The sampling train was leak checked before and after each run.
A constant sampling rate (+ 10 percent) was used in withdrawing the
integrated gas sample.
The sampling train was purged prior to sample collection.
The sampling port was sealed to prevent air inleakage.
Analysis
Analytical quality control for Method 3 included the following:
The Orsat analyzer was leak-checked prior to use.
The Orsat analyzer was leveled and fluid levels zeroed prior to each
use.
6-7
-------
The Orsat analyzer was thoroughly purged with sample prior to
analysis.
Analyses were repeated until analysis agreed within 0.2% absolute.
Orsat solutions were changed when more than six passes were required
to obtain a stable reading for any component.
Validation of Results
The validity of Orsat analysis results were confirmed based on a
combustion stoichiometry method described in Reference 16. First, the
ultimate CO concentration was calculated based on an ultimate analysis of the
fuel. Since ultimate analyses were not performed on the refuse from this
site, data were used from a similar MSW. The ultimate analyses data used was
an average of twelve analyses. Then, on an 0^ versus C0? axis, a line was
drawn connecting the 0. intercept, 20.9 %, with the CO^ intercept and the
ultimate percent CQ . The Orsat results should fall within 10 percent of this
line. This analysis for the Orsat data is shown in Figure 6-1.
6.2.3 Quality Control Procedures for Moisture Determination
The moisture content of the gas streams was determined using the
technique specified in EPA Method 4. The following internal QC checks were
performed as part of the moisture determinations:
Each impinger was weighed to the nearest 0.1 grams before and after
sampling.
The field balance was rezeroed after every 10 weighings.
Only fresh, dry silica gel was used in the silica gel container.
Ice was kept in the ice bath to keep the gas exit temperature below
68 degrees F while sampling.
The sample train was leak checked before and after each run.
Dry gas meter readings were correctly recorded on the proper data
sheet.
6.2.4 Quality Control For CDD/CDF Testing
This section summarizes the quality control activities for CDD/CDF
testing at Marion County, OR. The aforementioned general quality control
6-8
-------
22
20
18
«f 16
CO
ct5
.0
b 14
c
CD
o
CD
Q.
CD
J3
O
C
0)
O)
>,
X
o
12
10-
8
6
4-
2-
Key
• ORSAT Analyses Boiler Outlet
3,5,6
8
S
i 2 4 6 8 10 12 14 16 18 20
CO2 (volume percent, dry basis)
'Ultimate CO, and O2 values were determined from data from a similar MSW
:igure 6 -1. Validation of Fixed Gas Analysis for Marion County 02 and CO2 Data
6-9
-------
activities also apply, as they do to all sampling methods based on EPA
Method 5. The specific CDD/CDF QC activities discussed in this section
include sampling preparation, sampling operations, surrogate and internal
standard recoveries and sample blanks.
CDD/CDF Equipment and Sampling Preparation
Pre-test calibrations or inspections were conducted on pitot tubes,
temperature sensors, dry gas meters, and balances. Precleaning procedures for
sample train glassware and amber glass sample bottles were followed as
specified in Section 5.1.1. After cleaning, each piece of the sampling
glassware was sealed with solvent cleaned aluminum foil to prevent
contamination. Sample bottles were sealed with teflon lined lids.
The XAD-2 and filter cleanup residual analyses results for the XAD resins
and filters are contained in Appendix 1.2. The residual analysis is a
requirement of the CDD/CDF sampling protocol.
CDD/CDF Sampling Operations
CDD/CDF sampling operations followed standard Method 5 operating
procedures (Appendix K.4) with the addition of the following:
1) Only precleaned aluminum foil or ground glass caps were used to
cover sample train components during train assembly, disassembly,
and leak checks.
2) The temperature of the gas entering the sorbent trap (XAD) remained
below 68°F.
3) The sampling rate averaged 0.59 dscfm.
4) All train components which were recovered were made of glass or
teflon.
5) All train components and sample bottles were marked according to the
cleaning procedure used.
CDD/CDF sampling results for isokinetics and leak checks are presented
in Table 6-2.
6-10
-------
TABLE 6-2. CDD/CDF TESTING ISOKINETICS3 AND LEAK CHECKb
SUMMARY BOILER OUTLET AT MARION COUNTY
Percent Sampling
Date Run # Isokinetics Leak Check Port
9/22/86 1 104.6 Initial
Port Change
Continue
Final
9/23/86 2 103.0 Initial
Port Change
Continue
Final
9/24/87 3 99.0 Initial
Port Change
Continue
Final
A
A
B
B
A
A
B
B
A
A
B
B
Leak Rate
(ft3/min)
0.007
0.007
0.007
0.005
0.015
0.013
0.013
0.010
0.016
0.019
0.012
0.012
Pressure
(in H20)
15
14
14
16
15
14
14
17
15
18
15
15
The QC requirement was isokinetics of 100 + 10 percent.
''The QC objective for leak checks was a leak-free train or leakage rate less
than or equal to 0.02 cfm or less than 4% of the average sampling rate
(whichever is less).
6-11
-------
Sample Recovery
CDD/CDF sample recovery followed the procedure presented in Section 5.
In addition to the general recovery QC procedures, these procedures were also
implemented:
All instruments used in the recovery process were either teflon or
stainless steel.
All instruments which came in contact with the sample, or the
sampling surfaces were cleaned according to the CDD/CDF cleaning
procedure.
No sealing grease was used on the sampling train. Reagent,
laboratory proof, and field blanks were taken.
Each sample container lid was individually sealed with teflon tape
to prevent leakage.
Each container lid was covered with an integrity seal (strip of
tape) to prevent tampering.
Samples were weighed in the field and again in the laboratory to
indicate possible sample loss.
^
All sample containers were packaged for transport in Ziplock
plastic bags, wrapped in bubble wrap, and placed into a second
ziplock bag.
CDD/CDF Surrogate and Internal Standard Recoveries of the Test Samples
CDD/CDF samples were spiked with internal standards and surrogates
prior to extraction. The internal standards were added in the soxhlet
extraction step. The surrogates were added to the impinger (condensate)
fraction. The internal standard 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, since the surrogate
results are adjusted using the internal standard recoveries.
The internal standard recoveries are summarized in Table 6-3. The QC
objective as required by the ASME/EPA protocol is + 50 percent. The front
13 13
half fractions of the flue gas samples and the C.. -CDD and C19-HxCDD
recoveries for the back half flue gas samples did not meet these requirements.
The internal standard recoveries were extremely low such that the native
species results were not reportable. However, the internal standard
recoveries for the cyclone ash, baghouse ash, and quality control samples were
all within the acceptable range.
6-12
-------
TABLE 6-3. INTERNAL STANDARDS RECOVERY RESULTS
FOR MARION COUNTY CDD/CDF ANALYSES
Recovery (%)
, _ 2378-13C10-TCDD
Sample Type 12
Flue Gas Samples
Run 01 0
Run 02 3
Run 03 0
Cyclone Ash
Run 01
Run 02
Run 03
Baghouse Ash
Run 01
Run 02
Run 03
Quality Control Samples
Recovery Efficiency
Blank
FH/BHa
.84/84.5
.22/81.7
(72.9)b
.01/87.4
100
98
94
101
100
97 (92)b
85
Laboratory Proof Blank 94
Soxhlet Blank #1
Soxhlet Blank #2
Aqueous Blank
105
101
92
13
FH/BHa
0.01/11.3
1.27/11.6
0.02/23.8
60
87
114
62
73
82 (72)b
28
50
74
86
62
C12-HxCDD
FH/BHa
0.01/17.2
1.27/17.6
o
( ID . 1 )
0.03/29.3
67
99
119
65
72
76 (71)b
32
55
73
80
63
13C12-HpCDD
FH/BHa
0.01/84.6
0.72/94.1
(83.1)b
0.02/82.8
90
150
134
83
150
100 (73)b
89
95
86
78
79
13C12-OCDD
FH/BHa
0.01/72.9
0.40/95.7
(84.9)b
0.29/87.3
86
136
129
82
151
89 (71)b
83
98
81
72
79
FH/BH = front half/back half percentages of internal standard recoveries.
Values in parenthesis are recoveries for duplicate analyses.
6-13
-------
The results indicate that a matrix effect of the front half fraction caused
the loss of the internal standards. The good recoveries of the other samples
indicate that it was not a systematic analytical laboratory error.
The internal standards are used to adjust the responses for extraction
efficiency and variable instrument performance. Since the internal standards
13
are spiked as a known amount (20 ng except for C-2-OCDD which is spiked at
40 ng) , the recovery is determined using the external standard and the
following equation. External standards are the same compounds used for
internal standards except that they are in a pure organic matrix rather than a
sample matrix.
Internal standard _ (area counts) internal standard
recovery (area counts) external standard
The sample results are calculated:
results reported (ng) = RF x F x (area counts) sample
,
where
F = the inverse of the internal standard recovery
KhiO
RF = response factor determined with external standards
(ng per area count)
The adjustment is done by computer so that the results reported by
Triangle Laboratories are already adjusted. Table 6-4 presents the internal
standard adjustment factors for the CDD/CDF samples.
13 13
The low recoveries of the C..--CDD and C..--HXCDD in the back half
fractions of the MM5 inlet samples are outside the acceptable range suggested
by the ASME protocol. However, this may not be entirely due to low extraction
or recovery efficiency of the standards. The operating conditions of the mass
spectrometer are computer controlled and reset for the tetra, penta/hexa, and
hepta/octa-CDD's . The consistent but low recoveries could indiciate that the
computer was not exactly resetting the analytical operating conditions for
penta/hexa CDDs to the external calibration conditions. In this case, the
internal standards are used to compensate for poor mass spectroscopy
performance as well as sample loss during recovery or extraction.
6-iU
-------
TABLE 6-4. FACTORS ADJUSTING RESPONSES FOR EXTRACTION EFFICIENCY
AND VARIABLE INSTRUMENT PERFORMANCE
^
Factors
-, _, 2378-l3C10-TCDD 13C10-PCDD
Sample Type 12 12
Flue Gas Samples
Run 01
Run 02
Run 03
Cyclone Ash
Run 01
Run 02
Run 03
Baghouse Ash
Run 01
Run 02
Run 03
FH/BH
119/1.18
31.1/1.22
10,000/1.14
1.0
1.02
1.06
0.99
1.0
1.03
FH/BH
10,000/8.85
78.7/8.62
5,000/4.20
1.67
1.15
0.88
1.61
1.37
1.22
13C12-HxCDD
FH/BH
10,000/5.81
78.7/5.68
3,333/3.41
1.49
1.01
0.84
1.54
1.39
1.32
13C12-HpCDD
FH/BH
10,000/1.18
139/1.06
5,000/1.21
1.11
0.67
0.75
1.20
0.67
1.0
13C12-OCDD
FH/BH
10,0001/1.
250/1.
345/1.
1.16
0.74
0.78
1.22
0.66
1.12
37
04
15
Quality Control Samples
Recovery
Efficiency Blank
Laboratory Proof
1.18
3.45
3.13
1.12
1.20
Blank
Soxhlet Blank #1
Soxhlet Blank #2
Aqueous Blank
1.06
0.95
0.99
1.09
2.0
1.35
1.16
1.61
1.82
1.37
1.25
1.59
1.05
1.16
1.28
1.27
1.02
1.23
1.39
1.27
2378- C10-TCDD is used to adjust 2378-TCDD and total TCDD, 2378-TCDF, and total TCDF.
13
C -PCDD is used to adjust 12378-PCDD, 12378-PCDF, 23478-PCDF, total PCDD and
total PCDF.
13C -HxCDD is used to adjust 123478-HxCDD, 123678-HxCDD, 123789-HxCDD, 123478-HxCDF,
123678-HxCDF, 123789-HxCDF, total HxCDD and total HxCDF.
13C. -HpCDD is used to adjust 1234678-HpCDD, 1234678-HpCDF, 1234789-HpCDF, total HpCDD
and total HpCDF.
C -OCDD is used to adjust total OCDD and total OCDF.
3FH/BH - front half/back half factors.
6-15
-------
The surrogate recoveries are summarized in Table 6-5. The surrogate
recoveries of the front half flue gas fractions did not meet the QC
requirement of + 50 percent. The very high recoveries reported for the front
half fractions are considered to be a result of the very low internal standard
recoveries. The surrogate results are adjusted by the internal standards.
Thus, a very low internal standard recovery results in a very large adjustment
factor such as shown previously in Table 6-4.
The back half flue gas fractions, the cyclone ash, the baghouse ash and
the quality control samples did meet the requirements except for the
10 13
recoveries of C -HxCDF. The low recoveries of C -HxCDF are considered
by Triangle Laboratories to be due to the chromatographic retention time of
the specific HxCDF isomer used. The technicians were not controlling the time
from loading the sample in the GC/MS to the start of the elution which was
causing the compound to elute before the peak could be recorded.
CDD/CDF Audit Results
Audit samples were prepared by Radian's Organic Chemistry Laboratory and
submitted along with the field samples. The audit samples consisted of one
water sample and one XAD trap spiked with 2378-TCDD and 2378-TCDF as well as
one blank water sample and one blank XAD trap. The results are summarized in
Table 6-6. The recoveries for the spiked samples ranged from 61 to 100
percent which were within the QC requirement of 50 to 150 percent recovery.
The blank samples contained very low amounts of the target compounds which
were less than ten times the detection limit. These values are in the
background noise range and are considered acceptable.
CDD/CDF Sample Blanks
Recovery efficiency and laboratory proof blanks were collected and
analyzed to evaluate contamination from the glassware, handling of the train
and field recovery. The recovery efficiency blank was a second recovery of a
previously used sample train and just includes the solvent rinses. The
laboratory proof blank quantified the background levels of CDD/CDF from the
glassware and recovery equipment. The laboratory proof blanks were obtained
6-16
-------
TABLE 6-5. SURROGATE RECOVERIES FOR MARION COUNTY CDD/CDF ANALYSES
Sample Type
MM5 Boiler Outlet Sample
Run 01
Run 02
13
FH/BHa
219/114
149/107
Recoveries (%)
37C1. -TCDD
4
FH/BHa
126/99
96/96
13C12-HxCDF
FH/BHa
42/0.02
0.22/0.07
(0.04)b
Run 03
2,690/113
1,495/103
42/0.06
Cyclone Ash
Run 01
Run 02
Run 03
96
100
106
110
109
112
0.02
1.5
9.8
Baghouse Ash
Run 01
Run 02
Run 03
98
113
109 (114)1
106
110
109 (115)1
1.1
0.04
0.05 (0.72)1
Quality Control Samples
Recovery Efficiency Blank
Laboratory Proof Blank
Soxhlet Blank #1
Soxhlet Blank #2
Aqueous Blank
117
100
96
95
101
110
100
106
107
109
0.67
0.40
0.73
0.02
0.32
aFH/BH = front half/back half percentages of surrogate recoveries.
"Values in parenthesis are recoveries for duplicate analyses.
6-17
-------
TABLE 6-6. MARION COUNTY CDD/CDF AUDIT SAMPLES
Sample #
13821
13822
13823
13824
Component
Water
solution
Water
solution
Blank XAD
Spike XAD
Amt.
Spiked
0
760
0
1500
2378 TCDDa
Amt. Percent
Meas . Recovery
ND
676 89
0.14b
1354 90
Amt.
Spiked
0
300
0
3000
2378
Amt
Meas
0.06
300
0.06
1824
TCDFa
Percent
Recovery
b
100
b
61
Values reported in nanograms (ng).
Values which are less than ten times the detection limit are in the
background noise range and are considered acceptable.
6-18
-------
from a complete set of sample train glassware and recovery equipment (brushes,
spatulas, etc.) that has been cleaned according to the specified precleaning
procedure. This glassware, which consisted of the probe liner, filter holder,
condenser coil, and impingers, was loaded and then recovered according to the
standard recovery method. Also, blanks of each solvent lot and filters used
at the test site (reagent blanks) were saved for potential analysis.
The results of the recovery efficiency and laboratory proof blanks are
summarized in Table 6-7. The laboratory proof blank results indicate some
contamination in the hepta-CDD, octa-CDD and tri-CDF. However, these values
which are less than ten times the detection limit are in the background noise
range and are considered acceptable.
The recovery efficiency blank for Run 1 also contains some significant
2378-TCDD and 2378-TCDF contamination. The level of contamination (2 ng) is
greater than seen in any of the flue gas samples. In addition, the
contamination is all the 2378-specific isomer. No other TCDD or TCDFs were
detected. The same pattern is also seen in the blank XAD audit sample
although to a lower level of contamination. These considerations indicate
contamination somewhere in the analytical procedure rather than poor recovery
efficiency, and therefore the results are not considered to affect the
validity of the CDD/CDF flue gas results.
The efficiency of the sample recovery procedure is calculated in
Table 6-8. The results from Run 2 are used in this calculation due to the
analytical problems with the Run 1 sample. The data show good recovery of the
sample except for 2378-TCDD. However, this 2378-TCDD result is considered
questionable as discussed in the previous paragraph.
Triangle Laboratories analyzes several method blanks with a batch of
samples. The purpose of the method blank is to measure contamination caused
by the analytical procedure. Basically, CDD/CDF samples are either solids
extracted in a soxhlet extraction or aqueous samples which are first
solvent-exchanged with hexane. Therefore, two method blanks are analyzed; a
6-19
-------
TABLE 6-7. SUMMARY OF CDD/CDF RECOVERY EFFICIENCY AND
LABORATORY PROOF BLANK RESULTS
o\
o
ISOMER
DIOXIN
Mono-CDD
D1-CDD
TM-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hepta-CDD
Octa-CDD
TOTAL PCDD
FURAN
Mono-CDF
D1-CDF
TrM-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Octa-CDF
a De-tect 1
are det
RECOVERY
AMOUNT
(ng)
ND
ND
ND
2.55
0.00
ND
ND
ND
ND
ND
ND
0.28
0.62
ND
ND
ND
1.18
0.00
ND
ND
ND
ND
ND
ND
ND
ND
ND
on limits are
EFFICIENCY
DETECT. a
LIM.
0.004
0.007
0.011
0.035
0.035
0.043
0.041
0.047
0.043
0.003
0.005
0.008
0.018
0.027
0.021
0.022
0.023
0.028
0.025
0.013
0.023
not rep o r
BLANK
RATIO
1.000
1.000
1.000
0.790
0.790
1.000
1.000
1.000
1.000
1.000
1.000
0.923
0.839
1.000
1.000
1.000
0.768
0.768
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
ted for resu
LAB
AMOUNT
(ng)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.29
0.50
ND
ND
0.06
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Its where na
PROOF BLANK
DETECT.3
LIM.
0.004
0.006
0.010
0.003
0.008
0.019
0.019
0.024
0.023
0.027
0.025
0.003
0.004
0.006
0.006
0.009
0.014
0.011
0.012
0.013
0.016
0.014
0.012
0.019
t i v e spec 1 e
RATIO
1.000
1 .000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.909
0.883
1.000
1.000
1.087
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
s
-------
TABLE 6-8. RECOVERY EFFICIENCY BLANK DIOXIN/FURAN
DATA FOR MARION COUNTY MM5 SAMPLES
ISOMER
DIOXIN
Mono-CDD
Di-CDD
Tri-CDD
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hepta-CDD
Octa-CDD
FURAN
Mono -CDF
Di-CDF
Tri-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Octa-CDF
RECOVERY
EFF.
BLANK
AMOUNT
(ng)
ND
ND
ND
2.55
0.00
ND
ND
ND
ND
ND
ND
0.28
0.62
ND
ND
ND
1.18
0.00
ND
ND
ND
ND
ND
ND
ND
ND
ND
EFFICIENCY
RUN 2 OF SAMPLE
VALUE RECOVERY
AMOUNT
(ng)
ND
0.28
0.37
1.37
1.90
ND
2.02
ND
ND
ND
29.98
26.27
26.50
6.03
3.68
87.16
17.08
15.41
ND
8.46
5.35
ND
ND
ND
6.07
ND
0.24
PERCENT
0.00
100.00
100.00
34.95
100.00
0.00
100.00
0.00
0.00
0.00
100.00
98.95
97.71
100.00
100.00
100.00
93.54
100.00
0.00
100.00
100.00
0.00
0.00
0.00
100.00
0.00
100.00
ND - Not detected.
a. Values reported in ng/sample for total analyses
b. The recovery efficiency blank was performed on
glassware used for Run 1. However, due to the
analytical problems with the Run 1 sample,
results from Run 2 are used for this calcu-
lation. The calculation is based on total
train results.
c. Percentage - 100% * Run 2/(Run 2 + Recovery
efficiency blank)
6-21
-------
blanked filter for the Soxhlet blank and a sample of HPLC water for the
aqueous blank. The results are presented in Table 6-9 and indicate that no
contamination occurred other than some values in the background noise region.
GDP/CDF Duplicate Analyses
Triangle Laboratories analyzed one flue gas sample (Run 2 back half)
and one ash sample (Run 3 baghouse ash) in duplicate. The quality control
requirements for a CDD/CDF duplicate analysis is a percent difference of + 50
percent based on the analytical precision for 95 percent of the analyses. The
results for the two samples are presented in Table 6-10. The duplicate
analyses meet these criteria except for two instances for the Run 2 back half
sample, where in one analysis the congener was not detected and in the second
analysis it was. Overall, based on total CDD/CDF results the difference was
less than 20 percent.
6.2.5 Quality Control for Particulate Testing
Particulate sampling was incorporated into the HC1 train (Runs 1-3) and
Pb/Cd train (Runs 4,5, and 6). The HC1 and Pb/Cd results are not reported
here, but the particulate fraction of those sampling efforts are reported.
Only quality control methods dealing with the particulate aspect of the two
trains will be considered in this section.
Particulate Equipment and Sampling Preparation.
Preparation for HCl/particulate sampling is described in the next
section, but it is important to note here, for the particulate QC, that tared
filters were used. Filters were tared in accordance with EPA Method 5 and
placed in sealed, precleaned, glass petri dishes prior to leaving for Marion
County, Oregon. The analytical balance used to tare the filters, and used in
later particulate analyses, was calibrated with standard weights (NBS Class
S). Measured values agreed within + 0.1 mg. The balance was calibrated prior
to making measurements. This calibration data can be found in the particulate
analysis logbook in Appendix D.4.
6-22
-------
TABLE 6-9. SUMMARY OF CDD/CDF SOXHLET AND AQUEOUS BLANK RESULTS
a\
ro
u>
ISOMER
DIOXIN
Mono-CDD
D1-CDD
Trl-CDD
2378 TCDD
Other TCDO
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
Hepta-CDD
Octa-CDD
TOTAL PCDD
FURAN
Mono-CDF
D1-CDF
Trl-CDF
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
Other HxCDF
Hepta-CDF
Octa-CDF
SOX. BLANK II
AMOUNT DETECT.3 RATIO
(ng) LIM.
SOX. BLANK §2
AQUEOUS BLANK
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.07
ND
0.003
0.005
0.009
0.069
0.008
0.013
0.013
0.019
0.018
0.021
0.019
0.028
1.000
1.000
1.000
0.125
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.889
1.000
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.003
0.004
0.006
0.005
0.005
0.006
0.010
0.008
0.010
0.010
0.012
0.011
0.014
0.023
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
AMOUNT
(ng)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.28
ND
ND
ND
0.09
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
DETECT.
LIM.
0.003
0.005
0.008
0.152
0.152
0.010
0.010
0.015
0.015
0.017
0.015
0.019
0.002
0.004
0.006
0.005
0.008
0.006
0.008
0.008
0.010
0.009
0.013
0.023
RATIO
1.000
1.000
1.000
0.303
0.303
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.857
1.000
1.000
1.000
0.765
0.765
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
AMOUNT
(ng)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.08
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
DETECT.
LIM.
0.004
0.006
0.009
0.065
0.065
0.014
0.014
0.020
0.019
0.022
0.021
0.137
0.003
0.004
0.007
0.006
0.006
0.007
0.011
0.009
0.010
0.011
0.013
0.012
0.014
0.022
RATIO
1.000
1.000
1.000
0.067
0.067
1.000
1.000
1.000
1.000
1.000
1.000
0.750
0.875
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
a Detection limits are not reported where native species
are not detected.
-------
TABLE 6-10. CDD/CDF DUPLICATE ANALYSES FOR MARION COUNTY MWC
RUNI3 BAGHOUSE ASH SAMPLE
AMOUNT DETECTED (NG)
RUN#2 BACKHALF FLUE GAS SAMPLE
AMOUNT DETECTED (NG)
TRIAL 1 TRIAL 2 AVERAGE DIFFERENCE (%) TRIAL 1 TRIAL 2 AVERAGE DIFFERENCE (7.)
MONO-CDD
DI-CDD
TRI-CDD
2378 TCDD
OTHER TCDD
12378 PCDD
OTHER PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
OTHER HxCDD
HEPTA-CDD
OCTA-CDD
Total PCDDs
MONO-CDF
DI-CDF
TRI-CDF
2378 TCDF
OTHER TCDF
12378 PCDF
23478 PCDF
OTHER PCDF
123478 HxCDF
123678 HxCOF
123789 HxCDF
OTHER HxCDF
HEPTA-CDF
OCTA-CDF
(0.001)
0.03
0.09
0.02
0.19
0.04
0.36
0.04
0.08
0.15
0.78
0.77
0.59
3.14
(0.001)
(1.30)
0.98
0.38
0.44
0.07
0.12
0.10
0.01
(0.002)
(0.017)
0.10
0.03
0.05
(0.001)
0.04
0.13
0.02
0.21
0.03
0.32
0.02
0.05
0.13
0.70
0.76
0.52
2.93
(0.001)
(1.34)
1.38
0.38
0.46
0.07
0.11
0.14
0.01
(0.001)
(0.017)
0.09
0.03
0.06
ND
0
0
0
0
0
0
0
0
0
0
0
0
3
ND
ND
1
0
0
0
0
0
0
ND
ND
0
0
0
.04
.11
.02
.20
.04
.34
.03
.07
.14
.74
.77
.56
.04
.18
.38
.45
.07
.12
.12
.01
.10
.03
.06
0
-29
-36
0
-10
29
12
67
46
14
11
1
13
7
0
0
-34
0
-4
0
9
-33
0
0
0
11
0
-18
(0.003)
0.28
0.37
0.22
0.07
(0.053)
2.02
(0.050)
(0.049)
(0.056)
3.25
1.77
1.76
9.74
0.31
3.68
8.26
2.20
2.03
(0.027)
1.55
4.74
(0.026)
(0.027)
(0.033)
2.25
(0.008)
0.24
(0
(0
(0
(0
(1
(0
(0
(0
(0
(0
.002)
0.26
0.39
0.22
0.05
.618)
1.28
.029)
0.21
.035)
3.69
1.71
1.78
9.59
0.39
1.8)
8.47
2.47
2.28
.016)
1.47
4.79
.015)
.016)
.020)
1.89
.011)
0.18
ND
0
0
0
0
ND
1
ND
0
ND
3
1
1
9
0
7
8
2
2
ND
1
4
ND
ND
ND
2
ND
0
.27
.38
.22
.06
.65
.13
.47
.74
.77
.67
.35
.74
.37
.34
.16
.51
.77
.07
.21
0
7
-5
0
33
0
45
0
-123
0
-13
3
-1
2
-23
-105
-3
-12
-12
0
5
-1
0
0
0
17
0
29
Total PCDFs
2.28
2.73
2.51
-18
25.26
21.94
23.60
14
-------
Particulate Sampling Operations
Sampling operations for the particulate sampling train were identical to
Method 5 sampling operations. HCl/Particulate and Pb/Cd/particulate sampling
results for isokinetics and leak checks are presented in Table 6-11 and 6-12,
respectively. None of the sampling trains were determined to have a final
leakrate above the maximum allowable leakrate (0.02 acfm) except for the last
port of Run 6. A sample volume correction
calculations are included in Appendix A.4.
port of Run 6. A sample volume correction of 0.9 ft was made and the
Particulate Sample Recovery
HCl/particulate sample recovery followed the standard Method 5 QC format.
With the particulate aspect of the HCl/particulate and Pb/Cd/particulate
sampling effort in mind, it is important to note here that the initial front
half washes were performed with acetone, as EPA Method 5 dictates.
Particulate Sample Analysis
The QC analyses for particulate involved evaluations of the recovery
efficiency, laboratory proof, and reagent particulate blanks. The sample
fractions evaluated were the nozzle and probe/filter holder acetone washes and
the filter. Table 6-13 shows the values for these fractions, and their
respective sums. All of the blanks contained measurable trace amounts of
particulate. Recovery efficiency blanks had the highest concentration of
particulate matter of any of the blanks; however, background particulates
found in the blanks will not significantly affect particulate analyses since
particulate mass collected during boiler outlet testing is typically at least
three orders of magnitude higher than the highest blank value.
6.2.6 Quality Control for HCl/Particulate Testing
HC1 and particulate samples were collected simultaneously in the same
train. The HC1 portion of the QC results are discussed in this section.
HCl/particulate tests were performed during Runs 1, 2 and 3.
6-25
-------
TABLE 6-11. HC1/PARTICULATE TESTING ISOKINETICS AND LEAK
CHECK SUMMARY BOILER OUTLET AT MARION COUNTY
Percent Sampling
Date Run # Isokinetics Leak Check Port
9/22/86 1 100.0 Initial
Port Change
Continue
Final
9/23/86 2 101.6 Initial
Port Change
Continue
Final
9/24/87 3 99.3 Initial
Port Change
Continue
Final
B
B
A
A
B
B
A
A
B
B
A
A
Leak Rate
(ft3/min)
0.011
0.005
0.005
0.000
0.013
0.013
0.011
0.015
0.015
0.012
0.010
0.013
Pressure
(in H20)
16
16
16
16
16
16
16
16
17
15
15
15
The QC objective for isokinetics was 100 + 10 percent.
The QC objective for leakchecks was a leak-free train or leakage rate less
than or equal to 0.02 cfm or less than 4% of the average sampling rate
(whichever is less) .
6-26
-------
TABLE 6-12.
Pb/Cd/PARTICULATE TESTING ISOKINETICS AND LEAK
CHECK SUMMARY BOILER OUTLET AT MARION COUNTY
Percent Sampling
Date Run # Isokinetics Leak Check Port
9/26/86 4 98.8 Initial
Port Change
Continue
Filter Change
Continue
Final
9/29/86 5 97.8 Initial
Port Change
Continue
Final
9/30/87 6 98.2 Initial
Port Change
Continue
Final
B
A
B
B
B
B
B
B
A
A
B
B
A
A
Leak Rate
(ft3/min)
0.010
0.012
0.012
0.013
0.017
0.014
0.010
0.013
0.013
0.013
0.014
0.007
0.012
0.025C
Pressure
(in H20)
_ .
15
15
16
16
17
16
12
15
12
15
13
15
8
The QC objective for isokinetics was 100 + 10 percent.
DThe QC objective for leakchecks was a leak-free train or leakage rate less
than or equal to 0.02 cfm or less than 4% of the average sampling rate
(whichever is less).
'0.9 ft sample volume correction was required to compensate for leakage at
the end of Run 6.
6-27
-------
TABLE 6-13. SUMMARY OF UNCONTROLLED PARTICULATE BLANK RESULTS
ACETONE
ACETONE
PROBE RINSE FREE RINSE
OA
1
IM
CX>
TRAIN AND
BLAN< TYPE
HO. SAM1_IN3
ACETONE RE/GENT
LAB PROOF
RECOVER!1 EFF.
Pb/Cd SAMTJN3
ACETONE RE/CENT
LAB PROOF
RHDVERY EFF.
FILTER PWnCLLATE
(G) (G)
0.0005
0.0000 0.0086
0.0105
0.0006
0.0001 0.0026
0.0076
\O_l>E
(M.)
435
165
490
260
150
525
ACETONE
BLAN<
CORRECTION
FACTOR
(G/M.)
l.OE-06
l.CE-06
l.OE-06
3.0E-06
3.0E-06
3.0E-06
ACETONH
ELAN<
CORRECTION
(G)
_
0.0002
0.0005
-
0.0005
0.0016
1
CORRECTED
ACETONE
FRCEE RINSE
PAKTiaLATE
(G)
_
0.0088
0.0101
_
0.0021
0.0060
ACETCIt ACETOe ACO>£ BLAN<
NOZZLE RINSE NOZZLE RINSE OORRBCTICN
PARTIOLATE VttOt
(G) (M.)
_
NOZZLE WA9CS ARE IHCLUDS)
NOZZLE HA9£S ARE INaiEffi
_
0.0011 120
NOZZLE WASFES ABE INailH)
ACETOhE
BLANX
FACTOR CORRECTION
(G/M)
_
INTtf FRONT
IN THE FRNT
_
3.0E-06
INTV£ FRONT
(G)
.
HALF
HtiF
_
0.0004
HALF
1
CDRRECTFD ACETONE
ACETONE FREE AM)
NOZZLE RINSE NOZZLE RINSES
PARTICULATE PAKTiaiAU
(G) (G)
_
0.0083
0.0101
_
0.0007 0.0028
WSS OF
TOTAL 2
HARnOJLATE
(G)
_
0.0083
0.0101
_
0.0029
0.0060
1) OORRECTH) ACETONE WEIGHT = UKDRFECTH) ACETDN5 WEIGHT (G) - [VCLU€ ACETONH (M.) X ACETONE ELANX CORRECTION FACTOR (CVM.)]
2) IWLLCES NOZZLE AN3 FREE RINSES + FILTER MASSES.
-------
HC1 Equipment and Sampling Preparation
Calibrations and/or inspections were made on all equipment prior to
sampling, as with CDD/CDF test preparation. Sample train glassware and
high-density polyethylene sample bottles were precleaned as described in
Section 5. All cleaned glassware was then sealed with glass plugs or parafilm
to prevent contamination.
HC1 Sampling Operations
Sampling operations for the HC1 train followed EPA Method 5 except that
0.1 N NaOH replaced the water in the impingers and an empty impinger was
inserted as the first impinger.
A summary of isokinetics and leak checks was previously presented in
3
Table 6-11. All leak checks were below 0.02 ft /min, which satisfies the QC
requirement. Also, isokinetics were within the QC requirement of 100% + 10%.
HC1 Sample Recovery
Sample recovery for the HCl/particulate train followed the procedure
outlined in Chapter 5. Standard Method 5 Quality Control measures were taken
with the following modifications:
All train components in contact with the sample, except the nozzle,
were made of glass or teflon.
All instruments used in recovery were either stainless steel or
teflon, and had been cleaned by the HC1 cleaning procedure.
Recovery efficiency, laboratory proof and NaOH reagent blanks were
taken.
Each sample container lid was individually sealed with teflon tape
to prevent leakage.
Each container lid was covered with an integrity seal (strip of
labeled tape) to prevent tampering.
Samples were weighed prior to and after shipping to indicate
possible sample loss.
j^
All sample containers were packaged for transport in Ziplock bags,
wrapped in bubble wrap, and placed in a second Ziplock bag. Sample
containers were then packed and shipped in an upright position.
6-29
-------
HC1 Analysis
The HC1 samples were analyzed via ion chromatography. The acetone rinses
and filters were first analyzed for particulate matter, but were then
resuspended in 0.1 N NaOH and combined with the front half NaOH rinse.
Although both the front half and the back half fractions were analyzed for
chloride ions, only the back half was used to determine the HC1 flue gas
concentration. Any chloride ions present in the front half were assumed to be
salts associated with the particulate. Each analysis was performed twice and
the average of the two values was reported.
Presence of chloride ions was not detected in either the laboratory proof
blank, the recovery efficiency blank, or the reagent blank. The analytical
detection limit for the ion chromatographic analysis was 1 ug/ml. Table 6-14
lists these results.
The Radian Laboratory also analyzed two audit samples. The results of
the audit sample analysis are presented in Table 6-15. Known concentrations
of 35 ug/ml and 120 ug/ml were chosen because they represented characteristic
expected sample concentration ranges. Each analysis was within 10% of the
true value with the most severe deviation being 8.9 percent.
6.2.7 Quality Control for Pb/Cd/Testing
Lead, cadmium and particulate samples were collected simultaneously in
the same train. The lead and cadmium portion of the QC results are discussed
in this section.
Equipment and Sampling Preparation
Calibrations and/or inspections were made on all equipment prior to
sampling, as with CDD/CDF and HC1 test preparation. Sample train glassware
and high-density polyethylene sample bottles were cleaned using the same
procedure as HC1 precleaning except that 0.1 N nitric acid replaced the NaOH
and no aluminum foil was used to cap the glassware. Instead, parafilm or
rubber caps were used where glass caps were not appropriate.
6-30
-------
TABLE 6-14.
SUMMARY OF UNCONTROLLED HCL BLANK RESULTS
FOR THE MARION COUNTY MWC
BLANK
RADIAN
FIELD
NUMBER
TRIAL
NUMBER
SAMPLE
VOLUME
(ML)
1,2
DILUTION
FACTOR
(WATER/SAMPLE)
ANALYTICAL
READING
(UG/ML)
BLANK
TOTAL
(UG/SAMPLE
LAB PROOF BLANK
FRONT HALF
cr\
I
BACK HALF
MAR-16
MAR-18
1
2
1
2
1000
1000
UNDILUTED
UNDILUTED
UNDILUTED
UNDILUTED
ND
ND
ND
ND
ND
ND
ND
ND
RECOVERY
EFFICIENCY BLANK
FRONT HALF MAR-64
BACK HALF MAR-65
1
2
1
2
1000
1000
UNDILUTED
UNDILUTED
UNDILUTED
UNDILUTED
ND
ND
ND
ND
ND
ND
ND
ND
REAGENT BLANK
0.1 N NaOH
MAR-2
1
2
1000
UNDILUTED
UNDILUTED
ND
ND
ND
ND
1) ND = NOT DETECTED
2) DETECTION LIMIT FOR HCL ANALYSIS BY 1C IS 1 UG/ML.
-------
TABLE 6-15. SUMMARY OF HC1 AUDIT SAMPLE RESULTS MARION COUNTY
SOLID WASTE-TO-ENERGY FACILITY, BOILER OUTLET
Absolute
Radian No. Audit Sample Measured Percent Difference
and True Value Value of True Between
Trial No. (ug/mL) (ug/mL) Value Trials (%)
436
Trial 1 35.0 33.2 94.9 4.0
Trial 2 35.0 31.9 91.1
437
Trial 1 120.0 117.9 98.3 0.8
Trial 2 120.0 118.8 99.0
6-32
-------
Pb/Cd Sampling Operations
Pb/Cd sampling operations followed standard Method 5 procedures with the
following modifications:
1) Only ground glass caps, parafilm, or rubber caps were used to cover
sampling train components during assembly, leak checks, and
disassembly of the sampling train.
2) The sampling rate averaged 0.56 dscfm.
3) All train components and sample bottles had been marked previously
according to the procedure used.
Table 6-12 previously summarized the isokinetics and leak check results
during the three Pb/Cd/particulate sampling runs. The final leakrate during
3 3
Run 6 was 0.025 ft /min which was above the prescribed 0.02 ft /min and
3
required a sample volume correction of 0.9 ft . The volume correction
calculation is presented in Appendix A.4.
Pb/Cd Sample Recovery
Pb/Cd/particulate sample recovery followed the procedure outlined in
Chapter 5. In addition to the general recovery QC procedures, the following
QC procedures were also implemented:
All train components that were in contact with the sample, except
the nozzle, were made of glass or teflon. Nozzle washes were
recovered into separate sample bottles to prevent the possibility of
contaminating the front half metals sample. All instruments used in
the recovery process were either teflon or teflon coated. All
instruments which came in contact with the sample or the sampling
surfaces were cleaned according to the metals/particulate cleaning
procedure.
Reagent, laboratory proof, and recovery efficiency blanks were
taken.
Each sample container lid was individually sealed with teflon tape
to prevent leakage.
Each container lid was covered with an integrity deal (strip of
tape) to prevent tampering.
Samples were weighed prior to and after shipping to indicate
possible sample loss.
6-33
-------
Tl
All sample containers were packaged for transport in Ziplock bags,
wrapped in bubble wrap, and placed into a second Ziplock bag.
Nitric acid rinses were placed in poly-propylene bottles.
Acetone rinses were placed in glass bottles.
Pb/Cd Analysis
Pb/Cd train samples were analyzed by Atomic Absorption for the presence
of lead and cadmium. Lab proof blanks and recovery efficiency blanks were
also analyzed as part of the QC program. The laboratory proof blank sample
was analyzed as front half and back half fractions for the purpose of EPA
investigation of the test protocol. The front half fraction consisted of
probe rinses, cyclone rinses, and front half filter rinses. The back half
fraction consisted of the back half filter rinses and impinger rinses.
Table 6-16 shows the blank results. Lead and cadmium were detected in
significant levels in the recovery efficiency train; however, they were not
detected in any of the other blanks. The recovery efficiency blank was
performed after Run 5. However, since the recovery efficiency blanks consist
of one percent or less of the amount detected in the total trains, the
sample recovery procedure is considered acceptable.
Audit samples prepared by RTI were analyzed in order to verify the
integrity of the atomic absorption analysis. In each case, the results were
within ten percent of the true value. These results are presented in
Table 6-17. During the audit sample analysis, preliminary results indicated
that it was necessary to add HF to the calibration standards. Otherwise, the
results were suppressed and would not be within the required range. The
preliminary results can also be found in Table 6-17 and a discussion of the
problem in Appendix F.3.
Filters were spiked with lead and cadmium for use as internal laboratory
QA samples. The spiked filters were carried through the digestion procedures
to determine the efficiency of the digestion process and analysis. As shown
in Table 6-18, the average percent recovery for lead was 99.5% and the average
percent recovery for cadmium was 92%.
-------
TABLE 6-16. LABPROOF BLANK AND RECOVERY EFFICIENCY BLANKS FOR Pb/Cd TRAIN
Lab Proof Blank3
uE/samDle
Front Back
Element Half Half
Total
Train
Recovery Efficiency Blank
ug/ sample
Total Train
Lead
Cadmium
ND [3.6] ND [1.8] ND
ND [1.6] ND [0.8] ND
19.60
2.10
ND = Not detected. Minimum detection limits are indicated in brackets based
on the following analytical minimum detection limits and 200 ml of sample:
Pb - 0.018 ppm
Cd - 0.008 ppm
Total train analysis only.
6-35
-------
TABLE 6-17. RESULTS OF Pb/Cd AUDIT SAMPLE ANALYSIS FOR MARION COUNTY MWC
EPA Audit Digestion
Sample # Procedure Metal
i
u>
a\
EPA-9C
High 57C
Low 59C
High 30C
Low 17°
EPA-3d
EPA-4d
a
a
a
f
f
b
b
Pb
Pb
Pb
Pb
Pb
Cd
Cd
Spiked
Value
(ug)
500
1000
500
1000
500
160
480
True^
Value
(ug)
465
995
465
995
465
161
477
Initial
Analysis
without HF
Range in Standard
(450-550)
(900-1100)
(450-550)
(900-1100)
(450-550)
(147-173)
(442-518)
327
716
367
598
345
158
486
Percent Final
of Analysis
True With HF in
Value Standard
70 506
72 e
79 e
60 e
74 e
98
102
Percent
of True
Value
109
e
e
e
e
-
a 18
Digestions were performed as described in Test Plan.
Dilution of ampule. HF not included in standard.
Spiked on filter.
Spiked in solution.
eSamples were spilled before the final analysis could be performed.
Nitric digestion only.
^This value represents the mean recovery based on six determinations at the end of a 90 day stability
study.
-------
TABLE 6-18. INTERNAL LABORATORY QA SAMPLES FOR Pb/Cd
ANALYSES BY ATOMIC ABSORPTION
Radian
Filter #
1
2
3
4
Average
Theoretical Analytical Recovery
True Value (ppm) Value (ppm) (%)
Pb Cd Pb Cd Pb Cd
2.00 2.00 1.95 1.82 98 91
2.00 2.00 2.06 1.83 103 92
2.00 2.00 1.99 1.82 99.5 91
2.00 2.00 1.95 1.86 98 93
2.00 2.00 1.99 1.83 99.5 92
Filters were spiked with 0.2 ml of 1000 ppm Pb and 0.2 ml of
1000 ppm Cd. Samples were carried through digestion procedures,
6-31
-------
6.2.8 Quality Control for Cr/Ni Testing
Cr/Ni Equipment Preparation
Equipment preparation was identical to the preparation procedures
followed for the HCl sampling train as discussed in Section 6.2.6.
Cr/Ni Sampling Operation
Operation of the Cr/Ni sampling train was identical to the sampling
operations for the HCl train except for the following:
0.5 N NaOH replaced the 0.1 N NaOH in the impingers, and
The train did not have a glass fiber filter section.
Some sodium carbonate is formed during operation as carbon dioxide in the
flue gas reacts with the sodium hydroxide. To correct for this carbon dioxide
loss, 1.5 dscf is added to each sampled volume. This adjustment was
calculated based on the expected volume of CO- that would be collected during
the sampling period. Leak checks and isokinetics are summarized in
Table 6-19.
Cr/Ni Sample Recovery
Sample recovery for the Cr/Ni train followed the procedure given in
Chapter 5. Standard Method 5 quality control measures were taken with the
following exceptions:
Not more than 200 ml of 0.1 N NaOH were used to recover both the
probe liner and impingers.
Following recovery, the alkalinity of the samples was adjusted to a
pH of 8.
All instruments used in recovery were either teflon or teflon
coated.
Each sample container lid was individually sealed with teflon tape
to prevent leakage.
Each container lid was covered with an integrity seal strip to
prevent tampering.
Samples were weighed prior to and after shipping to indicate
possible sample loss.
6-38
-------
TABLE 6-19. Cr/Ni TESTING ISOKINETICSa AND LEAK CHECK.b
SUMMARY BOILER OUTLET AT MARION COUNTY
Percent Sampling
Date Run # Isokinetics Leak Check Port
9/26/86 4 101.6 Initial
Port Change
Continue
Final
9/29/86 5 101.0 Initial
Port Change
Continue
Final
9/30/87 6 101.6 Initial
Port Change
Continue
Final
B
B
A
A
B
B
A
A
B
B
A
A
Leak Rate
(ft3/min)
0.009
0.011
0.017
0.008
0.011
0.016
0.000
0.005
0.009
0.010
0.012
0.006
Pressure
(in. H20)
_ _
15
15
15
15
15
15
15
15
—
aThe QC objective for isokinetics was 100 + 10 percent.
The QC objective for leakchecks was a leak-free train or leakage rate less
than or equal to 0.02 cfm or less than 4% of the average sampling rate
(whichever was less).
-- indicates that data were not recorded.
6-39
-------
13
All sample containers were packaged for transport in Ziplock bags,
wrapped in bubble wrap, and placed in a second Ziplock bag.
Cr/Ni Analysis
Cr/Ni train samples were analyzed by Atomic Absorption for nickel and
trivalent chromium. Analysis for hexavalent chromium was done by colorimetric
techniques. The sum of the trivalent chromium and hexavalent chromium
concentration is reported as total chromium.
Lab proof, recovery efficiency, and reagent blanks were analyzed as part
of the QC program. These results are presented in Table 6-20. Chromium
(+III) was found in the lab proof blank, 0.5 N NaOH reagent blank and the
0.1 N NaOH blank. Since the NaOH reagents are used in the recovery efficiency
blank, chromium would also be expected in the recovery efficiency blank.
However, chromium was not detected which indicates that the contamination may
have occurred during the analytical procedure rather than being background
levels in the reagents. Chromium (+VI) was not detected in any of the blanks.
Nickel was present in the lab proof blank, recovery efficiency blank, and the
0.1 N NaOH reagent blank. Nickel showed the same inconsistent results as
found with the chromium (+III). Also, the nickel results are close to the
detection limit where the analytical precision is worst.
Additional quality control measures included the analysis of audit
samples and internal laboratory QA samples. Audit samples served as a test
for the analysis integrity, while internal lab samples were used to determine
the efficiency of the digestion and analysis procedures. Summaries of audit
sample and internal lab sample results are shown in Tables 6-21 and 6-22,
respectively. Audit samples were all within six percent of the true values.
During the audit sample analysis, preliminary results indicated that it was
necessary to add HF to the calibration standards. Otherwise, the results were
suppressed and would not be within the required range. The preliminary
results can be found in Appendix F.3. Average percent recoveries in the
laboratory QA samples for chromium (+VI), chromium (+III), and nickel were
67%, 168.5%, and 97.8% respectively.
6-Uo
-------
TABLE 6-20. LABORATORY PROOF BLANK, RECOVERY EFFICIENCY
BLANK AND REAGENT BLANK RESULTS FOR Cr/Ni TRAIN
Recovery HO 0.5 N NaOH 0.1 N NaOH
Lab Proof Efficiency Reagent Reagent Reagent
Element Blank Blank Blank Blank Blank
(ug/sample) (ug/sample) (ug/sample) (ug/sample) (ug/sample)
Chromium (+III) 50.60 ND [0.4] ND [0.4] 2.25 34.50
Chromium (+VI) ND [20] ND [20] ND [20] ND [20] ND [20]
Nickel 5.18 8.58 ND [2.4] ND [2.4] 10.40
ND = Not detected. Minimum detection limits are indicated in brackets based
on the following analytical minimum detection limits and 100 ml of
sample: Ni = 0.024 ppm
Cr (+III) = 0.004 ppm
Cr (+VI) =0.2 ppm
6-Ui
-------
TABLE 6-21. RESULTS OF Cr/Ni AUDIT SAMPLE ANALYSIS FOR MARION COUNTY MWC
EPA Audit
Sample #
EPA-100
-10
-10
Low 17
i
K> Low 17
High 30
High 30
Digestion
Procedure Metal
b
b
b
d
d
d
d
Cr (Vi:
Cr
Ni
Ni
Cr
Ni
Cr
Spiked
Value
(ug)
) 0
250
2500
1250
125
2500
250
True
Value
(ug)
0
252
2406
1172
118
2406
252
Initial
Analysis
without HF
Range in Standard
-
(225-275)
(2250-2750)
(1125-1375)
(113-137)
(2250-2750)
(225-275)
ND
192
2390
1430
153
2720
307
Percent Final
of Analysis
True With HF in
Val ue Standard
ND
76 265
99 2390
122 e
61 e
113 e
122 e
Percent
of True
Value
-
105
99
e
e
e
e
Audit samples were supplied by Research Triangle Institute, Inc. A more detailed discussion of the
audit analyses, problems, and results can be found in Appendix F.3.
Digestions were performed as described in Test Plan.
"Spiked on filter.
Nitric acid digestion.
3
'Samples were not reanalyzed.
-------
TABLE 6-22. INTERNAL LABORATORY QA SAMPLES FOR Cr/Ni ANALYSIS BY AA
Theoretical
Radian True Value (ppm)
Filter # Cr (VI)a Cr(III)a
1 2.00 8.61
2 2.00 6.32
3
4
Analytical Value (ppm) Recovery (%)
Ni
2.00b
2.00b
2.00°
2.00°
Cr (VI) Cr(III)
1.28 14.74 1
1.40 10.51 2
1
2
Ni Cr (VI) Cr(III)
.11 64 171
.72 70 166
.95
.02
Ni
56
136
98
101
aOnly two Cr (III) and Cr (VI) filters were spiked.
Nickel results by Cr (III) digestion.
Q
Nickel results by Cr (VI) digestion.
NOTE: Analytical Report is included in Section 1.3.
6-U3
-------
6.2.9 Quality Control for Continuous SO. Determination
A TECO Model 40 S09 analyzer was used in conjunction with a Beckman 755
0 analyzer and Orsat C0? data to determine S0? concentration. The presence
of carbon dioxide and oxygen caused quenching of the SO response in the
analyzer. Therefore, the equation below, provided by the manufacturer of the
SO analyzer, was used to correct for quenching via a Radian computer program:
NEWS02 = KF x OLDS02
where
KF = 1 + 0.02139 x 0 + 0.0143 x C02 = Quenching Factor
NEWS02 = S09 concentration (ppmv)
OLDS02 = observed SO concentration (ppmv)
0- = corresponding oxygen concentration (%V)
C09 = average carbon dioxide concentration (%V) determined by EPA
Method 3
EPA Method 6C was used as a guideline for the S02 testing, and the
following quality control measures from Method 6C were used at Marion County:
Continuous SO. System Leak Checks
The CEM sampling system was leak-checked periodically by introducing an
oxygen free gas at the sample line or at the manifold. Any oxygen measured
was considered to be proportional to the system leakage. An oxygen
concentration of 0.5 percent or less was required for QC purposes. Table 6-23
presents the leak check results. Manifold leak checks test the manifold and
gas conditioner for leaks. Line leak checks test the entire system.
Continuous SO. Multipoint Calibration
z
Prior to sampling at the Marion County facility, a linearity check was
performed on the SO analyzer. Four certified calibration gases were used: a
zero, low-range, mid-range, and a high-range of the span. The acceptance
criterion for the linearity checks was a correlation coefficient (r) of
r > 0.9950. A correlation coefficient of r = 0.99925 was observed which was
well within the minimum quality control requirements. Figure 6-2 graphically
depicts the results.
-------
TABLE 6-23. LEAK CHECK SUMMARY FOR MARION
COUNTY S02 CEM SYSTEM
Line Check
Average % 0~
Run 1 0.6
Run 2 a
Run 3 0.2
Manifold
Average
0.0
0.0
0.0
Check
%o2
a
Line leakcheck was not performed for
Run 2.
-------
500-
®
-£- 400-
Q.
Q.
c
o
+-»
CO
*^
c
0)
o
c
o
300-
200-
CO
•3
•4-^
O
100-
100 200 300 400
Measured Concentration (ppm)
500
Actual
(ppmv SO2)
(y)
0
95.2
239
505
Measured
(ppmv SO2)
(x)
0
101.8
251.5
494.6
Figure 6 - 2. Multipoint GEM Calibrations for Marion County
cc
o
o
f^
-------
Continuous SO Daily Calibrations
At the beginning of each test, the CEM operator performed a two point
calibration using a zero gas (nitrogen) and a span gas, making adjustments as
necessary. A calibration check in which no adjustments were made followed
each test. Daily calibrations are summarized in Table 6-24. Daily drift
requirements according to Method 6C, for both zero and span were + 2% of span
for each run. The SOQ analyzer met this criterion for all runs. The oxygen
analyzer had one run slightly above 2% drift for the span. Results are
summarized in Table 6-25.
Continuous SO- Daily Quality Control Check
2
Daily calibration checks were made after the initial calibration with
mid-range standard gases. The concentration used for S0_ daily quality
control checks was 230 ppmv. The acceptance criteria for the daily quality
control standard determination was agreement within + 10% of the input
concentration. No instrument adjustments were allowed. The results of the QC
checks appear in Figure 6 - 3.
Continuous SO. Sampling System Bias Check
2
The S09/09 CEM operator performed the sampling system bias check on
September 24, 1986 with the high range (949 ppm S02) calibration gas. The
difference between the concentrations at the analyzer and the sample line
inlet was 24.5 ppm or 2.45% of the span. This is an acceptable difference
under EPA Method 6C; however, this method technically requires the use of a
mid-range calibration gas. A mid-range calibration at 239 ppm was performed
at the sample line inlet at 9:24 am on September 24, 1986. There was,
however, no calibration check made at 239 ppm S02 at the inlet manifold to the
instrument on that day.
For this reason, a comparison was made to a calibration check performed
at 10:16 am on September 23, 1986, which was introduced at the instrument
manifold. The observed concentrations for the manifold and sample line inlet
were 278.3 ppm SO and 240.7 ppm SO respectively. The difference between
these values is 37.6 ppm which is 3.8 % of the span. While this value is
6-UT
-------
TABLE 6-24. SUMMARY OF SO AND 0 INSTRUMENT CALIBRATIONS
Initial Final Response Instrument Calibration
Date Run Parameter Response Factor Factor Range Gas
9/22/86
S02 (ppmv)
0 (% vol)
7.870433
0.1970443
8.072487
0.1988661
0-500
0-25
zero, 505
zero, 19
9/23/86 2
S02 (ppmv)
00 (% vol)
7.90504
0.198123
8.0703
0.2026487
0-1000
0-25
zero, 949
zero, 19
9/24/86
S02 (ppmv)
00 (% vol)
7.903393
0.1968232
8.000563
0.1913394
0-1000
0-25
zero, 949
zero, 19
6-U8
-------
TABLE 6-25. SUMMARY OF SO AND 0 GEM DRIFT CHECK
RESULTS FOR MARION COUNTY, OREGON
Zero Drift
Test
Date
09-22-86
09-23-86
09-24-86
09-22-86
09-23-86
09-24-86
Test
Run
01
02
03
01
02
03
Parameter
°2
°2
°2
so2
so2
so2
Check
Span Drift
Check
Instrument drift Meets, Instrument drift Meets,
(% of span)a QC? (% of span)a QC?
-0
0
0
0
-0
0
.01
.03
.01
.00
.01
.01
Yes
Yes
Yes
Yes
Yes
Yes
0.
-1.
2.
-1.
-2.
1.
68
67
19
10
00
15
Yes
Yes
No
Yes
Yes
Yes
a
Instrument drift is defined as the difference between the instrument response to
the input concentration at the beginning and end of the test run. It is
expressed in this table as a percent of the span.
QC criterion for daily instrument drift was 2% of span.
6-H9
-------
290
c:
o
c
o
o
T3
01
ro
OJ
C
ro
0)
200
+ 10 %
Input
Concentration
-10 %
Day 1 Day 2 Day 3
Date
Input Cone.
Measured
Cone.
1
2
3
Mean Cone.
CV
9/22/86
239
251.4
251.6
251.0
251.3
0.001
9/23/86
239
278.4
278.8
278.4
278.5
0.001
9/24/86
239
243.8
240.7
238.4
241.0
0.011
Run 1: 0 - 500 range on instrument
Runs 2 and 3: 0 - 1000 range on instrument
Figure 6-3. Daily Quality Control Check
6-50
-------
higher than the recommended 3.0%, it should be noted that instrument
adjustments had been made between the two points. The data for these results
can be found in the GEM strip charts (Appendix B.5) and GEM printouts
(Appendix B.3).
Response Times
Response times for the S0? and 0? analyzers were checked as part of the
GEM QC procedure; the results are shown in Table 6-26. Oxygen response time
for obtaining 95% of the ultimate concentration lagged the SO^ monitors
response by 25 seconds. Since the CEM/computer interface reads 180 responses
per minute and reports averages each minute, this lag will have only a minor
effect on SO^ quenching calculations, and causes no other significant effects.
6.2.10 QC Procedures for Process Sampling
Two types of ash samples for CDD/CDF and metals analysis were taken at
the Marion County facility: cyclone ash and baghouse ash. Lime slurry and
Tesisorb samples were also collected for metals analysis. Quality control
measures for these samples were identical.
Process Sampling Equipment and Preparation
The sampling vessels and amber glass bottles designated for CDD/CDF ash
sampling were cleaned according to the standard CDD/CDF cleaning procedures
as discussed in Section 5.1.2. The high-density polyethylene containers used
for the metals samples were cleaned with nitric acid rinses.
Sampling Operations
During the four-hour to six-hour flue gas sampling periods, ash samples
of equal volume were taken each hour. At the end of the test the samples were
composited and a 500 ml aliquot was taken. The baghouse ash sample was taken
at the bottom of the baghouse ash screw conveyor. Likewise, the cyclone ash
was sampled at its screw conveyor. After each sample was taken, it was
covered and the date, time, and comments were recorded.
6-51
-------
TABLE 6-26. CEM RESPONSE TIMES AT MARION COUNTY
SO (§95.2 ppm
85%
90%
95%
100%
Response
Time
30 sec
45 sec
60 sec
180 sec
02 @ 19%
..
90%
95%
100%
Response
Time
..
64 sec
85 sec
180 sec
K.esponse time checks were performed on 9/23/86.
5-52
-------
One grab of lime slurry and Tesisorb, each, were collected during Runs 4,
5 and 6. The grab samples for each run were composited into a single lime
slurry and single Tesisorb sample. The lime slurry and Tesisorb samples were
analyzed for the full spectrum of metals by NAA.
Analysis
Baghouse and cyclone ash samples from Runs 1, 2 and 3 were analyzed for
CDD/CDF according to the same procedure as the front half fraction of CDD/CDF
flue gas samples. Results for surrogate and internal standards were presented
previously in Tables 6-3 and 6-4 respectively. The baghouse and cyclone ash
samples from Runs 4, 5 and 6 were analyzed for hexavalent and trivalent
chromium following the same procedures as the flue gas samples.
As part of an EPA in-house study, extracts from the chromium analysis
along with the lime slurry and Tesisorb samples were analyzed by NAA for the
full spectrum of metals. EPA audit samples were submitted along with the
samples along with blanks.
6-53
-------
7.0 REFERENCES
1. Zurlinden, Ronald A., Henry P. Von Dem Fange and Jeffrey L. Hahn (Ogden
Projects, Inc) Compliance with Permit Conditions for Marion County Solid
Waste-to-Energy Facility: September 22 - October 8. 1986. Prepared for
Ogden Martin Systems of Marion, Inc. November 21, 1986. Report
Number 107.
2. Zurlinden, R.A., H.P. Von Dem Fange and J.L. Hahn (Ogden Projects, Inc.)
Engineering Data: Heavy Metal Emissions for the Marion County Solid
Waste-to-Energy Facility. Prepared for Ogden Martin Systems of Marion,
Inc. Emeryville, California. Report number 117- January 6, 1987.
3. Memorandum from Hahn, Jeffrey L., Ogden Projects, Inc., to Riley, C.
Gene, U.S. EPA. March 20, 1987. Addendum to Report No. 117.
4. Reference 1.
5. Reference 2.
6 . Reference "i.
7. Sampling for the Determination of Chlorinated Organic Compounds in Stack
Emissions: Draft Protocol. Prepared by Environmental Standards Workshop.
Sponsored by the American Society of Mechanical Engineers of the U.S.
Department of Energy and the U.S. Environmental Protection Agency.
December 31, 1984.
8. Draft EPA Method for the Determination of Cadmium from Stationary
Sources. Prepared by the U.S. Environmental Protection Agency. Research
Triangle Park, North Carolina.
9. EPA Protocol for Emissions Sampling for Both Hexavalent and Total
Chromium. Prepared by the U.S. Environmental Protection Agency.
Research Triangle Park, North Carolina. February 22, 1985.
10. Analytical Procedures to Assay Stack Effluent Samples and Residual
Combustion Products for Polvchlorinated Dibenzo-p-Dioxins (PCPD) and
Polychlorinated Dibenzofurans (PCDF). Prepared by Environmental
Standards Wowkshop. Sponsored by the American Society of Mechanical
Engineers of the U.S. Department of Energy and U.S. Environmental
Protection Agency. December 31, 1984.
11. Letter Report from Katherine L. Wertz and William P. Gergen, Radian
Corporation to C. E. Riley, U.S. Environmental Protection Agency. Metals
Analysis bv NAA Results for the Marion County Emissions Test in
1986.
7-1
-------
12. National Incinerator Testing and Evaluation Program (NITEP) P.E.I.
Testing Program: Volume II. (Concord Scientific Corporation). Prepared
for Environmental Canada. July 1985. Table 6.2.2.
13. Procedures for Estimating Risks Associated With Polychlorinated
Dibenzofurans (CDD and CDF). Prepared by the U.S. Environmental
Protection Agency, Washington, D.C. April 1986.
14 Natrella, Mary Gibbons, Experimental Statistics. National Bureau of
Standards Handbook. 1963. pp 17-2 to 17-3.
15. Reference 13.
16. R.T. Shigehara, R.M. Neulicht and U.S. Smith, "Validating Orsat Analysis
from Fossil-fuel-fired Units" Stack Sampling Technical Information: A
Collection of Monographs and Papers. Volume I. United States
Environmental Protection Agency. EPA-450/2-78-042a. October 1978.
17 Reference 12.
18 Jamgochian, C.L., W.E. Kelly and D.J. Holder (Radian Corporation).
Revised Sampling and Analytical Plan for the Marion County Solid
Waste-to-Energy Facility Boiler Outlet. EPA Contract No. 68-02-4338.
DCN: 86-222-124-02-05. September 16, 1986.
7-2
-------
8.0 ENGLISH TO METRIC CONVERSION TABLE
Metric
English
0.028317 dscm
0.028317 dscnm
0.45359 kg/hr
1 ng/dscm
1 mg/dscm
°F
101325 Pa
1 ng/kg
1 ng/g
1 dscf
1 dscfm
1 Ib/hr
4.3699 x 10 "" grains/dscf
4.3699 x 10"4 grains/dscf
<°C x 9/5) + 32°F
10
1 atm
-9
6.9998 x 10 grains/lb
6.9998 x 10"6 grains/lb
6.9998 x 10"3 grains/lb
3-1
------- |