EPA-450/4-84-014h
National DIoxin Study Tier 4
       Combustion  Sources

  Engineering Analysis Report
                   By
               Radian Corporation
           Research Triangle Park, NC 27709
             Contract No. 68-02-3889
        U.S. ENVIRONMENTAL PROTECTION AGENCY
             Office Of Air And Radiation
        Office Of Air Quality Planning And Standards
           Research Triangle Park, NC 27711

                September 1987

-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U.S.
Environmental. Protection Agency, and approved for publication as received from the contractor.
Approval does not signify that the contents necessarily reflect the views and policies of the Agency,
neither does mention of trade names or commercial products constitute endorsement or
recommendation for use.
                                EPA-450/4-84-014h

-------
                               TABLE OF CONTENTS
Chapter                              Title                                Page

  1.0   EXECUTIVE SUMMARY	1-1

  2.0   BACKGROUND	.	2-1
        2.1  The National Dioxin Strategy	! ! 2-1
             2.1.1  Purpose of the National  Dioxin Strategy 	 2-1
             2.1.2  Management of the National  Dioxin Strategy	2-2
             2.1.3  Tier 4:  Combustion Sources.	2-2
        2.2  Tier 4 Overview.	 2-3
        2.3  Background Information on Chlorinated Dibenzo-p-Dioxins
             and Chlorinated Dibenzofurans	     2-5
             2.3,1  Structure	.  .	.2-5
             2.3.2  Nomenclature Used in This Report. .	2-6
             2.3.3  Report Organization  	 2-9

  3.0   LITERATURE REVIEW	3-1
        3.1  Overview of the Literature  Data Base	!  !  ! . ! 3-2
             3.1.1  Summary of Stack PCDD and PCDF Emissions.  .  ... ! ! 3-3
             3.1.2  Summary of PCDD and  PCDF in Ash Emissions	 3-9
        3.2  Emissions Data for Individual Source Categories.  .  	 3-15
             3.2.1  Municipal  Solid Waste Incinerators	 3-15
                    3.2.1.1   United States and  Canada .  .  .  .	3-17
                    3.2.1.2  Europe . .	3-18
                    3.2.1.3   Japan	 3-18
             3.2.2  Sewage Sludge Incinerators	3-18
             3.2.3  Fossil Fuel  Combustion	3-20
                    3.2.3.1   Coal  Combustion	3-20
                    3.2.3.2   Oil  and Coal  Combustion	3-22
                    3.2.3.3   Oil  Combustion	.....". 3-22
                    3.2.3.4   Natural  Gas Combustion 	 3-22
                    3.2.3.5   Coal  and Refuse-Derived Fuel  Combustion. .  . 3-23
             3.2.4  Wood Combustion	3-23
                    3.2.4.1   Residential  Wood Combustion.  ......!!  3-23
                    3.2.4.2   Treated Wood  Combustion	3-25
             3.2.5  Boilers  Co-firing Wastes.  ....  	  3-27
             3.2.6  Hazardous  Waste Incinerators.  ...  	3-30
                    3.2.6.1   Land-based  Incinerators	3-30
                    3.2.6.2   Incinerator Ships	3-35
             3.2.7  Lime/Cement  Kilns	  3-36
             3.2.8  Hospital  Incinerators  	  .....  3-38
             3.2.9  Wire Reclamation  Incinerators  	  ......  3-38
             3.2.10  PCB Fires	3-41
             3.2.11  Automobile Emissions	3-43
             3.2.12  Activated  Carbon  Regeneration  Furnaces	3-46
             3.2.13  Experimental Studies.	  .  3.43
                                       i i i

-------
Chapter
TABLE OF CONTENTS

       Title
Page
        3.3  PCDD Formation Hypotheses and Factors Affecting
               Emissions for Combustion Sources 	 3-50
             3.3.1  A Summary of Formation Hypotheses for PCDD
                    from Combustion 	 3-50
             3.3.2  Factors Affecting PCDD Emissions from Combustion
                    Sources	3-52
                    3.3.2.1  PCDD in Feed	3-52
                    3.3.2.2  Precursors in Feed	3-53
                    3.3.2.3  Chlorine in Feed	3-54
                    3.3.2.4  Combustion Conditions. ... 	 3-54
                    3.3.2.5  Combustion Temperature 	 3-55
                    3.3.2.6  Residence Time	3-55
                    3.3.2.7  Oxygen Availability	3-56
                    3.3.2.8  Feed Processing. ....... 	 3-56
                    3.3.2.9  Supplemental Fuel.	„	3-57

  4.0   TEST PROGRAM DEVELOPMENT:
          SOURCE CATEGORY RANKING AND TEST SITE SELECTION PROCEDURES. .  . 4-1
        4.1  Source Category Selection and Ranking	4-2
             4.1.1  Development of Ranked Source Category List	4-2
                    4.1.1.1  Preliminary Ranked Source Category List. .  . 4-2
                    4.1.1.2  Modification of Preliminary List 	 4-4
                    4.1.1.3  Final Ranked List	4-4
             4.1.2  Rationale for the Source Category Ranking 	 4-8
                    4.1.2.1  Rank A Sources 	 4-8
                    4.1.2.2  Rank B Sources	4-8
                    4.1.2.3  Rank C Sources	4-10
                    4.1.2.4  Rank D Sources	4-11
        4.2  Test Site Selection	•	4-11
             4.2.1  Background	4-11
             4.2.2  Site Selection Methodology	4-13
             4.2.3  Test Sites Selected	4-13
                    4.2.3.1  Sewage Sludge Incinerators 	 4-15
                    4.2.3.2  Black Liquor Boilers 	 4-15
                    4.2.3.3  Industrial Incinerators	4-15
                    4.2.3.4  Wire Reclamation Incinerators	4-18
                    4.2.3.5  Carbon Regeneration Furnaces 	 4-18
                    4.2.3.6  Secondary Copper Blast Furnaces	4-18
                    4.2.3.7  Wood-Fired Boilers 	 4-18
                    4.2.3.8  Drum and Barrel  Reclamation Furnaces .... 4-23
        4.3  Summary Description of All Test Sites.  .  .	4-23

  5.0   TIER 4 EMISSIONS TESTING RESULTS.	5-1
        5.1  Sampling Matrix	5-1
             5.1.1  CCD/CDF Emissions Sampling	5-4
                                        IV

-------
                              TABLE OF CONTENTS
Chapter
                                                                          Page
             5.1.2  HC1 Sampling	 5-5
             5.1.3  Continuous Monitoring of Combustion Gases .' .' . . .  '. 5-5
             5.1.4  In-Plant Ambient Air Sampling 	  .5-6
             5.1.5  Feed Sampling	[•[ 5.5
             5.1.6  Ash Samples	 5-6
             5.1.7  Soil Samples	     5-6
             5.1.8  Auxiliary Process Samples	•.'.'.	5-7
        5.2  COD/CDF Emissions Data	...'!.' 5-7
             5.2.1  Flue Gas Concentration Data .......'.I'.'.'.'. 5-7
                    5.2.1.1' Outlet Flue Gas Emission Concentrations. '.  '. 5-7
                    5.2.1.2  Control  Device Inlet Flue Gas
                             Concentrations .........	   5-11
             5.2.2  Mass Emission Rate Data	  .            ' 5-13
                    5.2.2.1  Outlet PCDD/PCDF Emission Rates. ..!!!! 5-13
                    5.2.2.2  Control  Device Inlet PCDD/PCDF
                             Mass Flow Rates	      5-13
             5.2.3  Scaled Mass Emissions Data.  .......           ' 5-15
        5.3  CDD/CDF Precursor Data	'.'.'.' 5-16
             5.3.1  Sewage Sludge Incinerator Test Sites.  .......    5-ig
             5.3.2  Black Liquor Boiler Test Sites.  .	 5-23
             5.3.3  Wood Combustion Test Sites	      5-23
             5.3.4  Metals Recovery Test Sites	        	5-24
             5.3.5  Other Test Sites	...!!.'!''' 5-25
        5.4  HC1  Sampling Data.  .	 5-26
        5.5  Continuous Emissions  Monitoring Data .  .            	5-28
        5.6  Process Data	'.'.'.'.'.'. 5-31
             5.6.1  Combustion Device Operating  Data.  .  ......... 5-31
             5.6.2  Emissions  Control  Device Operating Data    ••*"'* 5_31
        5.7  Ash  Sampling Data	          .... ^
        5.8  Ambient Air Sampling  Data	 5-39

 6.0    DATA  ANALYSIS	                      g.!
        6.1  Factors Affecting CDD/CDF Emissions!  !.*!.*!!"*"*•'" 6-1
             6.1.1  Rank Order Statistical Analysis  	            6-3
        6.2  Qualitative Observations  ..'...	[  5_8
             6.2.1  Qualitative Analysis  for Sewage  Sludge Incinerators  !  6-11
             6.2.2  Qualitative Data Analysis  for  Black Liquor Boilers.  .  6-13
             6.2.3  Qualitative Data Analysis  for  Municipal Waste
                      Incinerators	    5_13
                    6.2.3.1  Description  of  the MWI  Data Base  ......  6-15
                    6.2.3.2  Rank Order Analysis of  Flue Gas  Emissions
                              of CDD's and  CDF's	6-19
                    6.2.3.3  Summary	                "  5.25
       6.3   PCDD/PCDF Homologue Distributions	!!!!!!!' 6-27

-------
Chapter
TABLE OF CONTENTS

       Title
        6.4  Control Device Efficiency Results	6-37
             6.4.1  Overview of the Removal Efficiency Data 	 6-38
             6.4.2  Uncertainty Analysis	6-42
             6.4.3  Evaluation of Specific Control Devices	6-42
                    6.4.3.1  Electrostatic Precipitators (Sites BLB-A,
                               BLB-B, and BLB-C)	6-42
                    6.4.3.2  Baghouse (Site WFB-A).	 6-43
                    6.4.3.3  Spray Dryer/Baghouse Combination 	 6-44
                    6.4.3.4  Water Scrubber 	 6-44
                    6.4.3.5  Afterburner	6-45
             6.4.4  Summary of Control Efficiency Observations	6-45

  7.0   QUALITY ASSURANCE PROGRAM	'	7-1
        7.1  QA Program Objectives	7-1
        7.2  Summary of QA/QC Activities	7-3
             7.2.1  Sampling Activities	7-5
             7.2.2  Analytical Activities 	 7-12
             7.2.3  Audit Activities	7-14
        7.3  Data Quality Assessments	7-16
             7.3.1  Flue Gas CDD/CDF Analyses	7-16
             7.3.2  Ash CDD/CDF Analyses	7-22
             7.3.3  Feed CDD/CDF Precursor Analyses 	 7-27
             7.3.4  Flue Gas Combustion Parameters Monitoring 	 7-29

  8.0   ASH SAMPLING PROGRAM	8-1
        8.1  Overview of the Ash Sampling Program 	 8-1
        8.2  Previous Work	8-6
        8.3  Ash Sampling Site Selection	8-6
        8.4  Sample Acquisition 	 8-11
        8.5  Summary of Results 	 8-13
             8.5.1  Ash Sampling Program Data base	8-13
             8.5.2  PCDD/PCDF Ash Results for the Source Test Sites .  .  . 8-26
        8.6  Statistical Analysis of Tier 4 Ash Data Bases	8-26
             8.6.1  Non-parametric Rank Order Statistical Analysis of
                      the Source Test Ash Data	8-30
             8.6.2  Non-parametric Rank Order Statistical Analysis of
                      the Ash Sampling Program Data	8-32
             8.6.3  Summary of the Ash Sampling Program	 8-32
        8.7  Evaluation of Ranking of Source Categories 	 8-34

  9.0  REFERENCES .	9-1

  APPENDIX A:  Tier 4 Emissions Data	A-l
                                       VI

-------
                              LIST OF TABLES
Number
                                                                          Page
 1-1   Summary of PCDD/PCDF Stack Emissions by Source Category. ..... 1-3
 1-2   Average Emission Control Efficiencies and Inlet Concentrations
         for Total PCDD and Total PCDF	1-8

 2-1   Nomenclature and Schedule of Theoretical Chlorinated
         Dioxin Isomers 	 ......           2-8
 2-2   Nomenclature and Schedule of Theoretical Chlorinated
         Furan Isomers	2-10

 3-1   Summary of PCDD/PCDF Stack Emissions Data from the Literature. .  . 3-4
 3-2   Summary of Literature Data on PCDD and PCDF Contents of
         Combustion Ash	          3.10
 3-3   Source Category: Municipal Solid Waste Incinerators! !!!.'.'!' 3-16
 3-4   Source Category: Sewage Sludge Incinerators	       3-19
 3-5   Source Category: Fossil Fuel  Combustion. 	         ' 3-21
 3-6   Source Category: Wood Combustion .	'     3-24
 3-7   Source Category: Boilers Co-firing Wastes	     3-28
 3-8   Source Category: Hazardous Waste Incinerators	       '    3-31
 3-9   Source Category: Lime/Cement  Kilns	•..'!' 3-37
 3-10   Source Category: Hospital  Incinerators 	 .      '    3.39
 3-11   Source Category: Wire Reclamation Incinerators  	          3-40
 3-12   Source Category: PCB Fires	'    3.42
 3-13   Source Category: Automobile Emissions	    	3.44
 3-14   Source Category: Activated Carbon Regeneration  ...      .....
 3-15   Source Category: Experimental	          ....  3_^ •
 3-16   Summary of PCDD/PCDF Formation Hypotheses.  ............  3-51

 4-1    Preliminary Ranked  Source  Category  List  (October  1984)              4-5
 4-2    Modification of Preliminary List	  .  °  '  '  '  '  4-6
 4-3    Final  Ranked Source Category  List	                  4.7
 4-4    Information Collected at Sewage Sludge Incinerator  Sites
        during Pre-Test Surveys  for  Tier  4	                  4-16
 4-5    Information Collected at Black Liquor Boiler Sites  during	
        Pre-Test  Surveys  for Tier 4	    4-17
 4-6    Information Collected at the  Industrial  Incinerator Test Site
        during the Pre-Test Survey  	              4.19
 4-7    Information Collected at the Wire Reclamation Incinerator'sites'
        during the Pre-Test Survey for  Tier 4	4-20
 4-8    Information Collected at the Carbon Regeneration  Furnace Test
        Site  during the Pre-Test	      4.21
 4-9    Information Collected at Secondary Copper Blast Furnace'sites
        ^during Pre-Test Surveys  for  Tier 4  	                  4-22
 4-10   Information Collected at Wood-Fired Boiler Sites  during
        Pre-Test  Surveys  for  Tier 4	4.24
 4-11   Information Collected  at Drum  and Barrel Reclamation Furnace
        Sites  during  Pre-Test  Surveys for Tier 4	            4-25
 4-12  Summary  Description of  the  Tier 4 Test Sites  ....            "    4-26
                                      vii

-------
                                 LIST OF TABLES

Number                               Title                                Page

 5-1   Source Sampling Matrix for the Tier 4 Test Program	5-2
 5-2   Average PCDD/PCDF Outlet Emission Concentrations from
         the Tier 4 Test Sites	5-8
 5-3   Toxic Equivalency Factors Used in Estimating 2378-TCDD
         Equivalents	5-10
 5-4   Average Control Device Inlet PCDD/PCDF Emission Concentrations
         from the Tier 4 Test Sites 	 5-12
 5-5   Average Outlet PCDD/PCDF Mass Emission Rates for the
         Tier 4 Test Sites	5-14
 5-6   Average Control Device Inlet PCDD/PCDF Mass Flow Rates for
         the Tier 4 Test Sites	5-15
 5-7   Average Outlet Scaled PCDD/PCDF Mass Emissions for the
         Tier 4 Test Sites	5-17
 5-8   Average Control Device Inlet Scaled PCDD/PCDF Mass Flow for the
         Tier 4 Test Sites	5-18
 5-9   Dioxin/Furan Precursor Data Summary	5-20
 5-10  HC1 Train Emissions Data Summary	5-27
 5-11  Mean Values and Standard Deviations of the Continuously
         Monitored Stack Gases	5-29
 5-12  Mean Combustion Device Operating Parameters during the
         Test Runs	5-32
 5-13  Summary of Control Device Operating Parameters for Sewage
         Sludge Incinerators	5-34
 5-14  Summary of Control Device Operating Parameters for Black
         Liquor Boilers  	 5-35
 5-15  Summary of Control Device Operating Parameters for Wood Combustion,
         Metals Recovery and Miscellaneous Source Categories	5-36
 5-16  Summary of Ash Sample PCDD/PCDF Data for the Tier 4 Test Sites  .  . 5-38
 5-17  Summary of Ambinet PCDD/PCDF Data for the Tier 4 Test Sites.  .  .  . 5-40

 6-1   Tier 4 Surrogate Measurements for Factors Affecting
         Dioxin Emissions	6-2
 6-2   Data Matrix for Statistical Analysis 	  .... 6-4
 6-3   Statistical Analysis Input Data Sources	6-5
 6-4   Data Comparison for the Tier 4 Test Sites	6-9
 6-5   Data Comparison for the Three Tier 4 Sewage Sludge Incinerator
         Test Sites	6-12
 6-6   Data Comparison for the Three Tier 4 Black Liquor Boiler
         Test Sites	6-14
 6-7   Dioxin and Furan Emission Data Summary for United States and
         Canadian MSW Incinerators	6-16
 6-8   Dioxin and Furan Emission Data Summary for European MSW
         Incinerators 	 6-17
 6-9   MWI Emission Studies Not Included in Data base	6-18
 6-10  Ranking of Flue Gas Emissions of PCDD's and PCDF's with
         Furnace Temperature. 	 6-21
 6-11  Ranking of PCDD in Fly Ash with Furnace Combustion Temperature  .  . 6-24
 6-12  Emission Control Efficiency Results	6-39
 6-13  Average Emission Control Efficiences and Inlet Concentrations
         for Total PCDD and Total PCDF. •	6-41

                                       viii

-------
                                 LIST OF TABLES
Number
 7-1   Test Site Numbers and Combustion Device Codes. .                    7-2
 7-2   Summary of Tier 4 Data Quality Objectives. 	 	 7.4
 7-3   Tier 4 Glassware Precleaning Procedure 	                 7.5
 7-4   Analysis Results for Tier 4 Modified Method 5 Proof Blanks !  !  !  ' 7-8
 7-5   Summary of Isokinetic Calculations and Leak Check Results for
         MM5 and HC1	                         7.9
 7-6   Summary of CDD/CDF Analysis Quality Control'checks !!!!*''* 7-13
 7-7   Summary of Analytical Audit Results for 2378-TCDD.                 7-18
 7-8   Analysis Results for Fortified Flue Gas Quality Control  Samples!  ! 7-19
 7-9   Analysis Results for Tier 4 CDD/CDF Laboratory Blanks. .           7-20
 7-10  Analysis Results for Tier 4 Modified Method 5 Field Blanks .  .      7-21
 7-11  Spiked Surrogate Recoveries for Tier 4 Flue Gas CDD/CDF  Samples.  . 7-23
 I'M  Jnajvsis Results for Fortified Ash Quality Control Samples .... 7-24
 7-13  Spiked Surrogate Recoveries for Tier 4 Ash Samples .....        7-25
 7-14  Comparison of Surrogate Recoveries for Large and  Small
         Sample Sizes	                7.23
 7-15  Continuous Emission Monitoring System (CEM)'Audit"Results!  !  !  !    7-31
 7-16  Summary of Instrument Drift Check Results  for Flue Gas
         Parameters  	  	 .....                7-32
 7-17  Summary of Quality Control  Standard Analyses  for  Flue*Gas	
         Parameters	          7_36

 8-1    Summary of Ash Sampling Sites.  ........                      8-3
 8-2    Comparisons of Flue Gas and Fly Ash Dioxin and  Dibenzofuran'
         Contents	^                8_7
 8-3    Combustion Source  Categories  Sampled in*Ash Program!  !            '  8-9
 8-4    Source Characteristics  of Interest for Dioxin Test Program  !  !  !  '  8-10
 8-5    Sampling  Organizations  for  Ash  Sites ...                          Q  i?
 8-6    Site  Codes .	        	g  15
 8-7    Summary of Ash  Sampling Process Data ..!!!!!!!!!*'"'  8-16
 8-8    Average Combustion  Zone Temperatures for the Combustion'Device
         Categories	^                g_lg
 8-9    Average Temperature  in  the  Control  Devices  at'the'Ash'samplinq
         Point	                v   y      8  20
 8-10   Summary of Ash  Sampling  Results.  .  .  	  !!!!!!          8-21
 8-11   Summary of Ash  Sampling  Results Sorted by Type of*Ash! 	8-24
 8-12   Comparison  of Tier 4 Ash  and Flue  Gas Data  .                        8  27
 8-13   Summary of Source Test  Site Ash Results	  ^^
 8-14   Examination of Ash/Flue Gas Rank Non-parametric Spearman
        Correlations	;	      8_31
 8-15   Examination of Ash Sampling Program Rank Non-parametric	
        Spearman  Correlations Sorted by Type of Ash	        8-33
8-16  Summary of Ash Sampling Program Results Sorted by 2378-TCDD*
        Toxic Equivalents	                   g_35
                                       IX

-------

-------
Number
LIST OF FIGURES

     Title
 2-1   Structural Formula of Dioxin and Furan Nuclei	2-7

 4-1   Development of Ranked Source Category List 	         4.3
 4-2   Procedures Used for Selecting a Generic Test Site. ..!'!.*."!! 4-14

 6-1   Rank Order Plot of PCDD Emissions vs. Combustion Temperature
         for Municipal Waste Incinerators	             6-22
 6-2   Rank Order Plot of PCDF Emissions vs. Combustion Temperature
         for Municipal Waste Incinerators .....  	            6-23
 6-3   Rank Order Plot of Fly Ash PCDD Content vs.  Combustion
         Temperature for Municipal  Waste Incinerators . . 	 6-25
 6-4   Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
         from Sewage Sludge Incinerators (Inlet)	    6-28
 6-5   Dioxin/Furan Homologue Distributions of Controlled Emissions
         from Sewage Sludge Incinerators (Outlet)	 6-29
 6-6   Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
         from Black Liquor Boilers  (Inlet)	    6-30
 6-7   Dioxin/Furan Homologue Distributions of Controlled Emissions
         from Black Liquor Boilers  (Outlet) 	 6-31
 6-8   Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
         from Wood Combustion Processes (Inlet)  	    6-32
 6-9   Dioxin/Furan Homologue Distributions of Controlled Emissions
         from Wood Combustion Processes (Outlet)	         6-33
 6-10   Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
         from Miscellaneous Combustion Sources (Inlet)	     6-34
 6-11   Dioxin/Furan Homologue Distributions of Controlled Emissions
         from Miscellaneous Combustion Sources (Outlet)  	        6-35
 6-12   Dioxin/Furan Homologue Distributions of Controlled Emissions
         from Metals Recovery Processes (Outlet)	6-36

 7-1    Drift  Check Results for CEM  Parameters	 7-33

 7-2    Control  Standard  Analysis  Results for CEM Parameters  ......  .7-37

-------

-------
                                    CHAPTER 1
                                EXECUTIVE SUMMARY

      The Air Management Technology Branch of the U.  S.  Environmental
 Protection Agency's (EPA)  Office of Air Quality Planning and Standards (OAQPS)
 was responsible for implementing Tier 4 of the National  Dioxin Study.   Radian
 Corporation,  under task order contract, provided technical  support for this
 program.   The technical  support included developing  literature surveys,
•preparing sampling protocols,  conducting stack tests, coordinating ash
 sampling efforts and preparing this and other reports.
      The National  Dioxin Study was focused on the study  of  chlorinated
 dibenzo-p-dioxins  (CDD's),  and in particular on polychlorinated
 dibenzo-p-dioxins  containing  four or more chlorine atoms (PCDD's).  The
 acronyms  CDD  and PCDD will  be  used in this report to denote these  species.
 Section  2.3.2 contains a complete explanation of the nomenclature  used in this
 report.
      The  primary objective  of  Tier 4 was  to study combustion  processes as
 potential  sources  of PCDD emissions  and the related compounds  chlorinated
 dibenzofurans (CDF's).   The following questions  were addressed:
      o     Do  combustion  sources  emit PCDD's?
      o     If  so, how much?
      o     Are these  emissions  significant?
 A secondary objective of the Tier  4  sampling  program was to attempt to deter-
 mine  what  factors  affect PCDD/PCDF emissions, and to determine the effective-
 ness  of conventional  control devices  for controlling PCDD/PCDF emissions.
      The Tier 4  study began in November 1983 with an extensive literature
 survey.  The purposes  of the literature survey were to 1) summarize previous
 research done on emissions of CDD's and chlorinated dibenzofurans (CDF's) from
 combustion processes,  and 2) summarize available CDD and CDF emissions data
 from combustion processes.   After the literature survey was completed, emis-
 sions source testing  and combustion ash sampling programs were initiated under
                                      1-1

-------
Tier 4 to investigate the hypotheses identified in the literature.  This
report summarizes the data gathered during the Tier 4 study and provides a
data analysis which is primarily intended to illustrate the extent to which
combustion sources emit CDD's.
     The literature survey identified 13 broadly defined source categories
(described in Section 3.1) for which some PCDD and polychlorinated dibenzo-
furan (PCDF) emissions data had been collected prior to the Tier 4 study.
Those source-categories for which stack emissions data were available are
shown in Table 1-1 along with the source categories tested under Tier 4.  The
data presented reflects as-measured and as-reported concentrations.
Normalization to a reference oxygen level has not been made since insuffient
data were available.  The literature survey indicated that municipal solid
waste incinerators were the most frequently tested source category and had the
highest levels of PCDD's and PCDF's in stack gas emissions.  Commercial
boilers co-firing spiked waste oil had the next highest PCDD concentrations in
stack gas emissions, followed by emissions from one tested unit combusting
pentachlorophenol-treated wood.
     PCDD's and PCDF's were not found in stack emissions from all combustion
sources for which data were reported in the literature.  Two utility boilers
firing PCB-spiked waste oil, a lime kiln, a coal-fired utility boiler and the
incinerator ship M/T Vulcanus were all tested and had less than detectable
PCDD emissions.  In general, the literature survey showed that where
combustion sources did emit PCDD's or PCDF's, these emissions appear to be
dependent on the types of fuel or waste being fired and whether or not the
combustion units were designed and operated specifically to achieve the
destruction of compounds such as PCDD's and PCDF's.
     Emissions from combustion devices burning clean fuel  such as natural gas
and distillate oil have not been adequately characterized.  However, these
sources are unlikely to emit high levels of PCDD/PCDF's because of the small
amounts of precursors, specifically chlorine, present in the fuel.   Some high
temperature (above 2,500°F firebox temperature)  combustion sources such as
utility boilers and cement kilns were also found to emit less than detectable
amounts of PCDD/PCDF under the test conditions.
                                      1-2

-------
                         TABLE 1-1.  SUMMARY OP PCDD/PCDF STACK EMISSIONS BY SOURCE CATEGORY
Number of Units Tested* BanK» of pnpp Emission]
Source Category Tier 4
1.


2.


3.
4.


5.
6.

7.
8.

9.
10.

11.
12.


13.
14.
15.
16.
17.



IS.
Municipal Waste Incinerators
European
U.S. and Canada
Boilers Cofiring Waste
Commercial
Industrial
Secondary Copper Cupola Furnace
Wood Combustion
PCP-Treated Wood
Salt-Laden Wood-Fired Boiler
Sewage Sludge * Incinerators
Wire Reclamation Incinerator
(wire and transformer feed)
Industrial Solid -Waste Incinerator
Wire Reclamation Incinerator
(wire-only feed)
Hospital Incinerators
Hazardous Waste Incinerators
Rotary Kiln
Drum & Barrel Reclamation Incinerator
Carbon Regeneration Furnace
without Afterburner
with Afterburner
Black Liquor Boiler
Cement Kilns
Lime Kilns
Utility Boiler Co-firing Waste
Fossil Fuel Combustion
Coal-Fired Utility
Pulverized Coal
011-Fired Utility
Incinerator Ship

0
0

0
0
1

0
1
3

1
1

1
0

0
1

0
1
3
0
0
0

0
0
0
0
Literature , 3
(as measured, ng/m )

8 71 - 48,997
10 3.3 - 11,686

3 1,400 - 17,000
5 <0.002b - 76.4
0 11,900

2 <17° - 1,520
0 195
2 NDd - 812

0 704
0 625

0 173
4 15-69

2 7.7 - 8.6
0 5

1 0.18
1 1.6 - 3.7
0 0.8 - 2.9
3 
-------
     The literature review indicated that several factors affect CDD/CDF
emissions.  These factors include the PCDD content of the feed, precursor
content of the feed, chlorine content of the feed, combustion device
temperature, combustion device residence time, combustion device oxygen
availability, feed processing and supplemental fuel.  These factors were
addressed during the analysis of the Tier 4 data.  As a whole, the statistical
analysis did not provide quantitative relationships between CDD/CDF emissions
and the independent factors considered.  However, the analysis did identify a
few potentially meaningful associations, including strong inverse associations
between maximum measured combustion chamber temperature and total CDD/CDF.
Composition of feed materials also influenced the magnitude of CDD/CDF
emissions.  These analyses are discussed in Chapter 6.
     In conjunction with the literature survey, a source test plan was devel-
oped to expand the PCDD/PCDF emissions data base and to address the objectives
of the Tier 4 program.  Potential source categories to be included in the test
program were identified by a ranking process based on the potential to emit
PCDD's.  The ranking was subjective and was based on many factors, including
potential for PCDD content in the feed, the size of the source category, the
presence of precursors in the feed, typical operating conditions of the
combustion device and the number of previous source tests performed.   Specific
test sites within a source category were selected based on inputs from State
pollution control agencies and results of on-site pre-test surveys.  A total
of 31 pre-test surveys were conducted to select the 13 Tier 4 test sites.
     The 13 source tests were conducted during 1984 arid 1985.  Nine different
types of combustion units were tested.  These included three sewage sludge
incinerators, three black liquor boilers,  one wire reclamation incinerator,
one industrial  solid waste incinerator, one carbon regeneration furnace, one
secondary copper blast furnace,  one wood-fired boiler, one drum and barrel
reclamation furnace,  and one residential woodstove.
     The Modified Method 5 (MM5) protocol  specified by the December 1984
Environmental Standards Workshop was followed as closely as possible  for
                                      1-4

-------
PCDD/PCDF sampling.   Three MM5 test runs were performed at each test site
(except for the wire reclamation incinerator site, where six MM5 runs were
performed).  At a minimum, MM5 sampling of the outlet emissions stream was
performed at each test site.  Where feasible, MM5 sampling was also performed
at the inlet to the control devices in order to obtain PCDD/PCDF removal
efficiency data.  MM5 sample components were analyzed for PCDD and PCDF
content using high resolution gas chromatography/high resolution mass
spectrometry.  While the sampling and analysis methods used in this study were
state-of-the-art, they are nevertheless evolutionary.  During the course of
the study, it was  sometimes found that the analysis methods could not cope
with high levels of interfering contamination from other pollutants.  This
caused difficulty in achieving the desired validity and precision of results.
Also, the MM5 stack sampling method is currently undergoing validation
testing. Preliminary results indicate that under some conditions, recovery
efficiencies from the sampling method may be low and variable, with possibly
less than half of the CDDs and CDFs in the stack emissions being collected by
the stack sampling method.  Additional validation testing is currently
underway.                                                                  •
      Combustion gases were continuously monitored for oxygen, CO, CCL,  total
hydrocarbons (THC), NOX, and S02 to document the stability of combustion
conditions at each site.  In addition, at some or all of the sites the
following types of sampling were performed:  feed sampling, hydrochloric acid
emissions sampling, combustion air sampling, combustion ash sampling, soil
sampling, and auxiliary process sampling.
     Table 1-1 summarizes the range of PCDD/PCDF emissions data for the 18
source categories that have been sampled to date, either by the Tier 4 program
or those identified in the literature.  The results of the Tier 4 sampling and
analysis showed that all combustion sources tested under Tier 4 emit PCDD's
      Analytical Procedures to Assay Stack Effluent Samples and Residual
Combustion Products for Polychlorinated Dibenzo-p-Dioxins (PCDD) and
Polychlorinated Dibenzofurans (PCDF).  Prepared by Group C- Environmental
Standards Workshop.  Sept. 18, 1984, revised Dec. 31, 1984.
                                      1-5

-------
and PCDF's.  The total PCDD emissions vary over four orders of magnitude
between the test sites.  Emissions of 2378-TCDD generally account for less
than 1 percent of the total PCDD emissions.   PCDD and PCDF emissions are
usually on the same order of magnitude.
     An engineering analysis of the Tier 4 test data and equivalent data
obtained from the literature review was performed.  The analysis attempted to
address some of the secondary objectives of Tier 4, including:
     o    What factors affect PCDD/PCDF emissions?
     o    How effective are conventional emissions control devices for
          controlling PCDD/PCDF emissions?
     Unfortunately, the data set was too small to allow these questions to be
completely addressed and in addition, the analytical results for PCDD and PCDF
are not exact (+ 50%).  However, indications of factors that can have a
significant influence on PCDD emissions may be obtained by studying the
composite results from the'literature survey and testing effort.  At the very
broadest level of inspection the data obtained from the literature and Tier 4
indicate the following:
     o    The highest PCDD emissions concentrations appear to be associated
          with low temperature combustion processes whose function is to
          recover energy or other resources (e.g., metal values) by combustion
          of waste materials.  Examples of these types of processes are
          municipal solid waste incinerators and the secondary copper blast
          furnace.
     o    Combustion temperature alone appears to have an inverse association
          with the total PCDD emission levels.  This is supported by an
          analysis of the municipal solid waste incinerator data presented in
          6.2.3.  This is not an unexpected effect, since lower combustion
          temperatures lead to increased emissions of products of incomplete
          combustion, while high temperatures coupled with long residence
          times have been used to destroy hazardous wastes at very high
          efficiencies.
     o    The composition of feed materials is also expected to have a marked
          influence on the magnitude of PCDD/PCDF emissions and this is
                                      1-6

-------
           partially supported by the Tier 4 results.   There have been a number
           of theories advanced in the literature concerning which components
           of the feed contribute to PCDD/PCDF formation.   Components most
           often identified are specific precursors  (chlorobenzenes,
           chlorophenols,  chlorinated biphenyls),  organic  chlorine or total
           chlorine.   In  addition,  several  authors have suggested that lignin
           and plastics content may also be important.   Feed materials from
           some of the Tier 4  test  sites were analyzed  for specific precursors,
           total  organic  halogens (TOX),  and total chloride content.   Of these
           factors,  TOX analysis showed  the strongest association with total
           PCDD emissions.   The amount of plastics or lignins in  each feed
           material was not specifically analyzed  for.   However,  the  highest
           PCDD emissions were measured  at  sites containing plastics  in the
           feed.
     There are very  few data  in the  literature concerning control  of PCDD/PCDF
emissions  since  most sampling efforts have focused on  emissions  at the stack
and not at the control device inlet.  Several approaches  to controlling  PCDD
emissions  have been  postulated.  .These  include operating  the combustion  device
to minimize  PCDD emissions, use of high  temperature/long  residence time
afterburners  or  secondary  combustion  chambers to  reduce PCDD, or use  of
conventional  particulate control devices.
     PCDD/PCDF are semi-volatile organic compounds with high  boiling  points'
and can exist  in  the  gas or solid  phase depending on the  flue gas  temperature.
Several studies  have  shown that  PCDD  is destroyed at temperatures  above 800°C.
However, without  proper mixing  or  sufficient residence time, all  PCDD/PCDF may
not be destroyed  even at these  high temperatures.
     If the PCDD/PCDF are  in particulate form, an efficient particulate
control device is needed to control these emissions  to low levels.  Table 1-2
summarizes the control device and control device efficiency for the Tier 4
tests.   Based on the table and other data presented  in Chapters 5 and 6, the
following observations can be made:
     o     Three combustion devices tested during Tier 4 were equipped with
          afterburners operating at temperatures above the reported
                                      1-7

-------
        TABLE 1-2.  AVERAGE EMISSION CONTROL EFFICIENCIES AND INLET
                    CONCENTRATIONS FOR TOTAL PCDD AND TOTAL PCDF

Control Device/
Test Site
Total PCDD
Inlet
Concentration
(ng/dscm @ 3% 02)
.El.ectrost.at.i c Preci pi tators
BLB-A 1.8
BLB-B
BLB-C
Baqhouse
WFB-A
Spray Drver/Baghouse
CRF-A
Water Scrubber
SSI -A
Afterburner
DBR-A
17.1
9.0

102

28.8

101

687

Average
Measured
Efficiency
(Percent)
46
(23)
62

-130

88

40

99
Total PCDF
Inlet
Concentration
(ng/dscm @ 3% 02)
1.5
1.1
15.1

154

70.1

455
•
2,170

Average
Measured
Efficiency
(Percent)
58
(-101)
40

39

95

(-19)

99
Note: Values in parentheses (  ) indicate averages calculated from positive
      and negative values.
                                        1-8

-------
           destruction temperature for PCDD (i.e.,  800°C or 1,470°F)  and were
           emitting PCDD's.   This finding indicates that,  in addition to
           temperature,  other factors such as residence time and mixing must be
           accounted for when controlling CDD's.
      o     Results  for one of the tested  afterburners  on a drum and barrel
           reclamation furnace showed a control device efficiency greater than
           98  percent for reducing PCDD emissions.   Control  device efficiency
           was not  measured  for the other combustion devices using afterburners
           because  inlet MM5 sampling was not performed at these sites.
      o     A spray  dryer/fabric filter combination  installed on a carbon
           regeneration  furnace was more  effective  at  PCDD control  than  a
           fabric filter alone installed  on a wood-fired boiler.   This  indi-
           cates that perhaps  cooling/nucleation/condensation effects are
           needed to  effectively control  PCDD.  Alternatively,  this may  imply
           that removal  of reactive species such as  HC1  prior to  particulate
           control  is needed to improve PCDD/PCDF removal  efficiency.
      o     Fabric filters, which  are  generally considered  to be effective
           particulate control  devices, appear to be ineffective  PCDD/PCDF
           emission control devices when  used  alone.
      A statistical analysis was  also  performed using  the data  obtained  during
the Tier 4 testing.  The  purpose of  the  analysis was  to determine if PCDD,
PCDF, and  2378-TCDD  emissions  were related to source  type, combustion condi-
tions, feed type and other factors.   Non-parametric rank order statistical
techniques were used since the data were not well distributed  across all
possible values of the variables.  Rank order plots of the dependent variables
(i.e., 2378-TCDD, total PCDD and total PCDF emissions) versus  the independent
variables  (source type, feed total organic chloride,  feed chlorine, feed
precursors, combustion temperature, etc.) were developed.  The degree of
association between paired variables was tested using the Spearman Rank
Correlation Coefficient.  The analysis yielded the following results:
     o    Using all available data, there seems to be a strong association
          between total PCDD and 2378-TCDD emissions  (R = 0.93).  There  is
                                      1-9

-------
          also a strong association between total PCDD and total PCDF
          emissions (R = 0.86).
     o    There is a moderate association between total HC1 emissions and
          outlet PCDD emissions in the Tier 4 data set (R = 0.73), if the data
          from one site (MET-A) are excluded.  However, this correlation does
          not hold for the inlet emissions data.
     o    Using Tier 4 control device inlet data there is a potential strong
          inverse association between maximum temperature and inlet total
          PCDD/PCDF levels (R » -0.85).  An inverse association was also seen
          between the rankings of combustion temperature and PCDD emissions
          for the municipal solid waste combustor data set.  A weaker inverse
          association was also found for the PCDF emissions data.
     o    Using the maximum of outlet or inlet data from the Tier 4 test
          sites, rank correlation shows a moderate association between total
          organic halide (TOX) content of the feed and both maximum PCDD and
          maximum PCDF emissions (R = 0.81).
A more complete discussion of the results of the engineering analysis is
contained in Chapter 6 of this report.
     A comprehensive quality assurance program was implemented for the Tier 4
sampling effort.  The program emphasized 1) adherence to prescribed sampling
procedures, 2) careful documentation of sample collection and field analytical
data, 3) use of chain-of-custody records, 4) adherence to prescribed
analytical procedures, and 5) implementation of independent systems and
performance audits.  Overall, the quality assurance program was successful, in
that data were generally controlled within acceptable limits and sufficient
data were provided to assess the uncertainty in the results.  A more complete
description of the quality assurance program for Tier 4 is in Chapter 7 of
this report.
     Ash sampling was conducted under the Tier 4 study as a complimentary
effort to the more expensive stack sampling.  While it was recognized that
stack sampling was the rigorous means to determine CDD and CDF emissions from
combustion sources, ash sampling was believed to be a useful indicator of
                                     1-10

-------
 stack  emissions.   Therefore,  a  larger  number of combustion  source  categories
 were evaluated  using  ash  sampling.
     Chapter 8  of  this  report presents ash data from 75 separate combustion
 sources  representing  twenty-two different source categories.  The  types of ash
 sampled  included baghouse  ash,  multicyclone ash, electrostatic precipitator
 ash, afterburner ash, settling  chamber ash, bottom ash, economizer ash, and
 particulates filtered from scrubber water.  CDD's and CDF's were found in
 about  one-third of the  bottom ash and fly ash samples and about one-half of
 the scrubber effluent samples.  The highest concentrations were typically
 found  in fly ash samples.  Twelve of the twenty-two source categories had one "
 or more  ash samples with detectable concentrations.  Combustion conditions
 were also recorded when the samples were collected.  Chapter 8 discusses the
 effects of combustion parameters on CDD/CDF in the ash.
     A subset of the ash data base involves sources where both ash samples and
 stack samples were collected  simultaneously.  This subset covers ten
 combustion sources generally consisting of three samples of both ash and stack
 samples.  Chapter 8 presents an analysis of the ash data vs. the stack data.
 In general, it was observed that the presence of CDD's and CDF's in the fly
 ash appears to be a useful indicator of the presence of CDD's and CDF's in the
 stack emissions.  When PCDD/PCDF homologues were measured in the stack gas,
they were also detected in the control  device ash indicating that control
device ash has potential use as a screening tool.   However,  no quantitative
relationship was observed that could reliably predict the magnitude of CDD/CDF
emissions in the stack gases.
                                      1-11

-------

-------
                                   CHAPTER 2
                                   BACKGROUND
     This chapter presents background information on the Tier 4 Program and is
organized as follows:  Section 2.1 gives an overview of the purpose and
organization of the National Dioxin Study.  Section 2.2 outlines Tier 4 and
the steps taken to fulfill the directives of the National Dioxin Strategy for
Tier 4.  Background information on CDD's and CDF's, as well as definitions of
the nomenclature used in this report, follow in Section 2.3.

2.1  THE NATIONAL DIOXIN STRATEGY
2.1.1  Purpose of the National Dioxin Strategy
     On December 15, 1983, EPA released a National Dioxin Strategy which
provided a framework under which EPA was to:
     1.  Study the nature and extent of environmental contamination of
         2378-TCDD and the associated risks to humans and the environment;
     2.  Implement or compel necessary clean-up actions at contaminated
         sites; and
     3.  Further evaluate regulatory alternatives to prevent future
         contamination, as well as disposal alternatives to alleviate
         current problems.
     To implement the strategy, EPA formed seven study tiers, ordered by the
decreasing potential for 2378-TCDD contamination:
     Tier 1 - 2,4,5-trichlorophenol (245-TCP) production sites and associated
              waste disposal sites;
     Tier 2 - Sites (and associated waste disposal sites) where 245-TCP was
              used as a precursor to make pesticide products;
     Tier 3 - Sites (and associated waste disposal sites) where 245-TCP and
              its derivatives were formulated into pesticidal products;
                                       2-1

-------
     Tier  4  -  Combustion  sources;
     Tier  5  -  Sites  where pesticides  derived  from 245-TCP  have  been  and  are
               being  used  on  a  commercial  basis;
     Tier  6  -  Certain  organic  chemical  and  pesticide manufacturing facilities
               where  improper quality  control  on certain  production processes
               could  have  resulted  in  the  formation of  a  2378-TCDD contaminated
               product  waste  stream; and
     Tier  7  -  Control  sites  where  contamination from 2378-TCDD  is not
               suspected.
2.1.2  Management  of the  National  Dioxin  Strategy
     Overall management responsibility  for  the National  Dioxin  Strategy  was
assigned to  the Assistant Administrator for the Office of  Solid Waste  and
Emergency  Response (OSWER).  OSWER also managed the investigations for Tier 1
and Tier 2 of  the  study.   The  Office  of Water Regulations  and Standards  (OWRS)
managed Tiers  3, 5,  6  and 7, while the  Office of  Air and Radiation (OAR)
oversaw Tier 4.  The Office  of Research and Development  (ORD) was responsible
for overall  sampling and  analytical guidance.  ORD also  provided analytical
support for  Tiers  3  through  7  through two of  three EPA laboratories
collectively known as  Troika.  The EPA  laboratory at Bay St. Louis performed
all of the extractions for the Tier 4 samples, and PCDD/PCDF analyses  were
performed  at EMSL, Research  Triangle  Park.  EPA's Regional Offices were
responsible  for implementing various  aspects  of the National Dioxin  Strategy,
including  portions of  Tier 4.  A Tier 4 Work  Group composed of  representatives
from throughout the  Agency functioned in an advisory role  in the formulation
of plans and in the  review of  the  results.
2.1.3  Tier  4: Combustion  Sources
     The National  Dioxin  Strategy  directed Tier 4 to focus on "combustion
sources, such  as incineration  of hazardous and municipal  waste  (including
sewage sludge), wire reclamation facilities,  internal  combustion engines, home
heating units  (i.e.,  wood  burning  stoves), industrial  fossil fuel-fired
boilers, and inadvertent combustion sources."  This broad directive covered
literally millions of  individual  sources.   Since  it would have been
                                       2-2

-------
economically and otherwise impractical to test each source or even each source
category, considerable thought and judgment went into planning how best to
answer the following questions:
     1.  Which, if any, combustion source categories are likely to
         emit 2378-TCDD and other chlorinated isomers of CDD's and CDF's?
     2.  At what concentrations are these compounds emitted to the
         environment?
     3.  Do the ambient air concentrations resulting from emissions of these
         compounds pose an unreasonable risk to the public?
     The results from Tier 4 have been summarized in a report to Congress.
The results will provide data necessary to help make a decision on whether to
list "dioxin" as a hazardous air pollutant under Section 112 of the Clean Air
Act.
2.2  TIER 4 OVERVIEW
     The Air Management Technology Branch (AMTB) within the EPA's Office of
Air Quality Planning and Standards (OAQPS) in conjunction with the Hazardous
Waste Environmental Research Laboratory, Cincinnati, was responsible for the
development and implementation of the source testing program for Tier 4 of the
National Dioxin Study.
     The Tier 4 Project began in November 1983.  The project was performed in
a series of interrelated and overlapping steps.  These were as follows:
     1.  Perform an initial literature review; collect and review available
         literature on combustion related emission sources of PCDD's and
         PCDF's.  Develop a source test-plan based on the available CDD/CDF
         emissions data.  Adopt a "state-of-the-art" test method.
     2.  Develop a project plan incorporating the initial literature review
         findings.
     3.  Solicit regional and State inputs for potential  source test sites
         and prepare ash sampling, source sampling,  and quality assurance
         plans for the source sampling efforts.
         Implement the source test plans and the ash sampling plans.
         Collect and analyze all  the  data generated; prepare Engineering
         Analysis and Project Summary reports.
4.
5.
                                       2-3

-------
Twenty-one EPA documents, including this report, have been produced  as  a
result of the above-mentioned steps.  These are as follows:
     National Dioxin Study Tier 4 - Combustion Sources
       o  Project Plan - EPA-450/4-84-014a
       o  Initial Literature Review and Testing Options  - EPA-450/4-84-014b
       o  Sampling Procedures - EPA-450/4-84-014c
       o  Ash Sampling Program - EPA-450/4-84-014d
       o  Quality Assurance Project Plan - EPA-450/4-84-014e
       o  Quality Assurance Evaluation - EPA-450/4-84-014f
       o  Engineering Analysis Report - EPA-450/4-84-014h
       o  Final Literature Review - EPA-450/4-84-014i
       o  Final Test Report - Site 1, Sewage Sludge  Incinerator SSI-A
          EPA-450/4-84-014J
       o  Final Test Report - Site 2, Industrial Solid Waste  Incinerator
          ISW-A - EPA-450/4-84-014k
       o  Final Test Report - Site 3, Sewage Sludge  Incinerator SSI-B -
          EPA-450/4-84-0141
       o  Final Test Report - Site 4, Black Liquor Boiler BLB-A -
          EPA-450/4-84-014m
       o  Final Test Report - Site 5, Black Liquor Boiler BLB-B -
          EPA-450/4-84-014n
       o  Final Test Report - Site 6, Wire Reclamation Incinerator WRI-A  -
          EPA-450/4-84-0140
       o  Final Test Report - Site 7, Wood-Fired Boiler WFB-A -
          EPA-450/4-84-014p
       o  Final Test Report - Site 8, Black Liquor Boiler BLB-C -
          EPA-450/4-84-014q
       o  Final Test Report - Site 9, Carbon Regeneration Furnace CRF-A -
          EPA-450/4-84-014r
       o  Final Test Report - Site 10, Secondary Copper Recovery Cupola
          Furnace MET-A - EPA-450/4-84-014s
       o  Final Test Report - Site 11, Drum and Barrel Reclamation Furnace
          DBR-A - EPA-450/4-84-014t
                                       2-4

-------
        o   Final  Test Report -  Site 12,  Sewage Sludge Incinerator SSI-C -
           EPA-450/4-84-014U
        o   Final  Test Report -  Site 13,  Residential  Woodstove WS-A -
           EPA-450/4-84-014v
      Radian  Corporation,  under task order contract,  provided support  to AMTB
 and  Tier  4 in  the following areas:
      1.   Development of a project  plan  for Tier 4.
      2.   Collection  and review of  available literature  data  on  combustion-
          related sources  of PCDD and PCDF emissions.
      3.   Development of a source test plan.
      4.   Preparation of a sampling procedures document.
      5.   Development of a QA plan  for the sampling  effort.
      6.   Emissions testing at  13 combustion facilities.
      7.   Implementation of an  ash  sampling program.
      8.   Preparation of source test and other reports.
      The  purposes of this report are to summarize all of the data gathered
 during  the Tier  4 study and to analyze  the data.  The analysis  addresses the
 following questions:
      1.   Which combustion source categories  emit chlorinated
          dibenzo-p-dioxins  and dibenzofurans  to the  atmosphere?
      2.   What range  of  concentrations of chlorinated dibenzo-p-dioxins  and
          dibenzofurans  are  emitted from these  source categories?
      3.   What factors affect emissions  of chlorinated dibenzo-p-dioxins  and
          dibenzofurans  from combustion  sources?
      4.   How does  the distribution  of the  tetra through octa chlorinated
          CDD and  CDF homologues  vary between combustion sources?
      5.   How effective  are  conventional  emission control devices  in reducing
          chlorinated CDD  and CDF emissions from combustion sources?

2.3   BACKGROUND  INFORMATION ON CHLORINATED DIBENZO-p-DIOXINS AND CHLORINATED
DIBENZOFURANS
2.3.1  Structure
     Compounds which are generally labeled by the public as "dioxins" are
members of a family of organic compounds known chemically as
                                       2-5

-------
dibenzo-p-dioxins.  The common aspect of all dibenzo-p-dioxin compounds is
that they have a three ring nucleus consisting of two benzene rings
interconnected by a pair of oxygen atoms.  The structural formula of the
dioxin nucleus and the convention used in numbering its substituent positions
are shown in Figure 2-la.  In general, the term "dioxins" is used to mean the
chlorinated isomers of dibenzo-p-dioxin.  One to eight chlorine atoms can
occur at dioxin substituent positions such that 75 chlorinated dioxin isomers
are possible.  Each isomer has its own physical and chemical properties and
differs from others in the number and relative position of its chlorine atoms.
The potential chlorinated dioxin isomers are listed in Table 2-1.
     One of the 22 isomers with four chlorine atoms is 2,3,7,8-tetrachloro-
dibenzo-p-dioxin (2378-TCDD).  This isomer is the principal  focus of the
Tier 4 study for three reasons:
     1.  2378-TCDD is believed to be the most toxic of the chlorinated
         dioxins,
     2.  2378-TCDD is the isomer most often associated with  exposure and
         potential health risks to humans,  and
     3.  sufficient associated health and exposure information is
         available on 2378-TCDD to allow a targeted study to be developed.
     The compounds generally referred to as "furans" are members of a family
of organic compounds known chemically as dibenzofurans.  They have a similar
structure to the dibenzo-p-dioxins except that the two benzene rings in the
nucleus are interconnected with a five member ring containing only one oxygen
atom.  The structural formula of the furan nucleus and the convention used in
numbering its substituent positions are shown in Figure 2-lb.  The chlorinated
furan group can contain up to 135 different structural isomers, each with
varying physical and chemical properties.  The potential chlorinated furan
isomers are listed in Table 2-2.
2.3.2  Nomenclature Used in This Report
     Throughout this document the terms CDD and CDF will be  used to
generically indicate chlorinated dibenzo-p-dioxin or dibenzofuran compounds as
distinguished from specific chlorinated CDD or CDF isomers.   The abbreviations
PCDD and PCDF are used to indicate polychlorinated dibenzo-p-dioxins (PCDD)
and polychlorinated dibenzofurans (PCDF) with four or more chlorine atoms.  In
                                       2-6

-------
  Dibenzo - p - Dioxin Configuration
          9                  1
Figure 2-la.  Structural Formula of the Dioxin Nucleus
     Dibenzofuran Configuration
 Figure 2-lb.  Structural  Formula of the Furan Nucleus
                    2-7

-------
        Table 2-1.  Nomenclature and Schedule  of Theoretical
                    Chlorinated Dioxin  Isomers
Chlorinated Dioxin Compound  (abbreviation)
No. of Isomers
Honochlorodibenzo-p-dioxin  (Mono-CDD)
Di chlorodi benzo-p-di oxi n (Di-CDD)
Trichlorodibenzo-p-dioxin (Tri-CDD)
Tetrachlorodibenzo-p-dioxin (TCDD)
Pentachlorodi benzo-p-di oxi n (Penta-CDD)
Hexachlorodibenzo-p-dioxin  (Hexa-CDD)
Heptachlorodi benzo-p-di oxi n (Hepta-CDD)
Octachlorodi benzo-p-di oxi n  (Octa-CDD)
                    TOTAL ISOMERS
      2
     10
     14
     22
     14
     10
      2
      1
     75
                                  2-8

-------
the discussion of emissions data the terms total PCDD and total PCDF represent
the sum of the emissions of the tetra through octa homologues.  The term
"chlorinated CDD/CDF homologue" will be used to indicate the family of CDD/CDF
isomers with a fixed number of chlorine atoms.  For example, the tetra chlor-
inated CDD homologue consists of all CDD isomers containing four chlorine
atoms.  The abbreviations used for chlorinated CDD/CDF homologues are included
in Tables 2-1 and 2-2.
2.3.3  Report Organization
     The remainder of this report is organized as follows:  Chapter 3 presents
a summary of the literature survey, and the development of the source test
plan is outlined in Chapter 4.  The results of the emissions test program are
presented in Chapter 5, and an analysis of the emissions test program data is
summarized in Chapter 6.  The Quality Assurance Program for Tier 4 is
described in Chapter 7, and the Ash Sampling Program is described in Chapter
8.
                                       2-9

-------
        Table  2-2.   Nomenclature  and Schedule of Theoretical
                     Chlorinated Furan Isomers
Chlorinated  Furan  Compound  (abbreviation)
No. of Isomers
Honochlorodi benzofuran  (Mono-CDF)
Dichlorodibenzofuran  (Di-CDF)
Trichlorodibenzofuran (Tri-CDF)
Tetrachlorodibenzofuran  (TCDF)
Pentachlorodibenzofuran  (Penta-CDF)
Hexachlorodi benzofuran  (Hexa-CDF)
Heptachlorodi benzofuran  (Hepta-CDF)
Octachlorodi benzofuran  (Octa-CDF)
                    TOTAL ISOMERS
      4
     16
     28
     38
     28
     16
      4
    135
                                2-10

-------
                                  CHAPTER 3
                               LITERATURE REVIEW
     A literature review  for  sources  of PCDD  and  PCDF  air emissions was made
in 1983 by  the  Pollutant  Assessment Branch (OAQPS/PAB) of the United States
Environmental Protection  Agency  (EPA)A  This review was used as a  starting
point for  a more focused literature  search concerning PCDD emissions  from
combustion  sources  in early  1984.  The 1984  literature  review  identified
sources of  PCDD  and  PCDF  air  emissions and provided the  data base  on which
the Tier 4  testing program  was developed.8 At the conclusion of the source
testing efforts,  a  final  update of this review was conducted  to  include.
COD/CDF emissions information available through July of 1985.
     This  chapter summarizes the   final  literature review.   Industrial,
commercial,  and  residential  combustion sources that  have been tested  for
CDD/CDF emissions are  identified.   Quantitative  data are  presented  on
emissions of 2378-TCDD, PCDD's,  and PCDF's.   Qualitative information on the
source characteristics,   feed composition, and  sampling  and  analytical
methodologies are also presented.
     Sections 3.1  and 3.2  present  a  summary of the  literature  review.
Emissions data  are  presented for  2378-TCDD,  PCDD,  and  PCDF,  along with
qualitative  information   on  source  characteristics.   CDD/CDF  formation
hypotheses are summarized in Section 3.3.  Section  3.4 presents a discussion
of factors affecting CDD emissions.
     M
      Brooks, G. W.  Summary of a Literature Search to Develop Information on
Sources of  Chlorinated Dioxin  and  Furan Air  Emissions.   Final  Report
         No*  68-°2-3513-   u-  s-  Environmental  Protection Agency, October
    .                                       .
     B
      (Radian Corporation)  National Dioxin  Study—Tier 4 Combustion  Sources
1984    L1terature Review and Testing  Options.   EPA-450/4-84-014B.   October
                                     3-1

-------
3.1  OVERVIEW OF THE LITERATURE DATA BASE
     A review of  the  literature identified thirteen broadly defined  source
categories for which some PCDD and PCDF emissions data have been collected:
     o    municipal solid waste incinerators,
     o    sewage sludge incinerators,
     o    fossil fuel  combustion,
     o    wood combustion,
     o    boilers cb-firing wastes,
     o    hazardous waste incinerators,
     o    hospital incinerators,
     o    lime/cement kilns,
     o    wire reclamation incinerators,
     o    PCS fires,
     o    automobile emissions,
     o    activated carbon regeneration furnaces, and
     o    experimental studies.
     The PCDD and PCDF emissions data for each source category are presented
in the following sections.  The primary purposes of this presentation are  to
identify combustion sources that emit  PCDD's  and PCDF's  and,  as  a result of
comparisons of emissions  and  source  characteristics,  to  identify combustion
sources that are unlikely to  emit  PCDD's  and  PCDF's.  A  positive finding in
the literature  (i.e., detectable  CDD  emissions) may suggest the tested
combustion source emits CDD,  but  this  conclusion can only be reached after
consideration of the  quality  of the  reported  data  or research.   Likewise,  a
negative finding in the literature (i.e., less than detectable CDD emissions)
may be misleading if high detection limits  or  inappropriate sampling  and
analysis methods  were used.   Direct  comparison  of data reported in the
literature by  various researchers is  difficult  and  should  be done  with
caution for several reasons.  The  data  are often reported  on  different  bases
which cannot easily be interconverted because  of the  lack  of  source-specific
information.  A  variety  of sampling and  analytical  procedures were  often
used, and detection limits are frequently not specified.   Facility design and
operation may  vary considerably  and this type  of information  is  often
incompletely reported.  Also, some data are  available from  draft reports,
                                     3-2

-------
 emission tests,  and other unpublished documents which have not been subjected
 to peer or editorial review.   In this chapter, all available literature data
 were summarized.   No attempt  was made to exclude any reference because of the
 quality of the data.  For this  reason,  analysis  or full  explanation of the
 literature data is  not  possible.   However,  for the purpose of the Tier 4
 Study,  broad generalizations  can be made.  Elsewhere  in  this  document where
 literature data  are used quantitatively,  only data of verifiable-high  quality
 were used.
      In the  following  sections,  emission  test  results  are   presented
 separately for each  study along with  appropriate tables.  If  available,
 information is provided in the text  about the  combustion unit, analysis of
 feed samples,  or  identification  of precursors present in  the feed materials.
 When available, detection limits are specified.  If analytical methods other
 than gas chromatography-mass  spectrometry (GC-MS)  were used, that  information
 is  noted in the text.
 3-1.1   Summary of  Stack  PCDD  and  PCDF Emissions
      Table  3-1 presents a summary  of stack  PCDD  and  PCDF emissions  data
 available  from the literature for  combustion sources.  The data  for  each
 source  category  are discussed in detail  in  Section 3.2.   For  each source
 category the total  number of  facilities tested is shown,   as is  the type of
 sample.  The ranges  of detected  emissions and detection limits are shown for
 the  2378-TCDD isomer, PCDD's, and  PCDF's.   The .column labeled "Detected
 Range"  represents the  range of average concentrations  reported  by  the  various
 literature  references.   If only one  value is  shown it  represents the average
 value of all  tests for one unit.  The  number of sources  having detectable
.emissions  is  shown  in parentheses under the range values.  Detection  limits
 are  shown  in the  column labeled  "Detection Limits"  when  a source had less
 than  detectable PCDD or PCDF  emissions.  The number of sources  having less
 than detectable emissions is  shown in parentheses  under the detection  limits.
 Comments are provided  if the results  were  affected  by  such  things  as
 contaminated  fuel,  upset operating  conditions, or  use of a  nonspecific
 analytical  technique.  In the table,  the  source categories are ordered by the
 highest  value  of the PCDD emissions range.
                                     3-3

-------
0) | -O >*-
*. 1 r- f !_ O O











«
»
U
^
*
1
1

»—


p


U.
i
§
*/»
V)
ac
tu


S
v»
u>
S
5
9
O

c5
V

i
s
T

Ul
-


















V

£

c
X.
e
£
-
 4-iiQU1 CD -^ (J U "O 4^ t.
« c ^- — . a>  4J — a)
S— O O CD f— U) 4J Wl O C Q Ul ^J Ul •— 4J -^- £4>>
t. -M Q U UQO^OO "5 +J CD « TJ C —
^» i- co a c o co ui CD o o x •— T3»o«a-— c

* O Ul I- O t- O — O O 0) Ul TT 3CU14JL.C

•-* ui ui ^- -CDC*
» »C7> CO Ul Q. >— +J ••Xl^**-WUl^(O CL**" fO
«^t __00^0 ..0 ^°
m CM 3 ^ o vo CM
^ en ~« eo nr en CM
„
o a
s 2
II < CM T ^ A S §:£
O -v
• O
O v
V
O
^ o in
mo •
co ^ •> *-* o — » m*^- P-**
»eo CM en p- •-• i en o co <->
O»«— CM •— ^* *-» m -^ z m •**
I I —«
P* Ul •
m • o
S 2
•^ d
i*. —. ^»-^. i ^* a.—
II •< fl 1 ^ i— I -^ CX^^
2"^ CM*^ fO*^ !"*•*-'
g =>. •?
9
V
o
r*- «3 o «*

^ ««»*** rC*^ )2 -* S«
eo co <^ at •— i m i in Q m*^
•cr-w* ^ •— i »- <7 ^> Z • **
i i a *o „ ^
r** • ? cs
m *
0
in —»
r-t * O
|| 1 CM ^ ^H ^-t I
g- s-
O C3
V V
in CM
i T S i g- o i
*S - o-
0

(,3 CJ O C5 C3 O
U. U. U. U. U. U_

If
1
c i-t ' CM
C (— I CD
e 4-J *o
*c -o «: i-» r-i o
c « x m in z
CO » +j CD
CX O CD Ul — 3 ^
»- O • C € 3 ^- J3 1

-UJ3 t-0 w 3 CJQ.
C Q
*COO^O O O GO
IS i






























^
ID
3
C

C
o
U





























3-4

-------


                                               v» g> +J fp

                                               ^§se
                                               U. U «~ fl3

                                                                    iS3"
  5S^f eQ°iu!J=

  •— -Q •*•» (D O Q ~ •  — •

  3 g U1 V Q CJ

  r- O C >4- U r

  u_ o T- <« cu :
:i^SS!


    o
                                 SS
8
£•?
O +•
O) 
-------
      I       4J  in   <^
      U  . Ul CD  C   O
      <3 — C VI  O


'b'c (Contlmtotf)
1
i
iu
.J
i
u*
I
V*
s
VI
VI
3
Ul
•^
fe
1
i
1
Ul




i
!•
,
•1
1
u
£


1
C
i

t      r-t
                                                                             8
                                                               o

                                                                o
                                            3-6

-------








—
1
e
"c


*"*
u
J3
'ui
I
s
h-
j
4*
P



u.
<
<
z
o

CO
LU

^
H-
trt


s
^
&
C/3
*
1
m
UJ

.













C.

I
c





m"
c
*-
ti.
CL


§


S-
s *
0) __
o

» CD
4-* '
u e
(D tl
4-» a
Q





^
c
c
c





e
a
4J *|
U •*•
•§'
°

•o

ij s
"at
0


i

c
c
H
00
r-
e
0
*- U!
4J 4^
O —

1 =

) Detected
Ranne
u/
E >
us J—
V)







r-
>.-=
u a
O -H
O) Ul

U fl
3 O
0 h-
V) u.
O ui
"Q S "3 J) £
. U ^J O 4J Ul 3 •*-
•^ S "e ** jl ' c * * "*
C L. — C •— 1 O 'Jl "O CO
O» OJ O
X — en — +•» GL
 O •*- H- O U_ CJ Q. U.
O O O O O O
r-l 
I i s
1 •

•*-» Ul
' ^- C
e o
3 -2 ' '
Ul
| *
t_
A Ul
ui • .u c
03 31 "in
- S |
"So
' H- ^ «—
O +> JD .
O U) 4J
•— O S
> *~ 'o
01 -0
a =
Q) -^ e • -
tl > IT)
0) J= -M
19 Ul Ul
 
£ e • iq
(D Ul 3 ui .c
t. e
(D VI ui O ui
1 ui "ui ii
«j (D o ui c
4J 3 ^ i- 3
"O fl C S >*-
>  ui ui +j in ui
•^ 4-» C Ul -^ CD rtj
Ul O O 0) E t- S
>» 0> •»- r- — Q.
,- 4J Ul ^- (D -o
«J CD Ol "O (_ m
C -O -w na c 4J
Ul 5 T- O) Q
0> r- 0> 4J C Q.
in (D (D  o fl fl
O fl 4-» £ m
•— u d} e +4 a c
•O ID "O  O > X 3 Ul +J
•o  .QQ
 - a
OS t- -^ u Q < O
ui 5 c CD o n_
e is ^ oe -a +* 3 +J
*O 0>E i- B g O i— O "O
2 -MS wo 3 z LU za>
S oc uj= e 4J
® Q 01 II II II U
S 4^ ^Q£
3-7

-------
     PCDD's and PCDF's were not found in stack emissions  from  all  combustion
sources.   Utility  boilers firing  PCB-spiked waste oil,  a lime  kiln,  a
coal-fired utility  boiler,  and the incinerator ship M/T  Vulcanus had less
than detectable PCDD and (if analyzed for) PCDF emissions.
     In general, where combustion  sources  did emit PCDD's or  PCDF's, these
emissions  appeared  to  be dependent on  the types  of fuel  or  wastes being
fired, and whether  or not the  combustion  unit was designed  and operated
specifically to  achieve  the  destruction  of  potentially  hazardous wastes
including  PCDD's and PCDF's.   However, sampling  and  analysis methods  and
facility design and operation may  vary considerably between studies, which
makes direct comparisons of emissions test data difficult.  For most of the
sources  tested,  the magnitude  of  PCDD emissions  was  comparable  to the
magnitude  of  PCDF   emissions.   Most  of the  studies  analyzed  samples for
PCDD's; some studies quantitated 2378-TCDD and  the PCDF homologues as well.
Samples were taken  primarily  at the stack outlet  location and not  at  the
control device  inlet.  The range of measured PCDD and PCDF concentrations
varied from one to  four orders of magnitude.
     Based on  the   Tier  4 literature review,  two municipal  solid waste
incinerators were the  most frequently  tested source  and  had  the highest
levels of  PCDD's and PCDF's  in stack gas  emissions.  Facilities  located  in
Europe had higher PCDD emissions than North  American facilities.   Commercial
boilers co-firing spiked waste oil  had the next highest PCDD concentration in
stack gas  emissions, followed by emissions from one tested unit  combusting
PCP-treated wood.   However, another  combustion  unit designed  to  incinerate
PCP-treated wood had less  than detectable PCDD or PCDF emissions.   Of  the
remaining  source categories  tested,  sewage  sludge incinerators,  hospital
incinerators,  and two rotary  kiln  hazardous  waste  incinerators had the  next
highest PCDD emissions, respectively.
     Emissions data for the 2378-TCDD isomer were  available  for five source
categories.  Municipal  waste incinerators  had the highest  stack emissions of
2378-TCDD, followed by a  fluidized bed  system used to regenerate  activated
carbon, and an  industrial  boiler.   One coal-fired  utility boiler and  the
                                     3-8

-------
 incinerator  ship  M/T Vulcanus  had  nondetectable levels  of the 2378-TCDD
 isomer.
 3.1.2  Summary of PCDD and PCDF in Ash Emissions
     Table 3-2 presents  a summary of data on the PCDD  and PCDF content of
 combustion ash samples available  in the literature.  The data for each source
 category are discussed in detail  in Section 3.2.  As  in Table 3-1,  Table 3-2
 shows the number of facilities  tested  and the type  of sample.   The range of
 detected emissions  and detection  limits for the  2378-TCDD  isomer,  PCDD's and
 PCDF's are also  shown for each source category.  The source categories  are
 ordered by the highest value of the PCDD range for each category.
     PCDD's  and  PCDF's  were not  found in  ash samples from all  combustion
 sources.  Three rotary kilns, thirteen coal-fired boilers, two diesel cars, a
 lime kiln, and a sewage sludge incinerator had less than detectable levels of
 PCDD's or PCDF's (if  analyzed for) in ash or particulate samples.
     A  rotary kiln operated  without  supplemental  fuel   fired  in  the
 afterburner  had  the  highest  PCDD concentration  in particulate  samples.
 However, according  to the author  these results were skewed high by the use of
 a nonspecific GC-MS packed  column analytical  method.  The  same  rotary  kiln
 operated with  supplemental  fuel  had  significantly lower  PCDD  levels  in
 particulate samples.  PCB fires had the second highest  PCDD concentration  in
 soot samples, and also had the highest PCDF levels.   Unlike flue gas samples,
 the level of PCDD's found in ash  samples were less than PCDF levels  for  most
 of the combustion  sources where both PCDD's and PCDF's were  analyzed for.
 Municipal waste incinerators had the next highest concentrations of PCDD's in
 ash samples.  In  general, European facilities had higher  levels of PCDD's
 than North American or Japanese facilities.  However, two European facilities
 had less than detectable  PCDD's  in  ash samples.  North American facilities
 had higher PCDF levels in ash  samples  than European facilities.  A  natural
gas-fired residential heater had the next highest PCDD levels in particulate
material, followed  by a  commercial boiler  co-firing used  automobile  oil
 spiked with  organic compounds, and  a  hospital  incinerator.   Residential
woodstoves had detectable levels  of PCDD's  in  ash  samples, but  some of  the
fuel  was  reported  to have  been potentially  contaminated.   Other source
                                     3-9

-------
t
u
*
ft
I
§
§
1
CONTENTS OF t
s
|
S
|
i
I
S
i
7
(U
1



V)
4->
C
0
1
C
<
O
c
o
L.
<
ec



0
V
i
Q.
C
1
i


c
c
1
*
J*
§*
4J *•
g a
4-» C
gS
r
§
•*. u

T»
+•• C
1°
1.
e 1
a _
a
Si
*j a
fl
&1
0 4-
If
3 t:
1  Q e  4) G
O'
O *^ C Q t. IQ>*B Q tj ^ m — g
j£Oi— O*Jg-OO) *J fl ^- >QU4JO
CL (J t. O f— W U QIC g (B _J U •*-•
IU1 VI ^ O 4J &*• 3 EE— >. O -M U1 4-» fl
4-» C -M **- ^»*> J3 0) U fU* 0> 3 L. r— -O 0) 0> . »— 0> 3
3 C ^ ^ (D tfl ^ O O. O C Q '"l. -f-1 °. O * J
I/I <9 O L. O -M <^ U L. Q ifl C g C F—
0)>*Cin£O3t*- 3—3 03'O'QO'Q
o;^fl3-Mz«*-ig v> >^ .a ovicuiou
» ° ° ° " °
M CM (N O 3
*O O ^O >-* i-4


1 111 <5
z ^*
i
q) ^.
. , < =

O 0)
in ** r* —• \o ^» — *
mi 8= ?= 2= §
en o*

• v= A- , ,
CS| •— . v
V «•••
V
o
eo^ Q g i i
1*» V <•» .c *J
^ U L. ui L.
>9 19 19 < «a
Q. Q. Q. Q.
t? O n
4- U. *ffl UJ
1 * - %
C 4-> ^ +S
» r- 1 e r^ «a n3 m
t ^^ g r-» +* r- T L. rn
C W g UC ^40 bJ
H- Q (D 4_J C
a ^ *"" c § "^ c
4-t^O. d^. » >% a 3 co >\
t. o u v> aa ^- u
tf « "s. iO -v* ••- 03
= 4-» » 4^X t_ J3 4J
co o 'o a o
•= a £ £ i a
JO O O O O
S
3-10

-------


«1
+J
c
a
c.
1
1
£
a


"c
1
U.
1


c
c
e
c

_
c
c
H
00
e
o
t- U
O i-
£J
o
•o
V C
o c
Is
!«
1-
ll
a
I,
81
a> -
o
•W C
u c
*— a
E »
CO


i—
II
3*
£5
L.
C
U
o
g
£
2 £
« CM
°* O
O
vo

1

o
o
o
§~
t
o
2-
V
. 1
«-t *—
s
1

§-
10 CM
o\ w
in
1

s
1
u
1
c
£
13 O
•M VI
VI Q)
•O il)
^ «I r-
Olf -J3
^- U
J= 0 Q
•*-» > +J























           a.    ,«a-(->
43°
r-  V)

SS
      i-
                              ui ^^   O> ^
                                      -
K ^  2 —
^ !C  *^?3    i

m —   i  *•»
                                     i".
                                     e     f-i

                                     32
                                     3



                                     O
                      3-11

-------








1
5
1

u
JO
•i
5
5

8
fe


S
•£
8
&
9
1
\
I-™
S
a
s
_i

&
|





pj
m
1


















u1
4-1
C
(
1
C
<
)
c
p
i
9
V
1
Cr



^
c
i
|


§

11

a
?<
c
i
a




«•
c
1
1



c
1
1
£

5
e
o
4-* 4-
el

o —

2 1
I1
i
— u
** 4-
|-

|l

I5
o
•— fl
O. c


07





EC
C
4-
V
0 S
O "*
e r-
o *
*" *

en 4.
>* » 1 "O
•o — ej -3 e  c u c
o >* .-a — »o
O ^- .a £ u t.
r*» O O O 4J in
X g tn >»•— .0 CM
!O OJ « -= JJ C C P-*
^»£ £ e u c^
\ooc a> — o c to o
z o < u. ^= o -^ a. *• .
^>
CM CM OV


S

o ^*
1 ' ic;
o
0
V

1 1 O


0

is
0^
o
o
V
S~ ^-j



1 1 1



• 2 1 I
o~



u. , «0
ij3 +j g ,— 4-»
o — 5 o c o
o m » u z o o
^ ° °
CO



1


P:«
P«- ^H




1



£ 2



1







<
u.



r-i
L.
O
o
CQ
^_
fl


i
s
J

0


                 I "a
                 3^
3-12

-------











—
3
e
v
c
o
o


u

J3

fl

%
Z
o

to
i
5

H-
Z
UJ
Z
8
u_

i
§
§
LU
Ql
I
E"-

s
>_
o:




CM
1
en
3

h-
















in
•M
c
a
E
•s.
2
1
£
 -O C S E
tj a) o t. fl  r— >»

« •— u :o a. c • +J o -H-^ e
*«5E uijaot- v-^ -—^ o
r**o e-"-o> ui ^ tnoi"
vO^4J U. -0 O -H = O0>* O>* C
^i»c a o •— fl t. a.>-o o.-a ^
!* 3o ON-MF-S EOJEO® o
**^o H- — in ^- .Q o m N O'^N wi
Ul -*.-03<0U U>»O>«

—( fe 0) Q 3 S tn +J 03 O 'O tDO'io O
OO O »— O C «*- CUC Ct-c O
« to .a h-u.O><-'O O *^ <« O *^ ^ H-

O O O « O " O O
^r .m «4^4CMCMCM^H
U) —4 CMCMia O *~> O •" • O -—* ^ E <"^
OO5^ * t 1 OCMOCMOt-HOfOCM
oo*- . ^. . « . « . tn *-.
• • O O O O "*v
O O V V V V Ol

0
•

is a= °:2 js iiii
§- ^
o
*•* +> O) C3) CB +J
- uu (- u . c c e u o
x: +J n>«a oca ™- — •— m «
M u +J (_ +J t. a.- Q. a. +j u.

a. -^x -»-x ut-t_— x
(— I U.UJU_U4UOOLl_UJ
•^ to tn w>
u C9

(rt •-
§4-
L. i— t
•ut a ^
in c uj r-i
•0 OJ O m 1-1
a c t_ ^ ^ CM CM
x a: c am ^^ CM
u. ->-u -« U V 19
IS ^S 4 " • 1 )-^0>-"
Ul (D CJ «*- -O T3 *— t3 T3 ,—

•S •*• "5 ^••§ a> tn o •o . tn
EJtn QO -^ *(j r« Q>r**q +- < j2 oj e ^ c 5 —
UK ^ » C _| 3 Q 3 _J O
Q O ^ ° t; ° O O O O O
3 '< <
3-13

-------








9
9
jl
*J
I
0
J3
S
S


j.
o
V)
S
».
§
u.

3f
|

g
S


i




g
tu
H"
»•*
*
>~


1
CM


Ul
i











«j
e
§
C
V.
e
§
£
s
a



..
^
c
"-
g



I,
§1
I:

"i
II
s
a




-
1


|





~




c
H
 -a o •
e o "a — -a *J
O .c co "a in CD c
*»eeuiaioi *•* CD
U *J • 5 ^ 4J f— 4-»
W =3 O Q CD "O *- Ul *O .0
Lu O U O •"• CO C O*O
O **• £ jQ "O Of O
S CDr— O O +J U Ul C >»
*" O^£ZQ 030 u.ja
.. O " 0 O O
O CO CD ^f ^
S S S S S



II III



^ *••» cO ""^
.^ .^ 1 II


l/l
*"J
o
§!-• < ^^ < ^^
^- z-
o
V



»3 • i i i
0



li ill






ii i it


CD
• Ul Ul
*J 34-1 CD
< t O ui • < ^
"• £ f-S "• <
CD


C
4J
r-i 03
Is "

ig *H c


"52 ^ Si
" c ^ e
! | 1 j
ai •
c +• g
| 3 3
5s 5
° s
•M 5
U) U
3 O CD
.C d
X Ul iQ
LU -f* •*-*
Ul (J
U CD 0)
CD -M -H
C >— -a
3 *(5 C
X 4
II O 4-»
? 3 S

0 CD _C
•M CD JS
*4J CD • U
X £ Ul L.
LU 4-» CD 3
3 O
U Ul f— U)

**• ui CD
U. CD O) U
CD a, «j .a
•o t- §
*3 4J 5 C
o *• +•• c
1 II "
a. j= e •
Ul 3 Ul
It C
Ul Ul 0
U Ul Ul
19 CD CD Ul
4-» P- -M g
x 
. fe SS. S
CD 4^ O fl
TS tZ r- **" «
c e 4->
1 2 °I 5
f*™ >-*(/! C
fl
Ul f
. v ** .
•o ui
Ul Ul CD Ul Ul +*
^ 3 4-* C Ul •—
ui o u o CD g
,_ w 4J *U1 *" JH
«J «J 4> Ul *Q
e a) xS •»- *• Ul CO -^
O O i1™ CD 4^
Ul CD CD CJ U
3 CD > *— U CD
A3 Ul CD J3 3 4-»
u * Ul "O
J3 Q. CD O
O O  ^; L.
f— 3 <9 Ol » <£
« ^ «. -5 c c
> 4J O > X 3
o L. • CD c — x
S ui g c g J
X H +J CD 4-»
1 CD £ CD M
* U. * A *""
3-14

-------
categories having detectable levels of PCDD's or PCDF's in participate or ash
samples  were activated  carbon regeneration,  automobile  emissions,  wire
reclamation, and a cement kiln.
     Concentration data  for  the  2378-TCDO isomer were available for ash and
soot taken from  seven  source categories.   A rotary kiln burning tars, solid
waste, and natural gas and operated without  supplemental  fuel  being fired in
the  afterburner  had  the highest concentration  of  2378-TCDD  in particulate
samples.  However, the same  rotary kiln had  less than detectable emissions of
2378-TCDD when  supplemental  fuel  (tars and  natural  gas)  was  fired  in  the
afterburner.  PCB fires  had  the second highest  concentration  of 2378-TCDD in
soot samples'foilowed by a fluidized bed combustion system used to regenerate
activated  carbon.   Municipal  waste incinerators  had  the  next  highest
detectable levels of 2378-TCDD in fly ash samples, followed by filter extract
samples  from vehicles  burning  leaded  and unleaded gasoline.   Other  sources
having detectable levels  of  2378-TCDD  were  a natural gas-fired  residential
heater and residential  woodstoves.   Particulate matter samples  from a tar
burner had less than detectable levels of 2378-TCDD.

3.2  EMISSIONS DATA FOR INDIVIDUAL SOURCE CATEGORIES
     In  this section, the available PCDD/PCDF  emissions data for individual
source categories  are discussed.   Sections 3.2.1  through  3.2.12  cover
municipal solid waste incinerators, sewage  sludge  incinerators,  fossil  fuel
combustion,  wood  combustion,  boilers  co-firing wastes,  hazardous  waste
incinerators, lime/cement  kilns,  hospital  incinerators,  wire  reclamation
incinerators, PCB  fires,  automobile  emissions,  and  activated  carbon
regeneration furnaces,  respectively.   Various  experimental  studies are
discussed in Section 3.2.13.
3.2.1  Municipal  Solid Waste Incinerators
     Table 3-3  presents the emissions data for municipal  solid waste
incinerators.  These data  include  information  available  in  the literature
through  July of  1985.   Subsequent  to this report  EPA  has  compiled
information,  including emissions data,  for a number of municipal solid waste
                                    3-15

-------


















i
1
g
g
IU
t/)

X
0
*"«
s
£
o




£
5
Ul

s

O
1
V>

T


u>
1-




















W
1
1
e
o







"I

c
e
V.
4
<
I
•e


O
7
(fl
l_
+J
i
c

c
c
u-
m
U.
l^
C

^
g
OL











g
jj



&
c

ee



§
S








c
g;



&
H
o:





S
£



•





g
T
eo
£



8,
e
•a
» 19
l-u.
» m ^H
*»• in » ro co
r- o •-! » r-

CM O^ O •-• O *O <-»
CO (O » Csl ^
5^2 ^.4SS g cQJ^^
» » » *
S22 5SSS 5 23S S
o o o
vo o o r-
o m —<  in m
- m •
<-> CM

co_ **
s ci

• III
in vo
cn I
1 00
CM •»
O O

"V , . Ja
a) CL
C 11:1
00
^H cn
cn CM
•-4



C3 < U> < <
U. U. U. U- U.


cn ^* eo ** CM
CM m




,.
to
ce «(
O (J

S a
a i | a
3 * 8 • g
2 t S S

g a uj -g








































_
VI
0
 O
in < u
SO 34-t
3 >> U» O
§i- i- (0 2
U_ u. m
6 u
e H n T
1 W) 0
1 CD < * z
B u. i^ 
-------
incinerators.  These data will be presented  in  a  forthcoming report entitled
"Municipal Waste Combustion Study" (8 volumes).
     In  1978,  TCDD's were  detected  in the  emissions  from  the  Hempstead
municipal waste  incinerator (MWI)  on  Long  Island.   Since that  date,  this
source category  has  received considerable attention in  the  United States.
The Canadian Government has  identified  MWI's as one of the major combustion
sources of PCDD's  in  the  Canadian  environment.73  Numerous  tests  have  also
been conducted in Europe and Japan.
     MWI's can be classified as either  large mass burn units,  refuse-derived
fuel (RDF) units, or small modular units.  There are approximately 43 facili-
ties with modular units, 45 mass burn  facilities  and 8 RDF boiler  facilities
currently operating  in the  United States  and Canada.0   The mass  burn
facilities are responsible for the majority  of waste burned.
     The following subsections summarize the PCDD and PCDF flue gas emissions
and fly  ash  content  data for MWI's operating  in North America  (including
Canada), Europe  and  Japan.   The  emissions data presented  are  based upon  a
review of  the currently  available  literature  that  reports  CDD  and CDF
emissions from MWI's.  Thirty-eight articles were reviewed.
     Emissions data  are  available  for 62 MWI  facilities.  Fifty-seven  are
reported to use  an electrostatic precipitator  to  control  particulate matter
emissions from flue gases.  Twenty-six  of the  facilities  are located in the
United States and Canada,  34 in Europe, and 2 in Japan.
     3-2.1.1  United States and Canada.  PCDD emissions from stack testing of
9 facilities located  in  North America  ranged  from not  detected (NO)  to
22,000 ng/m .  PCDF  stack emissions for  9  facilities ranged  from ND  to
15,060 ng/m .  Detection  limits were  not specified.   Both of these  upper
emission values  were  reported in an  EPA study of  a  modular  incinerator
located in Lang!ey,  Virginia.   The incinerator is  characterized as  being
susceptible to upsets caused by burning grass clippings or wet refuse stored
in an open pit.    Two  facilities  are known to emit  low levels of  PCDD's
     'Resource Recovery Activities,  Citv Currents.   April  1985.
                                    3-17

-------
 (35  to 146 ng/m3) and  PCDF's  (50 to 246  ng/m3).   One is a  modular unit
 equipped with  a secondary chamber for combustion of off gases and the  other
 uses  RDF that  is  stored in  a silo and is very dry when combusted.  '
      The PCDD  content of fly ash samples  from 21 MWI's ranged  from  <0.5 to
 2,300 ppb.   PCDF's from 14  facilities ranged from <0.5 to 3,100 ppb.
      3.2.1.2  Europe.   Flue gas  emissions  of  PCDD's from  eight  MWI's located
 primarily  in Italy ranged from ND to 48,900  ng/m3.  Flue gas emissions  of
 PCDF's from  seven facilities ranged  from 37 to 7,460 ng/m3.   The highest  PCDD
 and  PCDF emissions were reported  for six MWI's located in the Lombardy  region
                   A *3
 of northern  Italy.    The report  contained no  information  describing  feed
 composition,- combustion  design  or  operating  conditions.   However,  each
 facility does  use an  ESP.
      The PCDD  content  of fly ash  samples from 31 facilities  ranged from <0.5
 to 3,540 ppb while PCDF's from 19 facilities ranged  from  ND to 1,770  ppb.
 Detection limits  were  not specified.
      3.2.1.3   Japan.   The PCDD content of  fly ash samples from two facilities
 ranged from  2.4 to 4.8  ng/g.
 3.2.2 Sewage  Sludge  Incinerators
      Table 3-4 presents emissions data from  two studies of  sewage  sludge
             24 235
 incineration.   '     An unpublished  study  reported emissions from a single
 multiple hearth sludge  incinerators with  a  water  scrubber.235   Operating
 temperatures were reported  to be  in  excess of 1,000°C with a  feed  rate  of 13
 to 15.5 short  tons/hour.  The results were not reported in terms  of  specific
 homologues,  references  only to CDD's and dibenzofurans were made. ' For  three
 sample periods flue  gas samples  contained CDD's ranging  from 483 ng/m3  to
 1,140  ng/m   with  an  average of 739  ng/m  . Dibenzofuran  concentrations  in
 flue  gas samples  ranged from 501  ng/m  to 2,248 ng/m3 with  an average of
 1,213  ng/m .   No detection  limits were specified  for these unpublished
 results.
     The second study    analyzed  emissions  from incineration  of aerobic
 sludge.  Fly ash  was  collected by means  of "dust abatement",  organic vapors
were  trapped with a  water condenser, and ashes from the combustion  process
were  collected by grab  sample.   The investigators  reported  PCDD's  to  be
                                    3-18

-------






















g
p
1


^4
g
t-H
S
3
5*

LU

iJj
CO


^

§



o

|

•
i
m
Ul
1—























Ul
o
o
i
ID
V.
8.








+J
(D
•H

O
0
a
1

e i i

m
CM
*-4

o
Scn6
—1^1 1
1 0)
m e
CO


a
m
e
^ ^ o
U1PO Q O
C — Z Z
o»
m







i i i






i i I


_ _
Ul Ul


3 Ul
UL <



1-1 "*







5
U
® in

•— Ul
1 §
















































•
e
Ul
i
3
|
o •
£ ^
•O 3
S -P
U)
•a
c c
C "O
Ul Q>
f t
»- (D
•o t.
c
in
Ul IO
a
>  —
Ul r-
O -0
e e ID 
S+J 
-------
"absent" from both  "ashes"  and  "fumes."   Detection  limits  and information on
the incinerator were not reported.
3.2.3  Fossil Fuel  Combustion
     Table 3-5 presents the emissions  data  from fossil  fuel-fired combustion
units.
                                             94
     3.2.3.1  Coal  Combustion.   Haile  et al_.    have  reported results from
research conducted  as  part  of a nationwide study of  organic  emissions  from
utility coal combustion.  Results were reported  for four of  the seven plants
comprising the complete survey.   Samples analyzed included samples from  the
flue gas outlet (downstream of the particulate emissions control device), fly
ash emissions, and  coal feed.   PCDD  and  PCDF homologues were not  identified
in any  sample from the four coal-fired  plants.  To  maximize  the  method
sensitivity, all samples were analyzed using five-day composites.   Detection
limits for PCDD and PCDF homologues  in the  flue  gas analyses  were 0.25  ng/m3
for mono- through tri-CDD;  0.10 ng/m3 for tetra-CDD; 0.50 ng/m3 for penta-CDD
and hexa-CDD; and 0.70 ng/m3 for hepta- and octa-CDD.   For solid feed and fly
ash samples, detection limits for the PCDD and PCDF homologues were .025  ng/g
for mono- through  tri-CDD;  0.010 ng/g  for TCDD;  0.050  ng/g  for penta- and
hexa-CDD; and 0.070 ng/g for hepta and octa-CDD.
                      97
     Harless and Lewis   tested  fly  ash  samples from  seven coal-fired power
plants and found the samples had non-detectable  levels of TCDD  at an  average
detection limit of 0*002  ng/g.  Also,  in an unrelated study,  DeRoos and
Bjorseth   analyzed one fly ash sample collected from a coal-fired combustion
unit for TCDD's.  None were detected at a detection limit of 0.002 ng/g.
     These results  are in  agreement with  those reported  by  Kimble  and
Gross    who analyzed stack-collected fly ash from a  typical  commercial  coal
combustion facility burning a  low  sulfur,  high  ash  coal.   The  chlorine
content of the input  coal was  50 ug/g and  the  sample was  taken downstream
from the electrostatic precipitator.   At a  detection  limit of  0.0006  ng/g,
TCDD's were not detected.   Kimble and Gross conclude that  a fossil-fueled
power plant  is  not  a  large source  of TCDD.   This  contrasts  with  the
conclusions presented by Dow Chemical,  which analyzed fly ash  samples from  a
                                               gn
coal- and oil-fired chemical  plant powerhouse.    The  results of the Dow
                                    3-20

-------


















z
o
1— 1
V)
=3
m
Z
§
_j
LU
=>
UB
^ J
I—I
ts>
V)
2

••
2;
Q
Lu
t
O

LU
1
CO
in
m
LU
_i
?













in
CD
CD
i_
CD
CD
a:
.u
c
CD
•p
c
 CD i-
•p -p
<+- in > 10
r- U.


TJ- ^ r-»  ^H •-(


IO ^ SM »
5 CD CD t-
r— r- 3

>» t- "5 "5 r-
•P CD CD m 3
N- r- CO

•^ o    >
i-   i— O


CM VO tn CM
^o o^ ^o
I— 1
«
in

i— i
i I i i




o
,00




1 11 1


CD O)
O) v.
•c3, g
c o t

GO 5


III!




O>
^^
• O)
o o c
Z 1 Z
*o
•
o



•p
U CD CD <
rtJU-U.LL.
Q.
C C
CD CD
"O «C < -O «C
CD i-H CO i-HCO i-ICD i-ICO
CO = = * =

u
CD
f— Lu r— * i—

o S « o 10 *^
•am CD -P
•o -o CD c
C C (. * r- CD
10 10 -^ y 10 t3
*i3 r— I— U" «-•>-+•

O «9 IO r- in -P CD C
o o o z
3
O)
o
§*~
.
o
-C

o
o
"S8
a
0
O t-
o o
lO
•p -o
Q. 0)
CD N
I >>
O C
o 10
.
X •—
CD C






*c o
O CD <«-
^^
IL. IO *O
O r- CD
O 3 N
H- 0 >»
GO -P IO
I-- t- C

CM O.

S~ JC O
o u c
4- IO
•P 0
in in o
— o
in •• i
>> • ia
•- -p -p
IO t- C
C IO CD
< Q. a.
0 -0 CD

ia

•o
O i—
C TJ C
CD O
m -P
c o c
IO 
-------
study are presented below  in  Section  3.2.3.2.   Kimble and Gross suggest the
difference in TCDD emissions between their study and the Dow study may be the
nature of the fuel sources, including total chlorine content.
     Ah!berg et  a].,   analyzed flue gas  samples from a  265  MW pulverized
coal-fired boiler equipped with  an electrostatic precipitator.  The  boiler
was firing Polish coal with a low sulfur, high ash content.  No 2378-TCDD was
detected at detection limits ranging  from <5.4  to <6.8  ng/m  .   The  2378-TCDF
isomer was not detected at detection limits ranging from <0.86 to <1.1 ng/m .
     3.2.3.2  Oil  and Coal Combustion.  Particulates  from the stack of  a
coal- and oil-fired powerhouse at  a Dow  Chemical  plant were tested for PCDD
emissions.62  TCDD, hexa-CDD, hepta-CDD, and OCDD emissions  ranged  from  2 to
38  ng/g with  TCDD's  and OCDD detected at  levels of 38  ng/g and  24 ng/g,
respectively. The concentration  of total PCDD's was 68 ng/g.  The 2378-TCDD
isomer was not  detected.   Detection  limits in this study  were 20  ng/g for
TCDD and 10 ng/g for  2378-TCDD.  Detection limits were  not  specified  for  the
other homologues, which were analyzed by electron capture gas chromatography.
The study did not report fuel analysis or operating conditions of the boiler.
     3.2.3.3  Oil Combustion.  Ahlberg et al.  analyzed flue gas samples from
a 250 MW boiler fired with  a low ash, 2  percent  sulfur,  heavy  fuel  oil.   The
sample was  taken after the heat exchanger  and before the  electrostatic
precipitator.  The 2378-TCDD isomer was not  detected at detection limits
ranging from <4.2 to  <7.9  ng/m .  The 2378-TCDF isomer was  not detected  at
detection limits ranging from <0.67 to <1.3 ng/m3.
     3.2.3.4  Natural Gas Combustion.   Dow Chemical  tested particulate matter
which  had  been  removed from  a  home  electrostatic  precipitator  in   a
residential, natural  gas-fired forced-air heating  system.   The collected
material represented the accumulation of material from  six  spring and  summer
months of  operation  of the  precipitator.   The  particulate  matter  sample
contained 34 ng/g hexa-CDD,  430  ng/g  hepta-CDD, and 1,300 ng/g OCDD,  for a
total of 1,764 ng/g.   The 2378-TCDD isomer was present at a level  of 0.6 ng/g
with a detection limit of 0.2 ng/g for the analysis.  Other TCDD isomers were
detected at a level of 0.4  ng/g, which was also  the detection  limit for  this
                                    3-22

-------
sample.  No detection  limits  were  specified for the other homologues which
were analyzed for by electron capture gas chromatography.
     3.2.3.5  Coal and Refuse-Derived Fuel Combustion.  Analysis of  flue  gas
emissions from a  coal  and RDF-fired facility located  in  Ames,  Iowa, found
less than  detectable levels  of  TCDD,  which was  the only CDO  homologue
analyzed for.     The detection limit for TCDD was 5 ng/m  for vapor samples.
This is  a suspension-fired boiler  that  burns coal  with  15  percent  RDF.
Small,  uniform, 2-5  cm pieces of RDF are  produced  in  a shredding and  air
classification process.  The facility operates with  a  combustion temperature
of approximately  1,200°C  and  produces 35  MW of  electrical  power from steam.
The unit  is reported  to  be operated at  approximately 22 percent excess  air
and uses an ESP.  Another study describing emissions testing at this facility
                                                                    196
                                                                         The
reported that PCDD's and  PCDF's  were not detected in the flue gas.
detection limit for PCDD and PCDF was 0.25 ng/m  for vapor samples.
3.2.4  Wood Combustion
     Table 3-6  presents the emissions  data  for combustion  units burning
PCP-treated wood and firewood.
     3.2.4.1  Residential Wood Combustion.  Four studies have been  conducted
on PCDD formation from  the combustion of firewood.54'62'155'167   Ash  samples
were collected from 24  woodstoves and two fireplaces.  The woodstoves  tested
were located  in rural   areas in  three different regions  of the  country.
Presumably the  wood  being burned was untreated, that  is,  it had not  been
exposed to  fungicides,  herbicides,  or  wood  preservatives.   For  the  24
woodstoves tested, PCDD concentrations in ash samples ranged from 0.007  ng/g
to 210 ng/g, with a mean concentration of 23.4 ng/g.  The penta-CDD homologue
was not analyzed  for.   '     The 2378-TCDD isomer  was  analyzed  for in  17
samples.  Two samples had non-detectable  levels of  2378-TCDD with detection
limits  ranging  from 0.0009  to  0.0014 ng/g.   The  other  15  samples  had
concentrations of 2378-TCDD varying  from  0.001 to 0.20  ng/g  with an average
concentration of 0.05 ng/g.  The authors  of one of  the  studies,165 in which
18 woodstoves were tested, attributed some of the variability in  the  results
to differences in woodstove design and sampling points.  They also  suggested
                                    3-23

-------



















g
£
0

J
§

g

w

cc
w
c?


10

tn

g















in
ID
I

cS



^
c


c
rt









tl_
|




&
B
a
cx

g
£






i



§)
I


B
I







c
c

t~
£
m
CM




ffl
cn


cr.







E
10



a
CO


in
^•ol
**— Ul (0

• _j
0 d
O 4*
r~
 5S« §m-2
S cni ill i «"_; _ i '°*™
OOP~P~ _2°^) ,^2-?
• rH*. ^m^ y2:r* ™
O CM «— * ^ ^«fnOI OCMwl
ON5 (_(c CMC
u u u
O> O) O>
g^gNl1 •S-S'oo olom
cQ ^M' ^ "o o) 01 r^r Q) o ID m v
^**^ m r*» ^zC c ^^CM ^I*""^'^!CM
m m I-H 0^*0 OJ*^ CM oi«-* oj 01^--
CM CMCM CMCMm-v. oi'^.O'1';-
mmcMO) mocMOi

o
CM

«v r- i . l ill i i i i

c *••*
rO
3



oj 01 o)
Ol O) Dl
B -O-~ C — C
in <-i i i i i i i i
Sfi o ** oo
~. . .
• •-* o
0
^^
+j j:  Ul
c a
^^ CM^gi-I^H-H'^ •-! CM

JC ,«. T) U
•o -o i- -o o t.
C O O 10 T3 l— 
ISO IO-OID ID l— U L, N U»U«
•^43 ^E3C 4JOCOQ) *^"€ ID 0*00 IOID
+J Ul Q. ID lCt.CC •Q
•OO •»••»- 0 KOBE r- >» — £ •»" £ .
a. .. U' • • • •
S a.































Ul
c
o
+J
IO
"3 in

13 'o.
U E

u in u
<£ £
O
2 T3
L. tO
10 ® N
N ^ >»
E •—
ID 3 fl
.0 c c
o in
•o in c
• E in © >s
<0 3 ID 3 i—
•P Ul J= Ol C
IO Ul 4-> O O
•O tf E r-
ii - " means no
*Not Detected.
Number In pare
^Penta-CDD homo
°TCDD/TCDF scan
3-24

-------
that  some of  the variability  could  potentially  be  attributed to  fuel
contamination, although feed samples were not analyzed for PCDD content.
     Ash  samples  from the  chimneys of two  fireplaces  were  analyzed  for
       C 0
PCDD's.    One fireplace  was 12 years old  and  one was 25 years old.   The
25-year-old fireplace  had total PCDD concentrations of 44.7  ng/g  including
1 ng/g of  2378-TCDD.   Ash samples from the  12-year-old fireplace  contained
1.79 ng/g  PCDD.   No TCDD isomers were detected  at  a detection limit  of
0.04 ng/g.  The penta-CDD homo!ogue was not  analyzed  for  in either of these
samples.
     Ash  samples  scraped  from the  flue  pipe  of a  residential  heater
combusting both oil  and wood were analyzed for PCDD's.  After burning  only
oil, the PCDD level in the  ash  was  0.280 ng/g.  By comparison, after  burning
only wood, the PCDD level was 0.97  ng/g.  After co^firing wood and  oil,  21.7
ng/g PCDD  were detected,  including 0.8 ng/g of the  2378-TCDD isomer.  The
penta-CDD homo!ogue was not analyzed for in any of these samples.
3.2.4.2  Treated Wood Combustion.  Chlorophenols are produced for use as wood
preservatives, slimacides,  bactericides,  and as  starting material  for the
chlorinated phenoxy  acids 2,4-D and 2,4,5-T.   Chlorophenols  may  either be
contaminated  with PCDD's and  PCDF's,  or  PCDD's  can  be  formed by the
dimerization  of  chlorophenates during  pyrolysis.  The following  section
discusses the results  of  several  studies where  Chlorophenols were combusted
with wood or wood products.
     Two studies concerned  the  combustion of pentachlorophenol  (PCP)-treated
military ammunition boxes.220'233  At  the  Los Alamos  National Laboratory in
Los Alamos, New Mexico, PCP-treated wood was  incinerated  under a variety of
test conditions in  a controlled  air  incinerator.220   The incinerator  had
modulated  burners,  steam  injection capability,  and  enhanced mixing  of
secondary air with the primary chamber effluent.  .Ash samples were taken from
the hot zone between the  primary  and secondary  combustion chambers.   Neither
TCDD's nor TCDF's  were detected at a detection limit of 17 ng/g.
     At the Tooele Army Depot  in  Tooele, Utah,  PCP-treated  ammunition boxes
and explosive-contaminated wastes were incinerated.233  The  incinerator was
designed to decontaminate metal parts containing  explosive  residue.  The
                                    3-25

-------
incinerator  has  an unfired  afterburner  (refractory  lined  duct) with  a
combustion residence time  of 0.3 seconds.  Four tests were performed while
the incinerator was firing:   1)  no  waste  fuels,  2)  wood  freshly coated with
PCP, 3)  40  percent by weight PCP-treated wood  and  60  percent by weight
contaminated waste  (including wood,  cloth,  metal,  and rubber).  Results of
the analysis of stack  emissions  for two baseline tests showed  average  PCDD
                      3                                          3
emissions of 5.0  ng/m   and  average PCDF emissions  of 9.82  ng/m  .   The
analysis of stack emissions for three tests conducted while the 40/60 mix was
fired showed average PCDD  emissions  of  125  ng/m  and average  PCDF  emissions
             3
of 14.2  ng/m .  Analysis of stack emissions for three tests  while  freshly
coated wood  was  fired showed  average  PCDD emissions of  8,215 ng/m  and
average  PCDF emissions  of 426 ng/m .   When  only ammunition  boxes  were
incinerated, afterburner samples were taken.  Analysis of afterburner samples
                                   3                               3
showed PCDD emissions of 1,420 ng/m  and PCDF emissions of 587 ng/m .
     A pilot scale  incinerator was  used to burn wood chips which  had been
mixed with  technical  grade  tri-  and tetrachlorophenate.   At combustion
temperatures of 500 to 800°C (932 to 1,472°F), the formation of PCDD's was
demonstrated.  At  higher temperatures,  the formation of  PCDD's decreased.
When wood chips and trichlorophenate were burned,  stack emissions of total
PCDD's were  111,540 ng/g feed.  When tetrachlorophenate was burned with wood
chips, stack emissions contained 350,200 ng/g feed.  Addition of copper salts
to the  tetrachlorophenate formulation  and  increasing the  residence time
within the incinerator reduced the emission of PCDD's.
     In  another study, fly ash samples  from a fluidized bed  system  burning
PCP-treated  wood,   painted wood,  and   hypochlorite-treated   paper  were
         170
analyzed.     Total  PCDD's and  PCDF's  detected in fly ash  samples  after
burning  painted  wood were 177 ng/g  and 217  ng/g,  respectively.    When
PCP-treated wood was burned,  PCDD levels in the fly  ash were 324  ng/g and
PCDF levels  were 241 ng/g.  When the hypochlorite-treated  paper was burned,
large amounts  of chlorine were  present  but  PCDD and PCDF  levels  were
relatively low with 24 ng/g of PCDD detected and 12 ng/g PCDF detected.  The
addition of  pentachlorophenol  to  these  fuels  did not increase  PCDD or  PCDF
emissions.
                                    3-26

-------
     In  a  pilot scale study, two  chlorophenate  formulations,  Servarex and
Kymmene KY-5, were  sprayed  over  wood wool  and birch leaves and combusted in
              190
an open  fire.      These formulations  are  mixtures of 2,4,6 tri-,  2,3,4,6
tetra- and pentachlorophenate as  sodium salts.   PCDD's  and PCDF's  were
detected in  these two formulations  at concentrations of 20 and  150 ppm,
respectively.   When Servarex and  KY-5 were each  burned  separately,  high
levels of  PCDD's  were  formed. When burned  alone, the Servarex formed 21,600
ng/g of PCDD and  the KY-5 formed 11,600 ng/g of  PCDD.  Each of these was then
sprayed over birch  leaves and wood wool  and combusted in  an open fire.  One
gram of  chlorophenate  was dissolved in 20 ml of water  and  sprayed over 30
grams of birch  leaves  or wood wool.   Smoke  gases  were  trapped  in charcoal
filters and  analyzed.  When birch  leaves sprayed with Servarex were burned,
213,300 ng/g feed of  PCDD's were formed.  When wood wool  and Servarex  were
burned, 392,000 ng/g feed of PCDD's  were formed.   When birch leaves  and KY-5
were  burned,  205,000  ng/g feed  of  PCDD's were  formed.   Purified
chlorophenates  were  also  burned  with  birch   leaves.   When  2,4,6
trichlorophenate and pentachlorophenate were burned with birch leaves,  levels
of  PCDD's  formed were  1,115,000  ng/g  feed and  957,200  ng/g  feed,
respectively.
3.2.5  Boilers Co-firing Wastes
     Table 3-7  represents the emissions data for  boilers co-firing wastes.
EPA's Hazardous Waste Engineering  Research Laboratory (HWERL)  (formerly
Industrial  Environmental Research  Laboratory -  Cincinnati  (IERL)) conducted
studies on  industrial  boilers co-firing waste  products.41  Four boilers
co-firing chlorinated wastes such  as creosote sludge,  chlorinated solvents,
and waste oil were  tested.   Stack  emissions from three of the four  boilers
were tested  for PCDD's  at  a detection  limit  of  1,000 ng/m3 but none were
detected.  The  fourth boiler was  a steam generator  firing waste wood
contaminated with pentachlorophenol.   Stack emissions of 2378-TCDD from this
boiler ranged from  <0.4  to  <1.5 ng/m3.  Total PCDD  stack  emissions  ranged
from 74.6 to 76.4 ng/m3 and averaged 75.5 ng/m3.
     A second study for EPA's HWERL tested waste fuels and stack gas emission
samples from five industrial boiler test  sites  co-firing hazardous waste
                                    3-27

-------























CO
1 1 1

co
^^
3
C3 1
I— i
cc

u.

0
u
CO
cc
UJ
•—1
>— 4
CO
CO

..
£>•
dljl
§
III
1 1 III
o

UJ
o
a;
§
CO

*
r--
i
CO
UJ
CD
JS


.












U)
o
g
E
«£

e
ia
cr





e
n
1











a.



o
01
e
(O
cr






i








Q
g
T"
CO

CM



O)
c






i
£


C9
1
CO

U)
•P *"^
T- e
e -o o
ID a> *^
•p -P
<4- UJ (O
® i® y
• -1
o *^
z


°f

E S ^ 0)
•v. v. 0) >x
TO— O) C O)
i c - o r^ o
• ^ r^ o
1-1 i-i  n

•o
• E
m « v.
E r- i o>
•x. me
110) IE 0
C 1 •*. 0 Q I t I
a \o v> *r o
z ^ . c ^H o
•-i r- 1-4


•o

m E E E
•x. P) O) CO X. «s.
O) C ^^ C ^m* C ^^ O) C31
efQCN* mace o
m in • m •— i o
r^ » in -i o
o r*** p1* o^ ^o




mE
1 O)
II 1C
• in i ii i i
V i-l
v
u
m
t- "E
C —4 O)
a CM i
Z g CO 1 1 1 1 . 1
• •
o
V

OC3 C3C3 C3O<< C3
U.U. U_ U_ U- U. U- LL. U_


T3
C
<
C r—
O C
O
O Ul
t- 
^\ *v ^*
r- E 0 —
c i- a o
0 r- 0 A
C C 4J
•^ O t- e
•^ O 01
"S ** ca u
•P cB o>«-
0 -P C *;
4-> TJ «- -5
 t-
oo o i- o
^^ ^1 (0 ^^
m > o
O» i-l 
ex
i
in

•§
>2

^
•»-
•p -a

fm. ^J ^J
U «
U) (D -O
io IB in

coo



-------
      40
fuels.     Among the  wastes being  fired were  creosote  sludge,  carbon
tetrachloride,  chlorobenzene,  methanol,  toluene,  and trichloroethylene.   A
watertube  boiler co-firing  wastes  and No.  6 oil was the  only  boiler  having
detectable levels of 2378-TCDD  in flue gas  emissions.   However, the measured
value for  the emission of this  isomer was equal to the  0.002 ng/m3 detection
limit.  This boiler also had the highest total  CDF emissions of 5.5 ng/m3  in
one of  two samples.   A creosote  sludge/wood-fired  stoker had the highest
total PCDD stack emissions  (75  ng/m3) but  the creosote  sludge co-fired with
wood waste in  this  boiler was  found  to  contain 7,400 ng/g of total  PCDD.
PCDD and PCDF homologues were not detected  in any other chlorinated waste  at
detection  limits ranging  from 0.045  to 4.6  ng/g.   Stack concentrations of
PCDD from  the  other  four boilers ranged from  less  than detectable to  1.1
ng/m  at detection limits ranging from 0.0022 ng/m3 to 0.019 ng/m3.
     In  another  study,   Buser,  Bosshardt,  and  Rappe33  report  the
identification  of 600 ng/g  and  300  ng/g  of PCDD's  and PCDF's,  respectively,
in the  fly ash of an  "industrial  heating  facility."   This facility  was
generating steam by co-firing  used  industrial  oils.   PCDD's and PCDF's were
detected in other fly ash  samples  as  well  but the samples with the highest
concentrations were the only ones reported.
     The EPA tested six  commercial  boilers  firing spiked waste oil.81  The
boilers were  in the size  range of 0.4 to  25 million  Btu/hr  heat  input
capacity.  The  fuel was  used automobile oil spiked with organic  compounds
such as chloroform,  trichlorobenzene,  chlorotoluene,  and trichloroethylene at
levels ranging from 1,500 ppm to 10,000 ppm.
     Of the  six boilers,  only  one had  detectable  levels  of  OCDD with
4,500 ng/m  and 17,000 ng/m  detected in one of three samples  of  stack gas.
Detection limits were not specified for these samples.   Only one  of  the six
boilers had detectable levels of TCDF with  170  ng/m3  detected  in  stack gas.
Detection  limits were not  specified  for the other  samples  with less  than
detectable levels.   The feed samples  of waste oil basestock and the "spiked"
waste oil  were  tested  and  no PCDD's  or  PCDF's  were  detected at detection
limits ranging from 0.04 ng/g to 2.0  ng/g.   Fly ash samples collected  from a
Scotch firetube boiler  did not have  detectable levels  of  TCDD,  but  the
                                    3-29

-------
concentration of penta through  octa  homologues  ranged  from not detectable to
230 ng/g.  For  three  fly ash samples total  PCDD's were 911 ng/g.  Detection
limits ranged from 0.5 to  10 ng/g.   Concentrations  of  PCDF homologues ranged
from not detectable to 1,000  ng/g for a total of 3,777 ng/g in three  samples.
Detection limits ranged  from  0.5 to  10 ng/g.
     A 233 MW utility boiler  was tested while firing No. 6 oil and PCB-spiked
waste oil.    The waste  oil  comprised 10  percent of the total fuel.  PCDD's
and PCDF's were not detected in  stack gas emissions  at  detection limits
ranging from 0.031 to 0.10 ug/m .
3.2.6  Hazardous Waste Incinerators
     Table 3-8  presents  the  emissions data for land-based  incinerators  and
incinerator ships.
     3.2.6.1  Land-based Incinerators.  Eleven  incinerators firing  hazardous
wastes were the focus of ten studies.  Among  the types of units tested were
rotary  kilns,   with  and  without  afterburners,  a  mobile  rotary  kiln
incinerator,  and a  tar  burner.  Wastes being fired typically consisted of
chlorine-containing liquid organic wastes, herbicides, and  wastes  containing
PCB's.
     An incinerator  was  tested while firing feed  containing 3,000  ug/g
      1 ^fi
PCB's.     Cyclone outlet  samples were analyzed by  selective  ion monitoring
gas chromatograph/mass spectroscopy  (GC-MS).    PCDD's  and  PCDF's were not
detected at detection limits  ranging from 0.03 to 0.06 ug/m3.   Because PCDD's
and PCDF's were not detected in  the cyclone samples,  analyses of stack
samples were not performed.
     One study  tested two  rotary  kilns with  afterburners.234  Three  tests
each were performed  at  incinerator facilities  at El Dorado,  Arkansas,  and
Deer Park, Texas.   Test results were reported in terms of total quantities
present in the analyzed sample because leakage and loss of unknown quantities
of most samples occurred during shipment preventing the calculation of actual
concentrations.  The  facilities are  operated by Energy  Systems Company
(ENSCO) and Rollins Environmental  Services, respectively.  Wastes during  the
first test at  each facility included hydrocarbon  wastes,  paint and ink
manufacturing wastes, pesticide process  wastes,  and  vinyl chloride  still
bottoms.  The Rollins facility had total  TCDD and TCDF levels of 6.94 ng  and
                                    3-30

-------











to
en
°
<
Ul
z
o

}—
to
*
13
o
a
S

a

a.
E
to
Ul
— "c
C "O o
t- Ul » nj
1- U.


jOj»jjrvoco fl^-oocsi
" 0) "*E
c c ui •? i
CS . O)
( CM  -5*
CO)!- CO)
^ C^4^— (Vf-> C — .
z *2 2 ^i * ** °° ** ' i
••1 • = <-i in in
. - CM o a, oo
0V


O) o> fr» oi
•» oo i- o> i ' i*
i o» »r i . c i i o
"3,0 in ^ §
|| ^ ^ *
' ' ^- r*»
i § s
•o
- "E me
en en "^ "^ ^* - O) •*- O)
A C(_)C.-»4"' C c ^ c ^
CSJ ** <-» S *° m "^ ^ *"
^Oujojcninosco



0)
1 0)
' 1 1 1 1 1 1 10
Zl
O)
O)
*^
0)
1 1 1 0 1 1 t 1 0
•~> oo
1—4
(D
C
o ^ j- 4!

^.u. u. u. u. 2 < < «
u n- -

t/) ^^
-H „;= -1 rtg »
o 5 "" "1 " ^9
£ ^
^
QJ IK (B
•at ai >- c c c '•^ OCP
IP! ! ! I I























f^L
•o

-------























5

3
c
1 »
1
*

cc
2
^c
g-

<>""i
y

IU
I—
VI
CO



fg
«c
*»
2:
•*

Q£

£5
}*;
5

1

«n
CO
i
C1
Ul

>-





















in

1*-
0)
cr










-p
c
ID



°
a
r-
§
10
CO
1


•o
10

I

4J
10
u
+J

CJ
c
c
_0

in
Ul
e
a

£3

Q.

C
g
°-











U.
g
a.







o

c
IO
,

O -r-
r—

c

II I I I m

I

m
m o
m

Ol
i = I I

*
o


CM TT
r>-  o o
i "o> *"* "01 '
c i c
to
in fn
CM «— 1



j. _,.
O) O>
O5 O> ^ .Q
c -^ c -- a o a Q
3 2 z z z z

O to
^r CM







II 1 1 1 1





Q *~ Q — i i i a
z 3 z £ z




L. L. CS C3 O CO
IO IO U. U. U. U.
a. a.




r-4 O •-< O CM 1-4
a a



Ql

r- -C
IB CO
c +J c e
L. r— C i— r- U
ID **— ID **" *— " O

3 >><—'>,>,<.
CQ C. .C Q.r- L. ^ a>
10 +J a. ID 10 ie c
a o * in u- o o TJ
i— cr a: ce c
h-4

• O
e —
o t-
•^ ^
IO ID
i— a.
3 VI
U 1
r— C •
« o >>
o c •—
c
CO O
O -H
•r- Ul
+•>  S c
C -r- O
a> .c . •
> O ID
a> •o • a 3
U (0 -0 O 0)
Q.J: o i o
J= IO p—
ffl IA +> +J O

^ Wl B ® O
10 +J >— ^
in , i »- O) -C i-

Ul
ID
3

o

t. u> *o
ID C g
js o 5
3 *M
C 10 Q
in "3 o >•
i- O O r—
r- C
in 10 • o
•,- o Q
in a ID
eg c o 3
.C -^ 1 <3I
•M 10 O
C "O +» r—
ID ID O. O
1. Ul ID S
a 3 .c 5
a. j=
in -
_c ffl Q u.
(D rt3 *O • fl
^ U} V >»4J
§G i— C
U- 0) C 0)
,
IO 3 IO t—
+J « r- C
co in a o
•o < 0
(— C
O • 10 10
c -o o o
ID in
in +J c
c u — u-
10 a> a

•o a> ^
+J N S
i o a) o
Z J= I-
S IO J2
3-32

-------
13.5 ng, respectively.  The second test  consisted  of  the same wastes as the
first test with  the addition of liquid  PCB  wastes.   The Rollins  facility
during the second test had total TCDD  and TCDF  levels of 1.42 ng and 22 ng,
respectively.  The ENSCO facility during the  second test had total TCDD and
TCDF levels of 0.48 ng and  6  ng,  respectively.   For the third test,  liquid
PCB wastes were  fired with  clean  fuel  oils.   The Rollins facility had  less
than detectable TCDD's at detection limits ranging from  0.48  to 0.9  ng  and  2
ng for TCDF.   The ENSCO facility had less than detectable levels of  TCDD and
TCDF at detection limits of"0.2 to 0.45 ng for TCDD's  and from 0.08 to 0.5 ng
for TCDF's.
     A rotary  kiln  operating at 1,200°C  was tested  while  burning Silvex
herbicide.246  The 2378-TCDD isomer was not detected  at  a detection  limit of
1 ppb  (by volume).   The  penta-CDD,  hepta-CDD,   and OCDD homologues  were
detected at a total  of 75 ng/MM5 train.  The  penta-CDF homologue was  present
at a concentration of 190 ng/MM5 train.
     In an unrelated study, stack emissions  from a rotary kiln operating at
                          Four  tests were  conducted while PCB's were being
1,200°C were analyzed.88
incinerated.  The average concentrations of  PCDD's  and  PCDF's  were 20.3 and
11.2 ng/m  ,  respectively.   However, TCDD's  and OCDD's  were the  primary
homologues detected.   Detection  limits for  the  other  homologues were  not
reported.
     The incinerator  exhaust  of the rotary  kiln waste  incinerator at  Dow
Chemical was tested.153*  This  incinerator destroys 20  tons/day of liquid
waste in  the 1,025°C  afterburner  and  185 tons/day  of solid  and liquid
combustible trash including 1.5  tons/day  of  chlorophenolic wastes from  the
2,4-dichlorophenol and 2,4-D processes.  The  average concentrations of PCDD's
and PCDF's  detected  in three test  runs  were 36.6  ng/m3  and 93.8  ng/m3,
respectively.
     At another facility, used  transformer oil  (supposedly containing  less
than 50 ppm PCB's) is  fired in  an  incinerator.74  A spot check on the  used
oil detected one  sample with 90  ppm PCB's.   The  incinerator,  which has
secondary combustion chambers and  an afterburner,  burns off the  insulation
from the  aluminum or copper  windings  from dismantled  transformers.   One
                                    3-33

-------
composite  ash  sample was analyzed and found to contain  538  ng/g PCDD's and
2,853 ng/g PCDF's.
     A mobile  incinerator was tested  while  firing CDD-contaminated liquid
still bottoms  and  soil  during one test and CDD-contaminated lagoon sediment
(containing  1-21 ppb 2378-TCDD)  during a  second test.108  The  only homologue
detected was OCDD  at a  total  of  91.3 ng/g  in three samples.  These detectable
levels were  suspected to be  from contaminated  solvent used in the analyses.
It was unlikely the  OCDD was  formed during the incineration process.
     Dow Chemical  tested an  industrial  solid waste incinerator (rotary kiln)
                 CO
and a tar burner.     The tar  burner was a 72 million Btu/hr unit with  natural
gas burned as  a supplemental  fuel.  Four tests were  conducted  while the unit
was firing natural gas  and tars.  The  2378-TCDD isomer and other TCDD's were
not detected in particulate matter samples.  Detection limits  ranged from 1.3
to 3.0 ng/g  for the  2378-TCDD isomer  and from 0.7 to 1.2 ng/g  for other
TCDD's.  Total concentrations of the hexa-CDD, hepta-CDD, and  OCDD homologues
ranged from 256 to 572  ng/g  for  the four  tests with  an  average of 406  ng/g.
The penta-CDD  homologue  was not analyzed for.
     The rotary kiln  incinerator Dow Chemical tested was a 70  million  Btu/hr
unit.  This  unit  is capable of incinerating  both  solids and  liquids.
Supplemental fuel  is also  burned in this  unit in  the rotary  kiln and the
secondary  combustion chamber to  maintain  combustion temperatures.  Three
tests were  performed while the  kiln  was  burning  tars,  solid waste,  and
natural gas,  but  without  supplemental  fuel  in  the  secondary combustion
chamber.   Particulate matter from the  first test  was analyzed  for PCDD's
using a  nonspecific  GC-MS packed column  method  and  very high levels  of
2378-TCDD were detected.  In the other two tests,  a capillary  column specific
for 2378-TCDD  was  used,  so the results of the  first  test are not comparable
with the second and  third test.  During the first test,  an average of  5500
ng/g of  2378-TCDD  was detected  in particulate matter.   The average total
concentration  of other TCDD's and the hexa-CDD, hepta-CDD and OCDD homologues
in the first test  was 847,400 ng/g.  The  total concentration of PCDD's (not
including penta-CDD)  for the  other  two tests were 9,710 and 113,600  ng/g.
The 2378-TCDD  isomer was detected in  the  second  test at a concentration of
                                    3-34

-------
 110 ng/g.  This  isomer was not detected in the third test but the detection
 limit was 260 ng/g..
     Five tests were then  conducted on  this rotary kiln while oil and natural
 gas, and tars and  natural gas  were  fired as  supplemental  fuel in  the
 secondary combustion chamber.  The 2378-TCDD isomer and other. TCDD's were not
 detected in particulate matter from any of the  five  tests.   Detection  limits
 ranged from 2 to 5 ng/g for the  2378-TCDD  isomer and from 2 to 8 ng/g for the
 other TCDD  isomers.   Total concentrations of the  hexa-CDD,  hepta-CDD,   and
 OCDD homologues  ranged from 13 to 1,064 ng/g with an  average  for the five
 tests of 267 ng/g.  The penta-CDD homologue was not analyzed for.
     Results from three  tested rotary kilns were reported in one  study.151
 Only one of the kilns  had  detectable TCDF  emissions at a concentration of 0.7
 ng/m .   However, TCDF's were detected  in the  fuel,  which was liquid organic
 waste containing 0.4 to 1  percent chlorine.  The other two kilns were firing
 liquid organic solvents with chloride  concentrations  ranging from 0.2 to 16
 percent chlorine.  No TCDD's or TCDF's were detected in the flue gas or  feed.
 Detection limits were unavailable.
     3.2.6.2  Incinerator  Ships.  Two  studies  were conducted with  the  M/T
 Vulcanus incinerator  ship.1'2   The first  study was conducted  during  the
 incineration  of Herbicide Orange  contaminated with  2378-TCDD.1   The
 incinerators were heated up to a flame temperature of 1280°C with fuel  oil
 before the burner was changed over to the Herbicide Orange feed.  TCDD levels
 in the feed ranged from less than detectable to 2,800 ng/ml, with an average
 concentration of 1,820 ng/ml   injected.  The detection  limit for the feed
 samples  was  20  ng/ml  injected.   No TCDD's  were  detected in  the stack
 emissions.   Detection  limits for the  TCDD's were very variable  due  to  the
 complexity of  the  samples.  Detection  limits  ranged  from  0.0009  ng/ml
 injected to 0.086 ng/ml injected into the GC-MS for analysis.
     The second study  was  conducted during a PCB burn.2  TCDD's, including
the 2378-TCDD isomer, were not detected in the  feed or in stack emissions.
Detection limits ranged  from  2 to 22  ng/g.   TCDF's were detected  in  all
samples of waste and in several samples  of stack gas.  The analytical method
could not distinguish  the  2378-TCDF isomer from the  other 37 TCDF isomers.
                                    3-35

-------
Total concentrations of  TCDF's  in  stack gas samples were reported to  range
from <0.3 to <3 ng/m .
3.2.7  Lime/Cement Kilns
     Table 3-9 presents emissions data for lime/cement kilns.
     Four studies have addressed PCDD and PCDF emissions from  lime or  cement-
kilns co-firing wastes.    '''      The combustion temperature of  this
process is about 1500°C with a typical residence time of 1.5 seconds.
     A cement  kiln  at San Juan  Cement  was tested  while  co-firing  liquid
                                                                         184
organic wastes containing  from  6.5  to 35.5  percent chlorine (by weight).
Flue gas  and  particulate samples were taken.  One of the four SASS train
samples had detectable levels of hexa-CDF and hepta-CDF.  The  concentrations
                                      3               3
of these two homologues were  1.35  ng/m   and 0.74 ng/m ,  respectively.   None
of the other homologues were detected at detection limits ranging from 1.6 to
4.9 ng/m .  Similarly, one of the  EPA Method  5  filters used for particulate
analysis contained  11.0  ug/m  of penta-CDF,  25.7  ng/m3  hexa-CDF, and  8.1
    3
ng/m  hepta-CDF.  None of the other particulate samples had detectable PCDF's
at detection  limits ranging  from  5  to  15 ng/filter.  These  detectable
emissions occurred  when  the  kiln  was fed  waste  containing 21.4 percent
chlorine which corresponds to a chlorine  input of  3.5 percent  of total fuel
input.  This resulted  in  a potentially  kiln-damaging condition.  The  study
maintains the detectable  emission  occurred  only during "upset" conditions.
Under other conditions PCDF's were  not emitted,  and PCDD's  were not  emitted
under any  condition including  the  "upset"  conditions.  Detection  limits
ranged from 1.6  to  4.9 ng/m  for  stack -gas emissions, and  from  5 to  15
ng/filter for the particulate analysis.
     A wet-process  cement kiln at  General  Portland  Cement,  Inc.,  and  a
dry-process cement  kiln  at Lone Star Cement  were  tested.20a'20b  Both  of
these facilities  were  co-firing hydrocarbon  solvents,  chlorine-containing
wastes, and wastes spiked with Freon  113.  No PCDD's  or PCDF's were detected
in stack emissions.at the detection limit of 1 ng/ul injected (into the GC-MS
for analysis).
     A lime kiln at Rockwell  Lime Company was tested while  firing petroleum
coke and waste  fuel  consisting  of  lacquer  thinner solvents,  alcohols,  and
                                    3-36

-------






















CO
z
I-H
NX*

I-
z
UJ
1
UJ
jr
H- *
_J

••
^—
g
CD
UJ
<£
CJ

UJ
ty
o
CO


0%
I
on
UJ
CD

1—








































































































in
CD
c
CD
£_
CD
CD
•P
C
CD
-P
C
O
0

CD

"o
E
0}
CO
-C
in
«c
-Q
C
10

c
o
?
•p
c
CD
O
C
O
O
c
O
in
in

g
UJ
u_
§
X^
o
8
o





u.
Q
O
CL





CD
O)
e
(O
ce





c

CD
s.








g
CL





CD

C

M- Ul fl
O CD O
1— 0

6 ~




>^
«*- 4->
O T-
^»

c <-» c
<— < Q .. Q O O
O» »" CO Z Z Z Z
O
CM ^t




III 1 II





 i— i.
*^* C I'* ^y f~ (O C r"*
JtfiO^. J^i— C Jtf-P'- i— r—
3 -p ram cO-pf-CD
^^ PO C ^) ^ c"" ^P C v* 2 ^"*
C CD CCD-P CCDCD ^. (0
0 CO O Q) CD CL. cU 	 1 O •*•• rff '
°" °" °" J ^
































































. '

-------
              KQa
paint wastes.      The  wastes  contained approximately 3 percent chlorine  (by
volume).  PCDD's were  not detected  in  baghouse  dust  or EPA Method 5 filters.
For baghouse  dust  samples,  detection  limits ranged from 0.005 to 0.25 ng/g.
For  EPA Method 5  filters,  detection  limits were  converted to stack  gas
concentration  and  ranged from 0.034 to 2.0  ng/m  .
3.2.8   Hospital Incinerators
     Table 3-10 presents emissions data from hospital  incinerators.
     High temperature  incineration  is  the preferred method for disposal  of
hospital wastes containing  infectious  or  hazardous materials.   Most hospital
incinerators  of  older design  are incapable of destroying all  hazardous
materials and  have inefficient combustion leading to  emission  of hazardous
air pollutants.  Hospital wastes  are  also highly variable in content.  They
usually contain 20 percent  plastics, compared to municipal  solid  waste which
contains  3  to 7 percent  plastics.   Combustion  of  plastics  composed of
polyvinyl chloride and other halogenated polymers and copolymers can be a
major generator of toxic air emissions.
     A  1983 stack  test on  a Canadian hospital  incinerator  found  PCDD's  and
PCDF's  to be  emitted  at   average  levels  of  69 ng/m3  and 156  ng/m3,
respectively.  a   The  test  was performed on  a  high combustion  efficiency
controlled-air, two-chamber incinerator.  Small  amounts  of PCDD's and  PCDF's
were detected  in the bottom ash, with much  higher levels in the fly ash.
                 62a
     Doyle et  a]..    reported results from  three hospital incinerators in the
United  States.   Stack test filter  samples  were analyzed  and  had average
levels  of PCDD's  and  PCDF's  of  15 ng/m3  and  25 ng/m3,  respectively.
According to Doyle et al_.,   these levels probably represent less than one-half
the actual emissions because more than 50 percent of PCDD's and PCDF's can be
found in  the  vapor phase,  which  was  not  analyzed  in  these  particulate
screening tests.
3.2.9  Wire Reclamation Incinerators
     Table 3-11 presents emissions data from one study conducted on a wire
reclamation incinerator.103  Wire insulation incinerated during this process
often contains PCB's and polyvinyl chloride.  Analyses of  inorganics  in the
stack and furnace  samples from the  three  furnaces revealed high  levels  of
                                    3-38

-------





















trt
1
£
LU

hH


t— 1

1
5
I-
s
o
X
£
Q£
o
C3

t
O

LU

1
co

O
i— 1
1
tn
LU
1-


















U)
O
u
c
2
•2
CD
cr





•p
q
1
CD

C
E

c/:
V
*
•o
c
1C

e
c

+;
R>
L.
a
c
c
c
O
c
C

in
if,
"e
LL

u.
c
|c
c
c










1,
Un
°"





S,
e

•M *~
t- C
c -o o
3 CD •»-
"S S g
0 d


0^"
a> ^
0. U
>, <0
1- U.




|Q lO
•-< CN
PO V3



i— I
m i
1 E

r-- C3)
•-I C

"E ^S
O>^» O)*1^
c f-i •em


tn c


en PO
g: g
>x ^.
O>**^ O)^^
c «-i cm
CT> m
vo I-H







g





i



a o
u. u.






i™4 m



1
•o •
>a co
c

U =
































































IO
o
c
in
10
i
cs
S

1
r
3-39

-------








fn
Q*
p
§

t-t
1


g
HH
J-

*E
3f
d
K
LU
rr*
H-t
3:

«•
>;
Q
CD
r~
^C
O

P
Q£
*™>
rS
CO

,4
r-t
I
ai

S
j-




















U>
0)
ci
(D
t-

s:


a>
Q.
s
ia
CO
in

t- "c
c -a o
=> a> t-
• j ^
M- in (U
oou
a' d



^^
<»- 4J
O •*•
a> -^
ex o
H"l2



rn PO
0 0
i-H <-<




(O  C3)
O) O
c c

• r*-.
i-H C3





1 1



O) O
"o> *o>
c e

~H ao
^r in

<3 0




i t







I i


~* "3
 CO
=

e c-
0 0
•^- +>
-P ia
S m
10 C
 c
3: Qi HH




















































_>>
c
O

in
m c
•p 
-------
copper  and  lead as well as  85,500  ppm of chloride  in  one  of the furnace
samples.
     Total TCDD and total TCDF concentrations in stack  fly ash scrapings were
0.41 ng/g and  11.6  ng/g,  respectively.   Bottom  ash samples from the furnace
contained 0.058 ng/g total TCDD's and 0.730  ng/g total  TCDF's.   The  analyses
did not distinguish the 2378-TCDD or 2378-TCDF isomers; only total TCDD's and
TCDF's were measured.
3.2.10  PCB Fires
     Table  3-12  presents emissions  data concerning  several  studies.   In
September 1978, 18  capacitors containing PCB's  were  burned  in  a fire  at a
transformer station near Stockholm, Sweden.     Several types of sampl'es were
taken.  Liquid from inside an exploded capacitor contained 75,000 ng/g PCDF.
     In Binghamton, New York,  in  1981,  an electrical transformer containing
about 1,100 gallons of PCB's was involved  in  an explosion.60  Total  PCDF
homologues in soot were initially found to be as high as 2,160,000 ng/g.  The
2378-TCDF isomer  accounted  for 12,000 ng/g  of total  PCDF's.   The hexa-CDF
homologue alone accounted  for 965,000 ng/g  of total  PCDF's.   Total  PCDD's
were found at a concentration of 20,000 ng/g including 600 ng/g 2378-TCDD.
     In January 1982,  an electrical fire  involving PCB's broke out  in a
Boston, Massachusetts, office building.60  One bulk soot  sample contained a
total of  115,000  ng/g  PCDF's including 60,000 ng/g TCDF.   No PCDD's  were
detected at a detection limit of 100 ng/g.
     In March  1982,  a fire  broke  out  in a  capacitor battery in a metal
treatment factory in Skovde, Sweden.192  The capacitors contained mineral  oil
and PCB's.  Wipe  tests were taken from several  locations  and results  were
reported in terms  of  unit  area.   Samples taken  from  the  floor,  0.5  meters
from the capacitor, had the highest levels of PCDF's with 100 ng/m2 2378-TCDF
and 772 ng/m2 PCDF's.
     In Miami, Florida, during April 1982, a fire and explosion occurred when
an underground transformer vault exploded releasing approximately 100 gallons
of PCB transformer oil onto  the floor."  Smoke ejector fans  were  set  up  to
ventilate the vault.   Samples of  soot  and other  residue from the fire were
collected.  Wipe samples were also taken  from surfaces  near  the fire scene.
                                    3-41

-------


















CO
i
u.



>-

§

1

LU
gg
g
u,
CM

1
UJ
€
}M























in
o

1

0

u-.
•g
<

C
-


.^
"Si

ll

t-u.


OO OCM OV  •-*
,-1 ^H I-H CM

CM
€
O)
I e
II II I 00 I 1
5 °^
• °> — ^
*s» cn o
O) O) ^ O)
-s. c cn CM •v.
cn c CM cn s cn cn
§O o cn c c o c
• o c o
C3 C3 *^ C3 €3*0 ^h ^T
* *O U) CM •— I OO •— • » •
\f\ r*i r-4 f*» P^ ^ CM 00 •*•<

CM

II II II 1 1


cn

cn cn cn
c *** -^
cn cn
o -a e. c
1 § 1 i ,
. CM
Cfi cn 10


II II II 1 1



cn cn
t v^ »^
cn cn
c c
10 II II CTV 1
o in
\o


•o in in
t- +J 4-> Q +J 0) 4^ -C
3 O O Q. O Q. O VI
^co cos cox «n ffl
+J
» (O
o +->
. . U CO
c < u in
BO Z« _iio i- c
•— 4-> (DCU.ECU O
o E .-am E i> c «c -(->
-c >a>- c > -o . <3t? >ao cn
•-ue i-t.cz —to f-ioo >-4-^ ^-i j= a I-HU i-ic
y cn 4-> ^: x E ra s u. ^
O c tncoconuco J=
4J — o — 3 c in
CO CQ ED Z CO "O «

u u L. t- t-
c.§ i i- i i- i i
o >» u L. o >, c o >i u E
+J I- O O +-> I- O +J 1- O O ^

~U4J I/If— U)r— *U4-> 10 I— Tj 4-> Ulr- U) CD
IO+> C 3 C 3 « 4-> C 3 R>4-> C 3 Ct.

O03!— ^ H-^O00 1— ^ O 1— * I—













•>»
i
in
ID
01
o

i
Q


o
•a
c
<0

1
in
f^
o
J=



>l
1
e
(O
•o u
o in

u u.
a> o
a> I—

z§
.00





IO
+J in

-------
Four bulk  samples  of soot and other residue, and three  hexane  wipe  samples
were analyzed.  No PCDD's were detected in these samples at a detection limit
of 10 ng/g.   PCDF's  from  tri-CDF  to  hexa-CDF were detected  in two  of the  six
samples.   The soot  and  dust sample taken  from a cable  support  bracket
contained  1,710 ng/g tetra- through octa-PCDF homologues.  Soot taken  from
the ejector fan contained 670 ng/g  tetra- through octa-PCDF homologues.   The
2378-TCDF  isomer  was  not detected  at  detection  limits of 10  ng/g  and
100 ng/g.
     In September  1982, molten steel  at a  steel  mill  in Surahammar, Sweden,
ignited a  500-unit capacitor battery.    The capacitors were  filled  with  two
tons of  PCB's and three  tons  of mineral  oil.  Wipe  samples  from several
locations were analyzed.  Results were  reported  in  terms of unit  area.  Two
samples from  the capacitor  room had  an  average of 620 ng/m2 2378-TCDF and  an
average 7,480 ng/m2  of tetra- through octa-PCDF homologues.
     In 1983, in San Francisco,  California,  a  fire  started in a transformer
vault containing three transformers  filled  with PCB's.158  It  was reported
that only  one transformer leaked.  The  liquid remaining contained 127  ng/g
total TCDD's and 59  ng/g 2378-TCDD.
     A fire  in  Washington  State in  1984,  involved transformer  oil  and
      ^ 1 Q
cores.     A  grab  sample  of the  ash  was analyzed and found to  contain  41.4
ng/g PCDF's and 2.7  ng/g  and 2.5  ng/g of the hepta-CDD  and OCDD homologues,
respectively.
3.2.11  Automobile Emissions
     Table 3-13  presents data from automobiles.
     Dow Chemical  and the U. S. EPA each conducted a study  on emissions from
            C O O 1 O
automobiles.   '       Dow  Chemical  collected particulate  solids  from  seven
types of mufflers.   Results were reported for three  cars burning gasoline and
two trucks using diesel  fuel.   Samples  were analyzed by GC-MS  and  GC-EC.
Results from the GC-MS analyses are reported here except where noted.
Samples from one car which  used  leaded  gasoline and no catalytic  converter
had no detectable 2378-TCDD and 0.004 ng/g  other TCDD's.   The detection limit
for the 2378-TCDD  isomer  was 0.002 ng/g.   Hexa-CDD and  hepta-CDD  were  not
                                    3-43

-------






















CO
1

CO
co

z
UJ
UJ

3
S
55
j-
£
cc
1
o
Ul
£
g
CO
CO

1
CO
UJ

Oj
pj





















in
CD
u
CD
CD
•S-
CD
Cd




•^
C

c?

(X









g
CD





Q
{•:
i
oo
c-
JS,




&
C
ro
K





f»
CD




CD
Q-
co

in
C "O O
3 (8 -r-
<*- Ul rtJ

• _1



>.
<*- +»
O T-
CD £
Q. 0

Hu.




SSS22S S2.
0*4 C^J 0*4 ^1




i i i i i ii i







i i i i i i ii



o
•
O O)

c
CT
CO
•
o
o «
o) TJ o -o o o>
^ D) O) O) ^O O) ^»
O) >•. V. V, O) "^ O>
C J3 O> O) O) -^ O) C
*-« C C C *•» OJ*-" C -^ "O
•»i-H rH C r-l IOCMQ
o • o • in o m
• r^* • r^* • • •
o ^r o en p~ o o


i
i *
0
1 1 1 1 1 1 Q 1


O)
Q) O) O)
"5, ^ "S,
C O) C IT)
e ft

O O* O ^f»HCN4i^Q OC^O
C>J fH ^ O *"* O "^


o> +*o> +»-pd> o> in
c uuc uut-uc c t- 4->
CX +J U Q. +J U +J I_ Q. Q. 4-> «3
(O r""^^ rw4JpM*<4-' j£
in n>R>iaU+>C9'o'io i— i—
C Q) O 'O ^J *O «W tJ O TJ * ^^ ^^ ^J fa* CJ r™" CJ ^J O Q)
3 m CD CD c c c c f- i-


A CO
CD "?
g>8
o
r- >.
O -a
O-o
gl
^? *c
(T3 O
x u
>, CD
e— I •(->
C O
O t- C
O
c in
IB in 
g *« 'V
A<8









•k
.g
c o
t- -a -p
CD CD a.
.a in CD
E * I
=1 -Q
C *
CD O
CD u a
U (O O
(O 1
in a
in c x
CD o CD
in t- i
CD -P
•P i— Q
C 3 Q
CD OCJ
rtJ at >.
Q. O «- r—
o e
c in <*- o
•^ CD
r- U) Q



jQ U» r-
e « -o
3 <«- C C
•z o <  O
c o <*-
<0 CD
81»2
•0 CD
S N

i a CD
Zo
^J
s >a
3-44

-------
detected at detection  limits  of 0.014 ng/g  and  0.006 ng/g, respectively,
while 0.016 ng/g of the OCDD homologue were present.
     The second car had been burning unleaded gasoline and was equipped with
a catalytic converter.  The 2378-TCDD,  other TCDD's  and  hexa-CDD  were not
detected at detection  limits of 0.003  ng/g, 0.001 ng/g,  and  0.01 ng/g,
respectively.   The hepta-CDD and OCDD homologues  were detected  at  levels  of
0.014 ng/g and 0.068 ng/g, respectively.
     The third  car  sampled was burning unleaded  gasoline  with  a catalytic
converter and  had  relatively low mileage  (-15,000 miles).  The 2378-TCDD
isomer was not detected at detection limits of 0.0002 ng/g.  Concentration of
the other TCDD's were  0.0001  ng/g which equaled  the  detection  limit.   The
hexa-CDD homologue  was detected at  0.0005 ng/g  by electron capture  gas
chromatography  (GC-EC)  but these  results were  not  confirmed  by  GC-MS
analysis.  GC-EC analysis  of  the particulate matter  samples detected  0.002
ng/g hepta-CDD and 0.008 ng/g OCDD.  These positive results were confirmed by
GC-MS.
     For samples from  one of the diesel mufflers,  GC-MS analysis  did  not
detect 2378-TCDD, other TCDD's or hexa-CDD at detection limits of 0.003 ng/g,
0.007 ng/g, and 0.025 ng/g, respectively.  Levels  of  hepta-CDD  and  OCDD were
0.110 ng/g  and 0.280 ng/g,  respectively.   The second diesel  muffler had
0.003 ng/g 2378-TCDD, 0.02 ng/g TCDD, 0.02 ng/g hexa-CDD, 0.10 ng/g
hepta-CDD, and 0.26 ng/g OCDD.
     The U. S.  EPA  analyzed four composites  of filter extracts from several
automobiles.   Each  of  the extracts was analyzed  by GC-MS  with  a detection
limit of 0.04  ng.   Results  were  reported  in  terms of  ng  per sample with the
weight in grams  of  the  sample also noted.   For the purposes of  this  report
the reported values were  recalculated to reflect  the  concentration  of  TCDD's
in ng/g.
     A pooled  sample from two  diesel  cars  contained no detectable  2378-TCDD
or other  TCDD's.  Pooled  filter  extracts  from three  cars burning leaded
gasoline contained  2.98 ng/g  of  the 2378-TCDD isomer and 47.7 ng/g of four
other unspecified TCDD isomers.  Samples from  10  cars with catalysts  burning
unleaded gasoline  comprised the third  sample.  The  2378-TCDD  isomer was
                                    3-45

-------
 detected at a  concentration  of 1.4  ng/g.   Nine other TCDD  isomers were
 detected at a  concentration of 37.4 ng/g.   The fourth sample was composed of
 a  filter extract- from a catalytic  car burning  unleaded  gasoline.   The  car was
 malfunctioning  and had excessive oil consumption.  It was tested  separately
 because  its extractable part.iculate emissions  were  so high its  full  inclusion
 in  the  catalyst pool would have skewed  the data.  It was included  in the
 catalyst composite pool at one-tenth its normal emission rate.  Particulate
 extracts from this vehicle contained 0.28 ng/g 2378-TCDD and 7.5 ng/g of 10
 other unspecified  TCDD  isomers.
 3.2.12   Activated  Carbon Regeneration Furnaces
     Table  3-14 presents two  studies  which were  conducted on  activated carbon
 regeneration at the  Cincinnati Waterworks, Cincinnati, Ohio.13'156   The first
 study tested emissions  from the fluidized bed system before  an afterburner
              13
 was installed.     The carbon  regenerated during the first study had been in
 service  for approximately  one  year.  Pre-chlorination of  the wastewater
 (relative to the granular  activated carbon bed)  was in use.
    .Concentrations  of  2378-TCDD in the flue gas ranged  from 0.01 to 0.21
 ng/m3 with  an  average of  0.1  ng/m3.   TCDD concentrations in the  flue  gas
 ranged from 0.06 to  0.3 ng/m3 with an average concentration  of 0.17 ng/m3.
 Flue gas concentrations of TCDF ranged  from 0.08  to 0.51  ng/m3 with  an
 average  of  0.3  ng/m  .  For particulate samples,  concentrations of the 2378-
 TCDD  isomer ranged  from  4.3 to 51  ng/g  with  an  average  of  25  ng/g.
 Concentrations  of TCDD ranged  from 36  to 66  ng/g with  an  average
 concentration of 48  ng/g.   For TCDF's, concentrations ranged from less than
 detectable  to  245 ng/g with  an  average concentration  of  103  ng/g in
 particulate samples.  The  detection limit for TCDF's was 4 ng/g.
     The second  study at this facility took place after  installation of  an
 afterburner.     The  afterburner is located in the off-gas stream from the
 fluidized bed reactivation unit.  The average temperature of  the afterburner
during the  test period was  2500°F.  Post-chlorination of the  wastewater.
 (relative  to  the  granular  activated  carbon  bed),  rather   than
pre-chlorination,  was in  use during  the second study.    The  carbon  being
regenerated had been  in use 200 days.
                                    3-46

-------




















2
1— t
h-
cc

21
UJ
O
1
CJ
Q
£
=»
5

^T



UJ
i
LU

fy
o
CO



%
m

UJ
«j
>-



















U)
ID
O
e
ID
ID
K









•
c
0
4*
C
0)
f""
c
IT
C/>
£
U
<
•a
c
c
c
J
•(->
s
c
3
c
c
It

^J
g
u.
u.
s
a.

a
8
a.















u_

Q*





*• in 10
O Q> O
-5
0 ~


**- 5*
o i—
ll
£2.



ro m \o *O vO vO
1—1 *~* tn in m in
•-H »— (


in m ro m

O CO C*4 "s,, Q
i • ™ i -^ i %= '
' 0) ' 0) ' g '
S C j^ C • tO o\
^^ ^5 ^3 ^* ^4
o z

^^

_ »" ' m m m
*^ E E E E
^ ^v O) O) O) O3
_,2 _"*!£} r»^ *t O>CM
ci'— S "^ • "I ** "^ f* *~
O — 1 o t-3 C3 C3



0)

o m c "^ *t
O 1 Q ^) Q "o>
0 «3 Z = Z C
CO

CO O O
^p co_ co ro
O) C3) "^ ^ ^ **B
o^5 g>§ = =3 = o>?3
1-1 _ CO Ol CM
• co in in -H r»
O TT . . o .



i-H
CM CM
• m .
OS -HO)
1 O O) 1 i i t
C | C

o co
o ^r



"^
O) IO •»»

^ 5 53 i i i i
O CM
ID (D
*^ C C
•P "0 §• 3 * *• '
O ^~ C9  3
— "» "£ _N v> J3
— -o S 2 * ®
3 ID (^ 3 © it.
IZ iZ CD "*









£2
0
in
ID
>< 3
eg1
o ^~
o
v> E
ID o
3 f-
O)
l§
u_

§v
(O
+* J^
^f C^^™
C ID C
(0 I O
O *• C
9U. 10
a u
O in
a i
•M 10 Q
Q. X Q
ID ID O
cc re ^~
0 -0 ID




in
e
o
•r-
3
O


J3
C

in

in
in

« £
'g
0 L.
C 10 -0
Q. 0)
tn +j
S "^ -2
E t- 0
0) ~o
'll
s 10 ja
3-47

-------
     Emissions from the stack, afterburner, and recuperator were tested.  The
2378-TCDD isomer was  not  detected in any  of  the  samples.   PCDD levels  in
stack samples were  1.58  ng/m3,  but only the  hepta-CDD  and  OCDD homologues
were present.  PCDF concentrations in stack samples were 0.5 ng/m .  In these
samples, the hexa-CDF, hepta-CDF and OCDF homologues were present.   Detection
limits for  these samples  were  not specified.  However,  information  was
available for  sample MM5  train  blanks  and  sample train volumes  making
calculation of detection limits possible.  The calculated detection limit for
                                                   3
the 2378-TCDD and other TCDD  isomers was 0.006 ng/m .   For  the  TCDF  isomers,
the calculated detection limit was 0.007 ng/m .
3.2.13  Experimental Studies
     Experimental studies have been conducted on PCDD and PCDF formation from
combustion  of  chlorinated  aromatics  (see Table  3-15).  Chlorobenzenes,
chlorophenols and the  effect of inorganic chlorine on  PCDD emissions have
been studied.  Buser investigated the formation of PCDD's and PCDF's from the
pyrolysis of  chlorobenzenes.    Both PCDD's  and  PCDF's were  detected in
pyrolyzed samples.  The formation mechanism proposed  included  a chlorophenol
intermediate.  Buser also investigated the formation of PCDF's from pyrolysis
of  PCB's.35'36'37   The yields  of PCDF's  were estimated to range  from
0.1 percent to several percent.  The proposed mechanism is  an  intramolecular
cyclization.
     In a pilot  incineration study,  Jansson  combusted  chlorophenol-treated
wood.     PCDD's were detected at levels ranging from 0.2-155 ug/g feed.
     Three reports  have dealt with PCDD formation from combustion  processes
in the  presence  of  inorganic chlorine.    ''      Tiernan et  al_. found  no
                                                                         ??fi
detectable PCDD's or  PCDF's emitted  from the combustion of virgin  pine.
However, in  the presence  of HC1,  significant quantities  of TCDD's  were
detected.  Mahle  et al_.  present similar results  when burning coal  in  the
                               149
                                    Liberati et al. studied the combustion of
presence of C1,, HC1, and NaCl
           145
vegetables.     When inorganic chlorine or polyvinyl chloride (PVC) is added,
PCDD's and PCDF's were detected in the emissions.
     Chlorophenol combustion was  studied  at  various combustion  temperatures
and residence  times by  Environment  Canada.      Combustion  of  2,4,5-TCP,
                                    3-48

-------






























^1
^2
£

*p
s
LU
0.
X
LU


••
%•
ce
§
O

UJ
Q£
3
CO

0
in


m
LU
flrt
i—



























Ul
CD
u
c
CD

CD
CD
tr









+j
CD
•H
e
o
O

CD
c
E
A3
CO

j:
Ul

•o
c
ID
C
c

•»•:

L.
+J
C
8
I
c
o
Ul
U!
E
U.
Lu

^
OL
X,
C
C
£













Q
£








CD
TO
C

t£










i
0}
JT












Q

Fj
^





CD
TO
C
a
C£









C
®
*






c

2
00
f-
R



CD
TO
C
t?




C

<*• ui 10
O CD U
1— 0
o d
z


>»

0 —
CD £
O. U
>> n)
h- Lu




T •-< in o> o
m •-! T T o
i-H rH r-l CM


in T>

00 P~ CD
-• 1 •*- 1 »
|
1 TO
.-J TO
Ot 3
CM

e" ^
A3 (D
Ul M-

TO^^ TO
e in »>. *^
«— TOOO
•-< 1 3^1 1
00
O f-4


TO
Q. "^ TO O -0
E TO >v O O CD
iQ 3 TO C7t T CD
Ul 3 CM PO <*-
>^ in
TO in 13 -H -a i i TO
c -• a> o CD \
CD •-< CD CM -> »- CD O CCD J3 -r- CC
J 1— C _UI 4->f^X CDOi— 3< CD_O

^Ul L.Qt3 ^4^IO +p— O*4^
N CD O >• > -r-cot- CD o :» 3*3 -co




































^
c
o

•o
CD
Ul
10
jQ
CD
!_
Ul
C
o
tjj
10
•5
u
15
u

Ul
CD

i
ia
Ul
^_
0

u,
CD
E
e
CD


-------
 Alchem 4135, Woodbrite 24, and  diptank sludge generated  from Woodbrite 24
 preservation resulted  in PCDD  concentrations  in  flue  gas ranging  from
 0.6-3400 ug/g feed.

 3.3  PCDD FORMATION HYPOTHESES AND FACTORS AFFECTING EMISSIONS FOR COMBUSTION
      SOURCES
      This section -presents  a summary  of  the most  common PCDD formation
 hypotheses  for  PCDD  emissions  from  combustion  sources.   The  section
 summarizes the hypotheses contained  in the literature and also presents  a
 discussion of combustion  device  operating  parameters and  fuel  characteristics
 that  may affect  PCDD emissions.
 3.3.1  A Summary of Formation Hypotheses for PCDD from Combustion
      One of the  earliest  combustion device PCDD formation  hypotheses  advanced
 was that of Dow  Chemical  Company entitled  "The Trace  Chemistries  of  Fire--A
 Source  of  and  Routes  for the  Entry of  Chlorinated  Dioxins  into  the
              O T  RO
 Environment."  '     This  hypothesis was advanced  based on  sampling conducted
 by  Dow  Chemical  Company on samples taken from around the Midland facility  and
 sampling of a wide  range  of combustion  devices.
      This  hypothesis suggested that  PCDD's/PCDF's  in  combustion  effluents
 were  ubiquitous  and  were due to  the trace chemistries  of fire.   This
 hypothesis,  in conjunction with the findings of PCDD/PCDF  in  ashes and stack
 gases  from municipal  solid  waste combustors,  led   to the inclusion  of
 combustion  sources  in the  National Dioxin  Study.
     A  significant  amount of  effort  has been expended  in an attempt  to
 explain  how or why  PCDD's/PCDF's  are  formed in  combustion processes.  Table
3-16  summarizes  hypotheses contained  in the literature.   Much  of the effort
has been directed toward  municipal solid waste  incinerators.   A significant
number of studies and hypotheses have tried  to link  specific precursors  with
PCDD  formation.   The most prevalent  precursors cited  include  chlorinated
phenols and chlorobenzenes.  A considerable amount  of  work has also focused
on the chlorine content of the fuel.   None of the  hypotheses advanced to date
have been proven.
                                    3-50

-------

U
2 P
i • ^
>.Q
3.








§

L. +J
V) (D
ii
00
+J
•a
85
(_
O ^
V) 0)
u.



e o

4J tj
« •<
1 ^
L. U)
O -^
Dioxin F
Hypothes

o
i— i i— i «
on csi rS




U)
CD
6 c

°» s"o .a 2
gj g 0 0 «
5 fell i Jl
W •*•» 4J .C »— 4J<5
o 1 * § ~ « «g 2 5 1 « !«
II ^|5'I - ilH ii
u in **»og o c ^ ^ * c f2 »
*£c •o!ee*o uSo.^®3* -cu*
§ O -M C O O *J 4JC U3*O ®
^ "^ C <9 ^  ^- ** 01 N
tn u ^ uj t_ In *u  -2 a> "Jo am
3 XS -^ 3 3 C O ^- C U 4J Q. ul ^
3 o o o o o o ^ u o o -^ ^!e
z zotkoito




c
TJ
a> -^
in u
*" 9
t i
la -a
« 0 C
| °." s
o o -^ wi
v> o u 
M ^v r«* S
i— tfl -~. — g
•3 o +j o S
< u. a .0 cj

1
i^
o *_
I 1
t 4^ *S rij
•M 5 2 ^
ut o >»-M

II ^ * .
5" «l §1
JD * 3 1«
g| 5 gR
tl S ii!
«
«

0
a\
r-4







2
L.
a.
g
•M U.
o
o o
= ?
ui o
I»




o
§
44
3 Ul
^ s
s s

ul .a
11 II
^- t^ I*.
* "O O O
04 -2 .£
£ 5 is
!• &• ^» ^™
X S >t >»
ua u a. a.
•si
SI
i
U)





a>
e •
1o| £ ^

«l * t
e *
•O iO
U ^ rt)
« £ 2
•»• n •*-
t- X -l

r*» o
S 155'

ig

0 - §
4J — in
fl +* «a
•— W» O)
3 3
£ f N
" ^ u §i ^
a. j= ^ °- "*~
ui t. o> c tn o) c
w wi e •«— ^ m 4— H—
og°e ^j ^2 £CS
*" > ^ 3 3  g "O T3 U
•MOOS' oo -2*0 ^c2o°'
^^•^
isJs II i£ IIII
2 11? si S^ olll

-------
3.3.2  Factors Affecting PCDD Emissions From Combustion Sources
     This  section  discusses  the  following factors  identified  in  the
literature that may affect PCDD emissions:
     o    PCDD in feed,
     o    precursors in feed,
     o    chlorine  in feed,
     o    combustion temperature,
     o    residence time,
     o    oxygen availability,
     o    feed processing, and
                •
     o    supplemental fuel.
The interaction of  these factors  during  the formation of PCDD's is not well
defined.  Therefore, each of the factors is discussed separately below.
     3.3.2.1  PCDD  in Feed.  2378-TCDD is  an  impurity that results from the
manufacture of trichlorophenol, which is used to make the  herbicide  2,4,5-
trichlorophenoxy  acetic  acid (245-T).   Pentachlorophenol  (PCP) production
will also result in a PCDD contaminant, primarily OCDD.  The  primary  end  use
for PCP  is  as a wood preservative.   It  is anticipated  that  limited  PCDD
contamination will  also  occur during the manufacturing of other similar
chlorinated  aromatics,  particularly  if  the  manufacturing  process  is
inefficient or not well controlled.   Therefore,  PCDD's are  expected to  enter
the environment  as a contaminant of commercial  products, such  as  wood
preservatives and pesticides.
     The widespread  use  of  these products increases  the  possibility  of
finding PCDD's in the feed of a combustion process.   For example, PCP-treated
wood may be'used  to fire boilers.   Runoff may carry pesticides  to  water
treatment facilities where the organics are incorporated into a sludge.  The
sludge may then be  incinerated.   Likewise, contaminated waste  streams  from
manufacturing processes may be incinerated as  an energy  recovery procedure.
One example is PCP sludge incinerators used at wood  preserving facilities.
     If PCDD's are found in the feed  of  an inefficient  or  poorly controlled
combustion process, it is  very likely that they will be  released to the
atmosphere.
                                    3-52

-------
     3.3.2.2   Precursors  in  Feed.   Although the  Dow "Chemistry  of Fire"
theory is backed by  a  considerable  amount  of  experimental  data,  many of the
reviewed  studies  focused  on the formation  of  PCDD's  and  PCDF's  from
precursors.  Experiments  by  Buser,  Rappe,  and others are described  in  more
detail in Section 3.2.11.  Esposito et al..  presented detailed descriptions of
the formation mechanisms  of  chlorinated  CDD's  from precursors.     This work
organizes COD precursors  into three classes:
   Class I -  Polyhalogenated phenols, primarily with a halogen  ortho  to  the
     hydroxyl group, with a high probability of CDD formation.
   Class  II  -  Ortho-halophenols  and ortho-halophenyl  esters  where  the
     substituted groups are a mixture of halogens and nonhalogens.
   Class III -  Other  chemicals having the possibility, but  less likelihood,
     of CDD formation.  These include chlorinated aromatic compounds.
     The majority of experimental work to date has  centered  on three classes
of-precursors:  chlorinated phenols, chlorinated benzenes, and PCB's.
     PCDD formation  from the combustion of chlorinated phenols has been
tested extensively by  Rappe190, Jansson111, and Ahling6'7.  Dechlorination of
the highly chlorinated homologues can result  in  the more  toxic TCDD isomers.
Chlorinated phenols are used as wood preservatives, herbicides,  and  sap stain
control.  Wood or vegetation sprayed with chlorophenols may be disposed of by
incineration or used as a supplemental fuel in boilers.
     Buser investigated the formation of PCDD's and PCDF's from  the  pyrolysis
of chlorobenzenes.     The formation mechanism included  a chlorophenol  and a
polychlorinated diphenyl  ester (PCDPE) intermediate.  Chlorobenzenes are  used
in solvents, dyes,  Pharmaceuticals,  and  rubber production.   These  products
make up  much  of the  organic chlorine found  in  feed of  municipal  waste
incinerators.  The associated waste product may also be  disposed of in  an
incinerator or boiler.
     Buser also investigated the  formation of PCDF's from the pyrolysis  of
      O VI *3 C *5 ff                                                 rw»/
PCB's.  '  '    No experimental work  has been identified on  PCDD formation
from PCB's.   However,  studies have been  identified  that found PCDD's emitted
from PCB  fires.60'158  In  addition,  PCB's  are  often  in solution  with
hexachlorobenzenes that have been shown to form PCDD's.  Up until 1975, PCB's
                                    3-53

-------
were used  as dielectric fluids in transformers  and  capacitors.   PCB's have
also been  used  in hydraulic  fluids, plasticizers, and dyes.  The  incineration
of PCB's at  waste  disposal  facilities or in boilers may  result  in  PCDD and
PCDF emissions.
     3.3.2.3  Chlorine  in Feed.  The  chlorine content of  fuel  is  obviously  an
important  parameter affecting the formation of PCDD's or  PCDF's.  Shih  et al_.
developed  a  ranked priority  list of conventional  combustion  systems emitting
                                                        011
polycyclic organic  matter including  PCDD's  and  PCDF's.     The  rationale
presented  for source ranking is based on  fuel  characteristics  and combustion
conditions.  Shih's work  places great emphasis on both the chlorine content
of the feed and the concentration of  aromatics in the feed.
     Other authors  have  demonstrated the  effect of  chlorine  on  PCDD
emissions.   Mahle  et  al_.  demonstrated that  PCDD's  were  emitted  from  coal
combustion only when  chlorine was  added.     Tiernan  et al_.  found  PCDD
formation  during the combustion of pine in the presence of HC1, but no  PCDD's
were detected during the  combustion  of pine alone.226  Liberti  studied the
combustion of vegetables.      When  inorganic chlorine or PVC  were  added  to
the vegetables  and the mixture burned, PCDD's and PCDF's  were  detected  in the
ash.
     While the  precursor  theory has  received  widespread  acceptance,  these
inorganic  chlorine studies demonstrate that  the  specific  mechanisms involved
in PCDD formation are  complex and  not well  understood.   However, it can  be
generally  stated that chlorine must be present for the formation  of PCDD, and
general trends  indicate that increased chlorine concentrations in  the  feed
improve the possibilities of PCDD emissions.
     3.3.2.4  Combustion Conditions.  The remaining factors identified  in the
literature that affect PCDD emissions are  combustion conditions. These
include combustion  temperature,  residence  time, supplemental  fuel,  fuel
processing, and oxygen availability.  Combustion efficiency is a  function of
all of these factors.   In order to destroy PCDD's or prevent their formation,
the combustion efficiency must be high.  This requires a  combination of high
temperatures, available oxygen,  high  heating value  fuel, and long mixing
times.
                                    3-54

-------
     3.3.2.5   Combustion  Temperature.   Experimental evidence  suggests  that
temperatures of 500-800°C promote  PCDD  formation,  while  temperatures greater
than 800°C destroy  PCDD's.6'37'62  Buser  e_t  al_.  showed that  PCB pyrolysis at
550 to 650°C forms  PCDF.192  However, pyrolysis at  a temperature greater  than
700°C causes 99 percent destruction  of  PCB's and no PCDF formation.  Ahling
et  al_.  produced similar  results  for both  PCDD's  and PCDF's  during the
combustion of chlorophenols.
     Several other  studies  have also been  identified that  demonstrate the
effects  of  high   combustion   temperatures  on   PCDD's  and   PCDD
           3 184 234
precursors. '   'a    For example, PCDD's were not  detected  in  the emissions
of the Vulcanus incinerator  ship  during the combustion  of PCDD-contaminated
                 •3
Herbicide Orange.   The combustion temperature during this study was 1600°C.
     Combustion temperature is a function of the heating value of the fuel or
supplemental fuel, the available air, and the degree of fuel processing.  Two
examples of  combustion  sources whose PCDD  combustion efficiencies  differ
because of operating/design differences are  municipal waste  incinerators  and
hazardous waste incinerators.  Municipal waste incinerators  are  considered  a
major combustion  source  of  PCDD's.228   The  large mass  burn  units  are
characterized by low  combustion temperatures.  This is  due  in part  to  high
moisture, low heating value fuel; poor  air/feed mixing as a  result  of a lack
of  feed  processing; and  lack  of  supplemental  fuel.  By  comparison, many
hazardous waste incinerators  and  high efficiency  boilers are  designed for
efficient combustion  and  thus  have high PCDD combustion efficiencies  (e.g.
the Vulcanus incinerator  ship3).   These units  burn high  heating value fuels
or add high heating value supplemental  fuels, and  even if the  air/fuel  ratio
is low,  the air/fuel mixing is efficient.  Typically, the fuels are processed
to decrease moisture  and  improve  mixing.  In many cases,  high temperature
afterburners are  used  for  the combustion  of offgases  which  increases
combustion efficiency.
     3.3.2.6  Residence Time.  The residence time necessary to destroy PCDD's
and the  combustion temperature are  inversely related.   The  higher the
combustion temperature, the  shorter  the required  residence  time for PCDD
destruction.   Likewise,  a low temperature  source will  require  a  long
                                    3-55

-------
residence time  for destruction of PCDD's.   Sachdev  et aJL showed  that  an
increase in  both  temperature and residence time decreased the  formation  of
PCDD's from  chlorophenol  combustion.      Similar results  have been found at
hazardous waste incinerators that run with 1.5 to 2.0  second residence times.
Combustion sources with longer residence times and high temperatures are less
likely to form products of incomplete combustion, such as PCDD's.
     3.3.2.7  Oxygen Availability.  Oxygen availability is a function of both
the air/fuel  ratio and air/fuel mixing  efficiency,  both of  which  are of
concern when  burning solid fuels.  Solid  fuels  and high viscosity liquid
fuels such as waste  tars  burn  as particulates or large droplets; therefore,
portions of  the fuel  are  burned in low oxygen or pyrolysis conditions.   An
insufficient  supply  of oxygen or  poor  air/fuel  mixing will  promote poor
combustion conditions  and PCDD  formation.   Jansson demonstrated  that an
insufficient  air  supply   increases  PCDD  emissions .from chlorophenol
combustion.     Municipal waste  incinerators are usually fired  with  excess
air.  However, large mass burn units may have poor air/fuel mixing due to the
lack of  fuel processing  or poorly  designed  air distribution  systems.
Activated carbon  regeneration  and wire  reclamation  incinerators are  both
designed to  be operated with low excess  air.  All of  these cases have been
shown to emit CDD's.100'102'103
     3.3.2.8  Feed Processing.   The feed material for  a combustion source may
be a liquid,  a  solid,  or a gas.  Both liquid and gas  fuels can be  easily
mixed with air resulting in a high combustion efficiency; solid feeds usually
require some  processing  to improve combustion.  Often solid  feeds  require
drying, shredding,  or separation to  improve combustion.   Similarly,  high
viscosity fuels (i.e., waste tars)  require preparations such as preheating
and atomization prior to combustion.
     Feed processing  will  determine in  part  both oxygen availability and
residence time.   Fine,  homogeneous feed particles will  improve air/fuel
mixing and combustion.  Larger particles will require  longer  residence times
and may  result  in local  oxygen deficiencies  due  to  poor mixing.   High
moisture will  also  decrease combustion  efficiency.  Therefore,  highly
                                    3-56

-------
processed homogeneous feeds  are  less  likely to emit products of  incomplete
combustion, such as PCDD's.
     3.3.2.9  Supplemental Fuel.  When a low Btu fuel is burned, the addition
of supplemental  fuel  will  increase the combustion  temperature  and improve
combustion.  Haile et a].,  tested a boiler co-firing RDF  with  coal.92  The
boiler temperature was  1200°C,  and no PCDD's  were  detected.   Dow Chemical
tested an industrial  incinerator burning waste tars without supplemental fuel
and found ppb  levels  of TCDD's  in the fly ash.62   After  the  addition of a
supplemental fuel, no TCDD's were detected.
                                    3-57

-------

-------
                                   CHAPTER 4
                           TEST PROGRAM DEVELOPMENT:
           SOURCE CATEGORY RANKING AND TEST SITE SELECTION PROCEDURES
     This chapter describes the procedures used to select the combustion
source categories and the specific facilities tested under the Tier 4
emissions testing program.  These procedures were designed to address the
question, "What combustion sources emit CDD's?"  The Tier 4 budget restricted
the ability to answer this question thoroughly, in that it financially limited
both the number of source categories and the number of sites within each
source category that could be tested.  It was not possible to select a
statistically valid, representative sample of combustion sources given the
limited number of tests that could be performed.  The decision was made to
select source categories with a high probability of having CDD emissions, i.e.
"worst case."
     The sections that follow discuss 1) which source categories were selected
for testing, 2) the criteria used to rank source categories in terms of
potential for CDD emissions, 3) the ranking assigned each source category, and
4) the procedure used to select emissions test sites.
     The source category and site selection tasks evolved as the program
progressed.  New source categories such as drum reclamation furnaces, blast
furnaces, and industrial incinerators were added at various times in the stack
testing program.  Similarly, charcoal manufacturing operations and PCP sludge
incinerators that were initially considered for stack testing were dropped
from the program as additional  information was gathered.
     The ash sampling and analysis program was originally envisaged to be an
integral  part of the source category ranking and site selection process.
Unfortunately,  PCDD/PCDF analyses of the ash samples could not be conducted
quickly enough  to serve in the  ranking and site selection process.   However,
                                      4-1

-------
results of ash analysis for some source categories reported by other agencies
played a part in site selection as, for example, in the inclusion of a
secondary metals blast furnace in the test program.

4.1  SOURCE CATEGORY SELECTION AND RANKING
     This section discusses the approach used to select and rank combustion
source categories for the Tier 4 emission test program.  The initial ranked
list of source categories is presented, and modifications that were made to
this list are discussed.  Details on the rationale used in ranking the various
sources are also provided.
4.1.1  Development of Ranked Source Category List
     4.1.1.1  Preliminary Ranked Source Category List.  The Initial Tier 4
Literature Review provided information on 1) previous CDD emissions studies,
and 2) experimental studies of CDD formation mechanisms.  This review produced
a list of 13 broadly defined source categories for which some CDD emissions
data had been collected (see Section 3.2).  Based on a review of the emissions
data and experimental studies done, the initial review hypothesized that the
following factors affect CDD emissions from combustion sources:
     1.   PCDD in the feed,
     2.   PCDD precursors in the feed (chlorophenols, chlorobenzenes, PCB's),
     3.   total chlorine in the feed,
     4.   combustion conditions (temperature, residence time, and oxygen
          availability),
     5.   feed processing, and
     6.   supplemental fuel.
The general method used to select the combustion source categories tested, and
the number of facilities to be tested in each category is outlined in
Figure 4-1.  Based on information contained in the field test reports and
experimental studies reported in the literature, a list of combustion source
categories believed to have the greatest potential to emit CDD's was
developed.  This list was modified to a ranked list, in which source
categories were grouped according to their estimated potential to emit CDD's.
The three main criteria used to rank, or group, the sources were source
                                      4-2

-------
               Consider All Combustion Sources
                             I
Look at Previous
Studies Which Have
Generated PCDD Data
                                                 1
Look at all Experimental
Studies to Determine Factors
Affecting CDD Emissions
                  Establish List of Sources
                  Which Should Potentially
                  Be Included in The Program
            Develop "Preliminary Ranked Source
            Category List" Based on The Potential
            For Each Source Category to Emit
            CDD'.s to The Environment, and the
            Number of Tests Previously Conducted
            in Each Category
                    Modify List as Further
                 Information Becomes Available
  FIGURE 4-1.   DEVELOPMENT OF RANKED SOURCE CATEGORY LIST.
                           4-3

-------
category size, precursor level in the feed, and the number of valid tests
previously conducted at sites within a source category.  Other factors
considered were those listed above (identified by the experimental studies),
and the location of facilities within the source category (i.e., the potential
for human exposure).
     Source categories were ranked A through D.  Rank A source categories were
expected to emit CDD's and to have the greatest potential for human exposure.
Rank B sources were expected to have lower CDD emission potential based on the
size of the category or the precursor level in the feed.  Rank C sources were
expected to have even less CDD emission potential than Rank B sources.  Rank D
sources were considered to be adequately tested for the purposes of Tier 4.
The Preliminary Ranked Source Category List (from the Initial Literature
Review, October 1984), along with the description of each of the ranking
categories, is shown in Table 4-1.
     4.1.1.2  Modification of Preliminary List.  Over the course of the stack
emissions testing program, the preliminary ranked list was modified as a
result of both 1) the inputs received from national, regional, state and local
agencies, and 2) further study of each of the combustion source categories.  A
few source categories were shifted from one rank to another, while other
source categories not identified in the Initial Literature Review were added
to the list.  The source categories that experienced a change in ranking were
PCP sludge incinerators, industrial incinerators, charcoal manufacturers and
wood-fired boilers.  Those categories that were added to the list for the
first time were drum and barrel reclamation furnaces, secondary metal recovery
furnaces and hospital waste incinerators.  Table 4-2 summarizes the changes
made from the preliminary list, and the reasons for these changes.
     4.1.1.3  Final Ranked List.  The final ranked list of source categories
(as used for the testing program) is shown in Table 4-3.  The objective of the
testing program was to test Rank A sources three times each, Rank B sources
once each, and as many Rank C sources as possible.  The categories that were
tested are indicated in Table 4-3.  The rationale for the ranking assigned to
each source category is discussed in the next section.
                                      4-4

-------
     TABLE  4-1.   PRELIMINARY  RANKED  SOURCE CATEGORY  LIST  (October  1984)
Rank A Source Categories

     Sewage Sludge  Incinerators
     Black Liquor Boilers

Rank B Source Categories

     PCP Sludge  Incinerators
     Carbon Regeneration (Industrial)
     Charcoal Manufacturing
     Wire Reclamation  Incinerators

Rank C Source Categories

     Commercial  Boilers Firing Fuels Contaminated with Chlorinated Organic
      Wastes
     Woodstoves                                                  .
     Wood Boilers (Firing PCP-Treated Wood)
     Mobile Sources
     Small Spreader-Stoker Coal Boilers
     Hazardous Waste Incinerators
     Lime/Cement Kilns Cofired with Chlorinated Organic Wastes
     Industrial  Incinerators
     Open Burning (Agricultural)
     Apartment House Incinerators
     Other Sources  Recommended by State and Regional Offices  (e.g., landfill
      flares, forest fires)

Rank D Source Categories

     Municipal Solid Waste (MSW) Incineration
     Industrial Boilers Cofiring Wastes


Rank A -  Large source categories (greater than 1 million tons of fuel
          and/or waste burned annually) with elevated dioxin precursor
          contamination of feed/fuel.  These categories are judged to have the
          highest potential to emit TCDD.
Rank B -  Small source categories (less than 1 million tons of fuel and/or
          waste burned annually) or source categories with limited dioxin
          precursor contamination of feed/fuel.  These categories have a
          high potential  to emit TCDD.
Rank C -  Source categories less likely to emit TCDD.
Rank D -  Source categories that have already been tested three or more
          times.

SOURCE:   Initial Literature Review
                                     4-5

-------
                 TABLE 4-2.  MODIFICATION OF PRELIMINARY LIST
    Source Category
  Modification
 Reason for Modification
PCP Sludge Incinerators
Industrial Incinerators
Charcoal Manufacturing
  Operations
Wood-fired Boilers
Drum and Barrel
  Reclamation Furnaces
Secondary Metal
  Recovery Furnaces
Hospital Waste
  Incinerators
Rank B to Rank C
Rank C to Rank B
Rank B to Rank C
Rank C to Rank B
Add as Rank B
Add as Rank B
Add as Rank C
No longer practiced.  This
sludge is being incinerated
in hazardous waste
i.ncinerators.

Many units operating in the
UoSo  Potential high level
of chlorinated plastics in
the feed.

PCP-treated wood is not used
to a large extent in this
source category.

Very large source cateogy.
Some units burn wood stored
in salt water (i.e., wood
contains Cl").

Over 100 facilities in U.S.
Precursor levels may be high.
Poor combustion conditions.

Chlorinated plastics often
present in the feed.  PCDD
detected in baghouse catch.

Burning chlorinated plastics.
                                      4-6

-------
                TABLE 4-3.   FINAL RANKED SOURCE CATEGORY  LIST
Rank A


Rank B
Rank C
Rank D
Sewage Sludge  Incinerators  (3)
Black Liquor Boilers  (3)

Industrial Incinerators  (1)
Wire Reclamation Incinerators (1)
Carbon Regeneration Furnaces (1)
Secondary Metal Blast Furnaces  (1)
Wood-Fired Boilers (1)
Drum and Barrel Reclamation Furnaces (1)

Mobile Sources3 (1)
Wood Stoves2 (1)
Spreader-Stoker Coal  Fired Boilers
Commercial Boilers Burning Chlorinated Organic Wastes
Lime Kilns
Cement Kilns
Hazardous Waste Incinerators
Hospital  Waste Incinerators
Apartment House Incinerators
Charcoal  Manufacturing Operations
Open Burning

Municipal Waste Incinerators
Commercial Waste Boilers
(3) Indicates that the source category was tested three times under Tier 4,

(1) Indicates that the source category was tested once under Tier 4.


 Indicates that the source category was tested once under Tier 4 in
 conjunction with another test program.
                                     4-7

-------
4.1.2  Rationale for the Source Category Ranking
     4.1.2.1  Rank A Sources.  The two Rank A source categories are sewage
sludge incinerators and black liquor boilers.  Three Tier 4 source tests were
performed in each of these categories.
     There are over 200 sewage sludge incinerators operating in the United
States.  Most sludge incinerators are large units (burning several tons/hr)
and many operate 24 hrs/day.  There is a potential for the feed to contain CDD.
precursors at some facilities, especially where the wastewater influent
contains significant input from the industrial sector.  Some plants receive as
much as 60 percent industrial flow.      Precursors (chlorophenols,
chlorobenzenes, and PCB's) have been found in parts-per-million (ppm)
concentrations in the wastewater and/or sludge at some facilities.  Some
sludges also contain several hundred ppm of inorganic chlorine.  Additionally,
combustion conditions in these incinerators would tend to promote CDD
formation.  The feed characteristics can vary widely and maximum furnace
temperatures are typically 700-800°C.  In an earlier Canadian study, PCDD's
                                 23
were detected in stack emissions.
     The black liquor boiler source category is similar to sewage sludge
incineration in that both have numerous large units burning material capable
of forming CDD's.  Approximately 42 million tons of black liquor are burned in
boilers annually.105  Initially, it was felt that the chief sources of
precursors in the black liquor circuit would be PCP-treated wood, biocides
used in the pulping operation, or wood stored in salt water; however, the
primary source of chlorine in the black liquor at many facilities is spent
acid from C102 generation systems.  The spent acids are added to the black
liquor circuit as a source of make-up sodium and sulfur.  An alternate
significant source is makeup water.  Black liquor chlorine contents are as
high as 2 to 3 percent by weight for some plants.  No previous CDD emission
tests have been performed in the black liquor boiler source category.
     4.1.2.2  Rank B Sources.  Six source categories ultimately ended up as
Rank B sources: industrial incinerators,  wire reclamation incinerators, carbon
regeneration furnaces, secondary metal blast furnaces, wood-fired boilers, and
drum and barrel reclamation furnaces.  One emissions test was performed in
each of these source categories.
                                      4-8

-------
     Thousands of batch-fed solid waste industrial incinerators have been
installed in the last ten years.  While the feed to these units is highly
site-specific, some feed streams contain plastics, PCP-treated wood, and other
materials that may contain chlorinated-organics.  Most units have multiple
combustion chambers, the first of which is a "starved-air" chamber which could
be conducive to COD formation.  Trace amounts of CDD were found in the only
                                           41
previous test of an industrial incinerator.
     The size and general practices of the wire reclamation incineration
source category are largely undefined.  Many facilities have switched from
incineration to wire chopping as a means of reclaiming the metal in the wire.
Of those facilities still -incinerating wire, many do not have state permits
(according to state agencies) and/or do not belong to the industry trade
organizations, such as the National Association of Recycling Industries or the
Wire Association International.  Therefore, information is limited.  Most wire
reclamation incinerators currently in use are dual chambered, with a starved
air combustion chamber and a gas-fired afterburner chamber.  The major
precursors being burned are polyvinylchloride (PVC) in wire coatings, and
PCB's in transformers.  Industry and trade organization contacts indicated
that the burning of PVC-coated wire is discouraged; however, this wire
continues to comprise at least a small fraction of the total feed to most wire
incinerators.  In previous tests, parts-per-trillion (ppt) levels of TCDD's
were detected in stack scrapings.103
     There are few furnaces regenerating industrial carbon operating in the
United States.  Those facilities that regenerate carbon used to treat
industrial waste streams (especially for industries handling CDD precursors)
have a high potential for CDD emissions.  This is due to the low air/fuel
ratio and the low combustion temperatures, and also the heterogeneity of the
waste streams treated by the carbon.  PCDD was found at ppt levels in previous
analyses of furnace fly ash samples.  The carbon regenerated in these tests
had been used to treat a municipal water supply, and contained precursor
levels far below what would be found in carbon used to treat many industrial
streams.
                                      4-9

-------
     The secondary copper subcategory of the secondary metal blast furnace
source category  is also small.  The operating facilities have a good potential
for CDD emissions based on feed materials and combustion conditions.  The feed
to these furnaces is made up of different types of metallic scrap, and
typically contains significant amounts of plastic-bearing materials (i.e.,
PVC-coated wire, telephone scrap, etc.).  Combustion conditions are poor,
since scrap is charged on a batch basis.  Feed/air mixing tends to be poor,
and the temperatures throughout the furnace vary.
     Approximately 60 million tons of wood are burned annually in wood-fired
boilers in the United States.  This was the largest source category tested in
terms of the amount of material burned.  Some facilities in the Pacific
Northwest store wood in salt water prior to combustion in the boilers.  Thus,
feed to wood-fired boilers can contain several weight percent of inorganic
chlorine.  However, the organic chlorine content of the wood burned is
expected to be very low.  No facilities were identified that regularly burn
wood that has been treated with pentachlorophenol or other chlorinated
organics.
     The drum and barrel reclamation furnace source category is made up of
approximately 250 facilities nationwide that recondition drums by burning
and/or washing processes.  Over 14 million drums are processed annually using
drum reclamation furances.  The amount of waste residue removed by burning has
been estimated at 35,000 tons annually, making this a fairly small source
category.  The material being burned however, can contain most different kinds
of wastes, including chlorinated organics.  Some facilities burn drums
containing wastes generated during pesticide manufacturing.  Additionally,
many drums are coated with phenolic resins.  Most furnaces are equipped with
afterburners.  Temperature in the furnaces average 650°C (1,200°F), while
afterburners achieve average temperatures of approximately 750°C (1,380°F).
Residence times of less than one second in the combustion zone are common for
this industry.
     4.1-2.3  Rank C Sources.  The Rank C source categories are as follows:
mobile sources, woodstoves,  small spreader stoker coal-fired boilers,
commercial boilers burning chlorinated organic wastes,  PCP sludge
                                     4-10

-------
incinerators, lime kilns, cement kilns, hazardous waste incinerators, hospital
waste incinerators, apartment house incinerators, charcoal manufacturing
operations and open burning.   With the exception of mobile sources and
woodstoves, these sources were excluded from the stack testing program for one
or more of the following reasons:
     1.   Minimal CDD precursors present in the feed.
     2.   Source category believed to be no longer existent or considerably
          different from what it was when identified in the Initial Literature
          Review.
     3.   High combustion temperatures and destruction efficiencies required
          by law indicate low potential for CDD emissions.
     4.   Insufficient budget to test all source categories of interest.
     It should be noted that both mobile sources and woodstoves were sampled
for CDD emissions through other on-going testing programs.
     4.1.2.4  Rank D Sources.  The Rank D source categories are municipal
waste incinerators and commercial waste boilers.  Several  CDD emission tests
have been performed in each of these categories.  The categories were
considered to be adequately tested for the purpose of Tier 4.   Data from the
previous tests have been incorporated into the literature review (Chapter 3).
Where high quality, complete stack tests are available for these sources, the
data have been incorporated in the Tier 4 data analyses (Chapters 5 and 6).

4.2  TEST SITE SELECTION
4.2.1  Background
     The objective of the test program was to focus the testing efforts on the
source categories believed to have the greatest potential  for CDD emissions.
Therefore, Rank A sources were tested three times each, Rank B sources were
tested once each, and Rank C sources were tested only when testing could be
done easily in conjunction with other programs.  In the Initial Literature
Review, it was stated that the results of the Ash Sampling Program analyses
would probably be used to determine which of the Rank B and C categories would
be emission tested; however, these results were unavailable until later in the
program.  Ultimately, the final ranked list (Table 4-3) was used as the basis
                                     4-11

-------
for selecting which source categories would be tested, and the extent to which
each category would be tested.
     The differences in the extent of testing desired for each source category
dictated that a different approach be taken toward site selection for each
category.  For source categories where only one test would be conducted, a
facility was chosen that was considered to have a greater potential for CDD
emissions than most others in that source category (these were termed "high
potential" facilities).  This selection was based upon a qualitative
assessment of source characteristics and factors affecting PCDD/PCDF
emissions.  This is described in more detail below.  At least one "high
potential" facility was chosen for each Rank A source category.  Other Rank A
sites chosen were believed to possess average or above average CDD emission
potential, as compared to other sources in that category.  These facilities
were referred to as "average potential" or "average/high potential" sites,
respectively.
     A major part of the site selection process was, therefore, source
category characterization.  The following variables were examined in each
source category:
     1.   average size of combustion units,
     2.   range of combustion unit sizes in source category,
     3.   subcategories of combustion units (e.g., multiple hearth or
          fluidized bed for sewage sludge incinerators),
     4.   precursor levels and other feed stream characteristics,
     5.   supplemental fuel usage,
     6.   normal combustion unit operating characteristics (% 02,
          temperatures, feed/air mixing, operating schedules),
     7.   control device types, and
     8.   control device design and operation.
This information was used to subjectively rank the sources within a source
category by the potential to emit CDD and,  therefore, determine which
facilities would be tested. Due to time constraints,  only information that was
readily available was obtained.
                                     4-12

-------
4.2.2  Site Selection Methodology
     The general approach used  in selecting test sites is shown  schematically
in Figure 4-2.  Specific site selection procedures for each source category
will be briefly discussed in Section 4.2.3.
     The initial step taken was to develop a list of facilities  within a
source category.  The next step, often taken concurrently with the first, was
to characterize the  source category in terms of the variables listed  in the
previous section (4.2.1).  Then, site-specific information was gathered on
units within the source category.  Of primary interest was the level  of.CDD
precursors in the feed to these units.  Site-specific information came
primarily from the following sources:
     1.   recommendations from  federal, state and local environmental
          agencies,
     2.   previous studies of specific facilities, and
     3.   industry trade organization contacts.
     Selected facilities were contacted by telephone to confirm  the
information gathered about the  site, to determine their willingness to
participate in the Tier 4 test  program, and to discuss factors which would
affect sampling.
     The last step taken, prior to selecting a site for testing was to conduct
a pre-test survey.  Pre-test surveys were conducted at several of the
facilities telephoned (typically three pre-tests were conducted for every test
site selected).  The functions of each pre-test survey were as follows:
     1.   To determine if a site was suitable for the Tier 4 testing program.
          Precursor information, combustion conditions,  and sampling locations
          were examined to estimate both the approximate potential for CDD
          emissions and the ease with which the facility could be sampled.
     2.   To explain the Tier 4 program to site personnel  (i.e.,  the purpose
          of the program,  the time involvement necessary,  etc.).
     3.   To collect ash samples for analysis.
4.2.3  Test Sites Selected
     Test sites were selected based on the information gathered during
pre-test survey visits.   A summary of the information collected at all tested
and pre-tested sites is  presented in this section.
                                     4-13

-------
        Develop  list of  sites in the source category.
       ! Use Source  Category Surveys, Trade Organization
       jDirectories, Background Information Documents
       :and other sources.
       Characterize the source category.  Determine what
       constitutes a "typical" facility in terms of the
       following characteristics:
         o  Combustion Unit Size (feed rate)
         o  Unit Design (e.g., multiple hearth or
            fluidlzed bed for sewage sludge incinerators)
         o  Feed stream characteristics (precursor
            levels, percent solids,  organic content)
         o  Combustor Operating Characteristics (XO,,
            temperatures, feed/air mixing, operating
            schedule)
         o  Supplemental  Fuel  Usage
         o  Control Device Design and Operation
       Gather Site  Specific Information.  Focus on the Precursor
       Level  in  the Feed to Typical Units.  Primary Information
       Contributors:
            o    Federal, state, and local environmental  agencies
            o    Previous studies
            o    Industry trade organization contacts
       Contact a group  of  potential test sites based on infor-
       mation gathered  in  previous step.  The functions of these
       contacts (made by telephone) were:

            o    Gather further  information about site to
                 determine suitability for testing (i.e.,
                 precursor levels, testing considerations)
            o    Determine willingness of site to participate in
                 the testing program
   Conduct pre-test survey at several of these facilities..
   The purposes of the pre-test visits were:

           o    Further determine the suitability of the site
                for testing
           o    Explain the Tier 4 program to site personnel
                (i.e., purpose of the program, time involvement
                necessary, etc...)
           o    Collect ash samples for analysis
            Select test site from information collected
FIGURE 4-2.   PROCEDURES  USED  FOR  SELECTING  A  GENERIC TEST  SITE.
                                    4-14

-------
     4.2.3.1  Sewage Sludge Incinerators.  For the Tier 4 program, pre-test
surveys were conducted at 11 sewage sludge incinerators.  Of these, three were
selected for testing (Sites SSI-A, SSI-B, and SSI-C).  Two of the test
facilities were judged to have "average" potential for CDD emissions with
respect to other sludge incinerators.  The third test site was judged to be a
"high potential" site based upon consideration of the factors affecting
PCDD/PCDF emissions.
     The characteristics of the pre-tested facilities are summarized in
Table 4-4.  The common unit type was a multiple hearth furnace controlled by a
scrubber system.  Typical precursor levels were ppm or parts per billion (ppb)
levels of various chlorinated organics.  Combustion temperatures were in the
500 to 800°C (930 to 1,470°F) range.
     4.2.3.2  Black Liouor Boilers.  Six black liquor boilers were pre-test
surveyed.  Three facilities were chosen for Tier 4 testing (Sites SSI-A,
SSI-B, and SSI-C).  The CDD emissions potential of the test sites ranged from
"average" to "high."  This potential was based on the the amount of chlorine
found in the feed to these units.
     Pre-test facility characteristics are shown in Table 4-5.  There were
several major differences among the units including the boiler manufacturer,
the evaporation technique employed, and the source of the make-up chemical
used.  All units were controlled by electrostatic precipitators.  Sources of
chlorine- in the black liquor include C102 generation waste acids that are
added to the black liquor as make-up chemical, salt in the wood chips used in
the pulping process (resulting from salt-water storage of wood), and chloride
contained in the ground water that is ultimately used as makeup in the black
liquor circuit.
     4.2.3.3  Industrial  Incinerators.  Only one industrial incinerator site
was pre-test surveyed,  and this site was subsequently tested.  The site was
identified by the state and regional EPA offices as a good candidate for the
program because of the high precursor concentration in the feed, and the
site's willingness to participate.  The applicability of this source category
for the Tier 4 program was explored and subsequently confirmed in several
                                     4-15

-------












1
CO

o
1

1
o
2-»
UJUJ
SP
Eg;
">£
feco

8 :S
O 2
Hi CO

8 UJ

G J£
is


•-•a
.
i

UJ

P















J,
CO
CO

CO
CO


1
r-t
CO
CO

x

CO


2
CO
CO

£
CO
CO




UJ
CO
CO

a
1
,i«
CO
CO



£5
CO
CO



m
CO
CO


CO
CO


e
TJ
s
©
+*
V)

o
—1


0)


.e
en
X


•N*. .G
o> ei
< x


w) O)
> •»-
•< X


en
x




*
1,
>



01
5





X J,
X



v. jr
x f |>



O)
X 3



O T3 UI r-
+* a e <9
CO « •*- *- 5
1 X in c
4* *™ O U) ffl
S « 5"i ?
i— u ui a.
a:
£ m L,



x en S
S «


cc
£u
. CO



a:
X O
£ CO
X


£ m 1
f

O
5 g
5^- CO
fe *





£ §





X u, «





£ a |




X 0, Si g
S • L, CO
CM •O 2



tn o
£ £ S


8
§e in >
O ^ (D
S 1*1 °
S 5=5^ -3
+* ^ C O 4>*
— 0 O • C
C 0 +> 0
3 U. — « O

3t
O


f— e
M ** 3 O O
 J3
in «-
CM 3 C.
•O +J
M O t- 3 O O
• \O  *- in »
SO 0 3 U ^-1
1 Q. -T *O •**
in c c
t-H O
O
e
^- O
•o -^
8^ ,0 o
Ijj-J- 0

c c
i— C
>9 O


N U) J3
3 •»-
•o fc.
C 4*
"§
o
|g

S5 § S
01 5r o
• ^— ••< ^H
0 O


S N
Q. C
°-s § s
S.'_ S 3
e> o

s • N
a. in oL c
a.* o. o o o

I-H a. 01 •-<
o do o u
e §"« u a.
o St. o E
t- — |-0 -MO
0 +« _ C J»l-
£1 lo OJ4*
3 E EC CO
O «4- X U. O 4-> U.
t. C 10 CO £ 30^




































in

UI >l
ID C
e o
o o js
cS5"
5£-°
3 *O
T3 !*• 0)
0 +>
19 +J e t.
5 ^ " 5
L. O O J3
O f r— 3
O 1= O O O
Q)O ^ >» UI
19  i— X O CO
< O 4 CL X

O
"o
E
4-16

-------












co
£
1— 1
&
UJ
t— t
§

«»
DC
1— 1 UJ

Xi *~
< o
^J u_
CO
1— >•
•t. UJ
o 5
UJ 3
§"
If
oi£
P CO
•t z
§1-4
o|

1-1


•
in
i


UJ
§
t—



















x
si

u_

5J




Q

Si








U
m

OQ




to











«c

_J
CO








CD
•o
o
0
CD
•P
co






0 1

§
O) O
* s
m




JC
O)








§
D) O
X > •
< o
o
en



8
£ O
X i 8
i— i









^ i
O) o> o
X 3E o
CM




IS ~
C 5 J=
•^ C ^
XS 5
S£ c
o -o in CD
•P CD C -P
•"• -P O <0
co s -S- ce
e in
•p *- in -o
in -p — co
CD in E o>
|_ UJ UJ u.
o

•p 2

•o •—
O O)
S.S1
X C
UJ

>^
CN L.
• -o






m.£
i-l -0








^^
>^
CO "••
*3




^_^
U) >.
en -a










-
Ul L»
*3






^
e ^*
O hv
fS"S
CO U.
c
o "5
54=
CO
1
o
in

c
c
i















5 u
CD
•o -p
O fl)
L. X
O

in p-
in in


§ °
<4- U
CO
CO C
JC CO
IO t3)

•Vf^4
F- O
10 i—
CO O


^

§-(J
fl
<4- £.
Q) C
y (I)
iO O>
u
*» CS|
•5°
CO O

CO
c
t.
o
t- CD
O U.
U J=
u -p
3
O C
CO «F-






1 1 1

o
o
Q- en i-
CO | 7
uj csa o
en

*
J3
£ o
U U1
in en
£ « i
co o
ui en
>,
u
O



Q.
co en in
UJ 1 -<
IN en
L.
O


o.
CO -. CJ1"
0 X -^ -P ^
•P. O M CO
> w S ^ 
-------
reports.40'41  The characteristics of the facility (Site ISW-A) tested are
summarized in Table 4-6.
     4.2.3.4  Wire Reclamation Incinerators.  Three wire reclamation
incinerators were pre-test surveyed, and one facility was chosen for testing
(Site WRI-A).  The facility chosen was judged to be a "high" potential site
because the feed contained some PVC-coated wire and PCB-contaminated
transformer cores.
     Pre-test facility characteristics are shown in Table 4-7.  The typical
wire incinerator is a batch-fed, natural gas-fired unit equipped with an
afterburner for emissions control. • The EPA's authority under Section 114 of
the Clean Air Act was used to gain entrance to two of the pre-tested
facilities in the wire reclamation source category.
     4.2.3.5  Carbon Regeneration Furnaces.  A single site was pre-test
surveyed and eventually tested.  This site was chosen because it was felt to
be representative of other facilities in the source category.  The facility
was judged to have "average/high" potential for CDD emissions based on the
heterogeneity of the feed stream, and the potential for precursors to be
present.  A summary of the characteristics of the facility tested is given in
Table 4-8.
     4.2.3.6  Secondary Copper Blast Furnaces.  Two blast furnaces were
pre-tested, and one was chosen for testing.  The facility chosen was estimated
to have "high" potential for CDD emissions due to the abundance of chlorinated
plastics in the feed.  The other facility processes little plastic-bearing
copper scrap in its blast furnace and was judged to have "low" potential for
CDD emission.  Characteristics of the pre-tested facilities are given in
Table 4-9.
     4.2.3.7  Wood-Fired Boilers.  Four wood-fired boilers were pre-tested.
One site was chosen for testing.  This facility was judged to have "high"
potential for CDD emissions with respect to the rest of the source category
because it fires wood which has been stored in salt water (i.e., salt-laden
wood.  The inorganic chlorine content in the feed to this unit is high.  The
other facilities pre-tested also reported using some salt-laden wood, but to a
lesser degree than the site chosen for testing.
                                     4-18

-------
 TABLE 4-6.  INFORMATION COLLECTED AT THE INDUSTRTAI
  INCINERATOR TEST SITE DURING THE PRE-TEST SURVEY
 Site Code

 Estimated Dioxin
 Emissions Potential
 ISW-A

 High
 Unit  Design
Feed Rate
Dual Chamber
Controlled-Air
Incinerator

2300 Ib/hr
Control Device
                                       Afterburner
Feed Material/
Precursor Information
Wood scraps,
Plastic (polyvinyl
chloride), Paint Sludge
Maximum Furnace Temperature
1800°F
                      4-19

-------
             TABLE 4-7.  INFORMATION COLLECTED AT WIRE RECLAMATION
             INCINERATOR SITES DURING PRE-TEST SURVEYS FOR TIER 4
Site Code
  WRI-A
   WRI-B
   WRI-C
Test Site
Estimated Dioxin
Emissions Potential
X
High

Avg/
High

Avg/
High
Unit Design
Dual-chambered,  Propri etary
batch-fed
                Dual-chambered
                batch-fed
Feed Rate (Ib/hr)
  800
Control Device
Afterburner
Afterburner/    Afterburner
baghouse
Feed Material/
Precursor Information
Wire,
Transformers
Wire, other
metal scrap
Wire, other
metal scrap
Igcinerator Temperature
  1000
Afterburner Temperature
(6F)
  1900
                  1600-1700
                                     4-20

-------
TABLE 4-8.  INFORMATION COLLECTED AT THE CARBON REGENERATION
            FURNACE TEST SITE DURING THE PRE-TEST
 Site Code

 Estimated Dioxin
 Emissions Potential
CRF-A

High
 Unit Design
Multiple Hearth
 Feed Rate
63,000 Ib/day
 Control  Device
Afterburner, spray cooler,
     baghouse
 Feed Material/
 Precursor Information
Spent carbon potentially
 containing chlorobenzenes
 and other chlorinated
 organics
 Maximum Gas
 Temperature
1800°F
                             4-21

-------
       TABLE 4-9.   INFORMATION COLLECTED AT SECONDARY COPPER BLAST FURNACE
                    SITES DURING PRE-TEST SURVEYS FOR TIER 4
 Site Code
    MET-A
                                                               MET-B
 Test  Site
Estimated Dioxin
Emissions Potential
      X


    High
    Low
Feed Rate
Control Device
Feed Material/
Precursor Information
Maximum Temperature (°F)
2400 tons/day


 Baghouse
Plastic-bearing
copper scrap, coke
    1500
200 tons/day


  Baghouse
Non-plastic-bearing
copper scrap,
metallurgical slags,
coke
                                    4-22

-------
     The characteristics of the pre-tested facilities are summarized in
Table 4-10.  All of these used Dutch oven "hogged-fuel" boilers.  The common
control device used for these units is a baghouse.
     4.2.3.8  Drum and Barrel Reclamation Furnaces.  Four drum and barrel
reclamation facilities were visited and one was chosen for testing.  The
testing, and some of the pre-test surveys, were conducted under authority of
Section 114 of the Clean Air Act.  As is noted in the table, a change was made
in the code for the sites between the site selection process and the source
testing.  The site labeled DBR-A was the subject of testing.
     A summary of the information collected during pre-tests is given in
Table 4-11.  All facilities used a continuous-feed conveyor furnace equipped
with natural gas-fired burners and an afterburner.  Precursor levels were
expected to be significant at all facilities pretested.

4.3  SUMMARY DESCRIPTION OF ALL TEST SITES
     A brief summary description of the 13 test sites in terms of
manufacturer, feed rate and control  device is given in Table 4-12.  Further
site specific design and operating data are contained in the individual  site
test reports.
                                    4-23

-------
         TABLE 4-10.  INFORMATION COLLECTED AT WOOD-FIRED BOILER SITES
                      DURING PRE-TEST SURVEYS FOR TIER 4
Site Code
    WFB-A
   WFB-B
   WFB-C
WFB-I
Test Site
Estimated Dioxin        High
Emissions Potential
                    High
                  High
                 High
Unit Design
Dutch oven
Dutch oven     Dutch oven    Dutch oven
Feed Rate (Ib/hr)     45,000
                  175,000
                 50,000
Control Device
 Cyclone/
 Baghouse
Multiclone/      Baghouse
Electrostatic
gravel bed
Feed Material/
Precursor
Information
Hogged-fuel,
primarily salt-
laden wood
Hogged-fuel,
some salt-
laden wood
Hogged-fuel,  Hogged-fuel,
some salt-    some salt-
laden wood    laden wood
Maximum Temp.
in Boiler (°F)
  1800
 2300-2600
                                    4-24

-------
        TABLE  4-11.   INFORMATION  COLLECTED AT  DRUM AND  BARREL  RECLAMATION
                     FURNACE  SITES  DURING PRE-TEST SURVEYS  FOR TIER  4
Site  Code
   DBR-A
DBI-A
                                                     DBI-C
                DBI-F
Test Site
Estimated Dioxin
Emissions Potential
    High
High
High
Avg
Unit Design
Conveyor
Conveyor     Conveyor
             Conveyor
Feed Rate (drums/hr)
   125
   150
  180
                                               100
Source of Barrels/     Paint industry,  Textile     Chemicals,     Orange
industry,  Chemicals,  orange juice     plants,     textiles, oil  juice,
Precursor Information  concentrate      chemicals,                 textiles
                                        paint
                                        industry
Incinerator Temp. (°F)
    1000
1100
 1200
                                              1400
Afterburner Temp. (°F) 1450-1600     1200-1300
                                1350
                             2600
*The pretest site code for site DBR-A was DBI-B.
                                     4-25

-------
            TABLE 4-12.  SUMMARY DESCRIPTION OF THE TIER 4 TEST SITES
                           SEWAGE SLUDGE INCINERATORS
SSI-A
SSI-B
SSI-C
3.75 wet TPH Envirotech six-hearth multiple hearth
incinerator controlled by a cyclone followed by a
venturi/impingement tray scrubber.  Average industrial
contribution to treatment plant influent.

12.5 wet TPH Envirotech nine-hearth multiple hearth
incinerator controlled by a cyclone followed by a waste
heat boiler, a venturi scrubber and a subcooler.  Average
industrial contribution to treatment plant influent.
Installed in 1983.

13.5 wet TPH Nichols twelve-hearth multiple hearth
incinerator controlled by a 3 tray impingement scrubber.
High industrial contribution to treatment plant influent.
Installed in 1974.
BLB-A
BLB-B
BLB-C
          BLACK LIQUOR BOILERS

900 TPD low odor B&W recovery boiler controlled by a wet
bottom ESP.  Average chloride content in black liquor.
Installed in 1983.

800 TPD direct contact CE recovery boiler controlled by a
dry bottom ESP.  High chloride content in black liquor.
Installed in 1962.

850 TPD low odor CE recovery boiler controlled by a dry
bottom ESP.  Average chloride content in black liquor.
Installed in 1973.
Wood Fired
Boiler WFB-A
Woodstove
WS-A

Industrial
Solid Waste
Incinerator
ISW-A
             WOOD COMBUSTION

100,000 Ib/hr steam B&W Dutch oven boiler controlled by
a baghouse.  Salt-laden wood fired in the boiler.
Installed in 1939.

Atlanta Stove Works free-standing non-catalytic woodstove.
Oak and pine were burned during the tests.

18 MMBtu/hr Kelley Model 2500 two-chamber controlled air
incinerator with an afterburner.  Incinerator feed
contains polyvinyl chloride and pentachlorophenol-treated
wood.  Installed in 1981.
                                       4-26

-------
                             TABLE 4-12.  Continued.
Secondary
Copper Blast
Furnace MET-A
Wire
Reclamation
Incinerator
WRI-A
        SECONDARY METALS RECOVERY

625 TPO (scrap input) rectangular shaft blast furnace with
oxygen enrichment.  Emissions controlled by afterburners
and a baghouse system.  Furnace feed contains
plastic-bearing copper scrap.

1200 Ib/hr United Corporation Model G-466 batch feed wire
reclamation incinerator controlled by a natural gas-fired
afterburner.  Incinerator feed consists of coated wire
and/or drained transformer cores containing PCB's.
Installed in 1978.
Drum and
Barrel
Reclamation
Incinerator
DBR-A (DBI-B)
Carbon
Regeneration
Furnace CRF-A
          MISCELLANEOUS SOURCES

150 drum/hr ECO Model 100 drum reclamation furnace
controlled by a natural gas-fired afterburner.
Combustible materials in feed include solvent residues and
drum coatings.  Drums originally contained lacquer,
organic solvents, inks, enamel-type paints, etc.
Installed in 1974.

109,000 Ib/day (bare carbon) multiple hearth furnace
controlled by an afterburner, a sodium carbonate spray
cooler, and a baghouse.  Spent carbon feed was used for
treatment of industrial wastewaters known to contain
chlorinated organics.  Furnace rebuilt in 1980.
                                     4-27

-------

-------
                                     CHAPTER 5
                          TIER 4  EMISSIONS  TESTING  RESULTS

      The  purpose  of this  chapter is  to  describe  the  Tier 4  emissions  testing
 program and  to  summarize  the  results of the program.  The summary  of  results
 includes  data on  homologue-specific  CDD/CDF emissions, CDD/CDF  precursor
 contents  of  combustion  device feed materials, continuous monitoring of  various
 combustion products,  HCl/total chloride emissions, combustion device  and
 control device  operating  conditions,  and CDD/CDF contents of combustion device
 and  control  device  ash  samples.   The CDD/CDF emissions data are presented  in
 site-average form in  this chapter, with run-specific data contained in
 Appendix  A.  Additional testing  details can  be obtained in the  site-specific
 test  reports that have  been prepared for each site.  An analysis of the test
 data  is presented in  Section  6-1  of  this report.
      The  remainder  of this chapter consists  of six sections.  Section 5.1
 describes the overall sampling matrix.   The  various samples taken  at each Tier
 4 test site  are listed.   In Section  5.2, the CDD/CDF emissions data are
 presented, followed by the CDD/CDF precursor data in Section 5.3.  The HC1
 emissions data and  continuous monitoring data are presented in Sections 5.4
 and 5.5,  respectively.  Process data documenting the operation of the
 combustion devices and the control devices during the test periods are
 summarized in Section 5.6.  The CDD/CDF contents of ash samples taken at the
 Tier 4 emissions  test sites are presented in Section 5.7,  and in-plant ambient
 CDD/CDF concentration data developed for three Tier 4 test sites are presented
 in Section 5.8.

 5.1  SAMPLING MATRIX
     Table 5-1 summarizes the sampling performed at the Tier 4 emissions test
 sites.  It should be noted that not all  of the samples shown in Table 5-1  were
 analyzed.   An analysis priority was assigned to each stream sampled.   Low
priority streams such as feed samples and soils were often  not analyzed
                                      5-1

-------



§
CD
O
cc
a.
H-
CO
UJ
cc
UJ

UJ
o
u.
X
1—4
en
1
a^

**c
CO
UJ
CJ
0 .
tst

«
1
UJ
s






l£«
sl
- 2
1*1



"?
«i
|




a
c
1
tn
1

« .-"a
3 S£
i- r
£:s
«•••«•
l"i

ai

—
»/»
!«4
|

*«
S

z
irt
•
m
i
_
1




8.5
1-
II
u* ^


5
8
Botto* Ash (U
Scrubber Blood

S-

Ob^
i~



*

_
3

rs
8
<
t/>
VI
•»•
O
•
3
.»
8




ff
u> O
II




j
J5
M ««
«• o
|o~
iz



1

^
8


z
<
V)
O
,**
8

i




8.
a
|


r
if
Bottom Ash (1)
Scrubber Blood
sollds(l). fll

M
0

|g~
5*5



z

^
s


o
X
CO
VI
X.
o
>.
M

i

i— I
£*
s
§•
JC
s
1




,

M
v
||
i =



•

„
w

2
M
<
i
0
5
>•

i

««
2*
o
J(
a




i

•
5
*i
i =



*


M

•M
«
m
i
e
o
z
M

i

<^
**
15
isulatlon F
meer Feed
££
si
• H*
^
fi
I_ ^
PrlMry Chwoei
i Settling Chart)
••*
M ^
^ ^J
J"M
SP



M


S


z
<
J,
S
i
s
I

I




*•
£
1
X


s
s!
3-
a A
3
fi ™ *
« •
JLJ,
i,



*


8

2
M
<
Jj
O
•
i
_
i

^^
^
o
*
wo
Is
VI U.



U

M
•
if
-3



8


5

2
M

i
i
*
a

i



5
?
s
s
v>

s
§-
Reactivated Cat
Baghouse Oust 1

i ^
( O
i .J
iz-



s


8
^
^
1

•
o
z
I

8



1
U.
— W
all

O Q*


•M
Baghous* Dust (

yl
•
II
to **
-I



M
e


8
^

O
Z
^

*
o
o
z
m
*

I
•



83.
II
M «
£3
55
0 Q

ss
M <
Furnace Inlet f
Furnace Outlet

M
e
II
i-



8


M
e

=
>•

i
i













T3
<1)
3
+3
c
o
o


















5-2

-------

















-o

LU
CQ
1— '












x o>
u ui e
<• tn •
III
= 0.",

u

.J

1L
|5



Ul
"5
VI
I
u.



8
I
to
1


**?
Contlnuou
Emission
Monitor 1
at
Q ^»
t/1
"a
C
e *•»
1 +*
3,3
e •
3
o




i


5

8

5
a




—
i
9
CA
•o
O
£
8
-if

«52
* • **
'Is
3&£

s
p

£


„
M
s.
3
V
(/I
V)
™


s




o
z

i






e je
15


«* r*

« »
8*
'S M
0>  r*
a. a. i- x
u i 5
•** ** u o
*e «>• § "u
« o o a.
SS ^ |
*• *• M
S3 S 2
?• 1 "
- 8S 8 ?
i SS i I
••si 11 * s
«-». 0.0. ,.
> • M > . O t>
:«* is o :
2i-£ SS g ?
?St uu „ 2
o u -o e e e e
u * * * o
SO*- 99 X U
^ — « «*. £ o •
9 »• y — ^ . ^
O Q, B » w vi ^
uiS - " N -
^^5 _££ 8 5
S5|I5i 8 C
= =*«•§,§£ ^,. s
»»^~g** 
ifffff fll
01 • ^ ^ _>, >, ID X Q.

• Z**!*^^ kW^
- * S^^J^ £ " J
2j: 5 •o — Z . ^ * 5
*»jie — — ui 01 u
g S jj •i'.w ** ^>.~ to £
!lijjj!il|
af-sTT?1"!!!
^•-c^CMm ^^^
frFaxxx >«
v> T u TI o fr
• in o O o 9»a
*; e ^ a. a, a. 4-» y -o
• *•• s c Q e
*~ ** 9 o & e

5-3

-------
because of the low CDD/CDF concentration measured in the stack samples.  Three
valid test runs were performed at each site except wire reclamation
incinerator site WRI-A.  At site WRI-A, three valid test runs were performed
for each of two feed materials, which made a total of six valid test runs.
Process operating conditions during the valid test runs at all sites were
representative of normal operations; no major process upsets occurred during
testing.  The various sample types taken at each facility are discussed below.
5.1.1  CDD/CDF Emissions Sampling
     Sampling for CDD/CDF emissions was the primary focus of the testing
                     *
effort.  The CDD/CDF sampling generally followed the Modified Method 5 (MM5)
protocol now being adopted by the American Society of Mechanical Engineers
(ASME) for measuring emissions of chlorinated organic compounds from municipal
waste combustors.  This stack sampling method is currently undergoing
validation testing.  Preliminary results indicate that recovery efficiencies
from the sampling method may be low and variable, with possibly less than half
of the CDD's and CDF's in the stack emissions being collected by the MM5
method.  Additional validation testing is currently underway.  Minor
modifications were made to the protocol for the purposes of the Tier 4
program.  The method used did not allow determination of particulate loading
as a conventional MM5 protocol would.  These modifications are discussed in
the Tier 4 Quality Assurance Project Plan.
     MM5 samples were taken downstream of the emission control device (i.e.,
at the control device outlet location) at each test site to measure controlled
emissions of CDD's and CDF's.  At least 240 minutes of on-line MM5 sampling
were performed at the control device outlet location of each test site.  Where
applicable and feasible, MM5 samples were also taken upstream of the control
device (i.e., at the control device inlet location) to develop information on
uncontrolled CDD/CDF emissions and CDD/CDF removal efficiencies of the various
control devices.  As shown in Table 5-1, inlet MM5 samples were taken at eight
      National Dioxin Study—Tier 4 Combustion Sources.
Project Plan.  EPA-450/4-84-014e.
Quality Assurance
                                      5-4

-------
of the thirteen Tier  4  test  sites.   Inlet MM5  sampling was  performed  at  these
sites during the  same time period as the outlet MM5  sampling.  Inlet MM5
samples were not  taken  at Sites  ISW-A, SSI-B,  WRI-A, MET-A,  and WS-A.  There
was no control device at Site WS-A.  Sampling .ports  were not available and/or
not readily accessible  at the control device inlet locations of the remaining
sites.
5.1.2  HC1 Sampling
     Sampling for chloride emissions was performed using the HC1 train at alt
test sites except for the three  sewage sludge  incinerator sites (Sites SSI-A
SSI-B, and SSI-C) and the woodstove site (Site WS-A).  The HC1 train  is
similar to the EPA Method 5  train except that the water in the impingers is
replaced with a sodium  hydroxide or potassium hydroxide solution.  The intent
of the HC1 sampling was to provide an indicator of the amount of chlorine-
containing species present in the combustion gases.  In each case, HC1
sampling was performed  at the control device outlet  location.
     As mentioned above, HC1 sampling was not performed at sewage sludge
incinerator sites SSI-A, SSI-B, and SSI-C.  It was expected that only small
quantities of HC1 would pass through the water scrubbers controlling emissions
from these incinerators.  For this reason, HC1  sampling was thought not to be
cost-effective at these sites.  HC1  sampling was also not performed at
woodstove Site WS-A.
5-1.3  Continuous Monitoring of Combustion Gases
     Continuous monitoring of as many as six species (02, CO, CCu, total
hydrocarbons (THC), NOX and S02) was performed to document the stability of
combustion conditions at each test site.   The monitoring was performed as
close as possible to  the combustion device outlet location (i.e.,  control
device inlet)  so that the potential  for air in-leakage would be minimized.   At
several  test sites, either S02 or S02 and NOX monitoring was omitted.   In
these cases, it was felt that continuous  monitoring of SO, and/or NO
concentrations would not provide significant additional  information on the
stability of combustion conditions,  relative to that provided by the 02,  CO,
C02 and  THC monitoring.
                                      5-5

-------
5.1.4  In-PIant Ambient Air Sampling
     In-plant ambient air sampling was performed near the combustion air inlet
at seven test sites where CDD/CDF or CDD/CDF precursors were potentially
present either in the pre-combustion air or in a dilution air stream that was
ultimately combined with the combustion device exhaust gas prior to
atmospheric discharge.  These samples were not intended to represent downwind
ambient air samples from the source.  The ambient air samples were taken using
a sample train containing XAD-2 resin.  Four of the ambient air samples were
analyzed for CDD/CDF content (Sites MET-A, CRF-A, SSI-C and DBR-A).  The
remainder were not analyzed due to insufficient laboratory support.
5.1.5  Feed Sampling
     Feed samples were taken at each test site, typically on an hourly basis.
A composite of the hourly samples was prepared and split into three parts:
one part was analyzed for CDD/CDF precursors including chlorophenols,
chlorinated biphenyls, chlorobenzenes, and total halogenated organics; one
part was analyzed for total chloride; and one part was retained for potential
CDD/CDF analysis.  Due to time and budget constraints, none of the feed
samples have been analyzed for CDD's and CDF's.
5.1.6  Ash Samples
     Bottom ash and/or fly ash samples were obtained at eleven of the test
sites.  These samples were submitted for CDD/CDF analysis.  The purpose
of the ash sampling effort was to determine whether the presence or absence of
CDD/CDF in ash samples can be used as an indicator of the presence or absence
of CDD/CDF in flue gas emissions.  Electrostatic precipitator (ESP) catch
samples for black liquor boiler Sites BLB-A and BLB-B were not obtained
because a wet bottom ESP was used at Site BLB-A and at Site BLB-B the ESP
catch is recycled directly into the black liquor circuit, and access points
were not available for sampling.  These were the only test sites for which ash
samples were not obtained.
5.1.7  Soil Samples
     Soil samples were taken at all Tier 4 test sites except Site CRF-A and
Site WS-A.  The soil samples were originally taken under the Tier 4 program,
but most of them (10 out of 11) were transferred to Tier 7 for analysis.   The
Site MET-A sample was the only soil sample analyzed by Tier 4 for CDD/CDF
content.
                                      5-6

-------
 5.1.8  Auxiliary Process Samples
     Auxiliary process samples were taken at the three black liquor boiler
 test sites  (Sites BLB-A, BLB-B, and BLB-C).  These samples, which consisted of
 various black liquor circuit  intermediates, were analyzed for total chloride
 content.  The reason for analyzing these samples was to determine the major
 input sources of chlorine to  the black liquor circuit.

 5.2  CDD/CDF EMISSIONS DATA
     This section summarizes  the CDD/CDF emissions data developed in the
 Tier 4 test program.  Data are presented for the outlet flue gas streams
 emitted to the atmosphere for each site, and for flue gas streams upstream of
 a control device for sites where these data were obtained.  Controlled and
 uncontrolled emission concentrations, mass emission rates, and scaled mass
 emission rates (i.e., emission factors) are presented for 2378-TCDD, total
 PCDD, and total PCDF.  The concentration, emission rate, and scaled mass
 emission rate data are contained in Sections 5.2.1, 5.2.2, and 5.2.3,
 respectively.  Contributions  of the various homologues to the total PCDD and
 total PCDF emissions are discussed in Section 6.3 of Chapter 6.  Homologue-
 specific emissions data for the various test sites are contained in Appendix
 A.
 5.2.1  Flue Gas Concentration Data
     In this section, average CDD/CDF emission concentration data are
 presented for each Tier 4 test site.  Flue gas concentrations are adjusted to
 a reference oxygen level  (3% 02) to eliminate the effect of excess air
 dilution when comparing emissions from different test sites.  Controlled
 emission concentrations are presented in Section 5.2.1.1,  and uncontrolled
 emission concentrations are presented in Section 5.2.1.2.   Quality
 assurance/quality control  results for these data are discussed in Chapter 7.
     5.2.1.1  Outlet Flue Gas Emission Concentrations.   Table 5-2 lists the
 average outlet 2378-TCDD,  total  PCDD,  total  PCDF mass emission concentrations,
 and 2378-TCDD equivalent  emission rates for each Tier 4 test site.
 Estimations of 2378-TCDD  equivalents were derived using "2378-TCDD toxic
equivalency factors (TEFs)"  to compare the relative potency of one mixture of
                                      5-7

-------










CO

i—
CO
.
V)
Ul
l-
*«•
o:
UJ
p

U
CM


g z

toC
S L_
Pz

t£p


g§
Ig
CO O
uj a
_) UJ
§1
O CO
Q Ul

Q CO
1
*
^
1
IO
UJ
s
«"*













«
-M S
g§°:o
Slo § "u
co — in •v.
r» 3 in o>
m cr-^ *--
CMIU E
UJ


g
Q,
,_

o u> 4J
of o i2
^ ••- •-
O -4-* *l

23 t«
O. +> M O
4j a> o
O U 91 Q.

4* O il i~
O « +>
c -o o
o>5 "o *~
10 in c
u in —
ID —
•Si S
p

s
M*
cQ







ID
Q.
s>
u
ID
a
2

I
§.
>.
t-
o
u
I




CD
JJ
CO

K-



cz
o
ji+j

i— ** -SC
t»° c12





u
a
CO


8
t*+
O
U3






S

«
f— 4








CM
CM





S
*^
urnace
u.
IO
O

10
o.
a.
S
Secondary






£








rH





CM
,—4
O
o




vo
CO








S
r~









^
d








S
o *-
+j •a

|£
Tj ID
<= S
H-4 k
^ £
C !4—
o u)
4-> .
u "c •
ID O
"ID
2i
3 —






•
KH
Ct








in





vo CT» pn m ^ o i**
cn^HocMin^rfncM i
CMOOOCMOCJO I
CMOOOOOOO
C]
•s
Ul
f
4-*
inOfn^*«or*vo -^
vomp»cncMf»ooz c
•» ^ CM CM X
T 0
•g

in
ID
'—
>

r-*vOOf*»OAvOCMCO
CMCT^inmcMf-Hf-iozin
in •-* c
3
U U
• o
U) "* fl
5-2 S
"D • a.
p— C f-
i— 
L. (D >» 4J
ZZZZZ -l-> •»" U) 0) T3 O
i- %-»->
U C -C 3 13 G
O O) t. S- . "X O O >^ ^ "O
t. t- 4J IO L. 4J —
o o 10 e o >
4-»4->L.Ua_4->O-ll. >,-^ —
>aia
U L.BU.UJ L. UJ UJ ID UIC O«-
celjeuct.^^ 4J "~ o
**" ^" C O ID "^* Q) 9 O 10 t» M ^
c cM5^ g^^i ^,2 ^f

fflUaQ GQCDO Oi/l L.
 ffl uca>€(d
a}O)tucL.o)t.i_c 3 u a.
"O *O IOI9 O ^ O O3 "Or— Ul
3 3 Q O) 3 3 3 3 "x. •^•'O'OC
*•• i^~ -*4 »-i ce u. eo 1-4 co co i vvt-ffluiOr—
COCOCQCCjCOjIjCO ^StCUlO
cocoaoocomcas coooou
5) in o > •—
L. 10 L. to 4J
0 «- 3 Q ~ >.
D. ID o a >—
> JS O "0 «
10 ia i—  O) 1 +J <0 ID
IO O C CO U C
•O 4-» •— (— ID "O L.
1— "O IO CM (D ^ J3
10 C L. -0 fl L.
O 0 (D C > o 00*4.
10 t, vo m it u u
c o ^ 10
< o co en a ce en
10 J3 U Z Z <
5-8

-------
CDD's and CDF's with a different mixture of CDD's and CDF's.  The use of TEF's
permits an estimation of the carcinogenicity of the mixture of CDD and CDF
compounds relative to the known carcinogenicity of 2378-TCDD.  The TEF's used
in this analysis are presented in Table 5-3.  The test sites are ranked in the
table by average outlet total PCDD concentration.  Data are available for all
sites except Site WS-A (woodstove).  Surrogate recovery data for Site WS-A
were below acceptable limits developed in the Tier 4 analytical quality
assurance document, and therefore data for this site were not reported by
Troika.  Two sets of data are shown for Site WRI-A, one for conditions of wire
and transformer feed and one for conditions of wire-only feed.
     Table 5-2 shows that PCDD's and PCDF's were detected in the outlet
emissions from all test sites for which data were available.  Average total
PCDD outlet emission concentrations varied over four orders of magnitude,
ranging from 0.8 ng/dscm 0 3% 02 for Site BLB-A to 11,900 ng/dscm @ 3% 0- for
Site MET-A.  The average total PCDD outlet emission concentration from
secondary copper cupola furnace MET-A was more than a full order of magnitude
higher than that of the next highest test site, wire reclamation incinerator
WRI-A with wire and transformer feed.  In general, run-to-run variations for
each test site were less than the site-to-site variations shown in Table 5-2.
     The 2378-TCDD isomer was detected in the outlet emissions from 7 of the
12 sites for which data were available.  Detectable emission concentrations of
2378-TCDD ranged from 0.05 ng/dscm @ 3% 02 for Site DBR-A to 232 ng/dscm @ 3%
02 for Site MET-A.  Detection limits at sites for which 2378-TCDD were not
detected ranged from 0.01 ng/dscm 03% 02 for Site BLB-C to 0.2 ng/dscm (3 3%
02 for Site CRF-A.  In general, the contribution of the 2378-TCDD isomer to
the total  PCDD outlet emissions was small  but was found to be highly variable
between test sites.  The 2378-TCDD isomer ranged from 0.06 percent (Site
WRI-A, wire-only feed) to 1.9 percent (Site MET-A) of the total PCDD outlet
concentration at those sites for which 2378-TCDD was detected.
     Outlet total  PCDF emissions were typically of the same order of magnitude
as total PCDD outlet emissions, although in most cases the total  PCDF
concentrations exceeded the total  PCDD concentrations.  Average total PCDF
outlet emission concentrations varied over more than four orders of magnitude
                                      5-9

-------
           TABLE  5-3.  TOXIC  EQUIVALENCY  FACTORS  USED  IN  ESTIMATING

                            2378-TCDD  EQUIVALENTS3
Compound (s)
2378-TCDD
Other TCDDs*
Penta-CDDs
Hexa-CDDs
Hepta-CDDs
Octa-CDDs
2378-TCDF
Other TCDFs*
Penta-CDFs
Hexa-CDFs
Hepta-CDFs
Octa-CDFs
Toxic Equivalency Factor
1.0
0.01
0.5
0.04
0.001
0.000
0.1
0.001
0.1
0.01
0.001
0.000
*In situations where 2378-TCDD or TCDF were not chemically analyzed in the
 sample, total TCDDs and TCDFs will have a relative potency factor of 1.0 and
 0.1, respectively.
3
 Reference:  U.S. Environmental Protection Agency.  Interim Procedures for
 Estimating Risks Associated with Exposures to Mixtures of
 Chlorinated-Dibenzodioxins and -Dibenzofurans (CDD and CDF).  The Risk
 Assessment Forum, Washington, D.C.  EPA 625/3-87-012.
                                     5-10

-------
from 60,700 ng/dscm 9 3% 02 for Site MET-A to 0.6 ng/dscm @ 3% 02 for Site
BLB-A.
     5.2.1.2  Control Device  Inlet Flue Gas Concentrations.  Table 5-4  shows
the average control device inlet 2378-TCDD, total PCDD, and total PCDF  flue
gas concentrations for the eight Tier 4 test sites for which sampling was
performed at the air pollution control device inlet location.  Total PCDD and
total PCDF data were reported by Troika for each of these sites except
Site SSI-A.  Surrogate recovery efficiencies for the scrubber inlet samples at
Site SSI-A were below acceptable limits developed in the Tier 4 analytical
quality assurance document, and therefore inlet flue gas PCDD/PCDF data were
not reported by Troika for this site.  Speciation for the 2378-TCDD isomer was
performed by Troika for inlet samples from only four of the test sites  due to
time and budget constraints.  However, the Troika analyst indicated that the
2378-TCDD isomer was less than 25 percent of the measured total TCDD values.
     As shown in Table 5-4, PCDD's and PCDF's were detected in the control
device inlet flue gas from all of the seven test sites for which data were
available.  Inlet total PCDD flue gas concentrations from these sites ranged
from 1.8 ng/dscm @ 3% 02 for Site BLB-A to 687 ng/dscm @ 3% 02 for Site DBR-A.
In most cases, the inlet total PCDD concentrations shown in Table 5-4 exceed
the outlet total PCDD concentrations shown in Table 5-2, reflecting the
removal efficiencies of the various emission control devices.  However, Site
WFB-A is.a case in which the measured outlet total PCDD concentrations  were
actually greater than the inlet total PCDD concentrations.  These results will
be further addressed in Section 6.0.
     The 2378-TCDD isomer was detected in the control  device inlet flue gas
from two of the four sites for which data were available.  The average  inlet
2378-TCDD flue gas concentrations for Sites DBR-A and CRF-A were 16.3 ng/dscm
@ 3% 02 and 0.09 ng/dscm 
-------
CO


CO
J_
co
Ul
*"" '


cr.
Ul



£
f—


U.
CO —

2s
ii
zz
II
ZQ

—• o
CO O_
CO
51

c! (—
v. -j


cj £5
0- j_

J O
5 *i

*-<
>• CO
g£
_J (O
s *"*
g
8
tlJ
1

T
tn
Ul
1

£
°-

fl
Q "o

Q» *"
§in c

O M O
Q. •— IT O
•M Q.

ffl U r—
•— 4J S «
C! C? tl U
5 I fe o
o ci^1-
§)§ 0
2 ^ ^ Q
fl> Q
5 H-
1

a




ID
0.
8
>
£

o
u
•M
C
g
O
ID
a.

1-
1
o
CO

ID
CO

in
i£

§
O "^
p a
*c "a ®  en
-»-> ^ >s O 0 ^
(O t. U -M ec -P -^- c -^ -^" +j

rt3 4- CD CT t- C U CO I- L. c

^"f«r"»— QJCTf" W ^' (D
occnen'ooS'^en'1^^- >

c/)oocncnaaffi3: in tor^oooig^^cr)^ ^ o £- Q. Tj u a. o U -M 03 (O >4-» j^ 3 (O Q 0) L. U ® TJ O Q.*— ^ ui u) o u u it ii 5 or a a. O U C r- 0) "o u OL •a *n t. in >


-------
      Control  device inlet total  PCDF concentrations were typically of the same
 order of magnitude as the inlet  total  PCDD concentration,  although in most
 cases the total  PCDF concentration exceeded the total  PCDD concentration.
 Black liquor  boiler Site BLB-B was the only notable exception to this trend.
 The average control device inlet total PCDF flue gas concentrations for the
 seven sites for  which data were  available ranged from 1.1  ng/dscm @ 3% CL for
 Site BLB-B to 2,170 ng/dscm @ 3% 02 for Site DBR-A.
 5.2.2  Mass Emission Rate Data
      In this  section,  average CDD/CDF  emission  rate data are presented for
 each Tier 4 test site.   Discussion of  the data  is limited  to the magnitude of
 mass emission rates.   Relationships between 2378-TCDD,  total  PCDD,  and total
 PCDF mass emission rates for individual  test sites are  identical  to those
 noted in Section 5.2.1  since the mass  emission  rate  of  each  species is merely
 a multiple of the mass  emission  concentration.
      5-2.2.1   Outlet  PCDD/PCDF Emission  Rates.   Table 5-5  shows  the average
 outlet  2378-TCDD,  total  PCDD,  and total  PCDF mass emission rates  for each Tier
 4 test  site.   The test  sites are ranked  in  the  table by average  outlet total
 PCDD mass  emission  rates.   This  ranking  differs  from that  in  Tables  5-2 and
 5-4  because of the  difference  in  flue  gas flow  rates between  test sites.
      The mean  outlet total  PCDD  mass emission rates range from a high  of
 283,000  ug/hr  for Site MET-A to  a low  of 12  ug/hr for Site SSI-A.  This
 represents a between-site variation of more  than  four orders  of magnitude.
 2378-TCDD mass emission  rates  range from a high of 5,360 ug/hr for Site  MET-A
 to a  low of 0.1  ug/hr for Sites WRI-A  and SSI-A.  Detection limits for the
 2378-TCDD mass emission  rates  for  the  5  sites for which 2378-TCDD was  not
detected ranged  from 1 ug/hr for  Site  SSI-B  to 5  ug/hr  for Site BLB-B.   Total
 PCDF outlet mass  emission rates ranged from  1,420,000 ug/hr for Site MET-A to
29 ug/hr for Site CRF-A, or five orders  of magnitude.
     5-2.2.2  Control Device Inlet PCDD/PCDF Mass Flow  Rates.  Table 5-6 shows
the average control device  inlet 2378-TCDD, total PCDD,  and total PCDF mass
flow rates for the eight Tier 4 test sites for which sampling was performed at
the air pollution control device inlet location.  As discussed in Section
5.2.1.2, total PCDD and total PCDF analytical data were reported by Troika for
                                     5-13

-------







V)

»-*

UJ
r~


ce
UJ

i-^
UJ V>
ig
V) >-4
UJ CO
5">
«5 »•*
§ CO
u? <
in X
58
CO 0.
»- en
^ to
S 52
*s
Ul
"*
u>
1
to
UJ
I





„.
ui a.
•P *—
IT -P
O
G (-
O
In
ui a
— .- o
g L O
ui £ a.
in c f—
Ul 3 IO
10 «• 4^
1. O
. r-

la


< fp
00
S



wirce Type/Control Device Type
w
V)


•P
v>
•p
Ul

c
a In je
4J O W C
•2 °* H cS
•P r— UJ
O 4J Ul +J
Q tn 10

s §
o 'oo
0 ^H
CM

•
•"*

o at
O "-"

CA ^^

CM




co
o m
en

cS

icondary Copper Cupola Furnace/AB,
»lt-Laden Wood Fired Boller/BH
CO V>



I I
yj u.




--4 CM



O O
co r^
r» ui
\o 



< u
5 «




m *



o
m
m






£








3
£
o
z



S;
UJ
"•s.
U
r—
I
u
o
3
cr
u
CD



0
3




Ul



CM

m ^r
CM f-*








<-< CM
S D
i §



lack Liquor Boiler/ESP
lack Liquor Boiler/ESP
CD CD



CD <
4 4
CO 03




r- oo



3
CM






m
f-4








r^
O




Ire Reclamation Inclnerator/AB
Hire-only feed)
x~



i




<*


•» OJ
SCO
CM






in  O



f i
v» u




o *H



a
~*





00
?i








m
o




1
o
•p
IQ
L.
(D
C
H4
£
£-
 a:



S f
v> X




m ^




01
c
3 C


P "O
in ID
*"(0 1-
^-5 o
^ C P
ID O 10
U P
L. ••-
O 0) Q>

fl "u
10 ID
•*- in u
c. e Q-
%+J
O ^ 3 IO O
C U L. ID
««• f U Our—
in in  Ul ID
a e ^
" 2 «
S* 2
c in 
1-* U (O S t-
S 0 i- -S o
3: >*- i— &
+J ID
•a in ui e i-
c m  14- +J
^ " "o •§ P
in j-> a. <
i. L. o) •-
t. c "o a
IO O •»- 3) *- C
4-> U ^ +J 0) *O
5 C § S £ IDT,
X Q.4J 3 Ul —
^- o in a> .a 3 —
IO £ ID f t- O IO
o ui s- CD ^ >
,1 1. +J +J 01
4-> in o o <*• fl o
^« Q) O C IO ^3 C
^ 3 °
lO r— '
-------

CO


HH
CO

E
*~
^
ce
LU
t~4


Z -~
1— CO
LU
£ ee
CO Z
£2
< CO
ce co

^_ V
O UJ
1
D_ CO
CO
CO <
a
u. O
o a.



§°
H-
1- UJ
UJ — 1
1 f—
z g
LU >-
O CO
t—t

LU LU
^1
1— CO
Z LU
8 1
CO
LU — .
CD
£
LU
>
VO

uo
UJ
m













!














































U_
a
a
Q in
a CD
O •*-*
•p * 11
(0 3 g
Is ^- c
^
(D Ul
OT UL
£1

>

































§1
«§i
O -P Ul
O Ul
x
U-
8
Q.

. <0
,2

a
8
a.
"ia
•p
o
i-


§
?
ao





(D
a.

1-

 2
I- "OC*»— Ct*-^"«»-C^*'C+J C4*>
g^-g^-M 03W ° O 0-Ot-*  0) (O 9 t> O U
O2V) O) L. tot- &. t- <0 <4- O)C U 55 C
O "O O E -M O O £ t3®«-O3

t.-O^-(/>*- OC^-1^* O C (O Q£ -O CO >

« W M w> CO X ^ CO CO ^~ CO O Q W> 3e





<<co





i— t i-* ^H >H i— (










,
o
5

Q.
y
OJ
CL
U
CD itl
C -M
t.  ,
•o CD a. t.
— +-> E -0
r— CD 19
ia "a ui >i
O 0 O Q.
c c c in
II II II II
ce a co a
Z ZZ CO
5-15

-------
only seven of these sites because surrogate recovery efficiencies for Site
SSI-A were below acceptable limits.  Troika reported 2378-TCDD speciation for
only four of these sites due to time and budget constraints.
     The mean control device inlet total PCDD mass flow rates for the seven
sites for which data are available range from 3,380 ug/hr for Site BLB-B to
310 ug/hr for Site CRF-A.  The mean inlet 2378-TCDD mass flow rates for Sites
CRF-A and DBR-A were 1 ug/hr and 24 ug/hr, while 2378-TCDD was not detected in
the control device inlet flue gas from Sites BLB-A and BLB-B.  The mean inlet
total PCDF mass flow rates ranged from 283 ug/hr for Site BLB-A to 7,900 ug/hr
for Site SSI-C.
5.2.3  Scaled Mass Emissions Data
     Average outlet and inlet PCDD/PCDF scaled mass emissions (i.e., emissions
per unit of feed) are presented in Tables 5-7 and 5-8 for each Tier 4 test
site.  The scaled mass emissions were calculated by dividing the PCDD/PCDF
emission rates (ug/hr) by a representative mass feed rate to the combustion
device (e.g., kg/hr dry solids for the black liquor boilers).  Scaled mass
emissions for various source types are not directly comparable because they
are based on different feed materials.
     Tables 5-7 and 5-8 give the feed rate basis for each test site along with
the outlet and inlet scaled mass emissions for 2378-TCDD, total  PCDD,  and
total PCDF.  Homo!ogue-specifie scaled mass emissions are contained in the
individual test report for each site.  The test sites are grouped by source
category in Tables 5-7 and 5-8.

5.3  CDD/CDF PRECURSOR DATA
     This section summarizes the CDD/CDF precursor analyses performed  on feed
samples from the Tier 4 test sites.  Homologue-specific precursor analyses were
performed using gas chromatography and mass spectrometry (GC/MS).  Specific
precursors analyzed for using the GC/MS were chlorobenzenes, chlorinated
biphenyls, and chlorophenols.   Chapter 7 of this report includes a discussion
of the difficulties encountered during the precursor analyses, along with a
summary of the quality assurance/quality control  results.
                                     5-16

-------








V)
S



s r-
O C (O
en 3 •*•»
*-• o
•M t-
£
3
0


< GO

S


10
u
o

(J
IO
u.
O>
C

• 13
u

CD

•a
or
"§
,2
o
in
in
IO
m











ID

(A
I




r-4 CM
1 1 O
o o o
r— 1 *— * 1—
XXX

m m o
CM CO tr,





222
XXX

o oo 10
• • *
t-t ^ m


^^

1 1 1
o o o
^-t ^4 t—t

XXX

P4 ^ ^>
pg m o*

z







(D 
o a o



in
i_
O


c.
CD i/l
C t.
— a

c •*•
i- O
CD a
C t.
-o a
-- < —i r**

t~- Oi -<
z


V CM
0 t

X X

ai m
o? a: -<








5
— *-* a
4-> -M or
CD CD
x x ~a
— — a)
CD
CD CD U.

10 
a
< < CO
i < i or
U- co en in
+•
X


-1- 1 O
0 O O

XXX

pr> m OO
in «3 ^





Vsl
XXX

^ o ^

IO CO i-l




o o o

XXX

m co p*
-H CM fn







CD
10 CD CD
or •*->•*->
<0 10
•a or or
CD
CD "O -O
CO CD
aiu. u.
c
t- in in
t- -o •o
en "o "o
i to to
CD i— ^
a. > u CD
r™ +* U
e u
O «4 3
0
CD CD c/1
U U
••- •»• in
XX 3
-- i
« <: c
u

X


'o'o

X X

t~i r*>
•-< CM




^ CM
o'o
i-H r-l
X X

o *->

CM  e i- o
10 o »-
u •»- »
 in i c
n» t- or o
— e m —
CD a in
•o in
CD in CD —
•p in -p s
o fl f S
CD E to
+j in
CD "O &- in
•0 CD C( §
* S*. T,
c in ex CD
ii <— 8 13
•— x u
§< CD to
IO ^
5-17

-------


1

£ i

CO

is
,

cc '
Ul
*~*



R •
g
u.
1
u.

CO
CO
^ i
Q
c
Q

t£
ou
o I
Ul >
3 '
%
UJ


s

i
J

!••
s
"

S
g
Ul
«c

CO
1
v\
Ul
f-


3K '*i

i— S

Ul •—
*O 4J
1 £

5:



^1
4«* ^s. ^

*-" ~ 4J
C "— O

I
IO
u
as
o
10 H-
1 £
< ro
CM

IO
L.
O

f9
U.
C)
c

IS
o
C/J
®
OS a.

•o
8
Lu
**.
o
Ul
Ul

CD







(D
S
CO
S



o
0

X
*o
or co m
z z


t— i

o
*•"«
X

CM

ce co co





izi








553
ce ce ce
O 09 G)
Q  t/) cn


qj qj Q -p -P
u. u. u. 25
ce ce
Ul Ul Ul
•»-•»-— a> ,>.>< 00
U U U O O
a a o xx









c
o

Ul
< CQ O -C <
III E 1 <
CD CO CO O CO 1
_l -J.J O U. CO
CO CD CO XX
a
a






CO CO CO CO
z z z z










CO CO CO CO
z zz z





CO CO CO CO
z z z z





5
0)  u_ o> ai
u. cnu. u.
c
Ul v Ul Ul
•a t_ -o -u

^— 01 ^ i—
O CO O O
CO 1 CO CO
L.
i— . t- 09
•— -P U
C !-
0-> §
>t ID 
6Q- ia

—
•Sai c.11
A 01 <
U| CXZ
-»., C "O •»
05 
** § °" u
*C 01 -M
C II i; C
•^ T> "O CO
a> -M
T3 VI «*• 
O (fl O *O
 o
U) L. L. *-
4) CL (D "O
t- x .a u
a. CD g i-
X 3 -P
09 Ul C >>
IO •—
Ul X CD iO
o ^ c
•p ID -pa
2 2 § "
•o ce
StJ "O z
ID 09
i^, q) Ul •»
t- a •o
O) -Q<0
55 .-5.
Ul +J I- E
3 lO IO
-C Ul
•o o <
09 — 1 -P
•p js ce o
* » co c
"3 t. ui
U O 09 C
•- H- +» ^
o • co ui
< Ul
Ul 1 t. —
u ce o g
Q CO **• 0}

O Ul "O
10 09 L. 49
u. +J O ^~
•^ *> r—
e co u o
3 u 5 .p
ui o e
ui O A Z
5-18

-------
      Total  organic halogen (TOX)  analyses  were performed on selected samples
 using short column gas  chromatography with a Hall  detector.  The TOX procedure
 is  incapable of speciation and  gives  a positive response to any organic
 halide.   The TOX procedure was  applied primarily to  sample  media for which  the
 GC/MS analysis  was not  completely successful.   Samples  analyzed using the TOX
 procedure included the  cardboard/paper/wood samples  and the wood/plastic  parts
 samples  from Site ISW-A,  the  electrical  switching  gear  sample and circuit
 board samples from Site MET-A,  the wire  coating/wire insulation samples and
 transformer core samples  from Site WRI-A,  the  drum coating  samples from Site
 DBR-A, and  the  spent  carbon samples from Site  CRF-A.  Additional  TOX analyses
 were  carried out on-at  least  one  sample  of each medium  type.   These additional
 TOX samples included  fuel  oil samples  from Sites SSI-A,  ISW-A,  and BLB-B; a
 sewage sludge sample  from Site  SSI-C;  a  strong black liquor sample from Site
 BLB-B; and  a wood sample  from Site WFB-A.
      Table  5-9  summarizes  the GC/MS and  TOX precursor results.   The data  show
 a wide range of precursor contents across  the  various source  categories.
 Precursor data  within source  categories  are generally consistent.   The  data
 for each  source category  are  discussed below.
 5.3.1  Sewage Sludge  Incinerator  Test  Sites
      Sewage sludge  samples  from the sewage  sludge  incinerator  test sites
 SSI-A, SSI-B, and SSI-C were  found to  contain  small  quantities  of
 chlorobenzenes.   The total  chlorobenzene content of  individual  sewage sludge
 samples ranged  from less  than the  detectable level  to about 40  ppb.
 Dichlorobenzene  was the only  chlorobenzene  detected.  Site SSI-B  sludge
 samples appear  to have contained more chlorobenzenes  than Site  SSI-A or SSI-C
 samples.  Detectable quantities of chlorobenzenes were found in each of the
 three sludge  samples analyzed from Site SSI-B, whereas chlorobenzenes were  not
 detected  in  at  least one sludge sample from each of Sites SSI-A and SSI-C.
 Chlorinated  biphenyls and chlorophenols were not detected in the sludge
 samples from  any  of the three sewage sludge incinerator sites.
     A sludge sample from Site SSI-C and a fuel oil sample from Site SSI-A
were analyzed using the TOX procedure.  These samples were found to contain
less than the detectable level of total TOX which in this analysis was 10  ppm.
                                     5-19

-------












1
t/7

1
0
K
1
i
u.


X
o
o
•
T
in
Ul
?








L. •»-
o> in
e o >,
u
o-->
CD
"io en
4J 3

t-
o
+i X

£ I/)
•§£•
r-B £
r~ ^
O «
H^
13 3
o
»-
/O
X





in
in
r"
IO
<
ursor
^
a.-
2

i



c
o
**?
2
u.
in
•o
^


e
0
«
2
in
ra

I

in
m
in
*^
11
25
O
i
^^

1
CO
*^
o
a> o
o +•
8^
I

§inOO OOOO O tf» C*
m 5 o o co in r~- co CM ^
CM m 01 CM m ft oo —




Si i i s i zi i Si



38.-
o « es
is i i T 7 Ts i ii
O Q Q




§s § i i i is i ii




1-4 o m
0 • 0
• o •
Ts • T i i is i ii
i '°. i
o
U (. L.
•0 -0 T3 O 0 O
ID O 
^c: "*• "- I] I] U— S
q> o ffl m O i*-
O OiCD 'o'tJOr— "O ®
's'fflSS flflflO OJCC
r— 3 ?•" i— »^f-»r—3 q j? T*
toi^tnco mmmu. zwcu
c
_a
oiui ^ -Jj
30< m03ui-4>r~a>mmcQ irPf!%
UCO COC/J-l— -J.J-J ,St»
aa)cn coco f-mmm uxac
«3 ijcr ^
5e £ |
CO M ™ ^


















ID
e
£
0















5-20

-------













^^
•o

3
C
C
o
0
^



C/J
^
1™

Q
a:
S
cc
§

OL

Z
Q£
=
z

X
o
o

M
1
LLJ

<
*~


















fa

c 01 'ui
o o ><
M *> t—
i <

o —
01
v
IB

S

3
O
2
a.

^

€3






c
o
u
IB
l_
u.
Ul
•o
"^
^

Ul
"o
§ "oi

"^L C3)
S-5
0
*^
o




c

u

la-

in
2

3
a
1
IB
CO


•o
c e 01
u jr o>
o a. 3

o ^


ID
e
CO
N -^
.a 01

O

o


1

i-



co


^
§"5

IB tO

IB Ul
O ®
(. 1—
3
O



li sis is Sz
o>


CN

IB CN -d
i~ CM m
cs z z i z z "2

^ Q *»
z



m
rH l«- fO O
^^ " ^^ -*
o - °,i §§ li
i i i
0
Sp« Q 0
• Z IB
in t.




 01
3 i i_ e
^» 0} ffl 
Q. IB IB « 3 I. Ul
f— et 4J U O Ul IB
.a. c •^o+'O'-
-o -5,^ oicu25«
t_ *O ^J (D IB r^* tv IB Q (J
IBOC UIQ.— 3i— O.+J —
OO"B IB O r-i — «^t-
J3X U1.OX ^— O 0) 3 -P
S -0 -P 1 8^- fl OC 0 01-
(-•O O U i— -P 
2 — «
ui ai a
3 3 < U <

•° C ' £C |i
O C »-H Ul X
O i—
•0 CJ IB
O •" -P
o 
-------













1
5
i

|

V)

^f

g

o
P
1
_
^q

£
z
X
o
s
*
01
in
Ul


*-


















Q.
10 U)

cn in
IHl3
it .£
t.

O *•»
0)
13 oj
4-> 3
O "**
H*
o
1.2
4J in
CO >t

«
13
x
l— "o>


lO 3
2~







IA

in
•^
^
t-
0
in
L.
§
u
a*

*»•*
8



§
4ji
2
u.
IA
•o

"o

U)
*o
§«-»
o>
11
u —
o

f~
CJ



e
o
IO
ll

in
y .
4->
I
1
in
ig
CD

CD IA
•H F—
cl c"c>
"u.2'o>
o Q. a

t^


ID
N *"»
J3 O)
0 3
U *-•
O
^
o



l
t—

^
o
en o
CO 4J
o
co in
U CO
u h-
§
01




< o o o o
2 r» o in o
CM r» ir
•v >o

^
rH CS TT
o I o  +A
0} C£ *O S F— E C
uta 10 a ml 30 =i a>
^•e u o ID uin t. Q.

i-
3
O
>> VI
U
CO «^ U)
> -a a
o co o
U = < CO < <
CO C 1 C 1 1
as — HH «3 a: u.
4J Q; r- ff3 Q£
IA C 3E •- Q O


4J — f
X I











.

".1
in *^ +J
"a.
10 .a 10
in »
•o ^- co
o -o r-.—
O C 3 —
X ra u- *-
in o
•o in in u
c ID ID a.
« r- O
to u
3 —
u a in j=
o in ix
a. -a ia
a -a CD i-
a. u NO)
10 >> O
• O f— -M
•o ^3 a ia
So a §
o »- t-
J3 3 ID JC
•O U -Q U
t_ U .
S — vac
u o o
c *o .^
ID CO C V
j: j= -o n o
4-> 4J i— CO
3 C 4->
i^ t^ O CO ID
O O U o)-o
O
••- • i- "Q.X s
in in in • s +J
o co o in 10 o a.
a.r— a. co in •» a
§0. g i— co
E 8 a. in «
« o € *- en c
in ia ^: u o
a id in v o u
in 4->
c 4J e i. <^ i— o
o u o id o id co
Id CD 4-> r—
•a Q.-O S) c o CD
1° Is1*!'!?
sttsim
i. Id U O I- 3
CO r- CO 4-> <•- -0
Ou ^ CL*^ CO >t
• in N .e

Q) ID Id C (O Id ^IB Id
U^'-a *"«4JC en
JJ 'fl — O — ^ (D -P
*5j?3£$!S|

ec§5S'35i2'o


a^pSSSviSwcj)
Z Z iQ -a O
5-22

-------
5.3.2  Black Liquor Boiler Test Sites
     Strong black liquor samples from test sites BLB-A, BLB-B, and BLB-C were
found to contain small quantities of chlorophenols.  Pentachlorophenol was the
only chlorophenol homologue detected.  The total chlorophenol content of
individual black liquor samples ranged from less than the detectable level to
about 14 ppb.  Chlorophenols were not detected in at least one black liquor
sample from each of the black liquor boiler test sites.  Chlorobenzenes and
chlorinated biphenyls were not detected in the strong black liquor samples
from any of the sites.
     A black liquor sample and a fuel oil  sample from site BLB-B were analyzed
using the TOX procedure.  These samples were found to contain less than the
detectable level of total TOX (i.e., <10 ppm).
5.3.3  Wood Combustion Test Sites
     Wood samples from wood-fired boiler WFB-A and woodstove WS-A were found
to contain less than detectable quantities of all precursors analyzed for
(i.e., Chlorobenzenes, chlorinated biphenyls, and chlorophenols).  A wood
sample from Site WFB-A analyzed using the  TOX procedure was found to contain
less than 10 ppm of total TOX (i.e., <10 ppm).
     Feed materials from industrial  solid  waste incinerator site ISW-A, which
is included in the wood combustion category, were found to contain small
quantities of chlorinated biphenyls  and small to large quantities of chloro-
phenols.  Chlorobenzenes were not detected in any of the feed materials from
this site.  Chlorinated biphenyls were detected in the wood/plastic parts
samples and in the latex paint sludge samples..  The total  chlorinated biphenyl
content of individual samples of these materials ranged from non-detectable to
50 ppb.  Chlorophenols were the most prevalent precursor species present in
the feed materials from site ISW-A.   Small quantities of chlorophenols
(ranging from non-detectable to about 100  ppb) were found in the
cardboard/paper/wood samples, the oil-based paint sludge samples, and the
latex paint sludge samples.  Large quantities of tetra- and penta-
chlorophenols were found in the wood/plastic parts samples.  The total
chlorophenol content of individual wood/plastic parts samples ranged from
5,700 to 19,700 ppb.
                                     5-23

-------
     A composite of the cardboard/paper/wood  samples  and  wood/plastic  parts
 samples from Site ISW-A was  analyzed  using  the  TOX  procedure.   The  composite
 sample was  found to contain  about  13  ppm total  TOX.   The  fuel  oil samples  from
 each test run at site  ISW-A  were also analyzed  using  the  TOX procedure.  One
 of  the samples (Run 02)  was  found  to  contain  about  24 ppm total  TOX, while the
 other  two samples contained  less than detectable  levels of total TOX.  The
 detection limit for the TOX  analysis  was 10 ppm.
 5.3.4   Metals Recovery Test  Sites
     Samples of the plastic-bearing feed materials  from Site MET-A  were  found
 to  contain  small  quantities  of  chlorinated  biphenyls.  The total chlorinated
 biphenyl content of individual  samples ranged from  less than the detectable
 level  for the metallurgical  coke and  circuit  board  samples to  260 ppb  for  the
 electrical  switching gear  samples.  Chlorobenzenes  and chlorophenols were  not
 detected in  any of the samples  from Site MET-A.
     A composite of" the  circuit board sample  and  the  electrical  switching  gear
 samples from Site MET-A  was  analyzed  using  the TOX  procedure.  The  circuit
 board/electrical  switching gear composite sample  was  found to  contain
 approximately 4,300  ppm  total TOX, which was the  highest TOX content of  any
 sample  analyzed  in  the Tier  4 program.   Thus, although the specific precursors
 analyzed for (i.e.,  chlorobenzenes, chlorinated biphenyls,  and chlorophenols)
were found in  only  small quantities,  there were significant quantities of
 halogenated  species  present  in the feed material  from site MET-A.  This
 suggests that  either 1)  the  specific  precursors analyzed for were present  in
the samples  but were not easily detected using the 6C/MS procedure due to the
complexity of the sample matrix, or 2) significant quantities of halogenated
species other  than the specific precursors analyzed for were present in the
samples.  One  potential  source of these  "other" halogenated species is
polyvinyl chloride.
     Feed materials from wire reclamation incinerator WRI-A included wire
coating/insulation and transformer core parts (cardboard,  paper, and wood
components of the transformer cores).   Small quantities of chlorophenols
 (i.e.,  less than 200 ppb) were found  in both materials.  Chlorobenzenes and
chlorinated biphenyls were not detected in the wire samples, but these species
                                     5-24

-------
could not be successfully analyzed for in the transformer samples using the
GC/MS procedure due to the complexity of the sample matrix.  Based on
conversations with plant personnel at Site WRI-A, it is believed that
chlorinated biphenyls were present in the transformer samples, but the amount
is unknown.
     The total TOX contents of the wire and transformer samples from Site
WRI-A were approximately 200 ppm and 20 ppm, respectively.  Again, this
suggests that either 1) the specific precursors analyzed for were present in
the samples but were not easily detected using the GC/MS procedure due to the
complexity of the sample matrix, or 2) significant quantities of halogenated
species other than the specific "precursors analyzed for were present in the
samples.  One potential source of these "other" halogenated species is
polyvinyl chloride.
5.3.5  Other Test Sites
     Combustible materials in the feed to drum and barrel reclamation
incinerator DBR-A included solvent residue contained in the drums and the
coating on the surface of the drums.  Chlorobenzenes were detected in both
materials, but chlorinated biphenyls and chlorophenols were not detected in
either material.  The total chlorobenzene content of solvent residue samples
from individual test runs ranged from below the detectable level (Run 03) to
nearly 70,000 ppb (Run 01).  This reflects the random nature of drum types
processed in the unit.  Trichlorobenzenes were the predominant species found
in the solvent residue samples, with small quantities of diChlorobenzenes also
detected.
     The total chlorobenzene content of the drum coating was approximately
220 ppb, with triChlorobenzenes being the only species detected.
     A solvent residue sample from Site DBR-A analyzed using the TOX procedure
was found to contain about 800 ppm total  TOX,  which was the second highest
total  TOX content found in the Tier 4 program.
     Spent carbon samples from carbon regeneration furnace CRF-A were found to
contain variable quantities of Chlorobenzenes,  but chlorinated biphenyls and
chlorophenols were not detected.   The total  chlorobenzene content of
individual  spent carbon samples ranged from 160 ppb (Run 02) to 6,630 ppb
                                     5-25

-------
 (Run  03).  Trichlorobenzenes were the predominant  species present, with small
 quantities of di-  and tetra-chlorobenzenes also detected.  A spent carbon
 sample  analyzed  using the TOX procedure was found  to contain about 150 ppm
 total TOX.

 5.4   HC1 SAMPLING  DATA
      This section  summarizes the chloride emissions data developed using the
 HC1 train.  The  data are reported as "front-half", "back half", and "train
 total"  chloride  emissions.  The front-half emissions represent chlorides
 captured in the  probe rinse/filter fraction of the HC1 train, which may
 include metal chlorides contained in the particulate matter (e.g., NaCl) and
 the back-half emissions include chlorides captured in the impingers,  which
 would include HC1  plus any metal chlorides that pass through the sample train
 filter.  The train-total emissions represent the sum of the front-half and
 back-half emissions.
     The data in Table 5-10 show a wide range of chloride emissions for the
 various test sites.  The mean train-total chloride concentrations range from a
 high of nearly 10,800 mg/dscm @ 3% 02 for industrial  solid waste incinerator
 ISW-A to a low of  less than 5 mg/dscm @ 3% 02 for carbon regeneration furnace
 CRF-A and black  liquor boiler BLB-B.  Other sites with relatively high
 chloride emissions included wire reclamation incinerator WRI-A, which had
 chloride emissions of approximately 1,000 mg/dscm @ 3% 02.  Chloride emissions
 from the black liquor boiler source category showed considerable variation
 between test sites.  Total chloride emissions from Site BLB-B were 2.9 mg/dscm
 @ Z% 02, considerably lower than those from sites BLB-A and BLB-C, which were
 112 and 185 mg/dscm @ 3% 02, respectively.  Site BLB-B is a direct contact
 black liquor boiler, while Sites BLB-A and BLB-C are low-odor black liquor
 boilers.  The direct flue gas/weak liquor contact at Site BLB-B apparently
 acted as a scrubber in removing chloride species from the flue gas.
     The distribution of total  chloride emissions between the front-half and
 back-half of the sample train varied between test sites.   The total chloride
was predominantly distributed in the back-half of the sample train for Sites
 BLB-A, BLB-C, ISW-A, and DBR-A.   Greater than 90 percent of the total
                                     5-26

-------
                 TABLE 5-10.  HC1 TRAIN  EMISSIONS DATA SUMMARY
Source Category/ Mean Chloride Concentration (ma/dscm @ 3% 0,,)
Test Site
Sewage Sludge
Incinerators
SSI-A
SSI-B
SSI-C
Black Liquor Boilers
BLB-A
BLB-B
BLB-Ca
Wood Combustion
WFB-A
WS-A
ISW-A
Metals Recovery
MET-A
WRI-A (wire only)
WRI-A (wire and transformers)
Miscellaneous Sources
DBR-A
CRF-A
Front-Half

NS
NS
NS

1.5
2.2
12.5

39.5
NS
168

33.6
438
249

0.39
1.7
Back-Half

NS
NS
NS

110
0.7
173

155
NS
10,600

26.1
385
1,050

92.7
1.7
Total L

NS
NS
NS

112
2.9
185

195
NS
10,800

59.7
820
1,300

93
3.4
M** •* •* i«<«1 •*«**» *K IK .»••_. ^ •. .^oj^_mn^*__^.._t^ _i_«__ f ** A « . • A ,*v -•
NS = not sampled
                                     5-27

-------
chlorides was found  in the  back-half of the sample train for these  sites.
This  indicates the potential  for  a  high proportion of HC1 emissions.  The
total chloride was more  evenly distributed between the  front-  and back-halves
of the sample trains for Sites BLB-B, WFB-A, MET-A, WRI-A,  and CRF-A.   For
these test sites  a significant portion of the total chloride emissions  may
have  been associated with less reactive metal chloride  species, rather  than
the more reactive HC1.

5.5   CONTINUOUS EMISSIONS MONITORING DATA
      This section summarizes  the  continuous monitoring  data obtained for 02,
CO, THC, S02, NOX, and COg.   The  data are summarized by giving the  mean and
standard deviation of the individual 5-minute concentration values  obtained
during the test runs.  The  mean concentrations and standard deviations  of the
six species monitored for at  the  various test sites are shown  in Table  5-11.
Graphic displays  and tabular  summaries of the 5-minute  values  are contained in
the site-specific test reports.   Concentrations of CO,  THC, S02, NO , and C02
have  been corrected  to a reference  oxygen level (3% 02) to  eliminate the
effect of excess  air dilution when  comparing data between various test  sites.
      Mean oxygen  concentrations measured at the five test sites for which
monitoring was performed at the control device inlet location  ranged from
6.0%  02 for Site  BLB-A to 13.2% 02  for Site SSI-C.  The range  for control
device outlet locations  was 5.7%  02 for Site WRI-A (wire-only  runs) to  20.2%
02 for Site MET-A.   Thus, there were significant variations in excess
combustion air and/or flue gas dilution air between test sites.  Normalization
of the CO, THC, S02,  and NOX concentrations to 3% 02 minimizes the  effects of
excess air dilution  and  allows for  a fair comparison of data between test
sites.
     Mean CO concentrations.varied over several orders of magnitude.  Sites
CRF-A, BLB-A, DBR-A, WFB-A,  and ISW-A had the lowest mean CO concentrations,
with mean values  ranging from about 100 to 400 ppmv CO @ 3% 02.  Sites WS-A,
MET-A, and BLB-B  had the highest mean CO concentrations, with mean values
exceeding 10,000  ppmv CO @ 3% 02.  Several  of the test sites showed
significant CO variations within test runs and/or between test runs.
                                     5-28

-------
TABLE 5-11.  MEAN VALUES AND STANDARD DEVIATIONS OF THE
          CONTINUOUSLY MONITORED STACK GASES4
Monitoring 0.
Test Site Location % VOl
CO
ppmv
@3%02
THC
ppmv
@3%0
?
soz
ppfflv
@3%0.
N°X
.ppmv
?3%02
CO-
% v6
@3%0
1
i
Sewage Sludge Incinerators
SSI -A

SSI-8

SSI-C

Black
BLB-A

BLB-B

8L8-C

Scrubber Inlet

Scrubber Inlet

Scrubber Inlet

Liquor Boilers
ESP Inlet

ESP Inlet

ESP Outlet

11
(1
13
(0
13
(1

6
(0
12
(0
12
(1
.9
.7)
.1
•5)
.2
•1)

.0
•3)
.2
.8)
.2
.1)
2,473
(860)
4,677
(750)
3,113
(723)

136
(98)
10,900
(2,400)
862
(360)
152.
(100)
22.
(9.
NS


3.
(4.
57.
(39)
23.
(11)
3

7
2)



5
1)
7

0

1,587
(560)
NSb

504
(72.2)

94
(77)
830
(220)
80
(41)
465.3
(210)
788.0
(65)
419
(54.8)

83.2
(5.4)
42.6
(12)
40.9
(2.5)
29.
(8.
18.
(2.
13.
(1.

15.
(1.
17.
(1.
17.
(0-
5
9)
9
8)
1
1)

6
2)
4
5)
0
6)
Wood Combustion
WFB-A

WS-A

ISW-A

Metals
MET-A

WRI-A
(wire


Baghouse Outlet

Woodstove Outlet

Incinerator
Outlet
Recovery
Baghouse Outlet

Afterburner
only) Outlet


13
(0
17
(0
11
(1

20
(0
s
(2


.0
.9)
.0
•4)
.4
.S)

.2
• i)
.7
.9)


273
(120)
36,900
(6,000)
370
(140)

30,800
(23,500)
4,657
(3,800)
(Continued)
5-29
1.
(1.
9
9)
49,500
NS

NS
0.3
(0.3)
NS
(7,700)
7.
(2.

371
(284)
276.
(260)


2
4)



5



NS


7,290
(1,760)
NS



128.3
(27)

940
(230)
89.2
(40)


17.
(0.
15.
(1.
9
6)
3
7)
15.9
(1.6)

43.

-8
(5.0)
12
(1


.9
•4)



-------
        TABLE 5-11.   MEAN  VALUES AND  STANDARD  DEVIATIONS OF THE
            CONTINUOUSLY MONITORED  STACK GASES9  (Continued)
Test Site
Monitoring
Location
WRI-A Afterburner
(wire & Outlet
transformers)
°2
X V61
7.4
(2.7)
CO
ppmv
5,605
(3,500)
THC
ppmv
877.5
(1,100)
so2
ppmv
NS
ppfftv
126.1
(50)
CO,
% V6l
14.1
(2.5)
Miscellaneous Sources
DBR-A
CRF-A
Afterburner
Outlet
Baghouse Outlet
13.5
(1.1)
13.9
(0.3)
234
(163)
89
(120)
5.6
(11.3)
3.0
(2.9)
18.9
(18.1)
NS
132
(32.2)
NS
11.8
(0.7)
13.2
(0.4)
 Mean value shown on top, with  standard deviation  shown  below in  parenthesis.
b>
 NS - not sampled.
                                   5-30

-------
     Mean THC emissions also showed significant variability between test
sites.  With the exception of Site WS-A, mean THC concentrations ranged from
less than 10 ppmv @ 3% 02 for Sites WFB-A, CRF-A, BLB-Ai DBR-A, and ISW-A to
greater than 100 ppmv @ 3% QZ for Sites WRI-A, MET-A, SSI-A, and BLB-B.  The
THC concentration for Site WS-A (49,500 ppmv @ 3% 02) was more than.50 times
greater than that of the next highest site (WRI-A/wire and transformers,
880 ppmv @ 3% 02).  In general, flue gas concentrations of THC and CO tended
to be positively correlated.  Sites with low mean CO concentrations tended to
have low mean THC concentrations, and sites with high mean CO concentrations
tended to have high mean THC concentrations as would be expected from
fundamental combustion theory.
     Mean concentrations of S02 and NOX also showed significant variations
between test sites.  The S02 concentrations ranged from about 20 ppmv @ 3% 0?
for Site DBR-A to 4,440 ppmv (I 3% 02 for Site MET-A.  The NOX concentrations
ranged from less than 1 ppmv @ 3% 02 for Site WFB-A to about 800 ppmv @ 3% 02
for Sites MET-A and SSI-B.

5.6  PROCESS DATA
     This section summarizes the process data obtained at each test site.
Combustion device operating data are presented in Section 5.6.1, and control
device.operating data are presented in Section 5.6.2.
5.6.1  Combustion Device Operating Data
     Table 5-12 summarizes the combustion device operating data obtained at
each test site.  Key operating parameters included process rate data,  feed
characterization data, process temperature data, and excess combustion air
data.  These data are discussed in detail in the individual site-specific test
reports.
5.6.2  Emissions Control  Device Operating Data
     Tables 5-13, 5-14,  and 5-15 summarize the emissions control device
operating data obtained at each test site.   These data are discussed in detail
in the individual site-specific test reports.
                                     5-31


-------























^.^
•O
C
J
U
*""*

1
1-
UJ
r—

i
i
S
§
en
UJ
S
1
1 DEVICE OPERATING PAR
2
r—
V)


Q

z
CM
i— t
1
in

i
r—












CM
O
«

C
o
0)
5*
0





L.
SO
.e

«

o
i
H
Ul
w §

2 "x

1

«
Ul

0)
&!
— o»
"x i.
' 3 *-
< u.
s
u_

Feed Characteristics

0
•M
S

T)
0}
O
U.
Ul
tA
0
2
CL


Q.
Combustion Device

•^
t
O) O
0 -M
•M •*•
5"
O Ul
F







<0 Ql
-M C
Ul ••-
O
CM 0>
• O
O £.
CM -O





1
1




C
o 2

C:*^
00 O
a.
3
~



t.
01
O 0
mO

i
U O Ul (A Ul
o a. +j -M —
& a. u i. ui
 > J3
20 (D » 0 a)
•_ C M Ul b. ^ (D
-M ^ in ui ^ 'ui a>
OQ*aa. JC
M "M O I. MO
^"oi^o,^.,;0
B •» c *• e ** v e
CM O 0) O 0) •^r*


t.

5**2

O £
m ja





i
u
3
U_
"o
a.


£
1
1 T
a H-
Ul ^
IS

u
0)
c
u


ID

*» 2

r->
• 3
in o




>* u
O t- 3)
0"^ « J3

1/1 a. "5



c


0 « ^

3§


U)
a
at
u
3
C
>*o
m

12.5 wt.X combustible
materials: 100X from
wire


^•o

^ »2 «2
Ol
""3^
m ** +*





545 kg/hr Mire
United Model G-466
Wire Reclanatlon
Incinerator (1978)

^^
>»
§
f 2
gi



>D
C
t_
3

01

*S 2

r»
^ ^
r- o




>t U
JP u *
un "^ «
1/1 Q."5



ol


0 ffl 2
,-«.(•»+*
>*• 3
<9 O


Ul

O)
3
* u
°o 1 s
CO ••- fl
^ OL-g



O
CJ 3 — .
o .a +*

CM O r-

2 o

Ul


*
3
C
1
ui
2*

Drums containing
lacquer & organic
solvents. Inks,
enamel-type paint
and orange juice

Ul
£ ^
*x. C
Ul "^

£8
si
Ul



•d
4 MM Btu/hr (Input:
ECO Model 100 drum
barrel reclamation
Incinerator (1974)




T
S

t_
03

3

t_


«*.

(0

<*"! 3
^ O
•o
o> >*
St- d
ID

C 3 — t_
u +* wi 
U 0) 0) O •*• ^

**- 3 (0 L. U ^~
«3 O +^ « O. OL




o"3 5
m t-
in 

(A -M
e u

1o "x
U (9
•p
c c
8 *
C 01
85
e a

x1 —

o-
Ul >t
o


S5«
l°s
5-2-°
°1|-
4J Ul
Ul •*»
COO
— O)
m m
from plant-maintained
the other source cat*
active-type system an
u t.
•o o -*-»
* •
5-33

-------






j3
o
i™
UJ
g
1
MJ
g
^^
_J
CO
£
CO
g
u.

CO
cc
ui
•C
5?
c2
•if!
0.

2
H><

§
£5

s

i

=1
p
§s
Q
[,_
C3
._
§
1


*
CO
1
in

UJ

j|












































































1  3
0. U

1_
3
10
L.
0)
Q
1
l_
*
3
t_


U
XI
3
U
U
CO
in
in
o o
u u.
3 U
u><
in
o a
I- O
a. u
a

 CD
i "oK
CD *- •_»

UJ


+•
i S



CD O
f— O
C «-»
1— 1
in
10 .—
m e
•• O O
•O -^ in
*^* *P t3
3 10 >^
O* ^^^^
_J +*


f^
O Q) ^™*
I- U IO
c ">^

II5


CO
•4J 
3 O
*o
a.




^^
^^
u
0
4J 

z it IO * • < vo CO Q.I-ICO •P 10 L. 1 "o C g •o 3 < CD O CO HH 1-1 1—1 CO CO CO 0) CO CO CO o IO CD col L. 0 *o 8 X) 3 in e -P IO CD C r— IO •»- X) r— L. IO ^> *^ §C iO 0 > > 10 <*- L. -P us O o +» 4- e 10 •o -o 10 0 -P e c 10 o>->- o t- XI


-------





CO

LU

§
ce
o
^^

i— i


M
m
1
CO
cr
UJ
^H
LU
I
g
0.

o
J5
1-
CC
LU
85
LU
U
i
I
§
"
0

^c
J»J
3
CO

a
^
1
LO
LU
«J
CD
H*







(j
^B
1^_
Q. ••-
CO CJ
LU CD
a
co


CD
L.
3
-P
ie
u
CD
Q
1-
85
UJ

CD
•^
a
C f—
0) 3
*- O
in i-
CD -P
O (.

a.


in
ie
a
CD
LU

























C
o -^

^j Q} £
U I- U
^aj < _«
.1— CM
O E
O —

a

u
I— C
ia CD

CD £ ~
Q£ ^^
LU

3 CD
I- U
5 ">
*
g,
CD CD
•P -P
O CO
%X
<- CD
3 ^~
O
CO



O (T)
CM -*
O O «C





LO . r- \o
r~- m «r






oo ^t >o
r» >o in
>-l >-H l-l





<"*i ^Q
Ol O^ ^






in o o
O CM ^




85 8; 8;
UJ LU LU
§ § §
•P +> +J
•P -P -P
o o o
co m m
•P >. >%
CD t. t.
X Q 0


ml
CD
^
«^»
o
3 < flQ O
0 1 1 1
•^ CD CD ffl
CO CD ^Q
u
itf
3









































L.
CD

3
U
ia
M-
C CD
10 t—
ia
£~-

CD ie
o 
-------










g
£


i



gs

OCUl
£fe
5"
l§
Q. 3
CO tO
5 ui
UIZ
H
Ul Ul
£ X
Q
|*
*i


E (/>

Ic
w
ui

in

UI









i.
0
C CO O t.
3f- OI— £
J3 O ItJ r- -V
U 3 Mr- =
O U. 3— *
•H X CC
»*- *^
t.
CO CD •-
c u -c
u c c c
3 en co eg c
ja •*" X! B c
u in — -S 5
**« CC
£ Ul CC
0 *> 1C I.
in o o Q >€
3 .-- i-
40 *J g c
~ lO 1 K
U C 0 .C
a ~- ic
CO (9 CM
u
°!c
Ceo
c§§.
1
£
3
Q 4) O
r™> t- O
C ffi
»-« a.
i

O Q.
^JS
> -a
Q) ^*
•a o re
^=S
2S~
o a.
S * 5


U "O |

» 5-- c
O) on
CO 4-» +»
U_ CJ



c
O r~
— O CD
4J t- O
3 4-» •>•
r« O CO
000
a.



8 o S
n -i>-io>oio 01 o o r-
CMCMO O O ^ i— I *+ i-4 C7> tnmCn
«5 >O CO p» »
CO CO
o^-o^
*O "O 2 "O "O

tu" u coco*' ct.
— U O — t. J3 1. •— I_ — !- L.
ICCJU ICr-4CM (O*^>CO 1C ICr— 0>
CD r— t. «O 3 ulU^k^lfc ^IC U1L. {ACOU1

oo o^^- CO^COCD (O3 co c3 1. O -Q O C U J=
U3 r-COCCD r— CO 3 3 C U t. r- (B i— 0) >, O)
c] f— cn ^-<4Jf— >4 3oi 4->u*^ a: 3< 3M -MCO
•v-ZCQCMX CDfOCQCOCO-M t/lid (Q
•4-1 ** >Z U1IO3ZZ
ui o z a
3O CO
£1 CO C
l|f « 5 f ill

O 2 HH +J Z 2 Ul Q O
1 i i





































.
e


in
»- Ul r-
ui +j i. a
** II
•M •!-> »*- >
CO U) >Q
Of— C
o. o ;o *;
CD •• CO
C C •M C
CD "S i— T) 18
•tJ C C +J
•o e o 10 us
c o> •o
x •»• <» eo
O Ul C C II
r- CO — i-
Jf-f?#£
5-36

-------
5.7  ASH SAMPLING DATA
     Table 5-16 summarizes the  PCDD/PCDF content of ash  samples  collected  at
the Tier 4 test sites.  These data are discussed below for each  source
category grouping.

Sewage Sludge  Incinerators
     For all three sewage sludge  incinerators 2378-TCDD  was not  detected in
the bottom ash.  Total PCDD and total PCDF were not detected at  Site SSI-A,
and each were  no more than 70 ppt at Sites SSI-B and SSI-C.  At  Site SSI-C,
filterable solids from the scrubber water and the filtrate were  analyzed
separately.  The filterable solids did not contain detectable quantities of
2378-TCDD, but 0.7 ng/liter of  total PCDD and 13 ng/liter of total  PCDF were
detected.  The filtrate contained much less PCDD/PCDF; 2 x 10"5  ng/liter of
total PCDD and 3 x 10"4 ng/liter  of total PCDF were detected.

Black Liquor Boilers
     The ESP ash was sampled and  analyzed only at Site BLB-C.  At Site BLB-A,
particulates were controlled with a wet bottom ESP; therefore ash samples
could not be collected.  At site BLB-B, particulates were controlled with  a
dry bottom ESP but there was no accessible ash sampling  location.   The ESP ash
at Site BLB-C  did not contain detectable quantities of 2378-TCDD, but
contained 20 ppt of total PCDD and 20 ppt of total PCDF.                ..   •

Wood Combustion
     The bottom ash and baghouse dust from Site WFB-A and the bottom ash from
Site ISW-A were analyzed for PCDD/PCDF.
     The baghouse dust from WFB-A contained 100 ppt of 2378-TCDD, 1.1 x 106
ppt of total  PCDD and 3.2 x 105 ppt of total  PCDF.  The  bottom ash from the
primary chamber and secondary chamber of boiler WFB-A were analyzed
separately.   Furans were not detected in either bottom ash.   However, 150 ppt
of PCDD's were detected in the primary chamber bottom ash and 100 ppt of
PCDD's were detected in the secondary chamber bottom ash.
     For Site  ISW-A,  the bottom ash contained 140 ppt of 2378-TCDD,  1.4 x 105
ppt of total  PCDD and 7,400 ppt of total  PCDF.
                                     5-37

-------
  TABLE 5-16.  SUMMARY OF ASH SAMPLE  PCDD/PCDF DATA FOR THE TIER 4 TEST SITES

PCDD/PCDF Content of Sample
(DDt}
Source
Category
Sewage Sludge
Incinerators
Site
SSI-A
SSI-C
Ash Sample Type
Bottom Ash
Bottom Ash
2378-TCDD
ND
ND
Scrubber water filterable ND




Black Liquor
Boilers3
Hood
Combustion


Metals






Miscellaneous



SSI-B
BLB-C
ISW-A
WFB-A


WRI-A





MET-A
CRF-A
solids (ng/1)
Scrubber water filtrate
(ng/1 scrubber water)
Bottom Ash
Dry bottom ESP Ash
Bottom Ash
Bottom Ash - Primary
Bottom Ash - Secondary
Baghouse Dust
Bottom Ash - Primary
Wire Only
Wire & Transformer
Settling Chamber
Wire Only
Wire & Transformer
Baghouse Dust
Baghouse Ash

(ppt)ND
ND
ND
ND
140
NR
NR
100 1

20
ND

20
6
150
ND
Total
PCDD
ND
20
0.7

0.3
2 x 10"3
10
20
138,200
150
100
,143,600

240,000
19,500

521,000
231,000
106.,.60Q
110
Total
PCDF
ND
70
13

4.0
3 x 10"*
50
20
7,400
ND
ND
315,600

730,000
82,000

2,610,300
657,600
571,700
50
 At Site BLB-A, particulates were controlled with a wet bottom ESP; therefore,
 ash samples could not be collected.  At Site BLB-B, the dry bottom ESP did not
 have an accessible ash sampling location.

For Site DBR-A, contamination prevented meaningful and valid results.
DBR
SSI
BLB
ISW
WFB
WRI
CRF
Drum and barrel incinerator
Sewage sludge incinerator
Black liquor boiler
Industrial solid waste incinerator
Wood-fired boiler
Wire reclamation incinerator
Carbon regeneration furnace
                                     5-38

-------
Metals Recovery
     Bottom ash and settling chamber ash were sampled at Site WRI-A at two
feed conditions:  wire-only feed and wire and transformer feed.  For the
wire-only feed condition, the bottom ash contained 20 ppt of 2378-TCDD, 2.4 x
105 ppt of total PCDD and 7.3 x 105 ppt of total PCDF.  The settling chamber
ash contained 20 ppt of 2378-TCDD, 5.2 x 105 ppt of total PCDD and 2.6 x  106
ppt of total PCDF.
     For the wire and transformer feed condition, the bottom ash did not
contain detectable quantities of 2378-TCDD.  However, 19,500 ppt of total PCDD
and 82,000 ppt of total PCDF were detected.  The settling chamber ash
contained 6 ppt of 2378-TCDD, 2.3 x 105 ppt of total PCDD and 6.6 x 105 ppt of
total PCDF.
     At Site MET-A, the baghouse dust contained 150 ppt of 2378-TCDD,
1.2 x 105 ppt of total PCDD and 5.7 x 105 ppt of total PCDF.
Miscellaneous Combustion Sources
     The baghouse ash at Site CRF-A was analyzed.
The baghouse ash did not
contain detectable quantities of 2378-TCDD.  However, 1.1 x 10  ppt of total
PCDD and 5.7 x 105 ppt of total PCDF were detected.

5.8  AMBIENT AIR SAMPLING DATA
     Table 5-17 summarizes the in-plant ambient air PCDD/PCDF concentration
data developed for the four Tier 4 test sites for which samples were taken and
analyzed (Sites MET-A, DBR-A, SSI-C, and CRF-A).  Samples were also taken at
Sites ISW-A and BLB-C but these samples were not analyzed.
     No 2378-TCDD was detected in the ambient samples from any of the four
sites for which data were available.  Ambient total PCDD concentrations ranged
from just above detectable at Site CRF-A and Site SSI-C (0.02 ng/dscm) to
about 0.4 ng/dscm at Site DBR-A.  Ambient total PCDF concentrations were
considerably higher, ranging from 0.04 ng/dscm at Site CRF-A to 5.3 ng/dscm at
Site DBR-A.
                                     5-39

-------
   TABLE 5-17.  SUMMARY OF AMBIENT PCDD/PCDF DATA FOR THE TIER 4 TEST SITES

Sitea .
MET-Ab
DBR-AC
CRF-Ad
SSI-C
Ambient
2378-TCDD
NR
ND (0.03)
ND (0.1)
ND (0.0004)
PCDD/PCDF Concentration
(nq/dscm)
Total PCDD
0.15
0.39
0.02
0.23

Total PCDF
1.1
5.3
0.04
0.16

 were not analyzed.

 Ambient air samples at Site MET-A were taken in the vicinity of the dilution
 air intake damper.

°Ambient air samples at Site DBR-A were taken in the vicinity of the furnace
 exit location.

 Ambient air samples at Site CRF-A were taken in the vicinity of the spray
 cooler air atomizer intake.
ND = Not detected at specified limits of detection.
NR = Not reported due to large quantities of native  CDD's  and CDF's  which
     interfered with the isomer-specific analysis.
                                    5-40

-------
                                   CHAPTER 6
                                 DATA ANALYSIS
     As discussed in Chapters 1 and 2, the primary purpose of the Tier 4
program was to address the questions:  Do combustion sources emit CDD's?  If
so, how much?  These questions have been largely answered by the results of
the literature review and the stack testing program.  A secondary objective of
the Tier 4 program was to attempt to address some other questions including:
What factors affect PCDD/PCDF emissions?  How effective are conventional
control devices for controlling PCDD/PCDF emissions?
     This chapter will address these questions using statistical analysis of
the stack gas data and subjective assessments.  Statistical analysis of the
ash data as well as correlations between ash samples and stack samples are
presented in Chapter 8.  The statistical analysis is presented in Section 6.1,
and subjective assessments of the data are presented in Section 6.2.  A
discussion of PCDD/PCDF homologue distributions is presented in Section 6.3,
and control of PCDD emissions is considered in Section 6.4.

6.1  FACTORS AFFECTING CDD/CDF EMISSIONS
     Based on the literature review, there are many variables that have been
hypothesized to affect CDD emissions.  These are listed in Table 6-1.  In
addition to determining PCDD/PCDF emissions, the Tier 4 sampling program was
designed to gather information on these variables or on surrogates for these
variables so that the PCDD/PCDF formation hypotheses and factors affecting
emissions could be tested.  Data sets from seven other test sites were found
in addition to the 13 Tier 4 test sites with sufficient information to be
included in the statistical analysis.  Other emission data sets reported in
the literature did not contain sufficient information on feed characteristics,
combustion temperatures, and combustion parameters (i.e.,  CO, 02, etc.) to be
included in the analysis.  Average emissions and operating data were used from
each site.  A complete listing of the data matrix used in  the statistical
                                      6-1

-------
            TABLE 6-1.  TIER 4 SURROGATE MEASUREMENTS FOR FACTORS
                        AFFECTING DIOXIN EMISSIONS
  Factors Affecting
  Dioxin Emissions
   Variable or Surrogate Variable
   Measured in the Tier 4 Program
PCDD in feed


Precursors in feed


Chlorine in feed

Combustion temperature


Residence time

Oxygen availability


Feed processing



Supplemental fuel
None (some feed samples archived for
potential PCDD analysis)

Chlorobenzenes, chlorophenols, PCB's
and TOX in feed

Chlorine in feed and HC1 emissions

Combustion chamber temperature or
flue gas temperature

Not measured, no surrogate
   content of stack gas,
   emissions, THC emissions
Broad characterizations of feed
types, moisture content of feeds (if
appropriate), feed rate

Fuel flow and heat input recorded,
if available
NOTE:  TOX » total organic halogen content of feed
       THC » total hydrocarbon concentration in flue gas
                                     6-2

-------
analysis  is given in Table 6-2.  The data sets included in the matrix are
identified in Table 6-3.
     The  remainder of this section presents an analysis of the factors
affecting the magnitude of CDD/CDF emissions from combustion sources.  The
results of a rank order statistical analysis of the Tier 4 data are considered
first, followed by a discussion of the sewage sludge incinerator, black liquor
boiler and municipal solid waste combustor data.
6.1.1  Rank Order Statistical Analysis
     A non-parametric "rank order" statistical analysis was performed on the
Tier 4 data base in an attempt to determine the factors affecting CDD/CDF
emissions.  Rank order statistics investigate associations between variables
by looking at the relative ordering (i.e., ranking) of the variables rather
than their absolute magnitudes.
     Parametric statistics based on magnitudes were not used as a starting
point in the data analysis because non-parametric rank-order statistics were
considered to be more valid for the existing data base.  Key considerations
are as follows:
     o    Conventional  parametric statistics (Pearson product-moment
          correlations)  assume a bivariate normal  population.   Because of the
          limited number of test sites  in the data base and the wide range of
          CDD/CDF emissions found,  the  data  are  not well distributed across
          all  possible  values of variables (e.g.,  emissions from Site MET-A
          are  more than  an order of magnitude higher than  those from the  next
          highest site).
     o    Rank order non-parametric statistics  (Spearman rank correlations)
          assume a continuous bivariate population,  with no assumption  of
          normality.  Thus,  the non-parametric tests require fewer assumptions
          about  the  distribution  of the population  being considered.   In
          addition,  there  is  no minimum sample size  required for most
          non-parametric methods  to be  valid  and reliable.   Finally,  the
          non-parametric tests  provide  an adequate means of treating
          non-detectable analytical  values (they are treated as  "ties").
                                     6-3

-------


O O O O O o O OOO « "
O O 0 o O O O o O « o o
J; 2- fs, o o O O 03 ^ o '^ O ^
*~ » fj d d o' ~ ;j ' o* » « ft o' 	
— O (s- ill O 0
ft - M CD
fe o c o o o o o o o o Q o o

tj r j o O M «r 61 n in «r o- ft «r ft
_ . ° — rl — K>r?WN^rftreftrO
(O *4 — «•
to
I-H
to
>- O OQOOOOOOO OQ
	 1 . O OOOOOOOOO OO
5* ri o omooooMNo • in «•
«C t; oo-oosaino- NCB
o ts. — M in K> ft tom
* 2 "
O
I"* • o o a o o o o o o o o o o o o o o c i
CT . . o o o o o o o o o o o Q oooooc;
to u o ri PS m rs 2i ui o- o o o so o-ocr-o — — <
___ mftmr-.ps.r-i PS o —
£ - NIB n g
to »
QC
O 3 O O O O O O O O OOO OOOOOC.
It. O O O O O O O O 3 OOO OOO O O C
**~ g !••*>•» o ft -o c< — n a- ooo in 4 ri <> -a o •-• o — o>  D-.
»r1sB-.>s3s5ft!B CO ft — 0- in 111
n ^" c* ^ m o w so
tsl
i
VO fit o O Q O 0 O 0 0 O 030 O O O O Q C
J- ooooooooo ooo o o o o o o
UJ OOOOOOOOO OOO OOOOOC
gj | 8?H|||g|= 2|g 823§SP
& o o o o o o o o o o o o o o o o o o o
S o o o o o o o o o o o o o oooooc
u o o o o o o o o o o o o o o o o o o c
X CDpO^OOODftOOOOO^ P* IO CD CO CO CD
S «Ts.m2 — Scoso — SsciyiKi S o- co oa — -c

o- in -" cj so «• «r
"> ooooooooooooooooooo:
s ooooo ooo oo oooooooo oc
o oooooooooo o o o o o o o o o c
— Mrt>rinsaps,ajo-o — ftio»insopscoo-c
& -iNKt«n44NCDO>n-4-O
0 °W
u Xfn i
^ ^m) "^
T3 C ^"
> o
c E ui
O 2.T3
+J — Ol
Ul g
3 U) —
-Q c
§° 5 -
••- o —
o v^ -
*^> t.  e
= ^8
IS 8§
I- O
(DC a. -fl
J5S^ § J
e s- it o 5
i »§-
x zo£
ii o o 3 pS
Q_ ^CC  a> o -o
 0 3 0 t.
i-M-i 4;0
5 § 8 J"S i o •
0 % ,. - "g if 2 ,
0 5 -P C - -
1—  = "* '
co •a j= x
u !K — o -o o
ID m u c t-
J3 CN O >i IO
i •— "- n,
3m .c tg . ^^ CD _ ,
CgOfiUIU. .s. «j
O ^5^ T -O — CJ ° '
£ 4J X  u
.«D2dfe22. *
s 8 s i .*; g a . - f
 (B r*> *
s
o o in — o o o — o o o o
So in n o o o -a o o o o
o — — m co 
50-OOOOOps.OOOO
?Rgo°SS255§2:N
OOOO — -*NOOO — 03
SOOOOO-fl^OOONN
T'OOOr'?NI^«OOOtn
sO CD ft M — ft
ft -0
O O O O 0 O
03 — -c «r — PS
ft > o in o o
in co CD -. o- ft
w T 
-------
           TABLE 6-3.  STATISTICAL ANALYSIS INPUT DATA SOURCES
Observation Number
   in Table 6-2
Data Source/Test Site
       1
       2
       3
       4
       5
       6
       7
       8
       9
      10
      11
      12
      13
      14
      15
      16
      17
      18
      19
      20
   Tier 4/SSI-A
   Tier 4/ISW-A
   Tier 4/SSI-B
   Tier 4/BLB-A
   Tier 4/BLB-B
   Tier 4/WRI-A (wire only)
   Tier 4/WRI-A (wire and transformers)
   Tier 4/WFB-A
   Tier 4/BLB-C
   Tier 4/CRF-A
   Tier 4/MET-A
   Tier 4/DBR-A
   Tier 4/SSI-C
   Tier 4/VIS-A
               1
               2
Ref 168/MWI
Ref 168/MWI
   Ref 49/MWI, Normal operation
   Ref 49/MWI, Long cycle
   Ref 49/MWI, High temperature
   Ref 49/MWI, Low temperature
                                  6-5

-------
      For purposes of the analysis, the following combustion variables were
 considered to potentially affect or be related to the magnitude of CDD/CDF
 emissions: total chloride content of the feed, total organic halide (TOX)
•content of the feed, maximum combustion temperature, flue gas carbon monoxide
 concentration, flue gas total hydrocarbons (THC) concentration, and flue gas
 total chloride concentration.  These data were available for the Tier 4 test
 sites, but were generally not available for the literature data.  Dioxin/furan
 emissions from each site were characterized by flue gas concentrations of
 2378-TCDD, other-TCDD, total PCDD, 2378-TCDF, other-TCDF, and total PCDF,
 correcte'd to 3% 02.
      Rank order correlations were developed between the combustion variables
 and three sets of flue gas PCDD/PCDF data.  Outlet flue gas PCDD/PCDF
 concentration data were considered first, followed by the control  device inlet
 data and a data set corresponding to the maximum PCDD concentration measured
 at each site (i.e., whether measured at the inlet or outlet location).  Each
 of the data sets has advantages and limitations.  The outlet flue gas
 PCDD/PCDF concentration data set is of particular interest because it
 represents actual emissions to the atmosphere and because these data are
 available  for all  test sites.  A major disadvantage of the outlet data set is
 that effects of the control device are not taken into account by the
 combustion variables.   This limitation is removed by looking at the control
 device inlet data set, but the number of test sites included in the inlet data
 set is much smaller.  Also, several of the highest PCDD emitters are removed
 from consideration  since inlet sampling was not performed at these sites
 (i.e.,  no inlet data were obtained for Sites MET-A, WRI-A,  ISW-A).  The
 maximum concentration  data set has some advantages over both the outlet
 emissions data set  and the control  device inlet data set because 1) some of
 the control  device  effects are removed from the analysis,  and 2)  all  of the
 test sites are represented in the data set.   However,  the maximum
 concentration data  set has the disadvantage of including inlet data from some
 sites (i.e.,  sites  for which inlet sampling was performed)  and outlet  data
 from other sites (i.e.,  sites for which outlet-only sampling was performed).
                                       6-6

-------
      The  data  base  used  for  the  non-parametric  tests  is  given  in  Table 6-2.
A  Spearman  rank  correlation  matrix  showing  the  association  found  between  each
pair  of variables was developed.  The  association  between variables  is
measured  by the  Spearman  Rank  Correlation Coefficient  "R".   Values of R larger
than  about  0.8 were considered to indicate  a reasonable  level  of  association
between variables for this analysis.
      The  following  observations  can be made concerning the  results of the  rank
order analysis.
Outlet Flue Gas  Concentration  Data
      o    Using  the entire data  base, there are strong associations  between
          2378-TCDD and total  PCDD emissions (R -  0.93)  and  between  2378-TCDF
          and  total PCDF  (R  =  0.93).
      o    Using  only the  Tier  4 data (except Site  MET-A) there is a  moderate
          association between  flue gas total chloride concentration  and total
          PCDD concentration (R = 0.73), and between flue gas total  chloride
          concentration and  total PCDF concentration (R  = 0.8).  However, when
          data from Site  MET-A are included, the association becomes weaker
          since  MET-A was characterized by  extemely high PCDD emissions and
          relatively low  flue gas total chloride concentrations.
     o    Other  variables considered (i.e.,  maximum combustion temperature,
          CO, THC, total   feed chloride, and TOX) did not show significant
          association with outlet PCDD/PCDF emissions.
Control  Device Inlet Flue Gas Concentration Data
     o    Using all  of the available inlet flue gas concentration data, there
          are strong inverse associations between maximum temperature and
          total PCDD (R = -0.85)  and between maximum temperature and  total
          PCDF (R =  -0.93).   However,  this association may be misleading since
          one or  more of  the maximum temperatures  listed  in  the data  base
          occur downstream of the inlet location.
     o    Using all  of the available inlet emissions data, there  is a strong
          relationship  between  total  PCDD and total PCDF  (R  = 0.86).
                                     6-7

-------
     o    Other variables considered (i.e., CO, THC, flue gas total chloride,
          and total feed chloride) did not show significant association with
          uncontrolled PCDD/PCDF emissions.
Maximum (Controlled, Uncontrolled) Emissions Data
     o    Using all of the available data there is a moderate association
          between TOX and Max PCDD (0.81) and between TOX and Max PCDF
          (0.81).
     o    Other variables considered (i.e., maximum combustion temperature,
          CO, THC, total feed chloride, and flue gas chloride) did not show
          significant association with Max PCDD/Max PCDF emissions.
     As a whole, the statistical analysis did not provide quantitative
relationships between CDD/CDF emissions and the independent variables
considered.  However, the analysis did identify a few potentially meaningful
associations (e.g., PCDD vs. Max temp and TOX).  Further investigation and
testing are needed to confirm or refute the hypotheses suggested by the
analysis.
6.2  Qualitative Observations
     Table 6-4 gives an overall data comparison for the Tier 4 test sites,
showing source characteristics, average inlet and outlet total PCDD emission
concentrations, combustion characteristics, and feed characteristics.  The
test sites are ranked in order of the maximum average total PCDD emissions
concentration measured at each site, regardless of whether it was measured at
the inlet or the outlet location (i.e., Max PCDD).  The purpose of the table
is to provide a qualitative look at the factors affecting CDD emissions.
Although simple comparisons are difficult to make in a data base of this
complexity, a few observations can be made.
     For example, sites with the highest PCDD emission concentrations (MET-A,
WRI-A, DBI-A, and ISW-A) all contained significant quantities of organic
chlorine in the feed, as measured by either the TOX procedure or the GC/MS
precursor analysis.  Conversely, sites with the lowest PCDD emission
concentrations (SSI-A, BLB-B, BLB-C, BLB-A, and SSI-B) contained only trace
quantities of organic chlorine in the feed.  The wood-fired boiler (WFB-A) and
one of the three sewage sludge incinerators (SSI-C) were unique in that no TOX
                                      6-8

-------





z
o
I
z
i
g
to
C/l
1—4
E
UJ
H-
UJ
_J
S
UJ
_J
z
1
^
cc
Q
^

CC
to
s
ce
o
CO
to
UJ
*-*
V)
t— i
V) 1
LU i
*—
*
iii !
H-
UJ
CC
0
u. 1

§ i
to
s
c_
^ :
8
1 1
1
. |
*v
1
c
O M
••- m
ui
UJ -3D

Se
U)


SoN
O
•»- «
ui m
Wi
I*
Ul >
oi
"S-
c|
= ^ fl
•~ 3 O
x .a S.
5 S I
z 8£
IM
or
MJ-M
rri o
3
8°
if,
^
*v
C
C

01
O 4-1
«.£
in e
^
£
8
S


31
2 2
e a>

e j2
o
"•- 4^
U) <0
tn —
i— 4>
£ 3
UJ O




iC
u | I a
h-







1
3
O
CO
V.
(D
4->

to
! ..
•s
! ff


0
\o




i
a.
S
o
m

<§>
r*>
r^
CD



o
o
0*
f*
^






1
i
1






IL
03 O
*-" vn
m
t- ^-4
Q) —
So,
a in
J3 3
U O
o -g,
u. ,
L.
IQ d>
T) U
C 
•*" +J E
» O
4J t_
ss
T3 *•
C U
i-* e


< * O

-4-t t. O)
cT
>o
S5

^
u
VJ



-M
wu
 *-«
Ul «—
fife
fO UJ
0 §

-*J O
U CD
$
2. ^
^

8s
LU
§
•M

o
CQ

>»
u
Ul

a.
t/i
UJ
S
4J
4-»
Q
CD

+J
                               !£   ai
                                          a
                                          i
                                          CO
                         «c  cc c
                         ca  !-• "u
6-9

-------

•


^

3
c

1
i
1

§
o
o
§
VI
c/l
2
H>
UI
I
i
S
1
i !
s
Q
3

S
M
1
V)
*" *



v>
t— j


UJ
H-




1
1 <
V) -
€


t~.
§
Ul

























i I
t
(,
L
_£
C
r
(S
£







i
















|

t


1
LU
u
«Q
C
X
^


I
1
^

«o
**
o
£




UI
u
3
2
Q.




|

g
h-















e
o
CL
fc
0
Q
1
u.



4J
V)
*J
ui
O
>-


It i
S " CO
t3 \o < 

CL — » 3 *x, O)r— *J O t- » U> Q -4J f -M ^- £ »*- Q) J£CLOU AM E t- iO ^*2 5 ** "^ OMMM O1X4-* L.-QU JCX-»JXX?355cL. O OXXX •-• 3g OCO "'^^j^ °i f 5 f -F i Uj a 01 to LU (A O) §"" (D o >o o -u i. i CL CL 1 - i , IS S S ^ In *^ ^ ^ S 0 — • r- *- 5) C <& (A 3 -O -O t 3 — •o en x > c *o (0 C JC *X O O> OCM O *O ** ^H * U » O O < o < < cn 5 v> a v> -a i co O to ca >> 1 ^ "*• m • a O i.i 0 0 m •— • CO & i CL 1 0 O a i ! i 1 ^~ >. ^ m fl i ~ L> O o>® §N SO


-------
and only small quantities of precursors were detected, yet PCDD emissions were
relatively high.  This indicates that while TOX and/or precursor analyses may
be useful as screening tools in identifying potential PCDD emission sources,
they cannot be used alone as quantitative predictors of PCDD emissions.  The
relatively high levels of total chloride found at the high and low ends of the
PCDD emissions spectrum indicate that total chloride analysis alone does not
appear to be a useful indicator of PCDD emissions when comparing across source
categories.
     Further research is needed to improve state-of-the-art clean-up and
extraction procedures for precursor and TOX analysis of complex feed materials
such as those encountered in Tier 4.  The analytical matrices of some of the
samples analyzed (e.g., transformer combustibles from Site WRI-A) were very
complex, and surrogate recoveries were low.
6.2.1  Qualitative Analysis for Sewage Sludge Incinerators
     Table 6-5 gives a "data comparison for the three sewage sludge
incinerators tested in the Tier 4 program, including controlled CDD/CDF
emissions per unit of feed, sludge characteristics, combustion
characteristics, and incinerator and scrubber design parameters.  The test
sites are ranked in order of decreasing controlled total  PCDD emissions from
left to right across the page (i.e., SSI-C, SSI-A, SSI-B).  Mass emission rate
divided by sludge feed rate is used as the comparative emission parameter in
Table 6-5 in order to remove the effect of incinerator size (i.e.,  feed rate)
when comparing emissions from various test sites.
     The data show a wide variation in CDD/CDF emissions  between the three
test sites, with average controlled total  PCDD emissions  ranging from 0.36 ug
total PCDD emitted per kg of dry solids feed (i.e., ug/kg) for Site SSI-C to
0.005 ug/kg for Site SSI-B.  Emissions from Site SSI-B are significantly lower
than emissions from Sites SSI-A and SSI-C.  The major difference in operations
at Site SSI-B relative to SSI-A and SSI-C is the much higher solids content of
the sludge (36 wt% solids for SSI-B vs. approximately 21  wt% solids for SSI-A
and SSI-C).  This allowed for autogenous sludge combustion, higher
temperatures on the individual  hearths, and also a higher exhaust gas
temperature for the top hearth.   In addition,  the total chloride content of
the sludge at Site SSI-B (660 ppm,  dry basis)  was  much lower than that at
                                     6-11

-------




CQ
i— i
CO
CO
*


CO
UJ
r™*
1— <
CO
I
co
UJ
eg. ^
i ^
i
0
H- 4
UJ
CD
0
=3
CO
UJ
cs
^»
LU
CO
LU 0
»— 1 1
1— I— 1
CO
UJ CO
UJ
C£
or
UJ
1—


2~
l~
^^
^,
o
co
3
0,
S
CJ
S
1
•
LO L.
H*^ «j
UJ g
CD L.
< rtJ
H- Q.

"01
Ol
3
S
o
0
o


2



O)
vx
Ol
3
CM
S
o
o

•




Ol
•V.
01
o\
S
o
*
o




Ul
c_
o
p
0
rO
u_
c '
o
Ul
Ul

l5 Q
c o
"O C»
CD 3 CO
r- U. f-
r- »<^ m
O C CM

•P 1<
C O
O -r-
O Q
CD S
01 01 ^ -o 0 j; 0
"oi"oi +J+JQ.Q. OCJQ- 01 1- >»
33 3 I Q. Q. Q- i? ^£. 3 CJrQ
OO<-^ PO =
tnm CM r— o •-< CT»CMO« oo c .-H
§00 ..VO« vOCMOm O CD CMvO
o vo-a-voo eovor-i-t *Q > o
. . on r- « « o
oo « ^r CM o-
m
o
o
p
CQ |
o s
t. CO S
01 01 S > J= 01 0
£ ^ 13. ei S a. s "N. e ui
-s.^. Q. CL OCJQ- Ol '•" "O
C101 +J^Q. Q.&J5 •* 0. >, CJ ^
33 3 3 un OO C7> 'VO SflOcn
OO ^H^O* O H-iS-iOS
om r- » « >- O
OO CMVOCMI CM 3 •
Q |




S s
u P o
CD CD S > JS C Ul
^^ BSJ^SJEO. a "v <» "0
^.^ O.CX OOO. 01 i^,,""*
0101 -P-PQ. Q.S? -^ ^^^rp
33 5310 OO CM 0) * 0 E
O •— 1 i— • ^T O • CMO Ci-CM
voo (f» t-t o - * ^^om I-HO -i-(— cMr>»
(n« • • in o r»^i-4r-4 vo o. o
• m om« * » E o
o CMui'Hi on CM t-t •
§
*^ Ul
—.01 o
S. Ul CD P
•o IB s- ui
— jQ 3 •'-
•P £-
•P >» rtJ CO *^ CD
CI- Ul S-t- >>-P —>
CDT3 (J C03 t-O >«
4J *^ •*- CL -P "O , nj «^--- t-o
o ue-^'-PCO'-^-c P 3-1-
•i- co t- efo +J o >» ui -P p
Ul T3CPPCDI0C '^-^CD >-Oi
•P- •i-OUc--PUlCDOl UCDO. CD
s- uic-u 10 > o. in
Qn CD COO t- CD.CO U1^IQ.OH- S CD -p iO fO CO O
OO O Ult-.CO -C t-UlOQ t-o t- i- +••
Q.Q. «J -0-PCJU1 O SJ8U1-P « « CO «
S_ t- ••-
(OiaJZOOrtJUO'i-UJiOP T3, 3 i—3
•ppocos>pco-'- Xo. <9 «3 t- s- -per
OO OS--P
-------
 Sites  SSI-A  and  SSI-C  (2,000  ppm  and  1,500 ppm, respectively).  These  factors
 may  be responsible  for the  reduced  CDD/CDF emissions at Site SSI-B.
     Differences in other factors considered,  such as precursor content of the
 sludge,  CO emissions,  excess  air, incinerator  size, and scrubber operation did
 not  seem to  affect  the magnitude  of CDD/CDF emissions.
 6.2.2   Qualitative  Data Analysis  for  Black Liauor Boilers
     Table 6r6 gives a data comparison for the three black liquor boilers
 tested in the Tier  4 program.   Included are total PCDD and total PCDF
 emissions per unit  of  feed, black liquor feed characteristics, combustion
 characteristics,  boiler design  parameters, and electrostatic precipitator
 design parameters.  The data  show that controlled total PCDD emissions per
 unit of feed ranged from about  0.3  x  10"2 ug PCDD per kg black liquor solids
 feed for Site BLB-A to  1.6  x  10"2 ug/kg for Site BLB-C.  This represents a
 range  of approximately  a factor of  5.
     The process  data  shown in  Table  6-6 do not provide a simple explanation
 for the  observed  range  of PCDD  emissions.  The three black liquor boilers had
 similar  solids content  in the feed  (63 to 70 percent solids) and similar
 precursor characteristics (trace quantities of chlorophenols were detected for
 each site).  Site BLB-A showed  lower  total chloride in the black liquor feed
 (1,400 ppm, dry weight  basis) than Sites BLB-B or BLB-C (7,500 ppm and
 5,300  ppm, respectively), which may account for the lower measured emissions.
 Combustion temperature  data were not  available for any of the black liquor
 boiler test sites.
 6.2.3  Qualitative  Data Analysis for Municipal  Waste Incinerators
     This section summarizes the municipal waste incinerator data base and
 presents  an analysis of factors affecting PCDD and PCDF emissions from
municipal waste incinerators  (MWPs).   The original  focus of the analysis was
 to evaluate these emissions in  light of key incinerator design and operating
 variables that may  affect the formation and emission of PCDD's and PCDF's.
The data  base for the analysis consisted of MWI CDD/CDF emissions testing
 identified in the literature review.  Unfortunately,  complete operating and
design data were often not obtained and/or reported for the MWI test sites in
                                     6-13

-------
en en
O> CD





^f
CQ
en



*




•




CO
LU
H- CQ
CO CO
K- CO1
CO
LU
t—
o:
LU
(—4
§
1
g
1
^f
* V
LU ca
j-i _j
1^™ £2
LU
LU
cc
LU
f—
g
LL.

,*"!
O
cn
>— «
a;
^c
Qj
j*l
^j
o
^c
r""
^C
O
*
1 0>
*O +j
Llj g
*C IB
H- Q.
3 3
CM CM

O O
i—l i— 1

X X
CO •-(
CM CM
• *
O O




en en
^f.^.
en en
3 3

CM CM

O O


X X
1— 1 *J-
T CM
• *
O O




en en
en en
3 3
CM CM
o'o
I— 1 r-4
X X
VD Ck

O














in
C

U7 *•*
tflf, l~\ 1 1
•*•• J^\ Cj
Sti_ O CJ
LU ^ Q. Q.
"O i_\ r^ r^
(D*^ *O fl3
^^"^™ InJ *J
r- E O O
O3 (— I—

O ^^
o
en en ^
3 3 Q. °
CM CM O 63
II > ro £- S
00 ^5 S S £ "00 0
rH r-4 CL CL CL ^ CL "O P O
P CL CL CL 53 P O P
X X 3: S OO
OLO VOVOO OS jQLD
voin •TO r-4 cr\ ~O G\ —I Pr-^
• • tn ** • ^» w
O O vo <-l O O) S
H
1
CM
i 3
en en
^t^l CM
en en O
3 =
a?
CM CM ro P
II > O
OO S3£ SO) •» ^~*
O l~-(~-h— O£  CM t^ in CL"O P
XX OO VO«T3 PO OO
m •— i «mo co ox •<*
«• ro* s mo >i>— i
CM T to in I oo _i '—
0 Q
Q LO
ZZ CM
U)
U)
jQ
-» p uj "a.
U> Ifl (D O <—
O -P •— 3 •*- 3 CD
•t- c uJ U» — • P Q. t- £_
p CD O « i— CQ)|— P Q)t-
c fl) (jinc-"— -^t-po-p— • ••" Q.OJ
O P irJO)Ul!--CI3<'-O CL >>CL
f- o CB^apiOM— 'OWrO>» -^ PS
(/)"ej i_ ••— uio o >fl ^ *^  O
S,2 QU_ -C O-^ -^ t- O^( 3 >l C SP>— U) "0(1) rtjt. -i-CO
1^"*™^ ^ P o CMO a o o o s- 3
P = i— i— -^i— a. u> ooi— t- ctcn «- CQ O CO LU
                      6-14

-------
 the literature.   However,  by including information  from several  sources  some
 common  information was  found for a  few of the  sites.
      6.2.3.1   Description  of the MWI  Data Base.   Thirty-eight  articles were
 reviewed for  information describing emissions  testing  for  PCDD's and  PCDF's at
 municipal  waste  incinerators.   Of these,  28  studies contained  sufficient data
 on  CDD/CDF emissions, CDD/CDF fly ash content, and  process operating
 characteristics  to be useful  in this  data analysis.  Table 6-7 summarizes ODD
 and CDF flue  gas emissions  data from  nine MWI's  located in the United States
 and Canada.   Pertinent  design  and operating  data are also  summarized  as
 available.  An in-depth comparison  of additional  operating variables  cannot be
 made since the data were not  available for each  of the  tested  units.
 Table 6-8  summarizes CDD and  CDF flue gas emissions data from  eight MWI's
 located in Europe.  In  addition,  the  literature  review  identified a facility
 in  Tsushima,  Japan, where MWI  emissions testing  had been performed.51
 However, the  results of the emissions  testing at  the Tsushima  facility were
 removed from  this  analysis because  the test  results have not been verified.
 The  remaining  studies in the literature were not  used in the data analysis
 because these  reports contained  data  reported elsewhere or because emissions
 data  could.not be  reduced into  suitable units for this  analysis  (e.g., in some
 cases data were  reported as ng/sample).   Data contained in the rejected
 studies  were not entered into the literature data base.  However, these
 studies  and their  reasons for rejection are-presented in Table 6-9.
      It  is important to understand that the data used in this analysis are the
 "as-published" data.  No adjustments were made to CDD and CDF concentrations
 reported in the  literature to account for  1)  unreported recovery efficiencies
 of surrogate compounds used in the analytical procedures, 2) incomplete
 analysis of the  tetra- through octa-PCDD/PCDF homologues, or 3) variations in
 excess  combustion air (i.e., excess air dilution).  These limitations in the
data base are discussed below.
     Because of the complexity of PCDD/PCDF analysis,  isotopically labeled
 internal standards  (i.e.,  surrogate standards)  are used to estimate  actual
recoveries of PCDD and PCDF isomers in each sample analysis.  However, less
than 5 percent of the emissions studies that  were reviewed report surrogate
                                     6-15

-------
                TABLE 6-7.  DIOXIN AND FURAN EMISSION DATA SUMMARY FOR
                            UNITED STATES AND CANADIAN MSW INCINERATORS
 Facility Location
 (Faci1i tv Number)
                                         Average   Average
Operating         Emissions   Average     PCDD      PCDF
 Capacity   Fuel   Control  Combustion  Emissions Emissions
 Tons/Day   Type   Device     Temp. C     nq/m	nq/m   REF
 1.  Albany,  NY (11)
    a.  condition lu
    b.  condition 1

 2.  Charlottetown,
    P.
3.
a. condition l5
b. condition 2°
c. condition 3^
d. condition 4
Hampton, VA (35)
a. (1983 tests)
b. (1981 tests)
c. (1982 tests)
d. (1984 tests)
 4. Akron, OH (36)

 5. Dyersburg, TN (38)
   600
   600
                           42
                           42
                           45
                           41
   125
   125
   125
   125

   600

    50
RDF
RDF
            MSW
            MSW
            MSW.
            MSW
MSW
MSW
MSW
MSW

RDF

MSW
 ESP
 ESP
       Settling
        hopper
 ESP
 ESP
 ESP
 ESP

 ESP

None
           1,000
           1,000
           1,130
             900
                                                        810
                                                        810
 165.0
  73.8
 107.0
 103.0
  62.0
 123
2,300
4,200
  245
5,950
   17.3
    8.5
  143.0
  156.0
   95.0
   98.0
11,000
 7,400
   384
 8,300
 32
 32
 49
 49
 49
 49
 92
174
101
205
  174       458    101

  11.2      72.5   101
6.
7.
8.
9.
Philadelphia, PA
(N.W. Unit 1) (46)
(N.W. Unit 2) (46)
Dayton, OH (48)
Canada (58)
Chicago, IL (59)
(Northwest)
--
225
270
1,600
MSW
MSW
RDF
RDF
MSW
ESP
ESP
ESP
ESP
ESP
940 2,400
990 1,333
47.6
700 2,680
1,100 45
2,500 168
1,066 168
110.0 174
181
460 196
^condition 1 » with overfire natural gas
"condition 2 » without overfire natural gas
jCondition 1 = normal operation (i.e., normal secondary chamber temperature)
"condition 2 ~ long feed cycle operation
^condition 3 ~ high secondary chamber operation
^condition 4 « low secondary chamber operation
9A combustion air preheater and a stack gas carbon monoxide monitoring system were
 installed in between the 1983 and 1985 tests at the Hampton facility.
NOTE:  PCDD/PCDF concentrations in this table are at as-measured oxygen conditions.
                                         6-16

-------
                  TABLE 6-8.  DIOXIN AND FURAN EMISSION DATA SUMMARY
                              FOR EUROPEAN MSW INCINERATORS
Operating Emissions
Facility Location Capacity Fuel Control
(Facility Number) Tons/Dav Tvne Devir.P
European Facilities
1. Italy (13)
2. Italy (14)
3. -Italy (15)
4. Italy (16)
5. Italy (17)
6. Italy (18)
7. Italy (33)
a. Condition la
b. Condition 2
8. Severn, Belgium
^condition 1: dry MSW
Condition 2: wet MSW

-_
ESP
ESP
ESP
ESP
ESP

Dry MSW ESP
Wet MSW ESP
MSWC


Average Average
Average PCDD PCDF
Combustion Emissions Emissions
Terno. C nq/nT na/nr RFF

473
49,000
7,500
4,410
1,030
588

269
7,360
51.5



61.5
7,900
1,440
3,600
472
51.0

397
6,910
34.3



117
117
11-7
117
117
117

118
118
119


                                            *
Dash (--) indicates data were not reported.
NOTE:  PCDD/PCDF concentrations in this table are at as-measured oxygen conditions.
                                        6-17

-------
          TABLE 6-9.  MWI EMISSION STUDIES NOT INCLUDED IN DATA BASE
Reference
 Number
Reason for Rejection
  44      Flue gas emissions reported as ppb/sample, sample size unknown.
  51      Dry repeat not published.
  97      Flue gas emissions reported as ng/sample, sample size unknown.
 142      Duplicate of reference 144 data.
 148      It is not possible to determine which European facilities were
          tested.  The report states that emissions testing was performed at
          "at least 25 facilities."  Emissions are presented for facilities
          that were found to be lowest,  average and highest emitters.
 202      Duplicate of reference 203 data.
 221      Dioxin and furan emissions were not contained in the report.
 228      Emissions reported as ng/sample, sample size unknown.
 243      Analysis for the presence of PCDD's and PCDF's was not performed on
          stack gas samples.
                                     6-18

-------
 recovery values.  Thus,  there may be variability in the data base attributable
 to analytical  procedure  differences between the various laboratories
 represented in the literature studies.
      In some studies flue gas and fly ash samples were analyzed for only
 specific homologues (most commonly the  tetra-  and octa-CDD and CDF
 homologues).12'26'43'71'115  This is because isotopically-labeled standards
 for all  of the CDD/CDF homologues have  become  available only recently.   The
 emissions data contained in these studies are  included in  the analysis  but  no
 adjustments were made to specific homologue concentrations to obtain emission
 values  representing total  concentrations  (i.e.,  all  homologues)  of PCDD's and
 PCDF's.   Additionally,   the 2378-TCDD and 2378-TCDF isomers are not routinely
 analyzed for in all  samples.   Less than 3 percent of the facilities tested
 report  data for these isomers.205'32'196'101'168
      Variations in  excess  combustion air  between the test  facilities affect
 the measured CDD/CDF concentration by both volumetric stack gas  dilution and
 at  the  basic combustion  level.   Ideally,  the volumetric stack gas dilution
 effect  can  be  accounted  for by correcting the  data to a reference oxygen or
 carbon  dioxide level  (e.g.,  3% 02 or 12%  C02).   However, since oxygen or
 carbon  dioxide concentration  data were not reported  for all  test facilities,
 the data could not  be uniformly  treated.   For  this  reason,  none  of the  CDD/CDF
 emission concentration data were  corrected to  a  reference  oxygen level.
      6-2.3.2   Rank  Order Analysis of Flue Gas  Emissions  of CDD's and CDF's.
      The focus  of this section is to  relate  flue  gas  and fly  ash emissions  of
 CDD and  CDF  to  key MWI design and operating  parameters.  As described in
 Chapter  4, many  factors are believed  to be responsible  for  the formation of
 CDD and  CDF from municipal waste  combustion.  Two that were  identified  and  are
 related,  are feed composition and  furnace  temperature.  This  section will show
 that there is a possible relationship between furnace temperature, which is
 largely  controlled by the heat content of the fuel, and flue gas emissions  of
 PCDD's and PCDF's.  However, this relationship is based on limited data.  Less
than 25 percent of the MWI emissions studies that were reviewed contain
 information describing both feed material  or operating conditions.
                                     6-19

-------
     Because many  factors  are  involved  in the  formation of CDD's and CDF's
from MWI's, the  reaction mechanism  and  factors affecting the mechanism are not
well understood.   However,  a convenient method to test which variables play a
key role  is a Rank Order plot.  Table 6-10 summarizes the rank order of flue
gas PCDD  and PCDF  emissions and combustion temperature for the seven MWI units
in the data base for which  both flue gas PCDD/PCDF emissions data and
combustion temperature data are available.  Figure 6-1 presents a rank order
plot of flue gas emissions  of  PCDD's versus decreasing combustion temperature.
The seven facilities included  in the table and figure are all located in the
United States and  Canada.   Figure 6-2 presents a similar Rank Order plot of
flue gas  emissions of PCDF's versus decreasing combustion temperature for 11
facilities located in North America, Europe and Japan.  Figures 6-1 and 6-2
both indicate that ranked flue gas emissions of PCDD's and PCDF's increase
with ranked decreasing furnace temperature.
     Spearman rank correlation coefficients were computed for combustion
temperature versus PCDD's  (Spearman correlation coefficient = -0.99) and for
combustion temperature versus  PCDF's (Spearman correlation coefficient
« -0.62).  As discussed earlier, the Spearman rank correlation coefficient
measures the degree of correspondence between variable rankings, instead of
between actual variable values.  However, it can still be considered a measure
of association between samples and an estimate of the association between
variables in a continuous bivariate population.
     The hypothesis of no agreement between sets of rankings was tested for
both PCDD and PCDF and was rejected at  a significance level of 10 percent.
Therefore, an inverse relationship exists between combustion temperature rank
and PCDD and PCDF emission ranks.  Thus, the data indicate that MWI facilities
with high combustion temperatures tend  to have lower CDD/CDF emissions than
MWI facilities with low combustion temperatures.
     Table 6-11 summarizes the rank order of fly ash PCDD content and
combustion temperature for the 16 MWI units in the data base for which both
fly ash data and combustion temperature data are available.  Figure 6-3
presents a rank order plot of fly ash PCDD content versus decreasing furnace
temperature.   The plot indicates that no relationship exists between these two
                                     6-20

-------










K
I —
«c
OL
Q!
LU

<
Ct
3

t—
2
CO

Ll_
§
Q
en
Q
Q
U
o.
fe
C/l

cn
CO

Z
UJ

CO
s
«J
u_
0

1

cc

O

1
xO
UJ
1













id
LU Z
U.
CO
C3I— O
Z  Q, Ht C
«c s:
UJ
cn
UJ Z
cc o cn ^s.
uj o co cn
> Z c


z
o
IK
So £
"
_j
o
ex



en MH
CO °
CO
£
LU
a?

H-
_J
^
U_


h- HH Q
III




o

u
o
i

h- a
31
o o
£ z

V ««• ^3- Oi




m «-H CM \o


m ^< Ln P-*



m CM v cj>



on ^H Ln r-*




m Ln eo o
S * m 8
CM


r** CM m o
O \o CM o
f-< ^H O
-H"
^^



 oj
Q. a. Q.
CL a. Q.
o o o
X X X Q_
cn
Ol CD Ol LU
C C C
555
QJ 0) fl>
cn co co


3S X- X X
£ S t? S





CM Ln o in
•v ^r *r CM




re re
c ~o c *o c re
n re * re * "o
Sc o c o re <
re -M re -M c >
Q) O o o QJ re
-M -M +-> O »
4^ » 4J » +J c
O HH O HH O t-i O
fes fes *is f
j= j= £ re
O CJ O X


o o o in
CM CM CM m


co co •-! »O Ln

•— 1 i—l ^H i-H CM



^* un i ^ eo


^ ^e i CM o
t



vo m r^ 1-4 co



^ ^«r 4 eo CM VD




o Ln o o
m r- i *Q \o
in r*- i ^r CM
* •>
CM CO


o Ln o m o
r» ^- co «* m
m r- *o cf»
CM CM in




Cf» O^ Ch O ^-t
Cft C?\ \O ^H CO
~







Q- a. a. CL o_
cn cn cn co cn
UJ UJ UJ UJ m





% * Li_ X 3£ '
CO CO g CO CO





! ! § § id
CN (O ^H

10 10


a. a. £

 o
«— xB I5 5 |
• • 3: £ re
Z Z CO O X


^ ^ in m m


































^
£9
** 5
L. re
OJ I-
a. Q)

v E
4J <]}
4J
L.
OJ t-
"l|
"S >> °
t- t- >%
3 10 U
10 e "O
t- 0 C
0) U O
i-88
0) ^- 01
c
u ^o c


E re -M
re ^ re
.c a> i-
O Q. OJ
O CL O
>» o •—
•0 3 U —
C +J 3 -^
o re •*-* re
UL.ro>
(D flj L. re
u> a. . L.
re re t. QJ
t- "o re *
OJ C -0
Q. o c re
O U O -4-*
a; u re
^- m a> o
10 tn
ex: u
O) *
0—01
Z X -J I
«o ^ u
6-21

-------
1O
                                                   1
                                                   7
T
8
            3456789    10

Ranking in Ord«r of D«cr«a«lng Combustion T»mp«ratur«
   Figure 6-1.  Rank Order Plot of PCDD Emissions vs.  Combustion
                Temperature for Municipal Waste Incinerators
                            (Data in Table  6-10)
                               6-22

-------
10 -I
                  i
                  2
I
3
I
4
i
5
1      2     34      5     6      789     10

  Ranking In Order of D*cr«a«ing Combustion T«mp«ratur«
   Figure 6-2.  Rank Order Plot of PCDF Emissions  vs.  Combustion
               Temperature for Municipal  Waste  Incinerators
                         (Data in Tabls 6-1Q)

                             6-23

-------
   gl
   '
SfcJe

§§§

SSI
                                   _    _H    _,    _-4    (M
                             co    *r    m
                                                                                   «»   *C    CM
                                                                                         O    O
                                                                                         m    o
                                                     a    2
                              8.
                                    Si    ft
85
                                                                                   w    v    ~-    —     C
                                                                              O    —    r-    ^-    t—
                  o    o    o    c    c
                  +J    +*    4J    O    O
                        S,   h- -w o
                                                                                                                       •V*  1-  U  >*   C
                                                                                                                       (O  3  (O  L. M- to -P
                                                                                                                       £ +J -a  fl  o o c
                                                                                                                       +j  us  e -o
                                          z
                                             t_  O  C  0)  (D  <
                                                o  o  o  a. i
                                           ) ^J    —    t.
                                                                                                                                 .
                                                                                                                                O  C  0>  C i—
                                                                                                                       — 10  i_  o
                                          §
                                          »— w ^  ra  Ji.
                                             fl) 0)  U »— •
                                          «— Ul O. 0)  <0
                                                                                                                                a  o> +•»  ID —
                                                                                                                       
                                                                                                                       ce z x -J u. «c  u_
                                                                                                                      TO JD  u ~a o> 4- en
                                                             6-24

-------
IP -
18-
14-
13-
12-
i "•
o 10-
40
e
?
o g _
O
£ 8-
5 7-
|
"o
1 6"
6
£
4-
3-
2-
1 -
u
0 0
O
o
o
o
0

0
0
o


o


o
o
0
o
i 1 I 1 I 1 1 	 1 	 1 	 1 	
12348878910
           ftankktg hi Order of D*cr«Mlnfl Combustion T«mp«ratur*

Figure  6-3.  Rank  Order Plot of Fly,Ash  PCDD Content vs. Combustion
             Temperature for Municipal Waste Incinerators
                           {Data in Tabie  6-11)

                               6-25

-------
variables.  However, since the fly ash/combustion temperature data base listed
in Table 6-11 contains facilities with potentially widely different designs
and ash handling schemes, other factors such as temperature history of the ash
may have obscured any possible relationship between fly ash PCDD content and
combustion temperature.
     6.2.3.3  Summary.  The analysis presented in this section in conjunction
with the literature review shows an association between two variables that may
affect flue gas emissions of CDD's and CDF's from MWI's.  These factors are
summarized as follows:
     1.   Flue gas emissions of PCDD's and PCDF's appear to be inversely
          related to furnace temperature.  Two factors that directly affect
          furnace temperature are excess air and the heat content of municipal
          waste fuels.  Values of these parameters are not normally reported
          in emission test reports.  However, the data suggest that there is a
          relationship between the heat content of fuels and the emissions of
          CDD and CDF's in flue gases.  Elevated concentrations of PCDD's and
          PCDF's were found in the flue gas samples of two facilities located
          in Hampton, Virginia, and in northern Italy.174'205'87  Emissions
          testing at both facilities was performed over a period of typical
          operation and when the feed was wet.  Maximum CDD and CDF emission
          levels were reported to be approximately 10 times higher when the
                                                                     87
          feed was wet during normal operating conditions (dry feed).    This
          also corresponds to periods of low combustion temperatures.
     2.   Higher levels of CDD's and CDF's are found in the flue gases of
          MWI's that were reported to have flawed incinerator design and
          suffer from poor maintenance.  For example, the literature reports
          that the rear grate seals of a Canadian facility (Facility 181) fit
          poorly and permit undergrate air to by-pass the grates and enter the
          furnace at the rear wall.  Furthermore, blockage of holes in the
          grates also results in poor distribution of combustion air into the
          (lower) furnace.
     3.   The PCDD and PCDF contents of fly ash from MWI's are not related to
          furnace temperature alone, but are probably a function of several
                                     6-26

-------
          design  and operating variables.   In general, however, large
          concentrations of CDD's  and CDF's  in stack gases are accompanied by
          large concentrations of  CDD's and  CDF's  in the fly ash.
     4.   PCDD and PCDF emissions  levels reported  for European facilities are
          higher  than those for United States and  Canadian facilities.
          Although these differences cannot  be explained by this analysis,
          these differences may be due to emissions testing at facilities that
          were located in rural areas where  a large proportion of the waste
          contains agricultural products such as vegetables.  It may also
          reflect differences in incinerator design, incinerator age, sampling
          methodologies, analytical methodologies, or other factors.

6.3  PCDD/PCDF HOMOLOGUE DISTRIBUTIONS
     This section presents a qualitative analysis  of the PCDD/PCDF homologue
distributions for the Tier 4 test  sites.  The purpose of the analysis is to
determine whether a given test site or combustion  source category has a
characteristic homologue "signature."
     Homologue distributions (mole fraction basis) of the inlet (i.e.,
upstream of the control device) and outlet (i.e., downstream of the control
device) emissions from the Tier 4  test sites are shown in Figures 6-4 through
6-12.  The test sites are grouped  by source category to facilitate comparison
between similar sites.  Inlet and  outlet data are presented on facing pages to
facilitate comparisons between uncontrolled and controlled homologue
distributions from each site.  The average magnitude of total  PCDD and total
PCDF emission concentrations for each site are shown on the homologue
distribution figures using the nomenclature PCDD and PCDF.   The homologue
distribution data are presented using the mole fraction basis because this is
more relevant to a consideration of combustion chemistry than the mass
fraction basis.  The sum of the mole fractions for the various PCDD and PCDF
homologues add up to 1.0 for each test run.
     The following general  observations can be made regarding the homologue
distribution results:
                                     6-27

-------
  SEWAGE SLUDGE INCINERATORS
                         INLET
                       Site SSI-A
     Analytical Data were not Reported by Troika for the Inlet MM5
     Samples taken at Site SSI - A due to Unacceptable Surrogate
                    Recovery Results
                      Site SSI - B
      Uncontrolled Emissions from Site SSI • B were not Sampled
       DIOXINS
                      Site SSI - C
0;i *
0;7-
i o...
1 »-
B O.4 •
0,3-
0:5-
9.1 •
PCDD - 1 14 ng/dscm at 3% Oa




J vM
I
i
1
1


I
I
I
i,
FURANS
                                o.


                                0.* -


                                0.3


                                0.2-


                                O.t
                                  PCDF - 507 ng/dscm at 3% O2 I
3371 rCSO Otnv TCOO »«iu-C£0 Hm-COO Hou-COO Otl.-COO
                                 2371 TCOT Other TCOT Pwita-COP HM«-COr Hmtv-COP Oet«-COT
    Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
                   Sewage Sludge Incinerators
                   (Sites SSI - A, SSI - B, SSI - C)

                        Figure 6 - 4.
           from
                         6-28

-------
     SEWAGE SLUDGE INCINERATORS
                           OUTLET
          DIOXINS
                           Site SSI • A
                                           FURANS
   . PCDD = 19.6 ng/dscm at 3%O2
0.7-

0.»-

OJ-


0.*-

O.3

o.a
                                    0.7

                                    a.*
                                    0.4


                                    OJ
                                     PCDF = 43.5 ng/dscm at 3% O2

   2371 TCOO OMMT TCOO f*\tm COO I
                                               -COT !!•• CUT
                     1771 RUN 09
                              U771 RUN 10.
                           Site SSI - B
  0.7-

  a.*-

  OJ
     PCDD = 1.6 ng/dscm at 3% O2
   U7i reao otMr TCOOP
                    CT7I MUM O1
                                  O.7-

                                  0.1-

                                  OJ-


                                  o.«-

                                  OJ-

                                  0.3-

                                  0.1 -
                                    PCDF= 27.6 ng/dscm at 3% O2
                                                 FMM-COT nwu-car o««-car
                                KUN O3
                                       BOB MM OS
                           Site SSI - C
  0.9 -


  o.s -


  0.7 -


  0.9 •
! «-
3 0..-
  0.3 -


  0.2 -


  0.1 -
     3CDD = 52.7 ng/dscm at 3% O2
                                  o.i -

                                  0.7 -
                                  0.3 -

                                  o.a
                                     PCDF - 446 ng/dscm at 3% O2
                                   237* rear Om«r rcor p«nt«-CDT M«M«-COP H«»M-COF Oeta-COr

                              RUN OS   K7B RUN 03
       Om«r TCOO P«f««-COD M«>«*COO Hwt«*COO Octa-COD
                    [771 RUN 01
       Dioxin/Furan Homologue Distibutlons of Controlled Emissions from
                      Sewage Sludge Incinerators
                      (Sites SSI - A, SSI - B, SSI - C)

                           Figure 6-5.
                               6-29

-------
      BLACK LIQUOR BOILERS/INLET
           DIOXINS
                                         FURANS
                          SiteBLB-A
              ng/dscm at 3%
   «7* TCOO OMrTCOOPMi-Caa M*a COO 	 COO Onm-OO

                    P"yi mm 01
                                     PCDF = 1.5 ng/dscm at 3% O2 |
                                         NUN O3
                          Site BLB - B
O.1

 0
          -17.1 ng/dscm at 3% O2
     TCOO OVMP TEBO M
                      RUN O1
                                  1 -
                                   01

                                   0.1 -

                                    0
                                     PCDF - 1.1 ng/dscm at 3% O2
                                     ICT
                                RUN OZ
                                         RUN 03
                          SiteBLB-C
... J PCDD - 9.0 ng/dscm at 3% O2 !
                                    ... j PCDF = 15.1 ng/dscm at 3% O2
*  —j
I  ••'•{
§  o,«^
  0,3 4
                       o«i»-cao

                   C77T RUN 01
                               RUN 02
                                    rear ou>«- TCDT »«it»-car M...-COF »..to-c=r

                                    ren RUN O3
    Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
                    from Black Liquor Boilers
                  (Sites BLB - A, BLB • B, BLB • C)
                          Figure 6 - 6.
                          6-30

-------
   BLACK LIQUOR BOILERS/OUTLET
       DIOXINS
                                            FURANS
                        SiteBLB-A
   PCDD = 0.7 ng/dscm at 3%
0.9-
0.4-
0.3-
0.3-
0.1 -
                  HMU-COD 2
                                   2371
                              RUN 02    KB
                                     rear owv Tear taw-cor MM-COT HMO-COT o««-cor
                                          03
                        Site BLB - B
   PCDD = 1.2 ng/dscm at 3% O2
0.7-
0.«-
0.3-
0.*-


0.2-
0.1 -
 0-
 237> TCOO OVMr TCOO ftftm COO MOM-COO
                                  o.t-
                                  0.*
                                  0.3 -
                                  0.3
                                  0.1 -
                                   0
   PCDF-0.7 ng/dscm at 3% O2
                     RUN 01
                              RUN 02
                                     rear OMT rear
                                     KB RUN O3
                        Site BLB -C
   PCDD=2.9 ng/dscm at 3%
O.4-
OJ-
0.3-
0.1 -
                       ^1
                       Kf
04
0.7
0.1-
041-
0^-
OJ
o.a-
0.1 -
                                     PCDF-2.1 ng/dscm at 3%
                    •coo o«>« coo
                    mm 01
                           EZ9 HUN 02
                                    xiTSTeer oiMrTeerp
                                     BS RUN O3
       Dioxin/Furan Homologue Distributions of Controlled Emissions
                    from Black Liquor Boilers
                 (Sites BLB - A, BLB - B, BLB - C)
                        Figure 6-  7.
                                                      flj^.
                          6-31

-------
        WOOD COMBUSTION/INLET
         DIOXINS
                         Site WFB - A
     FURANS
    PCDD - 102 ng/dscm at 3% O2
                         2Eb!
                                  1  -
                                    oa

                                    0.1
PCDF - 156 ng/dscm at 3% O2
                   cm RUN 01
                               mm 02   EX RUM aa
                          SitelSW-A
         Uncontrolled Emissions from Site ISW - A were not Sampled
                          SiteWS-A
Analytical Data were not Reported by Troika for MMS Samples taken at Site WS - A
            due to Unacceptable Surrogate Recovery Results
        Dioxin/Furan Homologue Distributions of Uncontrolled Emissions
                   from Wood Combustion Processes
                    (Sites WFB - A, ISW - A, WS - A)
                           Figure 6 - 8.


                            6-32

-------
       WOOD COMBUSTION/OUTLET
         DIOXINS
                         SiteWFB-A
                                              FURANS
 0.3 -


 O.4 -


 0.3-


 0.2-


 O.I -
   PCDD = 195 ng/dscm at 3%
                                       PCDF=83.2 ng/dscm at 3% O2|
         r TCOO PwtM-COO H««-COO Haeta«COD OcM-COO

                    E3 *UN on
                                    237B TCOf OtfMT

                            E23 RUN 02   tog RUN 03
                          SitelSW-A
1 -
   PCDD=625 ng/dscm at 3% O2
  2371 TCOO Othw TCOO F
                                   0.7 -


                                   O.i-


                                   0.3-


                                   O.*-


                                   0.3-


                                   0.2-


                                   0.1 -
                                       PCDF-2390 ng/dscm at 3% O2
                     RUN O2
                                    237* rear «MT rear p«>t»-csr Mn-cor

                            ^3 RUN 03   one RUN 04
                          SiteWS-A
Analytical Data were not Reported by Troika for MMS Samples taken at Site WS - A
            due to Unacceptable Surrogate Recovery Results
            Dioxin/Furan Homologue Distributions of Controlled
              Emissions from Wood Combustion Processes
                    (Sites WFB - A, ISW - A, WS - A)
                          Figure 6 - 9.


                           6-33

-------
   MISCELLANEOUS COMBUSTION
             SOURCES/INLET
       DIQXINS
     FURANS
                    Site DBR - A
  PCDD - 570 ng/dscm at 3 % O2
                            u-

                            0.1

                            0.1 •
PCDF - 2040 ng/dscm at 3%O2
                              -rj^Un
                                             _>S 71
              1771 Run O1
                          O2
                                 Run O3
                    SiteCRF-A
OJ

0-3

0.1
      : 28.8 ng/dscm at 3%

PCDF = 70.1 ng/dscm at 3% OU
                 RUN 01   EEJ "UN oa  oa KUN os
        Dioxin/Furan Homologue Distributions of Uncontrolled
        Emissions from Miscellaneous Combustion Sources
                 (Sites DBR - A, CRF - A)
                     Figure 6 -10.

                      6-34

-------
   MISCELLANEOUS COMBUSTION
           SOURCES/OUTLET
       DIOXINS
  FURANS
                    Site DBR - A
  PCDD = 5.0 ng/dscm at 3%
0.7 -


O.I -
PCDF = 27.0 ng/dscm at 3%
 I37» TCOO atMr TCOO »«nl«-COO
              HWU-COO 0«U-COO

              E3 KUN O1
                        RUN O3
                              GQl RUN 03
                    SiteCRF-A
  PCDD= 3.7 ng/dscm at 3% O2
PCDF= 3.3 ng/dscm at 3% O2
      Dioxin/Furan Homologue Distributions of Controlled
      Emissions from Miscellaneous Combustion Sources
              (Sites DBR • A, CRF - A)
                   Figure 6 - 11.


                    6-35

-------
   METALS RECOVERY PROCESSES
                     OUTLET
O.T

O.i

O.J
         DIOXINS
                      Site MET-A
          = 11,900 ng/dscm
         at 3% O0
                      i
                 H«»«"COO 04U-COO

                 P^l RUN 02
0.3

o.*
o.a-

o.t
        FURANS
     PCDF=60,700 ng/dscm
          at 3% O2
1
                            RUN 03
     RUN 04
                 Wzz
                    Site WRI-A (Wire Only)
o.t-
e.7-
r
i ":
i -•
0.1 -
PCDD=173 ng/dscm at 3% O2


I
I
I
I
^ ?c
If
n
»7> TCOO OOxr Tg»>»iU CBO IU«« COO MwH-CaO OU>-COO
IZ3 RUN 01
                                0.1-

                                0.7-

                                O.i

                                0.3-

                                0.«

                                0.3

                                0.1-
                                   PCDF- 305 ng/dscm at 3% O2
                                         -0JSL
                            RUN 02
                                     OMT TCOT Mm«-car

                                     RUN 0«
               Site WRI- A (Wire and Transformers)
                                OJ

                                03
                                   PCDF - 866 ng/dscm at 3% O2
  UTITCSO o*^ reap
                                 «?• rcor
                    RUM 03
                           E3 RUN 04
                                    BZIRUNM
    Dioxin/Furan Homologue Distributions of Controlled Emissions from
                  Metals Recovery Processes
     (Sites MET - A, WRI - A Wire Only, WRI - A Wire and Transformers)

                        Figures- 12.
                           6-36

-------
     1.   In general, reasonably good precision was found between test
          runs for a given site.  This implies that each test site has a
          characteristic PCDD/PCDF "fingerprint."  However, these fingerprints
          could not be readily tied to specific precursors in the feed
          materials.
     2.   A wide variation in PCDD/PCDF fingerprints was found between the
          Tier 4 test sites.  This has important implications in health risk
          analysis since each homologue has a different unit risk factor.
     3.   The homologue distribution of PCDD's was not necessarily similar
          to that of PCDF's for a given site (e.g., high octa-CDD's did not
          imply high octa-CDF's for the black liquor boilers).
     4.   In general, the homologue distributions did not shift significantly
          across the control devices for which inlet/outlet data are available
          (e.g., compare the inlet homologue distribution for Sites SSI-C,
          BLB-A, BLB-B, BLB-C, WFB-A, DBR-A, and CRF-A to the outlet homologue
          distribution for these sites).  This implies that there do not
          appear to be significant differences in the controllability of
          emissions of the various PCDD/PCDF homologues.
     5.   There appear to be some similarities in the homologue distributions
          for source categories that were tested more than once in the Tier 4
          program (i.e., sewage sludge incinerators and black liquor boilers).
          However, there are also some notable differences from source
          categories.  For example, the inlet homologue distribution for Site
          BLB-C (West Coast pulp mill) is noticeably different from that for
          Sites BLB-A and BLB-B (East Coast pulp mills).  These differences
          may be attributable to variations in feed materials within source
          categories (e.g., different types of wood).

6.4  CONTROL DEVICE EFFICIENCY RESULTS
     This section presents air pollution control device PCDD/PCDF removal
efficiency results for the seven Tier 4 test sites for which sampling and
analysis were performed upstream and downstream of a control  device.  There
were no studies identified in the literature that reported control device
                                     6-37

-------
 efficiency data for PCDD and/or PCDF emissions.   The control  devices tested in
 the Tier 4 program include three electrostatic precipitators  (Sites BLB-A,
 BLB-B,  and BLB-C), one baghouse (Site WFB-A),  one dry scrubber/baghouse
 combination (Site CRF-A),  one water scrubber (Site SSI-C),  and one afterburner
 (Site DBR-A).
      Section 6.4.1 provides an overview of the removal  efficiency data, and
 Section 6.4.2  considers uncertainties in the results.  The  specific control
 devices tested are discussed in Section 6.4.3.   General  conclusions and
 observations are presented in Section 6.4.4.
 6.4.1  Overview of the Removal  Efficiency Data
      RUn-specific removal  efficiency data for  2378-TCDD,  total  PCDD,  and total
 PCDF are shown in Tables 6-12 and 6-13 for the seven control  devices  tested in
 the Tier 4 program.   In general,  the data indicate a wide range of measured
 removal  efficiencies  for the seven devices tested.   Average measured  removal
 efficiencies for total  PCDD's ranged from a low  of -130  percent for the
 baghouse at Site WFB-A (i.e.,  an  increase in total  PCDD  across  the baghouse)
 to  a high  of +99 percent for the  afterburner at  Site DBR-A  (i.e.,  a decrease
 in  total  PCDD  across  the afterburner).   Average  measured  removal  efficiencies
 for total  PCDF's ranged from values  less than  zero  for the  water  scrubber at
 Site SSI-C  and  the electrostatic  precipitator  at Site BLB-C to  a  high of +99
 percent  for the afterburner  at  Site  DBR-A.
      Between-run variations  were  large  for all devices tested except the  dry
 scrubber/baghouse combination at  Site CRF-A and  the  afterburner at  Site  DBR-A.
These two devices showed consistently high  removal efficiencies for all  test
runs  (i.e.,  80  to 100 percent).   Data for  the  individual  PCDD and  PCDF  homo-
logues varied considerably for  specific  test runs at  all  sites  except CRF-A
and  DBR-A, which  also showed  consistently  high removal efficiencies for  all
homologues.  Much of the variability observed for the remaining five control
devices  is  accounted for by  the relatively  large  (i.e., +50 percent)
uncertainty  in  the CDD/CDF analytical data  combined with  the lower
efficiencies of these devices.  These considerations  are  addressed  in the
following section.
                                     6-38

-------
TABLE 6-12.
EMISSION CONTROL EFFICIENCY RESULTS
(INDIVIDUAL RUN DATA)
Control Device/Test
Site/Run Number
Electrostatic Precioitators
BLB-A
Run 1
Run 2
Run 3
Average
BLB-B
Run 1
Run 2
Run 3
Average
BLB-C
Run 1
Run 2
Run 3
Average
Baqhouse
WFB-A
Run 1
Run 2
Run 3
Average
Spray Drver/Baahouse
CRF-A
Run 1
Run 2
Run 3
Average
Emission Control
2378 TCDD


ND/1,0
ND/I,0
ND/1,0

ND/1,0
ND/1,0
ND/1,0
__

ND/1,0
ND/1,0
ND/1,0


NR/I
70.9
56.3
63.6


100.0
100.0
100.0
100.0
Device Removal
Total PCDD


5.5
43.7
88.7
46.0

-103.8
75.8
98.1
(23.4)

89.4
44.3
52.6
62.1


-243.2
-137.6
-8.0
-129.6


82.0
89.8
91.7
87.8
Efficiency (%}
Total PCDF


32.0
47.3
93.7
57.7

-466.3
82.2
80.2
(-101)

97.5
74.3
-52.9
(39.6)
•

9.2
35.6
73.5
39.4


93.1
95.1
96 9
j\j • y
95.0
                  (Continued)
                     6-39

-------
               TABLE 6-12.  EMISSION CONTROL EFFICIENCY RESULTS
                         (INDIVIDUAL RUN DATA) (Continued)
  Control Device/Test
   Site/ Run Number
Emission Control Device Removal  Efficiency (%}
2378 TCDD        Total  PCDD         Total  PCDF
 Water Scrubber

     SSI-C
          Run 1
          Run 2
          Run 3
         Average

 Afterburner

     DBR-A
          Run 1
          Run 2
          Run 3
         Average
   NR/I
   NR/I
   NR/I
   99.6
   99.3
   99.5
   99.6
NR/I
27.9
51.4
39.7
99.1
98.4
99.6
99.0
  NR/I
 -21.6
  27.9
( 3.1)
  98.2
  98.0
  99.4
  98.5
Note: Values in parenthesis ( ) indicate averages calculated from positive and
      negative values.
ND « species not detected.
NR « data not reported by Troika.
/I,0 =* applicable to inlet and outlet sampling locations.
/I » applicable to inlet sampling location.
/O » applicable to outlet sampling location.
                                       6-40

-------
        TABLE 6-13.  AVERAGE EMISSION CONTROL EFFICIENCIES AND INLET
                     CONCENTRATIONS FOR TOTAL PCDD AND TOTAL PCDF
                               Total PCDD
                                      Total  PCDF
Control Device/
  Test Site
                    Average                       Average
     Inlet         Measured        Inlet         Measured
  Concentration   Efficiency    Concentration   Efficiency
(ng/dscm @ 3% 02)   (Percent)  (ng/dscm 0 3% 02)  (Percent)
Electrostatic Precipitators
    BLB-A                    1.8
    BLB-B
    BLB-C
     17.1
      9.0
  46
 (23)
  62
 1.5
 1.1
15.1
   58
(-101)
   40
Baqhouse
    WFB-A
      102
-130
                                                           154
                 39
Spray Drver/Baqhouse
    CRF-A
     28.8
  88
                                                          70.1
                 95
Hater Scrubber
    SSI-C
      101
  40
                                                           507
               (3.1)
Afterburner
    DBR-A
      687
  99
                                                         2,170
                 99
Note: Values in parentheses (  ) indicate averages calculated from positive
      and negative values.
                                         6-41

-------
 6.4.2  Uncertainty Analysis
      An uncertainty analysis for control  device efficiency calculations is
 detailed in Appendix A.4.   Assuming an uncertainty of ±50 percent for a given
 analytical  result, the uncertainty analysis indicates the following
 conclusion:
           the minimum and maximum bounds (Emin,  Emax)  on the "true"  efficiency
           corresponding to a given measured efficiency (E    )  for an
                                                         'meas'
           individual  CDD/CDF homologue are expressed by the relationships:
                        max
                       "min
                              (200 + Emaac)/3,    and
                              3E
'meas
      meas
     - 200
      In  practical  terms,  these  relationships  suggest  that  measured  control
efficiencies greater  than 67  percent  represent  definite  positive  control
(i.e., Em..n >  0),  measured efficiencies  between -200  percent  and  67 percent
represent potentially positive  or  negative  control  (i.e.,  Em,v  >  0
                                                           luaX
but Em.jn < 0),  and measured efficiencies  less than  -200  percent represent
definite negative  control.  With the  exception  of the dry  scrubber/baghouse
combination at Site CRF-A and the  afterburner at Site DBR-A,  the  control
devices tested in  the Tier 4 program  generally  exhibited measured control
efficiencies in the intermediate range (i.e., -200  <  E < 67).   This implies
that  these devices were only marginally effective or  potentially  not effective
at all in reducing CDD/CDF emissions  under  the  conditions  tested.   The
consistently high  measured control efficiencies for the  dry scrubber/baghouse
combination at Site CRF-A and the  afterburner at Site DBR-A indicates that
these control  devices  were very effective in reducing CDD/CDF emissions.
6.4.3  Evaluation  of  Specific Control Devices
      In this section,  each  of the  control device types tested is  considered in
more detail.
     6.4.3.1   Electrostatic Precioitators (Sites BLB-A.  BLB-B.  and  BLB-CK
PCDD/PCDF control  efficiencies for three electrostatic precipitators were
measured in the Tier 4 program  (Sites BLB-A, BLB-B, and  BLB-C).   All of the
ESP's were used to control particulate emissions from black liquor  boilers.
As shown in Table  6-12, these ESP's operated at average  outlet  temperatures
                                     6-42

-------
 ranging from 146°C to 175°C (295°F to 350°F),  and each was designed for high
 particulate removal  efficiency (salt cake recovery).   The units at Sites BLB-B
 and BLB-C are dry bottom ESP's,  and the unit at Site  BLB-A is a wet bottom
 ESP.
      Viewed as a whole,  the removal  efficiency results in Table 6-12 and
 Table 6-13 suggest that  a limited degree  of control  (i.e.,  60 percent or
 less) was achieved across the  ESP's  for both PCDD's and PCDF's.   Negative
 efficiency values were calculated for some homologues and for some test runs,
 but the measured efficiency values were in the range  where analytical
 uncertainties could  have caused  these as artifacts of the data rather  than
 indicating true  results.   Another problem confounding the evaluation of the
 ESP removal  efficiency data is the relatively  low PCDD/PCDF  inlet
 concentrations associated with the black liquor boilers.   In  many cases,
 measured  concentration values for specific homologues were at or  near  (i.e.,
 within  a  factor  of 3  or  4)  the analytical  detection limits.
      6-4-3-2   Baghouse  (Site WFB-AK   PCDD control efficiencies  were  measured
 for one baghouse in the  Tier 4 program  (Site WFB-A).   The baghouse was  used  to
 control particulate emissions from a  boiler firing salt-laden wood.  As  shown
 in  Table  5-14, the baghouse operated  at  an  average temperature of
 approximately 224°C  (435°F) during the test  runs.
      The  removal  efficiency results in Table 6-12 and  Table 6-13  suggest  that
 the baghouse exhibited negative control  for  PCDD's (i.e.,  increases  in
 concentration across  the  baghouse) and positive  control for PCDF's.  The
 average measured  total PCDD removal efficiency of the  baghouse was  -130
 percent,  and the  average measured  total  PCDF removal   efficiency was +39
 percent.  The measured control  efficiency values for  individual PCDD
 homologues were  typically within the range explainable by  low positive removal
 efficiency and the ±50 percent analytical uncertainty.  However, the
 consistently negative values indicate little or no control as  a likely
 explanation of the data.   The measured control efficiency values for PCDF's
 indicate a limited positive degree of control.  Inlet  concentrations to the
baghouse were large relative to the minimum detectable levels.
                                     6-43

-------
     6.4.3.3  Spray Drver/Baqhouse Combination.  PCDD/PCDF control
efficiencies for one spray dryer/baghouse combination were measured in the
Tier 4 program  (Site CRF-A).  The spray dryer/baghouse was used to control
particulate and HC1 emissions from a carbon regeneration furnace equipped with
an afterburner.  In the spray dryer, an alkali solution (sodium carbonate) was
injected into the hot gas stream leaving the afterburner.  The cooled gas
stream was then passed through the baghouse, which operated at an outlet
temperature of  177°C (350°F) during the test runs.  Sampling and analysis were
performed downstream of the afterburner at the inlet and outlet of the spray
dryer/baghouse  combination.
     The removal efficiency results in Table 6-12 and Table 6-13 show that the
spray dryer/baghouse combination was very effective in reducing PCDD/PDCF
emissions leaving the carbon regeneration furnace afterburner.  The average
measured total  PCDD removal efficiency of the spray dryer/baghouse was +87.8
percent, and the average measured total PCDF removal efficiency was +95.0
percent.  There was little variation between test runs for either PCDD or PCDF
removal efficiency.  The measured efficiency values were high enough to safely
conclude that the spray dryer/baghouse combination provided positive control
of PCDD and PCDF emissions, even when potential analytical uncertainties are
considered.  Inlet PCDD and PCDF concentrations to the spray dryer/baghouse
were large relative to the minimum detectable levels.
     6.4.3.4  Water Scrubber.  PCDD/PCDF control efficiencies were measured
for one water scrubber in the Tier 4 program (Site SSI-C).  The scrubber was
used to control particulate and organic emissions from a multiple hearth
incinerator burning sewage sludge.  As shown in Table 5-12, the outlet gas
temperature from the scrubber was approximately 22°C (72°F).
     The removal efficiency results in Tables 6-12 and 6-13 suggest that the
scrubber provided a limited degree of control for PCDD's but little or no
control for PCDF's.  The average measured total PCDD removal efficiency was
+40 percent, and the average total PCDF removal efficiency was 3 percent.  The
measured control efficiency values were low enough that the true values were
most likely obscured by analytical uncertainties.  Inlet PCDD and PCDF
                                     6-44

-------
concentrations to the scrubber were large relative to the minimum detection
levels.
     6.4.3.5  Afterburner.   PCDD/PCDF control efficiencies for one natural
gas-fired afterburner were measured in the Tier 4 program (Site DBR-A).  The
afterburner was used to control hydrocarbon emissions from a drum and barrel
reclamation furnace. As shown in Table 5-14, the afterburner operated at an
average outlet temperature of 827°C (1520°F) during the test runs.
     The removal efficiency  results in Tables 6-12 and 6-13 show that the
afterburner was very effective in reducing PCDD/PCDF emissions from the drum
and barrel reclamation furnace.  The average measured total PCDD removal
efficiency of the afterburner was +99.0 percent, and the average measured
total PCDF removal efficiency was +98.5 percent.  There was little variation
between test runs for either PCDD or PCDF removal efficiency.  The measured
efficiency values were high enough to safely conclude that the afterburner
provided positive control of PCDD and PCDF emissions, even when potential
analytical uncertainties were considered.  Inlet PCDD and PCDF concentrations
to the afterburner were large relative to the minimum detectable levels.
6.4.4  Summary of Control Efficiency Observations
     The PCDD/PCDF emissions control data discussed in the previous sections
show a wide range of.PCDD/PCDF removal  efficiencies for the control  devices
tested.   The findings of the control efficiency data can be summarized as
fol1ows;
     o
The afterburner at Site DBR-A exhibited good control of PCDD's (99.0
percent) and PCDF's (98.5 percent) when subjected to high inlet
concentrations of these species (687 and 2,170 ng/dscm @ 3% 02,
respectively).
The spray dryer/baghouse combination at Site CRF-A showed good
control of PCDD's (88 percent) and PCDF's (95 percent) when
subjected to intermediate concentrations of these species (29 and
70 ng/dscm @ 3% 02, respectively).
The electrostatic precipitators at Sites BLB-A, BLB-B, and BLB-C
exhibited a limited degree (i.e.,  less than 60 percent) of control
for both PCDD's and PCDF's although the data are not conclusive.
                                     6-45

-------
          The inlet PCDD and PCDF concentrations to these devices were low
          (maximums of 17 and 15 ng/dscm @ 3% 02 for PCDD and PCDF,
          respectively).
     o    The water scrubber at Site SSI-C exhibited a limited degree of
          control for PCDD's (approximately 40 percent) and little or no
          control for PCDF's (3 percent) when subjected to intermediate inlet
          concentrations of these species (101 and 507 ng/dscm @ 3% CL,
          respectively).
     o    The baghouse at Site WFB-A exhibited negative control of total
          PCDD's (-130 percent) and a limited degree of control for total
          PCDF's when subjected to intermediate inlet concentrations of these
          species (102 and 154 ng/dscm @ 3% 02, respectively).
     Absolute comparisons of the removal efficiency data between source
categories are not quantitatively valid because the control devices were
subjected to varying inlet flue gas conditions (e.g., widely varying
PCDD/PDCF, HC1 and moisture concentrations).  However, several general
observations regarding control of PCDD and PCDF emissions can be made.  These
observations are discussed below.
     Perhaps the most important conclusion to be drawn from the Tier 4 control
efficiency data is that effective particulate emissions control does not
necessarily imply effective PCDD/PCDF control.  This is most clearly
demonstrated by the baghouse at Site WFB-A, which actually showed an increase
in PCDD concentration across the control device.  The data suggest that
PCDD/PCDF formation/destruction reactions can occur within the control device.
Baghouses would seem ideally suited for this type of mechanism, since
continuous flue gas/particulate matter contact is established at the bag
surface.  Reactive chlorine-containing species (e.g., HC1) present in the flue
gas may interact with compounds present in the particulate matter to produce
PCDD's.and PCDF's.  In contrast to the baghouse at Site WFB-A, the spray
dryer/baghouse combination at Site CRF-A showed good control for both PCDD's
and PCDF's.  Two possible reasons for this difference are 1) the water and
alkali solution injected in the spray dryer, and 2) the relative temperatures
of the two baghouses.  The first hypothesis is consistent with the control
                                     6-46

-------
 device  reaction  mechanism suggested  above.   The  purpose  of the  alkali  spray  at
 Site  CRF-A  is  to remove  HC1  from  the flue gas  (as  Nad)  prior to  particulate
 control  by  the baghouse.   Thus, the  reactive HC1 species is not available  to
 participate in potential  flue  gas/particulate  matter  reactions  on the  bag
 surface.  -Also,  the gas  cooling provided by  the  spray dryer [(204°C  to 927°C
 (400  F  to 1700°F))] may  provide nucleation sites for  vapor phase  PCDD/PCDF
 condensation onto particulate  matter surfaces, which  would also be expected  to
 improve  PCDD/PCDF removal  efficiency across  the  spray dryer/baghouse
 combination.
      The relative temperature  of  the baghouses at  Sites  CRF-A [(177°C
 (350°F))] and  WFB-A [(224°C  (435°F))] may also play a role in the relative
 PCDD/PCDF control  effectiveness of these two devices.  However, a temperature
 differential of  this magnitude would not be  expected  to  cause such a large
 difference  in  control efficiency.  The data  developed for  the water  scrubber
 at Site SSI-C  indicate that  gas cooling alone  does not necessarily imply
 effective PCDD/PDCF control.   The outlet temperature  of  the water scrubber
 [(22°C  (72°F))]  was the  lowest of all the control devices  tested,  yet  the
 water scrubber did not provide good  PCDD/PCDF  control.   The operating
 temperature/removal efficiency data  for the  three electrostatic precipitators
 tested seem to fit the trend of decreasing PCDD/PCDF  control efficiency with
 increasing temperature.  However, the validity of this trend may  be  obscured
 by the variability in the measured efficiency data for these sites.
     Another important observation that can  be made from the Tier  4  control
 efficiency data  is that properly operated afterburners can  be very effective
 PCDD/PCDF control devices.  This is  clearly demonstrated by the data from  Site
DBR-A, which exhibited average afterburner removal  efficiencies of 99.0 and
98.5 percent for PCDD's and PCDF's,  respectively.
     Homologue distribution shifts across the control  devices tested were  not
found to be significant.   This is illustrated by qualitative comparisons of
the inlet/outlet homologue distribution pairs in Section 6.3.  Also,  large
differences were not found between control  efficiencies for PCDD's and PCDF's.
The exception to this  rule was the baghouse at Site WFB-A, which showed poorer
control  for PCDD's than for PCDF's.
                                     6-47

-------

-------
                                    CHAPTER 7
                            QUALITY ASSURANCE PROGRAM
      This section discusses the Quality Assurance and Quality Control
  (QA/QC) program implemented for the Tier 4 study.  The objectives of the QA
•  program and a summary of the QC activities are discussed.  An assessment of
  data quality for key measurement parameters is also provided.  Quality
  assurance support to the Tier 4 program was provided by Research Triangle
  Institute (RTI).  Results of their quality assurance effort are presented in
  the document entitled "National Dioxin Study Tier 4:  Combustion Sources."
  Quality Assurance Evaluation.  EPA 450/4-84-014f.
      Throughout this chapter, the Tier 4 test sites are designated by
  numbers (01 through 13), corresponding to the chronological sequence in
 which the tests were performed.  This was done so that changes occurring
 during the course of the program (e.g., sampling protocol, data quality,
 etc.) could be easily tracked for the various test sites.  A table matching
 the site codes and the test numbers for the various test sites is provided
 in Table 7-1.

 7.1  QA PROGRAM OBJECTIVES
      For any measurement effort,  there always exists some degree of
 uncertainty associated with the measured data due to inherent limitations  of
 the measurement system.   The utility of the measured data is  largely depen-
 dent on the degree to which the magnitude of this uncertainty is known.  The
 Tier 4  testing  program incorporated a comprehensive  QA/QC plan.   The overall
 objective  of this  plan was  to  produce complete,  representative,  and
 comparable data of known quality.
      The Tier 4  QA/QC program  emphasized 1)  adherence to  prescribed  sampling
 procedures,  2)  careful  documentation  of sample collection  and  field
                                     7-1

-------
TABLE 7-1.  TEST SITE NUMBERS AND COMBUSTION DEVICE CODES

Site
Number
01
02
03
04
05
06
07
08
09
10
11
12
13
Site
Code
SSI -A
ISW-A
SSI-B
BLB-A
BLB-B
WRI-A
WFB-A
BLB-C
CRF-A
MET-A
DBR-A
SSI-C
WS-A
                            7-2

-------
 analytical  data,  3)  use of chain-of-custody records,  4)  adherence to
 prescribed  analytical  procedures,  and 5)  implementation  of independent
 systems  and performance audits.   In  combination,  these QA/QC efforts served
 two  purposes.   First,  the QA/QC  efforts  provided  the  mechanism for
 controlling data  quality to within acceptable  limits.  Second,  they formed
 the  basis for  estimates of uncertainty by providing the  necessary
 information for defining error limits associated  with the  measured data.
      The QA objectives for each  key  parameter  measured during the Tier 4
 test program are  shown in Table  7-2.   These objectives were  based largely on
 the  accuracy and  precision typically achieved  for the methods selected.  For
 all  GC/MS analyses  (MM5 CDD/CDF,' chlorobenzenes,  chlorophenols,  and
 polychlorinated biphenyls)  standard  extraction and analysis  procedures were
 initially used.   Difficulties were encountered for certain feed  materials
 and  MM5  samples containing high  levels of contamination.  Therefore,  even
 though extraction/cleanup procedures  were modified to improve spiked
 surrogate recoveries,  QA objectives were  still not met for some  samples.
 Despite  the  fact  that  QA objectives were  not achieved for these  samples, the
 analytical results obtained  are  still considered  adequate to  address  the
 objectives of Tier 4.   Site  WS-A (13) is  the only  site for which  CDD/CDF
 emissions data were invalidated due to QA concerns.  Surrogate recoveries
 for  the MM5  samples were well below the QA  objectives for this site.
     The QA/QC activities  for the Tier 4  program-are summarized  in  Section
 7.2  and data quality assessments are  provided  in Section 7.3.

7.2  SUMMARY OF QA/QC ACTIVITIES
     This section provides a summary of QA/QC activities and results  for
three main areas:   field sampling,  field  and laboratory analysis, and
independent audits.   Additional  detail on the activities associated with the
                                     7-3

-------
I







I/I
£
s
i

t



i
**
Ul

H"



>_
1

w

«N
1
i
*—





















y

U



i

|
i





















' -o

4-*
*
















t_
Q
I
re
Q.



1 % 1 f, 1 1 §i 1 § 1 » «i § §





*8 ""g 'g 8 8 8 § 8 8 8 8 8 8 -"g







i
—
0)



mm o
N N 1-
r ** L. 0
^ to *to N o tn

•o m U££«'«S
«O 13 N to te C Z

1X1 -a£ "leccocQ.
•a co e n " o _o •

125e5o,g.S.!::5|
^ < £ <~ £ s — — o § o a _" ;g


££uizuiQ.zza.u.oc9U>U)






*
•S £


o re >•
I "" o -c
U U  o ocr^ta^
OC>U oOOt/Jh-NC-^O

Ul
O -M
> tn
•V- U
4J •*•
U IL
O
1 ,

i-
u
y


•a
c

c
Q

Ul

8
U
15


tl
^*
e
re ,

1
>-
§
re
CO
re £












\
< i


U)
a
re i—
> A3

C >-

U =
ID *J

, <
1
tn o
L. =

£ £

•§
O 1-
: ^
_
i

in
fl>
n
a
e
re

a>
j
re
D
3-
3


£
^
C
E
0
5
+J
c
CD
U
u
o
°- c



I
>
ffl
5
•a
CD
g i
o a
X 1
09
q =
^a ^
1 ?
CD t-
C C =
re u
•c b «
5 is1 i
3 ~C
1 C
A3
X 4J »
Ul '
*
5 5
in
(D (D
•M t-
O (D
3 (D .e
^ > it
re —
re o
^ u ^- •<
re o o.
3 XI E
•M re re <•
u »— u>
*O f-
0 p
l_
< V
"j 1 1
° -5
tn
i 4-» in
- i i
£a> t—
L. re
(D C
c c re
» 0
• "D i—
tn o *.
Sm re
re T)
>» J3
— t.
re T3 o
g s -
s 1 i
3 In "
re (D c
O
D 0 5
*- re re
t_ t-
-^ 3 U
K U «
-re >
J5 ? o
t- « 4J

-------
sampling and analytical efforts is provided in the Tier 4 Quality Assurance
Project Plan (QAPP)A and the Dioxin Analytical Procedures and Quality
Assurance Plan for Tier 4.
7.2.1  Sampling Activities
     Quality control for flue gas and process sampling centered around
1) equipment calibration, 2) glassware and sampling equipment cleaning,
3) procedural QC checks, and 4) sample custody procedures.  Key activities
and QC results in each of these areas are discussed below.
     Pre-test calibrations or inspections were conducted on pi tot tubes,
sampling nozzles, temperature sensors and analytical balances.  Both pre-
and post-test calibrations were also performed on the dry gas meters used
for MM5, HC1, and ambient train sampling.  All of this equipment met the
calibration criteria specified in the Tier 4 QAPP.  Differences between pre-
and post-test dry gas meter calibrations for each of the 13 tests were less
than 4.3 percent.
     An extensive pre-cleaning procedure was implemented for all sample
train glassware, sample containers and sampling tools.  This cleaning
procedure, which is outlined in Table 7-3, was implemented to minimize the
potential for sample contamination with substances that may have interfered
with the analysis for CDD's and CDF's.  All sample train glassware was
capped with foil prior to use and stored in a dust-free environment.  A
clean sample trailer was maintained for train assembly and sample recovery.
To document the effectiveness of the glassware pre-cleaning procedure, a
sample train "proof blank" was obtained for each site beginning with Site 04
      National Dioxin Study—Tier 4 Combustion Sources.
Project Plan.  EPA-450/4-84-014e.
                                                         Quality Assurance
      Analytical Procedures and Quality Assurance Plan for the Analysis of
Tetra Through Octa Chlorinated Dibenzo-p-Dioxins and Dibenzofurans in
Samples from Tier 4 Incineration Processes.  Addendum to:  "Analytical
Procedures and Quality Assurance Plan for the Analysis of 2378-TCDD in Tier
3-7 Samples of the U. S. Environmental Protection Agency National Dioxin
Strategy."  EPA/600/3-85/019, April 1985.
                                     7-5

-------
              TABLE 7-3.  TIER 4 GLASSWARE PRECLEANING PROCEDURE
Soak all glassware in hot soapy water  (AlconoxR) 50°C or higher.
Distilled/deionized HgO rinse  (X3).a
Distilled/deionized H20 rinse  (X3).
Chromerge  rinse if glass, otherwise skip to 6.
High purity liquid chromatography grade H20 rinse (X3).
Acetone rinse (X3), (pesticide grade).
Methylene chlorideb rinse (X3), (pesticide grade).
Oven dry (110°C - 2 hrs).
Cap glassware with clean glass plugs or MeClg-rinsed aluminum foil
1.
2.
3.
4.
5.
6.
7.
8.
9.
a(X3) - three times.
 For Sites 01 through 04 hexane was used as the solvent for this last rinse.
                                       7-6

-------
0
0
0
0


0
0
 (BLB-A).   The  proof blank was  obtained  from a  complete  set  of MM5  sample
 train  glassware  that had  been  cleaned by  the procedure  in Table  7-3.   The
 pre-cleaned  glassware was rinsed  with the sample  recovery solvents (acetone
 and methylene  chloride) and  the rinses  were analyzed  for CDD/CDF.   Results
 of the proof blank analyses  are presented in Table  7-4.
     Procedural  QC activities  during manual  gas sampling for  CDD/CDF  and HC1
 focused on:
     o    visual equipment inspections,
           use  of sample train  blanks,
           ensuring the proper  location  and  number of  traverse  points,
           conducting pre-test  and post-test sample  train leak  checks,
           maintaining proper temperatures at the  filter housing, sorbent
           trap,  and  impinger train,
           maintaining isokinetic  sampling rates,  and
           recording  all data on preformatted field  data sheets.
     Results of  isokinetic calculations and  leak  checks for MM5  and HC1
 sampling at the  Tier 4 sites are presented  in Table 7-5.  The  QA objectives
 for each test  run  were a  leak  free sample train and an average isokinetic
 sampling rate  of 100  +10  percent.
     As shown  in Table 7-5, the isokinetics  objective was exceeded for at
 least  one  MM5  test run at  Sites 01, 02, 06,  08 and  12 and for  at least one
 HC1 test run at  Sites  02,  04,  05, 06 and 07.  Of  the 64 MM5 test runs, nine
 test runs  or 14  percent exceeded the isokinetics  QA objective.   For the HC1
 train,   13  of the 33  test  runs  were outside  the target 100 +10 percent
 isokinetic sampling  rate.  In  most cases, problems  in achieving  the
 isokinetics QA objective were  caused by the  normal but wide fluctuations in
 stack  gas  flowrates  and moisture contents typical  of batch processes.  This
was particularly true  for  the  batch type combustion processes tested at
Sites  02 (ISW-A) and 06 (WRI-A).  In other cases the isokinetics objective
was exceeded because of unexpected changes  in the flue gas moisture content
between test runs.
     Sampling outside the  isokinetics objective potentially affects the
total  mass and size distribution of particulate matter collected on the
                                7-7

-------






CO
Z
§
U-
8
rt*
in
g
a

o
UJ
HH
u.

Q
Q
j^
^

ce
HH


g
U.
CO
t-
d
CO
s
CO
CO



^t"
I
III

«p
L«
*""*


















•p
§
u
CD
Q_
•>
U)
CD

t_
CD

O

cc

CD
jj
•t^
nJ
0)
o
u
23
CO























•p tn
••- i-H
CO

CD
-P CM
I— I-H
CO

CD f-H
•P «-H

CO

CD
•P 0
•«— i-H
CO


CD
£§
CO




CD
£g
CO



CD
•P r-
•r- O
CO


CD
£8
CO


CD
•p in
•*- o
CO


CD
£S
CO





c
3
0
Q.
Q



Z Z Z Z



CO ^3 ^Q ^^
z «-H r» r-
i-H

z z z §



CM CM OO m
OO CO O O^
f-H



CM C3 ^O O^
eo oo >-H r>-
i-H





«c «c «c  o~v — i
ZZZZOOZOOi-HCM^H






zzzzzzzzzzzz





zzzzzzzzzzzz




zzzzzzzzzzzz



QQQQQQQQQQQQ
ZZZZZZZZZZZZ


in
Q a a Q a >oaoooo
zzzzzozzzzzz






go.
Q
O O O U_
i=- a a o i- Q u. u.
eo P S g o g i ^ o g 8 g
r- t-  CD CD CD U » -P CD CD CD U
cMoa.xxocMoa.a:xo
*
c
10
s_
-p
CD
a.
(O
(0

o
CL
1
^»
C
(0
c
o
o
**
*— (
*
o



*•
M"
•o
CD
O)
c
10 •
S- CD

c O.
• 0 E
(D «^  j=
T3 +J
»
10 in j£ => i—
in n- E ^ c
•p -P o a o
(o (o CL a.
g-p s
o o
CD CO O Z U
JiC J£
•^ •»- II II II
a. a.
CO CO CO Q <

-------
o
       u
       u
       u
   x  11
i: o  n
< LU  n
LU X  II
_f r^  [t
       0
       U
       II
       U
                   II
                UJ II
             U > II
             O M II
             V) O
                   II
O
U_
                               o o o
                               O O O
                               O O O
                               S3 CS CD
O Z Z O    O 00
O 1-1 "-i O    OO O
o < < o    oo o
cs a a: o    CD o cs
                                            cn cn
                                            tu UJ O O
                                            >- >- z z
                                                                       o a o a
                                                                       o o o o
                                                                       o oo o
                                                                       o o CD o
                                                         CO CO

                                                         LU UJ O O
                                                         >- >- Z Z
                                                                         o o a
                                                                         oo o
                                                                         o o o
                                                                         O Cfl C3
                                          CO
                                          UJ O O
                                          >• z z
                                                       o o o a o o o
                                                       o o o o o o o
                                                       o o oo o o o
                                                       . z
                                                                0 O 0
                                                                o o o
                                                                o o o
                                                                o a o
                                                               co    cn
                                                               LU O UJ
                                                               >- Z >
             oo o
             oo o
             oo o
             (g (9 (9
             cn cn co
             UJ UJ UJ
             >- J- >-
o a o    o o o
o o o    o o o
o o o    oo o
CD CD CD    eg eg eg
                                                                                                                                    co cn co
                                                                                                                                    LULU LU
                                                                                                                  cn cn cn
                                                                                                                  LU LU LU
                                                                                                                       • >•
—1
GO
UJ
a:
^
o
LU
:5
^
<:
LU
««l

Q
<
GO
O
I-H
_J UJ
U Z
o
IM

O UJ
o >•
HH
CO h-
LU UJ
UJ *"j
£ CD
O
o
1— t
UJ UJ
_J Z
Z HH
HH ^
* m m V'S,>N. r* «"o in ai cv m
ZZZ 00^*-4 ZZZ OVO~H~* 0 ^ ^




u
>- >- zzzz zzz >- >- > z >->->-



.
1
i co ov GO in p^ r-» ov co »o
j to cn ^ x. >* x. x. x, x. x. CM eo v x. m CN fsi
i o^o^*-< zzzz zzz oiaioz ooo
r^mCMOkCNir^o ^tn^o oo ^*f-
zzzzzzz ooo o2o
o «r at in oo CM
w d o» ov o o
•"* *^




to t/> co < < <
UJ UJ UJ X. X. X.




?n ^o o*
in ^ r* x, x. x,
ooo zzz
             O LU U
             I— O U
             Z CO
                o
                                           cn co o o
                                           UJ UI Z Z
                                                                       CO CO CO

                                                                       UJ UI LU '
                                                                                    o o o o co cn <
                                                                                    Z Z Z Z UJ UJ "V
                                                                                                >• >- z
                                                                              co cn co
                                                                              LU UI UI
                                                                           co cn cn
                                                                           LU UJ UJ
                                                                                                                                   cn co cn
                                                                                                                                   LU LU UJ
                                    co cn co
                                    LULU UJ
               CO u

               U II
            I- i-l II
             =5 •-
             O ^
                              tn *r m
                              o> o> o
                                            t m CM
                                            o o •-• •
                                                           I ** 1C


                                                           . «3 p.
                                                                                    o co i—    in m ~* >-• CM in
                                                         _* KO CO -X
                                                         cn oo z
                                          co ^ *n
                                          cn O o
                                       o ^< o ^* o m x.
                                       CM •-< — i M o O Z
                                                                                                            ^•^1*^    vmcM    cno^    ^^CMr^-

                                                                                                            o tn —    mincn.   OICMO    r>CMm
            en ut cn.
            o o o
                                                                                                                                   cncM o
                                                                                                                                   ooo
—J UJ  H
O O  II
cr i-<  ii

z LU  u
o o  u
o     u
LO
 I
I—


LU

CD
                              ce.
                              LU
                              CD
                              m
ce
UJ
z
ce

m
ce
UJ
ee
UJ
m
ca

ce
u
                                                                     a.
                                                                     cn
                                                                                                 ui
                                                                                                 z
                                                                                                m
                                                                                                e
UI
cn

O
^
eg

ca
                                                                                                                                   LU
                                                                                                                                   CO
                                                                                                                                   a
                                                                                                                                   o
                                                                                                                                   3:
                                                                                                                                   CD

                                                                                                                                   m
                                                                                                                                                             0
                                                                                                                                                             X
                                                                                                                                                             cr
                                          rt CM m ^    -^
                                                                     *-• CM m v    rt CM m    ••< CM m «r in 10 r-    r* CM m    -< CM
                                                                                                                                                — i CM m    — i CM m
            t— UJ
            co i—
            UJ HH
                              <
                               I
                                                          CO
                                                           I
                                                         CO
                                                         cn
                                                                      m
                                                                      _j
                                                                      co
                                                                        CO
                                                                         I
                                                                        m
                                                                                                                          i
                                                                                                                         CO
                                                                                          o

                                                                                          m
                                                                                                                                    ce
                                                                                                                                    u
                                                                                    7-9

-------
































0
LU
ra
|_,
1—
CD
0

in
i
r--.

UJ
i
CO
?


























M O
<: ut
Ul X
-JO

u
o >
a -i

CO O
t-UI
uin
Ul CD
Z O


CO
o
to*
H-
— 1 Ul
0 Z
X 1-1
g
CO
_
001
o >•

COI-
t-0
Ul Ul
Ul ">
I m
o

0
t-H
_! Z
Z M
"I
1-4
OLU
O ^
coi-
1-0
Ul 111
Ul t-y
a: m
o

to
O
Ul t-
ss
o
to
*"*


••1 til
00
cr I-H
I— >
ZUI
O Q
0




1- Z
CO O
Ul cr



t-UI
Ul 1-"
t- to
•o
Ul
— 1
O Q O ZOO Q Q Q
OOO MOO O O O
OOO < O O OOO
CDC9O CC O O C9C9C9
1—
Ul



tototo <<<< <<<
UIUIUI %vX.**^ x.*^-^
>•>->- zzz zzz







^- ^-co
• * • * en en co
C* CD CM "XX.X. O 0> 0)
OOO ZZZ
r^ ^4 *H




en en en en •>>>- >•>->- zzz





eo r* in ^t r- r-
CO O CD in O *O - >• >- >- ZZZ




oo  •
o 9 o en oo o




LU tX
Z LU
o: m LU
= m z
m 3 . o
C£ 0* z
LU O
h- en
LU
<

r-«CNrn «-i CM m «-icNrn





< o
i i t
m to to
o to x
^ ^ >»
—• CM m



^
L.
C t<
O 0)
^~
4* 
to «-
4i 'I

U
•M ?
O 
u en o
t- E +•>
o co in
!o o
OOO
1-1 Q,i
je ro
O 0 r-l
o O
U 4-"
o ^ to
10
> r~ .C +»
•^ *» ID
4«* U- &•
CJ O O)

••"> C »- O
J3 O «- •—
O •*- 3 *k
4-> -0
co O ID en
r— O C CD ID
•O ID "^y u
<0 +* r- ID
•0 O ID CL m CD
CD — .£= X t-
i— »— +J CD en ^
CL CL ID ID
E Q. en L. B —
ID ID CO O U
en -^ <•- 4-> ID
-P r- CL >
4J o CL-4-) ^
o c e x 5 -o
C -r- CD +J C
II +J 4-> ID
II = ID
•C O CD B
CO -^ Z CO O O
Z Z = to Z i—
ID J2 O -O 0)
7-10

-------
 filter.   However,  the measured exceedances in  the isokinetics QA objective
 were generally small  (i.e.,  an additional  +10  percent  above the QA objective
 of 100 +10 percent),  and are not expected  to have had  a significant effect
 on measured concentrations  of CDD/CDF or HC1.   Estimates for chlorine
 emitted  with particulate matter based on the HC1  sampling trains may be
 somewhat in error  for cases  where the isokinetics QA objective was not
 achieved.  However,  no test  data were  invalidated  on  the basis of poor
 isokinetic sampling rates.
      Initial,  final,  and port change  leak  checks  for the MM5 and HC1  sample
 trains were performed for each test run.   The  only leaks detected were at
 Sites 02 and 12.   At  Site 02,  the glass  probes on the  HC1  trains we're found
 to be broken at the end  of  test Runs  2 and 3.   The actual  time that the
 breakage occurred  during sampling is  not known, but  the breakage is most
 likely to  have occurred  either during the  port change  or when the train was
 removed  from the stack for disassembly.  Following a review of the chloride
 analysis results,  the HC1 data for Run 2 were  invalidated because the HC1
 values determined  for this run were unreasonably  low compared with Runs 1
 and  4.   Results for Run  3 were very similar to those for Runs 1  and 4 and  it
 appears  that this  probe  was  not broken until the  sampling was completed.
      At  Site 12, Run  1,  a leak in the MM5  sampling train was  detected during
 testing  at  the scrubber  inlet.   During this sampling period,  six ports were
 traversed  at the scrubber inlet.   A leak check performed after the first
 four  ports  showed no  indication of leakage.  However,  a leak  check performed
 after traversing the  last two  ports revealed that  the  filter  holder was
 broken.  It  is not  known  if  the breakage occurred  before,  during or after
 testing  of the last two  ports.   Rather than discard  the  entire test run, a
 decision was made to use the results  to provide a  range  of CDD/CDF
 concentrations entering the  scrubber.   The  higher  concentration  will  be
 based on the sample volume at  the  end  of the fourth  port and  the  lower
 concentration will  be based on  the volume at the end of  the sixth and final
 port.  All other test runs at  Site 12  showed no indication of leakage.
     Sample custody procedures  used during this program  emphasized  careful
documentation of the samples collected and the use of chain-of-custody
                                    7-11

-------
records for samples transported to the laboratory for analysis.  Steps taken
to Identify and document samples collected included labeling each sample
with a unique alphanumeric code and logging the sample in a master logbook.
All samples shipped to Troika or returned to Radian-RTP were also logged on
chain-of -custody records that were signed by the field sample custodian upon
shipment and also signed upon receipt at the laboratory.  Each sample
container lid was individually sealed to ensure that samples were not
tampered with.
7.2.2  Analytical Activities
     Laboratory and field analyses were conducted during the Tier 4 study to
determine 1) the CDD/CDF contents of flue gas and process samples, 2) the
CDD precursor contents of combustion device feed samples, 3) the HC1
concentrations of the impinger catch from HC1 sampling trains, and 4) the
SC
NOX and total hydrocarbons (THC) in the
concentrations of CO, CQy,
flue gas.
     Quality control (QC) activities for each of these analyses are
described in detail in the Tier 4 QAPP and the Dioxin Analytical Procedures
and QA Plan.  Activities for all of these analyses included daily instrument
calibration with certified standards to establish calibration or response
factors and to verify proper instrument operation.  Other QC activities
specific to CDD/CDF analyses, CDD precursor analyses, and continuous
monitoring of flue gas parameters are summarized in this section.  Results
of QC checks are discussed in Section 7.3.
     CDD/CDF samples were extracted and analyzed by EPA's Troika
laboratories in sets of twelve.  Internal QC activities for the CDD/CDF
analyses are summarized in Table 7-6.  These activities include the use of
labeled surrogates, sample blanks, control samples and duplicate extractions
and analyses.  Specific criteria for defining minimum detectable limits and
for confirming positive identification of 2378-TCDD, other CDD's, and CDF's
were also employed as part of the QA/QC effort.
     Quality control activities for samples analyzed for CDD precursors
(i.e., chlorobenzenes,  chlorophenols, chlorinated biphenyls, and total
chlorine) were similar to those for the CDD/CDF analyses.  With the
                                    7-12

-------




















CO
o
LU
5
g
^~
Z
8

HH
_I
1
co
HH
co
_j
z

u.
8

Q
O

Li.
O
S
§
CO



VO
1
r-
LU
§





























IO
•1—
i.
•p
U
8
c
IO
•p
Q.
(D
U
£








Ul

Ul
>t
*io
c
n-
o
(J
c
ID
3
o-
£
Lu






o
CD
x:
O

"o
L.
0
o

•p


10







ID
L.
0)
-p
E
IO
10
Q.








>,

CD

O
8
I-
CD
O

to
a.

0
r-l
O
•P

O
in












.
ID
r~
Q.
E
IO
Ul
0
IO
LU

CD

.^
0.
Ul

CD
•p
IO
o>
O
3
Ul
•o
CD

CD
XI










Q
Q
O
1-
00
r-.
ro
CM





>,
L.
CD

O
O
CD
!_
C
8
i.
2.

0
1— 1
o

o













CD
o.
E
IO
in
o
10
LU

CD

^
Q.
Ul

CD
•p
10
O)
o
Ul
T3
ID

CD
XI
IO
— 1

Ul
u.
Q
O
•o
c
IO

Ul
Q
8


¥"i
10 •»-
£^
O CD
0 •—
>»- IO
O -P
U

CD
Ul
O C
f IO
•»-> 1—
U XI
s- X
CO -P
10
fee
x: x:

f
S- ID

rmm
IO
u

^J
(D
O
CD
5


.p
c
CD
s
ID
CL
O
+1


fc
•P
CD •
Ul CD
c "o.
0 E
••- IO
•p in
u
u. "o
**
0 c
fe8
x: x:

^
t- (D
O) +j
CD C
U
CD CD
U -P
C i—
0 «



c
IO
f"—
XI

X
t_
•p
10
E
^
CD
u-

+J
O
U.
















•
CD
3

IO
>
,_
ie
u

^%
CD
1
4->
K-

4J
C
(D
U
CD
Q.
O



«
•P X

Ul t-
•P
C IO
5 E

4^ T3
0 _<£
1- >*-
X -P
ID U
O
ID
x: x:

^
£" (D
O -P
CD C •
I- ^
CD CD C
C >— r-
O IO XI









CD
i
IO
Ul

"o

4J
1























<^.

• *
•J
CD
U
C


«^.
«^-
•o
-p
c
CD
u
u
CD
Q.


.
^J

e
o

4J
U
1
CD
L.
x:
+^

>>
CD
CD
CD
j—
0



x»
IO

^^

a
i
10
Ul
ID
IO
u

^^
a.
o































4J
C
CD
U
s.
CD
Q.

0
in


IO


F—
Q.
3
T3

CD

Ul
o
U
x:
**
^
0)
•p •
c -o
u t—
CD CD
^™ ^"
«3 *^







































•f—

.j
*»
8
CD

(^ •
•»- -P
•o e
CD
•p u
C I-
CD CD
!_
CD O


•t
•P CD

Ul Q.
d
O U)

^J fli
U -P
IO IO
*" ^
CD a.
3
I- TJ
x: x:
"o ^
2 .
£•0,
CD -P
CD C •
L. «^

C r— f—

•o
5

^^

CD

IO
U
a.
3
•o
CD

"ui
o

o


































































•
t|
o
Ul
^.J

in
c
•o
CD
•P
ffl
^

X
*•*
CD
t.
CD
*
Ul
•—
££
ia
CO
IO
7-13

-------
exception of total chlorine, QC activities for CDD precursors  included
surrogate spikes, blanks, matrix spikes, and duplicates.  Sample blanks and
duplicate analyses were used as QC checks for total chlorine.
     Quality control checks for the continuous monitoring of CO, CCL, 0,,,
S02, NOX and THC included calibration drift checks and daily analysis of a
control standard.  The control standard was a different standard than that
used for instrument calibration.
7.2.3  Audit Activities
     The audit activities for the Tier 4 program consisted of  independent
systems and performance audits of both the field and laboratory efforts.
Radian performed systems and performance audits on field activities at
Site 03 (SSI-B) and on laboratory activities for CDD precursors.  Research
Triangle Institute (RTI) conducted systems audits on field activities at
Sites 02 (ISW-A), 05 (BLB-B), and 10 (MET-A).C  RTI also performed a systems
audit on the CDD precursor laboratory activities and provided  audit samples
for CDD/CDF analysis and total chloride analysis.
     Objectives of the systems and performance audits were to  1) evaluate
adherence to the test plans and QAPP, 2) document test and analytical
procedures used, 3) provide information to allow assessment of data quality,
and 4) make recommendations that could improve data quality.  The results of
field audits are presented in detail in the individual test reports for
Sites 02 (ISW-A), 03 (SSI-B), 05 (BLB-B), and 10 (MET-A).  A summary of
audit activities and systems audit results is presented in this section.
Performance audit results, with the exception of example MM5 calculations
and dry gas meter audits, are discussed in Section 7.3.
     Field systems and performance audits centered around manual sampling
methods (e.g., MM5, HC1, and grab sampling), continuous flue gas monitoring,
data documentation, and sample custody and handling.  The systems audits
showed that all field activities were being performed with strict adherence
      National Dioxin Study—Tier 4 Combustion Sources.  Quality Assurance
Evaluation.  EPA-450/4-84-014f.  January 1986.
                                    7-14

-------
to the test plans and QAPP.  Following each audit, recommendations were made
for further improving data quality but no problems serious enough to
invalidate test results were identified.
     Two modifications from the ASME MM5 protocol were noted by the RTI
auditors.  Both modifications had been previously approved by EPA.  The
first modification relates to the orientation of the condenser coil in the
MM5 train.  The ASME protocol specifies that both the condenser coil and the
resin trap be oriented vertically.  Radian has found, however, that
substitution of a horizontal condenser (but not trap) works equally well
without causing significant holdup of condensate in the coil or channeling
of the condensate through the resin.  The horizontal coil has an added
advantage of reducing the space required for traversing the sampling train.
A horizontal condenser coil was used throughout the Tier 4 test program.
     The second modification implemented for Tier 4 relates to the solvents
used for glassware cleaning and for rinsing the nozzle, probe, filter holder
and cyclone (if used).  The types of solvents used were changed several
times during the period that the first five CDD source tests were conducted.
For Site 01 (SSI-A) recovery of the glassware involved rinsing with
deionized (DI) water, acetone, and then hexane.  This procedure was changed
for Sites 02 (ISW-A) and 03 (SSI-B) to eliminate the water rinse due to a
revision of the ASME protocol (April 16,  1984 versus October, 1984 version).
Water was reinstituted as the first of three solvents for Sites 04 (BLB-A),
05 (BLB-B), and 08 (BLB-C), all of which were.black liquor boilers.  The use
of water for these particular sources was necessary to ensure recovery of
acetone insoluble particulate from the glassware.
     The last change in the glassware rinsing scheme was the substitution of
methylene chloride for hexane to improve recovery of CDD present in the
sample.  This change was implemented for all  test sites beginning with
Site 05 (BLB-B).  As will be discussed further in Section 7.3.1,  relatively
high quantities of octa-CDD's and CDF's were found in the field blank sample
train for Site 01 and the substitution of methylene chloride for hexane was
intended to improve recovery of these compounds.
                                    7-15

-------
     During two  of  the  audits,  an  example MM5 data  sheet was  provided  to the
test team to  evaluate the  computer program  used  for data reduction.  In both
cases the necessary calculations were  accurately performed.
     Dry gas  meters used for MM5,  HC1,  and  ambient  sampling were
also audited  at  three test sites.   A calibrated  orifice was used  at Sites 02
(ISW-A) and 05  (BLB-B)  to  check three  dry gas meters  used  in  the  Tier  4
study.  All three dry gas  meters agreed with the orifice flow rate to  within
±2.0 percent.  The  dry  gas meter used  at Site 03 (SSI-B) was  audited by
direct comparison to a  transfer dry gas meter which had been  referenced to
an  independent wet  test meter.  A  flow rate of approximately  0.4  cfm was
used as a reference.  The  two dry  gas  meters agreed within ±2.7 percent,
based on three 15-minute runs exhibiting a  coefficient of  variation of less
than 1.3 percent.
     Laboratory  systems audits  for the CDD  precursor  analyses were also
conducted by  Radian and RTI.  In general, the results of these audits  showed
that the quality of the precursor  analysis  efforts  was sufficient to meet
Tier 4 objectives.   However, a  number  of recommendations were made relating
to  sample handling,  documentation  of extraction  procedures, and data
reduction and review.

7.3  DATA QUALITY ASSESSMENTS
     This section provides an estimate  of the uncertainty  in  measurements
made for 1) flue gas CDD/CDF concentrations, 2)  ash CDD/CDF concentrations,
3)  feed material precursor contents, and 4) flue gas  combustion parameters.
The uncertainty  assessments are based  on performance  audit results and QC
data.
7.3.1  Flue Gas  CDD/CDF Analyses
     Data available to  assess the  quality of flue gas CDD/CDF concentrations
include audit results,  fortified QC sample  analyses,  lab blanks,  field
blanks and sample surrogate recoveries.  Based on the QA/QC data  presented
in this section, the accuracy achieved  for  the CDD/CDF analyses should be
within +50 percent, except  for certain  samples as noted.
                                    7-16

-------
     CDD/CDF audit  sample results are summarized  in Table 7-7.   Four samples
containing 2378-TCDD were provided to Troika for  analysis.  As  shown in
Table 7-7, the difference between expected and measured values  for
detectable quantities of 2378-TCDD ranged from 0.3 to 40 percent.
     Fortified QC sample results are presented in Table 7-8.  The acceptance
criterion for these samples was a measured value within +50 percent of the
amount  spiked on the sample.  As shown  in Table 7-8, this criterion was met
for all species in  each QC sample with  the exception of 2378-TCDF in QC
samples for Sites 09 and 10.  The high  values of TCDF for Sites  09 and 10
are probably due to instrument background caused  by large amount of TCDF in
the Site 10 source  samples.   In reviewing CDD/CDF analytical results for
Site 09, it should  be recognized that reported values for TCDF  may be
somewhat higher than the actual values.
     Analysis results for laboratory and field blanks are presented in
Table 7-9 and Table 7-10, respectively.  Spiked surrogate recoveries for the
labeled TCDD's were within the target range of 40 to 120 percent.  Surrogate
recoveries for the  labeled hepta- and octa-CDD's were also acceptable for
all but one field blank.  The recovery  for hepta-CDD in the Site 09 inlet
field blank (138 percent) was somewhat  above the target range of 40 to
120 percent.  This  is not expected to have any significant effect on other
samples from Site 09.
     As shown in Table 7-9, the laboratory blanks were found to  be clean
with the exception  of some OCDD and OCDF.  The amount of OCDD or OCDF found
in the blanks was generally at or near  the minimum limit of detection.  As
shown in Table 7-10, the field blanks were also found to be clean with the
exception of Sites  01, 02, 08, 10 and 11.  For Sites 01, 02, 10  and 11,
field blank values  of up to 40 ng were  detected for various isomers.
However, for each of these sites, the field blank values were small relative
to the MM5 test run values and no major problems related to sample recovery
or contamination are suggested for samples from these sites.
     For Site 08,  hepta- and octa-CDD's and CDF's in the 3.5 to  109 ng range
were found in the outlet field blank.  These values are higher than those
found in the MM5 test run samples from the control device inlet  and outlet
                                    7-17

-------
      TABLE 7-7.  SUMMARY OF ANALYTICAL AUDIT RESULTS FOR 2378-TCDD
Samp! e
Description
2378-TCDD in NBS
Urban Part icul ate
2378-TCDD in NBS
Urban Part icul ate
NBS 2378-TCDD in Octane
NBS 2378-TCDD in Octane
Expected
Value
0.05 ppb
0.007 ppb
67.8 ng/mL
67.8 ng/mL
Measured
Value
0.07 ppb
NDa
68 ng/mL
69 ng/mL
Percent
Difference
40
--
0.3
1.8
Reported method detection limit was 0.04 ppb.
                                   7-18

-------








CO
LU
Q_
21
CO
_l
s
t;
8
f^
~
<
3
CO
O
UJ
3
Lu
O
LU
HH
HH
C£.
O
^
g
CO
1—
=3
CO
LU
CC.
CO
I—*
CO
<
«c

CO

LU
_J
£?
H-
















c
CD
U
(_
CO
Q.
I/)
CO
•*••
CO
o
U
CD
CC
CD
•P
(0
O)
O
[_
L.
CO












CO
:*
a.
CO












•P CM
CO
CD
•P ^
CO

CD
•P 0
t- i-H
CO

CD

CO
CD
CO

CD
•P !*•
-- 0
CO
CD
4J VO
•— O
CO

CD
-P U1
•*- o
CO

CD
•P *
•r- o
CO

CD
•p en
•*- o
CO
CD
•P CM
••- O
CO

CD
•p 1-1
£°

IO
•f— *-•.
•P 0)
c c
 ^t ^r


VO CM .H O
O> O "S- «<»•
r- <

5 S § S



o o o •*
in • O
i-H f-H

0 CM P^ CM
o o> in r~»
i— i


CM CM en CM

CM CO «3- r-
o o vo m
i— i •— i
,

VO VO VO O
CO CO CO O*



in in o o
CM CM




§ 0
V 8
10 I
Q Q -P 10
O O Q. -P
O O CD U
1- 1- I O
•«r CM TT CM
r- i-Hr— i— 1
0000
r- en r— en
en I-H en >-H







•P
c
CD
CD
Q.

T3
CD
-P
•P
CD
O
•P
e
o

<
c
— ivo^«f
CM | ||
1

co en i— i
encocow «vomco «o\en •
i-HZZenoi-Hi-izvoi-ii-ien
1 ll III



^- VO^-CMf OO VO •T CM
ooo-HCMenooo-HCMen





O Lu
ggg Lu O Lu Lu
ii-ioiioV*Trt.'oVioV
cocD-pio-piecocD-pio-pie
r~-.cc Xcx+Jr--^c xex-p
PO-PCOCDCDOri-l-PCOCOCOO
CMOCLXIOCMOQ-XIO
in
+*
3

10
CD •
'o.
o a
in
p- CM
IO r»4
e—
CD

*C CO
CD
• -p
U)
f C
in o
^— ^
10 CD
C JC
r\
•o in
n CD
si
•p w

|s
CD "O
g"
O (O
•p
in
o o
18 .
CX 1 CD
CD X *0.
"ex-c i
10 .c
in o> in
f- rt /-
0 t. -P
CD -P O
0 1^
-M <0 -^
C j»
o> o
TH JJ _V
ffl *^^
i- CD in
in "CD -p
x> o
•o *- L. O
o o o.
n- e
3 ^
JO II
in co
7-19

-------







CO
=£
^^
1
CO
oE
1
m
iff*
-J

ji a
O
^p
{~^
Q|
O

^.
ce
UJ
l-l
J—

cc
o
u.
co
(—

CO
UJ
cc

CO
1— 1
CO
|!
<
•
cn
l


UJ
i^



























•p
S
o
(D
fv

•t
(fl

O
O
0

(D
U

O
u
co
















(D
•P CM
1- rH
CO
0
•P i-H
1- rH
CO

0
•P O
1- i-H
CO
0
•p o>
•r- O
CO
CD
•P co
T- 0
CO


•P r-
T- O
CO


a>
•P VO
•»- o
CO

(D
•p m
f O
co

CD
•P •«*•
CO


(D
•p cn
•r- O
CO
-P CM


a>

•r— C3
CO








T3
C

.


0

O O i-H O
o ^- •*
1-4

vo eo cn CM
a* cn ^^ ^f



vo *^ CM at
a> at at co


co oo o o
I-. OO i-H 0
i-H i-H
CM CO CM OO
o at r** co
i-H



^ co i— i cn
at o vo vo
i-H



OO VO CM VO
r-» o vo ^r
F-H


o CM vo m
o o in co
i-H i-H


at at in vo




00 •*!• CO ^
at at r*- oo

vo o r» vo
at •-« m m


•«3- vo r- co
cn at at oo


-0
C^ ^3
o a
O Q
1 0
(O (O (O I
Q Q -P <«
Q Q Q. -P
o o a> o
h- I— in o
i i i i
^ a ^ s
pa* f«^ ^^ ft
o o o o
r~ cn r- cn
cn I-H m 1-1













U)
^
o»
o
c
R>
2:
«,
•o
(D
•JJ
(J
(D
-P
(D
O
-P
C
o
s
































QQQQQCMQQQQCIQ
ZZZZZ^Z    Z





QQQQQi-HQQQQQQ

                CD




QQQQQQQClClQaQ
                       o    o
       ooooocMooaao
       zzzzz   «zzzzz
                       o
       QQQQQi-HQOQQOO
       QQQQQOOQQQQQ
       ZZZZZ   -ZZZZZZ
                        o
       OOOOOCMOOOOOO
       zzzzz   -zzzzzz
                        o
       QQQQQi-HQQQQQQ
       ZZZZZ  *ZZZZZZ
                        o
       aooooooooooo
       zzzzzzzzzzzz
       QQQQQQQC1QQQQ
       zzzzzzzzzzzz
           zzzzzzzzzz
           O                 U-
        Q Q Q     O    U.O
        r- -e  =  X a.4Jr- j= c  x  a..
                                                           U)
                                                            c

                                                            c
                                                     O
                                                    -p
                                                    o
                                                    o
                                                            E
                                                           •a
                                                            a>
                                                            O)
                                                     i_

                                                     o
                                                            0)
                                                           T3
                                                     U)
                                                     p

                                                •  a>  ^
                                               (D p—  i—
                                              i—  a.r—
                                               a. E
                                               E     z z
7-20

-------
















CO
1
	 1
CO
o
UJ
*H
u.
in
8
P=
3.

S
t-H
U_
t— 1
8

^f

ce
UJ
1—4
1-
ce
0
u.
CO


CO
UJ
o:

CO
>— 4
CO
_J
o
(—4
1
r-
UJ


*












































•p
(D
O

t
S
L.
2
>
0
g
ce

•p
10
I"
L.
3
CO
























050
•P CM
CO """'
•p
05

*-H -p
O
•P 4J
*- CO
CO i-
c
I-H
U -P

•P O .2
•— rt +»
*" 0
•P
O

°i
- ID
CO •—
c
1 %
o

co +>
°c3
£^
CO •—
c


.2
° =
«- 05
CO 1—
c
f~t

U +*
05 ID
•P *O •""
s;es
o

ID
in 4->
O 3
O
* 2

•"- 05
CO r-
c
t— 4
0)
T -p
2§
•^ 0>
CO •—
c
*"*
o +>
05 0)

°

m
10



(—4
o
f-4



S
p^




o
r-

\o
o

»
S



CM
O\




CO
f^.




0>



0>
CO





f^j
o
•-1
S


CM
ON



S
--—1


§




to
en



1C
a

CM
f-4
o
r-4

<


co
<£>


in



§


i^
>o

CO
m
1-1

CM



m


^
en

o
Cft



CO
CO



r-






r-




rr>
oo

1
O
IO
Q.
05
,J*
O
*^
•n

<


^^
O


>0
en


in


CM
^


cn


m



CN
in


CO


CM
en



o




en





CO
en

r«


co



en
en


o
IO




en
CO

J3
a
V
•p

2
u
m
 g
 I
05

"a.
«£

"S
        o r^ r*" *o to o

           o o >H CM x-i   »
                  o o   o o es   o
          o o >H in en
                  CM o
         '   «Z Z z * Z "I Z S 't "1
          o         o    o      o o
        aooa-HrooaoocMCM
        z z z z  •  • z z z z  •  •
                  o o            o o
                  OO    OOOOO
        aoooainacMocsioo

        Z2zzzozozozz
       aaooaaoocMoooa
       ZZZZZ  -Z  -ZZZZ
   go a   a    u.0 u.
   oaooooooi
o^OQoao^'


r- £ c x Hii S 5 c x a.+i

CMOCLXXOCSIOQ.XXO


        7-21
                                                          U)
                                                          E
                                                          1C



                                                          I
                                                          1C
                                                          c
                                                   >>    —i


                                                   O     C3


                                                   S     §
                                                  «•-     L.
                                                   C   05 ro
                                                  •P   N I.
                                                   C   >>
                                                      81— C
                                                       a o

                                                   05   ic
                                                   o  ^-S

                                                  •p  +J 1-
                                                   05   O O


                                                   3   in 4J

                                            •   »   °  .2 Ts
                                           03  i—   03     «-

                                          11--;

                                           n   in  4->   03 3
                                           B)       1C  .p E

                                          f   U  *O  CO C
                                           O   1C   05     --
                                           16   9  +l  t-X
                                           9       U  O
                                               C   3  «*-
                                           C  •*-  T3      • •  •
                                          •*-       C   05 05 CO
                                           O)   C   U   +> N
                                           C          -O O t. >,
                                              o   in     05 o i—
                                          to  IN   10  jc *> a. 
                                                            o o
                                                           Z Z

-------
sample locations.  The source of the Site 08 field blank contamination is
not known, but there  is no  indication that other samples from this site were
similarly contaminated.
     Surrogate recoveries for flue gas MM5 samples are presented in
Table 7-11.  Recovery of the four labeled surrogate compounds satisfied QA
requirements for all  samples except the inlet MM5 samples from Sites 11 and
12 and the outlet MM5 samples from Sites 06 and 10.  Surrogate recoveries
for Sites 06 and 10 (outlet), and 11 and 12 (inlet), were below the target
range of 40 to 120 percent.  Problems with method efficiency for the Sites
06, 11 and 12 samples resulted from contamination present in the sample
extracts and corresponding  difficulties in achieving acceptable
chromatographic separations.  Problems with method efficiency for Site 10
were due to the large amounts of CDD's and CDF's present in the MM5 samples.
Analytical results reported for MM5 samples from Sites 06 and 10 (outlet
only) and 11 and 12 (inlet  only) must be considered estimated values.
Actual flue gas CDD/CDF concentrations are probably higher than those
determined based on the reported results.
7.3.2  Ash CDD/CDF Analyses
     Data available to assess uncertainty in the ash CDD/CDF concentration
measurements include  the audit results discussed in Section 7.3.1, fortified
QC samples, and sample surrogate recoveries.
     Fortified QC results and sample surrogate recoveries are presented in
Tables 7-12 and 7-13, respectively.  For the fortified ash QC samples,
results are not reported for each site because samples were subjected to
extraction and clean-up procedures as a "set" of 10, 11, or 12 samples.  QA
samples were then incorporated into each set.  The QC ash sample results
presented in Table 7-12 satisfy the QA requirements of + 50 percent with the
exception of hexa-CDF in the Site 06 sample and 2378-TCDF in Site 07 and
Site 10 ash QC samples.
     The spiked surrogate recoveries are shown in Table 7-13.   The surrogate
recoveries satisfied the QA requirements of 40 to 120 percent with the
exception of low values for one baghouse sample at Site 07,  and one bottom
ash sample at Site 13.  Because triplicate runs were performed at these
                                    7-22

-------
             TABLE 7-11.  SPIKED SURROGATE RECOVERIES FOR TIER 4  FLUE GAS COD/CDF SAMPLES

Test Site
01


02



03



04





05





06








07





08





09





10


11





12




13


Test
09
10
11
01
02
03
04
01
03
05°
OS
01
02
03
01
02
03
01
02
03
01
02
03
01
02
04
OS
03
03
03
06
06
06
01
02
03
01
02
03
01
02
03
01
02
03
01
02
03
01
02
03
02
03
04
01
02
03
01
02
03
01
02
03
01
02
03
01
02
03
Run Sample Location9
Outlet


Outlet



Outlet



Outlet


Inlet


Outlet


Inlet


Outlet



(liquid)
(partlculates)
(XAO 4j filter)
(liquid)
(partlculates)
(XAD & filter)
Inlet


Outlet


Inlet


Outlet


Inlet


Outlet


Outlet


Inlet


Outlet


Inlet


Outlet

Outlet


37Cl4-TCDDb
98
100
96
96
94
96.
102
78
92
106
NS
94
85
100
90
100
96
96
96
104
96
107
86
13
10
5
92
0
NS
NS
0
NS
NS
52
112
88
92
108
104
100
84
84
96
96
100
70
80
92
84
82
68
0
0
0
84
76
92
94
102
94
NS
NS
NS
NS
NS
NS
0
0
0
13C12-TCDDb
100
98
100
98
92
88
82
106
94
NS
102
88
91
98
90
106
62
100
77
94
97
101
80
80
36
76
96
NS
76
54
NS
66
0
52
106
88
96
112
114
94
68
66
96
96
92
96
86
94
90
80
94
0
120
0
66
68
106
100
88
96
0
90
94
98
112
112
0
0
0
37Cl4-Hepta-CODc
89
85
84
88
81
98
86
48
52
53
NS
47
34
48
55
40
S3
60
54
55
55
63
39
0
0
20
53
0
NS
NS
0.2
NS
NS
4
78
48
75
54
66
69
74
74
42
55
58
114
122 '
102
80
116
90
0
0
0
0
0
26
45
47
52
0
25
42
42
68
40
o
o
0
13C12-Octa-CDOc
89
83
76
73
68
54
41
99
79
NS
78
49
51
63
76
66
42
79
64
62
76
66
28
18
11
4
61
NS
115
10
NS
81
o
6
66
64
58
41
40
56 .
65
66
59
62
42
84
66
64
82
77
105
0
58
33
11
98
28
40
41
42
0
14
26
40
46
43


 Spiked at 5 ng In each sample.

NS » Compound was not spiked  Into this sample.
C3p1ked at 20 ng 1n each sample
 Aqueous portion
eXAO-2 resin portion
                                              7-23

-------

CO
UJ
1
CO
g
fe
8
g
1
X
CO
o
UJ
u.
1
g
U-
CO
i
S
CO
CO
_J
^
<

CM
I-l
1
f-
LU
§
}M








•JSCM


CD
+•> 0
Z*




CD
•"- 0
CO





CD
£8
CO






o
-t-> "U-
•»- o
CO














co *>•
en in


CM f—
en in





CM O
^^ C7k
i-4






f- CO
at in







-« £
o\ co











(O A
§ §
H™ Cj
CM CM
i-H i-H
0 0
m m
i— 1 r-<



43
CD
U
CD
0.
•o
CD
y
«
o
~


•o
10
•o
1
"o.
CO
•P
c

.«. Q_
II
 c
c
0
in CM
jj
•o -o
CD CD

"ex's.
co co
a .0
7-24

-------
TABLE 7-13.  SPIKED SURROGATE RECOVERIES FOR TIER 4 ASH SAMPLES
Test
Site
01


02


03

04
06



07







08


09


Test
Run
09
10
11
01
03
04
03
04
Presurvey
1
2
3
2
3
1
3
1
2
3
1
2
3
01
02
03
01
02
03
Surrogate Recovery, Percent
Ash Type
Furnace
Furnace
Furnace
Bottom
Bottom
Bottom
Scrubber water solids
Scrubber water solids
Economizer
Primary chamber
Primary chamber
Primary chamber
Settling chamber
Settling chamber
Primary chamber
Primary chamber
Secondary chamber
Secondary chamber
Secondary chamber
Baghouse
Baghouse
Baghouse
ESP
ESP
ESP
Baghouse
Baghouse
Baghouse
13C12-TCDDa
90
82
78
99
100
94
84
98
89
78
110
71/100°
94
100
92
62
60
89
62
99
89
90
96
98
94
86
94
116
13C12-Octa-CDDb
91
88
86
43
56
69
63
73
93
51
70
44/54c
85
66
42
44
46
61
54
26
49
53
54
56
62
51
73
78
                         (Continued)
                             7-25

-------
 TABLE 7-13.  SPIKED SURROGATE RECOVERIES FOR TIER 4 ASH SAMPLES (Continued)

10





11


12





13


02
02
03
03
04
04
01
02
01
01
02
03
01
02
03
01
02
03
#1 Baghouse dust
#2 Baghouse dust
#1 Baghouse dust
#2 Baghouse dust
#1 Baghouse dust
#2 Baghouse dust
Furnace outlet
Furnace outlet
Furnace outlet
Filterable scrubber solids
Filterable scrubber solids
Filterable scrubber solids
Bottom
Bottom
Bottom
Bottom
Bottom
Bottom
100
102
104
90
96
68
61
50
88
102
98
106
98
86
84
94
93
99
73
56
63
64
64
45
Od
42
0
46
48
54
74
67
63
68
72
25
|:Spiked at 5 ng in each sample.
^Spiked at 20 ng in each sample.
^Duplicate samples
 Contaminated samples
                                    7-26

-------
 sites, the low surrogate recoveries for these single samples were not
 expected to affect the results.  For Site 11, the furnace ash samples from
 Run 01 did not show any surrogate recoveries because of sample
 contamination.
 7.3.3  Feed CDD/CDF Precursor Analyses
      CDD/CDF precursors analyzed in the feed samples include chlorobenzenes
 (CB), chlorophenols (CP),  chlorinated biphenyls (PCB),  total  elemental
 chlorine and total  halogenated organics (TOX).   Data available to assess
 uncertainty in the CB,  CP  and PCB analyses  include lab  blanks and sample
 surrogate recoveries.   Data  available to assess the quality  of the chlorine
 analyses include  lab blanks  and QC samples.
      Surrogate recoveries  for the CB,  CP and PCB analyses  are reported  for
 feed  samples in the site specific test reports  for the  Tier  4 program.
 These results  will  not  be  repeated in  this  section.   In general,  the
 surrogate recoveries for the CB,  CP and PCB  analyses were  below the QA
 objective of 50 percent and  the reported analytical  results must  be
 considered as  estimates.   Where measurable concentrations  of  CB,  CP and PCB
 were  found in  the  feed  samples,  the actual values  are probably three  or more
 times greater  than  the  reported value.
      There are  several  reasons  for the comparatively low precursor surrogate
 recoveries reported  for the  Tier  4 feed samples.   First, the  complex  nature
 of the samples  required extensive  clean-up procedures prior to  6C/MS
 analysis,  which increased  the potential  for  losses of the  surrogate
 compounds  (and  analytes) during sample  preparation.   There were no standard
 extraction/analysis  procedures  available for the specific  sample types to be
 analyzed.  And, although attempts were  made, ideal procedures could not be
developed  for all  sample types  in  the time frame available.  Second,  large
 sample sizes (25 to  50  gram) were  required to increase method sensitivity
for the target analytes and to ensure that representative portions of the
samples were analyzed.   Due to the high cost of labeled surrogates, it was
not desirable to spike the large sample sizes with surrogates in proportion
to that normally used for smaller samples.  Results of supplemental
laboratory studies are shown in Table 7-14.   These data show that  when
                                    7-27

-------









t
ia
CM ^ D.
o -o
0 U

•p X -p
J- •* in
to *
r—
a.


CO




CO
o
in



CM ffl
O -P 0
c -o
a> t- 3
•P  IB c
•P ex o
•F- to -a
to (-
3
CO
i-H



CO
o
in

CO


CO
o
in

*s
U
(0
in •— tn
oca t-
Q
(D O) 3
•PCC
to t- — J
to
•G
(D
(D
CM LU
r-4
-H -a
tO F^



CO
**

CO
o
in

CO

CO
o
in



I/)
C
cc















•o
c

o
a.
§
O




o o> m «* CM cn
^t in o> «* oo r~




co o vo o o in
F-4 rn PO 2^ F^ ro



««r o» ^ in vo FH
m in r~- CM tj» o
1^


r-4 CM «O O O Q
rH rH Z Z Z

CM CM r- CM in in
m •» co in oo to


in in o ^ co ro
1-4 CM z CM to in


CM r- o CM \o oo
co ^r CM in oo ao


FH t*- o* r-

r«4 o o* r^ r** co
co ^r in «» o\ FH
>>. -V. ^. "^ ^. 1-4
CM in r«- in o •>.
co m r-- CM oo r-i
o
i— i

"*" CM CM FH



*o ch co in r** P"-
co m »o co FH I-H
r-4 i-H


in r- o co »o 1-4
CM r* \o it FH I-H
FH i-H





"o
c
 •—
r- CX C >•>
O O ffl C
C U N d>
(DOC -C
.c F- a> f- a.
Q. ^ XI >> -F-
o o o c .Q
t. ia ^ 0> o
r- o -P 0 -e E
OF- e F- a. o
c .e CD ^ -F- t,
0) y ex u -O £*
a. CM 10 X) E i-
1 1 U | 0 -P
vo «* co •* »-  FH -O J3 -P
























in
0)
N
•^
(A
tn
a>
co'a.
i-4 E
10
O
•o -p
a> o
•p ^
U)
•F- C
in o
c
O TJ
U fl>
in -F-
O Q.
r— U)
a,
E W
IQ 10
in *
F^ *CO
(0 C
S o
o
•0 CM
C «-•
* in
co a>
O -P
10 S)
14- O
0 L.
t_
TJ 3
a> in
W "5
in
C >i

° +»
in c
Q) f0
r— 3
a. cr
E
ro CD
U) E
(0
8,1"
1- 
-------
  sample  size was  restricted  to  1  gram  and  the  amount  of surrogate  spiked  was
  held  fixed, surrogate  recoveries improved substantially.
       In  spite of the relatively  low surrogate recovery values  for some of
  the feed samples, the  resulting  analytical sensitivity for the target
  analytes was considered acceptable for the purpose of  this study.- The
  instrumental detection limit ranged from  about 100 to  500 picograms
  on-column for the 1 microliter of final extract injected into the GC/MS.  At
  a method recovery efficiency of  100 percent for a 50 gram solid sample
  cleaned up to a final extract volume of 1 milliliter,  the overall analytical
  sensitivity would be approximately 2 to 10 ppb in the  solid sample.  For
  samples with surrogate recoveries as low as 1 percent, the overall
 analytical  sensitivity of the method would still  be 200 to 1,000 ppb, or 0.2
 to 1.0 ppm.   Thus, even in a worst-case'situation the analytical  procedures
 used provide information on the precursor content of the feed samples down
 to the ppm  level.
      Results from laboratory blanks  for the CB, CP and PCB analyses showed
 no detectable  target compounds  at a  detection  limit of 10 ppb.
      Total  chloride  analyses were performed by Research Triangle  Institute.
 Blank  analysis values obtained  for the Parr bomb  combustion/ion
 chromatography technique were 36, 0, 56,  and  18 ppm chloride.   A
 commercially available  coal  standard containing 2,600 ppm chloride was
 analyzed  as  a daily  QC  standard.   Reported values  were  2,500, 2,500,  2,500
 and  2,400 ppm.
     Quality control for the total organic halide  (TOX) data consisted of a
 single QC sample  analysis.   The sample contained 2-chlorophenol and
 pentachlorophenol  at a  TOX level  of 376 ppm.  The  total  TOX measured  for  the
 sample was 251 ppm, showing  67 percent recovery.
 7-3-4  Flue Gas Combustion Parameters Monitoring
     Continuous emissions monitoring (CEM) of flue gas  CO, C02, 02, NO ,
S02, and total hydrocarbons  (THC) was performed at most of the Tier 4 test
sites.  The purpose of the CEM data was to 1) observe fluctuations in flue
gas parameters, and 2) provide an indication of combustion conditions.  Data
available to assess the quality of the continuous  monitoring efforts include
                                    7-29

-------
1) performance audit results, 2) drift check results, and 3) quality control
data.
     The results of the CEM performance audit are presented in Table 7-15.
Generally, the audit results showed the CEM instrumentation to be calibrated
well within the specified program accuracy objectives.  There were two
exceptions.  The C0/C02 instrument exhibited a high signal to noise ratio
and high bias for both channels with the CO channel exceeding the
±20 percent accuracy target by a small margin.  There was a degree of non-
linearity below 70 ppm CO; however the majority of actual testing data was
above 1000 ppm.  Also the NO  instrument showed a high response bias
                            A
exceeding the ±20 percent accuracy target for this parameter.  Both
instruments had a linear response across the scale (correlation coefficient,
r>0.9950).
     Accuracy problems resulting from the high signal-to-noise ratio for the
C0/C02 instrument were reduced for sites tested after the performance audit
(Site 03) by an improved data acquisition system.  The new system collected
integrated one minute data as opposed to instantaneous one minute values.
Data for CO concentrations at the sites using the original data acquisition
system should be flagged as possibly having a high bias (>20 percent).  The
NO  instrument was repaired following the audit and no further problems were
encountered with this instrument.
     Pre- and post-test analysis of mid-point calibration standards were
used to track daily instrument drift for each of the different parameters.
The mean percent drift for each parameter is presented in Table 7-1-6 and
ranged from -0.121 to 5.943 percent, well within the ±10 percent acceptance
criterion for instrument drift.  The extreme cases were the 0« and SO^
instruments exhibiting average or mean drifts of -0.121 and 5.943 percent
respectively.  Data showing instrument drift for each of the 13 test sites
are presented in Figure 7-1.
     Day-to-day variability for the continuous emission monitors was
determined by daily (pre-test) analysis of a control standard.
Acceptability of the QC check was based on agreement of the daily analysis
result with the average concentration measured for the QC standard on
                                    7-30

-------
TABLE 7-15.
CONTINUOUS EMISSION MONITORING SYSTEM
   (CEM) AUDIT RESULTS
Parameter
CO





CO,
Cf
v .

°2



NOX




THC



Input
Concentration
(ppm)
0.0
60.5
259.0
1002.0
2491.0
2480.0
0.00
2.01
7.61
10.50
0.0
4.02
7.99
9.96
0.0
51.7
100.0
225.0
705.0
0.0
9.63
20.50
80.40
Measured
Concentration
(ppm)
-0.8
115.4
325.9
1239.5
2933.8
3048.4
0.45
2.16
8.99
12.01
-0.05
4.23
8.40
10.34
1.3
62.6
121.5
279.5
795.9
1.11
9.84
19.50
75.16
Relative
Error
(*)

90.7
25.8
23.7
17.8
22.9
—
7.5
18.1
14.4
.
5.3
5.1
3.8
.
21.1
21.5
24.2
12.9

2.2
-4.9
-6.5
Accuracy
Target
(*)


±20





±20



±20



±20




±25

                        7-31

-------
            TABLE 7-16.  SUMMARY OF INSTRUMENT DRIFT CHECK RESULTS
                            FOR FLUE GAS PARAMETERS
Parameter
Average Drift (%)
                                                        Standard Deviation
°2
CO
co2
S02
NOX
THC
                                   -0.121
                                    2.016
                                    5.572
                                    5.943
                                    4.861
                                    4.689
                                  0.758
                                  8.514
                                 10.463
                                  8.515
                                 12.073
                                  7.186
                                      7-32

-------
                20 •
                1S -
                                    Oxygen
                10-
                5 -
            Q
            O)
               -5 -
            to
            c
               -1O •
               -15 -

               2O •
                15 H
                10
                Sn



                O
               -5 -
            i.
                   01   02  03  04 05   06    07  OB OB  1O  11  12
                              Carbon Monoxide
                  01   O2  O3  O*  OS   OS   O7  OB  OS  1O  11   12
s*"
4->
<4-
•^
a
+j
c
i_
4J
V)
C
HH

15 -
10-
5 -
-5 -
-1O -
-15 -
-20-
0 Carbon Dioxide 0
O
1
__= *
ii — - a —
a a a° «
** ° ° *
• ^
* A
A
4
                  01   02  03  0*  OS   08    07  08  Ofl  1O  11  12
                              Site Number


Notes:  Square represents even numbered  sites; triangle represents odd

        numbered sites.   Drift check acceptance criterion  was  +. 10% drift.


FIGURE 7-1.  DRIFT CHECK RESULTS  FOR CEM PARAMETERS
                                         7-33

-------



£

a
c

i-
c
*— 1

20 -i

15-
5 -



-5-

-15 -


= Nitrogen Oxides


° o °
» °- 0
0 a *• % * ° ' o
a A



                O1  O2  O3  O*  OS    O0   O7  O8 09  1O  11   12
13-
10-
14-
s-
c
-13-
Total Hydrocarbons
a
"a" * a **
o
a


3O-
13-
%« 3 -
0
—3 -
•M
C
i.
4J-13-
M
c
•--so.
Sulfur Dioxide
a0
O
a
*

                             Site Number
Notes:  Square represents even numbered sites;  triangle represents odd
        numbered sites.  Drift check acceptance criterion  was + 10% drift.
 FIGURE 7-1.  DRIFT CHECK RESULTS FOR CEM  PARAMETERS (continued)
                                       7-34

-------
previous test days at that test site (i.e., the running mean concentration).
Table 7-17 summarizes the results of these QC checks.  The average percent
difference from the running mean for the 13 sites ranged from -0.016 to
0.993 percent and only two percent of the QC checks fell outside the control
limits of. ±10 percent of the running mean.  Data showing the QC check
results for the 13 sites are presented in Figure 7-2.
                                    7-35

-------
          TABLE 7-17.  SUMMARY OF  QUALITY CONTROL STANDARD ANALYSES
                            FOR FLUE GAS  PARAMETERS
Parameter
°2
CO
co2
so2
NOX
THC
Average Difference
from Running Mean (%)
0.146
0.307
0.644
0.993
0.538
-0.016
Standard Deviation
1.079
6.821
4.209
4.391
3.081
2.571
Average based on Sites 01 through 13.  Running mean determined separately
for each site based on consecutive daily analyses of the quality control
standard.
                                    7-36

-------
«— - iO -
C
cn 10 -
c
C 5 .
i
o -
£-,.
CU
u —10 -
0)
CO -15 -
M-

ro 15 -
CU
1O -
O1 "
C
'= 5-
I -5-
1 O2 03 O4 OS OB 07 O8 09 1O 11 12
                             Site Number

Notes:  Square represents even numbered sites; triangle represents odd
        numbered sites.  Control standard acceptance criterion was
        difference from running mean within +_ 10%.

FIGURE 7-2.  CONTROL  STANDARD ANALYSIS  RESULTS FOR CEM  PARAMETERS

                                         7-37

-------
                                  Carbon  Dioxide
                                                      —s-
                    ».  o
                   O1   O2  O3  O4  OS    06   O7 O8  09  1O  11  12
                                Nitroggn  Oxide
                                       oa
                   O1   O2  O3  O4  OS    OO   O7  O8  O»  to  11  12
            O)
                15
                1O
            CD


            =   5,
               -1O
            o
               -15-


            «
            («-  -20
                                Total Hydrocarbons
            2     O1   °2  O3  O*  OS   OS   O7  O8 O0  1O   11  12

                                 Site Number


Notes:  Square  represents even numbered sites; triangle represents odd
        numbered  sites.   Control standard acceptance criterion was

        different from running mean within +_ 10%.


 FIGURE  7-2.  CONTROL STANDARD ANALYSIS RESULTS FOR CEM PARAMETERS
              (continued)
                                     7-38

-------
                                   CHAPTER 8
                              ASH  SAMPLING PROGRAM

 8.1   OVERVIEW  OF  THE  ASH  SAMPLING PROGRAM
      The  ash sampling program had two major  goals.   First,  for  ash  samples
 collected at the  source test  sites,  the  intent  of the  ash  sampling  program was
 to relate CDD/CDF concentrations  in  the  ash  to  CDD/CDF concentrations  in the
 flue  gas.   Rank order statistical  analyses were used to determine
 statistically  significant relationships  between the  CDD/CDF concentrations in
 the ash and flue  gas.  The hypothesis was if significant relationships  between
 the ash and flue  gas  emissions were  found, then ash  sampling could  potentially
 be used as  a quick and relatively inexpensive method for qualitatively
 screening combustion  devices  for  possible CDD/CDF emissions in  the  flue gas.
 The second  major  goal  was to  use  the results of the  ash sampling program as
 the basis  for  re-evaluating the ranking  of source categories as potential
 emitters  of CDD/CDF.
      Previous  to  this  study,  several studies, as discussed  in the
.Initial Literature Review,  had initially attempted to  assess CDD/CDF emissions
 by taking  ash  samples  only.   However, differences in analytical methods,
 sampling  methods,  congeners reported, and  limited process  information for the
 CDD/CDF data from these studies caused comparisons of  these results to  be
 questionable.  Therefore,  additional investigation was  considered necessary.
     The  use of ash samples for screening  combustion sources has several
 advantages:
     1.  There is  usually  a significant  quantity of easily  accessible ash in
         the hoppers of the control devices  on  combustion sources.
     2.  The fly  ash is usually homogeneous  and can be  readily  extracted
         without  complex sample preparation  techniques.
         Ash sampling does not require sophisticated sampling equipment.
         Ash sampling supplies and samples can be shipped easily.
3,
4.
                                      8-1

-------
      5.   A PCDD/PCDF (tetra-  through  octa-  congeners)  analysis  can  be
          performed for approximately  $500 to  $1,000.
      The main  disadvantage  of using ash  screening  as a means  of assessing
 CDD/CDF  emissions  in the flue gas  is  the lack of data  to  qualitatively  support
 the  relationship.   Work previous to the  Tier  4 Study had  not  demonstrated  ash
 sampling to be a proven,  reliable  screening technique.
      For the Tier  4 ash sampling program, ash samples  were  collected from  a
 range of combustion devices at sites  located  across the United  States.  The
 types of ash sampled included baghouse ash, multicyclone  ash, electrostatic
 precipitator ash,  afterburner ash, settling chamber ash,  economizer ash,
 particulates filtered from  scrubber water and bottom-ash.   If possible, the
 ash  was  sampled at a point  in the  process where the particulate matter  was the
 finest and coolest because  semivolatiles like PCDD/PCDF are expected to
 condense on cool,  fine particulate.   Thus,  the sampling point was usually  the
 farthest accessible point downstream  of  the process or control  device.  Bottom
 ash  was  collected  from uncontrolled combustion devices  or where control device
 ash  sampling was not practical.  At some sites,  ash samples were taken  from
 multiple sampling  locations in a process (e.g.,  bottom ash  and  scrubber water
 filterable solids  for a sewage sludge incinerator).  If available from  plant
 instrumentation, process  data such as feed  rate, feed  composition and flue gas
 characteristics  such as temperature,  oxygen content, and  carbon  monoxide
 content  were also  collected for each  site where  ash samples were collected.
     Ash  samples were  collected from  75  sites  (including  CDD  source test
 sites) located  in  21  states and in eight EPA  regions.   Twenty-two combustion
 source categories  were  included with  one to nine sampling sites  in each
 category.  The target  number  of ash sites for  each category was  three.  More
 samples were obtained  for some categories because EPA Regional offices
 requested more sites  (e.g., wood-fired boilers) or several pre-test surveys
were done for the  emissions testing program (e.g.,  sewage sludge incinerators
 and black liquor boilers).  For some  of the Rank C source categories,  fewer
than three sites were sampled because only one or two sites were identified
for sampling.  Table 8-1 summarizes the ash sampling sites.
                                      8-2

-------
                   TABLE 8-1.  SUMMARY OF ASH SAMPLING SITES
     Combustion
  Device Category
Radian
 Site
 Code
     Sample Type
     Solid/Slurry
  Sampling Organization*
Sewage Sludge
Incinerator
SSI-A
SSI-B
SSI-C
                    SSI-D
                    SSI-F

                    SSI-G
                    SSI-H
                    SSI-I
                    SSI-J

Black Liquor Boiler BLB-A
                    BLB-C
                    BLB-D
                    BLB-E
                    BLB-F
                    BLB-G
Wire Reclamation
Incinerator
Secondary Copper
Recovery
WRI-A

WRI-B
WRI-C
WRI-D

MET-A
MET-B
Carbon Regeneration CRF-B
Furnace             CRF-B
Bottom Ash
Bottom Ash
Filterable Solids
Filtrate
Bottom Ash
Scrubber Water
Filterable Sol Ids
Filtrate
Bottom Ash
Bottom Ash
Bottom Ash
Scrubber Solids

Economizer Ash
ESP Ash
ESP Ash
ESP Catch
Economizer Ash
ESP Ash

Settling Chamber Ash
Primary Chamber Ash
Baghouse Dust
Settling Chamber Ash
Fly Ash/afterburner

Baghouse Ash
Baghouse Dust

Baghouse Dust
Filterable Solids
Filtrate
Radlan/Source
Radian/Source
Radian/Source Test
Test
Test
Radian/Region V
Radian site survey

Radian site survey
Radian site survey
Radian site survey
Radian site survey

Radian site survsry
Radian/Source Test
Radian site survey
Radian/Region V
Radian site survey
CARB

Radian/Source Test

Radian site survey
Radian site survey
Radian/Region V

Radian/Source Test
Radian site survey

Radian/Source Test
Radian/Region III

Wood-fired Boiler








Drum and Barrel
Reclamation
Incinerator




CRF-C
WFB-A

WFB-B
WFB-C
WFB-D
WFB-E
WFB-F
WFB-G
WFB-H
DBI-A
DBR-A.
DBI-C
DBI-D
DBI-E


Afterburner Ash
Baghouse Dust
Bottom ash
Dust
Baghouse Dust
Scrubber Water
Multlclone Ash
Multlclone Ash
Multlclone Ash
Multlclone Ash
Bottom Ash
Bottom Ash
Bottom Ash
Bottom Ash
Bottom Ash
(Continued)
8-3
Radian/Region III
Radian/Source Test

EPA Region X
Radian site survey
Radian/Region V
CARB
CARB
CARB
CARB
Radian site survey
Radian/Source Test
Radian site survey
EPA Region V
Radian/Region V



-------
TABLE 8-1.  SUMMARY OF ASH SAMPLING SITES (Continued)
Combustion
Device Category
Hazardous Waste
Incinerator
Hospital Waste
Incinerator
Open Burn
Sulflte Liquor
Boiler
Woodstove
Spreader Stoker
Boiler
Commercial Boiler
Utility Boiler
Apartment House
Incinerator
Charcoal
Manufacturing
Cement Kiln
Radian
Site
Code
HWI-A
HWI-B
HWI-C
WIH-A
WIH-B
WIH-C
WIH-D
OB-A
OB-B
SLB-A
SLB-B
SLB-C
WS-A
WS-B
WS-C
SSB-A
SSB-B
SSB-C
CB-A
CB-B
UB-A
UB-B
UB-D
AHI-A
AHI-B
AHI-C
AHI-0
CM-A
CM-B
CK-A
CK-B
Sample Type
Solid/Slurry
Bottom Ash
Scrubber Water
Scrubber Water
Bottom Ash
Fly Ash
Bottom Ash
Primary Ash
Flyash
Ash
Filterable Solids
Filterable Solids
Scrubber Water
Solids/Filtrate
Bottom ash
Bottom Ash
Bottom Ash
Bottom Ash
Multi-Clone Ash
Multi-Clone Ash
Bottom Ash
Multi-clone ash
Multi-Clone Ash
Fly Ash
Baghouse Oust
ESP Catch
Bottom Ash
Bottom Ash
Bottom Ash
Bottom Ash
Bottom Ash
Afterburner Ash/
Boiler Fly Ash
Bottom Ash
ESP Catch
ESP Catch
Sampling Organization3
Radian/Region V
Radian/Region V
Radian/Region V
Radian/Region V
Radian/Region V
CENTEC
MDEQ.E
OSDA
FDER
EPA Region X
Radian/Region V
EPA Region V
Radian
Radian
RTI-Rad1an/Source Test
EPA Region IV
Radian
SODHEC
EPA Region IV
Memphis & Shelby County
Health Department
Radian
Radian/Region V
Radian
Hamilton County APCB
CENTEC
CENTEC
CENTEC
SODHEC
Region VII
Radian/Region V
Radian/Region V
                       (Continued)
                           8-4

-------
                  TABLE 8-1.  SUMMARY OF ASH SAMPLING SITES (Continued)
Combustion
Device Category
Municipal Solid
Waste Incinerator
Briquet Charcoal
Grill
Industrial Solid
Waste Incinerator
Rotary K1ln Dryer
Radian
Site
Code
MSWI-A
MSWI-B
MSWI-C
MSWI-D
BC-A
ISW-A
ADK-A
Sample Type
Solid/Slurry
ESP Catch
Bottom Ash
Filtered solids
Filtrate
Scrubber water
ESP Catch
Scrubber water
ESP Catch
Bottom Ash
Uncombusted Charcoal
Bottom Ash
Baghouse Dust
Sampling Organization3
Radian/Region V
Radian/Region V
Radian/Region V
Radian/Region V
EPA/OAQPS
Radian/Source Test
CARB
aCARB = California A1r Resources Board
 CENTEC = EPA Region II
 MDEQE = Massachusetts Department Environmental  Quality Engineering
 OSDA = Oregon Department of Agriculture
 FDER = Florida Department of Environmental  Regulation, Bureau of Air Quality Management
 SCDHEC = South Carolina Department of Health and Environmental Control
 Hamilton County APCB = Hamilton County Air Pollution Control  Board (Tennessee)
                                            8-5

-------
8.2  PREVIOUS WORK
     Previous work has not demonstrated ash sampling to be a proven, reliable
screening technique to provide conclusive estimates of CDD emissions in the
flue gas from combustion sources.  A summary of the ash data available in the
literature was presented previously in Table 3-2.
     A considerable fraction of CDD emissions is hypothesized to be present in
the gas phase or adsorbed onto the fine particulate.  If this is true, CDD
would generally not be concentrated in bottom ash or fly ash collected in an
air pollution control device.  Data are available that indicate there is
likely to be selective partitioning of homologues between the flue gas and the
particulate matter (fly ash and bottom ash).    These data imply that tetra
chlorinated CDD's concentrate in the flue gas, while the more chlorinated CDD
homologues concentrate in the particulate matter.
     Another study has shown that PCDD's tend to be enriched on smaller
particles.     Small particles may not be captured by some control devices and
may be emitted from the source.
     A third study has suggested that the absence of TCDD's in the ash samples
does not guarantee the absence of TCDD's in the flue gas.  At one source,
where both stack and ash data were collected, tetrachlorinated CDD's were
found in the flue gas, but no tetra through octa chlorinated CDD's were found
in the fly ash (i.e., a "false negative" condition).220
     Table 8-2 shows flue gas and fly ash analyses for several  combustion
categories where both types of samples were taken.  The data suggest that PCDD
ash analysis may be a useful indicator of the presence of TCDD emissions.
However,  based upon these data there is no quantitative relationship between
the PCDD content of the ash samples and the PCDD concentrations in the flue
gas emissions.

8.3  ASH SAMPLING SITE SELECTION
     The objective of the ash sampling site selection effort was to select
sites in combustion categories with potential for CDD/CDF emissions.  Also,  at
least three sites were selected within each category, if possible.
                                      8-6

-------
                                   TABLE 8-2:  =  -.  '
         COMPARISONS OF FLUE GAS AND FLY ASH DIOXIN AND DIBENZOFURAN CONTENTS
Stack (nq/m3) ESP Ash (nq/q)
Source Category
Municipal Solid
Waste



•





Hospital
Incinerator
Boilers Co- firing
Waste
Carbon Regeneration
Reference 1
43 (Italy #2)
43 (Italy #3)
43 (Italy #4)
43 (Italy #5)
43 (Italy #6)
49 (Canada)
92 (Hampton, VA)
168 (Philadelphia)
NW #1
168 (Philadelphia)
NW 12
178 (Zaanstad)
181 (Ontario)
31A (Canada)
41
13 (Cincinnati, OH)
PCDD
48,997
7509.9
4409.4
1030
587.8
107
2,284
2,366
745
1257.7
2,687
69
76.5
47. 7h
PCDF PCDD
599.9
7.32
3537.2
0.878
5.86
143 31?
ob
1307C
11,000 800
2,533 1,975
775 1,333
1116.5 5,173
23
155.6 3,369d.
1.6*
2.7f
48. 6g
103h 2.2M
PCDF
-
-
-
-
133h
°c
161C
3,000
1,500
1,066
2080.6
46.75
3,068^
3.6*
11. 1
-
1.7^
^Economizer ash.
 Incinerator ash.
^Boiler ash.
aFly ash from duct between the low temperature stack and heat exchanger.
^Combustion chamber ash.
£Bottom ash.
?Furnace ash.
JTCDD/TCDF scan only.
 Cyclone ash.
                                          8-7

-------
      The following factors were considered in ranking source categories  for
 ash sampling:
       a.  PCDD in feed,
       b.  precursors in  feed,
       c.  chlorine in feed,
       d.  combustion temperature,
       e.  residence time,
       f.  oxygen availability,
       g.  feed processing,
       h.  supplemental fuel,
       i.  valid testing  already performed,
       j.  size of source category.
      Details of the ranking process  were  discussed  in Chapter 4.   Also,  sites
 in  the Rank C  and Rank D categories  were  included in  order  to expand  the
 CDD/CDF data base for these categories.   Table 8-3  presents the final  ranking
 of  the combustion device categories  which was used  in selecting the sites  for
 the ash sampling program.   Also included  in  the table are the number  of  sites
 from which  samples were  collected for  each category.   The most ash samples
 were collected from Rank A and  Rank  B  categories.
      The combustion category rankings  and specific  selection  criteria  for  each
 combustion  source category are  presented  in  Table 8-4.  These category
 rankings and specific  selection  criteria  were sent  to EPA Regional Offices who
 provided inputs  regarding  the potential sampling sites.  The  Regional  Offices
 were asked  to  complete a questionnaire for each site  selected.  These
 questionnaires detailed  such information  as  feed material,  feed rate,
 pollution control  device,  types  of ash generated, potential for CDD and  CDD
 precursors  in the  feed,  combustion parameters  and ease of sampling.  The sites
that were recommended by EPA Regional Offices were  screened by OAQPS,  and a
mixture  of  sites believed  to be  generally "typical"  of those  in the rest of
the  category,  and  sites thought  to have a greater than average potential  for
CDD  emissions were  selected for  sampling.
                                      8-8

-------
        TABLE  8-3.  COMBUSTION SOURCE  CATEGORIES  SAMPLED  IN  ASH  PROGRAM
                                (August,  1985)
     Source  Category
                                                Number  of Ash  Sites
 Rank A
     Sewage  Sludge  Incinerators
     Black Liquor Boilers
 Rank B
     Industrial  Incinerators
     Carbon  Regeneration (industrial)
     Wire Reclamation
     Wood Boilers (firing PCP treated or salt-laden wood)
     Drum and Barrel
     Secondary Copper Smelters
 Rank C
     Hospital Waste Incinerators
     Charcoal Manufacturing
     Wood Stoves
     Small Spreader-Stoker Coal Boiler
     Chlorinated Organic Waste Incinerators
     Cement/Lime Kilns & Dryers Cofired w/Chlorinated Organic Wastes
     Commercial Boilers Firing Fuels Contaminated with
       Chlorinated Organic Wastes
     Open Burning
     Apartment House Flue-fed Incinerators
 Rank D
     Municipal Solid Waste (MSW) Incinerators
     Industrial Boilers Cofiring Wastes (Utility Boilers)
 Unranked                                                '
     Briquet Charcoal Grill
     Sulfite Liquor Boilers
     Residential Oil Burners Burning Waste Oil
                                                              9
                                                              6

                                                              1
                                                              3
                                                              4
                                                              8
                                                              5
                                                              2

                                                            V
                                                              2
                                                              3
                                                              3
                                                              3
                                                              3
                                                              2

                                                              2
                                                              3

                                                              4
                                                              3

                                                              1
                                                              3
                                                              1
Rank A -
Rank B -
Rank C -
Rank D -
Large source categories (greater than 1 million tons of fuel
and/or waste burned annually) with elevated dioxin precursor
contamination or feed/fuel.  These categories have a high
potential to emit TCDD, and population exposure is expected to be
relatively high compared to other source categories.
Small source categories (less than 1 million tons of fuel and/or
waste burned annually) or source categories with limited dioxin
precursor contamination of feed/fuel.  These categories have a
high potential to emit TCDD, but population exposures are expected
to be low.
Source categories less likely to emit 2378-TCDD.
Source categories which have been tested three or more times.
                                    8-9

-------
          TABLE 8-4. SOURCE CHARACTERISTICS OF INTEREST FOR DIOXIN TEST PROGRAM
        Source Category
      Source Characteristics of Greater Interest
 Apartment house incinerators
 Black liquor boilers
 Carbon  regeneration  furnace


 Cement/Lime  kilns  and dryers


 Charcoal manufacturing

 Chemical sludge  incinerators

 Commercial boilers

 Hazardous waste  incinerators


 Industrial boilers


 Open Burning


 Sewage  sludge incinerators
 Units with  a record of poor combustion  conditions
 and excessive waste loading and/or burning  waste oil,

 Paper pulp  mills  using wood stored in salt  water
 prior to  pulping  and/or producing  bleached  paper
 where bleach effluents are  combined with  the  black
 liquor.


 Activated carbon  used  to treat  industrial waste
 streams of  chlorinated organics.

 Firing chlorinated  organic  wastes  (e.g.,  Cl-phenols,
 Cl-benzenes).


 Using wood  scrap  treated with chlorinated organics.

 Burning wastes containing chlorinated organics.

 Burning chlorinated organic waste  contaminated fuels.

 Firing chlorinated  organic  wastes  (e.g.,  Cl-phenols,
 Cl-benzenes).


 Smaller than 50 million  Btu/hr  with a particulate
 control device.

 Open  burning of fields or wastes which  have been
 treated with chlorinated organics.

 All waste treatment plants  servicing areas with
 industry  manufacturing or using chlorinated organics.
Waste incinerators, hospitals      Burning chlorinated organics.
Wire reclamation incinerators
Burning PVC-coated wire or wire contaminated with
PCB's.
Wood/bark boilers
Burning pentachlorophenol treated wood or salt-laden
wood.
                                          8-10

-------
8.4  SAMPLE ACQUISITION
     The ash samples were picked up from the selected sites by several
different organizations including EPA Regional Offices, other local air
quality agencies, OAQPS/EPA and Radian. Radian collected ash samples during
source tests, source test pre-test surveys, and specific ash program pick-ups.
Table 8-5 lists the specific sampling organizations and number of samples
obtained by each organization.
     To standardize the sampling procedure as much as possible, an ash
sampling kit was developed and distributed to organizations picking up the ash
samples.  The ash sampling kits contained:
      1.  a copy of the ash sampling document,
      2.  two Tier 4-cleaned 950 ml amber glass bottles w/teflon lined lids,
      3.  sample labels, sample integrity seals and teflon tape to seal jar,
      4.  packing material to safely ship samples,
      5.  shipping labels and forms to ship samples back to Radian,
      6.  information, on how to select ash sampling location and sampling
          procedure, and
      7.  a form detailing process information needs.
     The samplers were asked to select a sampling location as far downstream
from the combustion device as possible.  Thus, pollution control device ash
was preferred over bottom ash.  Also, a sampling location before the  ash had
reached a storage bin was preferred (i.e., an ash outfall).
     The types of ash that were collected included:
      1.  fly ash (settling chambers, economizers, afterburners),
      2.  baghouse ash,
      3.  ESP ash,
      4.  bottom ash,
      5.  multi-cyclone ash,
      6.  particulate filtered from scrubber water.
     Scrubber water samples to be  analyzed by Troika were first returned to
Radian where they were filtered.   The filterable solids and filtrate  were then
sent to Troika to be analyzed separately.
                                     8-11

-------
                 TABLE 8-5.  SAMPLING ORGANIZATIONS FOR ASH SITES
 Sampling Organization
Number of Sites Sampled
 Radian
   Source Test Site Surveys
   Source Tests
     Subtotal

 EPA Region  II (CENTEC)
 EPA Region  III (Radian)
 EPA Region  IV
 EPA Region  V  (Radian)
 EPA Region  X
 EPA Region  VII
 OAQPS/EPA (In-house)
           (Radian)

 California  Air Resources  Board  (CARB)

 South  Carolina Department of Health
   and  Environmental Control (SCDHEC)

 Hamilton  County Air Pollution Control Board  (Tennessee)

 Massachusetts  Department  of Environmental Quality
    Engineering  (MDEQE)

 Memphis & Shelby County Health Department

 Florida Dept. of Environmental Regulation,
  Bureau  of Air Quality Management (FDER)

Oregon Department of Agriculture (OSDA)

     TOTAL
           14
           25

            4
            2
            2
           20
            2
            1
            1
            5
            2

            1


            1

            1
                                                               75
                                     8-12

-------
     Each ash sampling site was assigned a unique episode number and a range
of SCC (sample control center) numbers (for example, Episode 2677 and SCC
DQ002000-DQ002099).  Each ash sample was assigned a unique SCC number.  Also,
the samples were checked for proper labeling and packaging before they were
sent to the Troika laboratories.
     Sample, process and shipping information were logged into a process data
and sample tracking table stored on computer disk.  The data recorded for each
site included:
      1.  feed materials (suspected CDD precursors were noted),
      2.  feed rate,
      3.  type and number of samples,
      4.  sampling location,
      5.  characteristics of sampling location (e.g., temperature in ash pit),
          and
      6.  combustion parameters (firebox temp, % CO, % 02) obtained from plant
          instrumentation.

8.5  SUMMARY OF RESULTS
     The data base collected for the Tier 4 program consists of two subsets of
data:  1) process data and PCDD/PCDF ash results for the ash sampling program,
and 2) process data, PCDD/PCDF flue gas data and PCDD/PCDF ash data for the
twelve source test sites.  Two separate data bases were formed because the
source test sites had three samples of each type of ash analyzed and
corresponding flue gas PCDD/PCDF results, while for the ash sampling program
only one ash sample per site was analyzed.  Each data base will be discussed
separately in the following sections.
8.5.1  Ash Sampling Program Data base
     The ash sampling program data base includes process data and PCDD/PCDF
data for each site.  Due to the limited access points in most incinerator and
emissions control systems, plant instrumentation was used to collect the
incinerator and emissions control  device operating data.  For some sites plant
instrumentation was limited and data were not obtained for some parameters.
                                     8-13

-------
     The operating data collected was determined by the type of ash sampled.
For bottom ash, firebox temperature and CO and 02 at the incinerator outlet
were collected.  For flyash  (settling chambers, economizers, afterburners),
and control device ash, the  inlet and/or outlet flue gas temperature,
particulate loading, 02, CO, and firebox temperature were collected, if
available.  Data on the composition of the fuel and the fuel feedrate were
also collected.
     The sites were coded by type of combustion.  A key to the code is shown
in Table 8-6.  The process data collected for the ash sites are shown in Table
8-7.  For the source test sites which are included, an average of the three
test runs is reported.  The coolest combustion zone temperatures (<1,000°F)
were found in wire reclamation incinerators, drum & barrel reclamation
incinerators, spreader stoker boilers, apartment house incinerators and one
hospital waste incinerator.  The average firebox temperature and range for
each category is shown in Table 8-8.
     Combustion parameters such as oxygen and carbon monoxide concentrations
were recorded as available.  Oxygen data ranged from 0.5 to 19% 02; carbon
monoxide data ranged from 0 to 1,190 ppmv.
     The flue gas temperature at the point of collection is summarized in
Table 8-9.  For bottom ash, the firebox temperature or bottom hearth
temperature was used.  Baghouses, ESP's and multi-cyclones had the coolest ash
and afterburners and bottom ash had the hottest ash.
     Due to the volume of information and to aid the reader, the PCDD/PCDF ash
sampling program data base is summarized by site (Table 8-10) and by type of
ash sampled (Table 8-11).  Results are reported for the tetra- through
octa-chlorinated dioxin and furan homologues as well as the 2378-TCDD and
2378-TCDF isomers.  The total PCDD and total PCDF results are the sum of the
tetra- through octa-chlorinated homologues.  The PCDD/PCDF concentrations are
reported in units of nanogram per gram which is equivalent to parts per
billion by weight.
     In general, baghouse ash from wood-fired boilers, primary and settling
chamber ash from wire reclamation incinerators, baghouse ash from secondary
copper smelters and ESP ash from municipal solid waste incinerators contained
                                     8-14

-------
           TABLE 8-6.  SITE CODES
Code
Combustion Device Category
ADK and RKD
AHI
BC
BLB
CB
CK
CM
CRF
DBI and DBR
HWI
ISW
MET
MSWI
OB
SLB
SSB
UB
WFB
WIH
WRI
WS
Rotary Kiln Dryer
Apartment House Incinerator
Briquet Charcoal Grill
Black Liquor Boiler
Commercial Boiler
Cement Kiln
Charcoal Manufacturing
Carbon Regeneration Furnace
Drum and Barrel Reclamation Incinerator
Hazardous Waste Incinerator
Industrial Solid Waste Incinerator
Secondary Copper Recovery
Municipal Solid Waste Incinerator
Open Burn
Sulfite Liquor Boiler
Spreader Stoker Boiler
Utility Boiler
Wood-fired Boiler
Hospital Waste Incinerator
Wire Reclamation Incinerator
Woodstove
                     8-15

-------
      s  si

      if!!*
   5 s5  : » S
       SSSSSS  S  SS  S  5^  S  !2  S SS S  SS




       ssssss  s  Is  s  ss  £  is  a ss s  ss  ^sssss
                -!  2S  S  S"  T  |2  S 5S =  -S-  2-£g!2[
s s s  s-s:
       -
                            -S
    I  ssssss  I  53 I  s=  i  i s  s
                                             si  si sSSi
112 2  ii
: i 2  - -a i
: ss  -3.2 2 s
: •= .  sr s : * s
                         S  SSSSSS  iS
                                                s s  s  ss
                                                 = s
                                                          "2
                SSSSSS  S  Is  5
     g5  -2 S  2 f 3*£'S  S  2"






     • u_  * K "~  •*** *** tf flE K E
                  3? 8  f,
!  12

ill
ill
:


I Si
  **
lill
:l-3
                                       S  S  SS  "SS!





                                       S  5  S)3  2! S = '





                                      if  s  §1  Sis:






                                                 i ^
•                          t  t   s  t             s s        :

KiES:  bb&biii!       fe   =1  5   '  ?      fe  bt  -T        5

IJfl  Ifflll  I   If   ||  rlss'ls'lss  Isll  flasllll!




"                                                          •  ^
         : "ss           5


         jiff  ||sss«  I
             s  g ss s


                i
                              u
                             III! s  ss  ffl^
                                                           :  c

: w  i 4
s a  • is
il  1 --    e-  A  I   11^       ?.i   I-  S  Ssllifil^lis



i|  * |fl  ll  illitlllj  § I Ifs  if  |  ii||i^i|s|ji
* ^  tft ^ 5 **  ** -_.  "•   — —* «• »o ^ ^   9 «.  at »^   M-  .•  ^     « «M »   A ut a
« ^,  *^ ^ j3 MK^sS  33fS£'ow>«.2 S  2^£   S     "*     *• ^     **s
h

: a"

                                                           a  s IS  5
         S~      _    _            H  *   »      ™ —
         :3      J    5            s"=    2^5



         ti      s    Illllllllils'll51*^^^^^^^!

                ^5'5'^^JI!  -a  -g-3 -5  -fi-s  •  "£J  ? •£? •=  ??  a.^.^L^^S'a.'S^S
                     • ••  "•  3 ^ 
-------
  ii £ J

  ii | z
  ii s s
       £ -1 £ 8
         —  ii
       s s.  s
           jaa

           i!
                        £  ->  s
           !t
o
u

u] ii  -Si
o I!  S.'S


S!  "!
—I it
Q_ II  —
£ j. i jT "S'STS I^S 5?"3 5S H'i
X-S3&S s-i a-sa- ssa ass £s £^s 1 J = s S
ii
I iiii II ill £*•*• iii 21 m i|||!
ii
5 & s ii
S--3- 5 :=l_ii
— "S s ^ s w!I3 ™ •£•! i'
Ii I ?i 1 II 11= ii i!
                                    3-17

-------
i
1
I
i
VI


£
S
«
s
g
1
«
r^
to
bl
I













Cwlril htiti
in Etrm Ctiuctifiilici
llllt ftltllt
CO fwllMlltt iHf 02 N Nrticilitt
IritVI ltUin| IF) III |,MV) Ui((i|
i
<*
I
3 «
1 -£

f *i
I
S C
! ^
• »
s
: £ 5
i ?^

g
•
:

i
•
i S

:
i
: s _ -
ff-3
|
*
!| = £
[1-2
i
:
i
•I i
|!
E I
I
i
II
• 1
|
•

i
a
•
I
•
•
•
•
»
s
1
«
•
•
*
i
•
i
i
•
S
•
:
E
•
i
s
1
•
•
IS
95
* !
*i
ii


i
ii

5



^ s
i:
i SP



.^
i
V
_s
II


52

i
11
II
S 5
55
52
55
55
55 S
c* •" «n
Ss I
~ K



SS a

is s
1s ?



T S-
SE ^
§£ S







__ 1
J3 2

•«
aa E

I
! l
« u
J I
5
S
S
S
g


|
i

§
i



^
I
to







S
i

T
s
S
1
S
55
55
55
|S
is


g;


55 i



££ i
l~
~P


•3
•
i
•• *•
! 1
s5 -
3^ i
II 5

Vf
11 i

•s
Ji
S
• s
m
m
&
m


s


> i


£
~
R

u
3

f
*•
"S
•>•
•5*
f

—
2
2 5
1 *I
1 Ii
i .
s
i
i
1
i
i
i
i
:
i
:
«
j
s
!

•
S
•
•
i
:
:
:
•
:
:
:
:
3
3
:
i

!
3
3
:
|
!
:
•
•
m
:
•
•
•















9
3
^
•a
0
+*
g
tn
S
e
o
Ii
'a I
(A
O «^
C*-
s£
M tt
Ii
8-18

-------
      TABLE 8-8.  AVERAGE COMBUSTION ZONE TEMPERATURES FOR THE COMBUSTION
                               DEVICE CATEGORIES
Combustion Device Category
  #
Sites
  Average Combustion
Zone Temperature ( F)
Sewage Sludge Incinerator
Black Liquor Boiler
Wire Reclamation Incinerator
Secondary Copper Recovery
Carbon Regeneration Furnace
Wood-Fired Boiler
Drum & Barrel Reclamation Incinerator
Hazardous Waste Incinerator
Hospital Waste Incinerator
Sulfite Liquor Boiler
Spreader Stoker Boiler
Commercial Boiler
Utility Boiler
Apartment House Incinerator
Charcoal Manufacturing
Cement Kiln
Municipal Solid Waste Incinerator
Industrial Soli'd Waste Incinerator
Briquet Charcoal Grill
Open Burn
Woodstoves
  9
  7
  4
  2
  3
  8
  5
  3
  4
  3
  3
  2
  3
  4
  2
  2
  4
  1
  1
  2
  3
    1400 (1280-1500)
    2200
    1350 (1000-1700)
      NM
    1200
    2000 (1200-2500)
    1280 (1000-1600)
    1900 (1700-2000)
    1400 (800-1700)
    2200 (1950-2500)
     600
    1600 (1500-1750)
    2150 (1800-2500)
     600
      NM
    2600 (2500-2700)
    1800 (1600-2000)
    1340
      NM
      NM
      NM
NM = Not Monitored
                                    8-19

-------
            Table 8-9.  AVERAGE TEMPERATURE IN THE CONTROL DEVICES
                         AT THE ASH SAMPLING POINT
Sampling Point
# of Sites'
Gas Stream Temperature (°F)
Wet Scrubber

Multi -Cyclone
Baghouse
ESP

Settling Chambers
Afterburners
Economizers
Bottom ash
12

8
6
9

4
3
5
26
inlet: 620 (120-1800)
outlet: 120 (65-150)
460 (430-500)
360 (300-450)
350 (240-575)
1 ESP at 1600
900 (850-1000)
1700 (1500-2100)
600 (375-750)
1100 (102-2000)
 Includes sites for which temperature data at the ash sampling point  were
 available.   Data were not available for all  sites.
                                     8-20

-------
 f

 I
       a s « s  88
       l~ —d d  dd
                   388883  8S
               as  «ss  ?ss  Eqssass  sss
               oo  o o o  o o u>  o in o °2  22
                   |ssili
                    22222
           IS
                                 IS  ^33  S.§? R.a.  SS8."  "     s

                                 22  ''^d  020>' °'~  d°'2  d     -
      SSB3  35  ""5
      4d d  to d
      aass
      dddd
               «•  " 22
      300
          *-«i   i_>  S S <
                   §^ ^ n ^n ^^
                   S333S

                  222222
         22  22  22do22



      eseS1-!-!  °. "^  ^°.§§^^
         22  22  22||22
      S§§3  §§

      2^22  33
                             SB
                             as
                             dtn
                             §8
                             So
                             o

                              a
                                      >fflC^  ocvio  cstAr^i^oo  ooo  *r
                                      ^OO  v* ij **  m»>4<^Or«O  OCnto  f^
                                 o    odd »
               3!
               11
               22



               II
               22
                                     885
                                     odd  dor:
                                                             »-
                                 33
                                 22
          2


          S <

          2




          2



          gg  gg


          2s  22
 •SS  *S''S   *  -i -. -;

°22  °2S   "  °°2  s


 8C7 ^^  2 '^ fi  9 Q  O ^^ O  ^^
 • S3   *8 *   . .   . **5 .  S
°do  °orf  °®  °S^  ?
 u^ 2   Q             o
                                       22    2'
                                                   3  030  q
                                 d.2    22
                                           OO^i  &fV)QQr*
-------
•o
at
       (
     R-3ogg




      R S  R 5

      O O  JQ J£J
      8 8  R R

      "' •»  S3 S
88  8 °8S8 £888.  8 88 S 3838 88 B  H83
o o  o oooo 00*00  ornoooofqooo'o  o o o





sl  I s'l  s = ljd,  s'l^I  °§|1=!  ill
-.3
So1
•-« 3  " S " 5 ^^ P J-J 3 S 8
 • o  • M^OOO ss —s o • —i
O _•  O O _• O  • O O -• *^ O
 S  ^  SoS o S
            a
              §3  sl  g  5-3  3 §§!! §?i?
           88 88

           ~= S5S


           aa as
         ssasss
         a o ^ o CNJ in
                       s  ssss


             as |.22 a  Sail
             OQ    SS o  22SS
                                 S S o
                                 o o o
So  •  MoSS 8S9S S»S«»  »5°SS  §sS
--  °  °-°- -s--- ^rt^a°  s'3c>ss  s's's'
                         ss s'sss
       4 un .S O   rn «H


       3|sl   S3
                                   si   -'
                           S
                           o
         ra s a a R s
         s-^a^gs


         ssssss
             do  d  m d d d  dddd  d o o g d f-J g d d d d odd
                  OO  O  0  O £20 O
                   5*3 S  §S§S  S®3^*S ^oe9o 3 o §
                   pp _•  _•_•_•_•  8d_..o| "^^^3- ^^s-
          S1"* ft *"* P O
          o « or= N
^o
Sd  s aia  3- ssss 6



§i  i Idi^ iiii §
             ss § §2§i  §il|  g§ S  S§§|?§ |i§

             S3 S S2S-S  s'SS's  ss s'  S°ssSs sSs
     S   3

     !   !
     1   1

     III
                                          I
                       •o •o  -a

                      ,  2«J3«

                      ?  'a'al'a

                              i!  Ifi
                                f ¥ V 9 9 9 4- S" T T 7
                         8-22

-------
1
c
4-t

I
    i
    I
                      ^v( wj v i — m  CD O 03 c3 35 cK m  OOO
                      t"» ^ o rn f*  o? in CM o V c? o*  OOO
                         «"  'i5*a I
3||l3 §11|
 uSu  o £5 o £3

SSSSS 3333
a s.B. d d 8s.3.8.
RSS    §233
                           SSS

           agass  5588SSS52 asg
           ~3°g*  aagg^a  as|
         'ss""'65  s'ls'l  -4°°'s«* stg*gos"=*  s.||
                              ™    (s|    ^j

                      's.s^* sagss8.5  §§=
                      Jn^imru rtl ,X *s5 ^! X «^ °  PPP
                       ..
                      * o Ij en CM
                                S
                             333
         ^§s|§  § .tag ^is.a s.assa.ss  gsg
         Sss3*3  s°ss Nmo°- "asg^g0  sf.g
           S
           a
           "  sagas  sssasa
           =  oooo  05«oS--
aasa.ssssss.  sssas  assjssa* sss




aassssssgg  s.sgss!  s.?ss33s 333
In    ^^    od  ^^Q0,*"1
         <—I >—*    C**i rH


s.r^sasgggg  ssass?-
jCOOj£Jffl-^»"-«  O O O (N cH

                                     3*3*1
         ~S3 .^"JS"B§is ?•?*"?"?^ "?^^5®.^"^'^ isS'"'
        s.IgeasiSii aagss aiea
        92*2*85° 1111 »ml»° »«-
                           2 3.0*3



                        dgo
2* 23SSS*
                        .
                    — oea
                                  a
              I

              1
                                  I
               ||II  |l|f


               'iiii
                                     HI

til
                                           If
                                             1
                        8-23

-------
1



i
     8 S  9 8 S 8 S S a S 8 S § W 8.8 8. R S. 8 S. § § 8 3 S. 8 8 8 8 8.8. S S 8 8 8. S S S S 8. 8 8 8 8 8 8 S

     °^  °3fg° ° jflMS"* """
    3°,  Bss.igsgsa
    §°  °QS§o-c>1o-0-
     asss^s'
S8sss5ssigigs.sss§§gaga8a§ss=3si
0-00-lt><0sl3S0S0lsSS0S'niT13s^S°-§5=-
                                                          idcJS
S8. 3S?RS3
      ^  .•••S^2««.  ...SSG

                      333
                                                      ! 5S o
                                                      >So
    p
    ga  gs.e.  s  Rg 2s ggg  ssggggsggg
   a        a     333     CCQC scscs   agon   o   -'^Nc:ood



   sas.s.BRS.ss  sassSB^siasssssasassqsssssagsstsssssssss
i
     m o


     Offl
     sa
      f-I  or^tsooio^«r^^ o o o o •

          is o           *—*
           <_l
                              o o o o § o o o' o o o o c5 § -4 m
                                                        o o
    s5



    i;
        2.8.S§g8B.a.S |2g§gg|SaR.gGgg2ggg||g||gg§g§2.|R^8gggg
52. gasjggs. gg  sB§s33   a^9s.8s.o.ss§ss liss  ^i3.s.R.s.8.s.i3«3
|= l=sH- 33  alssp   S0s0l003ss|| HSS  °s°'<00-ssiil
n
                           !l
    gllllll
     fj_§53 3 3

     (3 CD ffi UJ lu DJ i
                !2 SS22
            i  iiiiiliiliiiiii
                              3-24

-------
I
I
£


I

S
I
     sas.ssssa sss  asesssss  3
     ddddodod r^std  d<>3ddda>t-d  d
                                stqses  s.ssss.ass.  sassssa  a
                                £J d V m d  ddddodod  d d — — —' — —'  —'
S§3SSS33  ^^S
d22*dd2*22  "  d

ISilSlil  |fj|


            sas
                                     Q ^-* r-i Q at r*i fcl r-t i—J r-i r-T^r^
                                     5§fSI |1|1§1§|
                                        2H   2  32d2d2
 •O«»"---  O *O •
o  oooooo  goo



t^tffisjcicDi  SS3
                                   8
                                   d
                                   8
                                   d
                               Q W Q Q CM  r^*-ir*ri—»r-if—ii—tm
                                ----*  SSSSSSSS

                                       oSSSSSSS
                                    2  SSSiessi
                                       • d d d d d o <
     gSjfr*joSo l-«S'-<  oSm"*mS*  S
     -• -• -•8 - °' ° ° £ o' d  °' rf ° d d a s d  °
        2d2
                                S^nSSm  %D^fi9'fl'"'3rQ
                                 • • • • •  O O O O O ^ O oJ

                                        d3SSS  S3
                                     asaas 355 3"
                                                             o

                                                             a
898.38338  a. S3 ^38S9S?S 8.
dddddddd  m n.® dujmddgflQO d


s s «? s ^3 s s s  s? 8 ea a s s s a s s? a s
8 d ~; §8 ^ b*S  eg~jd d en oi d d t>J cs d a
     o.  a ~ p
               S , 3'
                               8.S.8.S.S.  S283S888  SSfQa8§5 S
                               jo d V uj d  dddddddd  o? c4 jg CN o* IN iX co


                               Q CD SB Q fc  ^t^rifflfi*"11"^  QQCS^QO o
                               en a o 5f o  ci o g o o o sg g  "i ^ S . £3 . *~1  .
                               tn d d d d  o d ^ d d d ^ ^  <<
    -  •     -                s
^ r^ ^ ^n ^ ^^ ^i ^n  O O ^ ^ O ^^ f"t ^i Q O (™* ^^
§ = siisll  3i!| ISiSS&g §
                             s
                                      •  ^J O ** ^** Q )O CQ 5(  *^ 3 cK ^» S *O O  rn
                               8
-------
 higher levels of PCDD/PCDF.  PCDD/PCDF concentrations for these source
 categories ranged from 100 ppb total PCDD or total  PCDF to 3400 ppb total  PCDD
 and 19000 ppb total PCDF.  It is important to note  that these results are  for
 one ash sample collected for each site.  Analyses of multiple ash samples  from
 one site are available for the source sites only.
 8.5.2  PCDD/PCDF Ash Results for the Source Test Sites
      The results of the PCDD/PCDF analyses of the ash samples from the source
 test sites form a more detailed data base.  First,  corresponding flue gas  data
 are available.  Also,  three runs were collected for each type of ash which
 allowed for the variability of the results to be analyzed.
      The average ash and flue gas PCDD/PCDF results are summarized for the
 source test sites in Table 8-12.  Ash samples were  not collected at
 Site BLB-A because particulates at this site was controlled with a wet bottom
 ESP.   At Site BLB-B,  a dry bottom ESP was used,  but the ash sampling location
 was inaccessible.
      The ash results  are summarized by congener for each test run  in
 Table 8-13.   The flue  gas and incinerator operating results were previously
 summarized in Table 6-2.

 8.6  STATISTICAL ANALYSES OF  TIER 4 ASH DATA BASES
      The purpose of the  statistical  analyses was  to determine if significant
 relationships were  present in the data.   The rationale  for  selecting  the
 parameters analyzed and  the type of statistical  analyses  used were  explained
 previously in Section  6.0 and are only described  briefly  here.
      Based on previous studies,  certain  variables have  been  hypothesized to
 affect CDD emissions which were  listed  in  Table 6-1.  Of  these variables,
 combustion temperature,  flue  gas  temperature at the  ash  sampling point and the
 oxygen content of the flue gas were  collected for most  of the .ash sites and
 used  in  the  analyses.
      Non-parametric rank-order statistical analyses were  used  for the ash
data  bases because  the non-parametric tests  require fewer assumptions about
the distribution of the population being considered.  The ash  data  base is not
well-distributed across all possible values  (i.e., order of magnitude or more
                                     8-26

-------
a

c o

  o
g
to
S
o
tr
ui
               o
               a


               ?
               oo
               §
               £

               co

               S
               U)

               .c
               in
               o
               •H

               CO
            O t-
            o o
3 «
O +>
          O —I
           •  *
          o o
§   i
                                          S
                                  in
                                  i
                                m o
                      ooo
                      ZCM
                    ii§
                           p
                           2
                           01 o) o     rH rl tC
                                                                •«r t-«    CM m o
                                                                N      in ts -H
                                                                si
                                                                            s
                                              t
                                            £•5      £•
                                            «g
                                            £• w      t*   *• <0    •*- 4*

                                           ^"T«   ^    II    II
                                                  in in  in      in
                             ?    V      ff        f
                             S?    S      25 E        S
                             comc-ix        x
                                                                            <°<   <
                                                                             i  i    i
                                                                            8
                                                                            m
                                                    in         in
                                          co

                                                                                            O in
                                                                                            o 10
                                                                                            in
                                                                                            0) en
                                                      "S
                                                      o in
                                                      t.
                                                      o o
                                                      "S5

                                                      O +>

                                                      5 .a in
                                                                                   SS^
                                                                                   SS5

                                                                                   -&^
                                                                             . *  * *~

                                                                             O +J •(-" CD



                                                                             S -O -O S

                                                                            r-^- ^- §
                                                                               000

                                                                               0  U g

                                                                             o t-  £ a.
                                                                            **• ®  je
                                                                                         E a u
                                                                            I- C

                                                                            o S
                                                                                         O CO CO      II  II  II  II  II II  II

                                                                                        ,— +J .p O    CE >-< CD 3: CD i— U,
                                                                                                     mco
                                                            8-27

-------
CE


1
LU
t—

bo

ft
I
         gogo
         z~z
         liil
         aaas
                        -JPJ
                             iffl:
                          Qtcj-  «5"i-^ °i ^o?". g<222  22 2S
                                -  , CMUJ  • - • "'
                          asgr°s  §§il §1?

                >2!
                                               OJ-1QQ.
         aaaa
                                   SR'-S S^CMS
                             ojujujoj
         aaaa a
         ssss s
ssss


SSSESE
                                      HR ^"J8?^
                  ooo   o ** o
                  .-too   o
r-» i-if-i  ^

§11
sss
S S  >«l ^ °i Q
                          00a°
                                            s  aa  aasa
                                                            asaa
                                                      O z p in  o o o m
                                                      1  II  §.!!§
                                                      s  as  aaaa
                                                      a aa  aass
          ncRwtw  tj  Q Q QQ.QQ
                a  as ssas
                        a a
                            VQ U} i—i (•*;

                            SgSjS
                            m CNJ W o5
                                                           m
                                                         §^i
                                                                  P-38&
                                                         3SSS
                                       — 
-------
       s a B a ssss 5 § "2<
       dddd dddd s 5 ^ S
                             o o P-* r-  10 «»   ao —• P-J ^
                                                                dddd
I
       £3 _4 .-4 £1 jo § Q jS

       dddd dddd
                     *>.<*>. ™.™.
                      Q O Q O  ~^ •""•  f"*i r—t r-t m
                      8 2 fi 3   ^  ^   9 S3 iZ?  P R 55 P  •H^-i^H.-t  OQxOO
                      C** V W CD   (*1  m   (T> Q
       dddd  dddd  S3 S N *"
                                    «o a <==>   '          '
SSSS  dddd oiincstn inWm«v
                *


^j «•« 3 *M   • •  • 3  • •  • P 8 S S 8

SSSS      °     N -J---J



sc z z z  z z ^^ ^H o o o ^o S 5 S 9'

          do dddd dddd
         ooo  NNmm  ^d^d S2SS





         SS°  SSSS  ssss 3333


to*-4Mao  CM jo (vj u>  m M ^r r»  O5S5S5S nc?C3C?
o   d  o-|d|  riri -.  |§§§ |S||

                      • do'd Qodg
                     U_l
                                                 3333
                                                                        il
                                                                        J
                                                                        I'S
                                                                        II
                                                                       5^=^ *
                                                                       D 3 jj i
                                                                       *n\
                                                                         , .

                                                 VVVV V99V  VVVV
                                    8-29

-------
differences between values from different sites) and for this reason,
non-parametric analyses are believed to be more valid.
     A statistical software package was used to perform the analyses.  The
software generates a matrix of correlation coefficients corresponding to all
the possible combinations of the parameters to be analyzed.
     In evaluating the results, it was assumed that if the coefficient for a
possible combination was greater than or equal to 0.85, then a strong
association existed between the two parameters.  Coefficients between 0.7 and
0.85 indicated a moderate association and coefficients between 0.5 and 0.7
indicated a weak association.  Negative coefficients indicated an inverse
relationship.
     The source test ash data and the ash program data were analyzed
separately.  The source test ash data were analyzed for flue gas-ash and
operating data-ash relationships.  The ash program data were analyzed only for
operating data-ash relationships.  The results of the analyses are discussed
in the following sections.
8.6.1  Non-parametric Rank Order Statistical Analysis of the Source Test Ash
       Data
     The PCDD/PCDF data for the flue gas and ash as well as the process data
were correlated, first, by not differentiating between the type of ash and
secondly, by sorting the data into bottom ash and fly ash.
     The first correlation resulted in weak correlations at best (R < 0.7).
Examination of the data base indicated that bottom ash was a good example of
ash for which the PCDD/PCDF in the ash did not correlate with the PCDD/PCDF in
the flue gas (i.e., flue gas PCDD/PCDF did not increase as bottom ash
PCDD/PCDF increased).
     Then, the correlation was performed again, excluding the bottom ash data.
The associations were significant for this correlation and are summarized in
Table 8-14.  The control device ash PCDD/PCDF level was found to be strongly
associated with uncontrolled flue gas PCDD/PCDF levels and weakly associated
with controlled flue gas PCDD/PCDF levels.  Also, the maximum flue gas
temperature prior to collection was found to be inversely related to PCDD/PCDF
levels in the flue gas and control device ash.  It should be noted, however,
that these correlations were drawn based on data from five sites in five
                                     8-30

-------
          TABLE 8-14.   EXAMINATION  OF ASH/FLUE GAS RANK NON-PARAMETRIC
                              SPEARMAN  CORRELATIONS

           Assumptions:   Bottom  ash results not  included  in database.

                          Strong Association:  R > 0.85
                      Moderate Association:  0.7 < R < 0.85
                         Weak  Association: 0.5 < R < 0.7
       Parameter
Maximum  flue gas  temperature
  prior to collection  by  control
  device
           Relationship
1.  Moderate inverse association with
     uncontrolled flue gas PCDD/PCDF
     levels.

2.  Weak inverse association with
     controlled flue gas PCDD/PCDF
     levels.

3.  Weak inverse association with
     control device ash PCDD/PCDF
     levels.
Combustion Temperature in
 Firebox
Strong association with uncontrolled
 flue gas PCDD/PCDF levels
Oxygen level in the flue gas
Moderate association with controlled
 flue gas PCDD/PCDF levels
Carbon monoxide level in the
 flue gas
Moderate association with controlled
 flue gas PCDD/PCDF levels
Control device ash PCDD/PCDF
 level
    Strong association with uncontrolled
     flue gas PCDD/PCDF levels

    Weak association with controlled
     flue gas PCDD/PCDF levels
                                      8-31

-------
different source categories  (SSI-B, WFB-A, BLB-C, CRF-A, and MET-A).  The type
of ash analyzed included baghouse ash, ESP ash, and filterable solids from
scrubber water.
8.6.2  Non-parametric Rank Order Statistical Analysis of the Ash Sampling
       Program Data
     Based on the results of the source test data statistical analyses, the
ash sampling program data was first sorted by ash type and then correlated.
This step was necessary, since approximately 40 percent of the sites had
bottom ash samples analyzed.  Multicyclone ash, baghouse ash, ESP ash and
filterable solids from scrubber water were the next most prevalent types of
ash collected at 10 percent  of the sites, each.  The remaining 20 percent of
the samples were afterburner ash, economizer ash, and fly ash.
     The lack of reported operating data limited the sampling parameters to
temperature of the flue gas  at the sampling location, and firebox temperature.
The results of the statistical analysis by type of ash are summarized in
Table 8-15.  For ashes other than bottom ash, flue gas temperature at the
control device was weakly associated with the PCDD/PCDF level in the ash.
8.6.3  Summary of the Ash Sampling Program
     The ash sampling program attempted to qualitatively determine the
relationship between PCDD/PCDFs in the flue gas and PCDD/PCDFs in ash.
Corresponding ash and flue gas data were collected from ten sites which
included eight types of incinerators and four types of ash.  The data were
analyzed using non-parameter rank statistics excluding the bottom ash data.
The following conclusions were reached:

     1.   Control  device ash PCDD/PCDF results are a qualitative indicator for
          the presence of CDD/CDF in the uncontrolled flue gas.

     2.   Control  device ash PCDD/PCDF results are a weaker qualitative
          indicator for the presence of CDD/CDF in the controlled flue gas
          than CDD/CDF in uncontrolled flue gas.
                                     8-32

-------
              TABLE  8-15.   EXAMINATION OF ASH SAMPLING PROGRAM RANK
                      NON-PARAMETRIC SPEARMAN CORRELATIONS
                             SORTED BY TYPE OF ASH

                       Assumptions:  Sorted by Ash Type.

                         Strong Association:  R > 0.85
                      Moderate Association:  0.7 < R < 0.85
                        Weak Association: 0.5 < R < 0.7
Ash Type
       Parameter
                             Relationship
Bottom Ash


Baghouse Ash
No significant correlations
Firebox temperature and
 flue gas temperature at
 control device
                       Weak association with
                        PCDD/PCDF levels in the
                        ash
Economizer Ash
 and Afterburner
 Ash
Insufficient cases for
 valid correlation
ESP Ash
Flue gas
 control
temperature
device
at
Weak association with
 2378-TCDD equivalents
 in ash
Fly Ash
Firebox temperature and
 and flue gas temperature
 at control device
                       Moderate association with
                        2378-TCDD equivalents
                        in ash
Filterable Solids
 from Scrubber
 Water
    Inlet temperature
    to scrubber

    Outlet temperature
    to scrubber
                       Weak association with
                        PCDD/PCDF in ash

                       Moderate inverse
                        relationship with
                        PCDD/PCDF in ash
Multicyclone Ash
Flue gas
 control
temperature
device
at
Weak association with
 PCDD/PCDF in ash
                                     8-33

-------
     6.
Bottom ash is not a good indicator of PCDD/PCDF in the flue gas
because of the inconsistent results observed between sources.  Also,
one case of false negative results (i.e., not detected in the bottom
ash but detected in the flue gas) did occur for Site SSI-A.

False negatives and false positive results did not occur for the
control device ash (i.e., PCDD/PCDF were always detected in the
control device ash when measured in the stack gases.)

The maximum temperature achieved in the process stack, conveyor, or
firebox prior to collection of the control device ash is inversely
related to the PCDD/PCDF levels in the uncontrolled flue gas,
controlled flue gas and control device ash.

The flue gas temperature at the control device is weakly associated
with the PCDD/PCDF levels in control device ash.
8.7  EVALUATION OF RANKING OF SOURCE CATEGORIES
     In order to evaluate the original ranking of the source categories as
potential emitters of PCDD/PCDF, the ash results were converted to an
equivalent basis using 2378-TCDD toxic equivalents.  The original ranking was
presented previously in Table 8-3.
     The 2378-TCDD toxic equivalency factors used were developed by EPA and
were previously presented in Table 5-3.  The factors rank the toxicity of a
congener relative to the toxicity of 2378-TCDD which is given an equivalency
of one.  The sites are ranked in descending order by 2378-TCDD toxic
equivalents in Table 8-16.
     Of the 75 sites sampled, 19 of the sites had 2378-TCDD toxic
equivalencies above 1 ppb.  The sites above 1 ppb were mostly from the B
category with two A category sites, one C category site and one D category
site.  Using these results along with the conclusion that PCDD/PCDF in the
control device ash may qualitatively indicate PCDD/PCDF in the flue gas, the
                                     8-34

-------
          U LM *£•
           i  3E a
            5* u.
          s  s
 a)
 o
 a
 <_»
 co
 a
          -   S
S
o
>-
10
t—4
 I
CO
         is 3


                                               !9.


                                                                          SSS32SS   S   S 3 S   g  r7f7S3
                                       i 04   —«
s. sssasaa   sssass   ssss   ssrsssg-sssss   2   ~K
|«3°S|°5   s~~~-~   s""d   R 2~ -' - - ~ --' °' 2 °   •   °"2
                                                                                                        -   «_«,-

    SSSSgS   2SSSSS   SSSS   SSSSSgS2SS«S
    5 "" 2 s °~ =   « a s s s a   •= 5 « «   « o « - « - 0- <: « «• o 0-
                                                                                           «=•  «~     -  -• ---
                                                                                                "
                                                                                       8   s   ° = sss


                                SSSS22SSSSSS5?
                                -ooooo?^ o"-;oo^
                            S:

                            *
                                     "
                                  « -^
                               "5-a
                               l!
•   i
i   I1
                                                                     00  =
                                                                     J   i€

                              j=    £     M ^= .:; .S S;   ^
                            a ** 3 - * m«"5o«*:5-
                        •i«.'"«.'S*'"'"t*llll"M'--''*-
                        *!!?!9**!4KV4l«c;Mt«w
                                                                                     S fc I
                                                                                     J 5 s 1 £:

                                                                                     illll:
                   St4- C/l ^ LAJ CB Li- Cn 3K S " ife Cu W> En U3 Q- S
                   3*3C»ac^^ac^:»3K cj XT en wi 3c Si IE
                                                                                        tt="2SacS^c«3=*-^3=S
                                                                                        ^
-------
UJ


u
x
o
I
to
*•••*

ci


ui
tft
as-


j.
tj
5
£: &
iv »*-
*• x
g •§;

e
$ S
t u.
i •*
*» *-•
i 1
5«
a! „

^
_ *
£ S
*** £
r
s g
1 *
i §

5 S
« i.
f»* "


>•
•* te.
II
'Is
•H S g
pi


(X

fM
G

rx
o
5
e
g
S
•
S
2

y

s

s
*"*
s

^

o
ted


V*
I


M
&
U


g

O
0

0

g
u
~
-
•
2
g

s
M
0

5

s

y

0
^


•z
s
£

LJ


S











S
3

«



R



—

e
u
i
I

1.
i


u)


s

•


s
^
s
u
i"
-
s
0
s
s

u

s

s

0

0

d


o
m
u.

to
en


KSoS^SggSSggSSgSSSg^ogSgSgggggSSSSgggo^g

SS^S^gSSSSSgSo oSSSS^SSSSSSSSSSS^SSSSSSS
 o o o o o • o o o o • * o o o o o o • . ^ «» . . tad o o « o *dd
ted " . *•* ^^ ta-Mta-ta- te-M
22d^d^d^S^^S^^* SSSSg^8SSSSSSSSoo"^SSSo^oo
O O tad O 'OO • « O ^ ^ ^ ^ - ^^O^O^'^^.^i-'O^ . O O^
• • - -«OOO«X«>OOOO ^ ^ 0 O .O^*^OOOOOOOOOMOOOOOOO
dddd - d o o d d do-ddd2.dddoddddddc^dddddd
S*=SSS88SS5588 S8si258=5sI|i855i5|sSSiis
S S.£2SSS£££^22 2S££ S £ - £2 ££ 22SS2 S o £ S S S. 2 S
I2llilll|llll2|5lilsllllll||lllll|5lll=l
S ll ll li 1 111 I 111^11 11111 ill 1
ssssssssggsssssgsgsgsssgssssssssssssgsgg


2 ss s ss s s £ s^a 2 a
o o o o -^"^^ — ^^Soo^^^S^o^ »"^o^oooo^o^ .^o^tooo
0 ted ted *~* ""* «— •U-r — ^tedte-tedetedO^tedtedtadtadted
S|SSSSS§gSSSSS SS|SSSSSSSSS|5SSSS§SS=SSS



^S^S^^^^^^^s^^s^^?^^?*^^^^^^?*^^^^"1^*^^^^*1-^

• - • • d ^ • o • ^ « • • d • • d d tad * d tad • . . d . d d - d d d > d d
O GO^Otedtad O tedOtadOOOtadOO*_OtadOtadtadC><— >i_ti_rfOh««tad
I £ 1
"5 >-
If !
« ^ en
T3 "O
k~-M«Ck. •(£<£<•«£.£ .«= «E f £ f  — . — — i ^ «•• — — —
LU uj s en •*> *M 3 a 3—33 93
£«= = ,= £«:=.:*:

liiiiilililiiSiSISlIiiisiiiilliilsilllll























d
f
£^"
bU
en



rf:


I
o
0
g
0
s
°.
ted
s
0
o
d
§
d
S

g

g
d
S
0
o
ta—
0
"-
s
^



&
ft
f-l
'S
u

CD
























J
-
°
J=
w«
•i in
3 t:
a- chlorinated hoiolog
ion liiits for congenc
W Ul
2 •§
£ •
o mi
JS C
«a a
a at
< — ce
•fl J3


















(IB)
CO
ex
s
u
•e
"
f
«•
f
ai
I
mi
c
«
a)
r>
s .
a< •«
tf
-s s
fs
t s
a s
J £
i_i



























0»
o
n nanograis detected,

.
S -
o
a> *-
11
                                                            8-36

-------
ranking of Category A is not confirmed by the test data since only two of the
15 rank A sites were above 1 ppb of 2378-TCDD toxic equivalents.  The Rank B
and Rank C category assumptions were confirmed.  In addition, three of the
Rank D sites (MSWI-D, MSWI-B and MSWI-C) were in the top 19, as expected.
     Based on the ash sampling program results, source test sites in the Rank
B categories had the highest PCDD/PCDF concentrations (WFB-A, WRI-A, MET^A).
Some ash sampling sites in the Rank D category (MSWI-C,  MSWI-B, and MSWI-D)
also had some of the highest PCDD/PCDF concentrations.
                                    8-37

-------

-------
                                   CHAPTER 9

                                  REFERENCES

*Denotes draft or unpublished reports from which emissions data were available.

  1. Ackerman, D. 6., et a]..  (TRW, Inc.) At-Sea Incineration of Herbicide
     Orange Onboard the M/T Vulcanus.  EPA-600/2-78-086.  April 1978.

  2. Ackerman, D. G., et a!.  (TRW, Inc.) At-Sea Incineration of PCB
     Containing Wastes Onboard the M/T Vulvanus."  (Prepared for U. S
     Environmental Protection Agency, Office of Water) EPA600-7-83-024,
     April 1983.

  3. Ackerman, D. G.  (TRW, Inc.)  Destruction Efficiencies for TCDD During
     At-Sea Incineration of Herbicide Orange.  (Prepared for the Industrial
     Environmental Research Laboratory,  Office of Research and Development,
     U.S.  Environmental  Protection Agency, Research Triangle Park), Contract
     No. 68-02-2660, March 1979.

  4. Addis,  G. and R.Y.  Komai, eds.  Proceedings: 1983 PCB Seminar. EPRI
     EL-3581,  Electric Power Research Institute,  Palo Alto, Ca.  June 1984
     520 pp.

  5. Ahlberg,  M.  et al.   Chemical and Biological  Characterization of Emissions
     from  Coal and Oil-Fired Power Plants.  Environmental Health Perspectives.
     47: 85-102.   1983.

  6. Ah!ing,  B.  and A. Lindskog.   Emission of Chlorinated Organic Substances
     from  Combustion.   From:  Chlorinated Dioxins and Related Compounds Impact
     on  the  Environment.   0. Hutzinger,  R.W.  Frei,  E. Merian, F. Pocchiari
     (eds).   Pergamon Press, Ltd.  1982,  pp.  215-225.

  7. Ah!ing,  B.,  et al.   Formation of Polychlorinated Dibenzo-p-dioxins and
     Dibenzofurans During Combustion  of  a 2,4,5-T Formulation.   Chemosphere.
     Volume  6, 1977.   pp.  461-468.

  8. Alsberg,  T.,  et al.   Chemical and Biological  Characterization  of Organic
     Material  from Gasoline Exhaust Particles.   Environmental Science
     Technology,  Vol.  19,  pp.  43-50,  No.  1,  1985.

  9.  Arsenault,  R.'D.  Pentachlorophenol  and  Chlorinated Dibenzodioxins in the
     Environment  - A Study of Environmental  Fate,  Stability,  and Significance
     When  Used in  Wood Preservation.   Proceedings  of  the American Wood
     Preserving Association.   Volume  72,  1976,  pp. 122-148.

 10.  (ASME)  Study  on  State-of-the-Art  of  Dioxin from  Combustion Sources.
     (Prepared by  Arthur  D.  Little, Inc.)   1981.

 11.  Ballschmiter,  K., W.  Zoller,  C. Scholz,  and A. Nottrodt.    Occurrence and
     Absence of Polychlorodibenzofurans and Polychlorodibenzodioxins  in Fly
     Ash from  Municipal Incinerators.  Chemosphere, Vol.  12,  No. 4/5,
     pp. 585-594,  1983.

                                      9-1

-------
  12. Ballschmiter,  K.,  et  al.   Occurrence  and  Absence  of Polychlorodibenzo-
     furans  and  Polychlorodibenzodioxins in  Fly Ash  from Municipal
     Incinerators.   Paper  Presented  at  the 3rd International  Symposium  on
     Chlorinated Dioxins and Related Compounds.   Salzburg, Austria,
     October 12-14,  1982.

  13. (Battelle Columbus Division.)   Determination of Dioxin  Levels  in Carbon
     Reactivation Process  Effluent Streams.  EPA  Contract No.  68-02-3487.
     January 1984.

  14. Barnes,  D.  G.   Assessing TCDD Emissions from Municipal  Waste Combustors.
     Presented at the 3rd  International Symposium on Chlorinated Dioxins and
     Related Compounds.  Salzburg, Austria, October  12-14, 1982.
                                                  *
  15. Barnes,  D.  6.   Assessing TCDD Emissions from Municipa Waste Combustors.
     Chemosphere Vol. 12,  No. 4/5, 1983, pp. 645-655.

  16. Barnes,  D.  G.   Dioxin Production from Combustion  of Biomass and Wastes.
     (Presented  at  the  "Symposium on Energy From  Biomass and  Wastes VII," by
     The Institute  of Gas  Technology) Lake Buena  Vista,  Florida,
     January 24-28,  1983.

  17. Barnes,  D.G.   Possible Consequences of Sharing  An Environment with
     Dioxins.  Accidental  Exposure to Dioxins:  Human  Health  Aspects.
     F. Coulston and F. Pocchiari (eds).   Academic Press, 1983, pp. 259-266.

  18. Benfenati,  E.,  et  al.  Polychlorinated Dibenzo-p-dioxins  (PCDD) and
     Polychlorinated Dibenzofurans (PCDF)  in Emissions from  an Urban
     Incinerator, 2. Correlation Between Concentration of Micropollutants and
     Combustion  Conditions.  Chemosphere.  1983.  (Volume No.  12, No 9/10,
     pp. 1151-1157).

  19. Boddington, M. J., Douglas V. M., Duncan  C.  E., Gilbertson M.,
     Grant D. L., Hallett D., McClelland L., Roberts J.  R., Singh J.,
     Whittle, M.  Dioxins  in Canada.  Chemosphere. Vol.  12, No. 4/5,
     pp. 477-480, 1983.

 20. Bond,  D. H.  At-Sea Incineration of Hazardous Wastes.  Environmental
     Science Technology, Vol. 18, Nov. 5,  1984. pp.  148A-152A.

20a. Branscome, M. et al.  Evaluation of Waste Combustion in  a Wet-Process
     Cement Kiln at General Portland, Inc., Paulding, Ohio.   EPA Contract No.
     68-02-3149.  February 1985.

20b. Branscome, M. et. al.  Evaluation of Waste Combustion in a Dry-Process
     Cement Kiln at Lone Star Industries, Oglesby, Illinois.   EPA Contract No.
     68-02-3149.  December 1984.

 21. Brenner, K. S.  et al. Dioxin Analysis in Stack Emissions and in the Wash
     Water Circuit During High-Temperature Incineration  of Chlorine-Containing
     Industrial Wastes.   Bull.  Environ.  Contamin.  Toxicol.  33: 153-162. 1984.
                                      9-2

-------
 22. Bridle, T.R., et al_.  The Formation and Fate of PCDD's and PCDF's During
     Chlorophenol Combustion.  For presentation at the 77th annual meetinq of
     the Air Pollution Control Association.  San Francisco, California
     June 24-29, 1984.

 23. Bridle, T. R.  Assessment of Organic Emissions from the Hamilton Sewage
     Sludge Incinerator.  Environment Canada, Environmental Protection
     Service, Wastewater Technology Centre, Ontario.

 24. Brocco, D. et al_.  Evaluation of the Organochlorine Compounds in the
     Biological Sludges and in the Products of Transformation. Annali di
     Chimica 74, 1984. pp. 573-578.

 25' nrnr?°'.DV— — '  InteH aboratory Sampling and Analysis of PCDD's and
               Emiss1ons from Urban Incinerators.   Chemosphere 13 (12):
 26.  Brocco,  D.,  et al.   Polychlorodibenzodioxins and polychlorodibenzofurans
     in the Environment.   Physi co-Chemical  Behavior of Atmospheric Pollutants
     (Proceedings of the  Second European Symposium, Varese,  Italy,
     September 29 - October 1,  1981.)   B.  Versino and H.  Ott,  ed.  1981.
     pp.  82-88.

 27.  Brooks,  C.,  compiler.   Energy from Municipal Waste Research:  A Technical
     Review of Thermochemical  Systems -Workshop Proceedings.  ANL/CNSV-TM-144
     Argonne  National  Laboratory,  Argonne,  IL.  September 1984.   331 pp.

 28.  Brooks,  G. W.   (Radian Corp.)  Summary of a Literature  Search to Develop
     Information  on Sources of Chlorinated  Dioxin and Duran  Air  Emissions.
     Final  Report.  Prepared for  Strategies  and Air Standards Division,  U   S
     Environmental  Protection  Agency,  October 1983.

 29.  Bronzetti, G.,  Bauer,  C., Corsi,  C., ;DelCarratore,  R.,  Nieri,  R.,
         "'          *1'      Study  of TCDD and Ashes  from Urban Incinerator
                                'ng  Yeast °7 Strain'   ChemosPhere'  vol. 12,
30.  (Brown ing- Ferris  Industries,  Inc.)   Executive Summary:  Dioxins  and
     Waste-to-Energy Plants.  July 1982.

31.  Bumb, R. R., et al_.   (Dow Chemical)  Trace Chemistries of  Fire:  A  Source
     of Chlorinated Dioxins.  Science.  Volume 210.  October 1980
     pp. 383-391.

Sla.Bumbaco, M. J.  Report on a Stack Sampling Program to Measure the
     Emissions of Selected Trace Organic  Compounds, Particulates, Heavy
    Metals,  and HC1 From the Royal Jubilee Hospital Incinerator, Victoria,
    B.C.  Pollution Measurement Division, Environmental Protection  Proqrams
    Directorate, Ottawa, Ontario.  April 1983.

32. Bureau of Air Toxics, Division of Air Resources, New York State
    Department of Environmental  Conservation.  Preliminary  Report on Sheridan
    Avenue RDF Plant.  "Answers."  January 28, 1985.
                                     9-3

-------
33. Buser, H. R., et al.  Identification of Polychlorinated Dibenzo-p-dioxin
    Isomers Found in Fly Ash.  Chemosphere. Volume 7, 1978.  pp. 165-172.

34. Buser, H. R.  Formation of Polychlorinated Dibenzo-p-dioxins (PCDDs) and
    Dibenzofurans (PCDFs) from the Pyrolysis of Chlorobenzenes.  Chemosphere.
    Volume 8.  1979.  pp. 415-424.

35. Buser, H. R., et al.  Identification of Polychlorinated Dibenzofuran
    Isomers in Fly Ash and PCB Pyrolyses.  Chemosphere.  Volume 7, 1978.
    pp. 419-429.

36. Buser, H. R., et al.  Formation of Polychlorinated Dibenzofurans (PCDFs)
    from the Pyrolysis of PCB's.  Chemosphere.  Volume 7, 1978.  pp. 109-119.

37. Buser, H. R. and C. Rappe.  Formation of Polychlorinated Dibenzofurans
    from the Pyrolysis of Individual PCB Isomers.  Chemosphere.  Volume 8,
    1979.  pp. 157-174.

38. Buser, H.R. and C. Rappe.  Isomer-specific Separation of 2378-Substituted
    Polychlorinated Dibenzo-p-dioxins by High Resolution Gas Chromatography/
    Mass Spectrometry.  Anal. Chem. 1984, 56, pp. 442-448.

39. Cassitto, L. and P. Magnani.  2,3,7,8-TCDD Monitoring in Flue Gases from
    an Incinerating Plant.  Recent Developments in Mass Spectrometry in
    Biochemistry and Medicine.  (Proceedings of the 4th International
    Symposium.)  Alberto Frigerio, ed.  Volume 1, 1978.  pp. 563-568.

40. Castaldini, C.  Dioxin Emission from Industrial Boilers Burning Hazardous
    Materials.  EPA Contract No. 68-03-3241.  February 1985.

41. Castaldini, C., et al. (Acurex Corporation).  Emissions Testing of
    Industrial Boilers Cofiring Hazardous Wastes - Sites A, D, E, G.
    (Prepared for U.S. Environmental Protection Agency, IERL, Cincinnati),
    1983.

42. Cavallaro, A., s£ al.  Sampling, Occurrence, and Evaluation of PCDDs and
    PCDFs from Incinerated Solid Urban Waste.  Chemosphere.  Volume 9, 1980.
    pp. 611-621.

43. Cavallaro, A., et al. Summary of Results of PCDD Analyses from
    Incinerator Effluents.  Chemosphere, Volume 11, No. 9, 1982, pp. 859-868.

44. Chin, C. et al.  Polychlorinated Hydrocarbons from Power Plants, Wood
    Burning, and Municipal Incinerators. Chemosphere 12: 607-616, 1983.

45. Choudhry, G., et al.  Mechanisms in the Thermal Formation of Chlorinated
    Compounds Including Polychlorinated Dibenzo-p-dioxins.  Proceedings of
    Workshop held at Institute Superior de Fanita, Rome, Italy, October 1980.

46. Choudry, G. G. and 0. Hutzinger.  Mechanistic Aspects of The Thermal
    Formation of Halogenated Organic Compounds Including Polychlorinated
    Dibenzo-p-Dioxins.  Gordon and Breach Science Publishers, New York, 1983.
    Chapter 6, pp. 167-186.

                                     9-4

-------
 47. Clement, R. E., A. C. Viau, and F. W. Karasek.  Daily Variations in
     Composition of Extractable Organic Compounds in Fly-Ash from Municipal
     Waste Incinerators.

 48. Commoner B., T. Webster, K. Shapiro, and M. McNamara.  The Origins and
     Methods of Controlling Polychlorinated Dibenzo-p-Dioxin and Dibenzofuran
     Emissions from MSW Incinerators.  For Presentation at:  Air Pollution
     Control Association Annual Meeting, Detroit, Mich.  June 20, 1985.
     Center for the Biology of Natural  Systems, Queens College.

 49. (Concord Scientific Corporation.)   National Incinerator Testing and
     Evaluation Program (NITEP) Volume  4: Detailed Results. A Report to
     Environment Canada.  June 1985.

 50. Cooke,  M. et al_.   Candidate Sampling and Analysis Methods for Twenty-One
     Suspect Carcinogens in Combustion  Emissions.  For Presentation at the
  »   77th Annual  Meeting of the Air Pollution Control  Association, San
     Francisco,  Calif.   June 24-29, 1984.

 51. (Cooper Engineers,  Inc.)  Air Emissions and Performance Testing of a Dry
     Scrubber (Quench  Reactor), Dry Venturi  and Fabric Filter System Operating
     on  Flue Gas  from  Combustion of Municipal  Solid Waste at Tsushima, Japan.
     West County  Agency of Contra Costa County Waste Co.  - Disposal/Enerqy
     Recovery Project.   August 1984.

 52.  Coulston F.,  and  F.  Pacchiari,  eds.   Accidental  Exposure to Dioxins,
     Human Health  Aspects.   Academic Press 1983.

 53.  Crosby,  D. G.  and  A.  S.  Wong.   Photochemical  Generation of Chlorinated
     Dioxins.  Chemosphere.   Volumes,  1976.   p.  327.

 54.  Crummett, W.  B., T.J.  Nestrick,  and  L.L.  Lamparski.   Advanced/Good
     Analytical Techniques  Elaborated on  the Detection  of Polychlorinated
     Dibenzodioxins  in  Environmental  Samples.

 55.  Crummett, W.  B. (Dow Chemical).  Environmental Chlorinated  Dioxins  from
     Combustion - The Trace Chemistries of Fire Hypothesis.   From  Pergamon
     Series on Environmental  Science Volume 5  - Chlorinated  Dioxins  and
     Related Compounds Impact  on the Environment.   1982.   pp. 253-263.

56. Czuczwa,  J.M. and R. A.  Hites.  Environmental Fate of Combustion-
    generated Polychlorinated Dioxins and Furans.  Environmental Science
    Technology, Vol. 18, No.  6, 1984.

57. DaRos, B., et al_.  (Acurex Corporation) Measured Multimedia  Emissions from
    the Wood Preserving Industry.   (Prepared for U.S. Environmental
    Protection Agency, Cincinnati, Ohio) Contract No. 68-03-2567, Task 4028
    March 1981.
                                     9-5

-------
I
              58. Dawson G. W., J. H. Meuser, M. C. Lilga, Dioxin Transport From
                  Contaminated Sites To Exposure Locations: A Methodology for Calculating
                  Conversion Factors.  January 1985.  Prepared for. Office of Health and
                  Environmental Assessment, U.S. Environmental Protection Agency.

             58a. Day, D. R., L. A. Cox, and R. E. Mournighan.  Evaluation of Hazardous
                  Waste Incineration in a Lime Kiln:  Rockwell Lime Company.  EPA-600/
                  S2-84-132.  November 1984.

              59. DeRoos, F. L. and A. Bjorseth.  (Battelle-Columbus) TCDD Analysis of Fly
                  Ash Sample.  (Prepared for U.S. Environmental Protection Agency, Research
                  Triangle Park, NC)  EPA Contract No. 68-02-2686, Task 113. June 15, 1979.

              60. des Rosiers, P. E.  PCBs, PCDFs, and PCDDs Resulting from Transformer/
                  Capacitor Fires: An Overview. From: Proceedings: 1983 PCB Seminar.
                  Research Project 2028, Electric Power Research Institute, Palo Alto, Ca.,
                  June 1984.

              61. des Rosiers, P. E.  Remedial Measures for Wastes Containing
                  Polychlorinated Dibenzo-p-dioxins (PCDDs) and Dibenzofurans (PCDFs):
                  Destruction, Containment, or Process Modification.  British Occupational
                  Hygiene Society, Vol. 27, No. 1, pp 57-72, 1983.

              62. (Dow Chemical) The Trace Chemistries of Fire - A Source of and Routes for
                  the Entry of Chlorinated Dioxins into the Environment.  The Chlorinated
                  Dioxin Task Force, the Michigan Division of Dow Chemical U.S.A. 1978.

             62a. Doyle, B. W., D. A. Drum, and J. D. Lauber.  The Smoldering Question of
                  Hospital Wastes.  Pollution Engineering.  17(7), July 1985.  pp. 35-39.

              63. Dryden, F. E., et a]..  Dioxins.  Industrial Environmental Research
                  Laboratory, Cincinnati, OH.  Nov. 80.  EPA 600/2-80-197.

              64. Dryden, F. E., et al_.  (Walk, Hayde, & Associates).  Dioxins: Volume III
                  - Assessment of Dioxin-Forming Chemical Processes.  (Prepared for the
                  U.S. Environmental Protection Agency.  Cincinnati, Ohio).
                  EPA-600/2-80-158.  June 1980.

              65. Duckett, E. J.  Dioxins in Perspective:  Knowns, Unknowns, Resolving the
                  Issues.  Solid Wastes Management.  May 1981.  24(5) pp. 56-57, 88-89.

              66. Dykes, R. M.  (Radian Corp.)  A Review of Standards of Performance for
                  New Stationary Sources - Sewage Sludge Incinerators. (Prepared for
                  Environmental Protection Agency, Emissions Standards and Engineering
                  Division)  EPA Contract No. 68-02-3816, January 1984.

              67. Eiceman, G. A. and H. 0. Rghei.  Chiorination reactions of 1,2,3,4-Tetra-
                  chlorodibenzo-p-dioxin on fly ash with HC1 in air.  Chemosphere, Vol. II,
                  No. 9, pp. 833-839, 1982.
                                                   9-6

-------
 68. Eiceman, G. A. and H. 0. Rghei.  Presence of Nitro-chlorinated Dioxins on
     Fly Ash from Municipal Incinerators and From Laboratory Production
     Through Reactions Between N02 and T.CDD on Fly Ash.  Chemosphere,
     Vol. 13, No. 9, pp. 1025-1032, 19847

 69. Eiceman, G. A., et aj..  Variations in Concentrations of Organic Compounds
     Including Polychlorinated Dibenzo-p-dioxins and Polynuclear Aromatic
     Hydrocarbons in Fly Ash from a Municipal Incinerator.  Analytical
     Chemistry.   Volume 53, 1981.  pp. 955-959.

 70. Eiceman, G. A., et al.  Ultrasonic Extraction of Polychlorinated
     Dibenzo-p-dioxins and Other Organic Compounds from Fly Ash from Municipal
     Incinerators.   Analytical  Chemistry.   Volume 52, 1980.  pp. 1492-1496.

 71. Eiceman, G. A., et al.  Analysis of Fly Ash from Municipal Incinerators
     for Trace Organic Compounds.  Analytical Chemistry.  Volume 51, 1979
     pp.  2343-2350.

 72.  Eklund  G.,  and B.  Stromberg.  Detection of Polychlorinated Polynuclear
     Aromatics in Flue Gases from Coal  Combustion and Refuse Incinerators
     Chemosphere, Vol.  12,  No.  4/5,  pp.  657-660,  1983.

 73.  (Environment Canada),  Report of the Ministers'  Expert Advisory Committee
     on  Dioxins,  November  1983.

 7*'  !PADLut£er  from Anita  J> Frankel,  Pesticides and Toxic Substances Branch,
     to  Bob  Courson, Dana Davoli,  RE:   Dioxin and Furans in Ross Electric at
     Logan Hill  Incinerator Ash.   Jan.  15,  1985.

 75.  EPA  Memorandum  from: J. R. O'Connor, Strategies  and Air Standards
     Division,   to Dr.  Robert M.  Clark,  Drinking  Water  Research
     Laboratory/WERL.   RE:  Risk Assessment  on Preliminary Dioxin Data  for
     Cincinnati  Carbon  Regeneration Plant.   May  10,  1985.

 76.  EPRI CS  3308.   State-of-the-Art Review:  PCDDs and  PCDFs in Utility PCB
     Fluid.   November 1983.

 77.  Erickson, M. D. et al_.  Thermal Degradation  Products  from  Dielectric
     Fluids.  EPA-560/5-84-009, U.S. Environmental Protection Agency,
     Washington,  D.C.  December 1984.  91 pp.

 78.  Esposito, M. P., et al. (PedCo Environmental).  Dioxins: Volume I  -
     Sources, Exposure, Transport, and Control.   (Prepared for  the U S
     Environmental Protection Agency, Cincinnati, Ohio).   EPA-600/2-80-156
    June 1980.

79. Esposito, M. P. and D. R. Watkins.  Airborne Dioxins:   The  Problem  in
    Review.  Paper Presented at the 73rd Annual Meeting of  the Air Pollution
    Control Association.  Montreal, Quebec.  June 22-27,  1980.
                                     9-7

-------
80. Feeney, A. and J. Jubak.  Dow and Dirty.
    pp. 26-28, November 9, 1983.
Environ. Action, Volume 14,
81. Fennelly, P. F., e_t aj_.  Environmental Characterization of Disposal of
 *  Waste Oils in Small Combustors.  (Draft) U.S. Environmental Protection
    Agency, Contract No. 68-02-3168.  August 1983.

82. Fishbein, L. Trace Organic Contaminants in the Environmental Chlorinated
    Dioxins and Dibenzofurans.  International Journal of Ecology and
    Environmental Sciences.  Volume 2, 1976.  pp. 69-81.

83. (Fred C. Hart, Associates).  Assessment of Potential Public Health
    Impacts Associated with Predicted Emission of Polychlorinated Dibenzo-
    Dioxins and Polychlorinated Dibenzo-Furans from the Brooklyn Navy Yard
    Resource Recovery Facility. Prepared for New York City Department of
    Sanitation.  August 17, 1984.  387 pp.

84. (Fred C. Hart, Associates).  Conclusions of New York's Study on Dioxins
    and MSW Incineration.  Waste Age V. 15, No. 11, pp. 25-30, 1984.

85. (Fred C. Hart, Associates).  Impact of Burning Hazardous Waste in
    Boilers.  Prepared for SCA Chemical Services, Inc.,  August 1982.

86. Frounfelker, R. (Systech), A Technical, Environmental, and Economic
    Evaluation of Small Modular Incinerator Systems with Heat Recovery,
    (Prepared for U.S. Environmental Protection Agency, Office of Solid Waste
    and the California State Waste Management Board), 1979.

87. Gizzi, F., et al_.  Polychlorinated Dibenzo-p-dioxins (PCDD) and
    Polychlorinated Dibenzofurans (PCDF) in Emissions from an Urban
    Incinerator - 1.  Average and Peak Values.  Chemosphere.  Volume 11,
    1982,  pp. 577-583.

88. Gorman, P. G., G. J. Hennon and G. S. Kush.  Summary of Trial Burn
    Results at the SCA Incinerator.  For Presentation at the 77th Annual
    Meeting of the Air Pollution Control Association.  San Francisco, Ca.
    June 24-29, 1984.

89. Graham, S. J. et aj_.  Production of Polychlorinated Dibenzo-p-Dioxins
    (PCDD) and Dibenzofurans (PCDF) from Resource Recovery Facilities -
    Part I.  Sources, Emissions and Air Quality Standards.  Presented at the
    1984 National Waste Processing Conference - Engineering:  The Solution,
    Orlando, Fla., June 3-6, 1984.  pp. 345-376.

90. Gross, M. L. et al.  Analysis of Tetrachlorodibenzodioxins and
    Tetrachlorodibenzofurans in Fly Ash and Bottom Ash.  Department of
    Chemistry, University of Nebraska, Lincoln, N.E.

91. Gschwandtner, G. et al.  Sensitivity Analysis of Dispersion Models for
    Point and Area Sources, JAPCA, Volume 32, No. 10, October 1982.
                                     9-8

-------
  92.  Haile,  C.  et al (MRI).   Assessment of Emissions of Specific Compounds
      from a  Resource Recovery Municipal Refuse Incinerator.   1984.   EPA
      560/5-84-002.

  93.  Haile,  C.  L.,  et a]..   (MRI)   Comprehensive Assessment of the Specific
      Compounds  Present in  Combustion Processes.  Volume 3, National  Survey of
      Organic Emissions from  Coal  Fired Utility Boiler Plants,
      EPA-560/5-83-006, September  1983.

  94.  Haile,  C.L.,  et al.   Emissions  of Organic Pollutants  from Coal-Fired
      Utility Boiler Plants:   Identification  and Analysis of Organic  Pollutants
      in Air.  L.H.  Keith  (ed).  Butterwork Publishers,  1984,  pp.  443-458.

  95.  Hall, J. et  al (GCA Corp.).   Evaluation of PCB  Destruction Efficiency in
      an Industrial  Boiler.  .(Prepared for  IERL,  U.S.  Environmental Protection
      Agency)  April  1981.

  96.  Hall, J. M., et al (GCA Corp.).   Evaluation of  the PCB  Thermal  Disposal
  *   Process  in Unit 3 Boiler at  Northeast Utilities' Middletown  Station.
      Draft Final  Report, (Prepared for Northeast Utilities,  Hartford,
      Connecticut).   GCA-TR-81-82-G.

  97.  Harless, R.  L.  and R. G.  Lewis.   Quantitative Determination  of
      2,3,7,8-TCDD Residues by GC/MS.   U.S. Environmental Protection  Agency,
      Health  Effects  Research  Laboratory, Research Triangle Park,  NC, 1982.

  98.  Harless, R.  L.   High Resolution  Mass  Spectrometry  Methods  of Analysis  for
      Chlorinated Dibenzo-p-dioxins and  Dibenzofurans.   Human and  Environmental
      Risks of Chlorinated Dioxins and Related  Compounds. Plenum Publishing,
      1983.

  99.  Health Hazard Evaluation Report.   HETA 82-224-1336.   NIOSH Investigators:
      Stanley Salisbury  I.H. Miami Fire  Dept.   Miami, Fla., July 1983.

100. Higgins, G. H.  (Systech  Corporation) An Evaluation of Trace  Organic
      Emissions from Refuse Thermal Processing  Facilities.  (Prepared for U.S
      Environmental Protection Agency, Washington, D.C.) Contract  No.
     68-01-6071.  July 1982.

101. Howes, J. E., Jr. et al.  Characterization of Stack Emissions from
     Municipal Refuse-to-Energy Systems.  EPA  Contract No.  68-02-3458.

102. Howes, J. E., et al.   Determination of Dioxin Levels  in Carbon
     Reactivation Process Effluent Streams.  U.S. Environmental Protection
     Agency,  Contract No.  68-02-3487.

103. Hryhorczuk, D.  0., et al.  A Wire Reclamation Incinerator  as a Source of
     Environmental Contamination with Tetrachlorodibenzo-p-dioxins and
     Tetrachlorodibenzofurans.  Archives of Environmental Health.  Volume 36,
     1981.  pp.  228-234.
                                      9-9

-------
 104.  Hulman,  P.B.,  et  al-   (Radian  Corp.)   Screening  Study on Feasibility of
      Standards  of Performance for Wood  Charcoal  Manufacturing,  (Prepared for
      U.S.  Environmental  Protection  Agency,  Emissions  Standards  and Engineering
      Division,  Office  of Air Quality Planning  and  Standards), August 1978.

 105.  Hunter,  S.  C.  et  a]..   Application  of Combustion  Modifications to
      Industrial  Combustion  Equipment.   EPA-600/7-79-015a,  U.S.  Environmental
      Protection Agency.  Washington,  D.C. January 1979.   550 pp.

 106.  Hutzinger,  0.  (Editor)  Chlorinated Dioxin and Related Compounds Impact  on
      the  Environment.   Pergamon  Series  on Environmental  Science,  Vol.  5,  1982.

 107.  Hutzinger,  0.   Urban Incinerators: Causes of  Dioxins  in the  Environment.
      Presented  at the  Congress of La Risposta  Technologica Agli  Inquinamenti
      Chimici, Milan, Italy,  September 20-22, 1984.  10 pp.

 108.  (IT  Corporation.)   Dioxin Trial  Burn Data Package EPA Mobile Incineration
      System at  the  James Denney  Farm Site McDowell, Missouri.  EPA Contract
      68-03-3069.  June 21,  1985.

 108a.Jacobs,  L.,  M.  Zabik,  and J. Philips.  Concentrations  of Selected
      Hazardous  Chemicals in  Michigan  Sewage Sludges and  Their Impact on  Land
      Application.   Final Report.  Michigan State University,  East Lansing, MI.
      1981.

 109.  Janssens, J. et al_.  Qualitative and Quantitative Analysis of Emissions
      of a Municipal  Incineration  Installation.   Presented  at CEC
      Physico-chemical  Behaviour of Atmospheric Pollutants,  2nd Symposium,
      Italy, September  1981.

 110. Jansson, B.  and 6.  Sundstrom.   Formation  of Polychlorinated  Dibenzofurans
      (PCDFs) During a  Fire Accident  in Capacitors Containing Polychlorinated
     Biphenyls  (PCB).  From  Pergamon Series on  Environmental  Science  Volume 5
     - Chlorinated Dioxins and Related Compounds Impact  on  the Environment.
     1982.  pp. 201-208.

111. Jansson, B., e£ al_.  Formation of Polychlorinated Dibenzo-p-dioxins
     During Combustion of Chlorophenol Formulations.  The Science  of the Total
     Environment.  Volume 10, 1978,  pp. 209-217.

112. Jellum, E., A.K. Thorsrud, and F. W. Karasek.  Two  Dimensional
     Electrophoresis for Determining Toxicity  of Environmental Substances.
     Analytical Chemistry,  1983,  Volume 55,  pp. 2340-2344.

113. (JRB Associates).   Solid Waste Data:  A Compilation of  Statistics on
     Solid Waste Management Within the United States.   EPA Contract
     No.  68-01-6000, August  1981.

114. Josephson, J. Chlorinated Dioxins and Furans in the Environment.
     Environment Science Technology 17(3):  124A-129A.   1983.
                                      9-10

-------
 115. Junk, J. A. and J. J. Richard.  Dioxins Not in Effluents from Coal/Refuse
      Combustion.  Chemosphere, Volume 10, 1981.  pp. 1237-1241.

 116. Karasek, F. W., et al.  Distribution of PCDDs and Other Toxic Compounds
      Generated on Fly Ash Particulates in Municipal Incinerators.  Journal of
      Chromatography.  Volume 239, 1982,  pp.  173-180.

 117. Karasek, F. W.  Dioxins from Garbage:  Previously Unknown Source of Toxic
      Compounds is Being Uncovered Using  Advanced Analytical  Instrumentation
      Canadian Research. September 1980.   pp. 50-56.

 118. Karasek, F. W. et al.  Gas Chromatographic-Mass Spectrometric Study on
      the Formation of Polychlorinated Dibenzo-p-Dioxins and  Polychlorobenzenes
      fvnm Polyvinyl Chloride in a Municipal  Incinerator.   J.  of Chromatography
              -      1983.
 121
122,
 119.  Kayser,  R.,  et al.   2,3,7,8-TCDD - Intermedia Priority Pollutant Guidance
      Documents.   Prepared by the Office of Toxics Integration and the Office

      Jul   1982ideS  a0d Tox1c Substances-  u-s-  Environmental  Protection Agency.


 120.  Kearney,  D.  C.,  et  al.   Tetrachlorodibenzo-p-dioxin  in  the Environment-
      sources,  Fate,  and  Decontamination.   Environmental Health Perspectives
      Volume 5,  1973.   pp.  273-277.
     i E'/F 4l-  (Mitre Corporation)  Hazardous Waste Incineration:  A
     le of Existing Facilities, EPA Contract No. 68-03-3021, July 1982.
Kemp, Clinton C.,  Needed: Plain Talk About Dioxin.
v.15, No. 5 pp. 142-152.
                                                          Waste Age.  May  1985
123. Kim, N.  Building Reclamation After Dioxin Contamination by PCB Fires
     From:  Accidental Exposure to Dioxins:  Human Health Aspects.
     K Coulston and F. Pocchiari (eds).  Academic Press, 1983, pp. 185-186.

124. Kimble, B. J   Comments on "Chlorinated Dibenzo-p-dioxins and Related
     Compounds in Incinerator Effluents."  Chemosphere.  Volume 10, 1981
     pp. 243-244.


125. Kimble, B  J  and M  L. Gross.   Tetrachlorodibenzo-p-dioxin Quantisation
     in Stack-Collected Coal Fly Ash.  Science.  Volume 207.  January 1980.
     pp. 59-61.


126. Kimbrough, R.  D. ed   Halogenated Biphenyls,  Terphenyls, Naphthalenes,
     Dibenzodioxins, and Related Products.   Elsevier/North Holland Biomedical
     Press,  Amsterdam.   1980.  406 pp.

127. Kooke,  R., et  al.   Extraction Efficiencies of Polychlorinated
     Dibenzo-p-dipxins  and Polychlorinated  Dibenzofurans from Fly Ash.
     Analytical  Chemistry.  Volume 53, 1981.   pp.  461-463
                                      9-11

-------
128. Kriebel,  D.  Dioxins:  Toxic  and  Still  Troublesome.   Environment.
     Volume  23, Number  1.  January/February 1981.   pp.  6-13.

129. Krishnan  E.  R.,  and 6.  V. Hellwig.  Trace  Emissions from Coal  and Oil
     Combustion.   Environmental  Progress  (Vol.  1,  No.  4)  November  1982.
     pp. 290-296.

130. Kulkarni, S.   (TRW Inc.)  Pollution Aspects of the  Thermal  Regeneration
     of Spent  Activated Carbon.   U.  S.  Environmental  Protection  Agency,
     Research  Triangle  Park, N.C., February 1984.  39 pp.

131. Lahaniatis,  E. S., et al_.   On the  Occurrence  of Chlorinated Hydrocarbons
     in Fly  Ash from  Waste Combustion Facilities.   Chemosphere.   Volume 6,
     1977.   pp. 11-16.

132. Lamberton, J., et  a]_.   The  Determination of Polychlorodibenzo-p-dioxins
     in Pentachlorophenol  Wood Treatment Solutions.   Journal  of  the American
     Industrial Hygiene Association.  Volume 40, 1979.   pp. 816-822.

133. Lamparski, L.  L. et al_.  Photolysis of Pentachlorophenol-Treated  Wood.
     Chlorinated  Dibenzo-p-dioxin Formation.  Environmental Science and
     Technology.   Volume 14, No.  2.  February 1980.   pp. 196-200.

134. Lamparski, L.  L. and  T. J.  Nestrick.   Determination of Tetra-, Hexa-,
     Hepta-, and  Octachlorodibenzo-p-dioxin Isomers  and  Particulate Samples at
     Parts Per Trillion Levels.   Analytical  Chemistry.   Volume 52,  1980.
     pp. 2045-2059.

135. Lao, R. C. et 3]..  Investigation of PAH and Polychlorinated Organic
     Pollutant Emissions from Wood Combustion Sources.   From:  Polynuclear
     Aromatic  Hydrocarbons:  Formation, Metabolism  and Measurement,  7th
     International Symposium.  Columbus, Ohio,  Oct.  26-28, 1982

136. Lee, K. W. et a]_  (Radian Corporation)  J.  M.  Huber  PCB Destruction
     Process Trial Burn Report.   (Prepared  for  J.  M. Huber Corp.  Borger,
     Texas,) October  1983.

137. Leinster, C.  The Sweet Smell of Profits from Trash.  Fortune, Vol. Ill
     No. 7, pp. 150-154, April 1, 1985.

138. Levin, J. 0. and C. A. Nilsson.  Chromatographic Determination of
     Polychlorinated Phenols, Phenoxyphenols, Dibenzofurans, and
     Dibenzodioxins in Wood Dust  from Worker Environments.  Chemosphere.
     Volume 6, 1977.  pp.  443-498.

139. Levin, J. 0., et al_.   Use of Chlorophenols and  Fungicides in Sawmills.
     Scandinavian Journal  of Work Environment and  Health.  Volume 2, 1976.
     pp.  71-81.
                                      9-12

-------
 140. Liberti ,  A., and D. Brocco.  Formation of Polychlorodibenzo-Dioxins and
      Polychlorodibenzofurans in Urban Incinerators Emissions. In:
      Chlorinated Dioxins & Related Compounds Impact on the Environment.
      0.  Hutzinger, R. W. Frei,  E. Merian,  F. Pocchiari (eds.) Pergamon
      Press 1982, pp. 245-251.

 141. Liberti,  A., 6. Goretti, and M. V.  Russo.   Formation of Polychlorodi-
      benzodioxins and Polychlorodibenzofurans in Urban Wastes Incineration
      Gazzetta  Chimica Italiana, 113 No.  9-10, pp. 641-644, 1983.

 142. Liberti,  A., et al.  Determination  of Polychlorinated Dibenzo-p-dioxins
      and Dibenzofurans in the Emissions  from Urban Incinerators.   Mikrochimica
      Acta.   Volume I, 1981.   pp. 271-280.

 143. Liberti,  A., et al.  Evaluation of  Organochlorine Compounds  in the
      Emissions of Urban Incinerators.  From the Pergamon Series on
      Environmental  Science Volume 3 -  Analytical  Techniques in Environmental
      Chemistry.  .1980.  pp.  157-166.

 144. Liberti,  A.  and D.  Brocco.  Formation of Polychlorodibenzo-dioxins and
      Polychlorodibenzofurans in Urban  Incinerators Emissions:  Chlorinated
      Dioxins and Related Compounds Impact on the Environment.  0.  Hutzinger,
      R.W.  Frei,  E.  Merian,  F. Pocchiari  (eds).   Pergamon Press, 1982,
      pp.
150.
151
 145.  Liberti, A.,  et  al..   PCDD  and  PCDF  Formation  in  the  Combustion  of
      Vegetable Wastes.  Chemosphere,  Volume  12,  No. 4/5,  1983,  pp. 661-663.

 146.  Lindahl, R. et al  Formation of  Polychlorinated  Dibenzofurans (PCDFs)  and
      Polychlorinated  Dibenzo-p-dioxins (PCDDs) from the Pyrolysis of
      Polychlonnated  Diphenyl Esters.  Chemosphere, Vol.  9, pp.  351-361,  1980.

 147.  Liptak, B.G.  Environmental Engineers Handbook.  Volume 3,  Chilton  Book
      Company, Ontario, Canada,  1974,  p.  283.

 148.  Lustenhouwer, J., et  al.   Chlorinated Dibenzo-p-dioxins and Related
      Compounds in  Incinerator Effluents:  A Review of Measurements and
      Mechanisms of Formation.   Chemosphere.  Volume 9, 1980.  pp. 501-522.

 149.  Mahle, N. and L. Whiting (Dow Chemical).  The Formation of
      Chlorodibenzo-p-dioxins by Air Oxidation and Chlorination of Bituminous
      Coal.  Chemosphere.  Volume 9,  1980.  pp. 693-699.
Main, J.  Dow vs. The Dioxin Monster.  Fortune, Vol. 107 (11), pp. 83-89,
May 30, 1983.

Makasani, J.  Don't Let the Dioxin Issue Axe New Refuse-Fired Power
Plants.  Power, August 1984, pp. 87-89.
152. Mamantov, A.  Linear Correlation Between Photolysis and Toxicity of
     Polychlorinated Dibenzo-p-dioxins.  Environmental Science Technology,
     1984,  18, pp. 808-810.
                                      9-13

-------
153. Mantel!, C.L.  Carbon and Graphite Handbook.  Interscience  Publishers,
     New York, 1968.

153a. Memo from Amendola, 6. A., Project Manager,  Michigan Dioxin Studies,
  *   Region V to D. 6. Barnes, Chairman Coordinator, Dioxins Work Group.
      Michigan Dioxin Studies—Ambient Air/Dow Chemical Emissions Data.
      September 5, 1985.
154.
155.
156.
  *
157.
158.
  *
159.




160.



161.



162.


163.



164.
Memo from Cook, M.B., Dioxin Management Coordinator, U.S.  Environmental
Protection Agency, to Schatzow, S. and List of Addressees.
February 13, 1984.

Memo from Poppiti, J., Manager -Delisting Program, U.S.Environmental
Protection Agency, to Chlorinated Dioxins Workgroup and List of
Addressees.  May 23, 1985.  25 pp. RCRA Dioxin Delisting Petition
for the Mobile Incinerator System.

Miller, S. E. et a].. .Determination of Dioxin and Furan Levels in
Effluent Streams of the Cincinnati Waterworks Carbon Reactivation
System.  EPA Contract No. 68-01-6702.  July 1985.

Mix, T. W. and B. L. Murphy.  Risks Associated with Waste-Fuel Use  in
Cement Kilns.  Environmental Progress Vol. 3, No. 1, 1984.  pp. 64-70.

Moore, J. A. Assistant Administrator for Pesticides and Toxic
Substances.  (Memorandum)  Advance Notice of Proposed Rulemaking on
PCB Electrical Transformer Fires.

Morselli, L., et al_.  Analysis of the Effluents of an Urban-Solid Refuse
Incinerator:  Study of Methods of Extraction and Analysis  for the
Quantitative Determination of Polychlorodibenzo-p-dioxins.  Annals of
the Italian Chemical Society - 1981.  Volume 71, Nos. 9-10.  pp. 557-572.

(MRI).  Assessment of Emission of Specific Compounds from  a Resource
Recovery Municipal Refuse Incinerator.  EPA Contract No. 68-01-5915.
May 22, 1985.  EPA 560/5-84-002.

(MRI).  Performance Evaluation of Full Scale Hazardous Waste
Incinerators.  Draft Final Report. (Prepared for U.S. Environmental
Protection Agency, Incineration Research Division) January 1984.

(MRI Report).  PDCF Formation from PCB's Under Fire Conditions EPA
Contract No.68-02-3938.  Sept. 20, 1984.

National Research Council of Canada.  Polychlorinated Dibenzo-p-Dioxins:
Criteria for Their Effects on Man and His Environment.  NRCC No. 18574.
1981. 251 pp.

National Research Council of Canada.  Polychlorinate Dibenzofurans:
Criteria for Their Effects on Humans and the Environment.  NRCC No.
22846.  1984. 243 pp.
                                      9-14

-------
 165. Nestrick, T. J. and L. L. Lamparski.   Isomer-Specific Determination of
      Chlorinated Dioxins for Assessment  of  Formation and Potential
      9!oJrSS!£nt?lo?Blss1on from Wood Comblistion.  Analytical Chemistry. 54:
      &&9Z-Z299. 1982.

 166. Nestrick, T. J., et al.  (Dow Chemical)  Comments on Variations in
      Concentrations of Organic Compounds Including Polychlorinated
      Dibenzo-p-dioxins and Polynuclear Aromatic Hydrocarbons in Fly Ash from a
      Muni£"Pal Incinerator.  Analytical Chemistry.  Volume 54, 1982.
      pp. 823-824.

 167. Nestrick, T. J., et al.  (Dow Chemical)  Methodology and Preliminary
      Results for the Isomer-Specific Determination of TCDDs and Higher
      Chlorinated Dibenzo-p-dioxins in Chimney Particulates from Wood-Fueled
             :/urnaces Located in Eastern Central, and Western Regions of the
             states.   Presented at the International Symposium on Chlorinated
              and Related Compounds,  Arlington, VA, October 25-29,  1981.
 168'  ?nrlnol!hnR-   gioxin .Analysis of Philadelphia Northwest and East Central
      Incinerators:  Preliminary Summary of Results.  Midwest Research
      Institute.  EPA Contract 68-02-3891  July 1,  1985.
 169'  rnmJc? Stat?X DePar*ment of Environmental  Conservation Municipal  Refuse
      Combustion  -  Recommendations  for Standards  and Guidelines.  February 1985.
      cc.  pp.
                                                                           The
172* Ko F?Unty+ ReTling and Energy Recovery Center.  Risk Assessment for
     Trace Element and Organic Emissions. April 1984.

173. (Northrop Services, Inc.)  How to Operate the Hazardous Waste
     Incinerator.  In: Course Notes, Hazardous Waste Incinerator Session, Air
     Toxics Course 86.  November 10, 1983. 54 pp.                *«»iun, «ir

174  Wunn  A  R   r  1  4. •    c
   • v:  "' "• °'  ^valuation of HC1 and Chlorinated Organic Compound Emission-;
  *  from Refuse Fired Waste-to-Energy Systems. (Draft) Prepared"!? US
  -   Environmental Protection Agency, ORD, Research Triangle Park, NC, 1983.

175. O'Connell, W. L. Distribution of Data on the Emission of Dioxins from
     Incinerators.  Presented at Southwestern Ohio Chapter APCA Specialty
     Meeting on Air Toxics, May 31, 1984.                       oHBv.,«uLy

176.  O'Hara P.A., and R.  C. Ahlert.  Mechanisms of Formation of
     Polychlorinated Dibenzo-p-dioxins  in Municipal  Incinerators.
                                      9-15

-------
177. Olie, K., et «*]_.  Chlorodibenzo-p-dioxins and Chlorodibenzo-furans are
     Trace Components of Fly Ash and Flue Gas of Some Municipal Incinerators
     in the Netherlands.  Chemosphere.  Volume 6, 1977.  pp. 455-459.

178. Olie, K., et al_.  Polychlorinated Dibenzo-p-dioxins and Related Compounds
     in Incinerator Effluents.  From Pergamon Series on Environmental Science
     Volume 5 - Chlorinated Dioxins and Related Compounds Impact on the
     Environment.  1982.  pp. 227-244.

179. Olie, K., et al_.  Formation and Fate of PCDD and PCDF from Combustion
     Processes.  Chemosphere, Volume 12, No. 4/5, 1983, pp. 627-636.

180. Olie, K. and 0. Hutzinger.  Detection of Polychlorodibenzodioxins in Fly
     Ash and Flue Gas of Municipal Incinerators.  Recent Developments in Mass
     Spectrometry in Biochemistry and Medicine.  (Proceedings of the 4th
     International Symposium.) Alberto Frigerio, ed. Volume 1, 1978. pp.
     555-562.

181. Ontario Ministry of the Environment, Air Resources Branch.  Report on
     Combustion Testing Program at the Swaru Plant, Hamilton-Wentworth.
     (Prepared by Envirocon Limited) January 1984.

182. Ozvacic, V., Wong CK, Tosine H., Clement R. E., and J. Osborne
     Emissions of Chlorinated Organics from Two Municipal Incinerators in
     Ontario.  For Presentation at the 77th Annual Meeting of the Air
     Pollution Control Association.  San Francisco, Calif.  June 24-29, 1984.

183. Peters, J. A., T. W. Hughes and R. E. Mournighan.  Evaluation of
     Hazardous Waste Incineration in a Cement Kiln at San Juan Cement.  In:
     Incineration & Treatment of Hazardous Waste - Proceedings of the Ninth
     Annual Research Symposium.  EPA-600/9-84-015, U.S.  Environmental
     Protection Agency, Cincinnati, Ohio  July 1984.  p. 210-24.

     Peters, J. A. (Monsanto Research Corp.).  Evaluation of Hazardous Waste
     Incineration in Cement Kilns at San Juan Cement Company. (Prepared for
     Incineration Research Branch, Industrial Research Laboratory, U.S.
     Environmental Protection Agency, Cincinnati, Ohio 45268) Contract No.
     68-03-3025, January 1983.

185. Raloff, J.  Dioxin: Is Everyone Contaminated?  Science News  Vol. 128
     July 13, 1985.  pp. 26-29.

186. Rao, K. R. (editor).  Pentachlorophenol: Chemistry. Pharmacology, and
     Environmental Toxicology. Plenum Press, New York, 1978.

187. Rappe, C.  Analysis of Polychlorinated Dioxins and Furans.  Environmental
     Science Technology.  Vol. 18, No. 3, 1984  pp. 78-90.
184.
  *
                                      9-16

-------
 188.  Rappe,  C.   Source & Identification,  Especially With Respect to PCDDs and
      PCDFs Found in Emissions of Incinerators.

 189.  Rappe,  C.,  et si.  Dioxins,  Dibenzofurans,  and Other Polyhalogenated
      Aromatics:   Production,  Use,  Formation,  and Destruction.   Annals of the
      New York Academy of Sciences.   Volume 320.   pp.  1-18,  1979.

 190.  Rappe,  C.,  et a]..  Formation  of Polychlorinated  Dibenzo-p-dioxins (PCDDs)
      and Dibenzofurans (PCDF) by  Burning  or Heating Chlorophenates.
      Chemosphere.   Volume 7,  1978.   pp. 269-281.

 191.  Rappe,  C.,  et al.  Occupational  Exposure to Polychlorinated  Dioxins and
      Dibenzofurans.   From Pergamon  Series on  Environmental  Science  Volume 5  -
      Chlorinated Dioxins and  Related  Compounds  Impact on the Environment.
      1982.   pp.  495-513.

 192.  Rappe,  C.,  et al..   Polychlorinated Dioxins,  Dibenzofurans, and  other
      Polynuclear Aromatics Formed During  PCB  Fires.   Chemica Scripta,
      Volume  20,  1982,  pp.  56-61.                               .

 193.  Rappe,  C. and S.  Marklund.  Thermal  Degradation  of  Pesticides and
      Xenobiotics:  Formation of Polychlorinated Dioxins and  Dibenzofurans.
      Pesticide Chemistry: Human Welfare and Environment,  Proc. Int.  Congress
      Pesticide Chem.,  5th, Volume 3,  1983, pp. 317-322.

 194.  Rappe,  C.,   S. Marklund,  P. A. Bergquist, and M.  Hansson.  Polychlorinated
      Dibenzo-p-dioxins, Dibenzofurans, and Other Polynuclear Aromatics Formed
      During  Incineration and  Polychlorinated  Biphenyl Fires.  In: Chlorinated
      Di oxi n  and  Dibenzofurans in the Total Environment.  Chapter 7.
      Choudhary,   6., Keith, L.H. Rappe, C.   (eds)  Buttersworth Publishers  1983.

195. Rawls, R  L.  Dow Finds Support, Doubt for Dioxin Ideas.  Chemical and
     Engineering News.  February 12, 1979.  pp.  23-29.


196'     °r?' D
                     C' L' Haile and R' M' Lucas-  Emissions of PCDDs and
                           Sources. Presented at The International Symposium
                    ?^?ins and Re1ated ^pounds.  Arlington, Virginia,
                    1981.

197. Rghei, H. 0. and 6. A. Eiceman.  Adsorption and Thermal Reactions of
     1,2,3,4-Tetrachlorodibenzo-p-dioxin on Fly Ash from a Municipal
     Incinerator.  Chemosphere.  Volume 11, No. 6, 1982, pp. 569-576.

198. Rghei, H. 0. and G. A. Eiceman.  Adsorption and Chlorination of

     ?nr?nf^f;   Xi?K3nd lk chlo™libenxo-P-Dioxin on Fly Ash from Municipal
     Incinerators.  Chemosphere, Vol.  13,  No. 3, pp 421-426, 1984.
199'  ?hini?n!S0; ?"? 5: A- Eiceman-  Effect of Matrix on Heterogenous Phase
     Chlorine Substitution Reactions for Dibenzo-p-dioxin and HCL in Air
     Chemosphere Vol. 14, No.  2,  pp 167-171, 1985.
                                      9-17

-------
200. Sachdev, A.  K.  and  I. J.  Marvan.   Thermal  Destruction  of Chlorophenol
     Residues.   Prepared for  Environmental  Protection  Service,  Canada.   July
     1983.   107  pp.

201. Samuelson,  U. and A.  Lindskog.    Chlorinated  Compounds in  Emissions from
     Municipal Incineration.   Chemosphere,  Vol.  12,  No.  4/5,  pp 665-668,
     1983.

202. Sawyer, T.,  et  il_.   Bioanalysis of Polychlorinated  Dibenzofuran  and
     Dibenzo-p-dioxin Mixtures in  Fly  Ash.   Paper  Presented at  the  3rd
     International Symposium  on Chlorinated Dioxins  and  Related Compounds.
     Salzburg, Austria,  October 12-14,  1982.

203. Sawyer, T.,  et  a]..   Bioanalysis of Polychlorinated  Dibenzofuran  and
     Dibenzo-p-dioxin Mixtures in  Fly  Ash.   Chemosphere  12: 529-535,  1983.

204. Scharsach,  A.   Incinerator Emission Control Technology - Present and
     Future  Processes.   Ontario Ministry of Energy - Energo '83:  Energy
     Opportunities in Biomass  Symposium,  Toronto,  Mar. 6-9, 1983.

205. (Scott  Environmental  Services.)   Sampling  and Analysis of Chlorinated
     Organic Compound Emissions from the Hampton Waste-to-Energy System.
     March 1985.

206. Shaub. W. M.  Containment of  Dioxin Emissions from  Refuse  Fired  Thermal
     Processing  Units -  Prospects  and  Technical  Issues.  NBSIR 84-2872,  U.S.
     National Bureau of  Standards, Washington,  D.C.  April  1984.  150 pp.

207. Shaub, W. M.  Technical  Issues Concerned with PCDD  and PCDF Formation  and
     Destruction  in  MSW  Fired  Incinerators.  NBSIR 84-2975,  U.S.  National
     Bureau of Standards,  Washington,  D.C.,  November 1984.   41  pp.

208. Shaub, W. M. and Wing Isang.  Dioxin Formation  in Incinerators.
     Environmental Science and Technology,  Vol. 17,  p. 721,  December  1980.

209. Shen, T. T.  Air Quality  Impact of Trace Contaminants  from Coal
     Conversion.  From:  Proceedings of the  Governors Conference on  Expanding
     the Use of Coal in  New York State:   Problems  &  Issues,  May 21-22,  1981
     Albany, N.Y.

210. Shih, C., ei a],.  (TRW, Inc.) Emissions of Polychlorinated
     Dibenzo-p-dioxins (PCDDs)  and Dibenzofurans (PCDFs) from the Combustion
     of Fossil Fuels, Wood, and Coal Refuse.  (Prepared  for U.S.  Environmental
     Protection Agency,  Research Triangle Park, NC)  EPA  Contract
     No. 68-02-3138.  December  1980.

211. Shih, C., et al_.  (TRW, Inc.)  POM  Emissions  from Stationary Conventional
     Combustion Processes, with Emphasis  on Polychlorinated  Compounds of
     Dibenzo-p-dioxin (PCDDs),  Biphenyl  (PCBs), and  Dibenzofuran  (PCDFs).
     (Prepared for U.S.  Environmental  Protection Agency, Research Triangle
     Park, NC) EPA Contract No. 68-02-3138.  September 1979.
                                      9-18

-------
212. Short, H. Dioxin Cleanup Methods Get Worldwide Attention.
     Engineering, pp. 22-25, Nov. 26, 1984.
Chemical
213. Sigsby, J.E.  Dioxins in Mobile Sources Particles.  U.S. Environmental
  *  Protection Agency, Mobile Source Measurements Research Section.
     February 23, 1982.

214. Smith, R. J.  Dioxins Have Been Present Since the Advent of Fire, Says
     Dow.  Science.  Volume 202.  December 1978.  pp. 1166-1167.

215. Smith, R. M., et a]..  Analysis for 2,3,7,8-TCDD and 2,3,7,8-TCDF in a
     Soot Sample from a Transformer Explosion in Binghamton, NY.  Chemosphere,
     Volume 2, No. 8, 1982,  pp. 715-720.

216. Stanley, J. S. et alI.  Quality Assurance Program for Combustion Source
     Studies.  In: Identification and Analysis of Organic Pollutants in Air,
     L. H. Keith (ed.)  Butterworth Publishers 1984. pp. 459-473.

217. State of California Air Resources Board.  Air Pollution Control at
     Resource Recovery Facilities - Final Report.  May 24, 1984. 393 pp.

218. State of Washington.  Letter from C. Jonathan Neel, District Supervisor
  *  to Ms. Karen Karl re:  PCB burning ash residue.  Nov. 28, 1984.
219. Stehl, R. H., and L. L. Lamparski.  Combustion of Several
     2,4,5-Trichlorophenoxy Compounds:  Formation of 2,3,7,8-TCDD.
     Volume 197.  September 1977.  pp. 1008-1009.
    Science.
220. Stretz, L. A., et al.  Controlled Air Incineration of PCP-Treated Wood.
     (Prepared for U.S. Environmental Protection Agency, Industrial
     Environmental Research Laboratory) 1982.

221. (Swiss Federal Office of Environmental Protection)  Environmental
     Pollution Due to Dioxins and Furans from Communal Rubbish Incineration
     Plants.  Schriftenreihe Umweltschutz.  No. 5., 1982.  20p.

222. Taylor, M. L. et al.  Assessments of Incineration Processes as Sources of
     Supertoxic Chlorinated Hydrocarbons: Concentration of Polychlorinated
     Dibenzo-p-dioxins/Dibenzofurans and Possible Precursor Compounds in
     Incinerator Effluents.  Chapter 8.  In:  Chlorinated Dioxins and Dibenzo-
     furans in the Total Environment.  G.  Choudhary, L. H. Keith, and
     C. Rappe  eds.  Butterworth Publishers, Ann Arbor Science Book, 1983.

223. (TDI, Inc.)  The Dibenzofurans.  (Prepared for U.S. Environmental
     Protection Agency, Washington, D.C.) EPA Contract No. 68-03-2919. 1981.

224. Teller, A. J. and J. D. Lauber.  Control of Dioxin Emissions from
     Incineration.  Paper presented at the 76th Annual Meeting and Exhibition
     of the Air Pollution Control Association, Atlanta, Georgia.  June 19-24,
     1983.
                                      9-19

-------
 225.  Test  Report for Toxic Screening at the Philadelphia Northwest Municipal
      Incinerator.  Prepared by Engineering Science EPA Contract No. 68-02-3960,
      October 1980.

 226.  Tiernan,  M.  L.  et  al_.  Chlorodibenzodioxins, Chlorodibenzofurans,  and
      Related Compounds  in  the Effluent  from Combustion Processes.
      Chemosphere,  Volume 12,  No.  4/5,  1983,  pp.  565-606.

 227.  Tiernan,  T.  0.,  Chlorodibenzodioxins and Chlorodibenzofurans:  An
      Overview.   In:  Detoxification  of Hazardous  Wastes,  Ann  Arbor  Science.
      Publishers,  1982.   pp.  243-260.

 228.  Tiernan,  T.  0.,  et a]_.   Characterization of Toxic Components  in the
      Effluent  from a  Refuse-Fired Incinerator.   Resources and  Conservation.
      Volume  9,  1982,  pp. 343-354.

 229.  Tong, H.  Y.  et  al.   Identification of Orange Compounds  Obtained from
      Incineration of  Municipal Waste by High-Performance Liquid
      Chromatographic  Fractionation  and  Gas Chromatogrpahy-Mass Spectrometry.
      Journal of Chromatography 285  (1984)  423-441.

 230.  Townsend,  D. E.  (Dow  Chemical).  The  Use of Dioxin  Isomer Group Ratios to
      Identify  Sources and  Define  Background  Levels of Dioxins  in the
      Environment.  Journal  of Environmental  Science  and  Health.  Volume B15,
      1980.   pp.  571-609.

 231.  Townsend,  D. I.  Change  of Isomer  Ratio and Fate of Polychlorinated-p-
      dioxins in the Environment.  Chemosphere, Vol.  12,  No.  9/5, pp.  637-643,
      1983.

 232.  Tremblay,  J. W.  The  Design, Implementation,  and Evaluation of  the
      Industrial Hygiene  Program Used During  the  Disposal  of  Herbicide Orange.
      In: Human  and Environmental  Risks  of  Chlorinated Dioxins  and  Related
      Compounds, 1983. R. E. Tucker, A.  L.  Young,  and  A.  P. Gray (eds).

 233.  U. S. Army Environmental Hygiene Agency.  Incineration  of Pentachloro-
      phenol-Treated Wood,  Tooele Army Depot, Tooele,  Utah. 19-30 October 1981,
      and 14-30 June 1982.

 234.  U. S. Environmental Protection Agency.  PCB Disposal  by Thermal
      Destruction.  NTIS  -  PB82-241860,  U.S.  Environmental  Protection  Agency,
      Dallas, TX, June 1981.

235. U. S. Environmental Protection Agency.  Unpublished  EPA Sewage  Sludge
     Task Force Report.

236.  (U. S. Conference of Mayors)  Resource  Recovery Activities.   Waste Age,
     pg. 56, November 1983.
                                      9-20

-------
 237.  U.  S. DOE, Report on the 1980 Manufacturing Industries Energy Consumption
      Study and Survey of Large Combustors, DOE/EIA-0358, January 1983.

 238.  U.  S. Environmental Protection Agency.  Application of Combustion
      Modifications to Industrial  Combustion Equipment, EPA-600/7-79-015a
      January 1979.

 239.  U.  S. Environmental Protection Agency.  Kraft Pulping: Control  of TRS
      Emissions from Existing Mills.  EPA-450/2-78-003b,  March 1979.

 240.  Velzy,  C.O.   ASME's'Role in  Developing Standards for Measurement of
      Dioxin  Emissions.

 241.  Versar,  Inc.   Memorandum to  Dave Alexander,  Jay Wind from Shiv  Krishnan.
      Method  for Characterization  of Toxic Emissions from the Burning of Waste
      Oil.  February 13,  1984.

 242.  Viau, A.C., S.  M. Studak,  and F.  W.  Karasek.   Comparative Analysis of
      Hazardous Compounds on  Fly-Ash from  Municipal  Waste Incineration by Gas
      Chromatography/ Mass  Spectrometry.   Canadian  Journal  of Chemistry,
      Nov.  1984, 62 (11).

 243.  Vick, R.  D.,  et al.   Organic Emissions from Combustion of Combination
      Coal  Refuse to  Produce  Electricity.   Chemosphere.   Volume 7,  1978
      pp. 893-902.

 244.  Villaneuva, E.  C.,  et al.  Chlorodibenzo-p-dioxin Contamination  of Two
      Commercially  Available  Pentachlorophenols.  Journal  of Agricultural  and
      Food  Chemistry.  Volume 21,  No.  4. July/August 1973.   pp.  739-740.

 245.  Wang, D.  K. W.  et al.  Sampling  and  Analytical  Methodologies  for PCDDs
      and PCDFs  in  Incinerators  and  Wood Burning Facilities.   In  Environmental
      Science and Research, 1983,  Vol.  26  No.  Human and  Environmental  Risks
      of Chlorinated  Dioxins and Related Compounds   pp. 113-126.
246. Weston, R. F., Inc.  Results of the Silvex Test Burn DRE's
  *  September 9, 1983.
Draft Report.
247. Witness Statement from Ozvacic, V. M.,  Ministry of the Environment,
     concerning Magnitude of Dioxin and Furan Emissions from the Proposed
     Energy Waste Plant, Victoria Hospital Corporation. Westminster Campus,
     London Ontario.  January 11, 983.  10 pp.

248. Wong, T.S. Dioxin Formation and Destruction Combustion Processes
     For presentation at the 77th Annual Meeting of the Air Pollution Control
     Association, San Francisco, Calif.  June 24-29, 1984.

249. Yakowitz, H.  Background Information With Respect to Dioxin Issues
     Associated With Incineration of Municipal Solid Waste.  ENV/WMP/85 11
     Organization for Economic Co-Operation and Development, Paris, France.
     February 22, 1985.  34 pp.
                                      9-21

-------

-------
                        APPENDIX A

                   TIER 4 EMISSIONS DATA


A.I  Summary of PCDD/PCDF Emission Concentrations
     (As-Measured Basis)

A.2  Summary of PCDD/PCDF Emission Concentrations
     (Normalized to 3% Oxygen)

A.3  Summary of PCDD/PCDF Emissions Per Unit of Feed

A.4  Error Analysis:  Control Device Efficiency Calculations

-------
                APPENDIX A.I

Summary of PCDD/PCDF Emission Concentrations
             (As-Measured Basis)

-------
                    TABLE A.1.1.  SITE 1 (SSI-A)/OUTLET
                    DIOXIN/FURAN EMISSION CONCENTRATIONS
                            (As-measured values)
 Dioxin/Furan
     Isomer
     Isomer Concentration In Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg,
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
5.08E-03
1.70E+00
2.12E-02
ND( 4.06E-02)
3.79E-01
9.02E-01
3.01E+00

NR
4.91E+00
1.59E+00
ND( 9.44E-02)
6.21E-02
ND( 5.65E-02)
6.56E+00
9.12E-03
1.62E+00
4.19£-02
1.24E-01
3.03E-01
5.50E-01
2.65E+00

NR
4.46E+00
1.28E+00
ND( 7.30E-02)
7.82E-02
3.65E-02
5.86E+00
4.49E-03
1.38E+00
ND( 3.04E-02)
1.10E-01
4.13E-01
9.78E-01
2.88E+00

NR
4.99E+00
1.51E+00
7.29E-02
8.19E-02
ND( 4.11E-02)
6.66E+00
6.23E-03
1.56E+00
2.10E-02
7.80E-02
3.65E-01
8.10E-01
2.84E+00

NR
4.79E+00
1.46E+00
2.43E-02
7.41E-02
1.22E-02
6.36E+00
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.
NR  =  not reported by Troika.
ND  -  not detected (detection limit in parentheses).
ng  =  1.0E-09g
6000 operating hours per year
                                   A. 1-1

-------
 Dioxin/Furan
     Isomer
                      TABLE A.1.2.  SITE 2 (ISW-A)/OUTLET
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                              (As-measured values)
   •  Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 03          Run 04
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

4.12E-01
6.80E+00
1.03E+01
2.02E+01
2.41E+01
9.90E+00
7.18E+01

2.27E+00
6.95E+01
7.24E+01
1.15E+02
6.37E+01
1.09E+01
3.34E+02

1.47E+00
2.61E+01
3.26E+01
3.94E+01
5.03E+01
1.77E+01
1.68E+02

6.53E+00
1.46E+02
1.43E+02
1.32E+02
8.97E+01
1.83E+01
5.36E+02

1.01E+00
• 1.68E+01
2.29E+01
3.08E+01
6.33E+01
1.14E+01
1.46E+02

4.46E+00
1.36E+02
1.50E+02
1.41E+02
1.21E+02
1.24E+01
5.65E+02

9.67E-01
1.66E+01
2.20E+01
3.01E+01
4.59E+01
1.30E+01
1.29E+02

4.42E+00
1.17E+02
1.22E+02
1.30E+02
9.15E+01
1.39E+01
4.*78E+02
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =*  not detected (detection limit in parentheses).
ng  =  1.0E-09g
2200 operating hours per year
                                   A. 1-2

-------
 Dioxin/Furan
      Isomer
                      TABLE A.1.3.   SITE 3 (SSI-B)/OUTLET
                      DIOXIN/FURAN  EMISSION CONCENTRATIONS
                              (As-measured values)
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg,
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
*
ND( 1.15E-02)
1.92E-01
ND( 1.92E-02)
ND( 1.85E-01)
1.15E-01
3.08E-01
. 6.15E-01

7.31E-01
7.50E+00
2.08E+00
8.46E-01
ND( 2.38E-01)
3.85E-02
1.12E+01

ND
ND
ND
ND
ND


ND(
NDl
ND(
ND(


6.00E-02
6.00E-02
4.50E-02
1.02E-01
6.75E-02
1.75E-01
1.75E-01

2.50E-01
1.52E+00
4.50E-02
5.75E-02
7.25E-02
1.25E-02
1.77E+00

) ND( 1.15E-02)
) 2.87E-02
) ND( 5.75E-02)
) ND( 8.62E-03)
) ND( 4.02E-02)
1.72E-01
2.01E-01

4.02E-01
2.67E+00
7.18E-01
ND( 2.13E-01)
ND( 2.01E-02)
ND( 1.72E-02)
3.79E+00

.OOE+00
7.37E-02
.OOE+00
.OOE+00
3.85E-02
2.18E-01
3.31E-01

4.61E-01
3.90E+00
9.32E-01
2.82E-01
.OOE+00
1.28E-02
'5.59E+00
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                 A. 1-3

-------
                       TABLE A.1.4a.  SITE 4 (BLB-A)/INLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                                (As-measured values)
 Dioxln/Furan
     Isomer
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 1.94E-02)
ND( 1.94E-02)
ND( 7.52E-02)
1.21E-01
2.91E-01
8.01E-01
1.21E+00

ND( 1.46E-02)
2.43E-01
3.16E-01
4.37E-01
2.43E-01
7.28E-02
1.31E+00
ND( 1.98E-02)
1.98E-02
ND( 2.38E-02)
7.94E-02
1.98E-01
6.35E-01
9.33E-01

ND( 7.94E-02)
1.98E-01
2.38E-01
3.97E-01
1.19E-01
7.94E-02
• 1.03E+00
ND( 7.47E-02)
ND( 7.47E-02)
ND( 9.96E-02)
ND( 3.90E-01)
5.81E-01
2.03E+00
2.61E+00

ND( 1.66E-01)
4.15E-01
ND( 6.14E-01)
7.88E-01
3.73E-01
ND( 2.61E-01)
1.58E+00
.OOE+00
6.61E-03
.OOE+00
6.69E-02
3.57E-01
1.16E+00
1.59E+00

.OOE+00
2.85E-01
1.85E-01
5.41E-01
2.45E-01
5.07E-02
1.31E+00 .
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  »  not detected (detection limit in parentheses).
ng  »  1.0E-09g
8760 operating hours per year
                                   A. 1-4

-------
 Dioxin/Furan
     Isomer
                      TABLE A.1.4b.   SITE 4  (BLB-A)/OUTLET
                      DIOXIN/FURAN  EMISSION  CONCENTRATIONS
                              (As-measured values)
 Isomer Concentration in Flue Gas
           (ng/dscm)
01          Run 02          Run 03
                                                                   Avg,
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 6.00E-02)
ND( 6.00E-02)
ND( 2.51E-02)
1.55E-01
3.09E-01
7.16E-01
1.18E+00

ND( 7.74E-03)
3.29E-01
1.55E-01
1.35E-01
1.93E-01
7.74E-02
8.90E-01

ND( 1.69E-02)
5.06E-02
ND( 2.02E-02)
ND( 5.40E-02)
8.43E-02
3.71E-01
5.06E-01

ND( 3.37E-02)
5.06E-02
ND( 3.88E-02)
5.06E-02
1.69E-01
2.53E-01
5.23E-01

ND( 1.88E-02)
ND( 1.88E-02)
ND( 3.77E-02)
ND( 4.11E-02)
8.56E-02
2.23E-01
3.08E-01

ND( 1.71E-02)
3.42E-02
ND( 2.23E-02)
ND( 8.73E-02)
5.14E-02
1.71E-02
1.03E-01

.OOE+00
1.69E-02
.OOE+00
5.16E-02
1.60E-01
4.36E-01
6.65E-01

.OOE+00
1.38E-01
5.16E-02
6.20E-.02
1.38E-01
1.16E-01
5.05E-01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8760 operating hours per year
                                   A.1-5

-------
                        TABLE A.l.Ba.  SITE 5 (BLB-B)/INLET
                        DIOXIN/FURAN EMISSION CONCENTRATIONS
                                (As-measured values)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 3.95E-02)
ND( 3.95E-02)
ND( 3.95E-02)
ND( 3.95E-02)
1.75E-01
7.02E-01
8.77E-01

4.39E-02
4.39E-02
ND( 1.05E-01)
8.77E-02
4.39E-02
4.39E-02
2.63E-01
ND( 6.25E-02)
ND( 6.25E-02)
ND( 6.94E-03)
ND( 8.68E-02)
7.29E-01
2.33E+00
3.06E+00

ND( 6.25E-02)
- ND( 6.25E-02)
ND( 3.47E-02)
1.04E-01
3.47E-01
2.43E-01
6.94E-01
ND( 2.43E-02)
ND( 2.43E-02)
ND( 8.68E-02)
2.08E-01
6.81E+00
3.42E+01
4.13E+01

6.94E-02
4.51E-01
ND( 8.68E-02)
2.43E-01
6.25E-01
4.51E-01
1.84E+00
.OOE+00
.OOE+00
.OOE+00
6.94E-02
2.57E+00
1.24E+01
1.51E+01

3.78E-02
1.65E-01
.OOE+00
1.45E-01
3.39E-01
2.46E-01
9.33E-01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =-  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                   A. 1-6

-------
                       TABLE A.l.Sb.  SITE 5 (BLB-B)/OUTLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                               (As-measured values)
 Dioxin/Furan
     Isomer
          Isomer Concentration in Flue Gas
                     (ng/dscm)
     Run 01          Run 02          Run 03
                                                                   Avg,
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total  PCDD

 FURANS
ND(
ND(
ND(
3.01E-02)
3.01E-02)
4.51E-02)
1.88E-01
3.38E-01
1.05E+00

1.58E+00
ND( 1.18E-02)
ND( 1.18E-02)
ND( 3.29E-02)
ND( 8.47E-02)
    1.65E-01
    5.41E-01

    7.06E-01
            ND(
            ND(
            ND(
    2.49E-02)
    2.49E-02)
    1.36E-01)
    6.79E-02
    1.58E-01
    5.43E-01

    7.69E-01
              .OOE+00
              .OOE+00
              .OOE+00
             8.53E-02
             2.20E-01
             7.12E-01

             1.02E+00
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
ND(

ND(
3.76E-02)
2.63E-01
2.74E-01)
  14E-01
  63E-01
  52E-02
    7.
    2.
    7.
ND(
ND(
ND(
ND(
4.00E-02)
4.00E-02)
3.29E-02)
1.36E-01)
7.06E-02
4.71E-02
ND(
    1.32E+00
                1.18E-01
2.26E-02
6.79E-02
1.36E-01)
1.36E-01
9.05E-02
4.52E-02

3.62E-01
7.54E-03
1.10E-01
 .OOE+00
2.83E-01
1.41E-01
5.58E-02

5.98E-01
NOTE:  Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                 A. 1-7

-------
                      TABLE A.l.Sa.  SITE 6 (WRI-AJ/OUTLET
                                   (Wire only)
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                              (As-measured values)
 Dioxin/Furan
     Isomer
      Isomer Concentration in Flue Gas
                (ng/dscm)
 Run 01          Run 02          Run 06
                                                                   Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total PCDD

 FURANS
  NR
 .16E-01
  NR
 .10E+00
 .18E+01
1.54E+01

5.09E+01
  NR
  NR
  NR
  NR
2.15E+01
1.24E+01

3.39E+01
                9.29E-02
                  24E+00
                  04E+00
                8.82E+00
                1.39E+02
                1.26E+02

                2.77E+02
             9.29E-02
             8.78E-01
              .04E+00
              .96E+00
2.
5.
             6.41E+01
             5.13E+01

             1.24E+02
 2378 TCDF
 Other TCDF
 'Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
  NR
3.89E+00
5.28E+00
1.23E+01
4.74E+01
2.70E+01
9.58E+01
  NR
9.63E+00
  NR
  , 18E+00
  .12E+01
3.
6.
2.92E+01

1.03E+02
                3.72E-01
                1.63E+01
6.06E+01
2.54E+02
9.97E+01

4.57E+02
3.72E-01
9.94E+00
 .59E+01
 -54E+01
 .21E+02
 .20E+01
1,
2.
1.
5.
                             2.25E+02
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.
NR  *  not reported by Troika.
ND  »  not detected (detection limit in parentheses).
ng  *  1.0E-09g
2080 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the quality
       assurance requirements for all four labelled compounds.  Problems with
       method efficiency resulted from contamination present in the sample
       extracts and corresponding difficulties in achieving acceptable
       chromatographic separations.  The reported analytical results may
       actually represent lower bounds on the true values.  See Section 7.3.1
       for more details.
                                    A. 1-8

-------
                      TABLE A.1.6b.  SITE 6 (WRI-AJ/OUTLET
                             (Wire and Transformers)
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                              (As-measured values)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 03          Run 04          Run 05
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF.
5.12E-02
2.30E-01
4.27E+00
4.95E+01
3.41E+02
1.21E+03
1.61E+03

4.09E-01
2.59E+01
5.46E+01
1.77E+02
3.86E+02
8.08E+02
1.45E+03
NR
1.32E+00
NR
NR
6.19E+01
6.27E+01
1.26E+02

NR
5.42E+01
NR
3.78E+01
2.65E+02
1.37E+02
4.93E+02
1.15E-01
1.53E+00
3.23E+00
5.50E+00
2.07E+01
1.84E+01
4.95E+01

8.07E-01
2.15E+01
1.29E+01
2.10E+01
6.17E+01
4.61E+01
1.64E+02
8.31E-02
1.03E+00
3.75E+00
2.75E+01
1.41E+02
4.32E+02
6.05E+02

6.08E-01
3.39E+01
3.38E+01
7.85E+01
2.38E+02
3.30E+02
7.15E-I-02
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.
NR  =  not reported by Troika.
ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
2080 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the quality
       assurance requirements for all four labelled compounds.  Problems with
       method efficiency resulted from contamination present in the sample
       extracts and corresponding difficulties in achieving acceptable
       chromatographic separations.  The reported analytical results may
       actually represent lower bounds on the true values.  See Section 7 3 1
       for more details.                                                 '
                                  A.1-9

-------
 Dioxin/Furan
     Isomer
                      TABLE A.1.7a.  SITE  7  (WFB-A)/INLET
                      DIOXIN/FURAN EMISSION  CONCENTRATIONS
                               (As-measured values)
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

NR
4.78E+00
NR
4.92E+00
9.07E+00
8.77E+00
2.75E+01

NR
2.98E+01
4.70E+00
3.25E+00
1.31E+00
ND( 3.17E+00)
3.91E+01

3.70E-01
8.72E+00
1.02E+01
1.02E+01
8.12E+00
1.97E+00
3.96E+01

2.31E+00
4.33E+01
1.65E+01
5.90E+00
2.54E+00
2.56E-01
7.08E+01

3.18E-01
1.24E+01
1.16E+01
1.18E+01
6.62E+00
1.73E+00
4.45E+01

2.31E+00
4.75E+01
1.68E+01
6.81E+00
2.23E+00
2.31E-01
7.59E+01

3.44E-01
8.62E+00
1.09E+01
8.98E+00
7.94E+00
4.16E+00
4.09E+01

2.31E+00
4.02E+01
1.27E+01,
5.32E+00
2.02E+00
1.63E-01
6.27E+01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

NR  »  not reported by Troika.
ND  =*  not detected (detection limit in parentheses).
ng  ^  1.0E-09g
8760 operating hours per year
                                   A.1-10

-------
                        TABLE A.I.7b.   SITE 7 (WFB-A)/OUTLET
                        DIOXIN/FURAN EMISSION CONCENTRATIONS
                                (As-measured values)
  Dioxin/Furan
      Isomer
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                    Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
.Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
6.01E-02
1.39E+01
1.77E+01
1.82E+01
9.91E+00
2.82E+00
6.26E+01

3.90E-01
9.88E+00
6.79E+00
4.26E+00
1.80E+00
3.60E-01
2.35E+01
8.80E-02
2.07E+01
1.90E+01
1.94E+01
1.46E+01
3.70E+00
7.75E+01

5.87E-01
1.68E+01
9.93E+00
6.66E+00
3.06E+00
2.49E-01
3.73E+01
' 9.38E-02
7.58E+00
6.08E-I-00
6.23E+00
9.44E+00
3.22E+00
3.26E+01

5.63E-01
6.70E+00
4.00E+00
1.09E+00
l.OOE+00
1.88E-01
1.35E+01
*
8.06E-02
1.41E+01
1.42E+01
1.46E+01
1.13E+01
3.25E+00
5.76E-I-01

5.13E-01
1.11E+01
6.90E+00
4.00E+00
1.96E+00
2.66E-01
2.48E+01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                               A.l-11

-------
 Dioxin/Furan
     Isomer
                      TABLE A.l.Sa.   SITE 8  (BLB-C)/INLET
                      DIOXIN/FURAN  EMISSION  CONCENTRATIONS
                               (As-measured values)
   .  Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
.
ND( .OOE+00)
8.10E-01
1.03E+00
1.59E+00
1.40E+00
9.35E-01
5.76E+00

ND( .OOE+00)
6.17E+00
5.89E+00
4.36E+00
1.15E+00
1.56E-01
1.77E+01

ND( .OOE+00)
3.26E-01
4.81E-01
1.02E+00
7.92E-01
6.52E-01
3.28E+00

ND( .OOE+00)
6.37E-01
3.73E-01
5.12E-01
1.86E-01
9.32E-02
1.80E+00

ND( .OOE+00)
3.86E-01
4.50E-01
1.09E+00
7.72E-01
6.75E-01
3.38E+00

ND( .OOE+00)
7.07E-01
ND( 4.18E-01)
5.14E-01
1.61E-01
ND( 9.65E-02)
1.38E+00

= .OOE+00
5.07E-01
6.53E-01
1.24E+00
9.88E-01
7.54E-01
4.14E+00

.OOE+00
2.50E+00
2.09E+00
1.80E+00
5.00E-01
8.30E-02
6.97E+00
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8760 operating hours per year
                                   A.1-12

-------
                       TABLE A.l.Sb.  SITE 8  (BLB-C)/OUTLET
                       DIOXIN/FURAN EMISSION  CONCENTRATIONS
                                (As-measured values)
  Dioxin/Furan
      Isomer
           Isomer Concentration  in  Flue  Gas
                     (ng/dscm)
      Run  01           Run  02           Run  03
                                                                    Avg.
  DIOXINS


  2378 TCDD
  Other TCDD
  Penta-CDD
  Hexa-CDD
  Hepta-CDD
  Octa-CDD

  Total PCDD

  FURANS
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
ND(

ND(
ND(

ND(
 9.57E-03)
 4.78E-02
 2.39E-02)
 7.18E-02
 1.20E-01
 2.87E-01

 5.26E-01
1.44E-02)
1.20E-01
4.78E-02)
1.20E-01
9.57E-02
4.78E-02
ND(
4.52E-03)
1.24E-01
2.49E-01
4.75E-01
1.38E+00
2.71E+00

4.94E+00
    1.81E-02
    2.99E-01
    2.49E-01
    2.60E-01
    3.85E-01
    4.52E-02
ND( 2.51E-03)
    6.28E-02
    1.01E-01
    2.39E-01
    4.77E-01
    4.52E-01

    1.33E+00
            ND(
    2.51E-03)
    3.39E-01
    4.77E-01
    5.15E-01
    3.77E-01
    5.03E-02
  .OOE+00
  .84E-02
  .16E-01
  .62E-01
    3.83E-01        1.26E+00        1.76E+00
                                                 6.59E-01
                                                 1.15E+00

                                                 2.27E+00
6.03E-03
2.52E-01
2.42E-01
2.98E-01
2.86E-01
4.78E-02

1.13E+00
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  ?°Ld««ected (detection limit in parentheses).
ng  »  1.0E-09g                                    •
8760 operating hours per year
                                  A.1-13

-------
                         TABLE A.1.9a.  SITE 9 (CRF-AJ/INLET
                         DIOXIN/FURAN EMISSION CONCENTRATIONS
                                 (As-measured values)
  Dioxin/Furan
      Isomer
       Isomer Concentration  In  Flue Gas
                 (ng/dscm)
 Run 01          Run 02          Run 03
                                                                    Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total PCDD

 FURANS
5.12E-02
9.21E-01
  .02E+00
  .04E+00
  .95E+00
2.
6.
7.
9.18E+00

2.62E+01
4.95E-02
8.66E-01
4.70E-01
2.90E+00
6.26E+00
                1.06E+01

                2.12E+01
                                8.87E-02
3,
2.
3,
3,
2,
08E+00
51E+00
64E+00
97E+00
13E+00
                1.54E+01
6,
1,
1,
4.
6.
7,
31E-02
62E+00
67E+00
19E+00
06E+00
32E+00
             2.09E+01
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
8.95E-01
1.22E+01
8.77E+00
1.
7.
  04E+01
  29E+00
3.35E+00

4.29E+01
1.16E+00
1.73E+01
9.36E+00
1.23E+01
1.11E+01
4.88E+00

5.60E+01
                                9.31E-01
1.40E+01
1.25E+01
1.14E+01
8.96E+00
3.46E+00
                                                     5.12E+01
           9.97E-01
           1.45E+01
           1.02E+01
           1.13E+01
           9.12E+00
           3.90E+00
                                             5.01E+01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  -  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                  A.1-14

-------
 Dioxin/Furan
      Isomer
                       TABLE A.1.9b.  SITE 9 (CRF-AJ/OUTLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                               (As-measured values)
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01   .       Run 02          Run 03
                                                                    Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 1.79E-01)
5.87E-01
3.83E-01
6.12E-01
4.59E-01
4.08E-01
2.45E+00

ND( 1.53E-01)
5.36E-01
1.53E-01
2.81E-01
3.32E-01
2.30E-01
1.53E+00

ND( 5.29E-02)
ND( 7.94E-02)
1.32E-01
3.17E-01
3.44E-01
2.91E-01
1.08E+00

ND( 1.59E-01)
5.29E-01
2.91E-01
2.12E-01
2.12E-01
1.32E-01 :
1.38E+00

ND( 4.61E-02)
6.91E-02
ND( 1.15E-01)
2.07E-01
2.30E-01
2.30E-01
7.37E-01

ND( 9.22E-02)
3.23E-01
ND( 1.15E-01)
1.84E-01
1.61E-01
2.53E-01
9.22E-01

.OOE+00
2.19E-01
1.72E-01
3.79E-01
3.45E-01
3.10E-01
1.42E+00

.OOE+00
4.62E-01
1.48E-01
2.26E-01
2.35E-01
2.05E-01
1.28E+00
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g                                    '
8760 operating hours per year
                                  A.1-15

-------
                     TABLE A.1.10.  SITE  10  (MET-A)/OUTLET
                     DIOXIN/FURAN EMISSION CONCENTRATIONS
                              (As-measured values)
 Dioxin/Furan
     Isomer
      Isomer Concentration in Flue Gas
                (ng/dscm)
 Run 02          Run 03          Run 04
                                                                   Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total  PCDD

 FURANS
1.75E+01
3.53E+01
6.28E+01
1.45E+02
1.09E+02
6.51E+01

4.36E+02
8.50E+00
7.
1.
1,
2,
1,
01E+01
16E+02
08E+02
93E+02
85E+02
7.81E+02
  77E+00
  40E+01
  45E+01
  76E+01
1.48E-I-02
1.06E+02
              4.56E+02
1.06E+01
5.31E+01
8.46E+01
1.07E+02
1.84E+02
1.19E+02

5.58E+02
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
1.86E+02
6.28E+02
5.59E+02
5.10E+02
1.81E+02
1.23E+02

2.19E+03
2.53E+02
8.90E+02
9.27E+02
3.04E+02
5.32E+02
3.60E+02

3.27E+03
              2.65E+02
              1.33E+03
                .65E+02
                .58E+02
              1.97E+02
              1.85E+02
7.
2.
              3.00E+03
2.35E+02
9.48E+02
  .50E+02
  .57E+02
  .03E+02
  .22E+02
7.
3.
3.
2.
             2.82E+03
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  -  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8160 operating hours per year
NOTE:  Surrogate recoveries could not be determined for Run 02 and 04
       dioxin/furan samples because of the relatively large quantities of
       native CDD and CDF species present in the samples.  The reported
       analytical results may actually represent lower bounds on the true
       values.  See Section 7.3.1 for more details.
                                   A.1-16

-------
                      TABLEA.l.lla  SITE 11  (DBR-AJ/INLET
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                              (As-measured values)
  Dioxin/Furan
      Isomer
                            Isomer Concentration  in  Flue Gas
                                      (ng/dscm)
                           01           Run  02          Run 03
                                                                   Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
4.98E+00
1.72E+01
2.60E+01
2.53E+01
1.42E+02
5.29E+01
2.69E+02

1.44E+01
1.94E+02
1.69E+02
4.34E+01
1.32E+02
4.26E+01
5.95E+02
3.06E+00
1.05E+01
9.15E+00
5.94E+00
2.47E+00
2.07E+00
3.32E+01

1.16E+01
1.15E+02
5.86E+01
1.08E+01
4.49E+00
1.04E+00
2.02E+02
2.49E+00
2.06E+01
3.11E+01
5.36E+01
5.32E+01
1.16E+01
1.73E+02

1.26E+01
2.67E+02
1.67E+02
9.15E+01
5.18E+01
1.23E+01
6.02E+02
3.51E+00
1.61E+01
2.21E+01
2.83E+01
6.60E+01
2.22E+01
1.58E+02

1.29E+01
1.92E+02
1.31E+02
4.86E+01
6.28E+01
1 .86E+01
4.66E+02
NOTE: Isomer concentrations shown are at as -measured oxygen  conditions.

                    (detection 11m1t in Parentheses).
ng  "  i°
1536 operating hours per year
NOTE:  Several recoveries for the  MM5  train  samples did not meet the quality
       mPt'hnH^?/6^^6"16"*8/0!;  a11  four labe"led compounds   Problems with
       SJtSt! ffl5iency resulted  from contamination present in the sample
       extracts and corresponding  difficulties in achieving acceptable
       chromatographic separations.  The reported analytical results may

       fir more iT*?        r  ^""^ 0" th& trU6 Val"6S'  See Sectlw 7.3.1
                                 A.1-17

-------
                     TABLE A.I.lib.  SITE 11 (DBR-A)/OUTLET
                     DIOXIN/FURAN EMISSION CONCENTRATIONS
                             (As-measured values)
Dioxin/Furan
    Isomer
     Isomer Concentration in Flue Gas
               (ng/dscm)
Run 01          Run 02          Run 03
                                                                  Avg,
DIOXINS
2378 TCDO .
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.86E-02
6.86E-01
6.29E-01
5.71E-01
1.03E+00
5.71E-01
3.51E+00

6.86E-01
7.23E+00
4.26E+00
2.11E+00
1.46E+00
4.00E-01
1.61E+01
1.10E-02
3.47E-01
1.10E-01
1.65E-01
3.03E-01
3.03E-01
1.24E+00

2.20E-01
5.96E+00
1.79E+00
8.82E-01
5.79E-01
1.65E-01
9.60E+00
2.58E-02
4.64E-01
1.80E-01
2.71E-01
3.35E-01
2.84E-01
1.56E+00

2.32E-01
4.74E+00
1.79E+00
7.86E-01
5.15E-01
1.29E-01
8.20E+00
2.18E-02
4.99E-01
3.06E-01
3.36E-01
5.56E-01
3.86E-01
2.10E+00

3.79E-01
5.98E+00
2.61E+00
1.26E+00
8.50E-01
2.31E-01
1.13E+01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  »  not detected (detection limit in parentheses).
ng  -  1.0E-09g
1536 operating hours per year
                                 A.1-18

-------
  Dioxin/Furan
       Isomer
                        TABLE A.1.12a.  SITE  12  (SSI-O/INLET
                        DIOXIN/FURAN  EMISSION CONCENTRATIoSiF
                                (As-measured values)
                        Isomer Concentration in Flue Gas"
                   D   A,         (ng/dscm)
                   Run 01          Run 02          Run 03
                                                                    Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
NR
NR
NR
NR
NR
NR
NR

NR
NR
NR
NR
NR
NR
NR
WD
WK
1.39E+00
ND( 7.88E-01)
2.85E+00
2.16E+01
2.37E+01
4.95E+01

1.65E+01
4.00E+01
3.60E+01
4.09E+00
. 3.92E+01
3.93E+01
1.75E+02

NR
4.23E+00
2.42E-01
3.32E+00
9.82E+00
1.10E+01
2.86E+01

3.94E+01
5.86E+01
5.28E+01
4.95E+00
5.77E+00
6.62E+00
1.68E+02

NR
2.81E+00
1.21E-01
3.09E+00
1.57E+01
1.74E+01
3.91E+01

2.80E+01
4.93E+01
4.44E+01
4.52E+00
2.25E+01
2.30E+01
1.72E+02
ng
8760
   	 UJ  iru,R.     are at "measured oxygen conditions.
-   l°OE-§9gCt6d (detect1on Iim1t in Parentheses).
operating  hours per year



'
                                   A.1-19

-------
                       TABLE A.l.lZb.   SITE  12  (SSI-C)/OUTLET
                       DIOXIN/FURAN  EMISSION CONCENTRATIONS
                                (As-measured  values)
 Dioxin/Furan
     Isomer
      Isomer Concentration in Flue Gas
                (ng/dscm)
 Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total  PCDD

 FURANS
2.33E-02
1.26E+00
1.33E-01
8.80E-01
1.30E+00
1.06E+00
4.65E+00
 •30E-02
 .56E+00
1.97E-01
1.
7.
4.
56E+00
27E+00
90E+00
1.55E+01
3.62E-02
1.85E+00
 .08E-01
 .61E+00
 .55E+00
 .83E+00
3.
1,
3.
2.
              1.02E+01
                           2.75E-02
                             55E+00
                             13E-01
1.35E+00
4.
2.
04E+00
93E+00
             1.01E+01
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
l.OlE-f-01
2.25E+01
1.67E+01
2.77E+00
1130E+00
5.98E-01

5.41E+01
9.18E+00
2.95E+01
1.91E+01
8.59E+00
2.35E+01
1.83E+01

1.08E+02
              1.
              3,
  12E+01
  41E+01
              2.44E+01
              7.48E+00
              l.OOE+01
              7.36E+00

              9.45E+01
             1.02E+01
             2.87E+01
             2.01E+01
             6.28E+00
             1.16E+01
             8.74E+00

             8.56E+01
NOTE: Isomer concentrations shown are at as-measured oxygen conditions.

ND  -  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                  A.1-20

-------
                APPENDIX A.2

Summary of PCDD/PCDF Emission Concentrations
          (Normalized to 3% Oxygen)

-------
                                 APPENDIX A. 2
The formula used for correction to 3 percent 02 was:
     Pmx
  20.9 - KQ
  20.9 - %0,
where:
Pc
Pm
Ko
%0
   corrected value
   measured value
   reference 02 concentration
2 » percent by volume of 02
Reference:
               Stack Sampling Technical  Information / A Collection of
               Monographs and Papers, Volume 1, U.S. Environmental Protection
               Agency, Publication No. EPA-450/2-78-042a, October, 1978.

-------
Dioxin/Furan
    Isomer
                         TABLE A.2.1.  SITE 1 (SSI-AJ/OUTLET
                         DIOXIN/FURAN EMISSION CONCENTRATIONS
                             (values corrected to 3% 0?)



                             Isomer  Concentration  in  Flue Gas
                     Dl   rt1      »--«	 ~ 3% oxygen)
                     Kun 01          Run 02          Run
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

4.04E-02
1.35E+01
1.68E-01
ND( 3.23E-01)
3.02E+00
7.18E+00
2.39E+01

NR
3.91E+01
1.27E+01
ND( 7.52E-01)
4.95E-01
ND( 4.50E-01)
5.22E+01

7.78E-02
1.38E+01
3.57E-01
1.06E+00
2.59E+00
4.69E+00
2.26E+01

fcfn
NR
3.80E+01
1.10E+01
ND( 6.23E-01)
6.67E-01
3.11E-01 .
5.00E+01

1.90E-02
5.83E+00
ND( 1.29E-01)
4.64E-01
1.75E+00
4.14E+00
1.22E+01


NR
2.11E+01
6.41E+00
3.09E-01
3.47E-01
ND( 1.74E-01)
2.82E+01

4.58E-02
1.11E+01
1.75E-01
5.09E-01
2.45E+00
5.34E+00
1.96E+01


NR
3.28E+01
l.OOE+01
1.03E-01
5.03E-01
1.04E-01
4-.35E+01
{detect1on limit
ng  I
6000 operating hours per year
                                              to 3%

                                      Parentheses).
                                  A.2-1

-------
                    TABLE  A.2.2.   SITE  2  (ISW-AJ/OUTLET
                    DIOXIN/FURAN  EMISSION CONCENTRATIONS
                        (Values corrected to  3%  0-)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 03          Run 04
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.17E+00
3.58E+01
5.43E+01
1.06E+02
1.27E+02
5.21E+01
3.78E+02

1.19E+01
3.66E+02
3.81E+02
6.08E+02
3.35E+02
5.75E+01
1.76E+03
4.89E+00
8.67E+01
1.08E+02
1.31E+02
1.67E+02
5.87E+01
5.57E+02

2.17E+01
4.84E+02
4.76E+02
4.40E+02
2.98E+02
6.08E+01
1.78E+03
6.52E+00
1.08E+02
1.47E+02
1.98E+02
4.07E+02
7.30E+01
9.40E+02

2.87E+01
8.74E+02
9.64E+02
9.08E+02
7.78E+02
7.95E+01
3.63E+03
4.53E+00
7.69E+01
1.03E+02
1.45E+02
2.34E+02
6.13E+01
6.25E+02

2.08E+01
5.74E+02
6.07E+02
6.52E+02
4.71E+02
6.60E+01
2.39E+03
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  -  not detected (detection limit in parentheses).
ng  »  1.0E-09g
2200 operating hours per year
                                     A. 2-2

-------
 Dioxin/Furan
     Isomer
                       TABLE A.2.3.  SITE 3  (SSI-B)/OUTLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                            (Values corrected to 3% 02)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  -  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8760 operating hours per year
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 6.59E-02)
1.10E+00
ND( 1.10E-01)
ND( 1.05E+00)
6.59E-01
1.76E+00
3.52E+00

4.18E+00
4.29E+01
1.19E+01
4.84E+00
ND( 1.36E+00)
2.20E-01
6.40E+01

ND( 2.01E-01)
ND( 2.01E-01)
ND( 1.51E-01)
ND( 3.43E-01)
ND( 2.26E-01)
5.86E-01
5.86E-01

8.36E-01
5.10E+00
ND( 1.51E-01)
ND( 1.92E-01)
ND( 2.43E-01)
ND( 4.18E-02)
5.94E+00

ND( 3.90E-02)
9.74E-02
ND( 1.95E-01)
ND( 2.92E-02)
ND( 1.36E-01)
5.84E-01
6.82E-01

1.36E+00
9.06E+00
2.44E+00
N0( 7.21E-01)
ND( 6.82E-02)
ND( 5.84E-02)
1.29E+01

.OOE+00
3.99E-01
.OOE+00
.OOE+00
2.20E-01
9.76E-01
1.59E+00

2.13E+00
1.90E+01
4.77E+00
1.61E+00
.OOE+00
7.33E-02
2.76E+01
                                       A.2-3

-------
Dioxin/Furan
    Isomer
                     TABLE A.2.4a.   SITE 4 (BLB-A)/INLET
                     DIOXIN/FURAN EMISSION CONCENTRATIONS
                         (Values corrected to 3% 02)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                  Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD .
Hexa-CDD
Hepta-CDO
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 2.18E-02)
ND( 2.18E-02)
ND( 8.46E-02)
1.37E-01
3.28E-01
9.01E-01
1.37E+00

ND( 1.64E-02)
2.73E-01
3.55E-01
4.92E-01
2.73E-01
8.19E-02
1.47E+00

ND( 2.23E-02)
2.23E-02
ND( 2.68E-02)
8.93E-02
2.23E-01
7.14E-01
1.05E+00

ND( 8.93E-02)
2.23E-01
2.68E-01
4.46E-01
1.34E-01
8.93E-02
1.16E+00

ND( 8.40E-02)
ND( 8.40E-02)
ND( 1.12E-01)
ND( 4.39E-01)
6.54E-01
2.29E+00
2.94E+00

ND( 1.87E-01)
4.67E-01
ND( 6.91E-01)
8.87E-01
4.20E-01
ND( 2.94E-01)
1.77E+00

.OOE+00
7.44E-03
.OOE+00
7.53E-02
4.01E-01
1.30E+00
1.79E+00

.OOE+00
3.21E-01
2.08E-01
6.08E-01
2.76E-01
5.71E-02
1.47E+00
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  =  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8760 operating hours per year
                                       A. 2-4

-------
                       TABLE A.2.4b.  SITE 4  (BLB-A)/OUTLET
                       DIOXIN/FURAN EMISSION  CONCENTRATIONS
                            (Values corrected  to 3% 02)
 Dioxin/Furan
      Isoraer
     Isomer Concentration in Flue Gas
            (ng/dscm 0 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 6.75E-02)
ND( 6.75E-02)
ND( 2.83E-02)
1.74E-01
3.48E-01
8.05E-01
1.33E+00

ND( 8.70E-03)
3.70E-01
1.74E-01
1.52E-01
2.18E-01
8.70E-02
1. OOE+00

ND( 1.97E-02)
5.91E-02
ND( 2.37E-02)
ND( 6.31E-02)
9.86E-02
4.34E-01
5.91E-01

ND( 3.94E-02)
5.91E-02
ND( 4.53E-02)
5.91E-02
1.97E-01
2.96E-01
6.11E-01

ND( 2.03E-02)
ND( 2.03E-02)
ND( 4.05E-02)
ND( 4.42E-02)
9.21E-02
2.40E-01
3.32E-01

ND( 1.84E-02)
3.68E-02
ND( 2.40E-02)
ND( 9.40E-02)
5.53E-02
1.84E-02
1.11E-01

.OOE+00
1.97E-02
.OOE+00
5.80E-02
1.80E-01
4.93E-01
7.50E-01

.OOE+00
1.55E-01
5.80E-02
7.05E-02
1.57E-01
1.34E-01
5.74E-01
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g                                    '
8760 operating hours per year
                                       A. 2-5

-------
                    TABLE A.2.5a.  SITE 5  (BLB-B)/INLET
                    DIOXIN/FURAN  EMISSION  CONCENTRATIONS
                         (Values corrected  to 3% 02)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 4.15E-02)
ND( 4.15E-02)
ND( 4.15E-02)
ND( 4.15E-02)
1.84E-01
7.37E-01
9.22E-01

4.61E-02
4.61E-02
ND( 1.11E-01)
9.22E-02
4.61E-02
4.61E-02
2.77E-01
ND( 6.90E-02)
ND( 6.90E-02)
ND( 7.67E-03)
ND( 9.59E-02)
8.05E-01
2.57E+00
3.37E+00

ND( 6.90E-02)
ND( 6.90E-02)
ND( 3.83E-02)
1.15E-01
3.83E-01
2.68E-01
7.67E-01
ND( 2.77E-02)
ND( 2.77E-02)
ND( 9.89E-02)
2.37E-01
7.75E+00
3.90E+01
4.70E+01

7.91E-02
5.14E-01
ND( 9.89E-02)
2.77E-01
7.12E-01
5.14E-01
2.10E+00
.OOE+00
.OOE+00
.OOE+00
7.91E-02
2.91E+00
1.41E+01
1.71E+01

4.17E-02
1.87E-01
.OOE+00
1.61E-01
3.81E-01
2.76E-01
1.05E+00
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  -  not detected (detection limit in parentheses).
ng  «  1.0E-09g
8760 operating hours per year
                                       A.2-6

-------
                         TABLE A.2.5b.   SITE 5  (BLB-B)/OUTLET
                         DIOXIN/FURAN  EMISSION  CONCENTRATIONS
                             (Values corrected  to  3% 02)
  Dloxin/Furan
      Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm 9 3% oxygen)
Run 01          Run 02          Run 03
                                                                    Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( •
ND( 2
ND( !
2
4
1
1

ND( 4
3
ND( 3
8
3
8
1
I.59E-02)
I.59E-02)
L38E-02)
•24E-01
.03E-01
.25E+00
.88E+00

.48E-02)
.14E-01
.27E-01)
.51E-01
.14E-01
.96E-02
.57E+00
ND( 1
ND( 1
ND( 3
ND( 9
1.
6.
8.

ND( 4.
ND( 4.
ND( 3.
ND( 1.
8.
5.
1.
.36E-02
•36E-02
-80E-02
77E-02
90E-01
24E-01
14E-01

62E-02
62E-02
80E-02
57E-01
14E-02
43E-02
36E-01
) ND( 2
) ND( 2
) ND( 1
) 7
1
6
8

2
7
ND( 1
1
1
5
4
.85E-02)
.85E-02)
•56E-01)
•78E-02
.82E-01
.23E-01
.82E-01

.59E-02
.78E-02
.56E-01)
•56E-01
.04E-01
•19E-02
.15E-01
1
2
a
i

8
1
3
1
6
7
.OOE+00
.OOE+00
.OOE+00
.01E-01
.58E-01
.34E-01
.19E+00

.65E-03
.31E-01
.OOE+00
•36E-01
.66E-01
.53E-02
•06E-01
NOTE: Isomer concentrations shown are corrected to 3% oxygen.


ng  =  l°OE-09gCt6d (detect1on 11rait 1n Parentheses).
8760 operating hours per year
                                       A.2-7

-------
                    TABLE A.2.6a  SITE 6  (WRI-A)/OUTLET
                                 (Wire only)
                    DIOXIN/FURAN  EMISSION CONCENTRATIONS
                         (Values corrected to 3% 02)
 Dioxin/Furan
     Isomer
      Isomer Concentration in Flue Gas
             (ng/dscm @ 3% oxygen)
 Run 01          Run 02          Run 06
                                                                   Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total PCDD

 FURANS
  NR   .
 .59E-01
  NR
 .36E+00
 .45E+01
1.67E+01

5.52E+01
  NR
  NR
  NR
  NR
2.24E+01
1.29E+01

3.53E+01
1.38E-01
  84E+00
3.04E+00
1.31E+01
2.07E+02
1.87E+02
4.12E+02
1.38E-01
1,
3.
 .20E+00
 .04E+00
8.23E+00
8.79E+01
7.24E+01
1.73E+02
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
  NR
4.22E+00
5.72E+00
1.33E+01
5.14E+01
2.93E+01

1.04E+02
  NR
  •OOE+01
  NR
  .31E+00
  .37E+01
  •04E+01
1.07E+02
5.53E-01
2.42E+01
  96E+01
  02E+01
  78E+02
  48E+02
6.80E+02
5.
1,
2.
3.
  .53E-01
  .28E+01
  .27E+01
  .56E+01
1.64E+02
6.93E+01
3.05E+02
NOTE: Isomer concentrations shown are corrected to 3% oxygen.
NR  -  not reported by Troika.
ND  -  not detected (detection limit in parentheses).
ng  -  1.0E-09g
2080 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the quality
       assurance requirements for all four labelled compounds.  Problems with
       method efficiency resulted from contamination present in the sample
       extracts and corresponding difficulties in achieving acceptable
       chromatographic separations.  The reported analytical results may
       actually represent lower bounds on the true values.  See Section 7.3.1
       for more details.
                                        A. 2-8

-------
Dioxin/Furan
    Isomer
                        TABLE A.2.6b  SITE 6 (WRI-AJ/OUTLET
                              (Wire and Transformers)
                        DIOXIN/FURAN EMISSION CONCENTRATIONS
                            (Values corrected to 3% 0-)


                             Isomer  Concentration in Flue Gas
                                            @ 3% oxygen)
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

5.83E-02
2.62E-01
4.87E+00
5.64E+01
3.89E+02
1.38E+03
1.83E+03

4.66E-01
2.95E+01
6.22E+01
2.01E+02
4.40E+02
9.21E+02
1.65E+03

NR
1.68E+00
NR
NR
7.85E+01
7.95E+01
1.60E+02

NR
6.86E+01
NR
4.79E+01
3.35E+02
1.73E+02
6.25E+02

1.94E-01
2.57E+00
5.43E+00
9.26E+00
3.48E+01
3.10E+01
8.32E+01

1.36E+00
3.61E+01
2.17E+01
3.52E+01
1.04E+02
7.75E-I.01
2.76E+02
Mvg. .
1.26E-01
1.50E+00
5.15E+00
3.28E+01
1.67E+02
4.98E+02
7.05E+02

9.13E-01
4.48E+01
4.20E+01
9.48E+01
2.93E+02
3.90E+02
8.66E+02
                               are
ng  ^  l°OE-09|"ed (detect""1
2080 operating hours per year
NOTE:
                                  in parentheses).
                               B?
       method efficiency resulted  from
                                                             p1
                                                          .   Problems with
                                      A.2-9

-------
Dioxin/Furan
    Isomer
                    TABLE A.2.7a.   SITE 7 (WFB-A)/INLET
                    DIOXIN/FURAN EMISSION CONCENTRATIONS
                        (Values corrected to 3% 02)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
NOTE: Isomer concentrations shown are corrected to 3% oxygen.
NR  »  not reported by Troika.
ND  =  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8760 operating hours per year
                                                                  Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

NR
1.12E+01
NR
1.15E+01
2.12E+01
2.05E+01
6.44E+01

NR
6.97E+01
1.10E+01
7.60E+00
3.07E+00
ND( 7.41E+00)
9.13E+01

9.39E-01
2.21E+01
2.60E+01
2.59E+01
2.06E+01
4.98E+00
1.01E+02

5.85E+00
1.10E+02
4.19E+01
1.50E+01
6.43E+00
6.50E-01
1.80E+02

8.06E-01
3.13E+01
2.95E+01
2.99E+01
1.68E+01
4.40E+00
1.13E+02

5.86E+00
1.20E+02
4.26E+01
1.73E+01
5.64E+00
5.86E-01
1.92E+02

8.73E-01
2.15E+01
2.78E+01
2.25E+01
1.95E+01
9.96E+00
1.02E+02

5.86E+00
l.OOE+02
3.18E+01
1.33E+01
5.05E+00
4.12E-01
1.56E+02
                                         A.2-10

-------
                      TABLE A.2.7b.  SITE 7  (WFB-AJ/OUTLET
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                          (Values corrected to 3% 09)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF ..
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.12E-01
4.90E+01
6.24E+01
6.42E+01
3.50E+01
9.96E+00
2.21E+02

1.38E+00
3.49E+01
2.40E+01
1.51E+01
6.36E+00
1.27E+00
8.29E+01
2.73E-01
6.44E+01
5.89E+01
6.01E+01
4.54E+01
1.15E+01
2.40E+02

1.82E+00
5.22E+01
3.08E+01
2.07E+01
9.51E+00
7.74E-01
1.16E+02
3.52E-01
2.84E+01
2.28E+01
2.34E+01
3.54E+01
1.21E+01
1.22E+02

2.11E+00
2.51E+01
1.50E+01
4.10E+00
3.75E-I-00
7.03E-01
5.08E+01
2.79E-01
4.73E+01
4.80E+01
4.92E+01
3.86E+01
1.12E+01
1.95E+02

1.77E+00
3.74E-H01
2.33E+01
1.33E+01
6.54E+00
9.16E-01
8.32E+01
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

NO  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                       A.2-11

-------
Dioxin/Furan
    Isomer
                   TABLE A.2.8a.   SITE 8  (BLB-C)/INLET
                   DIOXIN/FURAN  EMISSION  CONCENTRATIONS
                        (Values corrected  to 3% 02)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  =  not detected (detection limit in parentheses).
ng  *  1.0E-09g
8760 operating hours per year
                                                                  Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( .OOE+00)
1.76E+00
2.23E+00
3.45E+00
3.04E+00
2.03E+00
1.25E+01

ND( .OOE+00)
1.34E+01
1.28E+01
9.46E+00
2.50E+00
3.38E-01
3.84E+01

ND( .OOE+00)
7.62E-01
1.13E+00
2.40E+00
1.85E+00
1.52E+00
7.66E+00

ND( .OOE+00)
1.49E+00
8.71E-01
1.20E+00
4.36E-01
2.18E-01
4.21E+00

ND( .OOE+00)
7.80E-01
9.10E-01
2.21E+00
1.56E+00
1.37E+00
6.83E+00

ND( .OOE+00)
1.43E+00
ND( 8.45E-01)
1.04E+00
3.25E-01
ND( 1.95E-01)
2.80E+00

.OOE+00
1.10E+00
1.42E+00
2.68E+00
2.15E+00
1.64E+00
9. OOE+00

.OOE+00
5.43E+00
4.55E+00
3.90E+00
1.09E+00
1.85E-01
1.51E+01
                                        A. 2-12

-------
 Dioxin/Furan
      Isomer
                    TABLE A.2.8b.  SITE 8  (BLB-C)/OUTLET
                    DIOXIN/FURAN EMISSION  CONCENTRATIONS
                         (Values corrected  to 3% 0-)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 2.39E-02)
1.20E-01
ND( 5.98E-02)
1.79E-01
2.99E-01
7.18E-01
1.32E+00

ND( 3.59E-02)
2.99E-01
ND( 1.20E-01)
2.99E-01
2.39E-01
1.20E-01
9.57E-01

ND( 3.90E-03)
1.07E-01
2.15E-01
4.10E-01
1.19E+00
2.34E+00
4.27E+00

1.56E-02
2.58E-01
2.15E-01
2.25E-01
3.32E-01
3.90E-02
1.08E+00

ND( 6.11E-03)
1.53E-01
2.44E-01
5.81E-01
1.16E+00
1.10E+00
3.24E+00

ND( 6.11E-03)
8.25E-01
1.16E+00
1.25E-I-00
9.17E-01
1.22E-01
4.28E+00

.OOE+00
1.27E-01
1.53E-01
3.90E-01
8.84E-01
1.39E+00
2.94E+00

5.21E-03
4.61E-01
4.59E-01
5.92E-01
4.96E-01
9.36E-02
2.11E+00
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

NO  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                       A.2-13

-------
                     TABLE A.2.9a.  SITE 9 (CRF-A)/INLET
                     DIOXIN/FURAN EMISSION CONCENTRATIONS
                         (Values corrected to 3% 02)
Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                  Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
6.86E-02
1.23E+00
2.71E+00
8.10E+00
1.07E+01
1.23E+01
3.51E+01

1.20E+00
1.63E+01
1.18E+01
1.40E+01
9.78E+00
4.49E+00
5.75E+01
6.12E-02
1.07E+00
5.82E-01
3.58E+00
7.75E+00
1.32E+01
2.62E+01

1.44E+00
2.14E+01
1.16E+01
1.52E+01
1.37E+01
6.03E+00
6.93E+01
1.44E-01
5.02E+00
4.08E+00
5.92E+00
6.46E+00
3.46E+00
2.51E+01

1.52E+00
2.28E+01
2.04E+01
1.85E+01
1.46E+01
5.63E+00
8.34E+01
9.14E-02
2.44E+00
2.46E+00
5.87E+00
8.29E+00
9.65E+00
2.88E+01

1.39E+00
2.02E+01
1.46E+01
1.59E+01
1.27E+01
5.39E+00
7.01E+01
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  -  not detected (detection limit in parentheses).
ng  »  1.0E-09g
8760 operating hours per year
                                       A.2-14

-------
                       TABLE A.2.9b.  SITE 9 (CRF-A)/OUTLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                           (Values corrected to 3% 0-)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 4.61E-01)
1.52E+00
9.88E-01
1.58E+00
1.19E+00
1.05E+00
6.32E+00

ND( 3.95E-01)
1.38E+00
3.95E-01
7.25E-01
8.56E-01
5.93E-01
3.95E+00
ND( 1.31E-01)
ND( 1.96E-01)
3.27E-01
7.84E-01
8.49E-01
7.19E-01
2.68E+00

ND( 3.92E-01)
1.31E+00
7.19E-01
5.23E-01
5.23E-01
3.27E-01
3.40E+00
ND( 1.30E-01)
1.95E-01
ND( 3.25E-01)
5.84E-01
6.49E-01
6.49E-01
2.08E+00

ND( 2.60E-01)
9.09E-01
ND( 3.25E-01)
5.19E-01
4.54E-01
7.14E-Q1
2o60E+00
.OOE+00
5.70E-01
4.38E-01
9.83E-01
8.95E-01
8.07E-01
3.69E+00

.OOE+00
1.20E+00
3.71E-01
5.89E-01
6.11E-01
5.44E-01
3.32E+00
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  =  not detected (detection limit in parentheses).
ng  =  1.0E-09g
8760 operating hours per year
                                       A.2-15

-------
                      TABLE A.2.10.  SITE  10  (MET-A)/OUTLET
                      DIOXIN/FURAN EMISSION CONCENTRATIONS
                           (Values corrected to 3% 02)
 Dioxin/Furan
     Isomer
      Isomer Concentration in Flue Gas
             (ng/dscm @ 3% oxygen)
 Run 02          Run 03          Run 04
                                                                   Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total PCDD

 FURANS
3.95E+02
7.
1,
3,
93E+02
41E+03
27E+03
2.46E+03
1.47E+03

9.80E+03
1.70E+02
1.40E+03
2.33E+03
2.15E+03
5.86E+03
3.70E+03
              1.56E+04
•1.30E+02
 1.22E+03
 1.68E+03
 1.52E+03
 3.34E+03
 2.39E+03

 1.03E+04
2.32E+02
1.14E+03
1.81E+03
2.32E+03
3.89E+03
2.52E+03

1.19E+04
 2378 TCDF
 Other TCDF
 Penta-CDF
 Hexa-CDF
 Hepta-CDF
 Octa-CDF

 Total PCDF
4.18E+03
1.41E+04
1.26E+04
1.15E+04
4.06E+03
2.76E+03

4.92E+04
                06E+03
                78E+04
                85E+04
                08E+03
              1.06E+04
              7.20E+03
              6.53E+04
                5.97E+03
                2.99E+04
                1.72E+04
                5.81E+03
                4.44E+03
                4.15E+03

                6.74E+04
               .07E+03
               .06E+04
               .61E+04
               .79E+03
              6.38E+03
              4.70E+03

              6.07E+04
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  -  not detected (detection limit in parentheses).
ng  -  1.0E-09g
8160 operating hours per year
NOTE:  Surrogate recoveries could not be determined for Run 02 and 04
       dioxin/furan samples because of the relatively large quantities of
       native CDD and CDF species present in the samples.  The reported
       analytical results may actually represent lower bounds on the true
       values.  See Section 7.3.1 for more details.
                                      A.2-16

-------
                       TABLE A.2.11a.  SITE 11 (DBR-A)/INLET
                       DIOXIN/FURAN EMISSION CONCENTRATIONS
                           (Values corrected to 3% 02)
 Dioxin/Furan
     Isomer
                     Isomer Concentration  in  Flue  Gas
                           (ng/dscm  @ 3%  oxygen)
               Run 01           Run 02           Run  03
                                                                    Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD*
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
1.76E+01
6.07E+01
9.19E+01
8.94E+01
5.03E+02
1.87E+02
9.49E+02

5.09E+01
6.85E+02
5.95E+02
1.53E+02
4.66E+02
1.50E+02
2.10E+03
1.84E+01
6.30E+01
5.49E+01
3.57E+01
1.48E+01
1.24E+01
1.99E+02

6.98E+01
6.92E+02
3.51E+02
6.47E+01
2.70E+01
6.22E+00
1.21E+03
1.32E+01
1.09E+02
1.65E+02
2.84E+02
2.81E+02
6.15E+01
9.14E+02

6.68E+01
1.41E+03
8.84E+02
4.85E+02
2.74E+02
6.49E+01
3.19E+03
1.64E+01
7.76E+01
1.04E+02
1.36E+02
2.66E+02
8.69E+01
6.87E+02

6.25E+01
9.30E+02
6.10E+02
2.34E+02
2.56E+02
7.38E-I-01
2.17E+03
NOTE: Isomer concentrations shown are corrected to 3% oxygen.
not detected (detection limit in parentheses).
ND
ng
1536 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the quality
       assurance requirements for all four labelled compounds.  Problems with
       method efficiency resulted from contamination present in the sample
       extracts and corresponding difficulties in achieving acceptable
       chromatographic separations.  The reported analytical results may
       actually represent lower bounds on the true values.  See Section 7.3.1
       for more details.
                                     A. 2-1.7

-------
Dioxin/Furan
    Isomer
                        TABLE A.2.lib.   SITE 11  (DBR-A)/OUTLET
                        DIOXIN/FURAN EMISSION CONCENTRATIONS
                            (Values corrected to 3% 02)
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
NOTE: Isomer concentrations shown are corrected to 334 oxygen.

ND  «  not detected (detection limit in parentheses).
ng  -  1.0E-09g
1536 operating hours per year
                                                                  Avg.  -
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

5.59E-02
1.58E+00
1.45E+00
1.32E+00
2.37E+00
1.32E+00
8.11E+00

1.58E+00
1.67E+01
9.82E+00
4.88E+00
3.36E+00
9.23E-01
3.73E+01

2.79E-02
8.80E-01
2.79E-01
4.19E-01
7.68E-01
7.68E-01
3.14E+00

5.59E-01
1.51E+01
4.54E+00
2.23E+00
1.47E+00
4.19E-01
»
2.43E+01

6.10E-02
1.10E+00
4.27E-01
6.41E-01
7.94E-01
6.71E-01
3.69E+00

5.49E-01
1.12E+01
4.24E+00
1.86E+00
1.22E+00
3.05E-01
1.94E+01
-
5.16E-02
1.19E+00
7.19E-01
7.93E-01
1.31E+00
9.19E-01
4.98E+00

8.97E-01
1.43E+01
6.20E+00
2.99E+00
2.02E+00
5.49E-01
2.70E+01
                                     A.2-18

-------
                     TABLE A.2.12a.  SITE 12  (SSI-C)/INLET
                     DIOXIN/FURAN EMISSION CONCENTRATIONS
                         (Values corrected to 3% 02)
  Dioxin/Furan
      Isomer
 Isomer Concentration in Flue Gas
        (ng/dscm @ 3% oxygen)
01          Run 02          Run 03
                                                                    Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
NR
NR
NR
NR
NR
NR
NR

NR
NR
NR
NR
NR
NR
NR
NR
3.98E+00
ND( 2.25E+00)
8.14E+00
6.16E+01
6.76E+01
1.41E+02

4.71E+01
1..14E+02
1.03E+02
1.17E+01
1.12E+02
1.12E+02
5.00E+02
NR
1.29E+01
7.37E-01
1.01E+01
3.00E+01
3.35E+01
8.72E+01

1.20E+02
1.79E+02
1.61E+02
1.51E+01
1.76E+01
2.02E+01
5.13E+02
NR
8.44E+00
3.69E-01
9.12E+00
4.58E+01
5.06E+01
1.14E+02

8.36E+01
1.47E+02
1.32E+02
1.34E+01
6.48E+01
6.61E+01
5.07E+02
                                      corrected to 3% oxygen'
ng  =  l°OE-09e°ted (detection 11mit 1n Parentheses).
8760 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the oualitv
       mPtShnHnC?/e?UlrementS for a11 four ^belled compounds   Problems with
       method efficiency resulted from contamination present  n the samole
       extracts and corresponding difficulties in  achieving acceptable
       chromatographic separations.   The reported
                                     A.2-19

-------
                        TABLE A.2.12b.  SITE 12 (SSI-C)/OUTLET
                        DIOXIN/FURAN EMISSION CONCENTRATIONS
                            (Values corrected to 3% 02)
 Dioxin/Furan
     Isomer
     Isomer Concentration in Flue Gas
            (ng/dscm @ 3% oxygen)
Run 01          Run 02          Run 03
                                                                   Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
. 1.55E-01
8.37E+00
8.86E-01
5.87E+00
8.64E+00
7.09E+00
3.10E+01

6.76E+01
1.50E+02
1.12E+02
1.85E+01
8.64E+00
3.99E+00
3.60E+02
1.30E-01
8.75E+00
1.11E+00
8.79E+00
4.09E+01
2.76E+01
8.72E+01

- 5.16E+01
1.66E+02
1.08E+02
4.83E+01
1.32E+021
1.03E+02
6.08E+02
1.42E-01
7.23E+00
1.21E+00
6.31E+00
1.39E+01
1.11E+01
3.98E+01

4.38E+01
1.33E+02
9.56E+01
2.93E+01
3.91E+01
2.88E+01
3.70E+02
1.42E-01
8.12E+00
1.07E+00
6.99E+00
2.11E+01
1.52E+01
5.27E+01

5.43E+01
1.50E+02
1.05E+02
3.20E+01
6.00E+01
4.52E+01
4.46E+02
NOTE: Isomer concentrations shown are corrected to 3% oxygen.

ND  •*  not detected (detection limit in parentheses).
ng  »  l.'OE-09g
8760 operating hours per year
                                     A. 2-20

-------
                 APPENDIX A.3

Summary of PCDD/PCDF Emissions Per Unit of Feed
          (ug emitted per kg of feed)

-------

-------
                       TABLE A.3.1.   SITE 1 (SSI-A)/OUTLET
                          DIOXIN/FURAN EMISSION FACTORS
                           (Dry Solids Feed Rate Basis)
  Dioxin/Furan
      Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                         Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.09E-04
7.00E-02
8.71E-04
ND( 1.67E-03)
1.56E-02
3.71E-02
1.24E-01 .

NR
2.02E-01
6.55E-02
ND( 3.89E-03)
2.56E-03
ND( 2.33E-03)
2.70E-01
3.09E-04
5.48E-02
1.42E-03
4.22E-03
1.03E-02
1.86E-02
8.97E-02

NR
1.51E-01
4.35E-02
ND( 2.47E-03)
2.65E-03
•1.24E-03
1.99E-01
1.49E-04
4.55E-02
ND( 1.01E-03)
3.63E-03
1.37E-02
3.23E-02
9.53E-02

MR
ni\
1.65E-01
5.01E-02
2.41E-03
2.71E-03
ND( 1.36E-03)
2.20E-01
2.22E-04
5.68E-02
7.64E-04
2.62E-03
1.32E-02
2.94E-02
1.03E-01

MD
INK
1.73E-01
5.30E-02
8.03E-04
2.64E-03
4.12E-04
2.30E-01
NR  =  not reported by Troika.
ND  =  not detected (detection limit in parentheses).
ug  =  1.0E-06g
6000 operating hours per year
                                  A.3-1

-------
                        TABLE A.3.2.  SITE 2 (ISW-A)/OUTLET
                           DIOXIN/FURAN EMISSION FACTORS
                              (Total Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 03          Run 04
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
5.69E-03
9.39E-02
1.42E-01
2.79E-01
3.33E-01
1.37E-01
9.90E-01

3.13E-02
9.58E-01
9.98E-01
1.59E+00
8.79E-01
1.51E-01
4.61E+00
2.12E-02
3.75E-01
4.69E-01
5.66E-01
7.23E-01
2.54E-01
2.41E+00

9.38E-02
2.09E+00
2.06E+00
1.90E+00
1.29E+00
2.63E-01
7.70E+00
1.18E-02
1.97E-01
2.68E-01
3.60E-01
7.39E-01
1.33E-01
1.71E+00

5.21E-02
1.59E+00
1.75E+00
1.65E+00
1.41E+00
1.45E-01
6.60E+00
1.29E-02
2.22E-01
2.93E-01
4.01E-01
5.98E-01
1.74E-01
1.70E+00

5.91E-02
1.55E+00
1.60E+00
1.71E+00
1.19E+00
1.86E-01
6.30E+00
ND  -  not detected (detection limit in parentheses)
ug  »  1.0E-06g
2200 operating hours per year
                                  A. 3-2

-------
  Dioxin/Furan
      Isomer
                        TABLE A.3.3.  SITE 3  (SSI-B)/OUTLET
                           DIOXIN/FURAN EMISSION FACTORS
                             (Dry Solids Feed  Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                         Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 1.77E-04)
2.96E-03
ND( 2.96E-04)
ND( 2.84E-03)
1.77E-03
4.73E-03
9.47E-03

1.12E-02
1.15E-01
3.19E-02
1.30E-02
ND( 3.67E-03)
5.92E-04
1.72E-01

ND 6.59E-04)
ND 6.59E-04)
ND 4.94E-04)
ND 1.12E-03)
ND( 7.41E-04)
1.92E-03
1.92E-03

2.74E-03
1.67E-02
ND( 4.94E-04)
ND( 6.31E-04)
ND( 7.96E-04)
ND( 1.37E-04)
1.95E-02

ND( 1.75E-04)
4.38E-04
ND( 8.77E-04)
ND( 1.31E-04)
ND( 6.14E-04)
2.63E-03
3.07E-03

6.14E-03
4.08E-02
1.10E-02
ND( 3.24E-03)
ND( 3.07E-04)
ND( 2.63E-04)
5.79E-02

.OOE+00
1.13E-03
.OOE+00
.OOE+00
5.92E-04
3.09E-03
4.82E-03

6.71E-03
5.76E-02
1.43E-02
4.34E-03
.OOE+00
1.97E-04
8.32E-02
ND  =  not detected (detection limit in parentheses).
ug  =  1.0E-06g                                    '
8760 operating hours per year
                                  A.3-3

-------
                        TABLE A.3.4a.   SITE 4 (BLB-AJ/INLET
                           DIOXIN/FURAN EMISSION FACTORS
                            (Dry Solids Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD'
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 8.28E-05)
ND( 8.28E-05)
ND( 3.21E-04)
5.18E-04
1.24E-03
3.42E-03
5.18E-03

ND( 6.21E-05)
1.04E-03
1.35E-03
1.86E-03
1.04E-03
3.11E-04
5.59E-03
ND( 8.18E-05)
8.18E-05
ND( 9.81E-05)
3.27E-04
8.18E-04
2.62E-03
3.84E-03

ND( 3.27E-04)
8.18E-04
9.81E-04
1.64E-03
4.91E-04
3.27E-04
4.25E-03
ND( 2.80E-04)
ND( 2.80E-04)
ND( 3.73E-04)
ND( 1.46E-03)
2.18E-03
7.62E-03
9.80E-03

ND( 6.22E-04)
1.55E-03
ND( 2.30E-03)
2.95E-03
1.40E-03
ND( 9.80E-04)
5.91E-03
.OOE+00
2.73E-05
.OOE+00
2.82E-04
1.41E-03
4.55E-03
6.27E-03

.OOE+00
1.14E-03
7.76E-04
2.15E-03
9.75E-04
2.13E-04
5.25E-03
ND  »  not detected (detection limit in parentheses).
ug  »  1.0E-06g
8760 operating hours per year
                                  A. 3-4

-------
  Dioxin/Furan
      Isomer
                       TABLE A.3.4b.  SITE 4  (BLB-A)/OUTLET
                          DIOXIN/FURAN EMISSION FACTORS
                            (Dry Solids Feed Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                         Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
«
ND( 2.52E-04)
ND( 2.52E-04)
ND( 1.06E-04)
6.51E-04
1.30E-03
3.01E-03
4.96E-03

ND( 3.26E-05)
1.38E-03
6.51E-04
5.70E-04
8.14E-04
3.26E-04
3.74E-03

ND( 7.06E-05)
2.12E-04
ND( 8.47E-05)
ND( 2.26E-04)
3.53E-04
1.55E-03
2.12E-03

ND( 1.41E-04)
2.12E-04
ND( 1.62E-04)
2.12E-04
7.06E-04
1.06E-03
2.19E-03

ND( 7.54E-05)
ND 7.54E-05)
ND 1.51E-04)
ND 1.64E-04)
3.43E-04
8.91E-04
1.23E-03

ND( 6.85E-05)
1.37E-04
ND( 8.91E-05)
ND( 3.49E-04)
2.06E-04
6.85E-05
4.11E-04

.OOE+00
7.06E-05
.OOE+00
2.17E-04
6.66E-04
1.82E-03
2.77E-03

.OOE+00
5.77E-04
2.17E-04
2.60E-04
5.75E-04
4.84E-04
2.11E-03
ND  -  not detected (detection limit in parentheses).
ug  =  1.0E-06g                                    '
8760 operating hours per year
                                  A. 3-5

-------
                       TABLE A.3.5a.   SITE 5 (BLB-B)/INLET
                          DIOXIN/FURAN EMISSION  FACTORS
                           (Dry Solids Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( 1.53E-04)
ND( 1.53E-04)
ND( 1.53E-04)
ND( 1.53E-04)
6.81E-04
2.72E-03
3.40E-03

1.70E-04
1.70E-04
ND( 4.09E-04) '
3.40E-04
1.70E-04
1.70E-04
1.02E-03
ND( 2.47E-04)
ND( 2.47E-04)
ND( 2.74E-05)
ND( 3.43E-04)
2.88E-03
9.19E-03
1.21E-02

ND( 2.47E-04)
ND( 2.47E-04)
ND( 1.37E-04)
4.11E-04
1.37E-03
9.60E-04
2.74E-03
ND( 9.60E-05)
ND( 9.60E-05)
ND( 3.43E-04)
8.23E-04
2.69E-02
1.35E-01
1.63E-01

2.74E-04
1.78E-03
ND( 3.43E-04)
9.60E-04
2.47E-03
1.78E-03
7.27E-03
.OOE+00
.OOE+00
.OOE+00
2.74E-04
1.01E-02
4.90E-02
5.94E-02

1.48E-04
6.51E-04
.OOE+00
5.70E-04
1.34E-03
9.71E-04
3.68E-03
ND  »  not detected (detection limit in parentheses).
ug  »  1.0E-06g
8760 operating hours per year
                                  A. 3-6

-------
 Dioxin/Furan
     Isomer
                       TABLE A.3.5D.   SITE  5  (BLB-B)/OUTLET
                          DIOXIN/FURAN EMISSION  FACTORS
                           (Dry Solids Feed Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
ND  =  not detected (detection limit in parentheses).
ug  =  1.0E-06g
8760 operating hours per year
                                                                        Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

ND( 1
ND( 1
ND( 1
7
1
4
6

ND( 1
1
ND( 1
2
1
2
5

.19E-04)
.19E-04)
.79E-04)
.46E-04
.34E-03
.18E-03
.27E-03

.49E-04)
.04E-03
.09E-03)
.83E-03
.04E-03
.98E-04
.22E-03

ND(
ND{
ND(
ND(


ND(
ND(
ND(
ND(


4
4
1
3
6
2
2

1
1
1
5
2
1
4

.58E-05)
.58E-05)
.28E-04)
.30E-04)
.42E-04
.11E-03
.75E-03

.56E-04
.56E-04
.28E-04
.32E-04
.75E-04
.83E-04
.58E-04

ND 1
ND 1
ND 5
2
6
2
3

9
2
ND( 5
5
3
1
1

.02E-04)
.02E-04)
.56E-04)
.78E-04
.49E-04
.22E-03
.15E-03

.27E-05
.78E-04
.56E-04)
.56E-04
.71E-04
.85E-04
.48E-03

3
8
2
4

3
4
1
5
2
2

.OOE+00
.OOE+00
.OOE+00
.41E-04
.78E-04
.84E-03
.06E-03

.09E-05
.41E-04
.OOE+00
.13E-03
.63E-04
.22E-04
.39E-03
                                  A.3-7

-------
  Dioxin/Furan
      Isomer
                          TABLE  A.3.6a   SITE 6 (WRI-A)/OUTLET
                                      (Wire  only)
                             DIOXIN/FURAN  EMISSION  FACTORS
                                (Total  Feed  Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 06
                                                                         Avg.
  DIOXINS

2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
*
NR
1.05E-03
NR
6.32E-03
6.50E-02
3.15E-02
1.04E-01

NR
7.94E-03
1.08E-02
2.50E-02
9.68E-02
5.51E-02
1.96E-01

NR
NR
NR
NR
5.49E-02
3.15E-02
8.64E-02

NR
2.46E-02
NR
8.11E-03
. 1.56E-01
7.45E-02
2.63E-01

2.83E-04
3.77E-03
6.23E-03
2.69E-02
4.24E-01
3.84E-01
8.45E-01

1.13E-03
4.96E-02
8.10E-02
1.85E-01
7.74E-01
3.04E-01
1.39E+00

2.83E-04
2.41E-03
6.23E-03
1.66E-02
1.81E-01
1.49E-01
3.56E-01

1.13E-03
2.74E-02
4.59E-02
7.26E-02
3.42E-01
1.44E-01
6.33E-01
NR  -  not reported by Troika.
ND  -  not detected (detection limit in parentheses).
ug  =*  1.0E-06g
2080 operating hours per year
NOTE:  Several recoveries for the MM5 train samples did not meet the quality
       assurance requirements for all four labelled compounds.  Problems with
       method efficiency resulted from contamination.present in the sample
       extracts and corresponding difficulties in achieving acceptable
       cnromatographic separations.  The reported analytical results may
       actually represent lower bounds on the true values.  See Section 7 3 1
       for more details.
                                   A. 3-8

-------
  Dloxin/Furan
      Isomer
 TABLE A.3.65  SITE 6 (WRI-A)/OUTLET
       (Wire and Transformers)
    DIOXIN/FURAN EMISSION FACTORS
	JJotal_ Fee_d_ Rate_ Bas i s)


       Dioxin/Furan  Emission  Factors  (ug/kg)

     Run 03          Run 04          Run 05
                                                                         Avg.
  OIOXINS


  2378 TCDD
  Other TCDD
  Penta-CDD
  Hexa-CDD
  Hepta-CDD
  Octa-CDD

  Total  PCDD

  FURANS
   1.16E-04
   5.21E-04
   9.68E-03
   1.12E-01
     73E-OI
   2.75E+00

   3.65E+00
   NR
 3.28E-03
   NR
   NRE+00)
 1.53E-01
 1.55E-01

 3.12E-01
 6.31E-04
 8.36E-03
 1.77E-02
 3.01E-02
   13E-01
1.01E-01

2.71E-01
              3
              4,
              1.
              7,
              3.
              1.
   74E-04
   05E-03
   37E-02
   11E-02
   47E-01
   OOE+00
                                                1.44E+00
  2378 TCDF
  Other TCDF
  Penta-CDF
  Hexa-CDF
  Hepta-CDF
  Octa-CDF

  Total PCDF
   9.27E-04
   5.88E-02
   1.24E-01
   4.00E-01
   8.76E-01
   1.83E+00

   3.29E+00
  NR
1.34E-01
  NR
9.38E-02
6.56E-01
3.39E-01

1.22E+00
4.
1.
7.
1,
  -42E-03
  .18E-01
  .07E-02
  •15E-01
3.38E-01
2.52E-01
                                                        8.97E-01
2.67E-03
1.04E-01
9.74E-02
2.03E-01
6.23E-01
8.07E-01

1.84E+00
NR  -  not reported by Troika.

ug  I  KE^r^ (detect10n I1rait 1n
2080 operating hours per year

       a"SraJcerere0qVS^enf?s for M  %P,SX}1 did
                                   A.3-9

-------
 Dioxin/Furan
     Isomer
                     TABLE A.3.7a.  SITE 7  (WFB-A)/INLET
                         DIOXIN/FURAN EMISSION FACTORS
                           (Wet Wood Feed Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
NR  -  not reported by Troika.
ND  -  not detected (detection limit in parentheses)
ug  «  1.0E-06g
8760 operating hours per year
                                                                        Avg.
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD'
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Pehta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF

NR
6.04E-02
NR
6.21E-02
1.15E-01
1.11E-01
3.48E-01

NR
3.76E-01
5.94E-02
4.11E-02
1.66E-02
ND( 4.00E-02)
4.93E-01

4.39E-03
1.03E-01
1.21E-01
1.21E-01
9.62E-02
2.33E-02
4.70E-01

2.74E-02
5.13E-01
1.96E-01
6.99E-02
3.01E-02
3.04E-03
8.40E-01

3.75E-03
1.46E-01
1.37E-01
1.39E-01
7.80E-02
2.04E-02
5.24E-01

2.72E-02
5.60E-01
1.98E-01
8.02E-02
2.62E-02
2.72E-03
8.94E-01

4.07E-03
1.03E-01
1.29E-01
1.07E-01
9.63E-02
5.15E-02
4.91E-01

2.73E-02
4.83E-01
1.S1E-01
6.37E-02
2.43E-02
1.92E-03
7.52E-01
                                    A.3-10

-------
                       TABLE  A.3.7b.   SITE  7  (WFB-A)/OUTLET
                           DIOXIN/FURAN  EMISSION  FACTORS
                             (Wet  Wood Feed Rate  Basis)
  Dioxin/Furan
      Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                         Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
7.60E-04
1.76E-01
2.24E-01
2.30E-01
1.25E-01
3.57E-02
7.92E-01

4.94E-03
1.25E-01
8.59E-02
5.40E-02
2.28E-02
4.56E-03
2.97E-01
1.08E-03
2.55E-01
2.33E-01
2.38E-01
1.80E-01
4.54E-02
9.52E-01

7.21E-03
2.07E-01
1.22E-01
8.18E-02
3.77E-02
3.06E-03
4.59E-01
1.12E-03
9.05E-02
7.26E-02 •
7.44E-02
1.13E-01
3.84E-02
3.90E-01

6.71E-03
8.00E-02
4.77E-02
1.31E-02
1.19E-02
2.24E-03
1.62E-01
9.87E-04
1.74E-01
1.77E-01
1.81E-01
1.39E-01
3.98E-02
7.11E-01

6.29E-03
1.37E-01
8.52E-02
4.96E-02
2.41E-02
3.29E-03
3.06E-01
NO  =  not detected (detection limit in parentheses)
ug  =  1.0E-06g                                    '
8760 operating hours per year
                                A.3-11

-------
                       TABLE A.3.8a.   SITE 8 (BLB-CJ/INLET
                          DIOXIN/FURAN EMISSION FACTORS
                          (Dry Solids Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND( .OOE+00)
4.68E-03
5.94E-03
9.18E-03
8.10E-03
5.40E-03
3.33E-02

ND( .OOE+00)
3.56E-02
3.40E-02
2.52E-02
6.66E-03
9.00E-04
1.02E-01
ND( .OOE+00)
2.02E-03
2.98E-03
6.35E-03
4.91E-03
4.04E-03
2.03E-02

ND( .OOE+00)
3.94E-03
2.31E-03
3.17E-03
1.15E-03
5.77E-04
1..12E-02
ND( .OOE+00)
2.29E-03
2.67E-03
6.49E-03
4.58E-03
4.01E-03
2.00E-02

ND( .OOE+00)
4.20E-03
ND( 2.48E-03)
3.05E-03
9.54E-04
ND( 5.73E-04)
8.21E-03
.OOE+00
3.00E-03
3.87E-03.
7.34E-03
5.86E-03
4.48E-03
2.45E-02

.OOE+00
1.46E-02
1.21E-02
1.05E-02
2.92E-03
4.92E-04
4.06E-02
ND  =*  not detected (detection limit in parentheses).
ug  »  1.0E-06g
8760 operating hours per year
                                  A.3-12

-------
  Dioxin/Furan
      Isomer
                          TABLE  A.3.85.   SITE 8  (BLB-O/OUTLET
                             DIOXIN/FURAN EMISSION FACTORS
                             (Dry Solids  Feed Rate Basis)
Dioxin/Furan Emission Factors (ug/kg)

   °1          Run 02          Run 03
                                                                         Avg.
  DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND = not detec
ug = 1.0E-06g
ND( 6.33E-05)
3.17E-04
ND( 1.58E-04)
4.75E-04
7.92E-04
1.90E-03
3.48E-03

ND( 9.50E-05)
7.92E-04
ND( 3.17E-04)
7.92E-04
6.33E-04
3.17E-04
2.53E-03
ted (detection limit

ND( 3.19E-05)
8.78E-04
1.76E-03
3.35E-03
9.73E-03
1.91E-02
3.49E-02

1.28E-04
2.11E-03
1.76E-03
1.84E-03
2.71E-03
3.19E-04
8.86E-03
in parentheses

ND( 1.76E-05)
4.40E-04
7.04E-04
1.67E-03
3.35E-03
3.17E-03
9.33E-03

ND( 1.76E-05)
2.38E-03
3.35E-03
3.61E-03
2.64E-03
3.52E-04
1.23E-02

' *
.OOE+00
5.45E-04
8.20E-04
1.83E-03
4.62E-03
8.07E-03
1.59E-02

4.26E-05
1.76E-03
1.70E-03
2.08E-03
2.00E-03
3.29E-04
7.91E-03


8760 operating hours per year
                                  A.3-13

-------
                        TABLE A.3.9a.  SITE 9  (CRF-A)/INLET
                           DIOXIN/FURAN  EMISSION  FACTORS
                           (Bare Carbon  Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
7.93E-04
1.43E-02
3.13E-02
9.36E-02
1.23E-01
• 1.42E-01
4.06E-01

1.39E-02
1.89E-01
1.36E-01
1.61E-01
1.13E-01
5.19E-02
6.65E-01
6.17E-04
1.08E-02
5.87E-03
3.61E-02
7.81E-02
1.33E-01
2.64E-01

1.45E-02
2.15E-01
1.17E-01
1.53E-01
1.39E-01
6.08E-02
6.99E-01
1.25E-03
4.34E-02
3.53E-02
5.12E-02
5.58E-02
2.99E-02
2.17E-01

1.31E-02
1.97E-01
1.76E-01
1.60E-01
1.26E-01
4.87E-02
7.21E-01
8.86E-04
2.28E-02
2.41E-02
6.03E-02
8.58E-02
1.02E-01
2.96E-01

1.38E-02
2.00E-01
1.43E-01
1.58E-01
1.26E-01
5.38E-02
6.95E-01
ND  -  not detected (detection limit in parentheses)
ug  -  1.0E-06g
8760 operating hours per year
                                  A.3-14

-------
   Dioxin/Furan
       Isomer
                           TABLE A.3.9b.   SITE 9 (CRF-A)/OUTLFT
                              DIOXIN/FURAN EMISSION FACTORS
                              (Bare Carbon Feed Rate Basis)
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02
ug  =  1.0E-06g
8760 operating hours per year
DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
ND - not det

ND( 4.18E-03)
1.37E-02
8.96E-03
1.43E-02
1.08E-02
9.56E-03
5.74E-02

ND( 3.59E-03)
1.25E-02
3.59E-03
6.57E-03
7.77E-03
5.38E-03
3.59E-02
ected {detection Mm

ND( 9.84E-04)
ND( 1.48E-03)
2.46E-03
5.90E-03
6.39E-03
5.41E-03
2.02E-02

ND( 2.95E-03)
9.84E-03
5.41E-03
3.94E-03
3.94E-03
2.46E-03
2.56E-02
T 4* 4 •* a* •* u — .^ L. 	 	

ND( 1.03E-03)
1.54E-03
ND( 2.57E-03)
4.62E-03
5.13E-03
5.13E-03
1.64E-02

ND( 2.05E-03)
7.19E-03
ND( 2.57E-03)
4.11E-03
3.59E-03
5.65E-03
2.05E-02
_ *
Mvg.
.OOE+00
5.09E-03
3.81E-03
8.29E-03
7.43E-03
6.70E-03
3.13E-02

.OOE+00
9.86E-03
3.00E-03
4.87E-03
5.10E-03
4.49E-03
2.73E-02

                                 A.3-15

-------
                       TABLE A.3.10.  SITE 10 (MET-AJ/OUTLET
                          DIOXIN/FURAN EMISSION FACTORS
                           (Coke-Free Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 02          Run 03          Run 04
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.22E-01
4.46E-01
7.94E-01
1.84E+00
1.38E+00
8.23E-01
5.51E+00

2.35E+00
7.94E+00
7.07E+00
6.45E+00
2.28E+00
1.55E+00
2.76E+01
9.05E-02
7.47E-01
1.24E+00
1.15E+00
3.12E+00
1.97E+00
8.32E+00

2.70E+00
9.48E+00
9.88E+00
3.24E+00
5.67E+00
3.83E+00
3.48E+01
6.76E-02
6.33E-01
8.73E-01
7.92E-01
1.74E+00
1.24E+00
5.35E+00

3.11E+00
1.56E+01
8.96E+00
3.03E+00
2.31E+00
2.16E+00
- 3.51E+01
1.27E-01
6.09E-01
9.70E-01
1.26E+00
2.08E-t-00
1.35E+00
6.39E+00

2.72E+00
1.10E+01
8.64E+00
4.24E+00
3.42E+00
2.52E+00
3.25E+01
Note: emission factor is based on feed rate to cupola furnace  (coke-free basis)
ND  -  not detected (detection limit in parentheses).
ug  »  1.0E-06g
kg -   1.0E+03g
8160 operating hours per year
NOTE:  Surrogate recoveries could not be determined for Run 02 and 04
       dioxin/furan samples because of the relatively large quantities of
       native CDD and CDF .species present in the samples.  The reported
       analytical results may actually represent lower bounds on the true
       values.  See Section 7.3.1 for more details.
                                   A.3-16

-------
                       TABLE A.3.11a.   SITE 11  (DBR-A)/INLET
                          DIOXIN/FURAN EMISSION FACTORS
                             (Drum Feed Rate Basis)
 Dioxin/Furan
     Isomer
                             Dioxin/Furan  Emission  Factors  (ug/kgj

                            Run  01           RUn  02           Run  03
                                                                        Avg.
 DIOXINS


 2378 TCDD
 Other TCDD
 Penta-CDD
 Hexa-CDD
 Hepta-CDD
 Octa-CDD

 Total  PCDD

 FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF

Total  PCDF
                         2.90E-01
                         l.OOE+00
                         1.52E+00
                         1.48E+00
                         8.31E+00
                         3.09E+00

                         1.57E+01
                         8.42E-01
                         1.13E+01
                         9.85E+00
                         2.53E+00
                         7.71E+00
                         2.48E+00

                         3.48E+01
ugD  ^  l?OE?06j°ted (detect1on
1536 operating hours per year
 2.15E-01
 7.39E-01
 6.43E-01
 4.18E-01
 1.73E-01
 1.46E-01

 2.33E+00
8.18E-01
8.11E+00
  12E+00
  58E-01
  16E-01
  29E-02
                                       1.42E+01
                                    in parentheses)
 1.20E-01
 9.92E-01
 1.50E+00
   59E+00
   56E+00
 5.61E-01

 8.32E+00
6.09E-01
1.29E+01
8.06E+00
4.41E+00
2.50E+00
5.91E-01

2.90E+01
 2.09E-01
 9.12E-01
 1.22E+00
 1.49E+00
 3.68E+00
 1.26E+00

 8.78E+00
7.56E-01
1.08E+01
7
2
3
1.
34E+00
57E+00
51E+00
05E+00
                                                                    2.60E+01
                                        .
                                A.3-17

-------
                    TABLE A.3.lib.  SITE 11 (DBR-A)/OUTLET
                        DIOXIN/FURAN EMISSION FACTORS
                           (Drum Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD *
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
2.68E-03
6.44E-02
5.90E-02
5.37E-02
9.66E-02
5.37E-02
3.30E-01

6.44E-02
6.79E-01
4.00E-01
1.99E-01
1.37E-01
3.76E-02
1.52E+00
1.18E-03
3.72E-02
1.18E-02
1.77E-02
3.25E-02
3.25E-02
1.33E-01

2.36E-02
6.40E-01
1.92E-01
9.46E-02
6.21E-02
1.77E-02
1.03E+00
2.39E-03
4.30E-02
1.67E-02
2.51E-02
3.11E-02
2.63E-02
1.45E-01

2.15E-02
4.40E-01
1.66E-01
7.29E-02
4.78E-02
1.19E-02
7.60E-01
2.09E-03
4.82E-02
2.92E-02
3.22E-02
5.34E-02
3.75E-02
2.03E-01

3.65E-02
5.86E-01
2.53E-01
1.22E-01
8.22E-02
2.24E-02
1.10E+00
ND  «  not detected (detection limit in parentheses)
ug  =  1.0E-06g
1536 operating hours per year
                                     A.3-18

-------
   Dioxin/Furan
       Isomer
                         TABLE A.3.12a.  SITE 12 (SSI-O/INLET
                             DIOXIN/FURAN EMISSION FACTORS
                              (Dry Solids Feed Rate Basis)
                            Dioxin/Furan  Emission Factors  (ug/kgj

                           Run 01          Run 02          Run 03
                                                                         Avg.
  DIOXINS


  2378 TCDD
  Other TCDD
  Penta-CDD
  Hexa-CDD
  Hepta-CDD
  Octa-CDD

  Total  PCDD

  FURANS
                           NR
                           NR
                           NR
                           NR
                           NR
                           NR

                           NR
ND(
  NR
3.27E-02
1.85E-02)
6.68E-02
5.06E-01
5.55E-01

1.16E+00
 NR
•34E-02
.19E-03
 77E-02
 70E-01
                    1.90E-01

                    4.96E-01
 NR
.31E-02
•10E-03
,23E-a2
-38E-01
 73E-01
                                                                      8.29E-01
  2378  TCDF
  Other TCDF
  Penta-CDF
  Hexa-CDF
  Hepta-CDF
  Octa-CDF

  Total PCDF
                          NR
                          NR
                          NR
                          NR
                          NR
                          NR

                          NR
NR
    3.87E-01
    9.39E-01
    8.44E-01
    9.60E-02
    9.19E-01
    9.22E-01

    4.11E+00
       not reported by Troika.

                    (d6tect1on 11mit 1n Parentheses)
                6.84E-01
                1.02E+00
                9.16E-01
                8.60E-02
                l.OOE-01
                1.15E-01

                2.92E+00
           5.-36E-01
           9.80E-01
           8.80E-01
           9.10E-02
           5.10E-01
           5.19E-01

           3.52E+00
8760 operating hours per year
                                  A.3-19

-------
                      TABLE A.3.12b.  SITE 12 (SSI-C)/OUTLET
                           DIOXIN/FURAN EMISSION FACTORS
                           (Dry Solids Feed Rate Basis)
 Dioxin/Furan
     Isomer
 Dioxin/Furan Emission Factors (ug/kg)

Run 01          Run 02          Run 03
                                                                        Avg.
 DIOXINS
2378 TCDD
Other TCDD
Penta-CDD
Hexa-CDD
Hepta-CDD
Octa-CDD
Total PCDD
FURANS
2378 TCDF
Other TCDF
Penta-CDF
Hexa-CDF
Hepta-CDF
Octa-CDF
Total PCDF
8.12E-04
4.39E-02
4.64E-03
3.08E-02
4.53E-02
3.71E-02
1.62E-01

3.54E-01
7.86E-01
5.85E-01
9.69E-02
4.53E-02
2.09E-02
1.89E+00
9.31E-04
6.29E-02
7.98E-03
6.32E-02
2.94E-01
1.98E-01
6.27E-01

3.71E-01
1.19E+00
7.73E-01
3.47E-01
9.50E-01
7.38E-01
4.37E+00
1.07E-03
5.44E-02
9.06E-03
4.74E-02
1.04E-01
8.32E-02
3.00E-01

3.29E-01
l.OOE+00
7.19E-01
2.20E-01
2.94E-01
2.16E-01
2.78E+00
9.37E-04
5.37E-02
7.23E-03
4.71E-02
1.48E-01
1.06E-01
3.63E-01

3.52E-01
9.94E-01
6.92E-01
2.21E-01
• 4.30E-01
3.25E-01
3.01E+00
ND  =  not detected (detection limit in parentheses)
ug  »  1.0E-06g
8760 operating hours per year
                                      A. 3-20

-------
                     APPENDIX A.4
Error Analysis:  Control  Device Efficiency Calculations

-------

-------
                                   APPENDIX A.4
              ERROR ANALYSIS: CONTROL DEVICE EFFICIENCY CALCULATIONS
 Objective:
       Let:
                   the analytical uncertainty of the CDD/CDF analyses
                   . accuracy), estimate the uncertainty of the *•««*«»••
             Cout,meas = Jhe measured concentration of a given CDD/CDF
                         homologue at the outlet location.

                         the measured concentration of a given CDD/CDF
                         homologue at the inlet location.
              in.meas
             "out,max
             "out,min
              in,max
             in,min
                         the maximum possible concentration of the CDD/
                         CDF homologue given the measured value C
                                                                 out,meas
                         the minimum possible concentration of the CDD/
                         CDF homologue given the measured value C
                                                                 out,meas*
                         the maximum possible concentration of the CDD/
                         CDF homologue,  given the measured value C
                                                                  in,meas*
                         the minimum possible concentration of the CDD/
                         CDF homologue,  given the measured value C
                                                                  in,meas'
            E = the removal efficiency of the control device
Assuming + 50 percent analytical accuracy:
            Cmin - cmeas ' °*5 Sneas = °'5 Cmeas
            Cmax = cmeas + °'5 Cmeas " 1<5 Cmeas
Note that:  EmaY = Cin.max " £»..*,..<• = 1 - cout.min
                         v»i _	             P t
                                              in,max
            "max
                          in,max
             max
                            in,meas
                                                   .  E
                                                      rneas-
                                    A.4-1

-------
and:
             min
       c       -  c
        in.min     out.max

             in,min



       1 -     out.meas
                       0.5 C
                            in,meas
               out.max

               in,min
         - 3
                            - Emeas>
Emin * 3 Emeas
                            " 2
Now,
Emin
positive control (i.e., emissions

reduction across the control device)
             (3Emeas  '  2>
                    Em
         ieas
Therefore,  if E _ao is larger than 66.7 percent,  the true removal  efficiency
                UlcaS

can  safely  be assumed to be greater than zero.
 And,
              ,nax
             V, + V
                    negative control  (i.e., emissions

                    increase across the control device)
                     3  meas
                < 0
                        meas




 Therefore,  if Emooo is less than -200 percent, the true efficiency can safely
                [flcaS

 be assumed  to be less than zero.
 To summarize:
             Emeas > 66'7 percent
                                       positive  control
             -200 < Emo.c < 66.7 percent
             Emeas < 20° percent
                                       no  definitive  conclusions

                                       can be drawn


                                       negative control
                                     A.4-2

-------
TABLE A. 4  VALUES OF Emax and
           FOR VARIOUS MEASURED CONTROL EFFICIENCIES
Control
me as
100
95
90
85
80
75
50
25
0
-25
-50
-100
-200
Device Efficiency ('
max
100
98.3
96.7
95.0
93.4
91.7
83.4
75.0
66.7
. 58.4
50.0
33.4
0
n
Emin
100
85
70
55
40
25
-50
-125
-200
-275
-350
-500
-800
 max
Emin ' 3Emeas - 20°
         A.4-3

-------

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
 REPORT NO.
    EPA-450/4-84-014H
                            2.
                                                          3. RECIPIENT'S ACCESSION NO.
 TITLE AND SUBTITLE
 National Dioxin  Study Tier 4
 Engineering Analysis  Report
 - Combustion Sources
5. REPORT DATE
  September 1987
                            6. PERFORMING ORGANIZATION CODE
 AUTHOR(S)
 Andrew J. Miles,  Martha H.  Keating, and
 Carol L. Jamgochian
                            8. PERFORMING ORGANIZATION REPORT NO.


                                87-203-061-07-09
 PERFORMING ORGANIZATION NAME AND ADDRESS
                                                          10. PROGRAM ELEMENT NO.
 Radian Corporation
 P.O. Box 13000
 Research Triangle Park,  NC
                             11. CONTRACT/GRANT NO.
27709
                                                             68-02-3889
2. SPONSORING AGENCY NAME AND ADDRESS
 Environmental  Protection Agency
 Office of Air  Quality Planning and Standards
 Monitoring and Data  Analysis Division
 Research Triangle  Park,  NC  27711
                             13. TYPE OF REPORT AND PERIOD COVERED
                               Final
                             14. SPONSORING AGENCY CODE
5. SUPPLEMENTARY NOTES
 EPA Project Officer:   William B. Kuykendal
6. ABSTRACT
     This report  summarizes the complete results  of Tier 4 (combustion sources) of  the
National Dioxin Study.   The purpose of the Tier 4 study was to address the following
questions:  Do combustion sources emit dioxin?  If so,  how much?  Are these emissions
significant?  A secondary objective was to attempt to determine what combustion para-
meters affect dioxin.emissions and to determine the effectiveness of conventional
control devices for controlling dioxin emissions.
     The report presents the results of a literature review containing 249 references,
the results of a  stack  sampling program, and the  results of an ash sampling program.
The stack sampling  program produced valid data from 12  sites covering 8 combustion
source categories.   Data are presented for dioxin (tetra through octa homologue +
2378-TCDD) and furan (tetra through octa homologue +2378-TCDF) emissions as well as
combustion conditions.   Where possible, data were obtained before and after control
devices.  Ash samples were collected from 74 sites covering 22 combustion source
categories.  Dioxin and furan data are presented  for the ash data.  Various data
correlations are  presented.
     Dioxins and  furans were detected in the stack emissions for all 12 sites tested.
These emissions varied  over 5 orders of magnitude.   About one half of the ash samples
contained dioxins and furans.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                                                           COS AT I Field/Group
  Air Emissions
  Combustion Sources
  Di oxi n
  Furans
  2,3,7,8 Tetrachlorodibenzo-p-dioxin
                Air  Pollution Emissions
                Data
18. DISTRIBUTION STATEMENT
   Release  Unlimited
                19. SECURITY CLASS (ThisReport)
                       Unclassified
                                                                              381
                2O. SECURITY CLASS (Tillspage)

                       Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (R«v. 4-77)   PREVIOUS EDITION is OBSOLETE

-------

-------
-------



















VJ
ft

i

g
s
1


£
g


•g
U



B

UJ
g

1
8
5
•


Ul
Ul
9












^S<






^
0




 ••-
333 O O O * L.


^•J=N£^£ 1 1 1 1 OSS
N O«4 ^ I I 1 1 ro *••"  ^ & f*» ^ ^
o o o a, u
•w ** ** *-
£: J= £
*• -M *»
L. t- (-
S S m u^ ^«
O^O^O^ O«t-O'o"
r^^aco??? aj 8 «o ao o fl
•o *o "o *•*• Ul U Ul U
3 *>» *S •—
t*. U ^ C
9 e eg o
N JC O
• Ul C • «*
o jz f 3 . f—
Ul Ul U»+* tj ^ *- UlUfl
o a »a •.«. ^- 13 —
?C £C ?*5 *0 0 S &J3 0
T3 +J 9 +1 *O > "* M VI •Q^^-M
« » o m • r» ui *• ^ ui o>^ v> 01^ ui w^ vi CT^» ui o> ui CT o *- *.
o) o 3 ao oo o o o o o 5 TJ o
JCVIO OV13 O Ul 3 O VI 3 O V) 3 O »+J 03^
f« eo r- u y ?» -^ AO s
* ov ^ o»  u a B e -^ m i- •-*
«— t_ O— '*-> O>»'«' U « — » ^ oi «— *-
o t- o» i- oi-vi swco urn 4>r*» -^ O m u ^- t.
v O <"— ' O 3^ Of «d Q£ Qt UJ O Ot UJ >Cn A Ql A9~O
(ff • * *^ U""* U *— 3 O p— fl
L r- S .— U r- L. fl OO O-M O9 0-M •*-• Xr- U
•S *^3 ^^ — ' % -S fro® &«« ^u^ ?°2 2 ^+* 5
1 *9c *3C *= c^i tn o S"o 00*0 S -a ^ o 3 "g 00*00 =
c 3C »-« £*-< a i-t*- • L o* — i GO oom QOOOD ^ aim ^H^OM
^- ^
<£ a
i «. i
xi a ui
= < CO 0 =< 03 0 = < <
c;i i ^. ^i i i |i i
v> vi tn — -3 -V -5 au. vt
Svicnco S m m cj s i-i
^
M (J T!
1 4 §
vi] ea a
5-32