EPA-453/R-96-012
Technical Support Document:
CHEMICAL RECOVERY COMBUSTION
SOURCES AT KRAFT AND SODA
PULP MILLS
Emission Standards Division
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
October 1996
-------
This report has been reviewed by the Emission Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication. Mention of trade names or commercial
products is not intended to constitute endorsement or
recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U. S. Environmental
Protection Agency, Research Triangle Ps.rk, NC 27711.
(919) 541-2777, or from National Technical Information Services,
5285 Port Royal Road, Springfield, VA 22161, (703) 487-4650.
11
-------
TABLE OF CONTENTS
1.0 INTRODUCTION 1-1
1.1 SCOPE OF THE TECHNICAL SUPPORT DOCUMENT .... l-l
1.2 ORGANIZATION OF THE TECHNICAL SUPPORT
DOCUMENT 1-1
2.0 COMBUSTION PROCESSES IN THE KRAFT PULP INDUSTRY . . 2-1
2.1 OVERVIEW OF CHEMICAL RECOVERY PROCESSES .... 2-1
2.2 COMBUSTION PROCESSES AND EQUIPMENT 2-9
2.2.1 Recovery Furnaces 2-9
2.2.2 Smelt Dissolving Tanks 2-21
2.2.3 Black Liquor Oxidation Systems 2-24
2.2.4 Lime Kilns 2-28
2.3 BASELINE EMISSIONS 2-34
2.3.1 Federal and State Regulations Affecting
Kraft Pulp Mill Combustion Sources ... 2-34
2.3.2 Baseline Emission Estimates 2-36
2.4 REFERENCES FOR CHAPTER 2 2-44
3;. 0 EMISSION CONTROL TECHNIQUES , 3-1
3.1 ADD-ON CONTROLS , 3-1
3.1.1 Electrostatic Precipitators 3-1
3.1.2 Wet Scrubbers 3-18
3.1.3 BLO Control 3-43
3.2 EQUIPMENT CHANGES/MODIFICATIONS 3-45
3.2.1 .Elimination of Black Liquor used in
NDCE Recovery Furnace ESP Control
Systems 3-45
3.2.2 Conversion from a DCE Recovery
Furnace System to an NDCE Recovery
Furnace .......... 3-50
3.3 REFERENCES FOR CHAPTER 3 3-60
4.0 MODEL PROCESS UNITS, CONTROL OPTIONS, AND ENHANCED
MONITORING OPTIONS 4-1
4.1 MODEL PROCESS UNITS 4-1
4.1.1 Recovery Furnace Models 4-1
4.1.2 Smelt Dissolving Tank Models ...... 4-11
4.1.3 Black Liquor Oxidation Unit Models ... 4-14
4.1.4 Lime Kiln Models 4-17
4.2 CONTROL OPTIONS 4-21
4.2.1 Recovery Furnace Control Options .... 4-21
4.2.2 SDT Control Options 4-28
4.2.3 BLO Unit Control Option 4-34
4.2.4 Lime Kiln Control Options 4-34
4.3 ENHANCED MONITORING OPTIONS 4-38
4.3.1 Recovery Furnace Enhanced Monitoring . . 4-41
4.3.2 Smelt Dissolving Tank Enhanced
Monitoring . ". 4-43
-------
TABLE OF CONTENTS (continued)
5.0
6.0
4.3.3 Black Liquor Oxidation Unit Enhanced
Monitoring
4.3.4 Lime Kiln Enhanced Monitoring . . .
4.4 REFERENCES FOR CHAPTER 4
5.2
,1.2
,1.3
,1.4
,1.5
.2.2
.2.3
MODEL PROCESS UNIT ENVIRONMENTAL AND ENERGY
IMPACTS
5.1 GENERAL APPROACH
5.1.1 Air Pollution Impacts
Energy Impacts
Water Pollution Impacts
Solid Waste Disposal Impacts ....
Other Impacts
RECOVERY FURNACE CONTROL'OPTIONS
5.2.1 PM Controls
Wet to Dry ESP System Conversion . .
Conversion of a DCE Recovery Furnace
System to an NDCE Recovery Furnace .
Addition of Packed-Bed Scrubber . .
BLACK LIQUOR OXIDATION UNIT CONTROL OPTION
5.3.1 Air Pollution Impacts
Energy Impacts
Water Pollution Impacts
Solid Waste Disposal Impacts ....
Other Impacts
SMELT DISSOLVING TANK CONTROL OPTIONS . . .
5.4.1 PM Controls
LIME KILN CONTROL OPTIONS
5.5.1 PM Controls ....
REFERENCES FOR CHAPTER 5
5.2.4
5.3
5,
5,
5,
5,
3.2
3.3
3.4
3.5
5.4
5.5
5.6
MODEL PROCESS UNIT CONTROL AND ENHANCED MONITORING
COSTS
6.1 CONTROL OPTION COSTS
6.1.1 General Costing Approach .......
Recovery Furnace Control Options . . .
Black Liquor Oxidation Unit Control
Options
Smelt Dissolving Tank Control Options
Lime Kiln Control Options
ENHANCED MONITORING COSTS
6.2.1 Recovery Furnace Enhanced
Monitoring
6.2.2 Black Liquor Oxidation Unit Enhanced
Monitoring
6.
6,
6
6
1.2
1.3
1.4
1.5
6.2
4-44
4-44
4-46
5-1
5-1
5-2
5-5
5-7
5-8
5-8
5-9
5-9
5-14
5-16
5-23
5-28
5-28
5-29
5-30
5-30
5-30
5-31
5-31
5-35
5-36
5-98
6-1
6-1
6-2
6-7
6-40
6-43
6-47
6-52
6-53
6-58
VI
-------
6.3
TABLE OF CONTENTS (continued)
6.2.3 Smelt Dissolving Tank Enhanced
Monitoring
6.2.4 Lime Kiln Enhanced Monitoring
REFERENCES FOR CHAPTER 6
Page
6-59
6-60
6-127
APPENDIX A.
APPENDIX B.
EVOLUTION OF THE TECHNICAL SUPPORT DOCUMENT
EMISSION MEASUREMENT AND CONTINUOUS MONITORING
FOR PULP AND PAPER COMBUSTION SOURCES
VI1
-------
LIST OF FIGURES
Figure 2-1
Figure 2-2
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10a
2-1 Ob
2-lla
2-lib
2-12
2-13
2-14
2-13
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
3-12
Figure 3-13
Distribution of U.S. kraft and soda pulp
mills •• •
Relationship of the chemical recovery
cycle to the pulping and product forming
processes
Kraft process--chemical recovery area
(DCE recovery furnace)
Soda process--chemical recovery area
(DCE recovery furnace)
Schematic of NDCE recovery furnace and
associated equipment
Schematic of DCE recovery furnace and
associated equipment , • •
Cascade design evaporator
Cyclone design evaporator .......
Recovery furnace zones and air stages
DCE recovery furnace age distribution
NDCE recovery furnace age distribution .
DCE recovery furnace size distribution .
NDCE recovery furnace size distribution
Smelt dissolving tank and wet scrubber .
Two-stage air-sparging black liquor
oxidation system .
Schematic of a lime kiln used at kraft
pulp mills
Schematic of a lime kiln used at kraft
pulp mills
Rigid-electrode ESP
PM emission data for recovery furnace
No. 3 at Mill A . .
PM emission data for recovery furnace
No. 4 at Mill B
PM emission data for recovery furnace
No. 3 at Mill B
PM emission data for lime kiln No. 5 at
Mill C
PM emission data for lime kiln No. 1 at
Mill D
Acid gas control system configuration on
recovery furnace exhaust gases
Schematic of a counterflow" packed-bed
scrubber
Schematic of a cross-flow packed-bed
scrubber
Schematic of a venturi scrubber ....
PM emission data for lime kiln at Mill A
PM emission data for lime kiln No. 4 at
Mill B
PM emission data for SDT No. 3 at Mill B
2-2
2-4
2-5
2-6
2-11
2-12
2-13
2-14
2-16
2-18
2-18
2-20
2-20
2-22
2-26
2-30
2-29
3-5
3-10
3-11
3-12
3-17
3-19
3-21
3-22
3-23
3-34
3-38
3-39
3-40
vixi
-------
LIST OF FIGURES (continued)
Page
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
3
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
-14
-15
-16
-17
-18
-19
-la
-Ib
-2a
-2b
-3
-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
PM emission data for SDT No. 4 at Mill B .
PM emission data for SDT No. 22 at -Mill C
Methanol emissions for NDCE recovery
furnaces
Methanol emissions for DCE recovery
furnace systems and NDCE recovery
furnaces ....
Emission sources for a DCE recovery
furnace system
Emission sources for an NDCE recovery
furnace
Size distribution for DCE recovery
furnaces
Size distribution for NDCE recovery
furnaces ....
DCE recovery furnace model size, ranges . .
NDCE recovery furnace model size ranges
Size distribution for lime kilns
Particulate matter emissions from model
NDCE recovery furnaces
Particulate matter emissions from model
DCE recovery furnaces
Gaseous organic HAP emissions from model
recovery furnaces
Total reduced sulfur emissions from
model recovery furnace systems
Hydrochloric acid emissions from model
recovery furnaces
Sulfur dioxide emissions from model
recovery furnaces
Gaseous organic HAP emissions from model
BLO units . .
Total reduced sulfur emissions from
model BLO units
Particulate matter emissions from model
smelt dissolving tanks
Particulate matter emissions from model
lime kilns
3-41
3-42
3-51
3-56
3-58
3-59
4-6
4-6
4-7
4-7
4-19
5-41
5-42
5-43
5-44
5-45
5-46
5-47
5-48
5-49
5-50
IX
-------
LIST OF TABLES
TABLE 2-1
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
2-2
2-3
2-4
2-5a
2-5b
2-6
3-1
3-2
3-3
TABLE 3-4
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
3-5
3-6
3-7
4-la
4-lb
4-2a
4-2b
4-3a
4-3b
4-4a
4-4b
4-5
4-6a
TABLE 4-6b
BLACK LIQUOR OXIDATION SYSTEMS: EQUIPMENT
TYPES
CHEMICAL COMPOSITION OF LIME MUD FEED . . .
NSPS AND EMISSION GUIDELINES FOR KRAFT PULP
MILL COMBUSTION SOURCES
HAP EMISSION FACTORS
(METRIC). NATIONWIDE BASELINE GASEOUS HAP
EMISSION ESTIMATES
(English). NATIONWIDE BASELINE GASEOUS HAP
EMISSION ESTIMATES
NATIONWIDE BASELINE PM, PM HAP, AND TRS
EMISSIONS
PARTICULATE MATTER HAP CONTROL TECHNIQUES .
GASEOUS HAP CONTROL TECHNIQUES ......
SUMMARY OF DATA ON KRAFT RECOVERY FURNACE
COMBUSTION GAS CHARACTERISTICS
SUMMARY OF ACID GAS SCRUBBER PERFORMANCE
DATA
HAP EMISSIONS FOR BLO UNITS •
HAP EMISSIONS FOR NDCE RECOVERY FURNACES
HAP EMISSIONS FOR DCE RECOVERY FURNACE
SYSTEMS AND NDCE RECOVERY FURNACES ....
(METRIC). RECOVERY FURNACE MODEL PROCESS
UNITS AND PROCESS PARAMETERS
(ENGLISH). RECOVERY FURNACE MODEL PROCESS
UNITS AND PROCESS PARAMETERS
(METRIC). SMELT DISSOLVING TANK MODEL
PROCESS UNITS AND PROCESS PARAMETERS . . .
(ENGLISH). SMELT DISSOLVING TANK MODEL
PROCESS UNITS AND PROCESS PARAMETERS . . .
(METRIC). BLACK LIQUOR OXIDATION UNIT
MODEL PROCESS UNITS AND PROCESS PARAMETERS
(ENGLISH). BLACK LIQUOR OXIDATION UNIT
MODEL PROCESS UNITS AND PROCESS PARAMETERS
(METRIC). LIME KILN MODEL PROCESS UNITS
AND PROCESS PARAMETERS
(ENGLISH). LIME KILN MODEL PROCESS UNITS
AND PROCESS PARAMETERS
RECOVERY FURNACE CONTROL OPTIONS
(METRIC). RECOVERY FURNACE MODELS:
LOW-ODOR CONVERSION CONTROL OPTION
(INCLUDES WET TO DRY ESP SYSTEM CONVERSION
AND PM CONTROL TO NSPS LEVEL)
(ENGLISH). RECOVERY FURNACE MODELS:
LOW-ODOR CONVERSION CONTROL OPTION
(INCLUDES WET TO DRY ESP SYSTEM CONVERSION
AND PM CONTROL TO NSPS LEVEL)
Page
2-28
2-31
2-35
2-38
2-43
2-43
2-44
3-2
3-2
3-28
3-30
3-46
3-49
3-55
4-2
4-3
4-12
4-12
4-15
4-15
4-18
4-18
4-22
4-24
4-24
-------
LIST OF TABLES (continued)
TABLE 4-7a
TABLE 4-7b
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
4-
4-
4-
4-
4-
4-
4-
4-
4-
8a
8b
9a
9b
lOa
lOb
lla
lib
12a
TABLE 4-12b
TABLE 4-13a
TABLE 4-13b
TABLE 4-14a
TABLE 4-14b
TABLE 4-15a
TABLE 4-15b
TABLE 4-16a
TABLE 4-16b
(METRIC). RECOVERY FURNACE MODELS: LOW-
ODOR CONVERSION CONTROL OPTION (INCLUDING
WET TO DRY ESP SYSTEM CONVERSION AND PM
CONTROL TO 0.034 G/DSCM) 4-25
(ENGLISH). RECOVERY FURNACE MODELS: LOW-
ODOR CONVERSION CONTROL OPTION (INCLUDING
WET TO DRY ESP SYSTEM CONVERSION AND PM
CONTROL TO 0.015 GR/DSCF) 4-25
(METRIC). RECOVERY FURNACE MODELS: WET TO
DRY ESP SYSTEM CONVERSION CONTROL OPTION . 4-26
(ENGLISH). RECOVERY FURNACE MODELS:
WET TO DRY ESP SYSTEM CONVERSION CONTROL
.OPTION 4-26
(METRIC). RECOVERY FURNACE MODELS: PM
CONTROL OPTIONS (0.10 G/DSCM) 4-27
(ENGLISH). RECOVERY FURNACE MODELS: PM
CONTROL OPTIONS (0.044 GR/DSCF) 4-27
(METRIC). RECOVERY FURNACE MODELS: PM
CONTROL OPTIONS (0.034 G/DSCM) 4-29
(ENGLISH). RECOVERY FURNACE MODELS: PM
CONTROL OPTIONS (0.015 GR/DSCF) 4-30
(METRIC). RECOVERY FURNACE MODELS: HC1
CONTROL OPTION (PACKED-BED SCRUBBER) ... 4-31
(ENGLISH). RECOVERY FURNACE MODELS: HC1
CONTROL OPTION (PACKED-BED SCRUBBER) . . . 4-31
(METRIC). RECOVERY FURNACE MODELS: HC1
CONTROL OPTION (PACKED-BED SCRUBBER AFTER
LOW-ODOR CONVERSION) , 4-32
(ENGLISH). RECOVERY FURNACE MODELS: HC1
CONTROL OPTION (PACKED-BED SCRUBBER AFTER
LOW-ODOR CONVERSION) . 4-32
(METRIC). SMELT DISSOLVING TANK MODELS:
PM CONTROL OPTIONS(0.10 KG/MG BLS) .... 4-33
(ENGLISH). SMELT DISSOLVING TANK MODELS:
PM CONTROL OPTIONS (0.20 LB/TON BLS) ... 4-33
(METRIC). SMELT DISSOLVING TANK MODELS:
PM CONTROL OPTIONS (0.06 KG/MG BLS) .... 4-35
(ENGLISH). SMELT DISSOLVING TANK MODELS:
PM CONTROL OPTIONS (0.12 LB/TON BLS) ... 4-35
(METRIC). BLACK LIQUOR OXIDATION UNIT
MODELS: METHANOL CONTROL OPTION
(INCINERATION) 4-36
(ENGLISH). BLACK LIQUOR OXIDATION UNIT
MODELS: METHANOL CONTROL OPTION
(INCINERATION) 4-36
(METRIC). LIME KILN MODELS: PM CONTROL
OPTIONS (0.15 G/DSCM) . 4-37
(ENGLISH). LIME KILN MODELS: PM CONTROL
OPTIONS (0.067 GR/DSCF) 4-37
-------
LIST OF TABLES (continued)
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
4-17a
4-17b
4-18
5-la
5-lb
5-2a
5-2b
5-3a
5-3b
5-4a
5-4b
5-5a
5-5b
5-6a
5-6b
5-7
5-8a
5-8b
5-9a
5-9b
5-10a
MODEL NDCE RECOVERY FURNACE
MODEL DCE RECOVERY FURNACE
TABLE 5-10b
TABLE 5-lla
(METRIC). LIME KILN MODELS: PM CONTROL
OPTIONS (0.023 G/DSCM) .........
(ENGLISH). LIME KILN MODELS: PM CONTROL
OPTIONS (0.010 GR/DSCF) . . . .
ENHANCED MONITORING OPTIONS
(METRIC). MODEL NDCE RECOVERY FURNACE
PARAMETERS
(ENGLISH).
PARAMETERS
(METRIC). MODEL DCE RECOVERY FURNACE
PARAMETERS
(ENGLISH).
PARAMETERS
(METRIC). MODEL NDCE RECOVERY FURNACE
CONCENTRATIONS AND EMISSION FACTORS ....
(ENGLISH). MODEL NDCE RECOVERY FURNACE
CONCENTRATIONS AND EMISSION FACTORS
(METRIC). MODEL DCE RECOVERY FURNACE
CONCENTRATIONS AND EMISSION FACTORS ....
(ENGLISH). MODEL DCE RECOVERY FURNACE
CONCENTRATIONS AND EMISSION FACTORS ....
(METRIC). PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL RECOVERY FURNACES . .
(ENGLISH). PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL RECOVERY FURNACES
(METRIC). SECONDARY EMISSIONS FOR PM
CONTROL FOR MODEL RECOVERY FURNACES ....
(ENGLISH). SECONDARY EMISSIONS FOR PM
CONTROL FOR MODEL RECOVERY FURNACES ....
ENERGY IMPACTS FOR PM CONTROL FOR MODEL
RECOVERY FURNACES
(METRIC). PRIMARY GASEOUS ORGANIC HAP
EMISSIONS FOR MODEL RECOVERY FURNACES . . .
(ENGLISH). PRIMARY GASEOUS ORGANIC HAP
EMISSIONS FOR MODEL RECOVERY FURNACES . . .
(METRIC). TOTAL REDUCED SULFUR COMPOUND
EMISSIONS FOR MODEL RECOVERY FURNACES . . .
(ENGLISH). TOTAL REDUCED SULFUR COMPOUND
EMISSIONS FOR MODEL RECOVERY FURNACES . . .
(METRIC). SECONDARY EMISSIONS FOR
LOW-ODOR CONVERSION FOR MODEL DCE RECOVERY
FURNACES
(ENGLISH). SECONDARY EMISSIONS FOR
LOW-ODOR CONVERSION FOR MODEL DCE RECOVERY
FURNACES
(METRIC). ENERGY IMPACTS FOR LOW-ODOR
CONVERSION FOR MODEL DCE RECOVERY
FURNACES
Page
4-39
4-39
4-40
5-51
5-52
5-53
5-54
5-55
5-56
5-57
5-58
5-59
5-60
5-61
5-62
5-63
5-64
5-65
5-66
5-67
5-68
5-69
5-70
XI1
-------
LIST OF TABLES (continued)
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5
TABLE 5'
-lib
•12a
•12b
•13a
•13b
•14
•15a
•15b
•16a
•,16b
•17a
TABLE 5-17b
TABLE 5-18a
TABLE 5-18b
TABLE 5-19
TABLE 5-20a
TABLE 5-20b
TABLE 5-21a
TABLE 5-21b
TABLE 5-22a
TABLE 5-22b
TABLE 5-23a
(ENGLISH). ENERGY IMPACTS FOR LOW-ODOR
CONVERSION FOR MODEL DCE RECOVERY FURNACES
(METRIC). PRIMARY ACID GAS EMISSIONS FOR
MODEL RECOVERY FURNACES
(ENGLISH). PRIMARY ACID GAS EMISSIONS FOR
MODEL RECOVERY FURNACES
(METRIC). SECONDARY EMISSIONS FOR HCL
CONTROL FOR MODEL RECOVERY FURNACES . . . .
(ENGLISH). SECONDARY EMISSIONS FOR HCL
CONTROL FOR MODEL RECOVERY FURNACES . . . .
ENERGY IMPACTS FOR HCL CONTROL FOR MODEL
RECOVERY FURNACES
(METRIC). WASTEWATER IMPACTS FOR HCL
CONTROL FOR MODEL RECOVERY FURNACES . . . .
(ENGLISH). WASTEWATER IMPACTS FOR HCL
CONTROL FOR MODEL RECOVERY FURNACES . . . .
(METRIC). MODEL BLACK LIQUOR OXIDATION
UNIT PARAMETERS AND EMISSION FACTORS . . .
(ENGLISH). MODEL BLACK LIQUOR OXIDATION
UNIT PARAMETERS AND EMISSION FACTORS . . .
(METRIC). PRIMARY GASEOUS ORGANIC HAP
EMISSIONS FOR MODEL BLACK LIQUOR OXIDATION
UNITS
(ENGLISH). PRIMARY GASEOUS ORGANIC HAP
EMISSIONS FOR MODEL BLACK LIQUOR OXIDATION
UNITS
(METRIC). SECONDARY EMISSIONS FOR MODEL
BLACK LIQUOR OXIDATION UNITS
(ENGLISH). SECONDARY EMISSIONS FOR MODEL
BLACK LIQUOR OXIDATION UNITS .......
ENERGY IMPACTS FOR MODEL BLACK LIQUOR
OXIDATION UNITS
(METRIC). TOTAL REDUCED SULFUR COMPOUND
EMISSIONS FOR MODEL BLACK LIQUOR OXIDATION
UNITS
(ENGLISH). TOTAL REDUCED SULFUR COMPOUND
EMISSIONS FOR MODEL BLACK LIQUOR OXIDATION
UNITS
(METRIC). MODEL SMELT DISSOLVING TANK
PARAMETERS AND PM EMISSION FACTORS . . . .
(ENGLISH). MODEL SMELT DISSOLVING TANK
PARAMETERS AND PM EMISSION FACTORS . . . .
(METRIC)
PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL SMELT DISSOLVING TANKS
(ENGLISH). PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL SMELT DISSOLVING TANKS
(METRIC). SECONDARY EMISSIONS FOR MODEL
SMELT DISSOLVING TANKS
Page
5-71
5-72
5-73
5-74
5-75
5-76
5-77
5-78
5-79
5-79
5-80
5-80
5-81
5-81
5-82
5-83
5-83
5-84
5-85
5-86
5-87
5-88
Xlll
-------
LIST OF TABLES (continued)
TABLE 5-23b (ENGLISH). SECONDARY EMISSIONS FOR MODEL
SMELT DISSOLVING TANKS ' - 5-88
TABLE 5-24 ENERGY REQUIREMENTS FOR MODEL SMELT
DISSOLVING TANKS 5-89
TABLE 5-25b (ENGLISH) . MODEL LIME KILN PARAMETERS AND
PM CONCENTRATIONS 5-91
TABLE 5-263. (METRIC) . PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL LIME KILNS • • 5-92
TABL3 5-26b (ENGLISH) . PRIMARY PM AND PM HAP EMISSIONS
FOR MODEL LIME KILNS 5-93
TABLE 5-27a (METRIC) . SECONDARY EMISSIONS FOR MODEL
LIME KILNS 5-94
TABLE 5-27b (ENGLISH) . SECONDARY EMISSIONS FOR MODEL
LIME KILNS 5-95
TABLE 5-28 ENERGY IMPACTS FOR MODEL LIME KILNS .... 5-96
TABLE 5-29a (METRIC) . WASTEWATER IMPACTS FOR MODEL
LIME KILNS 5-97
TABLE 5-29b (ENGLISH) . WASTEWATER IMPACTS FOR MODEL
LIME KILNS 5-97
TABLE 6-la (METRIC) . COSTS OF ESP REPLACEMENT TO
CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-62
TABLE 6-2a (METRIC) . COSTS OF ESP REPLACEMENT TO
CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-64
TABLE 6-2b (ENGLISH) . COSTS OF ESP REPLACEMENT TO
CONTROL PM TO 0.044 GR/DSCF FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-65
TABLE 6-3 SUMMARY OF ASSUMPTIONS USED IN ESP UPGRADE
COSTS 6-66
TABLE 6-4a (METRIC). COSTS OF SCHEDULE 1 ESP UPGRADE
TO CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-67
TABLE 6-4b (ENGLISH). COSTS OF SCHEDULE 1 ESP UPGRADE
TO CONTROL PM TO 0.044 GR/DSCF FOR MODEL
RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-68
XIV
-------
LIST OF TABLES (continued)
TABLE 6-5a (METRIC). COSTS OF SCHEDULE 2 ESP UPGRADE
TO CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-69
TABLE 6-5b (ENGLISH). COSTS OF SCHEDULE 2 ESP UPGRADE
TO CONTROL PM TO 0.044 GR/DSCF FOR MODEL
RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-70
TABLE 6-6a (METRIC). COSTS OF SCHEDULE 1 ESP UPGRADE
TO CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-71
TABLE 6-6b (ENGLISH). COSTS OF SCHEDULE 1 ESP UPGRADE
TO CONTROL PM TO 0.044 GR/DSCF FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-72
TABLE 6-7a (METRIC). COSTS OF SCHEDULE 2 ESP UPGRADE
TO CONTROL PM TO 0.10 G/DSCM FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-73
TABLE 6-7b (ENGLISH). COSTS OF SCHEDULE 2 ESP UPGRADE
TO CONTROL PM TO 0.044 GR/DSCF FOR MODEL
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-74
TABLE 6-8a (METRIC). COSTS OF ESP UPGRADE TO CONTROL
PM FROM 0.10 G/DSCM TO 0.034 G/DSCM FOR
MODEL RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-75
TABLE '6-8b (ENGLISH). COSTS OF ESP UPGRADE TO CONTROL
PM FROM 0.044 GR/DSCF TO 0.015 GR/DSCF FOR
MODEL RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) 6-76
TABLE 6-9a (METRIC). COSTS OF ESP UPGRADE TO CONTROL
PM FROM 0.10 G/DSCM TO 0.034 G/DSCM FOR
MODEL RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-77
TABLE 6-9b (ENGLISH). COSTS OF ESP UPGRADE TO CONTROL
PM FROM 0.044 GR/DSCF TO 0.015 GR/DSCF FOR
MODEL RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) 6-78
TABLE 6-10a (METRIC). WET- TO DRY-BOTTOM ESP
CONVERSION COSTS FOR MODEL NDCE RECOVERY
FURNACES 6-79
TABLE 6-10b (ENGLISH). WET- TO DRY-BOTTOM ESP
CONVERSION COSTS FOR MODEL NDCE RECOVERY
FURNACES 6-80
xv
-------
LIST OF TABLES (continued)
TABLE 6-lla
TABLE 6-lib
TABLE 6-12a
TABLE 6-12b
TABLE 6-13a
TABLE 6-13b
TABLE 6-14a
TABLE 6-14b
TABLE 6-15a
TABLE 6-15b
TABLE 6-16
(METRIC). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.10 G/DSCM) FOR MODEL
DCE RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES)
(ENGLISH). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.044 GR/DSCF) FOR MODEL
DCE RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES)
(METRIC). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.10 G/DSCM) FOR MODEL DCE
RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES)
(ENGLISH). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.044 GR/DSCF) FOR MODEL
DCE RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES) . . .
(METRIC). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.034 G/DSCM) FOR MODEL
DCE RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES)
(ENGLISH). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.015 GR/DSCF) FOR MODEL
DCE RECOVERY FURNACES (EXCLUDING PULP
PRODUCTION LOSSES) .
(METRIC). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.034 G/DSCM) FOR MODEL
DCE RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES)
(ENGLISH). CAPITAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.015 GR/DSCF) FOR MODEL
DCE RECOVERY FURNACES (INCLUDING PULP
PRODUCTION LOSSES)
(METRIC). MODEL DCE RECOVERY FURNACE/ESP
DESIGN PARAMETERS •
(ENGLISH). MODEL DCE RECOVERY FURNACE/ESP
DESIGN PARAMETERS
SCENARIO 1: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.10 G/DSCM [0.044 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (EXCLUDING
ANNUALIZED PULP PRODUCTION LOSSES) . . . .
6-81
6-82
6-83
6-84
6-85
6-86
6-87
6-88
6-89
6-90
6-91
xvi
-------
LIST OF TABLES (continued)
TABLE 6-17
TABLE 6-18
TABLE 6-19
TABLE 6-20
TABLE 6-21
TABLE 6-22a
TABLE 6-22b
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
6-23a
6-23b
6-24a
6-24b
6-25
6-26
6-27a
6-27b
6-28
SCENARIO 2: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.10 G/DSCM [0.044 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (INCLUDING
BLEACHED PULP PRODUCTION LOSSES) 6-92
SCENARIO 3: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.10 G/DSCM [0.044 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (INCLUDING
UNBLEACHED PULP PRODUCTION LOSSES) .... 6-93
SCENARIO 1: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.034 G/DSCM [0.015 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (EXCLUDING
ANNUALIZED PULP PRODUCTION LOSSES) .... 6-94
SCENARIO 2: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.034 G/DSCM [0.015 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (INCLUDING
BLEACHED PULP PRODUCTION LOSSES) 6-95
SCENARIO 3: ANNUAL COSTS OF LOW-ODOR
CONVERSION (INCLUDING ESP UPGRADE TO
CONTROL PM TO 0.034 G/DSCM [0.015 GR/DSCF])
FOR MODEL DCE RECOVERY FURNACES (INCLUDING
UNBLEACHED PULP PRODUCTION LOSSES) .... 6-96
(METRIC). GAS AND LIQUID STREAM PARAMETERS
FOR RECOVERY FURNACE MODEL PROCESS UNITS . 6-97
(ENGLISH). GAS AND LIQUID STREAM
PARAMETERS FOR RECOVERY FURNACE MODEL
PROCESS UNITS 6-98
(METRIC). PACKED-BED SCRUBBER DESIGN AND
OPERATING PARAMETERS 6-99
(ENGLISH). PACKED-BED SCRUBBER DESIGN
AND OPERATING PARAMETERS 6-100
(METRIC). UNIT COSTS FOR PACKED-BED
SCRUBBER 6-101
(ENGLISH). UNIT COSTS FOR PACKED-BED
SCRUBBER 6-101
PACKED-BED SCRUBBER CAPITAL COSTS FOR
MODEL RECOVRERY FURNACES 6-102
PACKED-BED SCRUBBER ANNUAL COSTS FOR
MODEL RECOVERY FURANCES 6-103
(ENGLISH). MODEL BLACK LIQUOR OXIDATION
UNIT DESIGN PARAMETERS 6-104
(ENGLISH). MODEL BLACK LIQUOR OXIDATION
UNIT DESIGN PARAMETERS 6-105
CAPITAL AND ANNUAL COSTS OF COLLECTION
AND INCINERATION OF BLO VENT GASES FOR
MODEL BLACK LIQUOR OXIDATION UNITS .... 6-106
xvi i
-------
LIST OF TABLES (continued)
TABLE 6-293.
TABLE 6-29b
TABLE 6-30a
TABLE 6-30b
TABLE 6-31
TABLE 6-32a
TABLE 6-32b
TABLE 6-33a
TABLE 6-33b
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
TABLE
6-34a
6-34b
6-35
6-36a
6-36b
6-37
6-38
6-39a
6-39b
6-40
6-41
(METRIC). SCRUBBER REPLACEMENT COSTS FOR
MODEL SDT'S
(ENGLISH). SCRUBBER REPLACEMENT COSTS FOR
MODEL SDT'S
MODEL SDT/SCRUBBER DESIGN
MODEL SDT/SCRUBBER DESIGN
(METRIC).
PARAMETERS
(ENGLISH).
PARAMETERS
CAPITAL AND ANNUAL COSTS TO REPLACE MIST
ELIMINATORS WITH SCRUBBERS FOR MODEL SDT'S
(METRIC). COSTS TO REPLACE SCRUBBERS
WITH ESP'S TO CONTROL PM TO 0.15 G/DSCM
FOR MODEL LIME KILNS
(ENGLISH). COSTS TO REPLACE SCRUBBERS WITH
ESP'S TO CONTROL PM TO 0.067 GR/DSCF FOR
MODEL LIME KILNS
(METRIC). COSTS TO REPLACE SCRUBBERS WITH
ESP'S TO CONTROL PM TO 0.023 G/DSCM FOR
MODEL LIME KILNS
(ENGLISH). COSTS TO REPLACE SCRUBBERS
WITH ESP'S TO CONTROL PM TO 0.010 GR/DSCF
FOR MODEL LIME KILNS
(METRIC). MODEL LIME KILN/SCRUBBER
DESIGN PARAMETERS
(ENGLISH). MODEL LIME KILN/SCRUBBER
DESIGN PARAMETERS
ANNUAL COSTS FOR EXISTING LIME KILN
SCRUBBERS
(METRIC). MODEL LIME KILN/ESP DESIGN
PARAMETERS .
(ENGLISH). MODEL LIME KILN/ESP DESIGN
PARAMETERS
ANNUAL COSTS FOR NEW LIME KILN ESP'S
CONTROLLING PM TO 0.15 G/DSCM
(0.067 GR/DSCF)
ANNUAL COSTS FOR NEW LIME KILN ESP'S
CONTROLLING PM TO 0.023 G/DSCM
(0.010 GR/DSCF)
(METRIC). COSTS TO UPGRADE ESP'S TO
CONTROL PM TO 0.023 G/DSCM FOR MODEL LIME
KILNS
(ENGLISH). COSTS TO UPGRADE ESP'S TO
CONTROL PM TO 0.010 GR/DSCF FOR MODEL LIME
KILNS . . .
SUMMARY OF ENHANCED MONITORING COSTS . . .
OPACITY AND HCL CONTINUOUS EMISSION MONITOR
COSTS
Page
6-107
6-108
6-109
6-110
6-111
6-112
6-113
6-114
6-115
6-116
6-117
6-118
6-119
6-120
6-121
6-122
6-123
6-124
6-125
6-126
XVlll
-------
'LIST OF ACRONYMS AND UNITS OF MEASURE
a.c.
acfm
acmm
ADMBP/d
ADMP
ADMUP/d
ADTBP/d
ADTP
ADTUP/d
APCD
As
Be
BLO
BLRBAC
Btu/lb
Ca (OH) ,
CAA
CaCOo
CaO
Cd
GEM
CFR
cm
Co
CO
CO,
Cr
CRF
CS9
d/yr
d.c.
DAC
DAS
DCE
DMF
EMTIC
EPA
ESP
FID
ft9
ftVl/000 acfm
ft2
FTIR
g/L
g/dscm
gal/acf
gal/yr
GC
GFC
gpm
gr/dscf
alternating current
actual cubic foot .(feet) per minute
actual cubic meter(s) per minute
air-dried megagram(s) of bleached pulp per day-
air-dried megagram(s) ton of pulp
air-dried megagram(s) of unbleached pulp per day
air-dried ton(s) of bleached pulp per day
air-dried ton(s) of pulp
air-dried ton(s) of unbleached pulp per day
air pollution control device
arsenic
beryllium
black liquor oxidation
Black Liquor Recovery Boiler Advisory Committee
British thermal unit per pound
calcium hydroxide
Clean Air Act as amended in 1990
calcium carbonate
calcium oxide
cadmium
continuous emission monitor
Code of Federal Regulations
cent imeter(s)
cobalt
carbon monoxide
carbon dioxide
chromium
capital recovery factor
carbon disulfide
day(s) per year
direct current
direct annual cost(s)
data acquisition system
direct contact evaporator
N,N-dimethylformamide
Emission Measurement Technical Information
Center
U. S. Environmental Protection Agency
electrostatic precipitator
flame ionization detector
foot (feet)
square foot (feet) per 1,000 actual cubic feet
per minute
square foot (feet)
fourier transform infrared
gram(s) per liter
gram(s) per dry standard cubic meter
gallon(s)
gallon(s) per actual cubic foot
gallon(s) per year
gas chromatography
gas filter correlation
gallon(s) per minute
grain(s) per dry standard cubic foot
xix
-------
HoO
HAP
HC1
Hg
HOMER
hp
hr/d
HSCSST
HVLC
HWI
IAC
IMS
in.
in. H20
ITAC
KC1
kg BLS/d
kg/ADMP
kg/Mg
kg PM/MWh
kJ/kg
kW
kWh
L/min
L
L/yr
L/acm
L/G
Ib PM/MM Btu
lb BLS/d
Ib/gal
Ib/ADTP
Ib/ton
LOQ
LVHC
mo i
nc/(nr/sec)
m3
n£
rrr/sec
MACT
MEE
mg
Mg/yr
min
MJ/yr
ml
mm Hg
MM Btu/yr
MM
Mn
MSD
MST
MWI
Na2C03
rate
water
hydrogen sulfide
hazardous air pollutant
hydrochloric acid
mercury-
hazardous organic mass emission
horsepower
hour(s) per day
heated summa canister source sampling train
high-volume, low-concentration
hazardous waste incinerator
indirect annual cost(s)
ion mobility spectroscopy
inch(es)
inch(es) of water
incremental total annual cost(s)
potassium chloride
kilogram(s) of black liquor solids per day
kilogram(s) per air-dried megagram of pulp
kilogram(s) per megagram
kilogram(s) of PM per megawatt-hour
kilojoule(s) per kilogram
kilowatt(s)
kilowatt-hour(s)
liter(s) per minute
liter(s)
liter(s) per year
liter(s) per actual cubic meter
liquid-to-gas
pound(s) of PM per million Btu
'' ' of black liquor solids per day
per gallon
per air-dried ton of pulp
c per ton
limit of quantitation
low-volume, high-concentration
meter(s) ,
square meter(s) per cubic meter per second
square meter(s)
cubic meter(s)
cubic meter(s) per second
maximum available control technology
multiple-effect evaporators
milligram(s)
megagram(s) per year
minute(s)
megajoule(s) per year
milliliter(s)
millimeter(s) of mercury
million Btu per year
million(s)
manganese
mass selective detector
Methanol Sampling Train
medical waste incinerator
sodium carbonate
pound(s)
pound(s)
pound(s)
pound(s)
xx
-------
Na2S
0,
NaCl
NaOH ,
NCASI
NCG
NDCE
NESHAP
Ni
^
NSPS
NSSC
°?
OAQPS
Pb
PCDD/PCDF
PM
ppm
ppmdv
ppmv
QA/QC
RRF
Sb
SCA
SDT
Se
sec
SIE. -
S02
T-R
TAG
TCI
TDS
THC
Tl
ton/yr
TPIEC
TRS
TSD
UV
IfOC
sodium sulfide
sodium thiosulfate
sodium sulfate
sodium chloride
sodium hydroxide
National Council of the Paper Industry for
Air and Stream Improvement, Inc.
noncondensible gases
nondirect contact evaporator
national emission standards for hazardous air
pollutants
nanogram(s)
nickel
nitrogen oxides
new source performance standards
neutral sulfite semichemical
oxygen
Office of Air Quality Planning and Standards
lead
polychlorinated dibenzo-p-dioxins
and dibenzofurans
particulate matter
part(s) per billion
part(s) per million.
part(s) per million dry volume
part(s) per million by volume
quality assurance/quality control
relative response factor
antimony
specific collecting area
smelt dissolving tank
selenium
second(s)
specific ion electrodes
sulfur dioxide
transformer-rectifier
total annual costs
total capital investment
total dissolved solids
total hydrocarbon
thallium
ton(s) per year
Texas Paper Industry Environmental Committee
total reduced sulfur
technical support document
mi crometer(s)
ultraviolet
volatile organic compound
xxi
-------
-------
1.0 INTRODUCTION
National emission standards for hazardous air pollutants
(NESHAP} are under development for combustion processes in the
chemical recovery area of kraft and soda pulp mills under
authority of Section 112(d) of the Clean Air Act (CAA) as amended
in 1990. This technical -support document (TSD) provides
technical information and analyses used in the development of
this NESHAP. Effluent guideline limitations and a separate
NESHAP covering pulp and paper manufacturing processes are being
developed concurrently under the authority of the Clean Water Act
and the CAA, respectively; the technical information and analyses
for these regulations are found in separate documents. The U. S.
Environmental Protection Agency (EPA) is coordinating these
efforts to produce air and water regulations, for'the pulp and
paper industry. The remainder of this chapter describes the
scope and organization of this TSD.
1.1 SCOPE OF THE TECHNICAL SUPPORT DOCUMENT
This TSD presents the technical information and the analyses
used in the development of the NESHAP for combustion processes in
the chemical recovery area of kraft and soda pulp mills. The
chemical recovery combustion processes included in this NESHAP
are (1) recovery furnaces (including associated smelt dissolving
tanks and black liquor oxidation systems) and (2) lime kilns.
1..2 ORGANIZATION OF THE TECHNICAL SUPPORT DOCUMENT
This TSD is organized into six chapters. Chapter 2.0
describes the processes and equipment associated with the
a.*.
chemical recovery cycle at kraft and soda pulp mills, identifies
the hazardous air pollutants (HAP's) emitted and the emission
1-1
-------
sources, and quantifies baseline HAP emissions. Chapter 3.0
describes the emission control techniques that can be used to
reduce HAP emissions from chemical recovery combustion processes,
including add-on control systems and equipment changes.
Chapter 4.0 presents model plants and control and enhanced
monitoring options for chemical recovery combustion processes.
Chapter 5.0 discusses the environmental and energy impacts on the
model process units from the application of the HAP control
options presented in Chapter 4.0. Finally, Chapter 6.0 presents
the model process unit control and enhanced monitoring costs for
the application of the control options. Appendix A lists the
evolution of this technical support document. Appendix B
summarizes the available HAP emissions data and discusses the
test methods and monitoring methods that could be used to
demonstrate compliance with proposed standards for chemical
recovery combustion sources at kraft and soda pulp mills.
1-2
-------
2.0 COMBUSTION PROCESSES IN THE KRAFT PULP INDUSTRY
This chapter describes the combustion processes and
equipment associated with the chemical recovery cycle at kraft
and soda pulp mills, as well as the HAP's emitted from these
processes and equipment. The chemical recovery process includes
four sources of HAP emissions that are the focus of this
document. These sources are (1) chemical recovery furnaces
(including direct contact evaporator [DCE] recovery furnaces and
nondirect contact evaporator [NDCE] recovery furnaces), (2) smelt
dissolving tanks (SDT's), (3) black liquor oxidation (BLO)
systems, and (4) lime kilns. Section 2.1 provides an overview of
the kraft chemical recovery process. More detailed information
on the combustion processes and equipment is included in
Section 2.2. Section 2.3 presents existing Federal and State
regulations affecting kraft pulp mills, process emission factors,
and baseline HAP emissions. The references to this chapter are
provided in Section 2.4.
2.1 OVERVIEW OF CHEMICAL RECOVERY PROCESSES
This section provides an overview of the chemical recovery
cycle at kraft and soda pulp mills. References 1 and 2 listed in
Section 2.4 provide more detailed descriptions of the chemical
recovery cycle.1'
TJiere are approximately 122 kraft pulp mills currently
operating in tlie United States.3 There are only two soda mills
currently operating in the United States.4 One soda mill is
located in Pennsylvania and the other is located in Tennessee.4
Figure 2-1 shows the number of kraft and soda pulp mills located
in each State. As shown in Figure 2-1, the majority (52 percent)
of kraft pulp mills are located in the southeastern United
2-1
-------
-H
r-l
ft
•d
o
m
-tJ
IW
rt
J-i
fl
O
-H
-H
5-1
4J
CQ
-M
O
tH
i
0)
cn
-H
fa
2-2
-------
States. The kraft process is the dominant pulping process in the
United States, accounting for approximately 85 percent of all
domestic pulp productipn.5 The soda pulping process is similar
to the kraft process, except that soda pulping is a nonsulfur
process. One reason why the kraft process dominates the paper
industry is because of the ability of the kraft chemical recovery
process to recover approximately 95 percent of the pulping
chemicals and at the same time produce energy in the form of
steam. Other reasons for the dominance of the kraft process
include its ability to handle a wide variety of wood species and
the superior strength of its pulp.
The production of kraft and soda paper products from wood
can be divided into three process areas: (1) pulping of wood
chips, (2) chemical recovery/ and (3) product forming (includes
bleaching). The relationship of the chemical recovery cycle to
the pulping and product forming processes is shown in Figure 2-2.
Process flow diagrams of the chemical recovery area at kraft and
soda pulp mills are shown in Figures 2-3 and 2-4, respectively.
Kraft and soda chemical recovery processes are similar, except
that the soda process does not require BLO systems, which are
installed to reduce total reduced sulfur (TRS) emissions from DCE
recovery furnaces. Both kraft and soda mills have chemical
recovery furnaces, SDT's, and lime kilns. This document focuses
on the four primary emission sources in the chemical recovery
area. The emission sources covered in this document are the BLO
system (kraft process only), the recovery furnace, the SDT, and
tiie lime kiln. These emission sources are shaded in both process
flow diagrams. Air emissions from these sources are expected to
be similar for both kraft and soda processes, with the exception
of TRS emissions. The soda process does not utilize sulfur-
containing compounds; therefore, no TRS emissions are generated
by soda mills.
Because of the similarities between the kraft and soda
process and the predominant use of the kraft process, the
information presented in the remainder of this document focuses
on the kraft chemical recovery process.
2-3
-------
WOOD
PULPING
PULP
PRODUCT
FORMING
PAPER OR
PULP PRODUCT
WHITE LIQUOR
BLACK LIQUOR
CHEMICAL RECOVERY
Figure 2-2. Relationship of the chemical recovery cycle
to the pulping and product forming processes.
2-4
-------
LU
o
o
111
o
s
UJ
X
o
CD
-z.
D.
_i
ID
D.
(U
u
nJ
0)
O
o
Q)
CJ
Q
ft
0)
O
o
cu
u
-H
g
0)
JCt
U
m
m
0)
u
O
0)
Cn
-H
2-5
-------
/
1
1
'
I
\
\
\
\
\
\
\
EC
UJ
^>
O
o
UJ
cc
_i
g
"^
rf^l
UJ
X
0
0)
o
&
0)
o
0)
u
p
rtf
QJ
u
0)
rt
o
-H
g
(U
,fi
u
I
I
m
CO
QJ
o
o
nl
•d
o
CO
CN
0
CD
•H
HOflOn 31IHM
2-6
-------
The purpose of the chemical recovery cycle is to recover
cooking liquor chemicals from spent cooking liquor. The process
involves concentrating black liquor,^. combusting organic
compounds, reducing inorganic compounds, and reconstituting
cooking liquor.
Cooking liquor, which is referred to as "white liquor," is
an aqueous solution of sodium hydroxide (NaOH) and sodium sulfide
(Na2S) that is used in .the pulping area of the mill. In the
pulping process, white liquor is introduced with wood chips into
digesters, where the wood chips are "cooked" under pressure. The
contents of the digester are then discharged to a blow tank,
where the softened chips are disintegrated into fibers or "pulp."
The pulp and spent cooking liquor are subsequently separated in a
series of brown stock washers; Spent cooking liquor, referred to
as "weak black liquor," from the brown stock washers is routed to
the chemical recovery area. Weak black liquor is a dilute
solution (approximately 12 to 15 percent solids) of wood lignins,
organic materials, oxidized inorganic compounds (sodium sulfate
[Na2SO4], sodium carbonate [Na2C03]), and white liquor (Na2S and
NaOH).
In the chemical recovery cycle, weak black liquor is first
directed through a series of multiple-effect evaporators (MEE's)
to increase the solids content to about 50 percent. The "strong"
(or "heavy") black liquor from the MEE's is then either oxidized
in the BLO system if it is further concentrated in a DCE or
routed directly to a concentrator (NDCE). Oxidation of the black
liquor prior to evaporation in a DCE reduces emissions of TRS
compounds, which are stripped from the black liquor in the DCE
when it contacts hot flue gases from the recovery furnace. The
solids content of the black liquor following the final
evaporator/concentrator typically averages 65 to 68 percent.
Concentrated black liquor is sprayed into the recovery
furnace, where organic compounds are combusted, and the Na2S04 is
reduced to Na2S. The black liquor burned in the recovery furnace
has a high energy content (13,500 to 15,400 kilojoules per
kilogram [kJ/kg] of dry solids [5,800 to 6,600 British thermal
2-7
-------
units per pound {Btu/lb} of dry solids]), which is recovered as
steam for process requirements, such as cooking wood chips,
heating and evaporating black liquor, preheating combustion air,
and drying the pulp or paper products. Particulate matter (PM)
(primarily Na2S04) exiting the furnace with the hot flue gases is
collected in an electrostatic precipitator (ESP) and added to the
black liquor to be fired in the recovery furnace. Additional
makeup Na2S04, or "saltcake," may also be added to the black
liquor prior to firing.
Molten inorganic salts, referred to as "smelt," collect in a
char bed at the bottom of the furnace. Smelt is drawn off and •
dissolved in weak wash water in the SDT to form a solution of
carbonate salts called "green liquor," which is primarily Na2S
and Na2CO3. Green liquor also contains insoluble unburned carbon
and inorganic impurities, called dregs, which are removed in a
series of clarification tanks.
Decanted green liquor is transferred to the causticizing
area, where the Na2CO3 is converted to NaOH by the addition of
lime (calcium oxide [CaO]). The green liquor is first
transferred to a slaker tank, where CaO from the lime kiln reacts
with water to form calcium hydroxide (Ca(OH)2). From the slaker,
liquor flows through a series of agitated tanks, referred to as
causticizers, that allow the causticizing reaction to go to
completion (i.e., Ca(OH)2 reacts with Na2C03 to form NaOH and
CaC03).
The causticizing product is then routed to the white liquor
clarifier, which removes CaC03 precipitate, referred to as "lime
mud." The lime mud, along with dregs from the green liquor
clarifier, is washed in the mud washer to remove the last traces
of sodium. The mud from the mud washer is then dried and
calcined in a lime kiln to produce "reburned" lime, which is
reintroduced to the slaker. The mud washer filtrate, known as
weak wash, is used in the SDT to dissolve recovery furnace smelt.
The white liquor (NaOH and Na2S) from the clarifier is recycled
to the digesters in the pulping area of the mill.
2-8
-------
At about 7 percent of kraft mills, neutral sulfite
semichemical (NSSC) pulping is also practiced.6 The NSSC process
involves pulping wood chips in a solution of sodium sulfite and
sodium bicarbonate, followed by mechanical defibrating. The NSSC
and kraft processes often overlap in the chemical recovery loop,
when the spent NSSC liquor, referred to as "pink liquor," is
mixed with kraft black liquor and burned in the recovery furnace.
In such cases, the NSSC chemicals replace most or all of the
makeup chemicals. For Federal regulatory purposes, if the •
weight percentage of pink liquor solids exceeds 7 percent of the
total mixture of solids fired and the sulfidity of the resultant
green liquor exceeds 28 percent, the recovery furnace is
classified as a "cross-recovery furnace."7,8 Because the pink
liquor adds additional sulfur to the black liquor, TRS emissions
from cross recovery furnaces tend to be higher than from straight
kraft black liquor recovery furnaces.9
2.2 COMBUSTION PROCESSES AND EQUIPMENT
The following section describes the process and
characterizes the equipment for: .(1) recovery furnaces,
(2) SDT's, (3) BLO systems, and (4) lime kilns.
The process descriptions presented in this section provide
overviews of the various processes associated with the chemical
recovery cycle. Much information is available on the kraft
chemical recovery cycle and associated equipment; therefore, the
process descriptions in this section do not cover every aspect of
the chemical recovery cycle. Additional information about the
chemical recovery' cycle can be found in References 1, 2, and 10
through 13 listed in Section 2.4.1/2,10-13
2.2.1 Recovery Furnaces
2.2.1.1 Process Description. The purpose of the kraft
recovery furnace is to (1) recover inorganic pulping chemicals
(e.g., Na2S) and (2) produce steam. Inputs to the furnace
include concentrated black liquor, combustion air, and auxiliary
fuel (usually, auxiliary fuel is only used during shutdown or
startup). Outputs include molten smelt (primarily Na2S and
Na2CO3), flue gases, and steam. The smelt exits from the bottom
2-9
-------
of the furnace into a SDT tank, where the recovery of cooking
chemicals continues. Particulate matter (primarily Na2S04 and
Na2C03) entrained in the flue gases is also recovered using an
ESP and subsequently added to the concentrated black liquor.
Steam produced by the recovery furnace is used in other processes
around the mill.
Prior to being fired in the recovery furnace, the black
liquor is concentrated using an NDCE or DCE. Figures 2-5 and 2-6
show the equipment associated with NDCE and DCE recovery
furnaces, respectively. The NDCE is an indirect, steam-heated
black liquor concentrator. Black liquor typically enters the
NDCE at a solids concentration of 50 percent and exits at a
concentration of 68 percent or higher. The DCE uses the hot
combustion gases exiting the furnace to increase the solids
content of the black liquor from about 50 percent to 65 percent.
Direct contact evaporators may be of the cascade or cyclone
design. Figures 2-7 and 2-8 present detailed diagrams of the
cascade and cyclone designs, respectively. The cascade
evaporator consists of a rotating assembly of tubes that are
alternately submerged in black liquor and exposed to hot flue
gases. A cyclone evaporator is a cylindrical vessel with a
conical bottom. Black liquor is sprayed into the side of the
evaporator, where it contacts the hot combustion gases that are
introduced tangentially, creating a "cyclone" effect. The flue
gases exit from the top of the evaporator, and the concentrated
black liquor drains down to and exits from the bottom of the
evaporator. .
To minimize the stripping of TRS compounds when the hot flue
gases contact the black liquor in the DCE, most DCE's are
preceded by BLO units, which stabilize the sulfur compounds in
the black liquor. Black liquor that is concentrated in NDCE's
does not contact the hot flue gases, and, therefore, does not
require oxidation. Because the NDCE recovery furnace typically
has lower TRS emissions than does the DCE recovery furnace, the
NDCE recovery furnace is also referred to as the "low-odor"
2-10
-------
i
a
-H
0)
4.)
ft!
-H
U
O
m
en
rt
T)
C
rt
0)
u
I
U
Q)
o
CM
O
O
-H
4J
U
CO
in
i
CM
0)
i^
tn
-H
fe
2-11
-------
cu
a
-H
QJ
•O
0)
.U
ttf
-H
O
O
CO
in
rt
OJ
o
cd
Q)
S
O
w
a
p
M-l
O
O
-H
JJ
nJ
OJ
X3
O
CO
VD
I
a)
5-1
•H
2-12
-------
FLUE GAS
BLACK LIQUOR
Figure 2-7. Cascade design evaporator.
2-13
-------
FLUE GAS
TO STACK
LIQUOR TRANSFER
TO SALT CAKE
MIX TANK
4-
/
MIXING TANK
LEVEL CONTROL
BLACK LIQUOR
INLET SPRAYS
FLUE
GASES
WALL WETTING
NOZZLES
BLACK LIQUOR
RECIRCULATING
PUMPS
CYCLONE
EVAPORATOR
LEVEL CONTROL
Figure 2-8. Cyclone design evaporator.
2-14
-------
design. Since the 1970's, most new recovery furnaces have been
designed with NDCE's.
Regardless of how the black liquor is concentrated, the
chemical reactions that take place inside an NDCE- or DCE-type
furnace are the same. The concentrated black liquor is sprayed
into the furnace through fixed or oscillating nozzles or "guns"
mounted in the walls of the furnace. Depending upon the design
and operation of the recovery furnace, the sprayed black liquor
may hit the opposing wall, where it dries and burns before
falling to the,hearth, or may fall short of the opposing wall and
dry and burn in suspension. Combustion air is generally supplied
to the furnace at three levels, with two levels located below the
black liquor nozzles and one above. Some furnaces may have only
two combustion air levels (i.e.,- one above and one below the
liquor nozzles).
The inorganic chemicals in the black liquor are recovered in
three distinct zones inside the recovery furnace. Figure 2-9
shows where each zone is located in the furnace and the chemical
reactions that take place in each zone. The area of the furnace
extending from the black liquor spray to just above the molten
smelt at the bottom of the furnace is referred to as the "drying
zone." The purpose of the drying zone is to evaporate the water
from the liquor droplets. The reducing zone is just below the
drying zone and includes the char bed. The reduction of Na2S04
to Na^S takes place in the reducing zone. Volatile gases that
are released in the drying and reducing zones of the furnace
travel to the highly turbulent upper section of the furnace,
referred to as the oxidizing zone, where the gases are
combusted.2 The heat generated from the combustion of these
gases is then used to generate steam as the combustion gases are
drawn through the heat exchanger section of the furnace (i.e.,
superheater, boiler bank, and economizer).
For NDCE recovery furnaces, the design economizer exit gas
temperature ranges from 177° to 190°C (350° to 375°F).1 For DCE
recovery furnaces, the heat from the recovery furnace is used to
evaporate -the black liquor. Thus, the required economizer exit
2-15
-------
OXIDIZING ZONE
CO + 1/2O2
H2S + 3/2 O2
SO2 + 1/2 O2
Na2S + 2C>2
SO3
3/2 O2 + CO2
CO2
SO2 + H2O
S03
N32SO4 + CO2
DRYING ZONE
ORGANICS + HEAT
+ CO2 + I
CO2
• PYROLYSIS PRODUCTS
• Na2CO3 + H2S
REDUCING ZONE
ORGANICS + HEAT-
Na2SO4 + C
N32SO4 + 2C
4C
PYROLYSIS PRODUCTS
Na2O + SO2 + CO
Na2S + 2CO2
Na2S + 4CO
TERTIARY
AIR
LIQUOR
GUNS
SECONDARY
AIR
I
PRIMARY
AIR
SMELT
Figure 2-9. Recovery furnace zones and air stages.
2-16
-------
gas temperature for DCE recovery furnaces, 370° to 430°C (700° to
800°F), is much higher than that for NDCE recovery furnaces.
Because a large portion of the combustion heat from DCE recovery
furnaces is required for black liquor evaporation, less
combustion heat is available to produce steam.
Flue gases exiting the economizer are routed either directly
to a PM control device (i.e., for NDCE recovery furnaces) or to a
DCE followed by a PM control device (i.e., for DCE recovery
furnaces). The recovered PM (primarily Na2S04 [saltcake] and
Na2C03) is subsequently added to the concentrated black liquor,
and the cleaned flue gas exits through the stack. Approximately
95 percent of the Na2S04 is recovered; additional makeup saltcake
is added to the concentrated black liquor as needed.
The inorganic chemical in the black liquor, Na^O^, is
reduced to Na2S, a cooking liquor chemical, in the reducing zone
(lower section) of the furnace. The Na2S and other inorganic
chemicals, predominantly Na2CC>2, drain as molten smelt from the
furnace bottom to the SDT, where reprocessing into cooking liquor
continues.'
2.2.1.2 Equipment Characterization. An estimated
211 recovery furnaces operate at U.S. kraft pulp mills.3
Detailed information is available on about 85 percent of the
estimated number of furnaces.14 Based on the available data,
61 percent of these recovery furnaces are NDCE recovery furnaces
and 39 percent are DCE recovery furnaces. Eight percent of the
recovery furnaces (76 percent of which are NDCE and 24 percent of
which are DCE recovery furnaces) are located at kraft pulp mills
that also practice NSSC pulping.6 The number of recovery
furnaces at these mills that can be classified as cross-recovery
furnaces is uncertain.
The nationwide distribution of DCE and NDCE recovery
furnaces by age (i.e., original installation date) is provided in
Figures 2-10a and 2-10b, respectively.3 As shown in
Figures 2-10a and 2-10b, most recovery furnaces installed since
the 1970's have been NDCE recovery furnaces. The first
installation of a new NDCE recovery furnace took place in 1969.15
2-17
-------
1947-49 1950-59 1960-69 1970-79 1981X89 1990-94
Range of year installed
Figure 2-10a. DCE recovery furnace age distribution.
1947-49 1950-59 1960-69 1970-79 1980-89 1990-94
Range of year installed
Figure 2-10b. NDCE recovery furnace age distribution.
2-18
-------
Although Figure 2-10b shows that approximately 22 NDCE recovery
furnaces currently in operation were installed in or before 1969,
most, if not all of these furnaces were originally installed as
DCE recovery furnaces and later converted to NDCE recovery
furnaces. The most recent installation of a new DCE recovery
furnace occurred in 1988; however, prior to 1988, the last
installation of a new DCE recovery furnace was in 1979.3
The nationwide distribution of DCE and NDCE recovery
furnaces by size (i.e., black liquor solids [BLS] feed rate) is
provided in Figures 2-lla and 2-lib, respectively.3 The DCE
recovery furnaces are not only typically older than NDCE recovery
furnaces, but, as shown in Figures 2-lla and 2-lib, DCE recovery
furnaces are also typically smaller. The majority of DCE
recovery furnaces have a firing rate less than or equal to
900,000 kilograms of BLS per day (kg BLS/d) (2 million pounds of
BLS/d [Ib BLS/d]); whereas, the majority of NDCE recovery
furnaces have firing rates greater than 900,000 kg BLS/d
(2 million Ib BLS/d).
The two furnace types also differ in the types of ESP's that
are used to control particulate emissions from the furnace. For
example, the ESP's that control PM emissions from DCE recovery
furnaces tend to be wet-bottom ESP's (i.e., the PM catch falls
directly into a pool of black liquor at the bottom of the ESP),
whereas ESP's on NDCE recovery furnaces tend to be dry-bottom
ESP's. Other differences between DCE and NDCE recovery furnaces
include the following: (1) the inlet loading of PM to the ESP
tends to be lower for DCE recovery 'furnaces than for NDCE
recovery furnaces due to the recovery of some PM as the flue
gases pass through the DCE; (2) emissions of TRS compounds tend
to be higher from DCE recovery furnaces; and (3) NDCE recovery
furnaces do not require black liquor oxidation, which eliminates
one source of emissions.
Due to State and Federal regulations regarding PM emissions
and the economic benefits of recycling PM captured from the
recovery furnace flue gases, all kraft recovery furnaces are
equipped with add-on PM control devices. Particulate matter
2-19
-------
4CK
Range of solids feed rates, MM ib BLS/d
Figure 2-lla. DCE recovery furnace size distribution.
>0.4 and <=1' >2 and <=3 >4 and <=5
>1 and <=2 >3 and <=4 >5
Range of solids feed rates, MM Ib BLS/d
Figure 2-llbi NDCE recovery furnace, size-, distribution.
2-20
-------
emissions from.- approximately 95 percent of all kraft recovery
furnaces are controlled with an ESP alone.3 The remaining
furnaces are controlled with ESP's followed by wet scrubbers
(4 percent) or with wet scrubbers alone (1 percent).3 Additional
information on recovery furnace air pollution control devices
(APCD's) and the effectiveness of these APCD's in controlling HAP
emissions is provided in Chapter 3.
In general, recovery furnaces are not used to incinerate waste
streams generated in other parts of the mill, with the exception
of a few recovery furnaces (approximately 3 percent) that receive
high-volume, low-concentration (HVLC) noncondensible gases
(NCG's) from the pulping area of the mill.14 Because of the
importance of the recovery furnace to the chemical recovery cycle
and the potential for catastrophic explosion (due to water
entering the furnace during operation), the industry's Black
Liquor Recovery Boiler Advisory Committee (BLRBAC) recommends
that the recovery furnace not be used for NCG incineration.1
2.2.2 Smelt Dissolving Tanks
2.2.2.1 Process Description. Molten smelt is one of the
main products from the combustion of black liquor. Smelt, which
is predominantly Na2S and Na2CO3, is formed in the bottom of the
furnace. Smelt, at approximately 1040° to 1150°C (1900° to
2100°F), filters through the char bed and is continuously
discharged through water-cooled smelt spouts into the SDT. In
the SDT, smelt is mixed with weak wash water from the
recausticizing area to form green liquor, an aqueous solution of
Na2C03 and Na2S in about a three to one ratio. The green liquor
is subsequently transferred to the recausticizing area for
reprocessing into cooking liquor (i.e., white liquor). A
schematic of an SDT with a wet scrubber is provided in
Figure 2-12.
The SDT is a large, covered vessel located below the
recovery furnace. Green liquor is maintained in the tank at a
level of about half the depth of the tank. As the smelt exits
the water-cooled smelt spouts and falls several feet into the
SDT, it is shattered by high-pressure steam or shatter sprays of
2-21
-------
WET SCRUBBER
TO STACK
SCRUBBER
BYPASS -
VALVE
SMELT
\
WATER-COOLED
SPOUT
/ SHATTER
/ SPRAY NOZZLE
FAN
Nfl
OVERFLOW
LINES -
1 1
A A A AAA
4
1
1
' X
\
A
\ A
1 1
/ 1
X
t L
/
WEAK
WASH
WATER
SCRUBBER WATER
DRAIN TO SDT
GREEN LIQUOR
TO CLARIFIER
AGITATOR
SMELT DISSOLVING TANK
Figure 2-12. Smelt dissolving tank and wet scrubber.
2-22
-------
recirculated green liquor. The steam or shatter sprays break the
smelt flow into small droplets and cool the smelt before it falls
into and reacts with the liquid in the SDT to form green liquor.
Large volumes of steam are generated when the molten smelt and
liquid mix. The vapor space above the liquid level provides an
opportunity for water vapor and PM resulting from the quenching
of smelt to settle out of suspension into the green liquor. An
induced-draft .fan*constantly draws the vapor and entrained PM
through a PM control device, generally a wet scrubber. Scrubber
water is sprayed into the scrubber and allowed to drain directly
into the SDT, where it reacts with smelt to form green liquor.
The SDT is constantly agitated to prevent formation of hot
spots on the surface of the liquor and to keep solids from
accumulating in the bottom of the tank. Surface liquor hot spots
can contribute to the formation of explosive hydrogen gas from
the dissociation of water reacting with the hot smelt.
Green liquor formed in the SDT is sent to the green liquor
clarifier in the causticizirig area. Green liquor is converted to
white liquor (i.e., NaOH and Na2S) in the causticizing area.
2.2.2.2 Equipment Characterization. An estimated 227 SDT's
operate at U.S. kraft pulp mills.3 Detailed information is
available on approximately 83 percent of the estimated number of
SDT's.6 Smelt dissolving tanks are basically large, covered
tanks. Some recovery furnaces have two SDT's, which explains why
the estimated number of SDT's is higher than the estimated number
of recovery furnaces. When there are two SDT's under one
furnace, the flue gases from these SDT's may be combined or
treated separately.
Due to State and Federal regulations regarding PM emissions,
all but two kraft SDT's are equipped with add-on PM control
devices. Particulate matter emissions from approximately
8? percent of kraft SDT's are controlled with wet scrubbers.14
Particulate matter emissions from most of the remaining SDT's are
controlled with mist eliminators alone.14 Additional information
on SDT APCD's and the effectiveness of these APCD's in
controlling HAP emissions is provided in Chapter 3.
2-23
-------
2.2.3 Black Liquor Oxidation Systems
2.2.3.1 Process Description. The BLO system reduces
malodorous TRS emissions from DCE recovery furnaces at kraft pulp
mills. Total reduced sulfur compounds (primarily hydrogen
sulfide [H2S]) are stripped from black liquor when hot flue gases
from the recovery furnace contact the black liquor in the DCE.
Hydrogen sulfide is stripped by the reaction between residual
Na2S in the black liquor and carbon dioxide (C02) and water (H20)
in the recovery furnace flue gas, as follows:
Na2S
sodium
sulfide
CO,
H20
Na2C03
sodium
carbonate
H2S
hydrogen
sulfide
The BLO system minimizes the stripping of TRS compounds in the
DCE by stabilizing the sulfur compounds in the black liquor prior
to evaporation in the DCE. The main reaction that takes place in
the BLO system is the oxidation of Na2S to nonvolatile sodium
thiosulfate (Na2S203), as follows:
2Na2S
sodium
sulfide
20-
HoO
Na2S203
sodium
thiosulfate
2NaOH
sodium
hydroxide
The oxidation efficiency of the BLO process is measured by
the percent conversion of Na2S to Na2S203 in the black liquor on
a gram per liter (g/L) (pound per gallon [Ib/gal]) basis.
Oxidation efficiencies greater than 99 percent are common. High
oxidation efficiencies require good mixing and the avoidance of
excessive oxygen (02) consumption to limit side reactions.
Sulfidity levels of the black liquor entering the DCE are
targeted to less than 0.2 g/L (0.002 Ib/gal) in order to meet TRS
emission limits for DCE recovery furnaces.1 Oxidizing black
liquor results in a slight increase (2 to 3 percent) in the
solids content of the black liquor and reduces its heating value
by 2 to 5 percent. °
2-24
-------
The BLO system is typically located after the MEE, a process
referred to as "strong" black liquor oxidation. A small number
of mills oxidize black liquor prior to evaporation in the MEE,
which is referred to as "weak" black liquor oxidation. Other
options are to oxidize both weak and strong black liquor or to
oxidize the black liquor between effects of the MEE. To avoid
excessive foaming caused by tall oil components in weak black
liquor, mills in the south that pulp pine oxidize strong black
liquor. With strong black liquor, the soaps and tall oil are
removed during the soap-skimming process that follows black
liquor evaporation in the MEE. With weak BLO systems, the
effects of partial reversing of the oxidation reaction (i.e.,
oxidized sulfur compounds reducing to H2S) that occurs in the MEE
can be minimized by adding a second oxidation step, such as
oxygen polishing of strong black liquor.
Black liquor can be oxidized using either air or pure
(molecular) 02. Because of economic considerations, the majority
of BLO systems use air as the oxidant. Air BLO systems have
higher capital costs than oxygen systems, but their operating
costs are usually much lower.
Air-sparging units are the predominant type of BLO equipment
used-at kraft pulp mills. Figure 2-13 shows a schematic of an
air-sparging BLO system. Air-sparging units operate by bubbling
air, which is sometimes preheated, through the black liquor using
multiple diffuser nozzles. Air systems require residence times
of several hours or more to obtain high oxidation efficiencies.
Because of this relatively long residence time, large oxidation
tanks--on the order of 9 meters (m) (30 feet [ft]) in diameter
and 12 m [40 ft] high) must be used. Air-sparging units have one
to three tanks (or stages) that operate in series. Air-sparging
units are equipped with mechanical foam breakers for foam
control; chemical defoamers (e.g., diesel oil, turpentine, and
kerosene) may also be used.1'12 Each oxidation tank is vented.
Gases exiting the BLO system flow through cyclone separators- to
have entrained water droplets removed prior to being vented to
2-25
-------
CO
H
I
0)
OJ
01
>1
CQ
fl
O
-H
4J
rt
O
-H
H
U
X)
Cn
a
-H
Cn
a
03
•H
rtf
Q)
Cn
rt
4->
CD
ro
H
0)
I
-H
2-26
-------
the atmosphere. Add-on APCD's typically are not used with air
oxidation units.
Oxygen BLO systems require only 30 seconds (sec) to
5 minutes (min) residence time, which enables the use of in-line
reactors. Since all the gas (i.e., pure O2) added to the system
is consumed in the oxidation reaction, a system vent is not
needed.16 The oxygen consumption for weak black liquor oxygen
systems is 25 to 65 kilograms per megagram (kg/Mg) of pulp (50 to
130 pounds per ton [Ib/ton] of pulp); strong black liquor oxygen
systems use 35 to 75 kg 02/Mg pulp (70 to 150 Ib 02/ton pulp.)17
Oxygen polishing is sometimes used as a supplement to air
oxidation systems to address (1) stricter TRS standards, (2) an
overloaded BLO system resulting from production increases, or
(3) peaks in TRS emissions resulting from process upsets or
temporary production increases. In-line 02 polishing systems are
used to oxidize strong liquor and may follow either a weak liquor
air system or a strong liquor air system. Polishing of strong
black liquor requires approximately 2.5 to 13 kg 02/Mg pulp (5 to
25 Ib 02/ton pulp) to achieve the desired incremental increases
in oxidation efficiency.17
2.2.3.2 Equipment Characterization. Of the 47 U.S. kraft
pulp mills that operate DCE recovery furnaces, only 1 (with 1
DCE recovery furnace) is known to operate without a BLO system.
Therefore, the nationwide number of BLO systems is estimated to
be 46. Detailed information is available on approximately
91 percent of the estimated U.S. population.14'18 The type of
equipment operated and the type of liquor oxidized for the BLO
systems is summarized in Table 2-I.3 Oxidation of strong black
liquor using air-sparging units is the predominant method of
black liquor oxidation and accounts for an estimated 89 percent
of the U.S. population.3 Seven percent of the mills operating
air-sparging units also have added 02 polishing systems. Of the
air-sparging BLO systems, 50 percent are two-stage systems,
31 percent are single-stage systems, and 18 percent have three
oxidation tanks.14
2-27
-------
TABLE 2-1. BLACK LIQUOR OXIDATION SYSTEMS: EQUIPMENT TYPES
Equipment type
Air sparging
Air sparging with 02
polishing
Air, sieve tray tower
Molecular 0^
Molecular 07
Molecular 09
Total
Liquor type
Strong
Strong
Weak
Strong
Weak
Weak and strong
Estimated
No. in
U.S.
38
3
1
2
1
1
46
Percent
83
7
2
4
2
2
100
Approximately 8.5 percent of BLO systems are molecular 02
systems.3 Strong black liquor is oxidized with 02 after the
second effect of the MEE (26 percent solids) at two of the four
mills; weak black liquor (i.e., prior to the MEE) is oxidized at
another mill; and both weak and strong liquor are oxidized with
02 at the fourth mill.3'14 One mill with a BLO system located
after the second effect of the MEE reported an oxygen usage rate
of 32 Mg/d (35 ton/d) (13 kg/Mg pulp [25 Ib/ton pulp]).14
Emissions from most air-based BLO systems (95 percent) are
uncontrolled.3 Based on available information, TRS emissions
from 5 percent of air-based BLO systems are currently controlled
by using a condenser or mist eliminator to remove the water vapor
and then venting the gas stream to a power boiler for
incineration.14 Molecular 02 systems do not require system vents
and, thus, have no emissions directly associated with the BLO
unit.13
2.2.4 Lime Kilns
2.2.4.1 Process Description. The lime kiln is part of the
causticizing process, in which green liquor from the SDT is
converted to white liquor. The function of the lime kiln is to
2-28
-------
convert lime mud (calcium carbonate [CaC03]) to reburned lime
(calcium oxide [CaO]), a process known as calcining:
CaC03
lime mud/
calcium carbonate
heat
CaO +
lime/
calcium oxide
[lime kiln]
Lime mud that is burned in the lime kiln is a final product of
the following causticizing reactions:
CaO
H20
lime/
calcium oxide
Ca(OH)2
calcium
hydroxide
Na2C03
sodium
carbonate
Ca(OH)2
calcium
hydroxide
NaOH 4
sodium
hydroxide
[s laker]
CaCO,
[causticizer]
lime mud/
calcium carbonate
Lime that is produced in the kiln is used in the causticizing
reaction that takes place in the green liquor slaker and
causticizers. The lime kiln typically produces about 95 percent
of the lime needed for the causticizing reaction. Either make-up
lime or limestone is purchased to account for losses.
Prior to calcining, lime mud from the causticizing tanks is
washed and dewatered. Lime mud washers reduce the sodium and
sulfide content of the lime mud, which lowers TRS emissions from
the lime kiln. The lime mud is typically dewatered to about
70 percent solids using a rotary vacuum precoat filter. The
precoat filter also helps reduce TRS emissions by oxidizing any
remaining sodium sulfide to sodium thiosulfate using air that is
pulled through the filter.
Rotary lime kilns, as shown in Figure 2-14, are typically
used at kraft pulp mills. In a rotary kiln, lime mud from the
precoat filter is introduced at the feed end (cold end) and flows
downward towards the discharge end (hot end). The chemical
composition of the lime mud is presented in Table 2-2. Natural
gas or fuel oil are the fuels typically used to fire the kiln.
2-29
-------
Q
Z
UJ
I
Q
LU
111
CD
n:
<
o
LU
Q
O
O
Q
01
Q
LU
LU
t
LU
ra
.H
g
a
rt
M
•d-
Q)
en
a
rH
-rH
0)
g
1-1
O
u
-H
4J
Q)
^3
O
CD
0)
tn
•H
2-30
-------
TABLE 2-2. CHEMICAL COMPOSITION OF LIME MUD FEED
20
Component
Calcium carbonate
Alumina
Iron oxide
Calcium oxide
Sodium oxide
Silica
Sodium sulfide
Composition (%)
>95
<0.5
<0.5
<0.5.
<0.2
<0.2
<0.01
Primary combustion air is introduced through a concentric tube
around the fuel pipe, and preheated secondary combustion air is
introduced through the bottom of the firing hood.
The majority of lime kilns at kraft pulp mills also burn
NCG's from various process vents, such as digester and evaporator
vents. The NCG streams may be introduced into the kiln- through a
dedicated nozzle or combined and fed with either primary or
secondary air. The components of the NCG stream include TRS
compounds, turpentine, methanol, acetone, alpha-pinene, water
vapor, nitrogen, and oxygen.1'19 To avoid excessive sulfur
dioxide (SO2) formation in the lime kiln, the NCG gas stream may
be scrubbed prior to incineration to remove TRS compounds.
Packed columns with white liquor as the scrubbing fluid are
commonly used. The TRS removal efficiency is typically about
70 percent.1
Rotary kilns have three internal zones: (1) the drying
zone, (2) the heating zone, and (3) the calcining zone. In the
drying zone, water is evaporated from the lime mud as it passes
througli metal chains that are suspended from the kiln shell in a
curtain or garland arrangement. The lime mud is dried to about
95 percent or greater solids as it passes through the chains that
are heated by the hot flue gases that flow countercurrent to the
lime mud. In addition to providing additional heat transfer area
for drying the lime mud, the chains also help reduce the amount
2-31
-------
of lime dust exiting with flue gases. In the heating zone, other
heat transfer devices, such as tumblers and lifters, are used to
heat the dried mud uniformly in preparation for calcining. The
calcining reaction, which takes place in the calcining zone,
requires a minimum temperature of 815°C (1500°F); temperatures
greater than about 1150°C (2100°F) can cause everburning, which
leads to a less-reactive lime product.1 Ideally, calcination
produces lime pellets that are about 2 centimeters (cm)
(0.75 inches [in.]) in diameter; however, if the lime mud is
improperly dried and heated, large lime balls may be produced.-1
The hot lime product is cooled by incoming secondary air as it
passes under the burner towards the discharge end of the kiln.
Newer kilns use integral tube coolers to preheat secondary
combustion air while cooling the discharged lime pellets. In
this heat exchange process, the air is heated to about 315°C
(600°F) and the lime is cooled to about 190°C (375°F). The
reburned lime product from the integral tube coolers, or from the
kiln discharge hoppers in the absence of coolers, is transported
to the lime storage bin and subsequently introduced into the
green liquor slaker.
Combustion gases exit the lime mud feed end of the kiln at
temperatures of approximately 150° to 200°C (300° to 400°F). The
exhaust gases consist of combustion products, CO2 released during
calcination, water vapor evaporated from the mud, and entrained
lime dust. Particulate matter in the exhaust gas is mainly
sodium salts, CaC03, and CaO. Add-on PM control devices are
required to meet Federal and State PM standards. Venturi
scrubbers are the most commonly used control device, and water is
typically used as the scrubbing fluid. The exhaust stream may be
scrubbed with a caustic solution with the added benefit of
lowering TRS and SO2 emissions. However, Federal and State TRS
standards can be met through good lime mud washing practices,
which reduces the sulfide content of the lime mud feed. If a wet
scrubber is used, a cyclone separator may be installed upstream.
The dust collected by the separator is returned directly to the
lime kiln. In recent years, the use of ESP's has been more
2-32
-------
prevalent. As>-with the cyclone separator, the PM catch from the
ESP is returned directly to the lime kiln.
2.2.4.2 Equipment Characterization. An estimated 192 lime
kilns operate at U.S. kraft pulp mills.3 Detailed information is
available on about 85 percent of the estimated number of lime
kilns.14 Rotary kilns are the most commonly used type of lime
kiln at kraft pulp mills, accounting for about 98 percent of the
kilns.6'21 Fluidized-bed calciners are.also used by the kraft
pulp industry. Industry representatives estimated the number of
fluidized-bed calciners at two to five, accounting for the
remaining 2 percent of lime kilns at kraft pulp mills.6'21 Based
on available data, the oldest operating lime kiln was installed
in 1940.14 New lime kilns were installed at a rate of
approximately three kilns per year from 1981 to 199I.14
The predominant fuel used by the industry is natural gas,
accounting for about 68 percent of lime kilns.14 Fuel oil is the
primary fuel for about 29 percent of kilns.1 Petroleum coke is
fired in about 3 percent of the kilns as the primary fuel.
Due to State and Federal regulations regarding PM emissions,
all lime kilns are equipped with add-on PM control devices.
Particulate matter emissions from the majority (90 percent) of
lime kilns at kraft pulp mills are controlled with wet scrubbers.
Two percent of these scrubbers are operated in series with a
second scrubber. Venturi scrubbers are the most commonly used
type of wet scrubber. Water is the typical scrubbing fluid,
but caustic and weak wash are also used.14 The scrubbing fluid
is recirculated, and the scrubber blowdown is recycled to the
lime mud washer. Particulate matter emissions from the remaining
10 percent of lime kilns are controlled by single-chamber ESP's
(9 percent) or the combination of an ESP and scrubber
(1 percent},1 Installing ESP's to control PM from lime kilns
has been more widespread in recent years; about half of the APCD
installations since 1990 have been ESP's.
An estimated 66 percent of lime kilns are either primary or
backup incineration devices for NCG streams.14 Of the lime kilns
that incinerate NCG's, 57 percent incinerate low-volume, high-
2-33
-------
concentration (LVHC) NCG streams; 28 percent incinerate HVLC NCG
streams; and 24 percent incinerate both LVHC and HVLC NCG
streams. Approximately 33 percent of the lime kilns in which
NCG's are incinerated use an NCG scrubber prior to
incineration.14 The majority of these NCG scrubbers use white
liquor as the scrubbing fluid; a small percentage use water.
2.3 BASELINE EMISSIONS
2.3.1 Federal and State Regulations Affecting Kraft Pulp Mill
Combustion Sources
Federal regulations that affect the pulp and paper industry
include new source performance standards (NSPS) for kraft pulp
mills. The NSPS does not include soda mills. The NSPS specifies
PM and TRS emission limits for various pulp mill equipment that
are constructed, modified, or reconstructed after September 24,
1976.7'8 Recovery furnaces,.SDT's, and lime kilns are regulated
under the kraft pulp mill NSPS.
The percentage of recovery furnaces, SDT's, and lime kilns
that are subject to the NSPS were estimated using installation
dates. These estimates, therefore, do not include those recovery
furnaces, SDT's, and lime kilns that are subject to the NSPS
through the modification/reconstruction provisions.
Approximately 4 percent (i.e., 3 of 83) of the DCE recovery
furnaces and 39 percent (50 of 128) of the NDCE recovery furnaces
are subject to the NSPS.3 Based on recovery furnace installation
dates, approximately 29 percent of the SDT's are subject to the
NSPS.3 Approximately 28 percent of the lime kilns are subject to
the NSPS.14 Nine new BLO systems have been installed since the
1976 NSPS proposal for kraft pulp mills; eight of these new BLO
systems are still in operation, making up 17 percent of the
18
existing population of operating BLO systems.
Table 2-3 lists the NSPS PM and TRS emission limits. The
EPA TRS emission limit guidelines for sources not subject to the
NSPS are also listed in Table 2-3. The TRS guidelines are the
same as the NSPS emission limits for NDCE and cross-recovery
furnaces. The NSPS emission limit is 5 parts per million dry
volume (ppmdv) at 8 percent O2. For DCE recovery furnaces, the
2-34
-------
rd
CO
w
C
D
o
CO
S
0
H
EH
CO
£D
i
o
o
H
s
O
CM
H
&
a
«
«
o
co
w
M
H
P
H
CD
O
H
CO
CO
H
W
B
IcT
f""H
CO
CO
K
— '
^
00
i
o.
OH
to
cT
fe^,
00
©
•a
ex
a,
ir>
0*
00
O
o
•3
00
•^j-
o
o
E
•a
~ob
0
*— *
CJ
u
a
0
O
w
u
Q
Z
•cT
vo
ES
oo
i
OH
OH
®
0^
fe*v
00
1
cu
»n
o"1
oo
©
c-
o
CO
•o
00
•^f
O
o
6
T3
"Sb
o
^-*
0
8
03
&
S
o
w
o
Q
cT
VS3
OO
•o
OH
OH
cs
f
oo
1
&
CN
cF
oo
®
o
CO
T3
00
•^t-
0
S
"T3
"ob
o
*-^
°"
o
o
CO
i
!
§
£
8
^^/
00
s
JM
vo
O i 1
O CQ
3
CQ
'a'
o
£
8
&0
--*
^
o
*— e
S3
-a
oo
_c
"o
.2
•5
1
CO
cs
0
fe"*
o
1
CN
s
o
i— <
1
o.
OH
OO
cs
o
CD
©
o
8
5
00
r-
o
o
o
CO
"oo
tjT)
*-«
°"
|
S
00
_e
1
3
CM
0
fe^-
0
,o»
^@
1
cs
cf4
!£
O
1
OH
OH
OO
C?
o
o
"tS
OO
en
O
1
"So
o
en
O
tC
•O
*3
s.
i
j
a
|
o
Z
o
o
Q
Z
e
1
°a
_o
a
"2
'S
o
VH
O
2*
S
g
2-35
-------
TRS guidelines are 20 ppmdv at 8 percent 02, which are higher
than the NSPS limit. Similarly, the TRS guidelines for lime
kilns (20 ppmdv at 10 percent 02) are higher than the NSPS limit
(8 ppmdv at 10 percent 02)- The TRS compounds are hydrogen
sulfide, methyl mercaptan, dimethyl sulfide, and dimethyl
disulfide. None of these TRS compounds are currently listed in
the Clean Air Act for regulation under Section 112. However, PM
emitted from kraft pulp mill sources does contain small amounts
of HAP compounds (e.g., HAP metals).
State regulations for PM and TRS emissions for new,
modified, and reconstructed recovery furnaces, lime kilns, and
SDT's are the same as Federal regulations, with a few
exceptions.22 The State of Georgia has promulgated process-
weighted standards for PM, which, for some Georgia mills, are
slightly more stringent than the NSPS PM limits. Also, the State
of Alabama requires new sources to use the "highest and best
practicable controls for PM."22 In addition, several States
regulate toxic air emissions from pulp mills. These regulations,
however, tend to limit the concentration of air toxics at the
pulp mill fence line, and thus do not directly regulate emissions
from specific equipment in the chemical recovery area of the
mill. For existing recovery furnaces that are not subject to the
NSPS, most States have emission standards equivalent to or more
stringent than the EPA TRS emission limit guidelines.22'23 For
existing lime kilns that are not subject to the NSPS, approxi-
mately half of the States with kraft mills have emission
standards equivalent to or more stringent than the EPA TRS
99 9^
emission limit guidelines.-"'^J
2.3.2 Baseline Emission Estimates
Section 2.3.2.1 presents the baseline gaseous HAP emission
estimates from combustion sources at kraft and soda pulp and
paper mills. Baseline gaseous HAP emissions were developed for
DCE recovery furnaces, NDCE recovery furnaces, and BLO systems.
Baseline gaseous HAP emissions were not developed for lime kilns
and SDT's because the gaseous HAP emissions from these units
result from the use of contaminated condensates in the makeup
2-36
-------
waters. The regulation of contaminated condensates is part of
the noncombustion source category for kraft and soda pulp and
paper mills. Section 2.3.2.2 presents baseline PM and TRS
emissions from combustion sources at kraft and soda pulp and
paper mills.
2.3.2.1 Baseline Gaseous HAP Emissions. Table 2-4 presents
gaseous HAP process emission factors for the two types of
recovery furnaces (DCE and NDCE) used in the kraft chemical
recovery loop. The process emission factors were calculated
based on available emission test data and process emission rates
for each of the recovery furnaces.24 In the case of DCE recovery
furnace systems, emission points include the recovery furnace/DCE
stack and the BLO vent. The only emission point associated with
NDCE recovery furnaces is the recovery furnace-stack. As shown
in Table 2-4, the gaseous HAP compounds emitted in the largest
quantities from DCE and NDCE recovery furnaces are.hydrochloric
acid (HC1) and methanol. The following sections discuss the
potential mechanisms by which HC1 and gaseous organic HAP's such
as methanol are emitted, as well as the nationwide emissions of
these pollutants.
2.3.2.1.1 HC1 emissions. Typically, HC1 emissions from
combustion processes are generated from oxidation reactions
involving chlorinated organic compounds in the fuel, and all of
the organic chlorine in the fuel is presumed to. be emitted as
HC1. However, chlorine in black liquor is expected to be in the
form of inorganic chlorine. Consequently, only a small fraction
(0 to 10 percent) of the black liquor chlorine is believed to be
emitted from the recovery furnace as HC1.25 A possible mechanism
for HC1 formation in the recovery furnace is gas phase reactions
of volatilized alkali chlorides with S02 in the presence of water
vapor and oxygen.25'26 The proposed reactions are as
follows:25'26
2NaCl(s) + S02(g) + 1/2 02 (g) + H20(g) -» 2HC1 (g) + Na2SO4 (s) (1)
2KCl(s) + S02 (g) + 1/2 02 (g) + H20(g) -* 2HC1 (g) + K2SO4(s) (2)
2-37
-------
CO
O
«!
En
52;
O
H
CO
CO
w
1ST
w
•tf
CS
m
1
1
IS
W
pa
to
E
"S
§
W
O
g
'S
*jjj
m
8
C3
E
a
b
4J
O
g
U
u
7-
p
?
en
cs
3
ecovery
CJ
u
"Ha
S*
o
5
&o
o
X
CS
_^
b
X
o\
o\
CS
o
x
»
*— 1
u
•c
>»
-g
c:
u
VO
b
X
t—
t-
VO
b
X
ON
VO
b
X
en
t-
_
o
X
^>
— <
VO
b
X
u
N
U
CQ
vo
b
X
rs
ON
VO
b
X
CO
CS
VO
o
X
en
_
b
X
S
CS
CT
o
X
Cv
CO
CS
a
^^
JG
-c
&
VO
b
X
{*
VO
VO
b
X
VO
r-
VO
b
X
s
_<
VO
b
X
en
— '
b
X
CS
~ J
5
^
•o
o
0
"^
c
j:
•a
ffi
•*
b
X
§
Tf
0
X
«0
CS
•*
b
X
-------
These reactions are believed to occur at temperatures below about
900°C (1650°F) in the furnace post-combustion zone and in higher
temperature areas in the ductwork downstream from the
furnace. ' Assuming the above reactions are the correct
mechanism for HC1 formation, HC1 emissions will be higher for
well-designed and operated combustion systems. Combustion system
parameters that could potentially affect the amount of HC1 formed
include higher combustion zone temperatures leading to greater
emissions of volatile sodium chloride (NaCl) and potassium
chloride (KCl), furnace S02 generation rates, and good mixing of
combustion gases with excess air in the post-combustion zone.
A review of the available emissions data indicates that HC1
emissions from recovery furnaces are highly variable. A recent
study of kraft recovery furnace HC1 emissions conducted by the
National Council of the Paper Industry and Air and Steam
Improvement, Inc. (NCASI) examined the relationships between
several recovery furnace operating parameters and HC1
emissions. Some earlier studies had suggested that there may
be strong correlations between stack S02 emissions and HC1
emissions and between stack O2 concentration and HC1
o c: o C.
emissions. 3/a The NCASI study, which was based on tests of 14
recovery furnaces conducted specifically for this study,' examined
the relationship between HC1 emissions and each of the following
parameters: stack SO2 concentration, as-firedrblack liquor
chloride content, stack O2 levels, chloride content of
precipitated ash, and furnace load or liquor firing rate.
Based only on the tests conducted in this study, NCASI found a
strong correlation using a Langmuir adsorption-type relationship
between HC1 emissions and S02 emissions for NDCE 'recovery
furnaces and found no clear correlation between these parameters
for DCE recovery furnaces. The study found no clear
correlations between HC1 emissions and other recovery furnace
operating parameters.27
As shown in Table 2-4, the HC1 emission factor presented for
both NDCE and DCE recovery furnace systems is the same. Based on
a statistical analysis of the available HC1 emissions data, it
2-39
-------
was concluded that the NDCE and DCE recovery furnace HC1 emission
factors should be averaged together. The range of HC1 emission
factors for NDCE recovery furnaces extends above and below the
range of factors for DCE recovery furnaces. Both the 90 and
95 percent confidence intervals of the difference between the DCE
and NDCE recovery furnace means include zero. Stated another
way, the hypothesis that the means are equal cannot be rejected
-------
high, then metihanol emissions from the DCE recovery furnace stack
were .low) . °
No emission data are available from DCE recovery furnaces
with molecular oxygen BLO systems for comparison with DCE
recovery furnaces with air-based BLO systems. Therefore, the
effect of molecular oxygen BLO systems on total emissions from
the DCE recovery furnace system is uncertain. Unlike air-based
systems', molecular, oxygen systems use pure oxygen, and thus, no
diluents are introduced that could strip organic compounds from
the black liquor; consequently, organic compounds not released
from the black liquor during the oxidation process could be
subsequently stripped, in theory, from the oxidized black liquor
when the black liquor enters the ESP or direct contact
evaporator.
Most DCE recovery furnaces and some older NDCE recovery
furnaces are equipped with wet ESP systems. Either black liquor
or HAP-contaminated process water may be present in the bottom of
these ESP's and in the system used to transport the collected PM
to the saltcake mix tank. Also, for wet ESP systems, the PM may
be collected as a dry product in the bottom of the ESP (i.e., a
dry-bottom ESP) but black liquor or HAP-contaminated water is
used in the PM return system. Because there is contact between
the hot recovery furnace flue gases and the black liquor or HAP-
contaminated process water in wet ESP systems, gaseous organic
HAP's are stripped from the liquor and emitted from the recovery
furnace stack. Most newer NDCE recovery furnaces (i.e., those
constructed after 1989) emit lesser quantities of methanol and
other gaseous organic HAP's because they tend to be equipped with
dry ESP systems (i.e., dry-bottom ESP's and dry PM return
systems).30
Data are not available to determine the effects of
converting a wet ESP system to a dry ESP system on overall
gaseous HAP emissions for a DCE recovery furnace system. Gaseous
organic HAP's that would have been stripped from the black liquor
in the wet ESP system potentially could be stripped from the
black liquor when it enters the DCE. Also, based on emission
2-41
-------
data for NDCE recovery furnaces with wet ESP systems, the ESP is
a less significant source of gaseous organic HAP emissions than
the BLO' or DCE. Data are available for NDCE recovery furnaces
with both wet ESP systems and dry ESP systems such that the
impacts that different ESP systems have on gaseous organic HAP
emissions from NDCE recovery furnaces can be determined. These
impacts can be more readily determined since the ESP represents
the only opportunity where stripping can occur for NDCE recovery
furnaces.
2.3.2.1.3 Nationwide gaseous HAP emissions. Table 2-5
presents the nationwide baseline gaseous HAP emissions from
recovery furnaces. The nationwide baseline total gaseous HAP
emissions from the estimated 128 NDCE recovery furnaces are
estimated to be 13,700 megagrams per year (Mg/yr) (15,100 tons
per year [ton/yr]). The nationwide baseline total gaseous HAP
emissions from the estimated 83 DCE recovery furnace systems
(which includes BLO systems) are estimated to be 15,700 Mg/yr
(17,300 ton/yr). The baseline emissions shown in Table 2-5 were
calculated assuming no control of gaseous HAP's (with the
exception of the two BLO systems with emissions controlled via
incineration) because gaseous HAP emissions from recovery
furnaces are highly variable and largely uncontrolled. The
baseline emissions were calculated by multiplying the gaseous HAP
emission factors by the process emission rate in kg BLS/hr
(Ib BLS/hr) for each mill for which data were available and then
extrapolating based on the available data to account for those
mills where data were unavailable. The baseline emissions were
also calculated based on the assumption that each emission source
operates 24 hours per day (hr/d) for 351 days per year (d/yr),24
2.3.2.2 Baseline PM and TRS Emissions. Table 2-6 presents
nationwide total PM, PM HAP, and TRS baseline emissions from each
combustion source. The HAP portion of total PM consists of HAP
metals and accounts for less than l percent of the total PM
emitted on average; therefore, only the total estimated quantity
of HAP metals are presented in Table 2-6, instead of presenting
emission estimates for each individual HAP metal.31 The baseline
2-42
-------
TABLE 2-5a (Metric). NATIONWIDE BASELINE GASEOUS HAP
EMISSION ESTIMATES24
Gaseous HAP
Acetaldehyde
Benzene
Formaldehyde
Methyl ethyl ketone
Methyl isobutyl ketone
Methanol
Phenol
Toluene
Xylenes
Hydrochloric acid
Total gaseous HAP's
Nationwide emissions, Mg/yr3
DCE recovery furnace systems
Air-sparging BLO units
333
27.3
6.5
95.3
39.6
3,890
43.3
32.3
96.1
N/A
4,560
DCE recovery furnaces
769
373
72.6
220
312
4,390
1,020
251
637
3,090
11,100
NDCE recovery
furnaces
588
454
525
376
466
2,120
924
422
968
6,840
13,700
aThe emissions in this table were calculated by multiplying the emissions in Table 2-5b by 0.9072.
TABLE 2-5b (English). NATIONWIDE BASELINE GASEOUS HAP
: EMISSION ESTIMATES24
Gaseous HAP
Acetaldehyde
Benzene
Formaldehyde
Methyl ethyl ketone
Methyl isobutyl ketone
Methanol
Phenol
Toluene
Xylenes
Hydrochloric acid
Total gaseous HAP's0
Nationwide emissions, ton/yra
DCE recovery furnace systems
Air-sparging BLO units
367
30.0
7.2
105
43.7
4,290
47.8
35.6
106
N/A
5,030
DCE recovery furnaces
848
411
80.0
242
343
4,840
1,120
276
702
3,410
12,300
NDCE
recovery
furnaces
648
501
578
414
513
2,330
1,020
465
1,070
7,540
15,100
nationwide emissions were calculated using the following equation:
Emissions, ton/yr = (emission factor, Ib/lb BLS) x (nationwide BLS firing rate, Ib/hr) x
, (35_I d/yr) x (1 ton/2,000 Ib).
Emlssiozss from the two BLO units that are controlled via incineration were multiplied by
for aa estimated 98 percent emission reduction.
cNumbers may not add exactly due to rounding.
(24 hr/d) x
0.02 to account
2-43
-------
PM and TRS emissions were calculated based on the sum of the PM
or TRS emission rates calculated for each combustion source at
each mill operating at the U.S. The baseline emissions were also
calculated based on the assumption that each combustion source
operates 24 hr/d for 351 d/yr.32
TABLE 2-6. NATIONWIDE BASELINE PM, PM HAP, AND TRS EMISSIONS
Combustion source
NDCE recovery furnaces
DCE recovery furnaces
Smelt dissolving tanks
BLO units
Lime kilns
Total0
Nationwide emissions, Mg/yr (ton/yr)
Total PM
29,400 (32,400)
16,000 (17,600)
7,680 (8,470)
—
8,830 (9,730)
61,900 (68,200)
PM HAP'sa
54(65)
31 (35)
5(5)
—
124 (136)
219 (241)
TRSb
658 (725)
2,720 (3,000)
7,350 (8,100)
658 (725)
648 (714)
12,000 (13,300)
aOn average, PM HAP's account for 0.2 percent of total PM emitted from recovery furnaces,
0.06 percent of total PM emitted from SDT's, and 1.4 percent of total PM emitted from lime kil
bAs H2S.
°Numbers may not add exactly due to rounding.
2.4 REFERENCES FOR CHAPTER 2
1. Green, R. and G. Hough (eds.). Chemical Recovery in the
Alkaline Pulping Processes. 3rd Edition. Prepared by the
Alkaline Pulping Committee of the Pulp Manufacture Division.
Atlanta, GA, TAPPI Press. 1992. 196 p.
2. Smook, G. Handbook for Pulp and Paper Technologists.
Montreal, Quebec, Canada. Canadian Pulp and Paper
Association. Atlanta, GA, TAPPI Press. 1987.
3. Memorandum from Nicholson, R., MRI, to Telander, J., ,
EPA/MICG. June 13, 1996. Addendum to Summary of Responses
to the 1992 NCASI "MACT" Survey.
4. Lockwood Post's Directory of the Pulp, Paper and Allied
Trades, 1993. San Francisco, Miller Freeman Publications.
1993. 966 p.
2-44
-------
5. Pulp, Paper, and Paperboard Industry--Background Information
for Proposed Air Emission Standards, Manufacturing Processes
at Kraft, Sulfite, Soda, and Semi-Chemical Mills. U. S.
Environmental Protection Agency, Research Triangle Park, NC.
Publication No. EPA-453/R-93-050a. October 1993.
6. Memorandum from Nicholson, R., MRI, to the project file.
September 20, 1996. Semichemical Pulping Operations Co-
Located at Kraft Pulp Mills.
7. U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 43 FR 7568.
Washington, DC. U.S. Government Printing Office.
February 23, 1978.
8. U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 51 FR 18538.
Washington, D.C. U.S. Government Printing Office.
May 20, 1986.
9. Review of New Source Performance Standards for Kraft Pulp
Mills. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/3-83-017.
September 1983. p. 4-7.
10. Stultz, S. and J. Kitto (eds.). Steam: Its Generation and
Use. 40th Edition. Babcock & Wilcox. New York. 1992.
Chapter 26.
11. Someshwar, A. and J. Pinkerton. Wood Processing Industry.
In: Air Pollution Engineering Manual, Air & Waste Management
Association. Buonicore, A. and W. Davis (eds.). New York,
Van Norstrand Reinhpld. 1992. pp. 835-849.
12. Proposed Standards of Performance for Kraft Pulp Mills. In:
Standards Support and Environmental Impact Statement.
Volume 1. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/2-76-014-a.
September 1976.
13. Environmental Pollution Control: Pulp and Paper Industry,
Part I, Air. U. S. Environmental Protection Agency.
Cincinnati, OH. Publication No. EPA-625/7-76-001.
October 1976.
14. Memorandum from Soltis, V., Nicholson, R., and Holloway, T.,
MRI, to Telander, J., EPA/MICG. July 29, 1994. Summary of
Responses to the NCASI "MACT" Survey--Kraft and Soda Pulp
Mills.
15. Barsin, J. Process Recovery Boiler Design Considerations.
, Babcock and Wilcox. (Presented at the Kraft Recovery
Operations Seminar. Orlando, FL. February 10-14, 1986.)
2-45
-------
16. Kirby, M. Economic and Process Considerations in the Use of
Oxygen for Black Liquor Oxidation. Union Carbide Canada
Limited. Ontario, Canada. (Presented at the 21st Pulp and
Paper ABCP Annual Meeting. Sao Paulo, Brazil.
November 21-25, 1988.) 10. p.
17. Cooper, H. TRS Control Implications Reviewed, Part II.
Southern Pulp and Paper. April 1981. p. 42-42.
18. Memorandum from Soltis, V., MRI, to the project file.
April 3, 1995. Kraft and Soda Pulp Mill Combustion Sources
Data Base.
19. Walther, J. and H. Amberg. Crown Zellerbach Corp. A
Positive Air Quality Control Program at a New_Pulp Mill.
Journal of the Air Pollution Control Association.
20(1):9-18. January 1970.
20. Reference 1, p. 156.
21. Memorandum from Ramsey, M. , MRI, to Telander, J. , EPA/ISB.
January 27, 1992. Meeting minutes from the August 1, 1991
meeting between National Council of the Paper Industry for
Air and Stream Improvement, Inc., American Paper_Institute,
the U. S. Environmental Protection Agency, and Midwest
Research Institute.
22 Memorandum from Subramarian, S. and Nicholson, R., MRI, to
the project file. September 13, 1993. State Regulations on
Particulate Matter Emissions from Kraft Lime Kilns, Recovery-
Furnaces, and Smelt Dissolving Tanks.
23. Memorandum from Gideon, L. and Watkins, S., Radian Corp., to
Lassiter, P., EPA/CPB. January 22, 1992. Summary of
Regulatory Survey for the Pulp and Paper NESHAP.
24. Memorandum from Randall, D., Jones, R., and Nicholson, R.,
MRI, to Telander, J., EPA/MICG. April 25, 1995. Nationwide
Baseline Emissions for Combustion Sources of Kraft and Soda
Pulp Mills.
25. Someshwar, A. and A. Jain. Emissions of Hydrochloric Acid
from Kraft Recovery Furnaces. National Council of the_Paper
Industry for Air and Stream Improvement, Inc. Gainesville,
FL. (Presented at the 1992 TAPPI International Chemical
Recovery Conference. Seattle, WA. June 18, 1992.) 14 p.
26. Warnqvist, B. and H. Norrstrom. Chlorides in the Recovery
Boiler and Mechanism for Chloride Removal. TAPPI.
59(11):89. 1976.
2-46
-------
27. Someshwar, A. A Study of Kraft Recovery Furnace
Hydrochloric Acid Emissions. National Council of the Paper
Industry and Air and Steam Improvement, Inc. New York.
Technical Bulletin No. 674. August 1994.
28. Memorandum from Randall, D., MRI. to the project file.
November 8, 1994. Comparison of EPA and NCASI HC1 Emission
Factors for Recovery Furnaces.
29. Telecon. Ramsey, M., MRI, with Morris, T., Westvaco Corp.,
Covington, VA. June 17 and 25, 1993. Supplemental
information for Westvaco emission test reports.
30. Memorandum from Nicholson, R., MRI to the project file.
March 31, 1995. Methanol Emissions from Kraft NDCE Recovery
Furnaces.
31. Memorandum from Holloway, T., MRI, to the project file.
June 14, 1996. Summary of PM and HAP Metals Data.
32. Memorandum from Holloway, T., MRI, to the project file.
May 20, 1996. Nationwide Baseline PM and TRS Emissions.
2-47
-------
-------
3.0 EMISSION CONTROL TECHNIQUES
This chapter discusses emission control techniques for
reducing HAP emissions from recovery furnaces, BLO systems,
SDT's, and lime kilns located at kraft pulp and paper mills.
Recovery furnaces include DCE recovery furnaces and NDCE recovery
furnaces. Add-on controls and equipment changes/modifications
are discussed in Sections 3.1 and 3.2, respectively. Both PM and
gaseous HAP control techniques are included in this discussion.
Section 3.3 contains the references cited in this chapter.
Tables 3-1 and 3-2 present applicable control techniques for
reducing PM and gaseous HAP emissions, respectively, from the
aforementioned combustion sources located at kraft pulp and paper
mills. These control techniques and their effectiveness in
reducing PM and gaseous HAP emissions are discussed below.
3.1 ADD-ON CONTROLS
This section describes the various add-on air pollution
control equipment that can be applied to the combustion sources
located in the chemical recovery areas at kraft pulp and paper
mills. Sections 3,1.1, 3,1.2, and 3.1.3 describe ESP's, wet
scrubbers, and BLO control, respectively; each section presents a
description of the control equipment and describes the
performance of the equipment in controlling HAP (PM and/or
gaseous) emissions from the applicable combustion sources.
3.1.1 Electrostatic Precipitators
Electrostatic precipitators are a demonstrated control
technique for reducing PM emissions from kraft recovery furnaces
and lime kilns. The PM emissions from over 98 percent of kraft
recovery furnaces and approximately 10 percent of kraft lime
kilns are controlled with ESP's.1 A general description of the
3-1
-------
TABLE 3-1. PARTICULATE MATTER HAP CONTROL TECHNIQUES
Source
Recovery furnace
Lime kiln
SDT
Control technique
ESP
ESP
Venturi scrubber
Wet scrubber
Percent PM reduction
99 +
99+
99 +
99+
TABLE 3-2. GASEOUS HAP CONTROL TECHNIQUES
Source
NDCE recovery
furnace systems
DCE recovery furnace
system (includes air-
sparging BLO system)
Lime kiln
SDT
Control technique
Packed-bed scrubber
Elimination of black liquor
from ESP system
Packed-bed scrubber
Conversion to NDCE recovery
furnace with dry- ESP system0
Incineration of BLO vent gases
Control makeup water quality
in causticizing area
Control makeup water quality
in causticizing area
Percent reduction
HC1
Up to 99% or
5 ppmva
—
Up to 99% or
5 ppmva
_d
_e
Other HAP's
_
72%
_
93%
38%*
_g
_g
aControl efficiency depends on inlet HC1 concentration. Based on manufacturers' guarantees,
scrubbers can achieve either a 99 percent control efficiency or an outlet concentration of 5
ppmv.
^Insufficient data are available to determine the emission reductions of other HAP's.
cConversion includes replacement of the DCE recovery furnace with an NDCE recovery furnace
, and elimination of the BLO unit. . . .
dHCl emissions are highly variable. Available data indicate no overall change in HC1 emissions
is expected from converting a DCE recovery furnace system to an NDCE recovery furnace.
?HCl is not emitted from BLO systems.
fThe overall percent reduction is based on average HAP emission factors for DCE recovery
furnaces and air-sparging BLO systems, and on a control efficiency of 98 percent for HAP s
emitted from the BLO system.
SWastewater control option performance is addressed in a separate wastewater regulatory
development program.
3-2
-------
types of ESP's''"used to control PM emissions from kraft recovery
furnaces and lime kilns is provided in Section 3.1.1.1, below.
The performance of these ESP's'is discussed in Section 3.1.1.2.
3.1.1.1 Description. The ESP's used to control PM
emissions from kraft recovery furnaces and lime kilns are
generally classified as plate-wire ESP's. In plate-wire ESP's,
the flue gas flows between parallel sheet metal plates and high-
voltage electrodes. The flue gas passes between collecting
plates into a field of ions that have been negatively charged by
o
the high-voltage electrodes located between the plates. Each
paired set of electrodes and plates forms a separate
electrostatic field within the ESP. Electrostatic precipitators
used to control PM emissions from kraft recovery furnaces
typically have two parallel precipitator chambers (i.e., flu'e gas
passages) with three or four electrostatic fields per chamber.
Lime kiln ESP's typically have one chamber with two or three
electrostatic fields.1
As the flue gas passes through each electrostatic field, the
particles suspended in the flue gas are bombarded by the ions,
imparting a negative charge to the particles. The negatively
charged particles then migrate towards the positively charged or
grounded "collecting" plates, where the particles transfer a
portion of their charge, depending upon their resistivity. The
particles are kept on the collecting plates by-*the electrostatic
field and the remaining charge. At periodic intervals, the
collection plates are knocked ("rapped"), and the accumulated PM
falls into the bottom of the ESP. The recovered PM is
subsequently recycled to the black liquor in recovery furnace
applications or,-in lime,kiln applications, fed back to the kiln.
The ESP's used on recovery furnaces may be designed with either a
wet or dry bottom. In wet-bottom ESP's, the collected PM falls
directly into a pool of liquid, which may be black liquor or
water, in the bottom of the ESP. In dry-bottom ESP's, the
collected PM falls to the (dry) bottom of the ESP and is
transferred from the ESP bottom to a mix tank (containing black
liquor) via drag-chain or screw 'conveyors. Black liquor is
3-3
-------
sometimes used to transport the dry collected PM to the mix tank.
More recent ESP installations employ a dry PM return system to
transport the PM to the mix tank. Because the PM removed by the
ESP is recycled to the black liquor in the mix tank, the ESP is
an integral part of the chemical recovery loop as well as an air
pollution control device.
The design of the plate-wire ESP's used to control PM
emissions from recovery furnaces and lime kilns may include
either weighted-wire electrodes cr rigid electrodes. With the
weighted-wire design, the wire electrodes are suspended inside
the ESP, and weights are attached to the wires to maintain
tension. In the rigid-electrode design, the discharge electrodes
are rigid tubes with pointed corona emitters welded to the
surface; each tube replaces two weighted wires. Although the
weighted-wire design has been available for more than 50 years,
rigid-electrode ESP's have only been available since the late
1970's. The rigid-electrode ESP represents the current stage of
development in ESP technology and offers the following advantages
over the weighted-wire design: a higher tolerance of in-service
abuse (no wires to break), better collection efficiencies, and
better cleaning characteristics.4 According to one manufacturer,
all new ESP installations in the pulp and paper industry since
1990 have rigid-electrode designs.5 Additionally, since 1981, 64
of the 66 (97 percent) recovery furnace ESP installations
supplied by this manufacturer have been of the rigid-electrode
design, and approximately 80 percent of the lime kiln ESP
applications use rigid-electrode designs.5 Figure 3-1 is a
schematic of a rigid-electrode, plate-wire ESP.
The average lifetime of an ESP in service on a kraft
recovery furnace varies depending upon the type of ESP bottom
(i.e., wet vs. dry), the inlet temperature of the gases, and
maintenance practices. The lifetime of ESP's used to control PM
emissions from recovery furnaces with NDCE's, which tend to
operate with dry-bottom ESP's, typically ranges from 12 to
15 years.4 After that point, major repairs or a rebuild may be
required. Recovery furnaces with DCE's tend to have cooler inlet
3-4
-------
Figure 3-1. Rigid-electrode ESP.
3-5
-------
gases and wet-bottom ESP's; these two factors promote corrosion
through condensation of acid gases and shorten the life of the
ESP. The lifetime of an ESP on a DCE recovery furnace typically
ranges from 7 to 10 years.4
Lime kiln ESP's operate in a "milder" environment (i.e.,
hotter temperatures prevent any acid gases from condensing and
corroding the ESP, and the primary constituent of the PM
collected is lime, which creates an alkaline environment that
further protects the ESP from acid gas corrosion). Therefore,
lime kiln ESP's have fewer corrosion problems than do recovery
furnace ESP's. The expected lifetime of a lime kiln ESP is
typically more than 15 years.6
The size of the ESP is often expressed in terms of the
specific collecting area (SCA), which is defined as the total
collecting plate surface area divided by the flue gas flow rate.
Specific collecting areas of ESP's used to control PM emissions
from kraft recovery furnaces and lime kilns typically range from
about 39 to 160 square meters per cubic meter per second
(m2/[m3/sec]) (200 to 800 square feet per 1,000 actual cubic feet
per minute [ft2/!,000 acfm]).1 The SCA's of ESP's used to
control emissions from DCE recovery furnaces tend to be somewhat
lower than those associated with NDCE recovery furnaces. The
average SCA for ESP's used to control emissions from DCE recovery
furnaces is approximately 15 percent lower than for NDCE recovery
furnaces.1 The primary reason for the difference is that the DCE
removes 20 to 40 percent of the PM before the ESP; therefore, the
inlet loading of PM to the ESP operating on a DCE recovery
furnace is lower than that of an NDCE recovery furnace.7 The
lower SCA's for DCE recovery furnaces may also be because DCE
recovery furnaces tend to be older than NDCE recovery furnaces,
and most are not subject to the NSPS for Kraft Pulp Mills.
3.1.1.2 Performance. Electrostatic precipitators can
control PM HAP emissions but provide no control of gaseous HAP
emissions. Properly designed and operated ESP's used on kraft
recovery furnaces and lime kilns routinely achieve PM removal
efficiencies of 99 percent or greater. Although emission test
3-6
-------
data from recovery furnace ESP's on PM HAP performance are
limited, data collected from other combustion sources on the
relative performance of ESP's for PM and metals indicate that
those systems that achieve the greatest PM removal also provide
the best performance for the HAP portion of the PM. Therefore,
total PM performance can be used as a surrogate for PM HAP's.
Because performance guarantees made by ESP manufacturers and
permit limits are usually in the form of outlet PM emission
concentrations rather than in the form of achievable control
efficiencies, outlet concentrations will be the focus of the
discussion of the performance of ESP's on kraft recovery furnaces
and lime kilns in the following sections. Following the format
of the NSPS, all PM concentrations for recovery furnace emissions
are corrected to 8 percent (%,, and those for lime kiln emissions
are corrected to 10 percent O2. The discussion below also
addresses the factors that affect ESP performance.
3.1.1.2.1 Recovery furnaces. The PM emitted from the
recovery furnace and subsequently recovered by the ESP is
primarily composed of sodium sulfate (i.e., "saltcake") and
sodium carbonate, neither of which are HAP's. Particulate matter
HAP's, which are primarily metals (although small quantities of
high-boiling-point semivolatile organic compounds also may be
emitted in particle form), are emitted from the ESP in small
quantities. For any single metal, the quantity emitted from
either NDCE or DCE recovery furnaces is typically less than
140 kg/yr (300 Ib/yr).9 The paragraphs below first discuss the
outlet concentrations achievable by ESP's based on information
compiled as a part of the previous NSPS development effort and on
long-term performance data compiled by the State of Washington.
Then factors that can affect ESP performance are described.
The NSPS for kraft recovery furnaces requires that PM
emissions from recovery furnaces constructed, reconstructed, or
modified after September 24, 1976 be less than or equal to
0.10 grams per dry standard cubic meter (g/dscm) (0.044 grains
per dry standard cubic foot [gr/dscf] of flue gas).10'11 Based
on installation dates, approximately 39 percent of NDCE recovery
3-7
-------
furnaces (i.e., 50 NDCE recovery furnaces) and 4 percent of DCE
recovery furnaces (i.e., 3 DCE recovery furnaces) are subject to
the NSPS.12 In addition, some NDCE and DCE recovery furnaces not
subject to the NSPS are subject to State PM permit limits that
are less than or equal to the NSPS limit.13 Therefore, about
25 percent of the 211 existing recovery furnaces nationwide are
required by regulation to emit PM in concentrations less than or
equal to 0.10 g/dscm (0.044 gr/dscf).
The NSPS emission limit was established based on the
performance of a weighted-wire ESP design, taking into
consideration maintenance issues such as wire breakage, the
corrosive and sticky nature of the PM removed, any long-term
deterioration in ESP performance, and the inherent variability of
ESP performance.14 In considering these issues, EPA found that
ESP's that formed the basis for the standard were installed with
added insulation or heated shells to maintain the gas temperature
throughout the precipitator to prevent corrosion.14 Also,
although some amount of wire breakage was noted, most occurred
soon after start-up and lessened in frequency with more operating
time. A loss of 5 to 10 percent of the wires was determined to
have no noticeable effect on the performance of the ESP.
Finally, about long-term deterioration in ESP performance,
manufacturers of ESP's were emphatic in stating that a properly
maintained precipitator should not deteriorate over the expected
life of the unit.14 Based on a review of all available
information collected from both the industry and control device
manufacturers, ESP systems were determined to be capable of
meeting an emission limit of 0.10 g/dscm (0.044 gr/dscf) on a
continuous, long-term basis.14
More recent information obtained from ESP manufacturers
indicates that since the NSPS was promulgated, the industry has
demanded further improvements to ensure continued compliance with
PM emission limits associated with the NSPS or State permit
requirements. Mills often require that ESP manufacturers provide
emission limit guarantees such that the furnace can meet the
permitted limit (e.g., 0.10 g/dscm [0.044 gr/dscf]) with only one
3-8
-------
chamber .of the ESP in operation. Two ESP manufacturers commonly
guarantee that the permitted emission limit will not be exceeded
while the recovery furnace is operating at 70 percent capacity
with one chamber of the ESP in operation.4'6 Requests by mills
for these types of guarantees began in the late 1980's and have
resulted in the installation of ESP's that are sized to ensure
that these emission limits are achieved continuously. This
design provides mills with the advantage of being able to make
ESP repairs to one chamber of the ESP without disrupting the
operation of the furnace, and thus avoiding production losses
while maintaining the integrity of the ESP. In addition, ESP's
that meet these design criteria routinely achieve PM outlet
emissions at levels less than or equal to half the NSPS (i.e.,
0.050 g/dscm [0.022 gr/dscf]) under normal operating conditions.4
The information supplied by manufacturers is supported by
long-term PM emission test data for recovery furnace ESP's
supplied by the State of Washington.13 Figures 3-2 through 3-4
show monthly PM emission data for two kraft NDCE recovery
furnaces with older ESP's over a 6-year period from 1988 to 1994
and one kraft NDCE recovery furnace with a newer ESP over a
5K-year period from 1990 to 1995. The figures provide an
indication of concentrations achievable by older systems designed
to meet the NSPS limits and by newer ESP's with improved design.
The data also demonstrate the variability in ESP performance.
The data in Figure 3-2 were generated by an ESP of weighted-
wire design that was designed to comply with a permit limit of
0.10 g/dscm (0.044 gr/dscf). The ESP was installed in 1974 and
has a design and operating SCA of about 92 m2/(m3/sec)
(467 ft2/!,000 acfm).15
The data in Figure 3-3 were generated by another ESP of
weighted-wire design that was designed to comply with a permit
limit of 0.10 g/dscm (0.044 gr/dscf). The recovery furnace was
originally installed as a DCE recovery furnace in 1975 and
converted to an NDCE recovery furnace in 1982.15 The ESP was
last upgraded in 1982 and has a design and operating SCA of about
75 m2/(m3/sec) (383 ft2/!,000 acfm).15 The control system is
3-9
-------
UJ
o
ie
oiH
3^
35 Q-
CO
UJ
O
*=*. r*> fr-
li?
" lu£
UJ
u_
O
_
«>-
P-
CO
Q
O
— °?
U. <
. u_
S o
O
a co
co h-
g£
g
CO
Q
UJ
O
o
¥
d
IT
TT
q
o
p
d
- >-
- O
.« ^
"^ H
: o
2
• _J
- Q.
-~ <
•c
-co
-co
CO
-co
Q)
O
a)
o
o
(U
rt
•d
c
o
-H
CD
CD
-H
CN
ro
0)
^
tn
-H
(dosa/yo)
3ivinoiiuvd
3-10
-------
lO
LU IS
O 2
cc
LU
m
m
13
CC
o
CO
So
.. CO
LU CO
*£ cc
Q ^
O CO
CO LU
LU
O
cc
Z)
u_
LU
cb
O <
LU 1? >
D- pC <>
leJ
O m
O DC
Q Q
°D
O
O CL
LU
O
oc
LU
CD
cc
o
CO
HI
CO
CO
LU
o
o
CO
o
o
o
LU
1- CO
CC LU
CC LU
Q O
-
CO <
LU O
in
o
o
CC
3
-en O
- I—
•c -
.§ CO
en
-co
-c
CO
-CO
O
O
-H
S
(ti
-------
E OF FURNACE: NDCE
NACE INSTALLED: 1956
Q. CC
— d
O
CD Q-
T. LL
Q CC
UJ CC
Q —
Q cr
< c
CC .j
,- < >
co >- C
7! < cc
LU h- CC
h- CC C
< LU ff
/-ODOR CONVERSION D
ACITY INCREASE AND T
D: DRY-BOTTOM FSP/OF
^ CL C
O < CL
ri o <
1 C
*
CO
LU
Q
-3
O
.
CC
LU
CO
CD
"™v
OSSFLOW SCRL
DC
0
INSTALLED: 1989
D_
CO
LU
5 i»
^~
^^
DC' cc
UJ Cu CC
m co z
m i O
Sfecj
0 >• UJ
CO UJ CC
Q b uj
LU ^i I
CQ Q_ H-
i u- r
Q Q 2
*£LL
2-SECTION PAC
WITH PERFORA'
SHOWFR RFTW
x— «
^
u_
O~< Q
LLJ 0 UJ
w § co
ID pr ^j
O u_ (j
. CO ^
o o d
co o oo
h- CO
< 7
0 <
CO O
0 cc
CO LU
LU D_
Q O
d d
_i _i
< <
0 0
c
CC
LU
f—
>
^*
•z.
CO,
05 0
CO I
Ox ^^
T~ «2!
S CD
— to
RUBBER INSTAL
ESSURE DROP:
O CC
CO 0-
^"
•Lu
O
•^
O
o
LT
d
CO
o
(M
D L
•) ^
r o
0 0
•
1
,^--«
k^_
•K—
^;B
W£~
\ BB^^^^
^;B
Jj[
•
i
"m^.
i
••
H
•™
I
i?
r1
i
JP
%
%
••
f
*-m
J
I
i
1
^
J i- C
in
-co
— "Q
~"O
—
—
—
LO ,
~~ O) M*\
"o.
~< H
~ r-l
- -H
~^f S
-o>
o ^
ro
co O
{pm
< 0)
co «< fl
9 UJ >-i
"eo ^ d)
12 § ^
I^"^"" m
l-i
•eE
~(N fZ ^
=§CD 5
~"O ~2L
~ i (tf
^ ^ TM
— Cn •e— • *-*
-2 < rt
_ 9- CO 5
< o
-H
- CD
m
—CO -H
— ^ 6
:
-------
augmented with- a combination low-energy venturi scrubber and
packed-bed scrubber in series installed in 1984.15 The scrubbers
provide some added PM removal and act as a heat-transfer unit "to
provide hot•water.1°
The data in Figures 3-2 and 3-3 indicate that although
emissions are quite variable, these units can readily meet the
NSPS limit of 0.10 g/dscm (0.044 gr/dscf) on a long-term,
continuous basis.
Outlet PM emission concentrations at least as low as
0.002 g/dscm (0.001 gr/dscf) for recovery furnaces with newer
ESP's have been documented in the Washington data as well as in
the data provided by individual pulp mills.13'15 However, the
ability of an ESP to maintain emissions on a continuous basis at
levels substantially below 0.034 g/dscm (0.015 gr/dscf) has not
been demonstrated.13 However, the data in Figure 3-4 do
demonstrate that a system can readily meet a level of
0.034 g/dscm (0.015 gr/dscf) on a long-term, continuous basis.
These data were generated by an NDCE recovery furnace with a
rigid electrode ESP followed by a packed-bed scrubber. The
recovery furnace was originally installed as a DCE recovery
furnace in 1956 and converted to an NDCE recovery furnace in
1981.1S Both the ESP and the scrubber were installed in 1989.15
While the data in Figure 3-4 for recovery furnace No. 3 at
Mill B are somewhat variable, they show considerably less
variability and consistently lower emissions (less than
0.034 g/dscm [0.015 gr/dscf]) than the data associated with the
two older ESP's. The ESP on recovery furnace No. 3 at Mill B was
designed to achieve an outlet concentration of 0,069 g/dscm
(0.033 gr/dscf). Furthermore, although the system was designed
with an SCA of 79 m2/(m3/sec) (400 ft2/!,000 acfm), the furnace
has operated at about 60 to 70 percent capacity over its life.16
Consequently, the operating SCA of the unit is estimated to be in
the range of 110 to 130 m2/(m3/sec) (570 to 670 ft2/!,000 acfm)
or 120 m2/(m3/sec) (620 ft2/!,000 acfm), on average. These data
demonstrate that systems comprising a recent ESP with an SCA of
about 120 m2/(m3/sec) (620 ft2/!, 000 acfm) followed by a packed-
3-13
-------
bed scrubber can achieve a maximum 'outlet concentration of
0.034 g/dscm (0.015 gr/dscf) on a long-term, continuous basis.
No single parameter in and of itself is a good predictor of
ESP performance. Both the design of the ESP and operating
variables affect the performance of the ESP. Although the SCA
may provide some indication of the expected efficiency of the
ESP, other design features, such as the location of the rappers
and plate spacing (which differs from one ESP manufacturer to
another), also affect the ESP's performance. For example,
differences in plate spacing from one manufacturer to another
have resulted in different manufacturers requiring different
SCA's for the same outlet PM concentration.2'4 Another factor
that complicates assessment of the effect of SCA on performance
is that the PM emission concentrations for newer ESP's may be
measured based on gas flow through one chamber with the recovery
furnace operating at about 70 percent capacity.4 The mechanism
used to report emissions and calculate the SCA (i.e., based on
gas flow at 70 or 100 percent capacity) can mask any relationship
between the SCA and PM emissions.4 Consequently, available data
for recovery furnaces show only a limited relationship between
outlet concentration and SCA.
The power input to the ESP can be a useful parameter in
monitoring ESP performance. The value of the power input for
each field and for the total ESP indicates how much work is being
done to collect the PM. The power supply to the ESP consists of
four basic components: (1) a step-up transformer (to step up the
line voltage), (2) a high-voltage rectifier (to convert
alternating current [a.c.] voltage to direct current [d.c.]),
(3) a control element, and (4) a control system sensor.17 Most
ESP's installed since the 1980's are equipped with primary-
voltage and current meters on the low-voltage (a.c.) side of the
transformer and secondary voltage and current meters on the high-
voltage rectifier (d.c.) side of the transformer.17 The power
input for each transformer-rectifier (T-R) set is calculated by
multiplying the voltage and current values of either the primary
or secondary side of the transformer. The ESP control system is
3-14
-------
designed to provide the highest possible voltage without causing
sustained sparking between the discharge electrode and the
collection surface. Spark meters" can be used to set the maximum
voltage/current levels for efficient operation. When the
resistivity of the particulate is normal to moderately high, ESP
performance improves as the total power input increases.
However, when particulate resistivity is either low or very high
or when spark rates are very high, improving the performance of
the ESP may require decreasing the power input.17 In addition,
newer, high-efficiency ESP's that are generously sized and
sectionalized are relatively unaffected by changes in power
input. The normal power input of some of these newer and larger
ESP's may be reduced by one-half to two-thirds without
substantially affecting performance.17 Therefore, power input
cannot be used as a sole indicator of ESP performance.
Other operating variables that reportedly impact the
performance of the ESP include temperature of the flue gas and
maintenance practices. When flue gas temperatures exceed 260°C
(500°F), the ESP's PM collection efficiency starts to decline.
At low temperatures (less than 150°C [300°F]), the ESP may
perform as desired, but corrosion will be accelerated (due to
condensation of acid gases).4 To obtain optimal control on a
continuous basis, the system should be operated within this
temperature envelope. «
Poor operating and maintenance practices may also result in
a decline in ESP performance, and differences in operating and
maintenance practices are expected to contribute to the
variability in emissions from mill to mill and over time at the
same mill. Repairs (e.g., replacing broken wires) are often
delayed, and routine on-line maintenance is sometimes neglected.
Mills commonly solicit bids for maintenance and repair work from
ESP manufacturers and contracting firms that provide ESP
maintenance services.4
3.1.1.2.2 Lime kilns. Although venturi scrubbers have
traditionally been the most common PM control device used on
kraft lime kilns, the use of ESP's to control PM emissions from
3-15
-------
lime kilns has steadily increased since about 1980. The ESP is
generally mounted on top of the lime kiln feed building, and the
captured dry PM is rerouted to the kiln by gravity feed.18 The
trend towards ESP's as PM control devices at new lime kiln
installations and as replacement control devices for older
scrubbers is primarily related to the lower energy costs, reduced
maintenance, and increased reliability of the ESP's in comparison
to venturi scrubbers that provide equivalent control. An added
benefit is that lime kiln ESP's produce a dry product that can be
recycled directly to the kiln. The wastewater produced by the
venturi scrubber is typically recycled to the mud washers before
the kiln to recover the lime particulate in the spent scrubbing
fluid. Additional energy is needed to remove the excess water in
the lime mud filter and to complete evaporation in the kiln.
(Additional information about venturi scrubbers is provided in
Section 3.1.2.2.)
The NSPS for kraft lime kilns requires that PM emissions
from gas-fired and oil-fired lime kilns constructed,
reconstructed, or modified after September 24, 1976 be less than
or equal to 0.15 and 0.30 g/dscm (0.067 and 0.13 gr/dscf),
respectively.10'11 Based on installation dates, approximately
28 percent of existing lime kilns are subject to the NSPS.
The PM emissions reported by the industry for lime kiln
ESP's show less variability month to month and kiln to kiln than
those associated with recovery furnace ESP's. Point-in-time PM
data were reported for nine facilities with ESP's controlling
lime kilns subject to the NSPS; these PM data showed outlet
concentrations ranging from 0.0057 to 0.059 g/dscm (0.0025 to
0.026 gr/dscf).15
These reported data are confirmed by monthly PM emission
data from the State of Washington for two gas-fired lime kilns
with ESP's at Mills C and D.13 Both kilns are subject to the
NSPS; the ESP's at Mills C and D were installed in 1982 and 1985,
respectively.15
As shown in Figure 3-5, the PM emission data from the kiln
at Mill C vary from 0.0025 to 0.0519 g/dscm (0.0011 to
3-16
-------
CM
CO
O)
Q
UJ
CO
LU
u_
00
UJ <
CO o
•^ °
«§ 0_
O fr
§
CO
LU
s-
JJ Q ^
UJ O O.
r> a. o:
u_ < u.
O)
q
o
H-
ZD
o
3
CO
q
ci
CD
o
q
o"
CO
q
o
d3iivi/M aivinonyvd
3-17
-------
0.0227 gr/dscf) during the period from November 1988 to
April 1996. The SCA of the ESP on this lime kiln is
90 m2/(m3/sec) (460 ft2/l,000 acfm).15 The data for Mill C
confirm that ESP's on existing lime kilns can readily meet the
NSPS limit of 0.15 g/dscm (0.067 gr/dscf) on a long-term,
continuous basis.
The lime kiln at Mill D (shown in Figure 3-6) consistently
emitted PM at or below 0.018 g/dscm (0.008 gr/dscf) during the
7-year period from March 1989 to April 1996. The SCA for the
lime kiln ESP at Mill D is 220 m2/(m3/sec) (1,120 ft2/
1,000 acfm) .15 The monthly PM emission data from Mill D
demonstrate that new ESP's on lime kilns can achieve a maximum
outlet concentration of 0.023 g/dscm (0.010 gr/dscf) on a long-
term, continuous basis.
3.1.2 Wet Scrubbers
The following discussion focuses on the control of gaseous
HAP emissions using packed-bed scrubbers and PM HAP emissions
using venturi scrubbers. Section 3.1.2.1 discusses packed-bed
scrubbers as the primary mechanism of HC1 control; the control of
gaseous organic HAP's by packed-bed systems will also be
discussed. Section 3.1.2.2 discusses venturi scrubbers as a
means of controlling PM emissions. In addition to venturi
scrubbers, the pulp and paper industry uses other types of wet
scrubbers, such as impingement plate, cyclone, flooded disc, and
packed-bed, to control PM emissions from lime kilns and SDT's.
However, venturi scrubbers are the most prevalent type of wet
scrubber used to control PM from these sources, and the other
scrubber types do not control PM emissions more effectively than
venturi scrubbers.
3.1.2.1 Packed-Bed Scrubbers. Test data indicate that both
NDCE and DCE kraft recovery furnaces emit HC1 in the furnace
exhaust gas at concentrations as high as 100 parts per million by
volume (ppmv), corrected to 8 percent 02.19'20 These exhaust
streams also contain SO0 at concentrations as high as 979 ppmv,
&
corrected to 8 percent O2.1/19 Typically, facilities that use
add-on control measures for acid gas emissions use packed-bed wet
3-18
-------
in
CO
CD
Q
LJJ
CO
2
UJ
u_
O 11-
CO
O q
co £j
o "^
CO C\J
p
Q CO
la UJ
CO 0-
< CQ
-U • Q =f O
111 O Dl —I
Z) Q_ CO <
U_ < USJ O
o
o
q
-------
scrubbers. Although the packed-bed scrubbers on kraft recovery
furnaces are not specifically designed to reduce HC1 emissions,
HC1 control can be achieved as an ancillary benefit.
This section focuses on control of gaseous HC1 emissions
with packed-bed scrubbers. Packed-bed systems have been
installed to a limited extent on kraft recovery furnaces for a
variety of reasons, including acid gas (i.e., SO2) control and
hot-water generation (i.e., to recover heat from the flue
gas),21/22 They have also been demonstrated to achieve high
levels of HC1 control on other combustion systems. The remainder
of the section describes packed-bed scrubbing systems, presents a
brief summary of the basic operating principles of packed-bed
systems and their theoretical performance for both HC1 and SO2
emission reductions, examines packed-bed scrubber performance
data for kraft recovery furnaces and for other combustion
sources, and discusses existing packed-bed scrubber applications
for recovery furnaces.
. 3.1.2.1.1 Description. As illustrated in Figure 3-7, the
recovery furnace exhaust gases are ducted to the acid gas control
system after PM has been removed in the ESP. Acid gas scrubbers
control HC1 and S02 emissions by mass transfer of these
pollutants from the exhaust gas stream to a liquid scrubbing
solution, typically either water or a caustic solution with
water. Packed-bed scrubbers are the most frequently used
absorption device because they provide excellent liquid-gas
contact, leading to efficient mass transfer.23'24
Schematics of two standard types of packed-bed scrubbers,
the counter-flow design and the cross-flow design, are shown in
Figures 3-8 and 3-9. Typically, the gases pass through a quench
section before entering the scrubber to saturate the gas stream.
The gases then enter the inlet plenum, which is designed to
distribute the gases uniformly across the bed cross-section. As
the gases flow through the bed, they contact a liquid film
created by a flowing liquid that is distributed evenly throughout
the bed. After the gas stream exits the bed, it passes through a
mist eliminator to remove entrained droplets that may contain
3-20
-------
CD
5 9
CO 3
o
w
Q.
OT
u
S a \
m UJ \
si W
S§ /
2 » /
CL
w
LU
a
a
5
EC W
O <
CO O
111
UJ
O
<
IT
U-
CQ
(U
to
ttJ
Cn
m
td
OJ
Q)
o
4-)
0)
u
0)
a
o
o
•H
tn
fl
O
o
Q)
JJ
m
>i
03
rH
o
5-4
J-)
CJ
O
u
CD
nJ
Cn
•d
-H
u
I
n
-------
Mist eliminator
Liquid sprays
Packing
Figure 3-8. Schematic of a counterflow packed-bed scrubber.
3-22
-------
g-
O
CQ
•d
0)
T)
0)
^
O
(ti
a
o
I
CD
m
o
^
o
nj
O
O
-H
4J
tti
6
Q)
4^
U
CO
CTi
i
CO
Cn
-H
fa
3-23
-------
alkaline salts (generated as the acid gases are scrubbed from the
exhaust gas), residual absorbed acids, and PM that was not
removed by the ESP or the packed bed. The paragraphs below
describe the role of the packing and the scrubbing liquid in more
detail.
A packed-bed scrubber is filled with packing material that
is used to produce a large liquid film surface area for contact
between the gas stream and scrubbing liquid. Packing materials
are available in a variety of forms and materials, each having
specific characteristics with respect to surface area, pressure
drop, weight, corrosion resistance, and cost.25 Specific packing
"materials are selected by the manufacturer to conform with the
gas and liquid stream properties of a particular application.
The gas stream moves through the packed bed, contacting the
liquid film on the surface of the packing. In the counterflow
configuration, the gas moves upward, while the liquid moves
downward through the packed-bed section. In a cross-flow
configuration, the gas moves in a horizontal direction,
perpendicular to the liquid stream, which moves downward through
the packing. The countercurrent system is more efficient for
acid gas removal, but height limitations or particle loadings
that can lead to plugging sometimes dictate use of the cross-flow
*? P,
configuration. °
An aqueous scrubbing solution, which may be either water or
an alkaline solution, is used in acid gas scrubbers. Because HC1
is highly soluble in water, adequate control can often be
achieved with water alone. However, because the exhaust from
kraft recovery furnaces also contains S02, alkaline solutions are
expected to be used for this source. In addition to providing
substantially better S02 control and somewhat better HC1 control,
an alkaline solution typically is used to maintain a constant
scrubber liquid pH at or slightly below 7.0 to prevent both scale
buildup and corrosion of scrubber components. The scrubbing
liquid that is expected to be used for acid gas control is
caustic solution (sodium hydroxide [NaOH]), although sodium
carbonate (Na2CO3) and calcium hydroxide (Ca[OH]2) (slaked lime)
3-24
-------
also can be used. When the acid gases are absorbed into the
scrubbing solution, they react with alkaline compounds to produce
neutral salts. The rate of absorption of the acid gases is
dependent on the solubility of the acid gases in the scrubbing
liquid. The neutralization reactions are essentially
stoichiometric, i.e., the stoichiometric ratio of alkaline
compounds added to the system to that required for complete
neutralization of the acid absorbed into the scrubber solution is
essentially 1:1 in packed-bed scrubbers.26
3.1.2.1.2 Theoretical, performance of packed-bed scrubbers.
Packed-bed scrubbers are chemical engineering unit operations
that provide mass transfer between a soluble gas component (in
the case of kraft recovery furnaces, HC1, S02/ and possibly some
organic HAP's) and.a solvent liquid scrubbing solution (water or
an alkaline solution). The driving force for absorption is the
difference between the partial pressure•of the soluble compound
in the gas mixture and the vapor pressure of the solute gas in
the liquid film in contact with the gas. The performance of
the system is a function of the mass transfer rate of the
pollutant from the gas to the liquid solute and of the contact
time between the gas and the liquid. These properties form the
basis for the design equations that are typically used to
characterize the theoretical performance of gas absorption
systems like packed-bed scrubbers.
Because the pollutant concentration in the gas stream is
very dilute, several assumptions that simplify these design
equations considerably are reasonable. The waste gas is assumed
to operate as an ideal gas; the scrubbing solution is assumed to
be an ideal solution; heat effects associated with absorption are
assumed to be negligible; and the reaction of the HC1 and S02
with the alkaline material in the scrubbing solution is assumed
to be fast compared to the rate-limiting absorption step. Also,
because equilibrium characteristics of a counterflow scrubber are
much less complex mathematically than the characteristics of a
cross-flow tower, the theoretical analyses focus on the
counterflow tower.
3-25
-------
The design and performance equations for counterflow packed-
bed scrubbers are based on the Whitman two-file theory, which is
described in detail in reference 28.28 The key element of this
theory for the analysis of kraft recovery furnace packed-bed
scrubbers is that the system is considered to be gas-film
controlling or liquid-film controlling, depending on the
pollutant to be controlled and the scrubbing liquid. For the
recovery furnace exhaust stream, HC1 absorption is liquid-film
controlling, while both the liquid film and the gas film provide
appreciable resistance to S02 absorption.28 These system
properties simplify the theoretical assessment of the performance
of the system for HC1 in that only gas-phase transfer needs to be
considered. They also partially explain the difference in
performance of packed-bed scrubbers for S02 and HC1.
An estimate of the theoretical performance of a packed-bed
scrubber for HC1 emissions can be developed by manipulating the
system design equations.29"31 For a specific set of inlet
conditions, these design equations can be modified to obtain the
following expression for the efficiency of a packed-bed scrubber:
Eff = 1-Yo/Yi
= 1 -exp-(Z/HOG)
where yo is the outlet gas mole fraction, y.j_ is the inlet gas
mole fraction, Z is the packing height, and HQG is the height of
a gas transfer unit.
These theoretical analyses have two important implications
for the performance of packed-bed scrubbers in controlling HC1
emissions from kraft recovery furnaces. First, the theoretical
efficiency of the system is limited by the concentration that can
be achieved in the exhaust gas from the scrubber when it is in
equilibrium with the scrubber liquor feed. At the concentrations
found in kraft furnace recovery exhaust gases and scrubbers, the
1 "5 O
Henry's Law coefficient is of the order of 10"J or less. If
the scrubber liquor feed is maintained at a pH of 6.5 or greater,
the molar concentration of HC1 in the feed is on the order of
3-26
-------
6 x 10" . Theoretically, the limiting HC1 concentration in the
exhaust gas is well below 1 ppm. However, this theoretical limit
is based only on ideal mass transfer considerations and does riot
address nonuniformities in gas and liquid flows that may be found
in operating systems.
Second, if for a given system, the height of a transfer unit
changes between pollutants by a factor of k, the penetration
(i.e, 1 - efficiency) for the two pollutants changes
exponentially with 1/k. The reported range of manufacturer's HQG
for plastic packings is 0.18, to 0-.33 meters (m) (0.6 to 1.1 feet
[ft]) for water as a scrubbing liquid and 0.15 to 0.31 m (0.5 to
0.7 ft) for caustic as a scrubbing liquid. Note that the same
reference reports a range of 0.15 to 0.61 m (0.7 to 2.0 ft) for
scrubbing S02 with caustic. These data indicate that the height
of a transfer unit for SC>2 is 2 to 3 times that for HCl, so that
the penetration for S02 will be the square root or the cube root
of that for HCl or, equivalently, that the penetration for HCl
will be the square or the cube of the penetration for SOo. For
example, if a system achieves 99 percent control for HCl, (i.e.,
the penetration is 0.01), the control efficiency for SC>2 would be
expected to be in the range of 78 to 90 percent.
3.1.2.1.3 Performance of HCl scrubbers for kraft recovery
furnaces. To provide a basis for subsequent analyses of HCl
performance potential, a typical sample of emission test data on
HCl concentrations and kraft recovery furnace combustion gas
characteristics is compiled in Table 3-3. The emission test data
are taken from tests conducted in 1992-93 by the National Council
of the Paper Industry for Air and Stream Improvement, Inc.
(NCASl).1^ A complete compilation of available HCl emission test
data for kraft recovery furnaces is presented in a separate
memorandum.33 The data in Table 3-3 reflect two factors that
complicate the design, operation, and characterization of
performance of packed-bed scrubbing systems--the relatively large
flue gas flow rates generated by these systems and the relatively
low and variable concentrations of both HCl and SC^ generated by
these systems. Because of the large flow rates, either large-
3-27
-------
TABLE 3-3. SUMMARY OF DATA ON KRAFT RECOVERY FURNACE COMBUSTION GAS
CHARACTERISTICS19
Flue gas concentrations8 ||
ppm CO® &% 02
o^1
«
CO
@
cT
CO
Q,
CU
c-»
O
«
CO
©
O
cu
cu
ft
V3
VO
4,352(153,705)
VO
M
O
3
G.
CO
""
W
H
o
<
•<
2
°S
vi
(S
0
CO
«s
VI
2,575 (90,970)
VI
«n
R)
a
0
u
o
a
en
a
a
1
vi
0
-------
diameter scrubbers or parallel systems are needed to maintain
proper flow conditions in the scrubber. The low HC1
concentrations such as those in Table 3-3 are reported to affect
the ability of the scrubber to achieve high removal efficiencies
because of limitations on the.outlet concentrations that can be
achieved with the systems."'
Only limited test data on acid gas control with packed-bed
scrubbers are available for kraft recovery furnaces. However,
packed-bed scrubbers are used widely on waste combustion systems,
and a substantial body of information is available on their
performance for these systems. These.data show that these
scrubbers can achieve removal efficiencies of 99 percent or
greater with relatively high concentrations of HC1. Efficiencies
of this order are confirmed'by scrubber manufacturer representa-'
tives, who indicate that properly designed and operated
packed-bed scrubbers can achieve HC1 emission reductions of at
least 99 percent or an HC1 outlet concentration not exceeding
5 ppmv for recovery furnace exhaust streams.23'34
Available data on scrubber performance are presented in
Table 3-4 for seven kraft DCE and NDCE recovery furnaces and five
medical waste incinerators (MWI's).15/21'22'35f36 With one
exception, the data in Table 3-4 indicate that packed-bed
scrubbers on waste combustion systems can achieve HC1 removal
efficiencies of 98.8 percent or greater. The one exception is a
cross-flow scrubber, which manufacturers indicate are less
efficient than counterflow systems. The high HC1 removal
efficiencies with packed-bed scrubbers are supported by a summary
of trial bum data for hazardous waste incinerators (HWI's).37
' This study contains limited data on HC1 control efficiencies for
10 HWI's based on feed rates of organic chlorine to the system
and outlet HC1 measurements.37 Because the reported HWI data are
insufficient to characterize inlet concentrations or scrubber
characteristics, they were excluded from Table 3-4. However,
efficiencies for all 10 systems are greater than 98 percent, and
6 of the 10 systems showed efficiencies of substantially greater
than 99 percent.37 More detailed data on two of these systems
3-29
-------
?
0
ri"
0
•^
>i
-T
>g
n
_i
g
g
Q
|yj
rJ
o
fe,
ry*
w
Oi
erf
w
CQ
Di
S
CD
H
O
CO
<*
i
n
1
w
a
•§
0
g
^
^
o*
CO
s
•o
s
|||
0 £»
o «
o "?
u g
" &
jg
E
U)
a
lit
w »&
lit
*"* ° A
i
« <=•
a s
•S -2
S "
0 -S
ilet gas flow,
cmm (acfm)
*"
crubber description
w
1
:=
*5
<2
o
£"
30
c-i
•S.
«n
2
2
2
Z
oo
p
O
Z
2
2
2
^^
fO
S
eo
C4
—
d
O
en"
s
s
cked-bed scrubber with 2 ci
iparated by spray chamber
§. ">
_g>-|
Ii
•§
3 "
11
fi
ii.
J,
41
O
2
w 8
Hi
z s
z
o
1
2
2
2
2
_
iQ
a
o
c4
»n
~™
360 (260,000)
"-'
S
41
cked-bed scrubber with pro
SO
0,
^
js
cu £
CO w
W 5
b
u
o
u) 1
u c
Q S
2
V
*
2
2
2
„.
2
,800 (522,049)
—
"" 1
ecovery furnaces followed 1
-bed scrubber with causticA
*" "S
bO O
C B
S |
to o
tu 2
C4 °
b
4J
>
0
8 1
en <2
2
n-
in
2
2
2
2
1
o
2 8
§1
s
C)
^1
O\
oo
8
^
CO
oo
e-
P
•^-
^
en
270 (9,600)
llowed by 2 packed-bed
<2
j>
1 .
"§ £
1 s
> S
S
o
oo
cS
s
5
?
oo
§
C-J
tn
§
tn
d
S
by cross-flow packed-bed
•S
1
S
S u
41
s i
2
2
2
2
en
2
8
^^
M
U
03
0.
£•
1
S
J
u
U3
•n
3
>
>
S
recovery furnace data corrected to 8 percent 02.
1
2
0)
d<
r-
S
•s
correcte
3
cd
•O
als 0.08 ppm.
i to plugged nozzles dunng the test.
tual outlet SC>2 equ
y be biased low du<
u
a
•|
•c
c
S
01
1
U
*O
£
a)
3
bic meters per minu
ctual cul
es
II
E
tual cubic feet.
ic feet per minute.
• actual cubic meter
cms per thousand ac
•o a =
5 » »
1 1 «
o -^ o
II p*^
E o ^
million volume.
o.
Q3
0.
II
S.
3-30
-------
that used a caustic scrubbing solution showed outlet
concentrations of 1.3 and 13 ppmv, with removal efficiencies of
>99.9 percent and 98.2 percent, respectively.38 Taken together,
the data in Table 3-4 and the HWI data show that a properly
designed and operated packed-bed scrubber can achieve HC1 removal
efficiencies in the range of 99 percent.
For two MWI systems in Table 3-4, outlet HC1 concentrations
of less than 1.5 ppmv were obtained with HC1 inlet concentrations
of greater than 1,000 ppmv. This same type of reduction appears
to have been achieved.by the one HWI system with an outlet
concentration of 1.3 ppmv. These MWI and HWI data support the
results from the theoretical analyses on the ability of these
systems to achieve low outlet concentrations. Also, systems with
caustic scrubbing solution are able to achieve outlet
concentrations not exceeding 5 ppmv.23'34
3.1.2.1.4 Performance data--gaseous organic HAP's. Data
were presented in Chapter 2 on levels of gaseous organic HAP
emissions expected from recovery furnaces. While there are no
available gaseous organic HAP emission test data that would allow
the determination of the performance of packed-bed scrubbers in
controlling HAP emissions from recovery furnaces, packed-bed
scrubbers designed primarily for HC1 control would be expected to
control emissions of the water-soluble gaseous organic HAP's
(such as methanol and formaldehyde) to a limited degree.
However, because of the low concentrations of gaseous organic
HAP's expected in the furnace exhaust streams, absorption of
these pollutants may be limited in packed-bed scrubbers designed
for HC1 control due to the small concentration gradient (driving
force} between the gas and scrubbing liquid.
In providing design and cost information for packed-bed
scrubbers for HC1 control, vendors were asked to provide
estimates of the degree to which the HAP's methanol,
formaldehyde, and acetaldehyde would be controlled with the same
equipment. One vendor estimated removal efficiencies of
50 percent for methanol and formaldehyde and 20 percent for
acetaldehyde but would not offer a performance guarantee for the
3-31
-------
three pollutants.34 While the packed-bed scrubber may remove
some gaseous organic HAP's from the gas stream, a portion of the
•pollutants absorbed into the scrubbing fluid may volatilize back
into the atmosphere from the wastewater treatment process, which
would decrease the overall system removal efficiencies.
Currently, data are insufficient to determine whether the
blowdown of the scrubber liquor is sufficient to maintain low
levels of HAP's in the scrubber water to maintain the
concentration gradient and to prevent stripping.
3.1.2.1.5 Existing applications of packed-bed scrubbers.
Currently,.two NDCE recovery furnaces, both at the same kraft
pulp mill, are each equipped with a cross-flow packed-bed
scrubber that operates downstream from an ESP. For other
recovery furnace applications, the packed-bed scrubber also would
be installed downstream from the PM control device, which would
remove over 99 percent of the PM from the emission stream before
scrubbing.
One of the packed-bed scrubbers currently in operation on an
NDCE recovery furnace was designed for S02 control and hot-water
generation (i.e., to recover heat from the flue gas).21'22 Fresh
water is used as the scrubbing medium, with some added caustic.
Although no HC1 emission data are available for the recovery
furnaces at this mill, S02 scrubber design criteria are available
for this recovery furnace. As shown in Table 3-4, the design
inlet S02 concentration was 150 ppmv, while the design outlet S02
concentration was 2.5 ppmv, for a design S02 emission reduction
of 98 percent.35 Because HC1 is more soluble in water than S02
and based on the theoretical discussion presented earlier, HC1
emission reductions greater than 98 percent would be expected for
this packed-bed scrubber. s
While only one facility has installed packed-bed scrubbers
on NDCE recovery furnaces, three other facilities have reported
that packed-tower systems are .being used to control emissions
from five DCE recovery furnaces.15 Although no information is
available on the control of HC1 achieved by these systems, the
available data indicate that they achieve S02 outlet
3-32
-------
concentrations -ranging from <1 to about 60 ppmv.1^ Furthermore,
the data indicate that the three packed-tower systems handle flow
rates in the range of 7,360 to 14,800 actual cubic meters per
minute (acmm) (260,000 to 522,000 acfm).15 These systems show
that counterflow packed-tower systems are compatible with both
DCE and NDCE recovery furnace exhaust streams and provide a
viable technology for control of HC1 emissions from these
furnaces,
3.1.2.2 Venturi Scrubbers. Venturi scrubbers are a
demonstrated control.technology for reducing PM emissions and,
therefore, PM HAP emissions from lime kilns and SDT's.
Particulate matter emissions from approximately 86 percent of
lime kilns and 35 percent of SDT's are controlled using venturi
scrubbers.^ However, • venturi scrubbers are used on only one
existing recovery furnace (an NDCE recovery furnace) as the sole
PM control device.12 Venturi scrubbers may also provide, to a
more limited extent, control of gaseous HAP emissions. This
section briefly describes venturi scrubbers and presents PM HAP
performance data.
3.1.2.2.1 Description. Venturi scrubbers are designed to
remove PM primarily by impaction through high-energy contact
between the scrubbing liquid and suspended PM in 'the gas
stream.26 A schematic of a venturi scrubbing system is shown in
Figure 3-10. A venturi scrubbing system typically consists of a
venturi scrubbing vessel and cyclonic separator. The venturi
scrubbing vessel has a converging section, a throat section, and
diverging section.
The exhaust gas enters the converging section and, as the
cross-sectional area of the scrubber vessel decreases, gas
velocity increases. Scrubbing liquid is introduced either at the
entrance to the converging section or at the throat.39 The
exhaust gas, which is pulled through the venturi vessel by the
system's induced draft fan and forced to move at extremely high
velocities in the throat, shears the scrubbing liquid from the
walls, atomizing the liquid.26,39 Particulate matter and gaseous
pollutants are transferred from the gas stream to the liquid
3-33
-------
Cyclonic
separator
Flooded
elbow
Figure 3-10. Schematic of a venturi scrubber.
3-34
-------
droplets via impaction and diffusive mass transfer in the throat
section as the exhaust stream turbulently mixes with the atomized
scrubbing liquid. The throat .'section may be constructed so that
the size of the throat opening is adjustable. With an
adjustable-throat (or variable-throat) venturi, the gas velocity
across the throat can be maintained at a constant speed as the
gas flow rate changes, thereby maintaining the desired PM
collection efficiency. From the throat section,, the exhaust
stream passes through the diverging section of the venturi
scrubbing vessel, where the velocity decreases. Diffusion, which
is an effective collection mechanism for fine particles and also
the primary mechanism of gaseous pollutants transfer, usually
occurs in the diverging section, where the velocities of the gas
stream and liquid droplets are almost equal. Collection of fine
particles by the.liquid droplets is possible because the path of
fine particles is influenced primarily by Brownian motion rather
than by the path of the gas stream.40
An entrainment separator, typically a cyclonic separator, is
needed to collect the PM entrained in the droplets because the
high velocity of the exhaust stream from the venturi scrubbing
vessel would have a tendency to exhaust the droplets.39
3.1.2.2.2 Performance data--PM HAP's. Available
information indicates that control systems that achieve the
greatest total PM collection efficiencies also achieve the best
performance for PM HAP emissions.8 Thus, although the following
information is presented in terms of total PM, the relative
performance of the different systems is applicable to that
portion of the total PM that is comprised of PM HAP's.
The performance of the scrubber in terms of PM collection is
strongly affected by the' pressure drop across the scrubber
throat, the liquid-to-gas (L/G) ratio, and the particle size
distribution. Particulate matter collection efficiency generally
increases as the throat velocity and turbulence of the gas stream
increase, as indicated by an increased pressure drop across the
scrubber.
3-35
-------
Typical venturi scrubber L/G ratios for PM control range
from 0.4 to 1.3 liters per actual cubic meter (L/acm) (3 to
10 gallons per 1,000 actual cubic feet [gal/103 acf]). While L/G
ratios up to 2.7 L/acm (20 gal/103 acf) can be used, increasing
the L/G ratio beyond 1.3 L/acm (10 gal/103 acf) usually does not
significantly improve PM collection efficiency. However, venturi
scrubbers with L/G ratios ranging from 2.7 to 5.3 L/acm (20 to
40 gal/103 acf) are used where absorption of gaseous pollutants
in addition to PM control is desired. Liquid-to-gas ratios below
0.4 L/acm (3 gal/103 acf) are usually not sufficient to cover the
throat section.39
For lime kiln applications, PM collection efficiencies for
venturi scrubbers average 99 percent.39 These results are
supported by data reported for four lime kiln scrubbers.15'41
One scrubber provided a removal efficiency of 98 percent with a
pressure drop of 26 millimeters of mercury (mm Hg) (14 inches of
water [in. H20]).41 The other three scrubbers provided removal
efficiencies of greater than 99.5 percent with scrubber pressure
drops in the range of 37 to 45 mm Hg (20 to 24 in. H20).15
Variable-throat venturi scrubbers with pressure drops that
range from 19 to 56 mm Hg (10 to 30 in. H20) and average 37 mm Hg
(20 in. H20) are typically used for controlling PM emissions from
lime kilns.1'40 Typical L/G ratios are between 1.3 and 2.7 L/acm
(10 and 20 gal/103 acf).40 Particle size distribution data,
based on limited lime kiln test data, show that approximately
17 percent of the particles entering the scrubber are less than
10 micrometers (/zm) in diameter, while 98 percent of particles
exiting the scrubber are less than 10 /zm in diameter.42 Lime mud
with a high soda content, which results from inefficient washing
of the lime mud, can produce a large quantity of PM in the 0.1 to
1.0 /zm size range.40 Particles in this size range are the most
difficult to remove and may require increased scrubber pressure
drops or process modifications to improve lime mud washing.
The NSPS for kraft lime kilns requires that PM emissions
from gas-fired and oil-fired lime kilns constructed,
reconstructed, or modified after September 24, 1976 be less than
3-36
-------
or equal to 0.15 and 0.30 g/dscm (0.067 and 0.13 gr/dscf ) ,
respectively.10'11 Based on installation dates, approximately
28 percent of existing lime kilns are subject to the NSPS.1
Long-term PM data are available for two gas -fired lime kilns
with venturi scrubbers that are subject to the NSPS PM limit of
0.15 g/dscm (0.067 gr/dscf). Figures 3-11 and 3-12 present
monthly PM emission data from the State of Washington for lime
kilns at Mills A and B over a 6-year period from 1988 to 1994. 13
The lime kilns at Mills A and B were installed in 1980 and 1979,
respectively.1^ The data demonstrate that existing venturi
scrubbers on lime kilns can readily meet the NSPS PM level of
0.15 g/dscm (0.067 gr/dscf) on a long-term, continuous basis.
For SDT applications, reported PM removal efficiencies for
venturi scrubbers range from 97 to greater than 99
percent. 40,42,43 The average pressure drop for SDT venturi
scrubbers is 15 mm Hg (8 in. I^O).1 Liquid-to-gas ratios range
from 1.0 to 1.3 L/acm (8 to 10 gal/103 acf).40 Weak wash (from
the lime mud washer) is the scrubbing fluid for the majority of
SDT venturi scrubbers.
The NSPS for kraft pulp mills require that PM emissions from
SDT's that are constructed, modified, or reconstructed after
September 24, 1976 be less than 0.10 kg/Mg (0.20 Ib/ton) of BLS
fired in the recovery furnace.10'11 Based on installation dates,
approximately 29 percent of existing SDT's are subject to the
NSPS.12
No long-term PM data are available for SDT's with venturi
scrubbers that are subject to the NSPS limit of 0.10 kg/Mg
(0.20 Ib/ton) BLS. However, long-term data are available for two
SDT's equipped with packed-bed scrubbers designed to meet a more
stringent PM permit limit of 0.06 kg/Mg (0.12 Ib/ton) BLS; high-
efficiency venturi scrubbers are expected to achieve equivalent
control. Figures 3-13 and 3-14 present monthly PM emission data
from the State of Washington for two SDT's with packed-bed
scrubber systems at Mill B over a 6 -year period from 1988 to
1994.13 Figure 3-15 presents monthly PM emission data from the
State of Washington for a new SDT with a combination venturi and
3-37
-------
o
oo
O)
Q
LLJ
i<£
HCD
111
LU
UL
cc
LJLJ
CD
Q
_
0 r^
co <
S co
> en
. . CQ
Q =3
O DC
0_ O
< CO
CC
LL!
OJ
DL
O
cn
Q
UJ
cc
ID
CO
lO
U
r
n
D
r
C\J
o"
CO
o
co
q
d
-05
-CCS
"CO
•CD
•c
.CO
•H
S
4J
(ti
' LLI 0)
i *
| ^
-OP
-CD UJ
: CD
- Q.
:0 <
•op CO
•c
-co
-co
• c
.CO'
O
•H
m
m
-H
CU
>
cn
(U
tn
-H
-co
ra
(dosa/yo)
3ivnnonyvd
3-38
-------
05
O
Q
LU
CO
CD
OJ
LU
^ zs
S
CC
X
UJ
CO
CC
LU
CO
m
13
CC
o
LU
=> 2:
£*
.. m
Q Z>
O CC
DL O
CO
DC
LU
co
CVJ
i!
D
U
r
D
X!
r
a.
r
JU
D
r
C\l
o
c\j
d
co
OJ
CO
o
•co
•OJ
•c
-CO
m
rH
rH
-H
a
^
. A CC rH
.i < -H
.-J LU -M
: ± §
• fe 3
- o ,.
•9 LU
:is *
: CD
- Q_
•c
-co
•c
• CO
'CO
•CO
(dosa/uo)
d
o
-H
m
CO
-H
£
0)
CN
H
i
m
0)
^
Cn
-H
3i\nnoiiuvd
3-39
-------
DC
HI
O
I
in
o
CO
o"
in
CM
CM
o
J
CD
~3T
o
CD
_-> LU
: o
•c -5
• CO ;£
•-> h-
: o
- Q.
m
r-i
rH
-H
m
E-i
Q
CO
4.)
tti
•d
o
•H
m
CD
-H
cn
-03
-CO
~
OJ
>-)
3
Cn
CO
-CO
-c
. CO
C) ~3
(sia NOi/ai)
aivnnonavd
3-40
-------
CC
UJ
II
CC
UJ
05
TT. Q:
Q O
Uil CC
LO
h-
CD
Q
LU
CC <;
CM > c/i>
CO -> -ry
05 O ^
Q Q S
UJ U-l
CC
UJ
CC
ID
CO
UJ
D_
D_
UJ
CC Q
a
cc
UJ
CD
m
o O cc
Q d. <: o
< ni co
so
CD
ID
•<*•
CO
O
in
CM
o
-c
-co
-J
to
o
CM
.ra <
.-3 LJJ
' >
; 1
: o
^
•55 yj
•c "^
.« ^c
•"> P
: o
z
_J
D_
, <
en
-co
-co
CO
-co
-c
. CO
O ~3
o
(sia Noi/ai) yanviAi
m
r-t
r-t
-H
S
O
a
Q
CO
O
M-t
ttJ
4J
rt!
T5
C
O
-i-t
ra
m
-H
S
I-t
I
0)
H
d
Cn
-H
fe.
3-41
-------
in
o
oj
0
ID
S ||
§ §i
UJ CM ill r?
ZJ != < i CD
hi^S^i
c
V.
o o a: cc
o. a. O o
< < co co
in
o
in
".
o
co
o
m
5
(sia Noi/ai) H3HVIAI 3i\nnonyvd
3-42
-------
packed-bed scrubber system at Mill C over a 2-year period from
1992 to 1994.13 These data demonstrate that high-efficiency wet
scrubbers on SDT's can achieve a maximum outlet PM level of
0.06 kg/Mg (0.12 Ib/ton) BLS on a long-term, continuous basis.
3.1.3 BLO Control
As explained in Chapter 2, BLO systems that use pure oxygen
instead of air to oxidize black liquor do not need a system vent
because all of the oxygen is consumed in the oxidation reaction.
As a result, there are no emission points for the molecular 02
BLO unit. Black liquor oxidation systems that use air to oxidize
black liquor require vents for each oxidation tank. Add-on
controls currently used to reduce gaseous organic HAP emissions
from air-sparging BLO units collect and incinerate exhaust gases
from the BLO vents. This section describes the add-on controls
and presents the reductions in gaseous organic HAP emissions that
can be achieved with the add-on controls.
'3.1.3.1 Description. Of the estimated 42 kraft mills with
air-based BLO systems, two mills reduce gaseous organic HAP
emissions from air-sparging BLO vents with vent gas collection/
incineration systems. The BLO vent gas collection/incineration
method used by one mill involves ducting the vent gases to two
condensers in series (to remove water from the gases), reheating
the gases, and then ducting the dry gases to a power boiler for
incineration.44'45 The control method used by the second mill is
identical except that a mesh-pad mist eliminator is used in place
of the condensers to remove water from the vent gas. "' Both
mills incinerate their BLO vent gases in order to reduce TRS
emissions. Because gaseous organic HAP's such as methanol,
acetaldehyde, and methyl ethyl ketone•are also emitted from BLO
vents, emissions of these HAP's would also be reduced by the
incineration of BLO vent gases. Particulate HAP's and acid gases
(e.g., HC1) are not emitted from BLO vents.
Before the BLO vent gases can be incinerated, they must
first be captured and transported to the incineration device.
Vent gases from a BLO unit can be hard-piped to the combustion
control device. The type of duct material that can be used is
3-43
-------
determined by the characteristics of the gas in the vent
stream.48 Two materials commonly used in the construction of
vent ducts in the pulp and paper industry are fiberglass and
stainless steel. Fiberglass ducts have the advantages of
relatively low cost, light weight, and corrosion resistance, but
are unable to be electrically grounded to prevent the buildup of
static charge and the absorption of hydrocarbons in its
fiberglass resin.48 Stainless steel is the preferred material of
construction for NCG cransport systems.49 Stainless steel has
the advantage of resisting corrosion by water and sulfur
compounds but is susceptible to corrosion by chlorides that may
be present in a vent gas control system that includes other vents
besides the BLO vent (e.g., bleach plant vent).50
The BLO vent gas may be conditioned to alter the moisture
content and/or the temperature of the stream before it is" vented
to the combustion control device. Condensers, knockout drums, or
entrainment separators may be used to condition the gas in the
AR
gas transport system.^0
Vent gases are preheated only when emissions are controlled
with a combustion device. Preheating is generally only practiced
on high-volume, low-concentration (HVLC) streams, where the risk
of explosion is sufficiently low.51 The vent gases from BLO
units can be classified as HVLC streams. Power boilers and
stand-alone thermal oxidizers are generally the preferred
combustion devices for incinerating HVLC streams. Power boilers,
which include coal, natural gas, oil, wood waste, or combination
fuel-fired boilers, are designed to produce heat, steam, and
electricity for pulp mill operations.48 A thermal oxidizer is a
refractory-lined chamber containing a burner or burners used to
oxidize vent streams containing volatile organic compounds
(VOC's).48 Both combustion devices have been demonstrated in the
A p
pulp and paper industry as control devices for vent emissions.
Combustion control devices, such as power boilers and
thermal oxidizers, destroy the chemical structure of organic
compounds by oxidation at elevated temperatures. These devices
operate on the principle that any VOC heated to a high enough
3-44
-------
temperature in* the presence of sufficient oxygen will oxidize to
predominantly C02 and H20.52 Combustion devices such as power
boilers and thermal oxidizers-can achieve VOC/HAP destruction
efficiencies of 98 percent or greater.53'54 Therefore, a net
removal efficiency of 98 percent was assumed for all HAP's for
the BLO incineration control system.
3-1-3.2 Performance data--gaseous organic HAP's. Because
neither of the two vent gas collection/incineration systems
currently existing has undergone emissions testing, the gaseous
organic HAP emission reductions associated with a BLO vent gas
control system were estimated based on gaseous organic HAP
emission data available from uncontrolled air-sparging BLO
systems, assuming a 98 percent destruction efficiency.55 The
gaseous organic-HAP emission -reductions resulting from BLO
control are presented in Table 3-5. For a BLO unit with a BLS
firing rate of 700,000 kilograms of BLS per day (kg BLS/d)
(1.5 million pounds of BLS per day [Ib BLS/d]) and 351 operating
days per year (d/yr), controlling emissions from the BLO vent
could reduce average gaseous organic HAP emissions by about
47 Mg/yr (54 ton/yr), of which methanol accounts for about
85 percent.
3.2 EQUIPMENT CHANGES/MODIFICATIONS
This section describes equipment changes/modifications that
could be applied•to recovery furnaces at kraft pulp and paper
mills to reduce gaseous organic HAP emissions. The two equipment
_changes/modifications are elimination of black liquor or HAP-
contaminated process water from ESP control systems and
conversion from a DCE recovery furnace system to an NDCE recovery
furnace, which are discussed in Sections 3.2.1 and 3.2.2,
respectively.
3-2.1 Elimination of Black Liquor used in NDCE Recovery Furnace
ESP Control Systems
"When hot recovery furnace flue gases come in contact with
ESP control systems, gaseous organic HAP's (primarily methanol)
may be stripped from any black liquor or HAP-contaminated process
water present in the system and emitted to the atmosphere.
3-45
-------
TABLE 3-5. HAP EMISSIONS FOR BLO UNITS55
Hazardous air pollutant
Total organic HAP'sb
Acetaldehyde
Benzene
Formaldehyde
Methanol
Methyl ethyl ketone
Methyl isobutyl ketone
Phenol
Toluene
Xylenes
Uncontrolled BLO unit
Emission factor,
mass/mass BLS
2.03 x 10"4
1.48 x 1(T5
1.21 x 1CT6
2.89 x 10"7
1.73 x 1CT4
4.24 x ID"6
1.77 x 1CT6
1.93 x 10-6
1.44 x KT6
4.28 x ID"6
Mass rate, kg/yr
(lb/yr)a
48,500
(107,000)
3,530
(7,790)
289
(637)
69
(152)
41,300
(91,100)
1,010
(2,230)
423
(932)
461
(1,020)
344
(758)
1,020
(2,250)
Controlled BLO unit
Emission factor,
mass/mass BLS
4.06 x 10T6
2.96 x 10'7
2.42 x 10'8
5.78 x 10'9
3.46 x W6
8.48 x 10'8
3.54 x 10'8
3.86 x KT8
2.88 x 10'8
8.56 x 10'8
Mass rate,
kg/yr (lb/yr)a
969
(2,140)
70.7
(156)
5.78
(12.7)
1.38
(3.04)
826
(1,820)
20.3
(44.6)
8.45
(18.6)
9.22
(20.3)
6.88
(15.2)
20.4
(45.1)
aMass rates are based on a BLS firing rate of 700,000 kg BLS/d (1.5 million Ib BLS/d) and 351 operating
d/yr.
kTotal organic HAP's include all gaseous HAP's in this table. Numbers may not add exactly due to
rounding.
3-46
-------
Electrostatic precipitator control systems include the ESP plus
the PM return system associated with the ESP. One method of
controlling gaseous organic HAP emissions from ESP systems is to
prevent their generation by eliminating the black liquor or
contaminated process water from the system. This equipment
modification for ESP's used with NDCE recovery furnaces and the
associated HAP emission reductions are discussed below» (This
equipment modification is not discussed as an option for reducing
gaseous organic HAP's from DCE recovery furnaces be'cause, as
discussed in Chapter 2, no data are available to determine if
eliminating the wet bottom ESP's that are typically used with DCE
recovery furnaces would actually decrease overall gaseous organic
HAP emissions.)
3.2.1.1 Description. Particulate matter emissions from
approximately 23 percent of all NDCE recovery furnaces are
controlled with wet-bottom ESP's that use either unoxidized black
liquor or process water in the ESP bottom. The majority
(77 percent) of NDCE recovery furnaces, however, are controlled
with dry-bottom ESP's.1 Unoxidized black liquor is used in about
87 percent of the wet-bottom ESP's on NDCE recovery furnaces, and
water is used in the remaining 13 percent.
Since the 1980's, the industry trend has been toward the use
of dry-bottom ESP's. Some mills that installed wet-bottom ESP's
on NDCE recovery furnaces in the-late 1970's and early 1980's
experienced difficulties in meeting the NSPS TRS emission limit
of 5 ppm. These mills attempted to comply with the TRS emission
limit through the use of one of the following techniques:
(1) modifying the bottom of the ESP to limit contact between the
flue gases and the, black liquor (e.g., using baffles),
(2) switching from the use of black liquor to using water in the
ESP bottom, or (3) converting from a wet-bottom ESP to a dry-
bottom ESP. In general, technique Nos. 2 and 3 proved to be the
most viable options for meeting the TRS emission limit on a
continuous basis.56'57 Because the mechanism for generating TRS
emissions in the wet-bottom ESP is the same mechanism for
generating HAP emissions, technique Nos. 2 and 3 would also
3-47
-------
prevent methanol and any other gaseous organic HAP's from being
generated in the ESP, provided the water used in technique No. 2
is not contaminated with substantial quantities of organic
compounds.
The major modifications involved in technique No. 3 (i.e.,
converting a wet-bottom ESP to the dry-bottom design) include
(1) removing the existing agitator paddles and liquor piping,
(2) installing a perpendicular drag scraper system, shallow
fallout hoppers, drag chain conveyors, rotary valves, ash mixing
tank, agitator, and associated instrumentation, and (3) making
£- Q CO
piping modifications. °'3y
The older design dry-bottom ESP control systems sometimes
use black liquor to sluice and transport the PM captured by dry-
bottom ESP's to the saltcake mix tank. As a result, gaseous
organic HAP's may be stripped from the black liquor as the hot
recovery furnace flue gases are pulled through the ESP by the
induced draft fan. The gaseous organic HAP emissions could be
controlled by converting to a dry PM return system. Dry PM
return systems eliminate the use of black liquor in the PM return
system, so that the captured PM does not contact any black liquor
until it reaches the mix tank. These hewer PM return systems
also are equipped with rotary valves (through which the dry
captured PM passes) that provide an air lock between the ESP and
the remainder of the PM return system and the mix tank (which
contains black liquor).60 According to one ESP manufacturer,
systems that included dry return of the captured PM were
initially installed around 1985.60
3.2.1.2 HAP Emission Reduction Potential. As shown in
Table 3-6, the elimination of black liquor from ESP control
systems results in substantial reductions for the majority of the
gaseous organic HAP compounds. (Note: Although HC1 emissions are
included in Table 3-6, HC1 emissions are unaffected by the
presence or absence of black liquor in the ESP system.) Although
the available data indicate that acetaldehyde emissions increase
with the elimination of black liquor from ESP control systems,
3-48
-------
TABLE 3-6. HAP EMISSIONS FOR NDCE RECOVERY FURNACES55
Hazardous air
pollutant
Total organic
HAP'se
Acetaldehyde
Benzene
Formaldehyde
Methanol
Methyl ethyl
ketone
Methyl isobutyl
ketone
Phenol
Toluene
Xylenes
HC1
ESP control system with black
liquor
Emission factor,
mass/mass
BLSa'b
1.30 x 1CT4
7.50 x 10"6
9.73 x 10'6
2.98 x 10'6
5.04 x 10'5
7.46 x 10'6
9.23 x 10~6
1.50 x 10'5
8.37 x 10~6
1.92 x 10'5
1.20 x 10"4
Mass rate,
kg/yr
0b/yr)c
31,000
(68,400)
1,790
(3,950)
2,320
(5,120)
712
(1,570)
12,000
(26,500)
1,780
(3,930)
2,200
(4,860)
3,580
(7,900)
2,000
(4,410)
4,590
(10,100)
28,700
(63,200)
ESP control system without
black liquor
Emission factor,
mass/mass
BLSb'd
3.67 x 10'5
1.52 x 10'5
1.73 x 1Q-6
1.93 x IOT6
4.93 x W6
1.80 x KT6
2.27 x 10"6
2.05 x 10-6
2.03 x 10"6
4.76 x 10"6
1.20 x KT4
Mass rate,
kg/yr
(lb/yr)c
8,760
(19,300)
3,630
(8,000)
413
(911)
461
(1,020)
1,180
(2,600)
430
(948)
542
(1,200)
490
(1,080)
485
(1,070)
1,140
(2,510)
28,700
(63,200)
Percent
reduction
72
-103
82
35
90
76
75
86
76
75
—
aEmission factors for gaseous organic HAP's are based on the average emission factors for NDCE
recovery furnaces equipped with wet-bottom ESP's or with dry-bottom ESP's with wet PM return
systems.
Emission factor for HCi is based on the average emission factor for all NDCE recovery furnaces.
GMass rates are based on a furnace size of 700,000 kg BLS/d (1.5 million Ib BLS/d) and 351 operating
d/yr.
Emission factors for gaseous organic HAP's are based on the average emission factors for NDCE
recovery furnaces equipped with dry-bottom ESP's and dry PM return systems.
^otal organic HAP's include all gaseous HAP's in this table except HCI. Numbers may not add exactly
due to rounding.
3-49
-------
the difference in the acetaldehyde levels may be an anomaly of
the limited data on acetaldehyde emissions.55
Emission data are available for 10 NDCE recovery furnaces
with black liquor in the ESP control system (i.e., 6 NDCE
recovery furnaces with wet-bottom ESP's and 4 NDCE recovery
furnaces with dry-bottom ESP's and wet PM return systems).55
Emission data are also available for three NDCE recovery furnaces
with dry-bottom ESP's that have dry return of the captured PM.
Figure 3-16 illustrates the differences in methanol emissions
between the two data sets. The methanol emission data from the
ten NDCE recovery furnaces with black liquor in the ESP control
system are significantly higher than those from the three NDCE
recovery furnaces with dry-bottom ESP's and dry PM return
systems.
Based on the available emission test data, eliminating the
black liquor from the ESP control system reduces total gaseous
organic HAP emissions by approximately 72 percent.55 In addition
to their lower HAP emission potential, NDCE recovery furnaces
with dry-bottom ESP's and dry PM return also have lower TRS
emissions than NDCE recovery furnaces with black liquor in the
ESP control system.15
For an NDCE recovery furnace with a furnace size of
700,000 kg BLS per day (1.5 million Ib BLS per day) and 351
operating d/yr, eliminating black liquor from the ESP control
system has the potential to reduce average gaseous organic HAP
emissions by about 22 Mg/yr (24 ton/yr). This emission reduction
represents the difference in emissions between NDCE recovery
furnaces with black liquor in the ESP control system (i.e., wet-
bottom ESP's or dry-bottom ESP's with wet PM return systems) and
NDCE recovery furnaces with dry-bottom ESP's and dry PM return
systems. Methanol emission reductions account for most of the
estimated HAP emission reduction.
3.2.2 Conversion from a DCS Recovery Furnace System to an NDCE
Recovery Furnace
The conversion from a DCE recovery furnace system to an NDCE
recovery furnace (often referred to as a "low-odor conversion")
3-50
-------
05
in
en
(U
o
nJ
fl
k
3
14-1
>i
k
a)
o
0)
m
d
o
-H
tn
m
-H
o
fl
ttf
U)
m
Q)
5-1
3
tn
(SIS NOi/ai) 1ONVH131AJ
3-51
-------
offers significant emission reduction and operational benefits.
This conversion is an effective measure for reducing HAP
emissions from a DCE recovery furnace system because it
eliminates two sources of HAP emissions, the DCE and BLO unit.
Conversion from a DCE recovery furnace system to an NDCE recovery
furnace is a common modification; an estimated 24 percent of
existing NDCE recovery furnaces were originally installed as DCE
recovery furnaces.1'61'62 These conversions have been performed
for several reasons, including compliance with Federal and State
TRS emission standards and increased energy efficiency. Other
factors influencing the decision to convert a DCE recovery
furnace system to an NDCE design include the age of the existing
DCE recovery furnace, the condition of the system, and whether
additional capacity is needed at the mill. The following
sections describe the major modifications involved in converting
a DCE recovery furnace system to an NDCE recovery furnace,
including operational benefits, and present the potential HAP
emission reductions.
3.2.2.1 Description. The major modifications involved in
converting a DCE recovery furnace system to an NDCE recovery
furnace are (1) replacing the DCE with a concentrator and
associated equipment, (2) extending or replacing the economizer,
(3) rebuilding or replacing the ESP, and (4) removing the BLO
unit. Removal of the DCE is the driving force behind the latter
three major modifications.
The DCE is replaced with a concentrator, which can achieve
the desired black liquor solids content without direct contact
between recovery furnace exhaust gases and black liquor.
Eliminating contact between the hot exhaust gases and black
liquor is desirable because the emissions that result from
stripping are also eliminated. Operational benefits of replacing
the DCE with a low-pressure, steam-driven concentrator include
the higher solids content achievable, improved energy
utilization, and reduced energy costs.63 The concentrator, which
may be one of several types, including falling film and forced
circulation, is installed as part of the evaporator plant. With
3-52
-------
current concentrator technologies, the final solids content of
the black liquor before firing is generally limited to
75 percent; however, concentrator technologies that can achieve
or exceed a solids content of 80 percent are currently in
operation.64 For existing DCE recovery furnace systems, the
average solids content of the fired black liquor is 65 percent;
for existing NDCE recovery furnaces, the average solids content
of the fired black liquor.is 68 percent.1 Increasing the solids .
content of the fired black liquor is advantageous because it
reduces the amount of energy required to evaporate residual
moisture from the black liquor solids.
The economizer is modified to .recover additional heat from
the flue gas that has become available with removal of the DCE.
The additional•heat recovered in the economizer can be used to
produce more low-cost, high-pressure steam. ^ To produce
additional steam, the recovery furnace economizer is either
expanded with the addition of modules or replaced (i.e., with a
long-flow economizer). The economizer is the last section of the
recovery furnace that absorbs heat from the flue gas for steam
production. The economizer optimizes the thermal efficiency of
the recovery furnace by lowering the exit temperature of the
exhaust gas as much as possible while minimizing cold-end
corrosion. Typical design exit temperatures for an NDCE recovery
furnace range from 176° to 190°C (350° to 375°F). For a DCE
recovery furnace system, an exit flue gas temperature of 370°C to
430°C (700°F to 800°F) is required to operate the DCE.S5 The
thermal efficiency of the recovery furnace increases by 1 percent
for every 22°C (40°F) drop in exit gas temperature. 4 "Direct
contact evaporator recovery furnaces typically operate at thermal
efficiencies of 53 to 58 percent, whereas NDCE recovery furnaces
typically operate at thermal efficiencies of 63 to 68 percent."
Auxiliary equipment that is required when expanding an economizer
includes additional soot blowers, hoppers, and conveyors, which
collect and transport the additional ash from the economizer.
The ESP is modified to handle the greater PM loading that
results from the removal of the DCE. The DCE acts as a PM
3-53
-------
control device, collecting 20 to 40 percent of PM emissions from
the recovery furnace.7 Thus, removal of the DCE, without
upgrading the ESP, would likely result in increased PM emissions
from the recovery furnace stack. Because of the increased PM
emissions and a possible change in gas flow rate, the ESP may
need to be either upgraded or replaced. As discussed in
Section 3.1.1.1, the characteristics of the gas stream for NDCE
recovery furnaces contribute to reduced ESP corrosion rates in
comparison to DCE recovery furnaces.
With the removal of the DCE, the BLO unit is no longer
needed; therefore, the emissions associated with the air-sparging
BLO units are eliminated. The operational benefits associated
with removing an air-sparging BLO unit are the elimination of the
high costs associated with operating this unit and the increased
heating value of the black liquor.
3.2.2.2 HAP Emission Reduction Potential. As shown in
Table 3-7, the conversion from a DCE recovery furnace system to
an NDCE recovery furnace results in substantial reductions for
all of the gaseous organic HAP compounds. Based on the available
emission test data, the conversion from a DCE recovery furnace
system to an NDCE recovery furnace reduces total gaseous organic
HAP emissions by approximately 93 percent.55 In addition to a
lower HAP emission potential, NDCE recovery furnaces also emit
lower TRS emissions.
Figure 3-17 presents methanol emission data for DCE recovery
furnace systems and for NDCE recovery furnaces equipped with dry-
bottom ESP's and dry PM return systems.54 For the DCE recovery
furnace systems presented in Figure 3-17, only the methanol
emission data from mills that measured emissions from both the
DCE recovery furnace and the BLO unit were included. There are
three cases in Figure 3-17 where more than one DCE recovery
furnace is present at a mill (M01 and M04, MP1/2 and MP3, and PBS
and PB4). In those cases, a portion of the total BLO methanol
emissions was attributed to each recovery furnace system based on
the BLS firing rates measured for the DCE recovery furnaces
during the emission tests. (Note: Emissions for MP1/2 are
3-54
-------
TABLE 3-7. HAP EMISSIONS FOR DCE RECOVERY FURNACE SYSTEMS
AND NDCE RECOVERY FURNACES55
Hazardous air
pollutant
Total organic
HAP's6
Acetaldehyde
Benzene
Formaldehyde
Methanol
Methyl ethyl
ketone
Methyl isobutyl
ketone
Phenol
Toluene
Xylenes
HC1
DCE recovery furnace system
Emission
factor, mass/mass
BLSa'b
s.isx ur4
4.47 x 1CT5
1.57 x lO"5
3.11 x icr6
3.43 x 1CT4
1.28 x 1CT5
1.39x 10"5
4. 14 x 1CT5
1.12x 10'5
2.90 x 1(T5
1.20 x 10-4
Mass rate,
kg/yr (lb/yr)c
123,000
(271,000)
10,700
(23,500)
3,750
(8,270)
742
(1,640)
81,900
(181,000)
3,050
(6,720)
3,310
(7,300)
9,890
(21,800)
2,670
(5,880)
6,920
(15,300)
28,700
(63,200)
NDCE recovery furnace
Emission factor,
mass/mass
BLSb,d
3.67 x 10'5
1.52 x 10'5
1.73 x 10"6
1.93 x ID"6
4.93 x 10'6
1.80 x W6
2.27 x lO"6
2.05 x 10"6
2.03 x ID"6
4.76 x 10"6
1.20 x HT6
Mass rate,
kg/yr
(lb/yr)c
8,760
(19,300)
3,630
(8,000)
413
(911)
461
(1,020)
1,180
(2,600)
430
(948)
542
(1,200)
490
(1,080)
485
(1,070)
1,140
(2,510)
28,700
(63,200)
Percent
reduction
93
66
89
38
99
86
84
95
82
84
—
aEmission factors for gaseous organic HAP's are based on the sum of the average emission factors for
DCE recovery furnaces and BLO units.
Emission, factor for HC1 is based on the average emission factor for all recovery furnaces since no
overall change in HC5 emissions is expected from converting a DCE recovery furnace system to an
NDCE recovery furnace.
GMass rates are based on a furnace size of 700,000 kg BLS/d (1.5 million Ib BLS/d) and 351 operating
d/yr.
Emission factors for gaseous organic HAP's are based on the average emission factors for NDCE
recovery furnaces equipped with dry-bottom ESP's and dry PM return systems.
^Total organic HAP's include all gaseous HAP's in this table except HC1. Numbers may not add exactly
due to rounding.
3-55
-------
CO
CO <§
0- g
CO 4J
UJ CQ
o
(sna Noi/ai)
3-56
-------
actually the combined emissions for DCE recovery furnaces MP1 and
MP2, which have a combined stack.)
: As shown in Figures 3-18 and 3-19, DCE recovery furnace
systems have more gaseous organic HAP emission points than NDCE
recovery furnaces (HAP emission points are shaded in the
figures). In the DCE system, gaseous organic HAP's such as
methanol can be stripped off in the DCE and wet-bottom ESP by
contact between the hot flue gases and the black liquor. The BLO
vents are also an emission source for methanol and other gaseous
organic HAP's because gaseous organic HAP's can be stripped from
the black liquor and vented as the oxidizing air is forced up
through the black liquor. As described in Section 3.3.1., the
wet-bottom ESP (commonly associated with DCE recovery furnaces)
can be converted to a dry-bottom ESP with a dry PM return system,
thereby eliminating the ESP system as a source of methanol
emissions.
For a DCE recovery furnace system with a furnace size of
700,000 kg BLS per day (1.5 million Ib BLS per day) and 351
operating d/yr, converting the furnace system to an NDCE design
has the potential to reduce average gaseous organic HAP emissions
by about 114 Mg/yr (126 ton/yr). This emission reduction
represents the difference in emissions between DCE recovery
furnace systems (including the BLO unit) and NDCE recovery
furnaces (with dry-bottom ESP's and dry PM return systems).
Methanol emissions reductions account for most of the estimated
gaseous organic HAP emission reduction.
3-57
-------
3-58
-------
Q)
O
rt
t-i
I
o
0)
o
M-4
0)
o
o
m
O
•H
m
ro
-H
I
CTi
H
OJ
-H
fa
3-59
-------
3.3 REFERENCES FOR CHAPTER 3
1. Memorandum from Soltis, V., Nicholson, R., and Holloway, T.,
MRI, to Telander, J., EPA/MICG. July 29, 1994. Summary of
Responses to the NCASI "MACT" Survey--Kraft and Soda Pulp
Mills.
2. APTI Course SI:412B: Electrostatic Precipitator Plan
Review, Self Instructional Guidebook. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. EPA-450/2-82-019. July 1983.
3. Environmental Pollution Control: Pulp and Paper Industry,
Part I, Air. U. S. Environmental Protection Agency.
Cincinnati, OH. Publication No. EPA-625/7-76-001. October
1976.
4. Memorandum from Nicholson, R.,,MRI, to Telander, J.,
EPA/ISB. May 28, 1993. Meeting minutes from the March 30,
1993 meeting between Environmental Elements Corp. and
Midwest Research Institute.
5. Telecon. Bullock, K., MRI, with Bringman, L., Environmental
Elements Corp. December 16, 1993. Rigid-electrode ESP's.
6. Memorandum from Nicholson, R., MRI, to Telander, J.,
EPA/MICG. June 15, 1995. Meeting minutes from the April
12, 1993 meeting between Research-Cottrell and Midwest
Research Institute.
7. Review of New Source Performance Standards for Kraft Pulp
Mills. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/3-83-017.
September 1983.
8. Mcllvane, R. Removal of Heavy Metals and Other Utility Air
Toxics. (Presented at the EPRI Hazardous Air Pollutant
Conference. 1993.)
9. Someshwar, A. Compilation of "Air Toxic" Emission Data for
Boilers, Pulp Mills, and Bleach Plants. National Council of
the Paper Industry for Air and Stream Improvement, Inc. New
York. Technical Bulletin No. 650. June 1993. 128 p.
10. U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 43 FR 7568.
Washington, DC. U.S. Government Printing Office.
February 23, 1978.
3-60
-------
11. U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 51 FR 18538.
Washington, DC. U.S. Government Printing Office. May 20,
1986.
12. Memorandum from Nicholson, R., MRI, to Telander, J.,
EPA/MICG. June 13, 1996. Addendum to Summary of Responses
to the 1992 NCASI MACT Survey.
13. Memorandum from Holloway, T., MRI, to the project file.
July 16, 1996.. State of Washington PM Data for Kraft
Recovery Furnaces, Smelt Dissolving Tanks, and Lime Kilns.
14. Proposed Standards of Performance for Kraft Pulp Mills. In:
Standards Support and Environmental Impact Statement.
Volume 1. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/2-76-014a.
September 1976. pp. 4-2 through 4-10.
15. Memorandum from Soltis, V., MRI, to the project file.
April 3, 1995. Kraft and Soda.Pulp Mill Combustion Sources
Data Base.
16. Telecon. Wallace, D., MRI, with Prouty, A., James River
Corp., Camas, WA. March 24, 1994. Performance of the James
River recovery furnace ESP's.
17. Operation and Maintenance Manual for Electrostatic
Precipitators. U. S. Environmental Protection Agency.
• Research Triangle Park, NC. Publication No.
EPA-625/1-85-017. September 1985.
18. Reference 4, Attachment 9: Sludge lime kiln air pollution
control equipment.
19. Someshwar, A. A Study of Kraft Recovery Furnace
Hydrochloric Acid Emissions. National Council of the Paper
Industry for Air and Stream Improvement, Inc. New York.
Technical Bulletin No. 674. August 1994. 100 p.
20. Somesiiwar, A. and A. Jain. Emissions of Hydrochloric Acid
from Kraft Recovery Furnaces. National Council of the Paper
Industry for Air and Stream Improvement, Inc. (Presented at
the 1992 TAPPI International Chemical Recovery Conference.
Seattle, WA. June 18, 1992.) 14 p.
21. Fact Sheet for Prevention of Significant Deterioration and
Notice of Construction: James River Camas Mill Energy and
Recovery Modernization. Washington Department of Ecology.
February 9, 1989.
3-61
-------
22.
23.
24.
25.
26.
27.
• 28.
29.
30.
31.
32.
33.
34.
Telecon. Soltis, V., MRI, with Bruno, J., Airpol Inc.
October 29, 1993. Information of scrubbers used to control
HC1 emissions from recovery furnaces at pulp and paper
mills.
Telecon. Soltis, V., MRI, with Bruno, J., AirPol, Inc.
July 2, 1993. Information about scrubbers used to control
HC1 emissions.
Calvert, S. and'H. Englund (eds.). Handbook of Air
Pollution Technology. New York, John Wiley & Sons, Inc.
1984. p. 137.
OAQPS Control Cost Manual. 4th Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-90-006. January 1990.
p. 9-9.
Anderson 2000
Incinerators.
February 1989
Inc. Emission Control Systems for
Bulletin TR 89-900239. Peachtree City, GA.
Reference 24, p. 136.
Reference 24, pp. 147-149.
Buonicore, A. Absorption. In: Air Pollution Engineering
Manual, Air and Waste Management Association. Buonicore, A.
and W. Davis (eds.). New York, Van Norstrand Reinhold.
1992. pp. 23-24.
Reference 25, pp. 9-28, 9-60, and 9-61.
Memorandum from Randall, D. and Holloway, T., MRI, to the
project file. October 4, 1994. Absorber Design and Cost
Algorithm.
Reference 25, p. 9-45.
Memorandum from Randall, D., MRI, to the project file.
November 8, 1994. Comparison of EPA and NCASI HC1 Emission
Factors for Recovery Furnaces.
Letter and attachments from Sanders, D., Andersen 2000,
Inc., to Soltis, V., MRI. November 9, 1993. Preliminary
proposal for three separate quench and packed-tower
absorption systems to control hydrochloric acid gas
emissions from recovery furnaces at kraft pulp and paper
mills.
3-62
-------
35.' Telecon. *Soltis, V., MRI, with Bruno, J., AirPol, Inc.
October 29, 1993. Information about scrubbers used to
control HC1 emissions from recovery furnaces at pulp and
paper mills.
36. Memorandum from Holloway, T., MRI, to the project file.
July 2, 1993. Data to Support HC1 Control Technology
Transfer from MWI's to Kraft Recovery Furnaces.
37. Handbook: Permit Writer's Guide to Test Burn Data:
Hazardous Waste Incineration. U. S. Environmental
Protection Agency. Cincinnati, OH. Publication No.
EPA-625/6-86-012. September 1986.
38. Trenholm, A., P. Gorman, and G. Junglcaus. Incinerator
Performance Evaluation Results. In: Performance Evaluation
of Full-Scale Hazardous Waste Incinerators, Volume II. U. S.
Environmental Protection Agency. Cincinnati, OH.
Publication No. EPA-600/2-84-181b. November 1984.
39. APTI Course ST:412C: Wet Scrubber Plan Review, Self
Instructional Guidebook. U. S. Environmental Protection
Agency. Research Triangle Park, NC. Publication No.
EPA-450/2-82-020. March 1984.
40. Kraft Pulp Mill Inspection Guide. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. EPA-340/1-83-017. January 1983.
41. Roy F. Weston, Inc. Emissions Testing of Combustion
Processes in a Pulp and Paper Facility: Champion
International.Corp., Roanoke Rapids, North Carolina.
Prepared for U. S. Environmental Protection Agency--Emission
Measurement Branch. Research Triangle Park, NC. EMB Report
No. 92-KPM-27. October 1992. Volume I.
42. Someshwar, A. and J. Pinkerton. Wood Processing Industry,
In: Air Pollution Engineering Manual, Air and Waste-
Management Association. Buonicore, A. and W. Davis (eds.).
New York, Van Norstrand Reinhold. 1992. p. 840.
43. Modetz, H. and M. Murtiff. Kraft Pulp Mill Industry
Particulate Emissions: Source Category Report. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-600/7-87-006. February 1987.
44. Letter and attachments from Black, C., Champion
International Corp., Courtland, AL, to Crowder, J., EPA/ISB.
June 10, 1993. Response to request for information on costs
for the BLO vent gas control system at the Champion facility
'in Courtland, AL.
3-63
-------
45. Memorandum from Soltis, V. and March, D., MRI, to Telander,
J. , EPA/ISB. May 17, 1993. Trip Report for Champion
International Corp., Courtland, AL.
46. Telecon. Nicholson, R., MRI, with Porritt, T., S.D. Warren
Co. May 27, 1993. The black liquor oxidation air pollution
control system in use at S.D. Warren.
47. Memorandum from Nicholson, R. and Holloway, T., MRI, to
Telander, J., EPA/MICG. September 6, 1995. Trip Report for
S.D. Warren Co., Muskegon, MI.
48. Pulp, Paper, and Paperboard Industry--Background Information
for Proposed Air Emission Standards, Manufacturing Processes
at Kraft, Sulfite, Soda, and Semi-Chemical Mills. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-453/R-93-050a. October 1993.
49. Collection and Burning of Kraft Non-Condensible Gases--
Current Practices, Operating Experience, and Important
Aspects of Design and Operation. National Council of the
Paper Industry for Air and Stream Improvement. New York.
Technical Bulletin No. 469. August 29, 1985. p. 40.
50. An Investigation of Corrosion in Particulate Control
Equipment. U. S. Environmental Protection Agency.
Washington, DC. Publication No. EPA-340/1-81-002.
February 1981. p. 38.
51. Reference 50, p. 51.
52. Hazardous Air Pollutant Emissions from Process Units in the
Synthetic Organic Chemical Manufacturing Industry--
Background Information for Proposed Standard. Volume IB,
Control Technologies. U. S. Environmental Protection
Agency. Research Triangle Park, NC. Publication
No. EPA-453/D-92-0166. November 1992. pp. 2-8 and 2-9.
53. Memorandum from Farmer, Jack R., EPA/CPB, to Ajax, B. ,
et al., EPA/CPB. August 22, 1980. Thermal Incinerators and
Flares.
54. Reference 52, p. 2-18.
55. Memorandum from Randall, D., Jones, R., and Nicholson, R.,
MRI, to Telander, J., EPA/MICG. April 25, 1995. Nationwide
Baseline Emissions for Combustion Sources at Kraft and Soda
Pulp Mills.
3-64
-------
56. Final Environmental Impact Statement: Kraft Pulp Mills--
Background Information for Promulgated Revisions to
Standards. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/3-85-020. May
1986.
57. Memorandum from Eddinger, J., EPA/ISB, to Durkee, K.,
EPA/ISB. July 8, 1983. Meeting minutes from the April 14,
1983 meeting between kraft pulp mill representatives, the
National Council of the Paper Industry for Air and Stream
Improvement, Inc., and the U. S. Environmental Protection
Agency. Discussion of problems mills have with TRS pickup
in wet-bottom ESP's using unoxidized black liquor.
Attachments.
58. Letter from Bringman, L., Environmental Elements Corp., to
Nicholson, R., MRI. October 12, 1993. Budgetary pricing to
convert from wet- to dry-bottom ESP.
59. RUST Environment & Infrastructure. MACT Cost Analysis
Report for NCASI. Prepared for the National Council of the
Paper Industry for Air and Stream Improvement, Inc. RUST
Contract No. 35-4735. October 15, 1993. 47 p.
60. Telecon. Nicholson, R., MRI, with Bringman, L.,
Environmental Elements Corp. November 30, 1994.
Information about the particulate return systems associated
with kraft recovery furnace ESP's.
61. Memorandum from Soltis, V., MRI to Telander, J., EPA/ISB.
July 14, 1992. Meeting minutes from the February 24, 1992
meeting between Gotaverken Energy Systems, Inc., the U. S.
Environmental Protection Agency, and Midwest Research
Institute.
62. Memorandum from Ramsey M. and Nicholson, R., MRI, to
Telander, J., EPA/ISB. November 3, 1992. Meeting minutes
from the June 24, 1992 meeting between the Babcock & Wilcox
Co., the U. S. Environmental Protection Agency, and Midwest
Research Institute.
63. Barsin, J., The Babcock and Wilcox Co., Barberton, OH; R.
Johnson, James River Corp., Camas, WA; and C. Rissler, James
River Corp., St. Francisville, LA. The St. Francisville
Recovery Low Odor Conversion and Capacity Upgrade.
(Presented at the International Chemical Recovery
Conference. Ottawa, Ontario, Canada. April 3-6, 1989.)
The Babcock and Wilcox Co. Technical paper BR-1372. 11 p.
64. Green R. and G. Hough (eds.). Chemical Recovery in the
Alkaline Pulping Process^,, 3rd Edition. Prepared by the
Alkaline Pulping Committee of the Pulp Manufacture Division.
Atlanta, GA, TAPPI Press. 1992. 196 p.
3-65
-------
65. Garner, J., Jaako Poyry. Conversion to Low Odor Improves
Recovery Boiler Efficiency and Life. Pulp and Paper.
63(7):91-95. July 1989.
3-66
-------
4.0 MODEL PROCESS UNITS, CONTROL OPTIONS, AND ENHANCED
MONITORING OPTIONS
This chapter defines model process units, identifies HAP
emission control options, and discusses enhanced monitoring
options for combustion sources in the chemical recovery area at
kraft pulp and paper mills. Model process units represent the
types of units that currently exist in the industry and the types
that may be constructed in the future. The control options
represent demonstrated emission control techniques, and the
enhanced monitoring options are methods of demonstrating
continuous compliance for a particular control option. The use
of model process units to characterize an industry allows the EPA
to evaluate the environmental and energy impacts and costs of
various control options for each combustion source. These
impacts and costs are presented in Chapters 5 and 6,
respectively.
4.1 MODEL PROCESS UNITS
This section presents the model process units that were
developed for recovery furnaces, SDT's, BLO units, and lime
kilns. The following parameters were evaluated to develop the
process units: process-specific characteristics, level of PM
emission control, stack gas characteristics, and HAP emission
levels. The model process unit parameters are typical for
equipment that currently exists or may be constructed in the
kraft pulp industry.
4.1.1 Recovery Furnace Models
The nine model process units that were developed to
characterize kraft recovery furnaces are presented in Table 4-1.
Two PM emission levels were developed for each of the nine models
4-1
-------
Tl
CO
Pi
w
^
y
3
CO
CO
H
O
§
D
g
H
CO
CO
0
0
ft
w
Q
2
^*
H
(St*
•^
i
fa
X
H
t)
•
O
H
1
^1^
rt
H
W
ro
1
8
21
V
•3
1
S
o
K
£
•s
S
i
s
o
D
0. u
a «3
a. *•
? s
*s* -a
ifl
o
.I1
X
§
CQ
® J>
E 0
•3)°°
1 jj>
§
P
O 13
is
£
a,
i*
H °
*j
•5 i
w
--1
0 «
§5 S
ato*
£
j,
0 U
s S
5 Z
,", 2
o
-o
£
ea
a
•o
cu
s
5
«T -a
23
c? ^
•c w
«=
O
•s "*
W
"3
11
o
to
VO
s
0
0
g
•§
£•
a.
Q
g
to
1
S
r*
o
U
z
li
o
0
rf
to
VO
CO
VO
Ov
-3-
s
s-
cu
1"
g
en
§
|
^.
o
u
o
Q
JO
t--
o
o
VO
g
s
w 's
I o
e u
> eu
-2 a
Is
1
a =
§ I
.0.
a
"
4-2
-------
CO
EH
H
PJ
CO
co
w
u
o
Pi
Ot
CO
EH
H
CO
rn
j PROCE
W
P
Q
ftt
D
Pn
Pi
W
g
U
H
Pi
co
H
i-J
1
.Q
i
1
S ^
o ""
'i ®
OH
3
1
§
8
1
^H
U
Q
i
cs
o
VO
CO
CO
S
S
en
|
CO
S
n.
a
1
g
s
t-
(S
U
Q
Of
O
0
en
P!
S
S
en
|
CO
Wet ESP system
o
1
t-
N
(D
U
Q
3
/)
Wl
OH
a
I
8
g
o\
CO
Ui
H
u-
oo
o
•0
55
OO
O
en
8
en
|
o\
i
OH
a
o
S
g
en
i
0
a)
H
tS
u.
a
S
0
CO
oo
o
CO
8
en
|
-
1
OH
a
1
o
a
g
en
s
S
Ov
O
W
H
is
ti
oo
o
0
w
g
rt
§
en
8
en
|
oo
0\
Wet ESP system
§
8
VJ
.— i
Ul
H
oo
04
oo
o
O
ss
en
CN
en
en
|
en
Wet ESP system
i?
i
c-
(S
H
UH
O
0
S!
en
0*
en
S
I
en
Wet ESP system
g
i
c-
CN
a
J3
O\
OH
1
x>
Cu
a
4)
C
S
0.
a
S
"a
J3
g
OH
a
4-3
-------
(RF-la and RF-lb, RF-2a and RF-2b, etc.). Model designations
ending in "a" represent pre-new source performance standards
(NSPS) furnaces with PM emissions above the NSPS PM limit
promulgated in 1978; designations ending in "b" represent
furnaces with PM emissions at or below the NSPS. Recovery
furnaces have been characterized based on the following
parameters: (1) type of final-stage black liquor evaporator
(i.e., DCE or NDCE); (2) size (i.e., BLS firing rate and pulp
production rate); (3) type of PM emission control device;
(4) stack gas characteristics (i.e., flow rate, temperature, and
moisture content); and (5) HAP emission levels (i.e., HC1,
methanol as a surrogate for gaseous organic HAP's, and PM as a
surrogate for PM HAP's). Each of these parameters is discussed
below.
4.1.1.1 Evaporator Type. The recovery furnace models were
characterized based on whether a DCE or an NDCE (i.e.,
concentrator) is used in the final stage of black liquor
evaporation. Models RF-l through RF-6 represent NDCE recovery-
furnaces, and models RF-7 through RF-9 represent DCE recovery
furnaces. Separate models were developed for these two types of
evaporators because the type of evaporator affects methanol
emissions. As discussed in Chapter 2, higher methanol emissions
are associated with DCE recovery furnace systems because methanol
can be stripped from the black liquor in the DCE and in the BLO
unit, which is only used with the DCE systems. Approximately
61 percent of existing recovery furnaces are NDCE recovery
furnaces, and the remaining 39 percent are DCE recovery furnace
systems.1 Based on the current trend in the industry, all new
recovery furnace installations are expected to be of the NDCE
design.
4.1.1.2 Recovery Furnace Size. Model recovery furnaces
were characterized by size, based on the BLS firing rate and
equivalent unbleached and bleached pulp production rates. The
BLS firing rates are expressed in terms of kg BLS/d (Ib BLS/d).
Equivalent unbleached pulp production rates are expressed in
terms of air-dried megagrams of unbleached pulp per day (ADMUP/d)
4-4
-------
(air-dried tons of unbleached pulp per day [ADTUP/d]). Similar
units (air-dried megagrams of bleached pulp per day [ADMBP/d] and
air-dried tons of bleached pulp per day [ADTBP/d]) are used to
express the equivalent bleached pulp production rates. The
equivalent pulp production rates are based on the BLS firing
rates and conversion factors of 1,500 kg BLS/ADMUP (3,000 Ib
BLS/ADTUP) and 1,800 kg BLS/ADMBP (3,600 Ib BLS/ADTBP).2'3
The models include three sizes of NDCE recovery furnaces and
three.sizes of DCE recovery furnaces. For the NDCE recovery
furnace models, the BLS firing rates are 0.7 million (MM),
1.2 MM, and 1.8 MM kg BLS/d (1.5 MM, 2.7 MM, and 3.9 MM Ib
BLS/d). Equivalent unbleached pulp production rates are 450,
820, and 1,200 ADMUP/d (500, 900, and 1,300 ADTUP/d). Equivalent
bleached pulp production rates are 380, 680, and 1,000 ADMBP/d
(420, 750, and 1,100 ADTBP/d). For the DCE recovery furnace
models, the BLS firing rates are 0.4 MM, 0.7 MM, and 1.2 MM kg
BLS/d (0.-9 MM, 1.5 MM, and 2.7 MM Ib BLS/d). Equivalent
unbleached pulp production rates are 270, 450, and 820 ADMUP/d
(300, 500, and 900 ADTUP/d). Equivalent bleached pulp production
rates are 230, 380, and 680 ADMBP/d (250, 420, and 750 ADTBP/d).
The distributions of BLS firing rates for DCE and NDCE
recovery furnaces are shown in Figures 4-la and 4-lb,
respectively. The sample populations of DCE and NDCE recovery
furnaces were divided into small, medium, and large model sizes.
The median BLS firing rate within each size range was then
selected as the production rate for the model. Figure 4-2a and
4-2b show the BLS firing rate ranges for DCE and NDCE recovery
furnaces, respectively, for the small, medium, and large size
categories and the corresponding medians for each size.
As shown in Figures 4-1 through 4-2, DCE recovery furnaces
tend to be smaller than NDCE recovery furnaces. The difference
in recovery furnace size reflects the fact that in recent years
(i.e., since 1980), mills have installed larger recovery
furnaces, and the majority of these mills selected the newer NDCE
recovery furnace technology.1'4 Consequently, the six model
recovery furnace sizes reflect the size variation between the DCE
4-5
-------
,<=2 ' >2, <=3 >3, <=4 >4, <=5
Range of sizes, million Ib BLS/d
>5
Figure 4-la. Size distribution for DCE recovery furnaces.
50
>0.4, <=1 >1.<=2 . >2, <=3 >3, <=4 >4, <=5 >5
Range of sizes, million Ib BLS/d
Figure 4-lb. Size distribution for NDCE recovery furnaces
4-6
-------
5-r"
CO
CD
.Q
o
CO
CD
§
O
2
LU
O
Q
Small DCE Medium DCE Large DCE
DCE recovery furnace model size
i i
I Median size
I Maximum size
Figure 4-2a. DCE recovery furnace model size ranges
Small NDCE Medium NDCE Large NDCE
NDCE recovery furnace model size
I j Minimum size JHH Median size
Maximum size
Figure 4-2b. NDCE recovery furnace model size ranges.
4-7
-------
and NDCE recovery furnaces. The 0.7 and 1.2 MM kg BLS/d (1.5 and
2.7 MM Ib BLS/d BLS firing rates for the medium (RF-8) and large
(RF-9) model DCE recovery furnaces correspond to the small (RF-1
and RF-4) and medium (RF-2 and RF-5) model NDCE recovery-
furnaces, respectively. A separate small DCE recovery furnace
model (RF-7) with a BLS firing rate of 0.4 MM kg BLS/d (0.9 MM Ib
BLS/d) and large NDCE recovery furnace models (RF-3 and RF-6)
with a BLS firing rate of 1.8 MM kg BLS/d (3.9 MM Ib BLS/d) were
also selected to characterize the existing population of kraft
recovery furnaces.
4.1.1.3 PM Emission Control Device. The PM emission
control device for all of the model, recovery furnaces is an ESP.
Electrostatic precipitators are the most prevalent PM emission
control device for recovery furnaces, installed on about
99 percent of recovery furnaces.1 The model recovery furnaces
are further characterized by the type of ESP system (i.e., with
or without black liquor or HAP-contaminated process water in the
ESP bottom or PM return system). The DCE recovery furnace
systems are represented by models RF-7 through RF-9; NDCE
recovery furnaces with dry ESP systems (i.e., dry-bottom ESP's
with dry PM return systems) are represented by models RF-1
through RF-3; NDCE recovery furnaces with wet ESP systems (i.e.,
wet-bottom ESP's or dry-bottom ESP's with wet PM return systems)
are represented by models RF-4 through RF-6. Nondirect contact
evaporator recovery furnaces equipped with wet ESP systems emit
higher quantities of gaseous organic HAP's than those equipped
with dry ESP systems, due to stripping of HAP's from the black
liquor or HAP-contaminated water in the ESP bottom or PM return
system. However, new ESP's tend to have dry ESP systems.5 For
NDCE recovery furnaces, models were developed for both types of
ESP systems. All of the DCE recovery furnace models have wet ESP
systems because 94 percent of DCE recovery furnaces are equipped
with wet-bottom ESP's, and information is not available to
determine if the remaining 6 percent that are equipped with dry-
bottom ESP's also have dry PM return systems.
4-8
-------
4.1.1.4 Stack Gas Characteristics. Model recovery furnaces
were characterized by stack gas characteristics (i.e., flow rate
at the ESP exit, temperature, and moisture content). The gas
flow rates were calculated from the BLS feed rates using
conversion factors of 13.9 MJ/kg BLS (6,000 Btu/lb BLS) and
0.242 dscm at 0 percent C>2/MJ (9,000 dscf at 0 percent
O2/106 Btu),3'6
The exhaust gas temperatures.used for-the model recovery
furnaces are based on ESP inlet gas temperatures from emissions
p
tests. Inlet.temperature is an acceptable estimator for outlet
temperature because the temperature does not vary significantly
across an ESP. For model NDCE recovery furnaces, the stack gas
temperature is 199°C (390°F). For model DCE recovery furnaces,
the stack gas temperature is 160°C (320°F).
Stack gas moisture contents for the recovery furnace models
are averages of measurements taken during emission tests of two
NDCE recovery furnaces and two DCE recovery furnaces. The
average stack gas moisture content based on the two NDCE recovery
furnace emission tests is 26 percent. The average stack gas
moisture content based on the two DCE recovery furnace emission
tests is 32 percent.
4.1.1.5 HAP Emissions. Model recovery furnaces were
characterized by the levels of HC1, methanol, which is a
surrogate for gaseous organic HAP's, and PM, which is a surrogate
for PM Hiy?'s. The HC1 emission levels for the model recovery
furnaces are based on the HC1 emission factor of 1.20 x 10"4 kg
of HCl/kg of BLS fired (1.20 x 10~4 Ib of HCl/lb of BLS fired)
presented in Chapter 2. As discussed in Chapter 2, HC1 emissions
from recovery furnace systems are highly variable, and
insufficient data exist to determine the process operating
parameters that significantly influence emissions. In addition,
the available data on HC1 emissions from NDCE and DCE recovery
furnaces show a significant overlap. Therefore, the same
emission factor is applied for both NDCE and DCE model recovery
furnaces.
4-9
-------
The methanol emission levels for model recovery furnaces are
also based on emission factors presented in Chapter 2. Methanol
emissions for the model DCE recovery furnaces are based on an
emission factor of 1.70 x 10'4 kg methanol/kg BLS (1.70 x 10'4 Ib
methanol/lb BLS). For model NDCE recovery furnaces with wet ESP
systems, methanol emission levels are based on an emission factor
of 5.04 x 10"5 kg methanol/kg BLS (5.04 x 10"5 Ib methanol/lb
BLS). For model NDCE recovery furnaces with dry ESP systems,
methanol emission levels are based on an emission factor of
4.93 x 10"6 kg.methanol/kg BLS (4.93 x 10'6 Ib methanol/lb BLS).
Separate methanol emission levels were developed because DCE
recovery furnaces have higher methanol emissions than NDCE
recovery furnaces due to the stripping of methanol from the black
liquor. Also, methanol emissions are higher from NDCE recovery
furnaces with wet ESP systems -than from NDCE recovery furnaces
with dry ESP systems, due to the same stripping effects.
The NSPS for kraft pulp mills establishes a PM emission
limit of 0.1 g/dscm (0.044 gr/dscf) at 8 percent O2 that is
applicable to recovery furnaces constructed, modified, or
reconstructed after September 24, 1976.8'9 Available data
indicate only the year, not the month, that recovery furnaces
were installed.1'2 Consequently, this analysis assumes that
those recovery furnaces installed during or after 1977 are
subject to the NSPS, and those recovery furnaces installed before
1977 are not subject to the NSPS. Also, recovery furnaces
subject to the NSPS are assumed to be in compliance with the NSPS
PM standard.
The "a" model recovery furnaces shown in Table 4-1
characterize those recovery furnaces with PM emissions above the
NSPS PM emission limit. Therefore, the "a" model PM emission
levels were based on the PM emission concentrations for those
recovery furnaces that are not subject to the NSPS and that emit
more than the NSPS PM limit. Different PM levels were calculated
for the DCE and NDCE recovery furnaces that are subject to these
parameters. For NDCE recovery furnaces, the PM emission level is
4-10
-------
0.27 g/dscm (0-. 12 gr/dscf) . For DCE recovery furnaces, the PM
emission level is 0.18 g/dscm (0.08 gr/dscf).10
The "b" model recovery furnaces represent those recovery
furnaces with PM emissions at or below the NSPS PM limit of
0.10 g/dscm (0.044 gr/dscf). The "b" model recovery furnace PM
emission level is equivalent to the NSPS limit and represents the
maximum emission level for these recovery furnaces. Although
only 30 percent of the recovery furnaces are subject to the NSPS,
the majority (approximately 80 percent) of all recovery furnaces
reportedly are meeting the NSPS PM emission level.1'2 Therefore,
the "b" models represent the majority of existing recovery
furnaces.
4.1.2 Smelt Dissolving Tank Models
The seven model' process units (SDT-1 through SDT-7) that
were developed to characterize existing SDT's are presented in
Table 4-2. Model process units SDT-1 through SDT-4 are also
representative of those SDT's that are expected to be constructed
in the future. The parameters selected to characterize typical
SDT's are (l) size (i.e., BLS,firing rate of the associated
recovery furnace, equivalent pulp production rate, and smelt
flow rate), (2) PM emission control device, (3) inlet and outlet
gas stream characteristics (i.e., flow rate, temperature and
moisture content), and (4) HAP emission levels (i.e., PM as a
surrogate for PMHAP's).
4.1.2.1 SDT Size. The SDT size is indicated by the BLS
firing rate and equivalent pulp production rate of the associated
recovery furnace and the smelt flow rate. The smelt flow rate
was calculated based on a conversion factor of 0.37 kg
smelt/kg BLS (0.37 Ib smelt/lb BLS).2 An equivalent BLS firing
rate was used as an indicator of SDT size because the SDT is an
integral part of the kraft recovery furnace. The SDT models
represent the majority of recovery furnace configurations, i.e.,
one SDT per recovery furnace. Approximately 8 percent of
recovery furnaces have two SDT's.1 The SDT models SDT-1 and
SDT-5 correspond to the small DCE recovery furnace model (RF-7);
SDT models SDT-2 and SDT-6 correspond to the medium DCE and small
4-11
-------
4-12
-------
NDCE recovery furnace models (RF-1, RF-4, and RF-8); SDT models
SDT-3 and SDT-7 correspond to the large DCE and medium NDCE
recovery furnace models (RF-2,rRF-5; and RF-9); SDT model SDT-4
corresponds to the large NDCE recovery furnace models (RF-3 and
RF-6) .
4.1.2.2 PM Emission Control Device. Two types of emission
control devices are predominantly used to control PM emissions
from SDT's--wet scrubbers and mist eliminators. Wet scrubbers
are used to control PM emissions from about 87 percent of the
SDT's, and mist eliminators are used to control PM emissions from
about 10 percent of the SDT's.11 The type of PM emission control
device will impact PM emission levels because wet scrubbers are
generally more effective than mist eliminators at controlling PM
emissions. The PM emission control device f or *SDT models SDT-l
through SDT-4 is a wet scrubber. These SDT models correspond to
all four recovery furnace model sizes and include sizes for both
DCE and NDCE recovery furnaces. The PM emission control device
for SDT models SDT-5 through SDT-7 is a mist eliminator. These
SDT models correspond to the three DCE recovery furnace model
sizes and the small and medium NDCE recovery furnace model sizes.
An SDT model with a mist eliminator that corresponds to the large
NDCE recovery furnace model size was not included because less
than 1 percent of recovery furnaces have this configuration.
4.1.2.3 Inlet and Outlet Gas Stream Characteristics. The
model SDT stack gas flow rates represent typical gas flow rates
within each size range. Available recovery furnace BLS firing
rates and corresponding SDT gas flow rates were used to develop a
factor of 1.04 x 10"5 m3/sec/kg BLS/d (0.01 acfm/lb BLS/d).2 The
model gas flow rates calculated with this factor were then
compared to, and found to be consistent with, the gas flow rates
at actual mills with comparable BLS firing rates.
The model stack temperature of 77°C (170°F) is based on
process information reported by mills and on measurements during
emission tests.1'12'1 The process information shows stack
temperatures range from 49° to 100°C (120° to 212°F). The
4-13
-------
emission test reports listed stack temperatures of 82°C (180°F)
and 76°C (168°F).
The stack gas moisture content for model SDT's is
36 percent, which is the average moisture content from four
emission tests.7 The average moisture contents of the stack
gases for the four SDT's that were tested are 35.5 percent,
46.7 percent, 25.4 percent, and 38.5 percent.7
The inlet gas flow rates were calculated from the stack gas
flow rates and estimated inlet temperature and moisture content.
The model inlet temperature of 93°C (200°F) was estimated based
*p
on available process information reported by individual mills.
4.1.2.4 HAP Emissions. Particulate matter emissions, as a
surrogate for PM HAP emissions, are characterized by the model
SDT's. The models characterize SDT's with PM emissions above the
NSPS PM emission limit for SDT's, i.e., 0.10 kg/Mg (0.20 Ib/ton)
BLS. Therefore, the model PM emission levels were based on the
PM levels for those SDT's that are not subject to the NSPS and
that emit more than the NSPS PM limit. The PM levels do not
represent the PM emission performance of typical SDT's because
approximately 75 percent of existing SDT's emit, on average, less
than the NSPS.2 The model SDT PM emission levels were selected
so that the emission reduction potential and control costs for
those SDT's with PM emission levels higher than the NSPS could be
evaluated.
Scrubbers are generally more effective at removing PM than
mist eliminators; therefore, different PM levels were calculated
for SDT's with wet scrubbers and SDT's with mist eliminators.
Average PM emissions are 0.18 kg/Mg (0.37 Ib/ton) BLS if a wet
scrubber is the PM emission control device, and 0.23 kg/Mg
(0.46 Ib/ton) BLS if a mist eliminator is the PM emission control
device.
4.1.3 Black Liquor Oxidation Unit Models
The three model process units that were developed to
characterize existing BLO units are presented in Table 4-3. With
the possible exception of the estimated one kraft pulp mill that
operates DCE recovery furnaces but does not have a BLO unit, no
4-14
-------
1
TO
E-"
H
&
TO
TO
W
O
O
CM
ft
K
Q
O
S
£-H
H
BnJ
TO
1 i PH
^^
C LIQUOR
PROCESS
*^-t
o
m
CJ
H
w
*G£*
S.
rt
m
i
^
S
§
^
1-
•5 .
s
•
•
1-
w ci
a -*
o
g.8
3 g
1?
3 §
0 ^
3 A
2 g
O •"
S|
I©
j.
a!
o 2 u
-^
^, «
3 '3
a. «i
Equivalent
production
S °
^.
03
Q
<
VDMUP/d
^*,
•o -£"
ea g
in
> u, 60
IF
o
>,
^
g
E
.£•
f
"o
•a
5
oo
O
t~
1
2
1
a
*o
en
°1
"*
O
a
§
|
d
Ml
"60
3
&,
^
'3
n
f
m
es
I
S
pa
CO
'""'
•8
1
§
O
p
to
en
"1
CO
O
CO
en
§
r-
C5
60
.S
3
CL,
i
"3
n
f
1
2
P3
T— t
ts
"2
1
g
J3
IT)
en
t-;
^-M
O
CO
CO
1
"
60
C
'So
u<
a
u.
'3
H
1
f
es
en
S
05
3
-------
new BLO units are expected to be built.11 The parameters
selected to characterize typical BLO units are (1) equipment
type, (2) size (i.e., equivalent BLS firing rate and pulp
production rate), (3) air pollution control device, (4) vent
characteristics (i.e., flow rate and moisture content), and
(5) HAP emission levels (i.e., methanol).
4.1.3.1 Equipment Type. Two types of BLO units are used at
kraft pulp mills--air-sparging units and molecular 02 systems.
Models were developed only for air-sparging units, which account
for an estimated 94 percent of the BLO units.1'11 The BLO models
were selected to characterize the majority of BLO systems;
therefore, BLO models were not developed for molecular 02
systems. Also, molecular 02 BLO systems are essentially closed
systems that do-not emit HAP's.
4.1.3.2 BLO Size. Three model sizes for BLO systems were
selected to correspond to the DCE recovery furnace model sizes.
Therefore, the BLS firing rates and pulp production rates for the
DCE recovery furnace models were selected as the parameters that
reflect the size of the BLO unit. The equivalent BLS firing
rates for models BLO-1, BLO-2, and BLO-3 are 0.4 MM, 0.7 MM, and
1.2 MM kg BLS/d (0.9 MM, 1.5 MM, and 2.7 MM lb BLS/d),
respectively.
4.1.3.3 Air Pollution Control Device. Air emissions from
approximately 95-percent of air-sparging BLO systems are
uncontrolled.11 Those few mills (about 2 mills) that control BLO
emissions incinerate the BLO vent gases in a power boiler. The
BLO models were selected to characterize the majority of BLO
systems; therefore, BLO model vent emissions are uncontrolled.
4.1.3.4 Vent Characteristics. The vent gas flow rates,
temperature, and moisture content for the model BLO units are
based on typical process data provided by mills for actual units
within each model size range. The vent gas flow rates for the
three model BLO units are 4.2, 8.5, and 12.7 m3/sec (8,900,
18,000, and 26,900 acfm).2 The exhaust gas temperature is 54°C
(130°F) .1:L The exhaust gas moisture content is 35 percent. '
4-16
-------
4.1.3.5 JHAP Emissions. As noted in Chapter 2, methanol
emission levels associated with the BLO models were calculated
using an emission factor of 1.73 x 10~4 kg/kg BLS
(1.73 x 10~4 Ib/lb BLS).
4.1.4 Lime Kiln Models
The six model process units (LK-1 through LK-6) that were
developed to characterize rotary lime kilns are presented in
Table 4-4. Current industry trends are (1) using ESP's rather
than scrubbers to control PM emissions and (2) longer and larger
diameter lime kilns.11 Therefore, the majority of new lime kiln
installations will likely be represented by the larger models
LK-5 and LK-6. The parameters selected to characterize lime
kilns are (1) size (i.e., equivalent pulp production rate, lime
production rate, length, and diameter), (2) PM emission control
device, (3) stack gas characteristics (i.e., flow rate,
temperature and moisture content), and (4) HAP emission levels
(i.e., PM as a surrogate for PM HAP's).
4.1.4.1 Lime Kiln Size. The lime kiln size is indicated by
the equivalent pulp production rate, lime (i.e., CaO) production
rate, length, and diameter.
The model sizes for lime kilns were determined using
available lime production rates. The size distribution, in Mg/d
(ton/d) of lime, for lime kilns is shown in Figure 4-3. The
sample population of lime kilns was divided into small, medium,
and large model sizes. The median lime production rate within
each size range was then selected as the lime production rate for
the model. Based on the data presented in Figure 4-3, the three
lime kiln model sizes are 90, 180, and 270 Mg/d (100, 200, and
300 ton/d) of CaO. The daily equivalent pulp production rate was
estimated using a conversion factor of 275 kg CaO/ADMP
(550 Ib CaO/ADTP).3
The length and diameter were determined based on the lime
production rates for the small, medium, and large models and
available data on lime kiln dimensions at similar production
rates.2 Lime kiln lengths range from 37 to 137 m (120 to 450 ft)
and average 82 m (270 ft); lime kiln diameters range from 2.1 to
4-17
-------
rt
W
i
u<
ro
rn
PROCE
_^
s
g
H
CO
CO
W
0
CM
n
Q
^
ks
s
n
H
t-1
*
^^^
O
H
I
ft
4.
K
t~
ff
<
E-<
j=0
"8© o
§3O
0 0
S5
ll"
S o
U
> •©
s> r
a-sfo
WJ-S? o
•^4 O
c^^g
_. o >
•^ o ^ Q
*§'g C 42©o
•i^sTs-Sov
§^ °° 2^
g|
"i-s
is
<^8
S
IH
C
U
5
S
J=
CXI
U
.ll
0.1
•a e _,
§ -2 §
> "5 3 'S S
3^ ^"Q f"
r?f fe ^
"o
•a
O o
Si
o
o
cn
O
o
C~
©
c-
-5J.
o
(-1
1
8
1
c-
vn
S
cs
cn
,_|
%
55
o
o
cn
O
O
©
rs
-1
,_i
S
j-
c
CO
3
1
O
CO
ex
CO
VO
cs
*
55
o
o
cn
O
o
f-
©
1— «
CS
r-<
co
1_i
o
|
•c
2
§
r-
£j
»— t
o
Q
?s
o
•—i
CO
"
T— C
o
W-l
cs
U
O
1
e
•
o
rt-
0
1
t-
VD
s
o
cs
CO
1Q
o
s
O
O
1
e
^(
S
,-.
o
cs
OH
W
O
CO
oc
o
CO
VO
-------
m
a
I
-H
O
O
-H
JJ
4-)
m
-H
T)
(1)
N
-H
CO
n
I
i 9LUj| ^O 'ON
4-19
-------
4.6 m (7 to 15 ft) and average 3.2 m (10.5 ft).11 The model lime
kiln dimensions were selected to cover these ranges.
4.1.4.2 PM Emission Control Device. Both wet scrubbers and
ESP's are used by the kraft pulp industry to control PM emissions
from lime kilns.11 Particulate matter emissions from the
majority of lime kilns (90 percent) are controlled with wet
scrubbers.11 Venturi scrubbers are the most prevalent type of
wet scrubber used (89 percent of all wet scrubbers.used to
control PM emissions are venturi scrubbers).11 Particulate
matter emissions from the remaining 10 percent of lime kilns are
controlled by ESP's.11 Lime kiln models were developed to
characterize both types of PM emission control devices. The PM
emission control device for models LK-1 through LK-3 is a venturi
scrubber. For models LK-4 through LK-6, the PM emission control
device is an ESP.
4.1.4.3 stack Gas Characteristics. The stack gas
flow rates for the lime kiln models with scrubbers represent
typical gas flow rates within each size range based on available
process data.2 The model gas flow rates for the small, medium,
and large models are 7.3, 14.2, and 24.1 m3/sec (15,500, 30,000,
and 51,000 acfm), respectively. The exhaust gas temperature of
71°C (160°F) is based on process data provided by mills. ' The
range of stack temperatures reported is 49° to 102°C (120° to
216°F).1:L The model exhaust gas moisture content of 30 percent
is based on test data.14
Model inlet gas flow rates were calculated from the stack
gas flow rates for the lime kiln models with scrubbers and
estimated inlet temperature and moisture content. An inlet
temperature of 250°C (480°F) was based on process data provided
by mills.11 The range of inlet temperatures is 150° to 315°C
(300° to 600°F)-11 An inlet moisture content of 25 percent was
assumed based on the moisture content from test results for a
lime kiln with an ESP.2 Because the gas stream temperature and
moisture content do not vary significantly across the ESP, the
inlet and outlet gas stream conditions are the same for the model
lime kilns with ESP's.
4-20
-------
4.1.4.4 .HAP Emissions. The model lime kilns shown in
Table 4-4 include PM emission levels as a surrogate for PM HAP
emissions. The PM emission level for model lime kilns with wet
scrubbers (LK-1 through LK-3) is 0.27 g/dscm (0.12 gr/dscf) at
10 percent 02.10 This emission level characterizes all lime .
kilns with PM emission levels above the NSPS PM emission limit
for lime kilns firing natural gas, i.e., 0.15 g/dscm
(0.067 gr/dscf); the emission level represents both natural
gas-fired and oil-fired lime kilns. The PM level does not
represent the PM emission performance of typical lime kilns
because approximately 64 percent of existing lime kilns emit less
than the NSPS limit for gas-fired lime kilns.2 The model lime
kiln PM emission level was selected so that the emission
reduction potential and control costs for those lime kilns with
PM emission levels higher than the NSPS could be evaluated.
Electrostatic precipitators are generally more effective at
removing PM than venturi scrubbers; therefore, different PM
levels were calculated for lime kilns with ESP's. Particulate
matter emission levels for lime kiln models with ESP's (LK-4
through LK-6) are equivalent to the NSPS level, 0.15 g/dscm
(0.067 gr/dscf). All existing lime kilns equipped with ESP's
reportedly have PM emissions less than or equal to the NSPS PM
limit for gas-fired lime kilns.2
4.2 CONTROL OPTIONS
Control options for recovery furnaces, SDT's, BLO units, and
lime kilns were developed based on the emission control
information presented in Chapter 3. This section identifies and
briefly describes these control options. Controlled emission
levels are presented for each applicable model process unit,
along with model control device parameters, where applicable.
4.2.1 Recovery Furnace Control Options
Table 4-5 presents the four control options that have been
evaluated for recovery furnaces. Table 4-5 also identifies the
recovery furnace type that is affected by the control option and
the pollutants that would be controlled. The four control
options are (1) conversion of a DCE recovery furnace system to an
4-21
-------
NDCE recovery furnace (i.e., low-odor conversion), (2) wet to dry
ESP system conversion, (3) PM controls, and (4) addition of a
packed-bed scrubber.
TABLE 4-5. RECOVERY FURNACE CONTROL OPTIONS
Control option
Affected recovery furnace type
Pollutants Controlled
Low-odor conversiona
DCE recovery furnaces
Gaseous organic HAP's and
PM HAP's
Wet to dry ESP system conversion
NDCE recovery furnaces with wet
ESP systems
Gaseous organic HAP's
(e.g., methanol)
PM Controls:
(a) ESP upgrade or replacement in
order to meet NSPS PM
emission level of 0.10 g/dscm
(0.044 gr/dscf)
(b) ESP upgrade or replacement
plus addition of packed-bed
scrubber in order to meet PM
level of 0.034 g/dscm
(0.015 gr/dscf).
NDCE and DCE recovery furnaces
currently emitting PM in quantities
above NSPS PM emission level
NDCE and DCE recovery furnaces
currently emitting PM in quantities
greater than 0.034 g/dscm
(0.015 gr/dscf).
PM HAP's
PM HAP's
Packed-bed scrubber
All recovery furnaces
HC1
aln addition to the elimination of the BLO unit and the conversion of the evaporator to the noncontact
design, the low-odor conversion option also includes an ESP upgrade or replacement (including wet to
dry ESP system conversion) to meet the applicable PM emission limit listed under "PM Controls."
4.2.1.1 Conversion of a DCE Recovery Furnace System to an
NDCE Recovery Furnace. Converting a DCE recovery furnace system
to an NDCE recovery furnace (or "low-odor conversion") was
evaluated as a control option for reducing gaseous organic HAP
emissions from DCE recovery furnace systems. Under this control
option, the DCE is eliminated from the chemical recovery process
and replaced with a concentrator, the BLO unit is eliminated, and
the wet ESP system is converted to a dry ESP system. Based on
the emission factors presented in Sections 4.1.1.5 and 4.1.3.5,
this control option reduces methanol emissions by 99 percent.
Because the DCE provides some PM control, as discussed in
Chapter 3, conversion to an NDCE recovery furnace has the
•potential to increase PM emissions. Therefore, depending on the
design characteristics of the ESP, an ESP upgrade or replacement
may be required to achieve compliance with PM emission limits.
4-22
-------
Two levels of PM control were included as part of the low-odor
conversion option. A PM control level of 0.10 g/dscm
(0.044 gr/dscf) at 8 percent 02 (i.e., the NSPS limit) was
evaluated for both "a" and "b" model recovery furnaces. Although
"b" model recovery furnaces have a baseline of 0.10 g/dcsm
(0.044 gr/dscf), conversion to an NDCE recovery furnace has the
potential to increase PM emissions above the baseline.
Therefore, PM controls are necessary to maintain a PM level of
0.10 g/dscm (0.044 gr/dscf) for the "b" model recovery furnaces.
A control level of 0.034 g/dscm (0.015 gr/dscf) at 8 percent G>2
was also evaluated for both "a" and "b" model recovery furnaces.
Tables 4-6 and 4-7 present the model recovery furnaces that were
evaluated for this control option. Controlled emission levels
also are included in these tables.
4.2.1.2 Wet to Dry ESP System Conversion. Converting wet
ESP systems to dry ESP systems was evaluated as a control option
for reducing methanol and other gaseous organic HAP emissions
from NDCE recovery furnaces. With this control option, stripping
of methanol from the black liquor or HAP-contaminated water in
the ESP bottom or PM return system is eliminated. Using the
methanol emission factors presented in Section 4.1.1.5 for wet
and dry ESP systems, methanol emissions are reduced by
72 percent. The model recovery furnaces used to evaluate the
impact of this option are presented in Table 4-•8.
4.2.1.3 PM Emission Controls. One PM emission control
option was evaluated that would reduce PM emissions from existing
NDCE and DCS recovery furnace systems to the NSPS PM limit. The
control option would involve either an ESP upgrade or an ESP
replacement to meet the NSPS limit of 0.10 g/dscm
(0.044 gr/dscf). The model recovery furnaces (NDCE and DCE "a"
models) analyzed for this control option are presented in
Table 4-9.
A second PM emission control option was evaluated that would
reduce PM emissions from NDCE and DCE recovery furnaces to a more
stringent level of 0.034 g/dscm (0.015 gr/dscf). The control
option would involve (1) upgrading the existing ESP and
4-23
-------
i1 §"•%
1«'3^
o e
0 S
E
£
ih
1
2
e
. ,
g
£
|
"S
O
j
0
2
e
c
U
U
s:
.w
_»j
•2- £
1.1
•| -f
g- 2
[2 a.
fl
a £
**
„-
0 §
•a S
'g "M
->1
U «
o. 8
a^
E
r- 0
O «H
11
°
g.S
's «
•5 ^
5 S
£ §
j?
. «T t>
E ^
eo m
c
•o
ca
§
S;
D
5
0
S o
0 0.
| o
o
rJ
•~*
o
*— i
d
oo
f—*
o
to
to
Cu
Q
E
>>
a,
^
S
S
o
0
s
g
m
H
Is*
fe
«n
-*
a
0
d
OO
~«
o
a
1
>
a,
£
2
s
cs
g
VO
1
w
Q
oj
o\
S
'5
.c
IS
O
S-s a"
11
<§e
i
a
-c
"S ^
.2 S
1"
s
•§3
1 |~
1
"S
S 'S -0
2 .a -a
•S C 04
0 8
U
ea* .^
SO ta
'S -o
PU w "r-
^ M
O
2*§
E 'S
u a,
62
J § s
I'll
1 SB
§ 2
^ o
0
JO
1|2
- 2 3
5
"5 "S
0. »
-j e
g .2
•« §
f S.
1
•o
CU
r^
"*
•o
ADTUP/
2 o
o a.
y-£-
,
° 6
S z
M-
•*'
o
m
?
O
d
oo
o
d
f
oT
1
:?
1
Q
I
ta
OH
*j
&•
S
o>
o
0
*
I
S
5?
8
ia
o
Q
at
jJ
(2
en
en
VO
S
„
d
oo
o
d
g
^
en
E
«"
A.
Q
I
ot
s
«
>
s
CN
0
t^»
1
U
o
Q
08
t*l
(2
S
CQ
S
8
r
4-24
-------
co
"8 ~s s"
c
- -_~T3
•S 8 w,
§ 8*.
03 .5
111
o
U
s-§-§
CU to
•3
«
o g
111
<3§ 8
§
^n. o
3 a
C, «-
•" C
•3 o
> 5
"II
tU CH
B3
s
2
u
CM
ts
u
OS
8
tt,
8.
ew
8
Q
a;
S
o
I
.S
*
I
|
a *j
oo c
«§s
HSS
rt pq
!l
1!
^l
1s3
|£
n >
Sfi
Jj
Jl
3-§
^•1
.si
H
U CO
'11
o ~
S £
A .£>
CD g
5CD
"H — of
a c §«
|i|S
III
•n o~
*3 -a .2 8
.g g S 5
u i
•3
«a « 5
a i .-*
o a ?
| g
11
i!
- -o
§< 2
U 0.
o
II
£
PI
to
s
o
d
o
d
o
d
o
d
00
o
S
o^
«
o
o
o
OH
OH
S
£>
Q
fr
a
o
0
•§
in
(S
o
m
O
U
W
O
w
o
u.
OH
On
(
t
tu
I
o
c
a
I
"o
J3
ttT1
4-25
-------
8
H
CO
Di
W
8
I
E-«
IS
CO
CQ
W
rt
»8
|g
Q O
O
O
3-s s
"s J-ll,
1 T
u '
<•
•
\
V
*
IS-
3 i*
o g
S
ft
f
t<
5 "3
3 u
3 g
i-il
3 W ^
i 1 "
-8-s
HI
S — «
^* Ui
11 ft
C
|
"o
•a
2
cu
Cu o
a. ii
1
>
f
^ d ">
S~* "to "^
- .2
§ g
.0
g 8
'.8 S
"2 *^
5 1
CM 0
u
•a
CM
CQ
S
5
•o
CM
D
S
Q
<
Ut
O -^ -j-j
f 22
•* e "_
s«^
I
2 «
o P-
Cu ^»
OS
"S
•a
S
en
en
*
ON
S
.S3
VI
PM
S
°
g
3
fr
Cu
1
O
00
en
O
S
f-
o
1
*z.
t
b
c-
o
vd
^
\o
09
VO
E
j>
"«
CM
00
&
Q
i
CA
CM
00
m
1
o
CO
vo
00
S
cs
~
U)
O
Q
Z
>o
ti
S
00
Tf
00
en
S
S
-2
CO
&H
w
Q
g
"S!
fr
CM
S
1
-T
1
|
oo
"
g
Q
Z
M>
[L
S
I
CO
85-
6
Ui
O
'5 S
il
1 §
— u
II
•8
ii e
I?
_o cu
U i-pi
11
U Ul
* CM
l!
o u
•a g.
2
o
H
CO
>
j^
0
o
w
CO
^H
co
co
w
p
o
m
5
^j
•• O
CO H
i— 1 EH
W CU
Q 0
Q
Hrv"
t-M
U EH
rt! &
*Z O
Pi O
g
>H
H
RECOV
CO "
H
o
S
w
A
co
i
sf
pq
EH
]3 -3
"o S
11
"3
S
^
1
st
o
53
CO
o
§
S
o
«
O *CJ
"Jo ^
1^
i .«
o 5
> a
0
^ eg
^ «
1 *,
2'1
1
9 -x
'a, S •§
1§ s
^
o
U
e
o
1
1
cu
•&.S
3 2
EquivalenI
*5j *O
CO
S
ctf
E3 cl>
o .a
•5 >
(A 4>
"2 *^
o "3
PM §
5
ADTBP
•n
ADTUP/<
w
tfj
^
Cl
cc
M"
C CQ
JS s2
2 u
o Pj
13
•a
^ °
^
t-:
vq
t-
g
s_
06"
OS
1
^
w
p^
Q
§
CO
CU
1
-§
S
S
VI
»-<
tt
c
•z,
t
b
e<
en
en
vo
en
1— H
g
8
^
CO
1
CO
w*
w
^
o
a
-------
CJ
CO
P
CD
H
o
' "
CO
*2J
o
H
Pi
O
i
&?
^
o
CJ
g
CO
W
Q
O
S
PQ
^
^
g
pH
>H
.ECOVER
OH
—
H
K
td
|
W
i
E-<
a
2
1
c3
i
fft
o f£
13
O
a
_o
i
-a
1
O-, „
*f3 -4-J
Equivalent
u
o
_o
"g
pa
a
icF
w"S?
o °°
'§©
o
-cf
C 0
O ®s>
'w °°
•3 ©
o g
CM "3
M
CO
en; -a
SB
« u
'I *
-2-1
W .g
C - —'
§ g
«s "S
Q_( O
o
^2
Ct,
pa
p
T3
ADMUP/i
00
*M
* O* "O
MCQ
*C
1C
o
2 «>
|§
W
•s
*tJ
o o
S z
o
0
r-~
cs
o
si
CL,
CO
UJ
o
oo
en
o
|
t-
o
ca
u
p
z
en
i
Pi
O
0
f^.
cs
o
CO
vo
1—1
CL,
CO
W
o
oo
VO
oo
1
^
CJ
CO
Q
CD
o
o
%
O
H
§
O
CJ
co
H
§
H
CTi
S^
CL, *"
"8 °°
11
§ s"
01
« v
§00
a
o
CS U
.11
11
1 "p
3
•a
2
a.
£-£
S
O
3
.3* 2 to
00
8
2 g
o P-
o
6
o
6
cs
cs
cs
oo
o
oo
o
00
O\
o\
>n
co
CM
co
PJ
8
a,
CO
co
(X
co
8
W
u
cs
U
U
Q
u
p
W
CJ
Q
t~-
cs
U
Q
B,
Pi
4-27
-------
installing a packed-bed scrubber or (2) installing a new ESP and
a packed-bed scrubber. The control option would be applicable to
both new and existing recovery furnaces. The model recovery
furnaces analyzed for this control option are presented in
Table 4-10.
Note that because the type of ESP bottom or PM return system
does not affect PM emission levels, this parameter was not
considered in evaluating either of the PM control options.
4.2.1.4 Packed-Bed Scrubber. The installation of
packed-bed scrubbers was evaluated as an HC1 control option for
existing and new NDCE and DCE recovery furnaces, and for DCE
recovery furnaces that have undergone a conversion to the NDCE
furnace design. Packed-bed scrubbers are capable of controlling
outlet HC1 emissions by 99 percent or to levels less than or
equal to 5 ppmv.16'17 This emission level (i.e., 5 ppmv)
corresponds to an emission factor of 6.20 x 10"5 kg HCl/kg BLS
(6.20 x 10"5 Ib HCl/lb BLS) for the model NDCE recovery furnaces
and converted DCE's. The corresponding emission factor for the
model DCE recovery furnaces is 6.53 x 10"5 kg HCl/kg BLS
(6.53 x 10"5 Ib HCl/lb BLS). Under this control option, HC1
emissions from the model NDCE recovery furnaces (and converted
DCE's) are reduced by 48 percent, and HC1 emissions from the
model DCE's are reduced by 46 percent. The model recovery
furnaces that were analyzed for this control option are listed in
Tables 4-11 and 4-12.
4.2.2 SDT Control Options
One PM emission control option was evaluated that would
reduce PM emissions from existing SDT's to the NSPS PM limit.
The control option would involve replacing the existing mist
eliminator or existing scrubber with a new wet scrubber designed
to meet the NSPS PM limit. The model SDT's analyzed for this
control option are presented in Table 4-13.
A second PM emission control option was evaluated that would
reduce PM emissions to a more stringent level of 0.06 kg/Mg BLS
(0.12 Ib/ton BLS). The control option would involve
(1) replacing the existing mist eliminator or scrubber with a new
4-28
-------
rt
^j-
S
co
Q
O
o
•
0
g
O
H
EH
n.
f-H
0
i_lj
8
1
o
2
CM
CO
W
P
O
S
o
^
fa
Pi
0
o
PJ
^
.^
o
H
g
EH
M-l
^>*
csj
o
t-3
1
w
pq
P_I
S ^
PH M- 00
""3 S (Sn
U O V3* _.
15 '8 g 0
& "§ OT
§ « 5
O w>
M-cT
C vs,
O sR
'« °°
•1 ®
c ^?
45 6
*S w
(^ *O
W)
i
t
s
"g .-g" „
o Pncn^ "S
*3 CO B *
V) W
O
« 8
O --H
•a j>
i-
§ 2
S "c
Q_) O
MH o
g
^o S
*3 *J J>-
5. 2 Q
eg <
«•> o
j« 'o -a
> 3 >-
fl i
§
i*
3 4-* •> —
_n* g co
•^ d
JS "S ^
3 ^
Ut
O
I o
o ».
W
d
-H
o
•^
en
O
O
cs
o
rf
si
§3
1
§
1
t-
o
o
Q
2
a
bu
Pi
o
o
o
o
Tf-
a
PH
CO
W
g
en
§
|
t-
O
8
p
z
JO
1
Pi
_x.
^•J
en
O
O
Si
o
oo
VO
T— <
PH
CO
W
00
vo
1
1
cs
__
U
P
rt
fT j
Pi
en
O
O
0
o
oo
~H
PH
00
o
oo
VO
1
s
cs
•^4
8
p
_o
[T t
oi
en
O
o
cs
o
9
cs
PH
CO
•"*
1
*~H
s
oo
_J
8
p
rt
CH
Pi
en
O
0
0
o
en
cs
1
r— t
8
•^-<
|
00
_
8
p
.0
[T .
K
en
o
0
oo
o
CS
vd
PH
£3
o
cs
o
55
|
Tj-
0
u
p
a
tin
Pi
en
O
o
0
o
cs
vd
PH
O
CS
|
|
Tf
0
8
p
^
fT l
Pi
en
O
O
oo
O
Tj-
en
o\
1
O
CO
en
O
|
f~
0
U
P
oi
CO
OH
Pi
en
O
0
o
o
•^~
en
o\
PH
O
co
en
O
1
t—
0
8
p
^
J"T .
s
en
O
O
oo
o
CO
*-«
PH
O
CO
vo
oo
1
cs
_l
8
p
a
o\
1
Pi
en
O
O
0
o.
00
VO
PH
£3
o
oo
vo
oo
1
CS
_i
8
p
J3
|
OH
Pi
<§
»—t
4-
u
S
_>
°5b
1
I
•*_»
D
g
"2
r
I
u
I
o
1
S
4-29
-------
O
CO
Q
CD
in
H
o
*
o
03
O
H
O
j
8
O
u
cu
w
Q
S
S
U
1
§
fe
XJ
ECOVEE
Oi
•
W
O3
H
O
S
2.
o
H
i
W
1
P-1 I/T 09
"So©
~ *5> <*- rC^
P .£2 o w
Js 3 *>
^2 n "o
(5 w>
w^ ^9
.1 "
'w (g
"g «g O
0 -3
§"B>
i|
2 °f
|1
^? fTi
C U
I'l
°l
CU °
"••« o
C
2 5
•3 a,
1 1
°* <
•S* °
g. 2
~s ^5.
1 P
t ^
§ «*5
:=* •" 2
J«i c* CO
^ -C! .0
S "^
^
ll
fTi
HI
d
"u
•a
o
0
o
cs
o
P
co
ON
CU
S
0
^
g
•g
^
U1
rt
U
Q
ri
S
IT)
O
o
^
o
P,
09
ON
»— i
CU
CO
W
o
"*
1
S
v>
^;
O
Q
1
o
o
co
o
P
t—
10
en
a,
W
o
^
1
S
t-»
cs
U
Q
&
1
10
0
o
I
o
g
t—
cu
W
0
t~
1
s
;g
t^
cs
S
Q
1
VI
O
O
rj
o
g
VI
Cu
W
g
1— 1
T— 1
g
en_
s
^
ON
en
O
Q
rt
i
V)
0
0
i
o
g
5?
CH
W
g
»— H
8
s
^
ON
en
O
S
V)
O
o
09
O
o
s
oT
cu
W
0
cs
g
en
s
S
ON
O
U
Q
rt
t~-
UH
Pi
Vl
O
o
o
o
g
°H
cu
CO
W
o
cs
g
en
,g
S
ON
O
W
o
Q
_0
V)
0
O
CO
o
o
g
09
ON
Ou
W
o
TT
i
g
S
V)
— ^
g
P
C3
09
[i
V,
O
o
o
o
g
CO
ON
T— <
CM
W
o
*
g
g
S
V)
^M
W
O
Q
A
09
Vl
O
o
09
O
0
g
en
Cu
W
o
t-
§
s
S
t-~
d
U]
o
P
fe
V)
O
o
1
o
g
f-
Vl
en
CH
W
o
1
s
§
t-
cs
s
P
XJ
4-30
-------
o
H
EH
O
O
O
o
ffi
0
H
rt
H
H
||
O ,w *
H *3
§§•*
"0
K
a"
.2
09 ^
o
a
1 S
c
§
C 1
vi-s
&l
•3
=
o
U
^E-c
tg'R 5
«0n
°w
§.«
31
5|
C
^
OH
£• ° ffl
"3 13 2
0. >H Q
1| <
Is 5
*S "O CU
§<£ D
w °- 2
Q
v, M
0-«
a gra
o WJS3
ej
z
•o
0
c
c
V£
ex
i
J
j
3
CO
•a
.£)
T
<
a.
eu
CO
[I
^.
fr
OS
OH
V)
tu
o
oo
0
vn
2
2
t~
o
^_,
(i
OH
t-
\f
t^
t-
1
J
J
3
ca
•a
.£>
"O
<
a
OH
a
tu
CO
vc
OH
co
tu
1
o
g
2
2
tx
eN
S
O
*—*
^
|
J
3
CO
t3
JO
T
(
a.
-i
0
u
m
eN
OH
CO
tu
|
r '
o
o
eN
*~*
2
2
CO
en
&
t~-
\l
e>
O
c
(
j
3
CO
•a
.0
•a
<
d.
OH
CO
tl
tN
V£
OH
co
tu
O
g
2
2
Tt
O
f.
1
tu
OJ
^
^;
VO
ex
UH
(
J
c
CO
•o
JO
i
a.
OH
CO
u
^
e*
OS
OH
r/i
tu
O
oo
en
o
2
2
t~
o
CO
L.
00
CTs
f^.
§
1
JO
I
Cj
ca
•a
JO
o
CO
tu
00
VO
OH
0
oo
vo
CO
2
2
eN
OS
&
I
'
§
I
'S
3
O
1
1
I
>
'§•
^j
0
H
E-t
o I
o
J
o
ptj
o
CJ
I-H
« ,
rn
t/J
Q P
ff
H D
0 Pi
«! C
Etj jt
m
>H '
Pi Q
W H
> «
O 0
u «;
pr"[ P[
Pi — '
•
w
CD
3
1
,Q
rH
*~^
4
pfj
£T?
l
2 «
11
co a,
Is
c fi
.2 •-
*M 5>
li
S o
°" 0
•o
OH
_g< jj n
"3 « f
0, VH Q
13 § <
« -g
1 = 5
tS* o. H
bD
>^ ^*j
u. J
§«.O
J*& 4)
-S w
CO
4
1
2
o
m
Os
o
CO
scrubber
1
1
a
C.
•|-
OH
CO
DJ
C
CO
Os
O.
O
^-
I
2
2
i-*
*7
L<
VI
~
T
f
CO
scrubber
1
e.
B
CL,
-{-
IX
CO
tu
o
0
o
a
OH
CO
W
o
r^
o
o
OS
2
2
CS
^
s
^
w
CO
v<
scrubber
packed-bed
-1-
o-
CO
tu
0
o
o
VI
OH
0
O
-^
1
2
2
OS
m
en
*>
CO
Cl
oo
c
~*
scrubber
1
1
a
&<
-f-
OH
CO
IU
§
Os
O-
O
VI
eN
§
cn
2
2
OS
o
17-
I-
OH
O
CO
OS
o
ex
• '
scrubber
•n
a
a
+
OH
CO
W
§
CO
Os
OH
CO
tu
0
•^J-
o
o
2
2
*— t
CO
I-
g
^j.
r
C"l
scrubber
packed-bed
~f*
£X
c/3
S
0
o
o
r-
VI
ri
OH
0
f^
o
0
OS
2
2
cs
OS
x!
4-31
-------
II
.
"S g
1 n .2 3
sSUs
o £
CJ S
CO*
•Z C
0 o-|5
H B .2 £
H
DJ
0
j
P3
FH *•— •»
o o
0 H
o w
o
•• o
CO
J Pi
w o
Q Q
S ?
ts
w o
O J
||
>^
Pi Pi
[x] H
^ CQ
o m
U JD
W Pi
Pi 0
co
• p
T* m
M-t M^
CO '
(ENGLI
; PACKED
03
H
i
W
g
o
"g
1
o
o
C O
i'*
o4
f
0
11
l-a"
a «
-------
M
3
3
i
I-
J3m
Ij CD
^ C5
— ' k^*
fl ^
3 H
-1 .
£~
-H
«H
EH
^0
^ J
J o
^
r S
-8
*
a
3
H
"o *
s''
u
§•
0*.;
'
Ci
e
j
*c
^
c
_c
*x
c
c
.frs
§,2
IJ
.§1
-C/
c ^™
* "55 W
^ OT *C
«il
n
5^2
5
Moisture
content. %
"o
5 ff-g
U) g cS
ill
a
3 -, 4)
i|.y
3 g 'Q
i ^ 5
' c
ll
^8
II
li
ADMBP/d
ADMUP/d
d
"3
o
o
o
O!
0
VO
en
n-
1
d
CO
g
scrubber
$
g
in
o
o
is
rt
/D
O
O
oo
o
VO
en
r-
1
2
w
n
Q
CO
o
o
en
c
0
VO
en
r-
u
d
I
•c
1
CO
Q
g
H.
s
o
CO
CO
H
Q
CO
CO
H
J
i
-^
Uj
ffl
EH
CQ
h-J
o
OJ
0
-a -co
* n J
•g .<-. 0 CO
2 2'w c
C t;
i— i_O CQ
•a c
.ll
C.
leg
cs
&o *
*•-» -*y
•9 "
S f ^
•|J -2
ij ° 8
§2§
C3 o
11
§1
21
o —
"3 .0"
•S*.2 £
•aa <:
8.2 <
"tti 0 _.
> 3 -2
•3"H Sr
ty S 3
WO. P
Q
d
"3
*o
0
o
o
^
o
VO
en
g
^
c7\
^
1
O
£
t-l
1
g
S
|
en
en
en
§
S
v
3
o
o
r-
o
VO
en
g
C*^
Vl
^-*
s
s
CO
£
*H
I
to
CD
g
in
in
o
VO
en
a
c^
Q
^
1
5
CO
£
t-(
-D
S
O
CO
s
1
1
S
1
t
p
o
o
VO
O
VO
en
g
~^-
rh
S
s
c
CO
£
(-1
O
B
C
'§
CO
S
§
en
en
en
8
n
*?
Q
CO
o
o
VO
o
VO
en
8
>n
(U
u:
M
^
u<
td
75
CO
S
g
in
in
in
i
n
VO
i
Q
CO
o
C5
VO
d
VO
en
g
en
oo
tN
1
CO
1
fH
P
"o
CO
§
1
1
n
1
rp
,0
4-33
-------
wet scrubber or (2) installing a new wet scrubber. The control
option would be applicable to both new and existing SDT's. The
model SDT's analyzed for this control option are presented in
Table 4-14.
4.2.3 BLO Unit Control Option
Two control options, (1) conversion of a DCE recovery-
furnace system to an NDCE recovery furnace and (2) incineration
of BLO vent gases, were evaluated for controlling gaseous organic
HAP emissions from air-sparging BLO units. Converting the DCE
recovery furnace eliminates the BLO unit from the chemical
recovery process. Thus, this control option results in a
100 percent reduction in emissions of HAP's, such as methanol,
from BLO units. As discussed in Section 4.2.1.1, the overall
methanol emission reduction for the recovery furnace system is
approximately 99 percent. The model BLO units were not used to
assess the impact of this control option; the economic and
environmental impacts, such as cost for equipment removal and the
associated emission reduction, were included in the impacts for
the DCE recovery furnace models.
The second control option to reduce gaseous organic HAP
emissions from BLO units is to incinerate the BLO emissions, most
likely in a power boiler. This control option reduces' methanol
emissions from the BLO unit by 98 percent and from the DCE
recovery furnace system by 49 percent. The model BLO units
analyzed for this control.option are presented in Table 4-15.
4.2.4 Lime Kiln Control Options
One PM emission control option was evaluated that would
reduce PM emissions from existing lime kilns to the NSPS PM
limit. The control option would involve replacing the existing
scrubber with a new ESP. Because lime kilns with ESP's are
already achieving the NSPS PM level, they are not included under
this control option. The model lime kilns analyzed for this
control option are presented in Table 4-16.
A second PM emission control option was evaluated that would
reduce PM emissions from lime kilns to a more stringent level of
0.023 g/dscm (0.010 gr/dscf). The control option would involve
4-34
-------
CO
W
S
•a
a;
n
be
•s
es
CO
bei
CO
bei
en
IT)
"
CO
es
co
VO
VO
.3
I
"Eb
I
-4_>
"3
•s
1
"8
ts
1
I
£
I
o
1
o
"S
I
co
w
Q
g
«
CO
H.
o
H
E-i
w
w
J
"8
o,.| 5
^ en E
P* « §
OT t3
O
|a
If
8-|.
"s 1'
^s
o *
(J&,
CO
S.8
1.2
d
TB
es
es
es
es
ts
es
cs
s
5?
VO
vo
vo
VO
CO
es
S
o
na
IT)
VI
0\
8
es
VT)
CO
Q
co
CO
CO
Q
CO
CO
4-35
-------
•3-3.
11*
si
•8
1
w
o
S
•*
o
.S
e
o
•S,
•i
oo
•a
u
1
u
1
i
to
"-S
c
1
I
>
I
CO
W
O
S
E-i
H
§ --»
g
§ H
2 H
OXIDAT]
NCINERA
£K LIQUOR
OPTION (I
« °
MH
f-H
J^O
w
CO j
S M
w p
111
5^*
i n
u I
H
i
^f
TABLE
t3
"S "3 -"
O o
e *
> CO
i §
° 1 1
"c
O
H £
5 "
1 1
1
^
(*
*
.
'
" E
sl
u
2
S
1 8
a'g
1-
S*|
^
r
0
Methanol control
device
1
e
0
•»
§
•o
2
a,
"3
a<
'S
"1
"3
o"
BJ
^
«
Equivalent b
,.,
^
M
g
T3
S
H
Q
^
«r
a,,
~
s5
(3 2
r
_^^^^^__^_
^H
CO
VO
c
•S
§
e
|
Uncontrolled
o
c^
o
c
o
C
a
ft
>n
1
e
§
"o
c
Uncontrolled
o
fv
^j
o
c
?
o
^
a
CO
OS
i
c
•S
1
J:
[ Uncontrolled
o
VI
r^
0,
o
OS
2
1
o
J
CO
4-36
-------
nS
i
cc
IT
*~
O
c/:
c
EH
0
1
O
u
§
CO
^
Q
i
§
H
W
S
H
.
CJ
H
EH
1
rt
H
i
EH
a
"O (to
55 „ ^
lls
. c!|@
o
a
If
§ "Q
45 £j
O
.S
PH 4j
- e —
•2 S
18
||
a|
o
e v
*~* o
o .a
|l
o o
S "3
OH o
0
1 s
C3
S
.a
r
a -o
.2 ~5b
nj 3 ^
° l£
S, 2
Mf!
d
"o
•g
d
e-5
d
OH
sa
s
wf^
s
o
CO
•c
"a
>
t-
\o
S
en
_1
d
o5
d
PH
00
S3
^O
I
•c;
2
1
CD
CO
o
CO
O
CO
vo
§
d
cs
d
00
W
Ui
JD
1
•c
I
en
o
o
1
en
VO
4
I
f2
.S
a
CD
'Eb
i
c
•s
I
3
CO
"3
o
I
•8
o
o
ID
1
I
_e
CO
1
a
§*
•_c
"u
C
CO
Q
py
P-i
CD
c~-
o
o
CO
C
1-
o
J
c
t~j
g
o
u
s
CO
M
1
§
H
w
w
H
.
w
co
H
J
CD
iz;
w
X!
V£>
H
i
w
EH
Controlled PM
missions, gr/dscf
@ 10% O9
u
s
•i 'S
CO CO
*3 -5
«s
o
en
8
en
8
*— i
r— (
en
W
J
4-37
-------
(1) replacing the existing scrubber with an ESP, (2) upgrading
the existing ESP, or (3) installing a new ESP. The more
stringent PM control option is applicable to both new and
existing lime kilns. The model lime kilns that were analyzed for
this control option are presented in Table 4-17.
4.3 ENHANCED MONITORING OPTIONS
The practice of enhanced.monitoring allows a facility to
demonstrate continuous compliance with the emission limits
established by the emission standard. The most direct means of
monitoring compliance is the use of continuous emission monitors
(CEM's) to measure the emissions of each pollutant on a
continuous basis. In the event that CEM's for specific
pollutants are not applicable because of cost and/or technology
constraints, alternative approaches to ensure continuous
compliance can be adopted. The best alternative approach is to
use CEM's to monitor surrogate pollutants with emission profiles
that closely match those of the pollutants of concern. Where
CEM's are not applicable for surrogate pollutants, the next best
option is to use process monitors to measure those process or
add-on control device operating parameters that impact emissions
of the pollutants of concern. In some cases, the very presence
of specific processing equipment will ensure continuous
compliance with the emission standard. Pollutant emissions can
also be tested on a periodic basis (e.g., semiannually).
The following sections describe how the approach described
above was used to develop the enhanced monitoring options for
combustion sources in the pulp and paper industry. Table 4-18
summarizes the enhanced monitoring options for recovery furnaces,
SDT's, BLO units, and lime kilns, based on the control options
presented in Section 4.2. Enhanced monitoring options were not
developed for control options other than those developed in
Section 4.2; if a CEM is not applicable, facilities that choose
to meet the emission limits through the application of other
control options must develop an enhanced monitoring plan that
demonstrates the ability of the selected parameter to gauge a
change in emissions.
4-38
-------
rt
p;
CT
ro
CN
0
•
0
g
o
H
EH
•04
O
-i
i— 1
0
EH
0
u
§
co
H
P
i
%
H
^4
H
H
^
^
CU
a
rt
rH
I
«
§
2 ^
!§ -2 © cv
•i '1 1
o ° 33
U ao
2§g
SJ:3
g 60
o
.2 g.§
I'll
*""* 6 t~~|
8 * 2
"S S "3
0^8
a °
O -r;
|l
g -g
g 1
g
jf
i
Q
G
j£
%>
£
a ^
9 1 s*
^ "o 1 8
•3-0 <
6
~S
•o
o
en
g
0
d
cu
w
IH
.§
J3
I
•c
2
g
t~;
es
,_(
vo
s
e^
en
^
en
g
d
d
Cu
w
1
•c
2
g
q
en
^j.
00
g
*•*
o
00
§
§
d
55
d
cu
CO
u
u
JO
I
CO
•c
t~
en
t-
O
0
o
d
cu
w
CM
%
(~~
en
t~.
o
0
eN
|
S
•-
o
IS
g
I
I
I
•8
8
o
o
1
g
1
1
•"^
C
CO
P
C5
o
H
0
0
CO
o
H
E-i
Ot
O
j
O
fH
0
U
g
* *
CO
w
g
s
s
H
i
H
.-i
i— J
»?
co
H
H
*"
X!
H
i
"^J
W
§
|a2
^ .2 © es
*O CO C4_ O
III
U 60
fc
S J 1
CU w ^
-------
1
s
o
EH
H
S5
SI
r-1
3
J
c
o
•f
(
*(
<
1
If '
I J
II 1
-
II 1
II •'
II '<
i
=
II
I
||
i?
§
!i
Opacity monitor
i
o
JO
1
£
o
g
a
c
•c
c
netcrs according
c
c
«
c
4
Cu
c/
U
o
be
c
'C
^
"E
<
cu
ffl
'3*
fi
o
•o
|
JQ
JO
|
ressure i
a
arameters (i.e.,
c
en
c
Monitoring of scrubber operati
,
j
i
i
o
1
1
c
*£
M
c
•n
c
£
c
c
'i
c.
c
"3
j
•c
4
7
4
~4
c
bC
c
•c
c
"E
<
O
c
•£
2
L4
O
o
"O
c
"5
c
JS
cu
II
c
c
•I
—
JO
CU
s
S
i
Equipment in place
Cu
|S
°,
w 8 £
Pli
o
Methanol CEM (e.g., FTIR)
1
'S
oj
E?
O
a"
Ik
OS
HCI CEM
1
1
•g
«3
5
^
LI
«j
xi
x
E
c
mrameters (i.e.,
bo
c
Monitoring of scrubber operat
.n
1
U
to
•3
i
*
u
C
'i
I
1
1
C
cc
"E
bO
c
"1
'c
o
E
«e
o
BO
C
1
[
u.
j
1
«
O-
4
t
J
"fl
O
be
c
•n
c
"S
<
1
c^
•o
o
JC
1
<
CM
e
c
•j
u
O
u
s
&.
0 *
o E
-------
4.3.1 Recovery Furnace Enhanced Monitoring
Enhanced monitoring options that can be used to demonstrate
compliance with recovery furnace emission limits for PM or PM
HAP's, total gaseous organic HAP's, and HC1 are presented in the
following sections.
4.3.1.1 Enhanced Monitoring for PM or PM HAP's Controlled
with an ESP. Because opacity is the surrogate measurement that
best characterizes the level of recovery furnace PM emissions,
installation of an opacity monitor after the ESP is one option
being considered as a means of demonstrating compliance with a PM
or PM HAP emission limit for recovery furnaces. For those
recovery furnaces with a wet scrubber following the ESP, an
opacity monitor must be located after the ESP but prior to the
scrubber. Method 5, Method 29, or Method 17 compliance tests
could be performed periodically as a substitute for an opacity
monitor.
Another option being considered is for the facility to
develop a monitoring plan that specifies ESP operating parameters
to be monitored. Operating parameters for the ESP would be site-
specific and would be based on the parameters measured during a
three-run, EPA Method 5, Method 29, or Method 17 compliance test
that showed the facility to be in compliance with the applicable
PM or PM HAP emission limit. Under this option, operation
outside the ranges of the ESP operating parameters would not
represent noncompliance with the applicable emission limit but
instead would require the facility to take corrective actions, if
necessary, to return the ESP parameters to the levels established
during the compliance test. The corrective action procedures
would be documented in the facility's startup, shutdown, and
malfunction plan.
4.3.1.2 Enhanced Monitoring for PM or PM HAP's Controlled
with a Wet Scrubber. For those recovery furnaces that can comply
with a PM or PM HAP emission limit with existing wet scrubbers,
the use of an opacity monitor to demonstrate compliance with the
PM or PM HAP emission limit may be inappropriate. The exhaust
from the recovery furnace wet scrubber will have a high moisture
4-41
-------
content and will interfere with the readings from an opacity
monitor. Monitoring scrubber operating parameters (i.e.,
pressure drop and scrubber liquid flow rate) is an alternative
enhanced monitoring option for showing compliance with a PM or PM
HAP emission limit for recovery furnaces. The pressure drop and
liquid flow rate are indirect measurements of the performance of
the scrubber. Pressure drop and scrubber liquid flow rate levels
would be site-specific and would be based on the operating
parameters measured during a three-run, EPA Method 5, Method 29,
or Method 17 compliance test that showed the facility to be in
compliance with the applicable PM or PM HAP emission limit.
Method 5, Method 29, or Method 17 compliance tests could also be
performed periodically as a substitute for monitoring scrubber
operating parameters.
4.3.1.3 Enhanced Monitoring for Gaseous Organic HAP's.
Control of gaseous organic HAP emissions from recovery furnaces
can be achieved by using NDCE recovery furnaces equipped with dry
ESP systems. Therefore, enhanced monitoring for recovery furnace
gaseous organic HAP emissions can be achieved simply by
confirming that the recovery furnace is an NDCE recovery furnace
with a dry ESP system. If the recovery furnace is a DCE recovery
furnace or an NDCE recovery furnace equipped with a wet ESP
system, the facility could measure methanol emissions with a
methanol CEM (e.g., a fourier transform infrared [FTIR]
spectroscopy monitoring system).
4.3.1.4 Enhanced Monitoring for HC1. Hydrochloric acid
emissions can be measured directly using an HC1 CEM. An HC1 CEM
could be installed after the packed-bed scrubber to demonstrate
continuous compliance with an HC1 emission standard. An HC1 CEM
could also be used after the ESP for those, recovery furnaces, that
could comply with an HC1 emission limit without a packed-bed
scrubber. The feasibility of using HC1 CEM's to demonstrate
compliance with an HC1 standard has not been determined. The low
HC1 concentrations and high moisture content associated with the
recovery furnace flue gas may make the use of HC1 CEM's more
4-42
-------
difficult. However, additional information is needed before an
HC1 CEM can be definitively ruled out for recovery furnaces.
Because HC1 emissions can be controlled with a packed-bed
scrubber, monitoring scrubber operating parameters is another
monitoring option being considered for those recovery furnaces
that comply with an HC1 emission limit using a packed-bed
scrubber. The scrubber operating parameters to be monitored are
the scrubber liquid pH and scrubber liquid flow rate. Scrubber
liquid flow rate and pH levels would be site-specific and would
be based on the operating parameters measured during a three-run,
EPA Method 26 HC1 compliance test that showed the facility to be
in compliance with an HC1 emission limit.
Alternative enhanced monitoring options are also available
to demonstrate compliance for those recovery furnaces that could
comply with an HC1 emission limit without a packed-bed scrubber.
One option would require the facility to develop a monitoring
plan that specifies operating parameters to be monitored. The
operating parameters would be site-specific and would be based on
the parameters measured during a three-run, EPA Method 26 HC1
compliance test that showed the facility to be in compliance with
an HC1 emission limit. Under this option, operating outside the
ranges of the operating parameters would not represent
noncompliance with the applicable emission limit but instead
would require the facility to take corrective actions, if
necessary, to return the parameters to the levels established
during the compliance test. The corrective action procedures
would be documented in the facility's startup, shutdown, and
malfunction plan. A second option would require periodic
Method 26 HC1 compliance tests to demonstrate compliance.
4.3.2 Smelt Dissolving Tank Enhanced Monitoring
This section presents the enhanced monitoring options that
can be used to demonstrate compliance with an SDT emission limit
for PM or PM HAP's.
4.3.2.1 Enhanced Monitoring for PM or PM HAP's Controlled
with a Wet Scrubber. Because the exhaust from the SDT wet
scrubber will have a high moisture content and will interfere
4-43
-------
with the readings from an opacity monitor, the use of an opacity
monitor to demonstrate compliance with a PM or PM HAP emission
limit for SDT's may be inappropriate. Monitoring scrubber
operating parameters (i.e., pressure drop and scrubber liquid
flow rate) is an alternative enhanced monitoring option for
showing compliance with a PM or PM HAP emission limit for SDT's.
The pressure drop and liquid flow rate are indirect measurements
of the performance of the scrubber. Pressure drop and scrubber
liquid flow rate levels would be site-specific and would be based
on the operating parameters measured during a three-run, EPA
Method 5, Method 29, or Method 17 compliance test that showed the
facility to be in compliance with the applicable PM or PM HAP
emission limit. Method 5, Method 29, or Method 17 compliance
tests could also be performed periodically as a substitute for
monitoring scrubber operating parameters.
4.3.3 Black Liquor Oxidation Unit Enhanced Monitoring
This section presents the enhanced monitoring options that
can be used to demonstrate compliance with a total gaseous
organic HAP emission limit for DCE recovery furnace systems
(which include the BLO unit).
One control option presented for the BLO unit involves the
removal of this piece of equipment from the chemical recovery
process by converting a DCE recovery furnace to an NDCE recovery
furnace equipped with a dry ESP system. Demonstrating that this
conversion has been completed assures compliance with the
applicable total gaseous organic HAP emission limit.
A second control option involves incineration of the BLO
emissions. Enhanced monitoring for BLO incineration could be
achieved simply by affirming that the BLO control equipment is in
place. Another enhanced monitoring option would be for the
facility to monitor the temperature of the power boiler or other
incineration device.
4.3.4 Lime Kiln Enhanced Monitoring
Enhanced monitoring options that can be used to demonstrate
compliance with a lime kiln emission limit for PM or PM HAP's are
presented in the following sections.
4-44 ,
-------
4.3.4.1 Enhanced Monitoring for PM or PM HAP's Controlled
with an ESP. Because opacity is the surrogate measurement that
best characterizes the level of lime kiln PM emissions,
installation of an opacity monitor after the ESP is one option
being considered as a means of demonstrating compliance with a PM
or PM HAP emission limit for lime kilns controlled with ESP's.
For those lime kilns with a wet scrubber following the ESP, an
opacity monitor must be located after the ESP but prior to the
scrubber. Method 5, Method 29, or Method 17 compliance tests
could be performed periodically as a substitute for an opacity
monitor.
Another option being considered is for the facility to
develop a monitoring plan that specifies ESP parameters to be
monitored. Operating parameters for the ESP would be site-
specific and would be based on the parameters measured during a
three-run, EPA Method 5, Method 29, or Method 17 compliance test
that showed the facility to be in compliance with the applicable
PM or PM HAP emission limit. Under this option, operation
outside the ranges of the ESP operating parameters would not
represent noncompliance with the applicable emission limit but
instead would require the facility to take corrective actions, if
necessary, to return the ESP parameters to the levels established
during the compliance test. The corrective action procedures
would be documented in the facility's startup, shutdown, and
malfunction plan.
4.3.4,2 Enhanced Monitoring for PM or PM HAP's Controlled
with a Wet Scrubber. For those lime kilns that can comply with a
lime kiln PM or PM HAP emission limit with existing wet
scrubbers, the use of an opacity monitor to demonstrate
compliance with a PM or PM HAP emission limit may be
inappropriate. The exhaust from the lime kiln wet scrubber will
have a high moisture content and will interfere with the readings
from an opacity monitor. Monitoring scrubber operating
parameters (i.e., pressure drop and scrubber liquid flow rate) is
an alternative enhanced monitoring option for showing compliance
with the applicable PM or PM HAP emission limit for lime kilns.
4-45
-------
The pressure drop and liquid flow rate are indirect measurements
of the performance of the scrubber. Pressure drop and scrubbing
liquid flow rate levels would be site-specific and would be based
on the operating parameters measured during a three-run, EPA
Method 5, Method 29, or Method 17 compliance test that showed the
facility to be in compliance with the applicable PM or PM HAP
emission limit. Method 5, Method 29, or Method 17 compliance
tests could also be performed periodically as a substitute for
monitoring scrubber operating parameters.
4.4 REFERENCES FOR CHAPTER 4
1. Memorandum from Nicholson, R., MRI, to Telander, J.,
EPA/MICG. June 13, 1996. Addendum to Summary of Responses
to the 1992 NCASI MACT Survey.
2. Memorandum from Soltis, V., MRI, to the project file
April 3, 1995. Kraft and Soda Pulp Mill Combustion Sources
Data Base.
3. Someshwar, A. Compilation of "Air Toxic" Emission Data for ,
Boilers, Pulp Mills, and Bleach Plants. National Council of
the Paper Industry for Air and Stream Improvement, Inc.,
New York. Technical Bulletin No. 650. June 1993. 128 p.
4. Green, R. and G. Hough, (eds.). Chemical Recovery in the
Alkaline Pulping Process. 3rd Edition. Prepared by_the_
Alkaline Pulping Committee of the Pulp Manufacture Division.
Atlanta, GA, TAPPI Press. 1992. 196 p.
5. Telecon. Nicholson, R., MRI, with Bringman, L,
Environmental Elements Corp. November 30, 1994. _
Information about the particulate return systems associated
with kraft' recovery furnace ESP's.
6. Memorandum from Soltis, V., MRI, to the project file.
October 20, 1994. Calculation of Recovery Furnace Stack Gas
Flow Rate.
7. Proposed Standards of Performance for Kraft Pulp Mills. In:
Standards Support and Environmental Impact Statement.
Volume 1 U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/2-76-014a.
September 1976.
8 U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 43 FR 7568.
Washington, DC. U.S. Government Printing Office.
February 23, 1978.
4-46
-------
10
11.
9. U. S. Environmental Protection Agency. New Source
Performance Standards for Kraft Pulp Mills. 51 FR 18538.
Washington, DC. U.S. Government Printing Office. May 20,
1986.
Memorandum from Holloway, T., MRI, to the project files.
June 14, 1996. Summary of PM and HAP Metals Data.
Memorandum from Soltis, V., Nicholson, R., and Holloway, T.,
MRI, to Telander, J., EPA/MICG. July 29, 1994. Summary of
Responses to the NCASI "MACT" Survey--Kraft and Soda Pulp
Mills (Data Base Summary Memo).
12. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Champion International
Corp.--Sheldon, Texas. Prepared for Texas Paper Industry
Environmental Committee. Report No. 06848-001-001.
January 1993. Volume 3.
13. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Simpson-Pasadena--Pasadena, Texas.
Prepared for Texas Paper Industry Environmental Committee.
Report No. 06848-001-001. January 1993. Volume 5.
14. Environmental Pollution Control, Pulp and Paper Industry,
Part I, Air. U. S. Environmental Protection Agency.
Cincinnati, OH. Publication No. EPA-625/7-76-001.
October 1976.
15. Roy F. Weston, Inc. Emissions Testing of Combustion
Processes in a Pulp and Paper Facility: Champion
International Corp.- -Roanoke Rapids, NC. Prepared for U. S.
Environmental Protection Agency--Emission Measurement
Branch. Research Triangle Park, NC. EMB Report
No. 92-KPM-27. October 1992. Volume I.
16. Telecon. Soltis, V., MRI, with Bruno, J., AirPol, Inc.
July 2, 1993. Information about scrubbers used to control
HC1 emissions.
17. Telecon. Soltis, V., MRI, with Sanders, D., Andersen 2000,
Inc. August 11, 1993. Cost and efficiency information
about HC1 control for recovery furnaces.
4-47
-------
-------
5.0 MODEL PROCESS UNIT ENVIRONMENTAL AND ENERGY IMPACTS
This chapter discusses the environmental and energy impacts
of controlling HAP emissions from new and existing combustion
sources at kraft and soda pulp and paper mills. The
environmental and energy impacts of each control option discussed
in Chapter 4 are presented for the individual model combustion
process units. The total nationwide environmental and energy
impacts associated with each control option are presented in a
separate memorandum.^
Section 5.1 of this chapter discusses the general approach
used to determine the environmental and energy impacts associated
with each control option. Sections 5.2 through 5.5 present the
environmental and energy impacts of each control option for model
process units representing recovery furnaces, BLO units, SDT's,
and lime kilns. Section 5.6 contains the references cited in
this chapter.
5.1 GENERAL APPROACH
This section introduces the types of environmental and
energy impacts associated with the control options and discusses
how these impacts were determined for each of the model process
units. The types of impacts discussed include air pollution,
energy consumption, water pollution, solid waste disposal, and
other impacts (i.e., noise, visual, odor, and irreversible and
irretrievable commitment of resources). Impacts were calculated
for each model process unit on an annual basis. In calculating
the annual impacts, all process units were assumed to operate
24 hr/d for 351 d/yr, which is equivalent to 8,424 operating
hr/yr. This operating time accounts for 14 days of scheduled
shutdown annually for maintenance and repair.
5-1
-------
The procedure for estimating air pollution'impacts is
described in the following section. The general approaches used
to develop energy impacts, water pollution impacts, solid waste
disposal impacts, and other impacts are provided in
Sections 5.1.2, 5.1.3, 5.1.4, and 5.1.5, respectively.
5.1.1 Air Pollution Impacts
This section presents the methodology used to determine both
primary and secondary air impacts.
5.1.1.1 Primary Emissions. For this impact analysis,
primary air impacts include the reduction of emissions directly •
attributable to the control option (i.e., the reduction of
emissions due to the use of APCD's or process modifications).
Primary emissions of PM, PM HAP's, gaseous organic HAP's, HC1,
and S02 were estimated based on model concentrations and emission
factors at baseline and at control levels. Gaseous organic HAP's
include acetaldehyde, benzene, formaldehyde, methanol, methyl
ethyl ketone, methyl isobutyl ketone, phenol, toluene, and
xylenes.
As discussed later in this chapter, PM is used as a
surrogate measure for PM HAP's, i.e, trace metals, such as
antimony (Sb), arsenic (As), beryllium (Be), cadmium (Cd),
chromium (Cr), cobalt (Co), lead (Pb), manganese (Ma), mercury
(Hg), nickel (Ni), and selenium (Se). The primary emission
estimates for PM were derived for each model recovery furnace and
lime kiln by multiplying the model PM concentration by the model
flow rate corrected to dry conditions at standard temperature
HC1.
gas
and by 351 operating d/yr.
Primary emission estimates of gaseous organic HAP's
and S02 from model recovery furnaces and PM from model SDT's were
determined by multiplying their respective model emission factors
by the model BLS firing rates and by 351 operating d/yr.
Primary emission estimates of PM HAP's were determined as
a percentage of PM for each model recovery furnace, SDT, and lime
kiln; the percentages were derived based on a comparison of PM
and PM HAP emissions for kraft and soda recovery furnaces, SDT's,
and lime kilns. The percentages of PM were estimated to be
5-2
-------
0.2 percent for recovery furnaces, 0.06 percent for SDT's, and
1.4 percent for lime kilns.2
The control level HC1 emission factor was derived from an
outlet HC1 emission concentration of 5 ppmv guaranteed by packed-
bed scrubber manufacturers for inlet HC1 concentrations less than
500 ppmv.3'4 However, the actual outlet HC1 emission level
achieved at a particular mill will be site-specific. The
baseline HC1 emission estimates (and percent HC1 reduction) in
this impact analysis may be underestimated because the baseline
HC1 emission factor includes emissions data from recovery
furnaces with HC1 emissions at or below 5 ppmv.
The S02 emission reduction obtained with the packed-bed
scrubber was estimated for each model recovery furnace by
applying a percent S02 reduction to average uncontrolled S02
emission estimates. Uncontrolled S02 emission factors for
recovery furnaces- have already been developed and are presented
in the Air Pollution Engineering Manual.5 As shown in the
manual, the average uncontrolled S02 emission factor for NDCE
recovery furnaces is 2.1 kilograms per air-dried megagram of pulp
(kg/ADMP) (4.2 pounds per air-dried ton of pulp [lb/ADTP]).5
Using a conversion factor of 1,700 kg BLS/ADMP (3,400 Ib
BLS/ADTP) (the average for both bleached and unbleached pulp
mills), the average uncontrolled S02 emission factor for NDCE
recovery furnaces is equivalent to 1.24 x 10"3 kg/kg BLS (1.24 x
10~3 Ib/lb BLS).6 The average uncontrolled SO2 emission factor
for DCE recovery furnaces is 1.8 kg/ADMP (3.5 lb/ADTP), which is
equivalent to 1.03 x 10"3 kg/kg BLS (1.03 x 10"3 Ib/lb BLS).5
Based on information from individual mills, at least 50 percent
S02 control was assumed to be achievable with packed-bed
rj
scrubbers. The control level S02 emission estimates were
determined using the uncontrolled emission factors and the
50 percent SO2 control.
To estimate the incremental reduction in emissions for each
primary pollutant (i.e., PM, PM HAP's, gaseous organic HAP's,
HC1, and S02), the control level emission estimate for each
5-3
-------
primary pollutant was compared to its corresponding baseline
emission estimate.
5.1.1.2 Secondary Emissions. Secondary air impacts include
the indirect or induced impacts resulting from implementing a
control option. These indirect or induced impacts include the
following:
(1) changes in power boiler emissions of criteria
pollutants such as PM, S02, nitrogen oxides (NOX), and carbon
monoxide (CO) resulting from the generation of energy required to
operate APCD's or other equipment included in the control
options; and
(2) changes in power boiler emissions of S02 resulting from
the incineration of TRS compounds present in the BLO vent gases
routed to the power boiler to control gaseous organic HAP
emissions.
The power boiler secondary emissions of PM, SO2/ NOX, and CO
resulting from the generation of electricity to operate APCD's or
other equipment were estimated based on emission factors related
to electricity usage. The emission factors used are
0.23 kilograms of PM per megawatt-hour (kg PM/MWh) (0.15 pounds
of PM per million Btu [Ib PM/MM Btu]) , 1.1 kg S02/MWh
(0.73 Ib S02/MM Btu), 0.45 kg NOx/MWh (0.29 Ib NOX/MM Btu), and
0.85 kg CO/MWh (0.55 Ib CO/MM Btu).8 The secondary emission
estimates were calculated as a product of these emission factors
and the model process unit electricity impacts discussed in the
next section.
Secondary SO2 emissions can be generated under the BLO vent
gas control option when TRS compounds present in the vent gases
are incinerated in a power boiler or other incineration device.
All of the TRS from the BLO unit is assumed to be routed to a
power boiler or other incineration device; none is "lost" as
condensate along the way. In deriving the chemical equation for
the conversion of TRS to SO2, the TRS was assumed to be in the
form of H2S. During combustion, the H2S would be combined with
02 to produce S02 and H20. The chemical equation is as follows:
5-4
-------
SO
H2O
As shown in this equation, one mole of S02 is formed for each
mole of H2S combusted. This molar ratio was used in the
following equation to estimate the mass ratio of S02 to TRS:
(1 kg-mole SO2 formed/'l kg-mole H2S combusted) x
(1 kg-mole H2S/34.08 kg H2S) x (64.06 kg SO2/kg-mole SO2)
~ 1.88 kg S02/kg H2S
The English unit equivalents for this equation are the same as
the metric unit values presented in the equation. To determine
the secondary S02 emission estimates, the ratio of SO2 to TRS was
multiplied by the estimate of TRS emissions incinerated by the
BLO control system. These secondary S02 emission estimates
represent worst-case estimates because some control of S02
emissions is likely to be achieved through the application of wet
scrubbers on incineration devices such as power boilers and lime
kilns that could be used to incinerate BLO vent gases.
To estimate the change in emissions for each secondary
pollutant (i.e., PM, S02, NOX/ and CO), the control level
emission estimate for each secondary pollutant was compared to
its corresponding baseline emission estimate.
5.1.2 Energy Impacts
The energy impacts of the combustion source control options
include changes in electricity and steam requirements. The
increases in electricity requirements from the operation of
APCD's were calculated using electricity cost equations from the
EPA's Office of Air Quality Planning and Standards (OAQPS)
Control Cost Manual.9 These electricity requirements are divided
into fan, pump, and operating electricity requirements and were
calculated assuming 8,424 operating hr/yr. The electricity
requirements to operate a BLO vent gas control system were
calculated based on information from a kraft pulp mill.10 The
methods used, to calculate the BLO control system energy impacts
are described in Section 5.3.2.
5-5
-------
The fan electricity requirement (applicable to ESP's and
scrubbers) is equal to a numerical factor (0.00018) times the
product of the gas flow rate, pressure drop, and operating
hr/yr>9 The gas flow rate varies with each model process unit.
The pressure drop is based on information from individual mills.
Although the pressure drop is not the sole parameter that
determines PM collection efficiency for lime kiln scrubbers, for
the purposes of estimating impacts, the pressure drop was used as
an indicator of PM collection efficiency for lime kiln scrubbers.
(Note: A different scrubber design, rather than a higher pressure
drop, was used to improve the PM collection efficiency for SDT
scrubbers.)
The gas flow rate and pressure drop do not change for ESP's
relative to current operation when the ESP's are upgraded or
replaced to improve PM collection. Therefore, the fan
electricity requirements for ESP's do not change relative to
current operation. However, if a scrubber is added after an ESP,
the gas flow rate would be reduced, thereby reducing the fan
electricity requirements for the ESP.
The pump electricity requirement (applicable to packed-bed
scrubbers) is equal to a numerical factor (0.000188) times the
product of the liquid flow rate, amount of head pressure, and
operating hr/yr divided by the pump-motor efficiency.9 The
liquid flow rate varies with each model process unit. A head
pressure of 18 m (60 ft) and a pump-motor efficiency of
70 percent were assumed.
The operating electricity requirement (applicable to ESP's)
is equal to a numerical factor (0.00194) times the product of the
ESP plate area and operating hr/yr.9 The ESP plate area is
calculated as a product of the exhaust gas flow rate and the ESP
SCA. The gas flow rate varies with each model process unit, and
the SCA is based on information from individual mills. Although
the SCA is not the sole parameter that determines PM collection
efficiency for ESP's, for the purposes of estimating impacts, the
SCA was used as an indicator of PM collection efficiency.
5-6
-------
Steam energy savings result from the increased steam flow
that occurs with the gaseous organic HAP control option of
converting the DCE recovery furnace system to an NDCE recovery
furnace.11 With NDCE recovery furnaces, more heat is available
for steam production than with DCE recovery furnaces because
combustion gases are not used to concentrate black liquor. The
increased steam production resulting from converting the DCE
recovery furnace system to the NDCE design reduces the power
boiler steam production requirements. A fuel savings can be
realized from this steam energy savings because the amount of
fuel required for steam production in the power boiler decreases.
The steam energy savings are partially offset by low-pressure
steam requirements for the concentrator, which is a part of the
NDCE recovery furnace system. The low-pressure steam would be
used to concentrate the black liquor. These concentrator steam
requirements were determined based on the low-pressure steam flow
rate and the change in enthalpy between the entering feedwater
and the steam leaving the superheater.1^
Another energy savings from converting the DCE recovery
furnace system to an NDCE recovery furnace is the electricity
saved by removing the BLO unit. The BLO operating energy savings
were determined by dividing the BLO electricity cost savings
presented in Chapter 6 by the unit cost of electricity in the
U. S. EPA Handbook: Control Technologies for Hazardous Air
Pollutants.13
The exact methods used to determine the energy impacts of
converting DCE recovery furnace systems to NDCE recovery furnaces
are presented in Section 5.2.3.2 and corresponding tables.
5.1.3 Water Pollution Impacts
Some of the control options for recovery furnaces, SDT's,
and lime kilns may have a significant impact on the amount of
wastewater generated, treated, and disposed. If lime kilns
reduce PM emissions by replacing the existing scrubber with a new
ESP, the existing scrubber discharge would be completely
eliminated. On the other hand, if recovery furnaces reduce HC1
emissions by adding a packed-bed scrubber, a new scrubber
5-7
-------
discharge would be generated. If SDT's reduce PM emissions by
replacing the existing PM control device with a new scrubber, the
wastewater discharge may be increased because of the increased
amount of PM in the scrubber liquid. The additional discharge
from the recovery furnace and SDT scrubbers may simply be
recycled and used elsewhere in the plant. Alternatively, the
scrubber discharge may require treatment and disposal. The
wastewater disposal impacts and the exact methods used to
estimate them are discussed for each of these control options in
their respective sections.
5.1.4 Solid Waste Disposal Impacts
The recovery furnace and lime kiln PM control options
include ESP's for PM control and, therefore, will generate a PM
catch. The PM catch from the ESP may simply be recycled back
into the process. Potential solid waste disposal impacts are
discussed for each of these control options in their respective
sections.
5.1.5 ni-.her Impact
Dther impacts <
r—.—* 4-Vi <-*' ci c-s-n+- -vr^n f"\7"\t~ n OT1 W
Impacts, visual impacts, odor impacts,
-WJ.J.a a..u.v~.i.u.v*>_
aoise : _
Irretrievable c
•x-P -v~A« /"M T T~f"«O G
doise impacts may occur as a :
:o accommodate the new or i
jccur as a result of 2
i.e.,
steam) <
j CLO Q. J- *—• *-* *•*••*- « *»* •*• —•——— ^j — ^f
T-irr-raded APCD's. Visual impacts may
.-, _he amount of PM and moisture
•u— -.-i—i /-.IT- ririi-it- n Trrna fl~s mav c
i result of i
.mpacts i
^ir^oY-rmQ TT?S emissions.
, U. S3 AtltAjr ^s^^*\*.^. V*.*-'
,= o,ari nn
__j_ \^\^^\^ j« -^^ WLkv' ^ —i™ —
-.r^,-; i =v>i o fpc! om-i ssions data.
V^VriL^S J-
rere ea:
nformation v
sily estimated quantitatively.
^< +- -t rv»r^ +- /= V» r> T C! d
^ O.J_\^»^J-*—
jj^ VVcto ctv&.J.a.cix/j-^s *-^-* %—i-f **—«.— '- — .
aantitatively; therefore, only a qualitative c
,\JL \^\^*+\mf*-
mpacts q
.f these impacts was done for e
pplicable.
,i j_ y t-t *^ wfc»p^p—w •« »"»-P •-- — - — — --
j_ _——. n ^.*-vi— -I r-N'M T.rln ^ Y*O
1DJ.6 .
ior impacts were determined for the gaseous organic I
T"* O *-*»vi T r^ f^ *i ^^Vl C! 3
ontrol options by comparing 1
ontrol levels for those control options.
CE and NDCE recovery furnaces were d
t baseline and at
. • J & v.-.wt rpTD O earm C! C! T OT1 C!
O
-------
data (in parts per million [ppm] as H2S) provided by individual
pulp and paper mills.6 These TRS concentrations were converted
to annual TRS emissions using the ideal gas law and the model BLS
firing rates, assuming 351 operating d/yr. Total reduced sulfur
emissions from BLO units are generally emitted in the range of
0.04 to 0.06 kg/ADMP (0.08 to 0.13 lb/ADTP).14 A mid-range value
of 0.05 kg/ADMP (0.10 lb/ADTP) was used to calculate the annual
TRS emissions from model BLO units based on the model pulp
production rates, assuming 351 operating d/yr. The mid-range
value for BLO units was used because the only data available for
BLO TRS emissions were the bounds of the range.
5.2 RECOVERY FURNACE CONTROL OPTIONS
The following sections discuss the model environmental and
energy impacts of implementing control options designed to reduce
emissions of PM (as a surrogate for PM HAP's), methanol (as a
surrogate for gaseous organic HAP's), and HC1. These control
options include PM controls (Section 5.2.1), wet to dry ESP
system conversion (Section 5.2.2), conversion of a DCE recovery
furnace system to an NDCE recovery furnace (Section 5.2.3), and
addition of a packed-bed scrubber (Section 5.2.4). Table 5-1
presents the model recovery furnace sizes and operating
parameters.
5.2.1 PM Controls
Two PM control options were evaluated for model NDCE
recovery furnaces RF-1 through RF-6 and model DCE recovery.
furnaces RF-7 through RF-9. The control options apply to both
new and existing recovery furnaces and are described below.
One PM control option that was evaluated would reduce PM
emissions from existing recovery furnaces to the NSPS level of
0.10 g/dscm (0.044 gr/dscf). The PM control option evaluated
would involve replacing or upgrading the recovery furnace ESP.
A second PM control option that was evaluated would reduce
PM emissions from existing recovery furnaces to 0.034 g/dscm
(0.015 gr/dscf). This more stringent PM control option would
involve replacing or upgrading the recovery furnace ESP and
adding a packed-bed scrubber. The second PM control option also
5-9
-------
applies to new recovery furnaces; the option could be used to
evaluate the impact on new sources subject to a more stringent
standard (0.034 g/dscm [0.015 gr/dscf]) than the current NSPS.
The PM control impacts for model NDCE recovery furnaces with
dry ESP systems (i.e., RF-1 through RF-3) are assumed to be
identical to the PM control impacts for model NDCE recovery
furnaces with wet ESP systems (i.e., RF-4 through RF-6) because
PM emissions are not affected by whether or not black liquor is
used in the ESP system. The environmental and energy impacts
associated with the PM control options are presented for the
model NDCE and DCE recovery furnaces in the following sections.
5.2.1.1 air Pollution Impacts. This section presents the
primary and secondary air impacts resulting from implementing PM
controls.
5.2.1.1.1 Primary emissions. Although emission test data
from recovery furnace ESP's on PM HAP performance are limited,
data collected from other combustion sources on the relative
performance of APCD's for. PM and PM HAP's indicate that systems
that achieve the greatest PM removal also provide the best
performance for the HAP portion of the PM.15 Therefore, PM
performance can be used as a surrogate for PM HAP's. Because
emission test data from recovery furnace ESP's indicate that PM
emissions are reduced with PM controls, PM HAP emissions would
also be reduced.6 As stated in Section 5.1.1.1, PM HAP emissions
from recovery furnaces were estimated to be 0.2 percent of PM
based on a comparison of PM and PM HAP emission data for recovery
furnaces.
Tables 5-1 and 5-2 present the operating parameters for
model NDCE and DCE recovery furnaces, respectively. Tables 5-3
and 5-4 present PM concentrations and PM HAP emission factors for
model NDCE and DCE recovery furnaces, respectively. Table 5-5
presents the annual PM and PM HAP emission estimates.
Figures 5-1 and 5-2 illustrate the annual PM emission estimates
for model NDCE and DCE recovery furnaces, respectively.
The baseline PM concentration for model NDCE recovery
furnaces RF-la through RF-6a is 0.27 g/dscm (0.12 gr/dscf).2 By
5-10
-------
controlling PM emissions from baseline to the NSPS level
(0.10 g/dscm [0.044 gr/dscf]), PM emissions would be reduced by
63 percent. On an annual basis, PM emissions would be reduced by
about 227 to 593 Mg/yr (250 to 654 ton/yr); PM HAP emissions
would be reduced by about 0.5 to 1.2 Mg/yr (0.5 to 1.3 ton/yr).
By controlling PM emissions from baseline to a more stringent PM
control level of 0.034 g/dscm (0.015 gr/dscf), PM emissions would
be reduced by 88 percent. On an annual basis, PM emissions would
be reduced by about 313 to 818 Mg/yr (345 to 902 ton/yr); PM HAP
emissions would be reduced by about 0.6 to 1.6 Mg/yr (0.7 to
1.8 ton/yr).
The baseline PM concentration for model NDCE recovery
furnaces RF-lb through RF-6b is the NSPS PM level (0.10 g/dscm
[0.044 gr/dscf]). By controlling PM emissions from the NSPS
baseline to 0.034 g/dscm (0.015 gr/dscf), PM emissions would be
reduced by 66 percent. On an annual basis, PM emissions would be
reduced by about 86 to 225 Mg/yr (95 to 248 ton/yr); PM HAP
emissions would be reduced by about 0.2 to 0.4 Mg/yr (0.2 to
0.5 ton/yr).
The baseline PM concentration for model DCE recovery
furnaces RF-7a through RF-9a is 0.18 g/dscm (0.08 gr/dscf).2 By
controlling PM emissions from baseline to the NSPS level, PM
emissions would be reduced by 45 percent. On an annual basis, PM
emissions would be reduced by about 65 to 194 Mg/yr (71 to
214 ton/yr); PM HAP emissions would be reduced by about 0.1 to
0.4 Mg/yr (0.1 to 0.4 ton/yr). By controlling PM emissions from
baseline to a more stringent control level of 0.034, g/dscm
(0.015 gr/dscf}, PM emissions would be reduced by 81 percent. On
an annual basis, PM emissions would be reduced by about 117 to
350 Mg/yr (129 to 386 ton/yr); PM HAP emissions would be reduced
by about 0.2 to 0.7 Mg/yr (0.3 to 0.8 ton/yr). The PM emission
reductions at both PM control levels are lower for model DCE
recovery furnaces than for model NDCE recovery furnaces because
the baseline PM concentration for model DCE recovery furnaces is
lower than that for model NDCE recovery furnaces.
5-11
-------
The baseline PM concentration for model DCE recovery
furnaces RF-7b through RF-9b is the NSPS PM level (0.10 g/dscm
[0.044 gr/dscf]). By controlling PM emissions from the NSPS
baseline to 0.034 g/dscm (0.015 gr/dscf), PM emissions would be
reduced by 66 percent. On an annual basis, PM emissions would be
reduced by about 52 to 156 Mg/yr (57 to 172 ton/yr) ; PM HAP
emissions would be reduced by about 0.1 to 0.3 Mg/yr (0.1 to
0 . 3 ton/yr) .
5.2.1.1.2 Secondary emissions. Secondary emissions of PM,
S02, NOX, and CO generated under the PM control options were
estimated for model NDCE and DCE recovery furnaces. Table 5-6
presents the annual secondary emission estimates. As shown in
the table, the secondary emissions generated under the PM control
options are small. The increases in secondary PM emissions are
especially insignificant compared to the reductions in primary PM
emissions. The more stringent PM control option includes both an
ESP upgrade and the addition of a packed-bed scrubber. The
increases in secondary S02 emissions under the more stringent PM
control option are insignificant compared with the large
reductions in primary S02 emissions that result from adding the
packed-bed scrubber. The reductions in primary S02 emissions
using a packed-bed scrubber are presented
discussed in Section 5.2.4.1.1
in Table 5-12 and
5.2.1.2
Impacts. The energy requirements of the PM
control options were estimated for model NDCE and DCE recovery
furnaces. The energy requirements were estimated based on model
SCA values for the ESP. Tables 5-1 and 5-2 present the baseline
and control level SCA values for model NDCE and DCE recovery
furnaces, respectively. Table 5-7 presents the annual energy
requirements
requirements
As shown in the table, the annual energy
of the PM control options are small,
on a
5.2.1.3 Water Pol.ini-.ion Impacts. Because the ESP operates
dry basis, no water pollution impacts are associated with an
ESP replacement or upgrade used to improve PM collection
Section 5.2.4.3 presents a discussion of the water pollution
5-12
-------
impacts resulting from adding a packed-bed scrubber (which is
included in the more stringent PM control option).
5.2.1.4 Solid Waste Disposal Impacts. As mentioned in
Chapter 2, the PM catch from the recovery furnace ESP is
primarily Na2S04 (i.e., saltcake) andNa2C03. These chemicals
are subsequently added to the concentrated black liquor in a mix
tank (i.e., recycled back into the process) in order to conserve
chemicals. Approximately 95 percent of the Na2SO4 is recovered.
The Na2S04 recycled back to the recovery furnace is reduced to
Na2S, a cooking liquor chemical, in the reducing zone of the
furnace. Reprocessing of the Na2S and Na2CO3 into cooking liquor
continues in the SDT. The recovery process was assumed to have
sufficient capacity to absorb the additional PM resulting from
the ESP replacement or upgrade. Thus, no solid waste disposal
impacts are expected with an ESP upgrade or replacement. Also,
no solid waste disposal impacts are expected with the addition of
a packed-bed scrubber (which is included in the more stringent PM
control option).
5.2.1.5 Other Impacts. Limited information was available
to estimate quantitatively impacts such as noise, visual, odor,
and irreversible and irretrievable commitment of resources.
Beneficial visual impacts are expected to result from the reduced
PM emissions coming out of the recovery furnace stack. However,
there also may be negative visual impacts under the more
stringent PM control option. The more stringent PM control
option would involve adding a packed-bed scrubber, which would
add to the moisture (i.e., steam) coming out of the stack.
Adding a packed-bed scrubber also would require additional
equipment (i.e., larger fans to overcome pressure drops and
pumps) that would increase noise levels. However, these
incremental noise increases are expected to be small compared to
the typical background noise levels at pulp and paper mills. The
other impacts, if any, are expected to be minimal as a result of
implementing the recovery furnace PM control options.
5-13
-------
5.2.2 Wet to Dry ESP System Conversion
Two control options were evaluated for reducing emissions of
gaseous organic HAP's such as methanol from existing NDCE
recovery furnaces. These control options are (1) converting an
ESP system that uses unoxidized black liquor or HAP-contaminated
process water in the ESP bottom or PM return system (referred to
as a wet ESP system) to an ESP system that uses "clean" water
(i.e., water uncontaminated with methanol and other gaseous
organic HAP's) in the ESP bottom or PM return system; and
(2) converting a wet ESP system to a dry-bottom ESP with a dry PM
return system (referred to as a dry ESP system). With these two
control options, the potential stripping of methanol and other
gaseous organic HAP's from the black liquor or HAP-contaminated
process water in the ESP system would be eliminated.
Only the impacts for the second control option, converting
from a wet to a dry ESP system, were evaluated. This decision
was based on (1) the lack of emissions data from ESP systems that
use "clean" water in the ESP bottom or PM return system; and
(2) the fact that very few mills use water in the ESP system.
The wet to dry ESP system conversion control option applies
to model NDCE recovery furnaces RF-4 through RF-6, which
represent existing NDCE recovery furnaces with wet ESP systems.
These models represent existing NDCE recovery furnaces only,
because no wet ESP systems are expected to be installed on new
NDCE recovery furnaces. The environmental and energy impacts
associated with this control option are presented in the
following sections.
5.2.2.1 Air Pollution Impacts: Primary Emissions. This
section presents the primary air impacts for model NDCE recovery
furnaces RF-4 through RF-6 resulting from a wet to dry ESP system
conversion. As discussed below in Section 5.2.2.2, no additional
electricity requirements are expected in order to operate a dry
ESP system. Therefore, no additional secondary emissions are
expected to be generated from operation of a dry ESP system.
The impact of the wet to dry ESP system conversion on
emissions of gaseous organic HAP's was evaluated. Table 5-3
5-14
-------
presents the gaseous organic HAP emission factor for the model
NDCE recovery furnaces. Table 5-8 and Figure 5-3 present the
annual gaseous organic HAP.emission estimates for model NDCE
recovery furnaces.
Gaseous organic HAP emissions are reduced by about
72 percent with a wet to dry ESP system conversion. On an annual
basis, gaseous organic HAP's are reduced by about 22 to 58 Mg/yr
(25 to 64 ton/yr). Approximately 49 percent of these reductions
result from the reduction of methanol emissions.
5-2-2.2 Energy Impacts. The tradeoff in horsepower
requirements between the wet and dry ESP systems is expected to
be approximately equal. For example, a wet-bottom ESP design
requires four agitators and a recirculation pump, whereas a
dry-bottom ESP design requires four drives for the drag system.16
Therefore, additional energy requirements are not expected in
order to operate dry ESP systems.
5-2.2.3 Water Pollution Impacts. The conversion to wet ESP
systems that use "clean" water in the ESP bottom or PM return
system was not considered as a control option for the reasons
cited in Section 5.2.2. Therefore, no water impacts are expected
for the wet to dry ESP system conversion control option.
5-2.2.4 Solid Waste Disposal Impacts. The wet to dry ESP
system conversion control option is not expected to have an
effect on PM control by the ESP. Furthermore, the PM catch from
the ESP is recycled back into the process, and the recovery
process is assumed to have sufficient capacity to absorb the
additional PM. Therefore, no solid waste disposal impacts are
expected for this control option.
5.2.2.5 Other Impacts. Limited information was available
to estimate impacts such as noise, visual, and irreversible and
irretrievable commitment of resources. The impacts, if any, are
expected to be minimal under this control option.
Odor impacts were determined for the wet to dry ESP system
conversion control option by comparing TRS emissions before and
after the ESP conversion. Wet ESP systems are different from dry
ESP systems in that they have the capability of increasing TRS
5-15
-------
18
emissions in the furnace gas stream by stripping TRS from the
black liquor.17 Table 5-9 and Figure 5-4 present the annual TRS
emission estimates.
The TRS emission estimates were derived from TRS emissions
data (in ppm as H2S) provided by individual pulp and paper
mills.6 Baseline TRS emissions were estimated based on the
10 percent trimmed mean TRS concentration (2.2 ppm) for NDCE
recovery furnaces assumed to be equipped with wet ESP systems.
This trimmed mean is a compromise between the mean and 'the
median. Using a trimmed mean with a moderate trimming
proportion, such as 10 percent, yields a measure which is neither
as sensitive to outlying values as the mean (since any small
number of outlying values will be deleted before averaging) nor
as insensitive as the median. The 10 percent trimmed mean was
estimated by eliminating the smallest 10 percent and the largest^
10 percent of the sample and then averaging what was left over.
Control level TRS emissions were estimated based on the available
TRS concentration data (1 ppm) for an NDCE recovery furnace known
to be equipped with a dry ESP system.6 Using these baseline and
control level numbers, it was estimated that TRS emissions would
be reduced by approximately 55 percent under the wet to dry ESP
system conversion control option. On an annual basis, the TRS
emissions for models RF-4 through RF-6 would be reduced by about
3.0 to 7.8 Mg/yr (3.3 to 8.6 ton/yr).
523 Conversion of a DCE Recovery Furnace System to an NDCE
Recovery Furnace
Converting a DCE recovery furnace system to an NDCE recovery
furnace (or "low-odor conversion") was evaluated as a control
option for reducing emissions of gaseous organic HAP's from DCE
recovery furnace'systems. The conversion of a DCE recovery
furnace system to an NDCE design involves removing the DCE and
BLO unit, adding a concentrator, and extending or replacing the
boiler economizer. Additional upgrades are included in the
low-odor conversion control option, i.e., an ESP upgrade to
improve PM collection and a wet to dry ESP system conversion to
reduce gaseous organic HAP emissions.
5-16
-------
At stated in Chapter 3, the DCE provides some PM control.
Therefore, with the removal of the DCE, the recovery furnace ESP
often must be upgraded or replaced during a low-odor conversion
in order to meet applicable PM emission limits. For the purposes
of this impact analysis, an ESP upgrade PM control option that
would maintain or reduce' PM emissions to the NSPS level of
0.10 g/dscm (0.044 gr/dscf) has been evaluated for existing DCE
recovery furnaces that have baseline PM emissions at or above the
NSPS level. This PM control option applies to model DCE recovery
furnaces RF-7a/7b through RF-9a/9b. These models represent
existing sources only, because no new DCE recovery furnaces are
expected to be built.
A PM control option that would reduce PM emissions to
0.034 g/dscm (0.015 gr/dscf) has also been evaluated for DCE
recovery furnaces that have PM emissions at or below the NSPS
level but greater than 0.034 g/dscm (0.015 gr/dscf). This PM
control option includes an ESP upgrade coupled with the addition
of a packed-bed scrubber and applies to model DCE recovery
furnaces RF-7b through RF-9b.
The environmental and energy impacts associated with the
low-odor conversion control option are presented for model DCE
recovery furnaces in the following sections.
5-2.3.1 Air Pollution Impacts. This section presents the
primary and secondary air impacts resulting from implementing the
low-odor conversion control option.
5.2.3.1.1 Primary emissions. The impact on emissions of
gaseous organic HAP's from converting a DCE recovery furnace
system to an NDCE design and converting the wet ESP system to a
dry ESP system was evaluated. The impact on PM and PM HAP
emissions from upgrading the ESP was also studied. Table 5-2
presents the operating parameters, and Table 5-4 presents the
model concentrations and emission factors for the model DCE
recovery furnaces. Table 5-5 and Figure 5-2 present the annual
emission estimates for PM and PM HAP's; Table 5-8 and Figure 5-3
present the annual emission estimates for gaseous organic HAP's.
5-17
-------
The PM and PM HAP emission reductions achievable by
upgrading the ESP for the model DCE recovery furnaces were
discussed in Section 5.2.1.1.1.
With a low-odor conversion, gaseous .organic HAP emissions
from model DCE recovery furnace systems (including BLO units)
would be reduced by about 93 percent; on an annual basis,
emissions would be reduced by 69 to 206 Mg/yr (76 to 227 ton/yr).
Approximately 70 percent of these reductions result from the
reduction of methanol emissions.
5.2.3.1.2 Secondary emissions. Secondary emissions
generated under the low-odor conversion control option were
estimated for model DCE recovery furnaces. Table 5-10 presents
the annual secondary emission estimates. The secondary emission
reductions resulting from the removal of the BLO unit and the
secondary emission increases resulting from the implementation of
the low-odor conversion PM controls are discussed separately
below.
As a result of the removal of the BLO unit, secondary
emissions are reduced for model DCE recovery furnaces RF-7
through RF-9. The secondary PM emission estimates range from
-567 to -1,100 kg/yr (-1,250 to -2,430 Ib/yr); the secondary S02
emission estimates range from -2,770 to -5,350 kg/yr (-6,110 to
-11,800 Ib/yr); the secondary NOX emission estimates range from
-1,100 to -2,130 kg/yr (-2,430 to -4,700 Ib/yr); and the
secondary CO emission estimates range from -2,090 to -4,050 kg/yr
(-4,600 to -8,920 Ib/yr).
The procedures for estimating secondary emissions associated
with the PM control options were presented previously in Section
5.2.1.1.2 and are the same for both converted and unconverted
model DCE recovery furnaces; therefore, only the differences
between the converted and unconverted models that impact
secondary emissions are discussed below. The model SCA values
for upgraded ESP's on DCE recovery furnaces converted to the NDCE
design are higher than the model SCA values for ESP's on
unconverted DCE recovery furnaces. Removal of the DCE, which
removes a portion of the PM prior to the ESP, is the primary
5-18
-------
reason for increasing the existing ESP's SCA after the low-odor
conversion. The increased SCA results in an increase in the
electricity requirements, and, subsequently, an increase in
secondary emissions. Although the gas flow rates from converted
DCE's are lower than the gas flow rates from unconverted DCE's
(which reduces the fan electricity requirements), the associated
decrease in electricity requirements is more than offset by the
larger increase in electricity requirements due to the higher
SCA. Therefore, the secondary emission estimates presented in
Table 5-10 for converted DCE recovery furnaces are higher than
those presented in Table 5-6 for unconverted DCE recovery
furnaces. For example, the incremental PM secondary emissions
for model RF-7a for the NSPS PM control option are equal to 90
kg/yr (199 Ib/yr) for a converted DCE recovery furnace and
45 kg/yr (100 Ib/yr) for an unconverted DCE recovery furnace.
Although secondary PM emissions are higher when the PM control
options are applied to the converted DCE recovery furnaces than
to the unconverted DCE recovery furnaces, the secondary emissions
are insignificant when compared to the primary PM emission
reductions achieved with the PM control options.
5-2-3-2 Energy Impacts. Energy impacts for the low-odor
conversion control option were estimated for model DCE recovery
furnaces. The energy impacts include electricity impacts from
both the elimination of the BLO unit and the PM controls and
steam energy impacts from the furnace conversion. Table 5-2
presents the baseline and control level SCA values used to
estimate the PM control electricity impacts for each model DCE
recovery furnace. Table 5-11 presents the annual estimates of
the energy impacts.
5.2.3.2.1 Electricity impacts. The incremental electricity
impacts from the elimination of the BLO unit and the low-odor
conversion PM controls are discussed separately below.
With the elimination of the BLO unit, the BLO operating
electricity savings were estimated to range from -2,450 to
-4,750 MWh/yr for models RF-7 through RF-9. The operating
electricity savings were estimated by dividing the annual BLO
5-19
-------
operating cost savings by the unit cost of electricity. The
annual BLO operating cost savings are discussed in Chapter 6 and
range from $147,000/yr to $285,000/yr for models RF-7 through
RP-9. The unit cost of electricity is $0.06/kilowatt-hour
(KWh).13 Most of the BLO operating costs (about 60 percent) is
for power to operate the blowers and pumps. The remaining
40 percent is for operating the reheater.
The procedures for estimating the electricity requirements
associated with the PM control options were presented previously
in Section 5.2.1.1.3 and are the same for both converted and
unconverted model DCE recovery furnaces; therefore, only the
differences between the converted and unconverted models that
impact electricity requirements are discussed below. As
discussed in Section 5.2.3.1.2, the model SCA values for upgraded
ESP's on converted DCE recovery furnaces are higher than the
model SCA values for ESP's on unconverted DCE recovery furnaces.
The increased SCA results in an increase in the electricity
requirements. Although the gas flow rates from converted DCE
recovery furnaces are lower than the gas flow rates from
unconverted DCE recovery furnaces (which reduces the fan
electricity requirements), the associated decrease in electricity
requirements is more than offset by the larger'increase in
electricity requirements due to the higher SCA. Therefore, the
electricity requirements for the PM control options are higher
for the converted DCE recovery furnaces shown in Table 5-11 than
for the unconverted DCE recovery furnaces shown in Table 5-7.
For example, the electricity requirements for model RF-7a for the
NSPS PM control option are equal to 391 MWh/yr for a converted
DCE furnace and 191 MWh/yr for an unconverted DCE recovery
furnace. Although electricity requirements are higher when the
PM control options are applied to the converted DCE recovery
furnaces than to the unconverted DCE recovery furnaces, these
requirements are insignificant when compared to the energy
savings obtained from removing the BLO unit or the energy
requirements of'pulp and paper mills.
5-20
-------
5.2.3.2.2 Steam energy impacts. Implementing the low-odor
conversion control option may increase steam energy production.11
Table 5-11 presents the estimates for steam energy production for
unconverted DCE recovery furnaces and the net increase in steam
energy production after a low-odor conversion. The increased
steam energy production for model DCE recovery furnaces RF-7
through RF-9 ranges from 1.97 x 108 megajoules per year (MJ/yr)
(1.87 x 105 million Btu per year [MM Btu/yr]) to 5.91 x 108 MJ/yr
(5.61 x 105 MM Btu/yr).
The net increase in steam energy production was estimated to
be equal to the steam production from a low-odor conversion minus
the steam requirements to operate the concentrator.
To determine the steam production from a low-odor
conversion, the increase in the thermal efficiency that results
from the low-odor conversion was estimated. The heat input to
the system was calculated using a BLS heat content of
13,900 kJ/kg (6,000 Btu/lb) of BLS fired.19 The increase in the
thermal efficiency that results from a low-odor conversion was
estimated to be 10 percentage points, based on average thermal
efficiencies of 56 percent for DCE recovery furnaces and
66 percent for NDCE recovery furnaces.20'21 Therefore, the
increase in the amount of heat input that can be converted to
steam with a low-odor conversion was estimated by multiplying the
heat input by the 10 percentage point increase in thermal
efficiency. The heat input value was divided by the thermal
efficiency for a power boiler (about 85 percent) to determine the
required energy input for a power boiler. The required energy
input is equivalent to the increased steam production. The
increased steam production was converted to annual steam
production estimates for each model DCE recovery furnace,
assuming 351 operating d/yr. The energy input for the power
boiler can be converted to fuel savings by using the heat
contents for specific power boiler fuels.
To determine the concentrator steam requirements, the steam
flow for each model DCE recovery furnace was multiplied by the
change in enthalpy between the entering feedwater (439.8 kJ/kg
5-21
-------
[189.2 Btu/lb]) and the steam leaving the superheater
(3,095.0 kJ/kg [1,331.5 Btu/lb]).12 The model steam flows were
based on a steam flow of approximately 4,500 kg/hr (10,000 ib/hr)
of low-pressure steam for a 1.1 million kg BLS/d
(2.4 million Ib BLS/d) furnace.22 The steam flow was
extrapolated for each model DCE recovery furnace size, assuming
that steam usage is proportional to the amount of black liquor
concentrated.
5.2.3.3 Water Po3 int-.i on Impacts. The conversion to ESP
systems that use "clean" water in the ESP bottom or PM return
system was not evaluated as a control option for the reasons
cited in Section'5.2.2. Therefore, no water impacts are expected
from the wet to dry ESP system conversion included under the
low-odor conversion control option. The water pollution impacts
resulting from adding a packed-bed scrubber (included under the
more stringent PM control option) are discussed in
Section 5.2.4.3.
5.2.3.4 Solid Waste Disposal Impacts. -As noted in
Section 5.2.1.4, the PM catch from the ESP is recycled back into
the process. The recovery process was assumed to have sufficient
capacity to absorb the additional PM resulting from upgrading the
ESP. Therefore, no solid waste impacts are expected to be
associated with the low-odor conversion PM control options.
5.2.3.5 nt-.her Impacts. Odor impacts were determined for
the low-odor conversion control option by comparing TRS emissions
before and after the low-odor conversion. Table 5-9 presents the
baseline TRS emission estimates for unconverted model DCE
recovery furnace systems (which include BLO units) and the TRS
emission reductions for converted model DCE recovery furnaces.
Figure 5-4 illustrates the annual TRS emission estimates.
In order to determine the baseline TRS emissions from DCE
recovery furnace systems, the baseline TRS emissions from DCE
recovery furnaces and BLO units were combined. The DCE recovery
furnace baseline TRS emission estimate was derived from TRS
emissions data (in ppm as H2S) provided by individual pulp and
paper mills.6 Based on the available TRS emissions data, the
5-22
-------
10 percent trimmed mean TRS concentration for DCE recovery
furnaces is approximately 12 ppm.6 The BLO unit TRS emission
estimate was derived from the pulp and paper NSPS supporting
document.14 Based on the bounds of TRS values given in the
supporting document, a median.value of 0.05 kg/ADMP
(0.10 Ib/ADTP) was used as the baseline TRS emission estimate for
BLO units. The control level TRS emission estimate (1 ppm) for
converted DCE recovery furnaces was derived from TRS emissions
data provided for an NDCE recovery furnace known to be equipped
with a dry ESP system.6 Emissions of TRS compounds are expected
to be reduced by approximately 94 percent under the low-odor
conversion control option. On an annual basis, the TRS emission
reductions would be 22 to 67 Mg/yr (25- to 74 ton/yr).
Limited information was available to estimate quantitatively
impacts such as noise, visual, and irreversible and irretrievable
commitment of resources. Beneficial visual impacts are expected
under the low-odor conversion control option. The beneficial
visual impacts include the reduction in PM emissions, the
elimination of the BLO vent gas stacks, and the reduction in the
amount of moisture (i.e., steam) coming out of the stack with the
elimination of the DCE. This reduction in moisture content could
be completely eliminated if the more stringent PM control option
is implemented. The more stringent PM control option includes a
packed-bed scrubber, which would increase the moisture content in
the stack. Adding a packed-bed scrubber also requires additional
equipment (i.e., larger fans to overcome pressure drops and
pumps) that would increase noise levels. However, these
incremental noise increases are expected to be small compared to
the typical background noise levels at pulp and paper mills. The
other impacts, if any, are expected to be minimal as a result of
implementing the low-odor conversion control option.
5.2.4 Addition of Packed-Bed Scrubber
The addition of a packed-bed scrubber downstream of the ESP
is included in two of the control options examined for recovery
furnaces. These control options are (1) the use of an ESP plus a
packed-bed scrubber to meet an outlet PM emission level of
5-23
-------
0.034 g/dscm (0.015 gr/dscf); and (2) the use of a packed-bed
scrubber to reduce HC1 emissions from recovery furnaces.
The impacts of combining an ESP upgrade or ESP replacement
with a packed-bed scrubber to control PM emissions to
0.034 g/dscm (0.015 gr/dscf) are presented in Section 5.2.1. The
impacts presented in the sections below are based on the use of a
packed-bed scrubber to control HC1 emissions from recovery
furnaces. The applicable model recovery furnaces for the packed-
bed scrubber HC1 control option are model NDCE and DCE recovery
furnaces RF-1 through RF-9. The model DCE recovery furnaces
include both DCE recovery furnaces converted to the NDCE design
and unconverted DCE recovery furnaces. Tables 5-1 and 5-2
present the sizes and operating parameters for model NDCE and DCE
recovery furnaces, respectively.
5.2.4.1 Air Pollution Impacts. This section presents the
primary and secondary air impacts resulting from adding a
packed-bed scrubber to control HC1 emissions from model recovery
furnaces.
5.2.4.1.1 Primary emissions. The impact of the packed-bed
scrubber control option on emissions of HC1 and another acid gas,
S02, was evaluated for model recovery furnaces. Tables 5-3 and
5-4 present the HC1 and S02 emission factors for model NDCE and
DCE recovery furnaces, respectively. Table 5-12 presents the
annual emission estimates for HC1 and S02; Figures 5-5 and 5-6
illustrate the annual emission estimates for HC1 and S02,
respectively.
The incremental HC1 emission reductions for model NDCE and
converted model DCE recovery furnaces RF-1 through RF-9 range
from 8.3 to 36 Mg/yr (9.1 to 40 ton/yr). The incremental HC1
emission reductions for the unconverted model DCE recovery
furnaces RF-7 through RF-9 are slightly lower because of
different furnace characteristics (i.e., higher model gas flow
rates). The HC1 emission reductions for unconverted models RF-7
through RF-9 range from 7.8 to 24 Mg/yr (8.6 to 26 ton/yr).
The incremental S02 emission reductions for model NDCE and
converted model DCE recovery furnaces RF-1 through RF-9 range
5-24
-------
from 59 to 384 'Mg/yr (65 to 423 ton/yr). 'The incremental S02
emission reductions for the unconverted model DCE recovery
furnaces RF-7 through RF-9 are slightly higher because the
control level S02 emission factor associated with DCE recovery
furnaces is lower than the control level SO2 emission factor
associated with NDCE recovery furnaces. The SO2 emission
reductions for unconverted models RF-7 through RF-9 range from 74
to 221 Mg/yr (81 to 244 ton/yr). Overall, the S02 emission
reduction achievable with a packed-bed scrubber is approximately
10 times greater than the HC1 emission reduction.
Based on information from a scrubber manufacturer, emissions
of gaseous HAP's such as methanol, formaldehyde, and acetaldehyde
may be reduced with a packed-bed scrubber, but no emissions data
were available to estimate potential reductions for these
HAP's.23 Therefore, HC1 and S02 were assumed to be the only
gaseous pollutants reduced with a packed-bed scrubber.
5.2.4.1.2 Secondary emissions. Secondary emissions
generated under the packed-bed scrubber control option were
estimated for model recovery furnaces. The baseline secondary
emissions for the packed-bed scrubber control option are equal to
zero (i.e., no packed-bed scrubber at baseline). Table 5-13
presents the incremental secondary emissions relative to baseline
for each model recovery furnace.
The increases in secondary PM emissions from operating a
recovery furnace packed-bed scrubber are insignificant compared
to the reductions in primary PM emissions from implementing the
recovery furnace PM control options. The secondary PM emission
estimates for model NDCE and converted DCE recovery furnaces RF-1
through RF-9 range from 138 to 594 kg/yr (304 to 1,310 Ib/yr).
The secondary PM emission estimates for unconverted model DCE
recovery furnaces RF-7 through RF-9 range from 145 to 435 kg/yr
(320 to 960 Ib/yr).
The increases in secondary S02 emissions from operating a
recovery furnace packed-bed scrubber are insignificant compared
to the reductions in primary S02 emissions from operating a
packed-bed scrubber. For model NDCE/converted DCE recovery
5-25
-------
furnaces RF-1 through RF-9, the secondary S02 emission estimates
for the packed-bed scrubber control option range from 671 to
2,900 kg/yr (1,480 to 6,390 Ib/yr) . The secondary SO2 emission
estimates for unconverted model DCE recovery furnaces RF-7
through RF-9 range from 708 to 2,120 kg/yr (1,560 to
4,670 Ib/yr) .
For model NDCE and converted DCE recovery furnaces RF-1
through RF-9, the secondary NOX emission estimates range from 266
to 1,150 kg/yr (587 to 2,540 Ib/yr). For unconverted model DCE
recovery furnaces RF-7 through RF-9, the secondary NOX emission
estimates range from 281 to 844 kg/yr (619 to 1,860 Ib/yr).
For model NDCE and converted DCE recovery furnaces RF-1
through RF-9, the secondary CO emission estimates range from 503
to 2,190 kg/yr (1,110 to 4,820 Ib/yr). For unconverted model DCE
recovery furnaces RF-7 through RF-9, the secondary CO emission
estimates range from 531 to 1,600 kg/yr (1,170 to 3,520 Ib/yr).
5,2.4.2 Kngray Impacts. The increases in energy impacts
for the packed-bed scrubber control option were determined for
model recovery furnaces. The baseline energy level for the
packed-bed scrubber control option is equal to zero (i.e., no
packed-bed scrubber at baseline). Table 5-14 presents the
incremental energy impacts relative to baseline.
The energy impacts for the packed-bed scrubber control
option range from 593 to 2,570 MWh/yr for NDCE/ converted DCE
models RF-1 through RF-9 and 625 to 1,870 MWh/yr for unconverted
DCE models RF-7 through RF-9.
5,2.4.3 Water Pollution Impacts. Adding a packed-bed
scrubber creates a new wastewater stream for recovery furnaces,
thereby affecting the water balance at the mill. Whether there
are any significant wastewater disposal impacts for this option
depends on whether or not the scrubber discharge could be
recycled and reused elsewhere in the mill. Wastewater impacts
were estimated for model recovery furnaces assuming there was no
prior recycle or reuse. Tables 5-1 and 5-2 present the model
wastewater flow rates used to estimate wastewater impacts for
model NDCE and DCE recovery furnaces, respectively. The baseline
5-26
-------
wastewater impacts for the packed-bed scrubber control option are
equal to zero (i.e., no packed-bed scrubber at baseline).
Table 5-15 presents the incremental wastewater impacts relative
to baseline.
Information was available from two sources to determine the
amount of wastewater generated by the packed-bed scrubber control
option. Wastewater flow rate values were estimated using
procedures in the OAQPS Control Cost Manual and using wastewater
flow rate values provided by a scrubber manufacturer.9'23 The
wastewater flow rate values calculated using the OAQPS Control
Cost Manual are presented in Chapter 6 in Table 6-23. A scrubber
manufacturer provided wastewater flow rate values of 21.24,
53.07, and 84.79 liters per minute (L/min) (5.61, 14.02, and
22.40 gallons per minute [gpm]) for three model recovery furnaces
with scrubber inlet gas flow rates of 47.2, 118, and 189 m3/sec
(100,000, 250,000, and 400,000 acfm), respectively.23 The
wastewater flow rates from the scrubber manufacturer were
corrected to correspond to the model sizes used in this impact
analysis and then were averaged with the wastewater flow rates
obtained using the OAQPS Control Cost Manual to determine the
average wastewater impacts for the packed-bed scrubber control
option. The increased wastewater impacts presented below for the
recovery furnace packed-bed scrubber control option are
insignificant compared to the reduced wastewater impacts
presented in Section 5.5.1.3 for the lime kiln PM control option,
which involves replacing existing wet scrubbers with ESP's.
The wastewater impacts for NDCE/converted DCE models RF-1
through RF-9 range from 6.1 to 26 million liters per year (L/yr)
(1.6 to 7.0 million gallons per year [gal/yr]). The wastewater
impacts for unconverted DCE models RF-7 through RF-9 range from
6.3 to 19 million L/yr (1.7 to 5.0 million gal/yr).
5.2.4.4 Solid Waste Disposal Impacts. No solid waste
disposal impacts are expected from adding a packed-bed scrubber.
5.2.4.5 Other Impacts. Adding a packed-bed scrubber
requires additional equipment (i.e., larger fans to overcome
higher pressure drops and pumps) that would increase, noise
5-27
-------
levels. However, these incremental noise increases are expected
to be small compared to the typical background noise levels at
pulp and paper mills. Limited information is available to
estimate quantitatively the visual, odor, and other impacts.
However, because adding a packed-bed scrubber increases the
moisture (i.e., steam) coming out of the stack, there may be
negative visual impacts. Beneficial odor impacts may result from
the reduction in acid gas emissions. The other impacts, if any,
are expected to be minimal as a result of implementing this
control option.
5.3 BLACK LIQUOR OXIDATION UNIT CONTROL OPTION
Two control options, (1) conversion of a DCE recovery
furnace system to an NDCE recovery furnace and (2) incineration
of BLO vent gases, were evaluated for controlling gaseous organic
HAP emissions from air-sparging BLO units. The environmental and
energy impacts of the first option--converting DCE recovery
furnace systems to NDCE recovery furnaces--were presented in
Section 5.2.3. The following sections present the environmental
and energy impacts of the second BLO control option--incineration
of BLO vent gases. This BLO control option applies to model BLO
units BLO-l through BLO-3, which represent existing BLO units
associated with DCE recovery furnaces. These models represent
only existing BLO units because no new DCE recovery furnace
systems with BLO units are expected to be installed.
5.3.1 Air Pollution Impacts
This section presents the primary and secondary air impacts
for model BLO units resulting from implementing the BLO vent gas
control option.
5.3.1.1 Primary Emissions. The impact of the BLO vent gas
control option on emissions of gaseous organic HAP's is discussed
below for the model BLO units. Table 5-16 presents the operating
parameters, as well as the model emission factors, for each model
BLO
unit. Table 5-17 and Figure 5-7 present the gaseous organic
HAP annual emission estimates for model BLO units
Gaseous organic HAP
emissions are assumed to be reduced by
about 98 percent from the model BLO units with the incineration
5-28
-------
of BLO vent gases in a power boiler or other incineration device.
For models BLO-1 through BLO-3, gaseous organic HAP emissions are
estimated to be reduced by 29 to 86 Mg/yr (31 to 94 ton/yr).
Approximately 85 percent of these reductions result from the
reduction of methanol emissions.
5.3.1.2 Secondary Emissions. The secondary emissions
generated under the BLO vent gas control option were estimated
for the model BLO units. The baseline secondary emissions for
the BLO vent gas control option are equal to zero (i.e., no BLO
control at baseline). Table 5-18 presents the incremental
secondary emissions relative to baseline.
For models BLO-1 through BLO-3, the secondary PM emission
estimates range from 400 to 1,210 kg/yr (883 to 2,670 Ib/yr); the
secondary NOX emission estimates range from 776 to 2,340 kg/yr
(1,710 to 5,160 Ib/yr); and the secondary CO emission estimates
range from 1,470 to 4,440 kg/yr (3,240 to 9,780 Ib/yr).
The secondary SO2 emission estimates associated with
controlling BLO vent gas emissions range from 9,710 to
29,200 kg/yr (21,400 to 64,300 Ib/yr) for models BLO-1 through
BLO-3. As discussed in Section 5.1.1.2, the secondary SO2
emissions associated with the BLO vent gas control option include
(1) S02 emissions resulting from the generation of energy
required to collect and incinerate the BLO vent gases and (2) S00
£
emissions generated when the TRS in the BLO vent gases is
combusted. All of the TRS collected is assumed to be combusted.
Also, all of the S02 that is formed from the combustion of TRS
compounds is assumed to be emitted to the atmosphere; this
assumption is a worst-case assumption since many power boilers
are equipped with scrubbers for S02 control.
5-3.2 Energy Impacts
The energy impacts for the BLO vent gas control option were
estimated for the model BLO units. The baseline energy impacts
for the BLO vent gas control option are equal to zero (i.e., no
BLO control at baseline). Table 5-19 presents the incremental
energy impacts relative to baseline.
5-29
-------
The model energy impacts were estimated based on the
information provided by an individual mill regarding the total
horsepower requirements to operate the mill's BLO vent gas
control equipment.10 The energy requirements for the BLO control
system at this mill were scaled for the model BLO units, assuming
a direct relationship between BLO vent gas flow rate and energy
requirements. The mill has a BLO vent gas flow rate of
7.7 m3/sec (16,327 acfm).6 Based on information supplied by the
mill, 980 kilowatts (kW) (100 horsepower [hp]) are required to
operate the mill'water booster pump motor, 29 kW (3 hp) to
operate the BLO condenser condensate pump motor, and 3,900 kW
(400 hp) to operate the BLO off gas. blower motor.10 The annual
energy requirements were estimated assuming the BLO system
operates 8,424 hr/yr and range from 1,720 to 5,210 MWh/yr for
models BLO-1 through BLO-3.
5.3.3 Water Pollution Impacts
Although some condensate is collected from the BLO vent
gases, the quantity is negligible.24 Therefore, the water
impacts for the BLO control option are expected to be negligible.
5.3.4 Solid Waste Disposal Impacts
No solid waste impacts are expected to be associated with
the BLO control option.
5.3.5 Otiher Impacts
Beneficial odor impacts are expected to result from
implementing the BLO vent gas control option and are discussed
below for the model BLO units. Table 5-20 and Figure 5-8 present
the annual TRS emission estimates.
As discussed in Section 5.1.5, a mid-range value of
0.05 kg/ADMP (0.10 Ib/ADTP) was used to calculate the baseline
TRS emissions from BLO units.14 The baseline TRS emissions are
assumed to be reduced by about 98 percent with the incineration
of BLO vent gases in a power boiler or other incineration device.
On an annual basis, the TRS emissions were estimated to be
reduced by about 4.1 to 12 Mg/yr (4.6 to 14 ton/yr) for models
BLO-1 through BLO- 3.
5-30
-------
Beneficial visual impacts are expected to result from the
elimination of BLO vent gas stacks. Limited information was
available to estimate impacts such as noise and irreversible and
irretrievable commitment of resources. The impacts, if any, are
expected to be minimal as a result of implementing the BLO vent
gas control option.
5.4 SMELT DISSOLVING TANK CONTROL OPTIONS
This section discusses the environmental and energy impacts
resulting from implementing the SDT PM control options. Two PM
control options that would reduce PM emissions from SDT's have
been evaluated. The first option would reduce PM emissions from
existing SDT's to the NSPS level of 0.10 kg/Mg (0.20 Ib/ton) of
BLS. The second option would reduce PM emissiqns from existing
SDT's to a more stringent level of 0.06 kg/Mg (0.12 Ib/ton) of
BLS; scrubbers installed on new SDT's would also be required to
meet a PM level of 0.06 kg/Mg (0.12 Ib/ton) BLS.
For mills with existing SDT scrubbers, the environmental
impacts of both PM control options were estimated based on
replacing the existing scrubber with a new scrubber. These
impacts were estimated for SDT models SDT-1 through SDT-4. The
environmental impacts of installing scrubbers on new SDT's under
the second, more stringent PM control option also apply to SDT-1
through SDT-4.
For mills with existing SDT mist eliminators, the
environmental impacts of both PM control options were estimated
based on replacing the existing mist eliminator with a new
scrubber. These impacts were estimated for SDT models SDT-5
through SDT-7. The impacts of installing new mist eliminators
were not examined because mist eliminators are not assumed to be
installed on new SDT's.
5.4.1 PM Controls
Table 5-21 presents the model SDT sizes and operating
parameters for the PM control options. The environmental and
energy impacts associated with the control options are discussed
in the following sections.
5-31
-------
5.4.1.1 Air Pollution Impacts. This section presents the
primary and secondary air impacts for model SDT's resulting from
the implementation of the PM control options.
5.4.1.1.1 Primary emissions. Although emission test data
from SDT's on PM HAP performance are limited, available
information indicates that APCD's that achieve the greatest PM
removal also provide the best performance for the HAP portion of
the PM.15 Therefore, PM performance can be used as a surrogate
for PM HAP's. Because emission test data from SDT scrubbers
indicate that PM emissions are reduced with PM controls, PM HAP
emissions would also be reduced.6 As stated in Section 5.1.1.1,
PM HAP emissions from SDT's were estimated to be 0.06 percent of
PM based on a comparison of PM and PM HAP emission data for
SDT's.
Table 5-21 presents the model PM emission factors for models
SDT-1 through SDT-7. The baseline PM emission factor is
0.18 kg/Mg (0.37 Ib/ton) BLS for model SDT's with scrubbers
(models SDT-1 through SDT-4).2 The baseline PM emission factor
is 0.23 kg/Mg (0.46 Ib/ton) BLS for model SDT's with mist
eliminators (models SDT-5 through SDT-7).2 The control level PM
emission factors are 0.10 kg/Mg (0.20 Ib/ton) BLS and
0.06 kg/Mg (0.12 Ib/ton) BLS.
Table 5-22 and Figure 5-9 present the annual PM and PM HAP
emission estimates. By controlling PM emissions from a baseline
of 0.18 kg/Mg (0.37 Ib/ton) BLS to a control level of 0.10 kg/Mg
(0.20 Ib/ton) BLS, PM emissions would be reduced by 46 percent
from models SDT-1 through SDT-4. On an annual basis, PM
emissions would be reduced by 'about 12 to 53 Mg/yr (13 to
58 ton/yr); PM HAP emissions would be reduced by about 0.007 to
0.03 Mg/yr (0.008 to 0.03 ton/yr)-.
By controlling PM .emissions from 0.18 kg/Mg (0.37 Ib/ton)
BLS to a more stringent control level of 0.06 kg/Mg (0.12 Ib/ton)
BLS, PM emissions would be reduced by 68 percent from models
SDT-1 through SDT-4. On an annual basis, PM emissions would be
reduced by about 18 to 78 Mg/yr (20 to 86 ton/yr); PM HAP
5-32
-------
emissions would be reduced by about 0.01 to 0.05 Mg/yr (0.01 to
0.05 ton/yr).
By controlling PM emissions from a baseline of 0.23 kg/Mg
(0.46 Ib/ton) BLS to a control level of 0.10 kg/Mg (0.20 Ib/ton)
BLS, PM emissions would be reduced by 57 percent from models
SDT-5 through SDT-7. On an annual basis, PM emissions would be
reduced by about 19 to 56 Mg/yr (21 to 62 ton/yr); PM HAP
emissions would be reduced by about 0.01 to 0.03 Mg/yr (0.01 to
0.04 ton/yr).
By controlling PM emissions from 0.23 kg/Mg (0.46 Ib/ton)
BLS to a more stringent control level of 0.06 kg/Mg (0.12 Ib/ton)
BLS, PM emissions would be reduced by 74 percent from models
SDT-5 through SDT-7. On an annual basis, PM emissions would be
reduced by about.24 to 73 Mg/yr (27 to 81 ton/yr); PM HAP
emissions would be reduced by about 0.01 to 0.04 Mg/yr (0.02 to
0.05 ton/yr).
5.4.1.1.2 Secondary emissions. As discussed in the
following section, there is only a small increase in energy
requirements from implementing PM controls for SDT's. The
increases in energy requirements from a scrubber replacement were
considered negligible; because the secondary emissions were
estimated based on the increase in energy requirements, they were
also considered negligible and are not presented in this impact
analysis for models SDT-l through SDT-4. A different scrubber
design, rather than a higher pressure drop, is used in this
impact analysis to estimate the impacts for improved PM control
for SDT scrubbers.
Although small, the difference in pressure drop between the
existing mist eliminator and the new design scrubber replacing it
was considered large enough to justify estimating the increase in
APCD energy requirements for those model SDT's replacing existing
mist eliminators with new wet scrubbers (models SDT-5 through
SDT-7). The increases in secondary emissions were then estimated
for those SDT models based on the increase in APCD energy
requirements. Table 5-23 presents the annual secondary emission
estimates for models SDT-5 through SDT-7. As shown in the table,
5-33
-------
the increases in secondary emissions associated with replacing an
existing mist eliminator with a new scrubber are small; the
increases in secondary PM emissions are especially insignificant
compared to the reductions in primary PM emissions associated
with the SDT PM control options.
5.4.1.2 Knerov Impacts. For those SDT's replacing existing
scrubbers with new scrubbers, the resulting increases in energy
requirements are not included in this impact analysis; the
increases are considered negligible. The pressure drop for the
existing scrubber design, i.e., 12 mm Hg (6.5 in. of H2O), is
only slightly lower than the pressure drop for the new scrubber
design, i.e., 13 mm Hg (7 in. H20),6'25 A different scrubber
design, rather than a higher pressure drop, is used in this
impact analysis to estimate the impacts for improved PM control
for SDT scrubbers. Therefore, no energy impacts are presented
for models SDT-1 through SDT-4.
As discussed in the previous section, the difference in
pressure drop between the existing mist eliminator and the new
design scrubber replacing it was considered large enough to
justify estimating the incremental APCD energy requirements for
models SDT-5 through SDT-7. Based on information from mills, the
baseline pressure drop for existing mist eliminators is 1.3 mm Hg
(0.7 in. H20); the pressure drop for the new scrubber design is
estimated to be 13 mm Hg (7 in. H20).6'25 .The increase in.
pressure drop from 1.3 to 13 mm Hg (0.7 to 7 in. H2O) was used to
estimate the incremental energy requirements. Table 5-24
presents the annual energy requirements for models SDT-5 through
SDT-7. As shown in the table, the incremental energy
requirements associated with replacing an existing mist
eliminator with a new scrubber are small.
5.4.1.3 water Pollution Impacts. By increasing the amount
of PM removed from the exhaust gases, the SDT PM control options
would increase the amount of PM in the blowdown. Slowdown rates
would have to increase because of the greater amount of PM,
thereby increasing the amount of wastewater generated. Because
wastewater from SDT scrubbers and mist eliminators would
5-34
-------
typically be allowed to drain into the SDT to react with the
smelt, the additional wastewater generated under this control
option is expected to be reused. The SDT is assumed to have
sufficient capacity to absorb the additional wastewater and PM
resulting from implementing the PM control options; therefore, no
wastewater impacts are expected.
5.4.1.4 Solid Waste Disposal Impacts. No solid waste
disposal impacts are expected from implementation of the PM
control options for SDT's.
5.4.1.5 Other Impacts. Beneficial visual impacts are
expected from the reduced PM emissions coming out of the stack.
Replacing an existing mist eliminator with an SDT scrubber that
has a higher pressure drop would require a larger fan to overcome
the higher pressure drop. The larger fan would increase noise
levels. However, these incremental noise increases would be
small compared to typical background noise levels at pulp and
paper mills. Limited information is available to determine the
other impacts; however, the impacts, if any, are expected to be
minimal as a result of implementing the SDT PM control options.
5.5 LIME KILN CONTROL OPTIONS
This section discusses the environmental and energy impacts
resulting from implementing control options designed to reduce PM
emissions from lime kilns. Two PM control options were evaluated
for existing and new lime kilns. These control options are
described below.
One PM control option that has been evaluated for existing
lime kilns would reduce PM emissions to the NSPS level of
0.15 g/dscm (0.067 gr/dscf). For existing lime kilns with wet
scrubbers, the control option would involve replacing the
existing scrubber with an ESP. The actual control device (i.e.,
ESP or high-efficiency scrubber) selected by a particular mill
would actually be site-specific. The impacts for this PM control
option were estimated for model lime kilns LK-1 through LK-3,
which represent existing lime kilns controlled with wet
scrubbers.
5-35
-------
Based on PM emissions data supplied by mills, lime kilns
controlled with ESP's already achieve a PM level of 0.15 g/dscm
(0 067 gr/dscf).6'25 Therefore, the impacts for the control
option reducing PM emissions to.0.15 g/dscm (0.067 gr/dscf) were
not estimated for lime kilns controlled with ESP's (represented
by models LK-4 through LK-6).
A second PM control option that was evaluated for new and
existing lime kilns would reduce PM emissions to 0.023 g/dscm
(0.010 gr/dscf). For existing lime kilns with wet scrubbers, the
control option would involve replacing the existing scrubber with
an ESP; impacts would be estimated for model lime kilns LK-1
through LK-3. For existing lime kilns with ESP's, the control,
option would involve upgrading the existing ESP. For new lime
kilns, the control option would involve installing a new ESP
capable of achieving the 0.023 g/dscm (0.010 gr/dscf) PM level.
The impacts for upgrading or installing an ESP were estimated for
model lime kilns LK-4 through LK-6. The actual control device
selected by a particular mill would actually be site-specific.
5.5.1 PM Controls
Table 5-25 presents the model lime kiln sizes and operating
parameters used in estimating the impacts of the lime kiln PM
control options. The environmental and energy impacts associated
with the control options are presented in the following sections.
5.5.1.1 A-iy Pollution Impacts. This section presents the
primary and secondary air impacts estimated for each model lime
kiln.
5.5.1.1.1 P-rimarv emissions. Emission test data from lime
kiln ESP's on PM HAP performance are limited. As mentioned in
Section 5.4.1.1.1, PM performance can be used as a surrogate for
PM HAP's, and because emission test data from lime kiln ESP's
indicate that PM emissions are reduced with PM controls, PM HAP
emissions would also be reduced.6 As stated in Section 5.1.1.1,
PM HAP emissions from lime kilns were estimated to be 1.4 percent
of PM based on a comparison of PM and PM HAP emission data for
lime kilns.
5-36
-------
Table 5-25 presents the PM emission concentrations for the
PM control options. Table 5-26 and Figure 5-10 present the
annual PM and 'PM HAP emission estimates.
The baseline PM concentration is 0.27 g/dscm (0.12 gr/dscf)
for model lime kilns with wet scrubbers (models LK-1 through
LK-3). By controlling PM emissions from a baseline of
0.27 g/dscm (0.12 gr/dscf) to an NSPS control level of
0.15 g/dscm (0.067 gr/dscf), PM emissions would be reduced by
44 percent from models LK-1 through LK-3. On an annual basis, PM
. emissions would be reduced by about 16 to 53 Mg/yr (18 to
58 ton/yr); PM HAP emissions would be reduced by about 0.2 to
0.7 Mg/yr (0.2 to 0.8 ton/yr).
By controlling PM emissions from 0.27 g/dscm (0.12 gr/dscf)
to a more stringent control level of 0.023 g/dscm
(0.010 gr/dscf), PM emissions would be reduced by 92 percent from
models LK-1 through LK-3. On an annual basis, PM emissions would
be reduced by about 33 to 110 Mg/yr (37 to 121 ton/yr); PM HAP
emissions would be reduced by about 0.5 to 1.5 Mg/yr (0.5 to
1.7 ton/yr).
The baseline PM concentration for model lime kilns with
ESP's (models LK-4 through LK-6) is the NSPS PM level
(0.15 g/dscm [0.067 gr/dscf]). By controlling PM emissions from
0.15 g/dscm (0.067 gr/dscf) to a control level of 0.023 g/dscm
(0.010 gr/dscf), PM emissions would be reduced by 85 percent from
models LK-4 through LK-6. On an annual basis, PM emissions would
be reduced by about 17 to 57 Mg/yr (19 to 63 ton/yr); PM HAP
emissions would be reduced by about 0.2 to 0.8 Mg/yr (0.3 to
0.9 ton/yr).
5.5.1.1.2 Secondary emissions. The incremental secondary
emissions were estimated based on the difference between the
baseline and control level APCD energy requirements for lime
kilns. Table 5-27 presents the annual secondary emission
estimates associated with the PM control options.
Less energy is needed to operate ESP's than to operate
scrubbers. Therefore, if the existing scrubber is replaced with
an ESP to improve PM control, APCD energy requirements would be
5-37
-------
reduced. Because the secondary emissions were estimated based on
the APCD energy requirements, they would also be reduced. As
shown in Table 5-27, the reductions in secondary emissions are
small; the reductions in secondary PM emissions are especially
small compared to the reductions in primary PM emissions under
the lime kiln PM control options. The reductions'are smaller
under the more stringent PM control option since that option
would require a larger ESP with more energy requirements.
If the existing ESP is upgraded to reduce PM emissions to a
more stringent level, secondary emissions would be increased
because the upgraded ESP would have higher energy requirements.
As shown in Table 5-27, the increases in secondary emissions
resulting from the ESP upgrade are small; the increases in
secondary PM emissions are especially insignificant compared to
the reductions in primary PM emissions associated with this more
stringent PM control option.
5.5.1.2 Knar-ay Impacts. Table 5-28 presents the annual
APCD energy requirements associated with the lime kiln PM control
options.
Because ESP's require less energy to operate than scrubbers,
energy requirements are reduced for lime kilns that reduce PM
emissions by replacing the existing scrubber with an ESP. The
baseline energy requirements for the existing scrubber were
determined based on a baseline pressure drop of 39 mm Hg
(21 in. H20) for
lime kiln scrubbers.6 The energy requirements
for ESP's
controlling PM emissions to 0.15 g/dscm (0.067 gr/dscf)
were
determined based on an SCA of 90 m2/(m3/sec)
(460 ft2/!,000 acfm) from an ESP with long-term PM emission data
below 0.15 g/dscm (0.067 gr/dscf).6'25 The energy requirements
for ESP's controlling PM emissions to 0.023 g/dscm
(0.010 gr/dscf) were determined based on an SCA of
220 m2/(m3/sec) (1,120 ft2/!,000 acfm) from an ESP with long-term
PM emission data at or
below 0.023 g/dscm (0.010 gr/dscf)
6,25
As shown in Table 5-28, the reductions in energy
•quirements associated with replacing an existing scrubber with
small. The reductions are even smaller when baseline
an ESP are
5-38
-------
PM emissions are controlled even further, to a level of
0.023 g/dscm (0.010 gr/dscf); additional energy is required for
the ESP, thereby reducing the energy savings from replacing the
existing scrubber with an ESP.
Energy requirements are increased if existing ESP's are
upgraded to improve PM control. An increase in SCA from
90 m2/(m3/sec) (460 ft2/!,000 acfm) to 220 m2/(m3/sec)
A
(1,120 ft^/1,000 acfm) was used to estimate the increase in
energy requirements associated with an ESP upgrade. As shown in
Table 5-28, the increases in energy requirements associated with
an ESP upgrade are small.
5.5.1.3 Water Pollution Impacts. If a wet scrubber is
replaced with an ESP to improve PM control, the wastewater stream
from the scrubber would be eliminated, thereby affecting the
water balance at the mill. Whether there would be any
significant reduction in wastewater disposal for this option
would depend on whether or not the scrubber discharge was
previously recycled and reused. Wastewater impacts were
estimated for model lime kilns assuming there was no prior
recycle or reuse.
The wastewater impacts were estimated based on a factor of
2,250 kilograms of wastewater per oven-dried megagram of pulp (kg
wastewater/ODMP) (4,500 pounds of wastewater per oven-dried ton
of pulp [Ib wastewater/ODTP]) for lime kiln scrubber blowdown.21
This factor was converted to annual wastewater impacts using
conversion factors of 1.0 kg H20/L (8.345 Ib H20/gal) and
0.9 ODMP/ADMP (0.9 ODTP/ADTP) and multiplying by the product of
the model ADMP/d (ADTP/d) and 351 operating d/yr. Table 5-29
presents the annual wastewater impacts for model lime kilns LK-1
through LK-3. With the replacement of existing wet scrubbers
with ESP's, the wastewater discharge would be reduced; the
reductions in wastewater impacts were estimated to range from
-226 to -709 million L/yr (-60 to -187 million gal/yr).
Because ESP's operate on a dry basis, no water pollution
impacts are associated with lime kiln PM control if the control
involves an ESP upgrade.
5-39
-------
5.5.1.4 Solid Waste Disposal Impacts. Dry PM catch would
be generated if the existing scrubber is replaced with an ESP to
improve PM control. Also, a larger PM catch would be generated
if the existing ESP is upgraded to improve PM control. As stated
in Chapter 2, existing lime kilns with ESP's return the PM catch
directly to the lime kiln. The lime kiln is expected to have
sufficient capacity to absorb the PM catch resulting from
implementing the lime kiln PM control options.
5.5.1.5 nt-.her Impacts. Beneficial visual impacts are
expected as a result of the reduced PM emissions and reduced
moisture (i.e., steam) coming out of the stack. The moisture
content in the stack is lower with an ESP as the control device
than with a scrubber. If the existing scrubber is replaced with
an ESP to improve PM control, the noise from the larger fans used
to overcome the higher pressure drop would be eliminated, thereby
reducing noise impacts. However, the reduction in noise is not
expected to be noticeable due to the high background noise levels
typically associated with pulp and paper mills. Limited
information is available to determine the other impacts, but the
impacts, if any, are expected to be minimal as a result of
implementing the lime kiln PM control options.
5-40
-------
X \ \ \ \ \
u.
oc
O O
O O
T- O
O
O
O)
O
O
00
O
O
o
CD
o
o
LO
CO
0)
u
o
o
•SJ-
O
CO
O
C\J
o
o
(U
o
(U
o
0)
•d
o
g
M
H-i
m
c
o
-H
ra
m
-H
e
0)
4J
4J
Q)
4J
U
-H
OJ
^
(ti
IT)
-------
CO
Q)
O
5-)
0)
O
0)
u
Q
r-H
QJ
T)
i
8
M
M-l
m
fl
o
•H
m
CD
-H
I
0)
(U
4J
o
CM
OQ
i
in
0)
Cn
-H
Cn
IO
5-42
-------
o|UB6jo snoaseo
03
0)
U
4-J
S-l
1
o
0)
Q)
•O
i
s
o
M
M-i
m
C
o
-H
m
m
-H
e
0)
o
-H
tn
CD
d
O
Q)
m
rt
O
in
0)
tn
-H
5-43
-------
to
CD
O
C
13
o
o
CD
HI
O
Q
"CD
to
CD
CD
o
o
s
HI
o
Q
CD
•a
o
'spunodujoo jnjins paonpaj
01
g
Q)
JJ
CO
>i
CQ
0)
O
rtJ
B
(U
o
0)
I
CQ
CQ
m
-H
g
cu
CQ
TJ
0)
O
Q)
H
r-f
nS
4J
O
•*
i
in
Q)
tn
•rH
to
5-44
-------
\ \ \ \
\ \ \ \ \ \ \ \
m
(U
u
rt
§
o
(D
(1)
g
O
o
-H
m
CD
-H
H
-------
\x\\
d)
o
rt
fi
C/)
CD
O
03
§2
o
o
CD
LU
O
Q
"CD
T3
o
CO
CD
O
CO
C
CD
O
O
LJJ
O
Q
CD
TD.
O
o
QJ
rH
(U
-d
o
m
«
o
-H
m
m
-H
g
0)
Q)
13
•H
X
O
-H
H
3
CO
in
tn
-H
fa
5-46
-------
\x\\\\\\\\
X \
\ \ \ \ \ \
o
o
o
en
o
CO
o
CO
•7-
o
in
o
CO
~1 r~
o o
C\J T-
jA/uo; 'dVH OjUBBjo
IT)
tn
-H
5-47
-------
\ X X \
CO
u
-H
CM
0)
•d
o
S
e
o
M
CO
a
o
-H
m
m
•H
S
QJ
CQ
-d
d)
o
(U
cti
4-)
O
EH
co
i
in
0)
Cn
•H
jA/uo} 'spunodujoo jniins paonpaj
5-48
-------
CQ
^
rt
(6
OJ
tn
C
-H
O
CQ
m
-H
r-l
§
m
Q)
TJ
i
e
o
CQ
C
O
-H
CQ
m
-H
g
0)
0)
JJ
0)
4-)
rt
o
-H
AJ
M
rt
04
in
-------
jA/uo} '
5-50
-------
rrj
CO
fe
o
u
w
u
Q
§
H
etj
E-i
rd
H
I
in
1
1
I
E
n
ffi
content, %
Moisture
0
8
m
1
1
s
o
O
1
"m
I
1
CL
I
'c
a
I
5
cr
1,
5
5
Jj
||
0- C
M
IS
Baseline
S "3
P
jj a
a °
°- «S
S3 m
11
u o
a I
0 S"
lion rate
ADMBP/d
produc
ADMUP/d
ttnng rate,
kg BLS/d
oi 9
i!
recovery
(umaces
gjisiiia'ssssjg
ssssssssasss
888888888888
. § - § . § . I . I . i
SSSSS8SSSS8S
WCMCMCUCJ
-------
5-52
-------
O
H
in
Wastewater
flow rate,
L/min
ESP SCA, m2/(m3/sec)
Moisture content, %
Temperature, degrees C
i
3
a
Gas flow r
PM controls-
0.034 g/dscm
PM controls-
0.10 g/dscm
Baseline
J|
11
Quench
outlet
§5-*
UJ 0
1 1
§515
Ul 0
III
*~ 'o m1
Equivalent pulp
bon rate
ADMBP/d
produc
ADMUP/d
f If
5 c"33
s « •*
u.
Model
recovery
furnaces
If) T- in T- T *- *Tt- O1Q O)Q
TJ- »- •q-*- r*. CM r^ CM oJ
&
•o
1
1
1
1
1
8
£
1
-------
8S SS S3 ?S SS SS
ni n n n ui in in m o> o> o> 01
O
c3
in
2 5
8
I
§Tf
88 88
88 88
5 S5 SS SS
(2 tO tO CO
SS
jj O CO CO
UJ O
O O) O) O)
8.8. 8.8.
S3
CO CM
fC i
s
ffi
CO
H
IS IS
01
I
in
§
8 8
o
8.
o
8|
! !
o o
8|
5-54
-------
w
Q
g
U
H
LO
w
S
O
•o ^3 w
TO ^^
O3
IS
2 o
c J-
o
o
tn
CD CM
S O
'5 en
CD ~ra
C ^£
"CD 7^
8 i
m
Isis
S I -i s
^ "rri
o m E5-*
f\ C/D
QJ (/) ^r i
~ 0 X CO
CD CO C -"-^
m o> « ro
0
« I
I 1
2 o St
°- e S o
§ I 0--0
— -Q | _
o ob .52 £
~ o w
f*t "*~* ^^
a § "S
0 o
S5
jlf
*ii
CD
H
U.
_ §T g
CD > Ss
-0 0 2
0 0 £
«£ l_ *+
3
LU
00
co
to
CD
LU
CM
CD
CO
cp
LLJ
CM
9
LU
0
CM
10
cp
LLJ
£
CO
to
0
LJJ
co
CO
-------
(D
|rf
O 0)
"S I
-9 a.
•O CO
JD UJ
8 &
as
of :§
"o S
c- V
o <"
.2 o
:-i± co
U)
E »
W .2 m
E g§
2 &"o
« a. *"-
w co "5
n UJ C
ffl &-S
UJ 13 Q.
Ill"
ii2.
s gf §
Sal-i
Ib BLS/AD
tr
co"
O)
c
I
(A
s
«
0)
3
o
.0
ja
D
O
in
-a
(D
J3
MODEL
ENGLISH
ABLE 5-3b
c c o o
S t:o
• 5 r
so:£0'->--s«
£ « i" £
(D
_C
"5
3
m
s"
D-
0)
0
c
o
O
w -—
Q. ^
g$^3
g.Jjm
o c ^a
CJl O =r
0>J?
5-
O
D)
O
Ul
r-
co
1^ 5-
u in o
Sg
! "-
Q. o"
(D u_
C 0
~» _r en
•*
TT
o
d
CM
a) S -01 'I
w n ^i o
UJ
£
CO
O
o"
•
-------
O
H
LO
CO
CD
CO
CO
9 OQ
_
If
s
o
O
8
'5 D)
C3 .*
CD 01
"CD
CO
1 *
~ O
"5 CD
*- CO
*C *8
S en
O
CD W
1 °
W (0
CO (0
CD cn
2*
•5 E
> o
CO W
— "D
C
0
o
c
"CD
m
o
1
i
"CD
. |
'
c =£.
co cn
cn je
k.
o
_«"
X 5
•— V
c =i
CO CD
01-*
o
LI
o w
" TO
c ~£n
° s
2 o
ix o
A £
9 w
•S 5.
o cn
0 o
2 i:
CL 0
E
-= °
2 w
CD
a_ E
co a3
4 j •4-'
•** V)
CD
§ a.
D -^
j.
3 |
U
ss
lit LU
in co
in CD
m in
9 9
LU LU
S c\i
co' CD'
8
LU
s
cp
LU
^3
T-'
in
9
jij
CO
CO
LU
in
iri
i
o'
T_
O
CO
o
to £«
? ~o
LU
0
Q
S|
Q ^
c
8
CO
tv.
UL
cc
ss
LU LU
in co
in co
in in
9 9
LU LU
S cvi
CD' CD'
CO
cp
LU
s.
cp
LU
0
OJ
in
cp
LU
CO
co'
o
LU
in
in
CO
o
o
T-
o'
T-
o
«) cT"
5 -o
LU
O
Q
LU -0
O 3
Q <5
g
o
o
JD
LL.
o:
S3
LU LU
in co
in co
in in
9 9
LU LU
S CNJ
CD' co'
s
LU
s
T~
s
LU
^3
•^
in
cp
LU
h-
CO
co'
S
LU
in
in
i
c5
^_
o'
CO
o
"53 £«
S "o
LU
O
Q
0 "§
Q ffl
c
0
o
(0
co
uL
o:
ss
LU LU
in co
in co
in in
9 9
LU LU
S w
CD' co'
CO
cp
LU
8
T"
cp
LU
O
CVJ
in
0
LU
CO
co'
cp
LU
in
iri
S
o
o
1—
0
T-
d
"S £•
LU
O
Q
LU "O
o .2
Q S
o
CJ
op
LL.
oc
ss
LU LU
in co
in co
ss
LU LU
u5 c\i
CD' co°
9
LU
8
"^
S
LU
^'
in
cp
UJ
CD
co'
§
iii
in
iri
§
o
t
o
CO
o
03 £<
5^3
LU
0
Q
Ul "g
Q Iii
c
o
o
CO
2
DC
3S
LU L
in oo
in co
in in
9 9
LU L
co' cc
9
LU
8
T~
3
LU
in
cp
LU
CD
co'
cp
LU
in
T~
iri
I
0
^
o
^-
o
o> £•
5 "a
UJ
O
Q
LU "O
O S
Q a
g
o
o
JD
op
LL.
tr
CO
CD
O
|
,O
in
J5
S
o
J3
S
CD
3
(3
I
CD
in
CD
_2
en
LU
•a
CD
co
o
CD
.C
2
T3
CD
•e
CD
I
U
CD
I
U>
!c
_c
.i
I
g
a)
2
"c?
5-57
-------
co
H
I
i
LO
to
),_
i «
CD m
i «=-
!
U
S jo
fct
CO
CO
CO
1 s
8 g
0 S
ij 01
0
p co
is O
CD O
id co
CO O
2
Q.
CO Tj
> c/>
ca -g
1 §
c
o
o
1
cc
cr
re
CC
a
,s
H
"a
fr
^:
CM
O
CO
0
CM
O
CO
0
^ «
if
0
w" ,_
b- ,
o m
2 o
Q- C
A P
o -S
§*" &
T
2 o
Q. 0
"1
2™ u
!O "i
££
0
At
* **~ ^
CD
CO
UL
. CD
3 > S
: CD _
SS
LU LU
in co'
in in
9 9
LU LU
8 CM
CO CO
8
LU
s.
1—
o
LU
O
CM
T—
m
o
LU
f-
co
CO'
o
LU
in
iri
in
o
o
o
o
8
o
03 £"
S T)
LU
0
Q
LU *O
O S
Q a5
c
o
o
ca
DC
S3
LU LU
in co
in co
in in
9 9
LU LU
8 CM
CO CO
CO
S
8.
o
LU
"*""
in
o
LU
f-
co
co'
S
LU
in
in
in
o
0
.
o
o
5
d
% T3
LU
0
Q
LU "a
O £
C
o
u
A
^
DC
33
LU LU
m co
in co
in in
9 9
LU LU
CO CO
LU
8.
o
LU
"""
LU
CO
co'
o
LU
in
in
in
o
o
§
o
co
o
o
"5 &•
S -a
LU
o
Q
LU "0
OQ3
C
c
o
o
ca
CO
u
DC
??
LU LU
in co
in co
in in
9 9
LU LU
SS
CO CO
1.03E-03
S
LU
s
in
0
LU
co'
n-
o
LU
in
iri
in
o
o
^.
S
o
5
d
**"* ^s
5 T3
LU
O
Q
LU ~°
OQ3
•f-'
Q »
c
ja
CO
u.
DC
S3
LU LU
in co
in co
in in
9 9
LU LU
S c\S
CO CO
9
LU
8
T™
s
w
in
0
LU
co'
o
LU
in
iri
in
o
o
^4-
s
o
CO
o
o
flj C*
o
Q
Q m
c
o
ca
0)
u.
DC
ss
LU LU
m co
in co
in in
9 9
LU LU
S S
CO CO
9
LU
8
T-
S
LU
O
CJ
in
CD
LU
&
CO
o
LU
in
iri
in
o
0
^
S
o
5
o
o
tu &
o
a
o .
Q
a>
U-
DC
•a
CD
JO
•a
co
1
a.
ca
"5
o
jt;
CD
c
^
2
"o
CD
a.
CD"
o
'S3
Q
<
<$
ca
.g
o
o
tj-
co"
01
c
E
3
CO
£
V)
J3 a>
•5 E | |
3 £ — §
O §.^T3
' co •55 ^ g >.
_ ,±1 ^- Cu w
£
§
W -3
(D £.
•2 8S
CD
.a
XI
o
ca
TO
(D
JQ
i
u
: CO
; a-
ca
ggco^oocrg
^^g-o-Scogu]
D- S- ca •£ -^ •£ OQ
Q) Qj
1_ i-
C. ^- C 5 £ O £ ^J *Z o?
X n»^- wiri+f.,. n fl)*^
O
(U 0)
•§•§
D) O3 o
§•§•£
Q_ Q. LU
CO CO O
LU LU Q
ca co £
CD co „
(0 CO O
81-gr *•
IP
X. co
D *" r"
•S « E
CD O
£ 1
C
CO
CO
E.I
LU = 2 o
Q CO CO 2
LU
£.§.*
LJ "n *=
-5-0 « Iz
o .S
oto
"o 13 CD
to to oi
"o ^ £
O) O) >
«• in to
^- T- CD
o q .n
d d Z.
00°
o-n
£-D^
* "3 —
'
.S f c
OT Q) 3
I S£
S s r
CD - O
O) CD ^
£ ^, CD
? £ 2
*- o
E •-
o £
S -2
ca c
"o o
o !2
'« 'E
.52 CD
E cj
co o
O OT
¥ ID
?2
CD ^
CO ""
a> m
II
in c
•g 2
£ CD
CO O
CO "^
ag
0) 05
C C. -i **i (D K^ iu
111.» 11 -5
E E co o CD .« t3
-------
Emission
reduction,
%
Emission reduction, Mg/yr
missions, Mg/yr
D
"5
>
_aj
5
5
SCL
a. <
c
V)
^2:
0. <
a.
"o c
•fc .2
§I
0 °
? «
22o.
«n.<
5
"w
w
)
)
i a-
i
i
Model
recovery
furnaces
co co to co ao to co co to
toco to co co to toco to
into CN co r- co CM CD *t
do d d r^ d ^ t-^ o
r-» co tn co ••* co coco in
cs <- ~ oto in a>T- CM
CM co ro rf un r- in co CM
^ o o "2 ^ °i ^ ^ ^
d Q o' do d do d
5 in w S- - 5" "
P* * S« « co- £
2 o u ~ o o s a o
c E £ c E E c £ E
f=(J (J C0 o C0 L
"« M hf"' "> ;: <» w
^•a -a 43"° "° -S"0 "°
•g,a a --o a ^cs o
i-i"3" •* «-> •* "* ri'* •*
°n n °n n °n n
^O O ^O o *^o o
9o CD 9 CD o 9o o'
^2w in —In in —in in
So o So o So"o
§c c Sc c §c c
Ho o Ro o So o
0 0 0 °(J 0 °0 0
5^ -s 5^ ^ S^ 5
ff «£ 4^ n ^i ^. n *& ^i
^-Q. £L ^-OL Q- °-CL CL
r^ n co in en . r»
O O r-' O r^ O
CO r- ITS tO -^f t-
m co •* co ro ^
CO T- tO cs| CO CO
a ^i a -3 a -O
•«• jf in in co to
1J ra J3 a xi ra J3
J i— r- CM CN CO CO
u. u_ li. u_ uL li.
c cr or cc cc cr
tn^- co ini-co inr- co
•j^ CN ^i^ f
do d do d do d
in1"" oj ^^ r^ ^ O <°
S- . 2 °^ ES ^S 12
^S 0 "0 0 "I" «
0 d d °d d °° °
cor-. r» Ilin in 5^^ r-
r«N CM J2rt •* J3co co
2" 02° °S" «
§1 1 § § I § 1 1
1-8 -S |-S -S |-S -S
"o05 a "S0 o "£,= =>
r-)'* •* (-.•* •* <-,•* <
°CO CO °CO CO °CO CO
^o o ^o o *^o o
9d o 9d 9' 9d o
i? in in 21 in in £ 'in in
So 0 So "o S"o "o
*— -*— ' +-I .— 4— '
o u
1 S.
"3 «
£2 «
S -i.
4) *2 4-.
•o c c
3 CO o 3 3
•S tl a- Q-
*^ o co to
_C •»• UJ UU
'« -9 c c
4^ in co co
jj in u t>
CO CL) CO C3
IS II
SsII
•c ^ o u
5-59
-------
5-60
-------
ti
eo
W
^
i*
OS
H
I
W
CtJ
J
fx]
P
i
cti
O
fe
J
O
(*
O
u
§
&
o
fe
§
o
H
CO
CO
H
S
w
o
u
w
CO
l_
emissions, kc
Incremental
O
§
S"
_o
Baseline em
£>
« §
j O
c
c
c\
«
o_
c
c
8
i
CM
O
C/3
5
furnaces
6i i IB § el !
in r-. CM co i- co ••- ° o
T— in co cMcn co co00- c
O O O O ft
fe° S R!-* N- co o i?
JS co co co co co 10 S •«-
co - en co - ^- en -
••- CM T- OJ CO CM
cofc r: coco o in-* en
S;cq en cor- 3 cncn en
^~ CM -i- -r-tj- CO 1- CO •*
IT E E IT E E "C"E E
ll i ll I ll I
~Bi ro ro ~6> ro ro "Bj01 ro
.— * it" '^f* — •, ^rf* 5j" __ ^" *Q
°CO CO °CO CO °CO CO
l^o o ^.o o ^ o o
0 CD 0 9 0 CD 9 CD 0
•01 u> co J2co co -£! o Sea o
|^^S | ^ ^ |-S -?
"ra"^ "°* "05 ^ ~3) "3> "Si "B>
°CO CO °CO CO °CO CO
.0 o .0 o .0 o
9 o o 9 9 9 9 o CD
-2ci) to J2tii w J2J) j
Eo o 2o o 2o"c
1 1 1 ll 1 11 1
°0 0 °00 °0 O
°-D- 0. °- O. D- 0-CL. Q.
*^* ' ^3 CD ^D C
g p; r- in CM Z-
'*" ^ T^" T-" CM" cv
CM CO -r- r- CM •§
r- in CM co i-i S
CO rf CO r- ^j- -
_ o o o o o
2 co co CM -^- N.
ST •»— in en co ^j-
T-" -r-" W CM" co"
Sco o in co CM
co CM en r— T—
i- CM co co in r-
j r— r— co co en 05
J LL. LlL LL. LL. LL LL.
Q DC DC cr cc or DC
CO
"o
'o
o
c§
in
.2
CO
CO
JO
CO
C
CO
to
co
.C
.22
"5)
LLJ
•o
co
CO
XI
_o
a
u
CD
.£
O
I
I
O
u
2
CO
O
U)
t>
CO
. ^1
c -n
CD CD
E -g
CD o
" Q.
it
CD CO
T3 "D
11
Q. Q.
« -I -i
C « 0)
_eo co co
to co CD
•^ 2 9
•C <0 03
Iff
5-61
-------
w
w
H
J
g
w
VD
in
W
J
I
£
5
of
.i
OJ
.2
"c
m
o
C
~5
8
1
t
c
•s
CC
• CO
1
a
s
CO
_ 1
-S I
O c.
*5 Q
3
|
CM
O
CO
Q_
C
o
"c
o
r.
CN
C/
I
•»
i
)
)
g| 1 II i II i
"> CM" ••-" T-' co" CM" "- in •*
o 2 o o co S o g w
s 5 a S3 " SS 5
o£ § gg § S| |
00 of cvT i-7 io~ co" oir^ w
8s s §§ s |l i
•t— in *4- co ^j- i*- ^ T-^ •»-
-8-8 ^ -05. 5. 5. 3>. 5
CD CD CD CD CD CD CDCD C
•q- in in •<*• in in 3|J2 S
oo o oo o oo q
o o o o o' o o o cj)
0)0) tO tOtO 0) tijJO A
~o ~5 o "oo o go o
1! 1 11 1 11 1
22 2 22 2 22 2
570- O- Q-O. O- Q-D- O-
0 00 § CM C
CU CO S 00 CO 0
co" co" m" CD" co" en
o oo 22 S
g E S S 8. S
^" c\T co" co" •* to
2 g g ? 1 I
C\J 0 CD -r-. ~- _
Tf in" t-- en ^- T-
00 00 0
SI S fn Eo CM °
03 ^T ^-" T-" CM" CV
^ ^ in *2- 4 '•
O T- V °V ^ °? ^
Q LL. U. LI. LL U. a
Z DC EC DC DC CC 1
c-l o O O O O O
g| $° gs s Ste 5
CO - to ^j- ^ ^- ^j- j»5
o o _ o o
?S S co^3~ ^j- "'CM i-
0°, S 0§ g gg g
coo in Oco in •^•i— co
•* CM" -r-" m co" CM --co •*
8 CO CO CD^f CO Oio £2
S- T- cocn CM 2w ffi
^•TJ- co T- CD in co^- en
£.£, ^i. £.-H- -H. £.-H- -2-
D)O) D) DJD) CD OJCD C
^•in in ^m in ^;in in
oo o oo o oq q
11 1 11 1 ll 1
11 1 !§ | || |
iEi? s? C?D- Q- o-Q- Q-
' § II 1 § §
^_- T-" CM" co" •* to
° g g g • i
CM 0 CO CO •* 0
03 ^J" T-" -r-^ CM CO
o oo 22 S
& 83 S 8 S
CM" CM" co" Tj- CD r>-
^. Tt CD CM °. r
^! SI S Eo Jj- ^_-
03 J3 ffl ^3 • ffl ^
m ^ ^ " £ 2 i
a cc. cc D: cc DC cc
E
*
2 ir
« en *-
p* ^N £
o o oo
IIS
£ «= to
o 2 ,to
S $ "
11! ^ 1
*03 m C "O
s "E P i
i
0> Q.
ffl 03
« g. 3
S^TBL
CO CO (D
•g 03 CD
£-2 £
OS ffl 03
T3 I- a
D) CD
Q_ n
D 3
"r- C. 0- Q_
§ | | ? ffl
— ~c -S .£ .£
I § = 2 1
"5 °- . p c
•g £ 2 -B ~
2 ^^ ^^
U O)
05 m 8 i
0
2 o to
•0*5
O J2 «
'«§•!•
in E E
5-62
-------
TABLE 5-7. ENERGY IMPACTS FOR PM CONTROL FOR MODEL RECOVERY
FURNACESa
Model
recovery
furnaces
NDCE
RF-1a/4a
RF-1b/4b
RF-2a/5a
RF-2b/5b
RF-3a/6a
RF-3b/6b
DCE
RF-7a
RF-7b
RF-8a
RF-66
RF-9a
RF-9b
Baseline
energy impacts,
MWh/yr
1,700
2,030
3,070
3,650
4,430
5,270
829
. 1,020
1,380
1,700
2,490
3,070
Control
option
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.015 gr/dsd) (c)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.01 5 gr/dscf) (c)
PM controls-0.034 g/dscm (0.01 5 gr/dscf) (c)
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.10 g/dscm (0.044 gr/dscf) (b)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
PM controls-0.034 g/dscm (0.015 gr/dscf) (c)
Control level
energy impacts,
MWh/yr
2,030
2,850
2,850
3,650
5,120
5,120
5,270
7,410
7,410
1,020
1,640
1 1,640
1,700
2,730
2,730
3,070
4,920
4,920
Incremental
energy impacts,
MWh/yr (c)
330
1,150
820
580
2,050
1,470
840
2,980
2,140
191
811
620
320
1,350
1,030
580
2,430
1,850
(a) Numbers in parentheses represent negative values, indicating that energy impacts are reduced by that amount
PM control energy impacts at baseline and at 0.044 gr/dscf = [0.00018 x model gas flowrate x 1 in. H2O pressure
drop x 8,424 hr/yr x 1 MWh/1,000 kWh] + [0.00194 x model gas flow rate x SCA x 8,424 hr/yr x 1 MWh/1,000 kWh].
PM control energy impacts at 0.015 gr/dscf = [0.00018 x model gas flow rate x 4 in. H2O pressure drop x 8,424 hr/yr
x 1 MWh/1,000 kWh] + [0.00194 x model gas flow rate x SCA x 8,424 hr/yr x 1 MWh/1,000 kWh] + [0.746 kW/hp x
Equid flow rate x 60 ft head x spec. grav. H2O x 1/70% pump motor effici. x 8,424 hr/yr x 1 MWh/1,000 kWh].
(b) Impacts were estimated for ESP upgrade/replacement
(c) Impacts were estimated for ESP upgrade/replacement plus packed-bed scrubber.
5-63
-------
5-64
-------
ffi
CO
H
J
§
oo
ui
§ §
E -a °
UJ CD
— ° !r
Ill
UJ CD
Control level
emissions,
ton/yr
o c
4= .9
0 Q.
o o
CD «
~ O >x
CD 'co "c
% -E 42
m E
CD
>% co
>— CD
~O O ^
o o E
^ CD 3
W CM CM
Is. |s. |s.
LO -^f -st-
CM •* CD
f^ is. m
O) -V- CM
ccc
o o o
co co co
CD CD CD
C C C
O O O
O O O
E E E
CD CD CD
to "co to
to co co
Q- D_ CL
CO CO CO
UJ UJ UJ
T3 -D TD
O O O
4-1 4-< 4-<
"CD " "CD "CD
"3- CM CD
CO CO CO
UJ
O <* 10 co
^ t i i
Q U. U- LL.
z rr cc cc
CO CO CO
CO CO CO
^ S^ c\
cq rs. ^
LO CD . T-
c c c
g p o
"to "co "co
CD CD CD
C C C
O O O
o' o o
o o o
T3 TD T3
O O O
i i i
333
_ CD ^
« .2. $
u N. cp en
O LL. Li 1[
D CC OC CC
CD CD"
2 §
» c
CO
u.
CC
JC
O)
2 co o
-^ -°
CO co" c
—I "D CD
m >, 3
CD CO j3
TD T3 __-
O TO O
E "CD c:
^.sf
CO CD .
-J "O CD
CQ ^ C
=^. J= "CD
" x -a
•s -y
LL.
cc
_co
CD
T5
O
o u_
•*- oc
If
^ o
CD "5
"to
CO
CD
CO >, CO
o "CD CD
W E CD
CD-^
£ K o
c _
'
CD 3
Q.T3
,n O
|8§? =
co ^1 *-• co o
CO ^ CD CD -75
'E 42 ^ •!= .t2
a. x c £ Q
< :: CB p CD
O T3 CD O •*-
"c -^ E c °
CO LO . 2 -p
05 CO _CD £ §
O X
T3 CD
CD
CD
Q.
%* E
CD £ o
"cF
^) T3
X CO
5-65
-------
Cd
O
u
H
rti
in
c c
.2 o
to " ^
'r"j ,**~
LU CD
c cf
0 0 i-
"to "f? •<:
to CJ 0j
E -a S
LU CD
"55 -
> to
CD C v_
— ° <^
g "to D)
o "
"o c
i= o
BH
0 °
CD 2
r- i— 1
•— O >»
co 'to "oi
to to ^
co •;= .>
m E
CD
>% to
_ a? g
CD > fe
•oo?
O o b
22-
.
in m m
in m in
o •* cq
co in t^'
in m in
CM' -V CD
c c c
o o .g
"to "to 'to
CD a CD
> > >
000
000
E E E
CD CD JD
"to "w co
to to co
D_ Q- 0.
(D V) C7D
LU LU LU
>>>>>>
T> T) "O
2 £ 2
"CD ^ CD
in a> ^j-
in cri •r-
LU
Of LO CD
Q LL U_ U_
z: a: a: cc
a> o 05
CM Is- ^
CM CO CD
in m in
T-* CM -t"
C C C
g _g .g
"co 'to "co
CD CD CD
C C C
O O O
000
i — i — i—
O O O
T5 TJ T3
99 9
555
555
^ 0 -r-
CM •* r^-
LLI h- CO O)
0 LL LL. LL.
Q CC CC CC
C
CD
"5)
en
CD
_3
CQ
+->
"c
13
LU
-a
CD
_o
CO
o
CD
JT
-n co
CD B
1± O
CD C
CO .0
q>
in
CD
CD
_CD OC
CO .
> .a
CD
CD
CO
JD
CO
5-66
-------
CO
u
Q
Q
W
K
CO
H
en
i
LO
c c
o o
8 og5
E-S
LU CD
§§£
11 f
LU 03 O
Control level
emissions,
ton/yr (b)
O C
C '•P
0 0-
0 °
— o ~
p 'w >
> w
•7; S? °
•o o 2
out:
^ o 3
in in in
in in LO
en en CD
co in' ccS
r>> q T-
CN in K
c c c
o o o
WWW
03 03 03
C C C
O O O
o o o
E E E
03 CD 03
WWW
WWW
n_ Q_ D_
CO CO CO
LU LU LU
•o TJ -a
o o o
•f-1 +-*•*-•
•*-'•*-' 4— '
03 03 03
O T- CD
to *- <-
LU
O sj- in CD
Q LL. LL LL
Z DC CC CC
> >
C C C
000
o o o
I— l~ 1_
O O 0
•o -a -o
o o o
i I I
555
o o o
_! _J _I
CD sj- en
LU r^ CD en
J LL LL LL.
Q DC CC CC
c
o
co
+-•
|
CN
CN
C
O
. CO
w —
E E
« a.
w Q-
>-CN
w VN
!!"•§
tl ^Q r*
O
w "^"
~"5 2
— "o co
a o -a
O C
c c
k. U. t—'
c o. ^
*— •"— ^—
r Is
C C3 03
W
W
^S^i
i§5^
1 -a ^
' £ •= E
E
O 03
U. -4-J
t- W
o
co ro
a.
—. "O CO
O 03
CO
w -a
W 03
c
o co w
-o w "
03 03 -Q
W O
co ro
-Q E en
.^auL
CD - ^
JC g O
Imr^
£ O ,,'
"- >- -^
cc o °2
s- ^
4: ro "
nf ^ E >.
(— O L_ Q. ^-
«L.L ^^
W Q O T- ra
13 < ^ s- JZ
•o 2: ^ o .t;
W CD
c
6 " = I o
™ £ o 2 a.
£ O — c --
5 O i OJ '3
•e- o cr
^•^-co c °
g.E CD
0 E .E
_
w
'
i= <— C
-"I
ro c o
•° o c
P -t: n
03
T>
O
E
*-• F +3
ro ro
"o i^ "o
c "- c
.2 co .2
o c
'
3 C CO ro
o .a. ££
^1?^
O Q5 LL O
"*- *- DC >
a. c o
'""•—"§)
~ " 13 i-
o CO < .52 o „
^|^£§£^
E o E 2 w sJ-z
Or- O 'T F n _
5-67
-------
o
P
§
H
CO
OS
8
, ^
I1""
en v
C J?
0 O
en 2
_«
CO o
CD t*- T-
^5 ^^, *
S S S
g g §
•« "en w
0) 0) 0
1 O O O
L 0 0 0
1 1 1
ill
•
o o o
D en 3- in
^ 0 CO 0_
CM" CM M-
>< S § S
D ^- in t-
Z -rT T- CM
^i O O O
^ •*• •*• Sf
2 ^- m- S
JJ CM CO W
2t-- to c
co r** T-
o
ca t^ co a
1 £ fe a
.8 CMO coS .8
SS SS S^I- CM" cn^j- ?^I-
o _,
E^i s8 §s ?£ 8? §§
11 8-g 8| S| II 8|
*
in o -r- CM co^- CMCM gco
Sg -$$2 J2eS ^CM CMS T-rt
^^ ^=5- ^.=5- •s-'S- •5-£o -s-gj
jj }} If |j If If
1 1 1 1 5 1 5 1 ? § ? I
14 li li If 11 II
II 11 11 |1 |1 ||
n ^ n «2s o ^ ol J? n p d. o
a. a. a- .
8 P S 1 S 1
r~- co ^ ^r CM" CM"
§ 1- s S 2- 5-
o o q i o o
en CD to CM rr C:
co •?- in en « •*
en ^j- ^J" T- CM co
g « s s s ^
? c3 co co m t^-
(0 XI CO J3 CO XI
= fe fe fe fe c =
D 3 3
5 O Q. a.
E CQ {S S
.§ £ c c
"S *o 52 S2
"C 0) 0)
1111
S 1 5 §
Illl
XI en — ,_
S 5 £ £
x: co CD tu
c * to w
<2 1 1 1
c 8 a? a)
"S CD CD CU
.S CD CD CD
S" co w en
0000
'•§§.§.§.
2 I £ -i
"~* — *~** ^S"
*—* "^^ "^"^ * '
5-68
-------
CO
H
J
g
X)
o
H
i
in
•$
w
.2
"&
E
CD
CO
•£
CD
£
O
_c
1
0
O
f
?"
.0
seline err
CO
CD
£.
£ k_
c
(g
-^
c\
U
Q.
C
Q.
O
O
C)
X
O
CM
O
CO
2
8!
lurnac
O CO C\
CO CM 0
^ SB- S
o o o
S? g £
C\f CO, ^
*— •* *•— * r?
£ 5 8
co" co" ^
**•"*
in T- c;
CJ r-_ rf
LQ f} J^
C C C
o o o
03 W V
0) 03 Q)
O 0 O
o o o
•0 -D -0
III
O O O
O CO W
CD CM en
M- CO 00
o o o
5 S g
CM CO •**•
_. zz o
£ CO M
co co j;
o o o
10 T- CO
CM r- tf
i-- T-- CM"
N- op en
i f 1 1 1 1
cc cc cc
— O O OO _ O OO OO
Sm POT T-r- gco cnio o cc
CvCD P** CM CMP*. O-r- T- CD T- C
T-" " T-~ T-" CM" ^ CM" CM" •*" T-" cc
en en o o o § o °- g ° o g
COCO OCO ^ T± CM T- -r-CO CO o
COCO CMCO CD- CO- - - r in -
.-,0 o oo _ o oo oo
po o CM T- t>- O|>^ T-r-. cocv
g*CM OTf^ COCO T-QQ COin ^"T-
CM" • i-^ T^ n ^ CM" CM" CD" -^10"
OCM OCO ^f^f COCO Oirt Oi/
mm oio coin coco Orn o?
T-^ T-CO cor-- -r-in co- co -
T- T-
O T3 O T3 "tT" ^* *O* ^^ *O* ^* ~Q~ ^
*S*c3 *o*G *c3*c3 *G*o TStS *5*c3
coco coco coco coco coco coco
TJTJ "OT3 TJT3 T3T3 "OT3 T3T3
COO) COO) O)O) COCO COO) COO)
^f 10 3 m ^ in ^~ m ^ m ^ if,
oo oo oo oo oo oo
o o' ca ! o o o' cp i ci o' o o" o"
cow cow ww cow cow wco
go oo oo oo oo oo
cc cc cc cc cc'cc
oo oo oo oo oo oo
oo oo oo oo oo oo
22 22 22 55 22 22
O.CL Q.Q. O.Q. CLQ. D_D_ D_O_
O O O O O O
CO CM O) O f- CO
in o) in CM co i-»
T- T- CM co -sr in
o o o o o
T^ T- h- en CD Tf
gJ 0 CO CD Tf 0
ro T-" -T-" ^ CM" co"
o o o o o o
r-- 10 TI- rj- o m
o_ in •* CM CM CD
CM" CM" co" rr" CD" N."
5 S g 'S | §
CO J3 CO _O 03 XJ
r^ IY CO CO O) O)
U- 11. LL U. LI. 11.
o: cc o: cc cc cc
£
w
o
IB
0
3 E
— CO
IS
.c
•*- CD
=e §
.2 3 «
S ^ w
i 11
C Q C
03 CO ®
^—' <^ C"
c — co
CD CO -O
E Is- c
s °' 8 c
^ b" S o
£ £ « 1
tit I
•c 2 -S "
tS o_ to ^
"CD — =5 o
o
CD
XI
JO
o
CO
•g
•§
I
O
II
2
o
•s
CO •=;
Si co -i=
CO "*• CO
EC CO
.— O CD
"w c
tn ~
•^ \J
is
* "E
T3
CD
O
CO
O.
w
J3
Q.
_ CD
T3 T>
CO CO
O) O)
§•§•
0_ 0.
(D
J3
^ CD
CD £
CO CO
CD >
O O O
~~ o o
c £ o o
co _ -a T)
O- o O O
o .£
f> S CD ca •- t
"CD S •§ ~° "*- "-
> S 5 "o "a TJ
^ o c3 "cj o
CD 65 — ca co co
<2 fe in o- o. a.
m c w. J E E
_ 05 O _— —
5-69
-------
to e ^,
Q. CO ^
•— CC ^
§5 S
CB 55
1-f
0^ CO ^
J LU
co en
o S _
2 c >.
a CD ^
s E 2
7! CD
S co
CD
ill
c •£ 2
"o c:
o "2
85 CO ,_
o c >
t^ ^x
C «£
& 55
CD 2? !r
III
SO >
Q) 5»
CO LU
o o E
CO
o
LU
en
•i-:
S"
in
CNJ
cn
o
LU
5
T—
0
c
'2
Low-odor con
en
0
LU
CM
CO
O
CM"
t--
li
oc
§
LU
CO
CM
co
CD"
Q
en
o
LU
CM
in
CM
0
0
'{2
CO
^>
Low-odor con
0)
o
LU
CO
c\i
0
8-
co"
i?
CO
o
LU
CO
in
S"
R.
^j ^
01
0
LU
S
••*•
o
2
c
1
Low-odor con
o>
o
u
u
O
CO
c
c
L
: i
CD CO
CO CO
t :
0 0
§I_P
— ?•
It
5 §
lU
2 2
||
,
CO
CO
fe
1 1
0 O
O CO
CM CO
1 1
0 0
Si?;
T- T"
-e
It
tf
e -o
u
1
o_
li
cc
! I
CO "*-
| ;
o o
co in
O CO
c\f CM"
•cT^
E E
It
13
Si
U
!
!
CO
CO
li
cc
: i
°s
: i
co in
CM" CM"
•3- "3"
11
9 o
li
H
2 2
!
O
T—
J3
U.
oc
! !
1 §
i-~ CM
i 1
0 0
{S ?!
co" in"
-c?§
E E
S O
l-s
2 S
PM controls~0.
PM controls-O.C
i
§
T
CM"
CO
li
CC
1 1
0 °
CO o
r i
o o
in CM
CO T-
co" m"
17 S
It
J2 J
2 S
SI
1
1
co"
JO
cp
u.
cc
in
CD
(U
CO
_CD
.a
c
0)
'en
CO
CO
.0
1 fe
I
o
CO
o
o
CO
1
> C T3
.•e fc CD
C 5 J3
•o
CD
(0
ca
cc
D.
LU
1
1
S t
CD E?
£ CD
P~ Q)
O C
T3
CD
•e
co ^
^ CQ
o 2 |
Sll
II5
CD
1
a.
CO
"a.
co' CD
•a TI
D5 D)
a. Q-
co co
LU LU
c c
° .2
'in co
.
o o
o o
»— fc_
•8-8
o o
jj •«
S o
05
m
il^
£ o co
tl c ca
£ £
"O *O
(U CD
||ll '1
•3 5 o.
o
-co £ S S-
•I 8« s
LU JE —
5-70
-------
CJ
p
Q
i
§
H
CO
O
Pi
O
o
CO
cu
H
W
K
CO
H
J
i
H
H
i
in
W
a
w
«
t
a
p
0!
C
•3
oremen
_c
J3
CO
E
EP
CD
c
o
1
c
«!
o
i
^
1 :
j
1 1
o ~
c
0) ££
m ^
p_ i
CO
.
||
If
c •§
o 3
p 03
3 ^
/?
5 3
18 i
u
-S o
?£
: 3
is
S-*
3 *T
S 2
K a
!l
S S
UJ UJ
r-- c\i
co i-
*- co
st a s
SCO CO
o o
*f* <4- .
LU UJ UJ
•» co o;
»- CM •*•
0 O O
a. £. S
c c c
o o o
2 12 12
III
8 88
| | |
SCO CO
o o
•f* 4- +
UJ HI LU
tf> CO IT,
CM O S.
^- CM CO
o o o
r^» op CT)
a: cc CE
ii ii ii ii ii ii
8§ 11 §1 g? II §1
t— T- f- CM CM
II II II II II II
oo po oo oo oo oo
c\ii^ c\ii--I o"5 com IOCM mcv
*- •*- *- •"- CM CM" CM c\f co" in co" w"
— — „
9*^ *o*o t>*o "0*0 "o*c5 t5*o
5^ :£•§ §-8 -S-S -S-S ||
Cu GJ O>D) OO> O3D) D)n» D>O3
oo oo oo oo oo oo
° o" o o o o* o" o o o" o" o
|| || Ii Ii II li
88 88 8§ 88 8§ 88
22 22 22 22 22 22
O.O. Q.Q. Q.Q. O-O. O.QL CLQ.
1 1 1 1 1 1
0 0 g g 0
*~ ^- T-1 CM" to"
r2 f2 S S S -°
= fe fe fe £ £
5-71
-------
8?
I
"a
•a
ULJ
-= 2
5 *
>>
= .2
I* -n
O
in
O
to
O
in
o
in
o
m
CO
CO
CO
CO
o -5
2 «
CO
CO
CO
04
s
CO
cn
in
in
CM
co
co
CO
cd
CO
CO
CD
CM
CO
CO
CO
CO
CO
CM
_O
II
in
CT1
CO
cn
CO
T- CM
«- -S
5 2
3 g
co H3
o
~
o •
o
•o
>l
O
Ul
T3
CJ
CO
a.
O
CO
•a
cu
J3
•a
a>
cu
JD
O
(O
•o
CO
.a
•a
cu
jtf
o
ca
D.
CU
JD
TJ
CO
TJ
CO
.2£
c
cs
a.
O
CO
T3
CO
J3
•6
CO
^£
O
ca
a.
TJ
CO
^£.
O
co
a.
in
cn
CM
CD
r--
CD
Tf
CM
CO
-
CU
JO
O
CO
•a
cu
jp
"g
•a
co
_Q
-a
co
^i:
s
a.
CD
CM
05
CM
CM
in
in
cn
CM
•a o 2
o o E
•52 =
ul
o =.
Q S
CM
LL
a:
to
c?
uL
tr
U-
cr.
LL
cn
a:
I-- cn
U CM
8 UJ
c O
ID Q
op
uL
DC
cn
CM
in
_co
1=
c
CO
"co
"5)
LU
•o
CO
o «2
s J
03 ca
o o
i co
•a -a
CO 3
•c "o
—
o o
o 2
co
cu 'o
II
CO
JO CM
.cu in
ca cu
.0
"S £
5-72
-------
CO
H
CN
H
in
c
"t
g
C
"5
"i
UJ
->•
c
g
'c
•3
£
c
o
"55
en
c
UJ
•£
o
en*
O
"en
.52
5
"55
-
3
5
5
O
5 :
n~
5
5
D ;
>
i •
—
t
O <
•S C
•^ t
3
CO :
_O
O •
5.
TJ
X
.5 -
"3 ,
w =5
.0
_o
O T
O fl
TJ
X'.
i_ 0)
3 T!
H
O
5
3 •-
D n
5
i"
"o c
It
U O
3 T)
11
/J TJ
O
5
3 ~
5
^ en
•" flj
» 0
3 I
U _3
O O O O O
in in in ^ -^r •<;
CO CO CO CO CO CO
co co co )ri co in
& °j w S o 5
o r-- o ^ m i-~.
"r" CAJ ^" CJ5 T- CV
s g a g s s
T- CM •<*• OT ?- 0
co m co co CD CD
T- CM •* en •>- cv
CU CU CD CD CD CU
XI XI XI XI XI X
"s s s 1 "i 1
O O 0 0 0 0
en eo en en en en
CD CD CU 0) 0) ffi
XI XI XI XI XJ XI
TJ T) T3 T! TJ T!
CD CU CU 0> 0) 01
o o "o "o "o "o
co co co co co co
CL Q. a. n. Q. a.
in tn tn co t- co
C\J CO TJ- cc cc cc cc cc cc
000
in in in
CD CD CD
^r Tf T
5 J2 J
CD ^ CD
CO ••— CM
^- W r:
•»- CM
O h- -^
T- T- CO
cu cu cu
,^> Tf~t J^
XI XI X
O O O
en en t/,
TJ T) T!
CD CU 0)
XI XI X
TJ TJ TJ
CD 00 CC
o "o ij
CO CO CO
Q_ OL a.
CO 1- CO
co r~- co
•^- CM TT
a> CM i>-
• i- co in
UJ r^- oo 05
C) a. u. u.
Q cc cc cc
o
o
o
X
=5-
^
ID
Xj
B
'ro
en
m
"53
•a
o
m
XI
o
CO
o
cu
en
c
O
"«
.52
0
CM
O
g
O
X
5-73
-------
8
•4
W
P
8
w
w
u
H
rci
i
in
i
v_
to"
O
'co
CO
E
CD
CO
C
CD
E
CD
O
o
o
*
^-
CM
O
CO
D_
~o c
.b o
O °
Model
recovery
furnaces
8 • ° °> 0
00 TT-" CM"
Nt" co in CD
5 • £ -- &
1
% 8 8 -
T- O C> £
T^ CM" CM"
O) CM •* CO
CM •<- cn co
CM •* m T-
0) CD CD .- Q;
O O O X
JD & -Q -Q
13 rs 3 5
o o o o
CO CO CO CO
T3 T3 T3 TD
CD CD CD CD
•9 -9 9 -9
-a -a -a n
CD CD CD CD
"0 ^ ^ 0
CO CO CO CO
Q. D_ D_ D-
g Q CO C£
UJ T3 51 12 ^
(J) Q) T— CM CO rj-
Q "£ LL. UL UL U.
z > nc cc cc cc
-t- O5 o
CO CO CD
in co T_-
5 CD •$;
CM ^ CO
o o
g co CM
T— CM"
in CM m
^ -^f CO
T- CM •*
CD CD CD
J3 JD -Q
225
O O O
CO W CO
T3 T3 T5
CD CD CD
-9-9-9
T3 "a T3
CD CD CD
"o o o
CO CO CO
a. a. a.
T3
CD
•e
CD
C
c o u. u. uL
ID Q CC CC CC
CD
_2
co
-*-»
"c
13
.C
UJ
T5
CD
•4— •
O
CO
o
CD
£= O
TJ £
CD O
•e o
CD •*-
1!
CD $2
| in
> _CD
^ JD
JD CO
CO 1—
-*-'
.<£ CD
£ co
W CO
_CD in
CO )
CD
CO
5-74
-------
CO
H
H
t
in
.>,
.Q
to"
O
'to
to
E
CD
lg
f«"
CD
E
CD
b
JZ
O
O
X
~2L
CM
0
CO
2
Q_
"o c
-b O
C 5=
0 0-
O 0
>^ to
5r CD
— CD e\
s s §
is>I
O O O O
in co CM i-
CO CO CO T-
T-" CO" M-" T-
m ° °
£ £ S fe
03 •' .-" CM" W
o o o o
CD CM 0) CO
•— 1— >_ 1_
CD CD CD CD
-Q _Q X3 JD
S E S E
0000
to to to to
"D "O T3 T3
CD CD CD CD
J3 J3 JD JQ
1 1 1 1
T5 T3 T3 T3
CD CD CD CD
0 0 "0 "0
CO CO CO CO
Q- a. a. D_
c ^
j Q CO 0)
^ ^ T «? e? f>
Q 03 u. u_ u. u.
?L > CT (T EC CC
o o o
r^- CD CM
T-_ C> IT
•«-~ T-" CO"
^3 C
en g cc
CD ^ ».
O 0 0
CD O 1^
in CD CD
T-" CM" -^-"
O CO 0
CM CO CD
co in a>
CD CD CD
O O O
-O JD JD
E E E
O 0 0
to to co
T3 T5 -D
CD CD CD
•9-9-9
~O TD TD
CD CD CD
o o "o
CO CO CO
n_ a. a.
D
D
>
•5 uj r^- co en
^-* * ^ * 1 i
c O u_ u_ u_
Z) Q DC DC CC
— : to
CO c
to
C
O
-
CD "to -55
C CO 13
•*—»
m
O
o
.Q
"D CO
§1
— O
CD o
> CD
^) to
CO CD
§ s
— -^ o c
li
8s
i— o
£ o
CD CD
CD E
~
CD
CD
to .S2 m
to to __
c 52 2
g E 2
CD ^J
C CD O
s^i
-u J-J
c •*' ~
CD !Z cn
c V CM
^— IO _- *
CD Ul* O
.h: _CD • -
§"2 s
I— I v
J3
*-• "~ •*-"' — ^
10 CD 'o TD ^.
O CO S S g
g 2 ^ § CO
CD -Q CD ^~
r— r* v_
t Q Q. CO
p T-i CD •
t— *-J i_ o
CO
to
•2 S « ^ H
to .2 co £3 ^
to to -Q o CD
"E ~ -a g .5
^ £ B c ^
>, CD to E 2
co c^ E >« 2
"o ro "i^ E* D-
g "5 CD CD -
O O CD
CD O br
CO CD y
--^ w ^
CO
? —
CD m
V_ T
O X
5-75
-------
TABLE 5-14. ENERGY IMPACTS FOR HCL CONTROL FOR MODEL RECOVERY
FURNACES51
Model
recovery
furnaces
NDCE/con-
verted DCE
RF-1/4/8
RF-2/5/9
RF-3/6
RF-7
Unconverted
DCE
RF-7
RF-8
RF-9
Control
option
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Incremental
energy impacts,
MWh/yr
AQQ
yoo
1,770
2,570
593
625
1,040
1,870
(a) Baseline energy impacts are equal to zero. Packed-bed
scrubber energy impacts = [0.00018 x model gas flow rate
x 3 in. H20 pressure drop x 1 MWh/1,000 kWh x 8,424 hr/yr]
+ [0.746 kW/hp x liquid flow rate x 60 ft head x specific
gravity H2O x hp/3,960 gal x 1/(70% pump motor effic.) x
1 MWh/1,000 kWh x 8,424 hr/yr].
5-76
-------
TABLE 5-15a (METRIC). WASTEWATER IMPACTS FOR HCL CONTROL FOR
MODEL RECOVERY FURNACES3-
Model
recovery
furnaces
NDCE/con-
verted DCE
RF-1/4/8
RF-2/5/9
RF-3/6
RF-7
Unconverted
DCE
RF-7
RF-8
RF-9
Control
option
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed Scrubber
Packed-bed scrubber
Incremental
wastewater
impacts,
million L/yr
10
18
26
6.1
6.3
10
19
(a) Metric equivalents in this table were converted from the
calculated English unit values given in Table 5-15b. See
Table 5-15b for footnotes, which include calculations.
5-77
-------
TABLE 5-15b (ENGLISH) . WASTEWATER IMPACTS FOR HCL CONTROL FOR
MODEL RECOVERY FURNACESa
Model
recovery
urnaces
vIDCE/con-
verted DCE
RF-1/4/8
RF-2/5/9
RF-3/6
RF-7
Unconverted
DCE
RF-7
RF-8
RF-9
, —
Control
option
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
..
Packed-bed scrubber
Packed-bed scrubber
Packed-bed scrubber
—
Incremental
wastewater
impacts,
million gal/yr
__ " ~~
2.7
4.8
7.0
1.6
_ 1 —
1.7
2.8
5.0
. —
(a) Wastewater impacts are zero at baseline; therefore,
control level and incremental wastewater impact numbers
are the same. Control level wastewater impacts =
wastewater flow rate (gpm) x 60 min/hr x 8,424 hr/yr.
5-78
-------
§
H
I
H
O CO
P4
Pi O
O EH
D U
Ortl
H fo
CQ
Q
g
H
fti
VD
H
I
LO
— co~
CD n rn
CD 3 : en CQ
{0 *C ^
CQ 5>-
_,....
C
0)
£ o>
Is
cr
LU
"o
°S O "
•? — ' °E
S m D
CO
o
LU
CO
p
. "*
,3.
cp
LU
8
c\i
in
CO
tn
CM
•cf'
O
CO
CM
O
r--
CM
o
o
o
o"
o
^^
CO
. c
8, E>
co co
to CL
• CO
CM ^
'co
^m
6
CD
CO
o
LU
CO
CD
"*
^j-
cp
LU
8
CM
in
in
in
CO
0
CO
CO
o
in
•«-
o
§
o"
o
r--
en
. _c
ffi CO
'5
CM
6
CO
CO
cp
LU
CD
CD
"*
rj-
O •
LU
8
CM
in
CO
in
P*-
Cvj
O
CO
CD
O
CM
CO
O
O
1
en
. c
8> "ff
co co
"co Q-
i CO
"3
co
6
CD
in
J2
JD
c
CD
>
"en
CO
a>
en
LU
•a
CD
_O
CO
o
CD
•a
CD
•c
CD
>
C
O
o
CD
CD
_CD
JD
a
tn
"c
_Q
I
cr
CD
O
H
Q
H
X!
O
D O
O«<
H fc
ra
Q
i
CO
H
VD
H
LO
W
— -m" —
J 3 < c7
2 § Is 2
§ & § S
0 S5 :9
O
o_ "eo
l§25
I 1-IS
CD OJ tD ^~
E? -^
o
2 *^
D ^
•r ff ^e
.;2 r* *^
i 8
tO 5 (f)
O) CO CD
*- *— a
m 0. C
>|-s
co m-
® *-- r—
co 2 ^
II s
TD
a a, fe
S. 2 £
•s c: , C
JZ. O
i- o
CD" .£
C O
O .£
i: o
| c
t CD
_- CJ
C *
^f^ CO
-s m
E. i
CD 3
T3 CO
!-<
CD
E 5
g o
CD *^
N £
c S
CD H
_Q CD
CD" °-
-a S
CD CO
2 H
'
S a.
CD C
Tl O
" c
~ _m
o cu
CD C
.5 3
5-79
-------
I
8
a
H
I
ll
t "o t^
= 3
,., ~°
UJ g
_ en"
li*
n o D)
u "§
•3
> e/j
~" O >,
2l!|
"£ E
o S
o
ll
o c
0 °
g S ^
.5 o >,
CD "en "3=
cn en ^
eo -F S
CQ E
CD
"53 ,
"B3-1
2 m §
CO
O)
cn
CM
CD
O
"6
l_
"c
o
0
S
D5
"c
CD
o
_]
CQ
c^
§
CQ
CO
cn
CO
0
•^
—
L_
"c
0
o
cn
CO
o>
0
O
_i
CQ
CO
3
CQ
CO
cn
'
CO
CO
f^.
•r^
"o
L«
"c
o
o
cn
CO
o>
g
O
_
CQ
CO
CO
C
a
c
CD
"o>
cn
CD
"ca
-C O
en ~
•g s
1-S
3 3
.2 "o
CO C
8
"81
•e -5
Q) O
> o
co ii
CD J3
CD ^
JS
3 ^
co -2
-^-» co
en H-
_
CO
.>
CO
H
I
W
r- cn"
!•! o
0 "O c»^
E :3
UJ li>
r- ">"
° § £
en -.r; .
2 !« 1
"c c: •—
o ^
"o c
"c '-5
° S-
0 °
Baseline
imissions,
ton/vr
CD
"53
•§31
^— ••;?
•S CQ 3
co
.r_
CO
CD
O
"5
"El
O
o
cn
CO
TO
CD
g
CQ
CM
CO
,_
3
CQ
CO
cn
CJ
m
,_
T—
"o
"c
o
o
cn
CO
TO
"c
CD
g
CQ
CO
CM
3
CQ
CO
O3
^~
O)
cn
•^
"o
"c
o
0
en
CO
TO
'c
C
C
5
CD
cn
CO
C
5
CD CD
« S
i— 0)
f I 8
iff
CD M c •
•o 2 co
o ca
E -5 g
x y CD
^ re ^
co ^ ^
CQ ^ _-
o "O O
= c c
"3- — CD
= cn .E
ro. a.
£ < CD"
o X c
CO o O
"= Is 1
O CO ^
"en 21 >,
_en o ~g
E w XI
§30
O cn
II CD ~ —•
sJUl
-> *-* CD
Slo^
o °- *< c
"55 £J CD CD
P O >*
«-£
O- « »
o.
C T-
5. w
ra co
o x
,
• en
•£ CD
CD cn
E'g.
CD' E
•o o
03 O
-a c
— ^-^ »w .«•
S m E |
O =9. o -5
in
5-80
-------
TABLE 5-18a (METRIC). SECONDARY EMISSIONS FOR MODEL BLACK
LIQUOR OXIDATION UNITSa
Mode!
BLO
units
BLO-1
BLO-2
BLO-3
Control
option
BLO vent gas control
BLO vent gas control
BLO vent gas control
Incremental emissions, kg/yr
PM
400
807
1.210
SO2
9,710
16,900
29,200
NOx
776
1,560
2,340
CO
1,470
2,970
4,440 -.
(a) Metric equivalents in this table were converted from the calculated English unit values
given in Table 5-18b. Refer to Table 5-18b for footnotes.
TABLE 5-18b (ENGLISH). SECONDARY EMISSIONS FOR MODEL BLACK
LIQUOR OXIDATION UNITSa
Model
BLO
units
BLO-1
BLO-2
BLO-3
Control
option
BLO vent gas control
BLO vent gas control
BLO vent gas control
Incremental emissions, Ib/yr
PM
883
1,780
2,670
SO2
21,400
37,200
64.300
NOx
1,710
3,450
5,160
CO
3,240
6,540
9,780
(a) Secondary emissions are zero at baseline; therefore, control level and incremental
secondary emission numbers are the same. Secondary emissions for PM, NOx, and CO
were estimated based on energy and emission factors for PM, NOx, and CO. Secondary
emissions for SO2 were estimated based on (1) energy requirements and the emission
factor for SO2 and (2) 1.88 Ib of SO2 generated for each Ib of TRS combusted.
Calculations for energy impacts are presented in Table 5-19. Emission factors =
0.15 Ib PM/MM Btu; 0.73 Ib SO2/MM Btu; 0.29 Ib NOx/MM Btu; and 0.55 Ib CO/MM Btu.
5-81
-------
TABLE 5-19. ENERGY IMPACTS FOR MODEL BLACK LIQUOR
OXIDATION UNITS
Model
BLO
units
BLO-1
BLO-2
BLO-3
Control
option
BLO vent gas control
BLO vent gas control
BLO vent qas control
Incremental
energy impacts,
MWh/yr
1,720
3,480
5,210
(a) BLO control energy impacts = 503 hp x 0.746 kW/hp x
8,424 hr/yr x 1 MWh/1,000 kWh x model vent gas flow
rate/16,327 acfm. The hp requirements and vent gas
flow rate were provided by an individual mill that
controls vent gas emissions.
5-82
-------
TABLE 5-20a (METRIC). TOTAL REDUCED SULFUR COMPOUND EMISSIONS
FOR MODEL BLACK LIQUOR OXIDATION UNITSa
Model
BLO
units
BLO-1
BLO-2
BLO-3
Baseline
emissions,
Mg/yr
4.2
7.0
13
Control
option
BLO vent gas control
BLO vent gas control
BLO vent gas control
Control level
emissions,
Mg/yr
0.08
0.14
0.25
Emission
reduction,
Mg/yr
4.1
6.9
12
Emission
reduction,
%
98
98
98
(a) Metric equivalents in this table were converted from the calculated English unit values
given in Table 5-20b. Refer to Table 5-20b for footnotes.
TABLE 5-20b (ENGLISH). TOTAL REDUCED SULFUR COMPOUND EMISSIONS
FOR MODEL BLACK LIQUOR OXIDATION UNITS
Model
BLO
units
BLO-1
BLO-2
BLO-3
Baseline
emissions,
ton/yr (a)
4.6
7.7
14
Control
option
BLO vent gas control
BLO vent gas control
BLO vent gas control
Control level
emissions,
ton/yr (b)
0.09
0.15
0.28
Emission
reduction,-
ton/yr
4.6
7.6
14
Emission
reduction,
' %(b)
98
98
98
(a) The baseline TRS emissions for BLO units are based on an average TRS emission factor
of 0.10 to/ADTP (assuming an average 3,400 Ib BLS/ADT of bleached and unbleached pulp).
(b) The control level TRS emissions for BLO units are based on 98 percent TRS control.
5-83
-------
tt
5
j
c
J
o 5
H
co :
CO
i
S
CM
Q
)7
§
CO
w
Trl
S
S
*"~J
S
g
^3
EH
CD
>«*
2 w
f^ "
cog
H^
I'M
£_(
(4
FT!
CO
Q
t
o
H
oi
§
"— '
(ti
H
05
IO
w
EH
ill
**l
U «
3 8 "^
§1°
d
CO
1 — - °
.X
&
liE
S E
S
a.
0
c?"5 x
&
S o
£ g o
f»
cs
III
g S co
^ o E
CO ^
5
a.
_ n
S.2!^
11 -n
111
<
§ 2 3
g S, "
S g c
S « *
\\\
\\
G 4
CD
•§
O
g S S S g. ° 1
d d d o o o o
ooooo o o
d d o' o o o o
en o> en m co
-------
H
CO
O 05
CO O
CO EH
H CJ
CO
I
CO
H
H
OJ
in
CO
32?
CD C C
0 ° -°
g 2 CM
•g , C7
US §
cS g CM
O
g «
o 2 c
« n_ S
m ^
^ -D O
I §2
a
CL
g
= o> S!
B *- i
a.
"o to
c 75
J3 ti 0
II
S.
£
-* *~ *5
Jo
_.
Sr
-9- o" >-
— c
a
III
112
2=a
= .§ o
Baselino
APCD
-§ £
S co
CM
O
O
CM
O
o
f.
r-
S
o"
o
X
g
OJ
0
g
8"
Ol
o
1
to
scrubber
i—
Q
CO
CM
O
CM
O
O
t^
r-
8
in
s
o
in
1
s
in
8
O
O
o
1
.
1
v>
CM
CO
CM
0
CM
O
t«-
«
O
f-
t^
o
CO
CM
s
5
w
s
N.
s
o
1
o
o
r-
CJ
s
?
s
scrubber
n
CO
CM CM
O 0
O O
CM CM
O O
f~ CO
o o
K ^
h-
*"• d
S 8
§ 5
8
° O
g o-
o
§ g
1-" n
§ g
8 |
5 8
3 S
I 1
S S
o 2
% S c!
2 E 1
0 -S
in ID
^ 10
CO CO
CM CM
0 O
CM S
O O
CO CO
o o
r*- r*-
h- ^
d d
S 8
to co
g
S o
in r-
§ s
g g
o o
g g
S p^.
- w
I 1
2 B
o o
0) (0
0 0
II II
E E E E
1 "•§
to f~.
CO CO
5-85
-------
H
O
CO
CO
H
P
a
CO
O
H
(TJ
cx
CN
in
ta
§ g
•^ 3 °
E "o
UJ CO
t_
0
T3
CO
k_
o 2
"to O_
1
"Si
5Uf
^T rt
CO
"§
B o.
c
c
c
15 c
i: O
§1
O °
t~
>,
OL ^
CO ^
c
.2
co
E
CD
1 1
CD
CO
CO
CD
•5 y
E E
5 c7
CD CO
•q- CD
O °
^ O
CM 00
o> in
88
d d
•4- CO
T- CO
'S"n
CO CO
—1 — 1
m CD
CO CD
"CD co
O CO
T-_ O
0 0
"c. "c
o o
0 0
D- D.
S
d
> \-
> co
CO 00
•sr CD
T- CM
0 0
d d
o o
CM CO
5§
°. 0
0 d
c\i ^
S~S"
CO CO
_i —i
CD CD
C35 O)
oi"co
O CO
T- O
O CD
0 0
"c "c
o o
0 0
^? ••—
O_ Q-
co
0
d
^
CM
£
CO
CO CO
-3- CD
d d
co in
CO CM
0 0
d d
co co
TT CM
£.=£.
CO CO
CD CD
CD CD
]?]?
d o
0 0
c c
o o
0 0
*5 "5
0. 0-
in
o
d
o
03
CO
SB
CO
(O CO
-3- CD
co in
0 0
d d
CO 00
in f-
•>«• CM
0 0
d d
CM r»
co co
.0 J3
co eg
CD CD
??
c? _ra
O CO
•^- o
CD CD
0 0
O O
0 0
2 2
D. 0.
d
in
^ —
CO
m r^
0 0
d d
O) •*
T- CM
o> in
0 0
0 0
d d
T- CO
.0..0,
j j
CD CD
CO CD
CD CD
O CD
T- O
dc?
o o
0 0
0 CJ
^> ^>
D- O-
CM
O
O
CO .
CO
in
CO
in t^-
CM CM
o o
d d
co •£•
s°.
0 0
CM T-
£,£,
COCO
CD CD
CD CD
CD CD
O CO
T- O
22
o o
0 0
^> ^>
0- 0-
8
d
in
in
CD
CO
m h-
CO '3-
0 0
d d
co co
in h-
CO CM
o o
d d
CO CD
•*3- CM
£,£,
3 53
CD CD
CD CD
S S
O CD
•^- O
22
S 2
c c
o o
0 0
5 2
Q- Q.
CD
O
d
CD
CD
Q
CO
.a
CM
in
co
I
CD
CO
CD
2
CO
*-*
'c
co
ill
(D
o
CD
JD
H r- co
O £
co
co £
2 J5 c •=
o
CO ^
t re £ o
p o •> co
-*= co * .S
•£ o
5 w
o •£
c
o —
i: CD CD
£ 5 .£ =
| CO' S S
> CO X X
03 «~ CD CD
— Q n_ .,
S S S m
in
CO CD
CO
CD CO CO
'R-« s S
^^^ £
•gr 8 8
co a.
a:
co
.a o
5-86
-------
CN
CN
in
c d"
0 0
to tS ^
E -o
LU 03
-^ to
C -s- i™
2 ?T <
O
o
T3
£
o 2
'55 a.
CO
LJJ
u.
I .«
~1 35 o-
co
to
'E
CO
"55
— -^
O Q_
t_
O
O C
is o
O o
:-. _co
J *— ^g,
•^ n
I Q
2 co
co co
"3- CO
o "-
8 q
d °
co o
*- CM
cn CD
o o
o o
d d
CD m
••- cn
co eg
m m
c c
o o
o o
O CM
CM -r-
9 9
2 2
c c
o o
O 0
n. Q.
•s
d
O)
CVJ
Q
CO
CO CO
•* co
i- CM
0 0
0 O
88
°" d
CD CO
CM i-
o _o
co eg
m m
c c
o o
2? 2?
0 CM
CM i-
9 9
2 2
c c
0 0
O 0
.2 .5:
Q. D_
CO
O
d
O)
CM
CO
CD CO
•* CD
CM •*
O O
0 0
O O)
•* in
CO CM
0 0
d d
r*** oo
•sf CM
3"n
CO CO
m m
c c
o o
2? 2?
CM 2
9 9
2 2
c c
o o
o o
0_ Q.
in
o
d
00
CO
co
CO
$ §
co in
o o
0 0
CO CD
in co
ss
d d
CO T-
co •<«-
JD S3
CO CO
m m
c c
0 O
O O
O CM
CM i-
9 9
1 t
2 2
c c:
o o
o o
0. Q.
§
d
r--
CM
T—
CO
in N
oS
o o
CM CM
CD CD
O O
O O
d d
CD m
••- oj
o o
CO CO
m m
c. c
O 0
S3 S3
O CM
CVJ T-
9 9
2 2
c c
o o
o o
a. a.
S
o'
CO
co
in
Q
CO
fe.it
CM CO
O O
0 0
•* in
co -3-
d
CD CD
CM T-
0 0
CO CO
m m
c c
o o
O _O
^3 ^^
^D ^D
2 2
c c
o o
0 0
0. 0.
o
d
(D
CO
Q
CO
in r~-
^r in
0 0
o o
CM i-
co co
CO CM
0 0
d d
r»- co
•^- CM
o c
CO CO
m CQ
c c
o o
O f~
O CM
CM t-
9 9
o o
c c
o o
0 O
a. a_
o
d
§
Q
CO
m
o
o
o
CM"
m
c
c
2
CO
co
_J
m 5
.Q 0_
CD <->
ro c
"~ 03 v.'
C35 O 03
S^fci
% q £ g
m o a $
^« I I
CO ~
cf
CO
m
c o
o
•^-^ -_ ^r ^_
i- fc -Q ==
O 03 S3 03
O Q. E ]«
•p ^ lo E
5 _ D5 O)
'to •— ~
— .52 "w
E -^' x 'x
g3 ^ CD 03
2 !a "° "°
^i|l
•eS-« §
•^•Si o o
o Q. 'o.'a.
5I-- 03 CU
^ C3 to to
Ixll
E ^.o §
03 T? 0 CJ
a. w a. a.
"aT S~ "o"
5-87
-------
to
co
H
1
CO
CO
H
Q
CO
>J
H
Q
Pi
g
CO
H
CO
CO
H
U
w
co
u
H
IS
ctf
m
CN
i
in
§
S
o
5
o
o
~
D
ra
5
D
5
—
o
o
X
O
^
CM
O
CO
"5>
Q.
~o c
£s o
11
"ra
eo~
0
'to
CO
E
CO
CD
"53
CO
CO
m
O
O
X
*^
c\
u.
~
c
_
"o F-
0 Q
2 CO
CD
0
CM
O
•i —
-
CM
2s
^-^
"o
c
o
o
D_
in
co"
in
^*r
T-
co
CM"
in
D
CO
CO
CM
CO"
o
in
CO
'S'
'
"o
c
o
o
Q.
?
in
r**
en
en
co'
CD
CO
o
CO
CM
CM
CD
o
CO
CO
CO
"S"
" — *
"o
c
c
a.
CO
CM
CO
CO
o
^
^
Q
CO
CO
CM
in
"o
CO
c
CD
"o>
CO
CO
"co
'c
a
CO
c?
UJ
•a
J°
o
ca
o
CO
E
o
"D
CD
•c
CD
C
o
o
CO
CD
CD
J3
a
CO
Ic
.„
CO
c
CO
CO
"5
cr
CD
o
"S
S
CD
£l
^2
0
CO
5
CO
CO
JC
•"§
,_
o
ro
.»
_E
"CD
1o
E
c
*^z
.52
CO
"5
"c
CD
E
CO
o
_co
Q.
CD
' . C
CO O
CD
O CO
S w
- "S
0 £
•*— CO
.0 f—
co -S
CM tn
in ^
co £
_o co
ca $
i__
£
2 ^
CO C
"S E
o: ~
»^^
Tj
CO
fe
f3j
E-«
H
j>
•4
O
CO
CO
H
P
1
CO
t "I
w
P
g
0
CO
o
H
co
CO
s
w
s_4
cyi
*?r
|
O
U
CO
.
w
CO
H
1
ft
' ro
05
i •
in
w
1
E""'
^
J
o
5
5
z
D
CO
—
0
D
5
~—
O
o
X
O
CM
o
CO
2E
D_
5 c
It
^^
ID
co"
o
'co
.22
E'
CO
CD
"55
C/]
CO
CD
O
X
CM
O
CO
•*£
c
""""
fj f-
O c
2 cr
en CM
CO CO
•r- CM
§ !
•* "3-
CM r-
CM CO
CD !*•
TT r-
S" ~S~
—
"o "o
c c
o o
0 0
2 2
o_ o_
2 5
en f-
cri •*-
in CM
CM •*
T- in
in co
in CD
H H
0 Q
CO CO
CO
0
in
CO
CO
CM
£
CO
CO
CO
' •£
**— *
"o
c
o
o
^
Q-
to
in
0
CO
in
in
"•"
C
a;
6
CO
x"
O
^- if
CM" S
O ^;
W- ^
1 ?
£ -°
w in
O "T
t5 °
45 u
C CO
-° 0
_co CJ
CD C
o
C CO
co .w
CO E
0 UJ
ca
a. ra-
E ^ "^
D) CO
CD -Q
C ^"
CD H
c £
0 -o
"D CO
CD -t:
CO »r-
ca ^
i |
E co
"•^ CO
X? 0
Qj tJ
CD Q
ii
CO ^
c E
.2 "^
"crt C—
CO CD
E o
CO ^
*^- CO
^ C
ca o
"D *2
cz ca
o n
o a
CO TP
co 2
CO
o
CO
CD
C
to
CO
_ _c
'« ^ J
S ^
it
II
S 'x
m <°
. - CD
D O
O -a
CO
si
CM
co
CD
CO i2
co -
N- E
5-88
-------
TABLE 5-24. ENERGY REQUIREMENTS FOR MODEL SMELT DISSOLVING TANKS
Model
SDT's
SDT-5
SDT-6
SDT-7
Baseline
energy impacts,
MWh/yr (a)
10
17
30
Control
option
PM control (c)
PM control (c)
PM control (c)
Control level
energy impacts,
MWh/yr (b)
100
167
300
Incremental
energy impacts,
MWh/yr (b)
90
150
270
(a) Baseline mist eliminator energy impacts = 0.00018 x model inlet gas flow
rate x 0.7 in. H2O pressure drop x 8,424 hr/yr x 1 MWh/1,000 kWh
(b) Incremental energy impacts = (control level scrubber energy impacts) -
(baseline mist eliminator energy impacts). Control level scrubber energy
impacts = 0.00018 x model inlet gas flow rate x 7 in. H2O pressure drop x
8,424 hr/yr x 1 MWh/1,000 kWh
(c) Impacts were estimated based on replacement of existing mist eliminator with
a new scrubber.
5-89
-------
«.p f~\ — '
- O =1
g
o
iss
14*
0 0 -
to fi —
o ir ts
£ & O
Cn
•8
0)
S Q ^
0 0 £
§81
19! 8
0>
u.
0 Q —
^ O J>
(i
Equivalent pulp]
production rate
ADMP/d 1
1
|»s
|l8
Baseline
APCD
III
0 0
CO CO
in in
CM CM
P P
cn cn
T T
CM CM
CO rr
r-' •«-
o o
5?- CM
CM CO
CO CD
o
O 03
a. CL
S3 ft
1 1
JD J3
. § §
u> co
) T- CM
i s s
o in m in
CO CM CM CM
in in w in
CM CM CM CM
- 5 3 ?
!*• CM CM CM
cn cn cn cn
•cr ^r ^~ *^"
CM CM CM CM
"* ° CM CO
^- o o **
CO T- OJ CO
g 0 0 g
8. g § £
g § 1 g
0_ CL Q- 0.
CO CO CO CO
UJ UJ Ul LU
0)
^ Q. a. Q.
-e co co i
2 LU ai ui
o
CO
co "f in to
_J — 1 — I — 1
3
»
3
k
1.
01
)
3
5
O
CD
D
n
i E
•5 g
11
0 CO
2 CM
O
O
_L E
o "
_L. co
§1
21—
^~
a. o
5^ °
a. P
CD ro
CO
1
E
^}
CO
0-
ft
CD
j. §
O t^
— "C
§•&
0 CO
5g
°- d
! E
o t
"c 3
S "
D. 0
a>
"c
tn
co
CO
tr
co I
J= 3 E
11!
ca a. i
•c
"53
"D a) '
CO CO CO CO CO CO
d d d d d d
in in in in m w
d d d d d o'
^ h- N. w m in
CM CM CM -r- ; t-; t-
d d o o o o
0 0 0 0 0 C
Ri CM W CM ol CM
§§§§§§
• 0 0 0
i i • cn cn cn
en cn cn j j •
CO CO CO
T- CM co •* up <*?
j _i — i — i — i — '
in
0}
I
I
cn
cn
CD
O)
HI
T)
CO
_
(0
u
0)
.c
•o
a>
•e
a>
o
o
CO
.£
JO
to
c
01
1
cr
a>
5-90
-------
CO
H
LD
CN
LO
W
5?
r
eu
c
(1
o
£
3
u>
O
2
LL
CD
0
0)
(I)
TJ
perature
t-
•2
E
a
CD
CO
«
CO
O
Q.
"c
CD
CO
IT
LU
6
T)
2
D.
O
t
2
C
O
o
"55
•o
0
2
Q «
11
Q »-
O a
Q- c
^ *
81
CL 3
*< O
Q ~
O .£
CL C
81
CL D
< o
Q —
O .2
<-
2"
O Q-
= ?
o
Q.
- T3
CD A*
O C
18
•2 <
CD
.£ Q
•5 0
en CL
03 ^£
0 g
^= 2
8
in
CM
o
CO
0
CO
o
¥
o
o
o
CM
CM
0
in
CO
o
0
CL
CO
LU
crubber
T
—I
0
CO
in
CM
o
CO
o
CO
0
o
o
o"
CO
o
o
in
CM
o
m
o
o
CM
CL
S3
crubber
w
CM
1
8
m
CM
o
CO
T—
O
CO
0
o
o
in
o
o
CM
CM
o
o
T—
o
o
CO
CL
to
LU
crubber
CO
9
-J
in
CM
in
CM
o
CO
o
CO
0
o
o
CM
CM
O
O
o
CM
o
in
CO
8
CL
CO
LU
Q.
CO
LU
rt
_l
a
in
CM
o
CO
o
CO
o
o
in
CM
o
o
CM
O
in
0
CL
CO
LU
CL
CO
LU
in
_j
in
c\
in
CM
o
CO
o
CO
o
o
CM
ff
1
CM
h-
o
o
T-"
o
o
CO
CL
CO
LU
CL
CO
LU
CD
i
0
35.
O)
2"
"55
J
J
c
0
O
_L "5
11
o ,_
2 °.
Q. 0
1 t5
il
° CD
2 o
Q- d
CD ^_
i 2 -S
w n "^.
E
0
CO
o
o
o
I
75
CL
W
U
J. 0
O 05
*3 "D
c "Cr
0 CD
o ,_
2 °
CL 0
! o
S-8
1 »
o r~
_- CD
2 o
0. o
c~
u=
W
| 1 |
o o o o
d d d d
!%. («. h, h.
CD CO CO CD
o o o o
d d d d
CM CM CM fe
"-. T ^ o
o o o 0-
o o o o
CM CM CM CM
* " " "
O O O O
CO CO CO CO
0
! : : co
rr
O
d
i
d
§
d
o
CM
**
o
CO
rr
o
to
rj-
o
d
ID
o
d
h-
co
o
d
o
CM
••
0
CO
•fl-
o
CO
CM CM CM ! ' !
T^- CM CO rj-
-J — 1 — 1 — I
in
_j
CD
_l
5-91
-------
r
c
H
o
8
E
nJ
CO
§
»-
* 1
m 2
H c
c
7
*-l -i
W a
Q £
A "c
fr. r
0 fi
fe u
CO j
JHJ i
O 2
cn '
co !
H ~t
S '
r^ i
W '
<
CM
a
s
PM
Q f
3
§
X
Pj
CU
CM
O
H
1
§
nJ
VD
i
in
W
™4
cf
1 *
•o
2
"" CO
) *5 Q_
. £ <
r X
i
!
i
]
i
I 5
t n_
J
^
in
» .f
2 Q. <
5 x
/)
n
1)
5
— S
J
mm
5
j
C
o
S
o
"o
o
0
l«
•2: y
D) 2 H
2 a. <
I
s
o
tn
(O
CO ^
C -e
= C
0)
tn
CO
m
"5
•o o '
O c '
;§ .E =;
rj- CM •* CM
•q- cn -sr cn
CM in "* cn
do o o
$2 S? S S
co. g «q §
0 o 0 d
o q cn cn
CM co "m
^_ — -~-
JJ |j
" 1 " S
° ° ^ °
S "o 2 o
11 §!
^-* *c . -
IS -i
in o
: d -^
: CD o
; co i*-
2 T °V
\ ^ 3
? S S S
h- in CM in
d T^ d o
m ° h- CO
m ^ -r- co
_ ^-00
CJJ T-; Q o
o o d d
^ Q o cn
co T- co in
— S" "o" "o"
f¥ ¥ §
1^ 5. 5.
Jc? " "
m (,5 CM CM
•*-. O O O
° d cp' 9
"£ "5 "5 "o
= c c c
° o o o
" 0 0 0
D- CL 0. 0.
[^ CO CD
^ do
q o cn
CM CM CO
co T "?
^ J IJ
in
co
CO
d
k
o"
o
•5-
0
CO
"cT
CO
CM
O
O
"5
o
o
^5
^
D.
CT
C
r-
u
Cl
V
CD
CM
in
03
JD
d
03
'cn
co
CO
CO
32
3
r:
.S2
cn
UJ
T3
03
75
3 Sfe
S | m
03 co 5
£ = c
o
E "TO £
o o •>
*= 03 ?!
o E s
a ^ .a
s ° -Q
5 •£ E
C £ 0
R " w ol
2i-ffi
03 - Tn __
? S 1 £
.22 £ S B
£ o E (u
c ^ 0 "§
CO CD — ™
C CM g-g
03 m Si 3
CO 03 .CO CO
t_l 1 *-T Vr
O3 ~ C C
o 2 o o
? E.S ^
5 -s as.
5-92
-------
CO
H
J
i
vo
CN
i
LO
c c
.2 o
'co 'ft
co ^
E -o
UJ CD
^
°
-
O
o
"O
k.
c
o
"co
E
UJ
u.
o
tn
0
"co
co
£
CD
"CD
J2
2
c:
o
O
Q.
o
&^
co
o_
X
Q_
5
D-
CO
6-
X
0_
.x.
o
-
CO
c
O
"co
CO
'E
CD
CD
.C
"55
CO
CO
£Q
0.
c
~
Q.
O
S
0
o
CO
b_
X
5
o_
to
T3 CD
i -^
CO
12
^3-
»^
CM
0
03
CO
o"
OJ
CM
g
"o
co
33
O)
fs»
CD
O
CD
"o
"c
o
o
o_
co
o
o
5
_J
c\l
O)
in
o"
CO
8
o"
CO
co"
2-
"o
CO
5.
0)
o
CD
2
"c
o
o
Q-
^.
^~
in
o"
••a-
CO
CO
0
CO
g
T3
CO
33
0)
r-
CD
o
CD
"o
0
o
o_
^|
03
•
^~
CO
0
CO
in
q
^
h-
&
15
co
O)
fs«
CO
o
o"
!
O
o
o
o_
cq
CM
CO
CO
— 1
^1
O5
t^;
C\j
CM
O
,_
^~
2-
"o
CO
O5
O
CD
"o
O
O
D_
in
co
CO
o"
O)
in
o
CD
CO
CO
•5-
B"
CO
3p_
O)
o
CD
0
o
0
o_
co
o"
81
2
— '
in
co
in
o"
fo-
ol
o
CD
in
CD"
•5-
75
CO
33
O3
q
CD
"o
c
o
0
Q.
CD
o"
CO
in
i
— '
in
co
en
o"
CO
CD
o.
o'
T—
•5-
co
O)
o
9
s
s
CL
q
h-
CD
_J
CD
to
I
CO
CO
D)
03
o
"co
_co
CD
CD
X O
_CD _
.E S5 05
— o Q.
03 O r*
"° ^. 1-
E O" II
o —
O CD
.
Si
x o
«=•
O
3. 5
« =
E £
CD -
•S ^~,D-
a
c
03
o
o
Q
0_
II
0
CO
O
"co
_co
CD
p
D.
u_
O
CD
03
IM
£
03
Q.
03
J2
"53
TD
o
E
EC
CO
C\J
in
.g"
o
o
o_
"c?
o
x
r~
5
of
X
I
£
o
CD
i^
CO
O3
To
"x
03
"5
"c
03
03
O
CO
a.
03
_co
2
c
o
CJ
•^
o.
o_
CO
UJ
05
_c
To
"x
CD
"5
03
TD
CO
u.
05
CD-
CO
'o
O
o
•(—
Q.
"c?
5-93
-------
o
H
o;
CN
t
in
W
J
I
£8
ra
.it:
o
|S
CD
"c
g CM
|g
D.
"5 C
0 c
to
c
^
to
to
c
§
i
s
cc
cc
to
"O CD
2 -1
O) CD
CM CM
•<3- CM
in" co"
CM nr-
fT" CD"
CD CD
in, CM,
CD" -^
"* {Q
ll
"^CT
1 *%
* 2?
2 "o
'o
° o
ii
-> CD
< CD
^ in
X in
3 5
D CD
IO I1*"*
s S
D- T-
tn -<-f
S —i
CD T-
CM CO
CO "d"
*•—••'* **~*^
in" c5~
£^
R fe
in" co
II
"cn c"
in CO
•"? CM
^. O
3?
2 "o
"c -p
Q fS
° O
a- 1
o
in
co
o
CD
o
CO
^ —
^
CO
^1
\^
_1
Ig
*_
oT o"
CO CD
^_^
"-&
co" CM"
CO O
CO, CM,
ll
"cn *=*
S§
9 §
i i
2 "o
c -g
O Q
° o
a- 1
0
CD
cn
o
CO
o_
0
o
CD
CM
in
CO
in
CO
-1
CM
O
CM
f^w
^^
CD
CD
CM
in
in
o
C/)
"cn
CO
CM
o
o
i
o
c
o
0
D.
O
0
CD
in
CM
CM
CD
t
=L
CO
CD
CO
CD
in
o
E
o
to
"cn
co
s
o
i
o
c
o
o
D.
co
CM
CO
CO
in
CO
O)
CO
in
_i
CM
CD
CD
O
in
CO
o
co
CO
co
0
to
33
"cn
CO
CM
0
o
"o
c
o
o
D_
CO
in
in
O)
CM
0)
CO
CM
in
CD
— 1
in
03
"cn
CO
0)
2
CD
•4-*
"c
x:
tn
"cn
c
UJ
_0
CO
O
CO
.c
o
T)
CD
0)
O
o
CD
Q.
W
HI
CD
C
CO
-C
CD
n
JD
b
Q 7S
cn
^ CD
CD "n
E CD
CD -g
co 2
0) -
c c
® O CO CO
jS o
co to
-Q JD
•^ *~ T3
to h- g -E
•*-? r\i TX *tT
CM
CO W
^ in CD CD
(0 _CO CD CO
0)
O
1 to to
p. ts ts
'C i_ CO CO
0)
co
o
5-94
-------
rt
CO
H
W
W
S
H
J
J
w
Q
i
OJ
o.
&4
CO
S
o
H
CO
co
H
S
H
O
CJ
w
CO
ffi
co
H
J
i
w
£!
r-
CN
i
ID
H
J
CD
Ti
O
^^
2
^
—
to"
c
o
"co
.52
"E
CO
"co
"c
ncreme
—
—
t™
tl
o
o
>»
o
^j
to"
o
to
to
"E
CD
CD
"55
to
CO
m
CD
=
O
X
O
CM
O
CO
S
a.
c
.0
'S.
o
o
0
X
0
CM
o
CO
^
0.
to
5
CD
-d-
en.
CD"
en
in
°i
in
CM
_
S-
"o
to
"D
CD
f-
CD
O
O
"o
c
0
o
Q_
O
CM
CO_
CD
O
in
0)
m
CO
t
— '
oT
en
CD
o"
CD
CD,
In"
CO
&
CO
T£
O)
o
CD
"o
0
o
a.
o" • — •
CM S
°i en
"
ST ^
in o
2- Hi
C? O~
5- CM
CM. T£
CD" ^
en CD
•«*• CM
,~.^
Q, S]
"o *-
ll
05 D)
|°
"5 2
o °
n~ 0-
O
in
CM"
o
•sf-
co_
o
CO
co"
CO
en
CD
S
_J
O O
en co
• °. "I
% *
co S"
CD CO
i-" S-
^— '
o" o"
O CD
in" in"
co •*
. ,
& S"
1 o
03 CD
§°
CD 9
1 1
"o 2
II
0 °
CL ^
0
CM
C0_
O
co
CM
CM"
o
CO
f-
in
co
CO
— i
CD
in
CO
CM
CO
en
in
CM
CM
T—
,0,
to
T3
O)
O
CD
"o
"c
o
0
a.
CO
en
N-
en
CM
o
T—
2
—1
N-
CD
CO
in
in
o
i —
in
CO
CM
,0,
"o
CO
T!
O)
O
CD
"o
•4—1
O
0
n_
CO
CM
5
CO
o
CD
en
f^
™
in
— '
cc
CM
f^
^^»
O
Pi
T—
en
en
CO
2
"o
CO
TJ
CO
o
CD
"o
"c
o
o
a.
o
CO
CM_
CO
CD
O
CO
CD_
t—
in
CO
CO
CD
—i
CD
O
TJ
C
co
x~
O
^
CM"
O
CO
2
a.
,0
CO
o
0
m
c
o
to
to
'E
CO
TJ
ro
CO
J^
CD
£
cr
CD
•Q
Q)
m"
C
O
TJ
CD
CO
CO
.£2
TJ
0
"co
E
"to
CD
ro
*^n
CO
CO
c
o
to
CO
E
CD
to
c
o
o
CO
CO
-— '
2
S
2
a.
.0
, _
-r-
CD
II
CO
J5
o
c
"to
to
E
LU
03 -,'
CM 3
in m
JJ S
H O
C 0
TJ S
S m
c m
CD o
CO
CD TJ
£ ' •
ro 2
to tQ
C ^
p s
Is
a-2
Q) -D
.-^ CM
.2 CD
"o b"
CD ±i
"55 m
O ^
•*- .2
§§
•^= co
"5 ^
_o co
CO f^-
0 CD
o
CO
«
TJ
CO
O
TJ
£
£
co
to
w CO
CD
C £
o '$
CO ^
11
CD O
-E M a.'
H c w
.2 = LU
C '5? m
~ CD ~
to •»- to
CD O -x
J3 *- co
> § o
> l-s
'•S £ 2
|§ §
to TJ TJ
CD CD CO
O. m W
Q3 r**t i^i
, ' ' fl
to TJ TJ
CO CO CD
CO ^ CO
£
CO
to to
CD CO
CD CD
111
25 "o "o
J3 CO CO
E ^- ex
5-95
-------
re!
w
a
H
«
J
W
1
Pi
s
I
H
>H
Q
(H
W
&
w
CO
CN
I
in
I
tn
"co "9
~ co ,_
its
D >» ]>
5 E? 2
C CD
c
CD
cn~
CD t5 -7T
> co ,52.
its"
o j?^
o gS
CD
1 §
§'"2.
0 °
"o -H.
CD CO --
"53 .&. ^
co S; >•
CD E?€
CD ">•
C ^
CD
"53
•a CD 2
ill
O CD
%^^ ^— ^
O5 1^
CD CO
"0 T3
,~. ^—
"a "o
to to
en en
CD O
°. d
o •— *
E 1
o w
to "D
35. "en
en „
in CM
T- O
d d
i ;
0 O
"c "£
0 0
o o
CL 0-
o
^*"
_1
in CD
CD O
S- Hi
in -a-
CO •*
CO CO
^c? ~o
^— ^-^
t5 t5
to to
T5 TJ
en en
CD O
°. d
o *— '
E 1
o to
CO ~O
32. "en
in CM
T- O
d d
t i
i 1
0 0
"c "c
O 0
0 0
CL CL
O
in
co_
T—
CM
_J
^_^
o - — •
in o
in" *~~*
*° ^
o
in 2
CD
T3 T3
— £JZ-
t5 o
en to
TJ T3
en en
CD O
9 d
s. * —
to ~O
32. "en
en „
in CM
T- O
d d
' !
"o "o
"c "c
0 O
0 0
CL D-
O
O
CO
CM"
CO
XL
CO
CO
CM
CO
-C£-
J— .
t5
tn
32.
D)
o
S-
E
o
tn
"en
CO
CM
o
q
i
o
1—
-*—'
o
0
CL
CD
CD
^
-J
CD
in
CO
Q),
cr-
o
to
33
en
o
S-
0
CO
32.
co
CM
o
d
i
0
"c
o
0
CL
in
CO
co
LO
\f
_I
CD
r-
o
CO
,2,
5^
CO
L«
en
q
2.
E
o
CO
CM
q
d
i
i
0
"c
o
o
CL
in
CD
CD
IJ
hat energy impacts are reduced
-*-•
en
'-cS
o
'•a
to
CD
^3
negative vali
^ f
c
CD
CO
CD
Q.
CD
CO
CD
to
CD
•*-*
C
CD
CO
CX
to
CD
X)
E
'co"
X
Q.
0
•a
CD
a
to
to
CD
Q.
O
CM
T.
d
•t—
CM
X
CD
2
5
H^
to
CO
en
+-*
_CD
CD
JD
b
to
X
CO
o
o
o
d
II
en
"o
CO
CX
.i
en
CD
c
CD
CD
*-' X5
amoun
ie scru
-*-; ^=
CO CD
sz tn
•^ co
>,m
X5
d
£|
B 2 CL
CO ^ CO
H "^" C.
co -c co
"CD ^ %
"c co. ^
CO •— ' _D
1
O
o
§
2
X
i:
CM
co"
LU
X
co
o
o
q
II
to
t5
CO
ex
E
en
CD
CD
CL
CO
LU
"53
_CD
"o
o
0
T3
CO
CD
"53
to
CO
CQ
"o
s
CO
CL
CO
LU
X
s
'co
5
•+—
to
CO
en
'^•—
.—
CL
CO
LU
X
CD
O
O
CD,
+
CX
0
CD
to
CO
CD
Q.
b
tn
en
"to
"x
CD
"o
•epiacement
c
o
TJ
CD
CO
CO
T3
"co
"tn
CD
CD
1
tj
CO
CX
E
"D
CL
CO
LU
en
c
'«
"x
0
M —
O
CD
-a
CO
en
CX
— J
c
0
CD
to
CO
n
CD
"co
E
"co
CD
CD
o3
to
"o
CO
CX
E
CD
5-96
-------
TABLE 5-29a (METRIC). WASTEWATER IMPACTS FOR MODEL LIME KILNSa
Model
lime
kilns
LK-1
LK-2
LK-3
Control
option
,PM control (b)
PM control (b)
PM control (b)
Incremental
wastewater impacts,
million L/yr
(226)
(484)
(709)
(a) Metric equivalents in this table were converted
from the calculated English unit values given in
Table 5-29b. Refer to Table 5-29b for footnotes,
which include calculations.
(b) Impacts were estimated based on replacement of
the existing scrubber with a new ESP.
TABLE 5-29b (ENGLISH). WASTEWATER IMPACTS FOR MODEL LIME KILNS*
Model
lime
kilns
LK-1
LK-2
LK-3
Control
option
PM control (b)
PM control (b)
PM control (b)
Incremental
wastewater impacts,
million gal/yr
(60)
(128)
(187)
(a) Control level wastewater impacts are zero. Therefore,
incremental impacts are equal to baseline. Numbers
in parentheses represent negative values, indicating
that wastewater impacts are reduced by that amount.
Wastewater impacts = 4,500 Ib/ODTP x 0.9 ODTP/ADTP
x ADTP/d x 351 d/yr x gal/8.345 Ib
(b) Impacts were estimated based on replacement of
the existing scrubber with a new ESP.
5-97
-------
5.6 REFERENCES FOR CHAPTER 5
1. Memorandum from Holloway, T. and Nicholson, R., MRI, to
Telander, J., EPA/MICG. October9, 1996. Nationwide Costs,
Environmental Impacts, and Cost-Effectiveness of Regulatory
Alternatives for Kraft, Soda, Sulfite, and Semichemical
Combustion Sources.
2. Memorandum from Holloway, T., MRI, to the Project file.
June 14, 1996. Summary of PM and HAP Metals Data.
3 Telecon. Soltis, V., MRI, with Sanders, D.,_Andersen 2000,
Inc. August 11, 1993. Cost and efficiency information
about HC1 control for recovery furnaces.
4. Telecon. Soltis, V., MRI, with Bruno, J., AirPol, Inc.
July 2, 1993. Information about scrubbers used to control
HC1 emissions.
5 Someshwar, A. and J. Pinkerton. Wood Processing Industry.
In: Air Pollution Engineering Manual, Air & Waste Management
Association. Buonicore, A. and W. Davis (eds.). New York,
Van Norstrand Reinhold. 1992. pp. 835-849.
Memorandum from Soltis, V., MRI, to the project file.
April 3, 1995. Kraft and Soda Pulp Mill Combustion Sources
Data Base.
Memorandum from Holloway, T., MRI, to the project file.
May 3, 1995- Sulfur Dioxide Control by Recovery Furnace
Scrubbers.
Memorandum from Holloway, T., MRI, to the project file.
May 21, 1996. Secondary Emission Calculations.
OAQPS Control Cost Manual. 4th Edition. U. S.
Environmental Protection Agency. Research Triangle Park, NC
Publication,No. EPA-450/3-90-006. January 1990.
Letter and attachments from Black, C., Champion
International Corp., Courtland, AL, to Crowder, J., EPA/ISB.
June 10, 1993. Response to request for information on costs
for the BLO vent gas control system at the Champion facility
in Courtland, AL..
11 Kirby, M. Economic and Process Considerations in the Use of
Oxvaen for Black Liquor Oxidation. Union Carbide Canada
Limited. Ontario, Canada. (Presented at the 21st Pulp and
Paper ABCP Annual Meeting. Sao Paolo, Brazil.
November 21-25, 1988.) 10 p.
6.
7.
8.
9.
10,
5-98
-------
15.
16.
12. Stultz, S. and J. Kitto (eds.). Steam: Its Generation and
Use. 40th Edition. Babcock & Wilcox Co. New York. 1992.
Chapter 26, pp. 3-7.
e
13. Handbook: Control Technologies for Hazardous Air Pollutants.
U. S. Environmental Protection Agency. Cincinnati, OH.
Publication No. EPA-625/6-91-014. June 1991. Chapter 4.
14. Proposed Standards of Performance for Kraft Pulp Mills. In:
Standards Support and Environmental Impact Statement.
Volume 1. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/2-76-014-a.
September 1976. p. 2-9.
Mcllvane, R. Removal of Heavy Metals and Other Utility Air
Toxics. (Presented at the EPRI Hazardous Air Pollutant
Conference. 1993.).
Letter from Hurt, R., Radian Corp., to Holbrook, J. and
Bringman, L., Environmental Elements Corp. July 23, 1985.
Facsimile cover letter discussing utility and maintenance
costs for wet- vs. dry-bottom ESP designs.
17. Memorandum from Eddinger, J., EPA/ISB, to Durkee, K.,
EPA/ISB. July 8, 1983. Meeting minutes from the April 14,
1993 meeting between kraft pulp mill representatives, the
National Council of the Paper Industry for Air and Stream
Improvement, Inc., and the U. S. Environmental Protection
Agency. Discussion of problems mills have with TRS pickup
in wet-bottom ESP's using unoxidized black liquor.
Attachments.
18. Devore, J. Probability and Statistics for Engineering and
the Sciences. Brooks/Cole Publishing Co. Monterey,
California. 1982. pp. 15-17.
19. Someshwar, A. Compilation of "Air Toxic" Emission Data for
Boilers, Pulp Mills, and Bleach Plants. National Council of
the Paper Industry for Air and Stream Improvement, Inc. New
York. Technical Bulletin No. 650. June 1993. 128 p.
20. Garner, J., Jaako Poyry. Conversion to Low Odor Improves
Recovery Boiler Efficiency and Life. Pulp and Paper.
63(7):91-95. July 1989.
21. Green, R. and G. Hough (eds.). Chemical Recovery in the
Alkaline Pulping Process. 3rd Edition. Prepared by the
Alkaline Pulping Committee of the Pulp Manufacture Division.
Atlanta, GA, TAPPI Press, 1992. 196 p.
22. Telecon. Soltis, V., MRI, with Oscarsson, B., Gotaverken
Energy Systems, Inc. December 16, 1993. Concentrator steam
costs.
5-99
-------
23
24,
25
Letter and attachments from Sanders, D., Andersen 2000,
Inc to Soltis, V., MRI. November 9, 1993. Preliminary
proposal for three separate quench and packed-tower
absorption systems to control hydrochloric acid gas
emissions from recovery furnaces at kraft pulp and paper
mills.
Memorandum from Nicholson, . R. and Holloway, T., MRI, to
Slander, J., EPA/MICG. September 6, 1995. Trip Report for
S.D. Warren Co., Muskegon, MI.
Memorandum from Holloway, T.,_MRI, to the project file.
July 16, 1996. State of Washington PM Data for Kraft
Recovery Furnaces, Smelt Dissolving Tanks, and Lime Kilns.
5-100
-------
6.0 MODEL PROCESS UNIT CONTROL AND ENHANCED MONITORING COSTS
This chapter discusses the costs of controlling HAP
emissions from new and existing combustion sources at kraft and
soda pulp and paper mills. The costs of each control option
discussed in Chapter 4 are presented for the individual model
combustion process units. Also presented are the costs of the
enhanced monitoring options discussed in Chapter 4. The enhanced
monitoring options are methods of demonstrating continuous
compliance with the control options. The total nationwide costs
associated with each control option are presented in a separate
memorandum.1
Section 6.1 of this chapter presents the capital and annual
costs for each control option for model process units
representing recovery furnaces, BLO units, SDT's, and lime kilns.
Section 6.2 discusses the costs of the enhanced monitoring
options. Section 6.3 contains the references cited in this
chapter.
6.1 CONTROL OPTION COSTS
This section discusses the general costing approach used to
develop capital and annual costs for each control option and
presents the estimated capital and annual costs of each control
option as applied to the model process units. Section 6.1.1
describes the general costing approach. Section 6.1.2 provides
the capital and annual costs associated with each recovery
furnace control option and model recovery furnace.
Sections 6.1.3, 6.1.4, and 6.1.5 provide the capital and annual
costs applicable to the control options for model BLO units,
SDT's and lime kilns, respectively.
6-1
-------
6.1.1
Posting Approach
A number of assumptions were made in deriving the costs for
the control options. The year 1991 was used as the base year for
all costs. Capital costs were adjusted to 1991 dollars using the
Chemical Engineering Plant Cost Index or the Consumer Price
index, whichever was applicable.2'3 Operating and maintenance
personnel were assumed to work 3 shifts per day, 8 hours per
shift, for 365 d/yr, which is equivalent to 8,760 operating
hr/yr. All process units were assumed to operate 24 hr/d, for
351 d/yr, which is equivalent to 8,424 operating hr/yr. This
operating time accounts for 14 days of scheduled shutdown
annually for maintenance and repair. These 14 days could be
combined into a single 2-week annual shutdown period to provide
sufficient time for the modifications and upgrades both to
existing equipment and APCD's needed to comply with the control.
options. If the scheduled 2 -week shutdown did not provide
sufficient time for the modifications and upgrades, then pulp
production losses were calculated for the number of days beyond
the 2-week shutdown that were needed to finish the work,
The procedure used for estimating pulp production :
described in the following section. The general approaches
to develop capital and annual control costs are provided in
Sections 6.1.1.2 and 6.1.1.3, respectively,
losses is
used
6.1.1
.1 estimation of Pulp Production Losses. Pulp
production
losses were
calculated for each pulp type as a product
of the
pulp loss value and the total quantity of lost pulp
production.
Assuming
the kraft pulp mills achieve an earnings margin of
25 percent, the pulp loss value was estimated to be equal to
approximately 25 percent of the total market value. The 1989
market values of bleached and unbleached kraft pulp were
estimated to be approximately $712 and $423, respectively, per
ADMP ($646 and $384, respectively, per ADTP). The market values
.djusted to 1991 dollars using the Consumer Price Index.
estimate for each model recovery furnace the total
lost pulp production associated with the extended
were a
To
quantity
of
6-2
-------
recovery furnace downtime, the number of days of shutdown beyond
the scheduled 2-week shutdown were multiplied by the appropriate
model pulp production rate. Model pulp production rates were
determined for kraft bleached and unbleached pulp mills by
multiplying model BLS firing rates by the appropriate correlation
factor. A correlation factor of 1,800 kg BLS/ADMP (3,600 Ib
BLS/ADTP) was used for bleached pulp; a correlation factor of
1,500 kg BLS/ADMP (3,000 Ib BLS/ADTP) was used for unbleached
pulp.4
6.1.1-2 Development of Capital Cost-.a. The following
sources of cost information were used to develop capital costs
(i.e., total capital investment [TCI]) for the control options
being considered for kraft and soda pulp mill combustion sources:
1. Actual installed capital costs provided by individual
kraft and soda pulp and paper mills;
2. Cost equations and quotes supplied by ESP and scrubber
manufacturers;
3. Information supplied by recovery furnace manufacturers;
4. The U. S. EPA Handbook: Control Technologies for
Hazardous Air Pollutants;5 and
5. The U.S. EPA's OAQPS Control Cost Manual.6
Whenever possible, actual cost information from individual
mills was used to develop the capital cost estimates. Because
many combustion sources are subject to the NSPS for PM emissions
from kraft and soda pulp mills, actual mill-specific PM control
costs were already available for all of the control options
involving control of PM HAP's. For those control options where
actual costs were not available from individual mills or were
only available from one mill, or where mill-specific costs varied
widely, the EPA reference books (sources 4 and 5, above) were
used.
In most cases, cost algorithms were developed, relating
costs to the model process unit parameters (e.g., gas flow rate).
In a few cases where a direct relationship between the capital
cost and the model process unit was not apparent, the "six-tenths
rule" was used to extrapolate costs from one model to another.
6-3
-------
The "six- tenths rule" assumes a direct relationship between
capital cost and capacity taken to the six- tenths power (i.e.,
C /C2 - (Qi/Qa)0-6' where c and Q are caPital cost and caPacity
parameter, respectively) . The capital costs were extrapolated
using either gas flow rate or BLS firing rate as the capacity
parameter.
6.1.1.3 Dftvelotaneni-- of Annual Costs. Incremental total
annual costs (ITAC) were derived using the annual cost model
described in the OAQPS Control Cost Manual.6 Incremental total
annual costs refer to the incremental increase of total annual
costs (TAG) over current operation. Total annual costs include
both direct annual costs (DAC) and indirect annual costs (IAC) .
The cost components that comprise the DAC and IAC are discussed
in the following sections.
6.1.1.3.1 nirect annual costs. The DAC include operating
labor costs, maintenance labor and material costs, utility costs,
and wastewater treatment costs.6 Operating and maintenance labor
costs were calculated as a product of the number of working hr/d
to perform the required task, the number of operating d/yr, and
the hourly wage. Operating and maintenance labor costs were
calculated assuming 365 operating d/yr. The hourly wage and
number of working hr/d vary with each control option and are
based on information from the U. S. EPA' s Handbook: Control
Technologies for Hazardous Air Pollutants and the OAQPS Control
Cost Manual.5'6 The maintenance hourly wage is equal to
approximately 1.5 times the operating hourly wage. The
supervisory labor cost is approximately 15 percent of the
operating labor cost.. With the exception of ESP's, the
maintenance materials cost is estimated as approximately
100 percent of the maintenance labor cost.6 The maintenance
materials cost for ESP's is estimated as 1 percent of the flange-
to-flange purchased equipment cost (PEC) for ESP's, and the PEC
is estimated as 0.6 times the TCI.
Utility costs were broken down into electricity costs and
water costs. Electricity costs were calculated as a product of
the electricity unit cost and the electricity requirement. The
6-4
-------
electricity unit cost was assumed to be $0.06/kWh.5 Electricity
requirements were divided into fan, pump, and operating
electricity requirements and were calculated assuming
8,424 operating hr/yr.
The fan electricity requirement (applicable to ESP's and
scrubbers) is equal to a numerical factor (0.00018) times the
product of the gas flow rate, pressure drop, and operating
hr/yr.6 The gas flow rate varies with each model process unit.
The pressure drop is based on information from mills.4 Although
the pressure drop is not the sole parameter that determines PM
collection efficiency for lime kiln scrubbers, for the purposes
of estimating costs, the pressure drop was used as an indicator
of PM collection efficiency for lime kiln scrubbers. Note: A
different scrubber design, rather than a higher pressure drop,
was used to improve the PM collection efficiency for SDT
scrubbers.
The gas flow rate and pressure drop do not change for ESP's
relative to current operation when the ESP's are upgraded or
replaced to improve PM collection. Therefore, the fan
electricity requirements for ESP's do not change relative to
current operation. However, if a scrubber is added after an ESP,
the gas flow rate would be reduced, thereby reducing the fan
electricity requirements for the ESP.
The pump electricity requirement (applicable to packed-bed
scrubbers) is equal to a numerical factor (0.000188) times the
product of the liquid flow rate, amount of head pressure, and
operating hr/yr divided by the pump-motor efficiency.6 The
liquid flow rate varies with each model process unit. A head
pressure of 18 m (60 ft) and a pump-motor efficiency of
70 percent were assumed.
The operating electricity requirement (applicable to ESP's)
is equal to a numerical factor (0.00194) times the product of the
ESP plate area and operating hr/yr.6 The ESP plate area is
calculated as a product of the exhaust gas flow rate and the ESP
SCA. The gas flow rate varies with each process unit, and the
SCA is based on information from mills.4 Although the SCA is not
6-5
-------
the sole parameter that determines PM collection efficiency for
ESP's, for the purposes of estimating costs, the SCA was used as
an indicator of PM collection efficiency. The cost analysis does
not consider increasing the SCA to account for single-chamber ESP
operation at reduced gas flow during maintenance situations.
Instead, the regulation will allow for PM emission excursions
during maintenance situations as part of the "Startup,^Shutdown,
Malfunction Plan" described in the General Provisions.
Water costs were calculated as a product of the water unit
cost and the water requirement. The water unit cost was assumed
to be $0.05/m3 ($0.20/1,000 gal).5 The water requirement is
equal to a numerical factor (0.060) times the product of the gas
flow rate and operating hr/yr.6 The gas flow rate varies with
each model process unit.
For some control options (e.g., PM controls that include
upgrading or replacing an existing recovery furnace ESP or
replacing an existing SDT scrubber with a new scrubber), the
labor and maintenance costs are not expected to increase
significantly, and, therefore, the ITAC.is represented by the IAC
plus the difference in electricity costs before and after
implementation of the control options.
6.1.1.3.2 indirect annual costs. The IAC include overhead
costs, administrative charges, property taxes, insurance costs,-
and capital recovery costs. Overhead costs are
approximately
60
percent of all labor and maintenance material costs
Overhead costs are not applicable when labor and maintenance
costs
do not increase significantly. Administrative, insurance,
and property tax costs are approximately 4 percent of the TCI
Capital recovery costs are equal to a capital recovery factor
(CRF) multiplied by the TCI. The CRF is estimated using the
equation CRF - [id + i)n]/td + Dn - U , where i = interest
rate (assumed to be 7 percent) and n = equipment life of the
device being installed or modified.
6-6
-------
6.1.2 Recovery Furnace Control Options
The following sections discuss the model costs of four
control options evaluated for recovery furnaces. These control
options include PM controls (Section 6.1.2.1), wet to dry ESP
system conversion (Section 6.1.2.2), conversion of a DCE recovery
furnace system to an NDCE recovery furnace (Section 6.1.2.3), and
addition of a packed-bed scrubber (Section 6.1.2.4).
6.1.2.1 PM Controls. Two PM control options were evaluated
for model NDCE recovery furnaces RF-1 through RF-6 and model DCE
recovery furnaces RF-7 through RF-9. The control options apply
to new and existing recovery furnaces and are described below.
One PM control option that was evaluated would reduce PM
emissions from existing recovery furnaces to the NSPS level of
0.10 g/dscm (0.044 gr/dscf). The PM control option evaluated
would involve (1) replacing the recovery furnace ESP or
(2) upgrading the recovery furnace ESP. The control equipment
selected by a particular mill would be site-specific.
A second PM control option that was evaluated would reduce
PM emissions from existing recovery furnaces to 0.034 g/dscm
(0.015 gr/dscf). This more stringent PM control option would
involve replacing or upgrading the recovery furnace ESP and
adding a packed-bed scrubber. The second PM control option also
applies to new recovery furnaces; the option could be used to
evaluate the cost to new sources subject to a more stringent
standard (0.034 g/dscm [0.015 gr/dscf]) than the current NSPS.
The PM control costs for existing recovery furnaces with
baseline PM emissions above the NSPS level were estimated for
model NDCE recovery furnaces RF-la through RF-6a and model DCE
recovery furnaces RF-7a through RF-9a. The PM control costs for
new and existing recovery furnaces with baseline PM emissions at
or below the NSPS level were estimated for model NDCE recovery
furnaces RF-lb through RF-6b and model DCE recovery furnaces
RF-7b through RF-9b.
The PM control costs for model NDCE recovery furnaces with
dry ESP systems (i.e., RF-1 through RF-3) are assumed to be
identical to the PM control costs for model NDCE recovery
6-7
-------
furnaces with wet ESP systems (i.e., RF-4 through RF-6) because
PM emissions are not affected by whether or not black liquor is
used in the ESP bottom or PM return system. The capital and
annual costs to replace or upgrade ESP's are presented for the
model NDCE and DCE recovery furnaces in the following sections.
The capital and annual costs to install a new packed-bed scrubber
are presented in Section 6.1.2.4.
Based on the costs presented below, it is more expensive to
replace an existing ESP than to upgrade: it. However, a
replacement may be necessary, depending on the age and condition
of the existing ESP, in order to effectively control PM
emissions. Site-specific conditions often dictate the cost-
effectiveness of replacing or upgrading an ESP.
6.1.2.1.1 ESP replacement; capital costs. The ESP
replacement costs for models RF-la through RF-9a were calculated
based on recent ESP replacement costs provided by individual pulp
and paper mills (i.e., costs for ESP's replaced during or after
1989).4 The cost to dispose of the existing ESP was assumed to .
be included in the new ESP costs provided by the individual
mills. New ESP costs average $420/m2 ($39/ft2) of ESP plate area
for NDCE and DCE recovery furnaces.4 The ESP plate area for the
new ESP was derived from the model exhaust gas flow rates for
RF-la through RF-9a and the SCA for the replacement ESP.
For NDCE recovery furnaces, an SCA of approximately
100 m2/(m3/sec) (530 ft2/!,000 acfm) was assumed based on ESP SCA
information from NDCE recovery furnaces subject to the NSPS
(i.e., furnaces installed or replaced during or after 1977 and
required to have outlet PM emissions of 0.10 g/dscm
[0.044 gr/dscf] or lower).4 Because dry-bottom ESP's are used to
control PM emissions from approximately 80 percent of NDCE
recovery furnaces, it was assumed that all replacement ESP's for
NDCE recovery furnaces would be of the dry-bottom design. The
ESP replacement costs for the three model NDCE recovery furnaces
are presented in Table 6-1. The total capital costs range from
$4.12 million to $10.7 million for RF-la through RF-6a.
6-8
-------
For DCE recovery furnaces, an SCA of approximately
90 m2/{m3/sec) (430 ft2/!,000 acfm) was assumed based on ESP SCA
information from DCE recovery furnaces subject to the NSPS (i.e.,
furnaces installed or replaced during or after 1977 and required
to have outlet PM emissions of 0.10 g/dscm [0.044 gr/dscf] or
lower).4 Because wet-bottom ESP's are used to control PM
emissions from approximately 90 percent of DCE recovery furnaces,
it was assumed that all new replacement ESP's for DCE recovery
furnaces would be of the wet-bottom design.8 The ESP replacement
costs for the three model DCE recovery furnaces are presented in
Table 6-1. The TCI costs range from $2.01 million to
$6.03 million for RF-7a through RF-9a.
Installing a new fan and stack typically comprises
approximately 20 percent of the total capital costs of installing
a new ESP.9 Because the fan and stack usually do not need to be
replaced when an ESP is replaced, the ESP replacement costs
stated above are estimated to be 80 percent of the cost of a
completely new ESP. The costs to replace the fan and stack due
to the addition of the packed-bed scrubber are presented in
Section 6.1.2.4.2.
In some cases, installation of a replacement ESP can be
achieved using a "roll-in" procedure, in which the ESP is erected
adjacent to its final location and then rolled into position
using roller assemblies or rubber-tired dollies. The roll-in
technique has been used for a number of years on a wide range of
equipment sizes.9 However, the roll-in technique probably cannot
be used for the majority of recovery furnaces due to the location
of the ESP (e.g., an elevated position relative to the recovery
furnace). Because the ESP replacement costs were based on actual
ESP replacement costs provided by individual pulp and paper
mills, it was assumed that they were based on recovery furnaces
that cannot do ESP roll-ins. Therefore, no contingency factor is
needed to adjust ESP replacement costs to account for the
furnaces being unable to do ESP roll-ins. However, without the
ESP roll-in, an ESP replacement probably could not be completed
within the scheduled 2-week shutdown for maintenance. Because no
6-9
-------
information is available on the time needed to complete an ESP
replacement, the maximum time estimated to complete an ESP
upgrade, as described in Section 6.1.2.1.3, was used. The
estimated maximum time is 30 days. Therefore, pulp production
losses for replacing the ESP were calculated for that period of
time minus the 2-week (14-day) scheduled mill shutdown. Pulp
production losses for the 16-day period are presented in
Table 6-2.
For mills producing bleached pulp, the pulp production
losses range from $1.19 million to $3.12 million for RF-la
through RF-6a and $710,000 to $2.13 million for RF-7a through
RF-9a. For mills producing unbleached pulp, the pulp production
losses range from $844,000 to $2.19 million for RF-la through
RF-6a and $506,000 to $1.52 million for RF-7a through RF-9a. The
unbleached pulp production losses are lower for each model
recovery furnace because unbleached pulp has a lower market value
than bleached pulp.
6.1.2.1.2 ESP replacement: incremental annual costs.
Labor and maintenance requirements and costs are assumed to be
unchanged when the ESP is replaced. However, because the ESP
replacement option includes an increase in ESP plate area, .
electricity costs are increased. The increase in ESP electricity
costs resulting from the ESP replacement is based on an increase
in the SCA.
For NDCE and DCE recovery furnaces, the baseline SCA values
are approximately 90 m2/(m3/sec) (430 ft2/l,000 acfm) and
70 m2/(m3/sec) (330 ft2/!,000 acfm), respectively. These SCA
values are the average values common to ESP's on NDCE and DCE
recovery furnaces installed prior to 1977 (i.e., installed prior
to the NSPS and not subject to the NSPS PM standard of
0.10 g/dscm [0.044 gr/dscf]) that also have PM emissions greater
than 0.10 g/dscm (0.044 gr/dscf).4 As a result of the ESP
replacement, the baseline SCA would be increased to a value of
approximately 100 m2/(m3/sec) (530 ft2/!,000 acfm) for NDCE
recovery furnaces and 90 m2/(m3/sec) (430 ft2/l,000 acfm) for DCE
recovery furnaces. These SCA values are the average values
6-10
-------
common to ESP's on NDCE and DCE recovery furnaces installed
during or after 1977 (i.e., installed after the NSPS and subject
to the NSPS standard of 0.10 g/dscm [0.044 gr/dscf]) that also
have PM emissions less than or equal to 0.10 g/dscm
(0.044 gr/dscf) but greater than 0.034 g/dscm (0.015 gr/dscf)
(the more stringent PM control level).4
The incremental annual costs for ESP replacements include
the difference in electricity costs plus the indirect costs
affected by the TCI, which include the administrative, property
tax, and insurance costs plus the capital recovery cost. The
lifetime of dry-bottom ESP's typically ranges from 12 to
15 years.9 Therefore, an average 13.5-year life span was assumed
when calculating the capital recovery cost for dry-bottom ESP's
installed on NDCE recovery furnaces. The lifetime of wet-bottom
ESP's is typically 10 years.9 Therefore, an average 10-year life
span was assumed when calculating the capital recovery cost for
wet-bottom ESP's installed on DCE recovery furnaces.
The ITAC for the model NDCE and DCE recovery furnaces,
excluding annualized pulp production losses (i.e., capital
recovery costs for the pulp production losses), are presented in
Table 6-1. -The incremental annual costs range from $666,000/yr
to $1.73 million/yr for RF-la through RF-6a and $378,000/yr to
$1.14 million/yr for RF-7a through RF-9a. The ITAC for the model
NDCE and DCE recovery furnaces, including annualized bleached and
unbleached pulp production losses, are presented in Table 6-2.
The incremental annual costs, including annualized bleached pulp
production losses, range from $805,000/yr to $2.09 million/yr for
RF-la through RF-6a and $764,000/yr through $1.99 million/yr for
RF-7a through RF-9a. The incremental annual costs, including
annualized unbleached pulp production losses, range from
$764,000/yr to $1.99 million/yr for RF-la through RF-9a and
$451,000/yr to $1.36 million/yr for RF-7a through RF-9a.
6.1.2.1.3 ESP upgrade (to 0.10 g/dscm TO.044 gr/dscf1):
capital costs. The ESP upgrade costs for existing NDCE and DCE
recovery furnaces are based on May 1993 ESP upgrade costs
supplied by an ESP manufacturer.9 The ESP manufacturer supplied
6-11
-------
costs for a model NDCE recovery furnace with an exhaust gas flow
rate of 109 m3/sec (230,000 acfm) and a model DCE recovery
furnace with an exhaust gas flow rate of 160 m3/sec
(340,000 acfm).9 According to the ESP manufacturer, the proposed
upgrade would result in an ESP size adequate to meet an outlet PM
level of 0.10 g/dscm (0.044 gr/dscf).9
The ESP manufacturer supplied ESP upgrade costs for two
different ESP upgrade schedules (Schedules 1 and 2} . Schedule 1
is a 21-day, 7-day/week ESP upgrade schedule, including about
14 days of recovery furnace outage and about 3 or 4 days of
partial load on both sides of the outage.9 Three days of partial
load before and after the outage were assumed, for a total of
6 days of partial load. Assuming a 50 percent load for 6 days,
there would be 3 days of downtime during the period of partial
load. Added to the 14 days of recovery furnace outage, there is
a total downtime of 17 days, which is 3 days of downtime beyond
the annual 2-week shutdown for maintenance. Schedule 2 is a
30-day, 6-day/week upgrade schedule and has no periods of partial
load. There would be 30 days of recovery furnace outage.
Schedule 2 would have 16 days of downtime beyond the annual
2-week shutdown. For the model NDCE recovery furnace, the cost
for Schedule 1 is $1,292,000, and the cost for Schedule 2 is
$1,259,000.9 For the model DCE recovery furnace, the cost for
Schedule 1 is $1,504,750, and the cost for Schedule 2 is
$1,466,500.9
The nature of the ESP upgrades is identical for Schedules 1
and 2. A brief summary of the required modifications is
presented in Table 6-3. The ESP upgrades for these schedules
include replacing the weighted wire design with a rigid electrode
design.9 Rigid electrode ESP's generally operate at a higher
voltage than weighted wire ESP's and, in some cases, may require
replacement of the transformer and other controls. However, the
ESP upgrade cost estimate does not include those costs that vary
widely based on site-specific conditions, such as the need for
transformer replacement.
6-12
-------
The only difference between the two ESP upgrade schedules is
that Schedule 2 allows a longer downtime because major
modifications to the furnace would be performed at the same time.
The costs of the recovery furnace modification are not included
because they are not associated with the ESP upgrade. The ESP
upgrade work for Schedule 2 would be extended over the 30-day
shutdown period.
The ESP upgrade costs were adjusted to 1991 dollars and then
scaled using the six-tenths power rule described in
Section 6.1.1.2 to derive costs for the model NDCE and DCE
recovery furnaces. The model ESP upgrade costs, without pulp
production losses, for Schedules 1 and 2 are presented in Tables
6-4 and 6-5, respectively. The ESP upgrade capital costs for
Schedule 1 range from $1.20 million to $2.12 million for RF-la
through RF-6a and $811,000 to $1.57 million for RF-7a through
RF-9a. The ESP upgrade capital costs for Schedule 2 range from
$1.16 million to $2.07 million for RF-la through RF-6a and
$791,000 to $1.53 million for RF-7a through RF-9a.
The pulp production losses associated with shutting down the
mill to upgrade the ESP were calculated using the method
described in Section 6.1.1.1. The ESP upgrade costs that include
bleached and unbleached pulp production losses are presented in
Tables 6-6 and 6-7, respectively. The pulp production losses for
Schedules 1 and 2 are presented for the model recovery furnaces
in Tables 6-6 and 6-7, respectively.
The bleached pulp production losses for Schedule 1 range
from $224,000 to $585,000 for RF-la through RF-6a and $133,000 to
$399,000 for RF-7a through RF-9a. The unbleached pulp production
losses for Schedule 1 range from $158,000 to $411,000 for RF-la
through EF-6a and $94,900 to $285,000 for RF-7a through RF-9a.
The pulp production losses for mills producing unbleached pulp
are lower than for mills producing bleached pulp because
unbleached pulp has a lower market value than bleached pulp.
Because Schedule 2 is longer than Schedule 1, the pulp
production losses are higher for Schedule 2 than for Schedule 1.
The bleached pulp production losses for Schedule 2 range from
6-13
-------
$1.19 million to $3.12 million for RF-la through RF-6a and
$710,000 to $2.13 million for RF-7a through RF-9a. The
unbleached pulp production losses for Schedule 2 range from
$844,000 to $2.19 million for RF-la through RF-6a and $506,000 to
$1.52 million for RF-7a through RF-9a.
6.1.2.1.4 ESP upgrade (to 0-10 a/dsnm FO.04* gr/dsef1):
incremental annual costs. Labor and maintenance costs are
assumed to be unchanged when the ESP is upgraded to achieve.a PM
level of 0.10 g/dscm (0.044 gr/dscf). Therefore, the incremental
annual costs for the ESP upgrade include only the electricity
costs and the TCI-based indirect annual costs, which include
administrative, property tax, insurance, and capital recovery
costs. The capital recovery cost for NDCE recovery furnace ESP's
is based on an average 13.5-year life span for dry-bottom ESP's.
The capital recovery cost for DCE recovery furnace ESP's is based
on an average 10-year life span for wet-bottom ESP's.9 Although
the ESP plate area is not increased with an ESP upgrade, there
will be additional electricity costs associated with the new ESP
design. The actual increase in electricity costs with an ESP
upgrade is not currently known. Therefore, it was assumed that
the incremental increase in electricity costs for an upgraded ESP
would be the same as for a replacement ESP. Because the
electricity cost increases for the replacement ESP in
Section 6.1.2.1.2 were based on an increase in SCA from baseline
to control levels, the electricity cost increases for the
upgraded ESP were calculated in the same manner.4 The ITAC for
the model recovery furnaces are presented in Tables 6-4 through
6-7.
Table 6-4 presents the ITAC for the Schedule 1 ESP upgrade
for each model recovery furnace, without accounting for
annualized pulp production losses. As shown in the table, the
ITAC for the Schedule 1 ESP upgrade, without annualized pulp
production losses, range from $207,000/yr to $383,000/yr for
RF-la through RF-6a and $159,000/yr to $322,000/yr for RF-7a
through RF-9a. Table 6-5 presents the ITAC for the Schedule 2
ESP upgrade for each model recovery furnace, without accounting
6-14
-------
for annualized pulp production losses. As shown in the table,
the ITAC for the Schedule 2 ESP upgrade, without annualized pulp
production losses, range from $202,000/yr to $375,000/yr for
RF-la through RF-6a and $156,000/yr to $314,000/yr for RF-7a
through RF-9a.
Table 6-6 presents the ITAC for the Schedule 1 ESP upgrade
for each model recovery furnace and includes estimated production
losses associated with bleached and unbleached pulp. As shown in
the table, the ITAC for the Schedule 1 ESP upgrade, including
bleached pulp production losses, range from $233,000/yr to
$452,000/yr for RF-la through RF-6a and $178,000/yr to
$379,000/yr for RF-7a through RF-9a. The ITAC for the Schedule 1
ESP upgrade, including unbleached pulp production losses, range
from $226,000/yr to $431,000/yr for RF-la through RF-6a and
$173,000/yr to $363,000/yr for RF-7a through RF-9a.
Table 6-7 presents the ITAC for the Schedule 2 ESP upgrade
for each model recovery furnace and includes estimated production
losses associated with bleached and unbleached pulp. As shown in
the table, the ITAC for the Schedule 2 ESP upgrade, including
bleached pulp production losses, range from $341,000/yr to
$740,000/yr for RF-la through RF-6a and $257,000/yr to
$617,000/yr for RF-7a through RF-9a. The ITAC for the Schedule 2
ESP upgrade, including unbleached pulp production losses, range
from $300,000/yr to $631,000/yr for RF-la through RF-6a and
$228,000/yr to $530,000/yr for RF-7a through RF-9a.
6.1.2.1.5 ESP upgrade (to 0.034 a/dscm TO. 015 crr/dscf 1 ) :
capital costs. Control costs have been determined for those new
and existing recovery furnaces controlling PM emissions from the
NSPS level of 0.10 g/dscm (0.044 gr/dscf) to 0.034 g/dscm
(0.015 gr/dscf). These costs would include an ESP upgrade cost
and a packed-bed scrubber cost. Because no actual ESP upgrade
costs were available for controlling PM emissions to 0.034 g/dscm
(0.015 gr/dscf), the upgrade cost for the recovery furnace ESP
was instead based on the incremental cost difference between an
ESP capable of achieving a PM level less than or equal to
0.10 g/dscm (0.044 gr/dscf) but greater than 0.034 g/dscm
6-15
-------
(0.015 gr/dscf) and an ESP capable of achieving a PM level less
than or equal to 0/034 g/dscm (0.015 gr/dscf). This section
presents the ESP upgrade costs; the packed-bed scrubber costs are
presented in Section 6.1.2.4.
The ESP upgrade capital costs were based on recent ESP costs
provided by individual pulp and paper mills (i.e., costs for
ESP's installed or replaced during or after 1989).4 The ESP
costs average $420/ra2 ($39/ft2) of ESP plate area for recovery
furnaces.4 To determine the ESP upgrade cost, this cost per ESP
plate area was multiplied by the increase in ESP plate area
assumed to reduce PM emissions from 0.10 g/dscm (0.044 gr/dscf)
to 0.034 g/dscm (0.015 gr/dscf).
The ESP plate area for NDCE recovery furnace ESP's achieving
a PM level less than or equal to 0.10 g/dscm (0.044 gr/dscf) but
greater than 0.034 g/dscm (0.015 gr/dscf) is based on an average
SCA of approximately 100 m2/(m3/sec) (530 ft2/l,000 acfm), as
discussed in Section 6.1.2.1.2. The ESP plate area for DCE
recovery furnace ESP's achieving a PM level less than or equal to
0.10 g/dscm (0.044 gr/dscf) but greater than 0.034 g/dscm
(0.015 gr/dscf) is based on an average SCA of approximately
90 m2/(m3/sec) (430 ft2/!,000 acfm), as discussed in
Section 6.I.2.I.2.4 The ESP plate area for recovery furnace
ESP's achieving a PM level less than or equal to 0.034 g/dscm
(0.015 gr/dscf) is based on an SCA of approximately
120 m2/(m3/sec) (620 ft2/!,000 acfm). This is the SCA value for
an ESP achieving a PM emission level of 0.034 g/dscm
(0.015 gr/dscf) on a long-term basis.4'1
The capital cost attributable to the control option to
install new recovery furnace ESP's capable of achieving a PM
level of 0.034 g/dscm (0.015 gr/dscf) would not be the cost of a
new ESP, but only that portion associated with controlling PM
emissions from 0.10 g/dscm (0.044 gr/dscf) to 0.034 g/dscm
(0.015 gr/dscf), which is the same as the cost to upgrade
existing recovery furnace ESP's to achieve the same PM level of
0.034 g/dscm (0.015 gr/dscf).
6-16
-------
The capital costs of the ESP upgrade used to achieve the
0.034 g/dscm (0.015 gr/dscf) PM level are lower than the capital
costs of the ESP upgrade used to achieve the 0.10 g/dscm
(0.044 gr/dscf) NSPS PM level discussed in Section 6.1.2.1.3.
The capital costs for this more stringent ESP upgrade control
option, excluding pulp production losses, are presented in
Table 6-8 and range from $644,000 to $1.67 million for RF-lb
through RF-6b and $387,000 to $1.16 million for RF-7b through
RF-9b.
There are no pulp production losses for new recovery
furnaces to install an upgraded ESP because the ESP is upgraded
prior to installation. No information is currently available for
existing recovery furnaces on the amount of time required to
complete an ESP upgrade that would allow PM control to
0.034 g/dscm (0.015 gr/dscf). However, it was assumed that the
pulp production losses could be as high as those for Schedule 1
of the ESP upgrade control option discussed in Section 6.1.2.1.3.
The pulp production losses are presented in Table 6-9.
6.1.2.1.6 ESP upgrade (to 0.034 cr/dscm TO.015 ar/dscf 1 ) :
incremental annual costs. Labor and maintenance costs are
assumed to be unchanged when the ESP is upgraded to achieve a PM
level of 0.034 g/dscm (0.015 gr/dscf). Therefore, the
incremental annual costs for the ESP upgrade include only the
electricity costs and TCI-based indirect annual costs, which
include administrative, property tax, insurance, and capital
recovery costs. The capital recovery cost for NDCE recovery
furnace ESP's is based on an average 13.5-year life span for
dry-bottom ESP's. The capital recovery cost for DCE recovery
furnace ESP's is based on an average 10-year life span for
wet-bottom ESP's. The PM control electricity costs would include
both ESP upgrade and packed-bed scrubber electricity costs. The
ESP upgrade electricity costs are presented in this section;
Section 6.1.2.4 presents the packed-bed scrubber electricity
costs. The ESP upgrade electricity costs are based on an
increase in SCA from approximately 100 m2/(m3/sec) (530 ft2/
1,000 acfm) for NDCE recovery furnaces and 90 m2/(m3/sec)
6-17
-------
(430 ft2/!,000 acfm) for DCE recovery furnaces to approximately
120 m2/(m3/sec) (620 ft2/!,000 acfm) for all recovery furnaces.4
The electricity costs presented in this section are based on
gas flow rates and pressure drops in the absence of a packed-bed
scrubber. If a packed-bed scrubber was added after the ESP, the
fan electricity costs for the ESP would change slightly from
those presented in this section. However, because the total
electricity costs are only a small fraction (approximately
10 percent, of the ITAC, a slight change in fan electricity costs
would have a negligible effect on the ITAC.
The ITAC for the model recovery furnaces, excluding
annualized pulp production losses, are presented in Table 6-8 and
range from $117,000/yr to $304,000/yr for RF-lb through RF-6b and
$80,300/yr to $241,000/yr for RF-7b through RF-9b. The ITAC for
the model recovery furnaces, including annualized pulp production
losses, are presented in Table 6-9. The incremental annual
costs, including annualized bleached pulp production losses,
range from $143,000/yr to $373,000/yr for RF-lb through RF-6b and
$99,000/yr to $298,000/yr for RF-7b through RF-9b. The
incremental annual costs, including annualized unbleached pulp
production losses, range from $l36,000/yr to $352,000/yr for
RF-lb through RF-6b and $94,000/yr to $282,000/yr for RF-7b
through RF-9b.
6.1.2.2 Wgfc to Dry ESP System Conversion. Two control
options were evaluated for reducing emissions of gaseous organic
HAP's such as methanol from existing NDCE recovery furnaces.
These control options are (1) converting an ESP system that uses
black liquor or HAP-contaminated process water in the ESP bottom
or PM return system (referred to as a wet ESP system) to an ESP
system that uses "clean" water (i.e., water uncontaminated with
methanol and other gaseous organic HAP's) in the ESP bottom or PM
return system; and (2) converting a wet ESP system to a dry-
bottom ESP with a dry PM return system (referred to as a dry ESP
system). With these two control options, the potential stripping
of methanol and other gaseous organic HAP's from the black liquor
6-18
-------
or contaminated process water in the ESP system would be
eliminated.
Only the costs for the second control option, converting
from a wet to a dry ESP system, were evaluated. This decision
was based on (1) the uncertainty associated with the available
cost estimates for converting to an ESP system that uses "clean"
water in the ESP bottom or PM return system; and (2) the fact
that very few mills use water in the ESP system.
A cost estimate is available for converting a wet-bottom ESP
that uses black liquor in the ESP bottom to one that uses "clean"
water in the ESP bottom. This cost estimate is significantly
lower than the cost estimates available for converting to a
dry-bottom ESP, but the accuracy of this cost estimate is
questionable. According to a 1985 EPA estimate for a 900 ADMP/d
(1,000 ADTP/d) mill, the capital cost to convert a wet-bottom ESP
to one that uses water in the ESP bottom is $154,000. The annual
cost is $67,000/yr.11 These costs are lower than the wet- to
dry-bottom conversion costs presented below. Several pulp and
paper industry representatives commenting on the estimate
indicated that the costs to evaporate the added water should be
higher. They also noted that mills may experience a loss of
production to evaporate the extra water if excess capacity was
not available in the evaporators.12
The wet to dry ESP system conversion control option applies
to model NDCE recovery furnaces RF-4 through RF-6, which
represent existing NDCE recovery furnaces with wet ESP systems.
These models represent existing NDCE recovery furnaces only,
because no wet ESP systems are expected to be installed on new
NDCE recovery furnaces.
The model costs for the wet to dry ESP system conversion
control option are based on costs to convert NDCE recovery
furnace wet-bottom ESP's to dry-bottom ESP's. For the purposes
of this cost analysis, these wet- to dry-bottom ESP conversion
costs are assumed to apply also to those NDCE recovery furnaces
with dry-bottom ESP's and wet PM return systems. The ESP
conversion costs may be lower than those presented if an ESP
6-19
-------
upgrade to improve PM collection is also performed at the same
time. However, no information is currently available on the
extent of the cost reduction. The capital and annual costs for
the wet to dry ESP system conversion control option are presented
in the following sections for existing model NDCE recovery
furnaces.
6.1.2.2.1 Capital costs. The wet to dry ESP system
conversion capital costs are based on 1993 conversion costs from
an ESP manufacturer.13 These costs were adjusted to 1991 dollars
and then scaled, using the six-tenths power rule, to derive costs
for the three model NDCE recovery furnaces. The ESP model used
by the manufacturer to develop the conversion costs was stated as
being three fields in length and two chambers in width, each
6.1 m (20 ft) wide. The cost to remove the existing agitator
paddles and liquor piping and install a perpendicular drag
scraper system, shallow fallout hoppers, drag chain conveyors,
and rotary valves was estimated to be $560,000 for the material
and $285,000 for installation. The cost estimate does not
include the cost associated with (1) any removal of asbestos, if
applicable; (2) any piping beyond the rotary valves; or (3) any
equipment beyond the rotary valves, such as an ash mixing tank
with associated instrumentation. The ESP conversion costs from
the ESP manufacturer are based on working two 10-hour shifts for
about 10 days and converting both ESP chambers simultaneously.
Therefore, no downtime would be necessary beyond the annual
2-week shutdown, which means no pulp production losses would need
to be included in the model ESP conversion costs presented below.
The recovery furnace size used by the ESP manufacturer in
calculating the wet-bottom ESP conversion costs was stated to be
about 600 to 900 ADMP/d (700 to 1,000 ADTP/d)13. A size of
800 ADMP/d (900 ADTP/d) (the size for the mid-size model NDCE
recovery furnace RF-5) was used to scale the costs. The capital
costs for the three model NDCE recovery furnaces are presented in
Table 6-10. The wet to dry ESP system conversion costs range
from $596,000 to $1.06 million for RF-4 through RF-6.
6-20
-------
6.1.2.2.2 Incremental annual costs. Direct annual costs
are not assumed to increase as a result of the wet to dry ESP
system conversion. Although there may be some extra maintenance
costs, they are expected to be small compared to the increases in
capital recovery and other indirect costs. Past dry-bottom ESP
designs were associated with higher maintenance costs. Changes
in designs have eliminated many of those problems.14 The costs
from the ESP manufacturer used to develop the model costs are
based on the modern design. Furthermore, because wet-bottom ESP
designs are associated with greater corrosion, switching to dry-
bottom ESP designs results in a longer life span for the ESP.
Utility costs (i.e., electricity) also do not change
significantly because of an equal trade-off in horsepower
requirements between the wet and dry ESP system designs.14
Based on these assumptions, the ITAC for wet to dry ESP
system conversions should only include those indirect annual
costs affected by the TCI (i.e., administrative, property tax,
insurance, and capital recovery costs). Similar to the ESP
replacement costs, the capital recovery cost is also based on an
average 13.5-year life for dry-bottom ESP's operating on NDCE
recovery furnaces. The ITAC for the model NDCE recovery furnaces
are presented in Table 6-10 and range from $93,500/yr to
$166,000/yr for RF-4 through RF-6.
6.1.2.3 Conversion of a DCE Recovery Furnace System to an
NDCE Recovery Furnace. Converting a DCE recovery furnace system
to an NDCE recovery furnace (or "low-odor conversion") was
evaluated as a control option for reducing gaseous organic HAP
emissions from DCE recovery furnace systems. The conversion of a
DCE recovery furnace system to an NDCE design involves removing
the DCE and BLO unit, adding a concentrator, and extending or
replacing the boiler economizer. Capital and annual costs have
been evaluated for these three tasks. Additional upgrades are
included in the low-odor conversion control option, i.e., an ESP
upgrade to improve PM collection and a wet to dry ESP system
conversion to reduce gaseous organic HAP emissions. Separate
capital and annual costs were developed for the ESP upgrade and
6-21
-------
the wet to dry ESP system conversion. The ESP conversion costs
may be lower if an ESP upgrade is also performed at the same time
as the ESP conversion. However, no information is currently
available on the extent of the cost reduction. Therefore, the
costs were developed in the same way as those developed for
recovery furnaces in Sections 6.1.2.1 and 6.1.2.2.
Often other upgrades are performed at the same time as a
low-odor conversion. These upgrades usually provide additional
cost savings because of increased efficiency, increased process
capacity, and improved performance and safety. Possible upgrades
include combustion air system improvements, composite tubing, and
emergency drain and flame safety systems.15 The capital costs
and annual cost savings associated with these additional upgrades
have not been evaluated as part of the low-odor conversion
control option. The pulp production credits associated with
increased process capacity were not included in the low-odor
conversion total annual cost estimate because the additional
capacity increases may require significant modifications (e.g.,
expanding the recovery furnace bed or modifying the air system),
which would require additional capital expenses.
The low-odor conversion total annual cost estimates also do
not include (1) DCE maintenance cost savings, (2) ESP maintenance
cost savings, and (3) higher solids firing cost benefits for the
reasons described below.
Maintenance requirements associated with the DCE are
eliminated with the removal of this piece of equipment during a
low-odor conversion. The lower maintenance requirements
with an NDCE recovery furnace increase furnace
which allows for higher utilization of recovery
associated
availability
furnace capacity without additional costs. The DCE maintenance
cost savings were not included in the low-odor conversion cost
estimates because sufficient data are not available to quantify
the cost savings.
Lower corrosion rates are associated with NDCE recovery
furnace ESP's than with DCE recovery furnace ESP's. The lower
corrosion rates for NDCE recovery furnace ESP's are the result of
6-22
-------
a higher-temperature, lower-moisture content gas stream from NDCE
recovery furnaces compared to DCE recovery furnaces, as well as
from the predominant use of dry-bottom ESP's. Therefore,
converting to an NDCE design would eliminate the need for more
frequent ESP replacement resulting from ESP corrosion. The ESP
maintenance cost savings were not included in the low-odor
conversion cost estimates because sufficient data are not
available to quantify the cost savings.
Because concentrators can achieve higher BLS concentrations
than DCE's (i.e., 75 to 80 percent .vs. 65 percent), converting to
an NDCE recovery furnace provides the mill with an opportunity to
increase the solids content of the black liquor fired in the
furnace.4'16 However, increasing the solids content of the black
liquor to the upper limits requires additional capital expenses,
such as modifications to the fuel delivery system to handle a
more viscous liquid. Neither the potential cost credits nor the
additional capital expenses and any associated maintenance costs
associated with higher solids firing are included in the low-odor
conversion cost estimates.
Particulate matter control costs are included in the low-
odor conversion cost estimates. With the removal of the DCE,
which provides some PM control, as stated in Chapter 3, the ESP
often must be upgraded or replaced during a low-odor conversion
in order to meet applicable PM emission limits. For the purposes
of this cost analysis, an ESP upgrade PM control option that
would maintain or reduce PM emissions to the NSPS level of
0.10 g/dscm (0.044 gr/dscf) has been evaluated for those existing
DCE recovery furnaces that have PM emissions at or above the NSPS
level. This PM control option applies to model DCE recovery
furnaces RF-7a/7b through RF-9a/9b. These models represent
existing sources only, because no new DCE recovery furnaces are
expected to be built.
A PM control option that would reduce PM emissions to
0.034 g/dscm (0.015 gr/dscf) has also been evaluated for DCE
recovery furnaces that have PM emissions at or below the NSPS
level. This PM control option includes an ESP upgrade coupled
6-23
-------
with the addition of a packed-bed scrubber and applies to model
DCE recovery furnaces RF-7b through RF-9b, which, as cited above,
represent only existing sources.
The capital and annual costs for low-odor conversions are
presented in the following sections for model DCE recovery
furnaces.
6.1.2.3.1 Capital costs: jnt-.roduction. The total capital
cost of the low-odor conversion control option includes purchase
and installation costs of the extended economizer with associated
soot blowers and ash handling equipment; demolition costs for the
DCE and BLO unit; purchase and installation costs of the black
liquor concentrator; ESP upgrade capital costs; and wet to dry
ESP system conversion capital costs.
6 1.2.3.2 Capital costs; economizer expansion, demolition,.
and concentrator. The cost of economizer expansion and
demolition has been estimated at $6.5 million for a mid-size
recovery furnace. The cost estimate is based on cost data from
three sources--two recovery furnace manufacturers and one kraft
pulp mill, at which three low-odor conversions were completed
over a 3-year period. A 20 percent contingency factor was added
to the supplier costs to account for site-specific tie-in work.
Where applicable, the available cost data were adjusted to 1991
dollars and scaled for a 0.7 million kg BLS/d (1.5 million Ib
BLS/d) furnace using the six-tenths power rule. The $6.5 million
cost estimate is the average of the adjusted costs from the three
sources--$6 million, $4.8 million, and $8.6 million. ' '
The concentrator costs are based on cost information from a
concentrator manufacturer. The equipment costs for a falling
film concentrator range from $1.5 to $3 million. Total capital
costs, including installation, are approximately two times the
equipment cost.19 Based on this information, the average total
capital cost estimate for a concentrator is approximately
$4.5 million. The concentrator capital cost estimate does not
include liquor storage and piping costs or the cost for the
addition of a cooling tower cell.
6-24
-------
The total economizer expansion, demolition, and concentrator
cost for the mid-size model DCE recovery furnace (i.e., RF-8) is
approximately $11 million, equal to the sum of the $6.5 million
for the economizer expansion and demolition and the $4.5 million
for the concentrator. The six-tenths power rule was used to
calculate the capital costs for the small and large model DCE
recovery furnaces (i.e., RF-7 and RF-9). The low-odor conversion
capital costs (excluding the pulp production losses) are
presented in Table 6-11. The capital costs, excluding the ESP
upgrade and wet to dry ESP system conversion costs, range from
$8.09 million to $15.7 million for model DCE recovery furnaces
RF-7 through RF-9.
6.1.2.3.3 Capital costs: ESP upgrade. Electrostatic
precipitator upgrade costs to achieve total outlet PM emissions
of 0.10 g/dscm (0.044 gr/dscf) have been determined for the
applicable DCE recovery furnace models. The ESP upgrade capital
costs to control PM to NSPS levels for model DCE recovery
furnaces RF-7a through RF-9a are derived from the Schedule 1 ESP
manufacturer costs discussed in Section 6.1.2.1.3 and presented
in Table 6-4.9 The Schedule 1 costs were chosen because the ESP
upgrade could be completed within the scheduled time for the
low-odor conversion. The six-tenths power rule was used to
calculate the capital costs for the model DCE recovery furnaces.
The ESP upgrade costs to control PM emissions to NSPS levels for
the model DCE recovery furnaces are presented in Table 6-11,
excluding pulp production losses. The ESP upgrade costs range
from $881,000 to $1.70 million for RF-7a through RF-9a. The
bleached and unbleached pulp production losses are presented in
Table 6-12. Although NSPS PM emission levels are associated with
model DCE recovery furnaces RF-7b through RF-9b at baseline, once
they are converted to NDCE recovery furnaces, the ESP's must be
upgraded in order to maintain PM emissions at NSPS levels. The
ESP upgrade costs presented above for model furnaces RF-7a
through RF-9a were applied, as a worst-case cost estimate, for
model DCE recovery furnaces RF-7b through RF-9b.
6-25
-------
Control costs to achieve total outlet PM emissions of
0.034 g/dscm (0.015 gr/dscf) have been determined for the
applicable DCE recovery furnace models. These PM control costs
would include both ESP upgrade costs and packed-bed scrubber
costs. This section presents the ESP upgrade costs; the packed-
bed scrubber costs are presented in Section 6.1.2.4. The capital
costs for an ESP upgrade to control PM emissions to 0.034 g/dscm
(0.015 gr/dscf) were estimated by summing the Schedule 1 ESP
upgrade costs presented above and the ESP upgrade costs presented
in Table 6-8 and discussed in Section 6.1.2.1.5. These costs
apply to model DCE recovery furnaces RF-7a through RF-9a. The
ESP upgrade costs for model DCE recovery furnaces are presented
in Table 6-13, excluding pulp production losses, and range from
$9.80 million to $19.4 million for RF-7a through RF-9a. The
bleached and unbleached pulp production losses are presented in
Table 6-14. Using the same reasoning stated in the previous
paragraph, the ESP upgrade costs presented above for model
recovery furnaces RF-7a through RF-9a were applied, as a worst-
case cost estimate, for model DCE recovery furnaces RF-7b through
RF-9b.
6.1.2.3.4 Capital costs: wet to dry ESP system conversion.
The wet to dry ESP system conversion costs for model DCE recovery
furnaces RF-7 through RF-9 are based on ESP manufacturer costs
presented in Table 6-10 and discussed in Section 6.1.2.2.1 for
converting NDCE recovery furnace wet ESP systems to the dry ESP
system design.13 The ESP manufacturer costs were converted to
1991 dollars and then scaled for the model DCE recovery furnaces
using the six-tenths power rule. The conversion costs include
the costs to remove the existing agitator paddles and liquor
piping and install a perpendicular drag scraper system, shallow
fallout hoppers, drag chain conveyors, and rotary valves but do
not include the costs for asbestos removal or equipment or piping
beyond the rotary valves (e.g., an ash mixing tank and associated
equipment).13 The ESP system conversion costs for the model DCE
recovery furnaces are presented in Table 6-11 and range from
$439,000 to $849,000 for RF-7 through RF-9.
6-26
-------
6.1.2.3.5 Capital costs: total costs. The total capital
costs for the low-odor conversion option are equal to the sum of
the economizer expansion, demolition, and concentrator costs plus
the ESP upgrade and wet to dry ESP system conversion costs.
These costs, excluding pulp production losses, are presented in
Table 6-11 and range'from $9.41 million to $18.2 million for
RF-7a/7b through RF-9a/9b.
6.1.2.3.6 Capital costs: pulp production losses. The
production losses attributed to a low-odor conversion are
site-specific and depend on factors such as liquor storage
capacity, liquor trade or sell options, and coordination with
scheduled mill shutdowns.15 Pulp production losses were
calculated assuming an average additional shutdown period of
11 days beyond the scheduled 2-week shutdown period (i.e., a
total of 25 days of downtime). The pulp production losses were
calculated using the market values of bleached and unbleached
pulp discussed in Section 6.1.1.1 and an earnings margin of
25 percent.3
The average 25-day shutdown was estimated based on the
following information:
1. A time frame for completion of 21 to 30 days with proper
pre-shutdown planning and prefabrication;15
2. A case study where one mill completed three low-odor
conversions over a 4-week outage (i.e.., 31 days);20 and
3. A case study that involved two shutdowns; the first
shutdown was for several days to relocate ductwork, and the
second was for approximately 10 days to tie in the new systems.21
The total capital costs that include pulp production losses
associated with the low-odor conversion are presented in Tables
6-12 and 6-14. For mills producing bleached pulp, the pulp
production losses associated with a low-odor conversion range
from $488,000 to $1.46 million for RF-7a/7b through RF-9a/9b.
For mills producing unbleached pulp, the pulp production losses
associated with a low-odor conversion range from $348,000 to
$1.04 million for RF-7a/7b through RF-9a/9b.
6-27
-------
6.1.2.3.7 Incremental annual costs: introduction. The ITAC
estimate includes the following six primary components:
1. Capital recovery;
2. Administrative costs, taxes, and insurance;
3. Steam production credits;
4. Operating cost savings for the BLO;
5. Concentrator steam costs; and
6. Operating electricity costs for the ESP.
The annual costs of these six primary components were
estimated for the three model DCE recovery furnaces and summed to
determine the ITAC. As discussed in Section 6.1.2.3, certain
other potential cost savings were not quantified. These
potential cost savings include (1) maintenance cost savings
resulting from eliminating the DCE; (2) -ESP replacement cost
savings resulting from a less corrosive exit gas stream
associated with NDCE recovery furnaces; and (3) pulp production
credits for those mills that choose to provide for additional
capacity during the low-odor conversion.
6.1.2.3.8 Incremental annual costs: capital recovery. For
a low-odor conversion, the capital recovery costs for'the model
DCE recovery furnaces are based on the following:
1. An equipment life of 20 years;
2. The model capital costs presented in Tables 6-11 through
6-14;
3. The model bleached and unbleached pulp production losses
incurred during construction, which are presented in Tables 6-12
and 6-14; and
4. An interest rate of 7 percent.
Total capital recovery costs were calculated for each PM
control level for the following three scenarios:
1. Without annualized pulp production losses (Scenario 1);
2. With annualized bleached pulp production losses
(Scenario 2); and
3. With annualized unbleached pulp production losses
(Scenario 3).
6-28
-------
Design parameters for each model furnace are presented in
Table 6-15. *'
The low-odor conversion capital recovery costs for
Scenarios 1, 2, and 3 for model DCE recovery furnaces (including
an ESP upgrade to control PM emissions to NSPS levels) are
presented in Tables 6-16, 6-17, and 6-18, respectively. For
Scenario 1, capital recovery costs range from $918,000/yr to
$1.78 million/yr for RF-7a/7b through RF-9a/9b. For Scenario 2,
capital recovery costs range from $964,000/yr to $1.92 million/yr
for RF-7a/7b through RF-9a/9b. For Scenario 3, capital recovery
costs range from $951,000/yr to $1.88 million/yr for RF-7a/7b
through RF-9a/9b.
The low-odor conversion capital recovery costs for
Scenarios 1, 2, and 3 for model DCE recovery furnaces (including
an ESP upgrade to control PM emissions to 0.034 g/dscm
[0.015 gr/dscf]) are presented in Tables 6-19, 6-20, and 6-21,
respectively. For Scenario 1, capital recovery costs range from
$963,000/yr to $1.91 million/yr for RF-7a/7b through RF-9a/9b.
For Scenario 2, capital recovery costs range from
$1.01 million/yr to $2.05 million/yr for RF-7a/7b through
RF-9a/9b. For Scenario 3, capital recovery costs range from
$996,000/yr to $2.01 million/yr for RF-7a/7b through RF-9a/9b.
6.1.2.3.9 Incremental annual costs: administrative, taxes,
and insurance costs. Administrative, tax, and insurance costs
were estimated as 4 percent of the TCI and are presented in each
of the low-odor conversion annual cost tables, starting with
Table 6-16.6
6.1.2.3.10 Incremental annual costs: steam production
credits. Steam production credits result from the improved steam
flow that occurs with a low-odor conversion.22 The steam
production credit is assumed to be equal to the cost of the power
boiler fuel that has been displaced by black liquor in steam
generation. It was assumed that mills would first reduce the use
of higher-cost power boiler fuels, i.e., natural gas or fuel oil.
Therefore, only the reduction in the use of natural gas and fuel
oil was considered in determining the displaced fuel cost and not
6-29
-------
reductions in the use of lower-cost hogged wood or coal. To
determine the displaced fuel cost, the increase in the thermal
efficiency that results from a low-odor conversion was estimated.
The increase in the thermal efficiency that results from a
low-odor conversion is estimated to be 10 percentage points. The
supporting data for this estimate are as follows:
1. Direct contact evaporator recovery furnaces operate at
thermal efficiencies of 53 to 58 percent, whereas NDCE recovery
furnaces operate at thermal efficiencies of 63 to 68 percent;15
and
2. The thermal efficiency of a recovery furnace increases
approximately 1 percentage point for every 22 °C (40°F) drop in
exit temperature.16 The difference in the exit flue gas
temperatures before and after a low-odor conversion is about
200°C (400°F), which corresponds to a 10 percentage point
increase in thermal efficiency. For DCE recovery furnaces, the
exit flue gas temperature is 371°C to 427°C (700°F to 800°F).15
This high temperature range is needed to operate the DCE. For
NDCE recovery furnaces, the design exit flue gas temperature
range is 177°C to 357°C (350°F to 375°F).l6 The exit flue gas
temperature for NDCE recovery furnaces is limited by the optimum
operable range for the ESP (163°C to 204°C [325°F to 400°F]) and
recovery furnace operating and design parameters (163°C [325°F]
minimum).16
The steam production credit estimates are presented in each
of the low-odor conversion annual cost tables, starting with
Table 6-16. If natural gas is the displaced steam generation
fuel, the average steam production credit is estimated to range
from $758,000/yr to $2.27 million/yr for RF-7a/7b through
RF-9a/9b. If oil is the displaced fuel, the average steam
production credit is estimated to range from $1.19 million/yr to
$3.58 million/yr for RF-7a/7b through RF-9a/9b. These estimates
are based on the following information:
1. A thermal efficiency increase of 10 percentage points
(from 56 to 66 percent);
6-30
-------
2. Model BLS firing rates of 0.7, 1.2, and 1.8 million kg
BLS/d (0.9, 1.5, and 2.7 million Ib BLS/d), with a BLS heat
content of 13,900 kJ/kg (6,000 Btu/lb);8
3. A natural gas heat content of 38,100 kJ/m3
(1,024 Btu/ft3) and cost of $0.12/m3 ($3.48/1,000 ft3);23'24
4. A fuel oil heat content of 40,400 kJ/L (145,000 Btu/gal)
and cost of $0.20/L ($0.77/gal);23'25 and
5. A power boiler thermal efficiency of 85 percent.21
6.1.2.3.11 Incremental annual costs; BLO operating cost
savings. The high operating costs associated with air-sparging
BLO units are eliminated with the removal of the BLO unit during
a low-odor conversion. Most of the BLO operating costs (about
60 percent) is for power to operate the blowers and pumps. The
remaining 40 percent is for operating the reheater.26 These cost
savings are included as a credit in the total annual cost
estimate. The annual operating costs of a BLO system that
oxidizes black liquor for a DCE recovery furnace range from
$103,000 to $309,000 for RF-7 through RF-9. The cost savings
from removal of the BLO unit are presented in each of the low-
odor conversion annual cost tables, beginning with Table 6-16.
The cost savings are based on total annual costs of $251,900/yr
for a BLO 'unit that oxidizes black liquor fired in two DCE
recovery furnaces with a total black liquor firing rate of
1.0 million kg BLS/d (2.2 million Ib BLS/d)4.
6.1.2.3.12 Incremental annual costs: concentrator steam
costs. The concentrator'that replaces the DCE in a low-odor
conversion uses low-pressure steam to evaporate moisture from the
black liquor. The vapor from the concentrator can be used for
additional evaporation at lower black liquor solids levels, or
can be used to heat water. Concentrator steam costs were
estimated for each model DCE recovery furnace, assuming that
steam usage is proportional to the amount of black liquor
concentrated. The annual concentrator steam costs for the model
recovery furnaces range from $57,800 to $173,000 for RF-7 through
RF-9. The concentrator steam costs are presented in each of the
low-odor conversion annual cost tables, beginning with
6-31
-------
Table 6-16. Concentrator steam costs are based on the following
information:
1. A steam requirement of approximately 4,500 kg/hr
(10,000 Ib/hr) of low-pressure steam for a 1.1 million kg BLS/d
(2.4 million Ib BLS/d) furnace;27 and ^
2. A cost of $4.02/Mg ($3.65/ton) for low-pressure steam.
6.1.2.3.13 Incremental annual costs: electricity costs.
The increase in electricity costs for upgrading the ESP to
maintain or reduce PM emissions to the NSPS level of 0.10 g/dscm
(0.044 gr/dscf) was estimated for model DCE recovery furnaces
RF-7a/7b through RF-9a/9b. The actual increase in electricity
costs for the ESP upgrade is not known. It was assumed that the
incremental increase in electricity costs for an upgraded ESP
would be the same as for a replacement ESP. Because the
electricity cost increases for the replacement ESP in
Section 6.1.2.1.2 were based on an increase in SCA, the
electricity cost increases for the upgraded ESP were calculated
in the same manner.
-9a, with
PM
For DCE recovery furnaces RF-7a through
emissions above 0.10 g/dscm (0.044 gr/dscf), the ESP
baseline
electricity costs were estimated based on an increase in SCA
values from approximately 70 m2/(m3/sec) (330 ft2/!,000 acfm) to
approximately 100 m2/(m3/sec) (530 ft2/l,000 acfm).4 For DCE
recovery furnaces RF-7b through RF-9b, with baseline PM emissions
less than or equal to 0.10 g/dscm (0.044 gr/dscf) but greater
than 0.034 g/dscm (0.015 gr/dscf), the ESP electricity costs were
increase in SCA values from approximately
estimated based on an
90 m2/(m3/sec) (430 ft2/!,000 acfm) to
approximately
100 m2/(m3/sec) (530 ft2/!,000 acfm).4 The increase in ESP
electricity costs resulting from the maintenance or control of PM
emissions to the NSPS level are presented in Tables 6-16 through
6-18 for each of the model DCE recovery furnaces and range from
$70,000/yr for RF-7a through RF-9a and $ll,700/yr
$23,300/yr to
to $35,000/yr
for RF-7b through RF-9b.
The increase in electricity costs resulting from the
implementation of PM controls to reduce PM emissions to
6-32
-------
0.034 g/dscm (0.015 gr/dscf) was estimated for model DCE recovery
furnaces RF-7a/7b through RF-9a/9b. The PM control electricity
costs would include both ESP upgrade and packed-bed scrubber
electricity costs. The ESP upgrade electricity costs are
presented in this section; Section 6.1.2.4 presents the packed-
bed scrubber electricity costs.
For DCE recovery furnaces RF-7a through RF-9a, with baseline
PM emissions above 0.10 g/dscm (0.044 gr/dscf), the increase in
ESP electricity costs is based on an increase in the SCA from a
baseline value of approximately 70 m2/(m3/sec) (330 ft2/
1,000 acfm) to approximately 120 m2/(m3/sec) (620 ft2/
1,000 acfm}.4 For DCE recovery furnaces RF-7b through RF-9b,
with baseline PM emissions less than or equal to 0.10 g/dscm
(0.044 gr/dscf) but greater than 0.034 g/dscm (0.015 gr/dscf),
the ESP electricity costs were estimated based on an increase in
SCA values from approximately 90 m2/(m3/sec) (430 ft2/!,000 acfm)
to approximately 120 m2/(m3/sec) (620 ft2/!,000 acfm).4 The
increase in ESP electricity costs resulting from the control of
PM emissions to 0.034 g/dscm (0.015 gr/dscf) are presented in
Tables 6-19 through 6-21 for each of the model DCE recovery
furnaces and range from $33,100/yr to $99,200/yr for RF-7a
through RF-9a and $21,400/yr to $64,200/yr for RF-7b through
RF-9b.
As stated in Section 6.1.2.2.2, electricity costs do not
change significantly when a wet ESP system is converted to the
dry ESP system design because of an equal trade-off in horsepower
requirements between the wet and dry ESP system designs.
Therefore, no electricity costs for the wet to dry ESP system
conversion are presented in this cost analysis.
6.1.2.3.14 Incremental annual costs: total costs.
Incremental total annual costs were estimated for converting
model DCE recovery furnaces to an NDCE design for each of the PM
control levels, each of the scenarios, and each of the displaced
fuels. The scenarios were discussed in Section 6.1.2.3.8.
Scenario 1 excludes annualized pulp production losses from the
ITAC; Scenario 2 includes annualized bleached pulp production
6-33
-------
losses in the ITAC; and Scenario 3 includes annualized unbleached
pulp production losses in the ITAC.
The ITAC estimates for Scenario 1 for DCE recovery furnaces
with controlled PM emissions at NSPS levels are presented in
Table 6-16. For Scenario 1, with natural gas as the. displaced
steam generation fuel, the ITAC range from $520,000/yr to
$170,000/yr for RF-7a through RF-9a and $508,000/yr to
$140,000/yr for RF-7b through RF-9b. If fuel oil is the
displaced steam generation fuel, the ITAC for Scenario 1 range
from a cost of $90,000/yr for RF-7a to a cost savings of
$1.14 million/yr for RF-9a; for RF-7b through RF-9b, the ITAC for
Scenario 1 range from a cost of $80,000/yr to a cost savings of
$1.17 million/yr.
The ITAC estimates for Scenario 2 for DCE recovery furnaces
with controlled PM emissions at NSPS levels are presented in
Table 6-17. For Scenario 2, with natural gas as the displaced
steam generation fuel, the ITAC estimates range from $560,000/yr
to $310,000/yr for RF-7a through RF-9a and $548,000/yr to
$280,000/yr for RF-7b through RF-9b. If fuel oil is the
displaced steam generation fuel, the ITAC for Scenario 2 range
from $130,000/yr for RF-7a to $120,000/yr for RF-7a to a cost
savings of $1.00 million/yr for RF-9a; for RF-7b through RF-9b,
the ITAC for Scenario 2 range from a cost of $120,000/yr to a
cost savings of $1.03 million/yr.
The ITAC estimates for Scenario 3 for DCE recovery furnaces
with controlled PM emissions at NSPS levels are presented in
Table 6-18. For Scenario 3, with natural gas as the displaced
steam generation fuel, the ITAC range from $550,000/yr to
$270,000/yr for RF-7a through RF-9a and $538,000/yr to
$240,000/yr for RF-7b through RF-9b. If fuel oil is the
displaced steam generation fuel, the ITAC for Scenario 3 range
from a cost of $120,000/yr for RF-7a to a cost savings of
$1.04 million/yr for RF-9a; for RF-7b through RF-9b, the ITAC' for
Scenario 3 range from a cost of $110,000/yr to a cost savings of
$1.07 million/yr.
6-34
-------
The ITAC estimates for Scenario 1 for DCE recovery furnaces
with controlled PM emissions of 0.034 g/dscm (0.015 gr/dscf) are
presented in Table 6-19. For Scenario 1, with natural gas as the
displaced steam generation fuel, the ITAC range from $590,000/yr
to $380,000/yr for RF-7a through RF-9a and $578,000/yr to
$350,000/yr for RF-7b through RF-9b. If fuel oil is the
displaced steam generation fuel, the ITAC for Scenario 1 range
from a cost of $160,000/yr to a cost savings of $930,000/yr for
RF-9a; for RF-7b through RF-9b, the ITAC for Scenario 1 range
from a cost of $150,000/yr to a cost savings of $960,000/yr.
The ITAC estimates for Scenario 2 for DCE recovery furnaces
with controlled PM emissins of 0.034 g/dscm (0.015 gr/dscf) are
presented in Table 6-20. For Scenario 2, with natural gas as the
displaced steam generation fuel, the ITAC range from $630,000/yr
to $520,000/yr for RF-7a through RF-9a and $618,000/yr to
$490,000/yr for RF-7b through RF-9b. If fuel oil is the
displaced steam generation fuel, the ITAC for Scenario 2 range
from a cost of $200,000/yr for RF-7a to a cost savings of
$790,000/yr for RF-9a; for RF-7b through RF-9b, the ITAC for
Scenario 2 range from a cost of $190,000/yr to a cost savings of
$820,000/yr.
The ITAC estimates for Scenario 3 for DCE recovery furnaces
with controlled PM emissions of 0.034 g/dscm (0.015 gr/dscf) are
presented in Table 6-21. For Scenario 3, with natural gas as the
displaced steam generation fuel, the ITAC range from 620,000/yr
to $480,GOO/yr for RF-7a through RF-9a and $608,000/yr to
$450/000/yr for RF-7b for RF-9b. If fuel oil is the displaced
steam generation fuel, the ITAC for Scenario 3 range from a cost
of $190,000/yr for RF-7a to a cost savings of $830,000/yr for
RF-9a; for RF-7b through RF-9b, the ITAC for Scenario 3 range
from a cost of $180,000/yr to a cost savings of $860,000/yr.
6s. 1.2.4 Addition of Packed-Bed Scrubber. The addition of a
packed-bed scrubber downstream of the ESP is included in two of
the contrpl options examined for recovery furnaces. These
control options are (1) the use of an ESP plus a packed-bed
scrubber to meet an outlet PM emission level of 0.034 g/dscm
6-35
-------
(0.015 gr/dscf); and (2) the use of a packed-bed scrubber to
reduce HCl emissions from recovery furnaces.
The costs of replacing or upgrading ESP's to control PM
emissions are presented in Sections 6.1.2.1 and 6.1.2.3 for NDCE
and DCE recovery furnaces. This section discusses the design and
cost of packed-bed scrubbers for nine model recovery furnaces.
The design and cost of packed-bed scrubbers are presented for the
three model DCE recovery furnaces both with and without a low-
odor conversion. The applicable model recovery furnaces for the
packed-bed scrubber control option are RF-1 through RF-9.
Exhaust gas stream parameters for each model and associated
absorber are shown in Table 6-22.
6.1.2.4.1 Packed-bed scrubber design. Because only limited
information was available from a scrubber manufacturer regarding
the design parameters associated with a scrubber used to control
HCl emissions from kraft recovery furnaces, the model packed-bed
scrubbers were designed based on the procedures presented in
Chapter 9 of the OAQPS Control Cost Manual for counterflow
towers.6'29 Two assumptions were made to simplify the packed-bed
scrubber design analysis. First, it was assumed that the gas
stream exiting the ESP is cooled to saturation by a water spray
before it enters the packed-bed scrubber. As a result, the gas
stream flow rates into and out of the packed-bed scrubber are the
same. Insufficient design information was available from the
OAQPS Control Cost Manual and a scrubber manufacturer to include
in the design analysis a quench chamber for cooling the gas
stream. A second simplifying assumption was that the diffusivity
of HCl in the gas stream is approximated by the diffusivity of
HCl in the air.
Because the model gas flow rates are large and the inlet HCl
concentrations are low (only 9.7 ppmv for model NDCE and
converted model DCE recovery furnaces and 9.2 ppmv for
unconverted model DCE recovery furnaces), the diameters of the
model towers are 11 to 23 times the height of the packing. The
packing height was about 0.60 m (1.54 ft) for each model NDCE and
converted model DCE recovery furnace and 0.43 m (1.41 ft) for
6-36
-------
each unconverted model DCE recovery furnace. The tower diameters
ranged from 5.2 to 11 m (17 to 36 ft) for model NDCE and
converted model DCE recovery furnaces and 5.4 to 9.3 m (18 to
30 ft) for unconverted model DCE recovery furnaces. Design and
operating parameters for each model recovery furnace packed-bed
scrubber are shown in Table 6-23. The algorithm showing the
procedures and equations used to determine the packed-bed
scrubber design parameters is presented in a separate
memorandum.3 ®
6.1.2.4.2 Capital costs. Capital costs were calculated
based on procedures presented in the OAQPS Control Cost Manual.^
The unit costs used in calculating the capital and annual costs
were derived from the OAQPS Control Cost Manual and from
background information for the Medical Waste Incinerator
standard.6'31 The unit costs are shown in Table 6-24. Capital
costs in 1991 dollars are presented in Table 6-25 for each model
recovery furnace. These costs were compared to those obtained
from a scrubber manufacturer.2^
Capital costs consist of purchased equipment and
installation costs. Purchased equipment costs consist of
equipment, instrumentation, sales tax, and freight costs.
Equipment includes a fiberglass reinforced polyester stack,
fiberglass absorber tower, 5-cm (2-in.) randomly packed ceramic
Raschig rings, a liquid recirculating pump, an induced draft fan,
and a fan motor.
The addition of a packed-bed scrubber may result in
additional dissolved solids loading to the wastewater treatment
system. For those mills with restrictive total dissolved solids
(TDS) effluent limitations, the additional solids loading from
the scrubber may require internal process measures to reduce
dissolved loading from other areas of the mill. These internal
process measures will have associated engineering, equipment, and
construction costs. However, because the additional costs are so
site-specific, they cannot be estimated on a model basis and may
even be offset by any heat recovery benefits realized as a result
of adding an HC1 scrubber.
6-37
-------
All equipment costs were estimated as functions of various
design parameters. Specifically, the stack cost was based on the
length and diameter of the stack; the packed tower cost was based
on the surface area of the tower; the packing cost was based on
the volume of packing; the fan cost was based on the impeller
diameter; the fan motor cost was based on the horsepower rating;
and the pump cost was based on the design liquid flow rate. '
In most cases, the referenced costing equations were developed
for much smaller equipment. Thus, the costs for the models in
this analysis were developed by extrapolating well beyond the
largest parameter value for which the equations were developed.3
Instrumentation, sales tax, and freight were estimated to be
equal to 18 percent of the equipment costs. Installation was
estimated to be equal to 120 percent of the purchased equipment
costs.6 Total capital costs were estimated for each of the model
recovery furnaces. The model furnaces include both new and
existing furnaces. For model NDCE recovery furnaces RF-l through
RF-6, the total capital costs range from $1.10 million to
$2.58 million. For unconverted model DCE recovery furnaces RF-7
through RF-9, the total capital costs range from $736,000 to
$1.93 million. The total capital costs for the model DCE
recovery furnaces converted to the NDCE design are slightly lower
because of the change in furnace characteristics that occurs
after a low-odor conversion. The total capital costs for the
converted model DCE recovery furnaces RF-7 through RF-9 are
identical to those for comparably sized model NDCE recovery
furnaces and range from $707,000 to $1.85 million.
Capital costs for a packed-bed scrubber were also obtained
from a scrubber manufacturer. The manufacturer provided packed-
bed scrubber capital costs of $895,000, $1.69 million, and
$2.30 million for three model furnaces with scrubber inlet gas
flow rates of 47.2, 118, and 189 m3/sec (100,000, 250,000, and
400,000 acfm) , respectively. These costs include the cost for a
quench, packed tower,, mist eliminator, recirculation system
(including two recirculation pumps and recirculation piping),
7 Q
instrumentation, engineering costs, and exhaust stack.
6-38
-------
Corrected for differences in gas flow rates from the model costs
presented in the previous paragraph, the costs from the scrubber
manufacturer differ by approximately 5 to 25 percent from the
OAQPS costs presented in the previous paragraph for comparably
sized model furnaces. This cost difference is within the
±30 percent range of accuracy that OAQPS costs should have,
according to the OAQPS Control Cost Manual.6 However, it should
be noted that the OAQPS costs estimated here do not include
quench costs.
6.1.2.4.3 Annual costs. Annual costs for the packed-bed
scrubber were also calculated based on procedures presented in
the OAQPS Control Cost Manual. Unit costs used in the annual
cost calculations are shown in Table 6-24. Total annual costs in
1991 dollars are presented in Table 6-26 for the model recovery
furnaces.
Annual costs were developed for labor, maintenance
materials, water, caustic, wastewater disposal, electricity,
overhead, property taxes, insurance, administrative charges, and
capital recovery. Operator labor and maintenance labor were both
assumed to be 0.5 hr per 8-hr shift, with three shifts per day.
Supervisory labor costs were estimated to be equal to 15 percent
of the operator labor costs. Maintenance materials costs were
estimated to be equal to 100 percent of the maintenance labor
costs.6 The wastewater flow rate was estimated based on the
assumption that the NaCl concentration in the recirculating water
would be limited to 10 percent by weight. As a result, blowdown
is approximately 0.08 to 0.09 percent of the recirculating liquid
flow rate. Makeup water is needed for evaporative cooling before
the packed-bed scrubber and to replace the blowdown losses. A
stoichiometric amount of caustic is needed to react with all of
the HC1 in the exhaust gas stream; an additional amount of
caustic must be added to react with the S02 also present in the
exhaust gas stream (assuming 50 percent SO2 control).
Electricity usage by the fan was based on the gas flow rate out
of the quench shown in Table 6-22 and the pressure drop and
fan-motor efficiency shown in Table 6-23. Electricity usage by
6-39
-------
the pump was based on the gas flow rate out of the quench shown
in Table 6-22, the pump-motor efficiency shown in Table 6-23, and
an assumed pressure head of 18 m (60 ft). Overhead costs were
estimated to be equal to 60 percent of all labor and maintenance
materials costs.6' Collectively, property taxes, insurance, and
administrative charges were estimated to be equal to 4 percent of
the TCI.6 Capital recovery was estimated to be equal to a CRF
times the TCI.6 The CRF is 0.1098, based on an equipment life of
15 years and an interest rate of 7 percent.
For model NDCE recovery furnaces RF-1 through RF-6, the TAG
(as shown in Table 6-26) range from $348,000/yr to $790,000/yr.
For model unconverted DCE recovery furnaces RF-7 through RF-9,
the TAG range from $229,000/yr to $554,000/yr. The TAG for the
model DCE recovery furnaces converted to the NDCE design are
slightly higher because of higher costs for caustic, water, and
wastewater disposal. The higher costs are a result of the higher
S02 emission factor included in the equations for those costs.
Based on the limited information available, S02 emissions are
slightly higher, on average, from NDCE recovery furnaces than
from DCE recovery furnaces.32 The TAG for the converted model
DCE recovery furnaces RF-7 through RF-9 are identical to those
for comparably sized model NDCE recovery furnaces and range from
$234,000/yr to 571,000/yr.
6.1.3 Black Liquor Oxidation Unit Control Options
Two control options, conversion of a DCE recovery furnace
system to an NDCE recovery furnace and incineration of BLO vent
evaluated for controlling gaseous organic HAP
gases, were
emissions from air-sparging BLO
units. The cost of the first
option--converting DCE recovery furnace systems to NDCE recovery
furnaces--was presented in Section 6.1.2.3. The following
section presents the capital and annual costs of the second BLO
control option--incineration of BLO vent gases. This BLO control
option applies to model BLO units BLO-1 through BLO-3, which
represent existing BLO units associated with DCE recovery
furnaces. These models represent only existing BLO units because
6-40
-------
no new DCE recovery furnace systems with BLO units are expected
to be installed.
6.1.3.1 Capital Costs. The total capital costs to collect
BLO vent gases and incinerate them in a power boiler or other
incineration device are based on a 1990 BLO control cost estimate
of $4.8 million supplied by industry for a 730 ADMP/d
(800 ADTP/d) kraft pulp mill.33 The 1990 BLO control cost
estimate includes piping, fans, condensers, and safety-related
equipment, such as flame arresters, rupture disks, etc. No major
power boiler modifications (such as scrubber modifications) are
included. The BLO control cost estimate was adjusted to 1991
dollars and then scaled using the six-tenths power rule described
in Section 6.1.1.2 to derive BLO control costs for the model BLO
units. A conversion factor of 1,700 kg BLS/ADMP (3,400 Ib
BLS/ADTP) (the average for bleached and unbleached pulp mills
together) was assumed in scaling the cost. Design parameters for
each model BLO unit are presented in Table 6-27. The model BLO
control capital costs are presented in Table 6-28. The BLO
control capital costs range from $2.5 million to $4.83 million
for BLO-1 through BLO-3. Because the BLO collection and
incineration system for one mill was installed within a 1-week
maintenance shutdown, it was assumed that no downtime would be
necessary beyond the annual 2-week shutdown used in determining
costs.34 As a result, no pulp production losses are included in
these costs.
6.1.3.2 Annual Costs. Annual costs were estimated for
operating and supervisory labor, maintenance labor and materials,
electricity, steam, and indirect costs (e.g., overhead,
administrative, taxes, insurance, and capital recovery). The BLO
control annual costs were estimated for each model BLO unit based
on the following sources:
" 1. Annual operating requirements provided by one mill (Mill
A) for its BLO vent gas control system;2^
2. Operating labor costs from the U. S. EPA Handbook:
Control Technologies for Hazardous Air Pollutants;5 and
6-41
-------
3. Supervisory labor and maintenance costs and indirect
costs from the OAQPS Control Cost Manual.6
The operating labor costs were estimated assuming 0.5 hours
per shift per condenser for two condensers for three 8-hour
shifts per day.5 The operating labor hourly rate was assumed to
be $17/hr.5 The annual operating labor hours were assumed to be
365 d/yr. The supervisory labor costs were assumed to be
15 percent of the operating labor costs.6 The maintenance labor
costs were estimated at 1.5 times the operating labor costs.
The maintenance materials costs were estimated at 100 percent of
the maintenance labor costs.6
Electricity costs were estimated based on the total kW (hp)
requirements to operate the BLO vent gas control equipment at
Mill A and scaled for the model BLO units assuming a direct
relationship between BLO vent gas flow rate and electricity
costs. For Mill A, with a BLO vent gas flow rate of 7.7 m3/sec
(16,327 acfm), 980 kW (100 hp) are required to operate the mill
water booster pump motor, 29 kW (3 hp) to operate the BLO
condenser condensate pump motor, and 3,900 kW (400 hp) to operate
the BLO off gas blower motor.26 The model electricity costs were
estimated assuming 8,424 operating hr/yr and $0.06/kWh.
Steam costs were estimated based on the steam requirements
and unit steam cost for the BLO off-gas reheater at Mill A--
730 kg steam/hr (1,600 Ib steam/hr) and $7/Mg of steam
($3/1,000 Ib of steam), respectively--and scaled for the model
BLO units assuming a direct relationship between BLO vent gas
flow rate and steam costs.26 The model steam costs were
estimated assuming 8,424 operating hr/yr.
Indirect costs were estimated using assumptions in the OAQPS
Control Cost Manual. Overhead costs were estimated as 60 percent
of labor and maintenance costs. Administrative, taxes, and
insurance costs were estimated as 4 percent of the TCI. Capital
recovery costs were calculated as the product of the CRF and the
TCI.6 The CRF is 0.1424, based on a 10-yr equipment life for the
5 24
ductwork and condenser and an interest rate of 7 percent. '
6-42
-------
Total annual costs are presented in Table 6-28 and range
from $681,000/yr to $1.32 million/yr for BLO-1 through BLO-3.
6.1.4 Smelt Dissolving Tank Control Options
Two PM control options that would reduce PM emissions from
SDT's have been evaluated. The first option would reduce PM
emissions from existing SDT's to the NSPS level of 0.10 kg/Mg
(0.20 Ib/ton) BLS. The second option would reduce PM emissions
from existing SDT's to a more stringent level of 0.06 kg/Mg
(0.12 Ib/ton) BLS; the second option also applies to new SPTs;
the option could be used to evaluate the cost to new sources
subject to a more stringent standard (0.06 kg/Mg [0.12 Ib/ton]
BLS) than the current NSPS.
For mills with existing SDT scrubbers, the costs of both PM
control options were estimated based on replacing the existing
scrubber with a new scrubber. These-costs were estimated for SDT
models SDT-1 through SDT-4. For mills with new SDT scrubbers,
the costs of installing scrubbers under the second, more
stringent PM control option also apply to SDT-1 through SDT-4.
For the purposes of this cost analysis, the capital cost to
install a new SDT scrubber capable of meeting 0.06 kg/Mg
(0.12 Ib/ton) BLS was assumed to be the same as the cost to
replace an existing scrubber with a new scrubber capable of
meeting 0.06 kg/Mg (0.12 Ib/ton) BLS. However, that may be an
overestimate because the capital cost that would be attributable
to the 0.06 kg/Mg (0.12 Ib/ton) BLS control option would only be
that cost associated with controlling PM emissions from the
current NSPS level of 0.10 kg/Mg (0.2 Ib/ton) BLS to 0.06 kg/Mg
(0.12 Ib/ton) BLS. Such a cost is more similar to a scrubber
modification cost than a scrubber replacement cost.
For mills with existing SDT mist eliminators, the costs of
both PM control options were estimated based on replacing the
existing mist eliminator with a new scrubber. These costs were
estimated for SDT models SDT-5 through SDT-7. The costs of
installing new mist eliminators were not examined because mist
eliminators are not,assumed to be installed on new SDT's.
6-43
-------
Note: For at least one scrubber type (wetted-wheel
scrubbers), the scrubber replacement capital and annual costs
would be approximately equal to the cost of replacing a mist
eliminator with a scrubber. Similar to mist eliminators, the
total capital costs would be based on replacement of the entire
scrubber system. Also similar to mist eliminators, the
incremental annual costs would include electricity costs based on
replacing an existing scrubber that has a low pressure drop with
another that has a significantly higher pressure drop, as well as
indirect costs based on a capital cost of replacing the entire
scrubber system. The capital and annual costs to replace wetted-
wheel scrubbers will not be presented in this cost analysis
because there is insufficient information to estimate the costs
and because only 13 percent of SDT's have wetted-wheel
scrubbers.4
The following sections present the capital and annual
costs to replace existing wet scrubbers and mist eliminators with
new wet scrubbers under the SDT PM control options.
6.1.4.1 Replacement of Existing Scrubber with New Scrubber:
Capital Costs. The conditions under which replacing a scrubber
is more cost-effective than modifying an existing scrubber are •
very site-specific. To be conservative, only the cost to replace
a scrubber was evaluated. The cost to dispose of the existing
scrubber was assumed to be included in the scrubber replacement
costs. The capital costs to replace the scrubber are based on
recent costs provided by a pulp and paper mill for an SDT
scrubber.4 The available long-term PM emissions data for the
scrubber show that it is capable of consistently meeting both of
the SDT PM emission limits--0.10 kg/Mg (0.20 Ib/ton) BLS and
0.06 kg/Mg (0.12 Ib/ton) BLS.10 The SDT scrubber is a packed-
tower scrubber with a gas flow rate of 3.8 m3/sec (8,071 acfm) at
60 to 70 percent of recovery furnace capacity. The costs
provided by the pulp and paper mill are packed-tower scrubber
Hs and scrubber modification costs. The packed-tower scrubber
is $280,000 (1988 dollars); the scrubber'modification cost
50,000 (1991 dollars).4 The 1988 scrubber cost was adjusted
6-44
-------
to 1991 dollars using the Chemical Engineering Plant Cost Index
and then summed with the scrubber modification costs to obtain
the total capital cost in 1991 dollars. The six-tenths rule was
used to extrapolate the total capital cost for each of the model
SDT's. Based on information from a scrubber manufacturer, the
on-site work time to replace a scrubber is about 2 days.35
Therefore, no downtime beyond the annual 2-week shutdown is
necessary for the scrubber replacement, which means that no pulp
production losses are expected for this control option.
The model scrubber replacement costs are presented in
Table 6-29. Total capital investment costs for the SDT models
SDT-1 through SDT-4 range from $292,000 to $706,000.
6.1.4.2 Replacement of Existing Scrubber with New Scrubber:
Incremental Annual Costs. The ITAC for each SDT model include
the indirect costs associated with the TCI of the new scrubber
(i.e., administrative, property tax, and insurance costs plus the
capital recovery costs). The administrative, taxes, and
insurance costs were estimated to be equal to 4 percent of the
TCI.6 A properly designed and maintained venturi scrubber can
operate for 20 years. 5 Other types of scrubbers may have
different life spans. To be conservative, a 15-year life span
for replacement scrubbers was assumed. The capital recovery cost
was estimated based on a 15-year scrubber life and a 7 percent
interest rate. The increase in electricity costs from the
scrubber replacement is not included in the ITAC estimate because
it is not significant relative to the total annual cost.' The
average pressure drop for existing SDT scrubbers with PM
emissions greater than 0.10 kg/Mg (0.20 Ib/ton) BLS (i.e.,
baseline pressure drop) is approximately the same as the average
pressure drop for those SDT scrubbers capable of meeting both SDT
PM emission limits (0.10 kg/Mg [0.20 Ib/ton] BLS) and 0.06 kg/Mg
[0.12 Ib/ton] BLS) on a long-term basis.4'10 The baseline and
control level pressure drops are 12 mm Hg (6.5 in. H20) and 13 mm
Hg (7 in. H20') , respectively.4 All other direct costs (i.e.,
costs for operating labor, maintenance, water, and wastewater
treatment) are assumed to be the same as those incurred by the
6-45
-------
existing scrubber and, therefore, are not included in the ITAC
estimate.
The ITAC for the four model SDT's are presented in
Table 6-29. The ITAC estimates range from $43,800/yr to
$106,000/yr for SDT-1 through SDT-4.
6.1.4.3 Replacement of Existing Mist Eliminator with New
Scrubber: Capital Costs. The capital costs to replace an
existing mist eliminator with a new scrubber include the costs
for the new scrubber and all auxiliary equipment (i.e., fans,
ductwork, etc.) that would be required at a new source. Based on
information from a scrubber manufacturer, a completely new
scrubber system would cost about twice as much as replacing only
the scrubber, as described in Section 6.1.4.I.35 The cost to
dispose of the existing mist eliminator was assumed to be
included in these replacement costs. Therefore, the capital
costs for the three model SDT's are equal to twice the scrubber
replacement costs for the three corresponding size model SDT's
presented in Section 6.1.4.1. It is assumed that no downtime
beyond the annual 2-week shutdown is necessary for replacing a
mist eliminator with a scrubber.
The design parameters for the three SDT models are presented
in Table 6-30. Capital costs for replacing a mist eliminator
with a new scrubber are presented in Table 6-31. Total capital
investment costs for the SDT models SDT-5 through SDT-7 range
from $584,000 to $1
6.1.4.4 Repl
} million
acement (
Existing Mist Eliminator with New
for each SDT
Scrubber: Incremental Annual Costs. Incremental total annual
costs were estimated for each of the three model SDT's. The ITAC
model include both direct and indirect annual costs,
with the exception of operating labor. No changes in operator
and supervisor personnel were assumed to be required. The
increase in maintenance labor costs was estimated, assuming
3 hr/d at a wage rate of $25/hr.6 Maintenance materials were
estimated at 100 percent of maintenance labor costs. As
discussed in Section 6.1.1.3.1, for the purposes of calculating
costs, pressure drop is used as an indicator of PM collection
6-46
-------
efficiency. Electricity requirements and costs increase as a
result of increasing the pressure drop to reduce PM emissions to
0.10 kg/Mg (0.20 Ib/ton) BLS and to 0.06.kg/Mg (0.12 Ib/ton) BLS.
The average pressure drop is 1.3 mm Hg (0.7 in. H20) for those
SDT mist eliminators with PM emissions above 0.06 kg/Mg
(0.12 Ib/ton) BLS (i.e., baseline pressure drop); the average
pressure drop is 13 mm Hg (7 in. H2O) for those SDT scrubbers
capable of meeting both PM emission limits (0.10 kg/Mg
[0.20 Ib/ton] and 0.06 kg/Mg [0.12 Ib/ton] BLS) on a long-term
basis.4'10 Therefore, an increase in pressure drop from 1.3 to
13 mm Hg (0.7 to 7 in. H20) was used to estimate the increase in
electricity requirements and resulting increase in electricity
costs. The overhead cost was estimated to be equal to 60 percent
of the total maintenance cost.6 The administrative, taxes, and
insurance costs were estimated to be equal to 4 percent of the
TCI.6 The capital recovery cost was estimated to be equal to the
product of a CRF and the TCI.6 The CRF is 0.1098, based on a
15-year scrubber life and a 7 percent interest rate.
The ITAC for the SDT models SDT-5 through SDT-7 are
presented in Table 6-31 and range from $190,000/yr to
$301,000/yr.
6.1.5 Lime Kiln Control Options
Two PM control options have been evaluated for existing and
new lime kilns. One PM control option that has been evaluated
for existing lime kilns would reduce PM emissions to the NSPS
level for gas-fired lime kilns--0.15 g/dscm (0.067 gr/dscf). For
existing lime kilns with wet scrubbers, the control option would
involve replacing the existing scrubber with an ESP. However,
the actual control device (e.g., ESP or high-efficiency scrubber}
selected by a particular mill would be site-specific. The costs
for this PM control option were estimated for lime kiln models
LK-1 through LK-3, which represent existing lime kilns controlled
with wet scrubbers.
Based on PM emissions data supplied by mills, lime kilns
controlled with ESP's already achieve a PM level of 0.15 g/dscm
(0.067 gr/dscf).4'10 Therefore, costs were not estimated for the
6-47
-------
control option reducing PM emissions to 0.15 g/dscm
(0.067 gr/dscf) for lime kilns controlled with ESP's (represented
by models LK-4 through LK-6).
A second PM control option that was evaluated for new and•
existing lime kilns would reduce PM emissions to 0.023 g/dscm
(0.010 gr/dscf). For existing lime kilns with wet scrubbers, the
control option would involve replacing the existing scrubber with
an ESP; costs would be estimated for models LK-1 through LK-3.
For existing lime kilns with ESP's, the control option would
involve upgrading the existing ESP. For new lime kilns, the
control option would involve installing a new ESP capable of
achieving the 0.023 g/dscm (0.010 gr/dscf) PM level. The costs
for upgrading or installing a new ESP were estimated for models
LK-4 through LK-6. The actual control device selected by a
particular mill would actually be site-specific. The capital and
annual costs for each of these options are presented in the
following sections.
6.1.5.1 Replacement of Existing Scrubber with ESP: Capital.
Costs. The costs of replacing a scrubber with an ESP were
calculated based on recent ESP costs provided by individual pulp
and paper mills.4 The cost to dispose of the existing scrubber
was assumed to be included in the new ESP costs provided by the
individual mills. New lime kiln ESP costs average $484/m2
($45/ft2) of ESP plate area.4 The ESP plate area for each model
lime kiln was derived from the model gas flow rate at the ESP
inlet and the SCA for the new ESP. The SCA for model lime kiln
ESP's meeting a PM level of 0.15 g/dscm (0.067 gr/dscf) is
estimated to be 90 m2/(m3/sec) (460 ft2/!,000 acfm).4 The SCA
value is based on the SCA of an actual lime kiln ESP for which
long-term PM emissions data are available to demonstrate that its
PM emissions are consistently at or below 0.15 g/dscm
(0.067 gr/dscf).10 The SCA for model lime kiln ESP's meeting a
PM level of 0.023 g/dscm (0.010 gr/dscf) is estimated to be 220
m2/(m3/sec) (1,120 ft2/!,000 acfm).4 The SCA value is based on
the SCA of an actual lime kiln ESP for which long-term PM
6-48
-------
emissions data are available to demonstrate that its PM emissions
are consistently at or below 0.023 g/dscm (0.010 gr/dscf).10
Installing a new fan and stack typically comprises
approximately 20 percent of the total capital costs of installing
a new ESP. Because the fan and stack usually do not need to be
replaced when a scrubber is replaced with an ESP, the cost for
replacing a scrubber with an ESP is only about 80 percent of the
cost of a completely new ESP.9 Based on information from an ESP
manufacturer, the lifetime of the replacement ESP is about
15 years.36
Table 6-32 presents the model replacement costs to control
PM emissions to 0.15 g/dscm (0.067 gr/dscf). The costs range
from $457,000 to $1.50 million for LK-l through LK-3. Table 6-33
presents the model replacement costs to control PM emissions to
0.023 g/dscm (0.010 gr/dscf). The costs range from $1.11 million
to $3.65 million for LK-l through LK-3.
It was assumed that installation of the ESP could be
completed within the 2-week scheduled shutdown, and, therefore,
pulp production losses were not included in these cost estimates.
Further information is needed to determine the validity of this
assumption (e.g., the lack of available space complicating ESP
installation and thereby increasing costs).
6.1.5.2 Replacement of Existing Scrubber with ESP:
Incremental Annual Costs. To determine the incremental annual
costs of replacing the existing scrubber with a new ESP for each
model lime kiln, the annual costs for operating the existing
scrubber were subtracted from the annual costs for operating a
new ESP. To be conservative, the TCI-related indirect costs for
the existing scrubber (i.e., the administrative, property tax,
insurance, and capital recovery costs) were not included in the
cost comparison.
The incremental annual costs of replacing the existing
scrubber with a new ESP are presented in Tables 6-32 and 6-33.
Design parameters and costs for the existing scrubber at baseline
PM levels are presented in Tables 6-34 and 6-35, respectively.
Because over 80 percent of lime kiln scrubbers are venturi
6-49
-------
scrubbers, the scrubber design parameters presented in Table 6-34
are based on a venturi scrubber as the baseline control device.
Design parameters for the new ESP are presented in Table 6-36,
and the ESP costs at the 0.15 g/dscm (0.067 gr/dscf) and
0.023 g/dscm (0.010 gr/dscf) PM levels are presented in Tables
6-37 and 6-38, respectively.
Direct annual costs (i.e., operating labor costs,
maintenance costs, and utility costs) are reduced significantly
when the existing scrubber is replaced with an ESP. Overhead
costs are also reduced because they are a function of the labor
and maintenance costs. For model lime kilns LK-1 through LK-3
controlling PM emissions to 0.15 g/dscm (0.067 gr/dscf), the
reduction in direct annual costs and overhead costs obtained by
switching from a scrubber to an ESP was greater than the
TCI-related indirect annual costs for the new ESP. As a result,
the ITAC are actually cost savings of $l04,000/yr to $53,000/yr
for LK-1 through LK-3. The ITAC savings for the model lime kilns
are presented in Table 6-32.
For model lime kilns LK-1 through LK-3 controlling PM
emissions to 0.023 g/dscm (0.010 gr/dscf), the reduction in
direct annual costs and overhead costs was less than the TCI-
related indirect'costs for the ESP. As a result, there are
annual costs associated with replacing the existing scrubber with
a new ESP, ranging from $17,400/yr to $342,000/yr for LK-1
through LK-3. The TCI-related indirect annual costs for the ESP
were strongly influenced by the high SCA value used to estimate
the TCI. The SCA for new ESP's meeting a PM level of
0.023
„*.!_ « dbdkaiM — —• — — — — —
g/dscm (0.010 gr/dscf) is approximately 220 m2/(m /sec)
(1,120 ft2/!,000 acfm).4 The ITAC for the model lime kilns are
presented in Table 6-33.
Approximately 15 percent of lime kiln scrubbers for which
data are available have pressure drops below 19 mm Hg (10 in.
H20), or are low-energy, low-pressure, or ejector type
scrubbers.4 These scrubbers operate at a lower pressure drop
than the venturi scrubber used as the baseline control device.
The annual costs to replace these low-pressure drop scrubbers
6-50
-------
with ESP's would be higher than the costs and cost saving
presented above. The additional annual costs for replacing low-
pressure drop scrubbers were estimated based on the difference in
pressure drop between the low-pressure drop scrubbers and the
venturi scrubber used as the baseline control device. The
average pressure drop is 7.5 mm Hg (4 in. I^O) for those low-
pressure drop lime kiln scrubbers with PM emissions greater than
0.15 g/dscm (0.067 gr/dscf). The average pressure drop is 39 mm
Hg (21 in. H2O) for the baseline venturi scrubber. To estimate'
the incremental annual costs of replacing low-pressure drop lime
kiln scrubbers with ESP's, annual costs of $34,500/yr for LK-1,
$65,700 for LK-2, and $112,000/yr for LK-3 should be added to the
annual costs and cost savings presented above.
6.1.5.3 Upgrade of Existing ESP: Capital Costs. Because no
actual ESP upgrade costs were available for controlling PM
emissions from a level of 0.15 g/dscm (0.067 gr/dscf) to a level
of 0.023 g/dscm (0.010 gr/dscf}, the upgrade cost for the lime
kiln ESP was instead based on the incremental cost difference
between an ESP capable of achieving a PM level of 0.15 g/dscm
(0.067 gr/dscf) and one capable of achieving a PM level of
0.023 g/dscm (0.010 gr/dscf).
The ESP upgrade capital costs were based on the recent ESP
costs provided by individual pulp and paper mills (i.e., costs
for ESP's installed or replaced during or after 1989).4 The new
lime kiln ESP costs average $484/m2 ($45/ft2) of ESP plate area.4
To determine the ESP upgrade cost, this cost per ESP plate area
was multiplied by the increase in ESP plate area assumed to
reduce PM emissions from 0.15 g/dscm (0.067 gr/dscf) to
0.023 g/dscm (0.010 gr/dscf). The ESP plate area for each model
lime kiln was derived from the model gas flow rate at the ESP
inlet and the ESP SCA. The model SCA for an ESP meeting a long-
term PM level of 0.15 g/dscm (0.067 gr/dscf) is approximately 90
m2/(m3/sec) (460 ft2/!,000 acfm).4'10 The model SCA for an ESP
meeting a long-term PM level of 0.023 g/dscm (0.010 gr/dscf) is
approximately 220 m2/(m3/sec) (1,120 ft2/!,000 acfm).4'10
6-51
-------
For new lime kilns installing ESP's to control PM emissiosn
to 0.023 g/dscm (0.010 gr/dscf), the capital cost attributable to
the control option would not be the cost of a new ESP but only
that portion associated with controlling PM emissions from the
current NSPS level of 0.15 g/dscm (0.067 gr/dscf) to 0.023 g/dscm
(0.010 gr/dscf). Such a cost would be the same as the cost to
upgrade existing lime kiln ESP's to achieve the same PM level of
0.023 g/dscm (0.010 gr/dscf).
The capital costs, excluding pulp production losses, for the
ESP upgrade control option for new and existing model lime kilns
are presented in Table 6-39 and range from $654,000 to
$2.15 million for LK-4 through LK-6. No information is currently
available for existing lime kilns on the amount of time required
to complete an ESP upgrade that would achieve a PM control level
of 0.023 g/dscm (0.010 gr/dscf). As a result, no pulp production
losses were estimated for this ESP upgrade control option.
6 1.5.4 Upgrade of Existing ESP: Incremental Annual Costs.
Labor and maintenance costs are assumed to be unchanged when the
ESP is upgraded to achieve a PM level of 0.023 g/dscm
(0.010 gr/dscf). Therefore, the incremental annual costs for the
ESP upgrade include only the increase in electricity costs and
the TCI-based indirect annual costs, which include
administrative, property tax, insurance, and capital recovery
costs. The capital recovery cost is based on an average 15-year
life span for lime kiln ESP's.36 The increased electricity costs
are based on an increase in SCA from 90 m2/(m3/sec)
(460 ft2/!,000 acfm) to 220 m2/(m3/sec) (1,120 ft2/!,000 acfm) .
The ITAC for the new and existing model lime kilns, without
taking pulp production losses into account, are presented in
Table 6-39 and range from $112,000/yr to $369,000/yr for LK-4
through LK-6.
6.2 ENHANCED MONITORING COSTS
The following sections present the estimated costs for the
enhanced monitoring options discussed in Chapter 4. Table 6-40
presents a summary of the enhanced monitoring costs, and
Table 6-41 presents the itemized capital and annual costs for
6-52
-------
opacity monitors and HC1 CEM's. Enhanced monitoring costs were
not estimated for control options other than those presented in
Chapter 4; if facilities choose to meet the emission limits
through the application of other control options, and a CEM is
not applicable because of cost and/or technology constraints, the
facilities must develop an enhanced monitoring plan that
demonstrates the ability of the selected parameter to gauge a
change in emissions.
6.2.1 Recovery Furnace Enhanced Monitoring
The following sections present the costs of enhanced
monitoring options that can be used to demonstrate compliance
with recovery furnace emission limits for PM or PM HAP's, total
gaseous organic HAP's, and HC1.
6.2.1.1 Enhanced Monitoring for PM or PM HAP's Controlled
with an ESP. Because opacity is the surrogate measurement that
best characterizes the level of recovery furnace PM emissions,
installation of an opacity monitor after the ESP is one option
being considered as a means of demonstrating compliance with a PM
or PM HAP emission limit for recovery furnaces. For those
recovery furnaces with a wet scrubber following the ESP, an
opacity monitor must be located after the ESP but prior to the
scrubber. A computer program distributed by the Emission
Measurement Technical Information Center (EMTIC) of EPA was used
to estimate capital and annual costs for an opacity monitor.
The capital cost from EMTIC to purchase and install an
in-situ opacity monitor (i.e., an opacity monitor that measures
emissions in the stack or duct) is approximately $34,800. The
capital costs include planning, selecting the type of equipment,
providing support facilities, PEC, installing and checking CEM's,
performance specification tests (certification), and preparing
the quality assurance/quality control (QA/QC) plan required by
appendix F (40 CFR 60). The PEC includes the cost to purchase a
data acquisition system (DAS) which includes data reduction and
reporting hardware/software.37
The annual costs from EMTIC equal $16,500/yr and include
costs for operating and maintenance, reporting and recordkeeping,
6-53
-------
and annual review and update.37 Administrative, insurance,
property tax, and capital recovery costs were estimated
separately and added to the annual costs from the EMTIC program.
The administrative, insurance, and property tax costs were
calculated as 4 percent of the TCI, based on guidance in the
OAQPS Control Cost Manual.6 The capital recovery cost was
calculated as a product of a CRF and the TCI.6 The CRF was
calculated assuming a 20-year equipment life and 7 percent
interest. The total annual cost for the opacity monitor is
approximately $2l,200/yr.
Method 5, Method 29, or Method 17 compliance tests could be
performed periodically (e.g., semiannually) as a substitute for
an opacity monitor. The estimated cost for one three-run, EPA
Method 5 compliance test is $8,500.37 The estimated cost for one
three-run, EPA Method 29 compliance test is $12,000.3 No costs
are available for a three-run, EPA Method 17 compliance test. If
performed semiannually, the Method 5 tests would cost $17,000/yr;
the Method 29 tests would cost $24,000/yr.
Another option being considered is for the facility to
develop a monitoring plan that specifies ESP operating parameters
to be monitored. Operating parameters for the ESP would be site-
specific and would be based on the parameters measured during a
three-run, EPA Method 5, Method 29, or Method 17 compliance test
that showed the facility to be in compliance with the applicable
PM or PM HAP emission limit. The cost to implement a monitoring
plan has not been estimated. The estimated costs for three-run,
EPA Method 5 or Method 29 compliance tests were discussed in the
previous paragraph. No costs are available for a three-run, EPA
Method 17 compliance test.
6.2.1.2 Enhanced Monitoring for PM QT- PM HAP's Controlled
w-it-Ji a Wet Scrubber. For those recovery furnaces that can comply
with a PM or PM HAP emission limit with existing wet scrubbers,
the use of an opacity monitor to demonstrate compliance with the
PM emission limit may be inappropriate. The exhaust from the
recovery furnace wet scrubber will have a high moisture content
and will interfere with the readings from an opacity monitor.
6-54
-------
Monitoring scrubber operating parameters (i.e., pressure drop and
scrubber liquid flow rate) is an alternative enhanced monitoring
option for showing compliance with a PM or PM HAP emission limit
for recovery furnaces. The pressure drop and liquid flow rate
are indirect measurements of the performance of the scrubber.
The pressure drop across a wet scrubber can be determined by
using a basic magnehelic gauge coupled with the S-type pitot tube
and would require manual reading. The pressure drop can also be
determined using a similar arrangement with an electronic
magnehelic gauge that produces a digital signal. The cost for a
manual read-out system would be less than $300. The additional
cost for the electronic magnehelic gauge is not currently known.
There are various techniques for measuring liquid flow rate
in a wet scrubber. These techniques include ultra-sonic
detection mounted externally on the in-flowing water pipe and
turbine devices that are mounted within the pipe, both of which
generate an electrical signal that can be logically displayed in
a control room. The cost for these flow-measuring techniques can
vary from $2,000 to $25,000, depending on the device sensitivity
and distance from the control room.^7
Pressure drop and scrubber liquid flow rate levels would be
site-specific and would be based on the operating parameters
measured during a three-run, EPA Method 5, Method 29, or
Method 17 compliance test that showed the facility to be in
compliance with the applicable PM or PM HAP emission limit.
Method 5, Method 29, or Method 17 compliance tests could also be
performed periodically (e.g., semiannually) as a substitute for
monitoring scrubber operating parameters. The estimated costs
for three-run, EPA Method 5 and Method 29 compliance tests were
discussed in Section 6.2.1.1. No costs are available for a
three-run, EPA Method 17 compliance test.
6.2.1.3 Enhanced Monitoring for Gaseous Organic HAP's.
Control of gaseous organic HAP emissions from recovery furnaces
can be achieved by using NDCE recovery furnaces equipped with dry
ESP systems. Therefore, enhanced monitoring for recovery furnace
gaseous organic HAP emissions can.be achieved simply by
6-55
-------
confirming that the furnace is an NDCE recovery furnace with a
dry ESP system. No costs are associated with this enhanced
monitoring option.
If the recovery furnace is a DCE recovery furnace or an NDCE
recovery furnace equipped with a wet ESP system, the facility
could measure methanol emissions with a methanol GEM (e.g, a
fourier transform infrared, or FTIR, spectroscopy monitoring
system) . The capital and annual costs for an FTIR monitor have
been estimated to be approximately 160,000 and 71,500/yr,
respectively.39
6.2.1.4 Enhanced Monitoring for HC1. Hydrochloric acid
emissions can be measured directly using an HC1 GEM. An HC1 GEM
could be installed after the packed-bed scrubber to demonstrate
continuous compliance with an HC1 emission standard. An HC1 GEM
could also be used after the ESP for those recovery furnaces that
could comply with an HC1 emission limit without a packed-bed
scrubber. The capital and annual costs for an HC1 GEM were
determined using the EMTIC GEM cost program.
The capital cost from EMTIC to purchase and install an
extractive HC1 monitor (i.e., an HC1 GEM that extracts a sample
and transports the sample through a conditioning system and into
a gas analyzer) is approximately $126,900.37 The capital costs
include planning, selecting the type of equipment, providing
support facilities, PEC, installing and checking CEM's,
performance specification tests (certification), and preparing
the QA/QC plan required by Appendix F (40 CFR 60). The PEC costs
include the cost for a DAS for data reduction and reporting. A
performance specification for HC1 monitors is currently under
development for the CFR. However, one has already been developed
by the Northeast States for Coordinated Air Use Management, or
37
NESCAUM, and is being evaluated by EPA.
The direct annual costs from EMTIC equal $60,300/yr and
include costs for operating and maintenance, reporting and
recordkeeping, and annual review and update.37 Administrative,
insurance, property tax, and capital recovery costs were
estimated separately and added to the program costs. The
6-56
-------
administrative, insurance, and property tax costs were calculated
as 4 percent of the TCI, based on guidance in the OAQPS Control
Cost Manual.6 The capital recovery cost was calculated as a
product of a CRF and the TCI.6 The CRF was calculated assuming a
20-year equipment life and 7 percent interest.6 The total annual
cost for the HC1 CEM is approximately $77,400/yr.
The feasibility of using HC1 CEM's to demonstrate compliance
with an HC1 standard has not been determined. The low HC1
concentrations and high moisture content associated with the
recovery furnace flue gas may make the use of HC1 CEM's more
difficult. However, additional information is needed before an
HC1 CEM can be definitively ruled out for recovery furnaces.
Because HC1 emissions can be controlled with a packed-bed
scrubber, monitoring scrubber operating parameters is another
monitoring option being considered for those recovery furnaces
that comply with an HC1 emission limit using a packed-bed
scrubber. The scrubber operating parameters to be monitored are
scrubber liquid pH and scrubber liquid flow rate. The cost for a
pH monitoring system for scrubber water is approximately
$5,000.37 The cost to measure scrubber liquid flow rate was
discussed in Section 6.2.1.2 and ranges from $2,000 to $25,000,
depending on the devices' sensitivity and distance from the
control room.37 Scrubber liquid flow rate and pH levels would be
site-specific and would be based on the operating parameters
measured during a three-run, EPA Method 26A HC1 compliance test
that showed the facility to be in compliance with an HC1 emission
limit. Method 26A must be used when the measurement point is
downstream of a wet scrubber. The estimated cost for one three-
run, EPA Method 26 HC1 compliance test is $9,100 if performed
alone, or $600 if performed in conjunction with EPA Method 5,
Method 29, or Method 17 testing.40 The $600 would be in addition
to the cost of the PM or PM HAP testing.
Alternative enhanced monitoring options are also available
to demonstrate compliance for those recovery furnaces that could
comply with an HC1 emission limit without a packed-bed scrubber.
One option would require the facility to develop a monitoring
6-57
-------
plan that specifies operating parameters to be monitored.
Operating parameters would be site-specific and would be based on
the parameters measured during a three-run, EPA Method 26 or 26A
HC1 compliance test that showed the facility to be in compliance
with the applicable HC1 emission limit. Either Method 26 or 26A
could be used if the measurement point does not follow a wet
scrubber. The cost to implement a monitoring plan has not been
estimated. The estimated cost for a three-run EPA Method 26 HC1
compliance test would be the same as the estimated cost for a
three-run, EPA Method 26A HC1 compliance test, which was
discussed in the previous paragraph. A second option would
require periodic (e.g., annual) Method 26 or 26A HC1 compliance
tests to demonstrate compliance. The annual cost of EPA
Method 26 or 26A HC1 compliance testing is $9,100/yr if the test
is performed annually and without EPA Method 5, Method 29, or
Method 17 testing.40 The annual cost is $600/yr if performed
annually and in conjunction with EPA Method 5, Method 29, or
Method 17 testing.40 The $600 would be in addition to the annual
cost of the PM or PM HAP testing.
6.2.2 Black Liquor Oxidation Unit Enhanced Monitoring
This section presents the costs of the enhanced monitoring
options that can be used to demonstrate compliance with a total
gaseous organic HAP emission limit for DCE recovery furnace
systems (which include the BLO unit).
One control option presented for the BLO unit involves the
removal of this piece of equipment from the chemical recovery
process by converting a DCE recovery furnace system equipped with
'a wet ESP system to an NDCE recovery furnace equipped with a dry
ESP system. Demonstrating that this conversion has been
completed assures compliance with the applicable total gaseous
organic HAP emission limit. No costs are associated with this
enhanced monitoring option.
A second control option involves incineration of the BLO
emissions. Enhanced monitoring for BLO incineration could be
achieved simply by affirming that the BLO control equipment is in
place. No cost is associated with this enhanced monitoring
6-58
-------
option. Another enhanced monitoring option would be for the
facility to monitor the temperature of the power boiler or other
incineration device. The cost of a temperature monitor is
assumed to be zero because temperature monitoring is already
conducted by mills.
6.2.3 Smelt Dissolving Tank Enhanced Monitoring
This section presents the costs of the enhanced monitoring
options that can be used to demonstrate compliance with an SDT
emission limit for PM or PM HAP's.
6.2.3.1 Enhanced Monitoring for PM or PM HAP's Controlled
with a Wet Scrubber. Because the exhaust from the SDT wet
scrubber will have a high moisture content and will interfere
with the readings from an opacity monitor, the use of an opacity
monitor to demonstrate compliance with a PM or PM HAP emission
limit for SDT's may be inappropriate. Monitoring scrubber
operating parameters (i.e., pressure drop and scrubber liquid
flow rate) is an alternate enhanced monitoring option for showing
compliance with a PM or PM HAP emission limit for SDT's. The
pressure drop and liquid flow rate are indirect measurements of
the performance of the scrubber. The costs to monitor pressure
drop and scrubber liquid flow rate were discussed in
Section 6.2.1.2.
Pressure drop and scrubber liquid flow rate levels would be
site-specific and would be based on the operating parameters
measured during a three-run, EPA Method 5, Method 29, or
Method 17 compliance test that showed the facility to be in
compliance with the applicable PM or PM HAP emission limit.
Method 5, Method 29, or Method 17 compliance tests could also be
performed periodically (e.g., semiannually) as a substitute for
monitoring scrubber operating parameters. The estimated costs
for three-run, EPA Method 5 and Method 29 compliance tests were
discussed in Section 6.2.1.1. No costs are available for a
three-run, EPA Method 17 compliance test.
6-59
-------
6.2.4 Lime Kiln Enhanced Monitoring
The following sections present the costs of the enhanced
monitoring options that can be used to demonstrate compliance
with a lime kiln emission limit for PM or PM HAP's.
6.2.4.1 Enhanced Monitoring for PM or PM HAP's Controlled
with an ESP. Because opacity is the surrogate measurement that
best characterizes the level of lime kiln PM emissions,
installation of an opacity monitor is one option being considered
as a means of demonstrating compliance with a PM or PM HAP
emission limit for lime kilns controlled with ESP's. For those
lime kilns with a wet scrubber following the ESP, an opacity
monitor must be located after the ESP but prior to the scrubber.
The capital and annual costs for the opacity monitor were
discussed in Section 6.2.1.1. Method 5, Method 29, or Method 17
compliance tests could be performed periodically (e.g.,
semiannually) as a substitute for an opacity monitor. The
estimated costs for three-run, EPA Method 5 and Method 29
compliance tests were discussed in Section 6.2.1.1. No costs are
available for a three-run, EPA Method 17 compliance test.
Another option being considered is for the facility to
develop a monitoring plan that specifies ESP parameters to be
monitored. Operating parameters for the ESP would be site-
specific and would be based on the parameters measured during a
three-run, EPA Method 5, Method 29, or Method 17 compliance test
that showed the facility to be in compliance with the applicable
PM or PM HAP emission limit. The cost to implement a monitoring
plan has not been estimated. The estimated costs for three-run,
EPA Method 5 and Method 29 compliance tests were discussed in
Section 6.2.1.1. No costs are available for a three-run, EPA
Method 17 compliance test.
6.2.4.2 Enhanced Monitoring for PM or PM HAP's Controlled
with a Wet Scrubber. For those lime kilns that can comply with a
lime kiln PM or PM HAP emission limit with existing wet
scrubbers, the use of an opacity monitor to demonstrate
compliance with a PM or PM HAP emission limit may be
inappropriate. The exhaust from the lime kiln wet scrubber will
6-60
-------
have a high moisture content and will interfere with the readings
from an opacity monitor. Monitoring scrubber operating
parameters (i.e., pressure drop and scrubber liquid flow rate) is
an alternative enhanced monitoring option for showing compliance
with the applicable PM or PM HAP emission limit for lime kilns.
The pressure drop and liquid flow rate are indirect measurements
of the performance of the scrubber. The costs to monitor
pressure drop and scrubber liquid flow rate were discussed in
Section 6.2.1.2.
Pressure drop and scrubber liquid flow rate levels would be
site-specific and would be based on the operating parameters
measured during a three-run, EPA Method 5, Method 29, or
Method 17 compliance test that showed the facility to be in
compliance with the applicable PM or PM HAP emission limit.
Method 5, Method 29, or Method 17 compliance tests could also be
performed periodically (e.g., semiannually) as a substitute for
monitoring scrubber operating parameters. The estimated costs
for three-run, EPA Method 5 and Method 29 compliance tests were
discussed in Section 6.2.1.1. No costs are available for a
three-run, EPA Method 17 compliance test.
6-61
-------
o
fn
a D4
ri
w
I
IS
•a £
£5*
2
^1
S>
S §
^o
UJ
0 *»
o"
'E T3
O J2
to n
2 ^
2 CM
*s
c. §
J3 §
Gas llowtale-
ESP exit, m3/sec
@B% 02
1 5
li
_£<
1 05
c- a a
if -
5 = •*
o
1 |
3 £"
U,
f 1
Q- £•
- r§
•§ g 1
i ii
(O
V>
g
S
*r
*»
$165.000
S
f
01
v»
g
§
^
44
O
ro
g
g
C3
of
•
s
in
o
CM
in
in"
N
CO
^
o
1
*"
0
z
S
f1F-3a/Ga
$378,000
1
g
OJ
v>
1
o
s
1
4»
g
o"
5
S
O
CO
in
ca
^*
en
r-._
^•~
CN
S
§
R
o
o
m
0
r..
LI.
cc
o
g
o
f»-
r^
s
$134,000
o
o
^r
o>
w
g
o
s
n
£
o
C3
in
CO
£
Ol
rC
TT
ri
O)
0
CO
P3
C
c
g
S
es
a
u.
C
o"
«T
«»
g
I
$241,000
|
S
v>
I
1
g
m
CO
3
^"
CO
cc
g
to
200.000
o
o
m
to
a
d.
cc
$666.000
g
S
S
$165.000
g
0)
«»
|
S
1A
O
CO
i
CO
o
cc
of
*T
S
o
m
*tr
a
z
.0
^
03
2
e
u
cr
$1.200,000
I
en
$297.000
o
g
m
s
o
o
o
o
P3
V
fc"
O
n
«
o
i
r*^
i
§
]
2
c
~^
i£
c
L
V>
§
8.
V>
$428.000
i
s
v>
i
o
r*.
o"
w
g
s
CM
in
m"
w
CO
*r
OJ
o
o
c\
o
z
.0
c
RF-3a/6a
$378.000
§
g
CM
v>
0
o
^
g
f*
4*
g
O
o
o
cy
*^
o
CO
in
CO
«•
Cl
r*.
•v
CM
.S
o
N.
N
400.000
O
o
•i
(S
r^.
u.
c:
$630.000
g
f-.
r*
s
v"
C3
V>
o
o
'ff
v>
o
g
o
in
CO
s
8
in
CO
fx.
g
l^."
•"T
to
CT
S
T
700,000
a
^
c
~"
ca
a
LL
CC
o"
V
V5-
g
g
CD
W»
$241,000
§
sy
*»
g
CO*
44
o
CO
s
p;
r;
^
1
i
|
Q
c
0
C
u
cr
I
J
f
Ic
5
_o
I
.2
6-62
-------
OS
5*
Is
<
o
co
la
: E
gig
I 'Z #
£31
40,000
$3
$6
.730,
$1.140,
8
3
S
g
g
o
8
I
g
1
I
5
I
cf
g
O
§
O
to
I
5
8
o"
o
si
s
s
s
s °- fc »
in
in
O
o
z
LU
c
o
z
o
o
c
LU
o
o
z
LU
c
o
z
LU
o
o
LU
o
o
o
o
CM
cc
t»
c:
01 " o S
2 g o E
«? i _°-s
'3" 3"
£-§£§§
o o 3 ± o
H " "S. S "5
o S « "K =
£ S u g- g
" S a ° i
m .5 S «f •=
6-63
-------
Pi
o
fa
CO
P
o — -
H CO
• W
O CO
CO
O O
Pi O
H
A EH
go
Pi D
p-i D
§5 O
O Pi
U Oi
O CM
ca o
M H
Pi HH
Oi
CO CO
ia w
fa
-------
CO
Q
w
o
Q
O
Oi
CO
w
CO
EH
CO
co
w
u
?B
co
XI
CN
VD
is
ft; «
III
_ |- 8
"8 S E
2 s ^
8
5
3
I
£ £
•e-
E
I
•a
a
1 «=' I
6-65
-------
TABLE 6-3
SUMMARY OF ASSUMPTIONS USED IN ESP UPGRADE COSTS9
NDCE Recovery Furnace ESP Characteristics:
SSS=3=S=
Existing ESP:
• The existing ESP is assumed to contain 46
lanes on 10-inch centers (weighted wire design).
• The existing ESP is 3 fields long, with each
field 10 feet long and 30 feet tall.
• The existing ESP has a parallel drag scraper
arrangement (dry-bottom). ^
New ESP:
• The ESP is rebuilt with 42 lanes on 11-inch
centers (rigid electrode design).
• The size will be adequate for 99.4 percent PM
removal, yielding an outlet PM residual of 0.044
gr/dscf at 8 percent oxygen.
• The ESP will be 3 fields long, with each field
10 feet long and 30 feet high.
DCE Recovery Furnace ESP Characteristics:
Existing ESP:
• The existing ESP is assumed to contain 50
lanes on 10-inch centers (weighted wire design).
• The existing ESP is 3 fields long, with each
field 10 feet long and 30 feet tall.
• The existing ESP has a steel shell design (wet-
bottom).
New ESP:
• The ESP is rebuilt with 46 lanes on 11-inch
centers (rigid electrode design).
• The size will be adequate for 98.8 percent PM
removal, yielding an outlet PM residual of 0.044
gr/dscf at 8 percent oxygen.
• The ESP will be 3 fields long, with each field
10 feet long and 32 feet high.
Modifications:
• There is single-chamber operation while the other chamber is washed down, the roof removed, and
the internals gutted.
• The recovery furnace is taken off line; work continues on rebuild; one chamber is finished.
• The recovery furnace is at partial load through rebuilt chamber; the other chamber is finished.
Other Assumptions:
« The existing casing is in acceptable condition; only minor replating is needed.
• Wet- and dry-bottoms are reused "as is," except for minor repairs. Transformers, controls, and
inlet and outlet ductwork are also reused.
The roof insulation is replaced, but all other insulation is acceptable.
• The site location is in the southeast United States. As such, all field labor is based on merit shop,
1993 wage rates and work rules.
• No removal of asbestos or polychlorinated biphenyls (PCB's) is required.
• The scrap is taken to a site within the plant gates.
• There are no perforated plates. • The ESP is at grade.
• The ductwork and stack do not obstruct crane • There is no low voltage.
mobility.
6-66
-------
o" t
g^
l^fc
D. 5 -5"
3 |*
h-" w
08 ^
<-^
1*
Is
LU
P«
1 -D
D
Gas flowrate-
ESP exit, m3/sec
@8%02
Equivalent
ADMP/d
Black liquor
firing rate,
kg BLS/d
as ®
c Q-
S*
D. O
"5 g-
Q. £•
>. «
- 0 g
l§i
5 £ 5
§
r».
o
3
8
o
o
TJ-
*»
g
CO
s
8
en
«»
w
r-~
•^r
s
o
03
TO
I
UJ
o
Q
5
a
%
LL.
CC
8
CM
§
§
«»
|
CO
(O
<«•
8
o
in
£
$1,700.000
t^
CD
CO
O
1 ,200,000
ill
O
a
m
RF-2a/5a
8
o
CO
s
o
g
CO
ct;
§
CM
(O
t*
O
8
I
$1.570,000
o
1,200,000
UJ
o
o
55
a
en
uL
cc
§
r-.
o
8
o
•*r
t»
|
CO
Tf
w
$1,200,000
•*r
g
o
in
T
700,000
LU
O
a
i
a
%
LL.
CC
8
o
§
o
8
§
«»
o
8
co
ca
v>
g
o
8
$1,700,000
o
3
1,200,000
UJ
O
a
3
3
1L.
cc
g
o
1
o
§
CO
£
V)
I
TT
CO
*»
o
s
s
*»
$2,120,000
i^
CO
T
CM
O
O
CM
1,800,000
UJ
O
D
3
c
LL
EC
g
O
8
S
g
in
w
g
v
8
<»•
<&
8
T
0)
£
$1,100,000
r-.
Tf
§
8
TT
700,000
m
O
O
3
c
g
LL.
a:
o
g
«»
8
5
CM
V*
8
8
$1,570,000
i^
oo
ID
0
g
1,200,000
UJ
o
o
3
c
3
c
o>
iL
tr
en
.2
0
3
O
O
o
•8
i
5
lotnotes
j^
o
5
Table 6-
^
•£
o
a:
J3
TT
(£>
O
1
_C
§
D)
(0
ted English unit value
rom the calcule
were converted
> in this table
JU
c
o
§
'D
D"
O
0
~
T5
S
6-67
-------
rt
w
o w
tri CO
CO
JgJ o
QI I "^
J a
o o
pj
S o
O D
CJ Q
gi
M 0<
Pg
§
g
O
*g
P4 Q
W J
U
H X
W
W-'
J
D CO
Q H
W O
W rt!
o s
CO K
Ut b
°><
CO Dd
tn ca
co >
o o
o o
w
«
^.j
W H
CO Q
H O
J S
CD
s PC!
w o
^«&4
•° ?i
Tjt O
i co
VD Q
W C!
^ O
g
,- "*
E-< ^
o
§•11
o g «i
•i
I
LU
D'S
3
O
a o
^"a?
5 «
fe ©
§ «?3i
^g O (/;
5 = '
II
f I
t s
>• 10
s ?
2.
g
n
O)
o
s *
a-5 §s
•I L •» e>
is
•». CM
w* CO ^
! S x co"
H §|l
L — o •
Olin-SSgg-S-5
^— _ W JS O g Q D.3
OLU©.£«- -oc
«h-fc.-a)Sgjoc
o .oS . 2 * -o
S S
a s'
6-68
-------
os
^ t
3 £S
tft
H=D
"*" >
<*
f -*
•^
o .>•
UJ
p_~ &
1 -o
s
. 8
f is
l-s" »
Ife®
UJ
'C T,
o ^£
1|
§ a> TO
tr QI co
^ O) CD
« 1 s
1*
Q. CD
d I
ill
2 *- —
8
O
CM
g
o
8
g
g
co
S
8
•a-
o>
8
o
8
S
o
^
CO
0>
g
CO
o
o
o
I
til
I
5
CD
T
s
E-
§
in
O)
8
I
o
co
to
o
o
in
TO
8
o
o
CO
CD
g
8
to
1,200,000
UJ
0
o
. s
o
CM
Lu
a:
o
8
in
r-.
8
8
CM
1
g
CM
S
g
to
s
8
o
o
fc
u
s
CM
O
8
1 ,800,000
LU
O
Q
5
m
CO
11
8
to
5
8
1
8
,_T
3
8
r-
5
o
o
**
o
CM
S
o
CO
CM
O
8
8
LU
O
Q
m
o
u.
EC
O
O
o
^-
8
o
8
I
g
CM
R
O)
8
o
§
o
CO
^
co"
o
CO
CO
o
8
I
LU
o
Q
m
o
CO
cc
8
•«•
t»
o
o
o
CO
g
^r
8
g
o
I
8
g
in
o
CO
CO
to
§
to
1,200,000
UJ
O
Q
m
a
9
LL.
EC
§
CVJ
8
o
o
o
CO
CO
o
o
CO
R
en
8
8
S
o
CO
^.
to
co
s
8
o
1
Q.
s
1
(0
^-
3
LL.
EC
§
1/5
S
8
o
o
R
CO*
%
g
o
8
8
8
to
o
CO
s
co
1,200,000
D_
CO
LU
S
c
to
CM
U.
EC
8
in
S
8
o
I
R
CO
CM'
CO
R
in
o
8
g
CM
g
CO
s
o
8.
o"
§
EL.
(t)
LU
C.
a
to
*
LL.
DC
§
to
to
8
o
TO
g
^
S
O
O
T-
0
o!
«»
g
OJ
g
CM
8
o
o"
o
UJ
8
3
i2
LL.
EC
|
^"
8
8
o
1
R
CO
CM
S
O
o
s
o
8
g
g
^
s
o
in
o"
UJ
o
O
c
o
LL.
EC
8
TT
CO
8
-------
cu
in
ll
1 ts.
K'g
ts
UJ
5 §
|fg
J;s >P
g S
Sgj©
O w
_
1|
tr<
Ul
jr 1 o
P-
S
5 *
u.
a g
a. j
r§
III
223
|
CM'
o
§
CD
CO
g
to
v>
g
en
160.000
I»
0
g
to"
c
o
o
I
u
c
ca
i
L
CC
o
1
i
en
g
1
*
in
CO
v>
g
o
g
to
p
o
o
r-T
m
CO
g
r-
c
o
CM
LU
C
c
!£
CM
L
CC
§
in
5
|
1
g
CO
g
tn
o
o
ft
CM
VJ-
O
8
in
m
o
o
1
r
UJ
O
C
e
^
r
L
CC
g
1
§
n"
s
s
I
s
o
g
o>
u
o
g
a
g
C\
O
o
c
c
G>
U
Ll
cc
§
g
1
!
g
CO
E
g
en
0
g
I
v>
o
g
cr
o
o
o
o
o
g
'-
Ul
o
a
CO
L
CC
i
5
§
CO
g
s
o
g
3
O
O
O
o
CO
in
v>
o
p;
g
1
CM
Ul
C.
a
en
u.
a:
|
1
g
!
o
o
to
g
0)
o
g
o
(O
V*
o
g
CO
o
o
o
in
c
o
o
•*-
LU
C
C
s
3
<
L
CC
1
g
o
g
en
g
070,000
8
O
O
O
ir
tn
?
§
°
Ul
O
c
s
e
c
9
L
cc
|
to
m
g
CO
o
o
to
CO
1
5
5
«
O
0
c
o
c
o
o
UJ
o
o
s
=
L
cc
1
CM
§
CM
m
g
CM
g
070,000
v>
S
g
CO
c
g
o
*~
Ul
c
o
1.
=
C
U.
cc
1
§
1
CO
u
g
CM
CO
|
n
.530.000
«>
c,
g
ir
o
o
o
700.000
CM
UJ
O
Q
j
3
C
L
C
.
1 1" g
II 1
ll 1
" w ^ E
E ". c •§
2-E | §.
ftp-
5 ^ ^- m
I s ^- T
llff
f | 5 §
|ii|
0 ° ° |
g" g ^ ci
O II °- 1
S ^" Q tfT
E i 1 s i
•|o~.fg
11|1|
^ °~ u. ^ u,
SS 2
«-
6-70
-------
W
•a .
E
c xi
®
"5
g 6 -O
.g-S"
o> m
I St
1
5>
Ii
S
8
g
o
1
§
I
g
o
s
s
u
I
in
g
S
o
g
o
g
o
g
cT
UJ
O
o
Ul
o
o
UJ
o
a
I
Ul
o
o
I
1
M
f
I
O
CO
o
UJ
O
Q
S
a
0
a
o
o
§
0
UJ
0
o
CM
CC
cr
09
•g
«S
6-71
-------
w ca
CO Q
H O
I*
w o
•
u> o
i CO
vo D
•^
W Crf
i4 O
•a _
lclty,
(d)
+ tr «r
% 3
s
s
8"
tf
£1
g
18-S
ll§
IS®1
U aj
III
8
g
g
g
g
o
o
t^ OJ
CO
N
5
§
I
6-72
-------
o
.•S
o
-a
<
SSH
8
I
Z
-------
ri
CO
O W
E-< w
CO
§3
g§
01 H
OE>
U Q
g§
S3
a D
CU
g
CD
pa
H
P< Q
co 5
W J
cj
M S
H
H —
J
D CO
Q H
W U
ffi
o o
U CJ
w
Pi
*
— J
w w
CO Q
H O
O
S f*
w o
^-t±4
•Q ft
l> CJ
• CO
U) Q
H Cd
J O
g
,- •*
f <*
o
Is
I
$2
TCI+
productio
losses. $
-
lm
.
O2
Gas l
ESP
@
qulvffil
ADTP/
g-Ig
HS
tl
$2
$2
s
$3
n s
$2.
$3
z
22
s
ADTf
uled
d d
91/12
ff
5S x § § -o
- ™ - . 0
f- f,
£
5- S -
S-5-
< g.
f
|-g
i
.
s
-sis Ift
<-T
Si-SStJ x Erf'-S s
Si-i is x-jdStu
17 " t •§ i I Q. s
.
6-74
-------
H
I*
If *
I—" w
5"
1*
UJ
i-
o
•= —
0 - |
g CJ P3
0 ^2 •§.
£ •- 53
E
— S CM
§ CL
£ °- i
0 CO S
_S uj «
•S « CM
| E 0
0 ~ g
» ® ©
So.
C CO
UJ
C ^i
o 5
H
i s ^
cr « CC
=: *• — J
j: at to
5 = *
0
16
U_
Q. 0
cS i
>. «
•§ o 5
$117.000
&
g
a
8
I
0
I
CO
CM
**
^.
S
§
CO
1
c.
z
5
5
5
cc
$212.000
$136.000
8
I
g
s
g
I
CD
8
CM"
CO
CO
o
CO
CO
I
1
CD
tn
~
cc
$304.000
$195,000
g
*
g
S
p
I
CO
CO
co"
n
o
o_
I
z
3
g
r:
CC
o
g
in
g
en"
i?,
g
R
I
CD
CM
CM
CO
in
o
o
o
o
g
o
o
5
S
cc
$134,000
I
&
g
a
§
to"
z
o
CO
CO
n
*~
^r
S
§
o
g
g
o
o
5
«
cr
I
i
s
e
g
i
i
o"
(O
to
CD
in
CM-
CO
§
.200,000
O
a
m
jQ
O>
cr
$117,000
in
fc
1
in"
8
to"
v>
s
I
CO
CM
C3
T-
^.
8
1
700,000
0
z
s
.a
T
cc
$212.000
$136,000
S
I
g
*
g
o
to
r-.
(O
to
in
of
g
.200,000
u
o
z
s
jg
~
cc
$304,000
$195,000
g
co"
g
H
1
I
to
to
o"
n
CM
!-
,800.000
c
z
£
£
f.
CC
o
o
I
i
g
in
S
8
s
!'
CC
CM
CM
CO
tn
CM
g
o
8
o
o
c
.a
tr
$134.000
o
g
S
g
in
&
p
to"
e
to
to
0
•»—
^r
O)
o
1
o
o
|
«
cc
$241.000
$165.000
g
to
g
of
CM
V*
1
1
N.
CD
CO
m
CM"
CO
1
,200,000
o
a
c
13
cn
cc
S
4
s
•o
I
•o
€
s
o
6-75
-------
CO
§
•a _
351
8s 5
: E
5?*ii
2 S
fs§
ti.
l §
s
T-ll N.
91 5
65.
OJ
_EJL
•= .£• =-
^* ^- en
w •= £
•C Tf ffi
CD* W S
.£ c
•— Q
£t §!t
X !£, T3 O 03
£ o
X
^r
i f s -
xgfSS^X-r
||1°°° = ^
•a. E
E «e
PS
821
a g s * <•
og.g 5 =>•
X — OI ^
i I "
g
II O II
t " CJ
§ £•<
S a- u i g
S .2 3r ° — S
5 S '3 o S ^
3 n
« VJ - *" ^-
5 w g ® S
S 5 s 1 « .E
S tn s s
«*" I ^
^ •- § "J
» S 1 O
i «1
1 § I i i
ESS E ^
a 2 uj a c
LU 2 O ol 5
O .E g O
8S_^al2oS
5J^££3=58
SS S s S s s
6-76
-------
VD
E-i o
•a &•
IH
l 5
53
en E
£ S
S —
its
Is
1°
•
I
I
I
tf
|
I
a
es
€>
_3
>
I
6-77
-------
I
fs|
It
TCI+
production
losses, $
|Sl
-I «
i s
5. S
s s &
its
«» n.
Gas llowialo
ESP exit, acl
% O2
ivale
TP/d
liquor
(lilng tale.
Ib BLSW
lack
lilng
*
,450,
c-5 S f-
a. ~
-------
.
ga
— ?••
to s;
-tS ® t
0 £
H" ._
5s*
O ^9-
1—
o
i CD
CD -52.
•s co~ CM
g E 0
O ** w"
^ X CO
So!®
O CO
LU
CD
EC O-
— .2 S1
O Q- D-
0 0 CO
LU
.1 1
CO Q_
CD LL
C -T)
o> 4J.
Tg Q-
> 2
D- °
LU "*
I « 2.
ET £S CO
JC D) CC
8 •§ »
CD —
Q. CD
•3 §-
D- 3?
— co S
•s § §
o o fc
S £ *2
g
to"
1
of
!»
O
§
co
U
o
o
o
CD
o>
in
to-
^.
0?
F
o
JS
o
-o
_0
0
CO
o
CO
o
o
o
o
r-.
5
^
^
4
u.
DC
g
O
8
o
o
CM
8"
0
o
o
i
g
O)
CO
CO
F
0
1
T3
O
2
I
0
CO
CO
o
o
o
g
C\J
m
^
"S
in
LU
DC
o
CO
CO
o
8
s"
t&
o
o
s
§
.
v>
CO
Si
F
o
o
-o
o
o
CO
o
°
c
o
o
co_
CO
J2
C£
U.
DC
O
o
in
1
1
en"
CO
w-
o
§
»
o
8
to
in
,3.
S
E
o
|
T3
E
Q
CD
O
in
°
I
— ^
XI
0
U.
CC
o
8
1
o
en"
en
8
i
1
0>
^~
CO
CO
CO
E
0
1
•o
1
1
0
CC
c:
g
C\J
S
—^
XI
a
c
in
u.
oc
0
o
o
CO
CO
I
w-
o
o
CM
o
o
o
CO
0_
CO
E
o
o
•o
E
o
o
XI
CD
o
2-
o
o
o
o
o
co
c
XI
CC
CC
LL
DC
tn
CD
CO
CO
3
CO
CD
•5
CC
ci
CO
_CD
J3
O
I
O)
"5)
c
LU
T3
S
JS
3
_O
CO
u
c
£
I
•o
CD
CD
O
o
CD
I
JS
2
:£ .2
c B
s 3
i S
S CD
1
-g I
6-79
-------
w
o
A
w
p
tf
o
FK
p
CO
S
&
o
H
°3
Pi
M
CJrt
co
cu w
C7) CJ
w <:
o t>
pa
i
s t
's- I
o 8
in
erf
0
o
o
CD
8
05
O)
0
o
o
CD"
(0
$124,000
o
o
o
co"
CO
^
o
o
oa
cn"
0)
08 r
CO
a
8
CO
•W-
cvf
•fl-
•w-
O
O
CD
co"
CM
o
o
0
•<*"
CO
•w-
5
o
o
o
CD"
cn
in
o
o
o
cn"
S
o
o
o
o"
CO
o
o
o
o
CD"
cn
in
ffi
W
H
ca
03
5 .±T ^P
a 8»
cn Q_
offi
o
S
co"
©
in
CO
o
in
o
o
o
eo"
o
o
o
N."
m
CO
c '-g
O CL
O °
O
=c3 -^l ^
cn
CO
jj Ji
co P-.
| SI
at
.5" 2 co
CD
CO
E
o
o
CO
in
8
CM
E
S
?
X)
o
$
o
o
o
o
XI
in
03
co"
o
o
03
O3
CD
CD
24
$
CM"
o
o
o
o"
CO
o_
T—
1f>
o
o
o
in"
in
E
o
tt
o
XI
o
S
XI
i
o
o
co_
CM
03
co"
CO
cn
cn
O f—. -^ '
= f~ 03 o3
(0
vo
W
D- 51i
_ 0
-§ §
0 0
m
CD
CD
§
CO
in
»-i
cn
.52 —
CO
TeT
co
-5 ••
— E Q.
CO
CO
= 0-0
co" o
< O =
6-80
-------
S E
II
UJ
g
I
8
i*
8
O c C
III
O X CD
UJ CO "D
O CO
UJ
.> 5
£S
**
_g- o w
j«: O) CD
Q. 0
"5 §•
0. .&•
s
8
8"
in
8
8
o
8"
o
o*
o
m
8
o
I
to
o
in"
o
o"
S
S
8
o
o"
o
in
CD"
o
o
o
o"
in
CVJ
en"
co
<»
o
S
5
o
CVl
S
8
m
c
u.
0-
o
o
tn
,000
S
8
r-
0)
I
8
i
in"
8
S
Cvl
en
in
CVJ
8
8
CVJ_
a:
o
iL
DC
a
T>
O
o
i
JD
a
o
1
=
D)
c
UJ
o
o
o
£
S
|
o
6-81
-------
H- «»
o
rr
H
•* & c
111
d>
T>
2
E*»
39.
o c
# °
c Q.
UJ S
T3
o
I -o
$849.
8
$439,000
$596.
8
8
8
S
£ ~3 CM
goo
CO
©
0
8
go.
'a Q
§ S
8
00
1
m
$8
•»
8
8"
±L
ID
O
S
It
I?
w FA
^ C>
en S
«A P5
g S8
— .£ CD •*"**
g^ ^i*
oco x^|o
03 O
S CD
(fl —
II
g!
o p
^ .
< S £
2" § §
g 1 §
•S _Q < g 0>UJ
0 o S'f •§ g
c o n o S S
O O, •- T3 p S
11 f 111
E ^j s= c: o c*
•S Scg i o o
g 8 5 o 8 I
S g -§ 8 te- §
S E ° §= +
8" 5 III
6-82
-------
O
tr>
S
CM
W
5 **
Q-'S »
*§
II
o
I 1 §
|||-
_E
"£ ~o
8c!®
a ta
li
3 O
.5T<
§ S
a ~
5 =
o
g
o"
8
5
8
8
O
o"
S
o
*r
e>
8
"
S
o
o
S
$
o
S
o
S
8
en
^
5
8
S
o"
S
o
S
3
1
1
1
1
s
1
cc
I
i
£
E
0>
g
6-83
-------
»"M
S
I
sll *
°-"§ s =?
5.1 *»
s
§5
01
•8 -2
9 IS
g
g
111.
c 5 c
III
Ju
0
6.1
.2 ^3 CM
goo
Ifg
31*
is
~a o.
•IS
LU
| tf -0
cr ^5 35
— ^ — J
JC Oi rr
° *C ^
O. 0
If
ill
1 11
o_
s
in
CM
C)
1
o
o
o
g
5
.c
t
1 =
1
tD
v>
tn
CM
g
o
CO
I
1
g
*~
5
^
1
CO
d
1
8
8
tn
o
g
£
§
•*•
g
o
o
w
m
1
cc
o
g"
in
CM
119,000
c
In
O
o
o
o
_c
IJ3
ft.
n:
1
(0
v>
S
g
O
CO
Ui
c
g
o_
g
*"
^c
JD
1
C)
1
v>
tn
CM
| 357.000
O)
o
g
o"
CM
C
i
"S
o>
" '
o = —.
iiiiic
•r- E
6-84
-------
u
o
•is
$• E
I!
8;f «
UJ D)
Q.
s
£ c" c"
.t; o o
E •= '•=
.o
-— -^
S.-s
"n
£ S -S
-•§ fe
O CO
UJ
g. D
UJ **•
§ S
rr «
= *-
m "=
Q. O
"5 9:
Q. 2?
23
8
o
o"
o
co
f
cn"
o
8
8
§
8
o
8
8
o
I
in"
o
to
in
•tf
o
8
o"
CO
co"
o
o"
o
in
o
S
S
CM
s
CM
cn
JD
CO
I
O)
S
S3
CM
ID'
in
CM
(D
in
o
8
8
,200,00
8
o
8
8
o
8"
,200,000
I
I
u.
EC
S
U_
(T
cn
CE
CD
C
UJ
•g
o
o
o
_c
I
g-
a
|
"S"
6-85
-------
S ^
£• co §
ills
i.g 8
a
a. "5 -3-
ffi|»
9 IS
5 g «»
I
1 «»
^
o
O
g c" cf
.H o o
ill-
0 Q. g
UJ ° "O
co"
e
"E TJ
Q
! E
Us
il'S
S g: ©
O UJ
"c _
(D ^
11
cr «
UJ
>_
or o J
>* D) n
o •- !o
CO —
CL
3
Q. .£
^«
— o
•S §
Is-
g"
R
o
S*
S
c
o
8
1
S
CM
C
o
in
CM
g
o
en
a
f^.
r-
L
cr
3,400.000
**
0
in
8
o
8
8"
«
o
o
o
o
o
in
.•"
V*
8
1
CM
CO
0
1
o
in
c
.0
CO
CO
L
cr
9,400.000
**
R
en"
S
g
o
g
cq_
**
P
8"
f^.
IT.
8
1
CO
c
o
It
o
S
CM
357,000
o
u
o
o
o
r-
CM
C
e
(
i
EC
en"
R
en
g
o
S
**
g
c
cc
g
0
CO
CO
c
o
CO
119.000
o
o
co
g
o
en
3
^
i
r-
i
1
13,400.000
V5-
S
co"
1
o
s
**
8
1
s
c
§
^f
V5-
C
If
cc
o
o
o
CO
c
o
o
u
o
o
o
o
s
CO
CO
I
19.400,000
*»
o
o
o
en"
I
§
o
CO
CO
CM
v>
8
g
in
c
g
tc
v>
c
• S
c\
a
357.000
1
o
o
e>_
o
CM"
3
^
c
1
I
•
4^ o
o S^
f Hs
•s- g *»
^ «rt CO
003
S co
S o
S £2-
UJ O
o
o
i 5 o
S S §
£ S c
sas
sS?"^ffi
cliffll
O O *- "D C "^
-tS O O) w •^-, (-i
"o o c o c -V
I 10^ ;= ^- £ £>
m O m O *"*
ri ^ O O •->
^J I| CO 03 C? U
I "" m g§ "S
to o * tr co eo
fill!?
N == o n o S:
£ | Q H o W
o> •<-*
=- 8 •
u UJ
•
2 -s
§
£ o c
TT o o
Q- '§>
CO 0
.S -s: u —"
S UJ >
•§ 1 » « E
en 5
> 9-
Mi
m S3 "D
< -I
II8IB
6-86
-------
PJ C5
gl 8*»
"-So
§«
g
•g «
lr
-1 8
si = c"
.2 o O
o "5
. §
Us
I**
s£®
11
3 O
,fT<
§ 23=
.g- g co
jc pirn
5 = °
II
§ I
o
w
v>
o
g
o
9
1
1
i
g
o
I
o
g
s
<0
O
£
=
c?
,200,00
G>
iL
ir
i
.1
§•
o
o
I
6-87
-------
s
i
gag
S
tvl
g
OJ
S
f "Iffi-
"5 o
8 2
VDO
w
t?m
to 5 £j.
2.2..2.S
6-88
-------
tn
CO
o
co
c
II
m
1
CD
EC
CD
T3
O
JQ
9
u.
QC
co
9
u_
IX
JO
GO
uL
EC
CO
CO
ul
rr
X)
LL.
IT
CO
ul
rr
to
CD
m
E
CO
CO
Q
c.
£33
tn
Model furnace/ESP de
o
o
o
I
o
o
0
o"
o
CM_
1
0
o
o
o"
o
r--
o
o
o
o"
o
rr
o
o
o
o"
o
rr
T3
eg
m
TO
OS
cn
o
3
.cr
o
CO
m
Equivalent ADMP/d
88
co co
o o
Scvi
co
o o
co in
CO rr
§8
CO rr
0 0
CO C-
CM CM
o o
CO t-~
CM CM
Bleached pulp
Unbleached pulp
CO
CD
CO
CO
co
0)
co
O)
CM
CD'
in
CM
CD"
in
o
CD
tn
CO
E
1
CL
to
UJ
CD
2
_O
CO
O
in
co
(O
CD
in
co
co
co
in
CO
to
CD
ESP SCA, m2/(m3/sec
Baseline
O CM
in i-
0 CM
in T-
0 CM
in T-
O CM
in i-
O CM
in T-
O CM
E i
o to
to "D
o co
•^ q
to co
o o
c 'c
o o
o o
Q- D_
CO
co_
1
CO
°l
cn
CM
to"
cn
h-
rr"
CO
CO
CD
co"
Cvj
co"
CD
CO ^
CD ^=
•~ CD
_co to
£g
CO
Ul
cp_ co_
CO rr
8 S
S«
cn ^
81
CD ZI
O CM
O CM
cn co
m" CD"
O CM
O CM
cn co
in" CD"
F 5
to "O
5. oi
o co
T- O
o cr>
tn tn
S 8
"H "c
o o
O 0
D_ Q-
co
in
CO
in
CO
in
CO
ir-
in
co
in
co
CO
Operating days per ye
§
co"
of
co"
CM
rr
co"
CM
co"
rr
CM
rr
co"
CO
CD
h.
CD
Q.
S2
3
o
cn
2
CD
Q.
0
S
S
_CD
JO
CO
I
D3
to
CD
"5
'E
cn
UJ
•D
CD
CO
o
CD
E
o
v_
•o
CD
CD
C
o
o
CD
I
_03
CO
CO
.2
cr
CD
O
•c
"CD
2
"5"
6-89
-------
CO
H
>4
i
W
.Q
in
H
u>
en
8
rri
c
1 Recovery F
a
T
O
n
>
L™
LL
ra
r>
C
o
t
LJL
CO
CO
uL
CC
.0
u_
LL
CO
uL
CC
en
CD
CO
o"
Cj
h"*
CM
0
8
8
CM
8
^1
1
T""
O
I
"
0
8
o"
o
en
CO 13
QJ i3
CJ EQ
en
CD
•a
a
V.
u
"S
c
re
i_
"E
c
Black liquor firing rate, 1
o o
in o
r^ \ji
O 0
in o
h- en
88
-C- m
0 0
CM 0
-a- in
0 0
in o
CM CO
0 0
in o
CM CO
Equivalent ADTP/d
Bleached pulp
Unbleached pulp
8
o
r-~
in
co
o
8
te
CO
o
o
o
co"
en
i—
co"
en
'
o
o
0
en"
of
E
o
ra
CD
Q.
CO
UJ
1
"ro
£
CD
§
o
8
o
CM
CO
O
CM
CO
O
CM
CO
Li-
en
CD
CD
D5
CO
T5
m"
Exhaust gas temperatu
CM
CO
8
8
CM
CO
CM
CO
CM
CO
0
^1
"x
CD
Moisture content-ESP
8 8 £
•t in CD
CO CO h-
co co i-
co in CD
co co t*-
CO CO t-
TT m CD
co co r^
CO CO i—
co in CD
co co r^
CO CO T-
•^r in CD
CO CO N
CO CO T-
co in CD
0 0
en en
"D "D
t "^
E C3) en
"o •* m
« S 5
8 o o
O i >
- en en
C. O O
tt m § t:
< 1 8 8
(Tl Q_ Q.
CO
LU
O O O
o o m
•<3-" o" o"
f\1
'
o o in
O ^ i-
CO 0 0
i— O) CM
T- i- CM
rt 0 0
1-Ss
g 0 CM
^" l
8§|
Q r\T
h- N CO
CD CD CO
in T co
i-1 CO" CO"
in CD t^-
fv. fv. CO
CD CD CO
CD •*_ C0_
en" co" co"
co CD r-
73 o
en en
"D "O
D) O)
-a- in
So
a
-------
TABLE 6-16. SCENARIO 1: ANNUAL COSTS OF LOW-ODOR CONVERSION
(INCLUDING ESP UPGRADE TO CONTROL PM TO 0.10 G/DSCM
[0.044 GR/DSCF]) FOR MODEL DCE RECOVERY FURNACES (EXCLUDING
ANNUALIZED PULP PRODUCTION LOSSES)a
Costs
Total Capital Investment (TOO
Direct Annual Costs (DAC)
Steam production credits (b)
Natural gas
Fuel oil
BLO operating cost savings (c)
Concentrator steam costs (d)
ESP operating electricity costs (e)
Total DAC
Natural gas
Fuel oil
Indirect Annual Costs (IAC)
Administrative, taxes, and insurance (t)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Capital recovery (g)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Total tAC
Total Annual Costs
Natural gas
Fuel oil
FIF-7a
$9,410,000
(8758,000)
($1,190,000)
(£103,000)
$57,600
S23.3OO
(5780,000)
($1,210,000)
$324,000
835,200
$17,600
5377,000
8764,000
$103,000
851,300
$91B.OOO
£1,300,000
£520,000
£90,OOO
Model Recovery Furnaces
RF-7b
£9,410,000
($758,000)
(£1,190,000)
($103,000)
£57,800
£11,700
(5792,000)
(£1.220,000)
$324,000
£35.200
$17,600
£377,000
£764,000
$103,000
£51,300
$918.000
£1.300,000
5508.OOO
£80,OOO
RF-8a
£12,800,000
($1.260,000)
(£1,990,000)
(£172,000)
S96.3OO
£38,800
($1.300,000)
(£2.030.OOO)
$440,000
£48,OOO
$23,800
£512,000
£1.040,000
£140,000
£69,700
$1,250,000
£1.760,000
£46O,OOO
(£270. DOO)
RF-Sb
£12.800,000
(£1.260,000)
(£1,990,000)
($172,000)
£96.300
S19.4OO
(£1,320,000)
(£2,050.000)
£440,000
£48,000
£23.6OO
£512,000
£1,040,000
$140.000
£69.700
£1.250.000
£1.760.000
£440.000
(£290.000)
£18,200.000
(£2.270,000)
(£3,580.000)
(£309.000)
£173,000
£70,000
(£2,340.000)
(£3.650.000)
£628,000
$68.000
£34.0OO
£730.000
$1,480.000
5199,000
£99.2OO
£1,780.000
£2.510.000
£170,000
(51,140.000)
£18.200,000
(£2,270,000)
(£3,580,000)
($309.000)
£173,000
£35,000
(£2,370,000)
(£3,680.000)
£626,000
£68.000
£34,OOO
£730,000
£1,480,000
$199,000
$99,200
81,780,000
£2,510.000
£140.000
(£1,170,000)
(a) All costs in £1991. Numbers in parentheses represent cost savings.
(b) Steam production credit calculations (either natural gas or fuel oil):
Natural gas credit = (model BLS firing rate) x (6,000 Btu/lb) x (1 d/24 hr) x (66^56% thermal efficiciency)/(S5% power boiler efficiciency) x
£1 fan ,024 Btu) x (8,424 hr/yr) x (£3.48/1,000 ft3)
Fue) oil credit =-- (model SLS firing rate) x (6,000 Btu/lb) x (1 d/24 hr) x (66%.56% thermal efficiency)/(85% power boiler efficiency) x
(1 gal/144.000 Bto)x (8.424 hrftri} K ($0.77/gal) "if n *
(c) BLO operating cost savings = $251,900/yr x (model BLS firing rate/2,200.000 Ib BLS/d)
(d) Concentrator steam costs = S54,OOO/yr x (model BLS firing rate/2,400,000 Ib BLS/d)
(e) ESP operating electricity cost = 0.00194 x increase in ESP plate area x 8,424 hr/yr x SO.Oe/kWh
(f) Administrative, taxes, and insurance «= O.O4 x TCI
(g) For low-odor conversion, capital recovery cost = O.O944 x TCI (based on 20^r remaining service life and 7% interest)
For ESP upgrade and wet- to dry-bottom ESP conversion, capital recovery cost = 0.1169 x TCI (based on 13.5-yr ESP life and 7% interest)
6-91
-------
TABLE 6-17 SCENARIO 2: ANNUAL COSTS OF LOW-ODOR CONVERSION
(INCLUDING ESP UPGRADE TO CONTROL PM TO 0.10 G/DSCM
[0 044 GR/DSCF]) FOR MODEL DCE RECOVERY FURNACES (INCLUDING
BLEACHED PULP PRODUCTION LOSSES)a
Costs
Total Cacrtal Investment (TCI)
Direct Annual Costs (DAC)
Steam production credits (b)
Natural gas
Fuel oil
BUD operating cost savings (c)
Concentrator steam costs (d)
ESP operating electricity costs (d)
Total OAC
Natural gas
Fuel oil
Indirect Annual Costs (1AC)
Administrative, taxes, and insurance (e)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Capital recovery (1)
Low-odor conversion
ESP upgrade
Wet- to dry-Bottom ESP conversion
Production losses
Tota!
Total IAC
Total Annual Costs
Natural gas
Fuel oil _^__^_
59,900,000
($758,000)
(S1.190.000)
($103,000)
$57,800
$23,300
(5760,000)
(51. 21 0.000)
324,000
35,200
17,600
5377,000
5764,000
$103,000
551,300
546,100
5964,000
51.340.000
S560.000
5130,000
$9.900.000
($758,000)
($1.190.000)
(5103,000)
$57,800
511,700
(5792,000)
(51.220.000)
324,000
35,200
17,600
5377,000
5764.000
$103,000
$51,300
- 546,100
5964,000
51,340.000
$548,000
5120.000
RF-8a
513,600.000
($1,260,000)
($1.990,000)
(5172.000)
$96,300
$38,600
(51,300,000)
(52.030.000)
440,000
48,000
23,800
5512.000
51.040.000
5140,000
569,700
577,400
51,330.000
51.640.000
5540,000
(5190.000)
RMb T
513,600,000
($1,260,000)
($1,990,000)
($172,000)
596.300
519,400
(51.320,000)
(52,050.000)
440,000
48.000
23,600
5512.000
$1,040,000
$140,000
$69,700
$77,400
$1,330.000
$1,640.000
$520,000
(S21 0.000)
RF^9a T
$19,700,000
($2.270,000)
($3.580,000)
($309.000)
$173,000
570,000
($2,340,000)
(53.650,000)
628,000
68.000
34,000
5730.000
51,480,000
5199,000
$99.200
$138.000
51,920,000
$2.650.000
$310,000
(51.000.000)
RF-90
$19,700,000
($2,270,000)
(53.580.000)
($309,000)
$173,000
$35,000
($2,370.000)
(53,660,000)
628,000
68,000
34,000
5730.000
51,480.000
5199,000
599,200
5138,000
51,920.000
52.650.000
$280,000
(51.030.000)
(a) AH costs in S1991. Numbers in parentheses represent cost savings.
(b) Steam production credit calculations (either natural gas or fuel oil): .,,,.. „.
Natural gas credit = (model BLS firing rate) x (6,000 Btu/lb) x (1 d/24 hr) x (6S%-56% thermal efficiciency)/(85% power boiler efficiency) x
0 ft3M.024 Btu) X (8,424 hr/yr) X (53.48/1,000 ft3) ...,„,,
Fuel wl credit - (model BLS firing rate) x (S.OOO Btu/lb) x (1 d/24 hr) x (6S%-56% thermal efnciency)/(85% power boiler efficiency) x
(1 gal/144.000 Btu) x (8,424 hr/yr) x (50,77/gal)
(c) BLO operating cost savings = S251,900/yr x (model BLS firing rate/2,200.000 Ib BLS/d)
(d) Concentrator steam costs = 554,000/yr x (model BLS firing rate/2,400,000 Ib BLS/d)
(e) ESP operating elect/icity cost - 0.00194 x Increase in ESP plate area x 8,424 hr/yr X $0.06/kWh
m Administrative, taxes, and Insurance - 0.04 x TCI ._.,,..,
la) For low-odor conversion, capital recovery cost - 0.0944 x TCI (based on 2O-yr remaining service Me and 7% Intensl) .,,„,_„„
For ESP upgrade and wet-to dry-bottom ESP conversion, capital recovery cost = 0.1169 xTCI (based on 13.5-yr ESP life and 7% Interest)
6-92
-------
TABLE 6-18. SCENARIO 3: ANNUAL COSTS OF LOW-ODOR CONVERSION
(INCLUDING ESP UPGRADE TO CONTROL PM TO 0.10 G/DSCM
[0.044 GR/DSCF]) FOR MODEL DCE RECOVERY FURNACES (INCLUDING
UNBLEACHED PULP PRODUCTION LOSSES}a
Costs
Total Capital Investment (TCI)
Direct Annual Costs (DAC)
Steam production credits (b)
Natural gas
Fuel oil
BLO operating cost savings (c{
Concentrator steam costs (d)
ESP operating electricity costs (e)
Total DAC
Natural gas
Fuel oil
Indirect Annual Costs (IAC)
Administrative, taxes, and insurance (f)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Capital recovery (g)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Production losses
Total
Total IAC
btat Annual Costs
Natural gas
Fuel oil
RF-7a
$9,760,000
($758.000)
($1,190,000)
($103,000)
$57,800
$23,300
($760,000)
(81,210,000)
324,000
33,200
17,600
377,000
$764.000
$103,000
$51,300
$32,900
951.000
$1,330.000
$550,000
$120.000
Model Recovery Furnaces
RF-7b
$9.760,000
($758,000)
($1,190,000)
($103,000)
$57,800
$11,700
($792,000)
($1,220.000)
324,000
35,200
17,600
377,000
$764.000
$103,000
$51,300
$32,900
951.000
$1,330.000
$538,000
$110.000
RF-8a
$13.400.000
($1.260,000)
($1.990,000)
($172.000)
$96,300
$38,800
($1,300,000)
($2,030.0001
440,000
48,000
23.800
512,000
$1.040.000
$140.000
$69,700
$54,800
1,300,000
$1.820.000
$520.000
($210.000)
RF-Bb
$13,400,000
($1,260,000)
($1.990,000)
($172,000)
$96,300
$19,400
($1,320,000)
($2.050.000)
440.000
48,000
23,800
512,000
$1,040,000
$140,000
$69,700
$54,800
1,300,000
$1,820.000
$5OO.OOO
($230.000!
RF-9a
$19,200,000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$70,000
($2,340.000)
(S3.650.000)
628,000
68,000
34,000
730,000
$1.480,000
$199,000
$99,200
$98,2OO
1,880,000
$2.610.000
$270,000
(S1.040.OOO)
Rr-9b
$19.200.000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$35,000
($2,370,000)
($3,680,000)
628,000
68,000
34,000
730.OOO
$1,480,000
$199,000
$99,200
$96,200
1,880,000
$2.610,000
$240.000
($1.070.OOO)
(a) All costs in $1991. Numbers in parentheses represent cost savings.
(b) Steam production credit calculations (either natural gas or fuel oil):
Natural gas credit = (model BLS firing rate) x (6.00O Btu/lb) x (1 d/24 hr) x (66%r56% thermal efficiciency)/(85% power boiler efficiciencv) x
(1 ft3/1,024 8tu) x (8.424 hr/ysS X f£3.48/1,OOO«3!
fuel oil credit = (model BLS firing rate) x (S.OOO Btu/lb) x (1 d/24 hr) x f.66%-56% thermal efficiency)/(85% power boiler efficiency) x
(1 gal/144,000 Btu) x (8,424 hrfyi) x ($0.77/gaf)
(cj BLO operating cost savings = $251,900/yr x (model BLS firing rate/2.200,000 Ib BLS/d)
(dj Concentrator steam costs = $54,OOO/yr x (model BLS firing rate/2,400,000 Ib BLS/d)
-------
TABLE 6-19 SCENARIO 1: ANNUAL COSTS OF LOW-ODOR CONVERSION
(INCLUDING ESP UPGRADE TO CONTROL PM TO 0.034 G/DSCM
[0 015 GR/DSCF]) FOR MODEL DCE RECOVERY FURNACES (EXCLUDING
ANNUALIZED PULP PRODUCTION LOSSES)a
Total Caoital Investment (TCI)
Steam production credits (b)
Natural ga*
Fuel on
iLO operating cost savings (c)
Concentrator stum costs (d)
ESP operating electricity costs (e)
Total DAC
Natural gu
Indirect Annual Costs (1AC)
Admlnittrative. taxes, and insurance (0
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Casual recovery (g)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Total Annual Costs
Natural gas
SS.6OO.OOO
($758,000)
{S1.190.000)
(5103,000)
557,800
S33.100
(S770.000)
(S1.200.000)
324,000
50,800
17,600
392,000
$764.000
$148,000
$51,300
963,000
S1.360.000
$590,000
$160.000
S9.8OO.OOO
($758,000)
(51,190,000)
($103,000)
$57.800
S21.400
(8782,000)
($1.210.000)
324,000
50.800
17.600
392,000
$764.000
S148.000
$51,300
963.000
$1,360.000
$578,000
$150.000
RF-Sa
$13,400,000
($1,263,000)
($1,990.000)
($172,000)
$98,300
$55.000
($1,280,000)
($2,010.000)
440,000
73,600
23,800
537,000
$1,040.000
$215,000
$69.700
1,320.000
$1,660.000
$580,000
($150.000)
BF-Sb
$13,400,000
($1,260.000)
(S1.990.0OO)
($172,000)
$96,300
$35.600
(S1,3OO,000)
(52,030,000)
440.000
73.600
23.800
537.000
$1,040,000
$215.000
S69.700
1,320.000
$1.660.000
$560,000
(SI 70.000)
$19.400,000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$99,200
($2,310.000)
(S3.620.000)
628.000
114.000
34,000
776.000
S1, 480.000
$334.000
$99.200
1.910.000
$2.690,000
$380,000
($930.000)
$19.400,000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$64,200
($2,340,000)
($3.650.000)
628,000
114,000
34,000
776.0OO
$1.480,000
$334,000
$99.200
1,910.000
$2.690.000
$350.000
(S960.000)
(a) AH c«a in S1991. Numbers in parentheses represent cost savings,
(b! Steam production credit calculations (either natural gas or fuel oil): «•••„-*.
Natural gas credit - (model BUS firing rate) X (6.000 Btu/lb) x (1 d/24 hi) x (66%-5S% thermal effi=iciency)/(S5% power boiler efficiency) x
(1 M/1.024 Btuix (8.424 hr/yr)x ($3.48/1,000 ft3) ,
Fuel 0,1 credit - (model BLS firing rate) x (6.000 Btuilb) x (1 d/24 hi) x (66%-56% thermal efficiency>/(B5% power bo,ler effic.ency) x
(1 ga!rt44,000 Btu) x (8.424 hr/yr) x (S0.77/gal)
(c) BUD operating cost savings = S251.900/yrx (model BLS firing rate/2,200,000 Ib BLS/d)
(d) Concentrator steam costs - S54.OOO/yr x (model BLS firing rate/2,400.000 Ib BLS/d)
(e) ESP operating electricity cost - 0.00194 x increase In ESP plate area x 8,424 hr/yr x S0.06/KWh
(ft Administrative, taxes, and insurance « O.O4 x TCI
fa) For tew*dor conversion, capital recovery cost - 0.0944 x TCI (based on 20-yr remaining sennce life and T*!"*"*}
For ESP upgrade and wet-to dry-bottom ESP conversion, capital recovery cost = 0.1169 x TCI (based on 13.5-yr ESP We and 7% interest)
6-94
-------
TABLE 6-20. SCENARIO 2: ANNUAL COSTS OF LOW-ODOR CONVERSION
(INCLUDING ESP UPGRADE TO CONTROL PM TO 0.034 G/DSCM
[0.015 GR/DSCF]) FOR MODEL DCE RECOVERY FURNACES (INCLUDING
BLEACHED PULP PRODUCTION LOSSES)a
Costs
Total Capital Investment (TCI)
Direct Annual Costs (DAC)
Steam production credits (b)
Natural gas
Fuel oil
BLO operating cost savings (c)
Concentrator steam costs (d)
ESP operating electricity costs (d)
Total DAC
Natural gas
Fuel oil
Indirect Annual Costs (IAC)
Administrative, taxes, and insurance (e)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Capital recovery (f)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Production losses
Total
Total IAC
Total Annual Costs
Natural gas
Fuel oil
RF-7a
S10.300.0CX3
($756,000)
($1,190,000)
($103.000)
857,800
$33,100
(S770.000)
(SI. 200,000)
324,000
50.800
17,600
392.000
$764.000
$148,000
551,300
546,100
1,010,000
S1 .400,000
$630,000
S200.000
HF-7b
$10,300.000'
(5758.000)
($1,190,000)
($103.000)
557,800
' $21,400
($782,000)
(S1. 210.000)
324.000
50.8OO
17.6OO
392.000
$764.000
$143,000
551,300
$46.100
1,010.000
S1.4OO.OOO
$618.000
$190.000
Model Recovery Furnaces
RF-8a
$14.200,000
($1,260.000)
($1,990,000)
($172,000)
$96.300
$55,000
(51,280,000)
($2.010.000)
440.OOO
73,600
23.800
537,000
$1.040,000
5215,000
$69,700
$77,400
1,400,000.
$1.940.000
$660,000
($70.000)
RF-Sb
$14.200.000
(S1.260.000)
($1,990.000)
(S172.OOO)
$96.300
535,600
($1,300,000)
(52,030.000)
440.000
73,600
23.800
537,000
S1. 040.000
5215.000
S69.700
577.400
1,4OO.OOO
51.940.000
5640.000
(S90.000I
RF-Sa
$20.900,000
(52.270.000)
($3,580,000)
(5309.000)
S173.000
$99.200
($2,310,000)
(53,620.000)
628,000
114,000
34.000
776.000
$1,480,000
5334,000
599.200
5138.000
2.050.000
$2.830.000
5520.000
(5790.000)
RF-9b
$20,900.000
($2,270.000)
($3,580,000)
($309,000)
S173.000
$54,200
($2,340,000)
($3,650,000)
628.000
114.000
34,000
776,000
51,480,000
$334,000
S99.2OO
5138.000
2,050,000
$2.830.000
S490.OOO
(5820.000)
(a) All costs in 51991. Numbers in parentheses represent cost savings.
(b) Steam production credit calculations (either natural gas or fuel oil):
NituraS gas credit = (model BLS firing rate) x (6,000 Btu/lb) x (1 d/24 hr) x (66%-56% thermal efficiciency)/(85% power boiler efficiciency) x
p ft3/l,024Btu! x {8,424 hr/yr) x (S3.48/1.000 ft3)
Fuel oil credit = (model BLS firing rate) x (6.000 Btu/ib) x (1 d/24 hr) x (66%-56% thermal efficiency)/(85% power boiler efficiency) x
(1 gal/144,000 Btu) x (S.-424 hr;yr) x ($0.77/gal)
(c) BLO operating cost savings = S251.900ryr x (model BLS firing rate/2.200.000 Ib BLS/d)
(d) Concentrator steam costs = $54.000/yr x (model BLS firing rate/2.400,000 Ib BLS/d)
(e) ESP operating electricity cost = O.00184 x increase in ESP plate area x 8,424 hr/yr x $0.06/kWh
(f) Administrative, taxes, and insurance = 0.04 x TCI
(g) For low-odor conversion and production losses, capital recovery cost = 0.0944 x TCI (based on 20-yr remaining service life and 7% interest)
For ESP upgrade and wet- to dry-bottom ESP conversion, capital recovery cost = 0.1169 x TCI (based on 13.5-yr ESP life and 7% interest)
6-95
-------
TABLE 6-21
^I
[0 015
L° *
SCENARIO 3: ANNUAL COSTS OF LOW-ODOR CONVERSION
ESP UPGRADE TO CONTROL PM TO 0.034 G/DSCM
) FOR MODEL DCE RECOVERY FURNACES (INCLUDING
UNBLEACHED PULP PRODUCTION LOSSES) a
Total CacHa] Investment (TCI)
Stsam production credits (b)
Natural gas
Fuel oil
BIO operating cost livings (c)
Concentrator steam costs (d)
ESP operating electricity costs (e)
Total DAC
Natural gas
Indirect Annual Costs PAC)
Administrative, taxes, and Insurance (0
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Total
Capital recovery (g)
Low-odor conversion
ESP upgrade
Wet- to dry-bottom ESP conversion
Production losses
Total
Totai Annual Costs
Natura! gas
- edel Recovery
S1 0.100.000
($758,000)
C$1.190,000)
($103,000)
$57,800
$33,100
($770,000)
($1.200.000)
324.000
50,800
17.600
392.000
$764,000
$148.000
$5 1,300
$32,900
-996,000
$1.390,000
$620.000
$190.000
$10.100.000
($758,000)
($1,190.000)
($103,000)
$57,600
$21,400
($782,000)
($1.210.000)
324,000
50,800
17,600
332.000
$764.000
$148.000
$51,300
$32,900
996,000
$1.390,000
$608.000
$180000
RF-8a
$14,000.000
($1,260,000)
($1,990.000)
($172,000)
$96,300
$55,000
($1,280,000)
($2.010,000)
440.000
73,600
23,800
537,000
$1,040,000
$215,000
$69,700
$54.800
1,380,000
$1.920.000
$640.000
($90,000)
RF-Sb
$14.000.000
($1,260,000)
($1,990,000)
($172,000)
$96,300
$35,600
($1,300,000)
(S2.030.OOO)
440.000
73,600
23,800
537,000
$1,040,000
$215,000
$69.700
$54.800
1,380,000
S1.920.000
$620,000
($110.000)
$20.400,000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$99,200
($2,310,000)
($3,620.000)
626,000
114.000
34,000
776,000
$1,480,000
8334,000
$99,2OO
$98,200
2.010.OOO
S2.790.000
S480.0OO
($830.000)
$20,400.000
($2,270,000)
($3,580,000)
($309,000)
$173,000
$64.200
($2,340,000)
(S3.650 OOO)
628,000
114.000
34,000
77.6,000
$1,480,000
$334,000
$99,200
$98,200
2,010.000
$2,790,000
$450.000
($860.000)
(a) AH costs in $1991. Numbers in parentheses represent cost savings.
* ,66^56, therm, efficiency,/,** power bo ..... fT,cM x
(1 gal/144.000 Btu) X (8.424 hr/yr) x (S0.77/gal)
(c) BLO operating cost savings = $251,900/yr x (model BLS firing rate/2.200,000 Ib BLS/d)
(d) Concentrator steam costs - S54,000/yr x (model BLS firing rate/2.400,000 Ib BLS/d)
(e) ESP operating electricity cost - 0.00194 x Increase In ESP plate area x 8.424 hr/yr x $0.06/kWh
For ESP upgrade and wet- to dry-bottom ESP conversion, capital recovery cost = 0.1169 x TCI (based on
6-96
-------
TABLE 6-22a (METRIC). GAS AND LIQUID STREAM PARAMETERS FOR
RECOVERY FURNACE MODEL PROCESS UNITSa
Parameters
Gas Stream Prooerties:
ESP Outlet:
Flow rate, m^/sec
Temperature, CC
Moisture content, %
Out of Quench:
Flow rate, m /sec
Temperature, °C
Moisture content, %
Density, kg/m
Molecular weight,
g/gmole
Viscosity, kg/m-hr
HCI concentration, ppmv
HCI emissions, kg/d
Out of Absorber:
HCI concentration, ppmv
Liquid Stream Prooerties:
HCI concentration in
entering liquid, gmole HCI
per gmole pollutant-free
liquid
Density, kg/rn^
Moiecular weight,
g/gmole
Pollutant Properties:
Diffiisivity of HCI in air,
Diffusivitv of HCI in
water, m /hr
Model NDCE/Converted DCE Recovery Furnaces
RF-1/4/8
93
199
26
76
72
33
0.8946
25.3
0.0617
9.7
82
5
0
999
18
0.0673
9.4E-6
RF-2/5/9
168
199
26
136
72
33
0.8946
25.3
0.0617
9.7
147
5
0
999
18
0.0673
9.4-6
RF-3/6
243
199
26
197 .
72
33
0.8946
25.3
0.0617
9.7
212
5
0
999
18
0.0673
9.4E-6
RF-7
56
199
26
45
72
33
0.8946
25.3
0.0617
9.7
49
5
0
999
18
0.0673
9.4E-6
Model DCE Recovery Furnaces
RF-7
56
160
32
48
74
37
0.8762
25.0
0.0607
9.2
49
5
0
999
18
0.0673
9.4E-6
RF-8
93
160
32
80
74
37
0.8762
25.0
0.0607
9.2
82
5
0
999
18
0.0673
9.4E-6
RF-9
168
160
32
144
74 :
37
0.8762
25.0
0.0607
9.2
147
5
0
999
n
0.0673
9.4E-6
aMetric equivalents in this table were converted from the calculated English unit values given in Table 6-22b.
6-97
-------
TABLE 6-22b (ENGLISH). GAS AND LIQUID STREAM PARAMETERS FOR
. ii..-»-ii_ ^-M--» <-•*—• •• ^^-XT-H^-IT- T^T^ X"\^"*T3 O O TTKTTT^O
Parameters
— = — ^^^=
Gas Strom Properties:
ESP Outlet:
Flow rate, acfm
Temperature, *F
Moiiture content, %
Out of Quench*:
Flow rate, acfm
Temperature, °F
Moisture content, %
Density, lb/ft3
Molecular weight,
Ib/lbmole
Viscosity, lb/ft-hr
HC1 concentration, ppmv
HCI emissions, Ib/d
Out of Absorber:
HCI concentration, ppmv
Liauid Stream Properties:
HCI concentration in
entering liquid, Ibmole
HCI per Ibmole pollutant-
free liquid
Density, lb/ft3
Molecular weight,
Ib/lbmole
Pollutant Prooerties:
Diffusivity of HCI in air,
fl?/hr
Diffusiviry of HCI in
water, fir/hr
Model NDCE/Converted DCE Recovery Furnaces
RF-1/4/8
=====
198,000
390
26
161,000
161
33
0.05587
25.3
0.0415
9.7
180
5
0
62.4
IS
0.725
1.02E-4
RF-2/5/9
=====
357,000
390
26
289,000
161
33
0.05587
25.3
0.0415
9.7
324
5
0
62.4
18
0.725
1.02E-4
RF-3/6
=====
515,000
390
26
417,000
161
33
0.05587
25.3
0.0415
9.7
468
5
0
62.4
18
0.725
1.02E-4
RF-7
==
119,000
390
26
96,500
161
33
0.05587
25.3
0.0415
9.7
108
5
0
62.4
18
0.725
1.02E-4
Model DCE Recovery Furnaces
RF-7 1
-•'_ ' "-"—
119,000
320
32
102,000
165
37
0.05470
25.0
0.0408
9.2
108
5
0
62.4
18
0.725
1.02E-4
RF-8
=====
198,000
320
32
170,000
165
37
0.05470
25.0
0.0408
9.2
180
5
0
62.4
18
0.725
1.02E-4
RF-9
•
357,000
320
32
307,000
165
37
0.05470
25.0
0.0408
9.2
324
5
0
62.4
18
0.725
1.02E-4
«To simplify the packed-bed scrubber design analysis, it was assumed that the gas stream exiting the ESP was cooled to saturation
with water sprays.
6-98
-------
TABLE 6-23a (METRIC). PACKED-BED SCRUBBER DESIGN AND
OPERATING PARAMETERS51
Packed-Bed Scrubber
Parameters
Packed-Bed Scrubber Desien
Parameters:
Cross-sectional area, n?
Vessel diameter, m
Packing height, m
Tower height, m
Surface area, m^
Pressure drop, mm Hg
Auxiliary Equipment:
Stack height, m
Suck diameter, m
Fan impeller diameter, m
Fan motor efficiency,
fraction
Pump motor efficiency,
fraction
Power for fan, k\V
Power for pump, kW
Wastewater flow rate,
L/min
Caustic addition (dry
NaOH), kg/hr
Column Operation:
Minimum wetting rate,
m2/hr
Fraction of flooding gas
velocity
Operating hours per year
Equipment
Labor
Operating labor
requirement, hr/d
Maintenance labor
requirement, hr/d
Model NDCE/Converted DCE Recovery Furnaces
RF-1/4/8
36
6.7
0.47
8.4
249
2.4
30.5
3.1
2.7
0.7
0.7
34.4
28.1
6.25
25.6
0.12
0.7
8,424
8,760
1.5
1.5
RF-2/5/9
64
9.0
0.47
11
433
2.4
30.5
4.1
3.6
0.7
0.7
62.3
50.6
11.2
46.1
0.12
0.7
8,424
8,760
1.5
1.5
RF-3/6
93
11
0.47
13
615
2.4
30.5
5.0
4.1
0.7
0.7
89.5
73.1
16.2
66.6
0.12
0.7
8,424
8,760
1.5
1.5
RF-7
21
5.2
0.47
6.8
155
2.4
30.5
2.4
2.2
0.7
0.7
20.6
16.9
3.75
15.4
0.12
0.7
8,424
8,760
1.5
1.5
Model DCE Recovery Furnaces
RF-7
22
5.4
0.43
6.9
161
2.2
30.5
2.5
2.3
0.7
0.7
20.2
17.7
3.21
13.2
0.12
0.7
8,424
8,760
1.5
1.5
RF-8
37
6.9
0.43
8.5 .
259
2.2
30.5
3.2
2.8
0.7
0.7
33.6
29.5
5.36
22.0
0,12
0.7
8,424
8,760
1.5
1.5
RF-9
67
9.3
0.43
It
452
23.
30.5
43
3.7
0.7
0.7
60.7
53.1
9.65
39.6
0.12
0.7
8,424
8,760
1.5
1.5
"Metric equivalents in this table were converted from the calculated English unit values given in Table 6-23b.
6-99
-------
TABLE 6-23b
(ENGLISH). PACKED-BED SCRUBBER DESIGN AND
OPERATING PARAMETERS
Picked-Bed Scrubber
Parameters
=============
Packed-Bed Scrubber Design
Para meters:
CroJt-sectiontI area, ft
Vessel diameter, ft
Packing height, ft
Tower height, ft
Surface area, ft2
Pressure drop, in E^O
Auxiliary Eauioment;
Suck height, ft
Stack diameter, ft
Fan impeller diameter, in
Fan motor efficiency,
fraction
Pump motor efficiency,
fraction
Power for fan, kW
Power for pump, kW
Wastewater flow rate.
gpm
Caustic addition (dry
NaOH), Ib/hr
Column Oneration:
Minimum wetting rate,
f&hr
Fraction of flooding gas
velocity
Operating hours per year
Equipment
Labor
Operating labor
requirement, hr/d
Maintenance labor
requirement, hr/d
Model NDCE/Converted DCE Recovery Furnaces
RF-1/4/8
==
384
22.1
1.54
27.5
2,679
1.3
100
10.1
105
0.7
0.7
34.4
28.1
1.651
56.5
1.3
0.7
8,424
8,760
1.5
1.5
RF-2/S/9 |
690
29.6
1.54
35.2
4,657
1.3
100
13.6
140
0.7
0.7
62.3
50.6
2.971
102
1.3
0.7
8,424
8,760
1.5
1.5
RF-3/6
=====
998
35.6
1.54
41.3
6,624
1.3
100
16.3
160
0.7
0.7
89.5
73.1
4.291
147
1.3
0.7
8,424
8,760
1.5
1.5
RF-7
=====
231
17.1
1.53
22.4
1,672
1.3
100
7.8
85
0.7
0.7
20.6
16.9
0.990
33.9
1.3
0.7
8,424
8,760
1.5
1.5
Model DCE Recovery Furnaces
RF-7
====
242
17.6
1.41
22.7
1,735
1.2
100
8.1
90
0.7
0.7
20.2
17.7
0.849
29.1
1.3
0.7
8,424
8,760
1.5
1.5
RF-8
=
403
22.7
1.41
27.9
2,791
1.2
100
10.4
110
0.7
0.7
33.6
29.5
1.415
48.4
1.3
0.7
8,424
8,760
1..5
1.5
RF-9
===
725
30.4
1.41
35.8
4,864
1.2
100
14.0
145
0.7
0.7
60.7
53.1
2.548
87.2
1.3
0.7
8,424
8,760
1.5
1.5
6-100
-------
TABLE 6-24a (METRIC). UNIT COSTS FOR PACKED-BED SCRUBBER
Parameters
Operator wage rate, $/hr
Water cost, $/m3
Electricity, $/kWh
Wastewater disposal, $/m3
Caustic (dry NaOH), $/Mg
Packing material, $/m^
Value
15.64
0.05
0.06
1.0
441
706
aMetric equivalents in this table were converted from the calculated English unit values
given in Table 6-24b.
TABLE 6-24b (ENGLISH). UNIT COSTS FOR PACKED-BED SCRUBBERa'b
Parameters
Operator wage rate, $/hr
Water cost, $71,000 gal
Electricity, $/kWh
Wastewater disposal, $71,000 gal
Caustic (dry NaOH), $7ton
Packing material, $/rr
Value
15.64
0.2
0.06
3.8
400
20
fOAQPS Control Cost Manual
Medical Waste Incinerators-Background Information for Proposed Standards and
Guidelines: Model Plant Description and Cost Report for New and Existing Facilities.
U. S. Environmental Protection Agency. Research Triangle Park, NC. Publication
No. EPA-453/R-94-044a. July 1994. p. 62.
6-101
-------
F
«
g
Q
es
S
04
O
. a
S * a
s
s
a ^
B
1 1
1 1
|
§ 1
5
I
S
S
8. i
g S
i
S
2
s
I
8
g
1
S
i
«- °°- " ^
| 5 B S
§ g S
o
U
H S ••* p,
E > ^; p
.S- I 8 1
a E- 0- cu
O^
6
I
C.
g i
u. tu
I
CO
•o
60
IS
O
C3
D.
C
E
o
00
r
W 3
Ix
3 £
— O" _
O
UN R £ O
o^.^-S
\Q C
8 III
«*N ^-/
X
o
^ o
'« I I
en C A
II
Jh
'•= E
x •- e
v )S q" —
g «o 5 co
I " "S-'g-
* J -
_- = tt, ON
»w ^—
tt. on
2 <3 rt
01 «-! -K
6-102
-------
rrt
*0
CO
w
1
f^>
H
0
W
H
r*\
8
s
os
o
In
CO
C
U
!
g
§
OS
W
ff
ff
i —
C*
U
CO
Ct
m
I
Q
pt
C
<;
0)
U)
CXI
.
El
ft
p
%
£
1
o:
W
u
Q
a*
•o
u
3
Recovery Furn
w
c
c
4
•c
a
|
ii
u
Q
Z
g
-o
c
*
i
oo
t^
u.
u.
^
~l
s:
1
Gd
••• ' "• •
oo
IT
e^
§
o"
8
S
§
^
|
f*»
g
s
*;
o"
CO
V
1
s
oe
—
1
v>
1
-
•
-
u"
M
B
.
c.
Q
^?S-2
vooo s:r Er G S
00 ^^ r4 r» **l J5 *
a s a a S g g a |
^^ OXOS SS§g£S
<^_ f? 5_ 5_^ °. °- _: t-_ ^
co — •• o^ o^ ^/\ v« ^* y^ *^
r^Tj- o^o^ gcniQoc
8 S 5 5 «. S -. C *
1
s
o.
1 ""
c o
•*\. __ c2 .§•
"S o g £• k.
i 1 1 I = H 1 * s 1
« w cu-S-0^5-^ S «-s;£
t* ^3 C 83 ^1 , — — ^5. 2 (fl
g_O«-5JS:=Ui? =fe
cr
^
.
i
u
•^
.
_ r» o\
H. 8. S. B.
=0
—• «•> !Q 2
r^ ^ .
S 5 ^ yi
~ a S S;
M 9 S n.
^" a s a
*»«»*»«»
= & JS 2
c) o -
s 8 S -
.^T m" ri t*1
£ s S S
•" g - i
? ?- s a
S £» -£J £
= ° S §
M oo
X 0
1 i 1 i
fll C ft.
6 < u t-
|
g
§
§
|
2
$790,000
1
5
ex
T
V
4
b
.
c.
I
S-
•3
CO
S
I §
6-103
-------
H
I
H
O
8.
H
CQ
CO
6
m
(n
"c
g
m CM
•CD 6
O CQ
§
m
Design parameters
o
0
o
o"
o
CJ
""~
8
o
8
0
0
o
o"
•a
CO
m
D)
•2C
Black liquor firing rate,
0 0
CO CM
CO CO
0 0
co in
co •<*
o o
CO h~
CM CM
Equivalent ADMP/d
Bleached pulp
Unbleached pulp
CM
m
co
CM
O
CO
CO
CD
D)
CD
,-j.
in
©
o
CD
CO
CO
E
1
I
H—
CO
co
CD
•4-f
CD
T- in
in CD
CO CO
i- in
in CD
CO CO
T- in
in CD
CO CO
,
co
Operating days per ye
Equipment
Labor
•* o
CM CD
co" co"
""sj" O
CM CD
^i" r*"*"
co" co"
•* o
CM CD
•3- r--
co" co"
co
m
Operating hours peryi
Equipment
Labor
CD
JCD
.Q
CO
CD
CO
CD
D
.CO
"cb
c
UJ
T3
CD
.O
CO
o
CD
.C
CD
1±
CD
O
O
CD
CD
_
.D
CO
•*-•
CO
C
c
CD
CT
CD
.0
CO
6-104
-------
§
H
Q
H
X)
O
8
o
ffl
§
CO
CN
t
VD
co
2
CD
en
'c
3
3
CO CM
0 6
O CQ
g
m
esign parameters
Q
0
o
o
o"
CM"
o
o
0
o"
o
in
0
o
o
o"
o
0)
T3
CO
CD
.Q
O
^f O
CM CO
•
-------
TABLE 6-28. CAPITAL AND ANNUAL COSTS OF COLLECTION AND
INCINERATION OF BLO VENT GASES FOR MODEL BLACK LIQUOR
OXIDATION UNITSa
Total Capital Investment (TCI) (b)
Direct Annual Costs (DAC)
Operating labor (c)
Operator
Supervisor
Maintenance (d)
Labor
Material
Electricity (e)
Steam (f)
Total DAC
Indirect Annual Costs (IAC)
Overhead (g)
Administrative, taxes and insurance (h)
Capital recovery (i)
Total IAC
Total Annual Costs
Model BLO units
BLO-1
2,500,000
$18,600
$2,790
$27,400
$27,400
$103,000
$22,000
$179,000
$45,700
$100,000
$356,000
$502,000
$681,000
BLO-2
3,390,000
$18,600
$2,790
$27,400
$27,400
$209,000
$44,600
$285,000
$45,700
$136,000
$483,000
$665,000
$950,000
BLO-3
4.830,000
$18,600
$2,790
$27,400
$27,400
$312,000
$66,600
$388,000
$45,700
$193,000
$688,000
$927,000
$1 ,320,000
(a) All costs in $1991.
(b) TCI = $4.800,000 x [(model BLS rate)/(800 ADTP/d x 3,400 Ib BLS/ADTP)] ~ 0.6 x (361.3 $1991 /3S7.6 $1990)
(c) Operating labor ~ 0.5 hr/shifl/condenser x 3 sh'rfts/d x 2 condensers x 365 d/yr x $17/hr
Supervisor labor - 15% of operator labor
(d) Maintenance labor = 0.5 hr/shift/condenser x 3 shifts/d x 2 condensers x 365 d/yr x $25/hr
Maintenance materials = 100 percent of maintenance labor
(a) Electricity costs = Mill A electricity costs x (model vent gas flowrate/Mill A vent gas flowrate)
Mill A electricity costs = (100 hp + 3 hp + 400 hp) x 0.746 kW/hp x 8,424 hr/yr x $0.06/kWh
The Mill A electricity costs are based on 100 hp to operate the mill water booster pump motor, 3 hp to
operate the BLO condenser condensate pump motor, and 400 hp to operate the BLO off gas blower motor.
The Mill A vent gas flowrate - 16,327 acfm
(1) Steam cost = Mill A steam cost x (model vent gas flowrate/Mill A vent gas flowrate)
Mill A steam cost = 1,600 Ib steam/hr x 8,424 hr/yr x $3/1,000 Ib steam
The Mill A steam cost is based on steam requirements of 1,600 Ib steam/hr for the BLO off gas reheater.
The Mill A vent gas flowrate = 16,327 acfm
(g) Overhead = 0.6 x (labor + maintenance)
(h) Administrative, taxes, and insurance = 0.04 x TCI
(i) Capital recovery = 0.1424 x TCI. based on 7% interest and 10-yr equip, life for ductwork and condenser.
6-106
-------
CQ
Q
CO
w
1
g
CO
EH
CO
O
2
W
a
^
a
w
C4
OS
W
m
PQ
§
o
CO
CJ ^^
2 m"
5 if w
'5. > -2:
O a.
oS .>•
_r
*~
| | 0
= 8 f
8 °"
o <
|| g
o a. a_
c*) ^—} **•
.£ Q
15 O
co Q-
ca
—
. "5s fl-
.> S
U Q
UJ
.a" 15 co
_*: O> m
m •*-
a. a>
"5 g-
a. 2:
"a>
"a H
° Q
2 co
o
o
a
CO
<*
o
o
T—
&
o
o
N-
JI
f»
o
o
o
c\
^
CD
JD
JD
d
0
CD
JD
JD
E>
0
CO
o
CO
CM
O
O
O
o"
O
S
J5
CO
o
o
1C
en
IT
o
o
^
*
o
0
O)
in
49-
0
O
co
a>
CO
r^
CD
JD
JD
zs
o
CO
ffl
JD
JD
o
CO
§
CO
o
O
O
cf
O
r-.
m
CM
fe
CO
o
fs^
^.
CO
g
'f~
cc
cc
Si
fO-
o
o
o
cc
CO
I —
a>
JD
JD
D
o
CO
CD
JD
JD
o
VI
o
CO
CD
O
O
o
o"
o
CM
m
CO
£
CO
g
^^
CO
c
in
£
1
CO
CM
to-
O
O
O
CO
o
N.
CD
JD
JD
u
0
CO
CD
JD
JD
?
u
CO
o
o
o
0
cf
s
55
&
CO
o
o
a
CO
«
o
o
•^
g
o
o
f-
^~
««•
o
o
cv
*
CD
JD
JD
?
U
CO
CD
JD
JD
O
to
o
CM
O
O
O
o"
O
JD
C
z>
fc
CO
o
{£
cn
ir
0
tv
*
o
o
cn
in
w-
o
o
o
CO
I
N-"
CD
JD
JD
D
O
CO
CD
JD
JD
U
CO
o
in
o
JD
^
CM
Q
CO
R
f-.
rt-
3-
0
"~
cc
c
cc
CM
CM
t&
O
O
O
CO
CO
£
CO
0)
JD
JD
O
CO
(U
JD
JD
O
CO
o
CM
CO
O
O
o
8"
CVJ_
JD
^
CO
£
CO
o
o
CD
O
O
in_
1
CO
CM
o
o
o
§
CD
JD
JD
3
Q
to
CD
JD
JD
2
o
CO
o
0
CM
o
o
o
cf
o
co
JD
c
ZD
••s-
H
CO
o
Ic
o
'o
o
JD
O)
C\I
DC
JD"
O)
CVI
CO
JD
CO
c
ID
c
D
j;
co
T3
O
a
o
0)
j:
o
CD
(D
>
C
O
o
_
JD
a
c
<2 S
C O
(D «
3 ^a
cr to
CD O
6-107
-------
rt
CO
Q
W
g
CO
o
u
CO
H
>J
OJ
i
VD
W
S"?
afs
«$!•*
o 8 »
os tT
- -— — .
0 =£•
CD — "
Pi
o Q t3
= O °s
to n
CO oj
CB ~ ,0
CQ —
t |
Q- 2r
"S
•a H
0 Q
S co
1
co"
tj-
g
CM
CO
to-
|
=
Q
CV
en
o.
oi"
scrubber
a
~
o
CO
o
u
o
g
o
o
en
1-
o
o
CO
of
in
o
cj
te-
g
en
in
o
o
«
g
ir
scrubber
a
j.
~
0
CO
o
c
o
IT
5
CM
H
C
g
f-
§
o
o
CM
CO
to-
g
CO
CM"
V*
o
CD
CO
ts
o
c
CC
CM
scrubber
a
.1
o
CO
o
in
CM
m
CO
fe
g
o
co"
o
69-
g
in
r>-
w
o
o
CM
co"
CM
V*
O
O
O
CO
o
o
o:
o"
scrubber
CD
S.
3
O
CO
o
0
o
0
CO
5
1
D
CO
1
co"
o
o
CM
CO
|
O
o
o
CM
1
0
o
M-
en"
scrubber
CD
X
^
0
CO
cr
o
o
o
en
3
^
1-
u
o
o
CO
en
in
1
"
o
o
en
in
o
0
o
CO
o
o
f-
ir:
scrubber
a
^
2
o
en
ir
o
o
o
o
o
ir
j
c
~^
CM
g
g
f—
S"
to-
O
o
CM
O
0
Cp_
g
O
CO
CO
in
to
o
o
p;
co"
CM
scrubber
a
S.
2
o
CO
o
en
o
o
c
o
o
e\
'-,
~~^
CO
h-
U.
\ $106,000
o
in
r-
g
CM
co"
CM
to-
0
o
CO
to-
g
o"
o
en
0
X
2
o
CO
o
o
CT
0
O
o
cT
o
en
co"
'-
Tf
Q
CO
CO
C3
CO
cj^
I
CD
"S
c3
O)
55
Cj
"cu
O
c >>
eg m
CD >
:= O
X. O
.;-. Q CD £
8 t|l
o •* j^ a.
£S 8 S
« -
—
in
CM 0) $
-
co CD a> _
^ " o to
o CD" § o
o" •£ o ^
u
11
1—' II .—
= O I « g
< I- < O i
6-108
-------
TABLE 6-30a (METRIC) . MODEL SDT/SCRUBBER DESIGN PARAMETERS3-
Design parameters
Baseline APCD:
Control option APCD:
Black liquor firing rate, kg BLS/d:
Equivalent ADMP/d:
Bleached pulp
Unbleached pulp
Gas flowrate-APCD inlet (Q), m3/sec:
Existing pressure drop, mm Hg:
New pressure drop, mm Hg:
Operating days per year:
Equipment
. Labor
Operating hours per year:
Equipment
Labor
Model SDT's
SDT-5
mist
eliminator
scrubber
400.000
230
270
4.4
1.3
13
351
365
8,424
8,760
SDT-6
mist
eliminator
scrubber
700,000
380
450
7.4
1.3
13
351
365
8,424
8,760
SDT-7
mist
eliminator
scrubber
1,200,000
680
820
13.4
1.3
13
351
365
8,424
8,760
(a) Metric equivalents in this table were converted from the calculated English unit values
given in Table 6-30b.
6-109
-------
TABLE 6-30b (ENGLISH). MODEL SDT/SCRUBBER DESIGN PARAMETERS
Baseline APCD:
Control option APCD:
Black liquor firing rate, Ib BLS/d:
Equivalent ADTP/d:
Bleached pulp
Unbleached pulp
Gas flow rate-APCD inlet, acfm:
Existing pressure drop, in. H2O:
New pressure drop, in. H20:
Operating days per year:
Equipment
Labor
Operating hours per year:
Equipment
Labor
Model SDT's
SDT-5
mist
eliminator
scrubber
900,000
250
300
9,400
0.7
7
351
365
8,424
8,760
SDT-6
mist
eliminator
scrubber
1,500,000
420
500
15,700
0.7
7
351
365
8,424
8,760
SDT-7
mist
eliminator
scrubber
2,700,000
750
900
28,300
0.7
7
351
365
8,424
8,760
6-110
-------
TABLE 6-31. CAPITAL AND ANNUAL COSTS TO REPLACE MIST
ELIMINATORS WITH SCRUBBERS FOR MODEL SDT'Sa
Costs
Total Capital Investment (TCI) (b):
Direct Annual Costs (DAC) :
Operating labor (c):
Operator
Supervisor
Maintenance (d):
Labor
Material
Utilities (e):
Electricity
Water
Wastewater treatment (f) :
Total DAC:
Indirect Annual Costs (IAC):
Overhead (g):
Admin., taxes and insurance (h):
Capital recovery (i) :
Total IAC:
Total Annual Costs (TAC) :
Incremental Total Annual Costs (1TAC) (j):
Model SDT's
SDT-5
$584,000
$37,200
$5,580
$27,400
$27.400
$5,420
$9,500
$0
$1 1 3,000
$32,900
$23,400
$64,100
$120,000
$233,000
$190,000
SDT-6
$796,000
$37,200
$5,580
$27,400
$27,400
$9,050
•$15,900
$0
$123,000
$32,900
$31 ,800
$87,400
$152,000
$275,000
$232,000
SDT-7
$1,130,000
$37,200
$5,580
$27,400
$27,400
$16,300
$28,600
$0
$142,000
$32,900
$45,200
$124,000
$202,000
$344,000
$301 ,000
(a) All costs in $1991
{b) New scrubber TCI (models 5-7) = [scrubber replacement TCI (models 1-3)] x 2
(c) Operator labor = 6 hr/d x 365 d/yr x $17/hr
Supervisor labor = 15% of operator labor
{d} Maintenance labor = 3 hr/d x 365 d/yr x $25/hr
Maintenance materials = 100% of maintenance labor
(e) Electricity = 0.000181 x inlet gas flowrate x (7 - 0.7 in. H2O) x 8,424 hr/yr x $0.06/kWh
Water = 0.6 x inlet gas flowrate x 8,424 hr/yr x $0.20/1,000 gal
(f) Recycled to SDT
(g) Overhead = 0.6 x maintenance
(h) Administrative, taxes, and insurance = 0.04 x TCI
(i) Capital recovery = 0.1098 x TCI (based on 15-yr scrubber life and 7% interest)
(j) ITAC = TAC - operating labor costs.
6-111
-------
8
§
§
o
o
nJ
05
ro
i
H
118
•5 c
6-112
-------
CO
H
J
i
m
W
tH
if
1 g
CD ~ S
o § £
08 ,- S?
o 8
< o
0
_ £„
f i fe
<3g*
«-'S
<•"
o ^— '
H-
C Q
Jo
n
to **•
c
^ S
'r Q
cr <
UJ
E -S 1
5 S |
"5
T3 o «
O c J=
2 = J£
s-
S
s.
o"
8
S-
0
o"
o
o
co
o
o
o
r^
in
•a-
V)
S
^
in
o
8
o
c\f
CM
D_
CO
UJ
^
S
X)
D
CO
0
CO
o
o
v
_J
o
o
I
*~*.
8
in
I
R
o
&
8
co
in
o
o
o
CO
CO
CO
CM
ID
^
CM
CO
O>
8
in
CM"
O-
co
UJ
,_
o
XI
fe
10
s
r^
o
CM
_j
o
o
o
co"
„
8
s
CM
g
in"
CO
S
o
8
1
o
8
^
in
»
s
""
CO
CO
CO
CM"
r-
O-
co
UJ
L-
£
X)
(0
0
o
*•"
1
?
_J
8
-o
J
XI
CO ~= S^
UJ .2 i
f m
' ~D O CT- D)
6-113
-------
ri
m
ro
W
8
o
8"
CM
s
"8 „-
II
O
Q
8
o
CM
r~.
s.
CD
CO
0
8
§
CM
CM
O
||
11 £ii
O CO
UJ
1 oil
CO
0
- =
0 <
Iff
1
8
0
to
8
o"
s
8
H
c
o
_D
I
"c
D
JC
_a
D
_O
O
-------
ffi
CO
H
J
i
w
m
n
- ro
^ *•"
t*
T3
O (/)"
JZ D)
g | f
oo**
o
. .
.3 CD «£-
p *» «_
J5 o -^
.5'?
< <»
5 -a
H* ^
E
."§
S8
tP Q_ (§)
offi
"2 5 o
%-g o
« &<
CD
C Q
•5 0
§ £
m
I -D
o Q.
> i—
•5 b
uT<
I«I
a fi 1
"o
T CD 2
° E
o
-
5
o"
8
o>
•»
a
o
o
CM
CM
o
8
s
CM
CO
CM
T-
J^
*
CJ
rr
a.
CO
UJ
scrubber
o
m
o
C\J
_j
o
8
CM
^
in"
o
o
o
o
^.
i?
«»
1
*£>
O
8
g
CO
CO
8
T-
OD
8
5
R
CM
cJ
r-
a.
C/)
UJ
scrubber
o
-
8
CO
CO
_J
§
costs in $
"^
a
E
1
8
o
1
X
CD
S
O
c=
CO
en
"5
|
11
ca
fl>
P plate an
w
UJ
J3
•3-
£
CO
03
JS
n.
a
iS
•a-
X
CO
£
CO
JS
Q.
W
UJ
II
o
o
g
X
o
o
II
c«
^_
CO
1
•o
c
0
of
u
i
CO
CD"
ministrativ
**
o
• ESP life and 7% interest)
>^
IO
1
CO
£.
X
LL
cr
o
g
o
o
II
CD
£
"5.
CO
o
CD
= (ESP DAC + ESP overhead costs) - (scrubber DAC + scrubber overhead
D)
1
ct annual
b
ec-
hoed cost savings -f A,l & T + capital recovery
CD
g
-D
C
CO
O
2
ii
cT
<|
H
1
O
•5
C
CO
3
o
remental 1
^_
D)
6-115
-------
TABLE 6-34a (METRIC). MODEL LIME KILN/SCRUBBER DESIGN
PARAMETERS51
Design parameters
ipiiiiiiiiiHglif*8"'!! ill-' '-' .iii-...iiit...=a:^^^g:^s^^s:^^^^^^^^^
Baseline APCD
Lime production rate, Mg CaO/d
Equivalent ADMP/d
Gas flowrate-APCD inlet, m3/sec
Pressure drop, mm Hg
Operating days per year
Equipment
Labor
Operating hours per year
Equipment
Labor __
^— —— *^~?==^
Model Lime Kilns
LK-1
=—======
scrubber
90
320
10
39
351
365
8,424
8,760
LK-2
s==s^=^=s^^=:^s:
scrubber
180
680
20
39
351
365
8,424
8,760
LK-3
=====
scrubber
270
1,000
34
39
351
365
8,424
8,760
(a) Metric equivalents in this table were converted from the calculated English unit values
given in Table 6-34b.
6-116
-------
TABLE 6-34b (ENGLISH). MODEL LIME KILN/SCRUBBER DESIGN
PARAMETERS
Design parameters
Baseline APCD
Lime production rate, ton CaO/d
Equivalent ADTP/d
Gas flowrate-APCD inlet, acfm
Pressure drop, in. H20
Operating days per year
Equipment
Labor
Operating hours per year
Equipment
Labor
Model Lime Kilns
LK-1
scrubber
100
350
22,000
21
351
365
8,424
8,760
LK-2
scrubber
200
750
42,500
21
351
365
8,424
8,760
LK-3
scrubber
300
1,100
72,200
21
351
365
8,424
8,760
6-117
-------
TABLE 6-35. ANNUAL COSTS FOR EXISTING LIME KILN SCRUBBERS*
costs „_===
Direct Annual Costs (DAG)
Operating labor (b)
Operator
Supervisor
Maintenance (c)
Labor
Material
Utilities (d)
Electricity
Water
Wastewater treatment (e)
Total DAC
Indirect Annual Costs (IAC)
Overhead (f)
Total Annual Costs
LK-1
$37,200
$5,580
$27,400
$27,400
$42,000
$22,200
$0
$162,000
$58,500
$221,000
vlodel Lime Mins
LK-2
$37,200
$5,580
$27,400
$27,400
$81,200
$43,000
$0
$222,000
$58,500
$281,000
LK-3
$37,200
$5,580
$27,400
$27,400
$138,000
$73,000
$0
$309,000
$58,500
$368,000
(a) All costs in $1991
(b) Operator labor = 6 hr/d x 365 d/yr x $17/hr
Supervisor labor = 15% of operator labor
(c) Maintenance labor = 3 hr/d x 365 d/yr x $25/hr
Maintenance materials = 100% of maintenance labor
(d) Electricity = 0.00018 x model inlet gas flowrate x pressure drop x 8,424 hr/yr x $0.06/kWh
Water = 0.06 x model inlet gas flowrate x 8,424 hr/yr x $0.20/1,000 gal
(e) Recycled to mud washer
(f) Overhead = 0.6 x (labor + maintenance)
6-118
-------
TABLE 6-36a (METRIC) . MODEL LIME KILN/ESP DESIGN PARAMETERS51
Design parameters
Control option APCD
Lime production rate, Mg CaO/d
Equivalent ADMP/d
Gas flowrate~ESP exit, m3/sec
ESP plate area, m2
PM controls— 0. 1 5 g/dscm
PM controis~0.023 g/dscm
SCA, m2/(m3/sec)
PM controls-0.15 g/dscm
PM controls-0.023 g/dscm
Pressure drop, mm Hg
Operating days per year
Equipment
Labor
Operating hours per year
Equipment
Labor
Model Lime Kilns
LK-1
ESP
90
320
10
' 909
2,210
91
221
2
351
365
8,424
8,760
LK-2
ESP
180
680
20
1,818
4,420
91
221
2
351
365
8,424
8,760
LK-3
ESP
270
1,000
34
3,090
7,514
91
221
2
351
365
8,424
8,760
(aj Metric equivalents in this table were converted from the calculated English unit values
given in Table B-3Sb.
6-119
-------
TABLE 6-36b (ENGLISH). MODEL LIME KILN/ESP DESIGN PARAMETERS
Control option APCD
Lime production rate, ton CaO/d
Equivalent ADTP/d
Gas flowrate-ESP exit, acfm
ESP plate area (A), ft2
PM controls-0.067 gr/dscf
PM controls-0.01 gr/dscf
SCA,ft2/1, 000 acfm
PM controls-0.067 gr/dscf
PM controls--0.01 gr/dscf
Pressure drop, in. H20
Operating days per year
Equipment
Labor
Operating hours per year
Equipment
Labor
Model Lime Kilns
LK-1
ESP
100
350
22,000
10,157
24,699
462
1,123
1
351
365
8,424
8,760
LK-2
ESP
200
750
42,500
19,621
47,714
462
1,123
1
351
365
8,424
8,760
LK-3
ESP
300
1,100
72,200
33,332
81,058
462
1,123
1
351
365
8,424
8,760
6-120
-------
TABLE 6-37.
ANNUAL COSTS FOR NEW LIME KILN ESP'S CONTROLLING
PM TO 0.15 G/DSCM (0.067 GR/DSCF)a
Costs
Total Capital Investment 0"CI) (b):
Direct Annual Costs (DAC):
Operating labor (c):
Operator
Supervisor
Coordinator
Maintenance (d):
Labor
Material
Electricity (e):
Electricity-fan
Electricity-operating
Waste disposal:
Total DAC:
Indirect Annual Costs (IAC):
Overhead (f):
Admin., taxes and insurance (g):
Capital recovery (h):
Total IAC:
Total Annual Costs:
Model Lime Kilns
LK-1
$457,000
$13,100
$1,970
$4,320
$838
$2,740
$2,000
$10,000
$0
$35,000
$13,800
$18,300
$50,200
$82,300
$117,000
LK-2
$883,000
$13,100
$1,970
$4,320
$1,620
$5,300
$3,870
$19,200
$0
$49,400
$15,800
$35,300
$97,000
$148,000
$197,000
LK-3
$1,500,000
$13,100
$1,970
$4,320
$2,750
$9,000
$6,570
$32,700
$0
$70,400
$18,700
$60,000
$165,000
$244,000
$314,000
(a) All costs in $1991
(b) TCI = ESP plate area x $45/ft2 plate area
(c) Operator labor = 3 hr/day x 365 d/yr x $12/hr
Supervisor labor = 15% of operator labor
Coordinator labor = 33% of operator labor
(d) Maintenance labor = 0.0825 x ESP plate area
. Maintenance material = 0.01 x (0.6 x TCI)
(e) Electricity-fan = 0.00018 x model gas f lowrate x pressure drop x 8,424 hr/yr x $0.06/kWh
Electricity-operating = 0.00194 x ESP plate area x 8,424 hr/yr x $0.06/kWh
(f) Overhead.= 0.6 x (labor + maintenance)
(g) Administrative, taxes, and insurance = 0.04 x TCI
(h) Capital recovery = 0.1098 x TCI (based on 15-yr ESP life and 7% interest)
6-121
-------
TABLE 6-38.
ANNUAL COSTS FOR NEW LIME KILN ESP'S CONTROLLING
PM TO 0.023 G/DSCM (0.010 GR/DSCF)a
Costs
Total Capital Investment (TCI) (b):
Direct Annual Costs (DAC):
Operating labor (c):
Operator
Supervisor
Coordinator
Maintenance (d):
Labor
Material
Electricity (e):
Electricity-fan
Electricity-operating
Waste disposal:
Total DAC:
Indirect Annual Costs (IAC):
Overhead (f):
Admin., taxes and insurance (g):
Capital recovery (h):
Total IAC:
Total Annual Costs:
Model Lime Kilns
LK-1
$1,110,000
$13,100
$1,970
$4,320
$2,040
$6,660
$2,000
$24,200
$0
$54,300
$16,900
$44,400
$122,000
$183,000
$237,000
LK-2
$2,150,000
$13,100
$1,970
$4,320
$3,940
$12,900
$3,870
$46,800
$0
$86,900
$21,700
$86,000
$236,000
$344,000
$431,000
LK-3
$3,650,000
$13,100
$1,970
$4,320
$6,690
$21,900
$6,570
$79,500
$0
$134,000
$28,800
. $146,000
$401,000
$576,000
$710,000
(a) All costs in $1991
(b) TCI = ESP plate area x $45/ft2 plate area
(c) Operator labor = 3 hr/day x 365 d/yr x $12/hr
Supervisor labor = 15% of operator labor
Coordinator labor = 33% of operator labor
(d) Maintenance labor = 0.0825 x ESP plate area
Maintenance material = 0.01 x (0.6 x TCI)
(e) Electricity-fan = 0.00018 x model gas flowrate x pressure drop x 8,424 hr/yr x $0.06/kWh
Electricity-operating = 0.00194 x ESP plate area x 8,424 hr/yr x $0.06/kWh
(f) Overhead = 0.6 x (labor + maintenance)
(g) Administrative, taxes, and insurance = 0.04 x TCI
(h) Capital recovery = 0.1098 x TCI (based on 15-yr ESP life and 7% interest)
6-122
-------
i*
1 SB
f^ rt *^
tO g tO-
ff
-
08 ^
^~ *
£
.— >»
"5 «S
CD
UJ
H- **
13"
CD - a>
"> co c:
l£g~
0 CO ;=
_C UJ ro
0
' ^
T5 CO" C\i
s 1 o
*^- ^
= "x ob
O CO
UJ
l|
§•§
LU
g Ti 5
E 0 C3)
^J cLS
"CD
TD CD ™
0
CM
t»
0
s
*
V*
0
CM
CD
CM
«*
R
CO
^~
*
8
o
s
CD
o
CO
^
o
o
o
8
o
CD
_l
O
8.
CM
69-
0
8
co
CO
o
o
o
in
to-
o
in
t>-
CM
O
O
O
S
CM_
O
CO
CM
O
CO
CM"
o
o
co
CD
o
co
in
_i
o
o
o
ca-
o
o
o
CD
CO
CM
o
o
o
CO
co
o
CO
co"
tr\
O
o
o
o"
in
CM"
o
CO
^.
1
^-
0
o
o
o_
o
CM
CO
_J
XI
en
co
JO
co
c
-------
g
u
CO
Q
Crf
CD
O
r-{
O
*^^
S
CO
H
J
cf *S]
<< fcTl
™~ *>vl
1 §fs
§" 8 -^
2
(_" "5T
08 •_
<-£
^s-.
;5 S
*^ t
o ,^»
.2 «?
UJ
o"S
J— ^5.
E
o _ *n
w
g
co
co"
Tl-
o
g
o"
in
CM"
CD
CO
CO
CM
r^
!*
o
o
CM
£
CD
O
O
V)
X
co
X
E
•s
g J
CO
E
c3
"g-
g
ca D-
TD
o
Q. •=
g
,
s SB sill
i im
S. ~ x H
i g
a> — £ CD g _
_ns
^ CL
CD Q_
51 co
S uj
CD
_ O
co *•;
D °
g II
| S 5 I" I
-
-±
-
15" "
' C- O)
6-124
-------
TABLE 6-40. SUMMARY OF ENHANCED MONITORING COSTS
Enhanced monitoring system
Opacity monitor
Methanol CEM (FITR)
HC1 CEM
EPA Method 5 PM compliance test (three runs)
EPA Method 29 PM compliance test (three runs)
EPA Method 26 or 26A HC1 compliance test (three
runs)2
Monitoring plan
Confirmation of NDCE recovery furnace with dry
ESP system
Confirmation of BLO control equipment
Temperature monitor for power boiler or other
incineration device0
pH monitoring system
Flow rate measurement system°>e
Magnehelic gauge, S-rype pitot tube (pressure drop
measurement)"
Cost
Capital cost: $34,800
Annual cost: $21,200/yr
Capital cost: $160,000
Annual cost: $71,500/yr
Capital cost: $126,900
Annual cost: $77,400/yr
Initial compliance test cost: $8,500
Semiannual testing annual cost: $17,000/yr
Initial compliance test cost: $12,000
Seminnual testing annual cost: $24,000/yr
Initial compliance test cost: $9, 100*
$600b
Annual testing annual cost: $9, lOO/y^
$600b
Cost not available
$0
$0
$0
$5,000
$2,000 - $25,000
Manual read-out system: $300
Digital read-out system: Cost not available
aCost assumes EPA Method 26 or 26A HC1 testing is performed alone.
bCost assumes EPA Method 26 or 26A HC1 testing is performed in conjunction with EPA Method 5,
Method 29, or Method 17 testing.
cCost of monitoring power boiler or incinerator temperature is assumed to be zero because temperature
monitoring is already conducted by mills.
d Annual costs were not available for monitoring pH and flow rate of scrubber liquid or for monitoring
scrubber pressure drop.
eCost depends on the device sensitivity and distance from the control room.
6-125
-------
TABLE 6-41. OPACITY AND HCL CONTINUOUS EMISSION MONITOR
COSTSa
Parameters
Capital costs
Planning
Select type of equipment
Provide support facilities
Purchased equipment cost
Install and check CEM's
Performance specification tests (certification)
Prepare QA/QC plan
Total capital investment
Annual costs
Operating and maintenance
Annual RATAb
Supplemental RATA
Quarterly CGA's0
Reporting and recordkeeping
Annual review and update
Administrative, insurance, and property taxes
Capital recovery6
Total annual cost
CEM costs
Opacity CEM
=======^=
$1,200
$4,000
$100
$21,500
$300
$1,400
$6,300
$34,800
$7,000
$0
$0
$0
$5,900
$3,600
$3,300
$1,400
$21,200
HC1 CEM
$4,600
$10,500
$13,100
$61,500
$11,100
$14,300
$11,800
$126,900
$8,500
$10,000
$9,500
$3,500
$11,600
$17,200
$12,000
$5,100
$77,400
aAll costs are based on EMTIC's CEM program, except for administrative, insurance, property tax, and
capital recovery costs, which are based on procedures from the OAQPS Control Cost Manual.
^Relative accuracy test audit
cCylinder gas audits
^Administrative, taxes, and insurance = 0.04 x TCI
eCapital recovery = 0.0944 x TCI (based on 20-yr equipment life and 7% interest)
6-126
-------
6.3 REFERENCES FOR CHAPTER 6
1. Memorandum from Holloway, T. and Nicholson, R., MRI, to
Telander, J., EPA/MICG. October 9, 1996. Nationwide Costs,
Environmental Impacts, and Cost-Effectivness of Regulatory
Alternatives for Kraft, Soda, Sulfite, and Semichemical
Combustion Sources.
5.
6.
7.
8.
9.
10.
11
Economic Indicators--CE Plant Cost Index.
Engineering (New York). 1960-1993.
Chemical
Memorandum from Mathias, S.
EPA/ISB. February 2, 1994.
for Kraft Pulp Mills.
EPA/CEIS, to Telander, J.,
Value of Lost Production days
Memorandum from Soltis, V., MRI, to the project file.
April 3, 1995. Kraft and Soda Pulp Mill Combustion Sources
Data Base.
Handbook: Control Technologies for Hazardous Air Pollutants.
U. S. Environmental Protection Agency. Cincinnati, OH.
Publication No. EPA-625/6-91-014. June 1991. Chapter 4.
OAQPS Control Cost'Manual. 4th Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-90-006. January 1990.
U. S. Environmental Protection Agency. National Emission
Standards for Hazardous Air Pollutants: General Provisions.
59 FR 12408 et seq. Washington, DC. U.S. Government
Printing Office. March 16, 1994.
Memorandum from Soltis, V., Nicholson, R., Holloway, T.,
MRI, to Telander, J., EPA/MICG. July 29, 1994. Summary of
Responses to the NCASI "MACT" Survey--Kraft and Soda Pulp
Mills (Data Base Summary Memo).
Memorandum from Nicholson, R., MRI, to Telander, J.,
EPA/ISB. May 28, 1993. Meeting minutes from the March 30,
1993 meeting between Environmental Elements Corp. and
Midwest Research Institute.
Memorandum from Holloway, T., MRI, to the project file.
July 16, 1996. State of Washington PM Data for Kraft
Recovery Furnaces, Smelt Dissolving Tanks, and Lime Kilns.
Memorandum from Radian Corp. to EPA/ISB. September 18,
1985. Meeting minutes from the June 10, 1985 meeting
between kraft pulp mill representatives, the National
Council of the Paper Industry for Air and Stream
Improvement, Inc., the U. S. Environmental Protection
Agency, and Radian Corp.
6-127
-------
12. Letter and attachments from Eddinger, J., EPA/ISB, to
Blosser, R., National Council of the Paper Industry for Air
and Stream Improvement. May 22, 1985. Preliminary cost _
estimates for evaluating the three control options to bring
wet-bottom ESP's into compliance with the NSPS.
13. Letter from Bringman, L., Environmental Elements Corp., to
Nicholson, R., MRI. October 12, 1993. Budgetary pricing to
convert from wet- to dry-bottom ESP.
14. Letter from Burt, R., Radian Corp., to Holbrook, J. and
Bringman, L., Environmental Elements Corp. July 23, 1985.
Facsimile cover letter discussing utility and maintenance
costs for wet- vs. dry-bottom ESP designs.
15. Garner, J., Jaako Poyry. Conversion to Low Odor Improves
Recovery Boiler Efficiency and Life. Pulp and Paper.
63(7):91-95. July 1989.
16. Green, R. and G. Hough (eds.). Chemical Recovery in the
Alkaline Pulping Process. 3rd Edition. Prepared by the
Alkaline Pulping Committee of the Pulp Manufacture Division.
Atlanta, GA, TAPPI Press, 1992. 196 p.
17. Memorandum from Ramsey, M. and Nicholson, R., MRI, to
Telander, J., EPA/ISB. November 3, 1992. Meeting minutes
from the June 24, 1992 meeting between The Babcock and
Wilcox Co., the U. S. Environmental Protection Agency, and
Midwest Research Institute.
18. Telecon. Soltis, V., MRI, with Morgan, P., ABB Combustion
Engineering Services, Inc. August 6, 1993. Information
about cascade evaporators and low-odor conversion costs.
19. Telecon. Soltis, V., MRI, with Blake, J., Goslin-
Birmingham. August 27, 1993. Concentrator costs.
20. Williamson, F., Chief Power Engineer, Harmac Pulp Division.
Low Odor Boiler Conversion--By Threes. In: TAPPI Kraft
Recovery Operations Short Course. TAPPI Press.
January 3-8, 1993. pp. 229-247.
21. Hein, A. Kraft Recovery Boiler: Conventional to Low-Odor
Conversion. Pulp & Paper Canada. 88(12):223-226.
December 1987.
22. Kirby, M. Economic and Process Considerations in the Use of
Oxygen for Black Liquor Oxidation. Union Carbide Canada
Limited. Ontario, Canada. (Presented at the 21st Pulp and
Paper ABCP Annual Meeting. Sao Paolo, Brazil.
November 21-25, 1988.) 10 p.
6-128
-------
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Perry, R., D. Green, and J. Maloney (eds:).
Chemical Engineer's Handbook. 6th edition.
pp. 9-16, 9-18.
Perry's
1984.
Pulp, Paper, and Paperboard Industry--Background Information
for Proposed Air Emission Standards, Manufacturing Processes
at Kraft, Sulfite, Soda, and Semi-Chemical Mills. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-453/R-93-050a. October 1993.
National Petroleum News 1991 Fact Book.
June 1991. p. 114.
Volume 83, No. 7.
Letter and attachments from Black, C., Champion Interna-
tional Corp., Courtland, AL, to Crowder, J., EPA/ISB.
June 10, 1993. Response to request for information on costs
for the BLO vent gas control system at the Champion facility
in Courtland, AL.
Telecon. Soltis, V., MRI, with Oscarsson, B., Gotaverken
Energy Systems, Inc. December 16, 1993. Concentrator steam
costs.
Memorandum from Rovansek, W., Radian Corp., Herndon, VA, to
the record. September 1, 1993. Pulp and Paper Operating
Costs.
Letter and attachments from Sanders, D., Andersen 2000,
Inc., to Soltis, V., MRI. November 9, 1993. Preliminary
proposal for three separate quench and packed-tower
absorption systems to control HC1 gas emissions from
recovery furnaces at kraft pulp and paper mills.
Memorandum from Randall, D. and Holloway, T., MRI, to the
project file. October 4, 1994. Absorber Design and Cost
Algorithm.
Medical Waste Incinerators--Background Information for
Proposed Standards and Guidelines: Model Plant Description
and Cost Report for New and Existing Facilities. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-453/R-94-044a. July 1994. p. 62.
Someshwar, A. and J. Pinkerton. Wood Processing Industry.
In: Air Pollution Engineering Manual, Air and Waste
Management Association. Buonicore, A. and W. Davis (eds.).
New York, Van Norstrand Reinhold. 1992. pp. 835-849.
RUST Environment & Infrastructure. MACT Cost Analysis
Report for NCASI. Prepared for the National Council of the
Paper Industry for Air and Stream Improvement, Inc. RUST
Contract No. 35-4735. October 15, 1993. 47 p.
6-129
-------
34.
35.
36.
37.
38.
39.
40.
Memorandum from Nicholson, R. and Holloway, T., MRI, to
Telander, J., EPA/MICG. September 6, 1995. Trip Report for
S.D. Warren Co., Muskegon, MI.
Telecon. Soltis, V., MRI, with Bruno, J., Airpol, Inc.
April 22 and 27, 1993. Information on scrubbers used to
control particulate matter emissions from lime kilns and
smelt dissolving tanks at pulp and paper mills.
Memorandum from Nicholson, R., MRI, to Telander, J.,
EPA/MICG June 15, 1995. Meeting minutes from the
April 12, 1993 meeting between Research-Cottrell and Midwest
Research Institute.
Memorandum from Toney, M., EPA/EMB, to Telander, J.,
EPA/ISB. November 24, 1993. Enhanced Monitoring for Pulp
and Paper Combustion Sources.
Memorandum from Segall, R. EPA/EMB, to Copland, R., EPA/SDB.
October 14, 1992. Final Report on Emission Measurement and
Continuous Monitoring for Medical Waste Incinerator Study.
Memorandum from Toney, M., EPA/EMC, to Telander, J.,
EPA/MICG. January 13, 1995. Costs of Enhanced Monitoring
for Pulp and Paper Combustion Sources.
Telecon Harris, V., MRI-NCO, with Hosenfeld, J. , MRI-KCO.
May 26, 1995. Costs of EPA Method 26 stack tests.
6-130
-------
APPENDIX A.
EVOLUTION OF THE TECHNICAL SUPPORT DOCUMENT
The purpose of this study was to develop a basis for
supporting proposed NESHAP for combustion sources at kraft and
soda pulp and mills. This study was initiated in June of 1991
and focused on the following combustion equipment: chemical
recovery furnaces (including DCE's), SDT's, BLO systems, and lime
kilns. To accomplish the objectives of this study, the following
technical data were acquired for each type of combustion equip-
ment: (1) equipment design and operating parameters, (2) types
and quantities of HAP's emitted, (3) types and costs of emission
control techniques (including both process changes/controls and
add-on air pollution control devices), and (4) the effectiveness
of these control techniques in reducing HAP emissions. The bulk
of the information was gathered from the following sources:
1. Technical literature;
2. Responses to a maximum achievable control technology
(MACT) survey distributed by NCASI to U.S. pulp and paper mills;
3. Plant visits;
4. Emission test reports supplied by individual mills, State
agencies, and NCASI;
5. Recovery furnace manufacturers; and
6. Manufacturers of ESP's and wet scrubbers.
Significant events relating to the evolution of the TSD are
listed in Table A-l.
A-l
-------
TABLE A-l. EVOLUTION OF THE TECHNICAL SUPPORT DOCUMENT
Date
7/1/91-
7/31/91
8/1/91
9/17/91
9/23/91
9/24/91
9/25/91
10/28/91
1/10/92
1/14/92
2/7/92
2/24/92
4/9/92
4/30/92
5/30/92
6/24/92
7/2/92
9/22/92
9/23/92
9/30/92
10/8/92-
10/31/92
11/20/92
3/30/93
4/12/93
5/24/93
6/14/93
6/15/93
8/16/93
9/13/93
10/12/93
Event
Contacted State and local agencies to identify well-controlled combustion sources
Initial meeting with NCASI and AFPA regarding combustion sources NESHAP and "MACT"
survey
Site visit to Champion International Corp., Roanoke Rapids, NC
Site visit to Procter and Gamble, Perry, FL
Meeting with Florida Department of Environmental
mills in Florida and obtain copies of permits
Regulation to discuss kraft pulp and paper
Site visit to Packaging Corp., Clyattville, GA
Meeting with NCASI and AFPA to discuss industry
profile
Site visit to Longview Fibre Co., Longview, WA
Site visit to James River Corp., Camas, WA
NCASI MACT survey mailed to industry by NCASI
Meeting with Gotaverken to discuss design and operation of kraft recovery furnaces
Public meeting held in Washington, DC to discuss status of NESHAP and effluent guidelines
for pulp and paper mills
Initial responses to NCASI MACT survey received by EPA, and kraft combustion sources
data base created to tabulate information in these responses
Additional responses to NCASI MACT survey received by EPA
Meeting with Babcock & Wilcox to discuss design and operation of kraft recovery furnaces
Draft summary of responses to NCASI MACT survey prepared
Site visit to MEAD Coated Board, Phenix City, AL
Site visit to Champion International Corp., Courtland, AL
Public meeting in Washington, DC, to discuss status of NESHAP and effluent guidelines for
pulp and paper mills
Source test at Champion International Corp., Roanoke Rapids, NC, to identify HAP's emitted
from a recovery furnace, smelt dissolving tank, black liquor oxidation system, and lime kiln
Meeting with NCASI and industry representatives to discuss NCASI's source test plan for
combustion sources
Meeting with Environmental Elements Corp., Baltimore, MD, to discuss ESP designs,
modifications, and upgrades, and associated costs
Meeting with Research-Cottrell, Somerville, NJ, to
upgrades, and associated costs
discuss ESP designs, modifications, and
First EPA Work Group meeting to discuss status and direction of combustion sources
NESHAP
Monthly particulate matter emission data for kraft recovery furnaces, lime kilns, and SDT's
received from the State of Washington (1989-1992)
Emission test reports for Texas kraft mills received
from NCASI
Letters sent to Andersen 2000, Inc., and AirPol regarding the effectiveness of packed-bed
scrubbers on HC1 removal from kraft recovery furnace flue gas
Completed review and summary of State regulations regarding PM and gaseous HAP
emissions from kraft combustion sources
Meeting with NCASI and industry to discuss methanol emissions and effectiveness of steam
strippers
A-2
-------
TABLE A-l. (continued)
Date
10/31/93
12/2/93
12/16/93
1/1/94
2/3/94
2/17/94
3/11/94
3/21/94
4/11/94
4/27/94
5/12/94
7/13/94
7/29/94
7/29/94
8/24/94
8/31/94
9/1/94
9/14/94
9/21/94
9/30/94
10/4/94
11/30/94
12/20/94
1/18/95
1/31/95
3/7/95
4/20/95
Event
Review of first draft emission test report from NCASI-sponsored testing at kraft pulp
completed
Mill-specific control costs for MACT floor options submitted to EPA for impact analysis
Meeting with NCASI and industry to discuss status of NCASI source test program,
subcategories, MACT floors, and proposed format of the standard
Preliminary drafts of TSD Chapter 1 through 6 completed
Meeting with NCASI and industry to discuss status and findings of NCASI's test program
Preliminary draft of TSD Chapter 6 (Control Costs) sent to NCASI for review
Mill-specific control costs for beyond-the-floor options submitted to EPA for impact analysis
Industry comments on preliminary draft of TSD Chapter 6 received
Response to industry comments on preliminary draft of TSD Chapter 6 completed
Preliminary draft of Appendix B completed
An updated version of monthly particulate matter emission data for kraft recovery furnaces,
lime kilns, and SDT's received from the State of Washington (1988-1994)
Received and reviewed report from industry summarizing results from an industry survey of
recovery furnace manufacturers regarding time needed for recovery furnace conversions that
included a list of all kraft and soda recovery furnaces valid through December 1994
All HAP emissions data from NCASI's test program received to date compiled into emissions
data base for the project
Preliminary regulatory alternatives developed based on the subcategorization of recovery
furnaces for the source category
Combustion sources data base revised to account for mill-wide and recovery furnace
shutdowns and furnaces installed since the original MACT survey was completed based on an
industry inventory of kraft and soda recovery furnaces
Emissions data received from International Paper Co. compiled and added to emissions data
base for the project
Meeting with industry representatives to discuss the MACT standards under consideration and
the vendor capacity and resource requirements to rebuild/replace recovery furnaces.
Site visit to Federal Paper Board, Augusta, GA
Site visit to S.D. Warren Co., Muskegon, MI
Received and reviewed HC1 study (Technical Bulletin No. 674) conducted by NCASI
Additional mill-specific control costs for MACT floor and beyond-the-floor options submitted
to EPA for impact analysis
Second Work Group meeting to provide technical update
Third Work Group meeting to discuss the preliminary regulatory alternatives, the impacts and
economic analyses of those alternatives, and the EPA recommended regulatory alternative;
examined use of bubble approach to reduce the control costs for particulate control under the
EPA recommended regulatory alternative
Teleconference meeting with OECA on enhanced monitoring
Teleconference meeting with States on issues associated with the bubble approach
Teleconference meeting with OECA on enhanced monitoring protocols
Teleconference meeting with OECA on enhanced monitoring
A-3
-------
TABLE A-l. (continued)
Date
Event
Meeting with industry representatives regarding control options for the Pulp and Paper
NESHAP and the definition of a "clean mill"
5/22/95
8/2/95
Meeting with representatives of the Institute of Clean Air Companies to discuss the status of
the NESHAP
2/29/96
Submitted revised mill-specific control costs for new regulatory alternatives to EPA for impact
analysis
2/29/96
Submitted inputs for HC1 risk analysis to EPA to determine applicability of Section 112(d)(4)
of the CAA to HC1 controls for pulp and paper combustion sources
4/12/96
Nationwide environmental impacts inputs submitted to EPA for benefits analysis
6/17/96
Meeting with industry representatives to discuss applicability threshold and PM HAP emisison
limits for new sources
7/8/96
Teleconference meeting with OECA on enhanced monitoring
7/26/96
Draft TSD submitted to EPA and industry representatives for review
Teleconference meeting with industry representatives regarding enhanced
approaches
7/31/96
monitoring
Fourth Work Group meeting to provide update on status of the NESHAP
NCASI comments on the TSD received
A-4
-------
APPENDIX B.
EMISSION MEASUREMENT AND CONTINUOUS MONITORING FOR
PULP AND PAPER COMBUSTION SOURCES
B.1 INTRODUCTION
Between July 1991 and November 1994, EPA gathered emissions
data from a variety of information sources in support of a NESHAP
for combustion sources in the pulp and paper industry. Sources
of information included:
1. An emissions test program sponsored by NCASI;
2. An emissions test program for the Texas Paper Industry
Environmental Committee (TPIEC);
3. An emissions test program sponsored by the International
Paper Company; and
4. Selected emissions test reports in state agency
compliance files.
The industry-sponsored test programs were specifically
designed to provide data to: (1) determine the mass emission
rates of HAP's from recovery furnaces, BLO units, SDT's, and lime
kilns; and (2) determine control efficiencies of APCD's for
HAP's.
Emissions testing was performed primarily using manual
methods. Locations sampled included the inlet and outlet of the
control device(s) where pollution controls were applied, and the
effluent stack where they were not applied.
This appendix defines the methods used in the test programs
to collect emissions data, discusses alternatives for monitoring
emissions to indicate continuous compliance with any proposed
B-l
-------
standards, and recommends procedures to demonstrate compliance
with proposed standards.
B.2 MEASUREMENT METHODS USED IN DATA COLLECTION
This section presents a summary of measurement methods used
to generate emissions data in approximately 29 source test
reports. The data in these reports were used to support the
development of emissions standards for combustion sources in the
pulp and paper industry.
B.2.1 fi^opp of Test Programs
The types of samples and data collected at each test site,
the measurement locations, and test methods used are summarized
in Tables B-l through D-a.1"29 Tables B-2 through B-8 summarize
information and data collected as part of International Paper
Company's nine-mill emissions test program. Table B-l summarizes
information contained in the remaining 20 emission test reports.
Sample types and sampling locations are discussed in Section 2.1.
The specific test methods used for collecting and analyzing the
samples are discussed in Section 2.2.
B.2. 1.1 Types of Samples and Data Collected. The samples
collected during the test programs were primarily flue gas
samples before or after emission control devices. Because a
variety of source emissions test reports were used, a disparity
exists in the number and type of pollutants measured at each
source .
The source tests contained data for the following
pollutants :
1. Metals, including antimony (Sb) , arsenic (As) , beryllium
(Be) , cadmium (Cd) , cobalt (Co) , chromium (Cr) (total) , lead
(Pb)
thallium (Tl)
manganese (Mn), mercury (Hg), nickel (Ni), selenium (Se)
SO.
H.
A •
,S, methyl mercaptan, dimethyl sulfide, carbon
disulfide, sulfuric acid;
4. Terpenes, a-terpinol, a,b-pyrene, 3-curene, p-cyrene;
5. HC1;
6. CO ;
B-2
-------
7. NOX;
8. Polychlorinated dibenzo-p-dioxins and dibenzofurans
(PCDD/PCDF);
9. Selected organic HAP's, such as acrolein, acetaldehyde,
acetophenone, acetone, methanol, carbon tetrachloride,
chloroform, chlorobenzene, chloromethane, bromomethanol, isobutyl
ketone, methyl ethyl ketone, methylene chloride, 1,2-dichloro-
methane, trichlorofluoromethane, bromodichloromethane,
formaldehyde, 1,1,1-dichloroethylene, 1,1,2-trichloroethane,
1,2,4-trichloroethane, tetrachloroethylene, styrene, toluene,
xylenes, benzene; and
10. Semivolatile organics, such as acenaphthene, pyrene,
benzo-a-anthracene, chrysene, benzo-b-fluoroanthene,
benzo-a-pyrene, dibenzo-a,h-anthracene, benzo-g,h,i-pyrene,
indeno-1,2,3-pyrene.
B.2.1.2 Emission and Process Sampling Locations. Flue gas
stream samples consisted of uncontrolled gas streams from BLO
tanks, SDT's, and recovery furnaces, and controlled streams from
ESP outlets and scrubbers following lime kilns and recovery
furnaces. Additionally, one test report contained analysis of a
quench water process stream. This process stream was analyzed
for PCDD/PCDF and metals.
B.2.2 Sampling and Analytical Methods Used
Sampling and analysis methods used at each mill are given
in Tables B-l and B-2. The discussion in this section groups the
sampling and analytical methods into three categories:
(1) organic HAP methods, (2) metal HAP methods, and (3) HC1
methods.
B.2.2.1 Organic HAP Methods. The methods used to gather
organic HAP emissions data included EPA Methods 18 and 23 of 40
CFR 60 Appendix A; Methods 0010, 0011, and 0030 of SW-846; and
two methods developed by NCASI. One of the methods developed by
NCASI and validated by EPA Method 301 made use of a heated summa
canister source sampling train (HSCSST) for collection of the gas
sample for later laboratory analysis by gas chromatography (GC)
and flame ionization detector (FID) or mass selective detector
B-3
-------
(MSD).1/2,30,31 This method is capable of measuring a variety of
organic HAP's. The HSCSST used a Summa® canister, heated to
130°C, for collecting gaseous samples. The system employed for
the analysis of the Summa® canisters was consistent with the
requirements of EPA Method T014, except that the system bypasses
the Nafion® permeation dryer.
The other method developed by NCASI and validated by EPA
Method 301 made use of a water-filled impinger followed by silica
gel traps for capturing methanol and other water-soluble organics
such as acetaldehyde, acetone, and methyl ethyl ketone.1'32 The
collected samples are later analyzed by GC/FID. The method is
very similar to EPA's Method 308 for methanol, which is described
in Section 4.4.
Differences among the methods include the ability to provide
information on multiple pollutants and detection limits.
Methods 18, 0010, 0030, and the HSCSST provided information on a
greater number of pollutants. Methods 23 (dioxins), 0011
(aldehydes) , and 308 (methanol) provided information on a very-
limited number of pollutants. Detection limits for most of the
methods were generally acceptable. However, some of the data
gathered using EPA Method 18 did not provide adequate detection
limits because the GC set-up variables were not optimized.
B.2.2.2 Metal HAP Methods. At most plants, metal HAP
compounds were collected and analyzed using EPA draft Method 29
(also known as draft Method 0012 of SW-846, the EPA combined
metals train, or the EPA multimetals train). Differences were
noted in some of the older test reports with respect to blank
correction procedures and in the treatment of nondetect data.
These differences were reconciled before the data were used
further.33'34 The GARB Method 425 was used at six plants for
determining hexavalent Cr emissions.16'17'29 At one plant, EPA
Method 12 and a modified Method 5 train were used to determine Ni
and Mn, and hexavalent Cr emissions, respectively.10 At this
plant, a stainless steel probe was used instead of a glass probe.
For this reason, high levels of Cr and Ni were present in the
B-4
-------
blanks. The Cr and Ni emissions data at this mill were removed
from the NESHAP data pool.
B.2.2.3 Hydrogen Chloride Methods. Hydrogen chloride
emissions were determined using EPA Methods 26 and 26A in 40 CFR
Part 60 Appendix A. The two methods are very similar, except
that in Method 26A the gas sample is collected isokinetically,
whereas in Method 26 a nonisokinetic procedure is used. Method
26A is preferred at locations downstream of scrubbers where acid-
gas mists are present.
B.3 MONITORING SYSTEMS AND DEVICES
This section presents the various types of continuous
monitoring systems and monitoring devices, including data
acquisition and data processing systems, readily available on an
"off-the-shelf" basis that could be used to monitor emissions.
The applicability of these systems to the affected facilities or
similar facilities is discussed. This discussion covers such
factors as the accuracy, precision, repeatability, reliability,
complexity, maintenance, the difficulty of installation and
operation of these systems, and costs (capital and annualized).
Cost estimates include the costs of setting up and operating the
monitoring system, but not the cost of reporting any data
collected. Performance specifications are provided, also, if
available.
B.3.1 Opacity Monitoring
Opacity monitoring equipment has been routinely used to
monitor the opacity of emissions from combustion sources for many
years. These monitors are comparatively simple and are easy to
install. They are usually installed in the breeching duct or in
the discharge stack downstream of any particulate control device.
Choice of an appropriate measurement location depends upon
source-specific factors, such as effluent ductwork configuration,
presence of vibration, ease of access to the location, etc. A
detailed discussion of the accuracy, precision, repeatability,
reliability, and maintenance requirements for these instruments
will not be provided because this monitoring technique is well
established.
B-5
-------
Estimated initial costs (purchase, installation, performance
testing, quality assurance program plan preparation) and annual
costs (operation & maintenance, recordkeeping and reporting,
program update, quality assurance) for an opacity monitor
installed at a stack location are $39,600 and $18,800,
respectively (all dollar amounts are in 1994 dollars).35
Performance specifications for opacity monitors are
contained in Performance Specification 1, Appendix B, 40 CFR
Part 6C. These requirements were first promulgated on October 6,
1975; substantial revisions were issued on March 30, 1983.
B.3.2 Hydrogen Chloride Monitoring
Emissions monitoring equipment for measuring gaseous HC1 is
commercially available; monitoring techniques include gas filter
correlation (GFC) infrared analyzers, FTIR, ion mobility
spectroscopy (IMS), ultraviolet (UV) spectroscopy, on-line
specific ion electrodes (SIE), and mass spectrometric techniques.
The mass spectrometric technique is not included in the
discussion that follows because of its high cost. Detection
limits for these instruments range from the low part per billion
(ppb) range (e.g., IMS) to the low ppm range (e.g., FTIR). The
FTIR and SIE systems are fairly complex systems, whereas the GFC
and IMS are not. All of these systems are comparatively new,
and, therefore, source-specific installation instructions are not
well developed. All systems are prone to difficulties if
condensed water is present in the sample delivery lines. Sample
lines and other sample delivery components must be relatively
•ri anH should be heated to prevent condensation.
• Vio i <~,™ rvr-iTn r-zmap. some HC1 absorption
.nert to moist I
tor HC1 concentrations in 1
in the sample c
system start up. However, after an :
JLCXUJ.WUO -.«. - ,
eliverv system is likely to occur upon monitoring
ii ^ ^ ^fc^^A*4»**^3 W *-* ** V *•—. •— —• — — — — —
referred method of moisture removal where
iah. Gas dilution to reduce the moisture
:he HC1 instrument :
Iryers are' the i
moisture contents are t
:ontent below the dew point of the gas i
lowever, the latter t
spots are present in t
padinas should become stable. Permapure
L^^J.1J.L- W A- U-ii^- 3(-*-l~' *"-*•--——•—• — a
schnique is not always successful if cold
Vio <3=Tnr>i <= d^liverv svstem. The gas
_ c.
-------
dilution technique will dilute the sample below the HC1
guantitation limit for some of the instrumental methods.
Limited information can be provided at this time about the
accuracy, precision, repeatability, and reliability of HC1
monitoring systems as applied to gas streams from pulp and paper
mills. It should be noted that none of the aforementioned
methods were developed for, or have yet been successfully applied
to, measuring HC1 emissions from recovery furnaces at kraft pulp
mills. However, performance characteristics of GFC monitoring
systems at hazardous waste incineration facilities have been
studied by EPA and have been found to be acceptable.
The NCASI completed an evaluation of a TECO 15 GFC
monitoring system using the dilution approach for moisture
removal.
37
Method 26 data were gathered concurrently with the
monitoring data. The investigators were unable to establish a
good correlation between the GFC monitoring data and the EPA
Method 26 data. Several factors may have been responsible for
the poor results. For example, the HC1 content of the gas stream
may have been reduced to levels below the detection limit of the
instrument by the dilution gases. It is possible to compensate
for the lower gas concentrations by increasing the optical path
length. However, sufficient details were not provided in the
study to confirm this. Loss of HC1 in the sampling lines is also
a possibility due to the high moisture contents. Improved HC1
recoveries may be achieved by using other moisture management
schemes, such as a Nafion® drier at or near the stack. The
Nafion® drier would remove moisture while the gas stream is still
hot and significantly reduce the potential for water condensation
in the sampling lines. Furthermore, the use of the Nafion® drier
as an alternative to dilution gases lessens the potential for
reducing the HC1 concentrations below the GFC monitor detection
limits.
Based on information generated in an EPA study of GFC
monitoring systems, the initial costs (purchase, installation,
performance testing, quality assurance program plan preparation)
and annual costs (operation & maintenance, recordkeeping and
B-7
-------
reporting, program update, quality assurance) for a GFC monitor
installed at a stack location are $144,500 and $68,700,
respectively.35'36 The estimated initial costs and annual costs
associated with an FTIR monitoring system are $162,000 and
$50,000, respectively, and those for an IMS system are
approximately $149,500 and $68,700, respectively.38'39 The SIE
systems have not been costed because their performance
characteristics are not acceptable.
Performance specifications for HC1 monitoring systems in
stationary sources are contained in Performance Specification 13,
which was proposed on April 19, 1996 along with regulations for
40
hazardous waste combustion systems.
B.3.3 Methanol Monitoring
Emissions monitoring equipment for measuring vaporous
methanol is currently commercially available; HAP specific
monitors include on-line GC/FID and FTIR. The methanol
quantitation limit for the GC/FID system is 2 to 3 ppm.1 If
lower detection limits are needed, an Entech concentrator can be
used.
A GC-based monitoring system, called the hazardous organic •
mass emission rate (HOMER) monitor, is another type of monitoring
system that combines EPA Methods 25A and 18 using a direct
interface.41 A total hydrocarbon (THC) analyzer provides a.
continuous measurement of the THC concentration, as propane, in
an extracted gas sample. The analyzer output is recorded by a
personal computer-based data acquisition system (PC/DAS). A gas
chromatograph equipped with a FID is programmed to semi-
continuously analyze the same gas sample stream every 6 minutes,
with the GC/FID results being recorded by the PC-DAS. The HOMER
software program converts the continuous THC signal to provide a
continuous read out of the speciated VOC concentrations (in this
case, methanol) in the gas stream using the GC/FID results. This
is done by determining the relative response factors (RRF's) for
each target VOC on the THC analyzer from gaseous standards of the
compound at known concentrations. The GC/FID response determines
the proportion of each target VOC in the gas stream. The EPA
B-8
-------
41
recently sponsored a demonstration study of this monitoring
system and showed that the HOMER monitor is a viable way to
provide continuous information on speciated organic compounds.'
For emission sources where the correlation between methanol
emissions and THC emissions can be established, continuous on-
line FID monitors are readily available as surrogate monitors.
Current technical data indicate that the above monitoring
systems are applicable for monitoring methanol emissions from
SDT's and recovery furnace vents, as well as other pulp and paper
vent streams.1'2 Although no long-term continuous monitoring
data are available, sufficient short-term integrated data have
been gathered to establish that both the GC/FID system and the
FTIR system can successfully measure methanol emissions from the
above sources.1'2'42 Data were recently submitted to EPA to
demonstrate that the techniques can meet EPA Method 301
criteria.30'31'42 It should be noted that none of the monitoring
systems suggested for continuous monitoring of methanol emissions
have been used for long-term monitoring of combustion sources at
kraft or soda pulp mills. Therefore, long-term studies are
required to determine the accuracy, precision, repeatability, and
costs of setting up and operating the monitoring systems.
The initial costs (purchase, installation, performance
testing, quality assurance program plan preparation) and annual
costs (operation & maintenance, recordkeeping and reporting,
program update, quality assurance) for a HOMER monitoring system
installed at a stack location are $135,300 and $64,500,
respectively.39 The estimated costs associated with an THC
monitoring system would be approximately $115,300 and $59,800,
respectively,33 The estimated costs associated with a FTIR
monitoring system would be approximately $162,000 and $50,000,
respectively.39
Performance specification 8A for THC systems can be found in
the background information document which supports proposed
regulations for hazardous waste combustion systems.43 Additional
performance specifications for THC systems and GC systems were
proposed on October 22, 1993 in 58 FR 54648.44 The FTIR sampling
B-9
-------
and analysis protocols for measuring gaseous organic compounds,
such as methanol, can be found in EPA Method 318.45'46
B.3.4 Parametric Monitoring Systems
Scrubber performance is easily monitored with off-the-shelf
monitoring equipment for pH and liquid flow rate. There are
various techniques for measuring liquid flow rate in a wet
scrubber including ultra sonic detection mounted externally on
the inflowing water pipe and turbine devices that are mounted
within the pipe, both of which generate an electrical signal that
can be logically displayed in a control room.38 For liquid
streams containing less than 1 percent solids, a paddle wheel
sensor can be used to measure the flow rate. If the solids
content of the liquid flow stream is greater than 1 percent, an
insertion type of sensor must be used to determine flow rate.
The cost of the liquid flow rate monitoring equipment can
vary from $2,000 to $25,000 depending on the device sensitivity
and device location from the control room.35 A complete paddle
wheel flowrate monitoring system will cost approximately
$1,200.38 The initial cost of a flow rate monitoring system with
an insertion type sensor is approximately $3,750.38 The initial
cost for a pH monitoring system is estimated to be $5,000.
B.4 PERFORMANCE TEST METHODS
The following subsections discuss the test methods
recommended for measuring emissions of particulates,
Dpacity, methanol, and HC1.
TTK-! n A^ omioa-i on (
netal HAP's,
H^ l^Jtid*,LV^*i» / t**.XA^A J.AX-*.*- • • ••—.». — — —— —
>rocedures and analytical techniques. Costs to conduct
sampling ]
Derformance testing using
-
-
rhese costs :
nclude nreoaration of the final test report.
«7here alternate test methods or procedures
:ases '
evaluated, comments are ]
lata obtained by the c
rovided on the relationships between the
r _ ._ i_ «. u»« 4— ti —% 4—
:he proposed reference methods. As appropriate,
:ould prevent the use of the reference methods to determine
:ompliance are identified.
i_ T n
-------
B.4.1 Particulate Matter and Metal HAP Emissions
Emissions of HAP metals and PM can be measured
simultaneously using EPA Method 29.47/48 Because of potential
losses of Hg during sample desiccation, it is recommended that PM
and Hg not be measured simultaneously. If these two species must
both be measured, one of the two following options should be
used:
1. Method 549 or Method 1749 for PM and Method 2948 for Hg
and other trace metals of interest, or
2. Method 101A48 for Hg and Method 2948 for PM and other
trace metals of interest.
Method 3, using the integrated sampling techniques, or
Method 3A are recommended for measuring 02 and C02.49 The
concentrations of these gases are necessary for correcting the
particulate matter emission rate to a standard basis.
The sampling and analytical cost for three Method 5 runs
with associated Method 3 testing is estimated to be $8,500 (all
dollar amounts are expressed in 1994 dollars),47 Method 17 may
be used as an alternative to Method 5 if a constant value of
0.009 g/dscm (0.004 gr/dscf) is added to the results of Method 17
and the stack temperature is no greater than 205°C (400°F).49 No
sampling and analytical cost is available for Method 17 test
runs. If a three-run Method 29 test for all trace metals and Hg
is run in conjuction with a Method 5 test for PM, the additional
cost of the Method 29 test would be approximately $8,750.47 The
estimated cost for three Method 29 runs for PM and trace metals
with associated Method 3 testing is $12,000.47 The additional
cost for three Method 101A runs in conjunction with PM/metals
testing is estimated at $5,500.47
B.4.2 Opacity
Method 9 should be used to determine the opacity of
emissions.49 The estimated cost for conducting three Method 9
observations in conjunction with PM emissions testing is
$1,060.47
B-ll
-------
B.4.3 Hydrogen Chloride Emissions
Methods 26 or 26A should be used to determine the HC1
emissions.49 If the measurement point is downstream of a wet
scrubber, Method 26A must be used. The estimated cost to conduct
a three-run performance test using Method 26 or 26A is $9,100.
However, if the Method 26A test is run in conjunction with a
Method 5 test, the additional cost for the Method 26A test would
be approximately $600 (above the cost of a Method 5 test) for
analyzing the HC1 impingers after the PM filters.50
B.4.4 Methanol Emissions
Method 308 or the Methanol Sampling Train (MST) should be
used to determine methanol emissions.32'51 The Method 308
sampling train consists of an unheated Teflon® line, two
30-milliliter (mL) water-filled midget impingers in an ice bath,
and one glass adsorbent tube packed in two sections containing
520 milligrams (mg) and 260 mg of silica gel, respectively. The
gas sample is nonisokinetically extracted from the sampling point
in the stack, and methanol is collected in deionized distilled
water and adsorbed on silica gel. The sample is returned to the
laboratory, where the methanol in the water fraction is separated
from other organic compounds with a GC and is then measured by an
FID. The fraction adsorbed on the silica gel is extracted with
an aqueous solution of n-propanol and is then separated and
measured by GC/FID. The analytical limit of guantitation (LOQ)
in the water impinger samples is 0.50 nanograms (ng) (equivalent
to about 0.5 ppm methanol in a 20-liter (L) sample of source
gas). The overall LOQ for the method is 1.6 ppm of methanol for
a 20-L sample of source gas.51
The MST consists of a glass-lined heated probe, two
condensate knockout traps in an ice bath, and three sorbent
cartridges packed with Anasorb® 747. A 1:1 mixture of carbon
disulfide (CS2) and N,N-dimethylformamide (DMF) is used to desorb
methanol from the Anasorb® samples. The condensate traps also
collect a significant amount of methanol, if water is present.
Condensate and desorption samples are analyzed by GC/FID. The
analytical LOQ in the CS2/DMF samples is 0.69 ng (equivalent to
B-12
-------
about 0.5 ppm methanol in a 20-L sample of source gas). The
overall LOQ for the MST is 3 ppm of methanol for a 20-L sample of
source gas.51
The estimated cost to conduct a three-run performance test
using Method 308 in conjunction with other"performance testing is
$9,500.35 A three-run performance test using the MST in
conjunction with other performance testing should cost about
$9,500 and offers the additional advantage that other chemical
compounds can be measured simultaneously using the same sampling
train, if desired.5 Compounds in the emission matrix can be
identified by mass spectral library matching if the desorbed
samples are analyzed on a GC equipped with an MSD and appropriate
computer software.
B-13
-------
•11-i^
if
<= £ V S 5 o
2 £ if
B-14
-------
QJ
3
a
-H
o
o
H
i
CQ
w
1
£
1
I
•o
o
•S
4)
J
S.
.G*
£
a
co
c
o
•a
1
60
C
°E.
1
CO
e
o
•a
.°-
o
K
•a
E
•§
Jo1
o
o
•o
c
•a
3
O
£
•a
•g
8
E-
Hydrogen sulfide, methyl mercaptan,
dimethyl sulfide, carbon disulfide
dimethyl disulfide
VO
v— 1
2>
Sfi
S)
4)
3
E
^
o
1
c
a
05 *- .
S4) JJH
•C J3
* 3 -C nt
0-,-g E 5 i
— ; C- O —i
Qi ^
*"*C £i
11 C-> S * (
u • c M
5 K % 8-
cD D •
E II 5 * -j
41 5 '? J :
- ^"S "^ '
J 'I °^"'
I'i^^
0 J « g .;
^^ ^^ *o ^
§ frT^ 8
17 1|.
IfSlJ
o P '
E "2 ' — —
J2 3 CJ C
U -J ^ eS
S
tM 2" §
Methanol, ethanol, acetone, 2-propanol,
butanone, chloroform, benzene,
bromodichloromethane, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cumer
alpha-pinene, beta-pinene, 3-carene, terp
(unspecified), p-cymene
oo
S
(O
s>
4)
3
E
r-.S'f
? 41 C
1 " J E
£>fc.S £
•" C *^ *t *£*
3 « E « 2
j *S ^ ^
3 ^r" ° TJ ^
3 - » E -S
> !1 ™ — to
ifilJ
?'?' 60-2 0
i fe .s £ u
3 "1 "I-S5
n ^ ? 4i 3
i||J|
1*1^1
= £ 1 II
">. A
i ^§
Chloromethane, bromomethane, methyle
chloride, acetone, carbon disulfide,
chloroform, trichlorofluoromethane, dim
disulfide, dimethyl sulfide,
n-hexane, 2-butanone,
bromodichloromethane, trichloroethene,
benzene, toluene, ethylbenzene, a-pinene
pinene, p-cymene
o
^
op
£j
CO
00
a
60
D
3
E
Formaldehyde, acetaldehyde, acetone,
acetophenone, methyl isobutyl ketone,
acrolein, benzaldehyde
^H
t
i>
CO
60
U
3
E
Antimony, arsenic, barium, beryllium,
cadmium, chromium, copper, lead,
manganese, mercury, nickel, silver,
selenium, thallium
Hydrogen chloride
0\ VO
CM CM
S S
IO 00
a a
60 60
O U
3 3
E E
B-15
-------
Hydrog
lil'
il
Ooo-3 i
u D,
'O
(U
g
•H
4J
fi
O
O
i
CQ
W
I
oa
u
E
oo
i
E
§3
O
5 o
OJ2
•Is
1 S
:;3 ^
"O u
d-l > J-*
P c§ ^
co S o
ll
i
O 5- -O
O (C 3
w 43 t
.WO
TT p| ae
O O o
Z»**
5'g.S
§|-
'c o I' •
" = §••?
•gsll
^ g en,
CO O =3 _
M^ *-4 EQ ^^
l|s°
i gaii
US L,
'•3 b" J> 3
— ° -S S?
llli
o
U
s.
u £
o j.
s g "»• -o a " . *
u
B-16
-------
-
Q)
-r-l
4J
.-H
O
U
H
t
CQ
W
H^
EH
1
>
|||
VO
lo
o o
3 0
ES
CO
OJ
A. §1
1 |lg-
** •* •* c £
«5*§" §•§. V
«r s 2. ¥ *** •
c § ^ o «C
1 - g s" i i
w € s ~ 'S. E
~^«2 'o f?A S*
|Hri|
if! HI
u •S g c ^ c
00
1
60
4>
3
E
O
O
en
0
OJ W
C^ UJ
M -o -a .t; 3 |
iS^IP'SS^iJSS S°S
in^il" i§ * il
i|^|SJ||Ilt S ll
ti-l^l^lfs. g -1
be° 2 M^|2't5'^ .. © g-S
il.?!J 1!
iisgshj l!i
ff2|lse-5^5-f 1 1? |1|1 s^i
^oS^^^O^O^^S oafi /^v SS ta "* ce a -,
"C « $ ?^*^- S _- ^ °° 2 » ^«E c °°co "o 5
a»"— fe«»c>.vp5d*:-.^-
-------
0 .§
o ft. S
lai
.
f
•s
c
0
o
o
s
f>
V
>
0
o
Q
1
§••§
HS
II
III
111
!!!
K'-S'-S
VO
2
CO
o
E
^
c? o
*> ^
.2'!
T3 4J
-j -0
11
"s ~
'§ El_E J
§ a -§•,£ ^
e -tf = *K *u
-^ 4i 5 M "^
5 to 3 O
.£ # _C « —
•Q tfa o c 2 S
^* o g o. .Jj
ft ^ i ? §
j-'gj'g^ g i
e3 ^S ^ ° *^ 2
00 g «" "2 -3 e£ O
'> Ji M S i § .>
=3 1 S 1 1 g 1
llfllil
CO U ? O C,J3 K
•2 c S
I •§ '^
u £<§•
£>
•s.l
• • 1
.2^
> C
|l|
•a "§ £
j: Me >.
•£ <= g J
03 iJ S o
S 2 "3 .S
Co"
fsj o 3 ^
I' -"2 -S |
« §• §• J
.Sao*
«•§ g 1
ffl (o £ ft.
B-18
-------
TJ
C
•H
JJ
fl
O
u
ffl
w
1
g
CO
d,
1
i
•g
1
u
1
i
to
e
0
1
C
a
C,
g
M
C
o
•£3
_C.
O
1
S
S
to
1
O
O
Process unit and
CO
•a
«
•D
£
•s
•s
{2
Methanol, ethanol, acetone, 2-propanol, 2-
butanone, chloroform, benzene,
bromodichloromethane, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cumene,
alpha-pinene, beta-pinene, 3-carene, terpenes
(unspecified), p-cymene
oo
S
"i
to
d}
3
E
Antimony, arsenic, barium, beryllium,
cadmium, chromium, copper, lead;
manganese, mercury, nickel, phosphorous,
silver, selenium, thallium
CM
1
4)
E
M !>
B S
Hi
S "i §
^ C u
^ .. 'aj
CO S o
1
1
o
u
.if
I
on
C
J>
'•3
•S
00
0 - J>
111
III
Hydrogen sulfide, methyl mercaptan,
dimethyl sulfide, carbon disulfide
dimethyl disulfide
VO
•-H
S
a
00
4)
3
E
Methanol, ethanol, acetone, 2-propanol, 2-
butanone, chloroform, benzene,
brdmodichloromethane, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cumene,
alpha-pinene, beta-pinene, 3-carene, terpenes
(unspecified), p-cymene
oo
S
CO
o!
M
U
3
E
Bromomethane, methylene chloride, acetone,
carbon disulfide, acrolein, dibromomethane,
2-butanone, benzene, toluene, styrene
?
oo
S o
§*n
ES
1
1
to
B
?-3
4> t)
E •=>
J §
"1 s | i
o | O ^ "| 7
4) C _*? . O
c3 § "3 »£*•£ £
"-3 3, "« E, B "°
.sil*i§
H3 r* ^ cc
,g eo
•g22|^ feS |3^^^»
p,iig||i^s|
f.^2- 23 2 S^ x =•§ E
|-« Sflllg.lg.l-i'i'S
•55* € 8-g J S 1 £ c^ B
<- M.S « e § -3 g"-o-s^S
^l-sl-lt^lgSlig
IssJSHJlilrJs'
4)
r? *
V iTf^J (N
T3 CuO O\ C^
C C ^^ *^"
*S O ^ ^
B-19
-------
•d
ul
G
XJ
O
u
t
H
i
03
ta
i
•a
3
1
•a
o
f
E
•3
f
1
(5?
1
1
t
C/3
c
^?
o
1
E
te
§
u
1
rt
•s
3
M
8
3
1
*lfi
8
>,
g
0
ll
II
0 3
E .a
„ o
i|
§1
c
fti
£
^j
§
CO
o
E
u
i -r e
~ o &•
Methanol, ethanol, acetone, 2-propanol
butanone, chloroform, benzene,
bromodichloromethane, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cum
alpha-pinene, beta-pinene, 3-carene, tei
(unspecified), p-cymene
oo
i
1
CO
ti
3
E
-5,
o -5 .i
c u o
JJ E = fi
Chloromethane, bromomethane, methy
chloride, acetone, carbon disulfide,
chloroform, trichlorofluoromethane, di
disulfide, dimethyl sulfide,
iodomethane, isooctane, n-hexane, 1-
butanone, benzene, toluene, ethylbenze
styrene, m-p- xylene, acrolein, a-pinen
pinene, p-cymene
VO
Tt
OO
CO
Is
0 0
3 0
ES
^^
3
Formaldehyde, acetaldehyde, acetone,
acetophenone, 2-butanone, methyl isob
ketone, acrolein, benzaldehyde
VO
Tt
00
CO
l~
0 0
3 0
ES
o
o
1^
> w
o ..
u O*
0 CO
•S
* . "a «R i
g a. — m •=
P 1*4 ^ Q .5* CO
3 C f T ** P o;
g p \ C*J CU 3
u ^g ob "« til o
° "^ S "S o **•
g * • fig "i
M J2"c A -1
JJ-KwM=S d~E
J^M^fo 71s -2 gE
S,"5'dc4 g.»o "3 i- 'S'o "o g; &
^ • *- C P "" •£? "^ 3 **^ *— '
o •& S '-o " •'|>~ i) 2 -J*" •§ « ~ ^
•3 *J5 *a3 M *^ ** *^ ^3 PQ c *^ u to § '
V» s» «Sj y o« *™* ^* A u c *2 rt * P
• a a .
g C C ^ (x)
tra a oo in
._ « « C CJ
co Q- Cu
-------
0)
§
-H
O
u
m
w
Pollutants/parameters measured ||
1
's
1
JO
"S.
to
c
o
•a
1
1 *
"5,
t/o
c
o
•a
_Ci.
y
Efi
0)
T3
E
•g
CO
1
o
0
•g
rt
|
10
CO
8
o
f
«
5
*S5
CO
H
A-pinene, b-pinene, 1,2,4-trichlorobenzene, 11
a-terpineol, naphthalene, di-n-butylphthalate,
phenol, 2-methylphenol, 1 ,2,4-
trichlorobenzene, naphthalene, di-n-
butylphthalate, bis(2-ethylhexyl)phthalate ||
o
§
VO
00
to
"to
M
o
3
E
4J
O
tC *.
1 * .=
tSS 1
•o
•XI
c
o
u
8
1
t^,
£*
o
o
•-
o
D
Hydrogen sulfide, methyl mercaptan,
dimethyl sulfide, carbon disulfide, dimethyl
disulfide ||
vc
i
"3
M
D
3
E
Methanol, ethanol, acetone, 2-propanol, 2-
butanone, chloroform, benzene,
bromodichloromethane, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cumene,
alpha-pinene, beta-pinene, 3-carene, terpenes
(unspecified), p-cymene
CO
S
en
«
0
3
E
Chloromethane, bromomethane, methylene
chloride, acetone, carbon disulfide,
chloroform, dimethyl disulfide, dimethyl
sulfide, iodomethane, isooctane, n-hexane, 2-
butanone, 2-chloro-l,3-butadiene,
bromodichloromethane, trichloroethane, tert-
butyl methyl ether, tetrachloroethene,
benzene, toluene, ethylbenzene, sryrene, o-
xylene, m-p- xylene, a-pinene, b-pinene,
cumene, p-cymene
0
S
o
S
5?
00
1
to
"oo
60
O
3
E
A-pinene, b-pinene, 1,2,4-trichlorobenzene,
a-terpineol, naphthalene, di-n-butylphthalate,
butylbenzylphthalate, 2-methylnaphthalene,
benzyl alcohol, 1,2,4-trichlorobenzene,
naphthalene, di-n-butylphthalate, bis(2-
ethylhexyOphthalate
o
S
o
S
VO
oo
w
*eo
60
O
S
E
ts
T3
6fl 3
1 S
o -^
cc ^
ec 3
=3 |
"g -^
J I
tc
u.
"rt
_c
1
ss
en
3
J
^>
g
v2
|
a g
00 3
S H3
> u
"o E
w M
« JD
1 1
tO in
,
= c c «
2 1> 4) 3
I'S'S-S
E 0! IB C
CO ft- O- o
B-21
-------
TJ
ni
u/
C
•H
JJ
C
0
O
*
t"*1
m
w
i
E^
i
i
t
1
1
•g
•5
o
E
*J
&
0
1
CO
c
Q
•a
g
Ul
"E,
E
CS
co
e
•a
S
E
o
w
&
1
S
•g
«
"1
I
(3
"O
.1
"S
i
r-
|
C* 43*
§i
i -
„ '-3
> =
|l
Eg
o «
•a o
S2
3 t:
S8
s>l-s
ill
£?•- •-
35 -O T3
VO
S
1
t)
g
fi «T S
_r § &
S — E S
1 tir
Methanol, ethanol, acetone, 2-pi
butanone, chloroform, benzene,
bromodichloromethane, toluene,
benzene, m-, p- xylene, o-xylen
alpha-pinene, beta-pinene, 3-car
(unspecified), p-cymene
CO
s
CO
o
£
a
«T o
It
Formaldehyde, acetaldehyde, ac
acetophenone, 2-butanone, meth
ketone, acrolein, benzaldehyde
\o
oo
CO
04~
a o
3 O
E2
a.
C
5
c S
j i
- -* *^ a. "
2 U . S co
-23 E g-g
M .= |§^ « 0
•ici'S^sJ o z
i ^" T! "o -^ i ~ *s
•C^^o"^ ». O ^*
** u> eE°p y^i-c c>
!||^l igii J
••i-llSjl Isll ^
I^^E^ aSjoSfc^-i
2-lgSls I5«^ggi|
ltdls." SlIlH..."!!
ijHi iifflii!
Bcha2;>jg SceSSoOod
Z^5S?--o esutoaOZcoZ
IS
i pf 1>
i g 1
2 -3 -a
i"! §
B-22
-------
TJ
QJ
-H
OJ
d
o
u
m
Pollutants/parameters measured
1
£
o
"3.
g
V3
C
o
•a
1
c
I
C/3
C
o
•J3
^C.
•o
I
te
O
O
1
O
£
S
«
•o
.1
CO
4>
H
1
ft. gi
« -o
1 .1
"O
Is
Jil
•32= .
QQ 3 -S
§j! -
^•s •s
S
]2
*^
to
4>
E
0
CO
E
O 3
> O
0 ^
O CO
a: w
2
i 1.1 «L- .
•"" ^R " *O W5 ^^
*^ O &* *^ tTJ U
£ "etf ^ "73 £ t/5
^llsfli
g'g'|l«|.« i
I||2-S*?-| :|
8< g .£ &8| 1
» B "-1 S 1 2 § 8
•g -S |- g g. § A | g
JiliSitS I
Swea1" J-2 S^ e
«§ §°
liiinu i
ui-c.az^o a:
•o
c
jl H
i .r ' rJ
"a, c S S
S "S ^-
ej T3 ^ C;
H co r- r~
^ » =
Methanol, ethanol, acetone, 2-propanol,
butanone, chloroform, benzene,
bromodichloromeduule, toluene, ethyl
benzene, m-, p- xylene, o-xylene, cume
alpha-ptnene, beta-pinene, 3-carene, teq
(unspecified), p-cymene
CO
2?
"to
CO
eo
o
3
E
i
t «f
Mediylene chloride, acetone, n-hexane,
butanone, benzene, dibromomethane,
toluene, styrene, o-xylene, m-p- xylene,
pinene, b-pinene, p-cymene
TC
CO
03
"S
b£ r^
0 0
3 0
ES
s>
Formaldehyde, acetaldehyde, acetone,
acetophenone, 2-butanone, methyl isobu
ketone, acrolein, benzaldehyde
5?
^
C/D
CO
to ^^
0 0
3 0
ES
£>
eg
•a
c
0
CJ
- ' «r
"fiT. vo g -g ©
B-23
-------
TJ
(U
•H
JJ
CJ
O
CJ
CQ
•g
C/)
o
•a
B.
unit and
rolei
lorid
, di
thane, 1,2-dichlo
lfide, dimethyl s
ne, 1,1,2-trichlor
zene, tetrachlorob
it-
;;
3 s
thy
hanol, acetone,
aldehyde
= .2
•§ rf
£§
!j
•g-s
^^
ES
•3 ^
b =
ID
w
3
il* *§!§•
•
M W ^" ^
to 2 O °* °
< cp_. r- eo
B-24
-------
Hill!'
! 1-ji i fin
.* Hsl 1-2 -f
1 c-lltssgg^
ll Ills a g SI1
rf"
•g-S g^^.-.- _
SB ss^'fi s^s
llllilJI
« | J s 1 •§ -g -g
< S -g '-6 =3 '£ 'S 'S
"8
« s
« £
T)
a
-H
-U
C
o
o
m
w
j
s
«2 ^
S<
2^2
c •- c
S tN U
ill
Illi
Proc
•g >ss . -
tn en
taa^Mg.Uc-S
'in.
^^, o
t-g
\O tN tf
SP
fls
""
-------
13
0)
-H
XJ
S
O
H
CQ
ia
1
1
1
11 S
II ^
Ii U
11 "
II i
1.
PolluUnts/
ample type/method
II °°
g
2
1
=
hi.
1
r
ii
Process unit and control system desct
«
M
M
u
r
|
>\
5
.
^
To
B "
« "a
- >>
§1
* *§
o E
6J
itf
^ 0
II
"1
<«
ra
u
gas/National Coun
;et Train
o -^
E'S
1
c
I
>. 3
si
oa g
BLO system, continued
0
c
^ S
I o- rf 11
§• g i J§ ^ 2 •
W M C*^ (_ « Q ^D
e S *^ .g o •§ "
•— • 2 § , *o £
^".25 ^'fi ^ "
g •§ g o J^ «T-
_..^" I'-g^ S
e
& O
gJJ
" ¥ £•
•• O 3
o jfo
3 ^ '"*
S g> 5>
c p. «)
o" E" E- e
Jljitliili
O firt
gas/heated Summa
;ter source samplinj
o •- c
ES'I
01 eo
M*
•o S -c
M H5 o „
" ~* *^ ^)
o c m ••
21 « 2 c
^vf tr
— e« >
§ ||
No. 1 and No. 2 smelt dissolving tanks. SI
gases are routed through orifice scrubber, w
wash as scrubbing medium. Scrubber efflui
into the smelt dissolving tanks.
Test run-averace oneratine conditions:
^t "n^
— p ° '
OT g Q .E ^ n
1 II 1 1 1
Itl 1
No. 1 tank:
Recovery firing rate:
1.62 million Ib BLS/day
Scrubber makeup flow rate: N/A
No. 2 tank:
Recovery firing rate:
2.89 million Ib BLS/day
Scrubber makeup flow rate:
377 gal/min
*^
2*
E
0
e
2
eO ^
•s^"
1^
ll
II
•S o
II
S >s
It
na
o
; gas/National Cout
get Impinger Train
J.-H
u
CD
|
O
z
eo
a
to
3
E
g
E
"
s gas/Modified MS
s
E
u
o
c
u
> §
8a.
U CO
•5 oo u -
"S S = -g
? -0 C C3 rt
•O « c 0 T-I •§
S ^ E -a ox ^ ">
No. 8 B&W NDCE Recovery Boiler - steai
capacity: 250,000 Ib/hr. PM emissions co
Flakt electrostatic precipitator (maximum o
0.023 gr/dscf; inlet loading: 10 gr/dscf; ma
opacity: 10% © 182,000 ACFM; specific (
area = 28 ft2 CA/KACFM; gas velocity ••
ft/sec; treatment time = 1 1 .82 seconds). C
gas exhausted to atmosphere through radial
height = 249;', internal diameter = 7' - 1<
•r
0 §
U •*
-c **' er\ a\
0 g 00 00
ia§§
B-26
-------
O
M
£
I
•a 2
.la
I
J="0
o c
o. «
II f||
•111?!!
I
^3
it
H
f
s..2
V -
II
"8
•d
0)
H
•H
JJ
«
O
u
I
i
Is
« -a
M o
«sS
3 «J
ES
5?
1:
o oo
E w
li
-1
S ii
•aH
1*
P
si
ES
.
e v
o en
to o
.. 60
3 ."S
E2
00
_c
> D,
O CO
$ §
ft.'I
a
&0
£<£
O CO
8 w
z
z
-2 -
CO
M
e
o
•
c.
o
a
•s
I
o
o
fr
1
ID
U
Q
Z
I
o o =
!§•§
a a g
°'"3,
B3 a
S
o
o
?-|r«a|
O\ C w jj
c*^ O C •** >v^
CS O O ^ K
.. a -S g »
C.-S .2 S to
il
mm
5
V
O
O O ra •«• oo
^5 *-• *^ ~^- '—
Z O.2 oo co
" P '
— e , . m n
co « iai ov o\
•< to—. — in
B-27
-------
TJ
0)
•H
JJ
§
O
CO
I-I
1*
||
MM
§.f
E2
-
§§>
•J3 C
I I
to •—'
I!
22
O 00
ci tu
§
o 9
n
tr
& en
B
B
Fu
Sta
o « £
2 §|
eo° 2
• -S 8.
CQ 03 CQ
%
CaO/day
tent: 7595
277
olid
cy
(2
Te
Prod
Lim
Z c.2 — —
B-28
-------
•a
(D
O
u
rameters measured
«
1
•o
o
3
E
I)
|
i
OT
e
1
00
C
"t
E
M
jj
•£J
U
a>
1
1
o
•a
c
1
ee
OJ
^
*M3
W
<^
g 4
ne, acrolein, methanol
:arbon tetrachloride,
roform, dichlorometha
1 ,2-dichloroethylene,
limethyl sulfide 1,1,1-
,2-trichloroethane, 1,2
trachloroethylene,
R u O . ~ _7 «3
!ff|lff
"S S o "3 — 8 "o
"o .. « -^ o y "N *-•
8 H — 01 1 'S'S
< E-g-r--s'£"fi
1
to
0 '1
® *&3
c °*
E-1-!
cx
gj g
4> H
II
4> *ft «
t> w ^3
1 |1
"2 *- go
^* _H i ^ '
> o « S
Is ll
&
-^
> "S
tf ^ rt
8 ° ^
£13 tj
C 3
3 !5 t3
Two sources tested for same pollutants: •
1) Babcock & Wilcox NDCE Recovery F
bottom ESP, firing rate = 5,4 million poi
BLS/day. BL at 73 % solids is fed to pro
tank, then to furnace.
Representative operating parameters:
1 .. ,
** C ^*
i— • C _ r**i ON
^ tit f^ S
u o ?3 p^j ^
z fe.S
*>.
^ »" S"
iT c o
Jfi
«r s *>
i 1 ">•
tf rf !
e B iT
0 0 g
|!|o
•§ js •§ S
'i &.I n
s
•§
"§•
"o
s
S "g.,
BLS Production: 3600 Ib BLS/ODTP
Liquor sulfidity: 25%-30%
Soda and sulfur makeup:
Saltcake from R8 C1O2 generator, added
mix tank
No. of liquor guns: 4
Auxiliary fuel: Natural gas or No. 1 fuel
Combustion air: 4 levels of combustion ai
BLS firing rate: 237,000 Ib BLS/hr
BL temp: 252°F
ESPO,: 1.5 %
Stack TRS: Ippm
I
§
i
O 4}
i|
•s *
tJ o
11
o ">•>
< 0
1
u £
*M ™
C 4)
O bfl
it
C4 *9>
O *^
ES
o
!l. f
"2 f*** ^3 ,•••
O 4> 3 jyj
N~' § ^ ^
Oi tlJ *J
o
.."SbiS
o "S ^ «
^'5| 1
(continued)
2) Lime kiln; can use either natural gas o
receives NCG's; PM emissions controlled
chamber, 3-field ESP followed by wet sci
8% caustic for SO2 removal.
Test run-specific operating parameters (a\
I •• if
— E _ =
CO " ^> C
-ft *• ^"^ *ff
^* &0 i e
U g » o
r^ «• ^J (J
t
S3
O
k<
S
S
o«
u
CQ
Kiln throughput: 553 tons CaO/day
Cold end temp: 1210"F
Hot end temp: 2535°F
Exit TRS concentration: 3.9 ppm
Exit VOC concentration: 8.7 ppm
Exit SO2 concentration: 56 ppm
B-29
-------
TJ
0)
B
•rl
XJ
C!
O
rj
H
U
I
|
s
jj
»
i
,0.
c
a
.§
£
fci«
•§'
€
i
g
U
1
C5
W
C
O
M3
g
o
U)
c
ts
re
e
Process unit and control system descriptio
i
.£
^
'u
•s
O
*jy
Si
o
e
o
CO
o
Ui
•o
x1
(O
2
1
1)
C
^> ^3
-3 3
b§ £" §
> a. > ft.
O V} O C/3
g u5 g tu
" u 2 o
M ° 0 $
ll ll
»*•
Is
Two sources tested for HCI:
1) No. 2 recovery furnace, DCE (cyclone) with
bottom ESP; BLS firing rate: = 1 .6 million Ib B
BLS content = 65%.
Test run-specific operating parameters (average):
L*
g. <
3. u
J~ —1 ^C Ov
Il|§
u. ca < 10
E
o
Stack temperature: 351°F
Stack flow rate: 99,255 DCFM
02 concentration: 8.4%
2) No. 3 recovery furnace, NDCE with dry-bott
ESP; BLS firing rate =4.65 million Ib BLS/d;
BLS content = 65%.
Test run-specific operatinE parameters (average)
Stack temperature: 356°F
Stack flow rate: 175,414 DCFM
O9 concentration: 9.2%
3
'§
E
3
'I
2
J3
o
S
CM
,
3
H
m
9
S
1
o
"i
eo
3
U
U
E
^
gs^
4> 3
> 0
Q
O ^
& W
E
0
£i
O
J5
Recovery furnace No. 2; DCE (cascade) w/ wet
ESP; design capacity: 1 .5 million Ib BLS/d
Test run-specific operating parameters (average)
•o
V
•S c^
111
E"
'I
s
"§
1
'E
•a
g
E*
3
£.
JS
c
3 T3
'Sjo
oo a
OQ
2-g
CM O
11
CL, KJ
S *3
a!
u S
3 3
E E
Pulp production: 858 tons/d
BL firing rate: 106 gal/min
BL solids concentration: 70%
Stack temperature: 262° F
02 concentration: 6%
Stack gas flow: 51,000 dscfm
Average total PM emissions: 0.0069 gr/dscf
CO
B-30
-------
eth
•d
0)
c
i
S J 2 « 8
-.
9
"S
is
§
Process unit and c
I
•3
1
to
r
W *s 22 ^7
1 1 1 1 g
o cc 5 5 >
-O W u -a >2
b g 'I |_ £
~> "• S "
u "a 7*1 "^ «
C -• L^l *• ^
£ « CQ •§ 6
1 " 1-^ * !
fiili!
lull?
£„
-C ca
2s u.
— 2 e
S"^ ^
^.Q**^-
S T. ?5 0 y
S £<-:«= |
.. rt .. •• .&
e ** fi S g
H! I
a>slls
!§8§o
a j f^o =
35 « O U E
•8* S
o "* 2!i
* § Z «^
u|s '
i S -S
3; w _to g;
||2S
8
a
I"
If
81
Q o
o
1-
S "S
y g
4) *fl
"11
B-31
-------
•d
Q)
•H
XJ
g
U
H
i
n
1
PolluUnts/ptrtmeten
]j
u ~
E g
"O
1
Acetaldehyde, methanol, foi
acetone, methyl ethyl ketoni
ll
Is
O W
'«!
if
u
~Q
ES
Cl>
£3
is
i "
•- 3
J 0
.S
-J= 0 "S >,
rg'fj.s |-3:|g3
_Soa-°EO- ^\
Lime Kiln; lime mud from precoat filter fed U
clean mill water used as shower water on prec
contaminated evaporator condensate used in th
mud clarifierj foul condensate from evaporatoi
steam stripped, and gases from steam stripper
the kiln; all other NCG's burned in NCG ther
oxidizer. Kiln design firing rate: 400 tons Ca
PM emissions from kiln are controlled with at
Ooeratine conditions durine samnlinc (average
fr 1
•O to
9§§
^•§-CN"X ^
«••§«&§•«.«
c^gll?!-
O jj 6Q ^3 CO ••
"^ C *^ ^^ CI Cd C"*!
•fis
~ g f>
•< CO O
u sH ~
Z cs.2 —
B-32
-------
Q)
-H
.U
C
O
o
ffl
•n
1
S
in
E
2
1
2
•8"
r
4
3
£
•a
o
"S
c
*
J3
ES
CM „
«u ''m " «, " ° f 1
Q3 « -0 Q 3 -S CJ 3 Js ® fc »>
*C O* •*• t*^ "^ ^^ ^*^ ^1 —^ O t~4 ^Q
"o£^K °~fc'o Z b"3 ' "S ^ ^2 u a"
•=gd.o ^ o 8 Q o'S11' ^foSE 3
OStalc O'SJico Z£c2 OZiES aa
a ^ £T
c W-aE« L,Jj"3 "^S
U O ^* ** rrl ""^ "^3 « C 2 «-^ fc£ C CS C c"
ex *— 'rto*™ . oCO -— ^ f. ^^ C . w C
c H u^^j: o ...'is . 0 -j "S .-!C2 o> «
t)H "O^g^.0 Z.0!c"3"uE o1^*3. Z f— ^
«§ HTIi" ^t"° D.V. 3t-^r S*-?!! oQ •»
-3* 3 o I IT * S ^ •?• ° fi <2 •!? £>>b-2 -S10 M
iJ i! ! III! iii ii, I
E'C >, s e o, SSc-53— >-5S^3 S™_o o
U lii'i i!i]!S!i «i!5 il{ !
ii Siii! ASiliii !!!! si ii
Sii H'*i illiill? ilP ill il
111 ^ISSl c7§llfiil Ills li'l si
^
si ,
rt co __ f*
I'lll
^ U ?3 ^
'§"
"O — -^
O J*iti
. o
fc-a
|".i
f-
I S"
•sj
O
5 o 2^
2 «> «S
•5 S E
1
2
E
0
u
|'l
W) »
o 5
E S
Ij
Is
u -X
. o
c. •
i"J
E'&
E -1 ^
1 S E
2 •§•!
"S "3 JS
c 2 •" g
« 'S a S,
5 83 J^ S^
ai 2 o e$
X < E S
1
*n IS
a E
^ 0
03 o
a <
DO 60 W
4) 4> 2
3 3 o
E E E
E
•g
£
o
1
i
«n
c<
aa
oi
i
i
i>
3
E
o
3
03
1
Q
to
CL.
S
C
It dissolving tank
:ock and Wilcox DCE Recovery Furnace
ng rate: 60 tons BLS/day) equipped with a
wet scrubber.
1 -S -S -o
1 n&S
i i «
C "•" O e i; O
2 o s; o m s;
!i us n g_ o
o '§ < 5 i or < S
j£o— ioCJU--
B-33
-------
(U
-H
4J
C
O
o
I
CQ
§
I
E
2
£
arame
•§*
3
1
e/method |
£
n
CO
J
C9
J
CO
c
o
•j
^c.
u
1
u
w
"3
o
u
T3
ra
'£
g
K
ft
I~
I
•3
^
li"
~ 0
£'a
II
u .^
O
Arsenic, total chroi
mercury, selenium,
manganese
combined
<
Uj .5
!i fi
CO 69
4) *S
D 4>
E E
E
3
Hexavalent chromi
oa
es
u
•3
eo
u
3
E
c
3
1
U C*
C, u
cu *^
§1 I
*• - c3
co UUH
"if
- O
tf'2
° 3
E fr
I-8.
Arsenic, total chro
mercury, selenium,
manganese
combined
<
60 ra
o 3
3 U
u. E
E
Hexavalent chromi
CQ
Ci
QJ
"§
M
O
3
E
E
0
1
""
•o
•5
'i
li
CU ^/;
§• ^
8 s
c 2
iS 2
g>|
> o
8 •§
mD
ll
jf
u
_O
u «5
•3 J < g
^•2 g, 2
D U U U -
B-34
-------
* «*
H
S
Cc!
W
ff
oi
1
EH
S
?V
W
H
EH
f^
^!
rfi
K
0
CM
H
CO
w
EH
CO
o
erf
LJ
o
UJ
0
H
H
CO
O
r >
O
^
5J
rQ
CN
i
m
w
1
O
p
S
0}
2
o
&
D
•3
f
2
e
X
0
C 08
.2 g S
Q *2
M
•H "^e
1 •*
3 •"
I'i
e-s
is
« 5
0|
ftj
S "x
Z °
2 " £
2 •§ 1
g »
U to
*C '5
=§-
-o o
o S
*o S
S.S?
CO O
2-g-o
|,|'i
1
S
<
3 -o •£
•°«
2
2
«
tl
||
Process unit and
control system
description"
V)
oss Point, M
3-05/15/93
2 ^
IMoss Point Mill,
Test dates: 04/28
2
0
S
oo
2
o
2
o>
2
a
M
4>
U,
Combined
stack outlet
»Nos. 1 &2
recovery furnaces
VO
oo
2
VO
2
0.
(S
OS
3
U.
I
No. 3 East
recovery furnace
VO
0
2
OO
2
VO
r-i
2
Ov
2
M
U
E
O
[No. 3 West
recovery furnace
09
i
1
o
2
3
ll
O 3
CO O
D
W"t
cs
•o
3
|_]
Scrubber
solution
*-
Nos. 1 and 2 sme
dissolving tanks
=
e.
Z
O
i
oo
2
S>
u.
1
Black liquor
oxidation lank
00
2
Ov
S>
3
U.
Scrubber
outlet
a
M
•a
3
O"
Scrubber
outlet
1 Lime kiln
Jay, ME
3-07/23/93
.Ov
»Androscoggin Mil
Test dates: 07/09
00
2
VO
IN
2
5,
u.
Combined
stack
Recovery furnace
2
go
2
(O
H
o
3
U.
Scrubber
outlet
a
(S
•a
3
sr
_J
Weak wash
scrubber
solution
No. 1 smelt
dissolving tank
2
41
S
u.
1
II
CO O
A-side lime kiln
B-35
-------
tl
(U
•H
Jj
e
o
o
CQ
|
§
n
2
£
*
elhod U
5
oa
x"
0
•M
I1!
2 -a
III
2 j>
ll
w '-5
11
CJ |
t|
Z °
"? o P
f2| I
ll
11
I-P
t/j a
= -g
•
g
<
•s 8 §
•Co
~
s
«>
t/j
.5 §
'n/-
I-S
Process unit and
control system
descrintion1
1-08/31/93
SSI
•88
UJ
I-
CO
2
2
•S3
2
VO
rJ
2
us
E
u U
t \
CO C
^c
i
'•3
co £
00
2
Cv
^
V>
r-
c
E
i
u
i
et
o
CO
2
^
VI
2
a
£
J!
|| Lime Kiln
£8
£ a.
1S
ig i
5 -
S »
SH
•o
2
0
2
ll
CO
2
C-)
o>
2
M
CO
o
E
1
1
H »
o '
ZtE
2
o
2
ll
CO
S
1
2*
to
M
«)
E
o
I
a 0
TT
O
Z
2>
i
M
u.
"
J
J
c
i
^
2
09
U)
4)
3
u-
V
.0
"Z-.
tn
Q
c3
2
3
er
-J
Scrubber
liquid
b
0)
zl
B-36
-------
•o
(1)
fl
-H
4J
C
o
u
CN
i
ffl
H
J
9
a
c
J3
S
u
3
S
2
73
Si
1
g
1
S ».
§11
s s
w
i1!
§i
u
c -a
g'S
•e 2
U 2
0
f-S
z
"S
2 |a
(2-s i
O lio
I"!
•S'l
v £A
Q. C1
CO O
£•£•=
•o § •§
>»*3 «
S
"c
o
<
-si
3 "^ -o
E^^-S
5
U
0
fjl
CO ~
•Is
n
CO —
Process unit and
control system
description8
VO
S
s
•Is
1
•a
o
s
<
rs
w
ce
M
1 '
ii
CO 0
22
"H
"3
o"
1 Scrubber
solution
No. 2 Smelt
dissolving tank
a
a.
|
i
00
i
<
s
ce
B8
3
U.
4)
OA
2
ii
u, >
vo
CO
2
<
cs
tn
a
I
1 Second stage
vent
Black liquor
oxidation system
<->
2
OO
i
<
vn
2
cs
to
J
1 Scrubber
outlet
O
vn
S
1
Scrubber
solution
JS
1
o
2
00
2
^
*n
2
M
OB
U
E
s
11
U 3
CO O
n vn
vn o
22
'5
cr
-J
Scrubber
solution
JS
cs
0
iconderoga, N\
3-11/03/93
Ticonderoga Mill, T
Test dates: 10/21/9
i
m
s
X
&
C
vo
o
S
w
2
VO
00
2
VO
S
^
S
s
»
«J
o
3
U,
Recovery furnace
1
OO
2
1
^
ts
tt
an
E
C
V
M25D&
M305
"3
.2"
Smelt dissolving
tank
i
r-i
cs
o
2
UJ
r*
2
vo
2
OO
2
•«•
cs
1
1
0
E
0
1
.2"
Lime kiln
u, AL
3-09/24/93
IRiverdale Mill, Seln
Test dates: 09/09/9
00
2
1
o
3
U.
D
rs
i-i
£ .5*
Scrubber
outlet
_c
i
o
oo
2
vo
r4
S
tfl
CO
00
§
E
b
O
o
h. o
j , O
2*2
oo
2
VO
2
«
E
I1
1.
ll
B-37
-------
•d
0)
C
O
U
Black liquor
oxidation
*)
o>
r-i
2
ce
CO
tu
L.
V
_o
*5 «
11
i
d
B-38
-------
•d
cu
-H
4J
ffl
•g
c
S
=5
a.
£
S
S
o
»
D
O
f
2
c
O
«» 4)
ill
Q
>v-0 g
| i|
II
= 0
W -0
§1
s i
e
e
as
§•1
2 °
_•§£
£f 1
t> S
3 "°
t'l
en o
8-g-a
">, ° '"
«
c
E
i
<
|2 1
tn
^2
2
&
II
0?
ll
li
to —
Process unit an
control system
description8
!, AL
-05/17/93
5 §
Mobile Mill, Mo
Test date: 04/28
«
2
0
2
oo
2
VO
2
2
£
§
u.
11
CO O
No. 1 recovery
furnace
^s
2
o
2
00
2
(S
2
2
£
3
U,
4J
Stack B out
INo. 1 recovery
furnace
*o
2
o
2
CO
2
2
2
M
3
UL
1
b
1
•^ S
ll
OO
2
VO
2
"O
o
2
w
3
U.
Scrubber
outlet
»No. 1 smelt
dissolving tank
00
2
VQ
2
13
2
S>
3
U.
Scrubber
outlet
«No. 4 smelt
dissolving tank
X
0.
Z
a
oo
2
S>
3
U,
1
Black liquor
oxidation
oo
i
2
CO
3
U.
Scrubber
outlet
4)
6
oo
2
CN
S
M
3
U.
1
11
U 3
50 O
1
6
•s
I
I
§
ff
O
.^ 2
C w
111
111
°sl
« o *
121
i|l
"PI
III
° "S C
111
2 .2 u
S.S =5
III
§ o u
§•5^
E
•s
s
«
I
I
u
£«
O.IS
•r- "3
£2
5 «
°2
El
S =
1^
-° ^>
la
f!
|ft
is
•« I
*l
ll
^.H
2.2
« "O
« V
•s-s
<£<£<£ <°<°
H H
84> 4) rr, D O
4J IU JP- 4) «J
O CO 03 2 « C/5
J5 O "O 4>
B-39
-------
1
M
21
«
I
H
a
1
CO
H
CO
CO
H
Ql
n
0 0 ^
•O T3 C ^?
** .2 *" v&
0*° °
!--
>» c -C
rt
« oeR
C3 U
O
a 2 g^
f? OT "^
If,
'§ o-
§ e
D
u.
o
s i
a s
o g,
E
0
i S
O C-
•£•"0 jj£
0. <
"t
t
(
I
II
II
CM O CO O O O
O\ CO f~ d d JO
O CM CM 00 O\ CM
CM en en o O co
*-* ^^ *•"• CM CM
CM TJ- m vo p en
en en en —• m en
Tt CO CM •— CO CO
vo Q t— en •<*• en
vo o m ON cM O
«n ^ ob" cs O •*
co rt- in •<*• m vo
o r- t~ m r~ vo
ON CM in en o\ CM
CM M- m en c— vo
•a- t- t- •* CM cr
t*» vo vo t~- CM *^
iZ i: i: — en en
15
3
O
«-« Ul
JO J^
3 -2
o 5
ij ^> O
*3 *"* w *"
! 1 1 1
1 M -s i -
C P C ^ ** e
•2 £ -2 S e ^
b j^ & ^ -2 J
"• 5 H ® i
SU W *" W U, )
^ M .j- -K C^ 0 '
Q* ~ en co "* ^ "
a s d d s 8 E
| Z Z 2 Z ea J
«n o t~;
CM 6 t^
O O\ O
CJ\ C5 CM
CM —
CO O —
— 9 en
Svo o\
in m
-H vo_ r^
en" ^ CM"
Tf CO t~
— <_ en o\
oC
— in CM
OV t- \O
in T vo
en CM oo
O vo — -
Tf ~* C^
1
60
.S
S «3 "3
^^ C "O '
§ ^ 1 |
1 S - .i
•ago"
5 oi Z <
^5 i^ *•* *
O en CM
o m •<*
d t-^ d
CM *•"'
en O\ —<
CM — en
CM O O
co CM r-
o\ v\ vo_
t~" in" en"
vo "— ' en^
\o" -^
CO CO ~*
•* in vo
*•* en ~*~*
in m v~i
CM CM CJ
en en en
^-*
55
.
5"
3 «•
o j
1 ^
5 &
S -o
O r\
«i •;
^ 2
g %
CO
**• ° 'C
M o !
c « x
sis
g a !;
.2 £» —
^ T3 0 IS
S — "" '
in CM o o o m
O\ ~* O O O *"*
en O O
vo — oo oo co en
O\ f~ O O Q CO
en os ^
co CM in d en CM
CM •* CM CM m en
co co O O en p
co co O p O O
«n o\ co O ^ oo
d d co" rf' en" d
m m ON en — •-;
en CM •* co vo CM
•^- Tf «— ^" ~*
in p- CM — eM in
en -< m •* oo vo
en en ^^ ^ ^^ ^^
t- g> c 1
e3 rt c c Q Q
££•>>•£,.
•S <2 'g •§ "I £
en w >< J
S o o "S "S § ,
fe § § i 1 -2* S
2 en •* en •* ^
o cJ d d o ^ .
.S Z Z Z Z ca J
PH
B-40
-------
-d
QJ
a
•H
4J
C
O
O
m
w
§
-° •§ 7^ >
fe-s §"5
0 ° c i*
w tr; O 6^
« T3 o
a
c
s .
?r ^ ^
•If "s
3 8^
o
2 •- >
r?j .Is »^ *•
§o S?
O
|f
'x "^
§1
O
O
•a e"
° 3
Ul 'S
5 E.
E
§ S
•="1 -2 §
fv "o £ t— >
1 "
'H
3
O
S
.0.
jg
oo oo en
O o o o o — en
O •* O O O P- >D
»••* • ' *"^ ^^
O OO OO
O\ vo O O O t— CS
cs cs cs — '
VO »D
-H v> cs en
V|fe§§*§
- « •> »D" t* Q V^
^ ^ *" ^-T ^
oo o *o o o c*^ ON
en oo vn oo oo ID
•—" OO »— ' OS O\ CS O
vo vd •— o' o" od O
c >
7? " O
— > 00
3 ft> 5
« u o gJ «
O "Z3 IM "2 *O
"c g ^g ™ c
""* '"^ W Q
4J 4) E *C w *- ^_.
ill § ill
8 8 1. §. 5. 2 S
8 8 w e e | |
1 1 i 1 1 1 1
== g., ^ j» 'S 'S je .e
llllllll
£f . . . o o
Q O O Q ^ ,.<^ O Q
^
o cs
cs o o
•^ cs
V> 00 CS
^ 0 vo
>D '-' en
••a- >n >D
cs cs cs
|88
o\ vo Ov
en"
00 f~ O
O CS vo
TJ- »-M l— 1
CS Q O
•^ Ov OO
f» 00 O\
— cs cs
43 •"
— 0 00
~ ec c
S § |
« "° -^
^ S "o s
o X s E
_o ^ n •--
o r- cs o o
o o — o o
••M *"* *~^
cs a
v> cs cs oo oo
">»>«' 8 8
oo — < in ts «-<
en cs cs en •*
§vo o cs o
—• vo r~ o
Tf O «D OO
*— * OO V> ^ »•-<
f- cs — < en en
r~' cs o vo «n
vo" oT
O\ O OO »"•< ^J*
vo oo ID vo t—
^^ en en ^^. ^^
00000
>D en >D en ID
oo en oo en oo
>D CS CS CS CS
«-l *-.
5J O
S 5
o o
II
•f> JO
- 52
o ^ "^
•O u u OO OO
^ 3 a C C
2 E E "> •>
O -a -3 "5 "S
w tC (C O O
—• "^ t^* C-* -ri *ri
a o o ^ is
o e O O c c
— .5 O o E E
C5 S k. i- « 01
"S en — cs — cs
o
.2; o o o o o
X Z Z Z Z. X
csosoooooo
csenoooooo>
*—« '"^
voovoooooooooor-
C^IDOOOOOO\
cs cs cs cs cs
'-<«—r»csr~o
cscststscscs^-en
§!§§§§§§
vovocsovor»ooes
en" rf *J -^
1^ OO ^f ^^ C^ *"" ^D ^D
cnc*^— *"- *«-^-^ — ^^
gJ^gS^ooen
en en t"^ t^ oo ^^ en ON
cs cs — — « en —•
^ — M ^5
"5 t3 »3 —
13 ^* ^3 3
3 3 3 O
O O O u
^Illll
i §•§•§! i -
- a- o o o M «>
fe K J2 "' " -^ 5
« « DO M 00 C e XI
E S | .5 .0 ^ J -jj
f^'s's's M "§ 8
£ o .2 .2 .2 =3 'g JS
lil?^^1§«
N *" i-— K W w W -2" Jjjj
^fGfS^.^^fG'n-^
o -g
esOOOOOO^O
22ZZ2ZZ5Z
B-41
-------
•O
Q)
§
o
o
en
I
CQ
w
1
CO
o
o
s
O
_a
£
•8
u
_o
c
s>
x"
°
O
1
•g
5
C
o
•"1
•a
s
c.
„
u
1
o
^J1
c
"E
o
u
V
Is
8
i
1
0
•^
u
V}
5
«f
s
"J
^
^
"?
IS
.>
.g
£
•a
o
2
tu
o
H
3
L
^>
g
m
H
G
"c
3
tfl
CJ
S
c
I
CO
rf
'"^
ej\
en
es
en
S
ts
*—*
en"
en
en
•
8
en
-,.-.
O
1
1
55
.5
1 1
_O *"
IE
0 o
S 2
o
en
»— i
CM
t-
en
vo
en
O
CO
en"
en
en
vo
en
^.
J2
o
CQ
o
55
o
1
s
o
*—
«_(
o
2
r-
o\
es
o
en
CO
o
es
en
5?
en"
o
en
5-
en
1
0
g
en
"o
s
0
Ui
1
^!
1
M
[>
"3
tn
tA
^
t/s
f—i
O
O
o
09
s
en
en
o
vo
Ov
g
VO
en
co
2
13
o
1
*i
M
'>
"o
.a
•5
•4^
O
w
<^-
O
2
0
0
CO
g
•<*•
f-
~
l>
ON
K
1
xidation
o
0
.S"
•—
>^i
o
_B
en
vo
«n
f*
\n
en
en
CO
vo
oo
vo
vo
$
£
15
1
1
o
n
J3
S>
~"
*— «
d
2
ON
vo
—
VD
%
8
vo
1
t"~
CO
CO
^
^5
o
1;
A
X)
I
J3
'J2
u
™"
en
o'
e
o
B-42
-------
TABLE B-4. METALS ANALYZED BY EPA METHOD 29 IN TESTING AT
INTERNATIONAL PAPER MILLSa
Metals
Antimony
Arsenic
Beryl lium
Cadmium
Chromium
Cobalt
Copper
Lead
Manganese
Mercury-
Nickel
Selenium
Silver
Millsb
Moss Paint
and Mobile
/
S
S
S
s
s
s
s
s
s
s
Pine Bluff
/
/
/
/
^
^
/
/
^
/
/
/
Georgetown
/
/
/
/
^
/
/
/
/
/
S
Natchez
/
S
S
s
s
s
s
s
s
s
s
aThe same metals were analyzed for in all samples at a particular
mill, regardless of the process unit.
bNo metals testing was conducted at the Androscoggin, Erie, and
Riversdale Mills. At Ticonderoga, the metals compounds were not
identified.
B-43
-------
TABLE B-5.
VOLATILE ORGANIC COMPOUNDS SPECIATED BY METHOD 18 IN
TESTING AT INTERNATIONAL PAPER MILLS5
Ac«on«Mcrolemb
Chlorobenzene
Cumene
1,2-Dichlorethane
1 ,2-DichloroethyIene
Dimethyl disulfide
Dimethyl sulfide
Ethano!
Ethyl benzene
Ethyl benzene/
m.p-cresor
Hexichlorocyelopentadiene
Hexachloroethane
Hexachloroethane/
m,p-cresolb
n-Hextne
Isopropyl alcohol
Isopropyl alcohol/acrolein
Methano!
Melhanol/acetaldehyde
Methly chloroform
Melhylene chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl mercaptan
Napthalene/a-terpineol"
Phenol
a-Pinene
b-Pinene
a-Terpineol
Mills
/
/
/
/
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
Erie
/
/
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
/
s
s
s
s
Pine Bluff
and Natchez
/
/
/
/
/
/
^
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
Georgetown
/
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
Ticonderoga
/
/
/
/
/
/
/
/
/
/
s
s
/
/
/
/
s
s
s
s
s
s
s
s
s
/
s
s
Mobile
/
/
s
s
s
s
s
s
s
/
/
/
/
s
s
s
s
s
s
s
s
s
s
Rjverdale
/
/
/
s
s
s
/
/
/
J
s
s
s
s
s
s
s
s
s
s
s
s
/
B-44
-------
TABLE B-5. (continued)
Volatile organic compounds
Tetpenes
Toluene
Trichlorobenzene
1 , 1 ,2-trichloroethane
Trichloroethylene
m,p-Xy!ene
o-Xylene
Mills
Moss Point
/
/
/
/
/
/
/
Androscoggin
s
s
s
s
s
s
Erie
/
s
s
s
s
s
Pine Bluff
tnd Natchez
/
s
s
s
s
s
Georgetown
/
/
/
S
s
Ticonderoga
^
S
s
s
s
Mobile
/
/
/
/
/
/
Riverdale
/
/
S
s
s
s
"All samples analyzed by Method 18 at a particular mill were analyzed for the same compounds, regardless of the process unit from which the sample*
were collected.
''Compounds listed together were coelutants.
B-45
-------
TABLE B-6. VOLATILE ORGANIC COMPOUNDS SPECIATED BY LIQUID
INJECTION TESTING AT INTERNATIONAL PAPER MILLS
Volatile organic
compounds
Total VOC
Total chlorine
Acetaldehyde
Acetone
Acetone/Acrolein"
Acctophcnone
Acrolein
Benzene
Carbon tetrachloride
Carbon tetrachloride/
benzene'5
Catechol
Catfichol/Naphthaleneb
Chlorobenzene
Chlorofonn
Chloroform/n-Hexane"
m,p-Cresol
o-Cresol
Cumene
Cumene/a-terpineol
1 ,2-Dichloroethane
1 ,2-Dichloroethylene
Dimethyl disulfide
Dimethyl sulfide
Ethanol
Ethyl benzene
Ethyl benzene/
m,p-Cresolb
Hexachlorocyclo-
Hexachloroe thane
Hexachloroethane/
m,p-Cresolb
n-Hexane
Isopropyl alcohol
Mills8
Moss Point
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
^
/
Androscoggin
/
Pine Bluff
and
Natchez
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
^
/
/
/
/
Georgetown
Recovery
furnace &.
lime kiln
No. 1
/
/
/
Smelt
dissolving
tank&
limekiln
No. 2
/
/
/
/
/
/
/
/
/
/
Ticonderoga
Smelt
dissolving
tank
/
/
/
/
/
/
/
/
/
/
/
/
/
Limekiln
s
s
s
Riverdale
/
/
/"
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
B-46
-------
TABLE B-6. (continued)
Volatile organic
compounds
Isopropyl alcohol/
Acrolein"
Methanol
Methanol/
Acetaldehyde"
Methyl chloroform
Methylene chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl mercaptan
Naphthalene
Naphthalene/
a-Terpineolb
Phenol
a-Pinene
b-Pinene
a-Terpineol
Terpenes
Toluene
Trichlorobenzene
1 , 1 ,2-Trichloroeihane
Trichloroethylene
m,p-Xylene
o-Xylene
Mills*
Moss Point
/
/
/
/
/
/
/
/
/
/
y
s
s
s
s
Androscoggin
Pine Bluff
and
Natchez
^
/
/
/
/
/
/
^
/
/
/
/
/
/
/
Georgetown
Recovery
furnace &
lime kiln
No. 1
Smelt
dissolving
tank&
limekiln
No. 2
^
/
/
/
/
/
^
/
/
/
/
Ticonderoga
Smelt
dissolving
tank
/
/
/
/
/
/
/
/
^
/
/
Lime kiln
Riverdale
/
/
/
/
/
/
/
/
/
/
/
/
/
/
aNo liquid samples were collected from the recovery furnaces, lime kilns, smelt disolving tanks, or black liquor oxidation tanks at the
Mobile and Erie Mills.
''Compounds listed together were coelutants.
B-47
-------
TABLE B-7. ALDEHYDE AND KETONE COMPOUNDS ANALYZED BY RTI DNPH
. PROCEDURE IN TESTING AT INTERNATIONAL PAPER MILLS
Aldehydes and ketones
Acetaldehyde
Acetone
Acetophenone
Acrolein
Benzaldehyde
Crotonaldehyde
Cyclohexanone
Formaldehyde
Methyl ethyl ketone
Methyl isobutyl ketone
Pentanal
Propionaldehyde
Millsa
Moss Point, Pine Bluff,
Natchez, and Mobile
/
/
/
S
/
^
/
/
^
/
/
/
Georgetown
/
/
/
^
/
/
^
S
S
aNo aldehyde and ketone testing was conducted at the Androscoggin,
Erie, and Riverdale Mills. The aldehyde and ketone compounds
analyzed by this method at the Ticonderoga Mill were not identified.
B-48
-------
TABLE B-8.
DIOXIN AND FURAN COMPOUNDS ANALYZED BY METHOD 23 AT
INTERNATIONAL PAPER MILLSa
2378
12378
123478
123678
123789
1234678
2378
12378
23478
123478
123678
234678
123789
1234678
1234789
Tetrachlorodibenzo-p-dioxin
Pentachlorodibenzo-p-dioxin
Hexachlorodibenzo-p-dioxin
Hexachlorodibenzo-p-dioxin
Hexachlorodibenzo-p-dioxin
Heptachlorodibenzo-p-dioxin
Octachlorodibenzo-p-dioxin
(TCDD)
(PeCDD)
(HxCDD)
(HxCDD)
(HxCDD}
(HpCDD)
(OCDD)
Tetrachlorodibenzofuran (TCDF)
Pentachlorodibenzofuran (PeCDF)
Pentachlorodibenzofuran (PeCDF)
Hexachlorodibenzofuran (HxCDF)
Hexachlorodibenzofuran (HxCDF)
Hexachlorodibenzofuran (HxCDF)
Hexachlorodibenzofuran (HxCDF)
Heptachlorodibenzofuran (HpCDF)
Heptachlorodibenzofuran (HpCDF)
Octachlorodibenzofuran (OCDF)
Total
Total
Total
Total
Total
Total
Total
Total
Tetrachlorodibenzo-p-dioxins (TCDD)
Pentachlorodibenzo-p-dioxins (PeCDD)
Hexachlorodibenzo-p-dioxins (HxCDD)
Heptachlorodibenzo-p-dioxins (HpCDD)
Tetrachlorodibenzofurans (TCDF)
Pentachlorodibenzofurans (PeCDF)
Hexachlorodibenzofurans (HxCDF)
Heptachlorodibenzofurans (HpCDF)
aThis is a list of congeners from the analysis at the Georgetown Mill.
At Ticonderoga, the congeners were not identified.
B-49
-------
B.5 REFERENCES FOR APPENDIX B
1. Roy F. Western, Inc. Hazardous Air Pollutant Emission
Inventory for Mill C. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Revised
Draft Final. October 11, 1993. Volume 1, Section 4.6.
2. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill H. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. December 8, 1993. Volume 1, Sections 4.4-4.6.
3. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Champion International Corp., Lufkin,
Texas. Prepared for Texas Paper Industry Environmental
Committee. Report No. 06848-001-001. January 1993.
Volume 2.
4. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Champion International Corp.,
Sheldon, Texas. Prepared for Texas Paper Industry
Environmental Committee. Report No. 06848-001-001.
January 1993. Volume 3.
5. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Inland-Orange, Orange, Texas._
Prepared for Texas Paper Industry Environmental Committee.
Report No. 06848-001-001. January 1993. Volume 4.
6. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Simpson-Pasadena, Pasadena, Texas.
Prepared for Texas Paper Industry Environmental Committee.
Report No. 06848-001-001. January 1993. Volume 5.
7. Roy F. Weston, Inc. Texas Emissions Speciation Study
Emission Test Results: Temple-Inland, Inland, Texas._
Prepared for Texas Paper Industry Environmental Committee.
Report No. 06848-001-001. January 1993. Volume 6. .
8. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill G. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. May 20, 1994. Volume 1, Section 4.7.
9. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill D. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. December 13, 1993. Volume 1, Sections 4.4 and 4.6.
B-50
-------
10. Letter and attachments from Hershey, R., Potlatch Corp., to
Thorvig, L., Minnesota Pollution Control Agency. July 28,
1989. Transmittal of a copy of Interpoll Laboratories
Report No. 9-2757 containing the results of the April 1989
emissions test on the No. 8 recovery furnace at Potlatch
Corp.'s Cloquet, Minnesota mill.
11. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill J. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. December 30, 1993. Volume 1, Section 4.2.
12. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill K. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. April 19, 1994. Volume 1, Section 4.3.
13. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill L. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. March 17, 1994. Volume 1, Section 4.8.
14. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill M. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. May 25, 1994. Volume 1, Section 4.2.
15. Letter and attachments from McGaughey, J., Radian
Corporation, to Pearson, J., Federal Paper Board Company,
Inc. June 11, 1991. Transmittal of HC1 emissions test data
for the May 1991 emissions test of the Nos. 2 and 3 recovery
furnaces at Federal Paper Board's Augusta, Georgia mill.
16. Letter and attachments from Reitter, A., Consolidated
Papers, Inc., to Telander, J., EPA/ISB. June 22, 1993.
Transmittal of Interpoll Laboratories Report No. 1-3472E
containing the results of the December 1991 emissions tests
on the Nos. P2 and P3 power and Rl recovery boiler stacks at
the Consolidated Paper Kraft Plant in Wisconsin Rapids,
Wisconsin.
17. Letter and attachment from Reimer, D., Boise Cascade Paper,
to Telander, J., EPA/ISB. June 13, 1993. Transmittal of
Source Test Report: Boise Cascade Paper, International
Falls, Minnesota, Recovery Boiler Noncriteria Pollutant
Tests, February 5-9, 1991, Volume II.
18. Roy F. Weston, Inc. Hazardous Air Pollutant Emission
Inventory for Mill N. Prepared for National Council of the
Paper Industry for Air and Stream Improvement, Inc. Draft
Final. April 26, 1994. Volume 1, Section 4.2.
B-51
-------
19
Letter and attachments from Morse, T., Westvaco Bleached
Board Division to Telander, J., EPA/ISB. September 22, 1993.
Transmittal of the complete emissions test report from the
February 1990 methanol emissions test and the emission test
summaries from the April 1990 to April 1993 methanol
emissions tests for the Nos. 1 and 2 recovery furnaces, the
Nos. 1 and 2 SDT's, the BLO unit, and the lime kiln at
Westvaco's Covington, Virginia mill.
20,
21.
22
23
24,
25
26
27
28,
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Mississippi. September 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Mill. Jay, Maine. November 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Pennsylvania. December 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Arkansas. July 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Carolina. July 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
York. April 1993.
Roy F. Weston, Inc. Hazardous Air
Study. Prepared for International
Selma, Alabama. December 1993.
Pollutant Emissions
Paper. Moss Point,
Pollutant Emissions
Paper. Androscoggin
Pollutant Emissions
Paper. Erie,
Pollutant Emissions
Paper. Pine Bluff,
Pollutant Emissions
Paper. Georgetown, South
Pollutant Emissions
Paper. Ticonderoga, New
Pollutant Emissions
Paper. Riverdale Mill.
Roy F. Weston, Inc. Hazardous Air Pollutant Emissions
Study. Prepared for International Paper. Natchez,
Mississippi. September 1993.
Roy F. Weston, Inc. Hazardous Air Pollutant Emissions
Study. Prepared for International Paper. Mobile, Alabama.
September 1993.
B-52
-------
29. Air Toxics Emission Inventory Report, California Kraft Pulp
Mills. Prepared for California Air Resources Board.
Sacramento, California. Undated. Appendix B: Pooled Air
Toxics Source Test Program for Kraft Pulp Mills. The
appendix contains the November 14, 1990 emissions test
report for Gaylord Container Corp., Antioch, California; the
November 27, 1990 emissions test report for Simpson Paper
Company, Anderson, California; the December 10, 1990
emissions test report for Simpson Paper Company, Fairhaven,
California; and the December 13, 1990 emissions test report
for Louisiana Pacific Corp., Samoa, California.
30. Roy F. Weston, Inc. Field Evaluation of the Heated Summa®
Canister Source Sampling Train. Recovery Furnace, Mill D.
Prepared for the National Council of the Paper Industry for
Air and Stream Improvement, Inc. July 1993.
31. Roy F. Weston, Inc. Field Evaluation of the Heated Summa®
Canister Source Sampling Train. Smelt Dissolving Tank Vent,
Mill C. Prepared for the National Council of the Paper
Industry for Air and Stream Improvement, Inc. July 1993.
32. Environmental Protection Agency. Test Methods, Proposed
Method 308. Code of Federal Regulations. Title 40, Chapter
I, Subchapter C, Part 63, Appendix A (draft). December 17,
1993.
33. Memorandum from Harris, V., MRI, to the project file.
July 15, 1996. Revisions to Champion Sheldon and Champion
Lufkin Metals Emissions Data.
34. Memorandum from Harris, V., MRI, to the project file.
May 16, 1996. Review Findings of Metals Blank Data.
35. Memorandum from Toney, M., EPA/EMB, to Telander, J. ,
EPA/ISB. November 24, 1993. Enhanced Monitoring for Pulp
and Paper Combustion Sources.
36. Shanklin, S., L. Cone., and S. Steinsberger, Entropy
Environmentalists, Inc. Report: Evaluation of HC1
Measurement Techniques at a Hazardous Waste Incinerator.
Prepared for the U. S. Environmental Protection Agency,
Research Triangle Park. EPA Contract No. 68-02-4442.
EPA/ORD/AREAL. Undated.
37. Someshwar, A., A Study of Kraft Recovery Furnace
Hydrochloric Acid Emissions. National Council of the Paper
Industry for Air and Stream Improvement, Inc. New York.
Technical Bulletin No. 674. August 1994. pp. 11-13.
38. Memorandum from Toney, M., EPA/EMC, to Telander, J.
EPA/MICG. January 13, 1995. Costs of Enhanced Monitoring
for Pulp and Paper Combustion Sources.
B-53
-------
39. HAP GEMS Data Base, Emission Measurement Technical
Information Center Bulletin Board, Office of Air Quality
Planning and Standards Technology Transfer Network,
U. S. Environmental Protection Agency, Research Triangle
Park, North Carolina.
40. U. S. Environmental Protection Agency. Proposed Revisions
to Federal Regulations for Hazardous Waste Combustion
Devices. Code of Federal Regulations. Title 40, Chapter I,
Subchapter C, Part 264/265, Subpart O and Part 266,
Subpart H. April 19, 1996.
41. Steinsberger, S., W. Buynitzky, W. DeWees, J. Knoll,
M. Midgett, and M. Hartman, U. S. EPA Evaluation of
Hazardous Organic Mass Emission Rate (HOMER) System for
Measurement of Hazardous Organic NESHAP Emissions.
Submitted for presentation at the 86th Annual meeting of the
Air and Waste Management Association. Unpresented'.
June 1993.
42. Leavy, J., Analytical Sciences. Hazardous Air Pollutant
Testing at Natchez Mill by Merlab. International Paper
Technical Service Report No. 93.558T. Program/Project
No. 3271-14. June 16, 1993.
43. Draft Technical Support Document for HWC MACT Standards.
Volume IV: Compliance with the Proposed MACT Standards.
Appendix G. Performance Specification 8A--Performance
Specification for Total Hydrocarbon CEMS. U. S.
Environmental Protection Agency. Office of Solid Waste and
Emergency Response. February 1996.
44. U.S. Environmental Protection Agency. Test Methods,
Proposed Performance Specifications 101 and 102. Code of
Federal Regulations. Title 40, Chapter C, Proposed Part 64,
Appendix A. October 22, 1993.
45. U.S. Environmental Protection Agency. Test Methods,
Proposed Method 318. Measurement of Gaseous Formaldehyde,
Phenol, and Methanol Emissions by FTIR Spectroscopy. Code
of Federal Regulations. Title 40, Chapter C, Part 63,
Appendix A (draft). February 3, 1995.
46. U.S. Environmental Protection Agency. Test Methods, EPA
Protocol for the Use of Extractive Fourier Transform
Infrared (FTIR) Spectrometry in Analyses of Gaseous
Emissions from Stationary Industrial Sources. Code of
Federal Regulations. Title 40, Chapter C, Part 63, Appendix
A (draft). February 3, 1995.
B-54
-------
47
48
49
50,
51.
52
Memorandum from Segall, R., EPA/EMB, to Copland, R.,
EPA/SDB. October 14, 1992. Final Report on Emission
Measurement and Continuous Monitoring for Medical Waste
Incinerator Study.
U.S. Environmental Protection Agency. Test Methods, Method
29. Code of Federal Regulations. Title 40, Chapter I,
Subchapter C, Part 60, Appendix A. April 25, 1996.
U. S. Environmental Protection Agency. Test Methods. Code
of Federal Regulations. Title 40, Chapter I, Subchapter C,
Part 60, Appendix A. July 1, 1993.
Telecon. Harris, V., MRI-NCO, with Hosenfeld, J., MRI-KCO.
May 26, 1995. Costs of EPA Method 26 stack tests.
Peterson, M., B. Pate, E. Rickman, R. Jayanty, F. Wilshire,
and J. Knoll. Validation of a Test Method for the
Measurement of Methanol Emissions from Stationary Sources.
•Journal of the Air and Waste Management Association
(Pittsburgh). 45:3-11. January 1995.
Telecon. Harris, V., MRI, with Peterson, M., Research
Triangle Institute. May 24, 1995. Costs of Methanol
Sampling Train (MST).
B-55
-------
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-453/R-96-012
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Technical Support Document: Chemical Recovery Combustion
Sources at Kraft and Soda Pulp Mills
5. REPORT DATE
October 1996
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Air Quality Planning and Standards
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D1-0115
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air and Radiation
U. S. Environmental Protection Agency
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Interim Final (1991-1996)
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
National emission standards for hazardous air pollutants (NESHAP) are being proposed for the pulp and
paper industry under authority of Section 112(d) of the Clean Air Act as amended in 1990. This
technical support document provides technical data and information such as industry and equipment
descriptions, analyses of effectiveness and costs of emission control systems, and estimates of
environmental impacts of emission control options, that were used in the development of the proposed
NESHAP for chemical recovery combustion sources at kraft and soda pulp mills. A NESHAP for
noncombustion sources in the pulp and paper industry is being developed concurrently, and information
on these sources is contained in separate documents.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
black liquor
black liquor oxidation
chemical recovery
combustion scarce
hazardous air pollutants
kraft pulp mill
lime kiln
particulate matter
recovery furnace
smelt dissolving tank
soda pulp mill
NESHAP
air pollution control
pulp and paper mills
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
471
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOLETE
-------
-------
|