United States Office Of Air Quality EPA-452/D-01 -001
Environmental Protection Planning And Standards May 2001
Agency Research Triangle Park, NC 27711 DRAFT REPORT
Air
Control Measure Evaluations: The Control
Measure Data Base for the National
Emissions Trends Inventory (ControlNET)
Draft Report
-------
THIS DOCUMENT HAS NOT BEEN PEER OR ADMINISTRATIVELY REVIEWED WITHIN EPA AND IS FOR
AGENCY USE/DISTRIBUTION ONLY. DO NOT QUOTE, CITE, OR DISTRIBUTE. MENTION OF TRADE
NAMES ORCOMMERCIAL PRODUCTS DOES NOT CONSTITUTE ENDORSEMENT OR RECOMMENDATION
FOR USE.
CONTROL MEASURE EVALUATIONS:
THE CONTROL MEASURE
DATA BASE FOR THE NATIONAL
EMISSION TRENDS INVENTORY
(Control NET)
DRAFT REPORT
Prepared for:
Innovative Strategies and Economics Group
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Prepared by:
The Pechan-Avanti Group,
a unit of E.H. Pechan & Associates, Inc.
5528-B Hempstead Way
Springfield, VA 22151
September 1999
EPA Contract No. 68-D98-052
Work Assignment 1-12
Pechan Report No. 99.09.001/9004.112
-------
CONTENTS
Page
TABLES vii
ACRONYMS AND ABBREVIATIONS ix
CHAPTER I
INTRODUCTION AND BACKGROUND 1
CHAPTER II
EVALUATION OF CONTROL MEASURES FOR STATIONARY VOC
SOURCE CATEGORIES 13
A. AREA SOURCE VOC CONTROL MEASURES 13
1. Architectural and Industrial
Maintenance (AIM) Coatings 13
2. Consumer Products 19
3. Industrial Surface Coating 22
o
4. Aerosol Paints 29
5. Pesticides 30
6. Degreasing 30
B. POINT SOURCE CONTROL MEASURES 31
C. REFERENCES 32
CHAPTER III
STATIONARY SOURCE NOX 35
A. POINT SOURCE NOX CONTROL MEASURES 35
B. AREA SOURCE NOX CONTROL MEASURES 46
1. Agricultural Burning 46
2. Commercial and Residential
Water Heaters 46
C. REFERENCES 49
CHAPTER IV
STATIONARY SOURCE SO2 51
A. COST ESTIMATES FOR RETROFIT CONTROL
TECHNOLOGIES 51
1. FGD Scrubber Spreadsheet Cost
Model 54
2. Sulfuric Acid Plants 54
3. Coke Ovens 61
4. Sulfur Recovery Plants 61
B. SOURCE CATEGORY DESCRIPTIONS 61
1. Industrial Steam Generation —
Bituminous / Subbitumino
us Coal 61
2. Industrial Steam Generation —
Lignite 63
3. Industrial Steam Generation —
Residual Oil 64
4. Commercial/Institutional Steam
Generation —
Bituminous/Subbituminous Coal 66
5. Commercial/Institutional Steam
Generation — Residual Oil 67
6. Sulfuric Acid — Contact Process 69
-------
CONTENTS (continued)
CHAPTER V
Page
7. Primary Copper Smelters —
Copper Converter, Smelting
Furnace, and Roaster 70
8. Primary Zinc Smelters —
Sintering 71
o
9. Primary Lead Smelters —
Sintering 72
o
10. Petroleum Refineries — Fluid
Catalytic Cracking Units
(FCCU) 73
11. Petroleum Refineries — Claus
Sulfur Recovery 75
12. Natural Gas Processing 77
13. In-process Fuel Use —
Bituminous/Subbituminous Coal 78
14. Municipal Waste Combustors
(MWC) 79
15. Steam Generating Unit —
Coal/Oil 81
C. REFERENCES 83
STATIONARY SOURCE PM10 AND PM2 s 85
A. POINT SOURCE PM CONTROL MEASURES 85
1. Cement Manufacturing (Wet and
Dry Process) 86
2. Wood/Bark Waste (Industrial
Boilers) 95
3. Stone Quarrying - Processing 96
4. Taconite Iron Ore Processing 96
o
5. Bituminous/Subbituminous Coal
(Industrial Boilers) 97
6. Coal Mining, Cleaning, and
Material Handling 97
7. Steel Manufacturing 98
8. Iron Production 99
9. By-product Coke Manufacturing 99
10. Residual Oil (Industrial Boilers) 100
11. Fiberglass Manufacturing 100
12. Feed and Grain Terminal and
Country Elevators 100
13. Grey Iron Foundries 101
14. Catalytic Cracking Units 101
15. Glass Manufacture 102
16. Plywood/Particleboard
Operations 102
17. Asphalt Concrete 103
B. AREA SOURCE PM CONTROL MEASURES 103
1. Fugitive Dust - Unpaved Rural
Roads 104
2. Construction Activities 106
3. Fugitive Dust - Paved Roads 107
o
4. Residential Wood Combustion 108
-------
5. Waste Disposal & Recycling,
Open Burning, Residential Ill
C. REFERENCES Ill
-------
CONTENTS (continued)
Page
CHAPTER VI
STATIONARY SOURCE CARBON MONOXIDE CONTROL MEASURES 115
A. CARBON BLACK PRODUCTION 115
1. Description of Available Control
Options 115
2. Control Option Selected for
Analysis 115
B. IRON AND STEEL PRODUCTION 115
1. Description of Available Control
Options 117
2. Control Option Selected for
Analysis 117
C. PULP AND PAPER AND WOOD PRODUCTS 118
1. Description of Available Control
Options 118
D. ALUMINUM ORE PRODUCTION 118
E. BITUMINOUS/SUBBITUMINOUS COAL
COMBUSTION 118
1. Description of Available Control
Options 119
F. WOOD/BARK WASTE COMBUSTION 119
1. Description of Available Control
Options 119
2. Control Option Selected for
Analysis 119
G. NATURAL GAS COMBUSTION 119
1. Description of Available Control
Options 120
H. CHARCOAL MANUFACTURING 120
1. Description of Available Control
Options 120
2. Control Option Selected for
Analysis 120
I. MINERAL WOOL MANUFACTURING 120
1. Description of Available Control
Options 121
2. Control Option Selected for
Analysis 121
J. FLARES 121
1. Description of Available Control
Options 121
K. REFERENCES 121
CHAPTER VII
STATIONARY SOURCE AMMONIA CONTROLS 123
A. DESCRIPTION OF AVAILABLE CONTROL
OPTIONS 123
B. CONTROL OPTIONS SELECTED FOR ANALYSIS 123
1. Cattle 123
2. Poultry 126
3. Hogs 127
C. REFERENCES 127
CHAPTER VIII
-------
UTILITY SOURCE 129
A. ECU SOURCE NOX CONTROL MEASURES 129
B. ECU SOURCE SO2 CONTROL MEASURES 129
1. FGD Scrubbers 129
C. ECU SOURCE PM CONTROL MEASURES 131
D. REFERENCES 131
-------
CONTENTS (continued)
Page
CHAPTER IX
HIGHWAY VEHICLES 135
A. VEHICLE TECHNOLOGY 135
1. Tier 2 Emission Standards 135
B. IN-USE VEHICLES 136
1. Inspection and Maintenance 136
2. Heavy Duty Diesel Vehicle
Roadside Testing 137
3. Remote Sensing 139
4. Heavy Duty Retrofit Programs
for Highway Engines 140
5. Vehicle Retirement Programs 140
C. FUEL OPTIONS 141
1. Diesel 141
2. Alternative Fuel (CNG -
Compressed Natural Gas) 142
D. TRANSPORTATION SYSTEM MODIFICATIONS 142
1. Transit Improvements 142
2. Pricing Mechanisms 143
o
3. Employer Provisions of
Transportation/Buses to
Employees 144
4. Voluntary Adjustment of Work
Schedule 144
E. REFERENCES 144
CHAPTER X
NONROAD ENGINES AND VEHICLES 147
A. REGULATIONS 147
B. CONTROL OPTIONS 147
1. Diesel-Powered Engines 147
2. Gasoline-Powered Engines 147
3. Locomotives 149
4. Marine Compression Ignition
Engines (Commercial Marine
Vessels) 149
5. (Recreational) Marine Spark
Ignition Engines 149
6. Airport 151
7. Commercial/Industrial Mobile
Equipment 151
8. Lawn and Garden Equipment 151
9. Container Spillage Control
Measures 152
C. REFERENCES 153
-------
TABLES
Page
I-1 List of Control Measures 2
II-1 Summary of AIM Coating Control Measure Reductions 18
II-2 Summary of AIM Coating Control Measure Cost Effectiveness 18
II-3 Cost Effectiveness of CARB Consumer Product Regulations 22
o
II-4 Industrial Surface Coating Categories ~ CAA Baseline Controls 24
II-5 Industrial Surface Coating Categories ~ Additional Control Options 26
III-l Unit Costs for NOX Control Technologies for Non-Utility Stationary Sources 36
III-2 Size-Specific Cost Equations for Large Sources 44
III-3 Revised Low-NOx Burner Control Measure for Commercial and
Residential Water Heaters 48
IV-1 Retrofit Control Options 52
IV-2 Illustration of FGD Scrubber Cost Spreadsheet Model 55
IV-3 SO2 Groups and SCCs 56
V-l Stationary PM Controls and Cost Estimates 87
V-2 Stationary Point Source PM Groups and SCCs 92
V-3 Stationary Area Source PM Controls 105
VI-1 Carbon Monoxide Control Measures for Retrofit Applications 116
VII-1 Livestock Ammonia Controls and Cost Estimates 124
VII-2 Example Fractions of Confined and Unconfined Cattle Operations at
Specified Geographic Locations 126
VIII-1 NOX Control Costsfor Utility Boilers 130
VIII-2 SO2 Scrubber Control Costs for Utility Boilers 131
VIII-3 Equations for Estimating Capital Costsfor Fabric Filters on Utility
Boilers Fired with Oil or Coal 132
VIII-4 Equations for Estimating Annual Operating and Maintenance Costs
for Fabric Filters on Utility Boilers Fired With Oil 133
VIII-5 Equations for Estimating Annual Operating and Maintenance Costs
for Fabric Filters on Utility Boilers Fired With Coal 134
IX-1 Estimated Purchase Price Increases Due to Proposed Tier 2 Standards 136
X-l Federal Regulations Affecting Future Year Emissions 148
X-2 Control Measures for NO, Reductions from Locomotives 150
-------
ACRONYMS AND ABBREVIATIONS
|Im micrometers
acfrn actual cubic feet per minute
ACT Alternative Control Techniques
AIM Architectural and Industrial Maintenance
BAAQMD Bay Area Air Quality Management District
BACM Best Available Control Measures
BID background information document
Btu/lb British thermal units per pound
C&C consumer and commercial
CAA Clean Air Act
CARB California Air Resources Board
CATC Clean Air Technology Center
CNG compressed natural gas
CO carbon monoxide
CO2 carbon dioxide
COA Census of Agriculture
o
CTC Control Technology Center
CTG control technique guideline
DMA dimethylaniline
DPR Department of Pesticide Regulation
EPA U.S. Environmental Protection Agency's (EPA's
ESP electrostatic precipitator
FBC fluidized bed combuster
FCCU Fluid Catalytic Cracking Units
J o
FGD flue gas desulfurization
o
FIP Federal Implementation Plan
ft/min feet per minute
g/L grams per liter
gr/100 scf grains per 100 standard cubic feet
H2S hydrogen sulfide
HAP hazardous air pollutant
HBr hydrogen bromide
HDDVs heavy-duty diesel vehicles
HDV heavy-duty vehicle
HDVIP Heavy-Duty Vehicle Inspection Program
HF hydrogen fluoride
HMA hot mix asphalt
HPWH heat pump water heater
I/M Inspection and Maintenance
IAPCS Integrated Air Pollution Control System
ICI industrial, commercial, and institutional
IPM Integrated Pest Management
kg/head-yr kilograms per head per year
kPa kilopascals
LDGTs light-duty gasoline trucks
-------
ACRONYMS AND ABBREVIATIONS (continued)
LDGVs light-duty gasoline vehicles
LDTs light-duty trucks
LDVs light-duty vehicles
LEVs low-emission vehicles
LNG liquified natural gas
LP liquid propane
MACT maximum achievable control technology
mg/nm3 milligrams per normal cubic meters
MMBtu million British thermal units
MW megawatts
o
MWC Municipal Waste Combustors
NAAQS national ambient air quality standards
NESHAP National Emission Standard for Hazardous Air Pollutants
NET National Emission Trends
NGR natural gas reburn
o
NH3 ammonia
NO oxides of nitrogen
x O
NPI National Particulates Inventory
NSPS New Source Performance Standards
NSSC neutral sulfite semichemical
O&M operation and maintenance
OAQPS Office of Air Quality Planning and Standards
OTC Ozone Transport Commission
OTR Ozone Transport Region
PSIP Periodic Smoke Inspection Program
RACT reasonably available control technology
RDF refuse derived fuel
ROG reactive organic gases
RWC residential wood combustion
SCAQMD South Coast Air Quality Management District
SCC Source Classification Code
scfm standard cubic feet per minute
SCR selective catalytic reduction
SIC Standard Industrial Classification
SIP State Implementation Pkn
SNCR selective non-catalytic reduction
SO2 sulfur dioxide
SO3 sulfur trioxide
SOX sulfur oxide
tpd tons per day
tpy tons per year
TSP total suspended particulate
ULEV ultra-low emission vehide
UV/EB ultraviolet and electronic beam
UV ultraviolet
VMT vehicle miles traveled
VOCs volatile organic compounds
o L
-------
CHAPTER I
INTRODUCTION AND BACKGROUND
The U.S. Environmental Protection Agency's (EPA's) Office of Air Quality Planning
and Standards (OAQPS) established national ambient air quality standards (NAAQS) for criteria pollutants
under section 110 of the Clean Air Act (CAA). To support the development and implementation of the
NAAQS and associated policies, EPA must develop and maintain data on emission sources and potential
control measures. These databases assist EPA in analyzing the effects of different standards and/or control
strategies. This report documents development of control measures for inclusion in a comprehensive
control measure data base to support EPA analyses of policies and regulations. This data base is called
ControlNET — the control measure data base for the National Emission Trends (NET) inventory.
ControlNET will cover all criteria pollutants: oxides of nitrogen (NOX), sulfur
dioxide (SO2), carbon monoxide (CO), volatile organic compounds (VOCs), primary PM10, primary PM2 s,
and ammonia (NH3). The purpose of the ControlNET is to provide the data necessary for quick and
comprehensive regulatory impact analyses. A memorandum documenting the ControlNET interface is
provided separately. This report addresses the control measure research and evaluations conducted to
provide the parameters necessary (control measure effidency and cost) to model the impact of control
measures using data from the NET inventory.
Table 1-1 provides a complete listing of the control measures included in
ControlNET as well as those which have been researched and evaluated for future indusion (some
experimental and future measures are documented in this report; however, modeling parameters could not
be developed for indusion in ControlNET). Those with an asterisk are documented within this report. For
all other measures, refer to the Ozone/PM NAAQS report for documentation (Additional Control Measure
Evaluation Jor the Integrated Implementation of the Ozone and Paniculate Matter National Ambient Air Quality Standards, and Regional
Haze Program, Pechan, July 17, 1997).
The remainder of this report is organized by sector and pollutant as follows:
Chapter II: Stationary Source
voc
Chapter III: Stationary Source
NOX
Chapter IV: Stationary Source
SO2
Chapter V: Stationary Source
PM10 and PM2 s
Chapter VI: Stationary Source
CO
Chapter VII: Stationary Source
NH3
Chapter VIII: Utility - All Pollutants
Chapter IX: Highway Vehicles
Chapter X: Nonroad Engines and
Vehicles
-------
Table 1-1
List of Control Measures
This
Report Measure ID Source Category
Control Measure
Primary
Pollutant
Area Source Ammonia
A00101 Cattle Feedlots
A00201 Poultry Operations
A00301 Hog Operations
Point Source PM
PGELE Grain Elevators
PICIC ICI Boilers - Coal
PICIG ICI Boilers - Gas
PICIO ICI Boilers - Oil
PICIW ICI Boilers - Wood
P0101 Coke mfg - oven pushing
P0201 Coke sizing & screening - cold
P0301 Iron & steel - casthouses
P0402 Iron & steel - casthouses
P0501 Iron&steel-hot metal transfer
P0601 Mineral prod- dryers/furnaces
P0602 Mineral prod- dryers/furnaces
P0701 Phosphate rock calcining
P0801 Prim, metals-material handling
P0802 Prim, metals-material handling
P0803 Prim, metals-material handling
P0901 Min. prod. - material handling
P0902 Min. prod. - material handling
P0903 Min. prod. - material handling
P1001 Coal clean.-material handling
P1002 Coal clean.-material handling
P1003 Coal clean.-material handling
P1101 Surface mining-loading/storage
P1201 Mineral prod-loading/storage
P1301 Primary metals: vehicle travel
P1401 Surface mining: vehicle travel
P1501 Mineral prod. - vehicle travel
P1601 Kraft process
P1602 Kraft process
P1603 Kraft process
P1701 Coal cleaning - thermal dryers
Chemical Additives
Chemical Additives
Chemical Additives
Oil Suppression
Fabric Filter
Fabric Filter
Fabric Filter
Electrostatic Precipitator
Partial shed to baghouse
Total enclosure to baghouse
Total enclosure to baghouse
Local hood venting to baghouse
Movable canopy to baghouse
Venturi scrubber
Fabric filter system
Venturi scrubber
Local hood/venturi scrubber
Local hood/fabric filter
Water suppression
Local hood/venturi scrubber
Local hood/fabric filter
Water suppression
Local hood/venturi scrubber
Local hood/fabric filter
Water suppression
Water suppression
Water suppression
Chemical suppression
Chemical suppression
Chemical suppression
ESP
Scrubber
Demister
Venturi scrubber
Ammonia
Ammonia
Ammonia
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
-------
Table 1-1 (continued)
This
Report Measure ID
P1801
P1901
P2001
P10101
P10102
P10103
P11101
P12101
P12201
P13101
P14101
P14102
P14103
P14104
P15101
P16101
P17101
P17201
P18101
P18201
P19101
Source Category
Ore crushing
Ore grinding
Ore crushing/grinding
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
Residual Oil (Industrial Boilers)
Fiberglass Manufacturing
Fiberglass Manufacturing
Feed and Grain Country Elevators
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Catalytic Cracking Units
Feed and Grain Terminal Elevators
Glass Manufacture
Glass Manufacture
Plywood/Particleboard Operations
Plywood/Particleboard Operations
Asphalt Concrete
Control Measure
Fabric filter
Fabric filter
Fabric filter
Fabric Filter (Mech. Shaker)
Venturi Scrubber
Wet ESP - Wire Plate Type
Venturi Scrubber
Wet ESP - Wire Plate Type
Dry ESP-Wire Plate Type
Fabric Filter (Mech. Shaker)
Impingement-plate scrubber
Dry ESP-Wire Plate Type
Fabric Filter (Mech. Shaker)
Venturi Scrubber
Dry ESP-Wire Plate Type
Fabric Filter (Mech. Shaker)
Dry ESP-Wire Plate Type
Fabric Filter (Pulse Jet Type)
Wet scrubber system
Wet ESP - Wire Plate Type
Fabric filter or venturi scrubber
Primary
Pollutant
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
Area Source PM
PP110
PP130
PP150
PP170
PP190
PP210
PP230
PP250
PP270
PP290
PP310
PP330
PU150
Paved Road- Rural Interstate
Paved Road-Rural Oth Prin Art.
Paved Road-Rural Minor Art.
Paved Road-Rural MajorColl.
Paved Road-Rural MinorColl.
Paved Road - Rural Local
Paved Road- Urban Interstate
Paved Road-Urban Oth Freeway
Paved Road-Urban Oth Prin Art.
Paved Road-Urban Minor Art.
Paved Road - Urban Collector
Paved Road - Urban Local
Unpaved Road-Rural Minor Art.
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Vacuum Sweeping
Chemical Stabilization
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM,5
-------
Table 1-1 (continued)
This
Report Measure ID
PU170
PU190
PU210
PU270
PU290
PU310
PU330
Pagbu
Pagtl
Pcatf
Pcnst
Ppreb
Presw
Source Category
Unpaved Road-Rural Major Coll.
Unpaved Road-Rural Minor Coll.
Unpaved Road - Rural Local
Unpaved Rd-Urban Oth Prin Art.
Unpaved Road-Urban Minor Art.
Unpaved Road - Urban Collector
Unpaved Road - Urban Local
Agricultural Burning
Agricultural Tilling
Beef Cattle Feedlots
Construction Activities
Prescribed Burning
Residential Wood Combustion
Control Measure
Chemical Stabilization
Chemical Stabilization
Chemical Stabilization
Hot Asphalt Paving
Hot Asphalt Paving
Hot Asphalt Paving
Hot Asphalt Paving
Bale Stack/Propane Burning
Soil Conservation Plans
Watering
Dust Control Plan
Increase Fuel Moisture
Education and Advisory Program
Primary
Pollutant
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
PM10,
PM25
Point Source SO2
S00101
S00201
S00202
S00301
S00302
S00401
S00402
S00501
S00502
S00601
S00602
S00701
S00702
S00801
S00802
S00901
S01001
S01101
S01201
S01301
S01401
S01501
S01601
S01701
S01801
S01901
S02001
S02101
S02201
S02301
S02401
S02501
S02601
S02701
Point Source
N01101
N01103
N01104
Sulf. Acid - Contact Absorber (99.9%)
Sulf. Acid - Contact Absorber (99%)
Sulf. Acid - Contact Absorber (99%)
Sulf. Acid - Contact Absorber (98%)
Sulf. Acid - Contact Absorber (98%)
Sulf. Acid - Contact Absorber (97%)
Sulf. Acid - Contact Absorber (97%)
Sulf. Acid - Contact Absorber (93%)
Sulf. Acid - Contact Absorber (93%)
Elemental Sulfur Recovery - Qaus Stg 2
Elemental Sulfur Recovery - daus Stg 2
Elemental Sulfur Recovery - Claus Stg 3 (95-96%)
Elemental Sulfur Recovery - Claus Stg 3 (95-96%)
Elemental Sulfur Recovery - Claus Stg 3 (96-97%)
Elemental Sulfur Recovery - Claus Stg 3 (96-97%)
Sulfur Recovery - Sulfur Removal (99%)
Sulfur Recovery - Elemental Sulfur Prodctn.
Inorganic Chemical Manufacture
By-Product Coke Manufacturing
Process Heaters (Oil and Gas Production)
Primary Metals Industry
Secondary Metal Production
Mineral Products Industry
Pulp and Paper Industry (Sulfate Pulping)
Petroleum Industry
Ind. Boiler Bituminous/Subbituminous Coal
Ind. Boiler Residual Oil
Comm/lnst Boiler- Bituminous/Subbit. Coal
In-process Fuel Use - Bituminous Coal
Industrial Boilers - Lignite
Comm/lnst. Boilers - Residual Ol
Municipal Waste Combustors
Steam Generating Unit-Coal/Oil
Primary Copper Smelters
NOX
ICI Boilers - Coal/Wall
ICI Boilers - Coal/Wall
ICI Boilers - Coal/Wall
FGD
Dual absorption
Dual absorption + FGD
Dual absorption
Dual absorption + FGD
Dual absorption
Dual absorption + FGD
Dual absorption
Dual absorption + FGD
Amine Scrubbing
Amine Scrubbing + FGD
Amine Scrubbing
Amine Scrubbing + FGD
Amine Scrubbing
Amine Scrubbing + FGD
FGD
FGD
FGD
Vacuum Carbonate
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
FGD
Dual absorption
SNCR
LNB
SCR
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
SO2
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
S02
NOX
NOX
NOy
-------
Table 1-1 (continued)
This
Report Measure ID
N01201
N01301
N01401
N01402
N01403
N01404
N01501
N01502
N01503
N01504
N01601
N01602
N01603
N01604
N01701
N01702
N01703
N01704
N01705
N01801
N02001
N02101
N02104
N02201
N02204
N02207
N02210
N02211
N02212
N02301
N02302
N02401
N02402
N02403
N02404
N02405
N02406
N02501
N02502
N02503
N02504
N02505
N02506
N02507
N02601
N02602
N02603
N02604
N02605
N02606
N02607
N02701
N02702
N02703
N02704
N02705
N02706
N02707
N02801
N02802
N02901
N02902
N02903
N03001
N03002
Source Category
ICI Boilers -Coal/FBC
ICI Boilers - Coal/Stoker
ICI Boilers - Coal/Cyclone
ICI Boilers - Coal/Cyclone
ICI Boilers - Coal/Cyclone
ICI Boilers - Coal/Cyclone
ICI Boilers - Residual Oil
ICI Boilers - Residual Oil
ICI Boilers - Residual Oil
ICI Boilers - Residual Oil
ICI Boilers - Distillate Oil
ICI Boilers - Distillate Oil
ICI Boilers - Distillate Oil
ICI Boilers - Distillate Oil
ICI Boilers - Natural Gas
ICI Boilers - Natural Gas
ICI Boilers - Natural Gas
ICI Boilers - Natural Gas
ICI Boilers - Natural Gas
ICI Boilers - Wood/Bark/Stoker
ICI Boilers - MSW/Stoker
Internal Combustion Engines - Oil
Internal Combustion Engines - Oil
Internal Combustion Engines - Gas
Internal Combustion Engines - Gas
Internal Combustion Engines - Gas
Internal Combustion Engines - Gas
Internal Combustion Engines - Gas
Internal Combustion Engines - Gas
Gas Turbines - Oil
Gas Turbines - Oil
Gas Turbines - Natural Gas
Gas Turbines - Natural Gas
Gas Turbines - Natural Gas
Gas Turbines - Natural Gas
Gas Turbines - Natural Gas
Gas Turbines - Natural Gas
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Distillate Ol
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Residual Oil
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Process Heaters - Natural Gas
Adipic Acid Manufacturing
Adipic Acid Manufacturing
Nitric Acid Manufacturing
Nitric Acid Manufacturing
Nitric Acid Manufacturing
Glass Manufacturing - Container
Glass Manufacturing - Container
Control Measure
SNCR- Urea
SNCR
SNCR
Coal Reburn
SCR
NCR
LNB
LNB + FGR
SCR
SNCR
LNB
LNB + FGR
SCR
SNCR
LNB
LNB + FGR
OT + Wl
SCR
SNCR
SNCR- Urea
SNCR -Urea
IR
SCR
IR
AF RATIO
AF + IR
L-E (Medium Speed)
L-E (Low Speed)
SCR
Water Injection
SCR + Water Injecti
Water Injection
Steam Injection
LNB
SCR + LNB
SCR + Steam Injecti
SCR + Water Injecti
LNB
LNB + FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
LNB + FGR
LNB
SNCR
ULNB
LNB + SNCR
SCR
LNB + SCR
LNB
LNB + FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
Thermal Reduction
Extended Absorption
Extended Absorption
SCR
SNCR
Electric Boost
Gullet Preheat
Primary
Pollutant
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
-------
Table 1-1 (continued)
This
Report Measure ID
N03003
N03004
N03005
N03006
N03101
N03102
N03103
N03104
N03105
N03201
N03202
N03203
N03204
N03205
N03206
N03301
N03302
N03303
N03304
N03305
N03401
N03402
N03403
N03501
N03502
N03503
N03601
N03602
N03603
N03604
N03605
N03606
N03701
N03702
N03801
N03901
N04101
N04102
N04103
N04104
N04201
N04203
N04204
N04301
N04302
N04303
N04304
N04501
N04502
N04503
N04504
N04601
N04604
N04701
N04702
N04703
N04704
N04705
N04706
N04707
N04801
N04802
N04803
N04804
N04805
Source Category
Glass Manufacturing - Container
Glass Manufacturing - Container
Glass Manufacturing - Container
Glass Manufacturing - Container
Glass Manufacturing - Flat
Glass Manufacturing - Flat
Glass Manufacturing - Flat
Glass Manufacturing - Flat
Glass Manufacturing - Flat
Glass Manufacturing - Pressed
Glass Manufacturing - Pressed
Glass Manufacturing - Pressed
Glass Manufacturing - Pressed
Glass Manufacturing - Pressed
Glass Manufacturing - Pressed
Cement Manufacturing - Dry
Cement Manufacturing - Dry
Cement Manufacturing - Dry
Cement Manufacturing - Dry
Cement Manufacturing - Dry
Cement Manufacturing - Wet
Cement Manufacturing - Wet
Cement Manufacturing - Wet
Iron & Steel Mills - Reheating
Iron & Steel Mills - Reheating
Iron & Steel Mills - Reheating
Iron & Steel Mills - Annealing
Iron & Steel Mills - Annealing
Iron & Steel Mills - Annealing
Iron & Steel Mills - Annealing
Iron & Steel Mills - Annealing
Iron & Steel Mills - Annealing
Iron & Steel Mills - Galvanizing
Iron & Steel Mills - Galvanizing
Municipal Waste Combustors
Medical Waste Incinerators
ICI Boilers - Process Gas
ICI Boilers - Process Gas
ICI Boilers - Process Gas
ICI Boilers - Process Gas
ICI Boilers -Coke
ICI Boilers -Coke
ICI Boilers -Coke
ICI Boilers -LPG
ICI Boilers -LPG
ICI Boilers -LPG
ICI Boilers -LPG
ICI Boilers -Liquid Waste
ICI Boilers -Liquid Waste
ICI Boilers -Liquid Waste
ICI Boilers -Liquid Waste
1C Engines - Gas, Diesel, LPG
1C Engines - Gas, Diesel, LPG
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - Process Gas
Process Heaters - LPG
Process Heaters - LPG
Process Heaters - LPG
Process Heaters - LPG
Process Heaters - LPG
Control Measure
LNB
SNCR
SCR
OXY-Firing
Electric Boost
LNB
SNCR
SCR
OXY-Firing
Electric Boost
Gullet Preheat
LNB
SNCR
SCR
OXY-Firing
Mid-Kiln Firing
LNB
SNCR -Urea Based
SNCR- NH3 Based
SCR
Mid-Kiln Firing
LNB
SCR
LEA
LNB
LNB + FGR
LNB
LNB + FGR
SNCR
LNB + SNCR
SCR
LNB + SCR
LNB
LNB + FGR
SNCR
SNCR
LNB
LNB + FGR
OT + WI
SCR
SNCR
LNB
SCR
LNB
LNB + FGR
SCR
SNCR
LNB
LNB + FGR
SCR
SNCR
IR
SCR
LNB
LNB + FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
LNB
LNB + FGR
SNCR
ULNB
SCR
Primary
Pollutant
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
-------
Table 1-1 (continued)
This
Report Measure ID
N04806
N04807
N04901
N04902
N04903
N04904
N04905
N04906
N04907
N05001
N05002
N05401
N05402
N05403
N05404
N05501
N05502
N05503
N05504
N05505
N05601
N05602
N05603
N05604
N05605
N05801
N05802
N05803
N05804
N05805
N05901
N06001
N06101
N06102
N06103
N06104
N06105
N06202
N06302
N06402
N06503
N06602
N06703
N06802
N06902
N07001
N07101
N07201
N07301
N07401
N07503
N07603
N07702
N07802
N07901
N08012
N08103
N08203
N08301
N08402
N08501
N08602
N08701
N08801
N08901
Source Category
Process Heaters - LPG
Process Heaters - LPG
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Process Heaters - Other Fuel
Gas Turbines - Jet Fuel
Gas Turbines - Jet Fuel
Space Heaters - Distillate Oil
Space Heaters - Distillate Oil
Space Heaters - Distillate Oil
Space Heaters - Distillate Oil
Space Heaters - Natural Gas
Space Heaters - Natural Gas
Space Heaters - Natural Gas
Space Heaters - Natural Gas
Space Heaters - Natural Gas
Ammonia - NG-Fired Reformers
Ammonia - NG-Fired Reformers
Ammonia - NG-Fired Reformers
Ammonia - NG-Fired Reformers
Ammonia - NG-Fired Reformers
Lime Kilns
Lime Kilns
Lime Kilns
Lime Kilns
Lime Kilns
Comm./lnst. Incinerators
Indust. Incinerators
Sulfate Pulping - Recovery Furnaces
Sulfate Pulping - Recovery Furnaces
Sulfate Pulping - Recovery Furnaces
Sulfate Pulping - Recovery Furnaces
Sulfate Pulping - Recovery Furnaces
Ammonia Prod; Feedstock Desulfurization
Plastics Prod-Specific; (ABS) Resin
Starch Mfg; Combined Operatbns
By-Product Coke Mfg; Oven Underfiring
Pri Cop Smel; Reverb Smelt Furn
Iron Prod; Blast Furn; Blast Htg Stoves
Steel Prod; Soaking Pits
Fuel Fired Equip; Process Htrs; Pro Gas
Sec Alum Prod; Smelting Furn/Reverb
Steel Foundries; Heat Treating Furn
Fuel Fired Equip; Furnaces; Natural Gas
Asphaltic Cone; Rotary Dryer; Conv Plant
Ceramic Clay Mfg; Drying
Coal Cleaning-Thrml Dryer; Fluidized Bed
Fbrglass Mfg; Txtle-Type Fbr; Recup Furn
Sand/Gravel; Dryer
Fluid Cat Cracking Units; Cracking Unit
Conv Coating of Prod; Add Cleaning Bath
Natural Gas Prod; Compressors
In-Process; Bituminous Coal; Cement Kiln
In-Process; Bituminous Coal; Lime Kiln
In-Process Fuel Use;Bituminous Coal; Gen
In-Process Fuel Use; Residual Oil; Gen
In-Process Fuel Use; Natural Gas; Gen
ln-Proc;Process Gas;Coke Oven/Blast Furn
In-Process; Process Gas; Coke Oven Gas
Surf Coat Oper;Coating Oven Htr; Nat Gas
Solid Waste Disp; Gov; Other Incin; Sludge
Control Measure
LNB + SNCR
LNB + SCR
LNB + FGR
LNB
SNCR
ULNB
LNB + SNCR
SCR
LNB + SCR
Water Injection
SCR + Water Injection
LNB
LNB + FGR
SCR
SNCR
LNB
LNB + FGR
OT + WI
SCR
SNCR
LNB
LNB + FGR
OT + WI
SCR
SNCR
Mid-Kiln Firing
LNB
SNCR- Urea Based
SNCR -NH3 Based
SCR
SNCR
SNCR
LNB
LNB + FGR
OT + WI
SCR
SNCR
LNB + FGR
LNB + FGR
LNB + FGR
SNCR
LNB + FGR
LNB + FGR
LNB + FGR
LNB + FGR
LNB
LNB
LNB
LNB
LNB
LNB
LNB
LNB + FGR
LNB + FGR
LNB
SCR
SNCR- urea based
SNCR - urea based
SNCR
LNB
LNB
LNB + FGR
LNB
LNB
SNCR
Primary
Pollutant
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
-------
Table 1-1 (continued)
This
Report
Measure ID Source Category
Control Measure
Primary
Pollutant
Area Source NOX
N10001 Industrial Coal Combustion
N10002 Industrial Coal Combustion
N10101 Industrial Oil Combustion
N10102 Industrial Oil Combustion
N10201 Industrial NG Combustion
N10202 Industrial NG Combustion
N12201 Open Burning
N13201 Agricultural Burning
Point Source VOC
V0171 Terephthalic Acid Manufacture
V0211 Cellulose Acetate Manufacture
V0281 Vegetable Oil Manufacture
V0321 Carbon Black Manufacture
V0349 Beverage Can Coating
V0389 Plastic Parts Coating
V0399 Wood Furniture Coating
V0529 Aircraft Surface Coating
V0531 Whiskey Fermentation - Aging
V0541 Charcoal Manufacturing
V0561 SOCMI - Reactor Processes
V0571 SOCMI - Distillation
V0951 Bakeries
V0961 Urea Resins - General
V0971 Organic Acids Manufacture
V0981 Leather Products
V1089 Fabric Coating
V1701 Service Stations- Stage I
V1801 Printing-Letterpress
V1821 Printing - Lithographic
Area Source VOC
V21101 Dry Cleaning - perc
V21201 Dry cleaning - petroleum
V21701 Bulk Terminals
V22001 Architectural coatings
V22002 Architectural coatings
V22003 Architectural coatings
V22004 Architectural coatings
V22101 Traffic markings
V22102 Traffic markings
V22103 Traffic markings
V22104 Traffic markings
V22201 Industrial maintenance coating
V22202 Industrial maintenance coating
V22203 Industrial maintenance coating
V22204 Industrial maintenance coating
V22301 Metal coil & can coating
V22302 Metal coil & can coating
V22303 Metal coil & can coating
V22501 Wood furniture surface coating
V22502 Wood furniture surface coating
V22503 Wood furniture surface coating
V22601 Adhesives - industrial
V23201 Open top degreasing
V23202 Open top degreasing
V23203 Open top degreasing
V24001 Paper surface coating
V24101 Cold cleaning
V24102 Cold cleaning
V24103 Cold cleaning
V24401 Rubber and plastics mfg
V24501 Metal furn, appliances, parts
RACTto50tpy(LNB) NOX
RACT to 25 tpy (LNB) NOX
RACTto50tpy(LNB) NOX
RACTto25tpy(LNB) NOX
RACT to 50 tpy (LNB) NOX
RACTto25tpy(LNB) NOX
Episodic Ban (Daily Only) NOX
Seasonal Ban (Ozone Season) NOX
Incineration VOC
Carbon Adsorption VOC
Stripper and Equipment VOC
Flare VOC
Incineration VOC
Incineration VOC
Incineration VOC
Incineration VOC
Carbon Adsorption VOC
Incineration VOC
New CTG level control VOC
New CTG level control VOC
Incineration at Oven Vent VOC
RACT Extended to Other Areas VOC
RACT Extended to Other Areas VOC
RACT Extended to Other Areas VOC
Incineration VOC
Vapor Balance VOC
Carbon Adsorption VOC
New CTG to Other Areas VOC
MACT (condensers/adsorbers) VOC
MACT VOC
Balanced/Adsorber/Testing VOC
AIM Coating Federal Rule VOC
South Coast Phase I VOC
South Coast Phase II VOC
South Coast Phase III VOC
AIM Coating Federal Rule VOC
South Coast Phase I VOC
South Coast Phase II VOC
South Coast Phase III VOC
AIM Coating Federal Rule VOC
South Coast Phase I VOC
South Coast Phase II VOC
South Coast Phase III VOC
MACT VOC
BAAQMD Rule 11 Amended VOC
Incineration VOC
MACT VOC
New CTG VOC
Add-On Controls VOC
SCAQMDRule1168 VOC
MACT VOC
SCAQMD 1122 (VOC content limit) VOC
Airtight degreasing system VOC
Incineration VOC
NESHAP/MACT VOC
SCAQMD 1122 (VOC content limit) VOC
Airtight degreasing system VOC
SCAQMD - low VOC VOC
MACT VOC
-------
Table 1-1 (continued)
This
Report
Measure ID Source Category
Control Measure
Primary
Pollutant
V24502 Metal furn, appliances, parts
V24601 Automobile refinishing
V24602 Automobile refinishing
V24603 Automobile refinishing
V24701 Machn, electric, railroad ctng
V24702 Machn, electric, railroad ctng
V24803 Aerosols
V24901 Consumer solvents
V24902 Consumer solvents
V24903 Consumer solvents
V25001 Aircraft surface coating
V25101 marine surface coating
V25102 marine surface coating
V25301 Electrical/electronic coating
V25302 Electrical/electronic coating
V25401 Motor vehicle coating
V25402 Motor vehicle coating
V25901 SOCMI batch reactor processes
V26602 Open burning
V26901 Commercial adhesives
V26902 Commercial adhesives
V26903 Commercial adhesives
V27003 TSDFs
V27102 Bakeries
V27201 Cutback Asphalt
V27401 SOCMI fugitives
V27601 Petroleum refinery fugitives
V27702 Pharmaceutical manufacture
V27801 Synthetic fiber manufacture
V27901 Oil and natural gas production
V28001 Stage l-truck unloading
V28003 Stage l-truck unloading
V28401 Municipal solid waste landfil
V28501 Web Offset Lithography
V29502 Pesticide Application
V22401 Wood product surface coating
V22402 Wood product surface coating
V22403 Wood product surface coating
Utility Boilers
PUTIL Utility Boilers
SUTIL Utility Boilers
N00101 Utility Boilers
Nonroad Engines
PHDRET Nonroad Diesel Engines
VNRFG Nonroad Gasoline Engines
Highway Vehicles
mOT1 Highway Vehicles - LD Gas Trucks
mOT2 Highway Vehicles - Gasoline
mOT3 Highway Vehicles - LD Gasoline
mOT4 Highway Vehicles - LD Gasoline
mOT5 Highway Vehicles - HD Diesels
mOT6 Highway Vehicles - Gasoline
SCAQMD Limits VOC
Federal Rule VOC
CARB BARCT limits VOC
FIP Rule (VOC content & TE) VOC
MACT level of control VOC
SCAQMD Limits VOC
CARB Tier 2 standards - reform VOC
Federal Consumer Solvents Rule VOC
CARB mid-term limits VOC
CARB long-term limits VOC
MACT VOC
MACT VOC
Add-on control levels VOC
MACT VOC
SCAQMD Rule VOC
MACT VOC
Incineration VOC
New CTG VOC
Episodic ban VOC
Federal Consumer Solvents Rule VOC
CARB mid-term limits VOC
CARB long-term limits VOC
Phase I & II rules VOC
Incineration >100,000 Ibs brea VOC
Switch to emulsified asphalts VOC
Equipment and maintenance VOC
Equipment and maintenance VOC
RACT VOC
Carbon adsorber VOC
Equipment and maintenance VOC
Vapor balance VOC
Vapor balance + P/V valves VOC
RCRA standards VOC
New CTG VOC
Reformulation - FIP rule VOC
MACT VOC
SCAQMD Rule 1104 VOC
Incineration VOC
Fabric Filter PM10,
PM25
FGD Scrubber SO2
SCR NO,
Heavy Duty Retrofit Program PM10
Federal Reformulated Gasoline VOC
Tier 2 Standards VOC,NOX
Federal Reformulated Gasoline VOC,NOX
High Enhanced I/M NOX, VOC
Fleet ILEV VOC
HDDV Retrofit Program PM10
Transportation Control Package NOX, VOC
-------
Page Intentionally Blank
10
-------
CHAPTER II
EVALUATION OF CONTROL MEASURES FOR STATIONARY VOC
SOURCE CATEGORIES
This chapter evaluates potential source control measures for point and area source
VOC emitters. Table 1-1 presents a complete list of control measures incorporated into ControlNET,
differentiating between measures that are documented in this report and measures documented in previous
reports. This chapter identifies the newly developed control measures, as well as revisions to measures
developed in previous analyses. The general impetus for these revisions is the availability of new
information.
Area source measures that control VOCs are described first, followed by point
source control measures for VOC. Each subsection briefly describes the source category, available control
techniques, and the control options selected for the analysis. The discussion of the control options selected
for the analysis includes an evaluation of emission reductions and total annualized costs. Capital and
operation and maintenance (O&M) costs are discussed for those measures for which information was
identified to estimate these costs.
A. AREA SOURCE VOC CONTROL MEASURES
w.
This section discusses area source control measures for VOC. The penetration rate,
hich is an estimate of the fraction of emissions covered by each measure, accounts for source size cutoffs
and other exemptions from regulation. For most area source measures, the penetration rate is included
within the control measure efficiency.
1. Architectural and Industrial Maintenance (AIM) Coatings
Section 183 of the CAA requires EPA to list for regulation products within the
consumer and commercial (C&C) products category that account for at least 80 percent of VOC emissions
from this category in ozone nonattainment areas. In 1995, EPA completed a report to Congress which
prioritized into groups the products that would be subject to regulation (EPA, 1995a). The AIM coatings
source category is in the first grouping of C&C products to be regulated because it is one of the largest
product categories, accounting for about 9 percent of the VOC emissions from all C&C products
(60FR15264, 1995). AIM coatings are used by contractors, industry, and households, and include: interior
and exterior paints, industrial maintenance coatings, wood finishes, cement coatings, roof coatings, traffic
marking paints, and specialty coatings. Area source emissions for AIM coatings are generally classified
under Source Classification Code (SCC) 2401001000 (architectural coatings), SCC 2401100000 (industrial
maintenance coatings), and SCC 2401008000 (traffic markings).
a. Description of Available Control Options
AIM coatings are formulated with a variety of components including pigments,
resins, solvents, and different additives such as driers, anti-skinning agents, anti-sag agents, dispersing
agents, defoaming agents, preservatives, and fungicides. The primary source of emissions from AIM
coatings is the solvent component. Add-on control equipment is not a feasible control option for AIM
coatings because the application of these products involves a large number of primarily small, widespread
emission sources. Reductions in VOC emissions from AIM coatings are currently achieved in a number of
States through the use of product reformulation, product substitution, and the education of consumers
about the availability of low-VOC coatings (STAPPA/ALAPCO, 1993).
i. Federal Rule
The EPA proposed a national rule for reducing VOC emissions from specific types of
AIM coatings (61FR32729, 1996). The proposed rule, which provides uniformity over the State-level
11
-------
content limits that AIM coating manufacturers must meet, was expected to go into effect on April 1, 1997.
The final rule was promulgated in August of 1998, with compliance required one year after promulgation.
The proposed Federal rule is expected to reduce VOC emissions by 20 percent from 1990 levels (EPA,
1995a).
The proposed rule sets maximum allowable VOC content limits for 55 different
categories of AIM coatings, and would affect the manufacturers and importers of the coating products. The
EPA identified two options for manufacturers of AIM coatings to reduce VOC emissions: reformulate the
coating to increase the solids-to-solvent ratio, or reformulate the coating by substituting water for an
organic solvent (EPA, 1995a). The EPA estimated the cost of r eformulating noncompliant coatings based
on information provided by industry representatives during the regulatory negotiation process. Industry
representatives estimated the level of effort required by a representative firm to research and develop a new
prototype coating to be 2.5 scientist-years over a 3-year time period. Based on an assumed cost of
$100,000 per scientist-year over 2.5 years, EPA calculated an annualized cost of $17,772 (1991 dollars) per
reformulation (61FR32729, 19%). The EPA developed an aggregate cost-effectiveness value of $237/ton
(1991 dollars), using the estimated emission reduction of 106,000 tons of VOC per year. The EPA's cost
analysis does not include estimates of changes in equipment, materials, or labor costs associated with
product reformulation. In 1990 dollar terms, the cost effectiveness of the Federal rule is $228 per ton.1
For the final rule, EPA estimated the total annualized cost of roughly $32 million (in 1996 dollars) and the
cost effectiveness of $250 per ton (EPA, 1998a).
In its analysis of the proposed Federal rule, EPA assumed that the cost of product
reformulation would bring the VOC content limit for each noncompliant coating down to the level of the
standards. The EPA, however, noted the likelihood that some manufacturers will likely reduce the VOC
content of their coatings to levels significantly below the limits in the rule (EPA, 1996a). The "at-the-limit"
assumption, therefore, likely results in emission reductions being understated. The EPA did not account
for potential cost differences for reformulating coatings to various content limits. Instead, EPA assumed
that a reformulation has a certain cost to manufacturers regardless of the target content limit, or the
anticipated VOC reduction (Ducey, 1997). In its cost analysis, insufficient data were available for EPA to
distinguish reformulation costs between different coating types (i.e., the reformulation cost for flat paints is
equal to the reformulation cost for all other affected paint types). The EPA noted the likelihood of
reformulation costs varying from product to product (EPA, 1995b).
California
Regulations
In 1989, the California Air Resources Board (CARB) adopted a model architectural
coatings rule for use by California jurisdictions in developing their architectural coatings rules. In most
cases, the Federal VOC limits are equal to or less stringent than the limits in CARB's model rule. CARB's
estimated cost-effectiveness values range from a savings of $8,600 per ton (for pool finishes) to a cost of
$12,800 per ton of VOC reduced (for specialty enamels) (CARB, 1989). The cost differential is
attributable to the wide diversity of coatings and uncertainty about the necessary equipment modifications
associated with coating reformulation. In CARB' s cost analysis, quantifiable cost factors included price
changes and raw material costs. Factors that were not quantified in the cost analysis induded: research and
development costs, costs associated with increased surface preparation, and any necessary training costs for
using the reformulated coatings.
To date, 16 of California's air quality districts have adopted architectural coatings
rules (Jaczola, 1997). The architectural coatings rule, originally adopted by the South Coast Air Quality
Management District (SCAQMD) in 1977 (and amended several times since), is the most stringent in
California. In its most recent air quality management plan, the SCAQMD outlines a two-phase approach
for obtaining further reductions from this source category. Phase I is a recently adopted amendment to
SCAQMD's existing architectural coatings rule that establishes more stringent VOC content limits for flat,
multi-color, traffic, and lacquer coatings. Phase II represents an effort to lower the VOC content limits for
non-flat industrial maintenance primers and topcoats, sealers, undercoaters, and quick-dry enamels. Phase
I is expected to resultin a VOC emission reduction of 17.5 percent, and Phase II is expected to achieve an
'The 1991:1 990 producer price index for SIC code 2851 was applied to the 1991 cost-effectiveness value to
convert to 1990 dollars (BLS, 1996).
12
-------
additional 32 percent reduction in VOC emissions by 2008, for an overall emission reduction of nearly
50 percent beyond SCAQMD's previous architectural coatings rule (SCAQMD, 1996a). Most recently,
SCAQMD has adopted a three phase (rather than two) approach to amending AIM coating rules.
For the Phase I amendment, a SCAQMD report documents cost per gallon,
total annual cost, emission reduction and cost-effectiveness values for each of the four regulated
coating types (SCAQMD, 1996b). Cost data for Phase II controls are sparse and not well-
documented. The 1997 air quality management plan cites a cost effectiveness of $12,300 per ton of
VOC reduced; however, documentation of this estimate is not available. It was determined that this
estimate should be considered speculative because cost data have not been developed for the
individual coating categories to be covered by this measure. The South Coast notes that the process
of collecting reformulation cost data for these categories is very complex due to the resin technology
used in lower-VOC, high-performance industrial maintenance coatings (silicone-based resins, or
polyurethanes) and the number of resin systems involved (Berry, 1997).
The SCAQMD Governing Board adopted further amendments to its
architectural coatings rule (Rule 1113) at its May 14, 1999 meeting (SCAQMD, 1999a). The
amendments (phase 2 of planned revisions) will require further reformulations in two steps, first by
July 2002, and then again by July 2006, subject to studies of technical feasibility for the various paint
categories covered. When fully effective, the new requirements are expected to result in reductions
of 34 percent from baseline levels. Additional amendments (phase 3) are planned in the next decade
to further reduce emissions by about 24 percent.
b. Control Options Selected for Analysis
For the CAA baseline, a 20 percent control effectiveness is assumed for the
proposed Federal AIM coatings rule. Based on an assessment of available cost data for modeling
control options beyond the baseline, it was determined that sufficient data are available from
SCAQMD to model the Phase I amendment for flat, multi-color, traffic, and lacquer coatings. The
SCAQMD documentation for the Phase I amendment contains detailed information on the
incremental costs for tightening limits for the affected coating types beyond the limits established by
O O O J L J J
the proposed Federal rule (SCAQMD, 1996b). The Phase 2 amendment is also included as an
additional control measure with the recently available information from the South Coast. Phase 3
reductions are also included; however, reduction and cost data are highly uncertain.
The SCAQMD has identified the cost increase for different paint types
associated with its current architectural coatings content limits. For the coating types covered by
the SCAQMD Phase I amendment, the current SCAQMD coating limits are the same as the limits
in the Federal rulemaking. The content limits in the SCAQMD Phase I amendment are phased-in,
and are specified in terms of near-term limits, and future effective limits. Based on cost estimates
from paint manufacturers, SCAQMD developed per gallon costs of compliance with the near-term
and future effective limits for lacquers and flat paints. The SCAQMD estimated that manufacturers
would use an acetone formulation with an associated cost of $2 per gallon to meet the proposed 550
grams per liter (g/L) VOC limit for kcquers. To comply with a iuture effective standard of 275
g/L VOC limit for lacquers, the SCAQMD estimated that manufacturers would use water-borne
reformulations at an estimated cost of $4 per gallon. This cost estimate is based on EPA's Wood
Coatings National Emission Standard for Hazardous Air Pollutants (NESHAP) analysis (SCAQMD,
1996b; Berry, 1997). Forflats, South Coast estimated a zero cost for complying with the near-term
100 g/L limit since most flats sold in California are already in compliance with this limit. For the
future effective limit of 50 g/L, South Coast estimated an incremental cost of $4 per gallon based
O ' 1 O
on data supplied by paint manufacturers (SCAQMD, 1996b). The South Coast later determined
that flat paints were being reformulated at a lower cost, but maintained the original $4 per gallon
estimate as a worst-case scenario in its cost-effectiveness calculation. (The SCAQMD also noted
that the $4 per gallon estimate is supported by an estimate developed by New York's INFORM
group.)
These cost estimates represent the incremental cost of reformulating flats
and lacquers from the initial SCAQMD limits (equivalent to the limits in the proposed Federal rule)
to the Phase I SCAQMD limits. For the remaining two coating types regulated by the amendment
(multi-color and traffic coatings), the SCAQMD estimated that a cost savings was likely to be
associated with reformulation due to a decrease in the cost of input materials. (The estimated
13
-------
magnitude of the savings is not documented.) These per gallon costs were used as the basis for
calculating a cost effectiveness for the control option being modeled in this analysis. Because flats
and lacquers comprise 88 percent of paint sales for the categories covered by the SCAQMD Phase I
amendment, the omission of multi-color and traffic coatings from the cost-effectiveness calculation
is not expected to greatly impact the estimate developed for this analysis.
For each coating type and content limit, the required reduction in VOC
content was calculated. Based on the assumption that all VOCs would evaporate, the emission
reduction was calculated by converting the reduction per gallon to total tons reduced using national
sales data by paint type (EPA, 1996a). Costs were estimated by multiplying the cost per gallon data
described above to total gallons sold. These calcuktions result in cost-effectiveness values ranging
o o o
from $3,300 to $4,600 per ton depending on the specified limit and paint type. A weighted average
cost effectiveness of $4,500 per ton (1995 dollars) was calculated for this analysis based on national
sales data by paint type (EPA, 1996a). This value was converted to 1990 dollars using the
1995:1990 producer price index for Standard Industrial Classification (SIC) code 2851 (Paints and
Allied Products), resultingin a cost-effectiveness estimate of $3,940 per ton (BLS, 1996).
For the phase 2 amendments, the SCAQMD completed a socioeconomic
impact assessment (SCAQMD, 1999a). Staff conservatively assumed a 10 percent price increase
per gallon for compliant coatings meeting the interim standards in 2002 and a 20 percent increase
for compliant coatings meeting the final limits in 2006. It is assumed that the increase will continue
to 2015 for the final compliance data coatings, recognizing the fact that reformulation is typically a
one-time investment whose useful life goes beyond the year in which reformulation occurs. The
overall cost-effectiveness of the proposed amendments (total costs/total reductions) over the years
2002-2015, is estimated to be $13,317 per ton. Costs vary significantly among individual coatings
categories.
o
Phase 3 amendments are projected to achieve an additional 24 percent
reduction. There is currently no cost data available for the phase 3 amendments. The highest
incremental cost effectiveness estimate for any individual product for the phase 2 amendments is
$26,000 per ton (1998 dolkrs). This is about double theaverage phase 2 cost. This estimate will
be used to model phase 3 reductions. This cost estimate is highly uncertain as no specific cost data
are available. CARB is currently funding a study to examine zero-polluting stains, waterproofing
sealers, and clear wood finishes which will be used to comply with the third phase emission
reductions.
Emission
Reduction
A summary of the emission reductions by measure for AIM coatings is shown
in Table II-1.
Annual
Costs
The estimated incremental cost effectiveness and overall cost effectiveness
for each measure is shown in Table II-2.
14
-------
Table 11-1
Summary of AIM Coating Control Measure Reductions
Rule Overall Reduction (%)*
Federal AIM 20
SCAQMD Phase I 34
SCAQMD Phase II 47
SCAQMD Phase III 73
NOTE: 'Phase I is incremental reduction of 17.5% from baseline (Federal AIM); phase II is a
34% reduction from baseline; phase III is an additional 24 percent reduction from phase
Table 11-2
Summary of AIM Coating Control Measure Cost Effectiveness
Rule Incremental Cost Effectiveness Overall Cost Effectiveness
Federal AIM $228 $228
SCAQMD Phase I $3,179 $1,443
SCAQMD Phase II $10,746 $4,017
SCAQMD Phase III $20,981 $10,059
NOTE: 1990dolla-s.
15
-------
iii. Capital
Costs
Because capital cost information was not available, capital costs were not
estimated for this analysis.
It is important to highlight the fact that areas of uncertainty exist in
determining the costs for additional AIM coating regulations. First, it is difficult to quantify the
extent to which the Federal rule may result in coatings with VOC contents that are lower than the
levels specified in the proposed rule. In these cases, the incremental cost of achieving a tightened
standard may approach zero. Also, there are a number of quality considerations related to
reformulating AIM coatings that may affect the feasibility of achieving tightened standards. The
paint manufacturing industry has expressed concerns about the inability of some low-VOC coatings
to meet performance requirements that can be met by high-VOC coatings, and the undesirable
qualities of some low-VOC coatings, including longer drying times or the need for more thorough
surface preparation.
2. Consumer Products
Section 183(e) of the CAA requires EPA to regulate VOC emissions from
C&C products. In 1990, the C&C products source category accounted for approximately 6 million
tons of VOC, or about 28 percent, of all anthropogenic VOC emissions (60FR15264, 1995). In its
schedule for regulating the category, EPA divided the C&C category into four groups. In addition
to AIM coatings, the consumer products source category is a component of the first group to be
regulated under Section 183(e) of the CAA. Consumer products include (but are not limited to):
personal care products, household cleaners and disinfectants, automotive aftermarket products,
adhesives and sealants, lawn and garden products, and household insecticides. The 24 types of
products covered by EPA's 1996 proposed consumer products rule accounted for 331,000 tons of
VOC emissions in 1990 (60FR15264, 1995). The final rule for household consumer products was
promulgated in August of 1998.
L O O
For this analysis, control measures options were evaluated for the consumer
solvents source category. Consumer solvent emissions are classified under the following SCCs:
• 2461600000: Miscellaneous Non-Industrial: Commercial-
Adhesives and Sealants
• 2461850000: Miscellaneous Non-Industrial: Commercial-
Pesticide Application
• 2465000000: Miscellaneous Non-Industrial: Consumer - All
Products / Processes
• 2465100000: Miscellaneous Non-Industrial: Consumer -
Personal Care Products
• 2465200000: Miscellaneous Non-Industrial: Consumer -
Household Products
• 2465400000: Miscellaneous Non-Industrial: Consumer -
Automotive Aftermarket Products
• 2465600000: Miscellaneous Non-Industrial: Consumer -
Adhesives and Sealants
• 2465900000: Miscellaneous Non-Industrial: Consumer -
Miscellaneous Products - Not Elsewhere
Classified
• 2495000000: All Solvent User Groups
a. Description of Available Control
Options
Existing and proposed State and Federal regulations for consumer products
limit the VOC content of specific product types. Potential control techniques for consumer
products include product reformulation, repackaging, and product substitution. In its cost analysis
of consumer product reguktions, CARB determined that product reformulation is the most likely
16
-------
means of meeting VOC limits (CARB, 15)90). Several methods are available to reformulate
products, including the partial or complete replacement of VOC solvents with water or other non-
or low-VOC compounds.
Federal
Rule
In April 1996, EPA proposed a consumer products rule, which reduced the
inconsistency associated with the limits that had been adopted by individual States (61FR145 31,
1996). The final rule was promulgated in August of 1998 (EPA, 1998b). The Federal rule covers
those consumer products that EPA determined to be most amenable to reguktion, and were
capable of achieving significant VOC reductions without significant effects on product quality or
price (EPA, 1995b). The proposed rule sets maximum allowable VOC content limits for
24 consumer product categories. The EPA estimates a 20 percent reduction in VOC emissions
from the proposed consumer products rule (61FR14531, 1996). The EPA developed a cost-
effectiveness value (in 1991 dollars) for the consumer products rule of $237 per ton of VOC
reduced (EPA, 1996b). In 1990 dollar terms, the cost effectiveness of the Federal rule is $232 per
ton.2 The Federal rule required companies to do what they (in most cases) had already done to
comply with CARB's and other States' rules in existence before EPA's efforts. The cost data
developed by EPA does not allow for estimation of the costs of limits more stringent than those in
the proposed rule (Moore, 1997).
California
Regulation
The CARB plans to reduce VOC emissions from the consumer products
category using three types of control measures: near-term, mid-term, and long-term measures. Near-term
measures include VOC content limits for antiperspirants, Phase I consumer products, and Phase II
consumer products. The CARB is implementing the near-term measures as follows:
1) Initial VOC limits for:
Antiperspirants by 1993,
Phase I consumer products by 1994, and
Phase II consumer products by 1995;
2) More stringent VOC content limits for:
Antiperspirants by 1999,
Selected Phase I products by 1996 and 1999, and
Selected Phase II products by 1997 and 1998.
In its Federal rule, EPA essentially proposed CARB's initial VOC limits for
antiperspirants, Phase I consumer products, and Phase II consumer products. In total, CARB
expects to achieve a 30 percent VOC reduction in 1990 Statewide emissions for near-term measures
by 1999. The estimated cost effectiveness for its near-term measures vary by product type, and range
from a savings of $100 per ton to a cost of $2,100 per ton of VOC reduced (CARB, 1991). This
range is based on cost-effectiveness estimates for five product type categories with different cost
input assumptions and amortization periods.
Some of CARB's standards were identified as technology-forcing because
they cannot be met by manufacturers at the time of rule adoption, but can be met within the time-
frame provided by the regulation. At the time CARB was developing its rules, it did not have the
resources to research the costs associated with reformulating each of the individual consumer
products. Because the hairspray rule was controversial, CARB undertook an analysis to evaluate the
ability of manufacturers to meet the near-term hairspray standard (CARB, 1997a). The CARB analysis
of the hairspray manufacturing industry estimated an incremental cost-effectiveness range of $ 1,000
The cost estimate in the Federal rule was converted from 1991 dollars to 1990 dollars using the produc
price index for SIC code 284 (BLS, 1996).
17
-------
to $7,600 per ton. This cost-effectiveness range corresponds to the cost associated with the
tightening of the current standard (80 percent VOC content) to the future standard (55 percent
VOC content). The CARB does not plan on conducting in-depth studies of reformulating additional
product types, since the results of the hairspray analysis are reasonably close to CARB's original cost
estimates for reformulating a broad range of consumer product types (Billington, 1997).
The CARB's mid-term controls (Phase III) apply to additional consumer
products that are not affected by the near-term measures. These measures are to achieve an additional
25 percent reduction in overall VOC emissions from consumer products by 2005. In 1997, CARB
addressed the first round of mid-term measures, referred to as mid-term I. In September of 1999,
CARB proposed additional amendments for consideration by the Air Resources Board, referred to
as mid-term II. These will be considered by the Air Resources Board on October 28, 1999. The
CARB's long-term measures depend on future technological innovation and market incentive
methods that can be developed and implemented before 2010. CARB hopes to achieve an
additional 30 percent reduction from these measures, bringing total reductions to 85 percent
(CARB, 1997; 1999a).
b. Control Options Selected for Analysis
ControlNET will include a 20 percent control effectiveness in the baseline
based on the Federal consumer products rule. After assessing the available cost data for modeling
control levels beyond the baseline, the additional options include: CARB's near-term limits,
CARB's mid-term measures, and CARB's long-term measure.
' o
i. Emission
Reduction
CARB's near-term measures are slightly more stringent than the Federal rule
for some products. CARB's near-term measures are expected to result in an overall reduction of 30
percent, compared to the 20 percent reduction for the Federal rule.
CARB's mid-term measures are expected to result in an additional 25
percent reduction, bringing the total reduction to 55 percent. Mid-term I (plus other amendments
for specific products) brought estimated reductions to 40 percent. The mid-term II regulations
would reduce emissions of regulated products by about 50 percent. Regulated products contribute
approximately 20 percent of total consumer emissions in 1994. Because CARB has implemented
regulations in phases and reductions are sometimes expressed with respect to different baselines
(1990 versus 1994 versus 1997), it is difficult to determine the reduction from "uncontrolled"
emissions for each phase of regulation. Based on CARB's commitment, the following layers of
measures will be modeled:
Measure Overall Reduction
Federal AIM 20 percent
CARB Mid-term 55 percent
CARB Long-term 85 percent
ii. Annual
Costs
The Federal consumer solvents rule is estimated to have an average cost
o
effectiveness of $232per ton (1990 dollars). Table II-3 presents estimates of the cost-effectiveness
of CARB's consumer products reguktions. All costs are in 1997 dollars.
Table 11-3
Cost Effectiveness of CARB Consumer Product Regulations
18
-------
Year
1989
1990
1992
1995
1997
1997
1999
Regulation/Control Measure
Antiperspirants and Deodorants
Phase I Consumer Products
Phase II Consumer Products
Aerosol Coating Products
Hairspray
Mid-Term I Consumer Products
Mid-Term II Consumer Products
Cost-Effectiveness
($/Pound(lb) ofVOC
Reduced)
$0.54-$1.30
Net savings to $1.80
<$0.01 to $1.10
$2.85 to $3.20
$2.10 to $2.50
$0.00 to $7.10
$0.00 to $6.30
Average
$/lb
$0.92
$0.90
$0.55
$3.03
$2.25
$0.25
$0.40
Based on the emission reductions for each regulation, the overall cost
effectiveness of CARB's mid-term limits measure (includes all limits for near-term and mid-term) is
estimated at $2,340 per ton. In 1990 dollars, this is $2,129per ton.
There currently is no information available on the cost of the long-term
consumer products commitment. An incremental cost per ton of $4,680 is assumed, double the
average cost through the mid-term limits. In 1990 dollars, this is $4,257 per ton. Overall cost
effectiveness for this measure (combining near-term, mid-term, and long-term) is $2,880 per ton of
VOC reduced.
It should be noted that CARB expects costs to be incurred only through the
first 15 years or so of regulation, due to research and development and changes to production lines.
iii. Capital
Costs
Capital costs were not estimated for this analysis because capital cost
information was not available. The Federal rule indicates that there are virtually no capital costs
associated with the final rule, except for the development of new, reformukted products.
3. Industrial Surface Coating
This category covers the small industrial surface coating facilities not
captured in the point source inventory. The categories, along with information on CAA
requirements, are shown in Table II-4. Measures beyond CAA requirements that were developed
under prior modeling efforts are shown in Table II-5.
The following subsections address control measure options by individual
source category. This is folio wed by a discussion of emerging or future control options which may
be applicable across several industrial surface coating categories.
a. Wood Furniture Coating
The Wood furniture coating industry covers 10 SIC codes including: Wood
o J o
Kitchen Cabinets; Wood Household Furniture (except upholstered); Wood Household Furniture
(upholstered); Wood Television, Radios, Phonograph, and Sewing Machine Cabinets; Household
Furniture Not Classified Elsewhere; Wood Office Furniture; Public Building and Related Furniture;
' " O '
Wood Office and Store Fixtures; Furniture and Fixtures Not Elsewhere Classified; and Custom
Kitchen Cabinets. Area source emissions would typically account for the smaller facilities that are
not covered in the point source inventory.
Under the 1990 CAAA, the wood furniture industry is covered by both a
new
control technique guideline (CTG) and a maximum achievable control technology (MACT)
standard (EPA, 1996c). The MACT standard sets limits for hazardous air pollutant (HAP) content
and establishes work practices and was originally promulgated in December of 1995. VOC
reductions for the MACT standards are uncertain, as compliance with the HAP content standards
only precludes the use of certain organics which are also HAPs. The workpractice standards are
19
-------
expected to result in reductions of VOC. The new CTG, published in 1996, applies to ozone
nonattainment areas and the Ozone Transport Region (OTR). The overall level of reductions
expected for compliance with presumptive reasonably available control technology (RACT) is 47
percent at an estimated cost of $967 per ton of VOC reduced. This will affect facilities emitting 25
tons per year or more, and therefore, will not affect the entire area source inventory. The overall
reduction of 47 percent is applied to both the area and point source emitters since the breakdown of
area source emissions by size category (above vs. below 25 tons per year [tpy]) is not known.
The CTG examined alternative controls that go beyond the presumptive
RACT requirements. Hybrid waterborne systems can reduce emissions by 28 to 85 percent at a cost
effectiveness rangingfrom a savings of $462 per ton to a cost of $ 11,500 per ton reduced. A full
waterborne coating system (only technically feasible for plants with short finishing sequences) can
result in reductions of 60 to 93 percent, with a cost effectiveness ranging from $ 1860 per ton to
$8430 per ton. Add-on controls include thermal incinerators, catalytic incinerators, and a
combination of carbon adsorbers and
20
-------
^ Table 11-4
Industrial Surface Coating Categories ~ CAA Baseline Controls
Pod
223
224
225
240
245
247
250
251
253
SCCs
2401040000
2401045000
2401015000
2401020000
2401030000
2401025000
2401060000
2401090000
2401050000
2401055000
2401085000
2401075000
2401080000
2401065000
cs
1
1
1
2
1
1
1
1
1
Source Category/CAA Control Measure
Metal coil and can coating
MACT
Wood product surface coating
MACT
Wood furniture surface coating
MACT
New CTG
Paper surface coating
MACT
Percent
Reduction (%)
36
30
30
47
78
Metal furniture, appliances, misc. parts coating
MACT
Machinery, equipment, railroad coating
MACT level of control
Aircraft surface coating (aerospace)
MACT/CTG
Marine surface coating (shipbuilding)
MACT
Electronic & other electrical
36
36
60
24
Cost Per Ton
($/ton)
1,000
446
446
967
4,776
1,000
1,000
165
2,090
Notes
10-year MACT category. MACT has not
yet been proposed.
10-year MACT category: wood building
production (formerly flat wood paneling)
and Plywood and composite wood
products. MACT have not yet been
proposed. MACT reductions based on
wood furniture MACT.
MACT establishes HAP limits and sets
work practices. VOC emission
reductions are uncertain. Compliance
11/21/97.
Applies to nonattainment area & OTR
Size cutoff is 25 tpy. Requirements
include coating emission limits and work
practices. Compliance by 5/20/98.
10-year MACT category. MACT has not
yet been proposed.
10-year MACT for appliances, metal
furniture, and miscellaneous parts
categories.
1 0-year MACT for metal parts and
products. MACT has not yet been
proposed.
Options for compliance include work
practice standards for cleaning
operations, carbon adsorber use,
no-HAP strippers, and control of HAP
from spray coating and blast depainting
operations MACT promulgated 9/01/95.
Initial complance by
MACT promulgated 12/1 5/95 with initial
compliance 12/16/97. Costs for model
plants with less than 100 tpy are used for
the area source category.
References
Pechan, 1997a
Pechan, 1997a
Pechan-Avanti,
1998a&b
This report
Pechan, 1997a
Pechan, 1997a
Pechan, 1997a
Pechan-Avanti,
1998a&b
Pechan-Avanti,
1998a&b
-------
Table 11-4 (continued)
Pod
254
SCCs
2401070000
cs
1
1
Source Category/CAA Control Measure
MACT
Motor vehicle coating
MACT
Percent
Reduction (%)
36
36
Cost Per Ton
($/ton)
5,000
1,000
Notes
1 0-year MACT for metal parts and
products. MACT has not yet been
proposed. MACT reduction based on a
ROM analysis.
10-year MACT category. MACT has not
yet been proposed. Emission reductions
based on a ROM analysis.
References
Pechan, 1997a
Pechan, 1997a
NOTE:
SOURCES:
All costs are in 1990 dollars.
Pechan, 1997a; Pechan-Avanti, 1998a; Pechan-Avanti, 1998b.
-------
K Table jl-5
Industrial Surface Coating Categories ~ Additional Control Options
Pod
223
224
225
240
245
247
250
251
253
254
SCCs
2401040000
2401045000
2401015000
2401020000
2401030000
2401025000
2401060000
2401090000
2401050000
2401055000
2401085000
2401075000
2401080000
2401065000
2401070000
cs
2
3
2
3
3
2
2
2
2
2
Source Category/CAA Control Measure
Metal coil and can coating
San Francisco Bay Area Rule 11
Incineration
Wood product surface coating
SCAQMD Rule 1 104 Amended
Incineration
Wood furniture surface coating
Add-on Controls
Paper surface coating
No further controls identified
Percent
Reduction (%)
42
90
53
86
75
Metal furniture, appliances, misc. parts coating
SCAQMD Rule 11 07 Amended
Machinery, equipment, railroad coating
SCAQMD Rule 1 107 Amended
Aircraft surface coating (aerospace)
No further controls identified.
Marine surface coating (shipbuilding)
Incineration
Electronic & other electrical
SCAQMD Rule 1164
Motor vehicle coating
Incineration
55.2
55.2
90
70
90
Cost Per Ton
($/ton)
2,007
8,937
881
40,202
20,000
2,027
2,027
8,937
5,976
8,937
Notes
More stringent VOC content limits and
limits for additional coatings.
Based on the use of add-on controls
such as incineration.
The amended rule is expected to result
in an incremental reduction of 17
percent, with only one facility (of two in
the area) being out of compliance with
the more stringent levels.
High cost per ton is associated with the
low VOC concentration streams. Cost
is incremental to RACT-level low VOC
content requirements
Based on costs for small plants to install
add-on control options.
Amended rule sets more stringent VOC
content limits.
Amended rule sets more stringent VOC
content limits.
Based on the use of add-on controls
such as incineration.
Based on the use of add-on controls,
the use of low-VOC coatings, and
improved operating procedures.
Based on the use of add-on controls
such as incineration.
References
This report
Pechan, 1997b
This report
This report
This report
Pechan, 1997b
Pechan, 1997b
Pechan, 1997b
Pechan, 1997b
Pechan, 1997b
-------
Table 11-5 (continued)
Percent
Pod | SCCs \ CS \ Source Category/CAA Control Measure | Reduction (%)
NOTE: All costs are in 1990 dollars.
SOURCE: Pechan, 1997b.
Cost Per Ton
($/ton)
Notes
References
-------
incinerators. Emission reductions range from 67 to 98 percent, and the cost effectiveness ranges
from $468 per ton to more than $22,100per ton. Thehighest costs for add-on controls are
associated with the model plants with specialized casegoods and the smallest casegoods plant size.
Based on the above data, the control measures modeled for beyond CAAA
requirements will be based on the use of add-on controls. Emission reductions of 75 percent will be
modeled at a cost effectiveness of $20,000 per ton. Where facilities can achieve comparable
reductions through the use of hybrid waterborne systems, full waterborne systems, or other
alternative coatings (see the discussion on emerging controls at the end of the section on industrial
coatings), reductions may be higher and costs may be lower than those estimated based on the add-
to '' J o J
on control measure. For some of the smallest facilities (i.e., smaller than those covered in the
CTG), add-on controls may not be feasible.
b. Wood Product Coating
This is a 10-year MACT source category, so CAA baseline controls are based
on engineering judgement. This category is also covered by a CTG, although compliance with these
RACT requirements should already be incorporated within the base year emission inventory for
nonattainment areas. For other areas, it is assumed that the MACT requirements will bring all areas
into compliance with any limits established in the CTG as well.
The SCAQMD recently considered an amendment to Rule 1104
(SCAQMD, 1999b), establishing more stringent limits for inks and exterior siding coatings for
wood flat stock coatings. Rule 1104 already includes VOC content limits on interior wood
o J
paneling, specifications for application, and solvent cleaning requirements. There are only two
facilities in the South Coast district which would be subject to the rule. One of these facilities
already use coatings which comply with the lower limits. The amendments are expected to reduce
emissions by 17 percent over current baseline levels at a cost-effectiveness of $1802 per ton of VOC
reduced (1999 dollars). This results in an overall reduction of 53 percent at an incremental cost of
$1429 per ton (1990 dollars) for an overall cost per ton of $762.
The SCAQMD identified thermal oxidation as an alternative control
technique, with a destruction efficiency of at least 95 percent and a collection efficiency of at least
90 percent, for an overall reduction of 85.5 percent. For the one facility examined (which has
coatings above the proposed limits), cost effectiveness is estimated at $40,202 per ton reduced
(1999 dollars) for a reductions of 85.5 percent. The high cost effectiveness of add-on controls is
due to the fact that low-VOC coatings have been already been adopted for many of the coating
processes, lowering the baseline emissions from which reductions are measured.
The Bay Area and Placer County districts already have limits as stringent as
those proposed by the SCAQMD. Therefore, it is difficult to determine whether other areas of the
country, particularly ozone nonattainment areas, already have limits as stringent as those being
considered by SCAQMD. This would require an examination of all State regulations to determine
the VOC content limits of individual coatings for this industry. Since the California districts tend to
O J
be out front in efforts to reduce smog-forming air pollution, it will be assumed that other areas can
achieve similar incremental reductions for this industry.
c. Metal Coil and Can Coating
This is also a 10-year MACT source category and is also covered by a CTG
which requires RACT in ozone nonattainment areas. It is assumed that the MACT standard will
reduce emissions by 36 percent and that this will bring all areas into compliance with CTG
requirements as well. Further reductions will then be examined incremental to RACT levels which
set VOC content limits for various coating operations.
The San Francisco Bay Area has adopted more stringent VOC content limits
for body spray coatings for both two and three piece cans; set VOC limits for end sealing
compounds for non-food products; and set limits for interior and exterior body sprays used on
drums, pails, and lids (BAAQMD, 1999). This amendment to Rule 11 is expected to further
reduce emissions by 9 percent at a cost effectiveness of $8,400 per ton. The year of dollars is not
25
-------
given in the control measure summary, so 1997 dollars is assumed since this was the year of
adoption of the regulation. In 1990 dollars, this is $8,074 per ton. This brings theoverall
reduction to $2007 per ton at 42 percent reduction from uncontrolled emissions.
d. Emerging and Future Control Options
for Industrial Surface Coating
Most recent regulatory efforts for achieving emission reductions from
o J o
surface coating operations have relied on reformulation of the coatings and other materials used in
the surface coating operations (e.g., surface preparation products). These have included
waterborne, high solids, and low-VOC coatings. For some categories, waterborne and other
' o ' o o '
reformulations may not be applicable because of performance. Other types of coatings include
powder coatings and ultraviolet and electronic beam (UV/EB) coatings.
Biofiltration has been used in Europe, primarily for odor control. It has not
gained widespread use in the United States due to the lower cost associated with add-on controls
such as incineration and adsorption. Other treatment options for VOC streams include membrane
separation, corona destruction, and photo catalytic destruction (EC/R, 1994).
Zeolite has received recent attention as a means to concentrate the VOC
stream before the flow goes to the add-on control device, such as an incinerator (CATC, 1998;
1999). Since the cost of the add-on controls are directly related to the gas flowrate, concentrating
the VOC stream will reduce the flow requirements for controlling VOC from low-VOC streams.
This is particularly important for the surface coating industry since many sectors already use low-
VOC coatings to comply with RACT and other regulations. Costs of $1,000 to $3,000 per ton
have been reported for low-VOC concentration streams with the use of zeolite. Zeolite is also
replacing carbon in adsorbers. This increases efficiency from 95 to 99 percent for low-VOC
streams and also reduces the risk of carbon bed fires. Zeolite is currently being examined by CARB
under their Innovative Clean Air Technologies program (CARB, 1999b) and has also been the focus
of research by EPA's Clean Air Technology Center (CATC, 1998; 1999).
4. Aerosol Paints
The NAAQS analysis assumed a 60 percent reduction from aerosol paints for
CARB's aerosol coating rule which was passed in 1995 (Pechan, 1997b). The rule called for a
second tier of standards to take effect in 1999. In October of 1998, CARB proposed less stringent
limits for certain products, stating that the original limits were not technologically and
commercially feasible (CARB, 1998). For other products, the limits were unchanged, and for a
few, the limits were made more stringent. The final limits are expected to result in a 42 percent
reduction in aerosol coating emissions. The cost effectiveness is expected to range from less than
$1.00 per pound to about $3.00 per pound of VOC reduced. The overall average cost effectiveness
is estimated at $1.57 per pound (1988 dollars). Converting the 1990 dollars (using thePPI for
aerosol paints), this is $2732 per ton of VOC reduced.
The cost analysis separates costs into fixed costs and recurring costs. Fixed
costs account for research and development, which may include the purchase of an additional
propellant tank. Recurring costs are the additional costs associated with the new propellant used to
manufacture the complying product. Additional research is needed to develop fixed versus
recurring cost estimates for this category.
5. Pesticides
Control strategies for pesticides include reformulation (with low VOC
content), reducing fumigant usage, use of alternative and more effective application methods,
microencapsulation, implementation of better integrated pest management programs, the use of
alternative active ingredients, and reducing the use of crop oils (EPA, 1993).
Past modeling efforts have used the California Federal Implementation Pkn
(FIP) Rule as the basis for estimating emission reductions and costs for pesticide application. This
calls for a 20 percent reduction in pesticide emissions. California intends to reach their goal by
26
-------
switching to and/or encouraging the use of low-VOC pesticides and better Integrated Pest
Management (IPM) practices.
CARB formed the Department of Pesticide Regulation (DPR) in 1991 to
regulate all aspects of pesticide sales and use. The DPR has implemented a faster registration
process so that new pesticide products can be more quickly integrated. The DPR also encourages
better IPM practices by working with local agricultural agencies and rewarding those who
demonstrate good practice or innovation.
No new regulations have been developed for pesticides as the DPR believes
that the reduction goals will be met through reformulation (which is occurring without specific air
regulations) and better IPM practices (CDPR, 1999). As such, no estimates of the cost for meeting
the reduction goals has been estimated.
o
EPA's alternative control technique for pesticides was published in March of
1993 (EPA, 1993). While the document addresses cost, most of the information is cited as variable
and unknown. Any costs that are given are registration costs or costs for new equipments. No cost
per ton estimates, which can be applied in modeling, are available from the alternative control
technique. The cost for the 20 percent reduction for the California FIP was estimated at $9,300 per
ton. This cost is likely overestimated given the information available from California's DPR;
however, no new cost effectiveness estimates are available to date. If other States regulate
pesticides due to other environmental concern, then pesticide emissions may show a decrease
without any cost incurred for air pollution regulation.
For modeling, the 20 percent reduction at $9,300 per ton will be retained
until better information is available.
6. Degreasing
CARB indicates that South Coast Rules 1122 and 1171 represent the most
effective strategy for reducing emissions from solvent cleaning and degreasing.
Rule 1122 applies to both batch and conveyorized degreasing. The latest
amendments, from 1997, set lower VOC limits for batch loaded and conveyorized cold deaners at
50 grams of VOC per liter of material. The amendments are estimated to reduce emissions from
solvent degreasing tanks (as opposed to hand-held cleaning) by 76 percent by using widely available
no- or low-VOC solvents. The expected cost is $1,391 per ton of VOC reduced (1997 dollars)
(SCAQMD, 1997).
In the study to amend Rule 1122, the SCAQMD examined a more stringent
option that would require airless batch cleaning systems or air-tight cleaning systems. This would
reduce emissions by 98 percent. The incremental cost effectiveness of this option was estimated at
$53,360 per ton (beyond the amended rule).
Rule 1171 applies to solvent cleaning operations that do not include placing
the parts in a tank. The original rule was adopted in 1991 and amended in 1995. The last
amendments were adopted in the fall of 1996. These amendments lower the VOC limit for repair
and maintenance cleaning from 900 to 50 grams per liter and lowers the vapor pressure limit for the
cleaning of printing press equipment and polyester resin application equipment. Emission
reductions were estimated at 46 percent at a cost savings of ($1,066) to ($811) per ton of VOC
reduced. In the study, the SCAQMD also evaluated a more stringent option that would reduce
emissions by 61 percent. The incremental cost effectiveness of this option was estimated at
$12,936 per ton. This option targets additional emission reductions from ink application equipment
cleaning operations, reducing the compliance limit to 350 grams per liter (SCAQMD, 1996).
Since rules 1122 and 1171 are amendments to existing rules, it is possible
that some areas may achieve reductions higher than those in the amendments. For the purposes of
this study, it will be assumed that MACT and RACT will result in baseline levels equivalent to those
from which the SCAQMD evaluated this control option.
27
-------
Additional research is needed to determine the fixed versus recurring cost
breakout for these regulations. In general, if new degreasing agents are used, little or no capital
expenditures would be required. For the more stringent options, new equipment is required.
The 1996 NET inventory classifies degreasing emissions by industry as either
open top and conveyorized or cold cleaning. The SCAQMD regulatin distinguishes by whether it is
batch or hand-cleaning, with some requirements varying by industry for hand-cleaning. Since the
cold cleaning SCCs do not cover the printing and resin industry, it will be assumed that both the
open top/conveyorized and cold cleaning categories will be coveredby Rule 1122 limits, resulting
in a reduction of 76 percent at a cost of $ 1,249 per ton (1990 dollars). This will be applied
incremental to any existing RACT control levels in the inventory.
A second measure, the requirement for airless batch cleaning systems or air-
tight cleaning systems, will also be modeled at a 98 percent reduction and incremental cost of
$47,919. The overall cost effectiveness for this measure is $9,789 per ton of VOC reduced.
B. POINT SOURCE CONTROL MEASURES
No additional control measures for point source VOC emitters have been
developed at this time. The more stringent area source measures for industrial surface coating are
candidates for point source control. In addition, the use of zeolite may make it more economical to
control other point source emitters with low concentration VOC streams. Future effort on VOC
point source emitters should also focus on better identifying sources with "99" SCCs (miscellaneous
and other not classified).
C. REFERENCES
BAAQMD, 1999: Bay Area Air Quality Management District, San Francisco Ray Area Ozone Attainment Plan for the 1-
Hour National Ozone Standard, Appendix B - Control Measure Descriptions,
adopted June 1999.
Berry, 1997: N. Berry, South Coast Air Quality Management District, personal communication with D.
Crocker, E.H. Pechan & Associates, Inc., March 4, 1997.
Billington, 1997: J. Ellington, California Air Resources Board, personal communication with D. Crocker,
E.H. Pechan & Associates, Inc., February 17, 1997.
BLS, 1996: Bureau of Labor Statistics, U.S. Department of Labor, Producer Price Indices, Washington, DC,
1996.
CARB, 1989: California Air Resources Board, Stationary Source Division, ARB-CAPCOA Suggested Control Measure
Jor Architectural Coatings, Technical Support Document, July 1989.
CARB, 1990: California Air Resources Board, Stationary Source Division, Proposed Regulation to Reduce Volatile
Organic Compound Emissions from Consumer Products - Technical Support Document, August
1990.
CARB, 1991: California Air Resources Board, Stationary Source Division, Technical Support Documentfor Proposed
Amendments to the Statewide Regulation to Reduce Volatile Organic Compound Emissions from
Consumer Products - Phase E, October 1991.
CARB, 1997a: California Air Resources Board, Initial Statement ojReasonsJor Proposed Amendments Pertaining to
Hairspray in the California Consumer Products Regulation, February 7, 1997.
CARB, 1997b: California Air Resources Board, Initial Statement oj Reasons Jor Proposed Amendments to the California
Consumer Products Regulation, June 1997.
CARB, 1998: California Air Resources Board, Initial Statement oj Reasons Jor the Proposed Amendments to the Regulations
Jor Reducing Volatile Organic Compound EmissionsJrom Aerosol Coatings, Antiperspirants and
Deodorants, and Consumer Products, October 2, 1998.
28
-------
CARB, 1999a: California Air Resources Board, "Initial Statement of Reasons for Proposed Amendments to
the California Consumer Products Regulation," September 10, 1999.
CARB, 1999b: ICAT Funded Projects, CARB website: www.arb.ca.gov/research/icat/projects.html.
CATC, 1998: Clean Air Technology Center, U.S. Environmental Protection Agency, Zeolite.- A Versatile Air
Pollutant Adsorber, EPA 456/F-98-004, July 1998.
CATC, 1999: Clean Air Technology Center, U.S. Environmental Protection Agency, Choosing anAdsorption
System/or VOC: Carbon, Zeolite, or Polymers, EPA 456/F-99-004, May 1999.
CDPR, 1999: California Department of Pesticide Regulation website: www.cdpr.ca.gov.
Ducey, 1997: E. Ducey, U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, personal communication with D. Crocker, E.H. Pechan &
Associates, Inc., February 13, 1997.
EC/R, 1994: EC/R Incorporated, Control Costsfor VOC and NO^ Measures for Non-Traditional Sourcesfor Ozone NAAQS
Review, prepared for U.S. Environmental Protection Agency, September 30,
1994.
EPA, 1995a: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Economic
Impact and Regulatory Flexibility Analysis of the Proposed Architectural Coatings Federal Rule,
Research Triangle Park, NC, March 1995.
EPA, 1995b: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Study of
Volatile Organic Compound Emissions from Consumer and Commercial Products, Report to
Congress, EPA-4S3/R-94-966-A, Research Triangle Park, NC, March
1995.
EPA, 1996a: U.S. Environmental Protection Agency, Emission Standards Division, Office of Air and
Radiation, Architectural Coatings- Backgroundfor Proposed Standards, Draft Report, EPA-
453/R-95-009a, March 1996.
EPA, 1996b: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Economic Impact and Regulatory Flexibility Analysis of the Regulation of VOCsfrom Consumer
Products, Draft Report, EPA-453/R-96-014, Research Triangle Park, NC,
October 1996.
EPA, 1996c: U.S. Environmental Protection Agency, Control of Volatile Organic Compound Emissions Jrom Wood
Furniture Manufacturing Operations, April 1996.
EPA, 1998a: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Fact
Sheet: Final Air Regulation for Architectural Coatings," August 14, 1998.
EPA, 1998b: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, "Fact
Sheet: Final Air Regulations for Consumer Products," August 14, 1998.
o ' o '
60FR15264, 1995: Federal Register, "Consumer and Commercial Products: Schedule for Regulation,"
Volume 60, Number 56, March 23, 1995.
61FR32729, 1996: Federal Register, "National VOC Emission Standards for Architectural Coatings, Proposed
Rule," Volume 61, Number 171, June 25, 1996.
61FR14531, 1996: Federal Register, "National Volatile Organic Compound Emission Standards for Consumer
Products, Proposed Rule," Volume 61, Number 64, April 2, 19%.
Jaczola, 1997: M. Jaczola, California Air Resources Board, personal communication with D. Crocker,
E.H. Pechan & Associates, Inc., March 4, 1997.
29
-------
Moore, 1997: B. Moore, U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, personal communication with D. Crocker, E.H. Pechan &
Associates, Inc., February 24, 1997.
Pechan, 1997a: E.H. Pechan & Associates, Inc., "Integrated Ozone, Particukte Matter, and Regional Haze
' ' ' o ' ' O
Cost Analysis - Methodology and Results," preparedfor U.S. Environmental
Protection Agency, Innovative Strategies and Economics Group, Office of
Air Quality Planning and Standards, June 6, 1997.
Pechan, 1997b: E.H. Pechan & Associates, Inc., Download of 1990 National Particulates Inventory data,
February 1997.
Pechan-Avanti, 1998a: The Pechan-Avanti Group, Emission Projections for the Clean Air Act Section 812 Prospective
Analysis, June 1998.
Pechan-Avanti, 1998b: The Pechan-Avanti Group, Clean Air Act Prospective Cost Analysis - Draft Report, September
1998.
SCAQMD, 1996a: South Coast Air Quality Management District, Proposed Modifications to the Appendices of the Draft
1997 Air Quality Management Plan, October 1996.
SCAQMD, 1996b: South Coast Air Quality Management District, "Addendum to Staff Report, Final
Socioeconomic Impact Assessment, Proposed Amendments to Rule 1113,"
October 1996.
SCAQMD, 1996c: South Coast Air Quality Management District, Draft Staff Report for Proposed Amendments to Rule
1171- Solvent Cleaning Operations, July 17, 1996.
SCAQMD, 1997: South Coast Air Quality Management District, Draft Staff Report for Proposed Amendments to Rule
1122 - Solvent Degreasers, June 3, 1997.
SCAQMD, 1999a: South Coast Air Quality Management District, "Addendum to Staff Report: Final
Socioeconomic Impact Assessment, Proposed Amendments to Rule 1113,"
May 1999.
SCAQMD, 1999b: South Coast Air Quality Management District, Staff Report: Proposed Amended Ruk 1104 - Wood
Flat Stock Coating Operations, August 1999.
STAPPA/ALAPCO, 1993: State and Territorial Air Pollution Program Administrators/ Association of
Local Air Pollution Control Officers, Meeting the 15-Percent Rate-of-Progress
Requirement Under the CAA: A Menu of Options, September 1993.
30
-------
CHAPTER III
STATIONARY SOURCE NOV
This chapter evaluates potential source control measures for point and area
source NOX emitters. This chapter identifies the newly developed control measures, as well as
revisions to measures developed in previous analyses. The general impetus for these revisions is the
availability of new information.
Point source measures that control NOX are described first, followed by area
source control measures for NOX. Each subsection briefly describes the source category, available
control techniques, and the control options selected for the analysis. The discussion of the control
options selected for the analysis includes an evaluation of emission reductions and total annualized
costs, and concludes with a listing of the references examined. Capital and O&M costs are
discussed for those measures for which information was identified to estimate these costs.
A. POINT SOURCE NO, CONTROL MEASURES
X
Potential stationary source NOX control options were identified in a previous
report for an analysis of EPA's NOX State Implementation Plan (SIP) Call (Pechan-Avanti, 1998).
Table III-l shows the percentage emission reduction and size-specific cost-effectiveness values (in
annual 1990 dollars) for source category/control measure combinations. Boiler design capacity-
based cost equations were created to provide costs for krge industrial, commercial, and
institutional (ICI) boilers and gas turbines. Table III-2 shows source category/control measure
combinations and variables (in annual 1990 dollars) used for each to determine costs.
Several simplifying assumptions were made for this analysis. A discount rate
of 7 percent and a capacity factor of 65 percent were assumed for all sources. The equipment life of
each control was taken directly from the source category's respective Alternative Control
Techniques (ACT) document.
The ACT documents listed available controls and costs for a variety of model
plants, depending upon the source category. Model plants with emission levels of 1 ton per ozone
season day and less were considered representative of "small" plants, while model plants with
higher emissions were considered representative of "large" plants. From these determinations,
default cost per ton values were then assigned for small and large sources, as shown in Table III-l.
Each source category was also assigned a capital cost to annual cost ratio based upon information
provided in the respective ACT document. In cases where the default cost per ton value was
applied, a default capital and operating and maintenance cost could also be determined.
For large boilers and gas turbines, capacity-based information was available
in the respective ACT document. In those cases, capadty-based equations were calculated using
scaling factors for these source category/control measure combinations. The source
o o J
31
-------
Table 111-1
Unit Costs for NOX Control Technologies for Non-Utility Stationary Sources
Source Category/Control Measure
11
12
13
14
15
16
17
18
19
20
21
22
23
ICI Boilers -Coal/Wall
1
3
4
SNCR
LNB
SCR
ICI Boilers -Coal/FBC
1
SNCR - Urea based
ICI Boilers - Coal/Stoker
1
SNCR
ICI Boilers - Coal/Cyclone
1
2
3
4
SNCR
Coal Reburn
SCR
NCR
ICI Boilers - Residual Oil
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
ICI Boilers - Distillate Oil
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
ICI Boilers - Natural Gas
1
2
3
4
5
LNB
LNB +FGR
OT + Wl
SCR
SNCR
ICI Boilers - Wood/Bark/Stoker
1
SNCR - Urea based
ICI Boilers - Wood/Bark/FBC
1
SNCR- NH3 based
ICI Boilers - MSW/Stoker
1
SNCR- Urea based
Internal Combustion Engines - Oil
1
4
IR
SCR
Internal Combustion Engines -Gas
1
4
7
10
11
12
IR
AF RATIO
AF + IR
L-E (Medium Speed)
L-E (Low Speed)
SCR
Gas Turbines - Oil
1 Water Injection
Percentage
Emission
Reduction
40
50
70
75
40
35
50
80
55
50
60
80
50
50
60
80
50
50
60
65
80
50
55
55
55
25
80
20
20
30
87
87
90
68
Average Cost per Ton
Small*
1040
1460
1260
900
1015
840
1570
820
1570
400
1120
1480
2580
1180
2490
2780
4640
820
2560
680
2230
3870
1440
1320
1690
770
2340
1020
1570
1440
380
1680
2769
1290
Large*
840
1090
1070
670
817
700
300
700
300
430
390
810
1050
2070
760
1510
1890
650
590
320
1210
1570
930
960
1250
490
920
550
380
460
0
630
533
650
32
-------
Table 111-1 (continued)
Source Category/Control Measure
24
25
26
27
28
29
30
31
2
SCR + Water Injection
Gas Turbines - Natural Gas
1
2
3
4
5
6
Water Injection
Steam Injection
LNB
SCR + LNB
SCR + Steam Injection
SCR + Water Injection
Process Heaters - Distillate Oil
1
2
3
4
5
6
7
LNB
LNB +FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
Process Heaters - Residual Oil
1
2
3
4
5
6
7
LNB +FGR
LNB
SNCR
ULNB
LNB + SNCR
SCR
LNB + SCR
Process Heaters - Natural Gas
1
2
3
4
5
6
7
LNB
LNB +FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
Adipic Acid Manufacturing
1
2
Thermal Reduction
Extended Absorptbn
Nitric Acid Manufacturing
1
2
3
Extended Absorption
SCR
SNCR
Glass Manufacturing - Container
1
2
3
4
5
6
Electric Boost
Gullet Preheat
LNB
SNCR
SCR
OXY-Firing
Glass Manufacturing - Flat
1
Electric Boost
Percentage
Emission
Reduction
90
76
80
84
94
95
95
45
48
60
74
75
78
92
34
37
60
73
75
75
91
50
55
60
75
75
80
88
81
86
95
97
98
10
25
40
40
75
85
10
Average Cost per Ton
Small*
2300
1510
1040
490
2570
2010
2730
3470
4250
3180
2140
9230
3620
9120
3490
2520
1930
1290
2300
5350
5420
2200
3190
2850
1500
12040
3520
11560
420
90
480
590
550
7150
940
1690
1770
2200
4590
2320
Large*
1010
730
500
100
600
840
1130
970
1680
1720
610
6030
1880
5250
1380
710
1100
360
1240
3590
3160
1800
2470
1950
1200
8160
2590
8020
420
90
480
590
550
7150
940
1690
1770
2200
4590
2320
33
-------
Table 111-1 (continued)
Source Category/Control Measure
32
33
34
35
36
37
38
39
41
42
2
3
4
5
LNB
SNCR
SCR
OXY-Firing
Glass Manufacturing - Pressed
1
2
3
4
5
6
Electric Boost
Gullet Preheat
LNB
SNCR
SCR
OXY-Firing
Cement Manufacturing - Dry
1
2
3
4
5
Mid-Kiln Firing
LNB
SNCR - Urea Based
SNCR- NH3 Based
SCR
Cement Manufacturing - Wet
1
2
3
Mid-Kiln Firing
LNB
SCR
Iron & Steel Mills - Reheating
1
2
3
LEA
LNB
LNB +FGR
Iron & Steel Mills -Annealing
1
2
3
4
5
6
LNB
LNB +FGR
SNCR
LNB + SNCR
SCR
LNB + SCR
Iron & Steel Mills - Galvanizing
1
2
LNB
LNB +FGR
Municipal Waste Combustors
1
SNCR
Medical Waste Incinerators
1
SNCR
ICI Boilers - Process Gas
1
2
3
4
LNB
LNB +FGR
OT + Wl
SCR
ICI Boilers - Coke
1
3
4
SNCR
LNB
SCR
Percentage
Emission
Reduction
40
40
75
85
10
25
40
40
75
85
30
25
50
50
80
30
25
80
13
66
77
50
60
60
80
85
90
50
60
45
45
50
60
65
80
40
50
70
Average Cost per Ton
Small*
700
740
710
1900
8760
810
1500
1640
2530
3900
460
560
770
850
3370
420
530
2880
1320
300
380
570
750
1640
1720
3830
4080
490
580
1130
4510
820
2560
680
2230
1040
1460
1260
Large*
700
740
710
1900
8760
810
1500
1640
2530
3900
460
560
770
850
3370
420
530
2880
1320
300
380
570
750
1640
1720
3830
4080
490
580
1130
4510
650
590
320
1210
840
1090
1070
34
-------
Table 111-1 (continued)
Source Category/Control Measure
43
44
45
46
47
48
49
50
54
55
ICI Boilers - LPG
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
ICI Boilers - Bagasse
1
SNCR- Urea based
ICI Boilers - Liquid Waste
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
1C Engines - Gas, Diesel, LPG
1
4
IR
SCR
Process Heaters - Process Gas
1
2
3
4
5
6
7
LNB
LNB +FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
Process Heaters - LPG
1
2
3
4
5
6
7
LNB
LNB +FGR
SNCR
ULNB
SCR
LNB + SNCR
LNB + SCR
Process Heaters - Other Fuel
1
2
3
4
5
6
7
LNB +FGR
LNB
SNCR
ULNB
LNB + SNCR
SCR
LNB + SCR
Gas Turbines - Jet Fuel
1
2
Water Injection
SCR + Water Injection
Space Heaters - Distillate Oil
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
Space Heaters - Natural Gas
Percentage
Emission
Reduction
50
60
80
50
55
50
60
80
50
25
80
50
55
60
75
75
80
88
45
48
60
74
75
78
92
34
37
60
73
75
75
91
68
90
50
60
80
50
Average Cost per Ton
Small*
1180
2490
2780
4640
1440
400
1120
1480
2580
770
2340
2200
3190
2850
1500
12040
3520
11560
3470
4250
3180
2140
9230
3620
9120
3490
2520
1930
1290
2300
5350
5420
1290
2300
1180
2490
2780
4640
Large*
2070
760
1510
1890
930
430
390
810
1050
490
920
1800
2470
1950
1200
8160
2590
8020
970
1680
1720
610
6030
1880
5250
1380
710
1100
360
1240
3590
3160
650
1010
2070
760
1510
1890
35
-------
Table 111-1 (continued)
Source Category/Control Measure
56
57
58
59
60
61
62
63
64
65
66
67
68
69
1
2
3
4
5
LNB
LNB +FGR
OT + Wl
SCR
SNCR
Ammonia - NG-Fired Reformers
1
2
3
4
5
LNB
LNB +FGR
OT + Wl
SCR
SNCR
Ammonia - Oil-Fired Reformers
1
2
3
4
LNB
LNB +FGR
SCR
SNCR
Lime Kilns
1
2
3
4
5
Mid-Kiln Firing
LNB
SNCR - Urea Based
SNCR- NH3 Based
SCR
Comm./lnst. Incinerators
1
SNCR
Indust. Incinerators
1
SNCR
Sulfate Pulping - Recovery Furnaces
1
2
3
4
5
LNB
LNB +FGR
OT + Wl
SCR
SNCR
Ammonia Prod; Feedstock Desulfurization
2
LNB +FGR
Plastics Prod- Specific; (ABS) Resin
2
LNB +FGR
Starch Mfg; Combined Operations
2
LNB +FGR
By-Product Coke Mfg; Oven Underfiring
3
SNCR
Pri Cop Smel; Reverb Smelt Furn
2
LNB +FGR
Iron Prod; Blast Furn; Blast Htg Stoves
3
LNB +FGR
Steel Prod; Soaking Pits
2
LNB +FGR
Fuel Fired Equip; Process Htrs; Pro Gas
2
LNB +FGR
Percentage
Emission
Reduction
50
60
65
80
50
50
60
65
80
50
50
60
80
50
30
30
50
50
80
45
45
50
60
65
80
50
60
55
55
60
60
77
60
55
Average Cost per Ton
Small*
820
2560
680
2230
3870
820
2560
680
2230
3870
400
1120
1480
2580
460
560
770
850
3370
1130
1130
820
2560
680
2230
3870
2560
3190
3190
1640
750
380
750
3190
Large*
650
590
320
1210
1570
650
590
320
1210
1570
430
390
810
1050
460
560
770
850
3370
1130
1130
650
590
320
1210
1570
590
2470
2470
1640
750
380
750
2470
36
-------
Table 111-1 (continued)
Source Category/Control Measure
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
Sec Alum Prod; Smelting Furn/Reverb
1
LNB
Steel Foundries; Heat Treating Furn
1
LNB
Fuel Fired Equip; Furnaces; Natural Gas
1
LNB
Asphaltic Cone; Rotary Dryer; Conv Plant
1
LNB
Ceramic Clay Mfg; Drying
1
LNB
Coal Cleaning-Thrml Dryer; Fluidized Bed
3
LNB
Fbrglass Mfg; Txtle-Type Fbr; Recup Furn
3
LNB
Sand/Gravel; Dryer
2
LNB +FGR
Fluid Cat Cracking Units; Cracking Unit
2
LNB +FGR
Conv Coating of Prod; Acid Cleaning Bath
1
LNB
Natural Gas Prod; Compressors
12
SCR
In-Process; Bituminous Coal; Cement Kiln
3
SNCR - urea based
In-Process; Bituminous Coal; Lime Kiln
3
SNCR - urea based
In-Process Fuel Use; Bituminous Coal; Gen
1
SNCR
In-Process Fuel Use; Residual Oil; Gen
2
LNB
In-Process Fuel Use; Natural Gas; Gen
1
LNB
In-Proc; Process Gas; Coke Oven/Blast Furn
2
LNB +FGR
In-Process; Process Gas; Coke Oven Gas
1
LNB
Surf Coat Oper; Coating Oven Htr; Nat Gas
1
LNB
Solid Waste Disp; Gov; Other Incin; Sludge
1
SNCR
Percentage
Emission
Reduction
50
50
50
50
50
50
40
55
55
50
20
50
50
40
37
50
55
50
50
45
Average Cost per Ton
Small*
570
570
570
2200
2200
1460
1690
3190
3190
2200
2769
770
770
1260
2520
2200
3190
2200
2200
1130
Large*
570
570
570
1800
1800
1090
1690
2470
2470
1800
533
770
770
940
710
1800
2470
1800
1800
1130
37
-------
Table 111-1 (continued)
Source Category/Control Measure
Percentage
Emission
Reduction
Average Cost per Ton
Small*
Large*
Key:
IR
LPG
LEA
LNB
LNB+FGR
LNB+SCR
LNB+SNCR
OT+WI
SCR
SNCR
ULNB
Ignitbn
Timing
Retard
Liquefied
Petroleum
Gas
Low Excess
Air
Low-NOx
Burner
Low-NOx Burner + Flue Gas Recirculation
Low-NOx Burner + Selective Catalytic Reduction
Low-NOx Burner + Selective Noncatalytic
Reduction
Oxygen
Trim +
Water
Injection
Selective
Cata lytic
Reduction
Selective
Noncatalytic
Reduction
Ultra-Low
NOV Burner
NOTES:
'Small source cost per ton values are used to estimate control costs for all sources with 1996 NOX
emissions below 1 tons per day (tpd). If the ozone season daily 1996 baseline NOX value is 1 ton
or more, the cost per ton value for large sources is used.
38
-------
(A
(1)
O
V)
0)
o>
re
cv <2
= o
I?
2 iff
+j
(A
O
O
U
U
0)
Q.
CO
a>
39.N
(O
01
"3
(/)
TO"
CQ
1—
0
2
E
0
i
o
01
c
1
m
TT
•£
o
o
Q
z
_
0
«
o
0
"c
01
o
Q.
X
LLJ
- S
3 "5.
5 "5
^ 5
^J
c
c
o
a
X
LLJ
ral
"n •—
' f~ ^
s
"c
c
o
a
LU
•5.2
=1
£ ^
^ g
+J
c
a>
c
o
a
LLJ
^_
— 01
tl
*• ^
s
c
o
U)
li
UJ.=
0) °
co:
01
01
Q.
01
3
8
01
1
c
o
0
"Si
o
D)
*J
0
01
o
0
V)
CO
"co
0
o
to
'o
m
o
T_
T~
LO
a>
•vt-
o
CN
^
LO
CO
CN
0
00
CO
O)
o
CO
CO
[^
0
a>
0
^J-
co
CO
CN
0
CO
00
0
o
a
o
z
w
,_
CO
o
^
CO
00
T—
CO
0
^
00
CO
00
CO
LO
CO
0
T_
CO
00
•»-
CO
0
1^
00
CO
00
CO
LO
o
LO
m
z
CO
CO
a>
CO
o
LO
-tJ
o
00
LO
CO
0
CN
CN
o
o
a>
LO
00
00
0
CO
LO
LO
LO
LO
LO
CO
0
O3
0
o
CN
00
o
OL
o
^J.
o
m
LL
"co
o
o
to
'o
m
o
CN
CO
0
CN
0)
LO
o
CO
CO
0
00
CN
O)
LO
T
CO
0
LO
0
0)
CO
0
00
CN
O)
LO
T
LO
CD
to
CO
.Q
CO
CD
=>
OL
o
z
,_
/Stoker
CO
o
o
to
'o
m
o
CO
LO
O)
•vT
0
CN
^
[^
CO
CN
0
00
CO
0)
o
CO
CO
CO
0
a>
0
^J-
co
CO
CN
0
CO
00
0
o
OL
o
z
,_
b
"co
'to
CD
to
'o
m
o
LO
00
CO
CN
0
^
^
CN
LO
CO
CN
0
CO
CN
0
o
00
a>
CN
CN
0
^
CN
0
CN
CO
CN
0
00
00
CN
CO
o
LO
OL
o
z
^J.
o
LO
CO
0
LO
-tJ
00
^J-
LO
CO
0
CO
^
0)
00
0
CN
CO
0
T_
00
O)
CN
LO
CO
0
CO
CO
0
CN
CO
CO
o
00
OL
o
CO
b
CD
JO
"to
b
to
'o
m
o
CO
00
CO
CN
0
^
CN
LO
CO
CN
0
CO
CN
0
o
00
"*
CO
CN
0
^
CN
0
CN
CO
CN
0
00
00
CN
CO
o
LO
OL
o
z
^J.
o
LO
CO
0
LO
^
00
^J-
LO
CO
0
CO
t-J
O)
00
0
CN
CO
0
T_
00
a>
CN
LO
CO
0
CO
CO
0
CN
CO
CO
o
00
OL
o
CO
CD
"co
"co
z
to
"o
m
o
^
T~
00
CO
CN
0
^
CN
LO
CO
CN
0
CO
CN
0
o
00
"*
CO
CN
0
^
CN
Q
CO
CN
0
00
00
CN
CO
0
LO
OL
o
z
LO
o
LO
CO
0
LO
-tJ
00
^J-
LO
CO
0
CO
^
O)
00
0
CN
CO
0
T_
00
O)
LO
CO
0
CO
CO
0
CN
CO
CO
0
00
a
o
^J.
CD
-^
d/Bark/Stc
o
1
2
'o
m
o
00
CN
CO
CO
0
^
[^
T—
T —
CO
CO
0
T
o
CN
00
LO
CO
CN
CO
CO
0
T_
£
fs,.
•«-
[•X,.
0
CO
CO
0
T—
o
CN
00
LO
CO
LO
LO
CD
to
CO
.Q
CO
CD
OL
O
Z
,_
o
d/Bark/FB
0
o
2
'o
m
O
O)
CO
0
^
LO
00
^J-
co
0
CO
LO
LO
00
O)
CO
0
^
LO
00
CO
0
CO
LO
LO
00
O)
LO
LO
-a
CD
to
CO
.Q
i"
z
o
z
,_
b
to
CD
_Q
1—
to
CO
CD
CO
CN
O)
LO
0
O)
|s^
00
O3
CO
00
CO
LO
0
LO
CO
LO
^J-
LO
O)
LO
0
O)
00
CO
O)
CO
00
CO
LO
0
LO
CO
LO
^J-
LO
00
CO
c
o
'o
.CD_
fc
1
,_
s
LO
0
LO
CN
O)
00
CN
O)
LO
o
CO
0
0)
00
CO
LO
0
CO
LO
CO
0
CN
o
0
CO
CO
LO
CN
a>
LO
0
CN
0
00
0)
CO
CN
0
a>
c
o
"o
.CD_
fc
CO
o
CN
atural Gas
z
to
CD
_Q
1—
to
CO
CD
,.,.
CN
00
CO
T—
^
LO
CN
o
•^
CN
^
00
CN
00
CO
•^
^
LO
T —
CN
o
•"-
CN
00
CN
CO
^
c
o
••s
.CD_
-ffi
1
T_
CN
•^
T—
CO
^
CO
O)
O)
O)
o
T
CO
O)
CO
O)
CN
T —
•^
CO
CO
O)
O)
O)
o
T—
CO
O)
CO
O)
0
00
c
.0
"o
,CD_
E
CO
-CD
w
CN
LO
o
LO
0
CO
CO
CN
00
LO
o
LO
o
T
^
00
CN
^
LO
o
LO
o
CO
CO
CN
00
LO
o
LO
o
T—
00
CN
^
•*
00
m
z
CO
CO
CO
00
CO
0
0
CN
O)
CO
T—
LO
O)
CN
o
^
CO
0
CN
CO
CO
CO
LO
CO
o
^
CO
O)
O)
•»-
00
•vT
CO
o
00
CO
CD
00
^-
0>
m
z
OL
o
w
^J.
LO
CO
CO
00
0
O)
^
LO
CN
LO
00
o
00
00
CN
LO
"
CO
0
O)
CO
o
^
CO
CO
O)
LO
CN
LO
0)
CO
CO
o
CN
CO
0
CO
0>
LO
0)
c
.0
o
QJ
E
CO
CD
CO
o
CO
LO
-------
Table 111-2 (continued)
Source Category/Control Measure
| 6 | SCR + Water Injection
Percentage Emission
Reduction
95
Cost from No Control Baseline
Capital
Multiplier
121119
Exponent
0.5891
Annual
Multiplier
36298.9
Exponent
0.6308
Cost from RACT Baseline
Capital
Multiplier
18026.5
Exponent
0.8237
Annual
Multiplier
7607
Exponent
0.7828
Key:
LNB Low-NOx Burner
LNB+SCR Low-NOx Burner + Selective Catalytic Reduction
LNB+SNCR Low-NOx Burner + Selective Noncatalytic Reduction
SCR Selective Catalytic Reduction
SNCR Selective Noncatalytic Reduction
NOTE:
All costs in the form y=mxAb.
-------
category ACT documents also contained enough information to determine the separate costs
necessary for applying more stringent controls where RACT level controls already existed in the
base year. In the case of incremental controls, it was assumed that the new, more stringent control
levels modeled would be applied from uncontrolled emission levels.
Further detailed information on the calculation of these costs can be found in
Pechan-Avanti's Ozone Transport Rulemaking Non-Electricity Generating Unit Cost Analysis (Pechan-Avanti, 1998).
B. AREA SOURCE NOX CONTROL MEASURES
This section discusses area source control measures for NOX. Each
subsection describes the penetration rate, the fraction of emissions covered by each measure,
assumed in the analysis.
Information on controls previously developed and used in this analysis can be
found in Pechan-Avanti's Additional Control Measure Evaluation for the Integrated Implementation of the Ozone and
Paniculate-Matter National Ambient Air Quality Standards, and Regional Haze Program (Pechan, 1997a).
1. Agricultural Burning
Agricultural burning is defined as the intentional burning of agricultural
o o o o
fields for the purpose of waste reduction (Pechan, 1995).
a. Description of Available Control
Options
Emissions from agricultural burning may be reduced by limiting the types of
material that can be burned, or based on ambient conditions, limiting the days on which materials
' ' O J
can be burned. Because information was not available on the cost or emission reductions associated
with limiting the types of materials that can be burned, this type of control is not modeled.
Since burning can simply be shifted to other acceptable periods, emission
control costs are assumed to be zero for regulations that schedule the burning days where ozone
exceedances are not predicted. Costs may be incurred if personnel scheduled to participate in the
agricultural burning cannot be used elsewhere or if fire personnel or other profes sionals have been
scheduled to participate. Assuming full compliance with the regulation, ozone season daily emission
reductions from such a regulation would be 100 percent. However, annual emission reductions
would not be expected, because there would likely be a shift in the timing of the emissions, not a
reduction in the total amount of annual NOX emitted (as well as all other pollutants from
agricultural burning). A compliance rate of 80 percent is used in estimating daily reductions.
Annual emissions are not reduced (Pechan, 1997b).
b. Control Options Selected for Analysis
A seasonal ban on agricultural burning is selected for analysis. Control
o o J
efficiencies for ozone season daily emission reductions (there are no annual emission reductions) are
presented in Table III-3.
2. Commercial and Residential Water Heaters
Roughly one-quarter of all U.S. energy consumption is related to space
heating, water heating, and air conditioning (SCAQMD, 1997). The following SCCs are subject to
controls for water heating:
• 2103006000: Commercial and Institutional Natural Gas Water
Heaters
• 2104006000: Residential Natural Gas Water Heaters
a. Description of Available Control
Options
41
-------
The primary means of reducing emissions from natural gas-fired residential
and commercial water heaters would be low-NOx burners. Low-NOx burners are designed to
control the combustion process with proper air/fuel mixing and increased heat dissipation to
minimize thermal NOX formation.
The heat pump water heater (HPWH) is another method for reducing NOX
emissions from this source. The HPWH uses the same vapor compression refrigeration technology
as space conditioning heat pumps. Basically, the HPWH takes heat from the surrounding air and
transfers it to the water in the storage tank. This reduces the amount needed to heat the water.
The most common type of HPWH is the air to water variety. In addition, HPWH can provide
supplemental cooling effect. Heat pump water heaters are best suited for applications with high,
consistent, year-round water heating loads and a need for ventilation and/or space cooling and
dehumidifying (SCAQMD, 1997). Examples include commercial laundries, restaurant kitchens,
homes with large cooling loads, hot attics, etc.
o o ' '
Solar water heating could be another method in reducing NOX emissions
from this source. Non-concentrating solar collectors, such as flat-plate solar panels, are capable of
providing sufficient domestic water heating capabilities. Conventional natural gas-fired water
heaters (using low NOX burners emitting 10 nanograms per joule of heat output, or less) would still
continue to be used to supplement the solar component.
Other control technologies include the use of electric thermal storage
systems for commercial or multiple housing units. For example, an insulated tank is filled with
water to a predetermined level. The water to be used throughout the building is heated by passing
through a heat exchanger located below the level of the insulated tank. This system can either be
installed inside or outside a building, above or below ground level.
b. Control Options Selected for Analysis
Low-NOx burners control costs have been updated for this analysis and,
along with control efficiencies, are presented in Table III-3. Controls previously modeled, such as
programs designed for buying new water heaters, can be found in previous Pechan-Avanti
documents (Pechan, 1997a).
42
-------
Table 111-3
Revised Low-NOx Burner Control Measure for
Commercial and Residential Water Heaters
Low-NO, Water Heater
Year Rule Penetration (%) Control Efficiency (%) Overall Control (%) Cost Per Ton ($)
1996 0 75 0 595
2003 0 75 0 595
2006 40 75 30 595
2010 73 75 55 595
43
-------
C. REFERENCES
Pechan, 1995: The Pechan-Avanti Group, Regional Paniculate Strategies - Draft Report, prepared for U.S.
Environmental Protection Agency, Office of Policy Planning and Evaluation,
Washington, DC, EPA Contract No. 68-D3-0035, Work Assignment No.
1-54, September 29, 1995.
Pechan, 1997a: E.H. Pechan & Associates, Inc., Additional Control Measure Evaluation Jor the Integrated Implementation of
the Ozone and Paniculate Matter National Ambient Air Quality Standards, and Regional Haze
Proaram, prepared for U.S. Environmental Protection Agency, Office of Air
Cl ' 1 1 O J '
Quality Planning and Standards, Research Triangle Park, NC, EPA Contract
Nos. 68-D3-0035, Work Assignment No. III-100, July 1997.
Pechan, 1997b: E.H. Pechan & Associates, Inc., The Emission Reduction and Cost Analysis Model/or NOX (ERCAM-NOJ -
Revised Documentation, prepared for U.S. Environmental Protection Agency,
Ozone Policy and Strategies Group, Research Triangle Park, NC, EPA
Contract Nos. 68-D3-0035, Work Assignment No. 111-93, September 1997.
Pechan, 1998: The Pechan-Avanti Group, Ozone Transport Rulemaking Non-Electricity Generating Unit Cost Analysis,
prepared for U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Innovative Strategies and Economics Group,
Research Triangle Park, NC, EPA Contract Nos. 68-D4-0102 (WA 4-12)
and 68-D9-8052 (WA 0-3), September 17, 1998.
SCAQMD, 1997: South Coast Air Quality Management District, 1997 Air Quality Management Plan, Appendix IV-1:
Stationary Source Control Measures, November 1996.
44
-------
CHAPTER IV
STATIONARY SOURCE SO,
Control efficiencies and costs of systems used to control SO2 emissions from
point sources were based on the use of various flue gas desulfurization (FGD) scrubbing systems,
sulfur recovery plants, and sulfuric acid plants. The use of FGD scrubbers for controlling SO2
emissions has been applied to utility and industrial boilers for over 25 years and is now considered a
mature technology which can be designed to control SO2 emissions from most industrial sources.
Switching to burning a lower-sulfur fuel is another method used to lower SO2 emissions, but the
economic feasibility of obtaining of a lower-suliur fuel in sufficient quantities limits its application.
FGD scrubbers can be either wet or dry systems. In wet systems, a liquid
sorbent is sprayed into the flue gas in an absorber vessel. Limestone and lime-based reagents are
most frequently used in scrubbers in the United States. Dry and semi-dry FGD systems include
spray dryers, and dry injection into a duct or a combustion zone.
Sulfur recovery and sulfuric acid plants are generally used to control the SO2
in waste gas from processes involved in the nonferrous metal, elemental sulfur, and sulfuric acid
production industries. Most of the ores from which nonferrous metals such as copper, zinc, and
lead are extracted also contain sulfur which generally forms SO2 upon removal from the ore, usually
by heat. Sulfur recovery or sulfuric acid plants used to control emissions from elemental sulfur or
sulfuric acid plants amount to improving the existing process to recover more sakble material.
Control efficiency, capital costs, annual costs, and cost-effectiveness ranges
for control technologies were based on information taken from various sources and estimated using
professional judgement. Sources include journal articles, EPA documents, and reference materials.
Cost data was not available for all control technologies. The dollar years of the cost data, control
o J '
equipment lifetimes, and discount rates used to estimate annual costs are provided where available.
Section A details the control measure parameters used in ControlNET for
the additional SO2 measures. Section B of this chapter gives a description of the SO2 source
categories along with potential controls for each category. This includes control options for which
modeling parameters (costs) were not available.
A. COST ESTIMATES FOR RETROFIT CONTROL TECHNOLOGIES
Retrofit control options by source category are listed in Table IV-1. This
table lists the source category, the applicable SO2 technology, and the expected control efficiency
(Pechan-Avanti, 1999).
FGD is a control option for most of the source categories that were
L O
evaluated. Costs for FGD scrubbers were developed using a computer spreadsheet model provided
by EPA.
45
-------
V)
c
o
O
+->
0)
46
Expected SO2 Control
Efficiency (%)
Measure
S
•E
0
o
Strategy
Control
0)
ra
z
Q.
0
CD
Q.
1
CD
o"
(/)
S
a
CD
c
u
'e
CD
C
O
u
s?
o>
o>
o>
Absorber
o
CD
c
0
O
to
c
CD
Q.
T3
z
o
3
co
s
c
o
p-
o
to
£2
CD
"ro
D
C
O
'to
CD
>
C
0
O
s?
o>
o>
Absorber (
o
CD
c
0
O
to
c
CD
Q_
T3
'o
<
o
2
co
CN
s
Q
CD
LL
+
c
o
p-
o
to
_Q
CD
"CD
D
C
O
'to
CD
>
C
0
O
s?
o>
o>
Absorber (
o
CD
c
0
O
to
c
CD
Q.
T3
'o
<
o
2
co
CN
LO
0>
c
o
p-
o
to
_Q
CD
"CD
D
c
o
'to
tu
>
c
0
o
s?
00
o>
Absorber (
o
CD
C
0
O
to
c
CD
Q_
T3
'o
<
O
2
CO
m
LO
o>
o>
Q
CD
LL
+
C
o
p-
o
to
_Q
CD
"CD
D
c
o
'to
tu
>
c
0
o
s?
00
o>
Absorber
o
CD
c
0
o
to
c
CD
Q_
T3
z
o
IS
co
CO
1^
CO
o>
c
o
p-
o
to
_Q
CD
"CD
D
c
o
'to
CD
>
c
0
o
s?
1^
o>
Absorber
o
CD
c
0
o
to
c
CD
Q_
T3
z
o
IS
co
M-
1^
CO
o>
o>
Q
CD
LL
+
c
o
p
o
to
_Q
CD
"CD
D
c
o
'to
CD
>
c
0
o
s?
1^
o>
Absorber (
o
CD
c
0
o
to
c
CD
Q_
T3
'o
<
o
2
co
•*
CO
00
0>
c
o
p-
o
to
_Q
CD
"CD
D
c
o
'to
CD
>
c
0
o
s?
m
o>
Absorber (
o
CD
c
0
o
to
c
CD
Q_
T3
'o
<
o
2
w
LO
CO
00
O)
O)
Q
CD
LJ-
+
c
o
p-
o
to
^3
CD
"CD
D
c
o
'to
CD
>
c
0
o
s?
m
O)
Absorber (
o
CD
c
0
o
to
c
CD
Q_
-a
'o
<
o
2
w
LO
M-
00
O)
D)
c
!Q
D
o
CO
CD
c
"CD
>
o
CD
^o
LO
en
CN
O)
"o
c
o
o
0
g
CD
D)
.3
CN
to
D
CD
o
D
w
1
CD
HI
.2
c
CD
Recovery P
D
w
CO
s
O)
O)
Q
CD
LL
+
D)
c
!Q
D
o
w
CD
c
"CD
>
o
CD
^o
LO
en
CN
O)
"o
c
o
o
0
g
CD
D)
.3
CN
to
D
CD
o
D
w
1
CD
HI
.2
c
CD
Recovery P
D
w
CO
00
h^
0)
D)
c
!Q
D
o
w
CD
c
I
o
CD
s?
CO
en
LO
O)
"o
c
o
o
0
g
CD
D)
.3
CO
to
D
CD
o
^
D
w
1
CD
HI
I
CD
Recovery P
2
w
1^
00
1^
O)
O)
Q
CD
LL
+
D)
c
!Q
D
o
CO
CD
c
I
o
CD
s?
CO
en
LO
O)
"o
c
o
o
0
g
CD
D)
.3
CO
to
D
CD
o
^
D
w
1
CD
HI
I
CD
Recovery P
2
w
1^
h^
0)
D)
c
!Q
D
o
w
CD
c
I
o
CD
s?
1^
en
CO
O)
"o
c
o
o
0
g
CD
D)
.3
CO
to
D
CD
o
^
D
w
1
CD
HI
I
CD
Recovery P
2
w
00
i^
O)
O)
Q
CD
LL
+
D)
c
!Q
D
o
w
CD
c
"CD
>
o
CD
^o
h-
cn
CO
O)
"o
c
o
o
0
g
CD
D)
s
m
to
D
CD
o
"5
w
1
CD
HI
.2
c
CD
Recovery P
D
w
00
s
Q
CD
"CD
>
0
CD
s?
O)
Si-
Process
urRemova
D
to
c
CD
• Recovery P
n
w
O)
s
Q
CD
TD
E^
'to
to
CD
u
b
o
n
-a
Q
CL
~3
w
1
_cu
to
c
CD
• Recovery P
n
w
o
s
a
CD
cu
o
CD
1 Manuf
anic Chemice
S5
0
c
CN
00
_cu
"CD
8
CD
/acuum C
J2
CD
Q_
c
CD
>
CD
_^
0
o^
D>
c
b
45
3
c
CD
•=:
CD
^
8
"o
D
-o
0
Q-
>,
m
CD
CN
s
a
CD
to
3SS80OJC
b
0,
D)
c
b
45
3
c
CD
•=:
CD
^
8
"o
D
T3
0
0.
>^
m
^3
CN
s
a
CD
^
"to
D
-o
c
c
s Productb
CD
rn
-o
c
CD
0
ss Heaters (
o
CL
CO
s
Q
CD
£
'o
m
"CD
"to
3
-o
c
"CD
o
uminous
inous/Subbit
b
m
O)
s
Q
CD
_cu
0
m
"CD
•c
3
T3
c
6
"CD
D
2
CD
r±
o
s
Q
CD
to
_fc
o
m
CD
c
0
5
c
"CD
'tj
0
o_
"CD
o
o
uminou
inous/Subbil
b
m
s
Q
CD
§~
o
uminous
lous/Subbi
E
m
CD
Z>
"cu
3
LL
to
to
S
c
(N
S
Q
CD
't/T
_fc
'o
m
e (Industrial
'c
U)
_l
m
S
a
CD
to
_fc
titutional Be
to
c
mercial/
h
o
O_
O
"CD
D
'to
CD
C£
•vT
S
Q
CD
2
n
,ombus1
u
_cu
"to
1
CD
CL
'£
D
2
LO
-------
Table IV-1 (continued)
SO, Group
26
27
28
29
Group Name
Steam Generating Unit-Coal/Oil
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Lead Smelters - Sintering
Primary Zinc Smelters - Sintering
Control Strategy
1
1
1
1
Control Measure
FGD
Dual absorption
Dual absorption
Dual absorption
Expected SO2 Control
Efficiency (%)
90
99
99
99
-------
The model is based on a wet FGD system and was developed from data presented in a report that
analyzed the impacts of SO2 controls on the electric power generation sector (EPA, 1996). To
develop a method for estimating costs for emission sources in the 1990 emission inventory, the
spreadsheet model was used to develop capital and operating cost components using the stack gas
flow rate and stack gas temperature for emission points in the 1990 emission inventory as the
independent variable (Pechan, 1997). Table IV-2 presents an illustration of the FGD spreadsheet
model. The following describes the spreadsheet model and then describes the equations used to
estimate costs for emission sources. Table IV-3 lists the SCCs for the various SO2 groups.
All costs are expressed in 1990 dolkrs unless otherwise indicated. Capital
costs were updated from these base years by means of the Chemical Engineering Cost Index.
Operating costs were updated using the Producer Price Index.
1. FGD Scrubber Spreadsheet Cost Model
The key input parameters used as variables in the model include stack gas
flow rate and temperature entering the scrubber, and annual operating time. The inputs for capital,
fixed O&M, and variable O&M costs are used as constants in the model. These constants are based
on data for FGD scrubber cost assumptions for utility boilers with a 3 percent coal sulfur content
(EPA, 1996). The assumptions apply to capacities at or above 500 megawatts (MW)
[approximately 1,000,000 actual cubic feet per minute (acfin)]. For smaller sizes, the costs are
scaled down using the standard 0.6 power law. Thus, at lower capacities, capital costs [in dollars
per kilowatt or $/acfm] are proportionately higher. In the spreadsheet model, costs are scaled
down using the 0.6 power law if the gas flow rate is less than 1,028,000 acfrn. A gas flow rate
factor of 0.486 is used to convert costs from $/kilowatt to $/acfm. This factor was derived from
data in the Integrated Air Pollution Control System (IAPCS) model (version 5). For existing
emission sources, a retrofit factor of 1.1 is applied to the capital costs. A capital recovery factor of
0.1098 is used to estimate capital charges based on a 7-percent interest rate and a 15-year
equipment life. The FGD scrubber cost assumptions are in 1995 dollars. Capital and annual costs
are de-escalated to 1990 dollars using the ratio of Chemical Engineering Annual Plant Cost Indexes
for 1990 and 1995.
2. Sulfuric Acid Plants
Technology for SO, emissions control from sulfuric acid plants is well
cv z 1
established. The dual absorption process is operating successfully at many U.S. facilities. In
addition, several desulfurization processes apply to tail gases from a sulfuric acid plant. Using a dual
absorption process, a plant can convert 99.7 to 99.8 percent of the SO2 produced to SO3. The dual
absorption process has proved to be the SO2 control system of choice for the sulfuric acid industry
since the promulgation of the New Source Performance Standards (NSPS).
Cost equations for dual absorption (EPA, 1985):
Capital cost = $990,000 + $9.836 * Flow rate (in cubic feet per minute [ft?/minute])
Operating cost = $75,800 + $12.82 * Flow rate (in ft3/minute)
48
-------
Table IV-2
Illustration of FGD Scrubber Cost Spreadsheet Model
Model Inputs
— Gas flow rate at FGD scrubber hlet [standard cubic feet per minute (scfm)] 1
— Gas temperature at FGD scrubber inlet (°F) 1
- SO2 concentratbn at FGD scrubber inlet (vol. %)
- Capital cost (1995 $/kilowatt) *
- Fixed O&M cost (1995 $/kilowatt-year) 2
- Variable O&M cost (1995 $/kilowatt hour) 2
- Flow Rate factor (kilowatt/acfm) 3
- FGD scrubber operating time (hour/year) 1
- FGD scrubber control efficiency (%) 2
- FGD scrubber useful life (year)
— Interest (discount) rate (fraction)
- Retrofit factor 2
- De-escalation factor "
500,000
400
1.0
192
6.9
0.0015
0.486
8,736
90
15
0.07
1.10
0.9383
Variable
Variable
Variable
Constant
Constant
Constant
Constant
Variable
Constant
Constant
Constant
Constant
Constant
Model Outputs
- Gas flow rate at FGD inlet (acfm)
- SO2 inlet rate (tons/year)
- Capital cost (1995 $/acfm) 5
- Fixed O&M cost (1995 $/acfm-year)
- Variable O&M cost (1995 $/acfm-hour)
- Capital cost-retrofit (1995 $) 6
- Capital cost-retrofit (1990 $) 6
- Fixed O&M cost (1995 $/year) 6
- Variable O&M cost (1995 $/year) 6
— Capital recovery factor
- Capital recovery cost (1995 $/year) 6
- Total annual cost (1995 $/year) 6
- Total annual cost (1990 $/year) 6
- Cost effectiveness (1995 $ton)
- Cost effectiveness (1990 $/ton)
811,321
124,063
107.6
3.35
0.000729
96,028,000
90,103,000
2,718,000
2,957,000
0.1098
10,544,000
16,219,000
15,218,000
140
130
NOTES:
1Emission point-specific input value obtained from the National Particulates Inventory
(NPI). Note that the gas flow rate in the NPI is reported in units of acfm. The NPI values
were converted to scfm for the model.
2These costs are for the high sulfur (3 percent) case in the report. They apply
to capacities at or above 500 MW (approximately 1,000,000 acfm). For smaller
sizes, report suggests scaling down costs via the standard 0.6 power law.
Thus, at lower capacities the capital cost (in $/kilowatt or$/acfm) will be higher
(EPA, 1996).
3Factor derived from data in IAPCS model (version 5).
"Ratio of Chemical Engineering Plant Indexes (annual) for 1990 and 1995,
respectively (357.6/381.1). Ratio used for both capital and annual costs.
5lf flow rate < 1,028,000 acfm, cost is scaled down via 0.6 power law.
"Values are rounded to $1,000.
49
-------
Table IV-3
SO2 Groups and SCCs
sec
30102301
30102306
30102308
30102310
30102318
30103201
30103202
30103203
30103204
30103299
30199999
30100509
30300315
30300401
30300302
30300303
30300304
30300306
30300308
30300313
30300314
31000402
31000403
31000404
31000405
31000412
31000413
30300101
30300102
30300105
30300103
30300199
30300201
30399999
30300813
30300817
30300824
30300825
30300901
30300908
30300911
30300931
30300933
30300999
30301001
30301002
30301101
30301199
30301201
30303003
30499999
30500606
30500612
30500622
SO2 Group
01
02
03
04
05
06
07
08
09
10
11
11
12a
12a
12b
12b
12b
12b
12b
12b
12b
13
13
13
13
13
13
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
15
16
16
16
Group Name
Sulfuric Acid Plants - Contact Absorber (99.9% Conversion)
Sulfuric Acid Plants - Contact Absorber (99% Conversion)
Sulfuric Acid Plants - Contact Absorber (98% Conversion)
Sulfuric Acid Plants - Contact Absorber (97% Conversion)
Sulfuric Acid Plants - Contact Absorber (93% Conversion)
Sulfur Recovery Plants - Elemental Sulfur (Claus: 2 Stage w/o control
Sulfur Recovery Plants - Elemental Sulfur (Claus: 3 Stage w/o control
Sulfur Recovery Plants - Elemental Sulfur (Claus: 3 Stage w/o control
(92-95% removal))
(95-96% removal))
(96-97% removal))
Sulfur Recovery Plants - Sulfur Removal Process (99.9% removal)
Sulfur Recovery Plants - Elemental Sulfur Production (Not Classified)
Inorganic Chemical Manufacture
Inorganic Chemical Manufacture
By-Product Coke Manufacturing (Coke Oven Plants)
By-Product Coke Manufacturing (Coke Oven Plants)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
By-Product Coke Manufacturing (Other Processes)
Process Heaters (Oil and Gas Production Industry)
Process Heaters (Oil and Gas Production Industry)
Process Heaters (Oil and Gas Production Industry)
Process Heaters (Oil and Gas Production Industry)
Process Heaters (Oil and Gas Production Industry)
Process Heaters (Oil and Gas Production Industry)
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Primary Metals Industry
Secondary Metal Production
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
50
-------
Table IV-3 (continued)
sec
30500706
30500801
30501037
30501401
30501402
30501403
30501410
30501602
30501604
30501905
30502201
30502509
30599999
30501001
30501002
30501201
30501202
30501203
30501212
30501404
30501499
30700104
30700106
30700110
30600101
30600102
30600103
30600104
30600105
30600106
30600199
30600201
30600202
30600204
30600301
30600401
30600402
30600503
30600504
30600805
30600902
30600903
30600904
30600999
30601001
30601101
30601201
30601401
30609903
30609904
30699998
30699999
10200201
10200202
10200203
SO2 Group
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
17
17
17
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
19
19
19
Group Name
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Mineral Products Industry
Pulp and Paper Industry (Sulfate Pulping)
Pulp and Paper Industry (Sulfate Pulping)
Pulp and Paper Industry (Sulfate Pulping)
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Petroleum Industry
Bituminous/Subbituminous Coal (Industrial
Bituminous/Subbituminous Coal (Industrial
Bituminous/Subbituminous Coal (Industrial
Boilers)
Boilers)
Boilers)
51
-------
Table IV-3 (continued)
sec
10200204
10200205
10200206
10200210
10200212
10200213
10200217
10200219
10200221
10200222
10200223
10200224
10200225
10200226
10200229
10200401
10200402
10200404
10200405
10300205
10300206
10300207
10300208
10300209
10300211
10300214
10300216
10300217
10300221
10300222
10300223
10300224
10300225
10300226
39000288
39000289
39000299
10200301
10200302
10200303
10200304
10200306
10200307
10300401
10300402
10300404
50100101
50100102
50100103
50100104
50100105
10200101
10200104
10200107
10200501
SO2 Group
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
22
22
22
23
23
23
23
23
23
24
24
24
25
25
25
25
25
26
26
26
26
Group Name
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Residual Oil (Industrial Boilers)
Residual Oil (Industrial Boilers)
Residual Oil (Industrial Boilers)
Residual Oil (Industrial Boileis)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
Bituminous/Subbituminous Coal (Commercial/Institutional Boilers)
In-process Fuel Use - Bituminous/Subbituminous Coal
In-process Fuel Use - Bituminous/Subbituminous Coal
In-process Fuel Use - Bituminous/Subbituminous Coal
Lignite (Industrial Boilers)
Lignite (Industrial Boilers)
Lignite (Industrial Boilers)
Lignite (Industrial Boilers)
Lignite (Industrial Boilers)
Lignite (Industrial Boilers)
Residual Oil (Commercial/Institutional Boilers)
Residual Oil (Commercial/Institutional Boilers)
Residual Oil (Commercial/Institutional Boilers)
Municipal Waste Combustors
Municipal Waste Combustors
Municipal Waste Combustors
Municipal Waste Combustors
Municipal Waste Combustors
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
52
-------
Table IV-3 (continued)
sec
10200502
10200504
10200505
10201101
10201403
10201404
10299997
10300101
10300102
10300103
10300305
10300306
10300307
10300309
10300501
10300502
10300504
30300502
30300509
30300513
30300525
30300529
30300530
30300503
30300507
30300510
30300512
30300514
30300523
30300531
30300526
30300527
30300532
30300533
30300534
30300535
30300504
30300515
30300528
30300521
30301001
30303002
30303007
30303008
SO2 Group
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
28
29
29
29
Group Name
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Steam Generating Unit-Coal/Oil
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Copper Smelters (copper converter, smelting furnace, and roaster)
Primary Lead Smelters - Sintering
Primary Zinc Smelters - Sintering
Primary Zinc Smelters - Sintering
Primary Zinc Smelters - Sintering
53
-------
3. Coke Ovens
The coke-oven gases produced by the controlled pyrolysis of coal contain
or J L J J
reduced sulfur compounds, in addition to numerous hydrocarbons. About 25 to 30 percent of the
sulfur in the coal is emitted in gaseous form as a constituent of the coke oven gas. Almost all of this
sulfur is present as hydrogen sulfide (H2S), with minor amounts of mercaptans. Using the coke
oven gas to heat or underfirethe coke ovens, or as fuel for other combustion operations, results in
SO2 emissions unless the H2S is removed.
Several processes are suitable for removing H2S from coke oven gases. In the
vacuum carbonate process, H2S is absorbed into a 3.0 to 3.5 percent solution of sodium carbonate.
The H2S is then stripped by steam from the absorbent in a reactivating tower. The reactivation is
performed under vacuum to reduce the quantity of steam required. Conventional systems achieve
about 90 percent removal.
Cost equations for vacuum carbonate (Emmel, et al., 1986):
Capital cost = $3,449,803 + $135.86 * Flow rate (ftYminute)
Operating cost = $797,667 + $58.54 * Flow rate (ftVminute)
4. Sulfur Recovery Plants
Refinery sour gas streams are generally fed to a regenerative type of H2S
removal process. The concentrated acid gas is then sent to the sulfur recovery unit. The Claus
process is the most widely used method of producing sulfur from refinery H2S. The modified Claus
process is based on producing elemental sulfur by first converting one-third of the H2S feed by
precise combustion with air. The combustion products are then allowed to react thermally with the
remaining two-thirds of the H2S feed in the presence of a suitable catalyst to form sulfur vapor.
Cost equations for amine scrubbing:
Capital cost = $2,882,540 + $244.74 * Flow rate (ftYminute)
Operating cost = $749,170 + $148.40 * Flow rate (ftVminute)
B. SOURCE CATEGORY DESCRIPTIONS
1. Industrial Steam Generation -
Bituminous/Subbituminous Coal
Gaseous sulfur oxide (SOX) from coal combustion are primarily SO2, with a
much lower quantity of sulfur trioxide (SO3) and gaseous sulfates. These compounds form as the
organic and pyritic sulfur in the coal are oxidized during the combustion process. On average,
about 95 percent of the sulfur present in bituminous coal will be emitted as gaseous SOX, whereas
somewhat less will be emitted when subbituminous coal is fired. The more alkaline nature of the
ash in some subbituminous coals causes some of the sulfur to react in the furnace to form various
sulfate salts that are retained in the boiler or in the flyash. Emissions from these sources are
classified under SCC 102002xx (EPA, 1997).
a. Description of Available Control
Options
Several techniques are used to reduce SO2 emissions from coal combustion.
One way is to switch to lower sulfur coals, since SO2 emissions are proportional to the sulfur
content of the coal. This alternative may not be possible where lower sulfur coal is not readily
available or where a different grade of coal cannot be satisfactorily fired. In some cases, various coal
o J '
cleaning processes may be employed to reduce the fuel sulfur content. Physical coal cleaning
removes mineral sulfur such as pyrite but is not effective in removing organic sulfur. Chemical
cleaning and solvent refining processes are being developed to remove organic sulfur (EPA, 1997).
54
-------
Post-combustion FGD techniques can remove SO2 formed during
combustion by using an alkaline reagent to absorb SO2 in the flue gas. Flue gases can be treated
using wet, dry, or semi-dry desulfarization processes of either the throwaway type (in which all was
streams are discarded) or the recovery/regenerable type (in which the SO2 absorbent is regenerated
and reused). To date, wet systems are the most commonly applied (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
industrial steam generation by combustion of bituminous/subbitumino us coal were determined
from EPA documents and recent technical literature. Wet FGD systems are a general category of
control device in which, for SO2 control, a liquid solution or liquid/solid slurry is used to absorb,
and, in most cases, react with SO2 in a waste gas stream. A scrubbing vessel, into which both the
solution or slurry and the waste gas are introduced, is used to maximize the contact between the
SO2 in the waste gas and the reacting compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO, from the flue gas. Sodium scrubbers are
J z o
generally limited to smaller sources because of high reagent costs and can have SO, removal
o J o o z
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
2. Industrial Steam Generation - Lignite
The SOX emissions from lignite combustion are a function of the sulfur
content of the lignite and the lignite composition (i.e., sulfur content, heating value, and alkali
concentration). The conversion of lignite sulfur to SOX is generally inversely proportional to the
concentration of alkali constituents in the lignite. The alkali content is known to have a great effect
on sulfur conversion and acts as a built-in sorbent for SOX removal. Emissions from these sources
are classified under SCC 102003xx (EPA, 1997).
a. Description of Available Control
Options
55
-------
FGD systems are in current operation on several lignite-fired utility boilers.
Flue gases can be treated through wet, semi-dry, or dry desulfurization processes of either the
throwaway type (in which all waste streams are discarded) or the recovery (regenerable) type (in
which the SO2 absorbent is regenerated and reused). To date, wet systems are the most commonly
applied. Wet systems generally use alkali slurries as the SO2 absorbent medium and can be designed
to remove in excess of 90 percent of the incoming SO2. Lime/limestone scrubbers, sodium
scrubbers, spray drying, and dual alkali scrubbing are among the commercially proven FGD
techniques (EPA, 1997).
Spray drying is a dry scrubbing approach in which a solution or slurry of
alkaline material is sprayed into a reaction vessel as a fine mist and mixes with the flue gas. The SO2
reacts with the alkali solution or slurry to form liquid-phase salts. The slurry is dried by the latent
heat of the flue gas to about 1 percent free moisture. The dried alkali continues to react with SO2 in
the flue gas to form sulfiteand sulfate salts. The spray dryer solids are entrained in the flue gas and
carried out of the dryer to a particulate control device such as an electrostatic precipitator (ESP) or
baghouse (EPA, 1997).
Limestone may also be injected into the furnace, typically in a fluidized bed
combuster (FBC), to react with SO2 and form calcium sulfate. An FBC is composed of a bed of
inert material that is suspended or "fluidized" by a stream of air. Lignite is injected into this bed
and burned. Limestone is also injected into this bed where it is calcined to lime and reacts with SO2
to form calcium sulfate. Particulate matter emitted from the boiler is generally captured in a
cyclone and recirculated or sent to disposal. Additional PM control equipment, such as an ESP or
baghouse, is used after the cyclone to further reduce particulate emissions (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
industrial steam generation by combustion of lignite were determined from EPA documents and
recent technical literature. Wet FGD systems are a general category of control device in which, for
SO2 control, a liquid solution or liquid/solid slurry is used to absorb, and, in most cases, react with
SO2 in a waste gas stream. A scrubbing vessel, into which both the solution or slurry and the waste
gas are introduced, is used to maximize the contact between the SO, in the waste gas and the
o ' z o
reacting compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems include: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO, removal
o J o o z
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
56
-------
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
3. Industrial Steam Generation - Residual Oil
SOX emissions are generated during oil combustion from the oxidation of
sulfur contained in the fuel. The emissions of SOX from conventional combustion systems are
predominately in the form of SO2. Uncontrolled SOX emissions are almost entirely dependent on
the sulfur content of the fuel and are not affected by boiler size, burner design, or grade of fuel
being fired. On average, more than 95 percent of the fuel sulfur is converted to SO2, about 1 to 5
percent is further oxidized to SO3, and 1 to 3 percent is emitted as sulfate particulate. SO3 readily
reacts with water vapor (both in the atmosphere and in flue gases) to form a sulfuric acid mist.
Emissions from these sources are classified under SCC 102004xx (EPA, 1997).
a. Description of Available Control
Options
Commercialized FGD processes use an alkaline reagent to absorb SO2 in the
flue gas and produce a sodium or a calcium sulfate compound. These solid sulfate compounds are
then removed in downstream equipment. FGD technologies are categorized as wet, semi-dry, or
dry depending on the state of the reagent as it leaves the absorber vessel. These processes are either
regenerable (such that the reagent material can be treated and reused) or nonregenerable (in which
case all waste streams are de-watered and discarded) (EPA, 1997).
Wet regenerable FGD processes are attractive because they have the
potential for better than 95 percent sulfur removal efficiency, have minimal waste water discharges,
and produce a saleable sulfur product. Some of the current nonregenerable calcium-based processes
can, however, produce a saleable gypsum product (EPA, 1997).
To date, wet systems are the most commonly applied. Wet systems
generally use alkali slurries as the SO2 absorbent medium and can be designed to remove greater
than 90 percent of the incoming SO2. Lime/limestone scrubbers, sodium scrubbers, and dual alkali
scrubbing are among the commercially proven wet FGD systems. Effectiveness of these devices
depends not only on control device design but also on operating variables (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
industrial steam generation by combustion of residual oil were determined from EPA documents
and recent technical literature. Wet FGD systems are a general category of control device in
which, for SO2 control, a liquid solution or liquid/solid slurry is used to absorb, and, in most cases,
react with SO, in a waste gas stream. A scrubbing vessel, into which both the solution or slurry and
^ o o ' J
the waste gas are introduced, is used to maximize the contact between the SO, in the waste gas and
o ' ^ o
the reacting compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
57
-------
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects of inflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
4. Commercial/Institutional Steam Generation -
Bituminous/Subbituminous Coal
Gaseous SOX from coal combustion are primarily SO2, with a much lower
quantity of SO3 and gaseous sulfates. These compounds form as the organic and pyritic sulfur in the
coal are oxidized during the combustion process. On average, about 95 percent ofthe sulfur
present in bituminous coal will be emitted as gaseous SOX, whereas somewhat less will be emitted
when subbituminous coal is fired. The more alkaline nature ofthe ash in some subbituminous coals
causes some of the sulfur to react in the furnace to form various sulfate salts that are retained in the
boiler or in the flyash. Emissions from these sources are classified under SCC 103002xx.
a. Description of Available Control
Options
Several techniques are used to reduce SO2 emissions from coal combustion.
One way is to switch to lower sulfur coals, since SO2 emissions are proportional to the sulfur
content ofthe coal. This alternative may not be possible where lower sulfur coal is not readily
available or where a different grade of coal cannot be satisfactorily fired. In some cases, various coal
cleaning processes may be employed to reduce the fuel sulfur content. Physical coal cleaning
removes mineral sulfur such as pyrite but is not effective in removing organic sulfur. Chemical
cleaning and solvent refining processes are being developed to remove organic sulfur (EPA, 1997).
Post-combustion FGD techniques can remove SO2 formed during
combustion by using an alkaline reagent to absorb SO2 in the flue gas. Flue gases can be treated
using wet, dry, or semi-dry desulfurization processes of either the throwaway type (in which all was
streams are discarded) or the recovery/regenerable type (in which the SO2 absorbent is regenerated
and reused). To date, wet systems are the most commonly applied (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
commercial and institutional steam generation by combustion of bituminous/subbituminous coal
were determined from EPA documents and recent technical literature. Wet FGD systems are a
general category of control device in which, for SO2 control, a liquid solution or liquid/solid slurry
is used to absorb, and, in most cases, react with SO, in a waste gas stream. A scrubbing vessel, into
' ' ' z O O '
which both the solution or slurry and the waste gas are introduced, is used to maximize the contact
between the SO2 in the waste gas and the reacting compounds in the solution or slurry (EPA, 1981).
58
-------
The design of the scrubbing vessel and the manner in which the waste gas and
o o o
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO, removal followed by a regeneration step using lime or limestone to recover the
z J o L O
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
5. Commercial/Institutional Steam Generation - Residual
Oil
SO emissions are generated during oil combustion from the oxidation of
x o o
sulfur contained in the fuel. The emissions of SOX from conventional combustion systems are
predominately in the form of SO2. Uncontrolled SOX emissions are almost entirely dependent on
the sulfur content of the fuel and are not affected by boiler size, burner design, or grade of fuel
being fired. On average, more than 95 percent of the fuel sulfur is converted to SO2, about 1 to 5
percent is further oxidized to SO3, and 1 to 3 percent is emitted as sulfate particulate. SO3 readily
reacts with water vapor (both in the atmosphere and in flue gases) to form a sulfuric acid mist.
Emissions from these sources are classified under SCC 102004xx (EPA, 1997).
a. Description of Available Control
Options
Commercialized FGD processes use an alkaline reagent to absorb SO2 in the
flue gas and produce a sodium or a caldum sulfate compound. These solid sulfate compounds are
then removed in downstream equipment. FGD technologies are categorized as wet, semi-dry, or
dry depending on the state of the reagent as it leaves the absorber vessel. These processes are either
regenerable (such that the reagent material can be treated and reused) or nonregenerable (in which
case all waste streams are de-watered and discarded) (EPA, 1997).
Wet regenerable FGD processes are attractive because they have the
potential for better than 95 percent sulfur removal efficiency, have minimal waste water discharges,
and produce a saleable sulfur product. Some of the current nonregenerable calcium-based processes
can, however, produce a saleable gypsum product (EPA, 1997).
59
-------
To date, wet systems are the most commonly applied. Wet systems
generally use alkali slurries as the SO2 absorbent medium and can be designed to remove greater
than 90 percent of the incoming SO2. Lime/limestone scrubbers, sodium scrubbers, and dual alkali
scrubbing are among the commercially proven wet FGD systems. Effectiveness of these devices
O O J L J
depends not only on control device design but also on operating variables (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
commercial and institutional steam generation by combustion of residual oil were determined from
EPA documents and recent technical literature. Wet FGD systems are a general category of control
device in which, for SO2 control, a liquid solution or liquid/solid slurry is used to absorb, and, in
most cases, react with SO2 in a waste gas stream. A scrubbing vessel, into which both the solution
or slurry and the waste gas are introduced, is used to maximize the contact between the SO2 in the
waste gas and the reacting compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
O O O
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
J J O
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems include: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects of inflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
6. Sulfuric Acid - Contact Process
The contact process for creating sulfuric acid is designed for use in plants
where elemental sulfur is oxidized specifically in order to make acid. Nonferrous metals smelting
L J O
plants, as well as some petroleum refineries, typically have high enough concentrations of SO2 in
waste gases to warrant construction of a sulfuric acid plant. Emissions from these sources are
classified under SCC 301023xx (EPA, 1981; Friedman, 1981).
a. Description of Available Control
Options
60
-------
Nearly all sulfur dioxide emissions from sulfuric acid plants are found in the
exit stack gases. Extensive testing has shown that the mass of these SO2 emissions is an inverse
function of the sulfur conversion efficiency (SO2 oxidized to SO3 ). This conversion is always
incomplete, and is affected by the number of stages in the catalytic converter, the amount of catalyst
L ' J to J ' J
used, temperature and pressure, and the concentrations of the reactants (sulfur dioxide and
oxygen). The double contact process has generally been accepted as the best available control
technology for meeting NSPS emission limits. There are no byproducts or waste scrubbing
materials created, only additional sulfuric acid. Wellman-Lord scrubbing is an alternative control
technology for SO2 emissions not controlled by a single- or double-contact sulfuric acid plant (EPA,
1997).
b. Control Options Selected for Analysis
Emission control and cost data for double-contact sulfuric acid plants and
Wellman-Lord scrubbing systems applied to single contact sulfuric acid plants were determined
from EPA documents and recent technical literature. The double-contact, or double-absorption,
process for making sulfuric acid from waste gas containing SO2 is essentially the same as the contact
process with the addition of an interpass absorption tower. The waste gas is cleaned and dried as in
the single-contact process before entering the process. Upon leaving the second or third catalyst
bed, depending upon the process, the gas is cooled and introduced to a packed-bed, counter-current
absorption tower where it contacts 98 to 99 percent sulfuric acid. After the absorbing tower, the
gas is reheated and passed to the third or fourth catalyst bed, where approximately 97 percent of the
remaining SO2 is converted to SO3 and passed to the final absorption tower for conversion to
sulfuric acid as in the single-contact process. The typical control efficiency is from 98 to greater
than 99 percent (EPA, 1981; EPA, 1997; Friedman, 1981.)
In the Wellman-Lord scrubbing process an aqueous sodium sulfite solution is
used to absorb SO2, usually in a counter-current tray-type column absorber. Sodium bisulfite is
formed as the SO, is absorbed from the gas steam. The SO, is then released in a concentrated
z to z
stream in the stripping step, in which sodium sulfite is recovered and returned to the absorber loop.
The concentrated SOX stream with water vapor enters a condenser, where most of the water is
removed. If necessary, the resulting SO2 stream may be further dried in a concentrated sulfuric acid
drying tower. Sulfur compounds from the SOX stream may be recovered as liquid SO2, liquid SO3,
sulfuric acid, or elemental sulfur, as determined by potential use, market demand, and cost of
transportation to the destination (EPA, 1997; EPA, 1981).
Before reaching the tray-tower absorber, particulates or fly ash are removed
from the stream by an electrostatic precipitator, fabric filter, wet particulate scrubber, or other
device. The stream is normally cooled to its adiabatic saturation temperature in a wet scrubber or
presaturator. Humidification of the stream helps to reduce the evaporation of water in the
absorbing scrubber. This step also serves to remove most of the chlorides in the stream, which can
cause the scrubber water to become acidic, leading to stress corrosion. The typical control
efficiency range is from 90 to 99 percent (EPA, 1997; EPA, 1981).
Capital costs for a double-contact acid plant ranges from $15 to 50 million
(mid-1979 dollars) and operating costs range from $5 to $30 million (mid-1979 dollars) per year.
Capital costs for a Wellman-Lord scrubbing system range from $500,000 to $2,500,000 (mid-1979
dollars), and operating costs range from $300,000 to $900,000 (mid-1979 dollars). The cost
efficiency of Wellman-Lord scrubbing systems was estimated at $625 (1990 dollars) per ton of SO2
controlled, based a 15-year equipment lifetime and not adjusted for inflation. Equipment lifetimes
and discount rates for double contact acid plants were not found among the references (EPA, 1981;
Radcliffe, 1992).
7. Primary Copper Smelters - Copper Converter, Smelting
Furnace, and Roaster
Roasters, smelting furnaces, and converters are sources of SO . In the
' to ' x
standard Fierce-Smith copper converter, flue gases are captured during the blowing phase by the
primary hood over the converter mouth. To prevent the hood from binding to the converter with
splashing molten metal, a gap exists between the hood and the vessel. During charging and pouring
61
-------
operations, significant fugitives may be emitted when the hood is removed to allow crane access.
Remaining smelter operations process material containing very little sulfur, resulting in insignificant
SO2 emissions. Electrolytic refining does not produce emissions unless the associated sulfuric acid
tanks are open to the atmosphere. Emissions from these sources are classified under SCC 303005xx
(EPA, 1997).
a. Description of Available Control
Options
Control of SO2 from smelters is commonly performed by a sulfuric acid
plant. Use of a sulfuric acid plant to treat copper smelter effluent gas streams requires that
particulate-free gas containing minimum SO2 concentrations, usually of at least three percent SO2,
be maintained. Sulfuric acid plants also treat converter gas effluent. Some multiple hearth and all
fluidized bed roasters use sulfuric acid plants. Reverberatory furnace effluent contains minimal SO2
and is usually released directly to the atmosphere with no SO2 reduction. Effluent from the other
types of smelter furnaces contain higher concentrations of SO2 and are treated in sulfuric acid plants
before being vented. Absorption of the SO2 in dimethylaniline (DMA) solution has also been used
in domestic smelters to produce liquid SO2 (EPA, 1997).
b. Control Options Selected for Analysis
Emission control data for sulfuric acid plants and DMA scrubbing systems
applied to primary copper smelters were determined from EPA documents and recent technical
literature. The contact process is used to produce sulfuric acid from waste gas which contains SO2.
First, the waste gas must be pretreated, which usually involves dust removal, cooling, and scrubbing
for further removal of particulate matter and heavy metals, mist, and moisture. After
pretreatment, the gas is heated and passed through a catalytic converter (platinum mass units or
units containing beds of pelletized vanadium pentoxide) to oxidize the SO2 to SO3. The
exothermic, reversible oxidation reaction results in a conflict between high equilibrium conversions
at lower temperatures and high reaction rates at high temperatures. Because of this, the gas is
passed between the catalyst and two or three different heat exchangers in order to achieve
conversion of SO2 to SO3 of about 92.5 to 98 percent. The gas leaving the final catalyst stage is
cooled and introduced to an absorption tower by a stream of strong (98 to 99 percent) acid, where
the SO3 reacts with water in the acid to form additional sulfuric acid. Dilute sulfuric acid or water
is added to the recirculating acid to maintain the desired concentration (EPA, 1981; EPA, 1997).
The double-contact, or double-absorption, process for making sulfuric acid
from waste gas containing SO2 is essentially the same as the single-contact process with the addition
of an interpass absorption tower. The waste gas is cleaned and dried as in the single-con tact process
before entering the process. Upon leaving the second or third catalyst bed, depending upon the
process, the gas is cooled and introduced to a packed-bed, counter-current absorption tower where
it contacts 98 to 99 percent sulfuric acid. After the absorbing tower, the gas is reheated and passed
to the third or fourth catalyst bed, where approximately 97 percent of the remaining SO2 is
converted to SO3 and passed to the final absorption tower for conversion to sulfuric acid as in the
single-contact process. No cost data were available for either single- or double-contact sulfuric acid
plants controls (EPA, 1981; EPA, 1997).
The DMA absorption system is a cyclic-regenerative process that
incorporates an absorber with trays on which most of the incoming SO2 is absorbed in a
countercurrent stream of DMA. The residual SO, in the gases is scrubbed with a weak sodium
z o
carbonate solution to give sodium sulfite or sodium bisulfite. Liquid sulfur dioxide is recovered as a
product, and its absorbent is regenerated and recycled through the system. The typical control
efficiency range is from 95 to 99 percent.
Capital costs for a double-contact acid plant ranges from $ 15 to 50 million
(mid-1979 dollars) and operating costs range from $5 to $30 million (mid-1979 dollars) per year.
Capital costs for DMA scrubbing systems are estimated to be $50million (mid-1979 dollars), and
annual operating costs are estimated at approximately $12 million (mid-1979 dollars). No cost
efficiency estimate, equipment lifetime, or discount rate were available in the references for either
control system (AWMA, 1992; EPA, 1981; Radcliffe, 1992).
62
-------
8. Primary Zinc Smelters - Sintering
The processing of zinc concentrate into metallurgical lead involves three
major steps: sintering, reduction, and refining. The primary purpose of the sinter machine is the
reduction of sulfur content of the feed material. Emissions from these sources are classified under
SCC 30303003 (EPA, 1997).
a. Description of Available Control
Options
Control of SO2 from smelters is commonly performed in a sulfuric acid
plant. Use of a sulfuric acid plant to treat zinc smelter sinter effluent gas streams requires that
particulate-free gas containing minimum SO2 concentrations, usually of at least three percent SO2,
be maintained (EPA, 1997).
b. Control Options Selected for Analysis
Emission control data for sulfuric acid plants applied to primary zinc smelter
sintering processes were determined from EPA documents and recent technical literature. The
contact process is used to produce sulfuric acid from waste gas which contains SO2. First, the waste
gas must be pretreated, which usually involves dust removal, cooling, and scrubbing for further
removal of particulate matter and heavy metals, mist, and moisture. After pretreatment, the gas is
heated and passed through a catalytic converter (platinum mass units or units containing beds of
pelletized vanadium pentoxide) to oxidize the SO2 to SO3. The exothermic, reversible oxidation
reaction results in a conflict between high equilibrium conversions at lower temperatures and high
reaction rates at high temperatures. Because of this, the gas is passed between the catalyst and two
or three different heat exchangers in order to achieve conversion of SO, to SO, of about 92.5 to 98
o z J
percent. The gas leaving the final catalyst stage is cooled and introduced to an absorption tower by
a stream of strong (98 to 99 percent) acid, where the SO3 reacts with water in the acid to form
additional sulfuric acid. Dilute sulfuric acid or water is added to the recirculating acid to maintain
the desired concentration (EPA, 1981; EPA, 1997).
The double-contact, or double-absorption, process for making sulfuric acid
from waste gas containing SO2 is essentially the same as the single-contact process with the addition
of an interpass absorption tower. The waste gas is cleaned and dried as in the single-con tact process
before entering the process. Upon leaving the second or third catalyst bed, depending upon the
process, the gas is cooled and introduced to a packed-bed, counter-current absorption tower where
it contacts 98 to 99 percent sulfuric acid. After the absorbing tower, the gas is reheated and passed
to the third or fourth catalyst bed, where approximately 97 percent of the remaining SO2 is
converted to SO3 and passed to the final absorption tower for conversion to sulfuric acid as in the
single-contact process. No cost data were available for either single- or double-contact sulfuric acid
plants controls (EPA, 1981; EPA, 1997).
Capital costs for a double-contact acid plant ranges from $ 15 to 50 million
(mid-1979 dollars) and operating costs range from 5 to 30 million (mid-1979 dollars) per year.
Cost efficiency estimates, equipment lifetimes, and discount rates for double-contact acid plants
were not found among the references (EPA, 1981; Radcliffe, 1992).
9. Primary Lead Smelters - Sintering
The processing of lead concentrate into metallurgical lead involves three
major steps: sintering, reduction, and refining. The primary purpose of the sinter machine is the
reduction of sulfur content of the feed material. This feed material typically consists of the
following:
o
1. Lead concentrates, including pyrite concentrates that are high in sulfur
' o 1 J o
content;
2. Lime rock and silica, incorporated in the feed to main tain a desired sulfur
content;
3. High-lead-content sludge byproducts from other facilities; and
63
-------
4. Undersized sinter recycled from the roast exiting the sinter machine.
J o
Emissions from these sources are classified under SCC 30301029 (EPA, 1997).
a. Description of Available Control
Options
Control of SO2 from smelters is commonly performed by a sulfuric acid
plant. Use of a sulfuric acid plant to treat lead smelter sinter effluent gas streams requires that
particulate-free gas containing minimum SO2 concentrations, usually of at least three percent SO2,
be maintained. Absorption of the SO2 in DMA solution has also been used in domestic smelters to
produce liquid SO2 (EPA, 1997).
b. Control Options Selected for Analysis
Emission control data for sulfuric acid plants and DMA scrubbing systems
applied to primary lead smelter sintering processes were determined from EPA documents and
recent technical literature. The contact process is used to produce sulfuric acid from waste gas
which contains SO2. First, the waste gas must be pretreated, which usually involves dust removal,
cooling, and scrubbingfor further removal of particukte matter and heavy metals, mist, and
O' o L J ' '
moisture. After pretreatment, the gas is heated and passed through a catalytic converter (platinum
mass units or units containing beds of pelletized vanadium pentoxide) to oxidize the SO2 to SO3.
The exothermic, reversible oxidation reaction results in a conflict between high equilibrium
' O 1
conversions at lower temperatures and high reaction rates athigh temperatures. Because of this,
the gas is passed between the catalyst and two or three different heat exchangers in order to achieve
conversion of SO2 to SO3 of about 92.5 to 98 percent. The gas leaving the final catalyst stage is
cooled and introduced to an absorption tower by a stream of strong (98 to 99 percent) acid, where
the SO3 reacts with water in the acid to form additional sulfuric acid. Dilute sulfuric acid or water
is added to the recirculating acid to maintain the desired concentration (EPA, 1981; EPA, 1997).
The double-contact, or double-absorption, process for making sulfuric acid
from waste gas containing SO2 is essentially the same as the single-contact process with the addition
of an interpass absorption tower. The waste gas is cleaned and dried as in the single-contact process
before entering the process. Upon leaving the second or third catalyst bed, depending upon the
process, the gas is cooled and introduced to a packed-bed, counter-current absorption tower where
it contacts 98 to 99 percent sulfuric acid. After the absorbing tower, the gas is reheated and passed
to the third or fourth catalyst bed, where approximately 97 percent of the remaining SO2 is
converted to SO3 and passed to the final absorption tower for conversion to sulfuric acid as in the
single-contact process. No cost data was available for either single- or double-contact sulfuric acid
plants controls (EPA, 1981; EPA, 1997).
The DMA absorption system is a cyclic-regenerative process that
incorporates an absorber with trays on which most of the incoming SO 2 is absorbed in a
countercurrent stream of DMA. The residual SO, in the gases is scrubbed with a weak sodium
z o
carbonate solution to give sodium sulfite or sodium bisulfite. Liquid sulfur dioxide is recovered as a
product, and its absorbent is regenerated and recycled through the system. The typical control
efficiency range is from 95 to 99 percent. Capital costs are estimated to be $50 million (mid-1979
dollars), and operating costs are estimated at approximately $12 million (mid-1979 dollars). No
cost efficiency estimate, equipment lifetime, or discount rate were available in the references
(AWMA, 1992; EPA, 1981).
10. Petroleum Refineries - Fluid Catalytic Cracking Units
(FCCU)
Fluid catalytic cracking, using heat, pressure, and catalysts, converts heavy
oils into lighter products with product distributions favoring the more valuable gasoline and
distillate blending components. Feedstocks are usually gas oils from atmospheric distillation,
vacuum distillation, coking, and deasphalting processes. These feedstocks typically have a boiling
range of 340 to 540°C (650 to 1000°F). All of the catalytic cracking processes in use today can be
classified as either fluidized-bed or moving-bed units (EPA, 1997).
64
-------
The fluid catalytic cracking process uses a catalyst in the form of very fine
particles that act as a fluid when aerated with a vapor. Fresh feed is preheated in a process heater
and introduced into the bottom of a vertical transfer line or riser with hot regenerated catalyst. The
o J
hot catalyst vaporizes the feed, bringing both to the desired reaction temperature, 470 to 525°C
(880 to 980°F). The high activity of modern catalysts causes most of the cracking reactions to take
place in the riser as the catalyst and oil mixture flows upward into the reactor. The hydrocarbon
vapors are separated from the catalyst particles by cyclones in the reactor. The reaction products
are sent to a fractionator for separation (EPA, f 997).
The spent catalyst falls to the bottom of the reactor and is steam stripped as
it exits the reactor bottom to remove absorbed hydrocarbons. The spent catalyst is then conveyed
to a regenerator. In the regenerator, coke deposited on the catalyst as a result of the cracking
reactions is burned off in a controlled combustion process with preheated air. Regenerator
temperature is usually 590 to 675°C (f f 00 to f 250°F). The catalyst is then recycled to be mixed
with fresh hydrocarbon feed. Emissions from these sources are classified under SCC 3060020f
(EPA, 1997).
a. Description of Available Control
Options
Commercialized FGD processes use an alkaline reagent to absorb SO2 in the
flue gas and produce a sodium or a calcium sulfate compound. These solid sulfate compounds are
then removed in downstream equipment. FGD technologies are categorized as wet, semi-dry, or
dry depending on the state of the reagent as it leaves the absorber vessel. These processes are either
regenerable (such that the reagent material can be treated and reused) or nonregenerable (in which
case all waste streams are de-watered and discarded) (EPA, f 997).
Wet regenerable FGD processes are attractive because they have the
potential for better than 95 percent sulfur removal efficiency, have minimal waste water discharges,
and produce a saleable sulfur product. Some of the current nonregenerable calcium-based processes
can, however, produce a saleable gypsum product (EPA, 1997).
To date, wet systems are the most commonly applied. Wet systems
generally use alkali slurries as the SO2 absorbent medium and can be designed to remove greater
than 90 percent of the incoming SO2. Lime/limestone scrubbers, sodium scrubbers, and dual alkali
scrubbing are among the commercially proven wet FGD systems. Effectiveness of these devices
depends not only on control device design but also on operating variables (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
FCCU at petroleum refineries were determined from EPA documents and recent technical
literature. Wet FGD systems are a general category of control device in which, for SO2 control, a
liquid solution or liquid/solid slurry is used to absorb, and, in most cases, react with SO2 in a waste
gas stream. A scrubbing vessel, into which both the solution or slurry and the waste gas are
o o ' J o
introduced, is used to maximize the contact between the SO, in the waste gas and the reacting
' z o o
compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
o o o
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
J J o
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems include: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
65
-------
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects of inflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
11. Petroleum Refineries - Claus Sulfur Recovery
Sulfur recovery plants are used in petroleum refineries to convert the H2S
separated from refinery gas streams into the more disposable, and sometimes saleable, byproduct,
elemental sulfur. Most plants are now built with two catalytic stages, although some air quality
jurisdictions require three. From the condenser of the final catalytic stage, the process stream
passes to some form of tailgas treatment process. The tailgas, containing H2S, SO2, sulfur vapor,
and traces of other sulfur compounds formed in the combustion section, escapes with the inert gases
from the tail end of the plant. Thus, it is frequently necessary to follow the Claus unit with a tailgas
cleanup unit to achieve higher recovery. Emissions from these sources are classified under SCC
30603301 (EPA, 1997).
a. Description of Available Control
Options
Emissions from the Claus process may be reduced by: (1) extending the
Claus reaction into a lower temperature liquid phase; (2) adding a scrubbing process to the Claus
exhaust stream; or (3) incinerating the hydrogen sulfide gases to form sulfur dioxide. Currently,
there are five processes available that extend the Claus reaction into a lower temperature liquid
phase including the BSR/selectox, Sulfreen, Cold Bed Absorption, Maxisulf, and IFP-1 processes.
These processes take advantage of the enhanced Claus conversion at cooler temperatures in the
catalytic stages. All of these processes give higher overall sulfur recoveries of 98 to 99 percent
when following downstream of a typical two- or three-stage Claus sulfur recovery unit, and
therefore reduce sulfur emissions (EPA, 1997).
Sulfur emissions can also be reduced by adding a scrubber at the tail end of
J o
the plant. There are essentially two generic types of tailgas scrubbing processes: oxidation tailgas
scrubbers and reduction tailgas scrubbers. The first scrubbing process is used to scrub SO, from
o or t
incinerated tailgas and recycle the concentrated SO2 stream back to the Claus process for conversion
to elemental sulfur. There are at least three oxidation scrubbing processes: the sodium sulfite
(Wellman-Lord), Stauffer Aquaclaus, and IFP-2. Only the sodium scrubbing process has been
applied successfully to U.S. refineries (EPA, 1997).
In the second type of scrubbing process, sulfur in the tailgas is converted to
H2Sby
hydrogenation in a reduction step. After hydrogenation, the tailgas is cooled and water is removed.
The cooled tailgas is then sent to the scrubber for H2S removal prior to venting. There are at least
four reduction scrubbing processes developed for tailgas sulfur removal: Beavon, Beavon MDEA,
SCOT, and ARCO. In the Beavon process, H2S is converted to sulfur outside the Claus unit using a
66
-------
lean H2S-to-sulfur process (the Strefford process). The other three processes utilize conventional
arnine scrubbing and regeneration to remove H2S and recycle back as Claus feed (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for sodium FGD scrubbing systems applied to
Claus process sulfur plants at petroleum refineries were determined from EPA documents and
recent technical literature. In the Wellman-Lord scrubbing process an aqueous sodium sulfite
solution is used to absorb SO2, usually in a counter-current tray-type column absorber. Sodium
bisulfite is formed as the SO2 is absorbed from the gas steam. The SO2 is then released in a
concentrated stream in the stripping step, in which sodium sulfite is recovered and returned to the
absorber loop. The concentrated SOX stream with water vapor enters a condenser, where most of
the water is removed. If necessary, the resulting SO2 stream may be further dried in a concentrated
sulfuric acid drying tower. Sulfur compounds from the SOX stream may be recovered as liquid SO2,
liquid SO3, sulfuric acid, or elemental sulfur, as determined by potential use, market demand, and
cost of transportation to the destination (EPA, 1997; EPA, 1981).
Before reaching the tray-tower absorber, particulates or fly ash are removed
from the stream by an electrostatic precipitator, fabric filter, wet particulate scrubber, or other
device. The stream is normally cooled to its adiabatic saturation temperature in a wet scrubber or
presaturator. Humidification of the stream helps to reduce the evaporation of water in the
absorbing scrubber. This step also serves to remove most of the chlorides in the stream, which can
cause the scrubber water to become acidic, leading to stress corrosion. The typical control
efficiency range is from 90 to 99 percent (EPA, 1997; EPA, 1981).
Capital costs for a Wellman-Lord scrubbing system ranges from $500,000 to
$2,500,000 (mid-1979 dolkrs), and operating costs range from $300,000 to $900,000 (mid-1979
dollars). The cost efficiency of Wellman-Lord scrubbing systems was estimated at $625 (1990
dollars) per ton of SO2 controlled, based a 15-year equipment lifetime andnot adjusted for inflation
(EPA, 1981; Radcliffe, 1992).
12. Natural Gas Processing
Natural gas from high-pressure wells is usually passed through field
separators at the well to remove hydrocarbon condensate and water. Natural gasoline, butane, and
propane are usually present in the gas, and gas processing plants are required for the recovery of
these liquefiable constituents. Natural gas is considered "sour" if H2S is present in amounts greater
than 5.7 milligrams per normal cubic meters (mg/nm3) (0.25 grains per 100 standard cubic feet
[gr/100 scf]). The H2S must be removed (called "sweetening" the gas) before the gas can be
utilized. Many chemical processes are available for sweetening natural gas. At present, the amine
process (also known as the Girdler process), is the most widely used method for H2S removal.
Emissions from these sources are classified under SCC 310002xx (EPA, 1997).
a. Description of Available Control
Options
The H2S gas removed through processing natural gas may be: (1) vented; (2)
flared in waste gas flares or modern smokeless flares; (3) incinerated; or (4) utilized for the
production of elemental sulfur or sulfuric acid (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for elemental sulfur production using the
Claus process applied to natural gas processing were determined from EPA documents and recent
technical literature. The Claus process involves burning one-third of the available H2S with air in a
reactor furnace to form SO2. The furnace normally operates at combustion chamber temperatures
ranging from 980 to 1540°C (1800 to 2800°F) with pressures rarelyhigher than70 kilopascals (kPa)
(10 pounds per square inch absolute). Before entering a sulfur condenser, hot gas from the
combustion chamber is quenched in a waste heat boiler that generates high to medium pressure
steam. About 80 percent of the heat released could be recovered as useful energy. Liquid sulfur
67
-------
from the condenser runs through a seal leg into a covered pit from which it is pumped to trucks or
railcars for shipment to end users. Approximately 65 to 70 percent of the sulfur is recovered. The
cooled gases exiting the condenser are then sent to the catalyst beds (EPA, 1997; EPA, 1981).
The remaining uncombusted two-thirds of the hydrogen sulfide undergoes
o J o o
Claus reaction (reacts with SO2) to form elemental sulfur. The catalytic reactors operate at lower
temperatures, ranging from 200 to 315 ° C (400 to 600° F). Alumina or bauxite is sometimes used as
a catalyst. Because this reaction represents an equilibrium chemical reaction, it is not possible for a
Claus plant to convert all the incoming sulfur compounds to elemental sulfur. Therefore, two or
more stages are used in series to recover the sulfur. Each catalytic stage can recover half to two-
o J o
thirds of the incoming sulfur. The number of catalytic stages depends upon the level of conversion
desired (EPA, 1997; EPA, 1981).
The typical control efficiency range for a Claus process sulfur plant is from
95 to 99 percent. A two-stage catalytic Claus plant can achieve 94 to 96 percent efficiency, while a
three-stage ranges from 96 to 97.5 percent, and a four-stage from 97 to 98.5 percent. The addition
of a low-temperature stage after the final normal-temperature stage can extend control efficiencies
to the 98 to 99 percent range (EPA, 1997; EPA, 1981).
Capital costs for a Claus sulfur plant ranges from $ 1 to 4 million (mid-1979
dollars) and operating costs range from $100,000 to $400,000 (mid-1979 dollars) per year.
Operating costs are also estimated to be between $119 and $124 (1981 dollars) per ton of sulfur
produced. Cost efficiency estimates, equipment lifetimes, and discount rates for double contact
acid plants were not found among the references (EPA, 1981; Friedman, 1981).
13. In-process Fuel Use - Bituminous/Subbituminous Coal
Coal is commonly used as a fuel to fire kilns in the manufacture of both
cement and lime. SO2 may be generated both from the sulfur compounds in the raw materials and
from sulfur in the fuel. The sulfur content of both raw materials and fuels varies from plant to plant
and with geographic location. However, the alkaline nature of the cement and the presence of
calcium oxides from lime provides for direct absorption of SO2, thereby mitigating the quantity of
SO2 emissions in the exhaust stream. Depending on the process and the source of the sulfur, SO2
absorption ranges from about 70 percent to more than 95 percent. Emissions from these sources
are classified under SCC 390002xx (EPA, 1997).
a. Description of Available Control
Options
Commercialized FGD processes use an alkaline reagent to absorb SO2 in the
flue gas and produce a sodium or a caldum sulfate compound. These solid sulfate compounds are
then removed in downstream equipment. FGD technologies are categorized as wet, semi-dry, or
dry depending on the state of the reagent as it leaves the absorber vessel. These processes are either
regenerable (such that the reagent material can be treated and reused) or nonregenerable (in which
case all waste streams are de-watered and discarded) (EPA, 1997).
Wet regenerable FGD processes are attractive because they have the
potential for better than 95 percent sulfur removal efficiency, have minimal waste water discharges,
and produce a saleable sulfur product. Some of the current nonregenerable calcium-based processes
can, however, produce a saleable gypsum product (EPA, 1997).
To date, wet systems are the most commonly applied. Wet systems
generally use alkali slurries as the SO2 absorbent medium and can be designed to remove greater
than 90 percent of the incoming SO2. Lime/limestone scrubbers, sodium scrubbers, and dual alkali
scrubbing are among the commercially proven wet FGD systems. Effectiveness of these devices
depends not only on control device design but also on operating variables (EPA, 1997).
b. Control Options Selected for Analysis
68
-------
Emission control and cost data for wet FGD scrubbing systems applied to
FCCU at petroleum refineries were determined from EPA documents and recent technical
literature. Wet FGD systems are a general category of control device in which, for SO2 control, a
liquid solution or liquid/solid slurry is used to absorb, and, in most cases, react with SO2 in a waste
gas stream. A scrubbing vessel, into which both the solution or slurry and the waste gas are
introduced, is used to maximize the contact between the SO, in the waste gas and the reacting
' z o o
compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
o o o
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO, removal followed by a regeneration step using lime or limestone to recover the
z J o 1 o
sodium alkali and produce a calcium sulfite and sulfate sludge. SO 2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $ 260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates in journal artides range from
$340 to $630 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
14. Municipal Waste Combustors (MWC)
The chief acid gases of concern from the combustion of municipal solid waste
o L
are HC1 and SO2. Hydrogen fluoride (HF), hydrogen bromide (HBr), and SO3 are also generally
present, but at much lower concentrations. Concentrations of HC1 and SO2 in MWC flue gases
directly rdate to the chlorine and sulfur content in the waste. The chlorine and sulfur content vary
considerably based on seasonal and local waste variations. Emissions of SO2 from MWCs depend on
the chemical form of sulfur in the waste, the availability of alkali materials in combustion-generated
' J o
fly ash that act as sorbents, and the type of emission control system used. SO2 concentrations are
considered to be independent of combustion conditions. Sulfur is contained in many constituents of
MSW, such as asphalt shingles, gypsum wallboard, and tires. Emissions from these sources are
classified under SCC SOlOOlxx (EPA, 1997).
a. Description of Available Control
Options
Commercialized FGD processes use an alkaline reagent to absorb SO2 in the
flue gas and produce a sodium or a calcium sulfate compound. These solid sulfate compounds are
then removed in downstream equipment. FGD technologies are categorized as wet, semi-dry, or
69
-------
dry depending on the state of the reagent as it leaves the absorber vessel. These processes are either
regenerable (such that the reagent material can be treated and reused) or nonregenerable (in which
case all waste streams are de-watered and discarded) (EPA, 1997).
To date, spray dryers are most the common FGD systems applied to MWC
in the United States. Both spray dryers and dry sorbent injection systems are generally used with
fabric filters to remove spent sorbent as well as for control of other pollutants, such as particulate
matter and toxics. Lime/limestone scrubbers, sodium scrubbers, and dual alkali scrubbing are
among the commercially proven wet FGD systems, though they are most often used to control
MWC emissions in Japan and Europe. Effectiveness of these devices depends not only on control
device design but also on operating variables (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for spray dryers, dry sorbent injection, and
wet scrubbing FGD systems applied to MWC were determined from EPA documents and recent
technical literature. In a spray-dryer system, waste gas at air preheater temperatures (generally,
275°F to 400°F) is contacted with a solution or slurry of alkaline material, generally lime or sodium
carbonate, in a vessel for five to fifteen seconds residence time. The hot gas is adiabatically
humidified to within 50°F of its saturation temperature by the water evaporated from the solution
or slurry. As the slurry or solution is evaporated, liquid phase salts are precipitated and the
remaining solids are dried to generally less than one percent free moisture. These solids, along with
any ash in the waste stream, are entrained by the waste gas to a particulate collection device, either
an ESP or a fabric filter (baghouse). Reaction between the alkaline material and waste gas SO2
proceeds both during and following the drying process. The by-product is a dry mixture of sulfite,
sulfate, fly ash, and unreacted reagent. Generally, part of the by-product is recycled and mixed
with fresh reagent to enhance sorbent utilization (EPA, 1997; Smith, et al., 1994; EPA, 1981).
The typical control efficiency range for spray dryer systems is from 30 to 90
percent. Systems which operate in the higher efficiency range are those which treat waste gas
streams with a high enough concentration of SO2 to efficiently utilize reagent and recycle reagent
for maximum reagent usage (Smith, et al., 1994; Dennis, et al., 1993).
The dry sorbent injection process generally involves pneumatically
introducing a dry, powdery alkaline material, generally calcium-based (lime or limestone), although
sodium-based (nahcolite or trona) sorbents are also used, into a waste gas stream with subsequent
particulate collection. The injection point has been varied from the boiler furnace exit to the
entrance to an ESP or, more typically, a baghouse. The sorbent material reacts with the SO2 in the
gas stream and is collected on filter bags. Filter bags are more popular than ESPs since there is
evidence that the removal reaction between SO2 and the sorbent material takes place, in large part,
on the bag surface. The reacted sorbent is disposed of after periodic baghouse cleanings. The
typical control efficiency range for Dry Sorbent Duct Injection is from 50 to 90percent (EPA,
1997; Smith, etal., 1994; Soud, 1993; Sondreal, 1993; Torrens, 1990; EPA, 1981).
Wet FGD systems are a general category of control device in which, for SO2
control, a liquid solution or liquid/solid slurry is used to absorb, and, in most cases, react with SO2
in a waste gas stream. A scrubbing vessel, into which both the solution or slurry and the waste gas
o o ' J o
are introduced, is used to maximize the contact between the SO, in the waste gas and the reacting
' z o o
compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
o o o
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
' ' L J o J
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
70
-------
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO2 from the flue gas. Sodium scrubbers are
generally limited to smaller sources because of high reagent costs and can have SO2 removal
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for spray drying systems generally fall within the range of $140 to $200
(1990 dollars) per kilowatt. Operating costs for spray drying systems are between $300 and $550
(1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-year equipment
lifetime without the effects of inflation (Soud, et al., 1993, Smith, et al., 1994; Radcliffe, 1992;
Torrens, 1990).
Capital costs for sorbent injection systems generally fall within the range of
$80 to $120 (1990 dollars) per kilowatt. Operating costs for wet scrubbing systems are between
$500 and $720 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
Capital costs for wet scrubbing systems generally fall within the range of
$180 to $260 (1990 dollars) per kilowatt. Operating costs for wet scrubbing systems are between
$460 and $620 (1990 dollars) per ton of SO2 controlled. Annual costs were estimated using a 15-
year equipment lifetime without the effects ofinflation (Soud, et al., 1993, Smith, et al., 1994;
Radcliffe, 1992; Torrens, 1990).
15. Steam Generating Unit-Coal/Oil
Gaseous SOX from coal and oil combustion are primarily SO2, with a much
lower quantity of SO3 and gaseous sulfates. These compounds form as the organic sulfur in both
coal and oil, as well as pyritic sulfur in coal are oxidized during the combustion process.
Uncontrolled SOX emissions are almost entirely dependent on the sulfur content of the fuel and are
not affected by boiler size, burner design, or grade of fuel being fired. On average, about 95
J ' o ' o o o '
percent of the sulfur present in bituminous coal or oil will be emitted as gaseous SO2, whereas
somewhat less will be emitted when subbituminous coal is fired. About 1 to 5 percent of the sulfur
in oil is further oxidized to SO3, and 1 to 3 percent is emitted as sulfate particulate. SO3 readily
reacts with water vapor (both in the atmosphere and in flue gases) to form a sulfuric acid mist. The
more alkaline nature of the ash in some subbituminous coals causes some of the sulfur to react in the
furnace to form various sulfate salts that are retained in the boiler or in the flyash. Emissions from
these sources are not classified under any single identifiable SCC (EPA, 1997).
a. Description of Available Control
Options
Several techniques are used to reduce SO2 emissions from coal and oil
combustion. One way is to switch to lower sulfur coals or oils, since SO2 emissions are
proportional to the sulfur content of the fuel. This alternative may not be possible where lower
sulfur fuel is not readily available or where a different grade of fuel cannot be satisfactorily fired. In
some cases, various coal cleaning processes maybe employed to reduce the sulfur content. Physical
coal cleaning removes mineral sulfur such as pyrite but is not effective in removing organic sulfur.
71
-------
Processes are being developed to remove organic sulfur from coal using chemical cleaning and
solvent refining (EPA, 1997).
Post-combustion FGD techniques can remove SO2 formed during
combustion by using an alkaline reagent to absorb SO2 in the flue gas. Flue gases can be treated
using wet, dry, or semi-dry desulfiirization processes of either the throwaway type (in which all was
streams are discarded) or the recovery/regenerable type (in which the SO2 absorbent is regenerated
and reused). To date, wet systems are the most commonly applied (EPA, 1997).
b. Control Options Selected for Analysis
Emission control and cost data for wet FGD scrubbing systems applied to
industrial steam generation by combustion of bituminous/subbitumino us coal were determined
from EPA documents and recent technical literature. Wet FGD systems are a general category of
J o o J
control device in which, for SO2 control, a liquid solution or liquid/solid slurry is used to absorb,
and, in most cases, react with SO2 in a waste gas stream. A scrubbing vessel, into which both the
solution or slurry and the waste gas are introduced, is used to maximize the contact between the
SO2 in the waste gas and the reacting compounds in the solution or slurry (EPA, 1981).
The design of the scrubbing vessel and the manner in which the waste gas and
the solution or slurry are introduced to the vessel are the means by which the reagent contact is
controlled. Types of wet scrubber designs include: tray-type column, packed-bed column, mobile-
bed column, venturi, and spray tower. Reagents used in wet scrubber systems indude: calcium
oxide (from lime), calcium carbonate (from limestone), magnesium oxide, sodium carbonate (soda
ash), sodium hydroxide (caustic), sodium citrate, and ammonium hydroxide. Some wet scrubbing
systems use a reagent which can be treated and reused and/or produces a saleable product, while
other wet systems require that the spent reagent be treated and disposed of appropriately (EPA,
1981).
Wet FGD systems generally use alkali slurries as the SO2 absorbent medium
and can be designed to remove greater than 90 percent of the incoming SO2. Lime/limestone
scrubbers, sodium scrubbers, and dual alkali scrubbers are among the commercially proven wet
FGD systems. The effectiveness of these devices depends not only on control device design but also
on operating variables. The lime and limestone wet scrubbing processes use a slurry of calcium
oxide or limestone to absorb SO2 in a wet scrubber. Control efficiencies in excess of 91 percent for
lime and 94 percent for limestone over extended periods are possible (EPA, 1997).
Sodium scrubbing processes generally employ a wet scrubbing solution of
sodium hydroxide or sodium carbonate to absorb SO, from the flue gas. Sodium scrubbers are
J z o
generally limited to smaller sources because of high reagent costs and can have SO, removal
o J o o ^
efficiencies of up to 96.2 percent. The double or dual alkali system uses a dear sodium alkali
solution for SO2 removal followed by a regeneration step using lime or limestone to recover the
sodium alkali and produce a calcium sulfite and sulfate sludge. SO2 removal efficiencies of 90 to 96
percent are possible (EPA, 1997).
Costs for FGD systems are often expressed in dollars per energy output in
kilowatts. Capital costs for wet scrubbing systems generally fall within the range of $ 180 to $260
(1990 dollars) per kilowatt. Annualized cost effectiveness estimates for wet scrubbing systems
given by articles in scientific journals range from $340 to $630 (1990 dollars) per ton of SO2
controlled. Annual costs were estimated using a 15-year equipment lifetime without the effects of
inflation (Soud, etal., 1993, Smith, etal., 1994; Radcliffe, 1992; Torrens, 1990).
C. REFERENCES
AWMA, 1992: Air and Waste Management Association, Air Pollution Engineering Manual, Van Nostrand
Reinhold, New York, NY, 1992.
Dennis, etal., 1993: Dennis, R.A., N.W.J. Ford, and M.J. Cooke, "A Guide to Flue Gas
Desulphurization for the Industrial Plant Manager," from "Desulfiirization
72
-------
3," Institution of Chemical Engineers Symposium Series No. 131,
Warwickshire, UK, 1993.
Emmel, T.E., et al., 1986: "Cost of Controlling Directly Emitted Acidic Emissions from Major Sources,"
Radian Corporation, Research Triangle Park, NC, (EPA/600/7-88-012),
July 1986.
EPA, 1981: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Control Techniques for Sulfur Oxide Emissions from Stationary Sources,"
Second Edition, Research Triangle Park, NC, April 1981.
EPA, 1985: U.S. Environmental Protection Agency, "Sulfuric Acid: Review of New Source Performance
Standards for Sulfuric Acid Plants," Research Triangle Park, NC,
(EPA/450/3-85/012), March 1985.
EPA, 1996: U.S. Environmental Protection Agency, "Analyzing Electric Power Generation Under the
CAAA," Office of Air and Radiation, July 1996.
EPA, 1997: U.S. Environmental Protection Agency, "Compilation of Air Pollutant Emission Factors,
Volume I, Stationary Point and Area Sources," AP-42, Fifth Edition,
Research Triangle Park, NC, October 1997.
Friedman, 1981: L.J. Friedman, "Production of Liquid SO2, Sulfur and Sulfuric Acid," Sulfur Dioxide
Control in Pyrometallurgy, edited by T.D. Chatwin and N. Kikumoto,
Metallurgical Society of AIME, Warrendale, PA, February 1981.
Pechan, 1997: E.H. Pechan & Associates, Inc., "Additional Control Measure Evaluation for the Integrated
Implementation of the Ozone and Particulate Matter National Ambient Air
Quality Standards, and Regional Haze Program," prepared for the U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC, July 17, 1997.
Pechan-Avanti, 1999: The Pechan-Avanti Group, a Unit of E.H. Pechan & Associates, Inc., "Emissions
Information Acquisition and Verification - Other Point Source Costing,"
prepared for the Western Governors' Association, Denver, CO, March 3,
1999.
Radcliffe, 1992: Paul Radcliffe, "FGD Economics," EPRI Journal, Vol. 17, No. 6, September 1992.
Smith, et al., 1994: Smith, Irene M., Hermine N. Soud, and Mitsuru Takeshita, "Flue Gas Desulfurization
for Coal-Fired Power Generation in the Asia-Pacific Region: An Overview
o
of Best Available Options," from "The Clean and Efficient Use of Coal and
Lignite: Its Role in Energy, Environment and Life," Organisation for
Economic Co-Operation and Development/International Energy Agency,
Paris, France, 1994.
Sondreal, 1993: Everett A. Sondreal, "Clean Utilization of Low-Rank Coals for Low-Cost Power
Generation," from "Clean and Efficient Use of Coal: The New Era for Low-
Rank Coal," Organisation for Economic Co-Operation and Development/
International Energy Agency, Paris, France, 1993.
O-> O J ' ' '
Soud, et al., 1993: Soud, Hermine N., Mitsuru Takeshita, andlrene M. Smith, "FGD Systems and
Installations for Coal-Fired Plants," from "Desulfurization 3," Institution of
Chemical Engineers, Warwickshire, UK, 1993.
O ' ' "
Torrens, 1990: lanM. Torrens, "Environment," Vol. 32, No. 6, July/August 1990.
73
-------
Page Intentionally Blank
74
-------
CHAPTER V
STATIONARY SOURCE PM10 AND PM25
This chapter discusses control options for stationary sources of direct
particulate emissions (PM10 and PM2 s). The first section discusses options for point source emitters.
The second section discusses area source emitters.
A. POINT SOURCE PM CONTROL MEASURES
Control efficiencies and costs of systems used to control PM emissions from
point sources are based on the use of dry and wet ESPs, fabric filters, venturi scrubbers, and
impingement plate scrubbers. These add-on controls are in general use and have been for a number
of years, although improvements are continually implemented.
Control efficiency and cost data for control technologies were based on
information taken from various sources and professional judgement. Sources include primarily EPA
documents, and other agency documents.
' o J
The reference used for PM10 and PM2 s control efficiencies throughout this
section is "Evaluation of Fine Particulate Matter Control Technology" (EC/R, 1996). This
document is a compilation of data on achievable emissions reductions for major emission source
categories. This document was selected because data for "maximum potential control efficiency"
for both PM10 and PM2 s for each specific source/control combination are presented.
The primary references for cost data are the Air Pollution Technology
Factsheets posted on EPA's OAQPS, Clean Air Technology Center (CATC) website,
www.epa.gov/ttn/ catc/cica/cicaeng.html (EPA, 1999a). The factsheets present a complete
overview of each control technology; describing the applicable pollutants, achievable emission
limits/reductions, industrial applications, emission stream characteristics, cost information, theory
of operation, and advantages and disadvantages of each control device. The range of high and low
capital costs and O&M costs presented in the factsheets were calculated based on the OAQPS
Control Cost Manual and associated spreadsheets (EPA, 1996). The low costs are representative of
equipment sized based on the maximum flow rate recommended in the cost manual, with no exotic
materials. The high costs are representative of equipment sized based on the minimum flow rate
recommended in the cost manual, with no exotic materials. No optional pre- or post treatment
equipment costs are included. Unless otherwise noted, costs are based on third quarter 1995
dollars.
The cost estimates are approximate; many control technologies are highly
dependent upon specific application needs and varying inlet conditions, which may result in a wide
range of values of cost per ton of pollutant controlled.
When stack gas flow rate data was available, the costs and cost effectiveness
o '
were calculated using the typical values of normalized capital and O&M costs listed in Table V-l.
Total annualized costs were determined by adding the annualized O&M costs; fixed capital recovery
charges; and a fixed annual charge for taxes, insurance, and administrative costs. The fixed annual
o ' o ' '
charge for taxes, insurance, and administrative costs was estimated as 4 percent of the total capital
investment (EPA, 1990). Total installed capital costs were annualized using a capital recovery
factor, which is based on a 7 percent discount rate and the expected life of the control equipment.
When stack gas flow rate data was not available, default typical capital and O&M cost values based
on tons per year of PM removed were used. The typical cost effectiveness values ($ per ton of PM
removed) listed in Table V-1 were also used as default values when stack gas flow rate data was not
' O
available.
75
-------
Table V-2 lists the applicable SCCs for the various PM groups. For each of
these PM groups, PM control costs and cost effectiveness values were obtained.
1. Cement Manufacturing (Wet and Dry Process)
Portland cement plants produce portland cement powder, which is a
mixture of calcium silicates, calcium aluminate, calcium aluminoferrite, and calcium aluminate.
Portland cement is used inmost construction activities. The largest source of particulate emissions
at a cement plant is the kiln used to produce clinker. Cement kilns are rotary kilns, which are
slowly rotating refractory-lined steel cylinders inclined slightly from the horizontal. Raw materials
are fed into the top end of the kiln and spend several hours traversing the kiln. In wet process kilns
(SCC 30500706), the raw materials are fed as a wet slurry. During this time, the raw materials are
heated by a flame at the discharge end of the kiln. This heating dries the raw materials, converts
J o o '
limestone to lime, and promotes reaction between and fusion of the separate ingredients to form
clinker. Clinker exiting the kiln is fed to a dinker cooler (SCC 30500714) for cooling before
storage and further processing(STAPPA/ALAPCO, 1996).
a. Description of Available Control
Options
Particulate emissions from cement kilns are typically controlled using dry
ESPs or fabric filters. These control options are considered below. Given high inlet loadings and
required performance, both devices must be relatively large. Thus, a typical cement kiln baghouse
air-to-clothe ratio would be 1.5 feet per minute (ft/min) and a typical precipitator specific
collection area would be 350-500 ft2/1,000 acfm. Kiln gas flows are on the order of 50,000 to
300,000 acfm (STAPPA/ALAPCO, 1996). Clinker coolers may have significant uncontrolled
emissions. While most of the air flow through the clinker cooler serves as preheated air for the
o L
cement kiln, additional air used to cool the dinker further is vented to the atmosphere. Typical
vent stream flows may be 20,000 - 100,000 acfm and may have particulate loadings of 10 - 20 g/m3,
only 10 percent of which is PM10 or finer. ESPs and fabric filters provide control of clinker cooler
PM emissions to below 50 mg/m3. ESPs used for this purpose will specific collection areas of 450 -
500 ft3/acfm and baghouses will have air-to-cloth ratios of 2 ft/min if reverse air and 5 ft/min if
o
pulse jet.
b. Control Options Selected for Analysis
Control options, efficiencies (EC/R, 1996), and cost data are presented in
Table V-l.
76
-------
Table V-1
Stationary PM Controls and Cost Estimates
PM
Group
21
22
31
41
42
51
61
Source Category/Control Measure
Cement Manufacturing
Wet Process:
Portland (Constru ction) Ce ment -
Kiln; Clinker Cooler
Dry Process:
Portland (Constructbn)
Cement - Kiln;Clinker
Cooler; Ore Crushing;
GrindingScreening
Wood/Bark Waste (Industrial Bolters)
Wet scrubber
Stone Quarrying - Processing
Nonmetallic Mineral Processing -
Ore Crushing; Grinding/Screening
Calcine rs and D ryers
NSPS (Venturi scrubber
technology)
Fabric filter
High pressure drop scrubber
Taconite Iron Ore Processing
Metallb Mineral Processing
NSPS (Wet scrubbing or
baghouse technology)
Bituminous/Subbituminous Coal
(Industral Boilers)
ICI Steam Generation
Wet (venturi) scrubbers
Control
Strategy
^i
-J
2
1
1
1
2
3
1
1
Control Type
(Assumed for
Cost Estimate)
Dry ES P-Wire
Plate Type
Fabric Filter
(Mech. Shaker
Type)
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Fabric Filter
(Mech. Shaker
Type)
High pre ssure
drop scrubber
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
PM10
Control
Efficiency1
98
99.5
99.5
93
99
95
99
99.9
99
82
PM2.5
Control
Efficiency1
95
99
99
92
99
90
99
86
99
5U
Capital Costs2 ($/scfm)
Min
$15
$15
$15
$3
$15
$3
$15
$3
$15
$3
Max
$50
$150
$150
$28
$150
$28
$150
$28
$150
$28
Typical
$27
$60
$60
$11
$60
$11
$60
$11
$60
$11
O&M Costs 2($/scfm)
Min
$4
$3
$3
$4
$3
$4
$3
$4
$3
$4
Max
$40
$21
$21
$119
$21
$119
$21
$119
$21
$119
Typical
$16
$9
$9
$42
$9
$42
$9
$42
$9
$42
Annu alized C osts2 ($/scfm )
Min
$5
$5
$5
$5
$5
$5
$5
$5
$5
$5
Max
$40
$50
$50
$123
$50
$123
$50
$123
$50
$1 23
Typical
$17
$20
$20
$44
$20
$44
$20
$44
$20
$44
Cost Effectiveness2
($/ton PM Rem oved)
Min
$40
$36
$37
$76
$37
$76
$37
$76
$37
$^6
Max
$250
$340
$320
$2,100
$320
$2,100
$320
$2,100
$320
$2,1 00
Typical
$110
$137
$131
$751
$131
$751
$131
$751
$131
$^51
-------
Table V-1 (continued)
PM
Group
71
72
81
82
92
101
111
Source Category/Control Measure
NSPS (Baghouseor ESP
technology for largesources
and wet scrubbertechnology
for small sources)
Coal Mining, Cleaning, and Material
Handling
Coal Preparation Plants- Thermal
Dryers
NSPS (Centrifugal collectors
to venturi scrubber
technology)
Coal Preparation Plants -
Pneumatic Coal Cleaning
NSP S (Co Hectors to fabric
filtertechnology)
Steel Manufacturing
Iron and Steel - BasicOxygen
Process Furnace
NSPS (Closed hood /venturi
scrubber technology)
Steel Production-EAF Argon O2
Deca rb Vessels
NSPS (Hood/fabric filter
technology)
Iron Production
Metallic Mineral Processing
NSPS (Wet scrubbing or
baghouse technology)
By-product Coke Manufacturing
Fabric filter
Venturi sciubber
Wet ESP
Residual Oil (Industrial Boilers)
Wet sc rubbers
Control
Strategy
2
3
-J
1
2
1
1
1
2
1
2
3
1
Control Type
(Assumed for
Cost Estimate)
Dry ES P-Wire
Plate Type
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Fabric Filter
(Mech. Shaker
Type)
Fabric Filter
(Mech. Shaker
Type)
Venturi Scrubber
Wet ESP -Wire
Plate Type
Venturi Scrubber
PM10
Control
Efficiency1
98
99
99
99.5
73
99.9
73
99
92
93
99
92
PM25
Control
Efficiency1
95
99
98
99
25
99.9
25
99
92
89
99
89
Capital Costs2 ($/scf m )
Min
$15
$15
$3
$15
$3
$15
$3
$15
$15
$3
$30
$3
Max
$50
$150
$28
$150
$28
$150
$28
$150
$150
$28
$60
$28
Typical
$27
$60
$11
$60
$11
$60
$11
$60
$60
$11
$40
$11
O&M Costs 2($/scfm)
Min
$4
$3
$4
$3
$4
$3
$4
$3
$3
$4
$6
$4
Max
$40
$21
$119
$21
$119
$21
$119
$21
$21
$119
$45
$119
Typical
$16
$9
$42
$9
$42
$9
$42
$9
$9
$42
$19
$42
Annualized Costs2 ($/scf m )
Min
$5
$5
$5
$5
$5
$5
$5
$5
$5
$5
$10
$5
Max
$40
$50
$123
$50
$123
$50
$123
$50
$50
$123
$50
$123
Typical
$17
$20
$44
$20
$44
$20
$44
$20
$20
$44
$23
$44
Cost Effectiveness2
($/ton PM Removed)
Min
$40
$37
$76
$37
$76
$37
$76
$37
$37
$76
$55
$76
Max
$250
$320
$2,100
$320
$2,100
$320
$2,100
$320
$320
$2,100
$550
$2,100
Typical
$110
$131
$751
$131
$751
$131
$751
$131
$131
$751
$220
$751
-------
Table V-1 (continued)
PM
Group
121
122
131
141
151
161
171
Source Category/Control Measure
Fiberglass Manufacturing
Wool Fiberglass
Man ufactu ring-Rotary S pin
NSPS (Wet ESP technology)
Glass Manufacturing Plants -
Furnace
NSPS (ESP technology)
Feed and Grain Country Elevators
Fabric filters
Grey Iron Foundries
Impingement sciubber
ESP
Fabric filter
High energy sciubber
Catalytic C racking U nits
Petroleum Refineres - FCCU
NSPS (ESP technology)
Feed and Grain Terminal Elevators
Grain E levators
NSPS (Fabric filter
technology)
Glass M anufactu re
Glass Manufacturing Plants -
Furnace
NSPS (ESP technology)
Control
Strategy
1
^D
1
1
1
2
3
4
1
1
1
Control Type
(Assumed for
Cost Estimate)
Wet E SP - W ire
Plate Type
Dry ES P-Wire
Plate Type
Fabric Filter
(Mech. Shaker
Type)
Impin gem ent-pla
te
Dry ES P-Wire
Plate Type
Fabric Filter
(Mech. Shaker
Type)
Venturi Sciubber
Dry ES P-Wire
Plate Type
Fabric Filter
(Mech. Shaker
Type)
Dry ES P-Wire
Plate Type
PM10
Control
Efficiency1
99
99
99
64
90
94
94
82-99
99
99
PM25
Control
Efficiency1
95
99
99
64
90
93
94
82
99
99
Capital Costs2 ($/scf m )
Min
$30
$15
$15
$3
$15
$15
$3
$15
$15
$15
Max
$60
$50
$150
$71
$50
$150
$28
$50
$150
$50
Typical
$40
$27
$60
$26
$27
$60
$11
$27
$60
$27
O&M Costs 2($/scfm)
Min
$6
$4
$3
$3
$4
$3
$4
$4
$3
$4
Max
$45
$40
$21
$71
$40
$21
$119
$40
$21
$40
Typical
$19
$16
$9
$26
$16
$9
$42
$16
$9
$16
Annualized Costs2 ($/scf m )
Min
$10
$5
$5
$3
$5
$5
$5
$5
$5
$5
Max
$50
$40
$50
$71
$40
$50
$123
$40
$50
$40
Typical
$23
$17
$20
$26
$17
$20
$44
$17
$20
$17
Cost Effectiveness2
($/ton PM Removed)
Min
$55
$40
$37
$46
$40
$36
$76
$40
$36
$40
Max
$550
$250
$320
$1,200
$250
$340
$2,100
$250
$340
$250
Typical
$220
$110
$131
$431
$110
$137
$751
$110
$137
$110
-------
Table V-1 (continued)
PM
Group
172
181
182
191
Source Category/Control Measure
Nonmetallic Mineral Processing -
Crushing/Grinding
NSPS (Fabric filter
technology)
Plywood/Partcleboard Operations
Venee r Dryers
Wet scrubber system (spray
chamber, cycbnic coltector
and p acke d bed scrub ber in
series)
Waferboard Dryers
Wet ESP
Aspha It Concrete
Hot Mix Asphalt Plants
NSPS (Fabric filter or venturi
scrubber technology)
Control
Strategy
1
o
1
1
1
Control Type
(Assumed for
Cost Estimate)
Fabric Filter
(Pulse Jet Type)
Packed bed
scrubbe r x 1 .5
Wet E SP - W ire
Plate Type
FF - MS/RA
PM10
Control
Efficiency1
99
95
90-99
99
PM25
Control
Efficiency1
99
91
99
Capital Costs2 ($/scf m )
Min
$5
$16
$30
$15
Max
$130
$84
$60
$150
Typical
$47
$39
$40
$60
O&M Costs 2($/scfm)
Min
$4
$24
$6
$3
Max
$20
$108
$45
$21
Typical
$9
$52
$19
$9
Annualized Costs2 ($/scf m )
Min
$4
$26
$10
$5
Max
$45
$117
$50
$50
Typical
$18
$56
$23
$20
Cost Effectiveness2
($/ton PM Removed)
Min
$29
$73
$55
$36
Max
$300
$328
$550
$340
Typical
$119
$244
$220
$137
SOURCES:
1EC/R, 1996.
-------
Table V-2
Stationary Point Source PM Groups and SCCs
sec
30500706
30500714
30500606
30500614
30500609
30500611
30500613
10200901
10200902
10200903
10200904
10200905
10200906
10200907
30502001
30502004
30502012
30302301
30302302
30302303
30302304
30302306
30302311
10200201
10200202
10200203
10200204
10200205
10200206
10200212
10200213
10200217
10200219
10200221
10200222
10200223
10200224
10200225
10200226
10200229
30501001
30501002
30501003
30501004
30501005
30501006
30501007
30501013
30300914
30300928
30300812
30300818
30300820
30300821
30300823
30300827
PM Group
21
21
22
22
22
22
22
31
31
31
31
31
31
31
41
41
42
51
51
51
51
51
51
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
71
71
71
71
71
71
71
72
81
82
92
92
92
92
92
92
PM Group Name
Cement Manufacturing - Wet Process
Cement Manufacturing - Wet Process
Cement Manufacturing - Dry Process
Cement Manufacturing - Dry Process
Cement Manufacturing - Dry Process
Cement Manufacturing - Dry Process
Cement Manufacturing - Dry Process
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Wood/Bark Waste (Industrial Boilers)
Stone Quarrying -Processing
Stone Quarrying - Processing
Stone Quarrying - Processing
Taconite Iron Ore Processing
Taconite Iron Ore Processing
Taconite Iron Ore Processing
Taconite Iron Ore Processing
Taconite Iron Ore Processing
Taconite Iron Ore Processing
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Bituminous/Subbituminous Coal (Industrial Boilers)
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Coal Mining, Cleaning, and Material Handling
Steel Manufacturing
Steel Manufacturing
Iron Production
Iron Production
Iron Production
Iron Production
Iron Production
Iron Production
81
-------
Table V-2 (continued)
sec
30300842
30300302
30300303
30300304
30300305
30300306
30300307
30300308
30300309
30300310
30300311
30300312
30300313
30300314
30300315
30300316
30300401
10200401
10200404
10200405
30501204
30501205
30501201
30501202
30501203
30501207
30501211
30501212
30501213
30200603
30200604
30200605
30200606
30200607
30200608
30200609
30200610
30200611
30200699
30400301
30400302
30400303
30400304
30400310
30400315
30400320
30400325
30400331
30400340
30400350
30400351
30400352
30400353
30400360
30600201
30600202
30600204
PM Group
92
101
101
101
101
101
101
101
101
101
101
101
101
101
101
101
101
111
111
111
121
121
122
122
122
122
122
122
122
131
131
131
131
131
131
131
131
131
131
141
141
141
141
141
141
141
141
141
141
141
141
141
141
141
151
151
151
PM Group Name
Iron Production
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
By-product Coke Manufacturing
Residual Oil (Industrial Boilers)
Residual Oil (Industrial Boilers)
Residual Oil (Industrial Boilers)
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Fiberglass Manufacturing
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Feed and Grain Country Elevators
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Grey Iron Foundries
Catalytic Cracking Units
Catalytic Cracking Units
Catalytic Cracking Units
82
-------
Table V-2 (continued)
sec
PM Group PM Group Name
30600301 151 Catalytic Cracking Units
30200501 161 Feed and Grain Terminal Elevators
30200502 161 Feed and Grain Terminal Elevators
30200503 161 Feed and Grain Terminal Elevators
30200504 161 Feed and Grain Terminal Elevators
30200505 161 Feed and Grain Terminal Elevators
30200506 161 Feed and Grain Terminal Elevators
30200507 161 Feed and Grain Terminal Elevators
30200508 161 Feed and Grain Terminal Elevators
30200509 161 Feed and Grain Terminal Elevators
30200510 161 Feed and Grain Terminal Elevators
30200511 161 Feed and Grain Terminal Elevators
30200512 161 Feed and Grain Terminal Elevators
30501401 171 Glass Manufacture
30501402 171 Glass Manufacture
30501403 171 Glass Manufacture
30501404 171 Glass Manufacture
30501414 171 Glass Manufacture
30501413 172 Glass Manufacture
30700713 181 Plywood/Particleboard Operatbns
30700715 181 Plywood/Particleboard Operatbns
30700716 181 Plywood/Particleboard Operatbns
30700704 182 Plywood/Particleboard Operatbns
30500202 191 Asphalt Concrete
30500205 191 Asphalt Concrete
83
-------
2. Wood/Bark Waste (Industrial Boilers)
The burning of wood and bark waste in boilers is mostly confined to
industries where wood and bark waste is available as a byproduct. Wood and bark waste is burned
to obtain heat energy and to alleviate possible solid waste disposal problems. In boilers, the waste is
burned in the form of hogged wood, sawdust, shavings, chips, sanderdust, or wood trim. Bark is
CO ' ' O ' 1 ' '
the major type of waste burned in "power" boilers at pulp and paper mills. At lumber, furniture,
and plywood plants, either a mixture of wood and bark waste or wood waste alone is burned most
frequently (EPA, 1995).
Wood- and bark-waste fired boilers (SCC 10200901, Bark, >50K Ib steam;
10200902, Wood/Bark, >50K Ib steam; and 10200903, Wood, >50K Ib steam) are significant
sources of PM10 and PM2 s emissions.
a. Description of Available Control
Options
According to AP-42, the four most common control devices used to control
PM emissions from wood- or bark-waste fired boilers are cyclones, wet scrubbers, ESPs, and fabric
filters. Cydones use centrifugal force to mechanically separate particulates from gas streams. As
the flow stream passes through the cyclone, PM is impinged on the walls of the unit and fall into a
collection bin. Multitube cyclones (or multiclones) provide PM control for many hogged boilers.
Two multiclones are often used in series where the first unit collects the bulk of the dust and the
second unit collects smaller particles passing through the first unit. The total suspended particulate
(TSP) control efficiency of two multiclones in series ranges from 65 to 95 percent (EPA, 1995).
Scrubbers remove particulates by collecting them in liquid droplets. Venturi
scrubbers, with gas-side pressure drops exceeding 15 inches of water column, are the most widely
used type of scrubber for wood and bark fired boilers. Venturi scrubbers of this type can achieve
TSP control efficiencies of 90 percent or more (EPA, 1995).
ESPs are used when control efficiencies of 95 percent or more are required.
ESPs use electrical fields to remove particulate from boiler flue gas. An intense electric field is
maintained between high-voltage discharge electrodes. An electric discharge ionizes the gas passing
through the ESP, and these gas ions ionize fly ash or other particulates. The collector plates are
periodically mechanically cleaned to remove the collected particuktes (EPA, 1995;
STAPPA/ALAPCO, 1993). ESPs are often used downstream of a cyclone, which removes larger-
sized dust particles. ESPs can achieve TSP control efficiencies by approximately 90 percent
(STAPPA/ALAPCO, 1996).
According to AP-42, fabric filters are also used when control efficiencies of
o '
95 percent or more are required. However, fabric filters have had limited applications to wood and
bark fired boilers because of a potential fire danger (as perceived by potential users) associated with
the collection of combustible carbonaceous fly ash. To reduce this potential hazard, mechanical
collectors can be installed upstream of the fabric filter to remove large burning particles of fly ash.
Fabric filters are generally preferred for boilers firing salt-laden wood, which produces fine particles
with a high salt content (EPA, 1995).
b. Control Options Selected for Analysis
Multiple cyclones and Venturi scrubbers were modeled as a control option
for reducing PM emissions from wood- and bark-waste fired boilers.
o
Control efficiencies and estimated cost data are presented in Table V-l.
3. Stone Quarrying - Processing
Nonmetallic Mineral Processing (305020) - ore crushing, grinding, and
screening, and Calciners (SCC 305150) and Dryers (SCC 30502012) are considered in this
84
-------
category. Materials handling operations including crushing, grinding, and screening, can produce
significant PM emissions. Drying, the heating of minerals or mineral products to remove water,
and calcination, heating to higher temperatures to remove chemically bound water and other
compounds, are normally performed in dedicated, closed units. Emissions from these units will be
through process vents, to which PM controls can be applied relatively simply. Fugitive dust
emissions may come from paved and unpaved roads in plants and from raw material and product
loading, unloading, and storage (STAPPA/ALAPCO, 1996).
a. Description of Available Control
Options
In 1980, EPA published a background information document (BID) for
proposed NSPS for nonmetallic mineral processing plants (EPA, 1980). According to the BID,
fabric filtration is the predominant control technique used to control PM emissions from mineral
ore crushing, grinding, and screening operations. Other control techniques used by controlled
emission sources include gravity collectors, cyclones, electrostatic precipitators, wet and venturi
scrubbers, water spray, process enclosure, and many different combinations of these control
techniques.
b. Control Options Selected for Analysis
Fabric filters, high pressure drop scrubbers, Venturi scrubbers,
cyclone/fabric filter combinations, and multicyclone/wet scrubber combinations control
technologies were selected as the basis for estimating the emission reductions and costs of
controlling PM10 and PM2 s emissions from calciners and dryers.
Fabric filters were selected as the basis for estimating the emission
o
reductions and costs of controlling PM10 and PM2 s emissions from sources of crushing, grinding, and
screening operations. Control efficiencies and estimated cost data are presented in Table V-l.
4. Taconite Iron Ore Processing
Taconite iron ore processing involves the production of fluxed iron ore
pellets from the initial ground excavation to the final stage of shipping. The process is composed of
six interrelated operations: mining of the ore, ore crushing, ore concentrating, fluxstone
processing, pelletizing, and shipping. Crushing, concentrating, fluxstone processing, and pelletizing
generate fugitive PM emissions which are collected by local hoods and ducted to a control device.
a. Description of Available Control
Options
Two types of control equipment are used for the control of fugitive PMthat
can be collected with hoods: the fabric filters and wet scrubbers.
b. Control Options Selected for Analysis
The fabric filter is selected for analysis. It provides a high level of control
than wet scrubbing. Control efficiencies and estimated cost data are presented in Table V-1.
5. Bituminous/Subbituminous Coal (Industrial Boilers)
There are two major coal combustion techniques in industrial boilers -
suspension firing and grate firing. Suspension firing is the primary combustion mechanism in
pulverized-coal-firedand cyclone-fired units and overfeed stoker-fired units. Both mechanisms are
employed in spreader stokers. Pulverized-coal and cyclone furnaces are used primarily in utility and
large industrial boilers.
o
85
-------
Stokers constitute the most practical method of firing coal for small
industrial units. In spreader stokers, a flipping mechanism throws the coal into the furnace and onto
a moving fuel bed. Combustion occurs partly in suspension and partly on the grate. In overfeed
stokers, coal is fed onto a traveling bed or vibrating grate, and it burns on the fuel bed as it
progresses through the furnace (AWMA, 1992).
Particulate composition and emission levels are a complex function of firing
configuration, boiler operation, and coal properties. In pulverized-coal systems, combustion is
almost complete, and thus the particulate is largely composed of the inorganic ash residue. Because
of a mixture of fine and coarse coal particles is fired in spreader stokers, significant unburned carbon
can be present in the particulate. To improve efficiency, fly ash from collection devices (typically
multiple cyclones) is sometimes reinjected into spreader stoker furnaces, which can dramatically
increase PM emissions. Uncontrolled overfeed and underfeed stokers emit considerably less
particulate than do pulverized-coal units and spreader stokers, since combustion takes place in a
relatively quiescent fuel bed. Variables other than firing configuration and fly-ash reinjection can
also affect emissions from stokers. Particulate loadings will often increase as load increases and with
o
sudden load changes. PM can increase as the ash and fines contents increase (AWMA, 1992).
a. Description of Available Control
Options
The primary types of PM control used for coal combustion are multiple
cyclones, ESPs, fabric filters, and venturi scrubbers.
b. Control Options Selected for Analysis
Venturi scrubbers (small sources) and fabric filters and dry ESPs (large
sources) are selected for analysis. Control efficiencies and estimated cost data are presented in
Table V-l.
6. Coal Mining, Cleaning, and Material Handling
Coal mining, cleaning and material handling (305010) consists of the
preparation and handling of coal to upgrade its value. For the purpose of this study, thermal dryers,
pneumatic coal cleaning and truck/vehicle travel are the sources considered. Thermal dryers are
used at the end of the series of cleaning operations to remove moisture from coal, thereby reducing
freezing problems and weight, and increasing the heating value. The major portion of water is
removed by the use of screens, thickeners, and cyclones. The coal is then dried in a thermal dryer.
Particulate emissions result from the entrainment of fine coal particles during the thermal drying
process (EPA, 1995). Pneumatic coal-cleaning equipment classifies bituminous coal by size or
separates bituminous coal from refuse by application of air streams. Fugitive PM emissions result
when haul trucks or other vehicles travel on unpaved roads or surfaces.
a. Description of Available Control
Options
Particulate emissions from dryers and cleaning are typically controlled using
wet suppression, or process enclosure venting to a control device (e.g., venturi scrubber or fabric
filter). Potential control options for truck/vehicle travel on unpaved roads include water
suppression, speed reduction, chemical treatment, and asphalt paving.
b. Control Options Selected for Analysis
A combination of a centrifugal collectors in series with venturi scrubbers is
selected for thermal dryers. A combination of centrifugal collectors in series with fabric filters is
selected for pneumatic coal cleaning. Asphalt paving is selected for truck/vehicle travel. Control
efficiencies and estimated cost data are presented in Table V-1.
86
-------
7. Steel Manufacturing
Several processes within this industry were selected for control, Basic
Oxygen Process Furnace (SCC 30300914) and EAF Argon O2 Decarb Vessels (SCC 30300928).
Steel normally is produced in either basic oxygen process furnaces or electric arc furnaces. In the
basic oxygen process furnace, a mixture of 70 percent molten iron from the blast furnace and 30
percent iron scrap are melted together. Pure oxygen is blown across the top or through the molten
steel to oxidize carbon and oxygen impurities, thus removing these from the steel. Basic oxygen
process furnaces are large open-mouthed furnaces that can be tilted to accept a charge or to tap the
molten steel to a charging ladle for transfer to an ingot mold or continuous caster.
Because basic oxygen furnaces are open, they produce significant
uncontrolled particulate emissions, notably during the refining stage when oxygen is being blown.
Emissions from these furnaces may be controlled by over 99 percent using either open or closed
hoods vented to scrubbers or electrostatic precipitators. Furnace enclosures or local hoods collect
emissions during tapping and other operations. The emissions are vented to baghouses which
provide greater than 99 percent emissions reductions.
Electric arc furnaces use the current passing between carbon electrodes to
L O
heat molten steel, but also use oxy-fuel burners to accelerate the initial melting process. These
furnaces are charged largely with scrap iron. Significant emissions occur during charging, when the
furnace roof is open, during melting, as the electrodes are lowered into the scrap and the arc is
L ' O O' L
struck, and during tapping, when alloying elements are added to the melt. Direct evacuation of the
gases above the molten steel through a water-cooled duct provides good control of particulate
emissions, from an uncontrolled level of 50 pounds per ton to a baghouse-controled level of 0.043
pounds per ton for carbon steel. Particulate generated during charging and tapping normally is
captured with a canopy hood.
a. Description of Available Control
Options
Control options include fabric filters, wet scrubbers, gravity collectors, and
electrostatic precipitators to reduce emissions.
b. Control Options Selected for Analysis
Closed hood/venturi scrubber technology is selected for controlling
emissions from the Basic Oxygen Furnace and hood/fabric filter technology is selected for
->o cv
controlling the EAF. Control efficiencies and estimated cost data are presented in Table V-l.
8. Iron Production
The sources of interest in this category are Metallic Mineral Processing (SCC
303008) and Mineral Product and Primary metal Operations - Truck/Vehicle Travel. PM
emissions are generated when haul trucks or other vehicles travel on unpaved roads or surfaces.
O L
a. Description of Available Control
Options
Potential control options for the mineral processing category includes wet
scrubbing and fabric filters. Potential control options for the truck/vehicle travel includes water
suppression, speed reduction, chemical treatment and paving.
b. Control Options Selected for Analysis
The control options analyzed for the mineral processing category are wet
scrubbing and fabric filters. The control option for the truck/vehicle travel source is paving.
Control efficiencies and estimated cost data are presented in Table V-l.
87
-------
9. By-product Coke Manufacturing
By product coke production is used to manufacture metallurgical coke by
J L L o J
heating high-grade bituminous coal (low sulfur and low ash) in an enclosed oven chamber without
oxygen. The resulting solid material consists of elemental carbon and any minerals (ash) that were
present in the coal blend that did not volatilize during the process. Sources of air emissions consist
of coke oven doors, coke oven lids and off-takes, coke oven charging, coke oven pushing, coke oven
underfire stack, coke quenching, battery venting, and coke by-product-recovery plants.
a. Description of Available Control
Options
Potential control options for byproduct coke manufacturing include proper
equipment and sealing designs, work practices, maintenance procedures, mobile systems (such as
mobile scrubber cars and mobile fume suppression systems), devices connected to land-based
pollution controls, such as fabric filters, wet scrubbers, and wet ESPs. Potential control options for
the truck/vehicle travel includes water suppression, speed reduction, chemical treatment and
paving.
b. Control Options Selected for Analysis
Control options selected for this category include fabric filters, mobile
scrubber cars, stage or sequential charge, venturi scrubbers, and wet ESPs. Control efficiencies and
estimated cost data are presented in Table V-1.
10. Residual Oil (Industrial Boilers)
Heavier fuel oil derived from crude petroleum are referred to as residual oils
and are graded from No. 4 (very light residual) to No. 6 (residual). Emissions from fuel oil
combustion depend on the grade anc composition of the oil, the type and size of the boiler, firing
practices used, and the level of equipment maintenance.
a. Description of Available Control
Options
Potential PM control options for industrial boilers using residual oil include
wet scrubbers and ESPs.
b. Control Options Selected for Analysis
The control option selected for this category is wet scrubbing. Control
efficiencies and estimated cost data are presented in Table V-l.
11. Fiberglass Manufacturing
The production of fiberglass consists of two different forms of product,
continuous-filament fiberglass or textile products and fiberglass blown wool or insulation products.
The two fiberglass sources considered here are wool fiberglass manufacturing - rotary spin and glass
O O O J L O
manufacturing plants - furnace.
a. Description of Available Control
Options
Potential PM control options consist of wet scrubbers (packed tower,
venturi), and wet and dry ESPs.
b. Control Options Selected for Analysis
-------
The control options selected for this category are wet ESPs for the wool
fiberglass manufacturing - rotary spin and dry ESPs for glass manufacturing plants - furnace.
Control efficiencies and estimated cost data are presented in Table V-1.
12. Feed and Grain Terminal and Country Elevators
Grain elevators act as intermediate storage and transfer points for grain as it
is transported from the farms to the market. The receiving, handling and transfer, drying, and
loading and shipping of grain at elevators generate grain dust which contributes to PM emissions.
Grain elevators are classified as either country (SCC 302006), terminal (SCC 302005), or export
(SCC 302031). Country elevators accept most of their grain from farms, while terminal elevators
accept most of their grain from country elevators and a small portion from farms. Country
elevators sometimes clean or dry grain before it is transported to terminal elevators or processors.
Terminal elevators dry, clean, blend, and store grain before shipment to other terminals or
processors, or for export. Terminal elevators ship their grain to export terminals or directly to
processors. Export elevators are terminal elevators that load grain primarily onto ships for export.
Country elevators account for 67 percent of total uncontrolled PM10 and PM2 s grain elevator
emissions in the 1990 NPI. Terminal and export elevators account for 27 and 6 percent of the
uncontrolled gain elevator emissions, respectively.
a. Description of Available Control
Options
Traditional PM emission controls for grain elevators include cyclones, fabric
filters, and dust recirculation. Capture systems are used to capture and convey emissions to
cyclones and fabric filters. The high cost of these controls, especially for the smaller country
elevators, has generated interest in the use of oil suppression as a less costly method for controlling
dust at grain elevators. In 1994, EPA's Control Technology Center (CTC) published a report on
the use of oil suppression as an alternative to traditional controls for controlling PM emissions at
grain elevators (EPA, 1994). According to the CTC report, oil suppression has been used
worldwide for more than a decade to control dust at small country elevators as well as krge export
terminals. Use of oil to control dust has been approved by the Food and Drug Administration for
wheat, corn, soybeans, barley, rice, rye, oats, and sorghum (21 CFR 172.878).
b. Control Options Selected for Analysis
Oil suppression can provide 75 to 99 percent control of TSP emissions.
While the oil suppression system is favored because of costs, for the purpose of this study, fabric
filters are being considered because they can achieve greater than 99 percent control of TSP as well
as small and light particles. Control efficiencies and estimated cost data are presented in Table V-l.
13. Grey Iron Foundries
Grey iron is an alloy of iron, carbon, and silicon, containing a higher
percentage of the last two elements than found in malleable iron. The high strengths are obtained
by the proper adjustment of the carbon and silicon contents or by alloying.
a. Description of Available Control
Options
Two primary collection methods are used for foundry particulates - wet and
dry. Wet scrubbers include low- and high energy types. Dry collection includes fabric filters,
mechanical collectors, and ESPs.
b. Control Options Selected for Analysis
89
-------
The control options selected for this category are dry ESPs, fabric filters,
venturi scrubbers, and impingement plate scrubbers. Control efficiencies and estimated cost data
are presented in Table V-l.
14. Catalytic Cracking Units
FCCU contribute over 85 percent of refinery primary particulate emissions.
These units are used to convert heavy oils into high-octane gasoline. The fluid catalytic cracking
process uses a catalyst in the form of very fine particles that act as a fluid when charged with a
vapor. There are two sources of air emissions from catalytic cracking: Process heater and catalyst
regenerator.
o
a. Description of Available Control
Options
ESPs are in common use at refineries for the treatment of FCCU regenerator
off-gas. In atypical application, the off-gas is routed through a CO boiler for recovery of the heat
value of the gas and then sent to an ESP. Scrubbers are used to separate and purify the fluid catalytic
cracking off-gas stream containing high concentrations of VOC, SO2, and PM. A common system
on fluid catalytic cracking regenerators is a caustic scrubber, where PM removal and SOX absorption
takes place in a venturi scrubber. PM is removed by inertial impaction of the scrubbing liquid with
the entrained particles (AWMA, 1992).
b. Control Options Selected for Analysis
The control options selected for this category are dry ESPs, fabric filters,
venturi scrubbers, and impingement plate scrubbers. Control efficiencies and estimated cost data
are presented in Table V-1.
15. Glass Manufacture
The three major components of the manufacture of glass are rawmaterial
blending and transport, melting, and forming and finishing. The sources of interest in the present
study are crushing and grinding of the raw materials, which generate fugitive dust, and the furnace
associated with the melting process.
a. Description of Available Control
Options
Pulse jet fabric filters are used for collection of PM emitted at weigh-
blending stations (AWMA, 1992). Many types of controls are used for controlling PM and acid gas
emissions from glass furnaces: wet scrubbing - venturi and packed bed, fabric filters, ESPs, electric
o or'''
boost, dry scrubbing, semi-dry scrubbing.
b. Control Options Selected for Analysis
The control options selected for this category are pulse jet fabric filters for
crushing and grinding and dry ESPs for furnace emissions. Control efficiencies and estimated cost
data are presented in Table V-1.
16. Plywood/Particleboard Operations
Plywood is a building material consisting of thin veneers (thin wood layers or
piles) bonded with an adhesive. The manufacture consists of seven main process; however, the
present source of interest is veneer dryers. The termparticleboard also is used generically to
include waferboard. The source of interest for this product is waferboard dryers.
90
-------
a. Description of Available Control
Options
PM from plywood veneer dryers includes both filterable and condensible
PM. Control options include multiple spray chambers, packed tower combined with a cyclonic
collector, sand filter scrubbers, ionizing wet scrubbers, electrified filter beds, and wet ESPs.
' ' O ' '
Methods of controlling PM emissions from the particle dryer include multi-clones, packed bed
absorbers, fabric filters, electrified filter beds, wet ESPs, and incinerators (EPA, 1999b).
b. Control Options Selected for Analysis
At present, multiple spray chambers are the most common control
technology used on veneer dryers, however, because they provide only limited removal of PM and
CV J ' ' J 1 J
condensible organic emissions, they are being replaced with newer, more effective techniques. The
packed tower combined with a cyclonic collector is selected for analysis for this source. For the
particle dryer, the wet ESP is selected. Control efficiencies and estimated cost data are presented in
Table V-l.
17. Asphalt Concrete
Hot mix asphalt (HMA) paving material is a scientifically proportioned
mixture of graded aggregates and asphalt cement. The process of producing involves drying and
heating the aggregates to prepare them for the asphalt cement coating.
a. Description of Available Control
Options
Two types of control equipment are used for the control of PM from the
HMA mixing process: the fabric filter and the venturi scrubber.
b. Control Options Selected for Analysis
The fabric filter is selected for analysis. It is more commonly used, and
provides a high level of control with less maintenance required than the venturi scrubber (AWMA,
1992). Control efficiencies and estimated cost data are presented in Table V-l.
B. AREA SOURCE PM CONTROL MEASURES
Control efficiencies and costs of systems used to control PM emissions from
area sources usually require a fundamental change in the source or the equipment used, or how the
source is operated, rather than simply applying an add-on control.
Control efficiency and cost data for control methodologies were based on
information taken from various sources and professional judgement. Sources include EPA
documents, other agency documents, and contacts with vendors.
' O J '
The reference used for PM,n and PM,, control efficiencies throughout this
1U Z.i J^
section is "Evaluation of Fine Particulate Matter Control Technology" (EC/R, 1996), unless
otherwise noted. This document is a compilation of data on achievable emissions reductions for
major emission source categories. This document was selected because data for "maximum
potential control efficiency" for both PM10 and PM2 s for each specific source/control combination
are presented.
The cost estimates are approximate; many control technologies are highly
dependent upon specific application needs and varying inlet conditions, which may result in a wide
range of values of cost per ton of pollutant controlled. Control efficiencies, and costs as available
for each group are presented in Table V-3. Due to the wide variability and uncertainty of emission
rates from these sources, no meaningful cost effectiveness data are available.
91
-------
Thus, this section generally provides measures which may provide
reductions in addition to those achieved by measures which have already been assessed (Pechan,
1997). No new control measure parameters have been developed as a result of this investigation.
Additional work is needed to determine the overall reduction for an SCC and the cost per ton of
emissions reduced.
1. Fugitive Dust - Unpaved Rural Roads
Unpaved roads comprise a sizable percentage of total PM10/PM2 s emissions.
Unpaved roads, especially rural roads, do not generally experience the type of traffic volume
associated with paved roads.
a. Description of Available Control
Options
Many control options exist for reducing PM emissions from unpaved rural
roads. These include (1) controlling the source, e.g., through speed reduction; (2) treating the
surface, such as through chemical stabilization or road oiling; and (3) improving the surface by
adding gravel, or paving. Hot asphalt paving is the control capable of achieving the highest PM
reduction from unpaved rural roads. Hot asphalt paving is based on paving materials which meet
RACT requirements and thereby do not emit VOCs. Hot asphalt paving is a viable control option
for urban areas; however, for rural roads, this control technique does have a high cost relative to the
emission reduction potential, due to the fact that there is typically less traffic.
b. Control Options Selected for Analysis
Hot asphalt paving was modeled as the control option for reducing PM
emissions. For purposes of control cost modeling, a cost per vehicle miles traveled (VMT) is
required. Capital costs were estimated by assuming $90,000 per mile of paving (Rosenberger,
1995). Assuming a 40 year life and a 7 percent discount rate, the capital recovery factor for asphalt
paving is 0.075. On an annualized basis, this is equivalent to 0.075 x $90,000/mile =
$6,750/mile. Operation and maintenance costs were estimated by
92
-------
Table V-3
Stationary Area Source PM Controls
Source Category/Control Measure
Fugitive Dust-Unpaved Roads ^
Rural Roads
Asphalt paving
Agricultural Production - Crops
Tilling
Soil conservation plans
Fugitive Dust - Construction, All Processes
Wind Erosion
Wind barrier fences
Fugitive Dust - Paved Roads
Vacuum/sweeping (mitigative control strategy)
Treat unpaved access roads (preventive control strategy)
Fuel Combustion - Resdential Wood
Alternative fuel use: natural gas
Other Combustion - Prescribed Burning
Waste Disposal & Recycling, Open Burning, Residential
Bans
sec
2296000000
2801 000003
2294000000
21040080000
2810015000
2810015000
Control Type
(Assumed for
Cost Estimate)
Asphalt paving
See Reference2
Wind barrier
fences
Closed loop
regenerative air
system
Watering and
chemical dust
suppression of
unpaved access
roads
Replace wood
stove with
natural gas
fireplace
Convert wood
fireplace to
natural gas
fireplace
See Reference2
PM1D
Control
Efficiency1
99
90
80
95
100
100
99
PM,,
Control
Efficiency1
99
90
80
95
100
100
99
Capital Cost
Min
$120,000
n/a
$1 ,400
$300
n/a
Max
$190,000
n/a
$3,250
$1 ,050
n/a
Typical
$90,000/
mile
$10 per
running
foot of
fence
$155,000
n/a
$2,325
$675
n/a
O&M Cost
Min
n/a
-$100
-$100
n/a
Max
n/a
$200
$200
n/a
Typical
$4,200/
mile
none
n/a
-$100
-$100
n/a
Annualized Cost
Min
$25,000
n/a
$37
-$74
n/a
Max
$32,000
n/a
$547
285
n/a
Typical
$0.37/
VMT
$10.70
per
running
foot of
fence
$28,500
n/a
$292
$106
n/a
Cost Effec. ($/ton)
Min
Max
Typical
2Pechan,1997.
-------
assuming that resurfacing is required for rural roads after 17 years, at a cost of $40,800/mile
(Rosenberger, 1995). Capital recovery, based on a 7 percent discount rate and 17 year life, is
0. 102. On an annualized basis, this is equivalent to 0. 102 x $40,800/mile = $4,160/mile,
assuming no additional incremental maintenance costs. Thus, the total annualized cost is $6,750 +
$4,160 = $10,910 per mileper year. Assuming 29,200 vehicles per year as typicalfor rural roads,
the annualized cost becomes $10,910 per mile per year/29,200 vehicles per year = $0.37/VMT.
2. Construction Activities
Construction activities are significant sources of PM emissions. In 1990,
construction activity emissions comprised approximately 1 6 percent of total PM2 s emissions and
20 percent of total PM10 emissions (Pechan, 1997). Construction activities that produce PM
include:
Land clearing;
o'
Demolition;
Blasting;
Ground excavations;
Cut and fill operations;
Vehicle traffic;
Welding operations; and
Wind erosion.
Except for vehicle traffic and welding operations, which are based on VMT and tons processed,
respectively, construction emissions are estimated based on the total number of acres of
construction activity. It is important to note that emission factors are not available for some
construction activities and that construction emissions are not avaikble for the individual
construction activities listed above (Pechan, 1997).
a. Description of Available Control
Options
The following control techniques have been identified as potential
construction activity PM control measures:
Water treatment of disturbed soil;
Vacuum street sweeping of nearby paved areas;
Wind barrier fencing;
o'
Chemical stabilization;
Controls such as aprons and truck washing for reducing mud/dirt
carry-out onto paved roads;
Early paving of construction roads;
Speed and/or vehicle weight limits for unpaved roads at construction
sites; and
• Prohibiting grading on windy days.
o o o J J
b. Control Options Selected for Analysis
Wind barrier fencing was selected for the present analysis. The other
potential measures identified were analyzed in Pechan (1997).
The principle behind wind fencing is to provide an area of reduced wind
elocity which provides for both a reduction in the wind erosion potential of the exposed surface,
and allows for the gravitational settling of the larger particles already airborne (EPA, 1 986) . A
survey of studies indicated total particulate control efficiencies ranging fromO to 90 percent (EPA,
1986). This document also states that problems have been noted with the sampling method used in
each of these studies, thereby limiting the degree of confidence in the degree of control achievable
by wind fences for large open sources. The EPA notes that no data directly applicable to
construction/demolition activity wind fences are available (EPA, 1988). Guidance provided by
94
v
-------
EPA does not suggest a control efficiency for wind fencing, althoughit does note that a laboratory
wind tunnel study indicated a control efficiency of 60 to 80 percent for materials handling emissions
(EPA, 1988). EC/R (1996) reports a 90 percent control efficiency for PM10 and PM2 s and rates it
as the most effective control for construction activities.
The height of a wind fence is determined by the size and type of source being
controlled, level of control needed, and surrounding terrain. For construction applications without
elevated sources, a 6 foot high fence of porous polyester material is recommended (Raring, 1999).
Capital cost is approximately $10 per running foot. O&M costs are negligible for the life of the
fence. Fence life depends on the amount of exposure to ultraviolet (UV) light. Under worst case
desert conditions, fabric replacement is necessary after 8-10 years. Under average UV conditions,
fence life is approximately 20 years. However, for construction applications, the required
performance time is usually less than the potential life of the fence, hence assuming a one-year life
and 7 percent discount rate, the annualized cost is $10.70 per running foot. It is conceivable that
the fence fabric could be reused at another location, however.
3. Fugitive Dust - Paved Roads
Particulate emissions occur whenever vehicles travel over a paved surface,
such as a road or parking lot. In general terms, particulate emissions from paved roads originate
from the loose material present on the surface. In turn, that surface loading, as it is moved or
removed, is continuously replenished by other sources. At industrial sites, surface loading is
replenished by spillage of material and trackout from unpaved roads and staging areas. Various field
studies have found that public streets and highways, as well as roadways at industrial facilities, can
be major sources of the atmospheric particulate matter within an area.
Dust emissions from paved roads have been found to vary with what is
termed the "silt loading" present on the road surface as well as the average weight of vehicles
traveling the road. The term silt loading refers to the mass of silt-size material (equal to or less than
O O ^ 1
75 micrometers [|-lm] in physical diameter) per unit area of the travel surface. The total road
surface dust loading is that of loose material that can be collected by broom sweeping and
vacuuming of the traveled portion of the paved road.
a. Description of Available Control
Options
There are two types of control scenario options: mitigative and preventive.
The most practical method used for controlling PM emissions from paved roads using mitigative
measures is some type of surface cleaning. Surface cleaning control options include vacuum
sweeping, water flushing, and water flushing followed by sweeping. Certain advances in vacuum
sweeping have resulted in closed-loop regenerative air systems that are more efficient than most
other designs.
O
Preventive options include road design and controlling practices near the
paved road. Specific preventive options are: watering and chemical dust suppression of unpaved
access roads leading to paved roads, limiting carryover from unpaved areas to paved roads, and
truck covers to minimize deposition of materials being hauled on paved roads. Preventive options
would also include methods to reduce VMT, such as through promoting commute options.
b. Control Options Selected for Analysis
The closed-loop regenerative air vacuum system was selected for analysis for
the mitigative option. These systems use an air jet generated by a blower and distributed by the
floating pickup head to loosen particles in the surface cracks and crevices before drawing them into
an internal hopper. A mechanical broom precedes the vacuum section. No water is used. An
internal centrifugal dust separator retains and collects the PM for proper disposal. Capital costs
vary from $150K to $190K (1999 dollars) for compressed natural gas (CNG) fueled units. Diesel-
powered units are approximately $30K less (Harrison, 1999). Unit life is approximately 5 years;
however, with thorough maintenance, life can be extended to 8 years. For best performance,
' O ' J 1 '
95
-------
operating speed is limited to 5 miles per hour. Based on a 7 percent discount rate and 8-year life,
annualized costs are $2SK to $32K. O&M costs are approximately $16 to $18 per curb mile, based
on operation with CNG, a thorough maintenance regimen, and a wage scale of approximately
$13/hr (Clapper, 1999). The SCAQMD conducted a series of controlled tests for different types of
street cleaning designs (Layborn, 1999). Manufacturers voluntarily submitted their equipment for
testing. An 80 percent collection efficiency was required to obtain certification. The closed-loop
regenerative air vacuum system satisfied this criterion. However, by prior agreement with the
manufacturers, the SCAQMD wouldnot divulge the specific collection efficiencies actually
obtained (Layborn, 1999).
Watering and chemical dust suppression of unpaved access roads leading to
paved roads is selected as the preventive control option. This is reported to have a 95 percent
control efficiency (EC/R, 1996). No cost data are available.
4. Residential Wood Combustion
Residential wood combustion (RWC) emissions include those from
traditional masonry fireplaces, freestanding fireplaces (metal zero clearance), wood stoves, and
furnaces. In many areas of the country with PM10 nonattainment designations, residential wood
combustion devices account for a large fraction of PM emissions.
o
PM emissions from residential wood combustion sources are a result of
incomplete combustion. One method of reducing PM emissions is to improve combustion
efficiency. This can be accomplished by changing the combustion design or using a catalyst that
fosters greater combustion efficiencies at lower temperatures. Other methods used to reduce or
eliminate PM emissions from residential stoves include changing the fuel — or characteristics of the
fuel consumed.
The NSPS were promulgated in 1988 for wood heaters. These regulations
required all new wood heaters to be EPA certified to meet specific PM emission limits in two
phases. For Phase I, new woodheaters manufactured on or after July 1988, or sold after July 1990,
were required to meet emission limits of 5.5 grams per hour and 8.5 grams per hour for catalytic
and noncatalytic wood heaters, respectively. Phase II of the NSPS required all wood heaters
manufactured on or after July 1990, or sold on or after July 1992, to meet stricter limits. Phase II
emission limits are 4.1 grams per hour and 7.5 grams per hour for catalytic and noncatalytic wood
heaters, respectively (EPA, 1989).
a. Description of Available Control
Options
In addition to reductions associated with the NSPS, the following control
techniques can be used to further reduce emissions from residential wood combustion:
1. Alternative fuel use: change to natural gas;
o o '
2. Ban on existing RWC units;
O '
3. Mandatory curtailment during predicted periods of high PM
concentrations;
4. Voluntary curtailment during predicted periods of high PM
concentrations;
5. Home weatherization; and
6. Replace existing RWC unit with cleaner-burning RWC unit.
Because these programs are rektively new, ample data does not exist to precisely quantify the cost
of controls in terms of dolkrs per ton of PM reduced (cost effectiveness). Emission reduction and
cost estimates for these measures were obtained from EPA guidance documents and augmented
with quantitative information from control agencies with experience using these measures.
1. Switching to natural gas requires replacing wood stoves and wood
burning fireplaces with gas burning alternatives. Natural gas is the primary fuel alternative to the
96
-------
combustion of wood as a residential heating fuel. Wood burning fireplaces or wood stoves can be
replaced with fireplace inserts that provide heat and reduce PM emissions completely (i.e., 100
percent control of PM emissions). "Gas logs" are also available for installation into a masonry
fireplace, but they provide less heat than an insert (EPA, 1989; Vendor Sources, 1999).
2. Banning the use of wood stoves and wood burning fireplaces already
installed in homes is an option. While this measure would reduce emissions of all pollutants for
RWC units by 100 percent, implementation is unlikely due to a probably negative public reaction.
No cost information was found for banning existing RWC units.
3. Mandatory curtailment programs are episodic controls designed to
reduce emissions when ambient PM concentrations approach the NAAQS. Several PM
nonattainment areas implement mandatory curtailment programs. A large component of the
mandatory curtailment program is public education and awareness, another large component is the
forecasting system.
o J
4. Voluntary curtailment programs are similar to mandatory curtailment
programs except that there are no enforcement penalties involved. The cost for enforcement
would be eliminated; however, some tracking of participation in the program would be needed.
The EPA guidance document suggests a credit of only a 10 percent reduction for a voluntary
program. The Best Available Control Measures (BACM) guidance (EPA, 1992) suggests the use of
a mandatory program and notes that many ar eas which began with a voluntary program have since
switched to a mandatory curtailment program. Because of the low reduction potential and since
many areas have shown little success with voluntary curtailment, this will not be considered for the
cost analysis.
5. Home weatherization seeks to reduce the amount of wood used for home
heating by reducing energy consumption required to heat a home. Less heat required to heat a
home translates to less wood combusted (and other fuels as well), and lower emissions. Public
education, home energy audits, and utility-funded conservation programs are effective means of
encouraging weatherization through measures such as dual-pane windows, weatherstripping of
doors, and increasing the insulation of homes. No cost information was available for home
' o
weatherization programs or the expenses incurred by the public (CAPCOA, 1989).
6. Replacing existing RWC units with cleaner-burning RWC alternatives
would result in a range of control efficiencies, depending upon the unit being replaced as well as the
new unit. The emission reduction resulting from replacing a pre-Phase I fireplace ranges from 53 to
94 percent, and from replacing a wood stove the reductions range from 45 to 90 percent. Since
replacement with natural gas insert or gas log is a more effective option, no cost analyses of these
options will be performed.
b. Control Option Selected for Analysis
The control measure chose for analysis is the conversion of fireplaces and
wood stoves to burn natural gas. Capital costs for converting wood burning fireplaces to a gas-
burning fireplace insert depend upon several factors, such as the size and type of existing fireplace
(masonry or freestanding), the distance between the fireplace and the natural gas or propane tank,
and the type of insert chosen.
Wood stoves may also be replaced with free-standing gas fireplaces. The
average price for replacement ranges between $1,200 and $2,500 (in 1999 dollars) for the new unit
and between $200 and $750 (in 1999 dollars) for installation (depending upon distance to gas line),
according to vendors. "Gas logs" range from $250 to $400 (in 1999 dollars) and cost between $50
and $650 (in 1999 dollars) (depending upon distance to gas line) to install (Vendor Sources, 1999).
The cost range for the complete system is therefore $ 1,700 to $4,300. Based on a 7percent
discount factor and 30 year life, the annualized capital cost range is $137 to $347.
Another consideration when determining the cost effectiveness of this
o
measure is the relative operating costs associate with burning wood and natural gas. Operating
97
-------
costs are dominated by fuel costs, which have been determined to be $ 172 per household heating
with natural gas, and between $0 and $286 per household heating with wood, assuming equivalent
heat inputs. These estimates are based on an average wood consumption of 1.3 cords per household
per year, 79 ft3 of solid wood per cord (no air spaces), a wood density of 30.2 lb/ft3, a wood heating
value of 8,613 British thermal units per pound (Btu/lb), and a natural gas heating value of 1,032
Btu/ft3 (DOE, 1997; EPA, 1997).
The cost for a cord of wood is assumed to be between approximately $ 130
to $220 based on vendor sources. This cost is likely to vary between urban and rural areas, the
season in which the wood is purchased, and the quality of the wood (Vendor Sources, 1995; Vendor
Sources, 1999). The costof natural gas is estimated to be $6.63 per thousand cubicfeet (DOE,
1999). Many people may receive firewood free-of-charge, and for these cases, there would be an
additional expense associated with changing to a natural gas fired unit. The availability of natural gas
may also be a limiting factor for this control strategy in some areas of the country, particularly in
more rural areas. Depending upon the price paid for firewood, the net operating costs of
converting an RWC unit to burn natural gas could range from a savings of $ 100 per year (assuming
$220/cord firewood) to a cost of $200 per year (assuming firewood is free-of-charge).
Annualized costs were calculated assuming a 7 percent discount rate and 30-
year life. The annualized cost range of capital and O&M costs for replacing wood stoves with free-
standing natural gas fireplaces is therefore $37 to $547. The annualized cost range of capital and
O&M costs for converting wood fireplaces to gas fireplaces (gas logs plus installation) is therefore -
$74 to $285. The $74 savings is due to the scenario of saving the cost of firewood.
5. Waste Disposal & Recycling, Open Burning, Residential
Residential open burning material includes yardwaste, garbage, trash,
rubbish and other forms of solid waste, including but not limited to wastes resulting form
residential, agricultural, commercial, industrial, institutional, trade and construction. Depending
on local ordinances, residents are often allowed to open burn yardwaste such as leaves, tree
branches or twigs and grass trimmings, but must maintain some sort of minimum of minimum
distance from structures and neighbors.
o
a. Description of Available Control
Options
Open burning is regulated by local ordinances which usually include limits
and restrictions on pile material to be burned, size and configuration, moisture content, time of
day, wind speed and direction. Cooking fires are usually exempt. Residents are usually prohibited
from burning garbage, trash, and rubbish. Permits may be required. Regulations serve to manage
open burning, but its emissions control effectiveness varies. The most effective control is banning
open burning.
b. Control Option Selected for Analysis
Banning open burning is the most effective method of control. However,
public acceptance for this approach is questionable, and will likely vary from place to place.
Banning will conceivably lead to other disposal alternatives, such as landfills, or as refuse derived
fuel (RDF) in a MWC or incinerator. Presumably, effective PM controls on these equipment
would result in a significant decrease in PM emissions than would have otherwise occurred from
open burning. No cost data are available.
L O
C. REFERENCES
AWMA, 1992: Air & Waste Management Association, "Air Pollution Engineering Manual," edited by A.
Buonicore and W. Davis, Van Nostrand Reinhold, NY, NY, 1992.
CAPCOA, 1989: California Air Pollution Control Officers Association, "A Proposed Suggested Control
Measure for the Control of Emissions from Residential Wood Combustion:
-------
Technical Support Document," Residential Wood Combustion Committee
and Stationary Source Division, October 1989.
Clapper, 1999: W. Clapper, Sunline Transit Services, personal communication with J. Reisman, The
Pechan-Avanti Group, August 18, 1999.
DOE, 1997: U.S. Department of Energy, "Household Energy Consumption and Expenditures 1993,"
Energy Information Administration, Washington, DC., December 1997.
CV ' O ' '
DOE, 1999: U.S. Department of Energy, "Short-Term Energy Outlook," Energy Information
Administration, Washington, DC., July 1999.
EC/R, 1996: EC/R Incorporated, Evaluation of Fine Paniculate Matter Control Technology, Final Draft, Task 2 Report Source
Category Emission Reductions with Paniculate Matter Control Technologies, prepared for
EPA's Integrated Policy and Strategies Group, Research Triangle Park, NC,
September 1996.
EPA, 1980: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Non-
Metallic Mineral Processing Plants - Background Information for Proposed Standards, Draft EIS,
EPA-450/3-80-002a, Research Triangle Park, NC, September 1980.
EPA, 1986: U.S. Environmental Protection Agency, Air and Engineering Research Laboratory, Identification,
' O J ' O O J ' J '
Assessment, and Control of Fugitive Paniculate Emissions, EPA/600/8-86/023, prepared
by Midwest Research Institute, August 1986.
J ' O
EPA, 1988: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Control of
Open Fugitive Dust Sources, EPA-450/3-88-008, Research Triangle Park, NC,
September 1988.
EPA, 1989: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Guidance Document for Residential Wood Combustion Emission Control
Measures," EPA-450/2-89-015, Research Triangle Park, NC, September
1989.
EPA, 1990: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"OAQPS Control Cost Manual," Fourth Edition, EPA 450/3-90-006,
Research Triangle Park, NC, January 1990.
EPA, 1992: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Technical Information Document for Residential Wood Combustion Best
Available Control Measures," EPA-450/2-92-002, Research Triangle Park,
NC, September 1992.
EPA, 1994: U.S. Environmental Protection Agency, Control Technology Center, Emission Standards
Division, and Air and Energy Research Laboratory, Oil Suppression of Paniculate
Matter at Grain Elevators, EPA-453/R-94-049, Research Triangle Park, NC, July
1994.
EPA, 1995: U.S. Environmental Protection Agency, "Compilation of Air Pollutant Emission Factors," AP-
42, Volume I, Fifth Edition, Research Triangle Park, NC, January 1995.
EPA, 1996: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Control Cost Manual, 5th Edition," EPA-453/B/96/001, February 1996.
EPA, 1997: U.S. Environmental Protection Agency, "Emission Inventory Improvement Program, Volume
III, Chapter 2: Residential Wood Combustion," EP A-454/R-97-004c,
Research Triangle Park, NC, September 1997.
99
-------
EPA, 1999a: U.S. Environmental Protection Agency, Officeof Air Quality Planning and Standards, Clean
Air Technology Center (CATC), Air Pollution Technology Factsheets, accessed at
EPA's CATC website: www.epa.gov/ttn/catc/cica/cicaeng.htrnl, July,
1999.
EPA, 1999b: U.S. Environmental Protection Agency, Office of Air Quality Planningand Standards,
' o -/ ' *- ./ O "
"Compilation of Air Pollutant Emission Factors, Volume I Stationary Point
and Area Sources, Fifth Edition," accessed at EPA's internet website:
www.epa.gov/ttn/chief/ap42etc.html, Research Triangle Park, NC, July 1999.
Harrison, 1999: J. Harrison, GCS Western Power, personal communication with J. Reisman, The Pechan-
Avanti Group, August 18, 1999.
Layborn, 1999: M. Layborn, SCAQMD, personal communication with J. Reisman, The Pechan-Avanti
Group, August 18, 1999.
Pechan, 1997: E.H. Pechan& Associates, Inc., Additional Control Measure Evaluationfor the Integrated Implementation of
the Ozone and Paniculate Matter National Ambient Air Quality Standards, and Regional Haze
Program, prepared for U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle Park, NC, July 17, 1997.
Raring, 1999: D. Raring, The Raring Corporation (WA), personal communication with J. Reisman, The
Pechan-Avanti Group, August 20, 1999.
Rosenberger, 1995: T. Rosenberger, County of Sacramento, Public Works Agency, personal
communication with S. Roe, E.H. Pechan & Associates, Inc., April 12,
1995.
SCAQMD, 1994: South Cost Air Quality Management District, 1994 Air Quality Management Plan, Appendix I-D:
Best Available Control Measures PM-10 SIP for the South Coast Air Basin, April 1994.
STAPPA/ALAPCO, 1996: State and Territorial Air Pollution Program Administrators - Association of
Local Air Pollution Control Officials, Controlling Paniculate Matter Under the Clean
Air Act: A Menu of Options, Washington, DC, July 1996.
Vendor Sources, 1995: Personal Communication by E.H. Pechan & Associates, Inc., with vendors at:
ACME Stove Co., Springfield, VA; Heatwave, Fairfax, VA; The Wood
Stove Store, Manassas, VA, 1995.
Vendor Sources, 1999: Personal Communication by The Pechan-Avanti Group, with vendors at: All Year,
Sacramento, CA; Muschetto's Firewood & Wood Pellets, Citrus Heights,
' ' ' o '
CA; Solar Syndicate, Sacramento, CA; and Young's Fireside Shop,
Sacramento, CA, July 30, 1999.
100
-------
CHAPTER VI
STATIONARY SOURCE CARBON MONOXIDE CONTROL
MEASURES
This chapter presents information on stationary source CO control
measures. Carbon monoxide has not received much attention since the number of nonattainment
areas is small in comparison to other pollutants such as ozone and PM. CAA controls have focused
on motor vehicles, which is the largest emitting sector for CO. Most of the control measures in this
chapter date back to before passage of the 1990 Amendments. Research, including contact with the
Clean Air Technology Center, did not yield any new information on stationary source CO controls.
O-> ' J J J
Table VI-1 presents the control efficiency and cost parameters for stationary source CO emitters.
The sections below provide background information on the source categories.
A. CARBON BLACK PRODUCTION
Carbon black (SCC 30100504) is produced by the reaction of a hydrocarbon
fuel (usually oil or gas) with a limited supply of combustion air at temperatures of 1320 to 1540°C.
Carbon black production is attributed with 15.07 percent of annual CO production. The unburned
carbon is used as a reinforcing agent in rubber compounds and as a black pigment in printing inks,
surface coatings, paper, and plastics. The two major processes used to manufacture carbon black
are the oil furnace process (~90 percent of production) and the thermal process (~10 percent of
production). CO emissions tend to be higher for small particle production, corresponding with
lower yields (EPA, 1999).
1. Description of Available Control Options
The control option available for this source is thermal incineration with
primary heat recovery, with and without the add-on treatment of a C O boiler, incinerator, or flare.
2. Control Option Selected for Analysis
Both thermal incineration and incineration with add-on controls have been
considered. Thermal incineration alone has a reported efficiency of 99.5 percent (Pechan, 1988).
The use of a flare increases CO combustion efficiency to 99.8 percent (EPA, 1999). Cost equation
parameters are given in Table VI-1.
B. IRON AND STEEL PRODUCTION
Iron is produced in blast furnaces by the reduction of iron-bearing materials
by use of hot gas. Iron oxides, coke (coal heated in an oxygen-free atmosphere to remove volatile
components), and fluxes (limestone, dolomite, or sinter) react to form molten reduced iron, CO,
and slag. The production of sinter involves the combustion of fine-sized raw materials into an
agglomerated product for use in the blast furnace. The production of 1
101
-------
8 Table VI-1
Carbon Monoxide Control Measures for Retrofit Applications
(1985 Dollars)
CO Source
Carbon Black
Carbon Black
Iron Ore Sinter Plant-Windbox
Iron Ore Sinter Plant-Windbox
Carbon Steel Electric Arc Furnace
Gray Iron Cupola
Conical Wood Burner
Municipal Incinerator
Basic Oxygen Furnace
Prebake Aluminum Cells
Aluminum Anode Baking
Maleic Anhydride
Maleic Anhydride
Coke Oven Charging
Cyclohexanol
Cyclohexanol
Ethylene Dichloride
Control Device
Incin. w/PR
Incin. w/PR & CO Boiler
Incin. w/PR
Incin. w/P&SR
Direct Shell Evacuation
Thermal Incin.
O2 Analyzer
O2 Analyzer
Open Hood System
Incin. w/PR
Incin. w/PR
Incin. w/PR
Incin. w/P&SR
Stage Charging
No Heat Recovery
Incin. w/PR
Incin. w/PR
Retrofit Capital
a
35.0
449.0
276.0
206.0
534.0
4,160.0
8,060.0
272.0
229.0
65.1
2.3
3,100.0
1,630.00
458,000.0
10,600.0
110,000.0
254.0
b
0.98
0.85
0.73
0.76
0.65
0.15
0.00
0.40
0.73
1.06
1.09
0.57
0.65
0.04
0.24
0.11
0.60
Retrofit
a
4.07
-37.20
39.60
-0.06
40.60
0.99
4,310.00
138.00
-1.51
41.20
0.62
57.10
2.93
8,650.00
68.00
334.00
1.08
O&M
b
0.94
1.06
0.82
1.32
0.74
0.91
0.00
0.42
0.99
1.10
1.10
0.93
1.22
0.30
0.64
0.49
1.00
Control Eff. (%)
99.5
99.5
90.0
90.0
90.0
90.1
50.0
50.0
95.0
99.0
99.0
98.0
98.0
99.0
98.0
98.0
98.0
Default Cost/Ton
3
-47
172
-288
248
5
9
102
-21
824
83
50
45
2,613
38
43
-
NOTES:
SOURCE:
Equations are of the form COST = a*(SIZE)Ab. Size varies by source category and is in SCC units.
Incin. w/PR is a Thermal Incineratorwith Primary Heat Recovery.
Incin. w/P&SR is a Thermal Incineratorwith Primary and Secondary Heat Recovery.
Pechan, 1988.
-------
ton of iron can produce 2.5 to 3.5 tons of blast furnace gas that contains CO and up to 100 Ibs of
dust. The byproduct gas is collected, cleaned of particulates, and used as a fuel within the steel
plant. Iron production is attributed with 12.66 percent of annual CO emissions. Coke
manufacturing plants associated with iron and steel production facilities are attributed with 4.23
percent of annual CO emissions.
Gray (cast) iron is an iron-carbon-silicon alloy used for industrial machinery
and heavy transportation equipment. Cupolas, electric arc furnaces, and electric induction furnaces
are used for production. Only cupoks are a significant source of CO, with2.08 percent of annual
CO emissions attributed to gray iron foundries.
o J
Steel is produced using several methods. The basic oxygen process consists
of injecting high-purity oxygen into a furnace containing molten iron from a blast furnace and iron
scrap. Carbon and other impurities are removed from the metal. The electric arc furnace is used to
produce carbon and alloy steels using electric current to melt and refine scrap materials. The open
hearth furnace uses gas burners to melt and refine steel using scrap and molten iron. Steel
production is attributed with 12.35 percent of annual CO emissions.
The emission sources included are represented by the following SCCs:
• 30300304
• 30300801
• 30400304
• 30400711
1. Description of Available Control Options
Blast furnaces for iron production are equipped with capture systems to
collect exhaust gas. Two types of hoods are used, either a closed or open hood. A closed hood fits
against the furnace mouth and directs all particulates and CO to a wet scrubber gas cleaner. CO is
flared at the scrubber outlet stack. An open hood design allows dilution air into the hood, allowing
combustion of the CO in the hood system. Windboxes draw combusted air from the sinter
production into a gas cleaning device. There, CO emissions can be treated with thermal
incineration, with or without primary and secondary heat recovery.
Stage charging during coke oven charging can also lead to reductions in CO
o o o o o o
emissions. Coal is charged at a reduced rate and suction on the oven is maintained during the
o o
charging. CO emissions remain within the oven collection system (Pechan, 1988).
During gray iron production, BACT for CO reduction from cupolas is the
use of afterburners or hot top burners (>0.3 second retention time at >1400°F).
During the steelmakingprocess, the large quantities of CO are controlled by
combustion at the mouth of the furnace and vented to gas cleaning devices using either open or
o o or
closed hoods. For electric arc furnaces, five emission capture systems are used: fourth hold (direct
shell) evacuation, side draft hood, combination hood, canopy hood, and iurnace endosures (EPA,
1999).
2. Control Option Selected for Analysis
The use of thermal incineration with primary heat recovery, and with the
addition of secondary heat recovery during the sinter production phase, were considered for
analysis. The control efficiency of both controls is 90 percent.
Stage charging can control 99 percent of CO emissions.
For carbon steel electric arc furnaces, a direct shell evaluation system is the
control device considered. It controls up to 90 percent of the CO emissions by aspiring air through
103
-------
an air gap and combusting the CO. An open hood system achieves 90 percent removal by adding air
to ensure complete burning of the CO in the hood (Pechan, 1988).
For basic oxygen furnaces, the control measure considered is the open hood
system. Available cost parameters are given in Table VI-1.
C. PULP AND PAPER AND WOOD PRODUCTS
Chemical wood pulping involves the extraction of cellulose from wood by
dissolving the ligninthat binds cellulose fibers together. Chemical pulping processes are kraft,
sulfite, neutral sulfite semichemical (NSSC), and soda. The kraft process accounts for over 80
percent of the chemical pulp in the United States (EPA, 1999). CO emissions from the kraft
process include the recovery furnace (where cooking chemicals and heat are recovered after the
digesting process) and lime kilns (where lime mud is regenerated into quicklime). The major cause
of CO emissions is furnace operation above rated capacity (EPA, 1999). Kraft pulping accounts for
7.72 percent of annual CO emissions. Acid sulfite pulpingis accomplished in the same manner as
kraft pulping, except that different chemicals are used in the cooking liquor. Acid sulfite pulping is
attributed with 0.19 percent of annual CO emissions.
The emission sources induded are represented by the following SCCs:
• 30700104
• 30700299
1. Description of Available Control Options
The CO emissions control considered is the proper operations of furnaces.
Data on costs and reductions were not available.
D. ALUMINUM ORE PRODUCTION
Bauxite ore, a hydrated oxide of aluminum, is refined and electrolytically
reduced to form elemental aluminum. The CO produced is treated as carbon dioxide (CO2)
because it is assumed that it will eventually be oxidized after being emitted (EPA, 1999). CO
emissions from aluminum ore production account for 6.36 percent of annual emissions. The
emissions are represented by SCC 30300199. No available controls were identified.
E. BITUMINOUS/SUBBITUMINOUS COAL COMBUSTION
Coal rank increases as the amount of fixed carbon increases and the amount
of volatile matter and moisture decreases. Bituminous coals are the largest group and are
characterized as having a lower fixed carbon and higher volatile matter than anthracite.
o o
Subbituminous coals have higher moisture and volatile matter and lower sulfur content than
o
bituminous coals. Emissions depend on the rank and composition of the fuel, the type and size of
the boiler, firing conditions, load, type of control technologies, and the level of equipment
maintenance.
The emission sources induded are represented by the following SCCs:
L J o
• 10100212
• 10200204
• 10300205
• 39000201
1. Description of Available Control Options
The rate of CO emissions depends on the fuel oxidation efficiency of the
source. If the combustion process is not controlled or a unit not properly operated or maintained,
CO emissions may increase by several orders of magnitude. Smaller boilers, heaters, and furnaces
104
-------
typically emit more CO than larger combustors because they have less high-temperature residence
time, and therefore, less time than larger combustors to achieve complete combustion. If
improperly designed or implemented, NOX reduction equipment can increase CO emissions (EPA,
1999). No controls were identified for inclusion in ControlNET.
F. WOOD/BARK WASTE COMBUSTION
Wood and bark waste combustion is generally confined to industries where
o J
it is available as a byproduct (e.g., pulp mills, lumber, furniture, or plywood industries). CO
emissions are increased when proper drying is not achieved or when secondary combustion is not
complete thereby lowering the combustion temperature. Nearly all of the fuel carbon (99 percent)
in wood waste is converted to CO2 during the combustion process (EPA, 1999). However,
industrial wood/bark waste combustion is attributed with 3.61 percent of annual CO emission and
commercial wood/bark waste combustion is attributed with an additional 1.98 percent of annual
CO emission.
The emission sources induded are represented by the following SCCs:
• 10200905
• 10300902
1. Description of Available Control Options
The use of O2 analyzers can reduce CO emissions by optimizing combustion.
2. Control Option Selected for Analysis
The efficiency of O2 analyzers is a 50 percent reduction in CO emissions
(Pechan, 1988). Cost parameters are given in Table VI-1.
G. NATURAL GAS COMBUSTION
Natural gas combustion is used mainly for industrial process steam and heat
production; for commercial space heating; and for electric power generation. The rate of CO
emissions from boilers depends on the efficiency of natural gas combustion (EPA, 1999). Internal
combustion boilers using natural gas for industrial sources are attributed with 2.98percent ofannual
CO emission and external combustion boilers are attributed with an additional 1.26 percent.
Electric generation using natural gas is attributed with 1.25 percent (external combustion boilers)
and 0.69 percent (internal combustion boilers) of annual CO emission.
The emission sources induded are represented by the following SCCs:
• 20200202
• 10200602
• 10100604
• 20100202
1. Description of Available Control Options
In a properly tuned boiler, nearly all of the fuel carbon (99 percent) in
natural gas is converted to CO2 during the combustion process. The addition of NOX control
systems (e.g., low NOX burners and flue gas recirculation) may reduce combustion efficiency,
resulting in higher CO emissions relative to uncontrolled boilers (EPA, 1999). Some induced flue
gas recirculation systems can lower NOX without increasing CO levels (Cleaver Brooks, 1999). No
control options were identified for inclusion in ControlNET.
H. CHARCOAL MANUFACTURING
105
-------
Charcoal is the solid carbon residue following the carbonization or
o
destructive distillation of carbonaceous raw materials, most often medium to dense hardwoods.
Manufacturing is accomplished in batch or continuous kilns. Emission levels are variable from plant
to plant depending on the raw materials and the kiln type. Most emissions are from the kiln exhaust
(EPA, 1999). Charcoal manufacturing is attributed with 0.64 percent of annual CO emissions. The
SCC for emissions from charcoal manufacturing is 30100699.
1. Description of Available Control Options
Continuous production of charcoal is more amenable to emission control
than batch production because emission composition and flow rate are relatively constant.
Emissions control is usually achieved with afterburners. Batch-type kilns typically do not have
emission control devices, but some may use afterburners.
2. Control Option Selected for Analysis
Afterburners for continuous multiple hearth charcoal kilns are the control
option considered. Reductions achieved by this method are estimated to be at least 80 percent.
Cost parameters were not identified.
I. MINERAL WOOL MANUFACTURING
Mineral wool is manufactured for use in insulation and other fibrous building
materials that are used for structural strength or fire resistance. Mineral wool is a fibrous glassy
substance made from minerals, e.g., natural rock and slags from iron, copper, lead, and phosphate
production (EPA, 1999). Mineral wool production is attributed with 0.45 percent of annual CO
production. The SCC for emissions from mineral wool manufacturing is 30501701.
1. Description of Available Control Options
CO emissions occur primarily from coke combustion in the furnace where
most control measures are directed toward PM control. Afterburners have been used to control
CO emissions. A gas-fired melting option, in place of the coke-fired cupola, has been developed for
mineral wool production. Efficiency and cost parameters have not been identified.
2. Control Option Selected for Analysis
The control option considered is the gas-fired production unit. Compared
to a coke-fired cupola facility operating with a pollution abating afterburner producing fiber at
approximately $71 /ton, the gas fired unit operating at a melting efficiency of 4.5 million British
thermal units (MMBtu) per ton is about $41 /ton of mineral fiber (based on a natural gas cost of
$5/Mcf) (Ridderbusch, 1990). This process eliminates CO emission as well as offering other
production benefits.
J. FLARES
Flares are incendiary devices which ensure safe combustion of waste gases
when the blowdown volume exceeds the storage capacity of the recovery subsystem. Slowdown
systems are designed to provide safe containment or release of liquids and gases that must be
vented. It is the last opportunity to treat blowdown gases before they are released to the
atmosphere. Completeness of combustion in flares is determined by flame temperature, residence
time in the combustion zone, turbulent mixing of the components to complete the oxidation
process, and available oxygen (SCAQMD, 1996). Flares are attributed withO.51 percent of annual
CO emissions. The emission sources are included under SCC 30600903.
106
-------
1. Description of Available Control Options
NSPS for flares that operate continuously or for emergency purposes is a 98
percent combustion efficiency. The BACT guidelines for the South Coast list the following controls
as "achieved in practice," or "contained in EPA approved SIP for refinery flares": groundlevel,
shrouded and steam assisted (SCAQMD, 1996).
K. REFERENCES
Cleaver Brooks, 1999: Cleaver Brooks online @ www.cleaver-brooks.com/Emissionsl .html.
EPA, 1999: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Compilation of Air Pollutant Emission Factors, Volume I Stationary Point
and Area Sources, Fifth Edition," accessed at EPA's internet website:
www.epa.gov/ttn/chief/ap42etc.html. Research Triangle Park, NC, 1999.
Pechan, 1988: E.H. Pechan& Associates, Inc., National Assessment ofVOC, CO, and NO^ Controls, Emissions, and Costs,
prepared for U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, September 1988.
Ridderbusch, 1990: G.L. Ridderbusch, "Industrial Processes/Shearing the High Cost Out of Mineral Wool
' ' o o
Production," In: Frontburner, pp.5-6, November 1990-March 1991.
SCAQMD, 1996: South Coast Air Quality Management District, 1997 Air Quality Management Plan (Draft) Appendix
IV-A, August 1996.
107
-------
Page Intentionally Blank
108
-------
CHAPTER VII
STATIONARY SOURCE AMMONIA CONTROLS
Pechan-Avanti focused efforts on developing cost modeling parameters for
livestock emissions. This source sector contributes 55 percent of the 19% update to Trends
emissions (Roe and Strait, 1998). Within the overall livestock emissions sector, cattle operations
contribute most of the NH3 emissions. These emissions are classified under SCC 2805001000 (beef
cattle feedlots) and 2805010000 (dairies). Hog operations are classified under SCC 2805015000
and poultry operations are under 2805005000.
A. DESCRIPTION OF AVAILABLE CONTROL OPTIONS
Control options that have been demonstrated for livestock include diet
optimization, chemical additives (to animal wastes), and various add-on controls (i.e., capture of
ammonia that has been release to the air). With diet optimization, the goal is to reduce unnecessary
nitrogen intake of the animal (i.e., protein), such that the amount of nitrogen lost to urine and feces
is reduced. Typically, the protein content of the feed is reduced and essential amino acids are
supplemented.
Chemical additives include urease inhibitors which prevent the urease
enzyme (produced from bacteria in the manure) from hydrolyzing urea to ammonia. Another
chemical additive, alum (aluminum sulfate), has been used to stabilize poultry litter to reduce
ammonia emissions. Acid-forming compounds, such as alum, keep the pH of the poultry litter
below 7, which inhibits ammonia volatilization (ammonia volatilization is significant at pH greater
than 7) (Moore et al., 1995).
Add-on controls include biofilters and wet scrubbers placed on the building
exhaust of confined animal operations. Also included are biocovers which are some type of cover
material placed over a storage lagoon. Both of these control types involve adsorption of ammonia
and possible subsequent oxidation (chemical or biological). Some additional, although less studied
control alternatives are mentioned in the individual livestock category sections below.
B. CONTROL OPTIONS SELECTED FOR ANALYSIS
Livestock ammonia controls and cost estimates are listed in Table VII-1.
Descriptions of the control options selected for analysis are given below for each of the livestock
categories (cattle, poultry, and hogs).
1. Cattle
Ammonia emissions from cattle operations in the NPI are estimated using a
composited emission factor of 22.9 kilograms (kg) NH3 per head per year (kg/head-yr). The
composite emission factor accounts for (1) the relative proportion of cattle and calves; and (2) the
sum of emissions occurring during confinement (stable), storage of the waste
109
-------
W
0)
"re
E
•[^
V)
HI
"55
o
o
c
CB
H J2
5 o
reO
"~.2
'c
o
E
^
u
o
(A
10 >
|j
U)
U)
01
c_
>~a
'4= o
II
In
o
o
"IB
Ifl
o ^_ ^
"D O
fl> 'o
•— >»
15 ^*
c
c
^
"r"
o
c
01
'o
LU
"o
•g
O
o
o
o
Measure
2
o
o
01
01
•%
k.
2
0
o
01
c
ra
2
a
3
o
CD
l"
Z
a
0
CD
l"
Z
x"
z
0
ro
CN
CN
«>
I-
-6
CD
CD
.g
LO
00
CN
s
o
0
o
o
CN
O
LO
0
00
CN
Additives2
8
E
CD
O
73
"CD
O
0
_j?
z
0
0_
^
I-
-6
CD
CD
.g
LO
T—
0
LO
o
0
o
o
o
LO
0
00
CN
Additives3
8
E
CD
O
"5
o
0.
CN
0
j£
Z
c
2
«>
I-
-6
CD
CD
.g
1^
CO
0
s
o
0
o
LO
CN
O
LO
0
00
CN
Additives2
8
E
CD
O
to
D>
0
m
0
03
E
CD
O
O
CL
to
0
CL
0
£
2>
^
i!
Z
0
—
*^.E
||
.E|
CD01 c
CD
"to
0
O
C/j
LU
1
0
z
-------
(as solid or slurry), spreading of the waste (in crop fields) , and grazing. Confined operations refer
to cattle confinement (in dairy barns or feedlots), waste storage, and subsequent waste spreading.
Unconfined operations refer to grazing. The portion of the emission factor associated with confined
operations (~20 kg/head-yr) is nearly an order of magnitude higher than the emission factor for
unconfined operations (~3 kg/head-yr).
Assessment of control measures applicable to ammonia emissions for cattle
operations is still in the early stages (MPCA, 1999). Work conducted to date is often done to assess
control of odors and/or emissions of other substances (hydrogen sulfide). This workhas focused on
control of emissions from confined operations. Given that confined operations contribute most of
the emissions (and that control of emissions from unconfined operations is probably not feasible),
research was performed on cost modeling data and methods for confined operations only.
Potential control measures cited by MPCA (1999) include biological and
chemical additives (to eliminate or reduce NH3/odor formation), biofilters and biocovers (to adsorb
and oxidize NH3), electrical conductivity and non-thermal plasma treatment (to treat collected
gases), aeration (of stored slurry), anaerobic digestion (of stored slurry followed by energy recovery
from the biogas), and crust formation (on earthen manure storage structures). Other than the use
of chemical controls, specifically urease inhibitors, sufficient information was not available to
develop cost modeling data. Urease inhibitors act by blocking urea hydrolysis (which produces
ammonia) by the urease enzyme (produced by bacteria in the cattle waste).
Varel et al. (1999) reported on the ability of two types of urease inhibitors in
reducing ammonia losses from cattle feedlot waste (through measur ement of nitrogen
conservation). These chemicals need to be applied (sprayed) over the feedlots on a routine basis.
Pechan-Avanti contacted the manufacturer of the more effective inhibitor, N-(n-butyl)
thiophosphoric triamide (NBPT; trade name Conserve-N®). According to the manufacturer, the
control effectiveness at cattle feedlots is 50 percent and the cost per head-day is $0.0062
($2.26/head-yr; Axe, 1999). The manufacturer also reports that field tests are ongoing at dairies
and that the product should perform the same (50 percent control), but cost slightlymore
$0.0094/head-day ($3.43/head-yr; Axe, 1999). It was not clear why the costs would be higher at
dairies.
As noted above, the inventory currently does not differentiate between
confined and unconfined operations (Pechan-Avanti has developed methods to do this; however,
they have yet to be incorporated into EPA's inventories). Emissions need to be broken out into at
least two, if not three, source categories (e.g., feedlots, dairies, unconfined/grazing). Until this is
done, it is assumed that cattle in confined and unconfined operations are the same throughout the
United States. This assumption will result in reasonable estimates of cost and reductions in areas
with high proportions of confined operations (e.g., beef producing States), since the emission
estimates are driven by confined operations (as described above). However, in areas where the
proportion of confined and unconfined operations are more even, there will be overestimates of
cost and reductions (since the emissions have been overestimated). For areas where cattle are
primarily grazing, the controls should not be applied (controls are not viable in these situations).
Pechan-Avanti developed county-level popuktion fractions of confined cattle
(dairy and beef) and cattle in unconfined operations (i.e., grazing) from the 1997 Census of
Agriculture (COA). Table VII-2 provides a sample of the data from this assessment. COA data for
"cattle fattened on grain and concentrates" was used as a surrogate for cattle at feedlots. The COA
also contains data for milk cows at the county-level. These data were used along with the total
cattle and calves inventory to obtain the fractions shown in Table VII-2. In the United States, about
9 percent of the popuktion is dairy, 28 percent feedlot cattle, and the remaining 63 percent are
presumed to be grazing cattle. The data in the table also show the variability of these data at the
State and county levels. In Nebraska, a large beef producing State, the majority of cattle are shown
to be in feedlots. In a sampling of California counties, El Dorado county is shown to have little in
the way of confined operations, while the two other counties have significant dairy operations.
Table VII-2
in
-------
Example Fractions of Confined and Unconfined Cattle Operations at Specified
Geographic Locations
Area
United States
Nebraska
Wisconsin
El Dorado Co., CA
Riverside Co., CA
San Joaquin Co., CA
Confined Operations
Dairy
0.09
0.01
0.39
0.002
0.55
0.39
Feedlot
0.28
0.72
0.07
0.006
0.003
0.03
Unconfined Operations
0.63
0.27
0.54
0.99
0.44
0.58
For some counties, the totals for feedlots were higher than the county totals.
These differences were usually slight and probably due to some survey anomaly. In these instances,
it is assumed that all cattle in that county are in confined operations. For counties without complete
data (e.g., due to confidentiality issues), the State-level population fractions should be used. In
order to derive a county-level control factor, the 50 percent control efficiency multiplied by a
county-level penetration factor. This penetration factor is derived by multiplying 0.87 (20/23;
which, as mentioned above, accounts for the portion of the emission factor for confined operations)
by the county-specific confined operations fraction (dairy plus feedlot fractions).
To estimate costs, an average per head cost between dairy cattle and feedlot
cattle would be $2.85/head-yr (from the values given above). The emission factor for cattle is
about 23 kg/head-yr (0.025 ton/head-yr). A 50 percent control efficiency yields 0.0125 ton/head-
yr reduced). Hence, the cost factor would be $2.85/0.0125 ton or $228/ton of NH3 reduced.
2. Poultry
The most thoroughly studied and documented ammonia control option for
poultry operations is chemical addition of alum to poultry litter. M ost of this work has been
conducted on broiler raising operations; however, the control technique should be applicable to
layer (egg) operations, as well (although the application of alum would be performed somewhat
differently) (Moore, 1999). As information was only available for broiler operations, it is assumed
that the emission reductions and cost effectiveness model inputs are the same. Axe (1999) also
reports that the same urease inhibitor product described above for cattle (Conserve-N®) has
applicability for poultry; however, sufficient information to develop cost modeling inputs was not
available in time for this study.
The control effectiveness for alum treatment is estimated to be 75 to 80
percent (Moore, 1999). The control effectiveness is highest during the early part of the growing
cycle (i.e., >95 percent), when the young chickens are most susceptible to health problems from
high ammonia levels. The control effectiveness drops off during the grow-out (about two months).
Alum is then reapplied to the litter before the next grow-out begins (typically, there are 5 or 6
grow-outs per year). A 75 percent control parameter is selected for modeling purposes. There is
assumed to be 100 percent penetration.
Treatment costs are estimated to be about $0.025/head (Moore, 1999).
These costs do not factor in some benefits to the grower (e.g., reduced heating/ventilation costs
due to lower ammonia levels; higher value for fertilizer due to higher nitrogen levels). Assuming
six grow-outs per year, the costs would be $0.15/head-yr. The emission factor used for all poultry
is 0.394 Ib/head-yr (1.97 x 104 ton/head-yr). Assuming a 75 percent control efficiency for alum
treatment, the emission reduction would be 1.48 x 10 4 ton/head-yr reduced. Hence, the cost
parameter would be $0.15/1.48E-04 ton reduced or $l,014/ton NH3 reduced.
3. Hogs
112
-------
Cost information for control of hog operations using Conserve-N® were
given by Axe (1999) to be $0.0010/head-day or ($0.37/head-yr). The same 50 percent control
efficiency given above for cattle and poultry is assumed for hogs (Axe, 1999). The emission factor
for hogs is 20.3 Ib/head-yr. With the SO percent control efficiency, this equates to 10.15 Ib/head-
yr reduced (5.08 x 103 ton/head-yr reduced). Therefore, the cost parameter would be
$0.37/5.08E-3 ton or $73/ton NH3 reduced. There is assumed to be 100 percent penetration;
however, the modeling parameters are probably most applicable to large hog farming operations.
Hence, it may be more reasonable to apply the control in counties with large hog raising operations
(i.e., using CO A data).
C. REFERENCES
Axe, 1999: D. Axe, IMC Agrico Feed Ingredients, personal communication with S. Roe, The Pechan-
Avanti Group, June 1999.
MPCA, 1999: Feedlot Air Quality Summary: Data Collection, Enforcement, and Program Development, Minnesota Pollution
Control Agency, March 1999.
Moore et al., 1995: Moore, Jr., P.A., T.C. Daniel, D.R. Edwards, and D.M. Miller, Journal of Environmental
Quality, Volume 24, no. 2, pp. 293-300, 1995.
Moore, 1999: P. A. Moore, Jr., University of Arkansas, personal communication with S. Roe, The Pechan-
Avanti Group, June 1999.
Roe and Strait, 1998: Roe, S.M. andR.P. Strait, "Methods for Improving National Ammonia Emission
Estimates," proceedings of the Air & Waste Management Association's
Emission Inventory Conference, December 8-10, 1998.
Varel et al., 1999: Varel, V., J.A. Nienaber, and H.C. Erectly, "Conservation of Nitrogen in Cattle
Feedlot Waste with Urease Inhibitors, "Journal of Animal Science, Volume 77, pp.
1162-1168, 1999.
113
-------
Page Intentionally Blank
114
-------
CHAPTER VIM
UTILITY SOURCE
This chapter evaluates potential NOX, PM, and SO2 utility source control
measures. Table VIII-1 presents a complete list of control measures incorporated into ControlNET,
differentiating between measures that are documented in this report and measures documented in
previous reports. This chapter identifies the newly developed control measures, as well as revisions
to measures developed in previous analyses. The general impetus for these revisions is the
availability of new information.
A. ECU SOURCE NO, CONTROL MEASURES
X
Control cost equations used for estimating the costs of applying selective
catalytic reduction (SCR), selective non-catalytic reduction (SNCR), and natural gas reburn (NGR)
controls were developed for electric utility boilers from EPA's IPM (EPA, 1998). The cost
equations used in this analysis are based on cost equations developed to scale costs to smaller or
larger boilers than the model plant. Model plants were considered to have boiler design capacities
of 200 MW, except coal-burning plants applying SNCR for which model plants had capacities of
100MW.
Several simplifying assumptions were made in developing the costing
L J o L L O O
parameters used for this analysis. A high NOX rate (> 0.5 pounds per MMBtu) and a capacity
utilization factor of 65 percent were assumed for the utility boilers, as well as a 7-percent discount
rate and 20-year lifetime of the SCR controls. A control efficiency of 80 percent was assumed for
SCR controls on both coal and oil or gas-fired utility boilers. The costing information for applying
SCR controls to utility boilers is listed in Table VIII-1.
B. ECU SOURCE SO2 CONTROL MEASURES
1. FGD Scrubbers
Control cost equations used for estimating the costs of applying FGD
scrubbers were developed for electric utility boilers. The cost equations used in this analysis are
based on cost equations developed to scale costs to smaller or larger boilers than the model plant.
Model plants were considered to have boiler design capacities of 500 MW. Several simplifying
assumptions were made in developing the costing parameters used for this analysis. A capacity
utilization factor of 65 percent were assumed, as well as a 7-percent discount rate and 15-year
lifetime of the SCR controls. A control efficiency of 90 percent was assumed for scrubbers on all
utility boiler fuel types.
115
-------
Table VIII-1
NOX Control Costs for Utility Boilers
Total Capital Cost
Fuel Type ($/kilowatt)
SNCR1
(Coal - Cyclone)
SNCR2
(Coal - All others)
SNCR3
(Oil/Gas)
NCR4
(Coal)
NCR4
(Oil/Gas)
SCR4
(Coal)
SCR4
(Oil/Gas)
8.0
15.8
7.8
26.9
16.4
59.6
23.3
Fixed O&M Cost Variable O&M Cost
($/kilowatt/year) (millions/kilowatt hour) % Removal
0.12
0.24
0.12
0.41
0.25
5.30
0.72
1.05 35
0.73 35
0.37 50
50
0.02 50
0.33 80
0.08 80
NOTES:
'Scaling factor for SNCR (Coal - Cyclone) is (100/MW)A0.577.
'Scaling factor for SNCR (Coal - All others) is (100/MW)A0.681.
3Scaling factor for SNCR (Oil/Gas) is (200/MW)A0.577.
"Scaling factors for NCR and SCR (both Coal and Oil/Gas) are (200/MW)A0.35.
116
-------
Costing equations for 2 percent, 3 percent and 4 percent fuel sulfur content
levels were developed for this model. The costing information for applying wet scrubber controls
to utility boilers are listed in Table VIII-2.
Table VIII-2
SO2 Scrubber Control Costs for Utility Boilers1
c.
Fuel Type
Medium Sulfur
(2% S)
High Sulfur
(3% S)
Medium Sulfur
(2% S)
NOTE:
Total Capital Cost Fixed O&M Cost Variable O&M Cost
($/kilowatt) ($/kilowatt/year) (millions/kilowatt hour)
149
166
174
5.4
6.0
6.3
0.83
6.3
1.8
'Scaling factor for Scrubber is (500/MW)A0.6.
ECU SOURCE
PM CONTROL MEASURES
The costs for fabric filters used to control PM emissions in this analysis were
developed for previous analyses. The costing information listed in Tables VIII-3, VIII-4, and VIII-5
can also be found in Pechan-Avanti's Additional Control Measure Evaluation for the Integrated Implementation of the
Ozone and Paniculate Matter 'National Ambient Air Quality Standards, and Regional Haze Program (Pechan, 1997) and,
more specifically, Pechan-Avanti's Regional Paniculate Strategies - Draft Report (Pechan, 1995).
D. REFERENCES
EPA, 1996: U.S. Environmental Protection Agency, Analyzing Electric Power Generation Under the CAAA, Office of
Air and Radiation, Washington, DC, July 1996.
EPA, 1998: U.S. Environmental Protection Agency, Analyzing Electric Power Generation under the CAAA, Office of
Air and Radiation, Washington, DC, March 1998.
EPA, 1999: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
"Compilation of Air Pollutant Emission Factors, Volume I Stationary Point
and Area Sources, Fifth Edition," accessed at EPA's internet website:
www.epa.gov/ttn/chief/ap42etc.html. Research Triangle Park, NC, 1999.
Pechan, 1995: The Pechan-Avanti Group, Regional Paniculate Strategies - Draft Report, prepared for U.S.
Environmental Protection Agency, Office of Policy Planning and Evaluation,
Washington, DC, EPA Contract No. 68-D3-0035, Work Assignment No.
1-54, September 29, 1995.
Pechan, 1997: The Pechan-Avanti Group, Additional Control Measure Evaluation for the Integrated Implementation of the
Ozone and Paniculate Matter National Ambient Air Quality Standards, and Regional Haze
Program, 1997.
117
-------
Table VIII-3
Equations for Estimating Capital Costs for Fabric Filters on Utility Boilers Fired
with Oil or Coal*
Fabric Filter Type
Shaker
Reverse-Air
Pulse-Jet
Pulse-Jet
BoilerFuel Type
Total
Oil/Coal
Oil/Coal
Oil
Coal
Equation**
Purchased Equipment
5.7019X + 77489
5.7993X + 69721
1.9634X + 59341
2.4967 + 59491
R2
Cost Equations
Value
1.00
1
1
1
.00
.00
.00
NOTES: 'Applies to units fired with any type of oil or coal.
**The variable "x" is the actual airflow rate into the fabric filter (actual cubic feet per minute).
Multiply purchased equpment costs by 2.17 to estimate total installed capital costs.
Multiply total installed capital costs by 1.4 to estimate retrofit costs.
118
-------
£ Table VIII-4
Equations for Estimating Annual Operating and Maintenance Costs for Fabric Filters on Utility Boilers
Fired With Oil*
Fabric Filter
Type
Shaker
Reverse -Air
Pulse-Jet
0
0.
0
Electricity
Equation**
.1876x- 19.576
2809x + 542.09
.1962x- 21.837
Dust Disposal
R2 Value
1.00
1.00
1.00
Equation**
0.0007X + 0.1895
0.0007X + 0.1895
0.0007x + 0.1895
R2
Value
1.00
1.00
1.00
Bag Replacement,
Equation** R2
0.2411X+ 1224.2
0.2866X+ 1486.8
0.1152X + 1.7916
1
Value
1.00
1.00
1.00
Com pressed
Equation**
Not Applicable
Not Applicable
0.1659x- 0.6381
Air, $
R2 Value
1.00
NOTES:
'Applies to any type of oil.
**The variable "x" is the actual airflow rate into the fabric filter (actual cubic feet per minute).
-------
8 Table VIII-5
Equations for Estimating Annual Operating and Maintenance Costs for Fabric Filters on Utility Boilers
Fired With Coal*
Fabric Filter
Type
Shaker
Reverse -Air
Pulse-Jet
Electricity
Equation**
0.1941X- 15.956
0.2869X + 562.52
0.2126X- 1.8948
R2 Value
1.00
1.00
1.00
0.
0.
0.
Dust
Equation
,7406x + 1.
,7406x + 1.
,7406x + 1.
Disposal
**
1461
1461
1461
R2 Value
1.00
1.00
1.00
0.
0.
0.
Bag Replacement, $
Equation** R2 Value
2497x+ 1220.7
2952x+ 1488.7
,1465x + 1.1497
1.00
1.00
1.00
Com pressed
Equation**
Not Applicable
Not Applicable
0.1659x- 0.6381
Air, $
R2 Value
1.00
NOTES:
'Applies to any type of coal.
**The variable "x" is the actual airflow rate into the fabric filter (actual cubic feet per minute).
-------
CHAPTER IX
HIGHWAY VEHICLES
This chapter analyzes control measures, their cost, and cost effectiveness for
highway vehicle mobile sources. For the purposes of this report only additional control measures,
above and beyond the existing (baseline) CAA measures, are analyzed. The control measures
examined in this chapter fall into the following four source categories:
• Vehicle Technology;
cv '
• In Use Vehicles;
• Fuel Options; and
• Transportation System Modifications.
A. VEHICLE TECHNOLOGY
1. Tier 2 Emission Standards
a. Description
EPA has proposed Tier 2 motor vehicle emission standards and gasoline
sulfur control requirements. For cars and light trucks, the proposed program would apply a single
average exhaust emission standard that would cover both passenger cars and all light trucks. This
program builds on the recent technology improvements resulting from the successful National Low
Emission Vehicle program and would also improve the performance of these vehides through low
sulfur gasoline (EPA, 1999).
b. Emission Reduction Estimate
The combined benefit of Tier 2 emission standards and associated fuel sulfur
limits in 2007 is 9.9 percent for VOC exhaust and 25.5 percent for NOX. These benefits apply to
light-duty gasoline vehicles (LDGVs) and light-duty gasoline trucks (LDGTs) (EPA, 1999). These
estimates are from the EPA Tier 2 spreadsheet model for the 47-State example.
c. Cost and Cost Effectiveness
i. Cost per
Ton of
VOC and
NOX
Reduction
The discounted lifetime nonmethane hydrocarbon plus NOX cost
effectiveness for the first five years of the program is $2,134 per ton. For year six and beyond, the
cost effectiveness is $1,748 per nonmethane hydrocarbon plus NOX ton reduced (EPA, 1999).
ii. Total Cost
The costs of the Tier 2 proposal can be divided into vehicle costs for
technology improvements to meet the more stringent emission standards, and increased fuel costs
to meet the gasoline sulfur restrictions. Table IX-1 provides EPA's estimates of the per vehicle
increase in purchase price for light-duty vehicles (LDVs) and light-duty trucks (LDTs). The near -
121
-------
term cost estimates are for the first years that vehicles meeting the standards are sold, prior to cost
J o ' L
reduction resulting from lower production costs and retirement of fixed costs (EPA, 1999).
Table IX-1
Estimated Purchase Price Increases Due to Proposed Tier 2 Standards
Standards
Tailpipe Standards
Near term (year 1)
Long term (year 6 and beyond)
Evaporative Standard
LDV
$76
46
4
LDT1
$69
43
4
LDT2
$132
99
4
LDT3
$270
214
4
LDT4
$266
209
4
The EPA-estimated per gallon gasoline cost increases for the years 2004,
2010, and 2015 (EPA, 1999) are as follows.
Estimated Per Gallon Cost for Desulfurizing Gasoline in Future Years
o
Year
Cost (cents per gallon)
2004
2010
2015
B.
IN-USE VEHICLES
1. Inspection and Maintenance
a.
Description
1.7
1.5
1.4
Inspection and Maintenance (I/M) is a way to check whether the emission
control system on a vehicle is working properly. The measure is designed to ensure that vehicles
stay clean while in use. Periodic checks will result in required repairs for those vehicles that fail the
test. In addition, tamperingwith emission control devices will be reduced (EPA, 1993).
On November 5, 1992, EPA established performance standards and other
requirements for basic and enhanced vehicle I/M programs. In this rule, enhanced I/M programs
were required in serious, severe, or extreme ozone nonattainment areas with urbanized populations
of 200,000 or more; and all metropolitan statistical areas with a population of 100,000 or more in
the Northeast OTR. These requirements have changed somewhat since the November 1992 rule
was released, and implementation of enhanced I/M programs has been much slower than originally
planned. The high-tech emission test that EPA originally proposed that areas use in an enhanced
I/M program is commonly known as an IM240 test. This test involves running each vehicle through
a 240 second test cycle on a dynamometer under load. Enhanced I/M programs were also to
include pressure and purge tests of the evaporative emission control systems to ensure that there are
no leaks and that the evaporative canister operates properly.
In practice, many areas have implemented decentralized I/M programs that
use acceleration simulation mode equipment because this equipment costs much less than the IM240
system. The cost analysis that follows has estimates for three generic I/M program types: basic
I/M, low enhanced I/M, and high enhanced I/M, to represent some of the variety of programs that
are observed in practice.
122
-------
b. Emission Reduction Estimates
Emission reductions depend upon the I/M program utilized. Reductions can
range from 5 percent to 30 percent for vehicle-related hydrocarbon and CO. In addition, the
program can also lead to reductions in NOX emissions of up to 10 percent (EPA, 1993).
c. Cost and Cost Effectiveness
i. Basic I/M
Costs for basic I/M are based on the RIA for enhanced I/M (EPA, 1992c).
The total-per-vehicle cost is based on the inspection fee, average repair cost, and the fuel economy
benefit. The average-per-vehicle cost is estimated to be $5.70. This is applied to LDGVs,
LDGTls, and LDGT2s in areas where basic I/M is required. If a basic I/M program is required and
a county already has a current I/M program, then no additional cost is attributed to that area. Basic
I/M costs are divided evenly among VOC, NOX, and CO (Pechan, 1998).
ii. Low
Enhanced
I/M
Costs for low enhanced I/M and OTR low enhanced I/M have not been well
defined. Therefore, its cost is considered equivalent to those of basic I/M. The average cost per
vehicle for this program is estimated to be $5.70 per vehicle. This per-vehicle cost applies to all
registered LDGV, LDGT1, and LDGT2 (Pechan, 1998).
Enhanced
I/M
Estimates of enhanced I/M costs are subject to change as States make
decisions about their program designs. I/M program costs maybe higher or lower according to
program designs selected by States. Enhanced I/M costs were also taken from the regulatory
support document. The estimated per-vehicle cost is $15.70. This figure is based on a test fee of
$18, an average repair cost of $14.20 per vehicle, and anaverage fuel economy benefit of $16.50
per vehicle. Annual costs can be estimated by applying the per vehicle costs to the projected vehicle
registrations for an area (Pechan, 1998).
2. Heavy Duty Diesel Vehicle Roadside Testing
a. Description
In practice, roadside testing of heavy-duty diesel vehicles (HDDVs) is
performed by stopping the vehicle or running it at a controlled speed past a checkpoint. An opacity
meter measures the transmittal of light through the exhaust plume, thereby measuring the amount
of particulates in the plume. This is an indirect and rough measure of the amount of NOX and
particulates in the plume. Some States are conducting these tests on a random and/or targeted
basis, largely at truck weigh stations, safety pullovers and similar examinations. States appear to be
conducting this activity to reduce the number of severity of smoking vehicles, or those emitting high
amounts of particulates from poorly performing engines, in part to reduce particulates and in part to
provide a measure of equity with LDGV emission tests.
Currently, no EPA program provides emission credits for HDDV testing:
the linkage between exhaust opacity and NOX is not yet certain. Several States have initiated
programs despite lack of avaikble emission credits. AZ, CA, and CO use opacity tests and cite
violators, providing time period to repair and be re-tested without fine. NJ instituted self
inspection, periodic inspection, and roadside inspection programs which are currently under
revision for simplicity and inspector productivity reasons. In June 1999, the Ozone Transport
Commission (OTC) States signed a resolution regarding interstate cooperation on the testing of
123
-------
heavy-duty diesel-fueled trucks and buses. Through this resolution, the OTC endorsed the testing
of heavy-duty on-road diesel trucks and buses (OTC, 1999).
b. Emission Reduction Estimate
Neither MOBILES nor MOBILE6 will include the means to estimate heavy-
duty vehicle (HDV) inspection program benefits. The State of California has implemented the
"Heavy-Duty Vehicle Inspection Program" (HDVIP) in November 1991 and developed the
"Periodic Smoke Inspection Program" (PSIP) in 1993. The emissions of ROG (Reactive Organic
Gases), NOX, and particulates will be reduced by the HDVIP and PSIP programs.
The HDVIP and PSIP Emission Reductions (percent) are estimated as
follows for two calendar years (CARB, 1997):
ROG NO
PM
1999
2010
18.0
13%
13.4
2.8%
3.5
17.9%
c. Cost and Cost Effectiveness
The analysis that follows is based upon recent cost estimates for California
(CARB, 1997):
California estimated annual HDDV inspection program costs:
Annual Labor Cost
$1,642,385
Annual Smokemeter Capital Cost
$6,817,787
Annual Cost of Inspections
$14,027,474
Total Fleet Annual Administrative Cost = $22,487,646
Total Cost to Vehicle Owners:
Annual Total Repair Cost
$16,208,686
Annual Increased Maintenance Cost =
$2,942,920
Annual Lost Opportunity Cost of Time =
$607,135
Annual Savings Cost of Fuel
-$24,983,117
Total Annual Cost to Vehide Owners =
-$5,238,756
Total Program Cost for 2010 = Total Fleet Adm. Cost + Total Cost Vehicle
Owners
$22,487,646
+ (-$5,238,756)
$17,248,890
Total Program Cost per Day $17,248,890 / 365 days = $47,257 per day
124
-------
Using the emission reduction estimates for 2010 the cost effectiveness values are:
o
Cost per Ton (NOX) = $47,257 / 14.03 tpd = $3,368 / ton
Cost per Ton (NOX + ROG) = $47,257 / 19.33 tpd = $2,445 / ton
Total Cost per Ton (NOX + ROG + PM) = $47,257 / 22.53 tpd = $2,098 / ton
3. Remote Sensing
a. Description
Remote sensing instruments monitor individual vehicle pollutant emissions
by referencing a measurement of the pollutant's concentration to the carbon dioxide concentration
in the vehicle exhaust. These measurements are made simultaneously and immediately behind the
moving vehicle and, thus have the potential to provide the best possible information on actual in-use
vehicle emission profiles. Remote sensing can also be used as a way to identify high emitters to be
directed toI/M facilities for testing. Another option for remote sensing is to use it for clean
screening in I/M programs. EPA has recently completed a draft guidance document on the use of
clean screening in I/M programs. Clean screening is designed to exempt certain cars from the I/M
requirement, on the strength of other evidence of the high probability that they are clean enough to
pass anyway. Based on a preliminary assessment of data currently available, EPA believes that it is
possible to excuse up to one-third of cars from inspection each year, with only a 5 to 10 percent loss
in emission reductions (EPA, 1998).
The Arizona Department of Environmental Quality has had a legislatively
mandated gross emitter remote sensing program since 1995. Cars and light trucks that have two
remote sensing readings within a time period that are above set CO and HC values (which differ by
model year), are mailed anotice requiring them to have an emission test within 30days. Cars that
are not retested have their registration suspended. The Phoenix program operates inMaricopa
County, which has a centralized, IM240 dynamometer-based emissions inspection program. There
are six remote sensing units that gather measurements 4 days per week, 7 hours per day. This
results in 150,000 records per month being submitted to the State Motor Vehicle Bureau.
Historically, this has resulted in 9,000 vehicle re-tests per year. This has been with a paired
van/instrument deployment. A recent change with a single van set-up has reduced the number of
re-tests to about 3,000 per year. Annual program cost is $914 thousand (Grubbe, 1999).
b. Emission Reduction Estimates
An analysis of clean screening in one State indicates that by using remote
sensing cutpoints that excuse 37 percent of remotely tested vehicles with a fleet coverage of 80
percent, the loss in tailpipe HC benefit is about 4 percent and the loss in overall HC benefit is 5 to 8
percent. By using remote sensing clean screening to measure NOX with a tight NOX cutpoint, NOX
benefit losses can be limited to 6 percent, for the 80 percent coverage example. If only HC and CO
cutpoints are used, the NOX benefit loss can be as high as 22 percent. MOBILES with the remote
sensing utility calcuktes a 0.24 percent vehide VOC and a 1.47 percent vehicle CO emission
benefit in calendar year 1999 for the Phoenix gross emitter program (Grubbe, 1999; EPA, 1998).
c. Cost and Cost Effectiveness
Based on the Arizona program performance in 1999, the VOC only cost
effectiveness of a gross emitter program is $10,000 per ton. Contractor charges for performing
remote sensing measurements and supplying license plate numbers and emission readings are in the
range of 50 cents to one dollar per vehicle. Motorist costs to those who fail the RS test would
include time for an additional inspection, plus repair costs (Grubbe, 1999; EPA, 1998).
4. Heavy Duty Retrofit Programs for Highway Engines
125
-------
a. Description
Voluntary programs to upgrade or retrofit after-treatment devices to older
heavy-duty engines could be a source of PM reductions in the heavy-duty highway categories. The
J J O J J O J O
number of engines retrofitted would vary based on the design of the local program. Based on the
amount of time over which the program could be phased in, it is assumed that 25 percent of all pre-
1994 highway heavy-duty engines still in thefleet in 2010 could be retrofitted (Pechan, 1997b).
b. Emission Reduction Estimate
The resulting PM reduction was estimated to be 2.82 percent in 2010
(Pechan, 1997b).
c. Cost and Cost Effectiveness
The costs and emission reductions associated with these programs are
somewhat speculative. For this analysis, it was assumed that both highway and nonroad engines
subject to the program could achieve a 25 percent reduction in PM emissions (Dolce, 1997). The
cost for the heavy-duty retrofit program (for highway vehicles) is estimated at $25,500 per ton of
PM10 reduced. This is based on the final rule for the Urban Bus Retrofit Program (58 FR 21385).
These estimates are based on EPA's experience to date with the existing urban bus retrofit program,
which has achieved similar reductions at similar cost (Pechan, 1997b).
5. Vehicle Retirement Programs
a. Description
Recently, air pollution regulatory agencies and researchers have focused
attention on the issue of older vehicles - those built with less effective emission control devices than
today's cars and light-duty trucks. Several California air pollution control districts have adopted
voluntary accelerated vehicle retirement programs, also known as vehicle buy back, cash for
clunkers, or vehicle scrapping. The programs offer vehicle owners cash for their old vehicle, which
is then scrapped. The potential cost effectiveness of such programs depends on the amount paid for
scrapped vehicles, annual mileages for scrapped versus replacement vehicles, and respective
emission rates for scrapped vs. replacement vehicles.
One example program is in the San Francisco Bay Area. Starting in 1996,
the Bay Area Air Quality Management District (BAAQMD) set aside a portion of its motor vehicle
registration fee revenue to purchase and retire pre-1975 vehicles. The program expanded in 1997
to include model years 1979 and earlier and in 1998 to include model years 1981 and older.
Through the end of May 1998, 1836 vehicles had been scrapped, with a contract to scrap 2,839
more. The BAAQMD program operates through contractors. Vehicle owners receive $500 per
vehicle (Dill, 1999).
b. Emission Reduction Estimate
The percentage emission reduction from 2007 baseline levels has not been
estimated.
c. Cost and Cost Effectiveness
The BAAQMD estimated that the 1997/1998 program cost $3,527per ton
of VOC and NOX reduced. Only direct program costs were included in this calculation. Other
researchers have estimated cost effectiveness for scrappage programs ranging from $2,500 to
$12,500 per ton of HC and NOX removed. This range is affected by the bounty offered for the
vehicles, the number of vehicles scrapped, and other program components. This program would
cost any area at least the per car purchase price ($500 to $ 1,000) times the number of scrapped cars
126
-------
per year. Any worthwhile program would probably have to scrap 1,000 vehicles per year, at a
minimum (Dill, 1999).
C. FUEL OPTIONS
1. Diesel
a. Description
The quality and composition of diesel fuel can have important effects on air
pollution emissions. The fuel variables having the most important effects on emissions are sulfur
contant, cetane number, and the fraction of aromatic hydrocarbons contained in the fuel. The
CAAA, in Section 217, required that effective October 1, 1993, motor vehicle diesel fuel would be
limited to a sulfur concentration of 0.05 percent (by weight) and a cetane index minimum of 40.
California's vehicle diesel fuel regulation goes beyond the Federal requirement by establishing a 500
ppm sulfur limit and requires a reduction in the aromatic content of the fuel from 30 to 10 percent.
b. Emission Reduction Estimate
From 2007 baseline levels a 4 percent reduction in diesel-fueled vehicle NOX
emissions would be expected.
c. Cost and Cost Effectiveness
It has been estimated that California reformulated diesel would increase the
per gallon cost of diesel by 6 cents (Pechan, 1998) compared with baseline diesel.
2. Alternative Fuel (CNG - Compressed Natural Gas)
a. Description
This measure evaluates the possibility of requiring private fleets to purchase
and use alternative fueled vehicles. Alternative fuel means a fuel other than gasoline. This
o
assessment compares compressed natural gas vehicle benefits with those from gasoline-powered
light-duty vehicles.
o J
At a June 1999 conference on alternative fueled vehicles, EPA said that in
MOBILE6, CNG vehicle emissions will be equal to the emissions of ultra-low emission vehicle
(ULEV) gasoline vehicles. The NOX emission standard for low-emission vehicles (LEVs) and ULEVs
is the same, so using this analysis technique, there will be no NOX benefits for CNG passenger cars
in an area once LEV standard vehides are sold (Kremer, 1999).
b. Emission Reduction Estimate
The NOX emissions difference between a Tier 1 car and a ULEV car (CNG in
this case) is estimated to be 0.236 grams per mile at 65,000 accumulated miles. If the lifetime of a
car is 130,000 miles and ten years, then the lifetime NOX benefit of a CNG car compared with a
Tier 1 car is 0.337 tons. Purchasing 1,000 dedicated CNG vehicles in 2000 would be expected to
be a 3 ton per year NOX credit, or 0.008 tpd.
c. Cost and Cost Effectiveness
Current estimates are that dedicated CNG vehicle prices are about $2,500
more than those for gasoline-powered vehicles. This is based on published estimates for the CNG
Toyota Camry LE. This analysis ignores the potential costs of adding fueling stations (AN, 1999).
Using a 10 year vehicle life time, the cost effectiveness of a CNG vehide
compared with a Tier 1 car is ($2,500)7(0.0337) tons = $74,000 per ton of NOX.
127
-------
D. TRANSPORTATION SYSTEM MODIFICATIONS
The section describes transportation system modifications with the potential
for controlling emissions. Specific reductions and costs are not cited herein, as the research in this
area is ongoing and often dependent upon specific regional/local parameters. The control measures
described are as follows:
1. Transit Improvements;
2. Pricing Mechanisms;
o '
3. Employer Provisions of Transportation; and
4. Voluntary Adjustment of Work Schedule.
1. Transit Improvements
This measure assumes increases in the levels of transit service (but not the
geographic coverage from existing levels). It makes transit a more convenient and viable option
from the typical 20-30 bus frequencies typically found outside large cities. Emissions reductions are
generated from increase transit ridership, thus avoiding auto trips, cold starts, hot sinks and VMT.
Modeling would estimate new mode shares for each wait time improvement
based on the expected change in travel time for each traveler and the probability of a traveler
changing mode to a higher occupancy vehicle or transit. Probability of mode change is dependent
both on the measure being tested and the baseline mode share. The revised vehide estimates are
converted to VMT and the VMT and vehicle reductions are translated to emissions benefits.
Improved public transit is comprised of three main components, which may
effectively reduce congestion and improve air quality. The goal of improving public transit is to
provide incentives for single occupancy vehicle commuters to forgo driving for the convenient and
reasonably priced alternative of mass transit. The three major ways of increasing ridership on public
transit are (1) system/ service expansion, (2) system/service operational improvements, and (3)
inducements to travelers to increase ridership. Transportation planners should be aware that these
strategies vary in risk, cost, and potential benefits.
Projects may be extremely costly if they are capital intensive (e.g., building
rail lines) and rely on infrastructure changes; improvements involving transit schedules and public
awareness programs are much cheaper. Examples of capital intensive projects are a dual rail/bus
tunnel system in Seattle that will improve bus and rail service in the region, and a light rail line
servicing Houston. These projects cost 400 million and one billion dollars, respectively.
Improving transit is effective in increasing the use of buses versus driving
alone. While there is an increase in bus usage (typically 10 percent), only a portion of the increase
is accounted for from lone drivers. Some increase stems from people who formerly used other
high-occupancy or non-motorized modes of transportation. The overall auto trip reduction is
estimated to be just under one percent.
2. Pricing Mechanisms
Pricing programs, such as an increase in the State gas tax, affect air quality
primarily by reducing automobile trips and VMT. Reducing commuter trips not only reduces
emissions associated with VMT, but those associated with "cold starts," when commuters set out in
the morning and "hot soaks," when vehicles are parked at work and continue to produce
evaporative emissions even after the engines are turned off.
Tax parking benefits and parking lots increase the cost of driving a motor
vehicle by ensuring that some level of the costs of providing parking are paid by users. This is
usually a peak hour measure targeting single occupant automobiles, especially those parking in
suburban areas where there is usually no charge for parking. The measure also may seek to equalize
the costs of transit use with driving, and to encourage other HOV modes. The measure is often
128
-------
utilized in combination with incentives for non-SOV use, such as transit subsidies, flex time, etc.,
and may be analyzed in combination with these to capture synergistic effects.
Parking charges may be levied in a variety of manners: tax per parking space
(suburban), minimum parking charges in parking garages, collected by the employer and used to
cross subsidize other employee benefits (such as transit passes). The parking charge encourages
ridesharing, transit and non motorized modes to avoid the full charge (e.g., rideshare people divide
the cost among them) or all together (walk, bike access).
The increase in parking costs, gas taxes, and tolls result in a change in trip
destinations, travel modes, and travel time periods. The goal of effecting the peak periods of travel-
typically morning rush hour- and distributing overall trips, can be reached with such a measure. It
is estimated that regional auto travel is marginally effected but total transit share increases by four to
five percent (NCHRP, 1998).
3. Employer Provisions of Transportation/Buses to
Employees
This control measure provides for employers to voluntarily provide subsidies
to employees for transit and rideshare (carpool, van pool, etc.) use in lieu of drive access to the
work place. These subsidies are utilized extensively in a number of areas, including Philadelphia,
Washington, Boston, Baltimore, and many smaller locales. The concept seeks to avoid penalizing
those employees who desire to use non-HOV modes, while parking may be provided free or at a
subsidized cost to those who drive to work.
Employer based provision of transportation/buses can be simulated by
providing four levels of transit subsidy to an employee: $0.50, $ 1.00, $ 1.50 and $2 .00. The
subsidies can be analyzed in a voluntary framework where employers would provide the subsidies
on a voluntary basis entirely dependent on the employers perception of the benefits they might
accrue (NCHRP, 1998).
4. Voluntary Adjustment of Work Schedule
Employees are allowed/encouraged to alter their work schedules to
arrive/depart earlier/later at their work place. This reduces emissions by reducing congestion
during peak commuting hours, allowing vehicles to operate at more economical and steadier
speeds. This measure does NOT include flexing the days a person works, only the starting/ending
time of day. Usually synergistic with transit, parking charge and other flex hours/days measures,
the measure may be analyzed as such for synergistic effects.
Employer based provision of flex time work schedules should be simulated at
four levels of availability: 15 percent of employees eligible, 30 percent, 45 percent and 60 percent.
Regional averages for starting mode shares and average trip length can be used to provide a baseline
for modeling.
Telecommuting is another way to reduce work trips. It is important to note
that while the reduction in peak period trips (i.e. morning rush hour) is critical to the reduction in
emissions, there are also added non-work trips during the day that offset these benefits. Errands
that would have been done on the way to or from work are now done on added trips during the day.
Studies by the National Cooperative Highway Research Program show that with these offsets, there
can be an expected decrease in overall trips by approximately one percent.
E. REFERENCES
AN, 1999: Automotive News, "Natural Gas Camry debuts in fall," May 24, 1999.
CARB, 1997: California Air Resources Board, "Technical Support Document for the Proposed
Amendments to the California Regulation Governing the Heavy-Duty
129
-------
Vehicle Inspection Program and Periodic Smoke Inspection Program,"
October 1997.
Dill, 1999: Jennifer Dill, "Scrapping Older Vehicles to Improve Air Quality in the San Francisco Bay
Area," Transportation Research Board, 78th Annual Meeting, Washington,
DC, January 10- 14, 1999.
EPA, 1993: U.S. Environmental Protection Agency, "Clean Cars for Clean Air: Inspection and
Maintenance Programs," EPA400-F-92-016, Fact Sheet OMS-14, January
1993.
EPA, 1998: U.S. Environmental Protection Agency, "Clean Screening in Inspection and Maintenance
Programs," (EPA-420-F-98-023), Office of Mobile Sources, May 1998.
EPA, 1999: U.S. Environmental Protection Agency, "Control of Air Pollution from New Motor Vehicles:
' O J '
Proposed Tier 2 Motor Vehicle Emissions Standards and Gasoline Sulfur
Control Requirements," 40 CFR Parts 80 and 86, Notice of Proposed
Rulemaking, 1999.
Grubbe, 1999: Dan Grubbe, Personal Communication with Jim Wilson (The Pechan-Avanti Group),
Arizona Department of Environmental Quality, Phoenix, AZ, June 14,
1999.
Kremer, 1999: Janet Kremer, "MOBILE6 and Compressed Natural Gas Vehicles," presentation at the
U.S. Environmental Protection Agency's Alternative Fuel Modeling
Workshop, Office of Mobile Sources, May 26, 1999.
NCHRP, 1998: NCHRP Project 8-33, prepared for the National Cooperative Highway Research Program
by Cambridge Systematics, Inc., November 1998.
OTC, 1999: Ozone Transport Commission, "Resolution of the States of the Ozone Transport Commission
Regarding Interstate Cooperation on the Testing of Heavy Duty Diesel-
Fueled Trucks and Buses," June 16, 1999.
Pechan, 1997b: The Pechan-Avanti Group, Additional Control Measure Evaluation Jor the Integrated Implementation of the
Ozone and Paniculate Matter National Ambient Air Quality Standards, and Regional Haze
Program, prepared for U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle Park, NC, July 17, 1997.
Pechan, 1998: E.H. Pechan& Associates, Inc., "Clean Air Act Section 812 Prospective Cost Analysis,"
Draft Report, Springfield, VA, prepared for Industrial Economics, Inc.,
Cambridge, MA, January 5, 1998.
130
-------
CHAPTER X
NONROAD ENGINES AND VEHICLES
A. REGULATIONS
Nonroad sources are mobile (non-highway) emission sources including lawn
and garden equipment, construction equipment, agricultural equipment, industrial equipment,
aircraft and airport service vehicles, logging equipment, recreational vehicles, locomotives, and
marine vessels. Under the authority of the CAA, EPA has proposed or finalized regulations for a
number of nonroad engine categories. The applicable regulations are listed in Table X-1.
B. CONTROL OPTIONS
This section discusses potential control measures beyond Federal engine
standards. The nonroad control measures which require new engines to meet stricter standards will
require time for fleet turnover before the regulations become 100 percent effective. These new
standards may include implementation of some of the control techniques listed below.
1. Diesel-Powered Engines
Direct injection (turbocharged/aftercooled) is a control measure for PM and
NOX. The primary PM - fine control efficiency has been estimated to be between 50 percent and 80
percent (FACA, 1997). Low-sulfur diesel fuel for nonroad engines is a control measure for PM
with primary PM - fine control efficiency in the range 0-50 percent (FACA, 1997).
In 1993 CARB's diesel-fuel regulations took effect. The California diesel is
much more clean burning than conventional diesel fuel and its cost-effectiveness is comparable to
other measures adopted in California. Switching to California diesel from conventional diesel lead
to significant reductions in emission of pollutants from vehicles and equipment that used diesel fuel.
There was an 82 percent reduction in SO2, a 25 percent reduction in particukte matter, and a 7
percent reduction in NOX. On an average the wholesale diesel prices in California have been within
five cents per gallon of diesel as compared to the costs in neighboring States (CARB, 1997).
2. Gasoline-Powered Engines
Electric motor substitution is control measure for all pollutants having a
control efficiency between 80 percent and 100 percent (FACA, 1997).
131
-------
Table X-1
Federal Regulations Affecting Future Year Emissions
Source Category
Aircraft
Railroads
Commercial Marine
Recreational Marine
Compression Ignition
(all nonroad
categories)
Spark Ignition (all
nonroad categories)
Applicable Regulations
May 8, 1997 rulemaking consistent with standards set by the United National
International Civil Aviation Organization.
Final emissbn standards published April 16, 1998.
Notice of Proposed Rulemaking issued for compression ignition marine
engines.
Final rule published October 4, 1996. Affects spark ignition recreational
marine vessels. Aimed at reducing VOC emissions.
Tier 1 standards promulgated for engines > 50 horsepower in 1994.
Tier 1 standards for small engines and tier 2 standards for all engines
promulgated in August of 1998.
Phase 1 promulgated July 1 995 for small engines (at or below 1 9 kilowatt/25
horsepower).
Phase 2 proposed December 1997, final phase 2 standards for nonhand-held
small engines signed March 1999.
Phase 2 Supplemental Notice of Proposed Rulemaking for new handheld
engines at or below 19 kilowatt was signed June 30, 1999.
132
-------
3. Locomotives
Some of the possible control measure options for NOX reductions are listed
in Table X-2 (STAPPA/ALAPCO, 1994). The source for the data is CARB. The reductions and
cost information are based on California locomotive fleet parameters.
Among the new and emerging railroad technologies is the electric
CyberTran rail transportation system (Manufacturer: CyberTran International, Inc.) which consists
of several small, computer-controlled vehicles. It is a lightweight rail system, which can decrease
congestion and since it is electrically powered it does not emit pollutants. The capital cost of a
CyberTran system (2 to 4 million dolkrs per mile) is 10 to 50 percent of the cost of conventional
light rail systems and the operational costs should also be less as compared to conventional rail and
transit systems. The information about CyberTran has been obtained from the website developed
by the University of California Riverside (CE-CERT, 1999).
4. Marine Compression Ignition Engines (Commercial
Marine Vessels)
Selective catalytic reduction (NOX emission reduction: 70 to 90 percent)
and water/fuel emulsion (NOX emission reduction: up to 35 percent) are two of the control
measures (STAPPA/ALAPCO, 1994). The source for the data is CARB.
5. (Recreational) Marine Spark Ignition Engines
a. Marine Fuel Additive
SoyGold Marine™ (Manufacturer: AG Environmental Products, L.L.C.) is a
soybean oil-based dieselfuel additive, which can be used for marine engines. The use of a 20/80
blend of SoyGold Marine™ when compared with an oxidation catalyst and low-sulfur #2 diesel
showed a decrease in particulate matter emissions by 45 percent, a decrease in total hydrocarbon
emissions by 65 percent, a decrease in CO emissions by 41 percent, and an increase in NOX
emissions by 7 percent. The cultivation of soybeans in order to produce this product leads to a
consumption of six times more CO2 than emitted during its use. Generally this product will cost
$0.60 to 0.80 per gallon more than petroleum diesel. The information about SoyGold Marine™ has
been obtained from the website developed by the University of California Riverside (CE-CERT,
1999).
b. Credits for Replacement of Existing
Pleasure Craft Engines with New Lower
Polluting Engines
This control measure focuses on the accelerated repkcement of existing
pleasure craft engines with lower polluting engines. Pleasure craft are recreational marine vessels.
This control measure includes the development of an emission reduction credit program. This
would allow the governing body to issue emission reduction credits for programs that accelerate the
replacement of existing pleasure craft engines with lower polluting engines. Such a program would
be voluntarily implemented and would provide industry with more flexible and potentially more
cost effective approaches in complying with SIP requirements. Program operators would qualify for
emission reduction credits by replacing existing uncontrolled engines with new engines that meet
EPA emission
133
-------
Table X-2
Control Measures for NO, Reductions from Locomotives
Control Measure
Electric line haul, liquified natural gas (LNG) + SCR,
local/switch
Engine Modifications
LNG and SCR
LNG dual-fuel
LNG dual-fuel line haul, remanufacture/replace bcal/switch
LNG dual-fuel line haul, LNG, spark ignition local/switch
LNG, spark ignition
Low-aromatic diesel fuel
Selective catalytic reductbn
NOX Reduced
99%
38%
97%
70%
72%
80%
86%
10%
90%
Cost Effectiveness
($/ton)
14,800
3,200
1,900
900
1,700
900
1,400
5,000
2,900
134
-------
standards. This is based on SCAQMD Proposed Rule 1624. The cost per ton of VOC and NOX
reduction has not yet been determined (SCAQMD, 1999).
6. Airport
a. Electric Vehicles
The use of electric vehicles in airports is a new and emerging air pollution
control program. The hybrid-electric prototype T-1000 Neighborhood Truck (Manufacturer:
Coval H2 Partners) which can be used as a maintenance vehicle is powered by a fuel cell and is a
zero-emission vehicle. The ThunderVolt 701 AT electric power train (Manufacturer: ISE Research)
is a zero emission vehicle. It is used as a drive system for several types of aircraft specialty
equipment. As compared to regular power trains it has 10-20 percent lower maintenance costs.
Information about the electric vehicles has been obtained from the website developed by the
University of California Riverside (CE-CERT, 1999).
b. Liquified Petroleum Gas-Fueled Vans
for Rental Car Company Shuttles
This control measure is to substitute propane-fueled vehicles for the vans
that transport passengers from rental car lots to terminals and vice-versa. The emission benefits
(tpd) are 0.005 for VOC and 0.003 for NOX. For this control measure the cost-effectiveness has
been estimated to be $24,200/ton for VOC and $40,900/ton for NOX. The key assumptions in
determining the cost-effectiveness are: i) the incremental vehide cost is $3,000, amortized for 8
years, ii) the fueling station cost is $79,000, amortized for 10 years, and ii) the fuel cost savings are
$.0059 per mile (conventional gasoline vs. propane). Shuttle operations, with limited range
requirements and centralized fleet characteristics, can be a good application for alternative fuel
technologies. In this instance, the lower cost of propane (based on the U.S. Department of Energy
(DOE) study) indicates that an option that reduces emissions can also reduce cost. Cost-
effectiveness is greatly influenced by the fuel cost. However, the small number of vehicles and
miles generates very little VOC and NOX savings (Molle, 1992; DOE, 1996).
7. Commercial/Industrial Mobile Equipment
The Envirolift ECD (Emission Control Device, Manufacturer: Envirolift)
has been designed as an emission control device for emissions from four or six cylinder engines. It
can be used for gas and liquid propane (LP) equipment like Forklifts, Manlifts, and
Scrubber/Sweepers. It significantly reduces emissions of CO and HC. Its emission reduction
capacity is up to ten times more than a catalytic converter, after the converter has reached a
temperature of 600 degrees Farenheit. The ECD unit costs around $1500, plus installation.
Information about Envirolift was obtained from the website developed by the University of
California Riverside (CE-CERT, 1999).
8. Lawn and Garden Equipment
a. Electric Lawnmowers
Cordless, electric lawnmowers are available in the marketplace. The
Cordless Mulching Mower (Model CMM1000, Manufacturer: Black and Decker) is powered by
improved batteries, which increase the operation time. Cordless mowers cost a little more than
conventional gasoline mowers. The operational costfor battery charging is around $3/year.
During operation cordless mowers do not create emissions but some emissions are created during
the recharging. The LawnPup (Manufacturer: GrassMasters) is a compact electric lawn mower.
The LawnPup 1000 model costs around $ 180. The use of electrical engines can lead to a reduction
in emissions of particulates by 91 percent, CO by 76 percent, and VOCs by 82 percent. The
TrimmerPlus 1090r trimmer (Manufacturer: Ryobi Outdoor Products, Inc.) is equipped with a
Ryobi second generation gas-powered 4-cycle engine, which is muchmore efficient than other 4-
stroke engines and is cleaner than 2-stroke engines. Compared to standard mowers, it reduces
135
-------
exhaust pollutants by 80 percent and costs $250.00. Information about these lawn and garden
equipment was obtained from the website developed by the University of California Riverside (CE-
CERT, 1999).
A control measure to control emissions from lawn and garden equipment is
to provide incentives for electric lawnmowers and to subsidize them. During the spring and
summer of 1995, the Maryland Department of the Environment implemented Cash-for-Clippers, a
lawn and garden equipment trade-in program. Through Cash-for-Clippers, Maryland provided $75
rebates toward the purchase of environmentally friendly (electric or push mowers) lawn equipment
to individuals who scrapped their gasoline-powered equipment. For hand-held equipment, the
rebates were $25. Compared with a gasoline-powered mower meeting EPA's 1997 emission
standards, a cordless mower has 99.9 percent lower VOC emissions. The market penetration of
electric lawnmowers depends on their cost relative to gasoline-powered mowers (as long as their
performance is perceived to be the same as a gasoline-powered mower). Cost effectiveness
calculations are based on residential use replacement, and the rebate amount ($75 for kwnmowers)
is about equal to the price difference between electrics and gasoline-powered mowers (associated
SCCs: 2260004010, 2265004010). Emission reductions are based on 10 percent market
penetration by 2005 of battery-powered lawn and garden equipment. The cost-effectiveness when
considering only consumer equipment is estimated to be $1172 per ton of VOC (MDE, 1996a;
EPA, 1991; EPRI, 1996a; EPRI, 1996b).
b. Leakless Gasoline Can Nozzles
This measure involves using vapor recovery nozzles to control refueling
emissions from the refueling of lawn and garden equipment. Special nozzles are available with an
automatic stop device. They work by keeping the gasoline from pouring until the nozzle is inserted
in the tank, stopping flow automatically when the tank becomes full, and sealing the container when
the nozzle is removed from the tank. For a small nozzle in typical residential use, the cost-
effectiveness of the vapor recovery nozzle is $1,400 to $5,800 per ton of VOC depending on the
gasoline quantity used during the summer season. Because the nozzle provides fuel savings, more
gasoline usage produces a lower cost per ton. In commercial use, fuel savings outweigh the nozzle
cost, so the cost effectiveness is a savings of $130 to $290per ton (MDE, 1996b; VEMCO, 1996a;
VEMCO, 1996b).
9. Container Spillage Control Measures
This control measure would reduce VOC emissions from portable fuel
containers used to refuel off-road equipment by replacing the current style of portable containers
with a spill-proof container system. These spill-proof container systems would eliminate overfilling
spillage and would substantially mitigate container transport and storage, container diurnal, and
container permeation emissions (CARB, 1999). The proposed regulations would apply to new cans
and spouts that are sold in California starting January 1, 2001 (CARB, 1999). The measure will be
implemented through compliance testing and the attrition of noncompliant portable fuel containers
(Bloudoff, 1999). Full compliance with the proposed measure is expected by 2007, assuming an
average container service life of six years (Bloudoff, 1999). The associated SCCs for this control
measure are 2260004000 and 2265004000.
Spill-proof containers would automatically stop the flow of fuel before
overflow of the target fuel tank, automatically close and seal when removed from the target fuel
o ' J o
tank, and remain completely closed when not dispensing fuel. Filling and pouring would both be
through the same opening, and the minimum flow rate would be one-half gallon per minute for the
smallest portable fuel containers, and higher for larger containers (CARB, 1999). The container
would be made of materials which ensure that the vapor permeation rate does not exceed 0.4 grams
per gallon per day (CARB, 1999).
About 87 tpd of smog-forming ROG are released into California's air from
portable gas cans. The proposed regulations being adopted will lead to a 73 percent reduction in
ROG emissions from gas cans by 2010. CARB estimates that the cost of the proposed regulation
would be $4020 for each ton of ROG reduced (CARB, 1999).
136
-------
C. REFERENCES
Bloudoff, 1999: Dan Bloudoff, California Air Resources Board, personal communication with Eric
Albright, June 14, 1999.
CARB, 1997: California Air Resources Board, California Diesel Fuel Fact Sheet, March 1997.
CARB, 1999a: California Air Resources Board, Draft Proposed Portable Container Spillage Control Regulations, May 28,
1999.
CARB, 1999b: California Air Resources Board, Gas Can Fact Sheet, September 20, 1999.
CE-CERT, 1999: College of Engineering, Center for Environmental Research and Technology (CE-
CERT) Website (www.nutech.org), the University of California Riverside,
1999.
DOE, 1996: U.S. Department of Energy, "Assessment of Costs and Benefits of Flexible and Alternative
Fuel Use in the U.S. Transportation Sector, Technical Report 14: Market
Potential and Impacts of Alternative Fuel Use in Light-Duty Vehicles: A
2000/2010 Analysis," p. C-39, January 1996.
EPA, 1991: U.S. Environmental Protection Agency, "Nonroad Engine and Vehicle Emission Study -
Appendices," (21A-2001), Office of Air and Radiation, Washington, DC,
November 1991.
EPRI, 1996a: Electric Power Research Institute, "The Environmental and Energy Benefits of Cordless
Electric Lawn Mowers," EPRI Report TR1065S9, Palo Alto, CA, July 1996.
EPRI, 1996b: Electric Power Research Institute, EPRI Journal, p. 18, March/April 1996.
FACA, 1997: National and Regional Strategies Work Group of the FACA Subcommittee, "Opportunity
Matrix for Ozone, PM-fine, and Regional Haze Integration," March 31,
1997.
MDE, 1996a: Maryland Department of the Environment, "Cash-for-Clippers - A Lawn Mower Rebate and
Trade-in Program," The Air Quality Planning Program, Baltimore, MD,
September 12, 1996.
MDE, 1996b: Maryland Department of the Environment, "Facts About Refueling Your Lawn and Garden
Equipment," Baltimore, MD, 19%.
Molle, 1992: Robert Molle, Philadelphia International Airport, response to request for information
including vehicle classification counts from December 1992.
o
SCAQMD, 1999: South Coast Air Quality Management District, "Rule and Control Measure Forecast
Report," August 1999.
STAPPA, 1994: State and Territorial Air Pollution Program Administrators/Association of Local Air
Pollution Control Officials, "Controlling Nitrogen Oxides Under the Clean
Air Act: A Menu of Options," July 1994.
VEMCO, 1996a: VEMCO, Inc., Sure Pour Automatic Leakless Nozzle Brochure, Emmett, ID, 1996.
VEMCO, 1996b: VEMCO, Inc., "Report of the Initial Evaluation of the Sure Power Fuel Container
Spout," Emmett, ID, Henry J. Beaulieu, Industrial Hygiene Resources, Ltd.,
Boise, ID.
137
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO. 2.
EPA-452/D-0 1-001
4. TITLE AND SUBTITLE
Control Measure Evaluations: The Control Measure Data Base for t
Emission Trends Inventory (ControlNET)
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 2771 1
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 2771 1
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
May 2001
te National
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OFREPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report contains a description of control measures that are contained with the ControlNET database. The c
for reductions of emissions each of the criteria pollutants that the Environmental Protection Agency (EPA) reg
measures can be applied to sources that are contained in the National Emission Trends (NET) Inventory, a con
of criteria pollutant emission sources for which EPA has data. This database is meant to serve as a repository
measure data for use in analyses EPA conducts in support of its air pollution regulations.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Emission Controls
Emission Inventory
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution control
Control Measures
Control Costs
19. SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Page)
Unclassified
c. COSATI Field/Group
21. NO. OFPAGES
173
22. PRICE
antrol n
ilates.
prehens
of curre
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
138
------- |