SPA-SOO/7-3G-009a
Environmental Protection
Agency February 1986
Research and
Development
EVALUATION OF
CONTBOL TECHNOLOGIES FOR
HAZARDOUS AIR POLLUTANTS
Volume 1. Technical Report
Prepared for
Office of Air Quality Planning and Standards
Prepared by
Air and Energy Engineering Research
Laboratory
Research Triangle Park NC 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development U.S. Environmental
Protection Agency, have been grouped into nine series. These nine brotd cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development*
8. "Special" Reports
9. Miscellaneous Reports
This report nas been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of. control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public througn the National Technical Informa-
tion Service. Springfield, Virginia 22161.
-------
EPft-&00/7-6&-009a
EVALUATION OF CONTROL TECHNOLOGIES
FOR HAZARDOUS AIR POLLUTANTS
Volume 1. Technical Report
by
Robert f. Purcell
Pacific Environmental Services, Inc,
1905 Chapel Hill Road
Durham, North Carolina 27707
and
Gunseli Sagun Shareef
Radian Corporation
3200 Progress Center
Research Triangle Park, North Carolina 27709
EPA Contract No. 68-02-3981
Project Officer:
Bruce Tichenor
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared for:
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
ABSTRACT
The purpose of this manual is to help EPA regional, State, and
local air pollution control agency technical personnel to select,
evaluate, and cost air pollution control techniques for reducing or
eliminating the emission of potentially hazardous air pollutants (HAP's)
from industrial/commercial sources. The information provided by this
manual will be useful for reviewing permit applications or for informing
interested parties as to the type, basic design, and cost of available
HAP control systems.
Since the definition of a HAP is very broad and, thus, encompasses
potentially thousands of specific compounds, it is not possible for
this handbook to develop an all-inclusive list of HAP compounds and
compound-specific control techniques. However, the number of generic
air pollution control techniques available is small, and the factors
affecting the cost and performance of these controls as applied to many
noncriteria pollutants have been identified and discussed in the literature.
Therefore, the main focus of this manual is to provide sufficient
guidance to select the appropriate air pollution control system(s) for
an emission stream/source containing HAP's.
. The manual will help the user perform three distinct functions:
(1) to select the appropriate control technique(s) that can be applied
to each HAP emission stream generated at a specific facility, (2) to
determine the basic design parameters of the selected air pollution
control device(s) and accompanying auxiliary equipment, and (3) to
estimate order-of-magnitude control system capital and annualized
costs.
111
-------
TABLE OF CONTENTS
Page
Abstract iii
List of Figures xiv
List of Tables xvii
Nomenclature xx
Conversions From English To Metric Units xxviii
Acknowledgments xxix
Chapter 1: Introduction 1-1
1.1 Objective 1-1
1.2 How to Use the Manual 1-3
Chapter 2: HAP Emissions and Their Key Physical Properties ... 2-1
2.1 Identification of Potential HAP's and Emission
Sources 2-4
2.1.1 Solvent Usage Operations 2-5
2.1.2 Metallurgical Industries 2-6
2.1.3 Synthetic Organic Chemical Manufacturing
Industry (SOCMI) 2-8
2.1.4 Inorganic Chemical Manufacturing Industry . 2-11
2.1.5 Chemical Products Industry 2-11
2.1.6 Mineral Products Industry 2-12
2.1.7 Wood Products Industry 2-12
2.1.8 Petroleum Related Industries 2-12
2.1.9 Combustion Sources 2-14
2.1.10 References for Section 2.1 2-43
2.2 Identification of Key Emission Stream Properties. . 2-49
Chapter 3: Control Device Selection 3-1
3.1 Vapor Emissions Control 3-2
3.1.1 Control Techniques for Organ-ic Vapor
Emissions from Point Sources 3-2
3.1.1.1 Thermal Incinerators 3-8
3.1.1.2 Catalytic Incinerators 3-9
3.1.1.3 Flares 3-10
-------
TABLE OF CONTENTS
(continued)
3.1.1.4 Boilers/Process Heaters 3-10
3.1.1.5 Carbon Adsorbers 3-11
3.1.1.6 Absorbers (Scrubbers) 3-12
3.1.1.7 Condensers 3-14
3.1.2 Control Techniques for Inorganic Vapor
Emissions from Point Sources 3-15
3.1.2.1 Absorbers (Scrubbers) 3-16
3.1.2.2 Adsorbers 3-18
3.1.3 Control Techniques for Organic/Inorganic
Vapor Emissions from Process Fugitive
Sources 3-19
3.1.4 Control Techniques for Organic/Inorganic
Vapor Emissions from Area Fugitive
Sources 3-22
3.1.5 Control Device Selection for a Hypothetical
Facility 3-26
3.1.6 References for Section 3.1 3-35
3.2 Particulate Emissions Control ........... 3-36
3.2.1 Control Techniques for Particulate Emissions
from Point Sources 3-36
3.2.1.1 Fabric Filters 3-38
3.2.1.2 Electrostatic Precipitators .... 3-40
3.2.1.3 Venturi Scrubbers 3-41
3.2.2 Control Techniques for Particulate Emissions
from Fugitive Sources 3-44
3.2.2.1 Process Fugitive Particulate
Emission Control 3-45
3.2,2.2 Area Fugitive Emission Control from
Transfer and Conveying ...... 3-46
3.2.2.3 Area Fugitive Emission Control from
Loading and Unloading 3-48
3.2.2.4 Area Fugitive Emission Control from
Paved and Unpaved Roads 3-52
3.2.2.5 Area Fugitive Emission Control from
Storage Piles 3-55
-------
TABLE OF CONTENTS
(continued)
3.2.2.6 Area Fugitive Emission Control from
Waste Disposal Sites 3-57
3.2.3 References for Section 3.2 3-59
Chapter 4: HAP Control Techniques 4-1
4.1 Thermal Incineration 4.1-1
4.1.1 Data Required 4.1-3
4.1.2 Pretreatment of the Emission Stream:
Dilution Air Requirements 4.1-4
4.1.3 Thermal Incinerator System Design Variables. 4.1-5
4.1.4 Determination of Incinerator Operating
Variables 4.1-7
4.1.4.1 Supplementary Heat Requirements . . 4.1-7
4.1.4.2 Flue Gas Flow Rate 4.1-11
4.1.5 Combustion Chamber Volume 4.1-13
4.1.6 Heat Exchanger Size 4.1-14
4.1.7 Evaluation of Permit Application 4.1-17
4.1.8 References for Section 4.1 4.1-20
4.2 Catalytic Incineration 4.2-1
4.2.1 Data Required 4.2-3
4.2.2 Pretreatment of the Emission Stream:
Dilution Air Requirements 4.2-5
4.2.3 Catalytic Incinerator System Design
Variables 4.2-6
4.2.4 Determination of Incinerator System
Variables 4.2-8
4.2.4.1 Supplementary Heat Requirements . . 4.2-8
4.2.4.2 Flow Rate of Combined Gas Stream
Entering the Catalyst Bed 4.2-13
4.2.4.3 Flow Rate of Flue Gas Leaving the
Catalyst Bed 4.2-15
4.2.5 Catalyst Bed Requirement 4.2-16
VII
-------
TABLE OF CONTENTS
(continued)
Page
4.2.6 Heat Exchanger Size (for Systems with
Recuperative Heat Exchange Only) 4.2-17
4.2.7 Evaluation of Permit Application 4.2-20
4.2.8 References for Section 4.2 4.2-22
4.3 Flares 4.3-1
4.3.1 Data Required 4.3-3
4.3.2 Determination of Flare Operating Variables . 4.3-4
4.3.2.1 Supplementary Fuel Requirements . . 4.3-5
4.3.2.2 Flare Gas Flow Rate and Heat
Content 4.3-6
4.3.2.3 Flare Gas Exit Velocity 4.3-6
4.3.2.4 Steam Requirements 4.3-9
4.3.3 Evaluation of Permit Application 4.3-10
4.3.4 References for Section 4.3 4.3-12
4.4 Boilers/Process Heaters . 4.4-1
4.5 Carbon Adsorption 4.5-1
4.5.1 Data Required 4.5-4
4.5.2 Pretreatment of the Emission Stream 4.5-6
4.5.2.1 Cooling 4.5-6
4.5.2.2 Dehumidification 4.5-6
4.5.2.3 High VOC Concentrations 4.5-7
4.5.3 Carbon Adsorption System Design Variables. . 4.5-7
4.5.4 Determination of Carbon Adsorber System
Variables 4.5-9
4.5.4.1 Carbon Requirements 4.5-9
4.5.4.2 Carbon Adsorber Size 4.5-12
4.5.4.3 Steam Required for Regeneration . . 4.5-14
4.5.4.4 Condenser . 4.5-17
4.5.4.5 Recovered Product 4.5-19
4.5.5 Evaluation of Permit Application 4.5-20
4.5.6 References for Section 4.5 4.5-22
vm
-------
TABLE OF CONTENTS
(continued)
4.6 Absorption 4.6-1
4.6.1 Data Required 4.6-4
4.6.2 Absorption System Design Variables 4.6-5
4.6.3 Determination of Absorber System Design and
Operating Variables 4.6-6
4.6.3.1 Solvent Flow Rate 4.6-6
4.6.3.2 Column Diameter 4.6-8
4.6.3.3 Column Height 4.6-12
4.6.3.4 Pressure Drop Through the Column. . 4.6-17
4.6.4 Evaluation of Permit Application 4.6-18
4.6.5 References for Section 4.6 4.6-21
4.7 Condensation 4.7-1
4.7.1 Data Required 4.7-3
4.7.2 Pretreatment of the Emission Stream .... 4.7-4
4.7.3 Condenser System Design Variables 4.7-5
4.7.4 Determining Condenser System Variables . . . 4.7-5
4.7.4.1 Estimating Condensation
Temperature 4.7-8
4.7.4.2 Selecting the Coolant 4.7-9
4.7.4.3 Condenser Heat Load 4.7-9
4.7.4.4 Condenser Size 4.7-13
4.7.4.5 Coolant Flow Rate 4.7-14
4.7.4.6 Refrigeration Capacity 4.7-15
4.7.4.7 Recovered Product 4.7-15
4.7.5 Evaluation of Permit Application 4.7-16
4.7.6 References for Section 4.7 4.7-18
4.8 Fabric Filters 4.8-1
4.8.1 Data Required 4.8-2
4.8.2 Pretreatment of the Emission Stream 4.8-3
4.8.3 Fabric Filter System Design Variables. . . . 4.8-3
4.8.3.1 Fabric Type 4.8-4
ix
-------
TABLE OF CONTENTS
(continued)
Page
4.8.3.2 Cleaning Method .... 4.8-5
4.8.3.3 Air-to-cloth Ratio 4.8-11
4.8.3.4 Baghouse Configuration 4.8-14
4.8.3.5 Materials of Construction 4.8-16
4.8.4 Evaluation of Permit Application 4.8-17
4.8.5 Determination of Baghouse Operating
Parameters 4.8-18
4.8.5.1 Collection Efficiency 4.8-19
4.8.5.2 System Pressure Drop ....... 4.8-19
4.8.6 References for Section 4.8 4.8-20
4.9 Electrostatic Precipitators 4.9-1
4.9.1 Data Required 4.9-2
4.9.2 Pretreatment of the Emission Stream 4.9-3
4.9.3 ESP Design Variables 4.9-3
4.9.3.1 Collection Plate Area 4.9-3
4.9.3.2 Materials of Construction 4.9-6
4.9.4 Evaluation of Permit Application 4.9-6
4.9.5 Determination of ESP Operating Parameters. . 4.9-7
4.9.5.1 Electric Field Strength 4.9-7
4.9.5.2 Cleaning Frequency and Intensity. . 4.9-8
4.9.5.3 ESP Collection Efficiency 4.9-8
4.9.6 References for Section 4.9 4.9-9
4.10 Venturi Scrubbers 4.10-1
4.10.1 Data Required 4.10-2
4.10.2 Pretreatment of the Emission Stream 4.10-3
4.10.3 Venturi Scrubber Design Variables 4.10-3
4.10.3.1 Venturi Scrubber Pressure Drop . . 4.10-4
4.10.3.2 Materials of Construction .... 4.10-4
4.10.4 Sizing of Venturi Scrubbers 4.10-8
4.10.5 Evaluation of Permit Application 4.10-12
4.10.6 References for Section 4.10 4.10-13
-------
TABLE OF CONTENTS
(continued)
Page
Chapter 5: Cost Estimation Procedure 5-1
5.1 Total Capital Cost 5-1
5.1.1 Estimation of Major Equipment Purchase Cost 5-2
5.1.2 Estimation of Auxiliary Equipment
Purchase Cost 5-5
5.1.2.1 Ductwork Purchase Cost 5-5
5.1.2.2 Fan Purchase Cost 5-7
. 5.1.2.3 Stack Purchase Cost 5-8
5.1.3 Estimation of Total Purchased Equipment
Cost 5-9
5.1.4 Estimation of Instrumentation and Controls
Plus Freight and Taxes 5-10
5.1.5 Estimation of Total Purchased Cost .... 5-10
5.1.6 Calculation of Total Capital Costs .... 5-11
5.2 Annualized Operating Costs. . 5-12
5.2.1 Direct Operating Costs 5-13
5.2.1.1 Determine Utility Requirements . . 5-13
5.2.1.2 Determine Remaining Direct
Operating Costs 5-15
5.2.2 Indirect Operating Costs 5-16
5.2.3 Credits 5-17
5.2.4 Net Annualized Costs 5-18
5.3 References for Chapter 5 5-62
-------
TABLE OF CONTENTS
(continued)
Appendices
A.I New York State Air Guide-1 A. 1-1
A.2 Chemical Hazard Information Profiles A.2-1
A. 3 Common Synonyms for Potential HAP's A. 3-1
A.4 Potential HAP's for Solvent Usage Operations A.4-1
A.5 Additional Information for the SOCMI Source Category . . . A.5-1
A.6 Additional Information on Petroleum Related Industries . . A.6-1
A.7 Additional Information on Controls for Process Fugitive
Emissions A.7-1
A.8 Control Techniques for Industrial Process Fugitive
Particulate Emissions (IPFPE) A.8-1
A.9 List of Chemical Dust Suppressants A.9-1
B.I Unit Conversion Factors B.l-1
B.2 Procedures for Calculating Gas Stream Parameters B.2-1
B.3 Dilution Air Requirements B.3-1
B.4 Thermal Incinerator Calculations B.4-1
B.5 Heat Exchange Design B.5-1
B.6 Catalytic Incinerator Calculations B.6-1
B.7 Flare Calculations B.7-1
B.8 Carbon Adsorption Data B.8.1
8.9 Absorption Calcualtions B.9-1
B.10 Condenser System Calculations B.10-1
B.ll Gas Stream Conditioning Equipment B.ll-1
C.I HAP Emission Stream Data Form C.l-1
C.2 Calculation Sheet for Dilution Air Requirements C.2-1
C.3 Calculation Sheet for Thermal Incineration C.3.1
C.4 Calculation Sheet for Catalytic Incineration C.4-1
C.5 Calculation Sheet for Flares C.5.1
C.6 Calculation Sheet fo'r Carbon Adsorption C.6.1
C.7 Calculation Sheet for Absorption C.7-1
-------
TABLE OF CONTENTS
(concluded)
Appendices
C.8 Calculation Sheet for Condensation C.8.1
C.9 Calculation Sheet for Fabric Filters C.9-1
C.10 Calculation Sheet for Electrostatic Precipitators C.10-1
C.ll Calculation Sheet for Venturi Scrubbers C.ll-1
C.12 Capital and Annualized Cost Calculation Worksheet C.12-1
xiii
-------
LIST OF FIGURES
Figure
1-1 Steps used when responding to inquiries 1-4
1-2 Steps used when reviewing permits 1-5
2-1 An example of a partially completed "HAP emission
data form" for one of six HAP emission streams (#1)
generated at a fictitious company 2-7
2-2 Potential emission points for a vacuum
distillation column using steam jet ejectors
with barometric condenser 2-10
3-1 Percent reduction ranges for add-on control
devices 3-4
3-2 Effluent characteristics for emission stream #1 3-28
3-3 Effluent characteristics for emission stream #2 3-29
3-4 Effluent characteristics for emission stream #3 3-30
3-5 Effluent characteristics for emission stream #4 3-31
3-6 Effluent characteristics for emission stream #5 3-33
3-7 Effluent characteristics for emission stream #6 3-34
3-8 Effluent characteristics for a municipal incinerator
emission stream 3-43
4.1-1 Schematic diagram of a thermal incinerator system .... 4.1-2
4.1-2 Supplementary heat requirement vs. emission
stream heat content (dilute stream/no combustion
air) 4.1-9
4.1-3 Supplementary heat requirement vs. emission
stream heat content (no oxygen in emission
stream/maximum combustion air » . . 4.1-12
4.1-4 Heat exchanger size vs. emission stream flow
rate (dilute stream/no combustion air). 4.1-16
4.1-5 Heat exchanger size vs. emission stream heat
content (no oxygen in emission stream/maximum
combustion air) 4.1-18
4.2-1 Schematic diagram of a catalytic incinerator
system 4.2-2
4.2-2 Supplementary heat requirement vs. emission
Stream heat content (dilute stream/no combustion air) . . 4.2-11
xiv
-------
LIST OF FIGURES
(continued)
Figure Page
4.2-3 Supplementary heat requirement vs. emission
stream heat content (no oxygen in emission
stream/maximum combustion air) 4.2-14
4.2-4 Heat exchanger size vs. emission stream heat content . . 4.2-18
4.3-1 A Typical steam-assisted flare system 4.3-2
4.5-1 Adsorption isotherms for toluene/activated carbon
system 4.5-2
4.5-2 A typical fixed-bed carbon adsorption system 4.5-3
4.5-3 Carbon requirement vs. HAP Inlet Concentration 4.5-11
4.5-4 Steam requirement vs. carbon requirement 4.5-16
4.5-1 A typical countercurrent packed column absorber system . 4.6-3
4.6-2 Correlation for flooding rate in randomly-packed
towers 4.6-9
4.6-3 NQQ for absorption columns with constant
absorption factor AF . 4.6-13
4.7-1 Flow diagram for a typical condensation system
with refrigeration 4.7-2
4.7-2 Vapor pressure-temperature relationship 4.7-6
4.10-1 Venturi scrubber collection efficiencies 4.10-5
4.10-2 Psychrometric chart, temp, range 0-500°F, 29.92
in. Hg pressure 4.10-11
5-1 Prices for thermal incinerators, including fan and
motor, and instrumentation and control costs 5-19
5-2 Prices for thermal oxidation recuperature heat
exchangers 5-20
5-3 Prices for catalytic incinerators, less catalyst .... 5-21
5-4 Prices for carbon adsorber packages. Price, includes
carbon, beds, fan and motor, and instrumentation
and controls 5-22
5-5 Prices for custom carbon adsorbers, less carbon. Price
includes beds, instrumentation and controls, and
a steam regenerator 5-23
5-6 Prices for absorber columns, including manholes, skirts,
and painting 5-24
xv
-------
LIST OF FIGURES
(concluded)
Figure Page
5-7 Prices for adsorber platforms and ladders . . 5-25
5-8 Total capital costs for cold water condenser
systems 5-26
5-9 Additional capital cost for refrigerant
condenser systems 5-27
5-10 Prices for negative pressure, insulated fabric filter
systems, less bags 5-28
5-11 Prices for insulated electrostatic precipitators .... 5-29
5-12 Prices for venturi scrubbers, including scrubber, elbows,
separator, pumps, and instrumentation and controls.
Price based on 1/8" carbon steel 5-30
5-13 Required steel thicknesses for venturi scrubbers 5-31
5-14 Price adjustment factors for venturi scrubbers.
For use with Figure 5-12 5-32
5-15 Carbon steel straight duct fabrication price, at
various thicknesses 5-33
5-16 Stainless steel straight duct fabrication price,
at various thicknesses 5-34
5-17 Fan prices 5-35
5-18 Carbon steel stack fabrication price for 1/4"
plate 5-36
5-19 Carbon steel stack fabrication price for 5/16"
and 3/8" plate 5-37
5-20 Completed cost calculation worksheets for the thermal
incinerator example case 5-51
xvi
-------
LIST OF TABLES
Table ' Page
2-1 Source Category Classifications and Information
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
2-11
2-12
2-13
2-14
2-15
2-16
2-17
2-18
2-19
2-20
2-21
2-22
3-1
3-2
3-3
3-4
3-5
Potential HAP's for Solvent Usage Operations
Potential HAP's for Metallurgical Industries
Emission Sources for Metallurgical Industries. ....
Potential HAP's for Inorganic Chemical Manufacturing
Emission Sources for Inorganic Chemical Manufacturing
Industry
Potential HAP's for the Chemical Products Industry . .
Emission Sources for the Chemical Products Industry. .
Potential HAP's for the Mineral Products Industry. , .
Emission Sources for the Mineral Products Industry . .
Potential HAP's for the Wood Products Industry ....
Emission Sources for the Wood Products Industry. . . .
Potential HAP's for Petroleum Related Industries . . .
Potential HAP's for Petroleum Refining Industries. . .
Emission Sources for Petroleum Related Industries. . .
Key Properties for Inorganic Vapor Emissions
Key Emission Stream Characteristics and HAP
Characteristics for Selecting Control Techniques for
Other Considerations in Control Device Selection for
Current Control Methods for Various Inorganic Vapors .
Summary of Control Effectiveness for Controlling
2-15
2-16
2-17
2-19
2-20
2-21
2-25
2-29
2-31
2-32
2-34
2-35
2-36
2-37
2-38
2-40
2-41
2-42
2-50
2-51
2-52
3-3
3-5
3-17
3-21
3-24
xvii
-------
LIST OF TABLES
(continued)
Table
3-6 Key Emission Stream Characteristics for Participate
Emission Streams 3-37
3-7 Advantages and Disadvantages of Participate Control
Devices 3-39
3-8 Control Technology Appplications for Transfer and
Conveying Sources 3-47
3-9 Control Technology Applications for Loading Operations 3-49
3-10 Control Technology Applications for Unloading
Operations 3-50
3-11 Control Technology Applications for Plant Roads. . . . 3-54
3-12 Control Technology Applications for Open Storage Piles 3-56
3-13 Control Technology Applications for Waste Disposal
Sites 3-58
4.1-1 Thermal Incinerator System Design Variables 4.1-6
4.1-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Thermal Incineration . . . 4.1-19
4.2-1 Catalytic Incinerator System Design Variables. .... 4.2-7
4.2-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Catalytic Incineration . . 4.2-21
4.3-1 Flare Gas Exit Velocities 4.3-7
4.3-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Flares 4.3-11
4.5-1 Carbon Adsorber System Design Variables 4.5-8
4.5-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Carbon Adsorption 4.5-21
4.6-1 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Absorption 4.6-19
4.7-1 Coolant Selection ' 4.7-7
4.7-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Condensation ....... 4.7-17
xviii
-------
LIST OF TABLES
(concluded)
Table Page
4.8-1 Characteristics of Several Fibers Used In Fabric
Filtration ..... 4.8-6
4.8-2 Comparisons of Fabric Filter Bag Cleaning Methods. . . 4.8-8
4.8-3 Recommended Air-to-cloth Ratios (Ft/Min) for Various
Dusts and Fumes by Cleaning Method 4.8-12
4.8-4 Factors to Obtain Gross Cloth Area from Net Cloth Area 4.8-15
4.8-5 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Fabric Filters 4.8-18
4.9-1 Typical Values for Drift Velocity for Various
Particulate Matter Applications 4.9-5
4.9-2 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for ESP's 4.9-7
4.10-1 Pressure Drops for Typical Venturi Scrubber
Applications 4.10-6
4.10-2 Materials of Construction for Typical Venturi
Scrubber Applications. . . 4.10-9
4.10-3 Comparison of Calculated Values and Values Supplied
by the Permit Applicant for Venturi Scrubbers 4.10-13
5-1 Identification of Design Parameters and Cost Curves
for Major Equipment 5-38
5-2 C.E. Fabricated Equipment Cost Indices (FE) 5-39
5-3 Unit Costs for Various Materials (June 1985 Dollars) . 5-40
5-4 Price of Packing for Absorber Systems 5-41
5-5 Bag Prices (Dec. 1977 Dollars/Gross Square Feet) . . . 5-42
5-6 Identification of Design Parameters and Cost Curves
for Auxiliary Equipment 5-43
5-7 Assumed Pressure Drops Across Various Components . . . 5-44
5-8 Capital Cost Elements and Factors 5-45
5-9 Unit Costs to Calculate Annualized Cost 5-46
5-10 Utility/Replacement Operating Costs for HAP Control
Techniques . 5-47
5-11 Additional Utility Requirements 5-48
5-12 Estimated Labor Hours Per Shift and Average
Equipment Life 5-50
xix
-------
NOMENCLATURE3
a = packing constant
2
A - heat exchanger surface area, ft
2
Abed = carbon bec* cross sectional area, ft
2
^column = absorber column cross sectional area, ft
2
A = condenser surface area, ft
Anc = net cloth area' ft2
2
A = collection plate area, ft
2
A. = venturi scrubber throat area, ft
2
At = total cloth area, ft
ABS = abscissa (Figure 4.6-2)
AC = adsorption capacity of carbon bed, Ib HAP/100 Ib carbon
2
A/C = air to cloth ratio for baghouse, acfm/ft
AF = absorption factor
b » packing constant
c = packing constant
C = annual credits, $/yr
C = amount of carbon required, Ib
CP^-V. = average specific heat of air, Btu/scf-°F
cll i
average specific heat of air, Btu/lb-mole-°F
Cp = average specific heat of combined gas stream, Btu/scf-°F
-pcoolant » average specific heat of coolant, Btu/lb-°F
Cp - average specific heat of emission stream, Btu/scf-°F
'Cp = average specific heat of emission stream, Btu/lb-°F
Cpf = average specific heat of supplementary fuel (natural gas), Btu/lb-°F
Cpr- = average specific heat of flue gas, Btu/scf-°F
aEnglish units are used throughout this report. Appendix B.I provides
conversion factors for English to Metric units.
XX
-------
-pf|q * avera9e specific heat of flare gas, Btu/lb-°F
£p » average specific heat, of water, Btu/1b-°F
W
CpHAp - average specific heat of HAP, Btu/lb-mole-°F
CE - collection efficiency (based on mass), percent
CRF - capital recovery factor
CRF - weighted average capital recovery factor
W
d - packing constant
D - annual direct labor costs, $/yr
"bed * carbon bed diameter, ft
"column = absorber column diameter, ft
DJ t = duct diameter, in.
D • mean particle diameter, m
Dt - venturi scrubber throat diameter, ft
Dt- • flare tip diameter, in.
2
OG » diffusivity in gas stream, ft /hr
DL = diffusivity in liquid, ft /hr
D, • annual operating labor cost, $/yr
D2 - annual supervision labor cost, $/yr
DE » destruction efficiency, percent
rePor*ed destruction efficiency, percent
DP - stream dew point, °F
ex • excess air, percent (volume)
f » fraction
FE - fabricated equipment cost index
FER » fan electricity requirement, kWh
g » packing constant
xx i
-------
2
gc - gravitational constant, = 32.2 ft/sec
G - gas (emission stream) flow rate, Ib/hr
G, « * gas (emission stream) flow rate based on column cross sectional area,
area lb/sec-ftz
G .. - gas (emission stream) flow rate at flooding conditions based on
' column cross sectional area, Ib/sec-ft
Snol * gas (em
-------
HAP,, m - quantity of HAP in the emission stream exiting the condenser,
°'m Ib-mole/min
HP - fan power requirement, hp (horsepower)
HR - heat recovery in the heat exchanger, percent
HRS - number of hours of operation per year
L - solvent flow rate, Ib/hr
n
L - solvent flow rate based on absorber column cross sectional area,
lb/hr-ftz
L.J « solvent flow rate, gal /mi n
Lmol * so^vent ^ow rate> Ib-mole/hr
Ly - liquid flow rate in venturi scrubber, gal/min
Ly/Qe a - liquid to gas ratio, gal/10 acf
LEL » lower explosive limit, percent (volume)
m « slope of the equilibrium curve
M - annual maintenance costs, S/yr
M - moisture content of emission stream, percent (volume)
M, - annual maintenance labor cost, $/yr
M~ * annual maintenance supervision cost, $/yr
M3 =• annual maintenance materials cost, $/yr
MW - average molecular weight of a mixture of components, Ib/lb-mole
MW - average molecular weight of emission stream, Ib/lb-mole
MWfla * avera9e molecular weight of flare gas, Ib/lb-mole
MW i t - molecular weight of solvent, Ib/lb-mole
MWU.D « molecular weight of HAP (average molecular weight if a mixture of
HAP HAPs is present), Ib/lb-mole
N « number of carbon beds
Nnr - number of gas transfer units (based on overall gas film
Ub coefficients)
xxiii
-------
0? = oxy9en content of emission stream, percent (volume)
ORD - ordinate (Figure 4.6-2)
AP - total pressure drop for the control system, in.H20
2
AP, - absorber column pressure drop, Ib/ft -ft
a
P - emission stream pressure, mm Hg
P rtial - partial pressure of HAP in emission stream, mm Hg
P = vapor pressure of HAP in emission stream, mm Hg
VelpOF
A P. . , = absorber column total pressure drop, in.H-O
AP * pressure drop across venturi, in.H^O
PC = purchased equipment cost, $
Q - flow rate of gas stream at actual conditions, acfm
cL
Q » combustion air flow rate, scfm
Qcom = flow rate of combined gas stream entering the catalyst bed, scfm
^coolant = coolant flow rate> Whr
Q i * cooling water flow rate, Ib/min
Q - emission stream flow rate, scfm
Q = saturated emission stream flow rate, acfm
e, s
Qf - supplementary fuel (natural gas) flow rate, scfm
Qfa " f ue 9as ^ow rate» sc^m
Qr_ , = flue gas flow rate at actual conditions, acfm
i g» a
- flare gas flow rate, scfm
- flare gas flow rate at actual conditions, acfm
,a
~ quantity of HAP recovered, Ib/hr
Q = steam flow rate, Ib/min
Q = cooling water flow rate, gal/min
W
r = packing constant
XXIV
-------
R * gas constant, * 0.73 ft -atm/lb-mole °R; - 1.987 cal/g-mole °K
Rhum - relative humidity, percent
Ref * refrigeration capacity, tons
RE * removal efficiency, percent
RE t j - reported removal efficiency, percent
s * packing constant
S » annual cost of operating supplies, $/yr
SCg » Schmidt number for HAP/emission stream
Sc, * Schmidt number for HAP/solvent system
St - steam ratio, Ib steam/1 b carbon
SV - space velocity, hr
t - cleaning interval, min
t - residence time, sec
T » temperature, °F
T * combustion temperature, °F
T . - temperature of combined gas stream entering the catalyst bed, °F
T - temperature of flue gas leaving the catalyst bed, °F
Tcon = condensation temperature, °F
T i . • inlet temperature of coolant, °F
T - outlet temperature of coolant, °F
T - emission stream temperature, °F
T. c - temperature of saturated emission stream, °F
e, s
Via " flare 9as temperature, °F
L - emission stream temperature after heat exchanger, °F
T =» reference temperature, - 70°F
Tsti * inlet steam temperature, °F
xxv
-------
T . = condensed steam outlet temperature, °F
\M * inlet cooling water temperature, °F
Wl
Tum = outlet cooling water temperature, °F
WO
AT... » logarithmic mean temperature difference, °F
Th , » absorber column thickness, ft
U = overall heat transfer coefficient, Btu/hr-ft2-°F
U. = drift velocity of particles, ft/sec
U. t * velocity of gas stream in the duct, ft/mi n
U » emission stream velocity through carbon bed, ft/min
Uo c = throat velocity of saturated emission stream, ft/sec
e, s
IJflq = flare gas exit velocity, ft/sec
Um*v = maximum flare gas velocity, ft/sec
IUaA
Ut = annual utility costs, $/yr
VG = combustion chamber volume, ft
Vcarbon = vo1ume of carbon Ded> ft
V.j = catalyst bed requirement, ft
V . . = absorber column packing volume, ft
W = particle grain loading, gr/acf
Wtcolumn = a'3sor')er column weight, Ib
x = mole fraction of solute in solvent, moles solute/(moles solute +
moles solvent)
X = mole fraction of gaseous component in liquid, moles solute/ moles
solvent
y = mole fraction of solute in air, moles solute/ (moles solute + moles
air)
Y = packing constant
Y = mole fraction of solute in air, moles solute/moles air
= carb°n be<* depth, ft
xxv i
-------
« » packing constant
x» latent heat of vaporization for steam, Btu/lb
*j - fan efficiency, percent
P.. = density of carbon bed, Ib/ft
ft - density of carbon steel plate, Ib/ft
PG =» density of gas (emission stream), Ib/ft
p. * density of solvent, Ib/ft
9ads * cycle time ^or adsorPtion, hr
ereq * cyde time ^or regeneration, hr
M, - viscosity of solvent, centipoise
ML" - viscosity of solvent, Ib/ft-hr
xxv n
-------
ACKNOWLEDGEMENTS
The authors express their appreciation to Dr. Bruce A. Tichenor, EPA
Project Officer, for his advice and technical support throughout this project.
We also wish to acknowledge the following persons for their assistance in
producing various sections of this manual: Mr. Vishnu S. Katari,
Ms. Karin C. C. Gschwandtner, Mr. Michael K. Sink, and Ms. Charlotte R. Clark
of Pacific Environmental Services, Inc.; and Mr. Andrew J. Miles, Mr. D. Blake
Bath, and Ms. Glynda E. Wilkins of the Radian Corporation.
xxv m
-------
CHAPTER 1
INTRODUCTION
1.1 OBJECTIVE
The objective of this manual is to present a methodology for
determining the performance and cost of air pollution control techniques
for reducing or eliminating the emission of potentially hazardous air
pollutants (HAP's) from industrial/commercial sources. [Note: The
term "hazardous" in this manual is very broad. It is not limited to the
specific compounds listed under current regulations (i.e., the Clean
Air Act, the Resource Conservation and Recovery Act, and the Toxic
Substances Control Act).] This manual is to be used by EPA regional,
State, and local air pollution control agency technical personnel for
two basic purposes: (1) to respond to inquiries from interested parties
(e.g., prospective permit applicants) regarding the HAP control require-
ments that would be needed at a specified process or facility, and
(2) to evaluate/review permit applications for sources with the potential
to emit HAP's. It should be noted that this manual provides general
technical guidance on controls and does not provide guidance for compli-
ance with specific regulatory requirements for hazardous air pollutants.
Specifically, the manual does not specify design requirements necessary
to achieve compliance with standards established under specific programs
such as Section 112 of the Clean Air Act or standards established under
the Resource Conservation and Recovery Act. Such requirements vary
with the hazardous air pollutant emitted and with the emission source;
thus, regulatory-specific detailed specifications are beyond the
scope of this manual.
Section 1.2 discusses the use of this manual. Chapter 2 assists
the user in identifying HAP's and their respective emission sources.
Chapter 2 also identifies the key emission stream characteristics
necessary to select appropriate control techniques. Chapter 3 provides
additional information to assist the user in the control technique
selection process for each HAP emission source/stream. Chapter 4
1-1
-------
presents simple step-by-step procedures to determine basic design
parameters of the specific control devices and auxiliary equipment.
Chapter 5 provides the necessary data a-nd procedures to determine
order-of-magnitude estimates (-60 to +30 percent) for the capital
and annualized costs of each control system.
There are numerous appendices that provide pertinent information
not found in the main text. The appendices are divided into three groups.
Appendices in "Group A" present supplementary data that clarify/expand
the information discussed. For example, Appendix A.I contains the "New
York State Air Guide - 1," which is presented to illustrate (this is not
an endorsement) one of several methods (other State agencies utilize
somewhat different approaches) that could be used when developing a HAP
program. "Group B" appendices contain derivations of equations, calculation
procedures, and unit conversion techniques for emission stream physical
characteristics. Blank worksheets to be used while performing the
functions of this manual are found in the "Group C" appendices. These
worksheets are masters; thus, copies should be used.
A good-source of current information pertaining to HAP's is the
"Air Toxics Information Clearinghouse," which was established by EPA in
response to State and local agency requests for assistance in the
exchange of information on toxic air pollutants. The Clearinghouse is
operated by EPA's Office of Air Quality Planning and Standards (OAQPS)
in close coordination with the State and Territorial Air Pollution
Program Administrators (STAPPA) and the Association of Local Air Pollu-
tion Control Officials (ALAPCO). The Clearinghouse collects and
disseminates information from State and local agencies, as well as
making users aware of air toxics information available from EPA and
other Federal agencies. Specifically, the Clearinghouse collects the
following air toxic information from State and local agencies: regula-
tory program descriptions, acceptable ambient concentrations on ambient
standards, toxic pollutant research, source permitting, ambient
monitoring, toxicity testing, and source testing.
The Clearinghouse provides an on-line data base containing all
toxic-related information submitted by State and local agencies,
bibliographic citations for relevant reports by EPA and other Federal
agencies, and references for ongoing EPA air toxic projects. It also
1-2
-------
publishes a quarterly newsletter with articles on current air toxics
concerns. Finally, the Clearinghouse periodically publishes various
special reports on topics of- interest to users. For further informa-
tion regarding the "Air Toxics Information Clearinghouse," contact the
appropriate EPA regional office air toxics contact, or EPA/OAQPS,
Pollutant Assessment Branch, MD-12, Research Triangle Park, North
Carolina 27711. (919) 541-5645 or FTS 629-5645.
1.2 HOW TO USE THE MANUAL
Figure 1-1 is a flowchart of the steps performed when responding
to inquiries; Figure 1-2 presents the same type of flowchart when
reviewing permits. As shown by these figures, these two functions are
basically the same; the only substantive difference is that the review
process also compares the determined/calculated parameters with the
corresponding parameters stated in the permit application to ensure that
the control system(s) proposed by the applicant will provide the required
reduction of HAP emissions.
Once an inquiry or permit application is received, determine the
HAP's applicable to the source category in question (Section 2.1).
The HAP's are categorized under four headings: organic vapor, organic
particulate, inorganic vapor, and inorganic particulate. [Mote: For
each HAP group, a list of potentially or suspected hazardous compounds
that may be emitted as a HAP from the source category is provided.
This listing is neither all-inclusive nor a declaration that the
compounds presented are hazardous.] Next, identify the potential
emission sources for each HAP group (Section 2.1). The HAP emission
sources are listed under one of three classifications: process point
sources, process fugitive sources, and area fugitive sources. [Note:
See Section 2.1 for classification definitions.] After each emission
source is determined, identify the key HAP emission stream characteristics
(e.g., HAP concentration, temperature, flow rate, heat content, particle
size) needed to select the appropriate control technique(s) (Section 2.2).
Obtain the actual values for these characteristics from the owner/operator
or from available literature if the owner/operator cannot provide the
necessary data. If two or more emission streams are combined prior to
-
-------
Manual Location to
Perform Step
Section 2.1
2Sect1on 2.2
3 None
4 Appendix B.2
5ttone
6Chapter 3
7Chapttr 4
8Chapter 5
Information Requested
on HAP Control for
a Specific Facility
1
. Obtain Available
• Plant-Specific Data
1
Define MAP'S1
i
Define Emission Sources
Generating HAP's1
i
Define Characteristics
Needed for Each HAP
Emission Stream?
1
Combine HAP Streams3
i
Define Characteristics
Of Combined Streams1
i
^Eacn Single/Combined ^
^ HAP Emission Stream j
I
Define HAP Control
Requirements^
• 1 Inquirer Assistance
Control Numerical Technology Other
Device Limit Forcing 1 Requirement
» ' '., „ • ' *
Select Appropriate __ Determine R
Control Technique! s)6 control tffi
;
/ *r* \
/Basic Design Parameter s\ No
\ Requested /~
\ 1 /
equlred / Is \ Regulation Agency '
ciency5 Yes / Cost of Control a \_Np Policy Decision i
\ ' /
, ,,
Agency Determination
of Cost Constraints5
Select Appropriate
Control Technlque(s)6
i Yes i 1
Determine Basic Design
Parameters for
Control System(s)7
Select Appropriate
Control Technlquels)6
1
Determine Basic Design
Parameters and Cost for
Control System! s)'«8
i
Select Control System
Having Most Stringent
Level of Control Ml thin
Given Cost Constraints
i
Recomnend Appropriate
Control Technlque(s)
i
/ Last \
/ HAP Emission \
\ Stream /
\ ? /
i Y«
Recommend HAP
Control Program
Determine Basic Design
Parameters for
Control System(s)'
i
Select Control System 1
Having Most Stringent '
Level of Control
l
i
^No
Figure 1-1. Steps used when responding to Inquiries.
1-4.
-------
Pern-it Application
for Review/Approval
Manual Location to
Perform Step
1on Z.I
^Section 2.2
3None
4Append1x 8.2
5None
6Chapter 3
7Chapter 4
8Chapter 5
HI Hff's
Addressed1
No
Obtain Additional Data
from Applicant
iw-v \
All HAP Emission, \ No
Sources Addressed /
Yes 1.
Obtain Additional Data
from Applicant
Arc
All HAP Emission \ No
Stream Characteristics)
Provided?
Obtain Additional Data
from Applicant
Yes
Arc
Any HAP
Emission Strews
Combined3
Yes
Are
Combined Stream
Characteristics
Correct4
No
Yes
c
Each Single/Combined
HAP Emission Stream
Define HAP Control
Requirements5
Control
Device
Numerical
Llnlt
Select Appropriate
Control Techn1que(s)6
Obtain Additional Data
from Applicant
Determine Required
Control Efficiency5
_L
Other
Requirement
Technology
Forcing
Regulatory Agency
Policy Decision5
Determine Basic Design
Parameters for
Control System(s)7
Is
Cost of Control a
Decision Factor
7
TTo
Is
Permit Control System \ No
Appropriate /
\ ? /
Recommend Appropriate
Control Technlqucis)
JJfL
' Is
Penan System Design \_Nj
Appropriate
Recommend Appropriate
Basic Design Parameters
Yes
Agency Determination
of Cost Constraints'
Select Appropriate
Control Technique!s)°
Select Appropriate
Control Technlque(s)6
3.
Determine Basic Design'
Parameters for
Control System(s)7
Determine Basic Design
Parameters and Cost for
Contro] System(s)'-3
Select Control System I
Having Most Stringent
Level of Control
Select Control System
Having Most Stringent
Level of Control W1tMn
Given Cost Constraints
\
Last
HAP Emission
Stream
TYes
\ No
P.ermlt Approval or
Provide Recommendations
Figure 1-2. Steps used when reviewing permits.
1-5
-------
entry into an air pollution control system, determine the characteristics
of the combined emission stream (Appendix B.2).
Depending upon the specific regulation and the type/characteristics
of the HAP emission source/stream, the remaining steps in the methodology
will differ. There are four basic "formats" for a regulation: (1) a
particular "control device" may be required, (2) a "numerical limit"
may be specified, (3) a "technology forcing" requirement may be imposed,
and (4) a specific work practice or "other" related practice may be
required. The regulation format will define the steps that lead to the
selection of the appropriate control technique(s). The "control device"
and "other" formats specify the appropriate control technique(s). A
"numerical limit" format requires the determination of the HAP removal
efficiency before the appropriate control technique(s) can be identified.
Lastly, the "technology forcing" format has two paths: one where the
cost of the control system is a factor in the decision, and one where
cost is not a factor. If control system cost is a factor, the agency
must determine the cost constraints that will be imposed on the control
technique selection process. The steps that occur in defining the HAP
control requirements will depend upon each agency's regulatory policies.
The HAP emission stream characteristics, in conjunction with the
limitations imposed by the applicable regulations, are used to select
the appropriate control techniques (Chapter 3) for each HAP emission
source/stream. General guidelines are provided that match specific
control devices with specified emission stream properties (e.g., HAP
content, temperature, moisture, heat content, particle size, flow rate).
Basic design parameters are then determined to provide general design
conditions that should be met or exceeded for each selected control
technique to achieve the specified HAP removal efficiency (Chapter 4).
This exercise also identifies which of the selected control techniques
will not achieve the desired HAP control requirements. The basic design
parameters also can be used to obtain an order-of-magnitude cost
estimate for each control device (Chapter 5). As noted above, this
cost information can be an integral part of the HAP control system
selection process. After completing the above process, a HAP control
program can be recommended or evaluated.
1-6
-------
• EXAMPLE CASE
To guide the user through the steps and calculations
described in this manual, examples are provided throughout
the text. As shown here, each example is always highlighted
by a surrounding box. The primary example case pertains
to a hypothetical plant owner requesting assistance in
determining the type of control system that should be used
on an emission stream generated by a paper coating drying
oven. This example is carried through the entire manual.
Additional example emission streams are introduced in
Chapter 3 to illustrate fully the control technique selec-
tion process and to clarify the design procedures of
Chapter 4.
1-7
-------
CHAPTER 2
HAP EMISSIONS AND THEIR KEY PHYSICAL PROPERTIES
This chapter's primary goal is to identify the following:
(1) potential HAP's for a given source category and the specific
sources that may emit the potential HAP's - Section 2.1, and (2) key
emission stream physical properties needed to select appropriate control
strategies and size control devices for the HAP emission sources -
Section 2.2. Specific source categories are divided into nine general
classifications in this manual. [Note: The general classification
system is a hybrid of the classification systems used by the following
documents: (1) U.S. EPA, Compilation of Air Pollutant Emission Sources
- Third Edition: Supplements 1-15, AP-42, January 1984; and (2) the
U.S. EPA, BACT/LAER Clearinghouse - A Compilation of Control Technology
Determinations, April 1983.] Every possible source category cannot be
listed; however, similarities exist between many categories. Thus, the
user should be able to obtain some guidance for any specific facility.
Only those source categories that are known to emit potential HAP's are
presented in this manual.
The specific source categories listed under the general classifica-
tions are identified in Sections 2.1.1 through 2.1.9; the list of refer-
ences used in Section 2.1 is presented in Section 2.1.10. Table 2-1
presents the nine generic source category classifications and the
location of specific information within Section 2.1. For further
information, Appendix A.2 presents a listing of chemical hazard
information profiles (CHIP's) and CAS numbers. Individual source
categories have been classified based on the manufacturing process
associated with emissions of potential HAP's. The Solvent Usage
Operations classification includes processes dependent on solvents,
such as surface coating and dry cleaning operations. Metallurgical
Industries include processes associated with the manufacture of
metals, such as primary aluminum production. Processes and opera-
tions associated with the manufacture of organic and inorganic
2-1
-------
oo
o
^_
^£
o
o
o
^—
a:
Oi
O
u_
z
i— i
Q
**
OO
Z.
0
^
o
1-^
u_
,.4
oo
oo
J
o
>-
o
CD
LU
1—
O
LU
CJ
Q£
o
oo
•
e— 1
|
CM
LU
CD
^£
1—
Z
o
l_^
I—
u
•z.
o
*— 4
1—
^c
^?*
OfT
o
u.
z
1— • 1
in
a. cu
< u
X C-
3
i— O
in
O
i— O
a. cu
CLOO
<
oo
z
o
IM^
1—
CJ
n— •!
u.
*•_!
C/}
OO
5
0
VO
r-t
1
CM
«
. a.
A
CO
1
CM
cu
r—
.Q
ro
1—
if)
i—i
1
CM
•
a.
«<
CM
1 X
CM •!-
•a
cu c
r- 0)
^Q ^^
TO QL,
l— <
U1
i
CM
a.
A
f— <
•
*— 1
CM
C
O
^j
(J
OJ
OO
c
cu
o
rjn
!— 4
1
CM
•
a.
-
""
A
'•fi
1
csa
OJ
^«
^
i—
P^
i— i
i
CM
•
a.
M
>
'i +3
CU CO
J= 3
o -a
c
o *-•
•r^
c a»
3^ '^~
&. IM
O 3
4^
U U
•(-J ^—
CU 3
j= C
c S
oo
LO
CM
1
CM
•
a. .
A
CO
1
CM
CU
)^—
r^
(O
1 —
^H
CM
1
CM
•
a.
A
f^^
1
CM
ai
r**
J2
t—
r— t
i-H
1
CM
a.
A
^"
•
^1
CM
C
O
+-J
u
cu
oo
"-Fl
C
.p—
t_
^
4J
u
ro
<*-
3
ro
z:
pM~
^
O (-
•r~ -M
C 10
ro 3
O^ "O
1_ C
O i-i
c
>— '
r-l
oo
1
CM
9
a.
A
o
r— 1
1
CM
CU
c*s
ro
1 —
CTi
CM
1
CM
•
a.
A
cn
i
CM
a>
^3
ro
I—
r- 4
r— t
1
CM
a.
A
LO
•
i— I
CM
C
C
+J
a
cu
oo
^^
t_
4.J
1/5
3
•o
C
co
^_>
O
3
•a
o
t_
a.
^^
rO
O
•r*
E
cu
CJ
^J-
ro
i
CM
•
a.
A
CM
i— I
1
CM
CU
r*"
.a
ro
1—
CM
CO
CM
•
a.
^^
<— i
CM
(U
^»
_Q
ro
t—
CM
i— l
1
CM
a.
A
tJD
•
•"•
CM
C
O
.JJ
o
cu
oo
^1^
t_
+J
in
3
•a
c
»— 4
co
4-1
U
3
•a
o
c_
a.
p~
ro
£»
C
•r~
Z
VO
00
1
CM
»
a.
A
^
r— 4
1
CV4
OJ
»—
o
ro
LO
CO
|
CVJ
•
a.
CO
»— i
i
CM
^
^»
f*^
^
c_
4^
I/I
3
-^
C
1/1
1 <
O
3
"8
t_
a.
•o
o
o
2
0
|
CM
•
a.
A
p»s.
r-H
1
CM
CD
f— •
f*>
ro
1—
P~- CO
co ro
i i
CM CM
• •
a. a.
Lfi vo <:
t-4 r-t
1 1 X
CM CM -r-
^3
cu cu c
r-— « f*-' CJJ
J3 -Q Q-
(O ^3 ^^>
i— t— <
CM
1—4
1
CM
Q.
A
CO
•
^H
CM
C
O
^J
(_)
u
oo
CO
cu
•r-
t_
4_l
CO
3
•a
c
^3
cu
ro
f—
1)
a:
E
3
cu
*-^
o
£_
CU
a.
CM
^^
i
CM
•
Q.
A
CT>
i— i
1
CM
CU
^»
i^
ra
1—
r— 1
r£*
1
CM
•
a.
CO
r-l
1
CM
CU
^~
jQ
(t3
H-
^.
^^
|
CM
2.
A
cy^
•
r-4
CVJ
c
o
4J
U
-------
chemicals have been grouped into the Synthetic Organic and Inorganic
Chemical Manufacturing classifications, respectively. Industries using
chemicals in the formulation of products are classified as Chemical
Products Industries. The Mineral and Wood Products Industries classifi-
cations include operations such as asphalt batch plants and kraft pulp
mills, respectively. The Petroleum Related Industries classification
is defined as oil and gas production, petroleum refining, and basic
petrochemicals production. Combustion Sources are utility, industrial,
and residential combustion sources using coal, oil, gas, wood, or
waste-derived fuels.
To assist the manual user in recording the pertinent information,
a worksheet has been provided. A copy of this worksheet, the "HAP
Emission Stream Data Form," is presented in Appendix C.I. An example
of a partially completed worksheet is shown in Figure 2-1 on page 2-7.
This worksheet is designed to record information pertaining to one
emission stream, be it a single stream or a combined stream consisting
of several single streams.
EXAMPLE CASE
Information has been requested by a paper coating
plant owner regarding the control of an emission stream
from his facility's drying operations. The most likely
generic classification listed in Table 2-1 to include a
paper coating plant (one of many surface coating industries)
would be Solvent Usage Operations. To determine if a
paper coating facility is listed within this category,
Section 2.1.1 and Table 2-2 should be reviewed (as directed
by Table 2-1). An inspection of Table 2-2 indicates that the
initial choice was correct (i.e., paper coating is listed
under SC-Paper, Tapes, Labels) and the information retrieval
process can begin.
2-3
-------
2.1 IDENTIFICATION OF POTENTIAL HAP'S AND EMISSION SOURCES
The purpose of this section is to-present general information on
emissions of potential HAP's by source category. Within each of the
nine general classifications, information is presented on the types of
potential HAP's that may be emitted by a particular source category.
This information includes the names of specific compounds, the classifi-
cation of the compounds (i.e., organic or inorganic), and the form
in which these compounds would be emitted (i.e., vapor or particulate).
Owing to process variations, actual emissions from specific facilities
may differ from the general information presented. Complete identifica-
tion of HAP emissions is best accomplished with assistance from the
owner/operator of the facility. Since many of the potential HAP's are
also known by other names, Appendix A.3 provides a cross reference
table where trade names and common synonyms for HAP's are listed.
This section also presents information pertaining to the sources
(e.g., processes) within each specific source category that have the
potential to emit HAP's. In this manual, emission sources are broadly
classified into three groups: process point sources, process fugitive
sources, and area fugitive sources. Point sources are, in general,
individually defined. Reactors, distillation columns, condensers,
furnaces, and boilers are typical point sources which discharge emissions
through a vent-pipe or stack. These sources can be controlled through
the use of add-on control devices. Process fugitive sources, like
process point sources, are individually defined. Emissions fron these
sources include dust, fumes, or gases that escape from or through
access ports and feed or discharge openings to a process (e.g., the
open top of a vapor degreaser). Process fugitive sources also can
include vent fans from rooms or enclosures containing an emissions
source (e.g., a vent fan on a dry cleaner). These sources can be
controlled by add-on control devices once the emissions are captured by
hooding, enclosures, or closed vent systems and then transferred to a
control device.
Area fugitive sources are characterized by large surface areas
from which emissions occur. In addition, process equipment such as
pumps, valves, and compressors are considered area fugitive sources;
2-4
-------
emissions from these sources occur through leaks during process operation.
Although these sources are small, they are usually found in large
numbers dispersed over a wide area in a given process. Area fugitive
sources also include large and undefined emission sources such as waste
treatment lagoons, raw material storage piles, roads, etc.
The sources listed in this manual generate emissions; however, a
definitive statement as to whether they emit a HAP cannot be made. As
in the case of identifying the potential HAP's emitted at a specific
facility, communication with the owner/operator is useful in identifying
each source that emits a HAP. Do not assume the listings found in this
section are all-inclusive; a specific facility may have an emission-
producing operation that is not common to its industry and, thus, the
source may not be included here. The references used in compiling the
"Potential HAP's" table and the "Emission Sources" table for each general
source classification are shown on each source's respective "Potential
HAP's" table.
2.1.1 Solvent Usage Operations
Solvent usage operations are defined as manufacturing processes
that use solvents, including such processes as surface coating operations,
dry cleaning, solvent degreasing, waste solvent reclaiming, and graphic
arts. Table 2-2 (page 2-15) lists source categories within this group
of operations that have been identified as sources of volatile organic
compound emissions that may include potential HAP's. As is shown by
Table 2-2, all solvent usage operations generate organic vapor emissions.
[Note: Some of the emission sources generate aerosols (i.e., organic
particulate); however, the aerosols evaporate in a short time and the
emissions normally are controlled as a vapor. Therefore, Table 2-2
does not indicate the presence of organic particulates.] Due to the
large number of potential HAP's associated with these types of operations,
tne format of Table 2-2 prohibits the inclusion of compound-specific
data. Potential HAP's that may be emitted by sources in Table 2-2 are
summarized in Appendix A.4; this appendix lists both specific compounds
and classes of compounds that may be emitted by sources within the
category. Appendix A.4 can be used to determine whether a particular
2-5
-------
solvent usage operation may emit a specific potential HAP or yroup of
potential HAP's, as well as to determine all solvent use operations
that may emit a particular potential HAP. Table 2-3 (page 2-16) presents
the emission sources that may emit potential HAP's.
EXAMPLE CASE
As directed by Section 2.1.1, Appendix A.4 is used to
determine the potential HAP's. The potential HAP's for paper
coating operations are as follows:
Specific Compounds Generic Compounds
toluene mineral spirits
xylene other aromatics
ethylene glycol alcohols
acetone cellusolves
methyl ethyl ketone ketones
methyl isobutyl ketone esters
ethyl acetate
Upon reviewing data from the solvent vendor, the owner
determined that only toluene is present in the solvent
being evaporated by the ovens. Table 2-2 indicates that
toluene is an organic compound, and it would be emitted as
a vapor. This information is then listed on the "HAP
Emission Stream Data Form" provided in Appendix C.I (see
Figure 2-1).
Also for the example case, the source (oven: a process
point source) was identified. If the user were interested
in the other emission-releasing processes at a paper coating
operation, Table 2-3 indicates that the remaining sources
include solvent transfer, solvent storage, application
areas, and solvent/coating mixing.
2.1.2 Metallurgical Industries
The metallurgical industries can be broadly divided into primary,
secondary, and miscellaneous metal production operations. The majority
2-6
-------
CM
O
01
•o
§
in
0) O g
§ " "
•C >> V.
& u O
Ol C
^ O) •
oi g> o
i— < z
01—01
01 >! W
i. w , I
V)
3
X
UJ
01
S
2
5j
c
o
(9
U
<*-
1
c
a.
^^
,_
Ol
1
i
01
t/1
e
o
VI
5
<
i
i^— •
,
^
o
0)
e
^
o
01
S
a
IB
U
O
c
o
I/I
UJ
1
ca
i
«««
i
.a
e
Q
rocess
a
IB
c
o
10
u
%
u
S
1
>1^
J=
oi
Ol
3
b.
3
U
01
i
1
-i
't
Ol
5
o
§
a
»
«J
S
§
a
5
41
*
c
OJ
I
u
c
01
1.
o
a.
M
CM
^^
m.
CM
*"*
C
01
c
S
01
L.
3
V)
i
3£
•
"o
»*
o
CM
I/I
3
CD
^
O
^•t
C
01
c
cS
^
IB
3!
O
1
«
^
01
b.
3
IB
Ol
1
_l
^—
(*1
e
01
e
8
01
IB
3
U
a.
ee.
X
IB
g
S
U
in
LA
CM
—•
01
ai
X
o
u.
JC
^^
PO
o
c
(0
n
0)
u
a.
W1
u
L.
01
a
1
_ __
M
.*^*,
01
3
(/t
01
U
a.
z
1
3
^ ^
01
01
0
c
0
^
f-
3
3-
3.
u)
|
4O
S
2
4*1
C
3
^
^
01
"J
VI
3
O
u
e • •
o -o c
U 01 O
01 —
•oc in
Oi •• in
«j in 3 •*-
u ~- uj E
oi j=3 ai
Oi TS m
in oi e in
•U IB 01
01 O u
j= e* o
w U u
133 a.
o c
.M IB (/> 01
C 3.W
^" A^JS
>» in irt 3
L. - U
<-> CL C"-
C < — *J
01 X U.
IB IB
o oi w a •
w Ol > .a
c in -o E
"^ * ° n
S IB Ul OX
« a. L. i
oi in IB o
^ 4^ C VI
I/I IB ^3 01
C II —
O •*- IB CM m
*^> W £ 01
in 5 eT2
•»- -a ai o —
Al Q in in
c ~- oi »- ^
IB ^ T3 | |
U § "" in °
«*. o wi wi S
Ol J£ U 01
in
3. . U— 5
sji'i oi-^
IB 13 3 t. C
^ O C O V
£ O)
Ol w 01 " t-
J= Ol J= S
>-=!——•«
< *
Of
o
(O
= Q.
VI
4-> O
•0 t-
a -4-1
C 4J
o o
•!•• »r—
WI «*-
VI
•g-
LU 4->
10
01
T3 S-
£X
r- I/I
no
-------
of this industry is covered under SIC Codes 331, 332, 333, 334 and 336.
The term primary metals refers to production of the metal from ore.
The secondary metals industry includes-the recovery of metal from
scrap"and salvage and the production of alloys from ingots. The mis-
cellaneous subdivision includes industries with operations that produce
or use metals for final products. Table 2-4 (page 2-17) presents the
potential HAP's for these industries and Table 2-5 (page 2-19) presents
the industry-specific emission sources.
2.1.3 Synthetic Organic Chemical Manufacturing Industry (SOCMI)
The SOCMI is a large and diverse industry producing several thousand
intermediate.and end-product chemicals from a small number of basic chemi-
cals. Most of the chemicals produced by this industry fall under SIC Code
286. Due to the complexity of the SOCMI, a general approach is used in
this section to describe generic emission sources and specific emission
source types. This approach is identical to the approach used by EPA
in its efforts to develop new source performance standards for the SOCMI.
A large proportion of the emissions from the SOCMI occur as organic
vapors. However, organic particulate emissions may be generated in
some processes (usually during the manufacture of chemicals that exist
as solids at ambient conditions). The emissions typically contain raw
materials (including impurities) used in and intermediate and final products
formed during the manufacturing process. Many of these emission streams may
contain HAP's; however, due to the great number of compounds manufactured
in the SOCMI, consult the references listed in Table A.5-1 (Appendix A.5)
to determine the specific emissions for different SOCMI processes.
Potential emissions from this industry can be described generically
as follows:
(a) storage and handling emissions,
(b) reactor process emissions,
(c) separation process emissions,
(d) fugitive emissions, and
(e) secondary emissions (e.g., from waste treatment).
Emissions can potentially occur from raw materials and product storage
tanks as working and breathing losses through vents. Emissions from
2-8
-------
handling result during transportation or transfer of the volatile
organic liquids. Reactor processes and separation processes are the
two broad types of processes used in manufacturing organic chemicals.
Reactor processes involve chemical reactions that alter the molecular
structure of chemical compounds. A reactor process involves a reactor,
or a reactor in combination with one or more product recovery devices.
Product recovery devices include condensers, adsorbers, and absorbers.
Emissions from reactor processes occur predominantly from venting of
inert gases from reactors and product recovery devices, or from the
release of organic compounds that cannot be recovered economically.
Typical emission sources in reactor processes include point sources
(e.g., vents on reactors and product recovery devices), process fugi-
tive sources such as disposal of bottoms from the reactor or the
product recovery devices, and area fugitive sources (e.g., pumps,
valves, sampling lines, and compressors). Some of the reactor
processes that contribute to the majority of emissions from the SOCMI
are listed in Table A.5-2 (Appendix A.5). These processes are further
broken down in Tables A.5-3 and A.5-4 (Appendix A.5) by the type of
chemical produced.
Separation processes often follow reactor processes and divide
chemical product mixtures into distinct fractions. Emissions from
separation processes are associated primarily with absorption,
scrubbing, and distillation operations. Other separation processes
that nay contribute to emissions include drying, filtration, extraction,
settling, crystalization, quenching, evaporation, ion exchange, dilu-
tion, and mixing/blending. One of the more commonly employed separation
techniques is distillation. Depending on the type of distillation
system used (i.e., vacuum or nonvacuum), typical emission points can
include condensers, accumulators, hot wells, steam jet ejectors, vacuum
pumps, and pressure relief valves. Emission points from a vacuum
distillation system are shown in Figure 2-2.
Although fugitive emissions are listed as a separate group, they
can occur from storage and handling, reactor processes, and separation
processes. Area fugitive sources include groups of valves, pressure
relief devices, pumps and compressors, cooling towers, open-ended lines,
and sampling systems. Process fugitive sources include hotwells,
2-9
-------
s.
o
u
U
t/1 S»
•<- 3
•o o
u
aj u
•M at
o •'->
a. a;
OsJ
i
CM
OJ
u
c
O
-------
accumulators, and process drains from reactors, product recovery
devices, and separation equipment.
Table 2-6 (page 2-20) presents information on specific emission
points and emission source types for each of the generic emission
source groups. Using this information, the user can identify the
potential emission sources pertaining to his specific situation. An
example case is provided in Appendix A.5 that illustrates the approach
outlined above and shows the emission sources and the HAP's potentially
emitted from a SOCMI process (ethylbenzene/styrene production).
2.1.4 Inorganic Chemical Manufacturing Industry
This industry includes the manufacture of the basic inorganic
chemicals before they are used in the manufacture of other chemical
products. Most of the chemicals produced by this industry fall under
SIC Code 281. Potential emissions from these processes may be high,
but because of economic necessity they are usually recovered. In some
cases, the manufacturing operation is run as a closed system, allowing
little or no emissions to escape to the atmosphere. Table 2-7
(page 2-21) presents the potential HAP's for these industries and
Table 2-8 (page 2-25) presents the industry-specific emission sources.
2.1.5 Chemical Products Industry
This industry includes the manufacture of chemical products, such
as carbon black, synthetic fibers, synthetic rubber and plastics, which
may be used in further manufacture. Also included are the manufacture
of finished chemical products for ultimate consumption such as Pharma-
ceuticals, charcoal, soaps and detergents; or products to be used as
materials or supplies in other industries such as paints, pesticides,
fertilizers and explosives. Most of the chemical products are covered
under SIC Codes 282, 283, 284, 285, 287 and 289. As in other chemical
industries, the potential emissions from these processes may be high,
but because of economic necessity they are usually recovered. Table 2-9
(page 2-29) presents the potential HAP's for this industry and Table 2-10
(page 2-31) presents the industry-specific emission sources.
2-11
-------
2.1.6 Mineral Products Industry
This industry involves the processing and production of various
nonmetallic minerals. The industry includes cement production, coal
cleaning and conversion, glass and glass fiber manufacture, lime manu-
facture, phosphate rock and taconite ore processing, as well as various
other manufacturing processes. A majority of this industry falls under
SIC Codes 142, 144, 145, 147, 148, 149, 321, 322, 323, 324, 325, and
327. Table 2-11 (page 2-32) presents the potential HAP's for these
industries, and Table 2-12 (page 2-34) presents the industry-specific
emission sources.
2.1.7 Wood Products Industry
The wood products industry involves industrial processes that
convert logs to pulp, pulpboard, hardboard, plywood, particleboard,
or related wood products and wood preserving. This industry falls
under SIC Codes 242, 243, 249, and 261. Chemical wood pulping involves
the extraction of cellulose from wood by dissolving the lignin that
binds the cellulose fibers together." The principal processes used in
chemical pulping are the kraft, sulfite, and neutral sulfite. Plywood
production involves the manufacturing of wood panels composed of several
thin wood veneers bonded together with an adhesive. The wood preserving
process is one in which sawn wood products are treated by injection of
chemicals that have fungi static and insecticidial properties or impart
fire resistance. Table 2-13 (page 2-35) presents the potential HAP's
for these industries and Table 2-14 (page 2-36) presents the industry-
specific emission sources.
2.1.8 Petroleum Related Industries
In this manual, the petroleum related industries source category
includes the oil and gas production industry, the petroleum refining
industry, and the basic petrochemicals industry; these industries fall
under SIC Codes 13 and 29.
2-12
-------
The oil and gas production industry includes the following
processes: exploration and site preparation, drilling, crude processing,
natural gas processing, and.secondary or tertiary recovery. The prin-
cipal' products of this industry are natural gas and crude oil.
The petroleum refining industry involves various processes that
convert crude oil into more than 2,500 products, including liquefied
petroleum gas, gasoline, kerosene, aviation fuel, diesel fuel, a variety
of fuel oils, lubricating oils, and feedstocks for the petrochemicals
industry. The different processes involved in the petroleum refining
industry are: crude separation, light hydrocarbon processing, middle
and heavy distillate processing, and residual hydrocarbon processing.
In the basic petrochemicals industry, hydrocarbon streams from the
oil and gas production and petroleum refining industries are converted
into feedstocks for the organic chemical industry. These feedstocks
include benzene, butylenes, cresol and cresylic acids, ethylene,
naphthalene, paraffins, propylene, toluene, and xylene. The main pro-
cesses used by this industry are separation, purification, and chemical
conversion processes. (Refer to Appendix A.6 for a breakdown of typical
processes involved in each of the three industries.)
Table 2-15 (page 2-37) presents the potential HAP's that may be
emitted from these industries. Table 2-16 (page 2-38) provides more
specific information on potential emissions from the petroleum refining
industry segment of this generic category. A large proportion of the
emissions occur as organic vapors; for example, benzene, toluene, and
xylenes are the principal organic vapor emissions. This is due to the
chemical composition of the two starting materials used in these indus-
tries: crude oil and natural gas. Crude oil is composed chiefly of
hydrocarbons (paraffins, naphthalenes, and aromatics) with small amounts
of trace elements and organic compounds containing sulfur, nitrogen,
and oxygen. Natural gas is largely saturated hydrocarbons (mainly
methane). The remainder may include nitrogen, carbon dioxide, hydrogen
sulfide, and helium. Organic and inorganic particulate emissions, such
as coke fines or catalyst fines, may be generated in some processes.
The emission sources within each of the petroleum related indus-
tries are given in Table 2-17 (see page 2-40)-. Sources of potential
HAP emissions from the oil and gas production industry include: blowouts
2-13
-------
during drilling operations; storage tank breathing and filling losses;
wastewater treatment processes; and fugitive leaks in valves, pumps,
pipes, and vessels. In the-petroleum refining industry, potential HAP
emission sources include: distillation/fractionating columns, catalytic
cracking units, sulfur recovery processes, storage tanks, fugitives,
and combustion units (e.g., process heaters). Fugitive emissions are a
major source of emissions in this industry. Emission sources in the
basic petrochemicals industry are similar to those from the petroleum
refining industry and the SOCMI (see Section 2.1.3).
2.1.9 Combustion Sources
The fuel combustion industry encompasses a large number of combustion
units generally used to produce electricity, hot water, and process
steam for industrial plants; or to provide space heating for industrial,
commercial, or residential buildings. The combustion units may differ
in size, configuration, and type of fuel burned. Coal, fuel oil, and
natural gas are the major fossil fuels burned, although other fuels
such as wood and various waste (e.g., waste oil) or byproduct fuels are
burned in relatively small quantities. Industrial applications of both
gasoline- and diesel-powered stationary internal combustion units such
as generators, pumps, and well-drilling equipment are also included in
this category.
The waste incineration category includes combustion processes
whereby municipal solid wastes or sewage treatment sludges are disposed.
Table 2-18 (page 2-41) presents the potential HAP's for the above
combustion sources and Table 2-19 (page 2-42) presents the facility-
specific emission sources.
2-14
-------
TABLE 2-2. POTENTIAL HAP'S FOR SOLVENT USAGE OPERATIONS
Source Category
Solvent Degreasing
Dry Cleaning
Graphic Arts3
Waste Solvent Reclaiming
SCb-Flatwood Paneling0
SC-Machineryd
SC-Appliancese
SC-Metal Furniture
SC-Auto/Truckf
SC-Fabrics
SC-Cans9
SC-Paper, Tapes, Labels
Magnetic Tape Coating
SC-Electrical Insulation
SC-Marine Vessel sh
Vinyl & Acrylic Coatings1
SC-Wood Furniture
SC-Trans. Vehicles^
Machine Lubricants
Rubber Tire Manufacturing
Hazardous Air Pollutants
Organic Inorganic
Vapor Particulate Vapor Particulate
X
X
X
X
X
X X
X X
X X
X X
X
X
X
X
X
X X
X
X
X X
X
X
References
(Page 2-43)
2,3,4,15
3,5,9,15
3,5,6,15
5
1,5,6
1,7
1,5
1,12
1,5,12
5,15
1,5,6,8 '
1,5,13
10,11
1,15
1,3,20
3,16
1,17
1,20
33
34
aCategory incl
bSC: surface
GCategory incl
^Category incl
eCategory incl
^Category incl
SCategory incl
^Category incl
1 Category incl
^Category incl
udes flexography, lithography, offset printing, and textile printing,
coating.
udes coating of other flat stock.
udes coating of misc. metal parts, machinery, and equipment.
udes all categories of appliances: large and small.
udes coating of automobiles and light-duty trucks.
udes surface coating of coils, cans, containers, and closures.
udes coating and maintenance of marine vessels.
udes vinyl, acrylic, and nitrocellulose coatings.
udes coating of trucks, buses, railroad cars, airplanes, etc.
2-15
-------
TABLE 2-3. EMISSION SOURCES FOR SOLVENT USAGE OPERATIONS
Potential HAP Emission Sources
Source Category
Solvent Deyreasing
Dry Cleaning
Graphic Arts
Waste Solvent Reclaiming
SC-Flatwood Paneling
SC-Machinery
SC-Appliances
SC-Meta^ Furniture
SC- Auto/Truck
SC-Fabrics
SC-Cans
SC-Paper, Tapes, Labels
Magnetic Tape Coating
SC-Electrical Insulation
SC-Marine Vessels
Vinyl & Acrylic Coatings
SC-Wood Furniture
SC-Trans. Vehicles
Machine Lubricants
Rubber Tire Manufacturing
Process
Point
C
F
F
N
0
0
0
0
0
0
0
0
F
0
0
0
F
Process
Fugitive
A.B.D
E,G,H,I
L,M,N
I
L,P
Q,R
Q.R
Q,R
S,R
D,K,Q,R,T
Q,U
B,I,Q,T
I.Q.T
Q
L,P
S.R.T
S,R
I.V.W
Area
Fugitive
K
J,K
J,K
Q
Q
Source Key
A - bath evaporation
B - solvent transfer
C - ventilation
D - waste solvent disposal
E - washer
F - drying
G - still, filtration
H - cooker
I - solvent storage
J - pipes, flanges, pumps
K - transfer areas
L - rollers
M - ink fountains
N - condenser
0 - oven
P - coaters
Q - application area
R - flashoff area
S - spray booth
T - solvent/coating mixing
U - quench area
V - green tire spraying
W - sidewal1/tread end/undertread
cementing
2-16
-------
TABLE 2-4. POTENTIAL HAP'S FOR METALLURGICAL INDUSTRIES
Source Category
Hazardous Air Pollutants
Organic Inorganic References
Vapor Particulate Vapor Particulate (Page 2-43)
Primary Aluminum Production
Primary Cadmium Production
Metallurgical Coke
Primary Copper Smelting
Ferroalloy Production
Iron and Steel Production
Primary Lead Smelting
i
5,23,25
36
c,h,m,s,u, r
w,z,A,B,D
d,C a,e,f,g,i,n, 5,24,25,27,
o,p,q,t,v 28,31,35,36
a,e,i,k,
n,o,t,x
q.v.x
5,18,23,24,25
30,31,32,36
5,19,21,22,
24,25,27,30,
32,35
f.l.J.M. 5,23,24,25,'
p,q,v,x 26,30,32,35,36
a,l a.e.i.k,
n,o,t
5,18,23,24,
25,30,32,36
Primary Zinc Smelting
Manganese Production
Nickel Production
Secondary Aluminum Operations
Secondary Copper Operations
(Brass and Bronze Production)
Gray Iron Foundries b,c,m,
s,u,w
Secondary Lead Smelting
Steel Foundries
Secondary Zinc Processing
r a,l a,i ,k,n,
o,t,x
r p
r a,l a,i,n,q,t,x
1 l.q
r x i,k,n,
q.t.x
k,n,o,p,q,
v,x,y
r a,n,p,t
p»q,y
x i,o,q,t,x
5,18,23,24,
25,30,31,32,36
23,24,26,35
23,24,35
5,27
5,23,25,27,
30,32
5,21,23,24
25,26,27,30,
32,35
5,21,23,24,25
5,24,25,26,
27,35
5,24,25,27,37
(Continued)
2-17
-------
TABLE 2-4. POTENTIAL HAP'S FOR METALLURGICAL INDUSTRIES
(concluded)
Source Category
Hazardous Air Pollutants
Organic Inorganic References
Vapor Particulate Vapor Particulate (Page 2-43)
Lead Acid Battery Production
Cadmium-Nickel Battery Prod.
Dry Battery Production
Misc. Lead Products
n n 5,30
i,n 24,27,35
p 24,26,35
n e,n 5
Pollutant Key
a
b
c
d
e
f
9
h
i
j
- arsenic
- acrolein
- acetaldehyde
- ammonia
- antimony
- barium
- beryl lium
- benzene
- cadmium
- chromium
k
1
m
n
0
P
q
r
s
- copper
-fluoride
- formaldehyde
- lead
- mercury
- manganese
- nickel
- polycyclic organic
matter (POM)
- phenol
t
u
V
w
X
y
z
A
B
C
D
- selenium
- toluene
- vanadium
- xylene
- zinc
- iron
- cresols
- cyanides
- pyridine
- hydrogen sulfide
- methyl mercaptan
2-18
-------
TABLE 2-5. EMISSION SOURCES FOR METALLURGICAL INDUSTRIES
Source Category
Primary Aluminum Production
Primary Cadmium Production
Metallurgical Coke
Primary Copper Smelting
Ferroalloy Production
Iron and Steel Production
Primary Lead Smelting
Primary Zinc Smelting
Manganese Production
Nickel Production
Secondary Aluminum Operations
Secondary Copper Operations
Gray Iron Foundries
Secondary Lead Smelting
Steel Foundries
Secondary Zinc Processing
Lead Acid Battery Production
Cadmium-Nickel Battery Prod.
Dry Battery Production
Misc. Lead Products
Potential HAP Emission
Process Process
Point Fugitive
A,I,J,M,N,R H,K,D
J,E 0,P
B C,0,X
F,J,T G,H,K,0,P,X
J H,K,0,P
B.J.V C,H,K,0,X
J,V H,K,0,P
E.J.T.S 0
J H,K,M,P
A,I,J,M,T P
J H.K.P
J H,K,P
J,Y H,K,G,P
J H,K,P
J,Y G,H,K,P
J.E.S H,K,L,P
O.P
V N,0
M,N,0
G.O.P
Sources
Area
Fugitive
N.Q.U.Z
N,Z
N.D.Q.U
N,Q,U,W,Z
N,Q,W
D,N,Q,U,W,Z
N,Q,U,W,Z
N,Q,U,W,Z
N,Q,Z
N.Q.Z
U
U
U
U,Q
U
U
Source Key
A - calciner
B - coke oven
C - coke oven charging/pushing
D - coke quenching
E - condenser
F - converter
G - converter charging/ etc.
H - furnace tapping
J - furnace
K - furnace charging
L - galvanizing vessel
M - material crusher/mill
N - material storage and
0 - material preparation
P - metal casting
Q - outdoor storage pile
R - reduction eel 1
S - retort
T - roaster
U - service road
V - sintering machine
W - slag dumping
X - vessel leakage
Y - foundry mold & core
Z - mining operations
handling
decomposition
2-19
-------
TABLE 2-6. EMISSION SOURCES FOR THE SOCMI3
Source Category
(Generic Source)
Storage and Handling
Reactor Processes
Separation Processes
Fugitives
Potential HAP Emission Sources
Process Process
Point Fugitive
A
E,F G
F,L G,M,N
G,M,N
(Specific)
Area
Fugitive13
B,C,D
C,D,H
K
B,C,D,H,I
J,K,M,N,0
aSources: References 14, 38, 39, 40, 41, 42, and 43.
bGroups of small point sources (e.g., valves, compressors, pumps, etc.)
at a SOCMI plant are considered as area fugitive sources in this manual
Source Key
A - storage, transfer, and handling I
B - spills J
C - valves K
D - flanges I
E - reactors
F - product recovery devices
(absorber, adsorber,
condenser) M
G - process drains N
H - pumps 0
compressors
sampling lines
pressure relief devices
separation devices
(distillation column,
absorber, crystalizer,
dryer, etc.)
hotwel1
accumulator
cooling tower
2-20
-------
TABLE 2-7. POTENTIAL HAP'S FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
Hazardous Air Pollutants
Source Category Inorganic
Vapor Particulate
Aluminum chloride f,l
Aluminum fluoride s
Ammonia ' a
Ammonium acetate a
Ammonium-nitrate, sulfate a
thiocyanate, formate, tartrate
Ammonium phosphate a,s
Antimony oxide g
Arsenic-disulfide, iodide d d
pentafluoride, thioarsenate
tribromide, trichloride,
trifluoride, trioxide
orthoarsenic acid
Barium-carbonate, chloride h
hydroxide, sulfate, sulfide
Beryllium-oxide, hydroxide i
Boric acid and Borax k
Bromine j,l
Cadmium (pigment) -sulfide q
sulfoselenide, lithopone
Calcium-carbide, arsenate . e,s d
phosphate
Chlorine 1 8
Chlorosulfonic acid u,K
Chromic acid n m,n
References
(Page 2-43)
25
25
5,25,44
25 '
25,45,46
25
25
15,47
25,48
15,24,25
25
25
15,25,49
15,24,25,47
5,25,29,49
25
22,24,25
(continued)
2-21
-------
TABLE 2-7. POTENTIAL HAP'S FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(continued)
-Hazardous Air Pollutants
Source Category Inorganic
Vapor Particulate
Chromium-acetate, borides m
halides etc.
Chromium (pigment) -oxide m
Cobalt-acetate, carbonate o
halides, etc.
Copper sulfate p
Fluorine s
Hydrazine a,P
Hydrochloric acid u,v v
Hydrofluoric acid s
Iodine (crude) 1 0
Iron chloride 1 ,v v
Iron (pigment) -oxide Q
Lead-arsenate, halides e d,x
hydroxides, dioxide,
nitrate
Lead chromate y
Lead (pigments) -oxide x
carbonate, sulfate
Manganese dioxide A z
(Potassium permanganate)
Manganese sulfate z
Mercury-halides, nitrates, B
oxides
Nickel -halides, nitrates, C
oxides
References
(Page 2-43)
15,22,24,25
15,22,24,25
25
25,32
25
25
5,25
5,25
25
25
25
15,24,25,30,47
25
5,25,30
15,24,25,26,35
15,24,25,26,35
15,24,25,29,31
15,24,25,27
2-22
(continued)
-------
TABLE 2-7. POTENTIAL MAP'S FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(continued)
Source Category
Nickel sulfate
Nitric acid
Phosphoric acid
Wet Process -
Thermal Process -
Phosphorus
Phosphorus oxychloride
Phosphorus pentasulfide
Phosphorus trichloride
Pot ass i um-bi chromate ,
chromate
Potassium hydroxide
Sodium arsenate
Sodium carbonate
Sodium chlorate
Sodium chromate-
di chromate
Sodium hydrosulfide
Sodi urn-si licofluoride,
fluoride
Sulfuric acid
Sulfur monochloride-
dichloride
Zinc chloride
Hazardous Air Pollutants
Inorganic
Vapor Particulate
D
E
1,5
d,G,t
s
1
H,F
1,1, F
r
1
a
1
r
t
s
J,K
1
M,x
C
E
G
F
F
r
B
d
r
J
X
References
(Page 2-43)
15,24,25,27
5,25,50
5,15,23,25
25
25
25
25
22,24,25
15,24,25
15,25,47
5,25,51
25
22,2.4,25
25
25
5,25,52,53
25
25
(continued)
2-23
-------
TABLE 2-7. POTENTIAL HAP'S
FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(concluded)
Source Category
Hazardous Air Pollutants
Inorganic
Vapor Particulate
References
(Page 2-43)
Zinc chromate (pigment)
Zinc oxide (pigment)
L
N
25
25
Pollutant Key
a - ammonia
d - arsenic
e - arsenic trioxide
f - aluminum chloride
g - antimony trioxide
h - barium salts
i - beryllium
j - bromine
k - boron salts
1 - chlorine
m - chromium salts
n - chromic acid mist
o - cobalt metal fumes
p - copper sulfate
q - cadmium salts E
r - chromates (chromium) F
s - fluorine G
t - hydrogen sulfide H
u - hydrogen chloride I
v - hydrochloric acid J
x - lead K
y - lead chromate L
z - manganese salts M
A - manganese dioxide N
B - mercury 0
C - nickel P
D - nickel sulfate Q
nitric acid mist
phosphorus
phosphoric acid mist
phosphorus pentasulfide
phosphorus trichloride
sulfuric acid mist
sulfur trioxide
zinc chromate
zinc chloride fume
zinc oxide fume
iodine
hydrazine
iron oxide
2-24
-------
TABLE 2-8. EMISSION SOURCES FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
Source Category
Aluminum chloride
Aluminum fluoride
Ammonia
Ammonium acetate
Ammonium-nitrate, sulfate
thiocyanate, formate, tartrate
Ammonium phosphate
Antimony oxide
Arsenic-disulfide, iodide
pent afluo ride, thioarsenate
tribromide, trichloride,
trifluoride, trioxide,
orthoarsenic acid
Barium-carbonate, chloride
hydroxide, sulfate, sulfide
Beryllium-oxide, hydroxide
Boric acid and Borax
Bromine
Cadmium (pigment) -sulfide
sulfoselenide, lithopone
Calcium-carbide, arsenate
phosphate
Chlorine
Chlorosulfonic acid
Chromic acid
Potential
Process
. Point
X
X
B.O.E
X
C,F,I,L
X
X
H,U
C,E,G,I,L,U
X
X
X
X
H
H,C
X
H
HAP Emission
Process
Fugitive
X
X
K
X
Q
X
X
K,Q,T
N,P,Q,T
X
X
X
X
K,P
- K,R
X
K,N,0,Q
Sources
Area
Fugitive
J,S
J,S
J
J,S
(continued)
2-25
-------
TABLE 2-8. EMISSION SOURCES FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(continued)
Source Category
Chromium-acetate, borides
halides, etc.
Chromium (pigment) -oxide
Cobalt-acetate, carbonate
halides, etc.
Copper sulfate
Fluorine
Hydrazine
Hydrochloric acid
Hydrofluoric acid
Iodine (crude)
Iron chloride
Iron (pigment) - oxide
Lead-arsenate, halides
hydroxides, dioxide,
nitrate
Lead chromate
Lead (pigments) -oxide
carbonate, sulfate
Manganese dioxide
(Potassium permanganate)
Manganese sulfate
Mercury -halides, nitrates,
oxides
Nickel -halides, nitrates,
oxides
Potential
Process
Point
X
X
X
X
X
X
B
8,6
X
X
X
G,L
G,R
G,R
G,L
G,L
X
HAP Emission
Process
Fugitive
X
X
X
X
X
X
K,R
X
X
X
P,Q
P.Q
P,Q
Q.P.T
Q,P,T
X
- P,Q
Sources
Area
Fugitive
2-26
(continued)
-------
TABLE 2-8. EMISSION SOURCES FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(continued)
• Source Category
Nickel sulfate
Nitric acid
Phosphoric acid
Wet* Process -
Thermal Process -
Phosphorus
Phosphorus oxychloride
Phosphorus pentasulfide
Phosphorus trichloride
Potassium-bichromate,
chromate
Potassium hydroxide
Sodium arsenate
Sodium carbonate
Sodium chlorate
Sodium chromate-
di chromate
Sodium hydrosulfide
Sodlim-silicofluoride,
fluoride
Sulfuric acid
Sulfur monochloride-
dichloride
Zinc chloride
Potential
Process
Point
L
B,H
H,C,W
B,6
X
X
X
X
I
X
H
I.L.V
X
G.I.L.M
X
X
A,B,C,H
X
X
HAP Emission
Process
Fugitive
Q,T
K.N.R
K.N.P.T
K,N,R,T
X
X
X
X
X
K,P
P
X
P,Q
X
X
K,R
" X
X
Sources
Area
Fugitive
J,S
J,S
J,S
J,S
(continued)
2-27
-------
TABLE 2-8. EMISSION SOURCES FOR INORGANIC CHEMICAL MANUFACTURING INDUSTRY
(concluded)
Potential HAP Emission Sources
Source Category
Zinc chromate (pigment)
Zinc oxide (pigment)
Process
Point
X
X
Process
Fugitive
X
X
Area
Fugitive
Source Key
A - converter
B - absorption tower
C - concentrator
D - desulfurizer
E - reformer
F - neutralizer
G - kiln
H - reactor
I - crystal lizer
J - compressor and pump seals
K - storage tank vents
L - dryer
M - leaching tanks
N - filter
0 - flakers
P - milling/grinding/crushing
Q - product handling and
packaging
R - cooler (cooling tower,
condenser)
S - pressure relief valves
T - raw material unloading
U - purification
V - calciner
W - hot wel1
X - no information
2-28
-------
TABLE 2-9. POTENTIAL HAP'S FOR THE CHEMICAL PRODUCTS INDUSTRY
Source Category
Carbon Black
Charcoal
Explosives
Fertilizers
Paint & Varnish
Pharmaceutical
Plastics
Printing Ink
Pesticides
Soap and Detergents
Synthetic Fibers
Synthetic Rubber
Hazardous Air Pollutants
Organic Inorganic
Vapor Particulate Vapor Particulate
n,o,u,P,x 0 a a,g,j ,k,B,C,K
d,w,D 0
w
w,z,R W
P.v.E.T f,B,Q,V
q,r,E,H,T B
w,G,I,M,P,X,Y,Z
b,p,A,P,T,S
i,p,q,r,t,y, B
F,J,M,U,
h,H 0 a a,e
c,h,rn,n ,s,
w,F,L,N,P,
T,x
c,l,r,t,v, 0
G,H,I,J,T,W
References
(Page 2-43)
15,24,25
15,24,25,54
24,35
5,15,24,
35,55
15,23,24,35
15,24,35,
56,57,58
15,24,25,35,59
5,15,24,35
15,24,35,
57,58,60
23,24,35
15,24,25,
35,61
15,24,25,
35,61,62
Pollutant Key (see next page).
(continued)
2-29
-------
TABLE 2-9. POTENTIAL MAP'S FOR THE CHEMICAL PRODUCTS INDUSTRY
(concluded)
Pollutant Key
a - arsenic
b - acrolein
c - acrylonitrile
d - acetic acid
e - boron
f - barium
g - beryllium
h - benzene
i - cresols
j - cadmium
k - chromium
1 - chloroprene
m - caprolactum
n - carbon disulfide
o - carbonyl sulfide
p - carbon tetrachloride
q - chloroform
r - dichlorobenzene
s - dimethylformamide
t - dimethyl amine
u - ethylene
v - ethylene dichloride
w - formaldehyde
x - hydrogen sulfide
y - hexachlorocyclopentadiene
z - hydrogen fluoride
A - ketones
B - mercury
C - manganese
D - methanol
E - methyl chloroform
(1,1,1-trichloroethane)
F - maleic anhydride
G - butadiene, 1,3-
H - morpholine
I - methylene chloride
J - nitrosomines
K - nickel
L - perchloroethylene
M - phosgene
N - phthalic anhydride
0 - polycyclic organic matter
P - phenol
Q - selenium
R - silicontetrafluoride
S - terpenes
T - toluene
U - xylene
V - zinc
W - ammonia
X - vinyl chloride
Y - toluene diisocyanate
I - pyridine
2-30
-------
TABLE 2-10. EMISSION SOURCES FOR THE CHEMICAL PRODUCTS INDUSTRY
Potential HAP Emission Sources
Source Category
Carbon Black
Charcoal
Explosives
Fertilizers D
Paint and Varnish
Pharmaceutical
Plastics
Printing Ink
Pesticides
Soap and Detergents
Synthetic Fibers A
Synthetic Rubber A
Process
Point
B,H
E
A,C,H
,H,R,S,V
N,0
A,H,U,W
A.P.V
Q
A,H,0,X
M,N,0
,H,J,0,U,
v.x.z
,H,0,P,X,Z
Process Area
Fugitive Fugitive
G,K,L I
K
K.T
L
G,L F
K,L F,I
G F,I
K,L
G,K I
Y F
Source Key
A - reactor
B - furnace
C - concentrator
D - neutral izer
E - kiln
F - compressor and pump seals;
flanges, open ended lines,
lines
G - storage tank vents
H - dryer
I - spills
J - spin cell or bath
K - product handling, finishing
and packaging
L - raw material transport and
M - spray dryer
N - kettle
0 - mixing tank (blend tank)
P - polymerization vessel
Q - cooking vessel
R - prill tower
S
T
U
V
W
valves, X
sampling Y
Z
»
unloading
- granulator
- screen
- distillation
- cooler (condenser)
- crystal lizer
- filter
- mil ling/blending/
compounding
- flash tank
2-31
-------
TABLE 2-11. POTENTIAL HAP'S FOR THE MINERAL PRODUCTS INDUSTRY
Hazardous Air Pollutants
Source Category Organic Inorganic
Vapor Particulate Vapor Particulate
Asbestos Products c,j,q
Asphalt Batching Plants b,h,m r
Brick, Ceramic, and j,l
Related Clay Prod. u,z g,j,l
Refractories j,l j,l
Cement Manufacture r j,l, g.i.j.n,
q,w o.q.x.y
Coal Cleaning (Dry) v
Coal Cleaning (Wet) a,e,f,g,i,j,k,
n,p,q,t',u,x
Coal Conversion h,s, r d,w a,e,g,i,n,
A.B ' o,p,q,t,C
Glass Fiber Manuf. m,s s f.t.u
Frit Manufacturing 1 1
Glass Manufacturing a,d,l,n,z a,e,f,l,
n,t,u
Lime Manufacturing o o
Mercury Ore Processing o o
Mineral Wool Manuf. m,s 1 ,w
Perlite Manufacturing 1 1
Phosphate Rock Processing f»t,u f,t,u
Taconite Ore Processing c
References
(Page 2-43)
23,24,27
5,15,23,
24,25,63
5,15,22,23,
24,25,64
5,15,22,
24,65
5,15,22,24,
25,26,27,30,
31,49,64,66
25,28
28,67
5,27,28,67
5,23,68,69
5,23
5,15,23,24,
25,47,68
15,24,31,70
71
29,31
5,25,69,72
5,15,73
15,24,25,74
75
Pol 1utant Key (see next page).
(continued
2-32
-------
TABLE 2-11. POTENTIAL MAP'S FOR THE MINERAL PRODUCTS INDUSTRY
(concluded)
Pollutant Key
a
b
c
d
e
f
g
h
i
j
arsenic
aldehydes
asbestos
ammoni a
antimony
barium
be ry11i urn
benzene
cadmium
chromium
k - copper t
1 - fluoride u
m - formaldehyde v
n - lead w
o - mercury x
p - manganese y
q - nickel z
r - polycyclic organic A
matter (POM) B
s - phenol C
selenium
boron
coal dust
hydrogen sulfide
zinc
iron
chlorine
cresols
toluene
phosphorus
2-33
-------
TABLE 2-12. EMISSION SOURCES FOR THE MINERAL PRODUCTS INDUSTRY
Source Category
Asbestos Products
Asphalt Batching Plants
Brick, Ceramic, and
Related Clay Products
Refractories
Cement Manufacture
Coal Cleaning (Dry Process)
Coal Cleaning (Wet Process)
Coal Conversion
Glass Fiber Manufacturing
Frit Manufacturing
Glass Manufacturing
Lime Manufacturing
Mercury Ore Processing
Mineral Wool Manufacturing
Perlite Manufacturing
Phosphate Rock Processing
Taconite Ore Processing
Potential
Process
Point
B
B.E.C
B,E
E
B,C
B,H
C.O
B,C
C
E,T
C
C,0
B,C
A.B.Q
C,Q
. HAP Emission Sources
Process Area
Fugitive Fugitive
D,N
F,J,M
D,F,N
D.F.N
F,G,N,S
M,N,R
M,N
F,G,M,N
D,F,G,N,P
S
0,F,M,N
G,R,S
G.N
D.G.P
G,M,N,S
F,M,N,R
F,M,N,R
I.L
I
I.L
I
I.L
I.L
I.L
I.L
I
I.L
I
I.L
I.L
I.L
I.L
I.L
I.L
Source Key
A - calciner
B - dryer
C - furnace
D - end-product forming and finishing
E - kiln
F - raw material preparation/mixing
G - cooling
H - reactor
I - storage pile
J - saturator
L - mining operations
M - raw material handling/
transport
N - raw material crusher/mill
0 - oven
P - resin application
Q - washers
R - screening
S - end-product handling/
grinding/bagging
T - hydrator
2-34
-------
TABLE 2-13. POTENTIAL HAP'S FOR THE WOOD PRODUCTS INDUSTRY
Source Category
• ' Hazardous Air Pollutants
Organic Inorganic References
Vapor Particulate Vapor Particulate (Page 2-43)
Chemical Wood Pulping
Kraft Pulp Mill
a,b,c,d 5,15,23,
24,25,76
Sulfite Pulp Mill
Neutral Sulfite Pulp Mill
Plywood, Particleboard, h,l,
Hardboard o,p
Wood Preservative j,g,m,n
f,k a,b,c,d 5,15,23,
24,25
a,c,d
5,23,25.
5,24,35
75
24,35,58
Pollutant Key
a - arsenic
b - asbestos
c - chromium
d - mercury
e - polycyclic organic matter (POM)
f - chlorine
g - chlorobenzene
h - formaldehyde
i - methyl mereaptan
j - dioxin
k - hydrogen sulfide
1 - phenol
m - pentachlorophenol
n - cresols
o - abietic acid
p - pinene
2-35
-------
TABLE 2-14. EMISSION SOURCES FOR THE WOOD PRODUCTS INDUSTRY
Source Category
- ' Potential
Process
Point
HAP Emission Sources
Process Area
Fugitive Fugitive
Chemical Wood Pulping
Kraft Pulp Mill A,B,C,D
Sulfite Pulp Mill A.B.C
Neutral Sulfite Pulp Mill A.C.E
Plywood, Particleboard, G
Hardboard
Wood Preservative
F
F
Source Key
A - recovery furnace
B - digester
C - blow tank
D - lime kiln
E - fluidized bed reactor
F - resin and/or adhesive application
G - dryer
2-36
-------
TABLE 2-15. POTENTIAL HAP'S FOR PETROLEUM RELATED INDUSTRIES*
(General Listing for Entire Source Category)
Organic
Vapor Participate
Hazardous Air Pollutants
Inorganic
Vapor
Particulate
Parafins
Cycloparafins
Aromatics (e.g.,
benzene, toluene
xylene)
Phenols
Sulfur containing
compounds (e.g.,
mercaptans,
thiophenes)
Coke
Sulfides Catalyst fines
(e.g., hydrogen
sulfide, carbon
disulfide,
carbonyl sulfide)
Ammoni a
aSource: References 28, 77, 78,79, 80, 81, and 82.
2-37
-------
TABLE 2-16. POTENTIAL MAP'S FOR PETROLEUM REFINING INDUSTRIES3
(Specific Listing for Petroleum Refining Segment)
Process
Hazardous Air Pollutants
Organic
Vapor Particulate Vapor
Inorganic
Particulate
Crude Separation
Light Hydrocarbon
Processing
Middle and Heavy
Distillate
Processing
Residual
Hydrocarbon
processing
Auxiliary
Processes
*
a,b,d,e, o c,m,t,u, P,I,Q,R
f,g,h,i, v,x,y,L
o.A.B.C,
D.E.F.J
g,h,i,n, R t,v G.H.Q
N.O.P
a.d.e.f, o,R m,t,u,v, p,q,G,H,
9,h,i,j, x.y.L I.Q.U
k.l.F.J,
K,0,P,S,
T
a,d,e,f, o,R m,s,t,u, p,q,G,H,
g.h.i.j, v,x,y,L I,Q,U
k,l,n,F,
J.M.N.P,
S.T
a,b,d,e, o,R c,m,s,u, p,q,r,z,
f.g.n.i, y,L I
A,B,C,D,
J,K,M,T
aSource: Reference 28.
Pollutant Key (see next page).
(continued;
2-38
-------
TABLE 2-16. POTENTIAL HAP'S FOR PETROLEUM REFINING INDUSTRIES3
(Specific Listing for Petroleum Refining Segment)
(concluded)
Pollutant Key
a - malelc anhydride A
b - benzole acid B
c - chlorides C
d - ketones D
e - aldehydes E
f - heterocyclic compounds F
(e.g., pyridlnes) G
g - benzene H
h - toluene I
1 - xylene J
j - phenols K
k - organic compounds containing L
sulfur (sulfonates, sulfones) M
1 - cresol N
m - inorganic sulfides 0
n - mercaptans P
o - polynuclear compounds (benzo Q
pyrene, anthracene, etc.) R
p - vanadium S
q - nickel T
r - lead U
s - sulfuric add
t - hydrogen sulfide
u - ammonia
v - carbon disulfide
x - carbonyl sulfide
y - cyanides
z - chromates
acetic acid
formic acid
methy1ethyl amine
diethyl amine
thlosulfide
methyl mercaptan
cobalt
molybdenum
zinc
cresylic acid
xylenols
thiophenes
thiophenol
nickel carbonyl
tetraethyl lead
cobalt carbonyl
catalyst fines
coke fines
formaldehyde
aromatic amines
copper
2-39
-------
TABLE 2-17. EMISSION SOURCES FOR THE PETROLEUM RELATED INDUSTRIES
Source Category
OIL AND GAS PRODUCTION
Exploration, Site Preparation
and Drilling
Crude Processing
Natural Gas Processing
Secondary and Tertiary
Recovery Techniques
PETROLEUM REFINING INDUSTRY
Crude Separation
Light Hydrocarbon Processing
Middle and Heavy Distillate
Processing
Residual Hydrocarbon Processing
Auxiliary Processes
BASIC PETROCHEMICALS INDUSTRY
Olefins Production
Butadiene Production
Benzene/Tol uene/Xylene
(BTX) Production
Naphthalene Production
Cresol/Cresylic Acids Production
Normal Paraffin Production
Potential HAP Emission Sources
Process Process Area
Point Fugitive Fugitive
A C D,E
G F,H
G,J,K H I
G I
G,J,L F,H,M,N I
0,G F,H Q
G,0,P,R F,H I
B,G,K,0,R H I
G F,H I
G,K,0 F,H I
G,J,L,0,R F,H,N I
G,K,0,R F,Q I
G.L.O F,H I
G,L F,H
G,0 F,H I
Source Key
A - blowout during drilling
B - visbreaker furnace
C - cuttings
D - drilling fluid
E - pipe leaks (due to corrosion)
F - wastewater disposal (process
drain, blowdown, cooling water)
G - flare, incinerator, process
heater, boiler
H - storage, transfer, and handling
I - pumps, valves, compressors,
fittings, etc.
J - absorber
K - process vent
L - distil lation/fractionation
M - hotwells
N - steam ejectors
0 - catalyst regeneration
P - evaporation
Q - catalytic cracker
R - stripper
2-40
-------
TABLE 2-18. POTENTIAL HAP'S FOR COMBUSTION SOURCES
Source Category
Coal Combustion
Oil Combustion
Natural Gas Combustion
Gasoline Combustion
Diesel Combustion
Wood Combustion
Waste Oil Combustion
Municipal Refuse
Incineration
Sewage SI udge
Incineration
PCB Incineration
Hazardous Air Pollutants
Organic Inorganic
Vapor Particulate Vapor Particulate
n,y,B s a,b,h, a,b,e,f,
i.m.q h,i,j,k,o,
A p,r,t,v,x
n s m,q,A a,b,e,f,h,
r,v,x,C
n s
l,n l,s q o
1 l,s f,r
c,d,l,n,y l,s A p,t
g,l,u,w,z l,s f,h,i,o,r
1 Us q,A f,h,i,k,o,
1 l,s q a,f,h,i,o,
P.r
l,u l,s D
References
(Page 2-43)
5,22,23,25,
26,27,28,30,
31,32,35,36,
55,74,83,84
5,22,23,25
26,27,32,35,
36,55,74,83,84
5,23,37,84
5,15,25,35
5,15,35 '
5,15,25,35
5,30,83
5,15,22,25,26,
27,29,30,32,36
5,15,22,25,26
27,30,36,85
35,86
Pollutant Key
a - arsenic
b - antimony
c - acetaldehyde
d - acetic acid
e - barium
f - beryl! ium
g - benzene
h - cadmium
i - chromium
j - cobalt
k - copper
1 - dioxin
m - fl uoride
n - formaldehyde
o - lead
p - manganese
q - mercury
r - nickel
s - polycyclic organic matter (POM)
t - phosphorus
u - polychlorinated biphenyls (PCB)
v - radionuclides
w - trichloroethylene
x - zinc
y - phenol
z - ethyl benzene
A - chlorine
B - pyridine
C - vanadium
D - dibenzofuran
2-41
-------
TABLE 2-19. EMISSION SOURCES FOR COMBUSTION SOURCES
Potential HAP Emission Sources
Source Category
Process
Point
Process
Fugitive
Area
Fugitive
Coal Combustion A,B
Oil Combustion A,B,E
Natural Gas Combustion A,B,E,F
Gasoline Combustion G
Diesel Combustion G
Wood Combustion A,B,C
Waste Oil Combustion A,B,D
Municipal Refuse Incineration D
Sewage Sludge Incineration D
PCB Incineration D,B
Source Key
A - furnace
B - boiler
C - woodstove/fireplace
D - incinerator
E - gas turbine
F - reciprocating engine
G - industrial engine and/or equipment
H - coal storage pile
I - ash handling system
2-42
-------
2.1.10 References for Section 2.1
1. National Paint and Coatings Association. Section III'. Pfl^nt and
Coatings Markets. Table A-6. Estimated Consumption of Solvents in
Paints and Coatings, by Market - 1981. pp. 208-209. (no date).
2. U.S. EPA. Organic Solvent Cleaners - Background Information for
Proposed Standard (Draft).EPA-450/2-78-045a.October 1979.
3. U.S. EPA. End Use of Solvents Containing Volatile Organic Compounds.
EPA-450/3-79-032. May 1979.
4. U.S. EPA. Source Assessment: Solvent Evaporation - Degreasing
Operations. EPA-600/2-79-019f. August 1979.
5. U.S. EPA. Compilation of Air Pollutant Emission Sources. Third
Edition: Supplements 1-15. AP-42. January 1984.
6. U.S. EPA. Guidance for Lowest Achievable Emission Rates for 18
Major Stationary Sources of Particulates, Nitrogen Oxides,
Sulfur Dioxide, or Volatile Organic Compounds. EPA-450/3->9-024.
April 1979.
7. U.S. EPA. Control of Volatile Organic Emissions from Existing
Stationary Sources - Vol. VI: Surface Coating of Miscellaneous
Metal Parts and Products. EPA-450/2-78-015. June 1978.
8. U.S. EPA. Control of Volatile Organic Emissions from Existing
Stationary Sources - Volume II: Surface Coating of Cans, Coils,
Paper, Fabrics, Automobiles, and Light Duty Trucks.
EPA-450/2-77-008.May 1977.
9. U.S. EPA. Control of Volatile Organic Compound Emissions from
Large Petroleum Dry Cleaners.EPA-450/3-82-009.September 1982.
10. Bob Buzenburg, Development Planning and Research Associates, Inc.,
to Sill Johnson and Bob Short, EPA. Trip Report - Plant Visit of
September 24, 1981, to IBM, Boulder, Colorado.
11. Bob Buzenburg, Development Planning and Research Associates, Inc.,
to Bill Johnson and Bob Short, EPA. Trip Report - Plant Visit of
November 9, 1981, to 3-M Company, St. Paul, Minnesota.
12. D. Salman, EPA/CPB, to L. Zaragoza, EPA/PAB. Memo - Volatile Organic
Compound Potentially Emitted from Topcoats at Ford's Twin Cities
Plant. February 7, 1985.
13. U.S. EPA. Pressure Sensitive Tape and Label Surface Coating Industry -
Background Information for Proposed Standards"! EPA-450/3-80-003a.
September 1980.
14. U.S. EPA. Hazardous/Toxic Air Pollutant Control Technology: A
Literature Review. EPA-600/2-84-194. December 1984.
2-43
-------
15. U.S. EPA. Nonindustrial Sources of Toxic Substance Emissions and Their
Applicability to Source Receptor Modeling. Draft Report, EPA Contract
No. 68-02-3509, Task No. 42. July 27, 1983.
16. U.S. EPA. Control Technique Guidelines for the Control of Volatile
Organic Emissions from Wood Furniture Coating (Draft).April 1979.
17. U.S. EPA. Flexible Vinyl Coating and Printing Operations - Background
Information for Proposed Standards^EPA-450/3-81-016a.January 1983.
18. U.S. EPA. Background Information for New Source Performance Standards:
Primary Copper, Zinc, and Lead Smelters - Volume 1; Proposed Standards.
EPA-450/2-74-002a.October 1974.
19. U.S. EPA. Background Information for Standards of Performance:
Electric Submerged Arc Furnaces for Production of Ferroalloys -
Volume 1: Proposed Standards. EPA-450/2-74-018a. October 1974.
20. U.S. EPA. Control Techniques for Volatile Organic Compound Emissions
from Stationary Sources - Third Edition (Draft).April 1985.
21. U.S. EPA. A Method for Characterization and Quantification of Fugitive
Lead Emissions from Secondary Lead Smelters, Ferroalloy Plants, and
Gray Iron Foundries. EPA-450/3-78-003. January 1978.
22. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Chromium. EPA-450/4-84-007g. July 1984.
23. U.S. EPA. A Survey of Emissions and Controls for Hazardous and Other
Pollutants. EPA-R4-73-021. February 1973.
24. U.S. EPA. Industrial Sources of Hazardous Air Pollutants - Draft.
September 1983.
25. U.S. EPA. Source Assessment: Noncriteria Pollutant Emissions (1978
Update). EPA-600/2-78-OU4T.July 1978.
26. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Manganese (Draft). September 1984.
27. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Nickel. EPA-450/4-84-007F. March 1984.
28. U.S. EPA. Potentially Hazardous Emissions from the Extraction and
Processing of Coal and Oil.EPA-650/2-75-038.April 1975.
29. U.S. EPA. Review of National Emission Standards for Mercury.
EPA-450/3-84-01. December 1984.
30. U.S. EPA. Status Assessment of Toxic Chemicals: Lead.
EPA-600/2-79-210h. December 1979.
2-44
-------
31. U.S. EPA. Status Assessment of Toxic Chemicals; Mercury.
EPA-600/2-79-2101. December 1979.
32. U.S. EPA. Sources of Copper Air Emissions. EPA-600/2-85-046.
April 1985. .
33. U.S. EPA. The Use and Fate of Lubricants, Oils, Greases and Hydraulic
Fluids in The Iron and Steel Industry.EPA-600/2-78-101.May 1978.
34. U.S. EPA. Rubber Tire Manufacturing Industry - Background Information
for Proposed Standards.EPA-450/3-81-008a. July 1981.
35. U.S. EPA. Human Exposure to Atmospheric Concentrations of Selected
Chemicals. EPA Contract No. 68-02-3066.February 1982.
36. U.S. EPA. Survey of Cadmium Emission Sources. EPA-450/3-81-013.
September 1981.
37. U.S. EPA. Source Category Survey: Secondary Zinc Smelting and
U.S. tHA. source Category Survey: Secondary ZT
Refinery Industry.EPA-450/3-80-012.May 1980.
38. U.S. EPA. Air Oxidation Processes in Synthetic Organic Chemical
Manufacturing Industry - Background Information for Proposed Standards.
EPA-450/3-82-OUU.October 1983.
39. U.S. EPA. Reactor Processes in Synthetic Organic Chemical Manufac-
turing - Background Information for Proposed Standards (Draft).
October 1984.
40. U.S. EPA. VOO Emissions from Volatile Organic Liquid Storage Tanks -
Background Information for Proposed Standards (Draft).
EPA-450/3-81-003.July 1984.
41. U.S. EPA. VOC Fugitive Emissions in Synthetic Organic Chemicals
Manufacturing Industry - Background Information for Promulgated
Standards. EPA-450/3-80-033b. June 1982.
42. U.S. EPA. Distillation Operations in Synthetic Organic Chemical
Manufacturing - Background Information for Proposed Standards.
EPA-450/3-83-005a.December 1983.
43. U.S. EPA. Organic Chemical Manufacturing Volume 6: Selected
Processes. EPA-450/3-80-028a. December 1980.
44. U.S. EPA. Source Category: Ammonia Manufacturing Industry.
EPA-450/3-80-014. August 1980.
45. U.S. EPA. Source Assessment: Ammonium Nitrate Production.
EPA-600/2-77-107i.September 1977.
46. U.S. EPA. Ammonium Sulfate Manufacture - Background Information
for Proposed Standards.EPA-450/3-79-034a.December 1979.
47. U.S. EPA. Preliminary Study of Sources of Inorganic Arsenic.
EPA-450/5-82-005. August 1982.
2-45
-------
48. U.S. EPA. Source Assessment: Major Barium Chemicals.
EPA-600/2-78-004b. March 1978.
49. U.S. EPA. Emission Factors for Trace Substances. EPA-450/2-73-001.
December
50. U.S. EPA. Review of New Source Performance Standards for Nitric Acid
Plants. EPA-450/3-84-011. April 1984.
51. U.S. EPA. Sodium Carbonate Industry - Background Information for
Proposed Standards. EPA-450/3-80-029a. August 1980.
52. U.S. EPA. Industrial Process Profiles for Environmental Use: Sulfur,
Sulfur Oxides and Sulfuric ACKL EPA-600/2-77-023w. February 1977.
53. U.S. EPA. Final Guideline Document: Control of Sulfuric Acid Mist
Emissions from Sulfuric Acid Production Plants. EPA-450/2-77-019.
September 1977.
54. U.S. EPA. Source Assessment: Charcoal Manufacturing.
EPA-600/2-78-004Z. December 1978.
55. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Formaldehyde. EPA-450/4-84-Q07e. March 1984.
56. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Chloroform. EPA-450/4-84-007c. March 1984.
57. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Carbon Tetrachloride. EPA-450/4-84-007b. March .1984.
58. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Chlorobenzenes. (Draft). September 1984.
59. U.S. EPA. Plastics and Resins Industry - Industrial Process Profiles
for Environmental Use. EPA-600/2-77-023J. February 1977.
60. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Phosgene (Draft). September 1984.
61. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Acrylonitrile. EPA-450/4-84-007a. March 1984.
62. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Ethylene Dichloride. EPA-450/4-84-007d. March 1984.
63. U.S. EPA. Asphalt Roofing Manufacturing Industry - Background
Information for Proposed Standards (Draft). EP-A-450/3-80-021a.
June 1980.
64. U.S. EPA. Trace Pollutant Emissions from the Processing of
Nonmetallic Ores. EPA-650/2-74-122. November 1974.
65. U.S. EPA. Source Category Survey: Refractory Industry.
EPA-450/3-80-006. March 1980.
2-46
-------
66. U.S. EPA. A Review of Standards of Performance for New Stationary
Sources - Portland Cement Industry. EPA-450/3-79-012. March 1979.
67. U.S. EPA. Background Information, for Standards of Performance:
Coal Preparation Plants Volume I; Proposed Standards.
EPA-450/2-74-021a.October 1974.
68. U.S. EPA. Glass Manufacturing Plants, Background Information:
Proposed Standards of Performance (Draft). EPA-450/3-79-005a.
June 1979.
69. U.S. EPA. Wool Fiberglass Insulation Manufacturing Industry -
Background Information for Proposed Standards (Draft)T
EPA-450/3-83-002A. December 1983.
70. U.S. EPA. Standards Support and Environmental Impact Statement
Volume I: Proposed Standards of Performance for Lime Manufacturing
Plants.EPA-450/2-77-007a.April 1977.
71. U.S. EPA. Final Standards Support and Environmental Impact
Statement Volume II; Promulgated Standards of Performance for
Lime Manufacturing Plants.EPA-450/2-77-007b.October 1977.
72. U.S. EPA. Source Category Survey: Mineral Wool Manufacturing
Industry. EPA-450/3-80-016. March 1980.
73. U.S. EPA. Source Category Survey: Perlite Industry.
EPA-450/3-80-005. May 1980.
74. U.S. EPA. Radionuclides - Background Information Document for Final
Rules. Volume I.EPA-520/1-84-022-1.October 1984.
75. Standards Support and Environmental Impact Statement for the Iron
Ore Benefication Industry (Draft).Battelle Columbus Laboratories.
December 1976.
76. U.S. EPA. Kraft Pulping - Control of TRS Emissions from Existing
Mills. EPA-450/2-78-003b. March 1979.
77. U.S. EPA. Industrial Process Profiles for Environmental Use:
Chapter 2. Oil and Gas Production Industry. EPA-600/2-77-023b.
February 1977.
78. U.S. EPA. Industrial Process Profiles for Environmental Use:
Chapter 3. Petroleum Refining Industry"EPA-600/2-77-023C.
January 1977.
79. U.S. EPA. Industrial Process Profiles for Environmental Use:
Chapter 5. Basic Petrochemicals Industry.EPA-600/2-77-023e.
January 1977.
80. U.S. EPA. VOC Fugitive Emissions in Petroleum Refining Industry -
Background Information for Proposed Standards. EPA-4bO/3-81-015a.
November 1982.
2-47
-------
81. U.S. EPA. VQC Species Data Manual, Second Edition.
EPA-450/4-80-115. July 1980.
82. U.S. EPA. Bulk- Gasoline Terminals— Background Information for
Proposed Standards (Draft).EPA-450/3-80-038a.December 1980.
83. U.S. EPA. Air Toxics Emission Patterns and Trends - Final Report.
EPA Contract No. 68-02-3513, Task 46.July 1984.
84. U.S. EPA. Hazardous Emission Characterization of Utility Boilers.
EPA-650/2-75-066. July 1975.
85. U.S. EPA. Thermal Conversion of Municipal Wastewater Sludge
Phase II: Study of Heavy Metal Emissions. EPA-600/2-81-203.
September 1981.
86. U.S. EPA. Locating and Estimating Air Emissions from Sources of
Polychlorinated Blphenyls (Draft).November 1984.
2-48
-------
2.2 IDENTIFICATION OF KEY EMISSION STREAM PROPERTIES
This section identifies the emission stream physical properties
needed to select the appropriate control technique(s) and to size the
control device(s) for each identified HAP emission stream generated by
a process source, be it either a process point source or a process
fugitive source. Design and costing techniques for area fugitive
emission control methodologies are outside the scope of this manual;
however, control techniques for vapor emissions and particulate emissions
from area fugitive sources are discussed in Sections 3.1.4 and 3.2.2,
respectively.
The actual/estimated values for the process emission stream
properties should be obtained from the owner/operator or from available
literature if the owner/operator cannot supply the necessary data. The
values obtained are used in conjunction with the guidelines given in
Chapter 3 to perform the control technique selection process. Table 2-20
(page 2-44) lists the required information for organic vapor emissions,
Table 2-21 (page 2-45) for inorganic vapor emissions, and Table 2-22
(page 2-46) for particulate emissions. After obtaining the values for
the key physical properties for each HAP emission stream, record the data
on the "HAP Emission Stream Data Form" found in Appendix C.I (see
Figure 2-1, page 2-7). Appendix 8,1 provides unit conversion factors,
and Appendix B.2 presents calculation procedures for determining the
heat content of an emission stream.
There will be occasions when it would be prudent for the owner/
operator to combine similar emission streams. For example, if two or
more emission streams require the use of the same control technique,
it will normally be more cost effective to combine the streams and
use just one control device as opposed to using a control device for
each separate emission stream. If the owner/operator decides to
combine emission streams, Appendix 8.2 provides calculation procedures
to determine the key effluent properties of- combined emission streams.
2-49
-------
TABLE 2-20. KEY PROPERTIES FOR ORGANIC VAPOR EMISSIONS
Emission Stream Properties- ' HAP Properties3
(preferred units of measure)
HAP Content (ppm by volume) Molecular Weight
Organic Content'5 (ppm by volume) Vapor Pressure
Heat Content0 (Btu/scf) Solubility (graph)
Oxygen Content (% by volume) Adsorptive Properties
(isotherm plot)
Moisture Content (% by volume)
Halogen / Metal Content (yes or no)
Flow Rate (scfm)
Temperature (°F)
Pressure (mm Hg)
aThese properties pertain to the specific HAP or mixture of HAP's
in the emission stream.
''Primary properties that affects control technique selection.
Organic content is defined as organic emission stream combustibles
less HAP emission stream combustibles.
cHeat content is determined from HAP/Organic Content (see Appendix B.2
for calculation procedures).
2-50
-------
TABLE 2-21. KEY PROPERTIES FOR INORGANIC VAPOR EMISSIONS
Emission Stream Properties HAP Properties3
(preferred units of measure)
HAP Content*5 (ppm by volume) Molecular Weight
Moisture Content (% by volume) Vapor Pressure
Halogen / Metal Content (yes or no) Solubility (graph)
Flow Rate (scfm) Adsorptive Properties
(isotherm plot)
Temperature (°F)
Pressure (mm Hg)
aThese properties pertain to the specific HAP or mixture of HAP's
in the emission stream.
^Primary properties that affects control technique selection.
2-51
-------
TABLE 2-22. KEY PROPERTIES FOR PARTICULATE EMISSIONS
Emission Stream Properties - • HAP Properties3
(preferred units of measure)
HAP Content (% by mass) (None)
Particulate Content13 (Ib/acf)
Moisture Content (% by volume)
503 Content (ppm by volume)
Flow Rate (acfm)
Temperature (°F)
Particle Mean Diameter0 (^m)
Drift Velocity0 (ft/sec)
aThese properties pertain to the specific HAP or mixture of HAP's
in the emission stream.
include total particulate loading and principle particulate
constituent.
cThese properties are necessary only for specific control techniques.
2-52
-------
EXAMPLE CASE
The emission stream from the coating oven has been
identified as containing organic vapors. Therefore, the
emission properties required are listed in Table 2-20. The
necessary information was provided by the owner/operator;
a source test was performed for a very similar operation at
another plant owned by his company. The information is
recorded on the worksheet presented in Appendix C.I. The
source test provided the following data (see Figure 2-1,
page 2-7):
1. HAP content
2. Organic content
3. Moisture content
4. Halogen content
5. Metal content
6. Temperature
7. Pressure
8. Flow rate
9* Heat content
10. Oxygen content
11. HAP molecular weight
12. HAP vapor pressure
13. HAP solubility
73 ppm (vol.) toluene
44 ppm (vol.) methane
4 ppm (vol.) other
2% (vol.)
None
None
120°F
Atmospheric
15,000 scfm (max.)
0.4 Btu/scf
20.6% (vol.)
92 Ib/lb-mole
28.4 mm Hg @ 77°F
Insoluble in water
2-53
-------
CHAPTER 3
CONTROL DEVICE SELECTION
This chapter presents guidelines that will enable the user to select the
control technique(s) that can be used to control HAP's. The control tech-
niques that can be applied to control HAP emissions from a specific emission
source will depend on the emission source characteristics and HAP characteris-
tics. Therefore, the chapter is divided into two sections, each pertaining to
specific HAP groups: Section 3.1 - Vapor Emissions Control and Section 3.2 -
Particulate Emissions Control. The discussion of control technique selection
within each section is according to type of HAP (organic or inorganic) and
emission source (point, process fugitive, or area fugitive).
In the following sections, guidelines for selecting controls for point
sources are discussed in detail. Point sources are typically controlled by
add-on control devices. For each control technique, ranges of applicability
with respect to emission stream characteristics, HAP characteristics,
performance levels (e.g., removal efficiency), and other considerations that
are important in control device selection are described in detail.
Work practices, including equipment modifications, play a key role in
reducing emissions from process fugitive and area fugitive sources. These
sources can also be controlled by add-on control devices if the emissions can
be captured by hooding or enclosure or collected by closed vent systems and
then transferred to a control device. Note that the overall performance of
the control system will then be dependent on both the capture efficiency of
the fugitive emissions and the efficiency of the control device.
To illustrate the control device selection process, several emission
stream scenarios are presented throughout this chapter. The emission stream
from the paper coating drying oven introduced in Chapter 2 is one of the
scenarios presented. The data necessary for control device selection are
recorded on the "HAP Emission Stream Data Form" (see Figure 2-1).
3-1
-------
3.1 VAPOR EMISSIONS CONTROL
3.1.1 Control Techniques for Organic Vapor Emissions from Point Sources
The most frequent approach to point source control is the application of
add-on control devices. These devices can be of two types: combustion and
recovery. Applicable combustion devices are thermal incinerators, catalytic
incinerators, flares, and boilers/process heaters. Applicable recovery
devices include condensers, adsorbers, and absorbers. The combustion devices
are the more commonly applied control devices, since they are capable of high
removal (i.e., destruction) efficiencies for almost any type of HAP (organic
vapor). The removal efficiencies of the recovery techniques generally depend
on the physical and chemical characteristics of the HAP under consideration.
Applicability of the control techniques depends more on the individual
emission stream under consideration than on the particular source category
(e.g., degreasing vs. surface coating in solvent usage operations source
category). Thus, selection of applicable control techniques for point source
emissions is made on the basis of stream specific characteristics and desired
control efficiency. Table 3-1 identifies the key emission stream characteris-
tics and HAP characteristics that affect the applicability of each control
technique and presents limiting values for each of these characteristics.
Matching the specific characteristics of the stream under consideration with
the corresponding values in Table 3-1 will help the user to identify those
techniques that can potentially be used to control the emission stream. The
list of potentially applicable control techniques will then be narrowed
further depending on the capability of the applicable control devices to
achieve the required performance levels. Figure 3-1 identifies the expected
emission reduction from the application of each control technique on the basis
of the total VOC concentration in the emission stream. Very little data
regarding control device removal efficiency for specific HAP's are available.
Therefore, without actual source test data for a specific emission stream and
control system, HAP removal efficiency is assumed to equal total volatile
organic compound (VOC) removal efficiency. Table 3-2 identifies other con-
siderations that, while not necessarily affecting the technical applicability
3-2
-------
£
V)
U» O
V) 2
UJ O
«C 2J
§2
. S
X U.
2 K
u> <
S"
K S
is
51*"
y Uj
LU 3
fc|
§w
^
V) _J
fig
LU Z
-5
1 Ul
m —j
IU f>
10
in
u
^
4J
Ul
^
L.
4
4J
U
A
k
fl
£
CJ
%
z
5
u
L.
Q
g
«
£
O
i
4"
§
Ul
II
S
u
O in
> O
H
Ul 0
at
•.si
O in
a. in g
w e B
> k «•»
0.
4«*
3
•5
k
a i
r- 4* A ~-
3 •= — £
0 — J3 "o
i- IP f- I
•*•
O
k
3
g£
•2
o
o ^^
u.
|o +>
i. e
3 0 **
4* 4J M
U) C —
"rt r^
;
* *> ^
o e s
x o +•>
u CQ
in
||i
1
o
u
1
*o
!_
c
c3
B |
ol^ll
w3 in j -O '
S 4J -o a a
o
m
I
in
§
I
§g
.
O O O >. O
§o o 10 i
O » 9 O
-• -^ tS 4J O
v v v ui m
-
^j
m
a .c
§o
in
A A
_l — _l O _l
UJ O UJ - Ul
___! O _ J O — 1
a**- • -4
MM 0 M
m I tn- O in
1- 0
•
> k e
E r- O O O
k a k r- .0
£ IO *— O IQ
K U U. IB U
1
—
O
« *» k
*i^ 3
0§^
—•Ok
Ul
k *•
>, 0 C
•— 4J O
»- 10 >
« * "o
o c in
000
£if
in p-
3 O k
£ in o
o
o
o
o
o
s §
§ °.
* CM
»-* V
§
o
c
o
1 O
o °>
to \f\
u
a
k
a
f
u
o
+J
5
a>
S.
41
•
ontent
u
|
g
•o
>>
X
termined from H
2
^
2
§
u
•o
c
>
3 0 4->
<•*>••
,. g 5
- 0 E
«•!•>.=
p- "8
A e »
10 JS >
a „ s
I 1 •;
C3> £ —
3-3
-------
c
o
a^
*
rt3
OJ
C
•r-
O
C
h—
r—
4
r<3
^
t5
^*
T3
C_3
LO
CT»
i
0
f
•
v
CT
C
,!
i
s
^
LO
O
1
H
•»
O
LO
J
h
•••
c
o
-1
p
0.
k.
o
m
•p
<
a:
c
o
<
(0
3*
00
LO
CT>
i
i
a*
o
cn
1
C
0
•r~
-4-5
0.
i»
o
^
•*
LO
cr>
i ,
'""""
o
00
I
o
LO
J
c
0
•F-
•M
fO
t/J
C
O)
T3
O
O
O
O
CM
0
O
0
^^
O
o
o
LO
0
o
-o
CO
o
"o
evT ^
I
CL
Q.
O •— '
O
O Ol
« CL.
C
O
O -r-
LO ra
s_
•4-*
C
O)
O 0
-0 C
n o
0
o "S>
— O ,.;.
v>4 j—
0
LO
_o
.0
a>
o
o
o
c
o
aj
s_
o
a>
01
n)
s_
c:
o
•a
O)
-------
V)
Ui
1
t£
£
g
u.
v»
&£
U
1
g
£
Z
£
.ECTION
«*
UJ
u
U
»•*
s
N CONTROL
VI
S
8
1— 1
VI
§
ce
X
S
t
tM
1
pri
LL)
§
r-
in
VI
O 0)
o in
x u c
+J O. O
> C +J
+» -5
•t- Ul c
ui « o
C CIO
v>
•M
Iin
•*• +•>
O- g
uj £
^% ^
IO 3
— ce.
u
5
in
*§
§«s
u
22
— Ul
j<§
fety
orations
10 ~o
V> w
Ul
1
x
01 X
W'~
K J3
a in
r— Ul
U 0
xo.
o
s
ce
o
u
1
.-
I.
^1
o
§
c
u
o
o
+3
3
in
8 -M
ja e «
c +J c *u
<0 •»• "- 3
U X E <••
fl p-
C S +J 3
o S e in
••• P" O
Sis?
SioS.2
o> -S 3 ji
3 5) *• *» «
p- a> i— 8
t- c 3 e 10
in — o>
*- Q) v. L.
O J3 Ul JJ +J
CCUI
Ol X 9 9
C (O Ol Ul C
^™ £ O 9 O
J3 /— I- —
J3 Ul 18 O. Ul
3 O .C Ul
i- in o —
o 10 •*- u e
v> 01** a 9
1.
t_ tn in ,—
O Ul 3 0
>» •^ 10 X C
•o s o —
e S *-
10 x o ?+n .
c a >o
a t. E — u 01
01 3 10 Ul O
o •»• a> in «s 01
i^ p" ^ ^— ^ 3
S in in 9 10 C
I
Js^iu*!
o o i- "S a «••
r- O Ul +* Jt O
S-*§'"g>.«
a> « •»• c »- ^ u>
ui .M o C N
O -o 10 -^ •»• -o
*J t. U. Ul « p- M
« +J Ul +J 3 ~.
SlMC-^COr- _J
3
X
&»
o
2
-1
i ^
O M
j£
O
at ui x
> in a 10 n
" in ® 5 O 1 *>
*J
§&• **B o> in ui
a. +J x^ o
in .. c i- c —
c in o a ia
X" Co-*-1 X •
fin i- v o 10 +J **
o -M c •—
CD" O 0) £ 0) >
kC'OJSUI^S"
•P lO-C 3 O O--M
XJ=oeou
UJOO'-UOUlfl
Ol
o S i*- i S
+J Ul O 'O U>
" ui -o e »•
— •— a. e +< f a.
^ " «F- O U>
in ? e in -v ^- ui
10 u Ul — 3 -H IO
OUIOiQiOr— CLU
•a
0)
ID -0
3 O •
•—OH E
•to c I- O 10
iO .C O
t- O -P U
c c ui
Ol X 0> •*•
C 10 0) C
— s o *> 3
J3 P- C "*•
^ ui 10 Q> in
3 a> £ ui in
L. ui at f
Si * fc u S
•o
o
IO
1- •
•(-» <3
»• > C ffl
r- 9 it- I.
•0 C -P O 4->
c — c ui
JZ Ul O 4J C
§ U)
** » S -2
• •^ * c — 1
.a -M •*• »*- o
iO U
a. 3 •o •o o
10 <— c e f
ui x
to E C
(•• ^"
& Ul
O 9 Ul •
u in E o
•o « a> e
o e f 3 N
•«v 5 <*- o
c u o a. I
U 9 9 P™
01 U V iO
10 x c t. j;
s JT •- u +J
ffl
C •— CO)
SO nj n- O fl C
3 +> O *• O "
L. (D C C
•o -a +J £ .c i- j= ja
3 a) *j v -c o -t-1 •
O Q) iO Ul CM
1 CBUI—t-UI^-ffl
j: +J 3 o +J •—
0 « ffl ffl O) ^O
c g "a.-— e a. 3 -H
j: XE o +J 5 o ui t.
IOOI.XP-UOIO
P- J= -^» J3
ai j=
iO Q) O CO X X
ui 4^ u e u o • •
§10 9 3 3 O 01 ^
t- Ul T3 CT U> •* °
•M O)"o L. U "3 E 3
V> iO C O.H- C tO -O
x|
E 1i • *> -0
in in c u
ui — ID E 01 10
.* 5 +J 5 o> » Q. •— C
fOS Ul C 13 IO •
in r~ 5 +J o c in
r?4J- i- u
o
c
z
o
10
IZ
. 01
* 1 1-
ui o JMI •£ 10
lOp-uuioui uii»*
O) iQio-MiO —
•O O U) T3 t- 1- O +J
ot - •*• 10 3 3 10 I-vt-UI
<*. 4) a. o~f— o " "O S
j: & i. O +J O>
+J • .0 J£ IO3UI
OXiOXOU 9 Q"
£t- OllOklO +J O Ul
X "^ iO Ul 6 Q.J3 V> U "
I.
u in ui
O Ul 3 CX
>K— * O
•§ § u<0
10 X t •
c 10 O in
in " s -^ c
e -o O
9 u e e •—
01 3 10 « tn
a a ** w §
^^
•o
(D
C
§
S
-------
.
i
3
£
|
W*
£
5
o
i
a.
K
£
o
r™ *•»
O -0
Ul 0
V) —
i-i o
i"
|
8
z
RATIONS
LU
O
V)
8
K
0
•N
1
f>
Ul
^
^»
in
in
O 0
44 y
6 in
X 1- C
44 Qu O
> c 44
^™ (A C
in 0 o
v> «J
S
0
^ C
i— C
il
££
0 O
s°
5
e
Ancillary Equlpme
Requi rements
m
"5 o
44 ^
c 44
ii
22
— in
eg
uj u
£
£
X44
-w a
o u
— 9
f
in
|
x
^
§=?
+•*
«•£
o: A
J2 S
u o
xo.
0
o
S
e
o
0
008 8 i
^ O *3 E ® E fe ^
44UI.OOOCO .
— CLOfVOUiO'^Vlin
Ul C44^44Ol»^t»
§c — in in 000
*~~eoee — a
ui in o o A 8 O o
xo c — 0 — 44 jc .0 .c
p- O) O in 44 in c 9
o g — in 10 in o »^ in
44 C O * *• O O
ui 2 8 — C5 J "S S a.
1
.0 "»• ui c
f— -O 0 O
S3 0 U —
in 44 a. in
u a in
«:M«.
in $ — ui 0 E
o i- u 3 e -w
O & o Q.— in
o u o
0 *• O "x. '-
O J3 Ul 44 4->
ecu)
Ol X 0 0
c g 5) ui c
— i S o o
ja in 13 Q.'in
3 o .c in
L. Ul 0 —
U <0 w- I- 6
to O)— a O
c
•2 0
u "ui Ul i—
O in 3 o
-.. — 10 x c
•o 5 o —
c 8 <*- .
fl X O (MUl
c -^ 9
>— >— U — -O 3
* 3 V E C •—
8 in
•a in
E ^ « OO
v 0 ^ y
e 0 -o r= +J 5
— £ C 0 £
in in a — —
§0 fi UJ |^
c 44 o m o
— 0 * *>Z 9
4J— UCO.I.UI.
(O Ul 0 X 3 O O
— UI**J0*J— 4->
WB-So 0 — "o *
~
c
5
o
u
o
t.
44
1
Ul
S
u
2
a. u
u 44
o a
^ X
o
m
g
C — . 0
— «*- 44 4=
C -D 44 9 .
UJgOCgO^g
— Im O " C Ul
•o ui u u u u in
— O 3 O C
— I- 3 -0 u JO
" * * tn r- •£ *"
^•^44 £'*"'••• ^ in
x S 2 *» 1 U 8. *
in
ox o
59 44 U O
S — 0 c
* « 44
>*• E *• ^ m 144 c
O « — — 0 x in j: 0
44 t x o-'o at'o "S x- c
c *> t. — — e a ui jr o
0 Ul a •— 44 — O
6 in u •— • 0 10
« o 8 c a.^ o 2 ui t.
«. "ui $ 15 ts i 'c S j: -3
4JUIC4J— en— in
0— c f— ^) 01 u. -o —
.
"0 x "c o
S'^ffOO O0in44 !S_S fe—ttc
S !_ r- Ul O003 Or— O
U30 J3 *> e 0 44 O —
— 44 > e I044L. lov-uin
•D a a 9 O x-— 44 c — , ui e -- e v-000
i- o S — ^ — ' u u c — 'oo'uo
0 J3 44 -O i e 0M0UI 0 — f
F- — 44 O, Ul Ul 4J £ 44 4J
o x 9 E o o f— 0— in u
UB44.C44IO O. -- O. 8 X -O Ok —
i- S
o in •n
44 0 44 0
8* * Q a 3 a
co 6 <-
=» J= 0 . t. 0
^_ 44 — C 0 ^ S=^
Ova— co JdoiO
44 44 O 44 . U 0
x o c i. -a
U ,— 10 .— 44 X g U 0
§— 5 — a o •— 4-> 0 o
0-044^444 e -c a.
U Ul Ul 44 Ul O ID 44 Ul
0 — x— x « u 5.— —
, -i
tc t- o i
t— 0 3 Ul
in o o CXTJ 3
O O E f UOg^
— o ^ 3 t. 44 +5 (.Bog
ui o e — a. o « 44 ui S
oo u«o>te o — u
>— p m 44 ^ o x ui -o t-
Cv> C M Ol C 0 **•
S.lc55S -§,55S-°
Ul 44 O C r\| — 44 » E 44
44 L. L. in io i— KIO i. o in 0 .
«J 44 Ul 44 3 5 . C 44 > 0 j= C
0NC— COr— _J 0 C •— L. a
3
44 u U O
e . o i. e t.
.tfl
oi.io044«-ce —
U 44 ^ 0 C 0 o
0, « "* O> 44 8
X C Ol M » L. 44 Q.X
•0 O X— . O Ol 0OO .
ui ui "xoin T3 5 ?
0— UI(3->OC>O 44
— 8— x v> x.e o 'o a.
S44 — C 0 C T!
cuig-o-oooui
^
^
3
c
c
o
u
3-6
-------
Ul
2
g
in
fe
M
2
u.
1
u
i
o
o o
51
Ul U
_i e
V) U
DEVICE
^J
C
1
z
V)
5
.
cc
S
h-t
to
1
«
o
rj
i
m
Ul
t-
Ul
Ul
O 9)
•w o
O in
>> u c
•p ex o
> c +*
^" IP* *^
+J •a
— in c
in o o
e oo
CD C
u
o
x5
zl
£ "*
a. e
o o
3
1
a. ui
3 C
Ul E
>> u
~ a
1
Ul
5 §
§S
E 1-
C CD
o -o
*- I/I
> c
c o
Ul O
in
e
o
•p a
* -o
to «•
I/I
c
"
x
o >•
§£
11
Ul
a:
e
u
1
o
§
u
01 E
u oi
O) r~ S In
«> ffl ~ E vi
1"-a5 S°
§= S T" «-
fO "*-
•— O (D ¥ -°
e c k m o ui
«i e a o §"•— <•>
U > -P i "S "o
01
c
o
c
o
-
+* t. tn
03 C >9lO -M 3
^ d t» JS Q £ • O •»• •*•>** ^3O "
>^^->UlC3J= OU) t. «1
04- ojocrx+jo::— O-MC
a>ciu a> e-M^» +J c O
O3 CQOi--OO O) O *•
c u -^ c c c > -a G*OU-M
0.^3 — o -a o o -- c Q.-P a, y
a.uio^cu(ouiuE&u^*^
— O *> OV-4J >, 3 — 0 X •-
t-uic>>a.ccc0y.— u ^ &
+J *^ O , ui — >> o>
**• Ul JO .0 ^- 3
> O <— O •
•P u £ 1 OI
•*- ui 01 -i-> ui u in
*• JZ - 10 CB C >>
Ul *O W 3 ^^ O Ul
I
u
8
>
'
t
t JJ
0) A
€2
O (J
Ul V)
<
*J 01
= 01 U
5* 5 3 ^ §
§ C lo — E
0 I- U O
E O *-
H« 0.+J I- •
o e c o £
:» u o a a. oi
C Ul O 01 Ul
— •o o -s >>
C C O -H W
Ul Q 10
en ui • j: o ^
C Ul 'CD V V -4-*
JO *- -P >*•
66 » t- t- t-
8 t, o 10 o
« oi
O c
I- C JL CB •• I it. -^
01 » o 3 v t. atom
•P E +> « « 1. 3 •
(O /O £ _ i_ <*- • £ U
ini+JiSSSSui^Suii^
4^CD -HUUIUICD'Q
CB c c .c -a ci 3 0.13 —
i- — o aicc-P-cgsc E
OI >
O CD 3 O t «
a •<-> cr-o i»> E
« 01
•Q Ul U, J3
8 CD .p •— +J
» Ul E ffl U -P -P
"O <— O Ul «l 1- 9C
4>O • J3 CO CO -P •— •(-> ^ O
C O 3 O > Ul "3
§roffl> -a""?) c*"
£« ceD*->-WQ.U1 OI
r-»Q.<»»O+'u<«" -P*O4^
Or-UI U U CD C i— O
)— ^ -S •-• — +J «• Q.+J E MI i
2-
CD
§
a
8
L.
CD
Ul
c
o
•o
o.
• 1 <-
•0 CO t. « CO .
CO I- O O a,r—
3 e *> <^ t. ui
10 ui • >.-o a. a. o
10 O » -M C 3 —
It-
o
§
a
0
a.
o
Ul
O
o
•M
a.
3
Ul
§
+J
10
u
c
o
u
§
^
S
•a
in
f
|
§""
0
^.
i
e
o
a
-------
of the control device for the emission stream in question, may affect the
desirability or consideration of using the control technique. Therefore,
Figure 3-1 and Tables 3-1 and 3-2 provide the information needed to select the
appropriate control technique.
3.1.1.1 Thermal Incinerators--
Thermal incinerators are used to control a wide variety of continuous
emission streams containing VOC's. Compared to the other techniques, thermal
incineration is broadly applicable; that is, it is much less dependent on HAP
characteristics and emission stream characteristics. Destruction efficiencies
up to 99+ percent are achievable with thermal incineration. Although they
accommodate minor fluctuations in flow, thermal incinerators are not well
suited to streams with highly variable flow because the reduced residence time
and poor mixing during increased flow conditions decrease the completeness of
combustion. This causes the combustion chamber temperature to fall, thus
decreasing the destruction efficiency.
Thermal incineration is typically applied to emission streams that are
dilute mixtures of VOC and air. In such cases, due to safety considerations,
concentration of the VOC's is generally limited by insurance companies to 25
percent of the LEL (lower explosive limit) for the VOC in question (see
Section 4.1.2 for more details). Thus, if the VOC concentration is high,
dilution may be required.
When emission streams treated by thermal incineration are dilute (i.e.,
low heat content), supplementary fuel is required to maintain the desired
combustion temperatures. Supplementary fuel requirements may be reduced by
recovering the energy contained in the hot flue gases from the incinerator.
For emission streams with high heat contents (e.g., >150 Btu/scf), the
possibility of using the emission stream as fuel gas should be considered.
Packaged single unit thermal incinerators are available in many sizes to
control emission streams with flow rates from a few hundred up to about
100,000 scfm.
3-8
-------
3.1.1.2 Catalytic Incinerators--
Catalytic incinerators are similar to thermal incinerators in design and
operation except that the former employ a catalyst to enhance the reaction
rate. Since the catalyst allows the reaction to take place at lower tempera-
tures, significant fuel savings are possible with catalytic incineration.
Catalytic incineration is not as broadly applied as thermal incineration
since performance of catalytic incinerators is more sensitive to pollutant
characteristics and process conditions than is thermal incinerator performance.
Materials such as phosphorus, bismuth, lead, arsenic, antimony, mercury, iron
oxide, tin, zinc, sulfur, and halogens in the emission stream can poison the
catalyst and severely affect its performance. [Note: Some catalysts can
handle emission streams containing halogenated compounds.] Liquid or solid
particles that deposit on the catalyst and form a coating also reduce the
catalyst's activity by preventing contact between the VOC's and the catalyst
surface. Catalyst life is limited by thermal aging and by loss of active
sites by erosion, attrition, and vaporization. With proper operating
temperatures and adequate temperature control, these processes are normally
slow, and satisfactory performance can be maintained for three to five years
before replacement of the catalyst is necessary.
Catalytic incineration is generally less expensive than thermal incinera-
tion in treating emission streams with low VOC concentrations. Emission
streams with high VOC concentrations should not be treated by catalytic
incineration without dilution since such streams may cause the catalyst bed to
overheat and lose its activity. Also, fluctuations in the VOC content of the
emission stream should be kept to a minimum to prevent damage to the catalyst.
Destruction efficiencies of up to 95 percent of HAP's are generally
achieved with catalytic incineration. Higher destruction efficiencies (99
percent) are also achievable, but require larger catalyst volumes and/or
higher temperatures.
Catalytic incinerators have been applied to continuous emission streams
with flow rates up to about 100,000 scfm.
3-9
-------
3.1.1.3 Flares--
Flares are commonly used for disposal of waste gases during process
upsets (e.g., start-up, shut-down) and emergencies. They are basically safety
devices that are also used to destroy waste emission streams.
Flares can be used for controlling almost any VOC emission stream.
They can be designed and operated to handle fluctuations in emission VOC
content, inerts content, and flow rate. There are several different types of
flares including steam-assisted, air-assisted, and pressure head flares.
Steam-assisted flares are very common and typically employed in cases where
large volumes of waste gases are released. Air-assisted flares are generally
used for moderate relief gas flows. Pressure head flares are small; they are
used in arrays of up to 100 individual flares. Normally, only a few of the
flares operate. The number of flares operating is increased as the gas flow
increases.
Flaring is generally considered a control option when the heating
value of the emission stream cannot be recovered because of uncertain or
intermittent flow as in process upsets or emergencies. If the waste gas to be
flared does not have sufficient heating value to sustain combustion, auxiliary
fuel may be required.
Based on studies conducted by EPA, 98 percent destruction efficiency
can be achieved by steam-assisted flares when controlling emission streams
with heat contents greater than 300 Btu/scf. Depending on the type of flare
configuration (e.g., elevated or ground flares), the capacity of flares to
treat waste gases can vary--up to about 100,000 Ib/hr for ground flares and
2,000,000 Ib/hr or more for elevated flares. The capacity of an array of
pressure head flares depends on the number of flares in the array.
3.1.1.4 Boilers/Process Heaters--
Existing boilers or process heaters can be used to control emission
streams containing organic compounds. These are currently used as control
devices for emission streams from several industries (e.g., refinery operations,
SOCMI reactor processes and distillation operations, etc.).
3-10
-------
Typically, emission streams are controlled in boilers or process
heaters and used as supplemental fuel only if they have sufficient heating
value (greater than about 150 Btu/scf). In some instances, emission streams
with high heat content may be the main fuel to the process heater or boiler
(e.g., process off-gas from ethylbenzene/styrene manufacturing). Note that
emission streams with low heat content can also be burned in boilers or
process heaters when the flow rate of the emission stream is small compared to
the flow rate of the fuel/air mixture.
When used as emission control devices, boilers or process heaters
can provide destruction efficiencies of greater than 98 percent at small
capital cost and little or no fuel cost. In addition, near complete recovery
of the emission stream heat content is possible.
There are some limitations in the application of boilers or process
heaters as emission control devices. Since these combustion devices are
essential to the operation of a plant, only those emission streams that will
not reduce their performance or reliability can be controlled using these
devices. Variations in emission stream flow rate and/or heating value could
adversely affect the performance of a boiler or process heater. By lowering
furnace temperatures, emission streams with large flow rates and low heating
values can cause incomplete combustion and reduce heat output. The perform-
ance and reliability of the process heater or boiler may also be affected by
the presence of corrosive compounds in the emission stream; such streams are
usually not destroyed in these devices.
3.1.1.5 Carbon Adsorbers--
Carbon adsorption is commonly employed as a pollution control and/or a
solvent recovery technique. It is applied to dilute mixtures of VOC and air.
Removal efficiencies of 95 to 99 percent can be achieved using carbon
adsorption. The maximum practical inlet concentration is usually about 10,000
ppmv. The inlet concentrations are typically limited by the adsorption
capacity of the carbon bed or safety problems posed by high bed temperatures
produced by heat of adsorption and presence of flammable vapors. Outlet
3-11
-------
concentrations of 50 to 100 ppmv can be routinely achieved with state-of-the-
art systems; concentrations as.low as 10 to 20 ppmv can be achieved with some
compounds. In contrast to incineration methods whereby the VOC's are destroyed,
carbon adsorption provides a favorable control alternative when the VOC's in
the emission stream are valuable.
High molecular-weight compounds that are characterized by low
volatility are strongly adsorbed on carbon. The affinity of carbon for these
compounds makes it difficult to remove them during regeneration of the carbon
bed. Hence, carbon adsorption is not applied to such compounds (i.e., boiling
point above 400°F; molecular weight greater than about 130). Highly volatile
materials (i.e., molecular weight less than about 45) do not adsorb readily on
carbon; therefore, adsorption is not typically used for controlling emission
streams containing such compounds.
Carbon adsorption is relatively sensitive to emission stream condi-
tions. The presence of liquid or solid particles, high boiling organics, or
polymerizable substances may require pretreatment procedures such as filtra-
tion. Dehumidification is necessary if the emission stream has a high humidity
(relative humidity >50 percent) and cooling may be required if the emission
stream temperature exceeds 120 - 130°F.
To prevent excessive bed temperatures resulting from the exothermic
adsorption process and oxidation reactions in the bed, concentrations higher
than 10,000 ppmv must frequently be reduced. This is usually done by conden-
sation or dilution of the emission stream ahead of the adsorption step.
Exothermic reactions may also occur if incompatible solvents are mixed in the
bed, leading to polymerization. If flammable vapors are present, the VOC
concentrations may be limited by insurance companies to less than 25 percent
of the LEL. If proper controls and monitors are used, LEL levels up to 40-50
percent may be allowed.
Packaged carbon adsorption systems are available that can handle
emission streams with flow rates from a few hundred to above 100,000 scfm.
3.1.1.6 Absorbers (Scrubbers)--
Absorption is widely used as a raw material and/or a product recovery
technique in separation and purification of gaseous streams containing high
3-12
-------
concentrations of VOC's. As an emission control technique, it is much more
commonly employed for inorganic vapors (e^g., hydrogen sulfide, chlorides,
etc.) than for organic vapors. Using absorption as the primary control
technique for organic vapor HAP's is subject to several limitations and
problems as discussed below.
The suitability of absorption for controlling organic vapor emissions
is determined by several factors; most of these factors will depend on the
specific HAP in question. For example, the most important factor is the
availability of a suitable solvent. The pollutant in question should be
readily soluble in the solvent for effective absorption rates and the spent
solvent should be easily regenerated or disposed of in an environmentally
acceptable manner.
Another factor that affects the suitability of absorption for
organic vapor emissions control is the availability of vapor/liquid equi-
librium data for the specific HAP/solvent system in question. Such data are
necessary for design of absorber systems. For uncommon HAP's, these data are
not readily available.
Another consideration involved in the application of absorption as a
control technique is disposal of the absorber effluent (i.e., used solvent).
If the absorber effluent containing the organic compounds is discharged to the
sewer, pond, etc., the air pollution problem is merely being transformed into
a water pollution problem. Hence, this question should be addressed (e.g.,
are there chemical/physical/biological means for treating the specific
effluent under consideration?). In solvent recovery, used organic solvents
are typically stripped (reverse of absorption) and recycled to the absorber
for economic reasons. However, in HAP control applications, stripping
requirements will often be very expensive because the residual organic
concentrations in the solvent must be extremely low for it to be suitable for
reuse. Also, if the VOC's in the effluent from the absorber have appreciable
vapor pressure (e.g., > 0.1 mm Hg), the possibility of .VOC emissions to the
atmosphere should be considered.
In organic vapor HAP control applications, low outlet concentrations
will typically be required. Trying to meet such requirements with absorption
3-13
-------
alone will lead to impractically tall absorption towers, long contact times,
and high liquid-gas ratios that may not be cost effective. Therefore,
absorbers will generally be effective when they are used in combination with
other control devices such as incinerators.
Removal efficiencies in excess of 99 percent can be achieved with
absorption.
3.1.1.7 Condensers--
Condensers are widely used as raw material and/or product recovery
devices. They are frequently applied as preliminary air pollution control
devices for removing VOC contaminants from emission streams prior to other
control devices such as incinerators, adsorbers, or absorbers.
Condensers are also used by themselves for controlling emission
streams containing high VOC concentrations (usually > 5,000 ppmv). In these
cases, removal efficiencies obtained by condensers range from 50 to 90 percent.
The removal efficiency of a condenser is dependent on the emission stream
characteristics including the nature of the HAP in question (vapor pressure-
temperature relationship) and HAP concentration, and the type of coolant used.
Note that a condenser cannot lower the inlet VOC concentration to levels below
the saturation concentration (or vapor pressure) at the coolant temperature.
When water, the most commonly used coolant, is employed, the saturation
conditions represent high outlet concentrations. For example, condenser
outlet VOC concentrations are often limited to above 10,000 to 20,000 ppmv due
to the saturation conditions of most of the organic compounds at the tempera-
ture of the cooling water. Therefore, it is not possible for condensation
with water as the coolant to achieve the low outlet concentrations that would
be required in HAP control applications.
Removal efficiencies above 90 percent can be achieved if lower
temperatures than those possible with cooling water are employed. These low
temperatures can be obtained with coolants such as chilled water, brine
solutions, or chlorofluorocarbons. However, for extremely low outlet HAP
concentrations, condensation will usually be economically infeasible.
3-14
-------
Depending on the type of condenser used, there may be potential
problems associated with the disposal of the spent coolant. Therefore, using
contact condensers that generate such effluents for controlling HAP emissions
is not recommended.
Flow rates up to about 2,000 scfm can be considered as representa-
tive of the typical range for condensers used as emission control devices.
Condensers for emission streams with flow rates above 2,000 scfm and contain-
ing high concentrations of noncondensables will require prohibitively large
heat transfer areas.
3.1.2 Control Techniques for Inorganic Vapor Emissions from Point Sources
Inorganic vapors make up only a small portion of the total HAP's
emitted to the atmosphere. Potential sources of the various inorganic vapors
found in the atmosphere are discussed in Chapter 2. Inorganic HAP vapors
typically include gases such as ammonia, hydrogen sulfide, carbonyl sulfide,
carbon disulfide, metals with hydride and carbonyl complexes, chlorides,
oxychlorides, and cyanides.
In many cases, although the inorganic HAP's are emitted as vapors at
the emission source, they may condense when passing through various ducts and
form particulates. Prior to discharge to the atmosphere, these particulates
are typically controlled by methods that will be discussed in Section 3.2. In
this section, the discussion will be based on control techniques for HAP's
that are emitted as vapors to the atmosphere.
There is only a limited number of control methods which are applicable
to inorganic vapor emissions from point sources. The two most commonly used
control methods are absorption (scrubbing) and adsorption. Absorption is the
most widely used and accepted method for inorganic vapor control. Although
combustion can be used for some inorganic HAP's (e.g., hydrogen sulfide,
carbonyl sulfide, nickel carbonyl), typical combustion methods such as thermal
and catalytic incineration are generally not applied. In some cases, for
example, in controlling hydrogen sulfide emissions from gas wells and gas
processing, flares are used.
3-15
-------
Applicability of absorption and adsorption as control methods
depends on the individual emission stream characteristics. The removal
efficiencies that can be achieved will be determined by the physical and
chemical properties of the HAP under consideration. Other factors (e.g.,
waste disposal, auxiliary equipment requirements), while not necessarily
affecting the technical feasibility of the control device, may affect the
decision to use that particular control method.
In the following two subsections, the applicability of absorption
and adsorption for controlling inorganic vapor emissions will be discussed.
3.1.2.1 Absorbers (Scrubbers)--
Absorption is the most widely used recovery technique for separation and
purification of inorganic vapor emissions. The removal efficiency achievable
with absorbers can be greater than 99 percent. It will typically be
determined by the actual concentrations of the specific HAP in gas and liquid
streams and the corresponding equilibrium concentrations. Table 3-3
summarizes the reported efficiencies for various inorganic vapors employing
absorption as the control method.
As discussed in Section 3.1.1.6 for organic vapors, the suitability
of absorption for controlling inorganic vapors in gaseous emission streams is
dependent on several factors. The most important factor is the solubility of
the pollutant vapor in the solvent. The ideal solvent should be non-volatile,
non-corrosive, non-flammable, non-toxic, chemically stable, readily available,
and inexpensive. Typical solvents used by industry for inorganic vapor
control include water, sodium hydroxide solutions, amyl alcohol, ethanolamine,
weak acid solutions, and hypochlorite solutions. Other factors which may
affect inorganic vapor absorption are similar to those for organic vapor
absorption (see Section 3.1.1.6).
Water is the ideal solvent for inorganic vapor control by absorption.
It offers distinct advantages over other solvents, the main one being its low
cost. It is typically used on a once-through basis and then discharged to a
wastewater treatment system. The effluent may require pH adjustment to
precipitate metals and other HAP's as hydroxides or salts; these are typically
less toxic and can be more easily disposed of.
3-16
-------
TABLE 3-3. CURRENT CONTROL METHODS FOR VARIOUS INORGANIC VAPORS*
ABSORPTION
ADSORPTION
Reported Removal Reported Removal
Inorganic Vapor Efficiency (%) Solvent Efficiency (%) Adsorbent
Mercury (Hg)
Hydrogen chloride
(HC1)
Hydrogen sulfide
(H2S)
Calcium fluoride
(CaF2)
Silicon tetra-
fluoride (SiF4)
Hydrogen fluoride
(HF)
Hydrogen bromide
(HBr)
Titanium
tetrachloride (TiCl4)
Chlorine (C12)
Hydrogen cyanide
(HCN)
95
95
98
95
95
85-95
99.95
99
90
Brine/hypochlorite
solution
Water
Sodium carbonate/
Water
Water
Water
Water
Water
Water
Alkali solution
90 Sulfur- impregnated
activated carbon
100 Ammonia- impregnated
activated carbon
99 Calcined alumina
Ammonia-impregnated
activated carbon
Source: Reference 1.
3-17
-------
3.1.2.2 Adsorbers--
When the removal of inorganic vapors is especially difficult using
absorption methods, adsorption may prove to be more effective. Adsorbents
such as activated carbon, impregnated activated carbon, silica gel, and
activated alumina are capable of adsorbing various inorganic vapors and gases.
The degree of adsorption is dependent not only on the waste stream characteris-
tics, but also on the different characteristics of the adsorbents.
Carbon adsorption, using conventional and chemically-impregnated carbons,
is widely used for controlling inorganic vapors such as mercury, nickel
carbonyl, phosgene, and amines. For example, when mercury vapors are passed
through a bed of sulfur-impregnated carbon, the mercury vapors react with the
sulfur to form a stable mercuric sulfide. Over 95 percent of the mercury
removed in this way can be recovered for reuse.
Important factors to consider when choosing an adsorbent for inorganic
vapor control are very similar to those for organic vapor control, which are
discussed in Section 3.1.1.5. Some of these factors include the amount of
adsorbent needed, temperature rise of the gas stream due to adsorption, ease
of regeneration, and the useful life of the adsorbent. Most of the reported
removal efficiencies for inorganic vapors are for activated carbon and impreg-
nated activated carbon, and range from 90 to 100 percent. Table 3-3 summarizes
removal efficiencies reported for various inorganic vapors controlled by
adsorption.
Activated carbons are the most widely used adsorbents for inorganic vapor
control. In several cases, they have to be treated (i.e., impregnated with
chemicals) for effective application. Since activated carbons are relatively
sensitive to emission stream conditions, pretreatment of the emission stream
may be necessary. Pretreatment methods such as filtration, cooling, and
dehumidification may be required depending on the emission stream conditions.
Filtration is used to prevent plugging of the adsorber bed by any solids or
particles which may be in the emission stream. Ideal adsorption conditions
for impregnated activated carbons are relative humidities less than 50 percent
and gas stream temperatures below 130°F. Inorganic vapor concentrations are
not recommended to exceed 1,000 ppmv (preferably, less than 500 ppmv) when
activated carbon is used as an adsorbent.
3-18
-------
3.1.3 Control Techniques for Organic/Inorganic Vapor Emissions From
Process Fugitive Sources
Process fugitive emissions are defined in this report as emissions
from a process or piece of equipment that are being emitted at locations other
than the main vent or process stack. Process fugitive emissions include fumes
or gases which escape from or through access ports and feed and/or discharge
openings to a process. Examples include the open top of a vapor degreaser,
the slag or metal tap opening on a blast furnace, and the feed chute on a ball
mill. Process fugitive emission sources can also include vent fans from rooms
or enclosures containing an emissions source. An example would be a vent fan
on a perch!oroethylene dry cleaner or the vent fan on a press room. Other
examples of process fugitive sources include cooling towers and process
drains.
These sources can be controlled by add-on control devices once the
emissions from the sources are captured by hooding, enclosures, or closed vent
systems and then transferred to a control device. Because of the nature of
the opening (e.g., for access or maintenance), the opening through which
emissions escape cannot be totally enclosed or blanked off. Operators have to
access the equipment or materials have to be fed or discharged from the
process. For this reason, hoods or partial enclosures are used to control
emissions from such openings.
Proper hood design requires a sufficient knowledge of the process or
operation so that the most effective hood or enclosure can be installed to
provide minimum exhaust volumes for effective contaminant control. In theory,
hood design is based upon trying to enclose the process and keep all openings
to a minimum and located away from the natural path of containment travel.
Where possible, inspection and maintenance openings should be provided with
doors. In practice, hoods are designed using the capture velocity principle
which involves creation of an air flow past the source of containment suffi-
cient to remove the highly contaminated air from around the source or issuing
from that source and draw the air into an exhaust hood. The capture velocity
principle is based on the fact that small dust particles travel very short
3-19
-------
distances (in the order of inches) when thrown or emitted from a source and
therefore can be assumed to follow air currents. Vapors and gases exhibit the
same effects.
In practice, hood capture efficiency is very difficult to determine and
therefore, when evaluating a hood, one of the few parameters that can be
considered is the capture velocity. Standard design values of capture
velocity are available from the American Conference of Government Industrial
o
Hygienists in the Industrial Ventilation Manual (see Table 3-4). Once a
capture velocity has been determined, the volume of air required should be
based on maintaining this capture velocity at the emissions point furthest
from the hood. This capture velocity should be sufficient to overcome any
opposing air currents. For additional information, refer to Table A.7-1 in
Appendix A.7, where specific guidelines on hood design (type of hood, capture
velocity, etc.) are presented for several industries.
Very few measurements of hood capture efficiency have been conducted
(see References 3, 4, and 5). Hood capture efficiencies of between 90 and 100
percent are possible depending on the situation and the particular process
fugitive sources being controlled. For sources where operator access is not
needed and where inspection doors can be provided, efficiencies toward the
upper end of the range are achievable. For sources where emissions are more
diffused, for example, from printing presses, capture efficiencies of 90
percent may be difficult. In the flexible vinyl and printing industry, 90
percent is typically the upper bound for capture efficiency on coating
presses. In the publication rotogravure industry, capture efficiencies of 93
A
to 97 percent have been demonstrated based on material balances.
Once the process fugitive emissions are captured, the selection of the
control device will be dependent on the emission stream characteristics, HAP
characteristics, and the required overall performance levels (e.g., removal
efficiency). Note that the required performance level for the control device
will be determined by the capture efficiency. The factors that affect the
control device selection process are the same as for point sources; therefore,
refer to Sections 3.1.1 and 3.1.2.
3-20
-------
TABLE 3-4. RANGE OF CAPTURE VELOCITIES'
Condition of Dispersion
of Contaminant
Examples
Capture
Velocity, fpmc
Released with practically no
velocity into quiet air.
Evaporation from tanks;
degreasing, etc.
50-100
Released at low velocity into
moderately still air.
Spray booths; intermittent
container filling; low
speed conveyor transfers;
welding; plating; pickling
100-200
Active generation into zone
of rapid air motion.
Spray painting in shallow
booths; barrel filling;
conveyor loading; crushers
200-500
Released at high initial velocity
into zone of very rapid air motion.
Grinding; abrasive blasting; 500-2,000
tumbling
In each category above, a range of capture velocity is shown. The proper choice
of values depends on several factors:
Lower End of Range
1. Room air currents minimal or favorable to capture.
2. Contaminants of low toxicity or of nuisance value only.
3. Intermittent, low production.
4. Large hood-large air mass in motion.
Upper End of Range
1. Disturbing room air currents.
2. Contaminants of high toxicity.
3. High production, heavy use.
4. Small hood-local control only.
Source: Reference 2.
Jfpm =» feet per minute.
3-21
-------
For process fugitive emission sources such as process drains, the control
alternatives involve a closure.or a seal. A common method involves the use of
a P-leg in the drain line with a water seal. A less common, but more effective
method, is a completely closed drain system. Several factors affect the
performance of water-sealed drains in reducing organic emissions: drainage
rate, composition and temperature of the liquid entering the drain, diameter
of the drain, and ambient atmospheric conditions. Emission reductions of 40
to 50 percent may be achieved with water-sealed drains. In a completely
closed drain system, the system may be pressured and purged to a control
device to effectively capture all emissions. The control efficiency would
then depend on the efficiency of the control device; 95 percent should be
achievable.
Process fugitive emissions from cooling towers have not been reliably
quantified due to difficulties encountered in measuring them and no specific
control guidelines have been developed. Probably, the best control technique
currently available is close monitoring of heat exchangers and other equipment
to detect small leaks as they occur.
3.1.4 Control Techniques for Organic/Inorganic Vapor Emissions from Area
Fugitive Sources
The control measures that can be employed for controlling organic or
inorganic vapor emissions from area fugitive sources are basically the same.
Organic fugitive emissions have been more extensively studied than inorganic
fugitive emissions. Therefore, the following discussion will be primarily
based on organic vapors. However, control techniques for inorganic vapor
emissions will also be discussed.
Fugitive emissions of organic vapors occur in plants processing organic
liquids and gases, such as petroleum refineries, chemical plants, and plants
producing chemically-based products such as plastics, dyes, and drugs. One
group of emission sources found in plants of this type is commonly referred to
as equipment leaks. Fugitive emissions of this type result from incomplete
sealing of equipment at the point of interface of process fluid with the
3-22
-------
environment. Control techniques for equipment leaks include leak detection
and repair programs and equipment installation or configuration. The follow-
ing sections contain information about control techniques for common types of
processing equipment found in plants processing organic materials. Control
techniques and control efficiencies for common types of processing equipment
are summarized in Table 3-5.
Pumps--Several types of equipment or equipment configurations can be
used to eliminate or capture all organic vapors leaking from pump seals.
There are, first of all, leakless pumps, pumps designed with no interface
between the process fluid and the environment, such as diaphragm seal, and
canned pumps. These pumps effectively limit fugitive emissions. However,
they are limited in application.
Sophisticated pump seals can also be used to capture or eliminate
fugitive emissions. Dual seal systems with pressurized barrier fluids or low
pressure systems vented to control devices may be used in some applications.
Another approach which may be used involves venting the entire seal area to a
control device. Capture efficiencies should be virtually 100 percent in both
systems vented to control devices. Then, the overall control efficiency would
be limited by the efficiency of the control device.
Another approach to reducing (but not eliminating) organic fugitive
emissions from pumps is leak detection and repair programs. A leak detection
and repair system modeled after the one EPA developed for the New Source
Performance Standards (NSPS) of the synthetic organic chemical industry
g
(SOCMI) should achieve about 60 percent control efficiency. The efficiency
of leak detection and repair programs is dependent on several factors such as
frequency of monitoring, effectiveness of maintenance, action level, and
underlying tendency to leak. These factors and their effect on control
efficiency have been studied and discussed in References 8, 9, 10, and 11.
Also, models are available for calculating the effectiveness of leak detection
and repair programs (e.g., see Reference 12).
Valves--As with pumps, control of fugitive emissions from valves may
be accomplished by installing equipment designed to isolate the process fluid
from the environment. But, also as with pumps, leakless valves such as
diaphragm valves are limited in their application.
3-23
-------
TABLE 3-5. SUMMARY OF CONTROL EFFECTIVENESS FOR CONTROLLING
ORGANIC AREA FUGITIVE EMISSION SOURCES
Emission Source
Control
Technique/Equipment Modification
Control
Effectiveness
(percent)
Pumps
Valves
- gas
- light liquid
Pressure relief
valves
Open-ended lines
Compressors
Monthly leak detection and repair
Seal less pumps
Dual mechanical seals
Closed vent system
Monthly leak detection and repair
Diaphragm valves
Monthly leak detection and repair
Diaphragm valves
Rupture disk .
Closed vent system
Caps, plugs, blinds
Mechnical seals with vented degassing
reservoirs .
Closed vent system
Sampling connections Closed purge sampling
61
100
100
100
73
100
46
100
100
100
100
100
100
100
Source: Reference 8.
'closed vent systems are used to collect and transfer the fugitive emissions to
add-on control devices such as flares, incinerators, or vapor recovery systems.
3-24
-------
Leak detection and repair programs have been used to reduce fugitive
emissions from valves. As indicated above for pumps, the control efficiency
of a leak detection and repair program depends on the frequency of monitoring,
the effectiveness of maintenance, the action level, the underlying tendency to
leak, and other factors. A leak detection and repair program modeled after
the one developed by EPA for the NSPS of the SOCMI should achieve a control
efficiency of about 70 percent for valves in gas service and about 50 percent
g
for valves in light liquid service.
Pressure Relief Valves—Fugitive emissions from pressure relief
valves may be virtually eliminated through the use of rupture disks to prevent
leakage through the seal. Fugitive emissions may also be added to gases
collected in a flare system by piping the relief valve to a flare header. The
control efficiency, then, depends on the destruction efficiency of the flare.
If flares are operated in accordance with flare requirements recently estab-
lished by EPA for sources complying with NSPS, at least 98 percent control
efficiency should be achieved.
Open-ended Lines--Leakage of organic vapors through valve seats to
the open ends of pipes can be eliminated by the installation of caps, plugs,
or blind flanges. The control efficiency should be 100 percent as long as the
plugs and caps remain in place.
Compressors—Fugitive emissions from compressor seals can be
controlled by venting the seal area to a flare or other control device.
Barrier fluid systems can also be used to purge the seal area and convey
leakage to a control device. Capture efficiency should be 100 percent and,
therefore, the overall control efficiency would depend on the efficiency of
the control device.
Sampling Connections—Fugitive emissions from sampling connections
can be controlled by returning the purged material to the process or by
disposing of it in a control device. The practice of returning purged
material to the process in a closed system should achieve almost 100 percent
control efficiency. The control efficiency achieved by diverting the
collected purge material to a control device depends on the efficiency of the
device.
3-25
-------
Other area fugitive emission sources Include lagoons and ponds where
liquid waste streams containing organic compounds are disposed of. Emissions
and emission rates of organic vapors from such sources are not well documented,
and such sources are not easily controlled. The best method currently avail-
able for reducing emissions from lagoons and ponds is enhancing upstream
treatment processes, thereby minimizing the amount of organic material reach-
ing the lagoons and/or ponds.
Area fugitive emissions of inorganic vapors may be found in plants
processing inorganic chemicals, metals, electronics, and other products.
Although extensive work has not been done to quantify equipment leaks in
plants processing inorganic chemicals, it is expected that they would be
similar to equipment leaks encountered in plants processing organic chemicals.
Therefore, plants processing highly volatile compounds such as hydrogen
chloride or ammonia would be expected to benefit by the same control tech-
niques applied to reduce or eliminate fugitive emissions containing organic
compounds. There are a couple of differences to keep in mind, however.
First, for those control techniques that employ a control device to treat
collected vapors, the control device will probably differ. Instead of a
combustion device, an absorber, condenser, or an adsorber may be a more
appropriate choice. Another difference would be relevant to leak detection
and repair programs. Leak detection and repair programs for organic vapors
were developed using portable organic analyzers. Portable analyzers that
respond to the inorganic vapors of concern would have to be used for leak
detection of inorganic materials.
Controlling inorganic vapor emissions from area sources such as lagoons
and/or ponds where liquid waste streams containing volatile inorganic
compounds are disposed of is quite difficult: The best control method
currently available is minimizing the quantity of inorganic compounds reaching
the lagoon and/or pond by improving upstream treatment processes.
3.1.5 Control Device Selection for a Hypothetical Facility
This subsection illustrates the control device selection process discussed
in the previous sections for a hypothetical facility with several emission
3-26
-------
streams. Assume that the owner/operator of this facility has requested
assistance regarding the control of these emission streams. The data supplied
by the owner/operator are presented in Figures 3-2 through 3-7.
Emission Stream 1 (see Figure 3-2)--This stream is the same as that
described in Chapter 2 in the Example Case. Assuming the HAP control require-
ment to be 99 percent reduction, from Figure 3-1, the only applicable control
technique for this level of performance at concentration levels of -100 ppmv
is thermal incineration. The HAP concentration is less than 25 percent of the
LEL for the HAP (see Table B.2-1); hence, the concentration limit indicated in
Table 3-1 will not be exceeded. Also, the flow rate of Emission Stream 1
falls in the range of application indicated for thermal incinerators in Table
3-1.
Emission Stream 2 (see Figure 3-3)--Assume the HAP control requirement
for Emission Stream 2 to be 95 percent reduction. For this level of perform-
ance, the applicable control techniques for inlet concentrations of -500 ppmv
are thermal incineration, catalytic incineration, and absorption. If either
of the incineration techniques are applied, the concentration limit indicated
in Table 3-1 will not be exceeded since the HAP concentration is less than 25
percent of the LEL for the HAP (see Table B.2-1). The flow rate of Emission
Stream 2 falls in the range indicated as applicable in Table 3-1 for these
control techniques. The final selection of the control technique should be
based on the information presented in Table 3-2 and costs (see Chapter 5).
Emission Stream 3 (see Figure 3-4)--Assume the HAP control requirement
for Emission Stream 3 is 98 percent reduction. In this case, the inlet HA?
concentration falls outside the range indicated in Figure 3-1; therefore, none
of the control devices in this figure are applicable. According to Section
3.1.1.3, flares can be used to control emission streams with high heat contents;
hence, flaring can be considered an option. Also, if a boiler or a process
heater is available on site, it can be used to control Emission Stream 3.
Emission Stream 4 (see Figure 3-5)--Assume the HAP control requirement
for Emission Stream 4 to be 95 percent reduction. For this level of perform-
ance, the applicable control techniques for inlet concentrations of -1,000
ppmv are thermal incineration, catalytic incineration, carbon adsorption, and
absorption. If either of the incineration methods or carbon adsorption is
3-27
-------
CM
O
f
U)
Ul
U)
41
41
•o
in
U
3
c
o
O
Ul
ce
>—
1/1
i/i
1/1
C JU
•— 41
k
5 VI
I/I
e •»-
O E
O Ul
>> u-
= °
CT1 O
£?
I
o
3
•r»
i
u
o.
rvi
at *^ Qj
41 >i «*
i £ 3
o o
w
VI
3
3
X
LU
C
41
O
n
an
e
o
^m
S
c
41
»•*
(J
a.
k
trearo
e
Emissio
«£
1
1
a
=
4)
O
e
i
u
41
a
?*
8
3
O
I/I
e
o
1
Ul
a.
z
ca
i
i
a
„
e
I
(/I
VI
S
o
t.
a
e
o
u
VI
VI
US
<3
Source
u
i
1
a
0)
e
41
3
75
VI
^
z
s
1
I/I
e
Emissio
a
i
i
-O
V.
O
a
2
a
CM
*
•^
§•
a.
1
a
VI
<
a.
—
i
i
.a
41
1
Q
CM
en
CM
•H
w
w
41
ae
s.
3
U
*o
a_
-s
o
S
a
^*
M
*4
z
>
I
a
^
^
*H
•k
•K
*
*•»
C
2
c
o
u
e
en
o
Q-
"o
M
CM
f*y
CM
^
^
on ten
41
3
VI
o
3£
_;
o
w
*°.
a
CM
^^
^
U
VI
3
CO
«r
0
*"*
^^v
u
e
S
§
u
CM
a
^
J
3
c
a
o
cu
13
3
U
U
ce
^•,
x
,§
vS
U
VI
o
§
U)
(*)
e\T
*
S
U
41
Q.
VI
O
§
« VI i/l 3
s$s5
413: t.
T> «
o 41 ** a.
u> 41 «
u. -O II
O •* 41 <*>
^ e <-> -o
Q. fl C
£ e a
~ ** 0
VI -
S 4) en e
e vi
VI * OJ "^
• a. S
o cn 41 4t
41 • T3
C 41 VI
§>1T- U
W r— O
O
C I/I T3
.„ .^ _
MI i/i iq o vi
C 3 e 3
vi a 0 " S
§n O
<•» u
O i- 1O CM
•— tJ J= Q.
vi'S * ez
•*• TS 41 O
S « M •«- VI
*• C VI 41
e •** 41 t- »—
« •oe
VI 01 VI
O U 41 I/I •—
*u> e) vi vi jd
*• 41 41 i-
41 C U *•»
u s « e
VI 3 ^ > IQ
41 41
t. VI u U
ex. • u •— *->
— • 41 C «l
«~- S O> U
ifl O 3 U •*-
•o 5 c o c
41 *•» 41 II §1
(5 § (5 -i o
•ft ^< &
* -fi
01
O
i/l
irt
(J
Ol
-M
U
n3
ro
CD
LU
OJ
I
CO
(U
3
Ol
3-28
-------
Si
- §
»-i . u -H
U O
o
u
3
U
8
o
l/t
e
o
VI
VI
UJ
J^
*
oa
i
ja
i^
e
o
a
VI
VI
Ol
a
e
o
a
a
a
o
in
*
*
CM
*«t
fcl
C
3
0
o
a.
S
u.
i
i
.0
U.
f"-
e»
01
§
CO
CM
~
^J
0)
k
3
VI
V)
01
k
a.
«
3>
•*_
*
Cfl
1
1
J3
C^
3
e
•*••
^B
j3
"o
e
10
>>-
5
3
"o
1^,
=
3;
1
0
^3
1
O
u
«>
-•««•
CM
.
ex
o
a.
S
w
a.
V.
o
VI
13
a.
S
_
1
1
n
Ol
o
|
^
CM
9)
(0
*«i
CM
*
*,
01
01
t_
X
e
3
§
u
Ol
X
10
3
U
k
a.
ae
X
1
g
u
VI
o
o
o
o"
CM
2
k
Q.
V>
U
k
1/1
1
CM
k
3
VI
VI
Ol
z
5
5?
^^
X
S
u
a
01
**
k
a
h-
41
e
o
s
"*»
1
^
c^
^
VI
3
£
I
o
i
Ol
k
M
in
Ol
i
3
t/1
ul
^_
"oi
01
"3
k
1-1
e
o
o
1
's
Ol
se.
^
e
o
a
VI
"
e
o
x
b
Ol
e
u
^
u
^
3
a
u
•
o
s
Ol
e
u
e
1
01
i/i
o
5
Ol
o
k
e
0
u
a
Ol
u
01
Ol
I/)
2
O
k
c •
o -o c
*J Ol O
U i-
Q C vl
2*. VI
ins t-
o -F- uj S
Ol J= i Ol
r— W
Ol ^ vi
"3 S S
Si, g
f*
C Qb ^
•^ «• 4J a. e -i-
c <•-<->
01 S k
Soi ** a.
01 c vl
VI " AJ *pa
• S. e
TJ 01 Ol 01
S" ^3
49 VI
jQ -j -.
§>!•>- U
k P* O
u « a. k
i/i ex ex
k v» 3
O Ol VI k
u o
01 oi Oi ex
*— e a a
0) >
c vi a
•»• 'f- i— U VI
VI 3 •»- Oi
— Ol O C •—
u JE « a
IS.1" ?«
01 I/I IQ O VI
k *•> e 3
vi « o s
e no
cow u
o i— •—
01 JS O **
k *> « O vi
* k C k 3
X 01 0.^3
0 X k S
Ol 41 * k S
«-> a. o o
c at a.
oi vi e « s
vi 3 ••- >
(J
It)
i.
c
(U
CO
I
CO
a>
3
a>
3-29
-------
i
o
u.
t—
a
5
Ul
oc
h-
Z
0
,
i
«
U
's
i
s
-
41
4.
°
C
3
u
o
00
"»3
i/i
"^
e
o
•*•
t^
y
_)
e
o
Ul
£Z
O
-s
a
2
Ul
I.'
X
M
Contac
x
u
41
<
£
2
i
O
I/I
J
O
IO
1
41
a?
V.
e
Ul
S
41
b
V)
e
o
Ul
's
Ul
o
•
z
"a.
S
3
i/>
— •
e
41
t^
41
ul
en
c
•£
3
U
3
C
X
41
•o
41
•o
V
41
U
^
2
|
u
Identifi
a
a.
41
S
3
Z
n Stream
o
VI
VI
1 • f
«(
1
1
J3
,j
C
absorber
flj
S
4.
3
O
|
••••
S
Ul
a.
^^
03
1
1
A
,j
e
o
a
process
e
o
«
u
•*•
»**
VI
s
u
VI
{J
-
1
1
.a
u
^
^
.2
•c
4*
methvlen
Ul
a.
x
S
I
0
Ul
VI
Ul
Q
1
1
J3
u
a
*
organic
8
4.
C
*O
o
a.
35
Ul
1
1
A
I
a
o
8
«
<•>
•
3
c
u
0.
^g
u.
1
1
0
u.
£
en
I
CM
41
4.
u
u
^
»
a.
^™
53
1
1
ja
i.
^
•0
c
41
insolubl
CM
^
|
r~
(/I
a.
^^*
3-
1
1
a
e
41
4J
O
c
CM
ex
o
L.
|
VI
<
a.
'^
»
i
i
^a
41
1
d
in
03
CM
Ol
•^
U
^w
X
a.
32
T
^
U
^_
O
M
CO
^
^m+
^
«
«
I
C
O
U
'c
en
o
a.
41
O
C
CM
M
^
u
e Con ten
j»
M
Ul
^*
ve
41
C
O
c
**•.
It.
u
Ul
3
^J
CO
0
2
^
"^ •
e
3
§
u
CM
O
41
ca-
ii.
o
o
fl
«
•
41
3
tl
(U
!
^1^
_4
^
n
•*^
e
e
o
4?
HI
artlcu
°-
0£
^^
•a
2
u
41
a
X
41
U
Ul
0
s
o"
CO
tn
•^
3
(Q
05
X
lj^
£
^
c^
"^^
e
i~
Q
S
4)
X
V
"u
k
i
y
o
*4>
w
•»«•
L.
O
)—
41
e
o
e
*^
41
C
O
c
CM
»*
„,
41
c
S
0
^
Q
VI
|
•2
3
ce
- ui S
41 S, s 41
P— w
41 ^1 "I
ui 4i e ui
••> . c o
4) O •— i- j-
4. s -o ui »r
few °
Ul . 4) -~ -f-
• a S c/1
41 • ^ *"
e 4i M 'g
^3 V >»- u
O V. i— O
U O. I/I
— C £> n ti
en > ._
c ui ri . •"
•*• -~ r— U Ul 4->
Ul 3 ^ 41 I/I
""* S 5
g m ^» .p. (/I *^
o o w> w c»
>^ c ui 41 (U
ul 41 ul r~
«. S ui 41 4-
O 4. 41 VI r— U_
*. o ui ui ja iTi
at .c o n
1. S. o o ro
c 4i a.
4i ui c 10 e -.
41 ^" S !-
4. Ul U 4. -I
ft • i. •*• ^
-» 41 C VI O1
3 VI a -0 •!-
— e 01 u 1 1
3^ = 0^
41 ** 41 II Ol
fiatJC 4.
i— a H. -« o
* *
He
3-30
-------
LO
in
U!
u o
3 z
s s
o o
-3
£
41
li
CO
si
3 VI
M
§ i
U Ul
X *-
o o
m -. j
in
e
o
VI
I/I
"a
Ul
^
1
1
a
VI
VI
41
k.
a
e
e
a
8
1
o
I/I
VI
Ul
a_
<
03
1
ja
e
U
a
VI
proces
e
o
•*•
3
**»
•*•»
*ft
Ul
*o
o
8
s
o
«7
O
1
1
a
41
e
41
3
3
a.
S
I
I/I
e
o
VI
VI
'e
Ul
a
i
|
J3
O
a
•a
u
°e
10
at
o
a
k.
o
u.
o-
41
2
k.
«
3
4(
X
a.
=
-^
41
C
O
c
._,
*
«
41
M
e
2
c
o
u
"c
o>
u
o
a.
x
•a
~s
s
41
k.
O
<*)
a
CM
ntent
o
o
41
3
VI
O
£
^
^J
O
M
*o
o
CM
U
I/I
3
ca
CM
*
~
- _ir
e
2
e
o
o
CM
O
19
41
cr
u.
o
01
CO
CM
41
3
2
41
a.
a
41
_j
-
—
^
2
e
o
u
41
4-1
(O
f^
3
U
k.
|Q
S.
Of
X
J
vt
O
S
-H
CM
•*+
-*^
3
IT)
oe
§
*•
u.
X
«
*—
,
s
•»»
o
1
41
u
k.
A)
Q.
in
u
i/i
|
CM
^^
•«•#
41
3
VI
VI
u
k.
a.
z
5
CO
o
,,
kl
O
41
*J
t_
O
1-
41
e
o
e
^
o
e
*—*
CM
£
3
41
X
"^
i
o
S.
0
VI
^^
e
o
*3
41
at
41
2
u
"a.
0,
**
=
«
i
41
k.
a
assume
^B
41
>
41
*O
k.
e
o
o
•o
3
O*
41
ae
^
c
Q
'3
a
o
i/i
J= CJ O
<-> Ok.
o e
W « VI 41
e O.S
X VI VI 3
fc a. c -ii
C < -p- *»
41 Z k.
O 41 3 Q.
** 41 «O
k. 13 M
k. f
o «* 41 en
w c w -o
Q, us e
^5§«
VI .
S 41 0 C
O k. C O
41 O •— •—
k. e -o v>
4J C VI
r a.'s
O 01 41 41
S' O
41 VI
^~- 0 VI
0 k?i^ 0
U « 0. k.
vi a. a.
k. VI 3
O 41 VI k.
kl 0
41 4) 41 O.
cr> >
C VI -0
T- i- r— U Irt
VI 3 f- 41
— • 4» 0 C —
U £
41 VI j= a.
VI f. 5 . «t
VI 13 C z
f- -a 41 o
S fl 4J -^ VI
4- e vi 41
C T- 41 — r-
IQ OS
in 41 vi
k. S "l u
O k. 41 vi »—
>4» O VI VI A
ia k. c k. 3
4^ ^) ^1 r\
0 X k. S
241 18 J. O
O. O kl
e 4i a.
41 vi e ia S
vi 3 •>- > «
41 41
k. I/I U k.
CX • k. i- U
— 41 e vi
IO VI 2 lO
u— s 01 u
« TJ 3 k. i-
o o e o c
4i ** 41 n at
•C 41 •= k.
h- St- -H 0
* *t -*t
* *
*
ra
(U
c
o
0)
o
u.
t/l
o
(T3
fO
o
c
a»
0)
3
3-31
-------
applied, the concentration limit indicated in Table 3-1 will not be exceeded
since the HAP concentration is less than 25 percent of the LEL (see Table
B.2-1). . The flow rate of Emission Stream 4 falls in the range indicated as
applicable in Table 3-1 for these control techniques. The final selection of
the control technique should be based on the information presented in Table
3-2 and costs (see Chapter 5).
Emission Stream 5 (see Figure 3-6)--Assume the HAP control requirement
for Emission Stream 5 to be 98 percent reduction. Since this emission stream
contains inorganic HAP's, incineration techniques are not applicable. The
only control technique that is applicable for this level of performance and
inlet HAP concentrations of -20,000 ppmv is absorption. The flow rate of
Emission Stream 5 falls in the range indicated as applicable for absorption.
Emission Stream 6 (see Figure 3-7)--Assume the HAP control requirement
for Emission Stream 6 to be 90 percent. Also, assume that the owner/operator
has indicated his preference to recover the HAP in the emission stream. For
this level of performance, the applicable control techniques for inlet
concentrations of -13,000 ppmv are absorption and condensation. The final
selection among these techniques should be based on the information presented
in Table 3-2 and costs (see Chapter 5).
3-32
-------
«
X
g
i—
a
5
Ul
ac
z
0
Ul
3
a.
.
l
c
IO
0
IO
u
u
s
3
1
1
cj
i
Ul
in
in
en
e
41
e
o
a.
Of
(•*
(2
41
k
°
=
(3
u
O
CO
"?
41
S
I/)
e
o
i
.j
*•>
1
1
s
k
Ul
i
4^
u
e
e
o
o
41
O*
jj
^5
£
S
9
o
IO
1
1
u
41
o
e
Streams
e
o
VI
VI
5
0
0'
*
EL
M
3
VI
-*
3
S
X
Ul
VI
at
i
o
S
IO
k
o
a
Ul
10
41
k
^
in
e
o
1
e
P>4
w
C2
at
,_
41
i
I
(/I
Q
•£
VI
Ul
<
i
i
.a
VI
10
T
!£
a
S
IO
o
a.
IO
41
—
S
i
t/>
c
o
VI
VI
^B
i!
a.
CO
1
1
.a
w
e
0
a
Wl
S
o
k
01
e
o
ssifica
10
u
s
1
o
1
1
.a
IO
e
o
§
IO
M
a.
IO
41
is
e
o
VI
Ul
"a
Ul
a
i
>
a
k
i
u
c
Ol
k
1
a
0
u.
e
10
VI
iO
a.
Ul
i
i
-d
s
|
§
o"
CM
*
CM
*
S
e
o
o
a.
u.
i
i
A
u.
00
VO
Ob
3
•a
•9
00
_,
CM
—4
41
Pressur
a
ex
S
1
0
1
1
-a
41
a
k
a
CM
*
3
!••
O
a.
s
i
i
J3
e
01
o
B
-""*
CM
••4
a.
k
a.
4>
>
fc»
ex
0
v>
T3
1
••
1
—
1
.a
41
O
I
^
*^
IO
^^
CM
__.
Ol
41
3
k
10
3
U
41
O
X
a.
"3
—
j.
on tent**
jj
.ji
e
o
a.
0
*»
CM
— »
"1
CM
on tent
u
41
k
3
"o
^
^^
^^
Content
CM
0
w
^*
o-
u.
in
00
S
41
i.
3
*»
IO
k
41
t—
—J
«
^
c
s
§
o
s
iO
3
U
*-
u
VI
o
O
O
fl
•1
CM
~*
S
*ts
ce
o
u.
X
-
*"^
a
IO
5
e
3
41
•f—
k
Q.
U
k
QL
VI
§
IO
CM
•^
1>
k
3
VI
£
0.
X
^^
5
«
cTv
2?
o
41
£
O
H-
C
O
e
4»
e
o
s
_^
CM
3
41
X
e
%
o
<0
o
^^
VI
e
o
IO
en
41
ee
41
4)
k
M
SO
cn
i
VI
VI
IO
^—
41
41
«J
"o
k
i
o
41
3
o-
a.
e
o
o
VI
.a
VI
•o
o
o
k
I
o
4)
U
4)
41
3
'o
k
C •
00 C
u 4i a
41 •—
Ol ..vl
*5 vi « *f^
l M VI 3 1^
«-> a. c •»• ^
e < •— • «•> =
01 Z k =
0 4) 3 ex a;
k O H
O ** 41 <*> C/l
k C 5 TJ _
^SeS o
~ - ° . '-
S 41 cn c 'C
10 k C 0 VI
41 O -r- •— -f-
k S T3 w J=
VI * Ol •*• Q)
• a. S
O 01 41 41 J_
01 • O o
C 41 "1 ,P
•—— T3 VI «*-
S X-— 0 (/,
u 10 ex k ^
vi ex ex T-
k VI 3 j_i
O 41 V» k 7n
U O ^
41 41 01 CX >r~
"en 2 rti
C VI 0 • J"
•r- <^ ^- U VI +-1
VI 3 f- 41 U
— , C 3 u
vl iO O S
e o <•» u c
'vi •£ 'i . < 3
VI O C Z r—
i- O 41 O (^
S ia •-> •— vi (T
JS U *J **>
w
ox2*l S
S4I 10 k O ^
CX O U CT1
C 41 CX -r-
oi « e 10 '^^
41 41
k VI U k
CX • k f» fc»
— 4i e vi
S o e o 'e
Ol ** 41 H 01
J= U j= k
1— S K- -* O
* *
3-33
-------
I
If)
1
£
k
>
5
IO
41
O
3
§
o
** o
0-1-
« VI
fcS vi
e —
O S
O UJ
** a. u o
e 4i e
IO r- 41 •
•— 41 Ol O
I
o
s
a
3
a.
M
41—4)
2 9 S
^*
e
41
u
I
41
O
e
o
o
k
S
s
s
**
""
U3
e
o
1
!J3
1
u
J
a.
k
§
I
e
o
VI
VI
f~
S
UJ
<
1
1
J3
C
41
in
e
41
e
o
u
S
1
s
•*•
v>
VI
UJ
a.
^
CO
1
1
a
w
e
o
a
V)
VI
41
2
a
c
o
sslfica
o
S
k
3
O
CJ
1
1
.0
^
CM
•-4
u
01
^u
at
3
U
41
"o
<^
«(
3j
T
e
o
e
3
*
*
4-k
S
e
o
u
u
a
o
a.
4)
3
**•
w
d
*
CM
^—
on tent
sture
-*»
o
^
"o
&*
^
o
CM
"^
»*-
U
(A
3
oa
U)
^4
10
3
u
c
s
o
CM
o
iO
or
O
rt
A
CM
*
*-*
3
0
k
s
a
41
_J
^
CO
c
2
c:
o
CJ
2
«0
3
U
k
a.
o:
-™*
X
J
u
VI
0
a
a
*
CM
m
CM
I
at
Q
fBH
U.
X
_,
s
ia
o
e
0
k
41
Q.
VI
O
§
CM
*
ssure
41
k
a.
z
,^
m
<*»
o
^
M
u
o
1
J;
k
a
i—
41
C
O
e
-
%
o
e
(M
*
£J
VI
3
41
I
O
IO
X
o
VI
e
o
c
o
VI
e
41
e
a
u
e
a
k
c
V
VI
O
o
I
41
Z
ontrol 1
•o
S
u
41
V
•x
1
e .
o -a e
U k
79 s CL
Sg v.4.
6 £L *•*
i- • 4j n>
>,"« VI *3
** a. e •—
C < f- *>
41 X k
3iO
_ _ a.
"|3 ii
O l-> d) f-)
^ c 5 •«
a 10 e
j= e «
•^ ** o
VI •
S 41 ot e
3 k c o
ai g -^ •••
k S n v»
f C V»
VI « 41 *^
. a. e
•0 01 41 41
41 • •O
CD VI
T- -0 VI
1 k ^ 0
u « a k
vi a. a.
k m a
O 11 vi k
U O
41 U 41 Ck
— C A 10
e vi a •
VI 3 •»* 41
— ai o c: —
u ^: « o
£ iO VI Ot *^
fO Q. k fc»
41 VI ffl O VI
k ••» C 3
vi 10 O E
c no
c o ••» u
O •^ IO CM
•— »J A a.
VI t- » • <
VI t3 C Z
•— O K O
S —
«^ O VI VI A
v*-£%-Z
k O VI
q k C k 3
K 41 O..Q
•0 X k S
241 IO k O
a. o u
e 4) a.
4) VI C 10 S
v> 3 t- » 10
41 41
k VI U k
0. • k -v **
— > 41 e vi
IQ vi ja 10
TT O C O 'c
•C 10
41 ** 41 II C3>
« 41 J= k
^- SI— -t 0
* *
1C
0)
c
o
(U
S-
OJ
-!->
u
(O
03
a>
3-34
-------
3.1.6 References for Section 3.1
1. Control Technology for Toxic and Hazardous Air Pollutants.
McFarland, A. R. (ed.). Illinois Institute for Environmental Quality.
Chicago, 111. 1975.
2. Committee on Industrial Ventilation. Industrial Ventilation: A
Manual of Recommended Practice. 17th Edition. Lansing, MI. 1982.
3. U.S. EPA. Flexible Vinyl Coating and Printing Operations - Background
Information for Proposed Standards. EPA-450/3-81-016a. January 1983.
4. U.S. EPA. Publication Rotogravure Printing - Background Information for
Proposed Standards. EPA-450/3-80-03U. October 1980.
5. U.S. EPA. Measurement of Process Capture Efficiency. Draft Report of
Laboratory Testing. EPA Contract No. 68-03-3038. January 1984.
6. U.S. EPA. VOC Emissions from Petroleum Refinery Wastewater
Systems—Background Information for Proposed Standards. Research
Triangle Park, N.C. July 1984.
7. U.S. EPA. Assessment of Atmospheric Emissions from Petroleum Refining
Appendix B: Detailed Results. EPA-600/2-80-075c. April 1980.
8. U.S. EPA. VOC Fugitive Emissions in Synthetic Organic Chemicals
Manufacturing Industry--Background Information for Promulgated Standards.
EPA-450/3-80-0335. June 1982.
9. U.S. EPA. Fugitive Emission Sources of Organic Compounds-rAdditional
Information on Emissions. Emission Reductions, and Costs.
EPA-450/3-82-010. April 1982.
10. Wilkins, G. E., J. H. E. Stelling, and S. A. Shareef. Monitoring
and Maintenance Programs for Puroos and Valves in Petroleum and
Chemical Processing Plants: Costs and Effects on Fugitive
Emissions. Presented at Vlth World Congress on Air Quality, IUAPPA,
Paris. May 1983.
11. Wilkins, G. E., and J. H. E. Stelling. Monitoring and Maintenance
Programs for Control of Fugitive Emissions from Pumps and Valves in
Petroleum and Chemical Processing Plants. Presented at 1984
Industrial Pollution Control Symposium ASME. New Orleans, LA. February
1984.
12. U.S. EPA. VOC Fugitive Emission Predictive Model - User's Guide.
EPA-600/8-83-029. October 1983.
13. U.S. EPA. Evaluation of the Efficiency of Industrial Flares. Background
- Experimental Design - Facility. EPA-600/2-83-070. August 1983.
•
3-35
-------
3.2 PARTICULATE EMISSIONS CONTROL
Section 3.2.1 discusses add-on participate control devices, and
presents guidelines that are used to determine the applicability of
each control device. Section 3.2.2 discusses control techniques that
are used to reduce fugitive particulate emissions.
3.2.1 Control Techniques for Particulate Emissions from Point Sources
Three types of control devices applicable to particulate-laden
emission streams from point sources are discussed below: fabric filters
(baghouses), electrostatic precipitators (ESP), and venturi scrubbers.
The control efficiencies and applicability of these devices depend on the
physical and chemical/electrical properties of the airborne particulate
matter under consideration. Brief descriptions of each of these control
devices appear in the subsections that follow.
Selection of the control devices themselves depends on the specific
stream characteristics and the parameters (e.g., required collection
efficiency) that affect the applicability of each control device.
Table 3-6 identifies some key emission stream characteristics that
affect the applicability of each device. Matching the characteristics
of. the emission stream under consideration with the corresponding infor-
mation presented in Table 3-6 will identify those techniques most suited
to control the stream. This does not imply that a given control device
should be excluded at this point, however. In general, the parameters
listed in Table 3-6 are given as typical guidelines and should not be
taken as absolute, definitive values. Gas stream pretreatment equip-
ment can be installed upstream of the control device (i.e., cyclones,
precoolers, preheaters) which enable the emission stream to fall within
the parameters outlined in Table 3-6.
The temperature of the emission stream should .be within 50 to 100°F
above its dew point if the emission stream is to be treated (i.e.,
particulate matter collected) by an ESP or a fabric filter. If the
emission stream temperature is below this range, condensation can occur;
condensation can lead to corrosion of metal surfaces, blinding and/or
deterioration of fabric filter bags, etc. If the emission stream is »
3-36
-------
oo
^"
LU
i^r*
i—
oo
^•^
O
OO
oo
z:
LU
LU
I—
^J
^5
Cj
i_^
i
a.
3£
o
LL.
00
O
P
OO
3
LU
H-
CJ
^C
3£
Z
0
^r
^£
LU
a;
2
O
t/}
00
1— t
z
LU
LU
**
^Q
1
ro
LU
h-
4)
t » >
W +•>
3 C
4>> Ol
V) 4->
•r- C
O O
X 0
^s^
co
V) >>
41 4-)
C ••-
4J >
> f-
•r- 4J
U) CO
O -r-
8. 0-1
i. 41
O OS
U
4)
U
3
TO
^
4)
a.
4)
r—
C
4) O
O 4) 4-»
4-> •!- 4->
C_ OO -r-
Q- •!-
^J
4) >>
•— U
XJ C
TO 4J 4)
> •*- cn
4) U C
.C <*- O£.
U <«-
< LU
O 4)
C_ U
^j .r-
C >
O 4)
O 0
00
4J
am-
c o
TO
U C
o> •
2 "c TO
O 41
4) O C.
> 4J
•r- 4) VI
•r- 3 CO
CO 4-> C
4) ••- -r-
10 O co
^ -12
O C £
•o
4J 1
t- (/I
1 -I- -r-
0 3 E
^ QJ
*J ^ (/>
C OJ -r- E
u o o>
»— 'Q ^ •*— *
fl £ O t/1
-r* O
U C C
OJ O t- O
OO 4-> <•!-. t/>
•
Ul
r— CT
^3 ^3
4) 4-»
C ^3
4) 4->
Ol-r-
E
O -f-
Z1~
4) 4J E
^-> •— a.
TO O
t_ -r- ITS
44 +•* *
f^ t« ^^
O TO
CL c
• TO
f 4-> 4-> «-
'•o 5 t- -»->
*" ** c
c i/i c- ^
4) O) fl T-
t3 ^ r— "O
»*
•^
c^
^^
o
+J
Q.
•r^
C.
3
+->
C
r- ^J <*_ yl
+J !_ -i- 1 4)
•r* 3 *^~ Ol *
3 4-> - — • 1/1 c e
VI ^— C !T3 TO
E O > on u (.
TO E ** .-*->
U ^ OO "3 2. ^
• +j cn o 3 c
(/»•<-« +J O
.C 4) • t/l -r-
»•-• • "^ «^~ t/1
O >>••- 4> O ^
<- r-~-" C E "~
+J 4-> -f- O 4)
O "*
TO O) O O -r~ 4—
C_3 t- O l/l 4-J O
+J • CO
c -a i t_ - 4>
TO 4) -i- o a. in t_
4^ C* ^7 If— QJ SS) •!— *r-
CO *r* C > LU O IJ f~ *4-»
•r— 3 O I/I ->"o 'O3O'r-4->
Or— 3 ^4IC C03
•r~ TO O* Ol f"™ O 4—* '"• +^ •
CO *r» 4) C ^"* O >fm O O t» >r~ ^)
O ^ ^ *r~ »•"• '^* I * C C» 4) -4J ^_
L* 4J C f~ 4^ *r™ ] • ( * c/^ TO
{^ 4-J >•) O Ol ^ ^3 4J C 4«J C ?S
Q TO TO »r~ *r~ TO ^3 f o m f~i m
<~> E sc 4-> x: o,<: TO u E o .c
a.
3
•
* LL.
^^O
— O
TO O
c_ •>
^J (— ^
c
4) O
£
co 4)
a. u 4) <-
OO mr— I/I Ni O 4)
I i i (4M O) *r* 4J 4^
14- i— CO 41
• 4) 0 E e
r— T- C ~1 <«
TO 4-> 4-> -i- -i-
L. CO (- CVJ "O
c 41 a. c d e
04 f— -r- 3
4) *J C O LO
C t- ••— its (_ •
p— i TO 5 t- <*- O
*«
•T*
O^
O
-C C •<—
•r- O 4J t. CO
5 co o O C
4) ^ t—
U CO 4-> •!- C >4—
•^- CO I/I -r- O
<4- C -i- C
14— O O 4) CO -M
4) Bf^ E I/I 4) C
CO O> 4)
C- CO f >, C 4->
o E T- 41 i: o
a. 4) .e > u o
.
>.
t
t- TO
4) to
*4IB CJ 4^ C
4) co O
r™" C *r* *r~
TO CO CO
•f- co 4J O
O 4> «- «-
4» Q. !-
a. >, o o
oo +-> +J o
+J
3 ai4->
c -a c 3 •
o •«- o u
4J "O .C 4J
4J Q. 4J •*-* »^
O) 4-> O 5 O
•a x o
C ^ 4) Lu OJ
4) 4) O t-
Q.J3 4-1 o a.
OJ •!— O U">
Q«4- C LO TO
4-*
C t/l
<&*&€:
•G-3*
•r- -r- oo
<4- 4-> • •
'aj ^ ° a!
a, o 4->
4-> +J 4)
fQ +>J ^*^ *^3
O) -r- • •!-
— 1 2 O -O
a*
+
^^
^^
o
^
a.
4)
3
O
.C
31
S
•
PS
TO
41
4^
CO
, 3-37
-------
above this range, optimal HAP collection may not occur; by lowering
the emission stream temperature, the vapor component of the HAP is
reduced and, thus, an ESP or fabric filter will collect the HAP more
effectively. Procedures for determining the dew point of an emission
stream are provided in Appendix B.2; brief discussions of gas stream
pretreatment equipment is presented in Appendix B.ll.
Table 3-7 identifies general advantages and disadvantages for
each particular control device. Table 3-7 is used to provide additional
information on other considerations that, while not necessarily
affecting the technical feasibility of the control device for the
stream, may affect the overall desirability of using the device for a
given emission stream. Thus, Tables 3-6 and 3-7 used in conjunction
provide guidelines to determine if a particular control device could and
should be used for a given emission stream. Further design criteria
(Chapter 4) and cost (Chapter 5) must be considered to enable a
complete technical evaluation of the applicability of these devices
to an emission stream.
3.2.1.1 Fabric Filters —
Fabric filters, or baghouses, are an efficient means of separating
particulate matter entrained in a gaseous stream. A fabric filter is
typically least efficient collecting particles in the range of 0.1 to
0.3 urn diameter. For particles larger and smaller than this, a fabric
filter can collect more than 99 percent of particles in an emissions
stream. Fabric filters used to control emissions containing HAP's
have two special constraints. First, they should have a closed,
negative-pressure (suction) configuration to prevent accidental
release of the gas stream and captured HAP's; and second, of the three
principal fabric cleaning methods (i.e., mechanical shaking, reverse
air flow, and pulse-jet cleaning), pulse-jet type cleaning is not
recommended for HAP control situations. As explained in Section 4.8,
pulse-jet cleaned filters are not as efficient as those cleaned by
other methods, and emissions are not as constant over a filtration
cycle as those from filters using the other two cleaning methods.
However, pulse-jet cleaning is widely used in general industrial
fabric filter applications and, therefore, Section 4.8 does include
3-38
-------
TABLE 3-7. ADVANTAGES AND DISADVANTAGES OF PARTICULATE CONTROL DEVICES
Type of
Control
Advantages-
Disadvantages
Baghouse - very efficient at removing
fine participate matter from a
gaseous stream; control efficiency
can exceed 99 percent for most
applications.
- lower pressure drop than venturi
scrubber when controlling fine
particulates; i.e., 2 to 6" HgO
compared with >^ 40" HgO. *
- can collect electrically resistive
particles.
- with mechanical shaking or reverse
air cleaning, control efficiency
is generally independent of inlet
loading.
- simple to operate.
ESP
Venturi
can control very small (<0.1 /^m)
particles with high efficiency.
low operating costs with very low
pressure drop (0.5" ^0).
can collect corrosive or tar mists.
power requirements for continuous
operation are low.
wet ESP's can collect gaseous
pollutants.
low initial investment.
takes up relatively little space.
can control sticky, .flammable,
or corrosive matter with few
problems.
can simultaneously collect
particulates and gaseous matter.
control efficiency is independent
of particle resistivity.
simple to operate, few moving
parts.
cannot control high temp
stream (>550°F) without a
precooler.
cannot effectively control
stream with high moisture
content.
highly erosive particles
can damage the filter,
mechanical collectors gener-
ally required upstream if
significant amounts of large
particulates (>20^m) are
present.
needs special or selected
fabrics to control
corrosive streams.
least efficient with particles
between 0.1 to 0.3 jjm
diameter.
high initial capital
investment.
not readily adaptable to
changing conditions.
conditioning agents may be
necessary to control
resistive particles.
more sensitive to particle
loading than other two devices
space requirements may be
greater than that for a
fabric filter or venturi
scrubber.
high operating cost due to
high pressure drop.
(40" H£0 or greater),
particularly for smaller
(<1 ^m) particles.
has wastewater and cleaning/
disposal costs.
le.ast efficient with particles
less than 0.5 pm diameter.
3-39
-------
information on pulse-jet cleaning, further, Chapter 5 includes pulse-jet
cost information to allow for review of permit applications indicating
the use of pulse-jet baghpuses. Fabric, filters usins mechanical shaking
or reverse air cleaning are fundamentally different from ESP's and
venturi scrubbers in that they are not "efficiency" devices. A properly"
designed and operated fabric filter using one of these two cleaning
methods will yield a relatively constant outlet particle concentration,
regardless of inlet load changes. The typical outlet particle concentra-
tion range is between 0.003 to 0.01 grains/scf (gr/scf), averaging
around 0.005 grains/scf. These numbers can be used to ascertain an
expected performance level (see the "Example Case" on page 3-41). This
is not meant to be an absolute, definitive performance level. A vendor
should assist in any attempt to quantify an actual performance level.
Variables important to achieve a given performance (i.e., air-to-cloth
ratio, cleaning mechanism, fabric type) are discussed in detail in
Section 4.8. Fabric filters are sensitive to emission stream temperature
and a precooler or preheater may be required, as discussed previously.
Fabric filters operate at low pressure drops, giving them low operating
costs. They are generally not a feasible choice to control emission
streams with a high moisture content.
3.2.1.2 Electrostatic Precipitators ~
Electrostatic precipitator particle removal occurs by charging
the particles, collecting the particles, and transporting the collected
particles into a hopper. ESP's are less sensitive to particle size
than the other two devices, but very sensitive to those factors that
affect the maximum electrical power (voltage) at which they operate.
These are principally the aerosol density (grains/scf) and the electrical
resistivity of the material. The electrical resistivity of the particles
influences the drift velocity, or the attraction between the particles
and the collecting plate. A high resistivity will cause a low drift
velocity which will decrease the overall collection efficiency. Elec-
trostatic precipitators are discussed further in Section 4.9.
3-40
-------
3.2.1.3 Venturi Scrubbers --
Venturi scrubbers use an aqueous stream to remove particulate
matter from an emissions stream. The performance of a venturi scrubber
is not affected by sticky, flammable, or corrosive particles. Venturi
scrubbers are more sensitive to particle size distribution than either
ESP or fabric filters. In general, venturi scrubbers perform most
efficiently for particles above 0.5 jim in diameter (see Section 4.10
for further detail). Venturi scrubbers have a lower initial cost than
either fabric filters or ESP's, but the high pressure drop required for
high collection efficiencies contributes to high operating costs.
EXAMPLE CASE
Assume a facility is required to achieve an emission
limit for particulate emissions from a municipal waste
incinerator. The emissions stream particles consist
primarily of fly ash; however, 102 of these particles is a
HAP: cadmium. The characteristics of the emission stream
after exiting a heat exchanger are shown in Figure 3-8 on
page 3-43. From Figure B.2-1 (Appendix B), the dew point of
an emission stream containing 200 ppmv $03 and 5 percent
moisture is approximately 327°F. Thus, the emission stream
temperature .(400°F) is within 50 to 100°F above its dew point,
thus minimizing the amount of the HAP in vapor form and
eliminating condensation problems.
Calculate the allowable outlet particle concentration:
110,000 acf x 3.2 gr x Ib x 60 min - 3^017 JJL
min acf 7,000 gr hr ' hr
3,017 I! x (1 - 0-999) = 3.017 1L.
hr hr
Since the HAP constitutes 1.0% of the total particulate matter,
the outlet concentration of the HAP is:
0.10 x 3.017 IA = 0.3017 1b HAP
hr hr
3-41
-------
This value assumes that the HAP is in particulate form and
that it is collected as. efficientT-y as the other particles.
A fabric filter will generally control particles to a
limit of 0.005 gr/scf. Converting gr/scf to Ib/hr yields:
O.Q05 gr x 110.000 acf x 53Q°R-scf x 60 min-lb = 2 9 H
scf min (400 + 460)"R-acf 7000 gr-hr ' hr
To calculate the HAP outlet emission rate:
2.9 Jl x 0.10 = 0.29 1b HAP
hr hr
This value again assumes that the HAP is in particulate form
and that it is collected as efficiently as the other gas
stream particles. The calculated HAP emission rate from a
fabric filter is less than the allowable rate, indicating
that use of a fabric filter is an appropriate control technique
for this emission stream.
In general, an electrostatic precipitator can achieve
control efficiencies of 99.9 percent, provided the particle
resistivity is not "high." The drift velocity of the particles
(0.30 ft/sec) is indicative of particles with an "average"
resistivity; therefore, an ESP can probably be used to
control this stream and, thus, it also is an appropriate
control technique for this emission stream.
A venturi scrubber has difficulty controlling particles
below 0.5 urn diameter. The emission stream presented contains
particles generally above this limit; therefore, a venturi
scrubber may be an acceptable control device for this example
and, thus, all three particulate control devices are appro-
priate control techniques for this emission stream. To deter-
mine the basic design parameters and actual applicability of
each control device, Section 4.8 (Fabric Filter), Section 4.9
(ESP's), and Section 4.10 (Venturi Scrubbers) must be examined.
3-42
-------
«
2.
<
5
o
5
LJ
as
""
g
LjJ
Z
VI
U
01
.c
4W
O
*••»
f»
cf
•
£
4-1
U
*
c
3
4J
^
o.
o
c
a
e
0
4-1
IQ
U
Ol
C
(J
c
IQ
0.
g
^f
CM
U3
IA
1
U1
1/1
in
^*
en
en
en
•
S.
Ol
g
j=
&
01
H-
4->
O)
O)
t.
•w
VI
e
>Q
X
«
CM
•«
*»
01
01
(/I
c
o
'4-1
u
o
•o
o
e
S
•
x
+*
>
U
•*
1
>
Ol
Q»
u
Ol
•o
c
3
VI
§
01
u
4_>
V)
e
o
VI
U)
"e
>*.
o
o
S
•
a.
P
i
4J
VI
3
a
'x
UJ
e
o
4-*
IQ
1
U
c
^
fp>4
c
o
IO
£;
jj
1
g
a-
01
1
z
1
Ol
GO
e
a
V)
VI
i
<
i
,
J3
U.
o
t.
Ol
e
e
•*•
_
OL
u
c
s
to
O)
u
u
1
c
0
VI
°i
a.
co
i
i
49
c
I
in
vi
Ol
fc.
a.
c
o
IQ
U
VI
I/I
U
Source
ej
i
i
.a
§
1
U
(A
|
Ol
u
e
o
VI
VI
LU
=
1
,
JB
01
fO
's
4-1
(Q
a
u
e
IQ
01
O
c
£
I
«
VI
VI
u
a.
UJ
i
i
n
0
«
•ta
ro
«_,
c
Ol
4-1
e
3
u_
1
,
A
CM
01
ressu
Q,
O
ex.
«
1
1
J3
CM
«
5
•"•
3
"o
to
1
I
1
,
J=
CM
*— *
§•
O.
01
a.
u
o
a.
—
|
1
&
CSJ
'**'
Of*
1
3
U
Ol
I
-3
*
^^
«**
4*1
C
01
c
3
c
Ol
o
a.
_j
'o
X
CM
*
Ol
c
3
Ol
u
Molstu
^
-.
c
Ol
c
3
CM
O
1
Cr
u.
|
<*>
CM
Ol
U
3
4>»
>Q
i
£
_i
JC
VI
.i1
*
(^^
U
9
(A
C
U
CM
->
i_.t
c
Ol
c
3
01
10
u
L.
a.
cc
5
L>
VI
§
es
0
m
CM
—•
01
a:
£
S
O
m
•
S
IQ
a
e
Q
01
fc.
M
en
en
gs
I/I
I/I
"a!
01
— j
"a
c
3
1
lw
3
cr
Ol
a:
»
Ol
u
o
VI
^
3
4-)
C
Ol
A
U
0
4-1
Ol *O VI u)
vi a> c vi
4J 10 01 _
oi o u C
S CL£ °
•oa a. ••—
0 C t vi vi 3 cl)
c<-si; s_
01 Z <- O
10 "0 , ,
O 01 w & • *^
w oi 10 u. - •= "° Jii
O 4J Olf*> C V
-~ s: m C
i- c 4-> -a .f.
ex 10 co ,.
s. e 10 oi oi c r>
oi --a i «
C Ol vi o •»"
i- — '•o vi j: c
^ Of Oj Vt ^
E >>•»• u . 2
U IQ 'cl U Ol
VI CL S**~ .a
<- VI 3 J3 l°
0 01 VI 1. •-
O O *J C_
oi oi oi a. vi ~
— C J3 IQ 3 O
•F- •*" •>»> U O
IA 3 •*- 0 ">
— 0) O C O
U •£ « O. «.
| g_ V, Olg ^
Si^^gv, 42
VI "0 'O Ol >-
C II — Ql
C 0 4.) ^J
vi ^- Z •— (H
VI -0 C J3 ?
T- -O Ol O — »
E •*- 4J vi *•
^ o vi vi s O»
<•- oi oi IQ 3
oi .c u oi -j
t. IQ w o i_ r^
IQ t. C U 4J M-
4J oj a. vi i^*
« 5 « <- c LU
4-1 a. o o
c oi a.-*-
Ol VI c *Q VI
vi 3 •!- > VI *
v — 00
k • in u S i
•^ • k -^- OJ
^ S c CO
(Q VI .Q 'Q U
iiQ ^T 3 1- C 5*
, 0 C 0^ W
^ « ^ " o 0»
IT ST<"I £
* * * Uo
* •«
3-43
-------
3.2.2 Control Techniques for Particulate Emissions from Fugitive Sources
Fugitive emission sources may be broken down into two source
categories: process sources and area sources, as defined in Section 2.1
on page 2-4. The methods used to control process sources of fugitive
particulate emissions are generally different from those applied to
area sources. Basically, process fugitive sources can employ conven-
tional measures (i.e., capture techniques and add-on control devices)
while area fugitive sources either cannot 'use conventional measures
or the use of conventional measures is precluded due to cost. For
example, fugitive emissions from unpaved roads cannot use conventional
control measures, but fugitive emissions from area sources such as
pumps and valves can be captured and ducted to a control device, although
the costs may be prohibitive. Area sources are often controlled by
preventive techniques rather than capture/control techniques.
Section 3.1.3 dicusses methods of hooding and capture of process
emissions. The remaining fugitive particulate emission control method-
ologies (i.e., nonconventional techniques) can be applied to multiple
fugitive emission sources — both for process and area sources.
The following subsections discuss the different types of fugitive
particulate emission control techniques that can be applied to general
process and area fugitive particulate emission sources (i.e., sources
common to many industries). Appendix A.8 presents industry-specific and
source-specific information for fugitive particulate emission control.
Appendix A.9 provides information on chemical dust suppressants.
An extensive review of available literature on fugitive emissions
revealed that a specific reference included almost all necessary
information pertinent to the. scope of this manual. Consequently, most
of the following subsections are taken directly from the following
document: "Technical Guidance for Control of Industrial Process Fugitive
Particulate Emissions."1 An April 1985 draft final-document from EPA on
fugitive emissions^ has been distributed and was reviewed. Although no
significant changes were made to this section as a result of that
document, it is referenced and does provide the reader with another
comprehensive view of the subject.
3-44
-------
Throughout the discussions, control efficiencies are stated for
many of the control techniques. It is important to note that the
efficiency values are estimates. The ability to quantify accurately
the emission rates from a fugitive emission source has not yet been
fully realized.
3.2.2.1 Process Fugitive Particulate Emission Control —
Control of HAP process fugitive emissions may be accomplished by
capturing the particulate material and venting it to an add-on control
device (i.e., venturi scrubbers, fabric filters, and ESP's). Venting
emissions is accomplished by exhausting particle-laden air through
fixed or movable ducting under negative pressure. The airflow into the
ducting must be sufficient to maintain particle size capture velocity
and to overcome opposing air currents. The effect of opposing air
currents can be eliminated by complete enclosure, or can be reduced by
minimizing the opening of capture enclosures, or by utilizing curtains
or partitions to block room air currents. If enclosure or installation
of fixed hoods is not feasible due to space limitations or operational
procedures, movable hoods may be a viable alternative. Movable hoods
can be placed over the fugitive emissions source as the production
cycle permits. For example, movable hoods can be placed over the
filling hatch in some types of trucks and rail cars during loading.
Movable hoods have also been applied in the production of coke, whereby
a movable hood follows quench cars during coke pushing. Another alter-
native is to evacuate an enclosed building to a control device.
Once the emissions are captured, the selection of the control
device will be dependent on the emission stream characteristics, HAP
characteristics, and the required performance levels (i.e., removal
efficiency). It is important to remember that the required performance
level for the control device is influenced by the efficiency of the
capture system. The factors that affect the control device selection
process are the same as for point sources, and are discussed in
Section 3.3.1.
3-45
-------
3.2.2.2 Area Fugitive Emission Control From Transfer and Conveying —
Loss of material from conveyors is primarily at the feeding,
transfer, and discharge points and occurs due to spillage or windage.
The majority of particulate emissions are generally from spillage and
mechanical agitation of the material at uncovered transfer points.
However, emissions from inadequately enclosed systems can be quite
extensive. Table 3-8 presents control techniques applicable to these
emission sources.
Control by wet suppression methods includes the application of
water, chemicals, and foam. The point of application is most commonly
at the conveyor feed and discharge points, with some applications at
conveyor transfer points. Wet suppression with water only is a rela-
tively inexpensive technique; however, it has the inherent disadvantage
of being short-lived. Control with chemicals (added to water for
improved wetting) or foam is longer lasting but more expensive than
water alone.
Foam is effective in dust suppression because small particles (in
the range of 1 to 50jim diameter) break the surface of the bubbles in
the foam when they come in contact, thereby wetting the particles.
Particles larger than 50 ^m only move the bubbles away. The small
wetted particles then must be brought together or brought in contact
with larger particles to achieve agglomeration. If foam is injected
into free-falling aggregate at a transfer point, the mechanical
motion provides the required particle to bubble contact and subsequent
particle-to-particle contact.
Highly diluted chemical wetting agents are applied by water jets
ahead of any points in the conveying system where dusting occurs. The
wetting agent breaks down the surface tension of the water, allowing
it to spread further, penetrate deeper, and wet the small particles
better than untreated water. With mechanical agitation of the material,
the small particles agglomerate. For effective control, the spray
should be applied at each point where the particles might be fractured,
allowed to free fall, or subjected to strong air currents.
3-46
-------
TABLE 3-8. CONTROL TECHNOLOGY APPLICATIONS FOR
TRANSFER AND CONVEYING SOURCES3
Emission Points
Control Procedure
Conveyor system (belt,
bucket elevator, etc.)
Transfer and transition
points
Enclosure
1. top covered (marginal control)
2. sides and top covered (good control)
3. completely enclosed (excellent control)
Wet suppression (water, chemical, foam)
at conveyor feed points.
Belt scrapers and wipers
Mechanical belt turnovers
Replacement with pneumatic system or screw
conveyor
Enclosure
Hoods, covers, or canopies with exhaust to
removal equipment (fabric filters, and
wet-col lectors).
Wet suppression (water, chemical, foam).
aSource: Reference 1.
3-47
-------
3.2.2.3 Area Fugitive Emission Control From Loading and Unloading —
Loading and unloading bulk material is common to many processing
industries. Loading and unloading operations can be either for
external transportation of material to or from a facility or for
internal transportation within a facility (for example, internal trans-
portation might consist of loading of a mining haul truck with ore via
a front-end loader for subsequent unloading to a crushing process).
Appendix A.8 should also be consulted for industry-specific informa-
tion on loading and unloading for internal transportation.
Various control technology applications for loading and unloading
operations are presented in Tables 3-9 and 3-10, respectively. These
techniques can be used alone or at times in various combinations.
Generally, the simultaneous use of more than one technique will provide
increased levels of control.
Rail car and truck loading - To minimize particulate emissions
from rail car and truck loading, the entire operation can be enclosed
by the use of doors on the loading shed. This prevents a wind tunnel
effect and allows dust emitted in the enclosure to settle to the ground
within the enclosure. By venting the entire enclosure to a control
device, dust leakage around the doors and any other openings can be
prevented, thus ensuring near 100 percent control.
Exhausting the car or truck body to a dust removal device reduces
emissions if the body is fairly well enclosed. In open type rail or
truck bodies this technique is not too effective.
Choke-feed eliminates free fall of material into the car or
truck. In this technique the mouth of the feed tube is immersed in the
material being unloaded. This technique only works for fairly free-
flowing dry material. A telescopic chute or spout also essentially
eliminates the free-fall distance of the material being loaded. This
type of system can be used on all types of material. Both the choke-
feed and telescopic chute methods are only partially effective in
eliminating emissions since the surface of the loaded material is
constantly disturbed by new material. This surface is subject to wind
and dust entrainment.
Movable hoods exhausted to a dust removal system can be placed
over the filling hatch in some types of trucks and railcars during
3-48
-------
TABLE 3-9. CONTROL TECHNOLOGY APPLICATIONS FOR
LOADING OPERATIONS3
Emission Points
Control Procedure
Railcar, truck
Barge and ship
Drive through enclosure with doors at
both ends
Exhaust of entire enclosure to dust
removal equipment
Movable hood over hatch opening
Exhaust of car hopper to dust removal
equipment
Choke-feed or telescopic chute to confine
and limit free-fall distance (gravity
loading)
Wet suppression (water, chemicals)
Use of tarpaulins or covers over the
holds
Canopy and exhaust system over the loading
boom, with attached tarps around the
hatch
Exhaust of ship hold to dust removal
equipment
Choke-feed or telescopic chute to confine
and limit free-fall distance
For tanker types, use of gravity filler
spouts with concentric outer exhaust
duct to control equipment
Wet suppression (water, chemicals)
aSource: Reference 1.
3-49
-------
TABLE 3-10. CONTROL TECHNOLOGY APPLICATIONS FOR
UNLOADING OPERATIONS3
Emission Points
Control Procedure
Railcar, truck
Barge and ship
Drive-through enclosure with doors at both
ends.
Exhaust of enclosure to dust removal
equipment.
Exhaust air from below grating of receiving
hopper to removal equipment.
Choke feeding to receiving pit (hopper car
and hopper truck).
Unloading with screw conveyor (box car).
Wet suppression (water, chemicals).
Utilize a pneumatic unloading system.
Enclosure of top of clamshell bucket with
transparent material and maintenance
of closure seals and teeth on bottom
of bucket.
Enclosure of shoreside receiving hopper.
Exhaust of enclosed shoreside receiving
hopper to dust removal equipment.
aSource: Reference 1.
3-50
-------
loading. By keeping other openings on the body closed, any dust
generated in loading must be emitted through the single open hatch. A
hood with sufficient airflow mounted around this opening could capture
most of the dust generated.
Wet suppression techniques, when applied to loading operations,
can reduce airborne dust to some extent. The loading process naturally
breaks up surface coatings, but some small dust particles will adhere
to larger pieces so as not to become entrained. Many materials cannot
be readily wetted and this technique could not be used for these
materials.
Barge and ship loading - Due to their larger size, barge and ship
loading present unique problems for dust control. However, a number of
control techniques have been developed and utilized, especially at some
of the larger shipping terminals.
The use of tarpaulins or similar covers over hatches on ships and
enclosed barges reduces airborne emissions by preventing their escape.
Air, displaced by the material being loaded, causes the hold to become
slightly pressurized during loading, and the hold must be vented at
some point if the hatches are air-tight. Thus, a more effective
control system incorporates an exhaust system for the hold. This
exhaust system is connected to a dust control system such as fabric
filter with the collected material being returned to the hold. Such a
system can practically eliminate loading emissions if carefully main-
tained and properly operated. The use of a canopy hood and exhaust
system over the loading boom is less effective than a totally enclosed
system, but can still reduce emissions and is a viable alternative
for open barges. Effective utilization of this technique requires some
type of wind break to increase the hood capture efficiency.
Choke feed and telescopic chutes or spouts as previously described
can also be used for loading both enclosed and open ships or barges.
Wet suppression techniques may also help reduce airborne emissions if
the product specifications do not prohibit use of this technique.
Rail car and truck unloading - Many of the unloading dust control
techniques are identical to the loading techniques. When a rail car or
truck is tilted and materials are dumped into an underground chamber
through a grating, exhausting air from this chamber through a control
3-51
-------
device will effectively reduce emissions. By causing air to flow down
through the grating, dust emissions are contained. The face velocity
of air through the grating is a critical design parameter in this
technique. Unloading cars with a screw conveyor causes less distribu-
tion of the material and thereby less dust. Problems of material
handling and time requirements limit the application of this technique.
Pneumatic unloading of very fine materials is an effective and widely
used technique that practically eliminates dust emissions. With this
system, careful maintenance of hose fittings and the fabric filter
through which the conveying air exhausts is required.
Barge and ship unloading - Control of barge and ship unloading
requires enclosure of the receiving point on the shore and possibly
exhausting of that enclosure to a control device. A good enclosure with
an exhaust system can provide essentially 100 percent capture. For
open ships and barges which use buckets and conveyors, a partially
enclosed bucket will reduce windblown dust. When observation of the
bucket by the operator is required, a transparent heavy plastic sheet
can be used as a cover. This system is only partially effective and
must usually be supplemented with other controls, such as tighter
fitting covers, wind breaks, or possibly wet suppression.
3.2.2.4 Area Fugitive Emission Control From Paved and Unpaved Roads —
Dust on the surface of paved roads is deposited by such processes
as mud track-out on vehicle tires, atmospheric fallout, spillage or
leakage from trucks, pavement wear and decomposition, runoff or wind
erosion from adjacent land areas, deposition of biological debris, wear
from tires and brake linings, and wear of anti-skid compounds. This
material is reentrained by contact with tires and by the air turbulence
created by passing vehicles.
On unpaved roads, the road base itself serves as the main source
of dust. As with paved roads, the dust becomes airborne by contact
with vehicles' tires and by air turbulence from passing vehicles.
Also, some of the fugitive dust from unpaved roads is attributed to
wind erosion. On both paved and unpaved roads, traffic movement causes
the continuing mechanical breakdown of large particles on the road
surface, thus providing new material in the suspended particulate size
3-52
-------
range. Available procedures for reducing emissions from plant roads
and their estimated efficiencies are presented in Table 3-11.
Paved streets and roads in a plant area can be cleaned on a
frequent schedule to reduce the amount of particulate material on the
surface that is available for reentrainment. Flushers and vacuum-
type motorized street cleaners are both quite effective in removing
surface material and thereby reducing emission rates from vehicles
using the cleaned streets. Because raw material accumulates rapidly on
the streets, the overall effectiveness of a street-cleaning measure
is a function of the frequency of cleaning and the removal efficiency
of the equipment.
For plants with small amounts of paved roads, industrial vacuum
sweepers or contracted sweeping programs (such as many shopping centers
use) would be more appropriate than the larger vacuum street cleaning
equipment used on public streets. Mechanical broom sweepers have been
shown to be ineffective from an air pollution control standpoint in
that they redistribute material into the active traffic lanes of the
streets and they remove almost none of the fine material (less than
43 urn) that is subject to reentrainment.
Many street sweepers depend upon the material being concentrated
in the gutter in order to achieve good collection efficiency, and
therefore cannot be used on streets without curbs and gutters. However,
the smaller industrial sweepers are usually designed for use in ware-
house and storage areas that are not curbed. A factor which might
limit the applicability of street flushers in plants is that unpaved
areas adjacent to the streets would be wet by the water spray and then
become subject to mud track-out onto the streets by equipment and
vehicles driving through these areas.
Good housekeeping practices include the rapid removal of spills on
roadways and at conveyor transfer points. Preventive measures
include covering of truck beds to prevent windage losses, cleaning of
truck tires and undercarriages to reduce mud track-out onto paved
roads, and minimizing the pick-up of mud by trucks.
The paving of unpaved roadways is the most permanent of the
various types of controls. However, the degree of effectiveness of
3-53
-------
TABLE 3-11. CONTROL TECHNOLOGY APPLICATIONS FOR PLANT ROADS3
Emission Points
Control Procedure
Efficiency
Paved streets
Street cleaning
Housecleaning programs to
reduce deposition of material
on streets
Vacuum street sweeping (daily)'5
Speed reduction
No estimate
No estimate
25 %b
Variable
Unpaved roads
Paving 85%
Chemical stabilization 50%
Watering 50%
Speed reduction Variable
Oiling and double chip surface 85%
Road shoulders
Stabilization
80%
aSource: Reference 1.
bSource: Reference 2.
3-54
-------
this technique is highly dependent on prevention of excessive surface
dust loading.
Watering of unpaved roads is effective only when carried out on a
regular basis. The schedule depends on climate, type of surface
material, vehicle use, and type of vehicles.
Oiling unpaved roads is more effective than watering and needs to
be applied less often. However, special precautions must always be
taken so as not to add to surface water runoff problems.
3.2.2.5 Area Fugitive Emission Control From Storage Piles —
Most dust arises from stockpile areas as the material is dumped
from the conveyor or chute onto the pile, and as bulldozers move the
pile. During periods with high wind speeds [greater than about
6 m/sec (13 mph)] or low moisture, wind erosion of a non-weathered
surface may also cause emissions. Applicable control techniques for
open storage piles are presented in Table 3-12.
Enclosing materials in storage is generally the most effective
means of reducing emissions from this source category because it allows
the emissions to be captured. However, storage bins or silos may be
very expensive. Storage buildings must be designed to withstand wind
and snow loads and to meet requirements for interior working conditions.
One alternative to enclosure of all material is to screen the material
prior to storage, sending the oversize material to open storage and the
fines to silos.
Wind screens, or partial enclosure of storage piles, can reduce
wind erosion losses but do not permit capture of the remaining storage
pile fugitive emissions. Earthen berms, vegetation, or existing
structures can serve as wind screens.
Telescoping chutes, flexible chute extensions, and traveling booms
are used to minimize the free fall of material onto the pile and
resulting emissions. Similarly, emissions due to loadout can be reduced
by reclaiming the material from the bottom of the pile with a mechanical
plow or hopper system. The use of telescoping chutes and flexible
chute extensions for piles with high material flow rates may require
closer control of operations because of the possibility of jamming.
3-55
-------
TABLE 3-12. CONTROL TECHNOLOGY APPLICATIONS FOR OPEN STORAGE PILES3
Emission Points
Control Procedure
Efficiency
Loading onto piles
Enclosure 70-99%
Chemical wetting agents or foam 80-90%
Adjustable chutes 75%
Movement of pile
Enclosure
Chemical wetting agents
Watering
Traveling booms to distribute
material
95-99%
90%
50%
No estimate
Wind erosion
Enclosure
Wind screens
Chemical wetting agents or foam
Screening of material prior to
storage, with fines sent
directly to processing or to
a storage silo
95-99%
very low
90%
No estimate
Loadout
Water spraying
Gravity feed onto conveyor
Stacker/reel aimer
50%
80%
25-50%
aSource: Reference 1.
3-56
-------
Traveling or adjustable booms can handle high flow rates, but have
greater operating costs.
Wetting agents or foam-that are sprayed onto the material during
processing or at transfer points retain their effectiveness in subse-
quent storage operations. Wetting agents retain surface moisture for
extended periods, thereby preventing dusting. Spraying of the material
prior to storage may not be possible in cases where product contamination
could result (e.g., Portland cement clinker) or where the material is
water soluble. However, such materials are generally not placed in
open storage anyway. Steam has also been found to be an effective dust
suppressant for some short-term storage operations.
3.2.2.6 Area Fugitive Emission Control From Waste Disposal Sites —
Fugitive dust can occur anywhere dusty waste material is dumped
for disposal. This includes overburden piles, mining spoils, tailings,
fly ash, bottom ash, catch from air pollution control equipment, process
overload discharges, building demolition wastes, contaminated product,
etc. Like open storage, emissions come from dumping and from wind
erosion across unprotected surfaces. Since waste piles are generally
not disturbed after dumping, there are no emissions from an activity
comparable to loading out of the storage pile. However, there may be
emissions from transporting the waste material on-site (if it is dry
when it is produced) or from a reclamation process such as landfill
covering associated with the waste disposal operation. If the surface
of the waste material does not include a compound that provides
cementation upon weathering, or if the surface is not compacted, or if
an area of very little rainfall, wind erosion of fines can occur with
winds greater than about 21 km per hour (13 mph). Table 3-13 presents
control techniques for waste disposal sites.
3-57
-------
TABLE 3-13. CONTROL TECHNOLOGY APPLICATIONS FOR WASTE DISPOSAL SITES3
Emission Points
Control Procedure
Efficiency
Handling
Keep material wet 100%
Covered or enclosed hauling No estimate
Minimize free fall of the material No estimate
Dumping
Spray bar at dump area
Minimal free fall of material
Semi-enclosed bin
50%
No estimate
No estimate
Wind Erosion
Covering with dirt or stable
materi al
Chemical stabilization
Revegetation
Rapid reclamation of newly
filled areas
100%
80%
25% - 100%
No estimate
Grading
Watering
50%
aSource: Reference 1.
3-58
-------
3.2.3 References for Section 3.2
1. U.S. EPA. Technical Guidance for Control of Industrial Process
Fugitive Particulate Emissioins"! EPA-450/3-77-010. March 1977.
2. U.S. EPA. Identification, Assessment, and Control of Fugitive
Particulate Emissions.Draft final report.EPA Contract No.
68-02-3922.April 10, 1985. Contact Mr. Dale Harmon of EPA at
(919) 541-2429.
3-59
-------
CHAPTER 4
HAP CONTROL TECHNIQUES
This section describes and illustrates the procedures used to calculate
the basic design and operating variables of HAP control techniques in terms of
commonly employed design principles and values. For each technique, the
manual provides: (a) a brief description of how the technique works, (b)
definitions of input data required, and (c) a step-by-step calculation
procedure showing where each number used in the procedure comes from and how
it is to be used. The procedures described in this manual will result in
conservatively designed control systems. In instances in which less
conservatively designed control systems may achieve the target control level,
more detailed calculation procedures requiring compound-specific data would be
needed. This level of specificity is beyond the scope of this manual.
The data for HAP the emission stream to be controlled are taken from the
"HAP Emission Stream Data From" given in Chapter 2. In case of a permit
evaluation, however, these data should be supplied by the applicant. The
reviewer may wish to confirm the completeness of the applicant's data by
referring to Chapters 2 and 3.
The step-by-step calculation procedures are illustrated for each control
technique using data based on Emission Streams 1 through 7 described in
Chapter 3. In permit reviews, the calculated values are compared to the
values in the permit application to determine the adequacy of the applicant's
proposed design.
Further details on the calculations and supplementary information are
given in Appendices 8.3 through B.ll. Appendices C.2 through C.ll contain
blank calculation sheets to use in applying the calculations described for
each control technique. Finally, if control systems costs are required, see
Chapter 5.
4-1
-------
4.1 THERMAL INCINERATION
Thermal incineration (Figure 4.1-1) is a widely used air pollution
control technique whereby organic vapors are oxidized at high temperatures.
The most important variables to consider in thermal incinerator design are the
combustion temperature and residence time because these design variables
determine the incinerator's destruction efficiency. Further, at a given
combustion temperature and residence time, destruction efficiency is also
affected by the degree of turbulence, or mixing of the emission stream and hot
combustion gases, in the incinerator. In addition, halogenated organics are
more difficult to oxidize than unsubstituted organics; hence, the presence of
halogenated compounds in the emission stream requires higher temperature and
longer residence times for complete oxidation. Thermal incinerators can
achieve a wide range of destruction efficiencies. This discussion is focused .
on efficiencies of 98 to 99+ percent.
The incinerator flue gases are discharged at high temperatures and
contain valuable heat energy. Therefore, there is a strong economic incentive
for heat recovery. Typical recovery methods include heat exchange between the
flue gases and the emission stream and/or combustion air and use of the
available heat for process heat requirements (e.g., recycling flue gases to
the process, producing hot water or steam, etc.). In most thermal incinerator
applications, the available enthalpy in the flue gases is used for preheating
the emission stream. This discussion will be based on a thermal incineration
system where the emission stream is preheated.
The incineration of emission streams containing organic vapors with
halogen or sulfur components may create additional control requirements. For
example, if sulfur and/or chlorine are present in the emission stream, the
resulting flue gas will contain sulfur dioxide (SO-) and/or hydrogen chloride
(HC1). Depending upon the concentrations of these compounds in the flue gas
and the applicable regulations, scrubbing may be required to reduce the
concentrations of these compounds. The selection and design of scrubbing
systems are discussed in Section 4.6.
4.1-1
-------
(J
•M
U1
o>
*
s_
—
(U iQ
u Q.
xo
LU- — •
O
£
S-
fd
•r»
-o
o
4->
Q.
(/I
CD
3
CT
I?
03
-4-J *
in S-
3 -t-
J3 *£.
O r- I
f S l~t
Q.
01
(U
4.1-2
-------
In this subsection, the calculation procedure will be illustrated using
Emission Stream 1 described in Chapter 3,
4.1.1 Data Required
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on the "HAP Emission Stream Data
Form" and the required HAP control as determined by the applicable regulations.
EXAMPLE CASE
1. Maximum flow rate, Q » 15,000 scfm
2. Temperature, T = 120°F
3. Heat content, h * 0.4 Btu/scf
4. Oxygen content, 02 - 20.6%
5. Moisture content, M - 2%
6. Halogenated organics: Yes No X
Based on the control requirements for the emission stream:
Required destruction efficiency, DE - 99%
If dilution air is added to the emission stream upon exit from process,
the data that will be used in the calculations are the resulting
characteristics after dilution.
In the case of permit review for a thermal incinerator, the data outlined
below should be supplied by the applicant; the calculations in this section
will then be used to check the applicant's values.
Thermal incinerator system variables at standard conditions (70°F,
1 atm):
1
Reported destruction efficiency, DEreDOrted
2. Temperature of the emission stream entering the incinerator,
T » _ °F (if no he
6 — ~"~ —
exchanger is employed).
T » _ °F (if no heat recovery); T,a = _ °F (if a heat
6 — ~"~ — ric """
Combustion temperature, T » _ °F.
4.1-3
-------
4.
5.
6.
7.
8.
9.
10.
11.
Residence time, t « sec
I "~ "'~
Maximum emission stream flow rate, Q - scfm
Excess air, ex * %
Fuel heating value (assume natural gas), he »
Supplementary heat requirement, Hr = Btu/min
ft3
Btu/scf
Combustion chamber volume, V -
Flue gas flow rate, Q
fg
scfm
Heat exchanger surface area (if a heat exchanger is employed),
A- ft2
4.1.2 Pretreatment of the Emission Stream: Dilution Air Requirements
In HAP emission streams containing oxygen/air and flammable vapors, the
concentration of flammable vapors is generally limited to less than 25 percent
of the lower explosive limit (LEL) to satisfy safety requirements imposed by
insurance companies. [Note: The LEL for a flammable vapor is defined as the
minimum concentration in air or oxygen at and above which the vapor burns upon
contact with an ignition source and the flame spreads through the flammable
gas mixture.] In some cases, flammable vapor concentrations up to 40-50
percent of the LEL are permitted if on-line monitoring of VOC concentrations
and automatic process control and shutdown are provided.
In general, emission streams treated by thermal incineration are dilute
mixture of VOC and air, and typically do not require dilution. For emission
streams with oxygen concentrations greater than 16 percent and heat contents
greater than 13 Btu-scf (corresponding to flammable vapor concentrations of
approximately 25 percent of LEL), the calculation procedure in this manual
assumes that dilution air is required (see Appendix B.3 for calculation of
dilution air requirements).
EXAMPLE CASE
Since 02 = 20.6% and he » 0.4 Btu/scf, no dilution
air is required.
4.1-4
-------
4.1.3 Thermal Incinerator System Design Variables
Table 4.1-1 presents suggested combustion temperature and residence time
values for thermal incinerators to achieve a given destruction efficiency.
Two sets of values are shown in the table, one set for nonhalogenated emission
streams and another set for halogenated emission streams. The combustion
temperature and residence time values listed are conservative and assume
adequate mixing of gases in the incinerator. The criteria in this table are
not the only conditions for achieving the specified destruction efficiencies.
For a given destruction efficiency, it may be possible to incinerate HAP
emission streams at lower temperatures with longer residence times.
Based on the required destruction efficiency (DE), select appropriate
values for TC and tr from Table 4.1-1.
EXAMPLE CASE
The required destruction efficiency is 99% and the
HAP emission stream is nonhalogenated, therefore:
Tc - 1800°F (Table 4.1-1)
tr - 0.75 sec. (Table 4.1-1)
In a permit evaluation, if the reported values for T and t are
sufficient to achieve the required DE (compare the applicant's values with the
values from Table 4.1-1), proceed with the calculations. If the reported
values for TC and tr are not sufficient, the applicant's design is
unacceptable. The reviewer may then wish to use the values for T and T from
Table 4.1-1. [Note: If DE is less than 98 percent, obtain information from
literature and incinerator vendors to determine appropriate values for T and
tr.]
4.1-5
-------
TABLE 4.1-1. THERMAL INCINERATOR SYSTEM DESIGN VARIABLES'
Required
Destruction
Efficiency
DE (X)
98
99
Nonhalogenated Stream
Combustion Residence
Temperature Time
Tc (°F) tr (sec)
1,600 0.75
1,800 0.75
Halogenated Stream
Combustion Residence
Temperature Time
Tc (°F) tr (sec)
2,000 1.0
2,200 1.0
Source: References 1 and 2.
4.1-6
-------
4.1.4 Determination of Incinerator Operating Variables
4.1.4.1 Supplementary Heat Requirements--
Supplementary fuel is added to the thermal'incinerator to attain the
desired combustion temperature (T ). For a given combustion temperature, the
quantity of heat needed to maintain the combustion temperature in the thermal
incinerator is provided by: (a) the heat supplied from the combustion of
supplementary fuel, (b) the heat generated from the combustion of hydrocarbons
in the emission stream, (c) the sensible heat contained in the emission stream
as it leaves the emission source, and (d) the sensible heat gained by the
emission stream through heat exhange with hot flue gases. Typically, items
(b) and (c) are very small and can often be neglected.
In general, emission streams treated by thermal incineration are dilute
mixtures of VOC and air, and typically do not require additional combustion
air. For preliminary calculations, it can be assumed that no additional
combustion air is required if the oxygen concentration of the emission stream
exceeds 16 percent. Depending on the heat content of the emission stream and
the desired combustion temperature, combustion air requirements may be zero
even when the oxygen concentration is below 16 percent. Hence, this cut-off
value will lead to a conservative design. Use the following simplified
equation for dilute streams to calculate supplementary heat requirements
(based on natural gas):
Hf - 1.1 hf e
Q(l + 0.002
hr - 1.4 Cp . (T - T )
"f 1>n ^air K c V
where:
Hr - supplementary heat requirement, Btu/min
hr • heating value of natural gas, Btu/scf
Q « maximum emission stream flow rate, scfm
M = moisture content of the emission stream, %
Cp . « average specific heat of air over a given temperature
interval (Tr to T), Btu/scf-°F
4.1-7
-------
T » combustion temperature, °F
T = reference temperature, = 70°F
TL = emission stream temperature after heat recovery, °F
hg » heat content of the emission stream, Btu/scf
See Appendix B.4 for derivation of this equation. The factor 1.1 in the
equation accounts for an estimated heat loss of 10 percent in the incinerator.
Supplementary heat requirements are based on maximum emission stream flow
rate, and hence will lead to a conservative design. In this manual, it is
assumed that the minimum supplementary heat requirement is 5 Btu/min per scfm
of emission stream.
A graph of Equation 4.1-1 is shown in Figure 4.1-2, where the ratio H^/Qg
is plotted against the emission stream heat content (h ) for four different
6
combustion temperatures (T ). Instead of evaluating Equation 4.1-1,
Figure 4.1-2 can be used directly to determine supplementary heat
requirements. This graph is based on the following assumptions:
(1) temperature of the emission stream (Tg) is 100°F, (2) moisture content of
the emission stream (M ) is 2 percent, (3) preheat temperature of the emission
stream (T^ ) is based on 50 percent heat recovery in the heat exchanger, and
(4) hf - 882 Btu/scf (based on the lower heating value for natural gas--see
Appendix B.4 for its composition).
4.1-8
-------
o o o o
o o o o
O O Q O
CM O CO
c
Q)
•<->
c
o
o
a)
-M
e-a
Ol
S_ 3
nj .—
c -o
o.
o.
3
(U
s-
(uips)
4.1-9
-------
EXAMPLE CASE
Using Equation 4.1-1:
Since the emission stream is very dilute and
has an oxygen content greater than 16%, Equation 4.1-1
is applicable. The values to be inserted in the
equation are:
Qe = 15,000 scfm
Me = 2%
hg = 0.4 Btu/scf
TC = 1,800°F (Table 4.1-1)
Tr - 70°F
The = 96°0fr (APPendix B.5--based on a heat recovery of 50%)
Cp,.„ = 0.0196 Btu/scf-°F for the interval 70-1,800°F
air
(Table B.4-1)
Cpair a °-0187 Btu/scf-°F for the interval 70-960°F
(Table B.4-1)
hf = 882 Btu/scf
H
H
[Note: Hf is greater than the minimum supplementary heat
requirement assumed in this manual.]
Using Figure 4.1-2:
For he - 0.4 Btu/scf and TC = 1800°F, Hf/Qe
as indicated in the figure is about 20 Btu/min/scfm.
Multiplying 20 by Qe, (20 x 15,000), yields an
approximate value of 300,000 Btu/min for Hf.
4.1-10
f *
f '-
1.1 x 882
"[15,000(33.91
882
• 295,000 Btu/min
- 16.64
- 47.5
- 0.4)]"
-
-------
For emission streams that do not contain sufficient quantities of oxygen
to satisfy the combustion air requirements (e.g., process emissions), refer to
Figure 4.1-3 which shows a plot of Hr/Qe versus h to obtain a conservative
estimate for H^. Figure 4.1-3 is based on the same assumptions as those
stated for Figure 4.1-2. In addition, the oxygen content of the emission
stream (02) is assumed as zero; this corresponds to maximum combustion air
requirements for the thermal incinerator system. If the oxygen content of a
particular emission stream falls between zero and 16 percent, use Figure 4.1-3
to obtain a conservative estimate of Hf/Q .
4.1.4.2 Flue Gas Flow Rate--
Flue gas is generated as a result of the combustion of supplementary fuel
and the emission stream. Use the following equation to calculate flue gas
flow rate:
Qfg-Qe + Qf + QC (4.1-2)
where:
Qft. * ^ue 9as ^ow rate> sc^m
Qf • natural gas flow rate, scfm
Q » combustion air requirement, scfm
Calculate Q from the following equation:
Hf/hf
(4.1-3)
As indicated earlier, Q will typically be zero for dilute emission streams
with oxygen contents (02) greater than 16 percent.
percent, refer to Appendix B.4 for calculating Q .
If CL is less than 16
4.1-11
-------
4-
U
o>
01
•u
to
o
-------
EXAMPLE CASE
Using Equation 4.1-3:
295,000 Btu/min
hf - 882 Btu/scf
Qf - 295,000/882
Using Equation 4.1-2;
330 scfm
Me
"c
Qfg
qfg
15,000 scfm
0 (since 0- is greater than 16%)
15,000 +330+0
15,330 scfm
4.1.5 Combustion Chamber Volume
The flue gas flow rate (Qfa) determined by Equation 4.1-2 is expressed at
standard conditions. In order to calculate the combustion chamber size, Qf
has to be expressed at actual conditions, i.e., temperature effects must be
considered (assume pressure effects are negligible). Use the following
equation to convert
^
from "scfm" to "acfm":
where:
Q
fg,a
Qfg>a - Qfg [(Tc + 460)/530]
flue gas flow rate at actual conditions, acfm.
(4.1-4)
The volume of the combustion chamber (V ) is determined from the
residence time (t ) and flue gas flow rate at actual conditions (Q- ,)
r . i y, a
according to the following equation:
(4.1-5)
4.1-13
-------
The factor "1.05" is used in Equation 4.1-5 to increase the chamber volume by
5 percent. This technique is an accepted industry practice and allows for
fluctuations in the operating conditions (e.g., flowrate, temperature, etc.).
The smallest commercially available incinerator has a combustion chamber
If the calculated V^ is less than 36 ft3,
volume of about 36 ft (1 m ).
define Vc as 36 ft'
EXAMPLE CASE
Using Equation 4.1-4:
Tc - 1800°F
Qf - 15,330 scfm
fg,a
15,330 [(1800 + 460)/530]
65,370 acfm
Using Equation 4.1-5:
tr - 0.75 sec (from Table 4.1-1)
Vc = [(65,370/60) 0.75] 1.05
Vc - 860 ft3
4.1.6 Heat Exchanger Size
The size of the heat exchanger required for preheating the emission
stream to T^e before it enters the thermal incinerator is based on the heat
exchanger design. Use the following expression to calculate the required
size, i.e., surface area, of the heat exchanger:
A
where:
A
[60
0.002 Me)
(T
he
(4.1-6)
heat exchanger surface area, ft
emission stream temperature, °F
overall heat transfer coefficient, Btu/hr-ft2-°F
4.1-14
-------
AT,M - logarithmic mean temperature difference, F
Refer to Appendix B.5 for the details of the heat exchanger design, and
for the values of the variables in Equation 4.1-6. The equation has been
evaluated for dilute emission streams that require no additional combustion
air as shown in Figure 4.1-4. In the figure, heat exchanger surface area (A)
is plotted against the emission stream flow rate (Q ). The assumptions
inherent in this figure are the same as those described for Figure 4.1-2. The
overall heat transfer coefficient is assumed as 4 Btu/hr-ft -°F.
EXAMPLE CASE
Using Equation 4.1-6:
Qe = 15,000 scfm
M. = 2%
CPair
U
AT
Tc
AT
LM
LM
960 F (Appendix B.5--based on heat
recovery of 50%)
120°F (input data)
* 0.0187 Btu/scf-°F for the interval
120-960°F (Table B.4-1)
4 Btu/hr-ft2-°F
- Tc - The (see Appendix B.5)
» 1800°F
- 1800 - 960 =• 840°F
Substituting in Equation 4.1-6:
A - F60 x 15.000 x 1.004 x 0.0187 (960 - 120)1
(4 x 840)
A * 4,200 ft2
Using Figure 4.1-4:
For Q = 15,000, the value of A from the
2
figure is about 4,000 ft .
4.1-15
-------
15,000
12,500- •
OJ
10,000-
O!
t 7,500^
3
oo
Ol
CT)
(J
X
UJ
na
HI
5,000,
2,500,
Emission Stream Flow Rate, Q (scfm)
Figure 4.1-4. Heat exchanger size vs. emission stream flow rate.
(dilute stream/no combustion air)
50,000
4.1-16
-------
For emission streams that are not dilute and require additional combus-
tion air, use Figure 4.1-5 to obtain an estimate of the heat exchanger surface
area. In Figure 4.1-5, the ratio A/Q is plotted against the emission stream
content (h ) for four different combustion temperatures (T ). The assumptions
inherent in this figure are the same as those stated for Figure 4.1-4; in
addition, maximum combustion air requirements are assumed (i.e., 0- =0). If
the conditions represented in Figures 4.1-4 and 4.1-5 are not directly
applicable for a particular emission stream, use Figure 4.1-4 to obtain a
conservative estimate.
4.1.7 Evaluation of Permit Application
Using Table 4.1-2, compare the results from the calculations and the
values supplied by the permit applicant. The calculated values in the table
are based on the example. The value of the combustion chamber volume (Vc) is
determined indirectly from the flue gas flow rate (Qra)> and Qr_ is determined
from the emission stream flow rate (Qe)» combustion air requirement (Qc), and
supplementary fuel requirement (Qr). Therefore, if there are differences
between the calculated and reported values for Qr_ and V , these will be
dependent on the differences between the calculated and reported values for Q
and Qr. If the calculated values for Q , Hr, Qr, and A differ from the
reported values for these variables, the differences may be due to the
assumptions involved in the calculations. Therefore, further discussions with
the permit applicant will be necessary to find out about the details of the
design and operation of the proposed thermal incinerator system.
If the calculated values and the reported values are not different, then
the design and operation of the proposed thermal incinerator system may be
considered appropriate based on the assumptions used in this manual.
4.1-17
-------
+J 01
fO 3
CU -Q
-C E
O
£ U
Ifl
£1
X
c fa
O £
0)
C
> o
0)
N
cu
CU C
-C CU
U CT»
X >s
Ol X
rtJ O
-------
TABLE 4.1-2. COMPARISON OF CALCULATED VALUES AND VALUES
SUPPLIED BY THE PERMIT APPLICANT FOR
THERMAL INCINERATION
Calculated
Value
(Example Case)'
Reported Value
Supplementary heat requirement, Hr 295,000
Supplementary fuel flow rate, Qf 330 scfm
Flue gas flow rate, Q.
Combustion chamber size, V
Heat exchanger surface area, A
15,330 scfm
860 ft3
4,200 ft2
Based on Emission Stream 1.
4.1-19
-------
4.1.8 References for Section 4.1
1. U.S. EPA. Organic Chemical Manufacturing. Volume 4: Combustion Control
Devices. EPA-450/3-80-026. December 1980.
2. U.S. EPA. Reactor Processes in Synthetic Organic Chemical Manufacturing
Industry - Background Information for Proposed Standards. Draft EIS.
October 1984.
4.1-20
-------
4.2 CATALYTIC INCINERATION
Catalytic incineration (Figure 4.2-1) is an air pollution control
technique whereby VOC's in an emission stream are oxidized with the help of a
catalyst. A catalyst is a substance that accelerates the rate of a reaction
at a given temperature, without being appreciably changed during the reaction.
Catalysts typically used for VOC incineration include platinum and palladium;
other formulations are also used, including metal oxides for emission streams
containing chlorinated compounds. The catalyst bed (or matrix) in the
incinerator is generally a metal mesh-mat, ceramic honeycomb, or other ceramic
matrix structure designed to maximize catalyst surface area. The catalysts
may also be in the form of spheres or pellets. Before passing through the
catalyst bed, the emission stream is preheated, if necessary, in a natural
gas-fired preheater.
The performance of a catalytic incinerator is affected by several factors
including: (a) operating temperature, (b) space velocity (reciprocal of
residence time), (c) VOC composition and concentration, (d) catalyst
properties, and (e) presence of poisons/inhibitors in the emission stream. In
catalytic incinerator design, the important variables are the operating
temperature at the catalyst bed inlet and the space velocity. The operating
temperature for a particular destruction efficiency is dependent on the
concentration and composition of the VOC in the emission stream and the type
of catalyst used. Space velocity is defined as the volumetric flow rate of
the combined gas stream (i.e., emission stream + supplemental fuel +
combustion air) entering the catalyst bed divided by the volume of the
catalyst bed. As such, space velocity also depends on the type of catalyst
used. At a given space velocity, increasing the operating temperature at the
inlet of the catalyst bed increases the destruction efficiency. At a given
operating temperature, as space velocity is decreased (i.e., as residence time
in the catalyst bed increases), destruction efficiency increases. Catalytic
incinerators can achieve overall VOC destruction efficiencies up to about 98
percent and HAP destruction efficiencies up to about 95 percent with space
-1134
velocities in the range 30,000 to 100,000 hr . ' ' However, the greater
catalyst volumes and/or higher temperatures required for higher destruction
4.2-1
-------
*
i.
c
o
o
=3
"^"
a
O)
03 (O
-C C
o o
X -i-
m 4->
a.
-M O
S.
O
+->
us
S-
(U
O
c
u
•^
+J
a
s_
O)
•(->
03
O)
0)
Q.
*
i.
c
o
O
O
X3
a>
ca
fO
e
01
+J
to
>>
t/l
s_
O)
c
•r»
o
c
o
+J
>>
*J
u
IT>
M-
O
£
(O
rO
•r»
•o
o
rO
e
OJ
0)
a
en
ai
3
a>
r—
Q.
ui
O
4->
ro
4->
(/I
U
(J
at
a.
T3
(U
a-
cu
4.2-2
to
-------
efficiencies (i.e., 99 percent) may make catalytic incineration uneconomical.
This discussion will be based on HAP destruction efficiencies of 90 and 95
percent.
The performance of catalytic incinerators is sensitive to pollutant
characteristics and process conditions. In the following discussion, it is
assumed that the emission stream is free from poisons/inhibitors such as
phosphorus, lead, bismuth, arsenic, antimony, mercury, iron oxide, tin, zinc,
sulfur, and halogens. [Note: Some catalysts can handle emission streams
containing halogenated compounds.] It is also assumed that the fluctuations
in process conditions (e.g., changes in VOC content) are kept to a minimum.
The energy in the flue gases leaving the catalyst bed may be recovered in
several ways including: a) use of a recuperative heat exchanger to preheat the
emission stream and/or combustion air, or b) by use of the available energy
for process heat requirements (e.g., recycling flue gases to the process,
producing hot water or steam, etc.). In recuperative heat exchange, only a
limited preheat is possible due to the temperature rise across the catalyst
bed as a result of the combustion of VOC in the emission stream. High preheat
temperatures accompanied by an increase in temperature due to combustion,
result in high operating temperatures at the catalyst bed, causing the
catalyst bed to overheat and eventually lose its activity.
The following discussion will be based on fixed bed catalytic incinerator
systems with no heat recovery and with recuperative heat exchange (i.e.,
preheating the emission stream). The calculation procedure will be
illustrated using Emission Stream 2 described in Chapter 3.
4.2.1 Data Required
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on the "HAP Emission Stream Data
Form" and the required HAP control as determined by the applicable regulations,
4.2-3
-------
1.
2.
3.
4.
5.
EXAMPLE CASE
Maximum flow rate, Qe - 20,000 scfm
Temperature, Tg - 120°F
Heat content, fr - 2.1 Btu/scf
C
Oxygen content, 0« - 20.6%
Moisture content, Ma « 2%
Based on the control requirements for the emission
stream:
Required destruction efficiency, DE = 95%
If dilution air is added to the emission stream upon exit from the
process, the data that will be used in the calculations are the resulting
characteristics after dilution.
In the case of a permit review for a catalytic incinerator, the following
data should be supplied by the applicant. The calculations in this section
will then be used to check the applicant's values.
Catalytic incinerator system variables at standard conditions (70 F, 1
atm):
1.
2.
3.
4.
Reported destruction efficiency, DEreDor+eci = %
Temperature of the emission stream entering the incinerator,
F (if no heat recovery); T.
F (if emission
stream is preheated)
Temperature of flue gas leaving the catalyst bed,
co
Temperature of combined gas stream (emission stream +
supplementary fuel combustion products) entering the catalyst
bed,a
If no supplementary fuel is used, the value for this variable will be
the same as that for the emission stream.
4.2-4
-------
5.
6.
7.
8.
9.
10.
11.
12.
Space velocity, SV - hr
Supplementary heat requirement, H
Flow rate of combined gas stream
Qcom " scfm
Combustion air flow rate, (L *
c
Excess air, ex - %
Catalyst bed requirement, V.. -
Fuel heating value, hr -
c = Btu/min
entering the catalyst bed,a
scfm
n3
Btu/scf
Heat exchanger surface area (if a heat exchanger is employed
A - ft2
4.2.2 Pretreatment of the Emission Stream: Dilution Air Requirements
In general, catalytic incineration is applied to dilute emission streams.
If emission streams with high VOC concentrations (i.e., heat content above 10
Btu/scf for air + VOC mixtures and above 15 Btu/scf for inert + VOC mixtures)
are treated by catalytic incineration, they may generate enough heat upon
combustion leading to deactivation of the catalyst. Therefore, dilution of
the emission stream with air is necessary to reduce the concentration of the
VOC's.
Typically, the concentration of flammable vapors in HAP emission streams
containing air is limited to less than 25 percent of the LEL (corresponding to
a heat content of 13 Btu/scf) for safety requirements. In order to meet the
safety requirements and to prevent damage to the catalyst bed, it is assumed
in this manual that catalytic incineration is applicable if the heat content
of the emission stream (air + VOC) is less than or equal to 10 Btu/scf. For
emission streams that are mixtures of inert gases and VOC ( i.e., containing
no oxygen), it is assumed that catalytic incineration is applicable if the
heat content of the emission stream is less than or equal to 15 Btu/scf.
Otherwise, dilution air will be required to reduce the heat content to levels
below these cut-off values (i.e., 10 and 15 Btu/scf). For emission streams
that cannot be characterized as air + VOC or inert + VOC mixtures, apply the
4.2-5
-------
10 Btu/scf cut-off value for determining dilution air requirements. See
Appendix B.3 for calculating dilution air requirements.
EXAMPLE CASE
Since the heat content of the emission stream
(hg) is 2.1 Btu/scf, no dilution is necessary.
4.2.3 Catalytic Incinerator System Design Variables
Table 4.2-1 presents suggested values and limits for the design variables
of a fixed bed catalytic incinerator system to achieve a given destruction
efficiency. For specific applications, other temperatures and space
velocities may be appropriate depending on the type of catalyst employed and
the emission stream characteristics (i.e., composition and concentration).
For example, the temperature of the flue gas leaving the catalyst bed may be
lower than 1,000°F for emission streams containing easily oxidized compounds
and still achieve the desired destruction efficiency. (Refer to Appendix B.6;
Table B.6-1 and Figure B.6-1 present data on temperatures typically required
for specific destruction efficiency levels for several compounds.)
Note that the destruction efficiency for a given compound may vary
depending on whether the compound is the only VOC in the emission stream, or
it is part of a mixture of VOC's. The destruction efficiency for a given
compound in different VOC mixtures may also vary with mixture composition (see
Figures B.6-2 and 3 where compound-specific destruction efficiency data are
presented for two different VOC mixtures).
Based on the required destruction efficiency (DE), specify the
appropriate ranges for TC-., TCQ, and select the value for SV from Table 4.2-1.
4.2-6
-------
TABLE 4.2-1. CATALYTIC INCINERATOR SYSTEM DESIGN VARIABLES'
Required
Destruction
Efficiency
DE (%)
90
Temperature
at the Catalyst
Bed In1etb
Te1 <°0
600
Temperature
at the Catalyst
Bed Outlet0
Tco <°f)
1,000-1,200
Space
Velocity
SV (hr"1)
40,000^
95
600
1,000-1,200
30,000*
Source: References 2, 3, and 4.
Minimum temperature of combined gas stream (emission stream + supplementary
fuel combustion products) entering the catalyst bed is designated as 600 F to
ensure an adequate initial reaction rate.
GMinimum temperature of flue gas leaving the catalyst bed is designated as
1,000 F to ensure an adequate overall reaction rate, to achieve the required
destruction efficiency. Note that this is a conservative value; it is in
general a function of the HAP concentration (or heat content) and a
temperature lower than 1,000 F may be sufficient to achieve the required
destruction level. Maximum temperature of flue gas leaving the catalyst bed
is limited to 1,200 F to prevent catalyst deactivation by overheating.
Corresponds to 1.5 ft of catalyst per 1,000 scfm of combined gas stream.
Q 3
Corresponds to 2.0 ft of catalyst per 1,000 scfm of combined gas stream.
4.2-7
-------
EXAMPLE CASE
The required destruction efficiency is 95%; therefore:
Tci (minimum) - 600°F
TCQ (minimum) » 1,000°F
TCQ (maximum) - 1,200°F
SV « 30,000 hr"1
In a permit evaluation, determine if the reported values for T ., T ,
C1 GO
and SV are appropriate to achieve the required destruction efficiency by
comparing the applicant's values with the values in Table 4.2-1. The reported
value for T . should equal or exceed 600°F in order to obtain an adequate
initial reaction rate. To ensure that an adequate overall reaction rate can
be achieved to give the desired destruction efficiency without damaging the
catalyst, check whether TCQ falls in the interval 1,000 - 1,200°F. Note that
1000°F is a conservative value. Then check if the reported value for SV is
equal to or less than the value in Table 4.2-1. If the reported values are
appropriate, proceed with the calculations. In some cases it may be possible
to achieve the desired destruction efficiency at a lower temperature level.
Otherwise, the applicant's design is considered unacceptable. In such a case,
the reviewer may then wish to use the values in Table 4.2-1.
4.2.4 Determination of Incinerator System Variables
4.2.4.1 Supplementary Heat Requirements--
Supplementary fuel is added to the catalytic incinerator system to
provide the heat necessary to bring the emission stream up to the required
catalytic oxidation temperature (T .) for the desired level of destruction
efficiency. For a given T ., the quantity of supplementary heat needed is
provided by: (a) the heat supplied from the combustion of supplementary fuel,
(b) the sensible heat contained in the emission stream as it enters the
catalytic incinerator system, and (c) the sensible heat gained by the emission
stream through heat exchange with hot flue gases. If recuperative heat
exchange is not practiced at a facility, then item (c) will be zero.
As mentioned earlier, emission streams treated by catalytic incineration
are dilute mixtures of VOC and air, and typically do not require additional
4.2-8
-------
combustion air. As a conservative cut-off value, it can be assumed that no
additional combustion air is required if the emission stream oxygen content
(CL) is greater than or equal to 16 percent.
Before calculating the supplementary heat requirements, the temperature
of the flue gas leaving the catalyst bed (T ) should be estimated to ensure
that an adequate overall reaction rate can be achieved to give the desired
destruction efficiency without damaging the catalyst. In other words, check
whether TCQ falls in the interval 1,000 - 1,200°F. Use the following
expression to calculate T , taking into consideration the temperature rise
across the catalyst bed due to heat generation from combustion of VOC in the
emission stream:
Tco = Tci + 50 he (4.2-1)
where:
hg = heat content of the emission stream, Btu/scf.
In this expression, it is assumed that the heat content of the emission stream
and the combined gas stream is the same. Refer to Appendix B.6 for details of
the equations used in Section 4.2. Inserting TC-. * 600°F, if TCQ is in the
range 1,000 - 1,200°F, then T . - 600°F is satisfactory. If T is less than
CO
1,000 F, use the following equation to determine an appropriate value for T .
(above 600°F) and use this value in the following calculations:
Tci = 1,000 - 50 he (4.2-2)
[Note: Emission streams with high heat contents will be diluted based on the
requirements discussed in Section 4.2.2. Therefore, values for TCQ exceeding
1,200°F should not occur.]
To calculate supplementary heat requirements (based on natural gas as the
fuel), use the following simplified equation for dilute emission streams that
require no additional combustion air:
4.2-9
-------
Hf - l.lhf X Qe(l+0.002Me)
(4.2-3)
V1-4 cP.lr
where:
Hr = supplementary heat requirement, Btu/min
hr * heating value of natural gas, Btu/scf
Q = maximum emission stream flow rate, scfm
M. * moisture content of the emission stream, percent
6
Cp . - average specific heat of air over a given temperature
interval, (Tr -T) Btu/scf-°F
T . * temperature of combined gas stream entering the catalyst bed, F.
T * reference temperature, = 70 F
T. = emission stream temperature after heat recovery, F
Note that for the case of no heat recovery, T. = T . The factor 1.1 accounts
for an estimated heat loss of 10 percent in the incinerator. Supplementary
heat requirements are based on maximum emission flow rate, and hence will lead
to a conservative design. In contrast to thermal incineration, there is no
minimum supplementary heat requirement specified for catalytic incineration
since no fuel is needed for flame stabilization. Depending on the VOC
concentration, emission stream temperature, and level of heat recovery,
supplementary heat requirements may be zero when heat recovery is practiced.
A plot of Equation 4.2-3 is shown in Figure 4.2-2 where the ratio of
Hr/Q. is plotted against the emission stream heat content for two levels of
heat recovery (zero and 50 percent). As an alternative to Equation 4.2-3, use
Figure 4.2-2 directly to determine Hr. The figure is based on the following
assumptions: (1) Moisture content of the emission stream (M ) is 2 percent,
(2) emission stream temperature (Tg) is 100°F, (3) preheat temperature of the
emission stream ("L ) is based on 50 percent heat recovery in the heat
exchanger, and (4) net heating value of supplementary fuel (natural gas) is
882 Btu/scf.
4.2-10
-------
5
u
<=?
-------
EXAMPLE CASE
Using Equations 4.2-1. 2. and 3:
Since the emission stream is dilute (h =2.1 Btu/scf)
and has an oxygen concentration greater than 16%
(CL = 20.6%}, these equations are applicable.
a. Determine if TCQ falls in the range 1,000-1200°F:
Tci = 600°F (Table 4'3)
he =2.1 Btu/scf (input data)
Tco = 600+(50 x 2-!) " 705°F
Since T is less than 1,000°F, use Equation 4.2-2
to calculate an appropriate value for T .:
Tc. = 1,000 - (50 x 2.1) = 895°F
b. Determine H^ (assume recuperative heat
recovery will be employed):
Qe = 20,000 scfm
Me . 2%
Tr - 70°F
The = 550°F (APPendix B-5' based
on heat recovery of 50%)
Cpair " °'0187 Btu/scf-°F for tne
interval 70-895°F (Table B.4-1)
cPa,-v " 0.0183 Btu/scf-°F for the
A
interval 70-550°F (Table B.4-1)
(15.43 - 8.78)
(882 - 21.60)
Hf - 1.1 x 882 x 20,000 x 1.004
Hf = 150,500 Btu/min.
Using Figure 4.2-2:
For h = 2.1 Btu/scf and using the curve
for 50% heat recovery, Hr/Q from the figure is
about 7.5 Btu/min/scfm. Multiplying 7.5 by Q ,
(7.5 x 20,000), yields an approximate value of
150,000 Btu/min for Hf.
4.2-12
-------
For emission streams that do not contain sufficient quantities of oxygen
to satisfy combustion air requirements (e.g., process emissions), refer to
Figure 4.2-3 which shows a plot of H£/Q versus h for two levels of heat
recovery (i.e., no heat recovery and where TV is 550°F). In this figure, the
oxygen content of the emission stream (Q*) is assumed as zero; this
corresponds to maximum combustion air requirements. The emission stream
moisture content (M ), emission stream temperature (T ) and the fuel heating
G 6
value (hf) are as specified for Figure 4.2-2. The preheat temperature of the
emission stream (Tue) is 550°F for the heat recovery case. If CL for a
particular emission stream is between 0 and 16 percent, use Figure 4.2-3.
4.2.4.2 Flow Rate of Combined Gas Stream Entering the Catalyst Bed--
In order to calculate the quantity of catalyst required, the flow rate of
the combined gas stream (emission stream + supplementary fuel combustion
products) at the inlet to the catalyst bed has to be determined. Use the
following equation:
where:
Q » flow rate of the combined gas stream, scfm
Qr - natural gas flow rate, scfm
Qc » combustion air requirement, scfm
Calculate Qr from the following expression:
Qf - Hf/hf (4.2-5)
As indicated earlier, Q will typically be zero for dilute emission streams
with oxygen contents (02) greater than 16 percent. If 02 is less than 16
percent, refer to Appendix B.6 for calculating Qc<
4.2-13
-------
(1) No heat recovery
(2) Heat recovery
(T, = 550°F)
(1) No heat recovery
(2) Heat recovery
(Tho = 550°F)
1
Emission Stream Heat Content, h. (Btu/scf)
Figure 4.2-3
Supplementary heat requirement vs. emission stream heat content.
(no oxygen in emission stream/maximum combustion air)
4.2-14
-------
EXAMPLE CASE
Using Equation 4.2-5:
Hf - 150,500 Btu/min
hf » 882 Btu/scf
Qr - 170 scfm
Using Equation 4.2-4:
Qe
'c
q
20,000 scfm
0 (since 02 is greater than 16%)
20,000 +170+0
20,170 scfm
4.2.4.3 Flow Rate of Flue Gas Leaving the Catalyst Bed--
In order to determine costs for incinerators and to size a heat exchanger
for preheating the emission stream, the flow rate of flue gas leaving the
catalyst bed has to be determined.
Assume that the flow rate of the combined gas stream entering the
catalyst bed is approximately equal to the flow rate of the flue gas leaving
the catalyst bed at standard conditions. The volume change across the
catalyst bed due to the combustion of the VOC in the mixed gas stream is small
especially when dilute emission streams are treated. Therefore,
Q
M
where:
QrQ - flow rate of the flue gas leaving the catalyst bed, scfm
While figuring costs, assume that catalytic incinerators are designed for a
minimum Q. of 500 scfm. Therefore, if Qrq is less than 500 scfm, define Q^
as 500 scfm.
4.2-15
-------
In order to determine operating costs, the flue gas flow rate (Qf ) has
to be expressed at actual conditions. Use the following equation to convert
Qf from "scfm" to "acfm":
(4.2-6)
fg.a • co
where Qfn , is the flue gas flow rate at actual conditions (acfm).
• a» a
EXAMPLE CASE
Using Equation 4.2-6:
Q
fg ~ Mcom
20,170 scfm
CO
fg,a
1,000UF
[20,170(1,000 + 460)/530]
55,600 acfm
4.2.5 Catalyst Bed Requirement
The total volume of catalyst required for a given destruction efficiency
is determined from the design space velocity as follows:
where:
"bed ' 60 «con/sv
Vbed = volume
bed required, ft"
EXAMPLE CASE
Using Equation 4.2-7:
Qcom
SV
Vbed
Vbed
= 20,170 scfm
= 30,000 hr"1 (Table 4.2-1)
= 60 x 20,170/30,000
40 ft3
(4.2-7)
4.2-16
-------
4.2.6 He aft Exchanger Size (for Systems with Recuperative Heat Exchange
Only)
To determine the size of the heat exchanger required for preheating the
emission stream to T. , use the following expression:
A - [60 Qe(l+0.002Me)Cpa.r(The-Te)]/UATLM (4.2-8)
where:
2
A - heat exchanger surface area, ft
U - overall heat transfer coefficient, 8tu/hr-ft2-°F
ATLM - logarithmic mean temperature difference, °F
Refer to Appendix B.5 for details of the heat exchanger design and for
the values to be used in the equation.
Alternatively, Figure 4.2-4 can be used to determine the heat exchanger
size. In this figure, the broken line represents Equation 4.2-8 evaluated for
dilute emission streams that do not require additional combustion air. The
assumptions inherent in this case are the same as those stated for Figure
4.2-2. The overall heat transfer coefficient is assumed as 4 Btu/hr-ft2-°F.
4.2-17
-------
o
o
o
350 -
340 -
330 -
S 320 -
O
oi
c
o
•I"
I/)
in
•r*
UJ
CM
(O
-------
EXAMPLE CASE
Using Equation 4.2-8:
Qe - 20,000 SCFM
Me - 2%
The * 550°F (APPendix B.5--based
on heat recovery of 50%)
Te - 120°F (input data)
Cpair - 0.0183 Btu/scf-°F for the interval
120-550°F (Table B.4-1)
U - 4 Btu/hr-ft2-°F
Since the emission stream is dilute, calculate
AT... as follows:
ATLM ' Tco-Tbe
°
1,000UF
ATLM « 1,000-550 - 450F
The heat exchanger surface area from Equation 4.2-8
then becomes:
A -
'60 x 20,000 x 1.004 x 0.0183 x (550-100)
4x450
U2
A - 5,500 ft'
Ujinq Figure 4.2-4:
For all values of he, A/Qe is about 275 x 10"3,
Thus, multiplying Qfi with A/Qg, (20,000 x
0.275) yields 5,500 ft2.
4.2-19
-------
For emission streams that require additional combustion air, use the
solid line in Figure 4.2-4 to obtain an estimate of the required heat
exchanger size. As in Figure 4.2-3, the' solid line is based on maximum
combustion air requirements (i.e., no oxygen in the emission stream). If
Figure 4.2-4 is not directly applicable in a particular situation, use the
broken line in the figure to obtain a conservative estimate for A.
4.2.7 Evaluation of Permit Application
Compare the results from the calculations and the values supplied by the
permit applicant using Table 4.2-2. The calculated values in the table are
based on the example case.
If the calculated values for Hf, Q , Qcom, Vbe£j, and A differ from the
reported values for these variables, the differences may be due to the
assumptions involved in the calculations. If that is the case, the reviewer
may wish to discuss the details of the proposed design with the permit
applicant.
If the calculated values agree with the reported values, however, then
the design and operation of the proposed catalytic incinerator system may be
considered appropriate based on the assumptions used in this manual.
4.2-20
-------
TABLE 4.2-2 COMPARISON OF CALCULATED VALUES AND VALUES
SUPPLIED BY THE PERMIT APPLICANT FOR
CATALYTIC INCINERATION
Calculated
Value Reported Value
(Example Case)
Supplementary heat requirement,
Supplementary fuel flow rate, Q
Combustion air flow rate, Q
Combined gas stream flow rate, Q
Catalyst bed volume, V. .
com
Heat exchanger surface area (if
recuperative heat recovery is used), A
150,500 Btu/min
170 scfm
0
20,170 scfm
40 ft3
5,500 ft2
Based on Emission Stream 2.
4.2-21
-------
4.2.8 References for Section 4.2
1. U.S. EPA. Parametric Evaluation of VOC/HAP Destruction Via Catalytic
Incineration. EPA-600/2-85-041. April 1985.
2. U.S. EPA. Organic Chemical Manufacturing. Volume 4: Combustion Control
Devices. EPA-450/3-80-026. December 1980.
3. U.S. EPA. Control of Volatile Organic Emissions from Existing Stationary
Sources - Volume I: Control Methods for Surface Coating Operations.
EPA-450/2-76-028. November 1976.
4. U.S. EPA. Guideline Series. Control of Volatile Organic Compound
Emissions from Manufacture of High-Density Polyethylene. Polypropylene.
and Polystyrene Resins. EPA-450/3-83-008. November 1983.
4.2-22
-------
4.3 FLARES
Flares use open flames for disposing of waste gases during normal
operations and emergencies. They are typically applied when the heating value
of the waste gases cannot be recovered economically because of intermittent or
uncertain flow, or when process upsets occur. In some cases, flares are
operated in conjunction with baseload gas recovery systems (e.g., condensers).
Flares handle process upset and emergency gas releases that the baseload
system is not designed to recover.
There are several types of flares, the most common of which are
steam-assisted, air-assisted, and pressure head flares. Typical flare
operations can be classified as: "smokeless," "non-smokeless," and "fired" or
"endothermic." For smokeless operation, flares use outside momentum sources
(usually steam or air) to provide efficient gas/air mixing and turbulence for
complete combustion. Smokeless flaring is required for destruction of
organics heavier than methane. Non-smokeless operation is used for
organic or other vapor streams which burn readily and do not produce smoke.
Fired, or endothermic, flaring requires additional energy in order to ensure
complete oxidation of the waste streams such as for sulfur tail gas and
ammonia waste streams.
In general, flare performance depends on such factors as flare gas exit
velocity, emission stream heating value, residence time in the combustion
zone, waste gas/oxygen mixing, and flame temperature. Since steam-assisted
smokeless flares are the most frequently used, they will be the focus of this
discussion. A typical steam-assisted flare system is shown in Figure 4.3-1.
First, process off-gases enter the flare through the collection header. When
water or organic droplets are present, it may be necessary to pass the
off-gases through a knock-out drum, since these droplets can create problems.
Water droplets can extinguish the flame and organic droplets can result in
burning particles. Once the off-gases enter the flare stack, flame flashback
can occur if the emission stream flow rate is too low. Flashback may be
prevented, however, by passing the gas through a gas barrier, a water seal, or
a stack seal. Purge gas is another option. At the flare tip, the emission
stream is ignited by pilot burners. If conditions in the flame zone are
optimum (oxygen availability, adequate residence time, etc.), the VOC in the
4.3-1
-------
Emission
Stream
Gas Collection Header
and Transfer Line
I
Knock-out Drum
Steam
Nozzles
Pilot Burners
Gas
Barrier
Flare
Stack
Steam Line
Ignition Device
Air Line
Gas Line
Figure 4.3-1. A typical steam-assisted flare system.
4.3-2
-------
emission stream may be completely burned (-100 percent efficiency). In some
cases, it may be necessary to add supplementary fuel (natural gas) to the
emission stream in order to achieve destruction efficiencies of 98 percent and
greater if the net heating value of the emission stream is less than 300
Btu/scf.1'4
Typically, existing flare systems will be used to control HAP emission
streams. Therefore, the following sections describe how to evaluate the
destruction efficiency of an existing flare system under expected flow
conditions (e.g., continuous, start-up, shut-down, etc.). The discussion will
be based on the recent regulatory requirements of 98 percent destruction
efficiency for flares. The calculation procedure will be illustrated for
Emission Stream 3 described in Chapter 3 using a steam-assisted flare system.
Note that flares often serve more than one process unit and the total flow
rate to the flare needs to be determined before the following calculation
procedure can be applied.
4.3.1 Data Required
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on the "HAP Emission Stream Data
Form," flare dimensions, and the required HAP control as determined by the
applicable regulations.
EXAMPLE CASE
Expected emission stream flowrate, Qe = 30,000 scfm
Emission stream temperature, Tg * 100° F.
Heat content, hg =» 180 Btu/scf
Mean molecular weight of emission stream,
MWe =33.5 Ib/lb-mole
Flare tip diameter, D^ = 48 in
Based on the control requirements for the
emission stream:
Required destruction efficiency, DE <* 98%
4.3-3
-------
In the case of a permit review, the data outlined below should be
supplied by the applicant. The calculations in this section will then be used
to check the applicant's values.
Flare system variables at standard conditions (70° F, 1 atm):
1. Flare tip diameter, 0^. - _ in
2. Expected emission stream flowrate, Q. - _ scfm
6
3. Emission stream heat content, h - _ Btu/scf
4. Temperature of emission stream, T » _ °F
g — — —
5. Mean molecular weight of emission stream, MW. - _ Ib/lb-mole
6 "~ ""
6. Steam flowrate, Q = _ Ib/min
7. Flare gas exit velocity, Ur-, - _ ft/sec
8. Supplementary fuel flow rate,3 Q~ * _ scfm
9. Supplementary fuel heat content, hf = _ Btu/scf
10. Temperature of flare gas, Tf1/, » _ ° F
h T|a -
11. Flare gas flow rate, 0-, - _ scfm
-
12. Flare gas heat content, h^, * _ Btu/scf
4.3.2 Determination of Flare Operating Variables
Based on studies conducted by EPA, relief gases having heating values
less than 300 Btu/scf are not assured of achieving 98 percent destruction
efficiency when they are flared in steam- or air-assisted flares.0 Therefore,
the first step in the evaluation procedure is to check the heat content of the
emission stream and determine if additional fuel is needed.
a This information is needed if the emission stream heat content is less
than 300 Btu/scf.
If no auxiliary fuel is added, the value for this variable will be the
same as that for the emission stream.
cFor unassisted flares, the lower limit is 200 Btu/scf.
4.3-4
-------
In a permit review case, if the heat content of the emission stream is
less than 300 Btu/scf and no supplementary fuel has been added, then the
application is considered unacceptable. The reviewer may then wish to follow
the calculations described below. If the reported value for the emission
stream heat content is above 300 Btu/scf, however, then the reviewer should
skip to Section 4.3.2.3.
4.3.2.1 Supplementary Fuel Requirements--
If the emission stream heat content is less than the 300 Btu/scf required
to achieve a destruction level of 98 percent, it is assumed that natural gas
will be added to the emission stream to bring its heat content to 300 Btu/scf.
Calculate the required natural gas requirements using the following equation:
where:
h. =
[(300 - he)Qe]/582
emission stream flow rate, scfm
natural gas flow rate, scfm
emission stream heat content, Btu/scf
(4.3-1)
(Refer to Appendix B.7 for details of all the equations used in Section 4.3.)
If the emission stream heat content is greater than or equal to 300 Btu/scf,
then Q-r » 0.
EXAMPLE CASE
Using Equation 4.3-1:
Since hg is less than 300 Btu/scf,
supplementary fuel is needed.
he » 180 Btu/scf
Qe » 30,000 scfm
Qf = (300-180)(30,OOOJ/582
Q = 6,200 scfm
4.3-5
-------
4.3.2.2 Flare Gas Flow Rate and Heat Content--
The flare gas flow rate is determined from the flow rates of the emission
stream and natural gas using the following equation:
where:
fl
Qe + Qf
flare gas flow rate, scfm.
(4.3-2)
Note that if Qf - 0, then Qfl - Qg.
The heat content of the flare gas (h
fig
) is dependent on whether
supplementary fuel is added to the emission stream.
or equal to 300 Btu/scf, then
When h is greater than
^-j
If h is less than 300 Btu/scf, since
supplementary fuel is added to increase h to 300 Btu/scf,
f-j
300 Btu/scf.
EXAMPLE CASE
Using Equation 4.3-2:
Qe - 30,000 scfm
Qr = 6,200 scfm
'ng
Since h.
36,200 scfm
180 Btu/scf, hflg » 300 Btu/scf.
4.3.2.3 Flare Gas Exit Velocity--
The flare gas exit velocity values presented in Table 4.3-1 to achieve at
least 98 percent destruction efficiency in a steam-assisted flare system are
based on studies conducted by EPA. Flare gas exit velocities are expressed
as a function of flare gas heat content. Determine the maximum allowable exit
velocity using the equation presented in Table 4.3-1.
4.3-6
-------
TABLE 4.3-1. FLARE GAS EXIT VELOCITIES3
Flare Gas Heat Content Maximum Exit Velocity
hflg (Btu/scf) Umax (ft/sec)
hflg < 300
300 < hf1g < 1,000 3.28 [10(°-00118 hflg
hr,n > 1,000 400
fig -
a Source: Reference 1.
If no supplementary fuel is used, hri_ =h .
c Based on studies conducted by EPA, waste gases having heating values
less than 300 Btu/scf are not assured of achieving 98 percent
destruction efficiency when they are flared in steam-assisted flares.
4.3-7
-------
EXAMPLE CASE
Since hr-j_ - 300 Btu/scf, use the equation
in Table 4.3-1 to calculate U :
max
Umay - 3.28[lo(°-00118hflg * °'908>]
lUClX
max
3
60 ft/sec
x 30° + °-908)
]
From the emission stream data (expected flow rate, temperature) and
information on flare diameter, calculate the flare gas exit velocity (U^ );
Use the following equation to calculate Uf-,n:
T 1 y
compare this value with U
ufig
where:
Uflg " exit velocity °f flare gas, ft/sec
= flare gas flow rate at actual conditions, acfm
> *
(4.3-3)
D^. - flare tip diameter, in
Use the following expression to calculate Qf, :
Qn« , = CQ*i. (T«« + 460)]/530
(4.3-4)
If U^-j is less than Umax, then the 98 percent destruction level can be
achieved. However, if U~ exceeds Umax, this destruction efficiency level
cannot be achieved. This indicates that the existing flare diameter is too
small to accommodate lower exit velocities for the emission stream under
consideration. Note that at very low flare gas exit velocities, flame
The minimum flare gas exit velocity for a stable flame
o
is assumed as 0.03 ft/sec in this manual. Thus, if U-i- is below 0.03
instability may occur.
is assumed as 0.03 ft/
ft/sec, the desired destruction efficiency may not be achieved. In summary,
4.3-8
-------
Ufl should fall in the range 0.03 ft/sec and Umax for a 98 percent
destruction efficiency level.
In a permit review case, if Ur-.. exceeds U_ax> then the application is
not acceptable. If Ur-iQ is below U and exceeds 0.03 ft/sec, then the
proposed design is considered acceptable and the reviewer may proceed with the
calculations.
EXAMPLE CASE
Using Equations 4.3-3 and 4;
Qfl - 36,200 scfm
Tflg - 95°F (see Appendix B.7)
Qflg,a * t36'200 (95 + 460)]/530
Qflg!a ' 37'900 acfm
Dt1p - 48 in.
Uflg - (3.06 x 37,900)/(48)Z
Uflg = 50 ft/sec
Since 0.03 ft/sec < Uf, - 50 ft/sec < Umax - 60 ft/sec,
the required level of 98% destruction efficiency can be
achieved under these conditions.
4.3.2.4 Steam Requirements--
Steam requirements for steam-assisted flare operation depend on the
composition of the flare gas and the flare-tip design. Typical values range
from 0.15 to 0.50 Ib steam/1b flare gas. In this manual, the amount of steam
required for 98 percent destruction efficiency is assumed as 0.4 Ib steam/lb
flare gas. Use the following equation to determine steam requirements:
Qs - 1.03 x 10'3 x Qflg x MWflg (4.3-5)
4.3-9
-------
where:
Q - steam requirement, Ib/min
EXAMPLE CASE
Using equation 4.3-5:
36,200 scfm
Qflg =
MWflg
30.6 Ib/lb-mole (see Appendix B.7)
,-3
Qs = 1.03 x 10 J x 36,200 x 30.6
Qc = 1,140 Ib/min
4.3.3 Evaluation of Permit Application
Compare the results from the calculated and reported values using Table
4.3-2. If the calculated values of Qr, Ur-|a, Qfia» and Q are different from
the reported values for these variables, the differences may be due to the
assumptions (e.g., steam to flare gas ratios, etc.) involved in the
calculations. In such a case, the reviewer may wish to discuss the details of
the proposed system with the permit applicant.
If the calculated values agree with the reported values, then the
operation of the proposed flare system may be considered appropriate based on
the assumptions made in this manual.
4.3-10
-------
TABLE 4.3-2. COMPARISON OF CALCULATED VALUES AND
VALUES SUPPLIED BY THE PERMIT APPLICANT
FOR FLARES
Calculated Value
(Example Case)
Reported Value
Supplementary fuel flow rate,
6,200 scfra
Flare gas exit velocity, U
fig
Flare gas flow rate, Or-,
Steam flow rate, Qt
50 ft/sec
36,200 scfm
1,140 Ib/min
Based on Emission Stream 3.
4.3-11
-------
4.3.4. References for Section 4.3
1. Federal Register. Volume 50. April 16, 1985. pp. 14941-14945.
2. U. S. EPA. Organic Chemical Manufacturing Volume 4: Combustion Control
Devices. EPA-450/3-80-026. December 1980.
3. U. S. EPA. Reactor Processes in Synthetic Organic Chemical Manufacturing
Industry—Background Information for Proposed Standards. Draft EIS.
October 1984.
4. U.S. EPA. Evaluation of the Efficiency of Industrial Flares: Test
Results. EPA-600/2-84-095. May 1984.
4.3-12
-------
4.4 BOILERS/PROCESS HEATERS
The application of boilers and/or process heaters as emission control
devices is very site-specific (see Section 3.1.1.4). Tire level of detail
required in the calculations for sizing such devices is beyond the scope of
this manual and, thus, are not presented.
4.4-1
-------
4.5 CARBON ADSORPTION
Adsorption Is a surface phenomenon whereby hydrocarbons and other
compounds are selectively adsorbed on the surface of such materials as
activated carbon, silica gel, or alumina. Activated carbon is the most widely
used adsorbent. The adsorption capacity of an adsorbent for a given VOC is
often represented by adsorption isotherms that relate the amount of VOC
adsorbed (adsorbate) to the equilibrium pressure (or concentration) at
constant temperature (see Figure 4.5-1). Typically, the adsorption capacity
increases with the molecular weight of the VOC being adsorbed. In addition,
unsaturated compounds are generally more completely adsorbed than saturated
compounds, and cyclical compounds are more easily adsorbed than linearly
structured materials. Also, the adsorption capacity is enhanced by lower
operating temperatures and higher concentrations. VOC's characterized by low
vapor pressures are more easily adsorbed than those with high vapor pressures.
Carbon adsorption is used for pollution control and/or for solvent
recovery in several industries. It is usually a batch operation, involving
multiple beds. The two main steps in the adsorption operation include
adsorption and regeneration, usually performed cyclically. For control of
continuous emission streams, at least one bed remains on line in the
adsorption mode while the other is being regenerated.
A typical batch operation (see Figure 4.5-2) can be described as follows:
The VOC-laden waste gas is passed through the carbon bed where the VOC's are
adsorbed on the bed surface. As the adsorption capacity of the bed is
approached, traces of VOC's appear in the exit stream, indicating that the
breakthrough point of the bed has been attained. The emission stream is then
directed to a parallel bed containing regenerated adsorbent, and the process
continued. Concurrently, the saturated bed is regenerated by the passage of
hot inert gases, low-pressure steam, or a combination of vacuum and hot gas.
Since adsorption is a reversible process, by supplying heat (equivalent to the
amount of heat released during adsorption), the VOC's on the bed can be
desorbed. A "heel" is always left on the bed because complete desorption is
technically difficult to achieve and economically impractical. During the last
part of the steam regeneration cycle, the hot bed saturated with water vapor
4.5-1
-------
O)
4->
l/l
O
.a
s_
c •—«
a> j—i
3
i— a»
o o
•M c
a)
s_ s_
o tu
<+- (4.
a>
w a:
S-
O 3
00 O
••- 00
i.
O
I/I
-------
S £
•M .c
X HJ
LU <£
1,
*
«
+
f
«
t
&.
V
(•—DO—-
£
4J
OO
•HXh
1/1
s_
O)
o
I/)
-CO—-
-HXI-f
O>
C
O
Q.
S_
O
C
O
.Q
S_
(TJ
U
•c
O)
JD
O)
X
fO
U
o
CM
t
LO
O)
s-
4.5-3
-------
is dried and cooled, usually with air. When steam is used as the regenerant,
the desorbed VOC's in the steam effluent are typically condensed. Then, the
VOC's are recovered either by simple decantation, in the case of
water-insoluble materials, or by distillation, in the case of water-soluble
materials. If high purity is required for the recovered VOC's, complex
distillation systems may be necessary, especially in cases where the VOC's
consist of mixtures of solvents.
In other designs, continuous adsorption can be accomplished by fluidized
bed adsorption. The fresh adsorbent flows down the adsorption section that
consists of a series of fluidized trays. The emission stream enters at the
bottom of the adsorption section and flows upward. The VOC's are
progressively adsorbed and the exit gas is discharged from the top stage. The
saturated adsorbent is continuously removed from the bottom and transferred to
the desorption section, where it is regenerated and returned to the adsorption
system.
For a given emission stream, the performance of a carbon adsorber as a
control device is affected by several variables including: (a) the adsorption
capacity of the carbon for the specific VOC in question (as determined from
the adsorption isotherm), (b) operating temperature, (c) adsorption and
regeneration cycle time, (d) amount and type of regenerant, and (e)
contaminants.
The discussion in the following sections will be based on a fixed-bed
carbon adsorption system with 2 parallel beds. Regeneration of the beds will
be carried out with low pressure steam. It is assumed that the desorbed VOC's
and steam will be condensed and the bed will be dried and cooled with air.
Another assumption is that the emission stream is free from liquid and/or
solid particles that may potentially blind the carbon beds. Emission Stream 4
described in Chapter 3 will be used to illustrate the calculation procedure.
4.5.1 Data Required
/
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on the "HAP Emission Stream Data
Form" and the required HAP control as determined.by the applicable regulations.
4.5-4
-------
EXAMPLE CASE
1.'Maximum flowrate, Qe * 15,000 scfm
2. Temperature, Tg » 90°F
3. Relative humidity, Rhym = 40%
4. HAP - toluene
5. Maximum HAP content, HAPg » 1,000 ppmv
Based on the control requirements
for the emission stream:
Required removal efficiency, RE - 95%
If dilution air is added to the emission stream upon exit from process,
the data that will be used in the calculations are the resulting characteris-
tics after dilution.
In a permit review case for a carbon adsorber, the following data
outlined below should be supplied by the applicant. The calculations in this
section will later be used to check the applicant's values.
Carbon adsorber (fixed-bed) system variables at standard conditions
(70°F, 1 atm):
1.
2.
3.
4.
5.
6.
7.
8.
Reported removal efficiency, REreported
HAP content, HAP,
ppmv
Emission stream flow rate, Qe »
Adsorption capacity of carbon bed,
AC - Ib HAP/100 Ib carbon
Number of beds, N -
Amount of carbon required, C =
Cycle time for adsorption, 9 j -
Cycle time for regeneration, 0
scfm
. lb
hr
reg
hr
4.5-5
-------
9. Emission stream velocity through the carbon bed,
U = ft/min
6 ~~~"~~
10. Bed depth, Zfaed = ft
11. Bed diameter, Dbed - ft
12. Steam ratio, St = Ib steam/1b carbon
4.5.2. Pretreatment of the Emission Stream
4.5.2.1 Cooling--
Adsorption of VOC's is favored by lower temperatures. If the temperature
of the emission stream is significantly higher than 100°F, a heat exchanger
may be needed to cool the emission stream to 100°F. Refer to Appendix B.5 for
determining the size of a heat exchanger required for such applications.
EXAMPLE CASE
The temperature of the emission stream is 90°F, which
is below 100°F. Therefore, cooling is not necessary.
4.5.2.2 Dehumidification--
Since water vapor competes with the VOC's in the emission stream for
adsorption sites on the carbon surface, emission stream humidity levels
exceeding 50 percent (relative humidity) are not desirable. In this manual,
it is assumed that if the relative humidity level of the emission stream is
above 50 percent, it will be reduced to 50 percent using additional equipment.
Dehumidification may be carried out by cooling and condensing the water
vapor in the emission stream. A shell-and-tube type heat exchanger can be
employed for this purpose. Refer to Section 4.7 where calculation procedures
for sizing condensers are described.
Another alternative for dehumidification is adding dilution air to the
emission stream if the dilution air humidity is significantly less than that
of the emission stream. However, since this will increase the size of the
adsorber system required, it may not be cost effective.
4.5-6
-------
EXAMPLE CASE
Since the relative humidity of the emission stream
is less than 50%, dehumidification is not necessary.
4.5.2.3 High VOC Concentrations--
If flammable vapors are present in emission streams that are mixtures of
VOC and air, the VOC content may be limited to below 25 percent of the LEL by
insurance companies. In some cases, it can be increased to 40 - 50 percent of
the LEL if proper monitoring and controls are used. In addition, since high
bed temperatures may occur due to heat released during adsorption, high VOC
concentrations may need to be reduced. The maximum practical inlet VOC
concentration is usually about 10,000 ppmv. In this manual, it is assumed
that the VOC content will be limited to less than 25 percent of the LEL. See
Table B.2-Xin Appendix B.2 for a list of LEL values for several compounds.
'2.
EXAMPLE CASE
The HAP concentration of the emission stream is 1,000
ppmv (toluene). This is below the 25% of the LEL for
toluene, which is 3,000 ppmv (see Table %
4.5.3 Carbon Adsorption System Design Variables
Table 4.5-1 presents suggested values for the design variables of a
carbon adsorber system to achieve a given outlet HAP concentration. If the
emission limit requirement is expressed as removal efficiency, the outlet HAP
concentration can be calculated from the required removal efficiency and the
inlet HAP concentration.
For specific applications, other values may be appropriate depending on
the emission stream characteristics and the type of carbon bed. For example,
the adsorption capacity for a given carbon bed is dependent on several factors
including the type of VOC in the emission stream, and the temperature and
humidity levels. Typically, the adsorption capacity is determined from the
'4.5-7
-------
TABLE 4.5-1 CARBON ADSORBER SYSTEM DESIGN VARIABLES1
Outlet . Adsorption Degeneration Steam Requirement
Concentration Cycle Time Cycle Time for Regeneration
HAPQ (ppmv)
ad
8 (hr) St(lb steam/1 b carbon)
70 2 2
10-12 2 2
0.3
1.0
Source: References 2 and 3.
'Emission stream exiting the carbon adsorber.
'Regeneration cycle also includes the time necessary for drying and
cooling the bed.
4.5-8
-------
adsorption isotherm of the compound under consideration. Refer to Figures
B.8-1 to B.8-3 in Appendix B.8 where adsorption isotherms for several
compounds are presented. Also, refer to Table B.8-1 where adsorption
capacities at specific conditions are listed for several compounds.
Based on the required removal efficiency, determine the outlet HAP
concentration using the following equation:
where:
HAPQ =• HAPe (1-0.01 RE) (4.5-1)
HAPQ « HAP content of the emission stream exiting the
adsorber, ppmv.
RE » Removal efficiency, %
Next, specify the appropriate values for 5,., &„._, and St using
aU i cCJ
Table 4.5-1.
HAPg
HAP »
o
EXAMPLE CASE
Using Equation 4.5-1:
RE = 95 percent
1,000 ppmv
1,000 (1-0.95)
HAPQ = 50 ppmv
Assuming the conditions for HAPQ = 70 ppmv
are approximately the same as those for
HAPQ = 50 ppmv, from Table 4.5-1,
*ad " * hrs
*reg ' 2 hrs
St * 0.3 Ib steam/1b carbon
4.5.4 Determination of Carbon Adsorber System Variables
4.5.4.1 Carbon Requirements--
In sizing a carbon adsorber system, the quantity of carbon required is
determined from the adsorption capacity of the carbon bed (based on the
4.5-9
-------
adsorption isotherm of the HAP in question) using the emission stream flow
rate and HAP concentration. For a fixed-bed adsorption system with N parallel
beds and a specified adsorption cycle 0 j, the following equation can be used
to calculate the carbon requirements:
where:
2[1.55 x 10'5N *adQe(HAPe-HAP0)MWHAp/AC] (4.5-2)
C » carbon requirement, Ib carbon
N - number of carbon beds
9. = adsorption cycle time, hr
Q * emission stream flow rate, scfm
MWHAP - molecular weight of HAP, Ib/lb-mole (for a mixture of
HAPs, MW^Ap will be defined as mean molecular weight)
AC * Adsorption capacity of the carbon bed,
Ib HAP/100 Ib carbon
For design purposes, the carbon requirement is generally multiplied by a
A
factor of 2 as indicated in Equation 4.5-2. This safety factor is an
allowance for build-up of a heel during regeneration (which results in a
reduced capacity); fluctuations in emission stream characteristics (e.g., HAP
concentration and composition, humidity, etc.). The value for AC is typically
determined from the adsorption isotherm of the specific HAP/carbon system. If
an isotherm for the HAP in question is not available, the isotherm for another
compound of similar molecular weight and boiling point may be used as an
approximation. See Table B.2-2 in Appendix B.2 for molecular weight and
boiling point data for several compounds. For additional data, see
Reference 5. In this manual, the isotherms shown in Figures B.8-1 to 3 or the
adsorption capacities from Table B.8-1 can be used to estimate AC. If no data
are available, use a conservative value of 5 Ibs HAP/ 100 Ib carbon for AC.
As an alternative, Figure 4.5-3 can be used to determine carbon
requirements. The figure is based on Equation 4.5-2 and evaluated at HAP =
10 and 70 ppmv for several inlet concentrations.
4.5-10
-------
u
l/l
V
4->
(T3
o
03
-------
If the emission stream contains a mixture of HAP's, Equation 4.5-2 should
be evaluated using appropriate adsorption- isotherms for each component and
then summed to determine C .
EXAMPLE CASE
Using Equation 4.5-2:
Assume N = 2
9&d = 2 hrs (from Table 4-5"1)
Qe = 15,000 scfm
HAPg =• 1,000 ppmv
HAPQ = 50 ppmv
MWHAP - MWtoluene " 92 lb/lb mole
AC = 20-25 Ib toluene/100 Ib carbon
This value is estimated from Figure 4.5-1
where adsorption isotherms for toluene are
plotted at different temperatures. To
obtain a conservative estimate for C_
assume AC = 20:
C = [2x1.55xlO~5x2x2xl5,000(1,000-50)92/20]
Creq = 8'100 1b
Using Figure 4.5-3:
'req'
For HAP
1,000 ppmv, HAP * 50 ppmv,
and AC = 20 Ib, C
is about 0.55
assuming the curve for HAP = 70 ppmv is
applicable. Thus,
C = 0.55 x 15,000 = 8,250 Ibs
4.5.4.2 Carbon Adsorber Size--
The size of the adsorber is determined using a two step calculation.
First, using the actual flow rate of the emission stream and its velocity,
4.5-12
-------
calculate the bed area. Typically, velocities used in industry range from 50
to 100 ft/rain depending on the system pressure. At high velocities, the bed
pressure- drop becomes too high for standard blowers; at lower velocities, the
bed becomes too large and expensive. A value of 100 ft/min is assumed in this
discussion. Use the following equation to calculate the required bed area:
Abed ' We <4-5-3'
where:
2
A^ . » bed area, ft
Q. , » emission stream flow rate at actual conditions, acfm
c, a
Ug « emission stream velocity, ft/min
In this expression, Q_ , is determined as follows:
e, a
Qe,a = Qe[Te + 46°)/53°] (4-5-4)
where Q and T are the flow rate and temperature of the emission stream.
From the bed area, calculate the bed diameter assuming a circular vessel; use
the following expression:
Dbed ' 2CAbed^°<5 =
where:
Diameter, ft
To calculate the bed depth, determine the volume occupied by carbon in
each bed. Assume a carbon bed density of P.^ , and use the following
equation to calculate the volume of carbon (per bed):
"carbon ' bed <4-5-6>
Having calculated vcarbon> tne bed depth can be determined as follows:
Zbed ' Vcarbon/Abed (4-5'7)
4.5-13
-------
where Z. . is the bed depth, ft. Hence, the required adsorber size for an
adsorption cycle time 6 , for obtaining RE percent removal efficiency for an
emission stream with flowrate Q is: D ft (diameter) by Z. . ft straight side
(minimum). Note that in cases where large flows ( > 20,000 scfm) of off-gases
are handled, three or more parallel beds may be used, reducing the bed size.
The cycle times for adsorption and regeneration will change accordingly.
EXAMPLE CASE
Using Equations 4.5-3. 4, 5. 6. and 7
= 90°F
*e,a
15,000 scfm
= 15,000[(90 + 460J/530]
= 15,565 acfm
U = 100 ft/min
"bed
Dbed
Assume P
15,565/100
bed
carbon
-bed
1.13 x (155.7)
- 30 lb/ft3
(8,100/2)/30 *
135/155,7 = 0.88
155.7 ft
0.5
2
14 ft.
135 ft'
• 1 ft
4.5.4.3 Steam Required for Regeneration--
Carbon beds may be regenerated by various means; the most common
regenerant used is steam. In this manual, regeneration with steam is followed
by condensation. The quantity of steam required for regeneration depends on
the required removal efficiency, (or outlet concentration) and on how much
material is to be desorbed from the bed. A certain amount of steam is
required to raise the bed to its regeneration temperature and provide the heat
of desorption. The major portion of the steam flow, about 60 - 70 percent,
acts as a carrier gas for the desorbed VOC's. It is not cost-effective to
achieve complete desorption; acceptable working capacities of adsorption can
be obtained without consuming large quantities of steam. For solvent recovery
4.5-14
-------
systems, a requirement of 0.25 to 0.35 Ib steam/lb carbon has usually been
specified. For applications where VOC concentrations are. low (e.g., odor
control), steam usage ratios are higher.
In this manual, it is assumed that with a steam ratio of 0.3 Ib steam/lb
carbon, a HAP outlet concentration of 70 ppmv can be achieved after regener-
ation, and with a ratio of 1 Ib steam/lb carbon, a HAP outlet concentration of
10-12 ppmv can be achieved (Table 4.5-1). The regeneration cycle time, 0req>
is dependent on the time required to regenerate, dry, and cool the bed. The
flow rate of the steam used for regeneration can be determined using the
following expression:
(4.5-8)
where:
Q = steam flowrate, Ib/min
^dry-cool = cyc^e time ^or drying anc* cooling the bed, hr
Typically, cooling and drying the bed with air can be carried out in about 15
minutes. Figure 4.5-4 can also be used to estimate steam requirements. This
figure is based on 9 = 2 hrs and ^rv-cool = ^'^ hrs'
Steam flow rates based on cross-sectional area of the bed (Qs/Au d) are
generally limited to less than 4 Ib steam/min-ft to prevent the carbon from
being fluidized in the bed. If Qs/Abec| exceeds 4, the regeneration cycle time
or the steam ratio may need to be modified.
4.5-15
-------
1,000
500
-Q
c
j
o>
s_
cr
01
cc.
-------
EXAMPLE CASE
Using Equation 4.5^8:
RE = 95%
HAP.
St
e
Assuming e
reg
= 50 ppmv
0.3 Ib steam/lb carbon (Table 4.5-1)
= 2 hrs (Table 4.5-1)
0.25 hrs:
dry-cool
Qs = [0.3(8,100)/(2-0.25)]/60
Qs - 23 Ib/min
QC/A
'bed
23/155.7 = 0.15 Ib steam/min-ft'
Since Qs/Abec| is less than 4 Ib steam/min ft ,
fluidization in the carbon bed is not expected.
Using Figure 4.5-4:
At C
req
Q,
8,100 Ib and St
20 Ib/min
0.3,
4.5.4.4 Condenser--
The steam used for regenerating the carbon bed containing the desorbed
VOC's is typically condensed. The heat duty of the condenser is based on the
amount of steam required to regenerate the bed; the amount of heat absorbed by
the bed and later removed by the drying and cooling air can be considered
negligible.
Use the following expression to calculate the condenser size:
con
H
load'
,/UAT
LM
(4.5-9)
where:
con
H
load
condenser surface area, ft
= condenser heat load, Btu/hr
U
AT
2 or
LM
overall heat transfer coefficient, Btu/hr-ft - F
logarithmic mean temperature difference, °F
4.5-17
-------
In this manual, it is assumed that following regeneration, steam will be
condensed and subcooled to 100°F with cooling water. Note that Equation 4.5-9
is an approximate expression; condensation and subcooling processes are
combined and average values are used for U and AT, M. See Appendix B.6 for
calculation of AT..J. To calculate H-, ,, use the following equation:
"load'1'1 X60x(>s ^+£pw{Tst. -Tsto^ <4-5
where:
X = latent heat of vaporization, Btu/lb
Cp.. » average specific heat of water, Btu/lb-°F
n
T = steam inlet temperature, F
T . = condensed steam outlet temperature, °F
In this expression, the condenser heat load is oversized by 10 percent. In
the following calculations, steam available for regeneration is at atmospheric
pressure and 212°F. The latent heat of vaporization of steam at these
conditions is 970 Btu/lb. The temperature rise of the cooling water
available at 80°F is 50°F. The overall heat transfer coefficient is assumed
as 150 Btu/hr-ft2-°F.2
To determine cooling water requirements, use the following equation:
where:
QC001 = cooling water flow rate, Ib/hr
T . = cooling water inlet temperature, °F
A
T * cooling water outlet temperature, F
WO
Q , can be expressed in terms of gal/min as follows:
Qw = Qcool,w X C(1/60) x d/62.43) x 7.48] = 0.002 x Qcool>w (4.5-12)
where the factor "62.43" is the density of water and the factor "7.48" is used
for converting from "ft3" to "gal" basis.
4.5-18
-------
EXAMPLE CASE
Using Equation 4.5-10:
Qs - 23 Ib/min
x - 970 Btu/lb (Reference 6)
Cpw - 1 Btu/lb-°F (Reference 6)
T$t1 - 212°F
Tsto ' 10° F
Hload " ^x60*23 [970+lx(212-100)3
Hload = ^fi*2*500 Btu/hr
Using Equation 4.5-9:
U = 150 Btu/hr-ft2-°F
ATLM - (212-130) - (100-80)
|_ln[(212-130)/(100-80)]
See Appendix B.8 for details.
AT,
'LM
44°F
1,642,500/(150 x 44)
,-™
con
Acon ' 25° ft
Using Equations 4.5-11 and 12:
AT « 50°F
Qw =• 0.002[1,642,500/(lx50)]
Q - 66 gal/min
4.5.4.5 Recovered Product--
To calculate costs, the quantity of recovered product that can be sold
and/or recycled to the process has to be calculated. Use the following
equation:
4.5-19
-------
Qrec = 60 x [Qe x (HAPe x 10'°)(1/387)(0.01RE)MWHAD]
Qrec - 1.55 x 10'9 Qe x HAPe x RE x MWHAp
HAp
(4.5-13)
where Q is the quantity of recovered product, Ib/hr. In this equation, the
factor "387" is the volume (scf) occupied by 1 Ib-mole of ideal gas at
standard conditions (70°F and 1 atm).
EXAMPLE CASE
Using Equation 4.5-13:
Qe = 15,000 scfm
HAPe = 1,000 ppmv
RE = 95 %
MWHAp = 92 Ib/lb-mole
Q
rec
Vec
1.55x10 Jxl5,OOOxl,000x95x92
200 Ib/hr
4.5.5 Evaluation of Permit Application
Compare the results from the calculated values and reported values using
Table 4.5-2. If the calculated values of Creq, Dbed, Zbed, Qs, ACQn, Qw, and
Qrec are different from the reported values for these variables, the
differences may be due to the assumptions involved in the calculations.
Therefore, the reviewer may wish to discuss the details of the proposed design
with the permit applicant.
If the calculated values agree with the reported values, then the design
and operation of the proposed carbon adsorber system may be considered
appropriate based on the assumptions made in this manual.
4.5-20
-------
TABLE 4.5-2
COMPARISON OF CALCULATED VALUES AND VALUES
SUPPLIED BY THE PERMIT APPLICANT FOR
CARBON ADSORPTION
Calculated Value
(Example Case)1
Reported Value
req
Carbon requirement, C
Bed diameter, D. .
Bed depth, Zbed
Steam rate, Q
Condenser surface area, A
con
Cooling water rate, Q,
w
Recovered product, Q
rec
8,100 Ib
14 ft
1 ft
23 Ib/min
250 ft2
66 gal/min
200 Ib/hr
aBased on Emission Stream 4.
4.5-21
-------
4.5.6 References for Section 4.5
1. Chandrasekhar R. and E. Poulin. Control of Hydrocarbon Emissions From
Cotton and Synthetic Textile Finishing Plants. EPA Contract No.
68-02-3134. May 1983.
2. U. S. EPA. Organic Chemical Manufacturing Volume 5: Adsorption.
Condensation, and Absorption Devices. EPA-450/3-80-027. December 1980.
3. Parmele, C.S., W.L. O'Connell, and H.S. Basdekis. Vaoor-phase Adsorption
Cuts Pollution, Recovers Solvents. Chemical Engineering.
December 31, 1979. pp. 58-70.
4. Vatavuk, W.M. and R.B. Neveril. Part XIV: Costs of Carbon Adsorbers.
Chemical Engineering. January 24, 1983. pp. 131-132.
5. Chemical Engineer's Handbook. Perry, R.H. and C.H. Chilton (eds.) Fifth
Edition. McGraw-Hill Book Company. New York, NY. 1973.
6. Smith, J.M. and H.C. Van Ness. Introduction to Chemical Engineering
Thermodynamics. Second Edition. McGraw-Hill Book Company, Inc. and
Kogakusha Company, Ltd. Tokyo. 1959.
4.5-22
-------
4.6 ABSORPTION
Absorption is an operation in which one or more components of a gas
mixture are selectively transferred into a relatively nonvolatile liquid.
Absorption of a gaseous component by a liquid occurs when the liquid contains
less than the equilibrium concentration of the gaseous component. The
difference between the actual concentration and the equilibrium concentration
provides the driving force for absorption. The absorption rate depends on the
physical properties of the gaseous/liquid system (e.g., diffusivity, viscosity,
density) and the absorber operating conditions (e.g., temperature, flow rates
of the gaseous and liquid streams). It is enhanced by lower temperatures,
greater contacting surface, higher liquid-gas ratios, and higher concentrations
in the gas stream.
Absorption can be physical or chemical. Physical absorption occurs when
the absorbed compound simply dissolves in the solvent. When there is a
reaction between the absorbed compound and the solvent, it is termed chemical
absorption. Liquids commonly used as solvents for organic and inorganic
compounds include water, mineral oils-, nonvolatile hydrocarbon oils, and
aqueous solutions (e.g., sodium hydroxide).
The types of equipment commonly used for gas/liquid contact operations
include packed towers, plate or tray towers, spray chambers, and venturi
scrubbers. These devices are designed to provide maximum contact between the
gas and liquid streams in order to increase the mass transfer rate between the
two phases. A packed tower is filled with packing material that is designed
to expose a large wetted surface area to the gas stream. Plate towers use
plates or trays that are arranged so that the gas stream is dispersed through
a layer of liquid on each plate. Bubble-cap plates have been widely used;
other types of plates include perforated trays and valve trays. In a spray
tower, the gas mixture is contacted with a liquid spray. In a venturi
scrubber, the gas and liquid streams come into contact at the throat of the
venturi nozzle; venturi scrubbers are typically used for removal of particulate
matter (see Section 4.10).
4.6-1
-------
Several different configurations of absorber systems are used for
controlling vapor emissions. The simplest configuration is one in which the
solvent (usually water) is used on a once-through basis, and is then either
discharged to a wastewater treatment system or introduced as a process water
stream (see Figure 4.6-1). The possibility of using solvents other than water
on a once-through basis may exist when fresh solvent is available in large
quantities as a process raw material or fuel. Another configuration involves
using the solvent (usually water) on a once-through basis and stripping it
(reverse of absorption) before discharging. In yet another configuration, an
organic liquid is used as a solvent and recycled to the absorber after being
stripped.
The efficiency of absorption for removing pollutants from gaseous streams
depends on several factors including: (a) solubility of the pollutant in a
given solvent, (b) concentration, (c) temperature, (d) flow rates of gaseous
and liquid streams (liquid to gas ratio), (e) contact surface area, and (f)
efficiency of stripping (if solvent is recycled to the absorber).
Determination of the absorber system variables (absorber column diameter,
height, etc.,) is dependent on the individual vapor/liquid equilibrium
relationship for the specific HAP/solvent system and the type of absorber to
be used (packed or plate tower, etc.). Note that equilibrium data may not be
readily available for uncommon HAP's.
Detailed design procedures for all types of absorbers are not appropriate
for this manual; therefore, important design considerations for one type of
absorber will be briefly discussed. Since packed towers are commonly used in
air pollution control, the discussion will be based on packed tower absorbers.
For illustration purposes, a simple configuration is chosen for the absorber
system: a packed tower absorber 2-inch ceramic Raschig rings as the packing
material with water used as the absorbent on a once-through basis. The
effluent from the absorber is assumed to be discharged to a wastewater
treatment facility. The treatment in the following subsections is equally
applicable to both organic and inorganic vapor emissions control. For more
information on gas absorption, see References 1, 2, 3, 4, and 5.
4.6-2
-------
Emission .
Str.eam Outlet
Solvent Inlet
Emission Stream
Inlet
•*• Solvent
Outlet
Figure 4.6-1. A typical countercurrent packed column absorber system.
4.6-3
-------
As indicated in Chapter 3, absorption is the most widely used control
method for inorganic vapor emissions; therefore, Emission Stream 5 containing
inorganic vapors will be used in the example case to illustrate the
calculation procedures.
4.6.1 Data Required
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on.the "HAP Emission Stream Data
Form" and the required HAP control as determined by the applicable
regulations.
EXAMPLE CASE
1.
2.
3.
4.
5.
Based
Maximum flow rate, Q - 3,
Temperature, T = 85 F
HAP * ammonia
HAP concentration, HAPg -
Pressure, P = 760 mm Hg
on the control requirements
000 scfm
20,000 ppmv
for the emission stream:
Required removal efficiency, RE = 98%.
In the case of a permit review for an absorber, the following data
outlined below should be supplied by the applicant. The calculations in this
section will then be used to check the applicant's values.
Absorption system variables at standard conditions (70°F, 1 atm):
1.
2.
3.
4.
Reported removal efficiency, REreporte(j * _
Emission stream flow rate, Q. - scfm
e ~~~~—
Temperature of emission stream, T =
HAP =
4.6-4
-------
5. HAP concentration, HAP « ppmv
6 ~
6. Solvent used » _.
7. Slope of the equilibrium curve, m =»
8. Solvent flow rate, L.O * gal/mi n
gai -
9. Density of the emission stream, p- • Ib/ft
10. Schmidt No. for the (HAP/emission stream) and (HAP/solvent)
systems:
ScG
ScL -
(Refer to Appendix B.9 for definition and calculation of SCQ
and SC,)
11. Properties of the solvent:
Density, PL - Ib/ft3
Viscosity, ML - centipoise
12. Type of packing used -
13. Packing constants:
a » b - c « d « < -
Y « s » g - r »
14. Column diameter, D -ju' * ft
15. Tower height (packed), Htcolumn - ft.
16. Pressure drop,APt t i - in HgO.
4.6.2. Absorption System Design Variables
In absorption, the removal efficiencies (or outlet concentrations) are
limited by the driving force available from gas to the liquid phase. The
driving force for a given set of operating conditions is determined by the
difference between the actual HAP concentrations in the gas stream and solvent
and the corresponding equilibrium concentrations. When the slope (m) of the
equilibrium curve is small for a given HAP/solvent system, indicating that the
HAP is readily soluble in the solvent, the driving force for absorption is
large and absorption occurs readily. On the other hand, if "m" is large
(e.g., > 50), the HAP is not readily soluble in the solvent and the driving
4.6-5
-------
force for absorption is small; therefore, long contact times, tall absorption
towers, and/or high liquid-gas ratios are required for adequate performance
(high removal efficiency and/or low outlet concentrations). Hence, as a
conservative guideline, assume that if "m" is greater than about 50 for a
given HAP/solvent system at atmospheric pressure, then high removal
efficiencies (- 99 percent) are not possible.
4.6.3 Determination of Absorber System Design and Operating Variables
In most applications involving the absorption of a gaseous pollutant from
an effluent gas stream, the inlet conditions (flow rate, composition, and
temperature) are usually known. The composition of the outlet gas is
specified by the control requirements. The conditions of the inlet liquid are
also known. The main objectives, then, in the design of an absorption column
will be the determination of the solvent flow rate and the calculation of the
principal dimensions of the equipment (column diameter and height to
accomplish the absorption operation) for a selected solvent.
To keep the discussion simple, the following assumptions are made: (1)
there are no heat effects associated with the absorption operation, and (2)
both the gas and liquid streams are dilute solutions (i.e., flow rates are
constant throughout the absorption column and the equilibrium curve can be
approximated as a straight line). All of the data (e.g., packing factors,
Schmidt numbers, etc.) required in the calculation of the design variables are
presented in Appendix B.9.
4.6.3.1 Solvent Flow Rate--
The quantity of solvent to be used is typically estimated from the
minimum liquid-gas ratio as determined from material balances and equilibrium
considerations. As a rule of thumb for purposes of rapid estimates, it has
been frequently found that the most economical value for the absorption factor
2
(defined below) will be in the range from 1.25 to 2.0.
AF ' m Gmol t4-6'1)
4.6-6
-------
where:
AF - absorption factor
LmoT * 11c'u''d (solvent) flow rate, Ib-moles/hr
Gmol * gas stream fl°w rate» Ib-moles/hr
m - slope of the equilibrium curve
The value of "m" is determined from the equilibrium data at a specific
temperature level for the HAP/solvent system under consideration. See
Figures B.9-1, 2, and 3 where equilibrium data for a number of systems are
plotted. For information on other systems, see References 1, 4, and 5.
Assuming a value of 1.6 for AF, use Equation 4.6-1 to calculate the solvent
flow rate:
Lmol = J-6 m Smol (4'6-2>
The variable G , can be expressed in terms of Q as follows (see
Appendix B.I):
Gmol - 0-155 Qe (4.6-3)
Note that Lmol can be converted to gal/hr basis as follows:
Lgal ' tLmol * ""solvent x <"V » 7'W> <^6^
where:
L i - solvent flow rate, gal/mi n
solvent * m°lecu1ar weight of solvent, Ib/lb-mole
PL - density of solvent (liquid), lb/ft3
The factor "7.48" is used to convert from "ft " to "gal" basis. For water as
the solvent, PL - 62.43 lb/ft3 (Reference 1) and MWSQlvent - 18 Ib/lb mole;
then,
L9al ' °-036 Lmol
4.6-7
-------
EXAMPLE CASE
Using Equations 4.6-2. 3, and 5;
m = 1.3
(for the operating conditions in
the system; see Figure B.9-1)
Qg - 3,000 scfm
-0.155 x 3,000 x 60
amol
3mol
LmQl - 1.6 x 1.3 x 465
Lmol = 97° I
-gal
•gal
0.036 x 970
35 gal/min
4.6.3.2 Column Diameter--
Once the gas and liquid streams entering and leaving the absorber column
and their concentrations are identified, flow rates calculated, and operating
conditions (type of packing) determined, the physical dimensions of the column
can be calculated. The column must be of sufficient diameter to accommodate
the gas and liquid streams.
The calculation of the column diameter is based on flooding
considerations, the usual operating range being taken as 60 to 75 percent of
the flooding rate. One of the commonly used correlations in determing the
column diameter is shown in Figure 4.6-2. The procedure to calculate the
column diameter is as follows: First, calculate the abscissa (ABS):
ABS - (L/G) (PG/PL)°'5
where:
L * solvent flow rate, Ib/hr
G = gas stream flow rate, Ib/hr
(4.6-6)
4.6-8
-------
l.U
C5
o.
_1
Q,
O
CP>
S 0.1
o
_j
3.
CO
<
>n
CM
*; 0-01
0}
2
-------
3
PQ = density of emission stream, Ib/ft
The values for the variables l.and G can be calculated by multiplying L , and
G , with their respective molecular weights. Then proceed to the flooding
line in Figure 4.6-2 and read the ordinate (ORD), and solve the ordinate
expression for Gav.aa f at flooding:
area y T
ORD ' ^G2 W3)(*i'ZW ' G PL *c ^'^
Thus,
Garea,f = {[ORD PG PL 9c]/[(a/e3>(ML°'2)]}0'5 - (4'6'8)
where:
G,-«, f = gas stream flow rate based on column cross sectional area
area, I n
(at flooding conditions), Ib/ft -sec
a, « * packing factors (see Table B.9-1)
M, = viscosity of solvent, centipoises
2
g - gravitational constant, ft/sec
v*
Assuming f as the fraction of flooding velocity appropriate for the proposed
operation, the gas stream flow rate (based on cross sectional area) can be
expressed as:
Garea = f Garea,f <4-6-9>
The usual column operating range for f is taken as 0.60 to 0.75. Calculate
the column cross sectional area by the following expression:
Acolumn ' G/(3'600 Garea>
The column diameter is then determined by:
"column ' <4/")°-5 • '
where:
Dcolumn = column diameter> ft
4.6-10
-------
EXAMPLE CASE
Using Equations 4.6-6. 7. 8. 9. 10. and 11;
L ' ""solvent x Lmol ' 18 * 97° " 17*460 lb/hr
G - MWe x Gm()1 « 28.4 x 465 * 13,200 Ib/hr
(see Appendix B.I for calculating MWe)
PG - 0.071 lb/ft3 (from ideal gas law at 85°F)
(See Appendix B.9 for calculating PG)
PL - 62.18 lb/ft3 (Reference 1; at 85°F)
ABS - (17,460/13,200)(0.071/62.18)°'5 = 0.045
From Figure 4.6-2, at ABS » 0.045, the value of ORD at
flooding conditions is about 0.15. For 2-inch ceramic
Raschig rings, from Table 8.9-1 in Appendix B.9:
a » 28
« - 0.74
Also,
Thus,
gc - 32.2 ft/sec2
M, - 0.85 cp (Reference 1, at 85°F)
** f - ffO.15 x 0.071 x 62.18 x 32.21]0'5
' •» o ?
I [28/(0.74r](0.85)°-Z J
„ f - 0.56 lb/sec-ft2 (at flooding)
Assuming f * 0.60
Thus,
Garea * °'60 x °'56 * °'34
Acolu«, • ",200/(3,600 x 0.34)
10'8 ft
„,
Dcolu«, - 1-» <10-8> ' 3-7 -
4.6-11
-------
4.6.3.3 Column Height--
The column must be of sufficient height to ensure that the required
removal efficiency is achieved. The height of a packed column is calculated
by determining the required number of theoretical transfer units and
multiplying by the height of a transfer unit. A transfer unit is a measure of
the difficulty of the mass transfer operation and is a function of the
solubility and concentrations of the solute in the gas and liquid streams. It
is expressed as NQG or N«. depending on whether the gas film or liquid film
resistance controls the absorption rate. In emission control applications,
gas film resistance will typically be controlling, therefore NOG will be used
in the following calculations.
The expression for the column height (packed) is:
Htcolumn ' NOG HOG . (4-6'12>
where:
Ht -I « packed column height, ft
NQ» = number of gas transfer units (based on overall gas film
coefficients)
^OG = ne^9n^ °f an overall gas transfer unit (based on overall gas film
coefficients), ft
Although the determination of NQ6 is usually complicated, when dilute
solutions are involved, NQg can be calculated using the following equation:
NQG - In {(HAPe/HAPQ)[l - (1/AF)] + (1/AF)}/[1 - (1/AF)] (4.6-13)
This expression is simplified based on the assumption that no HAP is present
in the solvent as it enters the column (see Appendix B.9 for details).
Alternatively, use Figure 4.6-3 directly to determine NQG.
The variable HQG is generally calculated from the following equation:
HOG = HG + <1/AF) HL (4.6-14)
4.6-12
-------
NOG
40
32
24
16
8
10
100 1,000 10,000
HAPe/HAP0
Figure 4.6-3. NQG for absorption columns with constant absorption factor AF
(Source: Reference 6)
4.6-13
-------
where:
HS = height of a gas transfer unit, ft
H, » height of a liquid transfer unit, ft
Generalized correlations are available to calculate H« and H,; these are based
on the type of packing and the gas and solvent flow rates. The correlations
for H- and H, are as follows:
u L
HG - [b(3,600 Garea)C/(L")d](ScG)°-5 (4.6-15)
HL =. Y(LY<*L")S(ScL)°'5 (4.6-16)
where:
b, c, d, Y, and s » empirical packing constants (see Tables B.9-2 and 3)
L" = liquid flow rate, lb/hr-ft2
ML" - liquid viscosity, Ib/ft-hr
Sc~ » Schmidt number for the gas stream
u J
Sc, » Schmidt number for the liquid stream
The values for SCQ and Sc. are listed for several compounds in Tables B.9-4
and 5 (see Appendix B.9 for definition and calculation of ScG and Sc.). In
the calculations, it is assumed that the effect of temperature on Sc is
negligible. The value for the variable L" in this equation is calculated as
follows:
L" ' L/Acolumn <4-6-17>
Use the following expression to calculate the total column height (H% tai)«
Httotal = Htcolumn + 2 + °'25 Dcolumn . (*'*
4.6-14
-------
EXAMPLE CASE
a. Calculation of NQQ
Using Equation 4.6-13:
HAP = 20,000 ppmv
HAPQ = 20,000 (1-0.98) - 400 ppmv
NQG - ln[(20,000/400)0.375+0.625]/0.375
NOG = 7'9
Using Figure 4.6-3:
HAPe/HAPQ = 20,000/400 = 50
At AF - 1.6, 1/AF - 0.63, and
b. Calculation of Hftr:
Ub
Using Equations 4.6-14. 15. 16, and 17:
L" = 17,460/10.8 = 1,617 lb/hr-ft2
3,600 G = 1,224 lb/hr-ft2
From Tables B.9-2 and 3, the packing
factors are:
b - 3.82
c - 0.41
d = 0.45
Y = 0.0125
s » 0.22
Although 1,224 lb/hr-ft2 is outside the
range shown in the table, assume that
the packing factors are applicable and
the error introduced into the calcula-
tions will be negligible.
From Tables B.9-4 and 5:
ScQ = 0.66
ScL = 570
4.6-15
-------
Also,
MLH =-0.85 x 2.42 = 2.06 Ib/ft-hr
(The factor "2.42" is used to convert from
"centipoise" to "Ib/ft-hr".)
Hence,
HG = [3.82(1,224)°-41/(1,617)°-45](0.66)°-5
- 2.06
HL = 0.0125(1, 617/2. 06)°'22(570)°-5
= 1.29
Using AF = 1.6,
HQG - 2.06 + (1/1.6) 1.29 = 2.87 - 2.9
c. Calculation of HtC{jlufnn:
Using Equation 4.6-12:
Ht
co1umn - 7'9
22'9 - 23 ft
d. Calculation of
Using Equation 4.6-18:
Ht
total
= 23 + 2 + (0.25 x 4) = 26 ft
For costing purposes, it is necessary to calculate the column weight,
Use the following equation (see Appendix B.9 for details):
^
where:
column - <48 Column x Httotal) * 39 ^column)
Wtcolumn = column wei9ht' lb
Also, to determine packing costs, volume occupied by the packing material
(V -^ ) has to be calculated. Use the following expression:
packing
acking
X ^
x Ht
column
column
(4.6-20)
4.6-16
-------
EXAMPLE CASE
Usino Equation 4.6-19:
°column ' 4 ft
Ht
Wt
Wt
total
column
26 ft
» (48 x 4 x 26) + 39(4)J
column a 5>500 lb
Using Equation 4.6-20:
Ht
column
packing
packing
23 ft
0.785 x (4)'
290 ft3
x 23
4.6.3.4 Pressure Drop Through the Column--
The pressure drop through a packed column for any combination of liquid
and gas flows in the operable range is an important economic consideration in
the design of such columns. For a particular packing, the most accurate data
will be those available from the manufacturer. For purposes of estimation,
use the following correlation:
APa - (gxlO-8)[10
-------
EXAMPLE CASE
Using Equation 4.6-21:
From Table B.9-6:
g - 11.13
r - 0.00295
Also,
L" = 1,617 lb/hr-ft2
3,600 Garea - 3,600 x 0.34
- 1,224 lb/hr-ft2
PQ = 0.071 lb/ft3
PL - 62.18 lb/ft3
Thus,
F11.13 x IP"8 x 10(0-00295 x 1,617/62.18)^^2,
(0.071)
APa = 2.8 Ib/ft2-ft
at
Using Equation 4.6-22:
Htco1umn ' 23 ft
APtotal - 2.8 x 23 - 64.4 lb/ft*
APtotal - 64.4/5.2 = 12 in,H20
(The factor "5.2" is used to convert
from "Ib/ft2" to Hin.H2CP.)
4.6.4 Evaluation of Permit Application
Compare the results from the calculations and the values supplied by the
permit applicant using Table 4.6-1. The calculated values in the table are
based on the Example Case. If the calculated values of L •,, D n ,
gal column
Htco1umn> Httotal• APtotaT Wtcolumn> and packing are different from the
reported values for these variables, the differences may be due to the
assumptions involved in the calculations. Therefore, the reviewer may wish to
discuss the details of the proposed design with the permit applicant.
4.6-18
-------
TABLE 4.6-1. COMPARISON OF CALCULATED VALUES AND VALUES
SUPPLIED BY THE PERMIT APPLICANT FOR
ABSORPTION
Calculated Value Reported Value
(Example Case)
Solvent flow rate, L •, 35 gal/min
Column diameter, D i 4 ft.
Column height, HtCQlumn 23 ft.
Total column height, Htt t -, 26 ft
Packing volume, Vpacking 290 ft3
Pressure drop, Aptota] 12 in HgO
Column weight, WtCQlumn 5,600 Ib
aBased on Emission Stream 5.
4.6-19
-------
If the calculated values agree with the reported values, then the design
and operation of the proposed scrubber system may be considered appropriate
based on .the assumptions made in this manual.
4.6-20
-------
4.6.5 References for Section 4.6
1. Chemical Engineer's Handbook. Perry, R.H. and C.H. Hilton (eds.)
Fifth edition. McGraw-Hill Book Company. New York, NY 1973.
2. Treybal, R.E. Mass-Transfer Operations. Third edition. McGraw-Hill
Book Company. New York, NY 1980.
3. Buonicore, A.J. and L. Theodore. Industrial Control Equipment for Gaseous
Pollutants. Volume I. CRC Press, Inc. Cleveland, OH. 1975.
4. Kohl, A. and F. Riesenfeld. Gas Purification. Second edition.
Gulf Publishing Company. Houston, TX. 1974.
5. U. S. EPA. Wet Scrubber System Study. Volume I: Scrubber Handbook.
EPA-R2-72-118a. August 1972.
6. U. S. EPA. Organic Chemical Manufacturing. Volume 5: Adsorption.
Condensation, and Absorption Devices. EPA-450/3-80-027. December 1980.
7. Vatavuk, W.M. and R.8. Neveril. Part XIII. Costs of Gas Absorbers.
Chemical Engineering. October 4, 1982. pp. 135-136.
4.6-21
-------
4.7 CONDENSATION
Condensation is a separation technique in which one or more volatile
components of a vapor mixture are separated from the remaining vapors through
saturation followed by a phase change (see Figure 4.7-1). The phase change
from gas to liquid can be accomplished in two ways: (a) the system pressure
may be increased, at a given temperature, or (b) the system temperature may be
reduced at constant pressure.
The design and operation of a condenser is significantly affected by the
number and nature of the components present in the emission stream. In a two-
component vapor system where one of the components is noncondensible (e.g.,
air), condensation occurs at dew point (saturation) when the partial pressure
of the condensible compound (e.g., benzene) is equal to its vapor pressure.
In most HAP control applications, the emission stream will contain large
quantities of noncondensible and small quantities of condensible compounds.
To separate the condensible component from the gas stream at a fixed pressure,
the temperature of the gas stream must be reduced. The more volatile a
compound (i.e., the lower the normal boiling point), the larger the amount
that can remain as vapor at a given temperature; hence the lower the
temperature required for saturation (condensation).
When condensers are used to control emissions, they are usually operated
at the constant pressure of the emission source, which is normally close to
atmospheric. Depending on the temperatures required for condensation, a
refrigeration unit may be necessary to supply the coolant (see Section
4.7.3.2). The two most common types of condensers used are surface and
contact condensers. Surface condensers are usually shell-and-tube heat
exchangers. The coolant typically flows through the tubes and the vapors
condense on the shell outside the tubes. The condensed vapor forms a film on
the cool tubes and is drained to a collection tank for storage or disposal.
In contrast to surface condensers where the coolant does not contact either
the vapors or the condensate, in contact condensers, the vapor mixture is
cooled by spraying a cool liquid directly into the gas stream.
4.7-1
-------
T3
a
o
o
u
ITJ
o
s-
o
CT)
03
O
E
O E 4-»
•r-
to O) r—
C/> i. C
•r— -M l—t
£00
-------
Design calculations for condenser systems vary in complexity depending on
the nature and number of components present in the emission stream. For
detailed .information on condenser design, consult References 1 and 2. In the
following discussion, Emission Stream 6 consisting of a single condensible
component and a single noncondensable component will be used to illustrate the
calculation procedure for surface condensers. It will be assumed that the
moisture content of the emission stream is negligible (i.e., no ice is
expected to form on the tubes in the condenser). The design procedure will
involve determining the condensation temperature required, selection of
coolant, and calculation of condenser size and coolant requirements.
4.7.1. Data Required
The data necessary to perform the calculations consist of HAP emission
stream characteristics previously compiled on the "HAP Emission Stream Data
Form" and the required HAP control as determined by the applicable regulations.
EXAMPLE CASE
1. Maximum flow rate, Q » 2,000 scfm
2. Temperature, T - 90 F
6
3. HAP - styrene
4. HAP concentration, HAP - 13,000 ppmv
(corresponding to saturation conditions)
5. Moisture content, M_ - negligible
6
6. Pressure, P - 760 mm Hg
6
Based on the control requirements for the emission stream:
Required removal efficiency, RE » 90%
4.7-3
-------
In the case of a permit review for a condenser, the following data should
be supplied by the applicant. The calculations in this section will then be
used to check the applicant's values.
Condenser system variables at standard conditions (70°F, 1 atm):
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Reported removal efficiency, REreported
Emission stream flow rate, Q
Temperature of emission stream, T
HAP «
scfm
HAP concentration, HAP
ppmv
CQn
Moisture content, Ma - _ %
s — — — —
Temperature of condensation, T
Coolant used = _
Inlet temperature of coolant, T ,
Coolant flow rate, Qcoolant •
Refrigeration capacity, Ref »
Condenser surface area, A
Ib/hr
tons
ft2
4.7.2 Pretreatment of the Emission Stream
If water vapor is present in the emission stream, ice may form on the
condenser tubes when coolants such as chilled water or brine solutions are
used, decreasing the heat transfer efficiency and thus lowering the
condenser's removal efficiency. In such cases, dehumidifying the emission
stream is necessary. This can be carried out in a heat exchanger prior to the
condenser.
EXAMPLE CASE
Since the moisture content of
the emission stream is negligible,
no pretreatment is necessary.
4.7-4
-------
4.7.3 Condenser System Design Variables
The key design variable in condenser system design is the required
condensation temperature for a given removal efficiency or outlet
concentration. A condenser's removal efficiency depends on the nature and
concentration of emission stream components. For example, compounds with high
boiling points (i.e., low volatility) condense more readily compared to those
with low boiling points. Assume, as a conservative starting point, that
condensation will be considered as a HAP emission control technique for VOC's
with boiling points above 100°F.
The temperature necessary to achieve a given removal efficiency (or
outlet concentration) depends on the vapor pressure of the HAP in question at
the vapor/liquid equilibrium. Once the removal efficiency for a given HAP is
specified, the required temperature for condensation can be determined from
data on its vapor pressure-temperature relationship. Vapor pressure-
temperature data can be represented graphically (Cox charts) as shown in
Figure 4.7-2 for typical VOC's. The coolant selection is then .based on the
condensation temperature required. See Table 4.7-1 for a summary of practical
limits for coolant selection.
In a permit evaluation, use Table 4.7-1 to determine if the values
reported for the condensation temperature (T ) and the type of coolant
selected are consistent. Also, check if the coolant inlet temperature is
based on a reasonable approach temperature (a conservative value of 15°F is
used in the table). If the reported values are appropriate, proceed with the
calculations. Otherwise, the applicant's design is considered unacceptable.
The reviewer may then wish to use the guidelines in Table 4.7-1 and follow the
calculation procedure outlined below.
4.7.4 Determining Condenser System Design Variables
The condenser system evaluated in this manual consists of a shell and
tube heat exchanger with the hot fluid (emission stream) in the shell side and
4.7-5
-------
(1) Styrene
(2) Toluene
(3) Ethylene dichloride
(4)
0.1
0.00125 0.00150 0.00175
0.00200
[1/(T
CQn
0.00225 0.00250
460)]
0.00275 0.0030
Figure 4.7-2. Vapor pressure-temperature relationship,
4.7-6
-------
TABLE 4.7-1 COOLANT SELECTION
Required
Condensation
Temperature
T (°F)a
fnn \ '/
con
Coolant
Coolant
Temperature u
cool,i ' '
Comments
Tcon: 60 " 80 Water
60 > TCQn > 45 Chilled water
45 > T > -30 Brine solutions (e.g.,
calcium chloride,
ethylene glycol)
- -90 T .. < -30 Chlorofluorocarbons
con
(e.g., Freon-12)
Tcon-15
Tcon-15
Tcon'15
System efficiency
limited by seasonal
conditions.
Simple chilling
system needed.
Refrigeration systems
needed; potential
corrosion problems.
Refrigeration system
needed; potential
corrosion problems.
u Also emission stream outlet temperature.
„ Assume the approach as 15 F.
Summer limit.
4.7-7
-------
the cold fluid (coolant) in the tube side. The emission stream is assumed to
consist of a two-component mixture: one condensable component (HAP) and one
noncondensable component (air). Typically, condensation for such a system
occurs nonisothermally. To simplify the calculations, it is assumed that
condensation occurs isothermally.
4.7.4.1 Estimating Condensation Temperature--
In the following calculations, it is assumed that the emission stream
entering the condenser consists of air saturated with the HAP in question.
Calculations for cases involving mixtures of HAP's and supersaturated streams
are quite complex and will not be treated here; for additional information,
consult References 1 and 2.
For a given removal efficiency, the first step in the calculation
procedure is to determine the concentration at the outlet of the condenser.
Use the following expression:
partial = 760{(1-0.01 RE)/[(1 - (RExKr8xHAPe)]}HAPe x 10'6 (4.7-1)
where PDartja] is the partial pressure (mm Hg) of the HAP in the exit stream
assuming the pressure in the condenser is constant and at atmospheric. At
equilibrium between the gas and liquid phases, the partial pressure of the HAP
is equal to its vapor pressure at that temperature. Therefore, by determining
the temperature at which this condition occurs, the condensation temperature
(T ) can be specified. To carry out this calculation, vapor pressure-
temperature data for the specific HAP are required (see Figure 4.7-2). Such
data can be obtained from References 3 and 4.
4.7-8
-------
EXAMPLE CASE
Using Equation 4.7-1 and Figure 4.7-2;
HAPe - 13
RE - 90%
HAPg - 13,000 ppmv (styrene)
Ppartial - 760{[l-(0.01x90)J/[(l-(90xlO~8xl3,000)]}13,000 x 10'6
partial - 1-° « Hg
For styrene, the value of [l/(TC(jn + 460)] corres-
ponding to 1.0 mm Hg in Figure 4.7-2 is about 0.00208.
Solving for T :
T onOr
Tcon ' 20
4.7.4.2 Selecting the Coolant--
The next step is to select the coolant based on the condensation
temperature required. Use Table 4.7-1 to specify the coolant type. For
additional information on coolants and their properties, see References 3 and
4.
EXAMPLE CASE
Based on TCQn - 20°F, the appropriate
coolant is a brine solution. Assume
the brine solution is a 29% (wt) calcium
chloride solution which can be cooled
down to -40°F (see Reference 3).
4.7.4.3 Condenser Heat Load--
Condenser heat load is defined as the quantity of heat that must be
extracted from the emission stream to achieve a certain level of removal. It
is determined from an energy balance, taking into account the heat of
condensation of the HAP, sensible heat change of the HAP, and the sensible
heat change in the emission stream. The calculation steps are outlined below:
4.7-9
-------
a. 1. Calculate moles of HAP in the. inlet emission stream
(Basis: 1 min):
HAPg n - (Qe/387) HAPe x 10"6 (4.7-2)
The factor "387" is the volume (ft ) occupied by 1 Ib-mole of
ideal gas at standard conditions (70°F and 1 atm).
2. Calculate moles of HAP remaining in the outlet emission stream
(Basis: 1 min):
HAP0)m
where Pvapor 1s e(^ual to partial'
3. Calculate moles of HAP condensed (Basis: 1 min)
HAPcon - HAPe,m - HAPo,m <4'7-4>
b. 1. Determine the HAP's heat of vaporization (AH): Typically
the heat of vaporization will vary with temperature. Using
vapor pressure-temperature data as shown in Figure 4.7-2,
AH can be estimated by linear regression for the vapor
pressure and temperature range of interest. See Appendix
B.10 for details.
2. Calculate the enthalpy change associated with the condensed HAP
(Basis: 1 min):
"con ' "WconN"1 + C"HAP 'Te ' Tcon>] I4'7'5'
where Cp is the average specific heat of the HAP for the
CQn
temperature interval T - T (Btu/lb-mole-°F).
4.7-10
-------
3. Calculate the enthalpy change associated with the uncondensed
HAP (Basis: 1 m1rr>:
Huncon - HAPo,m %AP ^'7'^
4. Calculate the enthalpy change associated with the noncondensable
vapors (i.e., air) (Basis: 1 min):
Hnoncon ' «V387J ' HAPe,m^ ^air (Te ' Tcon^ (4'7'7)
where Cpa- is the average specific heat of air for the
temperature interval Tcon - Tg (Btu/lb-mole-°F).
c. 1. Calculate the condenser heat load (Btu/hr) by combining Equations
4.7-5, 6, and 7:
"load • '•' * 60 ("con * "uncon * "noncon' <4'7-8'
The factor "1.1" is included as a safety factor.
4.7-11
-------
EXAMPLE CASE
Using Equations 4.7-2 to 8;
a. 1. Qe - 2,000 scfm
HAPQ - (2,000/387) 13,000 x 10
HAPe'm " °-06718 lb-moles/min
Pvapor - l-° "" "9
Pg = 760 mm Hg
HAPQ m - 5.1008 [1.0/(760 - 1.0)]
HAPQ'm - 0.00672 Ib-moles/min
3. HAPcon = 0.06718-0.00672 » 0.0605 Ib-moles/min
b. 1. AH - 17,445 Btu/lb-mole
See Appendix B.10 for details.
2. MWHAp - 104.2 Ib/lb-mole
24 Btu/lb-mole-°F
(extrapolated from data in Reference 4)
Hcon " °-0605tl7,445 + 24(90-20)]
Hcon " 1)157 Btu/min
3. HunCQn = 0.00672 x 24 x (90 - 20)
Huncon - »-3 Btu/min
4. Cpa. = 6.96 Btu/lb-mole-°F (Table B.4-1)
a I r
Hnoncon = [(2,000/387)-0.06718] 6.96 x (90-20)
Hnoncon ' 2'485 Btu/m1n
c. H1(jad = 1.1 x 60 (1,157 + 11.3 + 2,485)
Hload = 241'100 Btu/hr
4.7-12
-------
4.7.4.4. Condenser Size--
Condenser systems are typically sized based on the total heat load and
the overall heat transfer coefficient estimated from individual heat transfer
coefficients of the gas stream and the coolant. An accurate estimate of
individual coefficients can be made using physical/chemical property data for
the gas stream, the coolant, and the specific shell and tube system to be
used. Since this level of detail is not appropriate for the manual, the value
used for the overall heat transfer coefficient is a conservative estimate.
For additional information on how to calculate individual heat transfer
coefficients, consult Reference 1.
To size condensers, use the following equation to determine the required
heat transfer area:
Acon ~ H/UAT (4.7-9)
where:
2
AC(Jn - condenser (heat exchanger) surface area, ft
U = overall heat transfer coefficient, Btu/hr-ft2-°F
ATyj = logarithmic mean temperature difference, °F
In calculating AT, „» refer to Appendix B.10. Assume that the approach
temperature at the condenser exit is 15°F. In other words, T , . =
(T - 15). Also, the temperature rise of the coolant fluid is specified as
25 F' 1-e-» Tcool,o = (Tcool,o + 25) where Tcool,o is the coolant exit
temperature. In estimating A..,,, the overall heat transfer coefficient can be
9 n
conservatively assumed as 20 Btu/hr-ft - F; the actual value will depend on
the specific system under consideration. This is based on Reference 2 in
which guidelines on typical overall heat transfer coefficients for condensing
vapor-liquid media are reported.
4.7-13
-------
EXAMPLE CASE
Using Equation 4.7-9:
(Refer to Appendix B.10 for calculating T,M)
Te . 90°F
con
Tcoo1,o-Tcool,i+25-30F
AT
AT
LM
LM
f (90 - 30) - (20 - 5)
[In [(90 - 30)/(20 - 5)]
32°F
"load ' 241>1DO ?tu/hr
U = 20 Btu/hr-ftz-°F
x 32)
con
Acon ' 375
4.7.4.5 Coolant Flow Rate--
The quantity of heat extracted from the emission stream is transferred to
the coolant. By a simple energy balance, the flow rate of the coolant can be
calculated as follows:
Qcoolant ' Hload/^coolant .lb/hr
-pcoolant * avera9e specific heat of the coolant over the temperature
1ntervalTcool,i t0Wo> Btu/lb-°F.
Specific heat data for coolants are available in References 3 and 4.
4.7-14
-------
EXAMPLE CASE
Using Equation 4.7-10:
"load * 241'100 Btu/hr
Tcool,ia5°F
Tcool,o-30F
^coolant " °'65 Btu/lb-°F (Reference 3)
Qcoolant " 241,100/[0.65 (35-10)]
'coolant ' 14'840 1b/hr
4.7.4.6 Refrigeration Capacity--
A refrigeration unit is assumed to supply the coolant at the required
temperature to the condenser. For costing purposes, the required
refrigeration capacity is expressed in terms of refrigeration tons as follows:
Ref ' Hload/12'000
where Ref is the refrigeration capacity, tons.
(4.7-11)
EXAMPLE CASE
Using Equation 4.7-11:
Hload * 241»100 Btu/hr
Ref - 241,100/12,000
Ref « 20 tons
4.7.4.7 Recovered Product--
To calculate costs, the quantity of recovered product that can be sold
and/or recycled to the process must be determined. Use the following
equation:
Qrec - 60 x HAPCQn x MWHAp (4.7-12)
where Q is the quantity of product recovered, Ib/hr.
4,7-15
-------
EXAMPLE CASE
Using Equation 4.7-12:
HAPcon = 0.0605 Ib-moles/min
MW,
HAP
Vec
• 104.2 Ib/lb-mole
60 x 0.06065 x 104.2
378 Ib/hr
4.7.5 Evaluation of Permit Application
Compare the results from the calculations and the values supplied by the
permit applicant using Table 4.7-2. The calculated values in the table are
based on the Example Case. If the calculated values of TCQn, coolant type,
' ^
coolant'
' an(* ^rec are different ^rom tne reported values for these
variables, the differences may be due to the assumptions involved in the
calculations. Therefore, the reviewer may wish to discuss the details of the
proposed design with the permit applicant.
If the calculated values agree with the reported values, then the design
and operation of the proposed condenser system may be considered appropriate
based on the assumptions made in this manual.
4.7-16
-------
TABLE 4.7-2 COMPARISON OF CALCULATED VALUES AND
VALUES SUPPLIED BY THE PERMIT APPLICANT
FOR CONDENSATION
aBased on Emission Stream 6.
Calculated Value Reported Value
(Example Case)
Condensation temperature, TCQn 20°F
Coolant type Brine solution
Coolant flow rate, Qcoolant 14,840 Ib/hr
Condenser surface area, A 375 ft
Refrigeration capacity, Ref 20 tons
Recovered product, Q 378 Ib/hr
rsc
4.7-17
-------
4.7.6 References for Section 4.7
1. Kern, D.Q. Process Heat Transfer. McGraw-Hill Book Company, Inc. and
Kogakusha Company, Ltd. 1950.
2. Ludwig, E.E. Volume III. Applied Process Design for Chemical and
Petrochemical Plants. Gulf Publishing Company. Houston, TX. 1965.
3. Chemical Engineer's Handbook. Perry, R.H. and C.H. Hilton (eds.). Fifth
edition. McGraw-Hill Book Company. New York, NY. 1973.
4. Lanoe's Handbook of Chemistry. Dean, J.A. (ed.). Twelfth edition.
McGraw-Hill Book Company. New York, NY. 1979.
4.7-18
-------
4.8 FABRIC FILTERS
Fabric filter collectors (also known as baghouses) are one of
the most efficient means of separating participate matter from a gas
stream. Fabric filters are capable of maintaining mass collection effi-
ciencies of greater than 99 percent down to a particle size approaching
Q.3(j.m in most applications. 1»2» 3 7n-jS efficiency is largely insensitive
to the physical characteristics of the gas and dust, and, depending on
fabric cleaning method, to the inlet dust loading.^4 Physical limita-
tions of the fabric materials to the temperature, moisture content, and
corrosivity of the gas stream reduce the applicability of fabric
filters. Variables considered in baghouse design include fabric
type, cleaning method, air-to-cloth ratio, and equipment configuration.
The filter fabric, cleaning method, and air-to-cloth ratio all
should be selected concurrently; choice of these parameters is mutually
dependent.1 Equipment configuration is of secondary importance unless
site-specific space limitations exist that require configuration to
be of primary importance in the fabric filter design.
Fabric filter systems typically are designed on the basis of
empirical information obtained through testing and long-term actual
operating experience for similar combinations of cleaning method,
fabric type, and dust rather than by analytical methods,* Although
theoretical equations exist to predict the performance of filtering
systems under various conditions, these equations are not very useful
as design tools. Therefore, discussion of baghouse design in this
section provides qualitative guidance rather than predictive equations.
Generally, fabric filter design for HAP's is no different than fabric
filter design for control of any other type of particulate matter.
However, due to the hazards associated with HAP's, greater care must
be taken to ensure that control is consistently of high efficiency
and that the control device is leak-proof, thus preventing accidental
release of the gas stream and captured pollutants. For these reasons,
design of a fabric filter for HAP's should only consider selected
fabric cleaning methods, and the design should specify an induced draft
fan (i.e., a negative pressure or suction baghouse) rather than a
4.8-1
-------
forced draft fan (i.e., a positive pressure baghouse). Information
presented in this section can be used to provide guidance for or to eva-
luate the appropriateness of baghouse design for certain HAP applications.
Appendix C.9 provides a worksheet-to record the information obtained
during the performance of the fabric filter design procedures.
4.8.1 Data Required
The data necessary to perform the design steps consist of the HAP
emission stream characteristic previously compiled on the "HAP Emission
Stream Data Form" and the required HAP control as determined by the
applicable regulations.
EXAMPLE CASE
Fabric filtration was one of the selected control
techniques for the municipal incinerator. The pertinent
data for these procedures are found on the "HAP Emission
Stream Data Form" (see Figure 3-8, page 3-43).
1. Flow rate, Qe>a = 110,000 acfm
2. Moisture content, Me = 5% vol.
3. Temperature, Te = 400°F
4. Particle Mean dia. = 1.0 ^m
5. $03 content = 200 ppm (vol)
6. Particulate Content = 3.2 grains/scf - flyash
7. HAP Content = 10% (mass) cadmium
In the case of a permit review for a fabric filter, the following
data should be supplied by the applicant. The design criteria and
considerations discussed in this section will be used to evaluate the
reasonableness of the applicant's proposed design.
1. Filter fabric material '
2. Cleaning method
3. Air-to-cloth ratio ft/min
4. Baghouse configuration
4.8-2
-------
4.8.2 Pretreajtment of the Emission Stream
As discussed In Section. 3.2.1, the temperature of the emission
stream should be within 50 to 100°F above the stream dew point.
Procedures for determining the dew point of an emission stream are
provided in Appendix B.2. If the emission stream temperature does not
fall within the stated range, pretreatment (i.e., emission stream
preheating or cooling) is necessary. Methods of pretreatment are briefly
discussed in Appendix 8.11. If pretreatment is performed,'the emission
stream characteristics will be altered. The primary characteristics
affecting baghouse design are emission stream temperature and flow
rate. Therefore, after selecting a temperature for the emission stream,
the new stream flow rate must be calculated. The calculation method
depends upon the type of pretreatment performed; use approropriate
standard industrial equations. The use of pretreatment mechanical dust
collectors may also be appropriate. If the emission stream contains an
appreciable amount of large particles (20 to 20jum), pretreatment with
mechanical dust collectors is typically performed. Appendix B.ll
further describes the use of mechanical dust collectors.
4.8.3 Fabric Filter System Design Variables
Successful design of a fabric filter depends on the proper selec-
tion of fabric and cleaning method and on an adequate air-to-cloth
ratio. All fabric filter systems share the same basic features and
operate using the principle of aerodynamic capture of particles by
fibers. Systems vary, however, in certain key details of construction
and in the operating parameters. The design variables of particular
interest are filter bag material, fabric cleaning method, air-to-cloth
ratio, baghouse configuration, and materials of construction.
As stated earlier, the first three variables should be considered
concurrently. The configuration and construction materials are important,
but secondary, considerations. The following subsections discuss
step-by-step procedures for selecting each of these design variables
as they may apply to a specific particulate HAP control situation.
4.8-3
-------
Because HAP control is similar to participate control in general, a
good verification of these procedures can be accomplished by consulting
the section about the particular industry in a document entitled "Control
Techniques for Particulate Emissions from Stationary Sources - Volume II,'
or in the Mcllvaine Fabric Filter Manual.4»5 [Note: Because these
design variables are considered concurrently, the "Example Case" is
presented at the end of Section 4.8.3.]
4.8.3.1 Fabric Type --
Several types of natural and synthetic fabric are used in baghouse
systems. Gas stream characteristics, such as temperature, acidity,
alkalinity, and particulate matter properties (e.g., abrasiveness and
hygroscoposity), determine the fabric type to be used.l»6 in nany
instances, several fabric types will be appropriate, and a final selec-
tion will be chosen only when cleaning method and the desired air-
to-cloth ratio are considered.
Most of the principal synthetic fibers have been adapted for use
as filtering fabrics while the only natural fibers in common use are
cotton and wool. Some of the more common synthetic fibers in commer-
cial use are nylon (aromatic and polyamide), acrylic, polyester,
polypropylene, fiberglass, and fluorocarbon. Natural fibers can be
used for gas temperatures up to 200°F and have only moderate resistance
to acids and alkalis contained in the gas stream.1»4»6 Synthetics can
operate at temperatures up to 550°F and generally have greater chemical
resistance.I»4f6 Therefore, while the initial cost of the synthetic
filter fabric is greater than the cost for natural fibers, the increased
service life and improved operating characteristics of the synthetics
make them a preferred choice in a wide range of industrial situations.
Almost all of the filter fabrics can be constructed in either a
woven or a felted manner (cotton and fiberglass can be constructed
in a woven manner only). Woven fabrics are made up of yarn in one of a
variety of patterns that allow spaces between the fibers, whereas felted
fabrics are composed of a thick mat of randomly oriented fibers. When
woven fabrics are new, particles penetrate the pores of the fabric
fairly easily. As filtering continues, however, more particles are
retained on the filter threads and on the particles already collected.
4.8-4
-------
As this dust layer or "cake" builds up, particle penetration drops to
a very low level. Cleaning of woven fabrics must be performed so that
a layer of this dust cake remains on the fabric, enabling particle
penetration to remain low.1*5 Felted fabrics are thick enough that a
dust cake does not need to remain on the fabric in order to maintain a
good collection efficiency.1»4 This difference between woven and
felted fabrics has important implications for selection of fabric
cleaning method, as described in Section 4.8.3.2. Felted fabrics are
more expensive than woven fabrics.
Table 4.8-1 presents information on the maximum continuous operating
temperature and resistance characteristics of commonly used filter
fabrics. Knowing the emission stream characteristics, Table 4.8-1 can
be used to select an appropriate fabric filter type (or types). Although
the information presented is qualitative, Table 4.8-1 provides a good
basis either for selecting a fabric or for evaluating the appropriateness
of a fabric in a permit application.
When a number of fabrics are suitable for an application, the
relative cost of the fabrics may be the key decision criterion. In
general, fluorocarbon and nylon aromatic bags are the most expensive,
followed by wool and fiberglass. The remaining commonly used synthetics
are generally less expensive than fiberglass (polypropylene, polyester,
acrylic, nylon polyamide, and modacrylic), while cotton is generally
the least expensive fabric.2.3,4,7
4.8.3.2 Cleaning Method --
As dust accumulates on the filtering elements, the pressure drop
across the bag compartment increases until cleaning of the bags occurs.
A timer can be used to control the cleaning cycle or pressure drop can
be monitored so that cleaning occurs when some maximum desirable value
is reached. At this point the bags in the compartment are cleaned to
remove the collected dust and the cycle is then repeated. The two
basic mechanisms used to accomplish bag cleaning are flexing of the
fabric to break up and dislodge the dust cake, and reversed air flow
through the fabric to remove the dust.* These may be used separately
or in conjunction with one another. The three principal methods used
to accomplish fabric cleaning are mechanical shaking (manual or
4.8-5
-------
TABLE 4.8-1. CHARACTERISTICS OF SEVERAL FIBERS USED
IN FABRIC FILTRATION^
Max. Continuous
Fiber Operating
Typeb Temp., °F Abrasion
Cottond
Wool6
Modacrylic6
(Dynel)
Polypropylene8
Nylon Polyamide6
(Nylon 6 & 66)
Acrylic6
(Orion)
Polyester6
(Dacron, Creslan)
Nylon Aromatic6
(Nomex)
Fl uorocarbon6
(Teflon, TFE)
Fibergl assd
180
200
175
200
220
260
275
450
500
550
G
F/G
F
E
E'
G
E
E
F/G
P,Gh
Mineral
Acids
P
F
VG
E
P
G
G
F
£9
VG
Resistance*-
Organic
Acids Alkalies
G
F
VG
E
F
G
G
G
£9
E
G
P/F
G
E
VG
F
G
VG
E9
P
Solvent
E
G
G
G
E
E
E
£
£9
E
References 1, 2, 4, 6, 8, and 9. Where data differed, a representative
category was chosen.
bRepresents the major categories of filtration fibers. Names in
parentheses indicate some principal trade names.
CP = poor resistance, F = fair resistance, G = good resistance, VG = very
good resistance, and E = excellent resistance.
dWoven fabrics only.
6Woven or felted fabrics.
fConsidered to surpass all other fibers in abrasion resistance.
9The most chemically resistant of all these fibers.
hAfter treatment with a lubricant coating.
4.8-6
-------
automatic), reverse air flow, and pulse-jet cleaning. The first
method uses only the fabric flexing mechanism; the latter two methods
use a combination of the reverse air flow and fabric flexing mechanisms.
Selection of a cleaning method is based on the type of fabric
used, the pollutant collected, and the manufacturer's, vendor's, and
industry's experiences. A poor combination of filter fabric and
cleaning method can cause premature failure of the fabric, incomplete
cleaning, or blinding of the fabric.1 Blinding of a filter fabric
occurs when the fabric pores are blocked and effective cleaning can not
occur. Blinding can result because moisture blocks the pores or
increases the adhesion of the dust, or because a high velocity gas
stream imbeds the particles too deeply in the fabric.1 The selection
of a cleaning method may be based on cost, especially where more than
one method is applicable. Table 4.8-2 contains a comparison of cleaning
methods. Cleaning methods are discussed individually below.
With mechanical shaking, bags are hung on an oscillating framework
that periodically shakes the bags at timed intervals or at a predefined
pressure drop level.1»3»6 The shaker mechanisms produce a violent action
on the fabric filter bags and, in general, produce more fabric wear
than the other types of cleaning mechanisms.3 For this reason, mechan-
ical shaking is used in conjunction with heavier and more durable
fabric materials, such as most woven fibers.3»10 Bags with poor or fair
abrasion ratings in Table 4.8-1 (such as fiberglass) should not be chosen for
fabric filters cleaned by mechanical shaking unless they are treated
with a special coating before use. Although shaking is abrasive to the
fabric, it does allow a dust cake to remain on the fabric, thus
maintaining a high collection efficiency.
Bags are usually taken off-line for cleaning by mechanical shaking
so that no gas flows through the bags being cleaned. Thus, reentrainment
of particles is minimized. Because dust dislodgement is not severe
(i.e., a light dust cake remains on the fabric), and because cleaning
occurs off-line, outlet concentrations are almost constant with varying
inlet dust loading and through entire cleaning cycles when using
mechanical shaking.! Further control efficiency is very high, and, in
fact, properly selected woven fabrics cleaned by mechanical shaking can
provide much greater particle collection than pulse-jet cleaned felted
fabrics in many applications.4 For these reasons, mechanical shaking
4.8-7
-------
TABLE 4.8-2. COMPARISONS OF FABRIC FILTER BAG CLEANING METHODS3
Cleaning Method
Mechanical
Parameter Shake
Cleaning on- or
off-line
Cleaning time
Cleaning uniformity
Bag attrition
Equipment ruggedness
Fabric type
Filter velocity
Power cost
Oust loading
Maximum temperature15
Collection efficiency
Off-line
High
Average
Average
Average
Woven
Average
Low
Average
High
Good
Reverse
Airflow
Off-line
High
Good
Low
Good
Woven
Average
Low to
Medium
Average
High
Good
Pulse-jet Pulse-jet
Individual Compartmented
Bags Bags
On-line
Low
Average
Average
Good
Felt
High
High
Very high
Medium
Lower
Off-line
Low
Good
Low
Good
Felt
High
Medium
High
Medium
Lower
aSource: Reference 4.
^Fabric limited.
4.8-8
-------
is a good method to clean fabric filters controlling emissions containing
HAP's.4
Reverse air flow clean-ing is used to flex or collapse the filter
bags by allowing a large volume of low pressure air to pass countercur-
rent to the direction of normal gas stream flow during filtration.3,6
Reverse air is provided either by a separate fan or by a vent in the fan
damper, which allows a backwash of air to clean the fabric filters.3»6
Reverse air flow cleaning usually occurs off-line. Reverse air
cleaning allows the use of fragile bags, such as fiberglass, or light-
weight bags, and usually results in longer life for the bags.3 As with
mechanical shaking, woven fabrics are used, and because cleaning is less
violent than with pulse-jet cleaning and occurs off-line, outlet
concentrations are almost constant with varying inlet dust loading and
throughout the cleaning cycle. Reverse air flow cleaning is, therefore,
a good choice for fabric cleaning in HAP control situations.
In pulse-jet cleaning, a high pressure air pulse is introduced into
the bag from the top through a compressed air jet.3»6 This rapidly
expands the bag, dislodging the particles. Thus, the fabric is cleaned
thoroughly through a vibration effect. The pulse of air cleans so
effectively that no dust cake remains on the fabric to contribute to
particulate collection. Because such a cake is essential for effective
collection on woven fabrics, felted fabrics are generally used in
pulse-jet cleaned fabric fliters.1 All of the fabric materials may be
used with pulse-jet cleaning except cotton or fiberglass.
Because the cleaning air pulse is of such high pressure (up to
100 psi) and short duration (£0.1 sec), cleaning is usually accomplished
on-line. Extra bags are not necessary, therefore, to compensate for
bags off-line during cleaning. Cleaning occurs more frequently than
with mechanical shaking or reverse air flow cleaning, which permits
higher air velocities (higher A/C ratios) than the other cleaning
methods. Further, because the bags move less during cleaning, they may
be packed more closely together. In combination, these features allow
pulse-jet cleaned fabric filters to be installed in a smaller space,
and thus, at a lower cost, than fabric filters cleaned by the other
methods.1»6 This cost savings may be somewhat counterbalanced by the
greater expense and more frequent replacement required of felted bags,
4.8-9
-------
the higher power use that may occur, and the installation of the fabric
filter framework that pulse-jet cleaning requires.^»^
Pulse-jet cleaning is not, however, recommended for HAP control
situations for several reasons. First, because cleaning occurs on-
line, the rapid high-pressure pulse generated during cleaning causes
increased emissions from the bags. Mass emissions can vary by as much
as 100 times over a filtration cycle.4 Second, although collection
efficiencies of pulse-jet cleaned fabric filters are in the 99.9 to
99.99 percent range, filtering efficiency of pulse-jet cleaned filters
is inferior to that of mechanically shaken or reverse air cleaned
filters that have a good cake buildup.4 In one study, average outlet
concentrations were two to three orders of magnitude higher for pulse-
jet cleaned filters than for mechanically shaken filters.4 Third,
emissions from pulse-jet systems are strongly dependent on the inlet
concentration; thus, the collection efficiency rather than the effluent
concentration tends to be relatively constant for fabric filters using
pulse-jet cleaning.1»4 For these reasons, outlet emission levels are
not as constant or as low when using 'pulse-jet cleaning as when using
either mechanical shaking or reverse air flow cleaning. Pulse-jet
cleaning is, therefore, not recommended for fabric filters used in HAP
control situations or for high inlet loadings involving fine particulate
matter.2
In cases of permit evaluation where pulse-jet cleaning is believed
to be adequate to meet specific regulations in specific applications,
several options are available to minimize the disadvantages of pulse-jet
cleaning. First, pulse-jet filter bags can be compartmentalized to
permit off-line cleaning; additional bags must be installed to allow
this.4 Second, reduced filtration velocity, or pulse intensity, will
decrease average outlet concentration.4 Third, bags should be flexible,
lightweight, and inelastic, with uniform pore structure, to obtain
maximum particle collection.4 These changes, in effect, alter the
typical pulse-jet baghouse such that it behaves (i.e., cleans the bags)
in a manner that is very similar to that of a reverse air baghouse.
4.8-10
-------
4.8.3.3 Air-to-cloth Ratio --
The air-to-cloth (A/C) ratio, or filtration velocity, is a tradi-
tional fabric filter design parameter defined as the actual volumetric
flow rate (acfm) divided by the total active, or net, fabric area
(ft^). The A/C ratio is an important indicator of the amount of air
that can be filtered in a given time when considering the dust to be
collected, cleaning method and fabric to be used, and the character-
istics of the gas stream to be filtered for an individual situation.
Selection of an appropriate range of A/C ratios is not based on any
theoretical or empirical relationship, but, rather, is based on
industry and fabric filter vendor experience from actual fabric filter
installations. A ratio is usually recommended for a specific dust and
a specific cleaning method. For typical design calculations, the A/C
ratio must be obtained from the literature or the manufacturer.
Table 4.8-3 summarizes the ranges of recommended A/C ratios by typical
bag cleaning method for many dusts and fumes. These ranges are meant
to serve as a guide; A/C ratios may vary from those reported. Fabric
filter size and cost will vary with A/C ratio; lower A/C ratios, for
example, will require that a larger and more expensive fabric filter
be installed. [Note: Pulse-jet cleaning is not recommended for HAP
control situations; the A/C ratio for control of streams containing
HAP's will, therefore, be fairly low.] In addition to evaluating a
particular fabric filter application, the A/C ratio and the emission
stream flow rate (Qeja) are usec' to calculate net cloth area (Anc):
A/C ratio
where: Qe a = emission stream flow rate at actual conditions acfm
A/C ratio = air-to-cloth ratio, acfm/ft^ or ft/min
o
Apc = net cloth area, ft
Net cloth area is the cloth area in active use at any point in time,
Gross cloth area (A^c)» by comparison, is the total cloth area contained
in a fabric filter, including that which is out of service at any point
in time for cleaning or maintenance. In this manual, costing of the
fabric filter structure uses net cloth area, while costing of fabric
4.8-11
-------
TABLE 4.8-3. RECOMMENDED AIR-TO-CLOTH RATIOS (FT/MIN) FOR VARIOUS
DUSTS AND FUMES BY CLEANING METHOD3
Dust
or Fume
Abrasives
Alumina
Al urn in urn
Aluminum Oxide
Asbestos
Bauxite
Blast Cleaning
Carbon
Carbon Black
Chrome
Coal
Coke
Dyes
Fertilizer
Fl i nt
Fly Ash
Foundry
Glass
Graphite
Gypsum
Iron Ore
Iron Oxide
Iron Sulfate
Lead Oxide
Leather
Lime
Limestone
Machining
Manganese
A/ C Ratios Recommended for Cleaning Method
Shaker Reverse Air Pulse-Jet
2.0
2.25
2.5
2.25
3.0
1.2
1.5
1.5
2.0
2.0
1.5
2.0
2.0
2.0
2.0
2.0
3.5
2.0
2.0
- 3.0 *b
- 3.0 *
3.0
2.0 *
- 4.0 *
- 3.2 *
- 3.5 *
- 2.5 *
- 2.5 1.1 - 1.5
- 2.5 *
- 3.0 *
2.5 *
2.0 *
- 3.5 1.8 - 2.0
2.5 *
2.0 2.1 - 2.3
* *
2.5 *
- 3.0 1.5 - 2.0
- 3.5 1.8 - 2.0
- 3.5 *
- 3.0 1.5 - 2.0
- 2.5 1.5 - 2.0
- 2.5 1.5 - 1.8
- 4.0 *
- 3.0 1.5 - 2.0 -
- 3.3 *
3.0 *
2.25 *
9
*
16
*
9-16
"8-10
*
5-7
8-12
9-12
12-16
9-12
10
8-10
*
9-10
8-12
*
7-9
10-16
11-12
8-16
6-8
6-9
15-20
10-16
8-12
16
*
(continued)
4.8-12
-------
TABLE 4.8-3. RECOMMENDED AIR-TO-CLOTH RATIOS (FT/MIN) FOR VARIOUS
DUSTS AND FUMES BY CLEANING METHOD3
(concluded)
A/C Ratios Recommended for Cleaning Method
Dust
or Fume Shaker Reverse Air Pulse-Jet
Metal Fumes 1.5 1.5-1.8
Metal Powders 2.0 *
Mica 2.25 - 3.3 1.8 - 2.0
Paint Pigments 2.0 *
Paper 3.5 - 4.0 *
Perchlorates * *
Plastics 2.0 - 3.0 *
Polyethylene * *
PVC * *
Resin 2.0 *
Silica 2.25 - 2.8 1.2 - 1.5
Silica Flour 2.0 - 2.5 *
Silicates * *
Silicon Carbide * *
Slate 2.5 - 4.0 *
Starch 2.25 *
Talc 2.25 *
6-9
9-10
9-11
*
10-12
10
7-10
10
7
8-10
7-12
*
9-10
10
12-14
*
*
a Source: References 1 and 9.
b * = no information available.
4.8-13
-------
filter bags uses gross cloth area. Table 4.8-4 presents factors to
obtain gross cloth area from net cloth area:
.Anc x Factor = A^c
where: Factor = value from Table 4.8-4, dimension less
Atc = gross cloth area, ft2
Fabric filters with a higher A/C ratio require fewer bags to
accomplish cleaning, and, therefore, require less space and may be less
expensive. Other costs, such as more expensive (felted) bags, bag
framework structure, use of increased pressure drop and corresponding
increased power requirements, etc., may counterbalance to some degree
the savings of high A/C ratio systems.
4.8.3.4 Baghouse Configuration —
The basic configuration of a baghouse varies according to whether
the gases are pushed through the system by a fan located on the up-
stream side (forced draft fan), or pulled through by locating the fan
on the downstream side (induced draft fan). A baghouse using forced
draft fans is called a positive-pressure baghouse; one using induced
draft fans is called a negative-pressure or suction baghouse.
Positive-pressure baghouses may be either open to the atmosphere or
closed (sealed and pressure-isolated from the atmosphere). Negative-
pressure baghouses can only be of the closed type. Only the closed
suction design should be selected for a HAP application to prevent
accidental release of captured pollutants.6 The higher the gas stream
dew point, the greater the precaution that must be taken to prevent
condensation, which can moisten the filter cake, plug the cloth, and
promote corrosion of the housing and hoppers. In a suction-type
fabric filter, infiltration of ambient air can occur, which can lower
the temperature below design levels. Therefore, the structure walls
and hoppers of this type of baghouse should be insulated to minimize
the possibility of condensation.
4.8-14
-------
TABLE 4.8-4. FACTORS TO OBTAIN GROSS CLOTH AREA
FROM NET CLOTH AREAa
Net Clotb Area, Anc Factor to Obtain
(ft') Gross Cloth Area, At
(ft2) t
1 - 4,000 Multiply by 2
4,001 - 12,000 Multiply by 1.5
12,001 - 24,000 Multiply by 1.25
24,001 - 36,000 Multiply by 1.17
36,001 - 48,000 Multiply by 1.125
48,001 - 60,000 Multiply by 1.11
60,001 - 72,000 Multiply by 1.10
72,001 - 84,000 Multiply by 1.09
84,001 - 96,000 Multiply by 1.08
96,001 - 108,000 Multiply by 1.07
108,001 - 132,000 Multiply by 1.06
132,001 - 180,000 Multiply by 1.05
180,001+ Multiply by 1.04
aSource: Reference 3.
4.8-15
-------
4.8.3.5 Materials of Construction --
The most common material used in fabric filter construction is
carbon steel. In cases where the gas stream contains high concentrations
of $03 or where liquid-gas contact areas are involved, stainless
steel may be required. Stainless steel will increase the cost of the
fabric filter significantly when compared to carbon steel.3 However,
by keeping the emission stream temperature above the dew point and by
insulating the baghouse, the use of stainless steel should not be
necessary.
EXAMPLE CASE
Table 4.8-1 indicates that filter fabrics that can
withstand the 400°F emission stream temperature are
nylon aromatic (Nomex), fluorocarbon (Teflon), and
fiberglass. Because there is a high potential for acid
damage (i.e., a high $03 content), however, Nomex bags
should not be considered. Because HAP's are present,
only mechanical shaking or reverse air flow cleaning
methods are advisable. Using Table 4.8-3 for fly ash
type dust, a low A/C ratio is expected for the two
acceptable cleaning methods (2 to 2.3 ft/min). Because
a fiberglass bag would provide the most protection during
temperature surges, and because fiberglass bags may be
less expensive, it may be the fabric of choice for an
installation with these emissions characteristics.
Fiberglass bags would require that reverse air cleaning
be used. Teflon bags with mechanical shaking could also
be a possibility. The documents that describe experience
in certain industry applications support the choice of
fiberglass bags with reverse air flow cleaning.^»5
4.8-16
-------
4.8.4- Evaluation of Permit Application
Using Table 4.8-5, compare the results from this section and the
data supplied by the permit applicant. The calculated values are based
on the example case. As pointed out in the discussion on fabric
filter design considerations, the basic design parameters are generally
selected without the involved, analytical approach that characterizes
many other control systems, such as an absorber system (Section 4.6).
Therefore, in evaluating the reasonableness of any system specifications
on a permit application, the reviewer's main task will be to examine
each parameter in terms of its compatibility with the gas stream and
particulate conditions and with the other selected parameters. The
following questions should be asked:
1. Is the temperature of the emission stream entering the
baghouse within 50 to 100°F above the stream dew point?
2. Is the selected fabric material compatible with the conditions
of the emission stream; that is, temperature and composition
(see Table 4.8-1)?
3. Is the baghouse cleaning method compatible with the selected
fabric material and its construction; that is, material type
and woven or felted construction (see" Section 4.8.3.2 and
Table 4.8-2)?
4. Will the selected cleaning mechanism provide the desired control?
5. Is the A/C ratio appropriate for the application; that is,
type of dust and cleaning method used (see Table 4.8-3)?
6. Are the values provided for the gas flow rate, A/C ratio,
and net cloth area consistent? The values can be checked with
the following equation:
Q
A/C ratio = e»a
^nc
where: A/C ratio = air-to-cloth ratio, ft/min
Qe a 3 emission stream flow rate at actual
conditions, acfm
Anc = net cloth area, ft2
7. Is the baghouse configuration appropriate; that is, is it a
negative-pressure baghouse?
4.8-17
-------
TABLE 4.8-5. COMPARISON OF CALCULATED VALUES AND VALUES
SUPPLIED BY THE PERMIT APPLICANT FOR FABRIC FILTERS
Calculated Value Reported Value
(Example Case)3
Emission Stream Temp. Rangeb 365 - 415°F . . .
Selected Fabric Material fiberglas or Teflon . . .
Baghouse Cleaning Method mechanical shaking or
reverse air flow . . .
A/C ratio _ Q-e,a 2 - 2.3 ft/min . . .
AHC
Baghouse Configuration negative pressure . . .
aBased on the municipal incinerator emission stream.
bSee Section 3.2.1.
A particular manufacturer/customer combination may employ some-
what different criteria in their selection of design parameters (such
as lower annualized costs of operation at the expense of higher initial
costs), and so a departure from the "rules-of-thumb" discussed here
may still be compatible with achieving the needed high collection
efficiencies. Further discussions with the permit applicant are
recommended to evaluate the design assumptions and to reconcile any
apparent discrepancies with usual practice.
4.8.5 Determination of Baghouse Operating Parameters
Many times, optimization of a fabric filter's collection efficiency
occurs in the field after construction. The following discussion
does not pertain to the preliminary design of a fabric filtration
control system; however, the information presented should be helpful
in achieving and maintaining the desired collection efficiency for
the installed control system.
4.8-18
-------
4.8.5.1 Collection Efficiency —
A wel1 designed fabric filter can achieve collection efficiencies
in excess of 99 percent, although optimal performance of a fabric filter
system may not occur for a number of cleaning cycles as the new filter
material is "broken in." The fabric filter collection efficiency is
related to the pressure drop across the system, component life, filter
fabric, cleaning method and frequency, and A/C ratio. These factors
should be reevaluated if fabric filter performance is less than
permitted. Modifications to improve performance include changing the
A/C ratio, using a different fabric, or replacing worn or leaking
filter bags. Collection efficiency can be improved by decreasing the
frequency of cleaning or allowing the system to operate over a greater
pressure drop before cleaning is initiated.
4.8.5.2 System Pressure Drop —
The pressure drop across the operating fabric filter system is a
function of the difficulty with which the gas stream passes through the
filter bags and accumulating dust cake, how heavy the dust deposit is
prior to bag cleaning, how efficient cleaning is, and if the filter
bags are plugged or blinded. Normally, the value of this parameter is
set at about 3 to 4 inches of water, although pressure drops in excess
of 10 inches have been used.2 in actual operation, variations in
pressure drop outside of the design range may be indicative of problems
within the fabric filter system. Higher than expected pressure differ-
entials may indicate: (1) an increase in gas stream volume; (2) blinding
of the filter fabric; (3) hoppers full of dust, thus blocking the bags;
and/or (4) inoperative cleaning mechanism. Lower than expected pressure
differentials may indicate: (1) fan or motor problems, (2) broken
or undamped bags, (3) plugged inlet ducting or closed damper, and/or
(4) leakage between sections of the baghouse.
As the dust cake builds up during filtration, bot the collection
efficiency and system pressure drop increase. As the pressure drop
increases toward a maximum, the filter bags (or at least a group of the
bags contained in one isolated compartment) must be cleaned to reduce
the dust cake resistance. This cleaning must be timed and performed so
as to accomplish the following: (1) to keep the pressure drop and thus
4.8-19
-------
operating costs, within reasonable limits; (2) to clean bags as gently
and/or infrequently as possible to minimize bag wear and to maximize
efficiency; and (3) to leave-a sufficient dust layer on the bags to
maintain filter efficiency and to keep the instantaneous A/C ratio
immediately after cleaning from reaching excessive levels.
In practice, these various considerations are balanced using engi-
neering judgment and field trial experience to optimize the total
system operation. Changes in process or in fabric condition through
fabric aging will cause a shift in the cleaning requirements of the
system. This shift may require more frequent manual adjustments to the
automatic control to achieve the minimum cleaning requirements.
4.8.6 References for Section 4.8
1. Siebert, P.C. Handbook on Fabric Filtration. IIT Research
Institute. Chicago, IL. April 1977. Chapters 1 and 2.
2. U.S. EPA. Handbook of Fabric Filter Technology, Volume I; Fabric
Filter Systems Study. APTD 0690. December 1970. 623 pp.
3. U.S. EPA. Capital and Operating Costs of Selected Air Pollution
Control Systems. EPA-450/5-80-002. December 1978. pp. 5-19
through 5-31.
4. The Fabric Filter Manual. The Mcllvaine Company. Northbrook, IL.
1975. Chapter III.
5. U.S. EPA. Control Techniques for Particulate Emissions from Stationary
Sources - Volume 2. EPA-450/3-81-005b. September 1982. 540 pp.
6. U.S. EPA. Control Techniques for Particulate Emissions from
Stationary Sources - Volume I. EPA-450/3-81-005a. September
1982.Section 4.4.
7. Strauss, W. Industrial Gas Cleaning, 2nd Edition. Pergamon Press,
Oxford, England. 1975. Chapter 8.
8. U.S. EPA. Procedures Manual for Fabric Filter Evaluation.
EPA-600/7-78-113. June 1978. 451 p. '.
9. U.S. EPA. Air Pollution Engineering Manual. AP-40. Hay 1973. 987 p.
10. U.S. EPA. Particulate Control Highlights: Research on Fabric
Filtration Technology.EPA-600/8-78-005d.June 1978.20 p.
4.8-20
-------
4.9 ELECTROSTATIC PRECIPITATORS
Electrostatic precipitators (ESP's) use an electrostatic field to
charge participate matter contained in the gas stream. The charged
particles then migrate to a grounded collecting surface. The collected
particles are dislodged from the collector surface periodically by
vibrating or rapping the collector surface, and subsequently collected
in a hopper at the bottom of the ESP.
There are two basic types of ESP's: single stage and two stage.1»2
In the single stage precipitator, which may be wet or dry, ionization
and collection are combined, whereas in the two stage precipitator,
ionization and collection are done in separate steps. Dry, single
stage ESP's are the most common. Wet electrostatic precipitators,
while not as common as dry ESP's, can be used to remove both solid and
gaseous pollutants.
The most important variable considered in the design of an ESP is
collection plate area; this assumes that the ESP is provided with an
optimum level of secondary voltage. Collection plate area is a function
of the desired collection efficiency, gas stream flow rate and particle
drift velocity.1.2,3,5 other design details to be estimated by the vendor
include (but are not limited to) expected secondary voltage and current,
electrical sections alignment in direction of gas flow. In this document,
an approximate method to size an ESP is given.
Particle drift velocity is a complicated function of particle
size, gas velocity, gas temperature, particle resistivity, particle
agglomerization, and the physical and chemical properties of the parti-
culate matter. The theoretical relationship of-the drift velocity to the
variables is discussed extensively in the literature.1»2»3,5 Unfortunately,
there are no empirical equations readily available to calculate drift
velocity directly from these variables. Therefore, in determining drift
velocity for a given emission stream, equipment vendors often rely upon
historical data for similar streams and data established from pilot
plant tests. Published information on drift velocity (based on design
data for actual installations to represent typical gas characteristics)
are available for several industrial emission, streams.2
4.9-1
-------
Appendix C.10 provides a worksheet to record the information
obtained during the performance of the ESP design procedures.
4.9.1 Data Required
The data necessary to perform the design steps consist of the HAP
emission stream data characteristics previously compiled on the "HAP
Emission Stream Data Forms" and the required HAP control as determined
by the applicable regulations.
EXAMPLE CASE
Electrostatic precipitation was one of the selected
control techniques for the municipal incinerator stream.
The pertinent data for these procedures are found on the "HAP
Emission Stream Data Form" (see Figure 3-8, page 3-43).
1. Flow rate,
= 110,000 acfm
2. Emission stream temperature, Te = 400°F
3. Particulate content
4. Moisture content, Me
5. HAP content
6. Drift velocity of particles,
7. Collection efficiency, CE
= 3.2 grains/scf - flyash
= 5% (vol)
= 10% (mass) cadmium
= 0.3 ft/s
= 99.9% mass
In the case of a permit review for an ESP, the following data
should be supplied by the applicant. The design criteria and considerations
discussed in this section will be used to evaluate the reasonableness
of the applicant's proposed design.
4.9-2
-------
1. Reported collection efficiency =
2. Reported drift velocity of particles = ft/sec
3. Reported collection-plate area-= ft^
4.9.2 Pretreatment of the Emission Stream
As discussed in Section 3.2.1, the temperature of the emission
streams should be within 50 to 100°F above the stream dew point. Proce-
dures for determining the dew point of an emission stream are provided
in Appendix B.2. If the emission stream temperature does not fall
within the stated range, pretreatment (i.e., emission stream preheat or
cooling) is necessary. Methods of pretreatment are briefly discussed in
Appendix B.ll. The primary characteristics affecting ESP sizing are
drift velocity of the particles and flow rate. Therefore, after selecting
a temperature for the emission stream, the new stream flow rate must be
calculated. The calculation method depends upon the type of pretreatment
performed; use appropriate standard industrial equations. The use of
pretreatment mechanical dust collectors may also be appropriate. If
the emission stream contains an appreciable amount of large particles
(20 to 30 ^m), pretreatment with mechanical dust collectors is typically
performed. Appendix B.ll further describes the use of mechanical dust
collectors.
4.9.3 ESP Design Variables
Estimating the collection plate area is the important aspect of
sizing an ESP. A secondary consideration is the material of construction.
4.9.3.1 Collection Plate Area --
Although precise specification of collection plate area is best
left to the vendor, an approximate collection plate area can be calcu-
lated using the available drift velocity value for the gas stream.
As noted earlier, collection plate area is a function of the emis-
sion stream flow rate, the particulate drift velocity, and desired control
efficiency. The Oeutsch-Anderson equation relates these variables as
fo 11 ows : 1»2
4.9-3
-------
O
where: A = collection plate area, ft*-
Qe>a = emission stream flow rate at actual conditions
as it enters the control device, acfm
Ud = drift velocity of particles, ft/s
CE = required collection efficiency, decimal fraction
Published data on drift velocities for a number of industrial
applications are presented in Table 4.9-1. When unavailable, a drift
velocity value for an industrial application can be obtained from an ESP
vendor or from literature sources.6 If no value for drift velocity is
known, 0.30 ft/s for particles of "average" resistivity (approximately
107 to 2 x 1010 ohm-cm) and 0.10 ft/s for particles having a "high"
resistivity (10^ to 10^ ohm-cm) can be used. 2
Particles with low resistivities impose special design considera-
tions on an ESP. Such particles (resistivities from 10* to 107 ohm-cm)
are difficult to collect in an ESP because the particles tend to lose
their charge and drop off the collector plate and become reentrained in
the gas stream. In such cases, specially designed collecting plates or
coatings may be used to reduce reentrainment.l>2 Particles with high
resistivities also can cause ESP operating difficulties. High resistivity
particles accumulate on the collection plates and insulate the collection
plate, thus reducing the attraction between the particles and the
collecting plate. In these cases, oversizing an ESP and more frequent
cleaning or rapping of the collector plates are necessary. An alternative
to a larger ESP is the use of conditioning agents to reduce the resistivity
of the particles. Consult a vendor for advice concerning conditioning
agents.
4.9-4
-------
TABLE 4.9-1. TYPICAL VALUES FOR DRIFT VELOCITY
FOR VARIOUS PARTICULATE MATTER APPLICATIONS3
Application Drift Velocity, ft/s
Pulverized coal 0.33 to 0.44
Paper mills 0.25
Open-hearth furnace 0.19
Secondary blast furnace (80% foundry iron) 0.41
Gypsum 0.52 to 0.64
Hot phosphorous 0.09
Acid mist (H2S04) 0.19 to 0.25
Acid mist (TlOg) 0.19 to 0.25
Flash roaster 0.25
Multiple-hearth roaster 0.26
Portland cement (wet manufacturing) 0.33 to 0.37
Portland cement (dry manufacturing) 0.19 to 0.23
Catalyst dust 0.25
Gray-iron cupola (iron-coke ratio = 10) 0.10 to 0.12
a-Source: Reference 2.
4.9-5
-------
EXAMPLE CASE
Flow rate, Qe>a = 110,000 acfm
Drift velocity of particles, U^ = 0.30 ft/s
Collection efficiency, CE = 0.999
From the Deutsch-Anderson equation:
A = -110.000 acfm [In (1 - 0.999)]
P 60 x 0.30 ft/s
A- = 42,200 ft2 of collection plate area.
4.9.3.2 Materials of Construction —
The most common material used in ESP construction is carbon steel.
In cases where the gas stream contains high concentrations of $03 or
where liquid-gas contact areas are involved, stainless steel may be
required.1»2,3,4,5 However, by keeping the emission stream temperature
above the dew point and by insulating the ESP (the temperature drop
across an insulated ESP should not exceed 20°F) the use of stainless
steel should not be necessary.
4.9.4 Evaluation of Permit Application
Using Table 4.9.4, compare the results from this section and the
data supplied by the permit applicant. The calculated values are based
on the example. In evaluating the reasonableness of ESP design specifi-
cations in a permit application, the main task will be to examine each
parameter in terms of its compatibility with the gas stream conditions.
If the applicant's collection plate area is less than the calcu-
lated area, the discrepancy will most likely be the selected drift
velocity. Further discussions with the permit applicant are recommended
to evaluate the design assumptions and to reconcile any apparent
discrepancies.
4.9-6
-------
TABLE 4.9-2 COMPARISON OF CALCULATED VALUES AND
VALUES SUPPLIED BY THE PERMIT
APPLICANT FOR ESP'S
Dri
Col
Col
ft velocity of particles, U
-------
usually operate with a small amount of sparking to ensure that the
voltage is in the correct range and the field strength is maximized.
Automatic voltage controls can control sparking to a specified sparking
frequency (typically 50 to 150 sparks per minute per section of ESP).2
As the spark rate increases, a greater percentage of the input power is
wasted in the spark current. Consequently, less useful power is
applied to the collecting electrode.
4.9.5.2 Cleaning Frequency and Intensity —
Particles accumulating on the collecting plates must be removed
periodically. In wet ESP's the liquid flowing down the collector sur-
face removes the particles.5 In dry ESP's, the particles are removed
by vibrating or rapping the collector plates. For dry ESP's this is a
critical step in the overall performance because improperly adjusted or
operating rappers can cause reentrainment of collected particles or
sparking due to excessive particulate buildup on the collection plates
or discharge electrodes. In normal operation, dust buildup of 6 to
25 mm is allowed before rapping of a given intensity is initiated.1 In
this way, collected material falls off in large clumps that would not
be reentrained. If rapping is initiated more frequently or if the
intensity of rapping is lowered, the resulting smaller clumps of parti-
culate matter are more likely to be reentrained, reducing the collection
efficiency of the ESP. Optimal adjustment of the ESP can best be made
by direct visual inspections through sight ports.
4.9.5.3 ESP Collection Efficiency ~
ESP collection efficiencies less than permitted can be the result
of operational problems, mechanical troubles, or improper design.
Typical operational problems include improper electrical settings,
badly adjusted rappers, full or nearly full dust hoppers, and process
upsets. Mechanical difficulties typically are the result of electrode
misalignment or excessive dust buildup on the electrodes. Basic design
problems include undersized equipment, reentrainment, or high resist-
ivity particles. The permit applicant should carefully examine each of
these items if the ESP is emitting particulate emissions from his
facility that are in excess of permitted levels.
4.9-8
-------
4.9.6 References for Section 4.9
1. Liptak, B.C. Editor. Environmental Engineers' Handbook. Volume II
Air Pollution. Chilton Book Company.Radnor, Pennsylvania.1974.
Pp. 692-718.
2. U.S. EPA. Air Pollution Engineering Manual. 2nd Edition. AP-40.
May 1973.
3. U.S. EPA. A Manual of Electrostatic Precipitator Technology, Part 1
Fundamentals"! APTD 0610. 19707 Pp. 203-213.
4. U.S. EPA. Capital and Operating Costs of Selected Air Pollution
Control Systems. EPA-450/5-80-002. December 1978. Pp. 5-1 through
5-8.
5. Perry, R.H. and D. Green, Editors. Perry's Chemical Engineers'
Handbook. Sixth Edition. McGraw-Hill Book Company. New York,
New York. 1984. Pp. 20-110 through 20-120.
6. The Electrostatic Precipitator Manual. The Mcllvaine Company.
Northbrook IL. T9T^ Chapter IX.
4.9-9
-------
4.10 VENTURI SCRUBBERS
Venturi scrubbers are designed to serve as a control device for
applications requiring very high collection efficiencies of particles
generally between 0.5 to 5.0/iim in diameter. They employ gradually
converging and then diverging sections to clean an incoming gaseous
stream. The section connecting the converging and diverging sections
of the scrubber is called the throat. In general, the longer the
throat, the higher the collection efficiency at a given pressure drop,
provided the throat is not so long that fractional losses become signi-
ficant.1 Typically, a liquid (usually water) is introduced upstream of
the throat and flows down the converging sides into the throat where it
is atomized by the gaseous stream; this method is called the "wetted
approach." Alternatively, the liquid can be injected into the throat
itself by use of nozzles directed at the throat; this approach is
called the "nonwetted approach."1 The nonwetted approach works well
when a gas is already close to saturation; however, this method requires
that the liquid be free of particles that could clog the nozzles.
Where inlet gases are hot and a significant amount of liquid needs to
be evaporated, the wetted approach is preferred.
Once the liquid is atomized, it begins to collect particles from
the gas impacting into the liquid as a result of the difference in
velocities of the gas stream and the atomized droplets. As the mixture
decelerates in the expanding section, further impaction occurs causing
the droplets to agglomerate. Once the particles have been trapped by
the liquid, a separator (e.g. cyclone, demisters, swirl vanes) can
readily remove the scrubbing liquid from the cleaned gas stream.
Appendix C.ll provides a worksheet to record the information obtained
during the performance of the venturi scrubber design procedures.
4.10-1
-------
4.10.1 Data Required
The data necessary ta perform the design steps consist of the HAP
emission stream characteristics previously compiled on the "HAP Emis-
sion Stream Data Forms," and the required HAP control as determined by
the applicable regulations.
EXAMPLE CASE
A venturi scrubber was one of the selected control
techniques for the municipal incinerator emission stream.
The pertinent data for these procedures are found on the "HAP
Emission Stream Data Form" (see Figure 3-8, page 3-43).
1. Flow rate Qe>a = 110,000 acfm
2. Temperature, Te = 400°F
3. Moisture content, Me = 5% vol.
4. Required collection efficiency, CE = 99.9%
5. Particle mean diameter, Dp = 1.0 jim
6. Particulate content = 3.2 grams/scf flyash
7. HAP content = 10% (mass) cadmium
In the case of a permit review for a venturi scrubber, tne following
data should be supplied by the applicant.
1. Reported pressure drop across venturi * "
2. Performance curve applicable to the venturi scrubber.
3. Reported collection efficiency _ ' %
4.10-2
-------
4.10.2 Pretreatment of the Emission Stream
As discussed in Section 3.2.1, the temperature of the emission
stream should be within 50 to 100°F above the stream dew point. Pro-
cedures for determining the dew point of an emission stream are provided
in Appendix B.2. If the emission stream temperature does not fall with-
in the stated range, pretreatment (i.e., emission stream preheating or
cooling) is necessary. Methods of pretreatment are briefly discussed
in Appendix B.ll. If pretreatment is performed, the emission stream
characteristics will be altered. The primary characteristic affecting
venturi scrubber design is the saturated gas flow rate (Qe,s)» a ^unc-
tion of the emission stream temperature (Te) and flow rate at actual
conditions (Qe a). Thus, if the temperature of the emission stream
changes, thus changing the actual flow rate, the saturated gas flow
rate must be based on the new actual flow rate. The calculation method
depends upon the type of pretreatment performed; use appropriate stan-
dard industrial equations. The use of pretreatment mechanical dust
collectors may also dust collectors may also be appropriate, particu-
larly if a "nonwetted" venturi scrubber is used. Appendix B.ll further
describes the use of mechanical dust collectors.
4.10.3 Venturi Scrubber Design Variables
To design a venturi scrubber, any one of three paths may be chosen:
(1) rely on previous experience with an analogous application, which is
best for plants lacking effluent data; (2) test a scrubber on the
source itself; or (3) collect sufficient data about source stream
characteristics, such as particle size distribution, flow rate, and
temperature, to utilize existing "performance curves" for a given
venturi scrubber. This section is concerned with the third path.
Thus, the most important consideration becomes the pressure drop across
the venturi. A secondary consideration is materials of construction.
4.10-3
-------
4.10.3.1 Venturi Scrubber Pressure Drop --
Performance curves are typically logarithmic plots relating venturi
collection efficiency, pressure drop, and particle size.2»3.5,6 Collec-
tion (control) efficiency is usually plotted versus pressure drop across
the venturi (dPv) for a particle mean diameter (Dp). Figure 4.10-1
is a plot of venturi scrubber pressure drops for a given collection
efficiency and particle mean diameter for venturi scrubbers manufactured
by a specific vendor. Thus, if the particle mean diameter for an
emission stream and required collection efficiency is known, the pressure
drop across the venturi can be estimated. Figure 4.10-1 is representative
of plots likely to be used by vendors, and does not necessarily represent
characteristics for all venturi scrubbers.
Estimating the pressure drop gives an indication as to whether
a venturi scrubber is a feasible control device for a given stream.
Venturi scrubbers are used in applications where pressure drops of
between 10 and 80 inches water gauge occur across the venturi. Venturi
scrubbers can operate at pressure drops higher than 80 inches; however,
in general, a pressure drop exceeding 80 inches ^0 indicates that a
venturi scrubber will have difficulty collecting the particles.1
Therefore, if the pressure drop indicated on the performance curve
is greater than 80 inches ^0, assume that the venturi scrubber
cannot accomplish the desired control efficiency.
Table 4.10-1 lists typical pressure drops for venturi scrubbers
for a variety of applications. The pressure drops are listed to
provide general guidance for typical values that occur in industry.
The values are not meant to supersede any specific information known,
and given application may have a pressure drop outside those listed in
Table 4.10-1.
4.10.3.2 Materials of Construction —
Proper selection of the materials in constructing a venturi
scrubber ensures long-term operation with minimal downtime for repair.
The materials are generally chosen based on the corrosive or erosive
nature of the emission stream, and to a lesser degree, the temperature
of the gas stream. For any given application, a vendor should be
contacted to ensure correct selection of materials. A venturi scrubber
4.10-4
-------
TVAOW3U JJG3JI3d
a
o es o o o o c;
\\
\\
\\
^
V
\
\
\
\
\\
\
\
V
\
\
\
\
\
V
\\
\
\\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\\\\
\
\
\
O O O O
* ca r«. «
a so
en
O
00
o'
in
a
„ 23
Figure 4.10-1. Venturi scrubber collection efficiencies,
4.10-5
-------
TABLE 4.10-1.
PRESSURE DROPS FOR TYPICAL VENTURI SCRUBBER
APPLICATIONS3
Application
Pressure drop
(in. H20)
Boilers
Pulverized coal
Stoker coal
Bark
Combination
Recovery
Incinerators
Sewage sludge
Liquid waste
Solid waste
Municipal
Pathological
Hospital
Kilns
Lime
Soda Ash
Potassium chloride
15-40
10-12
6-10
10-15
30-40
18-20
50-55
10-20
10-20
10-20
15-25
20-40
30
Coal Processing
Dryers
Crushers
Dryers
General spray
Food spray
Fluid bed
Mining
Crushers
Screens
Transfer points
25
6-20
20-60
20-30
20-30
6-20
6-20
6-20
(continued)
4.10-6
-------
TABLE 4.10-1. PRESSURE DROPS FOR TYPICAL VENTURI SCRUBBER APPLICATIONS
(concluded)
Pressure drop
Application (in.
Iron and steel
Cupolas 30-50
Arc furnaces 30-50
BOF's 40-60
Sand systems 10
Coke ovens 10
Blast furnaces 20-30
Open hearths 20-30
Nonferrous metals
Zinc smelters 20-50
Copper and brass 20-50
smelters
Sinter operations 20
Aluminum reduction 50
Phosphorus
Phosphoric acid
Wet process 10-30
Furnace grade 40-80
Asphalt
Batch plants—dryer 10-15
Transfer points 6-10
Glass
Container 25-60
Plate 25-60
Borosilicate 30-60
Cement
Wet process kiln 10-15
Transfer points 6-12
aSource: Reference 1.
4.10-7
-------
will generally be constructed of either carbon or stainless steel or a
nickel alloy; it may also be lined with another material (e.g., ceramics)
Table 4.10-2 lists materials of construction for various industries and
is intended to serve as a general guide rather than a definitive state-
ment on the types of materials used in industry.
EXAMPLE CASE
The required collection efficiency is 99.9% and the
particle mean diameter in the municipal waste incinerator
emission stream is estimated to be 1.0/Ltm; therefore:
APV = 47 in. H20 (Figure 4.10-1)
Since the estimated venturi pressure drop value of APV is not
greater than 80 in. 1^0, this venturi scrubber should be able
to accomplish the desired control efficiency. Table 4.10-2
indicates the venturi scrubber should be constructed of 316L
stainless steel.
4.10.4 Sizing of Venturi Scrubbers
If a venturi scrubber is found to be a feasible control choice for
a given emission stream, it is then sized. Venturi scrubbers can be
sized using either the flowrate at inlet conditions (Qe a) or the
saturated gas flowrate (Qe s). Vendors may use either parameter; the
cost data presented in Chapter 5 are based on Qe a. However, more
current cost curves based on Q6jS may be available; therefore, Qe s
should be calculated. A psychrometric chart (Figure 4.10-2) can be used
to determine the saturated gas temperature (Te s), and Qe s can then be
calculated using the following formula:
Qe.s = Qe,a * (Te,s + 460)/(Te + 460)
where: Qe>$ = saturated emission stream flow rate, acfm
~Te,s = temperature of the saturated emission stream, °F
4.10-8
-------
TABLE 4.10-2. MATERIALS OF CONSTRUCTION FOR TYPICAL
VENTURI SCRUBBER APPLICATIONS3
Application
Material of
Construction
Boilers
Pulverized coal
Stoker coal
Bark
Combination
Recovery
Incinerators
Sewage sludge
Liquid waste
Solid waste
Municipal
Pathological
Hospital
316L stainless steel
316L stainless steel
Carbon steel
316L stainless steel
Carbon steel or
316L stainless steel
316L stainless steel
High nickel alloy
316L stainless steel
316L stainless steel
High nickel alloy
Soda Ash
Potassium chloride
Coal Processing
Dryers
Crushers
Dryers
General spray dryer
Food spray dryer
Fluid bed dryer
Mining
Crushers
Screens
Transfer points
Carbon steel or
stainless steel
Carbon steel or
stainless steel
Carbon steel or
stainless steel
304 stainless steel or
316L stainless steel
Carbon steel
Carbon steel or
stainless steel
Food-grade stainless
steel
Carbon steel or
stainless steel
Carbon steel
Carbon steel
Carbon steel
(continued)
4.10-9
-------
TABLE 4.10-2. MATERIALS OF CONSTRUCTION FOR TYPICAL
VENTURI SCRUBBER APPLICATIONS3
(concluded)
Application
Material of
Construction
Iron and steel
Cupolas
Arc furnaces
BOF's
Sand systems
Coke ovens
Blast furnaces
Open hearths
304-316L stainless steel
316L stainless steel
Carbon steel (ceramic lined)
Carbon steel
Carbon steel
Carbon steel (ceramic lined)
Carbon steel (ceramic lined)
Nonferrous metals
Zinc smelters
Copper and brass
smelters
Sinter operations
Aluminum reduction
Stainless
Stainless
steel
steel
or
or
high
high
nickel
nickel
Stainless steel
High nickel
or high nickel
Phosphorus
Phosphoric acid
Wet process
Furnace grade
Asphalt
Batch plants--dryer
Transfer points
Glass
Container
Plate
Borosilicate
Cement
Wet process kiln
Transfer points
316L stainless steel
316L stainless steel
Stainless steel
Carbon steel
Stainless steel
Stainless steel
Stainless steel
Carbon steel or stainless steel
Carbon steel
aSource: Reference 1.
4.10-10
-------
I l\/ U\l ' ' * • !'! 1 A I ' It
Figure 4.10-2. Psycnrometric chart, temp, range 0-500°F, 29.92 in. Hg pressure.
4.10-11
-------
EXAMPLE CASE
Determination of saturated gas flow rate:
Emission stream flow rate, Qe>a = 110,000 acfm
Moisture content, Me = 5% vol.
Emission stream temperature, Te * 400°F
Convert Me to units of Ib ^O/lb dry air, decimal fraction
(Me/100)(18/29) » (5/100)(18/29) = 0.031 Ib H20/lb dry air
T6jS = 127°F (Figure 4.10-2)
Qe>s = (110,000) x (127 + 460)/(400 + 460) = 75,000 acfm
4.10.5 Evaluation of Permit Application
Using Table 4.10-3, compare the results of this section and the
data supplied by the permit applicant. The calculated values in the
table are based on the example. Compare the estimated &PV and the
reported pressure drop across the venturi, as supplied by the permit
applicant.
If the estimated and reported values differ, the differences may
be due to the applicant's use of another performance chart, or a discre-
pancy between the required and reported collection efficiencies.
Discuss the details of the design and operation of the system with the
applicant. If there are no differences between the estimated and
reported values for APV, the design and operation of the system can be
considered appropriate based on the assumptions employed in this manual.
4.10-12
-------
TABLE 4.10-3 COMPARISON OF CALCULATED VALUES AND VALUES SUPPLIED
BY THE PERMIT APPLICANT FOR VENTURI SCRUBBERS
Calculated Value Reported
(Example Case)3 Value
Particle Mean Diameter, Dp = 1.0 /im
Collection Efficiency, CE = 0.999
Pressure drop across venturi, APV = 47"
aBased on the municipal incinerator emission stream.
4.10.6 References for Section 4.10
1. Cheremisinoff, P.N. and Young, R.A. Editors. Air Pollution Control
and Design Handbook; Part 2. Marcel Dekker, Inc. New York, NY.
1977. Pp. 747-777.
2. Liptak, B.G. Editor. Environmental Engineers' Handbook. Volume II
Air Pollution. Chi!ton Book Company.Radnor, Pennsylvania.1974.
Pp. 642-662.
3. U.S. EPA. Uet Scrubber System Study, Volume I: Scrubber Handbook.
EPA-R2-72-118a. August 1972.
4. U.S. EPA. The Cost Digest. Cost Summaries of SelectedEnvironmental
Control Technologies. EPA-600/8-84-010. October 1984.
5. U.S. EPA. Wet Scrubber Performance Model. EPA 600/2-77-172.
August 1977.
6. U.S. EPA. TI-59 Programmable Calculator Programs for Opacity, Venturi
Scrubbers and Electrostatic Precipitators. EPA-600/8-80-024. May
T985T
4.10-13
-------
CHAPTER 5
COST ESTIMATION PROCEDURE
This chapter provides generalized procedures for estimating capital
and annualized costs (June 1985 dollars) for a given add-on HAP control
system. [Note: Calculation of the cost of HAP waste disposal is outside
the scope of this manual; however, this cost must be included in any
rigorous control cost estimation.] The procedures are presented in a
step-by-step format and they are illustrated at each step with cost calcu-
lations pertaining to the thermal incinerator system example discussed in
Section 4.1. A complete cost estimation for the example case (Figure 5-20)
is also presented on the standard cost calculation worksheets at the end
of this chapter beginning on p. 5-51; blank standard cost calculation
worksheets are provided in Appendix C.12.
Only the process HAP control systems presented in Chapter 4 are
discussed in this chapter. The cost of fugitive emission controls a^e
outside the scope of this manual; however, an EPA report entitled "Identi-
fication, Assessment, and Control of Fugitive Particulate Emissions"! can
be used for estimating costs of fugitive emission controls.
5.1 TOTAL CAPITAL COST
In this manual the total capital cost is defined to include only
battery limit costs; thus, it excludes offsite costs. The total capital
cost of a control system is the sum of direct costs, indirect costs, and
contingency costs. Direct costs include the total purchased equipment
cost (i.e., the major equipment purchased cost plus the auxiliary equip-
ment purchased cost), instrumentation and controls, freighted taxes, and
installation costs (i.e., foundation and supports, erection and handling,
electrical, piping, insulation, and painting). [Note: The summation of
the total purchased equipment cost, the cost of instrumentation and
controls, and freight and taxes is defined as the total purchased cost.]
Indirect costs consist of inhouse engineering design and supervision
5-1
-------
costs, architect and engineering contractor expenses, contractor fees,
construction expenses, and preliminary testing costs. An example of
contingency costs are penalties incurred for failure to meet completion
dates or performance specifications.
The capital cost estimation procedure presented in this manual is
for a "factored" or "study" estimate. Usual reliability for a study
type estimate is _+ 30 percent. To determine the total capital cost by a
factored cost estimate, a reliable estimate of the total purchased cost
is calculated and predetermined factors are applied to determine all
other capital cost elements. Therefore, the procedure to estimate the
total capital cost is as follows: (1) obtain the total purchased equip-
ment cost by estimating the purchased cost of major and auxiliary equip-
ment; (2) estimate the cost of instrumentation and controls plus freight
and taxes as a percentage of the total purchased equipment cost; (3) esti-
mate the total purchased cost by adding (1) and (2) above; and (4) estimate
total capital cost by applying a predetermined cost factor to the total
purchased cost.
5.1.1 Estimation of Major Equipment Purchased Cost
The major equipment purchased cost (i.e., the cost of the major
components that comprise the control system) is related to a specific
equipment design parameter and can be expressed either analytically or
graphically. Table 5-1 (p. 5-38) presents a list of the design parame-
ters needed for costing the HAP control equipment, and it identifies the
figure that presents the applicable purchased cost curve. Gathering
current costs from vendors was beyond the scope of this project and,
thus, necessitated use of dated cost data compiled by others. In general,
cost estimates should not be escalated beyond 5 years. If more recent
cost data are available, they should be substituted for the cost curve
data presented. These cost curves should not be extrapolated beyond
their range. The cost data presented in these figures were obtained
from cost information published in EPA reports.2.3 TO escalate the cost
data to June 1985 dollars, multiply the cost estimate by the ratio of
the Chemical Engineering Fabricated Equipment (FE) cost indices for June
1985 and the date of the cost data. For example, if a cost is given in
5-2
-------
December 1977 dollars, it is converted to June 1985 dollars using a
factor of 1.49 [336.0 (June 1985)7226.2 (Dec. 1977)]. Table 5-2
(p. 5-39) presents the monthly FE cosfindices from December 1977
through June 1985.
Using the specific value for the design variable, obtain purchased
costs from the specific cost curve for each major control system compo-
nent. Presented below are brief descriptions of the equipment costs
included in each HAP control cost curve.
The cost curve for thermal incinerators (Figure 5-1, p. 5-19)
includes the fan plus instrumentation and control costs, in addition to
the major equipment purchased cost. If the HAP control system includes
a heat exchanger, its cost (Figure 5-2, p. 5-20) is part of the major
equipment purchased cost and, thus, must be added. The remaining auxiliary
equipment (ductwork and stack) purchased costs and costs of freight and
taxes must be added to obtain the total purchased cost.
The cost curve for catalytic incinerators (Figure 5-3, p. 5-21)
provides the cost of an incinerator less catalyst. Catalyst costs
(Table 5-3, p. 5-40) and the cost of a heat exchanger, if applicable,
(Fig. 5-2, p. 5-20) must be added to obtain the major equipment purchased
cost. All auxiliary equipment (ductwork, fan, and stack) purchased
costs, the cost of instrumentation and controls, and freight and taxes
must be added to obtain the total purchased cost.
Two cost curves are presented for carbon adsorbers: Figure 5-4
(p. 5-22) for packaged carbon adsorbers and Figure 5-5 (p. 5-23) for
custom carbon adsorbers. The cost curve for packaged carbon adsorbers
includes the fan plus instrumentation and control costs, in addition to
the major equipment purchased cost. The cost of the remaining auxiliary
equipment (ductwork and stack), as well as, costs of freight and taxes
must be added to obtain the total purchased cost. The cost curve for
custom carbon adsorbers does not include the cost of carbon (part of the
major equipment purchased cost), however, it does include the cost of
instrumentation and controls; the cost of carbon is obtained from Table 5-3
(p. 5-40). All auxiliary equipment (ductwork, fan, and stack) purchased
costs and freight and taxes must be added to obtain the total purchased
cost.
5-3
-------
The cost curve for Absorbers (Figure 5-6, p. 5-24) does not include
the cost of packing, platforms, and ladders. The cost of platform and
ladders (Figure 5-7, p. 5-25) and packing (Table 5-4, p. 5-41) must be
added.to obtain the major equipment purchased cost. All auxiliary
equipment (ductwork, fan, and stack) purchased costs, the cost of instru-
mentation and controls, and freight and taxes must be added to obtain
the total purchased cost.
The cost curve for condensers (Figure 5-8, p. 5-26) yields the total
capital cost for cold water condenser systems. For systems needing refrig-
erant (ethylene glycol), the applicable cost from Figure 5-9 (p. 5-27)
must be added to the cost obtained from Figure 5-8. Since a total capital
cost is determined, no additional cost estimates are necessary; therefore,
proceed to Section 5.2 to calculate annualized operating costs.
The cost curve for a negative pressure fabric filter (Figure 5-10,
p. 5-28) does not include the the cost of bags (Table 5-5, p. 5-42),
which depend upon type of fabric used. This cost must be added to obtain
the major equipment purchased cost. All auxiliary equipment (ductwork,
fan, and stack) purchased costs, the cost of instrumentation and controls,
and freight and taxes must be added to obtain the total purchased cost.
The cost curve presented in Figure 5-11 (p. 5-29) provides the major
equipment purchased cost for an insulated electrostatic precipitator.
All auxiliary equipment (ductwork, fan, and stack) purchased cost, the
cost of instrumentation and controls, and freight and taxes must be
added to obtain total purchased cost.
The cost curve for venturi scrubbers (Figure 5-12, p. 5-30) includes
the cost of instrumentation and controls, in addition to the major equip-
ment purchased cost. This cost curve is based on a venturi scrubber
constructed from 1/8-inch carbon steel. Figure 5-13 (p. 5-31) is used
to determine if 1/8-inch steel is appropriate for a given application (use
the higher curve). If thicker steel is required, Figure 5-14 (p. 5-32)
yields a price adjustment factor for various steel thicknesses; this fac-
tor is used to escalate the cost obtained from Figure 5-12. In addition,
if stainless steel is required (see Section 4.10.3.2) multiply the
scrubber cost estimate by 2.3 for 304L stainless steel or by 3.2 for
316L stainless steel. Costs of all auxiliary.equipment (ductwork, fan,
and stack) and freight and taxes must be added to obtain the total
purchased cost.
5-4
-------
EXAMPLE CASE
The example thermal Incinerator system case (see
Section 4.1) consists of an incinerator with a combustion
chamber volume (Vc) of approximately 860 ft , and a primary
heat exchanger with a surface area (A) of approximately
4,200 ft^. From the cost data presented in Figures 5-1 and
5-2, June 1985 cost estimates are obtained as follows:
(a) Incinerator plus instrumentation and control costs -
$98,000 x (336.2/226.2) = $145,700
[Note: 12/77 dollars escalated to reflect 6/85 dollars.]
(b) Heat exchanger cost -
$85,000 x (336.2/273.7) = $104,400
[Note: 12/79 dollars escalated to reflect 6/85 dollars.]
5.1.2 Estimation of Auxiliary Equipment Purchase Cost
The auxiliary equipment purchase cost .is related to emission stream
and equipment parameters. Table 5-6 (p. 5-43) presents the parameters
that must be known for costing the auxiliary equipment. Figures 5-15
through 5-19 (pages 5-33 through 5-37) present the December 1977 costs
for ductwork, fans, and stacks. The cost information presented in the
figures are from available published data and must be escalated to reflect
June 1985 dollars.
5.1.2.1 Ductwork Purchase Cost —
The ductwork purchase cost is typically porportional to the ductwork
weight, which is a function of: (1) the material of construction, (2)
length, (3) diameter, and (4) thickness. Carbon steel ducts are normally
used for noncorrosive flue gases at temperatures below 1,150°F. Stainless
steel ducts are generally used with gas temperatures between 1,150°F to
1,500°F, or if the gas stream contains corrosive materials. Figures 5-15
(p. 5-33) and 5-16 (p. 5-34) present purchase costs for carbon steel and
stainless steel ducts, respectively. It is assumed that the major portion
5-5
-------
of ductwork is utilized to transport the emission stream from the process
to the control system; therefore, the flow rate of the emission stream at
actual conditions (Qe>a) is used to size tne ductwork.
Without specific information, assume the following items to sim-
plify the costing procedure:
(1) The ductwork is constructed with 3/16-inch thick plate.
(2) The duct length equals 100 feet.
(3) The duct diameter is calculated using a duct gas velocity
of 2,000 ft/min. Therefore:
D » 12 ( JL I£ii ) * 0.3028 (Q )1/2 (5-1)
duct \ rr U
-------
5.1.2.2 Fan Purchase Cost --
The fan purchased cost (Figure 5-17, p. 5-35) is a function of the
flow rate moved by the fan and the pressure drop (AP) across the control
system. The fan is assumed to be located downstream of the final control
device in the control system. Therefore, the fan capacity must be based on
the final control device's exit gas flow rate at actual conditions (Qfgsa)-
Control system pressure drop (dP) is the total of the pressure drops across
the various control system equipment, including the stack and ductwork.
Table 5-7 (p. 5-44) presents conservative pressure drops across specific
control system components which can be used if specific data are not availa-
ble. Using the actual flow rate and total AP parameters, obtain the fan
purchased cost from Figure 5-17. Fans are categorized into Classes I to IV
according to control system pressure drop. Guidelines are presented on
Figure 5-17 to determine which class of fan to use. There is some overlap
between the classes. The lower class fan is generally selected due to
cost savings. To estimate the cost of a motor for the fan, multiply the
fan cost by 15 percent. [Note: The fan and motor costs are included in
the cost curves for thermal incinerators and packaged carbon adsorbers.]
EXAMPLE CASE
For the example case, the fan and motor costs
(Figure 5-17) included in the thermal incinerator cost curve;
however, these costs can be calculated separately. The total
pressure drop across the control system is 7 in. 1^0 (obtained
from summing the values from Table 5-7 for the incinerator,
heat exchanger, ductwork, and stack). The flow rate exiting
the heat exchanger is approximately 40,000 acfm (Qfg,a)- The
pressure drop the guidelines on Figure 5-17 indicate that a
Class II fan (the lower class fan) is appropriate. The esti-
mated fan and motor costs are as follows:
(a) Fan cost -
$5,000 x (336.2/226.2) = $7,400
[Note: 12/77 dollars escalated to reflect 6/85 dollars.]
(b) Motor cost --
$7,400 x 0.15 = $1,100
5-7
-------
5.1.2.3 Stack Purchase Cost --
The stack purchase cost is a function of: (1) the material of
construction, (2) height, (3) diameter, and (4) stack thickness.
In addition, minimum stack exit velocities should be at least 1.5 times
the expected wind velocity; or for instance, in the case of 30 mph winds,
the minimum exit velocity should be at least 4000 ft/mi n.2 For purposes
of this manual, the stack is designed and costed with respect to the
final control device's exit gas flow rate at actual conditions (Qfg,a)«
Figures 5-18 (p. 5-36) and 5-19 (p. 5-37) present purchased costs for
unlined, carbon steel stacks.
Without specific information, assume the following items to
simplify the costing procedure:
(1) The stack is constructed with 1/4-inch thick carbon steel
plate.
(2) The stack height equals 50 feet.
(3) The stack diameter is calculated using a stack exit velicity
of 4000 ft/min. Therefore:
Dstack = 12 J_ x ,,Qf9'a^ = 0.2141 (Q. }1'2 (5-2)
where: ^stack = stack diameter, in.
Qfg,a = flue gas flow rate at actual conditions, acfm
Dstack = velocity of gas stream in stack, ft/min
EXAMPLE CASE
In the example case, since no specific data for the stack
are available, use the above assumptions to cost the stack.
The stack diameter is calculated according to item (3) above.
The actual gas flow rate exiting the heat exchanger (flue gas
flow rate) equals approximately 40,000 acfm. Therefore, the
stack diameter is calculated as follows:
Dstack = 0.2141 (40,000)1/2 = 43 in.
5-8
-------
With the stack diameter known, use the appropriate
curve in Figure 5-18 (use the closest curve: 42 inches)
to estimate the stack cost as follows:
$4,500 x (336.2/226.2) = $6,700
[Note: 12/77 dollars escalated to reflect 6/85 dollars.]
5.1.3 Estimation of the Total Purchased Equipment Cost
The total purchased equipment cost equals the sum of the major
equipment purchased cost and the auxiliary equipment purchased cost.
[Note: The major equipment purchased cost curves for thermal incinerators,
carbon adsorbers, and venturi scrubbers also include the cost of instrumen-
tation and controls; therefore, this cost must be subtracted to estimate
the total purchased equipment cost for these control devices. Calculate
the total purchased equipment cost for thermal incinerators, carbon adsor-
bers, and venturi scrubbers as follows: (1) multiply the summation of the
major equipment purchased cost and the auxiliary equipment purchased cost
by a factor of 0.091 to obtain the cost of instrumentation and controls
(the cost of instrumentation and costs is estimated to equal 10 percent of
the total purchased equipment); and (2) subtract this cost from the summation
of the major equipment purchased cost and the auxiliary equipment purchased
cost.
EXAMPLE CASE
For the example case, the total purchased equipment
cost is estimated as follows:
(a) Total purchased equipment cost plus cast of instrumen-
tation and controls (included in cost curve) -
$145,700 + $104,400 + $7,700 + $6,700 = $264,500
5-9
-------
(b) Cost of instrumentation and controls -
$264,500 x 0.091 = $24,100
(c) Total purchased equipment cost -
$264,500 - $24,100 = $240,400
5.1.4 Estimation of Instrumentation and Controls Plus Freight and Taxes
For the majority of control equipment, instrumentation costs are a
small part of the total purchased cost. Instrumentation requirements
for a control system depend upon control and safety requirements. When
no specific cost data are available, estimate the instrumentation and
controls costs at 10 percent of the total purchased equipment cost.
The cost of equipment freight and taxes depend upon the location of
control system and the location of the supplier. Without specific data,
estimate the cost of freight and taxes at 8 percent of the total
purchased equipment cost.
EXAMPLE CASE
For the example case, the cost of instrumentation and
controls and the cost of freight and taxes are as follows:
(a) Instrumentation and controls -
$240,400 x 0.10 = $24,000
(b) Freight and taxes -
$240,400 x 0.08 = $19,200
5.1.5 Estimation of Total Purchased Cost
The summation of the total purchased equipment cost, the
cost of instrumentation and controls, and the cost of freight and
taxes equals the total purchased cost.
5-10
-------
EXAMPLE CASE
The total purchased cost for the example case is as
follows:
$240,400 + $24,000 + $19,200 = $283,600.
5.1.6 Calculation of Total Capital Costs .
The sum of the total purchased equipment cost, other direct costs,
indirect costs, and contingency costs represents the total capital cost.
Obtain the total capital cost for the control system by multiplying the
total purchased cost by the appropriate factor listed in Table 5-8
(p. 5-45). This factor accounts for the other direct costs, the indirect
costs, and the contingency costs. For control systems employing multiple
control devices, use the largest applicable factor. Retrofit applications
will likely be more expensive (see Footnote "e" of Table 5-8). Each
component that comprises the total capital cost is listed in Table 5-8 to
allow insertion of specific cost data if they are known.
EXAMPLE CASE
The example case consists of a control system using
one control device: a thermal incinerator. Therefore,
by using the factor of 1.63 for thermal incinerators
from Table 5-8, the total capital cost for the example
case is as follows:
$283,600 x 1.63 = $462,300
5-11
-------
5.2 ANNUALIZED OPERATING COSTS
The annualized cost of a control system can be divided into direct
operating costs, indirect operating costs, and credits. In this manual,
the inflation effect on costs is not considered, annualized costs are
assumed to be constant in real dollars, and the total annualized cost is
estimated on a before-tax basis. The direct operating costs consist of
utilities, operating labor charges, maintenance charges, and replacement
parts and labor charges. Utilities (i.e., fuel, electricity, water,
steam, and materials required for the control system) are annual costs
that vary depending upon the control system size and operating time.
They are calculated using gas stream characteristics and control equip-
ment capacity data. Operating labor costs consist of operator labor and
supervision, while maintenance costs consist of maintenance labor and
materials. The direct operating costs are established by estimating
annual quantities of utilities consumed and operator and maintenance
labor used and by applying unit costs to these quantities. The annual
quantities of utilities and labor requirements are assumed to be propor-
tional to the annual operating hours for the control system. Operating
labor supervision and maintenance materials are taken as percentages of
the operator and maintenance labor costs. Costs of replacement parts are
estimated as applicable, and the cost of replacement labor is assumed to
equal the cost of the replacement parts.
The indirect operating costs include overhead costs, property tax,
insurance, administration costs, and the capital recovery costs. Overhead
costs are estimated as a percent of operating labor and supervision costs
plus maintenance labor costs. Property tax, insurance, and administration
costs are estimated as a percent of the total capital cost. The capital
recovery cost is estimated as the product of the capital recovery factor
times the total capital cost. The factor for capital recovery costs (the
total of annual depreciation and interest on capital) is determined from
the expected life of the control device and the interest rate at which the
capital is borrowed. The expected life of a given control device depends
on the type of control application, maintenance service, and operating
duty. For costing purposes, preestablished expected life values are used.
5-12
-------
Some control techniques recover the HAP's from a given emission
stream as a salable product. Therefore, any cost credits associated
with the recovered material must be deducted from the total annualized
cost to obtain the net annualized cost for the system. The amount,
purity, and commercial value of the recovered material determine the
magnitude of credits.
5.2.1 Direct Operating Costs^
Table 5-9 (p. 5-46) presents June 1985 unit costs for utilities,
operator labor, and maintenance labor as well as cost factors for
other direct operating cost elements. The procedure used to estimate
direct operating costs (including utilities, direct labor, maintenance,
and replacement costs) and indirect operating costs (including overhead,
property tax, insurance, administration, and capital recovery cost) was
taken directly from Reference 2. These unit costs and cost factors
are applied to estimated quantities of utilities consumed, labor
expended, and parts used to obtain total direct operating costs.
If a given control system contains two or more control devices,
the direct operating costs must be calculated for each device and
summed. The capital recovery cost for a multiple control device system
should be calculated using a weighted average capital cost factor.
Unless specified, use 8,600 hours per year, 8 hours per shift,
and 24 hours per day, as necessary, to the estimate annual costs for
utilities consumed, operator labor, and maintenance labor.
5.2.1.1 Determine Utility Requirements --
The utility requirements for a control system are obtained from
each component's design calculations. Use the costing information in
Table 5-9 (p. 5-46), Table 5-10 (p. 5-47), and Table 5-11 (p. 5-48)
to estimate the total utility costs. A procedure to estimate fan
electricity costs is provided below, since these costs are applicable
to all control techniques. The fan horsepower requirements are
calculated as follows:
5-1.
-------
Fan horsepower, HP = Q>000157 x Qf x AP (5-3)
where: HP = fan horsepower requirement, hp.
AP = pressure drop across the control system, in. H£0
T) = fan efficiency (usually 60-70%)
Assuming a 65 percent fan efficiency and a 10 percent additional capacity
requirement for miscellaneous purposes, and using the conversion factor of
0.746 kilowatt hour per horsepower-hour, estimate the fan electricity
requirement as follows:
PER = 2.0 x ID'4 (Qfg>a) UP) (HRS) (5-4)
where: PER = fan electricity requirement, kWh
HRS = hours of operation per year
EXAMPLE CASE
For the thermal incinerator example case, since no
specific information is available, the control system is .
assumed to operate 8,600 hours per year. According to the
design calculations in Section 4.1, and Table 5-10, the only
utility requirement for the thermal incinerator, in addition
to fan electricity, is 330 scfm of natural gas. Using
Equation 5-4, the electricity requirement for a 40,000 acfm
fan at 7 in. H20 pressure drop is as follows:
2.0 x lO'4 (40,000 acfm) (7 in.) (8,600 hrs) = 481,600 kWh.
Applying the unit costs from Table 5-9 and the equation for
fuel requirements from Table 5-11, the utility costs are:
(a) Natural gas cost -
60 (330 scfm)(8,600 hrs)($0.00425/ft3) = $723,700
(b) Electricity cost -
481,600 kWh ($0.059/kWh) = $28,400.
5-14
-------
5.2.1.2 Determine Remaining Direct Operating Costs --
The remaining direct operating costs include replacement parts
and labor, operating labor (i.e., the summation of operator labor
and supervision labor), and maintenance (i.e., the summation of mainte-
nance labor and materials). Tables 5-9 (p. 5-46) and 5-10 (p. 5-47)
provide the necessary information to calculate the cost of replacement
parts and labor. Table 5-12 (p. 5-50) presents available data on estimated
labor requirements for various control devices. The labor requirements
presented as "hours per shift" must be converted to annual requirements.
Total annual operator and maintenance labor costs are obtained by
multiplying the estimated annual labor requirements with the applicable
unit costs from Table 5-9. These costs must be determined for each control
device in the control system. Operating labor supervision and maintenance
materials are estimated as a percentage of operator labor and maintenance
labor, respectively. Again, these costs must be determined for each
control device if a multiple control device system is used.
EXAMPLE CASE
Table 5-10 indicates that thermal incinerators do not
require replacement parts or labor. From Tables 5-9 and 5-12,
both the estimated operator and maintenance labor requirements
and their unit costs for thermal incinerator systems are 0.5
hours/shift and $11.53/hr, respectively. Therefore, the annual
operator and maintenance labor costs for the example are the
same and each cost is estimated as follows:
8,600 hr/yr x 0.5 hr/shift x 11.53/hr •» 8 hr/shift = $6,200.
Estimated other direct operating costs for the example
case using the Table 5-9 factors are as follows:
Operating Labor Supervision = $6,200 x 0.15 = $900
Maintenance Materials = $6,200 x 1.00 = $6,200
5-15
-------
Therefore, the total annual direct operating cost for the
example case is estimated by suming the utility costs and the
remaining direct operating costs:
$723,700 + $28,400 + $6,200 + $900 + $6,200 + $6,200 = $771,600.
5,2.2 Indirect Operating Costs
The indirect operating costs include overhead costs, property tax,
insurance, administration, and the capital recovery cost. Estimate the
overhead costs as 80 percent of the direct labor cost (the summation of
operating labor and supervision labor costs) and the maintenance labor
cost. The property tax estimate is calculated as 1 percent of the total
capital cost, insurance as 1 percent of the total capital cost, and
administration as 2 percent of the total capital cost. Estimate the
capital recovery cost portion of the fixed capital charges by multiplying
the total capital cost by a capital recovery factor. The capital recovery
factor (CRF) is calculated as follows:
CRF = [1(1 + i)"] / [(1 + 1)" - 1] (5-5)
where: i = interest rate on borrowed capital, decimal
n = control device life, years
For the purpose of this manual, an interest rate of 10 percent is
used. Table 5-12 (p. 5-50) contains data on expected control device
life (n). Calculated capital recovery factors at 10 percent interest
rate are 0.163, and 0.117 for 10-, and 20-year control device life-
times, respectively. If more than one control device is used by the
control system, use a weighted average capital recovery factor. A
weighted average capital recovery factor (CRFW) determined as follows:
CRFW - CRF! [Pq/CPCi + PC2)3 + CRF2 [PC2/(PCi + PC2)]
where: CRF^ = the capital recovery factor for control device #1
= the capital recovery factor for control device #2
= the purchased equipment cost for control device #1
PC2 = the purchased equipment cost for control device #2
5-16
-------
EXAMPLE CASE
As estimated in Section 5.1.5, the total capital cost
for the thermal incinerator control system example is $462,300.
Section 5.2.1.2 estimated that the direct and maintenance labor
cost is $13,300 (i.e., $6,200 + $900 + $6,200). Therefore,
the indirect operating costs are estimated as follows:
(a) Overhead costs -
$13,300 x 0.80 = $10,600
(b) Property tax -
$462,300 x 0.01 = $4,600
(c) Insurance -
$462,300 x 0.01 = $4,600
(d) Administration charges -
$462,300 x 0.02 = $9,200
(e) Capital recovery -
The capital recovery factor is calculated using Equation 5-5
obtain the expected equipment lifetime for an incinerator
from Table 5-12 (10 years), and assume an interest rate of
10 percent.
[0.1 (1 + 0.1)10] / [(i + o.l)10 - 1] = 0.163
The capital recovery cost is then estimated as follows:
$462,300 x 0.163 = $75,400
(f) Total indirect operating costs -
$10,600 + $4,600 + $4,600 + $9,200 + $75,400 - $104,400.
5.2.3 Credits
Credits for recovery of a salable product or energy must be
included in determining the net annualized cost of the control system.
The design calculations for the specific control devices include the
quantity of recovered product. This information, along with product
cost data from the inquirer/applicant, is used to calculate the credits
on an annual basis.
5-17
-------
EXAMPLE CASE
For the example case, there are no recovered products
since a thermal incinerator destroys the organic vapors
contained in the emission stream.
5.2.4 Net Annualized Costs
The direct and indirect operating costs less credits received equal
the net annualized cost of the HAP control system.
EXAMPLE CASE
Total direct and indirect operating costs for the
example case are $771,600 and $104,400, respectively.
There are no recovery credits. Thus, the net annualized
cost of the example HAP control system is as follows:
$771,600 + $104,400 - $0 = $876,000
5-18
-------
no
2 90 4-
s.
8
(U
80 ••
-------
§
O
O
, o
_s
u
X
o>
as.
as.
UJ
(3
UJ
2
in
.8
CM
a.—»
3 CM
U
-------
i i t i i .1 i
II I
L I I
o
o
o
o
o
o
10
T3
*J
O
VI
0)
Z
'-J_
VI
o a-
C 0)
•fw O
-------
41
u
V
a
ui
30 -
20 -
10
CARBON WEIGHT, Creq. (1,000 IB)
Figure 5-4. Prices for carbon adsorber packages.
Price includes carbon, beds, fan and motor, and instrumentation and controls
(Source: Reference 8)
. 0-7
—i-i.
-------
06
Ul
03
oe
o
!/>
Q
20 40 60 80 100 120 140 160
CARBON WEIGHT, Creq. (1,000 Its)
180 200
Figure 5-5. Prices for custom carbon adsorbers, less carbon.
Price includes beds, instrumentation and controls, and a steam regenerator,
(Source: Reference 8)
5-23
-------
1,000
500
o
o
00
91
I 100
-3
O
O
O
~ 50
oe
a.
o
GO
«c
10
~7
10
50
100
500
1,000
COLUMN WEIGHT, Wtcoiumn U.OOO Ib)
Figure 5-6. Prices for absorber columns, including manholes,
skirts, and painting.
(Source: Reference 9)
5-24
-------
o
a
ao
at
41
•5
ac
a.
COLUMN DIAMETER, Oco,umn (FT.)
Figure 5-7. Prices for absorber platform and ladders,
(Source: Reference 9)
5-25
-------
50 100
500 1,000
5,000 10,000
CONDENSER SYSTEM AREA, ACQru (SQ. FT.
Figure 5-8. Total capital costs for cold water condenser systems.
(Source: Reference 10)
5-26
-------
1 1.00Q
ifl
500
A
r«»
o>
8
100
50
fi
UJ
a
s
_| I I I I LI
10
50 100
500 i,uOO
REFRIGERATION CAPACITY, Sef. (TONS)
Figure 5-9. Additional capital cost for refrigerant condenser systems.
(Source: Reference 10)
5-27
-------
400
o
a
-------
10,000c
5,000
1.004
-------
o
o
CVI
o
00
0 §
CNJ 2
3 O
U U
U *J
i« c
o
o> u
U C •—
C O Ol
•r- •!- OJ
4-> 4J
•
t_ C C CM
OJ Ol O
-C E J2 OJ
^ 3 4- U
3 <- fl3 C
(_ >J U O)
o to t-
01 C 5 01
•«- oo «-
•r- ^ O)
t_ -a ^-4 a:
3 C
> in -a o
CL
-------
J I
CM
8
o -»
in z
u.
o
o
§
o
u.
2 5
fl** gj
ac
U) £
t_
OJ
-------
if
i
•3
3/8"
1/4*
3/16'
20 40 60 30
100 120 140
EMISSION STREAM FLUWRATE, Qefd (1,000 AC?M)
Figure 5-14. Price adjustment factors for venturi scrubbers.
For use with Figure 5-12.
(Source: Reference 2)
5-32
-------
oe
UJ
1/1
•U
•s\
3
o
L.
rt>
Ol
O
a.
c •—•
O fVJ
4-J OJ
re o
o c
•f- Hi
~ u^ re
"* uj *-
3 ••
•3 01
O
—1 L.
J= 3
J> C
01
X
c
o
m
i
O)
3
j*ww»a) -jj wan/t
5-33
jjno
-------
8
•8
UJ
O
•O >—
03 U
>o
CM
OJ
to
0)
c
en
3
O
01
u
c
O *-»
•r- CSJ
4-1
ro 0)
U U
f- c
(_ 0)
OJ
a:
Ol
U
"a!
Ol
(/I
to
o>
01
D
O)
(SJ«UOQ USI
*JJ
'SJIbd
5-34
-------
\\ \ \\ \ \
M§
I? 5
*!s
Is
_8'.
^ V J
|i§
1
*I*
•* u *«•
i5ki
H'l"
5.3
V *
w «
Is
sfisrsl
I:
-S52(o;l
r*
V. Ol
C 01
U) S-
U. 0)
-b.
Ol
rv.
~^
I
o>
3 O
a -
5-35
-------
s-
Ifl
o
0)
o
o
o
Q£
a.
o
H-
1/1
20
30
40-
50
60
70
30
90
STACK HEIGHT, H (FT.
Figure 5-13.
Carbon steel stack fabrication price for 1/4" plate.
(Source: Reference 2)
5-36
-------
1.
"a
9
a
-------
TABLE 5-1. IDENTIFICATION OF DESIGN PARAMETERS AND COST CURVES
FOR MAJOR EQUIPMENT
Control
Equipment
Thermal incinerator
Heat exchanger
Catalytic incinerator
Carbon adsorber
Absorber
Condenser
Fabric Filter
Electrostatic Precipitator
Venturi Scrubber
Design
Parameter3
vc
A
Qfg
Creq
Wtcol .
Dcol.
Aeon
Ref
Anc
Ap
Qe,a
Cost Curve
Figure No.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
5-1 (p. 5-19)
5-2 (p. 5-20)
5-3 (p. 5-21)
5-4 (p. 5-22)b
5-5 (p. 5-23)c
5-6 (p. 5-24)
5-7 (p. 5-25)
5-8 (p. 5-26)
5-9 (p. 5-27)
5-10 (p. 5-28)
5-11 (p. 5-29)
5-12 (p. 5-30)
5-13 (p. 5-31)
5-14 (p. 5-32)
aSee nomenclature list for definitions of variables.
bPackaged carbon adsorbers.
cCustom carbon adsorbers.
5-38
-------
TABLE 5-2. C.E. FABRICATED EQUIPMENT COST INDICES (FE)a
Date
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
1977
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1978
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1979
1980
1980
1980
1980
1980
1980
aSource:
FE
226.
226.
233.
233.
237.
237.
237.
238.
243.
243.
243.
244.
245.
245.
252.
253.
253.
258.
259.
262.
264.
266.
271.
272.
273.
273.
276.
277.
289.
290.
291.
Chemical
Date
2
6
0
6
1
3
4
6
3
2
8
1
2
2
5
1
7
3
9
6
2
6
6
6
7
8
9
7
3
9
3
Engi
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
neering
1980
1980
1980
1980
1980
1980
1981
1981
1981
1981
1981
1981
1981
1981
1981
1981
1981
1981
1982
1982
1982
1982
1982
1982
1982
1982
1982
1982
1982
1982
1983
FE
296.
297.
298.
301.
302.
304.
305.
307.
314.
321.
321.
322.
325.
325.
326.
330.
329.
328.
324.
323.
324.
327.
329.
327.
327.
326.
326.
325.
324.
325.
324.
Date
7
3
1
2
5
0
9
1
7
9
6
9
6
7
7
8
4
9
5
4
1
8
1
5
1
2
7
8
8
1
4
. McGraw-Hill
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
1983
1983
1983
1983
1983
1983
1983
1983
1983
1983
1983
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1984
1985
1985
1985
1985
1985
1985
1985
1985
FE
327.
326.
326.
327.
327.
327.
327.
328.
327.
328.
330.
331.
333.
332.
333.
334.
333.
335.
335.
335.
335.
335.
336.
336.
336.
336.
6
8
6
1
3
0
1
0
8
9
I
5
0
9
8
6
8
4
1
9
0
4
5
9
,5
6
338.0
336.0
336.2
Publications.
5-39
-------
TABLE 5-3. UNIT COSTS FOR VARIOUS MATERIALS
(June 1985 Dollars)
Chemical Cost
Refrigerant $0.31/lba
(Ethylene Glycol)
Activated Carbon $1.92/lbb
Catalyst (Platinum-based) $2,750/ft3c
aSource: Reference 12.
bSource: Reference 13.
cSource: Reference 14.
5-40
-------
TABLE 5-4. PRICE OF PACKING FOR ABSORBER SYSTEMS3
Packing Type and
Material
Cost/Cubic Foot
(June 1981 dollars)
Packing diameter, inches
1.5
Pall rings:
Carbon steel
Stainless steel
Polypropylene
Berl saddles
Stoneware
Porcelain
Intalox saddles
Polypropylene
Porcelain
Sto.neware
Packing rings
Carbon steel
Porcelain
Stainless steel
24.3
92.1
21.9
28.1
34.5
21.9
19.4
18.2
30.3
13.2
109.0
16.5
70.3
14.8
21.7
25.6
--
14.8
13.3
19.8
10.6
82.6
15.1
0.8
13.8
--
—
13.6
13.3
12.2
17.0
9.7
22.9
--
-_
--
--
--
7.0
12.2
11.0
13.9
8.1
~ "
aSource: Reference 9.
5-41
-------
IB
^-v
O
O
U.
Ol
(O
3
o-
oo
I/I
in
O
t_
CJ3
tS)
e
,2
r—
o
Q
o!
•-*
.
o
Ol
Q
UJ
CJ
>— *
ce
OU
1
LO
i
^f>
LU
i
33
<
h-
C
O
4.J
O
C_3
01
Ol
a.
o
0.
'o
O.
U1
V)
,_
0
X
E
O
c
o
f~~
>>
c
o
C_
o
c
o
o
ra
a
01
a.
i—
in
ui
(0
^»
f ^
LO
.
O
LO
LO
O
O
LO
*
o
LO
,_(
•
^
LO
1 —
M
o
LO
LO
o •
o
•
o
c_
Ol
\s
*B
CiNI
OO *J
,_ **"
no O
0 O
•r- O
C "
03 O
f CM
O V
i
•a
ft3
•a
c
'"O
^J
oo
o
,
o
LO
LO
CD
LO
^
O
LO
o
*
f-H
O
f%fc
«
o
o
LO
o
LO
CO
•
o
c
cu
•s^
f^
-CC\J
c/> ^
,__
ftj ^D
(J O
••- o
c -
fQ ^^
J= CM
O A
•a
oo
1
o
r->-
O
i
o
•
r— 1
1
I
LO
O
o
LO
•
o
-Q
4_)
01
•<—)
Ol
r—
a.
•o
PO
•a
c
^o
4_)
oo
0
.
o
LO
oo
O
o
oo
o
LO
LO
•
o
LO
^«
e
0
LO
OO
0
LO
CM
•
O
t.
O)
^^
rt3
JE
oo
,_
fl3
U
•^*
C
fQ
f~
o
01
ff
0
<_>
I/)
3
U
o
.
o
LO
oo
o
o
oo
o
'XI
•
a
LO
f
O
LO
OO
O
LO
CM
•
0
•r™
^£
QJ
CO
t_
QJ
>
0)
Oi
4J
in
3
O
•
LO
*
1)
O
t_
3
O
>l
a.
• «^"
CM 4->
r«»
Ol 3
LJ E
c
cy *
0) £
<4- 1J
oi ii-
as
na
Ol O)
u .c
c
3 t-
o o
oo u_
Its o
5-42
-------
TABLE 5-6. IDENTIFICATION OF DESIGN PARAMETERS AND COST CURVES
FOR AUXILIARY EQUIPMENT
Auxiliary Equipment
Design Parameters
Cost Curve
Figure No.
Ductwork
Fan3
Stack
Diameter
Length
Material of construction
Actual air flow rate
Pressure drop
Gas stream velocity
Diameter
Length
Material of construction
Fig. 5-15 (p. 5-30)
Fig. 5-16 (p. 5-31)
Fig. 5-17 (p. 5-32)
Fig. 5-18 (p. 5-33)
Fig. 5-19 (p. 5-34)
aAssumed to be located downstream of the control system and ductwork.
5-43
-------
TABLE 5-7. ASSUMED PRESSURE DROPS ACROSS VARIOUS COMPONENTS
Pressure Drop
System Component (in.
Stack 0.5
Ductwork 0.5
Thermal incinerator 4
Heat exchanger 2
Catalytic incinerator 6
Absorber Variable3
Carbon adsorber 6
Condenser 3
Fabric filter 6
Electrostatic precipitator 0.5
Venturi scrubber APV
aUse equation 4.6-21 on page 4.6-17 (Section 4.6) to determine pressure drop,
5-44
-------
TABLE 5-8. CAPITAL COST ELEMENTS AND FACTORSa'b
Control technique
Cost Elements
Direct Costs
Purchased equipment cost0
Other direct costs:
Foundation & supports
Erection & handling
Electrical
Piping
Insulation
Painting
Total Direct Cost
Indirect Costs
Engineering and
supervision
Construction and
field expenses
Construction fee
Start up
Performance test
Model study
Total Indirect Cost
Contingency^
TOTAL*
ESP
1.00
0.04
0.50
0.08
0.01
0.02
0.02
1.67
0.20
0.20
0.10
0.01
0.01
0.02
0.54
0.07
2.27
Venturi
Scrubbers
1.00
0.06
0.40
0.01
0.05
0.02
0.01
1.56
0.10
0.10
0.10
0.01
0.01
—
0.32
0.06
1.94
Fabric
Filters
1.00
0.04
0.50
0.08
0.01
0.07
0.02
1.72
0.10
0.20
0.10
0.01
0.01
—
0.42
0.07
2.21
Thermal 4
Catalytic
Incinerators
1.00
0.08
0.14
0.04
0.02
0.01
0.01
1.30
0.10
0.05
0.10
0.02
0.01
--
0.28
0.05
1.63
Adsorbers
1.00
0.08
0.14
0.04
0.02
0.01
0.01
1.30
0.10
0.05
0.10
0.02
0.01
--
0.28
0.05
1.63
Absorbers
1.00
0.12
0.40
0.01
0.30
0.01
0.01
1.85
0.10
0.10
0.10
0.01
0.01
--
0.32
0.07
2.24
Condensers
1.00
0.08
0.14
0.08
0.02
0.10
0.01
1.43
0.10
0.05
0.10
0.02
0.01
--
0.28
0.05
1.76
aSource: Reference 2.
bAs fractions of total purchased equipment cost. They must.be applied to the total purchased
equipment cost.
cTotal of purchased costs of major equipment and auxiliary equipment and others, which include
instrumentation and controls at 101, taxes and freight at 81 of the equipment purchase cost.
Contingency costs are estimated to equa.l 3J of the total direct and indirect costs.
eFoi- retrofit applications, multiply the total by 1.25.
5-45
-------
TABLE 5-9. UNIT COSTS TO CALCULATE ANNUALIZED COST
Cost Elements
Unit Costs/Factor
Direct Operating Costs
(1) Utilities15:
a. Natural gas
b. Fuel oil
c. Water
d. Steam
e. Electricity
f. Solvent
(2) Operating Labor:
a. Operator Labor
b. Supervision
(3) Maintenance:
a. Labor
b. Materials0
(4) Replacement:
a. 'Parts
b. Labor
Indirect Operating Costs
(1) Overhead
(2) Property Tax
(3) Insurance
(4) Administration
(5) Capital Recovery
Credits
$ 0.00425 per ft3
$ 1.025 per gal
$ 0.0003 per gal
$ 0.00504 per Ib
$ 0.059 per kWh
As applicable
(Ref. 6)
(Ref. 11)
(Ref. 2)
(Ref. 2)
(Ref. 4)
$ 11.53 per hour (Ref. 5)
15% of Operator Labor
$ 11.53 per hour (Ref. 5)
100% of Maintenance Labor
As applicable (see Table 5-10)
100% of Replacement Parts
80% of 2a + 2b + 3a + 4a
1% of Total Capital Cost
1% of Total Capital Cost
2% of Total Capital Cost
(CRFa) x Total Capital Cost
As applicable
aCRF = capital recovery factor.
For an average interest rate of 10 percent, the CRF for specific control
devices are listed below.
(1) ESP and fabric filter: CRF = 0.117 (based on 20-year life span).
(2) Venturi scrubber, thermal and catalytic incinerators, adsorber,
absorber, and condenser: CRF = 0.163 (based on 10-year lifespan).
DRefer to Tables 5-10 and 5-11 to estimate utility costs for each HAP control
technique.
Maintenance materials include operating supplies (e.g., lubrication, paper).
5-46
-------
TABLE 5-10. UTILITY/REPLACEMENT OPERATING COSTS FOR HAP CONTROL TECHNIQUES3
HAP Control Device
Utilities/Replacement Parts
Thermal Incinerator
Catalytic Incinerator
Carbon Adsorber Systems
Absorber Systems
Condenser System
Fabric Filter Systems
Electrostatic Precipitators
V'enturi Scrubbers
Natural gas or fuel oil13
Electricity (fan)
Catalyst costc (Vc.t, p. 4.2-16)
Natural gas or fuel oilb
Electricity (fan)
Carbond (Crea, p. 4.5-10)
Steamb M
Cooling waterb
Electricity (fan)
Absorbent*3 (water or solvent)
Electricity (fan)
Refrigerant6 (Ref, p. 4.7-15)
Electricity (fan)
Bagsf (Atc, p. 4.8-14)
Electricityb (fan + control device)
£lectricityb (fan + control device)
Water5
Electricity (fan)
aRefer to Table 5-9 for utility unit costs, Tables 5-3 and 5-5 for
replacement part unit costs, and Table 5-2 for FE cost indices.
bSee Table 5-11.
cAnnualized replacement catalyst costs are calculated as follows:
Annualized Cost = Vcat (ft3) x $/ft3 (Current FE/Base FE)
3 years
CAnnualized replacement carbon costs are calculated as follows:
Annualized cost = creq (ib) x $/lb. (Current FE/Base FE)
5 years
eRefrigerant replacement is due to system leaks, however, the loss rate of
refrigerant is very low and varies for every unit. Therefore, assume that
the cost of refrigerant replacement is negligible.
CAnnualized replacement bag costs are calculated as follows:
Annualized cost = Atc (ft2) x $/ft2 (Current FE/Base FE)
2 years
5-47
-------
TABLE 5-11. ADDITIONAL UTILITY REQUIREMENTS3
(1) Fuel Requirement for Incinerators,
[Note: The design sections for thermal and catalytic incinerators are
developed under the assumption that natural gas is used as
the supplementary fuel. Fuel oil could be used, however, the
use of natural gas is normal industry practice. If fuel oil
is used, the equation below can be used by replacing Qf with
the fuel oil flow rate in units of gallons per minute. The
product of the equation then equals gallons of fuel oil.
Fuel Requirement = 60 (Qf) x MRS
where: Qf = supplementary fuel required, scfm (p. 4.1-11 or p. 4.2-13)
HRS = annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
(2) Steam Requirement for Carbon Adsorber, Ib
[Note: Assume 4 Ib of steam required for each Ib of recovered product.]
Steam Requirement = 4 (Qrec) x HRS
where: Qrec = quantity of HAP recovered, Ib/hr (p. 4.5-20)
HRS = annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
(3) Cooling Water Requirement for Carbon Adsorber, gal
[Note: Assume 12 gal of cooling water required per 100 1bs steam.]
Water Requirement = 0.48 (Qrec) x HRS
where: Qrec = quantity of HAP recovered, Ib/hr (p. 4.5-20)
HRS = annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
(continued)
5-48
-------
TABLE 5-11. ADDITIONAL UTILITY REQUIREMENTS3
(concluded)
(4) Absorbent Requirement for Absorbers, gal
[Note: Assume no recycle of absorbing fluid (water or solvent).]
Absorbent Requirement = 60 (Lgai) x HRS
where: Lga] = absorbing fluid flow rate, gal/min (p. 4.6-7)
HRS = annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
(5) Water Requirement for Venturi Scrubbers, gal
[Note: Assume 0.01 gal H£0 are required per acf of emission stream.]
Water Requirement = 0.6 (Qe,a) x HRS
where: Qe>a = emission stream flow rate into scrubber, acfm
HRS = annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
(6) Baghouse Electricity Requirement, kWh
[Note: Assume 0.0002 kW are required per ft^ of gross cloth area.]
Baghouse Elect. Req. = 0.0002 (Atc) * HRS
f\
where: Atc = gross cloth area required, ft^ (p. 4.8-14)
HRS = annual operating hours, hr
[Note: Use 8,600 unless otherwise specified.]
(7) ESP Electricity Requirement, kWh
[Note: Assume 0.0015 kW are required per ft? of collection area.]
ESP Elect. Req. = 0.0015 (Ap) x HRS
where: Ap = collection plate area, ft2 (p. 4.9-4)
HRS = annual operating hours, hr
[Note: Use 8,600 unless otherwise specified.]
a
Source: Reference 2.
5-49
-------
TAbLE 5-12. ESTIMATED LABOR HOURS PER SHIFT AND AVERAGE EQUIPMENT LIFE^
Labor Requirements
(hr/shlft)
Control device
Operator
Labor
Maintenance
Labor
Average
Equipment Life
(yr)
Electrostatic precipitator 0.5
Fabric filter 2
Venturi scrubber 2
Incinerator 0.5
Adsorber 0.5
Absorber 0.5
Condenser 0.5
2
4
8
0.5
1
1
0.5
0.5
0.5
0.5
1
2
2
20
20
10
10
10
10
10
aSource: Reference 2.
5-50
-------
TABLE c.i2-i. oRiiiwMf CALCULATIONS FCR CAPITAL ::s: ALGC
(1! Calculation of 3uct Diameter, 0^uct (in.)
124 '/2
/(he-e: QetS » amission stream flow 'ate at actual conditions, acfm
'-'duct " velocity of ;as stream in duct, ft/mm
°duct * 12 / x
If velocity of gas stream in duct is unknown, use 2,300 ft/rain;
tne equation then becomes:
3duct .0.2028 (Q9,a)1/2
°duct * 0--028 ( IS. 500 _ }l/Z » 39 in.
(2) Calculation of Stack Diameter, Ostack (1r')
(
-------
TABLE C.12-2. ESTIMATE OF CAPITAL COSTS IN CURRENT DOLLARS
COST ELEMENTS
1. Major Equip. Pur. Cost
Thermal Incinerator*
Heat Exchanger0
Catalytic Incinerator0
Catalyst0, V_at »
Caroon Adsorber4
Carbon , C,.0_ «
Absorber*
Platforms and Ladders6
Packing8. ypac), »
Condenser^
Refrigerant^
Fabric Filta'S
Sags3, Atr =
ESPh
Venturi Scrubber'
Design Factors'
FIGURE OR
TABLE COST
$ 98,000
S 35,000
$
ft3 x S/ft3
S
Ib x S/lb
s
S
ft2 x S/ft3
$
S
S
ft2 x $/ft2
$
$
(Thickness Factor)
ESCALATION FACTOR
(Cur-ept FE/3ase Ft)
see Table 5-2, p. 5-39
x ( 336.2 / 226.2 )
x ( 336.2 / 226.2 )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
x ( / )
X
(Composition Factor)
SUBTOTAL
CURRENT COST
- S 145,700
» $ 104,400
' $
« S
» S
» S
* $
• J
* S
« S
* S
» $
» S
* S
X
» S
S 250,100
(continued)
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case,
(continued)
5-52
-------
TABLE C.12-2. ESTIMATE OF CAPITAL COSTS IN CURRENT DOLLARS
(ConeI uded)
COST ELEMENTS
FIGURE OR
TABLE COST
ESCALATION FACTOR
(Cur-ant FE/Sase FE)
see Table 5-2, p. 5-39
CURRENT COST
2. Aux. Equip. Purch. Cost
Fan*
I'.otor1
Stack1"
S 52 x 100 x ( 336.2 / 226.2
(Length)
x f I
(Fan Current Cost)
S 4,500
x 0.15
x ( 326.2 / 226.2 )
S 7.700
6.700
SUBTOTAL
S 14.400
•)
4.
5.
6.
t
i .
3.
P-e-Total Purch. Equip. Cost
Adjustments"
TOTAL Pure. Equip. Cost
Instrumentation and Controls
Freight and Taxes
TOTAL Purchased Cost
TOTAL CAPITAL COSTS
Item 1 Subtotal * Item 2 Subtotal
(Item 3) x -0.091
Item 3 * Adjustments
10S of Item 4
8S of Item 4
Item 4 * Item 5 * Item 6
F° x (Item 7); where F » 1.63
S
S
S
S
$
S
S
264,500
-24,100
240,400
24,000
19,200
283,600
462,300
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example casp.
(continued)
5-53
-------
FOOTNOTES TO TABL£ C.12-1
aThennal Incinerator: Figure 5-1 (p. 5-19), includes fan plus instru-
mentation and control costs for thermal incinerators, in addition to
the major equipment purchased cost. Additional auxilia'y equipment
(ductwork and stack) purchased costs and costs of f-eight and taxes
must be added to obtain the total purchased cost.
bHeat Exchangers: If the HAP control system 'equi-es a neat exchanger,
obtain the cost f'om Figure 5-2 (p. 5-20), escalate this cost using
the appropriate facto", and add to the major equipment purchased cost.
°Catalytic Incinerator: Figure 5-3 (p. 5-21) srovides the cost of a
catalytic incinerator, less catalyst costs. The "TABLE" catalyst cost
is estimated iy multiplying the volume of catalyst squired (VCS£,
p. 4.2-16) by the catalyst cost factor ($/ft3) found on Table 5-C
(p. 5-40). Catalyst costs, all auxiliary equipment (ductwork, fan, and
stack) purchased costs and the cost of instrumentation and controls,
and freight and taxes must be added to obtain the total purchased cost.
^Carbon adsorber: Figure 5-4 (p. 5-22) (packaged carbon adsoroer
systems) includes the cost of carbon, beas, fan and motor, instrumen-
tation and controls, and a steam 'agenerator. Adoptions! auxiliary
equipment (ductwork and stack) purchased costs and costs of freight
and taxes must be added to obtain the total purchased cost.
Figure 5-5 (p. 5-23) (custom carbon adsorber systems) includes beds,
instrumentation and controls, and a steam regenerator, less caroon.
The "TABLE" carbon cost for custom carbon adsorbers is estimated by
multiplying the weight of carbon required (C^M, p. 4.5-10) by the
caroon cost factor (S/1b) found on Table 5-3 (p. 5-40). Costs of carbon,
alt auxiliary equipment (duct, fan, stack) purchased costs, and freight
and taxes must be added to obtain the total purchased cost.
eAbsoroer: Figure 5-6 (p. 5-24} does not include the cost of packing,
platforms, and ladders. The cost of platforms and ladders (Fig. 5-7
p. 5-25) and packing must be added to obtain the major purchased equipment
cost. The "TABLE" packing cost is estimated by multiplying the volume
of packing required (Vgack. P« 4.5-16) by the appropriate packing cost
factor found on Table 5-4 (p. 5-41). All auxiliary equipment (ductwork,
fan, and stack) purchased costs, and costs of freight and taxes must
be added to obtain the total purchased cost.
fCondenser Systems: Figure 5-8 (p. 5-26) yields total capital costs
for cold water condenser systems.- For systems needing refrigerant,
the applicable cost from Figure 5-9 (p. 5-27) must be added to
obtain the total capital costs. In either case, the escalated cost
estimate is then placed on Line 3, "TOTAL CAPITAL COSTS."
SFabric Filter Systems: Figure 5-10 (p. 5-28) gives the cost of j
negative pressure, insulated bagfiouse. The curve does not include sag
costs. The "TABLE" bag cost is estimated by multiplying the gross
cloth area required (A^c, p. 4.3-14) by the appropriate bag cost factor
found on Table 5-5 (p. 5-42). Bag costs, all auxiliary equipment
(duct, fan, and stack) purchased costs, the cost of instrumentation
and controls, and freight and taxes must be added to obtain the
total purchased cost.
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-54
-------
FOOTNOTES TO . TABLE C.12-2 (Concluded)
Electrostatic P'ecipitators: Figure 5-il (?. 5-29) jrovides the cost
for an insulated ESP. All auxiliary equipment (duct, fan, ana stack)
purchased costs, the cost of instrumentation and controls, and
freight and taxes must oe added to ootain the total purchased cost.
'Venturi Scrubber: Figure 5-12 (p. 5-20) includes the cost of
instrumentation and controls in addition to the major equipment
purchased cost. This cost curve is based on a venturi sc'-ubbe"'
constructed from 1/8-inch carbon steel. Figure 5-13 (p. 3-21) is used
to determine if 1/8-incn steel is appropriate for a given application
(use the higher curve). If thicker steel is -equired. Figure 5-i4
(p. 5-32) yields an adjustment factor for various steel thicknesses;
this factor is used to escalate the cost obtained from Figure 5-12.
In addition, if stainless steel is -equi'ed (see Section 4,10.3.2)
iiultiply the scrubber cost estimate by 2.3 for 2041 stainless steel or
by 3.2 for 216L stainless steel. Costs of all auxiliary equipment
(ductwork, fan, and stack) and freight and taxes must be added to
obtain the total purchased cost.
JOuctwork: Figure 5-15 (p. 5-33) gives the cost of straight ductwork
made of carbon steel for various thicknesses, based on the 'equi'ed
duct diameter. Figure 5-15 (p. 5-24) gives the cost of straight
ductwork made of stainless steel for various thicknesses, based on the
-equired duct diameter. Preliminary calculations (duct diameter, see
Table C. 12-1) are necessary to estimate ductwork costs.
*Fan: Figure 5-17 (p. 5-35) gives the cost of a fan based on the gas
flow rate at actual conditions and the HAP control system pressure
drop (in inches of HjO). The applicable fan class is also based on the
HAP control system pressure drop. Calculation of the total system
pressure drop is presented in Table C.12-1.
^he cost of a motor is estimated as 151 of the fan cost.
""Stack: Figure 5-18 (p. 5-36) gives the cost of a caroon steel stack
at various stack heights and diameters. Figure 5-19 (p. 5-27) gives
the price of a stainless steel stack at various stack heights and
diameters. Preliminary calculations (stack diameter, see Table C.12-1)
are necessary to estimate stack costs. For both figures, use the curve
that best represents the calculated diameter.
"For thermal incinerators, carbon adsorbers, and venturi scrubbers, the
purchase cost curve includes the cost for instrumentation and controls.
This cost (i.e., the "Adjustment") must be subtracted out to estimate
the total purchased equipment cost. This is done oy adding the Item 1
subtotal and the Item 2 subtotal and multiplying the -esult by -0.091.
This value is added to the preliminary total purchased equipment cost
to obtain the total purchased equipment cost. For all other major
equipment, the "Adjustment'1 equals zero.
°0bta1n factor "F" from "TOTAL" line in Table 5-8 (p. 5-45).
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-55
-------
TABLE C.12-3. PRELIMINARY CALCULATIONS FOR AHNUALIZED COST ALGOfiiTHf,
(1) Calculation of Annual Elect. Requi-eraent, AER (Line 5, Table C.12-6)
a. Fan Electricity Requirement, FER
FER » 0.0002 (Qfgt4) x iP x MRS
whe^e: ^fg.a » actual flue gas flow 'ate, acfm
AP » total HAP control system pressure drop, in. H?0
(see Taole 5-7, p. 5-44)
HRS • annual operating hours, hr
[Note: use 3,600 unless otherwise specified.]
FER « 0.0002 ( .. 40,000 ) x 7
b. Saghouse Electricity Requirement, SER
[Note: Assume 0.0002 kW are requi-sd per ft2 of gross cloth area.]
8ER « 0.0002 (Atc) x HRS
where: Atc » gross cloth area required, ft2 (p. 4.8-14)
3ER * 0.0002 ( ) x =. kW
f.. ESP Electricity Requirement, EER
[Note: Assume 0.0015 kW are required per ft2 of collection area.]
EER * 0.0015 (Ap) x HRS
where: Ap * collection plate area, ft2 (p. 4.9-4)
EER * 0.0015 ( _ ) x _ « _ kWh
d. Annual Electricity Requirement, AER
AER . FER * BER + EER
AER - 481.600 * _ + _ - 481.600 kWh
(continued)
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-56
-------
TABLE C.12-3. PRELIMINARY CALCULATIONS FOR ANNUAHZEO COST ALGORITHM
(cone!udea)
(2) Calculation of Capital Recovery Factor, CRF (Line 13, Table C.12-6)
CRF > [1(1 * i)n] / [(1 + i)n - 1]
•ine'ei i « interest 'ate on borrowed capital, decimal fraction
[Note: Jnless otherwise specified use IQ percent.]
n * control device lifetime, years (see Table 5-12, a. 5-50)
CRF " [ Q.I x (I * 0.1 ) UQ1] / [(I <• 0.1 )UOi-l] , 0,162
(3) Calculation of Annual Operator Labor, OL (Line 9, Table C.12-6)
OL » (HRS) (operator hours per shift) / (operating hours per shift)
[Note: Obtain operator hr/snift value from Table 5-12, p. 5-50.]
OL * ( 8,600 ) x ( 0.5 ) / ( 3 ) - 537._S hr
(4) Calculation of Annual Maintenance Labor, ML (Line 11. Table C.12-6)
ML « (HRS) (maintenance hours per shift) / (operating Hours per shift)
[Note: Obtain maintenance hr/snift value from Table 5-12, p. 5-50.]
ML • ( 8,SOO_ ) x ( 0,5 .. ) / ( 8 ..) » 537.5 hr
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-57
-------
TA3L£ C.12-J. ADDITIONAL 'JTILITY
(I) Fuel Requirement for Incinerators (Line 1 or Line 2, Table C.12-6)
[Note: The design sections for Crtemal and catalytic incinerators are
developed under the assumption that natural gas is used as
the supplementary fuel. Fuel oil could be used, howeve-, the
use of natural gas is nomal industry practice. If fuel oil
is used, the equation below can be used by ~ap!acing 3f rfith
the fuel oil flow "ate in units of gallons per minute. The
resultant product of the equation (gallons of fuel oil -equi-ea)
is then used on Line 2 of Table C.12-6.]
Fuel Requi-ement » 50 (Qf) x HRS
wne^e: Qf * supplementary fuel 'equi"-ed, scfra (p. 4.1-11 o" p. 4.2-12)
HRS • annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
Fuel Requirement » SO ( 330 ) x 8.600. ' 170.280.000., ft3
(2) Steam Requirement for Carbon Adsorber (Line 4, Table C.12-6)
[Note: Assume 4 Ib of steam ••equi'-ed for each Ib of -ecovered product.]
Steam Requirement » 4 (Q^e<:) x HRS
where: qrec * quantity of HAP recovered, Ib/hr (p. 4.5-20)
HRS » annual operating hours, hr
[Note: Use 8,600 hours unless otherwise specified.]
Steam Requirement * 4 ( ) x * Ib
(3) Cooling Water Requirement for Carbon Adsorber (Line 3, Taole C.12-6)
[Note: Assume 12 gal of cooling water required per 100 1 bs steam.]
Water Requirement » 0.48 (Qrec) x HRS
where: Qrec » quantity of HAP --ecovered, Ib/hr (p. 4.5-20)
HRS * annual operating hours, hr
[Note: Use 3,600 hours unless otherwise specified.]
Water Requirement » 0.48 ( ) x * gal
(continued)
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-58
-------
TABLE C.12-4. ADDITIONAL 'JTILITY COSTS
(concluded)
(4) Absorbent Requirement for Absorbers (Line 2 or 6, Table C.12-6)
[Note: Assume no 'ecycle of absorbing fluid (water o* solvent).]
Absorbent Requirement * 60 (Lga]j x HRS
where: Lga] * absorbing fluid flow 'ate, gal/nnn (p. 4.5-7)
HRS » annual operating hours, hr
[Note: Use 8,600 nours unless otherwise specified.]
Absorbent Requirement « 60 ( _) x _____^ " 9al
(5) Water Requirement for Venturi Scrubbers (Line 3, Table C.12-6)
[Note: Assume 0.01 gal of water 'equired per acf of emission stream.]
Water Requirement • 0.6 (Qe>a) x HRS
where: Qe
-------
TABLE C.I 2-5. ESTIMATION OF REPLACEMENT PARTS ANNUAL IZEO COSTS
(1) Annualized Catalyst Replacement Costs (Line 7, Table C.12-6)
Over the lifetime of a catalytic incinerator, the catalyst is depleted
and must De replaced (assune catalyst lifetime is 3 years):
Annual Catalyst Cost » {Catalyst Current Cost3) / 3
Annual Catalyst Cost » ( ) / 2 * 3
(2) Annualized Carbon Replacement Costs (Line 7, Table C.12-6)
Over the lifetime of a carbon adsorber, the carbon is depleted and
must be replaced (assume carbon lifetime is 5 years):
Annual Carbon Cost • (Carbon Current Costd) / S
Annual Carbon Cost » ( ) / 5 • S
(3) Annualized Refrigerant Replacement Costs
Refrigerant in a condenser r>eeds to be "eplaced periodically due to
system leaks; however, the loss 'ate is typically very low. Therefore, assume
the cost of refrigerant replacement is negligible.
(*) Annualized Sag Replacement Costs (Line 7, Table C.12-6)
Over the lifetime of a fabric filter system the bags become *orn and
must be replaced (assume bag lifetime is 2 years):
Annual Sag Cost » (Sag Current Costa) / 2
Annual Sag Cost » ( ) / 2 « $
Table C.12-2.
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(continued)
5-60
-------
TABLE C.12-6. ESTIMATE OF ANNUALIZED COSTS IN CURRENT DOLLARS
COST
ELEMENTS
Direct Operating Costs
1. Natural Gas*
2. Fuel 011*
3. Water*
4. Steam*
5. Electricity6
6 Solvent*
7. Replacement Parts
3. Replacement Labor
9. Operator Labor0
10. Supervision Labor
11. Maintenance Labor13
12. Maintenance Materials
12. SUBTOTAL
Indirect Operating Costs
14. Overhead
15. Property Tax
16. Insurance
17. Administration
18. Capital Recovery^
19. SUBTOTAL
20. CREDITS
NET ANNUAL IZED COSTS
UNIT COSTS/FACTOR ANNUAL EXPENDITURE CURRENT DOLLARS
$0.00425 per ft3 x 170,280,000 ft; * $
SI. 025 per gal x gal - S
$0.0005 per gal x gal » J
SO. 00504 per Ib x Ib « S
SO. 059 per kWh x 481,600 kWh • S
S per galc x gal » S
As applicable (see Table C.12-5) S
1005 of Line 7 S
S11.53'per hr x 537.5 hr « S
15S of Line 9 S
$11.53 per hr x 537.5 hr « S
100S of Line 11 S
Add Items 1 through 12 S
80S of Sum of Lines 8, 9, 10, and 11 3
11 of Total Capital Costd S
IS of Total Capital Costd S
25 of Total Capital Costd . S
(CRF) x Total Capital Costd; where CRF « 0.163 S
Add Items 14 through 13 S
As applicable (see Section 5.2.3} S
Item 13 + Item 19 - Item 20 S
723,700
28,400
6,200
900
6,200
6,200
771,500
10,500
4,600
4,600
9,200
75,400
104,400
876, GOO
4See Table C.12-
Figure 5-20. Completed cost calculation worksheets for the
thermal incinerator example case.
(concluded)
5-61
-------
5.3 REFERENCES FOR CHAPTER 5
1. U.S. EPA. Identification, Assessment, and Control of Fugitive
Participate Emissions.Draft final report.EPA Contract No.
68-02-3922. April 10, 1985. Contact Dale Harmon of EPA at
(919) 541-2429.
2. U.S. EPA. Capital and Operating Costs of Selected Air Pollution
Control Systems. EPA-450/5-80-002. December 1978.
3. U.S. EPA. Organic Chemical Manufacturing Volume 4: Combustion
Control Devices. EPA-450/3-80-026. December 1980.
4. Monthly Energy Review. October 1982 electricity costs.
DOE/EIA-003583/01. January 1983.
5. Survey of Current Business. November 1982. Primary Metal Industry
Costs. December 1982.
6. Phillip Gennarelli, Manager, American Gas Association (AGA).
Monthly Gas Utility Statistical Report. AGA, 1515 Wilson Blvd.,
Arlington, Va.22209.July 1985.
7. Vatavuk, W.M. and Neveril, R.B. "Estimate the Size and Cost of
Incinerators." Chemical Engineering. July 12, 1982.
8. Vatavuk, W.M. and Neveril, R.B. "Costs of Carbon Adsorbers."
Chemical Engineering. January 23, 1983.
9. Vatavuk, W.M. and Neveril, R.B. "Costs of Gas Adsorbers."
Chemical Engineering. October 12, 1982.
10. U.S. EPA. Organic Chemical Manufacturing. Volume 5: Adsorption,
Condensation, and Absorption Devices. EPA-450/3-80-077.December
1980.
11. Bureau of Labor Statistics. August 1985. U.S. consumer price for
fuel oil.
12. Chemical Marketing Reporter. Schnel1 Publishing Company. New York,
NY. Volume 228. September 16, 1985.
13. Telecon. Charlotte Clark (PES) to David Chalmers (Calgon Corp.).
June 12, 1985.
14. Telecon. Gunseli Sagen Shareef (Radian Corp.) to Roy Uhlman
(Engelhart Specialty Chemicals). September 30, 1985.
5-62
------- |